00:02:25 nooo 00:02:39 It's a mild nerf :P 00:02:47 yay :) 00:02:47 "Nobody will even notice" 00:07:13 -!- stabwound has quit [Ping timeout: 245 seconds] 00:07:23 -!- stabwound has joined ##crawl-dev 00:19:11 -!- TGWi has quit [Read error: Connection reset by peer] 00:19:34 03greensnark * rf5497ce435ee 10/crawl-ref/source/effects.cc: Increase chances of magic contamination Glow effects. 00:20:23 greensnark: Are you attempting to make 0.7 harder? :) 00:20:35 All of this is master :P 00:21:09 I will begin cherry-picks this evening or tomorrow, but I haven't decided on these 00:21:28 I will probably stick pretty strictly to bugfixes only for 0.7 01:02:13 -!- stabwound has quit [Ping timeout: 245 seconds] 01:25:36 -!- ortoslon has quit [Remote host closed the connection] 01:39:55 blargh, you didn't wait for the apply_to_all_dungeons() rewrite 01:40:15 it can be called a bugfix (since it fixes the massive lags), but it's intrusive 01:40:19 Neutral monsters can be enslaved, act funny (https://crawl.develz.org/mantis/view.php?id=1828) by OG17 01:40:41 my ISP went belly down again, and I do all development on a remote server :( 01:45:42 -!- ortoslon has joined ##crawl-dev 01:49:25 KiloByte: Go ahead and commit to master. I'll review and apply if it looks safe-ish 01:49:37 I'd like that fixed 01:50:16 hi KiloByte 01:50:25 the actual delayed actions (about finished) are safe-ish, keeping counts (which I wanted to finish yesterday evening) may break 01:50:28 !tell Cryp71c Please fix any outstanding DS issues/incomplete pieces by Jul 8. 01:50:28 greensnark: OK, I'll let Cryp71c know. 01:50:47 hi due! 01:51:16 UI fixed fulsome! 01:51:30 greensnark: Can you cherry-pick the fulsome commit, or is that not a bugfix? 01:51:54 due: In the evening 01:51:59 I will take it, yes 01:52:09 but "safe-ish" doesn't mean "doesn't move large amounts of code around", since most of attitude-change.cc is gutted away and the actual functionality moved to a new file 01:52:12 Excellent 01:52:13 It's not a bugfix, but it looked safe enough and it's nice to have 01:52:30 It's more of a "stop complaints" bugfix. 01:52:31 greensnark: hmm, what is your time zone this week? 01:52:32 KiloByte: Send it in and we'll see :) 01:52:42 KiloByte: Same as ever: +0530 01:53:36 Is there anything I can do? I'm kinda bored/annoyed. 01:53:36 !lg 01:53:39 1316. bookofjude the Thaumaturge (L9 SpVM), killed by miscasting Venom Bolt on D:6 on 2010-06-26, with 2240 points after 10772 turns and 0:48:45. 01:53:50 I had it at good. 01:53:53 well, that suggests you tend to sit to 8am or so then, that makes me confused :p 01:54:25 KiloByte: I often have to take telephone calls from contacts in the US at weird hours 01:54:48 ah 02:00:10 among the counters, I didn't implement one: NON-HOLY non-unholy non-evil friendly gift-of-a-good-god. Was that even ever possible? 02:01:26 Huh 02:01:31 Don't think so 02:02:02 The counter is just a count of number of god gifts in the dungeon split by type, right? 02:02:10 *god-gift monsters 02:02:26 no, in no cases it's just "god gift" 02:02:27 Or orc friendlies and similar 02:03:05 Ok, monsters that will be affected by renouncing gods and piety breakpoints 02:03:13 Any others? 02:04:30 cathegories needing a message: holy, plants, slimes, god-gift-of-Beogh (orcies+sorcerer summons), enslaved souls+undead slaves, spellcasters, unholy+evil, unclean+chaotic 02:05:06 Mmm, long list 02:05:28 Done with the changes? 02:05:32 plus a longer one of actions not needing a counter due to no message given 02:05:37 not yet 02:05:58 there are several non-monster actions too, but they don't really need a message 02:06:28 I mean, strictly speaking Nemelex shuffle does need one, but I kludged that around since the old code was broken anyway 02:07:59 it used to trigger on any deck existing anywhere in the world (including your inventory), no matter if shuffled or not. I changed it to "did Nemelex give you any gifts?" which is inaccurate as well. 02:08:26 but hey, it's "Nemelex chuckles", not "Nemelex shuffles your decks" 02:10:14 I guess I'll fix one of my own bugs. 02:16:51 * due shuffles KiloByte. 02:19:38 How do you determine if a mimic has been identified or not? 02:19:53 oh, MF_KNOWN_MIMIC. 02:22:53 mons_is_known_mimic() 02:24:05 ... even better. 02:26:12 greensnark or KiloByte: best behaviour for autotravelling into a mimic: attack it, or reveal it and stop? 02:28:57 Or... not. 02:29:17 You never get prompted about mimics even with !w. 02:31:27 We'll go with current behaviour then. 02:34:54 Who tagged 0.8-a? 02:34:59 indeed, revealing them would leak information 02:35:07 03due * re16e327aee95 10/crawl-ref/source/ (fight.cc travel.cc): Don't identify mimics with autoexplore: #1199. 02:36:14 due: snark did, so trunk version numbers don't say 0.7 anymore 02:37:40 ah, good 02:37:46 I'll need to update the CDO scripts then 02:37:59 or at least the CDo windows trunk builds 02:38:07 Napkin: does anything need to be done with the actual build scripts with a new version? 02:39:20 by the way, Jiyva slime conversion is guaranteed, except if it went hostile when you abandoned Jiyva. This gives you massive enemies+penance when you go back -- shouldn't it work like Fedhas? 02:40:42 for Beogh, it makes sense since the orc knows you're a turncoat, but neither slimes nor plants have sentience 02:42:46 Huh, we're missing descriptions for most of the unrands. 02:42:56 * due rolls up sleeves. 02:43:08 Guess I'll do Joseph and those. 02:44:31 due CDO will stay on the branch to be released, right? 02:44:47 it should, IMO 02:44:54 it did for 0.6 at least 02:45:05 yeah, to iron out all bugs 02:45:13 unless you guys want it different this time? 02:54:51 that works for us 03:01:37 the scripts need to be changed to track stone_soup-0.7 instead of trunk, though 03:19:06 -!- ogaz has quit [Remote host closed the connection] 03:23:01 Who came up wiwth Cekugob? 03:23:37 Cekugob is ancient 03:23:58 it used to have rF+ rC+ in addition to everything it has now ;) 03:25:03 :( 03:25:43 Cekugob made its appearance between 3.20f and 3.30f 03:25:51 So a linleyism? 03:25:53 what does the 'f' mean? 03:26:41 no clue 03:29:03 yes, KiloByte - beep me, when the branching is done? 03:29:12 Napkin: it's been done! 03:29:30 hehe, ok 03:35:47 ... ! 03:35:52 greensnark: Did you do Robe of the night? 03:38:18 > git --bare prune 03:38:18 Unable to open directory /var/cache/git/crawl-ref.git/objects 03:38:18 Unable to open directory /var/cache/git/crawl-ref.git/objects/pack 03:38:21 --- 03:38:23 shit 03:38:45 I don't understand how git can be so picky about too many files?! 03:41:01 crawl-ref.git/objects/pack> ls -1 | wc -l 03:41:01 2160 03:43:54 -!- ortoslon has quit [Remote host closed the connection] 03:50:17 * due smacks git and tells it to play nice wih the other children? 03:58:10 -!- syllogism- has joined ##crawl-dev 04:01:53 heh, Don Brodale breaks Linley's license: there are many versions without the accompanying source 04:02:18 Master branch on CDO updated to: 0.7.0-a1-0-g7949946 (27.0) 04:02:57 which is sad since binaries are useless for the history 04:02:59 dgl updated, 0.7 mirrored locally and updated "update-cdo-trunk.sh" to use it 04:03:38 windows binaries next 04:03:40 <3! 04:03:48 KiloByte: doesn't sorear have a lot of sources? 04:04:25 sorear: do you have anything that's not on ftp.dungeoncrawl.org? 04:05:04 only one such thing 04:05:09 the 4.1.2alpha sources 04:05:45 ever thought about making a historical git repository of crawl? 04:06:05 maybe if we had a more complete set of sources 04:06:25 it's a pain to add a newly-found version in the middle of a git repo 04:06:33 * sorear <3 the historical git repository of perl 04:07:30 yeah, rebasing helps, although not completely painless 04:08:18 35,686 revisions over 25 years 04:09:20 it can be done quite easily with reparenting, but that alters all hashes of subsequent versions 04:10:14 i thought dcss was based on 4.0b26? 04:10:16 KiloByte: you aren't required to accompany the binary with the sources, but you have the give it to those who ask . So he hasn't violated the license ... yet, as there is no limit of long you can ask for the sources :) 04:10:48 So we should be approaching Don for sources? 04:10:58 Yes, we should. 04:11:06 However he insists ##crawl-dev is empty. 04:11:14 sorear, no, he does not 04:11:18 He did come in here, actually 04:11:20 bhaak: no, it's not GPL 04:11:22 When? 04:11:27 so.. how about taking the very first revision of dcss as a separate branch and then merge 4.1.2 on top? 04:11:34 I seem to have missed that 04:11:42 Oh look, people being assholes in ##crawl. 04:11:44 * due grumbles. 04:12:10 sorear, 2010-05-11 04:12:50 bhaak: the license states that you need to either: a) include the source, or b) (non-commercial only) provide an archive site 04:12:53 He said, "< dbrodale> just updating colloquy to the proper channel", and quit two hours later. 04:13:54 "proper" channel? 04:14:01 He had #crawl-dev, I think. 04:14:04 yes 04:14:38 so we haven't really gotten him to stay and talk in any forum yet, yes? 04:16:20 it seems like that 04:16:29 sorear: he talked a bit in crawl-ref-discuss 04:17:43 KiloByte: but the version of 3.3.0 I looked at had a website url for the sources in crawl.txt. That's enough for the license. 04:18:23 although linley was still involved with that versions. A copyright holder can't violate his own license anyway 04:19:45 there's a shitload of other copyright holders 04:21:01 DBrodale's packages of 1.03 and 1.04 don't even name an URL at all 04:22:29 well, there must be a shitload of other copyright holders, but are they named anywhere in those packages? 04:22:44 it is documented that Linley did not use anything remotely resembling version control in those days 04:23:07 so the sources are lost for good 04:23:11 gnah, copyright law never foresaw collaborative work with $number_of_workers > 5 04:23:31 unless Linley has some tarballs, which I doubt 04:24:15 the copyright law was made for a single printer company, so the Crown could gather extra fees 04:24:42 originally, yes. But there have been some revisions of the laws since :-) 04:25:35 hmmm, someone built windows binaries already? 04:25:37 yeah, but it never included the concept of collaboration at all 04:25:54 merely someone taking an existing work and changing it 04:26:26 heh... from crawl.faq in most versions: "Crawl is covered by the same GPL licence as Nethack, [...]" 04:26:27 as far as we know, the oldest known release of the program hack made it somehow on the usenix tapes from the early 80ies, but nobody who got the tapes, got appropriate reader hardware. 04:27:13 bhaak, what about rogue? 04:28:38 did you, due? 04:30:33 Zaba: rogue and hack don't share any code. hack was only inspired by hack. I'm not sure what code has been released for rogue as it was also released commercially early on 04:31:13 ah, okay 04:31:26 well, I knew that they shared no code, but I wasn't sure about its commerciality 04:34:09 hmm, there was website that had old rogue versions on it, but I'm not sure if those were the original code or re-implementations 04:34:38 here: http://rogue.rogueforge.net/ 04:37:27 it's really original code: "This is the final version of the original (Toy, Arnold, Wichman) rogue developed for the UNIX operating system. The source code was recovered by Ken Arnold and posted on Sourceforge which I then copied and restored here." http://rogue.rogueforge.net/rogue-5-4/ 04:37:43 nice 04:39:01 Napkin: nope 04:42:59 I wonder how much it sucked 04:43:25 the highlight of my Crawl 1.05 restoration project was when I discovered TC4 didn't enforce unique parameter names 04:43:36 lol 04:43:53 (the function in question had about 60 parameters) 04:45:14 Global variables are bad, so the level generation function accepted parameters for mgrid, etc, etc, etc 04:45:41 No structs or enums to be found in *this* codebase 04:46:05 as those rogue versions are checked into version control, you can easily see how much needed to be changed to let them run on modern unix 04:47:27 I wouldn't be surprised if c compilers on unix sucked way less than c compilers on win (apart from being K&R C probably) :-) 04:49:14 can I !tell to multiple people? 04:50:02 sorear: you have the sources for Crawl 1.05? dungeoncrawl.org has only binaries for 1.03 and 1.04, and source for 1.10 04:52:00 -!- ortoslon has joined ##crawl-dev 04:53:30 KiloByte: I have what dungeoncrawl.org calls 1.10 04:53:40 the source code, however, claims to be 1.05 04:54:12 alright, windows binaries are 0.7 branch compatible too 04:54:28 due: your python script to update the website rules :D 04:54:36 Napkin: easily updated? 04:55:01 very :) 04:55:08 go check 04:56:26 :D 04:56:58 * due rules. 04:57:08 anyone got by's email address at hand? 04:57:22 please send to me via query 04:57:25 git log 04:58:07 searching for "by" in git log is difficult, sorear ;) 05:02:15 -!- Anym has joined ##crawl-dev 05:25:57 Napkin: Search for 'Robert Vollmert' in git log :) 05:26:00 Thanks for updating the scripts <3 05:26:16 due: There's always someone bikeshedding about something :) 05:27:29 beh, any idea why the travel cache uses its own versioning system? 05:28:41 the last update there is by greensnark from 2007-07-20, so it's probably because it's just ancient musty code, right? 05:29:39 I love reading SA. 05:29:43 They're all whinging about my uniques. <3 05:29:58 due: <3 05:30:05 dead players = <3 05:32:05 ... someone got an orc warrior band on D:1. 05:32:08 *AWEEOSME*. 05:32:31 due to farming, or? 05:32:45 nope 05:32:51 http://forums.somethingawful.com/showthread.php?threadid=3285755&userid=0&perpage=40&pagenumber=5 05:32:54 last post 05:35:10 -!- Mu_ has joined ##crawl-dev 05:38:27 that's WARLORD not warrior 05:38:54 er 05:38:57 that's what i meant 05:38:57 but a mere warrior would be no less insane, too 05:39:12 and it's a band, which are not supposed to ever spawn OOD 05:40:15 yeah 05:40:16 Somethings up 05:42:09 KiloByte: Yeah, tc is b26 vintage 05:46:49 I made it use common versioning. This does mean it will be dropped during this save upgrade, which will be a nuisance, but probably acceptable. 05:46:49 Yes, needed to be done sometime 05:46:49 KiloByte: We don't happen to have any entry vaults or minivaults in the D:1 range featuring 8, do we? :) 05:46:49 * greensnark checks. 05:46:50 -!- Ashenzari has quit [Ping timeout: 258 seconds] 05:47:35 -!- Ashenzari has joined ##crawl-dev 05:47:37 -!- The topic of ##crawl-dev is: DCSS development channel! | 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. 05:48:49 pretty sure that was fixed already 05:48:50 the post is a few months old 05:48:50 Oh, I didn't check the date 05:48:50 Was it fixed? 05:49:32 at least one was 05:49:48 omg! the first star trek movie in 1080p :-O 05:56:46 too bad, it's still star trek 05:57:47 what? should it rather be star WARS? 05:58:51 Lucas should have named Episode I "A Lost Hope" to go with "A New Hope" 05:59:05 lol 06:02:02 I wouldn't mind a "lucas died in 1990 before his brain was completely eaten digital edition" of star wars 06:13:44 -!- Anym has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.4/20100611143157]] 06:48:24 -!- purge has quit [Ping timeout: 252 seconds] 06:49:49 I'm going to bed. 06:49:52 greensnark, luck with CAO. 07:06:07 -!- Zaba has quit [Ping timeout: 276 seconds] 07:13:11 -!- Zaba has joined ##crawl-dev 07:15:17 -!- doy has quit [Ping timeout: 248 seconds] 07:15:19 -!- doy has joined ##crawl-dev 07:20:04 -!- greensnark has quit [*.net *.split] 07:20:04 -!- Twinge has quit [*.net *.split] 07:20:04 -!- ivan has quit [*.net *.split] 07:20:05 -!- Napkin has quit [*.net *.split] 07:25:58 -!- greensnark has joined ##crawl-dev 07:25:59 -!- Twinge has joined ##crawl-dev 07:25:59 -!- ivan has joined ##crawl-dev 07:25:59 -!- Napkin has joined ##crawl-dev 07:26:57 -!- ortoslon has quit [Remote host closed the connection] 07:31:15 -!- Zaba has quit [Ping timeout: 240 seconds] 07:38:31 -!- Zaba has joined ##crawl-dev 07:46:56 -!- daftfad has joined ##crawl-dev 08:08:22 -!- Poor_Yurik has quit [Ping timeout: 240 seconds] 08:15:29 -!- daftfad has left ##crawl-dev 08:21:54 -!- Textmode has quit [Quit: Over the ages, man has been certain of a great many things. some of these things were even true.] 08:30:53 -!- badp has joined ##crawl-dev 08:31:24 Hello. When playing on local crawl 0.7.0-a0-2647-gc830e21, I got a labyrinth with no loot even after killing the minotaur 08:31:37 ##crawl seems to suggest this is a bug, so I'm trying to collect data 08:31:56 I have the char dump, where does crawl store save files? 08:33:00 heh, the saves/ folder may be a start 08:33:08 yeah :p 08:33:19 a file named bp-1000.tar.gz, perhaps? 08:33:32 exactly 08:34:16 ??mantis 08:34:17 mantis[1/1]: The replacement for sourceforge, mantis can be found at http://crawl.develz.org/mantis 08:37:17 No loot in labyrinth (https://crawl.develz.org/mantis/view.php?id=1829) by bp 08:37:54 close as invalid please 08:38:02 the loot was indeed on the upstairs 08:38:10 Hehe. 08:38:16 can I get a potion of gain int maybe? 08:38:54 You can have a complimentary restore abilities. 08:39:02 I'll take that. 08:39:15 don't worry, at least your mistakes can make only your characters die :p 08:42:19 there is a fake minotaur vault, BTW, and it was what I suspected 09:10:21 -!- purge has joined ##crawl-dev 09:16:55 -!- ogaz has joined ##crawl-dev 09:21:14 -!- badp has left ##crawl-dev 09:22:10 -!- eith has joined ##crawl-dev 09:36:02 ahhh 78291 just used meph cloud to drown all the dire elephants.... how boring :| 09:43:56 -!- stabwound has joined ##crawl-dev 09:49:11 -!- eith_ has joined ##crawl-dev 09:52:17 -!- eith has quit [Ping timeout: 276 seconds] 09:52:33 -!- eith_ is now known as eith 09:57:43 In the hells, up stairs don't appear at the top in the map (https://crawl.develz.org/mantis/view.php?id=1830) by elliptic 10:20:13 -!- Eronarn has quit [Ping timeout: 260 seconds] 10:21:53 -!- Eronarn has joined ##crawl-dev 10:32:13 -!- ogaz has quit [Ping timeout: 265 seconds] 10:32:55 Thief God (https://crawl.develz.org/mantis/view.php?id=1831) by eliotn 10:35:57 -!- Zaba has quit [Ping timeout: 265 seconds] 10:44:06 -!- Zaba has joined ##crawl-dev 10:54:14 !tell greensnark Tiles TODO list now current: https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:tiles:tiles_missing_for_0.7 10:54:14 Enne: OK, I'll let greensnark know. 10:54:23 Yay 10:54:23 greensnark: You have 1 message. Use !messages to read it. 10:55:14 Enne: feel free to merge tiles commits to 0.7 as needed 10:55:20 Will do. 10:56:05 I was really hoping to get around to changing the feature (and tile) for the level 1 up staircases, just to be more clear for new players. 10:56:31 Oh, players try to go up and get confused? 10:56:33 ...but I think the "adding new dungeon features" ship has already sailed. ;) 10:56:48 Yes, I'd really prefer not to break saves in 0.7 :) 10:57:11 Yeah, don't get me wrong--I have no intention of doing it right now. There's always 0.8. 10:57:19 Speaking of saves, it's time we fixed Crawl going zonko when the disk is full 10:57:34 Just writing to foo.tmp and renaming to foo ought to solve that for us 10:57:48 * greensnark will do this. 10:57:53 That's a good idea. :) 10:57:54 -!- ogaz has joined ##crawl-dev 10:58:11 What happens now? 10:59:12 Crawl overwrites the old .sav file and doesn't have space to write the new file 10:59:23 Same for any level files it wants to write 10:59:27 Oh, yuck. 10:59:49 I spent a fair bit of time this evening deleting 0 byte files for various players :P 10:59:56 Not fun 11:02:59 D:1 upstairs should be like Zot portals, but require the orb for passage :P 11:03:05 Autoexplore does not stop when entering LoS of Roxanne/oklob until they actually do something (https://crawl.develz.org/mantis/view.php?id=1832) by elliptic 11:07:25 Enne: spriggan assassins and warpers are merely enums; assassins are 0.8 material but require significant coding, warpers are probably not going to ever exist 11:07:50 Thanks, KiloByte. 11:08:31 Oh, you edited the wiki page too. Even better. :) 11:08:37 Enne: and defenders and the Enchantress are coded but never used. Knowing our Sprint guys, this means it is likely they will be there, but it's not like you can play tiles on CAO... 11:09:39 You can still select sprint from the tiles version. 11:09:58 I guess I need to add the sprint-only uniques to that list too. 11:10:27 hmm... this raises an issue: we probably should have SOME maps for the release 11:10:33 not just the tournament 11:10:57 Yes 11:11:26 Are Sprint maps supposed to be selectable by the player 11:11:32 There's a menu, right? 11:12:03 * greensnark would like to skip Sprint for tourney altogether. :P 11:12:09 It would save a lot of tourney scripting :P 11:12:42 I don't get a menu. 11:21:16 iirc there's a menu there in code, not sure if it's called 11:25:28 KiloByte: Do you have more than one sprint map? 11:25:59 I'm not sure what the idea is for the tournament: will the old map be disabled? 11:29:59 03greensnark * r62c85002ffac 10/crawl-ref/source/ (misc.cc mon-util.cc): [1832] Fix i_feel_safe ignoring stationary monsters like oklobs and Roxanne. (elliptic) 11:48:26 03enne.walker * reca484c69b55 10/crawl-ref/source/ (3 files in 3 dirs): Add hyperactive ballistomycete tile. 11:56:00 03enne.walker 07stone_soup-0.7 * r670a7d393971 10/crawl-ref/source/ (3 files in 3 dirs): Add hyperactive ballistomycete tile. 11:58:59 03due 07stone_soup-0.7 * r0656fd9e38bd 10/crawl-ref/source/ (spells4.cc spells4.h spl-cast.cc spl-cast.h): Allow victory dancing Fulsome Distillation. 11:59:01 03greensnark 07stone_soup-0.7 * r66b591c9e849 10/crawl-ref/source/mon-act.cc: Give monsters a better HD save vs mephitic cloud. 11:59:03 03greensnark 07stone_soup-0.7 * r2d8ba46d6b48 10/crawl-ref/source/ (misc.cc mon-util.cc): [1832] Fix i_feel_safe ignoring stationary monsters like oklobs and Roxanne. (elliptic) 12:06:08 Master branch on CDO updated to: 0.7.0-a1-4-g2d8ba46 (27.0) 12:21:08 * KiloByte really hates keyboard ports that stop working without warning. 12:22:19 a second normal keyboard didn't work, had to run to work to borrow an USB one which does 12:23:32 Artemis the Hunter Goddess (https://crawl.develz.org/mantis/view.php?id=1833) by gerryq 12:32:01 -!- Poor_Yurik has joined ##crawl-dev 12:48:40 Non evil necro spells should be allowed by good gods. (https://crawl.develz.org/mantis/view.php?id=1834) by casmith789 12:53:42 Spriggan tile (https://crawl.develz.org/mantis/view.php?id=1835) by coolio 13:10:11 -!- Chapayev has joined ##crawl-dev 13:10:30 We have an interesting crash on Chapayev's system: http://pastebin.org/360485 13:10:55 A static initialisation crash in skills2.cc 13:11:39 Anybody have ideas? It looks quite weird to me :) 13:15:19 Chapayev: Did 0.6 crash as well? 13:15:53 greensnark: i didn't get as far as trying it--i'm compiling 0.7 now, though 13:16:07 0.7 is practically identical to master, so you shouldn't see any diff 13:16:12 oh :P 13:16:53 http://www.acc.umu.se/~zao/cdo-hi.png 13:17:15 Looks like the highlight is slightly off, either due to some race or me mourning for a good while on the "You die..." prompt. 13:17:25 Zao: Looks like an off-by-one, can you file? 13:18:03 Could also be some other process sneaking in, yes 13:18:12 This is on CDO, so probably. 13:19:15 According to ##crawl bot spam, exactly one character with a higher score than me died inbetween. 13:19:43 Hrm no, two. 13:21:28 I'm not sure if I'm imagining it, or if this is actually much different than before, but Xom doesn't seem to do things very frequently right now 13:21:41 st_: The tension code was changed 13:21:50 700 turns between the thing he just did and the one before, and then 1800 turns to the one before that 13:22:00 Possibly he need to be made to act more often in general to compensate 13:22:47 st_: Please file a bug and mention the tension change 13:23:24 greensnark: 0.7 segfaults too, should i post the backtrace for that one or just move on to 0.6? 13:23:42 Chapayev: Hang on, let's try some minor variations 13:23:46 okay 13:25:31 Chapayev: Try http://sprunge.us/FCUa and recompile 13:25:41 Apply with curl http://sprunge.us/FCUa | patch -p3 from source dir as before 13:25:55 Just make debug again should be enough to recompile, don't clean :) 13:31:43 greensnark: still segfaults, want a backtrace? 13:31:51 Why not 13:32:11 It looks like you have something wrong with your setup though 13:33:01 http://pastebin.org/360517 13:33:18 greensnark: that wouldn't surprise me! i would like to know how to fix it though 13:34:51 Chapayev: Did you apply the last patch? 13:35:10 Just git checkout skills2.cc and apply with 13:35:12 curl http://sprunge.us/FCUa | patch -p3 13:35:40 yeah, i applied that patch 13:35:54 Because line numbers don't seem to match up 13:37:11 -!- Zaba has quit [Ping timeout: 276 seconds] 13:38:41 greensnark: okay, i just did it again (git checkout skills2.cc and then did curl and patch) and got a different line number in skills2.cc after make debug and getting a backtrace 13:38:45 http://pastebin.org/360528 13:39:03 i think i pasted the wrong thing into pastebin last time? not sure 13:39:07 sorry :P 13:39:41 -!- Zaba has joined ##crawl-dev 13:42:05 actually I'm not sure, he seems to be acting pretty frequently now, maybe it's just because there's not much to cause tension early on 13:42:30 Xom not going crazy on you in the early dungeon might be a good thing, I dunno 13:46:24 Chapayev: I'm stumped :) 13:46:34 We should get a Debian type to look at it, maybe KiloByte 13:47:01 greensnark: yeah, i'm not married with debian, i was planning on switching to a different distro in about a week, maybe now i'll do it sooner :P 13:47:06 married to debian, i mean 13:47:07 You didn't have anything obviously wonky with your setup, did you? 13:47:19 greensnark: not at all 13:47:48 greensnark: like i said it's a pretty clean install 13:48:13 the biggest change is that i am running dwm as a window manager but that shouldn't affect crawl 13:48:40 The crash is in trying to initialize a std::string from "Adj" in _skill_title_keys 13:48:47 Which really shouldn't be crashing afaik :) 13:48:53 yeah no kidding! 13:48:54 weird 13:51:31 0.6 segfaults too 13:52:33 and 0.6 is known to work on a large number of systems 13:52:56 it's totally feasible that i did something really stupid and weird that is causing this problem, but i have no idea what it is 13:53:33 including all Debian architectures, both release and toy (like hurd-i386) 13:53:58 heh, debian hurd, amazing :) 13:54:05 so it suggests something must be borked in your system, but it's a bit hard to see what 13:55:33 Neunon's zoology system used to detect it as Mac (since the check was for kernel being "mach"), but that's fixed now and it compiles out of the box :p 13:55:33 i'm trying it from the source code tarball linked to from crawl.develz.org now, maybe i screwed up using git somehow 13:56:05 do you have the sources for some other big C++ program you could test? 13:56:26 KiloByte: no, but i could download one and try it 13:56:33 Could this be caused by taking a bunch of lib updates without a reboot? 13:56:44 Don't see that it should be 13:57:16 Chapayev: You said you used libncurses from somewhere else, right? 13:57:28 not really, gcc and libs are loaded every time you start a process 13:57:39 greensnark: yeah libncurses is from the squeeze (debian unstable) repositories 13:57:52 most other packages on here are from the lenny (stable) repositories 13:58:21 I don't really see how that could be a problem either... 13:58:26 yeah me neither 13:59:04 Have tiles inventory provide more information. (https://crawl.develz.org/mantis/view.php?id=1836) by Krishh 13:59:04 anyway dpeg asked me to do some stuff with new sprint maps so i would rather call this a loss, wipe the hd, and install fedora or something so i can compile crawl and work on some sprint stuff 14:02:30 okay yeah it's not a problem with git, i get the same segfault when i run crawl compiled from the 0.6.1 source code on sourceforge 14:03:23 do you guys want any more information about this crash before i wipe the hd? 14:04:16 Can't think of anything 14:04:26 But I'm really curious as to what's causing it :) 14:04:44 Why don't you try a hello-world C++ program and see if it builds and runs 14:04:57 one with a std::string in it 14:05:44 yeah i have compiled and run c programs with no problems, i guess it needs to be something c++ specific like a std::string 14:06:04 or if something is borked on the disk, perhaps apt-get --reinstall install foo would help 14:06:15 Chapayev: Try building http://sprunge.us/hNRU 14:06:45 KiloByte: But what would foo be in this case :) 14:06:50 for x in `dpkg --get-selections|grep '\binstall$'|cut -f 1`;do apt-get --reinstall install $x;done 14:07:06 that's a blunt instrument, though 14:07:14 for x in `dpkg --get-selections|grep '\binstall$'|cut -f 1`;do apt-get --reinstall -y install $x;done 14:07:36 Chapayev: Try KiloByte's incantation right after you try building that code :) 14:08:01 gpp test.cc 14:08:01 bash: gpp: command not found 14:08:09 it will take a while, though, since it fetches and reinstalls every single package on the system 14:08:09 Chapayev: g++ 14:08:10 :PPPPPPPPPPPP 14:08:29 I forgot how to check the consistency using already installed tools 14:08:33 oh duh, sorry :( i use c not c++ when i have the choice 14:08:55 like the md5sum files in /var/lib/dpkg/info/ 14:09:18 greensnark: your test program compiles and runs normally 14:09:41 Well, that's about what I expected :) 14:10:52 md5sum -c /var/lib/dpkg/info/*.md5sums|grep -v 'OK$' 14:11:15 (cd /;md5sum -c /var/lib/dpkg/info/*.md5sums|grep -v 'OK$') 14:11:43 since these don't include the / at start 14:12:01 KiloByte: okay that's running 14:12:16