00:00:04 <04dracoomega> I mean, I generally agree with your statement, but that specifically doesn't matter for this 00:00:51 <04dracoomega> (I mostly only preserved the 'hide in searches' part since I was trying to a quick solution to a soft-lock and that was a quick non-buggy way to leave some more time to mull it over) 00:01:19 <04dracoomega> I certainly would not be opposed to removing the unobtainable flag in general 00:02:39 <07wizardike> Oh that's not as bad as I thought. I guess this would be good if it wasn't wrong half the time 00:03:03 <11Odds> I'm not even sure it would... I'd be confused if I noticed this as a player 00:03:42 <11Odds> Especially xv'ing being weird 00:04:32 <04dracoomega> (It's not like you can't already ctrl+f things that are not currently attainable in practice) 00:04:52 <04dracoomega> Like inside of certain very nasty vaults 00:05:11 <04dracoomega> I did fix that part today, at least 00:05:19 <04dracoomega> Since that is clearly just buggy behavior 00:05:26 <11Odds> Ah right, nice 00:05:29 <04dracoomega> But they still won't show up on ctrl+f 00:05:41 <04dracoomega> (Or affect ?acq, for how little that matters) 00:10:37 <07wizardike> Another option is just destroying the items if the player takes the transporter and it doesn't lead to them but I'm not sure this is any better than just showing them in item search 03:36:01 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 07:13:32 03CrawlOdds02 07* 0.35-a0-826-g970ca23689: Fix an issue with eldritch portals not disappearing 10(4 hours ago, 1 file, 0+ 3-) 13https://github.com/crawl/crawl/commit/970ca236895e 08:01:55 Centroid (L6 MiBe) Crash caused by signal #11: Segmentation fault (D:5) 09:24:32 <09gammafunk> I wonder what centroid is doing to regularly trigger crashes like this 09:24:35 <09gammafunk> &rc centroid 09:24:39 <04Cerebot> https://crawl.akrasiac.org/rcfiles/crawl-0.34/Centroid.rc 09:24:49 <09gammafunk> !crashlog centroid 09:24:54 <04Cerebot> 11. Centroid, XL6 MiBe, T:3513 (milestone): https://crawl.akrasiac.org/rawdata/Centroid/crash-Centroid-20260803-150155.txt 09:25:29 <09gammafunk> oh, is this related to the recent stash issue/discussion? 09:32:30 <11Odds> Yep, and DO fixed the actual crash 09:32:33 <11Odds> !versions 09:32:36 <04Cerebot> Latest versions seen in milestones this week: [LLD] => 0.34.1-4-g0e95e08, [CDO] => 0.35-a0-758-gf74bb9b, [CXC] => 0.35-a0-818-g31569e1, [CAO] => 0.35-a0-819-gf82e23d, [CUE] => 0.35-a0-820-ga221df3, [CBR2 CRG CBRG] => 0.35-a0-825-g2ba6933, [CDI CNC CPO] => 0.35-a0-826-g970ca23 09:33:04 <11Odds> 825 is the fix so CAO is still bad 09:41:45 <09gammafunk> rebuild trigger time I guess 09:46:20 <09gammafunk> wow did recent changes in the last 12 hours really invalidate the much ccache 09:46:36 <09gammafunk> I should check that CAO is actually properly using it, it certainly seems to be based on pathing 09:46:59 <09gammafunk> but that issue where rebuild CGI sees the correct python and cron rebuild apparently doesn't is weird 09:47:24 <09gammafunk> it's only 6 commits, let's see 09:49:06 <09gammafunk> only header change is to stash.h, so it certainly seems like it shouldn't 11:06:06 03Zigmound02 07https://github.com/crawl/crawl/pull/5354 * 0.35-a0-820-g0fb1d14f5c: Fix review comments 10(89 seconds ago, 2 files, 5+ 1-) 13https://github.com/crawl/crawl/commit/0fb1d14f5c84 11:38:09 03DracoOmega02 07* 0.35-a0-827-g91ee021ed7: More properly fix eldritch tentacle death cleanup 10(30 seconds ago, 1 file, 8+ 0-) 13https://github.com/crawl/crawl/commit/91ee021ed738 11:38:36 <04dracoomega> @Odds Thanks for looking into that bug (my apologies for not getting to it sooner), but I recalled that there was a reason it was doing extra cleanup in the death code rather than just wait for automatic cleanup ^^; 11:39:07 <04dracoomega> (Since during the major refactoring I did to it this version, I did also remove the death cleanup at one point, only to run into this issue and put it back - though apparently somewhat incorrectly at the time) 11:45:12 -!- MakMorn_ is now known as MakMorn 12:14:17 <11Odds> Ah fair enough! I thought about fixing it that way but couldn’t see why it mattered (and as it was usually doing nothing assumed it wasn’t important!) 12:14:53 <11Odds> But yes I see that if something dies late enough it would hang around for a player turn 12:24:17 <11Odds> However, even this version won’t fire if the tentacle is destroyed via a segment (see https://github.com/crawl/crawl/issues/5361) 12:43:50 <04dracoomega> Oh no, really? =/ 12:45:02 <11Odds> Yep, I’m pretty sure the code path that intends us to reach this block for killing a segment has been dead for a solid decade 12:45:33 <11Odds> (And you can easily see we don’t get the messages when doing so) 12:46:30 <04dracoomega> This looks like because mons_reset includes KILL_TENTACLE_CLEANUP. It's possible the portal cleanup check could allow specifically that 12:46:49 <04dracoomega> Which seems like it should be simple 12:47:17 <11Odds> Yeah, I thought that but when I realised it stopped the tentacle going in vanquished creatures I thought I’d think harder about the right structural fix 12:47:24 <04dracoomega> Since tentacles never give XP, there is no mechanical effect that I can think of. Wait, what about the ones that are normal monsters? 12:47:38 <04dracoomega> They surely give XP 12:47:43 <04dracoomega> (Like in the Golubria wizlab) 12:47:43 <11Odds> They don’t! 12:47:58 <04dracoomega> Even if you kill the head? 12:48:08 <11Odds> My read of the code was no, never 12:48:24 <11Odds> Tentacles 🤷 12:49:43 <04dracoomega> I am pretty sure you do 12:49:57 <04dracoomega> (At least, putting a statement in at the point where xp is granted shows that it is called) 12:50:21 <04dracoomega> But you are correct that if you kill a segment, you get none 12:50:25 <11Odds> Ah right. I didn’t check empirically, so you’ll be right 12:50:47 <04dracoomega> So there are some general structural problems about the head not properly dying in multiple ways on a segment kill 12:50:49 <11Odds> The mons_class_gives_xp or whatever it is sure looked like it excludes them 12:51:17 <08o____0> If I melee the eldritch tile itself I get xp. If I step back and kill the tentacle I do not 12:51:17 <08o____0> https://cdn.discordapp.com/attachments/747522859361894521/1533925241443717382/0v29k9x.png?ex=6a7242b4&is=6a70f134&hm=fbdd370b5c8fa5597d1ff6a8e7504c02baac9832ab6b9883409f8f7d5a5a5f40& 12:51:20 <11Odds> Exactly so I wanted to think about the right way to properly kill it 12:51:21 <04dracoomega> ....you're right that it does look like it should, though 12:51:29 <04dracoomega> Er, that it should not. But it clearly seems to. 12:51:44 <04dracoomega> (Design-wise, it obviously should give actual XP if it's an actual independent monster, of course) 12:51:50 <11Odds> I’ll certainly defer to reality on this one 12:52:29 <04dracoomega> (It looks like mons_class_gives_xp() isn't actually used in the xp-granting code) 12:52:37 <04dracoomega> But rather various places that describe them 12:52:49 <04dracoomega> And polymorph and a few other places 12:53:16 <11Odds> Kinda wonder if segments shouldnt die for themselves at all but just kick the death up to the head 12:53:17 <04dracoomega> ...wait C++ // Early out for no XP monsters. if (!mons_class_gives_xp(mc)) return 0; 12:53:48 <04dracoomega> (This may not matter, but as a general principle, I like to understand why 'unexpected' things are actually happening, just in case some other weirdness is afoot) 12:54:06 <11Odds> Oh for sure 12:54:12 <04dracoomega> As in: short-circuit their own death, but kill the head normally, and have the head call destroy_tentacle() 12:54:17 <11Odds> I’d be doing the same but I’m not at the code 12:54:21 <11Odds> Yeah 12:54:22 <04dracoomega> That sounds reasonable at first brush 12:54:35 <04dracoomega> Might be some non-obvious gotcha, but sounds reasonable 12:55:56 <04dracoomega> Oh, tentacles only return false for mons_class_gives_xp if a second argument is true 12:57:20 <04dracoomega> Which is only true for... xv-ing a monster and saying whether it normally gives xp, if a given instance of that monster is also unrewarding? 12:59:00 <11Odds> The zombie kraken case I guess 12:59:06 <04dracoomega> Oh, wait, I got thrown off by your description. What mons_class_gives_xp is doing is actually the opposite 12:59:35 <04dracoomega> If 'indirect' is true, it's saying that the segments do give XP (because killing them will kill the head, which gives XP) 12:59:43 <04dracoomega> Even though they themselves do not 13:00:48 <11Odds> Ahhh right so that bit makes sense, I just read it wrong 13:03:43 <04dracoomega> Well, that's one mystery solved, at least ^^; 13:28:05 <11Odds> (I fixed the issue to be more accurate) 13:28:27 <11Odds> Will have a little look at how we might do this cleanup better 14:10:11 <08o____0> https://discord.com/channels/205277826788622337/205316046230388737/1532191169298169866 So there's a cool Blorkula mitosis bug where if a batula gets shafted, it can create a 2nd copy of blorkula on another floor. I can't think of what the correct behavior should be though. It feels wrong to allow Blorkula to be duplicated like this but there is a logic to it haha. some possible fixes: Make blorkula bats shaft immune Make blorkula 14:10:11 bats shaft immune if there is at least 1 other blorkula bat on the level (so the last bat is not immune) (soul tether prevents the shaft) Kill blorkula bats that get shafted (soul tether is broken) (except for the last bat) Erase the blorkula properties from bats that get shafted so it's just a normal bat (except for the last bat) 14:26:57 <04dracoomega> I'm content to just not let them get shafted, tbh 14:27:07 <04dracoomega> No need for anything more complex 14:28:13 I mean yeah they're bats how they gonna go down shaft 14:36:42 <08o____0> (flying enemies are magically sucked down shafts!) 14:44:41 air currents? 14:49:09 <12ge0ff> Shafts are just a part of the Dungeon's central vacuum system: https://en.wikipedia.org/wiki/Central_vacuum_cleaner 14:50:07 hvac, they're why z:5 has air 15:01:00 -!- The topic of #crawl-dev is: Crawl Development | https://github.com/crawl/crawl | Logs: http://s-z.org/crawl-dev/, temporarily http://crawl.akrasiac.org/logs/cheibriados/ | People with +v have commit access, devs on bridged discord as well | General Crawl-related chat to #crawl | Long stuff to a pastebin service, please 15:06:00 03DracoOmega02 07* 0.35-a0-828-g5681d780c1: Don't retain cursor memory after casting hellfire mortar (adinsx) 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/5681d780c16e 15:19:31 orb of zot is an air conditioner, good to know 15:19:39 that would revive a god in this heat 15:20:06 ignis didn't die, they got out-powered by global warming 15:26:46 the orb dias is hooked up to pan, so they get mad when they lose their cool air and rush to you to get it back 15:27:48 why would fiery demons want cool air? 15:28:35 they're giant balls of fire, how else are they gonna stay cool? 15:28:48 more likely they're pissed that you're cooling the whole dungeon off with the orb 15:30:10 yeah I suppose they are the ones footing the bill 15:30:39 fire needs air to live perhaps 15:30:51 this is now my headcanon until proven otherwise 15:43:51 Unstable branch on underhound.eu updated to: 0.35-a0-827-g91ee021ed7 (34) 19:12:37 03Zigmound02 07https://github.com/crawl/crawl/pull/5354 * 0.35-a0-821-ge04186fff2: Add trap_is_safe_from_afar 10(2 minutes ago, 3 files, 9+ 1-) 13https://github.com/crawl/crawl/commit/e04186fff25e 19:21:11 03Zigmound02 07https://github.com/crawl/crawl/pull/5354 * 0.35-a0-822-gacb8341cf7: Added teleportation traps to `trap_is_safe_from_afar` 10(73 seconds ago, 1 file, 5+ 3-) 13https://github.com/crawl/crawl/commit/acb8341cf74f 19:27:35 New branch created: pull/5362 (1 commit) 13https://github.com/crawl/crawl/pull/5362 19:27:35 03autumn02 07https://github.com/crawl/crawl/pull/5362 * 0.35-a0-828-g3da4979052: extend explore_greedy_visit to control visiting shops and faded altars 10(6 minutes ago, 8 files, 67+ 22-) 13https://github.com/crawl/crawl/commit/3da497905281 20:44:39 03autumn02 07https://github.com/crawl/crawl/pull/5362 * 0.35-a0-829-g81cf389752: check for forlorn in faded_altar_needs_visit 10(37 seconds ago, 1 file, 2+ 1-) 13https://github.com/crawl/crawl/commit/81cf38975282 22:37:39 Unstable branch on crawl.develz.org updated to: 0.35-a0-828-g5681d780c1 (34) 23:02:24 Windows builds of master branch on crawl.develz.org updated to: 0.35-a0-828-g5681d780c1 23:12:34 Unstable branch on cbro.berotato.org updated to: 0.35-a0-828-g5681d780c1 (34) 23:59:54 Monster database of master branch on crawl.develz.org updated to: 0.35-a0-828-g5681d780c1