Merge strmatch, xlate, crypto and xml directories from apr-util.

git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@757729 13f79535-47bb-0310-9956-ffa450edef68

36 files changed, 16915 insertions, 0 deletions

diff --git a/xml/NWGNUmakefile b/xml/NWGNUmakefile
new file mode 100644
index 000000000..5f85964f6
--- /dev/null
+++ b/xml/NWGNUmakefile
@@ -0,0 +1,258 @@
+#
+# Declare the sub-directories to be built here
+#
+
+SUBDIRS = \
+	$(EOLIST) 
+
+#
+# Get the 'head' of the build environment.  This includes default targets and
+# paths to tools
+#
+
+include $(APR_WORK)\build\NWGNUhead.inc
+
+#
+# build this level's files
+
+#
+# Make sure all needed macro's are defined
+#
+
+#
+# These directories will be at the beginning of the include list, followed by
+# INCDIRS
+#
+XINCDIRS	+= \
+			$(APR)/include \
+			$(APR)/include/arch/NetWare \
+			$(APRUTIL)/include \
+			$(APRUTIL)/uri \
+			$(APRUTIL)/dbm/sdbm \
+			$(APRUTIL)/include/private \
+			$(APRUTIL)/xml/expat/lib \
+			$(EOLIST)
+
+#
+# These flags will come after CFLAGS
+#
+XCFLAGS		+= \
+			$(EOLIST)
+
+#
+# These defines will come after DEFINES
+#
+XDEFINES	+= \
+			$(EOLIST)
+
+#
+# These flags will be added to the link.opt file
+#
+XLFLAGS		+= \
+			$(EOLIST)
+
+#
+# These values will be appended to the correct variables based on the value of
+# RELEASE
+#
+ifeq "$(RELEASE)" "debug"
+XINCDIRS	+= \
+			$(EOLIST)
+
+XCFLAGS		+= \
+			$(EOLIST)
+
+XDEFINES	+= \
+			$(EOLIST)
+
+XLFLAGS		+= \
+			$(EOLIST)
+endif
+
+ifeq "$(RELEASE)" "noopt"
+XINCDIRS	+= \
+			$(EOLIST)
+
+XCFLAGS		+= \
+			$(EOLIST)
+
+XDEFINES	+= \
+			$(EOLIST)
+
+XLFLAGS		+= \
+			$(EOLIST)
+endif
+
+ifeq "$(RELEASE)" "release"
+XINCDIRS	+= \
+			$(EOLIST)
+
+XCFLAGS		+= \
+			$(EOLIST)
+
+XDEFINES	+= \
+			$(EOLIST)
+
+XLFLAGS		+= \
+			$(EOLIST)
+endif
+
+#
+# These are used by the link target if an NLM is being generated
+# This is used by the link 'name' directive to name the nlm.  If left blank
+# TARGET_nlm (see below) will be used.
+#
+NLM_NAME		= 
+
+#
+# This is used by the link '-desc ' directive. 
+# If left blank, NLM_NAME will be used.
+#
+NLM_DESCRIPTION	= 
+
+#
+# This is used by the '-threadname' directive.  If left blank,
+# NLM_NAME Thread will be used.
+#
+NLM_THREAD_NAME	= 
+#
+# If this is specified, it will override VERSION value in 
+# $(APR_WORK)\build\NWGNUenvironment.inc
+#
+NLM_VERSION		=
+
+#
+# If this is specified, it will override the default of 64K
+#
+NLM_STACK_SIZE	= 
+
+#
+# If this is specified it will be used by the link '-entry' directive
+#
+NLM_ENTRY_SYM	= 
+
+#
+# If this is specified it will be used by the link '-exit' directive
+#
+NLM_EXIT_SYM	= 
+
+#
+# If this is specified it will be used by the link '-check' directive
+#
+NLM_CHECK_SYM	=
+
+#
+# If this is specified it will be used by the link '-flags' directive
+#
+NLM_FLAGS		=
+ 
+#
+# If this is specified it will be linked in with the XDCData option in the def 
+# file instead of the default of $(APR)/misc/netware/apache.xdc.  XDCData can 
+# be disabled by setting APACHE_UNIPROC in the environment
+#
+XDCDATA         = 
+
+#
+# Declare all target files (you must add your files here)
+#
+
+#
+# If there is an NLM target, put it here
+#
+TARGET_nlm = \
+	$(EOLIST)
+
+#
+# If there is an LIB target, put it here
+#
+TARGET_lib = \
+	$(OBJDIR)/xmllib.lib \
+	$(EOLIST)
+
+#
+# These are the OBJ files needed to create the NLM target above.
+# Paths must all use the '/' character
+#
+FILES_nlm_objs = \
+	$(EOLIST)
+
+#
+# These are the LIB files needed to create the NLM target above.
+# These will be added as a library command in the link.opt file.
+#
+FILES_nlm_libs = \
+	$(EOLIST)
+
+#
+# These are the modules that the above NLM target depends on to load.
+# These will be added as a module command in the link.opt file.
+#
+FILES_nlm_modules = \
+	$(EOLIST)
+
+#
+# If the nlm has a msg file, put it's path here
+#
+FILE_nlm_msg =
+ 
+#
+# If the nlm has a hlp file put it's path here
+#
+FILE_nlm_hlp =
+
+#
+# If this is specified, it will override the default copyright.
+#
+FILE_nlm_copyright =
+
+#
+# Any additional imports go here
+#
+FILES_nlm_Ximports = \
+	$(EOLIST)
+ 
+#   
+# Any symbols exported to here
+#
+FILES_nlm_exports = \
+	$(EOLIST)
+	
+#   
+# These are the OBJ files needed to create the LIB target above.
+# Paths must all use the '/' character
+#
+FILES_lib_objs = \
+	$(OBJDIR)/apr_xml.o \
+	$(OBJDIR)/xmlparse.o \
+	$(OBJDIR)/xmlrole.o \
+	$(OBJDIR)/xmltok.o \
+	$(EOLIST)
+
+#
+# implement targets and dependancies (leave this section alone)
+#
+
+libs :: $(OBJDIR) $(TARGET_lib)
+
+nlms :: libs $(TARGET_nlm)
+
+#
+# Updated this target to create necessary directories and copy files to the 
+# correct place.  (See $(APR_WORK)\build\NWGNUhead.inc for examples)
+#
+install :: nlms FORCE
+
+#
+# Any specialized rules here
+#
+
+vpath %.c expat/lib
+
+#
+# Include the 'tail' makefile that has targets that depend on variables defined
+# in this makefile
+#
+
+include $(APR_WORK)\build\NWGNUtail.inc
+
diff --git a/xml/apr_xml.c b/xml/apr_xml.c
new file mode 100644
index 000000000..256223f44
--- /dev/null
+++ b/xml/apr_xml.c
@@ -0,0 +1,983 @@
+/* Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+#include "apr.h"
+#include "apr_strings.h"
+
+#define APR_WANT_STDIO          /* for sprintf() */
+#define APR_WANT_STRFUNC
+#include "apr_want.h"
+
+#include "apr_xml.h"
+
+#include "apu_config.h"
+
+#if defined(HAVE_XMLPARSE_XMLPARSE_H)
+#include <xmlparse/xmlparse.h>
+#elif defined(HAVE_XMLTOK_XMLPARSE_H)
+#include <xmltok/xmlparse.h>
+#elif defined(HAVE_XML_XMLPARSE_H)
+#include <xml/xmlparse.h>
+#else
+#include <expat.h>
+#endif
+
+#define DEBUG_CR "\r\n"
+
+static const char APR_KW_xmlns[] = { 0x78, 0x6D, 0x6C, 0x6E, 0x73, '\0' };
+static const char APR_KW_xmlns_lang[] = { 0x78, 0x6D, 0x6C, 0x3A, 0x6C, 0x61, 0x6E, 0x67, '\0' };
+static const char APR_KW_DAV[] = { 0x44, 0x41, 0x56, 0x3A, '\0' };
+
+/* errors related to namespace processing */
+#define APR_XML_NS_ERROR_UNKNOWN_PREFIX (-1000)
+#define APR_XML_NS_ERROR_INVALID_DECL (-1001)
+
+/* test for a namespace prefix that begins with [Xx][Mm][Ll] */
+#define APR_XML_NS_IS_RESERVED(name) \
+	( (name[0] == 0x58 || name[0] == 0x78) && \
+	  (name[1] == 0x4D || name[1] == 0x6D) && \
+	  (name[2] == 0x4C || name[2] == 0x6C) )
+
+
+/* the real (internal) definition of the parser context */
+struct apr_xml_parser {
+    apr_xml_doc *doc;		/* the doc we're parsing */
+    apr_pool_t *p;		/* the pool we allocate from */
+    apr_xml_elem *cur_elem;	/* current element */
+
+    int error;			/* an error has occurred */
+#define APR_XML_ERROR_EXPAT             1
+#define APR_XML_ERROR_PARSE_DONE        2
+/* also: public APR_XML_NS_ERROR_* values (if any) */
+
+    XML_Parser xp;              /* the actual (Expat) XML parser */
+    enum XML_Error xp_err;      /* stored Expat error code */
+};
+
+/* struct for scoping namespace declarations */
+typedef struct apr_xml_ns_scope {
+    const char *prefix;		/* prefix used for this ns */
+    int ns;			/* index into namespace table */
+    int emptyURI;		/* the namespace URI is the empty string */
+    struct apr_xml_ns_scope *next;	/* next scoped namespace */
+} apr_xml_ns_scope;
+
+
+/* return namespace table index for a given prefix */
+static int find_prefix(apr_xml_parser *parser, const char *prefix)
+{
+    apr_xml_elem *elem = parser->cur_elem;
+
+    /*
+    ** Walk up the tree, looking for a namespace scope that defines this
+    ** prefix.
+    */
+    for (; elem; elem = elem->parent) {
+	apr_xml_ns_scope *ns_scope = elem->ns_scope;
+
+	for (ns_scope = elem->ns_scope; ns_scope; ns_scope = ns_scope->next) {
+	    if (strcmp(prefix, ns_scope->prefix) == 0) {
+		if (ns_scope->emptyURI) {
+		    /*
+		    ** It is possible to set the default namespace to an
+		    ** empty URI string; this resets the default namespace
+		    ** to mean "no namespace." We just found the prefix
+		    ** refers to an empty URI, so return "no namespace."
+		    */
+		    return APR_XML_NS_NONE;
+		}
+
+		return ns_scope->ns;
+	    }
+	}
+    }
+
+    /*
+     * If the prefix is empty (""), this means that a prefix was not
+     * specified in the element/attribute. The search that was performed
+     * just above did not locate a default namespace URI (which is stored
+     * into ns_scope with an empty prefix). This means the element/attribute
+     * has "no namespace". We have a reserved value for this.
+     */
+    if (*prefix == '\0') {
+	return APR_XML_NS_NONE;
+    }
+
+    /* not found */
+    return APR_XML_NS_ERROR_UNKNOWN_PREFIX;
+}
+
+static void start_handler(void *userdata, const char *name, const char **attrs)
+{
+    apr_xml_parser *parser = userdata;
+    apr_xml_elem *elem;
+    apr_xml_attr *attr;
+    apr_xml_attr *prev;
+    char *colon;
+    const char *quoted;
+    char *elem_name;
+
+    /* punt once we find an error */
+    if (parser->error)
+	return;
+
+    elem = apr_pcalloc(parser->p, sizeof(*elem));
+
+    /* prep the element */
+    elem->name = elem_name = apr_pstrdup(parser->p, name);
+
+    /* fill in the attributes (note: ends up in reverse order) */
+    while (*attrs) {
+	attr = apr_palloc(parser->p, sizeof(*attr));
+	attr->name = apr_pstrdup(parser->p, *attrs++);
+	attr->value = apr_pstrdup(parser->p, *attrs++);
+	attr->next = elem->attr;
+	elem->attr = attr;
+    }
+
+    /* hook the element into the tree */
+    if (parser->cur_elem == NULL) {
+	/* no current element; this also becomes the root */
+	parser->cur_elem = parser->doc->root = elem;
+    }
+    else {
+	/* this element appeared within the current elem */
+	elem->parent = parser->cur_elem;
+
+	/* set up the child/sibling links */
+	if (elem->parent->last_child == NULL) {
+	    /* no first child either */
+	    elem->parent->first_child = elem->parent->last_child = elem;
+	}
+	else {
+	    /* hook onto the end of the parent's children */
+	    elem->parent->last_child->next = elem;
+	    elem->parent->last_child = elem;
+	}
+
+	/* this element is now the current element */
+	parser->cur_elem = elem;
+    }
+
+    /* scan the attributes for namespace declarations */
+    for (prev = NULL, attr = elem->attr;
+	 attr;
+	 attr = attr->next) {
+	if (strncmp(attr->name, APR_KW_xmlns, 5) == 0) {
+	    const char *prefix = &attr->name[5];
+	    apr_xml_ns_scope *ns_scope;
+
+	    /* test for xmlns:foo= form and xmlns= form */
+	    if (*prefix == 0x3A) {
+                /* a namespace prefix declaration must have a
+                   non-empty value. */
+                if (attr->value[0] == '\0') {
+                    parser->error = APR_XML_NS_ERROR_INVALID_DECL;
+                    return;
+                }
+		++prefix;
+            }
+	    else if (*prefix != '\0') {
+		/* advance "prev" since "attr" is still present */
+		prev = attr;
+		continue;
+	    }
+
+	    /* quote the URI before we ever start working with it */
+	    quoted = apr_xml_quote_string(parser->p, attr->value, 1);
+
+	    /* build and insert the new scope */
+	    ns_scope = apr_pcalloc(parser->p, sizeof(*ns_scope));
+	    ns_scope->prefix = prefix;
+	    ns_scope->ns = apr_xml_insert_uri(parser->doc->namespaces, quoted);
+	    ns_scope->emptyURI = *quoted == '\0';
+	    ns_scope->next = elem->ns_scope;
+	    elem->ns_scope = ns_scope;
+
+	    /* remove this attribute from the element */
+	    if (prev == NULL)
+		elem->attr = attr->next;
+	    else
+		prev->next = attr->next;
+
+	    /* Note: prev will not be advanced since we just removed "attr" */
+	}
+	else if (strcmp(attr->name, APR_KW_xmlns_lang) == 0) {
+	    /* save away the language (in quoted form) */
+	    elem->lang = apr_xml_quote_string(parser->p, attr->value, 1);
+
+	    /* remove this attribute from the element */
+	    if (prev == NULL)
+		elem->attr = attr->next;
+	    else
+		prev->next = attr->next;
+
+	    /* Note: prev will not be advanced since we just removed "attr" */
+	}
+	else {
+	    /* advance "prev" since "attr" is still present */
+	    prev = attr;
+	}
+    }
+
+    /*
+    ** If an xml:lang attribute didn't exist (lang==NULL), then copy the
+    ** language from the parent element (if present).
+    **
+    ** NOTE: elem_size() *depends* upon this pointer equality.
+    */
+    if (elem->lang == NULL && elem->parent != NULL)
+	elem->lang = elem->parent->lang;
+
+    /* adjust the element's namespace */
+    colon = strchr(elem_name, 0x3A);
+    if (colon == NULL) {
+	/*
+	 * The element is using the default namespace, which will always
+	 * be found. Either it will be "no namespace", or a default
+	 * namespace URI has been specified at some point.
+	 */
+	elem->ns = find_prefix(parser, "");
+    }
+    else if (APR_XML_NS_IS_RESERVED(elem->name)) {
+	elem->ns = APR_XML_NS_NONE;
+    }
+    else {
+	*colon = '\0';
+	elem->ns = find_prefix(parser, elem->name);
+	elem->name = colon + 1;
+
+	if (APR_XML_NS_IS_ERROR(elem->ns)) {
+	    parser->error = elem->ns;
+	    return;
+	}
+    }
+
+    /* adjust all remaining attributes' namespaces */
+    for (attr = elem->attr; attr; attr = attr->next) {
+        /*
+         * apr_xml_attr defines this as "const" but we dup'd it, so we
+         * know that we can change it. a bit hacky, but the existing
+         * structure def is best.
+         */
+        char *attr_name = (char *)attr->name;
+
+	colon = strchr(attr_name, 0x3A);
+	if (colon == NULL) {
+	    /*
+	     * Attributes do NOT use the default namespace. Therefore,
+	     * we place them into the "no namespace" category.
+	     */
+	    attr->ns = APR_XML_NS_NONE;
+	}
+	else if (APR_XML_NS_IS_RESERVED(attr->name)) {
+	    attr->ns = APR_XML_NS_NONE;
+	}
+	else {
+	    *colon = '\0';
+	    attr->ns = find_prefix(parser, attr->name);
+	    attr->name = colon + 1;
+
+	    if (APR_XML_NS_IS_ERROR(attr->ns)) {
+		parser->error = attr->ns;
+		return;
+	    }
+	}
+    }
+}
+
+static void end_handler(void *userdata, const char *name)
+{
+    apr_xml_parser *parser = userdata;
+
+    /* punt once we find an error */
+    if (parser->error)
+	return;
+
+    /* pop up one level */
+    parser->cur_elem = parser->cur_elem->parent;
+}
+
+static void cdata_handler(void *userdata, const char *data, int len)
+{
+    apr_xml_parser *parser = userdata;
+    apr_xml_elem *elem;
+    apr_text_header *hdr;
+    const char *s;
+
+    /* punt once we find an error */
+    if (parser->error)
+	return;
+
+    elem = parser->cur_elem;
+    s = apr_pstrndup(parser->p, data, len);
+
+    if (elem->last_child == NULL) {
+	/* no children yet. this cdata follows the start tag */
+	hdr = &elem->first_cdata;
+    }
+    else {
+	/* child elements exist. this cdata follows the last child. */
+	hdr = &elem->last_child->following_cdata;
+    }
+
+    apr_text_append(parser->p, hdr, s);
+}
+
+static apr_status_t cleanup_parser(void *ctx)
+{
+    apr_xml_parser *parser = ctx;
+
+    XML_ParserFree(parser->xp);
+    parser->xp = NULL;
+
+    return APR_SUCCESS;
+}
+
+APU_DECLARE(apr_xml_parser *) apr_xml_parser_create(apr_pool_t *pool)
+{
+    apr_xml_parser *parser = apr_pcalloc(pool, sizeof(*parser));
+
+    parser->p = pool;
+    parser->doc = apr_pcalloc(pool, sizeof(*parser->doc));
+
+    parser->doc->namespaces = apr_array_make(pool, 5, sizeof(const char *));
+
+    /* ### is there a way to avoid hard-coding this? */
+    apr_xml_insert_uri(parser->doc->namespaces, APR_KW_DAV);
+
+    parser->xp = XML_ParserCreate(NULL);
+    if (parser->xp == NULL) {
+        (*apr_pool_abort_get(pool))(APR_ENOMEM);
+        return NULL;
+    }
+
+    apr_pool_cleanup_register(pool, parser, cleanup_parser,
+                              apr_pool_cleanup_null);
+
+    XML_SetUserData(parser->xp, parser);
+    XML_SetElementHandler(parser->xp, start_handler, end_handler);
+    XML_SetCharacterDataHandler(parser->xp, cdata_handler);
+
+    return parser;
+}
+
+static apr_status_t do_parse(apr_xml_parser *parser,
+                             const char *data, apr_size_t len,
+                             int is_final)
+{
+    if (parser->xp == NULL) {
+        parser->error = APR_XML_ERROR_PARSE_DONE;
+    }
+    else {
+        int rv = XML_Parse(parser->xp, data, (int)len, is_final);
+
+        if (rv == 0) {
+            parser->error = APR_XML_ERROR_EXPAT;
+            parser->xp_err = XML_GetErrorCode(parser->xp);
+        }
+    }
+
+    /* ### better error code? */
+    return parser->error ? APR_EGENERAL : APR_SUCCESS;
+}
+
+APU_DECLARE(apr_status_t) apr_xml_parser_feed(apr_xml_parser *parser,
+                                              const char *data,
+                                              apr_size_t len)
+{
+    return do_parse(parser, data, len, 0 /* is_final */);
+}
+
+APU_DECLARE(apr_status_t) apr_xml_parser_done(apr_xml_parser *parser,
+                                              apr_xml_doc **pdoc)
+{
+    char end;
+    apr_status_t status = do_parse(parser, &end, 0, 1 /* is_final */);
+
+    /* get rid of the parser */
+    (void) apr_pool_cleanup_run(parser->p, parser, cleanup_parser);
+
+    if (status)
+        return status;
+
+    if (pdoc != NULL)
+        *pdoc = parser->doc;
+    return APR_SUCCESS;
+}
+
+APU_DECLARE(char *) apr_xml_parser_geterror(apr_xml_parser *parser,
+                                            char *errbuf,
+                                            apr_size_t errbufsize)
+{
+    int error = parser->error;
+    const char *msg;
+
+    /* clear our record of an error */
+    parser->error = 0;
+
+    switch (error) {
+    case 0:
+        msg = "No error.";
+        break;
+
+    case APR_XML_NS_ERROR_UNKNOWN_PREFIX:
+        msg = "An undefined namespace prefix was used.";
+        break;
+
+    case APR_XML_NS_ERROR_INVALID_DECL:
+        msg = "A namespace prefix was defined with an empty URI.";
+        break;
+
+    case APR_XML_ERROR_EXPAT:
+        (void) apr_snprintf(errbuf, errbufsize,
+                            "XML parser error code: %s (%d)",
+                            XML_ErrorString(parser->xp_err), parser->xp_err);
+        return errbuf;
+
+    case APR_XML_ERROR_PARSE_DONE:
+        msg = "The parser is not active.";
+        break;
+
+    default:
+        msg = "There was an unknown error within the XML body.";
+        break;
+    }
+
+    (void) apr_cpystrn(errbuf, msg, errbufsize);
+    return errbuf;
+}
+
+APU_DECLARE(apr_status_t) apr_xml_parse_file(apr_pool_t *p,
+                                             apr_xml_parser **parser,
+                                             apr_xml_doc **ppdoc,
+                                             apr_file_t *xmlfd,
+                                             apr_size_t buffer_length)
+{
+    apr_status_t rv;
+    char *buffer;
+    apr_size_t length;
+
+    *parser = apr_xml_parser_create(p);
+    if (*parser == NULL) {
+        /* FIXME: returning an error code would be nice,
+         * but we dont get one ;( */
+        return APR_EGENERAL;
+    }
+    buffer = apr_palloc(p, buffer_length);
+    length = buffer_length;
+
+    rv = apr_file_read(xmlfd, buffer, &length);
+
+    while (rv == APR_SUCCESS) {
+        rv = apr_xml_parser_feed(*parser, buffer, length);
+        if (rv != APR_SUCCESS) {
+            return rv;
+        }
+
+        length = buffer_length;
+        rv = apr_file_read(xmlfd, buffer, &length);
+    }
+    if (rv != APR_EOF) {
+        return rv;
+    }
+    rv = apr_xml_parser_done(*parser, ppdoc);
+    *parser = NULL;
+    return rv;
+}
+
+APU_DECLARE(void) apr_text_append(apr_pool_t * p, apr_text_header *hdr,
+                                  const char *text)
+{
+    apr_text *t = apr_palloc(p, sizeof(*t));
+
+    t->text = text;
+    t->next = NULL;
+
+    if (hdr->first == NULL) {
+	/* no text elements yet */
+	hdr->first = hdr->last = t;
+    }
+    else {
+	/* append to the last text element */
+	hdr->last->next = t;
+	hdr->last = t;
+    }
+}
+
+
+/* ---------------------------------------------------------------
+**
+** XML UTILITY FUNCTIONS
+*/
+
+/*
+** apr_xml_quote_string: quote an XML string
+**
+** Replace '<', '>', and '&' with '&lt;', '&gt;', and '&amp;'.
+** If quotes is true, then replace '"' with '&quot;'.
+**
+** quotes is typically set to true for XML strings that will occur within
+** double quotes -- attribute values.
+*/
+APU_DECLARE(const char *) apr_xml_quote_string(apr_pool_t *p, const char *s,
+                                               int quotes)
+{
+    const char *scan;
+    apr_size_t len = 0;
+    apr_size_t extra = 0;
+    char *qstr;
+    char *qscan;
+    char c;
+
+    for (scan = s; (c = *scan) != '\0'; ++scan, ++len) {
+	if (c == '<' || c == '>')
+	    extra += 3;		/* &lt; or &gt; */
+	else if (c == '&')
+	    extra += 4;		/* &amp; */
+	else if (quotes && c == '"')
+	    extra += 5;		/* &quot; */
+    }
+
+    /* nothing to do? */
+    if (extra == 0)
+	return s;
+
+    qstr = apr_palloc(p, len + extra + 1);
+    for (scan = s, qscan = qstr; (c = *scan) != '\0'; ++scan) {
+	if (c == '<') {
+	    *qscan++ = '&';
+	    *qscan++ = 'l';
+	    *qscan++ = 't';
+	    *qscan++ = ';';
+	}
+	else if (c == '>') {
+	    *qscan++ = '&';
+	    *qscan++ = 'g';
+	    *qscan++ = 't';
+	    *qscan++ = ';';
+	}
+	else if (c == '&') {
+	    *qscan++ = '&';
+	    *qscan++ = 'a';
+	    *qscan++ = 'm';
+	    *qscan++ = 'p';
+	    *qscan++ = ';';
+	}
+	else if (quotes && c == '"') {
+	    *qscan++ = '&';
+	    *qscan++ = 'q';
+	    *qscan++ = 'u';
+	    *qscan++ = 'o';
+	    *qscan++ = 't';
+	    *qscan++ = ';';
+	}
+	else {
+	    *qscan++ = c;
+	}
+    }
+
+    *qscan = '\0';
+    return qstr;
+}
+
+/* how many characters for the given integer? */
+#define APR_XML_NS_LEN(ns) ((ns) < 10 ? 1 : (ns) < 100 ? 2 : (ns) < 1000 ? 3 : \
+                            (ns) < 10000 ? 4 : (ns) < 100000 ? 5 : \
+                            (ns) < 1000000 ? 6 : (ns) < 10000000 ? 7 : \
+                            (ns) < 100000000 ? 8 : (ns) < 1000000000 ? 9 : 10)
+
+static apr_size_t text_size(const apr_text *t)
+{
+    apr_size_t size = 0;
+
+    for (; t; t = t->next)
+	size += strlen(t->text);
+    return size;
+}
+
+static apr_size_t elem_size(const apr_xml_elem *elem, int style,
+                            apr_array_header_t *namespaces, int *ns_map)
+{
+    apr_size_t size;
+
+    if (style == APR_XML_X2T_FULL || style == APR_XML_X2T_FULL_NS_LANG) {
+	const apr_xml_attr *attr;
+
+	size = 0;
+
+	if (style == APR_XML_X2T_FULL_NS_LANG) {
+	    int i;
+
+	    /*
+	    ** The outer element will contain xmlns:ns%d="%s" attributes
+	    ** and an xml:lang attribute, if applicable.
+	    */
+
+	    for (i = namespaces->nelts; i--;) {
+		/* compute size of: ' xmlns:ns%d="%s"' */
+		size += (9 + APR_XML_NS_LEN(i) + 2 +
+			 strlen(APR_XML_GET_URI_ITEM(namespaces, i)) + 1);
+	    }
+
+	    if (elem->lang != NULL) {
+		/* compute size of: ' xml:lang="%s"' */
+		size += 11 + strlen(elem->lang) + 1;
+	    }
+	}
+
+	if (elem->ns == APR_XML_NS_NONE) {
+	    /* compute size of: <%s> */
+	    size += 1 + strlen(elem->name) + 1;
+	}
+	else {
+	    int ns = ns_map ? ns_map[elem->ns] : elem->ns;
+
+	    /* compute size of: <ns%d:%s> */
+	    size += 3 + APR_XML_NS_LEN(ns) + 1 + strlen(elem->name) + 1;
+	}
+
+	if (APR_XML_ELEM_IS_EMPTY(elem)) {
+	    /* insert a closing "/" */
+	    size += 1;
+	}
+	else {
+	    /*
+	     * two of above plus "/":
+	     *     <ns%d:%s> ... </ns%d:%s>
+	     * OR  <%s> ... </%s>
+	     */
+	    size = 2 * size + 1;
+	}
+
+	for (attr = elem->attr; attr; attr = attr->next) {
+	    if (attr->ns == APR_XML_NS_NONE) {
+		/* compute size of: ' %s="%s"' */
+		size += 1 + strlen(attr->name) + 2 + strlen(attr->value) + 1;
+	    }
+	    else {
+		/* compute size of: ' ns%d:%s="%s"' */
+                int ns = ns_map ? ns_map[attr->ns] : attr->ns;
+                size += 3 + APR_XML_NS_LEN(ns) + 1 + strlen(attr->name) + 2 + strlen(attr->value) + 1;
+	    }
+	}
+
+	/*
+	** If the element has an xml:lang value that is *different* from
+	** its parent, then add the thing in: ' xml:lang="%s"'.
+	**
+	** NOTE: we take advantage of the pointer equality established by
+	** the parsing for "inheriting" the xml:lang values from parents.
+	*/
+	if (elem->lang != NULL &&
+	    (elem->parent == NULL || elem->lang != elem->parent->lang)) {
+	    size += 11 + strlen(elem->lang) + 1;
+	}
+    }
+    else if (style == APR_XML_X2T_LANG_INNER) {
+	/*
+	 * This style prepends the xml:lang value plus a null terminator.
+	 * If a lang value is not present, then we insert a null term.
+	 */
+	size = elem->lang ? strlen(elem->lang) + 1 : 1;
+    }
+    else
+	size = 0;
+
+    size += text_size(elem->first_cdata.first);
+
+    for (elem = elem->first_child; elem; elem = elem->next) {
+	/* the size of the child element plus the CDATA that follows it */
+	size += (elem_size(elem, APR_XML_X2T_FULL, NULL, ns_map) +
+		 text_size(elem->following_cdata.first));
+    }
+
+    return size;
+}
+
+static char *write_text(char *s, const apr_text *t)
+{
+    for (; t; t = t->next) {
+	apr_size_t len = strlen(t->text);
+	memcpy(s, t->text, len);
+	s += len;
+    }
+    return s;
+}
+
+static char *write_elem(char *s, const apr_xml_elem *elem, int style,
+			apr_array_header_t *namespaces, int *ns_map)
+{
+    const apr_xml_elem *child;
+    apr_size_t len;
+    int ns;
+
+    if (style == APR_XML_X2T_FULL || style == APR_XML_X2T_FULL_NS_LANG) {
+	int empty = APR_XML_ELEM_IS_EMPTY(elem);
+	const apr_xml_attr *attr;
+
+	if (elem->ns == APR_XML_NS_NONE) {
+	    len = sprintf(s, "<%s", elem->name);
+	}
+	else {
+	    ns = ns_map ? ns_map[elem->ns] : elem->ns;
+	    len = sprintf(s, "<ns%d:%s", ns, elem->name);
+	}
+	s += len;
+
+	for (attr = elem->attr; attr; attr = attr->next) {
+	    if (attr->ns == APR_XML_NS_NONE)
+		len = sprintf(s, " %s=\"%s\"", attr->name, attr->value);
+            else {
+                ns = ns_map ? ns_map[attr->ns] : attr->ns;
+                len = sprintf(s, " ns%d:%s=\"%s\"", ns, attr->name, attr->value);
+            }
+	    s += len;
+	}
+
+	/* add the xml:lang value if necessary */
+	if (elem->lang != NULL &&
+	    (style == APR_XML_X2T_FULL_NS_LANG ||
+	     elem->parent == NULL ||
+	     elem->lang != elem->parent->lang)) {
+	    len = sprintf(s, " xml:lang=\"%s\"", elem->lang);
+	    s += len;
+	}
+
+	/* add namespace definitions, if required */
+	if (style == APR_XML_X2T_FULL_NS_LANG) {
+	    int i;
+
+	    for (i = namespaces->nelts; i--;) {
+		len = sprintf(s, " xmlns:ns%d=\"%s\"", i,
+			      APR_XML_GET_URI_ITEM(namespaces, i));
+		s += len;
+	    }
+	}
+
+	/* no more to do. close it up and go. */
+	if (empty) {
+	    *s++ = '/';
+	    *s++ = '>';
+	    return s;
+	}
+
+	/* just close it */
+	*s++ = '>';
+    }
+    else if (style == APR_XML_X2T_LANG_INNER) {
+	/* prepend the xml:lang value */
+	if (elem->lang != NULL) {
+	    len = strlen(elem->lang);
+	    memcpy(s, elem->lang, len);
+	    s += len;
+	}
+	*s++ = '\0';
+    }
+
+    s = write_text(s, elem->first_cdata.first);
+
+    for (child = elem->first_child; child; child = child->next) {
+	s = write_elem(s, child, APR_XML_X2T_FULL, NULL, ns_map);
+	s = write_text(s, child->following_cdata.first);
+    }
+
+    if (style == APR_XML_X2T_FULL || style == APR_XML_X2T_FULL_NS_LANG) {
+	if (elem->ns == APR_XML_NS_NONE) {
+	    len = sprintf(s, "</%s>", elem->name);
+	}
+	else {
+	    ns = ns_map ? ns_map[elem->ns] : elem->ns;
+	    len = sprintf(s, "</ns%d:%s>", ns, elem->name);
+	}
+	s += len;
+    }
+
+    return s;
+}
+
+APU_DECLARE(void) apr_xml_quote_elem(apr_pool_t *p, apr_xml_elem *elem)
+{
+    apr_text *scan_txt;
+    apr_xml_attr *scan_attr;
+    apr_xml_elem *scan_elem;
+
+    /* convert the element's text */
+    for (scan_txt = elem->first_cdata.first;
+	 scan_txt != NULL;
+	 scan_txt = scan_txt->next) {
+	scan_txt->text = apr_xml_quote_string(p, scan_txt->text, 0);
+    }
+    for (scan_txt = elem->following_cdata.first;
+	 scan_txt != NULL;
+	 scan_txt = scan_txt->next) {
+	scan_txt->text = apr_xml_quote_string(p, scan_txt->text, 0);
+    }
+
+    /* convert the attribute values */
+    for (scan_attr = elem->attr;
+	 scan_attr != NULL;
+	 scan_attr = scan_attr->next) {
+	scan_attr->value = apr_xml_quote_string(p, scan_attr->value, 1);
+    }
+
+    /* convert the child elements */
+    for (scan_elem = elem->first_child;
+	 scan_elem != NULL;
+	 scan_elem = scan_elem->next) {
+	apr_xml_quote_elem(p, scan_elem);
+    }
+}
+
+/* convert an element to a text string */
+APU_DECLARE(void) apr_xml_to_text(apr_pool_t * p, const apr_xml_elem *elem,
+                                  int style, apr_array_header_t *namespaces,
+                                  int *ns_map, const char **pbuf,
+                                  apr_size_t *psize)
+{
+    /* get the exact size, plus a null terminator */
+    apr_size_t size = elem_size(elem, style, namespaces, ns_map) + 1;
+    char *s = apr_palloc(p, size);
+
+    (void) write_elem(s, elem, style, namespaces, ns_map);
+    s[size - 1] = '\0';
+
+    *pbuf = s;
+    if (psize)
+	*psize = size;
+}
+
+APU_DECLARE(const char *) apr_xml_empty_elem(apr_pool_t * p,
+                                             const apr_xml_elem *elem)
+{
+    if (elem->ns == APR_XML_NS_NONE) {
+	/*
+	 * The prefix (xml...) is already within the prop name, or
+	 * the element simply has no prefix.
+	 */
+	return apr_psprintf(p, "<%s/>" DEBUG_CR, elem->name);
+    }
+
+    return apr_psprintf(p, "<ns%d:%s/>" DEBUG_CR, elem->ns, elem->name);
+}
+
+/* return the URI's (existing) index, or insert it and return a new index */
+APU_DECLARE(int) apr_xml_insert_uri(apr_array_header_t *uri_array,
+                                    const char *uri)
+{
+    int i;
+    const char **pelt;
+
+    /* never insert an empty URI; this index is always APR_XML_NS_NONE */
+    if (*uri == '\0')
+        return APR_XML_NS_NONE;  
+
+    for (i = uri_array->nelts; i--;) {
+	if (strcmp(uri, APR_XML_GET_URI_ITEM(uri_array, i)) == 0)
+	    return i;
+    }
+
+    pelt = apr_array_push(uri_array);
+    *pelt = uri;		/* assume uri is const or in a pool */
+    return uri_array->nelts - 1;
+}
+
+/* convert the element to EBCDIC */
+#if APR_CHARSET_EBCDIC
+static apr_status_t apr_xml_parser_convert_elem(apr_xml_elem *e,
+                                                apr_xlate_t *convset)
+{
+    apr_xml_attr *a;
+    apr_xml_elem *ec;
+    apr_text *t;
+    apr_size_t inbytes_left, outbytes_left;
+    apr_status_t status;
+
+    inbytes_left = outbytes_left = strlen(e->name);
+    status = apr_xlate_conv_buffer(convset, e->name,  &inbytes_left, (char *) e->name, &outbytes_left);
+    if (status) {
+        return status;
+    }
+
+    for (t = e->first_cdata.first; t != NULL; t = t->next) {
+        inbytes_left = outbytes_left = strlen(t->text);
+        status = apr_xlate_conv_buffer(convset, t->text, &inbytes_left, (char *) t->text, &outbytes_left);
+        if (status) {
+            return status;
+        }
+    }
+
+    for (t = e->following_cdata.first;  t != NULL; t = t->next) {
+        inbytes_left = outbytes_left = strlen(t->text);
+        status = apr_xlate_conv_buffer(convset, t->text, &inbytes_left, (char *) t->text, &outbytes_left);
+        if (status) {
+            return status;
+        }
+    }
+
+    for (a = e->attr; a != NULL; a = a->next) {
+        inbytes_left = outbytes_left = strlen(a->name);
+        status = apr_xlate_conv_buffer(convset, a->name, &inbytes_left, (char *) a->name, &outbytes_left);
+        if (status) {
+            return status;
+        }
+        inbytes_left = outbytes_left = strlen(a->value);
+        status = apr_xlate_conv_buffer(convset, a->value, &inbytes_left, (char *) a->value, &outbytes_left);
+        if (status) {
+            return status;
+        }
+    }
+
+    for (ec = e->first_child; ec != NULL; ec = ec->next) {
+        status = apr_xml_parser_convert_elem(ec, convset);
+        if (status) {
+            return status;
+        }
+    }
+    return APR_SUCCESS;
+}
+
+/* convert the whole document to EBCDIC */
+APU_DECLARE(apr_status_t) apr_xml_parser_convert_doc(apr_pool_t *pool,
+                                                     apr_xml_doc *pdoc,
+                                                     apr_xlate_t *convset)
+{
+    apr_status_t status;
+    /* Don't convert the namespaces: they are constant! */
+    if (pdoc->namespaces != NULL) {
+        int i;
+        apr_array_header_t *namespaces;
+        namespaces = apr_array_make(pool, pdoc->namespaces->nelts, sizeof(const char *));
+        if (namespaces == NULL)
+            return APR_ENOMEM;
+        for (i = 0; i < pdoc->namespaces->nelts; i++) {
+            apr_size_t inbytes_left, outbytes_left;
+            char *ptr = (char *) APR_XML_GET_URI_ITEM(pdoc->namespaces, i);
+            ptr = apr_pstrdup(pool, ptr);
+            if ( ptr == NULL)
+                return APR_ENOMEM;
+            inbytes_left = outbytes_left = strlen(ptr);
+            status = apr_xlate_conv_buffer(convset, ptr, &inbytes_left, ptr, &outbytes_left);
+            if (status) {
+                return status;
+            }
+            apr_xml_insert_uri(namespaces, ptr);
+        }
+        pdoc->namespaces = namespaces;
+    }
+    return apr_xml_parser_convert_elem(pdoc->root, convset);
+}
+#endif
diff --git a/xml/expat/COPYING b/xml/expat/COPYING
new file mode 100644
index 000000000..fc97b02d9
--- /dev/null
+++ b/xml/expat/COPYING
@@ -0,0 +1,21 @@
+Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+                               and Clark Cooper
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be included
+in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
diff --git a/xml/expat/Makefile.in b/xml/expat/Makefile.in
new file mode 100644
index 000000000..bbf0a9f71
--- /dev/null
+++ b/xml/expat/Makefile.in
@@ -0,0 +1,156 @@
+################################################################
+# Process this file with top-level configure script to produce Makefile
+#
+# Copyright 2000 Clark Cooper
+#
+#  This file is part of EXPAT.
+#
+#  EXPAT is free software; you can redistribute it and/or modify it
+#  under the terms of the License (based on the MIT/X license) contained
+#  in the file COPYING that comes with this distribution.
+#
+# EXPAT IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+# SOFTWARE OR THE USE OR OTHER DEALINGS IN EXPAT.
+#
+#  ---
+#  I started using automake, but
+#		1) it seemed like overkill
+#		2) I don't want all the GNU policies
+#		3) I wanted more explicit control over what gets built
+#
+#  So I'm doing my Makefile.in files manually. But a fair part is based
+#  on what I learned from perusing the Makefile.in's generated by automake,
+#  and the automake authors still get my kudos.
+#
+
+SHELL = @SHELL@
+
+srcdir = @srcdir@
+top_srcdir = @top_srcdir@
+VPATH = @srcdir@
+prefix = @prefix@
+exec_prefix = @exec_prefix@
+
+bindir = @bindir@
+sbindir = @sbindir@
+libexecdir = @libexecdir@
+datadir = @datadir@
+sysconfdir = @sysconfdir@
+sharedstatedir = @sharedstatedir@
+localstatedir = @localstatedir@
+libdir = @libdir@
+infodir = @infodir@
+mandir = @mandir@
+includedir = @includedir@
+oldincludedir = /usr/include
+
+top_builddir = .
+
+
+AUTOCONF = autoconf
+
+INSTALL = @INSTALL@
+INSTALL_PROGRAM = @INSTALL_PROGRAM@
+INSTALL_DATA = @INSTALL_DATA@
+INSTALL_SCRIPT = @INSTALL_SCRIPT@
+
+NORMAL_INSTALL = :
+PRE_INSTALL = :
+POST_INSTALL = :
+NORMAL_UNINSTALL = :
+PRE_UNINSTALL = :
+POST_UNINSTALL = :
+host_alias = @host_alias@
+host_triplet = @host@
+
+CC = @CC@
+
+LIBTOOL = @LIBTOOL@
+LN_S = @LN_S@
+PACKAGE = @PACKAGE@
+RANLIB = @RANLIB@
+VERSION = @VERSION@
+
+SUBDIRS = lib
+ACLOCAL_M4 = $(top_srcdir)/aclocal.m4
+CONFIG_HEADERS = config.h
+
+DISTDIR = $(PACKAGE)-$(VERSION)
+DISTRIBUTION = $(DISTDIR).tar.gz
+
+all: build-subdirs
+
+.PHONY: all build-subdirs clean distclean extraclean maintainer-clean dist install \
+        uninstall distdir
+
+Makefile: Makefile.in config.status
+	CONFIG_FILES=$@ CONFIG_HEADERS= $(SHELL) config.status
+
+config.status: configure
+	@if test -f $@; then \
+		$(SHELL) config.status --recheck ; \
+	else \
+		$(SHELL) configure ; \
+	fi
+
+configure: configure.in
+	$(AUTOCONF)
+
+config.h: config.h.in config.status
+	CONFIG_FILES= CONFIG_HEADERS=$(CONFIG_HEADERS) \
+	$(SHELL) ./config.status
+
+build-subdirs:
+	@list='$(SUBDIRS)'; \
+	for dir in $$list; do \
+	  cd $$dir; $(MAKE); cd ..; \
+	done
+
+clean:
+	@list='$(SUBDIRS)'; for dir in $$list; do \
+	  cd $$dir; $(MAKE) clean; cd ..; \
+	done
+	rm -f core *~
+
+distclean:
+	@list='$(SUBDIRS)'; for dir in $$list; do \
+	  cd $$dir; $(MAKE) distclean; cd ..; \
+	done
+	rm -f config.h config.status config.log libtool examples/Makefile xmlwf/Makefile Makefile
+
+extraclean: distclean
+	rm -f configure aclocal.m4
+
+maintainer-clean: distclean
+	rm -f $(DISTRIBUTION)
+	rm -rf $(DISTDIR)
+
+distdir: MANIFEST
+	test -d $(DISTDIR) && rm -rf $(DISTDIR); \
+	mkdir $(DISTDIR); \
+	flist=`sed -e "s/[ 	]:.*$$//" MANIFEST`; for file in $$flist; do \
+	  cp -P $$file $(DISTDIR); \
+	done
+
+$(DISTRIBUTION): distdir
+	tar cfz $(DISTRIBUTION) $(DISTDIR)	
+
+dist: $(DISTRIBUTION)
+
+install:
+	@list='$(SUBDIRS)'; for dir in $$list; do \
+	  cd $$dir; $(MAKE) install; cd ..; \
+	done
+
+uninstall:
+	@list='$(SUBDIRS)'; for dir in $$list; do \
+	  cd $$dir; $(MAKE) uninstall; cd ..; \
+	done
+
+depend:
+	echo SOMEONE SHOULD MAKE THIS DO SOMETHING!!!
diff --git a/xml/expat/README b/xml/expat/README
new file mode 100644
index 000000000..15bcdccb5
--- /dev/null
+++ b/xml/expat/README
@@ -0,0 +1,72 @@
+
+			Expat, Release 1.95.2
+
+This is expat, a C library for parsing XML, written by James Clark.
+Expat is a stream-oriented XML parser.  This means that you register
+handlers with the parser before starting the parse.  These handlers
+are called when the parser discovers the associated structures in the
+document being parsed.  A start tag is an example of the kind of
+structures for which you may register handlers.
+
+Windows users should use the expat_win32bin package, which includes
+both precompiled libraries and executalbes, and source code for
+developers.
+
+Expat is free software.  You may copy, distribute, and modify it under
+the terms of the License contained in the file COPYING distributed
+with this package.  This license is the same as the MIT/X Consortium
+license.
+
+Versions of expat that have an odd minor version (the middle number in
+the release above), are development releases and should be considered
+as beta software.  Releases with even minor version numbers are
+intended to be production grade software.
+
+To build expat, you first run the configuration shell script in the
+top level distribution directory:
+
+	./configure
+
+There are many options which you may provide to configure (which you
+can discover by running configure with the --help option).  But the
+one of most interest is the one that sets the installation directory.
+By default, the configure script will set things up to install
+libexpat into /usr/local/lib, expat.h into /usr/local/include, and
+xmlwf into /usr/local/bin.  If, for example, you'd prefer to install
+into /home/me/mystuff/lib, /home/me/mystuff/include, and
+/home/me/mystuff/bin, you can tell configure about that with:
+
+	./configure --prefix=/home/me/mystuff
+
+After running the configure script, the "make" command will build
+things and "make install" will install things into their proper
+location.  Note that you need to have write permission into the
+directories into which things will be installed.
+
+When building for use with C++, you may need to add additional
+compiler flags to support proper interaction with exceptions.  This
+can be done by setting the CFLAGS environment variable.  For example,
+when using GCC, you can use:
+
+	CFLAGS=-fexceptions ./configure
+
+Note for Solaris users:  The "ar" command is usually located in
+"/usr/ccs/bin", which is not in the default PATH.  You will need to
+add this to your path for the "make" command, and probably also switch
+to GNU make (the "make" found in /usr/ccs/bin does not seem to work
+properly -- appearantly it does not understand .PHONY directives).  If
+you're using ksh or bash, use this command to build:
+
+	PATH=/usr/ccs/bin:$PATH make
+
+A reference manual is available in the file doc/reference.html in this
+distribution.
+
+The homepage for this project is http://expat.sourceforge.net/.  There
+are links there to connect you to the bug reports page.  If you need
+to report a bug when you don't have access to a browser, you may also
+send a bug report by email to expat-bugs@lists.sourceforge.net.
+
+Discussion related to the direction of future expat development takes
+place on expat-discuss@lists.sourceforge.net.  Archives of this list
+may be found at http://www.geocrawler.com/redir-sf.php3?list=expat-discuss.
diff --git a/xml/expat/acconfig.h b/xml/expat/acconfig.h
new file mode 100644
index 000000000..1283ad61b
--- /dev/null
+++ b/xml/expat/acconfig.h
@@ -0,0 +1,23 @@
+/* This file is used by autoheader to add items to expat_config.h.in */
+
+#ifdef WORDS_BIGENDIAN
+#define XML_BYTE_ORDER 21
+#else
+#define XML_BYTE_ORDER 12
+#endif
+
+@BOTTOM@
+
+#define XML_NS
+#define XML_DTD
+
+#define XML_CONTEXT_BYTES 1024
+
+#ifndef HAVE_MEMMOVE
+#ifdef HAVE_BCOPY
+#define memmove(d,s,l) bcopy((s),(d),(l))
+#else
+#define memmove(d,s,l) ;punting on memmove;
+#endif
+
+#endif
diff --git a/xml/expat/buildconf.sh b/xml/expat/buildconf.sh
new file mode 100755
index 000000000..465a32523
--- /dev/null
+++ b/xml/expat/buildconf.sh
@@ -0,0 +1,67 @@
+#! /bin/sh
+
+#
+# Find libtoolize
+#
+libtoolize=`conftools/PrintPath glibtoolize1 glibtoolize libtoolize libtoolize15 libtoolize14`
+if [ "x$libtoolize" = "x" ]; then
+    echo "libtoolize not found in path"
+    exit 1
+fi
+
+#
+# Create the libtool helper files
+#
+# Note: we copy (rather than link) the files.
+#
+# Note: This bundled version of expat will not always replace the
+# files since we have a special config.guess/config.sub that we
+# want to ensure is used.
+echo "Copying libtool helper files ..."
+
+# Remove any m4 cache and libtool files so one can switch between 
+# autoconf and libtool versions by simply rerunning the buildconf script.
+#
+(cd conftools ; rm -f ltconfig ltmain.sh)
+rm -rf aclocal.m4 libtool.m4 ltsugar.m4 autom4te*.cache
+
+$libtoolize --copy --automake
+
+#
+# Build aclocal.m4 from libtool's libtool.m4
+#
+if [ -f libtool.m4 ]; then
+  ltfile=libtool.m4
+else
+  ltpath=`dirname $libtoolize`
+  ltfile=${LIBTOOL_M4-`cd $ltpath/../share/aclocal ; pwd`/libtool.m4}
+fi
+echo "Incorporating $ltfile into aclocal.m4 ..."
+echo "dnl THIS FILE IS AUTOMATICALLY GENERATED BY buildconf.sh" > aclocal.m4
+echo "dnl edits here will be lost" >> aclocal.m4
+cat $ltfile >> aclocal.m4
+
+if [ -f ltsugar.m4 ]; then
+  echo "Incorporating ltsugar.m4 into aclocal.m4 ..."
+  cat ltsugar.m4 >> aclocal.m4
+fi
+
+# Clean up again
+rm -f libtool.m4 ltsugar.m4
+
+cross_compile_warning="warning: AC_TRY_RUN called without default to allow cross compiling"
+
+#
+# Generate the autoconf header template (config.h.in) and ./configure
+#
+echo "Creating config.h.in ..."
+${AUTOHEADER:-autoheader} 2>&1 | grep -v "$cross_compile_warning"
+
+echo "Creating configure ..."
+### do some work to toss config.cache?
+${AUTOCONF:-autoconf} 2>&1 | grep -v "$cross_compile_warning"
+
+# Remove autoconf caches
+rm -rf autom4te*.cache
+
+exit 0
diff --git a/xml/expat/configure.in b/xml/expat/configure.in
new file mode 100644
index 000000000..04cbf083b
--- /dev/null
+++ b/xml/expat/configure.in
@@ -0,0 +1,109 @@
+dnl   configuration script for expat
+dnl   Process this file with autoconf to produce a configure script.
+dnl
+dnl   Copyright 2000 Clark Cooper
+dnl
+dnl   This file is part of EXPAT.
+dnl
+dnl   EXPAT is free software; you can redistribute it and/or modify it
+dnl   under the terms of the License (based on the MIT/X license) contained
+dnl   in the file COPYING that comes with this distribution.
+dnl
+
+AC_INIT(Makefile.in)
+AC_CONFIG_AUX_DIR(conftools)
+
+dnl
+dnl Follow the GNU/Linux convention of odd number minor version for
+dnl beta/development releases and even number minor version for stable
+dnl releases. Edit is bumped with each release and set to 0 with
+dnl change to major or minor version.
+dnl
+
+EXPAT_MAJOR_VERSION=1
+EXPAT_MINOR_VERSION=95
+EXPAT_EDIT=2
+
+EXPAT_VERSION=$EXPAT_MAJOR_VERSION.$EXPAT_MINOR_VERSION.$EXPAT_EDIT
+VERSION=$EXPAT_VERSION
+PACKAGE=expat
+
+dnl
+dnl Increment LIBREVISION if source code has changed at all
+dnl
+dnl If the API has changed, increment LIBCURRENT and set LIBREVISION to 0
+dnl
+dnl If the API changes compatibly (i.e. simply adding a new function
+dnl without changing or removing earlier interfaces), then increment LIBAGE.
+dnl 
+dnl If the API changes incompatibly set LIBAGE back to 0
+dnl
+
+LIBCURRENT=1
+LIBREVISION=0
+LIBAGE=1
+
+AC_CONFIG_HEADER(config.h)
+
+AC_CANONICAL_SYSTEM
+case "$host_os" in
+*os2*)
+    # Use a custom made libtool replacement
+    echo Using aplibtool
+    LIBTOOL="$srcdir/../../../apr/build/aplibtool"
+    ;;
+*)
+    AC_LIBTOOL_WIN32_DLL
+AC_PROG_LIBTOOL
+    ;;
+esac
+
+blddir=`pwd`
+AC_SUBST(blddir)
+
+AC_SUBST(PACKAGE)
+AC_SUBST(VERSION)
+AC_SUBST(EXPAT_MAJOR_VERSION)
+AC_SUBST(EXPAT_MINOR_VERSION)
+AC_SUBST(EXPAT_EDIT)
+
+AC_SUBST(LIBCURRENT)
+AC_SUBST(LIBREVISION)
+AC_SUBST(LIBAGE)
+
+dnl Checks for programs.
+AC_PROG_CC
+AC_PROG_INSTALL
+
+dnl Checks for libraries.
+
+dnl Checks for header files.
+AC_HEADER_STDC
+AC_CHECK_HEADERS(fcntl.h unistd.h string.h)
+
+dnl Checks for typedefs, structures, and compiler characteristics.
+dnl check for endianness
+if test "$cross_compiling" = "no"; then
+  AC_C_BIGENDIAN
+else
+  AC_DEFINE(AP_UNKNOWN_BYTE_ORDER,1,
+                [byte order is unknown due to cross-compilation])
+fi
+AC_C_CONST
+AC_TYPE_OFF_T
+AC_TYPE_SIZE_T
+
+dnl Checks for library functions.
+
+AC_FUNC_MEMCMP
+AC_FUNC_MMAP
+AC_SUBST(FILEMAP_OBJ)
+if test -z "$HAVE_MMAP"; then
+FILEMAP_OBJ=unixfilemap.o
+else
+FILEMAP_OBJ=readfilemap.o
+fi
+
+AC_CHECK_FUNCS(memmove bcopy)
+
+AC_OUTPUT(Makefile lib/Makefile lib/expat.h)
diff --git a/xml/expat/conftools/PrintPath b/xml/expat/conftools/PrintPath
new file mode 100755
index 000000000..68435f374
--- /dev/null
+++ b/xml/expat/conftools/PrintPath
@@ -0,0 +1,116 @@
+#!/bin/sh
+# Look for program[s] somewhere in $PATH.
+#
+# Options:
+#  -s
+#    Do not print out full pathname. (silent)
+#  -pPATHNAME
+#    Look in PATHNAME instead of $PATH
+#
+# Usage:
+#  PrintPath [-s] [-pPATHNAME] program [program ...]
+#
+# Initially written by Jim Jagielski for the Apache configuration mechanism
+#  (with kudos to Kernighan/Pike)
+#
+# This script falls under the Apache License.
+# See http://www.apache.org/docs/LICENSE
+
+##
+# Some "constants"
+##
+pathname=$PATH
+echo="yes"
+
+##
+# Find out what OS we are running for later on
+##
+os=`(uname) 2>/dev/null`
+
+##
+# Parse command line
+##
+for args in $*
+do
+    case $args in
+	-s  ) echo="no" ;;
+	-p* ) pathname="`echo $args | sed 's/^..//'`" ;;
+	*   ) programs="$programs $args" ;;
+    esac
+done
+
+##
+# Now we make the adjustments required for OS/2 and everyone
+# else :)
+#
+# First of all, all OS/2 programs have the '.exe' extension.
+# Next, we adjust PATH (or what was given to us as PATH) to
+# be whitespace seperated directories.
+# Finally, we try to determine the best flag to use for
+# test/[] to look for an executable file. OS/2 just has '-r'
+# but with other OSs, we do some funny stuff to check to see
+# if test/[] knows about -x, which is the prefered flag.
+##
+
+if [ "x$os" = "xOS/2" ]
+then
+    ext=".exe"
+    pathname=`echo -E $pathname |
+     sed 's/^;/.;/
+	  s/;;/;.;/g
+	  s/;$/;./
+	  s/;/ /g
+	  s/\\\\/\\//g' `
+    test_exec_flag="-r"
+else
+    ext=""	# No default extensions
+    pathname=`echo $pathname |
+     sed 's/^:/.:/
+	  s/::/:.:/g
+	  s/:$/:./
+	  s/:/ /g' `
+    # Here is how we test to see if test/[] can handle -x
+    testfile="pp.t.$$"
+
+    cat > $testfile <<ENDTEST
+#!/bin/sh
+if [ -x / ] || [ -x /bin ] || [ -x /bin/ls ]; then
+    exit 0
+fi
+exit 1
+ENDTEST
+
+    if `/bin/sh $testfile 2>/dev/null`; then
+	test_exec_flag="-x"
+    else
+	test_exec_flag="-r"
+    fi
+    rm -f $testfile
+fi
+
+for program in $programs
+do
+    for path in $pathname
+    do
+	if [ $test_exec_flag $path/${program}${ext} ] && \
+	   [ ! -d $path/${program}${ext} ]; then
+	    if [ "x$echo" = "xyes" ]; then
+		echo $path/${program}${ext}
+	    fi
+	    exit 0
+	fi
+
+# Next try without extension (if one was used above)
+	if [ "x$ext" != "x" ]; then
+            if [ $test_exec_flag $path/${program} ] && \
+               [ ! -d $path/${program} ]; then
+                if [ "x$echo" = "xyes" ]; then
+                    echo $path/${program}
+                fi
+                exit 0
+            fi
+        fi
+    done
+done
+exit 1
+
diff --git a/xml/expat/conftools/config.guess b/xml/expat/conftools/config.guess
new file mode 100755
index 000000000..6cfac28c8
--- /dev/null
+++ b/xml/expat/conftools/config.guess
@@ -0,0 +1,1344 @@
+#! /bin/sh
+# Attempt to guess a canonical system name.
+#   Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
+#   2000, 2001, 2002 Free Software Foundation, Inc.
+
+timestamp='2002-03-20'
+
+# This file is free software; you can redistribute it and/or modify it
+# under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+# General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+#
+# As a special exception to the GNU General Public License, if you
+# distribute this file as part of a program that contains a
+# configuration script generated by Autoconf, you may include it under
+# the same distribution terms that you use for the rest of that program.
+
+#####################################################################
+# This file contains changes for Apache, clearly marked below.
+# These changes are hereby donated to the public domain.
+#####################################################################
+
+# Originally written by Per Bothner <per@bothner.com>.
+# Please send patches to <config-patches@gnu.org>.  Submit a context
+# diff and a properly formatted ChangeLog entry.
+#
+# This script attempts to guess a canonical system name similar to
+# config.sub.  If it succeeds, it prints the system name on stdout, and
+# exits with 0.  Otherwise, it exits with 1.
+#
+# The plan is that this can be called by configure scripts if you
+# don't specify an explicit build system type.
+
+me=`echo "$0" | sed -e 's,.*/,,'`
+
+usage="\
+Usage: $0 [OPTION]
+
+Output the configuration name of the system \`$me' is run on.
+
+Operation modes:
+  -h, --help         print this help, then exit
+  -t, --time-stamp   print date of last modification, then exit
+  -v, --version      print version number, then exit
+
+Report bugs and patches to <config-patches@gnu.org>."
+
+version="\
+GNU config.guess ($timestamp)
+
+Originally written by Per Bothner.
+Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001
+Free Software Foundation, Inc.
+
+This is free software; see the source for copying conditions.  There is NO
+warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
+
+help="
+Try \`$me --help' for more information."
+
+# Parse command line
+while test $# -gt 0 ; do
+  case $1 in
+    --time-stamp | --time* | -t )
+       echo "$timestamp" ; exit 0 ;;
+    --version | -v )
+       echo "$version" ; exit 0 ;;
+    --help | --h* | -h )
+       echo "$usage"; exit 0 ;;
+    -- )     # Stop option processing
+       shift; break ;;
+    - )	# Use stdin as input.
+       break ;;
+    -* )
+       echo "$me: invalid option $1$help" >&2
+       exit 1 ;;
+    * )
+       break ;;
+  esac
+done
+
+if test $# != 0; then
+  echo "$me: too many arguments$help" >&2
+  exit 1
+fi
+
+
+dummy=dummy-$$
+trap 'rm -f $dummy.c $dummy.o $dummy.rel $dummy; exit 1' 1 2 15
+
+# CC_FOR_BUILD -- compiler used by this script.
+# Historically, `CC_FOR_BUILD' used to be named `HOST_CC'. We still
+# use `HOST_CC' if defined, but it is deprecated.
+
+set_cc_for_build='case $CC_FOR_BUILD,$HOST_CC,$CC in
+ ,,)    echo "int dummy(){}" > $dummy.c ;
+	for c in cc gcc c89 c99 ; do
+	  ($c $dummy.c -c -o $dummy.o) >/dev/null 2>&1 ;
+	  if test $? = 0 ; then
+	     CC_FOR_BUILD="$c"; break ;
+	  fi ;
+	done ;
+	rm -f $dummy.c $dummy.o $dummy.rel ;
+	if test x"$CC_FOR_BUILD" = x ; then
+	  CC_FOR_BUILD=no_compiler_found ;
+	fi
+	;;
+ ,,*)   CC_FOR_BUILD=$CC ;;
+ ,*,*)  CC_FOR_BUILD=$HOST_CC ;;
+esac'
+
+# This is needed to find uname on a Pyramid OSx when run in the BSD universe.
+# (ghazi@noc.rutgers.edu 1994-08-24)
+if (test -f /.attbin/uname) >/dev/null 2>&1 ; then
+	PATH=$PATH:/.attbin ; export PATH
+fi
+
+UNAME_MACHINE=`(uname -m) 2>/dev/null` || UNAME_MACHINE=unknown
+UNAME_RELEASE=`(uname -r) 2>/dev/null` || UNAME_RELEASE=unknown
+UNAME_SYSTEM=`(uname -s) 2>/dev/null`  || UNAME_SYSTEM=unknown
+UNAME_VERSION=`(uname -v) 2>/dev/null` || UNAME_VERSION=unknown
+
+# Note: order is significant - the case branches are not exclusive.
+
+case "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" in
+    *:NetBSD:*:*)
+	# NetBSD (nbsd) targets should (where applicable) match one or
+	# more of the tupples: *-*-netbsdelf*, *-*-netbsdaout*,
+	# *-*-netbsdecoff* and *-*-netbsd*.  For targets that recently
+	# switched to ELF, *-*-netbsd* would select the old
+	# object file format.  This provides both forward
+	# compatibility and a consistent mechanism for selecting the
+	# object file format.
+	#
+	# Note: NetBSD doesn't particularly care about the vendor
+	# portion of the name.  We always set it to "unknown".
+	sysctl="sysctl -n hw.machine_arch"
+	UNAME_MACHINE_ARCH=`(/sbin/$sysctl 2>/dev/null || \
+	    /usr/sbin/$sysctl 2>/dev/null || echo unknown)`
+	case "${UNAME_MACHINE_ARCH}" in
+	    arm*) machine=arm-unknown ;;
+	    sh3el) machine=shl-unknown ;;
+	    sh3eb) machine=sh-unknown ;;
+	    *) machine=${UNAME_MACHINE_ARCH}-unknown ;;
+	esac
+	# The Operating System including object format, if it has switched
+	# to ELF recently, or will in the future.
+	case "${UNAME_MACHINE_ARCH}" in
+	    arm*|i386|m68k|ns32k|sh3*|sparc|vax)
+		eval $set_cc_for_build
+		if echo __ELF__ | $CC_FOR_BUILD -E - 2>/dev/null \
+			| grep __ELF__ >/dev/null
+		then
+		    # Once all utilities can be ECOFF (netbsdecoff) or a.out (netbsdaout).
+		    # Return netbsd for either.  FIX?
+		    os=netbsd
+		else
+		    os=netbsdelf
+		fi
+		;;
+	    *)
+	        os=netbsd
+		;;
+	esac
+	# The OS release
+	release=`echo ${UNAME_RELEASE}|sed -e 's/[-_].*/\./'`
+	# Since CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM:
+	# contains redundant information, the shorter form:
+	# CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM is used.
+	echo "${machine}-${os}${release}"
+	exit 0 ;;
+    amiga:OpenBSD:*:*)
+	echo m68k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    arc:OpenBSD:*:*)
+	echo mipsel-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    hp300:OpenBSD:*:*)
+	echo m68k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    mac68k:OpenBSD:*:*)
+	echo m68k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    macppc:OpenBSD:*:*)
+	echo powerpc-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    mvme68k:OpenBSD:*:*)
+	echo m68k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    mvme88k:OpenBSD:*:*)
+	echo m88k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    mvmeppc:OpenBSD:*:*)
+	echo powerpc-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    pmax:OpenBSD:*:*)
+	echo mipsel-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    sgi:OpenBSD:*:*)
+	echo mipseb-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    sun3:OpenBSD:*:*)
+	echo m68k-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    wgrisc:OpenBSD:*:*)
+	echo mipsel-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    *:OpenBSD:*:*)
+	echo ${UNAME_MACHINE}-unknown-openbsd${UNAME_RELEASE}
+	exit 0 ;;
+    alpha:OSF1:*:*)
+	if test $UNAME_RELEASE = "V4.0"; then
+		UNAME_RELEASE=`/usr/sbin/sizer -v | awk '{print $3}'`
+	fi
+	# A Vn.n version is a released version.
+	# A Tn.n version is a released field test version.
+	# A Xn.n version is an unreleased experimental baselevel.
+	# 1.2 uses "1.2" for uname -r.
+	cat <<EOF >$dummy.s
+	.data
+\$Lformat:
+	.byte 37,100,45,37,120,10,0	# "%d-%x\n"
+
+	.text
+	.globl main
+	.align 4
+	.ent main
+main:
+	.frame \$30,16,\$26,0
+	ldgp \$29,0(\$27)
+	.prologue 1
+	.long 0x47e03d80 # implver \$0
+	lda \$2,-1
+	.long 0x47e20c21 # amask \$2,\$1
+	lda \$16,\$Lformat
+	mov \$0,\$17
+	not \$1,\$18
+	jsr \$26,printf
+	ldgp \$29,0(\$26)
+	mov 0,\$16
+	jsr \$26,exit
+	.end main
+EOF
+	eval $set_cc_for_build
+	$CC_FOR_BUILD $dummy.s -o $dummy 2>/dev/null
+	if test "$?" = 0 ; then
+		case `./$dummy` in
+			0-0)
+				UNAME_MACHINE="alpha"
+				;;
+			1-0)
+				UNAME_MACHINE="alphaev5"
+				;;
+			1-1)
+				UNAME_MACHINE="alphaev56"
+				;;
+			1-101)
+				UNAME_MACHINE="alphapca56"
+				;;
+			2-303)
+				UNAME_MACHINE="alphaev6"
+				;;
+			2-307)
+				UNAME_MACHINE="alphaev67"
+				;;
+			2-1307)
+				UNAME_MACHINE="alphaev68"
+				;;
+		esac
+	fi
+	rm -f $dummy.s $dummy
+	echo ${UNAME_MACHINE}-dec-osf`echo ${UNAME_RELEASE} | sed -e 's/^[VTX]//' | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
+	exit 0 ;;
+    Alpha\ *:Windows_NT*:*)
+	# How do we know it's Interix rather than the generic POSIX subsystem?
+	# Should we change UNAME_MACHINE based on the output of uname instead
+	# of the specific Alpha model?
+	echo alpha-pc-interix
+	exit 0 ;;
+    21064:Windows_NT:50:3)
+	echo alpha-dec-winnt3.5
+	exit 0 ;;
+    Amiga*:UNIX_System_V:4.0:*)
+	echo m68k-unknown-sysv4
+	exit 0;;
+    *:[Aa]miga[Oo][Ss]:*:*)
+	echo ${UNAME_MACHINE}-unknown-amigaos
+	exit 0 ;;
+    *:[Mm]orph[Oo][Ss]:*:*)
+	echo ${UNAME_MACHINE}-unknown-morphos
+	exit 0 ;;
+#########################
+# Apache changes
+#
+#   *:OS/390:*:*)
+#      echo i370-ibm-openedition
+#      exit 0 ;;
+    *:OS390:*:* | *:OS/390:*:*)
+       echo s390-ibm-os390
+       exit 0 ;; 
+    *:OS400:*:* | *:OS/400:*:*) 
+       echo as400-ibm-os400
+       exit 0 ;;
+    *:OS/2:*:*)
+       echo "i386-pc-os2_emx"
+       exit 0;;
+#
+# end Apache changes
+#########################
+    *:OS/390:*:*)
+	echo i370-ibm-openedition
+	exit 0 ;;
+    arm:RISC*:1.[012]*:*|arm:riscix:1.[012]*:*)
+	echo arm-acorn-riscix${UNAME_RELEASE}
+	exit 0;;
+    SR2?01:HI-UX/MPP:*:* | SR8000:HI-UX/MPP:*:*)
+	echo hppa1.1-hitachi-hiuxmpp
+	exit 0;;
+    Pyramid*:OSx*:*:* | MIS*:OSx*:*:* | MIS*:SMP_DC-OSx*:*:*)
+	# akee@wpdis03.wpafb.af.mil (Earle F. Ake) contributed MIS and NILE.
+	if test "`(/bin/universe) 2>/dev/null`" = att ; then
+		echo pyramid-pyramid-sysv3
+	else
+		echo pyramid-pyramid-bsd
+	fi
+	exit 0 ;;
+    NILE*:*:*:dcosx)
+	echo pyramid-pyramid-svr4
+	exit 0 ;;
+    sun4H:SunOS:5.*:*)
+	echo sparc-hal-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
+	exit 0 ;;
+    sun4*:SunOS:5.*:* | tadpole*:SunOS:5.*:*)
+	echo sparc-sun-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
+	exit 0 ;;
+    i86pc:SunOS:5.*:*)
+	echo i386-pc-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
+	exit 0 ;;
+    sun4*:SunOS:6*:*)
+	# According to config.sub, this is the proper way to canonicalize
+	# SunOS6.  Hard to guess exactly what SunOS6 will be like, but
+	# it's likely to be more like Solaris than SunOS4.
+	echo sparc-sun-solaris3`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
+	exit 0 ;;
+    sun4*:SunOS:*:*)
+	case "`/usr/bin/arch -k`" in
+	    Series*|S4*)
+		UNAME_RELEASE=`uname -v`
+		;;
+	esac
+	# Japanese Language versions have a version number like `4.1.3-JL'.
+	echo sparc-sun-sunos`echo ${UNAME_RELEASE}|sed -e 's/-/_/'`
+	exit 0 ;;
+    sun3*:SunOS:*:*)
+	echo m68k-sun-sunos${UNAME_RELEASE}
+	exit 0 ;;
+    sun*:*:4.2BSD:*)
+	UNAME_RELEASE=`(sed 1q /etc/motd | awk '{print substr($5,1,3)}') 2>/dev/null`
+	test "x${UNAME_RELEASE}" = "x" && UNAME_RELEASE=3
+	case "`/bin/arch`" in
+	    sun3)
+		echo m68k-sun-sunos${UNAME_RELEASE}
+		;;
+	    sun4)
+		echo sparc-sun-sunos${UNAME_RELEASE}
+		;;
+	esac
+	exit 0 ;;
+    aushp:SunOS:*:*)
+	echo sparc-auspex-sunos${UNAME_RELEASE}
+	exit 0 ;;
+    # The situation for MiNT is a little confusing.  The machine name
+    # can be virtually everything (everything which is not
+    # "atarist" or "atariste" at least should have a processor
+    # > m68000).  The system name ranges from "MiNT" over "FreeMiNT"
+    # to the lowercase version "mint" (or "freemint").  Finally
+    # the system name "TOS" denotes a system which is actually not
+    # MiNT.  But MiNT is downward compatible to TOS, so this should
+    # be no problem.
+    atarist[e]:*MiNT:*:* | atarist[e]:*mint:*:* | atarist[e]:*TOS:*:*)
+        echo m68k-atari-mint${UNAME_RELEASE}
+	exit 0 ;;
+    atari*:*MiNT:*:* | atari*:*mint:*:* | atarist[e]:*TOS:*:*)
+	echo m68k-atari-mint${UNAME_RELEASE}
+        exit 0 ;;
+    *falcon*:*MiNT:*:* | *falcon*:*mint:*:* | *falcon*:*TOS:*:*)
+        echo m68k-atari-mint${UNAME_RELEASE}
+	exit 0 ;;
+    milan*:*MiNT:*:* | milan*:*mint:*:* | *milan*:*TOS:*:*)
+        echo m68k-milan-mint${UNAME_RELEASE}
+        exit 0 ;;
+    hades*:*MiNT:*:* | hades*:*mint:*:* | *hades*:*TOS:*:*)
+        echo m68k-hades-mint${UNAME_RELEASE}
+        exit 0 ;;
+    *:*MiNT:*:* | *:*mint:*:* | *:*TOS:*:*)
+        echo m68k-unknown-mint${UNAME_RELEASE}
+        exit 0 ;;
+    powerpc:machten:*:*)
+	echo powerpc-apple-machten${UNAME_RELEASE}
+	exit 0 ;;
+    RISC*:Mach:*:*)
+	echo mips-dec-mach_bsd4.3
+	exit 0 ;;
+    RISC*:ULTRIX:*:*)
+	echo mips-dec-ultrix${UNAME_RELEASE}
+	exit 0 ;;
+    VAX*:ULTRIX*:*:*)
+	echo vax-dec-ultrix${UNAME_RELEASE}
+	exit 0 ;;
+    2020:CLIX:*:* | 2430:CLIX:*:*)
+	echo clipper-intergraph-clix${UNAME_RELEASE}
+	exit 0 ;;
+    mips:*:*:UMIPS | mips:*:*:RISCos)
+	eval $set_cc_for_build
+	sed 's/^	//' << EOF >$dummy.c
+#ifdef __cplusplus
+#include <stdio.h>  /* for printf() prototype */
+	int main (int argc, char *argv[]) {
+#else
+	int main (argc, argv) int argc; char *argv[]; {
+#endif
+	#if defined (host_mips) && defined (MIPSEB)
+	#if defined (SYSTYPE_SYSV)
+	  printf ("mips-mips-riscos%ssysv\n", argv[1]); exit (0);
+	#endif
+	#if defined (SYSTYPE_SVR4)
+	  printf ("mips-mips-riscos%ssvr4\n", argv[1]); exit (0);
+	#endif
+	#if defined (SYSTYPE_BSD43) || defined(SYSTYPE_BSD)
+	  printf ("mips-mips-riscos%sbsd\n", argv[1]); exit (0);
+	#endif
+	#endif
+	  exit (-1);
+	}
+EOF
+	$CC_FOR_BUILD $dummy.c -o $dummy \
+	  && ./$dummy `echo "${UNAME_RELEASE}" | sed -n 's/\([0-9]*\).*/\1/p'` \
+	  && rm -f $dummy.c $dummy && exit 0
+	rm -f $dummy.c $dummy
+	echo mips-mips-riscos${UNAME_RELEASE}
+	exit 0 ;;
+    Motorola:PowerMAX_OS:*:*)
+	echo powerpc-motorola-powermax
+	exit 0 ;;
+    Night_Hawk:Power_UNIX:*:*)
+	echo powerpc-harris-powerunix
+	exit 0 ;;
+    m88k:CX/UX:7*:*)
+	echo m88k-harris-cxux7
+	exit 0 ;;
+    m88k:*:4*:R4*)
+	echo m88k-motorola-sysv4
+	exit 0 ;;
+    m88k:*:3*:R3*)
+	echo m88k-motorola-sysv3
+	exit 0 ;;
+    AViiON:dgux:*:*)
+        # DG/UX returns AViiON for all architectures
+        UNAME_PROCESSOR=`/usr/bin/uname -p`
+	if [ $UNAME_PROCESSOR = mc88100 ] || [ $UNAME_PROCESSOR = mc88110 ]
+	then
+	    if [ ${TARGET_BINARY_INTERFACE}x = m88kdguxelfx ] || \
+	       [ ${TARGET_BINARY_INTERFACE}x = x ]
+	    then
+		echo m88k-dg-dgux${UNAME_RELEASE}
+	    else
+		echo m88k-dg-dguxbcs${UNAME_RELEASE}
+	    fi
+	else
+	    echo i586-dg-dgux${UNAME_RELEASE}
+	fi
+ 	exit 0 ;;
+    M88*:DolphinOS:*:*)	# DolphinOS (SVR3)
+	echo m88k-dolphin-sysv3
+	exit 0 ;;
+    M88*:*:R3*:*)
+	# Delta 88k system running SVR3
+	echo m88k-motorola-sysv3
+	exit 0 ;;
+    XD88*:*:*:*) # Tektronix XD88 system running UTekV (SVR3)
+	echo m88k-tektronix-sysv3
+	exit 0 ;;
+    Tek43[0-9][0-9]:UTek:*:*) # Tektronix 4300 system running UTek (BSD)
+	echo m68k-tektronix-bsd
+	exit 0 ;;
+    *:IRIX*:*:*)
+	echo mips-sgi-irix`echo ${UNAME_RELEASE}|sed -e 's/-/_/g'`
+	exit 0 ;;
+    ????????:AIX?:[12].1:2)   # AIX 2.2.1 or AIX 2.1.1 is RT/PC AIX.
+	echo romp-ibm-aix      # uname -m gives an 8 hex-code CPU id
+	exit 0 ;;              # Note that: echo "'`uname -s`'" gives 'AIX '
+    i*86:AIX:*:*)
+	echo i386-ibm-aix
+	exit 0 ;;
+    ia64:AIX:*:*)
+	if [ -x /usr/bin/oslevel ] ; then
+		IBM_REV=`/usr/bin/oslevel`
+	else
+		IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
+	fi
+	echo ${UNAME_MACHINE}-ibm-aix${IBM_REV}
+	exit 0 ;;
+    *:AIX:2:3)
+	if grep bos325 /usr/include/stdio.h >/dev/null 2>&1; then
+		eval $set_cc_for_build
+		sed 's/^		//' << EOF >$dummy.c
+		#include <sys/systemcfg.h>
+
+		main()
+			{
+			if (!__power_pc())
+				exit(1);
+			puts("powerpc-ibm-aix3.2.5");
+			exit(0);
+			}
+EOF
+		$CC_FOR_BUILD $dummy.c -o $dummy && ./$dummy && rm -f $dummy.c $dummy && exit 0
+		rm -f $dummy.c $dummy
+		echo rs6000-ibm-aix3.2.5
+	elif grep bos324 /usr/include/stdio.h >/dev/null 2>&1; then
+		echo rs6000-ibm-aix3.2.4
+	else
+		echo rs6000-ibm-aix3.2
+	fi
+	exit 0 ;;
+    *:AIX:*:[45])
+	IBM_CPU_ID=`/usr/sbin/lsdev -C -c processor -S available | sed 1q | awk '{ print $1 }'`
+	if /usr/sbin/lsattr -El ${IBM_CPU_ID} | grep ' POWER' >/dev/null 2>&1; then
+		IBM_ARCH=rs6000
+	else
+		IBM_ARCH=powerpc
+	fi
+	if [ -x /usr/bin/oslevel ] ; then
+		IBM_REV=`/usr/bin/oslevel`
+	else
+		IBM_REV=${UNAME_VERSION}.${UNAME_RELEASE}
+	fi
+	echo ${IBM_ARCH}-ibm-aix${IBM_REV}
+	exit 0 ;;
+    *:AIX:*:*)
+	echo rs6000-ibm-aix
+	exit 0 ;;
+    ibmrt:4.4BSD:*|romp-ibm:BSD:*)
+	echo romp-ibm-bsd4.4
+	exit 0 ;;
+    ibmrt:*BSD:*|romp-ibm:BSD:*)            # covers RT/PC BSD and
+	echo romp-ibm-bsd${UNAME_RELEASE}   # 4.3 with uname added to
+	exit 0 ;;                           # report: romp-ibm BSD 4.3
+    *:BOSX:*:*)
+	echo rs6000-bull-bosx
+	exit 0 ;;
+    DPX/2?00:B.O.S.:*:*)
+	echo m68k-bull-sysv3
+	exit 0 ;;
+    9000/[34]??:4.3bsd:1.*:*)
+	echo m68k-hp-bsd
+	exit 0 ;;
+    hp300:4.4BSD:*:* | 9000/[34]??:4.3bsd:2.*:*)
+	echo m68k-hp-bsd4.4
+	exit 0 ;;
+    9000/[34678]??:HP-UX:*:*)
+	HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
+	case "${UNAME_MACHINE}" in
+	    9000/31? )            HP_ARCH=m68000 ;;
+	    9000/[34]?? )         HP_ARCH=m68k ;;
+	    9000/[678][0-9][0-9])
+		if [ -x /usr/bin/getconf ]; then
+		    sc_cpu_version=`/usr/bin/getconf SC_CPU_VERSION 2>/dev/null`
+                    sc_kernel_bits=`/usr/bin/getconf SC_KERNEL_BITS 2>/dev/null`
+                    case "${sc_cpu_version}" in
+                      523) HP_ARCH="hppa1.0" ;; # CPU_PA_RISC1_0
+                      528) HP_ARCH="hppa1.1" ;; # CPU_PA_RISC1_1
+                      532)                      # CPU_PA_RISC2_0
+                        case "${sc_kernel_bits}" in
+                          32) HP_ARCH="hppa2.0n" ;;
+                          64) HP_ARCH="hppa2.0w" ;;
+			  '') HP_ARCH="hppa2.0" ;;   # HP-UX 10.20
+                        esac ;;
+                    esac
+		fi
+		if [ "${HP_ARCH}" = "" ]; then
+		    eval $set_cc_for_build
+		    sed 's/^              //' << EOF >$dummy.c
+
+              #define _HPUX_SOURCE
+              #include <stdlib.h>
+              #include <unistd.h>
+
+              int main ()
+              {
+              #if defined(_SC_KERNEL_BITS)
+                  long bits = sysconf(_SC_KERNEL_BITS);
+              #endif
+                  long cpu  = sysconf (_SC_CPU_VERSION);
+
+                  switch (cpu)
+              	{
+              	case CPU_PA_RISC1_0: puts ("hppa1.0"); break;
+              	case CPU_PA_RISC1_1: puts ("hppa1.1"); break;
+              	case CPU_PA_RISC2_0:
+              #if defined(_SC_KERNEL_BITS)
+              	    switch (bits)
+              		{
+              		case 64: puts ("hppa2.0w"); break;
+              		case 32: puts ("hppa2.0n"); break;
+              		default: puts ("hppa2.0"); break;
+              		} break;
+              #else  /* !defined(_SC_KERNEL_BITS) */
+              	    puts ("hppa2.0"); break;
+              #endif
+              	default: puts ("hppa1.0"); break;
+              	}
+                  exit (0);
+              }
+EOF
+		    (CCOPTS= $CC_FOR_BUILD $dummy.c -o $dummy 2>/dev/null) && HP_ARCH=`./$dummy`
+		    if test -z "$HP_ARCH"; then HP_ARCH=hppa; fi
+		    rm -f $dummy.c $dummy
+		fi ;;
+	esac
+	echo ${HP_ARCH}-hp-hpux${HPUX_REV}
+	exit 0 ;;
+    ia64:HP-UX:*:*)
+	HPUX_REV=`echo ${UNAME_RELEASE}|sed -e 's/[^.]*.[0B]*//'`
+	echo ia64-hp-hpux${HPUX_REV}
+	exit 0 ;;
+    3050*:HI-UX:*:*)
+	eval $set_cc_for_build
+	sed 's/^	//' << EOF >$dummy.c
+	#include <unistd.h>
+	int
+	main ()
+	{
+	  long cpu = sysconf (_SC_CPU_VERSION);
+	  /* The order matters, because CPU_IS_HP_MC68K erroneously returns
+	     true for CPU_PA_RISC1_0.  CPU_IS_PA_RISC returns correct
+	     results, however.  */
+	  if (CPU_IS_PA_RISC (cpu))
+	    {
+	      switch (cpu)
+		{
+		  case CPU_PA_RISC1_0: puts ("hppa1.0-hitachi-hiuxwe2"); break;
+		  case CPU_PA_RISC1_1: puts ("hppa1.1-hitachi-hiuxwe2"); break;
+		  case CPU_PA_RISC2_0: puts ("hppa2.0-hitachi-hiuxwe2"); break;
+		  default: puts ("hppa-hitachi-hiuxwe2"); break;
+		}
+	    }
+	  else if (CPU_IS_HP_MC68K (cpu))
+	    puts ("m68k-hitachi-hiuxwe2");
+	  else puts ("unknown-hitachi-hiuxwe2");
+	  exit (0);
+	}
+EOF
+	$CC_FOR_BUILD $dummy.c -o $dummy && ./$dummy && rm -f $dummy.c $dummy && exit 0
+	rm -f $dummy.c $dummy
+	echo unknown-hitachi-hiuxwe2
+	exit 0 ;;
+    9000/7??:4.3bsd:*:* | 9000/8?[79]:4.3bsd:*:* )
+	echo hppa1.1-hp-bsd
+	exit 0 ;;
+    9000/8??:4.3bsd:*:*)
+	echo hppa1.0-hp-bsd
+	exit 0 ;;
+    *9??*:MPE/iX:*:* | *3000*:MPE/iX:*:*)
+	echo hppa1.0-hp-mpeix
+	exit 0 ;;
+    hp7??:OSF1:*:* | hp8?[79]:OSF1:*:* )
+	echo hppa1.1-hp-osf
+	exit 0 ;;
+    hp8??:OSF1:*:*)
+	echo hppa1.0-hp-osf
+	exit 0 ;;
+    i*86:OSF1:*:*)
+	if [ -x /usr/sbin/sysversion ] ; then
+	    echo ${UNAME_MACHINE}-unknown-osf1mk
+	else
+	    echo ${UNAME_MACHINE}-unknown-osf1
+	fi
+	exit 0 ;;
+    parisc*:Lites*:*:*)
+	echo hppa1.1-hp-lites
+	exit 0 ;;
+    C1*:ConvexOS:*:* | convex:ConvexOS:C1*:*)
+	echo c1-convex-bsd
+        exit 0 ;;
+    C2*:ConvexOS:*:* | convex:ConvexOS:C2*:*)
+	if getsysinfo -f scalar_acc
+	then echo c32-convex-bsd
+	else echo c2-convex-bsd
+	fi
+        exit 0 ;;
+    C34*:ConvexOS:*:* | convex:ConvexOS:C34*:*)
+	echo c34-convex-bsd
+        exit 0 ;;
+    C38*:ConvexOS:*:* | convex:ConvexOS:C38*:*)
+	echo c38-convex-bsd
+        exit 0 ;;
+    C4*:ConvexOS:*:* | convex:ConvexOS:C4*:*)
+	echo c4-convex-bsd
+        exit 0 ;;
+    CRAY*Y-MP:*:*:*)
+	echo ymp-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    CRAY*[A-Z]90:*:*:*)
+	echo ${UNAME_MACHINE}-cray-unicos${UNAME_RELEASE} \
+	| sed -e 's/CRAY.*\([A-Z]90\)/\1/' \
+	      -e y/ABCDEFGHIJKLMNOPQRSTUVWXYZ/abcdefghijklmnopqrstuvwxyz/ \
+	      -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    CRAY*TS:*:*:*)
+	echo t90-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    CRAY*T3D:*:*:*)
+	echo alpha-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    CRAY*T3E:*:*:*)
+	echo alphaev5-cray-unicosmk${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    CRAY*SV1:*:*:*)
+	echo sv1-cray-unicos${UNAME_RELEASE} | sed -e 's/\.[^.]*$/.X/'
+	exit 0 ;;
+    F30[01]:UNIX_System_V:*:* | F700:UNIX_System_V:*:*)
+	FUJITSU_PROC=`uname -m | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz'`
+        FUJITSU_SYS=`uname -p | tr 'ABCDEFGHIJKLMNOPQRSTUVWXYZ' 'abcdefghijklmnopqrstuvwxyz' | sed -e 's/\///'`
+        FUJITSU_REL=`echo ${UNAME_RELEASE} | sed -e 's/ /_/'`
+        echo "${FUJITSU_PROC}-fujitsu-${FUJITSU_SYS}${FUJITSU_REL}"
+        exit 0 ;;
+    i*86:BSD/386:*:* | i*86:BSD/OS:*:* | *:Ascend\ Embedded/OS:*:*)
+	echo ${UNAME_MACHINE}-pc-bsdi${UNAME_RELEASE}
+	exit 0 ;;
+    sparc*:BSD/OS:*:*)
+	echo sparc-unknown-bsdi${UNAME_RELEASE}
+	exit 0 ;;
+    *:BSD/OS:*:*)
+	echo ${UNAME_MACHINE}-unknown-bsdi${UNAME_RELEASE}
+	exit 0 ;;
+    *:FreeBSD:*:*)
+	echo ${UNAME_MACHINE}-unknown-freebsd`echo ${UNAME_RELEASE}|sed -e 's/[-(].*//'`
+	exit 0 ;;
+    i*:CYGWIN*:*)
+	echo ${UNAME_MACHINE}-pc-cygwin
+	exit 0 ;;
+    i*:MINGW*:*)
+	echo ${UNAME_MACHINE}-pc-mingw32
+	exit 0 ;;
+    i*:PW*:*)
+	echo ${UNAME_MACHINE}-pc-pw32
+	exit 0 ;;
+    x86:Interix*:3*)
+	echo i386-pc-interix3
+	exit 0 ;;
+    i*:Windows_NT*:* | Pentium*:Windows_NT*:*)
+	# How do we know it's Interix rather than the generic POSIX subsystem?
+	# It also conflicts with pre-2.0 versions of AT&T UWIN. Should we
+	# UNAME_MACHINE based on the output of uname instead of i386?
+	echo i386-pc-interix
+	exit 0 ;;
+    i*:UWIN*:*)
+	echo ${UNAME_MACHINE}-pc-uwin
+	exit 0 ;;
+    p*:CYGWIN*:*)
+	echo powerpcle-unknown-cygwin
+	exit 0 ;;
+    prep*:SunOS:5.*:*)
+	echo powerpcle-unknown-solaris2`echo ${UNAME_RELEASE}|sed -e 's/[^.]*//'`
+	exit 0 ;;
+    *:GNU:*:*)
+	echo `echo ${UNAME_MACHINE}|sed -e 's,[-/].*$,,'`-unknown-gnu`echo ${UNAME_RELEASE}|sed -e 's,/.*$,,'`
+	exit 0 ;;
+    i*86:Minix:*:*)
+	echo ${UNAME_MACHINE}-pc-minix
+	exit 0 ;;
+    arm*:Linux:*:*)
+	echo ${UNAME_MACHINE}-unknown-linux-gnu
+	exit 0 ;;
+    ia64:Linux:*:*)
+	echo ${UNAME_MACHINE}-unknown-linux-gnu
+	exit 0 ;;
+    m68*:Linux:*:*)
+	echo ${UNAME_MACHINE}-unknown-linux-gnu
+	exit 0 ;;
+    mips:Linux:*:*)
+	eval $set_cc_for_build
+	sed 's/^	//' << EOF >$dummy.c
+	#undef CPU
+	#undef mips
+	#undef mipsel
+	#if defined(__MIPSEL__) || defined(__MIPSEL) || defined(_MIPSEL) || defined(MIPSEL)
+	CPU=mipsel
+	#else
+	#if defined(__MIPSEB__) || defined(__MIPSEB) || defined(_MIPSEB) || defined(MIPSEB)
+	CPU=mips
+	#else
+	CPU=
+	#endif
+	#endif
+EOF
+	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^CPU=`
+	rm -f $dummy.c
+	test x"${CPU}" != x && echo "${CPU}-pc-linux-gnu" && exit 0
+	;;
+    ppc:Linux:*:*)
+	echo powerpc-unknown-linux-gnu
+	exit 0 ;;
+    ppc64:Linux:*:*)
+	echo powerpc64-unknown-linux-gnu
+	exit 0 ;;
+    alpha:Linux:*:*)
+	case `sed -n '/^cpu model/s/^.*: \(.*\)/\1/p' < /proc/cpuinfo` in
+	  EV5)   UNAME_MACHINE=alphaev5 ;;
+	  EV56)  UNAME_MACHINE=alphaev56 ;;
+	  PCA56) UNAME_MACHINE=alphapca56 ;;
+	  PCA57) UNAME_MACHINE=alphapca56 ;;
+	  EV6)   UNAME_MACHINE=alphaev6 ;;
+	  EV67)  UNAME_MACHINE=alphaev67 ;;
+	  EV68*) UNAME_MACHINE=alphaev68 ;;
+        esac
+	objdump --private-headers /bin/sh | grep ld.so.1 >/dev/null
+	if test "$?" = 0 ; then LIBC="libc1" ; else LIBC="" ; fi
+	echo ${UNAME_MACHINE}-unknown-linux-gnu${LIBC}
+	exit 0 ;;
+    parisc:Linux:*:* | hppa:Linux:*:*)
+	# Look for CPU level
+	case `grep '^cpu[^a-z]*:' /proc/cpuinfo 2>/dev/null | cut -d' ' -f2` in
+	  PA7*) echo hppa1.1-unknown-linux-gnu ;;
+	  PA8*) echo hppa2.0-unknown-linux-gnu ;;
+	  *)    echo hppa-unknown-linux-gnu ;;
+	esac
+	exit 0 ;;
+    parisc64:Linux:*:* | hppa64:Linux:*:*)
+	echo hppa64-unknown-linux-gnu
+	exit 0 ;;
+    s390:Linux:*:* | s390x:Linux:*:*)
+	echo ${UNAME_MACHINE}-ibm-linux
+	exit 0 ;;
+    sh*:Linux:*:*)
+	echo ${UNAME_MACHINE}-unknown-linux-gnu
+	exit 0 ;;
+    sparc:Linux:*:* | sparc64:Linux:*:*)
+	echo ${UNAME_MACHINE}-unknown-linux-gnu
+	exit 0 ;;
+    x86_64:Linux:*:*)
+	echo x86_64-unknown-linux-gnu
+	exit 0 ;;
+    i*86:Linux:*:*)
+	# The BFD linker knows what the default object file format is, so
+	# first see if it will tell us. cd to the root directory to prevent
+	# problems with other programs or directories called `ld' in the path.
+	# Set LC_ALL=C to ensure ld outputs messages in English.
+	ld_supported_targets=`cd /; LC_ALL=C ld --help 2>&1 \
+			 | sed -ne '/supported targets:/!d
+				    s/[ 	][ 	]*/ /g
+				    s/.*supported targets: *//
+				    s/ .*//
+				    p'`
+        case "$ld_supported_targets" in
+	  elf32-i386)
+		TENTATIVE="${UNAME_MACHINE}-pc-linux-gnu"
+		;;
+	  a.out-i386-linux)
+		echo "${UNAME_MACHINE}-pc-linux-gnuaout"
+		exit 0 ;;		
+	  coff-i386)
+		echo "${UNAME_MACHINE}-pc-linux-gnucoff"
+		exit 0 ;;
+	  "")
+		# Either a pre-BFD a.out linker (linux-gnuoldld) or
+		# one that does not give us useful --help.
+		echo "${UNAME_MACHINE}-pc-linux-gnuoldld"
+		exit 0 ;;
+	esac
+	# Determine whether the default compiler is a.out or elf
+	eval $set_cc_for_build
+	sed 's/^	//' << EOF >$dummy.c
+	#include <features.h>
+	#ifdef __ELF__
+	# ifdef __GLIBC__
+	#  if __GLIBC__ >= 2
+	LIBC=gnu
+	#  else
+	LIBC=gnulibc1
+	#  endif
+	# else
+	LIBC=gnulibc1
+	# endif
+	#else
+	#ifdef __INTEL_COMPILER
+	LIBC=gnu
+	#else
+	LIBC=gnuaout
+	#endif
+	#endif
+EOF
+	eval `$CC_FOR_BUILD -E $dummy.c 2>/dev/null | grep ^LIBC=`
+	rm -f $dummy.c
+	test x"${LIBC}" != x && echo "${UNAME_MACHINE}-pc-linux-${LIBC}" && exit 0
+	test x"${TENTATIVE}" != x && echo "${TENTATIVE}" && exit 0
+	;;
+    i*86:DYNIX/ptx:4*:*)
+	# ptx 4.0 does uname -s correctly, with DYNIX/ptx in there.
+	# earlier versions are messed up and put the nodename in both
+	# sysname and nodename.
+	echo i386-sequent-sysv4
+	exit 0 ;;
+    i*86:UNIX_SV:4.2MP:2.*)
+        # Unixware is an offshoot of SVR4, but it has its own version
+        # number series starting with 2...
+        # I am not positive that other SVR4 systems won't match this,
+	# I just have to hope.  -- rms.
+        # Use sysv4.2uw... so that sysv4* matches it.
+	echo ${UNAME_MACHINE}-pc-sysv4.2uw${UNAME_VERSION}
+	exit 0 ;;
+    i*86:*:4.*:* | i*86:SYSTEM_V:4.*:*)
+	UNAME_REL=`echo ${UNAME_RELEASE} | sed 's/\/MP$//'`
+	if grep Novell /usr/include/link.h >/dev/null 2>/dev/null; then
+		echo ${UNAME_MACHINE}-univel-sysv${UNAME_REL}
+	else
+		echo ${UNAME_MACHINE}-pc-sysv${UNAME_REL}
+	fi
+	exit 0 ;;
+    i*86:*:5:[78]*)
+	case `/bin/uname -X | grep "^Machine"` in
+	    *486*)	     UNAME_MACHINE=i486 ;;
+	    *Pentium)	     UNAME_MACHINE=i586 ;;
+	    *Pent*|*Celeron) UNAME_MACHINE=i686 ;;
+	esac
+	echo ${UNAME_MACHINE}-unknown-sysv${UNAME_RELEASE}${UNAME_SYSTEM}${UNAME_VERSION}
+	exit 0 ;;
+    i*86:*:3.2:*)
+	if test -f /usr/options/cb.name; then
+		UNAME_REL=`sed -n 's/.*Version //p' </usr/options/cb.name`
+		echo ${UNAME_MACHINE}-pc-isc$UNAME_REL
+	elif /bin/uname -X 2>/dev/null >/dev/null ; then
+		UNAME_REL=`(/bin/uname -X|egrep Release|sed -e 's/.*= //')`
+		(/bin/uname -X|egrep i80486 >/dev/null) && UNAME_MACHINE=i486
+		(/bin/uname -X|egrep '^Machine.*Pentium' >/dev/null) \
+			&& UNAME_MACHINE=i586
+		(/bin/uname -X|egrep '^Machine.*Pent ?II' >/dev/null) \
+			&& UNAME_MACHINE=i686
+		(/bin/uname -X|egrep '^Machine.*Pentium Pro' >/dev/null) \
+			&& UNAME_MACHINE=i686
+		echo ${UNAME_MACHINE}-pc-sco$UNAME_REL
+	else
+		echo ${UNAME_MACHINE}-pc-sysv32
+	fi
+	exit 0 ;;
+    i*86:*DOS:*:*)
+	echo ${UNAME_MACHINE}-pc-msdosdjgpp
+	exit 0 ;;
+    pc:*:*:*)
+	# Left here for compatibility:
+        # uname -m prints for DJGPP always 'pc', but it prints nothing about
+        # the processor, so we play safe by assuming i386.
+	echo i386-pc-msdosdjgpp
+        exit 0 ;;
+    Intel:Mach:3*:*)
+	echo i386-pc-mach3
+	exit 0 ;;
+    paragon:*:*:*)
+	echo i860-intel-osf1
+	exit 0 ;;
+    i860:*:4.*:*) # i860-SVR4
+	if grep Stardent /usr/include/sys/uadmin.h >/dev/null 2>&1 ; then
+	  echo i860-stardent-sysv${UNAME_RELEASE} # Stardent Vistra i860-SVR4
+	else # Add other i860-SVR4 vendors below as they are discovered.
+	  echo i860-unknown-sysv${UNAME_RELEASE}  # Unknown i860-SVR4
+	fi
+	exit 0 ;;
+    mini*:CTIX:SYS*5:*)
+	# "miniframe"
+	echo m68010-convergent-sysv
+	exit 0 ;;
+    M68*:*:R3V[567]*:*)
+	test -r /sysV68 && echo 'm68k-motorola-sysv' && exit 0 ;;
+    3[34]??:*:4.0:3.0 | 3[34]??A:*:4.0:3.0 | 3[34]??,*:*:4.0:3.0 | 3[34]??/*:*:4.0:3.0 | 4850:*:4.0:3.0 | SKA40:*:4.0:3.0)
+	OS_REL=''
+	test -r /etc/.relid \
+	&& OS_REL=.`sed -n 's/[^ ]* [^ ]* \([0-9][0-9]\).*/\1/p' < /etc/.relid`
+	/bin/uname -p 2>/dev/null | grep 86 >/dev/null \
+	  && echo i486-ncr-sysv4.3${OS_REL} && exit 0
+	/bin/uname -p 2>/dev/null | /bin/grep entium >/dev/null \
+	  && echo i586-ncr-sysv4.3${OS_REL} && exit 0 ;;
+    3[34]??:*:4.0:* | 3[34]??,*:*:4.0:*)
+        /bin/uname -p 2>/dev/null | grep 86 >/dev/null \
+          && echo i486-ncr-sysv4 && exit 0 ;;
+    m68*:LynxOS:2.*:* | m68*:LynxOS:3.0*:*)
+	echo m68k-unknown-lynxos${UNAME_RELEASE}
+	exit 0 ;;
+    mc68030:UNIX_System_V:4.*:*)
+	echo m68k-atari-sysv4
+	exit 0 ;;
+    i*86:LynxOS:2.*:* | i*86:LynxOS:3.[01]*:* | i*86:LynxOS:4.0*:*)
+	echo i386-unknown-lynxos${UNAME_RELEASE}
+	exit 0 ;;
+    TSUNAMI:LynxOS:2.*:*)
+	echo sparc-unknown-lynxos${UNAME_RELEASE}
+	exit 0 ;;
+    rs6000:LynxOS:2.*:*)
+	echo rs6000-unknown-lynxos${UNAME_RELEASE}
+	exit 0 ;;
+    PowerPC:LynxOS:2.*:* | PowerPC:LynxOS:3.[01]*:* | PowerPC:LynxOS:4.0*:*)
+	echo powerpc-unknown-lynxos${UNAME_RELEASE}
+	exit 0 ;;
+    SM[BE]S:UNIX_SV:*:*)
+	echo mips-dde-sysv${UNAME_RELEASE}
+	exit 0 ;;
+    RM*:ReliantUNIX-*:*:*)
+	echo mips-sni-sysv4
+	exit 0 ;;
+    RM*:SINIX-*:*:*)
+	echo mips-sni-sysv4
+	exit 0 ;;
+    *:SINIX-*:*:*)
+	if uname -p 2>/dev/null >/dev/null ; then
+		UNAME_MACHINE=`(uname -p) 2>/dev/null`
+		echo ${UNAME_MACHINE}-sni-sysv4
+	else
+		echo ns32k-sni-sysv
+	fi
+	exit 0 ;;
+    PENTIUM:*:4.0*:*) # Unisys `ClearPath HMP IX 4000' SVR4/MP effort
+                      # says <Richard.M.Bartel@ccMail.Census.GOV>
+        echo i586-unisys-sysv4
+        exit 0 ;;
+    *:UNIX_System_V:4*:FTX*)
+	# From Gerald Hewes <hewes@openmarket.com>.
+	# How about differentiating between stratus architectures? -djm
+	echo hppa1.1-stratus-sysv4
+	exit 0 ;;
+    *:*:*:FTX*)
+	# From seanf@swdc.stratus.com.
+	echo i860-stratus-sysv4
+	exit 0 ;;
+    *:VOS:*:*)
+	# From Paul.Green@stratus.com.
+	echo hppa1.1-stratus-vos
+	exit 0 ;;
+    mc68*:A/UX:*:*)
+	echo m68k-apple-aux${UNAME_RELEASE}
+	exit 0 ;;
+    news*:NEWS-OS:6*:*)
+	echo mips-sony-newsos6
+	exit 0 ;;
+    R[34]000:*System_V*:*:* | R4000:UNIX_SYSV:*:* | R*000:UNIX_SV:*:*)
+	if [ -d /usr/nec ]; then
+	        echo mips-nec-sysv${UNAME_RELEASE}
+	else
+	        echo mips-unknown-sysv${UNAME_RELEASE}
+	fi
+        exit 0 ;;
+    BeBox:BeOS:*:*)	# BeOS running on hardware made by Be, PPC only.
+	echo powerpc-be-beos
+	exit 0 ;;
+    BeMac:BeOS:*:*)	# BeOS running on Mac or Mac clone, PPC only.
+	echo powerpc-apple-beos
+	exit 0 ;;
+    BePC:BeOS:*:*)	# BeOS running on Intel PC compatible.
+	echo i586-pc-beos
+	exit 0 ;;
+    SX-4:SUPER-UX:*:*)
+	echo sx4-nec-superux${UNAME_RELEASE}
+	exit 0 ;;
+    SX-5:SUPER-UX:*:*)
+	echo sx5-nec-superux${UNAME_RELEASE}
+	exit 0 ;;
+    Power*:Rhapsody:*:*)
+	echo powerpc-apple-rhapsody${UNAME_RELEASE}
+	exit 0 ;;
+    *:Rhapsody:*:*)
+	echo ${UNAME_MACHINE}-apple-rhapsody${UNAME_RELEASE}
+	exit 0 ;;
+    *:Darwin:*:*)
+	echo `uname -p`-apple-darwin${UNAME_RELEASE}
+	exit 0 ;;
+    *:procnto*:*:* | *:QNX:[0123456789]*:*)
+	UNAME_PROCESSOR=`uname -p`
+	if test "$UNAME_PROCESSOR" = "x86"; then
+		UNAME_PROCESSOR=i386
+		UNAME_MACHINE=pc
+	fi
+	echo ${UNAME_PROCESSOR}-${UNAME_MACHINE}-nto-qnx${UNAME_RELEASE}
+	exit 0 ;;
+    *:QNX:*:4*)
+	echo i386-pc-qnx
+	exit 0 ;;
+    NSR-[GKLNPTVW]:NONSTOP_KERNEL:*:*)
+	echo nsr-tandem-nsk${UNAME_RELEASE}
+	exit 0 ;;
+    *:NonStop-UX:*:*)
+	echo mips-compaq-nonstopux
+	exit 0 ;;
+    BS2000:POSIX*:*:*)
+	echo bs2000-siemens-sysv
+	exit 0 ;;
+    DS/*:UNIX_System_V:*:*)
+	echo ${UNAME_MACHINE}-${UNAME_SYSTEM}-${UNAME_RELEASE}
+	exit 0 ;;
+    *:Plan9:*:*)
+	# "uname -m" is not consistent, so use $cputype instead. 386
+	# is converted to i386 for consistency with other x86
+	# operating systems.
+	if test "$cputype" = "386"; then
+	    UNAME_MACHINE=i386
+	else
+	    UNAME_MACHINE="$cputype"
+	fi
+	echo ${UNAME_MACHINE}-unknown-plan9
+	exit 0 ;;
+    i*86:OS/2:*:*)
+	# If we were able to find `uname', then EMX Unix compatibility
+	# is probably installed.
+	echo ${UNAME_MACHINE}-pc-os2-emx
+	exit 0 ;;
+    *:TOPS-10:*:*)
+	echo pdp10-unknown-tops10
+	exit 0 ;;
+    *:TENEX:*:*)
+	echo pdp10-unknown-tenex
+	exit 0 ;;
+    KS10:TOPS-20:*:* | KL10:TOPS-20:*:* | TYPE4:TOPS-20:*:*)
+	echo pdp10-dec-tops20
+	exit 0 ;;
+    XKL-1:TOPS-20:*:* | TYPE5:TOPS-20:*:*)
+	echo pdp10-xkl-tops20
+	exit 0 ;;
+    *:TOPS-20:*:*)
+	echo pdp10-unknown-tops20
+	exit 0 ;;
+    *:ITS:*:*)
+	echo pdp10-unknown-its
+	exit 0 ;;
+    i*86:XTS-300:*:STOP)
+	echo ${UNAME_MACHINE}-unknown-stop
+	exit 0 ;;
+    i*86:atheos:*:*)
+	echo ${UNAME_MACHINE}-unknown-atheos
+	exit 0 ;;
+esac
+
+#echo '(No uname command or uname output not recognized.)' 1>&2
+#echo "${UNAME_MACHINE}:${UNAME_SYSTEM}:${UNAME_RELEASE}:${UNAME_VERSION}" 1>&2
+
+eval $set_cc_for_build
+cat >$dummy.c <<EOF
+#ifdef _SEQUENT_
+# include <sys/types.h>
+# include <sys/utsname.h>
+#endif
+main ()
+{
+#if defined (sony)
+#if defined (MIPSEB)
+  /* BFD wants "bsd" instead of "newsos".  Perhaps BFD should be changed,
+     I don't know....  */
+  printf ("mips-sony-bsd\n"); exit (0);
+#else
+#include <sys/param.h>
+  printf ("m68k-sony-newsos%s\n",
+#ifdef NEWSOS4
+          "4"
+#else
+	  ""
+#endif
+         ); exit (0);
+#endif
+#endif
+
+#if defined (__arm) && defined (__acorn) && defined (__unix)
+  printf ("arm-acorn-riscix"); exit (0);
+#endif
+
+#if defined (hp300) && !defined (hpux)
+  printf ("m68k-hp-bsd\n"); exit (0);
+#endif
+
+#if defined (NeXT)
+#if !defined (__ARCHITECTURE__)
+#define __ARCHITECTURE__ "m68k"
+#endif
+  int version;
+  version=`(hostinfo | sed -n 's/.*NeXT Mach \([0-9]*\).*/\1/p') 2>/dev/null`;
+  if (version < 4)
+    printf ("%s-next-nextstep%d\n", __ARCHITECTURE__, version);
+  else
+    printf ("%s-next-openstep%d\n", __ARCHITECTURE__, version);
+  exit (0);
+#endif
+
+#if defined (MULTIMAX) || defined (n16)
+#if defined (UMAXV)
+  printf ("ns32k-encore-sysv\n"); exit (0);
+#else
+#if defined (CMU)
+  printf ("ns32k-encore-mach\n"); exit (0);
+#else
+  printf ("ns32k-encore-bsd\n"); exit (0);
+#endif
+#endif
+#endif
+
+#if defined (__386BSD__)
+  printf ("i386-pc-bsd\n"); exit (0);
+#endif
+
+#if defined (sequent)
+#if defined (i386)
+  printf ("i386-sequent-dynix\n"); exit (0);
+#endif
+#if defined (ns32000)
+  printf ("ns32k-sequent-dynix\n"); exit (0);
+#endif
+#endif
+
+#if defined (_SEQUENT_)
+    struct utsname un;
+
+    uname(&un);
+
+    if (strncmp(un.version, "V2", 2) == 0) {
+	printf ("i386-sequent-ptx2\n"); exit (0);
+    }
+    if (strncmp(un.version, "V1", 2) == 0) { /* XXX is V1 correct? */
+	printf ("i386-sequent-ptx1\n"); exit (0);
+    }
+    printf ("i386-sequent-ptx\n"); exit (0);
+
+#endif
+
+#if defined (vax)
+# if !defined (ultrix)
+#  include <sys/param.h>
+#  if defined (BSD)
+#   if BSD == 43
+      printf ("vax-dec-bsd4.3\n"); exit (0);
+#   else
+#    if BSD == 199006
+      printf ("vax-dec-bsd4.3reno\n"); exit (0);
+#    else
+      printf ("vax-dec-bsd\n"); exit (0);
+#    endif
+#   endif
+#  else
+    printf ("vax-dec-bsd\n"); exit (0);
+#  endif
+# else
+    printf ("vax-dec-ultrix\n"); exit (0);
+# endif
+#endif
+
+#if defined (alliant) && defined (i860)
+  printf ("i860-alliant-bsd\n"); exit (0);
+#endif
+
+  exit (1);
+}
+EOF
+
+$CC_FOR_BUILD $dummy.c -o $dummy 2>/dev/null && ./$dummy && rm -f $dummy.c $dummy && exit 0
+rm -f $dummy.c $dummy
+
+# Apollos put the system type in the environment.
+
+test -d /usr/apollo && { echo ${ISP}-apollo-${SYSTYPE}; exit 0; }
+
+# Convex versions that predate uname can use getsysinfo(1)
+
+if [ -x /usr/convex/getsysinfo ]
+then
+    case `getsysinfo -f cpu_type` in
+    c1*)
+	echo c1-convex-bsd
+	exit 0 ;;
+    c2*)
+	if getsysinfo -f scalar_acc
+	then echo c32-convex-bsd
+	else echo c2-convex-bsd
+	fi
+	exit 0 ;;
+    c34*)
+	echo c34-convex-bsd
+	exit 0 ;;
+    c38*)
+	echo c38-convex-bsd
+	exit 0 ;;
+    c4*)
+	echo c4-convex-bsd
+	exit 0 ;;
+    esac
+fi
+
+cat >&2 <<EOF
+$0: unable to guess system type
+
+This script, last modified $timestamp, has failed to recognize
+the operating system you are using. It is advised that you
+download the most up to date version of the config scripts from
+
+    ftp://ftp.gnu.org/pub/gnu/config/
+
+If the version you run ($0) is already up to date, please
+send the following data and any information you think might be
+pertinent to <config-patches@gnu.org> in order to provide the needed
+information to handle your system.
+
+config.guess timestamp = $timestamp
+
+uname -m = `(uname -m) 2>/dev/null || echo unknown`
+uname -r = `(uname -r) 2>/dev/null || echo unknown`
+uname -s = `(uname -s) 2>/dev/null || echo unknown`
+uname -v = `(uname -v) 2>/dev/null || echo unknown`
+
+/usr/bin/uname -p = `(/usr/bin/uname -p) 2>/dev/null`
+/bin/uname -X     = `(/bin/uname -X) 2>/dev/null`
+
+hostinfo               = `(hostinfo) 2>/dev/null`
+/bin/universe          = `(/bin/universe) 2>/dev/null`
+/usr/bin/arch -k       = `(/usr/bin/arch -k) 2>/dev/null`
+/bin/arch              = `(/bin/arch) 2>/dev/null`
+/usr/bin/oslevel       = `(/usr/bin/oslevel) 2>/dev/null`
+/usr/convex/getsysinfo = `(/usr/convex/getsysinfo) 2>/dev/null`
+
+UNAME_MACHINE = ${UNAME_MACHINE}
+UNAME_RELEASE = ${UNAME_RELEASE}
+UNAME_SYSTEM  = ${UNAME_SYSTEM}
+UNAME_VERSION = ${UNAME_VERSION}
+EOF
+
+exit 1
+
+# Local variables:
+# eval: (add-hook 'write-file-hooks 'time-stamp)
+# time-stamp-start: "timestamp='"
+# time-stamp-format: "%:y-%02m-%02d"
+# time-stamp-end: "'"
+# End:
diff --git a/xml/expat/conftools/config.sub b/xml/expat/conftools/config.sub
new file mode 100755
index 000000000..043d45b39
--- /dev/null
+++ b/xml/expat/conftools/config.sub
@@ -0,0 +1,1507 @@
+#! /bin/sh
+# Configuration validation subroutine script.
+#   Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
+#   2000, 2001, 2002 Free Software Foundation, Inc.
+
+timestamp='2002-03-07'
+
+# This file is (in principle) common to ALL GNU software.
+# The presence of a machine in this file suggests that SOME GNU software
+# can handle that machine.  It does not imply ALL GNU software can.
+#
+# This file is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330,
+# Boston, MA 02111-1307, USA.
+
+# As a special exception to the GNU General Public License, if you
+# distribute this file as part of a program that contains a
+# configuration script generated by Autoconf, you may include it under
+# the same distribution terms that you use for the rest of that program.
+
+#####################################################################
+# This file contains changes for Apache, clearly marked below.
+# These changes are hereby donated to the public domain.
+#####################################################################
+
+# Please send patches to <config-patches@gnu.org>.  Submit a context
+# diff and a properly formatted ChangeLog entry.
+#
+# Configuration subroutine to validate and canonicalize a configuration type.
+# Supply the specified configuration type as an argument.
+# If it is invalid, we print an error message on stderr and exit with code 1.
+# Otherwise, we print the canonical config type on stdout and succeed.
+
+# This file is supposed to be the same for all GNU packages
+# and recognize all the CPU types, system types and aliases
+# that are meaningful with *any* GNU software.
+# Each package is responsible for reporting which valid configurations
+# it does not support.  The user should be able to distinguish
+# a failure to support a valid configuration from a meaningless
+# configuration.
+
+# The goal of this file is to map all the various variations of a given
+# machine specification into a single specification in the form:
+#	CPU_TYPE-MANUFACTURER-OPERATING_SYSTEM
+# or in some cases, the newer four-part form:
+#	CPU_TYPE-MANUFACTURER-KERNEL-OPERATING_SYSTEM
+# It is wrong to echo any other type of specification.
+
+me=`echo "$0" | sed -e 's,.*/,,'`
+
+usage="\
+Usage: $0 [OPTION] CPU-MFR-OPSYS
+       $0 [OPTION] ALIAS
+
+Canonicalize a configuration name.
+
+Operation modes:
+  -h, --help         print this help, then exit
+  -t, --time-stamp   print date of last modification, then exit
+  -v, --version      print version number, then exit
+
+Report bugs and patches to <config-patches@gnu.org>."
+
+version="\
+GNU config.sub ($timestamp)
+
+Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001
+Free Software Foundation, Inc.
+
+This is free software; see the source for copying conditions.  There is NO
+warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
+
+help="
+Try \`$me --help' for more information."
+
+# Parse command line
+while test $# -gt 0 ; do
+  case $1 in
+    --time-stamp | --time* | -t )
+       echo "$timestamp" ; exit 0 ;;
+    --version | -v )
+       echo "$version" ; exit 0 ;;
+    --help | --h* | -h )
+       echo "$usage"; exit 0 ;;
+    -- )     # Stop option processing
+       shift; break ;;
+    - )	# Use stdin as input.
+       break ;;
+    -* )
+       echo "$me: invalid option $1$help"
+       exit 1 ;;
+
+    *local*)
+       # First pass through any local machine types.
+       echo $1
+       exit 0;;
+
+    * )
+       break ;;
+  esac
+done
+
+case $# in
+ 0) echo "$me: missing argument$help" >&2
+    exit 1;;
+ 1) ;;
+ *) echo "$me: too many arguments$help" >&2
+    exit 1;;
+esac
+
+# Separate what the user gave into CPU-COMPANY and OS or KERNEL-OS (if any).
+# Here we must recognize all the valid KERNEL-OS combinations.
+maybe_os=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\2/'`
+case $maybe_os in
+  nto-qnx* | linux-gnu* | storm-chaos* | os2-emx* | windows32-* | rtmk-nova*)
+    os=-$maybe_os
+    basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`
+    ;;
+########################
+# changes for Apache
+#
+  tpf | os390 | vmcms)
+    os=-$maybe_os
+    basic_machine=s390;
+    ;;
+  os400)
+    os=-$maybe_os
+    basic_machine=as400;
+    ;;
+  mvs)
+    os=-mvs
+    basic_machine=i370;
+    ;;
+#
+# end Apache changes
+########################
+  *)
+    basic_machine=`echo $1 | sed 's/-[^-]*$//'`
+    if [ $basic_machine != $1 ]
+    then os=`echo $1 | sed 's/.*-/-/'`
+    else os=; fi
+    ;;
+esac
+
+### Let's recognize common machines as not being operating systems so
+### that things like config.sub decstation-3100 work.  We also
+### recognize some manufacturers as not being operating systems, so we
+### can provide default operating systems below.
+case $os in
+	-sun*os*)
+		# Prevent following clause from handling this invalid input.
+		;;
+	-dec* | -mips* | -sequent* | -encore* | -pc532* | -sgi* | -sony* | \
+	-att* | -7300* | -3300* | -delta* | -motorola* | -sun[234]* | \
+	-unicom* | -ibm* | -next | -hp | -isi* | -apollo | -altos* | \
+	-convergent* | -ncr* | -news | -32* | -3600* | -3100* | -hitachi* |\
+	-c[123]* | -convex* | -sun | -crds | -omron* | -dg | -ultra | -tti* | \
+	-harris | -dolphin | -highlevel | -gould | -cbm | -ns | -masscomp | \
+	-apple | -axis)
+		os=
+		basic_machine=$1
+		;;
+	-sim | -cisco | -oki | -wec | -winbond)
+		os=
+		basic_machine=$1
+		;;
+	-scout)
+		;;
+	-wrs)
+		os=-vxworks
+		basic_machine=$1
+		;;
+	-chorusos*)
+		os=-chorusos
+		basic_machine=$1
+		;;
+ 	-chorusrdb)
+ 		os=-chorusrdb
+		basic_machine=$1
+ 		;;
+	-hiux*)
+		os=-hiuxwe2
+		;;
+	-sco5)
+		os=-sco3.2v5
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-sco4)
+		os=-sco3.2v4
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-sco3.2.[4-9]*)
+		os=`echo $os | sed -e 's/sco3.2./sco3.2v/'`
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-sco3.2v[4-9]*)
+		# Don't forget version if it is 3.2v4 or newer.
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-sco*)
+		os=-sco3.2v2
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-udk*)
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-isc)
+		os=-isc2.2
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-clix*)
+		basic_machine=clipper-intergraph
+		;;
+	-isc*)
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
+		;;
+	-lynx*)
+		os=-lynxos
+		;;
+	-ptx*)
+		basic_machine=`echo $1 | sed -e 's/86-.*/86-sequent/'`
+		;;
+	-windowsnt*)
+		os=`echo $os | sed -e 's/windowsnt/winnt/'`
+		;;
+	-psos*)
+		os=-psos
+		;;
+	-mint | -mint[0-9]*)
+		basic_machine=m68k-atari
+		os=-mint
+		;;
+esac
+
+# Decode aliases for certain CPU-COMPANY combinations.
+case $basic_machine in
+	# Recognize the basic CPU types without company name.
+	# Some are omitted here because they have special meanings below.
+	1750a | 580 \
+	| a29k \
+	| alpha | alphaev[4-8] | alphaev56 | alphaev6[78] | alphapca5[67] \
+	| alpha64 | alpha64ev[4-8] | alpha64ev56 | alpha64ev6[78] | alpha64pca5[67] \
+	| arc | arm | arm[bl]e | arme[lb] | armv[2345] | armv[345][lb] | avr \
+	| c4x | clipper \
+	| d10v | d30v | dsp16xx \
+	| fr30 \
+	| h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \
+	| i370 | i860 | i960 | ia64 \
+	| m32r | m68000 | m68k | m88k | mcore \
+	| mips | mips16 | mips64 | mips64el | mips64orion | mips64orionel \
+	| mips64vr4100 | mips64vr4100el | mips64vr4300 \
+	| mips64vr4300el | mips64vr5000 | mips64vr5000el \
+	| mipsbe | mipseb | mipsel | mipsle | mipstx39 | mipstx39el \
+	| mipsisa32 | mipsisa64 \
+	| mn10200 | mn10300 \
+	| ns16k | ns32k \
+	| openrisc | or32 \
+	| pdp10 | pdp11 | pj | pjl \
+	| powerpc | powerpc64 | powerpc64le | powerpcle | ppcbe \
+	| pyramid \
+	| sh | sh[34] | sh[34]eb | shbe | shle | sh64 \
+	| sparc | sparc64 | sparc86x | sparclet | sparclite | sparcv9 | sparcv9b \
+	| strongarm \
+	| tahoe | thumb | tic80 | tron \
+	| v850 | v850e \
+	| we32k \
+	| x86 | xscale | xstormy16 | xtensa \
+	| z8k)
+		basic_machine=$basic_machine-unknown
+		;;
+	m6811 | m68hc11 | m6812 | m68hc12)
+		# Motorola 68HC11/12.
+		basic_machine=$basic_machine-unknown
+		os=-none
+		;;
+	m88110 | m680[12346]0 | m683?2 | m68360 | m5200 | v70 | w65 | z8k)
+		;;
+
+	# We use `pc' rather than `unknown'
+	# because (1) that's what they normally are, and
+	# (2) the word "unknown" tends to confuse beginning users.
+	i*86 | x86_64)
+	  basic_machine=$basic_machine-pc
+	  ;;
+	# Object if more than one company name word.
+	*-*-*)
+		echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
+		exit 1
+		;;
+	# Recognize the basic CPU types with company name.
+	580-* \
+	| a29k-* \
+	| alpha-* | alphaev[4-8]-* | alphaev56-* | alphaev6[78]-* \
+	| alpha64-* | alpha64ev[4-8]-* | alpha64ev56-* | alpha64ev6[78]-* \
+	| alphapca5[67]-* | alpha64pca5[67]-* | arc-* \
+	| arm-*  | armbe-* | armle-* | armv*-* \
+	| avr-* \
+	| bs2000-* \
+	| c[123]* | c30-* | [cjt]90-* | c54x-* \
+	| clipper-* | cydra-* \
+	| d10v-* | d30v-* \
+	| elxsi-* \
+	| f30[01]-* | f700-* | fr30-* | fx80-* \
+	| h8300-* | h8500-* \
+	| hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \
+	| i*86-* | i860-* | i960-* | ia64-* \
+	| m32r-* \
+	| m68000-* | m680[012346]0-* | m68360-* | m683?2-* | m68k-* \
+	| m88110-* | m88k-* | mcore-* \
+	| mips-* | mips16-* | mips64-* | mips64el-* | mips64orion-* \
+	| mips64orionel-* | mips64vr4100-* | mips64vr4100el-* \
+	| mips64vr4300-* | mips64vr4300el-* | mipsbe-* | mipseb-* \
+	| mipsle-* | mipsel-* | mipstx39-* | mipstx39el-* \
+	| none-* | np1-* | ns16k-* | ns32k-* \
+	| orion-* \
+	| pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \
+	| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* | ppcbe-* \
+	| pyramid-* \
+	| romp-* | rs6000-* \
+	| sh-* | sh[34]-* | sh[34]eb-* | shbe-* | shle-* | sh64-* \
+	| sparc-* | sparc64-* | sparc86x-* | sparclet-* | sparclite-* \
+	| sparcv9-* | sparcv9b-* | strongarm-* | sv1-* | sx?-* \
+	| tahoe-* | thumb-* | tic30-* | tic54x-* | tic80-* | tron-* \
+	| v850-* | v850e-* | vax-* \
+	| we32k-* \
+	| x86-* | x86_64-* | xps100-* | xscale-* | xstormy16-* \
+	| xtensa-* \
+	| ymp-* \
+	| z8k-*)
+		;;
+	# Recognize the various machine names and aliases which stand
+	# for a CPU type and a company and sometimes even an OS.
+	386bsd)
+		basic_machine=i386-unknown
+		os=-bsd
+		;;
+	3b1 | 7300 | 7300-att | att-7300 | pc7300 | safari | unixpc)
+		basic_machine=m68000-att
+		;;
+	3b*)
+		basic_machine=we32k-att
+		;;
+	a29khif)
+		basic_machine=a29k-amd
+		os=-udi
+		;;
+	adobe68k)
+		basic_machine=m68010-adobe
+		os=-scout
+		;;
+	alliant | fx80)
+		basic_machine=fx80-alliant
+		;;
+	altos | altos3068)
+		basic_machine=m68k-altos
+		;;
+	am29k)
+		basic_machine=a29k-none
+		os=-bsd
+		;;
+	amdahl)
+		basic_machine=580-amdahl
+		os=-sysv
+		;;
+	amiga | amiga-*)
+		basic_machine=m68k-unknown
+		;;
+	amigaos | amigados)
+		basic_machine=m68k-unknown
+		os=-amigaos
+		;;
+	amigaunix | amix)
+		basic_machine=m68k-unknown
+		os=-sysv4
+		;;
+	apollo68)
+		basic_machine=m68k-apollo
+		os=-sysv
+		;;
+	apollo68bsd)
+		basic_machine=m68k-apollo
+		os=-bsd
+		;;
+	aux)
+		basic_machine=m68k-apple
+		os=-aux
+		;;
+	balance)
+		basic_machine=ns32k-sequent
+		os=-dynix
+		;;
+	c90)
+		basic_machine=c90-cray
+		os=-unicos
+		;;
+	convex-c1)
+		basic_machine=c1-convex
+		os=-bsd
+		;;
+	convex-c2)
+		basic_machine=c2-convex
+		os=-bsd
+		;;
+	convex-c32)
+		basic_machine=c32-convex
+		os=-bsd
+		;;
+	convex-c34)
+		basic_machine=c34-convex
+		os=-bsd
+		;;
+	convex-c38)
+		basic_machine=c38-convex
+		os=-bsd
+		;;
+	cray | j90)
+		basic_machine=j90-cray
+		os=-unicos
+		;;
+	crds | unos)
+		basic_machine=m68k-crds
+		;;
+	cris | cris-* | etrax*)
+		basic_machine=cris-axis
+		;;
+	da30 | da30-*)
+		basic_machine=m68k-da30
+		;;
+	decstation | decstation-3100 | pmax | pmax-* | pmin | dec3100 | decstatn)
+		basic_machine=mips-dec
+		;;
+	decsystem10* | dec10*)
+		basic_machine=pdp10-dec
+		os=-tops10
+		;;
+	decsystem20* | dec20*)
+		basic_machine=pdp10-dec
+		os=-tops20
+		;;
+	delta | 3300 | motorola-3300 | motorola-delta \
+	      | 3300-motorola | delta-motorola)
+		basic_machine=m68k-motorola
+		;;
+	delta88)
+		basic_machine=m88k-motorola
+		os=-sysv3
+		;;
+	dpx20 | dpx20-*)
+		basic_machine=rs6000-bull
+		os=-bosx
+		;;
+	dpx2* | dpx2*-bull)
+		basic_machine=m68k-bull
+		os=-sysv3
+		;;
+	ebmon29k)
+		basic_machine=a29k-amd
+		os=-ebmon
+		;;
+	elxsi)
+		basic_machine=elxsi-elxsi
+		os=-bsd
+		;;
+	encore | umax | mmax)
+		basic_machine=ns32k-encore
+		;;
+	es1800 | OSE68k | ose68k | ose | OSE)
+		basic_machine=m68k-ericsson
+		os=-ose
+		;;
+	fx2800)
+		basic_machine=i860-alliant
+		;;
+	genix)
+		basic_machine=ns32k-ns
+		;;
+	gmicro)
+		basic_machine=tron-gmicro
+		os=-sysv
+		;;
+	go32)
+		basic_machine=i386-pc
+		os=-go32
+		;;
+	h3050r* | hiux*)
+		basic_machine=hppa1.1-hitachi
+		os=-hiuxwe2
+		;;
+	h8300hms)
+		basic_machine=h8300-hitachi
+		os=-hms
+		;;
+	h8300xray)
+		basic_machine=h8300-hitachi
+		os=-xray
+		;;
+	h8500hms)
+		basic_machine=h8500-hitachi
+		os=-hms
+		;;
+	harris)
+		basic_machine=m88k-harris
+		os=-sysv3
+		;;
+	hp300-*)
+		basic_machine=m68k-hp
+		;;
+	hp300bsd)
+		basic_machine=m68k-hp
+		os=-bsd
+		;;
+	hp300hpux)
+		basic_machine=m68k-hp
+		os=-hpux
+		;;
+	hp3k9[0-9][0-9] | hp9[0-9][0-9])
+		basic_machine=hppa1.0-hp
+		;;
+	hp9k2[0-9][0-9] | hp9k31[0-9])
+		basic_machine=m68000-hp
+		;;
+	hp9k3[2-9][0-9])
+		basic_machine=m68k-hp
+		;;
+	hp9k6[0-9][0-9] | hp6[0-9][0-9])
+		basic_machine=hppa1.0-hp
+		;;
+	hp9k7[0-79][0-9] | hp7[0-79][0-9])
+		basic_machine=hppa1.1-hp
+		;;
+	hp9k78[0-9] | hp78[0-9])
+		# FIXME: really hppa2.0-hp
+		basic_machine=hppa1.1-hp
+		;;
+	hp9k8[67]1 | hp8[67]1 | hp9k80[24] | hp80[24] | hp9k8[78]9 | hp8[78]9 | hp9k893 | hp893)
+		# FIXME: really hppa2.0-hp
+		basic_machine=hppa1.1-hp
+		;;
+	hp9k8[0-9][13679] | hp8[0-9][13679])
+		basic_machine=hppa1.1-hp
+		;;
+	hp9k8[0-9][0-9] | hp8[0-9][0-9])
+		basic_machine=hppa1.0-hp
+		;;
+	hppa-next)
+		os=-nextstep3
+		;;
+	hppaosf)
+		basic_machine=hppa1.1-hp
+		os=-osf
+		;;
+	hppro)
+		basic_machine=hppa1.1-hp
+		os=-proelf
+		;;
+	i370-ibm* | ibm*)
+		basic_machine=i370-ibm
+		;;
+# I'm not sure what "Sysv32" means.  Should this be sysv3.2?
+	i*86v32)
+		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
+		os=-sysv32
+		;;
+	i*86v4*)
+		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
+		os=-sysv4
+		;;
+	i*86v)
+		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
+		os=-sysv
+		;;
+	i*86sol2)
+		basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
+		os=-solaris2
+		;;
+	i386mach)
+		basic_machine=i386-mach
+		os=-mach
+		;;
+	i386-vsta | vsta)
+		basic_machine=i386-unknown
+		os=-vsta
+		;;
+	iris | iris4d)
+		basic_machine=mips-sgi
+		case $os in
+		    -irix*)
+			;;
+		    *)
+			os=-irix4
+			;;
+		esac
+		;;
+	isi68 | isi)
+		basic_machine=m68k-isi
+		os=-sysv
+		;;
+	m88k-omron*)
+		basic_machine=m88k-omron
+		;;
+	magnum | m3230)
+		basic_machine=mips-mips
+		os=-sysv
+		;;
+	merlin)
+		basic_machine=ns32k-utek
+		os=-sysv
+		;;
+	mingw32)
+		basic_machine=i386-pc
+		os=-mingw32
+		;;
+	miniframe)
+		basic_machine=m68000-convergent
+		;;
+	*mint | -mint[0-9]* | *MiNT | *MiNT[0-9]*)
+		basic_machine=m68k-atari
+		os=-mint
+		;;
+	mips3*-*)
+		basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`
+		;;
+	mips3*)
+		basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`-unknown
+		;;
+	mmix*)
+		basic_machine=mmix-knuth
+		os=-mmixware
+		;;
+	monitor)
+		basic_machine=m68k-rom68k
+		os=-coff
+		;;
+	morphos)
+		basic_machine=powerpc-unknown
+		os=-morphos
+		;;
+	msdos)
+		basic_machine=i386-pc
+		os=-msdos
+		;;
+	mvs)
+		basic_machine=i370-ibm
+		os=-mvs
+		;;
+	ncr3000)
+		basic_machine=i486-ncr
+		os=-sysv4
+		;;
+	netbsd386)
+		basic_machine=i386-unknown
+		os=-netbsd
+		;;
+	netwinder)
+		basic_machine=armv4l-rebel
+		os=-linux
+		;;
+	news | news700 | news800 | news900)
+		basic_machine=m68k-sony
+		os=-newsos
+		;;
+	news1000)
+		basic_machine=m68030-sony
+		os=-newsos
+		;;
+	news-3600 | risc-news)
+		basic_machine=mips-sony
+		os=-newsos
+		;;
+	necv70)
+		basic_machine=v70-nec
+		os=-sysv
+		;;
+	next | m*-next )
+		basic_machine=m68k-next
+		case $os in
+		    -nextstep* )
+			;;
+		    -ns2*)
+		      os=-nextstep2
+			;;
+		    *)
+		      os=-nextstep3
+			;;
+		esac
+		;;
+	nh3000)
+		basic_machine=m68k-harris
+		os=-cxux
+		;;
+	nh[45]000)
+		basic_machine=m88k-harris
+		os=-cxux
+		;;
+	nindy960)
+		basic_machine=i960-intel
+		os=-nindy
+		;;
+	mon960)
+		basic_machine=i960-intel
+		os=-mon960
+		;;
+	nonstopux)
+		basic_machine=mips-compaq
+		os=-nonstopux
+		;;
+	np1)
+		basic_machine=np1-gould
+		;;
+	nsr-tandem)
+		basic_machine=nsr-tandem
+		;;
+	op50n-* | op60c-*)
+		basic_machine=hppa1.1-oki
+		os=-proelf
+		;;
+	or32 | or32-*)
+		basic_machine=or32-unknown
+		os=-coff
+		;;
+	OSE68000 | ose68000)
+		basic_machine=m68000-ericsson
+		os=-ose
+		;;
+	os68k)
+		basic_machine=m68k-none
+		os=-os68k
+		;;
+	pa-hitachi)
+		basic_machine=hppa1.1-hitachi
+		os=-hiuxwe2
+		;;
+	paragon)
+		basic_machine=i860-intel
+		os=-osf
+		;;
+	pbd)
+		basic_machine=sparc-tti
+		;;
+	pbb)
+		basic_machine=m68k-tti
+		;;
+        pc532 | pc532-*)
+		basic_machine=ns32k-pc532
+		;;
+	pentium | p5 | k5 | k6 | nexgen | viac3)
+		basic_machine=i586-pc
+		;;
+	pentiumpro | p6 | 6x86 | athlon)
+		basic_machine=i686-pc
+		;;
+	pentiumii | pentium2)
+		basic_machine=i686-pc
+		;;
+	pentium-* | p5-* | k5-* | k6-* | nexgen-* | viac3-*)
+		basic_machine=i586-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	pentiumpro-* | p6-* | 6x86-* | athlon-*)
+		basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	pentiumii-* | pentium2-*)
+		basic_machine=i686-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	pn)
+		basic_machine=pn-gould
+		;;
+	power)	basic_machine=power-ibm
+		;;
+	ppc)	basic_machine=powerpc-unknown
+	        ;;
+	ppc-*)	basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	ppcle | powerpclittle | ppc-le | powerpc-little)
+		basic_machine=powerpcle-unknown
+	        ;;
+	ppcle-* | powerpclittle-*)
+		basic_machine=powerpcle-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	ppc64)	basic_machine=powerpc64-unknown
+	        ;;
+	ppc64-*) basic_machine=powerpc64-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	ppc64le | powerpc64little | ppc64-le | powerpc64-little)
+		basic_machine=powerpc64le-unknown
+	        ;;
+	ppc64le-* | powerpc64little-*)
+		basic_machine=powerpc64le-`echo $basic_machine | sed 's/^[^-]*-//'`
+		;;
+	ps2)
+		basic_machine=i386-ibm
+		;;
+	pw32)
+		basic_machine=i586-unknown
+		os=-pw32
+		;;
+	rom68k)
+		basic_machine=m68k-rom68k
+		os=-coff
+		;;
+	rm[46]00)
+		basic_machine=mips-siemens
+		;;
+	rtpc | rtpc-*)
+		basic_machine=romp-ibm
+		;;
+	s390 | s390-*)
+		basic_machine=s390-ibm
+		;;
+	s390x | s390x-*)
+		basic_machine=s390x-ibm
+		;;
+	sa29200)
+		basic_machine=a29k-amd
+		os=-udi
+		;;
+########################
+# changes for Apache
+#
+	as400*)
+		basic_machine=as400-ibm
+		;;
+# 
+# end Apache changes
+########################
+	sequent)
+		basic_machine=i386-sequent
+		;;
+	sh)
+		basic_machine=sh-hitachi
+		os=-hms
+		;;
+	sparclite-wrs | simso-wrs)
+		basic_machine=sparclite-wrs
+		os=-vxworks
+		;;
+	sps7)
+		basic_machine=m68k-bull
+		os=-sysv2
+		;;
+	spur)
+		basic_machine=spur-unknown
+		;;
+	st2000)
+		basic_machine=m68k-tandem
+		;;
+	stratus)
+		basic_machine=i860-stratus
+		os=-sysv4
+		;;
+	sun2)
+		basic_machine=m68000-sun
+		;;
+	sun2os3)
+		basic_machine=m68000-sun
+		os=-sunos3
+		;;
+	sun2os4)
+		basic_machine=m68000-sun
+		os=-sunos4
+		;;
+	sun3os3)
+		basic_machine=m68k-sun
+		os=-sunos3
+		;;
+	sun3os4)
+		basic_machine=m68k-sun
+		os=-sunos4
+		;;
+	sun4os3)
+		basic_machine=sparc-sun
+		os=-sunos3
+		;;
+	sun4os4)
+		basic_machine=sparc-sun
+		os=-sunos4
+		;;
+	sun4sol2)
+		basic_machine=sparc-sun
+		os=-solaris2
+		;;
+	sun3 | sun3-*)
+		basic_machine=m68k-sun
+		;;
+	sun4)
+		basic_machine=sparc-sun
+		;;
+	sun386 | sun386i | roadrunner)
+		basic_machine=i386-sun
+		;;
+        sv1)
+		basic_machine=sv1-cray
+		os=-unicos
+		;;
+	symmetry)
+		basic_machine=i386-sequent
+		os=-dynix
+		;;
+	t3d)
+		basic_machine=alpha-cray
+		os=-unicos
+		;;
+	t3e)
+		basic_machine=alphaev5-cray
+		os=-unicos
+		;;
+	t90)
+		basic_machine=t90-cray
+		os=-unicos
+		;;
+	tic54x | c54x*)
+		basic_machine=tic54x-unknown
+		os=-coff
+		;;
+	tx39)
+		basic_machine=mipstx39-unknown
+		;;
+	tx39el)
+		basic_machine=mipstx39el-unknown
+		;;
+	toad1)
+		basic_machine=pdp10-xkl
+		os=-tops20
+		;;
+	tower | tower-32)
+		basic_machine=m68k-ncr
+		;;
+	udi29k)
+		basic_machine=a29k-amd
+		os=-udi
+		;;
+	ultra3)
+		basic_machine=a29k-nyu
+		os=-sym1
+		;;
+	v810 | necv810)
+		basic_machine=v810-nec
+		os=-none
+		;;
+	vaxv)
+		basic_machine=vax-dec
+		os=-sysv
+		;;
+	vms)
+		basic_machine=vax-dec
+		os=-vms
+		;;
+	vpp*|vx|vx-*)
+               basic_machine=f301-fujitsu
+               ;;
+	vxworks960)
+		basic_machine=i960-wrs
+		os=-vxworks
+		;;
+	vxworks68)
+		basic_machine=m68k-wrs
+		os=-vxworks
+		;;
+	vxworks29k)
+		basic_machine=a29k-wrs
+		os=-vxworks
+		;;
+	w65*)
+		basic_machine=w65-wdc
+		os=-none
+		;;
+	w89k-*)
+		basic_machine=hppa1.1-winbond
+		os=-proelf
+		;;
+	windows32)
+		basic_machine=i386-pc
+		os=-windows32-msvcrt
+		;;
+        xps | xps100)
+		basic_machine=xps100-honeywell
+		;;
+	ymp)
+		basic_machine=ymp-cray
+		os=-unicos
+		;;
+	z8k-*-coff)
+		basic_machine=z8k-unknown
+		os=-sim
+		;;
+	none)
+		basic_machine=none-none
+		os=-none
+		;;
+
+# Here we handle the default manufacturer of certain CPU types.  It is in
+# some cases the only manufacturer, in others, it is the most popular.
+	w89k)
+		basic_machine=hppa1.1-winbond
+		;;
+	op50n)
+		basic_machine=hppa1.1-oki
+		;;
+	op60c)
+		basic_machine=hppa1.1-oki
+		;;
+	romp)
+		basic_machine=romp-ibm
+		;;
+	rs6000)
+		basic_machine=rs6000-ibm
+		;;
+	vax)
+		basic_machine=vax-dec
+		;;
+	pdp10)
+		# there are many clones, so DEC is not a safe bet
+		basic_machine=pdp10-unknown
+		;;
+	pdp11)
+		basic_machine=pdp11-dec
+		;;
+	we32k)
+		basic_machine=we32k-att
+		;;
+	sh3 | sh4 | sh3eb | sh4eb)
+		basic_machine=sh-unknown
+		;;
+	sh64)
+		basic_machine=sh64-unknown
+		;;
+	sparc | sparcv9 | sparcv9b)
+		basic_machine=sparc-sun
+		;;
+        cydra)
+		basic_machine=cydra-cydrome
+		;;
+	orion)
+		basic_machine=orion-highlevel
+		;;
+	orion105)
+		basic_machine=clipper-highlevel
+		;;
+	mac | mpw | mac-mpw)
+		basic_machine=m68k-apple
+		;;
+	pmac | pmac-mpw)
+		basic_machine=powerpc-apple
+		;;
+	c4x*)
+		basic_machine=c4x-none
+		os=-coff
+		;;
+	*-unknown)
+		# Make sure to match an already-canonicalized machine name.
+		;;
+	*)
+		echo Invalid configuration \`$1\': machine \`$basic_machine\' not recognized 1>&2
+		exit 1
+		;;
+esac
+
+# Here we canonicalize certain aliases for manufacturers.
+case $basic_machine in
+	*-digital*)
+		basic_machine=`echo $basic_machine | sed 's/digital.*/dec/'`
+		;;
+	*-commodore*)
+		basic_machine=`echo $basic_machine | sed 's/commodore.*/cbm/'`
+		;;
+	*)
+		;;
+esac
+
+# Decode manufacturer-specific aliases for certain operating systems.
+
+if [ x"$os" != x"" ]
+then
+case $os in
+        # First match some system type aliases
+        # that might get confused with valid system types.
+	# -solaris* is a basic system type, with this one exception.
+	-solaris1 | -solaris1.*)
+		os=`echo $os | sed -e 's|solaris1|sunos4|'`
+		;;
+	-solaris)
+		os=-solaris2
+		;;
+	-svr4*)
+		os=-sysv4
+		;;
+	-unixware*)
+		os=-sysv4.2uw
+		;;
+	-gnu/linux*)
+		os=`echo $os | sed -e 's|gnu/linux|linux-gnu|'`
+		;;
+########################
+# changes for Apache
+#
+	-os2_emx | -tpf* | -os390* | -vmcms* | -os400* )
+        	;;
+#
+# end Apache changes
+########################
+	# First accept the basic system types.
+	# The portable systems comes first.
+	# Each alternative MUST END IN A *, to match a version number.
+	# -sysv* is not here because it comes later, after sysvr4.
+	-gnu* | -bsd* | -mach* | -minix* | -genix* | -ultrix* | -irix* \
+	      | -*vms* | -sco* | -esix* | -isc* | -aix* | -sunos | -sunos[34]*\
+	      | -hpux* | -unos* | -osf* | -luna* | -dgux* | -solaris* | -sym* \
+	      | -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \
+	      | -aos* \
+	      | -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \
+	      | -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \
+	      | -hiux* | -386bsd* | -netbsd* | -openbsd* | -freebsd* | -riscix* \
+	      | -lynxos* | -bosx* | -nextstep* | -cxux* | -aout* | -elf* | -oabi* \
+	      | -ptx* | -coff* | -ecoff* | -winnt* | -domain* | -vsta* \
+	      | -udi* | -eabi* | -lites* | -ieee* | -go32* | -aux* \
+	      | -chorusos* | -chorusrdb* \
+	      | -cygwin* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
+	      | -mingw32* | -linux-gnu* | -uxpv* | -beos* | -mpeix* | -udk* \
+	      | -interix* | -uwin* | -rhapsody* | -darwin* | -opened* \
+	      | -openstep* | -oskit* | -conix* | -pw32* | -nonstopux* \
+	      | -storm-chaos* | -tops10* | -tenex* | -tops20* | -its* \
+	      | -os2* | -vos* | -palmos* | -uclinux* | -nucleus* \
+	      | -morphos* | -superux* | -rtmk* | -rtmk-nova*)
+	# Remember, each alternative MUST END IN *, to match a version number.
+		;;
+	-qnx*)
+		case $basic_machine in
+		    x86-* | i*86-*)
+			;;
+		    *)
+			os=-nto$os
+			;;
+		esac
+		;;
+	-nto*)
+		os=-nto-qnx
+		;;
+	-sim | -es1800* | -hms* | -xray | -os68k* | -none* | -v88r* \
+	      | -windows* | -osx | -abug | -netware* | -os9* | -beos* \
+	      | -macos* | -mpw* | -magic* | -mmixware* | -mon960* | -lnews*)
+		;;
+	-mac*)
+		os=`echo $os | sed -e 's|mac|macos|'`
+		;;
+	-linux*)
+		os=`echo $os | sed -e 's|linux|linux-gnu|'`
+		;;
+	-sunos5*)
+		os=`echo $os | sed -e 's|sunos5|solaris2|'`
+		;;
+	-sunos6*)
+		os=`echo $os | sed -e 's|sunos6|solaris3|'`
+		;;
+	-opened*)
+		os=-openedition
+		;;
+	-wince*)
+		os=-wince
+		;;
+	-osfrose*)
+		os=-osfrose
+		;;
+	-osf*)
+		os=-osf
+		;;
+	-utek*)
+		os=-bsd
+		;;
+	-dynix*)
+		os=-bsd
+		;;
+	-acis*)
+		os=-aos
+		;;
+	-atheos*)
+		os=-atheos
+		;;
+	-386bsd)
+		os=-bsd
+		;;
+	-ctix* | -uts*)
+		os=-sysv
+		;;
+	-nova*)
+		os=-rtmk-nova
+		;;
+	-ns2 )
+	        os=-nextstep2
+		;;
+	-nsk*)
+		os=-nsk
+		;;
+	# Preserve the version number of sinix5.
+	-sinix5.*)
+		os=`echo $os | sed -e 's|sinix|sysv|'`
+		;;
+	-sinix*)
+		os=-sysv4
+		;;
+	-triton*)
+		os=-sysv3
+		;;
+	-oss*)
+		os=-sysv3
+		;;
+	-svr4)
+		os=-sysv4
+		;;
+	-svr3)
+		os=-sysv3
+		;;
+	-sysvr4)
+		os=-sysv4
+		;;
+	# This must come after -sysvr4.
+	-sysv*)
+		;;
+	-ose*)
+		os=-ose
+		;;
+	-es1800*)
+		os=-ose
+		;;
+	-xenix)
+		os=-xenix
+		;;
+        -*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
+	        os=-mint
+		;;
+	-none)
+		;;
+	*)
+		# Get rid of the `-' at the beginning of $os.
+		os=`echo $os | sed 's/[^-]*-//'`
+		echo Invalid configuration \`$1\': system \`$os\' not recognized 1>&2
+		exit 1
+		;;
+esac
+else
+
+# Here we handle the default operating systems that come with various machines.
+# The value should be what the vendor currently ships out the door with their
+# machine or put another way, the most popular os provided with the machine.
+
+# Note that if you're going to try to match "-MANUFACTURER" here (say,
+# "-sun"), then you have to tell the case statement up towards the top
+# that MANUFACTURER isn't an operating system.  Otherwise, code above
+# will signal an error saying that MANUFACTURER isn't an operating
+# system, and we'll never get to this point.
+
+case $basic_machine in
+	*-acorn)
+		os=-riscix1.2
+		;;
+	arm*-rebel)
+		os=-linux
+		;;
+	arm*-semi)
+		os=-aout
+		;;
+	# This must come before the *-dec entry.
+	pdp10-*)
+		os=-tops20
+		;;
+        pdp11-*)
+		os=-none
+		;;
+	*-dec | vax-*)
+		os=-ultrix4.2
+		;;
+	m68*-apollo)
+		os=-domain
+		;;
+	i386-sun)
+		os=-sunos4.0.2
+		;;
+	m68000-sun)
+		os=-sunos3
+		# This also exists in the configure program, but was not the
+		# default.
+		# os=-sunos4
+		;;
+	m68*-cisco)
+		os=-aout
+		;;
+	mips*-cisco)
+		os=-elf
+		;;
+	mips*-*)
+		os=-elf
+		;;
+	or32-*)
+		os=-coff
+		;;
+	*-tti)	# must be before sparc entry or we get the wrong os.
+		os=-sysv3
+		;;
+	sparc-* | *-sun)
+		os=-sunos4.1.1
+		;;
+	*-be)
+		os=-beos
+		;;
+########################
+# changes for Apache
+#
+#	*-ibm)
+#		os=-aix
+#		;;
+#
+	*-ibm)
+		case $basic_machine in
+			s390*)
+				os=-os390;
+				;;
+			i370*)
+				os=-mvs;
+				;;
+			as400*)
+				os=-os400;
+				;;
+			*)
+				os=-aix
+				;;
+		esac
+                ;;
+# 
+# end Apache changes
+########################
+
+	*-wec)
+		os=-proelf
+		;;
+	*-winbond)
+		os=-proelf
+		;;
+	*-oki)
+		os=-proelf
+		;;
+	*-hp)
+		os=-hpux
+		;;
+	*-hitachi)
+		os=-hiux
+		;;
+	i860-* | *-att | *-ncr | *-altos | *-motorola | *-convergent)
+		os=-sysv
+		;;
+	*-cbm)
+		os=-amigaos
+		;;
+	*-dg)
+		os=-dgux
+		;;
+	*-dolphin)
+		os=-sysv3
+		;;
+	m68k-ccur)
+		os=-rtu
+		;;
+	m88k-omron*)
+		os=-luna
+		;;
+	*-next )
+		os=-nextstep
+		;;
+	*-sequent)
+		os=-ptx
+		;;
+	*-crds)
+		os=-unos
+		;;
+	*-ns)
+		os=-genix
+		;;
+	i370-*)
+		os=-mvs
+		;;
+	*-next)
+		os=-nextstep3
+		;;
+        *-gould)
+		os=-sysv
+		;;
+        *-highlevel)
+		os=-bsd
+		;;
+	*-encore)
+		os=-bsd
+		;;
+        *-sgi)
+		os=-irix
+		;;
+        *-siemens)
+		os=-sysv4
+		;;
+	*-masscomp)
+		os=-rtu
+		;;
+	f30[01]-fujitsu | f700-fujitsu)
+		os=-uxpv
+		;;
+	*-rom68k)
+		os=-coff
+		;;
+	*-*bug)
+		os=-coff
+		;;
+	*-apple)
+		os=-macos
+		;;
+	*-atari*)
+		os=-mint
+		;;
+	*)
+		os=-none
+		;;
+esac
+fi
+
+# Here we handle the case where we know the os, and the CPU type, but not the
+# manufacturer.  We pick the logical manufacturer.
+vendor=unknown
+case $basic_machine in
+	*-unknown)
+		case $os in
+			-riscix*)
+				vendor=acorn
+				;;
+			-sunos*)
+				vendor=sun
+				;;
+			-aix*)
+				vendor=ibm
+				;;
+			-beos*)
+				vendor=be
+				;;
+			-hpux*)
+				vendor=hp
+				;;
+			-mpeix*)
+				vendor=hp
+				;;
+			-hiux*)
+				vendor=hitachi
+				;;
+			-unos*)
+				vendor=crds
+				;;
+			-dgux*)
+				vendor=dg
+				;;
+			-luna*)
+				vendor=omron
+				;;
+			-genix*)
+				vendor=ns
+				;;
+			-mvs* | -opened*)
+				vendor=ibm
+				;;
+			-ptx*)
+				vendor=sequent
+				;;
+			-vxsim* | -vxworks*)
+				vendor=wrs
+				;;
+			-aux*)
+				vendor=apple
+				;;
+			-hms*)
+				vendor=hitachi
+				;;
+			-mpw* | -macos*)
+				vendor=apple
+				;;
+			-*mint | -mint[0-9]* | -*MiNT | -MiNT[0-9]*)
+				vendor=atari
+				;;
+			-vos*)
+				vendor=stratus
+				;;
+		esac
+		basic_machine=`echo $basic_machine | sed "s/unknown/$vendor/"`
+		;;
+esac
+
+echo $basic_machine$os
+exit 0
+
+# Local variables:
+# eval: (add-hook 'write-file-hooks 'time-stamp)
+# time-stamp-start: "timestamp='"
+# time-stamp-format: "%:y-%02m-%02d"
+# time-stamp-end: "'"
+# End:
diff --git a/xml/expat/conftools/install-sh b/xml/expat/conftools/install-sh
new file mode 100755
index 000000000..e9de23842
--- /dev/null
+++ b/xml/expat/conftools/install-sh
@@ -0,0 +1,251 @@
+#!/bin/sh
+#
+# install - install a program, script, or datafile
+# This comes from X11R5 (mit/util/scripts/install.sh).
+#
+# Copyright 1991 by the Massachusetts Institute of Technology
+#
+# Permission to use, copy, modify, distribute, and sell this software and its
+# documentation for any purpose is hereby granted without fee, provided that
+# the above copyright notice appear in all copies and that both that
+# copyright notice and this permission notice appear in supporting
+# documentation, and that the name of M.I.T. not be used in advertising or
+# publicity pertaining to distribution of the software without specific,
+# written prior permission.  M.I.T. makes no representations about the
+# suitability of this software for any purpose.  It is provided "as is"
+# without express or implied warranty.
+#
+# Calling this script install-sh is preferred over install.sh, to prevent
+# `make' implicit rules from creating a file called install from it
+# when there is no Makefile.
+#
+# This script is compatible with the BSD install script, but was written
+# from scratch.  It can only install one file at a time, a restriction
+# shared with many OS's install programs.
+
+
+# set DOITPROG to echo to test this script
+
+# Don't use :- since 4.3BSD and earlier shells don't like it.
+doit="${DOITPROG-}"
+
+
+# put in absolute paths if you don't have them in your path; or use env. vars.
+
+mvprog="${MVPROG-mv}"
+cpprog="${CPPROG-cp}"
+chmodprog="${CHMODPROG-chmod}"
+chownprog="${CHOWNPROG-chown}"
+chgrpprog="${CHGRPPROG-chgrp}"
+stripprog="${STRIPPROG-strip}"
+rmprog="${RMPROG-rm}"
+mkdirprog="${MKDIRPROG-mkdir}"
+
+transformbasename=""
+transform_arg=""
+instcmd="$mvprog"
+chmodcmd="$chmodprog 0755"
+chowncmd=""
+chgrpcmd=""
+stripcmd=""
+rmcmd="$rmprog -f"
+mvcmd="$mvprog"
+src=""
+dst=""
+dir_arg=""
+
+while [ x"$1" != x ]; do
+    case $1 in
+	-c) instcmd="$cpprog"
+	    shift
+	    continue;;
+
+	-d) dir_arg=true
+	    shift
+	    continue;;
+
+	-m) chmodcmd="$chmodprog $2"
+	    shift
+	    shift
+	    continue;;
+
+	-o) chowncmd="$chownprog $2"
+	    shift
+	    shift
+	    continue;;
+
+	-g) chgrpcmd="$chgrpprog $2"
+	    shift
+	    shift
+	    continue;;
+
+	-s) stripcmd="$stripprog"
+	    shift
+	    continue;;
+
+	-t=*) transformarg=`echo $1 | sed 's/-t=//'`
+	    shift
+	    continue;;
+
+	-b=*) transformbasename=`echo $1 | sed 's/-b=//'`
+	    shift
+	    continue;;
+
+	*)  if [ x"$src" = x ]
+	    then
+		src=$1
+	    else
+		# this colon is to work around a 386BSD /bin/sh bug
+		:
+		dst=$1
+	    fi
+	    shift
+	    continue;;
+    esac
+done
+
+if [ x"$src" = x ]
+then
+	echo "install:	no input file specified"
+	exit 1
+else
+	true
+fi
+
+if [ x"$dir_arg" != x ]; then
+	dst=$src
+	src=""
+	
+	if [ -d $dst ]; then
+		instcmd=:
+		chmodcmd=""
+	else
+		instcmd=mkdir
+	fi
+else
+
+# Waiting for this to be detected by the "$instcmd $src $dsttmp" command
+# might cause directories to be created, which would be especially bad 
+# if $src (and thus $dsttmp) contains '*'.
+
+	if [ -f $src -o -d $src ]
+	then
+		true
+	else
+		echo "install:  $src does not exist"
+		exit 1
+	fi
+	
+	if [ x"$dst" = x ]
+	then
+		echo "install:	no destination specified"
+		exit 1
+	else
+		true
+	fi
+
+# If destination is a directory, append the input filename; if your system
+# does not like double slashes in filenames, you may need to add some logic
+
+	if [ -d $dst ]
+	then
+		dst="$dst"/`basename $src`
+	else
+		true
+	fi
+fi
+
+## this sed command emulates the dirname command
+dstdir=`echo $dst | sed -e 's,[^/]*$,,;s,/$,,;s,^$,.,'`
+
+# Make sure that the destination directory exists.
+#  this part is taken from Noah Friedman's mkinstalldirs script
+
+# Skip lots of stat calls in the usual case.
+if [ ! -d "$dstdir" ]; then
+defaultIFS='	
+'
+IFS="${IFS-${defaultIFS}}"
+
+oIFS="${IFS}"
+# Some sh's can't handle IFS=/ for some reason.
+IFS='%'
+set - `echo ${dstdir} | sed -e 's@/@%@g' -e 's@^%@/@'`
+IFS="${oIFS}"
+
+pathcomp=''
+
+while [ $# -ne 0 ] ; do
+	pathcomp="${pathcomp}${1}"
+	shift
+
+	if [ ! -d "${pathcomp}" ] ;
+        then
+		$mkdirprog "${pathcomp}"
+	else
+		true
+	fi
+
+	pathcomp="${pathcomp}/"
+done
+fi
+
+if [ x"$dir_arg" != x ]
+then
+	$doit $instcmd $dst &&
+
+	if [ x"$chowncmd" != x ]; then $doit $chowncmd $dst; else true ; fi &&
+	if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dst; else true ; fi &&
+	if [ x"$stripcmd" != x ]; then $doit $stripcmd $dst; else true ; fi &&
+	if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dst; else true ; fi
+else
+
+# If we're going to rename the final executable, determine the name now.
+
+	if [ x"$transformarg" = x ] 
+	then
+		dstfile=`basename $dst`
+	else
+		dstfile=`basename $dst $transformbasename | 
+			sed $transformarg`$transformbasename
+	fi
+
+# don't allow the sed command to completely eliminate the filename
+
+	if [ x"$dstfile" = x ] 
+	then
+		dstfile=`basename $dst`
+	else
+		true
+	fi
+
+# Make a temp file name in the proper directory.
+
+	dsttmp=$dstdir/#inst.$$#
+
+# Move or copy the file name to the temp name
+
+	$doit $instcmd $src $dsttmp &&
+
+	trap "rm -f ${dsttmp}" 0 &&
+
+# and set any options; do chmod last to preserve setuid bits
+
+# If any of these fail, we abort the whole thing.  If we want to
+# ignore errors from any of these, just make sure not to ignore
+# errors from the above "$doit $instcmd $src $dsttmp" command.
+
+	if [ x"$chowncmd" != x ]; then $doit $chowncmd $dsttmp; else true;fi &&
+	if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dsttmp; else true;fi &&
+	if [ x"$stripcmd" != x ]; then $doit $stripcmd $dsttmp; else true;fi &&
+	if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dsttmp; else true;fi &&
+
+# Now rename the file to the real destination.
+
+	$doit $rmcmd -f $dstdir/$dstfile &&
+	$doit $mvcmd $dsttmp $dstdir/$dstfile 
+
+fi &&
+
+
+exit 0
diff --git a/xml/expat/conftools/missing b/xml/expat/conftools/missing
new file mode 100755
index 000000000..7789652e8
--- /dev/null
+++ b/xml/expat/conftools/missing
@@ -0,0 +1,190 @@
+#! /bin/sh
+# Common stub for a few missing GNU programs while installing.
+# Copyright (C) 1996, 1997 Free Software Foundation, Inc.
+# Franc,ois Pinard <pinard@iro.umontreal.ca>, 1996.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2, or (at your option)
+# any later version.
+
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
+# 02111-1307, USA.
+
+if test $# -eq 0; then
+  echo 1>&2 "Try \`$0 --help' for more information"
+  exit 1
+fi
+
+case "$1" in
+
+  -h|--h|--he|--hel|--help)
+    echo "\
+$0 [OPTION]... PROGRAM [ARGUMENT]...
+
+Handle \`PROGRAM [ARGUMENT]...' for when PROGRAM is missing, or return an
+error status if there is no known handling for PROGRAM.
+
+Options:
+  -h, --help      display this help and exit
+  -v, --version   output version information and exit
+
+Supported PROGRAM values:
+  aclocal      touch file \`aclocal.m4'
+  autoconf     touch file \`configure'
+  autoheader   touch file \`config.h.in'
+  automake     touch all \`Makefile.in' files
+  bison        create \`y.tab.[ch]', if possible, from existing .[ch]
+  flex         create \`lex.yy.c', if possible, from existing .c
+  lex          create \`lex.yy.c', if possible, from existing .c
+  makeinfo     touch the output file
+  yacc         create \`y.tab.[ch]', if possible, from existing .[ch]"
+    ;;
+
+  -v|--v|--ve|--ver|--vers|--versi|--versio|--version)
+    echo "missing - GNU libit 0.0"
+    ;;
+
+  -*)
+    echo 1>&2 "$0: Unknown \`$1' option"
+    echo 1>&2 "Try \`$0 --help' for more information"
+    exit 1
+    ;;
+
+  aclocal)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified \`acinclude.m4' or \`configure.in'.  You might want
+         to install the \`Automake' and \`Perl' packages.  Grab them from
+         any GNU archive site."
+    touch aclocal.m4
+    ;;
+
+  autoconf)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified \`configure.in'.  You might want to install the
+         \`Autoconf' and \`GNU m4' packages.  Grab them from any GNU
+         archive site."
+    touch configure
+    ;;
+
+  autoheader)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified \`acconfig.h' or \`configure.in'.  You might want
+         to install the \`Autoconf' and \`GNU m4' packages.  Grab them
+         from any GNU archive site."
+    files=`sed -n 's/^[ ]*A[CM]_CONFIG_HEADER(\([^)]*\)).*/\1/p' configure.in`
+    test -z "$files" && files="config.h"
+    touch_files=
+    for f in $files; do
+      case "$f" in
+      *:*) touch_files="$touch_files "`echo "$f" |
+				       sed -e 's/^[^:]*://' -e 's/:.*//'`;;
+      *) touch_files="$touch_files $f.in";;
+      esac
+    done
+    touch $touch_files
+    ;;
+
+  automake)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified \`Makefile.am', \`acinclude.m4' or \`configure.in'.
+         You might want to install the \`Automake' and \`Perl' packages.
+         Grab them from any GNU archive site."
+    find . -type f -name Makefile.am -print |
+	   sed 's/\.am$/.in/' |
+	   while read f; do touch "$f"; done
+    ;;
+
+  bison|yacc)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified a \`.y' file.  You may need the \`Bison' package
+         in order for those modifications to take effect.  You can get
+         \`Bison' from any GNU archive site."
+    rm -f y.tab.c y.tab.h
+    if [ $# -ne 1 ]; then
+        eval LASTARG="\${$#}"
+	case "$LASTARG" in
+	*.y)
+	    SRCFILE=`echo "$LASTARG" | sed 's/y$/c/'`
+	    if [ -f "$SRCFILE" ]; then
+	         cp "$SRCFILE" y.tab.c
+	    fi
+	    SRCFILE=`echo "$LASTARG" | sed 's/y$/h/'`
+	    if [ -f "$SRCFILE" ]; then
+	         cp "$SRCFILE" y.tab.h
+	    fi
+	  ;;
+	esac
+    fi
+    if [ ! -f y.tab.h ]; then
+	echo >y.tab.h
+    fi
+    if [ ! -f y.tab.c ]; then
+	echo 'main() { return 0; }' >y.tab.c
+    fi
+    ;;
+
+  lex|flex)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified a \`.l' file.  You may need the \`Flex' package
+         in order for those modifications to take effect.  You can get
+         \`Flex' from any GNU archive site."
+    rm -f lex.yy.c
+    if [ $# -ne 1 ]; then
+        eval LASTARG="\${$#}"
+	case "$LASTARG" in
+	*.l)
+	    SRCFILE=`echo "$LASTARG" | sed 's/l$/c/'`
+	    if [ -f "$SRCFILE" ]; then
+	         cp "$SRCFILE" lex.yy.c
+	    fi
+	  ;;
+	esac
+    fi
+    if [ ! -f lex.yy.c ]; then
+	echo 'main() { return 0; }' >lex.yy.c
+    fi
+    ;;
+
+  makeinfo)
+    echo 1>&2 "\
+WARNING: \`$1' is missing on your system.  You should only need it if
+         you modified a \`.texi' or \`.texinfo' file, or any other file
+         indirectly affecting the aspect of the manual.  The spurious
+         call might also be the consequence of using a buggy \`make' (AIX,
+         DU, IRIX).  You might want to install the \`Texinfo' package or
+         the \`GNU make' package.  Grab either from any GNU archive site."
+    file=`echo "$*" | sed -n 's/.*-o \([^ ]*\).*/\1/p'`
+    if test -z "$file"; then
+      file=`echo "$*" | sed 's/.* \([^ ]*\) *$/\1/'`
+      file=`sed -n '/^@setfilename/ { s/.* \([^ ]*\) *$/\1/; p; q; }' $file`
+    fi
+    touch $file
+    ;;
+
+  *)
+    echo 1>&2 "\
+WARNING: \`$1' is needed, and you do not seem to have it handy on your
+         system.  You might have modified some files without having the
+         proper tools for further handling them.  Check the \`README' file,
+         it often tells you about the needed prerequirements for installing
+         this package.  You may also peek at any GNU archive site, in case
+         some other package would contain this missing \`$1' program."
+    exit 1
+    ;;
+esac
+
+exit 0
diff --git a/xml/expat/conftools/mkinstalldirs b/xml/expat/conftools/mkinstalldirs
new file mode 100755
index 000000000..6b3b5fc5d
--- /dev/null
+++ b/xml/expat/conftools/mkinstalldirs
@@ -0,0 +1,40 @@
+#! /bin/sh
+# mkinstalldirs --- make directory hierarchy
+# Author: Noah Friedman <friedman@prep.ai.mit.edu>
+# Created: 1993-05-16
+# Public domain
+
+# $Id$
+
+errstatus=0
+
+for file
+do
+   set fnord `echo ":$file" | sed -ne 's/^:\//#/;s/^://;s/\// /g;s/^#/\//;p'`
+   shift
+
+   pathcomp=
+   for d
+   do
+     pathcomp="$pathcomp$d"
+     case "$pathcomp" in
+       -* ) pathcomp=./$pathcomp ;;
+     esac
+
+     if test ! -d "$pathcomp"; then
+        echo "mkdir $pathcomp"
+
+        mkdir "$pathcomp" || lasterr=$?
+
+        if test ! -d "$pathcomp"; then
+  	  errstatus=$lasterr
+        fi
+     fi
+
+     pathcomp="$pathcomp/"
+   done
+done
+
+exit $errstatus
+
+# mkinstalldirs ends here
diff --git a/xml/expat/lib/Makefile.in b/xml/expat/lib/Makefile.in
new file mode 100644
index 000000000..b5e4b3def
--- /dev/null
+++ b/xml/expat/lib/Makefile.in
@@ -0,0 +1,154 @@
+################################################################
+# Process this file with top-level configure script to produce Makefile
+#
+# Copyright 2000 Clark Cooper
+#
+#  This file is part of EXPAT.
+#
+#  EXPAT is free software; you can redistribute it and/or modify it
+#  under the terms of the License (based on the MIT/X license) contained
+#  in the file COPYING that comes with this distribution.
+#
+# EXPAT IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
+# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
+# SOFTWARE OR THE USE OR OTHER DEALINGS IN EXPAT.
+#
+
+
+SHELL = @SHELL@
+
+srcdir = @srcdir@
+top_srcdir = @top_srcdir@
+VPATH = @srcdir@
+prefix = @prefix@
+exec_prefix = @exec_prefix@
+
+blddir = @blddir@/lib
+bindir = @bindir@
+sbindir = @sbindir@
+libexecdir = @libexecdir@
+datadir = @datadir@
+sysconfdir = @sysconfdir@
+sharedstatedir = @sharedstatedir@
+localstatedir = @localstatedir@
+libdir = @libdir@
+infodir = @infodir@
+mandir = @mandir@
+includedir = @includedir@
+oldincludedir = /usr/include
+
+subdir = lib
+
+top_builddir = ..
+
+INSTALL = @INSTALL@
+INSTALL_PROGRAM = @INSTALL_PROGRAM@
+INSTALL_SCRIPT = @INSTALL_SCRIPT@
+INSTALL_DATA = @INSTALL_DATA@
+
+host_alias = @host_alias@
+host_triplet = @host@
+AS = @AS@
+CC = @CC@
+DLLTOOL = @DLLTOOL@
+LIBTOOL = @LIBTOOL@
+LN_S = @LN_S@
+OBJDUMP = @OBJDUMP@
+PACKAGE = @PACKAGE@
+RANLIB = @RANLIB@
+VERSION = @VERSION@
+
+LIBRARY = libexpat.la
+SOURCES = xmlparse.c xmltok.c xmlrole.c
+OBJECTS = $(SOURCES:.c=.o)
+LTOBJECTS = $(SOURCES:.c=.lo)
+
+TEMPLATES = xmltok_impl.c xmltok_ns.c
+APIHEADER = expat.h
+HEADERS = ascii.h iasciitab.h utf8tab.h xmltok.h asciitab.h latin1tab.h \
+          nametab.h xmldef.h xmlrole.h xmltok_impl.h
+
+mkinstalldirs = $(SHELL) $(top_srcdir)/conftools/mkinstalldirs
+CONFIG_HEADER = ../config.h
+CONFIG_CLEAN_FILES = 
+
+INCLUDES = -I$(srcdir) -I.. -I$(blddir)
+DEFS = @DEFS@ -DPACKAGE='"$(PACKAGE)"' -DVERSION='"$(PACKAGE)_$(VERSION)"'
+
+CPPFLAGS = @CPPFLAGS@
+LDFLAGS = @LDFLAGS@
+LIBS = @LIBS@
+CFLAGS = @CFLAGS@
+
+LIBREVISION = @LIBREVISION@
+LIBCURRENT  = @LIBCURRENT@
+LIBAGE      = @LIBAGE@
+
+COMPILE = $(CC) $(DEFS) $(INCLUDES) $(CPPFLAGS) $(CFLAGS)
+LTCOMPILE = $(LIBTOOL) --mode=compile $(CC) $(DEFS) $(INCLUDES) $(CPPFLAGS) $(CFLAGS)
+CCLD = $(CC)
+LINK = $(LIBTOOL) --mode=link  $(CCLD) -version-info $(LIBCURRENT):$(LIBREVISION):$(LIBAGE) $(CFLAGS) $(LDFLAGS) -o $@
+DIST_COMMON =  Makefile.in
+
+
+DISTFILES = $(DIST_COMMON) $(SOURCES) $(TEMPLATES) $(APIHEADER) $(HEADERS) 
+
+TAR = gtar
+GZIP_ENV = --best
+
+all: $(LIBRARY)
+
+.SUFFIXES: .c .lo .o
+.PHONY: all clean distclean maintainer-clean
+
+.c.o:
+	$(COMPILE) -c $<
+
+.c.lo:
+	$(LTCOMPILE) -c $<
+
+Makefile: $(srcdir)/Makefile.in  $(top_builddir)/config.status
+	cd $(top_builddir) \
+	  && CONFIG_FILES=$(subdir)/$@ CONFIG_HEADERS= $(SHELL) ./config.status
+
+$(top_builddir)/config.status: $(top_builddir)/configure
+	cd $(top_builddir) && $(MAKE) config.status
+
+$(top_builddir)/config.h: $(top_builddir)/config.h.in
+	cd $(top_builddir) && $(MAKE) config.h
+
+clean:
+	rm -f $(LIBRARY) *.o *.lo *~
+	rm -rf .libs _libs
+
+distclean: clean
+	rm -f Makefile expat.h
+
+maintainer-clean: distclean
+
+install: $(LIBRARY) $(APIHEADER)
+	$(mkinstalldirs) $(DESTDIR)$(libdir) $(DESTDIR)$(includedir)
+	$(LIBTOOL)  --mode=install $(INSTALL) $(LIBRARY) $(DESTDIR)$(libdir)/$(LIBRARY)
+	$(INSTALL_DATA) $(APIHEADER) $(DESTDIR)$(includedir)
+
+uninstall:
+	$(LIBTOOL)  --mode=uninstall rm -f $(DESTDIR)$(libdir)/$(LIBRARY);
+	rm -f $(DESTDIR)$(libdir)/$(APIHEADER)
+
+$(LIBRARY): $(LTOBJECTS)
+	$(LINK) -rpath $(libdir) $(LDFLAGS) $(LTOBJECTS)
+
+xmlparse.o \
+xmlparse.lo: xmlparse.c expat.h xmlrole.h xmltok.h $(top_builddir)/config.h
+
+xmlrole.o \
+xmlrole.lo: xmlrole.c ascii.h xmlrole.h $(top_builddir)/config.h
+
+xmltok.o \
+xmltok.lo: xmltok.c xmltok_impl.c xmltok_ns.c \
+          ascii.h asciitab.h iasciitab.h latin1tab.h nametab.h utf8tab.h \
+          xmltok.h xmltok_impl.h $(top_builddir)/config.h
diff --git a/xml/expat/lib/ascii.h b/xml/expat/lib/ascii.h
new file mode 100644
index 000000000..6376b1f31
--- /dev/null
+++ b/xml/expat/lib/ascii.h
@@ -0,0 +1,86 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+#define ASCII_A 0x41
+#define ASCII_B 0x42
+#define ASCII_C 0x43
+#define ASCII_D 0x44
+#define ASCII_E 0x45
+#define ASCII_F 0x46
+#define ASCII_G 0x47
+#define ASCII_H 0x48
+#define ASCII_I 0x49
+#define ASCII_J 0x4A
+#define ASCII_K 0x4B
+#define ASCII_L 0x4C
+#define ASCII_M 0x4D
+#define ASCII_N 0x4E
+#define ASCII_O 0x4F
+#define ASCII_P 0x50
+#define ASCII_Q 0x51
+#define ASCII_R 0x52
+#define ASCII_S 0x53
+#define ASCII_T 0x54
+#define ASCII_U 0x55
+#define ASCII_V 0x56
+#define ASCII_W 0x57
+#define ASCII_X 0x58
+#define ASCII_Y 0x59
+#define ASCII_Z 0x5A
+
+#define ASCII_a 0x61
+#define ASCII_b 0x62
+#define ASCII_c 0x63
+#define ASCII_d 0x64
+#define ASCII_e 0x65
+#define ASCII_f 0x66
+#define ASCII_g 0x67
+#define ASCII_h 0x68
+#define ASCII_i 0x69
+#define ASCII_j 0x6A
+#define ASCII_k 0x6B
+#define ASCII_l 0x6C
+#define ASCII_m 0x6D
+#define ASCII_n 0x6E
+#define ASCII_o 0x6F
+#define ASCII_p 0x70
+#define ASCII_q 0x71
+#define ASCII_r 0x72
+#define ASCII_s 0x73
+#define ASCII_t 0x74
+#define ASCII_u 0x75
+#define ASCII_v 0x76
+#define ASCII_w 0x77
+#define ASCII_x 0x78
+#define ASCII_y 0x79
+#define ASCII_z 0x7A
+
+#define ASCII_0 0x30
+#define ASCII_1 0x31
+#define ASCII_2 0x32
+#define ASCII_3 0x33
+#define ASCII_4 0x34
+#define ASCII_5 0x35
+#define ASCII_6 0x36
+#define ASCII_7 0x37
+#define ASCII_8 0x38
+#define ASCII_9 0x39
+
+#define ASCII_TAB 0x09
+#define ASCII_SPACE 0x20 
+#define ASCII_EXCL 0x21
+#define ASCII_QUOT 0x22
+#define ASCII_AMP 0x26
+#define ASCII_APOS 0x27
+#define ASCII_MINUS 0x2D
+#define ASCII_PERIOD 0x2E
+#define ASCII_COLON 0x3A
+#define ASCII_SEMI 0x3B
+#define ASCII_LT 0x3C
+#define ASCII_EQUALS 0x3D
+#define ASCII_GT 0x3E
+#define ASCII_LSQB 0x5B
+#define ASCII_RSQB 0x5D
+#define ASCII_UNDERSCORE 0x5F
diff --git a/xml/expat/lib/asciitab.h b/xml/expat/lib/asciitab.h
new file mode 100644
index 000000000..eb445cc52
--- /dev/null
+++ b/xml/expat/lib/asciitab.h
@@ -0,0 +1,37 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML,
+/* 0x0C */ BT_NONXML, BT_CR, BT_NONXML, BT_NONXML,
+/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM,
+/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS,
+/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS,
+/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL,
+/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI,
+/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST,
+/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB,
+/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT,
+/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER,
diff --git a/xml/expat/lib/config.hnw b/xml/expat/lib/config.hnw
new file mode 100644
index 000000000..de129d343
--- /dev/null
+++ b/xml/expat/lib/config.hnw
@@ -0,0 +1,23 @@
+/*================================================================
+** Copyright 2000, Clark Cooper
+** All rights reserved.
+**
+** This is free software. You are permitted to copy, distribute, or modify
+** it under the terms of the MIT/X license (contained in the COPYING file
+** with this distribution.)
+**
+**
+*/
+
+#ifndef CONFIG_HNW
+#define CONFIG_HNW
+
+#include <memory.h>
+#include <string.h>
+
+#define XML_NS 1
+#define XML_DTD 1
+#define XML_BYTE_ORDER 12
+#define XML_CONTEXT_BYTES 1024
+
+#endif /* ndef CONFIG_HNW */
diff --git a/xml/expat/lib/expat.dsp b/xml/expat/lib/expat.dsp
new file mode 100644
index 000000000..71aae06bd
--- /dev/null
+++ b/xml/expat/lib/expat.dsp
@@ -0,0 +1,239 @@
+# Microsoft Developer Studio Project File - Name="expat" - Package Owner=<4>
+# Microsoft Developer Studio Generated Build File, Format Version 6.00
+# ** DO NOT EDIT **
+
+# TARGTYPE "Win32 (x86) Dynamic-Link Library" 0x0102
+
+CFG=expat - Win32 Debug
+!MESSAGE This is not a valid makefile. To build this project using NMAKE,
+!MESSAGE use the Export Makefile command and run
+!MESSAGE 
+!MESSAGE NMAKE /f "expat.mak".
+!MESSAGE 
+!MESSAGE You can specify a configuration when running NMAKE
+!MESSAGE by defining the macro CFG on the command line. For example:
+!MESSAGE 
+!MESSAGE NMAKE /f "expat.mak" CFG="expat - Win32 Debug"
+!MESSAGE 
+!MESSAGE Possible choices for configuration are:
+!MESSAGE 
+!MESSAGE "expat - Win32 Release" (based on "Win32 (x86) Dynamic-Link Library")
+!MESSAGE "expat - Win32 Debug" (based on "Win32 (x86) Dynamic-Link Library")
+!MESSAGE 
+
+# Begin Project
+# PROP AllowPerConfigDependencies 0
+# PROP Scc_ProjName ""
+# PROP Scc_LocalPath ""
+CPP=cl.exe
+MTL=midl.exe
+RSC=rc.exe
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 0
+# PROP BASE Output_Dir "Release"
+# PROP BASE Intermediate_Dir "Release"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 0
+# PROP Output_Dir "Release"
+# PROP Intermediate_Dir "Release"
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MT /W3 /EHsc /O2 /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "EXPAT_EXPORTS" /Yu"stdafx.h" /FD /c
+# ADD CPP /nologo /MT /W3 /EHsc /O2 /D "NDEBUG" /D "WIN32" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "EXPAT_EXPORTS" /D "COMPILED_FROM_DSP" /Yu"stdafx.h" /FD /c
+# ADD BASE MTL /nologo /D "NDEBUG" /mktyplib203 /win32
+# ADD MTL /nologo /D "NDEBUG" /mktyplib203 /win32
+# ADD BASE RSC /l 0x409 /d "NDEBUG"
+# ADD RSC /l 0x409 /d "NDEBUG"
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LINK32=link.exe
+# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll
+# ADD LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll
+# Begin Special Build Tool
+TargetPath=.\Release\expat.dll
+SOURCE="$(InputPath)"
+PostBuild_Desc=Embed .manifest
+PostBuild_Cmds=if exist $(TargetPath).manifest mt.exe -manifest $(TargetPath).manifest -outputresource:$(TargetPath);2
+# End Special Build Tool
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 1
+# PROP BASE Output_Dir "Debug"
+# PROP BASE Intermediate_Dir "Debug"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 1
+# PROP Output_Dir "Debug"
+# PROP Intermediate_Dir "Debug"
+# PROP Ignore_Export_Lib 0
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MTd /W3 /Gm /EHsc /Zi /Od /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "EXPAT_EXPORTS" /Yu"stdafx.h" /FD /GZ /c
+# ADD CPP /nologo /MTd /W3 /EHsc /Zi /Od /D "_DEBUG" /D "COMPILED_FROM_DSP" /D "WIN32" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "EXPAT_EXPORTS" /FD /GZ /c
+# ADD BASE MTL /nologo /D "_DEBUG" /mktyplib203 /win32
+# ADD MTL /nologo /D "_DEBUG" /mktyplib203 /win32
+# ADD BASE RSC /l 0x409 /d "_DEBUG"
+# ADD RSC /l 0x409 /d "_DEBUG"
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LINK32=link.exe
+# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll /debug /pdbtype:sept
+# ADD LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll /debug /pdbtype:sept
+# Begin Special Build Tool
+TargetPath=.\Debug\expat.dll
+SOURCE="$(InputPath)"
+PostBuild_Desc=Embed .manifest
+PostBuild_Cmds=if exist $(TargetPath).manifest mt.exe -manifest $(TargetPath).manifest -outputresource:$(TargetPath);2
+# End Special Build Tool
+
+!ENDIF 
+
+# Begin Target
+
+# Name "expat - Win32 Release"
+# Name "expat - Win32 Debug"
+# Begin Group "Source Files"
+
+# PROP Default_Filter "cpp;c;cxx;rc;def;r;odl;idl;hpj;bat"
+# Begin Source File
+
+SOURCE=.\xmlparse.c
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# ADD CPP /EHsc- /Od /D VERSION=\"expat_1.95.2\"
+
+!ENDIF 
+
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmlrole.c
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ENDIF 
+
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok.c
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ENDIF 
+
+# End Source File
+# Begin Source File
+
+SOURCE=xmltok_impl.c
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# PROP Exclude_From_Build 1
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ENDIF 
+
+# End Source File
+# Begin Source File
+
+SOURCE=xmltok_ns.c
+
+!IF  "$(CFG)" == "expat - Win32 Release"
+
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ELSEIF  "$(CFG)" == "expat - Win32 Debug"
+
+# PROP Exclude_From_Build 1
+# ADD CPP /D VERSION=\"expat_1.95.2\"
+
+!ENDIF 
+
+# End Source File
+# End Group
+# Begin Group "Header Files"
+
+# PROP Default_Filter "h;hpp;hxx;hm;inl"
+# Begin Source File
+
+SOURCE=.\ascii.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\asciitab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\config.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\expat.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\iasciitab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\latin1tab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\nametab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\utf8tab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmlrole.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok_impl.h
+# End Source File
+# End Group
+# Begin Group "Resource Files"
+
+# PROP Default_Filter "ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe"
+# End Group
+# Begin Source File
+
+SOURCE=.\ReadMe.txt
+# End Source File
+# End Target
+# End Project
diff --git a/xml/expat/lib/expat.h.in b/xml/expat/lib/expat.h.in
new file mode 100644
index 000000000..9e440e200
--- /dev/null
+++ b/xml/expat/lib/expat.h.in
@@ -0,0 +1,742 @@
+/*
+Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+#ifndef XmlParse_INCLUDED
+#define XmlParse_INCLUDED 1
+
+#include <stdlib.h>
+
+#ifndef XMLPARSEAPI
+#  if defined(__declspec) && !defined(__CYGWIN__)
+#    define XMLPARSEAPI __declspec(dllimport)
+#  else
+#    define XMLPARSEAPI /* nothing */
+#  endif
+#endif  /* not defined XMLPARSEAPI */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef void *XML_Parser;
+
+/* Information is UTF-8 encoded. */
+typedef char XML_Char;
+typedef char XML_LChar;
+
+enum XML_Content_Type {
+  XML_CTYPE_EMPTY = 1,
+  XML_CTYPE_ANY,
+  XML_CTYPE_MIXED,
+  XML_CTYPE_NAME,
+  XML_CTYPE_CHOICE,
+  XML_CTYPE_SEQ
+};
+
+enum XML_Content_Quant {
+  XML_CQUANT_NONE,
+  XML_CQUANT_OPT,
+  XML_CQUANT_REP,
+  XML_CQUANT_PLUS
+};
+
+/* If type == XML_CTYPE_EMPTY or XML_CTYPE_ANY, then quant will be
+   XML_CQUANT_NONE, and the other fields will be zero or NULL.
+   If type == XML_CTYPE_MIXED, then quant will be NONE or REP and
+   numchildren will contain number of elements that may be mixed in
+   and children point to an array of XML_Content cells that will be
+   all of XML_CTYPE_NAME type with no quantification.
+
+   If type == XML_CTYPE_NAME, then the name points to the name, and
+   the numchildren field will be zero and children will be NULL. The
+   quant fields indicates any quantifiers placed on the name.
+
+   CHOICE and SEQ will have name NULL, the number of children in
+   numchildren and children will point, recursively, to an array
+   of XML_Content cells.
+
+   The EMPTY, ANY, and MIXED types will only occur at top level.
+*/
+
+typedef struct XML_cp XML_Content;
+
+struct XML_cp {
+  enum XML_Content_Type		type;
+  enum XML_Content_Quant	quant;
+  const XML_Char *		name;
+  unsigned int			numchildren;
+  XML_Content *			children;
+};
+
+
+/* This is called for an element declaration. See above for
+   description of the model argument. It's the caller's responsibility
+   to free model when finished with it.
+*/
+
+typedef void (*XML_ElementDeclHandler) (void *userData,
+					const XML_Char *name,
+					XML_Content *model);
+
+void XMLPARSEAPI
+XML_SetElementDeclHandler(XML_Parser parser,
+			  XML_ElementDeclHandler eldecl);
+
+/*
+  The Attlist declaration handler is called for *each* attribute. So
+  a single Attlist declaration with multiple attributes declared will
+  generate multiple calls to this handler. The "default" parameter
+  may be NULL in the case of the "#IMPLIED" or "#REQUIRED" keyword.
+  The "isrequired" parameter will be true and the default value will
+  be NULL in the case of "#REQUIRED". If "isrequired" is true and
+  default is non-NULL, then this is a "#FIXED" default.
+ */
+
+typedef void (*XML_AttlistDeclHandler) (void		*userData,
+					const XML_Char	*elname,
+					const XML_Char	*attname,
+					const XML_Char	*att_type,
+					const XML_Char	*dflt,
+					int		isrequired);
+
+void XMLPARSEAPI
+XML_SetAttlistDeclHandler(XML_Parser parser,
+			  XML_AttlistDeclHandler attdecl);
+
+
+  /* The XML declaration handler is called for *both* XML declarations and
+     text declarations. The way to distinguish is that the version parameter
+     will be null for text declarations. The encoding parameter may be null
+     for XML declarations. The standalone parameter will be -1, 0, or 1
+     indicating respectively that there was no standalone parameter in
+     the declaration, that it was given as no, or that it was given as yes.
+  */
+
+typedef void (*XML_XmlDeclHandler) (void		*userData,
+				    const XML_Char	*version,
+				    const XML_Char	*encoding,
+				    int			standalone);
+
+void XMLPARSEAPI
+XML_SetXmlDeclHandler(XML_Parser parser,
+		      XML_XmlDeclHandler xmldecl);
+
+
+typedef struct {
+  void *(*malloc_fcn)(size_t size);
+  void *(*realloc_fcn)(void *ptr, size_t size);
+  void (*free_fcn)(void *ptr);
+} XML_Memory_Handling_Suite;
+
+/* Constructs a new parser; encoding is the encoding specified by the
+external protocol or null if there is none specified. */
+
+XML_Parser XMLPARSEAPI
+XML_ParserCreate(const XML_Char *encoding);
+
+/* Constructs a new parser and namespace processor.  Element type
+names and attribute names that belong to a namespace will be expanded;
+unprefixed attribute names are never expanded; unprefixed element type
+names are expanded only if there is a default namespace. The expanded
+name is the concatenation of the namespace URI, the namespace
+separator character, and the local part of the name.  If the namespace
+separator is '\0' then the namespace URI and the local part will be
+concatenated without any separator.  When a namespace is not declared,
+the name and prefix will be passed through without expansion. */
+
+XML_Parser XMLPARSEAPI
+XML_ParserCreateNS(const XML_Char *encoding, XML_Char namespaceSeparator);
+
+
+/* Constructs a new parser using the memory management suit referred to
+   by memsuite. If memsuite is NULL, then use the standard library memory
+   suite. If namespaceSeparator is non-NULL it creates a parser with
+   namespace processing as described above. The character pointed at
+   will serve as the namespace separator.
+
+   All further memory operations used for the created parser will come from
+   the given suite.
+*/
+
+XML_Parser XMLPARSEAPI
+XML_ParserCreate_MM(const XML_Char *encoding,
+		    const XML_Memory_Handling_Suite *memsuite,
+		    const XML_Char *namespaceSeparator);
+
+/* atts is array of name/value pairs, terminated by 0;
+   names and values are 0 terminated. */
+
+typedef void (*XML_StartElementHandler)(void *userData,
+					const XML_Char *name,
+					const XML_Char **atts);
+
+typedef void (*XML_EndElementHandler)(void *userData,
+				      const XML_Char *name);
+
+
+/* s is not 0 terminated. */
+typedef void (*XML_CharacterDataHandler)(void *userData,
+					 const XML_Char *s,
+					 int len);
+
+/* target and data are 0 terminated */
+typedef void (*XML_ProcessingInstructionHandler)(void *userData,
+						 const XML_Char *target,
+						 const XML_Char *data);
+
+/* data is 0 terminated */
+typedef void (*XML_CommentHandler)(void *userData, const XML_Char *data);
+
+typedef void (*XML_StartCdataSectionHandler)(void *userData);
+typedef void (*XML_EndCdataSectionHandler)(void *userData);
+
+/* This is called for any characters in the XML document for
+which there is no applicable handler.  This includes both
+characters that are part of markup which is of a kind that is
+not reported (comments, markup declarations), or characters
+that are part of a construct which could be reported but
+for which no handler has been supplied. The characters are passed
+exactly as they were in the XML document except that
+they will be encoded in UTF-8.  Line boundaries are not normalized.
+Note that a byte order mark character is not passed to the default handler.
+There are no guarantees about how characters are divided between calls
+to the default handler: for example, a comment might be split between
+multiple calls. */
+
+typedef void (*XML_DefaultHandler)(void *userData,
+				   const XML_Char *s,
+				   int len);
+
+/* This is called for the start of the DOCTYPE declaration, before
+   any DTD or internal subset is parsed. */
+
+typedef void (*XML_StartDoctypeDeclHandler)(void *userData,
+					    const XML_Char *doctypeName,
+					    const XML_Char *sysid,
+					    const XML_Char *pubid,
+					    int has_internal_subset
+					    );
+
+/* This is called for the start of the DOCTYPE declaration when the
+closing > is encountered, but after processing any external subset. */
+typedef void (*XML_EndDoctypeDeclHandler)(void *userData);
+
+/* This is called for entity declarations. The is_parameter_entity
+   argument will be non-zero if the entity is a parameter entity, zero
+   otherwise.
+
+   For internal entities (<!ENTITY foo "bar">), value will
+   be non-null and systemId, publicID, and notationName will be null.
+   The value string is NOT null terminated; the length is provided in
+   the value_length argument. Since it is legal to have zero-length
+   values, do not use this argument to test for internal entities.
+
+   For external entities, value will be null and systemId will be non-null.
+   The publicId argument will be null unless a public identifier was
+   provided. The notationName argument will have a non-null value only
+   for unparsed entity declarations.
+*/
+
+typedef void (*XML_EntityDeclHandler) (void *userData,
+				       const XML_Char *entityName,
+				       int is_parameter_entity,
+				       const XML_Char *value,
+				       int value_length,
+				       const XML_Char *base,
+				       const XML_Char *systemId,
+				       const XML_Char *publicId,
+				       const XML_Char *notationName);
+				       
+void XMLPARSEAPI
+XML_SetEntityDeclHandler(XML_Parser parser,
+			 XML_EntityDeclHandler handler);
+
+/* OBSOLETE -- OBSOLETE -- OBSOLETE
+   This handler has been superceded by the EntityDeclHandler above.
+   It is provided here for backward compatibility.
+This is called for a declaration of an unparsed (NDATA)
+entity.  The base argument is whatever was set by XML_SetBase.
+The entityName, systemId and notationName arguments will never be null.
+The other arguments may be. */
+
+typedef void (*XML_UnparsedEntityDeclHandler)(void *userData,
+					      const XML_Char *entityName,
+					      const XML_Char *base,
+					      const XML_Char *systemId,
+					      const XML_Char *publicId,
+					      const XML_Char *notationName);
+
+/* This is called for a declaration of notation.
+The base argument is whatever was set by XML_SetBase.
+The notationName will never be null.  The other arguments can be. */
+
+typedef void (*XML_NotationDeclHandler)(void *userData,
+					const XML_Char *notationName,
+					const XML_Char *base,
+					const XML_Char *systemId,
+					const XML_Char *publicId);
+
+/* When namespace processing is enabled, these are called once for
+each namespace declaration. The call to the start and end element
+handlers occur between the calls to the start and end namespace
+declaration handlers. For an xmlns attribute, prefix will be null.
+For an xmlns="" attribute, uri will be null. */
+
+typedef void (*XML_StartNamespaceDeclHandler)(void *userData,
+					      const XML_Char *prefix,
+					      const XML_Char *uri);
+
+typedef void (*XML_EndNamespaceDeclHandler)(void *userData,
+					    const XML_Char *prefix);
+
+/* This is called if the document is not standalone (it has an
+external subset or a reference to a parameter entity, but does not
+have standalone="yes"). If this handler returns 0, then processing
+will not continue, and the parser will return a
+XML_ERROR_NOT_STANDALONE error. */
+
+typedef int (*XML_NotStandaloneHandler)(void *userData);
+
+/* This is called for a reference to an external parsed general entity.
+The referenced entity is not automatically parsed.
+The application can parse it immediately or later using
+XML_ExternalEntityParserCreate.
+The parser argument is the parser parsing the entity containing the reference;
+it can be passed as the parser argument to XML_ExternalEntityParserCreate.
+The systemId argument is the system identifier as specified in the entity
+declaration; it will not be null.
+The base argument is the system identifier that should be used as the base for
+resolving systemId if systemId was relative; this is set by XML_SetBase;
+it may be null.
+The publicId argument is the public identifier as specified in the entity
+declaration, or null if none was specified; the whitespace in the public
+identifier will have been normalized as required by the XML spec.
+The context argument specifies the parsing context in the format
+expected by the context argument to
+XML_ExternalEntityParserCreate; context is valid only until the handler
+returns, so if the referenced entity is to be parsed later, it must be copied.
+The handler should return 0 if processing should not continue because of
+a fatal error in the handling of the external entity.
+In this case the calling parser will return an
+XML_ERROR_EXTERNAL_ENTITY_HANDLING error.
+Note that unlike other handlers the first argument is the parser, not
+userData. */
+
+typedef int (*XML_ExternalEntityRefHandler)(XML_Parser parser,
+					    const XML_Char *context,
+					    const XML_Char *base,
+					    const XML_Char *systemId,
+					    const XML_Char *publicId);
+
+/* This structure is filled in by the XML_UnknownEncodingHandler
+to provide information to the parser about encodings that are unknown
+to the parser.
+The map[b] member gives information about byte sequences
+whose first byte is b.
+If map[b] is c where c is >= 0, then b by itself encodes the Unicode scalar
+value c.
+If map[b] is -1, then the byte sequence is malformed.
+If map[b] is -n, where n >= 2, then b is the first byte of an n-byte
+sequence that encodes a single Unicode scalar value.
+The data member will be passed as the first argument to the convert function.
+The convert function is used to convert multibyte sequences;
+s will point to a n-byte sequence where map[(unsigned char)*s] == -n.
+The convert function must return the Unicode scalar value
+represented by this byte sequence or -1 if the byte sequence is malformed.
+The convert function may be null if the encoding is a single-byte encoding,
+that is if map[b] >= -1 for all bytes b.
+When the parser is finished with the encoding, then if release is not null,
+it will call release passing it the data member;
+once release has been called, the convert function will not be called again.
+
+Expat places certain restrictions on the encodings that are supported
+using this mechanism.
+
+1. Every ASCII character that can appear in a well-formed XML document,
+other than the characters
+
+  $@\^`{}~
+
+must be represented by a single byte, and that byte must be the
+same byte that represents that character in ASCII.
+
+2. No character may require more than 4 bytes to encode.
+
+3. All characters encoded must have Unicode scalar values <= 0xFFFF,
+(ie characters that would be encoded by surrogates in UTF-16
+are  not allowed).  Note that this restriction doesn't apply to
+the built-in support for UTF-8 and UTF-16.
+
+4. No Unicode character may be encoded by more than one distinct sequence
+of bytes. */
+
+typedef struct {
+  int map[256];
+  void *data;
+  int (*convert)(void *data, const char *s);
+  void (*release)(void *data);
+} XML_Encoding;
+
+/* This is called for an encoding that is unknown to the parser.
+The encodingHandlerData argument is that which was passed as the
+second argument to XML_SetUnknownEncodingHandler.
+The name argument gives the name of the encoding as specified in
+the encoding declaration.
+If the callback can provide information about the encoding,
+it must fill in the XML_Encoding structure, and return 1.
+Otherwise it must return 0.
+If info does not describe a suitable encoding,
+then the parser will return an XML_UNKNOWN_ENCODING error. */
+
+typedef int (*XML_UnknownEncodingHandler)(void *encodingHandlerData,
+					  const XML_Char *name,
+					  XML_Encoding *info);
+
+void XMLPARSEAPI
+XML_SetElementHandler(XML_Parser parser,
+		      XML_StartElementHandler start,
+		      XML_EndElementHandler end);
+
+void XMLPARSEAPI
+XML_SetStartElementHandler(XML_Parser, XML_StartElementHandler);
+
+void XMLPARSEAPI
+XML_SetEndElementHandler(XML_Parser, XML_EndElementHandler);
+
+void XMLPARSEAPI
+XML_SetCharacterDataHandler(XML_Parser parser,
+			    XML_CharacterDataHandler handler);
+
+void XMLPARSEAPI
+XML_SetProcessingInstructionHandler(XML_Parser parser,
+				    XML_ProcessingInstructionHandler handler);
+void XMLPARSEAPI
+XML_SetCommentHandler(XML_Parser parser,
+                      XML_CommentHandler handler);
+
+void XMLPARSEAPI
+XML_SetCdataSectionHandler(XML_Parser parser,
+			   XML_StartCdataSectionHandler start,
+			   XML_EndCdataSectionHandler end);
+
+void XMLPARSEAPI
+XML_SetStartCdataSectionHandler(XML_Parser parser,
+                                XML_StartCdataSectionHandler start);
+
+void XMLPARSEAPI
+XML_SetEndCdataSectionHandler(XML_Parser parser,
+                              XML_EndCdataSectionHandler end);
+
+/* This sets the default handler and also inhibits expansion of
+internal entities.  The entity reference will be passed to the default
+handler. */
+
+void XMLPARSEAPI
+XML_SetDefaultHandler(XML_Parser parser,
+		      XML_DefaultHandler handler);
+
+/* This sets the default handler but does not inhibit expansion of
+internal entities.  The entity reference will not be passed to the
+default handler. */
+
+void XMLPARSEAPI
+XML_SetDefaultHandlerExpand(XML_Parser parser,
+		            XML_DefaultHandler handler);
+
+void XMLPARSEAPI
+XML_SetDoctypeDeclHandler(XML_Parser parser,
+			  XML_StartDoctypeDeclHandler start,
+			  XML_EndDoctypeDeclHandler end);
+
+void XMLPARSEAPI
+XML_SetStartDoctypeDeclHandler(XML_Parser parser,
+			       XML_StartDoctypeDeclHandler start);
+
+void XMLPARSEAPI
+XML_SetEndDoctypeDeclHandler(XML_Parser parser,
+			     XML_EndDoctypeDeclHandler end);
+
+void XMLPARSEAPI
+XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
+				 XML_UnparsedEntityDeclHandler handler);
+
+void XMLPARSEAPI
+XML_SetNotationDeclHandler(XML_Parser parser,
+			   XML_NotationDeclHandler handler);
+
+void XMLPARSEAPI
+XML_SetNamespaceDeclHandler(XML_Parser parser,
+			    XML_StartNamespaceDeclHandler start,
+			    XML_EndNamespaceDeclHandler end);
+
+void XMLPARSEAPI
+XML_SetStartNamespaceDeclHandler(XML_Parser parser,
+				 XML_StartNamespaceDeclHandler start);
+
+void XMLPARSEAPI
+XML_SetEndNamespaceDeclHandler(XML_Parser parser,
+			       XML_EndNamespaceDeclHandler end);
+
+void XMLPARSEAPI
+XML_SetNotStandaloneHandler(XML_Parser parser,
+			    XML_NotStandaloneHandler handler);
+
+void XMLPARSEAPI
+XML_SetExternalEntityRefHandler(XML_Parser parser,
+				XML_ExternalEntityRefHandler handler);
+
+/* If a non-null value for arg is specified here, then it will be passed
+as the first argument to the external entity ref handler instead
+of the parser object. */
+void XMLPARSEAPI
+XML_SetExternalEntityRefHandlerArg(XML_Parser, void *arg);
+
+void XMLPARSEAPI
+XML_SetUnknownEncodingHandler(XML_Parser parser,
+			      XML_UnknownEncodingHandler handler,
+			      void *encodingHandlerData);
+
+/* This can be called within a handler for a start element, end element,
+processing instruction or character data.  It causes the corresponding
+markup to be passed to the default handler. */
+void XMLPARSEAPI
+XML_DefaultCurrent(XML_Parser parser);
+
+/* If do_nst is non-zero, and namespace processing is in effect, and
+   a name has a prefix (i.e. an explicit namespace qualifier) then
+   that name is returned as a triplet in a single
+   string separated by the separator character specified when the parser
+   was created: URI + sep + local_name + sep + prefix.
+
+   If do_nst is zero, then namespace information is returned in the
+   default manner (URI + sep + local_name) whether or not the names
+   has a prefix.
+*/
+
+void XMLPARSEAPI
+XML_SetReturnNSTriplet(XML_Parser parser, int do_nst);
+
+/* This value is passed as the userData argument to callbacks. */
+void XMLPARSEAPI
+XML_SetUserData(XML_Parser parser, void *userData);
+
+/* Returns the last value set by XML_SetUserData or null. */
+#define XML_GetUserData(parser) (*(void **)(parser))
+
+/* This is equivalent to supplying an encoding argument
+to XML_ParserCreate. It must not be called after XML_Parse
+or XML_ParseBuffer. */
+
+int XMLPARSEAPI
+XML_SetEncoding(XML_Parser parser, const XML_Char *encoding);
+
+/* If this function is called, then the parser will be passed
+as the first argument to callbacks instead of userData.
+The userData will still be accessible using XML_GetUserData. */
+
+void XMLPARSEAPI 
+XML_UseParserAsHandlerArg(XML_Parser parser);
+
+/* Sets the base to be used for resolving relative URIs in system
+identifiers in declarations.  Resolving relative identifiers is left
+to the application: this value will be passed through as the base
+argument to the XML_ExternalEntityRefHandler, XML_NotationDeclHandler
+and XML_UnparsedEntityDeclHandler. The base argument will be copied.
+Returns zero if out of memory, non-zero otherwise. */
+
+int XMLPARSEAPI
+XML_SetBase(XML_Parser parser, const XML_Char *base);
+
+const XML_Char  XMLPARSEAPI *
+XML_GetBase(XML_Parser parser);
+
+/* Returns the number of the attribute/value pairs passed in last call
+to the XML_StartElementHandler that were specified in the start-tag
+rather than defaulted. Each attribute/value pair counts as 2; thus
+this correspondds to an index into the atts array passed to the
+XML_StartElementHandler. */
+
+int XMLPARSEAPI
+XML_GetSpecifiedAttributeCount(XML_Parser parser);
+
+/* Returns the index of the ID attribute passed in the last call to
+XML_StartElementHandler, or -1 if there is no ID attribute.  Each
+attribute/value pair counts as 2; thus this correspondds to an index
+into the atts array passed to the XML_StartElementHandler. */
+
+int XMLPARSEAPI
+XML_GetIdAttributeIndex(XML_Parser parser);
+
+/* Parses some input. Returns 0 if a fatal error is detected.
+The last call to XML_Parse must have isFinal true;
+len may be zero for this call (or any other). */
+int XMLPARSEAPI
+XML_Parse(XML_Parser parser, const char *s, int len, int isFinal);
+
+void XMLPARSEAPI *
+XML_GetBuffer(XML_Parser parser, int len);
+
+int XMLPARSEAPI
+XML_ParseBuffer(XML_Parser parser, int len, int isFinal);
+
+/* Creates an XML_Parser object that can parse an external general
+entity; context is a '\0'-terminated string specifying the parse
+context; encoding is a '\0'-terminated string giving the name of the
+externally specified encoding, or null if there is no externally
+specified encoding.  The context string consists of a sequence of
+tokens separated by formfeeds (\f); a token consisting of a name
+specifies that the general entity of the name is open; a token of the
+form prefix=uri specifies the namespace for a particular prefix; a
+token of the form =uri specifies the default namespace.  This can be
+called at any point after the first call to an
+ExternalEntityRefHandler so longer as the parser has not yet been
+freed.  The new parser is completely independent and may safely be
+used in a separate thread.  The handlers and userData are initialized
+from the parser argument.  Returns 0 if out of memory.  Otherwise
+returns a new XML_Parser object. */
+XML_Parser XMLPARSEAPI
+XML_ExternalEntityParserCreate(XML_Parser parser,
+			       const XML_Char *context,
+			       const XML_Char *encoding);
+
+enum XML_ParamEntityParsing {
+  XML_PARAM_ENTITY_PARSING_NEVER,
+  XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE,
+  XML_PARAM_ENTITY_PARSING_ALWAYS
+};
+
+/* Controls parsing of parameter entities (including the external DTD
+subset). If parsing of parameter entities is enabled, then references
+to external parameter entities (including the external DTD subset)
+will be passed to the handler set with
+XML_SetExternalEntityRefHandler.  The context passed will be 0.
+Unlike external general entities, external parameter entities can only
+be parsed synchronously.  If the external parameter entity is to be
+parsed, it must be parsed during the call to the external entity ref
+handler: the complete sequence of XML_ExternalEntityParserCreate,
+XML_Parse/XML_ParseBuffer and XML_ParserFree calls must be made during
+this call.  After XML_ExternalEntityParserCreate has been called to
+create the parser for the external parameter entity (context must be 0
+for this call), it is illegal to make any calls on the old parser
+until XML_ParserFree has been called on the newly created parser.  If
+the library has been compiled without support for parameter entity
+parsing (ie without XML_DTD being defined), then
+XML_SetParamEntityParsing will return 0 if parsing of parameter
+entities is requested; otherwise it will return non-zero. */
+
+int XMLPARSEAPI
+XML_SetParamEntityParsing(XML_Parser parser,
+			  enum XML_ParamEntityParsing parsing);
+
+enum XML_Error {
+  XML_ERROR_NONE,
+  XML_ERROR_NO_MEMORY,
+  XML_ERROR_SYNTAX,
+  XML_ERROR_NO_ELEMENTS,
+  XML_ERROR_INVALID_TOKEN,
+  XML_ERROR_UNCLOSED_TOKEN,
+  XML_ERROR_PARTIAL_CHAR,
+  XML_ERROR_TAG_MISMATCH,
+  XML_ERROR_DUPLICATE_ATTRIBUTE,
+  XML_ERROR_JUNK_AFTER_DOC_ELEMENT,
+  XML_ERROR_PARAM_ENTITY_REF,
+  XML_ERROR_UNDEFINED_ENTITY,
+  XML_ERROR_RECURSIVE_ENTITY_REF,
+  XML_ERROR_ASYNC_ENTITY,
+  XML_ERROR_BAD_CHAR_REF,
+  XML_ERROR_BINARY_ENTITY_REF,
+  XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF,
+  XML_ERROR_MISPLACED_XML_PI,
+  XML_ERROR_UNKNOWN_ENCODING,
+  XML_ERROR_INCORRECT_ENCODING,
+  XML_ERROR_UNCLOSED_CDATA_SECTION,
+  XML_ERROR_EXTERNAL_ENTITY_HANDLING,
+  XML_ERROR_NOT_STANDALONE,
+  XML_ERROR_UNEXPECTED_STATE
+};
+
+/* If XML_Parse or XML_ParseBuffer have returned 0, then XML_GetErrorCode
+returns information about the error. */
+
+enum XML_Error  XMLPARSEAPI
+XML_GetErrorCode(XML_Parser parser);
+
+/* These functions return information about the current parse location.
+They may be called when XML_Parse or XML_ParseBuffer return 0;
+in this case the location is the location of the character at which
+the error was detected.
+They may also be called from any other callback called to report
+some parse event; in this the location is the location of the first
+of the sequence of characters that generated the event. */
+
+int XMLPARSEAPI XML_GetCurrentLineNumber(XML_Parser parser);
+int XMLPARSEAPI XML_GetCurrentColumnNumber(XML_Parser parser);
+long XMLPARSEAPI XML_GetCurrentByteIndex(XML_Parser parser);
+
+/* Return the number of bytes in the current event.
+Returns 0 if the event is in an internal entity. */
+
+int XMLPARSEAPI
+XML_GetCurrentByteCount(XML_Parser parser);
+
+/* If XML_CONTEXT_BYTES is defined, returns the input buffer, sets
+   the integer pointed to by offset to the offset within this buffer
+   of the current parse position, and sets the integer pointed to by size
+   to the size of this buffer (the number of input bytes). Otherwise
+   returns a null pointer. Also returns a null pointer if a parse isn't
+   active.
+
+   NOTE: The character pointer returned should not be used outside
+   the handler that makes the call. */
+
+const char XMLPARSEAPI *
+XML_GetInputContext(XML_Parser parser,
+		    int *offset,
+		    int *size);
+
+/* For backwards compatibility with previous versions. */
+#define XML_GetErrorLineNumber XML_GetCurrentLineNumber
+#define XML_GetErrorColumnNumber XML_GetCurrentColumnNumber
+#define XML_GetErrorByteIndex XML_GetCurrentByteIndex
+
+/* Frees memory used by the parser. */
+void XMLPARSEAPI
+XML_ParserFree(XML_Parser parser);
+
+/* Returns a string describing the error. */
+const XML_LChar XMLPARSEAPI *
+XML_ErrorString(int code);
+
+/* Return a string containing the version number of this expat */
+const XML_LChar XMLPARSEAPI *
+XML_ExpatVersion(void);
+
+typedef struct {
+  int major;
+  int minor;
+  int micro;
+} XML_Expat_Version;
+
+/* Return an XML_Expat_Version structure containing numeric version
+   number information for this version of expat */
+
+XML_Expat_Version XMLPARSEAPI
+XML_ExpatVersionInfo(void);
+
+#ifndef XML_MAJOR_VERSION
+#define XML_MAJOR_VERSION 1
+#endif
+#ifndef XML_MINOR_VERSION
+#define XML_MINOR_VERSION 95
+#endif
+#ifndef XML_MICRO_VERSION
+#define XML_MICRO_VERSION 2
+#endif
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not XmlParse_INCLUDED */
diff --git a/xml/expat/lib/iasciitab.h b/xml/expat/lib/iasciitab.h
new file mode 100644
index 000000000..55dbc398b
--- /dev/null
+++ b/xml/expat/lib/iasciitab.h
@@ -0,0 +1,38 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+/* Like asciitab.h, except that 0xD has code BT_S rather than BT_CR */
+/* 0x00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x04 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x08 */ BT_NONXML, BT_S, BT_LF, BT_NONXML,
+/* 0x0C */ BT_NONXML, BT_S, BT_NONXML, BT_NONXML,
+/* 0x10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x14 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x18 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x1C */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0x20 */ BT_S, BT_EXCL, BT_QUOT, BT_NUM,
+/* 0x24 */ BT_OTHER, BT_PERCNT, BT_AMP, BT_APOS,
+/* 0x28 */ BT_LPAR, BT_RPAR, BT_AST, BT_PLUS,
+/* 0x2C */ BT_COMMA, BT_MINUS, BT_NAME, BT_SOL,
+/* 0x30 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x34 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* 0x38 */ BT_DIGIT, BT_DIGIT, BT_COLON, BT_SEMI,
+/* 0x3C */ BT_LT, BT_EQUALS, BT_GT, BT_QUEST,
+/* 0x40 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x44 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x48 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x4C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x50 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x58 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_LSQB,
+/* 0x5C */ BT_OTHER, BT_RSQB, BT_OTHER, BT_NMSTRT,
+/* 0x60 */ BT_OTHER, BT_HEX, BT_HEX, BT_HEX,
+/* 0x64 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 0x68 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x6C */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0x78 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0x7C */ BT_VERBAR, BT_OTHER, BT_OTHER, BT_OTHER,
diff --git a/xml/expat/lib/latin1tab.h b/xml/expat/lib/latin1tab.h
new file mode 100644
index 000000000..178b1d186
--- /dev/null
+++ b/xml/expat/lib/latin1tab.h
@@ -0,0 +1,37 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+/* 0x80 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x84 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x88 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x8C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x90 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x94 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x98 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0x9C */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA4 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xA8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER,
+/* 0xAC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xB0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xB4 */ BT_OTHER, BT_NMSTRT, BT_OTHER, BT_NAME,
+/* 0xB8 */ BT_OTHER, BT_OTHER, BT_NMSTRT, BT_OTHER,
+/* 0xBC */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 0xC0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xC4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xC8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xCC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xD0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xD4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0xD8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xDC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xE8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xEC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xF0 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xF4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 0xF8 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 0xFC */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
diff --git a/xml/expat/lib/map_osd_ebcdic_df04_1.h b/xml/expat/lib/map_osd_ebcdic_df04_1.h
new file mode 100644
index 000000000..93ece0c88
--- /dev/null
+++ b/xml/expat/lib/map_osd_ebcdic_df04_1.h
@@ -0,0 +1,18 @@
+static unsigned char ebcdic[] = { 
+/* 00 */ 0x00 ,0x01 ,0x02 ,0x03 ,0x85 ,0x09 ,0x86 ,0x7f ,0x87 ,0x8d ,0x8e ,0x0b ,0x0c ,0x0d ,0x0e ,0x0f ,
+/* 10 */ 0x10 ,0x11 ,0x12 ,0x13 ,0x8f ,0x0a ,0x08 ,0x97 ,0x18 ,0x19 ,0x9c ,0x9d ,0x1c ,0x1d ,0x1e ,0x1f ,
+/* 20 */ 0x80 ,0x81 ,0x82 ,0x83 ,0x84 ,0x92 ,0x17 ,0x1b ,0x88 ,0x89 ,0x8a ,0x8b ,0x8c ,0x05 ,0x06 ,0x07 ,
+/* 30 */ 0x90 ,0x91 ,0x16 ,0x93 ,0x94 ,0x95 ,0x96 ,0x04 ,0x98 ,0x99 ,0x9a ,0x9b ,0x14 ,0x15 ,0x9e ,0x1a ,
+/* 40 */ 0x20 ,0xa0 ,0xe2 ,0xe4 ,0xe0 ,0xe1 ,0xe3 ,0xe5 ,0xe7 ,0xf1 ,0x60 ,0x2e ,0x3c ,0x28 ,0x2b ,0x7c ,
+/* 50 */ 0x26 ,0xe9 ,0xea ,0xeb ,0xe8 ,0xed ,0xee ,0xef ,0xec ,0xdf ,0x21 ,0x24 ,0x2a ,0x29 ,0x3b ,0x9f ,
+/* 60 */ 0x2d ,0x2f ,0xc2 ,0xc4 ,0xc0 ,0xc1 ,0xc3 ,0xc5 ,0xc7 ,0xd1 ,0x5e ,0x2c ,0x25 ,0x5f ,0x3e ,0x3f ,
+/* 70 */ 0xf8 ,0xc9 ,0xca ,0xcb ,0xc8 ,0xcd ,0xce ,0xcf ,0xcc ,0xa8 ,0x3a ,0x23 ,0x40 ,0x27 ,0x3d ,0x22 ,
+/* 80 */ 0xd8 ,0x61 ,0x62 ,0x63 ,0x64 ,0x65 ,0x66 ,0x67 ,0x68 ,0x69 ,0xab ,0xbb ,0xf0 ,0xfd ,0xfe ,0xb1 ,
+/* 90 */ 0xb0 ,0x6a ,0x6b ,0x6c ,0x6d ,0x6e ,0x6f ,0x70 ,0x71 ,0x72 ,0xaa ,0xba ,0xe6 ,0xb8 ,0xc6 ,0xa4 ,
+/* a0 */ 0xb5 ,0xaf ,0x73 ,0x74 ,0x75 ,0x76 ,0x77 ,0x78 ,0x79 ,0x7a ,0xa1 ,0xbf ,0xd0 ,0xdd ,0xde ,0xae ,
+/* b0 */ 0xa2 ,0xa3 ,0xa5 ,0xb7 ,0xa9 ,0xa7 ,0xb6 ,0xbc ,0xbd ,0xbe ,0xac ,0x5b ,0x5c ,0x5d ,0xb4 ,0xd7 ,
+/* c0 */ 0xf9 ,0x41 ,0x42 ,0x43 ,0x44 ,0x45 ,0x46 ,0x47 ,0x48 ,0x49 ,0xad ,0xf4 ,0xf6 ,0xf2 ,0xf3 ,0xf5 ,
+/* d0 */ 0xa6 ,0x4a ,0x4b ,0x4c ,0x4d ,0x4e ,0x4f ,0x50 ,0x51 ,0x52 ,0xb9 ,0xfb ,0xfc ,0xdb ,0xfa ,0xff ,
+/* e0 */ 0xd9 ,0xf7 ,0x53 ,0x54 ,0x55 ,0x56 ,0x57 ,0x58 ,0x59 ,0x5a ,0xb2 ,0xd4 ,0xd6 ,0xd2 ,0xd3 ,0xd5 ,
+/* f0 */ 0x30 ,0x31 ,0x32 ,0x33 ,0x34 ,0x35 ,0x36 ,0x37 ,0x38 ,0x39 ,0xb3 ,0x7b ,0xdc ,0x7d ,0xda ,0x7e
+};
diff --git a/xml/expat/lib/nametab.h b/xml/expat/lib/nametab.h
new file mode 100644
index 000000000..b05e62c77
--- /dev/null
+++ b/xml/expat/lib/nametab.h
@@ -0,0 +1,150 @@
+static const unsigned namingBitmap[] = {
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0x00000000, 0x04000000, 0x87FFFFFE, 0x07FFFFFE,
+0x00000000, 0x00000000, 0xFF7FFFFF, 0xFF7FFFFF,
+0xFFFFFFFF, 0x7FF3FFFF, 0xFFFFFDFE, 0x7FFFFFFF,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFE00F, 0xFC31FFFF,
+0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF,
+0xFFFFFFFF, 0xF80001FF, 0x00000003, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFD740, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD,
+0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF,
+0xFFFF0003, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF,
+0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE,
+0x0000007F, 0x00000000, 0xFFFF0000, 0x000707FF,
+0x00000000, 0x07FFFFFE, 0x000007FE, 0xFFFE0000,
+0xFFFFFFFF, 0x7CFFFFFF, 0x002F7FFF, 0x00000060,
+0xFFFFFFE0, 0x23FFFFFF, 0xFF000000, 0x00000003,
+0xFFF99FE0, 0x03C5FDFF, 0xB0000000, 0x00030003,
+0xFFF987E0, 0x036DFDFF, 0x5E000000, 0x001C0000,
+0xFFFBAFE0, 0x23EDFDFF, 0x00000000, 0x00000001,
+0xFFF99FE0, 0x23CDFDFF, 0xB0000000, 0x00000003,
+0xD63DC7E0, 0x03BFC718, 0x00000000, 0x00000000,
+0xFFFDDFE0, 0x03EFFDFF, 0x00000000, 0x00000003,
+0xFFFDDFE0, 0x03EFFDFF, 0x40000000, 0x00000003,
+0xFFFDDFE0, 0x03FFFDFF, 0x00000000, 0x00000003,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFE, 0x000D7FFF, 0x0000003F, 0x00000000,
+0xFEF02596, 0x200D6CAE, 0x0000001F, 0x00000000,
+0x00000000, 0x00000000, 0xFFFFFEFF, 0x000003FF,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0xFFFFFFFF, 0xFFFF003F, 0x007FFFFF,
+0x0007DAED, 0x50000000, 0x82315001, 0x002C62AB,
+0x40000000, 0xF580C900, 0x00000007, 0x02010800,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0x0FFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0x03FFFFFF,
+0x3F3FFFFF, 0xFFFFFFFF, 0xAAFF3F3F, 0x3FFFFFFF,
+0xFFFFFFFF, 0x5FDFFFFF, 0x0FCF1FDC, 0x1FDC1FFF,
+0x00000000, 0x00004C40, 0x00000000, 0x00000000,
+0x00000007, 0x00000000, 0x00000000, 0x00000000,
+0x00000080, 0x000003FE, 0xFFFFFFFE, 0xFFFFFFFF,
+0x001FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x07FFFFFF,
+0xFFFFFFE0, 0x00001FFF, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0x0000003F, 0x00000000, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF, 0xFFFFFFFF,
+0xFFFFFFFF, 0x0000000F, 0x00000000, 0x00000000,
+0x00000000, 0x07FF6000, 0x87FFFFFE, 0x07FFFFFE,
+0x00000000, 0x00800000, 0xFF7FFFFF, 0xFF7FFFFF,
+0x00FFFFFF, 0x00000000, 0xFFFF0000, 0xFFFFFFFF,
+0xFFFFFFFF, 0xF80001FF, 0x00030003, 0x00000000,
+0xFFFFFFFF, 0xFFFFFFFF, 0x0000003F, 0x00000003,
+0xFFFFD7C0, 0xFFFFFFFB, 0x547F7FFF, 0x000FFFFD,
+0xFFFFDFFE, 0xFFFFFFFF, 0xDFFEFFFF, 0xFFFFFFFF,
+0xFFFF007B, 0xFFFFFFFF, 0xFFFF199F, 0x033FCFFF,
+0x00000000, 0xFFFE0000, 0x027FFFFF, 0xFFFFFFFE,
+0xFFFE007F, 0xBBFFFFFB, 0xFFFF0016, 0x000707FF,
+0x00000000, 0x07FFFFFE, 0x0007FFFF, 0xFFFF03FF,
+0xFFFFFFFF, 0x7CFFFFFF, 0xFFEF7FFF, 0x03FF3DFF,
+0xFFFFFFEE, 0xF3FFFFFF, 0xFF1E3FFF, 0x0000FFCF,
+0xFFF99FEE, 0xD3C5FDFF, 0xB080399F, 0x0003FFCF,
+0xFFF987E4, 0xD36DFDFF, 0x5E003987, 0x001FFFC0,
+0xFFFBAFEE, 0xF3EDFDFF, 0x00003BBF, 0x0000FFC1,
+0xFFF99FEE, 0xF3CDFDFF, 0xB0C0398F, 0x0000FFC3,
+0xD63DC7EC, 0xC3BFC718, 0x00803DC7, 0x0000FF80,
+0xFFFDDFEE, 0xC3EFFDFF, 0x00603DDF, 0x0000FFC3,
+0xFFFDDFEC, 0xC3EFFDFF, 0x40603DDF, 0x0000FFC3,
+0xFFFDDFEC, 0xC3FFFDFF, 0x00803DCF, 0x0000FFC3,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0xFFFFFFFE, 0x07FF7FFF, 0x03FF7FFF, 0x00000000,
+0xFEF02596, 0x3BFF6CAE, 0x03FF3F5F, 0x00000000,
+0x03000000, 0xC2A003FF, 0xFFFFFEFF, 0xFFFE03FF,
+0xFEBF0FDF, 0x02FE3FFF, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x00000000, 0x00000000,
+0x00000000, 0x00000000, 0x1FFF0000, 0x00000002,
+0x000000A0, 0x003EFFFE, 0xFFFFFFFE, 0xFFFFFFFF,
+0x661FFFFF, 0xFFFFFFFE, 0xFFFFFFFF, 0x77FFFFFF,
+};
+static const unsigned char nmstrtPages[] = {
+0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x00,
+0x00, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
+0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13,
+0x00, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x15, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+};
+static const unsigned char namePages[] = {
+0x19, 0x03, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x00,
+0x00, 0x1F, 0x20, 0x21, 0x22, 0x23, 0x24, 0x25,
+0x10, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x12, 0x13,
+0x26, 0x14, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x27, 0x16, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x17,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x18,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+};
diff --git a/xml/expat/lib/osd_ebcdic_df04_1.h b/xml/expat/lib/osd_ebcdic_df04_1.h
new file mode 100644
index 000000000..6f4b6b91a
--- /dev/null
+++ b/xml/expat/lib/osd_ebcdic_df04_1.h
@@ -0,0 +1,81 @@
+/* Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+/* Table for the OSD_EBCDIC_DF04_1 encoding  */
+
+/* 00 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 04 */ BT_OTHER, BT_S, BT_OTHER, BT_OTHER,
+/* 08 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_NONXML,
+/* 0c */ BT_NONXML, BT_CR, BT_NONXML, BT_NONXML,
+/* 10 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 14 */ BT_OTHER, BT_LF, BT_NONXML, BT_OTHER,
+/* 18 */ BT_NONXML, BT_NONXML, BT_OTHER, BT_OTHER,
+/* 1c */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 20 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 24 */ BT_OTHER, BT_OTHER, BT_NONXML, BT_NONXML,
+/* 28 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 2c */ BT_OTHER, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 30 */ BT_OTHER, BT_OTHER, BT_NONXML, BT_OTHER,
+/* 34 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_NONXML,
+/* 38 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* 3c */ BT_NONXML, BT_NONXML, BT_OTHER, BT_NONXML,
+/* 40 */ BT_S, BT_OTHER, BT_NMSTRT, BT_NMSTRT,
+/* 44 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 48 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_NAME,
+/* 4c */ BT_LT, BT_LPAR, BT_PLUS, BT_VERBAR,
+/* 50 */ BT_AMP, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 54 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 58 */ BT_NMSTRT, BT_NMSTRT, BT_EXCL, BT_OTHER,
+/* 5c */ BT_AST, BT_RPAR, BT_SEMI, BT_OTHER,
+/* 60 */ BT_MINUS, BT_SOL, BT_NMSTRT, BT_NMSTRT,
+/* 64 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 68 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_COMMA,
+/* 6c */ BT_PERCNT, BT_NMSTRT, BT_GT, BT_QUEST,
+/* 70 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 74 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 78 */ BT_NMSTRT, BT_OTHER, BT_COLON, BT_NUM,
+/* 7c */ BT_OTHER, BT_APOS, BT_EQUALS, BT_QUOT,
+/* 80 */ BT_NMSTRT, BT_HEX, BT_HEX, BT_HEX,
+/* 84 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* 88 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_OTHER,
+/* 8c */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* 90 */ BT_OTHER, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 94 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 98 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* 9c */ BT_NMSTRT, BT_OTHER, BT_NMSTRT, BT_OTHER,
+/* a0 */ BT_NMSTRT, BT_OTHER, BT_NMSTRT, BT_NMSTRT,
+/* a4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* a8 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_OTHER,
+/* ac */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_OTHER,
+/* b0 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_NAME,
+/* b4 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_OTHER,
+/* b8 */ BT_OTHER, BT_OTHER, BT_OTHER, BT_LSQB,
+/* bc */ BT_OTHER, BT_RSQB, BT_OTHER, BT_OTHER,
+/* c0 */ BT_NMSTRT, BT_HEX, BT_HEX, BT_HEX,
+/* c4 */ BT_HEX, BT_HEX, BT_HEX, BT_NMSTRT,
+/* c8 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_NMSTRT,
+/* cc */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* d0 */ BT_OTHER, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* d4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* d8 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_NMSTRT,
+/* dc */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* e0 */ BT_NMSTRT, BT_OTHER, BT_NMSTRT, BT_NMSTRT,
+/* e4 */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* e8 */ BT_NMSTRT, BT_NMSTRT, BT_OTHER, BT_NMSTRT,
+/* ec */ BT_NMSTRT, BT_NMSTRT, BT_NMSTRT, BT_NMSTRT,
+/* f0 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* f4 */ BT_DIGIT, BT_DIGIT, BT_DIGIT, BT_DIGIT,
+/* f8 */ BT_DIGIT, BT_DIGIT, BT_OTHER, BT_OTHER,
+/* fc */ BT_NMSTRT, BT_OTHER, BT_NMSTRT, BT_OTHER,
diff --git a/xml/expat/lib/utf8tab.h b/xml/expat/lib/utf8tab.h
new file mode 100644
index 000000000..9e3b6b83e
--- /dev/null
+++ b/xml/expat/lib/utf8tab.h
@@ -0,0 +1,38 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+
+/* 0x80 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x84 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x88 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x8C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x90 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x94 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x98 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0x9C */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xA8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xAC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB0 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB4 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xB8 */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xBC */ BT_TRAIL, BT_TRAIL, BT_TRAIL, BT_TRAIL,
+/* 0xC0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xC4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xC8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xCC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD0 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD4 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xD8 */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xDC */ BT_LEAD2, BT_LEAD2, BT_LEAD2, BT_LEAD2,
+/* 0xE0 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xE4 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xE8 */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xEC */ BT_LEAD3, BT_LEAD3, BT_LEAD3, BT_LEAD3,
+/* 0xF0 */ BT_LEAD4, BT_LEAD4, BT_LEAD4, BT_LEAD4,
+/* 0xF4 */ BT_LEAD4, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0xF8 */ BT_NONXML, BT_NONXML, BT_NONXML, BT_NONXML,
+/* 0xFC */ BT_NONXML, BT_NONXML, BT_MALFORM, BT_MALFORM,
diff --git a/xml/expat/lib/winconfig.h b/xml/expat/lib/winconfig.h
new file mode 100644
index 000000000..77b1ece8c
--- /dev/null
+++ b/xml/expat/lib/winconfig.h
@@ -0,0 +1,27 @@
+/*================================================================
+** Copyright 2000, Clark Cooper
+** All rights reserved.
+**
+** This is free software. You are permitted to copy, distribute, or modify
+** it under the terms of the MIT/X license (contained in the COPYING file
+** with this distribution.)
+**
+**
+*/
+
+#ifndef WINCONFIG_H
+#define WINCONFIG_H
+
+#define WIN32_LEAN_AND_MEAN
+#include <windows.h>
+#undef WIN32_LEAN_AND_MEAN
+
+#include <memory.h>
+#include <string.h>
+
+#define XML_NS 1
+#define XML_DTD 1
+#define XML_BYTE_ORDER 12
+#define XML_CONTEXT_BYTES 1024
+
+#endif /* ndef WINCONFIG_H */
diff --git a/xml/expat/lib/xml.dsp b/xml/expat/lib/xml.dsp
new file mode 100644
index 000000000..1c5b06762
--- /dev/null
+++ b/xml/expat/lib/xml.dsp
@@ -0,0 +1,312 @@
+# Microsoft Developer Studio Project File - Name="xml" - Package Owner=<4>
+# Microsoft Developer Studio Generated Build File, Format Version 6.00
+# ** DO NOT EDIT **
+
+# TARGTYPE "Win32 (x86) Static Library" 0x0104
+
+CFG=xml - Win32 Release
+!MESSAGE This is not a valid makefile. To build this project using NMAKE,
+!MESSAGE use the Export Makefile command and run
+!MESSAGE 
+!MESSAGE NMAKE /f "xml.mak".
+!MESSAGE 
+!MESSAGE You can specify a configuration when running NMAKE
+!MESSAGE by defining the macro CFG on the command line. For example:
+!MESSAGE 
+!MESSAGE NMAKE /f "xml.mak" CFG="xml - Win32 Release"
+!MESSAGE 
+!MESSAGE Possible choices for configuration are:
+!MESSAGE 
+!MESSAGE "xml - Win32 Release" (based on "Win32 (x86) Static Library")
+!MESSAGE "xml - Win32 Debug" (based on "Win32 (x86) Static Library")
+!MESSAGE "xml - x64 Release" (based on "Win32 (x86) Static Library")
+!MESSAGE "xml - x64 Debug" (based on "Win32 (x86) Static Library")
+!MESSAGE 
+
+# Begin Project
+# PROP AllowPerConfigDependencies 0
+# PROP Scc_ProjName ""
+# PROP Scc_LocalPath ""
+CPP=cl.exe
+RSC=rc.exe
+
+!IF  "$(CFG)" == "xml - Win32 Release"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 0
+# PROP BASE Output_Dir "LibR"
+# PROP BASE Intermediate_Dir "LibR"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 0
+# PROP Output_Dir "LibR"
+# PROP Intermediate_Dir "LibR"
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MD /W3 /O2 /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /FD /c
+# ADD CPP /nologo /MD /W3 /Zi /O2 /Oy- /I "." /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /D VERSION=\"expat_1.95.2\" /Fo"$(INTDIR)\" /Fd"$(OUTDIR)\xml" /FD /c
+# ADD BASE RSC /l 0x409
+# ADD RSC /l 0x409
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LIB32=link.exe -lib
+# ADD BASE LIB32 /nologo
+# ADD LIB32 /nologo /out:"LibR\xml.lib"
+
+!ELSEIF  "$(CFG)" == "xml - Win32 Debug"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 1
+# PROP BASE Output_Dir "LibD"
+# PROP BASE Intermediate_Dir "LibD"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 1
+# PROP Output_Dir "LibD"
+# PROP Intermediate_Dir "LibD"
+# PROP Ignore_Export_Lib 0
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MDd /W3 /Zi /Od /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /FD /EHsc /c
+# ADD CPP /nologo /MDd /W3 /Zi /Od /I "." /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /D VERSION=\"expat_1.95.2\" /Fo"$(INTDIR)\" /Fd"$(OUTDIR)\xml" /FD /EHsc /c
+# ADD BASE RSC /l 0x409
+# ADD RSC /l 0x409
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LIB32=link.exe -lib
+# ADD BASE LIB32 /nologo
+# ADD LIB32 /nologo /out:"LibD\xml.lib"
+
+!ELSEIF  "$(CFG)" == "xml - x64 Release"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 0
+# PROP BASE Output_Dir "x64\LibR"
+# PROP BASE Intermediate_Dir "x64\LibR"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 0
+# PROP Output_Dir "x64\LibR"
+# PROP Intermediate_Dir "x64\LibR"
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MD /W3 /O2 /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /FD /c
+# ADD CPP /nologo /MD /W3 /Zi /O2 /Oy- /I "." /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /D VERSION=\"expat_1.95.2\" /Fo"$(INTDIR)\" /Fd"$(OUTDIR)\xml" /FD /c
+# ADD BASE RSC /l 0x409
+# ADD RSC /l 0x409
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LIB32=link.exe -lib
+# ADD BASE LIB32 /nologo
+# ADD LIB32 /nologo /out:"x64\LibR\xml.lib"
+
+!ELSEIF  "$(CFG)" == "xml - x64 Debug"
+
+# PROP BASE Use_MFC 0
+# PROP BASE Use_Debug_Libraries 1
+# PROP BASE Output_Dir "x64\LibD"
+# PROP BASE Intermediate_Dir "x64\LibD"
+# PROP BASE Target_Dir ""
+# PROP Use_MFC 0
+# PROP Use_Debug_Libraries 1
+# PROP Output_Dir "x64\LibD"
+# PROP Intermediate_Dir "x64\LibD"
+# PROP Ignore_Export_Lib 0
+# PROP Target_Dir ""
+# ADD BASE CPP /nologo /MDd /W3 /Zi /Od /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /FD /EHsc /c
+# ADD CPP /nologo /MDd /W3 /Zi /Od /I "." /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /D VERSION=\"expat_1.95.2\" /Fo"$(INTDIR)\" /Fd"$(OUTDIR)\xml" /FD /EHsc /c
+# ADD BASE RSC /l 0x409
+# ADD RSC /l 0x409
+BSC32=bscmake.exe
+# ADD BASE BSC32 /nologo
+# ADD BSC32 /nologo
+LIB32=link.exe -lib
+# ADD BASE LIB32 /nologo
+# ADD LIB32 /nologo /out:"x64\LibD\xml.lib"
+
+!ENDIF 
+
+# Begin Target
+
+# Name "xml - Win32 Release"
+# Name "xml - Win32 Debug"
+# Name "xml - x64 Release"
+# Name "xml - x64 Debug"
+# Begin Group "Source Files"
+
+# PROP Default_Filter "cpp;c;cxx;rc;def;r;odl;idl;hpj;bat"
+# Begin Source File
+
+SOURCE=.\xmlparse.c
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmlrole.c
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok.c
+# End Source File
+# Begin Source File
+
+SOURCE=xmltok_impl.c
+# PROP Exclude_From_Build 1
+# End Source File
+# Begin Source File
+
+SOURCE=xmltok_ns.c
+# PROP Exclude_From_Build 1
+# End Source File
+# End Group
+# Begin Group "Header Files"
+
+# PROP Default_Filter "h;hpp;hxx;hm;inl"
+# Begin Source File
+
+SOURCE=.\ascii.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\asciitab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\config.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\expat.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\iasciitab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\latin1tab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\nametab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\utf8tab.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmlrole.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok.h
+# End Source File
+# Begin Source File
+
+SOURCE=.\xmltok_impl.h
+# End Source File
+# End Group
+# Begin Group "Generated Header Files"
+
+# PROP Default_Filter ""
+# Begin Source File
+
+SOURCE=.\expat.h.in
+
+!IF  "$(CFG)" == "xml - Win32 Release"
+
+# Begin Custom Build - Creating expat.h from expat.h.in 
+InputPath=.\expat.h.in
+
+".\expat.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\expat.h.in > .\expat.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - Win32 Debug"
+
+# Begin Custom Build - Creating expat.h from expat.h.in 
+InputPath=.\expat.h.in
+
+".\expat.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\expat.h.in > .\expat.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - x64 Release"
+
+# Begin Custom Build - Creating expat.h from expat.h.in 
+InputPath=.\expat.h.in
+
+".\expat.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\expat.h.in > .\expat.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - x64 Debug"
+
+# Begin Custom Build - Creating expat.h from expat.h.in 
+InputPath=.\expat.h.in
+
+".\expat.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\expat.h.in > .\expat.h
+	
+# End Custom Build
+
+!ENDIF 
+
+# End Source File
+# Begin Source File
+
+SOURCE=.\winconfig.h
+
+!IF  "$(CFG)" == "xml - Win32 Release"
+
+# Begin Custom Build - Creating config.h from winconfig.h 
+InputPath=.\winconfig.h
+
+".\config.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\winconfig.h > .\config.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - Win32 Debug"
+
+# Begin Custom Build - Creating config.h from winconfig.h 
+InputPath=.\winconfig.h
+
+".\config.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\winconfig.h > .\config.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - x64 Release"
+
+# Begin Custom Build - Creating config.h from winconfig.h 
+InputPath=.\winconfig.h
+
+".\config.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\winconfig.h > .\config.h
+	
+# End Custom Build
+
+!ELSEIF  "$(CFG)" == "xml - x64 Debug"
+
+# Begin Custom Build - Creating config.h from winconfig.h 
+InputPath=.\winconfig.h
+
+".\config.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)"
+	type .\winconfig.h > .\config.h
+	
+# End Custom Build
+
+!ENDIF 
+
+# End Source File
+# End Group
+# Begin Source File
+
+SOURCE=.\ReadMe.txt
+# End Source File
+# End Target
+# End Project
diff --git a/xml/expat/lib/xmlparse.c b/xml/expat/lib/xmlparse.c
new file mode 100644
index 000000000..43eb307f9
--- /dev/null
+++ b/xml/expat/lib/xmlparse.c
@@ -0,0 +1,4641 @@
+/*
+Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+static char RCSId[]
+  = "$Header: /home/cvs/apr-util/xml/expat/lib/xmlparse.c,v 1.4 2001/08/30 05:44:18 wrowe Exp $";
+
+#ifdef COMPILED_FROM_DSP
+#  include "winconfig.h"
+#  define XMLPARSEAPI __declspec(dllexport)
+#  include "expat.h"
+#  undef XMLPARSEAPI
+#else
+#include <config.h>
+
+#ifdef HAVE_STRING_H
+#  include <string.h>
+#endif
+
+#ifndef __CYGWIN__
+#ifdef __declspec
+#  define XMLPARSEAPI __declspec(dllexport)
+#endif
+#endif
+
+#include "expat.h"
+
+#ifdef __declspec
+#  undef XMLPARSEAPI
+#endif
+#endif /* ndef COMPILED_FROM_DSP */
+
+#include <stddef.h>
+
+#ifdef XML_UNICODE
+#define XML_ENCODE_MAX XML_UTF16_ENCODE_MAX
+#define XmlConvert XmlUtf16Convert
+#define XmlGetInternalEncoding XmlGetUtf16InternalEncoding
+#define XmlGetInternalEncodingNS XmlGetUtf16InternalEncodingNS
+#define XmlEncode XmlUtf16Encode
+#define MUST_CONVERT(enc, s) (!(enc)->isUtf16 || (((unsigned long)s) & 1))
+typedef unsigned short ICHAR;
+#else
+#define XML_ENCODE_MAX XML_UTF8_ENCODE_MAX
+#define XmlConvert XmlUtf8Convert
+#define XmlGetInternalEncoding XmlGetUtf8InternalEncoding
+#define XmlGetInternalEncodingNS XmlGetUtf8InternalEncodingNS
+#define XmlEncode XmlUtf8Encode
+#define MUST_CONVERT(enc, s) (!(enc)->isUtf8)
+typedef char ICHAR;
+#endif
+
+
+#ifndef XML_NS
+
+#define XmlInitEncodingNS XmlInitEncoding
+#define XmlInitUnknownEncodingNS XmlInitUnknownEncoding
+#undef XmlGetInternalEncodingNS
+#define XmlGetInternalEncodingNS XmlGetInternalEncoding
+#define XmlParseXmlDeclNS XmlParseXmlDecl
+
+#endif
+
+#ifdef XML_UNICODE_WCHAR_T
+#define XML_T(x) L ## x
+#else
+#define XML_T(x) x
+#endif
+
+/* Round up n to be a multiple of sz, where sz is a power of 2. */
+#define ROUND_UP(n, sz) (((n) + ((sz) - 1)) & ~((sz) - 1))
+
+#include "xmltok.h"
+#include "xmlrole.h"
+
+typedef const XML_Char *KEY;
+
+typedef struct {
+  KEY name;
+} NAMED;
+
+typedef struct {
+  NAMED **v;
+  size_t size;
+  size_t used;
+  size_t usedLim;
+  XML_Memory_Handling_Suite *mem;
+} HASH_TABLE;
+
+typedef struct {
+  NAMED **p;
+  NAMED **end;
+} HASH_TABLE_ITER;
+
+#define INIT_TAG_BUF_SIZE 32  /* must be a multiple of sizeof(XML_Char) */
+#define INIT_DATA_BUF_SIZE 1024
+#define INIT_ATTS_SIZE 16
+#define INIT_BLOCK_SIZE 1024
+#define INIT_BUFFER_SIZE 1024
+
+#define EXPAND_SPARE 24
+
+typedef struct binding {
+  struct prefix *prefix;
+  struct binding *nextTagBinding;
+  struct binding *prevPrefixBinding;
+  const struct attribute_id *attId;
+  XML_Char *uri;
+  int uriLen;
+  int uriAlloc;
+} BINDING;
+
+typedef struct prefix {
+  const XML_Char *name;
+  BINDING *binding;
+} PREFIX;
+
+typedef struct {
+  const XML_Char *str;
+  const XML_Char *localPart;
+  int uriLen;
+} TAG_NAME;
+
+typedef struct tag {
+  struct tag *parent;
+  const char *rawName;
+  int rawNameLength;
+  TAG_NAME name;
+  char *buf;
+  char *bufEnd;
+  BINDING *bindings;
+} TAG;
+
+typedef struct {
+  const XML_Char *name;
+  const XML_Char *textPtr;
+  int textLen;
+  const XML_Char *systemId;
+  const XML_Char *base;
+  const XML_Char *publicId;
+  const XML_Char *notation;
+  char open;
+  char is_param;
+} ENTITY;
+
+typedef struct {
+  enum XML_Content_Type		type;
+  enum XML_Content_Quant	quant;
+  const XML_Char *		name;
+  int				firstchild;
+  int				lastchild;
+  int				childcnt;
+  int				nextsib;
+} CONTENT_SCAFFOLD;
+
+typedef struct block {
+  struct block *next;
+  int size;
+  XML_Char s[1];
+} BLOCK;
+
+typedef struct {
+  BLOCK *blocks;
+  BLOCK *freeBlocks;
+  const XML_Char *end;
+  XML_Char *ptr;
+  XML_Char *start;
+  XML_Memory_Handling_Suite *mem;
+} STRING_POOL;
+
+/* The XML_Char before the name is used to determine whether
+an attribute has been specified. */
+typedef struct attribute_id {
+  XML_Char *name;
+  PREFIX *prefix;
+  char maybeTokenized;
+  char xmlns;
+} ATTRIBUTE_ID;
+
+typedef struct {
+  const ATTRIBUTE_ID *id;
+  char isCdata;
+  const XML_Char *value;
+} DEFAULT_ATTRIBUTE;
+
+typedef struct {
+  const XML_Char *name;
+  PREFIX *prefix;
+  const ATTRIBUTE_ID *idAtt;
+  int nDefaultAtts;
+  int allocDefaultAtts;
+  DEFAULT_ATTRIBUTE *defaultAtts;
+} ELEMENT_TYPE;
+
+typedef struct {
+  HASH_TABLE generalEntities;
+  HASH_TABLE elementTypes;
+  HASH_TABLE attributeIds;
+  HASH_TABLE prefixes;
+  STRING_POOL pool;
+  int complete;
+  int standalone;
+#ifdef XML_DTD
+  HASH_TABLE paramEntities;
+#endif /* XML_DTD */
+  PREFIX defaultPrefix;
+  /* === scaffolding for building content model === */
+  int in_eldecl;
+  CONTENT_SCAFFOLD *scaffold;
+  unsigned contentStringLen;
+  unsigned scaffSize;
+  unsigned scaffCount;
+  int scaffLevel;
+  int *scaffIndex;
+} DTD;
+
+typedef struct open_internal_entity {
+  const char *internalEventPtr;
+  const char *internalEventEndPtr;
+  struct open_internal_entity *next;
+  ENTITY *entity;
+} OPEN_INTERNAL_ENTITY;
+
+typedef enum XML_Error Processor(XML_Parser parser,
+				 const char *start,
+				 const char *end,
+				 const char **endPtr);
+
+static Processor prologProcessor;
+static Processor prologInitProcessor;
+static Processor contentProcessor;
+static Processor cdataSectionProcessor;
+#ifdef XML_DTD
+static Processor ignoreSectionProcessor;
+#endif /* XML_DTD */
+static Processor epilogProcessor;
+static Processor errorProcessor;
+static Processor externalEntityInitProcessor;
+static Processor externalEntityInitProcessor2;
+static Processor externalEntityInitProcessor3;
+static Processor externalEntityContentProcessor;
+
+static enum XML_Error
+handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName);
+static enum XML_Error
+processXmlDecl(XML_Parser parser, int isGeneralTextEntity, const char *, const char *);
+static enum XML_Error
+initializeEncoding(XML_Parser parser);
+static enum XML_Error
+doProlog(XML_Parser parser, const ENCODING *enc, const char *s,
+	 const char *end, int tok, const char *next, const char **nextPtr);
+static enum XML_Error
+processInternalParamEntity(XML_Parser parser, ENTITY *entity);
+static enum XML_Error
+doContent(XML_Parser parser, int startTagLevel, const ENCODING *enc,
+	  const char *start, const char *end, const char **endPtr);
+static enum XML_Error
+doCdataSection(XML_Parser parser, const ENCODING *, const char **startPtr, const char *end, const char **nextPtr);
+#ifdef XML_DTD
+static enum XML_Error
+doIgnoreSection(XML_Parser parser, const ENCODING *, const char **startPtr, const char *end, const char **nextPtr);
+#endif /* XML_DTD */
+static enum XML_Error storeAtts(XML_Parser parser, const ENCODING *, const char *s,
+				TAG_NAME *tagNamePtr, BINDING **bindingsPtr);
+static
+int addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, const XML_Char *uri, BINDING **bindingsPtr);
+
+static int
+defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *,
+		int isCdata, int isId, const XML_Char *dfltValue,
+		XML_Parser parser);
+
+static enum XML_Error
+storeAttributeValue(XML_Parser parser, const ENCODING *, int isCdata, const char *, const char *,
+		    STRING_POOL *);
+static enum XML_Error
+appendAttributeValue(XML_Parser parser, const ENCODING *, int isCdata, const char *, const char *,
+		    STRING_POOL *);
+static ATTRIBUTE_ID *
+getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start, const char *end);
+static int setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *);
+static enum XML_Error
+storeEntityValue(XML_Parser parser, const ENCODING *enc, const char *start, const char *end);
+static int
+reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, const char *start, const char *end);
+static int
+reportComment(XML_Parser parser, const ENCODING *enc, const char *start, const char *end);
+static void
+reportDefault(XML_Parser parser, const ENCODING *enc, const char *start, const char *end);
+
+static const XML_Char *getContext(XML_Parser parser);
+static int setContext(XML_Parser parser, const XML_Char *context);
+static void normalizePublicId(XML_Char *s);
+static int dtdInit(DTD *, XML_Parser parser);
+
+static void dtdDestroy(DTD *, XML_Parser parser);
+
+static int dtdCopy(DTD *newDtd, const DTD *oldDtd, XML_Parser parser);
+
+static int copyEntityTable(HASH_TABLE *, STRING_POOL *, const HASH_TABLE *,
+			   XML_Parser parser);
+
+#ifdef XML_DTD
+static void dtdSwap(DTD *, DTD *);
+#endif /* XML_DTD */
+
+static NAMED *lookup(HASH_TABLE *table, KEY name, size_t createSize);
+
+static void hashTableInit(HASH_TABLE *, XML_Memory_Handling_Suite *ms);
+
+static void hashTableDestroy(HASH_TABLE *);
+static void hashTableIterInit(HASH_TABLE_ITER *, const HASH_TABLE *);
+static NAMED *hashTableIterNext(HASH_TABLE_ITER *);
+static void poolInit(STRING_POOL *, XML_Memory_Handling_Suite *ms);
+static void poolClear(STRING_POOL *);
+static void poolDestroy(STRING_POOL *);
+static XML_Char *poolAppend(STRING_POOL *pool, const ENCODING *enc,
+			    const char *ptr, const char *end);
+static XML_Char *poolStoreString(STRING_POOL *pool, const ENCODING *enc,
+				  const char *ptr, const char *end);
+
+static int poolGrow(STRING_POOL *pool);
+
+static int nextScaffoldPart(XML_Parser parser);
+static XML_Content *build_model(XML_Parser parser);
+
+static const XML_Char *poolCopyString(STRING_POOL *pool, const XML_Char *s);
+static const XML_Char *poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n);
+static const XML_Char *poolAppendString(STRING_POOL *pool, const XML_Char *s);
+static ELEMENT_TYPE * getElementType(XML_Parser Paraser,
+				     const ENCODING *enc,
+				     const char *ptr,
+				     const char *end);
+
+#define poolStart(pool) ((pool)->start)
+#define poolEnd(pool) ((pool)->ptr)
+#define poolLength(pool) ((pool)->ptr - (pool)->start)
+#define poolChop(pool) ((void)--(pool->ptr))
+#define poolLastChar(pool) (((pool)->ptr)[-1])
+#define poolDiscard(pool) ((pool)->ptr = (pool)->start)
+#define poolFinish(pool) ((pool)->start = (pool)->ptr)
+#define poolAppendChar(pool, c) \
+  (((pool)->ptr == (pool)->end && !poolGrow(pool)) \
+   ? 0 \
+   : ((*((pool)->ptr)++ = c), 1))
+
+typedef struct {
+  /* The first member must be userData so that the XML_GetUserData macro works. */
+  void *m_userData;
+  void *m_handlerArg;
+  char *m_buffer;
+  XML_Memory_Handling_Suite m_mem;
+  /* first character to be parsed */
+  const char *m_bufferPtr;
+  /* past last character to be parsed */
+  char *m_bufferEnd;
+  /* allocated end of buffer */
+  const char *m_bufferLim;
+  long m_parseEndByteIndex;
+  const char *m_parseEndPtr;
+  XML_Char *m_dataBuf;
+  XML_Char *m_dataBufEnd;
+  XML_StartElementHandler m_startElementHandler;
+  XML_EndElementHandler m_endElementHandler;
+  XML_CharacterDataHandler m_characterDataHandler;
+  XML_ProcessingInstructionHandler m_processingInstructionHandler;
+  XML_CommentHandler m_commentHandler;
+  XML_StartCdataSectionHandler m_startCdataSectionHandler;
+  XML_EndCdataSectionHandler m_endCdataSectionHandler;
+  XML_DefaultHandler m_defaultHandler;
+  XML_StartDoctypeDeclHandler m_startDoctypeDeclHandler;
+  XML_EndDoctypeDeclHandler m_endDoctypeDeclHandler;
+  XML_UnparsedEntityDeclHandler m_unparsedEntityDeclHandler;
+  XML_NotationDeclHandler m_notationDeclHandler;
+  XML_StartNamespaceDeclHandler m_startNamespaceDeclHandler;
+  XML_EndNamespaceDeclHandler m_endNamespaceDeclHandler;
+  XML_NotStandaloneHandler m_notStandaloneHandler;
+  XML_ExternalEntityRefHandler m_externalEntityRefHandler;
+  void *m_externalEntityRefHandlerArg;
+  XML_UnknownEncodingHandler m_unknownEncodingHandler;
+  XML_ElementDeclHandler m_elementDeclHandler;
+  XML_AttlistDeclHandler m_attlistDeclHandler;
+  XML_EntityDeclHandler m_entityDeclHandler;
+  XML_XmlDeclHandler m_xmlDeclHandler;
+  const ENCODING *m_encoding;
+  INIT_ENCODING m_initEncoding;
+  const ENCODING *m_internalEncoding;
+  const XML_Char *m_protocolEncodingName;
+  int m_ns;
+  int m_ns_triplets;
+  void *m_unknownEncodingMem;
+  void *m_unknownEncodingData;
+  void *m_unknownEncodingHandlerData;
+  void (*m_unknownEncodingRelease)(void *);
+  PROLOG_STATE m_prologState;
+  Processor *m_processor;
+  enum XML_Error m_errorCode;
+  const char *m_eventPtr;
+  const char *m_eventEndPtr;
+  const char *m_positionPtr;
+  OPEN_INTERNAL_ENTITY *m_openInternalEntities;
+  int m_defaultExpandInternalEntities;
+  int m_tagLevel;
+  ENTITY *m_declEntity;
+  const XML_Char *m_doctypeName;
+  const XML_Char *m_doctypeSysid;
+  const XML_Char *m_doctypePubid;
+  const XML_Char *m_declAttributeType;
+  const XML_Char *m_declNotationName;
+  const XML_Char *m_declNotationPublicId;
+  ELEMENT_TYPE *m_declElementType;
+  ATTRIBUTE_ID *m_declAttributeId;
+  char m_declAttributeIsCdata;
+  char m_declAttributeIsId;
+  DTD m_dtd;
+  const XML_Char *m_curBase;
+  TAG *m_tagStack;
+  TAG *m_freeTagList;
+  BINDING *m_inheritedBindings;
+  BINDING *m_freeBindingList;
+  int m_attsSize;
+  int m_nSpecifiedAtts;
+  int m_idAttIndex;
+  ATTRIBUTE *m_atts;
+  POSITION m_position;
+  STRING_POOL m_tempPool;
+  STRING_POOL m_temp2Pool;
+  char *m_groupConnector;
+  unsigned m_groupSize;
+  int m_hadExternalDoctype;
+  XML_Char m_namespaceSeparator;
+#ifdef XML_DTD
+  enum XML_ParamEntityParsing m_paramEntityParsing;
+  XML_Parser m_parentParser;
+#endif
+} Parser;
+
+#define MALLOC(s) (((Parser *)parser)->m_mem.malloc_fcn((s)))
+#define REALLOC(p,s) (((Parser *)parser)->m_mem.realloc_fcn((p),(s)))
+#define FREE(p) (((Parser *)parser)->m_mem.free_fcn((p)))
+
+#define userData (((Parser *)parser)->m_userData)
+#define handlerArg (((Parser *)parser)->m_handlerArg)
+#define startElementHandler (((Parser *)parser)->m_startElementHandler)
+#define endElementHandler (((Parser *)parser)->m_endElementHandler)
+#define characterDataHandler (((Parser *)parser)->m_characterDataHandler)
+#define processingInstructionHandler (((Parser *)parser)->m_processingInstructionHandler)
+#define commentHandler (((Parser *)parser)->m_commentHandler)
+#define startCdataSectionHandler (((Parser *)parser)->m_startCdataSectionHandler)
+#define endCdataSectionHandler (((Parser *)parser)->m_endCdataSectionHandler)
+#define defaultHandler (((Parser *)parser)->m_defaultHandler)
+#define startDoctypeDeclHandler (((Parser *)parser)->m_startDoctypeDeclHandler)
+#define endDoctypeDeclHandler (((Parser *)parser)->m_endDoctypeDeclHandler)
+#define unparsedEntityDeclHandler (((Parser *)parser)->m_unparsedEntityDeclHandler)
+#define notationDeclHandler (((Parser *)parser)->m_notationDeclHandler)
+#define startNamespaceDeclHandler (((Parser *)parser)->m_startNamespaceDeclHandler)
+#define endNamespaceDeclHandler (((Parser *)parser)->m_endNamespaceDeclHandler)
+#define notStandaloneHandler (((Parser *)parser)->m_notStandaloneHandler)
+#define externalEntityRefHandler (((Parser *)parser)->m_externalEntityRefHandler)
+#define externalEntityRefHandlerArg (((Parser *)parser)->m_externalEntityRefHandlerArg)
+#define unknownEncodingHandler (((Parser *)parser)->m_unknownEncodingHandler)
+#define elementDeclHandler (((Parser *)parser)->m_elementDeclHandler)
+#define attlistDeclHandler (((Parser *)parser)->m_attlistDeclHandler)
+#define entityDeclHandler (((Parser *)parser)->m_entityDeclHandler)
+#define xmlDeclHandler (((Parser *)parser)->m_xmlDeclHandler)
+#define encoding (((Parser *)parser)->m_encoding)
+#define initEncoding (((Parser *)parser)->m_initEncoding)
+#define internalEncoding (((Parser *)parser)->m_internalEncoding)
+#define unknownEncodingMem (((Parser *)parser)->m_unknownEncodingMem)
+#define unknownEncodingData (((Parser *)parser)->m_unknownEncodingData)
+#define unknownEncodingHandlerData \
+  (((Parser *)parser)->m_unknownEncodingHandlerData)
+#define unknownEncodingRelease (((Parser *)parser)->m_unknownEncodingRelease)
+#define protocolEncodingName (((Parser *)parser)->m_protocolEncodingName)
+#define ns (((Parser *)parser)->m_ns)
+#define ns_triplets (((Parser *)parser)->m_ns_triplets)
+#define prologState (((Parser *)parser)->m_prologState)
+#define processor (((Parser *)parser)->m_processor)
+#define errorCode (((Parser *)parser)->m_errorCode)
+#define eventPtr (((Parser *)parser)->m_eventPtr)
+#define eventEndPtr (((Parser *)parser)->m_eventEndPtr)
+#define positionPtr (((Parser *)parser)->m_positionPtr)
+#define position (((Parser *)parser)->m_position)
+#define openInternalEntities (((Parser *)parser)->m_openInternalEntities)
+#define defaultExpandInternalEntities (((Parser *)parser)->m_defaultExpandInternalEntities)
+#define tagLevel (((Parser *)parser)->m_tagLevel)
+#define buffer (((Parser *)parser)->m_buffer)
+#define bufferPtr (((Parser *)parser)->m_bufferPtr)
+#define bufferEnd (((Parser *)parser)->m_bufferEnd)
+#define parseEndByteIndex (((Parser *)parser)->m_parseEndByteIndex)
+#define parseEndPtr (((Parser *)parser)->m_parseEndPtr)
+#define bufferLim (((Parser *)parser)->m_bufferLim)
+#define dataBuf (((Parser *)parser)->m_dataBuf)
+#define dataBufEnd (((Parser *)parser)->m_dataBufEnd)
+#define dtd (((Parser *)parser)->m_dtd)
+#define curBase (((Parser *)parser)->m_curBase)
+#define declEntity (((Parser *)parser)->m_declEntity)
+#define doctypeName (((Parser *)parser)->m_doctypeName)
+#define doctypeSysid (((Parser *)parser)->m_doctypeSysid)
+#define doctypePubid (((Parser *)parser)->m_doctypePubid)
+#define declAttributeType (((Parser *)parser)->m_declAttributeType)
+#define declNotationName (((Parser *)parser)->m_declNotationName)
+#define declNotationPublicId (((Parser *)parser)->m_declNotationPublicId)
+#define declElementType (((Parser *)parser)->m_declElementType)
+#define declAttributeId (((Parser *)parser)->m_declAttributeId)
+#define declAttributeIsCdata (((Parser *)parser)->m_declAttributeIsCdata)
+#define declAttributeIsId (((Parser *)parser)->m_declAttributeIsId)
+#define freeTagList (((Parser *)parser)->m_freeTagList)
+#define freeBindingList (((Parser *)parser)->m_freeBindingList)
+#define inheritedBindings (((Parser *)parser)->m_inheritedBindings)
+#define tagStack (((Parser *)parser)->m_tagStack)
+#define atts (((Parser *)parser)->m_atts)
+#define attsSize (((Parser *)parser)->m_attsSize)
+#define nSpecifiedAtts (((Parser *)parser)->m_nSpecifiedAtts)
+#define idAttIndex (((Parser *)parser)->m_idAttIndex)
+#define tempPool (((Parser *)parser)->m_tempPool)
+#define temp2Pool (((Parser *)parser)->m_temp2Pool)
+#define groupConnector (((Parser *)parser)->m_groupConnector)
+#define groupSize (((Parser *)parser)->m_groupSize)
+#define hadExternalDoctype (((Parser *)parser)->m_hadExternalDoctype)
+#define namespaceSeparator (((Parser *)parser)->m_namespaceSeparator)
+#ifdef XML_DTD
+#define parentParser (((Parser *)parser)->m_parentParser)
+#define paramEntityParsing (((Parser *)parser)->m_paramEntityParsing)
+#endif /* XML_DTD */
+
+#ifdef COMPILED_FROM_DSP
+BOOL WINAPI DllMain(HINSTANCE h, DWORD r, LPVOID p) {
+  return TRUE;
+}
+#endif /* def COMPILED_FROM_DSP */
+
+#ifdef _MSC_VER
+#ifdef _DEBUG
+Parser *asParser(XML_Parser parser)
+{
+  return parser;
+}
+#endif
+#endif
+
+XML_Parser XML_ParserCreate(const XML_Char *encodingName)
+{
+  return XML_ParserCreate_MM(encodingName, NULL, NULL);
+}
+
+XML_Parser XML_ParserCreateNS(const XML_Char *encodingName, XML_Char nsSep)
+{
+  XML_Char tmp[2];
+  *tmp = nsSep;
+  return XML_ParserCreate_MM(encodingName, NULL, tmp);
+}
+
+XML_Parser
+XML_ParserCreate_MM(const XML_Char *encodingName,
+		    const XML_Memory_Handling_Suite *memsuite,
+		    const XML_Char *nameSep) {
+  
+  XML_Parser parser;
+  static
+  const XML_Char implicitContext[] = {
+    XML_T('x'), XML_T('m'), XML_T('l'), XML_T('='),
+    XML_T('h'), XML_T('t'), XML_T('t'), XML_T('p'), XML_T(':'),
+    XML_T('/'), XML_T('/'), XML_T('w'), XML_T('w'), XML_T('w'),
+    XML_T('.'), XML_T('w'), XML_T('3'),
+    XML_T('.'), XML_T('o'), XML_T('r'), XML_T('g'),
+    XML_T('/'), XML_T('X'), XML_T('M'), XML_T('L'),
+    XML_T('/'), XML_T('1'), XML_T('9'), XML_T('9'), XML_T('8'),
+    XML_T('/'), XML_T('n'), XML_T('a'), XML_T('m'), XML_T('e'),
+    XML_T('s'), XML_T('p'), XML_T('a'), XML_T('c'), XML_T('e'),
+    XML_T('\0')
+  };
+
+
+  if (memsuite) {
+    XML_Memory_Handling_Suite *mtemp;
+    parser = memsuite->malloc_fcn(sizeof(Parser));
+    mtemp = &(((Parser *) parser)->m_mem);
+    mtemp->malloc_fcn = memsuite->malloc_fcn;
+    mtemp->realloc_fcn = memsuite->realloc_fcn;
+    mtemp->free_fcn = memsuite->free_fcn;
+  }
+  else {
+    XML_Memory_Handling_Suite *mtemp;
+    parser = malloc(sizeof(Parser));
+    mtemp = &(((Parser *) parser)->m_mem);
+    mtemp->malloc_fcn = malloc;
+    mtemp->realloc_fcn = realloc;
+    mtemp->free_fcn = free;
+  }
+
+  if (!parser)
+    return parser;
+  processor = prologInitProcessor;
+  XmlPrologStateInit(&prologState);
+  userData = 0;
+  handlerArg = 0;
+  startElementHandler = 0;
+  endElementHandler = 0;
+  characterDataHandler = 0;
+  processingInstructionHandler = 0;
+  commentHandler = 0;
+  startCdataSectionHandler = 0;
+  endCdataSectionHandler = 0;
+  defaultHandler = 0;
+  startDoctypeDeclHandler = 0;
+  endDoctypeDeclHandler = 0;
+  unparsedEntityDeclHandler = 0;
+  notationDeclHandler = 0;
+  startNamespaceDeclHandler = 0;
+  endNamespaceDeclHandler = 0;
+  notStandaloneHandler = 0;
+  externalEntityRefHandler = 0;
+  externalEntityRefHandlerArg = parser;
+  unknownEncodingHandler = 0;
+  elementDeclHandler = 0;
+  attlistDeclHandler = 0;
+  entityDeclHandler = 0;
+  xmlDeclHandler = 0;
+  buffer = 0;
+  bufferPtr = 0;
+  bufferEnd = 0;
+  parseEndByteIndex = 0;
+  parseEndPtr = 0;
+  bufferLim = 0;
+  declElementType = 0;
+  declAttributeId = 0;
+  declEntity = 0;
+  doctypeName = 0;
+  doctypeSysid = 0;
+  doctypePubid = 0;
+  declAttributeType = 0;
+  declNotationName = 0;
+  declNotationPublicId = 0;
+  memset(&position, 0, sizeof(POSITION));
+  errorCode = XML_ERROR_NONE;
+  eventPtr = 0;
+  eventEndPtr = 0;
+  positionPtr = 0;
+  openInternalEntities = 0;
+  tagLevel = 0;
+  tagStack = 0;
+  freeTagList = 0;
+  freeBindingList = 0;
+  inheritedBindings = 0;
+  attsSize = INIT_ATTS_SIZE;
+  atts = MALLOC(attsSize * sizeof(ATTRIBUTE));
+  nSpecifiedAtts = 0;
+  dataBuf = MALLOC(INIT_DATA_BUF_SIZE * sizeof(XML_Char));
+  groupSize = 0;
+  groupConnector = 0;
+  hadExternalDoctype = 0;
+  unknownEncodingMem = 0;
+  unknownEncodingRelease = 0;
+  unknownEncodingData = 0;
+  unknownEncodingHandlerData = 0;
+  namespaceSeparator = '!';
+#ifdef XML_DTD
+  parentParser = 0;
+  paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
+#endif
+  ns = 0;
+  ns_triplets = 0;
+  poolInit(&tempPool, &(((Parser *) parser)->m_mem));
+  poolInit(&temp2Pool, &(((Parser *) parser)->m_mem));
+  protocolEncodingName = encodingName ? poolCopyString(&tempPool, encodingName) : 0;
+  curBase = 0;
+  if (!dtdInit(&dtd, parser) || !atts || !dataBuf
+      || (encodingName && !protocolEncodingName)) {
+    XML_ParserFree(parser);
+    return 0;
+  }
+  dataBufEnd = dataBuf + INIT_DATA_BUF_SIZE;
+
+  if (nameSep) {
+    XmlInitEncodingNS(&initEncoding, &encoding, 0);
+    ns = 1;
+    internalEncoding = XmlGetInternalEncodingNS();
+    namespaceSeparator = *nameSep;
+
+    if (! setContext(parser, implicitContext)) {
+      XML_ParserFree(parser);
+      return 0;
+    }
+  }
+  else {
+    XmlInitEncoding(&initEncoding, &encoding, 0);
+    internalEncoding = XmlGetInternalEncoding();
+  }
+
+  return parser;
+}  /* End XML_ParserCreate_MM */
+
+int XML_SetEncoding(XML_Parser parser, const XML_Char *encodingName)
+{
+  if (!encodingName)
+    protocolEncodingName = 0;
+  else {
+    protocolEncodingName = poolCopyString(&tempPool, encodingName);
+    if (!protocolEncodingName)
+      return 0;
+  }
+  return 1;
+}
+
+XML_Parser XML_ExternalEntityParserCreate(XML_Parser oldParser,
+					  const XML_Char *context,
+					  const XML_Char *encodingName)
+{
+  XML_Parser parser = oldParser;
+  DTD *oldDtd = &dtd;
+  XML_StartElementHandler oldStartElementHandler = startElementHandler;
+  XML_EndElementHandler oldEndElementHandler = endElementHandler;
+  XML_CharacterDataHandler oldCharacterDataHandler = characterDataHandler;
+  XML_ProcessingInstructionHandler oldProcessingInstructionHandler = processingInstructionHandler;
+  XML_CommentHandler oldCommentHandler = commentHandler;
+  XML_StartCdataSectionHandler oldStartCdataSectionHandler = startCdataSectionHandler;
+  XML_EndCdataSectionHandler oldEndCdataSectionHandler = endCdataSectionHandler;
+  XML_DefaultHandler oldDefaultHandler = defaultHandler;
+  XML_UnparsedEntityDeclHandler oldUnparsedEntityDeclHandler = unparsedEntityDeclHandler;
+  XML_NotationDeclHandler oldNotationDeclHandler = notationDeclHandler;
+  XML_StartNamespaceDeclHandler oldStartNamespaceDeclHandler = startNamespaceDeclHandler;
+  XML_EndNamespaceDeclHandler oldEndNamespaceDeclHandler = endNamespaceDeclHandler;
+  XML_NotStandaloneHandler oldNotStandaloneHandler = notStandaloneHandler;
+  XML_ExternalEntityRefHandler oldExternalEntityRefHandler = externalEntityRefHandler;
+  XML_UnknownEncodingHandler oldUnknownEncodingHandler = unknownEncodingHandler;
+  XML_ElementDeclHandler oldElementDeclHandler = elementDeclHandler;
+  XML_AttlistDeclHandler oldAttlistDeclHandler = attlistDeclHandler;
+  XML_EntityDeclHandler oldEntityDeclHandler = entityDeclHandler;
+  XML_XmlDeclHandler oldXmlDeclHandler = xmlDeclHandler;
+  ELEMENT_TYPE * oldDeclElementType = declElementType;
+
+  void *oldUserData = userData;
+  void *oldHandlerArg = handlerArg;
+  int oldDefaultExpandInternalEntities = defaultExpandInternalEntities;
+  void *oldExternalEntityRefHandlerArg = externalEntityRefHandlerArg;
+#ifdef XML_DTD
+  int oldParamEntityParsing = paramEntityParsing;
+#endif
+  int oldns_triplets = ns_triplets;
+
+  if (ns) {
+    XML_Char tmp[2];
+
+    *tmp = namespaceSeparator;
+    parser = XML_ParserCreate_MM(encodingName, &((Parser *)parser)->m_mem,
+				 tmp);
+  }
+  else {
+    parser = XML_ParserCreate_MM(encodingName, &((Parser *)parser)->m_mem,
+				 NULL);
+  }
+
+  if (!parser)
+    return 0;
+
+  startElementHandler = oldStartElementHandler;
+  endElementHandler = oldEndElementHandler;
+  characterDataHandler = oldCharacterDataHandler;
+  processingInstructionHandler = oldProcessingInstructionHandler;
+  commentHandler = oldCommentHandler;
+  startCdataSectionHandler = oldStartCdataSectionHandler;
+  endCdataSectionHandler = oldEndCdataSectionHandler;
+  defaultHandler = oldDefaultHandler;
+  unparsedEntityDeclHandler = oldUnparsedEntityDeclHandler;
+  notationDeclHandler = oldNotationDeclHandler;
+  startNamespaceDeclHandler = oldStartNamespaceDeclHandler;
+  endNamespaceDeclHandler = oldEndNamespaceDeclHandler;
+  notStandaloneHandler = oldNotStandaloneHandler;
+  externalEntityRefHandler = oldExternalEntityRefHandler;
+  unknownEncodingHandler = oldUnknownEncodingHandler;
+  elementDeclHandler = oldElementDeclHandler;
+  attlistDeclHandler = oldAttlistDeclHandler;
+  entityDeclHandler = oldEntityDeclHandler;
+  xmlDeclHandler = oldXmlDeclHandler;
+  declElementType = oldDeclElementType;
+  userData = oldUserData;
+  if (oldUserData == oldHandlerArg)
+    handlerArg = userData;
+  else
+    handlerArg = parser;
+  if (oldExternalEntityRefHandlerArg != oldParser)
+    externalEntityRefHandlerArg = oldExternalEntityRefHandlerArg;
+  defaultExpandInternalEntities = oldDefaultExpandInternalEntities;
+  ns_triplets = oldns_triplets;
+#ifdef XML_DTD
+  paramEntityParsing = oldParamEntityParsing;
+  if (context) {
+#endif /* XML_DTD */
+    if (!dtdCopy(&dtd, oldDtd, parser) || !setContext(parser, context)) {
+      XML_ParserFree(parser);
+      return 0;
+    }
+    processor = externalEntityInitProcessor;
+#ifdef XML_DTD
+  }
+  else {
+    dtdSwap(&dtd, oldDtd);
+    parentParser = oldParser;
+    XmlPrologStateInitExternalEntity(&prologState);
+    dtd.complete = 1;
+    hadExternalDoctype = 1;
+  }
+#endif /* XML_DTD */
+  return parser;
+}
+
+static
+void destroyBindings(BINDING *bindings, XML_Parser parser)
+{
+  for (;;) {
+    BINDING *b = bindings;
+    if (!b)
+      break;
+    bindings = b->nextTagBinding;
+    FREE(b->uri);
+    FREE(b);
+  }
+}
+
+void XML_ParserFree(XML_Parser parser)
+{
+  for (;;) {
+    TAG *p;
+    if (tagStack == 0) {
+      if (freeTagList == 0)
+	break;
+      tagStack = freeTagList;
+      freeTagList = 0;
+    }
+    p = tagStack;
+    tagStack = tagStack->parent;
+    FREE(p->buf);
+    destroyBindings(p->bindings, parser);
+    FREE(p);
+  }
+  destroyBindings(freeBindingList, parser);
+  destroyBindings(inheritedBindings, parser);
+  poolDestroy(&tempPool);
+  poolDestroy(&temp2Pool);
+#ifdef XML_DTD
+  if (parentParser) {
+    if (hadExternalDoctype)
+      dtd.complete = 0;
+    dtdSwap(&dtd, &((Parser *)parentParser)->m_dtd);
+  }
+#endif /* XML_DTD */
+  dtdDestroy(&dtd, parser);
+  FREE((void *)atts);
+  if (groupConnector)
+    FREE(groupConnector);
+  if (buffer)
+    FREE(buffer);
+  FREE(dataBuf);
+  if (unknownEncodingMem)
+    FREE(unknownEncodingMem);
+  if (unknownEncodingRelease)
+    unknownEncodingRelease(unknownEncodingData);
+  FREE(parser);
+}
+
+void XML_UseParserAsHandlerArg(XML_Parser parser)
+{
+  handlerArg = parser;
+}
+
+void
+XML_SetReturnNSTriplet(XML_Parser parser, int do_nst) {
+  ns_triplets = do_nst;
+}
+
+void XML_SetUserData(XML_Parser parser, void *p)
+{
+  if (handlerArg == userData)
+    handlerArg = userData = p;
+  else
+    userData = p;
+}
+
+int XML_SetBase(XML_Parser parser, const XML_Char *p)
+{
+  if (p) {
+    p = poolCopyString(&dtd.pool, p);
+    if (!p)
+      return 0;
+    curBase = p;
+  }
+  else
+    curBase = 0;
+  return 1;
+}
+
+const XML_Char *XML_GetBase(XML_Parser parser)
+{
+  return curBase;
+}
+
+int XML_GetSpecifiedAttributeCount(XML_Parser parser)
+{
+  return nSpecifiedAtts;
+}
+
+int XML_GetIdAttributeIndex(XML_Parser parser)
+{
+  return idAttIndex;
+}
+
+void XML_SetElementHandler(XML_Parser parser,
+			   XML_StartElementHandler start,
+			   XML_EndElementHandler end)
+{
+  startElementHandler = start;
+  endElementHandler = end;
+}
+
+void XML_SetStartElementHandler(XML_Parser parser,
+				XML_StartElementHandler start) {
+  startElementHandler = start;
+}
+
+void XML_SetEndElementHandler(XML_Parser parser,
+			      XML_EndElementHandler end) {
+  endElementHandler = end;
+}
+
+void XML_SetCharacterDataHandler(XML_Parser parser,
+				 XML_CharacterDataHandler handler)
+{
+  characterDataHandler = handler;
+}
+
+void XML_SetProcessingInstructionHandler(XML_Parser parser,
+					 XML_ProcessingInstructionHandler handler)
+{
+  processingInstructionHandler = handler;
+}
+
+void XML_SetCommentHandler(XML_Parser parser,
+			   XML_CommentHandler handler)
+{
+  commentHandler = handler;
+}
+
+void XML_SetCdataSectionHandler(XML_Parser parser,
+				XML_StartCdataSectionHandler start,
+			        XML_EndCdataSectionHandler end)
+{
+  startCdataSectionHandler = start;
+  endCdataSectionHandler = end;
+}
+
+void XML_SetStartCdataSectionHandler(XML_Parser parser,
+                                     XML_StartCdataSectionHandler start) {
+  startCdataSectionHandler = start;
+}
+
+void XML_SetEndCdataSectionHandler(XML_Parser parser,
+                                   XML_EndCdataSectionHandler end) {
+  endCdataSectionHandler = end;
+}
+
+void XML_SetDefaultHandler(XML_Parser parser,
+			   XML_DefaultHandler handler)
+{
+  defaultHandler = handler;
+  defaultExpandInternalEntities = 0;
+}
+
+void XML_SetDefaultHandlerExpand(XML_Parser parser,
+				 XML_DefaultHandler handler)
+{
+  defaultHandler = handler;
+  defaultExpandInternalEntities = 1;
+}
+
+void XML_SetDoctypeDeclHandler(XML_Parser parser,
+			       XML_StartDoctypeDeclHandler start,
+			       XML_EndDoctypeDeclHandler end)
+{
+  startDoctypeDeclHandler = start;
+  endDoctypeDeclHandler = end;
+}
+
+void XML_SetStartDoctypeDeclHandler(XML_Parser parser,
+				    XML_StartDoctypeDeclHandler start) {
+  startDoctypeDeclHandler = start;
+}
+
+void XML_SetEndDoctypeDeclHandler(XML_Parser parser,
+				  XML_EndDoctypeDeclHandler end) {
+  endDoctypeDeclHandler = end;
+}
+
+void XML_SetUnparsedEntityDeclHandler(XML_Parser parser,
+				      XML_UnparsedEntityDeclHandler handler)
+{
+  unparsedEntityDeclHandler = handler;
+}
+
+void XML_SetNotationDeclHandler(XML_Parser parser,
+				XML_NotationDeclHandler handler)
+{
+  notationDeclHandler = handler;
+}
+
+void XML_SetNamespaceDeclHandler(XML_Parser parser,
+				 XML_StartNamespaceDeclHandler start,
+				 XML_EndNamespaceDeclHandler end)
+{
+  startNamespaceDeclHandler = start;
+  endNamespaceDeclHandler = end;
+}
+
+void XML_SetStartNamespaceDeclHandler(XML_Parser parser,
+				      XML_StartNamespaceDeclHandler start) {
+  startNamespaceDeclHandler = start;
+}
+
+void XML_SetEndNamespaceDeclHandler(XML_Parser parser,
+				    XML_EndNamespaceDeclHandler end) {
+  endNamespaceDeclHandler = end;
+}
+
+
+void XML_SetNotStandaloneHandler(XML_Parser parser,
+				 XML_NotStandaloneHandler handler)
+{
+  notStandaloneHandler = handler;
+}
+
+void XML_SetExternalEntityRefHandler(XML_Parser parser,
+				     XML_ExternalEntityRefHandler handler)
+{
+  externalEntityRefHandler = handler;
+}
+
+void XML_SetExternalEntityRefHandlerArg(XML_Parser parser, void *arg)
+{
+  if (arg)
+    externalEntityRefHandlerArg = arg;
+  else
+    externalEntityRefHandlerArg = parser;
+}
+
+void XML_SetUnknownEncodingHandler(XML_Parser parser,
+				   XML_UnknownEncodingHandler handler,
+				   void *data)
+{
+  unknownEncodingHandler = handler;
+  unknownEncodingHandlerData = data;
+}
+
+void XML_SetElementDeclHandler(XML_Parser parser,
+			       XML_ElementDeclHandler eldecl)
+{
+  elementDeclHandler = eldecl;
+}
+
+void XML_SetAttlistDeclHandler(XML_Parser parser,
+			       XML_AttlistDeclHandler attdecl)
+{
+  attlistDeclHandler = attdecl;
+}
+
+void XML_SetEntityDeclHandler(XML_Parser parser,
+			      XML_EntityDeclHandler handler)
+{
+  entityDeclHandler = handler;
+}
+
+void XML_SetXmlDeclHandler(XML_Parser parser,
+			   XML_XmlDeclHandler handler) {
+  xmlDeclHandler = handler;
+}
+
+int XML_SetParamEntityParsing(XML_Parser parser,
+			      enum XML_ParamEntityParsing parsing)
+{
+#ifdef XML_DTD
+  paramEntityParsing = parsing;
+  return 1;
+#else
+  return parsing == XML_PARAM_ENTITY_PARSING_NEVER;
+#endif
+}
+
+int XML_Parse(XML_Parser parser, const char *s, int len, int isFinal)
+{
+  if (len == 0) {
+    if (!isFinal)
+      return 1;
+    positionPtr = bufferPtr;
+    errorCode = processor(parser, bufferPtr, parseEndPtr = bufferEnd, 0);
+    if (errorCode == XML_ERROR_NONE)
+      return 1;
+    eventEndPtr = eventPtr;
+    processor = errorProcessor;
+    return 0;
+  }
+#ifndef XML_CONTEXT_BYTES
+  else if (bufferPtr == bufferEnd) {
+    const char *end;
+    int nLeftOver;
+    parseEndByteIndex += len;
+    positionPtr = s;
+    if (isFinal) {
+      errorCode = processor(parser, s, parseEndPtr = s + len, 0);
+      if (errorCode == XML_ERROR_NONE)
+	return 1;
+      eventEndPtr = eventPtr;
+      processor = errorProcessor;
+      return 0;
+    }
+    errorCode = processor(parser, s, parseEndPtr = s + len, &end);
+    if (errorCode != XML_ERROR_NONE) {
+      eventEndPtr = eventPtr;
+      processor = errorProcessor;
+      return 0;
+    }
+    XmlUpdatePosition(encoding, positionPtr, end, &position);
+    nLeftOver = s + len - end;
+    if (nLeftOver) {
+      if (buffer == 0 || nLeftOver > bufferLim - buffer) {
+	/* FIXME avoid integer overflow */
+	buffer = buffer == 0 ? MALLOC(len * 2) : REALLOC(buffer, len * 2);
+	/* FIXME storage leak if realloc fails */
+	if (!buffer) {
+	  errorCode = XML_ERROR_NO_MEMORY;
+	  eventPtr = eventEndPtr = 0;
+	  processor = errorProcessor;
+	  return 0;
+	}
+	bufferLim = buffer + len * 2;
+      }
+      memcpy(buffer, end, nLeftOver);
+      bufferPtr = buffer;
+      bufferEnd = buffer + nLeftOver;
+    }
+    return 1;
+  }
+#endif  /* not defined XML_CONTEXT_BYTES */
+  else {
+    memcpy(XML_GetBuffer(parser, len), s, len);
+    return XML_ParseBuffer(parser, len, isFinal);
+  }
+}
+
+int XML_ParseBuffer(XML_Parser parser, int len, int isFinal)
+{
+  const char *start = bufferPtr;
+  positionPtr = start;
+  bufferEnd += len;
+  parseEndByteIndex += len;
+  errorCode = processor(parser, start, parseEndPtr = bufferEnd,
+			isFinal ? (const char **)0 : &bufferPtr);
+  if (errorCode == XML_ERROR_NONE) {
+    if (!isFinal)
+      XmlUpdatePosition(encoding, positionPtr, bufferPtr, &position);
+    return 1;
+  }
+  else {
+    eventEndPtr = eventPtr;
+    processor = errorProcessor;
+    return 0;
+  }
+}
+
+void *XML_GetBuffer(XML_Parser parser, int len)
+{
+  if (len > bufferLim - bufferEnd) {
+    /* FIXME avoid integer overflow */
+    int neededSize = len + (bufferEnd - bufferPtr);
+#ifdef XML_CONTEXT_BYTES
+    int keep = bufferPtr - buffer;
+
+    if (keep > XML_CONTEXT_BYTES)
+      keep = XML_CONTEXT_BYTES;
+    neededSize += keep;
+#endif  /* defined XML_CONTEXT_BYTES */
+    if (neededSize  <= bufferLim - buffer) {
+#ifdef XML_CONTEXT_BYTES
+      if (keep < bufferPtr - buffer) {
+	int offset = (bufferPtr - buffer) - keep;
+	memmove(buffer, &buffer[offset], bufferEnd - bufferPtr + keep);
+	bufferEnd -= offset;
+	bufferPtr -= offset;
+      }
+#else
+      memmove(buffer, bufferPtr, bufferEnd - bufferPtr);
+      bufferEnd = buffer + (bufferEnd - bufferPtr);
+      bufferPtr = buffer;
+#endif  /* not defined XML_CONTEXT_BYTES */
+    }
+    else {
+      char *newBuf;
+      int bufferSize = bufferLim - bufferPtr;
+      if (bufferSize == 0)
+	bufferSize = INIT_BUFFER_SIZE;
+      do {
+	bufferSize *= 2;
+      } while (bufferSize < neededSize);
+      newBuf = MALLOC(bufferSize);
+      if (newBuf == 0) {
+	errorCode = XML_ERROR_NO_MEMORY;
+	return 0;
+      }
+      bufferLim = newBuf + bufferSize;
+#ifdef XML_CONTEXT_BYTES
+      if (bufferPtr) {
+	int keep = bufferPtr - buffer;
+	if (keep > XML_CONTEXT_BYTES)
+	  keep = XML_CONTEXT_BYTES;
+	memcpy(newBuf, &bufferPtr[-keep], bufferEnd - bufferPtr + keep);
+	FREE(buffer);
+	buffer = newBuf;
+	bufferEnd = buffer + (bufferEnd - bufferPtr) + keep;
+	bufferPtr = buffer + keep;
+      }
+      else {
+	bufferEnd = newBuf + (bufferEnd - bufferPtr);
+	bufferPtr = buffer = newBuf;
+      }
+#else
+      if (bufferPtr) {
+	memcpy(newBuf, bufferPtr, bufferEnd - bufferPtr);
+	FREE(buffer);
+      }
+      bufferEnd = newBuf + (bufferEnd - bufferPtr);
+      bufferPtr = buffer = newBuf;
+#endif  /* not defined XML_CONTEXT_BYTES */
+    }
+  }
+  return bufferEnd;
+}
+
+enum XML_Error XML_GetErrorCode(XML_Parser parser)
+{
+  return errorCode;
+}
+
+long XML_GetCurrentByteIndex(XML_Parser parser)
+{
+  if (eventPtr)
+    return parseEndByteIndex - (parseEndPtr - eventPtr);
+  return -1;
+}
+
+int XML_GetCurrentByteCount(XML_Parser parser)
+{
+  if (eventEndPtr && eventPtr)
+    return eventEndPtr - eventPtr;
+  return 0;
+}
+
+const char * XML_GetInputContext(XML_Parser parser, int *offset, int *size)
+{
+#ifdef XML_CONTEXT_BYTES
+  if (eventPtr && buffer) {
+    *offset = eventPtr - buffer;
+    *size   = bufferEnd - buffer;
+    return buffer;
+  }
+#endif /* defined XML_CONTEXT_BYTES */
+  return (char *) 0;
+}
+
+int XML_GetCurrentLineNumber(XML_Parser parser)
+{
+  if (eventPtr) {
+    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
+    positionPtr = eventPtr;
+  }
+  return position.lineNumber + 1;
+}
+
+int XML_GetCurrentColumnNumber(XML_Parser parser)
+{
+  if (eventPtr) {
+    XmlUpdatePosition(encoding, positionPtr, eventPtr, &position);
+    positionPtr = eventPtr;
+  }
+  return position.columnNumber;
+}
+
+void XML_DefaultCurrent(XML_Parser parser)
+{
+  if (defaultHandler) {
+    if (openInternalEntities)
+      reportDefault(parser,
+	            internalEncoding,
+		    openInternalEntities->internalEventPtr,
+		    openInternalEntities->internalEventEndPtr);
+    else
+      reportDefault(parser, encoding, eventPtr, eventEndPtr);
+  }
+}
+
+const XML_LChar *XML_ErrorString(int code)
+{
+  static const XML_LChar *message[] = {
+    0,
+    XML_T("out of memory"),
+    XML_T("syntax error"),
+    XML_T("no element found"),
+    XML_T("not well-formed (invalid token)"),
+    XML_T("unclosed token"),
+    XML_T("unclosed token"),
+    XML_T("mismatched tag"),
+    XML_T("duplicate attribute"),
+    XML_T("junk after document element"),
+    XML_T("illegal parameter entity reference"),
+    XML_T("undefined entity"),
+    XML_T("recursive entity reference"),
+    XML_T("asynchronous entity"),
+    XML_T("reference to invalid character number"),
+    XML_T("reference to binary entity"),
+    XML_T("reference to external entity in attribute"),
+    XML_T("xml processing instruction not at start of external entity"),
+    XML_T("unknown encoding"),
+    XML_T("encoding specified in XML declaration is incorrect"),
+    XML_T("unclosed CDATA section"),
+    XML_T("error in processing external entity reference"),
+    XML_T("document is not standalone"),
+    XML_T("unexpected parser state - please send a bug report")
+  };
+  if (code > 0 && code < sizeof(message)/sizeof(message[0]))
+    return message[code];
+  return 0;
+}
+
+const XML_LChar *
+XML_ExpatVersion(void) {
+  return VERSION;
+}
+
+XML_Expat_Version
+XML_ExpatVersionInfo(void) {
+  XML_Expat_Version version;
+
+  version.major = XML_MAJOR_VERSION;
+  version.minor = XML_MINOR_VERSION;
+  version.micro = XML_MICRO_VERSION;
+
+  return version;
+}
+
+static
+enum XML_Error contentProcessor(XML_Parser parser,
+				const char *start,
+				const char *end,
+				const char **endPtr)
+{
+  return doContent(parser, 0, encoding, start, end, endPtr);
+}
+
+static
+enum XML_Error externalEntityInitProcessor(XML_Parser parser,
+					   const char *start,
+					   const char *end,
+					   const char **endPtr)
+{
+  enum XML_Error result = initializeEncoding(parser);
+  if (result != XML_ERROR_NONE)
+    return result;
+  processor = externalEntityInitProcessor2;
+  return externalEntityInitProcessor2(parser, start, end, endPtr);
+}
+
+static
+enum XML_Error externalEntityInitProcessor2(XML_Parser parser,
+					    const char *start,
+					    const char *end,
+					    const char **endPtr)
+{
+  const char *next;
+  int tok = XmlContentTok(encoding, start, end, &next);
+  switch (tok) {
+  case XML_TOK_BOM:
+    start = next;
+    break;
+  case XML_TOK_PARTIAL:
+    if (endPtr) {
+      *endPtr = start;
+      return XML_ERROR_NONE;
+    }
+    eventPtr = start;
+    return XML_ERROR_UNCLOSED_TOKEN;
+  case XML_TOK_PARTIAL_CHAR:
+    if (endPtr) {
+      *endPtr = start;
+      return XML_ERROR_NONE;
+    }
+    eventPtr = start;
+    return XML_ERROR_PARTIAL_CHAR;
+  }
+  processor = externalEntityInitProcessor3;
+  return externalEntityInitProcessor3(parser, start, end, endPtr);
+}
+
+static
+enum XML_Error externalEntityInitProcessor3(XML_Parser parser,
+					    const char *start,
+					    const char *end,
+					    const char **endPtr)
+{
+  const char *next;
+  int tok = XmlContentTok(encoding, start, end, &next);
+  switch (tok) {
+  case XML_TOK_XML_DECL:
+    {
+      enum XML_Error result = processXmlDecl(parser, 1, start, next);
+      if (result != XML_ERROR_NONE)
+	return result;
+      start = next;
+    }
+    break;
+  case XML_TOK_PARTIAL:
+    if (endPtr) {
+      *endPtr = start;
+      return XML_ERROR_NONE;
+    }
+    eventPtr = start;
+    return XML_ERROR_UNCLOSED_TOKEN;
+  case XML_TOK_PARTIAL_CHAR:
+    if (endPtr) {
+      *endPtr = start;
+      return XML_ERROR_NONE;
+    }
+    eventPtr = start;
+    return XML_ERROR_PARTIAL_CHAR;
+  }
+  processor = externalEntityContentProcessor;
+  tagLevel = 1;
+  return doContent(parser, 1, encoding, start, end, endPtr);
+}
+
+static
+enum XML_Error externalEntityContentProcessor(XML_Parser parser,
+					      const char *start,
+					      const char *end,
+					      const char **endPtr)
+{
+  return doContent(parser, 1, encoding, start, end, endPtr);
+}
+
+static enum XML_Error
+doContent(XML_Parser parser,
+	  int startTagLevel,
+	  const ENCODING *enc,
+	  const char *s,
+	  const char *end,
+	  const char **nextPtr)
+{
+  const char **eventPP;
+  const char **eventEndPP;
+  if (enc == encoding) {
+    eventPP = &eventPtr;
+    eventEndPP = &eventEndPtr;
+  }
+  else {
+    eventPP = &(openInternalEntities->internalEventPtr);
+    eventEndPP = &(openInternalEntities->internalEventEndPtr);
+  }
+  *eventPP = s;
+  for (;;) {
+    const char *next = s; /* XmlContentTok doesn't always set the last arg */
+    int tok = XmlContentTok(enc, s, end, &next);
+    *eventEndPP = next;
+    switch (tok) {
+    case XML_TOK_TRAILING_CR:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      *eventEndPP = end;
+      if (characterDataHandler) {
+	XML_Char c = 0xA;
+	characterDataHandler(handlerArg, &c, 1);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, end);
+      if (startTagLevel == 0)
+	return XML_ERROR_NO_ELEMENTS;
+      if (tagLevel != startTagLevel)
+	return XML_ERROR_ASYNC_ENTITY;
+      return XML_ERROR_NONE;
+    case XML_TOK_NONE:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      if (startTagLevel > 0) {
+	if (tagLevel != startTagLevel)
+	  return XML_ERROR_ASYNC_ENTITY;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_NO_ELEMENTS;
+    case XML_TOK_INVALID:
+      *eventPP = next;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_PARTIAL:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_UNCLOSED_TOKEN;
+    case XML_TOK_PARTIAL_CHAR:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_PARTIAL_CHAR;
+    case XML_TOK_ENTITY_REF:
+      {
+	const XML_Char *name;
+	ENTITY *entity;
+	XML_Char ch = XmlPredefinedEntityName(enc,
+					      s + enc->minBytesPerChar,
+					      next - enc->minBytesPerChar);
+	if (ch) {
+	  if (characterDataHandler)
+	    characterDataHandler(handlerArg, &ch, 1);
+	  else if (defaultHandler)
+	    reportDefault(parser, enc, s, next);
+	  break;
+	}
+	name = poolStoreString(&dtd.pool, enc,
+				s + enc->minBytesPerChar,
+				next - enc->minBytesPerChar);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	entity = (ENTITY *)lookup(&dtd.generalEntities, name, 0);
+	poolDiscard(&dtd.pool);
+	if (!entity) {
+	  if (dtd.complete || dtd.standalone)
+	    return XML_ERROR_UNDEFINED_ENTITY;
+	  if (defaultHandler)
+	    reportDefault(parser, enc, s, next);
+	  break;
+	}
+	if (entity->open)
+	  return XML_ERROR_RECURSIVE_ENTITY_REF;
+	if (entity->notation)
+	  return XML_ERROR_BINARY_ENTITY_REF;
+	if (entity) {
+	  if (entity->textPtr) {
+	    enum XML_Error result;
+	    OPEN_INTERNAL_ENTITY openEntity;
+	    if (defaultHandler && !defaultExpandInternalEntities) {
+	      reportDefault(parser, enc, s, next);
+	      break;
+	    }
+	    entity->open = 1;
+	    openEntity.next = openInternalEntities;
+	    openInternalEntities = &openEntity;
+	    openEntity.entity = entity;
+	    openEntity.internalEventPtr = 0;
+	    openEntity.internalEventEndPtr = 0;
+	    result = doContent(parser,
+			       tagLevel,
+			       internalEncoding,
+			       (char *)entity->textPtr,
+			       (char *)(entity->textPtr + entity->textLen),
+			       0);
+	    entity->open = 0;
+	    openInternalEntities = openEntity.next;
+	    if (result)
+	      return result;
+	  }
+	  else if (externalEntityRefHandler) {
+	    const XML_Char *context;
+	    entity->open = 1;
+	    context = getContext(parser);
+	    entity->open = 0;
+	    if (!context)
+	      return XML_ERROR_NO_MEMORY;
+	    if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+				          context,
+					  entity->base,
+					  entity->systemId,
+					  entity->publicId))
+	      return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+	    poolDiscard(&tempPool);
+	  }
+	  else if (defaultHandler)
+	    reportDefault(parser, enc, s, next);
+	}
+	break;
+      }
+    case XML_TOK_START_TAG_WITH_ATTS:
+      if (!startElementHandler) {
+	enum XML_Error result = storeAtts(parser, enc, s, 0, 0);
+	if (result)
+	  return result;
+      }
+      /* fall through */
+    case XML_TOK_START_TAG_NO_ATTS:
+      {
+	TAG *tag;
+	if (freeTagList) {
+	  tag = freeTagList;
+	  freeTagList = freeTagList->parent;
+	}
+	else {
+	  tag = MALLOC(sizeof(TAG));
+	  if (!tag)
+	    return XML_ERROR_NO_MEMORY;
+	  tag->buf = MALLOC(INIT_TAG_BUF_SIZE);
+	  if (!tag->buf)
+	    return XML_ERROR_NO_MEMORY;
+	  tag->bufEnd = tag->buf + INIT_TAG_BUF_SIZE;
+	}
+	tag->bindings = 0;
+	tag->parent = tagStack;
+	tagStack = tag;
+	tag->name.localPart = 0;
+	tag->rawName = s + enc->minBytesPerChar;
+	tag->rawNameLength = XmlNameLength(enc, tag->rawName);
+	if (nextPtr) {
+	  /* Need to guarantee that:
+	     tag->buf + ROUND_UP(tag->rawNameLength, sizeof(XML_Char)) <= tag->bufEnd - sizeof(XML_Char) */
+	  if (tag->rawNameLength + (int)(sizeof(XML_Char) - 1) + (int)sizeof(XML_Char) > tag->bufEnd - tag->buf) {
+	    int bufSize = tag->rawNameLength * 4;
+	    bufSize = ROUND_UP(bufSize, sizeof(XML_Char));
+	    tag->buf = REALLOC(tag->buf, bufSize);
+	    if (!tag->buf)
+	      return XML_ERROR_NO_MEMORY;
+	    tag->bufEnd = tag->buf + bufSize;
+	  }
+	  memcpy(tag->buf, tag->rawName, tag->rawNameLength);
+	  tag->rawName = tag->buf;
+	}
+	++tagLevel;
+	if (startElementHandler) {
+	  enum XML_Error result;
+	  XML_Char *toPtr;
+	  for (;;) {
+	    const char *rawNameEnd = tag->rawName + tag->rawNameLength;
+	    const char *fromPtr = tag->rawName;
+	    int bufSize;
+	    if (nextPtr)
+	      toPtr = (XML_Char *)(tag->buf + ROUND_UP(tag->rawNameLength, sizeof(XML_Char)));
+	    else
+	      toPtr = (XML_Char *)tag->buf;
+	    tag->name.str = toPtr;
+	    XmlConvert(enc,
+		       &fromPtr, rawNameEnd,
+		       (ICHAR **)&toPtr, (ICHAR *)tag->bufEnd - 1);
+	    if (fromPtr == rawNameEnd)
+	      break;
+	    bufSize = (tag->bufEnd - tag->buf) << 1;
+	    tag->buf = REALLOC(tag->buf, bufSize);
+	    if (!tag->buf)
+	      return XML_ERROR_NO_MEMORY;
+	    tag->bufEnd = tag->buf + bufSize;
+	    if (nextPtr)
+	      tag->rawName = tag->buf;
+	  }
+	  *toPtr = XML_T('\0');
+	  result = storeAtts(parser, enc, s, &(tag->name), &(tag->bindings));
+	  if (result)
+	    return result;
+	  startElementHandler(handlerArg, tag->name.str, (const XML_Char **)atts);
+	  poolClear(&tempPool);
+	}
+	else {
+	  tag->name.str = 0;
+	  if (defaultHandler)
+	    reportDefault(parser, enc, s, next);
+	}
+	break;
+      }
+    case XML_TOK_EMPTY_ELEMENT_WITH_ATTS:
+      if (!startElementHandler) {
+	enum XML_Error result = storeAtts(parser, enc, s, 0, 0);
+	if (result)
+	  return result;
+      }
+      /* fall through */
+    case XML_TOK_EMPTY_ELEMENT_NO_ATTS:
+      if (startElementHandler || endElementHandler) {
+	const char *rawName = s + enc->minBytesPerChar;
+	enum XML_Error result;
+	BINDING *bindings = 0;
+	TAG_NAME name;
+	name.str = poolStoreString(&tempPool, enc, rawName,
+				   rawName + XmlNameLength(enc, rawName));
+	if (!name.str)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&tempPool);
+	result = storeAtts(parser, enc, s, &name, &bindings);
+	if (result)
+	  return result;
+	poolFinish(&tempPool);
+	if (startElementHandler)
+	  startElementHandler(handlerArg, name.str, (const XML_Char **)atts);
+	if (endElementHandler) {
+	  if (startElementHandler)
+	    *eventPP = *eventEndPP;
+	  endElementHandler(handlerArg, name.str);
+	}
+	poolClear(&tempPool);
+	while (bindings) {
+	  BINDING *b = bindings;
+	  if (endNamespaceDeclHandler)
+	    endNamespaceDeclHandler(handlerArg, b->prefix->name);
+	  bindings = bindings->nextTagBinding;
+	  b->nextTagBinding = freeBindingList;
+	  freeBindingList = b;
+	  b->prefix->binding = b->prevPrefixBinding;
+	}
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      if (tagLevel == 0)
+	return epilogProcessor(parser, next, end, nextPtr);
+      break;
+    case XML_TOK_END_TAG:
+      if (tagLevel == startTagLevel)
+        return XML_ERROR_ASYNC_ENTITY;
+      else {
+	int len;
+	const char *rawName;
+	TAG *tag = tagStack;
+	tagStack = tag->parent;
+	tag->parent = freeTagList;
+	freeTagList = tag;
+	rawName = s + enc->minBytesPerChar*2;
+	len = XmlNameLength(enc, rawName);
+	if (len != tag->rawNameLength
+	    || memcmp(tag->rawName, rawName, len) != 0) {
+	  *eventPP = rawName;
+	  return XML_ERROR_TAG_MISMATCH;
+	}
+	--tagLevel;
+	if (endElementHandler && tag->name.str) {
+	  if (tag->name.localPart) {
+	    XML_Char *to = (XML_Char *)tag->name.str + tag->name.uriLen;
+	    const XML_Char *from = tag->name.localPart;
+	    while ((*to++ = *from++) != 0)
+	      ;
+	  }
+	  endElementHandler(handlerArg, tag->name.str);
+	}
+	else if (defaultHandler)
+	  reportDefault(parser, enc, s, next);
+	while (tag->bindings) {
+	  BINDING *b = tag->bindings;
+	  if (endNamespaceDeclHandler)
+	    endNamespaceDeclHandler(handlerArg, b->prefix->name);
+	  tag->bindings = tag->bindings->nextTagBinding;
+	  b->nextTagBinding = freeBindingList;
+	  freeBindingList = b;
+	  b->prefix->binding = b->prevPrefixBinding;
+	}
+	if (tagLevel == 0)
+	  return epilogProcessor(parser, next, end, nextPtr);
+      }
+      break;
+    case XML_TOK_CHAR_REF:
+      {
+	int n = XmlCharRefNumber(enc, s);
+	if (n < 0)
+	  return XML_ERROR_BAD_CHAR_REF;
+	if (characterDataHandler) {
+	  XML_Char buf[XML_ENCODE_MAX];
+	  characterDataHandler(handlerArg, buf, XmlEncode(n, (ICHAR *)buf));
+	}
+	else if (defaultHandler)
+	  reportDefault(parser, enc, s, next);
+      }
+      break;
+    case XML_TOK_XML_DECL:
+      return XML_ERROR_MISPLACED_XML_PI;
+    case XML_TOK_DATA_NEWLINE:
+      if (characterDataHandler) {
+	XML_Char c = 0xA;
+	characterDataHandler(handlerArg, &c, 1);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+    case XML_TOK_CDATA_SECT_OPEN:
+      {
+	enum XML_Error result;
+	if (startCdataSectionHandler)
+  	  startCdataSectionHandler(handlerArg);
+#if 0
+	/* Suppose you doing a transformation on a document that involves
+	   changing only the character data.  You set up a defaultHandler
+	   and a characterDataHandler.  The defaultHandler simply copies
+	   characters through.  The characterDataHandler does the transformation
+	   and writes the characters out escaping them as necessary.  This case
+	   will fail to work if we leave out the following two lines (because &
+	   and < inside CDATA sections will be incorrectly escaped).
+
+	   However, now we have a start/endCdataSectionHandler, so it seems
+	   easier to let the user deal with this. */
+
+	else if (characterDataHandler)
+  	  characterDataHandler(handlerArg, dataBuf, 0);
+#endif
+	else if (defaultHandler)
+	  reportDefault(parser, enc, s, next);
+	result = doCdataSection(parser, enc, &next, end, nextPtr);
+	if (!next) {
+	  processor = cdataSectionProcessor;
+	  return result;
+	}
+      }
+      break;
+    case XML_TOK_TRAILING_RSQB:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      if (characterDataHandler) {
+	if (MUST_CONVERT(enc, s)) {
+	  ICHAR *dataPtr = (ICHAR *)dataBuf;
+	  XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
+	  characterDataHandler(handlerArg, dataBuf, dataPtr - (ICHAR *)dataBuf);
+	}
+	else
+	  characterDataHandler(handlerArg,
+		  	       (XML_Char *)s,
+			       (XML_Char *)end - (XML_Char *)s);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, end);
+      if (startTagLevel == 0) {
+        *eventPP = end;
+	return XML_ERROR_NO_ELEMENTS;
+      }
+      if (tagLevel != startTagLevel) {
+	*eventPP = end;
+	return XML_ERROR_ASYNC_ENTITY;
+      }
+      return XML_ERROR_NONE;
+    case XML_TOK_DATA_CHARS:
+      if (characterDataHandler) {
+	if (MUST_CONVERT(enc, s)) {
+	  for (;;) {
+	    ICHAR *dataPtr = (ICHAR *)dataBuf;
+	    XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+	    *eventEndPP = s;
+	    characterDataHandler(handlerArg, dataBuf, dataPtr - (ICHAR *)dataBuf);
+	    if (s == next)
+	      break;
+	    *eventPP = s;
+	  }
+	}
+	else
+	  characterDataHandler(handlerArg,
+			       (XML_Char *)s,
+			       (XML_Char *)next - (XML_Char *)s);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+    case XML_TOK_PI:
+      if (!reportProcessingInstruction(parser, enc, s, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_COMMENT:
+      if (!reportComment(parser, enc, s, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    default:
+      if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+    }
+    *eventPP = s = next;
+  }
+  /* not reached */
+}
+
+/* If tagNamePtr is non-null, build a real list of attributes,
+otherwise just check the attributes for well-formedness. */
+
+static enum XML_Error storeAtts(XML_Parser parser, const ENCODING *enc,
+				const char *attStr, TAG_NAME *tagNamePtr,
+				BINDING **bindingsPtr)
+{
+  ELEMENT_TYPE *elementType = 0;
+  int nDefaultAtts = 0;
+  const XML_Char **appAtts;   /* the attribute list to pass to the application */
+  int attIndex = 0;
+  int i;
+  int n;
+  int nPrefixes = 0;
+  BINDING *binding;
+  const XML_Char *localPart;
+
+  /* lookup the element type name */
+  if (tagNamePtr) {
+    elementType = (ELEMENT_TYPE *)lookup(&dtd.elementTypes, tagNamePtr->str,0);
+    if (!elementType) {
+      tagNamePtr->str = poolCopyString(&dtd.pool, tagNamePtr->str);
+      if (!tagNamePtr->str)
+	return XML_ERROR_NO_MEMORY;
+      elementType = (ELEMENT_TYPE *)lookup(&dtd.elementTypes, tagNamePtr->str, sizeof(ELEMENT_TYPE));
+      if (!elementType)
+        return XML_ERROR_NO_MEMORY;
+      if (ns && !setElementTypePrefix(parser, elementType))
+        return XML_ERROR_NO_MEMORY;
+    }
+    nDefaultAtts = elementType->nDefaultAtts;
+  }
+  /* get the attributes from the tokenizer */
+  n = XmlGetAttributes(enc, attStr, attsSize, atts);
+  if (n + nDefaultAtts > attsSize) {
+    int oldAttsSize = attsSize;
+    attsSize = n + nDefaultAtts + INIT_ATTS_SIZE;
+    atts = REALLOC((void *)atts, attsSize * sizeof(ATTRIBUTE));
+    if (!atts)
+      return XML_ERROR_NO_MEMORY;
+    if (n > oldAttsSize)
+      XmlGetAttributes(enc, attStr, n, atts);
+  }
+  appAtts = (const XML_Char **)atts;
+  for (i = 0; i < n; i++) {
+    /* add the name and value to the attribute list */
+    ATTRIBUTE_ID *attId = getAttributeId(parser, enc, atts[i].name,
+					 atts[i].name
+					 + XmlNameLength(enc, atts[i].name));
+    if (!attId)
+      return XML_ERROR_NO_MEMORY;
+    /* detect duplicate attributes */
+    if ((attId->name)[-1]) {
+      if (enc == encoding)
+	eventPtr = atts[i].name;
+      return XML_ERROR_DUPLICATE_ATTRIBUTE;
+    }
+    (attId->name)[-1] = 1;
+    appAtts[attIndex++] = attId->name;
+    if (!atts[i].normalized) {
+      enum XML_Error result;
+      int isCdata = 1;
+
+      /* figure out whether declared as other than CDATA */
+      if (attId->maybeTokenized) {
+	int j;
+	for (j = 0; j < nDefaultAtts; j++) {
+	  if (attId == elementType->defaultAtts[j].id) {
+	    isCdata = elementType->defaultAtts[j].isCdata;
+	    break;
+	  }
+	}
+      }
+
+      /* normalize the attribute value */
+      result = storeAttributeValue(parser, enc, isCdata,
+				   atts[i].valuePtr, atts[i].valueEnd,
+			           &tempPool);
+      if (result)
+	return result;
+      if (tagNamePtr) {
+	appAtts[attIndex] = poolStart(&tempPool);
+	poolFinish(&tempPool);
+      }
+      else
+	poolDiscard(&tempPool);
+    }
+    else if (tagNamePtr) {
+      /* the value did not need normalizing */
+      appAtts[attIndex] = poolStoreString(&tempPool, enc, atts[i].valuePtr, atts[i].valueEnd);
+      if (appAtts[attIndex] == 0)
+	return XML_ERROR_NO_MEMORY;
+      poolFinish(&tempPool);
+    }
+    /* handle prefixed attribute names */
+    if (attId->prefix && tagNamePtr) {
+      if (attId->xmlns) {
+	/* deal with namespace declarations here */
+        if (!addBinding(parser, attId->prefix, attId, appAtts[attIndex], bindingsPtr))
+          return XML_ERROR_NO_MEMORY;
+        --attIndex;
+      }
+      else {
+	/* deal with other prefixed names later */
+        attIndex++;
+        nPrefixes++;
+        (attId->name)[-1] = 2;
+      }
+    }
+    else
+      attIndex++;
+  }
+  if (tagNamePtr) {
+    int j;
+    nSpecifiedAtts = attIndex;
+    if (elementType->idAtt && (elementType->idAtt->name)[-1]) {
+      for (i = 0; i < attIndex; i += 2)
+	if (appAtts[i] == elementType->idAtt->name) {
+	  idAttIndex = i;
+	  break;
+	}
+    }
+    else
+      idAttIndex = -1;
+    /* do attribute defaulting */
+    for (j = 0; j < nDefaultAtts; j++) {
+      const DEFAULT_ATTRIBUTE *da = elementType->defaultAtts + j;
+      if (!(da->id->name)[-1] && da->value) {
+        if (da->id->prefix) {
+          if (da->id->xmlns) {
+	    if (!addBinding(parser, da->id->prefix, da->id, da->value, bindingsPtr))
+	      return XML_ERROR_NO_MEMORY;
+	  }
+          else {
+	    (da->id->name)[-1] = 2;
+	    nPrefixes++;
+  	    appAtts[attIndex++] = da->id->name;
+	    appAtts[attIndex++] = da->value;
+	  }
+	}
+	else {
+	  (da->id->name)[-1] = 1;
+	  appAtts[attIndex++] = da->id->name;
+	  appAtts[attIndex++] = da->value;
+	}
+      }
+    }
+    appAtts[attIndex] = 0;
+  }
+  i = 0;
+  if (nPrefixes) {
+    /* expand prefixed attribute names */
+    for (; i < attIndex; i += 2) {
+      if (appAtts[i][-1] == 2) {
+        ATTRIBUTE_ID *id;
+        ((XML_Char *)(appAtts[i]))[-1] = 0;
+	id = (ATTRIBUTE_ID *)lookup(&dtd.attributeIds, appAtts[i], 0);
+	if (id->prefix->binding) {
+	  int j;
+	  const BINDING *b = id->prefix->binding;
+	  const XML_Char *s = appAtts[i];
+	  for (j = 0; j < b->uriLen; j++) {
+	    if (!poolAppendChar(&tempPool, b->uri[j]))
+	      return XML_ERROR_NO_MEMORY;
+	  }
+	  while (*s++ != ':')
+	    ;
+	  do {
+	    if (!poolAppendChar(&tempPool, *s))
+	      return XML_ERROR_NO_MEMORY;
+	  } while (*s++);
+	  if (ns_triplets) {
+	    tempPool.ptr[-1] = namespaceSeparator;
+	    s = b->prefix->name;
+	    do {
+	      if (!poolAppendChar(&tempPool, *s))
+		return XML_ERROR_NO_MEMORY;
+	    } while (*s++);
+	  }
+
+	  appAtts[i] = poolStart(&tempPool);
+	  poolFinish(&tempPool);
+	}
+	if (!--nPrefixes)
+	  break;
+      }
+      else
+	((XML_Char *)(appAtts[i]))[-1] = 0;
+    }
+  }
+  /* clear the flags that say whether attributes were specified */
+  for (; i < attIndex; i += 2)
+    ((XML_Char *)(appAtts[i]))[-1] = 0;
+  if (!tagNamePtr)
+    return XML_ERROR_NONE;
+  for (binding = *bindingsPtr; binding; binding = binding->nextTagBinding)
+    binding->attId->name[-1] = 0;
+  /* expand the element type name */
+  if (elementType->prefix) {
+    binding = elementType->prefix->binding;
+    if (!binding)
+      return XML_ERROR_NONE;
+    localPart = tagNamePtr->str;
+    while (*localPart++ != XML_T(':'))
+      ;
+  }
+  else if (dtd.defaultPrefix.binding) {
+    binding = dtd.defaultPrefix.binding;
+    localPart = tagNamePtr->str;
+  }
+  else
+    return XML_ERROR_NONE;
+  tagNamePtr->localPart = localPart;
+  tagNamePtr->uriLen = binding->uriLen;
+  for (i = 0; localPart[i++];)
+    ;
+  n = i + binding->uriLen;
+  if (n > binding->uriAlloc) {
+    TAG *p;
+    XML_Char *uri = MALLOC((n + EXPAND_SPARE) * sizeof(XML_Char));
+    if (!uri)
+      return XML_ERROR_NO_MEMORY;
+    binding->uriAlloc = n + EXPAND_SPARE;
+    memcpy(uri, binding->uri, binding->uriLen * sizeof(XML_Char));
+    for (p = tagStack; p; p = p->parent)
+      if (p->name.str == binding->uri)
+	p->name.str = uri;
+    FREE(binding->uri);
+    binding->uri = uri;
+  }
+  memcpy(binding->uri + binding->uriLen, localPart, i * sizeof(XML_Char));
+  tagNamePtr->str = binding->uri;
+  return XML_ERROR_NONE;
+}
+
+static
+int addBinding(XML_Parser parser, PREFIX *prefix, const ATTRIBUTE_ID *attId, const XML_Char *uri, BINDING **bindingsPtr)
+{
+  BINDING *b;
+  int len;
+  for (len = 0; uri[len]; len++)
+    ;
+  if (namespaceSeparator)
+    len++;
+  if (freeBindingList) {
+    b = freeBindingList;
+    if (len > b->uriAlloc) {
+      b->uri = REALLOC(b->uri, sizeof(XML_Char) * (len + EXPAND_SPARE));
+      if (!b->uri)
+	return 0;
+      b->uriAlloc = len + EXPAND_SPARE;
+    }
+    freeBindingList = b->nextTagBinding;
+  }
+  else {
+    b = MALLOC(sizeof(BINDING));
+    if (!b)
+      return 0;
+    b->uri = MALLOC(sizeof(XML_Char) * (len + EXPAND_SPARE));
+    if (!b->uri) {
+      FREE(b);
+      return 0;
+    }
+    b->uriAlloc = len + EXPAND_SPARE;
+  }
+  b->uriLen = len;
+  memcpy(b->uri, uri, len * sizeof(XML_Char));
+  if (namespaceSeparator)
+    b->uri[len - 1] = namespaceSeparator;
+  b->prefix = prefix;
+  b->attId = attId;
+  b->prevPrefixBinding = prefix->binding;
+  if (*uri == XML_T('\0') && prefix == &dtd.defaultPrefix)
+    prefix->binding = 0;
+  else
+    prefix->binding = b;
+  b->nextTagBinding = *bindingsPtr;
+  *bindingsPtr = b;
+  if (startNamespaceDeclHandler)
+    startNamespaceDeclHandler(handlerArg, prefix->name,
+			      prefix->binding ? uri : 0);
+  return 1;
+}
+
+/* The idea here is to avoid using stack for each CDATA section when
+the whole file is parsed with one call. */
+
+static
+enum XML_Error cdataSectionProcessor(XML_Parser parser,
+				     const char *start,
+			    	     const char *end,
+				     const char **endPtr)
+{
+  enum XML_Error result = doCdataSection(parser, encoding, &start, end, endPtr);
+  if (start) {
+    processor = contentProcessor;
+    return contentProcessor(parser, start, end, endPtr);
+  }
+  return result;
+}
+
+/* startPtr gets set to non-null is the section is closed, and to null if
+the section is not yet closed. */
+
+static
+enum XML_Error doCdataSection(XML_Parser parser,
+			      const ENCODING *enc,
+			      const char **startPtr,
+			      const char *end,
+			      const char **nextPtr)
+{
+  const char *s = *startPtr;
+  const char **eventPP;
+  const char **eventEndPP;
+  if (enc == encoding) {
+    eventPP = &eventPtr;
+    *eventPP = s;
+    eventEndPP = &eventEndPtr;
+  }
+  else {
+    eventPP = &(openInternalEntities->internalEventPtr);
+    eventEndPP = &(openInternalEntities->internalEventEndPtr);
+  }
+  *eventPP = s;
+  *startPtr = 0;
+  for (;;) {
+    const char *next;
+    int tok = XmlCdataSectionTok(enc, s, end, &next);
+    *eventEndPP = next;
+    switch (tok) {
+    case XML_TOK_CDATA_SECT_CLOSE:
+      if (endCdataSectionHandler)
+	endCdataSectionHandler(handlerArg);
+#if 0
+      /* see comment under XML_TOK_CDATA_SECT_OPEN */
+      else if (characterDataHandler)
+	characterDataHandler(handlerArg, dataBuf, 0);
+#endif
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      *startPtr = next;
+      return XML_ERROR_NONE;
+    case XML_TOK_DATA_NEWLINE:
+      if (characterDataHandler) {
+	XML_Char c = 0xA;
+	characterDataHandler(handlerArg, &c, 1);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+    case XML_TOK_DATA_CHARS:
+      if (characterDataHandler) {
+	if (MUST_CONVERT(enc, s)) {
+	  for (;;) {
+  	    ICHAR *dataPtr = (ICHAR *)dataBuf;
+	    XmlConvert(enc, &s, next, &dataPtr, (ICHAR *)dataBufEnd);
+	    *eventEndPP = next;
+	    characterDataHandler(handlerArg, dataBuf, dataPtr - (ICHAR *)dataBuf);
+	    if (s == next)
+	      break;
+	    *eventPP = s;
+	  }
+	}
+	else
+	  characterDataHandler(handlerArg,
+		  	       (XML_Char *)s,
+			       (XML_Char *)next - (XML_Char *)s);
+      }
+      else if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+    case XML_TOK_INVALID:
+      *eventPP = next;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_PARTIAL_CHAR:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_PARTIAL_CHAR;
+    case XML_TOK_PARTIAL:
+    case XML_TOK_NONE:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_UNCLOSED_CDATA_SECTION;
+    default:
+      *eventPP = next;
+      return XML_ERROR_UNEXPECTED_STATE;
+    }
+    *eventPP = s = next;
+  }
+  /* not reached */
+}
+
+#ifdef XML_DTD
+
+/* The idea here is to avoid using stack for each IGNORE section when
+the whole file is parsed with one call. */
+
+static
+enum XML_Error ignoreSectionProcessor(XML_Parser parser,
+				      const char *start,
+				      const char *end,
+				      const char **endPtr)
+{
+  enum XML_Error result = doIgnoreSection(parser, encoding, &start, end, endPtr);
+  if (start) {
+    processor = prologProcessor;
+    return prologProcessor(parser, start, end, endPtr);
+  }
+  return result;
+}
+
+/* startPtr gets set to non-null is the section is closed, and to null if
+the section is not yet closed. */
+
+static
+enum XML_Error doIgnoreSection(XML_Parser parser,
+			       const ENCODING *enc,
+			       const char **startPtr,
+			       const char *end,
+			       const char **nextPtr)
+{
+  const char *next;
+  int tok;
+  const char *s = *startPtr;
+  const char **eventPP;
+  const char **eventEndPP;
+  if (enc == encoding) {
+    eventPP = &eventPtr;
+    *eventPP = s;
+    eventEndPP = &eventEndPtr;
+  }
+  else {
+    eventPP = &(openInternalEntities->internalEventPtr);
+    eventEndPP = &(openInternalEntities->internalEventEndPtr);
+  }
+  *eventPP = s;
+  *startPtr = 0;
+  tok = XmlIgnoreSectionTok(enc, s, end, &next);
+  *eventEndPP = next;
+  switch (tok) {
+  case XML_TOK_IGNORE_SECT:
+    if (defaultHandler)
+      reportDefault(parser, enc, s, next);
+    *startPtr = next;
+    return XML_ERROR_NONE;
+  case XML_TOK_INVALID:
+    *eventPP = next;
+    return XML_ERROR_INVALID_TOKEN;
+  case XML_TOK_PARTIAL_CHAR:
+    if (nextPtr) {
+      *nextPtr = s;
+      return XML_ERROR_NONE;
+    }
+    return XML_ERROR_PARTIAL_CHAR;
+  case XML_TOK_PARTIAL:
+  case XML_TOK_NONE:
+    if (nextPtr) {
+      *nextPtr = s;
+      return XML_ERROR_NONE;
+    }
+    return XML_ERROR_SYNTAX; /* XML_ERROR_UNCLOSED_IGNORE_SECTION */
+  default:
+    *eventPP = next;
+    return XML_ERROR_UNEXPECTED_STATE;
+  }
+  /* not reached */
+}
+
+#endif /* XML_DTD */
+
+static enum XML_Error
+initializeEncoding(XML_Parser parser)
+{
+  const char *s;
+#ifdef XML_UNICODE
+  char encodingBuf[128];
+  if (!protocolEncodingName)
+    s = 0;
+  else {
+    int i;
+    for (i = 0; protocolEncodingName[i]; i++) {
+      if (i == sizeof(encodingBuf) - 1
+	  || (protocolEncodingName[i] & ~0x7f) != 0) {
+	encodingBuf[0] = '\0';
+	break;
+      }
+      encodingBuf[i] = (char)protocolEncodingName[i];
+    }
+    encodingBuf[i] = '\0';
+    s = encodingBuf;
+  }
+#else
+  s = protocolEncodingName;
+#endif
+  if ((ns ? XmlInitEncodingNS : XmlInitEncoding)(&initEncoding, &encoding, s))
+    return XML_ERROR_NONE;
+  return handleUnknownEncoding(parser, protocolEncodingName);
+}
+
+static enum XML_Error
+processXmlDecl(XML_Parser parser, int isGeneralTextEntity,
+	       const char *s, const char *next)
+{
+  const char *encodingName = 0;
+  const char *storedEncName = 0;
+  const ENCODING *newEncoding = 0;
+  const char *version = 0;
+  const char *versionend;
+  const char *storedversion = 0;
+  int standalone = -1;
+  if (!(ns
+        ? XmlParseXmlDeclNS
+	: XmlParseXmlDecl)(isGeneralTextEntity,
+		           encoding,
+		           s,
+		           next,
+		           &eventPtr,
+		           &version,
+			   &versionend,
+		           &encodingName,
+		           &newEncoding,
+		           &standalone))
+    return XML_ERROR_SYNTAX;
+  if (!isGeneralTextEntity && standalone == 1) {
+    dtd.standalone = 1;
+#ifdef XML_DTD
+    if (paramEntityParsing == XML_PARAM_ENTITY_PARSING_UNLESS_STANDALONE)
+      paramEntityParsing = XML_PARAM_ENTITY_PARSING_NEVER;
+#endif /* XML_DTD */
+  }
+  if (xmlDeclHandler) {
+    if (encodingName) {
+      storedEncName = poolStoreString(&temp2Pool,
+				      encoding,
+				      encodingName,
+				      encodingName
+				      + XmlNameLength(encoding, encodingName));
+      if (! storedEncName)
+	return XML_ERROR_NO_MEMORY;
+      poolFinish(&temp2Pool);
+    }
+    if (version) {
+      storedversion = poolStoreString(&temp2Pool,
+				      encoding,
+				      version,
+				      versionend - encoding->minBytesPerChar);
+      if (! storedversion)
+	return XML_ERROR_NO_MEMORY;
+    }
+    xmlDeclHandler(handlerArg, storedversion, storedEncName, standalone);
+  }
+  else if (defaultHandler)
+    reportDefault(parser, encoding, s, next);
+  if (!protocolEncodingName) {
+    if (newEncoding) {
+      if (newEncoding->minBytesPerChar != encoding->minBytesPerChar) {
+	eventPtr = encodingName;
+	return XML_ERROR_INCORRECT_ENCODING;
+      }
+      encoding = newEncoding;
+    }
+    else if (encodingName) {
+      enum XML_Error result;
+      if (! storedEncName) {
+	storedEncName = poolStoreString(&temp2Pool,
+					encoding,
+					encodingName,
+					encodingName
+					+ XmlNameLength(encoding, encodingName));
+	if (! storedEncName)
+	  return XML_ERROR_NO_MEMORY;
+      }
+      result = handleUnknownEncoding(parser, storedEncName);
+      poolClear(&tempPool);
+      if (result == XML_ERROR_UNKNOWN_ENCODING)
+	eventPtr = encodingName;
+      return result;
+    }
+  }
+
+  if (storedEncName || storedversion)
+    poolClear(&temp2Pool);
+
+  return XML_ERROR_NONE;
+}
+
+static enum XML_Error
+handleUnknownEncoding(XML_Parser parser, const XML_Char *encodingName)
+{
+  if (unknownEncodingHandler) {
+    XML_Encoding info;
+    int i;
+    for (i = 0; i < 256; i++)
+      info.map[i] = -1;
+    info.convert = 0;
+    info.data = 0;
+    info.release = 0;
+    if (unknownEncodingHandler(unknownEncodingHandlerData, encodingName, &info)) {
+      ENCODING *enc;
+      unknownEncodingMem = MALLOC(XmlSizeOfUnknownEncoding());
+      if (!unknownEncodingMem) {
+	if (info.release)
+	  info.release(info.data);
+	return XML_ERROR_NO_MEMORY;
+      }
+      enc = (ns
+	     ? XmlInitUnknownEncodingNS
+	     : XmlInitUnknownEncoding)(unknownEncodingMem,
+				       info.map,
+				       info.convert,
+				       info.data);
+      if (enc) {
+	unknownEncodingData = info.data;
+	unknownEncodingRelease = info.release;
+	encoding = enc;
+	return XML_ERROR_NONE;
+      }
+    }
+    if (info.release)
+      info.release(info.data);
+  }
+  return XML_ERROR_UNKNOWN_ENCODING;
+}
+
+static enum XML_Error
+prologInitProcessor(XML_Parser parser,
+		    const char *s,
+		    const char *end,
+		    const char **nextPtr)
+{
+  enum XML_Error result = initializeEncoding(parser);
+  if (result != XML_ERROR_NONE)
+    return result;
+  processor = prologProcessor;
+  return prologProcessor(parser, s, end, nextPtr);
+}
+
+static enum XML_Error
+prologProcessor(XML_Parser parser,
+		const char *s,
+		const char *end,
+		const char **nextPtr)
+{
+  const char *next;
+  int tok = XmlPrologTok(encoding, s, end, &next);
+  return doProlog(parser, encoding, s, end, tok, next, nextPtr);
+}
+
+static enum XML_Error
+doProlog(XML_Parser parser,
+	 const ENCODING *enc,
+	 const char *s,
+	 const char *end,
+	 int tok,
+	 const char *next,
+	 const char **nextPtr)
+{
+#ifdef XML_DTD
+  static const XML_Char externalSubsetName[] = { '#' , '\0' };
+#endif /* XML_DTD */
+
+  const char **eventPP;
+  const char **eventEndPP;
+  enum XML_Content_Quant quant;
+
+  if (enc == encoding) {
+    eventPP = &eventPtr;
+    eventEndPP = &eventEndPtr;
+  }
+  else {
+    eventPP = &(openInternalEntities->internalEventPtr);
+    eventEndPP = &(openInternalEntities->internalEventEndPtr);
+  }
+  for (;;) {
+    int role;
+    *eventPP = s;
+    *eventEndPP = next;
+    if (tok <= 0) {
+      if (nextPtr != 0 && tok != XML_TOK_INVALID) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      switch (tok) {
+      case XML_TOK_INVALID:
+	*eventPP = next;
+	return XML_ERROR_INVALID_TOKEN;
+      case XML_TOK_PARTIAL:
+	return XML_ERROR_UNCLOSED_TOKEN;
+      case XML_TOK_PARTIAL_CHAR:
+	return XML_ERROR_PARTIAL_CHAR;
+      case XML_TOK_NONE:
+#ifdef XML_DTD
+	if (enc != encoding)
+	  return XML_ERROR_NONE;
+	if (parentParser) {
+	  if (XmlTokenRole(&prologState, XML_TOK_NONE, end, end, enc)
+	      == XML_ROLE_ERROR)
+	    return XML_ERROR_SYNTAX;
+	  hadExternalDoctype = 0;
+	  return XML_ERROR_NONE;
+	}
+#endif /* XML_DTD */
+	return XML_ERROR_NO_ELEMENTS;
+      default:
+	tok = -tok;
+	next = end;
+	break;
+      }
+    }
+    role = XmlTokenRole(&prologState, tok, s, next, enc);
+    switch (role) {
+    case XML_ROLE_XML_DECL:
+      {
+	enum XML_Error result = processXmlDecl(parser, 0, s, next);
+	if (result != XML_ERROR_NONE)
+	  return result;
+	enc = encoding;
+      }
+      break;
+    case XML_ROLE_DOCTYPE_NAME:
+      if (startDoctypeDeclHandler) {
+	doctypeName = poolStoreString(&tempPool, enc, s, next);
+	if (! doctypeName)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&tempPool);
+	doctypeSysid = 0;
+	doctypePubid = 0;
+      }
+      break;
+    case XML_ROLE_DOCTYPE_INTERNAL_SUBSET:
+      if (startDoctypeDeclHandler) {
+	startDoctypeDeclHandler(handlerArg, doctypeName, doctypeSysid,
+				doctypePubid, 1);
+	doctypeName = 0;
+	poolClear(&tempPool);
+      }
+      break;
+#ifdef XML_DTD
+    case XML_ROLE_TEXT_DECL:
+      {
+	enum XML_Error result = processXmlDecl(parser, 1, s, next);
+	if (result != XML_ERROR_NONE)
+	  return result;
+	enc = encoding;
+      }
+      break;
+#endif /* XML_DTD */
+    case XML_ROLE_DOCTYPE_PUBLIC_ID:
+      if (startDoctypeDeclHandler) {
+	doctypePubid = poolStoreString(&tempPool, enc, s + 1, next - 1);
+	if (! doctypePubid)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&tempPool);
+      }
+#ifdef XML_DTD
+      declEntity = (ENTITY *)lookup(&dtd.paramEntities,
+				    externalSubsetName,
+				    sizeof(ENTITY));
+      if (!declEntity)
+	return XML_ERROR_NO_MEMORY;
+#endif /* XML_DTD */
+      /* fall through */
+    case XML_ROLE_ENTITY_PUBLIC_ID:
+      if (!XmlIsPublicId(enc, s, next, eventPP))
+	return XML_ERROR_SYNTAX;
+      if (declEntity) {
+	XML_Char *tem = poolStoreString(&dtd.pool,
+	                                enc,
+					s + enc->minBytesPerChar,
+	  				next - enc->minBytesPerChar);
+	if (!tem)
+	  return XML_ERROR_NO_MEMORY;
+	normalizePublicId(tem);
+	declEntity->publicId = tem;
+	poolFinish(&dtd.pool);
+      }
+      break;
+    case XML_ROLE_DOCTYPE_CLOSE:
+      if (doctypeName) {
+	startDoctypeDeclHandler(handlerArg, doctypeName,
+				doctypeSysid, doctypePubid, 0);
+	poolClear(&tempPool);
+      }
+      if (dtd.complete && hadExternalDoctype) {
+	dtd.complete = 0;
+#ifdef XML_DTD
+	if (paramEntityParsing && externalEntityRefHandler) {
+	  ENTITY *entity = (ENTITY *)lookup(&dtd.paramEntities,
+					    externalSubsetName,
+					    0);
+	  if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+					0,
+					entity->base,
+					entity->systemId,
+					entity->publicId))
+	   return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+	}
+#endif /* XML_DTD */
+	if (!dtd.complete
+	    && !dtd.standalone
+	    && notStandaloneHandler
+	    && !notStandaloneHandler(handlerArg))
+	  return XML_ERROR_NOT_STANDALONE;
+      }
+      if (endDoctypeDeclHandler)
+	endDoctypeDeclHandler(handlerArg);
+      break;
+    case XML_ROLE_INSTANCE_START:
+      processor = contentProcessor;
+      return contentProcessor(parser, s, end, nextPtr);
+    case XML_ROLE_ATTLIST_ELEMENT_NAME:
+      declElementType = getElementType(parser, enc, s, next);
+      if (!declElementType)
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_ROLE_ATTRIBUTE_NAME:
+      declAttributeId = getAttributeId(parser, enc, s, next);
+      if (!declAttributeId)
+	return XML_ERROR_NO_MEMORY;
+      declAttributeIsCdata = 0;
+      declAttributeType = 0;
+      declAttributeIsId = 0;
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_CDATA:
+      declAttributeIsCdata = 1;
+      declAttributeType = "CDATA";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_ID:
+      declAttributeIsId = 1;
+      declAttributeType = "ID";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_IDREF:
+      declAttributeType = "IDREF";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_IDREFS:
+      declAttributeType = "IDREFS";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_ENTITY:
+      declAttributeType = "ENTITY";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_ENTITIES:
+      declAttributeType = "ENTITIES";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN:
+      declAttributeType = "NMTOKEN";
+      break;
+    case XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS:
+      declAttributeType = "NMTOKENS";
+      break;
+
+    case XML_ROLE_ATTRIBUTE_ENUM_VALUE:
+    case XML_ROLE_ATTRIBUTE_NOTATION_VALUE:
+      if (attlistDeclHandler)
+      {
+	char *prefix;
+	if (declAttributeType) {
+	  prefix = "|";
+	}
+	else {
+	  prefix = (role == XML_ROLE_ATTRIBUTE_NOTATION_VALUE
+		    ? "NOTATION("
+		    : "(");
+	}
+	if (! poolAppendString(&tempPool, prefix))
+	  return XML_ERROR_NO_MEMORY;
+	if (! poolAppend(&tempPool, enc, s, next))
+	  return XML_ERROR_NO_MEMORY;
+	declAttributeType = tempPool.start;
+      }
+      break;
+    case XML_ROLE_IMPLIED_ATTRIBUTE_VALUE:
+    case XML_ROLE_REQUIRED_ATTRIBUTE_VALUE:
+      if (dtd.complete
+	  && !defineAttribute(declElementType, declAttributeId,
+			      declAttributeIsCdata, declAttributeIsId, 0,
+			      parser))
+	return XML_ERROR_NO_MEMORY;
+      if (attlistDeclHandler && declAttributeType) {
+	if (*declAttributeType == '('
+	    || (*declAttributeType == 'N' && declAttributeType[1] == 'O')) {
+	  /* Enumerated or Notation type */
+	  if (! poolAppendChar(&tempPool, ')')
+	      || ! poolAppendChar(&tempPool, '\0'))
+	    return XML_ERROR_NO_MEMORY;
+	  declAttributeType = tempPool.start;
+	  poolFinish(&tempPool);
+	}
+	*eventEndPP = s;
+	attlistDeclHandler(handlerArg, declElementType->name,
+			   declAttributeId->name, declAttributeType,
+			   0, role == XML_ROLE_REQUIRED_ATTRIBUTE_VALUE);
+	poolClear(&tempPool);
+      }
+      break;
+    case XML_ROLE_DEFAULT_ATTRIBUTE_VALUE:
+    case XML_ROLE_FIXED_ATTRIBUTE_VALUE:
+      {
+	const XML_Char *attVal;
+	enum XML_Error result
+	  = storeAttributeValue(parser, enc, declAttributeIsCdata,
+				s + enc->minBytesPerChar,
+			        next - enc->minBytesPerChar,
+			        &dtd.pool);
+	if (result)
+	  return result;
+	attVal = poolStart(&dtd.pool);
+	poolFinish(&dtd.pool);
+	if (dtd.complete
+	    /* ID attributes aren't allowed to have a default */
+	    && !defineAttribute(declElementType, declAttributeId, declAttributeIsCdata, 0, attVal, parser))
+	  return XML_ERROR_NO_MEMORY;
+	if (attlistDeclHandler && declAttributeType) {
+	  if (*declAttributeType == '('
+	      || (*declAttributeType == 'N' && declAttributeType[1] == 'O')) {
+	    /* Enumerated or Notation type */
+	    if (! poolAppendChar(&tempPool, ')')
+		|| ! poolAppendChar(&tempPool, '\0'))
+	      return XML_ERROR_NO_MEMORY;
+	    declAttributeType = tempPool.start;
+	    poolFinish(&tempPool);
+	  }
+	  *eventEndPP = s;
+	  attlistDeclHandler(handlerArg, declElementType->name,
+			     declAttributeId->name, declAttributeType,
+			     attVal,
+			     role == XML_ROLE_FIXED_ATTRIBUTE_VALUE);
+	  poolClear(&tempPool);
+	}
+	break;
+      }
+    case XML_ROLE_ENTITY_VALUE:
+      {
+	enum XML_Error result = storeEntityValue(parser, enc,
+						 s + enc->minBytesPerChar,
+						 next - enc->minBytesPerChar);
+	if (declEntity) {
+	  declEntity->textPtr = poolStart(&dtd.pool);
+	  declEntity->textLen = poolLength(&dtd.pool);
+	  poolFinish(&dtd.pool);
+	  if (entityDeclHandler) {
+	    *eventEndPP = s;
+	    entityDeclHandler(handlerArg,
+			      declEntity->name,
+			      declEntity->is_param,
+			      declEntity->textPtr,
+			      declEntity->textLen,
+			      curBase, 0, 0, 0);
+	  }
+	}
+	else
+	  poolDiscard(&dtd.pool);
+	if (result != XML_ERROR_NONE)
+	  return result;
+      }
+      break;
+    case XML_ROLE_DOCTYPE_SYSTEM_ID:
+      if (startDoctypeDeclHandler) {
+	doctypeSysid = poolStoreString(&tempPool, enc, s + 1, next - 1);
+	if (! doctypeSysid)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&tempPool);
+      }
+      if (!dtd.standalone
+#ifdef XML_DTD
+	  && !paramEntityParsing
+#endif /* XML_DTD */
+	  && notStandaloneHandler
+	  && !notStandaloneHandler(handlerArg))
+	return XML_ERROR_NOT_STANDALONE;
+      hadExternalDoctype = 1;
+#ifndef XML_DTD
+      break;
+#else /* XML_DTD */
+      if (!declEntity) {
+	declEntity = (ENTITY *)lookup(&dtd.paramEntities,
+				      externalSubsetName,
+				      sizeof(ENTITY));
+	declEntity->publicId = 0;
+	if (!declEntity)
+	  return XML_ERROR_NO_MEMORY;
+      }
+      /* fall through */
+#endif /* XML_DTD */
+    case XML_ROLE_ENTITY_SYSTEM_ID:
+      if (declEntity) {
+	declEntity->systemId = poolStoreString(&dtd.pool, enc,
+	                                       s + enc->minBytesPerChar,
+	  				       next - enc->minBytesPerChar);
+	if (!declEntity->systemId)
+	  return XML_ERROR_NO_MEMORY;
+	declEntity->base = curBase;
+	poolFinish(&dtd.pool);
+      }
+      break;
+    case XML_ROLE_ENTITY_COMPLETE:
+      if (declEntity && entityDeclHandler) {
+	*eventEndPP = s;
+	entityDeclHandler(handlerArg,
+			  declEntity->name,
+			  0,0,0,
+			  declEntity->base,
+			  declEntity->systemId,
+			  declEntity->publicId,
+			  0);
+      }
+      break;
+    case XML_ROLE_ENTITY_NOTATION_NAME:
+      if (declEntity) {
+	declEntity->notation = poolStoreString(&dtd.pool, enc, s, next);
+	if (!declEntity->notation)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&dtd.pool);
+	if (unparsedEntityDeclHandler) {
+	  *eventEndPP = s;
+	  unparsedEntityDeclHandler(handlerArg,
+				    declEntity->name,
+				    declEntity->base,
+				    declEntity->systemId,
+				    declEntity->publicId,
+				    declEntity->notation);
+	}
+	else if (entityDeclHandler) {
+	  *eventEndPP = s;
+	  entityDeclHandler(handlerArg,
+			    declEntity->name,
+			    0,0,0,
+			    declEntity->base,
+			    declEntity->systemId,
+			    declEntity->publicId,
+			    declEntity->notation);
+	}
+      }
+      break;
+    case XML_ROLE_GENERAL_ENTITY_NAME:
+      {
+	const XML_Char *name;
+	if (XmlPredefinedEntityName(enc, s, next)) {
+	  declEntity = 0;
+	  break;
+	}
+	name = poolStoreString(&dtd.pool, enc, s, next);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	if (dtd.complete) {
+	  declEntity = (ENTITY *)lookup(&dtd.generalEntities, name, sizeof(ENTITY));
+	  if (!declEntity)
+	    return XML_ERROR_NO_MEMORY;
+	  if (declEntity->name != name) {
+	    poolDiscard(&dtd.pool);
+	    declEntity = 0;
+	  }
+	  else {
+	    poolFinish(&dtd.pool);
+	    declEntity->publicId = 0;
+	    declEntity->is_param = 0;
+	  }
+	}
+	else {
+	  poolDiscard(&dtd.pool);
+	  declEntity = 0;
+	}
+      }
+      break;
+    case XML_ROLE_PARAM_ENTITY_NAME:
+#ifdef XML_DTD
+      if (dtd.complete) {
+	const XML_Char *name = poolStoreString(&dtd.pool, enc, s, next);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	declEntity = (ENTITY *)lookup(&dtd.paramEntities,
+				      name, sizeof(ENTITY));
+	if (!declEntity)
+	  return XML_ERROR_NO_MEMORY;
+	if (declEntity->name != name) {
+	  poolDiscard(&dtd.pool);
+	  declEntity = 0;
+	}
+	else {
+	  poolFinish(&dtd.pool);
+	  declEntity->publicId = 0;
+	  declEntity->is_param = 1;
+	}
+      }
+#else /* not XML_DTD */
+      declEntity = 0;
+#endif /* not XML_DTD */
+      break;
+    case XML_ROLE_NOTATION_NAME:
+      declNotationPublicId = 0;
+      declNotationName = 0;
+      if (notationDeclHandler) {
+	declNotationName = poolStoreString(&tempPool, enc, s, next);
+	if (!declNotationName)
+	  return XML_ERROR_NO_MEMORY;
+	poolFinish(&tempPool);
+      }
+      break;
+    case XML_ROLE_NOTATION_PUBLIC_ID:
+      if (!XmlIsPublicId(enc, s, next, eventPP))
+	return XML_ERROR_SYNTAX;
+      if (declNotationName) {
+	XML_Char *tem = poolStoreString(&tempPool,
+	                                enc,
+					s + enc->minBytesPerChar,
+	  				next - enc->minBytesPerChar);
+	if (!tem)
+	  return XML_ERROR_NO_MEMORY;
+	normalizePublicId(tem);
+	declNotationPublicId = tem;
+	poolFinish(&tempPool);
+      }
+      break;
+    case XML_ROLE_NOTATION_SYSTEM_ID:
+      if (declNotationName && notationDeclHandler) {
+	const XML_Char *systemId
+	  = poolStoreString(&tempPool, enc,
+			    s + enc->minBytesPerChar,
+	  		    next - enc->minBytesPerChar);
+	if (!systemId)
+	  return XML_ERROR_NO_MEMORY;
+	*eventEndPP = s;
+	notationDeclHandler(handlerArg,
+			    declNotationName,
+			    curBase,
+			    systemId,
+			    declNotationPublicId);
+      }
+      poolClear(&tempPool);
+      break;
+    case XML_ROLE_NOTATION_NO_SYSTEM_ID:
+      if (declNotationPublicId && notationDeclHandler) {
+	*eventEndPP = s;
+	notationDeclHandler(handlerArg,
+			    declNotationName,
+			    curBase,
+			    0,
+			    declNotationPublicId);
+      }
+      poolClear(&tempPool);
+      break;
+    case XML_ROLE_ERROR:
+      switch (tok) {
+      case XML_TOK_PARAM_ENTITY_REF:
+	return XML_ERROR_PARAM_ENTITY_REF;
+      case XML_TOK_XML_DECL:
+	return XML_ERROR_MISPLACED_XML_PI;
+      default:
+	return XML_ERROR_SYNTAX;
+      }
+#ifdef XML_DTD
+    case XML_ROLE_IGNORE_SECT:
+      {
+	enum XML_Error result;
+	if (defaultHandler)
+	  reportDefault(parser, enc, s, next);
+	result = doIgnoreSection(parser, enc, &next, end, nextPtr);
+	if (!next) {
+	  processor = ignoreSectionProcessor;
+	  return result;
+	}
+      }
+      break;
+#endif /* XML_DTD */
+    case XML_ROLE_GROUP_OPEN:
+      if (prologState.level >= groupSize) {
+	if (groupSize) {
+	  groupConnector = REALLOC(groupConnector, groupSize *= 2);
+	  if (dtd.scaffIndex)
+	    dtd.scaffIndex = REALLOC(dtd.scaffIndex, groupSize * sizeof(int));
+	}
+	else
+	  groupConnector = MALLOC(groupSize = 32);
+	if (!groupConnector)
+	  return XML_ERROR_NO_MEMORY;
+      }
+      groupConnector[prologState.level] = 0;
+      if (dtd.in_eldecl) {
+	int myindex = nextScaffoldPart(parser);
+	if (myindex < 0)
+	  return XML_ERROR_NO_MEMORY;
+	dtd.scaffIndex[dtd.scaffLevel] = myindex;
+	dtd.scaffLevel++;
+	dtd.scaffold[myindex].type = XML_CTYPE_SEQ;
+      }
+      break;
+    case XML_ROLE_GROUP_SEQUENCE:
+      if (groupConnector[prologState.level] == '|')
+	return XML_ERROR_SYNTAX;
+      groupConnector[prologState.level] = ',';
+      break;
+    case XML_ROLE_GROUP_CHOICE:
+      if (groupConnector[prologState.level] == ',')
+	return XML_ERROR_SYNTAX;
+      if (dtd.in_eldecl
+	  && ! groupConnector[prologState.level]
+	  && dtd.scaffold[dtd.scaffIndex[dtd.scaffLevel - 1]].type != XML_CTYPE_MIXED
+	  ) {
+	dtd.scaffold[dtd.scaffIndex[dtd.scaffLevel - 1]].type = XML_CTYPE_CHOICE;
+      }
+      groupConnector[prologState.level] = '|';
+      break;
+    case XML_ROLE_PARAM_ENTITY_REF:
+#ifdef XML_DTD
+    case XML_ROLE_INNER_PARAM_ENTITY_REF:
+      if (paramEntityParsing
+	  && (dtd.complete || role == XML_ROLE_INNER_PARAM_ENTITY_REF)) {
+	const XML_Char *name;
+	ENTITY *entity;
+	name = poolStoreString(&dtd.pool, enc,
+				s + enc->minBytesPerChar,
+				next - enc->minBytesPerChar);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	entity = (ENTITY *)lookup(&dtd.paramEntities, name, 0);
+	poolDiscard(&dtd.pool);
+	if (!entity) {
+	  /* FIXME what to do if !dtd.complete? */
+	  return XML_ERROR_UNDEFINED_ENTITY;
+	}
+	if (entity->open)
+	  return XML_ERROR_RECURSIVE_ENTITY_REF;
+	if (entity->textPtr) {
+	  enum XML_Error result;
+	  result = processInternalParamEntity(parser, entity);
+	  if (result != XML_ERROR_NONE)
+	    return result;
+	  break;
+	}
+	if (role == XML_ROLE_INNER_PARAM_ENTITY_REF)
+	  return XML_ERROR_PARAM_ENTITY_REF;
+	if (externalEntityRefHandler) {
+	  dtd.complete = 0;
+	  entity->open = 1;
+	  if (!externalEntityRefHandler(externalEntityRefHandlerArg,
+					0,
+					entity->base,
+					entity->systemId,
+					entity->publicId)) {
+	    entity->open = 0;
+	    return XML_ERROR_EXTERNAL_ENTITY_HANDLING;
+	  }
+	  entity->open = 0;
+	  if (dtd.complete)
+	    break;
+	}
+      }
+#endif /* XML_DTD */
+      if (!dtd.standalone
+	  && notStandaloneHandler
+	  && !notStandaloneHandler(handlerArg))
+	return XML_ERROR_NOT_STANDALONE;
+      dtd.complete = 0;
+      if (defaultHandler)
+	reportDefault(parser, enc, s, next);
+      break;
+
+      /* Element declaration stuff */
+
+    case XML_ROLE_ELEMENT_NAME:
+      if (elementDeclHandler) {
+	declElementType = getElementType(parser, enc, s, next);
+	if (! declElementType)
+	  return XML_ERROR_NO_MEMORY;
+	dtd.scaffLevel = 0;
+	dtd.scaffCount = 0;
+	dtd.in_eldecl = 1;
+      }
+      break;
+
+    case XML_ROLE_CONTENT_ANY:
+    case XML_ROLE_CONTENT_EMPTY:
+      if (dtd.in_eldecl) {
+	if (elementDeclHandler) {
+	  XML_Content * content = (XML_Content *) MALLOC(sizeof(XML_Content));
+	  if (! content)
+	    return XML_ERROR_NO_MEMORY;
+	  content->quant = XML_CQUANT_NONE;
+	  content->name = 0;
+	  content->numchildren = 0;
+	  content->children = 0;
+	  content->type = ((role == XML_ROLE_CONTENT_ANY) ?
+			   XML_CTYPE_ANY :
+			   XML_CTYPE_EMPTY);
+	  *eventEndPP = s;
+	  elementDeclHandler(handlerArg, declElementType->name, content);
+	}
+	dtd.in_eldecl = 0;
+      }
+      break;
+      
+    case XML_ROLE_CONTENT_PCDATA:
+      if (dtd.in_eldecl) {
+	dtd.scaffold[dtd.scaffIndex[dtd.scaffLevel - 1]].type = XML_CTYPE_MIXED;
+      }
+      break;
+
+    case XML_ROLE_CONTENT_ELEMENT:
+      quant = XML_CQUANT_NONE;
+      goto elementContent;
+    case XML_ROLE_CONTENT_ELEMENT_OPT:
+      quant = XML_CQUANT_OPT;
+      goto elementContent;
+    case XML_ROLE_CONTENT_ELEMENT_REP:
+      quant = XML_CQUANT_REP;
+      goto elementContent;
+    case XML_ROLE_CONTENT_ELEMENT_PLUS:
+      quant = XML_CQUANT_PLUS;
+    elementContent:
+      if (dtd.in_eldecl)
+	{
+	  ELEMENT_TYPE *el;
+	  const char *nxt = quant == XML_CQUANT_NONE ? next : next - 1;
+	  int myindex = nextScaffoldPart(parser);
+	  if (myindex < 0)
+	    return XML_ERROR_NO_MEMORY;
+	  dtd.scaffold[myindex].type = XML_CTYPE_NAME;
+	  dtd.scaffold[myindex].quant = quant;
+	  el = getElementType(parser, enc, s, nxt);
+	  if (! el)
+	    return XML_ERROR_NO_MEMORY;
+	  dtd.scaffold[myindex].name = el->name;
+	  dtd.contentStringLen +=  nxt - s + 1;
+	}
+      break;
+
+    case XML_ROLE_GROUP_CLOSE:
+      quant = XML_CQUANT_NONE;
+      goto closeGroup;
+    case XML_ROLE_GROUP_CLOSE_OPT:
+      quant = XML_CQUANT_OPT;
+      goto closeGroup;
+    case XML_ROLE_GROUP_CLOSE_REP:
+      quant = XML_CQUANT_REP;
+      goto closeGroup;
+    case XML_ROLE_GROUP_CLOSE_PLUS:
+      quant = XML_CQUANT_PLUS;
+    closeGroup:
+      if (dtd.in_eldecl) {
+	dtd.scaffLevel--;
+	dtd.scaffold[dtd.scaffIndex[dtd.scaffLevel]].quant = quant;
+	if (dtd.scaffLevel == 0) {
+	  if (elementDeclHandler) {
+	    XML_Content *model = build_model(parser);
+	    if (! model)
+	      return XML_ERROR_NO_MEMORY;
+	    *eventEndPP = s;
+	    elementDeclHandler(handlerArg, declElementType->name, model);
+	  }
+	  dtd.in_eldecl = 0;
+	  dtd.contentStringLen = 0;
+	}
+      }
+      break;
+      /* End element declaration stuff */
+
+    case XML_ROLE_NONE:
+      switch (tok) {
+      case XML_TOK_PI:
+	if (!reportProcessingInstruction(parser, enc, s, next))
+	  return XML_ERROR_NO_MEMORY;
+	break;
+      case XML_TOK_COMMENT:
+	if (!reportComment(parser, enc, s, next))
+	  return XML_ERROR_NO_MEMORY;
+	break;
+      }
+      break;
+    }
+    if (defaultHandler) {
+      switch (tok) {
+      case XML_TOK_PI:
+      case XML_TOK_COMMENT:
+      case XML_TOK_BOM:
+      case XML_TOK_XML_DECL:
+#ifdef XML_DTD
+      case XML_TOK_IGNORE_SECT:
+#endif /* XML_DTD */
+      case XML_TOK_PARAM_ENTITY_REF:
+	break;
+      default:
+#ifdef XML_DTD
+	if (role != XML_ROLE_IGNORE_SECT)
+#endif /* XML_DTD */
+	  reportDefault(parser, enc, s, next);
+      }
+    }
+    s = next;
+    tok = XmlPrologTok(enc, s, end, &next);
+  }
+  /* not reached */
+}
+
+static
+enum XML_Error epilogProcessor(XML_Parser parser,
+			       const char *s,
+			       const char *end,
+			       const char **nextPtr)
+{
+  processor = epilogProcessor;
+  eventPtr = s;
+  for (;;) {
+    const char *next;
+    int tok = XmlPrologTok(encoding, s, end, &next);
+    eventEndPtr = next;
+    switch (tok) {
+    case -XML_TOK_PROLOG_S:
+      if (defaultHandler) {
+	eventEndPtr = end;
+	reportDefault(parser, encoding, s, end);
+      }
+      /* fall through */
+    case XML_TOK_NONE:
+      if (nextPtr)
+	*nextPtr = end;
+      return XML_ERROR_NONE;
+    case XML_TOK_PROLOG_S:
+      if (defaultHandler)
+	reportDefault(parser, encoding, s, next);
+      break;
+    case XML_TOK_PI:
+      if (!reportProcessingInstruction(parser, encoding, s, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_COMMENT:
+      if (!reportComment(parser, encoding, s, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_INVALID:
+      eventPtr = next;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_PARTIAL:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_UNCLOSED_TOKEN;
+    case XML_TOK_PARTIAL_CHAR:
+      if (nextPtr) {
+	*nextPtr = s;
+	return XML_ERROR_NONE;
+      }
+      return XML_ERROR_PARTIAL_CHAR;
+    default:
+      return XML_ERROR_JUNK_AFTER_DOC_ELEMENT;
+    }
+    eventPtr = s = next;
+  }
+}
+
+#ifdef XML_DTD
+
+static enum XML_Error
+processInternalParamEntity(XML_Parser parser, ENTITY *entity)
+{
+  const char *s, *end, *next;
+  int tok;
+  enum XML_Error result;
+  OPEN_INTERNAL_ENTITY openEntity;
+  entity->open = 1;
+  openEntity.next = openInternalEntities;
+  openInternalEntities = &openEntity;
+  openEntity.entity = entity;
+  openEntity.internalEventPtr = 0;
+  openEntity.internalEventEndPtr = 0;
+  s = (char *)entity->textPtr;
+  end = (char *)(entity->textPtr + entity->textLen);
+  tok = XmlPrologTok(internalEncoding, s, end, &next);
+  result = doProlog(parser, internalEncoding, s, end, tok, next, 0);
+  entity->open = 0;
+  openInternalEntities = openEntity.next;
+  return result;
+}
+
+#endif /* XML_DTD */
+
+static
+enum XML_Error errorProcessor(XML_Parser parser,
+			      const char *s,
+			      const char *end,
+			      const char **nextPtr)
+{
+  return errorCode;
+}
+
+static enum XML_Error
+storeAttributeValue(XML_Parser parser, const ENCODING *enc, int isCdata,
+		    const char *ptr, const char *end,
+		    STRING_POOL *pool)
+{
+  enum XML_Error result = appendAttributeValue(parser, enc, isCdata, ptr, end, pool);
+  if (result)
+    return result;
+  if (!isCdata && poolLength(pool) && poolLastChar(pool) == 0x20)
+    poolChop(pool);
+  if (!poolAppendChar(pool, XML_T('\0')))
+    return XML_ERROR_NO_MEMORY;
+  return XML_ERROR_NONE;
+}
+
+static enum XML_Error
+appendAttributeValue(XML_Parser parser, const ENCODING *enc, int isCdata,
+		     const char *ptr, const char *end,
+		     STRING_POOL *pool)
+{
+  for (;;) {
+    const char *next;
+    int tok = XmlAttributeValueTok(enc, ptr, end, &next);
+    switch (tok) {
+    case XML_TOK_NONE:
+      return XML_ERROR_NONE;
+    case XML_TOK_INVALID:
+      if (enc == encoding)
+	eventPtr = next;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_PARTIAL:
+      if (enc == encoding)
+	eventPtr = ptr;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_CHAR_REF:
+      {
+	XML_Char buf[XML_ENCODE_MAX];
+	int i;
+	int n = XmlCharRefNumber(enc, ptr);
+	if (n < 0) {
+	  if (enc == encoding)
+	    eventPtr = ptr;
+      	  return XML_ERROR_BAD_CHAR_REF;
+	}
+	if (!isCdata
+	    && n == 0x20 /* space */
+	    && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
+	  break;
+	n = XmlEncode(n, (ICHAR *)buf);
+	if (!n) {
+	  if (enc == encoding)
+	    eventPtr = ptr;
+	  return XML_ERROR_BAD_CHAR_REF;
+	}
+	for (i = 0; i < n; i++) {
+	  if (!poolAppendChar(pool, buf[i]))
+	    return XML_ERROR_NO_MEMORY;
+	}
+      }
+      break;
+    case XML_TOK_DATA_CHARS:
+      if (!poolAppend(pool, enc, ptr, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_TRAILING_CR:
+      next = ptr + enc->minBytesPerChar;
+      /* fall through */
+    case XML_TOK_ATTRIBUTE_VALUE_S:
+    case XML_TOK_DATA_NEWLINE:
+      if (!isCdata && (poolLength(pool) == 0 || poolLastChar(pool) == 0x20))
+	break;
+      if (!poolAppendChar(pool, 0x20))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_ENTITY_REF:
+      {
+	const XML_Char *name;
+	ENTITY *entity;
+	XML_Char ch = XmlPredefinedEntityName(enc,
+					      ptr + enc->minBytesPerChar,
+					      next - enc->minBytesPerChar);
+	if (ch) {
+	  if (!poolAppendChar(pool, ch))
+  	    return XML_ERROR_NO_MEMORY;
+	  break;
+	}
+	name = poolStoreString(&temp2Pool, enc,
+			       ptr + enc->minBytesPerChar,
+			       next - enc->minBytesPerChar);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	entity = (ENTITY *)lookup(&dtd.generalEntities, name, 0);
+	poolDiscard(&temp2Pool);
+	if (!entity) {
+	  if (dtd.complete) {
+	    if (enc == encoding)
+	      eventPtr = ptr;
+	    return XML_ERROR_UNDEFINED_ENTITY;
+	  }
+	}
+	else if (entity->open) {
+	  if (enc == encoding)
+	    eventPtr = ptr;
+	  return XML_ERROR_RECURSIVE_ENTITY_REF;
+	}
+	else if (entity->notation) {
+	  if (enc == encoding)
+	    eventPtr = ptr;
+	  return XML_ERROR_BINARY_ENTITY_REF;
+	}
+	else if (!entity->textPtr) {
+	  if (enc == encoding)
+	    eventPtr = ptr;
+  	  return XML_ERROR_ATTRIBUTE_EXTERNAL_ENTITY_REF;
+	}
+	else {
+	  enum XML_Error result;
+	  const XML_Char *textEnd = entity->textPtr + entity->textLen;
+	  entity->open = 1;
+	  result = appendAttributeValue(parser, internalEncoding, isCdata, (char *)entity->textPtr, (char *)textEnd, pool);
+	  entity->open = 0;
+	  if (result)
+	    return result;
+	}
+      }
+      break;
+    default:
+      if (enc == encoding)
+	eventPtr = ptr;
+      return XML_ERROR_UNEXPECTED_STATE;
+    }
+    ptr = next;
+  }
+  /* not reached */
+}
+
+static
+enum XML_Error storeEntityValue(XML_Parser parser,
+				const ENCODING *enc,
+				const char *entityTextPtr,
+				const char *entityTextEnd)
+{
+  STRING_POOL *pool = &(dtd.pool);
+  for (;;) {
+    const char *next;
+    int tok = XmlEntityValueTok(enc, entityTextPtr, entityTextEnd, &next);
+    switch (tok) {
+    case XML_TOK_PARAM_ENTITY_REF:
+#ifdef XML_DTD
+      if (parentParser || enc != encoding) {
+	enum XML_Error result;
+	const XML_Char *name;
+	ENTITY *entity;
+	name = poolStoreString(&tempPool, enc,
+			       entityTextPtr + enc->minBytesPerChar,
+			       next - enc->minBytesPerChar);
+	if (!name)
+	  return XML_ERROR_NO_MEMORY;
+	entity = (ENTITY *)lookup(&dtd.paramEntities, name, 0);
+	poolDiscard(&tempPool);
+	if (!entity) {
+	  if (enc == encoding)
+	    eventPtr = entityTextPtr;
+	  return XML_ERROR_UNDEFINED_ENTITY;
+	}
+	if (entity->open) {
+	  if (enc == encoding)
+	    eventPtr = entityTextPtr;
+	  return XML_ERROR_RECURSIVE_ENTITY_REF;
+	}
+	if (entity->systemId) {
+	  if (enc == encoding)
+	    eventPtr = entityTextPtr;
+	  return XML_ERROR_PARAM_ENTITY_REF;
+	}
+	entity->open = 1;
+	result = storeEntityValue(parser,
+				  internalEncoding,
+				  (char *)entity->textPtr,
+				  (char *)(entity->textPtr + entity->textLen));
+	entity->open = 0;
+	if (result)
+	  return result;
+	break;
+      }
+#endif /* XML_DTD */
+      eventPtr = entityTextPtr;
+      return XML_ERROR_SYNTAX;
+    case XML_TOK_NONE:
+      return XML_ERROR_NONE;
+    case XML_TOK_ENTITY_REF:
+    case XML_TOK_DATA_CHARS:
+      if (!poolAppend(pool, enc, entityTextPtr, next))
+	return XML_ERROR_NO_MEMORY;
+      break;
+    case XML_TOK_TRAILING_CR:
+      next = entityTextPtr + enc->minBytesPerChar;
+      /* fall through */
+    case XML_TOK_DATA_NEWLINE:
+      if (pool->end == pool->ptr && !poolGrow(pool))
+	return XML_ERROR_NO_MEMORY;
+      *(pool->ptr)++ = 0xA;
+      break;
+    case XML_TOK_CHAR_REF:
+      {
+	XML_Char buf[XML_ENCODE_MAX];
+	int i;
+	int n = XmlCharRefNumber(enc, entityTextPtr);
+	if (n < 0) {
+	  if (enc == encoding)
+	    eventPtr = entityTextPtr;
+	  return XML_ERROR_BAD_CHAR_REF;
+	}
+	n = XmlEncode(n, (ICHAR *)buf);
+	if (!n) {
+	  if (enc == encoding)
+	    eventPtr = entityTextPtr;
+	  return XML_ERROR_BAD_CHAR_REF;
+	}
+	for (i = 0; i < n; i++) {
+	  if (pool->end == pool->ptr && !poolGrow(pool))
+	    return XML_ERROR_NO_MEMORY;
+	  *(pool->ptr)++ = buf[i];
+	}
+      }
+      break;
+    case XML_TOK_PARTIAL:
+      if (enc == encoding)
+	eventPtr = entityTextPtr;
+      return XML_ERROR_INVALID_TOKEN;
+    case XML_TOK_INVALID:
+      if (enc == encoding)
+	eventPtr = next;
+      return XML_ERROR_INVALID_TOKEN;
+    default:
+      if (enc == encoding)
+	eventPtr = entityTextPtr;
+      return XML_ERROR_UNEXPECTED_STATE;
+    }
+    entityTextPtr = next;
+  }
+  /* not reached */
+}
+
+static void
+normalizeLines(XML_Char *s)
+{
+  XML_Char *p;
+  for (;; s++) {
+    if (*s == XML_T('\0'))
+      return;
+    if (*s == 0xD)
+      break;
+  }
+  p = s;
+  do {
+    if (*s == 0xD) {
+      *p++ = 0xA;
+      if (*++s == 0xA)
+        s++;
+    }
+    else
+      *p++ = *s++;
+  } while (*s);
+  *p = XML_T('\0');
+}
+
+static int
+reportProcessingInstruction(XML_Parser parser, const ENCODING *enc, const char *start, const char *end)
+{
+  const XML_Char *target;
+  XML_Char *data;
+  const char *tem;
+  if (!processingInstructionHandler) {
+    if (defaultHandler)
+      reportDefault(parser, enc, start, end);
+    return 1;
+  }
+  start += enc->minBytesPerChar * 2;
+  tem = start + XmlNameLength(enc, start);
+  target = poolStoreString(&tempPool, enc, start, tem);
+  if (!target)
+    return 0;
+  poolFinish(&tempPool);
+  data = poolStoreString(&tempPool, enc,
+			XmlSkipS(enc, tem),
+			end - enc->minBytesPerChar*2);
+  if (!data)
+    return 0;
+  normalizeLines(data);
+  processingInstructionHandler(handlerArg, target, data);
+  poolClear(&tempPool);
+  return 1;
+}
+
+static int
+reportComment(XML_Parser parser, const ENCODING *enc, const char *start, const char *end)
+{
+  XML_Char *data;
+  if (!commentHandler) {
+    if (defaultHandler)
+      reportDefault(parser, enc, start, end);
+    return 1;
+  }
+  data = poolStoreString(&tempPool,
+                         enc,
+                         start + enc->minBytesPerChar * 4, 
+			 end - enc->minBytesPerChar * 3);
+  if (!data)
+    return 0;
+  normalizeLines(data);
+  commentHandler(handlerArg, data);
+  poolClear(&tempPool);
+  return 1;
+}
+
+static void
+reportDefault(XML_Parser parser, const ENCODING *enc, const char *s, const char *end)
+{
+  if (MUST_CONVERT(enc, s)) {
+    const char **eventPP;
+    const char **eventEndPP;
+    if (enc == encoding) {
+      eventPP = &eventPtr;
+      eventEndPP = &eventEndPtr;
+    }
+    else {
+      eventPP = &(openInternalEntities->internalEventPtr);
+      eventEndPP = &(openInternalEntities->internalEventEndPtr);
+    }
+    do {
+      ICHAR *dataPtr = (ICHAR *)dataBuf;
+      XmlConvert(enc, &s, end, &dataPtr, (ICHAR *)dataBufEnd);
+      *eventEndPP = s;
+      defaultHandler(handlerArg, dataBuf, dataPtr - (ICHAR *)dataBuf);
+      *eventPP = s;
+    } while (s != end);
+  }
+  else
+    defaultHandler(handlerArg, (XML_Char *)s, (XML_Char *)end - (XML_Char *)s);
+}
+
+
+static int
+defineAttribute(ELEMENT_TYPE *type, ATTRIBUTE_ID *attId, int isCdata,
+		int isId, const XML_Char *value, XML_Parser parser)
+{
+  DEFAULT_ATTRIBUTE *att;
+  if (value || isId) {
+    /* The handling of default attributes gets messed up if we have
+       a default which duplicates a non-default. */
+    int i;
+    for (i = 0; i < type->nDefaultAtts; i++)
+      if (attId == type->defaultAtts[i].id)
+	return 1;
+    if (isId && !type->idAtt && !attId->xmlns)
+      type->idAtt = attId;
+  }
+  if (type->nDefaultAtts == type->allocDefaultAtts) {
+    if (type->allocDefaultAtts == 0) {
+      type->allocDefaultAtts = 8;
+      type->defaultAtts = MALLOC(type->allocDefaultAtts*sizeof(DEFAULT_ATTRIBUTE));
+    }
+    else {
+      type->allocDefaultAtts *= 2;
+      type->defaultAtts = REALLOC(type->defaultAtts,
+				  type->allocDefaultAtts*sizeof(DEFAULT_ATTRIBUTE));
+    }
+    if (!type->defaultAtts)
+      return 0;
+  }
+  att = type->defaultAtts + type->nDefaultAtts;
+  att->id = attId;
+  att->value = value;
+  att->isCdata = isCdata;
+  if (!isCdata)
+    attId->maybeTokenized = 1;
+  type->nDefaultAtts += 1;
+  return 1;
+}
+
+static int setElementTypePrefix(XML_Parser parser, ELEMENT_TYPE *elementType)
+{
+  const XML_Char *name;
+  for (name = elementType->name; *name; name++) {
+    if (*name == XML_T(':')) {
+      PREFIX *prefix;
+      const XML_Char *s;
+      for (s = elementType->name; s != name; s++) {
+	if (!poolAppendChar(&dtd.pool, *s))
+	  return 0;
+      }
+      if (!poolAppendChar(&dtd.pool, XML_T('\0')))
+	return 0;
+      prefix = (PREFIX *)lookup(&dtd.prefixes, poolStart(&dtd.pool), sizeof(PREFIX));
+      if (!prefix)
+	return 0;
+      if (prefix->name == poolStart(&dtd.pool))
+	poolFinish(&dtd.pool);
+      else
+	poolDiscard(&dtd.pool);
+      elementType->prefix = prefix;
+
+    }
+  }
+  return 1;
+}
+
+static ATTRIBUTE_ID *
+getAttributeId(XML_Parser parser, const ENCODING *enc, const char *start, const char *end)
+{
+  ATTRIBUTE_ID *id;
+  const XML_Char *name;
+  if (!poolAppendChar(&dtd.pool, XML_T('\0')))
+    return 0;
+  name = poolStoreString(&dtd.pool, enc, start, end);
+  if (!name)
+    return 0;
+  ++name;
+  id = (ATTRIBUTE_ID *)lookup(&dtd.attributeIds, name, sizeof(ATTRIBUTE_ID));
+  if (!id)
+    return 0;
+  if (id->name != name)
+    poolDiscard(&dtd.pool);
+  else {
+    poolFinish(&dtd.pool);
+    if (!ns)
+      ;
+    else if (name[0] == 'x'
+	&& name[1] == 'm'
+	&& name[2] == 'l'
+	&& name[3] == 'n'
+	&& name[4] == 's'
+	&& (name[5] == XML_T('\0') || name[5] == XML_T(':'))) {
+      if (name[5] == '\0')
+	id->prefix = &dtd.defaultPrefix;
+      else
+	id->prefix = (PREFIX *)lookup(&dtd.prefixes, name + 6, sizeof(PREFIX));
+      id->xmlns = 1;
+    }
+    else {
+      int i;
+      for (i = 0; name[i]; i++) {
+	if (name[i] == XML_T(':')) {
+	  int j;
+	  for (j = 0; j < i; j++) {
+	    if (!poolAppendChar(&dtd.pool, name[j]))
+	      return 0;
+	  }
+	  if (!poolAppendChar(&dtd.pool, XML_T('\0')))
+	    return 0;
+	  id->prefix = (PREFIX *)lookup(&dtd.prefixes, poolStart(&dtd.pool), sizeof(PREFIX));
+	  if (id->prefix->name == poolStart(&dtd.pool))
+	    poolFinish(&dtd.pool);
+	  else
+	    poolDiscard(&dtd.pool);
+	  break;
+	}
+      }
+    }
+  }
+  return id;
+}
+
+#define CONTEXT_SEP XML_T('\f')
+
+static
+const XML_Char *getContext(XML_Parser parser)
+{
+  HASH_TABLE_ITER iter;
+  int needSep = 0;
+
+  if (dtd.defaultPrefix.binding) {
+    int i;
+    int len;
+    if (!poolAppendChar(&tempPool, XML_T('=')))
+      return 0;
+    len = dtd.defaultPrefix.binding->uriLen;
+    if (namespaceSeparator != XML_T('\0'))
+      len--;
+    for (i = 0; i < len; i++)
+      if (!poolAppendChar(&tempPool, dtd.defaultPrefix.binding->uri[i]))
+  	return 0;
+    needSep = 1;
+  }
+
+  hashTableIterInit(&iter, &(dtd.prefixes));
+  for (;;) {
+    int i;
+    int len;
+    const XML_Char *s;
+    PREFIX *prefix = (PREFIX *)hashTableIterNext(&iter);
+    if (!prefix)
+      break;
+    if (!prefix->binding)
+      continue;
+    if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
+      return 0;
+    for (s = prefix->name; *s; s++)
+      if (!poolAppendChar(&tempPool, *s))
+        return 0;
+    if (!poolAppendChar(&tempPool, XML_T('=')))
+      return 0;
+    len = prefix->binding->uriLen;
+    if (namespaceSeparator != XML_T('\0'))
+      len--;
+    for (i = 0; i < len; i++)
+      if (!poolAppendChar(&tempPool, prefix->binding->uri[i]))
+        return 0;
+    needSep = 1;
+  }
+
+
+  hashTableIterInit(&iter, &(dtd.generalEntities));
+  for (;;) {
+    const XML_Char *s;
+    ENTITY *e = (ENTITY *)hashTableIterNext(&iter);
+    if (!e)
+      break;
+    if (!e->open)
+      continue;
+    if (needSep && !poolAppendChar(&tempPool, CONTEXT_SEP))
+      return 0;
+    for (s = e->name; *s; s++)
+      if (!poolAppendChar(&tempPool, *s))
+        return 0;
+    needSep = 1;
+  }
+
+  if (!poolAppendChar(&tempPool, XML_T('\0')))
+    return 0;
+  return tempPool.start;
+}
+
+static
+int setContext(XML_Parser parser, const XML_Char *context)
+{
+  const XML_Char *s = context;
+
+  while (*context != XML_T('\0')) {
+    if (*s == CONTEXT_SEP || *s == XML_T('\0')) {
+      ENTITY *e;
+      if (!poolAppendChar(&tempPool, XML_T('\0')))
+	return 0;
+      e = (ENTITY *)lookup(&dtd.generalEntities, poolStart(&tempPool), 0);
+      if (e)
+	e->open = 1;
+      if (*s != XML_T('\0'))
+	s++;
+      context = s;
+      poolDiscard(&tempPool);
+    }
+    else if (*s == '=') {
+      PREFIX *prefix;
+      if (poolLength(&tempPool) == 0)
+	prefix = &dtd.defaultPrefix;
+      else {
+	if (!poolAppendChar(&tempPool, XML_T('\0')))
+	  return 0;
+	prefix = (PREFIX *)lookup(&dtd.prefixes, poolStart(&tempPool), sizeof(PREFIX));
+	if (!prefix)
+	  return 0;
+        if (prefix->name == poolStart(&tempPool)) {
+	  prefix->name = poolCopyString(&dtd.pool, prefix->name);
+	  if (!prefix->name)
+	    return 0;
+	}
+	poolDiscard(&tempPool);
+      }
+      for (context = s + 1; *context != CONTEXT_SEP && *context != XML_T('\0'); context++)
+        if (!poolAppendChar(&tempPool, *context))
+          return 0;
+      if (!poolAppendChar(&tempPool, XML_T('\0')))
+	return 0;
+      if (!addBinding(parser, prefix, 0, poolStart(&tempPool), &inheritedBindings))
+	return 0;
+      poolDiscard(&tempPool);
+      if (*context != XML_T('\0'))
+	++context;
+      s = context;
+    }
+    else {
+      if (!poolAppendChar(&tempPool, *s))
+	return 0;
+      s++;
+    }
+  }
+  return 1;
+}
+
+
+static
+void normalizePublicId(XML_Char *publicId)
+{
+  XML_Char *p = publicId;
+  XML_Char *s;
+  for (s = publicId; *s; s++) {
+    switch (*s) {
+    case 0x20:
+    case 0xD:
+    case 0xA:
+      if (p != publicId && p[-1] != 0x20)
+	*p++ = 0x20;
+      break;
+    default:
+      *p++ = *s;
+    }
+  }
+  if (p != publicId && p[-1] == 0x20)
+    --p;
+  *p = XML_T('\0');
+}
+
+static int dtdInit(DTD *p, XML_Parser parser)
+{
+  XML_Memory_Handling_Suite *ms = &((Parser *) parser)->m_mem; 
+  poolInit(&(p->pool), ms);
+  hashTableInit(&(p->generalEntities), ms);
+  hashTableInit(&(p->elementTypes), ms);
+  hashTableInit(&(p->attributeIds), ms);
+  hashTableInit(&(p->prefixes), ms);
+  p->complete = 1;
+  p->standalone = 0;
+#ifdef XML_DTD
+  hashTableInit(&(p->paramEntities), ms);
+#endif /* XML_DTD */
+  p->defaultPrefix.name = 0;
+  p->defaultPrefix.binding = 0;
+
+  p->in_eldecl = 0;
+  p->scaffIndex = 0;
+  p->scaffLevel = 0;
+  p->scaffold = 0;
+  p->contentStringLen = 0;
+  p->scaffSize = 0;
+  p->scaffCount = 0;
+
+  return 1;
+}
+
+#ifdef XML_DTD
+
+static void dtdSwap(DTD *p1, DTD *p2)
+{
+  DTD tem;
+  memcpy(&tem, p1, sizeof(DTD));
+  memcpy(p1, p2, sizeof(DTD));
+  memcpy(p2, &tem, sizeof(DTD));
+}
+
+#endif /* XML_DTD */
+
+static void dtdDestroy(DTD *p, XML_Parser parser)
+{
+  HASH_TABLE_ITER iter;
+  hashTableIterInit(&iter, &(p->elementTypes));
+  for (;;) {
+    ELEMENT_TYPE *e = (ELEMENT_TYPE *)hashTableIterNext(&iter);
+    if (!e)
+      break;
+    if (e->allocDefaultAtts != 0)
+      FREE(e->defaultAtts);
+  }
+  hashTableDestroy(&(p->generalEntities));
+#ifdef XML_DTD
+  hashTableDestroy(&(p->paramEntities));
+#endif /* XML_DTD */
+  hashTableDestroy(&(p->elementTypes));
+  hashTableDestroy(&(p->attributeIds));
+  hashTableDestroy(&(p->prefixes));
+  poolDestroy(&(p->pool));
+  if (p->scaffIndex)
+    FREE(p->scaffIndex);
+  if (p->scaffold)
+    FREE(p->scaffold);
+}
+
+/* Do a deep copy of the DTD.  Return 0 for out of memory; non-zero otherwise.
+The new DTD has already been initialized. */
+
+static int dtdCopy(DTD *newDtd, const DTD *oldDtd, XML_Parser parser)
+{
+  HASH_TABLE_ITER iter;
+
+  /* Copy the prefix table. */
+
+  hashTableIterInit(&iter, &(oldDtd->prefixes));
+  for (;;) {
+    const XML_Char *name;
+    const PREFIX *oldP = (PREFIX *)hashTableIterNext(&iter);
+    if (!oldP)
+      break;
+    name = poolCopyString(&(newDtd->pool), oldP->name);
+    if (!name)
+      return 0;
+    if (!lookup(&(newDtd->prefixes), name, sizeof(PREFIX)))
+      return 0;
+  }
+
+  hashTableIterInit(&iter, &(oldDtd->attributeIds));
+
+  /* Copy the attribute id table. */
+
+  for (;;) {
+    ATTRIBUTE_ID *newA;
+    const XML_Char *name;
+    const ATTRIBUTE_ID *oldA = (ATTRIBUTE_ID *)hashTableIterNext(&iter);
+
+    if (!oldA)
+      break;
+    /* Remember to allocate the scratch byte before the name. */
+    if (!poolAppendChar(&(newDtd->pool), XML_T('\0')))
+      return 0;
+    name = poolCopyString(&(newDtd->pool), oldA->name);
+    if (!name)
+      return 0;
+    ++name;
+    newA = (ATTRIBUTE_ID *)lookup(&(newDtd->attributeIds), name, sizeof(ATTRIBUTE_ID));
+    if (!newA)
+      return 0;
+    newA->maybeTokenized = oldA->maybeTokenized;
+    if (oldA->prefix) {
+      newA->xmlns = oldA->xmlns;
+      if (oldA->prefix == &oldDtd->defaultPrefix)
+	newA->prefix = &newDtd->defaultPrefix;
+      else
+	newA->prefix = (PREFIX *)lookup(&(newDtd->prefixes), oldA->prefix->name, 0);
+    }
+  }
+
+  /* Copy the element type table. */
+
+  hashTableIterInit(&iter, &(oldDtd->elementTypes));
+
+  for (;;) {
+    int i;
+    ELEMENT_TYPE *newE;
+    const XML_Char *name;
+    const ELEMENT_TYPE *oldE = (ELEMENT_TYPE *)hashTableIterNext(&iter);
+    if (!oldE)
+      break;
+    name = poolCopyString(&(newDtd->pool), oldE->name);
+    if (!name)
+      return 0;
+    newE = (ELEMENT_TYPE *)lookup(&(newDtd->elementTypes), name, sizeof(ELEMENT_TYPE));
+    if (!newE)
+      return 0;
+    if (oldE->nDefaultAtts) {
+      newE->defaultAtts = (DEFAULT_ATTRIBUTE *)MALLOC(oldE->nDefaultAtts * sizeof(DEFAULT_ATTRIBUTE));
+      if (!newE->defaultAtts)
+	return 0;
+    }
+    if (oldE->idAtt)
+      newE->idAtt = (ATTRIBUTE_ID *)lookup(&(newDtd->attributeIds), oldE->idAtt->name, 0);
+    newE->allocDefaultAtts = newE->nDefaultAtts = oldE->nDefaultAtts;
+    if (oldE->prefix)
+      newE->prefix = (PREFIX *)lookup(&(newDtd->prefixes), oldE->prefix->name, 0);
+    for (i = 0; i < newE->nDefaultAtts; i++) {
+      newE->defaultAtts[i].id = (ATTRIBUTE_ID *)lookup(&(newDtd->attributeIds), oldE->defaultAtts[i].id->name, 0);
+      newE->defaultAtts[i].isCdata = oldE->defaultAtts[i].isCdata;
+      if (oldE->defaultAtts[i].value) {
+	newE->defaultAtts[i].value = poolCopyString(&(newDtd->pool), oldE->defaultAtts[i].value);
+	if (!newE->defaultAtts[i].value)
+  	  return 0;
+      }
+      else
+	newE->defaultAtts[i].value = 0;
+    }
+  }
+
+  /* Copy the entity tables. */
+  if (!copyEntityTable(&(newDtd->generalEntities),
+		       &(newDtd->pool),
+		       &(oldDtd->generalEntities), parser))
+      return 0;
+
+#ifdef XML_DTD
+  if (!copyEntityTable(&(newDtd->paramEntities),
+		       &(newDtd->pool),
+		       &(oldDtd->paramEntities), parser))
+      return 0;
+#endif /* XML_DTD */
+
+  newDtd->complete = oldDtd->complete;
+  newDtd->standalone = oldDtd->standalone;
+
+  /* Don't want deep copying for scaffolding */
+  newDtd->in_eldecl = oldDtd->in_eldecl;
+  newDtd->scaffold = oldDtd->scaffold;
+  newDtd->contentStringLen = oldDtd->contentStringLen;
+  newDtd->scaffSize = oldDtd->scaffSize;
+  newDtd->scaffLevel = oldDtd->scaffLevel;
+  newDtd->scaffIndex = oldDtd->scaffIndex;
+
+  return 1;
+}  /* End dtdCopy */
+
+static int copyEntityTable(HASH_TABLE *newTable,
+			   STRING_POOL *newPool,
+			   const HASH_TABLE *oldTable,
+			   XML_Parser parser)
+{
+  HASH_TABLE_ITER iter;
+  const XML_Char *cachedOldBase = 0;
+  const XML_Char *cachedNewBase = 0;
+
+  hashTableIterInit(&iter, oldTable);
+
+  for (;;) {
+    ENTITY *newE;
+    const XML_Char *name;
+    const ENTITY *oldE = (ENTITY *)hashTableIterNext(&iter);
+    if (!oldE)
+      break;
+    name = poolCopyString(newPool, oldE->name);
+    if (!name)
+      return 0;
+    newE = (ENTITY *)lookup(newTable, name, sizeof(ENTITY));
+    if (!newE)
+      return 0;
+    if (oldE->systemId) {
+      const XML_Char *tem = poolCopyString(newPool, oldE->systemId);
+      if (!tem)
+	return 0;
+      newE->systemId = tem;
+      if (oldE->base) {
+	if (oldE->base == cachedOldBase)
+	  newE->base = cachedNewBase;
+	else {
+	  cachedOldBase = oldE->base;
+	  tem = poolCopyString(newPool, cachedOldBase);
+	  if (!tem)
+	    return 0;
+	  cachedNewBase = newE->base = tem;
+	}
+      }
+    }
+    else {
+      const XML_Char *tem = poolCopyStringN(newPool, oldE->textPtr, oldE->textLen);
+      if (!tem)
+	return 0;
+      newE->textPtr = tem;
+      newE->textLen = oldE->textLen;
+    }
+    if (oldE->notation) {
+      const XML_Char *tem = poolCopyString(newPool, oldE->notation);
+      if (!tem)
+	return 0;
+      newE->notation = tem;
+    }
+  }
+  return 1;
+}
+
+#define INIT_SIZE 64
+
+static
+int keyeq(KEY s1, KEY s2)
+{
+  for (; *s1 == *s2; s1++, s2++)
+    if (*s1 == 0)
+      return 1;
+  return 0;
+}
+
+static
+unsigned long hash(KEY s)
+{
+  unsigned long h = 0;
+  while (*s)
+    h = (h << 5) + h + (unsigned char)*s++;
+  return h;
+}
+
+static
+NAMED *lookup(HASH_TABLE *table, KEY name, size_t createSize)
+{
+  size_t i;
+  if (table->size == 0) {
+    size_t tsize;
+
+    if (!createSize)
+      return 0;
+    tsize = INIT_SIZE * sizeof(NAMED *);
+    table->v = table->mem->malloc_fcn(tsize);
+    if (!table->v)
+      return 0;
+    memset(table->v, 0, tsize);
+    table->size = INIT_SIZE;
+    table->usedLim = INIT_SIZE / 2;
+    i = hash(name) & (table->size - 1);
+  }
+  else {
+    unsigned long h = hash(name);
+    for (i = h & (table->size - 1);
+         table->v[i];
+         i == 0 ? i = table->size - 1 : --i) {
+      if (keyeq(name, table->v[i]->name))
+	return table->v[i];
+    }
+    if (!createSize)
+      return 0;
+    if (table->used == table->usedLim) {
+      /* check for overflow */
+      size_t newSize = table->size * 2;
+      size_t tsize = newSize * sizeof(NAMED *);
+      NAMED **newV = table->mem->malloc_fcn(tsize);
+      if (!newV)
+	return 0;
+      memset(newV, 0, tsize);
+      for (i = 0; i < table->size; i++)
+	if (table->v[i]) {
+	  size_t j;
+	  for (j = hash(table->v[i]->name) & (newSize - 1);
+	       newV[j];
+	       j == 0 ? j = newSize - 1 : --j)
+	    ;
+	  newV[j] = table->v[i];
+	}
+      table->mem->free_fcn(table->v);
+      table->v = newV;
+      table->size = newSize;
+      table->usedLim = newSize/2;
+      for (i = h & (table->size - 1);
+	   table->v[i];
+	   i == 0 ? i = table->size - 1 : --i)
+	;
+    }
+  }
+  table->v[i] = table->mem->malloc_fcn(createSize);
+  if (!table->v[i])
+    return 0;
+  memset(table->v[i], 0, createSize);
+  table->v[i]->name = name;
+  (table->used)++;
+  return table->v[i];
+}
+
+static
+void hashTableDestroy(HASH_TABLE *table)
+{
+  size_t i;
+  for (i = 0; i < table->size; i++) {
+    NAMED *p = table->v[i];
+    if (p)
+      table->mem->free_fcn(p);
+  }
+  if (table->v)
+    table->mem->free_fcn(table->v);
+}
+
+static
+void hashTableInit(HASH_TABLE *p, XML_Memory_Handling_Suite *ms)
+{
+  p->size = 0;
+  p->usedLim = 0;
+  p->used = 0;
+  p->v = 0;
+  p->mem = ms;
+}
+
+static
+void hashTableIterInit(HASH_TABLE_ITER *iter, const HASH_TABLE *table)
+{
+  iter->p = table->v;
+  iter->end = iter->p + table->size;
+}
+
+static
+NAMED *hashTableIterNext(HASH_TABLE_ITER *iter)
+{
+  while (iter->p != iter->end) {
+    NAMED *tem = *(iter->p)++;
+    if (tem)
+      return tem;
+  }
+  return 0;
+}
+
+
+static
+void poolInit(STRING_POOL *pool, XML_Memory_Handling_Suite *ms)
+{
+  pool->blocks = 0;
+  pool->freeBlocks = 0;
+  pool->start = 0;
+  pool->ptr = 0;
+  pool->end = 0;
+  pool->mem = ms;
+}
+
+static
+void poolClear(STRING_POOL *pool)
+{
+  if (!pool->freeBlocks)
+    pool->freeBlocks = pool->blocks;
+  else {
+    BLOCK *p = pool->blocks;
+    while (p) {
+      BLOCK *tem = p->next;
+      p->next = pool->freeBlocks;
+      pool->freeBlocks = p;
+      p = tem;
+    }
+  }
+  pool->blocks = 0;
+  pool->start = 0;
+  pool->ptr = 0;
+  pool->end = 0;
+}
+
+static
+void poolDestroy(STRING_POOL *pool)
+{
+  BLOCK *p = pool->blocks;
+  while (p) {
+    BLOCK *tem = p->next;
+    pool->mem->free_fcn(p);
+    p = tem;
+  }
+  pool->blocks = 0;
+  p = pool->freeBlocks;
+  while (p) {
+    BLOCK *tem = p->next;
+    pool->mem->free_fcn(p);
+    p = tem;
+  }
+  pool->freeBlocks = 0;
+  pool->ptr = 0;
+  pool->start = 0;
+  pool->end = 0;
+}
+
+static
+XML_Char *poolAppend(STRING_POOL *pool, const ENCODING *enc,
+		     const char *ptr, const char *end)
+{
+  if (!pool->ptr && !poolGrow(pool))
+    return 0;
+  for (;;) {
+    XmlConvert(enc, &ptr, end, (ICHAR **)&(pool->ptr), (ICHAR *)pool->end);
+    if (ptr == end)
+      break;
+    if (!poolGrow(pool))
+      return 0;
+  }
+  return pool->start;
+}
+
+static const XML_Char *poolCopyString(STRING_POOL *pool, const XML_Char *s)
+{
+  do {
+    if (!poolAppendChar(pool, *s))
+      return 0;
+  } while (*s++);
+  s = pool->start;
+  poolFinish(pool);
+  return s;
+}
+
+static const XML_Char *poolCopyStringN(STRING_POOL *pool, const XML_Char *s, int n)
+{
+  if (!pool->ptr && !poolGrow(pool))
+    return 0;
+  for (; n > 0; --n, s++) {
+    if (!poolAppendChar(pool, *s))
+      return 0;
+
+  }
+  s = pool->start;
+  poolFinish(pool);
+  return s;
+}
+
+static
+const XML_Char *poolAppendString(STRING_POOL *pool, const XML_Char *s)
+{
+  while (*s) {
+    if (!poolAppendChar(pool, *s))
+      return 0;
+    s++;
+  } 
+  return pool->start;
+}  /* End poolAppendString */
+
+static
+XML_Char *poolStoreString(STRING_POOL *pool, const ENCODING *enc,
+			  const char *ptr, const char *end)
+{
+  if (!poolAppend(pool, enc, ptr, end))
+    return 0;
+  if (pool->ptr == pool->end && !poolGrow(pool))
+    return 0;
+  *(pool->ptr)++ = 0;
+  return pool->start;
+}
+
+static
+int poolGrow(STRING_POOL *pool)
+{
+  if (pool->freeBlocks) {
+    if (pool->start == 0) {
+      pool->blocks = pool->freeBlocks;
+      pool->freeBlocks = pool->freeBlocks->next;
+      pool->blocks->next = 0;
+      pool->start = pool->blocks->s;
+      pool->end = pool->start + pool->blocks->size;
+      pool->ptr = pool->start;
+      return 1;
+    }
+    if (pool->end - pool->start < pool->freeBlocks->size) {
+      BLOCK *tem = pool->freeBlocks->next;
+      pool->freeBlocks->next = pool->blocks;
+      pool->blocks = pool->freeBlocks;
+      pool->freeBlocks = tem;
+      memcpy(pool->blocks->s, pool->start, (pool->end - pool->start) * sizeof(XML_Char));
+      pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
+      pool->start = pool->blocks->s;
+      pool->end = pool->start + pool->blocks->size;
+      return 1;
+    }
+  }
+  if (pool->blocks && pool->start == pool->blocks->s) {
+    int blockSize = (pool->end - pool->start)*2;
+    pool->blocks = pool->mem->realloc_fcn(pool->blocks, offsetof(BLOCK, s) + blockSize * sizeof(XML_Char));
+    if (!pool->blocks)
+      return 0;
+    pool->blocks->size = blockSize;
+    pool->ptr = pool->blocks->s + (pool->ptr - pool->start);
+    pool->start = pool->blocks->s;
+    pool->end = pool->start + blockSize;
+  }
+  else {
+    BLOCK *tem;
+    int blockSize = pool->end - pool->start;
+    if (blockSize < INIT_BLOCK_SIZE)
+      blockSize = INIT_BLOCK_SIZE;
+    else
+      blockSize *= 2;
+    tem = pool->mem->malloc_fcn(offsetof(BLOCK, s) + blockSize * sizeof(XML_Char));
+    if (!tem)
+      return 0;
+    tem->size = blockSize;
+    tem->next = pool->blocks;
+    pool->blocks = tem;
+    if (pool->ptr != pool->start)
+      memcpy(tem->s, pool->start, (pool->ptr - pool->start) * sizeof(XML_Char));
+    pool->ptr = tem->s + (pool->ptr - pool->start);
+    pool->start = tem->s;
+    pool->end = tem->s + blockSize;
+  }
+  return 1;
+}
+
+static int
+nextScaffoldPart(XML_Parser parser)
+{
+  CONTENT_SCAFFOLD * me;
+  int next;
+
+  if (! dtd.scaffIndex) {
+    dtd.scaffIndex = MALLOC(groupSize * sizeof(int));
+    if (! dtd.scaffIndex)
+      return -1;
+    dtd.scaffIndex[0] = 0;
+  }
+
+  if (dtd.scaffCount >= dtd.scaffSize) {
+    if (dtd.scaffold) {
+      dtd.scaffSize *= 2;
+      dtd.scaffold = (CONTENT_SCAFFOLD *) REALLOC(dtd.scaffold,
+					      dtd.scaffSize * sizeof(CONTENT_SCAFFOLD));
+    }
+    else {
+      dtd.scaffSize = 32;
+      dtd.scaffold = (CONTENT_SCAFFOLD *) MALLOC(dtd.scaffSize * sizeof(CONTENT_SCAFFOLD));
+    }
+    if (! dtd.scaffold)
+      return -1;
+  }
+  next = dtd.scaffCount++;
+  me = &dtd.scaffold[next];
+  if (dtd.scaffLevel) { 
+    CONTENT_SCAFFOLD *parent = &dtd.scaffold[dtd.scaffIndex[dtd.scaffLevel - 1]];
+    if (parent->lastchild) {
+      dtd.scaffold[parent->lastchild].nextsib = next;
+    }
+    if (! parent->childcnt)
+      parent->firstchild = next;
+    parent->lastchild = next;
+    parent->childcnt++;
+  }
+  me->firstchild = me->lastchild = me->childcnt = me->nextsib = 0;
+  return next;
+}  /* End nextScaffoldPart */
+
+static void
+build_node (XML_Parser parser,
+	    int src_node,
+	    XML_Content *dest,
+	    XML_Content **contpos,
+	    char **strpos)
+{
+  dest->type = dtd.scaffold[src_node].type;
+  dest->quant = dtd.scaffold[src_node].quant;
+  if (dest->type == XML_CTYPE_NAME) {
+    const char *src;
+    dest->name = *strpos;
+    src = dtd.scaffold[src_node].name;
+    for (;;) {
+      *(*strpos)++ = *src;
+      if (! *src)
+	break;
+      src++;
+    }
+    dest->numchildren = 0;
+    dest->children = 0;
+  }
+  else {
+    unsigned int i;
+    int cn;
+    dest->numchildren = dtd.scaffold[src_node].childcnt;
+    dest->children = *contpos;
+    *contpos += dest->numchildren;
+    for (i = 0, cn = dtd.scaffold[src_node].firstchild;
+	 i < dest->numchildren;
+	 i++, cn = dtd.scaffold[cn].nextsib) {
+      build_node(parser, cn, &(dest->children[i]), contpos, strpos);
+    }
+    dest->name = 0;
+  }
+}  /* End build_node */
+
+static XML_Content *
+build_model (XML_Parser parser)
+{
+  XML_Content *ret;
+  XML_Content *cpos;
+  char * str;
+  int allocsize = dtd.scaffCount * sizeof(XML_Content) + dtd.contentStringLen;
+  
+  ret = MALLOC(allocsize);
+  if (! ret)
+    return 0;
+
+  str =  (char *) (&ret[dtd.scaffCount]);
+  cpos = &ret[1];
+
+  build_node(parser, 0, ret, &cpos, &str);
+  return ret;
+}  /* End build_model */
+
+static ELEMENT_TYPE *
+getElementType(XML_Parser parser,
+	       const ENCODING *enc,
+	       const char *ptr,
+	       const char *end)
+{
+  const XML_Char *name = poolStoreString(&dtd.pool, enc, ptr, end);
+  ELEMENT_TYPE *ret;
+
+  if (! name)
+    return 0;
+  ret = (ELEMENT_TYPE *) lookup(&dtd.elementTypes, name, sizeof(ELEMENT_TYPE));
+  if (! ret)
+    return 0;
+  if (ret->name != name)
+    poolDiscard(&dtd.pool);
+  else {
+    poolFinish(&dtd.pool);
+    if (!setElementTypePrefix(parser, ret))
+      return 0;
+  }
+  return ret;
+}  /* End getElementType */
diff --git a/xml/expat/lib/xmlrole.c b/xml/expat/lib/xmlrole.c
new file mode 100644
index 000000000..ac130392e
--- /dev/null
+++ b/xml/expat/lib/xmlrole.c
@@ -0,0 +1,1275 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+static char RCSId[]
+  = "$Header: /home/cvs/apr-util/xml/expat/lib/xmlrole.c,v 1.1 2001/02/28 14:41:26 gstein Exp $";
+
+#ifdef COMPILED_FROM_DSP
+#  include "winconfig.h"
+#else
+#  include <config.h>
+#endif /* ndef COMPILED_FROM_DSP */
+
+#include "xmlrole.h"
+#include "ascii.h"
+
+/* Doesn't check:
+
+ that ,| are not mixed in a model group
+ content of literals
+
+*/
+
+static const char KW_ANY[] = { ASCII_A, ASCII_N, ASCII_Y, '\0' };
+static const char KW_ATTLIST[] = { ASCII_A, ASCII_T, ASCII_T, ASCII_L, ASCII_I, ASCII_S, ASCII_T, '\0' };
+static const char KW_CDATA[] = { ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_DOCTYPE[] = { ASCII_D, ASCII_O, ASCII_C, ASCII_T, ASCII_Y, ASCII_P, ASCII_E, '\0' };
+static const char KW_ELEMENT[] = { ASCII_E, ASCII_L, ASCII_E, ASCII_M, ASCII_E, ASCII_N, ASCII_T, '\0' };
+static const char KW_EMPTY[] = { ASCII_E, ASCII_M, ASCII_P, ASCII_T, ASCII_Y, '\0' };
+static const char KW_ENTITIES[] = { ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_I, ASCII_E, ASCII_S, '\0' };
+static const char KW_ENTITY[] = { ASCII_E, ASCII_N, ASCII_T, ASCII_I, ASCII_T, ASCII_Y, '\0' };
+static const char KW_FIXED[] = { ASCII_F, ASCII_I, ASCII_X, ASCII_E, ASCII_D, '\0' };
+static const char KW_ID[] = { ASCII_I, ASCII_D, '\0' };
+static const char KW_IDREF[] = { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, '\0' };
+static const char KW_IDREFS[] = { ASCII_I, ASCII_D, ASCII_R, ASCII_E, ASCII_F, ASCII_S, '\0' };
+static const char KW_IGNORE[] = { ASCII_I, ASCII_G, ASCII_N, ASCII_O, ASCII_R, ASCII_E, '\0' };
+static const char KW_IMPLIED[] = { ASCII_I, ASCII_M, ASCII_P, ASCII_L, ASCII_I, ASCII_E, ASCII_D, '\0' };
+static const char KW_INCLUDE[] = { ASCII_I, ASCII_N, ASCII_C, ASCII_L, ASCII_U, ASCII_D, ASCII_E, '\0' };
+static const char KW_NDATA[] = { ASCII_N, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_NMTOKEN[] = { ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, '\0' };
+static const char KW_NMTOKENS[] = { ASCII_N, ASCII_M, ASCII_T, ASCII_O, ASCII_K, ASCII_E, ASCII_N, ASCII_S, '\0' };
+static const char KW_NOTATION[] = { ASCII_N, ASCII_O, ASCII_T, ASCII_A, ASCII_T, ASCII_I, ASCII_O, ASCII_N, '\0' };
+static const char KW_PCDATA[] = { ASCII_P, ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, '\0' };
+static const char KW_PUBLIC[] = { ASCII_P, ASCII_U, ASCII_B, ASCII_L, ASCII_I, ASCII_C, '\0' };
+static const char KW_REQUIRED[] = { ASCII_R, ASCII_E, ASCII_Q, ASCII_U, ASCII_I, ASCII_R, ASCII_E, ASCII_D, '\0' };
+static const char KW_SYSTEM[] = { ASCII_S, ASCII_Y, ASCII_S, ASCII_T, ASCII_E, ASCII_M, '\0' };
+
+#ifndef MIN_BYTES_PER_CHAR
+#define MIN_BYTES_PER_CHAR(enc) ((enc)->minBytesPerChar)
+#endif
+
+#ifdef XML_DTD
+#define setTopLevel(state) \
+  ((state)->handler = ((state)->documentEntity \
+                       ? internalSubset \
+                       : externalSubset1))
+#else /* not XML_DTD */
+#define setTopLevel(state) ((state)->handler = internalSubset)
+#endif /* not XML_DTD */
+
+typedef int PROLOG_HANDLER(PROLOG_STATE *state,
+			   int tok,
+			   const char *ptr,
+			   const char *end,
+			   const ENCODING *enc);
+
+static PROLOG_HANDLER
+  prolog0, prolog1, prolog2,
+  doctype0, doctype1, doctype2, doctype3, doctype4, doctype5,
+  internalSubset,
+  entity0, entity1, entity2, entity3, entity4, entity5, entity6,
+  entity7, entity8, entity9,
+  notation0, notation1, notation2, notation3, notation4,
+  attlist0, attlist1, attlist2, attlist3, attlist4, attlist5, attlist6,
+  attlist7, attlist8, attlist9,
+  element0, element1, element2, element3, element4, element5, element6,
+  element7,
+#ifdef XML_DTD
+  externalSubset0, externalSubset1,
+  condSect0, condSect1, condSect2,
+#endif /* XML_DTD */
+  declClose,
+  error;
+
+static
+int common(PROLOG_STATE *state, int tok);
+
+static
+int prolog0(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    state->handler = prolog1;
+    return XML_ROLE_NONE;
+  case XML_TOK_XML_DECL:
+    state->handler = prolog1;
+    return XML_ROLE_XML_DECL;
+  case XML_TOK_PI:
+    state->handler = prolog1;
+    return XML_ROLE_NONE;
+  case XML_TOK_COMMENT:
+    state->handler = prolog1;
+  case XML_TOK_BOM:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_OPEN:
+    if (!XmlNameMatchesAscii(enc,
+			     ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			     end,
+			     KW_DOCTYPE))
+      break;
+    state->handler = doctype0;
+    return XML_ROLE_NONE;
+  case XML_TOK_INSTANCE_START:
+    state->handler = error;
+    return XML_ROLE_INSTANCE_START;
+  }
+  return common(state, tok);
+}
+
+static
+int prolog1(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_PI:
+  case XML_TOK_COMMENT:
+  case XML_TOK_BOM:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_OPEN:
+    if (!XmlNameMatchesAscii(enc,
+			     ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			     end,
+			     KW_DOCTYPE))
+      break;
+    state->handler = doctype0;
+    return XML_ROLE_NONE;
+  case XML_TOK_INSTANCE_START:
+    state->handler = error;
+    return XML_ROLE_INSTANCE_START;
+  }
+  return common(state, tok);
+}
+
+static
+int prolog2(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_PI:
+  case XML_TOK_COMMENT:
+    return XML_ROLE_NONE;
+  case XML_TOK_INSTANCE_START:
+    state->handler = error;
+    return XML_ROLE_INSTANCE_START;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype0(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = doctype1;
+    return XML_ROLE_DOCTYPE_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype1(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_BRACKET:
+    state->handler = internalSubset;
+    return XML_ROLE_DOCTYPE_INTERNAL_SUBSET;
+  case XML_TOK_DECL_CLOSE:
+    state->handler = prolog2;
+    return XML_ROLE_DOCTYPE_CLOSE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+      state->handler = doctype3;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+      state->handler = doctype2;
+      return XML_ROLE_NONE;
+    }
+    break;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype2(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = doctype3;
+    return XML_ROLE_DOCTYPE_PUBLIC_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype3(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = doctype4;
+    return XML_ROLE_DOCTYPE_SYSTEM_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype4(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_BRACKET:
+    state->handler = internalSubset;
+    return XML_ROLE_DOCTYPE_INTERNAL_SUBSET;
+  case XML_TOK_DECL_CLOSE:
+    state->handler = prolog2;
+    return XML_ROLE_DOCTYPE_CLOSE;
+  }
+  return common(state, tok);
+}
+
+static
+int doctype5(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_CLOSE:
+    state->handler = prolog2;
+    return XML_ROLE_DOCTYPE_CLOSE;
+  }
+  return common(state, tok);
+}
+
+static
+int internalSubset(PROLOG_STATE *state,
+		   int tok,
+		   const char *ptr,
+		   const char *end,
+		   const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_OPEN:
+    if (XmlNameMatchesAscii(enc,
+			    ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_ENTITY)) {
+      state->handler = entity0;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc,
+			    ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_ATTLIST)) {
+      state->handler = attlist0;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc,
+			    ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_ELEMENT)) {
+      state->handler = element0;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc,
+			    ptr + 2 * MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_NOTATION)) {
+      state->handler = notation0;
+      return XML_ROLE_NONE;
+    }
+    break;
+  case XML_TOK_PI:
+  case XML_TOK_COMMENT:
+    return XML_ROLE_NONE;
+  case XML_TOK_PARAM_ENTITY_REF:
+    return XML_ROLE_PARAM_ENTITY_REF;
+  case XML_TOK_CLOSE_BRACKET:
+    state->handler = doctype5;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+#ifdef XML_DTD
+
+static
+int externalSubset0(PROLOG_STATE *state,
+		    int tok,
+		    const char *ptr,
+		    const char *end,
+		    const ENCODING *enc)
+{
+  state->handler = externalSubset1;
+  if (tok == XML_TOK_XML_DECL)
+    return XML_ROLE_TEXT_DECL;
+  return externalSubset1(state, tok, ptr, end, enc);
+}
+
+static
+int externalSubset1(PROLOG_STATE *state,
+		    int tok,
+		    const char *ptr,
+		    const char *end,
+		    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_COND_SECT_OPEN:
+    state->handler = condSect0;
+    return XML_ROLE_NONE;
+  case XML_TOK_COND_SECT_CLOSE:
+    if (state->includeLevel == 0)
+      break;
+    state->includeLevel -= 1;
+    return XML_ROLE_NONE;
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_BRACKET:
+    break;
+  case XML_TOK_NONE:
+    if (state->includeLevel)
+      break;
+    return XML_ROLE_NONE;
+  default:
+    return internalSubset(state, tok, ptr, end, enc);
+  }
+  return common(state, tok);
+}
+
+#endif /* XML_DTD */
+
+static
+int entity0(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_PERCENT:
+    state->handler = entity1;
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    state->handler = entity2;
+    return XML_ROLE_GENERAL_ENTITY_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int entity1(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    state->handler = entity7;
+    return XML_ROLE_PARAM_ENTITY_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int entity2(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+      state->handler = entity4;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+      state->handler = entity3;
+      return XML_ROLE_NONE;
+    }
+    break;
+  case XML_TOK_LITERAL:
+    state->handler = declClose;
+    return XML_ROLE_ENTITY_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int entity3(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = entity4;
+    return XML_ROLE_ENTITY_PUBLIC_ID;
+  }
+  return common(state, tok);
+}
+
+
+static
+int entity4(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = entity5;
+    return XML_ROLE_ENTITY_SYSTEM_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int entity5(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_CLOSE:
+    setTopLevel(state);
+    return XML_ROLE_ENTITY_COMPLETE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_NDATA)) {
+      state->handler = entity6;
+      return XML_ROLE_NONE;
+    }
+    break;
+  }
+  return common(state, tok);
+}
+
+static
+int entity6(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    state->handler = declClose;
+    return XML_ROLE_ENTITY_NOTATION_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int entity7(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+      state->handler = entity9;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+      state->handler = entity8;
+      return XML_ROLE_NONE;
+    }
+    break;
+  case XML_TOK_LITERAL:
+    state->handler = declClose;
+    return XML_ROLE_ENTITY_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int entity8(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = entity9;
+    return XML_ROLE_ENTITY_PUBLIC_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int entity9(PROLOG_STATE *state,
+	    int tok,
+	    const char *ptr,
+	    const char *end,
+	    const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = declClose;
+    return XML_ROLE_ENTITY_SYSTEM_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int notation0(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    state->handler = notation1;
+    return XML_ROLE_NOTATION_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int notation1(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_SYSTEM)) {
+      state->handler = notation3;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_PUBLIC)) {
+      state->handler = notation2;
+      return XML_ROLE_NONE;
+    }
+    break;
+  }
+  return common(state, tok);
+}
+
+static
+int notation2(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = notation4;
+    return XML_ROLE_NOTATION_PUBLIC_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int notation3(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = declClose;
+    return XML_ROLE_NOTATION_SYSTEM_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int notation4(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = declClose;
+    return XML_ROLE_NOTATION_SYSTEM_ID;
+  case XML_TOK_DECL_CLOSE:
+    setTopLevel(state);
+    return XML_ROLE_NOTATION_NO_SYSTEM_ID;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist0(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = attlist1;
+    return XML_ROLE_ATTLIST_ELEMENT_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist1(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_CLOSE:
+    setTopLevel(state);
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = attlist2;
+    return XML_ROLE_ATTRIBUTE_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist2(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    {
+      static const char *types[] = {
+	KW_CDATA,
+        KW_ID,
+        KW_IDREF,
+        KW_IDREFS,
+        KW_ENTITY,
+        KW_ENTITIES,
+        KW_NMTOKEN,
+        KW_NMTOKENS,
+      };
+      int i;
+      for (i = 0; i < (int)(sizeof(types)/sizeof(types[0])); i++)
+	if (XmlNameMatchesAscii(enc, ptr, end, types[i])) {
+	  state->handler = attlist8;
+	  return XML_ROLE_ATTRIBUTE_TYPE_CDATA + i;
+	}
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_NOTATION)) {
+      state->handler = attlist5;
+      return XML_ROLE_NONE;
+    }
+    break;
+  case XML_TOK_OPEN_PAREN:
+    state->handler = attlist3;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist3(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NMTOKEN:
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = attlist4;
+    return XML_ROLE_ATTRIBUTE_ENUM_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist4(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_PAREN:
+    state->handler = attlist8;
+    return XML_ROLE_NONE;
+  case XML_TOK_OR:
+    state->handler = attlist3;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist5(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_PAREN:
+    state->handler = attlist6;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+
+static
+int attlist6(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    state->handler = attlist7;
+    return XML_ROLE_ATTRIBUTE_NOTATION_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist7(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_PAREN:
+    state->handler = attlist8;
+    return XML_ROLE_NONE;
+  case XML_TOK_OR:
+    state->handler = attlist6;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+/* default value */
+static
+int attlist8(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_POUND_NAME:
+    if (XmlNameMatchesAscii(enc,
+			    ptr + MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_IMPLIED)) {
+      state->handler = attlist1;
+      return XML_ROLE_IMPLIED_ATTRIBUTE_VALUE;
+    }
+    if (XmlNameMatchesAscii(enc,
+			    ptr + MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_REQUIRED)) {
+      state->handler = attlist1;
+      return XML_ROLE_REQUIRED_ATTRIBUTE_VALUE;
+    }
+    if (XmlNameMatchesAscii(enc,
+			    ptr + MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_FIXED)) {
+      state->handler = attlist9;
+      return XML_ROLE_NONE;
+    }
+    break;
+  case XML_TOK_LITERAL:
+    state->handler = attlist1;
+    return XML_ROLE_DEFAULT_ATTRIBUTE_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int attlist9(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_LITERAL:
+    state->handler = attlist1;
+    return XML_ROLE_FIXED_ATTRIBUTE_VALUE;
+  }
+  return common(state, tok);
+}
+
+static
+int element0(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = element1;
+    return XML_ROLE_ELEMENT_NAME;
+  }
+  return common(state, tok);
+}
+
+static
+int element1(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_EMPTY)) {
+      state->handler = declClose;
+      return XML_ROLE_CONTENT_EMPTY;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_ANY)) {
+      state->handler = declClose;
+      return XML_ROLE_CONTENT_ANY;
+    }
+    break;
+  case XML_TOK_OPEN_PAREN:
+    state->handler = element2;
+    state->level = 1;
+    return XML_ROLE_GROUP_OPEN;
+  }
+  return common(state, tok);
+}
+
+static
+int element2(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_POUND_NAME:
+    if (XmlNameMatchesAscii(enc,
+			    ptr + MIN_BYTES_PER_CHAR(enc),
+			    end,
+			    KW_PCDATA)) {
+      state->handler = element3;
+      return XML_ROLE_CONTENT_PCDATA;
+    }
+    break;
+  case XML_TOK_OPEN_PAREN:
+    state->level = 2;
+    state->handler = element6;
+    return XML_ROLE_GROUP_OPEN;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT;
+  case XML_TOK_NAME_QUESTION:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_OPT;
+  case XML_TOK_NAME_ASTERISK:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_REP;
+  case XML_TOK_NAME_PLUS:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_PLUS;
+  }
+  return common(state, tok);
+}
+
+static
+int element3(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_PAREN:
+    state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE;
+  case XML_TOK_CLOSE_PAREN_ASTERISK:
+    state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE_REP;
+  case XML_TOK_OR:
+    state->handler = element4;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+static
+int element4(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = element5;
+    return XML_ROLE_CONTENT_ELEMENT;
+  }
+  return common(state, tok);
+}
+
+static
+int element5(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_PAREN_ASTERISK:
+    state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE_REP;
+  case XML_TOK_OR:
+    state->handler = element4;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+static
+int element6(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_PAREN:
+    state->level += 1;
+    return XML_ROLE_GROUP_OPEN;
+  case XML_TOK_NAME:
+  case XML_TOK_PREFIXED_NAME:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT;
+  case XML_TOK_NAME_QUESTION:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_OPT;
+  case XML_TOK_NAME_ASTERISK:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_REP;
+  case XML_TOK_NAME_PLUS:
+    state->handler = element7;
+    return XML_ROLE_CONTENT_ELEMENT_PLUS;
+  }
+  return common(state, tok);
+}
+
+static
+int element7(PROLOG_STATE *state,
+	     int tok,
+	     const char *ptr,
+	     const char *end,
+	     const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_CLOSE_PAREN:
+    state->level -= 1;
+    if (state->level == 0)
+      state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE;
+  case XML_TOK_CLOSE_PAREN_ASTERISK:
+    state->level -= 1;
+    if (state->level == 0)
+      state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE_REP;
+  case XML_TOK_CLOSE_PAREN_QUESTION:
+    state->level -= 1;
+    if (state->level == 0)
+      state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE_OPT;
+  case XML_TOK_CLOSE_PAREN_PLUS:
+    state->level -= 1;
+    if (state->level == 0)
+      state->handler = declClose;
+    return XML_ROLE_GROUP_CLOSE_PLUS;
+  case XML_TOK_COMMA:
+    state->handler = element6;
+    return XML_ROLE_GROUP_SEQUENCE;
+  case XML_TOK_OR:
+    state->handler = element6;
+    return XML_ROLE_GROUP_CHOICE;
+  }
+  return common(state, tok);
+}
+
+#ifdef XML_DTD
+
+static
+int condSect0(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_NAME:
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_INCLUDE)) {
+      state->handler = condSect1;
+      return XML_ROLE_NONE;
+    }
+    if (XmlNameMatchesAscii(enc, ptr, end, KW_IGNORE)) {
+      state->handler = condSect2;
+      return XML_ROLE_NONE;
+    }
+    break;
+  }
+  return common(state, tok);
+}
+
+static
+int condSect1(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_BRACKET:
+    state->handler = externalSubset1;
+    state->includeLevel += 1;
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+static
+int condSect2(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_OPEN_BRACKET:
+    state->handler = externalSubset1;
+    return XML_ROLE_IGNORE_SECT;
+  }
+  return common(state, tok);
+}
+
+#endif /* XML_DTD */
+
+static
+int declClose(PROLOG_STATE *state,
+	      int tok,
+	      const char *ptr,
+	      const char *end,
+	      const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_PROLOG_S:
+    return XML_ROLE_NONE;
+  case XML_TOK_DECL_CLOSE:
+    setTopLevel(state);
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+
+#if 0
+
+static
+int ignore(PROLOG_STATE *state,
+	   int tok,
+	   const char *ptr,
+	   const char *end,
+	   const ENCODING *enc)
+{
+  switch (tok) {
+  case XML_TOK_DECL_CLOSE:
+    state->handler = internalSubset;
+    return 0;
+  default:
+    return XML_ROLE_NONE;
+  }
+  return common(state, tok);
+}
+#endif
+
+static
+int error(PROLOG_STATE *state,
+	  int tok,
+	  const char *ptr,
+	  const char *end,
+	  const ENCODING *enc)
+{
+  return XML_ROLE_NONE;
+}
+
+static
+int common(PROLOG_STATE *state, int tok)
+{
+#ifdef XML_DTD
+  if (!state->documentEntity && tok == XML_TOK_PARAM_ENTITY_REF)
+    return XML_ROLE_INNER_PARAM_ENTITY_REF;
+#endif
+  state->handler = error;
+  return XML_ROLE_ERROR;
+}
+
+void XmlPrologStateInit(PROLOG_STATE *state)
+{
+  state->handler = prolog0;
+#ifdef XML_DTD
+  state->documentEntity = 1;
+  state->includeLevel = 0;
+#endif /* XML_DTD */
+}
+
+#ifdef XML_DTD
+
+void XmlPrologStateInitExternalEntity(PROLOG_STATE *state)
+{
+  state->handler = externalSubset0;
+  state->documentEntity = 0;
+  state->includeLevel = 0;
+}
+
+#endif /* XML_DTD */
diff --git a/xml/expat/lib/xmlrole.h b/xml/expat/lib/xmlrole.h
new file mode 100644
index 000000000..db3ebc848
--- /dev/null
+++ b/xml/expat/lib/xmlrole.h
@@ -0,0 +1,100 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+#ifndef XmlRole_INCLUDED
+#define XmlRole_INCLUDED 1
+
+#include "xmltok.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+enum {
+  XML_ROLE_ERROR = -1,
+  XML_ROLE_NONE = 0,
+  XML_ROLE_XML_DECL,
+  XML_ROLE_INSTANCE_START,
+  XML_ROLE_DOCTYPE_NAME,
+  XML_ROLE_DOCTYPE_SYSTEM_ID,
+  XML_ROLE_DOCTYPE_PUBLIC_ID,
+  XML_ROLE_DOCTYPE_INTERNAL_SUBSET,
+  XML_ROLE_DOCTYPE_CLOSE,
+  XML_ROLE_GENERAL_ENTITY_NAME,
+  XML_ROLE_PARAM_ENTITY_NAME,
+  XML_ROLE_ENTITY_VALUE,
+  XML_ROLE_ENTITY_SYSTEM_ID,
+  XML_ROLE_ENTITY_PUBLIC_ID,
+  XML_ROLE_ENTITY_COMPLETE,
+  XML_ROLE_ENTITY_NOTATION_NAME,
+  XML_ROLE_NOTATION_NAME,
+  XML_ROLE_NOTATION_SYSTEM_ID,
+  XML_ROLE_NOTATION_NO_SYSTEM_ID,
+  XML_ROLE_NOTATION_PUBLIC_ID,
+  XML_ROLE_ATTRIBUTE_NAME,
+  XML_ROLE_ATTRIBUTE_TYPE_CDATA,
+  XML_ROLE_ATTRIBUTE_TYPE_ID,
+  XML_ROLE_ATTRIBUTE_TYPE_IDREF,
+  XML_ROLE_ATTRIBUTE_TYPE_IDREFS,
+  XML_ROLE_ATTRIBUTE_TYPE_ENTITY,
+  XML_ROLE_ATTRIBUTE_TYPE_ENTITIES,
+  XML_ROLE_ATTRIBUTE_TYPE_NMTOKEN,
+  XML_ROLE_ATTRIBUTE_TYPE_NMTOKENS,
+  XML_ROLE_ATTRIBUTE_ENUM_VALUE,
+  XML_ROLE_ATTRIBUTE_NOTATION_VALUE,
+  XML_ROLE_ATTLIST_ELEMENT_NAME,
+  XML_ROLE_IMPLIED_ATTRIBUTE_VALUE,
+  XML_ROLE_REQUIRED_ATTRIBUTE_VALUE,
+  XML_ROLE_DEFAULT_ATTRIBUTE_VALUE,
+  XML_ROLE_FIXED_ATTRIBUTE_VALUE,
+  XML_ROLE_ELEMENT_NAME,
+  XML_ROLE_CONTENT_ANY,
+  XML_ROLE_CONTENT_EMPTY,
+  XML_ROLE_CONTENT_PCDATA,
+  XML_ROLE_GROUP_OPEN,
+  XML_ROLE_GROUP_CLOSE,
+  XML_ROLE_GROUP_CLOSE_REP,
+  XML_ROLE_GROUP_CLOSE_OPT,
+  XML_ROLE_GROUP_CLOSE_PLUS,
+  XML_ROLE_GROUP_CHOICE,
+  XML_ROLE_GROUP_SEQUENCE,
+  XML_ROLE_CONTENT_ELEMENT,
+  XML_ROLE_CONTENT_ELEMENT_REP,
+  XML_ROLE_CONTENT_ELEMENT_OPT,
+  XML_ROLE_CONTENT_ELEMENT_PLUS,
+#ifdef XML_DTD
+  XML_ROLE_TEXT_DECL,
+  XML_ROLE_IGNORE_SECT,
+  XML_ROLE_INNER_PARAM_ENTITY_REF,
+#endif /* XML_DTD */
+  XML_ROLE_PARAM_ENTITY_REF
+};
+
+typedef struct prolog_state {
+  int (*handler)(struct prolog_state *state,
+	         int tok,
+		 const char *ptr,
+		 const char *end,
+		 const ENCODING *enc);
+  unsigned level;
+#ifdef XML_DTD
+  unsigned includeLevel;
+  int documentEntity;
+#endif /* XML_DTD */
+} PROLOG_STATE;
+
+void XmlPrologStateInit(PROLOG_STATE *);
+#ifdef XML_DTD
+void XmlPrologStateInitExternalEntity(PROLOG_STATE *);
+#endif /* XML_DTD */
+
+#define XmlTokenRole(state, tok, ptr, end, enc) \
+ (((state)->handler)(state, tok, ptr, end, enc))
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not XmlRole_INCLUDED */
diff --git a/xml/expat/lib/xmltok.c b/xml/expat/lib/xmltok.c
new file mode 100644
index 000000000..e3fe3f788
--- /dev/null
+++ b/xml/expat/lib/xmltok.c
@@ -0,0 +1,1569 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+static char RCSId[]
+  = "$Header: /home/cvs/apr-util/xml/expat/lib/xmltok.c,v 1.1 2001/02/28 14:41:26 gstein Exp $";
+
+#ifdef COMPILED_FROM_DSP
+#  include "winconfig.h"
+#else
+#  include <config.h>
+#endif /* ndef COMPILED_FROM_DSP */
+
+#include "xmltok.h"
+#include "nametab.h"
+
+#ifdef XML_DTD
+#define IGNORE_SECTION_TOK_VTABLE , PREFIX(ignoreSectionTok)
+#else
+#define IGNORE_SECTION_TOK_VTABLE /* as nothing */
+#endif
+
+#define VTABLE1 \
+  { PREFIX(prologTok), PREFIX(contentTok), \
+    PREFIX(cdataSectionTok) IGNORE_SECTION_TOK_VTABLE }, \
+  { PREFIX(attributeValueTok), PREFIX(entityValueTok) }, \
+  PREFIX(sameName), \
+  PREFIX(nameMatchesAscii), \
+  PREFIX(nameLength), \
+  PREFIX(skipS), \
+  PREFIX(getAtts), \
+  PREFIX(charRefNumber), \
+  PREFIX(predefinedEntityName), \
+  PREFIX(updatePosition), \
+  PREFIX(isPublicId)
+
+#define VTABLE VTABLE1, PREFIX(toUtf8), PREFIX(toUtf16)
+
+#define UCS2_GET_NAMING(pages, hi, lo) \
+   (namingBitmap[(pages[hi] << 3) + ((lo) >> 5)] & (1 << ((lo) & 0x1F)))
+
+/* A 2 byte UTF-8 representation splits the characters 11 bits
+between the bottom 5 and 6 bits of the bytes.
+We need 8 bits to index into pages, 3 bits to add to that index and
+5 bits to generate the mask. */
+#define UTF8_GET_NAMING2(pages, byte) \
+    (namingBitmap[((pages)[(((byte)[0]) >> 2) & 7] << 3) \
+                      + ((((byte)[0]) & 3) << 1) \
+                      + ((((byte)[1]) >> 5) & 1)] \
+         & (1 << (((byte)[1]) & 0x1F)))
+
+/* A 3 byte UTF-8 representation splits the characters 16 bits
+between the bottom 4, 6 and 6 bits of the bytes.
+We need 8 bits to index into pages, 3 bits to add to that index and
+5 bits to generate the mask. */
+#define UTF8_GET_NAMING3(pages, byte) \
+  (namingBitmap[((pages)[((((byte)[0]) & 0xF) << 4) \
+                             + ((((byte)[1]) >> 2) & 0xF)] \
+		       << 3) \
+                      + ((((byte)[1]) & 3) << 1) \
+                      + ((((byte)[2]) >> 5) & 1)] \
+         & (1 << (((byte)[2]) & 0x1F)))
+
+#define UTF8_GET_NAMING(pages, p, n) \
+  ((n) == 2 \
+  ? UTF8_GET_NAMING2(pages, (const unsigned char *)(p)) \
+  : ((n) == 3 \
+     ? UTF8_GET_NAMING3(pages, (const unsigned char *)(p)) \
+     : 0))
+
+#define UTF8_INVALID3(p) \
+  ((*p) == 0xED \
+  ? (((p)[1] & 0x20) != 0) \
+  : ((*p) == 0xEF \
+     ? ((p)[1] == 0xBF && ((p)[2] == 0xBF || (p)[2] == 0xBE)) \
+     : 0))
+
+#define UTF8_INVALID4(p) ((*p) == 0xF4 && ((p)[1] & 0x30) != 0)
+
+static
+int isNever(const ENCODING *enc, const char *p)
+{
+  return 0;
+}
+
+static
+int utf8_isName2(const ENCODING *enc, const char *p)
+{
+  return UTF8_GET_NAMING2(namePages, (const unsigned char *)p);
+}
+
+static
+int utf8_isName3(const ENCODING *enc, const char *p)
+{
+  return UTF8_GET_NAMING3(namePages, (const unsigned char *)p);
+}
+
+#define utf8_isName4 isNever
+
+static
+int utf8_isNmstrt2(const ENCODING *enc, const char *p)
+{
+  return UTF8_GET_NAMING2(nmstrtPages, (const unsigned char *)p);
+}
+
+static
+int utf8_isNmstrt3(const ENCODING *enc, const char *p)
+{
+  return UTF8_GET_NAMING3(nmstrtPages, (const unsigned char *)p);
+}
+
+#define utf8_isNmstrt4 isNever
+
+#define utf8_isInvalid2 isNever
+
+static
+int utf8_isInvalid3(const ENCODING *enc, const char *p)
+{
+  return UTF8_INVALID3((const unsigned char *)p);
+}
+
+static
+int utf8_isInvalid4(const ENCODING *enc, const char *p)
+{
+  return UTF8_INVALID4((const unsigned char *)p);
+}
+
+struct normal_encoding {
+  ENCODING enc;
+  unsigned char type[256];
+#ifdef XML_MIN_SIZE
+  int (*byteType)(const ENCODING *, const char *);
+  int (*isNameMin)(const ENCODING *, const char *);
+  int (*isNmstrtMin)(const ENCODING *, const char *);
+  int (*byteToAscii)(const ENCODING *, const char *);
+  int (*charMatches)(const ENCODING *, const char *, int);
+#endif /* XML_MIN_SIZE */
+  int (*isName2)(const ENCODING *, const char *);
+  int (*isName3)(const ENCODING *, const char *);
+  int (*isName4)(const ENCODING *, const char *);
+  int (*isNmstrt2)(const ENCODING *, const char *);
+  int (*isNmstrt3)(const ENCODING *, const char *);
+  int (*isNmstrt4)(const ENCODING *, const char *);
+  int (*isInvalid2)(const ENCODING *, const char *);
+  int (*isInvalid3)(const ENCODING *, const char *);
+  int (*isInvalid4)(const ENCODING *, const char *);
+};
+
+#ifdef XML_MIN_SIZE
+
+#define STANDARD_VTABLE(E) \
+ E ## byteType, \
+ E ## isNameMin, \
+ E ## isNmstrtMin, \
+ E ## byteToAscii, \
+ E ## charMatches,
+
+#else
+
+#define STANDARD_VTABLE(E) /* as nothing */
+
+#endif
+
+#define NORMAL_VTABLE(E) \
+ E ## isName2, \
+ E ## isName3, \
+ E ## isName4, \
+ E ## isNmstrt2, \
+ E ## isNmstrt3, \
+ E ## isNmstrt4, \
+ E ## isInvalid2, \
+ E ## isInvalid3, \
+ E ## isInvalid4
+
+static int checkCharRefNumber(int);
+
+#include "xmltok_impl.h"
+#include "ascii.h"
+
+#ifdef XML_MIN_SIZE
+#define sb_isNameMin isNever
+#define sb_isNmstrtMin isNever
+#endif
+
+#ifdef XML_MIN_SIZE
+#define MINBPC(enc) ((enc)->minBytesPerChar)
+#else
+/* minimum bytes per character */
+#define MINBPC(enc) 1
+#endif
+
+#define SB_BYTE_TYPE(enc, p) \
+  (((struct normal_encoding *)(enc))->type[(unsigned char)*(p)])
+
+#ifdef XML_MIN_SIZE
+static
+int sb_byteType(const ENCODING *enc, const char *p)
+{
+  return SB_BYTE_TYPE(enc, p);
+}
+#define BYTE_TYPE(enc, p) \
+ (((const struct normal_encoding *)(enc))->byteType(enc, p))
+#else
+#define BYTE_TYPE(enc, p) SB_BYTE_TYPE(enc, p)
+#endif
+
+#ifdef XML_MIN_SIZE
+#define BYTE_TO_ASCII(enc, p) \
+ (((const struct normal_encoding *)(enc))->byteToAscii(enc, p))
+static
+int sb_byteToAscii(const ENCODING *enc, const char *p)
+{
+  return *p;
+}
+#else
+#define BYTE_TO_ASCII(enc, p) (*(p))
+#endif
+
+#define IS_NAME_CHAR(enc, p, n) \
+ (((const struct normal_encoding *)(enc))->isName ## n(enc, p))
+#define IS_NMSTRT_CHAR(enc, p, n) \
+ (((const struct normal_encoding *)(enc))->isNmstrt ## n(enc, p))
+#define IS_INVALID_CHAR(enc, p, n) \
+ (((const struct normal_encoding *)(enc))->isInvalid ## n(enc, p))
+
+#ifdef XML_MIN_SIZE
+#define IS_NAME_CHAR_MINBPC(enc, p) \
+ (((const struct normal_encoding *)(enc))->isNameMin(enc, p))
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) \
+ (((const struct normal_encoding *)(enc))->isNmstrtMin(enc, p))
+#else
+#define IS_NAME_CHAR_MINBPC(enc, p) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) (0)
+#endif
+
+#ifdef XML_MIN_SIZE
+#define CHAR_MATCHES(enc, p, c) \
+ (((const struct normal_encoding *)(enc))->charMatches(enc, p, c))
+static
+int sb_charMatches(const ENCODING *enc, const char *p, int c)
+{
+  return *p == c;
+}
+#else
+/* c is an ASCII character */
+#define CHAR_MATCHES(enc, p, c) (*(p) == c)
+#endif
+
+#define PREFIX(ident) normal_ ## ident
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+enum {  /* UTF8_cvalN is value of masked first byte of N byte sequence */
+  UTF8_cval1 = 0x00,
+  UTF8_cval2 = 0xc0,
+  UTF8_cval3 = 0xe0,
+  UTF8_cval4 = 0xf0
+};
+
+static
+void utf8_toUtf8(const ENCODING *enc,
+		 const char **fromP, const char *fromLim,
+		 char **toP, const char *toLim)
+{
+  char *to;
+  const char *from;
+  if (fromLim - *fromP > toLim - *toP) {
+    /* Avoid copying partial characters. */
+    for (fromLim = *fromP + (toLim - *toP); fromLim > *fromP; fromLim--)
+      if (((unsigned char)fromLim[-1] & 0xc0) != 0x80)
+	break;
+  }
+  for (to = *toP, from = *fromP; from != fromLim; from++, to++)
+    *to = *from;
+  *fromP = from;
+  *toP = to;
+}
+
+static
+void utf8_toUtf16(const ENCODING *enc,
+		  const char **fromP, const char *fromLim,
+		  unsigned short **toP, const unsigned short *toLim)
+{
+  unsigned short *to = *toP;
+  const char *from = *fromP;
+  while (from != fromLim && to != toLim) {
+    switch (((struct normal_encoding *)enc)->type[(unsigned char)*from]) {
+    case BT_LEAD2:
+      *to++ = ((from[0] & 0x1f) << 6) | (from[1] & 0x3f);
+      from += 2;
+      break;
+    case BT_LEAD3:
+      *to++ = ((from[0] & 0xf) << 12) | ((from[1] & 0x3f) << 6) | (from[2] & 0x3f);
+      from += 3;
+      break;
+    case BT_LEAD4:
+      {
+	unsigned long n;
+	if (to + 1 == toLim)
+	  break;
+	n = ((from[0] & 0x7) << 18) | ((from[1] & 0x3f) << 12) | ((from[2] & 0x3f) << 6) | (from[3] & 0x3f);
+	n -= 0x10000;
+	to[0] = (unsigned short)((n >> 10) | 0xD800);
+	to[1] = (unsigned short)((n & 0x3FF) | 0xDC00);
+	to += 2;
+	from += 4;
+      }
+      break;
+    default:
+      *to++ = *from++;
+      break;
+    }
+  }
+  *fromP = from;
+  *toP = to;
+}
+
+#ifdef XML_NS
+static const struct normal_encoding utf8_encoding_ns = {
+  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+  {
+#include "asciitab.h"
+#include "utf8tab.h"
+  },
+  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+#endif
+
+static const struct normal_encoding utf8_encoding = {
+  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "utf8tab.h"
+  },
+  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_utf8_encoding_ns = {
+  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+  {
+#include "iasciitab.h"
+#include "utf8tab.h"
+  },
+  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+#endif
+
+static const struct normal_encoding internal_utf8_encoding = {
+  { VTABLE1, utf8_toUtf8, utf8_toUtf16, 1, 1, 0 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "utf8tab.h"
+  },
+  STANDARD_VTABLE(sb_) NORMAL_VTABLE(utf8_)
+};
+
+static
+void latin1_toUtf8(const ENCODING *enc,
+		   const char **fromP, const char *fromLim,
+		   char **toP, const char *toLim)
+{
+  for (;;) {
+    unsigned char c;
+    if (*fromP == fromLim)
+      break;
+    c = (unsigned char)**fromP;
+    if (c & 0x80) {
+      if (toLim - *toP < 2)
+	break;
+      *(*toP)++ = ((c >> 6) | UTF8_cval2);
+      *(*toP)++ = ((c & 0x3f) | 0x80);
+      (*fromP)++;
+    }
+    else {
+      if (*toP == toLim)
+	break;
+      *(*toP)++ = *(*fromP)++;
+    }
+  }
+}
+
+static
+void latin1_toUtf16(const ENCODING *enc,
+		    const char **fromP, const char *fromLim,
+		    unsigned short **toP, const unsigned short *toLim)
+{
+  while (*fromP != fromLim && *toP != toLim)
+    *(*toP)++ = (unsigned char)*(*fromP)++;
+}
+
+#ifdef XML_NS
+
+static const struct normal_encoding latin1_encoding_ns = {
+  { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
+  {
+#include "asciitab.h"
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(sb_)
+};
+
+#endif
+
+static const struct normal_encoding latin1_encoding = {
+  { VTABLE1, latin1_toUtf8, latin1_toUtf16, 1, 0, 0 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(sb_)
+};
+
+static
+void ascii_toUtf8(const ENCODING *enc,
+		  const char **fromP, const char *fromLim,
+		  char **toP, const char *toLim)
+{
+  while (*fromP != fromLim && *toP != toLim)
+    *(*toP)++ = *(*fromP)++;
+}
+
+#ifdef XML_NS
+
+static const struct normal_encoding ascii_encoding_ns = {
+  { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
+  {
+#include "asciitab.h"
+/* BT_NONXML == 0 */
+  },
+  STANDARD_VTABLE(sb_)
+};
+
+#endif
+
+static const struct normal_encoding ascii_encoding = {
+  { VTABLE1, ascii_toUtf8, latin1_toUtf16, 1, 1, 0 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+/* BT_NONXML == 0 */
+  },
+  STANDARD_VTABLE(sb_)
+};
+
+static int unicode_byte_type(char hi, char lo)
+{
+  switch ((unsigned char)hi) {
+  case 0xD8: case 0xD9: case 0xDA: case 0xDB:
+    return BT_LEAD4;
+  case 0xDC: case 0xDD: case 0xDE: case 0xDF:
+    return BT_TRAIL;
+  case 0xFF:
+    switch ((unsigned char)lo) {
+    case 0xFF:
+    case 0xFE:
+      return BT_NONXML;
+    }
+    break;
+  }
+  return BT_NONASCII;
+}
+
+#define DEFINE_UTF16_TO_UTF8(E) \
+static \
+void E ## toUtf8(const ENCODING *enc, \
+		 const char **fromP, const char *fromLim, \
+		 char **toP, const char *toLim) \
+{ \
+  const char *from; \
+  for (from = *fromP; from != fromLim; from += 2) { \
+    int plane; \
+    unsigned char lo2; \
+    unsigned char lo = GET_LO(from); \
+    unsigned char hi = GET_HI(from); \
+    switch (hi) { \
+    case 0: \
+      if (lo < 0x80) { \
+        if (*toP == toLim) { \
+          *fromP = from; \
+	  return; \
+        } \
+        *(*toP)++ = lo; \
+        break; \
+      } \
+      /* fall through */ \
+    case 0x1: case 0x2: case 0x3: \
+    case 0x4: case 0x5: case 0x6: case 0x7: \
+      if (toLim -  *toP < 2) { \
+        *fromP = from; \
+	return; \
+      } \
+      *(*toP)++ = ((lo >> 6) | (hi << 2) |  UTF8_cval2); \
+      *(*toP)++ = ((lo & 0x3f) | 0x80); \
+      break; \
+    default: \
+      if (toLim -  *toP < 3)  { \
+        *fromP = from; \
+	return; \
+      } \
+      /* 16 bits divided 4, 6, 6 amongst 3 bytes */ \
+      *(*toP)++ = ((hi >> 4) | UTF8_cval3); \
+      *(*toP)++ = (((hi & 0xf) << 2) | (lo >> 6) | 0x80); \
+      *(*toP)++ = ((lo & 0x3f) | 0x80); \
+      break; \
+    case 0xD8: case 0xD9: case 0xDA: case 0xDB: \
+      if (toLim -  *toP < 4) { \
+	*fromP = from; \
+	return; \
+      } \
+      plane = (((hi & 0x3) << 2) | ((lo >> 6) & 0x3)) + 1; \
+      *(*toP)++ = ((plane >> 2) | UTF8_cval4); \
+      *(*toP)++ = (((lo >> 2) & 0xF) | ((plane & 0x3) << 4) | 0x80); \
+      from += 2; \
+      lo2 = GET_LO(from); \
+      *(*toP)++ = (((lo & 0x3) << 4) \
+	           | ((GET_HI(from) & 0x3) << 2) \
+		   | (lo2 >> 6) \
+		   | 0x80); \
+      *(*toP)++ = ((lo2 & 0x3f) | 0x80); \
+      break; \
+    } \
+  } \
+  *fromP = from; \
+}
+
+#define DEFINE_UTF16_TO_UTF16(E) \
+static \
+void E ## toUtf16(const ENCODING *enc, \
+		  const char **fromP, const char *fromLim, \
+		  unsigned short **toP, const unsigned short *toLim) \
+{ \
+  /* Avoid copying first half only of surrogate */ \
+  if (fromLim - *fromP > ((toLim - *toP) << 1) \
+      && (GET_HI(fromLim - 2) & 0xF8) == 0xD8) \
+    fromLim -= 2; \
+  for (; *fromP != fromLim && *toP != toLim; *fromP += 2) \
+    *(*toP)++ = (GET_HI(*fromP) << 8) | GET_LO(*fromP); \
+}
+
+#define SET2(ptr, ch) \
+  (((ptr)[0] = ((ch) & 0xff)), ((ptr)[1] = ((ch) >> 8)))
+#define GET_LO(ptr) ((unsigned char)(ptr)[0])
+#define GET_HI(ptr) ((unsigned char)(ptr)[1])
+
+DEFINE_UTF16_TO_UTF8(little2_)
+DEFINE_UTF16_TO_UTF16(little2_)
+
+#undef SET2
+#undef GET_LO
+#undef GET_HI
+
+#define SET2(ptr, ch) \
+  (((ptr)[0] = ((ch) >> 8)), ((ptr)[1] = ((ch) & 0xFF)))
+#define GET_LO(ptr) ((unsigned char)(ptr)[1])
+#define GET_HI(ptr) ((unsigned char)(ptr)[0])
+
+DEFINE_UTF16_TO_UTF8(big2_)
+DEFINE_UTF16_TO_UTF16(big2_)
+
+#undef SET2
+#undef GET_LO
+#undef GET_HI
+
+#define LITTLE2_BYTE_TYPE(enc, p) \
+ ((p)[1] == 0 \
+  ? ((struct normal_encoding *)(enc))->type[(unsigned char)*(p)] \
+  : unicode_byte_type((p)[1], (p)[0]))
+#define LITTLE2_BYTE_TO_ASCII(enc, p) ((p)[1] == 0 ? (p)[0] : -1)
+#define LITTLE2_CHAR_MATCHES(enc, p, c) ((p)[1] == 0 && (p)[0] == c)
+#define LITTLE2_IS_NAME_CHAR_MINBPC(enc, p) \
+  UCS2_GET_NAMING(namePages, (unsigned char)p[1], (unsigned char)p[0])
+#define LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p) \
+  UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[1], (unsigned char)p[0])
+
+#ifdef XML_MIN_SIZE
+
+static
+int little2_byteType(const ENCODING *enc, const char *p)
+{
+  return LITTLE2_BYTE_TYPE(enc, p);
+}
+
+static
+int little2_byteToAscii(const ENCODING *enc, const char *p)
+{
+  return LITTLE2_BYTE_TO_ASCII(enc, p);
+}
+
+static
+int little2_charMatches(const ENCODING *enc, const char *p, int c)
+{
+  return LITTLE2_CHAR_MATCHES(enc, p, c);
+}
+
+static
+int little2_isNameMin(const ENCODING *enc, const char *p)
+{
+  return LITTLE2_IS_NAME_CHAR_MINBPC(enc, p);
+}
+
+static
+int little2_isNmstrtMin(const ENCODING *enc, const char *p)
+{
+  return LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p);
+}
+
+#undef VTABLE
+#define VTABLE VTABLE1, little2_toUtf8, little2_toUtf16
+
+#else /* not XML_MIN_SIZE */
+
+#undef PREFIX
+#define PREFIX(ident) little2_ ## ident
+#define MINBPC(enc) 2
+/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
+#define BYTE_TYPE(enc, p) LITTLE2_BYTE_TYPE(enc, p)
+#define BYTE_TO_ASCII(enc, p) LITTLE2_BYTE_TO_ASCII(enc, p) 
+#define CHAR_MATCHES(enc, p, c) LITTLE2_CHAR_MATCHES(enc, p, c)
+#define IS_NAME_CHAR(enc, p, n) 0
+#define IS_NAME_CHAR_MINBPC(enc, p) LITTLE2_IS_NAME_CHAR_MINBPC(enc, p)
+#define IS_NMSTRT_CHAR(enc, p, n) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) LITTLE2_IS_NMSTRT_CHAR_MINBPC(enc, p)
+
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+#endif /* not XML_MIN_SIZE */
+
+#ifdef XML_NS
+
+static const struct normal_encoding little2_encoding_ns = { 
+  { VTABLE, 2, 0,
+#if XML_BYTE_ORDER == 12
+    1
+#else
+    0
+#endif
+  },
+  {
+#include "asciitab.h"
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+static const struct normal_encoding little2_encoding = { 
+  { VTABLE, 2, 0,
+#if XML_BYTE_ORDER == 12
+    1
+#else
+    0
+#endif
+  },
+  {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(little2_)
+};
+
+#if XML_BYTE_ORDER != 21
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_little2_encoding_ns = { 
+  { VTABLE, 2, 0, 1 },
+  {
+#include "iasciitab.h"
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+static const struct normal_encoding internal_little2_encoding = { 
+  { VTABLE, 2, 0, 1 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(little2_)
+};
+
+#endif
+
+
+#define BIG2_BYTE_TYPE(enc, p) \
+ ((p)[0] == 0 \
+  ? ((struct normal_encoding *)(enc))->type[(unsigned char)(p)[1]] \
+  : unicode_byte_type((p)[0], (p)[1]))
+#define BIG2_BYTE_TO_ASCII(enc, p) ((p)[0] == 0 ? (p)[1] : -1)
+#define BIG2_CHAR_MATCHES(enc, p, c) ((p)[0] == 0 && (p)[1] == c)
+#define BIG2_IS_NAME_CHAR_MINBPC(enc, p) \
+  UCS2_GET_NAMING(namePages, (unsigned char)p[0], (unsigned char)p[1])
+#define BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p) \
+  UCS2_GET_NAMING(nmstrtPages, (unsigned char)p[0], (unsigned char)p[1])
+
+#ifdef XML_MIN_SIZE
+
+static
+int big2_byteType(const ENCODING *enc, const char *p)
+{
+  return BIG2_BYTE_TYPE(enc, p);
+}
+
+static
+int big2_byteToAscii(const ENCODING *enc, const char *p)
+{
+  return BIG2_BYTE_TO_ASCII(enc, p);
+}
+
+static
+int big2_charMatches(const ENCODING *enc, const char *p, int c)
+{
+  return BIG2_CHAR_MATCHES(enc, p, c);
+}
+
+static
+int big2_isNameMin(const ENCODING *enc, const char *p)
+{
+  return BIG2_IS_NAME_CHAR_MINBPC(enc, p);
+}
+
+static
+int big2_isNmstrtMin(const ENCODING *enc, const char *p)
+{
+  return BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p);
+}
+
+#undef VTABLE
+#define VTABLE VTABLE1, big2_toUtf8, big2_toUtf16
+
+#else /* not XML_MIN_SIZE */
+
+#undef PREFIX
+#define PREFIX(ident) big2_ ## ident
+#define MINBPC(enc) 2
+/* CHAR_MATCHES is guaranteed to have MINBPC bytes available. */
+#define BYTE_TYPE(enc, p) BIG2_BYTE_TYPE(enc, p)
+#define BYTE_TO_ASCII(enc, p) BIG2_BYTE_TO_ASCII(enc, p) 
+#define CHAR_MATCHES(enc, p, c) BIG2_CHAR_MATCHES(enc, p, c)
+#define IS_NAME_CHAR(enc, p, n) 0
+#define IS_NAME_CHAR_MINBPC(enc, p) BIG2_IS_NAME_CHAR_MINBPC(enc, p)
+#define IS_NMSTRT_CHAR(enc, p, n) (0)
+#define IS_NMSTRT_CHAR_MINBPC(enc, p) BIG2_IS_NMSTRT_CHAR_MINBPC(enc, p)
+
+#include "xmltok_impl.c"
+
+#undef MINBPC
+#undef BYTE_TYPE
+#undef BYTE_TO_ASCII
+#undef CHAR_MATCHES
+#undef IS_NAME_CHAR
+#undef IS_NAME_CHAR_MINBPC
+#undef IS_NMSTRT_CHAR
+#undef IS_NMSTRT_CHAR_MINBPC
+#undef IS_INVALID_CHAR
+
+#endif /* not XML_MIN_SIZE */
+
+#ifdef XML_NS
+
+static const struct normal_encoding big2_encoding_ns = {
+  { VTABLE, 2, 0,
+#if XML_BYTE_ORDER == 21
+  1
+#else
+  0
+#endif
+  },
+  {
+#include "asciitab.h"
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+static const struct normal_encoding big2_encoding = {
+  { VTABLE, 2, 0,
+#if XML_BYTE_ORDER == 21
+  1
+#else
+  0
+#endif
+  },
+  {
+#define BT_COLON BT_NMSTRT
+#include "asciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(big2_)
+};
+
+#if XML_BYTE_ORDER != 12
+
+#ifdef XML_NS
+
+static const struct normal_encoding internal_big2_encoding_ns = {
+  { VTABLE, 2, 0, 1 },
+  {
+#include "iasciitab.h"
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+static const struct normal_encoding internal_big2_encoding = {
+  { VTABLE, 2, 0, 1 },
+  {
+#define BT_COLON BT_NMSTRT
+#include "iasciitab.h"
+#undef BT_COLON
+#include "latin1tab.h"
+  },
+  STANDARD_VTABLE(big2_)
+};
+
+#endif
+
+#undef PREFIX
+
+static
+int streqci(const char *s1, const char *s2)
+{
+  for (;;) {
+    char c1 = *s1++;
+    char c2 = *s2++;
+    if (ASCII_a <= c1 && c1 <= ASCII_z)
+      c1 += ASCII_A - ASCII_a;
+    if (ASCII_a <= c2 && c2 <= ASCII_z)
+      c2 += ASCII_A - ASCII_a;
+    if (c1 != c2)
+      return 0;
+    if (!c1)
+      break;
+  }
+  return 1;
+}
+
+static
+void initUpdatePosition(const ENCODING *enc, const char *ptr,
+			const char *end, POSITION *pos)
+{
+  normal_updatePosition(&utf8_encoding.enc, ptr, end, pos);
+}
+
+static
+int toAscii(const ENCODING *enc, const char *ptr, const char *end)
+{
+  char buf[1];
+  char *p = buf;
+  XmlUtf8Convert(enc, &ptr, end, &p, p + 1);
+  if (p == buf)
+    return -1;
+  else
+    return buf[0];
+}
+
+static
+int isSpace(int c)
+{
+  switch (c) {
+  case 0x20:
+  case 0xD:
+  case 0xA:
+  case 0x9:	
+    return 1;
+  }
+  return 0;
+}
+
+/* Return 1 if there's just optional white space
+or there's an S followed by name=val. */
+static
+int parsePseudoAttribute(const ENCODING *enc,
+			 const char *ptr,
+			 const char *end,
+			 const char **namePtr,
+			 const char **nameEndPtr,
+			 const char **valPtr,
+			 const char **nextTokPtr)
+{
+  int c;
+  char open;
+  if (ptr == end) {
+    *namePtr = 0;
+    return 1;
+  }
+  if (!isSpace(toAscii(enc, ptr, end))) {
+    *nextTokPtr = ptr;
+    return 0;
+  }
+  do {
+    ptr += enc->minBytesPerChar;
+  } while (isSpace(toAscii(enc, ptr, end)));
+  if (ptr == end) {
+    *namePtr = 0;
+    return 1;
+  }
+  *namePtr = ptr;
+  for (;;) {
+    c = toAscii(enc, ptr, end);
+    if (c == -1) {
+      *nextTokPtr = ptr;
+      return 0;
+    }
+    if (c == ASCII_EQUALS) {
+      *nameEndPtr = ptr;
+      break;
+    }
+    if (isSpace(c)) {
+      *nameEndPtr = ptr;
+      do {
+	ptr += enc->minBytesPerChar;
+      } while (isSpace(c = toAscii(enc, ptr, end)));
+      if (c != ASCII_EQUALS) {
+	*nextTokPtr = ptr;
+	return 0;
+      }
+      break;
+    }
+    ptr += enc->minBytesPerChar;
+  }
+  if (ptr == *namePtr) {
+    *nextTokPtr = ptr;
+    return 0;
+  }
+  ptr += enc->minBytesPerChar;
+  c = toAscii(enc, ptr, end);
+  while (isSpace(c)) {
+    ptr += enc->minBytesPerChar;
+    c = toAscii(enc, ptr, end);
+  }
+  if (c != ASCII_QUOT && c != ASCII_APOS) {
+    *nextTokPtr = ptr;
+    return 0;
+  }
+  open = c;
+  ptr += enc->minBytesPerChar;
+  *valPtr = ptr;
+  for (;; ptr += enc->minBytesPerChar) {
+    c = toAscii(enc, ptr, end);
+    if (c == open)
+      break;
+    if (!(ASCII_a <= c && c <= ASCII_z)
+	&& !(ASCII_A <= c && c <= ASCII_Z)
+	&& !(ASCII_0 <= c && c <= ASCII_9)
+	&& c != ASCII_PERIOD
+	&& c != ASCII_MINUS
+	&& c != ASCII_UNDERSCORE) {
+      *nextTokPtr = ptr;
+      return 0;
+    }
+  }
+  *nextTokPtr = ptr + enc->minBytesPerChar;
+  return 1;
+}
+
+static const char KW_version[] = {
+  ASCII_v, ASCII_e, ASCII_r, ASCII_s, ASCII_i, ASCII_o, ASCII_n, '\0'
+};
+
+static const char KW_encoding[] = {
+  ASCII_e, ASCII_n, ASCII_c, ASCII_o, ASCII_d, ASCII_i, ASCII_n, ASCII_g, '\0'
+};
+
+static const char KW_standalone[] = {
+  ASCII_s, ASCII_t, ASCII_a, ASCII_n, ASCII_d, ASCII_a, ASCII_l, ASCII_o, ASCII_n, ASCII_e, '\0'
+};
+
+static const char KW_yes[] = {
+  ASCII_y, ASCII_e, ASCII_s,  '\0'
+};
+
+static const char KW_no[] = {
+  ASCII_n, ASCII_o,  '\0'
+};
+
+static
+int doParseXmlDecl(const ENCODING *(*encodingFinder)(const ENCODING *,
+		                                     const char *,
+						     const char *),
+		   int isGeneralTextEntity,
+		   const ENCODING *enc,
+		   const char *ptr,
+		   const char *end,
+		   const char **badPtr,
+		   const char **versionPtr,
+		   const char **versionEndPtr,
+		   const char **encodingName,
+		   const ENCODING **encoding,
+		   int *standalone)
+{
+  const char *val = 0;
+  const char *name = 0;
+  const char *nameEnd = 0;
+  ptr += 5 * enc->minBytesPerChar;
+  end -= 2 * enc->minBytesPerChar;
+  if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr) || !name) {
+    *badPtr = ptr;
+    return 0;
+  }
+  if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_version)) {
+    if (!isGeneralTextEntity) {
+      *badPtr = name;
+      return 0;
+    }
+  }
+  else {
+    if (versionPtr)
+      *versionPtr = val;
+    if (versionEndPtr)
+      *versionEndPtr = ptr;
+    if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) {
+      *badPtr = ptr;
+      return 0;
+    }
+    if (!name) {
+      if (isGeneralTextEntity) {
+	/* a TextDecl must have an EncodingDecl */
+	*badPtr = ptr;
+	return 0;
+      }
+      return 1;
+    }
+  }
+  if (XmlNameMatchesAscii(enc, name, nameEnd, KW_encoding)) {
+    int c = toAscii(enc, val, end);
+    if (!(ASCII_a <= c && c <= ASCII_z) && !(ASCII_A <= c && c <= ASCII_Z)) {
+      *badPtr = val;
+      return 0;
+    }
+    if (encodingName)
+      *encodingName = val;
+    if (encoding)
+      *encoding = encodingFinder(enc, val, ptr - enc->minBytesPerChar);
+    if (!parsePseudoAttribute(enc, ptr, end, &name, &nameEnd, &val, &ptr)) {
+      *badPtr = ptr;
+      return 0;
+    }
+    if (!name)
+      return 1;
+  }
+  if (!XmlNameMatchesAscii(enc, name, nameEnd, KW_standalone) || isGeneralTextEntity) {
+    *badPtr = name;
+    return 0;
+  }
+  if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_yes)) {
+    if (standalone)
+      *standalone = 1;
+  }
+  else if (XmlNameMatchesAscii(enc, val, ptr - enc->minBytesPerChar, KW_no)) {
+    if (standalone)
+      *standalone = 0;
+  }
+  else {
+    *badPtr = val;
+    return 0;
+  }
+  while (isSpace(toAscii(enc, ptr, end)))
+    ptr += enc->minBytesPerChar;
+  if (ptr != end) {
+    *badPtr = ptr;
+    return 0;
+  }
+  return 1;
+}
+
+static
+int checkCharRefNumber(int result)
+{
+  switch (result >> 8) {
+  case 0xD8: case 0xD9: case 0xDA: case 0xDB:
+  case 0xDC: case 0xDD: case 0xDE: case 0xDF:
+    return -1;
+  case 0:
+    if (latin1_encoding.type[result] == BT_NONXML)
+      return -1;
+    break;
+  case 0xFF:
+    if (result == 0xFFFE || result == 0xFFFF)
+      return -1;
+    break;
+  }
+  return result;
+}
+
+int XmlUtf8Encode(int c, char *buf)
+{
+  enum {
+    /* minN is minimum legal resulting value for N byte sequence */
+    min2 = 0x80,
+    min3 = 0x800,
+    min4 = 0x10000
+  };
+
+  if (c < 0)
+    return 0;
+  if (c < min2) {
+    buf[0] = (c | UTF8_cval1);
+    return 1;
+  }
+  if (c < min3) {
+    buf[0] = ((c >> 6) | UTF8_cval2);
+    buf[1] = ((c & 0x3f) | 0x80);
+    return 2;
+  }
+  if (c < min4) {
+    buf[0] = ((c >> 12) | UTF8_cval3);
+    buf[1] = (((c >> 6) & 0x3f) | 0x80);
+    buf[2] = ((c & 0x3f) | 0x80);
+    return 3;
+  }
+  if (c < 0x110000) {
+    buf[0] = ((c >> 18) | UTF8_cval4);
+    buf[1] = (((c >> 12) & 0x3f) | 0x80);
+    buf[2] = (((c >> 6) & 0x3f) | 0x80);
+    buf[3] = ((c & 0x3f) | 0x80);
+    return 4;
+  }
+  return 0;
+}
+
+int XmlUtf16Encode(int charNum, unsigned short *buf)
+{
+  if (charNum < 0)
+    return 0;
+  if (charNum < 0x10000) {
+    buf[0] = charNum;
+    return 1;
+  }
+  if (charNum < 0x110000) {
+    charNum -= 0x10000;
+    buf[0] = (charNum >> 10) + 0xD800;
+    buf[1] = (charNum & 0x3FF) + 0xDC00;
+    return 2;
+  }
+  return 0;
+}
+
+struct unknown_encoding {
+  struct normal_encoding normal;
+  int (*convert)(void *userData, const char *p);
+  void *userData;
+  unsigned short utf16[256];
+  char utf8[256][4];
+};
+
+int XmlSizeOfUnknownEncoding(void)
+{
+  return sizeof(struct unknown_encoding);
+}
+
+static
+int unknown_isName(const ENCODING *enc, const char *p)
+{
+  int c = ((const struct unknown_encoding *)enc)
+	  ->convert(((const struct unknown_encoding *)enc)->userData, p);
+  if (c & ~0xFFFF)
+    return 0;
+  return UCS2_GET_NAMING(namePages, c >> 8, c & 0xFF);
+}
+
+static
+int unknown_isNmstrt(const ENCODING *enc, const char *p)
+{
+  int c = ((const struct unknown_encoding *)enc)
+	  ->convert(((const struct unknown_encoding *)enc)->userData, p);
+  if (c & ~0xFFFF)
+    return 0;
+  return UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xFF);
+}
+
+static
+int unknown_isInvalid(const ENCODING *enc, const char *p)
+{
+  int c = ((const struct unknown_encoding *)enc)
+	   ->convert(((const struct unknown_encoding *)enc)->userData, p);
+  return (c & ~0xFFFF) || checkCharRefNumber(c) < 0;
+}
+
+static
+void unknown_toUtf8(const ENCODING *enc,
+		    const char **fromP, const char *fromLim,
+		    char **toP, const char *toLim)
+{
+  char buf[XML_UTF8_ENCODE_MAX];
+  for (;;) {
+    const char *utf8;
+    int n;
+    if (*fromP == fromLim)
+      break;
+    utf8 = ((const struct unknown_encoding *)enc)->utf8[(unsigned char)**fromP];
+    n = *utf8++;
+    if (n == 0) {
+      int c = ((const struct unknown_encoding *)enc)
+	      ->convert(((const struct unknown_encoding *)enc)->userData, *fromP);
+      n = XmlUtf8Encode(c, buf);
+      if (n > toLim - *toP)
+	break;
+      utf8 = buf;
+      *fromP += ((const struct normal_encoding *)enc)->type[(unsigned char)**fromP]
+	         - (BT_LEAD2 - 2);
+    }
+    else {
+      if (n > toLim - *toP)
+	break;
+      (*fromP)++;
+    }
+    do {
+      *(*toP)++ = *utf8++;
+    } while (--n != 0);
+  }
+}
+
+static
+void unknown_toUtf16(const ENCODING *enc,
+		     const char **fromP, const char *fromLim,
+		     unsigned short **toP, const unsigned short *toLim)
+{
+  while (*fromP != fromLim && *toP != toLim) {
+    unsigned short c
+      = ((const struct unknown_encoding *)enc)->utf16[(unsigned char)**fromP];
+    if (c == 0) {
+      c = (unsigned short)((const struct unknown_encoding *)enc)
+	   ->convert(((const struct unknown_encoding *)enc)->userData, *fromP);
+      *fromP += ((const struct normal_encoding *)enc)->type[(unsigned char)**fromP]
+	         - (BT_LEAD2 - 2);
+    }
+    else
+      (*fromP)++;
+    *(*toP)++ = c;
+  }
+}
+
+ENCODING *
+XmlInitUnknownEncoding(void *mem,
+		       int *table,
+		       int (*convert)(void *userData, const char *p),
+		       void *userData)
+{
+  int i;
+  struct unknown_encoding *e = mem;
+  for (i = 0; i < (int)sizeof(struct normal_encoding); i++)
+    ((char *)mem)[i] = ((char *)&latin1_encoding)[i];
+  for (i = 0; i < 128; i++)
+    if (latin1_encoding.type[i] != BT_OTHER
+        && latin1_encoding.type[i] != BT_NONXML
+	&& table[i] != i)
+      return 0;
+  for (i = 0; i < 256; i++) {
+    int c = table[i];
+    if (c == -1) {
+      e->normal.type[i] = BT_MALFORM;
+      /* This shouldn't really get used. */
+      e->utf16[i] = 0xFFFF;
+      e->utf8[i][0] = 1;
+      e->utf8[i][1] = 0;
+    }
+    else if (c < 0) {
+      if (c < -4)
+	return 0;
+      e->normal.type[i] = BT_LEAD2 - (c + 2);
+      e->utf8[i][0] = 0;
+      e->utf16[i] = 0;
+    }
+    else if (c < 0x80) {
+      if (latin1_encoding.type[c] != BT_OTHER
+	  && latin1_encoding.type[c] != BT_NONXML
+	  && c != i)
+	return 0;
+      e->normal.type[i] = latin1_encoding.type[c];
+      e->utf8[i][0] = 1;
+      e->utf8[i][1] = (char)c;
+      e->utf16[i] = c == 0 ? 0xFFFF : c;
+    }
+    else if (checkCharRefNumber(c) < 0) {
+      e->normal.type[i] = BT_NONXML;
+      /* This shouldn't really get used. */
+      e->utf16[i] = 0xFFFF;
+      e->utf8[i][0] = 1;
+      e->utf8[i][1] = 0;
+    }
+    else {
+      if (c > 0xFFFF)
+	return 0;
+      if (UCS2_GET_NAMING(nmstrtPages, c >> 8, c & 0xff))
+	e->normal.type[i] = BT_NMSTRT;
+      else if (UCS2_GET_NAMING(namePages, c >> 8, c & 0xff))
+	e->normal.type[i] = BT_NAME;
+      else
+	e->normal.type[i] = BT_OTHER;
+      e->utf8[i][0] = (char)XmlUtf8Encode(c, e->utf8[i] + 1);
+      e->utf16[i] = c;
+    }
+  }
+  e->userData = userData;
+  e->convert = convert;
+  if (convert) {
+    e->normal.isName2 = unknown_isName;
+    e->normal.isName3 = unknown_isName;
+    e->normal.isName4 = unknown_isName;
+    e->normal.isNmstrt2 = unknown_isNmstrt;
+    e->normal.isNmstrt3 = unknown_isNmstrt;
+    e->normal.isNmstrt4 = unknown_isNmstrt;
+    e->normal.isInvalid2 = unknown_isInvalid;
+    e->normal.isInvalid3 = unknown_isInvalid;
+    e->normal.isInvalid4 = unknown_isInvalid;
+  }
+  e->normal.enc.utf8Convert = unknown_toUtf8;
+  e->normal.enc.utf16Convert = unknown_toUtf16;
+  return &(e->normal.enc);
+}
+
+/* If this enumeration is changed, getEncodingIndex and encodings
+must also be changed. */
+enum {
+  UNKNOWN_ENC = -1,
+  ISO_8859_1_ENC = 0,
+  US_ASCII_ENC,
+  UTF_8_ENC,
+  UTF_16_ENC,
+  UTF_16BE_ENC,
+  UTF_16LE_ENC,
+  /* must match encodingNames up to here */
+  NO_ENC
+};
+
+static const char KW_ISO_8859_1[] = {
+  ASCII_I, ASCII_S, ASCII_O, ASCII_MINUS, ASCII_8, ASCII_8, ASCII_5, ASCII_9, ASCII_MINUS, ASCII_1, '\0'
+};
+static const char KW_US_ASCII[] = {
+  ASCII_U, ASCII_S, ASCII_MINUS, ASCII_A, ASCII_S, ASCII_C, ASCII_I, ASCII_I, '\0'
+};
+static const char KW_UTF_8[] =	{
+  ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_8, '\0'
+};
+static const char KW_UTF_16[] =	{
+  ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, '\0'
+};
+static const char KW_UTF_16BE[] = {
+  ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_B, ASCII_E, '\0'
+};
+static const char KW_UTF_16LE[] = {
+  ASCII_U, ASCII_T, ASCII_F, ASCII_MINUS, ASCII_1, ASCII_6, ASCII_L, ASCII_E, '\0'
+};
+
+static
+int getEncodingIndex(const char *name)
+{
+  static const char *encodingNames[] = {
+    KW_ISO_8859_1,
+    KW_US_ASCII,
+    KW_UTF_8,
+    KW_UTF_16,
+    KW_UTF_16BE,
+    KW_UTF_16LE,
+  };
+  int i;
+  if (name == 0)
+    return NO_ENC;
+  for (i = 0; i < (int)(sizeof(encodingNames)/sizeof(encodingNames[0])); i++)
+    if (streqci(name, encodingNames[i]))
+      return i;
+  return UNKNOWN_ENC;
+}
+
+/* For binary compatibility, we store the index of the encoding specified
+at initialization in the isUtf16 member. */
+
+#define INIT_ENC_INDEX(enc) ((int)(enc)->initEnc.isUtf16)
+#define SET_INIT_ENC_INDEX(enc, i) ((enc)->initEnc.isUtf16 = (char)i)
+
+/* This is what detects the encoding.
+encodingTable maps from encoding indices to encodings;
+INIT_ENC_INDEX(enc) is the index of the external (protocol) specified encoding;
+state is XML_CONTENT_STATE if we're parsing an external text entity,
+and XML_PROLOG_STATE otherwise.
+*/
+
+
+static
+int initScan(const ENCODING **encodingTable,
+	     const INIT_ENCODING *enc,
+	     int state,
+	     const char *ptr,
+	     const char *end,
+	     const char **nextTokPtr)
+{
+  const ENCODING **encPtr;
+
+  if (ptr == end)
+    return XML_TOK_NONE;
+  encPtr = enc->encPtr;
+  if (ptr + 1 == end) {
+    /* only a single byte available for auto-detection */
+#ifndef XML_DTD /* FIXME */
+    /* a well-formed document entity must have more than one byte */
+    if (state != XML_CONTENT_STATE)
+      return XML_TOK_PARTIAL;
+#endif
+    /* so we're parsing an external text entity... */
+    /* if UTF-16 was externally specified, then we need at least 2 bytes */
+    switch (INIT_ENC_INDEX(enc)) {
+    case UTF_16_ENC:
+    case UTF_16LE_ENC:
+    case UTF_16BE_ENC:
+      return XML_TOK_PARTIAL;
+    }
+    switch ((unsigned char)*ptr) {
+    case 0xFE:
+    case 0xFF:
+    case 0xEF: /* possibly first byte of UTF-8 BOM */
+      if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+	  && state == XML_CONTENT_STATE)
+	break;
+      /* fall through */
+    case 0x00:
+    case 0x3C:
+      return XML_TOK_PARTIAL;
+    }
+  }
+  else {
+    switch (((unsigned char)ptr[0] << 8) | (unsigned char)ptr[1]) {
+    case 0xFEFF:
+      if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+	  && state == XML_CONTENT_STATE)
+	break;
+      *nextTokPtr = ptr + 2;
+      *encPtr = encodingTable[UTF_16BE_ENC];
+      return XML_TOK_BOM;
+    /* 00 3C is handled in the default case */
+    case 0x3C00:
+      if ((INIT_ENC_INDEX(enc) == UTF_16BE_ENC
+	   || INIT_ENC_INDEX(enc) == UTF_16_ENC)
+	  && state == XML_CONTENT_STATE)
+	break;
+      *encPtr = encodingTable[UTF_16LE_ENC];
+      return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+    case 0xFFFE:
+      if (INIT_ENC_INDEX(enc) == ISO_8859_1_ENC
+	  && state == XML_CONTENT_STATE)
+	break;
+      *nextTokPtr = ptr + 2;
+      *encPtr = encodingTable[UTF_16LE_ENC];
+      return XML_TOK_BOM;
+    case 0xEFBB:
+      /* Maybe a UTF-8 BOM (EF BB BF) */
+      /* If there's an explicitly specified (external) encoding
+         of ISO-8859-1 or some flavour of UTF-16
+         and this is an external text entity,
+	 don't look for the BOM,
+         because it might be a legal data. */
+      if (state == XML_CONTENT_STATE) {
+	int e = INIT_ENC_INDEX(enc);
+	if (e == ISO_8859_1_ENC || e == UTF_16BE_ENC || e == UTF_16LE_ENC || e == UTF_16_ENC)
+	  break;
+      }
+      if (ptr + 2 == end)
+	return XML_TOK_PARTIAL;
+      if ((unsigned char)ptr[2] == 0xBF) {
+	*nextTokPtr = ptr + 3;
+	*encPtr = encodingTable[UTF_8_ENC];
+	return XML_TOK_BOM;
+      }
+      break;
+    default:
+      if (ptr[0] == '\0') {
+	/* 0 isn't a legal data character. Furthermore a document entity can only
+	   start with ASCII characters.  So the only way this can fail to be big-endian
+	   UTF-16 if it it's an external parsed general entity that's labelled as
+	   UTF-16LE. */
+	if (state == XML_CONTENT_STATE && INIT_ENC_INDEX(enc) == UTF_16LE_ENC)
+	  break;
+	*encPtr = encodingTable[UTF_16BE_ENC];
+	return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+      }
+      else if (ptr[1] == '\0') {
+	/* We could recover here in the case:
+	    - parsing an external entity
+	    - second byte is 0
+	    - no externally specified encoding
+	    - no encoding declaration
+	   by assuming UTF-16LE.  But we don't, because this would mean when
+	   presented just with a single byte, we couldn't reliably determine
+	   whether we needed further bytes. */
+	if (state == XML_CONTENT_STATE)
+	  break;
+	*encPtr = encodingTable[UTF_16LE_ENC];
+	return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+      }
+      break;
+    }
+  }
+  *encPtr = encodingTable[INIT_ENC_INDEX(enc)];
+  return XmlTok(*encPtr, state, ptr, end, nextTokPtr);
+}
+
+
+#define NS(x) x
+#define ns(x) x
+#include "xmltok_ns.c"
+#undef NS
+#undef ns
+
+#ifdef XML_NS
+
+#define NS(x) x ## NS
+#define ns(x) x ## _ns
+
+#include "xmltok_ns.c"
+
+#undef NS
+#undef ns
+
+ENCODING *
+XmlInitUnknownEncodingNS(void *mem,
+		         int *table,
+		         int (*convert)(void *userData, const char *p),
+		         void *userData)
+{
+  ENCODING *enc = XmlInitUnknownEncoding(mem, table, convert, userData);
+  if (enc)
+    ((struct normal_encoding *)enc)->type[ASCII_COLON] = BT_COLON;
+  return enc;
+}
+
+#endif /* XML_NS */
diff --git a/xml/expat/lib/xmltok.h b/xml/expat/lib/xmltok.h
new file mode 100644
index 000000000..8b02324c3
--- /dev/null
+++ b/xml/expat/lib/xmltok.h
@@ -0,0 +1,299 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+#ifndef XmlTok_INCLUDED
+#define XmlTok_INCLUDED 1
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* The following token may be returned by XmlContentTok */
+#define XML_TOK_TRAILING_RSQB -5 /* ] or ]] at the end of the scan; might be start of
+                                    illegal ]]> sequence */
+/* The following tokens may be returned by both XmlPrologTok and XmlContentTok */
+#define XML_TOK_NONE -4    /* The string to be scanned is empty */
+#define XML_TOK_TRAILING_CR -3 /* A CR at the end of the scan;
+                                  might be part of CRLF sequence */ 
+#define XML_TOK_PARTIAL_CHAR -2 /* only part of a multibyte sequence */
+#define XML_TOK_PARTIAL -1 /* only part of a token */
+#define XML_TOK_INVALID 0
+
+/* The following tokens are returned by XmlContentTok; some are also
+  returned by XmlAttributeValueTok, XmlEntityTok, XmlCdataSectionTok */
+
+#define XML_TOK_START_TAG_WITH_ATTS 1
+#define XML_TOK_START_TAG_NO_ATTS 2
+#define XML_TOK_EMPTY_ELEMENT_WITH_ATTS 3 /* empty element tag <e/> */
+#define XML_TOK_EMPTY_ELEMENT_NO_ATTS 4
+#define XML_TOK_END_TAG 5
+#define XML_TOK_DATA_CHARS 6
+#define XML_TOK_DATA_NEWLINE 7
+#define XML_TOK_CDATA_SECT_OPEN 8
+#define XML_TOK_ENTITY_REF 9
+#define XML_TOK_CHAR_REF 10     /* numeric character reference */
+
+/* The following tokens may be returned by both XmlPrologTok and XmlContentTok */
+#define XML_TOK_PI 11      /* processing instruction */
+#define XML_TOK_XML_DECL 12 /* XML decl or text decl */
+#define XML_TOK_COMMENT 13
+#define XML_TOK_BOM 14     /* Byte order mark */
+
+/* The following tokens are returned only by XmlPrologTok */
+#define XML_TOK_PROLOG_S 15
+#define XML_TOK_DECL_OPEN 16 /* <!foo */
+#define XML_TOK_DECL_CLOSE 17 /* > */
+#define XML_TOK_NAME 18
+#define XML_TOK_NMTOKEN 19
+#define XML_TOK_POUND_NAME 20 /* #name */
+#define XML_TOK_OR 21 /* | */
+#define XML_TOK_PERCENT 22
+#define XML_TOK_OPEN_PAREN 23
+#define XML_TOK_CLOSE_PAREN 24
+#define XML_TOK_OPEN_BRACKET 25
+#define XML_TOK_CLOSE_BRACKET 26
+#define XML_TOK_LITERAL 27
+#define XML_TOK_PARAM_ENTITY_REF 28
+#define XML_TOK_INSTANCE_START 29
+
+/* The following occur only in element type declarations */
+#define XML_TOK_NAME_QUESTION 30 /* name? */
+#define XML_TOK_NAME_ASTERISK 31 /* name* */
+#define XML_TOK_NAME_PLUS 32 /* name+ */
+#define XML_TOK_COND_SECT_OPEN 33 /* <![ */
+#define XML_TOK_COND_SECT_CLOSE 34 /* ]]> */
+#define XML_TOK_CLOSE_PAREN_QUESTION 35 /* )? */
+#define XML_TOK_CLOSE_PAREN_ASTERISK 36 /* )* */
+#define XML_TOK_CLOSE_PAREN_PLUS 37 /* )+ */
+#define XML_TOK_COMMA 38
+
+/* The following token is returned only by XmlAttributeValueTok */
+#define XML_TOK_ATTRIBUTE_VALUE_S 39
+
+/* The following token is returned only by XmlCdataSectionTok */
+#define XML_TOK_CDATA_SECT_CLOSE 40
+
+/* With namespace processing this is returned by XmlPrologTok
+   for a name with a colon. */
+#define XML_TOK_PREFIXED_NAME 41
+
+#ifdef XML_DTD
+#define XML_TOK_IGNORE_SECT 42
+#endif /* XML_DTD */
+
+#ifdef XML_DTD
+#define XML_N_STATES 4
+#else /* not XML_DTD */
+#define XML_N_STATES 3
+#endif /* not XML_DTD */
+
+#define XML_PROLOG_STATE 0
+#define XML_CONTENT_STATE 1
+#define XML_CDATA_SECTION_STATE 2
+#ifdef XML_DTD
+#define XML_IGNORE_SECTION_STATE 3
+#endif /* XML_DTD */
+
+#define XML_N_LITERAL_TYPES 2
+#define XML_ATTRIBUTE_VALUE_LITERAL 0
+#define XML_ENTITY_VALUE_LITERAL 1
+
+/* The size of the buffer passed to XmlUtf8Encode must be at least this. */
+#define XML_UTF8_ENCODE_MAX 4
+/* The size of the buffer passed to XmlUtf16Encode must be at least this. */
+#define XML_UTF16_ENCODE_MAX 2
+
+typedef struct position {
+  /* first line and first column are 0 not 1 */
+  unsigned long lineNumber;
+  unsigned long columnNumber;
+} POSITION;
+
+typedef struct {
+  const char *name;
+  const char *valuePtr;
+  const char *valueEnd;
+  char normalized;
+} ATTRIBUTE;
+
+struct encoding;
+typedef struct encoding ENCODING;
+
+struct encoding {
+  int (*scanners[XML_N_STATES])(const ENCODING *,
+			        const char *,
+			        const char *,
+			        const char **);
+  int (*literalScanners[XML_N_LITERAL_TYPES])(const ENCODING *,
+					      const char *,
+					      const char *,
+					      const char **);
+  int (*sameName)(const ENCODING *,
+	          const char *, const char *);
+  int (*nameMatchesAscii)(const ENCODING *,
+			  const char *, const char *, const char *);
+  int (*nameLength)(const ENCODING *, const char *);
+  const char *(*skipS)(const ENCODING *, const char *);
+  int (*getAtts)(const ENCODING *enc, const char *ptr,
+	         int attsMax, ATTRIBUTE *atts);
+  int (*charRefNumber)(const ENCODING *enc, const char *ptr);
+  int (*predefinedEntityName)(const ENCODING *, const char *, const char *);
+  void (*updatePosition)(const ENCODING *,
+			 const char *ptr,
+			 const char *end,
+			 POSITION *);
+  int (*isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
+		    const char **badPtr);
+  void (*utf8Convert)(const ENCODING *enc,
+		      const char **fromP,
+		      const char *fromLim,
+		      char **toP,
+		      const char *toLim);
+  void (*utf16Convert)(const ENCODING *enc,
+		       const char **fromP,
+		       const char *fromLim,
+		       unsigned short **toP,
+		       const unsigned short *toLim);
+  int minBytesPerChar;
+  char isUtf8;
+  char isUtf16;
+};
+
+/*
+Scan the string starting at ptr until the end of the next complete token,
+but do not scan past eptr.  Return an integer giving the type of token.
+
+Return XML_TOK_NONE when ptr == eptr; nextTokPtr will not be set.
+
+Return XML_TOK_PARTIAL when the string does not contain a complete token;
+nextTokPtr will not be set.
+
+Return XML_TOK_INVALID when the string does not start a valid token; nextTokPtr
+will be set to point to the character which made the token invalid.
+
+Otherwise the string starts with a valid token; nextTokPtr will be set to point
+to the character following the end of that token.
+
+Each data character counts as a single token, but adjacent data characters
+may be returned together.  Similarly for characters in the prolog outside
+literals, comments and processing instructions.
+*/
+
+
+#define XmlTok(enc, state, ptr, end, nextTokPtr) \
+  (((enc)->scanners[state])(enc, ptr, end, nextTokPtr))
+
+#define XmlPrologTok(enc, ptr, end, nextTokPtr) \
+   XmlTok(enc, XML_PROLOG_STATE, ptr, end, nextTokPtr)
+
+#define XmlContentTok(enc, ptr, end, nextTokPtr) \
+   XmlTok(enc, XML_CONTENT_STATE, ptr, end, nextTokPtr)
+
+#define XmlCdataSectionTok(enc, ptr, end, nextTokPtr) \
+   XmlTok(enc, XML_CDATA_SECTION_STATE, ptr, end, nextTokPtr)
+
+#ifdef XML_DTD
+
+#define XmlIgnoreSectionTok(enc, ptr, end, nextTokPtr) \
+   XmlTok(enc, XML_IGNORE_SECTION_STATE, ptr, end, nextTokPtr)
+
+#endif /* XML_DTD */
+
+/* This is used for performing a 2nd-level tokenization on
+the content of a literal that has already been returned by XmlTok. */ 
+
+#define XmlLiteralTok(enc, literalType, ptr, end, nextTokPtr) \
+  (((enc)->literalScanners[literalType])(enc, ptr, end, nextTokPtr))
+
+#define XmlAttributeValueTok(enc, ptr, end, nextTokPtr) \
+   XmlLiteralTok(enc, XML_ATTRIBUTE_VALUE_LITERAL, ptr, end, nextTokPtr)
+
+#define XmlEntityValueTok(enc, ptr, end, nextTokPtr) \
+   XmlLiteralTok(enc, XML_ENTITY_VALUE_LITERAL, ptr, end, nextTokPtr)
+
+#define XmlSameName(enc, ptr1, ptr2) (((enc)->sameName)(enc, ptr1, ptr2))
+
+#define XmlNameMatchesAscii(enc, ptr1, end1, ptr2) \
+  (((enc)->nameMatchesAscii)(enc, ptr1, end1, ptr2))
+
+#define XmlNameLength(enc, ptr) \
+  (((enc)->nameLength)(enc, ptr))
+
+#define XmlSkipS(enc, ptr) \
+  (((enc)->skipS)(enc, ptr))
+
+#define XmlGetAttributes(enc, ptr, attsMax, atts) \
+  (((enc)->getAtts)(enc, ptr, attsMax, atts))
+
+#define XmlCharRefNumber(enc, ptr) \
+  (((enc)->charRefNumber)(enc, ptr))
+
+#define XmlPredefinedEntityName(enc, ptr, end) \
+  (((enc)->predefinedEntityName)(enc, ptr, end))
+
+#define XmlUpdatePosition(enc, ptr, end, pos) \
+  (((enc)->updatePosition)(enc, ptr, end, pos))
+
+#define XmlIsPublicId(enc, ptr, end, badPtr) \
+  (((enc)->isPublicId)(enc, ptr, end, badPtr))
+
+#define XmlUtf8Convert(enc, fromP, fromLim, toP, toLim) \
+  (((enc)->utf8Convert)(enc, fromP, fromLim, toP, toLim))
+
+#define XmlUtf16Convert(enc, fromP, fromLim, toP, toLim) \
+  (((enc)->utf16Convert)(enc, fromP, fromLim, toP, toLim))
+
+typedef struct {
+  ENCODING initEnc;
+  const ENCODING **encPtr;
+} INIT_ENCODING;
+
+int  XmlParseXmlDecl(int isGeneralTextEntity,
+			      const ENCODING *enc,
+			      const char *ptr,
+	  		      const char *end,
+			      const char **badPtr,
+			      const char **versionPtr,
+			      const char **versionEndPtr,
+			      const char **encodingNamePtr,
+			      const ENCODING **namedEncodingPtr,
+			      int *standalonePtr);
+
+int  XmlInitEncoding(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING  *XmlGetUtf8InternalEncoding(void);
+const ENCODING  *XmlGetUtf16InternalEncoding(void);
+int  XmlUtf8Encode(int charNumber, char *buf);
+int  XmlUtf16Encode(int charNumber, unsigned short *buf);
+
+int  XmlSizeOfUnknownEncoding(void);
+ENCODING  *
+XmlInitUnknownEncoding(void *mem,
+		       int *table,
+		       int (*conv)(void *userData, const char *p),
+		       void *userData);
+
+int  XmlParseXmlDeclNS(int isGeneralTextEntity,
+			        const ENCODING *enc,
+			        const char *ptr,
+	  		        const char *end,
+			        const char **badPtr,
+			        const char **versionPtr,
+				const char **versionEndPtr,
+			        const char **encodingNamePtr,
+			        const ENCODING **namedEncodingPtr,
+			        int *standalonePtr);
+int  XmlInitEncodingNS(INIT_ENCODING *, const ENCODING **, const char *name);
+const ENCODING  *XmlGetUtf8InternalEncodingNS(void);
+const ENCODING  *XmlGetUtf16InternalEncodingNS(void);
+ENCODING  *
+XmlInitUnknownEncodingNS(void *mem,
+		         int *table,
+		         int (*conv)(void *userData, const char *p),
+		         void *userData);
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* not XmlTok_INCLUDED */
diff --git a/xml/expat/lib/xmltok_impl.c b/xml/expat/lib/xmltok_impl.c
new file mode 100644
index 000000000..36d2065ce
--- /dev/null
+++ b/xml/expat/lib/xmltok_impl.c
@@ -0,0 +1,1768 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+#ifndef IS_INVALID_CHAR
+#define IS_INVALID_CHAR(enc, ptr, n) (0)
+#endif
+
+#define INVALID_LEAD_CASE(n, ptr, nextTokPtr) \
+    case BT_LEAD ## n: \
+      if (end - ptr < n) \
+	return XML_TOK_PARTIAL_CHAR; \
+      if (IS_INVALID_CHAR(enc, ptr, n)) { \
+        *(nextTokPtr) = (ptr); \
+        return XML_TOK_INVALID; \
+      } \
+      ptr += n; \
+      break;
+
+#define INVALID_CASES(ptr, nextTokPtr) \
+  INVALID_LEAD_CASE(2, ptr, nextTokPtr) \
+  INVALID_LEAD_CASE(3, ptr, nextTokPtr) \
+  INVALID_LEAD_CASE(4, ptr, nextTokPtr) \
+  case BT_NONXML: \
+  case BT_MALFORM: \
+  case BT_TRAIL: \
+    *(nextTokPtr) = (ptr); \
+    return XML_TOK_INVALID;
+
+#define CHECK_NAME_CASE(n, enc, ptr, end, nextTokPtr) \
+   case BT_LEAD ## n: \
+     if (end - ptr < n) \
+       return XML_TOK_PARTIAL_CHAR; \
+     if (!IS_NAME_CHAR(enc, ptr, n)) { \
+       *nextTokPtr = ptr; \
+       return XML_TOK_INVALID; \
+     } \
+     ptr += n; \
+     break;
+
+#define CHECK_NAME_CASES(enc, ptr, end, nextTokPtr) \
+  case BT_NONASCII: \
+    if (!IS_NAME_CHAR_MINBPC(enc, ptr)) { \
+      *nextTokPtr = ptr; \
+      return XML_TOK_INVALID; \
+    } \
+  case BT_NMSTRT: \
+  case BT_HEX: \
+  case BT_DIGIT: \
+  case BT_NAME: \
+  case BT_MINUS: \
+    ptr += MINBPC(enc); \
+    break; \
+  CHECK_NAME_CASE(2, enc, ptr, end, nextTokPtr) \
+  CHECK_NAME_CASE(3, enc, ptr, end, nextTokPtr) \
+  CHECK_NAME_CASE(4, enc, ptr, end, nextTokPtr)
+
+#define CHECK_NMSTRT_CASE(n, enc, ptr, end, nextTokPtr) \
+   case BT_LEAD ## n: \
+     if (end - ptr < n) \
+       return XML_TOK_PARTIAL_CHAR; \
+     if (!IS_NMSTRT_CHAR(enc, ptr, n)) { \
+       *nextTokPtr = ptr; \
+       return XML_TOK_INVALID; \
+     } \
+     ptr += n; \
+     break;
+
+#define CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr) \
+  case BT_NONASCII: \
+    if (!IS_NMSTRT_CHAR_MINBPC(enc, ptr)) { \
+      *nextTokPtr = ptr; \
+      return XML_TOK_INVALID; \
+    } \
+  case BT_NMSTRT: \
+  case BT_HEX: \
+    ptr += MINBPC(enc); \
+    break; \
+  CHECK_NMSTRT_CASE(2, enc, ptr, end, nextTokPtr) \
+  CHECK_NMSTRT_CASE(3, enc, ptr, end, nextTokPtr) \
+  CHECK_NMSTRT_CASE(4, enc, ptr, end, nextTokPtr)
+
+#ifndef PREFIX
+#define PREFIX(ident) ident
+#endif
+
+/* ptr points to character following "<!-" */
+
+static
+int PREFIX(scanComment)(const ENCODING *enc, const char *ptr, const char *end,
+			const char **nextTokPtr)
+{
+  if (ptr != end) {
+    if (!CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+    ptr += MINBPC(enc);
+    while (ptr != end) {
+      switch (BYTE_TYPE(enc, ptr)) {
+      INVALID_CASES(ptr, nextTokPtr)
+      case BT_MINUS:
+	if ((ptr += MINBPC(enc)) == end)
+	  return XML_TOK_PARTIAL;
+	if (CHAR_MATCHES(enc, ptr, ASCII_MINUS)) {
+	  if ((ptr += MINBPC(enc)) == end)
+	    return XML_TOK_PARTIAL;
+	  if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	    *nextTokPtr = ptr;
+	    return XML_TOK_INVALID;
+	  }
+	  *nextTokPtr = ptr + MINBPC(enc);
+	  return XML_TOK_COMMENT;
+	}
+	break;
+      default:
+	ptr += MINBPC(enc);
+	break;
+      }
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "<!" */
+
+static
+int PREFIX(scanDecl)(const ENCODING *enc, const char *ptr, const char *end,
+		     const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  case BT_MINUS:
+    return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_LSQB:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_COND_SECT_OPEN;
+  case BT_NMSTRT:
+  case BT_HEX:
+    ptr += MINBPC(enc);
+    break;
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_PERCNT:
+      if (ptr + MINBPC(enc) == end)
+	return XML_TOK_PARTIAL;
+      /* don't allow <!ENTITY% foo "whatever"> */
+      switch (BYTE_TYPE(enc, ptr + MINBPC(enc))) {
+      case BT_S: case BT_CR: case BT_LF: case BT_PERCNT:
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      /* fall through */
+    case BT_S: case BT_CR: case BT_LF:
+      *nextTokPtr = ptr;
+      return XML_TOK_DECL_OPEN;
+    case BT_NMSTRT:
+    case BT_HEX:
+      ptr += MINBPC(enc);
+      break;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+static
+int PREFIX(checkPiTarget)(const ENCODING *enc, const char *ptr, const char *end, int *tokPtr)
+{
+  int upper = 0;
+  *tokPtr = XML_TOK_PI;
+  if (end - ptr != MINBPC(enc)*3)
+    return 1;
+  switch (BYTE_TO_ASCII(enc, ptr)) {
+  case ASCII_x:
+    break;
+  case ASCII_X:
+    upper = 1;
+    break;
+  default:
+    return 1;
+  }
+  ptr += MINBPC(enc);
+  switch (BYTE_TO_ASCII(enc, ptr)) {
+  case ASCII_m:
+    break;
+  case ASCII_M:
+    upper = 1;
+    break;
+  default:
+    return 1;
+  }
+  ptr += MINBPC(enc);
+  switch (BYTE_TO_ASCII(enc, ptr)) {
+  case ASCII_l:
+    break;
+  case ASCII_L:
+    upper = 1;
+    break;
+  default:
+    return 1;
+  }
+  if (upper)
+    return 0;
+  *tokPtr = XML_TOK_XML_DECL;
+  return 1;
+}
+
+/* ptr points to character following "<?" */
+
+static
+int PREFIX(scanPi)(const ENCODING *enc, const char *ptr, const char *end,
+		   const char **nextTokPtr)
+{
+  int tok;
+  const char *target = ptr;
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_S: case BT_CR: case BT_LF:
+      if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      ptr += MINBPC(enc);
+      while (ptr != end) {
+        switch (BYTE_TYPE(enc, ptr)) {
+        INVALID_CASES(ptr, nextTokPtr)
+	case BT_QUEST:
+	  ptr += MINBPC(enc);
+	  if (ptr == end)
+	    return XML_TOK_PARTIAL;
+	  if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	    *nextTokPtr = ptr + MINBPC(enc);
+	    return tok;
+	  }
+	  break;
+	default:
+	  ptr += MINBPC(enc);
+	  break;
+	}
+      }
+      return XML_TOK_PARTIAL;
+    case BT_QUEST:
+      if (!PREFIX(checkPiTarget)(enc, target, ptr, &tok)) {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	return XML_TOK_PARTIAL;
+      if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	*nextTokPtr = ptr + MINBPC(enc);
+	return tok;
+      }
+      /* fall through */
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+
+static
+int PREFIX(scanCdataSection)(const ENCODING *enc, const char *ptr, const char *end,
+			     const char **nextTokPtr)
+{
+  static const char CDATA_LSQB[] = { ASCII_C, ASCII_D, ASCII_A, ASCII_T, ASCII_A, ASCII_LSQB };
+  int i;
+  /* CDATA[ */
+  if (end - ptr < 6 * MINBPC(enc))
+    return XML_TOK_PARTIAL;
+  for (i = 0; i < 6; i++, ptr += MINBPC(enc)) {
+    if (!CHAR_MATCHES(enc, ptr, CDATA_LSQB[i])) {
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  *nextTokPtr = ptr;
+  return XML_TOK_CDATA_SECT_OPEN;
+}
+
+static
+int PREFIX(cdataSectionTok)(const ENCODING *enc, const char *ptr, const char *end,
+			    const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_NONE;
+  if (MINBPC(enc) > 1) {
+    size_t n = end - ptr;
+    if (n & (MINBPC(enc) - 1)) {
+      n &= ~(MINBPC(enc) - 1);
+      if (n == 0)
+	return XML_TOK_PARTIAL;
+      end = ptr + n;
+    }
+  }
+  switch (BYTE_TYPE(enc, ptr)) {
+  case BT_RSQB:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_PARTIAL;
+    if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
+      break;
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_PARTIAL;
+    if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+      ptr -= MINBPC(enc);
+      break;
+    }
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_CDATA_SECT_CLOSE;
+  case BT_CR:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_PARTIAL;
+    if (BYTE_TYPE(enc, ptr) == BT_LF)
+      ptr += MINBPC(enc);
+    *nextTokPtr = ptr;
+    return XML_TOK_DATA_NEWLINE;
+  case BT_LF:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_DATA_NEWLINE;
+  INVALID_CASES(ptr, nextTokPtr)
+  default:
+    ptr += MINBPC(enc);
+    break;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: \
+      if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
+	*nextTokPtr = ptr; \
+	return XML_TOK_DATA_CHARS; \
+      } \
+      ptr += n; \
+      break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_NONXML:
+    case BT_MALFORM:
+    case BT_TRAIL:
+    case BT_CR:
+    case BT_LF:
+    case BT_RSQB:
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  *nextTokPtr = ptr;
+  return XML_TOK_DATA_CHARS;
+}
+
+/* ptr points to character following "</" */
+
+static
+int PREFIX(scanEndTag)(const ENCODING *enc, const char *ptr, const char *end,
+		       const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_S: case BT_CR: case BT_LF:
+      for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+	switch (BYTE_TYPE(enc, ptr)) {
+	case BT_S: case BT_CR: case BT_LF:
+	  break;
+	case BT_GT:
+	  *nextTokPtr = ptr + MINBPC(enc);
+          return XML_TOK_END_TAG;
+	default:
+	  *nextTokPtr = ptr;
+	  return XML_TOK_INVALID;
+	}
+      }
+      return XML_TOK_PARTIAL;
+#ifdef XML_NS
+    case BT_COLON:
+      /* no need to check qname syntax here, since end-tag must match exactly */
+      ptr += MINBPC(enc);
+      break;
+#endif
+    case BT_GT:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_END_TAG;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&#X" */
+
+static
+int PREFIX(scanHexCharRef)(const ENCODING *enc, const char *ptr, const char *end,
+			   const char **nextTokPtr)
+{
+  if (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_DIGIT:
+    case BT_HEX:
+      break;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+    for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+      switch (BYTE_TYPE(enc, ptr)) {
+      case BT_DIGIT:
+      case BT_HEX:
+	break;
+      case BT_SEMI:
+	*nextTokPtr = ptr + MINBPC(enc);
+	return XML_TOK_CHAR_REF;
+      default:
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&#" */
+
+static
+int PREFIX(scanCharRef)(const ENCODING *enc, const char *ptr, const char *end,
+			const char **nextTokPtr)
+{
+  if (ptr != end) {
+    if (CHAR_MATCHES(enc, ptr, ASCII_x))
+      return PREFIX(scanHexCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_DIGIT:
+      break;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+    for (ptr += MINBPC(enc); ptr != end; ptr += MINBPC(enc)) {
+      switch (BYTE_TYPE(enc, ptr)) {
+      case BT_DIGIT:
+	break;
+      case BT_SEMI:
+	*nextTokPtr = ptr + MINBPC(enc);
+	return XML_TOK_CHAR_REF;
+      default:
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "&" */
+
+static
+int PREFIX(scanRef)(const ENCODING *enc, const char *ptr, const char *end,
+		    const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  case BT_NUM:
+    return PREFIX(scanCharRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_SEMI:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_ENTITY_REF;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following first character of attribute name */
+
+static
+int PREFIX(scanAtts)(const ENCODING *enc, const char *ptr, const char *end,
+		     const char **nextTokPtr)
+{
+#ifdef XML_NS
+  int hadColon = 0;
+#endif
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+#ifdef XML_NS
+    case BT_COLON:
+      if (hadColon) {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      hadColon = 1;
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	return XML_TOK_PARTIAL;
+      switch (BYTE_TYPE(enc, ptr)) {
+      CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+      default:
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      break;
+#endif
+    case BT_S: case BT_CR: case BT_LF:
+      for (;;) {
+	int t;
+
+	ptr += MINBPC(enc);
+	if (ptr == end)
+	  return XML_TOK_PARTIAL;
+	t = BYTE_TYPE(enc, ptr);
+	if (t == BT_EQUALS)
+	  break;
+	switch (t) {
+	case BT_S:
+	case BT_LF:
+	case BT_CR:
+	  break;
+	default:
+	  *nextTokPtr = ptr;
+	  return XML_TOK_INVALID;
+	}
+      }
+    /* fall through */
+    case BT_EQUALS:
+      {
+	int open;
+#ifdef XML_NS
+	hadColon = 0;
+#endif
+	for (;;) {
+	  
+	  ptr += MINBPC(enc);
+	  if (ptr == end)
+	    return XML_TOK_PARTIAL;
+	  open = BYTE_TYPE(enc, ptr);
+	  if (open == BT_QUOT || open == BT_APOS)
+	    break;
+	  switch (open) {
+	  case BT_S:
+	  case BT_LF:
+	  case BT_CR:
+	    break;
+	  default:
+	    *nextTokPtr = ptr;
+	    return XML_TOK_INVALID;
+	  }
+	}
+	ptr += MINBPC(enc);
+	/* in attribute value */
+	for (;;) {
+	  int t;
+	  if (ptr == end)
+	    return XML_TOK_PARTIAL;
+	  t = BYTE_TYPE(enc, ptr);
+	  if (t == open)
+	    break;
+	  switch (t) {
+	  INVALID_CASES(ptr, nextTokPtr)
+	  case BT_AMP:
+	    {
+	      int tok = PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, &ptr);
+	      if (tok <= 0) {
+		if (tok == XML_TOK_INVALID)
+		  *nextTokPtr = ptr;
+		return tok;
+	      }
+	      break;
+	    }
+	  case BT_LT:
+	    *nextTokPtr = ptr;
+	    return XML_TOK_INVALID;
+	  default:
+	    ptr += MINBPC(enc);
+	    break;
+	  }
+	}
+	ptr += MINBPC(enc);
+	if (ptr == end)
+	  return XML_TOK_PARTIAL;
+	switch (BYTE_TYPE(enc, ptr)) {
+	case BT_S:
+	case BT_CR:
+	case BT_LF:
+	  break;
+	case BT_SOL:
+	  goto sol;
+	case BT_GT:
+	  goto gt;
+	default:
+	  *nextTokPtr = ptr;
+	  return XML_TOK_INVALID;
+	}
+	/* ptr points to closing quote */
+	for (;;) {
+	  ptr += MINBPC(enc);
+	  if (ptr == end)
+	    return XML_TOK_PARTIAL;
+	  switch (BYTE_TYPE(enc, ptr)) {
+	  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+	  case BT_S: case BT_CR: case BT_LF:
+	    continue;
+	  case BT_GT:
+          gt:
+	    *nextTokPtr = ptr + MINBPC(enc);
+	    return XML_TOK_START_TAG_WITH_ATTS;
+	  case BT_SOL:
+          sol:
+	    ptr += MINBPC(enc);
+	    if (ptr == end)
+	      return XML_TOK_PARTIAL;
+	    if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	      *nextTokPtr = ptr;
+	      return XML_TOK_INVALID;
+	    }
+	    *nextTokPtr = ptr + MINBPC(enc);
+	    return XML_TOK_EMPTY_ELEMENT_WITH_ATTS;
+	  default:
+	    *nextTokPtr = ptr;
+	    return XML_TOK_INVALID;
+	  }
+	  break;
+	}
+	break;
+      }
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+/* ptr points to character following "<" */
+
+static
+int PREFIX(scanLt)(const ENCODING *enc, const char *ptr, const char *end,
+		   const char **nextTokPtr)
+{
+#ifdef XML_NS
+  int hadColon;
+#endif
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  case BT_EXCL:
+    if ((ptr += MINBPC(enc)) == end)
+      return XML_TOK_PARTIAL;
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_MINUS:
+      return PREFIX(scanComment)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+    case BT_LSQB:
+      return PREFIX(scanCdataSection)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+    }
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  case BT_QUEST:
+    return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_SOL:
+    return PREFIX(scanEndTag)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+#ifdef XML_NS
+  hadColon = 0;
+#endif
+  /* we have a start-tag */
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+#ifdef XML_NS
+    case BT_COLON:
+      if (hadColon) {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      hadColon = 1;
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	return XML_TOK_PARTIAL;
+      switch (BYTE_TYPE(enc, ptr)) {
+      CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+      default:
+        *nextTokPtr = ptr;
+        return XML_TOK_INVALID;
+      }
+      break;
+#endif
+    case BT_S: case BT_CR: case BT_LF:
+      {
+        ptr += MINBPC(enc);
+	while (ptr != end) {
+	  switch (BYTE_TYPE(enc, ptr)) {
+	  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+	  case BT_GT:
+	    goto gt;
+	  case BT_SOL:
+	    goto sol;
+	  case BT_S: case BT_CR: case BT_LF:
+	    ptr += MINBPC(enc);
+	    continue;
+	  default:
+	    *nextTokPtr = ptr;
+	    return XML_TOK_INVALID;
+	  }
+	  return PREFIX(scanAtts)(enc, ptr, end, nextTokPtr);
+	}
+	return XML_TOK_PARTIAL;
+      }
+    case BT_GT:
+    gt:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_START_TAG_NO_ATTS;
+    case BT_SOL:
+    sol:
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	return XML_TOK_PARTIAL;
+      if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_EMPTY_ELEMENT_NO_ATTS;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+static
+int PREFIX(contentTok)(const ENCODING *enc, const char *ptr, const char *end,
+		       const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_NONE;
+  if (MINBPC(enc) > 1) {
+    size_t n = end - ptr;
+    if (n & (MINBPC(enc) - 1)) {
+      n &= ~(MINBPC(enc) - 1);
+      if (n == 0)
+	return XML_TOK_PARTIAL;
+      end = ptr + n;
+    }
+  }
+  switch (BYTE_TYPE(enc, ptr)) {
+  case BT_LT:
+    return PREFIX(scanLt)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_AMP:
+    return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_CR:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_TRAILING_CR;
+    if (BYTE_TYPE(enc, ptr) == BT_LF)
+      ptr += MINBPC(enc);
+    *nextTokPtr = ptr;
+    return XML_TOK_DATA_NEWLINE;
+  case BT_LF:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_DATA_NEWLINE;
+  case BT_RSQB:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_TRAILING_RSQB;
+    if (!CHAR_MATCHES(enc, ptr, ASCII_RSQB))
+      break;
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return XML_TOK_TRAILING_RSQB;
+    if (!CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+      ptr -= MINBPC(enc);
+      break;
+    }
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  INVALID_CASES(ptr, nextTokPtr)
+  default:
+    ptr += MINBPC(enc);
+    break;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: \
+      if (end - ptr < n || IS_INVALID_CHAR(enc, ptr, n)) { \
+	*nextTokPtr = ptr; \
+	return XML_TOK_DATA_CHARS; \
+      } \
+      ptr += n; \
+      break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_RSQB:
+      if (ptr + MINBPC(enc) != end) {
+	 if (!CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_RSQB)) {
+	   ptr += MINBPC(enc);
+	   break;
+	 }
+	 if (ptr + 2*MINBPC(enc) != end) {
+	   if (!CHAR_MATCHES(enc, ptr + 2*MINBPC(enc), ASCII_GT)) {
+	     ptr += MINBPC(enc);
+	     break;
+	   }
+	   *nextTokPtr = ptr + 2*MINBPC(enc);
+	   return XML_TOK_INVALID;
+	 }
+      }
+      /* fall through */
+    case BT_AMP:
+    case BT_LT:
+    case BT_NONXML:
+    case BT_MALFORM:
+    case BT_TRAIL:
+    case BT_CR:
+    case BT_LF:
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  *nextTokPtr = ptr;
+  return XML_TOK_DATA_CHARS;
+}
+
+/* ptr points to character following "%" */
+
+static
+int PREFIX(scanPercent)(const ENCODING *enc, const char *ptr, const char *end,
+			const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  case BT_S: case BT_LF: case BT_CR: case BT_PERCNT:
+    *nextTokPtr = ptr;
+    return XML_TOK_PERCENT;
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_SEMI:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_PARAM_ENTITY_REF;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+static
+int PREFIX(scanPoundName)(const ENCODING *enc, const char *ptr, const char *end,
+			  const char **nextTokPtr)
+{
+  if (ptr == end)
+    return XML_TOK_PARTIAL;
+  switch (BYTE_TYPE(enc, ptr)) {
+  CHECK_NMSTRT_CASES(enc, ptr, end, nextTokPtr)
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_CR: case BT_LF: case BT_S:
+    case BT_RPAR: case BT_GT: case BT_PERCNT: case BT_VERBAR:
+      *nextTokPtr = ptr;
+      return XML_TOK_POUND_NAME;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return -XML_TOK_POUND_NAME;
+}
+
+static
+int PREFIX(scanLit)(int open, const ENCODING *enc,
+		    const char *ptr, const char *end,
+		    const char **nextTokPtr)
+{
+  while (ptr != end) {
+    int t = BYTE_TYPE(enc, ptr);
+    switch (t) {
+    INVALID_CASES(ptr, nextTokPtr)
+    case BT_QUOT:
+    case BT_APOS:
+      ptr += MINBPC(enc);
+      if (t != open)
+	break;
+      if (ptr == end)
+	return -XML_TOK_LITERAL;
+      *nextTokPtr = ptr;
+      switch (BYTE_TYPE(enc, ptr)) {
+      case BT_S: case BT_CR: case BT_LF:
+      case BT_GT: case BT_PERCNT: case BT_LSQB:
+	return XML_TOK_LITERAL;
+      default:
+	return XML_TOK_INVALID;
+      }
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+static
+int PREFIX(prologTok)(const ENCODING *enc, const char *ptr, const char *end,
+		      const char **nextTokPtr)
+{
+  int tok;
+  if (ptr == end)
+    return XML_TOK_NONE;
+  if (MINBPC(enc) > 1) {
+    size_t n = end - ptr;
+    if (n & (MINBPC(enc) - 1)) {
+      n &= ~(MINBPC(enc) - 1);
+      if (n == 0)
+	return XML_TOK_PARTIAL;
+      end = ptr + n;
+    }
+  }
+  switch (BYTE_TYPE(enc, ptr)) {
+  case BT_QUOT:
+    return PREFIX(scanLit)(BT_QUOT, enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_APOS:
+    return PREFIX(scanLit)(BT_APOS, enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_LT:
+    {
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	return XML_TOK_PARTIAL;
+      switch (BYTE_TYPE(enc, ptr)) {
+      case BT_EXCL:
+	return PREFIX(scanDecl)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+      case BT_QUEST:
+	return PREFIX(scanPi)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+      case BT_NMSTRT:
+      case BT_HEX:
+      case BT_NONASCII:
+      case BT_LEAD2:
+      case BT_LEAD3:
+      case BT_LEAD4:
+	*nextTokPtr = ptr - MINBPC(enc);
+	return XML_TOK_INSTANCE_START;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  case BT_CR:
+    if (ptr + MINBPC(enc) == end)
+      return -XML_TOK_PROLOG_S;
+    /* fall through */
+  case BT_S: case BT_LF:
+    for (;;) {
+      ptr += MINBPC(enc);
+      if (ptr == end)
+	break;
+      switch (BYTE_TYPE(enc, ptr)) {
+      case BT_S: case BT_LF:
+	break;
+      case BT_CR:
+	/* don't split CR/LF pair */
+	if (ptr + MINBPC(enc) != end)
+	  break;
+	/* fall through */
+      default:
+	*nextTokPtr = ptr;
+	return XML_TOK_PROLOG_S;
+      }
+    }
+    *nextTokPtr = ptr;
+    return XML_TOK_PROLOG_S;
+  case BT_PERCNT:
+    return PREFIX(scanPercent)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+  case BT_COMMA:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_COMMA;
+  case BT_LSQB:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_OPEN_BRACKET;
+  case BT_RSQB:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return -XML_TOK_CLOSE_BRACKET;
+    if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
+      if (ptr + MINBPC(enc) == end)
+	return XML_TOK_PARTIAL;
+      if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_GT)) {
+	*nextTokPtr = ptr + 2*MINBPC(enc);
+	return XML_TOK_COND_SECT_CLOSE;
+      }
+    }
+    *nextTokPtr = ptr;
+    return XML_TOK_CLOSE_BRACKET;
+  case BT_LPAR:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_OPEN_PAREN;
+  case BT_RPAR:
+    ptr += MINBPC(enc);
+    if (ptr == end)
+      return -XML_TOK_CLOSE_PAREN;
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_AST:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_CLOSE_PAREN_ASTERISK;
+    case BT_QUEST:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_CLOSE_PAREN_QUESTION;
+    case BT_PLUS:
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_CLOSE_PAREN_PLUS;
+    case BT_CR: case BT_LF: case BT_S:
+    case BT_GT: case BT_COMMA: case BT_VERBAR:
+    case BT_RPAR:
+      *nextTokPtr = ptr;
+      return XML_TOK_CLOSE_PAREN;
+    }
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  case BT_VERBAR:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_OR;
+  case BT_GT:
+    *nextTokPtr = ptr + MINBPC(enc);
+    return XML_TOK_DECL_CLOSE;
+  case BT_NUM:
+    return PREFIX(scanPoundName)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+#define LEAD_CASE(n) \
+  case BT_LEAD ## n: \
+    if (end - ptr < n) \
+      return XML_TOK_PARTIAL_CHAR; \
+    if (IS_NMSTRT_CHAR(enc, ptr, n)) { \
+      ptr += n; \
+      tok = XML_TOK_NAME; \
+      break; \
+    } \
+    if (IS_NAME_CHAR(enc, ptr, n)) { \
+      ptr += n; \
+      tok = XML_TOK_NMTOKEN; \
+      break; \
+    } \
+    *nextTokPtr = ptr; \
+    return XML_TOK_INVALID;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+  case BT_NMSTRT:
+  case BT_HEX:
+    tok = XML_TOK_NAME;
+    ptr += MINBPC(enc);
+    break;
+  case BT_DIGIT:
+  case BT_NAME:
+  case BT_MINUS:
+#ifdef XML_NS
+  case BT_COLON:
+#endif
+    tok = XML_TOK_NMTOKEN;
+    ptr += MINBPC(enc);
+    break;
+  case BT_NONASCII:
+    if (IS_NMSTRT_CHAR_MINBPC(enc, ptr)) {
+      ptr += MINBPC(enc);
+      tok = XML_TOK_NAME;
+      break;
+    }
+    if (IS_NAME_CHAR_MINBPC(enc, ptr)) {
+      ptr += MINBPC(enc);
+      tok = XML_TOK_NMTOKEN;
+      break;
+    }
+    /* fall through */
+  default:
+    *nextTokPtr = ptr;
+    return XML_TOK_INVALID;
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+    case BT_GT: case BT_RPAR: case BT_COMMA:
+    case BT_VERBAR: case BT_LSQB: case BT_PERCNT:
+    case BT_S: case BT_CR: case BT_LF:
+      *nextTokPtr = ptr;
+      return tok;
+#ifdef XML_NS
+    case BT_COLON:
+      ptr += MINBPC(enc);
+      switch (tok) {
+      case XML_TOK_NAME:
+	if (ptr == end)
+	  return XML_TOK_PARTIAL;
+	tok = XML_TOK_PREFIXED_NAME;
+	switch (BYTE_TYPE(enc, ptr)) {
+	CHECK_NAME_CASES(enc, ptr, end, nextTokPtr)
+	default:
+	  tok = XML_TOK_NMTOKEN;
+	  break;
+	}
+	break;
+      case XML_TOK_PREFIXED_NAME:
+	tok = XML_TOK_NMTOKEN;
+	break;
+      }
+      break;
+#endif
+    case BT_PLUS:
+      if (tok == XML_TOK_NMTOKEN)  {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_NAME_PLUS;
+    case BT_AST:
+      if (tok == XML_TOK_NMTOKEN)  {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_NAME_ASTERISK;
+    case BT_QUEST:
+      if (tok == XML_TOK_NMTOKEN)  {
+	*nextTokPtr = ptr;
+	return XML_TOK_INVALID;
+      }
+      *nextTokPtr = ptr + MINBPC(enc);
+      return XML_TOK_NAME_QUESTION;
+    default:
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    }
+  }
+  return -tok;
+}
+
+static
+int PREFIX(attributeValueTok)(const ENCODING *enc, const char *ptr, const char *end,
+			      const char **nextTokPtr)
+{
+  const char *start;
+  if (ptr == end)
+    return XML_TOK_NONE;
+  start = ptr;
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: ptr += n; break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_AMP:
+      if (ptr == start)
+	return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_LT:
+      /* this is for inside entity references */
+      *nextTokPtr = ptr;
+      return XML_TOK_INVALID;
+    case BT_LF:
+      if (ptr == start) {
+	*nextTokPtr = ptr + MINBPC(enc);
+	return XML_TOK_DATA_NEWLINE;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_CR:
+      if (ptr == start) {
+	ptr += MINBPC(enc);
+	if (ptr == end)
+	  return XML_TOK_TRAILING_CR;
+	if (BYTE_TYPE(enc, ptr) == BT_LF)
+	  ptr += MINBPC(enc);
+	*nextTokPtr = ptr;
+	return XML_TOK_DATA_NEWLINE;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_S:
+      if (ptr == start) {
+	*nextTokPtr = ptr + MINBPC(enc);
+	return XML_TOK_ATTRIBUTE_VALUE_S;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  *nextTokPtr = ptr;
+  return XML_TOK_DATA_CHARS;
+}
+
+static
+int PREFIX(entityValueTok)(const ENCODING *enc, const char *ptr, const char *end,
+			   const char **nextTokPtr)
+{
+  const char *start;
+  if (ptr == end)
+    return XML_TOK_NONE;
+  start = ptr;
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: ptr += n; break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_AMP:
+      if (ptr == start)
+	return PREFIX(scanRef)(enc, ptr + MINBPC(enc), end, nextTokPtr);
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_PERCNT:
+      if (ptr == start) {
+	int tok =  PREFIX(scanPercent)(enc, ptr + MINBPC(enc),
+				       end, nextTokPtr);
+	return (tok == XML_TOK_PERCENT) ? XML_TOK_INVALID : tok;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_LF:
+      if (ptr == start) {
+	*nextTokPtr = ptr + MINBPC(enc);
+	return XML_TOK_DATA_NEWLINE;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    case BT_CR:
+      if (ptr == start) {
+	ptr += MINBPC(enc);
+	if (ptr == end)
+	  return XML_TOK_TRAILING_CR;
+	if (BYTE_TYPE(enc, ptr) == BT_LF)
+	  ptr += MINBPC(enc);
+	*nextTokPtr = ptr;
+	return XML_TOK_DATA_NEWLINE;
+      }
+      *nextTokPtr = ptr;
+      return XML_TOK_DATA_CHARS;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  *nextTokPtr = ptr;
+  return XML_TOK_DATA_CHARS;
+}
+
+#ifdef XML_DTD
+
+static
+int PREFIX(ignoreSectionTok)(const ENCODING *enc, const char *ptr, const char *end,
+			     const char **nextTokPtr)
+{
+  int level = 0;
+  if (MINBPC(enc) > 1) {
+    size_t n = end - ptr;
+    if (n & (MINBPC(enc) - 1)) {
+      n &= ~(MINBPC(enc) - 1);
+      end = ptr + n;
+    }
+  }
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    INVALID_CASES(ptr, nextTokPtr)
+    case BT_LT:
+      if ((ptr += MINBPC(enc)) == end)
+	return XML_TOK_PARTIAL;
+      if (CHAR_MATCHES(enc, ptr, ASCII_EXCL)) {
+	if ((ptr += MINBPC(enc)) == end)
+	  return XML_TOK_PARTIAL;
+	if (CHAR_MATCHES(enc, ptr, ASCII_LSQB)) {
+	  ++level;
+	  ptr += MINBPC(enc);
+	}
+      }
+      break;
+    case BT_RSQB:
+      if ((ptr += MINBPC(enc)) == end)
+	return XML_TOK_PARTIAL;
+      if (CHAR_MATCHES(enc, ptr, ASCII_RSQB)) {
+	if ((ptr += MINBPC(enc)) == end)
+	  return XML_TOK_PARTIAL;
+	if (CHAR_MATCHES(enc, ptr, ASCII_GT)) {
+	  ptr += MINBPC(enc);
+	  if (level == 0) {
+	    *nextTokPtr = ptr;
+	    return XML_TOK_IGNORE_SECT;
+	  }
+	  --level;
+	}
+      }
+      break;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+  }
+  return XML_TOK_PARTIAL;
+}
+
+#endif /* XML_DTD */
+
+static
+int PREFIX(isPublicId)(const ENCODING *enc, const char *ptr, const char *end,
+		       const char **badPtr)
+{
+  ptr += MINBPC(enc);
+  end -= MINBPC(enc);
+  for (; ptr != end; ptr += MINBPC(enc)) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_DIGIT:
+    case BT_HEX:
+    case BT_MINUS:
+    case BT_APOS:
+    case BT_LPAR:
+    case BT_RPAR:
+    case BT_PLUS:
+    case BT_COMMA:
+    case BT_SOL:
+    case BT_EQUALS:
+    case BT_QUEST:
+    case BT_CR:
+    case BT_LF:
+    case BT_SEMI:
+    case BT_EXCL:
+    case BT_AST:
+    case BT_PERCNT:
+    case BT_NUM:
+#ifdef XML_NS
+    case BT_COLON:
+#endif
+      break;
+    case BT_S:
+      if (CHAR_MATCHES(enc, ptr, ASCII_TAB)) {
+	*badPtr = ptr;
+	return 0;
+      }
+      break;
+    case BT_NAME:
+    case BT_NMSTRT:
+      if (!(BYTE_TO_ASCII(enc, ptr) & ~0x7f))
+	break;
+    default:
+      switch (BYTE_TO_ASCII(enc, ptr)) {
+      case 0x24: /* $ */
+      case 0x40: /* @ */
+	break;
+      default:
+	*badPtr = ptr;
+	return 0;
+      }
+      break;
+    }
+  }
+  return 1;
+}
+
+/* This must only be called for a well-formed start-tag or empty element tag.
+Returns the number of attributes.  Pointers to the first attsMax attributes 
+are stored in atts. */
+
+static
+int PREFIX(getAtts)(const ENCODING *enc, const char *ptr,
+		    int attsMax, ATTRIBUTE *atts)
+{
+  enum { other, inName, inValue } state = inName;
+  int nAtts = 0;
+  int open = 0; /* defined when state == inValue;
+		   initialization just to shut up compilers */
+
+  for (ptr += MINBPC(enc);; ptr += MINBPC(enc)) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define START_NAME \
+      if (state == other) { \
+	if (nAtts < attsMax) { \
+	  atts[nAtts].name = ptr; \
+	  atts[nAtts].normalized = 1; \
+	} \
+	state = inName; \
+      }
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: START_NAME ptr += (n - MINBPC(enc)); break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_NONASCII:
+    case BT_NMSTRT:
+    case BT_HEX:
+      START_NAME
+      break;
+#undef START_NAME
+    case BT_QUOT:
+      if (state != inValue) {
+	if (nAtts < attsMax)
+	  atts[nAtts].valuePtr = ptr + MINBPC(enc);
+        state = inValue;
+        open = BT_QUOT;
+      }
+      else if (open == BT_QUOT) {
+        state = other;
+	if (nAtts < attsMax)
+	  atts[nAtts].valueEnd = ptr;
+	nAtts++;
+      }
+      break;
+    case BT_APOS:
+      if (state != inValue) {
+	if (nAtts < attsMax)
+	  atts[nAtts].valuePtr = ptr + MINBPC(enc);
+        state = inValue;
+        open = BT_APOS;
+      }
+      else if (open == BT_APOS) {
+        state = other;
+	if (nAtts < attsMax)
+	  atts[nAtts].valueEnd = ptr;
+	nAtts++;
+      }
+      break;
+    case BT_AMP:
+      if (nAtts < attsMax)
+	atts[nAtts].normalized = 0;
+      break;
+    case BT_S:
+      if (state == inName)
+        state = other;
+      else if (state == inValue
+	       && nAtts < attsMax
+	       && atts[nAtts].normalized
+	       && (ptr == atts[nAtts].valuePtr
+		   || BYTE_TO_ASCII(enc, ptr) != ASCII_SPACE
+		   || BYTE_TO_ASCII(enc, ptr + MINBPC(enc)) == ASCII_SPACE
+	           || BYTE_TYPE(enc, ptr + MINBPC(enc)) == open))
+	atts[nAtts].normalized = 0;
+      break;
+    case BT_CR: case BT_LF:
+      /* This case ensures that the first attribute name is counted
+         Apart from that we could just change state on the quote. */
+      if (state == inName)
+        state = other;
+      else if (state == inValue && nAtts < attsMax)
+	atts[nAtts].normalized = 0;
+      break;
+    case BT_GT:
+    case BT_SOL:
+      if (state != inValue)
+	return nAtts;
+      break;
+    default:
+      break;
+    }
+  }
+  /* not reached */
+}
+
+static
+int PREFIX(charRefNumber)(const ENCODING *enc, const char *ptr)
+{
+  int result = 0;
+  /* skip &# */
+  ptr += 2*MINBPC(enc);
+  if (CHAR_MATCHES(enc, ptr, ASCII_x)) {
+    for (ptr += MINBPC(enc); !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) {
+      int c = BYTE_TO_ASCII(enc, ptr);
+      switch (c) {
+      case ASCII_0: case ASCII_1: case ASCII_2: case ASCII_3: case ASCII_4:
+      case ASCII_5: case ASCII_6: case ASCII_7: case ASCII_8: case ASCII_9:
+	result <<= 4;
+	result |= (c - ASCII_0);
+	break;
+      case ASCII_A: case ASCII_B: case ASCII_C: case ASCII_D: case ASCII_E: case ASCII_F:
+	result <<= 4;
+	result += 10 + (c - ASCII_A);
+	break;
+      case ASCII_a: case ASCII_b: case ASCII_c: case ASCII_d: case ASCII_e: case ASCII_f:
+	result <<= 4;
+	result += 10 + (c - ASCII_a);
+	break;
+      }
+      if (result >= 0x110000)
+	return -1;
+    }
+  }
+  else {
+    for (; !CHAR_MATCHES(enc, ptr, ASCII_SEMI); ptr += MINBPC(enc)) {
+      int c = BYTE_TO_ASCII(enc, ptr);
+      result *= 10;
+      result += (c - ASCII_0);
+      if (result >= 0x110000)
+	return -1;
+    }
+  }
+  return checkCharRefNumber(result);
+}
+
+static
+int PREFIX(predefinedEntityName)(const ENCODING *enc, const char *ptr, const char *end)
+{
+  switch ((end - ptr)/MINBPC(enc)) {
+  case 2:
+    if (CHAR_MATCHES(enc, ptr + MINBPC(enc), ASCII_t)) {
+      switch (BYTE_TO_ASCII(enc, ptr)) {
+      case ASCII_l:
+	return ASCII_LT;
+      case ASCII_g:
+	return ASCII_GT;
+      }
+    }
+    break;
+  case 3:
+    if (CHAR_MATCHES(enc, ptr, ASCII_a)) {
+      ptr += MINBPC(enc);
+      if (CHAR_MATCHES(enc, ptr, ASCII_m)) {
+	ptr += MINBPC(enc);
+	if (CHAR_MATCHES(enc, ptr, ASCII_p))
+	  return ASCII_AMP;
+      }
+    }
+    break;
+  case 4:
+    switch (BYTE_TO_ASCII(enc, ptr)) {
+    case ASCII_q:
+      ptr += MINBPC(enc);
+      if (CHAR_MATCHES(enc, ptr, ASCII_u)) {
+	ptr += MINBPC(enc);
+	if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
+	  ptr += MINBPC(enc);
+  	  if (CHAR_MATCHES(enc, ptr, ASCII_t))
+	    return ASCII_QUOT;
+	}
+      }
+      break;
+    case ASCII_a:
+      ptr += MINBPC(enc);
+      if (CHAR_MATCHES(enc, ptr, ASCII_p)) {
+	ptr += MINBPC(enc);
+	if (CHAR_MATCHES(enc, ptr, ASCII_o)) {
+	  ptr += MINBPC(enc);
+  	  if (CHAR_MATCHES(enc, ptr, ASCII_s))
+	    return ASCII_APOS;
+	}
+      }
+      break;
+    }
+  }
+  return 0;
+}
+
+static
+int PREFIX(sameName)(const ENCODING *enc, const char *ptr1, const char *ptr2)
+{
+  for (;;) {
+    switch (BYTE_TYPE(enc, ptr1)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: \
+      if (*ptr1++ != *ptr2++) \
+	return 0;
+    LEAD_CASE(4) LEAD_CASE(3) LEAD_CASE(2)
+#undef LEAD_CASE
+      /* fall through */
+      if (*ptr1++ != *ptr2++)
+	return 0;
+      break;
+    case BT_NONASCII:
+    case BT_NMSTRT:
+#ifdef XML_NS
+    case BT_COLON:
+#endif
+    case BT_HEX:
+    case BT_DIGIT:
+    case BT_NAME:
+    case BT_MINUS:
+      if (*ptr2++ != *ptr1++)
+	return 0;
+      if (MINBPC(enc) > 1) {
+	if (*ptr2++ != *ptr1++)
+	  return 0;
+	if (MINBPC(enc) > 2) {
+	  if (*ptr2++ != *ptr1++)
+	    return 0;
+          if (MINBPC(enc) > 3) {
+	    if (*ptr2++ != *ptr1++)
+      	      return 0;
+	  }
+	}
+      }
+      break;
+    default:
+      if (MINBPC(enc) == 1 && *ptr1 == *ptr2)
+	return 1;
+      switch (BYTE_TYPE(enc, ptr2)) {
+      case BT_LEAD2:
+      case BT_LEAD3:
+      case BT_LEAD4:
+      case BT_NONASCII:
+      case BT_NMSTRT:
+#ifdef XML_NS
+      case BT_COLON:
+#endif
+      case BT_HEX:
+      case BT_DIGIT:
+      case BT_NAME:
+      case BT_MINUS:
+	return 0;
+      default:
+	return 1;
+      }
+    }
+  }
+  /* not reached */
+}
+
+static
+int PREFIX(nameMatchesAscii)(const ENCODING *enc, const char *ptr1,
+			     const char *end1, const char *ptr2)
+{
+  for (; *ptr2; ptr1 += MINBPC(enc), ptr2++) {
+    if (ptr1 == end1)
+      return 0;
+    if (!CHAR_MATCHES(enc, ptr1, *ptr2))
+      return 0;
+  }
+  return ptr1 == end1;
+}
+
+static
+int PREFIX(nameLength)(const ENCODING *enc, const char *ptr)
+{
+  const char *start = ptr;
+  for (;;) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: ptr += n; break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_NONASCII:
+    case BT_NMSTRT:
+#ifdef XML_NS
+    case BT_COLON:
+#endif
+    case BT_HEX:
+    case BT_DIGIT:
+    case BT_NAME:
+    case BT_MINUS:
+      ptr += MINBPC(enc);
+      break;
+    default:
+      return ptr - start;
+    }
+  }
+}
+
+static
+const char *PREFIX(skipS)(const ENCODING *enc, const char *ptr)
+{
+  for (;;) {
+    switch (BYTE_TYPE(enc, ptr)) {
+    case BT_LF:
+    case BT_CR:
+    case BT_S:
+      ptr += MINBPC(enc);
+      break;
+    default:
+      return ptr;
+    }
+  }
+}
+
+static
+void PREFIX(updatePosition)(const ENCODING *enc,
+			    const char *ptr,
+			    const char *end,
+			    POSITION *pos)
+{
+  while (ptr != end) {
+    switch (BYTE_TYPE(enc, ptr)) {
+#define LEAD_CASE(n) \
+    case BT_LEAD ## n: \
+      ptr += n; \
+      break;
+    LEAD_CASE(2) LEAD_CASE(3) LEAD_CASE(4)
+#undef LEAD_CASE
+    case BT_LF:
+      pos->columnNumber = (unsigned)-1;
+      pos->lineNumber++;
+      ptr += MINBPC(enc);
+      break;
+    case BT_CR:
+      pos->lineNumber++;
+      ptr += MINBPC(enc);
+      if (ptr != end && BYTE_TYPE(enc, ptr) == BT_LF)
+	ptr += MINBPC(enc);
+      pos->columnNumber = (unsigned)-1;
+      break;
+    default:
+      ptr += MINBPC(enc);
+      break;
+    }
+    pos->columnNumber++;
+  }
+}
+
+#undef DO_LEAD_CASE
+#undef MULTIBYTE_CASES
+#undef INVALID_CASES
+#undef CHECK_NAME_CASE
+#undef CHECK_NAME_CASES
+#undef CHECK_NMSTRT_CASE
+#undef CHECK_NMSTRT_CASES
diff --git a/xml/expat/lib/xmltok_impl.h b/xml/expat/lib/xmltok_impl.h
new file mode 100644
index 000000000..da0ea60a6
--- /dev/null
+++ b/xml/expat/lib/xmltok_impl.h
@@ -0,0 +1,46 @@
+/*
+Copyright (c) 1998, 1999 Thai Open Source Software Center Ltd
+See the file COPYING for copying permission.
+*/
+
+enum {
+  BT_NONXML,
+  BT_MALFORM,
+  BT_LT,
+  BT_AMP,
+  BT_RSQB,
+  BT_LEAD2,
+  BT_LEAD3,
+  BT_LEAD4,
+  BT_TRAIL,
+  BT_CR,
+  BT_LF,
+  BT_GT,
+  BT_QUOT,
+  BT_APOS,
+  BT_EQUALS,
+  BT_QUEST,
+  BT_EXCL,
+  BT_SOL,
+  BT_SEMI,
+  BT_NUM,
+  BT_LSQB,
+  BT_S,
+  BT_NMSTRT,
+  BT_COLON,
+  BT_HEX,
+  BT_DIGIT,
+  BT_NAME,
+  BT_MINUS,
+  BT_OTHER, /* known not to be a name or name start character */
+  BT_NONASCII, /* might be a name or name start character */
+  BT_PERCNT,
+  BT_LPAR,
+  BT_RPAR,
+  BT_AST,
+  BT_PLUS,
+  BT_COMMA,
+  BT_VERBAR
+};
+
+#include <stddef.h>
diff --git a/xml/expat/lib/xmltok_ns.c b/xml/expat/lib/xmltok_ns.c
new file mode 100644
index 000000000..21859738a
--- /dev/null
+++ b/xml/expat/lib/xmltok_ns.c
@@ -0,0 +1,98 @@
+const ENCODING *NS(XmlGetUtf8InternalEncoding)(void)
+{
+  return &ns(internal_utf8_encoding).enc;
+}
+
+const ENCODING *NS(XmlGetUtf16InternalEncoding)(void)
+{
+#if XML_BYTE_ORDER == 12
+  return &ns(internal_little2_encoding).enc;
+#elif XML_BYTE_ORDER == 21
+  return &ns(internal_big2_encoding).enc;
+#else
+  const short n = 1;
+  return *(const char *)&n ? &ns(internal_little2_encoding).enc : &ns(internal_big2_encoding).enc;
+#endif
+}
+
+static
+const ENCODING *NS(encodings)[] = {
+  &ns(latin1_encoding).enc,
+  &ns(ascii_encoding).enc,
+  &ns(utf8_encoding).enc,
+  &ns(big2_encoding).enc,
+  &ns(big2_encoding).enc,
+  &ns(little2_encoding).enc,
+  &ns(utf8_encoding).enc /* NO_ENC */
+};
+
+static
+int NS(initScanProlog)(const ENCODING *enc, const char *ptr, const char *end,
+		       const char **nextTokPtr)
+{
+  return initScan(NS(encodings), (const INIT_ENCODING *)enc, XML_PROLOG_STATE, ptr, end, nextTokPtr);
+}
+
+static
+int NS(initScanContent)(const ENCODING *enc, const char *ptr, const char *end,
+		       const char **nextTokPtr)
+{
+  return initScan(NS(encodings), (const INIT_ENCODING *)enc, XML_CONTENT_STATE, ptr, end, nextTokPtr);
+}
+
+int NS(XmlInitEncoding)(INIT_ENCODING *p, const ENCODING **encPtr, const char *name)
+{
+  int i = getEncodingIndex(name);
+  if (i == UNKNOWN_ENC)
+    return 0;
+  SET_INIT_ENC_INDEX(p, i);
+  p->initEnc.scanners[XML_PROLOG_STATE] = NS(initScanProlog);
+  p->initEnc.scanners[XML_CONTENT_STATE] = NS(initScanContent);
+  p->initEnc.updatePosition = initUpdatePosition;
+  p->encPtr = encPtr;
+  *encPtr = &(p->initEnc);
+  return 1;
+}
+
+static
+const ENCODING *NS(findEncoding)(const ENCODING *enc, const char *ptr, const char *end)
+{
+#define ENCODING_MAX 128
+  char buf[ENCODING_MAX];
+  char *p = buf;
+  int i;
+  XmlUtf8Convert(enc, &ptr, end, &p, p + ENCODING_MAX - 1);
+  if (ptr != end)
+    return 0;
+  *p = 0;
+  if (streqci(buf, KW_UTF_16) && enc->minBytesPerChar == 2)
+    return enc;
+  i = getEncodingIndex(buf);
+  if (i == UNKNOWN_ENC)
+    return 0;
+  return NS(encodings)[i];
+}
+
+int NS(XmlParseXmlDecl)(int isGeneralTextEntity,
+			const ENCODING *enc,
+			const char *ptr,
+			const char *end,
+			const char **badPtr,
+			const char **versionPtr,
+			const char **versionEndPtr,
+			const char **encodingName,
+			const ENCODING **encoding,
+			int *standalone)
+{
+  return doParseXmlDecl(NS(findEncoding),
+			isGeneralTextEntity,
+			enc,
+			ptr,
+			end,
+			badPtr,
+			versionPtr,
+			versionEndPtr,
+			encodingName,
+			encoding,
+			standalone);
+}