00:01:40 <Eksell> Unstable branch on CRAWL.XTAHUA.COM updated to: 0.24-a0-22-g2a401c0 (34)
00:03:34 <Eksell> Stable (0.23) branch on CRAWL.XTAHUA.COM updated to: 0.23-b1-6-ga921848
00:33:13 <Cheibriados> 03advil02 07* 0.24-a0-23-g193e081: Randomness fix for swamp / pools 10(12 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/193e08148b4d
00:33:13 <Cheibriados> 03advil02 07* 0.24-a0-24-g12d111d: Fix some typos and lua mistakes 10(26 seconds ago, 1 file, 3+ 3-) 13https://github.com/crawl/crawl/commit/12d111dd0df5
00:37:36 <Cheibriados> 03advil02 07* 0.24-a0-25-g8489be9: Add a new lua test for seed stability 10(2 minutes ago, 1 file, 126+ 0-) 13https://github.com/crawl/crawl/commit/8489be958e62
00:41:25 <advil> excited to see what travis makes of that test
00:42:11 <advil> though maybe I'll have to find out in the morning
00:45:52 <ebering> advil: re _place_specific_trap: if a vault places on a branch end a shaft is meaningless;, but the function should return after the error and not try to place the trap
00:46:13 <ebering> it's an mprf error and not a die because that's how I found it
00:46:35 <ebering> retconned reason: vaults that have depths that pre-date branch shortenings might call for shafts that can no longer be there
00:47:02 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.24-a0-25-g8489be958e (34)
01:28:03 <Gretell> Unstable branch on crawl.develz.org updated to: 0.24-a0-25-g8489be9 (34)
01:57:24 <travis-ci> The build was broken. (master - 8489be9 #11020 : advil): https://travis-ci.org/crawl/crawl/builds/487280219
01:58:47 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.24-a0-25-g8489be9
02:59:47 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.24-a0-25-g8489be9
03:31:07 <Kelbell> Stable (0.23) branch on crawl.kelbi.org updated to: 0.23-b1-6-ga921848b48
08:35:36 <advil> oops
08:35:45 <advil> guess I haven't figured out the swamp randomization issues after all
09:48:11 <NP7> !lg *
09:48:12 <Sequell> 10048665. Xqipyl the Devastator (L15 FeVM of Uskayaw), succumbed to a ghost moth's poison on Spider:4 (grunt_spider_rune_island) on 2019-02-01 05:36:25, with 89082 points after 84156 turns and 5:04:43.
09:48:42 <NP7> According to Sequell, the last game completed online was completed nearly 10 hours ago...
09:48:57 <NP7> Looks like something with the online database might have gone wrong?
09:52:58 <Pinkbeast> Am I missing a wizmode command to dig the undiggable, please?
09:53:14 <NP7> just cast shatter
09:53:42 <|amethyst> Pinkbeast: if you use x<navigate>m to move yourself on top of a wall, it will be replaced with floor
09:54:01 <NP7> oh interesting
09:54:11 <NP7> that's really not obvious
09:55:02 <|amethyst> I discovered it kind of by accident
09:55:02 <Pinkbeast> THanks both. I tried to & B onto it to & ( it, but that works.
09:55:31 <Pinkbeast> Shatter might be a bit undirected in this case. :-)
09:56:08 <|amethyst> when I was going to use &( to put a floor there
09:56:08 <|amethyst> so used xm to move there... and now it was already a floor
09:57:12 <Cheibriados> 03advil02 07* 0.24-a0-26-g4429137: Fix potential iteration order issue for autofight 10(4 minutes ago, 1 file, 10+ 4-) 13https://github.com/crawl/crawl/commit/4429137cf8f2
09:57:28 <advil> I think there is a way of simply moving onto a wall position, isn't there?
09:58:36 <advil> yeah, XT moves without placing a floor
09:59:06 <advil> I honestly didn't know about wizmode xm
10:00:23 <advil> so that gives at least four special wizmode move things: &b (los, doesn't move through terrain), &B (map, doesn't move through terrain), xm (x mode, moves through tererain and places floor), XM (map mode, moves through terrain, no placement)
10:00:43 <advil> oh, I guess I know xm worked on monsters
10:01:02 <advil> so that must be why it places a floor, since monster in wall is more of an error than wizmode player in wall
10:01:39 <advil> er, XT, not XM
10:02:20 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.24-a0-26-g4429137cf8 (34)
10:02:25 <Pinkbeast> I've been using xm for ages to wall monsters up with casks of Amontillado and whatnot but I didn't know it worked on me until now
10:02:43 <advil> yeah, me too
10:02:43 <advil> well not the amontillado part
10:03:14 <|amethyst> Yes, Sigmund, for the love of God!
10:04:30 <|amethyst> (One of my friends in high school oh so many years ago had "Nemo me impune lacessit" as his senior yearbook quote)
10:05:19 <Pinkbeast> My parents didn't really stop me reading anything I could get down so I read Poe at a very tender age and it made a lasting impression
10:05:22 <|amethyst> (and "Zork lord" as his nickname in the yearbook)
10:14:30 <Pinkbeast> While I'm at it, does anyone know _why_ Lamia was removed? The commit message ("Remove Lamia") is a masterpiece of succinctness
10:24:39 <advil> it seems a bit derivative to me
10:24:53 <advil> but no, I don't know why
11:13:30 <travis-ci> The build is still failing. (master - 4429137 #11021 : advil): https://travis-ci.org/crawl/crawl/builds/487477437
11:19:09 <advil> yeah yeah
11:19:41 <advil> does anyone have a sense for why tags on map_def are still stored as just a raw string? does order ever matter for them?
11:22:21 <Lantell> Stable (0.23) branch on underhound.eu updated to: 0.23-b1-6-ga921848b48
11:22:43 <advil> or counting for that matter
11:23:01 <advil> I've just been assuming they are really a set, but I haven't interacted with this code enough to be sure
11:24:36 <advil> I already fixed one bug like this, so I'm not sure why it took me a long time to clock this, but a large set of stability problems I've been finding today are because the vault cache keeps getting rewritten with the list of tags increasing, there's never a duplicate tag clearing
11:25:11 <advil> any call to x.tags("blah") in lua adds more instances of "blah", perhaps each time crawl runs (but at least the first two times)
11:25:17 <advil> after a clean -builddb that is
11:26:18 <advil> so the obvious thing would be treat tags as a set, at least in terms of cleaning up duplicates, but this is so obvious I'm not sure why it wasn't already done
11:27:48 -!- amalloy is now known as amalloy_
11:28:52 <advil> also I would probably sort them
11:29:33 <advil> but again, so obvious I'm worried there's something I don't know about how tags work
11:32:26 <geekosaur> there's some horror about their having multiple uses. go look at the mess around "pan" being both a location and a unique
11:32:58 <geekosaur> which abused the tags *and* revealed some things code was assuming / requiring
11:32:58 <geekosaur> iirc
11:35:22 <gammafunk> advil: order definitely doesn't ever matter for tags, or shouldn't, at least, and yeah a set-based approach would be fine
11:35:44 <gammafunk> I'm not sure what they'd be stored as if not a string though
11:36:04 <advil> stored as in written to a file, yes, a string makes the most sense
11:36:05 <gammafunk> they don't correspond to any enum and otherwise have no natural mapping
11:36:29 <advil> but a set<string> would use lexicographic order
11:36:54 <advil> in  map_def the tags field is just a string, and when you call get_tags() it does a split by " " on the spot
11:37:08 <advil> half the internal code for that class does the split itself
11:37:35 <gammafunk> probably could be implemented in a better way then, yeah; certainly nothing should care about the particular ordering of tags and if it does, that's a bug
11:37:48 <gammafunk> those are simply used with "select vaults by tag"
11:38:16 <gammafunk> and all such callers just select one at random from the set of returned vaults
11:38:16 <gammafunk> at least afaik
11:38:38 <advil> hm, this does mean that currently, reordering things like temple_altars_12 and temple_altars_18 will have an effect on seed, so that's an argument for a stable sort
11:39:34 <advil> shouldn't really depend on how the vault writer does it
11:40:02 <gammafunk> yeah
11:40:39 <advil> I have to say I've really gotten a lot of bug-fixing mileage out of creating a whole new class of bugs
11:42:27 <gammafunk> you overtook ebering in commits for 0.23 I think
11:42:31 <gammafunk> I think I'll be the next to fall
11:43:05 <gammafunk> time to do a massive round of DES bug fixes before release
11:44:36 <gammafunk> 226 gammafunk
11:44:37 <gammafunk> 212 advil
11:44:37 <gammafunk> 207 ebering
11:44:37 <gammafunk> 74 Aidan Holm
11:44:49 <gammafunk> hrm, guess we never merged aidanh's branch did we
11:55:03 <ebering> which one?
12:01:58 <gammafunk> the new class selection one?
12:02:07 <gammafunk> or did we
12:02:57 <ebering> oh, no
12:03:04 <ebering> I did merge his pr-599 branch
12:03:16 <gammafunk> yeah I guess the one I'm referring to has no PR
12:03:25 <gammafunk> I think he did say a while back that it's probably ready to merge
12:03:56 <gammafunk> but it probably would be good to at least hear from him about any outstanding issues before attempting that, I'm guessing
12:04:00 <ebering> yep
12:06:40 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.24-a0-26-g4429137 (34)
12:22:39 -!- amalloy_ is now known as amalloy
12:46:22 <ebering> %git f54098542e9
12:46:22 <Cheibriados> 07MarvinPA02 * 0.17-a0-1520-gf540985: Let healing from fake potions work on Vine Stalkers 10(3 years, 7 months ago, 5 files, 52+ 42-) 13https://github.com/crawl/crawl/commit/f54098542e98
12:47:18 <ebering> MarvinPA: with the only source of fake potions left being gozag PP, and PP not giving berserk and haste to species that can't, I'm thinking of taking away VS Heal Wounds from potion petition
13:31:25 <advil> btw do we have a release date?
13:31:42 <advil> I guess we must if there's a t date
13:31:46 <ebering> I was hoping for feb 6
13:31:59 <ebering> so that cwz has time to get the released version and not run the t on the beta
13:32:26 <amalloy> ebering: powered by pain also gives fake potion effects, although never potions you couldn't drink normally
13:33:15 <amalloy> also some other weird things. BEAM_HEALING apparently gives !hw effects. what the heck is BEAM_HEALING
13:33:30 <amalloy> i guess it comes from those chaos demonspawn?
13:34:22 <ebering> amalloy: hm. I guess s/fake potions/fake healing potions/ but I'm still wrong
13:34:37 <amalloy> ah, a chaos weapon hitting you
13:35:00 <ebering> amalloy: yeah, I think it's ok for VS to get that one
13:35:18 <amalloy> also ely's self-heal appears to use a potionlike_effect
13:35:49 <amalloy> indirectly from _setup_healing_beam to BEAM_HEALING
13:36:00 <gammafunk> ebering: hong sets up the cwz when he wishes, expect it to have the beta
13:36:01 <amalloy> oh, maybe not
13:36:11 <amalloy> _setup_minor_healing might be for allied monsters with a heal spell
13:36:27 <amalloy> yeah
13:36:38 <amalloy> anyway that also uses potionlike_effect
13:36:39 <gammafunk> if you want him to run a thing you really need to actually communicate with him, but ime he just does what he does when he has time
13:36:58 <ebering> hm. dang
13:37:01 <gammafunk> definitely won't work to try timing something and hoping he just reacts to that, since he's pretty busy
13:37:46 <gammafunk> not to change anything about release or w/e, but yeah that's been my experience over like the last 5 tournaments or so
13:38:04 <gammafunk> !lg * cwz 1 x=cv
13:38:09 <Sequell> 1/1301676. [cv=0.16-a] tlatlagkaus the Insei (L1 OpTm), quit the game on D:1 (dpeg_arrival_houses_and_road) on 2015-02-10 03:48:05, with 0 points after 0 turns and 0:00:15.
13:38:21 <gammafunk> ok good, 5 wasn't way off the mark
13:39:01 <gammafunk> I have already contacted him with details about the branch, so he has the basic info he needs at least
13:39:39 <gammafunk> !lg * cwz tall s=cv
13:39:50 <Sequell> 39400 games for * (cwz tall): 9783x 0.22-a, 7662x 0.21-a, 7492x 0.18, 6520x 0.19, 3834x 0.17-a, 3268x 0.20, 841x 0.19-a
13:40:19 <gammafunk> looks like he god stable 3 out of 7 times
13:40:19 <gammafunk> s/god/got/
13:40:20 <gammafunk> er
13:40:29 <gammafunk> well, yeah, that's stable
13:40:43 <gammafunk> but how many times did he get the actual release, I wonder
13:40:48 <gammafunk> !lg * cwz tall s=vlong
13:41:00 <Sequell> 39400 games for * (cwz tall): 9783x 0.22-b1-59-g0aedee7, 7662x 0.21-b1-23-g1fb0a74, 7492x 0.18.1-1-g2f07953, 6520x 0.19.0-2-gae57169, 3834x 0.17-b1-35-g4ccdf60, 3268x 0.20.0-1-g1cb1d84, 841x 0.19-b1-3-g7dea586
13:41:16 <gammafunk> 2 out of 7
13:43:34 <gammafunk> oh I see, 0.19 he got a mix of both, nice
13:43:51 <gammafunk> I forget whey he did an update after tournament start that time
13:43:54 <gammafunk> wasn't meleebug
13:48:34 <gammafunk> https://crawl.develz.org/tournament/0.23/overview.html
13:48:51 <gammafunk> been running the script for the last few ours with 0.23 from all servers that currently have that
13:48:55 <gammafunk> so far no problems
13:49:53 <ebering> looks good
13:49:57 <gammafunk> if there is an extended outage of several hours where the script is not running, it could take an hour for it to process the backlog, depending on the length of the outage
13:50:22 <gammafunk> it's about 15 minutes to process 2 hours of tournament data taken from the first two hours (which are the heaviest in terms of traffic)
13:50:27 <gammafunk> using 0.22 as a test
13:50:55 <gammafunk> so normally it's only processing 7 minute batches, and it might take a minute to process each batch (probably less on average)
13:51:15 <gammafunk> hence the only major problems are that rebuilding the db from scratch during the tournament isn't really feasible to do on CDO
13:51:27 <gammafunk> we had done that in past tournaments to fix case issues, although I didn't last time
13:51:55 <gammafunk> even on cszo that takes some time, but it's significantly faster compared to cdo to the point where you could do it
13:52:36 <gammafunk> if for some reason we need to do a rebuild, perhaps we could use cszo to do the rebuild and copy over the sql dump/html files and resume from there
13:52:44 <gammafunk> but my plan is to just not rebuild during the tournament
13:53:26 <ebering> that seems like a reasonable contingency plan
13:53:45 <gammafunk> actually dobrazupa was down yesterday...
13:54:26 <gammafunk> yeah it's still down; if it comes to that we could look into getting help from alexjurkiewicz but we'll just hope to not have to rebuild
13:54:41 <ebering> If we need to rebuild I can do it on waterloo csclub
13:54:49 <gammafunk> ok
13:54:51 <ebering> that'll be zippy
13:55:26 <gammafunk> I'm not actually sure about said transfer of db; hopefully just a zipped sql dump of the db and tar of html files would be ok to work from
13:56:14 <gammafunk> don't actually need the html files I think, since those can just get rebuilt by cdo; we can just cross that bridge if we get to it
13:56:25 <ebering> yes
13:56:46 <gammafunk> ebering: you might want to ask Napkin about adding your ssh key to cdo so you can log in there
13:57:13 <gammafunk> I know advil has a login, but it would be good to have one other person available to restart the tournament script if there's an issue or just to generally have access
13:57:22 <gammafunk> I'm not sure if advil would want to mess with tournament stuff
13:57:57 <gammafunk> if you don't either, that's fine, but it's nice if someone else has access to the tournament and website stuff who's active
13:58:07 <advil> I don't mind doing it
13:58:14 <advil> I'll basically be around
13:58:37 <gammafunk> ok
13:58:42 <advil> !time
13:58:43 <Sequell> Time: Feb 01, 2019, 06:58:42 PM, UTC. The 2019 0.23 tournament starts in 7 days, 1 hour, 1 minute and 17 seconds.
13:59:00 <advil> heh my giant grant submission is also due on Feb 6
13:59:12 <advil> (that I have been working on rng stuff instead of)
13:59:19 <gammafunk> how much funding will be crawl-related?
14:00:40 <advil> if someone wanted to build a deep learning crawl bot that was taking the graphics as input, we could conceivably do something
14:01:04 <ebering> alpha-qw-zero
14:01:06 <gammafunk> alpha crawl
14:01:32 <advil> exactly
14:01:32 <advil> would have to have a computer vision component though
14:02:45 <advil> it would be nice to someday have scripts that didn't require reprocessing all the data from the beginning to get back to functional
14:03:55 <gammafunk> well, if you wipe the db, isn't that requiring processing all the data from the beginning?
14:04:12 <gammafunk> actually, we only did that in the past to fix this case issue
14:04:28 <gammafunk> which is a pretty minor cosmetic one; that could be fixed presumably
14:04:57 <gammafunk> only other thing I know of where we'd need to rebuild is if we blacklisted someone and then wanted to unblacklist them
14:05:22 <gammafunk> I only added player blacklist was only added last tournament, and I doubt that issue would come up
14:06:01 <gammafunk> maybe in general the scripts just need a faster implementation, better table design etc
14:11:55 <Cheibriados> 03ebering02 07* 0.24-a0-27-gca774cb: Give a more helpful message for 'G>' at branch end (3734) 10(3 minutes ago, 1 file, 12+ 4-) 13https://github.com/crawl/crawl/commit/ca774cb5d5d2
14:17:27 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.24-a0-27-gca774cb5d5 (34)
14:18:01 <gammafunk> 27 commits in 0.24, time to lock the branch
15:19:44 <travis-ci> The build is still failing. (master - ca774cb #11022 : Edgar A. Bering IV): https://travis-ci.org/crawl/crawl/builds/487601392
15:20:49 <gammafunk> hrm, is this error related to the seed tests?
15:21:24 <ebering> Yes
15:21:49 <gammafunk> ah, yeah I see
15:33:55 <twelwe> ah, yes
15:35:18 <advil> yeah, sorry
15:35:21 <ebering> should fedhas protect plants from LOS attack spells
15:36:19 <NP7> if the alternative is having them be hurt by LOS attack spells and causing penance, then yes
15:36:19 <advil> I finally replicated the specific failure there (it only happens on the first run after a clean -builddb, so that's why it took pushing the test to find it)
15:37:04 <NP7> I would say
15:37:19 <ebering> that's what I'm thinking
15:54:42 <advil> literally every time I want to know how to do something basic with strings in c++, including something I did like yesterday: I google it, it's not part of the c++ lib, and then find that the first answer is "use boost", and then I realize it's probably also implemented somewhere in stringutil.{cc,h}
15:56:09 <floraline> i have a user reporting that their changes to macros are being lost when they start up the game
15:56:20 <floraline> they have this in their rcfile: macros += M 1 ===CRAWL_COSPLAY_CONDUCT
15:56:25 <floraline> is this expected behavior?
15:57:47 <floraline> i reproduced this and the macro file is reset when the game is started, my changes are lost and only that 1 macro is still there
15:58:59 <NP7> so saved macros from previous games are being cleared when there's a macro set in the rcfile?
15:59:36 <NP7> s/games/sessions
15:59:39 <floraline> right
15:59:51 <NP7> that does not sound like good behaviour
16:00:02 <floraline> i can set macros during a session and save my macro file, but saving and reloading my game results in a new macro file with only the one in the rcfile
16:00:13 <NP7> right yea
16:00:17 * geekosaur wonders if some of the recent game state reset stuff is a bit overexuberant
16:00:31 <floraline> i don't know if this is intended behavior or if my configuration is bad
16:00:48 <NP7> fair chance it's neither of those things
16:00:54 <advil> the game state reset changes are mainly for tests
16:01:18 <advil> i.e. things that are calling debug.flush_map_memory() in lua
16:02:30 <advil> the only one that applies to regular games (and this would be offline only) is:
16:02:33 <advil> %git 0b81bb8f681ffe2
16:02:33 <Cheibriados> 07advil02 * 0.24-a0-19-g0b81bb8: Reset temple vault tags when doing temple setup 10(23 hours ago, 2 files, 18+ 1-) 13https://github.com/crawl/crawl/commit/0b81bb8f681f
16:02:34 <NP7> can't reproduce this in a 300-commit-old offline install
16:02:57 <gammafunk> floraline: yeah that seems like expected behaviour
16:03:35 <gammafunk> if they're setting their macro in their RC, that is going to override what they do in one session
16:03:35 <switch__> advil: re boost, this has happened to me at least 5 times in the past two weeks
16:03:35 <gammafunk> the solution seems to be removing those lines from their RC
16:04:47 <floraline> ok. can they still set that kind of macro in-game? it looks like it's calling a lua function?
16:04:52 <gammafunk> ebering: I'm not sure I understand re: los attack spells
16:05:07 <gammafunk> sure, they can set macros in-game
16:05:22 <gammafunk> but if their RC explicitely sets that macro, it's going to override it the next time the RC is loaded
16:06:15 <NP7> ah yea I guess if they're redefining a macro on the same letter as one set directly in the rcfile, that makes sense
16:06:39 <floraline> it's not just macro on that key, it's all macros that are being lost
16:07:33 <ebering> gammafunk: fedhas' description says "fire through allied plants" but in practice that means you can also fire at them, with e.g. fireball, and nothing happens to them
16:07:35 <gammafunk> yeah, the macro in the RC would ideally only affect that one macro
16:08:17 <ebering> so you might expect that LOS attack spells would do the same thing
16:08:23 <ebering> instead you get a pennance prompt
16:09:24 <ebering> imo fedhas should protect your plants from these effects too
16:09:26 <gammafunk> yeah, that's a bit tricky since the notion of "fire through" isn't really well-defined from a player perspective
16:09:59 <gammafunk> if you make it so fedhas systematically prevents things you do from damaging your plants, I suppose that would help
16:10:37 <gammafunk> however it's also arguably fine to have targetted spells and ranged attack be the only thing fedhas supports
16:10:56 <gammafunk> so the only thing I'd worry about in regards to changing this
16:11:07 <gammafunk> is whether this opens up a big can of worms for trying to add various other exceptions
16:11:24 <gammafunk> what about cloud spells and such?
16:12:04 <ebering> fedhas already protects your plants from your clouds
16:12:31 <gammafunk> then it sounds like extending protection is ok as long as there's nothing really nasty we'd have to add protection for
16:14:22 <ebering> the mace of Variability  already respects plants
16:14:55 <ebering> as does chain lightning
16:14:58 <gammafunk> things like damnation and devastator
16:15:03 <gammafunk> I imagine those are fine already
16:15:06 <ebering> yep
16:15:09 <gammafunk> glaciate?
16:15:11 <ebering> anything that uses a beam is already ok
16:15:28 <gammafunk> maybe chain lit; same chain code I think, so probably fine if variability is
16:15:39 <gammafunk> static discharge maybe? that also uses a beam internally iirc
16:15:41 <gammafunk> so probably fine
16:19:31 <ebering> SD is not currently fine
16:19:48 <gammafunk> oh, interesting
16:19:58 <gammafunk> I had a todo of converting SD over to the chain code
16:20:08 <gammafunk> its recursive behaviour is kind of complicated
16:20:43 <gammafunk> just sort of adapting the chain code to only arc range 1 and to have similar behaviour in terms of average number of arcs and whatnot
16:21:33 <gammafunk> however it might be easy to fix SD to respect fedhas passive without doing that right away
16:21:51 <gammafunk> that meaning converting it to the chain spell code
16:22:48 <ebering> yeah
16:22:49 <ebering> hm
16:30:08 <NP7> looks like this thing about all macros being cleared when there's any macro option in the rcfile may have existed for a long time...at least since 0.20
17:09:53 <floraline> sounds like a bug to me!
18:04:32 <advil> hm, turns out tag order matters for validating temple maps
18:05:41 <advil> gonna declare that a bug
18:08:39 <gammafunk> fun
18:08:46 <gammafunk> how does it matter?
18:09:09 <advil> for temple_overflow_generic vaults, the first tag is supposed to match the number of altars present when validating the map
18:09:32 <advil> so, "temple_overflow_generic_8" if before your subst stuff you have 8 altars
18:09:51 <advil> I'm just going to change it so it registers as ok if it finds a generic tag for the number of altars present
18:16:46 <gammafunk> ah right
18:17:15 <gammafunk> that code actually parses the number from the tag name, doesn't it
18:17:30 <advil> yeah
18:22:35 <Lantell> Unstable branch on underhound.eu updated to: 0.24-a0-27-gca774cb5d5 (34)
18:23:18 <advil> some of this layout code is such mad scientist stuff
18:23:38 <gammafunk> hehe
18:23:42 <gammafunk> yes, very much so
18:23:49 <gammafunk> I have not read it too carefully
18:24:14 <gammafunk> vaults layout code is pretty complicated and feels overly so
18:24:27 <gammafunk> but I think a lot of layout code is pretty complicated
19:08:38 <gammafunk> ebering: I realize we can make net traps part of gauntlets so that they're not 100% dispersal traps (for arenas that successfully roll to place traps at all)
19:08:56 <gammafunk> any weighting you'd like to see on that between the two types?
19:21:29 <ebering> hm
19:22:39 <ebering> perhaps if an arena rolls traps it should only place one or the other
19:22:51 <ebering> since we don't want a player getting dispersed out of a net in close quarters
19:42:31 <gammafunk> I guess that's more likely to happen with the only two trap types being those
19:42:56 <gammafunk> not sure it's actually a huge concern, but it doesn't matter to me too much
19:43:52 <gammafunk> dispersal / zot / tele / net / alarm available, but tele/alarm are useless and zot too nasty
19:44:36 <ebering> yeah; it might not be too likely
19:44:43 <ebering> half and half would probably work
21:19:28 -!- CaptainFruitcake is now known as CanOfWorms
21:25:57 -!- amalloy is now known as amalloy_
21:52:35 <Cheibriados> 03ebering02 07* 0.24-a0-28-g788b6e3: Extend Fedhas' protection to non-beam spells (11310) 10(51 minutes ago, 2 files, 54+ 9-) 13https://github.com/crawl/crawl/commit/788b6e38277a
21:57:20 <ebering> !bug 11717
21:57:21 <Sequell> https://crawl.develz.org/mantis/view.php?id=11717
22:02:15 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.24-a0-28-g788b6e3827 (34)
22:20:48 <alexjurkiewicz> !source place_monster
22:20:48 <Sequell> 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/mon-place.cc#L617
22:35:53 <alexjurkiewicz> for (monster_iterator mi; mi; ++mi)
22:35:55 <alexjurkiewicz> if (mons_zombifiable(**mi->base_monster))
22:36:03 <alexjurkiewicz> why doesn't this code work? I get the error "indirection requires pointer operand"
22:36:41 <alexjurkiewicz> I copied the code from _beogh_extra_description
22:37:49 -!- amalloy_ is now known as amalloy
22:39:19 <geekosaur> because in that case they want the whole monster, so **. here, you're applying -> to pointer-to-monster, so *mi->...
22:42:58 <alexjurkiewicz> I get the same error: dungeon.cc:2450:38: error: indirection requires pointer operand ('int' invalid)
22:43:15 <alexjurkiewicz> (pointing at the reaining '*')
22:44:47 <gammafunk> isn't mi just a monster pointer
22:45:07 <gammafunk> oh I guess you do have to do *mi to defer to get to the pointer itself
22:52:26 <Cheibriados> 03ebering02 07* 0.24-a0-29-g43a6aff: Disable mores during targeting (11717) 10(3 minutes ago, 1 file, 1+ 0-) 13https://github.com/crawl/crawl/commit/43a6aff242bf
22:52:41 <geekosaur> this sounds like it's doing the -> before the *
22:52:49 <geekosaur> (*mi)->...
22:52:50 <geekosaur> ?
22:59:55 <alexjurkiewicz> geekosaur: 10 points to you
23:01:33 * geekosaur has clearly spent too long away from C, had forgotten that one
23:02:10 <travis-ci> The build is still failing. (master - 788b6e3 #11023 : Edgar A. Bering IV): https://travis-ci.org/crawl/crawl/builds/487743058
23:02:22 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.24-a0-29-g43a6aff242 (34)
23:04:30 <gammafunk> it's still failing, ebering
23:04:32 <gammafunk> *still*!
23:18:03 <twelwe> points for trying tho
23:25:51 <ebering> rip