00:24:47 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-200-gb89f83b (34)
01:06:55 <Rotatell> ziggurnaut (L25 FoFi)  (Snake:1)
01:13:38 <Cheibriados> 03giann02 07https://github.com/crawl/crawl/pull/595 * 0.21-a0-201-g226e9f6: Several arrival vaults 10(5 days ago, 1 file, 73+ 0-) 13https://github.com/crawl/crawl/commit/226e9f60723e
01:14:09 -!- Bammboobies is now known as Bammboo
01:18:51 <Cheibriados> 03giann02 07https://github.com/crawl/crawl/pull/595 * 0.21-a0-201-g77f634a: Several arrival vaults 10(5 days ago, 1 file, 72+ 0-) 13https://github.com/crawl/crawl/commit/77f634a207ab
01:56:27 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.21-a0-200-gb89f83b
02:10:52 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-192-gcbc1537: Add docstring and comments 10(4 minutes ago, 1 file, 20+ 6-) 13https://github.com/crawl/crawl/commit/cbc1537a6113
02:21:30 -!- TAS-2012v is now known as TAS_2012v
02:26:32 <Lightli> ??plan
02:26:32 <Sequell> plan[1/1]: https://crawl.develz.org/wiki/doku.php?id=dcss:planning:release_plans
02:28:39 <gammafunk> We keep the plan in here, Lightli
02:28:43 * gammafunk points at his head
02:28:52 <gammafunk> Then it can never leak
02:29:00 <amalloy> plus there's plenty of space
02:29:13 <gammafunk> yeah, since I got the quantum disk upgrade, that's true
02:29:27 <Lightli> (seriously, development feels like it basically stopped recently)
02:29:32 <Lightli> (and it's leaving me down)
02:29:57 <gammafunk> (development goes through "ups and downs" in terms of activity)
02:30:02 <gammafunk> (so I wouldn't get too worked up about it)
02:30:33 <amalloy> basically whenever PF retires there's a lull
02:30:48 <gammafunk> I do have a copy of somewhat important projects I'd like to get in for this release, but I haven't thought of anything really release highlight worthy yet
02:31:03 <gammafunk> wand stacking and trap reform are important, but not any kind of super interesting thing
02:31:09 <bgiannan> did the steam discussion occured ? what came of it ?
02:31:26 <gammafunk> oh steam is not really a thing now, since they ended Greenlight
02:31:32 <gammafunk> now you have to pay to get on the platform
02:31:48 <bgiannan> oh shit
02:32:00 <gammafunk> It's theoretically possible for someone to do it still, but we're not really equipped as an "organization" to deal with money
02:32:02 <bgiannan> they've gone full evil it seems
02:32:11 <gammafunk> well the cost isn't very high
02:32:16 <gammafunk> but it's a cost all the same
02:32:35 <gammafunk> I want to say it's like $200 and you can recoup it from sales? Sad thing is our game is free
02:32:46 <alexjurkiewicz> I'm happy to cover the $200 straight up, fwiw
02:32:52 <alexjurkiewicz> but not do the work :)
02:33:01 <gammafunk> But also, we might not really be a good condidate for steam release when you look at the state of Tiles
02:33:18 <gammafunk> I'm not sure if they've also raised their standards or what
02:33:24 <amalloy> the game is free, but we could make bank on the hellcrawl dlc
02:33:43 <amalloy> man, there are a lot of bad games on steam. their standards can't be that high
02:33:47 <gammafunk> part of ending the program was to deal with the flood of "bad games" I think; not sure what they would think of crawl's graphical UI
02:34:03 <gammafunk> yeah, maybe someone motivated could look more into it
02:34:10 <amalloy> but i thought we also hadn't resolved the question of relicensing contributions from previous devs
02:34:27 <gammafunk> well, I don't think there's anny issue for that in terms of GPL
02:34:39 <gammafunk> the issue as I understood things was if we wanted to link to the steam library
02:34:40 <alexjurkiewicz> yeah, steam doesn't require special libs to be included or such
02:34:44 <gammafunk> for steam integration
02:34:50 <gammafunk> which we don't have to do
02:35:20 <gammafunk> for tiles art, I think we have all of that agreed upon by the art that we're actually using
02:35:37 <gammafunk> er *by the artists whose art
02:40:30 <alexjurkiewicz> that's just covered by gpl isnt it
02:54:42 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-200-gb89f83b
03:20:36 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-200-gb89f83b (34)
03:41:39 -!- amalloy is now known as amalloy_
05:24:06 -!- Amnesiac_ is now known as Amnesiac
08:26:29 -!- Mayoi is now known as erkin
08:48:29 <bgiannan> to fix https://crawl.develz.org/mantis/view.php?id=11181 i'd have to go through all possible ally summoning/creation spells/abilities/etc. ? or is there some more centralized way of fixing it ?
09:00:27 <advil> yeah, the steam fee ends up being not as bad as expected, so doesn't seem like a barrier. There might be some organizational worries.  I'd also be *extremely* worried about doing this given the dev state w.r.t. windows local tiles, which would presumably be the most popular version.
09:00:45 <advil> we have no devs who compile on windows, and very little attention is paid to windows tiles
09:00:58 <advil> we went basically a whole release where it didn't work on most graphics cards and no one noticed
09:01:20 <advil> er, very little attention is paid to local tiles
09:02:36 <advil> I believe |amethyst (?) raised the same worry before, but I think the fullscreen scaling bug really highlights the problem
09:06:35 <gammafunk> advil: how do you say it didn't work on most graphics cards?
09:07:05 <advil> %git b3e98cc5569973af76d871fdfdbd6a2b93bd11c3
09:07:05 <Cheibriados> 07advil02 * 0.21-a0-86-gb3e98cc: Ignore negative (relative) window sizes in fullscreen mode 10(7 weeks ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/b3e98cc55699
09:07:50 <advil> between that commit, and the commit it was fixing, as I understand it fullscreen on windows was broken.  Played out in different ways on different cards: in some cards it just ended up in a weird screen size, in some cards it didn't work at all.
09:08:23 <gammafunk> yeah, but I don't think it was the case that most windows tiles users couldn't play crawl
09:08:23 <advil> since fullscreen is default forced below a certain screen size, it was just flat broken on most laptops (since not working at all is more common on cheaper graphics cards)
09:08:46 <advil> I think it was a lot more than we realized
09:08:47 <gammafunk> I mean, windows tiles is the single-most popular format for playing crawl as I understand
09:09:16 <advil> for a lot of people, it worked but just looked bad
09:09:41 <gammafunk> well, either way, yeah our Tiles app is just in maintanence mode
09:09:44 <advil> if you knew how to force non-fullscreen or were playing at a big enough resolution, it was ok as I understand it
09:10:03 <gammafunk> I think we can fix bugs as they pop up but it might not really be up to steam standards
09:10:31 <gammafunk> some of its functionality is probably better than most steam games, but perhaps they'd have some requirements we'd need to implement in their review
09:11:08 <gammafunk> Either way I'm not sure that the benefit of going on steam is quite worth the hassle
09:11:51 <advil> %git 45bdf812ecb1ac09e19663bf9380a0f58f1c2b4c
09:11:51 <Cheibriados> 07PleasingFungus02 * 0.19-a0-1554-g45bdf81: Make tiles resolution options affect fullscreen 10(11 months ago, 2 files, 8+ 5-) 13https://github.com/crawl/crawl/commit/45bdf812ecb1
09:12:01 <advil> is the commit that broke fullscreen, I think
09:12:18 <gammafunk> I say this in part because steam is a thing that would have to be regularly managed/updated probably, and without someone really stepping up to deal with that longterm
09:12:18 <advil> so that's about 8 months to notice the bug
09:12:38 <gammafunk> it's possible that we'd get on steam and there'd be an issue eventually but no one would be around to deal with it
09:12:43 <advil> yeah
09:12:59 <advil> I agree, without a community manager it's problematic
09:14:10 <advil> / a release manager
09:15:41 <advil> iirc someone else needs to be the one who signs the complicated legal thing, too
09:15:46 <gammafunk> oh, yeah
09:15:47 <advil> er
09:15:51 <advil> *someone, not someone else
09:16:02 <gammafunk> Dealing with complaints from steam users sounds like a dream job!
09:16:11 <bgiannan> The only advantage of steam as far as i can see is exposure. and it's not even guaranteed if not properly released. Does crawl needs more exposure ?
09:16:24 <gammafunk> yeah, probably it doesn't, like you're implying, bgiannan
09:16:33 <gammafunk> not that more exposure is bad, but is it feasable
09:16:48 <gammafunk> that said, if someone who seemed reliable and motivated came along and offered to do it long-term
09:16:53 <gammafunk> I think we'd be supportive
09:17:12 <gammafunk> it's not a money issue, but probably more a time/commitment issue
09:17:54 <gammafunk> Some devs possibly don't like the steam licensing too much, or the idea of a gpl-licensed game going on a commercial platform like that
09:18:29 <gammafunk> now if my crawl 3D+VR patch works out, we're for sure going on steam
09:18:42 <advil> I think that's ok without linking against the steam lib (so you don't get the overlay stuff, which is sort of too bad)
09:18:47 <advil> that = gpl issues
09:19:13 <gammafunk> yeah, I know a couple have devs have expressed a bit of reservation about even that route
09:19:26 <gammafunk> Well, one I can think of, who's also a CS prof who knows the Jargon File really well
09:19:33 <advil> hm, we'd probably also have to deal with mac signing again to distribute the mac version on steam
09:19:34 <gammafunk> but I don't think that's actually a massive complaint
09:19:37 <gammafunk> oh, god
09:19:47 <gammafunk> Mac signing is basically a "no"
09:19:50 <advil> haha
09:19:55 <gammafunk> geekosaur tried valiantly for many versions
09:20:07 <gammafunk> until someone can well and truly figure out that nightmare...
09:20:09 <advil> I'm not unwilling to try to deal with mac signing again in the future
09:20:16 <advil> but I haven't done it yet
09:20:25 <advil> it's a lot better documented now than when geekosaur tried, I think
09:20:31 <gammafunk> Just let us know a few hours in advance before you try
09:20:47 <gammafunk> If something happens we want to have emergency personel on call
09:20:57 <gammafunk> *personnel
09:21:19 <advil> heh
09:21:40 <advil> (also if PF ever unretires again maybe he'll know how to do it from his solo gamedev project, he was working on that a few months ago)
09:21:57 <gammafunk> I recall PF considering giving up on mac-native builds due to all the hassle, actually
09:22:03 <advil> yes, there is that
09:22:03 <gammafunk> but possibly he's figured somet things
09:22:16 <gammafunk> Some games are asking mac/linux users to use Wine
09:22:22 <gammafunk> and that's working out OK for them
09:22:42 <gammafunk> Cogmind does that, for instance
09:22:42 <advil> ah
09:22:45 <advil> spelunky de facto does that
09:22:59 <gammafunk> For a game dev, it's a headache to deal with all the build platforms, esp if you consider linux support means maybe doing a few kinds of packages
09:22:59 <advil> I have only ever really bothered with wine for about 1 game, and that is spelunky
09:23:06 <gammafunk> yeah
09:23:21 <gammafunk> I played classic spelunky a ton but never really played much of spelunky HD
09:23:32 <gammafunk> in part due to it not working as well, maybe it's ok now
09:23:38 <advil> works quite well in wine! but that's a huge barrier for entry still
09:23:52 <advil> I had to be really talked into bothering
09:24:02 <gammafunk> I mean, on linux it's not too bad since you just install a package really
09:24:08 <gammafunk> For Mac maybe it's a bunch more setup?
09:24:25 <gammafunk> I know it has a port and a fink install, but I guess Mac users don't use those so much
09:24:30 <gammafunk> s/port/macport/
09:24:40 <advil> well, wine configuration is insane at the best of times
09:24:54 <gammafunk> well for the wine games I've used recently, it's been 0 wine config
09:25:06 <advil> steam/spelunky doesn't require anything special, but you do need to find a step-by-step recipe to follow on the internet somewhere
09:25:11 <gammafunk> just ran out of the box, as they say
09:25:19 <gammafunk> Yeah, maybe some games just require more setup
09:25:26 <advil> granted I haven't tried to do much with wine in a year or so, once I got a steam install going on one computer
09:25:26 <gammafunk> Cogmind is very painless
09:25:45 <gammafunk> I'll have to try spelunky HD again at some point
09:26:16 <advil> the game itself requires no effort so if other things work out of the box, that should too
09:26:22 <advil> it was really steam that was tricky
09:26:37 <gammafunk> oh, do you have wine running through steam?
09:26:42 <gammafunk> I have linux steam but I don't really use it
09:26:55 <advil> yeah, since that's how I own spelunky
09:27:54 <advil> I have played a few other things through it, got thief/thief 2 running, also (insanely) STALKER
09:29:00 <gammafunk> yeah, once you get into games with sophisticated graphics, I'm sure things get worse
11:01:55 <Eksell> Torious (L27 DEWz) ASSERT(can_reach()) in 'melee-attack.cc' at line 2345 failed. (Pan)
11:02:26 <advil> !crashlog
11:02:38 <Sequell> 17613. Torious, XL27 DEWz, T:55270 (milestone): http://crawl.xtahua.com/crawl/morgue/Torious/crash-Torious-20170816-150133.txt
11:02:53 <advil> I guess !nick may be somewhat slow
11:02:56 <advil> !crashlog *
11:02:57 <Sequell> 17908. Torious, XL27 DEWz, T:55270 (milestone): http://crawl.xtahua.com/crawl/morgue/Torious/crash-Torious-20170816-150133.txt
11:04:50 <gammafunk> shouldn't be
11:05:01 <gammafunk> it's possible just some temporary sequell slowness
11:05:14 <gammafunk> !crashlog
11:05:15 <Sequell> 17613. Torious, XL27 DEWz, T:55270 (milestone): http://crawl.xtahua.com/crawl/morgue/Torious/crash-Torious-20170816-150133.txt
11:05:20 <gammafunk> seems pretty fast, yeah
11:06:03 <gammafunk> I'm not sure what different arguments being present can cause slowdowns, however.
11:06:28 <gammafunk> usually lg queries are pretty fast and lm queries are quite slow
11:06:45 <advil> !crashlog
11:06:45 <Sequell> 17613. Torious, XL27 DEWz, T:55270 (milestone): http://crawl.xtahua.com/crawl/morgue/Torious/crash-Torious-20170816-150133.txt
11:06:53 <advil> 2 seconds, unacceptable!
11:07:08 <gammafunk> Where is our 99.999999% uptime?
11:07:27 <advil> heh
11:41:25 <advil> huh why does str_to_skill default to fighting?
11:57:51 <advil> wonder if there is much lua that relies on that
13:20:59 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.21-a0-200-gb89f83b (34)
13:37:26 -!- amalloy_ is now known as amalloy
13:50:08 <Cheibriados> 03giann02 {amalloy} 07* 0.21-a0-201-g8501b10: Several arrival vaults 10(5 days ago, 1 file, 72+ 0-) 13https://github.com/crawl/crawl/commit/8501b1010993
13:56:15 <gammafunk> amalloy: oh, I was going to remove no_item_gen from that one vault
13:56:24 <gammafunk> placing 4 gold pieces isn't really a reason to have no_item_gen
13:56:26 <gammafunk> but I can do that later
13:57:12 <gammafunk> also not really sure why that one has no_rotate
13:57:24 <gammafunk> doesn't much look like any special shape or anything
13:58:06 <gammafunk> maybe the decor on that graveyard one should be cleaned up as well
14:00:40 <amalloy> whoops.i didn't pay much attention to the tags this time
14:01:03 <amalloy> my job is just to do merges so shoddy that real devs like gammafunk are compelled to fix them up now that it's committed
14:01:57 <gammafunk> fine but I want a 33% cut of your pay, only fair
14:09:12 <Jorgrell> Unstable branch on crawl.jorgrun.rocks updated to: 0.21-a0-201-g8501b10 (34)
14:23:18 -!- Fixer_ is now known as Fixer
14:31:31 <amalloy> if you can negotiate a 50% raise for me, you can have 33% of the new total
14:35:30 -!- MarvinPA_ is now known as MarvinPA
14:38:18 <travis-ci> The build passed. (master - 8501b10 #8625 : Benoit Giannangeli): https://travis-ci.org/crawl/crawl/builds/265257386
15:08:17 <Cheibriados> 03advil02 07https://github.com/crawl/crawl/pull/585 * 0.21-a0-184-g340322d: Clean up some bool arguments for skill_level_to_diffs 10(24 hours ago, 3 files, 25+ 27-) 13https://github.com/crawl/crawl/commit/340322dc412c
15:08:17 <Cheibriados> 03advil02 07https://github.com/crawl/crawl/pull/585 * 0.21-a0-185-ge6a8068: Implement training target backend 10(11 minutes ago, 13 files, 239+ 9-) 13https://github.com/crawl/crawl/commit/e6a8068e981a
15:14:27 <gammafunk> ...soon to be rip target_skill.lua
15:14:29 <gammafunk> end of an era
15:16:33 <amalloy> there's still time. you could assassinate advil
15:16:52 <johnny0> advil: i have some local tiles fixes on the way for window resizing, mouse lag, ignoring input when not in focus, etc.
15:16:57 <amalloy> just start training stealth, and in about 3 XLs you'll have enough to sneak up on him
15:17:23 <advil> lol
15:17:28 <advil> johnny0: nice!
15:17:47 <advil> johnny0: are there still mouse lag problems?
15:19:15 <gammafunk> oh cool
15:19:29 <gammafunk> I had heard some report that there was still mouse lag in some ways
15:19:38 <gammafunk> but I didn't get many details
15:19:42 <johnny0> in master, targeting mode is a problem
15:19:57 <gammafunk> for mouse lag?
15:20:04 <johnny0> yeah
15:20:16 <gammafunk> hrm, I see, that's not an area I had looked at
15:20:30 <gammafunk> since the lag was more "global" in the issue I fixed
15:20:39 <gammafunk> but that's great
15:21:07 <gammafunk> and yeah one of the early things I hacked in for SDL2 was a fix so resizing wouldn't crash the app
15:21:16 <gammafunk> but some of the refactoring I did introduced that very lag
15:21:35 <gammafunk> since I don't understand SDL/graphics loops all that well
15:24:10 <johnny0> gammafunk: it basically boils down to sdl pushing mouse events faster than crawl can handle them -- your fix was the right idea, just not applied aggressively enough
15:24:32 <gammafunk> I see, will be cool to read your patch then
15:24:42 <gammafunk> yeah I had to go read my commit comment to remember what even was happening
17:58:40 <|amethyst> re above conversation, if we don't link to the proprietary steam library, and we include a written offer to obtain the full source code (i.e. put a link to our github somewhere in the docs), I don't see any licensing problems with a Steam version
17:58:40 <Sequell> |amethyst: You have 1 message. Use !messages to read it.
18:00:17 <|amethyst> AFAIK, unlike Apple, Valve doesn't try to impose extra restrictions on the code they distribute through their store
18:03:34 <Lantell> Unstable branch on underhound.eu updated to: 0.21-a0-201-g8501b10 (34)
18:06:21 <gammafunk> |amethyst: thanks for the clarification. Maybe I can remember to add a little "steam dossier" to the plan
18:06:39 <gammafunk> that might be more like an implementable, I guess
18:22:14 <|amethyst> !tell johnstein I'd suggest can_place_on_trap rather than the three-way or on the trap type... the only problem is that the function requires a mon_type, though that is only relevant to tentacle segments at the moment
18:22:14 <Sequell> |amethyst: OK, I'll let johnstein know.
18:32:14 <|amethyst> !tell johnstein purely formatting: I see a lot of long lines, and I also don't like putting a comment line between } and else, but the latter is just personal preference (I'd move it into the else { } block).  I'd also capitalise sentences and use complete punctuation, especially but not only in docstrings
18:32:14 <Sequell> |amethyst: OK, I'll let johnstein know.
18:33:12 <|amethyst> !tell johnstein Especially in docstrings because those get reflowed when formatted by doxygen, so the missing punctuation isn't at the end of the line anymore
18:33:13 <Sequell> |amethyst: OK, I'll let johnstein know.
18:33:35 <|amethyst> !tell johnstein and I don't like // without a space after it, but again personal preference
18:33:35 <Sequell> |amethyst: OK, I'll let johnstein know.
18:42:57 <|amethyst> !tell johnstein Oh, and it looks like that function is used only in spl-summoning.cc, so I'd make it function static, prepend an _, and move the declaration from the header to the top of the .cc file.
18:42:58 <Sequell> |amethyst: OK, I'll let johnstein know.
18:43:59 <|amethyst> !tell johnstein Or you could move the definition above the uses, so you don't need a separate prototype
18:43:59 <Sequell> |amethyst: OK, I'll let johnstein know.
18:44:37 <johnstein> thanks for the info |amethyst
18:44:37 <Sequell> johnstein: You have 6 messages. Use !messages to read them.
18:44:49 <|amethyst> johnstein:  I also added a comment to the PR with the same info
18:44:58 <|amethyst> johnstein:  since I guess that's better for posterity :)
18:47:15 <johnstein> what is the line length limit? I usually try not to be like get than other lines in the same files. I assume it's 72 or 80
18:55:36 <amalloy> "what is the line length limit" can only really be answered with "nebulous and unenforced"
19:00:38 <|amethyst> it's not very well-enforced, but it's 80
19:00:48 <|amethyst> it's mentioned in an aside in coding_conventions.txt
19:00:58 <|amethyst> "If the parameter list of a method runs longer than a line length (80 columns)"
19:01:17 <|amethyst> it just kind of assumes you're not going to write lines longer than 80 columns, because why would you do such a thing? :)
19:10:36 -!- amalloy is now known as amalloy_
19:13:39 -!- amalloy_ is now known as amalloy
19:48:28 <advil> because we use editors that render that requirement obsolete :-P
19:49:25 <advil> though now I usually do try to make it look basically ok in git diff in a regular sized terminal
19:54:33 <amalloy> there are much better things to do with your screen space than an ultra-wide editor window
19:55:10 <amalloy> and as someone who's visually impaired i strongly prefer lines no longer than 100 characters because i can fit that comfortably on my screen at a large size
19:56:39 <Floodkiller> pfft, I just have everything on a single line
20:00:22 <Cheibriados> New branch created: floodkiller (1 commit) 13https://github.com/crawl/crawl/tree/floodkiller
20:00:22 <Cheibriados> 03amalloy02 07[floodkiller] * 0.21-a0-202-g352b5c1: Floodkiller-approved style changes 10(25 seconds ago, 729 files, 729+ 418185-) 13https://github.com/crawl/crawl/commit/352b5c1190cb
20:00:44 <amalloy> the millimarvins flow through me
20:01:12 <Floodkiller> perfect editing environment
20:01:23 <|amethyst> advil: my editor runs in a regular sized terminal :)
20:02:01 <|amethyst> I could make it wider, but then I can't fit as many on the screen at the same time
20:05:15 <amalloy> oh right, github's contributions graph only includes commits to master. so i can leave that joke commit up without feeling bad about my fraudulent stats
20:09:13 <travis-ci> The build failed. (floodkiller - 352b5c1 #8627 : Alan Malloy): https://travis-ci.org/crawl/crawl/builds/265380138
20:09:34 <advil> at my current editor font size, my external monitor here is about 235 characters wide
20:10:40 <Floodkiller> the failed build makes it, I think
20:11:11 <|amethyst> advil:  so you could almost look at three files at a time at full screen height
20:20:44 <shmup> about how long does it take you to make a debug build?
20:24:12 <amalloy> shmup: https://gist.github.com/amalloy/e328dbbceacb4dae38a8ad6c2728dd48
20:25:14 <amalloy> er
20:25:55 <amalloy> sorry, failed to copy the timing part of the output there. i've updated the gist to include timing
20:26:45 <shmup> yeah, okay. msys2 is real bad at this.
20:27:13 <shmup> i should either make a nicer Dockerfile, or use the Linux Subsystem in Win10. regardless I wanna use work laptop when I'm hackin on this
20:27:17 <amalloy> well, this is definitely the fastest computer i've used to compile
20:27:39 <amalloy> i think on my old laptop it was like 4 minutes at -j6 or something
20:28:43 <amalloy> shmup: for some low-hanging fruit you should at least use -j if you're not already
20:30:25 <shmup> let's do some benchmarks
20:30:42 <|amethyst> probably don't use -j without a number after it
20:30:45 <|amethyst> :)
20:31:07 <shmup> i went with 4 to see what time i could get
20:31:18 <|amethyst> how many cores do you have?
20:31:18 <amalloy> |amethyst: good test of your fans
20:31:21 <shmup> 4
20:31:44 <|amethyst> yeah, 4 or 5 will probably be fastest
20:33:07 <shmup> 2m36s on an i7 7820HQ clocked at 2.9
20:33:17 <amalloy> that's not too bad
20:33:18 <shmup> you have a pretty nice setup, amalloy :)
20:53:32 <shmup> a thing i wanna tackle on msys2 is using gdb. i don't really know how to run crawl with it. maybe i have to run crawl.exe manually and attach a debugger to it?
20:53:47 <shmup> my only experience with debugging was made easy with things like visual studio, or pycharm
20:54:45 <gammafunk> shmup: to run gdb you can do something like "gdb crawl.exe"
20:54:49 <gammafunk> from the dir with the binary
20:54:58 <shmup> i've done that, gammafunk. and i can set breakpoints on the functions i care about
20:55:09 <shmup> but doing run comes up with problems
20:55:23 <gammafunk> does it not launch the console window?
20:55:30 <shmup> Error initialising console input mode.[Thread 25404.0x5934 exited with code 0]
20:55:33 <gammafunk> yes
20:55:33 <shmup> nah, that'd be nice!
20:55:40 <gammafunk> so for that
20:55:54 <gammafunk> hrm, I'm going to get that gdb attaching to a process won't work
20:55:56 <gammafunk> but you could try that
20:56:03 <gammafunk> alternately you could do a tiles build
20:56:12 <gammafunk> tiles would work fine for directly running crawl through gdb
20:56:19 <gammafunk> but you can also try attaching a process
20:56:22 <shmup> haha, i've been lazy about tackling that. was having a problem compiling one of the contribs. i forget which, sdl related (not sdl itself)
20:56:34 <gammafunk> oh, yeah you may need a patch for that
20:56:41 <gammafunk> I thinke johnstein or |amethyst knows which
20:56:42 <shmup> and so i was gonna just slip by with console for a bit. maybe not though, since you're offering a reason to "fix it"
20:56:58 <shmup> johnstein knows about a thing yeah, but it isn't related to this IF i remember correctly
20:56:58 <gammafunk> but you can try starting console crawl
20:57:01 <gammafunk> then attaching
20:57:03 <shmup> i'll bug him again (hey johnstein)
20:57:11 <gammafunk> not sure if it will work in msys2 though
20:57:12 <gammafunk> let me try it
20:57:21 <shmup> cool thanks
20:57:43 <johnstein> PINGS
20:57:43 <johnstein> I'm trying to assess the problem
20:57:46 <shmup> well let me try compiling first. hey johnstein
20:57:49 <johnstein> reading scrollback
20:59:44 <shmup> to remind you of my problem https://gist.githubusercontent.com/shmup/07cd4d40e3b776da320d79f9b62b676b/raw/4032fb830f305294485ce8b5a8193b65eaf56066/gistfile1.txt
21:01:12 <johnstein> ah the sdl patch
21:01:25 <shmup> i can get beyond that, however, removing something from the Makefile in sdl-mixer
21:03:18 <shmup> removing it leads to: https://gist.githubusercontent.com/shmup/74b19356ea2318d0d764fe522906eb36/raw/82a5ed3f415a90bdb5fcbcb08f5a40de00921307/gistfile1.txt
21:11:16 <gammafunk> I was able to attach to the crawl process
21:11:30 <gammafunk> but I need to compile a version with debugging symbols to see if I can actually set breakpoints etc
21:11:50 <gammafunk> so I'm compiling in msys2 now, which will take a bit
21:12:01 <gammafunk> I think you ahve to get the process ID from the windows task manager though
21:12:05 <johnstein> I only have 2 cores, so it takes a while
21:12:06 <gammafunk> couldn't seem to get it from ps aux
21:16:40 <shmup> thanks for lookin into it
21:21:52 -!- amalloy is now known as amalloy_
21:49:48 <johnstein> |amethyst, I can't find the spot where there's a comment between the } and else
21:54:05 <|amethyst> johnstein:  sorry, missed that it was fixed in a later commit
21:54:25 <johnstein> ah ok. yea, it had bugged me too
21:55:00 <gammafunk> shmup: ok, I tried it, and a breakpoint did work in a build with debugging symbols
21:55:09 <johnstein> yay, made my first static C++ function
21:55:19 <gammafunk> I built it with debug-lite target (gives you debugging symbols, not a full debug build with all the message spam though)
21:55:24 <gammafunk> but would work fine with full debug
21:55:38 <gammafunk> and then ran crawl console as just "start crawl.exe"
21:55:43 <gammafunk> you could launch it from wherever
21:55:46 <|amethyst> johnstein:  hm, so you have that commit about monster casters
21:55:52 <gammafunk> from there, go to the windows task manager and get the process ID
21:56:00 <gammafunk> and once you have that number N
21:56:11 <|amethyst> johnstein:  but does the monster code even call handle_trap_at_target_location?
21:56:14 <gammafunk> from the dir with your crawl.exe do: gdb -p N
21:56:33 <gammafunk> gdb loads the data it needs from crawl.exe but attaches to the process you give
21:57:02 <gammafunk> and then the running crawl app is "frozen" while you set up gdb breakpoints etc
21:57:25 <gammafunk> and you can continue with `c' when you're ready, which causes your crawl to continue execution
21:57:59 <|amethyst> johnstein:  oh, there isn't even a real monster version of the spell... it looks like it doesn't do anything, and any ghost that did get it (probably a transferred game) would just consider the spell to be always a waste of time
21:58:31 <|amethyst> johnstein:  so I think the commit about the monster handling isn't needed
21:59:49 <|amethyst> johnstein:  I'd just add to the documentation of handle_trap_at_target_location that it's only for player casters
22:00:25 <|amethyst> johnstein:  (a monster with one of those spells would presumably just be allowed to cheat and pick a non-trap location)
22:00:43 <johnstein> |amethyst, yea, that was my eventual conclusion. it would be ok if it cheated
22:03:53 <johnstein> |amethyst, took me a while to convince myself (and let myself be convinced) that I didn't have to do much to handle non-players casting it.
22:04:15 <johnstein> |amethyst, I'll add a note clarifying.
22:27:15 <|amethyst> johnstein:  I mean you don't have to do anything at all
22:27:25 <|amethyst> johnstein:  because this function won't be called for monsters
22:27:37 <|amethyst> johnstein:  so you can remove the 'caster' parameter and that if
22:28:00 <johnstein> |amethyst, ok. so the only reason I thought I needed it was just in case in the future, if a monster can cast it, it will show the messages
22:28:23 <johnstein> but I didn't know if that's ever something that is worried about.
22:30:53 <|amethyst> I think if we did (re-) implement one of those spells for monsters there wouldn't be a need to call the function
22:33:09 <johnstein> ok. so in the cast_lightning_spire() function, we'd just do: if (caster == player) handle_traps...
22:33:16 <johnstein> makes sense.
22:33:34 <|amethyst> johnstein:  monsters wouldn't call that function
22:33:42 <johnstein> ooh
22:33:43 <|amethyst> cast_* are for player casters only
22:34:41 <|amethyst> the monster version of the spell would 1. pick a valid spot, 2. create the monster there
22:35:16 <johnstein> I suppose I should pick a monster spell bug next
22:36:17 <johnstein> also, I renamed "spret_type handle_trap_at_target_location" to "static spret_type _cast_on_trap_location"
22:36:38 <johnstein> does that make sense? wasn't sure if I should avoid cast_*
22:36:49 <|amethyst> I think cast_* isn't good there
22:36:55 <johnstein> but figured since it was returing spret_spell, maybe it was ok. also made the name shorter
22:37:12 <|amethyst> hm
22:37:26 <johnstein> I was wondering about that. I've spent longer than I'd like to admit on trying to come up with a good name that wasn't too verbose and fit with the existing 'style'
22:38:22 <|amethyst> _check_fixed_summon_loc
22:39:03 <|amethyst> some of those other checks (monster on the grid, etc) could be moved there eventually
22:42:07 <|amethyst> I think something about "fixed summon" is more important than something about traps
22:44:11 <johnstein> yea, makes sense
22:59:09 <johnstein> is the _ for the static functions a C++ thing? or just a convention?
23:04:33 <geekosaur> just a convention
23:04:47 <geekosaur> mpre specifically, it's a C/C++ convention that leading underscore means private/internal. this overlaps with "static" storage class
23:04:48 <geekosaur> / linkage
23:18:13 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-193-g787a6b5: Make code cleaner 10(83 minutes ago, 2 files, 64+ 52-) 13https://github.com/crawl/crawl/commit/787a6b574cb1
23:18:13 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-194-g56c20aa: Remove caster argument from a function that only players cast 10(46 minutes ago, 1 file, 4+ 8-) 13https://github.com/crawl/crawl/commit/56c20aaa4f41
23:18:13 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-195-g2dc6b8b: Rename a function 10(36 minutes ago, 1 file, 13+ 11-) 13https://github.com/crawl/crawl/commit/2dc6b8bd8f9b
23:18:55 <johnstein> thanks for the help |amethyst.
23:23:24 <Cheibriados> 03johnstein02 07https://github.com/crawl/crawl/pull/594 * 0.21-a0-195-g1cf66f8: Rename a function 10(41 minutes ago, 1 file, 13+ 11-) 13https://github.com/crawl/crawl/commit/1cf66f8db8cd
23:46:40 <shmup> wonderful gammafunk, thanks for looking into it
23:48:18 <shmup> do devs not care about squashing
23:55:36 <nikheizen> shmup: that's a PR. the devs seem to like to squash when actually merging
23:56:12 <shmup> ive been asked to squash my commits in PRs is prob why i thought to ask but
23:56:20 <shmup> thats a good point
23:56:39 <shmup> in other projects i meant
23:59:12 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.21-a0-201-g8501b10 (34)