From 6c7a06a3facfa3f71606655871d0617d3cad80d9 Mon Sep 17 00:00:00 2001 From: Tom Tromey Date: Fri, 20 Mar 2009 23:04:40 +0000 Subject: gdb: 2009-03-19 Tom Tromey Julian Brown 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) : 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) : Add type argument. : 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) : Rewrite. (tempbuf): Now global. (tempbuf_init): New global. (parse_string_or_char): New function. (yylex) : Now global. : 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/c-valprint.c | 83 +++++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 58 insertions(+), 25 deletions(-) (limited to 'gdb/c-valprint.c') diff --git a/gdb/c-valprint.c b/gdb/c-valprint.c index 0b616f7..cfdf2c0 100644 --- a/gdb/c-valprint.c +++ b/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)); - if (TYPE_LENGTH (type) > 0 && TYPE_LENGTH (TYPE_TARGET_TYPE (type)) > 0) + unresolved_elttype = TYPE_TARGET_TYPE (type); + 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 < options->print_max; - temp_len++); + (temp_len < len + && temp_len < options->print_max + && 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 *). */ - if (TYPE_CODE (type) == TYPE_CODE_PTR - && TYPE_NAME (type) == NULL - && TYPE_NAME (TYPE_TARGET_TYPE (type)) != NULL - && strcmp (TYPE_NAME (TYPE_TARGET_TYPE (type)), "char") == 0) + are always exactly (char *), (wchar_t *), or the like. */ + if (TYPE_CODE (val_type) == TYPE_CODE_PTR + && TYPE_NAME (val_type) == NULL + && TYPE_NAME (TYPE_TARGET_TYPE (val_type)) != NULL + && (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); -- cgit v1.1