summaryrefslogtreecommitdiffstats
path: root/configure.ac
AgeCommit message (Collapse)AuthorFilesLines
2012-11-27makepkg: allow debug package suffix to be configurableAllan McRae1-0/+8
Signed-off-by: Allan McRae <allan@archlinux.org>
2012-11-27pactree: autodetect and use unicode line drawing charactersDave Reisner1-2/+2
Add a compile time check for langinfo.h so that we can possibly use unicode line drawing characters if the current locale is supportive of them. This can be explicitly disabled at runtime with the use of a new switch: -a, --ascii. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Allan McRae <allan@archlinux.org>
2012-11-27buildsys: eschew use of DEFS, prefer AM_CPPFLAGSDave Reisner1-1/+0
This is redundant, and any usage of -D should belong to CPPFLAGS. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Allan McRae <allan@archlinux.org>
2012-11-27configure.ac: cleanup duplication in --enable-git-versionDave Reisner1-5/+1
Avoid adding our own messaging, as autoconf will add this for us with the result of the AC_CHECK_FILE test. Reuse the cache variable from autoconf to set our local variable. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Allan McRae <allan@archlinux.org>
2012-11-27Fix thinko in configure.ac CFLAGS empty checkingDan McGee1-1/+1
Since commit d2669b47, CFLAGS specified on the command line haven't been respected at all, resulting in no optimization being applied to builds. This exposed one warning flag issue in some new code, which is also fixed here. Signed-off-by: Dan McGee <dan@archlinux.org> Signed-off-by: Allan McRae <allan@archlinux.org>
2012-08-01Do not enable _FORTIFY_SOURCE without optimizationAllan McRae1-0/+5
With glibc-2.16, using -D_FORTIFY_SOURCE requires that optimization (-O) be used or it will prodice a warning message. Enable -Werror in our test for _FORTIFY_SOURCE support to catch when a users specifies CFLAGS without optimization. The line to set CFLAGS="" when no CFLAGS are specified (either due to being unset or geniunely empty) is required as autoconf will use "-O2 -g" for its tests by defult when CFLAGS is unset, but will not add them to the CFLAGS used... Signed-off-by: Allan McRae <allan@archlinux.org>
2012-06-26Portability fixes for makepkgJeremy Huntwork1-0/+5
Allow makepkg to work correctly when used with find from busybox. Fix handling of cross directory symlinks. Signed-off-by: Jeremy Huntwork <jhuntwork@lightcubesolutions.com> Signed-off-by: Allan McRae <allan@archlinux.org>
2012-04-26libalpm: add pkg-config fileDave Reisner1-0/+1
No one seems to do this "correctly", but for the sake of having an easy method of detecting the presence and version of libalpm on a given system, we provide a straightforward .pc file. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-24configure: require bash >= 4.1 at compile timeDave Reisner1-1/+17
We've unofficially agreed to raise our minimum supported bash version to 4.1, and since added features that require it. Additionally, an earlier commit adds a syntax check to the builds of scripts/ and contrib/ which could conceivably fail with an earlier shell. Therefore, make this a hard requirement of the build process. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-24scripts/library: introduce parseoptsDave Reisner1-0/+1
This will replace our current options parser used in pacman-key, makepkg, and ideally elsewhere. It follows heuristics closer to that of GNU getopt long (and thus pacman itself), with the exception that it does not allow for options with optional arguments. Due to the way this parser will be used, this sort of functionality will not be needed. Instead of relying on eval+set, options are normalized into an array, OPTRET, which callers should expect to be populated after returning from parseopts. This avoids problems with quotes and spaces in arguments, assuming that the user quotes properly when passing into the application. A new test harness for parseopts is added in test/scripts. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-24configure: avoid linking against libsslDave Reisner1-1/+1
We're not linking to libssl, only libcrypto. -Wl,--as-needed will get rid of this, but there's no sense in checking for and linking against a library we don't need. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-04-09Add more warning flagsAllan McRae1-0/+1
Signed-off-by: Allan McRae <allan@archlinux.org>
2012-04-09Add a new configure option for excessive compiler warning flagsDan McGee1-0/+41
This adds a bunch of warning flags to the list used when compiling. Signed-off-by: Dan McGee <dan@archlinux.org>
2012-04-09buildsys: move autotools bloat to build-auxDave Reisner1-0/+1
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: cleanup libtool and autoconf initializationDave Reisner1-5/+2
- Use LT_INIT over AC_PROG_LIBTOOL, as the latter is a deprecated alias for the former. - Remove redundant macros which are called implicitly by LT_INIT. - Remove unneeded AC_PROG_CXX call (we don't use c++ anywhere) - Add AC_CONFIG_MACRO_DIR([m4]) -- not strictly necessary, but added for consistency with autogen.sh and Makefile.am ref: http://www.gnu.org/software/libtool/manual/html_node/LT_005fINIT.html Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: define warning CFLAGS in separate varDave Reisner1-3/+3
Continue the trend of not touching the environment CFLAGS, ensuring that the user always has the final say. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: use AC_DEFINE for CYGWIN macroDave Reisner1-1/+1
Instead of directly modifying CFLAGS, use config.h for its intended purpose. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: cleanup gpgme compile time checkDave Reisner1-20/+21
- handle gpgme libs and cflags separately rather than appending to CFLAGS and LDFLAGS - be consistent in AC_LINK_IFELSE check for gpgme 1.3.0 (though this is irrelephant since we don't actually run) - be consistent with usage of "have" and "with" variables (this actually ends up reducing SLOC) - when voluntary detection fails, unset GPGME_CFLAGS and GPGME_LIBS - when requested support fails the version check, complain about the min version. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: use pkg-config for libcurl detectionDave Reisner1-4/+17
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: use pkg-config for libarchive detectionDave Reisner1-3/+3
This also introduces a versioned dependency of >=2.8.0. Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-09buildsys: use pkg-config for openssl detectionDave Reisner1-13/+11
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
2012-04-07Check minimum required gpgme versionAllan McRae1-1/+1
We use interfaces first introduced in gpgme-1.3.0 so test we have at least that version. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-04-07Make selecting scriptlet shell workAllan McRae1-3/+3
The initial patch to implement this achieved nothing apart from adding a configure option. This patch makes that configure option do what it advertises. Note that specifing any shell apart from /bin/sh causes testsuite failures as /bin/sh is the only shell in the testing environment. Bug-found-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-04-07Merge branch 'maint'Dan McGee1-2/+2
Conflicts: lib/libalpm/signing.c lib/libalpm/sync.c
2012-04-07Revert "makepkg: calculate exact total file size"Dan McGee1-0/+1
This reverts commit b264fb9e9ddcc31dc8782390309421965e507383. With our "fix" of sleeping for BTRFS, we can go back to using `du` to calculate total installed size.
2012-04-074.0.3 release NEWS, version bumps, etc.v4.0.3working-maintDan McGee1-2/+2
Signed-off-by: Dan McGee <dan@archlinux.org>
2012-03-09Merge branch 'maint'Dan McGee1-3/+3
Conflicts: lib/libalpm/be_package.c
2012-03-05Do not dereference symlinks when calculating sizeAllan McRae1-3/+3
Passing the "-L" flag to stat means we get the size of the file being pointed to for symlinks instead of the size of the symlink. Keep "-L" usage in repo-add as we want the actual size of the package/delta/signature there. Signed-off-by: Allan McRae <allan@archlinux.org>
2012-02-13Merge branch 'maint'Dan McGee1-2/+2
Conflicts: lib/libalpm/sync.c src/util/pactree.c
2012-02-13Use AM_PATH_GPGME macro from gpgme instead of home-built gpgme detection.Nathan Phillip Brink1-11/+40
Fixes compilation on Gentoo, where CPPFLAGS=-I/usr/include/gpgme is necessary. The AC_SYS_LARGEFILE macro call has to be before the GPGME checks, otherwise the GPGME header gives an error about ABI incompatibilities. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-02-13Move ROOTDIR definition into config.hAllan McRae1-0/+1
This was the only variable of its kind when a define was done on the compiler command line. Move it into config.h instead. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-02-13Allow specifying shell for running scriptletsAllan McRae1-0/+9
Some distributions insist on using bash specific commands in their install scripts under the assumption that "sh" is a symlink to bash. This can causes issues if (e.g.) their users what to change sh to point at another shell, such as dash, that does not support these features. Add a configure option to explicitly set the shell being used to run install scripts. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-02-114.0.2 release updatesv4.0.2Dan McGee1-2/+2
Signed-off-by: Dan McGee <dan@archlinux.org>
2012-01-23Merge branch 'maint'Dan McGee1-1/+3
Conflicts: lib/libalpm/diskspace.c src/pacman/util.h
2012-01-23lib/dload: enforce usage of TCP keepalivesDave Reisner1-1/+2
This is particularly important in the case of FTP control connections, which may be closed by rogue NAT/firewall devices detecting idle connections on larger transfers which may take 5-10+ minutes. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2012-01-19Add diskspace checking support for Solaris/IllumosDan McGee1-1/+2
Was able to get my hands on one of these boxes today, so add yet another new way of doing this. I'm glad these calls are so standardized. This was compile tested on Linux and Illumos and seems to still be working in both places. Signed-off-by: Dan McGee <dan@archlinux.org>
2011-12-29makepkg: calculate exact total file sizeAllan McRae1-1/+0
The current calculation of the total file size for a package using "du" suffers from issues in portability and correctness. Especially on btrfs, this can result in clearly wrong package information such as: Download Size : 14684.29 KiB Installed Size : 7628.00 KiB Use an approach based on "stat" to calculate total file size. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2011-12-08Make automake generate silent rules by defaultDan McGee1-1/+2
This will require you to pass 'V=1' if you want the previous, more verbose output. Signed-off-by: Dan McGee <dan@archlinux.org>
2011-12-07Add 'silent-rules' to automake setup in configure.acDan McGee1-1/+1
This is awesome, and I don't know why we haven't already done this. It gives us the much more less verbose make output in a few different ways: * If you run `make V=0`, you will get the quiet output. * If you run `./configure --enable-silent-rules`, the quiet output is the default; verbose output can be had by passing V=1 to make. make[3]: Entering directory `/home/dmcgee/projects/pacman/lib/libalpm' CC add.lo CC be_local.lo CC be_package.lo CC be_sync.lo CC delta.lo ..... Signed-off-by: Dan McGee <dan@archlinux.org>
2011-11-16Merge branch 'maint'Dan McGee1-2/+2
2011-11-16Add helper method for creating and opening archive objectDan McGee1-0/+1
This moves the common setup code of about 5 different callers into one method. Error messages will now be common and shared in all places; several paths did not have any messages at all before. In addition, we now pick an ideal block size for the archive read based off the larger value of our default buffer size or the st.st_blksize field. For a filesystem such as NFS, this is often much larger than the default 8192- values such as 32768 and 131072 are common. Signed-off-by: Dan McGee <dan@archlinux.org>
2011-11-14Updates in preparation for 4.0.1 releaseDan McGee1-2/+2
Bump the version, update the translation template files, and fill in NEWS with relevant commits and changes since 4.0.0. Signed-off-by: Dan McGee <dan@archlinux.org>
2011-10-14Remove mcheck.h supportDan McGee1-2/+0
When was the last time anyone used this? That's what I thought. Signed-off-by: Dan McGee <dan@archlinux.org>
2011-10-12Bump version to 4.0.0v4.0.0Dan McGee1-1/+1
Signed-off-by: Dan McGee <dan@archlinux.org>
2011-09-22Update configure.ac version to rc2Dan McGee1-1/+1
Signed-off-by: Dan McGee <dan@archlinux.org>
2011-09-18configure: Fix quoting in SEDINPLACE on DarwinDave Reisner1-1/+1
single quotes expanded to nothing, leaving us with a command that assumed the sed expression was the backup suffix. Use a pair of escaped double quotes, which survives automake and ends up properly in makepkg. Signed-off-by: Dave Reisner <dreisner@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2011-08-17configure.ac: add checks for more types, functions and headersAllan McRae1-7/+15
This covers most types, functions and headers that we use in the code base. Currently we do not use any of these checks, but it is useful to have the configure output when looking at build issues on other peoples systems. Signed-off-by: Allan McRae <allan@archlinux.org> Signed-off-by: Dan McGee <dan@archlinux.org>
2011-08-11Bump version to 4.0.0rc1v4.0.0rc1Dan McGee1-5/+5
Signed-off-by: Dan McGee <dan@archlinux.org>
2011-08-11Merge branch 'maint'Dan McGee1-2/+2
Conflicts: scripts/repo-add.sh.in
2011-08-11configure: simplify CARCH generation madnessDan McGee1-30/+2
Rather than a hardcoded list of only a few select architectures (of the 250+ case statements in config.guess), simply define CARCH to be the first component of the "target triplet". This introduces one "regression"- powerpc will no longer become ppc. However, this is easily worked around in downstream distros if wanted. This was the only CPU architecture with this oddity so it was felt worth the price to make this change. Note that 'ppc64' wasn't handled in this same odd fashion before anyway. Signed-off-by: Dan McGee <dan@archlinux.org>