summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Fix repo-alias value to use right syntaxLars Wirzenius2012-05-041-1/+1
|
* Make cmdtests set up repo-aliases and use prefixed reponamesLars Wirzenius2012-05-043-6/+7
|
* Remove use of baseurls in LocalRepoCacheLars Wirzenius2012-05-042-33/+29
| | | | The tests, they still fail.
* Implement repo-alias expansionLars Wirzenius2012-05-043-7/+46
| | | | | | | | Changed the delimieter between push and pull patterns to be #, since = is reasonably frequent in URLs, and # should never be necessary in git urls. Oh my how many tests now fail.
* Add tests for repo-alias expansionLars Wirzenius2012-05-042-0/+13
|
* Update LocalRepoCache tests to use repo aliases instead of baseurlsLars Wirzenius2012-05-041-6/+6
|
* Add --repo-alias setting and default valueLars Wirzenius2012-05-041-0/+16
| | | | Nothing uses the setting yet.
* Make morph's _clone_to_directory use CachedRepo.checkoutLars Wirzenius2012-05-045-11/+13
| | | | | | | | | This requires changing CachedRepo.checkout to handle non-sha1 refs, which resulted in some further changes in tests and their expected outputs. Also, a fix to CachedRepo to use the cwd keyword argument instead of pwd.
* Relax CachedRepo.checkout requirement that ref is a sha1Lars Wirzenius2012-05-042-11/+2
|
* baserock-bootstrap: attempt to fix staging fillerRichard Maw2012-05-041-1/+1
| | | | --option=path*foo won't be glob expanded, but --option path*foo will
* baserock-bootstrap: fix path for staging fillerRichard Maw2012-05-041-7/+5
| | | | | | | The changes to the cache directory layout meant that this failed to find the strata. This also fixes it extracting both pass1 and pass2's strata.
* Fix a line being too long.Jannis Pohlmann2012-05-031-1/+2
|
* Integrate RemoteArtifactCache into the building process.Jannis Pohlmann2012-05-032-18/+46
| | | | | | | | | For now we do not provide an option to avoid downloading artifacts during build. Unsetting --cache-server is an option but it's not user-friendly. Of course if there is no network connection, then downloading will fail and things will be built locally. We may want to add a generic --no-network switch to replace/extend --no-git-update later.
* Add support for RemoteRepoCache in build, show-dependencies, make-patch.Jannis Pohlmann2012-05-031-26/+36
| | | | | Both, build and show-dependencies should still work. I didn't test make-patch though.
* Use dedicated <cachedir>/artficats directory for local artifact cache.Jannis Pohlmann2012-05-0312-16/+25
| | | | | | Make sure that this directory is actually created before trying to save artifacts to it. Also adjust all tests to use this new directory for looking up artifacts.
* Get rid of the old internal morph APIsLars Wirzenius2012-05-0218-2482/+57
|
* Remove now-unnecessary modules from __init__.pyLars Wirzenius2012-05-021-6/+0
|
* Remove distributed building subcommandsLars Wirzenius2012-05-021-143/+0
| | | | We need to re-implement this for the new internal APIs.
* Convert _clone_to_directory to use new APIsLars Wirzenius2012-05-021-10/+16
|
* Make petrify use the new internal APIsLars Wirzenius2012-05-021-8/+12
|
* Make baserock-bootstrap use imx53/x86_64-generic BSP strata.Jannis Pohlmann2012-05-011-2/+4
|
* Build items as late as possible in the build order.Jannis Pohlmann2012-05-012-39/+40
| | | | | | | | At the moment this is done by creating a reverse topological sorting starting with the artifacts that depend on nothing else. Artifacts are then added to build groups as late as possible. Fix show-dependencies test output for the new build order generation.
* Add test for branching off not from masterLars Wirzenius2012-04-306-8/+71
|
* build-system-cpan test: only check file in tarRichard Maw2012-04-302-11/+1
| | | | | | It is possible that a bunch of stuff will be put in the tarball. What matters is if the binary is in the tarball, so only print that line of output
* stagingarea: fix dir detectionRichard Maw2012-04-301-4/+4
|
* Improve artifact extraction logic when overwriting existing stuff on filesystemLars Wirzenius2012-04-301-3/+49
| | | | | | | | | We had a problem where something in the bootstrap would fail because two different chunk artifacts had a path (/tools/lib64) exist as a directory exist as a symlink (to /tools/lib) versus a directory, and morph confusing things. The new logic should handle this better.
* stagingarea: when extracting, clobber outputRichard Maw2012-04-271-0/+3
| | | | | | I really don't like this, but it causes the bootstrap to fail because /bin/bash exists for compatibility and the stratum also creates /bin/bash.
* bootstrap: fix another non-sbin PATHRichard Maw2012-04-271-1/+1
|
* morph: submodules should only use an absolute refRichard Maw2012-04-271-1/+1
|
* bootstrap: don't fail on /tree/{sys,proc} unmountRichard Maw2012-04-271-6/+6
|
* bootstrap: include sbin in path for chrootRichard Maw2012-04-271-1/+1
| | | | | | chroot is used in morph, an absolute path would cause problems later in the build, when /tools is the prefix, so PATH needs to be able to find chroot, so the usual sbin needs to be included.
* Add a bit more cmd_build progress infoLars Wirzenius2012-04-271-0/+3
|
* Add a little progress reporting when build caches reposLars Wirzenius2012-04-271-0/+1
|
* buildenvironment: pass through BOOTSTRAP_TOOLSRichard Maw2012-04-261-0/+1
| | | | This is used in eglibc's bootstrap morphology.
* stagingarea: remove path from chrootRichard Maw2012-04-261-1/+1
| | | | | | | chroot may not be installed to /usr/sbin/chroot, this breaks bootstrap because chroot is installed in /tree/tools/sbin. PATH should be set, so putting the absolute path in should be unnecessary
* Restructure code so coverage is happyLars Wirzenius2012-04-261-8/+7
|
* Monkey patch Python's tarfile.TarFile for proper errors, ignoring overwritesLars Wirzenius2012-04-261-0/+31
|
* Fix unmounting /proc in staging so it works when /proc wasn't mountedLars Wirzenius2012-04-261-2/+2
|
* Fix how we set TMPDIR in the build environmentLars Wirzenius2012-04-262-1/+5
|
* Fix how unmounting of /proc inside chroot is doneLars Wirzenius2012-04-261-5/+8
| | | | | Note to self, everyone: /path/to/chroot/proc does not actually exist inside the chroot.
* Unmounted /proc in staging before assembling chunks, and only if it is thereLars Wirzenius2012-04-251-3/+2
| | | | | We have no use for /proc while building chunk artifacts, and hopefully this will prevent accidentally including /proc in them.
* Log /proc pathname in chroot when unmountingLars Wirzenius2012-04-251-1/+1
| | | | This is to make debugging easier.
* Fix "morph build" when building stuff that uses submodulesLars Wirzenius2012-04-252-1/+8
|
* Move repo and submodule caching into a helper methodLars Wirzenius2012-04-251-0/+6
| | | | This will be useful for later fixing of cmd_build.
* Fix update-gits to update repos of submodules, tooLars Wirzenius2012-04-251-2/+32
|
* Make test output be quite quiet with squeeze's version of gitLars Wirzenius2012-04-251-1/+2
|
* Make test output more quite by telling git init to be quiteLars Wirzenius2012-04-251-1/+1
|
* Remove set -x to make test output be silentLars Wirzenius2012-04-251-1/+1
|
* Oops, that error message was _not_ actually expectedLars Wirzenius2012-04-251-1/+0
| | | | | | We should be failing the test, since the code is broken. The expected output should not include the error; when the error goes away the test will pass.
* Add morphlib.Error and make local repo cache exceptions be based on itLars Wirzenius2012-04-252-2/+11
| | | | | This way, cliapp will automatically print an error message, rather than a stack trace.
View Lorry Controller Status