Merge lp:~lifeless/launchpad/run into lp:launchpad
Proposed by
Robert Collins
on 2010-10-11
| Status: | Merged |
|---|---|
| Approved by: | Michael Hudson-Doyle on 2010-10-11 |
| Approved revision: | no longer in the source branch. |
| Merged at revision: | 11696 |
| Proposed branch: | lp:~lifeless/launchpad/run |
| Merge into: | lp:launchpad |
| Diff against target: |
282 lines (+50/-75) 3 files modified
Makefile (+0/-1) lib/canonical/launchpad/scripts/runlaunchpad.py (+45/-62) lib/canonical/launchpad/scripts/tests/test_runlaunchpad.py (+5/-12) |
| To merge this branch: | bzr merge lp:~lifeless/launchpad/run |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| Michael Hudson-Doyle | 2010-10-11 | Approve on 2010-10-11 | |
|
Review via email:
|
|||
Commit Message
Improvements and prep work for consolidating Service with TacTestSetup.
Description of the Change
Improvements and prep work for consolidating Service with TacTestSetup
To post a comment you must log in.
| Robert Collins (lifeless) wrote : | # |
| Michael Hudson-Doyle (mwhudson) wrote : | # |
166 + """kill process and BLOCK until process dies.
Should have an initial capital.
I also wonder if stop_process should get more forceful than TERM, but well, that's something for another branch.
The changes look nice, btw.
review:
Approve

Note that the changes to ignore should_launch are because some services deliberately ignore it - they don't run by default but when explicitly chosen they need to run; their 'should_launch' is defined as 'return False' : we already filter so that only wanted services are attempted, so checking for should_launch on setUp is a harmful (breaks consistency, causes more code) belts-and-braces.