From: Bruno Haible <bruno@clisp.org>
To: Christian Biesinger <cbiesinger@google.com>
Cc: bug-gnulib@gnu.org, Christian Biesinger <cbiesinger@gmail.com>
Subject: Re: time_r module does not work on MingW
Date: Sat, 16 Nov 2019 13:15:30 +0100 [thread overview]
Message-ID: <2192288.myS54ITcZ4@omega> (raw)
In-Reply-To: <1576792.iqhKLmYyPN@omega>
This patch fixes the issue. The cause was that the contents of <time.h>
in mingw depends on whether <unistd.h> or <pthread.h> was included before.
2019-11-16 Bruno Haible <bruno@clisp.org>
time_r: Fix for mingw.
Reported by Christian Biesinger <cbiesinger@google.com> in
<https://lists.gnu.org/archive/html/bug-gnulib/2019-11/msg00014.html>.
* lib/time.in.h: On mingw, include <unistd.h>.
* m4/time_r.m4 (gl_TIME_R): On mingw, include <unistd.h> before
<time.h>. Test for localtime_r in a way that works when it is defined
as an inline function.
diff --git a/lib/time.in.h b/lib/time.in.h
index 94e1da3..dc00503 100644
--- a/lib/time.in.h
+++ b/lib/time.in.h
@@ -37,6 +37,12 @@
# define _@GUARD_PREFIX@_TIME_H
+/* mingw's <time.h> provides the functions asctime_r, ctime_r, gmtime_r,
+ localtime_r only if <unistd.h> or <pthread.h> has been included before. */
+# if defined __MINGW32__
+# include <unistd.h>
+# endif
+
# @INCLUDE_NEXT@ @NEXT_TIME_H@
/* NetBSD 5.0 mis-defines NULL. */
diff --git a/m4/time_r.m4 b/m4/time_r.m4
index 5caeca7..72cc975 100644
--- a/m4/time_r.m4
+++ b/m4/time_r.m4
@@ -17,19 +17,54 @@ AC_DEFUN([gl_TIME_R],
dnl Some systems don't declare localtime_r() and gmtime_r() if _REENTRANT is
dnl not defined.
- AC_CHECK_DECLS([localtime_r], [], [], [[#include <time.h>]])
+ AC_CHECK_DECLS([localtime_r], [], [],
+ [[/* mingw's <time.h> provides the functions asctime_r, ctime_r,
+ gmtime_r, localtime_r only if <unistd.h> or <pthread.h> has
+ been included before. */
+ #if defined __MINGW32__
+ # include <unistd.h>
+ #endif
+ #include <time.h>
+ ]])
if test $ac_cv_have_decl_localtime_r = no; then
HAVE_DECL_LOCALTIME_R=0
fi
- AC_CHECK_FUNCS_ONCE([localtime_r])
- if test $ac_cv_func_localtime_r = yes; then
+ dnl We can't use AC_CHECK_FUNC here, because localtime_r() is defined as an
+ dnl inline function on mingw.
+ AC_CACHE_CHECK([for localtime_r], [gl_cv_func_localtime_r],
+ [AC_LINK_IFELSE(
+ [AC_LANG_PROGRAM(
+ [[/* mingw's <time.h> provides the functions asctime_r, ctime_r,
+ gmtime_r, localtime_r only if <unistd.h> or <pthread.h> has
+ been included before. */
+ #if defined __MINGW32__
+ # include <unistd.h>
+ #endif
+ #include <time.h>
+ ]],
+ [[time_t a;
+ struct tm r;
+ localtime_r (&a, &r);
+ ]])
+ ],
+ [gl_cv_func_localtime_r=yes],
+ [gl_cv_func_localtime_r=no])
+ ])
+ if test $gl_cv_func_localtime_r = yes; then
HAVE_LOCALTIME_R=1
AC_CACHE_CHECK([whether localtime_r is compatible with its POSIX signature],
[gl_cv_time_r_posix],
[AC_COMPILE_IFELSE(
[AC_LANG_PROGRAM(
- [[#include <time.h>]],
+ [[/* mingw's <time.h> provides the functions asctime_r, ctime_r,
+ gmtime_r, localtime_r only if <unistd.h> or <pthread.h> has
+ been included before. */
+ #if defined __MINGW32__
+ # include <unistd.h>
+ #endif
+ #include <time.h>
+ ]],
[[/* We don't need to append 'restrict's to the argument types,
even though the POSIX signature has the 'restrict's,
since C99 says they can't affect type compatibility. */
next prev parent reply other threads:[~2019-11-16 12:15 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-11 18:56 time_r module does not work on MingW Christian Biesinger
2019-11-11 19:53 ` Paul Eggert
2019-11-11 19:59 ` Bruno Haible
2019-11-11 20:47 ` Christian Biesinger
2019-11-12 2:28 ` Bruno Haible
2019-11-16 12:15 ` Bruno Haible [this message]
2019-11-16 22:41 ` wctype.h compile error on mingw with GNULIB_NAMESPACE Christian Biesinger
2019-11-17 3:05 ` Bruno Haible
2019-11-17 4:40 ` Christian Biesinger
2019-11-24 19:15 ` Bruno Haible
2019-11-25 23:28 ` Christian Biesinger
2019-11-25 23:53 ` Bruno Haible
2019-11-24 14:06 ` time_r module does not work on MingW Bruno Haible
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
List information: https://lists.gnu.org/mailman/listinfo/bug-gnulib
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2192288.myS54ITcZ4@omega \
--to=bruno@clisp.org \
--cc=bug-gnulib@gnu.org \
--cc=cbiesinger@gmail.com \
--cc=cbiesinger@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).