00:00:40 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.24-a0-254-gf493760 (34) 00:42:11 -!- turin is now known as thurin 00:51:51 -!- dexap is now known as paxed 01:41:14 advil: I've tested that icon change on Windows and it does indeed work :) 03:08:39 New branch created: pull/1042 (1 commit) 13https://github.com/crawl/crawl/pull/1042 03:08:39 03Joshua Gelbard02 07https://github.com/crawl/crawl/pull/1042 * 0.24-a0-255-gb49bde3: Fix a couple Engulf-related bugs 10(10 hours ago, 6 files, 20+ 36-) 13https://github.com/crawl/crawl/commit/b49bde37e4f5 04:39:16 -!- amalloy is now known as amalloy_ 09:16:56 alexjurk1 yeah, they're all extremely rube goldberg-y 09:17:08 also there may or may not be an infinite supply of them 09:18:07 gotm 2 seems a lot more stable but there's something weird going on with randart names that I can't figure out 11:22:08 Stable (0.23) branch on underhound.eu updated to: 0.23.1-50-ga50397a6bc 13:05:37 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-254-gf493760 (34) 14:01:42 -!- amalloy_ is now known as amalloy 14:04:18 advil: i saw https://fuglede.dk/en/blog/when-redundant-type-casts-matter/ on reddit yesterday and thought of your ordeal 14:17:43 oh heh 14:17:43 that looks worth reading 14:17:44 even if it's not c++ 14:23:54 looks like that's a whole series 14:24:21 https://fuglede.dk/en/blog/rounding-errors-and-compilers/ is also relevant 14:24:30 "No matter how careful one might be, compilers will find a way to make it all fall apart." 14:33:20 this isn't all the compiler, it's partly how the cpu behaves. standard store operations drop floating point precision 14:33:46 i wondered if the same problem occurs in java, which i use at work. it reminded me that strictfp exists as a keyword in java, and i have literally never seen it used in my life 14:34:05 gcc has some special fp types that tell it to do extra work to preserve the precision (e.g. long double) 14:34:42 but it's standard (and going back to x87) that intel floating point uses extra precision internally and drops it wnhen storing to memory 15:13:38 -!- MarvinPA_ is now known as MarvinPA 15:59:42 03nlavsky02 {MarvinPA} 07* 0.24-a0-255-g1d4c5a4: Fix local tiles monster panel showing summoner halo for all enemies (#1038) 10(8 seconds ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/1d4c5a47d16d 16:16:55 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-255-g1d4c5a47d1 (34) 16:39:36 03MarvinPA02 07* 0.24-a0-256-g86c1934: Fix a message colour entry 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/86c19343c97c 16:47:25 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-256-g86c19343c9 (34) 16:53:29 03kitchen-ace02 {MarvinPA} 07* 0.24-a0-257-g927fd9e: Minor changes to default force_more_messages (#1037) 10(11 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/927fd9e6cbee 17:00:30 03MarvinPA02 07* 0.24-a0-258-gb86c0f7: Clean up more default force_more_messages 10(2 minutes ago, 1 file, 4+ 4-) 13https://github.com/crawl/crawl/commit/b86c0f735e93 17:02:21 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-257-g927fd9e6cb (34) 17:17:29 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-258-gb86c0f735e (34) 17:59:01 in save_compatibility.txt, it says that if you're adding anything to the player class, you should consider storing it in the props hash, since that won't create save compat issues 18:00:04 then there's an example about what it looks like to not do that, where we walk through adding how to add a field you.xyzzy and "save it without breaking saves" 18:02:06 if player.xyzzy didn't need to be saved and restored, would we still have to do everything that's walked through there? 18:03:45 (and if not, would it still be better to use you.props anyway for other reasons?) 18:06:23 if you.xyzzy is transient information that is only relevant within the turn you can use a new field, but make sure it gets a sensible default on initialization 18:06:24 i ask because i have an old pull request that adds a "you.last_unequip" field (and aliases it to ';' in menus to make some in-game equipment management stuff easier), and i'm wondering if i should go back and change that to be an entry in you.props instead 18:07:02 for that one, imo it should be "null" until there was a last unequip in a given play session 18:07:14 I wouldn't want ';' to do something based on state if I came back after a month, eg 18:07:41 lol neither would i 18:09:31 it sounds like i'm ok with that one, then. thanks 18:20:27 a good question: https://github.com/crawl/crawl/blob/master/crawl-ref/source/player.cc#L5636 (also i guess this has some dead cigotuvi's embrace code) 18:21:50 Unstable branch on underhound.eu updated to: 0.24-a0-258-gb86c0f735e (34) 18:23:52 is the qw bot no longer in development? any reccommendations? 18:23:55 oh no, i found out the answer and it's 2014-marvin's fault 18:24:02 MarvinPA: clearly it is for the very important colouring in _print_stats_ac 18:24:40 freelikegnu: recommendations for what? 18:24:41 yup, added by me apparently 18:26:19 i see. we want to know whether the player would expect to have SH, independent of whether they have SH 18:26:35 yeah, the other place it's used in beam.cc it could just check the actual SH value but that doesn't work there 18:27:16 freelikegnu: Around line 5254, you want to change "read2(it, " c")" to "read2(it, "c")", and likewise for the line two lines below. Then it doesn't choke on reading acquirement. 18:28:27 MarvinPA: it's also used in _iood_shielded, which i think could also use SH 18:30:58 how do you all like to search for callsites for stuff like this? cscope? 18:31:40 ("like this" = where is shielded() used) 18:31:40 git grep mostly 18:32:07 personally at least 18:34:21 I use cscope for that kind of thing 18:35:32 <|amethyst> I use: 18:35:34 <|amethyst> vimgrep() { 18:35:34 <|amethyst> local pattern=$1 18:35:34 <|amethyst> shift; 18:35:34 <|amethyst> local delim=$'\001' 18:35:36 <|amethyst> vim -c "vimgrep $delim$pattern$delim $*" 18:35:39 <|amethyst> } 18:36:09 freelikegnu: oh, and I had to add https://github.com/damerell/crawl/commit/3e76bdcb7c9ce5d4d8d893e2c621964eba0ab644 and a check for it in qw's can_berserk, or it chokes every time it tries to berserk when engulfed. (0.22 stopped you using god abilities when engulfed and breathing.) 18:36:36 freelikegnu: But with those changes I have qw working in 0.22 and I think in vanilla's 0.23 18:37:04 <|amethyst> then I can vimgrep foo *.cc *.h and scroll through the sites with :cn and :cp (for which I have remaps on ctrl-up and ctrl-down) 18:37:33 |amethyst: i just tried that and i kind of love it 18:37:48 i had never thought to open vim with the initial grep 18:38:54 i've been using a mix of cscope and vimgrep once vim is open, but that seems like a pretty cool way to get started if i'm not already in the editor 18:41:40 <|amethyst> another one you might like, if you have a ctags file 18:41:43 <|amethyst> tag () 18:41:44 <|amethyst> { 18:41:44 <|amethyst> vim -t "$*" -c 'norm z.' 18:41:45 <|amethyst> } 18:42:23 i just have emacs open and use M-x find-grep (which i bind to a key). the command it generated this time was: find . -type f -not -name '*.o' -exec grep -nH -Pie '(\.|->)shielded' {} + 18:42:23 <|amethyst> (you might prefer something other than z.) 18:43:21 cscope sounds pretty cool though 18:44:27 <|amethyst> yeah, that's one productivity tool that I never learned but looks awesome 18:47:22 <|amethyst> I need to try deoplete-clang, because OmniCppComplete tries but often falls short 18:48:05 <|amethyst> and rtags, which is mostly for emacs but apparently has a vim frontend 18:48:18 i've never found a solution for either linting or code completion in vim that didn't make me sad in some way or another 19:01:26 Fork (stoatsoup) on crawl.kelbi.org updated to: 0.21.2-951-g24cbdcb5fc 19:08:22 03MarvinPA02 07* 0.24-a0-259-gc7211da: Remove some dead Cigotuvi's Embrace code 10(38 minutes ago, 9 files, 4+ 25-) 13https://github.com/crawl/crawl/commit/c7211da41105 19:09:12 thanks Pinkbeast 19:37:19 -!- amalloy is now known as amalloy_ 20:18:51 The build passed. (master - c7211da #11349 : Chris Campbell): https://travis-ci.org/crawl/crawl/builds/521953979 21:32:17 -!- amalloy_ is now known as amalloy 21:47:42 |amethyst: those are cool vim tips; do you use any form of autocompletion in vim, e.g. to complete function names or other symbols as you write code? 21:47:51 that's an area I've not really played with much 21:48:09 I'll definitely use vimgrep and maybe that shell function in the future 22:21:57 I have my secretary flip through the latest printed out trunk listings and tell me which files/lines to go to 22:33:39 ebering: Have you noticed how all the comments say /* Give my secretary a raise - ebering */ 22:43:56 03ebering02 07* 0.24-a0-260-g7384745: Teach monsters to use scattershot, cloud, and iceblast wands 10(6 days ago, 4 files, 41+ 74-) 13https://github.com/crawl/crawl/commit/7384745ea90e 22:44:11 I have not 23:02:12 Pinkbeast: It was chocking on creating a new character when chose from the random combos, sees I few no longer exits in the 0.21.1 23:03:16 freelikegnu: Ah, sorry. I run it with a specific combo list, so hadn't hit that bug. 23:03:54 also qw's hypercombogen.sh -list-combos no longer works 23:04:03 seems it was replaced with -playable-json 23:05:34 sorry my typing is so bad...:P 23:05:34 Oh, yes. I have hit that, but forgot about it. 23:06:54 I posted the issue in the github project in case anyone knows how to get around that 23:07:18 I cant parse my way out of a paper bafg 23:07:45 03Joshua Gelbard02 {ebering} 07* 0.24-a0-261-g13415af: Fix a couple Engulf-related bugs 10(30 hours ago, 6 files, 20+ 36-) 13https://github.com/crawl/crawl/commit/13415af0d45b 23:08:51 it sucks that github no longer shows a timeline summary of what the forks of a project are doing 23:09:20 I think I adhocced some Perl to mangle the output of -playable-json 23:13:05 03kitchen-ace02 {ebering} 07* 0.24-a0-262-g6d359e7: Add second brand to description for Thermic Engine and Storm Bow 10(4 days ago, 1 file, 14+ 0-) 13https://github.com/crawl/crawl/commit/6d359e791a8d 23:19:20 ooooh 23:26:21 03argonaut02 {ebering} 07* 0.24-a0-263-gbee43f9: Update Gyre and Gimble 10(4 days ago, 1 file, 2+ 3-) 13https://github.com/crawl/crawl/commit/bee43f90e2b7 23:40:58 looks like development of qw has progressed a bit more here: https://github.com/alexjurkiewicz/qw/commits/improvements 23:41:58 "Wrong repository" seems to be my specialist skill at the moment 23:43:17 :o