Merge lp:~stevenk/launchpad/sanity-check-bugwatch-bug-id into lp:launchpad
Status: | Merged | ||||
---|---|---|---|---|---|
Approved by: | Steve Kowalik | ||||
Approved revision: | no longer in the source branch. | ||||
Merged at revision: | 16116 | ||||
Proposed branch: | lp:~stevenk/launchpad/sanity-check-bugwatch-bug-id | ||||
Merge into: | lp:launchpad | ||||
Diff against target: |
186 lines (+31/-19) 4 files modified
lib/lp/bugs/interfaces/bugtracker.py (+1/-3) lib/lp/bugs/interfaces/bugwatch.py (+1/-3) lib/lp/bugs/model/bugwatch.py (+15/-12) lib/lp/bugs/tests/test_bugwatch.py (+14/-1) |
||||
To merge this branch: | bzr merge lp:~stevenk/launchpad/sanity-check-bugwatch-bug-id | ||||
Related bugs: |
|
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Ian Booth (community) | Approve | ||
Review via email: mp+128625@code.launchpad.net |
Commit message
Change the parsing methods that check if a remote URL for a bugwatch is valid to only accept bug numbers that are all digits.
Description of the change
Change the parsing methods that check if a remote URL for a bugwatch is valid. From what I can see, before I started this branch, only debbugs actually gave an error if the bug id supplied is not all digits, and the Trac and RoundUp parsers did something *awesome* and truncated the URL at the first non-digit character.
I think I've sorted it out so all parsers (except the e-mail address parser for obvious reasons) will croak when the remote bug id is not all digits. The LoC count can come off my something like -1000 count, but I have removed the pylint garbage and a comment that didn't add value in the slightest.