I have pushed the attached two patches, which replace internal uses of
AC_EGREP_CPP with simpler and/or robust checks. See the discussion at
the bottom of
https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.71/html_node/Running-the-Preprocessor.html
for why AC_EGREP_CPP is fragil
AC_TYPE_GETGROUPS is the last remaining use of AC_EGREP_HEADER in
stock Autoconf macros. It uses it only when cross compiling, as a
fallback from an AC_RUN_IFELSE check, testing for a bug in system
headers from the late 1980s or early 1990s, where gid_t *existed*
but the second argument to getgrou
On 2023-04-02 12:42, Zack Weinberg wrote:
Cross-referencing gnulib’s getgroups.m4 I see that there *are*
current-generation Unixes with bugs in getgroups that are worth
worrying about (notably, FreeBSD mishandles an error case even in
CURRENT).
That error case is not very important, as callin
On Sun, Apr 2, 2023, at 5:36 PM, Paul Eggert wrote:
> That error case is not very important, since programs are not likely
> to exercise it. The really serious failures are for long-dead systems
> (Ultrix 4.3, NextSTEP 3.2) that we no longer need to worry about.
>
> AC_FUNC_GETGROUPS has been docum
On 2023-04-02 18:30, Zack Weinberg wrote:
how about I revise this patch to remove the case for
freebsd* | darwin*, and make no other changes?
Yes, sounds good, thanks.
This patchset removes all remaining internal uses of AC_EGREP_CPP and
AC_EGREP_HEADER. (See the bottom of
https://www.gnu.org/savannah-checkouts/gnu/autoconf/manual/autoconf-2.71/html_node/Running-the-Preprocessor.html
for why this is desirable.)
Many of the modified tests are obsolete in some wa
Thanks, I took a quick look at all the patches and see no issues.
