[11:08:04] Hello all, currently i'm noticing currently that pop-ups are shown when I hover over links. are these pop-ups a modification of the Hovercards feature? I guess it's not because I tried disabling Hovercards but still am able to see the pop-ups. [11:22:55] did you enable the original navigation popups? [[WP:POPUPS]] [11:41:05] closedmouth: Thanks for reminding that. I almost forgot enabling that Gadget. [13:40:13] Is anyone else getting constant "loss of session data" errors? [13:41:50] I've tried logging out and logging back in, and actually got a "No active login attempt is in progress for your session" error when entering the 2FA code (and was logged back in regardless) [13:43:18] There's been a few reports on and off [13:44:53] I think I managed to login via meta. There it accepted the 2FA code [13:46:05] nope, still getting "loss of session data" trying to edit enwiki :( [13:46:49] waldir: https://phabricator.wikimedia.org/T151770 [13:48:44] thanks legoktm, reading [13:51:40] I guess I'll try clearing the cookies. I did manage to login and edit successfully using a private browsing window, but I'm not sure that is useful information. In any case, I started following that bug. [13:54:25] legoktm: clearing the cookies seems to work, as indicated in the bug discussion. thanks for the pointer :) [13:54:44] waldir: are you using Firefox? [13:54:56] yes [13:56:38] do you have any privacy/anti-tracking extensions installed? [13:57:07] or something that might limit the number of cookies? [13:58:10] hmm, just regular ad blockers [13:58:19] uBlock to be specific [13:58:55] one of the two users who posted their addons also uses uBlock... [13:59:19] I'll mention that on the bug then [13:59:57] thanks [14:01:25] my pleasure, I hope that may help in fixing this :) [19:50:03] I'm scraping every Phabricator task. [19:50:10] I can't take this horrible fucking search any longer. [19:52:31] Yvette: in HTML or structured format? [19:52:37] HTML. [19:52:43] Is there a way to download all comments expanded? [19:52:56] I think the HTML includes them all? [19:52:59] I hope??? [19:53:06] Jesus, I hope it doesn't lazy load them. [19:53:39] Looks lazy loaded to me. [19:54:03] GODDAMN IT I JUST WANT TO FIND A TASK. [19:56:57] Ok, maybe with e.g. https://phabricator.wikimedia.org/T3495 curl downloads everything [19:57:42] Yeah, also in your browser you see everything if JavaScript is disabled. [19:58:58] (Or just because I'm logged out?) [19:59:48] Logged out gives some advantage. [19:59:52] But I think there's still collapsing when logged out. [20:00:15] I'm running the curl anyway, and just hoping it includes what I want. [20:00:25] Nepomuk search, Recoll or similar may be productive for a local search. [20:01:14] grep regex search for comment authors etc. will not be so easy because newlines seem to be inserted randomly. Would be nicer if every comment had its own line. [20:02:29] Yeah, but I just need file matches. [20:02:37] The file name should tell me where to look in a browser. [20:02:42] This is so fucking terrible. [20:02:58] If you just need a full-text search, yes. [20:02:58] it's… phabricator :P [20:04:06] Platonides: There's a comment of mine referencing the horrible behavior of MobileFrontend. [20:04:12] And how it strips out anything marked class="navbox". [20:04:18] But I can't find the comment to save my life. [20:04:26] Such an awful time-sink. [20:04:38] Soon we'll have enough momentum to migrate to bugzilla. ;) It's the last new thing, started in '8 while Phabricator was thought before '5, when Gettext was invented. [20:04:41] The search is unbearably terrible. [20:05:46] There seem to be at least 5 tasks were you bring up the topic [20:06:59] https://phabricator.wikimedia.org/T124168#2295956 https://phabricator.wikimedia.org/T55437#2408682 https://phabricator.wikimedia.org/T124959#2033333 https://phabricator.wikimedia.org/T110613 https://phabricator.wikimedia.org/T105365#1573781 [20:07:21] I had found https://phabricator.wikimedia.org/T32406 but it's not about navbox :P [20:08:03] I specifically referenced this template: https://en.wikipedia.org/wiki/Template:Infobox_SCOTUS_case/courts [20:08:07] Which someone else is now asking about. [20:08:17] https://phabricator.wikimedia.org/T124168#2295956 then [20:08:37] Goddamn it. [20:08:55] How did you find that? [20:09:17] https://www.google.it/search?q=navbox+mobile+MzMcBride+site:phabricator.wikimedia.org [20:09:34] * Yvette cries. [20:09:45] Doesn't work as well for recent tasks or substrings, of course [20:09:53] hahaha [20:10:26] I left yet another note on https://phabricator.wikimedia.org/T146843 [20:11:10] Yvette, I tend to use the maniphest search. I have this bookmarked with keyword "man" https://phabricator.wikimedia.org/maniphest/query/5Qf2sg9JYxdC/ and then adapt it as needed. e.g. [20:11:11] https://phabricator.wikimedia.org/maniphest/query/pZgS2LZklZYu/ [20:12:04] quiddity: Part of the issue, I'm fairly certain, is that the search index is incomplete. [20:12:13] So when Nemo just subscribed to that task, it now shows up in search results. [20:12:32] O.O if so, yikes. [20:14:22] I hate that it only finds exact matches, e.g. I couldn't search for just "SCOTUS" or "scotus" in 'Contains Words' section; i had to search for the full/exact template. [20:15:19] I filed https://phabricator.wikimedia.org/T153603 [20:15:29] I've scraped 3,000 tasks so far. [20:16:12] Yvette: i've been pointed to https://phabricator.wikimedia.org/T151500 last time i complained about search being useless. not sure if that is helpful. [20:16:34] Yvette: if you do the scrape, please zip it up and upload somewhere. perhaps to phabricator itself. ;) [20:16:39] Shouldnt phab auto index stuff into its search? [20:16:49] indeed it should [20:17:06] Why dont it [20:17:39] why indeed [20:17:59] doesnt* (the apostrophe isn't important, but the word-choice is!) [20:18:06] MatmaRex: It's in my home dir on tools-bastion-03 for now. [20:18:14] I can probably put it in /shared or somewhere. [20:18:20] quiddity: i said it that way because im on mobile [20:18:20] It'll have to be re-run occasionally, though. Ugh. [20:18:36] Yvette: cron cough cough [20:19:21] Thx. [20:19:59] NP [20:20:45] We should put cirrcussearch on phab lol [20:21:01] Phab search is generally a mess. Someday someone may figure out how to make it use Elasticsearch and work properly [20:21:52] We still have the old bz dont we :P [20:22:33] You're not really helping here. [20:23:40] we don't have the bugzilla search, which was by the way excellent, only a static mirror of the site. [21:00:00] Maybe we should ask for regular Phabricator dumps in order to make up-to-date mirrors possible? [21:04:19] those dumps would need to be anonymised [21:04:25] is that supported OOB? [21:29:54] Of course not, Phacility is not free software concepts such as right to fork or open data [21:30:37] They won't provide such a thing spontaneously but have little reason to decline being paid for it. [21:43:26] have you actually looked into the database layout Nemo_bis, or is that just random guessing and rambling? [21:47:33] I checked the docs https://secure.phabricator.com/book/phabricator/article/configuring_backups/ [21:52:25] ugh https://secure.phabricator.com/book/phabricator/article/configuring_backups/#security [22:02:11] MatmaRex: Sure we do. http://bugs.wmflabs.org/ ! [22:02:36] Only works for tasks that were around during the BZ era though... [22:03:12] https://dumps.wikimedia.org/other/misc/phabricator_public.dump [22:03:17] no idea what format that is in though [23:07:11] Nemo_bis: couldnt we do the same thing we do with like enwiki db ?