00:03:02 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.28-a0-410-gbef64dd269 (34)
01:34:27 <Gretell> Unstable branch on crawl.develz.org updated to: 0.28-a0-410-gbef64dd269 (34)
01:55:28 <Gretell> Windows builds of master branch on crawl.develz.org updated to: 0.28-a0-410-gbef64dd269
02:23:14 <Cbrotell> Unstable branch on cbro.berotato.org updated to: 0.28-a0-410-gbef64dd269 (34)
02:53:37 <Gretell> Monster database of master branch on crawl.develz.org updated to: 0.28-a0-410-gbef64dd269
03:50:26 <Kelbell> Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-4369-gf34790d913
05:38:34 <Yermak> !hs * paca
05:38:35 <Sequell> 15. jt the Grim Reaper (L27 PaCA of Yredelemnul), escaped with the Orb and 4 runes on 2021-09-05 21:59:47, with 2804292 points after 53002 turns and 4:53:23.
05:38:47 <Yermak> sorry, wrong channel
06:03:21 <Cheibriados> Scorch doesn't work with autofight option. 13https://crawl.develz.org/mantis/view.php?id=12646 by Yermak
06:22:24 <Lantell> Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-4369-gf34790d913
12:39:51 <Lantell> RinnBerevan (L3 MfGl) ASSERT(!spells.empty()) in 'randbook.cc' at line 493 failed. (D:2)
12:42:28 <Lantell> RinnBerevan (L3 MfGl) ASSERT(!spells.empty()) in 'randbook.cc' at line 493 failed. (D:2)
13:07:21 <Henzell> Unstable branch on crawl.akrasiac.org updated to: 0.28-a0-410-gbef64dd (34)
13:18:13 <advil> !crashlog
13:18:14 <Sequell> 24009. RinnBerevan, XL3 MfGl, T:1150 (milestone): https://underhound.eu/crawl/morgue/RinnBerevan/crash-RinnBerevan-20210906-164226.txt
13:25:23 <advil> I take it there are no longer any level 1 hexes
13:26:00 <advil> that crash is from a vault that places `randbook owner:Plato numspells:1 slevels:1`, which leads to a crash if the book rolls hexes
13:27:09 <Discord|> <10P​leasingFungus> nice
13:27:18 <advil> I do see other books that are likely to crash because of that
13:27:59 <advil> not a huge #
13:28:01 <Discord|> <10P​leasingFungus> corona secretly the pillar holding crawl up?
13:28:19 <advil> apparently
13:30:02 <Discord|> <09h​ellmonk> time for 25 power cap slow at level 1
13:31:06 <advil> hm, there's some code that should head this off it looks like
13:38:30 <advil> I guess that's just a completely different code path for randbooks
13:38:42 <Discord|> <10P​leasingFungus> there is no level 1 tloc, is there?
13:38:50 <Discord|> <10P​leasingFungus> o
13:38:52 <Discord|> <10P​leasingFungus> apport
13:38:54 <Discord|> <10P​leasingFungus> duh
13:48:24 <advil> sort of tricky to see a general solution to this
13:48:42 <advil> given the way randbook vault spec is implemented
13:53:38 <perryprog> Probably could have a test suite that attempts to generate every vault just to make sure any blatant runtime errors can get caught.
13:54:27 <advil> there is such a test suite, but it takes hours to run
13:54:32 <perryprog> Oh
13:54:35 <perryprog> Ouch
13:55:00 <advil> also, vaults are highly randomized, and this crash isn't guaranteed with this vault
13:55:03 <perryprog> Aaa
13:55:35 <advil> the placement.lua test script does allow specifying the number of placements but you'd need a lot to get this particular crash; it's easy to replicate with the seed in the crash log though
13:55:41 <perryprog> Hrm, in that case could maybe try to generate the individual features that a specific vault has?
13:55:56 <perryprog> E.g., try to generate every item spec a vault has, make sure it doesn't explode, move on
13:56:12 <advil> you'd need to do that many times
13:56:41 <advil> this randbook spec chooses spell schools randomly, and I believe to trigger this crash it needs to roll hexes for both (books have up to 2 disciplines)
13:56:47 <perryprog> oo
14:05:43 <Cheibriados> 03advil02 07* 0.28-a0-411-gffd9c288d6: fix: reroll spell schools for crashing hex books 10(3 minutes ago, 1 file, 23+ 3-) 13https://github.com/crawl/crawl/commit/ffd9c288d6fe
14:14:26 <advil> anyways, tests are good but the degree of randomization and procedural variation makes it challenging to fully rely on them in dcss
14:14:53 <spicyCebolla> whhat's aaa level 1 hex look like? 15 auuuts of confusion? unable tto attack or cast spells for 20 auts?
14:15:18 <advil> it was corona, in case that's not clear
14:15:34 <advil> a test that ran through every item spec sounds like it could be interesting but I still expect it would be slow enough that it couldn't be run in CI, for example
14:15:50 <spicyCebolla> oh i assumed it was crashing cuz  corona got removed
14:15:59 <advil> yes, that's right
14:16:27 <perryprog> advil, the item spec parsing code is decently fast. I have it running at about 1500 calls per second (I'm messing around with it for fun at the moment)
14:16:43 <advil> I guess you're asking what a viable lvl 1 hexes spell is, not sure
14:16:52 <perryprog> Doing every permutation of an item spec is probably prohibitively expensive though
14:18:43 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.28-a0-411-gffd9c288d6 (34)
14:49:31 <Discord|> <10P​leasingFungus> i don’t mind hellmonk’s slow to lvl 1 suggestion
15:16:59 <Cheibriados> Passed after stabbing a monster after using passwall 13https://crawl.develz.org/mantis/view.php?id=12647 by Lici_the_Crawler
16:18:52 <advil> !lg * D:1 nrune=0 crawl max=xl -log
16:18:58 <Sequell> 3028486. arbitraryname, XL27 DsGl, T:0: https://crawl.kelbi.org/crawl/morgue/arbitraryname/morgue-arbitraryname-20201114-023934.txt
16:19:03 <advil> anyone know what's going on with that game?
16:19:07 <advil> (found by perryprog)
16:19:33 <advil> I looked at the ttyrec and it wasn't very illuminating, looks scripted maybe
16:42:40 <Cheibriados> 03PleasingFungus02 07* 0.28-a0-412-g9c2f8b1708: Reduce Searing Ray power scaling 10(28 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/9c2f8b1708eb
16:42:40 <Cheibriados> 03PleasingFungus02 07* 0.28-a0-413-gde365fb1df: Cut Storm Form power scaling 20% 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/de365fb1df8b
16:47:28 -!- allbery_b is now known as geekosaur
16:48:10 <Kelbell> Unstable branch on crawl.kelbi.org updated to: 0.28-a0-413-gde365fb1df (34)
18:23:36 <Lantell> Unstable branch on underhound.eu updated to: 0.28-a0-413-gde365fb1df (34)
19:52:33 <Discord|> <06a​dvil> !lg * gid=arbitraryname:cko:20201014023856S -log
19:52:34 <Discord|> <04C​erebot> 1. arbitraryname, XL27 DsGl, T:0: https://crawl.kelbi.org/crawl/morgue/arbitraryname/morgue-arbitraryname-20201114-023934.txt
19:53:33 <Discord|> <06a​dvil> just want to flag this again in case anybody has any ideas -- taking it at face value it appears there must be a way to run dlua outside of wizmode in an online game, what happens there looks exactly like the results of you.give_exp
19:56:36 <Discord|> <10P​leasingFungus> alternative is that they entered wizmode without it being properly noted/tracked?
19:57:02 <Discord|> <06a​dvil> I think the morgue would show that?
19:57:21 <Discord|> <06a​dvil> should be visible in the ttyrec too
20:11:36 <Discord|> <09g​ammafunk> woah
20:15:40 <Discord|> <09g​ammafunk> yeah ttyrec seems to imply it happened right when ready() first proced or something like that
20:15:48 <Discord|> <06a​dvil> I've checked the most obvious things and they seem to be working as expected
20:15:51 <Discord|> <06a​dvil> yeah, that's my thought too
20:16:05 <Discord|> <06a​dvil> I can easily replicate the behavior by putting you.gain_exp with a large number in a wizmode ready
20:16:28 <Discord|> <09g​ammafunk> current RC is custom but contains no obvious fancy stuff, just settings
20:17:24 <Discord|> <09g​ammafunk> oh hey and some of this is actually my RC settings
20:18:07 <Discord|> <13S​astreii> hi guys! i've finally awaken from my eternal slumber and i'm back doing some doodles
20:18:20 <Discord|> <09g​ammafunk> nice!
20:18:22 <Discord|> <13S​astreii> i've done some zombies for warming up
20:18:31 <Discord|> <13S​astreii> https://cdn.discordapp.com/attachments/747522859361894521/884593439080067082/Schermata_2021-09-07_alle_02.05.43.png
20:18:32 <Discord|> <13S​astreii> https://cdn.discordapp.com/attachments/747522859361894521/884593442305482752/Schermata_2021-09-07_alle_02.08.33.png
20:18:38 <Discord|> <09g​ammafunk> 0.26-a0-814-ge366a860e6  not sure if anything weird happened on the dlua/clua side of things around this time
20:19:02 <perryprog> Whoa, Sastreii those are dope
20:19:29 <Discord|> <09g​ammafunk> looks good, did you consider adding some "flesh ruining" to the hydras?
20:19:42 <Discord|> <09g​ammafunk> they are perhaps a bit too clean
20:20:13 <Discord|> <13S​astreii> yes, but in the original scale they look really messy unfortunately
20:20:15 <Discord|> <09g​ammafunk> love the orc zombie's expression
20:20:23 <Discord|> <09g​ammafunk> yeah
20:20:37 <perryprog> gammafunk, could be a POC of this vuln: https://github.com/crawl/crawl/security/advisories/GHSA-vjrj-g7gf-rgjv
20:20:40 <Discord|> <09g​ammafunk> there's probably no real risk of confusing them for some kind of "hydra statue" or w/e
20:21:29 <Discord|> <13S​astreii> yeah, the "oh shit i'm a zombie now" expression
20:21:32 * geekosaur was wondering about the vuln too
20:21:54 <Discord|> <09g​ammafunk> that's a known exploit that was fixed
20:22:00 <perryprog> Right timeframe, though
20:22:03 <Discord|> <09g​ammafunk> before 0.26
20:22:08 <perryprog> O
20:22:16 <Discord|> <09g​ammafunk> as it says on the page
20:22:33 <Discord|> <13S​astreii> here, in case you like em
20:22:33 <Discord|> <13S​astreii> https://cdn.discordapp.com/attachments/747522859361894521/884594453866741810/Hydra_zombie.zip
20:22:34 <Discord|> <13S​astreii> https://cdn.discordapp.com/attachments/747522859361894521/884594456198787092/zombie_pack.zip
20:22:39 <perryprog> gammafunk, reading is hard?
20:23:02 <Discord|> <09g​ammafunk> not to mention you can see that advil himself published it
20:23:33 <perryprog> Yeah I read the time and thought the vuln was published after that morgue file was made :P
20:23:38 <perryprog> Months are hard
20:25:18 <Discord|> <09g​ammafunk> thanks a lot, I'm sure those top ones will get used no problem, and I imagine the lower ones as well
20:25:21 <Discord|> <09g​ammafunk> definitely a big improvement