[21:48:33] Hi. There is an UBN! request to revert https://phabricator.wikimedia.org/rMWa02a7ff8eab9d589d77c03d4b4c58fc4a05e65c8 EditPage: Show EditFilterMergedContent hook errors in an errorbox at https://phabricator.wikimedia.org/T149473. Apparently, it prevents comfortable editing on wikis with abuse filter using warnings messages. [21:49:23] A screenshot from cs. is indeed frightening. [21:53:06] legoktm ^^ [21:53:06] Or Reedy ^^ [21:55:46] Dereckson: Why does this seem to only be a problem on none en wikis? [21:56:41] I guess these are the wikis using AbuseFilter to add edit warnings. [21:56:56] with HTML in the warnings [21:57:40] Reedy Dereckson https://gerrit.wikimedia.org/r/#/c/318770/ [21:59:11] TBH, I'd be tempted to just do a partial revert, and just revert lines 1642/1652 [21:59:52] * Dereckson concurs [22:00:08] Oh, so i should just keep the i18n messages? [22:00:16] Or, the whole file at least [22:00:23] Don't bother messing with the i18n mesages [22:00:27] Ok [22:00:51] I know it's not your fault, you just hit the revert button etc :) [22:01:42] Reedy done :) [22:01:58] it will need backport to REL1_28 too :) [22:02:06] I wouldn't bother with that [22:02:25] oh [22:02:27] ok [22:02:29] Revert it out of the deployment branch in the first instance [22:02:39] legoktm may get round to fixing it before we need to revert out of master [22:02:48] And as such, backport the fix to REL1_28 when it's done [22:03:11] Ok [22:03:30] Reedy https://gerrit.wikimedia.org/r/#/c/318772/ [22:03:33] Dereckson ^^ [22:04:30] Just trying to think if reverting it is gonna break anything else [22:05:20] Oh, Maybe we could do a test run on test-wiki quickly? [22:05:40] I don't think it'll break anything [22:05:45] Ok :) [22:05:52] But may make some of the scribunto messages problematic [22:06:33] Oh, would it cause any serious problem if it was a bit prolematic for scribuntu? [22:07:33] hi [22:07:36] Well, AbuseFilter errors could appear for anyone... Scribunto errors not so much [22:07:46] legoktm hi [22:07:47] legoktm: shit's broken yo :( [22:08:12] oh god that's a terrible error message [22:09:38] ok, so are we going to revert? [22:09:53] No other intended side effects of reverting? [22:10:11] I'll deploy a revert to the deployment branch happily [22:10:19] no it should be fine [22:10:20] yes please [22:10:42] the double escaping is definitely wrong... [22:21:13] FINALLY JERKINS [22:22:15] LOL [22:27:01] Deployed [22:33:04] Thankyou [22:33:18] AaronSchulz looking at https://github.com/wikimedia/mediawiki/search?l=PHP&p=1&q=dbPrefix&type=Code&utf8=%E2%9C%93 it seems that Postgres dosen't use that [22:34:11] But https://github.com/wikimedia/mediawiki/search?utf8=%E2%9C%93&q=unittest_ that brings me to db prefix [22:35:05] But it uses the prefix here https://github.com/wikimedia/mediawiki/search?l=PLSQL&q=dbPrefix&type=Code&utf8=%E2%9C%93 [22:35:39] that's oracle ^^ [23:02:40] AaronSchulz if we add the /*_*/ and /*i*/ code to postgres sql, would that work? [23:11:20] AaronSchulz i found a fix https://gerrit.wikimedia.org/r/#/c/318769/ [23:11:23] It is a temp fix [23:11:40] since the unit tests should be creating those prefix tables, but that should fix ci :) [23:11:45] https://travis-ci.org/paladox/SemanticMediaWiki/builds/171692712 [23:12:33] Notice this one [23:12:34] 2) SMW\Tests\Integration\SQLStore\TableBuilder\TableBuilderIntegrationTest::testUpdateTableWithNewField [23:12:34] DBQueryError: A database query error has occurred. Did you forget to run your application's database schema updater after upgrading? [23:12:34] Query: CREATE TABLE "sunittest_rdbms_test" (id SERIAL NOT NULL PRIMARY KEY,t_text BYTEA,t_num bigint,t_int bigint) [23:12:35] Function: SMW\SQLStore\TableBuilder\PostgresTableBuilder::doCreateTable [23:12:37] Error: 42P07 ERROR: relation "sunittest_rdbms_test" already exists [23:38:41] AaronSchulz nope that dosent fix it