02:12:26 -!- trumpet is now known as GeryonsHorn
02:13:27 <n1> any ##crawl mods want to come over and make this quilel person chill out?
02:13:40 <n1> *ops
03:50:02 -!- amalloy is now known as amalloy_
04:04:25 -!- amalloy_ is now known as amalloy
04:41:37 -!- amalloy is now known as amalloy_
05:16:22 -!- amalloy_ is now known as amalloy
09:55:52 <advil> who even has ops in ##crawl
10:01:46 <gammafunk> was there a problem?
10:02:14 <gammafunk> MarvinPA does, so does dpeg (who isn't around these days)
10:02:26 <advil> last night I guess, I don't have logs for that channel at that time though
10:03:02 <advil> "<n1> any ##crawl mods want to come over and make this quilel person chill out?"
10:04:53 <advil> I guess they did chill out eventually
10:06:58 <gammafunk> yeah, looking at the log, they were using slurs
10:11:50 <bhaak> where's dpeg btw? haven't seen him for a while
10:15:08 <gammafunk> hasn't stopped by for many months
12:16:27 <bh> !seen jpeg
12:16:27 <Sequell> Sorry bh, I haven't seen jpeg.
12:39:00 <MarvinPA_> i think amethyst also has ##crawl ops if needed
12:40:26 <MarvinPA_> i'm not sure if i also have the ability to add new ops, but if i do i could add some of the more currently-active devs too
12:40:46 -!- MarvinPA_ is now known as MarvinPA
12:43:37 <gammafunk> MarvinPA: yeah, if you wanted to add advil and I, that would be fine
12:43:57 <gammafunk> maybe only rax can do that
12:52:47 <MarvinPA> yeah, it looks like that's the case
12:53:07 <MarvinPA> since nobody else has +s
12:53:30 <MarvinPA> and i think that's what lets you set flags
12:54:17 <MarvinPA> it's definitely not letting me add anyone at least (although i can change stuff in here, and i do have +s here)
12:57:11 <gammafunk> ok, I'll ask rax about that
12:57:28 <gammafunk> ??rax
12:57:28 <Sequell> rax[1/2]: Barracks, cmon, do I have to tell you guys everything?
12:57:31 <gammafunk> ??rax[2
12:57:31 <Sequell> rax[2/2]: don't !tell; just email rachel at akrasiac dot org
13:02:29 <gammafunk> I sent her an email asking about it (and mentioned possibly giving MarvinPA +s there)
14:14:51 <amalloy> !tell hellmonk did you see https://www.reddit.com/r/dcss/comments/7d6wzy? hellcrawl is "a fork that doesn't add bad ideas". if only mainline crawl could achieve such a high standard
14:14:51 <Sequell> amalloy: OK, I'll let hellmonk know.
17:16:20 -!- jefus- is now known as jefus
18:00:39 <advil> %git 56314776da11a9ae22ec072b7c
18:00:39 <Cheibriados> 07|amethyst02 * 0.20-a0-397-g5631477: Actually prevent running/resting next to slime walls (doh) 10(11 months ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/56314776da11
18:03:09 <advil> |amethyst: is there any particular reason that you didn't change i_feel_safe in 5631477? (Which would have covered some of the cases, but not all)
18:03:36 <advil> going to add a slime wall check to i_feel_safe and I'm wondering if there's some reason this isn't already in there
18:04:03 <advil> (that's what caused the death in https://crawl.develz.org/tavern/viewtopic.php?f=5&t=24510 in a hilarious lua mistake)
18:04:40 <advil> (their butchering script just endlessly loops without succesfully chopping anything when next to a slime wall)
18:08:23 <advil> 17:56:00 <|amethyst> I kind of wonder if this might not be done better by changing i_feel_safe
18:08:24 <advil> ok
18:08:49 <advil> 17:56:33 <|amethyst> but would probably have side effects
18:08:59 <advil> let me know if you thought of what they would be
18:12:04 <advil> I think for the safe of lua feel_safe this function should not return true when next to a slime wall
18:12:08 <advil> *sake
18:17:02 <advil> I did a quick audit and didn't spot any major issues, the only minor one is you can't start interlevel travel
18:27:37 <Cheibriados> 03advil02 07* 0.21-a0-443-gaba41f5: Check for slime walls in i_feel_safe 10(8 minutes ago, 3 files, 12+ 6-) 13https://github.com/crawl/crawl/commit/aba41f5a6fd5
18:56:09 <|amethyst> advil:  I think the issue was running/travelling away from a slime wall
18:56:48 <|amethyst> advil:  ah, no, I didn't recognise that issue at the time
18:56:57 <|amethyst> advil:  so, yeah, no real reason, so good change :)
18:57:04 <|amethyst> :q
18:57:06 <|amethyst> doh
20:36:52 <wheals_> according to github that commit adds a binary file
20:40:14 <geekosaur> "crawl-asdf"
20:41:31 <amalloy> part of advil's secret plan to convert dcss to common lisp
20:45:56 <gammafunk> woah
20:45:59 <gammafunk> that's what that was!
21:06:56 <Cheibriados> 03gammafunk02 07* 0.21-a0-444-g29a2fec: Remove the secret Crawl ASDF project added by the Shadow Dev Team 10(4 minutes ago, 1 file, 0+ 0-) 13https://github.com/crawl/crawl/commit/29a2fec1a17c
21:07:57 <wheals_> nothing to see here, folks
21:08:32 <gammafunk> the first rule about crawl-asdf: no one talks about crawl-asdf
21:08:41 <wheals_> we definitely aren't planning to replace hjkl-keys with asdf-keys
21:09:27 <gammafunk> advil: I've exposed your evil plan
22:20:12 -!- gammafunk is now known as foobargl
22:20:28 -!- foobargl is now known as gammafunk
22:41:32 <Yermak> Btw, advil, about "you can't start interlevel travel": it would be nice if you couldn't press ctrl-G with monsters in sight - right now after choosing your destination you're notified that you can't travel with monsters in sight anyway, but you've already spent time looking for the letter you need and game has spent time building the route.
22:47:03 <neizenel> wheals_: good (bad) april fools idea there
22:53:13 <|amethyst> Yermak:  well
22:53:23 <|amethyst> Yermak:  it also remembers your destination doesn't it?
22:53:44 <|amethyst> so you can do G<enter> after the monsters are gone
22:53:53 <Yermak> Yes, but after you kill the monster, it still will rebuild the route.
22:54:04 <Yermak> It can take a while in Webtiles.
22:54:13 <|amethyst> hm
22:55:32 <|amethyst> yeah, with G it's not so bad, but I guess ^G could take some time
22:55:55 <|amethyst> err
22:55:58 <|amethyst> I mean, with X<arrowns><enter> it's not so bad
22:56:03 <|amethyst> s/wn/w/
22:57:08 <advil> haha oops
22:57:17 <advil> I typed "git add ." didn't I
22:59:00 <advil> (that's just a binary from testing a PR branch on multiple machines)
22:59:30 <advil> Yermak: yeah, reasonable
22:59:44 <geekosaur> git addled...
23:00:35 <advil> another suggestion I've seen is that things like G and ctrl-G could take a single step
23:01:27 <advil> with monsters in sight
23:01:27 <|amethyst> ?
23:01:27 <|amethyst> ah
23:02:03 <|amethyst> that sounds dangerous, and I think should be an option at most
23:02:20 <advil> yeah, it's not obvious how to do it right
23:02:42 <advil> I think the specific scenario was orbruns
23:21:25 <Yermak> Are there any plans to change current turn system into elegant queue?
23:23:31 <gammafunk> By turns you mean monster actions?
23:25:01 <Yermak> I mean the whole time thing.
23:25:28 <Yermak> Currently all revolves around players moves.
23:25:34 <Yermak> player's
23:28:00 <gammafunk> Well world action is pretty complicated, and yes it is on a two-part cycle of player action, world actions
23:28:21 <gammafunk> But exactly how a rework of that would look isn't clear to most of us, I think
23:28:35 <gammafunk> It's certainly a project someone could take up if they had a good design
23:29:18 <gammafunk> It's difficult to reason about monster behaviour from a programming perspective at times because you have to understand how monster energy works
23:30:47 <gammafunk> And monster energy isn't "how long it takes a monster to act", which might make more sense
23:31:28 <gammafunk> so I guess the short answer to that question is "no, but patches welcome as always"
23:31:33 <|amethyst> even with a queue-based system, I think you'd still want actions to happen "immediately"
23:31:45 <|amethyst> I mean, you'd want the delay to come after the action, not before
23:32:25 <|amethyst> since otherwise you have things like swinging at a monster that then moves away before the swing lands
23:32:57 <|amethyst> or a monster deciding to move, and another monster entering its desired space before the move finishes
23:34:44 <|amethyst> I think designing such a system wouldn't be too hard... the problem is then making it work with all the weirdnesses and edge cases in Crawl
23:35:02 <|amethyst> e.g. you'd have to reimplement battiness probably
23:41:21 <Yermak> Let's say we a have actors_queue, which is sorted by actor->delay (this means actor will act after actor->delay auts pass).
23:41:21 <Yermak> 1) Now check if head->delay is 0. If it's not, decrease all delays in actors_queue by one. Goto 1).
23:41:21 <Yermak> 2) Pop our queue. Popped actor decides what action to take. Action happens. Popped actor gets inserted into the queue with actor->auts value of x, where x is the cost of the action taken.
23:42:19 <Yermak> Battiness could use some bool like batty_can_attack which is set to true after player moves.
23:50:04 -!- thurin is now known as mormegil
23:50:42 <Yermak> s/actor->auts/actor->deley/
23:50:46 <Yermak> delay
23:53:50 <|amethyst> 1 is slow, you can do better with a priority queue
23:54:47 -!- mormegil is now known as thurin
23:54:50 <|amethyst> where the priority is the aut on which the monster would act
23:54:59 <|amethyst> so (now + action_delay)
23:55:55 <|amethyst> then you don't need to tick off each individual aut
23:56:21 <|amethyst> but just pop from the queue, and calculate how much time has elapsed
23:57:05 <|amethyst> things like poison, cloud damage, etc. could go onto the queue, but they don't have to
23:57:48 <|amethyst> (making those per-aut decreases variance compared to currently)
23:58:01 <Yermak> How many actors will be at the queue simultaneously? It's not a problem to decrease a dozen of variables.
23:59:15 <|amethyst> every monster on the level, however many that is
23:59:32 <|amethyst> the problem isn't so much decrementing everyone's count
23:59:42 <Yermak> Yeah, monsters out of los get into queue too, you're right
23:59:50 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-444-g29a2fec (34)
23:59:52 <|amethyst> as it is doing that several times in a row before anyone acts