This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: Configuring gdb_wchar.h
>>>>> "Eli" == Eli Zaretskii <eliz@gnu.org> writes:
Eli> There's no pexecute in the DJGPP library, but there's one in
Eli> libiberty, and it supports DJGPP as well (see pex-djgpp.c).
Thanks, somehow I missed that file when I looked.
Here's an updated patch that changes charset.c to use pexecute and to
detect errors. It also adds parsing for libiconv-style "iconv -l"
output.
I tried this with the glibc iconv, libiconv's iconv, and a phony iconv
that prints garbage and exits with an error.
If you want to give it a try, to make sure it still builds on DJGPP,
that would be nice. If it works for you I will check it in.
Tom
2009-04-14 Tom Tromey <tromey@redhat.com>
* 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.
diff --git a/gdb/c-lang.c b/gdb/c-lang.c
index 0c9e4f8..139d219 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 14862e7..d6e2b11 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"
@@ -527,7 +528,7 @@ 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");
@@ -698,35 +699,92 @@ 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 */
@@ -739,7 +797,7 @@ _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 0d5effa..ab0c840 100644
--- a/gdb/config.in
+++ b/gdb/config.in
@@ -70,6 +70,9 @@
/* Define to 1 if you have the <bp-sym.h> 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 51f35f5..e06a451 100755
--- a/gdb/configure
+++ b/gdb/configure
@@ -15776,10 +15776,11 @@ 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 821dffe..88bd8a6 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/gdb_wchar.h b/gdb/gdb_wchar.h
index 583140e..07a6c87 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 <iconv.h>
+#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 <wchar.h>
#include <wctype.h>
@@ -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 */