00:06:41 -!- amalloy_ is now known as amalloy
00:14:28 -!- amalloy is now known as amalloy_
00:15:53 -!- amalloy_ is now known as amalloy
00:16:51 <Cheibriados> It's possible to get two ammo gifts from Okawaru in the same action 13https://crawl.develz.org/mantis/view.php?id=11138 by Mattlistener
00:36:47 <ProzacElf> any idea why console would look like this https://snag.gy/w8f4tb.jpg ?
00:36:52 <ProzacElf> after a win10 update
00:37:14 <ProzacElf> (not my machine or game, just wondering if it's come up)
00:39:26 <gammafunk> does look like it's having trouble properly rendering those block characters; I assume those aren't #
00:44:38 <gammafunk> johnstein: I've noticed that cbro has 0.20 listed as 0.20.0-15
00:44:48 <gammafunk> is cbro not updating stable?
00:45:17 <gammafunk> %git stone_soup-0.20
00:45:17 <Cheibriados> 07gammafunk02 * 0.20.1-1-g8b1a95d: Fix some typos in the changelog 10(6 days ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/8b1a95d66d7d
00:45:21 <gammafunk> %git stone_soup-0.20~2
00:45:21 <Cheibriados> 07advil02 {gammafunk} * 0.20.0-45-g989a428: Convert a segfault into an ASSERT failure 10(8 days ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/989a428c66d2
00:45:31 <gammafunk> about 31 commits behind
01:04:00 <Wander> Question:    In a vault, is there a way to take a bunch of placeholder glyphs (like ABCDE), randomly select only one of the glyphs (say B), substitute it with something else (like a tree), and then turn all the other glyphs into something else (like floor)?))
01:04:10 <Wander> kind of like the opposite of SUBST
01:08:54 <geekosaur> I think you want NSUBST?
01:09:43 <Wander> doesn't NSUBST only work with one thing at a time?
01:10:20 <Wander> like you can't do   NSUBST: ABCDE:t / *:.
01:10:24 <Wander> I don't think
01:12:10 <Wander> I dunno, I'm confused lol
01:13:42 <geekosaur> there's at least one vault using multiple symbols
01:14:17 <geekosaur> yeh, there's several. TIAS
01:14:29 <Wander> :o  name of one?
01:15:47 <gammafunk> yeah you can use multiple in an NSUBST
01:15:52 <geekosaur> !vault nicolae_vaults_network_barrier
01:15:53 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/branches/vaults_rooms_standard.des#L1765
01:15:54 <gammafunk> and I think it treats them all like one sysmbol
01:15:56 <gammafunk> *symbol
01:16:03 <geekosaur> so NSUBST: ABCDE = 1:t / *:w or some such
01:16:15 <gammafunk> so one of those become a t, the rest become w
01:16:18 <geekosaur> probably not w for the replacement but I didn';t want to use a space
01:16:24 <geekosaur> becuase hard to read for an example
01:16:32 <geekosaur> but space is likely what you really want
01:16:59 <gammafunk> Wander: typically what you're describing is done via SHUFFLE
01:17:13 <gammafunk> you simply do e.g. SUBST: A = t, BCDE = .
01:17:22 <gammafunk> but before that you do SHUFFLE: ABCDE
01:17:31 <Wander> hm, let me try some things out
01:17:40 <gammafunk> so which symbol becomes the t is randomized from the SHUFFLE
01:17:57 <Wander> !  I think that's what I want
01:17:59 <Wander> thanks
01:21:07 <johnstein> gammafunk, I will look
01:22:41 <johnstein> cron job script looks good (which makes sense because updating was working before)
01:24:10 <johnstein> log says no updates since jun 25. could not lock ... dgamelaunch-config/locks/crawl-update.lock
01:24:22 <johnstein> hmmm
01:24:31 <johnstein> ??rebuild
01:24:32 <Sequell> rebuild[1/2]: http://crawl.akrasiac.org/rebuild/ http://underhound.eu:81/rebuild/ http://crawl.berotato.org/crawl/rebuild/ http://crawl.xtahua.com/rebuild/ https://crawl.jorgrun.rocks/rebuild/ Bug |amethyst or Nap.Kin for CDO. Use your powers wisely.
01:31:15 <johnstein> gammafunk, I don't know why the lock file didn't clear up after the last successful build (appears to be june24/25).
01:32:22 <johnstein> I cleared the lock and I'm trying again.
01:40:02 <Rotatell> Stable (0.20) branch on crawl.beRotato.org updated to: 0.20.1-1-g8b1a95d
01:47:00 <gammafunk> johnstein: cool, thanks
01:47:03 <gammafunk> !nick genericusername johnl023
01:47:04 <Sequell> Mapping genericusername => johnl023
01:47:22 <johnstein> thank you for the heads up
01:52:33 <Rotatell> Unstable branch on crawl.beRotato.org updated to: 0.21-a0-110-g8e1637b (34)
02:52:10 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.21-a0-110-g8e1637b
03:03:53 <Lightli> ??plan
03:03:53 <Sequell> plan[1/1]: https://crawl.develz.org/wiki/doku.php?id=dcss:planning:release_plans
03:33:40 <checkers> !tell hellmonk ice form is the only starting book tmut spell that has another school now, please either add summoning to sticks to snakes or remove ice
03:33:40 <Sequell> checkers: OK, I'll let hellmonk know.
05:04:06 -!- amalloy is now known as amalloy_
05:21:41 <Nivim> checkers: Is there a new rule about starting book spells other than "not too powerful but ok"?
05:23:33 <nikheizen> Nivim: hellcrawl no longer has poison as a spell school or a skill.
05:24:01 <nikheizen> So spider form, for example is pure tmut. Same with Alistairs.
05:24:56 <Nivim> So it's some concept of consistency, or perhaps which schools you're not encouraged to branch into?
05:25:32 <nikheizen> I think checkers is just teasing him.
05:26:11 <nikheizen> (or if you mean why the schools were removed, then the same answer as everything else in hellcrawl)
05:26:21 <nikheizen> because hellmonk wanted it that way
09:57:22 <advil> !source mlist_index_to_letter
09:57:23 <Sequell> Can't find mlist_index_to_letter.
09:57:25 <advil> hm
09:58:40 <advil> does that exist?? git grep gives nothing
10:00:07 <advil> maybe just vestigial
10:02:55 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-113-gd9d4306: Remove some tiles-related ifdefs from state.h 10(61 minutes ago, 2 files, 3+ 6-) 13https://github.com/crawl/crawl/commit/d9d43066f64d
10:02:55 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-114-g152c280: Remove an ifdef in message.h 10(53 minutes ago, 2 files, 5+ 4-) 13https://github.com/crawl/crawl/commit/152c28065330
10:02:55 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-115-gf2bfa64: Relocate an ifdef'd function from view.h to tiles-build-specific.h 10(13 minutes ago, 8 files, 14+ 6-) 13https://github.com/crawl/crawl/commit/f2bfa64d1e77
10:02:55 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-116-ged80c03: Remove an old function definition 10(35 seconds ago, 1 file, 0+ 4-) 13https://github.com/crawl/crawl/commit/ed80c03dcda4
12:02:30 <Palyth> anyone around that help critique some vault changes for a vault i submitted?
13:54:34 <bh> !seen lasty
13:54:35 <Sequell> I last saw Lasty at Mon Jul  3 22:12:12 2017 UTC (4d 19h 42m 22s ago) parting ##crawl, saying 'chanpart'.
14:51:44 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-117-g59789f0: Another batch of removing/relocating ifdefs from .h files 10(23 minutes ago, 14 files, 41+ 31-) 13https://github.com/crawl/crawl/commit/59789f0ffbf6
14:51:44 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-118-g60851c4: Use a different strategy for `flash_monster_colour` 10(2 minutes ago, 6 files, 9+ 13-) 13https://github.com/crawl/crawl/commit/60851c457829
14:55:04 <kenran> advil: you're the one who merged the "diagonal running" pr, right?
14:55:13 <advil> yep, that's me
14:56:14 <kenran> thanks first of all :)
14:56:39 <advil> no problem
14:57:46 <kenran> in reply to your examples: you were right about the last case being the one I had in mind. I actually tried another strategy first that - I believe - would have solved both other cases as well, but it was the one I considered "too annoying".
14:58:13 <kenran> I moved a bit through some random levels and it felt like there was no sense in moving diagonally anymore because I would be stopped "everywhere"
14:58:41 <kenran> I didn't see the first example you gave though, and it's very interesting:
14:58:45 <advil> yeah, that makes sense
15:00:25 <Wander> hi again
15:00:29 <advil> I think that kind of case would naturally come up somewhere like shoals, and maybe less so on most other levels
15:00:38 <advil> diamond D levels maybe
15:01:09 <kenran> nevermind, I thought I had found a bug, but it's nothing. I must have overlooked something
15:01:20 <advil> hello Wander
15:02:39 <advil> kenran: it wasn't clear to me that there's a reasonable solution to that first example that isn't annoying for the other cases; maybe if you track the last few moves?  So maybe there's nothing more to be done
15:03:39 <kenran> yes, thinking about it I came to the same conclusion: I don't think it's possible just from the data that we track right now, because it's always just checking turn to turn
15:04:29 <kenran> maybe one could make it rc-configurable to use "strict" diagonal shift-running (behaving exactly like the horizontal/vertical running)
15:05:33 <advil> that may not be worth the effort
15:05:38 <kenran> but I guess it might not be worth it; most probably people don't have a problem with how it works
15:06:08 <kenran> correcting the above: it's possible of course, but not without the mentioned drawbacks
15:09:23 <kenran> I think I might search mantis for another easy bug to fix :) I encountered one that already had an old ticket, but that looks too difficult for me to recreate I think
15:09:50 <advil> sounds good...was the diagonal movement one from mantis?
15:11:50 <kenran> no, it was something that appeared in one of my games. I looked in mantis but didn't find anything
15:12:12 <kenran> it might be worth looking there again in case I used bad keywords
15:12:57 -!- amalloy_ is now known as amalloy
15:13:12 <advil> I don't see anything...just asking in case I should close an issue
15:14:23 <Wander> I want to use the sword statue in a vault. It seems to be called "dngn_statue_demonic_bust_3" ingame, but  TILE: G = dngn_statue_demonic_bust_3  gives a "bad spec" error. Anyone know what's up?  (I also tried other things like dngn_statue_sword, etc))
15:14:37 <Wander> you can see the sword mentioned here:   https://github.com/crawl/crawl/blob/master/crawl-ref/source/rltiles/dc-feat.txt#L161
15:15:07 <kenran> Wander: no idea if that helps, but I've had one case where I needed to omit the dngn_ prefix
15:15:27 <Wander> ooh, lemme try
15:20:25 <Wander> "statue_sword", "statue_demonic_bust_3", "demonic_bust_3", "statue_demonic_bust_sword", and "demonic_bust_sword" all error out as well
15:20:30 <Wander> unfortunately
15:22:16 <Wander> Btw, I found out it's called "dngn_statue_demonic_bust_3" by finding one ingame, entering wizmode, and mousing over it to bring up its info
15:22:32 <Wander> weird that it doesn't work
15:31:16 <kenran> Wander: I'm still not sure how to really use the dc-* files to get exactly what I want. I'm interested in your case as well. it might be that it's because the sword doesn't have its own %syn line?
15:33:59 <Wander> yeah, maybe
15:34:24 <Wander> although I don't know what that %syn actually is (not a programmer)
15:34:43 <Wander> but the sugmund statue doesn't work either, for example
15:34:57 <geekosaur> looked to me like it just declared a synonym
15:34:58 <Wander> it's dngn_statue_demonic_bust_4 ingame
15:35:07 <Wander> sigmund*
15:35:16 <kenran> I don't know either, but from my guts (might be totally wrong) I'd say that at the top of this block (statue_dwarf DNGN_GRANITE_STATUE) it starts the possible statues, with different weights
15:35:19 <geekosaur> I am suspecting that those statues are only accessible as part of a group
15:35:29 <kenran> and that the %syn lines define synonyms so you can get those looks directly
15:35:40 <geekosaur> ^ more or less
15:35:47 <Wander> hmm
15:41:52 <advil> I think it's dngn_statue_demonic_bust_3 because it's the third one after the one with an explicit name of dngn_statue_demonic_bust
15:42:40 <kenran> yes I think so too; but if it's possible to use that in a .des, I would like to know how :)
15:42:41 <advil> you should try just adding one of those syn lines after the statue file name, and before the weight line
15:43:12 <advil> it's probably better to name it in case someone later inserts things into dc-feat.txt
15:43:48 <advil> I guess you might need to have the capability of building rltiles though
15:44:17 <kenran> I just tried what advil said (took a while since I had to fully recompile for some reason)
15:44:22 <kenran> it works indeed
15:46:41 <|amethyst> hm
15:46:46 <geekosaur> because it turns into, among other things, #defines for the tile code
15:46:50 <geekosaur> I think
15:48:59 <advil> in master rltiles/tiledef-dngn.h is included in tilesdl.h which is included in tiles.h which is included in AppHdr.h which is included in everything
15:49:17 <advil> coincidentally, this is exactly what my current branch is aiming to refactor :)
15:49:39 <geekosaur> sounds right
15:50:08 <kenran> oh, so tiles.h is even included when doing a console build?
15:50:15 <|amethyst> hmm
15:50:35 <advil> no, but because it's ifedf'd, AppHdr.h changes for everything when you switch to console
15:50:36 <|amethyst> Does adding the %syn to that feature mark the end of DNGN_STATUE_DEMONIC_BUST ?
15:51:16 <|amethyst> as in, will that cause things that ask for TILE: dngn_statue_demonic_bust   to no longer pick the sigmund tile?
15:51:27 <advil> if you're just building console I don't think the rltiles stuff impacts anything, it's switching back and forth that's the worst
15:51:48 <kenran> I actually don't think dngn_statue_demonic_bust did include anything besides the actual demonic bust
15:51:49 <advil> |amethyst: I was wondering that too, but the sword one doesn't make a lot of sense if you're asking for a demon bust in any case
15:51:54 <|amethyst> hm
15:52:04 <kenran> I've placed 6 of them and Ctrl+R a couple of times
15:52:20 <kenran> I think the %syn actually only affects the statue above, even though the ones below are named accordingly
15:52:29 <|amethyst> so I guess the tile name showing up like that is spurious, and really it should be dngn_granite_statue_50 or something
15:52:34 <kenran> yes
15:52:41 <geekosaur> kenran, that was how I understood it, yes
15:53:43 <|amethyst> btw, ancient_hero and ancient_evil are hilarious
15:54:11 <kenran> yes, I've placed ancient hero just now and had a laugh :) have to check out _evil
15:56:26 <geekosaur> *snrk*
15:56:43 <advil> I wonder why it doesn't just auto-name everything in there
15:59:57 <geekosaur> pro tip: separate checkouts for tiles vs. console. I did my Mac stuff that way (don't bother for linux, I don't build local tiles there)
16:00:46 <geekosaur> also I kinda miss shape although it failed often enough that it's probably for the best. (make with caching by configuration)
16:01:32 <advil> I end up needing to switch back and forth for testing too much, for a lot of things
16:02:18 <geekosaur> if I'd had to do that enough, I would have automated the shipping of patches between them and maybe also synchronization of checkouts
16:02:33 <advil> now that I've got ccache really working it's all pretty fast though
16:02:49 <|amethyst> we really need to support a separate build directory
16:03:14 <|amethyst> well, "really need" might be a bit much
16:03:19 <|amethyst> it would sure be nice
16:04:51 <advil> that would be quite nice
16:09:08 <advil> personally I'm a bit afraid of messing with the build infrastructure though
16:10:20 <advil> speaking of which, the msvc and xcode build things are not maintained, right?
16:11:02 <amalloy> i think at least one of them is. check git log though
16:13:34 <|amethyst> xcode isn't
16:14:00 <|amethyst> MSVC is, kind of, but hasn't been used in a year or two, so no idea if it works
16:15:06 <advil> hm, they both seem to get updated a few times a year or so
16:15:31 <geekosaur> every so often someone tries to use one of them and it gets updated as a result
16:18:14 <advil> just wondering whether I should try to fix them for this branch I'm working on, but maybe I won't
16:18:46 <|amethyst> when someone adds or renames a file, I have tried to keep the vcxproj up to date
16:19:05 <|amethyst> not the Xcode project though, because I have no idea how to do that without Xcode
16:19:21 <advil> in principle I could try to get the xcode one to work, but I have no idea how to use xcode
16:19:25 <geekosaur> it's kinda not worth trying because the file format changes at the drop of a hat
16:19:37 <geekosaur> although I guess vcproj does that also
16:20:15 <|amethyst> the problem is that the xcode project uses these numeric references for files
16:20:21 <|amethyst> and I have no idea how to generate them
16:20:35 <|amethyst> the vcxproj can at least be hand-edited
16:21:17 <advil> do you just edit crawl.vcxproj.filters and crawl.vcxproj?
16:22:30 <|amethyst> I haven't bothered with the .filters most of the time, so it's fairly out of date
16:23:21 <|amethyst> hm, it's also missing a few carriage returns
16:25:44 <gammafunk> any good dev memes in here?
16:28:42 <kenran> great job me: enable ccache, make clean, make TILES, make console, then realize that I'm still in an older terminal that hasn't got the updated PATH yet ><
16:30:45 <geekosaur> heh. if I used ccache I'd probably add an indicator to the horror I use for prompt foo
16:32:38 <gammafunk> geekosaur's prompt is an array of fields
16:33:12 <geekosaur> oh, it's much worse than that :) https://github.com/geekosaur/dotty/blob/master/.prompt.pl
16:34:18 <gammafunk> dang
16:38:36 <Cheibriados> 03advil02 07[header_simplification] * 0.21-a0-119-gcbf361d: An attempt to update the MSVC project 10(45 seconds ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/cbf361d3b181
16:47:54 -!- amalloy is now known as amalloy_
17:07:42 <Palyth> question, if i have an update to this vault i submitted, do you want me to just make another patch and attach to the issue?
17:07:46 <Palyth> https://crawl.develz.org/mantis/view.php?id=11061
17:12:45 <gammafunk> Palyth: yeah, just make a new patch and attach the file and make a note
17:13:13 <gammafunk> you can have something like _version2 in the filename, but if you make a note in the bug referring to the new file you uploaded, that will make things clear
17:14:01 <minmay> gammafunk: transporter vault idea: pac-man
17:15:06 <gammafunk> well what are the pellets
17:15:26 <gammafunk> we should add a new feature that you can step on to get a short one-time buff
17:15:49 <|amethyst> quad damage
17:16:54 <|amethyst> make them non-pickup-able like corpses, but still evocable from the floor
17:16:57 <gammafunk> YOU CRUSH BLINKY LIKE A GRAPE!!!!!
17:18:09 <minmay> the pellets can be single gold pieces
17:19:07 <gammafunk> that only works if you worship gozag; the goal is to simulate classic arcade games in crawl
17:19:18 <gammafunk> we need a vault that implementes a working Burger Time, stuff like that
17:19:23 <gammafunk> *implements
17:20:18 <minmay> I mean the regular pellts
17:20:40 <minmay> not sure what to do for the power pellets, quad damage is the most appropriate but also the silliest
17:21:01 <gammafunk> oh that's right, I forgot that the normal pellets do nothing in pacman
17:21:46 <minmay> I'm not getting any real work done today so I guess I'll probably go ahead and implement this joke
17:48:47 <minmay> okay here's this stupid thing https://pastebin.com/KMJPNW9G
18:30:00 -!- amalloy_ is now known as amalloy
19:16:29 -!- tat_ is now known as Guest80495
19:28:01 <Doesnty> nice vault
20:27:59 <johnstein> what does no_pickup mean in vaultese?
20:29:10 <geekosaur> monsters can't pick the item up
20:30:56 <johnstein> ah ok. hmm. what is the general convention on when to use that?
20:31:11 <johnstein> seems like it's not used much in elf:$
20:31:21 <johnstein> which kinda makes sense
20:39:41 <minmay> the convention is "the item's position is aesthetically important"
20:41:02 <minmay> so for example the guarded_unrand vaults have no_pickup on the unrand because they want the mace of variability to be found in the middle of the xom altars and fountains and chaos spawn, not in some orc's inventory
20:43:41 <Yermak> "Your unstable footing causes you to fumble your attack."
20:43:50 <Yermak> Tihs shouldn't be true in tree form
20:44:11 <gammafunk> Your unstable rooting
20:44:36 <gammafunk> get it?
20:44:40 <Yermak> yeah
20:44:44 <gammafunk> heh
20:44:53 <Yermak> btw, it would be good to implement this joke
20:45:29 <Yermak> At least it could make it clear why tree can miss standing on liquified ground.
20:49:32 <gammafunk> ??shallow_water
20:49:32 <Sequell> shallow water[1/2]: Waist-deep pools of water. Moving through this stuff is gonna be really slow! All but Ogre, Troll, Centaur, Gray draconian, Naga, Merfolk, or Octopode players in shallow water will fumble their attack 4-out-of-DEX times, and 20% of the time regardless as well. Monsters will fumble their attack 1/4 of the time.
20:49:49 <gammafunk> probably we could just make the player large in tree form
20:50:18 <gammafunk> some trees can get pretty big, after all
20:52:20 <Floodkiller> I thought trees were already large, which was a factor in a bug that let you wield a GSC as a spriggan
20:53:16 <gammafunk> they may be, it's possible the fumbling is conditioned on specific species
20:53:16 <Yermak> Trees don't miss in water
20:53:43 <Yermak> they miss on liquified ground
20:53:47 <gammafunk> oh
20:54:22 <Yermak> if I get it right there is no conditions when one can't miss standing there
20:55:29 <Cheibriados> Bonus from skill manuals is in  red. 13https://crawl.develz.org/mantis/view.php?id=11139 by MakMorn
22:09:15 <minmay> I'd like to share an "epic bug" from my own game project
23:14:56 <minmay> no takers eh
23:18:44 <johnstein> ??epic bugs
23:18:45 <Sequell> epic bugs[1/28]: if (you.religion = GOD_XOM)
23:20:54 <gammafunk> http://familyfoodandtravel.com/wp-content/uploads/2013/08/epic_activitysheet_insect_experiment-JPG.jpg
23:34:56 <johnstein> nice. got levdes syntax working on msys2. it's such an odd system, that msys2
23:44:09 <minmay> alright here goes
23:44:44 <minmay> in luajit, a lot of Lua standard library functions are builtin to the VM to be as fast as possible, instead of being real Lua functions
23:45:07 <minmay> one of these is math.floor(x), which just gives you x rounded down
23:45:44 <minmay> this is true even if you don't actually pass an x, in which case "x" is a might-as-well-be-random chunk of memory
23:46:27 <minmay> result: the player's attacks were hitting for 8.6548638537281*10^189 damage
23:46:52 <minmay> (and similarly large numbers)
23:47:30 <geekosaur> didn't we have something like that at one point? not for player attacks, but I think plusses on something because of buggy save compat or w/e
23:52:21 <nikheizen> i remember the +100 rings of foo
23:54:31 <minmay> sadly those were just bytes
23:55:03 <minmay> if you can reasonably express it without using an exponent, can it even count as a bug?