00:00:29 Unstable branch on CRAWL.XTAHUA.COM updated to: 0.20-a0-1062-g7b787ac (34) 00:06:44 -!- Guest80352 is now known as bh 01:20:20 !lg * 01:20:22 7178905. tylertfb the Slasher (L2 DsGl), slain by an adder on D:2 on 2017-04-27 05:19:55, with 21 points after 667 turns and 0:02:29. 01:33:06 somehow my scoreboard has found more games 01:33:08 Processed 7530000 games... 01:33:22 !lg * sprint 01:33:22 632266. Spurious the Charmwright (L1 HOSk), succumbed to a water moccasin's poison in Sprint (Sprint IX: "|||||||||||||||||||||||||||||") on 2017-04-27 05:31:39, with 6 points after 95 turns and 0:00:37. 01:33:22 !lg * !sprint 01:33:22 632266. Spurious the Charmwright (L1 HOSk), succumbed to a water moccasin's poison in Sprint (Sprint IX: "|||||||||||||||||||||||||||||") on 2017-04-27 05:31:39, with 6 points after 95 turns and 0:00:37. 01:33:30 !kw sprint 01:33:30 Built-in: sprint => game=sprint 01:33:39 !lg * x=game 01:33:40 7178932. [game=] GavinH the Magician (L1 MuWz), slain by a giant cockroach (kmap: minmay_heptomino_w4) on D:1 on 2017-04-27 05:32:09, with 6 points after 229 turns and 0:01:01. 01:33:43 !lg * game!= 01:33:44 undefined method `+' for nil:NilClass 01:33:51 !lg * game!='' 01:33:51 undefined method `+' for nil:NilClass 01:33:55 !lg * game!="" 01:33:56 undefined method `+' for nil:NilClass 01:34:00 !lg * s=game 01:34:01 ERROR: column logrecord.game does not exist 01:34:10 !lg * x=game 01:34:11 7178932. [game=] GavinH the Magician (L1 MuWz), slain by a giant cockroach (kmap: minmay_heptomino_w4) on D:1 on 2017-04-27 05:32:09, with 6 points after 229 turns and 0:01:01. 01:34:28 maybe something about it going to an entirely different table 01:34:49 !lg * game=zotdef 01:34:50 38509. AlexThePrettyGood the Caller (L7 DESu), slain by Prince Ribbit in Zot (Scenario V: Glassed Sandwich : zotvault5) on 2017-04-27 00:24:10, with 541 points after 1171 turns and 0:13:45. 01:34:58 those each have their own logfiles 01:35:03 yeah 01:35:05 so I think you can't treat it like a normal field 01:35:15 but zotdef, sprint and "" are probably the only possible values 01:36:23 https://www.reddit.com/r/dcss/comments/67oo8n/given_their_description_shouldnt_formicid_be_able/?st=j1zz6koi&sh=5209ec03 this is an interesting suggestion 01:47:27 "I've pitched that before in this subreddit and was met with an inexplicable amount of resistance." 01:47:52 gammafunk are you there? 01:50:46 i think it's a sort of frivolous change, but it does differentiate formicid further 01:51:15 thanks for reverting the monk changes btw 01:51:23 whoever was involved in that decision 01:53:06 alexjurkiewicz: it's also a lot of added code and interface complexity for what amounts to a flavour change 01:54:04 and means that now every gloves ego/unrand/etc. design has to account for "what if the player has two of them" 01:55:44 finger necklace and Op work because the code, interface, and design were already written around multiple rings being used at a time 01:56:14 and even then they still required quite a bit of work 01:57:13 just imagine the new design possibilities though 01:57:31 chestplate of hands 01:57:42 you can wear another pair of gloves on it 01:57:54 cloak of feet 02:01:46 i need help, build about https://crawl.develz.org/wiki/doku.php?id=dcss:server:setting_up_dgamelaunch_and_webtiles 02:01:51 https://docs.google.com/document/d/1dKzz2PPccP-OrnOM--8cRYtdO-J_gfWo_cFW2pLCnKk/edit 02:02:10 above url is current my status 02:02:56 ASCIIPhilia: told me I have to request access, which I did 02:03:00 but I guess you have to approve 02:03:06 https://docs.google.com/document/d/1dKzz2PPccP-OrnOM--8cRYtdO-J_gfWo_cFW2pLCnKk/edit?usp=sharing 02:03:16 i miss it setting public 02:03:19 haha.. 02:04:24 ASCIIPhilia: you are missing development packages 02:04:47 ??install 02:04:47 install[1/1]: See the following for installation and compilation: https://github.com/crawl/crawl/blob/master/crawl-ref/INSTALL.txt 02:04:48 crawl git INSTALL.txt prerequires? 02:04:50 yes 02:04:55 oh 02:05:00 you need to be able to compile crawl on that machine 02:05:03 so all those packages you need 02:06:10 ASCIIPhilia: the messages about "already on szorg" is fine, it's telling you that it's already on that branch 02:06:20 minmay: there are only two randart gloves and neither have compatibility issues 02:06:30 aha 02:06:33 I'm a bit worried there's a slippery slope to giving formicid 4 rings though 02:06:50 is default branch szorg? 02:06:55 probably it is 02:07:03 hrm 02:07:07 there is a master 02:07:10 so maybe not 02:07:15 maybe you had checkout out szorg already? 02:07:20 either way it's fine 02:07:25 szorg is the branch that you want 02:07:37 ah I see 02:07:42 the command already has you clone that branch 02:07:52 so there's no need to check out out afterwords 02:08:18 don't need 'git checkout szorg' cmd? 02:08:52 correct, because the git clone command had "-b szorg" 02:09:02 this means it will check out szorg already 02:09:11 basically the instructions should be corrected 02:09:16 anyhow it's no problem 02:09:45 ah, okay 02:09:53 dgamelaunch git download directory Does it matter anywhere? 02:10:18 i downloaded at /home/crawl-dev/ 02:10:43 yes, that's a good place 02:10:48 okay 02:12:11 ASCIIPhilia: by the way, what's the purpose of your new server? Is it mainly a development server? 02:12:37 probably yes :) 02:14:45 I want to have it. my dcss server that fully same public server spec. 02:15:20 For now, probably I will use it for development purposes. 02:16:29 sounds good 02:16:49 For fun after a long day of doing military exercises :) 02:17:44 draft D-26 02:17:47 ㅠㅠ 02:18:23 (TT) 02:21:35 hmm;; despite i installed all of prerequisites, it print same error 02:21:45 maybe it doesn't see them 02:23:34 ASCIIPhilia: the ./autogen.sh command doesn't work? 02:23:39 yep 02:23:45 crawl-dev@fec1b9a7a79b:~/dgamelaunch$ ./autogen.sh --enable-debugfile --enable-sqlite --enable-shmem 02:23:46 Generating configuration files... 02:23:46 ./autogen.sh: 6: ./autogen.sh: autoconf: not found 02:23:46 ./autogen.sh: 7: ./autogen.sh: autoheader: not found 02:23:46 ./autogen.sh: 10: ./autogen.sh: ./configure: not found 02:24:31 ASCIIPhilia: apt-get install build-essential 02:24:38 already installed 02:25:06 is this the same server for which you gave me a login? 02:25:06 if that still works, I can look 02:25:22 but I think build-essentials gets autotools 02:26:58 wait i will setting portfoward 02:27:48 ASCIIPhilia: one thing you can try is to install the autoconf package 02:28:12 apt-get install autoconf 02:28:15 yeah 02:29:38 oh, I guess you changed servers 02:29:47 before you were using CentOS 02:30:01 no it is centos 02:30:06 i use just docker 02:30:17 oh 02:40:43 oh, it work after i installed autoconf. :) 02:42:18 where is dgamelaunch binary path? 02:42:44 is it generated after make install? 02:47:57 make VIRUS error 02:48:03 i updated at google docs 02:48:19 *VIRUS=1 02:50:38 i wonder if it's called VIRUS because the original author hated vi 02:51:10 hah 02:51:25 dgl-common.c:17:20: fatal error: curses.h: No such file or directory 02:51:26 but i already install libncursesw5-dev (follow INSTALL.txt prerequires) 03:15:23 EyeSore (L11 HOMo) ERROR in 'dungeon.cc' at line 453: invalid map border at (14,0): granite_statue (D:10) 03:24:39 Unstable branch on crawl.beRotato.org updated to: 0.20-a0-1062-g7b787ac (34) 04:53:12 -!- Rjs- is now known as Rjs 05:01:08 -!- amalloy_ is now known as amalloy 06:22:10 Yo 06:22:27 bad interaction with shambling mangrove and dragon form 06:23:31 casting dragon form while rooted by Shambling Mangrove results in the "You fly out of the water as you turn into a fearsome dragon!" text but you do not in fact fly out of the water 06:23:45 you are still grounded 06:30:00 -!- amalloy is now known as amalloy_ 09:25:12 <|amethyst> !tell advil I don't like that ensure_valid_actor includes a baked-in const cast 09:25:13 |amethyst: OK, I'll let advil know. 09:25:38 <|amethyst> !tell advil IMO there should be two versions, one with argument and return const and the other non-const 09:25:38 |amethyst: OK, I'll let advil know. 09:26:49 well, that's doable, but I tried to figure out if there was a problem with that and as far as I could tell returning const vs non-const is a bit vestigial in current c++ 09:26:49 advil: You have 2 messages. Use !messages to read them. 09:26:50 <|amethyst> !tell advil also, I would consider using static_cast<> instead of C-style casting here, just to make sure const isn't accidentally casted away 09:26:50 |amethyst: OK, I'll let advil know. 09:26:53 <|amethyst> doh 09:27:03 <|amethyst> advil: that would be char * const vs not 09:27:16 <|amethyst> advil: const char * isn't returning a const, it's returning a pointer to const 09:27:39 <|amethyst> also, why is it a member function? 09:28:14 it's static, which I guess is unusual in crawlcode 09:28:14 advil: You have 3 messages. Use !messages to read them. 09:30:13 but why would it not be a static function in actor? 09:31:15 <|amethyst> it's simpler to be a free-floating function 09:31:25 hmm 09:31:42 in what way? 09:32:01 <|amethyst> don't have to write actor:: to call it from outside the actor class 09:32:14 <|amethyst> and it seems like the kind of thing you might want to call from beam.cc for example 09:34:09 I realize this is not the crawlcode aesthetic & already a lost battle but my preferences are for more typing with cleaner namespaces 09:35:19 <|amethyst> perhaps if it is to be static, ensure_valid 09:35:24 <|amethyst> since actor:: is already in the name 09:35:46 <|amethyst> I am sympathetic with cleaning namespaces 09:35:47 that works 09:35:59 I did consider that but then had some reason for not doing it, can't remember what 09:35:59 <|amethyst> though one could put it in a namespace instead of making it a static method 09:36:20 <|amethyst> and then people can even 'using' it if they want to reduce typing 09:36:32 I vacillated on nearly every aspect of that function (and actually read up on the cast types too) 09:36:39 <|amethyst> don't touch it just yet, I'm working on the const correctness 09:38:33 <|amethyst> Oh, and also, you should have /** or /// on doxygen-formatted comments; that's how doxygen knows to look at it 09:38:49 ah 09:38:53 yeah, I always forget the extra * 09:48:37 03|amethyst02 07* 0.20-a0-1063-gb8e8b7e: Provide a const version of actor::ensure_valid_actor. 10(6 minutes ago, 3 files, 15+ 7-) 13https://github.com/crawl/crawl/commit/b8e8b7e09fab 09:48:45 <|amethyst> advil: okay, pushed, do what you will with names and/or namespaces :) 09:50:03 <|amethyst> advil: I had to change that line in monster.cc to const actor *valid_agent = ... 09:50:51 <|amethyst> advil: but then I autofied it (so we don't have to change it if hurt and m_d_fineff change to take non-const agent) and put it in the if condition 09:51:01 yeah, that looks godo 09:51:04 *good 09:51:12 so I see how this is more explicit, but is the behavior different? 09:51:43 <|amethyst> advil: if you have an actually-const actor (like one defined as const monster blah{...} in function scope) 09:51:48 <|amethyst> this one isn't undefined behaviour 09:52:05 <|amethyst> also, it won't accidentally remove constness when the caller didn't intend to 09:53:11 <|amethyst> (actually-const as opposed to just being accessed though a const pointer, which would be the case almost always, since it is kind of rare that we have an actor that isn't either you() or in the mons array 09:53:16 <|amethyst> ) 09:53:24 <|amethyst> s/, which /; the latter / 09:54:38 oh I see 09:54:47 <|amethyst> ha, found something amusing 09:56:02 <|amethyst> checkwhite's "french"-spacing fix (replacing double with single space after sentences) only works when the following sentence starts with a capital letter 09:56:09 <|amethyst> which makes sense, but 09:56:19 haha I'm sure that has missed some cases for me before 09:56:32 <|amethyst> we have an awful lot of French spacing before sentences that start with function names or with digits 09:57:06 <|amethyst> We even have a double-space that can appear as output of the program (but not "in-game") 09:57:15 <|amethyst> printf("Fsck starting. %u chunks, ...") 09:57:44 btw also in the realm of dubious fixes did you see the change I made to _wu_jian_number_of_attacks? 09:58:28 <|amethyst> advil: yeah, seems reasonable 09:59:06 <|amethyst> unwind_var is a horribly nasty hack, but it's a horribly nasty hack that makes me smile 10:00:06 yeah I like it 10:00:17 I've been itching to use that somewhere ever since I first saw it 10:00:41 <|amethyst> I do agree that the attack_delay method just needs to change, and also that changing it is somewhat bug-prone because of the number of places it is used 10:01:01 I personally think unwind_var isn't itself a hack, it's just that any use of it is necessarily a hack :-) 10:01:08 <|amethyst> yeah 10:01:14 <|amethyst> that's what I meant :) 10:01:39 <|amethyst> I guess mostly it reminds me of Lisp or (old) Perl 10:01:53 <|amethyst> where dynamic scope was the norm 10:02:30 <|amethyst> s/Lisp/(old) Lisp/ 10:03:00 I haven't used lisp since college 10:03:22 oh that reminds me, I was wondering if you know why ASSERT uses do { } while (false)? 10:03:27 <|amethyst> me either :) 10:04:23 I guess it ensures a more restricted set of syntactic contexts than just { } ? 10:04:45 <|amethyst> {} would fuck up one rare case 10:04:50 <|amethyst> if (blah) 10:04:54 <|amethyst> ASSERT(foo); 10:04:56 <|amethyst> else 10:04:57 <|amethyst> bar 10:05:03 oh, yeah 10:05:05 <|amethyst> without braces 10:05:57 <|amethyst> see http://stackoverflow.com/a/154138 10:06:05 <|amethyst> for a full explanation 10:07:14 "Why is it that I feel worse off as a programmer after learning all this..." 10:08:34 Resting in desolation of salt takes much more time than anywhere else, is that "normal" ? 10:08:47 <|amethyst> Alarkh: it's known 10:09:14 <|amethyst> Alarkh: turns out various calculations that have to be done for opaque clouds every turn are kind of slow 10:09:29 Unstable branch on crawl.jorgrun.rocks updated to: 0.20-a0-1063-gb8e8b7e (34) 10:09:32 <|amethyst> I wonder if we could avoid LOS calculations out of player sight 10:09:45 <|amethyst> or more than 10 squares away or something 10:09:49 <|amethyst> let monsters fake it 10:10:32 Los is irrelevant for monsters outside of player interaction anyway, unless i miss something 10:10:56 it is, but cloud creation leads to a lot of rapid los changes that have to be calculated at some point 10:11:19 |amethyst: I thought part of the issue was that the lua cloud code itself is slow 10:11:30 <|amethyst> that too 10:11:45 PF had an (incomplete?) wip c++ version of some of that, it's in a branch 10:11:54 <|amethyst> I profiled it a while back but now I don't remember 10:13:42 Oh, and now that it happens, I sometimes loose focus on crawl while playing in Firefox (ctrl+f opens firefox search instead of crawl search for instance) and i don't know how to get the focus back on crawl, so i have to restart 10:14:08 Is that known / is there some way to get focus back to crawl ? 10:14:23 (I'm not sure how to reproduce it though) 10:14:24 there's a bug with that report, we've actually had some trouble exactly replicating it 10:14:38 https://crawl.develz.org/mantis/view.php?id=10938 10:14:39 can you add your browser / system info there? 10:15:06 I tried to replicate it in a bunch of browsers on OS X and just couldn't, but it definitely happens to some people sometimes 10:16:31 That's indeed how it breaks 10:16:33 also https://crawl.develz.org/mantis/view.php?id=10318 10:17:21 I added a note 10:17:25 thanks 10:18:28 Alarkh: which server? 10:18:34 Xtahua 10:20:04 I also have the bug with the ! and spectator joining 10:23:42 -!- Reverie is now known as Catiua 10:59:41 -!- amalloy_ is now known as amalloy 11:28:30 what is rfk, atc, boggle? 11:28:42 You might want to remove the entries for rfk, atc, and boggle. If you do, also edit the menus under chroot/data/menus/ You definitely need to change the uid= and gid= lines in config.py (the webtiles config) to match the numeric IDs of the 'crawl' user. Likewise shed_uid and shed_gid in dgamelaunch.conf. edit config.py 11:41:20 -!- amalloy is now known as amalloy_ 12:32:02 haha wow, this focus locking bug is triggered by resizing too 12:32:27 I think I figured out why it's happening, but a fix isn't entirely trivial 12:34:04 ASCIIPhilia: I think they're other games you can run under dgamelaunch 12:34:09 rfk = robotfindskitten 12:36:38 aha 12:37:25 thanks for your explain :) 13:02:21 -!- amalloy_ is now known as amalloy 13:24:47 Unstable branch on crawl.akrasiac.org updated to: 0.20-a0-1063-gb8e8b7e (34) 13:56:44 -!- Catiua is now known as Reverie|Away 14:59:22 -!- Reverie|Away is now known as Catiua 17:51:18 !cmd randomgame 17:51:18 Command: !randomgame => !lm * alive -random $* time>${now()-interval('10minute')} x=src fmt:'$(lookup ${x[0]} (watchhash))${name}' 17:51:34 !cmd randomgame !lm * alive !cdo -random $* time>${now()-interval('10minute')} x=src fmt:'$(lookup ${x[0]} (watchhash))${name}' 17:51:34 Redefined command: !randomgame => !lm * alive !cdo -random $* time>${now()-interval('10minute')} x=src fmt:'$(lookup ${x[0]} (watchhash))${name}' 18:01:19 gammafunk: do you have any insight into game.js:layout()? 18:02:24 advil: what do you want to know about it? 18:02:31 specifically why it rebuilds everything from html? 18:03:34 oh, hm, I guess it is blanking the areas to find their dimensions 18:04:53 gammafunk: why !cdo? 18:05:20 amalloy: oh, you watch a lot of webtiles games on cdo? 18:05:22 ZING! 18:05:32 (it's a function to watch a random webtiles game) 18:06:31 gammafunk: what about just adding `tiles` instead? it means you miss spectating console players via tiles, but that's probably good anyway because it's hard to communicate with them 18:06:57 amalloy: that seems to be a strictly worse solution to me 18:07:23 I mean, you can just do `!randomgame tiles` if you want that 18:07:34 !randomgame 18:07:35 http://crawl.berotato.org:8080/#watch-hotbaconsauce 18:08:03 but it literally won't work on cdo, so it makes more sense to just add !cdo 18:11:39 makes sense 18:15:49 Unstable branch on underhound.eu updated to: 0.20-a0-1063-gb8e8b7e (34) 18:19:00 -!- Catiua is now known as Reverie|Away 18:40:30 Are ghosts supposed to change their tile each time I take a step? 19:58:30 -!- yesno__ is now known as yesno 21:08:38 03advil02 07* 0.20-a0-1064-g127e8e2: Don't delete event handlers in messages on layout/resize 10(3 hours ago, 1 file, 5+ 2-) 13https://github.com/crawl/crawl/commit/127e8e2b14b2 21:10:26 I have to see that was a pretty satisfying bug to finally fix 21:11:02 (though I feel like it could be done more expediently if I knew js better) 21:11:11 *have tosay 21:11:17 *to say 21:13:27 -!- amalloy is now known as amalloy_ 21:22:29 advil: oh great! 21:22:36 did you finally fix that horrible spectator freeze bug 21:22:40 I think so 21:22:50 I'm confident I fixed the find bug, which should be the same thing 21:22:53 will this also fix the problem where sometimes menus stay open 21:23:02 when a spectator joins at the right time 21:23:11 hmm that could be related, but maybe different 21:23:15 is there a # for that? 21:23:29 hrm 21:23:37 the basic problem is that `layout` uses jquery .html to rewrite a whole bunch of stuff, and if there are event handlers, they vanish in that process 21:23:52 and layout is called on resize, and maybe other times 21:24:17 so it could well impact things other than the input box, if there are event handlers dynamically added somewhere 21:24:20 yes, I wouldn't be at all surprised if it were a related issue 21:24:29 preliminary search uncovers only https://crawl.develz.org/mantis/view.php?id=8057 21:24:32 which isn't very helpful 21:25:23 is /home/crawl-dev/dgamelaunch-config/bin/dgl update-trunk cmd not use -j[n] flag? 21:25:29 so slow.... 21:26:09 ah 21:26:10 gammafunk: yeah, it's a bit non-specific, though possibly related. I haven't yet actually looked into what happens when a spectator joins or why it calls layout, just tested the specific case in one of those bugs involving spectators 21:26:13 I've found some bugs 21:26:16 *bug reports 21:27:04 first one is probably not related 21:27:07 https://crawl.develz.org/mantis/view.php?id=8816 21:27:29 that one might simply be closed, I seem to recall this being a more common problem that got fixed 21:27:38 those symptoms sound the same 21:27:47 https://crawl.develz.org/mantis/view.php?id=8063 21:27:54 this is a bug report of the actual problem 21:28:19 ah interesting 21:28:33 that could be something else, if it's responsive to regular keys 21:28:40 but yeah, it seems to be a thing that can happen to spectators and not to the player 21:28:50 like it's maybe independent for each browser window 21:29:13 I'll be playing on console, with a designated browser window spectating so I can obs capture that for my twitch stream 21:29:28 and eventually menus will stay open permanently on said window 21:29:47 and interestingly if I go into different menus, the menus continue changing on the now-bugged browser window 21:30:02 interesting 21:30:04 so it's showing variously my wear, inventory, spell menus etc as I access those 21:30:15 but it never allows a menu to go away, just keeps the last opened one up 21:30:35 and my gameplay containues in the background (showing updates but the menu obscuring most of the gameplay area) 21:30:47 *continues 21:30:59 and it seems related to spectators joining the game 21:31:20 yeah, I'll have to look into what exactly happens when they join 21:32:31 I take this still happens on a semi-regular basis? 21:33:06 it didn't happen my last stream 21:33:25 it had happened at least a couple months ago, but yeah I'm not 100% certain that it still occurs 21:34:38 well, I don't know of any changes that would affect this in the last few months 21:34:50 hmm, unless it is somehow related to fake_cursor 21:35:49 what I'll do is look out for it 21:35:55 and maybe update pubby's bug 21:36:04 I guess your stream doesn't want to see you play around with the js debug console :-) 21:36:16 yeah, that sounds good 21:36:39 I have to use that to delete the chat html element before I start :) 21:36:43 but yeah they don't usually see that 21:36:54 huh 21:37:16 but it's a problem since I'm not watching this window, so if no one speaks up when it happens, I'll just happily play along in console while my vod is messed up from a buggy menu 21:37:38 deleting the chat element is just since it makes little distracting messages 21:38:04 about spectators and new messages, so it makes sense to just delete it 21:38:08 right 21:38:21 I actually keep open a third webtiles window spectating my game that I have up behind my console window 21:38:28 heh 21:38:38 so I can still see webtiles chat from time to time and look at webtiles if that's relevant 21:38:40 I do sometimes play with a spectating webtiles window open, basically for the minimap 21:39:05 I tried to use glyphs mode for a while but I just didn't like it 21:39:10 yeah I've heard of people doing that, but I use X map 21:40:01 I can't get any interactions with menus or windows and spectators locally, but that probably just means it's very timing-based 21:40:16 yeah 22:07:23 -!- ByronJoh1son is now known as ByronJohnson 22:09:26 Unstable branch on crawl.jorgrun.rocks updated to: 0.20-a0-1064-g127e8e2 (34) 22:35:39 03|amethyst02 07* 0.20-a0-1065-g8446734: Checkwhite 10(13 hours ago, 23 files, 28+ 27-) 13https://github.com/crawl/crawl/commit/8446734adda9 23:09:26 Unstable branch on crawl.jorgrun.rocks updated to: 0.20-a0-1065-g8446734 (34) 23:30:02 Good Lastyquitjoins 23:30:28 Hep's message on resuming your game is too long, it forces more on 80x24