Come together with the global Drupal community in Rotterdam, 28 Sept – 1 Oct 2026. Sessions, contribution, connection, and Early Bird savings until 8 June.After dbtng was committed the dblog tests now has two exceptions:
DBLog functionality: 309 passes, 0 fails, 2 exceptions
overall impressive that we still stand at:
4775 passes, 0 fails, 2 exceptions
| Comment | File | Size | Author |
|---|---|---|---|
| #4 | 298444-2.patch | 2.03 KB | c960657 |
Comments
Comment #1
boombatower commentedComment #2
boombatower commentedStill fails:
To be consistent with other test failures marking as critical.
Comment #3
lilou commentedComment #4
c960657 commentedThis removes the two exceptions.
The ids weren't being used for anything except for finding the number of rows, so I just did a COUNT(*) instead. The same approach could be applied later in the same function, but I assume that people prefer this patch to land soon in order to get rid of the last failed tests, so I tried not to solve too many problems in one patch.
Comment #5
lilou commentedI run DBlog test : 309 passes, 0 fails, 0 exceptions
Comment #6
webchickI would like to commit this as soon as it's RTBC so we go back to 100% test passing. It'd be great if Larry (Crell) and/or Jimmy (boombatower) could take a look. But it seems like it's simplifying the test, which is a good thing. The one thing I question is whether there's a more "PDO-like" way to do the count.
Comment #7
boombatower commentedCode looks fine, I'm not an expert on PDO stuff so it would be nice if Crell would take a look or someone with better knowledge in that area.
Comment #8
catchLooks fine and also fixes the exception for me, but I'm not that familiar with the new db layer either.
Comment #9
catchI think Crell's away at the moment, so I'm bumping this to RTBC, we can ping him when we get back to make sure it's OK and it looks like a decent cleanup anyway.
Comment #10
webchickI did some Googling and can't really find anything to refute this method of doing counts, so committed. Thanks!
Comment #11
webchickHrm. This seems to have cropped up again...
Two of these:
Undefined property: stdClass::$wid
Undefined property: stdClass::$wid
...on line 224.
Comment #12
c960657 commentedIs this line 224 in dblog.test? Which CVS revision? In HEAD, line 224 is a comment, right?
Comment #13
damien tournoud commented@webchick: I can't reproduce, neither running only the DBLog test nor running the whole test suite. Could you give us more information about the error?
Comment #14
webchickSure, I'll look into it more this evening. Could've been a fluke.
Comment #15
c960657 commentedLooks like this was just a fluke, so I am setting this back to fixed.