Merge lp:~brian-murray/launchpad/person-subscription-story into lp:launchpad
| Status: | Merged |
|---|---|
| Approved by: | Brian Murray on 2010-10-13 |
| Approved revision: | no longer in the source branch. |
| Merged at revision: | 11699 |
| Proposed branch: | lp:~brian-murray/launchpad/person-subscription-story |
| Merge into: | lp:launchpad |
| Diff against target: |
216 lines (+98/-83) 1 file modified
lib/lp/registry/stories/person/xx-person-subscriptions.txt (+98/-83) |
| To merge this branch: | bzr merge lp:~brian-murray/launchpad/person-subscription-story |
| Related bugs: |
| Reviewer | Review Type | Date Requested | Status |
|---|---|---|---|
| Leonard Richardson (community) | 2010-10-07 | Approve on 2010-10-13 | |
|
Review via email:
|
|||
Commit Message
Modify xx-person-
Description of the Change
I rewrote this test not to use or rely on sample data.
| Robert Collins (lifeless) wrote : | # |
| Leonard Richardson (leonardr) wrote : | # |
You mention Webster in the text before creating that person. To avoid confusion, the first mention of Webster should be something like "We're going to create a person called Webster... Any user can see..."
You mention bug #16 in the test text but you never mention it as a code literal, so you can remove it from the text.
You still mention bug 17 as a code literal, as well as the subscriber id 243656. Can you do better?
In a big code branch I'd consider these changes on the level of "approve with changes". But since the point of this branch is to improve the test, I'm marking this 'needs fixing'
| Brian Murray (brian-murray) wrote : | # |
I've modified the branch based off your feedback - thanks!
| Leonard Richardson (leonardr) wrote : | # |
"Any user can see that Webster is subscribed to some bugs. " => "Any user can see Webster's bug subscriptions." (Otherwise it reads like "As even a dummy can see, Webster is subscribed to some bugs.")
Sometimes Webster is "they" and sometimes "he". Pick a pronoun and stick with it.
Looks good otherwise.

Its good that you've done that but, the '16' and '17' literals are
still dependent on sample data.
If you make this into a pyunit test rather than a doctest, it will
actually be free of sample data dependencies.
(or use ..., but thats potentially risky).
-Rob