00:03:26 -!- amalloy is now known as amalloy_ 00:15:07 03advil02 07* 0.25-a0-1079-g858b6a6: Add a placeholder fix for trap+corruption+nymphs 10(4 minutes ago, 1 file, 9+ 0-) 13https://github.com/crawl/crawl/commit/858b6a6ff660 00:17:17 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1079-g858b6a6ff6 (34) 00:18:34 advil: thanks, I'll let them know that they can play their game again 00:18:42 hopefully 00:18:47 guess I can actually test their save first 00:18:58 ah good point I actually forgot we have their save 00:19:08 that ... would have been a good thing to use 00:19:39 haha 00:20:04 well I had the save and you got to the fix before I did 00:20:20 New branch created: webtiles-save-info (3 commits) 13https://github.com/crawl/crawl/tree/webtiles-save-info 00:20:20 03advil02 07[webtiles-save-info] * 0.25-a0-1062-gdb88980: Add a CLO that prints save info as json 10(26 hours ago, 5 files, 174+ 5-) 13https://github.com/crawl/crawl/commit/db88980feb20 00:20:20 03advil02 07[webtiles-save-info] * 0.25-a0-1063-g58f2a50: CLO to give game types in json 10(7 hours ago, 1 file, 49+ 1-) 13https://github.com/crawl/crawl/commit/58f2a5084680 00:20:20 03advil02 07[webtiles-save-info] * 0.25-a0-1064-gaa6d9e4: Add experimental save info to webtiles lobby 10(40 minutes ago, 4 files, 168+ 22-) 13https://github.com/crawl/crawl/commit/aa6d9e46a8dd 00:21:14 The water nymph dies! 00:21:14 Attempting to change terrain to a trap without a corresponding trap_def! 00:21:19 and no crash 00:21:22 ok good 00:21:23 thanks 00:22:18 oh man, is the next step letting us download a save? 00:22:43 I just walked a non-technical macos user through using CLI ssh to get that crash save 00:22:52 we got through it, but not having to do that would be amazing 00:23:10 no, that would be nice but this is something else 00:24:03 https://www.dropbox.com/s/tgs2b7w8yw9mprh/Screenshot%202020-05-15%2000.23.45.png?dl=0 00:24:55 oh, very cool 00:26:13 this is my Last Minute Feature for 0.25 00:27:13 though tbh I do also want to try to merge alex's webtiles config branch because a version change is a really good time to do that 00:27:38 right, I'm happy to help with CAO if you need, but I imagine it's a one person job and you have a better grasp of it than I do 00:33:34 in principle nothing needs to change, so whatever work there is will basically be figuring out what went wrong 01:36:22 save downloading is sitting in a side branch 01:37:54 it's not merged because it adds HTML templates, which iirc might clobber any custom templates server ops have 01:38:55 Unstable branch on crawl.develz.org updated to: 0.25-a0-1079-g858b6a6ff6 (34) 02:10:52 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-1079-g858b6a6ff6 02:21:27 advil: cool branch 02:22:06 i would like to display more info about in-progress games in the lobby. Rather than relying on tracking milestone file, which is flaky when people resume a game for instance 02:25:25 -!- amalloy_ is now known as amalloy 02:36:19 alexjurkiewicz: remember those player role colors on webtiles-changes? 02:36:25 what an awful hack 02:37:27 obviously the dream is to somehow export game data and cheevos through some api 02:42:13 from the crawl binary? 02:42:56 could probably send that over the socket between crawl and the webserver 02:46:33 aidanh: no, I mean things that we track in sequell based on analysis of games, like greatplayer etc 02:47:12 in that ancient branch I implemented a thing that would fetch data from sequell's public nick url and update an sqlite db that webtiles was modified to read from 02:47:24 so it could color player names based on this 02:48:26 in a more advanced world something like sequell or our general scoring system would export stuff over some API 02:56:59 ah, right 02:57:17 btw who owns the dcss.io domain? 02:59:59 I believe that's an advil creation 03:23:31 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-1079-g858b6a6 (34) 03:31:10 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3054-g1a545eca1e 03:32:07 Fork (bcadrencrawl) on crawl.kelbi.org updated to: 0.22.1-2694-g56f04153ee 05:14:46 -!- amalloy is now known as amalloy_ 06:57:41 dcss deepmind startup domain 07:40:53 03gammafunk02 07* 0.25-a0-1080-g72af00c: Simplify targeter classes 10(13 minutes ago, 3 files, 6+ 10-) 13https://github.com/crawl/crawl/commit/72af00c55adc 07:52:40 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1080-g72af00c55a (34) 09:34:27 alexjurkiewicz maybe this is obvious but why is the directory with the webtiles config yaml files called games.d? 09:36:19 (I would have never guessed that's where it would be) 09:40:12 I guess this is based on an /etc naming convention 09:41:40 lol at this phrasing of the question on SO: "Is there some mystical Linux rationale to this naming style?" 09:41:46 anyways, I think I get the idea 09:42:07 <|amethyst> AFAIK .d is for "directory" 09:43:25 right, and it also seems to now mean (or imply) "directory with config files that are modular and autoloaded" 10:03:25 03ebering02 07* 0.25-a0-1081-gdddbc3d: Restore the default auto_butcher_max_chunks = 10 10(11 minutes ago, 2 files, 6+ 1-) 13https://github.com/crawl/crawl/commit/dddbc3d8f2f9 10:05:38 03ebering02 07* 0.25-a0-1082-g38b645b: Document the auto_butcher_max_chunks default in options_guide.txt 10(45 seconds ago, 1 file, 3+ 2-) 13https://github.com/crawl/crawl/commit/38b645b55d9a 10:17:12 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1082-g38b645b55d (34) 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1083-gfc79a73: Enhancements to webtiles game definitions 10(6 weeks ago, 7 files, 395+ 53-) 13https://github.com/crawl/crawl/commit/fc79a7346936 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1084-gec0d3a7: Reduce log spam 10(6 weeks ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/ec0d3a76dc45 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1085-g8dcfb2f: Convert README to markdown and update 10(6 weeks ago, 2 files, 87+ 49-) 13https://github.com/crawl/crawl/commit/8dcfb2fc82a9 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1086-gf4f6b11: Remove version pin from Tornado dependency (advil) 10(6 weeks ago, 5 files, 21+ 18-) 13https://github.com/crawl/crawl/commit/f4f6b119312a 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1087-g9ecdaae: Improve some webtiles log messages 10(6 weeks ago, 2 files, 6+ 3-) 13https://github.com/crawl/crawl/commit/9ecdaae92ffc 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1088-g2a3374b: Add some missing type information 10(6 weeks ago, 2 files, 14+ 18-) 13https://github.com/crawl/crawl/commit/2a3374b39a96 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1089-g26357f4: Allow overriding some HTML templates 10(6 weeks ago, 7 files, 13+ 10-) 13https://github.com/crawl/crawl/commit/26357f45c02c 10:18:02 03Alex Jurkiewicz02 {advil} 07* 0.25-a0-1090-g90e1a17: Always show milestone column 10(6 weeks ago, 2 files, 16+ 13-) 13https://github.com/crawl/crawl/commit/90e1a17ddc7e 10:18:02 03advil02 07* 0.25-a0-1091-ge9962a3: Add back in all non-default templates for compatibility 10(49 minutes ago, 3 files, 8+ 0-) 13https://github.com/crawl/crawl/commit/e9962a393cac 10:20:27 FLAG DAY 10:24:12 re harm since someone asked about it, I was interested in working on converting the penalty to something more interpretable like a frail effect (and thinking about how to better convey the damage part ui-wise too) but that almost certainly won't happen today 10:24:29 I wonder if it would make sense to simply lower the penalty for 0.25? 10:26:13 also, should we merge the icons that cebolla submitted in https://github.com/crawl/crawl/issues/1387? 10:26:29 at least I assume that it's the same cebolla :-) 10:26:48 yep! 10:28:21 those look good for now; better than TO DO 10:31:36 a VI player showed up on discord with some UI requests so I'm compiling those notes 10:31:45 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1091-ge9962a3 (34) 10:31:56 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1091-ge9962a393c (34) 10:31:58 other harm idea: make it 10% / 10%, but it has some +slay on it when it's alone on an amulet, and it shows up on randarts for people to decide it's maybe worth-it with other properties 10:32:30 the whole point of harm is to get those sick 27 invo beogh smites at +30% damage 10:32:51 just making it into slay or archmagi misses that niche 10:33:02 . o O (harm + Absolute Zero) 10:35:36 Webtiles server stopped. 10:36:23 03ebering02 07* 0.25-a0-1092-g1993010: Add a recommended options section for blind players (#1407) 10(68 seconds ago, 1 file, 34+ 0-) 13https://github.com/crawl/crawl/commit/1993010d6657 10:42:24 03advil02 07* 0.25-a0-1093-gcd5d389: Python 2.7 compatibility 10(2 minutes ago, 1 file, 9+ 5-) 13https://github.com/crawl/crawl/commit/cd5d38940ce6 10:44:56 ??stone of tremors 10:44:56 stone of tremors[1/2]: The earth {elemental evoker} sends a shockwave through walls, which damages monsters and can (rarely) {shaft} them. Also summons up to 3 {earth elemental}s. Also breaks rock (but not stone) walls. Removed in 0.19. 10:45:19 through walls O_o 10:46:45 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1093-gcd5d389 (34) 10:46:58 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1093-gcd5d38940c (34) 10:53:02 it is never simple 10:53:10 03advil02 07* 0.25-a0-1094-gbec1141: Disable yaml games by default 10(89 seconds ago, 2 files, 4+ 1-) 13https://github.com/crawl/crawl/commit/bec1141c92d9 10:53:31 why do I ever think it would be simple 10:53:51 (tbc that is definitely *not* a comment on the patch, just on the process of updating cao) 10:55:23 alexjurkiewicz also, before that commit when I tried to start the webtiles server, I got an exception in load_games: "OSError: [Errno 40] Too many levels of symbolic links: '/chroot/crawl-master/webserver/games.d'" 10:55:30 do you have any idea what could cause this? 10:55:39 I don't see any symbolic links 10:55:56 maybe it's just some weird interaction of the sample yaml file with some files that happen to exist on cao 10:56:57 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1094-gbec1141 (34) 10:57:33 Webtiles server started. 10:59:05 how is there someone logged into my :8081 testing instance 10:59:14 oh heh probably just a console player 10:59:53 Webtiles server stopped. 11:02:11 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1094-gbec1141c92 (34) 11:03:11 03advil02 07* 0.25-a0-1095-gbac526d: Add a CLO that prints save info as json 10(2 days ago, 5 files, 174+ 5-) 13https://github.com/crawl/crawl/commit/bac526da1940 11:03:11 03advil02 07* 0.25-a0-1096-gb1a17ef: CLO to give game types in json 10(18 hours ago, 1 file, 49+ 1-) 13https://github.com/crawl/crawl/commit/b1a17ef1ac15 11:03:11 03advil02 07* 0.25-a0-1097-ge4e6ab6: Add experimental save info to webtiles lobby 10(11 hours ago, 6 files, 183+ 30-) 13https://github.com/crawl/crawl/commit/e4e6ab6fc041 11:03:39 Build failed for master @ cd5d3894 https://github.com/crawl/crawl/actions/runs/105781985 (failed: Webtiles Server) 11:06:14 03advil02 07* 0.25-a0-1098-g03feba3: Fix flake8 error 10(46 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/03feba329250 11:12:13 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1097-ge4e6ab6 (34) 11:15:29 four years ago, i made this alias, and i no longer have any idea what for. is this a common operation? rusty with git now. DB = "!f(){ git branch -d ${1} && git push origin --delete ${1}; };f" 11:16:03 I think it deletes a branch locally and upstream 11:16:23 yep 11:16:41 I have a similar alias 11:16:51 Webtiles server started. 11:19:17 Webtiles server stopped. 11:19:22 Webtiles server started. 11:19:40 Build failed for master @ bec1141c https://github.com/crawl/crawl/actions/runs/105790316 (failed: Webtiles Server) 11:21:36 weird, on CAO my json CLO thing is getting some escape codes before the json? 11:21:38 `\x1b[2J\x1b[H` 11:21:48 anyone have any guesses why that would happen? 11:21:50 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1098-g03feba3292 (34) 11:22:58 <|amethyst> advil: that looks familiar, trying to remember where it comes from. That's clear screen + go to upper left 11:23:27 wonder why it didn't happen locally 11:24:28 <|amethyst> advil: how is it being called now? 11:25:20 Build failed for master @ e4e6ab6f https://github.com/crawl/crawl/actions/runs/105801199 (failed: Webtiles Server) 11:26:21 <|amethyst> advil: crawl-{git,stable}-launcher.sh adds that 11:26:32 ugh really? 11:26:41 thanks, I would never have figured that out 11:27:39 <|amethyst> advil: it looks like the script does that unless you provide one of the -await-connection, -print-charset, and -print-webtiles-options flags 11:27:51 yeah, I'll need to add some stuff to that I guess 11:27:52 <|amethyst> advil: probably you'd want to add the new flag to the list 11:28:26 <|amethyst> advil: look for JUST_RUN_CRAWL_ALREADY 11:29:55 <|amethyst> advil: if you named your new option something starting with -print- you could probably simplify down those existing checks to just [[ "$@" =~ -print- ]] 11:30:07 I ... did not 11:30:23 <|amethyst> advil: I guess you have to do it the cleaner way then 11:32:51 Webtiles server stopped. 11:32:58 Webtiles server started. 11:35:40 Build failed for master @ 03feba32 https://github.com/crawl/crawl/actions/runs/105807496 (failed: Webtiles Server) 11:38:02 Webtiles server stopped. 11:38:06 Webtiles server started. 11:40:59 Webtiles server stopped. 11:41:14 <|amethyst> btw, for 0.25, do we want to change branch_generation_order to put Temple before D? The idea is to spend a small amount of extra time at the beginning of every game to avoid the significantly longer delay of generating the rest of D whenever the player enters T. There's already some precedent in that the temple map and gods are selected first anyway so that we can figure out which overflow altars 11:41:20 <|amethyst> need to be generated 11:42:57 <|amethyst> whether it's an overall increase in work done depends on the percentage of games that end without entering temple, but if it's less than around 80% it should be a gain 11:43:15 Webtiles server started. 11:43:22 <|amethyst> and even if not, it's still a gain from the user's point of view, because the delay when entering the temple is quite noticeable 11:43:54 <|amethyst> s/a gain/a decrease in work/ 11:46:48 Midn8 (L14 DsCK) ASSERT(item.quantity == 1) in 'items.cc' at line 2324 failed. (Lair:6) 11:47:54 Webtiles server stopped. 11:47:57 Webtiles server started. 11:49:11 Webtiles server stopped. 11:49:15 Webtiles server started. 11:52:26 Webtiles server stopped. 11:52:36 Webtiles server started. 11:52:45 Webtiles server stopped. 11:53:08 Webtiles server started. 11:53:48 Webtiles server stopped. 11:53:51 Webtiles server started. 11:55:41 I think I'm done 11:55:41 finally 11:56:30 |amethyst I don't suppose you remember why it is needed to send that control code in that script? 11:57:08 we really need to get better automation for this kind of thing :/ 11:58:59 <|amethyst> advil: I don't remember for sure, but I would guess to make sure that any messages coming from crawl before the screen is initialized aren't dumped on top of the dgl screen 11:59:20 <|amethyst> advil: particularly error messages on startup 11:59:35 ah 11:59:50 I wonder if it is actually needed for webtiles 12:01:09 <|amethyst> advil: it's not, which is why -await-connection turns it off 12:01:09 ohh I see 12:02:30 <|amethyst> advil: webtiles itself does a cls+home in process_handler.py:_ttyrec_id_header() 12:03:12 <|amethyst> (for the ttyrec, as the name might suggest) 12:04:12 <|amethyst> though that might be unrelated? 12:08:23 03advil02 07* 0.25-a0-1099-ga6940e9: Handle a bad dgl-config + webtiles interaction 10(5 minutes ago, 1 file, 5+ 2-) 13https://github.com/crawl/crawl/commit/a6940e959062 12:09:32 !crashlog Mind8 12:09:33 No milestones for Mind8 (crash). 12:09:39 !crashlog Midn8 12:09:40 2. Midn8, XL14 DsCK, T:19099 (milestone): http://crawl.berotato.org/crawl/morgue/Midn8/crash-Midn8-20200515-154638.txt 12:11:21 03advil02 07* 0.25-a0-1100-gb8ef3ea: Add a comment about JSON decoding errors 10(83 seconds ago, 1 file, 5+ 2-) 13https://github.com/crawl/crawl/commit/b8ef3eaf7f55 12:11:53 The build passed. (master - 03feba3 #13393 : advil): https://travis-ci.org/crawl/crawl/builds/687484408 12:17:10 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1100-gb8ef3eaf7f (34) 12:25:05 Build failed for master @ a6940e95 https://github.com/crawl/crawl/actions/runs/105859555 (failed: Webtiles Server) 12:36:55 Build failed for master @ b8ef3eaf https://github.com/crawl/crawl/actions/runs/105861302 (failed: Webtiles Server) 12:40:59 I guess we are doing PEP 8 now? 12:54:13 The build passed. (master - a6940e9 #13394 : advil): https://travis-ci.org/crawl/crawl/builds/687507823 12:56:12 03advil02 07* 0.25-a0-1101-g715d37c: Fix some pep 8 stuff 10(37 seconds ago, 1 file, 10+ 5-) 13https://github.com/crawl/crawl/commit/715d37cef93f 13:02:18 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1101-g715d37cef9 (34) 13:05:48 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-1101-g715d37c (34) 13:13:05 Build failed for master @ 715d37ce https://github.com/crawl/crawl/actions/runs/105891658 (failed: Webtiles Server) 13:14:22 does whoever added this storngly object if I just disable isort is a test that can be failed? 13:14:35 I think I discussed this one with someone in the past 13:17:34 not really reasonable as a linting thing imo, pep 8 can be kind of annoying but at least it is a standard that one could in principle memorize and spot by eye 13:22:22 The build passed. (master - b8ef3ea #13395 : advil): https://travis-ci.org/crawl/crawl/builds/687509230 13:35:04 03advil02 07[stone_soup-0.24] * 0.24.1-12-g7a292d0: Add a CLO that prints save info as json 10(2 days ago, 5 files, 173+ 4-) 13https://github.com/crawl/crawl/commit/7a292d092e67 13:35:04 03advil02 07[stone_soup-0.24] * 0.24.1-13-g45ce301: CLO to give game types in json 10(20 hours ago, 1 file, 49+ 1-) 13https://github.com/crawl/crawl/commit/45ce301c0230 13:35:04 03Alex Jurkiewicz02 {advil} 07[stone_soup-0.24] * 0.24.1-14-gf8d2b50: Add a version of json_mkstring that accepts string 10(1 year, 3 months ago, 2 files, 6+ 0-) 13https://github.com/crawl/crawl/commit/f8d2b509d90b 14:01:46 sophia (L15 GrFi) ASSERT(m) in 'mon-util.cc' at line 1978 failed. (Zot (ZotDef)) 14:04:04 /win 3 14:34:25 oh alexjurkiewicz I should ping you on bec1141c9; basically I needed a way to prevent default dgl-config setups from trying to load the sample .yaml file, but you will need to set that variable to True in CPO's config 14:34:28 %git bec1141c9 14:34:28 07advil02 * 0.25-a0-1094-gbec1141: Disable yaml games by default 10(4 hours ago, 2 files, 4+ 1-) 13https://github.com/crawl/crawl/commit/bec1141c92d9 14:35:29 well, if you're at the point where you can just merge master config.py maybe you wouldn't need to do anything, not sure 14:35:31 since it's True by default relative to master, but False by default if not set 15:17:22 03ebering02 07* 0.25-a0-1102-ga51e346: Allow customizing the travel trail 10(2 hours ago, 5 files, 43+ 9-) 13https://github.com/crawl/crawl/commit/a51e346381c1 15:35:05 Build failed for master @ a51e3463 https://github.com/crawl/crawl/actions/runs/106001531 (failed: Webtiles Server) 15:53:51 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1102-ga51e346381 (34) 16:33:40 03ebering02 07* 0.25-a0-1103-g21c11e7: Use the magic mapping glyph for unvisited stairs and teleporters 10(16 minutes ago, 3 files, 16+ 6-) 13https://github.com/crawl/crawl/commit/21c11e76e5d9 16:47:10 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1103-g21c11e76e5 (34) 17:29:03 03ebering02 07* 0.25-a0-1104-g1bb6b8d: Reveal invisible monsters that fire visible beams at the player (minmay) 10(3 minutes ago, 1 file, 5+ 0-) 13https://github.com/crawl/crawl/commit/1bb6b8da3464 17:32:22 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1104-g1bb6b8da34 (34) 17:46:00 Build failed for master @ 1bb6b8da https://github.com/crawl/crawl/actions/runs/106096768 (failed: Webtiles Server) 18:02:37 regarding python linting: I am not a particular pythonic person but I would like to stop getting fail e-mails for the webtiles ci 18:05:31 hrm, how are you getting those? I'm subscribed to project notification, so I get a lot of emails, but I don't think I get those 18:13:39 I suspect every push since the one that broke linting will send them, regardless of the committer 18:15:55 03advil02 07* 0.25-a0-1105-ga8d2428: Remove isort as a CI check 10(77 seconds ago, 1 file, 0+ 3-) 13https://github.com/crawl/crawl/commit/a8d24287ef72 18:18:08 I do basically also think that exact PEP8 compliance is too strict, there's a ton of specific rules that are not super easy to keep in your head 18:18:34 too strict for a per-push CI check at least 18:19:53 it's something I can put up with, I guess, though I expect to be constantly breaking it accidentally (especially things I find especially pointless, like the two lines surrounding global-level function rule) 18:21:45 also things like "E261 at least two spaces before inline comment" 18:23:23 Unstable branch on underhound.eu updated to: 0.25-a0-1104-g1bb6b8da34 (34) 18:32:08 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1105-ga8d24287ef (34) 18:35:47 03ebering02 07* 0.25-a0-1106-gec9769f: Update the changelog 10(28 seconds ago, 1 file, 53+ 7-) 13https://github.com/crawl/crawl/commit/ec9769f98ae6 18:38:55 Build failed for master @ a8d24287 https://github.com/crawl/crawl/actions/runs/106129556 (failed: Webtiles Server) 18:39:17 I am working on this linting stuff btw, it's just actually kind of non-trivial 18:45:48 03advil02 07* 0.25-a0-1107-gac44191: Fix flake8 errors 10(71 seconds ago, 1 file, 9+ 12-) 13https://github.com/crawl/crawl/commit/ac4419142352 18:46:57 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1106-gec9769f98a (34) 18:50:09 jfc 18:55:08 03advil02 07* 0.25-a0-1108-g7bb5287: Don't disallow % formatting 10(39 seconds ago, 1 file, 0+ 1-) 13https://github.com/crawl/crawl/commit/7bb5287bac9f 18:56:06 03advil02 07* 0.25-a0-1109-g7d3621f: checkwhite 10(69 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/7d3621fadc71 18:57:30 ugh, cancelled too much 18:57:51 I guess you can't restart jobs with gh actions 18:59:40 well, I guess we'll never know if I fixed it 19:02:03 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1109-g7d3621fadc (34) 19:07:23 advil: games.conf.d would have been clearer, but I was using that dir name in webtiles-changes already 19:08:19 thank you for merging the webtiles pr btw advil <3 19:09:28 about "Too many levels of symbolic links" usually it's a symlink pointing to a parent, leading to an infinite loop 19:12:51 about pylint formatting -- your editor might support format on save 19:12:58 vscode gives me underlines of things failing the pylint checks and will fix them on save 19:13:08 +advil | I guess you can't restart jobs with gh actions <-- you can, in the top right 19:13:51 oh, you have to re-run every job individually, you can't re-run the whole workflow 19:13:58 ah, I see it now 19:14:04 yeah, was looking for it somewhere else 19:14:23 huh, clicking on it for one job restarted everything? 19:14:32 heh 19:14:42 the linting that is currently happening now is not something I have so far replicated locally 19:14:52 yeah. GHA is clearly still only a few steps past MVP. There are plenty of rough edges 19:16:52 oh for flake8 linting locally i think i had to copy the stuff manually out of tox.ini into a .flake8.cfg file or something. I threw it away at some point while git resetting and didn't replicate. BTW I'm also happy to see the end of isort. If the basic code formatter doesn't do it, it seems extraneous to me 19:22:27 it would be best to have a linter that is easy to set up locally 19:22:27 Stable (0.24) branch on underhound.eu updated to: 0.24.1-14-gf8d2b509d9 19:23:18 I can run flake8 locally no problem, but getting it to do what CI is doing is kind of involved 19:30:10 Build failed for master @ 7d3621fa https://github.com/crawl/crawl/actions/runs/106148883 (failed: Webtiles Server) 19:31:05 in my recent PR I've removed the "(failed: xxx)" part of these messages. It seemed cool but mostly unnecessary, and made the message much longer. In exchange, I'm adding colour 19:31:18 -!- jfcaron_ is now known as jfcaron 19:31:42 can linting stuff be moved to a separate workflow? 19:32:00 I'm also like this close to disabling flake8 tbh 19:32:22 I think you should chat with aidanh about that, it seems like a dev team decision 19:32:32 it wasn't a dev team decision to enable it 19:32:44 so I don't feel too bad about making a devteam decision to disable it 19:32:51 *a non-devteam 19:33:27 I do like linting as part of the CI run, but I agree the current linting setup is hard to replicate locally. How about replacing it with checking only the code meets `black` formatting? That's very easy to replicate locally 19:34:08 maybe something like flake8 without any plugins except maybe mypy? 19:35:26 btw I *also* am unable to regenerate these requirement.txt files 19:35:27 tox makes it hard to replicate linters in your editor. Let me clean up this PR to get `make test` working in gha and then look into it 19:35:49 which is just something about how my py2 is set up I think 19:36:05 hm, it worked for me, but I haven't looked at the internals 19:40:15 love these ancient undocumented util perl scripts 19:40:21 how and why does util/gather_mons work?? 19:41:08 alexjurkiewicz: it's for CDO bot info lookup 19:41:34 I mean I know it's used as part of the monster compilation. But when I run it locally it includes junk output and errors 19:41:43 that's the why; the how is just that it's gathering info for vault-defined monsters after building monster 19:41:48 ❯ util/gather_mons | head 19:41:51 clang: error: no input files 19:41:53 Cigotuvi's Monster 19:41:55 Cloud Mage LUA 19:42:05 it definitely works if you have proper build flags; it's what CDO uses 19:42:35 it's possible it doesn't work with clang 19:42:44 but CDO is the consumer of that script, so I wouldn't worry about it 19:43:40 and to be clear, it's not actually used for monster compilation or necessary for a working copy of monster 19:43:52 it's something needed specifically by the CDO web infrastructure 19:44:07 it's used by Cheibriados, right? 19:44:16 how would Cheibriados use it? 19:44:25 %??orb guardian 19:44:25 Orb Guardian (06X) | Spd: 14 | HD: 15 | HP: 66-95 | AC/EV: 13/13 | Dam: 45 | 10doors, fighter, see invisible | Res: 06magic(120), 12drown | XP: 2769 | Sz: Giant | Int: human. 19:44:25 it's not part of monster proper 19:44:40 sorry, i might be unclear 19:44:42 it's assembling information about monster *tiles* 19:44:50 `util/gather_mons -v` creates input for util/monster/monster 19:44:55 specifically tiles for vault-defined monsters 19:44:58 which is the underlying tool Cheibriados uses for %?? lookups, right? 19:46:03 no, not needed for monster 19:46:12 not saying there wasn't some historical use for it 19:46:23 in monster proper, that is 19:46:29 right now we use it for getting that tile data 19:46:43 hmmmmmm 19:47:48 looks like the monster code is specifically designed to ignore junk data in the output of gather_mons 19:47:48 oh maybe it does need to be there for vault monster data for monster 19:47:55 !source _get_vault_monster 19:47:56 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/util/monster/monster-main.cc#L601 19:50:06 this pr now does: 1) enable make test in GitHub Actions (for TAG_MAJOR_VERSION 34) 2) remove travis CI 3) fix some issues I found while trying to run `make test` with TAG_MAJOR_VERSION 35 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1110-g740f4b2: Remove Travis CI 10(2 days ago, 3 files, 0+ 242-) 13https://github.com/crawl/crawl/commit/740f4b226498 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1111-ge84bb98: Revert "Temporarily disable make test test pending further ... testing" 10(2 days ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/e84bb98a3dd6 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1112-gb8f4c76: Set TERM for Lua tests 10(2 days ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/b8f4c76dae4d 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1113-g3b7d271: Remove references to Forest and other removed things 10(2 days ago, 8 files, 33+ 33-) 13https://github.com/crawl/crawl/commit/3b7d27121770 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1114-g935d994: Fix Pan entry placement in TAG_MAJOR_VERSION 35 10(2 days ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/935d994903a5 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1115-g6a795ba: Install GDB for FULLDEBUG runs 10(2 days ago, 2 files, 6+ 4-) 13https://github.com/crawl/crawl/commit/6a795baf71c6 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1116-g9504156: Simplify failure reporting. 10(2 days ago, 1 file, 8+ 31-) 13https://github.com/crawl/crawl/commit/950415621249 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1117-ge72626b: Print compiler version 10(2 days ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/e72626bf3f68 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1118-gfc4fd79: Pull out ccache setup to an external action 10(2 days ago, 1 file, 8+ 72-) 13https://github.com/crawl/crawl/commit/fc4fd7900b42 19:50:58 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1119-ga41de85: Remove two obsolete mutations from TAG 35 10(2 days ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/a41de85b8d97 19:50:58 ... and 4 more commits 19:52:43 advil: the issue with logging.warn("JSON error with output '%s'" % repr(m_json)) is not the % formatting, per se. They expect you to write logging.warn("JSON error with output '%r'", m_json) instead 19:53:06 the idea being that the latter form doesn't do unnecessary work if the logging level is configured to drop the line anyway 19:54:11 ` ...%s", repr(m_json)` would have also silence the linter, the linter isn't smart enough to pick up %s/%r differences 19:55:25 ah hm 19:56:26 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1124-gd0d48dc: Tweak a logging line 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/d0d48dc2ac5b 19:56:48 03advil02 07* 0.25-a0-1110-g22b210b: Another linting fix 10(44 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/22b210b1cc9e 19:58:32 alexjurkiewicz: "!learn del travis[2] 19:58:42 alexjurkiewicz: you musn't! 19:59:13 alexjurkiewicz absolutely will not stop, ever, until travis[2] is dead. 19:59:40 :thumbsupinlavaemoji: 20:02:09 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1110-g22b210b1cc (34) 20:02:36 advil: also, you may have won the commit count this time, what with your webtiles fixes and and your linting commits, but know that I shall rise again 20:02:54 brb going to tweak all arrival vaults as individual commits 20:04:17 believe me, I'd rather not be committing all these linting fixes 20:05:13 a fun prelude to working on tournament scripts 20:05:39 luckily, the tournament/scoring scripts conform to no python standards whatsoever 20:06:46 heh, right 20:07:08 I like to refer to the tournament scripts as Perlthon 20:08:04 not really sure I'm ready to convert my python bots over to all this type checking and linting etc 20:08:26 have converted them to python 3.7+ async syntax since this was required for newer discord module, but that's all for now 20:14:42 I don't understand how I keep messing up checkwhite 20:15:48 03advil02 07* 0.25-a0-1111-gf6331bb: checkwhite (again??) 10(39 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/f6331bbc0c58 20:16:10 it appears that the webserver is finally linting though 20:18:21 advil: aidanh: I'm going to submit a PR that replaces flake8 with individual calls to black, pylint and mypy, and enables them for all files. It also pulls the config for each into individual files rather than tox.ini, which should make it possible to run the same linting config in-editor 20:20:56 bonus points if you can make a git pre-commit-hook script that people can use like they do unbrace/checkwhite 20:22:19 would also appreciate it if anyone can tell me if util/checkcommit is needed; are there checks in git now or elsewhere for partial merge lines (i.e. >>>>>>> and ======= and <<<<<< and stuff in between) in commits? 20:24:41 ohhh 20:24:57 ok so in fairness some of my complaints are me be dumb 20:25:11 most files are whitelisted, but load_games.py was new 20:25:16 *me being 20:26:03 so if I'd known how this was set up I could have short-circuited a lot of pain just by adding it to the whitelist in tox.ini 20:29:00 Build failed for master @ 22b210b1 https://github.com/crawl/crawl/actions/runs/106180978 (failed: Linting) 20:31:34 so I'm much more on board with a linting approach that whitelists everything and then maybe gradually removes things from the whitelist, which seems to have been the original intent :-) 20:32:33 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-1111-gf6331bbc0c (34) 20:41:55 The build was broken. (master - 22b210b #13406 : advil): https://travis-ci.org/crawl/crawl/builds/687649900 20:44:35 I'm not too bothered if import linting goes, the impact for python is minimal 20:45:17 isn't there a 'make lint' rule though for local testing? 20:46:25 well, the intent was that new files would be conformant from the beginning 20:47:26 alexjurkiewicz: that sounds like a bit of a step backwards imho 20:48:27 the issue with `make lint` is it can't run in your editor 20:55:55 clearly everyone needs to start using vim 20:56:07 :) 20:57:00 pep8 is the standard; what more is required for other editors to work with that? 20:57:41 flake8 has a bunch of custom config and excludes in tox.ini 20:58:27 is there a more standard location for linter config? 20:59:54 i'd prefer to have centralised config; it makes it easier to have consistency 21:00:17 if you mean centralised as in every file uses the same config, 100% agree 21:00:46 It would be nice for all Python lint tools to use a single file to store their config, but I don't think the ecosystem is quite there yet 21:01:35 the python ecosystem is a tyre fire :/ 21:02:13 yes, especially around this stuff. It's all ramshackle wrapper on ramshackle tool on ramshackle packaging system 21:03:28 are you aiming for linting or formatting? 21:03:42 -!- EuphOria is now known as Euph0ria 21:03:49 I'm just writing an explanation of what I'd like now! 21:03:51 > Change the CI process to run a smaller set of more popularly supported tools: pylint and mypy for linting, black for formatting. 21:03:59 I would be happy with a presubmit check for linting (which is what I have now) 21:04:04 s/popularly/widely/ 21:04:54 s/presubmit check/prepush git hook/ 21:05:24 aidanh: I didn't even realize that tyre was the british spelling of tire :) 21:06:02 the python tyre is covered in petrol 21:06:56 wait, there's an American in our crawl channel!? 21:07:53 if anyone has a spare minute or several 21:07:58 i wrote a thing 21:08:05 and some eyes sanity checking it would be nice 21:08:27 https://github.com/crawl/crawl/pull/1409 21:08:32 aidanh: I'm just an ordinary bloke, don't mind me 21:09:08 Implojin: is this to replace boots of running? 21:09:25 well 21:09:27 yes 21:09:34 but i didn't want to go that far 21:09:39 that just adds the new sparm 21:09:51 i thought you'd want to talk amongst yourselves re: running 21:09:58 hrm, guess I'm not clear from commit message how any limitation occurs 21:10:01 aidanh: advil: I wrote this: https://github.com/crawl/crawl/pull/1410 it has a tl;dr at the top. For me, it would make code style conformity much easier 21:10:09 is there a cooldown? 21:10:22 I'm heading out now but I'll be around later for feedback 21:10:38 New branch created: pull/1409 (1 commit) 13https://github.com/crawl/crawl/pull/1409 21:10:38 03Implojin02 07https://github.com/crawl/crawl/pull/1409 * 0.25-a0-1058-g3d98f8c: New ego: Boots of lunging. 10(10 minutes ago, 27 files, 323+ 6-) 13https://github.com/crawl/crawl/commit/3d98f8cfb805 21:10:38 New branch created: pull/1410 (9 commits) 13https://github.com/crawl/crawl/pull/1410 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1110-gaab9244: Format all webserver code with black 10(69 minutes ago, 15 files, 900+ 578-) 13https://github.com/crawl/crawl/commit/aab9244887da 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1111-gf938986: Add black to py3 dependencies 10(55 minutes ago, 5 files, 52+ 45-) 13https://github.com/crawl/crawl/commit/f938986f0f8a 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1112-g9e7c69c: Update webserver test commands 10(55 minutes ago, 1 file, 3+ 18-) 13https://github.com/crawl/crawl/commit/9e7c69c29b95 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1113-g994c2f7: Remove webserver Makefile 10(55 minutes ago, 1 file, 0+ 9-) 13https://github.com/crawl/crawl/commit/994c2f714367 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1114-gf2568a0: Add initial pylintrc file 10(42 minutes ago, 1 file, 595+ 0-) 13https://github.com/crawl/crawl/commit/f2568a027228 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1115-g055df36: Add project-specific pylint excludes 10(21 minutes ago, 1 file, 10+ 1-) 13https://github.com/crawl/crawl/commit/055df362b669 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1116-gbf0bd4e: Run pylint as part of tox 10(20 minutes ago, 4 files, 30+ 60-) 13https://github.com/crawl/crawl/commit/bf0bd4e19788 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1117-g01a49f5: Add pylint disables to ws_handler.py 10(19 minutes ago, 1 file, 7+ 0-) 13https://github.com/crawl/crawl/commit/01a49f59bbe6 21:10:39 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1410 * 0.25-a0-1118-gb8090e4: Relax the quality threshold for pylint passing in CI 10(15 minutes ago, 1 file, 5+ 1-) 13https://github.com/crawl/crawl/commit/b8090e4b2f0b 21:10:50 the limitation is that moving towards monsters wastes turns you could be using for better things 21:12:01 well, can I just repeat my lunge over and over until I'm adjacent? 21:12:13 yes 21:12:32 it only grants one free tile of movement per turn: 21:12:53 the normal attack / move action still happens as usual 21:13:00 it's not, like, infinite jumping 21:13:23 oh ok 21:13:24 another way to think about it could be like "boots of .5 movedelay but only when moving directly towards a hostile" 21:13:33 so you have to act and "consume" the turn still 21:13:41 with some non-lunge action 21:13:41 except that i didn't write it that way because i explicitly didn't want players to have to worry about energy juggling 21:13:48 yes 21:13:59 in fact, there is no way to activate a lunge except with a standard movement input 21:14:05 zero extra keypresses 21:14:16 ok, that's kind of interesting 21:14:18 zero manual targeting. 21:14:49 number one question that must be answered: how good would this be in meatsprint 21:15:03 the boots grant a perma-status as boots of flying, can be disabled with an ability evoke 21:15:10 i have no idea how it will play in sprints 21:15:16 i just assumed you all had removed sprints already 21:15:21 fair 21:16:07 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1112-g2d4bdeb: Remove Travis CI 10(2 days ago, 3 files, 0+ 242-) 13https://github.com/crawl/crawl/commit/2d4bdeb97cba 21:16:07 03Alex Jurkiewicz02 07https://github.com/crawl/crawl/pull/1404 * 0.25-a0-1113-g948017e: Revert "Temporarily disable make test test pending further ... testing" 10(2 days ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/948017e130a9 21:16:07