00:02:36 -!- MarvinPA has quit [Ping timeout: 265 seconds] 00:06:50 -!- MarvinPA has joined ##crawl-dev 00:18:12 -!- MarvinPA_ has joined ##crawl-dev 00:18:33 -!- MarvinPA has quit [Ping timeout: 265 seconds] 00:24:51 -!- joosa has quit [Ping timeout: 276 seconds] 00:25:19 -!- Hehfiel has quit [Ping timeout: 260 seconds] 00:44:04 -!- psyshvl has quit [Quit: stuff] 00:47:14 -!- Twinge has quit [Ping timeout: 250 seconds] 01:25:54 -!- Twinge has joined ##crawl-dev 01:31:09 is there a reason enchanters start with darts and a short sword instead of a dagger and stabbing skill, despite them being heavily stabbing-oriented? 01:32:45 Darts+corona. 01:55:31 -!- ortoslon has quit [Quit: bye] 02:04:04 hm. 02:04:30 Not sure re: short sword. 02:05:17 due, I have an unfinished hive:2 map here 02:05:34 due, the problem is, I don't know where to put loot. And there should be loot on Hive:2. 02:06:07 sprunge and let me see? 02:06:40 http://sprunge.us/FdBH 02:07:21 Small hive is small. 02:07:50 Clearly you should have secret rooms on eithe rside of the queen-bee chamber. 02:08:02 This seems to me to be the logical location. 02:09:57 -!- galehar has joined ##crawl-dev 02:10:04 galehar!!! 02:10:16 galehar, I have pplaced two implementables with patches on the tracker. Would you be able to look over these for me? 02:10:52 sure 02:11:33 due, yeah, that's the idea I've had, too, but I am not sure how to do them properly 02:12:04 galehar: They're both marked [Patch] and tiles. 02:12:25 yes, I have found them. 02:12:49 Crash when going upstairs in the Slime Pits... (https://crawl.develz.org/mantis/view.php?id=3108) by Twilight 02:13:02 due, the fact there are two chambers is kind of problematic... and how to hide them? Secret doors? Wax walls with no hint whatsoever that there is a room behind it? 02:15:28 due: I don't really know much how tiles work. I have only played with the interface so far, not the cells themselves. But I'm interested. I'll look into it. 02:15:48 due: for #3106, you've uploaded the .swp file 02:15:52 :S 02:15:56 OOPS. 02:16:27 galehar: 3106 almost drove me to tears. 02:17:03 There we go! 02:17:40 due: is #3106 a step toward drawing clouds behind monsters? 02:17:48 galehar: Pos..sibly? 02:17:56 I think it would definitely help. 02:18:05 ok 02:18:06 It would allow for a specific cloud field in the packed_cell struct. 02:18:19 Which could then be drawn first and then the monster on top of it. 02:18:20 So sure! 02:18:48 The main implementations I was thinking of was allowing for further aereas (ie, liquefaction, tornado effects) without having to make new flags (impossible seeing as we've reached the limit). 02:19:47 (The other on ebeing my secret plan to make Swamp less ugly.) 02:20:38 hrm 02:21:13 I like the word "instatiated". 02:21:17 Almost as much as "burgeoning". 02:26:05 you mean instantiated? 02:26:44 Yeah, that. 02:26:46 :D 02:30:06 lol, 3103 doesn't even compile 02:30:16 dbg-asrt.cc:40:11: error: missing binary operator before token "(" 02:30:29 this one is hilarious: 02:30:29 dbg-asrt.cc:40:5: warning: "define" is not defined 02:34:22 :( 02:34:32 Should'v ebeen defined, I think. 02:34:54 Yeah, if define -> if defined 02:35:02 Like I said, I couldn't even test it to compile it. 02:37:41 ok 02:45:03 appart from the typo and a missing include (SDL_syswm.h), the windows tiles assert patch works great. 02:45:20 :D! 02:45:24 good work doing that blind! 02:45:25 Yes I wasn't sure about the include. 02:45:42 Wow, I'm actually quite proud of myself :D 02:45:53 So is that all the change that needs to be made? I might apply the patch, tweak it, and then push... 02:46:19 Unless you wish to? 02:46:32 no, do it 02:46:38 :D 02:48:01 flex 2.5.4 doesn't support --nounput (https://crawl.develz.org/mantis/view.php?id=3109) by ahpla 02:48:05 I've pasted the diff with the fix on mantis 02:53:02 galehar: Hm, doesn't appear to be posted yet 02:53:09 Oh, there it is 02:53:39 :) 02:53:49 (I was refreshing the wrong bug) 02:54:54 galehar: Are both SDL includes required? 02:55:02 I suppose it shouldn't hurt too much to include both of them... 02:55:12 let me test 02:57:05 Windows development builds on CDO updated to: 0.8.0-a0-4217-g51afd0a 02:58:56 SDL.h isn't necessary 02:59:16 Excellent, thanks. 03:02:09 Hooray assert dialogs! 03:02:12 03due * r6f4ba073f75a 10/crawl-ref/source/dbg-asrt.cc: Generate a Windows Alert dialog on assertions (Tiles). 03:03:11 Re flex: BSD :( 03:15:58 wtf... flex 2.5.4 is from 1996 03:16:05 "lol". 03:16:08 Good old BSD! 03:16:56 good things shouldn't be changed! 03:17:03 cf, nethack? 03:17:06 I had to dig around to find that, changelogs were trimmed in 2001 03:17:31 I'll just remove the --nounput 03:17:35 Thats a looong time. 03:17:49 due: let's say, 3.4.3 wasn't the ideal version to stop development 03:17:51 There's supposed to be a #define you can use to prevent yyunput from being generated, but support for that also seems patchy 03:18:16 greensnark: I guess there might be other problems, it might be safer to use prebuilt yaccage in such cases 03:18:31 03greensnark * rdd76e9b60799 10/crawl-ref/source/util/Makefile: Don't use flex --nounput flag since support for it varies and it's used only to get rid of a compiler warning. 03:20:14 I don't think the level compiler has issues with very old flex and bison 03:20:47 Mac OS 10.4 had a bison 1.something version and that used to produce some strange complaints, but the built compiler worked fine 03:21:08 Don't know if it still works though, you had to add a lot of empty actions { } everywhere to keep it happy 03:21:18 And as soon as I got off 10.4 I stopped caring :P 03:21:40 :) 03:22:53 -!- sorear has quit [Ping timeout: 264 seconds] 03:24:52 greensnark: apple feels the same about 10.4 :D 03:25:14 greensnark, why do I find it strange that vault tags in map_def are stored in a string? 03:25:29 Zaba: Change it to a set :) 03:25:44 (if it troubles you) 03:26:21 stringilious 03:26:33 But change it to a tag_set class that uses whatever set/xyz structure 03:42:11 -!- Unrodney has joined ##crawl-dev 03:42:18 Unrodney: asciidreams_vote 03:42:19 bhaak: ToME 4.0.0, 384, Dungeon Crawl Stone Soup, 294, Dwarf Fortress, 171, DoomRL, 95, Rogue Survivor, 69, Angband, 44, UnNetHack, 34, Goblin Camp, 34, Brogue, 32, POWDER, 31 (Votes: 981, Poll closed) 03:42:25 -!- Unrodney has left ##crawl-dev 03:43:02 ... wtf? 03:43:11 Did tthat go up by over 150 within the last 9 hours? 03:44:24 Hm, no, more like 90 or so votes. 03:44:34 I see NetHack doesn't even get a mention :) 03:44:49 how could it? it's about releases in 2010 :) 03:45:42 h :) 03:45:44 *Ah :) 03:45:48 I was unaware. 03:46:18 wow, TOME bugged on me and i needed to kill -9 it to get it to die 03:46:41 is tome any good 03:46:54 I don't really knoow. 03:47:00 "it's not crawl" :) 03:47:04 it's ok, except it uses MMO style combat 03:47:10 cooldowns and stuff 03:47:14 Meh. 03:47:34 ToME's based on Angband, which is already about as far from Crawl as you can get and still be a roguelike 03:47:43 and, IIRC, made it even further away 03:49:52 in fact, for a while I thought Crawl's design decisions were just based on being as unlike Angband as possible 03:51:19 -!- sorear has joined ##crawl-dev 03:51:23 i think, before crawl (or maybe dcss, I don't know how the original crawl played like), you had angband and nethack as the opposing ends of roguelike types 03:54:08 what do you have now? 03:55:23 crawl is further from angband than nethack is 03:55:28 :) 03:55:36 thus making a better typical example of a hacklike 03:55:54 a little ironic, really 03:56:54 I think we have a triangle 03:57:13 Well, crawl is on the "less obtuse" corner and nethack is on the "lol you died again sucks to be you" corner. 03:57:13 angband is grindy, nethack is spoily, crawl is strategy 03:57:58 at least I'm told it's strategy, for me it's still "lol you died again sucks to be you" :) 04:00:36 well, I mean, Crawl has the whole "you aren't meant to be invincible even if you've got ten runes already" thing going, Angband has the "you could go straight to the level before the final boss if you wanted, but first you'll have to spend eight months grinding to be powerful enough", NetHack is somewhere in between 04:01:07 Angband's all about collecting members of a huge set of resistances, by grinding until you get them 04:01:08 nethack: "lol, you died again, you should have read this spoiler", angband: "lol, you died again, you should have been more patient and kept grinding", crawl: "lol, you died again, you should have used your brain" 04:01:40 galehar: actually, Crawl kills you by bad luck more than the other games do; I suspect that's deliberate 04:01:57 in NetHack, at least when you die, it's nearly always your fault 04:02:09 (case in point: even fully spoiled players often take months until their first ascension) 04:16:15 ais: Eh? Admittedly I've only played DCSS and ADOM from the major rogue-likes, but past level 1 (and even a lot of the time there) your deaths are almost all your own fault. 04:16:24 And I feel that way with Crawl moreso than ADOM for sure. 04:18:15 well, the design documents, at least, imply that Crawl should try to give random challenges, and it's inevitable that some will be too hard for the player 04:18:21 although I find my Crawl games mostly die to typos 04:18:46 certainly, many of my Sprint games die to bad luck, but that's to be expected as I startscum for combat luck in the first couple of rooms 04:18:56 "Angband's all about collecting members of a huge set of resistances, by grinding until you get them" <- myth 04:19:57 hmm, really? 04:20:25 certainly, that's how most Angband players I know describe it, although I don't know many 04:21:34 you don't need most resists and they aren't that hard to get... you can certainly grind for them at shallow depths if you want, but it is better to dive to more interesting depths because item generation becomes a lot better as you go deeper 04:22:08 yep, I'm aware you can complete pretty much the whole game on dlvl 1 04:22:14 and that it's faster to go deeper 04:22:26 the resistance/depth concept certainly exists, at least 04:22:45 that looks like grinding to me :) go deeper to find what you need, pop up and then go deeper again. 04:22:46 but then, it does to some extent in pretty much all roguelikes 04:22:48 a lot of people die while grinding at "safe" depths to out of depth monsters while waiting for resistances 04:23:13 elliptic: well, this is one of the things I meant by saying that Crawl's the polar opposite of Angband 04:23:15 the point is that it is actually usually a better option to dive into "unsafe" depths so you are giving the RNG less time to kill you 04:23:18 it doesn't really have a concept of safe depth 04:23:25 neither does angband 04:23:28 that's my point 04:23:41 well, it's possible to formulate such a concept even if it's a myth 04:24:15 -!- Sequell has quit [Remote host closed the connection] 04:24:24 and analogous behavior does exist in crawl... people who grind the abyss for hours for an item that they think they absolutely need before they go to pan 04:24:28 in NetHack, you talk about the risk of going in certain directions 04:24:48 -!- Sequell has joined ##crawl-dev 04:24:53 e.g. going below minetown without reflection is perfectly possible, but people prefer not to as it's risky 04:26:05 but in nethack you can postpone going below minetown 04:26:12 indeed 04:26:34 in crawl, the devteam is actually trying to make grinding not worthwile 04:26:36 an interesting, and major, part of the strategy of the game is the risk/reward tradeoff for various branches 04:26:48 minetown, sokoban, dungeons, "low dungeons" like the Castle 04:26:57 and then it stops, and the whole game just dries up below there 04:27:01 sokoban has risk? :P 04:27:05 yes :) 04:27:10 elliptic: indeed 04:27:18 yes to elliptic and ais523 :) 04:27:25 if you go there too late or underequipped, definitely 04:27:27 Sokoban has risk in the same sense that engraving Elbereth is such a risky proposition that it balances itself :P 04:27:50 greensnark: not true :) a lot of characters die in sokoban 04:27:50 the individual monsters that spawn can be quite hard to deal with 04:28:00 and Sokoban's close to the furthest from ideal terrain possible for fighting 04:28:05 It never fails to amuse me how many NH players claim that Elbereth is perfectly ok because sometimes you screw it up and die :P 04:28:15 -!- Mu_ has joined ##crawl-dev 04:28:30 greensnark: I playedNetHack for several hours the other day and forgot about elbereth. 04:28:31 greensnark: if you use Elbereth as it was probably intended, writing it only in emergencies, it's balanced 04:28:44 what isn't balanced about it is spamming it 8 or 9 times every time you see motion in the distance 04:28:50 which is, unfortunately, the best tactic 04:29:09 perhaps it wasn't very well thought out 04:29:11 I happen to think it's hilarious broken when used in emergencies 04:29:12 ais523: Crawl's opinion in this regard is that, if something can be abused, it should be made so that it couldn't. 04:29:18 due: indeed 04:29:26 And that Elbereth's brokenness has led to further design idiocy in an attempt to alleviate its worst effects 04:29:29 That there are "Elbereth" technique pages... 04:29:33 I suspect it would be the nethack devteam's opinion too, except they're really bad at noticing brokenness 04:29:38 greensnark: The easiest option would be to delete it. 04:29:42 due: Agreed 04:29:45 as in, I suspect they honestly haven't noticed the abuse that exists 04:30:06 greensnark: what's the further design idiocy there in your opinion? 04:30:07 greensnark: "further design idiocy"? what do you mean/ 04:30:11 yeah, how is it not broken in emergencies either? it is essentially a really really good emergency spell with a low failure rate that anyone can use from turn 1 04:30:29 bhaak: Elaborate code to wipe out engravings, arbtrary list of monsters that ignore Elbereth 04:30:32 28% is a pretty high failure rate if it's the only thing between you and death all the time 04:30:40 it's likely to fail at least once within a few uses 04:30:47 greensnark: oh, I suspect those would have been there anyway 04:30:54 It's not the only thing between you and death all the time 04:31:22 the first is to prevent you fighting on Elbereth (unfortunately, it doesn't actually work as you can just stack Es first), the second is entirely flavour 04:31:45 the list isn't arbitrary at all; it's a list of things which, in the environment of Lord of the Rings, would be unlikely to be scared by the name of an elven goddess 04:31:51 any gameplay relevance is entirely coincidental 04:31:56 moin guys 04:32:25 galehar: does your public key end with "...E02pQkU/" ? IRC messages tend to be cut if too long. 04:32:39 unfortunately, vanilla nethack doesn't have atheistic demons :) 04:33:01 I love Python. It lets me do naughty things like class closures to do x,y swapping with regards multidimensional arrays :D 04:33:01 Napkin: no it doesn't 04:33:21 galehar: better send it via email at napkin@develz.org, please 04:33:27 ok 04:33:32 ais523: It's entirely arbitrary. You're telling me Demogorgon is going to be afraid of a flipping scribble in the dust? :P 04:33:53 greensnark: he's a demon, that's about as close to the Nazgul as you can get 04:34:16 Save us from NH types and their logic :) 04:34:30 due: you highlighted me a few hours ago - I am around now - what's up? :) 04:34:47 the list of E-ignorers is actually pretty short and simple: elves (and mostly-elf humanoids, but there aren't any); humans (and minotaurs, mostly human humanoids); angels 04:34:51 that's it IIRC 04:35:13 Napkin: Oh, galehar fixed it for me :) I needed to build a version of tiles for Windows with a patch in it, and then test to see whether it worked on a Windows platform :) 04:35:15 I could say that Crawl has an equally arbitrary list of monsters immune to mephitic cloud 04:35:24 Napkin: But it does work, so the patch is now in trunk :D 04:35:30 ofc, in vanilla, the list is too short. it should include all monsters :) 04:35:30 awesome :) 04:35:32 not to mention, harder to remember 04:35:42 ais523: You don't need to remember now, though, as it's listed in description. 04:35:45 xv is such an awkward keystroke 04:35:54 sucks to be you :) 04:35:54 Mephitic cloud isn't exactly a zero-cost spell that any idiot can use :) 04:35:57 ;) 04:36:09 greensnark: Unfortunately it's treated like one and every idiot wants it ;) 04:36:12 I do agree MC is still OP 04:36:18 due: I'm saying that, even though there are in-game spoilers in Crawl, you actually need to keep looking them up 04:36:39 But comparing it to Elbereth is like pointing out the speck in thy neighbour's eye, when look, there is a beam in thine own :P 04:36:45 I'm not sure if there's a way around that, other than silly things like incorporating poison resistance into the color of the monsters onscreen 04:36:52 greensnark: oh, I know Elbereth's problematic 04:36:53 That sentence always cracked me up 04:36:55 greensnark: I find it amusing that another translating uses "rafter". 04:37:04 due: I've also seen "plank" 04:37:05 greensnark: OH MY GOD SOMEONE'S WHITE WATER RAFTING IN YOUR EYE! 04:37:29 I have several Septuagint Greek grammars and a trigram version of the New Testament. 04:37:33 I'm just cheerfully sitting here in the knowledge that a) while disastrous, it's not as disastrous as Crawl devs make out, b) it's not my fault 04:38:16 ais523: I'm sure it could be your fault ;) 04:38:28 But yes, in my opinion we should just remove mephitic cloud. It removes all the worrying. 04:38:51 and what about evaporation 04:38:59 ais523: Variant maintainer who does not fix it => it's your fault :) 04:39:06 monky: That is less troublesome as it requires reagents. 04:39:14 At least Spork makes an effort to fix it 04:39:25 I do think ripping it out entirely would be best though 04:39:29 greensnark: the goal of my variant is explicitly not to fix those sort of balance problems in vanilla 04:39:33 which is rather convenient 04:39:44 ais523: Yeah, which is why I think your variant has issues of purpose :) 04:39:49 Things that are creepy: I have one of those small artist's mannequins. I never move it, but every time I look at it I notice that it's carefully posed *and I did not pose it thus*. :S 04:39:54 ais523: Anyway, it's a good first step 04:39:57 Also, that was suposed to go in ##crawl-offtopic, sorry. 04:40:00 I like your interface changes 04:40:04 thanks 04:40:13 I mean, interface changes and bugfixes is pretty much all it is 04:40:18 -!- syllogism has joined ##crawl-dev 04:40:20 But if the variant gets popular you'll eventually have to make it more 04:40:22 so there were only two things you could really complement me on 04:40:26 greensnark: ripping it out completely will get you the wrath of all vanilla players 04:40:37 bhaak: Surely vanilla players can play something else? :) 04:40:38 due: I think that meph could be balanced if the numbers were tweaked sufficiently, no need to remove it entirely 04:40:53 bhaak: So what? The good players will deal with it; the bad players will become better 04:41:00 the problem with it is just that it is too reliable and the confusion lasts too long against stuff that doesn't have large HD 04:41:12 both of those could be tweaked 04:41:15 Sprint I is good evidence for luck-based balance issues in Crawl, incidentally 04:41:20 although I suspect they're less of an issue in the full game 04:41:35 Well, reliability is impossible to determine; you can't just make it more likely to fail. Duration is easier, however. 04:41:36 I was playing it as a FeTM a lot, the start of it is massively influenced by which monsters happen to drop corpses 04:41:43 I'm still partial to that overnerf idea of making meph cloud only spawn clouds around the player 04:42:10 monky: at least one good Crawl player recommended the strategy of repeatedly mephing your own square in Hive to me 04:42:11 whats the obsession with making casters get in melee range of stuff 04:42:15 due: why can't you just make it more likely to fail? 04:43:00 such a nerf already happened once and it helped 04:43:06 really? 04:43:07 greensnark: you won't get players in the first place if all the loud people in the community will say "OMFG, $variant is so hard! it has no elbereth! don't play it before you've ascended all role/race/gender combinations in vanilla". There already is an issue that players _think_ they should start with vanilla 04:43:17 due: it has a straight-up chance-to-fail check 04:43:17 it just wasn't a large enough nerf, especially at the low end of things 04:43:29 Eronarn: The actual spell being cast or the cloud affecting monsters? 04:43:38 The latter makes sense; the former is ... odd to me. 04:43:40 hmm, variables you could change with meph: how long the cloud persists; how long a monster has to stay in the cloud until they're affected on average (i.e. success chance); chance a monster becomes immune to a particular cloud after saving from it; how long the monster stays confused once affected; the effect of confusion itself; effect of monster HD/resistances on all those 04:43:43 i think it's a property of the cloud rather than the spell 04:43:51 due: chance of a monster resisting each turn is currently HD/21, it used to be HD/27 04:43:53 but it's not like other things that generate meph clouds are balanced 04:43:58 and yeah, of course I mean the cloud 04:44:07 Ahhhh. 04:44:31 ais523: I like the "buildup" aspect. 04:44:38 ais523: Actually, I like all of these ideas. 04:45:04 lots of variables that can be played with, yeah 04:45:33 i like making it not-guaranteed-3x3 + making meph-induced confusion wear off much, much faster 04:45:37 the HD/27 -> HD/21 nerf was enough to make meph no longer ridiculous in late game 04:45:52 HD/17? 04:45:53 now we just need more nerfs to handle earlier on 04:45:54 :) 04:46:09 (Joking). 04:46:24 due: I was just listing all the knobs you could twiddle if you wanted to nerf or buff meph I could think of 04:46:27 oh, here's another: size of the cloud 04:47:03 another: nerf effect of confusion on monsters 04:47:05 ais523: Yes, but I haven't touched the code recently nor played with it in ages so it's difficult for me to remember all of the variables. Having you list them makes me think that there is indeed hope. 04:47:46 monky: I listed that one already 04:47:49 oh 04:47:53 due: some of them may not be explicit in the code 04:48:17 oh there it is, second to last on the big list 04:48:18 I know my experience from playing Crawl is that confusion was ridiculously powerful both against monsters and players, but that was several versions ago 04:55:42 I've... not played much recently 04:55:52 galehar: any luck with 3106/packed_cell? 04:56:05 I'm looking into it now 04:56:24 <3! 04:56:54 It's also possible thatmy instantiation of a new packed_cell from the member of screen_cell_t is issueous. 05:00:14 tilereg-dgn.cc:111 ? 05:03:02 hey greensnark? 05:03:27 anyway, going home. see u later. 05:04:12 o/ 05:04:23 Napkin: ? 05:04:36 I have horribly choppy DSL, so speak quick :) 05:04:39 did you get my messages about scoring page? 05:04:51 Oh, right, they should be fixed now. Forgot to tell you 05:04:56 oh, nothing important 05:05:00 oh, cool :) 05:05:01 thanks! 05:05:09 I didn't think anybody looked at the stable scoring pages :) 05:05:24 hehehe 05:05:38 so it has a good side that i play stable only after all :D 05:07:09 works great, thanks :) 05:09:00 03j-p-e-g * r1cf9778054ab 10/crawl-ref/source/ (8 files): Store vault-defined tile names in the save. 05:09:02 03j-p-e-g * r5efdb5dd4cbb 10/crawl-ref/source/ (dungeon.cc l_dgntil.cc mapdef.cc mapdef.h tags.cc): Also store default floor and rock tile names in the save. 05:09:03 03j-p-e-g * rf02884c98be4 10/crawl-ref/source/l_dgntil.cc: Also store tile names defined via lua triggers. 05:09:14 03j-p-e-g * r8a8900b6d6e8 10/crawl-ref/source/tags.cc: Storing tile names: fix console compilation. 05:09:46 soon the whole game will be saved in a savegame ;> 05:11:07 We need an in-game emacs where the elisp is saved in the .cs 05:11:20 hehe 05:11:34 You should be able to C-x C-f a new character 05:11:50 ewmacs? 05:11:54 FR: implement Crawl in emacs lisp 05:12:17 The display layer is handled for you right there 05:12:47 I thought you only supported modern OS? 05:12:57 Emacs is modern! 05:13:07 Wait, you're an emacs user too 05:13:36 hehe :) 05:13:41 * due uses gvim. :) 05:13:50 Which basically makes me hated by everyone. 05:13:50 I also use viper in emacs 05:14:03 vim users go "OMFG, gui vim? EW!"; emacs users go, "OMFG, vim? EW!" 05:14:08 * due editor-war pariah. 05:22:11 -!- elliptic has quit [Ping timeout: 240 seconds] 05:22:26 greensnark: I've started writing my own roguelike :) 05:26:59 due: Cool :) 05:27:15 Does it use Python? :P 05:27:26 -!- galehar_ has joined ##crawl-dev 05:27:50 there is something inconvenient about playing vampire necromancer 05:27:59 greensnark: yes :D 05:28:06 greensnark: johanna wants to help :) 05:28:19 I am psychic, I just knew it would be in Python! 05:28:29 Beware my mad psychic powers 05:28:43 I looove python. 05:28:56 I know :) 05:29:11 when chopping the body, you'll get a message, that there's a skeleton to your feet. but if you drink the corpse, no such message is shown, which is incovenient for animate skeletons. 05:29:42 Napkin: Sounds bugworthy 05:30:02 iekko> I looked for an existing bug in mantis, but did not found anything about that. 05:30:06 iekko> I will do so then. 05:30:07 excellent :) 05:31:26 -!- monky has quit [Quit: hello] 05:42:22 03galehar * r67f4a8195744 10/crawl-ref/source/windowmanager-sdl.cc: Don't treat pressing and releasing the mouse button as the same event. 05:47:53 -!- Mu_ has quit [Ping timeout: 276 seconds] 05:49:21 -!- Mu_ has joined ##crawl-dev 05:50:35 wb Mu 05:51:53 due, if/when I deem my hive:2 vault completed, shall I just commit it? 05:53:10 Zaba: I don't see why not? 05:54:13 oookay. 05:55:58 tx 05:57:46 Unstable branch on CDO updated to: 0.8.0-a0-4224-g67f4a81 (31.25) 05:58:31 due: do you want to work on the radiant cavern at any point 05:59:17 Whatever happene diwh thtat? 06:00:00 lamil worked on it, made some maps, then disappeared. then rkd made a map (which i don't like) and put it on mantis about 2 months ago. 06:00:23 i just noticed dpeg assigned rkd's mantis to me but i dunno what to really do with the vault tbh 06:00:28 https://crawl.develz.org/mantis/view.php?id=2684 06:01:53 the monster set on that map is all over the place 06:02:43 fwiw, I don't really like those maps 06:05:08 Non-latin characters not usable in usernames (https://crawl.develz.org/mantis/view.php?id=3110) by jpeg 06:05:08 or rather, that map 06:05:34 Does anyone have lamil's maps? 06:05:40 I remember liking it... 06:06:13 They're not particularly... cavernous, are they? 06:08:46 not especially 06:12:27 hm. 06:15:33 Zaba: I recommend a moratorium on messages such as "hm.", "hrm", "Hmm", "Hm", "Hmmm" to the channel 06:16:21 greensnark, why? 06:17:06 Zaba: " hm" "Oh, hey, Zaba said something, it must be good!" and then "Oh, I want to smack Zaba with a rolled-up newspaper." 06:17:44 If it's worth saying it's worth saying something meaningful! 06:17:52 greensnark, I'm still pondering where to best put loot on my brand new hive ending vault :P 06:18:02 Zaba: That's more like it :) 06:18:21 Btw, what happened to the plans to change Hive to a portal vault 06:19:20 hive is like okawaru 06:20:18 what the 06:20:26 I just transferred my save and now it won't even load 06:20:30 ..or like that ton of dungeon.cc code that I have removed during the last indefinite timespan 06:20:45 casmith789: what's the error? 06:20:51 everybody knows it's a problem, but nobody's willing to touch it and it just kinda sticks around 06:20:52 um 06:21:02 Looks like save compat issue. 06:21:07 I think it says it crashes 06:21:25 but it lasts for .05s or something 06:21:26 !lm casmith789 06:21:26 6089. [2010-12-31] casmith789 the Severer (L16 KeBe) killed Urug on turn 71691. (D:16) 06:21:38 Napkin: can you see what the crash is with casmith789's save? 06:23:11 ahhh 06:23:33 bet jpeg's recent change changed the map cache code but idn't increase the version 06:24:25 anything i can do? 06:24:32 need the core dump? 06:24:50 can you switch it back to some other version? 06:25:00 Napkin: just the crash message would be good.. 06:25:18 how should I see it? in the morgue? 06:26:05 cool, the first time I trigger a CDO update, it breaks save compat 06:26:09 I'm so proud! 06:26:35 but, I didn't commit anything significant recently, so I'm not even to blame :) 06:26:52 Napkin: should be in moruge 06:27:25 casmith789, would you please search for your morgue? 06:27:38 morgue? 06:27:43 ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. 06:27:51 ah 06:27:57 wait it's really that? 06:28:27 that is odd 06:28:33 oh i see now 06:29:01 I've already reported a bug with ASSERT(m.mid > 0) but it wasn't in tags.cc 06:29:17 kilobyte fixed the other one 06:29:18 didn't he 06:29:19 #3098 06:29:45 mid <= 0 is a bug, no matter where it gets caught 06:30:02 Yeah, it's a save with a borken monster ID, apparently. 06:30:09 340fef0858cf79cef08e5eea02703ea979f2e264 introduced the assert. 06:30:44 removing the assert would silence the symptom, but the problem'd surface later and elsewhere, possibly in an obscure manner 06:31:04 kilobyte presumably fixed the issue 06:31:11 but yes, it seems like the save has a broken mid. 06:31:24 should i remove that last version? 06:31:43 Napkin: perhaps for now... 06:31:43 4 players in it 06:31:43 Napkin: I'm not sure it's related 06:31:56 ow 06:31:58 when was the previous version? 06:31:59 maybe then 06:32:09 -!- galehar_ has quit [Quit: disconnect] 06:38:46 disabling savegame transfer is ok too, right? 06:39:17 due? 06:39:47 casmith789: what's your char's name? 06:39:52 casmith789 06:39:53 Napkin: I'm not sure what's up 06:39:56 kilobytearound? 06:40:14 well, i disabled savegame transfer for now. a new game just worked fine for me. 06:40:33 daekdroom apparently transferred too and it worked 06:40:52 > mv crawl-svn-67f4a81/saves/casmith789* crawl-svn-9319fad/saves/ 06:40:55 try, casmith789 06:41:14 it loads :) 06:41:17 thanks 06:41:31 did you get the message about transfer being disabled? 06:42:30 yes 06:42:37 great, thanks :) 06:42:44 must just be casmith's save that is broken 06:42:52 I think so, too 06:43:10 well, let me know when to re-enable transfers 06:43:30 we should bump major version and call it a day :P 06:43:51 cheater ;> 06:44:26 the version will still be on CDO and people will get that version as last playable compatible version ;) 06:44:52 I need to steal Napkin's save versioning system for cao 06:45:00 But first cao needs more disk space 06:45:18 Zaba: zsh is awesome <3 06:45:26 Shared history across all shells alone is worth it! 06:45:38 and there are 213 players on major version 31 :D 06:45:39 yes, that's really cool 06:45:50 Although I had to twiddle it a bit to make up/down go through local history while history search checks shared history 06:45:54 oh, that's a nice feature 06:46:06 I normally use a bunch of xterms at once, and shared history is very lovely indeed 06:46:28 Zaba: Haha, we all know you don't use X 06:46:33 Napkin, players are just that: Players. They die, soon. 06:46:33 Or at least claim not to use X 06:46:38 greensnark, these days, I do 06:46:47 Egads 06:46:48 greensnark, since all the sane framebuffer drivers conflict with fglrx in some obscure way :P 06:46:59 Zaba: So are you ready to admit you use screen yet? :P 06:47:13 hehe 06:47:27 When I used the open graphics driver stack, I was indeed stuck in framebuffer consoles, but that's because they are so nice with said graphics stack. 06:47:48 greensnark, I don't use screen. 06:47:50 Wow, Zaba actually using fglrx 06:48:03 Zaba: What radeon do you have? And don't the open drivers support it? 06:48:19 greensnark, they support it, but 3d performance is still not comparable with that which fglrx provides 06:48:33 Wait, you need 3d performance?! 06:48:37 :-O 06:48:48 You're shattering my mental image of a Luddite billygoat Zaba :P 06:48:52 the world is coming to a hold! 06:49:02 not often, but sometimes I do need it, yes 06:49:35 and I'm too lazy to compile all that drm stuff as modules, so I just kind of stick with fglrx even when I don't need it :P 06:50:17 well, guys, just don't forget to tell me about savegame transferability 06:50:19 it does occasionally reboot my system upon starting X and so on, so it's not like I can forget that it's fglrx. 06:51:19 greensnark, not using KMS is quite Luddite :P 06:53:01 greensnark, and using fglrx is quite obviously masochistic, so I can't possibly be shattering any mental images of myself. 06:57:20 -!- ortoslon has joined ##crawl-dev 07:00:27 crash+assert on taking stairs (https://crawl.develz.org/mantis/view.php?id=3111) by ortoslon 07:07:50 do savefiles contain settings files? would be useful for debugging 07:20:19 Leibowitz (L23 KoEn) (D:23) 07:20:21 -!- TGWiglaf has joined ##crawl-dev 07:24:52 -!- MaryPoppins is now known as Adeon 07:25:54 -!- Zaba has quit [Ping timeout: 260 seconds] 07:25:57 syllogism: no, I don't know what could have caused it, I merely added asserts so it would fail sooner 07:26:09 oh 07:26:20 didn't look too closely 07:26:23 I fixed duplicated ids on clones, but that's unrelated 07:26:51 kilobyte: did you checked #3111 07:26:57 it's the same ASSERT 07:27:19 Zao: KMS believes the EDID data like scripture, and doesn't allow you to override it if it's wrong. And in many monitors, it is. 07:27:39 -!- Zaba has joined ##crawl-dev 07:27:46 Yoou mean Zaba? 07:27:56 hmmm? 07:27:59 14:27 <+kilobyte> Zao: KMS believes the EDID data like scripture, and doesn't allow you to override it if it's wrong. And in many monitors, it is. 07:28:23 galehar: yeah, but it appears to be a run of the mill orc -- either spawned via band generation, solo or as Beogh reinforcements, all of which go through normal placement 07:28:46 Zao: of course, Zaba being off obviously broken tab completion 07:28:52 kilobyte, well, kms works fine here, I just choose to not use it 07:31:45 kilobyte, also, can't the mode be overridden with video= on the kernel command line? 07:37:26 Goodnight folks! 07:37:34 it ignores those modes, claiming they're impossible with the monitor's specs 07:37:41 due: bie! 07:38:10 including even the monitor's native mode 07:39:38 my opinion is that even contemplating any resolution but the native one (or an integer fraction of it) is lunacy and sabotaging one's own eyes 07:41:13 galehar: I have a hunch that the issue might come from upgrading saves being botched somehow... but I don't really know how 07:44:16 I might try to recover from broken mids rather than asserting away... 07:44:39 but, what's the matter with jpeg's changes: is the save compat broken after all? 07:45:13 well, in #3098, there wasn't any upgrading involved. Just started wizmode game, fool around, upstairs and crash. 07:45:27 which change? 07:45:59 -> the talk around 13:40 our time zone 07:46:08 about storing tile names? 07:47:39 * kilobyte re-reads the scrollback over and over. 07:54:20 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:54:29 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:55:02 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:55:22 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:55:44 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:56:43 soul (L5 HEPa) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 07:56:51 yay 07:57:25 I can paper over the bug's symptoms, but we need an assert elsewhere... 08:22:02 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:04 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:07 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:10 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:15 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:18 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:19 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:23 hmmm. 08:22:27 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:33 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:42 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:22:47 wya's brother 08:22:58 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:23:11 I guess most players don't realize that it spams ##crawl-dev :P 08:23:15 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:23:30 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:23:46 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:05 nice 08:24:17 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:25 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:54 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:55 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:56 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:57 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:58 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:59 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:24:59 This m.mid > 0 assert is gripping stuff, but I think we have a bit of a surfeit of it 08:25:00 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:04 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:06 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:09 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:12 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:14 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:17 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:20 Is he stuck in a loop or just really anxious to get back to his game :P 08:25:20 lkid (L5 MDCK) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:25:28 I think possibly a loop 08:25:40 greensnark, it shouldn't really appear, should it? Saves with mids should have proper mids, saves from before mids should have ones assigned by the tags.cc code... 08:25:40 nah 08:25:47 Maybe he assumes that the game will eventually give up and load 08:25:58 Zaba: I haven't looked at the code 08:26:29 Zaba: those games are new I think 08:26:49 !lm lkid -log 08:26:49 726. lkid, XL5 MDCK, T:3310 (milestone): http://crawl.develz.org/morgues/trunk/lkid/crash-lkid-20101231-142502.txt 08:27:29 always orcs? 08:27:40 I guess orc bands are common 08:29:52 kilobyte: Are you looking at the m.mid>0 thing 08:30:14 I am off to sleep 08:30:22 feel free to mode -q gretell if they have stopped spamming 08:30:23 night all 08:30:27 Nights 08:30:45 Happy new year 08:30:59 greensnark: with great power comes great responsibility! 08:31:47 I was thinking it was the ability to change the topic to say something of such colossal silliness that it will take dpeg a month to get over it 08:32:14 I just need to come up with something truly sublime 08:32:17 The channel is -t, anyone can set the topic :p 08:32:19 * due gone. 08:32:24 These hard-bouled Germans are hard to shock 08:32:26 Oh! 08:32:31 lol, so true 08:32:35 Ok, so I have a lot more time to think 08:32:46 "Home of the Vi-Kings!"? 08:32:57 I bet that was due 08:33:09 So this is the first time you're looking at the topic? :P 08:33:46 it's on the first line of the window, but the interesting bits are always on the bottom :) 08:34:56 !lm * crash noun=~m.mid 08:34:56 42. [2010-12-31] Ragdoll the Digger (L3 MuEE) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed on turn 3449. (D:3) 08:35:13 !lm * crash noun=~m.mid s=place 08:35:13 42 milestones for * (crash noun=~m.mid): 37x D:4, 5x D:3 08:36:14 is it always with orcs? Could it be related to the new herd code? 08:46:48 -!- greensnark changed the topic of ##crawl-dev to: Home of the Vi-Kings! | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please. 08:46:52 Oops 08:47:43 [09:46] <@greensnark> Oops| Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please. 08:47:45 do it 08:47:59 -!- MarvinPA_ has quit [Ping timeout: 265 seconds] 08:48:43 -!- greensnark changed the topic of ##crawl-dev to: Home of the Vi-Kings! NetHack, AD&D, TGWi fan club! | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please. 08:49:10 Maybe I should have mentioned Sokoban in that list 08:50:02 TG isn't a race, is it? 08:50:06 although Wi is a class 08:50:07 dpeg won't ever notice. Who looks at the topic? :D 08:50:22 bhaak: Yeah, but it'll be funny when I point it out after a month :P 08:50:24 I do, on occasion 08:50:35 <3 08:51:03 AD&D is actually pretty cool. 08:51:08 btw, why does Crawl do AC backwards? 08:51:35 (I was trying to figure out what NetHack and AD&D had in common...) 08:51:50 things dpeg doesn't like 08:52:00 Yeah, actually the third item on that list is out of place 08:52:12 hmm, Crawl owes a lot to D&D, if only indirectly 08:52:33 backwards? 08:52:37 I don't think dpeg really dislikes D&D, it's more of an elaborate in-joke with him 08:52:37 do serial vaults still exist? 08:52:41 I need to make one real quick 08:52:42 don't forget Tolkien! 08:52:48 galehar: Good point! 08:52:57 bhaak: in both NetHack and AD&D, AC goes down from 10 08:53:07 thus, going up from 0 is clearly backwards 08:53:08 ais523: I thought modern D&D doesn't do that anymore 08:53:12 it doesn't 08:53:13 that's a good idea for an MDFi nerf 08:53:16 but it isn't called AD&D either 08:53:21 okawaru fanclub 08:53:21 -!- greensnark changed the topic of ##crawl-dev to: Home of the Vi-Kings! Tolkien, NetHack, AD&D, Rammstein fan club! | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please. 08:53:31 There, that ought to hit all of dpeg's buttons 08:53:38 in all versions of the game that are actually called AD&D, it goes down from 10 08:53:42 ais523: I also thought they stopped doing that when it still was AD&D :) 08:53:45 nope 08:53:45 Sorry TGWiglaf, but I think dpeg quite likes you 08:53:50 happened in the change from third edition 08:53:54 *to third edition 08:53:59 and the leading A was dropped at the same time 08:55:00 -!- pointless_ has joined ##crawl-dev 08:55:22 hmm, does dpeg also dislike vi? 08:56:02 ais523: He doesn't, but due added that line and I'm keeping it around for hysterical raisins. 08:56:20 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 08:56:34 greensnark: fail! 08:56:42 !lm * crash noun=~m.mid s=name 08:56:42 43 milestones for * (crash noun=~m.mid): 25x lkid, 7x Daek, 6x soul, 5x Ragdoll 08:56:49 bhaak: What is? :P 08:57:12 if I make a serial vault that places a bunch of trees, some of which contain dropbears, would it be added? 08:57:14 greensnark: Gretell still spams :) 08:58:17 bhaak: Oh, I didn't say I'd fixed it, did I? :P 08:58:53 ah, so do you fix bugs around here :) 08:59:02 Oh no, I don't 08:59:17 You read too much between the lines :P 08:59:43 We need the actual game version in milestones files 08:59:55 !lm * crash noun=~m.mid s=v 08:59:55 43 milestones for * (crash noun=~m.mid): 43x 0.8.0-a0 08:59:57 Daek (L4 DrTm) ASSERT(m.mid > 0) in 'tags.cc' at line 2939 failed. (D:4) 09:00:00 ^ not very useful 09:00:13 !lm Daek crash -tv 09:00:14 8. Daek, XL4 DrTm, T:3383 (milestone) requested for FooTV. 09:00:20 I will descrease the interline space of my irc client :) 09:00:40 !lm Daek crash -2 -tv 09:00:40 7. Daek, XL4 DrTm, T:3383 (milestone) requested for FooTV. 09:00:58 -!- Textmode has quit [Ping timeout: 260 seconds] 09:02:10 Is there a bug report with a copied save for this problem 09:02:42 #3111 09:02:43 two 09:03:08 Ok, thanks 09:03:43 sorry, been busy 09:04:07 the copied save is of no use, the damage is already done there with no traces why 09:04:47 Is the guilty monster interesting 09:06:45 So the monster operator= doesn't copy mids? 09:07:14 Because at least the transit list relies on operator= copying everything of importance 09:07:22 So perhaps this is a follower issue? 09:07:58 Reload game in a new Crawl, go downstairs with a follower next to you, maybe? 09:08:01 * greensnark tries. 09:09:18 greensnark: the one time I had the bug, I took stairs with monsters next to me 09:12:04 I got it just now 09:12:21 as a beogh worshipper 09:12:54 with an orc priest following me downstairs 09:14:22 easy to reproduce even 09:16:09 -!- OG17 has quit [Read error: Connection reset by peer] 09:16:13 No Skeleton information after a VpNe blood drink (https://crawl.develz.org/mantis/view.php?id=3112) by iekko 09:16:58 -!- dpeg has joined ##crawl-dev 09:17:05 only short visit 09:17:05 dpeg: You have 1 message. Use !messages to read it. 09:17:08 !messages 09:17:08 (1/1) Keskitalo said (11h 29m 52s ago): Slime plan == Charles' proposal for tweaking the levgen a bit (which now seems like a much smaller change than I thought previously) 09:17:17 !seen Keskitalo 09:17:18 I last saw Keskitalo at Fri Dec 31 03:57:11 2010 UTC (11h 20m 6s ago) saying ok, ill try to sleep instead :) on ##crawl-dev. 09:17:31 Sleep at 5am? Finns :) 09:17:38 !seen pointless_ 09:17:39 I last saw pointless_ at Fri Dec 31 14:55:00 2010 UTC (22m 39s ago) joining the channel. 09:17:44 pointless_: awake? 09:17:51 yes 09:18:00 it's 10 am in this part of the world 09:18:03 pointless_: so your plan would keep the slime walls intact? 09:18:10 pointless_: you Americans :) 09:18:13 moin dpeg 09:18:16 Hi! 09:18:26 Zaba: you wanted to catch me yesterday... About the shop vaults? 09:18:42 dpeg, no, 'twas about how I'm making a hive:2 vault 09:18:57 Zaba: cool. What's the question? 09:19:08 I'm still making it, though :P 09:19:08 pointless_: I believe that Slime level generation can be improved keeping the slimy walls. 09:19:25 dpeg: he's referring to the argument we had in that mantis issue, my proposal which I guess is probably on the wiki is to alter levgen for better connectivity, and I'll get around to implementing this whenever I feel like I have the time for it 09:19:28 pointless_: so what is your plan? =) 09:19:35 what is wrong with slime level generation? 09:19:38 pointless_: thanks, sounds good 09:19:55 Zaba: it could be more aware of the fact that walls are dangerous. 09:20:10 Eg. no (or fewer) crooks you can only explore cumbersomely. 09:20:34 due: I think I found the bug for your packed_cell patch 09:20:53 don't have the time to fix it right now, but I added a comment to the mantis 09:21:13 I did get the slimy rune after the change, and without Jiyva. I don't believe that the slimy walls ruin the rune, I rather think the effect is good and works as intended. I see that certain players feel the need to explore everything and we can do something for them without sacrificing the slimy walls. 09:21:19 galehar: hi 09:21:20 what's the function to get one of the player's coordinates? 09:21:38 or, how would I do that 09:21:43 (in lua) 09:23:15 ok leaving now 09:23:26 have a nice party tonight! 09:29:29 -!- galehar has quit [Quit: Page closed] 09:33:35 hey guys - is savegame transfer safe again? 09:39:12 Keene Ahnung! 09:43:10 Too many phone calls on the Happy New Year theme 09:43:13 dpeg! 09:43:29 greensnark: already in 2011? 09:43:37 dpeg: No, about 3 hours away 09:43:45 greensnark: everybody already drunk around you? 09:43:54 hint: there is still time for a commit =) 09:43:57 dpeg: I skipped the New Year's party this time :P 09:44:05 +1 09:44:11 greensnark: alone at home? 09:44:11 you mean, they've already started partying there? 09:44:12 I'm chasing down paperwork in Chennai, staying at my parents 09:44:23 people normally start partying only on 1st of january here 09:44:30 !send greensnark paper golem 09:44:30 Sending paper golem to greensnark. 09:44:38 Zaba: Well, the partying starts are 7pm on Dec 31 and carries on until dawn, pretty much 09:44:42 *starts at 09:44:45 uh oh. 09:45:15 And to make sure people who decided to skip the party get no sleep, there will be a lot of drunken calls around midnight 09:45:19 Well, starting after midnight 09:45:20 Just another excuse to get hopelessly drunk :) 09:45:32 greensnark: you can turn off the phones! 09:45:44 jpeg is coming over tonight 09:48:18 dpeg: Do you notice anythong *different* about this channel? 09:48:23 Anything subtle but awesome 09:48:42 Or not so subtle but nevertheless awesome 09:49:08 our @'s? 09:49:15 Oh, that's not awesome 09:49:24 yes greensnark was just bragging 09:49:24 We're the heros! 09:49:42 oh my god, the title 09:49:58 * dpeg goes puking in the old year :) 09:50:11 out of the four, I am not just which is the greatest crime 09:50:28 Hahaha 09:50:31 You found it! 09:50:51 dpeg: Let me know, and I'll sort it by descending order of heinousness 09:50:56 Assuming heinousness is a word 09:51:14 I believe it is 09:51:32 -!- TGWiglaf has quit [Quit: Page closed] 09:51:32 greensnark: there are many things in life which cannot be ordered linearly. 09:51:55 03greensnark * r0915ed9a552f 10/crawl-ref/source/monster.cc: [3111] Make monster operator= copy mid so that monsters placed from the transit list don't lose their mid. 09:51:56 I am afraid that heinousness of Tolkien, Nethack, AD&D and Rammstein is an instance. 09:52:07 After all, you wouldn't sort the cardinal sins either, right? 09:52:45 I can totally sort the cardinal sins 09:53:10 For instance, it is not at all clear to me why pride is a cardinal sin beyond the idiocy of the people who framed that list 09:53:35 Seems to me that constructing a list of cardinal sins is a very prideful thing to do 09:54:04 Also I totally admit I had to hit wikipedia to recall the list of sins 09:54:50 kilobyte: Do we need to modify the save shim to deal with old broken saves? 09:55:36 I bet kilobyte is partying like it's 2011 09:55:53 I hear they have this weird drink called vodka that tastes of nothing at all in the uncivilized parts of Europe 09:56:11 Much like we have arrack in south india 09:57:08 -!- psyshvl has joined ##crawl-dev 10:01:06 -!- ortoslon has quit [Quit: bye] 10:03:04 -!- TGWiglaf has joined ##crawl-dev 10:05:29 03greensnark * rb9a936f6c623 10/crawl-ref/source/ (tag-version.h tags.cc): Bump mid shim to handle newer saves. 10:05:39 03greensnark * rfc9de525ffda 10/crawl-ref/source/makefile: Add -Wno-unused-function to lexer cflags to shut up the warnings about yyunput. 10:10:38 Unstable branch on CDO updated to: 0.8.0-a0-4227-gfc9de52 (31.26) 10:11:13 -!- joosa has joined ##crawl-dev 10:14:53 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:14:53 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:14:55 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:14:55 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:14:56 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:14:58 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:15:01 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:15:19 Oh crap :P 10:15:25 I knew I made a minor mistake 10:20:50 03greensnark * rfd8673d1190a 10/crawl-ref/source/tags.cc: Use TAG_MINOR_FIX_MID correctly (mea culpa). 10:21:24 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:21:27 Unstable branch on CDO updated to: 0.8.0-a0-4228-gfd8673d (31.26) 10:21:43 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:21:46 phyphor (L2 FeCr) ASSERT(_read_offset+size <= _pbuf->size()) in 'tags.cc' at line 162 failed. (D:2) 10:23:20 So did that not fix it :P 10:23:25 hmm 10:23:46 having the version in the milestone would definitely help with that :P 10:24:13 Yeah 10:26:22 For some reason Napking plays 0.7 :P 10:26:29 I have no idea why 10:26:37 Acid Blob description is incorrect or tile is. (https://crawl.develz.org/mantis/view.php?id=3113) by Twilight 10:27:13 -!- ortoslon has joined ##crawl-dev 10:27:57 syllogism: Did you just pm Gretell 10:28:18 Gretell's log has an unhappy message involving your name 10:28:44 channel topic <-----------1km----------> relevance 10:28:44 hehe 10:29:09 Hey, somebody reads the topic around here 10:29:24 bhaak: Your allegations about nobody reading the topic have been directly refuted 10:29:36 I knew you secretly admired nethack and its greatness at everything 10:29:51 Truly, nothing can be hidden from Mary Poppins 10:31:12 so, what about this savegame transfer? 10:31:20 feels uncomfy to have it disabled! 10:31:24 help! :D 10:31:27 Napkin: it should work now! 10:31:36 I hadn't realised it was disabled 10:31:50 i did, when there was a slight panic earlier ;) 10:32:22 so, the update from 10 minutes should do the trick? 10:32:36 fd8673d should fix it 10:32:41 awesome! 10:32:45 thanks :) 10:33:18 greensnark: haven't PMed gretell today, as far as I can tell :P 10:33:20 alright, re-enabled (untested) 10:34:43 Windows development builds on CDO updated to: 0.8.0-a0-4228-gfd8673d 10:34:58 greensnark: you made him look for it :) 10:35:12 bhaak: ortoslon just joined! 10:35:15 what was the message ? 10:35:36 syllogism: I'm not sure it was you, it could just be something else broken in Gretell 10:35:45 "Received line :syllogism: also lots of data. It starts with: Y .7.1-1-g7ce9b19 Jeetuc @ix ossuary ossuary that is not IRC protocol" 10:35:56 weird 10:35:57 greensnark: oh, I thought you meant dpeg. As long as you don't prove that ortoslon is not nobody, I still might be right! ;-) 10:36:29 bhaak: i'm widely known in narrow circles :P youtube.com/ortoslon 10:41:55 03greensnark * r09d6ee8426bd 10/crawl-ref/source/tags.cc: Fix off-by-one check for fix-mid tag. 10:41:59 okay, okay: "nobody except ortoslon and sometimes ais523 reads the topic" 10:43:11 you can only add that much exceptions before the resulting statement won't fit in an irc message 10:44:25 -!- greensnark changed the topic of ##crawl-dev to: Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please | This is walrus appreciation week. 10:44:31 Ok, no more frivolity 10:44:32 I could rephrase it as: "only weird people read the topic" :-) 10:44:34 * greensnark is off. 10:45:10 -!- ortoslon has quit [Quit: bye] 10:46:25 @?? walrus 10:46:25 unknown monster: "walrus" 10:48:25 s/walrus/nethack/ 10:53:24 -!- MarvinPA has joined ##crawl-dev 10:55:09 we need a new monster: ostrich. when it sticks its head in the ground, you go blind 10:55:17 this is needed immediately 10:56:54 does Crawl have blindness as a status change? 10:56:55 -!- Zaba has quit [Ping timeout: 240 seconds] 11:01:28 -!- Zaba has joined ##crawl-dev 11:02:14 ais523: no 11:02:51 hmm, I suppose it has invisible stalkers to give that sort of challenge 11:02:58 but too many races are immune to them 11:15:47 -!- pointless_ has quit [Ping timeout: 240 seconds] 11:23:03 -!- TGWiglaf has quit [Quit: Page closed] 11:29:00 -!- TGWiglaf has joined ##crawl-dev 11:44:49 what the fuck is 'plojy' 11:50:59 -!- psyshvl has quit [Quit: stuff] 11:55:31 -!- MarvinPA has quit [Ping timeout: 265 seconds] 11:57:56 -!- TGWiglaf has quit [Quit: Page closed] 12:05:34 -!- pointless_ has joined ##crawl-dev 12:06:40 -!- Moriasc has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]] 12:19:20 -!- valrus has joined ##crawl-dev 12:24:26 valrus (L15 DDVM) (D:16) 12:24:46 so uh, was that a crash? 12:25:06 !lm valrus crash 12:25:06 1. [2010-12-31] valrus the Necromancer (L15 DDVM) ? (D:16) 12:25:12 neat 12:25:26 !lm valrus "-log" 12:25:26 Malformed argument: "-log" 12:25:28 er 12:25:43 !lm valrus crash -log 12:25:43 1. valrus, XL15 DDVM, T:33348 (milestone): http://crawl.develz.org/morgues/trunk/valrus/crash-valrus-20101231-182426.txt 12:26:11 that one has been around for a while 12:26:17 what causes it? 12:26:38 I think it just saved me from death :P 12:26:51 something related to feature (?) mimics, I think 12:27:30 huh 12:27:40 mons_is_known_mimic gets called on a NULL pointer, that is, get_cell_glyph_with_class is called on a cell that has no monster, and yet some code that calls it assumes that there -is- a monster 12:27:49 I haven't been able to figure it out beyond that :P 12:28:19 I'd be surprised if nobody's reported it yet 12:32:39 valrus, I could be blind, but I don't see anything like that in mantis---I guess you should report it... 12:33:21 ok 12:35:22 can I quote you in the report description? 12:36:23 well.. I guess. I'm pretty sure that I'm at least vaguely right :P 12:37:33 I mean, otherwise I basically leave the description empty. I have no idea 12:37:45 I would have guessed it was snorg-based, since he's who I saw when it happened 12:38:51 well, mons_is_known_mimic is present in the stack trace 12:39:33 psh, stack traces, like I know about that 12:39:47 I've just written two patches hurrrrr 12:42:23 Segfault, suspected related to feature mimics (https://crawl.develz.org/mantis/view.php?id=3114) by valrus 13:06:55 -!- Mu_ has quit [Ping timeout: 276 seconds] 13:07:29 -!- Mu_ has joined ##crawl-dev 13:16:13 -!- elliptic has joined ##crawl-dev 13:22:34 -!- OG17 has joined ##crawl-dev 13:39:09 -!- elliptic has quit [Quit: Leaving] 14:06:24 03zaba * r286102fb174c 10/crawl-ref/source/dungeon.cc: Simplify third pass in dgn_find_nearby_stair using rectangle_iterator. 14:06:28 03zaba * r0980a63ef14a 10/crawl-ref/source/dungeon.cc: Use radius_iterator in _has_vault_in_radius. 14:06:34 03zaba * rc940dc3321d9 10/crawl-ref/source/dat/des/branches/hive.des: Add a Hive:2 vault by me. 14:07:09 Just drank a potion and "A flood of memories washes over you."? 14:07:21 Is this a new potion in trunk? Or...? 14:08:09 experience on xl27 14:08:24 Ah.... That would be it. Thanks! 14:10:17 -!- psyshvl has joined ##crawl-dev 14:21:28 hi psyshvl 14:21:33 Hi 14:21:54 Been working on that portal vault 14:22:20 Not satisfied with how its is right now, but I think its alot better now. 14:25:15 keep it up :> 14:25:39 Yeah, I feel really stupid about it actually 14:25:47 I looked up the syntax document 14:26:02 and I got the metal wall and glass wall glyphs messed up 14:26:12 I really though m was metal walls for some reason 14:28:08 sounds quite logical, even :p 14:28:23 yeah 14:28:35 Can two fog machine markers be placed on the same square? 14:28:51 I can;t find anything on whether that works or not 14:28:57 did you try it and see? 14:29:06 No, I guess I should 14:29:37 Last time I tried, Crawl crashed on me though 14:29:43 wouldn;t open 14:29:55 didn;t get a crashlog (that I could at least find) 14:30:01 what OS are you on? 14:30:06 Mac OS X 14:30:21 do you use the tiles or the console version, and how do you start it? 14:30:38 -!- ais523 has quit [Remote host closed the connection] 14:30:40 Console, and by clicking the crawl soup pot icon 14:30:57 try starting it from console. When map compilation fails, it prints the error and exits. 14:31:05 okay 14:37:09 Wait, Zaba, you mean Terminal right? 14:37:14 yeah 14:37:17 Because that's what I ussually use to run Crawl 14:37:27 Yeah, uh, I tried that, and it didn;t work 14:37:35 I also tested with normal DCSS and it didn;t work 14:37:38 does it give any reason why it doesn't start? 14:37:45 No 14:38:09 well... then don't overuse fog machines, and call it a day :P 14:38:19 Well 14:38:30 the problem is, it isn;t running normal DCSS either 14:38:52 are you sure you're starting it right, and such? 14:38:56 Yeah 14:39:01 same as I normally do 14:39:05 and trying with terminal 14:39:29 you start it correctly from terminal, too, right? 14:39:55 it works when not reading your .des file, right? 14:39:59 No 14:40:11 It's not working period, with or without modded files 14:40:30 I'm ust going to re download DCSS 14:40:30 well, then odds are that you're not starting it right 14:40:34 yeah 14:40:46 but, I've no idea about os x 14:40:58 well, thanks for trying at least 14:41:50 I got it to work! 14:41:54 nice 14:42:04 I just had to open it, and open crawl with terminal 14:42:15 Now I'm going to copy it, and add the portal vault 14:42:20 See if it works 14:42:24 Thanks for the help 14:42:33 no problem.. 14:44:13 -!- valrus has quit [Remote host closed the connection] 14:49:33 -!- monky has joined ##crawl-dev 15:27:39 -!- Zchris13 has joined ##crawl-dev 15:32:41 -!- elliptic has joined ##crawl-dev 15:37:07 -!- pointless_ has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]] 15:38:42 -!- Zchris13 has quit [Ping timeout: 246 seconds] 15:52:49 03kilobyte * r36d5f39f50a7 10/crawl-ref/source/ (AppHdr.h rng.cc syscalls.cc syscalls.h): Use the Windows equivalent of /dev/urandom there. 15:56:02 03zaba * r85f114239205 10/crawl-ref/source/l_dgnbld.cc: Simplify dgn.spotty_map by using radius_iterator. 16:34:20 -!- Zchris13 has joined ##crawl-dev 17:00:14 -!- elliptic has quit [Quit: Leaving] 17:16:12 Zaba: Did you find the issue with mons_is_known_mimic and NULL pointers? 17:16:39 Okay, guess not. 17:18:36 -!- Zchris13_ has joined ##crawl-dev 17:19:08 -!- Zchris13__ has joined ##crawl-dev 17:21:50 -!- Zchris13 has quit [Ping timeout: 265 seconds] 17:22:04 -!- Zchris13__ is now known as Zchris13 17:23:40 -!- Zchris13_ has quit [Ping timeout: 276 seconds] 17:23:40 -!- psyshvl has quit [Remote host closed the connection] 17:49:10 Picking up a corpse's weapon breaks the link between it and the owner when zombified... (https://crawl.develz.org/mantis/view.php?id=3115) by Twilight 17:53:49 -!- valrus has joined ##crawl-dev 17:56:17 -!- valrus has quit [Remote host closed the connection] 18:12:24 -!- Zchris13 has quit [Ping timeout: 276 seconds] 18:14:30 03kilobyte * r93eb30cc1f84 10/crawl-ref/source/rng.cc: Fix a crash in read_urandom(). 18:47:02 -!- elliptic has joined ##crawl-dev 18:47:58 -!- Zchris13 has joined ##crawl-dev 18:50:27 -!- Tenebrys has joined ##crawl-dev 19:09:14 Whoa, what a fascinating crash. 19:09:26 #0 0xb7d54747 in std::string::c_str() const () from /usr/lib/libstdc++.so.6 19:09:26 #1 0x0875908b in _post_init (newc=false) at startup.cc:198 19:09:40 Which seems to be macro_init. 19:12:09 hi due! 19:12:21 hi deep leg! 19:16:49 hum, interesting crash. 19:17:34 did you hear about the LRD crashes that have been discovered? 19:17:38 yes 19:17:42 okay 19:18:00 The targetting systme needs to do bounds checking to disallow targettin gthe border 19:18:08 the beam structure needs to do bounds checking to disallow affecting the border 19:18:23 and the noise code needs to do bounds checking before causing a noise event on the border 19:21:26 -!- galehar has joined ##crawl-dev 19:21:51 happy new year! 19:23:26 Hi! 19:24:47 hey what happened to the new awesome channel topic? 19:26:32 It's walrus appreciation week, I see? 19:26:42 galehar: you have a family, you should sleep :) 19:26:57 galehar: I was so depressed that greensnark felt guilty and removed it. 19:27:15 hey! it's new year's eve! 19:27:18 :) 19:27:42 due: did you saw what I put on #3106? 19:27:45 03kilobyte * r6e6ab619dd63 10/crawl-ref/source/spl-tornado.cc: Restrict Tornado to a biconnected graph. 19:28:10 galehar: Yes, I'm having to do a clean build though because I can't start characters without getting weird std::string segfaults everywhere. 19:29:33 galehar: My solution was to move cell->tile.clear() into the radius iterator; it seemed cheaper. 19:31:30 good idea 19:31:33 bye folks 19:31:39 bye dpeg 19:31:43 -!- dpeg has quit [Quit: zzz] 19:36:11 -!- Zchris13 has quit [Ping timeout: 240 seconds] 19:36:35 galehar: Can you get it to compile? 19:36:40 I've got massive issues with segfaulting. 19:37:42 -!- Zchris13 has joined ##crawl-dev 19:40:52 Okay, it's not me. 19:41:48 seems to work all right 19:45:24 hm 19:45:32 So it's either my platform, which is unlikely, or my code... 19:45:41 What's your build platform? Windows, yeah? 19:45:52 no I'm under Linux here 19:45:56 Windows at work 19:45:56 oh 19:45:57 hm 19:46:14 To th evalgrind machine! 19:50:28 galehar: Do youdo debug or wizard builds? 19:50:46 usually debug 19:50:59 -!- elliptic has quit [Ping timeout: 240 seconds] 19:51:09 but I'm going to sleep now 19:51:18 let me now if you need more help on this 19:51:20 bye 19:51:22 -!- galehar has quit [Quit: disconnect] 19:51:32 hm this is odd 19:58:22 Really, really odd. 19:59:06 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 20:01:02 does it still keep crashing for you? 20:01:58 yes 20:02:01 random weird crashes 20:02:08 Score 20:02:12 Program received signal SIGSEGV, Segmentation fault. 20:02:12 0xb7cd3178 in std::string::assign(char const*, unsigned int) () 20:02:12 from /usr/lib/libstdc++.so.6 20:02:12 (gdb) bt 20:02:12 #0 0xb7cd3178 in std::string::assign(char const*, unsigned int) () 20:02:15 from /usr/lib/libstdc++.so.6 20:02:17 #1 0xb7cd3345 in std::string::operator=(char const*) () 20:02:20 from /usr/lib/libstdc++.so.6 20:02:22 #2 0x0844cf37 in game_options::reset_options (this=0x8b2fc40) 20:02:25 at initfile.cc:683 20:02:27 #3 0x0844f6f9 in read_init_file (runscript=true) at initfile.cc:1314 20:02:30 #4 0x0875748f in _post_init (newc=true) at startup.cc:251 20:02:32 #5 0x08759c62 in startup_step () at startup.cc:899 20:02:35 #6 0x0850b3f9 in _launch_game () at main.cc:345 20:02:37 #7 0x0850b2d4 in _launch_game_loop () at main.cc:323 20:02:39 #8 0x0850b161 in main (argc=1, argv=0xbffff464) at main.cc:285 20:02:42 Erm, whoops. I meant to paste in pastebin, not IRC. 20:02:59 http://pastebin.ca/2035417 20:03:22 This is with tiles on Linux. 20:03:41 It seems like any time std::string does something it crashes. 20:03:54 I've already done a clean build--complete rebuild. 20:05:00 Oh, possibly a ccache issue. 20:07:12 And a distclean followed by a clean left a few .o files. Huh? 20:08:36 unless you turned on ccache options with big fat warnings attached, it should never cause a misbuild 20:08:57 are you having any... oh, compile times are massive for you. 20:09:08 by default it even goes to such lengths that if the timestamp on the compiler changes, all builds are invalidated 20:10:06 Well, I tried clearing the cache anyway. 20:10:15 I meant that perhaps there was a buggy object flie in the ccache. 20:10:19 that .o files... what are they? 20:10:28 wnidowmanager-sdl.o 20:10:45 I did rm *.o after that, though... probably should've checked for other ones. 20:10:47 ah, remnants of a tiles builds 20:10:55 -!- syllogism has quit [] 20:11:05 -!- Tenebrys has quit [] 20:11:11 I still don't understand what the actual issue. 20:11:55 in what you posed... runscript=true ?? 20:12:28 Yeah. 20:12:46 I started it as gdb ./crawl -> run 20:23:13 I just built it on real i386... seems to work ok 20:28:02 -!- eith has joined ##crawl-dev 20:35:31 kilobyte: Segfault with a completely clean build. :/ 20:39:44 could you checkout an old commit that's known to be good and try it? 20:41:29 Yeah, going to do that if this fails. 20:41:43 (I'm dumping my packed_cell patch in the assumption that this is causing the issue.) 20:43:31 it'd have to do a nice fandango on core to break such unrelated parts, but I've seen better already 20:51:10 http://sprunge.us/BIMN 20:52:55 03kilobyte * r02d72b998fab 10/crawl-ref/source/ (stairs.cc stairs.h startup.cc): Don't take notes or cancel tornado when restoring the game. 20:52:57 03kilobyte * r2c7a5022d60c 10/crawl-ref/source/spl-damage.cc: Fix igniting poison clouds not giving xp. 20:53:07 03kilobyte * r452e414e47d5 10/crawl-ref/source/cloud.cc: Allow any clouds to overwrite tornado. 20:56:13 no crashes for me 20:59:47 -!- elliptic has joined ##crawl-dev 21:00:58 Nope, still segfaulting. 21:01:02 I don't understand. 21:01:37 -!- Zchris13__ has joined ##crawl-dev 21:02:37 at least in console, properties seem to work fine... my new comp still hasn't arrived and the lousy one I'm using to ssh is no good for tiles, so I can't check that 21:02:43 http://pastebin.ca/2035445 21:03:10 The code it's crashing with would seem to be platform independant. 21:03:35 -!- Zchris13__ has left ##crawl-dev 21:04:19 -!- Zchris13 has quit [Ping timeout: 260 seconds] 21:04:53 Ohh,, hm. 21:05:20 relevant: 21:05:31 Changing the parameters to pass by reference prevents the objects from being copied and thus they do not go out of scope when the constructor is called. The problem seems to be firmly rooted in the fact that whenever a std::string passes out of scope its destructor is called and a segmentation fault occurs. 21:05:42 So it's trying to call ->c_str() on a string that is already yout of scope. 21:07:39 valgrind would catch this, but doesn't... and no crash, too 21:07:55 So it's something only found in tiles. 21:08:14 can't help you there right now :( 21:08:17 Yeah. 21:08:28 I ran it through valgrind earlier but it didn't seem to pick anything interesting up. 21:09:05 const std::vector& files = Options.additional_macro_files; 21:09:35 Why is it using a reference here? 21:09:53 Or is the error something else. 21:10:18 got a n900 (tiles don't like it's X server) and a pentium 3 box that served as a router where I didn't even bother to get X running. 21:10:30 it's a reference to a vector, not a vector of references 21:10:36 Ah. 21:11:54 Okay, gonna revert to the version before jpeg added the tile name storage. 21:12:00 since Options is a global object and nothing destructs the vector inside, the reference to it never goes invalid 21:12:14 I don't think that is the actual issue. 21:12:19 As that code hasn't been changed in a very long time. 21:13:03 yeah... secondary damage most likely 21:16:05 Do you see anything in jpeg's commit that might be issueous? 21:16:27 not really, but I know little about tiles 21:16:43 Funny, 2011 feels exactly like 2010 to me 21:16:54 I don't see what all the hype is about 21:17:24 Also it is passing strange that kilobyte is up already or is still awake, as the case may be 21:17:33 uhm, it 21:17:36 uhm, it's only 4am 21:17:52 Heh, New Year celeb? 21:18:39 can you celebrate it better than by coding or reading IRC? 21:19:06 Questionable 21:21:05 -!- Zchris13 has joined ##crawl-dev 21:22:30 apparently bog mummies have shells into which they can withdraw 21:23:03 monky: huh? 21:23:24 The bog mummy withdraws into its shell! 21:23:28 yeah, just reproduced that 21:23:29 Where is this? 21:23:43 kilobyte: What is the issue? 21:23:44 st_ got it in his game (as clouded) a little bit ago 21:24:12 fAh. 21:24:26 mon-abil:2562 21:24:29 needs a break 21:24:35 Hehehe 21:24:41 Can we keep that message, please 21:24:42 can some fix it and commit? :) 21:24:44 It's hilarious 21:25:11 and horribly undefined 21:25:15 oooh 21:25:24 Nothing undefined about case fallthrough! 21:25:30 if you modify agate snails to have a number > 0 they will shoot spikes like manticores! 21:25:38 DOOO IT 21:26:00 I despise switch statements and fallthrough soooooooooooo much. 21:26:12 snails need the buff :P 21:26:27 elliptic: they're geting reflection 21:26:40 that seems weird 21:26:45 thunder snails? 21:26:50 03kilobyte * r2b21f2b8c3b3 10/crawl-ref/source/mon-abil.cc: Strip bog mummies out of their shells. 21:26:58 if stuff is being done with reflection, can you fix player reflection? 21:27:05 since it is nearly useless as is 21:29:17 casmith789: their shells are shiny; why is this weird? 21:29:20 03kilobyte * r9338ca061eaa 10/crawl-ref/source/mon-abil.cc: Take spikes from snails and gators, too. 21:29:32 casmith789: further justification: shields of reflection are made from agate snail shell. 21:29:37 hehehe 21:29:40 what does shiny have to do with reflection? 21:29:57 I like the flavor with shields being made from snail shells, though :P 21:29:57 it just seems completely random, that's all 21:29:58 elliptic: you can see your reflection in glossy surfaces. 21:30:02 but yes 21:30:06 i'll reverse the flavour. 21:30:12 cool 21:31:46 fingers crossed 21:32:48 Okay, the issue is with jpeg's commit. 21:32:54 Or rather 21:33:16 something in dd76e9b60799d107ae2731a6ff7baf97bf1c2f9f..HEAD 21:34:02 I'll do a manual bisect. 21:34:22 time to check if 2011 beds are ok 21:34:31 due, keep kicking ass. That's an order. 21:34:34 kilobyte: good luck 21:34:48 kilobyte: especially considering I don't know C++ very well... I'll try :) 21:38:11 -!- Zchris13 has quit [Ping timeout: 240 seconds] 21:38:29 -!- eith has quit [Ping timeout: 240 seconds] 21:42:11 -!- Zchris13 has joined ##crawl-dev 21:55:38 -!- Textmode has joined ##crawl-dev 22:05:35 Raising the to-skill to 27 (by victory dancing, say) does not end reskilling. (https://crawl.develz.org/mantis/view.php?id=3116) by elliptic 22:08:16 03greensnark * r86591b14709c 10/crawl-ref/source/rltiles/Makefile: Make rltiles use contrib SDL and png if CONTRIB_SDL and NO_PKGCONFIG are set. 22:09:00 due: I don't see any mystery crashes 22:09:08 What do you do to get it to crash 22:16:01 -!- valrus has joined ##crawl-dev 22:17:33 -!- valrus has quit [Remote host closed the connection] 22:17:55 greensnark: Start a new game. 22:18:10 I did! 22:18:11 greensnark: It gets to macro_init and then cries foul on std::string::c_str 22:20:02 Did you try tossing your macro file if any 22:20:06 It doesn't crash here 22:20:34 I don't ... have a macro file to toss? 22:20:38 Anway, I commented out that line. 22:20:45 And it fails somewhere else on std::string::c_str 22:24:25 I posted a couple of tracebacks up above earlier. 22:24:39 Is it Oh, I'm at work. 22:24:54 I can test on my machine here too. 22:25:04 Even if I have to fetch 3mb of commits 22:35:09 egad, it's somethign to do with my ssyste :/ 22:37:27 Okay. Looks like I get to wipe out crawl and start all over again when I get home. 22:37:32 And possibly a reboot. Ciao. 22:44:01 -!- ahpla has quit [Read error: Connection reset by peer] 22:44:19 -!- ahpla has joined ##crawl-dev 22:51:58 -!- psyshvl has joined ##crawl-dev 22:53:23 -!- Zchris13 has quit [Ping timeout: 240 seconds] 23:37:21 Windows development builds on CDO updated to: 0.8.0-a0-4242-g86591b1 23:51:53 Okay, reboot, cleaned everything. Complete recompile tile! 23:54:09 Unstable branch on CDO updated to: 0.8.0-a0-4242-g86591b1 (31.26) 23:56:55 -!- Zchris13 has joined ##crawl-dev