00:22:40 (Yep, the actual problem was at our end) 00:32:29 something is up with cdo, not sure what 00:32:35 very slow disk access 00:32:43 the disk is pretty full, it might just be that 00:36:21 does anyone remember how the version in https://crawl.develz.org/trunk/ is set? 00:36:21 I can't find whatever script it is that generates that page 01:08:24 Stable branch on crawl.develz.org updated to: 0.25.0-45-g978344163c (34) 01:38:17 Unstable branch on crawl.develz.org updated to: 0.26-a0-274-g7ed2249cc2 (34) 01:46:27 how are people feeling about delver? 01:46:32 unrelatedly, how are people feeling about palentonga? 01:56:38 the feedback i've seen is that people who have played delver seem quite happy with it; i'd sort of like to turn off portal vaults on d:5 for them, but other than that i feel happy with them 01:57:03 palentonga people seemed to like but think 'needed something more'. that was before they got bardings, so i'm not sure where people are at on that now 02:01:31 -!- myfreeweb_ is now known as myfreeweb 02:05:57 -!- neunon_ is now known as neunon 02:05:58 -!- dog_star_ is now known as dog_star 02:06:28 turning off portal vaults messes with seeded competitions 02:06:55 mb the portal doesn't open until the second time you enter the level? 02:07:50 delver fundamentally breaks seeds 02:08:10 it guarantees that no monsters spawn in sight of the player, which shifts stuff around 02:08:45 well, "fundamentally" is a strong term. "I'm not sure how much effort it's worth putting into getting this one background working for seeded competitions instead of banning it" is perhaps a better way to put it 03:12:08 Stable (0.25) branch on crawl.beRotato.org updated to: 0.25.0-45-g9783441 03:26:09 Unstable branch on crawl.beRotato.org updated to: 0.26-a0-274-g7ed2249 (34) 03:31:54 Stable (0.25) branch on crawl.kelbi.org updated to: 0.25.0-45-g978344163c 03:48:15 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-3439-g0cad18c694 07:45:31 03mainiacjoe02 07https://github.com/crawl/crawl/pull/1484 * 0.26-a0-261-g8cfadc0: Reduce starting Dodging to 1 10(3 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/8cfadc0823eb 09:45:08 !tell floraline dolemite99 wants their BcadrenCrawl save deleted; it's completely locked by an old crash. (issue with saves from between two minor_tag versions; doesn't happen any more; but still getting the occasional "some old save still stuck there") 09:45:09 Bcadren: OK, I'll let floraline know. 10:16:51 delver to me doesn't look like it fundamentally breaks seeding, the only issue might be if a unique gets zapped (but even that looks like it would be ok to me) 10:17:04 disabling portals doesn't sound ideal though, seems like part of the delver cost 10:17:21 oh, didn't even think as to how delvers actually place 10:17:22 easy way to do that with respect to seeding would be to remove the portal after levelgen 10:17:29 do they just spawn on a random tile? 10:20:54 they spawn on one of the downstairs to d:6 10:21:30 and, in addition to zapping monsters, they use your new code to avoid generating monsters nearby... (I could turn that back off though) 10:21:40 (that's the part that messes with seeds) 10:23:37 yeah, I guess both would cause seed problems wouldn't they 10:24:26 advil: hrm, can you do both zapping and replacement post-generation to avoid seed issues? 10:24:42 I guess zapping uniques seems like a problem, but you already mentioned that's fine 10:25:53 doing both those things (zapping for los, replacement for distance) does seem desireable for delver 10:25:53 actually, now that I think about it...oh 10:26:02 you need both; presumably delver can just place in some downstair vault or just in los of a vault 10:26:07 hence does need to zap things 10:26:18 the replacement thing is only for level placement monsters 10:26:48 um 10:28:29 when I try to compile trunk, compilation fails (full debug webtiles) http://dpaste.com/3F6YAEZ 10:28:54 gammafunk: cmd-name.h needs to be regenerated 10:29:17 already did make clean 10:29:43 joemaro had this problem last night and had to remove it explicitly; IDK if make clean doesn't DTRT 10:30:07 it got through macro.cc, seems to have fixed, thx 10:30:13 01:40 <+advil> ah needed to rm cmd-name.h 10:30:17 also, I'm getting this warning on my system now, which i don't recall 10:30:19 levcomp.ypp:40.1-7: warning: POSIX Yacc does not support %expect [-Wyacc] 10:30:19 40 | %expect 2 10:30:19 | ^~~~~~~ 10:30:36 ubuntu 20 system, so maybe this is a change with newer yacc 10:30:47 crawl is once again having yak problems 10:31:25 er 10:31:29 I meant to say "alright, did make clean", not "already did make clean" 10:31:32 make clean fixed it 10:42:48 %git stone_soup-0.25 10:42:48 07advil02 * 0.25.0-45-g9783441: Increase size of zig permastore the easy way 10(20 hours ago, 1 file, 0+ 0-) 13https://github.com/crawl/crawl/commit/978344163c4f 10:42:53 nice 10:46:39 Sleeping(?) Butterfly Has Batty Movement 13https://crawl.develz.org/mantis/view.php?id=12311 by MainiacJoe 10:46:45 03gammafunk02 07* 0.26-a0-275-g3af348c: Fix a monster equipment DES error 10(5 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/3af348c12c07 10:47:35 03gammafunk02 07[stone_soup-0.25] * 0.25.0-46-gbd7b085: Fix a monster equipment DES error 10(6 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/bd7b0854d241 11:01:15 fr: a CI test that fails when errors like that in DES are found by crawl 11:01:50 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-275-g3af348c12c (34) 11:12:02 is it not something that is caught by builddb? 11:14:28 right 9f4bf6adc57c6b9 is really not recommended for delver 11:14:36 %git 9f4bf6adc57c6b9 11:14:36 07PleasingFungus02 * 0.26-a0-231-g9f4bf6a: Delver: Fix monster spawn (WiseOldWolf) 10(5 days ago, 5 files, 32+ 14-) 13https://github.com/crawl/crawl/commit/9f4bf6adc57c 11:14:50 that will most likely mess up all aspects of the seed forever 11:14:53 can certainly revert that 11:15:31 does it cover something zapping doesn't? 11:15:31 iirc yes 11:15:49 one idea might be to always prepare a delver landing spot on D:5 for all backgrounds 11:15:55 there was something about monsters not getting zapped if they were within LOS radius but not actually in LOS 11:16:05 I actually sort of like that idea because it even covers the minor details 11:16:24 it's a little funny but i don't mind it 11:17:08 this is all stuff I can attend to btw 11:18:00 sometimes I have been known to passive aggressively whine when other commits make me do stuff but seeding-related changes are not one of those ;-) 11:18:24 that'd be great! 11:20:03 gammafunk: do you have any idea what to update on cdo to make the nightly build version change? It needs bumped to 0.26 and I'm *pretty* sure we have permissions, but I could not for the life of me find it 11:20:25 for the website that is 11:21:03 it's not detecting new builds because it's set to 0.25-a or something 11:22:09 am I getting it right that it just picks stairs down I 11:22:10 ? 11:22:51 that sounds right 11:24:49 hm is that intentionally wrapped in one_chance_in(8)? 11:25:31 looks like 7/8 times it will not do the starting_absdepth check 11:27:12 although at that point absdepth might actually be set to 0 11:27:41 now I'm very confused 11:29:28 is absdepth0 0 or 1 indexed? 11:29:31 not even sure 11:30:26 comments do indicate it might be 0 indexed 11:31:28 i believe it's 0 indexed, though i didn't check 11:32:05 yeah, I'm pretty sure it is 11:32:05 oops, i fumbled the logic there 11:32:15 i took several tries at it... i think i needed a flowchart 11:33:03 no, that check shouldn't be wrapped in a 1/8 chance 11:34:03 so, pre-delver the logic was: if on d:1, place always asleep and away from entrance. otherwise, on a 1/8 chance, place awake but not near the stairs. otherwise, outside pan, place asleep and potentially anywhere. otherwise, in pan, place awake and potentially anywhere. 11:34:11 it's a bit tricky but I think both the second and third condition should get an extra check to add on the proximity thing on starting absdepth 11:34:17 yes 11:35:33 with delver, the logic was supposed to be: if on d:1 or (a 7/8 chance and not in pan), place asleep; otherwise, place awake. SEPARATELY, if on starting_absdepth, place away from entrance; otherwise, if awake and not in pan, place away from stairs; otherwise, place wherever. 11:35:51 if that makes sense...? 11:38:55 what about: if D:1, place asleep and away from entrance, if 1 chance in 8: delver && starting absdepth ? away from entrance : away from stairs, else delver && starting absdepth ? away from entrance && asleep : asleep 11:39:00 that may well be what you just said :-) 11:39:26 add in the pan checks as needed 11:40:02 bloatcrawl species idea, superdelver = start on pan exit portal 11:40:09 ert 11:40:12 background idea 11:52:44 advil: doesn't any use of away_from_entrance after d:1 break seed compat? 11:54:10 it's sort of iffy. any monster that would try to place there would instead retry up to 45 (?) times 11:54:36 so we have issues if and only if it fails to find a valid location within those 45 tries 11:54:44 yes, probably 11:54:48 which probably only happens on one of those tiny weird levels 11:54:59 well, dungeon building is very much "a butterfly flaps its wings" 11:55:09 so even moving one monster may have a chaotic effect on everything else 11:55:26 since then other checks can condition on what's at the place the monster was or is 11:55:46 mm 11:55:57 anyway, i guess when you said you wanted to fix seed compat, i figured we'd revert this commit entirely 11:56:02 and instead just zap monsters more aggressively 11:56:22 maybe, yeah, I was going to play with it a bit since tbh I don't really know how that prox thing works 11:56:44 !source _zap_los_monsters 11:56:45 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/startup.cc#L211 11:56:49 re the stairs thing, it may work to check for the opaque kmask if that info is still around 11:56:50 the function is only called in one place 11:56:55 that one I do understand 11:57:04 i think it's as simple as just killing literally everything in radius 7 or w/e it is 11:57:25 oh, wait 11:57:28 that breaks funky entrance vaults 11:59:00 could be more aggressive about murder on d:5 instead and hope we don't tragically slaughter one of the weird altar vault's inhabitants? 11:59:41 or could shrug and say, hey, looks like a monster spawned within radius 7 of you on d:5. that's real rough. sorry to hear that 12:00:11 in terms of feelbad, i think the downstairs altar vaults or the portal vaults are worse 12:00:34 I think the one I was trying to remember is: 12:00:36 !vault minivault_12 12:00:37 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/dat/des/variable/mini_monsters.des#L7426 12:00:40 which doesn't place in the relevant range 12:00:56 convenient! 12:01:21 but that's what makes me think the kmask might be a shortcut for checking pathing 12:01:39 any non-connected vault should be masked as opaque 12:01:46 trying to enforce pathability to an upstairs is giving me wulndraste flashbacks 12:01:49 so we could pick the first downstairs without that mask 12:03:57 (keeping in mind that there might not be such a downstairs) 12:04:41 that won't be possible 12:05:13 i.e. a level like that would veto 12:05:13 huh 12:05:23 probably i'm misunderstanding something 12:05:25 I'm not sure of the exact constraints but that would mean that there's no path at all between an upstairs and a downstairs, old-orc-style 12:05:38 no path on the same level 12:05:41 i miss old orc 12:05:58 my first bug report was for a disconnected orc bubble! 12:06:51 nice 12:57:54 oh hm case 2 (the 1/8 chance) is even different than I thought, I think this should be unconditional away from stairs 12:58:15 (since for delver as currently implemented, away from stairs entails away from entrance, but not vice versa) 12:59:29 that is, awake monsters (1/8 chance) are supposed to always generate away from any stairs 12:59:32 not just downstairs 13:02:46 Stable (0.25) branch on crawl.akrasiac.org updated to: 0.25.0-46-gbd7b085 13:03:27 PleasingFungus: did you already start working on any fixes for these? If not I can push some stuff that brings it was what (I think) you had in mind, as a preface to figuring out what to do about seeding 13:03:37 all yours 13:23:45 03advil02 07[delpal] * 0.26-a0-274-g061143e: lint 10(10 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/061143efc956 13:23:45 03advil02 07[delpal] * 0.26-a0-275-g57b06d1: Fix some delver issues with initial monster placement 10(8 minutes ago, 2 files, 26+ 20-) 13https://github.com/crawl/crawl/commit/57b06d1c2606 13:31:58 Unstable branch on crawl.akrasiac.org updated to: 0.26-a0-275-g3af348c (34) 13:37:54 04Build failed for 08delpal @ 57b06d1c 06https://github.com/crawl/crawl/actions/runs/165771727 13:48:30 a d:5 delver entrance vault should have an unlucky delver's corpse when you're not a delver :D 14:04:19 The build is still failing. (delpal - 57b06d1 #13882 : advil): https://travis-ci.org/crawl/crawl/builds/707205925 14:25:21 03advil02 07[delpal] * 0.26-a0-276-gde8b981: Fix a tag_upgrade issue 10(2 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/de8b981d7d19 14:43:29 04Build failed for 08delpal @ de8b981d 06https://github.com/crawl/crawl/actions/runs/165807195 14:45:10 some day I need to figure out how to run the upgrade script locally 15:02:58 The build is still failing. (delpal - de8b981 #13883 : advil): https://travis-ci.org/crawl/crawl/builds/707217384 15:31:35 LRD does not auto-target behind the statues (it does manually) 13https://crawl.develz.org/mantis/view.php?id=12312 by hayenne 15:51:27 03kate-02 07* 0.26-a0-276-g21ffff0: Improve descriptions of Ash's passive abilities (#12245) 10(2 minutes ago, 2 files, 12+ 0-) 13https://github.com/crawl/crawl/commit/21ffff02790f 15:52:53 man, you.no_save really does not do what you'd think 15:56:52 LRD aim helper portrays safe tiles as being affected by the explosion 13https://crawl.develz.org/mantis/view.php?id=12313 by hayenne 15:57:22 er, Options.no_save 15:58:24 this green crystal level layout is seemingly bugged, I think I just killed a monster in the next room by blasting the wall from the other side 16:02:17 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-276-g21ffff0279 (34) 16:02:54 if it's an lrd issue it may be something about the spell + crystal (which generates a bigger explosion) than the level per se 16:06:21 right, I was under the impression LRD does not explode from the other side of the wall. might have been the case in the remote past when I played 16:06:29 that should be true 16:06:50 but I just checked and it does explode from the other side, even with the plain rock walls 16:08:25 how did you check? if wizmode xray vision I'd want to verify that that doesn't interfere 16:10:32 03advil02 07* 0.26-a0-277-gdb804b0: Don't recursively crash during tests 10(10 minutes ago, 1 file, 4+ 1-) 13https://github.com/crawl/crawl/commit/db804b0e95b8 16:10:32 03advil02 07* 0.26-a0-278-g3b0d5af: Don't write .prf files for tests 10(72 seconds ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/3b0d5afba560 16:10:56 did not use xray, just the monster detection: https://i.imgur.com/1zEr39U.png 16:16:55 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-278-g3b0d5afba5 (34) 16:17:31 I'm going to merge master into delpal in order to get those commits 16:17:42 (that's how we're doing this right?) 17:13:40 -!- hayenne_ is now known as hayenne 17:28:24 New branch created: charming-items (8 commits) 13https://github.com/crawl/crawl/tree/charming-items 17:28:24 03ebering02 07[charming-items] * 0.26-a0-276-g253b0db: Charm dispersal 🔕: Move Ozocubu's armour to ice only 10(5 hours ago, 6 files, 7+ 24-) 13https://github.com/crawl/crawl/commit/253b0dbaee70 17:28:24 03ebering02 07[charming-items] * 0.26-a0-277-g2a4b181: Charm dispersal 🔕: Move Swiftness to Air only 10(5 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/2a4b181fd51b 17:28:24 03ebering02 07[charming-items] * 0.26-a0-278-g0d274a8: Charm dispersal 🔕: Move Death's Door to Necromancy only 10(5 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/0d274a833fd8 17:28:24 03ebering02 07[charming-items] * 0.26-a0-279-g002bdb2: Charm dispersal 🔕: Move Excruciating Wounds to Necromancy only 10(4 hours ago, 4 files, 14+ 2-) 13https://github.com/crawl/crawl/commit/002bdb213f40 17:28:24 03ebering02 07[charming-items] * 0.26-a0-280-g95e380a: WIP Charm dispersal 🔕: Move Battlesphere to Conjurations only 10(5 minutes ago, 2 files, 2+ 2-) 13https://github.com/crawl/crawl/commit/95e380a8a83f 17:28:24 03ebering02 07[charming-items] * 0.26-a0-281-ga73a312: Charm dispersal 🔕: Move Song of Slaying to Tmut and Reflavour 10(4 hours ago, 20 files, 54+ 78-) 13https://github.com/crawl/crawl/commit/a73a3127c4d3 17:28:24 03ebering02 07[charming-items] * 0.26-a0-282-gfce1f60: Remove Sticks to Snakes 10(4 hours ago, 15 files, 9+ 158-) 13https://github.com/crawl/crawl/commit/fce1f603a566 17:28:24 03ebering02 07[charming-items] * 0.26-a0-283-gcc5f3d8: Charm dispersal 🔕: Spell removals 10(2 hours ago, 33 files, 21+ 448-) 13https://github.com/crawl/crawl/commit/cc5f3d896abc 17:42:21 04Build failed for 08charming-items @ cc5f3d89 06https://github.com/crawl/crawl/actions/runs/165910004 17:44:14 ooh nice, good stuff 17:45:47 good use of commit emoji 17:47:11 of the single-school ex-charms i'd guess ddoor would be pretty likely to want a level bump, going to single-school is a pretty considerable change at that level 17:48:40 and i feel like necromancy has long deserved to have one of ddoor/borg be l9 either way 18:23:13 Unstable branch on underhound.eu updated to: 0.26-a0-278-g3b0d5afba5 (34) 18:29:45 03advil02 07[delpal] * 0.26-a0-303-g05468a6: Merge branch 'master' into delpal 10(2 hours ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/05468a61bbe8 18:29:45 03advil02 07[delpal] * 0.26-a0-304-ga008e70: Fix a barding 10(2 hours ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/a008e701ff74 18:29:45 03advil02 07[delpal] * 0.26-a0-305-ge8178fa: Do something with centaur barding in makeitem test 10(27 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/e8178fa3a249 18:29:45 03advil02 07[delpal] * 0.26-a0-306-g97a777e: Fix another off-by-one error with depth 10(9 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/97a777e62f6b 18:41:57 ok, back to passing tests I think 19:27:34 03advil02 07* 0.26-a0-279-gdb952e5: Don't let Xom add to weight 0 innate mutations (Lightli) 10(2 minutes ago, 1 file, 3+ 0-) 13https://github.com/crawl/crawl/commit/db952e5dbcaf 19:32:11 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-279-gdb952e5dbc (34) 20:05:58 03advil02 07* 0.26-a0-280-g7e3f34e: Remove fast/slow metabolism 10(27 seconds ago, 9 files, 71+ 126-) 13https://github.com/crawl/crawl/commit/7e3f34eca08d 20:07:10 uh I didn't mean to actually push yet 20:07:11 well 20:07:14 we'll see how this goes 20:07:53 in fact 20:09:10 well, I was about to revert but it builds and seems to work :D 20:11:54 oh nope 20:12:05 03advil02 07* 0.26-a0-281-ga06e794: Revert "Remove fast/slow metabolism" 10(4 minutes ago, 9 files, 126+ 71-) 13https://github.com/crawl/crawl/commit/a06e7940fc23 20:15:38 04Build failed for 08master @ 7e3f34ec 06https://github.com/crawl/crawl/actions/runs/165993400 20:17:01 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-281-ga06e7940fc (34) 20:22:31 Stable (0.25) branch on underhound.eu updated to: 0.25.0-46-gbd7b0854d2 20:26:29 03advil02 07* 0.26-a0-282-geb11299: Really remove fast/slow metabolism 10(14 minutes ago, 10 files, 73+ 148-) 13https://github.com/crawl/crawl/commit/eb112998cb6b 20:33:34 The build was broken. (master - 7e3f34e #13889 : advil): https://travis-ci.org/crawl/crawl/builds/707271580 20:33:50 I wonder if there's a way to keep failures in tag_upgrade builds from interacting with non tag_upgrade builds? 20:34:14 not quite sure how actions decides when to give up on other builds based on failures in one 20:34:41 03advil02 07* 0.26-a0-283-gaa2aba8: One missed carnivorous case 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/aa2aba8d75c9 20:42:30 04Build failed for 08master @ eb112998 06https://github.com/crawl/crawl/actions/runs/166013472 20:46:53 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-282-geb112998cb (34) 20:47:26 03advil02 07* 0.26-a0-284-g73c0fb6: Add some tag version #ifdefs to mutation.cc 10(77 seconds ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/73c0fb61292d 20:54:45 04Build failed for 08master @ aa2aba8d 06https://github.com/crawl/crawl/actions/runs/166021263 21:02:37 Unstable branch on crawl.kelbi.org updated to: 0.26-a0-284-g73c0fb6129 (34) 21:09:03 i like how xom snakes to sticks will get to level up how obsolete it is with the charms stuff 21:10:06 getting to reference an ancient version of a spell that also doesn't even exist at all any more 23:41:39 oh no, looking into passage of golubria code... i had forgotten that it uses "ammo" as its duration counter 23:42:09 which also means that it expires based on actions rather than time 23:47:09 i guess it could theoretically be rescaled and changed to be based on time_taken while still using the ammo implementation, or something 23:48:57 I wish Ctrl+F could find items equipped by your followers. 23:49:25 Now I worry about using Yred's ability before I check the corpses..or I feel like I should check/kill all my zombies and skellies before I leave a level.