01:20:24 <twelwe> this is a non aggressive greeting message for you 01:51:37 <chequers> aidanh: yes, and moving all species logic into species-specific files, rather than put throughout the codebase 01:52:34 <chequers> if it's just hooks (like how unrands work) more hooks can be added on demand, and each hook is optional, so I would not be too worried about "overfitting" if that's what you mean there? 01:54:16 <aidanh> by over fitting I mean exposed hooks that are very species specific, so that in practice they're only usable by the species they were written for 01:56:06 <aidanh> ideally most of the species implementation would be in the code calling these hooks, not the hooks themselves 01:56:51 <aidanh> the difference between this and unrand hooks is that the latter have no compatibility requirements 01:59:33 <aidanh> in any case it'd be a good idea to think about what APIs one might need to implement some of crawl's more unique species; e.g. Dr, fe, mu, op, etc 02:43:19 <muffindrake> implementing APIs for things that are by definition exceptional is a fool's game 02:52:14 <Currency> aidanh: There are a few yesno calls that have nullptr's passed in for the strings. These should probably be changed to some sort of message now that they have their own window. 02:54:03 <aidanh> good thing that's exactly what I'm cautioning against, then 02:54:28 <Currency> hm? 02:56:49 <aidanh> Currency: that was a separate response; and thanks, I'll take a look 02:57:30 <aidanh> there's still lots of direct calls to cancellable_get_line, but mostly from places where it's not an issue 03:00:20 <Currency> Ah ok. Sorry just jumped in and posted. Didn't see the ongoing conversation. 03:54:48 -!- amalloy is now known as amalloy_ 05:53:44 -!- amalloy_ is now known as amalloy 06:09:35 -!- amalloy is now known as amalloy_ 07:38:33 -!- johannm is now known as Tungsten_ 09:50:06 <advil> yeah, I don't really see "species mods" that wouldn't require recompiling as particularly viable in crawl's codebase 09:58:29 <advil> also we need like a demigod faq that summarizes these circular conversations people are constantly having about them (though dg = atheist og is a new one to me, I have to admit) 11:09:00 <ebering> I thought about writing a patch that made altars disappear in puffs of smoke for Dg on sight 11:09:09 <ebering> pure flavour but might break the cycle :) 11:12:28 <ebering> also now that swamp worms can go on land nicolae_swamp_entry_splash_gauntlet might need to be looked at 11:12:35 <ebering> the worms wander off 11:16:48 <ebering> ya the worms should be made patrolling! 11:26:36 <Cheibriados> New branch created: pull/763 (1 commit) 13https://github.com/crawl/crawl/pull/763 11:26:37 <Cheibriados> 03ebering02 07https://github.com/crawl/crawl/pull/763 * 0.22-a0-423-gfbaaf6b: Make swamp worms patrol swamp entries 10(6 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/fbaaf6b9346f 11:41:09 <aidanh> advil: well, it'd require a lot of changes to various parts of crawl, but many of those changes could be worthwhile improvements in their own right 11:42:01 <aidanh> e.g. some sort of bare-bones mod system could make using various bits of lua more user-friendly by removing the config editing requirement 11:51:23 <advil> even the idea of extracting SP_ enum values from a c enum into something else (what?) seems basically intractable to me 11:51:30 <advil> Helix:source advil$ git grep SP_ *.cc | wc -l 11:51:30 <advil> 761 11:52:23 <advil> I mean, not to say that it's impossible, or that one couldn't design for a setup like that from scratch, but... 11:56:45 <aidanh> that would certainly be a formidable undertaking, but it's not strictly necessary 12:00:00 <aidanh> er, to clarify, if we provided support for runtime-loadable species data, it'd be nice if all "core" species used that interface, but not strictly necessary 12:18:00 -!- johannm is now known as Tungsten_ 13:18:03 -!- amalloy_ is now known as amalloy 14:40:41 <Cheibriados> excluding a huge gate that you have seen only part of yet doesn't exclude its unseen tiles 13https://crawl.develz.org/mantis/view.php?id=11479 by eb 15:11:34 <ebering> Hm. feature += awoken tree {,,lightred} doesn't work. looking at showsymb.cc this is because awoken forests always get colour ETC_AWOKEN_FOREST 15:13:02 <ebering> this colour alternates grean and lightgreen in animation. I'm not sure a good approach to making it configurable off the top of my head, but I think a default of red/lightred animation might be better 15:21:15 <amalloy> ebering: incidentally escape hatches are another feature you can't recolour 15:23:05 -!- NeremWorl is now known as NeremWorld 15:56:31 <ebering> amalloy: so they are 15:56:41 <amalloy> because they're always "highlighted" 15:56:45 <amalloy> for some dumb reason 16:01:40 <ebering> amalloy: oh, you can change that (after code reading) 16:01:49 <amalloy> oh? 16:01:52 <ebering> feature lets you change "emphasized colour" 16:02:47 <ebering> feature += escape { symbol, magic map symbol, colour, levelmap magic colour, levelmap seen colour, *this one* emphasized colour, *and this* emphasized levelmap colour } 16:03:10 <ebering> though the feature documentation could probably use a sentence pointing this out, it does work to do 16:03:22 <ebering> feature += escape { ,,,,,cyan} 16:04:03 <ebering> awoken forests are hard-coded set to this unconfigurable etc colour in a way that overrides player configured colours 16:06:10 <amalloy> dang, how did you know cyan is the colour i tried to use 16:06:50 <amalloy> oh no, i tried lightblue 16:07:28 <amalloy> ??rcfile 16:07:28 <Sequell> rcfile[1/4]: https://github.com/crawl/crawl/blob/master/crawl-ref/docs/options_guide.txt 16:09:17 <amalloy> i tried feature += escape hatch {,,,,lightblue} and it didn't work, which i later realised was because they're always emphasised. i didn't then go back to read the documentation again to see there's a way to change emphasised colour 16:16:55 <ebering> so, I think there's two things to do here. 1. easy: change defaults for awoken forest in console to make them stand out 2. more work: make abstract elemental colours (ETC_*) user configurable 16:25:35 <Cheibriados> New branch created: pull/764 (1 commit) 13https://github.com/crawl/crawl/pull/764 16:25:35 <Cheibriados> 03ebering02 07https://github.com/crawl/crawl/pull/764 * 0.22-a0-423-g84b5288: Recolour awoken forests in console. 10(7 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/84b528801de9 17:00:18 <Jozrael> Hello! Does anyone have experience setting up a crawl server locally? 17:02:39 <geekosaur> if by locally you mean same local LAN/WAN, just change webserver/config.py as appropriate and run webserver/server.py 17:02:58 <geekosaur> if you want things like scoreboards, you'd need a full server setup which is a lot more involved 17:03:33 <geekosaur> (note that the default config.py already allows anyone on the local network to connect, provided your router and/or firewall permit it) 18:11:43 <Jozrael> I think we’d want scoreboards :). Setting one up for my company, got a dozen or more players. 18:12:27 <Jozrael> Spoilers, my ultimate goal is going to get a dockerfile setup for this ;) 18:12:45 <Jozrael> Which we’ll share back! 18:12:57 <Jozrael> But DCSS-in-a-box sounds fun no? 18:19:24 <ebering> Jozrael: TZer0 was working on a crawl-server-docker a while back 18:19:37 <Jozrael> Ooo 18:19:51 <Jozrael> I am more than happy to shamelessly steal/crib if it already exists, thx 18:20:12 <Jozrael> https://bitbucket.org/TZer0/crawl-docker perhaps 18:28:46 <geekosaur> ??docker[1 18:28:46 <Sequell> docker[1/2]: https://bitbucket.org/TZer0/crawl-docker - functioning prototype with external storage settings (and thus import) working. 18:28:48 <geekosaur> ??docker[2 18:28:48 <Sequell> docker[2/2]: https://bitbucket.org/mattiasjp/crawl-docker by Mattias 18:29:02 <geekosaur> neither complete but you might be able to get something from one or both 18:37:37 <Cheibriados> 03advil02 07* 0.22-a0-423-g4197694: Don't leak information about passwall landing site via exclusions 10(12 minutes ago, 1 file, 26+ 8-) 13https://github.com/crawl/crawl/commit/4197694d7906 19:03:22 <ebering> advil: I think that still leaks information. I could place an exclusion in the direction I want to passwall to get a prompt if I'm going to be going further than I want to. I suppose it will charge me MP and a turn 19:08:26 <ebering> but, I could imagine being in a situation with a baddie 3 tiles away and wanting to "passwall into the unknown only if I won't get hit" 19:09:49 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.22-a0-423-g4197694d79 (34) 20:06:37 <amalloy> i thought the bug reported was about the opposite, the player wanting to avoid an overshoot 20:07:24 <amalloy> "i want to escape with passwall, but make sure i only spend 1 turn merging with the rock" 20:09:20 <ebering> amalloy: yes, I'm commenting on an unintended consequence of advil's change, which solves that bug at the expense of introducing the behavior I just described 20:09:36 <amalloy> i see 20:12:18 <advil> ebering: yes, the commit message mentions the possibility of an overshoot leak 20:12:52 <advil> the bug reported was about avoiding an undershoot 20:13:45 <advil> the overshoot leak would already have been present, as well 20:15:09 <ebering> oops, should have read the message. sorry 20:17:41 -!- amalloy is now known as amalloy_ 20:26:20 <advil> I think I know the basic check for the overshoot case but I couldn't really figure out any scenario where it mattered as much as the example in the reddit post (though I'm sure there could be some) 22:34:25 -!- Jozrael_ is now known as Jozrael 23:11:31 -!- n1k is now known as n1 23:59:28 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.22-a0-423-g4197694 (34)