01:50:35 -!- TAS-2012v is now known as TAS_2012v 02:50:22 -!- TAS-2012v is now known as TAS_2012v 03:47:05 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-201-gf205589 04:03:50 03CanOfWorms02 {GitHub} 07* 0.24-a0-202-g19c84a8: Update tgw_lugonu_bribe 10(10 seconds ago, 1 file, 8+ 8-) 13https://github.com/crawl/crawl/commit/19c84a8c1891 04:51:00 -!- amalloy is now known as amalloy_ 05:07:07 Exclusions on monsters made with auto_exclude disappear when monsters move. 05:07:13 FR: make exclusions that you put on monsters by hand follow the same behaviour. 05:07:24 It's really annoying to clear them afterwards. 05:13:27 The build has errored. (master - 19c84a8 #11251 : CanOfWorms): https://travis-ci.org/crawl/crawl/builds/510842756 06:08:12 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-202-g19c84a8c18 (34) 09:04:32 advil: I hesitated using the phrase because its such aa colloquialism 09:29:10 Hi there. Is this the place where i can speak to DCSS devs? Just i'm having a quick suggestion about accessibility, i'm a blind person who plays dcss with a screen reader and something's just came to my mind. 09:32:04 arnold: yes it is, please share 09:32:17 arnold: and in the future, no need to ask before asking 09:41:32 Basicaly what i'm thinking of is putting some indication of where monsters are relatively to your position. The solution in my opinion is quite simple, i thought of something like in the visible features list accessed by ctrl+x there could be a something like: A rat, 1 up, 1 right. The reason why i'm asking this is that sometimes when you face a pack of very dangerous monsters it would be so much easier to see or in my case hear w 09:43:25 Then i could more reliably determine for example the shortest escape routes, it is too cumbersome to fumble around in the map screen listening to every square when you're getting a whole pack of monsters. I lost five orb runs because if i took a wrong step so i came here to ask if something like could be made. The game is fantastic by the way, i love it. 09:45:33 arnold: how do coordinates "read" to you. For example, if a ctrl-x line was changed to include the coordinates: (1, 1) a rat 09:49:13 It'd be perfect. No need to be strings, they could be coordinates just i thought of strings in the first place, but coordinates probably are even easier to interpret. By the way, any idea where to translate crawl in to other languages? The transifex thing is defunct as far as i noticed. I thought of compiling crawl for myself and trying to make the changes myself but i would be too lazy to do it for every version :) 09:57:42 Note that i'm using the console version if it makes any difference. 10:49:21 New branch created: pull/1018 (1 commit) 13https://github.com/crawl/crawl/pull/1018 10:49:21 03NormalPerson702 07https://github.com/crawl/crawl/pull/1018 * 0.24-a0-202-g7ff1fe4: Don't place shafts in ebering_ghost_disaster_area (11926) 10(17 minutes ago, 2 files, 2+ 12-) 13https://github.com/crawl/crawl/commit/7ff1fe4476b7 12:41:36 -!- amalloy_ is now known as amalloy 13:05:39 Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-202-g19c84a8 (34) 15:06:23 Orphidia (L23 MiGl) ERROR in 'dbg-scan.cc' at line 680: Portal enter_abyss[54] didn't get generated. (Depths:4) 15:07:01 Orphidia (L23 MiGl) ERROR in 'dbg-scan.cc' at line 680: Portal enter_abyss[54] didn't get generated. (Depths:4) 15:28:14 03CanOfWorms02 {GitHub} 07* 0.24-a0-203-gf0bc1e9: More tweaks to tgw_lugonu_bribe 10(8 seconds ago, 1 file, 2+ 5-) 13https://github.com/crawl/crawl/commit/f0bc1e94aeb4 15:31:43 !crashlog 15:32:03 20762. Orphidia, XL23 MiGl, T:51541 (milestone): https://underhound.eu/crawl/morgue/Orphidia/crash-Orphidia-20190325-190700.txt 15:32:20 Unstable branch on crawl.kelbi.org updated to: 0.24-a0-203-gf0bc1e94ae (34) 16:40:12 The build passed. (master - f0bc1e9 #11253 : CanOfWorms): https://travis-ci.org/crawl/crawl/builds/511133702 19:22:13 Unstable branch on underhound.eu updated to: 0.24-a0-203-gf0bc1e94ae (34) 21:08:17 I have a kind of general question about crawl's item-related code 21:08:53 I've been staring at autopickup code a lot, so maybe I'm overestimating how much this comes up elsewhere 21:10:06 but it seems like there are a lot of functions that have to do with checking properties of an item type in general 21:10:19 but which take a reference to an actual instance of that item type 21:10:32 I should find an example 21:10:38 indeed, i was about to ask you to 21:11:36 I guess "_is_option_autopickup" in items.cc is a good one, actually 21:12:24 or _should_autobutcher is the same thing but shorter 21:12:34 !source _is_option_autopickup 21:12:35 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/items.cc#L2905 21:13:30 I see lua code calls in _is_option_autopickup so maybe those end up making the individual identity of the item relevant 21:13:45 _should_autobutcher is just the next function down 21:13:51 that seems like it uses the item, via _autopickup_item_name and userdef_annotate_item 21:13:59 yeah bad example 21:14:44 _should_autobutcher i think is just an example of a function with a narrow use case, so it was designed to have a convenient signature for the places it's called. in those places, an item_def& is already in scope 21:15:42 this is true 21:15:43 well, actually it relies on is_inedible and is_bad_food, which in principle could care about more than the base type. let's look at those 21:15:51 !source is_bad_food 21:15:51 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/food.cc#L709 21:15:53 !source is_edible 21:15:55 Can't find is_edible. 21:16:01 is_inedible 21:16:43 eventually this gets down to determine_chunk_effect, which checks the type of the monster that died 21:17:08 so even if the base type is corpse, there are edible corpses and inedible 21:17:34 maybe I was imagining this 21:18:34 but it sounds like there's certainly no intentional effort to do that, or anything 21:18:41 that was going to be my next question 21:19:13 maybe I should be avoiding writing functions that take an object_class_type and an int as an argument 21:19:25 since there just aren't that many of them 21:20:11 but it sounds like the actual answer is: individual item traits are relevant in more places than I thought 21:22:12 !source object_class_type 21:22:13 Can't find object_class_type. 21:22:17 what is a object_class_type 21:22:46 OBJ_POTIONS, OBJ_WEAPONS, etc 21:22:50 enum for base types 21:24:07 i think you should try to have your parameters include as little information as you can bear, but no less 21:26:12 I will do my best 21:26:52 I'm glad to have discovered that typing object_class_type all the time is not a universal crawl dev experience 21:27:53 maybe this really is only relevant to autopickup code 21:28:16 (I'm working on making Yermak's "extremely simple autopickup logic" idea a thing) 21:32:44 well, i haven't written a lot of crawlcode in a while. but apparently the only time i ever cared about object_class_type was in sept 2016 when i made it an enum class instead of an int enum, via ea4be0 21:35:12 -!- amalloy is now known as amalloy_ 21:35:14 mine seem to be ints 21:35:21 I am confused 21:37:16 well, it's fine. thanks for looking at the code I had questions about 21:52:44 -!- amalloy_ is now known as amalloy 22:14:22 fourba: https://github.com/crawl/crawl/blob/b306ea240a012dd66efd0c049ce1f1577385bee0/crawl-ref/source/object-class-type.h#L3 - an "enum class" has an int underlying it, but doesn't implicitly convert to an int like an old C-style enum does 22:35:47 I do like the whole enum class thing 22:35:58 scoping is cool 22:36:13 typing more than scoping, for me 22:36:15 but sure 22:54:14 Oh, fourba, since you are there (in autopickup area), could you take a look at this thing as well: when you backpack is full and chunks are off autopickup, autobutcher sometimes (always?) askes you if you want to pick up chunks from a corpse it just dissected. It's really annoying. 23:02:44 !kw statzero str<1|int<1|dex<1 23:02:45 Defined keyword: statzero => str<1|int<1|dex<1 23:03:41 !kw statzero str<1 || int<1 || dex<1 23:03:42 Defined keyword: statzero => str<1 || int<1 || dex<1 23:14:20 Does it make sense for autobutcher to be doing anything when chunks are off autopickup to begin with? 23:21:58 It does make sense for it to create them from available corpse, that's all, imo. 23:50:19 Mmm. If I have autobutcher on and chunks off autopickup, I think you have to interpret that as "make chunks, in case I feel hungry and one of the chunks I've made is lying around." 23:55:32 autobutcher thing has different states of satiation 23:55:40 mine is set to very hungry 23:56:00 so when it makes a chunk I autoeat it immediately 23:56:39 (autobutcher doesn't do anything until I'm very hungery) 23:56:50 That's true too, but even if someone has it set to Satiated but not to autopickup chunks, I don't see that admits of any interpretation but "make chunks and leave them behind" 23:58:04 Well, I can't see the reason for people to set autobutcher to satiated (for non-trolllike species) and then turn off autopickup for chunks. 23:59:32 I think we all want the same behavior here 23:59:33 It's like setting threshold for autofight to 99%, if they did it, it's their problem. 23:59:43 I hadn't even realized it, but this annoys me too (in zigs)