| Time |
Nick |
Message |
| 04:37 |
|
bshum joined #evergreen |
| 05:45 |
|
eglogbot joined #evergreen |
| 05:45 |
|
Topic for #evergreen is now Welcome to #evergreen (https://evergreen-ils.org). This channel is publicly logged. Logs for today: http://irc.evergreen-ils.org/evergreen/today |
| 07:28 |
|
Guest51 joined #evergreen |
| 08:39 |
|
mmorgan joined #evergreen |
| 09:58 |
|
sandbergja joined #evergreen |
| 10:27 |
sleary |
sandbergja I found this last night and I love how simple the custom tests are. https://biomejs.dev/blog/biome-v2 They're not quite done with the HTML piece, though -- https://github.com/biomejs/biome/pull/6907 |
| 10:29 |
eeevil |
sandbergja: I see you grabbed the might_have reporting patch. please do not commit that. it is incorrect. |
| 10:30 |
eeevil |
the fix is to use has_a instead of might_have. there is NEVER a case where an actual, extant column on a table should be described in an associated <link> entry with might_have. |
| 10:30 |
sandbergja |
sleary++ # that looks cool! |
| 10:30 |
sandbergja |
eeevil++ |
| 10:30 |
eeevil |
(jumping in here rather than just commenting on the bug just to catch you now, so we don't have to revert that later) |
| 10:31 |
sandbergja |
that might_have trips me up every time -- naming is hard |
| 10:31 |
sandbergja |
I remember it doesn't mean... quite what it sounds like it should mean |
| 10:31 |
eeevil |
I'll repost the "here's what those mean" in that bug, too |
| 10:31 |
sandbergja |
but never remember what it actually means hahaha |
| 10:31 |
sandbergja |
Yay thank you |
| 10:31 |
eeevil |
it's for the virtual side of a 1:1 relationship |
| 10:32 |
eeevil |
"that other table points at ME via MY pkey, put the object in this virtual field" |
| 10:32 |
eeevil |
"and there will only ever be exactly ONE remote object" |
| 10:32 |
sandbergja |
gotcha |
| 10:33 |
eeevil |
it's the 0-or-1 version of has_many, which is 0-or-more |
| 10:36 |
sandbergja |
eeevil: would you object to a perl test that parses the fm_IDL.xml and confirms that there are no non-virtual might_have links? It might be easier to remember *not* to do that if some automation helps us out. |
| 10:39 |
eeevil |
we should be able to do that with the xsd, and just call xmllint to test. that seems more correct since we're trying to enforce an xml schema. but I won't stand in the way of /any/ tests, if perl gets it done |
| 10:42 |
sandbergja |
ah that would work, I like that idea |
| 10:57 |
|
Dyrcona joined #evergreen |
| 11:06 |
Dyrcona |
It's too hot to think. |
| 11:23 |
|
jihpringle joined #evergreen |
| 11:23 |
|
Christineb joined #evergreen |
| 12:11 |
|
smayo joined #evergreen |
| 12:14 |
Bmagic |
Dyrcona++ |
| 12:27 |
smayo |
I were a fool |
| 12:29 |
smayo |
I fell for one of history's most classic blunders |
| 12:31 |
smayo |
eeevil++ |
| 12:42 |
Dyrcona |
smayo: You started a land war in Central Asia? |
| 12:43 |
|
jihpringle joined #evergreen |
| 12:46 |
Dyrcona |
"It was a hot one, like seven inches from the mid-day sun." |
| 14:38 |
jeffdavis |
Which bug has the incorrect might_have reporting patch? |
| 14:38 |
jeffdavis |
Is that bug 1161616 or something else? |
| 14:38 |
pinesol |
Launchpad bug 1161616 in Evergreen 3.14 "Reporter: incorrect link between hold_request & hold_request_cancel_cause" [Medium,Confirmed] https://launchpad.net/bugs/1161616 |
| 14:43 |
sandbergja |
jeffdavis: bug 1899777 |
| 14:43 |
pinesol |
Launchpad bug 1161616 in Evergreen 3.14 "duplicate for #1899777 Reporter: incorrect link between hold_request & hold_request_cancel_cause" [Medium,Confirmed] https://launchpad.net/bugs/1161616 |
| 14:43 |
jeffdavis |
aha! thanks |
| 14:45 |
sandbergja |
eeevil: I made a branch to get xmllint passing schema validation again, and put it into github actions so we can catch problems earlier. Would you mind taking a look if you have time? bug 2119034 |
| 14:45 |
pinesol |
Launchpad bug 2119034 in Evergreen "Get xmllint passing again and check it in github actions" [Undecided,New] https://launchpad.net/bugs/2119034 |
| 14:45 |
jeffdavis |
smayo++ # adding the explanation of might_have etc to the wiki |
| 14:47 |
sandbergja |
I came to the realization that I have No Idea At All how to add an assertion about the might_have must be virtual assertion to the schema, I'd be happy to write a Perl test, or just a ticket so somebody who knows how to do the xml schema approach can work their magic |
| 14:47 |
sandbergja |
smayo++ |
| 14:47 |
|
smayo joined #evergreen |
| 14:53 |
eeevil |
sandbergja: I'm not familiar with the GH bits, but the topmost commit looks great. we don't have any instances of unbounded, but that's the semantic default, so, +1 to that addition |
| 14:54 |
|
jihpringle joined #evergreen |
| 14:54 |
sandbergja |
eeevil: thanks for taking a look! |
| 14:55 |
eeevil |
I'll push a signoff for that commit now, unless you're planning to jump on a perl-ish test soon, in which case I'll commit to pushing the whole shebang (relatively) quickly! :) |
| 14:57 |
sandbergja |
I think the perl-ish will come later, I think I've spent all the XML time I can today haha |
| 14:58 |
eeevil |
FWIW, the reltype rules are in a comment at the top of fm_IDL.xml in the fix (and sandbergja's) branch. hopefully once that's in, it'll be a useful reminder |
| 14:58 |
eeevil |
but, wiki++ |
| 14:59 |
eeevil |
and, in particular |
| 14:59 |
eeevil |
smayo++ |
| 14:59 |
eeevil |
sandbergja++ |
| 15:00 |
sandbergja |
oh whoops, my branch has more commits than I intended -- it includes the commit that got this conversation all rolling in the first place |
| 15:01 |
sandbergja |
I can force push a fresh branch (and push bug 1161616 to main separately) |
| 15:01 |
pinesol |
Launchpad bug 1161616 in Evergreen 3.14 "Reporter: incorrect link between hold_request & hold_request_cancel_cause" [Medium,Confirmed] https://launchpad.net/bugs/1161616 |
| 15:08 |
sandbergja |
okay, pushed and force-pushed |
| 15:24 |
pinesol |
News from commits: LP#1161616: Wrong reltype attributes <https://git.evergreen-ils.org/?p=Evergreen.git;a=commitdiff;h=81015760b636b39d759234023a76c351f2d32da5> |
| 16:32 |
|
jihpringle joined #evergreen |
| 17:03 |
|
mmorgan left #evergreen |
