summaryrefslogtreecommitdiffstats
path: root/bin/misc-functions.sh
Commit message (Collapse)AuthorAgeFilesLines
* drop support for QA_DT_HASH/QA_STRICT_DT_HASHMike Frysinger2012-09-231-25/+0
| | | | | | | | These variables have been deprecated in favor of the new variables QA_STRICT_FLAGS_IGNORED/QA_FLAGS_IGNORED, and the tree has been converted over to the new ones, so drop the old vars. Signed-off-by: Mike Frysinger <vapier@gentoo.org>
* use `readlink -f` if it worksMike Frysinger2012-09-231-0/+13
| | | | | | | Rather than always re-implementing `readlink -f` in shell, probe the host tool first to see if it works. Signed-off-by: Mike Frysinger <vapier@gentoo.org>
* scan all lib subdirsMike Frysinger2012-09-231-5/+4
| | | | | | | Rather than hardcoding lib/lib32/lib64, scan all dirs starting with "lib". URL: https://bugs.gentoo.org/435834 Signed-off-by: Mike Frysinger <vapier@gentoo.org>
* doebuild: update environment.bz2 for pre/postinstv2.2.0_alpha129Zac Medico2012-09-171-1/+1
| | | | Also, remove unnecessary __dyn_preinst func.
* Convert funcs of ebuild.sh to __ prefixed namespace.Brian Harring2012-09-141-1/+1
|
* Convert funcs of phase*.sh to __ prefixed namespace.Brian Harring2012-09-141-5/+5
|
* Convert funcs of isolated-functions.sh to __ prefixed namespace.Brian Harring2012-09-131-39/+39
|
* misc-functions.sh: unwrap die messagesZac Medico2012-06-201-4/+3
| | | | | | | Jeremy Olexa <darkside@gentoo.org> complained that die was not handling these line continuations as it's supposed to. Maybe there's a bug in die's line continuation sed expression, but it won't hurt to unwrap these particular messages.
* Strip EPREFIX in RESTRICT=binchecks QA Notice.Zac Medico2012-05-081-2/+2
|
* NEEDED.ELF.2: include sonames from $ORIGINZac Medico2012-05-081-23/+2
| | | | | | | | | This reverts the behavior change from bug #205531 (see commit 5baa57a65368472c2c22b9bd2b30e634ef644b92). That special handling should be unnecessary since the changes from bug #289180 (see commit 296fc63fee262600811520fccf4692f47a39ffba), where the LinkageMapELF.findConsumers() exclude_providers argument is used to exclude irrelevant consumers.
* QA Notice for RESTRICT=binchecks + ELF filesZac Medico2012-05-051-0/+6
|
* Create NEEDED.ELF.2 despite RESTRICT=binchecks.Zac Medico2012-05-051-37/+43
| | | | | | This info is too useful not to have (it's required for things like preserve-libs), and it's tempting for ebuild authors to set RESTRICT=binchecks for packages containing pre-built binaries.
* Support /sys/fs/selinux/context for bug #410687.Sven Vermeulen2012-04-051-2/+4
|
* Bug #361759: QA check for /etc/udev/rules.dZac Medico2012-03-211-0/+12
|
* QA: detect pkg-config files that bleed LDFLAGSMike Frysinger2012-03-121-0/+7
| | | | Signed-off-by: Mike Frysinger <vapier@gentoo.org>
* install_qa_check: handle info dir rm failureZac Medico2012-02-011-1/+1
| | | | This will fix bug #400679.
* dyn_install: skip copy if possibleZac Medico2012-01-131-11/+16
| | | | | Also, fix EPREFIX breakage since commit 5d5364e982820fda56db5a40884e1c5f8e784b96.
* dyn_package: portable cp args by Mike FrysingerZac Medico2012-01-131-1/+3
|
* dyn_package: remove stale PROOT if necessaryZac Medico2012-01-131-0/+1
|
* Modify PKG_INSTALL_MASK to apply only to binpkg creation.Brian Harring2012-01-131-2/+13
| | | | | | | | | | | | Primarily useful for certain cases where cros needs to suppress debug symbols being included for specific pkgs, and/or suppressing binpkgs from carrying sensitive configuration data, but we still wish for it to be installed. BUG=chromium-os:25060 TEST=set PKG_INSTALL_MASK, check ${D} vs what's in the binpkg Change-Id: I0f81a9e394b7b0b64dbbcbe8d189876cb8421dfa
* Tweak CFLAGS scanelf loop.v2.2.0_alpha83Zac Medico2011-12-241-11/+13
|
* Don't check .GCC.command.line for kernel modules.Zac Medico2011-12-241-1/+12
| | | | | | | Linux kernel modules trigger the following warning when compiled with -frecord-gcc-switches: WARNING: mod.o (.GCC.command.line): unexpected non-allocatable section.
* Do CFLAGS check before strip for bug #395877.Zac Medico2011-12-241-71/+73
|
* Share variables for CFLAGS and LDFLAGS checks.Zac Medico2011-12-221-33/+44
| | | | | | | The rationale is that any exceptions to either of these checks will probably apply to both checks. So, QA_CFLAGS_IGNORED and QA_DT_HASH are merged into QA_FLAGS_IGNORED, and QA_STRICT_CFLAGS_IGNORED and QA_STRICT_DT_HASH are merged into QA_STRICT_FLAGS_IGNORED.
* Use *FLAGS for CFLAGS ignorance checks.Zac Medico2011-12-211-1/+6
| | | | | | | Current *FLAGS variables include CFLAGS, CXXFLAGS, FFLAGS, and FCFLAGS. Not that this check is only enabled if every one of these variables contains -frecord-gcc-switches, since otherwise the check could result in false positive results.
* s/QA_STRICT_DT_SWITCHES/QA_STRICT_CFLAGS_IGNORED/Zac Medico2011-12-211-1/+1
|
* s/QA_DT_SWITCHES/QA_CFLAGS_IGNORED/Zac Medico2011-12-211-5/+5
|
* install_qa_check: more RESTRICT=binchecksZac Medico2011-12-211-2/+4
|
* Support checking for ignored CFLAGS using -frecord-gcc-switches.Michał Górny2011-12-211-0/+40
|
* Support FEATURES=force-prefix.Zac Medico2011-12-081-8/+8
| | | | | | | This adjusts the logic from commit ff52f9dc31004becb8022e6437088d01917f413c to use FEATURES=force-prefix instead of USE=prefix. This has the advantage that we don't have to make any assumptions about the USE=prefix.
* Make USE=prefix enable EPREFIX in all EAPIs.Zac Medico2011-12-071-8/+16
| | | | | | | | This is safe because the prefix flag should be masked in all non-prefix profiles, and older EAPIs would otherwise be useless with prefix configurations. This brings compatibility with the prefix branch of portage, which also supports EPREFIX for all EAPIs (for obvious reasons).
* Use local EPREFIX for uniformity.Zac Medico2011-11-091-4/+3
| | | | | | | The EPREFIX variable is not readonly in EAPI 0, 1, and 2, so we can declare EPREFIX as a local variable. We must be careful not to let this variable leak into the environment of ebuild/eclass functions that we call, since bash propagates local variables to nested function calls.
* Use local ED for uniformity.Zac Medico2011-11-091-61/+54
| | | | | | | The ED variable is not readonly in EAPI 0, 1, and 2, so we can declare ED as a local variable. We must be careful not to let this variable leak into the environment of ebuild/eclass functions that we call, since bash propagates local variables to nested function calls.
* Add EPREFIX and ED support in all ebuild helpers.Zac Medico2011-11-011-51/+73
| | | | This allows our prefix tests to use helpers like insinto, doins, and dosym.
* Merge install_qa_check_prefix from prefix branch.Zac Medico2011-11-011-0/+124
|
* misc-functions.sh: condense world-writable outputZac Medico2011-10-201-7/+7
|
* misc-functions.sh: move sleep out of loopZac Medico2011-10-201-2/+2
| | | | | | Having sleep in the world-writable loop causes riduculous delays for prefix users with exotic filesystem constraints (like a FAT filesystem mounted with umask=000).
* Add QA_SONAME_NO_SYMLINK for bug #387053.Zac Medico2011-10-131-0/+4
|
* install_hooks: fix variable name for PORTAGE_CONFIGROOTFabian Groffen2011-09-231-1/+1
| | | | | in addition, omit the leading slash, since PORTAGE_CONFIGROOT contains a trailing slash
* Fix 'shopt: +s: invalid shell option name' error.Arfrever Frehtes Taifersar Arahesis2011-09-231-1/+1
|
* add install hooksBrian Harring2011-09-211-0/+15
|
* install_qa_check: adjust "poor programming" msgsZac Medico2011-09-181-10/+10
| | | | | Refer to "severe warnings" rather than "poor programming", in order to be more politically correct, as discussed in bug #374233.
* misc-functions.sh: fix incorrect has -o usageZac Medico2011-07-121-1/+1
|
* Remove all hasq/useq calls for bug #199722.Zac Medico2011-07-081-11/+11
|
* misc-functions.sh: remove stale NEEDED* filesZac Medico2011-06-301-0/+1
|
* Add a QA check for systemd unit file install.Michał Górny2011-06-241-0/+10
|
* install_qa_check(): Simplify code and make 'l' a local variable.Arfrever Frehtes Taifersar Arahesis2011-04-301-3/+3
|
* Check for references to ${ROOT} in install_qa_checks.David James2011-04-291-8/+26
| | | | | | | | | | | | | | | | | | | | | | | | When ROOT != /, binaries that reference ROOT will load their dependencies from ROOT first rather than from the system-configured path. This is a problem because the ROOT will be / on the target system. Besides the above, this patch also fixes incorrect parsing of scanelf output, where we would treat the RPATHs returned by scanelf as the names of binaries. TEST=When "stricter" FEATURE is enabled, verify that emerge fails when an ebuild references broken rpaths referencing ROOT. When "stricter" FEATURE is not enabled, verify that such references are automatically fixed. Also verify that ebuilds with non-broken RPATHs (e.g. RPATHs referencing $ORIGIN/../lib) are not touched by the change. BUG=chromium-os:14271 Change-Id: I4f29cc4ea9195a1255f080284da1f676e4a2c26b Review URL: http://codereview.chromium.org/6903153
* Move preinst/postinst_bsdflags from bash to pyDavid James2011-04-281-16/+0
| | | | | | | | | | | | | Moving these commands from shell to Python helps avoid an unnecessary call to misc-functions.sh for the postinst_bsdflags. This improves performance. BUG=none TEST=Run emerge-x86-generic -eg --jobs=16 libchrome Change-Id: I0c2af50b4d2f7644cabac84fde7fe4d682010c69 Review URL: http://codereview.chromium.org/6676107
* Bug #364467 - handle gcc-4.6-style QA warningsRyan Hill2011-04-261-18/+18
|
generated by cgit v1.2.3-1-g7c22 (git 2.25.1) at 2024-07-01 22:55:32 +0000