Merge lp:~vorlon/lp-ftbfs-report/report-only-regressions into lp:~wgrant/lp-ftbfs-report/production
Proposed by
Steve Langasek
Status: | Merged |
---|---|
Merged at revision: | 64 |
Proposed branch: | lp:~vorlon/lp-ftbfs-report/report-only-regressions |
Merge into: | lp:~wgrant/lp-ftbfs-report/production |
Diff against target: |
68 lines (+44/-0) (has conflicts) 1 file modified
source/build_status.py (+44/-0) Text conflict in source/build_status.py |
To merge this branch: | bzr merge lp:~vorlon/lp-ftbfs-report/report-only-regressions |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Matthias Klose | Pending | ||
William Grant | Pending | ||
Review via email: mp+343232@code.launchpad.net |
Commit message
Currently the archive rebuild reports list build failures that are not regressions vs the main archive. This is pointless redundancy and makes it difficult to identify which results should actually receive attention. The test archive rebuild reports should be focused on doing one thing well, namely identifying packages in the main archive that are not currently maintainable because they fail to build.
If someone feels these build failures should still be shown in the report, then I would at least want to categorize these separately in the output. But I don't think they belong here at all.
To post a comment you must log in.
As a data point, a test run of the report for test-rebuild- 20180408- bionic resulted in a 16% reduction in packages reported for main (45->36), and a 22% reduction for universe (1285->1002).