--- Log opened Thu Oct 30 00:00:03 2008 20081030 00:03:31 -!- Elvish_Pillager [n=eli@71-10-234-12.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!"] 20081030 00:35:10 Sirp_: something you may find useful... any paths::route can now be streamed via << to cout, cerr or the logging macros 20081030 00:35:40 Sapient: cool! :) 20081030 00:36:17 cya later, have fun. 20081030 00:36:17 -!- Sapient [n=sapien-x@c-71-59-35-203.hsd1.ga.comcast.net] has left #wesnoth-dev [] 20081030 00:40:53 -!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev 20081030 01:20:13 -!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)] 20081030 01:23:15 -!- ABCD [n=ABCD@wikipedia/ABCD] has joined #wesnoth-dev 20081030 01:40:25 -!- ilor [n=a@unaffiliated/ilor] has quit [Read error: 113 (No route to host)] 20081030 02:20:52 -!- noy [n=Noy@d75-157-43-119.bchsia.telus.net] has quit [] 20081030 03:27:33 -!- Sirp [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has joined #wesnoth-dev 20081030 03:27:44 -!- crimson_penguin [n=ben@64.201.60.220] has joined #wesnoth-dev 20081030 03:33:02 -!- Jetryl [n=Jetryl@mn-10k-dhcp1-151.dsl.hickorytech.net] has quit [Read error: 110 (Connection timed out)] 20081030 03:36:26 -!- [Relic] [n=[Relic]@76.229.220.201] has joined #wesnoth-dev 20081030 03:37:34 <[Relic]> Hello :) 20081030 03:38:18 -!- Jetryl [n=Jetryl@mn-10k-dhcp1-151.dsl.hickorytech.net] has joined #wesnoth-dev 20081030 03:45:39 -!- uso_ [n=uso@84-72-8-253.dclient.hispeed.ch] has quit [Read error: 104 (Connection reset by peer)] 20081030 03:59:57 -!- jon316-brb [n=jon316@dhcp-207-237.resnet.messiah.edu] has joined #wesnoth-dev 20081030 04:00:55 -!- crimson_penguin [n=ben@64.201.60.220] has quit ["Leaving"] 20081030 04:08:04 -!- jon316-brb [n=jon316@dhcp-207-237.resnet.messiah.edu] has left #wesnoth-dev [] 20081030 04:10:10 -!- grrrX [n=_@p57BBFDC8.dip.t-dialin.net] has quit ["Ex-Chat"] 20081030 04:10:54 -!- Ivanovic_ [n=ivanovic@dtmd-4db21a99.pool.einsundeins.de] has joined #wesnoth-dev 20081030 04:27:26 -!- Ivanovic [n=ivanovic@dtmd-4db2216a.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)] 20081030 04:28:52 -!- Ivanovic_ is now known as Ivanovic 20081030 04:35:57 -!- uso [n=uso@84-72-8-253.dclient.hispeed.ch] has joined #wesnoth-dev 20081030 04:51:00 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit [Remote closed the connection] 20081030 04:58:35 -!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev 20081030 05:00:08 -!- Psyche^ [n=Psyche@e177233172.adsl.alicedsl.de] has joined #wesnoth-dev 20081030 05:10:26 -!- Patterner [n=Psyche@e177239138.adsl.alicedsl.de] has quit [Read error: 110 (Connection timed out)] 20081030 05:10:27 -!- Psyche^ is now known as Patterner 20081030 05:14:47 -!- uso_ [n=uso@84-72-8-253.dclient.hispeed.ch] has joined #wesnoth-dev 20081030 05:14:59 -!- uso [n=uso@84-72-8-253.dclient.hispeed.ch] has quit [Read error: 110 (Connection timed out)] 20081030 05:15:15 -!- uso_ [n=uso@84-72-8-253.dclient.hispeed.ch] has quit [Read error: 104 (Connection reset by peer)] 20081030 05:16:45 -!- uso [n=uso@84-72-8-253.dclient.hispeed.ch] has joined #wesnoth-dev 20081030 05:21:05 -!- Jetryl is now known as Jetrel 20081030 05:26:52 -!- kthakore2 [n=kthakore@99.255.245.40] has joined #wesnoth-dev 20081030 06:38:22 -!- [Relic] [n=[Relic]@76.229.220.201] has quit ["Leaving"] 20081030 07:27:24 -!- Sirp [n=me@c-76-102-104-121.hsd1.ca.comcast.net] has quit ["leaving"] 20081030 07:46:45 esr * r30416 /trunk/data/campaigns/Legend_of_Wesmere/scenarios/24_Epilogue.cfg: Minor wording tweaks. 20081030 08:05:21 -!- BenUrban [n=benurban@unaffiliated/benurban] has quit ["Power failu"] 20081030 08:34:16 -!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev 20081030 08:34:38 morning 20081030 08:35:14 esr regarding the bay of pearls bug, it might be it's old (I think I saw it quite a while ago as well) maybe test with 1.4.0 20081030 08:36:01 I've never even *seen* 1.4.0 :-) 20081030 08:37:04 Oh, wait, I have. Misread the number. 20081030 08:38:21 But supposing that's true, it would reinforce my suspicion that suokko (and, later, Soliton) tickled a bad interaction that has been present for a very long time. 20081030 08:41:36 esr * r30417 /trunk/data/tools/trackplacer: trackplacer: Don't reread generated *_END macros as new data. 20081030 08:52:01 -!- loonybot [n=loonybot@217.10.38.214] has joined #wesnoth-dev 20081030 08:52:05 yes it might be bad interaction or just a stupid bug 20081030 08:52:30 just looked at BoP and the AI is indeed very timit 20081030 08:52:34 timid* 20081030 08:53:11 -!- loonycyborg [n=sergey@217.10.38.214] has joined #wesnoth-dev 20081030 08:55:26 -!- ettin_ [n=jorda@7.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev 20081030 09:08:35 -!- ettin [n=jorda@56.207.219.87.dynamic.jazztel.es] has quit [Read error: 110 (Connection timed out)] 20081030 09:33:43 -!- loonybot [n=loonybot@217.10.38.214] has quit [Read error: 110 (Connection timed out)] 20081030 10:15:17 -!- loonycyborg [n=sergey@217.10.38.214] has quit ["KVIrc 3.4.0 Virgo http://www.kvirc.net/"] 20081030 10:25:08 -!- EdB [n=EdB@79.88.117.20] has joined #wesnoth-dev 20081030 10:58:04 -!- Elvish_Pillager [n=eli@71-10-234-12.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081030 10:59:24 -!- zookeeper [n=l@dsl-tkubrasgw1-fe11fa00-51.dhcp.inet.fi] has joined #wesnoth-dev 20081030 11:26:34 zookeeper: http://gna.org/patch/?1109 somethin for you to look at, i think. 20081030 11:32:16 roger 20081030 11:32:48 Rh0nda: patch 1109 20081030 11:32:58 * Rh0nda . o O ( how do I trigger myself? ) 20081030 11:53:10 -!- Elvish_Pillager [n=eli@71-10-234-12.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!"] 20081030 11:54:50 -!- loonybot [n=loonybot@217.10.38.214] has joined #wesnoth-dev 20081030 11:55:28 -!- loonycyborg [n=sergey@217.10.38.214] has joined #wesnoth-dev 20081030 12:10:32 -!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev 20081030 12:12:16 -!- loonycyborg [n=sergey@217.10.38.214] has quit [Read error: 60 (Operation timed out)] 20081030 12:12:36 -!- loonycyborg [n=sergey@217.10.38.214] has joined #wesnoth-dev 20081030 12:32:48 -!- loonycyborg [n=sergey@217.10.38.214] has quit ["KVIrc 3.4.0 Virgo http://www.kvirc.net/"] 20081030 12:33:16 -!- loonybot [n=loonybot@217.10.38.214] has quit [Remote closed the connection] 20081030 12:55:18 -!- loonybot [n=loonybot@217.10.38.214] has joined #wesnoth-dev 20081030 12:59:09 -!- loonycyborg [n=sergey@217.10.38.214] has joined #wesnoth-dev 20081030 13:11:19 -!- grrrX [n=_@p57BBF279.dip.t-dialin.net] has joined #wesnoth-dev 20081030 13:14:06 -!- Sapient [n=patrickp@agentek.com] has joined #wesnoth-dev 20081030 13:14:14 o 20081030 13:14:37 hi there 20081030 13:14:39 -!- mjs-de [n=mjs-de@vpw.wh.Uni-Dortmund.DE] has joined #wesnoth-dev 20081030 13:14:42 hi Sapient 20081030 13:26:21 -!- EdB [n=EdB@79.88.117.20] has quit ["Konversation terminated!"] 20081030 13:28:23 -!- loonybot [n=loonybot@217.10.38.214] has quit [Remote closed the connection] 20081030 13:39:03 zookeeper: got my PM? 20081030 13:39:28 I posted it on the Useful WML Fragments page as well 20081030 13:40:01 under the "Advanced" section ;) 20081030 13:40:40 Sapient, i did, but i actually got it working my way after all 20081030 13:41:06 it took some careful reordering of all those storings and filterings 20081030 13:47:37 -!- loonycyborg [n=sergey@217.10.38.214] has quit [Read error: 110 (Connection timed out)] 20081030 13:48:22 ah well, I like my way better :) 20081030 13:49:17 congrats on getting yours working though 20081030 13:53:12 actually the loop itself is almost bizarrely simple now: http://rafb.net/p/nRmFSz61.html 20081030 13:53:54 like, no need for even the one [if] 20081030 13:55:35 yeah, yours is simpler 20081030 13:55:48 but mine can handle the generic case 20081030 13:56:24 so you could easily add 3 mp terrains, etc 20081030 13:57:11 it looks like urza is back in business 20081030 13:57:42 sure 20081030 13:58:06 maybe you should write an example of the exact use that macro too ;)? 20081030 13:58:33 i mean, like some simple event which puts an item image on all reachable hexes on moveto or something 20081030 14:00:19 maybe... I'd like to improve it so it can build move cost info for a unit automatically 20081030 14:00:47 then it would probably be more useful 20081030 14:08:06 -!- crimson_penguin [n=ben@64.201.60.220] has joined #wesnoth-dev 20081030 14:53:33 -!- Patterner [n=Psyche@e177233172.adsl.alicedsl.de] has quit [Read error: 113 (No route to host)] 20081030 14:54:51 zookeeper * r30418 /trunk/data/ (2 files in 2 dirs): Moved the SCATTER_UNITS macro to core. 20081030 15:05:12 -!- loonycyborg [n=sergey@79.139.246.10] has joined #wesnoth-dev 20081030 15:34:39 -!- Patterner [n=Psyche@e177233172.adsl.alicedsl.de] has joined #wesnoth-dev 20081030 16:05:38 -!- Dragonking [n=dk@dedikerad/dragonking] has quit [Read error: 60 (Operation timed out)] 20081030 16:08:32 -!- grrrX [n=_@p57BBF279.dip.t-dialin.net] has quit ["Ex-Chat"] 20081030 16:19:51 -!- stikonas [n=and@ctv-79-132-166-125.vinita.lt] has joined #wesnoth-dev 20081030 16:25:49 Is it normal that most of the wesnoth-umc-dev campaigns won't even load the first scenario? 20081030 16:26:58 * loonycyborg never tried any wesnoth-umc-dev campaigns except IftU 20081030 16:36:14 "Invasion from the Unknown Scenario '01_Border_Patrol'"..? 20081030 16:42:08 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 16:42:19 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has quit [Read error: 104 (Connection reset by peer)] 20081030 16:42:42 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 16:44:12 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has quit [Read error: 104 (Connection reset by peer)] 20081030 16:44:30 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 17:05:15 -!- irm [i=c9f6dfc8@gateway/web/ajax/mibbit.com/x-eaf83344e2d6d68c] has joined #wesnoth-dev 20081030 17:06:31 -!- irm is now known as Shadow_Master 20081030 17:08:49 Patterner: no, it isn't 20081030 17:09:26 unfortunately I won't be able to fix it any soon... if esr, Espreon, AI0867 or Mica want to fix stuff in IftU, feel free to do it for me, but I'd advice a wmllint run first 20081030 17:09:45 Shadow_Master: Okay, I can let the others know 20081030 17:10:43 Shadow_Master: I also am sorry to hear about what happened to your comp, hope you can get it fixed soon. But there's no rush. 20081030 17:12:14 yeah. I.m unsure whether I can afford repairs :( 20081030 17:12:47 I'll also need any of them to commit kitty's portraits for me, let hem know that 20081030 17:12:54 That is unfortunate. 20081030 17:13:02 Oh, I could do that. 20081030 17:13:40 okay, great 20081030 17:14:15 I also submitted a patch that was a pending commit lost in my laptop - hopefully the code monkeys are back this time? ;) 20081030 17:15:10 ah 20081030 17:15:51 Shadow_Master: What ones in particular did you want committed? 20081030 17:16:41 Mica: you cannot commit anything to the campaign images/portraits dir directly until kitty publishes the 205x205 versions 20081030 17:17:05 Shadow_Master: Okay, then I'll wait. :) Is there anything else you needed me to do? 20081030 17:17:36 however, you can commit the latest transparent Galas to images/portraits/transparent (make the dir) now, and wait for kitty to decide that Elynia is ready for commit that other potrait (and so on) 20081030 17:18:17 those are my last wishes I think 20081030 17:19:10 Okay 20081030 17:19:24 You sound like someone going to die 20081030 17:19:56 heh 20081030 17:20:02 Well, it might seem like it, because his comp is broken for now. 20081030 17:20:15 The difference is neglible. 20081030 17:20:28 Shadow_Master: Recently Sirp bought a laptop using wesnoth money. Perhaps he can finance you too.. 20081030 17:20:40 bah, I don't need that 20081030 17:21:08 loonycyborg: Man, gotta love those "business expenses," right? :p 20081030 17:21:20 I'd feel like a bug ("parásito"?) feeding upon Wesnoth instead of contributing to it (I'm no lead developer, nor senior developer) 20081030 17:21:58 I think you mean parasite 20081030 17:22:07 loonycyborg: btw last time I checked trunk (one week ago), just before I could not turn on the laptop anymore, I noticed that scons isn't detecting Python 2.6 20081030 17:22:12 s/2.6/2.5/ 20081030 17:22:30 Shadow_Master: Shame on you! Don't feel that way! You are very helpful around here. Maybe lately not for mainline, but UMC you are very, very helpful and in fact none of my contributions would be anywhere near where they are now if it weren't for you, and probably EoS and others as well. 20081030 17:22:36 I sure hope that was fixed in the mean time ;) 20081030 17:22:37 I'll have a look at your patch after dinner and some minor changes 20081030 17:23:09 Shadow_Master: It was never broken for me. 20081030 17:23:42 k, perhaps I should upgrade scons to 1.0 or stick to autotools 20081030 17:23:52 * Shadow_Master (me has had 20081030 17:23:57 Most likely it was just another sconf bug which could be overcame with --config=force 20081030 17:24:01 is the patch tested? 20081030 17:24:29 Mordante: yeah, it compiles, and the isolated bugs were fixedf 20081030 17:24:54 (that is version_info v1("0.2.2") and version_info v2("0.2.2") are no longer v1 < v2) 20081030 17:25:26 (and some other operator weirdness too) 20081030 17:26:16 besides that, it should be the only thing making Update Add-ons not work as expect 20081030 17:26:31 *ed (public cafe keyboards ftl) 20081030 17:27:37 anything needed in the changelogs? 20081030 17:30:52 um, mention "Fixed add-on version comparison logic" I guess. 20081030 17:31:23 There is no reported bug attached AFAIK (except a few complaints at the forum under OT section...) 20081030 17:32:46 I so tend to forget changelog updates until before a relelease eve. 20081030 17:36:27 well applied without problem :) 20081030 17:36:57 mordante * r30419 /trunk/src/help.cpp: Update doxygen comment style. 20081030 17:37:02 mordante * r30420 /trunk/src/intro.cpp: Update doxygen comment style. 20081030 17:37:06 mordante * r30421 /trunk/src/intro.hpp: Update doxygen comment style. 20081030 17:37:10 mordante * r30422 /trunk/src/ (loadscreen.cpp loadscreen.hpp): Update doxygen comment style. 20081030 17:37:15 mordante * r30423 /trunk/src/ (marked-up_text.cpp marked-up_text.hpp): Update doxygen comment style. 20081030 17:37:18 mordante * r30424 /trunk/src/ (network.cpp network.hpp): Update doxygen comment style. 20081030 17:37:25 mordante * r30425 /trunk/src/tests/test_network_worker.cpp: Update doxygen comment style. 20081030 17:37:27 mordante * r30426 /trunk/src/tools/cutter.cpp: Update doxygen comment style. 20081030 17:37:31 mordante * r30427 /trunk/src/tools/dummy_video.cpp: Update doxygen comment style. 20081030 17:37:34 mordante * r30428 /trunk/src/tools/key_test.cpp: Update doxygen comment style. 20081030 17:37:37 mordante * r30429 /trunk/src/wesconfig.h: Update doxygen comment style. 20081030 17:37:41 mordante * r30430 /trunk/src/widgets/scrollarea.cpp: Update doxygen comment style. 20081030 17:37:45 mordante * r30431 /trunk/src/widgets/scrollarea.hpp: Update doxygen comment style. 20081030 17:37:50 mordante * r30432 /trunk/src/widgets/ (scrollbar.cpp scrollbar.hpp): Update doxygen comment style. 20081030 17:37:52 mordante * r30433 /trunk/src/widgets/ (scrollpane.cpp scrollpane.hpp): Update doxygen comment style. 20081030 17:37:55 mordante * r30434 /trunk/src/loadscreen_empty.cpp: Update doxygen comment style. 20081030 17:38:02 mordante * r30435 /trunk/ (changelog src/version.cpp): Fixed addon update version logic (patch #1110). 20081030 17:38:13 mordante * r30436 /trunk/changelog: Fix alphabetic order in changelog. 20081030 17:38:13 * Shadow_Master looks forward to download the git snapshot someday 20081030 17:39:11 all doxygen comment has now finally been converted :) 20081030 17:39:44 What was wrong with old comment style? 20081030 17:39:52 shikadibot: file trunk/src/thread.hpp 20081030 17:39:52 Web interface URL to file trunk/src/thread.hpp: http://svn.gna.org/viewcvs/wesnoth/trunk/src/thread.hpp?view=log 20081030 17:40:21 mordante: not really, see that ^ 20081030 17:40:42 there are lots of non-doxygen comments that would do fine doxygenized there 20081030 17:41:01 specially the "// \pre f != NULL" ... look familiar ? ;) 20081030 17:42:47 meh 20081030 17:43:01 loonycyborg consistency 20081030 17:46:36 * Shadow_Master out 20081030 17:46:56 -!- Shadow_Master [i=c9f6dfc8@gateway/web/ajax/mibbit.com/x-eaf83344e2d6d68c] has left #wesnoth-dev [] 20081030 17:50:45 Leaving for now, bye. 20081030 17:50:48 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has quit ["What did you say? I can't hear you, my connection's gone."] 20081030 18:30:37 hi 20081030 18:31:15 hi Ivanovic 20081030 18:32:58 -!- Baufo [n=thomas@62-47-159-45.adsl.highway.telekom.at] has joined #wesnoth-dev 20081030 18:39:21 -!- Baufo [n=thomas@62-47-159-45.adsl.highway.telekom.at] has quit [Remote closed the connection] 20081030 18:40:21 -!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)] 20081030 18:50:31 -!- kthakore2 [n=kthakore@99.255.245.40] has quit ["Leaving"] 20081030 18:55:53 esr wesnoth 1.4.0 doesn't seem to have the problem with BoP 20081030 19:02:30 -!- stikonas [n=and@ctv-79-132-166-125.vinita.lt] has quit [Remote closed the connection] 20081030 19:03:35 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 19:07:26 -!- miq [n=miq@krlh-5d83abf9.pool.einsundeins.de] has joined #wesnoth-dev 20081030 19:09:08 -!- noy [n=Noy@d75-157-43-119.bchsia.telus.net] has joined #wesnoth-dev 20081030 19:16:39 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has quit ["What did you say? I can't hear you, my connection's gone."] 20081030 19:17:34 mordante: just heard some good news from the pandora guys 20081030 19:17:42 the pandora will have 256MB ram 20081030 19:17:44 what? 20081030 19:17:49 so wesnoth will be *no* problem at all 20081030 19:17:55 even lowmem should not be needed... 20081030 19:18:15 hmm, where's the challenge in that :( ;) 20081030 19:18:27 there is none 20081030 19:18:28 ha THoT can eat 400 MB 20081030 19:18:35 beside making the fonts readable 20081030 19:18:38 oh, that is bad 20081030 19:18:44 screw you, esr... 20081030 19:18:58 365 turns is also insane ;) 20081030 19:19:31 yes, it is 20081030 19:19:36 Ivanovic: ??? 20081030 19:19:38 esr: you should not allow this at all! 20081030 19:19:48 esr: (possible) ram usage with thot 20081030 19:19:49 Allow what? 20081030 19:20:22 And how am I to control that, or even know what it is? 20081030 19:20:24 * mordante knows quite some ways for UMC to increase the memory usage ;) 20081030 19:20:27 [19:18:28] ha THoT can eat 400 MB 20081030 19:20:29 [19:18:58] 365 turns is also insane ;) 20081030 19:20:41 there should be no *need* to use this many turns 20081030 19:21:10 Aha. I'll bet he's playing Underlevels. 20081030 19:21:23 probably 20081030 19:21:27 lucky guess :P 20081030 19:21:27 does the memory usage drop if you quit the game and just reload your turn 365 save? 20081030 19:21:40 no idea 20081030 19:21:47 not sure don't have the saves around anymore 20081030 19:21:48 if it does, then wouldn't that be a memory leak? 20081030 19:22:23 Well...I asked Taurus for a big final dungeon crawl, and I guess what I got was large in every sense. 20081030 19:23:20 anyways, i'd expect a lot of the memory usage in huge scenarios to come from a huge number of units in play at the same time, so one way to alleviate that would be to only spawn units when the player actually gets close to them. of course that's not always possible to do in a sane way. 20081030 19:23:20 All the ones I wrote, before that, are turn-limited. 20081030 19:23:35 yep but I think it's the only place where you get such huge memory usage 20081030 19:24:15 but at least Ivanovic now has a nice challenge to get Wesnoth working on the pandora :P 20081030 19:24:18 * mordante hides 20081030 19:24:42 i guess the replay data alone for such a huge number of units would be pretty massive too. 20081030 19:24:48 esr: Taurus is a sadist. How does he expact someone to manage a battle with around 100 unit under one's command? :) 20081030 19:25:12 mordante: i just won't ship this campaign 20081030 19:25:16 haha, easy fix! 20081030 19:25:36 I didn't use that many units, I even had a lot near the original keep idling 20081030 19:25:38 I' don't think I've ever seen it go to 100. I usually max out at about 30 units playing that. 20081030 19:25:56 esr: I was referring to NR. 20081030 19:26:12 ah yes that's also a huge one 20081030 19:26:33 Some of us like large battles, you know. 20081030 19:26:34 Ivanovic how else should we trick you into coding :( 20081030 19:27:21 mordante: a little will be needed for input stuff 20081030 19:27:24 ;) 20081030 19:27:28 I like huge battles as well :) 20081030 19:27:33 esr: I enjoy large battles too if I can control units effectively. 20081030 19:27:43 Ivanovic :) 20081030 19:27:49 But wesnoth is not an RTS. 20081030 19:27:58 beside this: i might try to optimize some stuff 20081030 19:28:00 I guess it must be my history as a hex wargamer. I wouldn't even find running 100 units very intimidating, not that a Wesnoth game ever gets that high. 20081030 19:28:00 -!- grrrX [n=_@p57BBF279.dip.t-dialin.net] has joined #wesnoth-dev 20081030 19:28:08 zookeeper how broken is ANL in trunk 20081030 19:28:38 Ivanovic no more than 1 #ifdef per function is allowed :P 20081030 19:29:00 mordante, the last time i tried it we ended up with massive OOS all over the place. 20081030 19:29:30 any ideas why? 20081030 19:29:34 mordante: :( 20081030 19:29:42 not that i remember really 20081030 19:29:52 Soliton, ? 20081030 19:29:54 mordante: only a single one, or a singly "type" 20081030 19:30:07 that is the "type" of ifdefs for lowmem 20081030 19:30:22 since this will be what i will probably using most time when optimizing for ram usage 20081030 19:31:10 just want to avoid you making wesnoth an ifdef hell, I know you know how to make an ifdef hell ;) 20081030 19:32:38 * Ivanovic has not seen this rule in any coding guidelines for wesnoth... 20081030 19:32:40 ;) 20081030 19:33:08 I was expecting a more creative answer ;) 20081030 19:33:24 hey, as long as it is not in the guidelines, it is allowed... 20081030 19:33:49 ^^ 20081030 19:34:33 * mordante stops using if, for and while and starts to good old goto :P 20081030 19:35:00 beside this: why should i try to find a creative answer beside: "hey this trick works and reduces mem usage" 20081030 19:35:03 ;) 20081030 19:36:33 I would have threatened to turn every place where I need an extra ifdef into a new function ;) 20081030 19:38:12 mordante, naturally it was most likely related to the custom menu item usage 20081030 19:38:29 i'm just not sure how 20081030 19:38:54 (well, it might have been something else too, but i'd always suspect menu items first ;) 20081030 19:39:04 :) 20081030 19:40:41 -!- boucman [n=rosen@50.83.102-84.rev.gaoland.net] has joined #wesnoth-dev 20081030 19:41:30 -!- boucman [n=rosen@50.83.102-84.rev.gaoland.net] has left #wesnoth-dev [] 20081030 19:44:18 -!- boucman [n=rosen@50.83.102-84.rev.gaoland.net] has joined #wesnoth-dev 20081030 19:45:00 -!- grrrX [n=_@p57BBF279.dip.t-dialin.net] has quit ["Ex-Chat"] 20081030 19:57:06 -!- stikonas [n=and@ctv-79-132-166-125.vinita.lt] has joined #wesnoth-dev 20081030 20:05:19 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 20:13:39 In case anyone is interested, I'm in the process of doing custom music playlists for all of mainline. I've done aOI, DiD, EI, and Liberty so far and plan to go over all the rest. 20081030 20:14:40 uh. 20081030 20:15:21 Was that "uh" directed at me? 20081030 20:15:27 leave out SoF, SotBE, HttT and TRoW. i've done those already, and would kinda like to keep them in my hands if you don't mind ;) i'll gladly take suggestions of course 20081030 20:15:47 Oh, OK. 20081030 20:16:42 * Ivanovic welcomes suggestinos to two brothers, too 20081030 20:16:51 though they should already have rather well working lists 20081030 20:16:58 How recently did you do them? Are Kaufman's latest pieces in your lists? 20081030 20:17:01 (the "really new music" might be missing, not sure) 20081030 20:17:04 -!- Dragonking [n=dk@dedikerad/dragonking] has joined #wesnoth-dev 20081030 20:17:16 Ivanovic: I'll give it a look. 20081030 20:18:38 esr, heroes rite is 20081030 20:19:30 -!- maxy [n=maxy@80-219-1-12.dclient.hispeed.ch] has joined #wesnoth-dev 20081030 20:19:45 i haven't wired in love_theme.ogg anywhere 20081030 20:19:55 so if you got suggestions on that, do tell 20081030 20:20:12 I'm actually doing DM now, not that that is mainline yet but I've been working hard at bringing it up to standard. I think we're going to cut another senario BTW, bring it down to 17 battles and 5 cut scenes. 20081030 20:20:26 (i've also wired in gmlion's pieces where i deemed appropriate) 20081030 20:21:28 I'm not using love-theme either, except for actual,ove scxenes. Though I can think of one deliberate exception I'm going to make to that in DM, involving Chasntal. 20081030 20:22:29 Which ones are gmlions? 20081030 20:22:29 damn, i knew that this was a mistake 20081030 20:22:35 ? 20081030 20:22:40 * Ivanovic is currently trying to catch up with forum posts he missed over the last week 20081030 20:22:44 deep path and dangerous symphony 20081030 20:23:36 -!- Elvish_Pillager [n=eli@71-10-234-12.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev 20081030 20:25:03 although i still need to add the deep path to the TRoW underground scenarios 20081030 20:25:25 Yeah. Deep Path is great stuff for scenes with necromancers in them; I'm going to lean on it pretty heavily for Iliah-Malal's appearances in DM. 20081030 20:25:55 i've mostly just kept it for underground scenarios 20081030 20:26:18 Works there, too. I used it that way in EI. 20081030 20:26:54 i figured it's better to not use the same piece in widely different kinds of settings, just in order to kinda associate the piece more strongly with a certain setting 20081030 20:29:17 I tend to do it the same way. I consider each piece to have a mood and try to be consistent about only applying it where the particular mood will fit. 20081030 20:29:56 are there pieces in the forum that are good to go ? 20081030 20:30:08 i don't recall there being any now 20081030 20:30:11 we should do a roundup of all WIP with west & co 20081030 20:30:22 Haven't checked lately. I should. 20081030 20:31:16 should be a good time to post a request for music on the forum... before you put too much work on your playlists 20081030 20:31:34 Also, I try to maintain a minimum-distance rule - no repeat of ther same piece in at least 4 scenarios. 20081030 20:33:09 boucman: I'm not too worried about that. Because the new MUSIC macros are easy to find by grep, incremental tweaks to playlists are really easy -- you just grep the while escenario and you can see every title in front of you. So folding new stuff in gracefully is easy. 20081030 20:33:38 s/escenario/scenario directory/ 20081030 20:34:09 esr: previous report (from the music dev, that is) was that the long part was actually playing the campaigns to get the mood :) 20081030 20:34:14 yes, that's what i've done, it's nice 20081030 20:34:25 that was from the thread where the music devs considered being responsible for wiring music 20081030 20:35:10 boucman: I've test-played all of mainline enough times that mostly I just have to *remember* what the mood is like. 20081030 20:35:19 true 20081030 20:35:53 i also got all the campaigns except the ones i don't deal with pretty much memorized :p 20081030 20:36:20 boucman can Jetrel's naga animation problem be solved with haloes? 20081030 20:36:35 It helps that so much of the intro and storyline prose has had my hands on it. Rer-reading it tends to remind me of the emotional effect I was after in writing it. 20081030 20:37:01 mordante: maybe 20081030 20:37:15 -!- stikonas [n=and@ctv-79-132-166-125.vinita.lt] has quit [Read error: 110 (Connection timed out)] 20081030 20:37:21 I havn't dealt with 1.4 for quite some time, so the documentation would be more reliable than my memory 20081030 20:37:31 (considering I did a big pass on 1.4 code to write it) 20081030 20:37:47 Jetrel: see mordante's comment above whe you're around 20081030 20:38:23 I'll already mention it before, but not sure whether he saw it, but I'm in the forum now so post there 20081030 20:38:32 k, thx 20081030 20:39:01 np 20081030 20:39:27 zookeeper: Is it OK if I wire in the victory and defeat music where you don't have it? 20081030 20:40:09 esr, i'd rather not see the VICTORY_AND_DEFEAT_MUSIC macro used anywhere, it's horrible :x 20081030 20:40:36 Really? Explain -- I might stop using it. 20081030 20:41:11 it inserts silly victory and defeat events into the scenario, which then trigger the music. the events can trigger before or after the scenario's regular events 20081030 20:41:49 the default victory and defeat music should be wired in [game_config], and an option to override it provided in [endlevel] 20081030 20:41:56 inserting some events into every scenario is really silly 20081030 20:43:01 Oh, so it's the implementation that's broken rather than the concept. I see. 20081030 20:43:01 and data/game_config.cfg actually does have default_victory_music= and default_defeat_music= defined as they should be 20081030 20:43:05 yes 20081030 20:43:11 having victory and defeat music is just fine 20081030 20:44:53 Do default_victory_music= and default_defeat_music= take a comma-separated list and randon-shuffle it? 20081030 20:45:17 yes 20081030 20:46:02 Then why does VICTORY_AND_DEFEAT_MUSIC exist at all? We should nuke it instantly. 20081030 20:46:34 good question ;) i think it was some cycholka's early hack 20081030 20:46:58 OK, I'm going to shoot it through the head *now*. 20081030 20:47:03 well 20081030 20:47:11 i'm not sure if the default mechanism works 20081030 20:47:16 i'll try it 20081030 20:47:23 OK. 20081030 20:47:38 but if it doesn't, it should be easy to fix i guess 20081030 20:48:17 Let me know either way. This cleanup should definitely get done before 1.6 feeze. 20081030 20:49:24 ah yes, seems to be working. 20081030 20:49:47 so unless i'm missing something, you can and should nuke the macro 20081030 20:49:51 -!- crimson_penguin [n=ben@64.201.60.220] has quit [Read error: 104 (Connection reset by peer)] 20081030 20:50:01 OK, I'll clean this mess up. Recursive grep is our friend. 20081030 20:50:07 -!- crimson_penguin [n=ben@64.201.60.220] has joined #wesnoth-dev 20081030 20:50:47 * esr has learned to be wary when he hears "cycholka hack" 20081030 20:50:58 he likes hacks 20081030 20:51:08 sometimes they're a bit...uh, ugly ;) 20081030 20:52:17 Wgat he did the the storyline text in aOI doesn't bear describing. People with a weak grasp of English should *not* try modifying my prose. 20081030 21:00:08 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has quit ["What did you say? I can't hear you, my connection's gone."] 20081030 21:01:15 -!- BenUrban [n=benurban@unaffiliated/benurban] has joined #wesnoth-dev 20081030 21:07:10 -!- grzywacz [n=grzywacz@moinmoin/developer/karol] has joined #wesnoth-dev 20081030 21:08:55 -!- crimson_penguin [n=ben@64.201.60.220] has quit [Connection timed out] 20081030 21:16:22 esr * r30437 /trunk/data/ (65 files in 8 dirs): Remove uneeded VICTORY_AND_DEFEAT_MUSIC macro, and fix a whitespace glitch. 20081030 21:20:46 -!- Sapient [n=patrickp@agentek.com] has quit [Read error: 110 (Connection timed out)] 20081030 21:26:11 zookeeper: have you had a look at http://www.wesnoth.org/forum/viewtopic.php?p=317736#p317736 ? 20081030 21:26:55 Ivanovic, yes, i'm going to change all that stuff somewhat soon 20081030 21:27:37 okay 20081030 21:27:43 -!- Mica [n=Manic_Mi@d14-69-222-189.try.wideopenwest.com] has joined #wesnoth-dev 20081030 21:38:55 -!- maxy [n=maxy@80-219-1-12.dclient.hispeed.ch] has quit [Remote closed the connection] 20081030 21:43:33 -!- crimson_penguin [n=ben@64.201.60.220] has joined #wesnoth-dev 20081030 21:53:37 esr just curious did you already publish your C++ paper? 20081030 21:53:41 hiya 20081030 21:53:46 oops, wc 20081030 21:54:08 watch out for the plague. 20081030 21:54:20 :) 20081030 21:54:40 mordante: Nope, we didn't finish it before my collabrator had to fly home. We're planning to work on it more over Christmas. 20081030 21:55:14 ok then I'll ask again next year ;) 20081030 21:55:34 -!- grrrX [n=_@p57BBF279.dip.t-dialin.net] has joined #wesnoth-dev 20081030 21:56:52 -!- loonycyborg_ [n=sergey@79.139.247.109] has joined #wesnoth-dev 20081030 21:57:11 -!- loonycyborg [n=sergey@79.139.246.10] has quit [Nick collision from services.] 20081030 21:57:15 -!- loonycyborg_ is now known as loonycyborg 20081030 22:05:47 esr didn't you already fix bug 12532? 20081030 22:06:16 Checking... 20081030 22:06:54 and I'm surprised you figured out bug 12524 was for me ;) 20081030 22:08:35 You're right, I did fix it. I've changed the status. 20081030 22:13:14 nice 20081030 22:33:05 -!- me [n=me@athedsl-4499278.home.otenet.gr] has joined #wesnoth-dev 20081030 22:33:11 -!- me is now known as kalenzz 20081030 22:36:22 kalenzz: Hey, I just finished making custom music lists for DM. 20081030 22:38:47 great 20081030 22:39:23 I also sent you a brief reply on the timeline issues. Is there a problem shifting the official timeline? 20081030 22:41:56 It might have ripple effects on other campaigns. I really don't want to mess with things like regnal dates. Are you really advocating shifting HttT rather than DM? If so, why? Seems to me there'd be less risk of disrupting other parts of the chronology if we moved DM. 20081030 22:42:24 (I think I agree about moving EI later, but that's a separate issue.) 20081030 22:44:04 why is there a chronology disruption? After HttT we really do not have much in mainline that will be seriously affected 20081030 22:44:48 The problem is really not DM per se, but the distance in the timeline between Delf's graduation and HttT 20081030 22:45:25 whether we absorb years in Land of the dead or in Methor training, it boils down to the same thing 20081030 22:45:26 -!- Sapient [n=sapien-x@c-71-59-35-203.hsd1.ga.comcast.net] has joined #wesnoth-dev 20081030 22:45:34 yo 20081030 22:45:37 Well, for one thing, I'd have to hack the Garardine Dynasty page, including a geneology tree that's a *graphic* (ugh). It would be mssy and I'd rather not. 20081030 22:45:50 kalenzz: did you manage to convert your maps yet? 20081030 22:46:07 Hey, Sapient. Was my anaysis of the AI problem helpful? 20081030 22:46:52 esr: I'm planning to hand that over to Sirp now... sorry 20081030 22:47:06 free time has been pwned 20081030 22:47:16 I understand. Ot's his code. 20081030 22:47:54 Sapient, in the meantime I had a recollection of the old map and did something close to it from scratch 20081030 22:48:07 kk 20081030 22:48:37 kalenzz: What would bother you about shifting DM forward 215-28 years. That would solve the problem. (Since you mentioned it, I think I'll also move EI, but as a separate operation. 20081030 22:48:47 s/215/25/ 20081030 22:48:52 esr, we have 55 years between the valey belongs to me and Prince of Wesnoth. If Delf is 30 years by the valley belong to me 20081030 22:49:05 -!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev 20081030 22:49:20 +55+15 years at least to raise Konrad, that makes 100 years 20081030 22:49:40 assuming HttT starts immediately after Prince of Wesnoth 20081030 22:50:22 Shifting DM would not bother me 20081030 22:50:31 OK, but now let's say we shift DM futureward 25 years. The Delf will be 80 in 517 - much morecreasonable. 20081030 22:50:58 esr: although I do thank you for that details summary of all the revisions 20081030 22:51:01 Right, and we do not need to make internal DM changes other than shifts 20081030 22:51:11 kalenzz: Correct. 20081030 22:51:23 it's fine with me 20081030 22:52:09 kalenzz: OK, I'll change as indicated, then. May require some very minor tweak to DM to write in Garard II, but otherwise it's just changes to the main timeline. 20081030 22:53:27 Ok, you may go ahead, but I will not touch DM until I get back from the States around Nov 11 and then if the AI is fixed, I will first check to make sure LoW works ok(=playtest) 20081030 22:53:39 Sapient: Do you agree with my guess that something is borked in the pathfinding/AI interaction, or do you suspect something else going on. 20081030 22:53:51 kalenzz: OK. 20081030 22:54:09 btw, Sapient, where are we with the AI issues in trunk? 20081030 22:54:20 esr: yes, it's either the pathfinding code itself or the ai move code 20081030 22:54:34 kalenzz: check the development mailing list 20081030 22:54:46 ok, thanks 20081030 22:55:10 https://mail.gna.org/public/wesnoth-dev/2008-10/msg00077.html 20081030 22:55:11 kalenzz: Only remaining issue is where Battle for Aethenwood should land. Do you want to pull it back into LoW somehow? 20081030 22:56:20 esr: I am highly suspicious of the pathfind or starsearch code because it would be likely to cause subtle problems with all the places in ai code that relied on historical behavior 20081030 22:57:02 Sapient: I share your suspcion, strongly. I think we need a major cleanup there. 20081030 22:57:33 I think we need OOP but I know Sirp would disagree :) 20081030 22:58:07 instead we have structs with uncommented public data members 20081030 22:58:27 which is, IMO, just asking for trouble 20081030 22:59:02 Yep. This is not linux kernel, you know :) 20081030 23:00:21 Sapient: If Sirp isn't willing to use OOP then why the hell are we in C++? 20081030 23:00:47 C++ has a lot of benefits over C if that's what you mean 20081030 23:01:07 Because there's smart_ptrs :) 20081030 23:01:18 * esr restrains himself from commenting, with difficulty. 20081030 23:01:22 esr, as Low is now, Battle for Aethenwood does not fit; I think it should be either in DM or axed 20081030 23:01:47 Axed, then. I'll put it in the resources tree. 20081030 23:01:52 ok 20081030 23:02:12 c is an old byte processing language without templates, exceptions, namespaces, constructors/destructors (and therefore RAII), virtual function polymorphism, references, operator/function overloading, reusable standard generic containers, or explicitly named casts. 20081030 23:02:27 * Sapient quoting from nolyC ;) 20081030 23:04:16 kalenzz: Thanks.. With these changes, I think DM will be in a nearly releasable state -- that is, all it will need is checks to make sure the WML is not buggy and balance testing. Of course we can't really do the latter wile the AI is broken... 20081030 23:05:10 * loonycyborg thinks that smart pointers make garbage collectors redundant 20081030 23:05:18 ...but I want to be able to tell Ivanovic it will freeze for translation before mainline does, so he won't scream too loudly when I merge it. 20081030 23:05:45 loonycyborg: I disagree... garbage collection is great for freeing other peoples memory bugs ;) 20081030 23:05:48 * esr thinks loonycyborg is horribly, *horribly* wrong about that. 20081030 23:05:56 (i added in syntax error's DiD story images and yes, i know someone might want to have those cropped/scaled differently...i'll look into it myself, please don't poke around them yet) 20081030 23:06:30 -!- zookeeper [n=l@dsl-tkubrasgw1-fe11fa00-51.dhcp.inet.fi] has quit [] 20081030 23:06:32 zookeeper * r30438 /trunk/data/campaigns/Descent_Into_Darkness/ (16 files in 3 dirs): Added and wired in the new DiD story images, and also added in theycallmerooster in the campaign art credits since he's been uncredited for his portraits so far. 20081030 23:06:32 esr: Those asterisks are *very* convincing ;) 20081030 23:07:20 smart pointers can help indeed in a lot of cases 20081030 23:07:20 zookeeper * r30439 /trunk/ (changelog players_changelog): Updated changelogs. 20081030 23:07:22 loonycyborg: for example, let's say there was a horrible memory leak in the freetype library. If we had garbage collection, it could save us 20081030 23:07:32 yes, although it may play, WML is sure to be buggy in DM; but translations could proceed. Of course there will be the extra scenario with Delf getting the book back and one may also revise the Iliah-Malal vs Delf war of words, but this will be relatively minor if the rest is translated ok 20081030 23:07:42 What are we talking about and whatnot? 20081030 23:08:02 * esr thinks sSapient is zeroing in on the referent of "horribly" rather nicely. 20081030 23:08:17 and GC also isn't without its own problems ;) 20081030 23:08:31 * Sapient is referring to an actual leak which occurred from freetype ;) 20081030 23:09:04 kalenzz: Have you seen the latest version of the "war of wards"? I'm rather proud of it. Where Delf gets his title "the Great". 20081030 23:09:17 Sapient: Isn't freetype a C library? 20081030 23:10:03 loonycyborg: C doesn't have GC and neither does C++. we're speaking of a hypothetical situation 20081030 23:10:48 Of course, smart pointers won't help freetype which isn't available in C. 20081030 23:11:43 nonetheless, if C++ had GC, you could compile C code with GC since C++ can compile C code 20081030 23:12:10 so your point is really irrelevant 20081030 23:12:31 Sapient: btw C GCs aren't hypothetical: http://en.wikipedia.org/wiki/Boehm_garbage_collector 20081030 23:13:04 right, right, I've also heard of C++ GC's 20081030 23:13:25 they're not standard with the language though 20081030 23:13:45 loonycyborg: They're not hypothetical except for the inconvenient fact that they *suck*. There are serious problems with Boehm GC in the presence of pointers. They're foundational, nit fixable with trickery or cleverness. 20081030 23:14:18 * esr is occasionally a language designer 20081030 23:14:28 -!- kalenzz [n=me@athedsl-4499278.home.otenet.gr] has quit [Remote closed the connection] 20081030 23:14:35 esr: Yes. That's why you'd better use smart pointers in your programs. 20081030 23:15:16 loonycyborg: Smart pointers are a band-aid on a slashed artery. Utterly inadequaye. 20081030 23:15:35 How they are inadequate? 20081030 23:15:49 Bjorne Strousup also said that he would like to see GC available with C++ if there can be a clean way of integrating it without excluding the possibility of manual memory management in tandem 20081030 23:16:33 Sapient: I'd say that shared_ptr is a small GC :) 20081030 23:16:47 I think you just don't like to admit defeat 20081030 23:17:08 shared_ptr is not going to save you for everything 20081030 23:17:15 for example, linked libraries 20081030 23:17:21 Neither will GC. 20081030 23:17:27 no but it can help in a lot of cases 20081030 23:17:56 GC is part of an overall solution I didn't say it would save you for everything 20081030 23:18:11 you are the one claiming shared_ptr solves it all ;) 20081030 23:18:25 No. I don't. 20081030 23:18:51 shared_ptr does not cover everything that GC covers 20081030 23:20:23 For now I don't know a situation where I'd prefer GC to smart pointer. That's why I said that I *think* that smart pointers make GC redundant. 20081030 23:21:50 smart pointers require smart people using them; GC is automatic and leaves a very small footprint in lines of code that you must write 20081030 23:23:05 not-smart people coding is the problem, GC or smart pointers have nothing to do with it 20081030 23:23:37 and when you're dealing with such people, you need languages where the compiler can check as much as possible for you 20081030 23:24:06 * loonycyborg prefers not to deal with such people :) 20081030 23:24:25 loonycyborg: you can choose your co-FOSS-developers 20081030 23:24:30 but not your colleagues 20081030 23:24:36 boucman: I agree on that part about the compiler 20081030 23:25:04 * boucman is coding some Ada and is amazed by what the compiler can do to help when the language provides the info 20081030 23:25:28 but I think GC is a valid real world strategy for dealing with co-workers or partners who don't understand or aren't careful with memory managment 20081030 23:25:31 and how little you need pointers in languages with variable sized arguments to function calls (and return values) 20081030 23:30:01 and... let's face it, even the old pros make the simple mistakes once in a while ;) if your product is already shipped out the door you may not have time to patch that memory leak in your own code 20081030 23:30:30 if there was GC for it you could lose less sleep until the next patch cycle 20081030 23:31:08 Memory leaks are a tiny fraction of mistakes you can make :) 20081030 23:31:23 GC won't save you much sleep. 20081030 23:33:01 zzz cya 20081030 23:33:07 -!- Sapient [n=sapien-x@c-71-59-35-203.hsd1.ga.comcast.net] has quit [] 20081030 23:33:31 Besides, I don't see how you can get memory leaks if you always assign pointers to smart_ptrs in situ nascendi. 20081030 23:33:59 circular references 20081030 23:34:48 Indeed. But it affects GCs too. 20081030 23:35:14 I know ;) 20081030 23:41:46 boucman loonycyborg found a problem in the cache item class, copying singular iterators is UB 20081030 23:42:11 so I'm wondering is it only allowed to use the position if loaded == true? 20081030 23:43:07 I think that's the case 20081030 23:43:29 I'm going to bed right now, so I can't check, but I seem to remember that I made sure that it was the case 20081030 23:44:58 oke then I can change some things in the code to avoid the problem 20081030 23:46:00 loonycyborg did your workaround work properly? That is can you also actually play without problems? 20081030 23:46:59 You mean the hack with dummy list? I didn't test much with it, though I don't see why would it break anything. 20081030 23:47:03 -!- Rh0nda changed the topic of #wesnoth-dev to: 49 bugs, 207 feature requests, 9 patches | logs: http://irclog.wesnoth.org and http://wesnoth.org/irclogs | Don't paste on IRC! Use a pastebin: http://rafb.net/paste, http://paste.debian.net, http://pastebin.ca 20081030 23:47:46 yes I mean that hack, but if you didn't test much I know I need to test before committing ;) 20081030 23:47:57 * mordante hugs Rh0nda --- Log closed Fri Oct 31 00:00:05 2008