--- Log opened Fri Dec 05 00:00:00 2008 --- Day changed Fri Dec 05 2008 20081205 00:00:00 Sirp__: Yes. I saw its message to that effect. 20081205 00:06:39 Sirp__: I assigned a bug that I guess is related to formula_ai to dragonking. Maybe you want to have a look at this as well. https://gna.org/bugs/?12668 20081205 00:06:58 fendrin: okay, thanks. 20081205 00:09:15 wesbot: bug 12668 20081205 00:09:16 AI0867: Are you sure with that number? Please check again... 20081205 00:09:30 eh? 20081205 00:10:36 AI0867: https://gna.org/bugs/?12668 20081205 00:11:05 wesbot is not uptodate sometimes... I have noticed that before. 20081205 00:11:27 I got the link, it's just that firefox is slow 20081205 00:11:31 so I prefer a summary 20081205 00:20:28 -!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has quit [Remote closed the connection] 20081205 00:20:46 i am off to bed, n8 20081205 00:28:51 -!- igel [n=igel@port-92-195-252-94.dynamic.qsc.de] has quit ["your ad here"] 20081205 00:37:09 -!- Sapient [n=sapien-x@c-71-59-39-246.hsd1.ga.comcast.net] has joined #wesnoth-dev 20081205 00:49:22 wesbot: man 20081205 00:49:27 wesbot: help 20081205 00:49:27 fendrin: I could tell you about bug(s), feature(s), task(s), patch(es), a log revision, update the topic, last seen of people... you name it. 20081205 00:49:45 wesbot: log 20081205 00:49:52 wesbot: log last 20081205 00:50:00 wesbot: log head 20081205 00:50:15 fendrin: wesbot doesn't have a 'log head' feature 20081205 00:50:23 you have to use specific numbers 20081205 00:50:24 shikadibot: log head 20081205 00:50:29 Revision 31274 (fendrin 2008-12-04 23:44:03 +0000 (Thu, 04 Dec 2008)): 20081205 00:50:32 LoW 14: Redone the village ownership assignment. 20081205 00:50:35 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31274 20081205 00:50:46 AI0867: thank you 20081205 00:58:18 -!- grrrX [n=_@p57BBFA64.dip.t-dialin.net] has quit ["Ex-Chat"] 20081205 01:08:32 -!- loonycyborg [n=sergey@79.139.246.148] has quit ["Zzzzzzzzzzzzzzzzzzzzz"] 20081205 01:09:14 -!- loonybot [n=loonybot@79.139.246.148] has quit [Remote closed the connection] 20081205 01:11:33 ESR_: I fixed an #enddef detection bug in wmlindent yesterday, but that caused a different one to surface: commented out #enddefs are still seen by wmlindent 20081205 01:12:07 methinks wmlindent should be using WmlIterator... 20081205 01:22:40 fixed 20081205 01:22:44 shikadibot: log head 20081205 01:22:46 Revision 31275 (ai0867 2008-12-05 00:22:29 +0000 (Fri, 05 Dec 2008)): 20081205 01:22:46 Stop wmlindent from using commented-out #enddefs. 20081205 01:22:46 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31275 20081205 01:24:06 AI: are you contributing to many open source projects? 20081205 01:27:36 -!- ilor [n=a@unaffiliated/ilor] has joined #wesnoth-dev 20081205 01:27:48 Sapient: just wesnoth and wes-umc-dev 20081205 01:29:04 ah ok 20081205 01:33:09 why? 20081205 01:33:49 just curious. wesnoth seems to attract a lot of developers 20081205 01:35:37 probably just a sign that it's a fun game with good replay value 20081205 01:36:07 Sapient: hopefully also that we have a friendly development community. :) 20081205 01:36:28 ah, yes. that too :) 20081205 01:44:08 loonycyborg: if it was just disk I/O, then why do I see the swap usage increase in kinfocenter? 20081205 01:44:27 yo Sapy 20081205 01:44:47 hello Shikadi 20081205 01:45:57 yeah, it's a fun game with good replay value 20081205 01:46:20 I've been commiting fixes to the campaign that made me fall in love with it... TRoW. Interesting. 20081205 01:47:49 to thy first love be true ;) 20081205 01:47:59 :) 20081205 01:48:13 is that the one with the red headed chick? 20081205 01:48:19 Yes. I don't think I have any favorite campaigns except TRoW and UtBS 20081205 01:49:01 or, should I say, 1.0.2 UtBS. It's been really long since I really played it to end. 20081205 01:49:09 (it was UMC back then) 20081205 01:49:22 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev 20081205 02:03:53 -!- fendrin_ [n=fabi@g227064058.adsl.alicedsl.de] has joined #wesnoth-dev 20081205 02:05:47 this prole] code is still pretty ugly 20081205 02:05:55 imma clean it up 20081205 02:07:41 prole]? 20081205 02:07:56 p => [ 20081205 02:08:49 in my keyboard I'd have obtained þrole] :) 20081205 02:09:43 -!- ilor [n=a@unaffiliated/ilor] has quit [Connection timed out] 20081205 02:12:29 -!- mjs-de [n=mjs-de@vpw.wh.Uni-Dortmund.DE] has quit [Remote closed the connection] 20081205 02:19:25 fendrin_: how does it feel to e maintainer of a mainline campaign? :) 20081205 02:19:56 Shadow_Master: why do you ask? 20081205 02:20:18 -!- fendrin [n=fabi@e179201116.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081205 02:20:18 wow... the [role] code doesn't even validate side number before using it as an array index 20081205 02:20:31 fendrin_: curiosity :) 20081205 02:20:57 or because I'm too bored right now 20081205 02:22:10 Shadow_Master: I am not satisfied with my work. Some bugs that have been reported by users were so obvious I should have found them before the release. 20081205 02:23:02 well, that happens even in C++ or graphics. it's not huge deal 20081205 02:23:51 you often need to get your stuff published for a broader audience in order to get all possible playtesting. that's why I suggested (and I see you did it) to publish DM for UMC ASAP 20081205 02:24:17 and that's why I need to get IftU usable again before xmas 20081205 02:25:25 Right, release early and often. What bugs me is that the campaign isn't playable until the end. So I loose feedback. 20081205 02:26:44 um, that's bad. I hope it isn't because of that foul story screen problem which was my fault, right? 20081205 02:27:11 shikadibot: blame Shadow_Master 20081205 02:28:15 there's some issue with delfador not being recalled as the leader 20081205 02:28:19 I have yet to see a bugtracker entry for it though 20081205 02:28:53 at some point Patterner had a journeyman mage as a leader, along with both a mage magister and a mage commander on his recall list 20081205 02:31:41 AI0867: strange 20081205 02:32:31 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"] 20081205 02:33:51 fendrin_: yeah, I think it may have something to do with the save_id switching 20081205 02:34:24 AI0867: There is only room for one persistant side? 20081205 02:35:15 no, there's room for more 20081205 02:35:28 Shadow_Master: no it's not your fault. I commited to many last minute changes. 20081205 02:35:33 but maybe not having a persistent side show up derails things somehow 20081205 02:35:45 I had some issues with that a while ago 20081205 02:36:27 the opponent of the player was persistent and in one scenario he didn't show up 20081205 02:36:33 in the next, wesnoth crashed 20081205 02:37:02 uh 20081205 02:38:30 noy: around? 20081205 02:38:35 yar! 20081205 02:39:39 noy: there's a new unit in 1.5.x named Dread Bat. Its description states 20081205 02:40:03 fendrin_: having a persistent side not show up in a scenario, caused problems when it reappeared, maybe something similar is happening here 20081205 02:40:41 "dread bats are fast and can darin the blood of those they attack, thereby healing themselves, even to the point where they become stronger than they behan the battl.e" 20081205 02:41:10 this is because when I made the unit in my UMC, it had regeneration. However, it no longer has regeneration and that comment does not make much sense to me. 20081205 02:41:29 (spelling mistakes from the mmanual copy to IRC aside) 20081205 02:41:29 they still have drain? 20081205 02:41:52 yesm, they have drain. but draining doesn't increase HP over the maximum, :) 20081205 02:43:58 so I was wondering if a fix for that would be possible in either side, description or actual stats 20081205 02:46:30 umm 20081205 02:46:51 can you link me to the description 20081205 02:47:15 shikadibot: file /trunk/data/core/units/bats/Dread.cfg 20081205 02:47:15 Web interface URL to file /trunk/data/core/units/bats/Dread.cfg: http://svn.gna.org/viewcvs/wesnoth/trunk/data/core/units/bats/Dread.cfg?view=log 20081205 02:47:27 hopefully that's the correct filename 20081205 02:47:49 shikadibot: file /trunk/data/core/units/bats/Bat_Dread.cfg 20081205 02:47:49 Web interface URL to file /trunk/data/core/units/bats/Bat_Dread.cfg: http://svn.gna.org/viewcvs/wesnoth/trunk/data/core/units/bats/Bat_Dread.cfg?view=log 20081205 02:49:18 so, http://svn.gna.org/viewcvs/wesnoth/trunk/data/core/units/bats/Bat_Dread.cfg?rev=30678&view=markup for short 20081205 02:51:28 noy: Shadow_Master: just give them the feeding ability that the ghols have. This way you don't need to brake the string freeze. 20081205 02:52:56 umm won't that break the string freeze too? 20081205 02:53:12 no. 20081205 02:53:45 there's a different magic for adding special notes that doesn't break string freezes. But this would create again the balance issue that forced me to remove feeding from them in the UMC version 20081205 02:53:57 hmm, upping gold of the gobbo will to? 20081205 02:54:01 though* 20081205 02:54:16 where does it say that? 20081205 02:54:23 well we forgot to 20081205 02:54:36 thats a separate issue 20081205 02:54:52 oh well, that's just a number. changing it doesn't interact with translations in any way 20081205 02:55:38 SM if you want to add the feeding ability, do so 20081205 02:56:32 -!- Mordocai [n=Mordocai@66.119.9.243] has joined #wesnoth-dev 20081205 03:09:14 hard to choose, asa I said, I had originaly replaced it with regeneration as it was too overpowering for long SP campaigns 20081205 03:12:06 * Shadow_Master would rather wish Espreon never had the idea of introducing it to mainline :/ 20081205 03:13:50 mmm... feeding 20081205 03:19:26 good night 20081205 03:19:31 gn, Shikadi 20081205 03:20:19 it seems that the wrong-thread-free()s are coming from gui2, not sure. but it seems to happen more often with --new-widgets 20081205 03:35:36 -!- Mordocai [n=Mordocai@66.119.9.243] has quit ["Leaving"] 20081205 03:45:53 Shadow_Master: hey, this is why I don't like adding things to mainline 20081205 03:46:21 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [] 20081205 03:54:32 -!- Sirp [n=user@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev 20081205 04:18:20 -!- Psyche^ [n=Psyche@e177234063.adsl.alicedsl.de] has joined #wesnoth-dev 20081205 04:21:32 -!- [Relic] [n=[Relic]@adsl-76-229-220-201.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev 20081205 04:22:42 <[Relic]> Hello :) 20081205 04:22:53 yo [Relic] 20081205 04:22:58 <[Relic]> :) 20081205 04:31:16 -!- Patterner [n=Psyche@e177239027.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081205 04:31:16 -!- Psyche^ is now known as Patterner 20081205 04:48:59 -!- Ivanovic_ [n=ivanovic@dtmd-4db26671.pool.einsundeins.de] has joined #wesnoth-dev 20081205 04:52:28 -!- isaac [n=isaac@debian/developer/isaac] has quit [Read error: 60 (Operation timed out)] 20081205 04:55:15 -!- isaac [n=isaac@debian/developer/isaac] has joined #wesnoth-dev 20081205 04:58:29 Sirp: there? 20081205 04:58:48 poolalloc.c is definitely not compiling for me 20081205 04:59:53 where am I supposed to get sys/mman.h and pthread.h ? 20081205 05:01:16 Sapient: ahh we can remove sys/mman.h 20081205 05:01:33 pthread.h should be part of the pthread library 20081205 05:01:37 is a pretty standard threads library 20081205 05:01:43 another dependency then? 20081205 05:02:18 I'll try searching for it on the web I guess 20081205 05:02:41 due to this, I am committing my changes to game_Events untested 20081205 05:05:30 -!- Ivanovic [n=ivanovic@dtmd-4db262d0.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 20081205 05:06:59 -!- Ivanovic_ is now known as Ivanovic 20081205 05:12:13 Sirp: does it matter what version of pthreads I should download? 20081205 05:14:57 -!- IRSeekBot3 [n=IRSeekLo@irseek/log-bot/IRSeekBot] has joined #wesnoth-dev 20081205 05:19:14 Sirp: it looks like you left some debug code in your r31277 20081205 05:19:23 I see lots of std::cerr 20081205 05:19:27 also it is asking what type of exception handling for pthread 20081205 05:19:38 SE - Structured EH, or 20081205 05:19:38 CE - C++ EH, or 20081205 05:19:38 C - no exceptions - uses setjmp/longjmp 20081205 05:23:20 pthread is used from c 20081205 05:23:30 so the third option seems like a good idea 20081205 05:23:42 probably takes less memory too 20081205 05:24:55 shikadibot: log head 20081205 05:24:59 Revision 31278 (sapient 2008-12-05 04:06:02 +0000 (Fri, 05 Dec 2008)): 20081205 05:25:02 massive cleanup for the [role] code, including crash prevention for invalid side numbers, consistent handling of precedence order, and removing excessive config copy operations. 20081205 05:25:06 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31278 20081205 05:25:37 AI0867: your change prompted me to do all that after I saw the surrounding code ;) 20081205 05:25:54 I would appreciate if someone could test that roles are still working in trunk 20081205 05:26:16 whilst I work on this dependency 20081205 05:33:44 Sapient: Damn, that looks like a *good* cleanup! 20081205 05:33:51 thanks, ESR 20081205 05:33:59 * ESR_ gives Sapient a cookie. 20081205 05:34:30 * Sapient cautiously grabs it, hides under a desk and nibbles it 20081205 05:34:50 * Patterner gives Sapient a durian cookie 20081205 05:35:11 * ESR_ is feeling very benign, having eaten abiut a pound of Brazilian barbecue warlier tonight. 20081205 05:36:03 ah, I never had Brazilian food... or, um, durian whatever that means :P 20081205 05:36:18 is the idea to highlight the button which gets selected by hitting enter still a good idea..? (e.g. on Quit: Yes No) 20081205 05:36:28 Durian is a fruit. 20081205 05:36:40 Sapient: U't a *really* stunky fruit from Indonesia. Tastes sweet but smells like fart gas. 20081205 05:36:56 s/stunky/stinky/ 20081205 05:37:10 Ah, I probably had some then; once had a Thai roommate 20081205 05:37:37 that lychee candy wasn't half bad 20081205 05:38:46 Mmmm...Thai food. 20081205 05:39:31 I wonder if it is a coincidence that the activists are called "PAD" which is also one of their famous dishes ;) 20081205 05:39:59 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev 20081205 05:46:45 south koreans definitely have the best activists, though 20081205 05:47:01 they take it as seriously as Americans and tailgating 20081205 05:47:16 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [Read error: 54 (Connection reset by peer)] 20081205 05:47:22 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev 20081205 05:48:55 I'd say phillipino protesters are colorful but not nearly as professional / impressive 20081205 05:51:48 hm? 20081205 05:52:18 you missed it noy, we were having a geopolitical discussion ! 20081205 05:52:23 ... 20081205 05:52:29 better check the irclog my friend 20081205 05:52:45 stupid competitive sports 20081205 05:52:49 take up all my time 20081205 05:53:14 you mean watching football? 20081205 05:54:58 no playing sports 20081205 05:57:58 cya later, have fun. 20081205 05:57:58 -!- Sapient [n=sapien-x@c-71-59-39-246.hsd1.ga.comcast.net] has left #wesnoth-dev [] 20081205 06:10:37 -!- [Relic] [n=[Relic]@adsl-76-229-220-201.dsl.milwwi.sbcglobal.net] has quit ["Leaving"] 20081205 08:34:03 -!- JonJon [i=hidden-u@mail2.hoforsnet.se] has joined #wesnoth-dev 20081205 08:34:11 -!- JonJon [i=hidden-u@mail2.hoforsnet.se] has left #wesnoth-dev [] 20081205 08:38:57 -!- miq [n=miq@krlh-5d854f21.pool.einsundeins.de] has joined #wesnoth-dev 20081205 08:45:18 -!- fendrin_ [n=fabi@g227064058.adsl.alicedsl.de] has quit [Remote closed the connection] 20081205 08:45:26 -!- Ivanovic_ [n=ivanovic@dtmd-4db26671.pool.einsundeins.de] has joined #wesnoth-dev 20081205 08:46:43 -!- Ivanovic [n=ivanovic@dtmd-4db26671.pool.einsundeins.de] has quit [Read error: 104 (Connection reset by peer)] 20081205 08:46:58 -!- fendrin_ [n=fabi@g227064058.adsl.alicedsl.de] has joined #wesnoth-dev 20081205 08:47:24 -!- Ivanovic_ is now known as Ivanovic 20081205 08:47:50 -!- ettin [n=jorda@234.207.219.87.dynamic.jazztel.es] has quit [Remote closed the connection] 20081205 08:49:34 -!- ettin [n=jorda@234.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev 20081205 09:13:24 -!- miq [n=miq@krlh-5d854f21.pool.einsundeins.de] has quit [Remote closed the connection] 20081205 09:21:08 Shadow_Master: there is nothing wrong with the description. drain allows a unit to "become stronger than they began the battle". 20081205 09:22:56 -!- Sirp [n=user@c-76-102-104-121.hsd1.ca.comcast.net] has quit ["leaving"] 20081205 09:29:35 -!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev 20081205 09:30:17 morning 20081205 09:55:38 I get lots of "cache size:, cache size max:, IMAGES: HEXED: SCALED_TO_HEX:" messages, the X server need 70% of CPU and the game is unplayable... 20081205 09:57:11 r31279,r2762,TSL:Scenario1 20081205 09:57:29 err.. r2772 20081205 09:58:48 Sirp__ ^ 20081205 09:59:12 Patterner try to comment out the std::cerr lines in image.cpp 20081205 09:59:57 I hope most of the problems are caused by the IO overhead of writing to the terminal (I had that problem with a slow terminal) 20081205 10:01:58 2>/dev/null helps :) 20081205 10:02:08 standard KDE konsole. 20081205 10:18:48 what log domain would be appropriate for those messages? 20081205 10:19:26 we have a cache domain but that is for the config cache afaict. 20081205 10:19:49 not sure maybe it's own domian 20081205 10:19:53 could make it engine or general otherwise. 20081205 10:19:53 domain* 20081205 10:21:07 actually i think i'll just add it to the cache domain. if it bothers someone they can split it. :-) 20081205 10:21:14 ok 20081205 10:21:46 also not sure whether they will stay, or are there for debugging 20081205 10:22:07 well, they'll go to debug level. 20081205 10:22:19 I assumed that 20081205 10:22:40 just that I sometimes also forget to remove my cerr debug messages 20081205 10:26:04 well, IMO that usually does not make sense knowing that we have a debug logdomain unless it's really trivial debug info. 20081205 10:51:06 -!- ilor [n=a@unaffiliated/ilor] has joined #wesnoth-dev 20081205 10:51:48 soliton * r31281 /trunk/src/image.cpp: put image cache debug messages in the cache domain 20081205 10:52:05 -!- loonybot [n=loonybot@79.139.246.148] has joined #wesnoth-dev 20081205 10:52:26 -!- loonycyborg [n=sergey@79.139.246.148] has joined #wesnoth-dev 20081205 11:10:03 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev 20081205 11:33:35 Soliton: Are you sure that it was absolutely necessary to remove regeneration from dread bat? 20081205 11:34:05 * loonycyborg thinks that their low hp and regeneration is an interesting combination. 20081205 11:34:35 no. 20081205 11:35:55 Is "no" answer to first or second remark? 20081205 11:37:31 i'm not going to say no to what you think. :-) 20081205 11:40:51 Some people say "no" to express disagreement. Glad you're not one of those :) 20081205 11:50:07 AI0867: http://www.wesnoth.org/forum/viewtopic.php?f=21&t=23172 20081205 11:55:52 -!- loonycyborg [n=sergey@79.139.246.148] has left #wesnoth-dev ["In Soviet Russia, channel /parts YOU!!"] 20081205 11:56:14 -!- ilor [n=a@unaffiliated/ilor] has left #wesnoth-dev [] 20081205 11:56:30 -!- loonybot [n=loonybot@79.139.246.148] has quit [Remote closed the connection] 20081205 11:59:59 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081205 12:04:05 hi 20081205 12:04:40 hi fendrin_ 20081205 12:29:09 zookeeper: http://gna.org/patch/?1115 20081205 12:30:17 Soliton, roger, will do later.. 20081205 12:40:24 are we really not going to fix http://gna.org/bugs/?12666 because of the *soft* string freeze? 20081205 12:41:48 i thought that is exactly what we have a soft string freeze for... to fix such bugs. 20081205 12:44:11 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit ["Hi! I'm a quit message virus vaccine. If you see a quit message virus, don't replace your quit message with it!"] 20081205 13:02:41 Soliton: It's not a big thing. Only anoying. 20081205 13:03:08 Soliton: If you prefer I go to fix it immediately. 20081205 13:08:50 AI0867: Is there a bug report for the persistent problem? 20081205 13:09:42 -!- JonJon [i=hidden-u@mail2.hoforsnet.se] has joined #wesnoth-dev 20081205 13:23:05 -!- mjs-de [n=mjs-de@holmes.E-Technik.Uni-Dortmund.DE] has joined #wesnoth-dev 20081205 13:25:42 -!- JonJon [i=hidden-u@mail2.hoforsnet.se] has quit ["ChatZilla 0.9.83 [Firefox 2.0.0.18/2008102918]"] 20081205 13:51:16 esr * r31282 /trunk/src/ (ai.cpp ai_move.cpp): 20081205 13:51:16 Address bug #12632 (Reproducible pathfinding crash) by backing out yet another 20081205 13:51:16 suokko change it turns out I shouldn't have merged. Sigh... 20081205 14:11:47 -!- wesbot changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 59 bugs, 210 feature requests, 8 patches | logs: http://wesnoth.org/irclogs | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page=add 20081205 14:16:25 ESR_: do you know about the persistent bug in DM? 20081205 14:16:43 No, what's that? 20081205 14:17:13 ESR_: after the protagonist is switched back to delfador he is a lvl1 unit again. 20081205 14:18:17 Yo, zookeeper! ^ <--- Does febdrin's report ring any bells? 20081205 14:19:29 esr * r31283 /trunk/src/ai_python.cpp: 20081205 14:19:29 Address bug #12638 (compilation crash). This is a theoretical fix based 20081205 14:19:29 on examining the 2.6 Python.h header. It will need testing. 20081205 14:20:44 fendrin_: OI think we should fix the pending LoW bugs. Soliton is right, the string freeze is still soft for a reason. 20081205 14:21:36 ESR_: okay I will take care about #12666 20081205 14:22:00 ESR_, no, i'm not familiar with DM 20081205 14:22:57 zookeeper: Didn't think you were, UI was wondering if you know of a common WML error pattern that can cause that sort of level revert. 20081205 14:26:43 ESR_: AI0867 has seen a similar thing before. 20081205 14:27:24 -!- EdB [n=EdB@188.117.88-79.rev.gaoland.net] has joined #wesnoth-dev 20081205 14:31:18 -!- fredrik_ [n=fredrik@81-234-91-123-o286.telia.com] has joined #wesnoth-dev 20081205 14:38:18 ESR_, but if someone tells me which scenario to look at, i can.. 20081205 14:40:19 scenarios 21-22? 20081205 14:40:57 zookeeper: Looking... 20081205 14:41:19 esr * r31284 /trunk/data/tools/wesnoth/wmltools.py: Cope with whole-repo checkputs better. 20081205 14:41:54 zookeeper: the protagonist is switched back to delfador in scenario 18 20081205 14:43:19 background=images/maps/wesnoth.png looks like a bad image path 20081205 14:43:48 fendrin_: Ack! That's an error. I think scenario 15 should be where it happens. In 14 Kalenz agrees to follow Del to the Ford of Abez. 20081205 14:45:44 ESR_: there seems to be nothing wrong with the wml. Looks like an engine bug. 20081205 14:46:06 Oh, great. 20081205 14:46:23 ESR_: thanks for assigning the TB bugreport to me 20081205 14:46:34 Another persistence problem. I was afraid that would be the case. 20081205 14:46:35 odd capitalization: "The elves fought well and have earned Our respect" 20081205 14:46:40 zookeeper: could you test if you can reproduce this one your windows machine? https://gna.org/bugs/?12683 20081205 14:46:44 eh, i'm looking into it, gimme a minute :) 20081205 14:46:47 Ivanovic: Yo're welcome. 20081205 14:47:25 looks like there are some double spaces in dialogue too 20081205 14:48:16 ESR_: that is: there is nothing at all that is special about this turn in this mission 20081205 14:48:23 so i am really suprised about problems with it 20081205 14:49:54 so kalenz is the leader in 15, then there's two sideless story scenarios, then a scenario with delfie as the leader. delfador won't be picked as the leader unless he has canrecruit=yes 20081205 14:50:21 zookeeper: In English, when a reigning monarch speaks in his official capacy, he may refer to himself as "we" (the "royal we"). In extremely formal contexts the "we" (and related posessive "our") may be capitalized to emphasize it. It's an archaic usage and even mst native speakers in the U.S. don't know of it (Brits are probably more familiar). 20081205 14:50:43 so, the game looks at the recall list, figures there's no id=Delfador leader to use, and so it uses the side def to create a new journeyman mage leader 20081205 14:50:55 ESR_, oh. silly monarchs. 20081205 14:51:06 Ivanovic: I susapect it's a Windows packaging error, wrong permission on some file. 20081205 14:51:28 Ivanovic, i don't have 1.5.6 20081205 14:51:39 err, 1.4.6 i mean 20081205 14:52:27 zookeeper: okay thanks. I can handle that bug now. 20081205 14:53:05 AFAIK leader carryover simply works so that the game looks at the id= in side def and then tries to find a matching canrecruit=yes unit from the recall list 20081205 14:53:26 I think the "royal we" also occurs in French. Don't know about other languages. 20081205 14:54:00 so whenever you want to have the leaders switched in scenario x, you need to flip their canrecruits around at the end of the previous one 20081205 14:54:21 (or just do the whole thing in prestart, whichever) 20081205 14:55:03 ESR_: uhm, does windows have this strange concept with "permissions" at all? 20081205 14:55:04 ;) 20081205 14:55:29 but if it is a packaging issue, loonycyborg should have a look at it since he bundles those 20081205 14:55:45 Ivanovic: I'm deducing that from the bug report, he says it worked when he went to Windows equivalent of root. 20081205 14:56:10 ESR_: still this is unlikely somehow since there is *nothing* new happening in this turn 20081205 14:56:16 as in no image loaded, nothing like this 20081205 14:57:36 Wacky. 20081205 14:58:05 ESR_ in Dutch we also use the royal we 20081205 15:00:37 mordante: I wonder what the common ancestor is? Maybe Roman emperors used it...rgat would be my guess. 20081205 15:00:52 s/rgat/that/ 20081205 15:01:16 zookeeper: I take it there;s no parallel form in Finnish? 20081205 15:04:04 eh, can't say i'm familiar with archaic finnish 20081205 15:05:04 You aren't one of the Swedish-speaking minority, are you? 20081205 15:05:22 nope 20081205 15:05:42 no idea where the we comes from 20081205 15:11:01 Wkipedia says the historical term for it is the Latin "pluralis majestatis", which suggests that my guess about Roman emperors is likely correct. 20081205 15:11:04 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has joined #wesnoth-dev 20081205 15:11:29 seems like it 20081205 15:15:22 AI0867: ping 20081205 15:23:52 esr * r31285 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/06_Acquaintance_in_Need.cfg: 20081205 15:23:52 Fix a typo and a content error. Adresses bug #12662: LoW: minor bugs 20081205 15:23:52 in Acquaintance in Need story. 20081205 15:28:29 esr * r31286 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/09_Bounty_hunters.cfg: Address bug #12666: LoW: Bounty Hunters - left_behind string starts with ", ". 20081205 15:29:38 ESR_: what did you do with 12666? 20081205 15:30:03 Just applied the improved patch. 20081205 15:30:24 okay, It's buggy as hell. ai made 3 mistakes. 20081205 15:30:34 I figured you got distracted by somerthing, so I did it. 20081205 15:30:41 Oopps. Sorry. 20081205 15:30:50 ESR_: I got distracted by it's bugginess: 20081205 15:30:52 :-) 20081205 15:30:59 Oh. 20081205 15:31:15 ESR_: no problem 20081205 15:31:56 fendrin_: I'm going to try to re-tune The Return of Trouble now. 20081205 15:32:45 ESR_: okay, I am going to fix the persistence bug of delfador 20081205 15:32:56 Good. 20081205 15:33:16 In wml arrays start with 0 or with 1? 20081205 15:35:37 fendrin_: Ask zookeeper, I'm not sure. 20081205 15:37:30 teh first element is array[0] 20081205 15:40:53 -!- boucman [n=rosen@159.83.102-84.rev.gaoland.net] has joined #wesnoth-dev 20081205 15:41:40 zookeeper: thanks 20081205 15:45:43 fendrin * r31287 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/09_Bounty_hunters.cfg: fixed bugs in ai's fix for bug #12666 20081205 15:58:51 fendrin_: Have you seen this? You might find it useful: http://www.catb.org/~esr/wesnoth/campaign-design-howto.html 20081205 16:04:41 ESR_: No that is new to me. Huge. 20081205 16:05:06 -!- Appleman1234 [n=Appleman@131.181.47.5] has joined #wesnoth-dev 20081205 16:05:21 I had a lot of fun writing it. 20081205 16:14:30 ESR_: and you promised me at the time that, once done, you would add it to the wiki :P 20081205 16:14:57 it's one of the best pieces of advice I know about campaign writing, but it's not at a place where it can be easily found 20081205 16:15:16 zookeeper: actually, by my experience, it picks *any* canrecruit=yes unit from the recall list of the corresponding save_id 20081205 16:15:24 though it was 1.2.8 when I last tried that 20081205 16:17:53 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has quit [Remote closed the connection] 20081205 16:17:59 AI0867: I am not sure what you wanted to do with your second patch for bug #12666. Sanitize the iterator? The code was broken. Maybe you have a look of what I have done with it. 20081205 16:18:43 I am away now, must buy some food. 20081205 16:18:53 fendrin_: I was unable to test it 20081205 16:19:12 but I switched from '$unit' to '$i' for the FOREACH's second var 20081205 16:19:18 AI0867: my biggest problem was that the case switch wml isn't linked in the wiki 20081205 16:19:30 it's not? 20081205 16:19:37 I got it from the wiki 20081205 16:19:39 I wasn't able to find it. 20081205 16:19:54 Maybe it's in there but the menu to the right doesn't know about it. 20081205 16:19:57 -!- fendrin_ is now known as fendrin 20081205 16:20:37 fendrin: that menu is often forgotten about 20081205 16:20:42 look up [if] and scroll down 20081205 16:20:48 anyway, I'm off again 20081205 16:21:38 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has joined #wesnoth-dev 20081205 16:23:17 AI0867, yeah, could be 20081205 16:23:41 but then again, it doesn't matter much since you shouldn't have two canrecruit=yes units at the same time anyway 20081205 16:29:37 Sirp__: Wesnoth doesn't work for me since r31222 (it freezes before titlescreen) and since r31277 it just segfaults on launch. 20081205 16:29:44 -!- loonybot [n=loonybot@79.139.246.148] has joined #wesnoth-dev 20081205 16:30:04 -!- loonycyborg [n=sergey@79.139.246.148] has joined #wesnoth-dev 20081205 16:32:06 -!- Appleman1234 [n=Appleman@131.181.47.5] has quit [Read error: 110 (Connection timed out)] 20081205 16:34:45 stikonas: when you compile with -DDISABLE_POOL_ALLOC it works though? 20081205 16:46:46 -!- Appleman1234 [n=Appleman@131.181.47.2] has joined #wesnoth-dev 20081205 16:47:13 Soliton: Wesnoth doesn't segfault when compiled with -DDISABLE_POOL_ALLOC, though it still freezes before showing titlescreen 20081205 16:48:11 stikonas before or after r31281 ? 20081205 16:48:39 stikonas: Are you sure that malloc.c and poolalloc.c aren't used in the compilation? 20081205 16:49:58 loonycyborg: I'm not I passed this option to CXX_FLAGS in cmake 20081205 16:50:24 loonycyborg: s/I'm not/I'm not sure/ 20081205 16:50:39 mordante: after r31281 20081205 16:50:55 Remove malloc.c and poolalloc.c from src/CMakeLists.txt 20081205 16:51:15 Or run "scons pool_alloc=false" :D 20081205 16:55:14 looncyborg: it works now, thanks. 20081205 16:55:52 * stikonas is restarting X server. 20081205 16:56:11 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has quit [Remote closed the connection] 20081205 17:02:53 new campaignd-1.4 profile: http://www.wesnoth.org/files/campaignd/1.4/gprof.20081126-180259.log 20081205 17:03:21 i've restarted it now after it was using 100% of one CPU for more than i day, i think. 20081205 17:10:19 Soliton: Great! It'll surely speed up compilations once that processor-hogging campaignd is killed. 20081205 17:11:57 -!- Appleman1234 [n=Appleman@131.181.47.2] has quit [Read error: 54 (Connection reset by peer)] 20081205 17:15:26 -!- EdB [n=EdB@188.117.88-79.rev.gaoland.net] has quit ["Konversation terminated!"] 20081205 17:23:55 http://www.wesnoth.org/wiki/Special:Log/block lolz. i guess i'm the official wiki police. 20081205 17:30:14 "Go spam yourself." 20081205 17:31:23 -!- fendrin is now known as fendrin_ 20081205 17:31:46 hello 20081205 17:31:50 And then ban yourself like uncyclopedia admins often do :) 20081205 17:40:22 -!- Appleman1234 [n=Appleman@131.181.47.2] has joined #wesnoth-dev 20081205 17:46:23 -!- IRSeekBot3 [n=IRSeekLo@irseek/log-bot/IRSeekBot] has quit [K-lined] 20081205 17:49:22 -!- mjs-de [n=mjs-de@holmes.E-Technik.Uni-Dortmund.DE] has quit [Remote closed the connection] 20081205 17:56:47 -!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit [] 20081205 17:57:06 -!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Read error: 104 (Connection reset by peer)] 20081205 17:58:33 -!- Appleman1234 [n=Appleman@131.181.47.2] has quit [Remote closed the connection] 20081205 18:00:45 mordante * r31288 /trunk/src/image.hpp: Fix a compiler warning. 20081205 18:04:41 -!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev 20081205 18:12:28 -!- Appleman1234 [n=Appleman@131.181.47.2] has joined #wesnoth-dev 20081205 18:44:52 -!- grzywacz [n=grzywacz@moinmoin/developer/karol] has joined #wesnoth-dev 20081205 18:55:47 -!- [Relic] [n=[Relic]@76.229.220.201] has joined #wesnoth-dev 20081205 18:56:48 <[Relic]> Hello :) 20081205 18:57:01 hi [Relic] 20081205 18:57:12 <[Relic]> :) 20081205 19:25:24 ai0867 * r31289 /trunk/src/ai_python.cpp: Fix some syntax errors in python2.6 ifdefs. 20081205 19:28:14 -!- Appleman1234 [n=Appleman@131.181.47.2] has quit [Read error: 110 (Connection timed out)] 20081205 19:28:41 -!- Appleman1234 [n=Appleman@131.181.47.2] has joined #wesnoth-dev 20081205 19:40:10 loonycyborg * r31290 /trunk/packaging/windows/Wesnoth.nsi.in: Made shortcuts created by NSIS installer pass --config-dir. 20081205 19:49:52 -!- fredrik_ [n=fredrik@81-234-91-123-o286.telia.com] has quit [Remote closed the connection] 20081205 20:02:17 -!- EdB [n=EdB@59.38.72-86.rev.gaoland.net] has joined #wesnoth-dev 20081205 20:06:52 -!- Baufo [n=thomas@62-47-150-239.adsl.highway.telekom.at] has joined #wesnoth-dev 20081205 20:07:21 hey all :) 20081205 20:07:27 Shadow_Master, you were looking for me? 20081205 20:08:33 yes, read your PM inbox in the forums :) 20081205 20:09:08 hi Baufo 20081205 20:09:17 hey mordante 20081205 20:09:36 Shadow_Master, ok, will do that 20081205 20:11:47 -!- wesbot changed the topic of #wesnoth-dev to: 1.4.7 planned for 12.12. | soft stringfreeze for most campaigns: http://www.wesnoth.org/wiki/TextdomainStatus | 1.6 release plan: http://www.wesnoth.org/wiki/Wesnoth1.6ReleasePlan | 56 bugs, 210 feature requests, 8 patches | logs: http://wesnoth.org/irclogs | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com, or for images: http://imagebin.org/index.php?page=add 20081205 20:24:50 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"] 20081205 20:27:47 loonycyborg: Have you seen bug #12683: aToTB: Permissions issue leads to freeze at the beginning of Guarded Castle turn 12? 20081205 20:28:22 -!- zookeeper2 [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev 20081205 20:30:01 -!- BenUrban [n=benurban@c-68-49-10-243.hsd1.md.comcast.net] has joined #wesnoth-dev 20081205 20:32:39 ESR_: Yes. Though naturally I couldn't repro it because I'm using an admin account and there's no limited accounts on my XP installation. 20081205 20:33:11 Maybe it's caused by some brain-dead security policy.. 20081205 20:33:59 How do you troubleshoot something like that? 20081205 20:34:16 make a limited account : ? 20081205 20:35:24 Good idea! Unfortunately I'll have to boot to windows to do that :( 20081205 20:35:40 -!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit [Read error: 145 (Connection timed out)] 20081205 20:35:43 And since it's a minor issue anyway.. 20081205 20:37:42 -!- zookeeper2 is now known as zookeeper 20081205 20:37:57 Sirp__ around? 20081205 20:38:03 mordante: Are you currently getting segfaults caused by poolalloc? 20081205 20:38:09 yes 20081205 20:38:21 and I also have some suspicious code 20081205 20:38:29 ESR reported on #wesnoth-umc-dev that he does too. 20081205 20:38:49 (not that changing it fixes the problem) 20081205 20:39:00 um, at least yesterday I had no problems 20081205 20:39:01 But strangely it works for me :/ 20081205 20:39:06 erm, 12 hours ago 20081205 20:39:24 I just filed a bug wuth a backtrace on this. 20081205 20:42:50 shikadibot: log r31288 20081205 20:42:53 Revision 31288 (mordante 2008-12-05 17:00:21 +0000 (Fri, 05 Dec 2008)): 20081205 20:42:56 Fix a compiler warning. 20081205 20:42:59 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31288 20081205 20:43:06 Oops! 20081205 20:43:10 shikadibot: bug r31288 20081205 20:43:24 you fail. 20081205 20:43:26 shikadibot: bug 31288 20081205 20:43:41 shikadibot doesn't know how to fetch that information 20081205 20:43:50 Shadow_Master: :( 20081205 20:43:58 wesbot: bug 31288 20081205 20:43:58 loonycyborg: Are you sure with that number? Please check again... 20081205 20:43:59 hey, I coded it as I needed it 20081205 20:44:15 wesbot: bug 12689 20081205 20:44:15 Bug #12689 Assigned to: David White Status: None Priority: 5 - Normal 20081205 20:44:19 Summary: Crash on startup when using poolalloc 20081205 20:44:21 Original submission: Program received signal SIGSEGV, Segmentation fault. 20081205 20:44:24 itching to Thread 0x7f8a51fb56f0 (LWP 29490)] 20081205 20:44:27 URL: https://gna.org/bugs/?12689 20081205 20:44:51 * loonycyborg never gives up :) 20081205 20:45:08 I have the impression some members are not initialized properly 20081205 20:45:19 In a possibly related development, wesnoth is now deeply and painfully slow. I'm actually able to watch partial screen updates happening. 20081205 20:45:29 indeed, but I thought it aws just me 20081205 20:45:36 even at -O3 :( 20081205 20:46:12 I'm getting extreme slowdowns even with -O2 -march=prescott 20081205 20:46:19 you updated to the latest head? 20081205 20:46:51 With unit number it was handling well previously. 20081205 20:47:11 *unit quantities 20081205 20:47:28 Thought that it was some local DM issue.. 20081205 20:47:34 * zookeeper scratches head 20081205 20:47:45 there seems to be something wrong with storing units from the recall list 20081205 20:47:52 mordante: I'm using latest head 20081205 20:49:17 loonycyborg: does -march introduce important optimizations? 20081205 20:49:24 ok 20081205 20:50:55 Shadow_Master: Everyone on gentoo pass -march appropriate to their cpu, so I guess it does :) 20081205 20:52:13 mordante * r31291 /trunk/src/poolalloc.c: 20081205 20:52:13 Fix two realloc issues. 20081205 20:52:13 realloc(*ptr, 0) now frees the pointer instead of looking at the current 20081205 20:52:13 size of the pointer. 20081205 20:52:13 if the malloc fails realloc should return NULL; 20081205 20:52:23 Sirp__ please review ^ 20081205 20:53:02 Somebody do a bisect. The slowdown is quite recent; I'm sure I wasn't seeing it yesterday. 20081205 20:57:44 * Shadow_Master getting at 70ºC, can't do bisects 20081205 20:59:28 ESR_: At which revision you definitely didn't have the slowdowns? 20081205 20:59:43 loonycyborg: Looking... 20081205 21:00:03 with git-svn making bisects is easier 20081205 21:02:16 loonycyborg: I'd suggest starting at 31161. Mordante did something that looks like it could be related one rev later. 20081205 21:04:12 While you do that, I'll see if Mordante's patch fixes my poolalloc crash. (That's why I',m not doing the bisect myself; you guys have seen the slowdown but so far only I have produced the starttup crash.) 20081205 21:04:27 -!- [Relic] [n=[Relic]@76.229.220.201] has quit ["Leaving"] 20081205 21:04:56 I'm not sure whether the crash will be fixed by it, my crash hasn't been fixed yet 20081205 21:05:19 ESR_ you might want to revert 31277 on image.cpp and image.hpp 20081205 21:05:53 or maybe revert it entirely to see whether it fixes both problems 20081205 21:07:10 shikadibot: log 31277 20081205 21:07:11 Revision 31277 (dave 2008-12-05 03:33:26 +0000 (Fri, 05 Dec 2008)): 20081205 21:07:11 made the memory allocator get memory from dlmalloc rather than using mmap(). Also made it free memory properly from threads other than the main one. 20081205 21:07:14 Web interface URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31277 20081205 21:08:00 Yeah, that looks like it could be implicated, all right. 20081205 21:08:04 ESR_ first revert r31281 or you'll get a conflict 20081205 21:08:19 Noted. 20081205 21:08:55 I'm not even sure the changes to image.cpp/hpp were intended to be commited 20081205 21:09:26 brb 20081205 21:10:01 Testing head with pool_alloc=true build=release...OK, mordante, your fix didn'tr prevent the startup crash. I'll try the reverts you suggested. 20081205 21:10:02 let0s see: scons extra_flags_release='-O3 -mtune=native -march=native' 20081205 21:10:23 since I'm such a bastard I don't know the exact family of this AMD64 20081205 21:11:21 mordante: so what was the problem? It wasn't freeing the memory if you tried to realloc() 0 ? 20081205 21:13:17 Sirp__: We've got two bugs at present. I'm seeing a startup crash with pooolalloc enabled, and everyone is seeing horrible slowdown whether poolalloc is enabled or not. 20081205 21:13:41 r31161 is unaffected by slowdowns for me. 20081205 21:14:15 loonycyborg: Yeah, I was pretty sure it would be. 20081205 21:14:57 I guess that definitely excludes DM-specific issue.. 20081205 21:15:02 * ESR_ means he was sure it would *not* be slow. 20081205 21:15:57 loonycyborg: You bisect, I'll back out the image.[ch]pp commits Mordante pointed out. Let's see who gets there first. 20081205 21:16:37 Sirp__: See bug #12689 for a backtrace on the startup poolalloc crash. 20081205 21:18:38 mordante tried a poolalloc cleanup that might have fixed it, but no joy. 20081205 21:19:35 mordante: Build faiuls messily with those two commits backed out. 20081205 21:22:15 -!- Sapient [n=sapien-x@c-71-59-39-246.hsd1.ga.comcast.net] has joined #wesnoth-dev 20081205 21:24:09 will memory management plans such as custom allocators affect our dates in the release plan? 20081205 21:25:08 I'm not necessarily in a hurry to get a new stable release out, but I thought it deserved asking before we got too far into it 20081205 21:25:30 -!- EdB [n=EdB@59.38.72-86.rev.gaoland.net] has quit ["Konversation terminated!"] 20081205 21:32:29 it is certainly slower than it was last night during animations 20081205 21:33:21 ah well, I still have the debug build for playtesting stuff ^__^ 20081205 21:37:22 -!- miq [n=miq@krlh-5d854f21.pool.einsundeins.de] has joined #wesnoth-dev 20081205 21:41:48 wesnoth is coredumping dnow 20081205 21:42:29 I suspect that it may be caused by the optimizations I asked the compiler to do :( 20081205 21:43:12 Dang! Now I'm not getting slowdowns even with HEAD. That means that my reproduction method is flawed :( 20081205 21:43:55 garg. 20081205 21:46:00 just to make sure, I'll try with a nonoptimized build with debug symbols now. the backtrace makes me suspect of poolalloc 20081205 21:47:56 torangan * r31292 /branches/1.4/po/wesnoth-l/fi.po: finnish translation update 20081205 21:48:04 Arrgh! If I get the slowdown, save the game and reload it, slowdown disappears! 20081205 21:49:09 torangan * r31293 /trunk/ (changelog po/wesnoth-l/fi.po): finnish translation update 20081205 21:52:55 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has joined #wesnoth-dev 20081205 21:54:11 Ivanovic: um, have you noticed that I sent you another Spanish translation update? 20081205 21:54:28 * Shadow_Master gets the slowdown from loadaed games 20081205 21:54:30 Shadow_Master: so far all i got were commited 20081205 21:54:39 Ivanovic: um. 20081205 21:54:40 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"] 20081205 21:54:47 how many did you get last two days? 20081205 21:54:53 the last one is about 27h ago 20081205 21:55:22 uh, okay 20081205 21:55:52 and i think i commited this one 20081205 21:56:38 nvm, It seems that I missed them in the git svn output 20081205 21:58:50 -!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081205 21:59:09 r31271, that's it. Got distracted by a bulky commit by mordante 20081205 22:02:36 Sapient! 20081205 22:02:44 hi zookeeper 20081205 22:03:01 waah, i ran into a situation where i really wanted [have_unit] to be able to filter on recalls 20081205 22:03:26 also, storing the recalls and using find_in= on the array didn't seem to work 20081205 22:03:37 (find_in= in [have_unit]) 20081205 22:03:39 hmm 20081205 22:03:49 let me check the src 20081205 22:07:37 -!- stikonas [n=and@ctv-213-164-109-68.vinita.lt] has quit [Remote closed the connection] 20081205 22:08:35 zookeeper: the only thing find_in checks is the id= 20081205 22:10:04 Shadow_Master: hey, sometimes i am fast with commiting this stuff 20081205 22:10:06 ;) 20081205 22:10:15 Sapient, i know 20081205 22:10:21 hmm 20081205 22:10:48 so does that mean that [have_unit] find_in=array level=2 [/have_unit] won't work? 20081205 22:10:55 or "shouldn't" work 20081205 22:11:34 uh... it should work if there is a level 2 unit with an id in array *on the gamemap* 20081205 22:11:49 duh! silly me :) 20081205 22:11:52 of course 20081205 22:12:47 i was thinking it'd evaluate to true if it'd find a level 2 unit in array 20081205 22:13:00 Finally! A bad commit that isn't HEAD. 20081205 22:15:16 Sapient, anyway, that recall list filtering would be pretty cool, i have to write much more hackish code now ;) 20081205 22:16:57 what about... [have_recall] ? 20081205 22:17:25 * zookeeper shrugs 20081205 22:17:26 i guess 20081205 22:17:49 or you could have the filtering for recall list active only if the filter has x,y=recall,recall 20081205 22:22:00 -!- fendrin__ [n=fabi@f051199091.adsl.alicedsl.de] has joined #wesnoth-dev 20081205 22:25:04 the problem I think is when most people use [have_unit] they intend to know if there's a living unit in play meeting those characteristics 20081205 22:25:27 so I wouldn't want to change the behavior from that expectation 20081205 22:28:14 -!- Appleman1234 [n=Appleman@131.181.47.2] has quit [Read error: 104 (Connection reset by peer)] 20081205 22:31:10 well, if you'd have the filtering for recall list active only if the filter has x,y=recall,recall, then that'd be true 20081205 22:31:49 technically, yes :) 20081205 22:34:52 why just "technically"? 20081205 22:35:46 well... it seems a little obscure, but it's not that bad 20081205 22:36:25 just my preference 20081205 22:38:21 i'd rather have it all working through [have_unit] because then you don't need to use a different tag if you just want a different filtering...like if you do some complicated [insert_tag]ing and such 20081205 22:38:28 -!- fendrin_ [n=fabi@g227064058.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081205 22:39:11 I added some changes to image caching and also added some logging (image.hpp, image.cpp) -- it didn't seem to slow things down on my machine, but if this slow behavior is persisting, people could try reverting it... 20081205 22:39:34 and i just think it makes sense to have the same SUF cover all unit filterings 20081205 22:40:03 zookeeper: ok then. it's rare enough to justify a little obscurity I suppose 20081205 22:40:11 cool :) 20081205 22:40:41 Sirp__: i already made that logging debug level so it won't print usually. 20081205 22:40:46 should be a really simple change, z 20081205 22:45:04 even debug level logging requires a conditional branch... just a simple boolean check though 20081205 22:45:27 if it was in a critical enough loop, it might impact something 20081205 22:46:14 31277 is the first bad commit. 20081205 22:46:17 then again, if you enabled such a frequent statement it would print way too much to read 20081205 22:47:08 Unless it's coused by reason different than logging and running "./wesnoth > /dev/null 2>&1" wasn't enough to exclude this factor. 20081205 22:47:26 *caused 20081205 22:49:10 So removing logging would be a good idea to try.. 20081205 23:05:43 -!- McShark [n=Simon_So@drsd-4db3f35e.pool.einsundeins.de] has joined #wesnoth-dev 20081205 23:06:01 Hello again! 20081205 23:06:09 I also changed some of the cache behavior, so I recommend trying with reverted image.cpp/image.hpp 20081205 23:06:30 This room seems to be quite full. That's good. 20081205 23:06:51 Hi Sirp, think I want to to to you about poolalloc 20081205 23:07:16 (to talk to...) 20081205 23:09:31 McShark: You're porting it to windows? :) 20081205 23:10:02 Maybe somebody can help we with poolalloc. I tried to use it in win environment (with pthreads-win32). I'm trying more to compile than to port. 20081205 23:10:46 From current svn I get the following warnings/errors and I'm not quite clear what to do about them: 20081205 23:10:46 D:\Spiele\WesnothTrunk\src\poolalloc.c||In function 'collect_memory_from_other_threads':| 20081205 23:10:46 D:\Spiele\WesnothTrunk\src\poolalloc.c|239|warning: comparison between pointer and integer| 20081205 23:10:46 D:\Spiele\WesnothTrunk\src\poolalloc.c||In function 'malloc':| 20081205 23:10:46 D:\Spiele\WesnothTrunk\src\poolalloc.c|327|error: invalid operands to binary ==| 20081205 23:10:46 D:\Spiele\WesnothTrunk\src\poolalloc.c||In function 'free':| 20081205 23:10:48 D:\Spiele\WesnothTrunk\src\poolalloc.c|386|error: invalid operands to binary !=| 20081205 23:12:34 Soliton, zookeeper: please guys, you are too funny on #wesnoth. I am pissing my pants. 20081205 23:12:45 :-) 20081205 23:18:19 McShark: Weird. It looks like return value of pthread_self() and pthread_t aren't comparable.. 20081205 23:19:23 Which would make sense in pthread_t is some struct on pthreads_win32 20081205 23:19:42 -!- Baufo_ [n=thomas@62-47-137-17.adsl.highway.telekom.at] has joined #wesnoth-dev 20081205 23:20:07 Which means that Sirp should have used pthread_equal instead if == 20081205 23:20:15 s/if/of 20081205 23:20:32 I'm back soon or later... 20081205 23:21:42 -!- Baufo_ [n=thomas@62-47-137-17.adsl.highway.telekom.at] has quit [Client Quit] 20081205 23:21:43 -!- Baufo [n=thomas@62-47-150-239.adsl.highway.telekom.at] has quit [Connection timed out] 20081205 23:31:55 -!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev 20081205 23:33:46 Hello people. 20081205 23:39:38 baufo * r31294 /trunk/src/server/ (forum_user_handler.cpp forum_user_handler.hpp): Making the forum user handler use the mysql C API rather than the C++ wrappers 20081205 23:40:03 -!- McShark [n=Simon_So@drsd-4db3f35e.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 20081205 23:40:53 I just tried compiling wesnoth with intel c++(non-commercial license) 20081205 23:41:03 And... 20081205 23:41:05 It seems to be as slow as g++ :( 20081205 23:41:22 That sucks. 20081205 23:41:25 But it didn't crash while compiling wesnoth :) 20081205 23:41:34 Yay. 20081205 23:41:41 Good job, intel :) 20081205 23:46:17 -!- happygrue [n=happygru@98.223.188.175] has joined #wesnoth-dev 20081205 23:47:04 -!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev 20081205 23:48:15 Now what? 20081205 23:49:06 Espreon: ? 20081205 23:49:37 -!- happygrue [n=happygru@98.223.188.175] has left #wesnoth-dev [] 20081205 23:49:43 Nothing. Sometimes I say things like "Now what?" or "Oh well..." when I am bored. 20081205 23:54:18 zookeeper * r31295 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/12_The_Final_Confrontation.cfg: Made a random unit (of the highest possible level) be auto-recalled at the entrance every turn after the central body has appeared, and did some minor cleanup. 20081205 23:55:58 ESR_, fendrin__, there's the auto-recall thing 20081205 23:56:10 balance or modify as you see fit 20081205 23:56:51 * Espreon stares at zookeeper. 20081205 23:57:54 -!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Client Quit] 20081205 23:59:52 zookeeper: Thanks! --- Log closed Sat Dec 06 00:00:21 2008