00:56:10 -!- amalloy is now known as amalloy_ 01:37:55 Unstable branch on crawl.develz.org updated to: 0.25-a0-166-g16436c4 (34) 02:08:37 Windows builds of master branch on crawl.develz.org updated to: 0.25-a0-166-g16436c4 02:58:54 Monster database of master branch on crawl.develz.org updated to: 0.24-a0-443-g80245de 03:11:52 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-166-g16436c4 (34) 03:30:52 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-2777-g6b872f2bae 05:49:08 DiscoBob (L19 DsFi) ASSERT(!invalid_monster(&mons)) in 'mon-death.cc' at line 1716 failed. (D (Sprint)) 08:06:30 -!- Tiobot is now known as Guest58365 08:56:28 !crashlog DiscoBob 08:56:29 No milestones for DiscoBob (crash). 08:56:33 !crashlog DiscoBob sprint 08:56:35 1. DiscoBob, XL19 DsFi, T:2967 (milestone): http://crawl.akrasiac.org/rawdata/DiscoBob/crash-DiscoBob-20191123-104906.txt 09:05:07 Processed 4068000 lines; 4.0GB from 128 logfiles remaining. 56 lines/s (25.5KB/s), ETA 1 day, 19:57:03 09:05:10 so much for speeding up 09:17:00 though I guess on the other hand it is still pretty close to on track for the original 2.5 day eta 09:51:34 ebering: not sure if you saw it but I half-seriously suggested turning cflame into a wand (with its current targeted behavior, even with renewal); having played a bit with the positional version I'm actually getting more into the wand idea 09:51:51 I did not see it 09:52:26 I think the positional version is reasonable powerwise but interface wise I'm not finding it to feel great (which is v subjective of course) 09:53:29 but limited charges solves a lot of the problems with the current spell in an entirely different way 09:55:52 another variant I had after playing with this version would be to instead give the player a flame trail that deposits clouds, length(/duration) scales with power, and doesn't spread if there's something in the square. At length/dur 1, that would have effectively the same behavior, but *might* have a different feel 09:55:57 *variant idea 09:58:08 the wand one does step on toes of a couple other wands, and I'm not sure I would really want to add a wand without removing one at this point 09:58:47 While that addresses the balance problems with indefinite usage, I'm also not a fan of its old targeting behavior 09:59:35 I think the implementation might be the issue. instead of the embers if it gives you a 1-turn status effect that tries to place a flame cloud where you were the casting turn 09:59:40 and succeeds if no monster 10:14:48 what's wrong with the old targeting behavior? 10:15:28 select one of up to 48 cells with perfect precision 10:15:42 but in a completely un-automatable way for its various use cases 10:17:38 well, I'm not sure I agree that making absolutely everything that uses targeting automatable is really the be all :-) 10:18:06 I agree with the general goal of simplifying a lot of it, and providing more things along those lines that don't have weird targeters, or can be done in a one-ish button way 10:24:10 actually, if fcloud had range 1 it would be fairly automatable 10:24:29 *cflam 10:24:29 e 10:25:11 here's another variant: produce 1-3 flame clouds depending on power, whose positions are randomly chosen within one tile of you and oriented towards a randomly chosen monster (up to 1 off from the straightest path) 10:44:01 hmm foxfires are seeking through transparent walls...I guess that might be int-related? 10:44:37 yes that might be int related 10:45:01 cbls do the same thing 11:01:17 Hello where is the code for sprint in github? Is it relatively easy to try to write your own sprint? 11:02:08 I have this sprint idea I want to try 11:06:59 sprints are kinda normal games with fixed (hard) vaults, I think 11:08:18 so most of the "code" for a sprint is in its vault definitions 11:09:13 geekosaur: but some sprints have spawns and weird stuff 11:09:20 which vaults can do 11:09:26 Okay got it, thank you 11:10:47 you can have arbitrary lua callbacks in vaults, which can do a bunhc of things. but you'll note for example that meatsprint is fairly glitchy because of bugs in how vault overrides for monster names work 11:10:47 and weapon names/stats 11:11:31 p0a: dat/des/sprints/* defines sprints - as long as you want to make a one-level sprint I think you can do everything in there and not touch C++. There some stuff about vault descriptions in the doc subdirectory of an install 11:12:06 And yeah, monster renaming in vaults isn't meant to do what meatsprint does with it (which isn't done with lua AFAIK, just the normal facility to rename a monster) 11:12:30 right, that was just a run-on sentence, not saying lua was involved 11:12:53 p0a: https://github.com/crawl/crawl/tree/master/crawl-ref/source/dat/des/sprint for the maps 11:12:54 vault definitions can do lots of weird things… but can uncover crawl bugs if you get too tricky 11:13:39 if you're doing something along the lines of existing sprints, you can do a lot in just the map + lua 11:13:41 (or just be slow, like some desolation of salt maps) 11:14:01 I want to do something similar 11:14:06 I want to try a pacific sprint 11:14:13 there's some custom stuff in the c++ codebase but it's pretty map-nonspecific, for example piety is scaled differently 11:14:16 ah sounds interesting 11:14:17 where you can only move (tele/blink/etc) 11:14:29 yeah let's see... heh 11:15:20 The way all the level is generated at once can have some amusing consequences like https://crawl.develz.org/mantis/view.php?id=12033 11:15:34 des syntax docs: https://github.com/crawl/crawl/tree/master/crawl-ref/docs/develop/levels 11:16:08 I've been trying to persuade Clare to read that stuff so I don't have to 11:21:30 ebering: here's yet another try for cflame. Fully targeted, range 1 (so automatable by dropping a cloud along a path to a monster); number of clouds allowed from the spell at one time is between 1 and some n, scaled by spellpower; newer casts fizzle with too many clouds, no renewing 11:21:30 maybe n=3 11:22:39 alright, thanks. I'll give it a shot! 11:23:44 Stable (0.23) branch on underhound.eu updated to: 0.23.1-91-gf373564dc4 12:12:31 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-166-g16436c4 (34) 12:34:15 alexjurkiewicz does footv automatically handle .xv? I think you implied this but I just want to double check that's what you meant 12:40:23 hm, so I optimistically said the other day about cao "only has 33G free; not crisis point yet, it should take a while to get through that". But now it has 25G free 12:40:27 in the chroot 13:11:27 does anyone know if I can delete /chroot/crawl-oldtrunk/ on cao 13:11:43 not entirely sure how anyone here *would* know that, but I feel compelled to ask 13:12:12 advil: yeah, possible r.ax would know obviously, but no idea 13:12:22 the versions are no later than 0.12a 13:12:47 did cao do builds at one point? 13:13:02 I think it's just trunk versions it was serving a long time ago? 13:13:59 it *looks* to me like at some point r.ax cleaned some old trunk builds but didn't delete them, that's exactly the kind of paranoia I would have too 13:16:06 -!- kate-- is now known as kate- 13:16:36 is there any way there'd be current games that depend on those versions? 13:18:04 I'd expect they'd have been unplayable with their actual game code moved elsewhere? 13:18:21 I don't know enough about dgl to know how to know for sure, but I *think* that everything playable is in /chroot/crawl-master/ 13:23:49 yeah geekosaur, that's what I assume, and unless there is some creative symmlinking, I think they do have to be in the main dir that advil gave 13:23:58 oh, well, with chroot symmlinks wouldn't matter 13:24:42 also symlink should have one m 13:24:42 I blame my constant typing of the word 'commit' 13:25:12 rename it, see if anything breaks, kill it after a week 13:25:33 heh good idea 15:02:08 Dharzhak (L17 MiFi) ASSERT(mgrd(mons->pos()) == s) in 'show.cc' at line 275 failed. (Snake:4) 15:04:48 !crashlog 15:05:07 21373. Dharzhak, XL17 MiFi, T:38492 (milestone): http://crawl.berotato.org/crawl/morgue/Dharzhak/crash-Dharzhak-20191123-200207.txt 15:08:15 -!- MakMorn_ is now known as MakMorn 15:44:32 advil: it seems like there's a bad interaction with removing spells/making them player only and the ghost permastore 15:45:02 In that if a spell has lost a monster version the game will happily load the ghost with the non-monster spell 15:45:10 then crash when fighting the ghost 15:45:34 if you know the right place and can push a fix to the positional-magic branch I'd be much obliged; if it must be done in save-compat code I'll get to it eventually 15:58:32 kate-: have you had a chance to look at the positional magic stuff? 15:58:39 I'd appreciate your input 16:32:31 advil: cpo works so i assume so 16:35:42 ebering: i haven't played with it any but i've tried to follow the branch a bit, yeah 16:37:26 i'm curious how well the targeting works for all the player-centred aoe stuff, is there any targeter? having to count tiles to check what's in range seems like it'd be fiddly if not, especially with things like veh affecting range in some cases 16:39:00 a very minor thing i noticed was sting is now the only dual-school starting spell, i think? which feels a little weird maybe 16:39:51 oh nevermind i'm completely wrong on that 16:40:10 thought shock/flame tongue were just the element for a sec 16:43:58 but yeah i'm definitely curious to try it out, it looks like a good direction, just curious how clear the targeting ends up being (eg for airstrike too with the power modifier, wonder if that could be indicated with colours in the targeter or something) 17:14:02 ebering: did you increment the minor tag at all in the branch? 17:15:57 and/or is there save compat for players that you can point me to? 17:17:14 also, do you have a save file for the ghost issue? 17:20:55 !source unmarshallSpells 17:20:56 1/1. https://github.com/crawl/crawl/blob/master/crawl-ref/source/tags.cc#L6997 17:21:16 I think what is probably needed is some save compat in that function, but I'd have to look at the specific case 17:21:32 (if this shows up for ghosts it may also show up for e.g. panlords) 17:25:12 in some cases if you don't remove the spell yet but just wrap it in #ifdef TAG_MAJOR_VERSION checks, that may be easier 17:25:49 e.g. I was just looking at what kate did for inner flame (since I don't remember these crashes there) and I think that was the solution 17:31:15 %git 78a4890d2848 17:31:15 07Chris Campbell02 * 0.17-a0-1550-g78a4890: Remove Inner Flame as a monster spell 10(4 years, 4 months ago, 2 files, 8+ 15-) 13https://github.com/crawl/crawl/commit/78a4890d2848 17:33:37 in that removal, when handle_mon_spell returns false instead of handling the spell, the monster just does something else 17:34:47 advil: I haven't incremeted the tag minor at all; there are definitely some other things that will warrant it 17:34:50 -!- amalloy_ is now known as amalloy 17:34:58 changing handle mon spell looks like the correct solution here though 17:35:00 thanks! 17:35:54 !tell kate- there's not a targeter for the AoE but there is a range check and I'm planning to add a proper hitfunc to check the exact target pattern 17:35:55 ebering: OK, I'll let kate- know. 17:38:10 advil: http://crawl.kelbi.org/saves/ebering-crawl-positional-magic-191123-2237.tar.bz2 17:38:21 the ghosts in the 5 ghost vaault in lair have non-monster spells 17:41:41 artmatt (L18 DrSu) ASSERT(!invalid_monster_type(type)) in 'monster.cc' at line 3264 failed. (Vaults:1) 17:43:27 !crashlog 17:43:28 21374. artmatt, XL18 DrSu, T:57314 (milestone): http://crawl.akrasiac.org/rawdata/artmatt/crash-artmatt-20191123-224132.txt 17:44:01 ah 17:44:02 yeah 17:44:07 gotta fineff that dismissal 17:44:20 force lance into guardian golem dismiss