applied UTF-16 encoding handling patch provided by Mark Itzcovitz more

* encoding.c: applied UTF-16 encoding handling patch provided by
  Mark Itzcovitz
* encoding.c parser.c: more cleanup and fixes for UTF-16 when
  not having iconv support.
Daniel

4 files changed, 34 insertions, 19 deletions

diff --git a/ChangeLog b/ChangeLog
index 059441fe..d2862c9a 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+Thu Aug 14 14:13:43 CEST 2003 Daniel Veillard <daniel@veillard.com>
+
+	* encoding.c: applied UTF-16 encoding handling patch provided by
+	  Mark Itzcovitz
+	* encoding.c parser.c: more cleanup and fixes for UTF-16 when 
+	  not having iconv support.
+
 Thu Aug 14 03:19:08 CEST 2003 Daniel Veillard <daniel@veillard.com>
 
 	* Makefile.am configure.in example/Makefile.am libxml.h nanoftp.c
@@ -8,7 +15,7 @@ Thu Aug 14 02:28:36 CEST 2003 Daniel Veillard <daniel@veillard.com>
 
 	* parser.c: fixed the serious CPU usage problem reported by
 	  Grant Goodale
-	* HTMLparser.c: applied patch from Oliver Kidman about a free
+	* HTMLparser.c: applied patch from Oliver Stoeneberg about a free
 	  missing in htmlSAXParseDoc
 
 Tue Aug 12 22:48:10 HKT 2003 William Brack <wbrack@mmm.com.hk>
diff --git a/encoding.c b/encoding.c
index e79a5f95..2d3ccd0f 100644
--- a/encoding.c
+++ b/encoding.c
@@ -875,6 +875,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
 {
     unsigned short* out = (unsigned short*) outb;
     const unsigned char* processed = in;
+    const unsigned char *const instart = in;
     unsigned short* outstart= out;
     unsigned short* outend;
     const unsigned char* inend= in+*inlen;
@@ -909,7 +910,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
       else if (d < 0xC0) {
           /* trailing byte in leading position */
 	  *outlen = (out - outstart) * 2;
-	  *inlen = processed - in;
+	  *inlen = processed - instart;
 	  return(-2);
       } else if (d < 0xE0)  { c= d & 0x1F; trailing= 1; }
       else if (d < 0xF0)  { c= d & 0x0F; trailing= 2; }
@@ -917,7 +918,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
       else {
 	/* no chance for this in UTF-16 */
 	*outlen = (out - outstart) * 2;
-	*inlen = processed - in;
+	*inlen = processed - instart;
 	return(-2);
       }
 
@@ -971,7 +972,7 @@ UTF8ToUTF16LE(unsigned char* outb, int *outlen,
 	processed = in;
     }
     *outlen = (out - outstart) * 2;
-    *inlen = processed - in;
+    *inlen = processed - instart;
     return(0);
 }
 
@@ -1086,6 +1087,7 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
 {
     unsigned short* out = (unsigned short*) outb;
     const unsigned char* processed = in;
+    const unsigned char *const instart = in;
     unsigned short* outstart= out;
     unsigned short* outend;
     const unsigned char* inend= in+*inlen;
@@ -1120,7 +1122,7 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
       else if (d < 0xC0)  {
           /* trailing byte in leading position */
 	  *outlen = out - outstart;
-	  *inlen = processed - in;
+	  *inlen = processed - instart;
 	  return(-2);
       } else if (d < 0xE0)  { c= d & 0x1F; trailing= 1; }
       else if (d < 0xF0)  { c= d & 0x0F; trailing= 2; }
@@ -1128,7 +1130,7 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
       else {
           /* no chance for this in UTF-16 */
 	  *outlen = out - outstart;
-	  *inlen = processed - in;
+	  *inlen = processed - instart;
 	  return(-2);
       }
 
@@ -1179,7 +1181,7 @@ UTF8ToUTF16BE(unsigned char* outb, int *outlen,
 	processed = in;
     }
     *outlen = (out - outstart) * 2;
-    *inlen = processed - in;
+    *inlen = processed - instart;
     return(0);
 }
 
@@ -1962,6 +1964,14 @@ xmlFindCharEncodingHandler(const char *name) {
         }
     }
 
+    /*
+     * If nothing was found and it is "UTF-16" then use the Little indian
+     * version.
+     */
+    if ((xmlStrEqual(BAD_CAST upper, BAD_CAST "UTF-16")) ||
+	(xmlStrEqual(BAD_CAST upper, BAD_CAST "UTF16")))
+        return(xmlUTF16LEHandler);
+
     return(NULL);
 }
 
diff --git a/include/libxml/parserInternals.h b/include/libxml/parserInternals.h
index 8507442b..aec4a823 100644
--- a/include/libxml/parserInternals.h
+++ b/include/libxml/parserInternals.h
@@ -295,7 +295,7 @@ void			xmlParseElement		(xmlParserCtxtPtr ctxt);
 xmlChar *		xmlParseVersionNum	(xmlParserCtxtPtr ctxt);
 xmlChar *		xmlParseVersionInfo	(xmlParserCtxtPtr ctxt);
 xmlChar *		xmlParseEncName		(xmlParserCtxtPtr ctxt);
-xmlChar *		xmlParseEncodingDecl	(xmlParserCtxtPtr ctxt);
+const xmlChar *		xmlParseEncodingDecl	(xmlParserCtxtPtr ctxt);
 int			xmlParseSDDecl		(xmlParserCtxtPtr ctxt);
 void			xmlParseXMLDecl		(xmlParserCtxtPtr ctxt);
 void			xmlParseTextDecl	(xmlParserCtxtPtr ctxt);
diff --git a/parser.c b/parser.c
index cb7fce4f..82caf10b 100644
--- a/parser.c
+++ b/parser.c
@@ -7518,7 +7518,7 @@ xmlParseEncName(xmlParserCtxtPtr ctxt) {
  * Returns the encoding value or NULL
  */
 
-xmlChar *
+const xmlChar *
 xmlParseEncodingDecl(xmlParserCtxtPtr ctxt) {
     xmlChar *encoding = NULL;
     const xmlChar *q;
@@ -7582,23 +7582,21 @@ xmlParseEncodingDecl(xmlParserCtxtPtr ctxt) {
         if ((encoding != NULL) &&
 	    ((!xmlStrcasecmp(encoding, BAD_CAST "UTF-16")) ||
 	     (!xmlStrcasecmp(encoding, BAD_CAST "UTF16")))) {
-	    if (ctxt->input->encoding != NULL)
-		xmlFree((xmlChar *) ctxt->input->encoding);
-	    ctxt->input->encoding = encoding;
-	    encoding = NULL;
+	    if (ctxt->encoding != NULL)
+		xmlFree((xmlChar *) ctxt->encoding);
+	    ctxt->encoding = encoding;
 	}
 	/*
 	 * UTF-8 encoding is handled natively
 	 */
-        if ((encoding != NULL) &&
+        else if ((encoding != NULL) &&
 	    ((!xmlStrcasecmp(encoding, BAD_CAST "UTF-8")) ||
 	     (!xmlStrcasecmp(encoding, BAD_CAST "UTF8")))) {
-	    if (ctxt->input->encoding != NULL)
-		xmlFree((xmlChar *) ctxt->input->encoding);
-	    ctxt->input->encoding = encoding;
-	    encoding = NULL;
+	    if (ctxt->encoding != NULL)
+		xmlFree((xmlChar *) ctxt->encoding);
+	    ctxt->encoding = encoding;
 	}
-	if (encoding != NULL) {
+	else if (encoding != NULL) {
 	    xmlCharEncodingHandlerPtr handler;
 
 	    if (ctxt->input->encoding != NULL)