00:15:28 -!- mikee___ is now known as mikee_ 01:25:27 what's up with this bailey entrance that needs flying/swimming? 01:26:44 i can't even dig in, stone walls instead of rock 01:31:08 Can't sacrifice corpses to Lugonu and Trog for piety 13https://crawl.develz.org/mantis/view.php?id=11265 by njwelc 01:32:32 heh 02:45:48 Monster database of master branch on crawl.develz.org updated to: 0.21-a0-401-g0b6bec4 03:02:34 -!- Bammboobies is now known as Bammboo 03:38:35 -!- amalloy is now known as amalloy_ 03:42:49 -!- amalloy_ is now known as amalloy 04:09:48 advil: all in the name of better gameplay 04:13:42 For webtiles dev, is there any way to a) auto-login, and b) kill any stale processes without waiting ten seconds? 04:17:24 aidanh: you can remove the file in rcs/running 04:17:42 that seems to be what ties it up 04:18:04 for auto-login, just click the button to save your login credentials 04:18:30 er click the checkbox on the webtiles lobby page 04:25:16 gammafunk: thanks, removing that file works; stay-logged-in doesn't seem to be working though 04:25:35 but it's not a huge deal; the stale processes interstitial was taking way longer 04:29:02 -!- amalloy is now known as amalloy_ 04:42:30 Is there any way to use separate build directories for different flags? So that I don't have to rebuild everything whenever I change target? 05:01:15 I think the way the makefile works, it will detect it as a argument change and rebuild, even if you specify rebuild dirs 05:01:25 but can't you just have a few different clones of the repo? 05:01:46 if you need commits in your local branch, you can make the repo with the commits be a remote in those other repos 05:02:16 so that you just have to git pull from the "origin" remote and then build 05:02:31 that's how I share commits between the different crawl repos I have locally 05:02:36 aidanh: ^ 05:03:58 Hmm.. I guess I could, but that seems a bit heavyweight 05:05:00 oh, are you using ccache? 05:05:19 ccache can make those rebuild very fast, at least 05:05:34 does the makefile detect it automatically? 05:05:36 doesn't speed up linking and it's not instantaneous, but it's good to have regardless 05:05:53 detect ccache? the makefile doesn't know about ccache 05:06:05 ccache gets added to the front of your path 05:06:23 and calls to gcc et all are through ccache 05:06:25 Ah, the g++ wrapper binaries? 05:07:06 yeah, it looks at a hash of the final source files after running through the preprocessor 05:07:51 and if it matches the hash corresponding to a binary it has in its cache (from a previous time when you built that exact file) 05:08:05 it simply copies that object file instead of compiling it 05:08:21 it will speed up compilation even when not changing branches or make args 05:08:50 although I guess those are the situations where the speedup is the most 05:08:54 anyhow most of us use ccache 05:10:31 Ok, thanks; enabled it 05:20:13 For menus, is it safe to assume that spans of <=52 items never have duplicated hotkeys, and >52 items always have them? 05:26:56 Excluding titles/subtitles/end_of_sections, that is 08:29:28 New branch created: pull/632 (7 commits) 13https://github.com/crawl/crawl/pull/632 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-402-ge212408: Remove menu.set_maxpagesize interface 10(2 days ago, 7 files, 3+ 38-) 13https://github.com/crawl/crawl/commit/e212408cdc92 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-403-ga4e6a71: Assert that menus have a valid selection flag 10(2 days ago, 2 files, 16+ 3-) 13https://github.com/crawl/crawl/commit/a4e6a719459b 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-404-gbadc0a6: Remove unused menu.f_drawitem hook 10(2 days ago, 2 files, 2+ 7-) 13https://github.com/crawl/crawl/commit/badc0a66e92e 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-405-gd874c07: Remove column_compoer set_pagesize() 10(2 days ago, 4 files, 2+ 20-) 13https://github.com/crawl/crawl/commit/d874c074705d 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-406-gbf147df: Fix webtiles section boundaries not being updated 10(20 hours ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/bf147df3c02f 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-407-g2ab32c8: Show branch tiles in lookup-help menu 10(19 hours ago, 3 files, 97+ 2-) 13https://github.com/crawl/crawl/commit/2ab32c8afa75 08:29:28 03Aidan Holm02 07https://github.com/crawl/crawl/pull/632 * 0.21-a0-408-gad50db3: Refactor Menu scrolling into MenuDisplay 10(2 hours ago, 5 files, 448+ 313-) 13https://github.com/crawl/crawl/commit/ad50db3e3741 09:41:51 -!- Guest82109 is now known as bh 09:55:10 is there any policy on PRs with a bunch of misc. fixes? 09:58:23 New branch created: pull/633 (1 commit) 13https://github.com/crawl/crawl/pull/633 09:58:23 03Aidan Holm02 07https://github.com/crawl/crawl/pull/633 * 0.21-a0-402-ga02d124: Fix title screen lingering when tile_skip_title = true 10(7 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/a02d124f828d 09:58:54 I don't know if there's a policy but it'd probably be easier to merge if any given PR is thematically coherent 09:59:31 if a sequence of commits doesn't have dependencies to earlier commits in the sequence, might be a candidate for breaking into multiple PRs 10:01:56 I ask because I'm discovering a lot of small fixes that are each a handful of lines 10:02:13 But they aren't necessarily related to each other 10:03:21 yeah...either way is fine I think, but if each of those is a single bitesize PR then it's really easy for someone to merge 10:06:42 alright, well, i'll decide case-by-case 10:07:03 if it's a bigger PR it probably means that a single dev will need to take a bit more time to look through everything carefully before merging the whole thing 10:07:33 but that's just from the merging side, it's kind of annoying to have a million tiny little branches 10:07:46 from the author side, that is 10:09:03 That, and I imagined it'd be easier to tab through a set of commits and then hit only a single merge button; if it's obvious that they're small/independent/etc 10:14:59 well, the dev who merges it takes some responsibility for correctness of all of them :) 10:19:02 of course 13:21:34 how do I filter sequell by date - e.g. filter by games in last 30 days? Thinking its using the 'start' keyword but not sure of the syntax 13:25:27 if you want games started in the last 30 days, you can do: 13:25:40 !lg * start>${now()-interval('30d')} 13:25:41 131792. SublimeSnake the Grave Robber (L2 MuNe), slain by a jackal on D:2 on 2017-10-22 17:22:53, with 20 points after 716 turns and 0:03:08. 13:27:10 thanks - is that postgres syntax? 13:27:48 I think some of sequell is in postgres but some is just sequellese 13:27:51 ??sequellese 13:27:51 sequellese[1/1]: https://github.com/crawl/sequell/blob/master/docs/commandline.md <|amethyst> It's like an unholy combination of Lisp and shell 13:28:01 is some of the documentation 13:28:55 ye, reading through the documentation at the moment - very impressed by how detailed it is 13:29:15 is it written purely for dcss or does it share code with other projects? 13:30:08 I think greens.nark basically wrote it all for this 13:30:43 !lg won s=species% start>${now()-interval('30d')} 13:30:44 74 games for won (start>${now()-interval('30d')}): 20x Formicid (27.03%), 17x Deep Dwarf (22.97%), 11x Octopode (14.86%), 10x Ogre (13.51%), 3x Deep Elf (4.05%), 3x Tengu (4.05%), 2x Demonspawn (2.70%), 2x Human (2.70%), 2x Gnoll (2.70%), 2x Ghoul (2.70%), Minotaur (1.35%), Vine Stalker (1.35%) 13:31:04 interesting to see that Formicid is taking ~ 1/3 of the wins 13:31:14 !lg * won s=species% start>${now()-interval('30d')} 13:31:15 907 games for * (won start>${now()-interval('30d')}): 141x Minotaur (15.55%), 136x Gnoll (14.99%), 82x Gargoyle (9.04%), 80x Demonspawn (8.82%), 45x Hill Orc (4.96%), 40x Vine Stalker (4.41%), 39x Deep Dwarf (4.30%), 37x Centaur (4.08%), 31x Formicid (3.42%), 31x Troll (3.42%), 28x Barachi (3.09%), 25x Deep Elf (2.76%), 19x Merfolk (2.09%), 18x Felid (1.98%), 18x Ogre (1.98%), 16x Naga (1.76%), 15... 13:31:26 that query was not doing what you wanted 13:31:45 ah, thought i had something wrong :) 13:31:46 !lg won 13:31:47 2107. won the Invulnerable (L27 MiBe of Cheibriados), blasted by a Tzitzimitl (dispel undead) (called by a doom hound) on Tar:1 on 2017-10-21 17:12:53, with 1125485 points after 91062 turns and 7:19:49. 13:31:56 it was doing the player named "won" instead of games that were won :) 13:32:54 ok, so it was just telling me that 'won' is very good with Formicid characters - that makes much more sense 13:33:53 well, actually just that 'won' has played a lot of formicids recently 13:34:01 nothing about how good they are with them 13:34:22 !lg won won s=species% start>${now()-interval('30d')} 13:34:23 One game for won (won start>${now()-interval('30d')}): Deep Dwarf (100.00%) 13:34:34 indeed 14:47:33 -!- amalloy_ is now known as amalloy 15:13:28 was there a decision on the purple chunk situation? I was thinking of getting back to that issue, but if purple chunks are leaving, I'll just submit the PR with the docstring updates 16:30:02 -!- Kittykai_ is now known as Kittykai 18:04:53 johnstein: it sounded to me like the best approach would be to remove purple chunks and maybe do some reforms to some classes of mutations 22:22:28 -!- tat is now known as Guest38639 22:23:00 alexjurkiewicz: please update when you have a chance