summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2016-07-27==== release 2.28.0 ====v2.28.0Natanael Copa1-1/+1
2016-07-27abuild: don't set and reset xterm title when splitting subpkgTimo Teräs1-2/+2
it's annoying flickering, and results the title being unset for the time when last split is executed and packages are created.
2016-07-26abuild: treat subpackages items as colon separated listsTimo Teräs1-15/+19
This is in preparation to support subpackages="pkg:split:arch" syntax, and just makes the current code ignore anything after the second colon if it exists. This allows to use the new syntax in aports git without running experimental abuild on the official builders.
2016-07-23abuild: use apk --no-scripts for target sysrootTimo Teräs1-3/+3
we probably will not be able to run the scripts as it can be incompatible system type
2016-07-23default to host gcc's machine for CBUILDTimo Teräs2-9/+2
2016-07-20Fix src buildsPaul Lorenz1-1/+19
Fix filename_from_uri call Ensure remote sources are fetched Include remote sources in source package Include pre/post scripts Include trigger scripts
2016-07-08abuild: rename distfile copy when checksum does not matchTimo Teräs1-4/+7
The previous deletion of file did not work properly at all. Fix this instead to rename the file, so we keep the previous download. This allows builders to redownload upstream file if the checksum has been changed in aports.
2016-07-08add aarch64 architectureTimo Teräs1-0/+1
2016-07-07abuild: support arch="all !armhf"Timo Teräs1-1/+10
this allows blacklisting architectures instead of just whitelisting them. useful when more architectures comeabout.
2016-07-06abuild: perform unzip operations quietlySören Tempel1-1/+1
unzip(1) is verbose by default since we don't run tar(1) in verbose mode we should run unzip with -q for consistency.
2016-07-05abuild: fix cross compilation undeps()Timo Teräs1-1/+1
"apk del" now returns error if the package name does not exist. Thus when cross compiling, always install the virtual .makedepends-$pkgname name for chroot also since undeps() will try to uninstall it always.
2016-06-20newapkbuild: redirect find(1) stderr to /dev/nullSören Tempel1-1/+1
If $sdir is not defined find writes an error message to stderr. For example when running: newapkbuild -n foo -a foo You will see a find error message because no source url was provided. This commit silences any find error message. An alternative solution would be to not invoke find unless $sdir is set.
2016-06-14==== release 2.27.2 ====v2.27.2Natanael Copa1-1/+1
2016-06-13newapkbuild: remove trailing spacesJakub Jirutka1-2/+2
2016-06-13newapkbuild: add empty line after builddir= and quote whole valueJakub Jirutka1-3/+1
2016-06-13newapkbuild: omit depends_dev if no *.h or *.hpp file foundJakub Jirutka1-0/+6
2016-06-13newapkbuild: fix undefined variable $_builddirJakub Jirutka1-1/+1
2016-06-13apkbuild-pypi: fix misspelled _pkg_realJakub Jirutka1-1/+1
2016-06-13apkbuild-pypi: rename _builddir to builddirJakub Jirutka1-3/+3
2016-06-13apkbuild-cpan: invoke default_prepare in prepare() functionJakub Jirutka1-0/+6
2016-06-13apkbuild-cpan: replace deprecated shell `backticks` with $(...)Jakub Jirutka1-4/+4
2016-06-13apkbuild-cpan: rename _builddir to builddirJakub Jirutka1-10/+9
2016-06-13abuild: fix apk_up2date checkNatanael Copa1-1/+1
fix check that all subpackages exists, not only main package
2016-06-13sample.initd: modernizeJakub Jirutka1-20/+7
2016-06-13sample.APKBUILD: rename _builddir to builddirJakub Jirutka1-8/+6
2016-05-20abuild-fetch: fix -Wformat-security warningsNatanael Copa1-3/+3
2016-05-20abuild-tar: improve portabilityNatanael Copa1-0/+4
splice is a linux thing.
2016-05-16==== release 2.27.1 ====v2.27.1Natanael Copa1-1/+1
2016-05-16abuild-fetch: retry to create lock on ESTALENatanael Copa1-2/+10
2016-04-14abuild-sudo: error check getlogin()Natanael Copa1-1/+2
2016-04-14abuild-sudo: make error msg "Not a member of group abuild" more clearJakub Jirutka1-2/+4
Few days ago some user on IRC pointed out that the current error message: abuild-apk: Not a member of group abuild is confusing. He was trying to build a package using abuild -r and didn't know what this message means. fixes #5408
2016-04-05abuild: don't cd to $builddir unless patches existsNatanael Copa1-0/+13
some packages don't have patches. We should not error if builddir is not set in this case.
2016-03-23==== release 2.27.0 ====v2.27.0Natanael Copa1-1/+1
2016-03-23newapkbuild: rename _builddir to builddirNatanael Copa1-6/+6
the _builddir should never been used bu abuild due to the _ prefix.
2016-03-23abuild: rename _builddir to builddirNatanael Copa1-1/+2
the _ prefix is reserved for APKBUILD. This is to avoid use variable names in APKBUILD that may clash with abuilds internal vars
2016-03-17abuild: set umaskKaarle Ritvanen1-0/+2
consistent result regardless of current setting
2016-03-14abuild: allow passing custom flags to patch(1) in prepareSören Tempel1-1/+1
To do so you need to set the patch_args variable to your desired command line flags.
2016-03-10abuild-fetch: use _exit after execvpNatanael Copa1-1/+1
2016-02-22==== release 2.26.0 ====v2.26.0Natanael Copa1-1/+1
2016-02-22abuild: fix getfattr matchTimo Teräs1-1/+1
it's a regular expression, not shell pattern. empty match will match everything (default is ^user\\.)
2016-02-22abuild: use gnu find supported way to find suid binariesTimo Teräs1-1/+1
2016-02-22newapkbuild: remove default prepare()Michael Zuo1-15/+0
Do not add prepare() to new APKBUILDs in the future, as this functionality is superseded by the addition of the default prepare patcher in abuild.
2016-02-22abuild: use standard-issue prepare by defaultMichael Zuo1-1/+14
A majority of APKBUILDs currently use exactly the prepare() provided by newapkbuild verbatim, even for packages where no patches exist, which has created the expectation that patches listed as sources are applied by default. This rev simply adds that default prepare() to abuild (and exposes it as default_prepare), with no configurability. If needed, default behaviour can be overridden by providing explicitly prepare().
2016-02-11abuild: Respect $reporev variable when fetching git repositoriesSören Tempel1-1/+2
For git repositories the $reporev variable is not really used since the $_rev variable is passed to git and $_rev was only set when $reporev was empty.
2016-02-11abuild: Fix indention of the snapshot() functionMichael Zuo1-44/+44
The snapshot function was indented with spaces while the rest of the code is indented with tabs.
2016-02-03abuild: minor speed improvement in symlink compressionNatanael Copa1-2/+2
instead of forking stat once force every file with one file argument, we fork it once with all the files as arg. This reduces number of forks.
2016-02-03abuild: fix finding already compressed manpagesNatanael Copa1-2/+3
2016-02-03abuild: avoid compress man pages doubleNatanael Copa1-0/+2
we should not compress man pages that are already compressed.
2016-02-03abuild: update symlinks before updating hardlinksSören Tempel1-7/+7
This should be a minimal performance improvement since hardlinks are replaced with symlinks and should thus already point to the correct file.
2016-02-03abuild: rewrite hardlink handling when compressing man pagesSören Tempel1-18/+22
The problem is that gzip refuses to run if it detects that a file has more than 1 link. Our existing solution (removing hardlinks, compressing the man page and recreating the hardlinks) made certain assumptions about inode order that are only given on Unix v7 like filesystems meaning it didn't work properly on 'tree-based' filesystems like BTRFS or ZFS. This patch has a different more bulletproof approach: It simply replaces all hardlinks with symlinks. This is way easier because symlinks (unlike hardlinks) can point to a file that doesn't exist, therefore we can update all links before compressing the file in an easy way.