fix: Use more robust filename/rev matching for meeting materials #5192
Conversation
There was a problem hiding this comment.
There's also been a discussion about disallowing new document names that end with -nn, and I'd be completely fine forcing that change. (Though it would irritate chairs that like to name their slides after a particular version of a draft, but it won't take long for them to adjust).
ietf/meeting/views.py
Outdated
| if doc is not None and doc.get_related_meeting() == meeting: | ||
| return doc, None | ||
| # try parsing a rev number | ||
| name, rev = name.rsplit('-', 1) |
There was a problem hiding this comment.
This isn't sufficient. There are drafts out there with names that are prefixes of other names. You need to make sure the thing after the last - looks like two digits before trying to look up what came before.
Codecov Report
@@ Coverage Diff @@
## main #5192 +/- ##
==========================================
+ Coverage 88.54% 88.58% +0.03%
==========================================
Files 295 295
Lines 40110 40111 +1
==========================================
+ Hits 35516 35531 +15
+ Misses 4594 4580 -14
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. |
Fixes #4731.
Note that this does away with a series of moderately complex regex checks in the revision matching. I don't know the motivation behind those checks, but it seems to me that the
get_related_meeting()check should be adequate to distinguish whether a document is suitable to be returned by this method.