[12:56:42] Any ideas? (re @Yetkin: We would like to sort wikitables by name if the table contains such a column. How to do it with a user script/gadget?) [14:02:24] Not sure I understand. Isn’t that already possible? (re @Yetkin: We would like to sort wikitables by name if the table contains such a column. How to do it with a user script/gadget?) [14:05:53] ? [14:16:22] Can you explain what you’re trying to achieve in a little more detail? (re @Yetkin: ?) [14:22:20] We would like to sort wikitables by name if the table contains such a column. How to do it with a user script/gadget? (re @rafsusername: Can you explain what you’re trying to achieve in a little more detail?) [14:32:02] Are you trying to do that for specific pages or one user script to apply for all pages? For specific pages you just make the wiki table sortable and you can set the default sort as you wish (re @Yetkin: We would like to sort wikitables by name if the table contains such a column. How to do it with a user script/gadget?) [14:51:01] We are trying to do it for all pages that have a name column [17:45:44] !log lucaswerkmeister@tools-bastion-15 tools.lexeme-forms deployed 3ba43046ee (avoid changing lemma if not necessary) [17:45:48] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools.lexeme-forms/SAL [21:27:41] How difficult would it be for me (or anyone) to implement recurring queries within Quarry? Has it been attempted before? [21:33:37] Alternatively, can I hack it by sending a request to a certain endpoint at regular intervals? [21:35:16] It is definitely something that we've discussed but not something we've implemented -- I'm pretty sure we don't have 'cron for PAWS' yet either. [21:35:35] I think a simple toolforge tool is probably the only good way to do what you want. [21:40:06] I wonder if something like https://en.wikipedia.org/wiki/Template:Database_report would work for your use case instead? [21:44:19] The return set of one of my queries has over a million rows [21:44:52] how does quarry like that size? [21:44:56] I take it Quarry doesn't have a submission API endpoint [21:45:23] Good question. At the moment I'm just using the replicas directly. (re @jeremy_b: how does quarry like that size?) [21:46:31] you do some postprocessing later? maybe something could be moved into SQL to shrink the result set [21:52:23] Quarry very explicitely does not support any programmatic usage [21:52:42] i.e. if you have some other software working with the results, said software should be talking to the replicas directly instead of going via Quarry [22:46:05] https://en.wikipedia.org/wiki/User:Bot1058/refreshlinks.php — a bot I'm running continuously which repeatedly queries the replica database