From 732f6a935c88164b258505f7b10091bafa6bf877 Mon Sep 17 00:00:00 2001 From: Tom Tromey Date: Wed, 15 Apr 2009 22:20:32 +0000 Subject: [PATCH] gdb * c-lang.c (c_emit_char): Use INTERMEDIATE_ENCODING. (c_printstr): Likewise. * charset.c: Include gdb_wait.h. (make_wchar_iterator): Use INTERMEDIATE_ENCODING. (find_charset_names): Use pexecute. Handle libiconv's output. Detect errors. (_initialize_charset): Use xstrdup. * gdb_wchar.h: Check HAVE_BTOWC. Split PHONY_ICONV and wchar cases. (INTERMEDIATE_ENCODING): New define. * configure, config.in: Rebuild. * configure.ac: Check for btowc. gdb/doc * gdb.texinfo (Character Sets): Document default character set. --- gdb/ChangeLog | 15 ++++++ gdb/c-lang.c | 4 +- gdb/charset.c | 110 +++++++++++++++++++++++++++++++++----------- gdb/config.in | 3 ++ gdb/configure | 3 +- gdb/configure.ac | 2 +- gdb/doc/ChangeLog | 4 ++ gdb/doc/gdb.texinfo | 4 +- gdb/gdb_wchar.h | 48 +++++++++++++++---- 9 files changed, 152 insertions(+), 41 deletions(-) diff --git a/gdb/ChangeLog b/gdb/ChangeLog index 8dfdfc8edf9..689552e7c93 100644 --- a/gdb/ChangeLog +++ b/gdb/ChangeLog @@ -1,3 +1,18 @@ +2009-04-14 Tom Tromey + + * c-lang.c (c_emit_char): Use INTERMEDIATE_ENCODING. + (c_printstr): Likewise. + * charset.c: Include gdb_wait.h. + (make_wchar_iterator): Use INTERMEDIATE_ENCODING. + (find_charset_names): Use pexecute. Handle libiconv's output. + Detect errors. + (_initialize_charset): Use xstrdup. + * gdb_wchar.h: Check HAVE_BTOWC. Split PHONY_ICONV and wchar + cases. + (INTERMEDIATE_ENCODING): New define. + * configure, config.in: Rebuild. + * configure.ac: Check for btowc. + 2009-04-15 Tom Tromey * c-lang.c (evaluate_subexp_c): Correctly handle EVAL_SKIP. diff --git a/gdb/c-lang.c b/gdb/c-lang.c index 86de9334ce0..027e9b292fb 100644 --- a/gdb/c-lang.c +++ b/gdb/c-lang.c @@ -296,7 +296,7 @@ c_emit_char (int c, struct type *type, struct ui_file *stream, int quoter) obstack_init (&output); make_cleanup_obstack_free (&output); - convert_between_encodings ("wchar_t", host_charset (), + convert_between_encodings (INTERMEDIATE_ENCODING, host_charset (), obstack_base (&wchar_buf), obstack_object_size (&wchar_buf), 1, &output, translit_char); @@ -562,7 +562,7 @@ c_printstr (struct ui_file *stream, struct type *type, const gdb_byte *string, obstack_init (&output); make_cleanup_obstack_free (&output); - convert_between_encodings ("wchar_t", host_charset (), + convert_between_encodings (INTERMEDIATE_ENCODING, host_charset (), obstack_base (&wchar_buf), obstack_object_size (&wchar_buf), 1, &output, translit_char); diff --git a/gdb/charset.c b/gdb/charset.c index 7dd0f40fb03..c77463814df 100644 --- a/gdb/charset.c +++ b/gdb/charset.c @@ -22,6 +22,7 @@ #include "gdbcmd.h" #include "gdb_assert.h" #include "gdb_obstack.h" +#include "gdb_wait.h" #include "charset-list.h" #include "vec.h" @@ -536,7 +537,7 @@ make_wchar_iterator (const gdb_byte *input, size_t bytes, const char *charset, struct wchar_iterator *result; iconv_t desc; - desc = iconv_open ("wchar_t", charset); + desc = iconv_open (INTERMEDIATE_ENCODING, charset); if (desc == (iconv_t) -1) perror_with_name ("Converting character sets"); @@ -707,35 +708,92 @@ find_charset_names (void) static void find_charset_names (void) { - FILE *in; + struct pex_obj *child; + char *args[3]; + int err, status; + int fail = 1; + + child = pex_init (0, "iconv", NULL); + + args[0] = "iconv"; + args[1] = "-l"; + args[2] = NULL; + /* Note that we simply ignore errors here. */ + if (!pex_run (child, PEX_SEARCH | PEX_STDERR_TO_STDOUT, "iconv", + args, NULL, NULL, &err)) + { + FILE *in = pex_read_output (child, 0); + + /* POSIX says that iconv -l uses an unspecified format. We + parse the glibc and libiconv formats; feel free to add others + as needed. */ + while (!feof (in)) + { + /* The size of buf is chosen arbitrarily. */ + char buf[1024]; + char *start, *r; + int len, keep_going; + + r = fgets (buf, sizeof (buf), in); + if (!r) + break; + len = strlen (r); + if (len <= 3) + continue; + /* Strip off the newline. */ + --len; + /* Strip off one or two '/'s. glibc will print lines like + "8859_7//", but also "10646-1:1993/UCS4/". */ + if (buf[len - 1] == '/') + --len; + if (buf[len - 1] == '/') + --len; + buf[len] = '\0'; + + /* libiconv will print multiple entries per line, separated + by spaces. */ + start = buf; + while (1) + { + int keep_going; + char *p; + + /* Find the next space, or end-of-line. */ + for (p = start; *p && *p != ' '; ++p) + ; + /* Ignore an empty result. */ + if (p == start) + break; + keep_going = *p; + *p = '\0'; + VEC_safe_push (char_ptr, charsets, xstrdup (start)); + if (!keep_going) + break; + /* Skip any extra spaces. */ + for (start = p + 1; *start && *start == ' '; ++start) + ; + } + } - in = popen ("iconv -l", "r"); - /* It is ok to ignore errors; we'll fall back on a default. */ - if (!in) - return; + if (pex_get_status (child, 1, &status) + && WIFEXITED (status) && !WEXITSTATUS (status)) + fail = 0; - /* POSIX says that iconv -l uses an unspecified format. We parse - the glibc format; feel free to add others as needed. */ - while (!feof (in)) - { - /* The size of buf is chosen arbitrarily. A character set name - longer than this would not be very nice. */ - char buf[80]; - int len; - char *r = fgets (buf, sizeof (buf), in); - if (!r) - break; - len = strlen (r); - if (len <= 3) - continue; - if (buf[len - 2] == '/' && buf[len - 3] == '/') - buf[len - 3] = '\0'; - VEC_safe_push (char_ptr, charsets, xstrdup (buf)); } - pclose (in); + pex_free (child); - VEC_safe_push (char_ptr, charsets, NULL); + if (fail) + { + /* Some error occurred, so drop the vector. */ + int ix; + char *elt; + for (ix = 0; VEC_iterate (char_ptr, charsets, ix, elt); ++ix) + xfree (elt); + VEC_truncate (char_ptr, charsets, 0); + } + else + VEC_safe_push (char_ptr, charsets, NULL); } #endif /* HAVE_ICONVLIST || HAVE_LIBICONVLIST */ @@ -748,7 +806,7 @@ _initialize_charset (void) /* The first element is always "auto"; then we skip it for the commands where it is not allowed. */ - VEC_safe_push (char_ptr, charsets, "auto"); + VEC_safe_push (char_ptr, charsets, xstrdup ("auto")); find_charset_names (); if (VEC_length (char_ptr, charsets) > 1) diff --git a/gdb/config.in b/gdb/config.in index 0d5effa81fd..ab0c84019f6 100644 --- a/gdb/config.in +++ b/gdb/config.in @@ -70,6 +70,9 @@ /* Define to 1 if you have the header file. */ #undef HAVE_BP_SYM_H +/* Define to 1 if you have the `btowc' function. */ +#undef HAVE_BTOWC + /* Define to 1 if you have the `canonicalize_file_name' function. */ #undef HAVE_CANONICALIZE_FILE_NAME diff --git a/gdb/configure b/gdb/configure index 51f35f5c1f7..e06a4510cc6 100755 --- a/gdb/configure +++ b/gdb/configure @@ -15774,12 +15774,13 @@ 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 iconvlist libiconvlist + ttrace wborder setlocale iconvlist libiconvlist btowc do as_ac_var=`echo "ac_cv_func_$ac_func" | $as_tr_sh` echo "$as_me:$LINENO: checking for $ac_func" >&5 diff --git a/gdb/configure.ac b/gdb/configure.ac index 821dffe2db3..88bd8a653af 100644 --- a/gdb/configure.ac +++ b/gdb/configure.ac @@ -796,7 +796,7 @@ 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 iconvlist libiconvlist]) + ttrace wborder setlocale iconvlist libiconvlist btowc]) AM_LANGINFO_CODESET # Check the return and argument types of ptrace. No canned test for diff --git a/gdb/doc/ChangeLog b/gdb/doc/ChangeLog index 89d7d114e66..68ba957f360 100644 --- a/gdb/doc/ChangeLog +++ b/gdb/doc/ChangeLog @@ -1,3 +1,7 @@ +2009-04-15 Tom Tromey + + * gdb.texinfo (Character Sets): Document default character set. + 2009-04-14 Pierre Muller * gdbint.texinfo: Change server name from sources.redhat.com to diff --git a/gdb/doc/gdb.texinfo b/gdb/doc/gdb.texinfo index 5269dfcfc93..ebba32fc2b7 100644 --- a/gdb/doc/gdb.texinfo +++ b/gdb/doc/gdb.texinfo @@ -7998,7 +7998,9 @@ Set the current host character set to @var{charset}. By default, @value{GDBN} uses a host character set appropriate to the system it is running on; you can override that default using the -@code{set host-charset} command. +@code{set host-charset} command. On some systems, @value{GDBN} cannot +automatically determine the appropriate host character set. In this +case, @value{GDBN} uses @samp{UTF-8}. @value{GDBN} can only use certain character sets as its host character set. If you type @kbd{@w{set target-charset @key{TAB}@key{TAB}}}, diff --git a/gdb/gdb_wchar.h b/gdb/gdb_wchar.h index 583140eeb53..07a6c877cf4 100644 --- a/gdb/gdb_wchar.h +++ b/gdb/gdb_wchar.h @@ -19,13 +19,36 @@ #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) - +/* We handle three different modes here. + + Capable systems have the full suite: wchar_t support and iconv + (perhaps via GNU libiconv). On these machines, full functionality + is available. + + DJGPP is known to have libiconv but not wchar_t support. On + systems like this, we use the narrow character functions. The full + functionality is available to the user, but many characters (those + outside the narrow range) will be displayed as escapes. + + Finally, some systems do not have iconv. Here we provide a phony + iconv which only handles a single character set, and we provide + wrappers for the wchar_t functionality we use. */ + + +#define INTERMEDIATE_ENCODING "wchar_t" + +#if defined (HAVE_ICONV) #include +#else +/* This define is used elsewhere so we don't need to duplicate the + same checking logic in multiple places. */ +#define PHONY_ICONV +#endif + +/* We use "btowc" as a sentinel to detect functioning wchar_t + support. */ +#if defined (HAVE_ICONV) && defined (HAVE_WCHAR_H) && defined (HAVE_BTOWC) + #include #include @@ -53,10 +76,15 @@ typedef int gdb_wint_t; #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 +/* If we are using the narrow character set, we want to use the host + narrow encoding as our intermediate encoding. However, if we are + also providing a phony iconv, we might as well just stick with + "wchar_t". */ +#ifndef PHONY_ICONV +#undef INTERMEDIATE_ENCODING +#define INTERMEDIATE_ENCODING host_charset () +#endif -#endif /* defined(HAVE_ICONV) && defined(HAVE_WCHAR_H) */ +#endif #endif /* GDB_WCHAR_H */ -- 2.30.2