gdb:

2009-03-19  Tom Tromey  <tromey@redhat.com>
	    Julian Brown  <julian@codesourcery.com>

	PR i18n/7220, PR i18n/7821, PR exp/8815, PR exp/9103,
	PR i18n/9401, PR exp/9613:
	* NEWS: Update
	* value.h (value_typed_string): Declare.
	(val_print_string): Update.
	* valprint.h (print_char_chars): Update.
	* valprint.c (print_char_chars): Add type argument.  Update.
	(val_print_string): Likewise.
	* valops.c (value_typed_string): New function.
	* utils.c (host_char_to_target): New function.
	(parse_escape): Use host_char_to_target, host_hex_value.  Update.
	Remove '^' case.
	(no_control_char_error): Remove.
	* typeprint.c (print_type_scalar): Update.
	* scm-valprint.c (scm_scmval_print): Update.
	* scm-lang.h (scm_printchar, scm_printstr): Update.
	* scm-lang.c (scm_printchar): Add type argument.
	(scm_printstr): Likewise.
	* printcmd.c (print_formatted): Update.
	(print_scalar_formatted): Update.
	(printf_command) <wide_string_arg, wide_char_arg>: New constants.
	Handle '%lc' and '%ls'.
	* parser-defs.h (struct typed_stoken): New type.
	(struct stoken_vector): Likewise.
	(write_exp_string_vector): Declare.
	* parse.c (write_exp_string_vector): New function.
	* p-valprint.c (pascal_val_print): Update.
	* p-lang.h (is_pascal_string_type, pascal_printchar,
	pascal_printstr): Update.
	* p-lang.c (is_pascal_string_type): Remove 'char_size' argument.
	Add 'char_type' argument.
	(pascal_emit_char): Add type argument.
	(pascal_printchar): Likewise.
	(pascal_printstr): Likewise.
	* objc-lang.c (objc_emit_char): Add type argument.
	(objc_printchar): Likewise.
	(objc_printstr): Likewise.
	* macroexp.c (get_character_constant): Handle unicode characters.
	Use c_parse_escape.
	(get_string_literal): Handle unicode strings.  Use
	c_parse_escape.
	* m2-valprint.c (print_unpacked_pointer): Update.
	(m2_print_array_contents): Update.
	(m2_val_print): Update.
	* m2-lang.c (m2_emit_char): Add type argument.
	(m2_printchar): Likewise.
	(m2_printstr): Likewise.
	* language.h (struct language_defn) <la_printchar>: Add type
	argument.
	<la_printstr, la_emitchar>: Likewise.
	(LA_PRINT_CHAR): Likewise.
	(LA_PRINT_STRING): Likewise.
	(LA_EMIT_CHAR): Likewise.
	* language.c (unk_lang_emit_char): Add type argument.
	(unk_lang_printchar): Likewise.
	(unk_lang_printstr): Likewise.
	* jv-valprint.c (java_val_print): Update.
	* jv-lang.c (java_emit_char): Add type argument.
	* f-valprint.c (f_val_print): Update.
	* f-lang.c (f_emit_char): Add type argument.
	(f_printchar): Likewise.
	(f_printstr): Likewise.
	* expprint.c (print_subexp_standard): Update.
	* charset.h (target_wide_charset): Declare.
	(c_target_char_has_backslash_escape, c_parse_backslash,
	host_char_print_literally, host_char_to_target,
	target_char_to_host, target_char_to_control_char): Remove.
	(enum transliterations): New type.
	(convert_between_encodings): Declare.
	(HOST_ESCAPE_CHAR): New define.
	(host_letter_to_control_character, host_hex_value): Declare.
	(enum wchar_iterate_result): New enum.
	(struct wchar_iterator): Declare.
	(make_wchar_iterator, make_cleanup_wchar_iterator, wchar_iterator,
	wchar_push_back): Declare.
	* charset-list.h: New file.
	* c-valprint.c (textual_name): New function.
	(textual_element_type): Handle wide character types.
	(c_val_print): Pass original type to textual_element_type.  Handle
	wide character types.
	(c_value_print): Use textual_element_type.  Pass original type of
	value to val_print.
	* c-lang.h (enum c_string_type): New type.
	(c_printchar, c_printstr): Update.
	* c-lang.c (classify_type): New function.
	(print_wchar): Likewise.
	(c_emit_char): Add type argument.  Handle wide characters.
	(c_printchar): Likewise.
	(c_printstr): Add type argument.  Handle wide and multibyte
	character sets.
	(convert_ucn): New function.
	(emit_numeric_character): Likewise.
	(convert_octal): Likewise.
	(convert_hex): Likewise.
	(ADVANCE): New macro.
	(convert_escape): New function.
	(parse_one_string): Likewise.
	(evaluate_subexp_c): Likewise.
	(exp_descriptor_c): New global.
	(c_language_defn): Use exp_descriptor_c.
	(cplus_language_defn): Likewise.
	(asm_language_defn): Likewise.
	(minimal_language_defn): Likewise.
	(charset_for_string_type): New function.
	* c-exp.y (%union): Add 'svec' and 'tsval'.
	(CHAR): New token.
	(exp): Add CHAR production.
	(string_exp): Rewrite.
	(exp) <string_exp>: Rewrite.
	(tempbuf): Now global.
	(tempbuf_init): New global.
	(parse_string_or_char): New function.
	(yylex) <tempbuf>: Now global.
	<tokptr, tempbufindex, tempbufsize, token_string, class_prefix>:
	Remove.
	Handle 'u', 'U', and 'L' prefixes.  Call parse_string_or_char.
	(c_parse_escape): New function.
	* auxv.c (fprint_target_auxv): Update.
	* ada-valprint.c (ada_emit_char): Add type argument.
	(ada_printchar): Likewise.
	(ada_print_scalar): Update.
	(printstr): Add type argument.  Update calls to ada_emit_char.
	(ada_printstr): Add type argument.
	(ada_val_print_array): Update.
	(ada_val_print_1): Likewise.
	* ada-lang.c (emit_char): Add type argument.
	* ada-lang.h (ada_emit_char, ada_printchar, ada_printstr): Add
	type arguments.
	* gdb_locale.h: Include langinfo.h.
	* charset.c (_initialize_charset): Set default host charset from
	the locale.  Don't register charsets.  Add target-wide-charset
	commands.  Call find_charset_names.
	(struct charset, struct translation): Remove.
	(GDB_DEFAULT_HOST_CHARSET): Remove.
	(GDB_DEFAULT_TARGET_WIDE_CHARSET): New define.
	(target_wide_charset_name): New global.
	(show_host_charset_name): Handle "auto".
	(show_target_wide_charset_name): New function.
	(host_charset_enum, target_charset_enum): Remove.
	(charset_enum): New global.
	(all_charsets, register_charset, lookup_charset, all_translations,
	register_translation, lookup_translation): Remove.
	(simple_charset, ascii_print_literally, ascii_to_control): Remove.
	(iso_8859_print_literally, iso_8859_to_control,
	iso_8859_family_charset): Remove.
	(ebcdic_print_literally, ebcdic_to_control,
	ebcdic_family_charset): Remove.
	(struct cached_iconv, check_iconv_cache, cached_iconv_convert,
	register_iconv_charsets): Remove.
	(target_wide_charset_be_name, target_wide_charset_le_name): New
	globals.
	(identity_either_char_to_other): Remove.
	(set_be_le_names, validate): New functions.
	(backslashable, backslashed, represented): Remove.
	(default_c_target_char_has_backslash_escape): Remove.
	(default_c_parse_backslash, iconv_convert): Remove.
	(ascii_to_iso_8859_1_table, ascii_to_ebcdic_us_table,
	ascii_to_ibm1047_table, iso_8859_1_to_ascii_table,
	iso_8859_1_to_ebcdic_us_table, iso_8859_1_to_ibm1047_table,
	ebcdic_us_to_ascii_table, ebcdic_us_to_iso_8859_1_table,
	ebcdic_us_to_ibm1047_table, ibm1047_to_ascii_table,
	ibm1047_to_iso_8859_1_table, ibm1047_to_ebcdic_us_table): Remove.
	(table_convert_char, table_translation, simple_table_translation):
	Remove.
	(current_host_charset, current_target_charset,
	c_target_char_has_backslash_escape_func,
	c_target_char_has_backslash_escape_baton): Remove.
	(c_parse_backslash_func, c_parse_backslash_baton): Remove.
	(host_char_to_target_func, host_char_to_target_baton): Remove.
	(target_char_to_host_func, target_char_to_host_baton): Remove.
	(cached_iconv_host_to_target, cached_iconv_target_to_host):
	Remove.
	(lookup_charset_or_error, check_valid_host_charset): Remove.
	(set_host_and_target_charsets): Remove.
	(set_host_charset, set_target_charset): Remove.
	(set_host_charset_sfunc, set_target_charset_sfunc): Rewrite.
	(set_target_wide_charset_sfunc): New function.
	(show_charset): Print target wide character set.
	(host_charset, target_charset): Rewrite.
	(target_wide_charset): New function.
	(c_target_char_has_backslash_escape): Remove.
	(c_parse_backslash): Remove.
	(host_letter_to_control_character): New function.
	(host_char_print_literally): Remove.
	(host_hex_value): New function.
	(target_char_to_control_char): Remove.
	(cleanup_iconv): New function.
	(convert_between_encodings): New function.
	(target_char_to_host): Remove.
	(struct wchar_iterator): Define.
	(make_wchar_iterator, make_cleanup_wchar_iterator, wchar_iterator,
	wchar_push_back): New functions.
	(do_cleanup_iterator): New function.
	(char_ptr): New typedef.
	(charsets): New global.
	(add_one, find_charset_names): New functions.
	(default_charset_names): New global.
	(auto_host_charset_name): Likewise.
	* aclocal.m4, config.in, configure: Rebuild.
	* configure.ac: Call AM_LANGINFO_CODESET.
	(GDB_DEFAULT_HOST_CHARSET): Default to UTF-8.
	(AM_ICONV): Invoke earlier.
	* acinclude.m4: Include codeset.m4.  Subst LIBICONV_INCLUDE and
	LIBICONV_LIBDIR.  Check for libiconv in build tree.
	* Makefile.in (LIBICONV_LIBDIR, LIBICONV_INCLUDE): New macros.
	(INTERNAL_CFLAGS_BASE): Add LIBICONV_INCLUDE.
	(INTERNAL_LDFLAGS): Add LIBICONV_LIBDIR.
	* gdb_obstack.h (obstack_grow_wstr): New define.
        * gdb_wchar.h: New file.
        * defs.h: Include it.
gdb/testsuite:
	* gdb.base/store.exp: Update for change to escape output.
	* gdb.base/callfuncs.exp (fetch_all_registers): Update for change
	to escape output.
	* gdb.base/pointers.exp: Update for change to escape output.
	* gdb.base/long_long.exp (gdb_test_long_long): Update for change
	to escape output.
	* gdb.base/constvars.exp (do_constvar_tests): Update for change to
	escape output.
	* gdb.base/call-rt-st.exp (print_struct_call): Update for change
	to escape output.
	* gdb.cp/ref-types.exp (gdb_start_again): Update for change to
	escape output.
	* gdb.base/setvar.exp: Update for change to escape output.
	* lib/gdb.exp (default_gdb_start): Set LC_CTYPE to C.
	* gdb.base/printcmds.exp (test_print_all_chars): Update for change
	to escape output.
	(test_print_string_constants): Likewise.
	* gdb.base/charset.exp (valid_host_charset): Check size of
	wchar_t.  Handle UCS-2 and UCS-4.  Add tests for wide and unicode
	cases.  Handle "auto"-related output.
	* gdb.base/charset.c (char16_t, char32_t): New typedefs.
	(uvar, Uvar): New globals.
gdb/doc:
	* gdb.texinfo (Character Sets): Remove obsolete text.  Document
	set target-wide-charset.
	(Requirements): Mention iconv.

1 files changed, 750 insertions, 72 deletions

diff --git a/gdb/c-lang.c b/gdb/c-lang.c
index 8b5410f8a6b..deab3f4fa0b 100644
--- a/gdb/c-lang.c
+++ b/gdb/c-lang.c
@@ -33,48 +33,304 @@
 #include "demangle.h"
 #include "cp-abi.h"
 #include "cp-support.h"
+#include "gdb_obstack.h"
+#include <ctype.h>
 
 extern void _initialize_c_language (void);
-static void c_emit_char (int c, struct ui_file * stream, int quoter);
+
+/* Given a C string type, STR_TYPE, return the corresponding target
+   character set name.  */
+
+static const char *
+charset_for_string_type (enum c_string_type str_type)
+{
+  switch (str_type & ~C_CHAR)
+    {
+    case C_STRING:
+      return target_charset ();
+    case C_WIDE_STRING:
+      return target_wide_charset ();
+    case C_STRING_16:
+      /* FIXME: UCS-2 is not always correct.  */
+      if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG)
+	return "UCS-2BE";
+      else
+	return "UCS-2LE";
+    case C_STRING_32:
+      /* FIXME: UCS-4 is not always correct.  */
+      if (gdbarch_byte_order (current_gdbarch) == BFD_ENDIAN_BIG)
+	return "UCS-4BE";
+      else
+	return "UCS-4LE";
+    }
+  internal_error (__FILE__, __LINE__, "unhandled c_string_type");
+}
+
+/* Classify ELTTYPE according to what kind of character it is.  Return
+   the enum constant representing the character type.  Also set
+   *ENCODING to the name of the character set to use when converting
+   characters of this type to the host character set.  */
+
+static enum c_string_type
+classify_type (struct type *elttype, const char **encoding)
+{
+  struct type *saved_type;
+  enum c_string_type result;
+
+  /* We do one or two passes -- one on ELTTYPE, and then maybe a
+     second one on a typedef target.  */
+  do
+    {
+      char *name = TYPE_NAME (elttype);
+
+      if (TYPE_CODE (elttype) == TYPE_CODE_CHAR || !name)
+	{
+	  result = C_CHAR;
+	  goto done;
+	}
+
+      if (!strcmp (name, "wchar_t"))
+	{
+	  result = C_WIDE_CHAR;
+	  goto done;
+	}
+
+      if (!strcmp (name, "char16_t"))
+	{
+	  result = C_CHAR_16;
+	  goto done;
+	}
+
+      if (!strcmp (name, "char32_t"))
+	{
+	  result = C_CHAR_32;
+	  goto done;
+	}
+
+      saved_type = elttype;
+      CHECK_TYPEDEF (elttype);
+    }
+  while (elttype != saved_type);
+
+  /* Punt.  */
+  result = C_CHAR;
+
+ done:
+  *encoding = charset_for_string_type (result);
+  return result;
+}
+
+/* Return true if print_wchar can display W without resorting to a
+   numeric escape, false otherwise.  */
+
+static int
+wchar_printable (gdb_wchar_t w)
+{
+  return (gdb_iswprint (w)
+	  || w == LCST ('\a') || w == LCST ('\b')
+	  || w == LCST ('\f') || w == LCST ('\n')
+	  || w == LCST ('\r') || w == LCST ('\t')
+	  || w == LCST ('\v'));
+}
+
+/* A helper function that converts the contents of STRING to wide
+   characters and then appends them to OUTPUT.  */
+
+static void
+append_string_as_wide (const char *string, struct obstack *output)
+{
+  for (; *string; ++string)
+    {
+      gdb_wchar_t w = gdb_btowc (*string);
+      obstack_grow (output, &w, sizeof (gdb_wchar_t));
+    }
+}
+
+/* Print a wide character W to OUTPUT.  ORIG is a pointer to the
+   original (target) bytes representing the character, ORIG_LEN is the
+   number of valid bytes.  WIDTH is the number of bytes in a base
+   characters of the type.  OUTPUT is an obstack to which wide
+   characters are emitted.  QUOTER is a (narrow) character indicating
+   the style of quotes surrounding the character to be printed.
+   NEED_ESCAPE is an in/out flag which is used to track numeric
+   escapes across calls.  */
+
+static void
+print_wchar (gdb_wint_t w, const gdb_byte *orig, int orig_len,
+	     int width, struct obstack *output, int quoter,
+	     int *need_escapep)
+{
+  int need_escape = *need_escapep;
+  *need_escapep = 0;
+  if (gdb_iswprint (w) && (!need_escape || (!gdb_iswdigit (w)
+					    && w != LCST ('8')
+					    && w != LCST ('9'))))
+    {
+      if (w == gdb_btowc (quoter) || w == LCST ('\\'))
+	obstack_grow_wstr (output, LCST ("\\"));
+      obstack_grow (output, &w, sizeof (gdb_wchar_t));
+    }
+  else
+    {
+      switch (w)
+	{
+	case LCST ('\a'):
+	  obstack_grow_wstr (output, LCST ("\\a"));
+	  break;
+	case LCST ('\b'):
+	  obstack_grow_wstr (output, LCST ("\\b"));
+	  break;
+	case LCST ('\f'):
+	  obstack_grow_wstr (output, LCST ("\\f"));
+	  break;
+	case LCST ('\n'):
+	  obstack_grow_wstr (output, LCST ("\\n"));
+	  break;
+	case LCST ('\r'):
+	  obstack_grow_wstr (output, LCST ("\\r"));
+	  break;
+	case LCST ('\t'):
+	  obstack_grow_wstr (output, LCST ("\\t"));
+	  break;
+	case LCST ('\v'):
+	  obstack_grow_wstr (output, LCST ("\\v"));
+	  break;
+	default:
+	  {
+	    int i;
+
+	    for (i = 0; i + width <= orig_len; i += width)
+	      {
+		char octal[30];
+		ULONGEST value = extract_unsigned_integer (&orig[i], width);
+		sprintf (octal, "\\%lo", (long) value);
+		append_string_as_wide (octal, output);
+	      }
+	    /* If we somehow have extra bytes, print them now.  */
+	    while (i < orig_len)
+	      {
+		char octal[5];
+		sprintf (octal, "\\%.3o", orig[i] & 0xff);
+		append_string_as_wide (octal, output);
+		++i;
+	      }
+
+	    *need_escapep = 1;
+	  }
+	  break;
+	}
+    }
+}
 
 /* Print the character C on STREAM as part of the contents of a literal
    string whose delimiter is QUOTER.  Note that that format for printing
    characters and strings is language specific. */
 
 static void
-c_emit_char (int c, struct ui_file *stream, int quoter)
+c_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
-  const char *escape;
-  int host_char;
+  struct obstack wchar_buf, output;
+  struct cleanup *cleanups;
+  const char *encoding;
+  gdb_byte *buf;
+  struct wchar_iterator *iter;
+  int need_escape = 0;
 
-  c &= 0xFF;			/* Avoid sign bit follies */
+  classify_type (type, &encoding);
 
-  escape = c_target_char_has_backslash_escape (c);
-  if (escape)
-    {
-      if (quoter == '"' && strcmp (escape, "0") == 0)
-	/* Print nulls embedded in double quoted strings as \000 to
-	   prevent ambiguity.  */
-	fprintf_filtered (stream, "\\000");
-      else
-	fprintf_filtered (stream, "\\%s", escape);
-    }
-  else if (target_char_to_host (c, &host_char)
-           && host_char_print_literally (host_char))
+  buf = alloca (TYPE_LENGTH (type));
+  pack_long (buf, type, c);
+
+  iter = make_wchar_iterator (buf, TYPE_LENGTH (type), encoding,
+			      TYPE_LENGTH (type));
+  cleanups = make_cleanup_wchar_iterator (iter);
+
+  /* This holds the printable form of the wchar_t data.  */
+  obstack_init (&wchar_buf);
+  make_cleanup_obstack_free (&wchar_buf);
+
+  while (1)
     {
-      if (host_char == '\\' || host_char == quoter)
-        fputs_filtered ("\\", stream);
-      fprintf_filtered (stream, "%c", host_char);
+      int num_chars;
+      gdb_wchar_t *chars;
+      const gdb_byte *buf;
+      size_t buflen;
+      int print_escape = 1;
+      enum wchar_iterate_result result;
+
+      num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+      if (num_chars < 0)
+	break;
+      if (num_chars > 0)
+	{
+	  /* If all characters are printable, print them.  Otherwise,
+	     we're going to have to print an escape sequence.  We
+	     check all characters because we want to print the target
+	     bytes in the escape sequence, and we don't know character
+	     boundaries there.  */
+	  int i;
+
+	  print_escape = 0;
+	  for (i = 0; i < num_chars; ++i)
+	    if (!wchar_printable (chars[i]))
+	      {
+		print_escape = 1;
+		break;
+	      }
+
+	  if (!print_escape)
+	    {
+	      for (i = 0; i < num_chars; ++i)
+		print_wchar (chars[i], buf, buflen, TYPE_LENGTH (type),
+			     &wchar_buf, quoter, &need_escape);
+	    }
+	}
+
+      /* This handles the NUM_CHARS == 0 case as well.  */
+      if (print_escape)
+	print_wchar (gdb_WEOF, buf, buflen, TYPE_LENGTH (type), &wchar_buf,
+		     quoter, &need_escape);
     }
-  else
-    fprintf_filtered (stream, "\\%.3o", (unsigned int) c);
+
+  /* The output in the host encoding.  */
+  obstack_init (&output);
+  make_cleanup_obstack_free (&output);
+
+  convert_between_encodings ("wchar_t", host_charset (),
+			     obstack_base (&wchar_buf),
+			     obstack_object_size (&wchar_buf),
+			     1, &output, translit_char);
+  obstack_1grow (&output, '\0');
+
+  fputs_filtered (obstack_base (&output), stream);
+
+  do_cleanups (cleanups);
 }
 
 void
-c_printchar (int c, struct ui_file *stream)
+c_printchar (int c, struct type *type, struct ui_file *stream)
 {
+  enum c_string_type str_type;
+  const char *encoding;
+
+  str_type = classify_type (type, &encoding);
+  switch (str_type)
+    {
+    case C_CHAR:
+      break;
+    case C_WIDE_CHAR:
+      fputc_filtered ('L', stream);
+      break;
+    case C_CHAR_16:
+      fputc_filtered ('u', stream);
+      break;
+    case C_CHAR_32:
+      fputc_filtered ('U', stream);
+      break;
+    }
+
   fputc_filtered ('\'', stream);
-  LA_EMIT_CHAR (c, stream, '\'');
+  LA_EMIT_CHAR (c, type, stream, '\'');
   fputc_filtered ('\'', stream);
 }
 
@@ -85,87 +341,208 @@ c_printchar (int c, struct ui_file *stream)
    printing LENGTH characters, or if FORCE_ELLIPSES.  */
 
 void
-c_printstr (struct ui_file *stream, const gdb_byte *string,
-	    unsigned int length, int width, int force_ellipses,
+c_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
+	    unsigned int length, int force_ellipses,
 	    const struct value_print_options *options)
 {
   unsigned int i;
   unsigned int things_printed = 0;
   int in_quotes = 0;
   int need_comma = 0;
+  int width = TYPE_LENGTH (type);
+  struct obstack wchar_buf, output;
+  struct cleanup *cleanup;
+  enum c_string_type str_type;
+  const char *encoding;
+  struct wchar_iterator *iter;
+  int finished = 0;
+  int need_escape = 0;
 
   /* If the string was not truncated due to `set print elements', and
      the last byte of it is a null, we don't print that, in traditional C
      style.  */
   if (!force_ellipses
       && length > 0
-      && (extract_unsigned_integer (string + (length - 1) * width, width)
-          == '\0'))
+      && (extract_unsigned_integer (string + (length - 1) * width, width) == 0))
     length--;
 
+  str_type = classify_type (type, &encoding) & ~C_CHAR;
+  switch (str_type)
+    {
+    case C_STRING:
+      break;
+    case C_WIDE_STRING:
+      fputs_filtered ("L", stream);
+      break;
+    case C_STRING_16:
+      fputs_filtered ("u", stream);
+      break;
+    case C_STRING_32:
+      fputs_filtered ("U", stream);
+      break;
+    }
+
   if (length == 0)
     {
       fputs_filtered ("\"\"", stream);
       return;
     }
 
-  for (i = 0; i < length && things_printed < options->print_max; ++i)
+  if (length == -1)
+    {
+      unsigned long current_char = 1;
+      for (i = 0; current_char; ++i)
+	{
+	  QUIT;
+	  current_char = extract_unsigned_integer (string + i * width, width);
+	}
+      length = i;
+    }
+
+  /* Arrange to iterate over the characters, in wchar_t form.  */
+  iter = make_wchar_iterator (string, length * width, encoding, width);
+  cleanup = make_cleanup_wchar_iterator (iter);
+
+  /* WCHAR_BUF is the obstack we use to represent the string in
+     wchar_t form.  */
+  obstack_init (&wchar_buf);
+  make_cleanup_obstack_free (&wchar_buf);
+
+  while (!finished && things_printed < options->print_max)
     {
-      /* Position of the character we are examining
-         to see whether it is repeated.  */
-      unsigned int rep1;
-      /* Number of repetitions we have detected so far.  */
-      unsigned int reps;
-      unsigned long current_char;
+      int num_chars;
+      enum wchar_iterate_result result;
+      gdb_wchar_t *chars;
+      const gdb_byte *buf;
+      size_t buflen;
 
       QUIT;
 
       if (need_comma)
 	{
-	  fputs_filtered (", ", stream);
+	  obstack_grow_wstr (&wchar_buf, LCST (", "));
 	  need_comma = 0;
 	}
 
-      current_char = extract_unsigned_integer (string + i * width, width);
+      num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+      /* We only look at repetitions when we were able to convert a
+	 single character in isolation.  This makes the code simpler
+	 and probably does the sensible thing in the majority of
+	 cases.  */
+      while (num_chars == 1)
+	{
+	  /* Count the number of repetitions.  */
+	  unsigned int reps = 0;
+	  gdb_wchar_t current_char = chars[0];
+	  const gdb_byte *orig_buf = buf;
+	  int orig_len = buflen;
 
-      rep1 = i + 1;
-      reps = 1;
-      while (rep1 < length
-	     && extract_unsigned_integer (string + rep1 * width, width)
-	     == current_char)
+	  if (need_comma)
+	    {
+	      obstack_grow_wstr (&wchar_buf, LCST (", "));
+	      need_comma = 0;
+	    }
+
+	  while (num_chars == 1 && current_char == chars[0])
+	    {
+	      num_chars = wchar_iterate (iter, &result, &chars, &buf, &buflen);
+	      ++reps;
+	    }
+
+	  /* Emit CURRENT_CHAR according to the repetition count and
+	     options.  */
+	  if (reps > options->repeat_count_threshold)
+	    {
+	      if (in_quotes)
+		{
+		  if (options->inspect_it)
+		    obstack_grow_wstr (&wchar_buf, LCST ("\\\", "));
+		  else
+		    obstack_grow_wstr (&wchar_buf, LCST ("\", "));
+		  in_quotes = 0;
+		}
+	      obstack_grow_wstr (&wchar_buf, LCST ("'"));
+	      need_escape = 0;
+	      print_wchar (current_char, orig_buf, orig_len, width,
+			   &wchar_buf, '\'', &need_escape);
+	      obstack_grow_wstr (&wchar_buf, LCST ("'"));
+	      {
+		/* Painful gyrations.  */
+		int j;
+		char *s = xstrprintf (_(" <repeats %u times>"), reps);
+		for (j = 0; s[j]; ++j)
+		  {
+		    gdb_wchar_t w = gdb_btowc (s[j]);
+		    obstack_grow (&wchar_buf, &w, sizeof (gdb_wchar_t));
+		  }
+		xfree (s);
+	      }
+	      things_printed += options->repeat_count_threshold;
+	      need_comma = 1;
+	    }
+	  else
+	    {
+	      /* Saw the character one or more times, but fewer than
+		 the repetition threshold.  */
+	      if (!in_quotes)
+		{
+		  if (options->inspect_it)
+		    obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
+		  else
+		    obstack_grow_wstr (&wchar_buf, LCST ("\""));
+		  in_quotes = 1;
+		  need_escape = 0;
+		}
+
+	      while (reps-- > 0)
+		{
+		  print_wchar (current_char, orig_buf, orig_len, width,
+			       &wchar_buf, '"', &need_escape);
+		  ++things_printed;
+		}
+	    }
+	}
+
+      /* NUM_CHARS and the other outputs from wchar_iterate are valid
+	 here regardless of which branch was taken above.  */
+      if (num_chars < 0)
 	{
-	  ++rep1;
-	  ++reps;
+	  /* Hit EOF.  */
+	  finished = 1;
+	  break;
 	}
 
-      if (reps > options->repeat_count_threshold)
+      switch (result)
 	{
-	  if (in_quotes)
+	case wchar_iterate_invalid:
+	  if (!in_quotes)
 	    {
 	      if (options->inspect_it)
-		fputs_filtered ("\\\", ", stream);
+		obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
 	      else
-		fputs_filtered ("\", ", stream);
-	      in_quotes = 0;
+		obstack_grow_wstr (&wchar_buf, LCST ("\""));
+	      in_quotes = 1;
 	    }
-	  LA_PRINT_CHAR (current_char, stream);
-	  fprintf_filtered (stream, _(" <repeats %u times>"), reps);
-	  i = rep1 - 1;
-	  things_printed += options->repeat_count_threshold;
-	  need_comma = 1;
-	}
-      else
-	{
-	  if (!in_quotes)
+	  need_escape = 0;
+	  print_wchar (gdb_WEOF, buf, buflen, width, &wchar_buf,
+		       '"', &need_escape);
+	  break;
+
+	case wchar_iterate_incomplete:
+	  if (in_quotes)
 	    {
 	      if (options->inspect_it)
-		fputs_filtered ("\\\"", stream);
+		obstack_grow_wstr (&wchar_buf, LCST ("\\\","));
 	      else
-		fputs_filtered ("\"", stream);
-	      in_quotes = 1;
+		obstack_grow_wstr (&wchar_buf, LCST ("\","));
+	      in_quotes = 0;
 	    }
-	  LA_EMIT_CHAR (current_char, stream, '"');
-	  ++things_printed;
+	  obstack_grow_wstr (&wchar_buf, LCST (" <incomplete sequence "));
+	  print_wchar (gdb_WEOF, buf, buflen, width, &wchar_buf,
+		       0, &need_escape);
+	  obstack_grow_wstr (&wchar_buf, LCST (">"));
+	  finished = 1;
+	  break;
 	}
     }
 
@@ -173,13 +550,27 @@ c_printstr (struct ui_file *stream, const gdb_byte *string,
   if (in_quotes)
     {
       if (options->inspect_it)
-	fputs_filtered ("\\\"", stream);
+	obstack_grow_wstr (&wchar_buf, LCST ("\\\""));
       else
-	fputs_filtered ("\"", stream);
+	obstack_grow_wstr (&wchar_buf, LCST ("\""));
     }
 
-  if (force_ellipses || i < length)
-    fputs_filtered ("...", stream);
+  if (force_ellipses || !finished)
+    obstack_grow_wstr (&wchar_buf, LCST ("..."));
+
+  /* OUTPUT is where we collect `char's for printing.  */
+  obstack_init (&output);
+  make_cleanup_obstack_free (&output);
+
+  convert_between_encodings ("wchar_t", host_charset (),
+			     obstack_base (&wchar_buf),
+			     obstack_object_size (&wchar_buf),
+			     1, &output, translit_char);
+  obstack_1grow (&output, '\0');
+
+  fputs_filtered (obstack_base (&output), stream);
+
+  do_cleanups (cleanup);
 }
 
 /* Obtain a C string from the inferior storing it in a newly allocated
@@ -298,7 +689,285 @@ c_get_string (struct value *value, gdb_byte **buffer, int *length,
 }
 
 
-/* Preprocessing and parsing C and C++ expressions.  */
+/* Evaluating C and C++ expressions.  */
+
+/* Convert a UCN.  The digits of the UCN start at P and extend no
+   farther than LIMIT.  DEST_CHARSET is the name of the character set
+   into which the UCN should be converted.  The results are written to
+   OUTPUT.  LENGTH is the maximum length of the UCN, either 4 or 8.
+   Returns a pointer to just after the final digit of the UCN.  */
+
+static char *
+convert_ucn (char *p, char *limit, const char *dest_charset,
+	     struct obstack *output, int length)
+{
+  unsigned long result = 0;
+  gdb_byte data[4];
+  int i;
+
+  for (i = 0; i < length && p < limit && isxdigit (*p); ++i, ++p)
+    result = (result << 4) + host_hex_value (*p);
+
+  for (i = 3; i >= 0; --i)
+    {
+      data[i] = result & 0xff;
+      result >>= 8;
+    }
+
+  convert_between_encodings ("UCS-4BE", dest_charset, data, 4, 4, output,
+			     translit_none);
+
+  return p;
+}
+
+/* Emit a character, VALUE, which was specified numerically, to
+   OUTPUT.  TYPE is the target character type.  */
+
+static void
+emit_numeric_character (struct type *type, unsigned long value,
+			struct obstack *output)
+{
+  gdb_byte *buffer;
+
+  buffer = alloca (TYPE_LENGTH (type));
+  pack_long (buffer, type, value);
+  obstack_grow (output, buffer, TYPE_LENGTH (type));
+}
+
+/* Convert an octal escape sequence.  TYPE is the target character
+   type.  The digits of the escape sequence begin at P and extend no
+   farther than LIMIT.  The result is written to OUTPUT.  Returns a
+   pointer to just after the final digit of the escape sequence.  */
+
+static char *
+convert_octal (struct type *type, char *p, char *limit, struct obstack *output)
+{
+  unsigned long value = 0;
+
+  while (p < limit && isdigit (*p) && *p != '8' && *p != '9')
+    {
+      value = 8 * value + host_hex_value (*p);
+      ++p;
+    }
+
+  emit_numeric_character (type, value, output);
+
+  return p;
+}
+
+/* Convert a hex escape sequence.  TYPE is the target character type.
+   The digits of the escape sequence begin at P and extend no farther
+   than LIMIT.  The result is written to OUTPUT.  Returns a pointer to
+   just after the final digit of the escape sequence.  */
+
+static char *
+convert_hex (struct type *type, char *p, char *limit, struct obstack *output)
+{
+  unsigned long value = 0;
+
+  while (p < limit && isxdigit (*p))
+    {
+      value = 16 * value + host_hex_value (*p);
+      ++p;
+    }
+
+  emit_numeric_character (type, value, output);
+
+  return p;
+}
+
+#define ADVANCE					\
+  do {						\
+    ++p;					\
+    if (p == limit)				\
+      error (_("Malformed escape sequence"));	\
+  } while (0)
+
+/* Convert an escape sequence to a target format.  TYPE is the target
+   character type to use, and DEST_CHARSET is the name of the target
+   character set.  The backslash of the escape sequence is at *P, and
+   the escape sequence will not extend past LIMIT.  The results are
+   written to OUTPUT.  Returns a pointer to just past the final
+   character of the escape sequence.  */
+
+static char *
+convert_escape (struct type *type, const char *dest_charset,
+		char *p, char *limit, struct obstack *output)
+{
+  /* Skip the backslash.  */
+  ADVANCE;
+
+  switch (*p)
+    {
+    case '\\':
+      obstack_1grow (output, '\\');
+      ++p;
+      break;
+
+    case 'x':
+      ADVANCE;
+      if (!isxdigit (*p))
+	error (_("\\x used with no following hex digits."));
+      p = convert_hex (type, p, limit, output);
+      break;
+
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+      p = convert_octal (type, p, limit, output);
+      break;
+
+    case 'u':
+    case 'U':
+      {
+	int length = *p == 'u' ? 4 : 8;
+	ADVANCE;
+	if (!isxdigit (*p))
+	  error (_("\\u used with no following hex digits"));
+	p = convert_ucn (p, limit, dest_charset, output, length);
+      }
+    }
+
+  return p;
+}
+
+/* Given a single string from a (C-specific) OP_STRING list, convert
+   it to a target string, handling escape sequences specially.  The
+   output is written to OUTPUT.  DATA is the input string, which has
+   length LEN.  DEST_CHARSET is the name of the target character set,
+   and TYPE is the type of target character to use.  */
+
+static void
+parse_one_string (struct obstack *output, char *data, int len,
+		  const char *dest_charset, struct type *type)
+{
+  char *limit;
+
+  limit = data + len;
+
+  while (data < limit)
+    {
+      char *p = data;
+      /* Look for next escape, or the end of the input.  */
+      while (p < limit && *p != '\\')
+	++p;
+      /* If we saw a run of characters, convert them all.  */
+      if (p > data)
+	convert_between_encodings (host_charset (), dest_charset,
+				   data, p - data, 1, output, translit_none);
+      /* If we saw an escape, convert it.  */
+      if (p < limit)
+	p = convert_escape (type, dest_charset, p, limit, output);
+      data = p;
+    }
+}
+
+/* Expression evaluator for the C language family.  Most operations
+   are delegated to evaluate_subexp_standard; see that function for a
+   description of the arguments.  */
+
+static struct value *
+evaluate_subexp_c (struct type *expect_type, struct expression *exp,
+		   int *pos, enum noside noside)
+{
+  enum exp_opcode op = exp->elts[*pos].opcode;
+
+  switch (op)
+    {
+    case OP_STRING:
+      {
+	int oplen, limit;
+	struct type *type;
+	struct obstack output;
+	struct cleanup *cleanup;
+	struct value *result;
+	enum c_string_type dest_type;
+	const char *dest_charset;
+
+	obstack_init (&output);
+	cleanup = make_cleanup_obstack_free (&output);
+
+	++*pos;
+	oplen = longest_to_int (exp->elts[*pos].longconst);
+
+	++*pos;
+	limit = *pos + BYTES_TO_EXP_ELEM (oplen + 1);
+	dest_type
+	  = (enum c_string_type) longest_to_int (exp->elts[*pos].longconst);
+	switch (dest_type & ~C_CHAR)
+	  {
+	  case C_STRING:
+	    type = language_string_char_type (current_language,
+					      current_gdbarch);
+	    break;
+	  case C_WIDE_STRING:
+	    type = lookup_typename ("wchar_t", NULL, 0);
+	    break;
+	  case C_STRING_16:
+	    type = lookup_typename ("char16_t", NULL, 0);
+	    break;
+	  case C_STRING_32:
+	    type = lookup_typename ("char32_t", NULL, 0);
+	    break;
+	  default:
+	    internal_error (__FILE__, __LINE__, "unhandled c_string_type");
+	  }
+	dest_charset = charset_for_string_type (dest_type);
+
+	++*pos;
+	while (*pos < limit)
+	  {
+	    int len;
+
+	    len = longest_to_int (exp->elts[*pos].longconst);
+
+	    ++*pos;
+	    if (noside != EVAL_SKIP)
+	      parse_one_string (&output, &exp->elts[*pos].string, len,
+				dest_charset, type);
+	    *pos += BYTES_TO_EXP_ELEM (len);
+	  }
+
+	/* Skip the trailing length and opcode.  */
+	*pos += 2;
+
+	if (noside == EVAL_SKIP)
+	  return NULL;
+
+	if ((dest_type & C_CHAR) != 0)
+	  {
+	    LONGEST value;
+
+	    if (obstack_object_size (&output) != TYPE_LENGTH (type))
+	      error (_("Could not convert character constant to target character set"));
+	    value = unpack_long (type, obstack_base (&output));
+	    result = value_from_longest (type, value);
+	  }
+	else
+	  {
+	    int i;
+	    /* Write the terminating character.  */
+	    for (i = 0; i < TYPE_LENGTH (type); ++i)
+	      obstack_1grow (&output, 0);
+	    result = value_typed_string (obstack_base (&output),
+					 obstack_object_size (&output),
+					 type);
+	  }
+	do_cleanups (cleanup);
+	return result;
+      }
+      break;
+
+    default:
+      break;
+    }
+  return evaluate_subexp_standard (expect_type, exp, pos, noside);
+}
 
 
 
@@ -396,6 +1065,15 @@ c_language_arch_info (struct gdbarch *gdbarch,
   lai->bool_type_default = builtin->builtin_int;
 }
 
+static const struct exp_descriptor exp_descriptor_c = 
+{
+  print_subexp_standard,
+  operator_length_standard,
+  op_name_standard,
+  dump_subexp_body_standard,
+  evaluate_subexp_c
+};
+
 const struct language_defn c_language_defn =
 {
   "c",				/* Language name */
@@ -405,7 +1083,7 @@ const struct language_defn c_language_defn =
   case_sensitive_on,
   array_row_major,
   macro_expansion_c,
-  &exp_descriptor_standard,
+  &exp_descriptor_c,
   c_parse,
   c_error,
   null_post_parser,
@@ -524,7 +1202,7 @@ const struct language_defn cplus_language_defn =
   case_sensitive_on,
   array_row_major,
   macro_expansion_c,
-  &exp_descriptor_standard,
+  &exp_descriptor_c,
   c_parse,
   c_error,
   null_post_parser,
@@ -562,7 +1240,7 @@ const struct language_defn asm_language_defn =
   case_sensitive_on,
   array_row_major,
   macro_expansion_c,
-  &exp_descriptor_standard,
+  &exp_descriptor_c,
   c_parse,
   c_error,
   null_post_parser,
@@ -605,7 +1283,7 @@ const struct language_defn minimal_language_defn =
   case_sensitive_on,
   array_row_major,
   macro_expansion_c,
-  &exp_descriptor_standard,
+  &exp_descriptor_c,
   c_parse,
   c_error,
   null_post_parser,