00:33:12 -!- Abuse has quit [] 00:45:45 after server disconnect, gecko no longer acts (https://crawl.develz.org/mantis/view.php?id=1438) by Tenaya 01:17:14 -!- eith has joined ##crawl-dev 01:21:59 -!- Henzell has quit [Ping timeout: 258 seconds] 01:29:20 -!- Henzell has joined ##crawl-dev 01:39:18 -!- by has joined ##crawl-dev 01:42:04 -!- eith has quit [Ping timeout: 268 seconds] 01:53:48 -!- Henzell has quit [Ping timeout: 258 seconds] 01:56:51 -!- Henzell has joined ##crawl-dev 01:59:29 !tell dpeg Cryp71c failed to understand the auxiliary unarmed code quite badly, and doesn't even acknowledge it. Don't trust him with probability analysis, even if he presents his numbers confidently... 01:59:29 by: OK, I'll let dpeg know. 02:00:40 D: 02:41:54 -!- Iainuki has quit [Read error: Operation timed out] 02:42:01 -!- Twinge_ has joined ##crawl-dev 02:43:39 -!- Twinge has quit [Read error: Connection reset by peer] 02:44:54 -!- ortoslon has joined ##crawl-dev 02:45:08 -!- Iainuki has joined ##crawl-dev 03:25:59 -!- dpeg has joined ##crawl-dev 03:26:11 mornings 03:26:27 dpeg: You have 1 message. Use !messages to read it. 03:30:27 !messages 03:30:28 (1/1) by said (1h 30m 58s ago): Cryp71c failed to understand the auxiliary unarmed code quite badly, and doesn't even acknowledge it. Don't trust him with probability analysis, even if he presents his numbers confidently... 03:30:45 by: you around? 03:31:46 yes; couldn't keep quiet after reading your "wow, you're good at this" in the log :) 03:32:27 mutation choice is not a long if/else, it's a perfectly fine weighted choice 03:32:38 there's no disadvantage to mutations at the end of the list 03:32:53 (unless I missed some bugs) 03:33:53 speaking of the stinger mutation, I'd give it a weight of at least 10 03:34:34 for something that's limited to just two rarely played races, it shouldn't be an exceedingly rare mutation. Weight 3 is hardly better than 1. 03:39:55 by: well, it is a bit more opaque to have a chain of conditional probalities than a list of plain chances? 03:40:14 KiloByte: I am also fine with 10, but would prefer to do this is several small steps. 03:44:44 do you have a clue what's the matter with Summon Dragon? In the earliest version in CVS, it was a half-implemented Xom-only spell, and that's still the case today. 03:46:12 dpeg: mutation-data.h has a plain list of weights; chance for a mutation is weight / (sum of weights of eligible mutations) 03:47:28 by: ah, thanks. 03:47:42 by: another question: did you implement the GDR boost? 03:51:17 hi 03:51:21 what a busy day! 03:53:41 dpeg: yes, I think that's in 03:53:55 29c85040c8629 03:57:16 39 emails, sigh 04:02:55 heavy armour getting a nice boost? 04:03:05 by: cool. Yesterday, I told players in ##crawl that we improved heavy armour. 04:03:09 cbus: yes 04:03:23 dpeg! What do you think of my tutorial changes? 04:03:26 due: 39? When I came back yesterday I had 100. 04:03:27 I'm not done yet, but I'm making headway. 04:03:42 due: they're awe-some! Didn't see you yesterday, would have told you right away. 04:03:57 :) 04:04:11 Yeah, was v. busy. Likewise today! 04:06:50 felirx: HAve you maade any commits yet? 04:07:30 dpeg, anywhere close to how armour used to be or? 04:09:05 cbus: it's completely different. GDR used to be marginally important (both players and developers thought too highly of it). Now, we have made it matter (more, at least). The boost is that you get GDR for your armour (no matter what type, but heavier is better, of course) regardless of Armour skill. In other words, that D:3 plate mail is as heavy and encumbering (sp?) as before, but you'll enjoy the full GDR right away. 04:09:35 due: Janne is busy with another coding assignment... something about Java IIRC. 04:09:44 ahh ok, and is there still this need x strength for y armour? 04:09:56 cbus: of course. That's a good idea, in my opinion. 04:10:01 dpeg: Ah 04:10:10 dpeg: Was going to update the CIA namelist, but no matter 04:10:26 Don't want any random weakling to wear plate. 04:10:37 I agree that str for armor is good, though I do think the penalties are too sharp and should be more gradual. That's something that can be looked at down the road though 04:10:57 Twinge_: sure. If you have proposals, tell us! (Not just IRC please, ideas get lost here.) 04:11:04 -!- Twinge_ is now known as Twinge 04:11:14 due: thanks for that. There seems to be an issue with c-r-c as well, but it's not that urgent, I hope. 04:11:47 Nat usualy fixes it up 04:11:50 *usually 04:12:03 Nothing offhand; I'd need to look into it more to figure out how I'd like to change it. Time is a little less free these days. 04:12:36 Anyone do MATLAB here? Just have a simple question about array dereferencing. 04:14:45 ixtli: no, last time I did Matlab was 17 years ago. Thank god. 04:15:15 One of my girlfriend's friends is having an issue with an array out-of-bounds error. 04:15:21 And I don't really know the language. 04:15:23 0-indexed? :) 04:15:27 yeah 04:15:28 HAH 04:15:31 that must be it 04:15:32 actually no 04:15:46 When in doubt, solution = 0-indexed. 04:15:57 Or the complete opposite, as in Lua, where it is 1-indexed and you try to loook at table[0]. 04:16:01 For the first item. 04:16:03 ARGH. >:| LUA >:| 04:17:28 Who changed the drop menu? 04:17:33 Ah, Johanna I guess. 04:17:45 There's no weight information in the top line anymore. 04:18:51 due: Is there an interactive debugger for MATLAB? 04:19:19 Is it known that autoexplore sometimes refuses to work? "Found a hand axe." ad infinitum... 04:19:19 is there an option to turn off the new cursor, btw? 04:19:30 due: I agree, though. Lua's array indexing is crazypants. 04:20:56 ixtli: I've never used MATLAB 04:21:13 Heh sorry. 04:21:15 :) 04:21:17 That was meant for dpeg. 04:21:20 Oh 04:21:22 You guys both have d-names! 04:21:24 >_< 04:21:26 I knoooow 04:21:32 it's more awesome with doy and myself thoguh 04:21:47 It /is/ more awesome. 04:22:09 -!- dpeg is now known as deg 04:23:01 :> 04:23:49 Okay 04:23:53 Now I have to go do other stuff 04:26:02 -!- ortoslon has quit [Quit: Leaving.] 04:26:47 Is Chapayev's sprint patch (the one which suppresses noise and teleportation scrolls) already in trunk? 04:29:20 don't bother, mail sent anyway 04:29:43 o 04:29:45 no 04:29:50 -!- ixtli is now known as dix 04:29:57 IT DOESN'T WORK FOR ME 04:30:04 * dix sobs 04:30:40 how about replacing the cursor so that " a - a short sword" becomes ".a - a short sword"? 04:31:03 by: sure, that's an option. The cursor is too heavy right now? 04:31:14 if you don't use it, definitely 04:31:45 -!- dix is now known as ixtli 04:31:48 the ".a" would mean that that item can be selected by . and a, which is nice, but maybe non-obvious 04:31:56 Having another nic is weird. 04:32:05 but hardly less obvious than not hinting at the use of . at all 04:33:50 also, shouldn't the cursor be on "a" at the start? 04:34:24 by: yes, it should. 04:35:00 I've been buggering jpeg to do a second round on the drop/pick up (including a help screen I've sent to her). She's back after some hiatus, so don't take action right now, please. 04:35:11 don't intend to 04:36:05 by: I am not sure that . and ' are the best keys, however. What about Down (moves down without selection, but scrolls like now), Right (moves down with selection, like . now)? If we do this, there'd be the obvious extensions to Up and Left, too. 04:36:25 the weights work fine as long as you don't use . 04:37:22 ah 04:38:04 I got a flail of distortion ... anyone knows when we allowed that to appear again? 04:45:49 don't see anything in the code or history that looks like it was ever disallowed 04:46:07 blunt *vampiricism* was disallowed 04:47:23 ah, so there was only ever talk about disallowing blunt distortion 04:47:27 fine with me 04:47:33 re the selection keys, I don't have an opinion; since I know where the letters are on the keyboard, it seems the feature is not that useful to me 04:47:34 !seen KiloByte 04:47:34 I last saw KiloByte at Tue Apr 27 09:35:14 2010 UTC (12m 20s ago) saying !learn edit clarity[1] s/confusion and forced berserk/ on ##crawl. 04:47:58 by: yes, the selection feature is decidedly for non-touch typers. 04:48:38 KiloByte: the new stat death code is in trunk, right? Did anyone get feedback about that... Don't want to lose my char to Int 0 right now :) 04:48:40 I imagine it could still be useful when you're stashing, but I've learned to go through the alphabet 04:49:12 deg: you should test that -- don't raise int 04:52:43 seems kilobyte didn't merge his branch 04:53:11 so all you get for zero stats is some paralysis and slowing for the moment 04:56:11 -!- ortoslon has joined ##crawl-dev 05:00:06 KiloByte: there was support for merging... 05:02:44 by: I merged only some penalties 05:03:30 int: no casting or reading scrolls, str: halved CC, 1/4 regen, dex: no stealth, stabbing or EV 05:04:01 KiloByte: there are more/other penalties in the branch? 05:06:20 staring at the wall with int=0 (effectively losing further 50% turns), but the code would have to be rewritten, including a change to saves (or reordering a bunch of other code) if we want to not penalize free actions, which we obviously want 05:07:09 other proposed penalties were crippling melee damage with str=0 and crippling to-hit with dex=0 05:07:35 but mere penalties are not the biggest issue 05:08:40 KiloByte: what's the biggest issue? 05:08:57 currently you get a vague message, get slowed and those smallish penalties. There is absolutely no hint you'll magically die out of a sudden after a while. 05:09:06 KiloByte: ah; I just checked for HEAD 05:10:01 we could just remove death completely 05:10:06 -!- ixtli has quit [Quit: ixtli] 05:10:10 KiloByte: we should print all of "Int: 0" in the same colour as Starving. 05:10:19 -!- Spads has quit [Ping timeout: 248 seconds] 05:10:45 by: that is true 05:11:06 assuming that stat 0 is very undesirable (which it is already, I think). 05:12:59 if we're to remove death, we'd have to ensure the penalties are really crippling 05:15:09 yeah 05:15:37 I like the idea that zero stat is similar to starvation.... but I may be overly biased here. 05:16:15 zero stat should shut down healing 05:16:17 it would not be hard to give messages and paralysis similarly to the way starving is done 05:17:09 yes 05:19:59 sorear: currently, only str does that, and only reduces to 1/4 instead of 0. That's not a bad idea. 05:20:13 1/4? eew 05:21:34 deg: perhaps even with 05:22:09 mm, HTMLCrawl 05:23:40 sorear: \e[5m -- it predates HTML by many, many years 05:24:24 xterm and gnome-terminal render it in a different way, by changing the foreground and background colors instead. Which is good too. 05:27:42 -!- Spads has joined ##crawl-dev 05:30:26 Sigh, I am unable to see the comments in the SF tracker. 05:30:33 https://sourceforge.net/tracker/?func=detail&aid=1928259&group_id=143991&atid=757516#commentTip 05:31:39 Can you access the comments? (Ignore the #commentTip at the end.) 05:33:18 do you see a line "Comments ( 14 )"? click on it doesn't do anything? 05:33:57 No, it doesn't :( 05:34:03 bhaak: also, hi! 05:34:14 then you have to view it with lynx :) 05:35:49 hi deg! which browser do you use? I can't reproduce that even with js switched off 05:49:36 KiloByte: does http://sprunge.us/AgBI 05:49:41 look good to you? 05:51:15 at that point time_taken is >0, right? 05:51:31 yes, it's the end of world_reacts 05:53:33 -!- Spads has quit [Ping timeout: 276 seconds] 05:54:10 -!- Spads has joined ##crawl-dev 05:55:07 end? There's a lot of places where world_reacts() returns without reaching the end. 05:55:35 (unless I'm confusing it with the other big function) 05:56:52 I don't think world_reacts returns at many points 05:57:41 right, it's _input() which does that 06:13:14 -!- syllogism has joined ##crawl-dev 06:13:40 -!- by has quit [Quit: Lost terminal] 06:26:29 bhaak: I am using konqueror, but via a ssh -X connection 06:30:03 deg: I can't imagine that ssh -X is the problem (as you're connecting to a server even when running it locally). maybe they used javascript that doesn't work on konqueror. if you disable javascript, you should see the comments 06:32:51 bhaak: thanks, works now 06:55:49 -!- ortoslon has quit [Read error: Connection reset by peer] 07:17:21 -!- by has joined ##crawl-dev 07:38:33 -!- ortoslon has joined ##crawl-dev 07:41:44 bah 07:42:09 deg :) 07:42:38 (: 07:43:01 <3 Denzi going to do tiles for the startup menu. 07:45:51 yes 07:46:01 due: Paul filed a bug report! 07:47:07 Paul? 07:47:18 Paul Paul? 07:47:49 That's cool. 07:48:36 Bah git gc. 07:51:28 Paul Du Bois, professional game coder. 07:51:40 He wrote the quiver code and added the HP/MP bars. 07:52:05 He dropped from development for time reasons, I am very happy he still (or again) plays the game. 07:52:17 :( 07:52:21 due: ? 07:57:36 Sad that he dropped out. 07:57:51 And I couldn't find my headphones, sorry. 08:02:13 03due * rc69456e8f199 10/crawl-ref/source/dat/des/tutorial/tutorial.des: Further tutorial tweaks. 08:08:44 due: maybe he'll slip back in :) 08:11:41 wait wait, Paul Du Bois works at Double Fine? 08:13:01 Paul DuBois as in the MySQL writer, or someone else? 08:13:25 probably not the Double Fine guy :P 08:13:43 but he said professional game coder 08:13:45 I do not know 08:13:54 Paul Du Bois (Senior Programmer, Double Fine) 08:14:01 Oh man, really? 08:14:11 Didn't they do Psychonauts? 08:14:13 yes 08:14:35 Rax was talking about Bru"tal Legend. 08:15:39 deg: is that him? 08:17:33 can't be :P 08:19:02 syllogism: yes 08:19:11 -!- Amonchakad has joined ##crawl-dev 08:19:20 Psychonauts, too 08:19:35 Why can't that be? 08:19:43 Has anyone observed yet that the Amulet of Guardian Spirit borders the "totally useless" zone? 08:19:52 Er 08:20:02 You aren't doing it right. 08:20:08 well it's just weird I hadn't heard about it 08:20:08 Amonchakad: the amulet is overpowered 08:20:14 it is? how so? 08:20:17 If you start on about how it's useless for casters, I'll hate you. 08:20:22 Forever. 08:20:31 It's useless for pretty much everyone. but that's my opinion, ofc 08:20:47 Fighters: oh look, in exchange of the amulet slot I get an extra 2-4% hp 08:20:53 and the inability to use any activated ability 08:20:58 Amonchakad: you're missing several important points. 08:21:02 Casters: plain useless. 08:21:09 Yeah well, I came here to learn them 08:21:16 no time, sorry 08:21:19 Crud 08:21:43 just two words: Vehumet; invocations; =magical power 08:22:07 Sif Muna's channeling. 08:24:44 Uhm...so, with Vehumet, you get back a couple MP every kill 08:24:50 I still fail to see how that helps 08:24:55 We have used it. 08:25:32 The amulet is very strong for certain builds and possibly overpowered even. 08:25:39 No, your TrFi will not use it. 08:26:51 I obviously came to the wrong place asking for info; apologies 08:27:09 Try ##crawl :) 08:28:01 Yes, you should find someone on ##crawl who has put "GS to good use. 08:28:15 dolorus added SUMMON_SWARM to the list of applicable ghost spells 08:28:22 09-11-18... 08:28:32 It's not implemented and a pan lord got it and crashed a game :) 08:28:41 :( 08:28:48 https://crawl.develz.org/mantis/view.php?id=1439 08:28:48 guardian spirit comibned with channeling effects can be nice 08:28:53 I'll replace with summon butterflies 08:28:59 Crash during orb run, after seeing random pan lord (https://crawl.develz.org/mantis/view.php?id=1439) by MarvintheParanoidAndroid 08:29:22 i don't think CDO has been updated in a while, incidentally? 08:29:43 i got demonic guardian from !mut earlier in my game, and that was marked as fixed on mantis a while ago 08:29:59 ah 08:30:10 I'll update 08:30:19 oh, by the way; could we get an update to the latest trunk for Tiles too? the last one has been on 13 april 08:30:31 does input work in trunk tiles win now? 08:32:41 I think there's problems with that 08:32:47 03due * r206561aada8e 10/crawl-ref/source/ghost.cc: Replace Summon Swarm with Summon Butterflies [#1439]. 08:33:04 -!- ais523 has joined ##crawl-dev 08:33:23 -!- Amonchakad has quit [Quit: Page closed] 08:33:32 It's compiling, will take a few minutes to update 08:33:38 loo break meanwhile 08:33:51 I'm combiling trunk tiles win right now to see if it works 08:33:51 my loo breaks all the time, too 08:35:00 Hm 08:35:08 files.cc: In function ‘std::string get_prefs_filename()’: 08:35:08 files.cc:867: error: ‘struct game_options’ has no member named ‘player_name’ 08:35:11 make: *** [files.o] Error 1 08:35:13 make: *** Waiting for unfinished jobs.... 08:35:37 Napkin: Ping! 08:35:52 Looks like DGL builds don't work any more. 08:36:41 by: Ping, you did changes to Options.player_name recently? It's needed for DGL, I think. 08:36:45 Anyway, I must sleep. 08:36:55 Poke Marc to update, he should be around. :) 08:38:57 tiledoll.cc:52: error: no matching function for call to 'stat::stat(const char*, stat*)' 08:39:00 on so en 08:39:02 *on 08:42:06 Napkin is playing Nethack these days. 08:42:35 what? not unnethack! that bastard! 08:44:22 !seen Napkin 08:44:22 I last saw Napkin at Mon Apr 26 21:47:38 2010 UTC (15h 56m 44s ago) saying !tell LordSloth Hint and photo about ssh compression added. thanks again for the screen capture :) on ##crawl. 08:45:44 is halberd two handed? 08:47:00 deg: yes 08:50:07 can someone close this https://crawl.develz.org/mantis/view.php?id=1409 it was just fixed 08:54:48 st_: will do 08:54:51 03dolorous * r3c5f833ad732 10/crawl-ref/source/dat/des/variable/ (altar.des mini_monsters.des): Update comments to account for demonic crawlers' glyph change. 09:06:39 -!- Spads has quit [Ping timeout: 260 seconds] 09:07:23 03j-p-e-g * rb20a49768870 10/crawl-ref/source/ (4 files in 4 dirs): Add tutorial training dummy tile. 09:12:44 -!- Spads has joined ##crawl-dev 09:14:00 KiloByte: I've been completely overhauling the page while you commented... 09:17:58 -!- ais523 has quit [Read error: Connection reset by peer] 09:18:12 -!- ais523 has joined ##crawl-dev 09:26:41 03dolorous * r42cec4d9a335 10/crawl-ref/source/mon-cast.cc: Implement Swarm as a monster spell. 09:26:43 03dolorous * r3510dad131f7 10/crawl-ref/source/ghost.cc: Add Summon Swarm back to the list of ghost spells. 10:00:26 -!- MarvinPA-2 has joined ##crawl-dev 10:00:56 -!- MarvinPA has quit [Ping timeout: 245 seconds] 10:03:22 -!- MarvinPA-2 is now known as MarvinPA 10:11:51 03by * rae5148397e2b 10/crawl-ref/source/ (files.cc files.h newgame.cc): Quarantine save game checking code in files.{cc,h}. 10:11:53 03by * r931c3e11d20f 10/crawl-ref/source/ (9 files): Unify startup menu with rest of character choice. 10:11:55 03by * r31fcf35416ef 10/crawl-ref/source/ (initfile.cc options.h): Remove obsolete Options.prev_game. 10:11:56 03by * re2788e7cd710 10/crawl-ref/source/files.cc: Update DGL to use Options.game.name instead of Options.player_name. 10:14:03 -!- shockwav3 has joined ##crawl-dev 10:14:22 -!- shockwav3 is now known as shockwave 10:15:04 felirx: stat code in files.cc is inside #ifdef HAVE_STAT and calls ::stat() instead of just stat() 10:16:32 it breaks windows compile for whatever the reason 10:16:50 Napkin: crawl should build for DGL again (at least the bug due ran into before should be fixed) 10:19:08 felirx: you may just want to #include 10:19:17 it used to be included in the old tilereg.cc 10:19:17 -!- ais523_ has joined ##crawl-dev 10:19:23 -!- ais523 has quit [Disconnected by services] 10:19:26 -!- ais523_ is now known as ais523 10:23:15 it compiled 10:23:27 mouse works, keyb not 10:24:07 -!- eith has joined ##crawl-dev 10:28:09 now we just need to make everything mousable 10:28:24 kiinda hard to type in a character name 10:28:27 needs a virtual keyboard! 10:28:45 crawl should have a random name generator anyway 10:29:02 random / viable 10:29:32 Welcome, mr. Ahuabambooman. 10:30:51 just use the pan lord name gen 10:34:11 -!- by has quit [Quit: Lost terminal] 10:44:52 -!- ixtli has joined ##crawl-dev 10:45:26 Good evening :) 10:46:29 Hi! 10:46:36 :D 10:46:38 oh god 10:46:46 deg: you're still in nic confusion mode 10:46:56 I just feel left out, since mine turns into dix 10:46:59 T_T 10:51:17 deg: can you review a patch for me? 10:51:20 it's to my code 10:51:36 It's small. 10:54:51 ixtli: I am not a coder :/ 10:54:55 -!- deg is now known as dpeg 10:55:08 how about dpeg 10:55:11 is dpeg a coder? 10:55:12 ^_^ 10:55:31 dpeg has coded some, but is amazingly unfamiliar with the Crawl code 11:01:10 -!- Zaba has quit [Ping timeout: 252 seconds] 11:03:29 -!- Zaba has joined ##crawl-dev 11:04:00 -!- ixtli has quit [Ping timeout: 276 seconds] 11:09:56 dpeg: probably the only way to stay sane 11:10:15 I felt ill after trying to sourceread bits of Crawl, I'm glad people are trying to improve it 11:16:17 I've heard some say Crawl's code is relatively clean and sane? 11:16:57 if you call 7 nested ternaries clean and sane. 11:18:02 Siber, the newest parts are better, yes 11:18:26 the older parts are the programming equivalent of goatse 11:18:40 you read them then wish you hadn't 11:19:24 but the current devteam's aware of the problem, and trying to make it better 11:22:21 03dolorous * ree31212ea4b7 10/crawl-ref/source/startup.cc: Fix typo. 11:44:37 well, at least it's better than nethack's code 11:44:38 (: 11:44:49 doy: I disagree 11:44:55 i'm sure you do 11:44:57 although maybe I was just reading a particularly bad bit 11:45:00 -!- shockwave has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539]] 11:46:29 * bhaak doesn't mention both don't have anything even resembling a testsuite or code coverage suite :-) 11:46:32 felirx: nested ternaries.. like (X ? Y : Z) where both Y and Z are each nested ternaries? or where just Z is another ternary expression 11:46:39 bhaak: crawl has tests 11:46:43 (and i suppose where X is also a ternary) 11:46:51 not very much coverage, but still 11:47:25 i find long chains where only Z is ternary-recursive to be quite readable (if newlined+indented properly) 11:47:35 * doy agrees 11:48:11 certainly better than the equivalent if-else structure would be... 11:48:18 a test does not a suite make :) 11:48:39 i write python that needs to be 2.4-compliant, i miss a real ternary expression all the time 11:50:45 thankfully I haven't found a ternary with both y and z being another ternary this far :) 11:50:45 -!- Madtrixr has quit [Read error: Connection reset by peer] 11:51:02 -!- Madtrixr has joined ##crawl-dev 11:51:56 -!- gal_bolle has joined ##crawl-dev 11:57:16 -!- Madtrixr has quit [Ping timeout: 252 seconds] 12:08:59 03j-p-e-g * r2497b14b2c06 10/crawl-ref/source/dat/des/tutorial/tutorial.des: A few minor tutorial tweaks. 12:08:59 03j-p-e-g * r8ef1d8e1441c 10/crawl-ref/source/dat/des/guide.txt: Update des/guide.txt to the sprint and tutorial folders. 12:10:13 yay 12:12:19 bhaak: it is of course trivial to write a test coverage tester for NetHack, just do something along the lines of int main() {printf("0\n"); return 0;} 12:13:44 ais523: "0.01% coverage is coverage, too" or what do you mean? :-) 12:14:11 bhaak: I mean that coverage testing is where you analyse the test cases in order to see how much they're actually testing 12:14:19 because NetHack has no test cases, coverage-testing it is therefore trivial 12:16:42 ais523: you can also do non-automatical code coverage 12:17:15 (only make sense in edge cases [which nethack is not one]) 12:24:01 -!- ixtli has joined ##crawl-dev 12:32:06 -!- Xiberia has quit [Ping timeout: 240 seconds] 12:32:51 -!- Cryp71c has joined ##crawl-dev 12:38:25 Morning. 12:43:51 Finished my rewrite of player_aux_unarmed, pending a bit of cleanup, anyone care to take a look? 12:45:06 -!- LordSloth has joined ##crawl-dev 12:50:23 When autoexploring with a weapon on !a it asks me whether I want to "attack" a mimic even though I can't tell it's a mimic (https://crawl.develz.org/mantis/view.php?id=1440) by casmith789 13:03:14 -!- Smee has joined ##crawl-dev 13:17:36 -!- Smee has quit [Ping timeout: 245 seconds] 13:21:58 -!- Smee has joined ##crawl-dev 13:36:44 no one is in at the moment? 13:36:56 I am 13:37:50 -!- Smee has quit [Quit: Leaving] 13:38:20 elly!!!! :) 13:38:26 dpeg: hey! what's up? 13:38:36 nuffink :) 13:39:45 so why are you asking? 13:40:52 dpeg, I am 13:41:03 -!- Henzell has quit [Read error: Connection reset by peer] 13:41:23 cao dying 13:42:04 or maybe not 13:42:15 it had a huge hickup there, IDK if it crashed or not. 13:42:45 bhaak: b/c there's been no activity for an hour (and I was away, too) 13:43:45 gosh, peope have lives beside rl :) 13:44:04 weirdos 13:44:11 I just sit here 24/7 and stare at ##crawl-dev 13:44:45 the benefit of having multiple monitors, I can actually do that! 13:45:43 elly: recently, I had to read elly-away. It broke my heart. 13:46:11 dpeg: what? I have never been elly-away 13:46:14 I have been elly|silent though 13:51:50 * elly disappears! 13:55:17 -!- Gernony has joined ##crawl-dev 13:58:28 My map suffers from maprot... and autoexplore won't work, occasionally. Anyone seen something similar? 14:00:11 Hello. 14:01:26 bhaak: is that besides real life, or rogue likes? 14:01:31 Dear ##crawl-dev, please please please fix the Crawl memory leak. <3, -r. 14:01:52 -!- Henzell has joined ##crawl-dev 14:02:24 If there's anything I can do to help, let me know. 14:02:24 rax: You have 1 message. Use !messages to read it. 14:02:26 LordSloth: I wondered the same. 14:02:33 I haven't found core files in the past, but I will check again. 14:02:35 rax: is this a well-known bug? 14:03:03 LordSloth: depends on the developer :D 14:03:27 It's a bug that I have seen like 50 times. I have another corefile for it now, I've saved it aside. 14:03:52 You're going to ask me next if it's in the tracker, and I am going to have no clue, because I have not even seen the new tracker, because I am basically a stop energy elemental. 14:04:02 But Crawl keeps eating six gigs of RAM and it's really hosing my server. 14:04:46 Okay, I'll try to ressurect Darshan then. 14:04:56 I know he was the last person to look at it. 14:05:34 dpeg, no, it appears nothing is in mantis about it. no results for 'memleak' 'leak' or 'memory' of any pertinence. 14:06:59 Does CDO have similar problems? Or hehfiel? 14:07:34 I don't recall it happening, but then again it's not thaaaat common 14:09:24 Napkin may have seen it once or twice, but not as regularly as I do. This is the first time I've seen it in 0.6. In theory it might be an 0.5 problem still? Though I don't think anyone was playing 0.5 at the time. 14:10:56 !lm * cv=0.5 src=cao 14:12:28 276994. [2010-04-27] firemonkey the Chopper (L3 MiBe) killed Jessica on turn 979. (D:2) 14:12:58 So it might be a 0.5 problem? 14:13:21 CDO had it quite a few times 14:13:29 don't remember when it started... 14:13:59 rax, would you like the ulimit patch for dgamelaunch? 14:14:07 Napkin: also recently? 14:14:11 that's the only thing that helped here 14:14:12 doesn't seem very likely considering how few people have been playing 0.5 14:14:44 not anymore, dpeg, since darshan and me put the ulimit (limits resources allowed to take, like memory etc) thingy into dgamelaunch 14:16:17 Napkin: rax is very busy, can you help her with the ulimit patch? 14:17:07 sure, but busy at the moment too 14:17:16 it's an easy +10 lines patch 14:17:46 will sent it tomorrow, busy at the moment too o/ 14:18:29 okay, so I won't send my mail to Darshan right now 14:18:44 shouldn't it be in Crawl itself? 14:19:16 10 lines... do you have something more than #ifdef DGL setrlimit(...); #endif ? 14:19:34 anyone know off hand if you can generate an ssh key for sorceforge so it doesn't prompt for password on each push? 14:21:09 Cryp71c: iirc, you need to import the ssh key on the sf website (in your profile or whatever) 14:22:25 -!- Abuse has joined ##crawl-dev 14:23:09 anyone feel like changing the reproducibility status on this bug: https://crawl.develz.org/mantis/view.php?id=993 14:23:16 i can do it every time. 14:24:33 Abuse: just add a note in the tracker about this 14:24:38 I am off for today. 14:25:04 Abuse, its an easy fix, I would do it but I havn't had the time. the most recent note on the tracker clearly indicates its reproducable. 14:25:23 -!- dpeg has quit [Quit: foood] 14:25:41 k 14:29:37 03Cryp71c * rf7fe397fb965 10/crawl-ref/source/ (mutation-data.h player.cc): Stealth bonus to thin skeletal structure 14:32:05 Napkin: I may or may not be able to take your patch, I am running old dgamelaunch. 14:32:16 I really need to fix this at some point, but probably not until I've moved to Indiana. :/ 14:34:10 -!- ixtli has quit [Quit: ixtli] 14:36:33 But if the patch applies I would take it. 14:36:46 That would be a fine solution. 14:37:28 Or honestly if you had a dgl source tree I could build that would fit in with my existing chroot and dgl-login file (versus database) I'd find the time to install that in order to stop having this memory leak. 14:46:59 rax: woo indiana 14:50:56 rax: you either insert a setrlimit() call in any random place in Crawl's startup, or run ulimit beforehand 14:51:54 KiloByte: I'm not comfortable doing that on my own without understanding what I'm doing. 14:53:21 -!- Madtrixr has joined ##crawl-dev 14:53:25 I'm not actually confident in "add an X call" into source code, and "man ulimit" doesn't return anything. I do see a cpulimit package in debian but I expect that's different. 14:54:25 it's in "man bash", since it must be a shell built-in (it wouldn't have any effect if set in a forked process) 14:54:47 I'm not sure that it would work in this case (shell script calling a crawl process). 14:55:00 But, see, this is why I can't just do it, because I don't know what it is. :) 14:56:33 rax: man 2 setrlimit is the appropriate manpage 14:56:36 if it's a shell script that calls crawl, you can use ulimit just fine 14:56:48 but it's a bit messy having to use it, I think 14:57:41 I don't care about messy. 14:57:48 I care about "I can do it in five minutes." 14:57:56 If you can give me a line to put in a shell script, I will put it there. 14:58:28 Right now there is "ulimit -c unlimited" in the shell script to get coredumps, which greensnark put there. 14:59:04 Would I want -l for total memory limit/ 14:59:10 Or do I also need a separate limit for swap? 14:59:23 Or maybe I actually want -d. 14:59:50 I don't know, and I don't have a testbed to find out in, or four hours to go find some bash book and read the chapter and look up all the words I don't know. 15:00:18 "Learn things half-assedly" is unfortunately a skill I don't have; I'm not good at being something other than very good or very bad at something. And I know how to change states, but it takes time. 15:02:34 re 15:02:39 don't worry, rax 15:02:58 it's really a most easy patch and fits to all versions of dgamelaunch 15:03:35 and doing those limits in dgl's source is better than the bash ulimit command 15:04:02 -!- syllogism has quit [] 15:05:10 http://git.develz.org/?p=dgamelaunch.git;a=commitdiff;h=36c21aada28ec5076d0821430149421157722ba7 15:05:40 that patch should apply to the dgl on cao cleanly 15:06:29 (click the "raw" link at the very top to have the patch as a text file) 15:07:14 ...oh, so I can't just wget it. :P 15:07:22 I was like "XML? The crap?" 15:07:31 you can :-P http://git.develz.org/?p=dgamelaunch.git;a=commitdiff_plain;h=36c21aada28ec5076d0821430149421157722ba7 15:08:13 it'll set the maximal coredump size to 150mb and maximal memory to 100mb 15:08:51 oooh, wait, wait 15:08:53 nonono 15:08:58 you can't do that, sorry 15:09:01 I forgot 15:09:19 sorry, no, it needs to be done in the shell script 15:09:46 otherwise some huge ttyrecs of players won't load properly 15:10:06 Ah. 15:14:17 And I had just manually merged the patch, too. :P 15:14:17 darshan rewrote the whole ttyplayback function because of that 15:14:17 arg! sorry :-$ 15:14:17 ulimit -S -c 153600 2>/dev/null 15:14:17 ulimit -S -v 102400 2>/dev/null 15:14:17 those two in the shell script should have the same effect 15:17:05 Added those, thank you! 15:19:12 And to crawl5-caller as well. Hopefully that will fix it. 15:20:34 i strongly believe so :) 15:20:44 03dolorous * r24671d041883 10/crawl-ref/source/mutation-data.h: Add punctuation fixes. 15:24:14 -!- ixtli has joined ##crawl-dev 15:33:12 -!- eith has quit [Quit: sleep] 15:46:37 -!- Gernony has quit [] 15:57:23 03j-p-e-g * r1f41d08f0edb 10/crawl-ref/source/rltiles/ (55 files in 3 dirs): Add more of Denzi's dungeon tiles. 15:57:30 03j-p-e-g * rbfeeec6d31be 10/crawl-ref/source/dat/des/portals/ossuary.des: Use sandstone floor tiles in the ossuary portal vaults. 16:36:08 idea: expand shopping list to be an "items of interest list" with items left on the floor 16:52:09 -!- ais523 has quit [Remote host closed the connection] 16:52:52 morning 16:53:34 Napkin! did you get to update CDO ok? 16:56:41 Imaginary eating haiku (https://crawl.develz.org/mantis/view.php?id=1441) by Kyrris 16:57:18 ...what? 17:00:51 -!- Cryp71c has quit [Quit: Leaving] 17:05:45 -!- ortoslon has quit [Read error: Connection reset by peer] 17:21:38 Okay... Imaginary Eating Haiku? 17:22:12 Oh wait, I think there might actually be a point, though Kyrris does his best to obscure it. 17:22:12 I have no idea. 17:23:33 I think the point he's making is that herbivore3 gets the same message as lesser herbivores, who can stomach a little meat. 17:23:40 Also, it is natural crawl-haiku 17:24:47 But it is uninformative at a glance. 17:25:18 If a more sane ticket was created, I'd suggest he meant that you should get a different message from being totally unable to eat something, then you would get eating something poorly. 17:25:38 It's a great way to hinder productive discussion without trolling, however. 17:26:01 Just imagine a world where all bugs and FRs were submitted in haiku. 17:26:32 * Madtrixr shudders 17:28:43 Following Kyriss 17:29:32 I'd sooner close down the tracker. 17:29:42 sorear/doy: i see you have entries on termcast, do you happen to do crawl dev work on there or something, or is it for walking people through code, or.. 17:29:43 May make our heads explode 17:29:59 Haiku is too much 17:32:02 -!- gal_bolle has quit [Ping timeout: 240 seconds] 17:32:18 due: close it? :-( 17:32:46 Posting obnoxious bug reports is obnoxious. 17:33:05 Especially in friggin Haiku. 17:37:53 Better: Patches with code formatted into haikus 17:49:40 i think limricks are more attractive than haiku 18:05:58 Master branch on CDO updated to: 0.7.0-a0-1501-gbfeeec6 (22.0)! 18:17:58 -!- casmith789 has joined ##crawl-dev 18:18:53 -!- casmith789 has left ##crawl-dev 18:19:52 Where'd travel stoppers go? 18:22:05 !tell by http://crawl.develz.org/morgues/trunk/bookofjude/crash--20100427-232031.txt 18:22:06 due: OK, I'll let by know. 18:22:08 -!- mspang has joined ##crawl-dev 18:23:21 -!- LordSloth has left ##crawl-dev 19:02:00 philsnow: yes I do crawl dev there from time to time 19:02:20 philsnow: it's nice when other people can occasionally point out me doing stupid stuff 19:30:17 ?? the orriflame is being lowered 19:30:17 I don't have a page labeled the_orriflame_is_being_lowered in my learndb. 20:09:54 philsnow: I'm ono termcast too, but that's just me playing scribble. 20:11:34 awesomeness 20:12:03 they just don't make stuff like that these days much 20:12:10 who doesn't? 20:12:17 I have to teach it new words 20:21:19 fun fact: I was first introduced to Crawl by doy on termcast 20:21:24 back in Dec 08 IIRC 20:26:53 :D cool 20:26:59 Two sevens in one game, not bad 20:30:56 Take that, computer, 464! 21:02:50 ??reflection 21:02:50 reflection[1/2]: A shield brand. The shield will reflect blocked things back the way they came from. 21:03:39 -!- Lemuel has joined ##crawl-dev 21:14:54 -!- Lemuel has quit [Ping timeout: 265 seconds] 21:20:38 -!- bmh has joined ##crawl-dev 21:20:52 should guardian spirit be marked as useless for DDs? 21:29:34 no 21:29:37 because it is not 21:29:46 Channeling + GS = free HP for Dd. 21:30:33 oh? When I don it without channeling, my MP just goes to 0 and my HP doesn't budge 21:39:14 -!- Enne has joined ##crawl-dev 21:42:29 exactly 21:42:37 if you have no use for the mana, you just saved some HP 21:42:59 bmh: potion of magic = free potion of mhealing 21:43:07 makhleb + "GS is... immensely overpowered on DD. 21:43:25 if you're melee 21:43:58 why would you be using GS if you rely heavily on mp and have no channel / crystal ball effects? 21:44:30 if this game had manapools in the traditional sense, where a mage type could have ten times more mana than HP, i'd see the use 22:10:12 :) 22:10:17 * due goes swimming at the manapool. 22:14:08 * due is disappointed to discover it's mana-pool, not man-apool. 22:17:05 -!- bmh has quit [Quit: bmh] 22:19:25 -!- Abuse has quit [Ping timeout: 240 seconds] 22:54:54 -!- Abuse has joined ##crawl-dev 23:04:43 -!- Enne has quit [Quit: zzz] 23:28:47 Into Sewer, out of Sewer, back into Sewer. Persistent portals (https://crawl.develz.org/mantis/view.php?id=1442) by Kyrris 23:29:52 -!- neunon has quit [Ping timeout: 246 seconds] 23:33:07 -!- Abuse has quit [Read error: Connection reset by peer] 23:35:19 -!- Abuse has joined ##crawl-dev 23:53:17 -!- neunon has joined ##crawl-dev