'-i' sed option is not portable, not specified by POSIX and does not
work on same platforms.
Replaced with simple and obvious workarounds.
ClosesMidnightCommander/mc#184.
Signed-off-by: Karlson2k (Evgeny Grin) <k2k@narod.ru>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Sometimes system default sed is not the best option.
With this modification user may easily override sed used by build system
without editing any file.
Signed-off-by: Karlson2k (Evgeny Grin) <k2k@narod.ru>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
configure: add explicit sed detection
Actually SED variable is set automatically by LT_INIT currently.
Having explicit check should help avoiding problems if/when LT_INIT
changed.
Signed-off-by: Karlson2k (Evgeny Grin) <k2k@narod.ru>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Found via `codespell -S
po,doc,./misc/syntax,./src/vfs/extfs/helpers/README.it -L
parm,rouge,sav,ect,vie,te,dum,clen,wee,dynamc,childs,ths,fo,nin,unx,nd,iif,iterm,ser,makrs,wil`
Co-authored-by: Yury V. Zaytsev <yury@shurup.com>
Signed-off-by: Kian-Meng Ang <kianmeng@cpan.org>
Signed-off-by: Yury V. Zaytsev <yury@shurup.com>
Before the change the description said:
Handle all compiler warnings as errors
Looks like a leftover from --enable-werror copy.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Ticket #2773: 'include' keyword (for command class def) have no effect
if it was defined before 'Include' keyword (for command def).
Ticket #3742: update comment in mc.ext.ini.
doc: remove mc.ext.ini format description from manual page.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Most of mc uses of PERL is to embed path to the interpreter in VFS
helpers. There we can use path to perl for --host. But
`date-of-man-include.am` is the place where perl is used for --build.
On most systems both paths are expected to be /usr/bin/perl.
But on some systems paths might differ a bit. Most prominent
example is NixOS, where packages get installed into unique prefixes:
$ file /nix/store/...-perl-5.36.0/bin/perl \
/nix/store/...-perl-powerpc64-unknown-linux-gnu-5.36.0/bin/perl
/nix/store/...-perl-5.36.0/bin/perl:
ELF 64-bit LSB executable, x86-64, version 1 (SYSV), ...
/nix/store/...-perl-powerpc64-unknown-linux-gnu-5.36.0/bin/perl:
ELF 64-bit MSB executable, 64-bit PowerPC or cisco 7500, ...
This allows running both binaries via qemu-user if needed for tests.
The change introduces PERL_FOR_BUILD (similar to autoconf's CC_FOR_BUILD
and friends) to allow passing both PERLs when needed.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
* configure.ac: test if "file" utility has -b option.
* (regex_check_type): don't compare file names if "file -b" is used.
Reduce variable scope. Rename variable.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Since release 2.70 of GNU Autoconf
> *** AC_PROG_CC_STDC, AC_PROG_CC_C89, AC_PROG_CC_C99 are now obsolete.
> Applications should use AC_PROG_CC.
Signed-off-by: Andreas Mohr <and@gmx.li>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
The mc's built-in samba library which is used to access data across
smb links is taken from an old samba version.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
FTBFS with following messages:
autoreconf: running: aclocal -I m4 --force -I m4
configure.ac:31: error: AC_INIT should be called with package and version arguments
/usr/local/Cellar/automake/1.16.4/share/aclocal-1.16/init.m4:29: AM_INIT_AUTOMAKE is expanded from...
configure.ac:31: the top level
autom4te: error: /usr/local/opt/m4/bin/m4 failed with exit status: 1
aclocal: error: /usr/local/Cellar/autoconf/2.71/bin/autom4te failed with exit status: 1
autoreconf: error: aclocal failed with exit status: 1
Initial changeset: partial revert of 32621b96d1.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
version.sh: move from maint/utils to the root of the source tree
and make it a part of distributable tarball.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
The build system runs configure script after switching to another branch
or changeset, which basically means that each git operation which
modifies HEAD will cause a full rebuild. This is highly annoying, in
particular during bisecting.
Solution:
Don't define version in config.h.
Add mc_version member to mc_global_t structure and use it instead of
VERSION macro.
Now only lib/global.c depends on MC version.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Lets minor bump gettext version to 0.18.2 (release 2012-12-25) to get rid of following warning:
configure.ac:269: warning: The 'AM_PROG_MKDIR_P' macro is deprecated, and its use is discouraged.
configure.ac:269: You should use the Autoconf-provided 'AC_PROG_MKDIR_P' macro instead,
configure.ac:269: and use '$(MKDIR_P)' instead of '$(mkdir_p)'in your Makefile.am files.
Signed-off-by: Andreas Mohr <and@gmx.li>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Fix deprecate warning
configure.ac:13: warning: AM_INIT_AUTOMAKE: two- and three-arguments forms are deprecated. For more info, see:
configure.ac:13: http://www.gnu.org/software/automake/manual/automake.html#Modernize-AM_005fINIT_005fAUTOMAKE-invocation
Support of m4_esyscmd_s() since autoconf-2.64 (release 2009-07-26).
Signed-off-by: Andreas Mohr <and@gmx.li>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
After 8857423e4e zip archives opened with
an error:
file -L -z archive.zip: Bad system call
This caused by using /usr/bin/file with -z option, because seccomp (a
security sandbox) doesn't allow it..
Solution: use -S option together with -z one.
The file command accepts the -S option since 5.33.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Embedding the configure time options into the executable can lead to
non-reproducible builds, since configure options often have embedded
paths. Add a configure time option to control if the configure args are
embedded so this can be disabled.
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
Bundled libintl did not support linking to internal static
libraries (libmc in our case): directly specified static
libraries are not pulled by libtool and are not usable for
dynamic libraries as PIC-related flags are not passed for
compilation.
This renders bundled libintl library unusable.
The change drops libintl bundling support and always relies
on external libintl (or falls back to disabled NLS).
On a related note gettext-0.20 drops support for bundling
or libintl and this change will ease migration to newer version.
The change is tested on x86_64-gentoo-linux-musl: mc builds
and links all tests successfully. A few tests fail for lack
of NLS support.
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
AX_GCC_FUNC_ATTRIBUTE detect attribute feature by compiler warnings.
Custom or unrecognized CFLAGS can causes extra warnings so attribute
check failed.
Reset CFLAGS and restore after AX_GCC_FUNC_ATTRIBUTE call.
Signed-off-by: Andreas Mohr <and@gmx.li>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
See gnulib a512e041120e9012e69afa2f5c3adc196ec4999a:
glibc 2.25 is deprecating the namespace pollution of <sys/types.h>
injecting major(), minor(), and makedev() into the compilation
environment, with a warning that insists that users include
<sys/sysmacros.h> instead. However, because the expansion of
AC_HEADER_MAJOR didn't bother checking sys/sysmacros.h until after
probing whether sys/types.h pollutes the namespace, it was not defining
MAJOR_IN_SYSMACROS, with the result that code compiled with -Werror
chokes on the deprecation warnings because it was not including
sysmacros.h.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
struct stat in libc for Linux kernel contains few fields more since 14+
years [1]. From bits/stat.h:
```
struct timespec st_atim; /* Time of last access. */
struct timespec st_mtim; /* Time of last modification. */
struct timespec st_ctim; /* Time of last status change. */
# define st_atime st_atim.tv_sec /* Backward compatibility. */
# define st_mtime st_mtim.tv_sec
# define st_ctime st_ctim.tv_sec
```
The conventional fields became an alias.
POSIX.1-2008 made struct stat st_[acm]tim mandatory [2].
OS takes care to initialize struct stat properly [3]. By not using an OS
syscall or a libc wrapper to fill struct stat, we have to take care of
initializing all fields (or at least those being used later) explicitly.
[1]: https://www.sourceware.org/ml/libc-alpha/2002-12/msg00011.html
[2]: https://www.sourceware.org/ml/libc-alpha/2009-11/msg00102.html
[3]: https://www.sourceware.org/ml/libc-alpha/2002-12/msg00013.html
Fixes: file timestamps not preserved (https://mail.gnome.org/archives/mc-devel/2017-April/msg00000.html)
Reported-By: Nerijus Baliunas <nerijus@users.sourceforge.net>
Signed-off-by: Yury V. Zaytsev <yury@shurup.com>
This is a preliminary step before being able to write tests for our rpm helper.
We introduce 'rpm2tags', a tool for converting RPM packages to parsable text
files. This will enable us to write tests that can run even where the 'rpm'
program isn't installed.
Signed-off-by: Mooffie <mooffie@gmail.com>
We introduce a mechanism by which tests can easily access configure-time
parameters (like @PERL@, @AWK@, ...).
It works by "sourcing" a file named config.sh (residing in the build tree):
. "$MC_TEST_EXTFS_CONFIG_SH"
$PERL -e 'print "hello"'
(Although config.sh has a shell syntax, Perl and Python programs too can
benefit from it, because it can be sourced into an .env_vars file and the
values exported from there.)
Of course, one can also use the traditional method, of *.in files listed in
configure.ac, but the mechanism introduced here is a more comfortable approach.
Signed-off-by: Mooffie <mooffie@gmail.com>
MC truncates timestamps during file copy and drops sub-second precision.
Make use of utimensat(), introduced in Linux kernel 2.6.22 (and since
2.6.26 compatible with POSIX-1.2008).
Signed-off-by: Andrey Gursky <andrey.gursky@e-mail.ua>
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
We introduce 'mc_parse_ls_l', a tool for parsing file-listings in format
similar to that of 'ls -l'. This format is used by the various extfs helpers.
We'll use this tool, in the next commit, to build a tester.
Signed-off-by: Andrew Borodin <aborodin@vmail.ru>
