00:27:30 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-196-g6a663c0 (34)
00:29:20 <johnstein> @??summoner
00:29:20 <Gretell> unknown monster: "summoner"
00:33:12 <johnstein> did some monsters used to be able to cast fulminating prism?
00:33:27 <johnstein> seems like there's some code to handle the situation
00:33:35 <johnstein> !source cast_fulminating_prism
00:33:36 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/spl-summoning.cc#L2950
00:33:37 <ProzacElf> uh
00:33:41 <ProzacElf> none that i know of
00:33:47 <ProzacElf> with the possible exception of ghosts
00:34:08 <johnstein> hmmm
00:34:10 <ProzacElf> ghosts definitely could now that i think about it
00:34:15 <ProzacElf> possibly still can
00:34:17 <johnstein> maybe that's what's going on
00:34:29 <johnstein> if (caster->is_player()
00:34:30 <johnstein> || (you.can_see(*caster) && you.see_cell(where)))
00:34:30 <johnstein> {
00:37:59 <johnstein> hm. is there a reliable way to create a ghost with particular stats?
00:38:25 <johnstein> or just kill a character with the stats I want, copy the bones file, and just keep reloading games till I see it?
00:39:46 <|amethyst> deleting other bones files would help
00:40:02 <|amethyst> if you build with DEBUG_BONES you can use &^G
00:40:32 <|amethyst> (which is enabled by FULLDEBUG)
00:40:45 <johnstein> ooh
00:44:56 <johnstein> !kw ghostly
00:44:56 <Sequell> Keyword: ghostly => !boring !won ktyp!= lg:place!=D:1 lg:place!=D:2 lg:place!=Temple lg:place!=abyss !mu !gh !vp lg:status!~~lich
00:45:04 <johnstein> shoot, D:2 also
00:46:10 <johnstein> hmm. I let a dude die on D:3 but nothing in the bones file
00:46:39 <johnstein> bones folder I mean
00:47:03 <johnstein> ??ghost
00:47:03 <Sequell> ghost[1/5]: Ghosts get damage, hitpoints, resistances, speed (not move delay), and spells very closely based on the former adventurer. All player ghosts are immune to poison, asphyxiation, torment, and negative energy.
00:48:58 <johnstein> I'd try the DEBUG_BONES, but I'm sure that will take 30 minutes or more for a full recompile
00:50:10 <johnstein> oh. maybe because it's wizmode?
00:54:34 <johnstein> so when do ghost spells get picked? when the player originally dies? when the bones file is loaded?
01:15:10 <johnstein> I can't seem to figure out how to make this ghost learn f. prism
01:59:40 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-189-gb10dd60: Add a parameter to a function 10(2 hours ago, 2 files, 6+ 5-) 13https://github.com/crawl/crawl/commit/b10dd60c2f15
01:59:40 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-190-g8d47aad: Handle ghosts casting these spells 10(28 minutes ago, 1 file, 10+ 6-) 13https://github.com/crawl/crawl/commit/8d47aad3f520
01:59:40 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-191-g91da654: Toggle autopickup when detecting monsters with spells 10(5 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/91da65494674
02:01:47 <alexjurkiewicz> "these spells" :P
02:02:04 <johnstein> yea, it's just a placeholder commit that will get squashed later.
02:05:36 -!- bug_sniper is now known as Elbereth
02:05:43 -!- Elbereth is now known as bug_sniper
02:47:03 <amalloy> i know as recently as a year or two ago, pan lords could cast f.prism
02:47:10 <amalloy> i think it got taken away though
02:47:46 <amalloy> %git 2d9ae9d7a9e0a4fcd3a021a97630a0e6cdeb7eb9
02:47:46 <Cheibriados> 07MarvinPA02 * 0.20-a0-175-g2d9ae9d: Don't let monsters cast Fulminant Prism 10(9 months ago, 1 file, 1+ 87-) 13https://github.com/crawl/crawl/commit/2d9ae9d7a9e0
02:49:14 <amalloy> johnstein: if you want to test creating player ghosts with particular spells, you can do something like: %mplayer ghost spells:magic_dart.200.wizard
02:49:21 <amalloy> &m, not %m of course
02:49:37 <amalloy> this is described in https://github.com/crawl/crawl/blob/master/crawl-ref/docs/develop/levels/syntax.txt#L937
02:50:13 <amalloy> (but it doesn't work for f.prism because as discussed monsters can no longer cast that)
02:51:50 <amalloy> monsters also can't cast lightning spire, which i imagine is the other spell you're thinking of
02:54:21 <gammafunk> they can't cast it....YET!
03:22:01 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-196-g6a663c0 (34)
05:00:15 <gammafunk> .gfgk
05:00:16 <Sequell> 681. deartome the Cleaver (L10 MiFi of Trog), slain by an orc knight (a +3 great sword) on D:8 (gammafunk_beogh_shrine) on 2017-08-15 06:50:19, with 4718 points after 8307 turns and 0:15:20.
05:00:26 <gammafunk> do you think this user's name is a love letter to TOME?
05:00:34 <gammafunk> .gfgk x=src
05:00:35 <Sequell> 681. [src=cwz] deartome the Cleaver (L10 MiFi of Trog), slain by an orc knight (a +3 great sword) on D:8 (gammafunk_beogh_shrine) on 2017-08-15 06:50:19, with 4718 points after 8307 turns and 0:15:20.
05:00:41 <gammafunk> on cwz so probably not
09:23:23 <johnstein> thanks amalloy. I was pretty confident non ghost monsters couldn't cast prism and spire since I couldn't find any monsters with that spell in their spell set
09:24:06 <johnstein> but I didn't know about player ghosts.
09:24:40 <johnstein> but sounds like player ghosts can't cast spells that monsters can't cast?
10:29:17 -!- Dracunos_ is now known as Dracunos
10:44:49 <advil> often the monster version is implemented separately, so that at least places a lower bound on it
10:45:32 <advil> pan lords can sometimes cast spells that regular monsters can't, though, and I think at least historical ghosts could too
10:45:53 <advil> not sure what the current state of things is
11:15:12 <johnstein> so for spells like f prism that were once castable by monsters but not anymore, are the cast_ functions ever updated to remove the code that handles monsters casting them?
11:15:32 <johnstein> like the if (caster=player) type of checks
11:26:18 <geekosaur> various past devs (grunt and DO come to mind) fleshed out a few spells for monsters that were later spiked, but simply removing checks for player v. monster may not work (runtime crash, or worse)
11:26:26 <geekosaur> pan lords and player ghosts are both monsters as far as crawl's internals are concerned, and quite a few spells are coded to only work with players (because they use things that are moissing or different in monster data)
11:27:25 <geekosaur> (and most of the spells that were changed back, were changed for a reason; often because of aforementioned crashes. prepare for hard-to-fix or hard-to-track-down bugs)
11:30:13 <johnstein> yea. that was kind of what I was assuming. might be risky to excise some of that code
13:16:12 <amalloy> imo if we never want monsters to cast those spells we should remove the code that handles it
13:19:23 <amalloy> the only reason i can think of *not* to do that would be save compat for games with pan lords who have those spells in their spellbook, but if that is a concern i think we should remove those spells from them on game load
13:22:52 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-196-g6a663c0 (34)
13:54:15 <amalloy> |amethyst: how would you feel about a patch replacing all instances of x=="" with x.empty() and x!="" with !x.empty()?
14:22:03 <amalloy> man, i don't like how far away from .empty() the ! is sometimes. in an expression like !foo.bar.f(x, y, z).empty() it's easy to lose track
14:22:48 <amalloy> -1 to C-family parser precedence. (not (empty ...)) is clearly better
14:23:28 <geekosaur> arguably it should be foo.bar.f(x, y, z).empty().not()
14:23:49 <amalloy> that would be more object-y, yes
14:25:45 <amalloy> which made me think: how does smalltalk do it? and it turns out that yes, smalltalk uses a postfix not. (2 >= 7) not
14:46:51 <amalloy> !source _mimic_vanish
14:46:52 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-util.cc#L1055
14:47:42 <amalloy> this appears to be protecting against _laughs_ not being found in monspeak.txt, but it is there and afaik should always be there. is it important that we keep the defensive clause in this function?
14:49:52 <advil> hm, fr generate a unit test from all instances of getSpeakString
14:54:26 <geekosaur> defensive programming
15:18:44 <advil> !crashlog
15:18:45 <Sequell> 17888. PublicChat, XL1 TrCK, T:1 (milestone): https://webzook.net/soup/morgue/0.19/PublicChat/crash-PublicChat-20170814-200001.txt
15:18:52 <advil> !crashlog !PublicChat
15:19:04 <Sequell> 17726. zahalkas, XL26 MiFi, T:46156 (milestone): https://webzook.net/soup/morgue/trunk/zahalkas/crash-zahalkas-20170814-200001.txt
15:19:21 <advil> !cmd !crashlog
15:19:21 <Sequell> Command: !crashlog => !lm ${*:-*} crash -log
15:23:14 <advil> what would be the easiest way to get !crashlog to exclude the spurious crash accounts?  PublicChat, Ziggurwhatever are the main ones I'm thinking of
15:27:32 <advil> though maybe what I really want is for SIGPIPE to not generate a milestone?
15:28:25 <gammafunk> advil: probably to create a nick
15:28:38 <gammafunk> and have the default command query that nick
15:29:19 <advil> is !whatever the right syntax for anyone but that alias?
15:29:26 <gammafunk> so it would like like: !lm ${*:-ignoredcrashers} crash -log
15:29:30 <gammafunk> if the nick was ignoredcrashers
15:29:47 <gammafunk> !lm ${*:-bot} crash -log
15:30:07 <Sequell> 75. drasked, XL7 HOMo, T:5715 (milestone): http://crawl.xtahua.com/crawl/morgue/drasked/crash-drasked-20170725-230747.txt
15:30:10 <gammafunk> !nick bot
15:30:11 <Sequell> Mapping bot => autorobin xw auto7hm rw qw ow qwrobin gw notqw jw parabodrick hyperqwbe cashybrid tstbtto parabolic oppbolic ew rushxxi gaubot cojitobot paulcdejean otabotab nakatomy testingqw beemell beem drasked phybot medicine dracbot whoyougonnacall khrogbot jwbot lovelain autocrawlbot swippen cotteux someone4956 nofunallowed
15:30:31 <gammafunk> yeah looks like that works fine
15:30:49 <gammafunk> then create the nick with something like !nick ignoredcrashers name1 name2 ...
15:31:15 <gammafunk> oh, right yeah if you want anyone but those
15:31:30 <advil> yeah, anyone but that nick
15:31:30 <gammafunk> using ${*:-!ignoredcrashers}
15:31:45 <gammafunk> !lmg ${*:-!bot} crash -log
15:31:45 <advil> ah
15:31:49 <gammafunk> !lg ${*:-!bot} crash -log
15:31:50 <Sequell> No keyword 'crash'
15:32:02 <gammafunk> er
15:32:04 <gammafunk> !lm ${*:-!bot} crash -log
15:32:07 <Sequell> 17813. PublicChat, XL1 TrCK, T:1 (milestone): https://webzook.net/soup/morgue/0.19/PublicChat/crash-PublicChat-20170814-200001.txt
15:32:13 <gammafunk> yeah looks good
15:32:25 <gammafunk> public chat is actually a bot, but it doesn't play games and isn't in that nick
15:32:48 <advil> what does what does ${*:-x} do? is that documented somewhere?
15:32:58 <gammafunk> yes, all that syntax is documented
15:32:59 <gammafunk> ??lg
15:32:59 <Sequell> listgame[1/5]: !lg command displays info about past games. The manual is available here: https://github.com/crawl/sequell/blob/master/docs/listgame.md
15:33:02 <gammafunk> in that manual
15:33:06 <gammafunk> but what that syntax means is
15:33:29 <gammafunk> "expand this variable ($*) and use the value after `:-' as a default value if that variable is undefined
15:33:32 <gammafunk> "
15:33:49 <advil> ah
15:33:59 <wheals_> yeah https://github.com/crawl/sequell/blob/master/docs/listgame.md#user-defined-commands
15:34:11 <advil> thanks
15:34:14 <wheals_> for the specific section
15:34:29 <wheals_> i guess it's inspired by shell script syntax
15:34:34 <gammafunk> ??sequellese
15:34:34 <Sequell> sequellese[1/1]: https://github.com/crawl/sequell/blob/master/docs/commandline.md <|amethyst> It's like an unholy combination of Lisp and shell
15:34:44 <gammafunk> another good reference
15:35:20 <gammafunk> that variable syntax is actually part of "sequellese" and isn't specific to !lg/!lm
15:35:42 <gammafunk> although the documentation about that particular syntax might be better in the lg manual
15:36:00 <advil> !nick ignoredcrashers PublicChat PublicControlBot ziggurnaut
15:36:01 <Sequell> Mapping ignoredcrashers => publicchat publiccontrolbot ziggurnaut
15:36:42 <advil> !cmd !crashlog !lm ${*:-!ignoredcrashers} crash -log
15:36:42 <Sequell> Redefined command: !crashlog => !lm ${*:-!ignoredcrashers} crash -log
15:36:45 <advil> !crashlog
15:36:46 <Sequell> 17596. zahalkas, XL26 MiFi, T:46156 (milestone): https://webzook.net/soup/morgue/trunk/zahalkas/crash-zahalkas-20170814-200001.txt
15:36:56 <advil> still a sigpipe but w/e
16:33:17 <Cheibriados> 03amalloy02 07* 0.21-a0-197-g68382d6: Fix a misplaced semicolon 10(11 minutes ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/68382d67e91f
16:33:17 <Cheibriados> 03amalloy02 07* 0.21-a0-198-ge013ddc: Use an rvalue reference to avoid some pointless copies 10(14 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/e013ddc241a4
16:33:17 <Cheibriados> 03amalloy02 07* 0.21-a0-199-g94b3988: Factor out some repeated map-building logic 10(14 minutes ago, 1 file, 7+ 6-) 13https://github.com/crawl/crawl/commit/94b3988db2b0
16:33:17 <Cheibriados> 03amalloy02 07* 0.21-a0-200-gb89f83b: Globally replace s == "" with s.empty() 10(15 minutes ago, 25 files, 36+ 36-) 13https://github.com/crawl/crawl/commit/b89f83b06519
16:38:35 <Cheibriados> New branch created: pull/598 (1 commit) 13https://github.com/crawl/crawl/pull/598
16:38:35 <Cheibriados> 03amalloy02 07https://github.com/crawl/crawl/pull/598 * 0.21-a0-201-g8d05549: Simplify code for making mimics cackle 10(12 minutes ago, 1 file, 2+ 4-) 13https://github.com/crawl/crawl/commit/8d05549928b5
16:55:33 <|amethyst> amalloy:  hm, it seems kind of weird to have an rvalue unwrap_desc but not an lvalue version, but I guess there's no point in having the latter if it's not being used
16:56:18 <|amethyst> amalloy:  I think you might be able to do slightly better with some compilers by changing the last line to    desc += "\n";  return desc;   because that gives you the named return value optimisation
16:56:35 <|amethyst> amalloy:  (second comment unrelated to the first other than being about the same function)
16:57:11 <amalloy> okay, i know about RVO, but what is named RVO?
16:58:12 <|amethyst> amalloy:  http://en.cppreference.com/w/cpp/language/copy_elision  ; search for NRVO
16:58:57 <|amethyst> amalloy:  basically, if you return a local object directly, the compiler is allowed to construct that object directly in the location that the return value will go
16:59:06 <|amethyst> amalloy:  so it can avoid a move
17:00:19 <|amethyst> amalloy:  avoid a move compared to the RVO even, since  desc + "\n"   requires a move while   desc += "\n"   does not
17:00:30 <|amethyst> s/requires a move/requires at least a move/
17:01:08 <|amethyst> hm
17:01:27 <|amethyst> I guess, though, nameless RVO is mandatory in C++17, while NRVO is not
17:03:01 <|amethyst> ohh
17:03:02 <|amethyst> never mind
17:03:11 <|amethyst> NRVO doesn't apply to parameters anyway
17:03:19 <|amethyst> hm
17:03:59 <amalloy> oh yeah, that's kinda funny. desc lives in the caller's stack frame already, right
17:04:22 <amalloy> because it's an rvalue reference. so you can't RVO to construct it in-place where the caller is expecting a return value
17:04:23 <|amethyst> it's probably not important anyway, because a move of a string is fairly cheap
17:04:37 <|amethyst> it's like two pointers
17:04:45 <|amethyst> (unless the small string optimisation applies, I guess)
17:06:18 <|amethyst> also, I highly approve of the != "" -> empty() change :)
17:06:54 <amalloy> |amethyst: i suspect it won't last. someone will commit another != "" soon
17:07:09 <|amethyst> add that to unbrace :)
17:07:24 <|amethyst> (I would say checkwhite, but that applies to things that aren't C++ code)
17:07:33 <amalloy> yes, i was thinking something like that. except i never run unbrace, myself, and therefore assume nobody else does. is it automated somehow?
17:07:47 <|amethyst> I run it reasonably often when I'm committing
17:08:04 <|amethyst> (which I don't do so often these days, so maybe not so often overall)
17:08:05 <amalloy> is there something that *is* run automatically? we could put it in a commit hook or something like that, or maybe travis?
17:08:44 <|amethyst> we don't have unbrace/checkwhite in a "mandatory" (default) commit hook, but you're welcome to put it in your own
17:09:14 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-200-gb89f83b (34)
17:09:15 <|amethyst> maybe we should... we do require Perl for building already, right?
17:10:28 <amalloy> we do, yes
17:11:15 <|amethyst> the problem with the != "" thing is that (as you pointed out in the scrollback, I see) it's fairly hard to do in the general case with just a regexp
17:11:24 <|amethyst> really you'd want something like a clang plugin
17:11:37 <|amethyst> and most people don't have clang
17:11:40 <amalloy> |amethyst: it's hard to *fix* with a regex, but it's easy to notice with a regex
17:11:54 <amalloy> and then issue an error message telling the human to fix it
17:13:50 <amalloy> which is why i suggested travis
17:15:02 <|amethyst> is there a way with Travis to distinguish different "levels" of failure?
17:15:07 <travis-ci> The build has errored. (master - b89f83b #8619 : Alan Malloy): https://travis-ci.org/crawl/crawl/builds/264899381
17:15:23 <|amethyst> I wouldn't like to conflate "the build failed" with "you have a formatting error"
17:15:33 <|amethyst> since that means you have to fix the latter before discovering the former
17:18:13 <advil> well, errored vs failed is already a distinction
17:19:23 <advil> I vaguely looked into this when I was a new dev and concluded there wasn't much to be done except tweak timeouts?
17:19:41 <amalloy> i'd love it if travis didn't fail 15% of builds due to failing to apt-get
17:21:26 <advil> there's also broken vs errored (neither of which is failed), but I don't know what the difference is
17:23:56 <|amethyst> we could improve that if we moved to the container-based infrastructure, but
17:24:07 <|amethyst> we'd need a whitelisted repo that had sdl2
17:24:18 <|amethyst> which probably exists by now?
17:24:43 <gammafunk> isn't sdl2 in debian stable now?
17:24:43 <advil> hm, sorry, broken groups with failed
17:25:07 <gammafunk> well, I guess I should say: "sdl2 is in debian stable"
17:25:10 <advil> failed = new branch, broken = failed in a branch with previous success
17:25:13 <gammafunk> but maybe that's not a repo we can use
17:25:50 <|amethyst> we really want the combination of 1. an oldish g++ and 2. sdl2
17:25:54 <|amethyst> which might be kind of hard
17:25:58 <gammafunk> oh I see
17:26:15 <|amethyst> alternatively, we could increase our minimum gcc version, but several servers would need upgrades
17:26:31 <|amethyst> on the gripping hand, those servers maybe need upgrades anyway?
17:27:40 <gammafunk> I'm sure they could use upgrades, but people would have to volunteer to do that I guess
17:27:49 <gammafunk> I assume CAO and CDO would
17:28:08 <gammafunk> cdo has 4.7.2
17:28:24 <gammafunk> I'm not sure to which version we'd be upgrading actually
17:28:29 <gammafunk> *which minimum version
17:29:02 <|amethyst> 5 supports pretty much all of C++14
17:29:04 <advil> "on the gripping hand" way to establish your nerd cred :-P
17:29:29 <|amethyst> I haven't even read any Niven/Pournelle
17:29:34 <advil> heh
17:29:38 <|amethyst> but I grew up on the Jargon File
17:29:53 <advil> ah I see...I have actually read a couple of those books, years ago
17:30:08 <advil> very 70s
17:30:25 <gammafunk> not sure what cbro is running either, nor if it would be a problem for cxc or cue
17:30:49 <|amethyst> (I have a copy of the New Hacker's Dictionary signed by ESR, and work with the original author of the Jargon File so could get it signed by him at any point; I need to find Guy Steele, though)
17:31:31 <|amethyst> s/author/editor/ # I suppose
17:31:52 <|amethyst> same guy who (co-)invented the quadtree
17:32:34 <advil> I suppose "way to establish your nerd cred" still applies
17:33:23 <advil> when does c++17 go live? I don't really understand the process still
17:33:42 <advil> final standard published on jul 31
17:34:23 <|amethyst> so it's already live :)
17:34:42 <advil> heh ok
17:34:47 <|amethyst> need gcc 7 for full support: https://gcc.gnu.org/projects/cxx-status.html
17:34:57 <|amethyst> or some version or another of clang
17:35:22 <|amethyst> though that's just language, not library
17:35:52 <|amethyst> https://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html  for library status
17:36:40 <|amethyst> looks like libstdc++ 7 has most but not all of C++17
17:37:13 <|amethyst> (relevant table is at the bottom of the page, labelled as C++ 201z)
17:37:15 <advil> no clang release with all of it yet though
17:38:45 <|amethyst> to be clear: I wouldn't like to require gcc 7 for building crawl
17:38:56 <|amethyst> gcc 5 I'd probably be happy with, assuming it's available in msys2
17:39:01 <gammafunk> we should require gcc 9?
17:39:04 <gammafunk> bleeding edge!
17:39:15 <|amethyst> even if that means I have to upgrade my home machine :/
17:39:22 <|amethyst> (still running 4.9 here)
17:39:30 <gammafunk> msys2 has a 5.x version, yeah
17:39:39 <gammafunk> although I think msys2 is having trouble building crawl these days
17:40:24 <gammafunk> or at least I'm not sure if there are still problems with sdl or other contribs; I had heard reports of people needing to apply patches
17:40:31 <|amethyst> you mean, crawl is having trouble being built by msys2 these days :)
17:40:51 <|amethyst> I wonder if just updating our contribs would fix that
17:42:10 <gammafunk> possibly, we tend to not update them until something is broken (or until we update to a new major version of SDL)
17:42:31 <|amethyst> gammafunk: I wonder how many security bugs we have in our outdated contribs :)
17:42:55 <|amethyst> probably fewer than we have in our own code...
17:42:56 <gammafunk> we should ask the hackers known as 4chan
17:43:44 <gammafunk> maybe I can try to update contribs and look at msys2 compilation at some point soon
17:43:56 <gammafunk> I'm not sure how many windows devs we have left that actually use it
17:44:36 <|amethyst> we could insist that people use WS4L and cross-compile :)
17:44:58 <|amethyst> but I suppose there are a lot of people not using Windows 10
17:45:21 <|amethyst> s/WS4L/WSL/  # I guess that's the abbreviation they use
17:47:24 <gammafunk> yeah, that would be cool if it worked in older windows versions
17:47:36 <gammafunk> but if msys2 having trouble is just some contribs that need updating, that's not too hard to fix
17:48:31 <|amethyst> In some cases (the setenv thing for example), it might just be that we don't run 'configure' and instead use our generated makefile
17:48:38 <amalloy> i wonder why it's WSL instead of LSW
17:49:05 <|amethyst> amalloy:  they've used that naming convention for a while
17:49:11 <|amethyst> Windows Services for UNIX for example
17:49:31 <|amethyst> I used the Windows NFS server once
17:49:48 <|amethyst> it ended up being less painful, and faster at run time, to just use SMB instead
17:49:48 <amalloy> maybe they're worried about confusion with Least Significant Word
17:49:59 <gammafunk> Microsoft: Putting the name of a competitor before our name?!
17:50:06 <|amethyst> amalloy:  I think it's more .... what gf said
17:50:14 <amalloy> yeah, my suggestion is clearly silly
17:50:40 <|amethyst> Windows S{ervices,ubsystem} for UNIX was abbreviated just "SFU"
17:51:14 <|amethyst> or "SUA" for the "Subsystem for UNIX-based Applications"
17:51:29 <|amethyst> but I think now "Windows" is actually part of the name
17:53:06 -!- Fixer_ is now known as Fixer
17:53:11 <|amethyst> I wonder if STL's mingw distro would work for building crawl?  Not sure if it has Perl though
17:53:14 <|amethyst> it does have sdl2
17:53:18 <|amethyst> https://nuwen.net/mingw.html
18:03:28 <|amethyst> sadly, it appears STL won't be giving a talk at this year's cppcon :(
18:04:25 <|amethyst> FR: rename mons_felid_revive() to meow()
18:14:36 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-200-gb89f83b (34)
19:14:48 <advil> did the setenv case get solved?  That seemed kind of mysterious / system specific
19:30:01 <amalloy> advil: i can't help but think of nancy drew or hardy boys or something. solving setenv cases
19:30:19 <advil> the case of the missing setenv
20:21:54 -!- amalloy is now known as amalloy_
20:49:26 -!- amalloy_ is now known as amalloy
22:38:01 -!- Chase is now known as ChaseSP