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, 62 insertions, 65 deletions

diff --git a/gdb/utils.c b/gdb/utils.c
index 0becfd9..5338868 100644
--- a/gdb/utils.c
+++ b/gdb/utils.c
@@ -1571,21 +1571,33 @@ query (const char *ctlstr, ...)
   va_end (args);
 }
 
-/* Print an error message saying that we couldn't make sense of a
-   \^mumble sequence in a string or character constant.  START and END
-   indicate a substring of some larger string that contains the
-   erroneous backslash sequence, missing the initial backslash.  */
-static NORETURN int
-no_control_char_error (const char *start, const char *end)
+/* A helper for parse_escape that converts a host character to a
+   target character.  C is the host character.  If conversion is
+   possible, then the target character is stored in *TARGET_C and the
+   function returns 1.  Otherwise, the function returns 0.  */
+
+static int
+host_char_to_target (int c, int *target_c)
 {
-  int len = end - start;
-  char *copy = alloca (end - start + 1);
+  struct obstack host_data;
+  char the_char = c;
+  struct cleanup *cleanups;
+  int result = 0;
+
+  obstack_init (&host_data);
+  cleanups = make_cleanup_obstack_free (&host_data);
 
-  memcpy (copy, start, len);
-  copy[len] = '\0';
+  convert_between_encodings (target_charset (), host_charset (),
+			     &the_char, 1, 1, &host_data, translit_none);
+
+  if (obstack_object_size (&host_data) == 1)
+    {
+      result = 1;
+      *target_c = *(char *) obstack_base (&host_data);
+    }
 
-  error (_("There is no control character `\\%s' in the `%s' character set."),
-	 copy, target_charset ());
+  do_cleanups (cleanups);
+  return result;
 }
 
 /* Parse a C escape sequence.  STRING_PTR points to a variable
@@ -1608,53 +1620,13 @@ parse_escape (char **string_ptr)
 {
   int target_char;
   int c = *(*string_ptr)++;
-  if (c_parse_backslash (c, &target_char))
-    return target_char;
-  else
-    switch (c)
-      {
+  switch (c)
+    {
       case '\n':
 	return -2;
       case 0:
 	(*string_ptr)--;
 	return 0;
-      case '^':
-	{
-	  /* Remember where this escape sequence started, for reporting
-	     errors.  */
-	  char *sequence_start_pos = *string_ptr - 1;
-
-	  c = *(*string_ptr)++;
-
-	  if (c == '?')
-	    {
-	      /* XXXCHARSET: What is `delete' in the host character set?  */
-	      c = 0177;
-
-	      if (!host_char_to_target (c, &target_char))
-		error (_("There is no character corresponding to `Delete' "
-		       "in the target character set `%s'."), host_charset ());
-
-	      return target_char;
-	    }
-	  else if (c == '\\')
-	    target_char = parse_escape (string_ptr);
-	  else
-	    {
-	      if (!host_char_to_target (c, &target_char))
-		no_control_char_error (sequence_start_pos, *string_ptr);
-	    }
-
-	  /* Now target_char is something like `c', and we want to find
-	     its control-character equivalent.  */
-	  if (!target_char_to_control_char (target_char, &target_char))
-	    no_control_char_error (sequence_start_pos, *string_ptr);
-
-	  return target_char;
-	}
-
-	/* XXXCHARSET: we need to use isdigit and value-of-digit
-	   methods of the host character set here.  */
 
       case '0':
       case '1':
@@ -1665,16 +1637,16 @@ parse_escape (char **string_ptr)
       case '6':
       case '7':
 	{
-	  int i = c - '0';
+	  int i = host_hex_value (c);
 	  int count = 0;
 	  while (++count < 3)
 	    {
 	      c = (**string_ptr);
-	      if (c >= '0' && c <= '7')
+	      if (isdigit (c) && c != '8' && c != '9')
 		{
 		  (*string_ptr)++;
 		  i *= 8;
-		  i += c - '0';
+		  i += host_hex_value (c);
 		}
 	      else
 		{
@@ -1683,14 +1655,39 @@ parse_escape (char **string_ptr)
 	    }
 	  return i;
 	}
-      default:
-	if (!host_char_to_target (c, &target_char))
-	  error
-	    ("The escape sequence `\%c' is equivalent to plain `%c', which"
-	     " has no equivalent\n" "in the `%s' character set.", c, c,
-	     target_charset ());
-	return target_char;
-      }
+
+    case 'a':
+      c = '\a';
+      break;
+    case 'b':
+      c = '\b';
+      break;
+    case 'f':
+      c = '\f';
+      break;
+    case 'n':
+      c = '\n';
+      break;
+    case 'r':
+      c = '\r';
+      break;
+    case 't':
+      c = '\t';
+      break;
+    case 'v':
+      c = '\v';
+      break;
+
+    default:
+      break;
+    }
+
+  if (!host_char_to_target (c, &target_char))
+    error
+      ("The escape sequence `\%c' is equivalent to plain `%c', which"
+       " has no equivalent\n" "in the `%s' character set.", c, c,
+       target_charset ());
+  return target_char;
 }
 
 /* Print the character C on STREAM as part of the contents of a literal