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.

gdb/ChangeLog

@@ -1,3 +1,217 @@
+2009-03-20  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.
+
 2009-03-20  Tom Tromey  <tromey@redhat.com>
 	    Jan Kratochvil  <jan.kratochvil@redhat.com>
 

gdb/Makefile.in

@@ -165,6 +165,8 @@ INTL_CFLAGS = @INCINTL@
 
 # Where is the ICONV library?  This can be empty if libc has iconv.
 LIBICONV = @LIBICONV@
+LIBICONV_INCLUDE = @LIBICONV_INCLUDE@
+LIBICONV_LIBDIR = @LIBICONV_LIBDIR@
 
 # Did the user give us a --with-sysroot option?
 TARGET_SYSTEM_ROOT = @TARGET_SYSTEM_ROOT@
@@ -388,7 +390,8 @@ INTERNAL_CFLAGS_BASE = \
 	$(CFLAGS) $(GLOBAL_CFLAGS) $(PROFILE_CFLAGS) \
 	$(GDB_CFLAGS) $(OPCODES_CFLAGS) $(READLINE_CFLAGS) \
 	$(BFD_CFLAGS) $(INCLUDE_CFLAGS) $(LIBDECNUMBER_CFLAGS) \
-	$(INTL_CFLAGS) $(INCGNU) $(ENABLE_CFLAGS) $(INTERNAL_CPPFLAGS)
+	$(INTL_CFLAGS) $(INCGNU) $(ENABLE_CFLAGS) $(INTERNAL_CPPFLAGS) \
+	$(LIBICONV_INCLUDE)
 INTERNAL_WARN_CFLAGS = $(INTERNAL_CFLAGS_BASE) $(GDB_WARN_CFLAGS)
 INTERNAL_CFLAGS = $(INTERNAL_WARN_CFLAGS) $(GDB_WERROR_CFLAGS)
 
@@ -400,7 +403,7 @@ LDFLAGS = @LDFLAGS@
 # I think it's perfectly reasonable for a user to set -pg in CFLAGS
 # and have it work; that's why CFLAGS is here.
 # PROFILE_CFLAGS is _not_ included, however, because we use monstartup.
-INTERNAL_LDFLAGS = $(CFLAGS) $(GLOBAL_CFLAGS) $(MH_LDFLAGS) $(LDFLAGS) $(CONFIG_LDFLAGS)
+INTERNAL_LDFLAGS = $(CFLAGS) $(GLOBAL_CFLAGS) $(MH_LDFLAGS) $(LDFLAGS) $(CONFIG_LDFLAGS) $(LIBICONV_LIBDIR)
 
 # If your system is missing alloca(), or, more likely, it's there but
 # it doesn't work, then refer to libiberty.

gdb/NEWS

@@ -3,6 +3,15 @@
 
 *** Changes since GDB 6.8
 
+* GDB now has support for multi-byte and wide character sets on the
+target.  Strings whose character type is wchar_t, char16_t, or
+char32_t are now correctly printed.  GDB supports wide- and unicode-
+literals in C, that is, L'x', L"string", u'x', u"string", U'x', and
+U"string" syntax.  And, GDB allows the "%ls" and "%lc" formats in
+`printf'.  This feature requires iconv to work properly; if your
+system does not have a working iconv, GDB can use GNU libiconv.  See
+the installation instructions for more information.
+
 * GDB now supports automatic retrieval of shared library files from
 remote targets.  To use this feature, specify a system root that begins
 with the `remote:' prefix, either via the `set sysroot' command or via
@@ -182,6 +191,11 @@ set target-async
   with GDB while the target is running.  "show target-async" displays the
   current state of asynchronous execution of the target.
 
+set target-wide-charset
+show target-wide-charset
+  The target-wide-charset is the name of the character set that GDB
+  uses when printing characters whose type is wchar_t.
+
 set tcp auto-retry (on|off)
 show tcp auto-retry
 set tcp connect-timeout

gdb/acinclude.m4

@@ -29,6 +29,9 @@ sinclude([../config/depstand.m4])
 dnl For AM_LC_MESSAGES
 sinclude([../config/lcmessage.m4])
 
+dnl For AM_LANGINFO_CODESET.
+sinclude([../config/codeset.m4])
+
 #
 # Sometimes the native compiler is a bogus stub for gcc or /usr/ucb/cc. This
 # makes configure think it's cross compiling. If --target wasn't used, then
@@ -174,14 +177,18 @@ AC_DEFUN([AM_ICONV],
   AC_ARG_WITH([libiconv-prefix],
 [  --with-libiconv-prefix=DIR  search for libiconv in DIR/include and DIR/lib], [
     for dir in `echo "$withval" | tr : ' '`; do
-      if test -d $dir/include; then CPPFLAGS="$CPPFLAGS -I$dir/include"; fi
+      if test -d $dir/include; then LIBICONV_INCLUDE="-I$dir/include"; CPPFLAGS="$CPPFLAGS -I$dir/include"; fi
-      if test -d $dir/lib; then LDFLAGS="$LDFLAGS -L$dir/lib"; fi
+      if test -d $dir/lib; then LIBICONV_LIBDIR="-L$dir/lib"; LDFLAGS="$LDFLAGS -L$dir/lib"; fi
     done
    ])
 
+  BUILD_LIBICONV_LIBDIR="-L../libiconv/lib/.libs -L../libiconv/lib/_libs"
+  BUILD_LIBICONV_INCLUDE="-I../libiconv/include"
+
   AC_CACHE_CHECK(for iconv, am_cv_func_iconv, [
     am_cv_func_iconv="no, consider installing GNU libiconv"
     am_cv_lib_iconv=no
+    am_cv_use_build_libiconv=no
     AC_TRY_LINK([#include <stdlib.h>
 #include <iconv.h>],
       [iconv_t cd = iconv_open("","");
@@ -200,6 +207,26 @@ AC_DEFUN([AM_ICONV],
         am_cv_func_iconv=yes)
       LIBS="$am_save_LIBS"
     fi
+    # Look for libiconv in the build tree.
+    if test "$am_cv_func_iconv" != yes && test -d ../libiconv; then
+      am_save_LIBS="$LIBS"
+      am_save_CPPFLAGS="$CPPFLAGS"
+      LIBS="$LIBS $BUILD_LIBICONV_LIBDIR -liconv"
+      CPPFLAGS="$CPPFLAGS $BUILD_LIBICONV_INCLUDE"
+      AC_TRY_LINK([#include <stdlib.h>
+#include <iconv.h>],
+        [iconv_t cd = iconv_open("","");
+         iconv(cd,NULL,NULL,NULL,NULL);
+         iconv_close(cd);],
+        am_cv_lib_iconv=yes
+        am_cv_func_iconv=yes)
+      LIBS="$am_save_LIBS"
+      if test "$am_cv_func_iconv" = "yes"; then
+         am_cv_use_build_libiconv=yes
+      else
+      	 CPPFLAGS="$am_save_CPPFLAGS"
+      fi
+    fi
   ])
   if test "$am_cv_func_iconv" = yes; then
     AC_DEFINE(HAVE_ICONV, 1, [Define if you have the iconv() function.])
@@ -229,7 +256,13 @@ size_t iconv();
   if test "$am_cv_lib_iconv" = yes; then
     LIBICONV="-liconv"
   fi
+  if test "$am_cv_use_build_libiconv" = yes; then
+    LIBICONV_LIBDIR="$BUILD_LIBICONV_LIBDIR"
+    LIBICONV_INCLUDE="$BUILD_LIBICONV_INCLUDE"
+  fi
   AC_SUBST(LIBICONV)
+  AC_SUBST(LIBICONV_INCLUDE)
+  AC_SUBST(LIBICONV_LIBDIR)
 ])
 
 dnl written by Guido Draheim <guidod@gmx.de>, original by Alexandre Oliva 

gdb/ada-lang.c

@@ -11022,9 +11022,9 @@ ada_language_arch_info (struct gdbarch *gdbarch,
 /* Not really used, but needed in the ada_language_defn.  */
 
 static void
-emit_char (int c, struct ui_file *stream, int quoter)
+emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
-  ada_emit_char (c, stream, quoter, 1);
+  ada_emit_char (c, type, stream, quoter, 1);
 }
 
 static int

gdb/ada-lang.h

@@ -255,12 +255,12 @@ extern int ada_value_print (struct value *, struct ui_file *,
 
                                 /* Defined in ada-lang.c */
 
-extern void ada_emit_char (int, struct ui_file *, int, int);
+extern void ada_emit_char (int, struct type *, struct ui_file *, int, int);
 
-extern void ada_printchar (int, struct ui_file *);
+extern void ada_printchar (int, struct type *, struct ui_file *);
 
-extern void ada_printstr (struct ui_file *, const gdb_byte *,
+extern void ada_printstr (struct ui_file *, struct type *, const gdb_byte *,
-			  unsigned int, int, int,
+			  unsigned int, int,
 			  const struct value_print_options *);
 
 struct value *ada_convert_actual (struct value *actual,

gdb/ada-valprint.c

@@ -269,7 +269,8 @@ printable_val_type (struct type *type, const gdb_byte *valaddr)
    (1 or 2) of the character.  */
 
 void
-ada_emit_char (int c, struct ui_file *stream, int quoter, int type_len)
+ada_emit_char (int c, struct type *type, struct ui_file *stream,
+	       int quoter, int type_len)
 {
   if (type_len != 2)
     type_len = 1;
@@ -366,10 +367,10 @@ ada_print_floating (const gdb_byte *valaddr, struct type *type,
 }
 
 void
-ada_printchar (int c, struct ui_file *stream)
+ada_printchar (int c, struct type *type, struct ui_file *stream)
 {
   fputs_filtered ("'", stream);
-  ada_emit_char (c, stream, '\'', 1);
+  ada_emit_char (c, type, stream, '\'', 1);
   fputs_filtered ("'", stream);
 }
 
@@ -411,7 +412,7 @@ ada_print_scalar (struct type *type, LONGEST val, struct ui_file *stream)
       break;
 
     case TYPE_CODE_CHAR:
-      LA_PRINT_CHAR ((unsigned char) val, stream);
+      LA_PRINT_CHAR ((unsigned char) val, type, stream);
       break;
 
     case TYPE_CODE_BOOL:
@@ -454,7 +455,7 @@ ada_print_scalar (struct type *type, LONGEST val, struct ui_file *stream)
  */
 
 static void
-printstr (struct ui_file *stream, const gdb_byte *string,
+printstr (struct ui_file *stream, struct type *elttype, const gdb_byte *string,
 	  unsigned int length, int force_ellipses, int type_len,
 	  const struct value_print_options *options)
 {
@@ -506,7 +507,7 @@ printstr (struct ui_file *stream, const gdb_byte *string,
 	      in_quotes = 0;
 	    }
 	  fputs_filtered ("'", stream);
-	  ada_emit_char (char_at (string, i, type_len), stream, '\'',
+	  ada_emit_char (char_at (string, i, type_len), elttype, stream, '\'',
 			 type_len);
 	  fputs_filtered ("'", stream);
 	  fprintf_filtered (stream, _(" <repeats %u times>"), reps);
@@ -524,7 +525,7 @@ printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("\"", stream);
 	      in_quotes = 1;
 	    }
-	  ada_emit_char (char_at (string, i, type_len), stream, '"',
+	  ada_emit_char (char_at (string, i, type_len), elttype, stream, '"',
 			 type_len);
 	  things_printed += 1;
 	}
@@ -544,11 +545,12 @@ printstr (struct ui_file *stream, const gdb_byte *string,
 }
 
 void
-ada_printstr (struct ui_file *stream, const gdb_byte *string,
+ada_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
-	      unsigned int length, int width, int force_ellipses,
+	      unsigned int length, int force_ellipses,
 	      const struct value_print_options *options)
 {
-  printstr (stream, string, length, force_ellipses, width, options);
+  printstr (stream, type, string, length, force_ellipses, TYPE_LENGTH (type),
+	    options);
 }
 
 
@@ -637,7 +639,7 @@ ada_val_print_array (struct type *type, const gdb_byte *valaddr,
           len = temp_len;
         }
 
-      printstr (stream, valaddr, len, 0, eltlen, options);
+      printstr (stream, elttype, valaddr, len, 0, eltlen, options);
       result = len;
     }
   else
@@ -817,7 +819,7 @@ ada_val_print_1 (struct type *type, const gdb_byte *valaddr0,
 		{
 		  fputs_filtered (" ", stream);
 		  ada_printchar ((unsigned char) unpack_long (type, valaddr),
-				 stream);
+				 type, stream);
 		}
 	    }
 	  return 0;

gdb/auxv.c

@@ -248,7 +248,8 @@ fprint_target_auxv (struct ui_file *file, struct target_ops *ops)
 	    get_user_print_options (&opts);
 	    if (opts.addressprint)
 	      fprintf_filtered (file, "0x%s", paddr_nz (val));
-	    val_print_string (val, -1, 1, file, &opts);
+	    val_print_string (builtin_type (target_gdbarch)->builtin_char,
+			      val, -1, file, &opts);
 	    fprintf_filtered (file, "\n");
 	  }
 	  break;

gdb/c-exp.y

@@ -141,6 +141,7 @@ void yyerror (char *);
     struct symbol *sym;
     struct type *tval;
     struct stoken sval;
+    struct typed_stoken tsval;
     struct ttype tsym;
     struct symtoken ssym;
     int voidval;
@@ -148,6 +149,7 @@ void yyerror (char *);
     enum exp_opcode opcode;
     struct internalvar *ivar;
 
+    struct stoken_vector svec;
     struct type **tvec;
     int *ivec;
   }
@@ -180,11 +182,13 @@ static int parse_number (char *, int, int, YYSTYPE *);
    Contexts where this distinction is not important can use the
    nonterminal "name", which matches either NAME or TYPENAME.  */
 
-%token <sval> STRING
+%token <tsval> STRING
+%token <tsval> CHAR
 %token <ssym> NAME /* BLOCKNAME defined below to give it higher precedence. */
 %token <voidval> COMPLETE
 %token <tsym> TYPENAME
-%type <sval> name string_exp
+%type <sval> name
+%type <svec> string_exp
 %type <ssym> name_not_typename
 %type <tsym> typename
 
@@ -522,6 +526,15 @@ exp	:	INT
 			  write_exp_elt_opcode (OP_LONG); }
 	;
 
+exp	:	CHAR
+			{
+			  struct stoken_vector vec;
+			  vec.len = 1;
+			  vec.tokens = &$1;
+			  write_exp_string_vector ($1.type, &vec);
+			}
+	;
+
 exp	:	NAME_OR_INT
 			{ YYSTYPE val;
 			  parse_number ($1.stoken.ptr, $1.stoken.length, 0, &val);
@@ -570,48 +583,64 @@ string_exp:
 			     string.  Note that we follow the
 			     NUL-termination convention of the
 			     lexer.  */
-			  $$.length = $1.length;
+			  struct typed_stoken *vec = XNEW (struct typed_stoken);
-			  $$.ptr = malloc ($1.length + 1);
+			  $$.len = 1;
-			  memcpy ($$.ptr, $1.ptr, $1.length + 1);
+			  $$.tokens = vec;
+
+			  vec->type = $1.type;
+			  vec->length = $1.length;
+			  vec->ptr = malloc ($1.length + 1);
+			  memcpy (vec->ptr, $1.ptr, $1.length + 1);
 			}
 
 	|	string_exp STRING
 			{
 			  /* Note that we NUL-terminate here, but just
 			     for convenience.  */
-			  struct stoken t;
+			  char *p;
-			  t.length = $1.length + $2.length;
+			  ++$$.len;
-			  t.ptr = malloc (t.length + 1);
+			  $$.tokens = realloc ($$.tokens,
-			  memcpy (t.ptr, $1.ptr, $1.length);
+					       $$.len * sizeof (struct typed_stoken));
-			  memcpy (t.ptr + $1.length, $2.ptr, $2.length + 1);
+
-			  free ($1.ptr);
+			  p = malloc ($2.length + 1);
-			  $$ = t;
+			  memcpy (p, $2.ptr, $2.length + 1);
+
+			  $$.tokens[$$.len - 1].type = $2.type;
+			  $$.tokens[$$.len - 1].length = $2.length;
+			  $$.tokens[$$.len - 1].ptr = p;
 			}
 		;
 
 exp	:	string_exp
-			{ /* C strings are converted into array constants with
+			{
-			     an explicit null byte added at the end.  Thus
+			  int i;
-			     the array upper bound is the string length.
+			  enum c_string_type type = C_STRING;
-			     There is no such thing in C as a completely empty
+
-			     string. */
+			  for (i = 0; i < $1.len; ++i)
-			  char *sp = $1.ptr; int count = $1.length;
-			  while (count-- > 0)
 			    {
-			      write_exp_elt_opcode (OP_LONG);
+			      switch ($1.tokens[i].type)
-			      write_exp_elt_type (parse_type->builtin_char);
+				{
-			      write_exp_elt_longcst ((LONGEST)(*sp++));
+				case C_STRING:
-			      write_exp_elt_opcode (OP_LONG);
+				  break;
+				case C_WIDE_STRING:
+				case C_STRING_16:
+				case C_STRING_32:
+				  if (type != C_STRING
+				      && type != $1.tokens[i].type)
+				    error ("Undefined string concatenation.");
+				  type = $1.tokens[i].type;
+				  break;
+				default:
+				  /* internal error */
+				  internal_error (__FILE__, __LINE__,
+						  "unrecognized type in string concatenation");
+				}
 			    }
-			  write_exp_elt_opcode (OP_LONG);
+
-			  write_exp_elt_type (parse_type->builtin_char);
+			  write_exp_string_vector (type, &$1);
-			  write_exp_elt_longcst ((LONGEST)'\0');
+			  for (i = 0; i < $1.len; ++i)
-			  write_exp_elt_opcode (OP_LONG);
+			    free ($1.tokens[i].ptr);
-			  write_exp_elt_opcode (OP_ARRAY);
+			  free ($1.tokens);
-			  write_exp_elt_longcst ((LONGEST) 0);
-			  write_exp_elt_longcst ((LONGEST) ($1.length));
-			  write_exp_elt_opcode (OP_ARRAY);
-			  free ($1.ptr);
 			}
 	;
 
@@ -1359,6 +1388,263 @@ parse_number (p, len, parsed_float, putithere)
    return INT;
 }
 
+/* Temporary obstack used for holding strings.  */
+static struct obstack tempbuf;
+static int tempbuf_init;
+
+/* Parse a C escape sequence.  The initial backslash of the sequence
+   is at (*PTR)[-1].  *PTR will be updated to point to just after the
+   last character of the sequence.  If OUTPUT is not NULL, the
+   translated form of the escape sequence will be written there.  If
+   OUTPUT is NULL, no output is written and the call will only affect
+   *PTR.  If an escape sequence is expressed in target bytes, then the
+   entire sequence will simply be copied to OUTPUT.  Return 1 if any
+   character was emitted, 0 otherwise.  */
+
+int
+c_parse_escape (char **ptr, struct obstack *output)
+{
+  char *tokptr = *ptr;
+  int result = 1;
+
+  /* Some escape sequences undergo character set conversion.  Those we
+     translate here.  */
+  switch (*tokptr)
+    {
+      /* Hex escapes do not undergo character set conversion, so keep
+	 the escape sequence for later.  */
+    case 'x':
+      if (output)
+	obstack_grow_str (output, "\\x");
+      ++tokptr;
+      if (!isxdigit (*tokptr))
+	error (_("\\x escape without a following hex digit"));
+      while (isxdigit (*tokptr))
+	{
+	  if (output)
+	    obstack_1grow (output, *tokptr);
+	  ++tokptr;
+	}
+      break;
+
+      /* Octal escapes do not undergo character set conversion, so
+	 keep the escape sequence for later.  */
+    case '0':
+    case '1':
+    case '2':
+    case '3':
+    case '4':
+    case '5':
+    case '6':
+    case '7':
+      if (output)
+	obstack_grow_str (output, "\\");
+      while (isdigit (*tokptr) && *tokptr != '8' && *tokptr != '9')
+	{
+	  if (output)
+	    obstack_1grow (output, *tokptr);
+	  ++tokptr;
+	}
+      break;
+
+      /* We handle UCNs later.  We could handle them here, but that
+	 would mean a spurious error in the case where the UCN could
+	 be converted to the target charset but not the host
+	 charset.  */
+    case 'u':
+    case 'U':
+      {
+	char c = *tokptr;
+	int i, len = c == 'U' ? 8 : 4;
+	if (output)
+	  {
+	    obstack_1grow (output, '\\');
+	    obstack_1grow (output, *tokptr);
+	  }
+	++tokptr;
+	if (!isxdigit (*tokptr))
+	  error (_("\\%c escape without a following hex digit"), c);
+	for (i = 0; i < len && isxdigit (*tokptr); ++i)
+	  {
+	    if (output)
+	      obstack_1grow (output, *tokptr);
+	    ++tokptr;
+	  }
+      }
+      break;
+
+      /* We must pass backslash through so that it does not
+	 cause quoting during the second expansion.  */
+    case '\\':
+      if (output)
+	obstack_grow_str (output, "\\\\");
+      ++tokptr;
+      break;
+
+      /* Escapes which undergo conversion.  */
+    case 'a':
+      if (output)
+	obstack_1grow (output, '\a');
+      ++tokptr;
+      break;
+    case 'b':
+      if (output)
+	obstack_1grow (output, '\b');
+      ++tokptr;
+      break;
+    case 'f':
+      if (output)
+	obstack_1grow (output, '\f');
+      ++tokptr;
+      break;
+    case 'n':
+      if (output)
+	obstack_1grow (output, '\n');
+      ++tokptr;
+      break;
+    case 'r':
+      if (output)
+	obstack_1grow (output, '\r');
+      ++tokptr;
+      break;
+    case 't':
+      if (output)
+	obstack_1grow (output, '\t');
+      ++tokptr;
+      break;
+    case 'v':
+      if (output)
+	obstack_1grow (output, '\v');
+      ++tokptr;
+      break;
+
+      /* GCC extension.  */
+    case 'e':
+      if (output)
+	obstack_1grow (output, HOST_ESCAPE_CHAR);
+      ++tokptr;
+      break;
+
+      /* Backslash-newline expands to nothing at all.  */
+    case '\n':
+      ++tokptr;
+      result = 0;
+      break;
+
+      /* A few escapes just expand to the character itself.  */
+    case '\'':
+    case '\"':
+    case '?':
+      /* GCC extensions.  */
+    case '(':
+    case '{':
+    case '[':
+    case '%':
+      /* Unrecognized escapes turn into the character itself.  */
+    default:
+      if (output)
+	obstack_1grow (output, *tokptr);
+      ++tokptr;
+      break;
+    }
+  *ptr = tokptr;
+  return result;
+}
+
+/* Parse a string or character literal from TOKPTR.  The string or
+   character may be wide or unicode.  *OUTPTR is set to just after the
+   end of the literal in the input string.  The resulting token is
+   stored in VALUE.  This returns a token value, either STRING or
+   CHAR, depending on what was parsed.  *HOST_CHARS is set to the
+   number of host characters in the literal.  */
+static int
+parse_string_or_char (char *tokptr, char **outptr, struct typed_stoken *value,
+		      int *host_chars)
+{
+  int quote, i;
+  enum c_string_type type;
+
+  /* Build the gdb internal form of the input string in tempbuf.  Note
+     that the buffer is null byte terminated *only* for the
+     convenience of debugging gdb itself and printing the buffer
+     contents when the buffer contains no embedded nulls.  Gdb does
+     not depend upon the buffer being null byte terminated, it uses
+     the length string instead.  This allows gdb to handle C strings
+     (as well as strings in other languages) with embedded null
+     bytes */
+
+  if (!tempbuf_init)
+    tempbuf_init = 1;
+  else
+    obstack_free (&tempbuf, NULL);
+  obstack_init (&tempbuf);
+
+  /* Record the string type.  */
+  if (*tokptr == 'L')
+    {
+      type = C_WIDE_STRING;
+      ++tokptr;
+    }
+  else if (*tokptr == 'u')
+    {
+      type = C_STRING_16;
+      ++tokptr;
+    }
+  else if (*tokptr == 'U')
+    {
+      type = C_STRING_32;
+      ++tokptr;
+    }
+  else
+    type = C_STRING;
+
+  /* Skip the quote.  */
+  quote = *tokptr;
+  if (quote == '\'')
+    type |= C_CHAR;
+  ++tokptr;
+
+  *host_chars = 0;
+
+  while (*tokptr)
+    {
+      char c = *tokptr;
+      if (c == '\\')
+	{
+	  ++tokptr;
+	  *host_chars += c_parse_escape (&tokptr, &tempbuf);
+	}
+      else if (c == quote)
+	break;
+      else
+	{
+	  obstack_1grow (&tempbuf, c);
+	  ++tokptr;
+	  /* FIXME: this does the wrong thing with multi-byte host
+	     characters.  We could use mbrlen here, but that would
+	     make "set host-charset" a bit less useful.  */
+	  ++*host_chars;
+	}
+    }
+
+  if (*tokptr != quote)
+    {
+      if (quote == '"')
+	error ("Unterminated string in expression.");
+      else
+	error ("Unmatched single quote.");
+    }
+  ++tokptr;
+
+  value->type = type;
+  value->ptr = obstack_base (&tempbuf);
+  value->length = obstack_object_size (&tempbuf);
+
+  *outptr = tokptr;
+
+  return quote == '"' ? STRING : CHAR;
+}
+
 struct token
 {
   char *operator;
@@ -1528,12 +1814,6 @@ yylex ()
   int namelen;
   unsigned int i;
   char *tokstart;
-  char *tokptr;
-  int tempbufindex;
-  static char *tempbuf;
-  static int tempbufsize;
-  char * token_string = NULL;
-  int class_prefix = 0;
   int saw_structop = last_was_structop;
   char *copy;
 
@@ -1605,46 +1885,6 @@ yylex ()
       lexptr++;
       goto retry;
 
-    case '\'':
-      /* We either have a character constant ('0' or '\177' for example)
-	 or we have a quoted symbol reference ('foo(int,int)' in C++
-	 for example). */
-      lexptr++;
-      c = *lexptr++;
-      if (c == '\\')
-	c = parse_escape (&lexptr);
-      else if (c == '\'')
-	error ("Empty character constant.");
-      else if (! host_char_to_target (c, &c))
-        {
-          int toklen = lexptr - tokstart + 1;
-          char *tok = alloca (toklen + 1);
-          memcpy (tok, tokstart, toklen);
-          tok[toklen] = '\0';
-          error ("There is no character corresponding to %s in the target "
-                 "character set `%s'.", tok, target_charset ());
-        }
-
-      yylval.typed_val_int.val = c;
-      yylval.typed_val_int.type = parse_type->builtin_char;
-
-      c = *lexptr++;
-      if (c != '\'')
-	{
-	  namelen = skip_quoted (tokstart) - tokstart;
-	  if (namelen > 2)
-	    {
-	      lexptr = tokstart + namelen;
-	      if (lexptr[-1] != '\'')
-		error ("Unmatched single quote.");
-	      namelen -= 2;
-	      tokstart++;
-	      goto tryname;
-	    }
-	  error ("Invalid character constant.");
-	}
-      return INT;
-
     case '(':
       paren_depth++;
       lexptr++;
@@ -1762,70 +2002,33 @@ yylex ()
       lexptr++;
       return c;
 
+    case 'L':
+    case 'u':
+    case 'U':
+      if (tokstart[1] != '"' && tokstart[1] != '\'')
+	break;
+      /* Fall through.  */
+    case '\'':
     case '"':
-
+      {
-      /* Build the gdb internal form of the input string in tempbuf,
+	int host_len;
-	 translating any standard C escape forms seen.  Note that the
+	int result = parse_string_or_char (tokstart, &lexptr, &yylval.tsval,
-	 buffer is null byte terminated *only* for the convenience of
+					   &host_len);
-	 debugging gdb itself and printing the buffer contents when
+	if (result == CHAR)
-	 the buffer contains no embedded nulls.  Gdb does not depend
-	 upon the buffer being null byte terminated, it uses the length
-	 string instead.  This allows gdb to handle C strings (as well
-	 as strings in other languages) with embedded null bytes */
-
-      tokptr = ++tokstart;
-      tempbufindex = 0;
-
-      do {
-        char *char_start_pos = tokptr;
-
-	/* Grow the static temp buffer if necessary, including allocating
-	   the first one on demand. */
-	if (tempbufindex + 1 >= tempbufsize)
 	  {
-	    tempbuf = (char *) realloc (tempbuf, tempbufsize += 64);
+	    if (host_len == 0)
-	  }
+	      error ("Empty character constant.");
-	switch (*tokptr)
+	    else if (host_len > 2 && c == '\'')
-	  {
-	  case '\0':
-	  case '"':
-	    /* Do nothing, loop will terminate. */
-	    break;
-	  case '\\':
-	    tokptr++;
-	    c = parse_escape (&tokptr);
-	    if (c == -1)
 	      {
-		continue;
+		++tokstart;
+		namelen = lexptr - tokstart - 1;
+		goto tryname;
 	      }
-	    tempbuf[tempbufindex++] = c;
+	    else if (host_len > 1)
-	    break;
+	      error ("Invalid character constant.");
-	  default:
-	    c = *tokptr++;
-            if (! host_char_to_target (c, &c))
-              {
-                int len = tokptr - char_start_pos;
-                char *copy = alloca (len + 1);
-                memcpy (copy, char_start_pos, len);
-                copy[len] = '\0';
-
-                error ("There is no character corresponding to `%s' "
-                       "in the target character set `%s'.",
-                       copy, target_charset ());
-              }
-            tempbuf[tempbufindex++] = c;
-	    break;
 	  }
-      } while ((*tokptr != '"') && (*tokptr != '\0'));
+	return result;
-      if (*tokptr++ != '"')
+      }
-	{
-	  error ("Unterminated string in expression.");
-	}
-      tempbuf[tempbufindex] = '\0';	/* See note above */
-      yylval.sval.ptr = tempbuf;
-      yylval.sval.length = tempbufindex;
-      lexptr = tokptr;
-      return (STRING);
     }
 
   if (!(c == '_' || c == '$'

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;
+  struct obstack wchar_buf, output;
-  int host_char;
+  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);
+  buf = alloca (TYPE_LENGTH (type));
-  if (escape)
+  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 (quoter == '"' && strcmp (escape, "0") == 0)
+      int num_chars;
-	/* Print nulls embedded in double quoted strings as \000 to
+      gdb_wchar_t *chars;
-	   prevent ambiguity.  */
+      const gdb_byte *buf;
-	fprintf_filtered (stream, "\\000");
+      size_t buflen;
-      else
+      int print_escape = 1;
-	fprintf_filtered (stream, "\\%s", escape);
+      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 if (target_char_to_host (c, &host_char)
+
-           && host_char_print_literally (host_char))
+  /* The output in the host encoding.  */
-    {
+  obstack_init (&output);
-      if (host_char == '\\' || host_char == quoter)
+  make_cleanup_obstack_free (&output);
-        fputs_filtered ("\\", stream);
+
-      fprintf_filtered (stream, "%c", host_char);
+  convert_between_encodings ("wchar_t", host_charset (),
-    }
+			     obstack_base (&wchar_buf),
-  else
+			     obstack_object_size (&wchar_buf),
-    fprintf_filtered (stream, "\\%.3o", (unsigned int) c);
+			     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,
+c_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
-	    unsigned int length, int width, int force_ellipses,
+	    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)
+      && (extract_unsigned_integer (string + (length - 1) * width, width) == 0))
-          == '\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)
     {
-      /* Position of the character we are examining
+      unsigned long current_char = 1;
-         to see whether it is repeated.  */
+      for (i = 0; current_char; ++i)
-      unsigned int rep1;
+	{
-      /* Number of repetitions we have detected so far.  */
+	  QUIT;
-      unsigned int reps;
+	  current_char = extract_unsigned_integer (string + i * width, width);
-      unsigned long current_char;
+	}
+      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)
+    {
+      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
-      rep1 = i + 1;
+	 single character in isolation.  This makes the code simpler
-      reps = 1;
+	 and probably does the sensible thing in the majority of
-      while (rep1 < length
+	 cases.  */
-	     && extract_unsigned_integer (string + rep1 * width, width)
+      while (num_chars == 1)
-	     == current_char)
 	{
-	  ++rep1;
+	  /* Count the number of repetitions.  */
-	  ++reps;
+	  unsigned int reps = 0;
-	}
+	  gdb_wchar_t current_char = chars[0];
+	  const gdb_byte *orig_buf = buf;
+	  int orig_len = buflen;
 
-      if (reps > options->repeat_count_threshold)
+	  if (need_comma)
-	{
-	  if (in_quotes)
 	    {
-	      if (options->inspect_it)
+	      obstack_grow_wstr (&wchar_buf, LCST (", "));
-		fputs_filtered ("\\\", ", stream);
+	      need_comma = 0;
-	      else
+	    }
-		fputs_filtered ("\", ", stream);
+
-	      in_quotes = 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;
+		}
 	    }
-	  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
+
+      /* NUM_CHARS and the other outputs from wchar_iterate are valid
+	 here regardless of which branch was taken above.  */
+      if (num_chars < 0)
 	{
+	  /* Hit EOF.  */
+	  finished = 1;
+	  break;
+	}
+
+      switch (result)
+	{
+	case wchar_iterate_invalid:
 	  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 ("\""));
 	      in_quotes = 1;
 	    }
-	  LA_EMIT_CHAR (current_char, stream, '"');
+	  need_escape = 0;
-	  ++things_printed;
+	  print_wchar (gdb_WEOF, buf, buflen, width, &wchar_buf,
+		       '"', &need_escape);
+	  break;
+
+	case wchar_iterate_incomplete:
+	  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 (" <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)
+  if (force_ellipses || !finished)
-    fputs_filtered ("...", stream);
+    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,

gdb/c-lang.h

@@ -29,9 +29,38 @@ struct language_arch_info;
 #include "macroexp.h"
 
 
-extern int c_parse (void);	/* Defined in c-exp.y */
+/* The various kinds of C string and character.  Note that these
+   values are chosen so that they may be or'd together in certain
+   ways.  */
+enum c_string_type
+  {
+    /* An ordinary string: "value".  */
+    C_STRING = 0,
+    /* A wide string: L"value".  */
+    C_WIDE_STRING = 1,
+    /* A 16-bit Unicode string: u"value".  */
+    C_STRING_16 = 2,
+    /* A 32-bit Unicode string: U"value".  */
+    C_STRING_32 = 3,
+    /* An ordinary char: 'v'.  This can also be or'd with one of the
+       above to form the corresponding CHAR value from a STRING
+       value.  */
+    C_CHAR = 4,
+    /* A wide char: L'v'.  */
+    C_WIDE_CHAR = 5,
+    /* A 16-bit Unicode char: u'v'.  */
+    C_CHAR_16 = 6,
+    /* A 32-bit Unicode char: U'v'.  */
+    C_CHAR_32 = 7
+  };
 
-extern void c_error (char *);	/* Defined in c-exp.y */
+/* Defined in c-exp.y.  */
+
+extern int c_parse (void);
+
+extern void c_error (char *);
+
+extern int c_parse_escape (char **, struct obstack *);
 
 /* Defined in c-typeprint.c */
 extern void c_print_type (struct type *, char *, struct ui_file *, int,
@@ -48,10 +77,10 @@ extern int c_value_print (struct value *, struct ui_file *,
 
 /* These are in c-lang.c: */
 
-extern void c_printchar (int, struct ui_file *);
+extern void c_printchar (int, struct type *, struct ui_file *);
 
-extern void c_printstr (struct ui_file * stream, const gdb_byte *string,
+extern void c_printstr (struct ui_file * stream, struct type *elttype,
-			unsigned int length, int width,
+			const gdb_byte *string, unsigned int length,
 			int force_ellipses,
 			const struct value_print_options *options);
 

gdb/c-valprint.c

@@ -55,6 +55,18 @@ print_function_pointer_address (CORE_ADDR address, struct ui_file *stream,
 }
 
 
+/* A helper for textual_element_type.  This checks the name of the
+   typedef.  This is bogus but it isn't apparent that the compiler
+   provides us the help we may need.  */
+
+static int
+textual_name (const char *name)
+{
+  return (!strcmp (name, "wchar_t")
+	  || !strcmp (name, "char16_t")
+	  || !strcmp (name, "char32_t"));
+}
+
 /* Apply a heuristic to decide whether an array of TYPE or a pointer
    to TYPE should be printed as a textual string.  Return non-zero if
    it should, or zero if it should be treated as an array of integers
@@ -77,6 +89,15 @@ textual_element_type (struct type *type, char format)
   /* TYPE_CODE_CHAR is always textual.  */
   if (TYPE_CODE (true_type) == TYPE_CODE_CHAR)
     return 1;
+  /* Any other character-like types must be integral.  */
+  if (TYPE_CODE (true_type) != TYPE_CODE_INT)
+    return 0;
+
+  /* Check the names of the type and the typedef.  */
+  if (TYPE_NAME (type) && textual_name (TYPE_NAME (type)))
+    return 1;
+  if (TYPE_NAME (true_type) && textual_name (TYPE_NAME (true_type)))
+    return 1;
 
   if (format == 's')
     {
@@ -115,7 +136,8 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 {
   unsigned int i = 0;	/* Number of characters printed */
   unsigned len;
-  struct type *elttype;
+  struct type *elttype, *unresolved_elttype;
+  struct type *unresolved_type = type;
   unsigned eltlen;
   LONGEST val;
   CORE_ADDR addr;
@@ -124,8 +146,9 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
   switch (TYPE_CODE (type))
     {
     case TYPE_CODE_ARRAY:
-      elttype = check_typedef (TYPE_TARGET_TYPE (type));
+      unresolved_elttype = TYPE_TARGET_TYPE (type);
-      if (TYPE_LENGTH (type) > 0 && TYPE_LENGTH (TYPE_TARGET_TYPE (type)) > 0)
+      elttype = check_typedef (unresolved_elttype);
+      if (TYPE_LENGTH (type) > 0 && TYPE_LENGTH (unresolved_elttype) > 0)
 	{
 	  eltlen = TYPE_LENGTH (elttype);
 	  len = TYPE_LENGTH (type) / eltlen;
@@ -135,7 +158,7 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	    }
 
 	  /* Print arrays of textual chars with a string syntax.  */
-          if (textual_element_type (elttype, options->format))
+          if (textual_element_type (unresolved_elttype, options->format))
 	    {
 	      /* If requested, look for the first null char and only print
 	         elements up to it.  */
@@ -143,15 +166,19 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 		{
 		  unsigned int temp_len;
 
-		  /* Look for a NULL char. */
 		  for (temp_len = 0;
-		       (valaddr + embedded_offset)[temp_len]
+		       (temp_len < len
-		       && temp_len < len && temp_len < options->print_max;
+			&& temp_len < options->print_max
-		       temp_len++);
+			&& extract_unsigned_integer (valaddr + embedded_offset
+						     + temp_len * eltlen,
+						     eltlen) == 0);
+		       ++temp_len)
+		    ;
 		  len = temp_len;
 		}
 
-	      LA_PRINT_STRING (stream, valaddr + embedded_offset, len, eltlen, 0, options);
+	      LA_PRINT_STRING (stream, unresolved_elttype,
+			       valaddr + embedded_offset, len, 0, options);
 	      i = len;
 	    }
 	  else
@@ -209,7 +236,8 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	  print_function_pointer_address (addr, stream, options->addressprint);
 	  break;
 	}
-      elttype = check_typedef (TYPE_TARGET_TYPE (type));
+      unresolved_elttype = TYPE_TARGET_TYPE (type);
+      elttype = check_typedef (unresolved_elttype);
 	{
 	  addr = unpack_pointer (type, valaddr + embedded_offset);
 	print_unpacked_pointer:
@@ -228,12 +256,11 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 
 	  /* For a pointer to a textual type, also print the string
 	     pointed to, unless pointer is null.  */
-	  /* FIXME: need to handle wchar_t here... */
 
-	  if (textual_element_type (elttype, options->format)
+	  if (textual_element_type (unresolved_elttype, options->format)
 	      && addr != 0)
 	    {
-	      i = val_print_string (addr, -1, TYPE_LENGTH (elttype), stream,
+	      i = val_print_string (unresolved_elttype, addr, -1, stream,
 				    options);
 	    }
 	  else if (cp_is_vtbl_member (type))
@@ -268,7 +295,7 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 		    }
 		  else
 		    {
-		      wtype = TYPE_TARGET_TYPE (type);
+		      wtype = unresolved_elttype;
 		    }
 		  vt_val = value_at (wtype, vt_address);
 		  common_val_print (vt_val, stream, recurse + 1, options,
@@ -442,11 +469,11 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	     Since we don't know whether the value is really intended to
 	     be used as an integer or a character, print the character
 	     equivalent as well.  */
-	  if (textual_element_type (type, options->format))
+	  if (textual_element_type (unresolved_type, options->format))
 	    {
 	      fputs_filtered (" ", stream);
 	      LA_PRINT_CHAR ((unsigned char) unpack_long (type, valaddr + embedded_offset),
-			     stream);
+			     unresolved_type, stream);
 	    }
 	}
       break;
@@ -468,7 +495,7 @@ c_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	  else
 	    fprintf_filtered (stream, "%d", (int) val);
 	  fputs_filtered (" ", stream);
-	  LA_PRINT_CHAR ((unsigned char) val, stream);
+	  LA_PRINT_CHAR ((unsigned char) val, unresolved_type, stream);
 	}
       break;
 
@@ -540,7 +567,7 @@ int
 c_value_print (struct value *val, struct ui_file *stream, 
 	       const struct value_print_options *options)
 {
-  struct type *type, *real_type;
+  struct type *type, *real_type, *val_type;
   int full, top, using_enc;
   struct value_print_options opts = *options;
 
@@ -553,7 +580,11 @@ c_value_print (struct value *val, struct ui_file *stream,
      C++: if it is a member pointer, we will take care
      of that when we print it.  */
 
-  type = check_typedef (value_type (val));
+  /* Preserve the original type before stripping typedefs.  We prefer
+     to pass down the original type when possible, but for local
+     checks it is better to look past the typedefs.  */
+  val_type = value_type (val);
+  type = check_typedef (val_type);
 
   if (TYPE_CODE (type) == TYPE_CODE_PTR
       || TYPE_CODE (type) == TYPE_CODE_REF)
@@ -561,11 +592,12 @@ c_value_print (struct value *val, struct ui_file *stream,
       /* Hack:  remove (char *) for char strings.  Their
          type is indicated by the quoted string anyway.
          (Don't use textual_element_type here; quoted strings
-         are always exactly (char *).  */
+         are always exactly (char *), (wchar_t *), or the like.  */
-      if (TYPE_CODE (type) == TYPE_CODE_PTR
+      if (TYPE_CODE (val_type) == TYPE_CODE_PTR
-	  && TYPE_NAME (type) == NULL
+	  && TYPE_NAME (val_type) == NULL
-	  && TYPE_NAME (TYPE_TARGET_TYPE (type)) != NULL
+	  && TYPE_NAME (TYPE_TARGET_TYPE (val_type)) != NULL
-	  && strcmp (TYPE_NAME (TYPE_TARGET_TYPE (type)), "char") == 0)
+	  && (strcmp (TYPE_NAME (TYPE_TARGET_TYPE (val_type)), "char") == 0
+	      || textual_name (TYPE_NAME (TYPE_TARGET_TYPE (val_type)))))
 	{
 	  /* Print nothing */
 	}
@@ -608,6 +640,7 @@ c_value_print (struct value *val, struct ui_file *stream,
             }
           type_print (type, "", stream, -1);
 	  fprintf_filtered (stream, ") ");
+	  val_type = type;
 	}
       else
 	{
@@ -653,7 +686,7 @@ c_value_print (struct value *val, struct ui_file *stream,
       /* Otherwise, we end up at the return outside this "if" */
     }
 
-  return val_print (type, value_contents_all (val),
+  return val_print (val_type, value_contents_all (val),
 		    value_embedded_offset (val),
 		    VALUE_ADDRESS (val) + value_offset (val),
 		    stream, 0, &opts, current_language);

gdb/charset.h

@@ -19,89 +19,129 @@
 #ifndef CHARSET_H
 #define CHARSET_H
 
-
 /* If the target program uses a different character set than the host,
    GDB has some support for translating between the two; GDB converts
    characters and strings to the host character set before displaying
    them, and converts characters and strings appearing in expressions
    entered by the user to the target character set.
 
-   At the moment, GDB only supports single-byte, stateless character
+   GDB's code pretty much assumes that the host character set is some
-   sets.  This includes the ISO-8859 family (ASCII extended with
+   superset of ASCII; there are plenty if ('0' + n) expressions and
-   accented characters, and (I think) Cyrillic, for European
+   the like.  */
-   languages), and the EBCDIC family (used on IBM's mainframes).
-   Unfortunately, it excludes many Asian scripts, the fixed- and
-   variable-width Unicode encodings, and other desireable things.
-   Patches are welcome!  (For example, it would be nice if the Java
-   string support could simply get absorbed into some more general
-   multi-byte encoding support.)
-
-   Furthermore, GDB's code pretty much assumes that the host character
-   set is some superset of ASCII; there are plenty if ('0' + n)
-   expressions and the like.
-
-   When the `iconv' library routine supports a character set meeting
-   the requirements above, it's easy to plug an entry into GDB's table
-   that uses iconv to handle the details.  */
 
 /* Return the name of the current host/target character set.  The
    result is owned by the charset module; the caller should not free
    it.  */
 const char *host_charset (void);
 const char *target_charset (void);
+const char *target_wide_charset (void);
 
-/* In general, the set of C backslash escapes (\n, \f) is specific to
+/* These values are used to specify the type of transliteration done
-   the character set.  Not all character sets will have form feed
+   by convert_between_encodings.  */
-   characters, for example.
+enum transliterations
+  {
+    /* Error on failure to convert.  */
+    translit_none,
+    /* Transliterate to host char.  */
+    translit_char
+  };
 
-   The following functions allow GDB to parse and print control
+/* Convert between two encodings.
-   characters in a character-set-independent way.  They are both
+
-   language-specific (to C and C++) and character-set-specific.
+   FROM is the name of the source encoding.
-   Putting them here is a compromise.  */
+   TO is the name of the target encoding.
+   BYTES holds the bytes to convert; this is assumed to be characters
+   in the target encoding.
+   NUM_BYTES is the number of bytes.
+   WIDTH is the width of a character from the FROM charset, in bytes.
+   For a variable width encoding, WIDTH should be the size of a "base
+   character".
+   OUTPUT is an obstack where the converted data is written.  The
+   caller is responsible for initializing the obstack, and for
+   destroying the obstack should an error occur.
+   TRANSLIT specifies how invalid conversions should be handled.  */
+void convert_between_encodings (const char *from, const char *to,
+				const gdb_byte *bytes, unsigned int num_bytes,
+				int width, struct obstack *output,
+				enum transliterations translit);
 
 
-/* If the target character TARGET_CHAR have a backslash escape in the
+/* These values are used by wchar_iterate to report errors.  */
-   C language (i.e., a character like 'n' or 't'), return the host
+enum wchar_iterate_result
-   character string that should follow the backslash.  Otherwise,
+  {
-   return zero.
+    /* Ordinary return.  */
+    wchar_iterate_ok,
+    /* Invalid input sequence.  */
+    wchar_iterate_invalid,
+    /* Incomplete input sequence at the end of the input.  */
+    wchar_iterate_incomplete,
+    /* EOF.  */
+    wchar_iterate_eof
+  };
 
-   When this function returns non-zero, the string it returns is
+/* Declaration of the opaque wchar iterator type.  */
-   statically allocated; the caller is not responsible for freeing it.  */
+struct wchar_iterator;
-const char *c_target_char_has_backslash_escape (int target_char);
 
+/* Create a new character iterator which returns wchar_t's.  INPUT is
+   the input buffer.  BYTES is the number of bytes in the input
+   buffer.  CHARSET is the name of the character set in which INPUT is
+   encoded.  WIDTH is the number of bytes in a base character of
+   CHARSET.
+   
+   This function either returns a new character set iterator, or calls
+   error.  The result can be freed using a cleanup; see
+   make_cleanup_wchar_iterator.  */
+struct wchar_iterator *make_wchar_iterator (const gdb_byte *input, size_t bytes,
+					    const char *charset,
+					    size_t width);
 
-/* If the host character HOST_CHAR is a valid backslash escape in the
+/* Return a new cleanup suitable for destroying the wchar iterator
-   C language for the target character set, return non-zero, and set
+   ITER.  */
-   *TARGET_CHAR to the target character the backslash escape represents.
+struct cleanup *make_cleanup_wchar_iterator (struct wchar_iterator *iter);
-   Otherwise, return zero.  */
-int c_parse_backslash (int host_char, int *target_char);
 
+/* Perform a single iteration of a wchar_t iterator.
+   
+   Returns the number of characters converted.  A negative result
+   means that EOF has been reached.  A positive result indicates the
+   number of valid wchar_ts in the result; *OUT_CHARS is updated to
+   point to the first valid character.
 
-/* Return non-zero if the host character HOST_CHAR can be printed
+   In all cases aside from EOF, *PTR is set to point to the first
-   literally --- that is, if it can be readably printed as itself in a
+   converted target byte.  *LEN is set to the number of bytes
-   character or string constant.  Return zero if it should be printed
+   converted.
-   using some kind of numeric escape, like '\031' in C, '^(25)' in
-   Chill, or #25 in Pascal.  */
-int host_char_print_literally (int host_char);
 
+   A zero result means one of several unusual results.  *OUT_RESULT is
+   set to indicate the type of un-ordinary return.
 
-/* If the host character HOST_CHAR has an equivalent in the target
+   wchar_iterate_invalid means that an invalid input character was
-   character set, set *TARGET_CHAR to that equivalent, and return
+   seen.  The iterator is advanced by WIDTH (the argument to
-   non-zero.  Otherwise, return zero.  */
+   make_wchar_iterator) bytes.
-int host_char_to_target (int host_char, int *target_char);
 
+   wchar_iterate_incomplete means that an incomplete character was
+   seen at the end of the input sequence.
+   
+   wchar_iterate_eof means that all bytes were successfully
+   converted.  The other output arguments are not set.  */
+int wchar_iterate (struct wchar_iterator *iter,
+		   enum wchar_iterate_result *out_result,
+		   gdb_wchar_t **out_chars,
+		   const gdb_byte **ptr, size_t *len);
 
-/* If the target character TARGET_CHAR has an equivalent in the host
+
-   character set, set *HOST_CHAR to that equivalent, and return
-   non-zero.  Otherwise, return zero.  */
-int target_char_to_host (int target_char, int *host_char);
 
+/* GDB needs to know a few details of its execution character set.
+   This knowledge is isolated here and in charset.c.  */
 
-/* If the target character TARGET_CHAR has a corresponding control
+/* The escape character.  */
-   character (also in the target character set), set *TARGET_CTRL_CHAR
+#define HOST_ESCAPE_CHAR 27
-   to the control character, and return non-zero.  Otherwise, return
-   zero.  */
-int target_char_to_control_char (int target_char, int *target_ctrl_char);
 
+/* Convert a letter, like 'c', to its corresponding control
+   character.  */
+char host_letter_to_control_character (char c);
+
+/* Convert a hex digit character to its numeric value.  E.g., 'f' is
+   converted to 15.  This function assumes that C is a valid hex
+   digit.  Both upper- and lower-case letters are recognized.  */
+int host_hex_value (char c);
 
 #endif /* CHARSET_H */

gdb/config.in

@@ -169,12 +169,18 @@
 /* Define if you have the iconv() function. */
 #undef HAVE_ICONV
 
+/* Define to 1 if you have the `iconvlist' function. */
+#undef HAVE_ICONVLIST
+
 /* Define if your compiler supports the #include_next directive. */
 #undef HAVE_INCLUDE_NEXT
 
 /* Define to 1 if you have the <inttypes.h> header file. */
 #undef HAVE_INTTYPES_H
 
+/* Define if you have <langinfo.h> and nl_langinfo(CODESET). */
+#undef HAVE_LANGINFO_CODESET
+
 /* Define if your <locale.h> file defines LC_MESSAGES. */
 #undef HAVE_LC_MESSAGES
 

gdb/configure

@@ -314,7 +314,7 @@ ac_subdirs_all="$ac_subdirs_all doc testsuite"
 ac_subdirs_all="$ac_subdirs_all gdbtk"
 ac_subdirs_all="$ac_subdirs_all multi-ice"
 ac_subdirs_all="$ac_subdirs_all gdbserver"
-ac_subst_vars='SHELL PATH_SEPARATOR PACKAGE_NAME PACKAGE_TARNAME PACKAGE_VERSION PACKAGE_STRING PACKAGE_BUGREPORT exec_prefix prefix program_transform_name bindir sbindir libexecdir datadir sysconfdir sharedstatedir localstatedir libdir includedir oldincludedir infodir mandir build_alias host_alias target_alias DEFS ECHO_C ECHO_N ECHO_T LIBS MAINTAINER_MODE_TRUE MAINTAINER_MODE_FALSE MAINT CC CFLAGS LDFLAGS CPPFLAGS ac_ct_CC EXEEXT OBJEXT CPP EGREP RANLIB ac_ct_RANLIB build build_cpu build_vendor build_os host host_cpu host_vendor host_os target target_cpu target_vendor target_os am__leading_dot DEPDIR CCDEPMODE MAKE GMAKE_TRUE GMAKE_FALSE SET_MAKE USE_NLS LIBINTL LIBINTL_DEP INCINTL XGETTEXT GMSGFMT POSUB CATALOGS DATADIRNAME INSTOBJEXT GENCAT CATOBJEXT localedir GL_COND_LIBTOOL_TRUE GL_COND_LIBTOOL_FALSE GNULIB_MEMMEM GNULIB_MEMPCPY GNULIB_MEMRCHR GNULIB_STPCPY GNULIB_STPNCPY GNULIB_STRCHRNUL GNULIB_STRDUP GNULIB_STRNDUP GNULIB_STRNLEN GNULIB_STRPBRK GNULIB_STRSEP GNULIB_STRSTR GNULIB_STRCASESTR GNULIB_STRTOK_R GNULIB_MBSLEN GNULIB_MBSNLEN GNULIB_MBSCHR GNULIB_MBSRCHR GNULIB_MBSSTR GNULIB_MBSCASECMP GNULIB_MBSNCASECMP GNULIB_MBSPCASECMP GNULIB_MBSCASESTR GNULIB_MBSCSPN GNULIB_MBSPBRK GNULIB_MBSSPN GNULIB_MBSSEP GNULIB_MBSTOK_R GNULIB_STRERROR GNULIB_STRSIGNAL HAVE_DECL_MEMMEM HAVE_MEMPCPY HAVE_DECL_MEMRCHR HAVE_STPCPY HAVE_STPNCPY HAVE_STRCHRNUL HAVE_DECL_STRDUP HAVE_STRNDUP HAVE_DECL_STRNDUP HAVE_DECL_STRNLEN HAVE_STRPBRK HAVE_STRSEP HAVE_STRCASESTR HAVE_DECL_STRTOK_R HAVE_DECL_STRERROR HAVE_DECL_STRSIGNAL REPLACE_STRERROR REPLACE_STRSIGNAL REPLACE_MEMMEM REPLACE_STRCASESTR REPLACE_STRSTR HAVE_LONG_LONG_INT HAVE_UNSIGNED_LONG_LONG_INT HAVE_INTTYPES_H HAVE_SYS_TYPES_H INCLUDE_NEXT NEXT_STDINT_H HAVE_STDINT_H HAVE_SYS_INTTYPES_H HAVE_SYS_BITYPES_H BITSIZEOF_PTRDIFF_T BITSIZEOF_SIG_ATOMIC_T BITSIZEOF_SIZE_T BITSIZEOF_WCHAR_T BITSIZEOF_WINT_T HAVE_SIGNED_SIG_ATOMIC_T HAVE_SIGNED_WCHAR_T HAVE_SIGNED_WINT_T PTRDIFF_T_SUFFIX SIG_ATOMIC_T_SUFFIX SIZE_T_SUFFIX WCHAR_T_SUFFIX WINT_T_SUFFIX STDINT_H NEXT_STRING_H GNULIB_WCWIDTH HAVE_DECL_WCWIDTH REPLACE_WCWIDTH WCHAR_H HAVE_WCHAR_H NEXT_WCHAR_H LIBGNU_LIBDEPS LIBGNU_LTLIBDEPS GNULIB_STDINT_H PACKAGE INSTALL_PROGRAM INSTALL_SCRIPT INSTALL_DATA CYGPATH_W VERSION ACLOCAL AUTOCONF AUTOMAKE AUTOHEADER MAKEINFO install_sh STRIP ac_ct_STRIP INSTALL_STRIP_PROGRAM mkdir_p AWK AMTAR am__tar am__untar am__include am__quote AMDEP_TRUE AMDEP_FALSE AMDEPBACKSLASH am__fastdepCC_TRUE am__fastdepCC_FALSE subdirs TARGET_OBS PKGVERSION REPORT_BUGS_TO REPORT_BUGS_TEXI LN_S YACC AR ac_ct_AR DLLTOOL ac_ct_DLLTOOL WINDRES ac_ct_WINDRES MIG ac_ct_MIG READLINE READLINE_DEPS READLINE_CFLAGS HAVE_LIBEXPAT LIBEXPAT LTLIBEXPAT PYTHON_CFLAGS ALLOCA CONFIG_LDFLAGS TARGET_SYSTEM_ROOT TARGET_SYSTEM_ROOT_DEFINE WARN_CFLAGS WERROR_CFLAGS SER_HARDWIRE WIN32LIBS LIBGUI GUI_CFLAGS_X WIN32LDAPP TCL_VERSION TCL_PATCH_LEVEL TCL_BIN_DIR TCL_SRC_DIR TCL_LIB_FILE TCL_LIB_FLAG TCL_LIB_SPEC TCL_STUB_LIB_FILE TCL_STUB_LIB_FLAG TCL_STUB_LIB_SPEC TCL_INCLUDE TCL_LIBRARY TCL_DEPS TK_VERSION TK_BIN_DIR TK_SRC_DIR TK_LIB_FILE TK_LIB_FLAG TK_LIB_SPEC TK_STUB_LIB_FILE TK_STUB_LIB_FLAG TK_STUB_LIB_SPEC TK_INCLUDE TK_LIBRARY TK_DEPS TK_XINCLUDES X_CFLAGS X_LDFLAGS X_LIBS GDBTKLIBS GDBTK_CFLAGS GDBTK_SRC_DIR SIM SIM_OBS ENABLE_CFLAGS PROFILE_CFLAGS CONFIG_OBS CONFIG_DEPS CONFIG_SRCS CONFIG_ALL CONFIG_CLEAN CONFIG_INSTALL CONFIG_UNINSTALL target_subdir frags nm_h LIBICONV LIBOBJS LTLIBOBJS gl_LIBOBJS gl_LTLIBOBJS gltests_LIBOBJS gltests_LTLIBOBJS'
+ac_subst_vars='SHELL PATH_SEPARATOR PACKAGE_NAME PACKAGE_TARNAME PACKAGE_VERSION PACKAGE_STRING PACKAGE_BUGREPORT exec_prefix prefix program_transform_name bindir sbindir libexecdir datadir sysconfdir sharedstatedir localstatedir libdir includedir oldincludedir infodir mandir build_alias host_alias target_alias DEFS ECHO_C ECHO_N ECHO_T LIBS MAINTAINER_MODE_TRUE MAINTAINER_MODE_FALSE MAINT CC CFLAGS LDFLAGS CPPFLAGS ac_ct_CC EXEEXT OBJEXT CPP EGREP RANLIB ac_ct_RANLIB build build_cpu build_vendor build_os host host_cpu host_vendor host_os target target_cpu target_vendor target_os am__leading_dot DEPDIR CCDEPMODE MAKE GMAKE_TRUE GMAKE_FALSE SET_MAKE USE_NLS LIBINTL LIBINTL_DEP INCINTL XGETTEXT GMSGFMT POSUB CATALOGS DATADIRNAME INSTOBJEXT GENCAT CATOBJEXT localedir GL_COND_LIBTOOL_TRUE GL_COND_LIBTOOL_FALSE GNULIB_MEMMEM GNULIB_MEMPCPY GNULIB_MEMRCHR GNULIB_STPCPY GNULIB_STPNCPY GNULIB_STRCHRNUL GNULIB_STRDUP GNULIB_STRNDUP GNULIB_STRNLEN GNULIB_STRPBRK GNULIB_STRSEP GNULIB_STRSTR GNULIB_STRCASESTR GNULIB_STRTOK_R GNULIB_MBSLEN GNULIB_MBSNLEN GNULIB_MBSCHR GNULIB_MBSRCHR GNULIB_MBSSTR GNULIB_MBSCASECMP GNULIB_MBSNCASECMP GNULIB_MBSPCASECMP GNULIB_MBSCASESTR GNULIB_MBSCSPN GNULIB_MBSPBRK GNULIB_MBSSPN GNULIB_MBSSEP GNULIB_MBSTOK_R GNULIB_STRERROR GNULIB_STRSIGNAL HAVE_DECL_MEMMEM HAVE_MEMPCPY HAVE_DECL_MEMRCHR HAVE_STPCPY HAVE_STPNCPY HAVE_STRCHRNUL HAVE_DECL_STRDUP HAVE_STRNDUP HAVE_DECL_STRNDUP HAVE_DECL_STRNLEN HAVE_STRPBRK HAVE_STRSEP HAVE_STRCASESTR HAVE_DECL_STRTOK_R HAVE_DECL_STRERROR HAVE_DECL_STRSIGNAL REPLACE_STRERROR REPLACE_STRSIGNAL REPLACE_MEMMEM REPLACE_STRCASESTR REPLACE_STRSTR HAVE_LONG_LONG_INT HAVE_UNSIGNED_LONG_LONG_INT HAVE_INTTYPES_H HAVE_SYS_TYPES_H INCLUDE_NEXT NEXT_STDINT_H HAVE_STDINT_H HAVE_SYS_INTTYPES_H HAVE_SYS_BITYPES_H BITSIZEOF_PTRDIFF_T BITSIZEOF_SIG_ATOMIC_T BITSIZEOF_SIZE_T BITSIZEOF_WCHAR_T BITSIZEOF_WINT_T HAVE_SIGNED_SIG_ATOMIC_T HAVE_SIGNED_WCHAR_T HAVE_SIGNED_WINT_T PTRDIFF_T_SUFFIX SIG_ATOMIC_T_SUFFIX SIZE_T_SUFFIX WCHAR_T_SUFFIX WINT_T_SUFFIX STDINT_H NEXT_STRING_H GNULIB_WCWIDTH HAVE_DECL_WCWIDTH REPLACE_WCWIDTH WCHAR_H HAVE_WCHAR_H NEXT_WCHAR_H LIBGNU_LIBDEPS LIBGNU_LTLIBDEPS GNULIB_STDINT_H PACKAGE INSTALL_PROGRAM INSTALL_SCRIPT INSTALL_DATA CYGPATH_W VERSION ACLOCAL AUTOCONF AUTOMAKE AUTOHEADER MAKEINFO install_sh STRIP ac_ct_STRIP INSTALL_STRIP_PROGRAM mkdir_p AWK AMTAR am__tar am__untar am__include am__quote AMDEP_TRUE AMDEP_FALSE AMDEPBACKSLASH am__fastdepCC_TRUE am__fastdepCC_FALSE subdirs TARGET_OBS PKGVERSION REPORT_BUGS_TO REPORT_BUGS_TEXI LN_S YACC AR ac_ct_AR DLLTOOL ac_ct_DLLTOOL WINDRES ac_ct_WINDRES MIG ac_ct_MIG LIBICONV LIBICONV_INCLUDE LIBICONV_LIBDIR READLINE READLINE_DEPS READLINE_CFLAGS HAVE_LIBEXPAT LIBEXPAT LTLIBEXPAT PYTHON_CFLAGS ALLOCA CONFIG_LDFLAGS TARGET_SYSTEM_ROOT TARGET_SYSTEM_ROOT_DEFINE WARN_CFLAGS WERROR_CFLAGS SER_HARDWIRE WIN32LIBS LIBGUI GUI_CFLAGS_X WIN32LDAPP TCL_VERSION TCL_PATCH_LEVEL TCL_BIN_DIR TCL_SRC_DIR TCL_LIB_FILE TCL_LIB_FLAG TCL_LIB_SPEC TCL_STUB_LIB_FILE TCL_STUB_LIB_FLAG TCL_STUB_LIB_SPEC TCL_INCLUDE TCL_LIBRARY TCL_DEPS TK_VERSION TK_BIN_DIR TK_SRC_DIR TK_LIB_FILE TK_LIB_FLAG TK_LIB_SPEC TK_STUB_LIB_FILE TK_STUB_LIB_FLAG TK_STUB_LIB_SPEC TK_INCLUDE TK_LIBRARY TK_DEPS TK_XINCLUDES X_CFLAGS X_LDFLAGS X_LIBS GDBTKLIBS GDBTK_CFLAGS GDBTK_SRC_DIR SIM SIM_OBS ENABLE_CFLAGS PROFILE_CFLAGS CONFIG_OBS CONFIG_DEPS CONFIG_SRCS CONFIG_ALL CONFIG_CLEAN CONFIG_INSTALL CONFIG_UNINSTALL target_subdir frags nm_h LIBOBJS LTLIBOBJS gl_LIBOBJS gl_LTLIBOBJS gltests_LIBOBJS gltests_LTLIBOBJS'
 ac_subst_files='host_makefile_frag'
 ac_pwd=`pwd`
 
@@ -885,6 +885,7 @@ Optional Packages:
   --with-libunwind            Use libunwind frame unwinding support
   --with-pkgversion=PKG   Use PKG in the version string in place of "GDB"
   --with-bugurl=URL       Direct users to URL to report a bug
+  --with-libiconv-prefix=DIR  search for libiconv in DIR/include and DIR/lib
   --with-system-readline  use installed readline library
   --with-expat            include expat support (auto/yes/no)
   --with-gnu-ld           assume the C compiler uses GNU ld default=no
@@ -899,7 +900,6 @@ Optional Packages:
   --with-tcl              directory containing tcl configuration (tclConfig.sh)
   --with-tk               directory containing tk configuration (tkConfig.sh)
   --with-x                use the X Window System
-  --with-libiconv-prefix=DIR  search for libiconv in DIR/include and DIR/lib
 
 Some influential environment variables:
   CC          C compiler command
@@ -9989,6 +9989,296 @@ if test "$ac_cv_search_dlgetmodinfo" != no; then
 fi
 
 
+
+
+
+# Check whether --with-libiconv-prefix or --without-libiconv-prefix was given.
+if test "${with_libiconv_prefix+set}" = set; then
+  withval="$with_libiconv_prefix"
+
+    for dir in `echo "$withval" | tr : ' '`; do
+      if test -d $dir/include; then LIBICONV_INCLUDE="-I$dir/include"; CPPFLAGS="$CPPFLAGS -I$dir/include"; fi
+      if test -d $dir/lib; then LIBICONV_LIBDIR="-L$dir/lib"; LDFLAGS="$LDFLAGS -L$dir/lib"; fi
+    done
+
+fi;
+
+  BUILD_LIBICONV_LIBDIR="-L../libiconv/lib/.libs -L../libiconv/lib/_libs"
+  BUILD_LIBICONV_INCLUDE="-I../libiconv/include"
+
+  echo "$as_me:$LINENO: checking for iconv" >&5
+echo $ECHO_N "checking for iconv... $ECHO_C" >&6
+if test "${am_cv_func_iconv+set}" = set; then
+  echo $ECHO_N "(cached) $ECHO_C" >&6
+else
+
+    am_cv_func_iconv="no, consider installing GNU libiconv"
+    am_cv_lib_iconv=no
+    am_cv_use_build_libiconv=no
+    cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+#include <stdlib.h>
+#include <iconv.h>
+int
+main ()
+{
+iconv_t cd = iconv_open("","");
+       iconv(cd,NULL,NULL,NULL,NULL);
+       iconv_close(cd);
+  ;
+  return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
+  (eval $ac_link) 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } &&
+	 { ac_try='test -z "$ac_c_werror_flag"
+			 || test ! -s conftest.err'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; } &&
+	 { ac_try='test -s conftest$ac_exeext'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; }; then
+  am_cv_func_iconv=yes
+else
+  echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+fi
+rm -f conftest.err conftest.$ac_objext \
+      conftest$ac_exeext conftest.$ac_ext
+    if test "$am_cv_func_iconv" != yes; then
+      am_save_LIBS="$LIBS"
+      LIBS="$LIBS -liconv"
+      cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+#include <stdlib.h>
+#include <iconv.h>
+int
+main ()
+{
+iconv_t cd = iconv_open("","");
+         iconv(cd,NULL,NULL,NULL,NULL);
+         iconv_close(cd);
+  ;
+  return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
+  (eval $ac_link) 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } &&
+	 { ac_try='test -z "$ac_c_werror_flag"
+			 || test ! -s conftest.err'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; } &&
+	 { ac_try='test -s conftest$ac_exeext'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; }; then
+  am_cv_lib_iconv=yes
+        am_cv_func_iconv=yes
+else
+  echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+fi
+rm -f conftest.err conftest.$ac_objext \
+      conftest$ac_exeext conftest.$ac_ext
+      LIBS="$am_save_LIBS"
+    fi
+    # Look for libiconv in the build tree.
+    if test "$am_cv_func_iconv" != yes && test -d ../libiconv; then
+      am_save_LIBS="$LIBS"
+      am_save_CPPFLAGS="$CPPFLAGS"
+      LIBS="$LIBS $BUILD_LIBICONV_LIBDIR -liconv"
+      CPPFLAGS="$CPPFLAGS $BUILD_LIBICONV_INCLUDE"
+      cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+#include <stdlib.h>
+#include <iconv.h>
+int
+main ()
+{
+iconv_t cd = iconv_open("","");
+         iconv(cd,NULL,NULL,NULL,NULL);
+         iconv_close(cd);
+  ;
+  return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
+  (eval $ac_link) 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } &&
+	 { ac_try='test -z "$ac_c_werror_flag"
+			 || test ! -s conftest.err'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; } &&
+	 { ac_try='test -s conftest$ac_exeext'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; }; then
+  am_cv_lib_iconv=yes
+        am_cv_func_iconv=yes
+else
+  echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+fi
+rm -f conftest.err conftest.$ac_objext \
+      conftest$ac_exeext conftest.$ac_ext
+      LIBS="$am_save_LIBS"
+      if test "$am_cv_func_iconv" = "yes"; then
+         am_cv_use_build_libiconv=yes
+      else
+      	 CPPFLAGS="$am_save_CPPFLAGS"
+      fi
+    fi
+
+fi
+echo "$as_me:$LINENO: result: $am_cv_func_iconv" >&5
+echo "${ECHO_T}$am_cv_func_iconv" >&6
+  if test "$am_cv_func_iconv" = yes; then
+
+cat >>confdefs.h <<\_ACEOF
+#define HAVE_ICONV 1
+_ACEOF
+
+    echo "$as_me:$LINENO: checking for iconv declaration" >&5
+echo $ECHO_N "checking for iconv declaration... $ECHO_C" >&6
+    if test "${am_cv_proto_iconv+set}" = set; then
+  echo $ECHO_N "(cached) $ECHO_C" >&6
+else
+
+      cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+
+#include <stdlib.h>
+#include <iconv.h>
+extern
+#ifdef __cplusplus
+"C"
+#endif
+#if defined(__STDC__) || defined(__cplusplus)
+size_t iconv (iconv_t cd, char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);
+#else
+size_t iconv();
+#endif
+
+int
+main ()
+{
+
+  ;
+  return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext
+if { (eval echo "$as_me:$LINENO: \"$ac_compile\"") >&5
+  (eval $ac_compile) 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } &&
+	 { ac_try='test -z "$ac_c_werror_flag"
+			 || test ! -s conftest.err'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; } &&
+	 { ac_try='test -s conftest.$ac_objext'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; }; then
+  am_cv_proto_iconv_arg1=""
+else
+  echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+am_cv_proto_iconv_arg1="const"
+fi
+rm -f conftest.err conftest.$ac_objext conftest.$ac_ext
+      am_cv_proto_iconv="extern size_t iconv (iconv_t cd, $am_cv_proto_iconv_arg1 char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);"
+fi
+
+    am_cv_proto_iconv=`echo "$am_cv_proto_iconv" | tr -s ' ' | sed -e 's/( /(/'`
+    echo "$as_me:$LINENO: result: ${ac_t:-
+         }$am_cv_proto_iconv" >&5
+echo "${ECHO_T}${ac_t:-
+         }$am_cv_proto_iconv" >&6
+
+cat >>confdefs.h <<_ACEOF
+#define ICONV_CONST $am_cv_proto_iconv_arg1
+_ACEOF
+
+  fi
+  LIBICONV=
+  if test "$am_cv_lib_iconv" = yes; then
+    LIBICONV="-liconv"
+  fi
+  if test "$am_cv_use_build_libiconv" = yes; then
+    LIBICONV_LIBDIR="$BUILD_LIBICONV_LIBDIR"
+    LIBICONV_INCLUDE="$BUILD_LIBICONV_INCLUDE"
+  fi
+
+
+
+
+
 # On alpha-osf, it appears that libtermcap and libcurses are not compatible.
 # There is a very specific comment in /usr/include/curses.h explaining that
 # termcap routines built into libcurses must not be used.
@@ -15444,11 +15734,12 @@ fi
 
 
 
+
 
 for ac_func in canonicalize_file_name realpath getrusage getuid \
                 getgid poll pread64 sbrk setpgid setpgrp setsid \
 		sigaction sigprocmask sigsetmask socketpair syscall \
-		ttrace wborder setlocale
+		ttrace wborder setlocale iconvlist
 do
 as_ac_var=`echo "ac_cv_func_$ac_func" | $as_tr_sh`
 echo "$as_me:$LINENO: checking for $ac_func" >&5
@@ -15550,6 +15841,70 @@ fi
 done
 
 
+  echo "$as_me:$LINENO: checking for nl_langinfo and CODESET" >&5
+echo $ECHO_N "checking for nl_langinfo and CODESET... $ECHO_C" >&6
+if test "${am_cv_langinfo_codeset+set}" = set; then
+  echo $ECHO_N "(cached) $ECHO_C" >&6
+else
+  cat >conftest.$ac_ext <<_ACEOF
+/* confdefs.h.  */
+_ACEOF
+cat confdefs.h >>conftest.$ac_ext
+cat >>conftest.$ac_ext <<_ACEOF
+/* end confdefs.h.  */
+#include <langinfo.h>
+int
+main ()
+{
+char* cs = nl_langinfo(CODESET);
+  ;
+  return 0;
+}
+_ACEOF
+rm -f conftest.$ac_objext conftest$ac_exeext
+if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
+  (eval $ac_link) 2>conftest.er1
+  ac_status=$?
+  grep -v '^ *+' conftest.er1 >conftest.err
+  rm -f conftest.er1
+  cat conftest.err >&5
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); } &&
+	 { ac_try='test -z "$ac_c_werror_flag"
+			 || test ! -s conftest.err'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; } &&
+	 { ac_try='test -s conftest$ac_exeext'
+  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
+  (eval $ac_try) 2>&5
+  ac_status=$?
+  echo "$as_me:$LINENO: \$? = $ac_status" >&5
+  (exit $ac_status); }; }; then
+  am_cv_langinfo_codeset=yes
+else
+  echo "$as_me: failed program was:" >&5
+sed 's/^/| /' conftest.$ac_ext >&5
+
+am_cv_langinfo_codeset=no
+fi
+rm -f conftest.err conftest.$ac_objext \
+      conftest$ac_exeext conftest.$ac_ext
+
+fi
+echo "$as_me:$LINENO: result: $am_cv_langinfo_codeset" >&5
+echo "${ECHO_T}$am_cv_langinfo_codeset" >&6
+  if test $am_cv_langinfo_codeset = yes; then
+
+cat >>confdefs.h <<\_ACEOF
+#define HAVE_LANGINFO_CODESET 1
+_ACEOF
+
+  fi
+
+
 # Check the return and argument types of ptrace.  No canned test for
 # this, so roll our own.
 gdb_ptrace_headers='
@@ -20776,230 +21131,11 @@ done
 
 
 
-
 cat >>confdefs.h <<\_ACEOF
-#define GDB_DEFAULT_HOST_CHARSET "ISO-8859-1"
+#define GDB_DEFAULT_HOST_CHARSET "UTF-8"
 _ACEOF
 
 
-
-
-
-# Check whether --with-libiconv-prefix or --without-libiconv-prefix was given.
-if test "${with_libiconv_prefix+set}" = set; then
-  withval="$with_libiconv_prefix"
-
-    for dir in `echo "$withval" | tr : ' '`; do
-      if test -d $dir/include; then CPPFLAGS="$CPPFLAGS -I$dir/include"; fi
-      if test -d $dir/lib; then LDFLAGS="$LDFLAGS -L$dir/lib"; fi
-    done
-
-fi;
-
-  echo "$as_me:$LINENO: checking for iconv" >&5
-echo $ECHO_N "checking for iconv... $ECHO_C" >&6
-if test "${am_cv_func_iconv+set}" = set; then
-  echo $ECHO_N "(cached) $ECHO_C" >&6
-else
-
-    am_cv_func_iconv="no, consider installing GNU libiconv"
-    am_cv_lib_iconv=no
-    cat >conftest.$ac_ext <<_ACEOF
-/* confdefs.h.  */
-_ACEOF
-cat confdefs.h >>conftest.$ac_ext
-cat >>conftest.$ac_ext <<_ACEOF
-/* end confdefs.h.  */
-#include <stdlib.h>
-#include <iconv.h>
-int
-main ()
-{
-iconv_t cd = iconv_open("","");
-       iconv(cd,NULL,NULL,NULL,NULL);
-       iconv_close(cd);
-  ;
-  return 0;
-}
-_ACEOF
-rm -f conftest.$ac_objext conftest$ac_exeext
-if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
-  (eval $ac_link) 2>conftest.er1
-  ac_status=$?
-  grep -v '^ *+' conftest.er1 >conftest.err
-  rm -f conftest.er1
-  cat conftest.err >&5
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); } &&
-	 { ac_try='test -z "$ac_c_werror_flag"
-			 || test ! -s conftest.err'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; } &&
-	 { ac_try='test -s conftest$ac_exeext'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; }; then
-  am_cv_func_iconv=yes
-else
-  echo "$as_me: failed program was:" >&5
-sed 's/^/| /' conftest.$ac_ext >&5
-
-fi
-rm -f conftest.err conftest.$ac_objext \
-      conftest$ac_exeext conftest.$ac_ext
-    if test "$am_cv_func_iconv" != yes; then
-      am_save_LIBS="$LIBS"
-      LIBS="$LIBS -liconv"
-      cat >conftest.$ac_ext <<_ACEOF
-/* confdefs.h.  */
-_ACEOF
-cat confdefs.h >>conftest.$ac_ext
-cat >>conftest.$ac_ext <<_ACEOF
-/* end confdefs.h.  */
-#include <stdlib.h>
-#include <iconv.h>
-int
-main ()
-{
-iconv_t cd = iconv_open("","");
-         iconv(cd,NULL,NULL,NULL,NULL);
-         iconv_close(cd);
-  ;
-  return 0;
-}
-_ACEOF
-rm -f conftest.$ac_objext conftest$ac_exeext
-if { (eval echo "$as_me:$LINENO: \"$ac_link\"") >&5
-  (eval $ac_link) 2>conftest.er1
-  ac_status=$?
-  grep -v '^ *+' conftest.er1 >conftest.err
-  rm -f conftest.er1
-  cat conftest.err >&5
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); } &&
-	 { ac_try='test -z "$ac_c_werror_flag"
-			 || test ! -s conftest.err'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; } &&
-	 { ac_try='test -s conftest$ac_exeext'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; }; then
-  am_cv_lib_iconv=yes
-        am_cv_func_iconv=yes
-else
-  echo "$as_me: failed program was:" >&5
-sed 's/^/| /' conftest.$ac_ext >&5
-
-fi
-rm -f conftest.err conftest.$ac_objext \
-      conftest$ac_exeext conftest.$ac_ext
-      LIBS="$am_save_LIBS"
-    fi
-
-fi
-echo "$as_me:$LINENO: result: $am_cv_func_iconv" >&5
-echo "${ECHO_T}$am_cv_func_iconv" >&6
-  if test "$am_cv_func_iconv" = yes; then
-
-cat >>confdefs.h <<\_ACEOF
-#define HAVE_ICONV 1
-_ACEOF
-
-    echo "$as_me:$LINENO: checking for iconv declaration" >&5
-echo $ECHO_N "checking for iconv declaration... $ECHO_C" >&6
-    if test "${am_cv_proto_iconv+set}" = set; then
-  echo $ECHO_N "(cached) $ECHO_C" >&6
-else
-
-      cat >conftest.$ac_ext <<_ACEOF
-/* confdefs.h.  */
-_ACEOF
-cat confdefs.h >>conftest.$ac_ext
-cat >>conftest.$ac_ext <<_ACEOF
-/* end confdefs.h.  */
-
-#include <stdlib.h>
-#include <iconv.h>
-extern
-#ifdef __cplusplus
-"C"
-#endif
-#if defined(__STDC__) || defined(__cplusplus)
-size_t iconv (iconv_t cd, char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);
-#else
-size_t iconv();
-#endif
-
-int
-main ()
-{
-
-  ;
-  return 0;
-}
-_ACEOF
-rm -f conftest.$ac_objext
-if { (eval echo "$as_me:$LINENO: \"$ac_compile\"") >&5
-  (eval $ac_compile) 2>conftest.er1
-  ac_status=$?
-  grep -v '^ *+' conftest.er1 >conftest.err
-  rm -f conftest.er1
-  cat conftest.err >&5
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); } &&
-	 { ac_try='test -z "$ac_c_werror_flag"
-			 || test ! -s conftest.err'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; } &&
-	 { ac_try='test -s conftest.$ac_objext'
-  { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
-  (eval $ac_try) 2>&5
-  ac_status=$?
-  echo "$as_me:$LINENO: \$? = $ac_status" >&5
-  (exit $ac_status); }; }; then
-  am_cv_proto_iconv_arg1=""
-else
-  echo "$as_me: failed program was:" >&5
-sed 's/^/| /' conftest.$ac_ext >&5
-
-am_cv_proto_iconv_arg1="const"
-fi
-rm -f conftest.err conftest.$ac_objext conftest.$ac_ext
-      am_cv_proto_iconv="extern size_t iconv (iconv_t cd, $am_cv_proto_iconv_arg1 char * *inbuf, size_t *inbytesleft, char * *outbuf, size_t *outbytesleft);"
-fi
-
-    am_cv_proto_iconv=`echo "$am_cv_proto_iconv" | tr -s ' ' | sed -e 's/( /(/'`
-    echo "$as_me:$LINENO: result: ${ac_t:-
-         }$am_cv_proto_iconv" >&5
-echo "${ECHO_T}${ac_t:-
-         }$am_cv_proto_iconv" >&6
-
-cat >>confdefs.h <<_ACEOF
-#define ICONV_CONST $am_cv_proto_iconv_arg1
-_ACEOF
-
-  fi
-  LIBICONV=
-  if test "$am_cv_lib_iconv" = yes; then
-    LIBICONV="-liconv"
-  fi
-
-
-
                               ac_config_files="$ac_config_files Makefile .gdbinit:gdbinit.in gnulib/Makefile"
           ac_config_commands="$ac_config_commands default"
 cat >confcache <<\_ACEOF
@@ -21880,6 +22016,9 @@ s,@WINDRES@,$WINDRES,;t t
 s,@ac_ct_WINDRES@,$ac_ct_WINDRES,;t t
 s,@MIG@,$MIG,;t t
 s,@ac_ct_MIG@,$ac_ct_MIG,;t t
+s,@LIBICONV@,$LIBICONV,;t t
+s,@LIBICONV_INCLUDE@,$LIBICONV_INCLUDE,;t t
+s,@LIBICONV_LIBDIR@,$LIBICONV_LIBDIR,;t t
 s,@READLINE@,$READLINE,;t t
 s,@READLINE_DEPS@,$READLINE_DEPS,;t t
 s,@READLINE_CFLAGS@,$READLINE_CFLAGS,;t t
@@ -21944,7 +22083,6 @@ s,@CONFIG_UNINSTALL@,$CONFIG_UNINSTALL,;t t
 s,@target_subdir@,$target_subdir,;t t
 s,@frags@,$frags,;t t
 s,@nm_h@,$nm_h,;t t
-s,@LIBICONV@,$LIBICONV,;t t
 s,@LIBOBJS@,$LIBOBJS,;t t
 s,@LTLIBOBJS@,$LTLIBOBJS,;t t
 s,@gl_LIBOBJS@,$gl_LIBOBJS,;t t

gdb/configure.ac

@@ -1,6 +1,6 @@
 dnl Autoconf configure script for GDB, the GNU debugger.
 dnl Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
-dnl 2005, 2006, 2007, 2008
+dnl 2005, 2006, 2007, 2008, 2009
 dnl Free Software Foundation, Inc.
 dnl
 dnl This file is part of GDB.
@@ -430,6 +430,8 @@ AC_SEARCH_LIBS(zlibVersion, z, [AC_CHECK_HEADERS(zlib.h)])
 # On HP/UX we may need libxpdl for dlgetmodinfo (used by solib-pa64.c).
 AC_SEARCH_LIBS(dlgetmodinfo, [dl xpdl])
 
+AM_ICONV
+
 # On alpha-osf, it appears that libtermcap and libcurses are not compatible.
 # There is a very specific comment in /usr/include/curses.h explaining that
 # termcap routines built into libcurses must not be used.
@@ -776,7 +778,8 @@ AC_FUNC_VFORK
 AC_CHECK_FUNCS([canonicalize_file_name realpath getrusage getuid \
                 getgid poll pread64 sbrk setpgid setpgrp setsid \
 		sigaction sigprocmask sigsetmask socketpair syscall \
-		ttrace wborder setlocale])
+		ttrace wborder setlocale iconvlist])
+AM_LANGINFO_CODESET
 
 # Check the return and argument types of ptrace.  No canned test for
 # this, so roll our own.
@@ -1930,17 +1933,10 @@ dnl Check for exe extension set on certain hosts (e.g. Win32)
 AC_EXEEXT
 
 dnl  Detect the character set used by this host.
-
+dnl  At the moment, we just assume it's UTF-8.
-dnl  At the moment, we just assume it's ISO-8859-1 (which is a
+AC_DEFINE(GDB_DEFAULT_HOST_CHARSET, "UTF-8",
-dnl  superset of ASCII containing the characters needed for French,
-dnl  German, Spanish, Italian, and possibly others), but if were
-dnl  *were* to support any host character sets other than ISO-8859-1,
-dnl  here's where we'd detect it.
-AC_DEFINE(GDB_DEFAULT_HOST_CHARSET, "ISO-8859-1",
           [Define to be a string naming the default host character set.])
 
-AM_ICONV
-
 AC_OUTPUT(Makefile .gdbinit:gdbinit.in gnulib/Makefile,
 [
 dnl Autoconf doesn't provide a mechanism for modifying definitions 

gdb/defs.h

@@ -61,6 +61,8 @@
 
 #include "gdb_locale.h"
 
+#include "gdb_wchar.h"
+
 /* For ``enum target_signal''.  */
 #include "gdb/signals.h"
 

gdb/doc/ChangeLog

@@ -1,3 +1,9 @@
+2009-03-20  Tom Tromey  <tromey@redhat.com>
+
+	* gdb.texinfo (Character Sets): Remove obsolete text.  Document
+	set target-wide-charset.
+	(Requirements): Mention iconv.
+
 2009-03-17  Hui Zhu  <teawater@gmail.com>
 
 	* gdb.texinfo: Change the introduce of "disassemble-next-line".

gdb/doc/gdb.texinfo

@@ -7959,13 +7959,17 @@ support:
 @table @code
 @item set target-charset @var{charset}
 @kindex set target-charset
-Set the current target character set to @var{charset}.  We list the
+Set the current target character set to @var{charset}.  If you type
-character set names @value{GDBN} recognizes below, but if you type
+@code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN}
-@code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
+will list the target character sets it supports.
-list the target character sets it supports.
+
-@end table
+@item set target-wide-charset @var{charset}
+@kindex set target-wide-charset
+Set the current target wide character set to @var{charset}.  The
+target wide character set is the character set used by @code{wchar_t}.
+If you type @code{set target-charset} followed by @key{TAB}@key{TAB},
+@value{GDBN} will list the target character sets it supports.
 
-@table @code
 @item set host-charset @var{charset}
 @kindex set host-charset
 Set the current host character set to @var{charset}.
@@ -7975,10 +7979,9 @@ system it is running on; you can override that default using the
 @code{set host-charset} command.
 
 @value{GDBN} can only use certain character sets as its host character
-set.  We list the character set names @value{GDBN} recognizes below, and
+set.  If you type @code{set target-charset} followed by
-indicate which can be host character sets, but if you type
+@key{TAB}@key{TAB}, @value{GDBN} will list the host character sets it
-@code{set target-charset} followed by @key{TAB}@key{TAB}, @value{GDBN} will
+supports.
-list the host character sets it supports.
 
 @item set charset @var{charset}
 @kindex set charset
@@ -8002,37 +8005,6 @@ Show the name of the current target charset.
 
 @end table
 
-@value{GDBN} currently includes support for the following character
-sets:
-
-@table @code
-
-@item ASCII
-@cindex ASCII character set
-Seven-bit U.S. @sc{ascii}.  @value{GDBN} can use this as its host
-character set.
-
-@item ISO-8859-1
-@cindex ISO 8859-1 character set
-@cindex ISO Latin 1 character set
-The ISO Latin 1 character set.  This extends @sc{ascii} with accented
-characters needed for French, German, and Spanish.  @value{GDBN} can use
-this as its host character set.
-
-@item EBCDIC-US
-@itemx IBM1047
-@cindex EBCDIC character set
-@cindex IBM1047 character set
-Variants of the @sc{ebcdic} character set, used on some of IBM's
-mainframe operating systems.  (@sc{gnu}/Linux on the S/390 uses U.S. @sc{ascii}.)
-@value{GDBN} cannot use these as its host character set.
-
-@end table
-
-Note that these are all single-byte character sets.  More work inside
-@value{GDBN} is needed to support multi-byte or variable-width character
-encodings, like the UTF-8 and UCS-2 encodings of Unicode.
-
 Here is an example of @value{GDBN}'s character set support in action.
 Assume that the following source code has been placed in the file
 @file{charset-test.c}:
@@ -24826,6 +24798,24 @@ The @samp{zlib} library is likely included with your operating system
 distribution; if it is not, you can get the latest version from
 @url{http://zlib.net}.
 
+@item iconv
+@value{GDBN}'s features related to character sets (@pxref{Character
+Sets}) require a functioning @code{iconv} implementation.  If you are
+on a GNU system, then this is provided by the GNU C Library.  Some
+other systems also provide a working @code{iconv}.
+
+On systems with @code{iconv}, you can install GNU Libiconv.  If you
+have previously installed Libiconv, you can use the
+@option{--with-libiconv-prefix} option to configure.
+
+@value{GDBN}'s top-level @file{configure} and @file{Makefile} will
+arrange to build Libiconv if a directory named @file{libiconv} appears
+in the top-most source directory.  If Libiconv is built this way, and
+if the operating system does not provide a suitable @code{iconv}
+implementation, then the just-built library will automatically be used
+by @value{GDBN}.  One easy way to set this up is to download GNU
+Libiconv, unpack it, and then rename the directory holding the
+Libiconv source code to @samp{libiconv}.
 @end table
 
 @node Running Configure

gdb/expprint.c

@@ -186,8 +186,8 @@ print_subexp_standard (struct expression *exp, int *pos,
 	   If necessary, we can temporarily set it to zero, or pass it as an
 	   additional parameter to LA_PRINT_STRING.  -fnf */
 	get_user_print_options (&opts);
-	LA_PRINT_STRING (stream, &exp->elts[pc + 2].string, nargs, 1, 0,
+	LA_PRINT_STRING (stream, builtin_type (exp->gdbarch)->builtin_char,
-			 &opts);
+			 &exp->elts[pc + 2].string, nargs, 0, &opts);
       }
       return;
 
@@ -205,8 +205,8 @@ print_subexp_standard (struct expression *exp, int *pos,
 	(*pos) += 3 + BYTES_TO_EXP_ELEM (nargs + 1);
 	fputs_filtered ("@\"", stream);
 	get_user_print_options (&opts);
-	LA_PRINT_STRING (stream, &exp->elts[pc + 2].string, nargs, 1, 0,
+	LA_PRINT_STRING (stream, builtin_type (exp->gdbarch)->builtin_char,
-			 &opts);
+			 &exp->elts[pc + 2].string, nargs, 0, &opts);
 	fputs_filtered ("\"", stream);
       }
       return;
@@ -291,8 +291,8 @@ print_subexp_standard (struct expression *exp, int *pos,
 	{
 	  struct value_print_options opts;
 	  get_user_print_options (&opts);
-	  LA_PRINT_STRING (stream, tempstr, nargs - 1, 1, 0,
+	  LA_PRINT_STRING (stream, builtin_type (exp->gdbarch)->builtin_char,
-			   &opts);
+			   tempstr, nargs - 1, 0, &opts);
 	  (*pos) = pc;
 	}
       else

gdb/f-lang.c

@@ -70,8 +70,9 @@ static SAVED_F77_COMMON_PTR allocate_saved_f77_common_node (void);
 static void patch_common_entries (SAVED_F77_COMMON_PTR, CORE_ADDR, int);
 #endif
 
-static void f_printchar (int c, struct ui_file * stream);
+static void f_printchar (int c, struct type *type, struct ui_file * stream);
-static void f_emit_char (int c, struct ui_file * stream, int quoter);
+static void f_emit_char (int c, struct type *type,
+			 struct ui_file * stream, int quoter);
 
 /* 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
@@ -80,7 +81,7 @@ static void f_emit_char (int c, struct ui_file * stream, int quoter);
    be replaced with a true F77 version.  */
 
 static void
-f_emit_char (int c, struct ui_file *stream, int quoter)
+f_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
   c &= 0xFF;			/* Avoid sign bit follies */
 
@@ -126,10 +127,10 @@ f_emit_char (int c, struct ui_file *stream, int quoter)
    be replaced with a true F77version. */
 
 static void
-f_printchar (int c, struct ui_file *stream)
+f_printchar (int c, struct type *type, struct ui_file *stream)
 {
   fputs_filtered ("'", stream);
-  LA_EMIT_CHAR (c, stream, '\'');
+  LA_EMIT_CHAR (c, type, stream, '\'');
   fputs_filtered ("'", stream);
 }
 
@@ -141,14 +142,15 @@ f_printchar (int c, struct ui_file *stream)
    be replaced with a true F77 version. */
 
 static void
-f_printstr (struct ui_file *stream, const gdb_byte *string,
+f_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
-	    unsigned int length, int width, int force_ellipses,
+	    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);
 
   if (length == 0)
     {
@@ -190,7 +192,7 @@ f_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("', ", stream);
 	      in_quotes = 0;
 	    }
-	  f_printchar (string[i], stream);
+	  f_printchar (string[i], type, stream);
 	  fprintf_filtered (stream, " <repeats %u times>", reps);
 	  i = rep1 - 1;
 	  things_printed += options->repeat_count_threshold;
@@ -206,7 +208,7 @@ f_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("'", stream);
 	      in_quotes = 1;
 	    }
-	  LA_EMIT_CHAR (string[i], stream, '"');
+	  LA_EMIT_CHAR (string[i], type, stream, '"');
 	  ++things_printed;
 	}
     }

gdb/f-valprint.c

@@ -256,7 +256,8 @@ f_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
     {
     case TYPE_CODE_STRING:
       f77_get_dynamic_length_of_aggregate (type);
-      LA_PRINT_STRING (stream, valaddr, TYPE_LENGTH (type), 1, 0, options);
+      LA_PRINT_STRING (stream, builtin_type (current_gdbarch)->builtin_char,
+		       valaddr, TYPE_LENGTH (type), 0, options);
       break;
 
     case TYPE_CODE_ARRAY:
@@ -293,7 +294,7 @@ f_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	      && TYPE_CODE (elttype) == TYPE_CODE_INT
 	      && (options->format == 0 || options->format == 's')
 	      && addr != 0)
-	    i = val_print_string (addr, -1, TYPE_LENGTH (elttype), stream,
+	    i = val_print_string (TYPE_TARGET_TYPE (type), addr, -1, stream,
 				  options);
 
 	  /* Return number of characters printed, including the terminating
@@ -365,7 +366,7 @@ f_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	    {
 	      fputs_filtered (" ", stream);
 	      LA_PRINT_CHAR ((unsigned char) unpack_long (type, valaddr),
-			     stream);
+			     type, stream);
 	    }
 	}
       break;

gdb/gdb_locale.h

@@ -41,4 +41,8 @@
 # define N_(String) (String)
 #endif
 
+#ifdef HAVE_LANGINFO_CODESET
+#include <langinfo.h>
+#endif
+
 #endif /* GDB_LOCALE_H */

gdb/gdb_obstack.h

@@ -45,4 +45,7 @@
 #define obstack_grow_str0(OBSTACK,STRING) \
   obstack_grow0 (OBSTACK, STRING, strlen (STRING))
 
+#define obstack_grow_wstr(OBSTACK, WSTRING) \
+  obstack_grow (OBSTACK, WSTRING, sizeof (gdb_wchar_t) * gdb_wcslen (WSTRING))
+
 #endif

gdb/gdb_wchar.h

@@ -0,0 +1,62 @@
+/* Wide characters for gdb
+   Copyright (C) 2009 Free Software Foundation, Inc.
+
+   This file is part of GDB.
+
+   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 3 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, see <http://www.gnu.org/licenses/>.  */
+
+#ifndef GDB_WCHAR_H
+#define GDB_WCHAR_H
+
+/* If this host has wchar_t and if iconv is available (perhaps via GNU
+   libiconv), then we arrange to use those.  Otherwise, we provide a
+   phony iconv which only handles a single character set, and we
+   provide wrappers for the wchar_t functionality we use.  */
+#if defined(HAVE_ICONV) && defined(HAVE_WCHAR_H)
+
+#include <iconv.h>
+#include <wchar.h>
+#include <wctype.h>
+
+typedef wchar_t gdb_wchar_t;
+typedef wint_t gdb_wint_t;
+
+#define gdb_wcslen wcslen
+#define gdb_iswprint iswprint
+#define gdb_iswdigit iswdigit
+#define gdb_btowc btowc
+#define gdb_WEOF WEOF
+
+#define LCST(X) L ## X
+
+#else
+
+typedef char gdb_wchar_t;
+typedef int gdb_wint_t;
+
+#define gdb_wcslen strlen
+#define gdb_iswprint isprint
+#define gdb_iswdigit isdigit
+#define gdb_btowc /* empty */
+#define gdb_WEOF EOF
+
+#define LCST(X) X
+
+/* This define is used elsewhere so we don't need to duplicate the
+   same checking logic in multiple places.  */
+#define PHONY_ICONV
+
+#endif /* defined(HAVE_ICONV) && defined(HAVE_WCHAR_H) */
+
+#endif /* GDB_WCHAR_H */

gdb/jv-lang.c

@@ -61,7 +61,8 @@ static char *get_java_utf8_name (struct obstack *obstack, struct value *name);
 static int java_class_is_primitive (struct value *clas);
 static struct value *java_value_string (char *ptr, int len);
 
-static void java_emit_char (int c, struct ui_file * stream, int quoter);
+static void java_emit_char (int c, struct type *type,
+			    struct ui_file * stream, int quoter);
 
 static char *java_class_name_from_physname (const char *physname);
 
@@ -796,7 +797,7 @@ java_value_string (char *ptr, int len)
    characters and strings is language specific. */
 
 static void
-java_emit_char (int c, struct ui_file *stream, int quoter)
+java_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
   switch (c)
     {

gdb/jv-valprint.c

@@ -230,7 +230,7 @@ java_value_print (struct value *val, struct ui_file *stream,
 
       value_free_to_mark (mark);	/* Release unnecessary values */
 
-      val_print_string (data + boffset, count, 2, stream, options);
+      val_print_string (java_char_type, data + boffset, count, stream, options);
 
       return 0;
     }
@@ -520,7 +520,7 @@ java_val_print (struct type *type, const gdb_byte *valaddr,
 	       || (TYPE_CODE (type) == TYPE_CODE_INT
 		   && TYPE_LENGTH (type) == 2
 		   && strcmp (TYPE_NAME (type), "char") == 0))
-	LA_PRINT_CHAR ((int) unpack_long (type, valaddr), stream);
+	LA_PRINT_CHAR ((int) unpack_long (type, valaddr), type, stream);
       else
 	val_print_type_code_int (type, valaddr, stream);
       break;

gdb/language.c

@@ -65,9 +65,11 @@ static void set_check (char *, int);
 
 static void set_type_range_case (void);
 
-static void unk_lang_emit_char (int c, struct ui_file *stream, int quoter);
+static void unk_lang_emit_char (int c, struct type *type,
+				struct ui_file *stream, int quoter);
 
-static void unk_lang_printchar (int c, struct ui_file *stream);
+static void unk_lang_printchar (int c, struct type *type,
+				struct ui_file *stream);
 
 static void unk_lang_print_type (struct type *, char *, struct ui_file *,
 				 int, int);
@@ -1065,20 +1067,22 @@ unk_lang_error (char *msg)
 }
 
 static void
-unk_lang_emit_char (int c, struct ui_file *stream, int quoter)
+unk_lang_emit_char (int c, struct type *type, struct ui_file *stream,
+		    int quoter)
 {
   error (_("internal error - unimplemented function unk_lang_emit_char called."));
 }
 
 static void
-unk_lang_printchar (int c, struct ui_file *stream)
+unk_lang_printchar (int c, struct type *type, struct ui_file *stream)
 {
   error (_("internal error - unimplemented function unk_lang_printchar called."));
 }
 
 static void
-unk_lang_printstr (struct ui_file *stream, const gdb_byte *string,
+unk_lang_printstr (struct ui_file *stream, struct type *type,
-		   unsigned int length, int width, int force_ellipses,
+		   const gdb_byte *string, unsigned int length,
+		   int force_ellipses,
 		   const struct value_print_options *options)
 {
   error (_("internal error - unimplemented function unk_lang_printstr called."));

gdb/language.h

@@ -186,14 +186,15 @@ struct language_defn
 
     void (*la_post_parser) (struct expression ** expp, int void_context_p);
 
-    void (*la_printchar) (int ch, struct ui_file * stream);
+    void (*la_printchar) (int ch, struct type *chtype, struct ui_file * stream);
 
-    void (*la_printstr) (struct ui_file * stream, const gdb_byte *string,
+    void (*la_printstr) (struct ui_file * stream, struct type *elttype,
-			 unsigned int length, int width,
+			 const gdb_byte *string, unsigned int length,
 			 int force_ellipses,
 			 const struct value_print_options *);
 
-    void (*la_emitchar) (int ch, struct ui_file * stream, int quoter);
+    void (*la_emitchar) (int ch, struct type *chtype,
+			 struct ui_file * stream, int quoter);
 
     /* Print a type using syntax appropriate for this language. */
 
@@ -381,13 +382,13 @@ extern enum language set_language (enum language);
 #define LA_VALUE_PRINT(val,stream,options) \
   (current_language->la_value_print(val,stream,options))
 
-#define LA_PRINT_CHAR(ch, stream) \
+#define LA_PRINT_CHAR(ch, type, stream) \
-  (current_language->la_printchar(ch, stream))
+  (current_language->la_printchar(ch, type, stream))
-#define LA_PRINT_STRING(stream, string, length, width, force_ellipses,options) \
+#define LA_PRINT_STRING(stream, elttype, string, length, force_ellipses,options) \
-  (current_language->la_printstr(stream, string, length, width, \
+  (current_language->la_printstr(stream, elttype, string, length, \
 				 force_ellipses,options))
-#define LA_EMIT_CHAR(ch, stream, quoter) \
+#define LA_EMIT_CHAR(ch, type, stream, quoter) \
-  (current_language->la_emitchar(ch, stream, quoter))
+  (current_language->la_emitchar(ch, type, stream, quoter))
 #define LA_GET_STRING(value, buffer, length, encoding) \
   (current_language->la_get_string(value, buffer, length, encoding))
 

gdb/m2-lang.c

@@ -29,8 +29,8 @@
 #include "valprint.h"
 
 extern void _initialize_m2_language (void);
-static void m2_printchar (int, struct ui_file *);
+static void m2_printchar (int, struct type *, struct ui_file *);
-static void m2_emit_char (int, struct ui_file *, int);
+static void m2_emit_char (int, struct type *, struct ui_file *, int);
 
 /* 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
@@ -39,7 +39,7 @@ static void m2_emit_char (int, struct ui_file *, int);
    be replaced with a true Modula version.  */
 
 static void
-m2_emit_char (int c, struct ui_file *stream, int quoter)
+m2_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
 
   c &= 0xFF;			/* Avoid sign bit follies */
@@ -88,10 +88,10 @@ m2_emit_char (int c, struct ui_file *stream, int quoter)
    be replaced with a true Modula version.  */
 
 static void
-m2_printchar (int c, struct ui_file *stream)
+m2_printchar (int c, struct type *type, struct ui_file *stream)
 {
   fputs_filtered ("'", stream);
-  LA_EMIT_CHAR (c, stream, '\'');
+  LA_EMIT_CHAR (c, type, stream, '\'');
   fputs_filtered ("'", stream);
 }
 
@@ -103,14 +103,15 @@ m2_printchar (int c, struct ui_file *stream)
    be replaced with a true Modula version.  */
 
 static void
-m2_printstr (struct ui_file *stream, const gdb_byte *string,
+m2_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
-	     unsigned int length, int width, int force_ellipses,
+	     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);
 
   if (length == 0)
     {
@@ -152,7 +153,7 @@ m2_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("\", ", stream);
 	      in_quotes = 0;
 	    }
-	  m2_printchar (string[i], stream);
+	  m2_printchar (string[i], type, stream);
 	  fprintf_filtered (stream, " <repeats %u times>", reps);
 	  i = rep1 - 1;
 	  things_printed += options->repeat_count_threshold;
@@ -168,7 +169,7 @@ m2_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("\"", stream);
 	      in_quotes = 1;
 	    }
-	  LA_EMIT_CHAR (string[i], stream, '"');
+	  LA_EMIT_CHAR (string[i], type, stream, '"');
 	  ++things_printed;
 	}
     }

gdb/m2-valprint.c

@@ -237,7 +237,8 @@ print_unpacked_pointer (struct type *type,
       && TYPE_CODE (elttype) == TYPE_CODE_INT
       && (options->format == 0 || options->format == 's')
       && addr != 0)
-    return val_print_string (addr, -1, TYPE_LENGTH (elttype), stream, options);
+    return val_print_string (TYPE_TARGET_TYPE (type), addr, -1,
+			     stream, options);
   
   return 0;
 }
@@ -294,7 +295,7 @@ m2_print_array_contents (struct type *type, const gdb_byte *valaddr,
 	   || ((current_language->la_language == language_m2)
 	       && (TYPE_CODE (type) == TYPE_CODE_CHAR)))
 	  && (options->format == 0 || options->format == 's'))
-	val_print_string (address, len+1, eltlen, stream, options);
+	val_print_string (type, address, len+1, stream, options);
       else
 	{
 	  fprintf_filtered (stream, "{");
@@ -359,7 +360,8 @@ m2_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 		  len = temp_len;
 		}
 
-	      LA_PRINT_STRING (stream, valaddr + embedded_offset, len, 1, 0,
+	      LA_PRINT_STRING (stream, TYPE_TARGET_TYPE (type),
+			       valaddr + embedded_offset, len, 0,
 			       options);
 	      i = len;
 	    }
@@ -547,7 +549,7 @@ m2_val_print (struct type *type, const gdb_byte *valaddr, int embedded_offset,
 	  else
 	    fprintf_filtered (stream, "%d", (int) val);
 	  fputs_filtered (" ", stream);
-	  LA_PRINT_CHAR ((unsigned char) val, stream);
+	  LA_PRINT_CHAR ((unsigned char) val, type, stream);
 	}
       break;
 

gdb/macroexp.c

@@ -23,6 +23,7 @@
 #include "macrotab.h"
 #include "macroexp.h"
 #include "gdb_assert.h"
+#include "c-lang.h"
 
 
 
@@ -320,14 +321,17 @@ get_character_constant (struct macro_buffer *tok, char *p, char *end)
      way GDB's C/C++ lexer does.  So we call parse_escape in utils.c
      to handle escape sequences.  */
   if ((p + 1 <= end && *p == '\'')
-      || (p + 2 <= end && p[0] == 'L' && p[1] == '\''))
+      || (p + 2 <= end
+	  && (p[0] == 'L' || p[0] == 'u' || p[0] == 'U')
+	  && p[1] == '\''))
     {
       char *tok_start = p;
       char *body_start;
+      int char_count = 0;
 
       if (*p == '\'')
         p++;
-      else if (*p == 'L')
+      else if (*p == 'L' || *p == 'u' || *p == 'U')
         p += 2;
       else
         gdb_assert (0);
@@ -339,7 +343,7 @@ get_character_constant (struct macro_buffer *tok, char *p, char *end)
             error (_("Unmatched single quote."));
           else if (*p == '\'')
             {
-              if (p == body_start)
+              if (!char_count)
                 error (_("A character constant must contain at least one "
                        "character."));
               p++;
@@ -348,10 +352,13 @@ get_character_constant (struct macro_buffer *tok, char *p, char *end)
           else if (*p == '\\')
             {
               p++;
-              parse_escape (&p);
+	      char_count += c_parse_escape (&p, NULL);
             }
           else
-            p++;
+	    {
+	      p++;
+	      char_count++;
+	    }
         }
 
       set_token (tok, tok_start, p);
@@ -370,16 +377,16 @@ static int
 get_string_literal (struct macro_buffer *tok, char *p, char *end)
 {
   if ((p + 1 <= end
-       && *p == '\"')
+       && *p == '"')
       || (p + 2 <= end
-          && p[0] == 'L'
+          && (p[0] == 'L' || p[0] == 'u' || p[0] == 'U')
-          && p[1] == '\"'))
+          && p[1] == '"'))
     {
       char *tok_start = p;
 
-      if (*p == '\"')
+      if (*p == '"')
         p++;
-      else if (*p == 'L')
+      else if (*p == 'L' || *p == 'u' || *p == 'U')
         p += 2;
       else
         gdb_assert (0);
@@ -388,7 +395,7 @@ get_string_literal (struct macro_buffer *tok, char *p, char *end)
         {
           if (p >= end)
             error (_("Unterminated string in expression."));
-          else if (*p == '\"')
+          else if (*p == '"')
             {
               p++;
               break;
@@ -399,7 +406,7 @@ get_string_literal (struct macro_buffer *tok, char *p, char *end)
           else if (*p == '\\')
             {
               p++;
-              parse_escape (&p);
+              c_parse_escape (&p, NULL);
             }
           else
             p++;

gdb/objc-lang.c

@@ -280,7 +280,7 @@ objc_demangle (const char *mangled, int options)
    for printing characters and strings is language specific.  */
 
 static void
-objc_emit_char (int c, struct ui_file *stream, int quoter)
+objc_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
 
   c &= 0xFF;			/* Avoid sign bit follies.  */
@@ -326,10 +326,10 @@ objc_emit_char (int c, struct ui_file *stream, int quoter)
 }
 
 static void
-objc_printchar (int c, struct ui_file *stream)
+objc_printchar (int c, struct type *type, struct ui_file *stream)
 {
   fputs_filtered ("'", stream);
-  objc_emit_char (c, stream, '\'');
+  objc_emit_char (c, type, stream, '\'');
   fputs_filtered ("'", stream);
 }
 
@@ -340,14 +340,16 @@ objc_printchar (int c, struct ui_file *stream)
    FORCE_ELLIPSES.  */
 
 static void
-objc_printstr (struct ui_file *stream, const gdb_byte *string, 
+objc_printstr (struct ui_file *stream, struct type *type,
-	       unsigned int length, int width, int force_ellipses,
+	       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);
 
   /* 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
@@ -395,7 +397,7 @@ objc_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("\", ", stream);
 	      in_quotes = 0;
 	    }
-	  objc_printchar (string[i], stream);
+	  objc_printchar (string[i], type, stream);
 	  fprintf_filtered (stream, " <repeats %u times>", reps);
 	  i = rep1 - 1;
 	  things_printed += options->repeat_count_threshold;
@@ -411,7 +413,7 @@ objc_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("\"", stream);
 	      in_quotes = 1;
 	    }
-	  objc_emit_char (string[i], stream, '"');
+	  objc_emit_char (string[i], type, stream, '"');
 	  ++things_printed;
 	}
     }

gdb/p-lang.c

@@ -97,7 +97,8 @@ pascal_main_name (void)
    but this does not happen for Free Pascal nor for GPC.  */
 int
 is_pascal_string_type (struct type *type,int *length_pos,
-                       int *length_size, int *string_pos, int *char_size,
+                       int *length_size, int *string_pos,
+		       struct type **char_type,
 		       char **arrayname)
 {
   if (TYPE_CODE (type) == TYPE_CODE_STRUCT)
@@ -114,8 +115,8 @@ is_pascal_string_type (struct type *type,int *length_pos,
 	    *length_size = TYPE_LENGTH (TYPE_FIELD_TYPE (type, 0));
           if (string_pos)
 	    *string_pos = TYPE_FIELD_BITPOS (type, 1) / TARGET_CHAR_BIT;
-          if (char_size)
+          if (char_type)
-	    *char_size = 1;
+	    *char_type = TYPE_TARGET_TYPE (TYPE_FIELD_TYPE (type, 1));
  	  if (arrayname)
 	    *arrayname = TYPE_FIELDS (type)[1].name;
          return 2;
@@ -126,7 +127,6 @@ is_pascal_string_type (struct type *type,int *length_pos,
           && strcmp (TYPE_FIELDS (type)[0].name, "Capacity") == 0
           && strcmp (TYPE_FIELDS (type)[1].name, "length") == 0)
         {
-	  struct type *char_type;
 	  if (length_pos)
 	    *length_pos = TYPE_FIELD_BITPOS (type, 1) / TARGET_CHAR_BIT;
 	  if (length_size)
@@ -134,13 +134,12 @@ is_pascal_string_type (struct type *type,int *length_pos,
 	  if (string_pos)
 	    *string_pos = TYPE_FIELD_BITPOS (type, 2) / TARGET_CHAR_BIT;
           /* FIXME: how can I detect wide chars in GPC ?? */
-	  char_type = TYPE_FIELD_TYPE (type,2);
+          if (char_type)
-	  if (char_size && TYPE_CODE (char_type) == TYPE_CODE_ARRAY)
 	    {
-	      *char_size = TYPE_LENGTH (TYPE_TARGET_TYPE (char_type));
+	      *char_type = TYPE_TARGET_TYPE (TYPE_FIELD_TYPE (type, 2));
+	      if (TYPE_CODE (*char_type) == TYPE_CODE_ARRAY)
+		*char_type = TYPE_TARGET_TYPE (*char_type);
 	    }
-	  else if (char_size)
-	    *char_size = 1;
  	  if (arrayname)
 	    *arrayname = TYPE_FIELDS (type)[2].name;
          return 3;
@@ -182,14 +181,15 @@ pascal_one_char (int c, struct ui_file *stream, int *in_quotes)
     }
 }
 
-static void pascal_emit_char (int c, struct ui_file *stream, int quoter);
+static void pascal_emit_char (int c, struct type *type,
+			      struct ui_file *stream, int quoter);
 
 /* 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
-pascal_emit_char (int c, struct ui_file *stream, int quoter)
+pascal_emit_char (int c, struct type *type, struct ui_file *stream, int quoter)
 {
   int in_quotes = 0;
   pascal_one_char (c, stream, &in_quotes);
@@ -198,7 +198,7 @@ pascal_emit_char (int c, struct ui_file *stream, int quoter)
 }
 
 void
-pascal_printchar (int c, struct ui_file *stream)
+pascal_printchar (int c, struct type *type, struct ui_file *stream)
 {
   int in_quotes = 0;
   pascal_one_char (c, stream, &in_quotes);
@@ -212,14 +212,16 @@ pascal_printchar (int c, struct ui_file *stream)
    had to stop before printing LENGTH characters, or if FORCE_ELLIPSES.  */
 
 void
-pascal_printstr (struct ui_file *stream, const gdb_byte *string,
+pascal_printstr (struct ui_file *stream, struct type *type,
-		 unsigned int length, int width, int force_ellipses,
+		 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);
 
   /* 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
@@ -273,7 +275,7 @@ pascal_printstr (struct ui_file *stream, const gdb_byte *string,
 		fputs_filtered ("', ", stream);
 	      in_quotes = 0;
 	    }
-	  pascal_printchar (current_char, stream);
+	  pascal_printchar (current_char, type, stream);
 	  fprintf_filtered (stream, " <repeats %u times>", reps);
 	  i = rep1 - 1;
 	  things_printed += options->repeat_count_threshold;

gdb/p-lang.h

@@ -48,12 +48,13 @@ extern void pascal_type_print_method_args (char *, char *,
 /* These are in p-lang.c: */
 
 extern int 
-  is_pascal_string_type (struct type *, int *, int *, int *, int *, char **);
+  is_pascal_string_type (struct type *, int *, int *, int *,
+			 struct type **, char **);
 
-extern void pascal_printchar (int, struct ui_file *);
+extern void pascal_printchar (int, struct type *, struct ui_file *);
 
-extern void pascal_printstr (struct ui_file *, const gdb_byte *,
+extern void pascal_printstr (struct ui_file *, struct type *, const gdb_byte *,
-			     unsigned int, int, int,
+			     unsigned int, int,
 			     const struct value_print_options *);
 
 extern struct type **const (pascal_builtin_types[]);

gdb/p-valprint.c

@@ -61,7 +61,7 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
   struct type *elttype;
   unsigned eltlen;
   int length_pos, length_size, string_pos;
-  int char_size;
+  struct type *char_type;
   LONGEST val;
   CORE_ADDR addr;
 
@@ -100,8 +100,9 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
 		  len = temp_len;
 		}
 
-	      LA_PRINT_STRING (stream, valaddr + embedded_offset, len, 
+	      LA_PRINT_STRING (stream, TYPE_TARGET_TYPE (type),
-			       eltlen, 0, options);
+			       valaddr + embedded_offset, len, 0,
+			       options);
 	      i = len;
 	    }
 	  else
@@ -175,8 +176,7 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
 	      && addr != 0)
 	    {
 	      /* no wide string yet */
-	      i = val_print_string (addr, -1, TYPE_LENGTH (elttype), stream, 
+	      i = val_print_string (elttype, addr, -1, stream, options);
-		    options);
 	    }
 	  /* also for pointers to pascal strings */
 	  /* Note: this is Free Pascal specific:
@@ -184,7 +184,7 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
 	     Pascal strings are mapped to records
 	     with lowercase names PM  */
           if (is_pascal_string_type (elttype, &length_pos, &length_size,
-                                     &string_pos, &char_size, NULL)
+                                     &string_pos, &char_type, NULL)
 	      && addr != 0)
 	    {
 	      ULONGEST string_length;
@@ -193,7 +193,7 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
               read_memory (addr + length_pos, buffer, length_size);
 	      string_length = extract_unsigned_integer (buffer, length_size);
               xfree (buffer);
-              i = val_print_string (addr + string_pos, string_length, char_size, stream, options);
+              i = val_print_string (char_type ,addr + string_pos, string_length, stream, options);
 	    }
 	  else if (pascal_object_is_vtbl_member (type))
 	    {
@@ -298,10 +298,10 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
       else
 	{
           if (is_pascal_string_type (type, &length_pos, &length_size,
-                                     &string_pos, &char_size, NULL))
+                                     &string_pos, &char_type, NULL))
 	    {
 	      len = extract_unsigned_integer (valaddr + embedded_offset + length_pos, length_size);
-	      LA_PRINT_STRING (stream, valaddr + embedded_offset + string_pos, len, char_size, 0, options);
+	      LA_PRINT_STRING (stream, char_type, valaddr + embedded_offset + string_pos, len, 0, options);
 	    }
 	  else
 	    pascal_object_print_value_fields (type, valaddr + embedded_offset, address, stream,
@@ -426,7 +426,7 @@ pascal_val_print (struct type *type, const gdb_byte *valaddr,
 	  else
 	    fprintf_filtered (stream, "%d", (int) val);
 	  fputs_filtered (" ", stream);
-	  LA_PRINT_CHAR ((unsigned char) val, stream);
+	  LA_PRINT_CHAR ((unsigned char) val, type, stream);
 	}
       break;
 

gdb/parse.c

@@ -352,6 +352,65 @@ write_exp_string (struct stoken str)
   write_exp_elt_longcst ((LONGEST) len);
 }
 
+/* Add a vector of string constants to the end of the expression.
+
+   This adds an OP_STRING operation, but encodes the contents
+   differently from write_exp_string.  The language is expected to
+   handle evaluation of this expression itself.
+   
+   After the usual OP_STRING header, TYPE is written into the
+   expression as a long constant.  The interpretation of this field is
+   up to the language evaluator.
+   
+   Next, each string in VEC is written.  The length is written as a
+   long constant, followed by the contents of the string.  */
+
+void
+write_exp_string_vector (int type, struct stoken_vector *vec)
+{
+  int i, n_slots, len;
+
+  /* Compute the size.  We compute the size in number of slots to
+     avoid issues with string padding.  */
+  n_slots = 0;
+  for (i = 0; i < vec->len; ++i)
+    {
+      /* One slot for the length of this element, plus the number of
+	 slots needed for this string.  */
+      n_slots += 1 + BYTES_TO_EXP_ELEM (vec->tokens[i].length);
+    }
+
+  /* One more slot for the type of the string.  */
+  ++n_slots;
+
+  /* Now compute a phony string length.  */
+  len = EXP_ELEM_TO_BYTES (n_slots) - 1;
+
+  n_slots += 4;
+  if ((expout_ptr + n_slots) >= expout_size)
+    {
+      expout_size = max (expout_size * 2, expout_ptr + n_slots + 10);
+      expout = (struct expression *)
+	xrealloc ((char *) expout, (sizeof (struct expression)
+				    + EXP_ELEM_TO_BYTES (expout_size)));
+    }
+
+  write_exp_elt_opcode (OP_STRING);
+  write_exp_elt_longcst (len);
+  write_exp_elt_longcst (type);
+
+  for (i = 0; i < vec->len; ++i)
+    {
+      write_exp_elt_longcst (vec->tokens[i].length);
+      memcpy (&expout->elts[expout_ptr], vec->tokens[i].ptr,
+	      vec->tokens[i].length);
+      expout_ptr += BYTES_TO_EXP_ELEM (vec->tokens[i].length);
+    }
+
+  write_exp_elt_longcst (len);
+  write_exp_elt_opcode (OP_STRING);
+}
+
 /* Add a bitstring constant to the end of the expression.
 
    Bitstring constants are stored by first writing an expression element

gdb/parser-defs.h

@@ -69,6 +69,22 @@ struct stoken
     int length;
   };
 
+struct typed_stoken
+  {
+    /* A language-specific type field.  */
+    int type;
+    /* Pointer to first byte of char-string or first bit of bit-string */
+    char *ptr;
+    /* Length of string in bytes for char-string or bits for bit-string */
+    int length;
+  };
+
+struct stoken_vector
+  {
+    int len;
+    struct typed_stoken *tokens;
+  };
+
 struct ttype
   {
     struct stoken stoken;
@@ -130,6 +146,8 @@ extern void write_exp_elt_intern (struct internalvar *);
 
 extern void write_exp_string (struct stoken);
 
+void write_exp_string_vector (int type, struct stoken_vector *vec);
+
 extern void write_exp_bitstring (struct stoken);
 
 extern void write_exp_elt_block (struct block *);

gdb/printcmd.c

@@ -48,6 +48,7 @@
 #include "solist.h"
 #include "solib.h"
 #include "parser-defs.h"
+#include "charset.h"
 
 #ifdef TUI
 #include "tui/tui.h"		/* For tui_active et.al.   */
@@ -276,10 +277,13 @@ print_formatted (struct value *val, int size,
       switch (options->format)
 	{
 	case 's':
-	  /* FIXME: Need to handle wchar_t's here... */
+	  {
-	  next_address = VALUE_ADDRESS (val)
+	    struct type *elttype = value_type (val);
-	    + val_print_string (VALUE_ADDRESS (val), -1, 1, stream,
+	    next_address = (VALUE_ADDRESS (val)
-				options);
+			    + val_print_string (elttype,
+						VALUE_ADDRESS (val), -1,
+						stream, options));
+	  }
 	  return;
 
 	case 'i':
@@ -374,7 +378,7 @@ print_scalar_formatted (const void *valaddr, struct type *type,
 	  print_hex_chars (stream, valaddr, len, byte_order);
 	  return;
 	case 'c':
-	  print_char_chars (stream, valaddr, len, byte_order);
+	  print_char_chars (stream, type, valaddr, len, byte_order);
 	  return;
 	default:
 	  break;
@@ -1958,7 +1962,8 @@ printf_command (char *arg, int from_tty)
 
     enum argclass
       {
-	int_arg, long_arg, long_long_arg, ptr_arg, string_arg,
+	int_arg, long_arg, long_long_arg, ptr_arg,
+	string_arg, wide_string_arg, wide_char_arg,
 	double_arg, long_double_arg, decfloat_arg
       };
     enum argclass *argclass;
@@ -2090,8 +2095,8 @@ printf_command (char *arg, int from_tty)
 	      break;
 
 	    case 'c':
-	      this_argclass = int_arg;
+	      this_argclass = lcount == 0 ? int_arg : wide_char_arg;
-	      if (lcount || seen_h || seen_big_l)
+	      if (lcount > 1 || seen_h || seen_big_l)
 		bad = 1;
 	      if (seen_prec || seen_zero || seen_space || seen_plus)
 		bad = 1;
@@ -2106,8 +2111,8 @@ printf_command (char *arg, int from_tty)
 	      break;
 
 	    case 's':
-	      this_argclass = string_arg;
+	      this_argclass = lcount == 0 ? string_arg : wide_string_arg;
-	      if (lcount || seen_h || seen_big_l)
+	      if (lcount > 1 || seen_h || seen_big_l)
 		bad = 1;
 	      if (seen_zero || seen_space || seen_plus)
 		bad = 1;
@@ -2159,6 +2164,15 @@ printf_command (char *arg, int from_tty)
 		last_arg[length_before_ll + lcount];
 	      current_substring += length_before_ll + 4;
 	    }
+	  else if (this_argclass == wide_string_arg
+		   || this_argclass == wide_char_arg)
+	    {
+	      /* Convert %ls or %lc to %s.  */
+	      int length_before_ls = f - last_arg - 2;
+	      strncpy (current_substring, last_arg, length_before_ls);
+	      strcpy (current_substring + length_before_ls, "s");
+	      current_substring += length_before_ls + 2;
+	    }
 	  else
 	    {
 	      strncpy (current_substring, last_arg, f - last_arg);
@@ -2223,6 +2237,76 @@ printf_command (char *arg, int from_tty)
 	      printf_filtered (current_substring, (char *) str);
 	    }
 	    break;
+	  case wide_string_arg:
+	    {
+	      gdb_byte *str;
+	      CORE_ADDR tem;
+	      int j;
+	      struct type *wctype = lookup_typename ("wchar_t", NULL, 0);
+	      int wcwidth = TYPE_LENGTH (wctype);
+	      gdb_byte *buf = alloca (wcwidth);
+	      struct obstack output;
+	      struct cleanup *inner_cleanup;
+
+	      tem = value_as_address (val_args[i]);
+
+	      /* This is a %s argument.  Find the length of the string.  */
+	      for (j = 0;; j += wcwidth)
+		{
+		  QUIT;
+		  read_memory (tem + j, buf, wcwidth);
+		  if (extract_unsigned_integer (buf, wcwidth) == 0)
+		    break;
+		}
+
+	      /* Copy the string contents into a string inside GDB.  */
+	      str = (gdb_byte *) alloca (j + wcwidth);
+	      if (j != 0)
+		read_memory (tem, str, j);
+	      memset (&str[j], 0, wcwidth);
+
+	      obstack_init (&output);
+	      inner_cleanup = make_cleanup_obstack_free (&output);
+
+	      convert_between_encodings (target_wide_charset (),
+					 host_charset (),
+					 str, j, wcwidth,
+					 &output, translit_char);
+	      obstack_grow_str0 (&output, "");
+
+	      printf_filtered (current_substring, obstack_base (&output));
+	      do_cleanups (inner_cleanup);
+	    }
+	    break;
+	  case wide_char_arg:
+	    {
+	      struct type *wctype = lookup_typename ("wchar_t", NULL, 0);
+	      struct type *valtype;
+	      struct obstack output;
+	      struct cleanup *inner_cleanup;
+	      const gdb_byte *bytes;
+
+	      valtype = value_type (val_args[i]);
+	      if (TYPE_LENGTH (valtype) != TYPE_LENGTH (wctype)
+		  || TYPE_CODE (valtype) != TYPE_CODE_INT)
+		error (_("expected wchar_t argument for %%lc"));
+
+	      bytes = value_contents (val_args[i]);
+
+	      obstack_init (&output);
+	      inner_cleanup = make_cleanup_obstack_free (&output);
+
+	      convert_between_encodings (target_wide_charset (),
+					 host_charset (),
+					 bytes, TYPE_LENGTH (valtype),
+					 TYPE_LENGTH (valtype),
+					 &output, translit_char);
+	      obstack_grow_str0 (&output, "");
+
+	      printf_filtered (current_substring, obstack_base (&output));
+	      do_cleanups (inner_cleanup);
+	    }
+	    break;
 	  case double_arg:
 	    {
 	      struct type *type = value_type (val_args[i]);

gdb/scm-lang.c

@@ -43,14 +43,14 @@ static int in_eval_c (void);
 struct type *builtin_type_scm;
 
 void
-scm_printchar (int c, struct ui_file *stream)
+scm_printchar (int c, struct type *type, struct ui_file *stream)
 {
   fprintf_filtered (stream, "#\\%c", c);
 }
 
 static void
-scm_printstr (struct ui_file *stream, const gdb_byte *string,
+scm_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string,
-	      unsigned int length, int width, int force_ellipses,
+	      unsigned int length, int force_ellipses,
 	      const struct value_print_options *options)
 {
   fprintf_filtered (stream, "\"%s\"", string);

gdb/scm-lang.h

@@ -59,7 +59,7 @@ extern void scm_scmval_print (LONGEST, struct ui_file *, int,
 
 extern int is_scmvalue_type (struct type *);
 
-extern void scm_printchar (int, struct ui_file *);
+extern void scm_printchar (int, struct type *, struct ui_file *);
 
 extern struct value *scm_evaluate_string (char *, int);
 

gdb/scm-valprint.c

@@ -187,7 +187,8 @@ taloop:
       if (SCM_ICHRP (svalue))
 	{
 	  svalue = SCM_ICHR (svalue);
-	  scm_printchar (svalue, stream);
+	  scm_printchar (svalue, builtin_type (current_gdbarch)->builtin_char,
+			 stream);
 	  break;
 	}
       else if (SCM_IFLAGP (svalue)

gdb/testsuite/ChangeLog

@@ -1,3 +1,28 @@
+2009-03-20  Tom Tromey  <tromey@redhat.com>
+
+	* 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.
+
 2009-03-19  Jerome Guitton  <guitton@adacore.com>
 
 	* gdb.ada/fixed_points/fixed_points.adb: Add a test on overprecise

gdb/testsuite/gdb.base/call-rt-st.exp

@@ -186,7 +186,7 @@ if {![gdb_skip_float_test "print print_two_floats(*f3)"] && \
 
 if ![gdb_skip_stdio_test "print print_bit_flags_char(*cflags)"] {
     print_struct_call "print_bit_flags_char(*cflags)" \
-            ".*alpha\[ \r\n\]+gamma\[ \r\n\]+epsilon\[ \r\n\]+.\[0-9\]+ = \\{alpha = 1 '\\\\001', beta = 0 '\\\\0', gamma = 1 '\\\\001', delta = 0 '\\\\0', epsilon = 1 '\\\\001', omega = 0 '\\\\0'\\}"
+            ".*alpha\[ \r\n\]+gamma\[ \r\n\]+epsilon\[ \r\n\]+.\[0-9\]+ = \\{alpha = 1 '\\\\1', beta = 0 '\\\\0', gamma = 1 '\\\\1', delta = 0 '\\\\0', epsilon = 1 '\\\\1', omega = 0 '\\\\0'\\}"
 }
 
 if ![gdb_skip_stdio_test "print print_bit_flags_short(*sflags)"] {

gdb/testsuite/gdb.base/callfuncs.exp

@@ -437,7 +437,7 @@ gdb_test "print t_small_values(1,3,5,7,9,11,13,15,17,19)" \
 	"The program being debugged stopped while.*" \
 	"stop at nested call level 4"
 gdb_test "backtrace" \
-	"\#0  t_small_values \\(arg1=1 '.001', arg2=3, arg3=5, arg4=7 '.a', arg5=9, arg6=11 '.v', arg7=13, arg8=15, arg9=17, arg10=19\\).*\#2  sum10 \\(i0=2, i1=4, i2=6, i3=8, i4=10, i5=12, i6=14, i7=16, i8=18, i9=20\\).*\#3  <function called from gdb>.*\#4  add \\(a=4, b=5\\).*\#5  <function called from gdb>.*\#6  add \\(a=2, b=3\\).*\#7  <function called from gdb>.*\#8  main.*" \
+	"\#0  t_small_values \\(arg1=1 '.1', arg2=3, arg3=5, arg4=7 '.a', arg5=9, arg6=11 '.v', arg7=13, arg8=15, arg9=17, arg10=19\\).*\#2  sum10 \\(i0=2, i1=4, i2=6, i3=8, i4=10, i5=12, i6=14, i7=16, i8=18, i9=20\\).*\#3  <function called from gdb>.*\#4  add \\(a=4, b=5\\).*\#5  <function called from gdb>.*\#6  add \\(a=2, b=3\\).*\#7  <function called from gdb>.*\#8  main.*" \
 	"backtrace at nested call level 4"
 gdb_test "finish" "Value returned is .* = 100" \
 	"Finish from nested call level 4"

gdb/testsuite/gdb.base/charset.c

@@ -20,11 +20,6 @@
    Please email any bugs, comments, and/or additions to this file to:
    bug-gdb@gnu.org  */
 
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-
-
 /* X_string is a null-terminated string in the X charset whose
    elements are as follows.  X should be the name the `set charset'
    command uses for the character set, in lower-case, with any
@@ -54,6 +49,21 @@ char iso_8859_1_string[NUM_CHARS];
 char ebcdic_us_string[NUM_CHARS];
 char ibm1047_string[NUM_CHARS];
 
+/* We make a phony wchar_t and then pretend that this platform uses
+   UCS-4 (or UCS-2, depending on the size -- same difference for the
+   purposes of this test).  */
+typedef unsigned int wchar_t;
+wchar_t ucs_4_string[NUM_CHARS];
+
+/* We also define a couple phony types for testing the u'' and U''
+   support.  It is ok if these have the wrong size on some platforms
+   -- the test case will skip the tests in that case.  */
+typedef unsigned short char16_t;
+typedef unsigned int char32_t;
+
+/* Make sure to use the typedefs.  */
+char16_t uvar;
+char32_t Uvar;
 
 void
 init_string (char string[],
@@ -62,7 +72,10 @@ init_string (char string[],
              char line_feed, char carriage_return, char horizontal_tab,
              char vertical_tab, char cent, char misc_ctrl)
 {
-  memset (string, x, NUM_CHARS);
+  int i;
+
+  for (i = 0; i < NUM_CHARS; ++i)
+    string[i] = x;
   string[0] = alert;
   string[1] = backspace;
   string[2] = form_feed;
@@ -85,13 +98,21 @@ fill_run (char string[], int start, int len, int first)
 }
 
 
+void
+init_ucs4 ()
+{
+  int i;
+
+  for (i = 0; i < NUM_CHARS; ++i)
+    ucs_4_string[i] = iso_8859_1_string[i] & 0xff;
+}
+
 int main ()
 {
 #ifdef usestubs
   set_debug_traps();
   breakpoint();
 #endif
-  (void) malloc (1);
   /* Initialize ascii_string.  */
   init_string (ascii_string,
                120,
@@ -146,5 +167,7 @@ int main ()
   /* The digits, at least, are contiguous.  */
   fill_run (ibm1047_string, 59, 10, 240);
 
-  puts ("All set!");            /* all strings initialized */
+  init_ucs4 ();
+
+  return 0;            /* all strings initialized */
 }

gdb/testsuite/gdb.base/charset.exp

@@ -47,13 +47,7 @@ proc parse_show_charset_output {testname} {
     global gdb_prompt
 
     gdb_expect {
-        -re "The current host and target character set is `(.*)'\\.\[\r\n\]+$gdb_prompt $" {
+        -re "The host character set is \"(.*)\"\\.\[\r\n\]+The target character set is \"(.*)\"\\.\[\r\n\]+The target wide character set is \"(.*)\"\\.\[\r\n\]+$gdb_prompt $" {
-            set host_charset $expect_out(1,string)
-            set target_charset $expect_out(1,string)
-	    set retlist [list $host_charset $target_charset]
-            pass $testname
-        }
-        -re "The current host character set is `(.*)'\\.\[\r\n\]+The current target character set is `(.*)'\\.\[\r\n\]+$gdb_prompt $" {
             set host_charset $expect_out(1,string)
             set target_charset $expect_out(2,string)
 	    set retlist [list $host_charset $target_charset]
@@ -81,76 +75,31 @@ proc parse_show_charset_output {testname} {
 }
 
 
-# Try the various `show charset' commands.  These are all aliases of each
+# Try the various `show charset' commands.
-# other; `show target-charset' and `show host-charset' actually print
-# both the host and target charsets.
 
 send_gdb "show charset\n"
 set show_charset [parse_show_charset_output "show charset"]
 
 send_gdb "show target-charset\n"
-set show_target_charset [parse_show_charset_output "show target-charset"]
+set show_target_charset \
+  [lindex [parse_show_charset_output "show target-charset"] 0]
 
-if {[lsearch $show_charset $show_target_charset] >= 0} {
+if {[lsearch -exact $show_charset $show_target_charset] >= 0} {
     pass "check `show target-charset' against `show charset'"
 } else {
     fail "check `show target-charset' against `show charset'"
 }
 
 send_gdb "show host-charset\n"
-set show_host_charset [parse_show_charset_output "show host-charset"]
+set show_host_charset \
+  [lindex [parse_show_charset_output "show host-charset"] 0]
 
-if {[lsearch $show_charset $show_host_charset] >= 0} {
+if {[lsearch -exact $show_charset $show_host_charset] >= 0} {
     pass "check `show host-charset' against `show charset'"
 } else {
     fail "check `show host-charset' against `show charset'"
 }
 
-
-# Get the list of supported (host) charsets as possible completions.
-send_gdb "set charset \t\t"
-
-# Check that we can at least use ASCII as a host character set.
-sleep 1
-gdb_expect {
-    -re "^set charset .*\r\nASCII.*\r\n$gdb_prompt set charset " {
-	# We got the output that we wanted, including ASCII as possible
-	# charset. Send a newline to get us back to the prompt. This will
-	# also generate an error message. Let's not check here that the error
-	# message makes sense, we do that below, as a separate testcase.
-        send_gdb "\n"
-        gdb_expect {
-	    -re ".*Requires an argument.*$gdb_prompt $" {
-		pass "get valid character sets"
-	    }
-	    -re ".*$gdb_prompt $" {
-		send_gdb "\n"
-		gdb_expect {
-		    -re ".*$gdb_prompt $" {
-			fail "get valid character sets"
-		    }
-		}
-	    }
-	    timeout {
-		fail "(timeout) get valid character sets"
-	    }
-	}
-    }
-    -re ".*$gdb_prompt $" {
-	# We got some output that ended with a regular prompt
-        fail "get valid character sets"
-    }
-    -re ".*$gdb_prompt set charset.*$" {
-	# We got some other output, send a cntrl-c to gdb to get us back
-        # to the prompt.
-	send_gdb "\003"
-        fail "get valid character sets"
-    }
-    timeout {
-        fail "get valid character sets (timeout)"
-    }
-}
-
 # Try a malformed `set charset'.
 gdb_test "set charset" \
          "Requires an argument. Valid arguments are.*" \
@@ -179,7 +128,7 @@ proc all_charset_names {} {
 
 proc valid_host_charset {charset} {
     global charsets
-    return $charsets($charset)
+    return [expr {[info exists charsets($charset)] && $charsets($charset)}]
 }
 
 send_gdb "set host-charset\n"
@@ -211,6 +160,13 @@ gdb_expect {
     }
 }
 
+# If gdb was built with a phony iconv, it will only have two character
+# sets: "auto" and the default.  In this situation, this set of tests
+# is pointless.
+if {[llength [array names charsets]] < 3} {
+    untested charset.exp
+    return -1
+}
 
 send_gdb "set target-charset\n"
 gdb_expect {
@@ -244,8 +200,10 @@ gdb_expect {
     }
 }
 
-# Make sure that GDB supports every host/target charset combination.
+# We don't want to test all the charset names here, since that would
-foreach host_charset [all_charset_names] {
+# be too many combinations.  We we pick a subset.
+set charset_subset {ASCII ISO-8859-1 EBCDIC-US IBM1047}
+foreach host_charset $charset_subset {
     if {[valid_host_charset $host_charset]} {
 
         set testname "try `set host-charset $host_charset'"
@@ -279,7 +237,7 @@ foreach host_charset [all_charset_names] {
 
         # Now try setting every possible target character set,
         # given that host charset.
-        foreach target_charset [all_charset_names] {
+        foreach target_charset $charset_subset {
             set testname "try `set target-charset $target_charset'"
             send_gdb "set target-charset $target_charset\n"
             gdb_expect {
@@ -404,23 +362,42 @@ gdb_expect {
 }
 
 
+# We only try the wide character tests on machines where the wchar_t
+# typedef in the test case has the right size.
+set wchar_size [get_sizeof wchar_t 99]
+set wchar_ok 0
+if {$wchar_size == 2} {
+    lappend charset_subset UCS-2
+    set wchar_ok 1
+} elseif {$wchar_size == 4} {
+    lappend charset_subset UCS-4
+    set wchar_ok 1
+}
+
 gdb_test "set host-charset ASCII" ""
-foreach target_charset [all_charset_names] {
+foreach target_charset $charset_subset {
-    send_gdb "set target-charset $target_charset\n" 
+    if {$target_charset == "UCS-4" || $target_charset == "UCS-2"} {
+	set param target-wide-charset
+	set L L
+    } else {
+	set param target-charset
+	set L ""
+    }
+    send_gdb "set $param $target_charset\n" 
     gdb_expect {
         -re "$gdb_prompt $" {
-            pass "set target-charset $target_charset"
+            pass "set $param $target_charset"
         }
         timeout {
-            fail "set target-charset $target_charset (timeout)"
+            fail "set $param $target_charset (timeout)"
         }
     }
 
     # Try printing the null character.  There seems to be a bug in
     # gdb_test that requires us to use gdb_expect here.
-    send_gdb "print '\\0'\n"
+    send_gdb "print $L'\\0'\n"
     gdb_expect {
-        -re "\\\$${decimal} = 0 '\\\\0'\[\r\n\]+$gdb_prompt $" {
+        -re "\\\$${decimal} = 0 $L'\\\\0'\[\r\n\]+$gdb_prompt $" {
             pass "print the null character in ${target_charset}"
         }
         -re "$gdb_prompt $" {
@@ -435,8 +412,14 @@ foreach target_charset [all_charset_names] {
     # a string in $target_charset.  The variable's name is the
     # character set's name, in lower-case, with all non-identifier
     # characters replaced with '_', with "_string" stuck on the end.
-    set var_name [string tolower "${target_charset}_string"]
+    if {$target_charset == "UCS-2"} {
-    regsub -all -- "\[^a-z0-9_\]" $var_name "_" var_name
+	# We still use the ucs_4_string variable -- but the size is
+	# correct for UCS-2.
+	set var_name ucs_4_string
+    } else {
+	set var_name [string tolower "${target_charset}_string"]
+	regsub -all -- "\[^a-z0-9_\]" $var_name "_" var_name
+    }
     
     # Compute a regexp matching the results we expect.  This is static,
     # but it's easier than writing it out.
@@ -444,12 +427,12 @@ foreach target_charset [all_charset_names] {
     set uppercase "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
     set lowercase "abcdefghijklmnopqrstuvwxyz"
     set digits "0123456789"
-    set octal_escape "\\\\\[0-9\]\[0-9\]\[0-9\]"
+    set octal_escape "\\\\\[0-9\]+"
 
     send_gdb "print $var_name\n"
     # ${escapes}${uppercase}${lowercase}${digits}${octal}${octal}
     gdb_expect {
-        -re ".* = \"(\\\\a|x)(\\\\b|x)(\\\\f|x)(\\\\n|x)(\\\\r|x)(\\\\t|x)(\\\\v|x)${uppercase}${lowercase}${digits}(\\\\\[0-9\]\[0-9\]\[0-9\]|x)(\\\\\[0-9\]\[0-9\]\[0-9\]|x).*\"\[\r\n\]+$gdb_prompt $" {
+        -re ".* = $L\"(\\\\a|x)(\\\\b|x)(\\\\f|x)(\\\\n|x)(\\\\r|x)(\\\\t|x)(\\\\v|x)${uppercase}${lowercase}${digits}(${octal_escape}|x)+\"\[\r\n\]+$gdb_prompt $" {
             pass "print string in $target_charset"
         }
         -re "$gdb_prompt $" {
@@ -461,22 +444,22 @@ foreach target_charset [all_charset_names] {
     }
 
     # Try entering a character literal, and see if it comes back unchanged.
-    gdb_test "print 'A'" \
+    gdb_test "print $L'A'" \
-             " = \[0-9-\]+ 'A'" \
+             " = \[0-9-\]+ $L'A'" \
              "parse character literal in ${target_charset}"
 
     # Check that the character literal was encoded correctly.
-    gdb_test "print 'A' == $var_name\[7\]" \
+    gdb_test "print $L'A' == $var_name\[7\]" \
              " = 1" \
              "check value of parsed character literal in ${target_charset}"
 
     # Try entering a string literal, and see if it comes back unchanged.
-    gdb_test "print \"abcdefABCDEF012345\"" \
+    gdb_test "print $L\"abcdefABCDEF012345\"" \
-             " = \"abcdefABCDEF012345\"" \
+             " = $L\"abcdefABCDEF012345\"" \
              "parse string literal in ${target_charset}"
 
     # Check that the string literal was encoded correctly.
-    gdb_test "print \"q\"\[0\] == $var_name\[49\]" \
+    gdb_test "print $L\"q\"\[0\] == $var_name\[49\]" \
              " = 1" \
              "check value of parsed string literal in ${target_charset}"
 
@@ -509,7 +492,7 @@ foreach target_charset [all_charset_names] {
         send_gdb "print $var_name\[$i\]\n"
         set have_escape 1
         gdb_expect {
-            -re "= \[0-9-\]+ '\\\\${escape}'\[\r\n\]+$gdb_prompt $" {
+            -re "= \[0-9-\]+ $L'\\\\${escape}'\[\r\n\]+$gdb_prompt $" {
                 pass "try printing '\\${escape}' in ${target_charset}"
             }
             -re "= \[0-9-\]+ 'x'\[\r\n\]+$gdb_prompt $" {
@@ -527,12 +510,12 @@ foreach target_charset [all_charset_names] {
         if {$have_escape} {
 
             # Try parsing a backslash escape in a character literal.
-            gdb_test "print '\\${escape}' == $var_name\[$i\]" \
+            gdb_test "print $L'\\${escape}' == $var_name\[$i\]" \
                      " = 1" \
                      "check value of '\\${escape}' in ${target_charset}"
 
             # Try parsing a backslash escape in a string literal.
-            gdb_test "print \"\\${escape}\"\[0\] == $var_name\[$i\]" \
+            gdb_test "print $L\"\\${escape}\"\[0\] == $var_name\[$i\]" \
                      " = 1" \
                      "check value of \"\\${escape}\" in ${target_charset}"
         }
@@ -540,10 +523,73 @@ foreach target_charset [all_charset_names] {
 
     # Try printing a character escape that doesn't exist.  We should 
     # get the unescaped character, in the target character set.
-    gdb_test "print '\\q'" " = \[0-9-\]+ 'q'" \
+    gdb_test "print $L'\\q'" " = \[0-9-\]+ $L'q'" \
              "print escape that doesn't exist in $target_charset"
-    gdb_test "print '\\q' == $var_name\[49\]" " = 1" \
+    gdb_test "print $L'\\q' == $var_name\[49\]" " = 1" \
              "check value of escape that doesn't exist in $target_charset"
 }
 
+# Reset the target charset.
+gdb_test "set target-charset UTF-8" ""
+
+# \242 is not a valid UTF-8 character.
+gdb_test "print \"\\242\"" " = \"\\\\242\"" \
+  "non-representable target character"
+
+gdb_test "print '\\x'" "\\\\x escape without a following hex digit."
+gdb_test "print '\\u'" "\\\\u escape without a following hex digit."
+gdb_test "print '\\9'" " = \[0-9\]+ '9'"
+
+# Tests for wide- or unicode- strings.  L is the prefix letter to use,
+# either "L" (for wide strings), "u" (for UCS-2), or "U" (for UCS-4).
+# NAME is used in the test names and should be related to the prefix
+# letter in some easy-to-undestand way.
+proc test_wide_or_unicode {L name} {
+    gdb_test "print $L\"ab\" $L\"c\"" " = $L\"abc\"" \
+      "basic $name string concatenation"
+    gdb_test "print $L\"ab\" \"c\"" " = $L\"abc\"" \
+      "narrow and $name string concatenation"
+    gdb_test "print \"ab\" $L\"c\"" " = $L\"abc\"" \
+      "$name and narrow string concatenation"
+    gdb_test "print $L\"\\xe\" $L\"c\"" " = $L\"\\\\16c\"" \
+      "$name string concatenation with escape"
+    gdb_test "print $L\"\" \"abcdef\" \"g\"" \
+      "$L\"abcdefg\"" \
+      "concatenate three strings with empty $name string"
+
+    gdb_test "print $L'a'" "= \[0-9\]+ $L'a'" \
+      "basic $name character"
+}
+
+if {$wchar_ok} {
+    test_wide_or_unicode L wide
+}
+
+set ucs2_ok [expr {[get_sizeof char16_t 99] == 2}]
+if {$ucs2_ok} {
+    test_wide_or_unicode u UCS-2
+}
+
+set ucs4_ok [expr {[get_sizeof char32_t 99] == 4}]
+if {$ucs4_ok} {
+    test_wide_or_unicode U UCS-4
+}
+
+# Test an invalid string combination.
+proc test_combination {L1 name1 L2 name2} {
+    gdb_test "print $L1\"abc\" $L2\"def\"" \
+      "Undefined string concatenation." \
+      "undefined concatenation of $name1 and $name2"
+}
+
+if {$wchar_ok && $ucs2_ok} {
+    test_combination L wide u UCS-2
+}
+if {$wchar_ok && $ucs4_ok} {
+    test_combination L wide U UCS-4
+}
+if {$ucs2_ok && $ucs4_ok} {
+    test_combination u UCS-2 U UCS-4
+}
+
 gdb_exit 

gdb/testsuite/gdb.base/constvars.exp

@@ -161,7 +161,7 @@ proc do_constvar_tests {} {
     gdb_test "print laconic" " = 65 'A'"
     local_compiler_xfail_check
     gdb_test "ptype laconic" "type = const char"
-    gdb_test "print laggard" " = 1 '.001'"
+    gdb_test "print laggard" " = 1 '.1'"
     local_compiler_xfail_check
     gdb_test "ptype laggard" "type = const unsigned char"
     gdb_test "print lagoon" " = 2"
@@ -209,7 +209,7 @@ proc do_constvar_tests {} {
     gdb_test "print *lewd" " = 65 'A'"
     local_compiler_xfail_check
     gdb_test "ptype lewd" "type = const char \\* const"
-    gdb_test "print *lexicographer" " = 1 '.001'"
+    gdb_test "print *lexicographer" " = 1 '.1'"
     local_compiler_xfail_check
     gdb_test "ptype lexicographer" "type = const unsigned char \\* const"
     gdb_test "print *lexicon" " = 2"
@@ -233,7 +233,7 @@ proc do_constvar_tests {} {
     gdb_test "print *languish" " = 65 'A'"
     local_compiler_xfail_check
     gdb_test "ptype languish" "type = const char \\*"
-    gdb_test "print *languor" " = 1 '.001'"
+    gdb_test "print *languor" " = 1 '.1'"
     local_compiler_xfail_check
     gdb_test "ptype languor" "type = const unsigned char \\*"
     gdb_test "print *lank" " = 2"

gdb/testsuite/gdb.base/long_long.exp

@@ -210,7 +210,7 @@ gdb_test_char "p/o *(char *)c" "01"
 gdb_test_char "p/t *(char *)c" "1"
 gdb_test_char "p/a *(char *)c" "0x1( <.*>)?"
 gdb_test_char "p/f *(char *)c" "1"
-gdb_test_char "p/c *(char *)c" "1 '.001'"
+gdb_test_char "p/c *(char *)c" "1 '.1'"
 
 gdb_test_short "p/x *(short *)s" "" "0x123" ""
 gdb_test_short "p/d *(short *)s" "" "291" ""
@@ -257,7 +257,7 @@ gdb_test "x/u w" "19088743"
 gdb_test "x/o w" "0110642547"
 gdb_test "x/t w" "00000001001000110100010101100111"
 gdb_test_xptr "x/a" { b "" } { h "" } { w "0x1234567" } { g "0x123456789abcdef" }
-gdb_test "x/c b" "1 '.001'"
+gdb_test "x/c b" "1 '.1'"
 if { $sizeof_double == 8 || $sizeof_long_double == 8 } {
     gdb_test "x/f &val.oct" "-5.9822653797615723e-120"
 } else {
@@ -273,7 +273,7 @@ gdb_test "x/2u g" "81985529216486895.*12046818088235383159"
 gdb_test "x/2o g" "04432126361152746757.*01234567123456701234567"
 gdb_test "x/2t g" "0000000100100011010001010110011110001001101010111100110111101111.*1010011100101110111001010011100101110111000001010011100101110111"
 gdb_test_xptr "x/2a" { b "" } { h "" } { w "0x1234567.*0xa72ee539" } { g "0x123456789abcdef.*0xa72ee53977053977" }
-gdb_test "x/2c b" "1 '.001'.*-89 '.'"
+gdb_test "x/2c b" "1 '.1'.*-89 '.\[0-9\]*'"
 if { $sizeof_double == 8 || $sizeof_long_double == 8 } {
     gdb_test "x/2f &val.oct" "-5.9822653797615723e-120.*-5.9041889495880968e-100"
 } else {
@@ -288,7 +288,7 @@ gdb_test "x/2bu b" "1.*167"
 gdb_test "x/2bo b" "01.*0247"
 gdb_test "x/2bt b" "00000001.*10100111"
 gdb_test_ptr "x/2ba b" "" "" "0x1.*0xffffffa7" "0x1.*0xffffffffffffffa7"
-gdb_test "x/2bc b" "1 '.001'.*-89 '.'"
+gdb_test "x/2bc b" "1 '.1'.*-89 '.\[0-9\]*'"
 gdb_test "x/2bf b" "1.*-89"
 
 gdb_test "x/2hx h" "0x0123.*0xa72e"
@@ -315,7 +315,7 @@ gdb_test "x/2gu g" "81985529216486895.*12046818088235383159"
 gdb_test "x/2go g" "04432126361152746757.*01234567123456701234567"
 gdb_test "x/2gt g" "0000000100100011010001010110011110001001101010111100110111101111.*1010011100101110111001010011100101110111000001010011100101110111"
 gdb_test_ptr "x/2ga g" "" "" "0x89abcdef.*0x77053977" "0x123456789abcdef.*0xa72ee53977053977"
-gdb_test "x/2gc g" "-17 '.'.*119 'w'"
+gdb_test "x/2gc g" "-17 '.\[0-9\]*'.*119 'w'"
 gdb_test "x/2gf g" "3.5127005640885037e-303.*-5.9822653797615723e-120"
 
 gdb_exit

gdb/testsuite/gdb.base/pointers.exp

@@ -389,7 +389,7 @@ gdb_expect {
 
 send_gdb "print *pUC\n"
 gdb_expect {
-    -re ".\[0-9\]* = 21 \'.025\'.*$gdb_prompt $" {
+    -re ".\[0-9\]* = 21 \'.25\'.*$gdb_prompt $" {
         pass "print value of *pUC"
       }
     -re ".*$gdb_prompt $" { fail "print value of *pUC" }

gdb/testsuite/gdb.base/printcmds.exp

@@ -137,12 +137,12 @@ proc test_print_all_chars {} {
     global gdb_prompt
 
     gdb_test "p ctable1\[0\]"   " = 0 '\\\\0'"
-    gdb_test "p ctable1\[1\]"   " = 1 '\\\\001'"
+    gdb_test "p ctable1\[1\]"   " = 1 '\\\\1'"
-    gdb_test "p ctable1\[2\]"   " = 2 '\\\\002'"
+    gdb_test "p ctable1\[2\]"   " = 2 '\\\\2'"
-    gdb_test "p ctable1\[3\]"   " = 3 '\\\\003'"
+    gdb_test "p ctable1\[3\]"   " = 3 '\\\\3'"
-    gdb_test "p ctable1\[4\]"   " = 4 '\\\\004'"
+    gdb_test "p ctable1\[4\]"   " = 4 '\\\\4'"
-    gdb_test "p ctable1\[5\]"   " = 5 '\\\\005'"
+    gdb_test "p ctable1\[5\]"   " = 5 '\\\\5'"
-    gdb_test "p ctable1\[6\]"   " = 6 '\\\\006'"
+    gdb_test "p ctable1\[6\]"   " = 6 '\\\\6'"
     gdb_test "p ctable1\[7\]"   " = 7 '\\\\a'"
     gdb_test "p ctable1\[8\]"   " = 8 '\\\\b'"
     gdb_test "p ctable1\[9\]"   " = 9 '\\\\t'"
@@ -150,24 +150,24 @@ proc test_print_all_chars {} {
     gdb_test "p ctable1\[11\]"  " = 11 '\\\\v'"
     gdb_test "p ctable1\[12\]"  " = 12 '\\\\f'"
     gdb_test "p ctable1\[13\]"  " = 13 '\\\\r'"
-    gdb_test "p ctable1\[14\]"  " = 14 '\\\\016'"
+    gdb_test "p ctable1\[14\]"  " = 14 '\\\\16'"
-    gdb_test "p ctable1\[15\]"  " = 15 '\\\\017'"
+    gdb_test "p ctable1\[15\]"  " = 15 '\\\\17'"
-    gdb_test "p ctable1\[16\]"  " = 16 '\\\\020'"
+    gdb_test "p ctable1\[16\]"  " = 16 '\\\\20'"
-    gdb_test "p ctable1\[17\]"  " = 17 '\\\\021'"
+    gdb_test "p ctable1\[17\]"  " = 17 '\\\\21'"
-    gdb_test "p ctable1\[18\]"  " = 18 '\\\\022'"
+    gdb_test "p ctable1\[18\]"  " = 18 '\\\\22'"
-    gdb_test "p ctable1\[19\]"  " = 19 '\\\\023'"
+    gdb_test "p ctable1\[19\]"  " = 19 '\\\\23'"
-    gdb_test "p ctable1\[20\]"  " = 20 '\\\\024'"
+    gdb_test "p ctable1\[20\]"  " = 20 '\\\\24'"
-    gdb_test "p ctable1\[21\]"  " = 21 '\\\\025'"
+    gdb_test "p ctable1\[21\]"  " = 21 '\\\\25'"
-    gdb_test "p ctable1\[22\]"  " = 22 '\\\\026'"
+    gdb_test "p ctable1\[22\]"  " = 22 '\\\\26'"
-    gdb_test "p ctable1\[23\]"  " = 23 '\\\\027'"
+    gdb_test "p ctable1\[23\]"  " = 23 '\\\\27'"
-    gdb_test "p ctable1\[24\]"  " = 24 '\\\\030'"
+    gdb_test "p ctable1\[24\]"  " = 24 '\\\\30'"
-    gdb_test "p ctable1\[25\]"  " = 25 '\\\\031'"
+    gdb_test "p ctable1\[25\]"  " = 25 '\\\\31'"
-    gdb_test "p ctable1\[26\]"  " = 26 '\\\\032'"
+    gdb_test "p ctable1\[26\]"  " = 26 '\\\\32'"
-    gdb_test "p ctable1\[27\]"  " = 27 '\\\\033'"
+    gdb_test "p ctable1\[27\]"  " = 27 '\\\\33'"
-    gdb_test "p ctable1\[28\]"  " = 28 '\\\\034'"
+    gdb_test "p ctable1\[28\]"  " = 28 '\\\\34'"
-    gdb_test "p ctable1\[29\]"  " = 29 '\\\\035'"
+    gdb_test "p ctable1\[29\]"  " = 29 '\\\\35'"
-    gdb_test "p ctable1\[30\]"  " = 30 '\\\\036'"
+    gdb_test "p ctable1\[30\]"  " = 30 '\\\\36'"
-    gdb_test "p ctable1\[31\]"  " = 31 '\\\\037'"
+    gdb_test "p ctable1\[31\]"  " = 31 '\\\\37'"
     gdb_test "p ctable1\[32\]"  " = 32 ' '"
     gdb_test "p ctable1\[33\]"  " = 33 '!'"
     gdb_test "p ctable1\[34\]"  " = 34 '\"'"
@@ -475,13 +475,13 @@ proc test_print_strings {} {
     gdb_test "p &ctable1\[0\]" \
 	" = \\(unsigned char \\*\\) \"\""
     gdb_test "p &ctable1\[1\]" \
-	" = \\(unsigned char \\*\\) \"\\\\001\\\\002\\\\003\\\\004\\\\005\\\\006\\\\a\\\\b\"..."
+	" = \\(unsigned char \\*\\) \"\\\\1\\\\2\\\\3\\\\4\\\\5\\\\6\\\\a\\\\b\"..."
     gdb_test "p &ctable1\[1*8\]" \
-	" = \\(unsigned char \\*\\) \"\\\\b\\\\t\\\\n\\\\v\\\\f\\\\r\\\\016\\\\017\"..."
+	" = \\(unsigned char \\*\\) \"\\\\b\\\\t\\\\n\\\\v\\\\f\\\\r\\\\16\\\\17\"..."
     gdb_test "p &ctable1\[2*8\]" \
-	" = \\(unsigned char \\*\\) \"\\\\020\\\\021\\\\022\\\\023\\\\024\\\\025\\\\026\\\\027\"..."
+	" = \\(unsigned char \\*\\) \"\\\\20\\\\21\\\\22\\\\23\\\\24\\\\25\\\\26\\\\27\"..."
     gdb_test "p &ctable1\[3*8\]" \
-	" = \\(unsigned char \\*\\) \"\\\\030\\\\031\\\\032\\\\033\\\\034\\\\035\\\\036\\\\037\"..."
+	" = \\(unsigned char \\*\\) \"\\\\30\\\\31\\\\32\\\\33\\\\34\\\\35\\\\36\\\\37\"..."
     gdb_test "p &ctable1\[4*8\]" \
 	" = \\(unsigned char \\*\\) \" !\\\\\"#\\\$%&'\"..."
     gdb_test "p &ctable1\[5*8\]" \
@@ -622,7 +622,7 @@ proc test_print_string_constants {} {
     set timeout 60;
 
     gdb_test "p \"a string\""		" = \"a string\""
-    gdb_test "p \"embedded \\000 null\"" " = \"embedded \\\\000 null\""
+    gdb_test "p \"embedded \\000 null\"" " = \"embedded \\\\0 null\""
     gdb_test "p \"abcd\"\[2\]"	" = 99 'c'"
     gdb_test "p sizeof (\"abcdef\")"	" = 7"
     gdb_test "ptype \"foo\""		" = char \\\[4\\\]"

gdb/testsuite/gdb.base/setvar.exp

@@ -121,7 +121,7 @@ proc test_set { args } {
 #    
 
 test_set "set variable v_char=0" "print v_char" ".\[0-9\]* = 0 \'.0\'"        "set variable char=0" 
-test_set "set variable v_char=1" "print v_char" ".\[0-9\]* = 1 \'.001\'"        "set variable char=1" 
+test_set "set variable v_char=1" "print v_char" ".\[0-9\]* = 1 \'.1\'"        "set variable char=1" 
 test_set "set variable v_char=7" "print v_char" ".\[0-9\]* = 7 \'.a\'"        "set variable char=7 (Bel)" 
 test_set "set variable v_char=32" "print v_char" ".\[0-9\]* = 32 \' \'"        "set variable char=32 (SPC)" 
 test_set "set variable v_char=65" "print v_char" ".\[0-9\]* = 65 \'A\'"        "set variable char=65 ('A')" 
@@ -132,7 +132,7 @@ test_set "set variable v_char=127" "print v_char" ".\[0-9\]* = 127 \'.177\'"
 # test "set variable" for type "signed char"
 #    
 test_set "set variable v_char=0" "print v_signed_char" ".\[0-9\]* = 0 \'.0\'"                 "set variable signed char=0" 
-test_set "set variable v_signed_char=1" "print v_signed_char" ".\[0-9\]* = 1 \'.001\'"        "set variable signed char=1" 
+test_set "set variable v_signed_char=1" "print v_signed_char" ".\[0-9\]* = 1 \'.1\'"        "set variable signed char=1" 
 test_set "set variable v_signed_char=7" "print v_signed_char" ".\[0-9\]* = 7 \'.a\'"        "set variable signed char=7 (Bel)" 
 test_set "set variable v_signed_char=32" "print v_signed_char" ".\[0-9\]* = 32 \' \'"        "set variable signed char=32 (SPC)" 
 test_set "set variable v_signed_char=65" "print v_signed_char" ".\[0-9\]* = 65 \'A\'"        "set variable signed char=65 ('A')" 
@@ -151,7 +151,7 @@ gdb_test "print v_signed_char" ".\[0-9\]* = -1 \'.377\'" \
 # test "set variable" for type "unsigned char"
 #
 test_set "set variable v_unsigned_char=0" "print v_unsigned_char" ".\[0-9\]* = 0 \'.0\'"        "set variable unsigned char=0" 
-test_set "set variable v_unsigned_char=1" "print v_unsigned_char" ".\[0-9\]* = 1 \'.001\'"        "set variable unsigned char=1" 
+test_set "set variable v_unsigned_char=1" "print v_unsigned_char" ".\[0-9\]* = 1 \'.1\'"        "set variable unsigned char=1" 
 test_set "set variable v_unsigned_char=7" "print v_unsigned_char" ".\[0-9\]* = 7 \'.a\'"        "set variable unsigned char=7 (Bel)" 
 test_set "set variable v_unsigned_char=32" "print v_unsigned_char" ".\[0-9\]* = 32 \' \'"        "set variable unsigned char=32 (SPC)" 
 test_set "set variable v_unsigned_char=65" "print v_unsigned_char" ".\[0-9\]* = 65 \'A\'"        "set variable unsigned char=65 ('A')" 

gdb/testsuite/gdb.base/store.exp

@@ -74,7 +74,7 @@ proc check_set { t l r new add } {
 	"${prefix}; print incremented l, expecting ${add}"
 }
 
-check_set "charest" "-1 .*" "-2 .*" "4 ..004." "2 ..002."
+check_set "charest" "-1 .*" "-2 .*" "4 ..4." "2 ..2."
 check_set "short" "-1" "-2" "4" "2"
 check_set "int" "-1" "-2" "4" "2"
 check_set "long" "-1" "-2" "4" "2"
@@ -102,7 +102,7 @@ proc up_set { t l r new } {
 	"${prefix}; print new l, expecting ${new}"
 }
 
-up_set "charest" "-1 .*" "-2 .*" "4 ..004."
+up_set "charest" "-1 .*" "-2 .*" "4 ..4."
 up_set "short" "-1" "-2" "4"
 up_set "int" "-1" "-2" "4"
 up_set "long" "-1" "-2" "4"

gdb/testsuite/gdb.cp/ref-types.exp

@@ -284,7 +284,7 @@ gdb_expect {
 
 send_gdb "print UC\n"
 gdb_expect {
-    -re ".\[0-9\]* = 21 '\.025'\.*$gdb_prompt $" {
+    -re ".\[0-9\]* = 21 '\.25'\.*$gdb_prompt $" {
         pass "print value of UC"
       }
     -re ".*$gdb_prompt $" { fail "print value of UC" }
@@ -557,7 +557,7 @@ gdb_expect {
 
 send_gdb "print rUC\n"
 gdb_expect {
-    -re ".\[0-9\]* = \\(unsigned char &\\) @$hex: 21 \'.025\'.*$gdb_prompt $" {
+    -re ".\[0-9\]* = \\(unsigned char &\\) @$hex: 21 \'.25\'.*$gdb_prompt $" {
         pass "print value of rUC"
       }
     -re ".*$gdb_prompt $" { fail "print value of rUC" }

gdb/testsuite/lib/gdb.exp

@@ -1162,9 +1162,12 @@ proc default_gdb_start { } {
     global gdb_prompt
     global timeout
     global gdb_spawn_id;
+    global env
 
     gdb_stop_suppressing_tests;
 
+    set env(LC_CTYPE) C
+
     verbose "Spawning $GDB $INTERNAL_GDBFLAGS $GDBFLAGS"
 
     if [info exists gdb_spawn_id] {

gdb/typeprint.c

@@ -236,7 +236,7 @@ print_type_scalar (struct type *type, LONGEST val, struct ui_file *stream)
       break;
 
     case TYPE_CODE_CHAR:
-      LA_PRINT_CHAR ((unsigned char) val, stream);
+      LA_PRINT_CHAR ((unsigned char) val, type, stream);
       break;
 
     case TYPE_CODE_BOOL:

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
+/* A helper for parse_escape that converts a host character to a
-   \^mumble sequence in a string or character constant.  START and END
+   target character.  C is the host character.  If conversion is
-   indicate a substring of some larger string that contains the
+   possible, then the target character is stored in *TARGET_C and the
-   erroneous backslash sequence, missing the initial backslash.  */
+   function returns 1.  Otherwise, the function returns 0.  */
-static NORETURN int
+
-no_control_char_error (const char *start, const char *end)
+static int
+host_char_to_target (int c, int *target_c)
 {
-  int len = end - start;
+  struct obstack host_data;
-  char *copy = alloca (end - start + 1);
+  char the_char = c;
+  struct cleanup *cleanups;
+  int result = 0;
 
-  memcpy (copy, start, len);
+  obstack_init (&host_data);
-  copy[len] = '\0';
+  cleanups = make_cleanup_obstack_free (&host_data);
 
-  error (_("There is no control character `\\%s' in the `%s' character set."),
+  convert_between_encodings (target_charset (), host_charset (),
-	 copy, target_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);
+    }
+
+  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))
+  switch (c)
-    return target_char;
+    {
-  else
-    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))
+    case 'a':
-	  error
+      c = '\a';
-	    ("The escape sequence `\%c' is equivalent to plain `%c', which"
+      break;
-	     " has no equivalent\n" "in the `%s' character set.", c, c,
+    case 'b':
-	     target_charset ());
+      c = '\b';
-	return target_char;
+      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

gdb/valops.c

@@ -1352,6 +1352,24 @@ value_array (int lowbound, int highbound, struct value **elemvec)
   return val;
 }
 
+struct value *
+value_typed_string (char *ptr, int len, struct type *char_type)
+{
+  struct value *val;
+  int lowbound = current_language->string_lower_bound;
+  int highbound = len / TYPE_LENGTH (char_type);
+  struct type *rangetype = create_range_type ((struct type *) NULL,
+					      builtin_type_int32,
+					      lowbound, 
+					      highbound + lowbound - 1);
+  struct type *stringtype
+    = create_array_type ((struct type *) NULL, char_type, rangetype);
+
+  val = allocate_value (stringtype);
+  memcpy (value_contents_raw (val), ptr, len);
+  return val;
+}
+
 /* Create a value for a string constant by allocating space in the
    inferior, copying the data into that space, and returning the
    address with type TYPE_CODE_STRING.  PTR points to the string

gdb/valprint.c

@@ -919,7 +919,8 @@ print_hex_chars (struct ui_file *stream, const gdb_byte *valaddr,
    Omit any leading zero chars.  */
 
 void
-print_char_chars (struct ui_file *stream, const gdb_byte *valaddr,
+print_char_chars (struct ui_file *stream, struct type *type,
+		  const gdb_byte *valaddr,
 		  unsigned len, enum bfd_endian byte_order)
 {
   const gdb_byte *p;
@@ -932,7 +933,7 @@ print_char_chars (struct ui_file *stream, const gdb_byte *valaddr,
 
       while (p < valaddr + len)
 	{
-	  LA_EMIT_CHAR (*p, stream, '\'');
+	  LA_EMIT_CHAR (*p, type, stream, '\'');
 	  ++p;
 	}
     }
@@ -944,7 +945,7 @@ print_char_chars (struct ui_file *stream, const gdb_byte *valaddr,
 
       while (p >= valaddr)
 	{
-	  LA_EMIT_CHAR (*p, stream, '\'');
+	  LA_EMIT_CHAR (*p, type, stream, '\'');
 	  --p;
 	}
     }
@@ -1315,7 +1316,8 @@ read_string (CORE_ADDR addr, int len, int width, unsigned int fetchlimit,
    whichever is smaller.  */
 
 int
-val_print_string (CORE_ADDR addr, int len, int width, struct ui_file *stream,
+val_print_string (struct type *elttype, CORE_ADDR addr, int len,
+		  struct ui_file *stream,
 		  const struct value_print_options *options)
 {
   int force_ellipsis = 0;	/* Force ellipsis to be printed if nonzero.  */
@@ -1325,6 +1327,7 @@ val_print_string (CORE_ADDR addr, int len, int width, struct ui_file *stream,
   int bytes_read;
   gdb_byte *buffer = NULL;	/* Dynamically growable fetch buffer.  */
   struct cleanup *old_chain = NULL;	/* Top of the old cleanup chain.  */
+  int width = TYPE_LENGTH (elttype);
 
   /* First we need to figure out the limit on the number of characters we are
      going to attempt to fetch and print.  This is actually pretty simple.  If
@@ -1378,7 +1381,7 @@ val_print_string (CORE_ADDR addr, int len, int width, struct ui_file *stream,
 	{
 	  fputs_filtered (" ", stream);
 	}
-      LA_PRINT_STRING (stream, buffer, bytes_read / width, width, force_ellipsis, options);
+      LA_PRINT_STRING (stream, elttype, buffer, bytes_read / width, force_ellipsis, options);
     }
 
   if (errcode != 0)

gdb/valprint.h

@@ -134,9 +134,10 @@ extern void print_decimal_chars (struct ui_file *, const gdb_byte *,
 extern void print_hex_chars (struct ui_file *, const gdb_byte *,
 			     unsigned int, enum bfd_endian);
 
-extern void print_char_chars (struct ui_file *, const gdb_byte *,
+extern void print_char_chars (struct ui_file *, struct type *,
-			      unsigned int, enum bfd_endian);
+			      const gdb_byte *, unsigned int, enum bfd_endian);
 
 int read_string (CORE_ADDR addr, int len, int width, unsigned int fetchlimit,
 		 gdb_byte **buffer, int *bytes_read);
+
 #endif

gdb/value.h

@@ -388,6 +388,8 @@ extern struct value *value_mark (void);
 
 extern void value_free_to_mark (struct value *mark);
 
+extern struct value *value_typed_string (char *ptr, int len,
+					 struct type *char_type);
 extern struct value *value_string (char *ptr, int len);
 extern struct value *value_bitstring (char *ptr, int len);
 
@@ -619,7 +621,7 @@ extern int common_val_print (struct value *val,
 			     const struct value_print_options *options,
 			     const struct language_defn *language);
 
-extern int val_print_string (CORE_ADDR addr, int len, int width,
+extern int val_print_string (struct type *elttype, CORE_ADDR addr, int len,
 			     struct ui_file *stream,
 			     const struct value_print_options *options);