00:15:12 -!- Enne has joined ##crawl-dev 00:15:13 -!- Enne has quit [Client Quit] 00:20:33 is it possible to have several miscast effects happen at the same time? 00:20:43 or it's always just one or none. 00:21:43 it's not possible to have several * happen at the same time 00:21:50 this is a single-threaded program 00:23:34 go figure 00:29:50 (in other words, I don't understand you) 00:30:28 suppose a player miscasts a necromancy spell 00:30:40 can then she get both "flickering shadows" and torment 00:32:40 reading _necromancy_miscast should answer that question 00:32:45 do you know how switches work in C 00:32:57 yeah I've heard something about them 01:32:11 -!- by has joined ##crawl-dev 01:48:50 by: hello! 01:51:23 hey rob 01:54:41 Condensation shield oddities (https://crawl.develz.org/mantis/view.php?id=1632) by OG17 02:05:26 -!- Poor_Yurik2 has joined ##crawl-dev 02:09:59 -!- Poor_Yurik2 has quit [Ping timeout: 260 seconds] 02:11:27 hi 02:43:38 by: I figured out the xom-induced assertion failure 02:43:57 < DChimera> Duvessa, [...] (berserk, resting) 02:43:59 < Pseudonut> zzzzz grrrrrr zzzzzzzz grrrrrrrr 02:45:53 That shouldn't happen 02:46:02 Pretty funny anyway :) 02:46:09 Especially Pseudonut's take on it :P 02:46:12 V. annoying, considering I deliberately wake her up. 02:48:05 xale: good 02:48:41 by: I think that should fix it: http://sprunge.us/BeCZ 02:48:54 there are a couple of places in MiscastEffects::_necromancy() where if draining fails it tries a different miscast 02:49:04 the problem is, avoid_lethal() which it uses to check whether miscast should succeed can start another do_miscast() 02:49:12 so what happens is: you avoid draining, and get something else instead 02:49:17 then it returns back to where it was trying to drain you and you get yet another miscast 02:49:23 but you already got one message (and one miscast effect), so it all crashes. 02:49:58 Divine wrath summons should work like royal jelly spawns (https://crawl.develz.org/mantis/view.php?id=1633) by OG17 02:53:40 03by * ra3f9f995d247 10/crawl-ref/source/format.cc: Don't cgotoxy a line too far down (#1591, #1604). 03:00:51 -!- ogaz has quit [Remote host closed the connection] 03:03:13 03by * r486b011a33bc 10/crawl-ref/source/spl-mis.cc: Avoid double necromancy miscasts. Fixes #1544. (xale) 03:03:57 hey so does crawl force power-of-two sized textures? 03:03:59 in Tiles? 03:04:06 or do I have to pad in my texture loader? 03:19:06 -!- ogaz has joined ##crawl-dev 04:23:10 -!- ogaz has quit [Remote host closed the connection] 04:34:32 ixtli: I'm not entirely sure, best bet woul dbe to askn Enne. 04:34:52 yeah i found that it passes an argument to the texture loading functions 04:36:59 -!- ortoslon has quit [Quit: bye] 04:56:35 -!- Zaba has quit [Ping timeout: 265 seconds] 05:02:29 -!- Zaba has joined ##crawl-dev 05:08:01 -!- syllogism has joined ##crawl-dev 05:52:58 by: do you have a moment? 06:00:49 c-69-250-105-237.hsd1.md.comcast.net: This is fucking stupid, you faggots. 06:00:59 commented on "Play-testing: Hit Me With Your Dowsing Rod" 06:01:39 trashed. 06:02:40 confirmation of statistics :D http://crawl.develz.org/wordpress/2009-dcss-survey-results/comment-page-1#comment-296 06:53:06 -!- Elvashi has joined ##crawl-dev 06:55:15 -!- Textmode has quit [Ping timeout: 260 seconds] 06:55:23 -!- Elvashi is now known as Textmode 07:22:03 -!- Textmode has quit [Ping timeout: 258 seconds] 07:36:37 USE_NEW_COMBAT_STATS is disabled in release builds, right? 07:37:28 actually, not. it is enabled, right? :-) 07:43:32 -!- Textmode has joined ##crawl-dev 07:44:54 it's enabled; I wonder if anybody uses these defines? 07:45:10 what does it do 07:45:39 dex/str weighting 07:50:11 there's also USE_SEMI_CONTROLLED_BLINK that allows disabling semi-controlled blink (for blink + tele control) 07:52:59 !seen greensnark 07:52:59 I last saw greensnark at Sat May 29 07:46:09 2010 UTC (5h 6m 50s ago) saying Especially Pseudonuts take on it :P on ##crawl-dev. 07:53:12 you alright, greensnarking? 07:54:31 Napkin: do you think it would be difficult to add a variant of the summary pages to mantis that considers only "Bug Report" issues? 07:56:19 hmm.. what would you like to know exactly, rob? 07:59:06 there's lots of basically useful information there, but I'm really just interested in bug reports, not in the ever growing number of feature requests 07:59:34 e.g., it would be nice to see how the number of open bug reports is developing 08:00:19 it doesn't really matter if the number of open feature requests keeps growing, but we should be keeping the number of bugs in check 08:01:03 -!- casmith789 has quit [Remote host closed the connection] 08:04:12 ok, I was just checking 08:04:21 I believe there is something like that already 08:04:57 go to "View Issues" and select "Source Cleanup" as filter 08:05:18 then click on the "Graph" link 08:05:48 "Summary by Status and Category", "Year to Date", "Show as Graph" and click "Show Graph" 08:06:02 ah, thanks, will try 08:06:19 it seems, those graphs are generated with the data you filtered for 08:06:27 definitely not perfect 08:06:40 -!- Mu_ has joined ##crawl-dev 08:06:44 and of course, we can have it summary changed in any way we like 08:06:56 it's just I'm not so geek with doing database querries 08:08:46 those graphs are great, thanks! 08:09:03 great :) 08:19:47 03by * r3f41f4974540 10/crawl-ref/source/stuff.cc: If unicode doesn't work, abort instead of ASCII fall-back. 08:23:00 03by * r62434dd8d55a 10/crawl-ref/source/stuff.cc: Fix non-unicode abort compile and message error. 08:54:57 -!- Enne has joined ##crawl-dev 09:03:03 -!- eith has joined ##crawl-dev 09:03:10 -!- Textmode has quit [Quit: Over the ages, man has been certain of a great many things. some of these things were even true.] 09:03:31 -!- by has quit [Disconnected by services] 09:03:32 -!- by_ has joined ##crawl-dev 09:36:02 -!- by_ is now known as by 09:54:19 Napkin: Doing great, thanks! 09:54:47 Napkin: Finally took the Hetzner server, but I haven't had the time to move everything onto it yet 09:54:57 Might move FooTV and friends now, hmm 09:55:11 The server is fast <3 10:09:09 to hit bonuses for various transformations are kinda weird 10:11:40 lich-form gets the same bonus as a dragon, spider or ice beast form. 10:18:16 :)) 10:18:33 Just wanted to make sure, that you didn't took a train last night - happy to read you :) 10:18:45 alright @ hetzner server :D 10:56:24 by: around? 10:56:30 half 10:57:00 I want to fix some broken death messages 10:57:24 when a monster kills itself before it kills the player, this is recorded as "killed by a program bug" 10:58:24 yes 10:58:28 but there doesn't seem to be any way to pass the monster's name to ouch() 10:58:35 and the env.mons entry is cleared 10:59:03 I'd add another optional argument to ouch() to handle that, but there probably is a better way? 10:59:37 it depends; for the mummy death curse, env.mons is still usable, but it'd probably be better to not access env.mons at all 11:00:31 for the cast I'm looking at (chain lightning) it isn't usable 11:00:37 s/cast/case/ 11:00:37 adding a parameter to ouch() sounds fine 11:01:09 alright, I'll try to do that then 11:02:24 -!- ais523 has joined ##crawl-dev 11:04:12 there's also the (theoretical?) possibility that death_source is not valid anymore, e.g. chain-lightning caster dies, chain-lightning kills other monster, that monster's death causes another monster to be summoned and placed in the free slot in env.mons 11:04:42 maybe it would be better if we only cleared the env.mons entries once per turn 11:05:03 that would also solve chain lightning, but not IOOD 11:07:03 no valid, as in, looks perfectly valid, but isn't? 11:08:59 why wouldn't it solve IOOD? 11:09:57 gosh, IOOD is a monster! 11:11:56 by: also, something I've noticed is that beam.source_name isn't used much if at all 11:12:21 by: it looks like a good place to put caster's name, but nothing seems to check it 11:16:54 and it looks like iood also uses beams to hit things, so it would help with that too. 11:18:15 so that's filled everywhere, then bolt::internal_ouch could pass it to ouch somehow 11:25:09 (if that's filled everywhere) 11:27:20 there are more cases like this 11:27:36 like, poison blaming handles only killer/kaux, not ckiller or ikiller 11:28:55 maybe it's better to make a list of these before trying to fix anything. 11:29:08 chain lightning, fire storm, fireball and such may last only one turn, but we have a number of effects that last longer 11:29:49 like, poison, confusion, clouds, IOOD, summons 11:30:25 and the current handling is inconsistent and full of duplicated code 11:30:54 so deaths by poison get broken messages too if you kill the poisoner? 11:30:54 like, summons store a lot of data, poison only two fields, clouds and confusion nothing 11:31:23 no, they are always ok but the data stored is only partial 11:46:58 -!- Poor_Yurik has quit [] 11:56:48 though I don't think #1628 should be fixed, it's hilarious. 12:03:00 -!- Zaba has quit [Ping timeout: 272 seconds] 12:05:21 -!- MarvinPA has joined ##crawl-dev 12:10:50 -!- Zaba has joined ##crawl-dev 12:18:41 okay, so.. I think for chain lightning and ood caster can be stored in source_name of the bolt/beam 12:18:59 and something like source_name can be added to clouds 12:20:47 that doesn't help with death curses at all, but I think they're unrelated? 12:22:32 Why not just delay the cleanup of monsters until some point later? 12:22:48 the problem is, if we store just the name for some ouch() calls and monster index for other some information will probably be lost 12:23:10 yeah, probably would be better to delay the clean-up but for how long exactly? 12:24:13 what happens if, say, something summons a monster that creates a cloud that will kill the player after several turns? 12:24:21 Refcount monsters? 12:24:23 perhaps monsters should be refcounted! 12:24:25 what! 12:24:34 :) 12:25:26 then we can remove the source_name and such and stop worrying 12:26:11 by, what do you think? 12:26:50 back 12:31:19 keeping dead monsters around does sound good 12:34:32 beams and clouds and orbs and summons can retain whatever created them 12:35:03 until they dissipate or die or whatever. 12:35:27 what data will they need of a dead monster? 12:35:40 source_name :-) 12:35:55 but maybe also source_name of the source 12:36:00 Well, more than that, too. 12:36:17 -!- Siber has quit [Read error: Connection reset by peer] 12:36:19 Some monsters have special name behavior. Mimics? 12:37:12 for mimics you could just store something like "a mimic (disguised as an old rag)" 12:37:30 but being able to follow a chain of summons would be useful 12:37:41 yes 12:37:49 "Killed by a cloud summoned by foo which was summoned by bar which was..." 12:37:54 yup 12:38:03 yes, those shouldn't break for dead monsters 12:39:33 refcounting should be doable 12:40:25 the only thing I'm kind of worried about is beams cleaning up after themselves 12:41:58 -!- Poor_Yurik has joined ##crawl-dev 12:43:43 what do you think could be a problem? 12:43:54 As long as the refcounting gets handled sanely during assignment and destruction, it seems like it should just work implicitly? 12:44:00 should be fine, really, yes 12:44:02 from what i've seen beams are created on stack 12:44:09 so they can clean up in dtor 12:44:09 03by * rf4de9dacf4c3 10/crawl-ref/source/beam.h: Update comment. 12:44:14 yup 12:44:19 03by * r3a0d0bd6bb76 10/crawl-ref/source/player.h: Remove unimplemented player::set_arena_los from header. 12:46:44 the only thing I'm worried about is how to test all this 12:47:21 just setting up things to test chain lightning was a bit complicated 12:47:35 xale: Use the arena with respawning monsters 12:47:37 had to make an immobile nikola with one hit point! 12:47:46 "Just add them to our comprehensive suite of unit tests." 12:48:01 Heh 12:48:43 This would be so much easier in a language with gc :P 12:49:38 * bhaak lols 12:49:59 I imagine writing the monster data out to disk would get more difficult, however 12:50:24 I believe boost includes refcounting pointers... But, but... 12:50:52 !coffee Zaba 12:50:52 * Henzell hands Zaba a cup of cappuccino, brewed by Ijyb. 12:52:04 Serialize monster pointers to monster index when writing out, and then fix-up on read? 12:52:50 Why does it complicate saving monster data? 12:53:15 I understood it as monsters having a refcount to keep dead monsters being flushed from the monsters vector 12:54:36 -!- ais523 is now known as zzo523 12:55:25 -!- Poor_Yurik has quit [] 12:56:26 -!- Siber has joined ##crawl-dev 12:56:32 -!- zzo523 is now known as ais523 13:04:25 -!- ogaz has joined ##crawl-dev 13:07:54 by: what actually happened in #1628? killed by a summon from a death curse? 13:20:56 -!- pointless_ has joined ##crawl-dev 13:22:11 -!- eith has quit [Ping timeout: 245 seconds] 13:25:31 -!- pointless_ has quit [Ping timeout: 245 seconds] 13:27:01 -!- pointless_ has joined ##crawl-dev 14:02:24 my comment before was re languages with garbage collection 14:03:02 xale: #1628 is just the normal pain death curse for a guardian mummy 14:06:30 greensnark: wasn't the stash tracker ignoring useless items fixed? 14:23:31 -!- ogaz has quit [Ping timeout: 240 seconds] 14:25:06 by: I can't reproduce #1627 14:25:23 I get almost normal death message, except that it's "a orb". 14:29:34 just started a fighter, &M boris, let the orb hit me, said 'y' to the death prompt, and got the "Blown up themselves" message again 14:30:56 oh, I thought you actually died to your own orb 14:31:08 no, sorry if the report was unclear 14:32:27 okay, I get the same message now. 14:36:07 -!- MarvinPA has left ##crawl-dev 14:36:07 -!- MarvinPA has joined ##crawl-dev 14:51:58 -!- casmith789 has joined ##crawl-dev 14:54:54 yuck, the "electricity arcs through the water" effect is performed through bolt::range_funcs 15:00:34 -!- MarvinPA has quit [Ping timeout: 252 seconds] 15:03:57 -!- MarvinPA has joined ##crawl-dev 15:05:31 KiloByte: instead of the range func _electricity_water_explosion, would just calling beam.finish_beam() from _charged_hit_victim not work? 15:15:37 -!- ais523 is now known as undecidable 15:17:22 -!- undecidable is now known as ais523 15:17:45 -!- Adeon is now known as MaryPoppins 15:19:54 -!- ogaz has joined ##crawl-dev 15:42:06 -!- daftfad has joined ##crawl-dev 15:50:27 -!- MarvinPA has quit [Ping timeout: 252 seconds] 15:53:41 03dolorous * r0b3a339f21c3 10/crawl-ref/source/dat/database/monspeak.txt: Fix grammar. 16:05:31 Arena crash w/multiple copies of one unique (https://crawl.develz.org/mantis/view.php?id=1634) by dolorous 16:06:39 do we have an integer square root function somewhere? (specifically, I want coord_def::abs() -> radius of circle it just fits in) 16:06:56 i.e., 65 -> 8, 66 -> 9 16:07:54 is (int)(ceil(sqrt(x-1)) sure to work here, or would I have to worry about rounding? 16:08:24 -!- ogaz_ has joined ##crawl-dev 16:08:35 -!- ogaz has quit [Disconnected by services] 16:08:40 -!- ogaz_ is now known as ogaz 16:20:36 Arena oddities with saved player settings (https://crawl.develz.org/mantis/view.php?id=1635) by dolorous 16:29:39 -!- Luca__ has quit [Remote host closed the connection] 16:29:56 -!- Luca__ has joined ##crawl-dev 16:48:55 -!- Luca__ has quit [Remote host closed the connection] 16:49:12 -!- Luca__ has joined ##crawl-dev 17:03:24 03by * rbdb4c33e37d7 10/crawl-ref/source/beam.cc: Fix tracking of bolt::bounces. 17:03:25 03by * rfb351136f5d3 10/crawl-ref/source/item_use.cc: Simplify electricity beam in water effect. 17:03:26 03by * r97c5eea3f298 10/crawl-ref/source/ (beam.cc beam.h): Euclidean range for beams. 17:03:29 03by * rac39ced3f196 10/crawl-ref/source/ (directn.cc mon-util.cc spl-util.cc view.cc): Adapt the interface to euclidean range. 17:22:30 -!- by has quit [Read error: Operation timed out] 17:28:12 -!- Iainuki has quit [Read error: Connection reset by peer] 17:28:53 -!- Iainuki has joined ##crawl-dev 17:29:01 -!- Iainuki has quit [Changing host] 17:29:01 -!- Iainuki has joined ##crawl-dev 17:34:02 -!- Nexos has joined ##crawl-dev 17:35:49 Hi. Is it possible on CAO for last log files to have sthg like symlinks: p.ex. rawdata/Nexos/morgue-Nexos.txt -> rawdata/Nexos/morgue-Nexos-xxxxx.txt 17:35:52 ? 17:36:04 -!- ais523 has quit [Remote host closed the connection] 17:49:34 yes 17:49:41 but I'm not going to do it 17:50:09 I've been burned too many times by any kind of change to the data storage system having unpredictable and fragile consequences 17:50:27 and we don't have any way to test changes 18:16:28 How can a symlink (or hardlink) have an effect on the storage system? The devil may lie in the details, but creating/altering a sym/hardlink when you create a file is not rocket science, is it? 18:20:58 you'd be amazed at how fragile the servers are 18:47:38 03dolorous * rb4ea0ed3942e 10/crawl-ref/source/delay.cc: Comment fixes. 18:47:39 03dolorous * r1c60ac9d3a4a 10/crawl-ref/source/spells1.cc: Clarify Ely's pacification messages. 19:03:04 03dolorous * r4aeb09a263be 10/crawl-ref/source/abl-show.cc: Change one message for Zin's Imprison. 19:06:20 03dolorous * r147c7959a208 10/crawl-ref/source/abl-show.cc: Simplify. 19:07:05 -!- casmith789 has quit [Remote host closed the connection] 19:09:21 03dolorous * r0a2a4a799219 10/crawl-ref/source/spells1.cc: Simplify further. 19:10:26 03dolorous * rf7a0984b1e0f 10/crawl-ref/source/spells1.cc: Add formatting fix. 19:13:49 -!- syllogism has quit [] 19:36:43 -!- Nexos has quit [Ping timeout: 260 seconds] 19:36:45 Who the hell are these wankers commenting on Denzi's excellence? 19:36:45 03dolorous * rcb14027732bd 10/crawl-ref/source/ (player.cc spells1.cc): Per Mantis 1447, scale the power of Ely's Divine Vigour a bit better. 19:36:49 Enne, rargh time! 19:38:42 03dolorous * r51bcbead6320 10/crawl-ref/docs/changelog.txt: Update the 0.7 changelog again. 19:41:07 well, I kind of agree with the hand comment 19:41:18 it's hard to tell which is which. 19:41:38 Denzi's art is impeachable! 19:41:48 yeah, it's very good. 19:43:52 mantis seem to remember the last query entered into it, no matter who did it? 19:46:02 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 20:29:23 due: here in the USA, "impeach" means something /entirely/ different 20:29:33 The word we use is "impeccable" 20:30:35 Er, unimpeachable is what I meant. 20:43:14 -!- ogaz has quit [Remote host closed the connection] 21:11:06 -!- straydusk has joined ##crawl-dev 21:14:49 -!- ogaz has joined ##crawl-dev 21:20:15 -!- xale has quit [Quit: xale] 21:22:46 -!- Textmode has joined ##crawl-dev 21:24:05 -!- straydusk has quit [Quit: Page closed] 21:52:17 Disabled characters should lose EV (https://crawl.develz.org/mantis/view.php?id=1636) by OG17 22:00:22 -!- Luca__ has quit [Remote host closed the connection] 22:00:37 -!- Luca__ has joined ##crawl-dev 22:07:22 Players don't have a petrify status (https://crawl.develz.org/mantis/view.php?id=1637) by OG17 22:36:55 -!- straydusk has joined ##crawl-dev 22:46:41 Sorry to bother, but would someone with creation permission on the Code and Docs Wiki mind creating the page dcss:feedback:god:makhleb? I have an entry for it saved as a template revision elsewhere and would like to place it there. 22:49:00 straydusk: File a ticket for permission? :) 22:51:44 due: Okay, thanks 22:57:05 -!- straydusk has left ##crawl-dev 23:08:07 -!- Enne has quit [Quit: zzz] 23:15:10 -!- jld has quit [Ping timeout: 240 seconds] 23:44:10 -!- jld has joined ##crawl-dev