702 Commits

Author SHA1 Message Date
Niels Thykier
5026e9ee6e Move old_libraries_format to britney_util
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-10 20:01:03 +02:00
Niels Thykier
568d5f809c Move "undo_changes" to britney_util
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-10 20:01:03 +02:00
Niels Thykier
df47518dae Move "constants" to a new consts module
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-10 20:01:03 +02:00
Niels Thykier
9e14cb520c Move same_source to separate module
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-10 20:01:03 +02:00
Niels Thykier
b505a5a357 Remove unused write_bugs
The method was last invoked in 2008 (commit 3dc3be1c7).

Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-10 20:01:03 +02:00
Niels Thykier
1942e00ee1 britney: Do not consider "smooth-updates-only" interesting
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-09 18:11:22 +02:00
Niels Thykier
2477033786 britney: Refactor arguments of method
Rewrite the arguments of find_upgraded_binaries to not use an instance
of MigrationItem.  We want to call it at a time where we have not
created MigrationItems yet.

Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-09 18:11:22 +02:00
Niels Thykier
3df18bcd53 britney: Refactor a part of doop_source into its own method
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-09 18:02:55 +02:00
Colin Watson
3e6fd501c9 aggregate all causes, but take the status from the latest version 2013-07-09 16:25:51 +01:00
Colin Watson
5d00fa864e fix pkglist lookup 2013-07-09 16:14:25 +01:00
Colin Watson
823899c6b3 Only record autopkgtest causes for the latest tested version. 2013-07-09 14:33:38 +01:00
Niels Thykier
14cefe379e INSTALL: Update package list to use python2.7
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-07 15:47:04 +02:00
Niels Thykier
1641706305 Migrate to python2.7
Signed-off-by: Niels Thykier <niels@thykier.net>
2013-07-07 15:37:15 +02:00
Adam D. Barratt
527d7f03e6 Remove two "the the"s
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-06 12:42:52 +00:00
Adam D. Barratt
141d62cca1 Correct a thinko in a comment in should_upgrade_srcarch()
A removal hint will generate both source and per-arch excuses if the
version of the source package differs between testing and unstable. If
the source versions are the same then only the per-arch excuses will
be generated.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-06 11:26:11 +00:00
Adam D. Barratt
6d8aec3a6a Improve comments in should_upgrade_srcarch()
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-06 11:21:53 +00:00
Colin Watson
33d905fc3a Apply undo list in reverse order.
In rare cases with hints with overlapping virtual packages provided by
different sources, this can make a difference.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-05 18:06:40 +00:00
Colin Watson
e99ea14601 Don't crash on unversioned unblock hints.
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-05 18:04:33 +00:00
Colin Watson
dddd9eb88b Don't linkify "-" versions. 2013-07-02 16:34:52 +01:00
Colin Watson
24d57f4f89 Show links to Launchpad package pages in excuses. 2013-07-02 15:39:24 +01:00
Colin Watson
59705dda61 Improve "Requested autopkgtest" logging. 2013-07-02 12:34:41 +01:00
Colin Watson
8584a061e1 Run autopkgtests as soon as amd64 and i386 are ready to go. 2013-07-02 11:59:29 +01:00
Colin Watson
cb7878320e Link to Jenkins logs for autopkgtests. 2013-07-02 11:39:57 +01:00
Colin Watson
8023ee2a40 Stop limiting adt-britney to amd64; this shouldn't make a difference anyway, per Jean-Baptiste. 2013-07-02 11:32:45 +01:00
Adam D. Barratt
84bad954ad Only record a package once in the source->binary package mapping
If there are multiple versions of an arch:all package in unstable (due
to outdated or no longer built arch:any packages) then only one of them
should be recorded in the list of binary packages built from the source
package. Otherwise we may try and remove the binary package from various
lists multiple times, leading to crashes.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-07-01 18:13:51 +00:00
Colin Watson
b45bc14d4f Split force-autopkgtest into force-badtest (tested package) and force-skiptest (triggering package). 2013-06-27 08:18:04 +01:00
Colin Watson
c04c71055e Only issue "From wrong source" for binaries not at the same version in testing (same-version could happen for merged binaries in partial-unstable mode). 2013-06-26 17:57:37 +01:00
Colin Watson
55d5a1651c Clarify force-autopkgtest excuse message. 2013-06-25 18:29:41 +01:00
Colin Watson
75b84eebe5 Run autopkgtests for blocked packages. 2013-06-25 18:15:43 +01:00
Colin Watson
9da36a9385 Don't crash on unversioned unblock hints. 2013-06-25 11:54:53 +01:00
Colin Watson
1c1d8cf891 Make force hints imply force-autopkgtest. 2013-06-25 11:14:26 +01:00
Colin Watson
c1e29f844f Make force-autopkgtest apply to the tested package, not the triggering package. 2013-06-24 13:25:29 +01:00
Colin Watson
2e34feee1f Log autopkgtest status collection. 2013-06-24 11:00:58 +01:00
Colin Watson
bc85ea45cb Add force-autopkgtest to HINTS_ALL. 2013-06-24 10:44:13 +01:00
Colin Watson
41821072e2 Make autopkgtest request logging more useful by moving it after adt-britney request. 2013-06-17 17:57:44 +01:00
Colin Watson
6da7f96ca2 Pass Britney instances through to AutoPackageTest. 2013-06-17 17:24:50 +01:00
Colin Watson
5ed767701b Log autopkgtest requests. 2013-06-17 16:59:57 +01:00
Colin Watson
1cf35e27c8 Fix cause handling in AutoPackageTest.request. 2013-06-17 14:04:23 +01:00
Colin Watson
f961f74ef3 Run autopkgtests for valid candidates, and wait for the results. 2013-06-17 13:56:21 +01:00
Adam D. Barratt
9ee37c5160 Enable support for p-u packages
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-06-12 19:39:27 +00:00
Colin Watson
111ab78d7d Apply undo list in reverse order; in rare cases with hints with overlapping virtual packages provided by different sources, this can make a difference. 2013-06-12 15:55:04 +01:00
Adam D. Barratt
a58f6d72c9 Ensure that binNMUs from superseded sources are not considered
Given a source which provides two packages and has different versions
in testing and unstable, binNMUs in unstable corresponding to the
older source version should not be considered as migration candidates.

For example:

testing
-------

source 1
bin 1 arch1
bin 1 arch2

unstable
--------

source 2
bin 2 arch1
bin 1+b1 arch2

The binary migration on arch2 should not be considered a candidate.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-06-08 15:40:07 +00:00
Adam D. Barratt
c7f27a8225 Re-enable smooth-updates
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-05-05 11:09:37 +00:00
Adam D. Barratt
858e8de218 same_source: handle being passed "None" as a version
Although this should never happen, rather than crashing if one of the
versions is none, simply indicate that they are unequal.

Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-17 10:33:59 +00:00
Adam D. Barratt
7933a720aa Completer: add "approve" to the command completion
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 17:19:49 +00:00
Adam D. Barratt
ce72b31bc6 Completer: offer tpu packages when completing "approve" hints
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 17:19:01 +00:00
Adam D. Barratt
af86002dea Completer: copyright foo
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 17:12:04 +00:00
Adam D. Barratt
c345dea3b5 Completer: when processing a "remove", only offer packages from testing
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 17:10:11 +00:00
Adam D. Barratt
de56c719b6 Completer: add forgotten "remove" to command completion
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 16:59:19 +00:00
Adam D. Barratt
d9a9a26c57 Completer: support all hints
Signed-off-by: Adam D. Barratt <adam@adam-barratt.org.uk>
2013-03-16 16:39:23 +00:00