00:33:12 Unstable branch on cbro.berotato.org updated to: 0.34-a0-1075-g5347199fb9 (34) 00:46:26 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-1062-g464970fc77 01:59:07 -!- indigaz24 is now known as indigaz2 02:02:00 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 02:02:35 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 02:03:53 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 02:53:54 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 03:05:32 <12g​e0ff> @gammafunk, there's a problem with https://crawl.develz.org/wordpress/blog: "Error establishing a database connection". The old wiki is also down, https://crawl.develz.org/wiki/doku.php. 03:25:23 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 03:28:03 YellowBlue (L6 HuIE) Crash caused by signal #8: Floating point exception (D:4) 04:13:02 03Implojin02 07* 0.34-a0-1076-gf77e99c820: Simplify Vehumet spell gift weighting 10(19 minutes ago, 2 files, 12+ 36-) 13https://github.com/crawl/crawl/commit/f77e99c82033 04:23:44 03Implojin02 07* 0.34-a0-1077-ga6fc723709: Fully remove a function 10(2 minutes ago, 1 file, 0+ 2-) 13https://github.com/crawl/crawl/commit/a6fc72370960 04:32:46 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 04:53:15 <09g​ammafunk> @napkin0815 Looks like like wordpress on crawl.develz.org has gone down. Can you take a look? 05:10:42 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-gc2108db771: Improve ranged combat tabbing 10(3 days ago, 7 files, 164+ 28-) 13https://github.com/crawl/crawl/commit/c2108db77188 05:14:23 <11O​dds> ^ That updates as discussed yesterday to move the "find a ray to hit a given monster" work to C++ 05:38:17 <03i​mplojin> we have radius_iterator and monster_near_iterator classes that are used heavily throughout the codebase for this kind of thing; we'd probably prefer to use one of them here instead of hand-rolling a re-implementation of the lua logic. the syntax you're looking for is gonna be something like for (monster_near_iterator mi(&you, LOS_SOLID_SEE); mi; ++mi) , there are loads of examples of both kinds of iterators if you grep the code 05:38:58 <04d​racoomega> Surely it should be LOS_NO_TRANS? 05:39:22 <04d​racoomega> Oh, wait 05:39:33 <03i​mplojin> no, it's searching for an autofire target, it doesn't want to try to aim at things that it won't be able to fire at 05:39:34 <04d​racoomega> That includes things on the other side of grates 05:39:44 <04d​racoomega> I keep forgetting that, tbh 05:39:58 <04d​racoomega> (Since the vast majority of spell effects don't care about that part) 05:40:00 <03i​mplojin> fair tbh, we use no_trans basically everywhere for this and surely a bunch of it is wrong 05:40:29 <04d​racoomega> Well, it is deliberate that most effects function across grates, but I am sure there are a few oversights 05:42:07 <09g​ammafunk> you also have to be wary of information leak, since I think the scoring here might be looking at a number of monster properties 05:42:32 <09g​ammafunk> this was a completely clua implementation before, so its code never had to worry about that 05:42:54 <09g​ammafunk> since all the underlying functions were ultimately querying a monster_info and not the actual underlying monster 05:43:59 <04d​racoomega> Those all look like things that are public knowledge anyway 05:44:17 <04d​racoomega> At least on a cursory inspection 05:55:48 <04d​racoomega> the difference here is that now Forgecraft spells with spflag::destructive will show up as Vehumet gifts more frequently There's actually some awkwardness here in general, tbh. People have complained in past that Veh can gift these at all because (and this is quite reasonable) almost nobody worshipping Veh will actually want them. I flagged them destructive since we've been pretty consistent about any non-necromancy that can do 05:55:48 damage to a thing counting, and so it makes some sense to support them, but it does feel like, as a gift, they're almost always wasted. (At one point, I even considered seperately excluding them from the gift list, while still providing passive support for them, but never quite decided if I was okay with that.) 05:56:18 <04d​racoomega> Like, Tempering and Fortress Blast in particular are weird here. (Kinetic Grapnel is whatever.) 05:57:28 <03i​mplojin> if veh players don't want them, i'd prefer to remove spflag::destructive from them instead of maintaining a separate distinction of spells-that-partially-work-with-veh 05:58:21 <11O​dds> The things that we now look at about monsters were all in lua, so I think should be safe 05:59:32 <04d​racoomega> Yeah, I considered that also. (There was a reason I hesitated on doing this.) It's just that the current state feels consistent, even if it's not, like, helpful to most players, if that makes sense? 05:59:54 <03i​mplojin> i mean, i think the state after that last commit is also consistent? if they're range boosted, they're being considered for gifting 05:59:58 <04d​racoomega> (And like, Tempering works with almost nothing else Veh gifts, but it does actually work quite positively with Servitor, which is a popular Veh spell) 06:00:08 <04d​racoomega> No, I'm not saying the new state is inconsistent 06:00:34 <04d​racoomega> It's just 'gift the things people have said they wish weren't gifted at all' just reminds me that I was a bit unsatisfied with that part. I am not sure any change is a specific improvement, though. 06:01:20 <03i​mplojin> i'm not super convinced that veh players will never be interested in these tbh, there were kind of a lot of them 06:01:38 <03i​mplojin> i believe that some players have complained, certainly 06:02:44 <04d​racoomega> (I mean, it might be fine. I did leave it alone this long, after all.) 06:03:14 <11O​dds> I don't think we'll give tempering or fortress blast much because people won't have trained forgecraft? 06:03:42 <03i​mplojin> oh, yeah, there is that also, the weights are determined by skill training 06:05:24 <04d​racoomega> Oh, maybe I misunderstood. I think I read as their gifting weights will no longer return 0 meaning that the baseline weight in the old system was somehow below even untrained spells in normal schools before and now wouldn't be. 06:05:42 <04d​racoomega> This is what I get for talking about code that I barely read the original of, right? >.>; 06:06:23 <04d​racoomega> Anyway, don't mind me too much. I'm not saying this is a Problem. Just me thinking about loud about something I'd felt conflicted about in past. 06:06:32 <03i​mplojin> that commit message could have been clearer, the previous weights were a flat amount + the skill weighting 06:06:49 <03i​mplojin> previously, the skill weighting portion was zero if the skill wasn't being supported iirc 06:06:53 <04d​racoomega> Yeah, I thought this was saying that in the old system, they also weren't getting the flat amount 06:07:06 <04d​racoomega> So they'd just become more common for everyone 06:15:30 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-gb91a849698: Improve ranged combat tabbing 10(3 days ago, 7 files, 159+ 28-) 13https://github.com/crawl/crawl/commit/b91a849698fd 06:32:33 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-g6e1e1ec984: Improve ranged combat tabbing 10(3 days ago, 8 files, 160+ 28-) 13https://github.com/crawl/crawl/commit/6e1e1ec98412 06:52:51 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-gaf2dc8fe47: Improve ranged combat tabbing 10(3 days ago, 8 files, 162+ 28-) 13https://github.com/crawl/crawl/commit/af2dc8fe4734 07:45:08 03DracoOmega02 07* 0.34-a0-1078-gbaffcba723: Give Slaughter form rMiasma (Acrobat) 10(73 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/baffcba723fd 08:03:18 <11O​dds> Fixed this one (though in fact it was a radius iterator we wanted, as we don’t only shoot at monsters) 08:03:52 <11O​dds> That was meant to reply to this 08:05:43 considersuicide (L3 MDMo) Crash caused by signal #8: Floating point exception (D:2) 08:15:27 <03i​mplojin> First pass looks mostly okay to me but as gammafunk mentioned earlier it might be better to setup the stuff in is_candidate_for_attack to use a monster_info instead of a monster for reasons of preserving infoleak boundaries in future. Crawl goes way out of its way to try to ensure that player-side ui info and clua binds don't have access to monster internals, which often contain state that we don't want players to directly view. 08:15:28 Setting this up as a monster now can mean that future extensions to this behavior may unintentionally infoleak things -- if we can get the same info we need by using a monster_info then we should, as long as we're trying to preserve this monster/monster_info boundary elsewhere 09:40:24 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-g83dbe36da0: Improve ranged combat tabbing 10(3 days ago, 10 files, 172+ 28-) 13https://github.com/crawl/crawl/commit/83dbe36da04f 10:03:51 03CrawlOdds02 07https://github.com/crawl/crawl/pull/4869 * 0.34-a0-1069-g2ff7cca182: Improve ranged combat tabbing 10(3 days ago, 10 files, 172+ 28-) 13https://github.com/crawl/crawl/commit/2ff7cca1826e 10:09:08 -!- lauregami is now known as laurie 11:49:06 HoriaBersec (L7 OpCj) Crash caused by signal #8: Floating point exception (Temple) 11:49:20 HoriaBersec (L7 OpCj) Crash caused by signal #8: Floating point exception (Temple) 11:49:43 HoriaBersec (L7 OpCj) Crash caused by signal #8: Floating point exception (Temple) 11:50:35 HoriaBersec (L7 OpCj) Crash caused by signal #8: Floating point exception (Temple) 11:50:56 HoriaBersec (L7 OpCj) Crash caused by signal #8: Floating point exception (Temple) 11:52:04 <03i​mplojin> a cue rebuild would fix that if anyone around has access 12:35:55 HoriaBersec (L8 DECj) Crash caused by signal #8: Floating point exception (Temple) 13:14:35 <09g​ammafunk> cue rebuild url doesn't work, but you can ping TZer0. @tzer0 if you have a chance and can trigger a cue rebuild for trunk, we'd appreciate it. It's nothing too bad, if you can't get to it 13:18:18 <02t​zer0> On it 13:19:25 <02t​zer0> Triggered 13:32:04 <09g​ammafunk> TY! 13:38:01 Sar (L8 ReIE) Crash caused by signal #8: Floating point exception (Temple) 13:38:14 Sar (L8 ReIE) Crash caused by signal #8: Floating point exception (Temple) 13:38:30 Sar (L8 ReIE) Crash caused by signal #8: Floating point exception (Temple) 13:42:26 uh 14:01:51 Unstable branch on underhound.eu updated to: 0.34-a0-1078-gbaffcba723 (34) 14:55:42 GordonOverkill (L12 OnDe) Crash caused by signal #8: Floating point exception (Lair:1) 14:59:08 GordonOverkill (L12 OnDe) Crash caused by signal #8: Floating point exception (Lair:2) 17:19:20 <03i​mplojin> !crashlog GordonOverkill 17:19:26 <04C​erebot> 8. GordonOverkill, XL12 OnDe, T:20969 (milestone): https://underhound.eu/crawl/morgue/GordonOverkill/crash-GordonOverkill-20251103-215908.txt 17:19:39 <03i​mplojin> !crashlog Sar 17:19:42 <04C​erebot> 24. Sar, XL8 ReIE, T:7045 (milestone): https://underhound.eu/crawl/morgue/Sar/crash-Sar-20251103-203828.txt 17:19:44 <03i​mplojin> oh hey Sar is still around, cool 17:19:55 <03i​mplojin> yeah these were all before the rebuild 17:20:02 <03i​mplojin> guess it just took a while to run 17:21:02 <03i​mplojin> !fastversions 17:21:04 <04C​erebot> Latest versions seen in milestones this week: [LLD] => 0.33.1-1-gc282f97, [CAO] => 0.34-a0-1061-g47010eb, [CDO] => 0.34-a0-1062-g464970f, [CXC] => 0.34-a0-1072-ga5e1a67, [CBR2] => 0.34-a0-1075-g5347199, [CDI CNC CPO CUE] => 0.34-a0-1078-gbaffcba 17:44:05 03Implojin02 07* 0.34-a0-1079-g4b5df66c0b: Calm down MUT_BERSERK (Lici) 10(4 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/4b5df66c0b5f 17:49:27 yeh, I realized there wasn't a problem when the update message came in. then the following two were just pre-update games that hit the crash late 18:43:11 horredbonez (L11 DsWn) Crash caused by signal #6: Aborted (Lair:1)