03:34:02 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-1398-g75a3a2ffe1 04:04:00 Skull (L6 GrEE) ERROR in 'bitary.h' at line 61: bit vector range error: 160 / 16 (D:4) 08:32:13 Reading a Scroll of Holy Word Does Not Generate Warning 13https://crawl.develz.org/mantis/view.php?id=11821 by Malacante 08:52:08 03ebering02 07* 0.23-a0-682-g68c7d63: Display real MP for Deep Dwarves when necessary (Ge0ff) 10(7 minutes ago, 5 files, 28+ 4-) 13https://github.com/crawl/crawl/commit/68c7d63de04a 08:52:08 03ebering02 07* 0.23-a0-683-g80e34f1: Fix coloring for current MP in Webtiles (Ge0ff) 10(7 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/80e34f145d62 08:52:08 03ebering02 07* 0.23-a0-684-gb3bac08: More permanent MP display (Ge0ff) 10(6 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/b3bac08b9776 08:52:08 03ebering02 07* 0.23-a0-685-gfe0cbb1: Add Ge0ff to CREDITS 10(5 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/fe0cbb1ec109 09:10:40 Unstable branch on crawl.kelbi.org updated to: 0.23-a0-685-gfe0cbb1ec1 (34) 09:21:02 is there any good reason to truncate message lines at 80 chars for #? 09:21:59 I guess the output of that mostly does obey 80 chars 12:15:40 Unstable branch on crawl.akrasiac.org updated to: 0.23-a0-685-gfe0cbb1 (34) 12:50:22 03Umer Shaikh02 07https://github.com/crawl/crawl/pull/936 * 0.23-a0-651-g0610bdb: Fix information leak for invisible monsters (11815) 10(6 days ago, 1 file, 8+ 3-) 13https://github.com/crawl/crawl/commit/0610bdbdffce 12:54:01 Skull (L3 GrEE) ERROR in 'bitary.h' at line 61: bit vector range error: 96 / 16 (D:2) 13:15:34 03Umer Shaikh02 {ebering} 07* 0.23-a0-686-g0964cf1: Fix information leak for invisible monsters (11815) 10(6 days ago, 1 file, 8+ 3-) 13https://github.com/crawl/crawl/commit/0964cf18199f 13:15:46 !bug 11815 13:15:46 https://crawl.develz.org/mantis/view.php?id=11815 13:21:03 !bug 11283 13:21:03 https://crawl.develz.org/mantis/view.php?id=11283 13:22:17 advil: the behavior described here seems to match 11780, I wonder if your fix also fixes this one? 13:22:31 I've added a mantis relationship and asked for feedback on the github issue 13:29:05 !bug 11431 13:29:06 https://crawl.develz.org/mantis/view.php?id=11431 13:32:03 Unstable branch on crawl.kelbi.org updated to: 0.23-a0-686-g0964cf1819 (34) 13:38:19 03Jacob Meigs02 {ebering} 07* 0.23-a0-687-gb50ad93: Decouple hi-score logic from ouch (11431) 10(9 months ago, 5 files, 85+ 51-) 13https://github.com/crawl/crawl/commit/b50ad93e3a5f 14:02:04 Unstable branch on crawl.kelbi.org updated to: 0.23-a0-687-gb50ad93e3a (34) 14:14:05 ebering yeah, it could well result from out-of-order water tile updates 14:14:37 there were a bunch of other things I noticed that didn't have visible consequences but were getting errors in chrome console 14:15:42 so who knows how long rare out-of-order issues have been happening 14:15:50 sometimes it just won't matter much, I think 14:56:44 this "bit vector range error" crash is possibly the craziest crawlcode bug I have worked on 14:56:49 which is saying quite a lot 14:57:35 Phial of Floods not marked as dangerous_item if sacrificed Love with Ru 13https://crawl.develz.org/mantis/view.php?id=11822 by kurzedmetal 14:59:45 the short version is that `show_update_at` is not reentrant and can cause memory corruption if it ends up recursing, which only sounds moderately crazy, but it's the stack trace leading to this situation that is insane 15:00:30 also the process involved in figuring this out 15:03:16 !source show_update_at 15:03:17 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/show.cc#L492 15:06:47 because the first thing it does is that clear_data() call, so anything higher in the stack trace that had an item set during the first call is now invalid 15:07:08 granted to operationalize this into memory corruption you need a *very specific* way of calling it 15:07:48 basically, a buggy autopickup lua script 15:09:40 well, anything that triggers a runrest activity interrupt can do it in principle, but most of those code paths don't do anything with the invalid item reference 15:10:41 oh maybe it would have to be an interrupt due to autopickup in some form, to set this up 15:17:13 it's hard to know how to fix this properly 15:17:24 one idea would be to just check for recursion in viewwindow 15:17:59 gammafunk: I'll probably land the zot-ification of net traps tonight. I thought about making them require lof and throw the net at the nearest player-aligned monster (not checking if the path is clear) but that seems very complicated for not much gain 15:19:04 ebering: sounds good, I can make sure to include net traps in my set of traps that hall_of_zot uses 15:19:20 excellent 15:19:31 I also have a review of ossuaries and tomb ready to go 15:20:03 only the ossuaries retain mechanical traps and those could maybe go away 15:20:14 could this be the version that sees the removal of mechanical traps??? 15:20:50 yeah, I was thinking about tomb; it seemed like just increasing the number of meaningful (non-mechanical) traps and removing the mechanical ones would be fine 15:20:53 is that what you did for tomb? 15:21:00 basically yes 15:21:18 also removed some of the placements that don't make sense even for meaningful traps 15:22:12 oh, and I kept disperal traps out of single tile corridors for sanity's sake 15:38:29 The build was broken. (master - b50ad93 #10814 : Jacob Meigs): https://travis-ci.org/crawl/crawl/builds/477968515 15:40:12 lol 15:40:12 I warned you 15:40:37 :) you did 15:40:47 I've actually been planning on doing that with a different CI so we aren't getting "The build was broken." on *every single PR* 15:40:55 you just beat me to merging it 15:41:02 probably circle CI? 15:41:44 Yeah, I don't know much about different CI options, I've just got itchy PR closing fingers 15:42:20 that one was my bad though, I rebased it locally and didn't unbrace and checkwhite 15:42:24 heh 15:42:30 because I somehow broke my pre-commit hooks 15:43:21 03ebering02 07* 0.23-a0-688-ge64fb27: unbrace and checkwhite 10(36 seconds ago, 2 files, 4+ 6-) 13https://github.com/crawl/crawl/commit/e64fb2763ee8 15:44:48 also, hot take, brogue has way better nets than we do 15:45:29 how do brogue nets work 15:47:15 the trap sprays a cloud of nets around, which are then terrain features that trap anyone in them or who later steps on them 15:47:18 the nets themselves break after a deterministic number of steps shown by a nice bar (I've been playing a lot of brogue lately and thinking about our ui inadequacies) 15:47:22 Unstable branch on crawl.kelbi.org updated to: 0.23-a0-688-ge64fb2763e (34) 15:47:36 oh yeah, I've played some brogue 15:48:10 not really intending to say that brogue nets would work wholesale in crawl, though maybe they could 15:48:58 but at least within the context of brogue, they're a very good design imo 15:49:55 probably the deterministic breaking ui could be a good fit, though (sort of like the changes to poison back in the day) 15:50:10 yeah; the aoe less so 15:50:25 also the fire interaction 15:50:32 I remember you can burn yourself out of a net in brogue 15:50:39 ah right, they do burn 15:50:54 or burn something else that's in one, or kill yourself because fire spreads across the net 15:51:06 yes 16:09:57 New branch created: pull/939 (1 commit) 13https://github.com/crawl/crawl/pull/939 16:09:57 03mgdelmonte02 07https://github.com/crawl/crawl/pull/939 * 0.23-a0-678-g96d62d8: added lua spells.path function 10(23 minutes ago, 1 file, 61+ 0-) 13https://github.com/crawl/crawl/commit/96d62d89866c 16:14:56 I'm not sure where that UI would fit; would you have to x-v each tile to see each tiles independent break chance or something? 16:15:37 or is that a UI you get when you become netted 16:18:20 the ui when you become netted 16:18:50 if I were working on crawl I would also be rewriting nets to not be items 16:18:56 er, nets on actors 16:19:15 more clearly: rewriting netted status to not involve an item 16:19:27 and having nets always mulch 16:19:30 sorry, but fake news 16:19:35 you have a +, you are working on crawl! 16:19:39 haha 16:19:51 I have been fixing bugs again 16:20:35 so I guess I'm not 100% retired yet 16:20:51 that's good, many of those are harder that most of the game design stuff 16:21:05 s/that/than/ 17:11:16 03advil02 07* 0.23-a0-689-g3016822: Don't chop messages at 80 chars for chardumps/crashes 10(8 hours ago, 1 file, 4+ 2-) 13https://github.com/crawl/crawl/commit/30168226961f 17:11:16 03advil02 07* 0.23-a0-690-g0a1fc62: Prevent recursion for `viewwindow` 10(22 minutes ago, 1 file, 90+ 69-) 13https://github.com/crawl/crawl/commit/0a1fc62c7c50 17:17:11 Unstable branch on crawl.kelbi.org updated to: 0.23-a0-690-g0a1fc62c7c (34) 17:47:44 -!- Tiobot is now known as Guest86322 18:21:37 Unstable branch on underhound.eu updated to: 0.23-a0-690-g0a1fc62c7c (34) 18:29:54 i can't figure out how to debug this lua 18:29:56 there has to be a way 19:12:15 -!- switch___ is now known as switch__ 21:04:15 03mgdelmonte02 07https://github.com/crawl/crawl/pull/939 * 0.23-a0-679-ge7882f0: fixed coding style 10(2 minutes ago, 1 file, 3+ 2-) 13https://github.com/crawl/crawl/commit/e7882f0a41b7 21:11:18 advil: while you warned me, I'd say we're 1-1 now :) 21:44:27 heh, looks like that was NP7 telling them, not travis ;-) 22:34:09 03mgdelmonte02 07https://github.com/crawl/crawl/pull/939 * 0.23-a0-680-g40aa3a6: added optional src coordinates -- allows determining path of spells cast from other locations 10(37 seconds ago, 1 file, 11+ 4-) 13https://github.com/crawl/crawl/commit/40aa3a6727b8 22:56:23 hrm, I wonder if firing an actual beam like that could have side-effects 22:59:23 also stuffing it into a char array like that is probably a bad idea 23:04:41 oh and I guess it should actually return data, not a string; an array of integer pairs probably