01:31:27 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-5140-g5775ae71e1 01:31:28 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-5140-g5775ae71e1 09:11:40 jorsephus (L1 HuIE) ASSERT(!you.turn_is_over) in 'state.cc' at line 199 failed. (D:1) 10:20:41 <04dracoomega> // Monster types that you can't gain experience from cannot // fight back, so don't bother having them do so. If you // worship Fedhas, create a ring of friendly plants, and try // to break out of the ring by killing a plant, you'll get // a warning prompt and penance only once. Without the // hostility check, the plant will remain friendly until it // dies, and you'll get a 10:20:42 warning prompt and penance once // *per hit*. This may not be the best way to address the // issue, though. -cao I feel this comment block may be a bit out of date 10:23:19 <06pleasingfungus> hmm 10:23:41 <06pleasingfungus> also, i don’t remember who cao is/was 11:51:38 maybe it's joshua, or, joshua joshua 11:52:03 iykwim 11:57:14 <07zureal> today I learned, shmup hangs out here. 11:57:59 off and on, over the decade, lol. nothing stays the same wif me ~ 12:10:33 <04dracoomega> So, I am wanting to simplify _mons_attacks_outside_los since not only is there mild bugs, but also over time more and more things keep being added into here as it is noticed that they've been overlooked, and I am wondering if the entire thing can just be replaced at this point with "If monster is permanently friendly, it can't attack outside of LoS". Are there any things remaining where we would want that to not be true? 12:11:01 <04dracoomega> I've been having trouble coming up with any that exist 12:11:20 <04dracoomega> (But it's possible I'm forgetting something, of course) 12:37:17 <04dracoomega> This function makes my eyebrows raise more, the more I look at it C++ bool monster_can_hit_monster(monster* mons, const monster* targ) { if (!summon_can_attack(mons, targ)) return false; if (!targ->submerged() || mons->has_damage_type(DVORP_TENTACLE)) return true; if (env.grid(targ->pos()) != DNGN_SHALLOW_WATER) return false; const item_def *weapon = mons->weapon(); return weapon && 12:37:18 item_attack_skill(*weapon) == SK_POLEARMS; } 12:38:17 <04dracoomega> First of all, I've never even seen has_damage_type(DVORP_TENTACLE) before, but also... if I understand this right, this is to specifically let monsters with polearms (and only polearms) attack enemies that are submerged in deep water or lava 12:38:40 reaching? 12:38:53 <06regret-⸸nde※> which are... currently just kraken tentacles?... which one can encounter... when using scrolls of summoning in shoals?... 12:38:54 <04dracoomega> This isn't about reaching itself. This is for adjacent targets. 12:44:25 <04dracoomega> I kind of feel like none of this needs to exist 12:45:03 <04dracoomega> Submerging barely exists (krakens could use some love, independently of this, imo). I don't think we need fine-grained checks for being able to attack it. Would it be a problem if monsters just couldn't? 12:45:49 <04dracoomega> Oops, I stand corrected. Monsters need a polearm to attack it in shallow water. Only tentacles can attack it in deep water or lava 12:58:06 <06pleasingfungus> lmao DVORP_TENTACLE 14:49:52 <04dracoomega> So, my first attempt at making retreating followers travel up stairs with you has not resulted in them being on the other side of the stairs with me. Even worse, it seems to have made them cease to exist on the other side of the stairs, too >.> 14:50:30 <04dracoomega> Currently possible to use this to just delete charmed monsters from existence. I think??? 16:41:40 <04dracoomega> So, I want to make Beogh recall instant (since it's only ever recalling 3 things, and the current version can always get 2 of them at once anyway, so splitting that across turns feels needless) and that would leave Yred the only one with the progressive recall behavior. I can't help but wonder if even that is necessary now? Like, originally it was to prevent your overwhelmingly large army (possibly across many floors) from dropping 16:41:41 all at once, but zombies already decay if you leave a floor and otherwise you have just the one bound soul. Would it be that big a deal if you could grab all your still-living zombies around you in one action instead of the way it works now? 16:41:59 <04dracoomega> In practice, they don't tend to be all that far from you anyway 16:42:05 <04dracoomega> Unless you just teleported, I guess 18:39:30 <06pleasingfungus> seems fine 18:39:48 <06pleasingfungus> both progressive and instant recall are cool in their own way 18:39:50 <06pleasingfungus> seems good to simplify 19:42:25 <06regret-⸸nde※> (also helps that zombie bodies won't be a yred ability cost anymore, so there's no incentive to keep the zombies away just to port them in to expend for bind soul) 20:35:24 <09gammafunk> > loaddb_1 | 2024-01-31 03:30:15,876 [INFO] Bad log line in logfiles/cko-logfile-0.31 at offset 4247937: 'rial_bayou_shore_a:seed=577291600169543377:tmsg=slain by a komodo dragon:vmsg=mangled by a komodo dragon\n' > loaddb_1 | 2024-01-31 03:30:15,876 [INFO] Bad log line in milestones/cko-milestones-0.31 at offset 31223003: 'scrollsused=0:potionsused=0:time=20240025041348S:type=uniq:milestone=killed Ijyb.\n' 20:36:30 <09gammafunk> not sure how, but we have partial logfile lines and partial milestone lines; just these two apparently 20:37:21 <09gammafunk> I think the newline at the end is just the line's typical newline, but it looks like something wasn't fully put into the xlog files here 20:43:01 Hi5aMiFi (L27 FoDe) ERROR in 'mon-util.cc' at line 696: bogus mc (no monster data): invalid monster_type 1000 (1000) (Slime:5) 21:31:02 what language is that written in? C has a buffer limit in `fgets` (please don't use `gets`, it's even worse) 21:33:31 most more modern languages do better 21:34:36 <05icoson> It’s python 21:34:53 huh. python should not show that behavior 21:35:47 <05icoson> I doubt it’s a bug reading the logfile