--- Log opened Thu Dec 25 00:00:36 2008
20081225 00:03:32-!- loonycyborg [n=sergey@79.139.247.111] has quit ["Zzzzzzzzzzzzzzzzzzzz"]
20081225 00:19:49< Ivanovic> time for me to head off to bed, n8
20081225 00:19:56< mordante> night Ivanovic 
20081225 00:32:52< mordante> night
20081225 00:33:50< CIA-54> mordante * r31711 /trunk/src/gui/widgets/window.cpp: 
20081225 00:33:50< CIA-54> Properly redraw dialogs.
20081225 00:33:50< CIA-54> For semi transparent dialogs it's important the background is also
20081225 00:33:50< CIA-54> restored before redrawing that part of the window. Otherwise the
20081225 00:33:52< CIA-54> semitransparent area will add to itself and darken the area.
20081225 00:34:05-!- mordante [n=chatzill@roadie.xs4all.nl] has quit ["ChatZilla 0.9.84 [Firefox 3.0.4/2008102920]"]
20081225 00:49:35-!- Appleman1234 [n=Appleman@CPE-124-177-116-97.qld.bigpond.net.au] has joined #wesnoth-dev
20081225 01:17:10-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Read error: 104 (Connection reset by peer)]
20081225 01:23:02-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20081225 01:32:58< CIA-54> ai0867 * r31712 /trunk/src/ (6 files): Rewrite some of my earlier changes to be less intrusive, fixing some side-effect bugs.
20081225 01:40:17-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit ["leaving"]
20081225 01:52:36-!- DDR [n=chatzill@66.183.125.196] has quit [Remote closed the connection]
20081225 01:53:39-!- DDR [n=chatzill@66.183.125.196] has joined #wesnoth-dev
20081225 01:59:56< AI0867> zookeeper: the turn event bug should be fixed now
20081225 02:16:20< zookeeper> AI0867, nice
20081225 02:17:38< AI0867> result of me overzealously rewriting the play_turn() code
20081225 02:20:02-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081225 02:20:29 * zookeeper completely misses the point of http://www.wesnoth.org/forum/viewtopic.php?p=327771#p327771
20081225 02:21:28< AI0867> I'll check in a few minutes
20081225 02:21:32< AI0867> when I'm done recompiling
20081225 02:27:34-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has quit []
20081225 02:43:00-!- crimson_penguin [n=ben@64.201.60.217] has quit ["Leaving"]
20081225 02:44:49< Soliton> can i include a C++ file in C project? i.e. there is a project in C that i want to extend by calling some C++ functions...
20081225 02:45:54< Soliton> well, more like does it make sense?
20081225 02:46:51-!- Blueblaze [n=Blueblaz@204.houston-04rh15-16rt.tx.dial-access.att.net] has quit [Nick collision from services.]
20081225 02:46:56-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 02:56:09< AI0867> Soliton: gcc (or g++) can link the two, but why would you want to?
20081225 02:56:39< Soliton> i'd rather write in c++.
20081225 02:56:41-!- cib0 [n=cib@p4FD0B8BB.dip.t-dialin.net] has joined #wesnoth-dev
20081225 02:56:59< cib0> wesnoth code can be so messy =/
20081225 02:57:31< Soliton> the c code is 20 years old and scary.
20081225 03:01:28-!- Appleman1234 [n=Appleman@CPE-124-177-116-97.qld.bigpond.net.au] has quit ["Leaving"]
20081225 03:06:16< AI0867> well, I'm not sure which IDE you're using (I guess you are using one since you used the word 'project'), but gcc can handle such things
20081225 03:08:20 * Soliton tried to use eclipse today and failed.
20081225 03:08:28 * Soliton went back to vim.
20081225 03:09:28< AI0867> so, makefiles and possibly a configure script?
20081225 03:09:42< Soliton> makefiles, yes.
20081225 03:10:05< Soliton> indeed the next question is how to integrate it there.
20081225 03:10:45< AI0867> well, change CFLAGS to CCFLAGS so they apply to .cpp files too
20081225 03:11:25< Soliton> looks like CPPFLAGS is already used.
20081225 03:11:26< AI0867> besides that, gcc should handle compiling everything into object files, so all you should need to do is get it linked in
20081225 03:11:32< AI0867> that helps
20081225 03:14:33-!- 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!"]
20081225 03:14:39< AI0867> someone should remove the accuracy era from wherever units.wesnoth.org generates its stats from
20081225 03:15:05< Soliton> you checked that is the culprit?
20081225 03:15:13< AI0867> it's incompatible with current svn and it seems to be interfering with wmlunits
20081225 03:15:21< AI0867> haven't checked, but the stats coincide
20081225 03:16:17< Soliton> yeah, i guess that's good enough.
20081225 03:16:23< AI0867> testing now
20081225 03:17:01< AI0867> okay, I may have encounted a bug in the addons.wesnoth.org tarball creating
20081225 03:17:21< AI0867> it doesn't include addon_name.cfg files
20081225 03:18:21< Soliton> oh.
20081225 03:19:27< AI0867> I just downloaded the AE tarball, and it included neither a _main.cfg nor an Accuracy_Era.cfg
20081225 03:19:30-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081225 03:20:08< Soliton> it should have the cfg.
20081225 03:21:04< AI0867> redownloading...
20081225 03:21:12< AI0867> nope, not in there
20081225 03:21:56< Soliton> yeah, i didn't mean you were wrong. :-)
20081225 03:22:27< AI0867> is the script generating it in trunk?
20081225 03:23:50< Soliton> don't know what you mean but data/tools/wesnoth_addon_manager:347 is the line to modify.
20081225 03:28:33< AI0867> I'm fairly sure 139 is the culprit ;)
20081225 03:30:15< Soliton> uh yeah, miscopy.
20081225 03:30:25< Soliton> would be nice if you can fix it.
20081225 03:32:25< CIA-54> espreon * r31713 /trunk/ (changelog players_changelog): * Updated the changelogs to mention the changes to Familiar's portrait.
20081225 03:36:31< AI0867> I have something that should work, but I'll need a few minutes to test it
20081225 03:40:53-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081225 03:40:53-!- crimson_penguin [n=ben@64.201.60.217] has quit [Client Quit]
20081225 03:41:03-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081225 03:41:04< AI0867> the --help wasn't very useful...
20081225 03:49:54< CIA-54> ai0867 * r31714 /trunk/data/tools/wesnoth_addon_manager: Include oldstyle add-on_name.cfg files in the tarballs.
20081225 03:54:49< Soliton> thanks.
20081225 03:56:11< AI0867> we should really get rid of those in 1.7
20081225 03:58:43< voris> The add-on_name.cfg files?
20081225 03:59:37< AI0867> the stuff that was replace with add-on_name/_main.cfg
20081225 03:59:55< voris> Right.
20081225 04:00:18< voris> I was verifying that is what you were saying should be gotten rid of.
20081225 04:00:19< voris> :)
20081225 04:01:54< AI0867> Soliton: confirmed that the accuracy era triggers the bug
20081225 04:02:13< Soliton> good, i already removed it.
20081225 04:03:31< AI0867> rerunning wmlunits now?
20081225 04:04:28< Soliton> too lazy.
20081225 04:04:52< AI0867> feh, I ran it thrice here...
20081225 04:15:50< ShikadiLord> merry christmas to all!! :)
20081225 04:15:56< ShikadiLord>  I've been given a new laptop.
20081225 04:16:04< AI0867> w00tness
20081225 04:16:09< Espreon> Yay!
20081225 04:16:52< ShikadiLord> it is an AMD Turion with double core :) 
20081225 04:17:02< Espreon> Sounds great.
20081225 04:17:31< ShikadiLord> so I should probably be able to wipe out Vista tomorrow and install Debian lenny there. with an adhoc wifi connection, transferring home fs should be rather quick and clean
20081225 04:17:37< Espreon> It is still Christmas Eve here.
20081225 04:17:42< CIA-54> cib * r31715 /trunk/src/ (4 files): bug #12780: fixed a bug where a function to get a unit type was checked against returning the end of the vector, when actually it would return a dummy unit, by replacing it with a new function
20081225 04:17:47< ShikadiLord> Espreon:  :) 
20081225 04:18:00< Espreon> But... I got some mochi ice cream!
20081225 04:18:15< ShikadiLord> Espreon: I actually was given this present at GMT-04:00 16:00 or so, so... earliy present
20081225 04:18:23< Espreon> Ah.
20081225 04:18:31< Espreon> Happy Vista slaying!
20081225 04:18:34< ShikadiLord> it's 17 past midnight now
20081225 04:18:43< ShikadiLord> Espreon: oh yeah! ^_^
20081225 04:20:14< ShikadiLord>  finally I'll be able to use a safer partitioning scheme, with /var and /usr not in root fs
20081225 04:22:56-!- crimson_penguin [n=ben@64.201.60.217] has quit ["Leaving"]
20081225 04:23:25< ShikadiLord> so, yeah, I hope my mother likes Wesnoth's soundtrack :P
20081225 04:23:36< Espreon> I hope she likes it too.
20081225 04:23:56< ShikadiLord> I already shown her Kitty's porrtraits and she says they are great
20081225 04:24:19< Espreon> :)
20081225 04:24:46< ShikadiLord> in particular the Elvish Shyde and Elynia's portraits, which we used to make a xmas card for the nighbors' children
20081225 04:25:11< Espreon> How adorable.
20081225 04:25:43< ShikadiLord> I see that mordante left a gift for us :)
20081225 04:26:05< Espreon> Really?
20081225 04:26:24< ShikadiLord> well, r314710
20081225 04:26:34< ShikadiLord> scrap the 4.
20081225 04:27:12-!- Ivanovic_ [n=ivanovic@dtmd-4db261d6.pool.einsundeins.de] has joined #wesnoth-dev
20081225 04:31:53< ShikadiLord> Espreon: you speak english natively. can you reword http://www.wesnoth.org/forum/viewtopic.php?p=327713#p327713 for me?
20081225 04:32:09< ShikadiLord> I don't understand a word of what they say.
20081225 04:33:14< Espreon> I think he means that Wesnoth is complaining that something is not where it is supposed to be.
20081225 04:33:19< Espreon> IDK.
20081225 04:33:34< ShikadiLord> sounds like a Windows-build screw-up
20081225 04:34:22< ShikadiLord> "pathing error"... not even M$ uses such obscure terms for their error messag- massages.
20081225 04:39:54< Espreon> ShikadiLord: Say, "Hable inglés por favor." and see how he reacts. ;)
20081225 04:41:17< ShikadiLord> no, better:
20081225 04:41:28< ShikadiLord> "Wesnoth.exe? is that a virus impersonating Wesnoth?"
20081225 04:41:36< Espreon> Heh...
20081225 04:42:22-!- Ivanovic [n=ivanovic@dtmd-4db2ad76.pool.einsundeins.de] has quit [Read error: 110 (Connection timed out)]
20081225 04:42:59< voris> "cost and path errors"?
20081225 04:43:10-!- Ivanovic_ is now known as Ivanovic
20081225 04:43:43< ShikadiLord> no, that's a too generic response
20081225 04:44:11< ShikadiLord> funny, I had to act as Santa Claus for the neighbors
20081225 04:44:29< ShikadiLord> risking myself to the wrath of their enormous dog 
20081225 04:46:23< Espreon> Really? How big was this enormous dog of which you speak?
20081225 04:46:46< ShikadiLord> the spanish name for the race is "San Bernardo" let me look it up...
20081225 04:48:06< Espreon> Saint Bernard?
20081225 04:48:17< ShikadiLord> probably
20081225 04:48:20< Espreon> Oh... those things are scary.
20081225 04:48:48< ShikadiLord> I always thought they were in gneneral nice dogs, since they were originally trained to rescue people and such.
20081225 04:48:59< Espreon> I think trained is the key word.
20081225 04:49:31< ShikadiLord> but I now think they were trained to harm people to give professional rescuers a excuse for their jobs :)
20081225 04:49:56< ShikadiLord> http://www.wesnoth.org/forum/viewtopic.php?f=23&t=23354 <- ah, I have been reminded of why I no longer pay attention to new art threads.
20081225 04:50:46< Espreon> Ugh...
20081225 04:50:55< Espreon> Do not remind me.
20081225 04:52:50< ShikadiLord> that animation  .... "I would remove the blood"
20081225 04:53:01 * ShikadiLord slaps Espreon with a fish
20081225 04:53:26< ShikadiLord> that thing is so cheap that that blood looks like... microscopic oconfetti.
20081225 04:53:38< Espreon> Fine... I'll change it to "blood."
20081225 04:54:55< ShikadiLord> I mean, um. I wonder if these people really have any intentions of learning, or they just throw random half-assed, pathetic attempts at spriting to say "I made a sprite, I am an artist".
20081225 04:55:26< ShikadiLord> 96% of these people never attempt at learning and improving.
20081225 04:55:40< Espreon> They are technically artists, just not good artists in that field of art.
20081225 04:56:11< ShikadiLord> yeah, that's the point. But they seem to  believe they r are "cool" for doing that.
20081225 04:56:41< ShikadiLord> you see how I started. It was pathetic as hell, but I invested time and efforts into learning and improving, and still do.
20081225 04:56:52< Espreon> I really wish that the guy would have at least made a proper image. Meaning that it still would not have the white stuff.
20081225 04:58:19< ShikadiLord> Esp	the erson obviously uses MSPaint.
20081225 04:58:36< Espreon> Ewww...
20081225 04:58:42< Espreon> M$ Paint...
20081225 04:58:45< ShikadiLord> and some freeware animator rogram
20081225 04:58:50< Espreon> :(
20081225 05:00:03< ShikadiLord> I see that the SSH lag made me spit invalid UTF-8 sequences
20081225 05:00:54-!- ShikadiLord [n=shadowm@unaffiliated/shadowmaster/x-275924] has quit ["Battle for Wesnoth [www.wesnoth.org]"]
20081225 05:01:09-!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081225 05:14:05< Shadow_Master> Sirp: do you know anything about Francisco Muñoz nowadays? (our initial artist)
20081225 05:14:16< Shadow_Master> I wonder if I could kidnap him for my translation team :P
20081225 05:14:25 * Espreon grins.
20081225 05:14:30< Shadow_Master> Or isaac ... :P 
20081225 05:14:39< Espreon> I'll get the flail.
20081225 05:15:46< Shadow_Master> those are the only Spanish-speaking developers that there are/were AFAIK.
20081225 05:40:46< Shadow_Master> lol @ topic
20081225 05:41:27< Shadow_Master> Ivanovic: lemme fix it ;)
20081225 05:41:31< Blueblaze> lol
20081225 05:42:23< Smar> it’s thursday, where’s my 1.5.7...
20081225 05:43:05< Shadow_Master> when Ivanovic wakes up 
20081225 05:43:09< Shadow_Master> :->
20081225 05:43:09< Smar> (oh, someone dard the fix the topic :-)
20081225 05:43:30< Smar> well, then it should be here when I wake up...
20081225 05:43:47< Smar> I might even actually install it... 
20081225 05:44:08< Shadow_Master> Good that we know Mordante doesn't use a  Windowss + kvirc 3.4 ;)  (long story)
20081225 05:44:35< Smar> :)
20081225 05:44:56< Smar> I didn’t really care about the details so I didn’t read them... oh well, I’ll go away again
20081225 06:06:05-!- cib01 [n=cib@p4FD0E4BF.dip.t-dialin.net] has joined #wesnoth-dev
20081225 06:06:53-!- cib0 [n=cib@p4FD0B8BB.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)]
20081225 06:26:06-!- cib01 is now known as cib0
20081225 06:38:58< Shadow_Master> hungry
20081225 06:39:07< Espreon> Grab some mochi ice cream.
20081225 06:39:32< Shadow_Master> it past 2:30 AM and I can't go to the fridge without awaking the second floor's troll
20081225 06:39:52< Shadow_Master> there are no snacks there either
20081225 06:40:00< DDR> A saw is your friend. Get into the fridge from the bottom!
20081225 06:41:05< Shadow_Master> the fridge isn't upstairs, but I still have to make some noise level to open it
20081225 06:42:45< Espreon> Who/what is this troll of which you speak?
20081225 06:43:29< Shadow_Master> my father. "He's a notorious troll!" :) I 'm not saying that he isn't a bad person, but he's just as annoying as a troll
20081225 06:44:10< Espreon> A Troll Warrior, a Great Troll, or a Troll Shaman?
20081225 06:44:12< Shadow_Master> specially t those ninja rocklobbers 
20081225 06:44:39< Shadow_Master> no, a ninja rocklobber. The variety that can parry all ranged attack hits from an Anceient Lich.
20081225 06:45:07< Shadow_Master> those are a rare breed.
20081225 07:11:51-!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has quit ["Battle for Wesnoth [www.wesnoth.org]"]
20081225 07:14:41-!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081225 07:18:40< Espreon> ESR__: I hate it when wmllint complains about potential misspellings with words that are correctly spelled that are next to single quotes and other punctuation marks.
20081225 08:20:35-!- mordante [n=chatzill@roadie.xs4all.nl] has joined #wesnoth-dev
20081225 08:21:12< mordante> morning
20081225 08:22:45< Espreon> Good morning mordante.
20081225 08:22:45< cib0> hoy mordante
20081225 08:23:00< mordante> hi Espreon, cib0 
20081225 08:23:30< Espreon> Are you having a very merry Christmas/other holiday you might be celebrating?
20081225 08:24:36< mordante> I'm not celebrating Christmas, it just happens to be Christmas ;-)
20081225 08:25:17< Espreon> You can't be sure these days.
20081225 08:25:51< mordante> well a lot of people in the Netherlands do celebrate Christmas
20081225 08:26:10< Blueblaze> Espreon, Merry Christmas 
20081225 08:26:36< Espreon> Blueblaze: And a very mery Christmas to you.
20081225 08:29:09 * mordante kickstarts his compiler and celebrates coding day :-)
20081225 08:29:55< Espreon> Imma restarting.
20081225 08:30:00-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20081225 08:37:15-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081225 09:21:12-!- loonybot [n=loonybot@79.139.247.111] has joined #wesnoth-dev
20081225 09:22:46-!- loonycyborg [n=sergey@79.139.247.111] has joined #wesnoth-dev
20081225 09:25:24-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has quit [Remote closed the connection]
20081225 10:16:18< CIA-54> mordante * r31716 /trunk/ (changelog src/text.cpp): 
20081225 10:16:18< CIA-54> Improved the drawing routine for the ttext class.
20081225 10:16:18< CIA-54> Loonycyborg pointed me at this side
20081225 10:16:18< CIA-54> http://cairographics.org/manual/cairo-image-surface.html#cairo-format-t
20081225 10:16:18< CIA-54> which explains why the text rendered looks ugly after one render cycle.
20081225 10:16:20< CIA-54> Used this information to improve the drawing, unfortunately it doesn't seem to
20081225 10:16:22< CIA-54> help on Windows.
20081225 10:16:24< mordante> loonycyborg maybe you want to play a bit with it on Windows ^
20081225 10:17:36< loonycyborg> Once Ivanovic releases I'll *have* to go to windows in order to make a release :)
20081225 10:18:02< loonycyborg> *windows release
20081225 10:18:14< mordante> sounds like you love Windows :-P
20081225 10:18:50< Espreon> I thought you could use NSIS on Linux.
20081225 10:18:57< Espreon> Compile the script and everything.
20081225 10:19:49< loonycyborg> But I need to test it on native windows, where most users will run it.
20081225 10:20:03< Espreon> Yeah, that is true.
20081225 10:55:09-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has quit []
20081225 10:57:00< Ivanovic> moin
20081225 10:57:12< Sirp> hi Ivanovic 
20081225 10:57:15< Espreon> Good morning Ivanovic.
20081225 10:57:18< Ivanovic> merry christmas
20081225 10:57:20< mordante> hi Ivanovic 
20081225 10:57:25< Espreon> Merry Christmas.
20081225 10:57:27< Sirp> merry Christmas.
20081225 10:58:18< Ivanovic> mordante: so how far are your changes?
20081225 11:04:00-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081225 11:04:25-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has left #wesnoth-dev []
20081225 11:04:48-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081225 11:08:34< mordante> Ivanovic I still need to get the MP server list working and have a few commits pending
20081225 11:08:54< mordante> once that's done I'd like some to see some testing
20081225 11:09:21< mordante> (I still have a few minor glitches left, but at least the new drawing code is done)
20081225 11:09:32< mordante> the other fixes are not that invasive
20081225 11:10:28< mordante> once the next few commits are in all features are done and I can start to polish and fix the bugs left
20081225 11:12:10-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20081225 11:13:14-!- DDR [n=chatzill@66.183.125.196] has quit [Read error: 110 (Connection timed out)]
20081225 11:13:42< Ivanovic> mordante: okay
20081225 11:18:56 * zookeeper wonders whether he should now wire in the ai controller everywhere or leave that to the next release
20081225 11:19:14< Espreon> Good night people.
20081225 11:19:21-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20081225 11:19:23< Ivanovic> zookeeper: your choice
20081225 11:19:42< Ivanovic> though i would probably wait for the next one before all the campaigns suddenly break due to this
20081225 11:20:55-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has quit []
20081225 11:34:14< mordante> Ivanovic if you need inspiration ;-) http://lkml.org/lkml/2008/12/24/105
20081225 11:34:38< mordante> if you announce tomorrow, santa can visit again ;-)
20081225 11:40:26< CIA-54> mordante * r31717 /trunk/src/gui/widgets/scroll_label.cpp: 
20081225 11:40:26< CIA-54> Enable the wrapping of the scroll label.
20081225 11:40:26< CIA-54> This is needed to make the scroll label wrap on smaller screens, this
20081225 11:40:26< CIA-54> works better as a horizontal scrollbar. (Also the horizontal scrollbar
20081225 11:40:26< CIA-54> didn't work as expected.)
20081225 11:40:32< CIA-54> mordante * r31718 /trunk/src/gui/widgets/scrollbar_container.cpp: 
20081225 11:40:32< CIA-54> Improve the rendering.
20081225 11:40:32< CIA-54> Make sure the foreground also is drawn inside the container and doesn't
20081225 11:40:34< CIA-54> leak out of it.
20081225 11:40:38< CIA-54> mordante * r31719 /trunk/src/gui/widgets/toggle_panel.hpp: 
20081225 11:40:40< CIA-54> Make the toggle panel draw properly.
20081225 11:40:42< CIA-54> A toggle panel has no fore and background and need to draw depending on
20081225 11:40:44< CIA-54> its state like a control, so use that drawing method.
20081225 11:41:49< CIA-54> mordante * r31720 /trunk/src/text.cpp: Remove some commented out code.
20081225 11:48:23< zookeeper> Ivanovic, meh, it can't really break anything (at least if the player doesn't use it ;) but i guess i'll leave it for the next release anyway
20081225 11:49:12< Ivanovic> hey, it is your choice
20081225 11:49:28< Ivanovic> though those changes might be the reason for a "new years release"
20081225 11:49:29< Ivanovic> ;)
20081225 11:52:01< zookeeper> i'll add in a inactive tip of the day for that though
20081225 11:57:14< CIA-54> mordante * r31721 /trunk/ (19 files in 3 dirs): 
20081225 11:57:14< CIA-54> Enable the new drawing engine by default.
20081225 11:57:14< CIA-54> The new engine is a complete rewrite of the old engine and seems to work
20081225 11:57:14< CIA-54> properly now. So the engine is enabled now. After more testing the #ifdef hell
20081225 11:57:14< CIA-54> will be cleaned up.
20081225 11:57:51< Ivanovic> good idea
20081225 11:59:02< CIA-54> zookeeper * r31722 /trunk/data/hardwired/tips.cfg: Added a commented-out tip of the day about the ai controller to get the string in.
20081225 11:59:15< Ivanovic> but commented out strings are not extracted into po files
20081225 11:59:30< Ivanovic> you got to add it in some other file like eg. core/english.cfg
20081225 11:59:31< zookeeper> using preprocessor #ifdef to comment it out
20081225 11:59:36< Ivanovic> ahhh, okay
20081225 11:59:38< Ivanovic> that works
20081225 12:00:14< zookeeper> i'm hoping that the thing is easy enough to use and find that it doesn't require any extra help text elsewhere
20081225 12:16:25< CIA-54> mordante * r31723 /trunk/ (4 files in 2 dirs): Enable the new dialogs and update changelogs.
20081225 12:30:18-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081225 12:54:25< fendrin_> zookeeper: your new ai_controller breaks low. The campaign is no longer startable. Please check.
20081225 12:54:48< zookeeper> argh, sorry
20081225 12:54:56< zookeeper> yeah, i'll fix it
20081225 12:55:18< fendrin_> Should be a simple naming conflict.Wait.
20081225 12:55:35< zookeeper> yeah, it is.
20081225 12:55:45< fendrin_> Don't rename the macro in core. I will rename the one in LoW.
20081225 12:56:21< fendrin_> This will enable your ai controller for elves last stand.
20081225 12:57:04< zookeeper> hmm, ok.
20081225 12:57:17< zookeeper> (it'll still need some fixes, but i'm gonna commit some soon)
20081225 13:06:23< CIA-54> mordante * r31724 /trunk/src/gui/dialogs/editor_settings.cpp: 
20081225 13:06:23< CIA-54> Force a redraw of the window.
20081225 13:06:23< CIA-54> The new drawing engine doesn't redraw the dialog properly (since it's
20081225 13:06:23< CIA-54> unaware of the dirtying) so set the dirty flag manual.
20081225 13:07:29< mordante> Ivanovic it looks like the dialogs are in working state now, still need some polishing
20081225 13:07:49< mordante> it would be nice to get some test coverage, to see whether there are bugs left
20081225 13:08:02< mordante> I'm aware of the I see too often a scrollbar problem
20081225 13:08:22< mordante> also the language dialog has a bug where items out of the listbox are drawn
20081225 13:08:36< mordante> off to lunch now
20081225 13:11:28< fendrin_> zookeeper: what would be a valid call to AI_CONTROLLER?
20081225 13:12:25< fendrin_> {AI_CONTROLLER ID-the id of what? 
20081225 13:27:47-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has joined #wesnoth-dev
20081225 13:39:25< zookeeper> fendrin_, if of that particular controller, if you're going to have several (with differently customized values for different ally sides for instance)
20081225 13:39:41< zookeeper> if you have just one, you can leave it blank ()
20081225 13:40:20< zookeeper> s/if/id
20081225 13:40:25< fendrin_> zookeeper: I should give it a default ai in a variable?
20081225 13:41:28< zookeeper> fendrin_, it has crude defaults so you don't need to, but you can and i suppose in many situations it'd be a good thing to do
20081225 13:42:23< zookeeper> i'm currently trying to fix one annoying bug preventing me from committing a slightly improved version, when i'm done with that i'll write some proper examples of how to use it
20081225 13:42:37< fendrin_> cool
20081225 13:42:52< fendrin_> I am going to commit to make LoW working again.
20081225 13:45:13< zookeeper> ok, great. i'll go afk for a bit and then i'll try to do all that..
20081225 13:45:27-!- loonybot [n=loonybot@79.139.247.111] has quit [Read error: 104 (Connection reset by peer)]
20081225 13:55:23< CIA-54> fendrin * r31725 /trunk/data/campaigns/Legend_of_Wesmere/ (3 files in 2 dirs): Replaced the old ai_controller with zookeeper's.
20081225 14:07:48-!- Sirp [n=me@222-154-40-244.jetstream.xtra.co.nz] has quit [Read error: 110 (Connection timed out)]
20081225 14:37:47< fendrin_> zookeeper: ping
20081225 14:38:44< Ivanovic> woho
20081225 14:38:49< Ivanovic> mordante: your changes seem great
20081225 14:38:57< Ivanovic> i get lovely game quits in two brothers
20081225 14:39:05< Ivanovic> "Abgebrochen"
20081225 14:39:38< Ivanovic> that is: once the dialog from bjarn at the start hangs
20081225 14:39:40< mordante> Ivanovic where, I used TB to test
20081225 14:39:43< Ivanovic> when getting out of it, crash
20081225 14:40:17< Ivanovic> later on when defeating the mage (debug cheat, killing him via an armageddon drake) and after he said some lines: crash
20081225 14:40:30< zookeeper> fendrin_, pong
20081225 14:40:43< Ivanovic> so yeah, it is not stable at all ATM
20081225 14:41:57< fendrin_> zookeeper: After a rebuild all story only scenarios in LoW show a "Game Over" popup and ask the player if he wants to continue the campaign. But Continuing the campaign is a horrible mess.
20081225 14:42:28< mordante> Ivanovic odd, I tested that part and it works for me (tm)
20081225 14:42:48 * mordante restarts wesnoth
20081225 14:42:53< zookeeper> fendrin_, err, what? i've never even heard of such a popup existing.
20081225 14:42:56< cib0> o.o
20081225 14:43:08< cib0> zookeeper, its for MP campaigns
20081225 14:43:16< zookeeper> oh, right.
20081225 14:43:23< cib0> observers and people who lost should get that message
20081225 14:43:32< mordante> Ivanovic can you try to reproduce it with debug info and in gdb so you can post a backtrace?
20081225 14:43:39< zookeeper> fendrin_, i have no idea of that. probably someone broke something recently :)
20081225 14:43:50< zookeeper> sounds like the perfect time for a release ;)
20081225 14:44:07< fendrin_> ohje
20081225 14:44:42< cib0> meh, ill look into it, maybe it has something to do with my endlevel extensions? fendrin_, when did it work for you properly the last time?
20081225 14:45:51< Ivanovic> mordante: got to rebuild for this first
20081225 14:46:05< fendrin_> cib0: Sorry, I can't tell you excactly. The last scons install is several days old.
20081225 14:46:58< zookeeper> waah. i can't seem to rid myself of this error: The maximum text width is less than 1. When reporting the bug please include the following error message : Condition 'max_width > 0' failed at src/marked-up_text.cpp:330 in function 'word_wrap_text'.
20081225 14:47:00< fendrin_> cib0: It's easy to reproduce. Start LoW and switch with cl to scenario 8.
20081225 14:47:19< zookeeper> mordante, what was the exact reason for that error again?
20081225 14:47:48< cib0> huh? error loading game?
20081225 14:48:53< mordante> Ivanovic ok please do, just did TB again and no crash here
20081225 14:49:57< Ivanovic> wow, make -j was a *bad* option for building wesnoth
20081225 14:50:03< Ivanovic> (at least without setting a niceness=
20081225 14:50:27< mordante> zookeeper not sure, but is it with the tips of the day?
20081225 14:50:39< zookeeper> mordante, no, with [message][option]s
20081225 14:51:09< mordante> do you have an empty text?
20081225 14:51:48< zookeeper> not that i can see...using that "&$side_i_leader.image~TC($temp_i|,$side_i_leader.flag_rgb|)=$side_i_leader.name|=(side $temp_i|)"+"= " syntax
20081225 14:51:59< mordante> make -j seems a bad option in general ;-)
20081225 14:52:12< zookeeper> maybe it's that space-only last column, although they usually work just fine, i'll test..
20081225 14:54:31< zookeeper> yeah, that did it. hrhm.
20081225 14:55:01< cib0> fendrin_: ugh, yeah, you're right..
20081225 14:56:20< fendrin_> cib0: After continuing the campaign it's no longer in a usefull state. 
20081225 14:57:31< Ivanovic> mordante: uhm, how to start a binary in gdb with parameters?
20081225 14:57:36< Ivanovic> that is i just got a cmake build
20081225 14:57:58< Ivanovic> and for this one i got to specify the dir to use as data dir
20081225 14:58:34< cib0> is endlevel ever called in the WML of that scenario?
20081225 14:59:09< mordante> r parameters 
20081225 14:59:26< mordante> the second time in that session you can just do r
20081225 14:59:31< cib0> hm, yep, it is..
20081225 14:59:33-!- EdB [n=EdB@115.117.88-79.rev.gaoland.net] has joined #wesnoth-dev
20081225 14:59:44< loonycyborg> Ivanovic: gdb --args
20081225 14:59:52< fendrin_> cib0: [event] name=prestart [endlevel] result=continue_no_save 
20081225 15:00:05 * mordante prefers the r option
20081225 15:02:58< loonycyborg> Ivanovic: Or better just use scons. It creates wesnoth in a place where it doesn't require parameters to run :)
20081225 15:04:16< Ivanovic> *lovely*
20081225 15:04:35< Ivanovic> now it does not even want to start since it thinks it can't read data/hardwired/fonts.cfg
20081225 15:04:53< fendrin_> Hell, everything is broken.
20081225 15:04:57< mordante> guess you gave the wrong parameters to run?
20081225 15:05:00 * fendrin_ cries.
20081225 15:05:52< loonycyborg> Ivanovic: It runs for me.
20081225 15:05:59< cib0> the problem seems to be that it is called within prestart
20081225 15:06:25-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081225 15:06:29< Ivanovic> mordante: yeah, wrong folder stuff
20081225 15:06:39< fendrin_> cib0: That's easy to fix. Which event should I call it in?
20081225 15:06:40< Ivanovic> how does this think with the "r option" work, mordante?
20081225 15:07:33< fendrin_> cib0: But that code isn't unique. I believe other campaigns have that kind of story only scenarios as well.
20081225 15:07:45< cib0> fendrin_: no, just leave it as it is, it's a bug with the source
20081225 15:07:46 * loonycyborg uses gdb --args ./wesnoth-debug -t to run test scenario
20081225 15:08:02< loonycyborg> Ivanovic: Try something along those lines.
20081225 15:08:12< cib0> by "problem" i only meant "bug triggering"
20081225 15:08:22< mordante> Ivanovic instead of ./wesnoth parameters use gdb wesnoth and r parameters inside gdb
20081225 15:10:15< Ivanovic> http://pastebin.com/m3602f936
20081225 15:11:06< Ivanovic> okay, now complete: http://pastebin.com/m7c3b9459
20081225 15:12:02< loonycyborg> Seems like double deletion of a gobject :)
20081225 15:12:52< mordante> yes looks like it 
20081225 15:13:10< Ivanovic> http://pastebin.com/m21a052b1
20081225 15:13:17< Ivanovic> data from the other threads, just to be sure
20081225 15:13:21< mordante> well I'm just busy to fix the copy problem for ilor, might also fix this issue
20081225 15:13:45< Ivanovic> with this bug it is not possible to get 1.5.7 out since the game just crashes
20081225 15:14:00< Ivanovic> if i can't playtest a single scenario on my computer, how could i release?
20081225 15:14:09< mordante> only very curious why it works for me ;-/
20081225 15:23:53< Ivanovic> some ugly black magic?
20081225 15:24:31< Espreon> mordante: You using powerful magic (like that used by Ancient Liches) does not count.
20081225 15:24:44< Ivanovic> this is on my 64bit machine with these CXXFLAGS:  -O0 -DDEBUG -ggdb3 -W -Wall -ansi
20081225 15:25:33 * mordante looks in his crystal ball, alas no answers there as well
20081225 15:25:40< Ivanovic> using pango-1.22.4
20081225 15:25:47 * mordante summons more powerful magic ;-)
20081225 15:25:57< Ivanovic> and cairo-1.8.6
20081225 15:26:46< Ivanovic> http://pastebin.com/m43a3a7f4
20081225 15:26:53< Ivanovic> output from cmake generation
20081225 15:27:09 * mordante tests his new build
20081225 15:27:33< CIA-54> zookeeper * r31726 /trunk/data/core/macros/ai_controller.cfg: Various tweaks and fixes.
20081225 15:30:14< zookeeper> fendrin_, ok, i'll write some examples for you now
20081225 15:34:32-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081225 15:36:35< CIA-54> mordante * r31727 /trunk/src/gui/ (28 files in 2 dirs): 
20081225 15:36:35< CIA-54> Let the window builder return a pointer.
20081225 15:36:35< CIA-54> The current design of twindow (and some other widget classes) have the
20081225 15:36:35< CIA-54> problem that they shouldn't be copied. The code doesn't copy the objects
20081225 15:36:35< CIA-54> as long as the compiler does a return value optimization, which
20081225 15:36:36< CIA-54> indicates a bad class design. By turning the return value into a pointer
20081225 15:36:40< CIA-54> the copying is no longer needed.
20081225 15:36:44< CIA-54> mordante * r31728 /trunk/src/gui/widgets/ (generator.hpp scrollbar_container.hpp widget.hpp): 
20081225 15:36:47< CIA-54> Make the twidget not copyable.
20081225 15:36:49< CIA-54> As a followup to my previous commit, also make sure a widget can no
20081225 15:36:51< CIA-54> longer be copied.
20081225 15:36:59< mordante> Ivanovic ^ that might fix your problem, ilor you also like that commit
20081225 15:38:16< loonycyborg> mordante: Why some dialogue windows can't be dismissed with a click and havescrollbars?
20081225 15:38:24< loonycyborg> now.
20081225 15:39:04< Espreon> fendrin_: How is LoW broken?
20081225 15:39:09< Ivanovic> mordante: lets see
20081225 15:39:10< mordante> they can be dismissed on my system
20081225 15:39:49< Espreon> *broken/unplayable/etc.
20081225 15:40:01< mordante> first testing something else as well, then I'll look into the scrollbars
20081225 15:40:17< fendrin_> Espreon: Switch to scenario8 and see what is happening.
20081225 15:40:34< Espreon> I shall do so.
20081225 15:40:40< mordante> with my previous commits it might be the ttext class can also be optimized by making it not copyable
20081225 15:41:08< fendrin_> Espreon: But the bug can't be fixed in wml. It's a c++ bug.
20081225 15:41:15< mordante> (which of course causes another full recompilation ;-) )
20081225 15:41:43< cib0> im working on it <.<
20081225 15:42:06< loonycyborg> mordante: There are some of those non-dismissable windows at start of "a Tale of Two Brothers"
20081225 15:42:36< loonycyborg> They all have new transparent portraits.
20081225 15:42:51< fendrin_> loonycyborg: That windows can be dismissed with a keypress. It's only the mouse pointer that can't dismiss them.
20081225 15:43:02< loonycyborg> fendrin_: I noticed.
20081225 15:43:25< loonycyborg> But if other windows can be dismissed with a keypress this is unacceptable.
20081225 15:43:25< Ivanovic> mordante: still a crash
20081225 15:43:43< mordante> the same crash?
20081225 15:43:52< zookeeper> fendrin_, http://wesnoth.pastebin.com/m443eb5a4
20081225 15:44:00< Ivanovic> mordante: a crash at a comparable place
20081225 15:44:09< Ivanovic> mordante: i started outside of gdb, just to check
20081225 15:44:22< Ivanovic> build a debug build now and then running it in gdb
20081225 15:45:11< loonycyborg> Weird. They are dismissable sometimes..
20081225 15:45:49< cib0> uh oh.. i updated svn.. and compiled.. does that mean ill be getting the same problems as you guys now? :p
20081225 15:45:57< Espreon> fendrin: How horrible. But it at least allows me to progress...
20081225 15:46:43< cib0> Espreon: it allows you to progress? over here it throws a plain SEGFAULT
20081225 15:47:18< loonycyborg> Ah. They cease to be dismissable after I click on the scrollbars.
20081225 15:47:32< fendrin_> cib0: The new drawing code is fine here. No crashes.
20081225 15:47:45< Espreon> I am not talking about that.
20081225 15:47:56< Ivanovic> mordante: http://pastebin.com/m5a8c59a6
20081225 15:47:58< Espreon> I was talking about the problems with LoW.
20081225 15:48:17< cib0> yeah, same here, im not getting crashes, either
20081225 15:48:42< cib0> i know, im getting a segfault when hitting end turn in LoW
20081225 15:48:50< Espreon> I am not.
20081225 15:49:12< fendrin_> Espreon: You are allowed to progress, but the game is in a strange state. Some scenarios run through the turns without letting the player a single turn to do. Some other are won or lost by the player at once.
20081225 15:49:17< CIA-54> zookeeper * r31729 /trunk/data/core/help.cfg: Added some stuff to the command mode help section.
20081225 15:49:20< Ivanovic> okay, so it looks really bad for a release today...
20081225 15:49:40 * loonycyborg agrees
20081225 15:50:10< Ivanovic> and seeing how buggy it is, we should give the code at least till saturday to mature and get soem testing
20081225 15:50:32 * loonycyborg wonders why people tend to commit their WIP stuff right before new release
20081225 15:51:05-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20081225 15:51:21< zookeeper> well who doesn't want their cool new stuff in a release ASAP :)?
20081225 15:52:18< CIA-54> mordante * r31730 /trunk/src/text.hpp: 
20081225 15:52:18< CIA-54> Make the class not copyable.
20081225 15:52:18< CIA-54> Now that widgets are not copyable, there's no need to have the ttext
20081225 15:52:18< CIA-54> class copyable. It might cause some problems, but those haven't been
20081225 15:52:18< CIA-54> confirmed yet.
20081225 15:52:30< CIA-54> mordante * r31731 /trunk/src/gui/README: Remove an obsolete document.
20081225 15:52:35< CIA-54> mordante * r31732 /trunk/src/gui/ (90 files in 2 dirs): Remove trailing whitespace.
20081225 15:52:57< mordante> Ivanovic 31730 might fix the problem, not 100% sure however
20081225 15:54:10< fendrin_> mordante: You should fix that portrait issue or remove the patch again. Portrait artists will be pissed off if they are not shown in the release.
20081225 15:54:48< mordante> fendrin_ well I need to get feedback from Kitty first, before I can fix the API
20081225 15:55:10< mordante> which kind of puts us in a vicious circle
20081225 15:55:13< zookeeper> urgh. a lot of _land_ units have _better_ defense on the coastal reef than on land :x
20081225 15:56:17< Ivanovic> http://pastebin.com/m14077d10
20081225 15:56:19< Ivanovic> still crashing
20081225 15:56:41< fendrin_> mordante: Can you explain why? Kitty doesn't care about the wml api. I bet she doesn't even have a glue about that wml thing.
20081225 15:58:02< mordante> Ivanovic and my commit is the only change on the system?
20081225 15:58:42< mordante> fendrin_ yes can, I explained it before and I think you was here as well ;-) But I'll explain again
20081225 15:59:58< mordante> at the moment I have one 400x400 image, which gets scaled down to 200x200. Kitty fears that the quality of the scaling is bad and her portraits will look bad
20081225 16:00:30< loonycyborg> mordante: Clicks seem to dismiss two messages. This is *really* bad.
20081225 16:00:41< mordante> in which case she rather provides a 300x300 or multiple in between sizes
20081225 16:00:54< mordante> which than can be scaled down
20081225 16:01:19< Ivanovic> mordante: when have your commits for the portrait stuff started?
20081225 16:01:30< Ivanovic> tell me the revision, and i test the one before to check if it is working there
20081225 16:01:35< mordante> so if we have multiple images I need another api as only image
20081225 16:03:02< Ivanovic> i'll try r31715
20081225 16:03:03< loonycyborg> (I.e. one click dismisses a dialogue message and one following it)
20081225 16:03:07< Ivanovic> that was before your changes today, mordante
20081225 16:03:28< mordante> Ivanovic no somewhere today, was looking in the log
20081225 16:03:42< mordante> 31710 should be save
20081225 16:03:48< fendrin_> mordante: That's not a good idea. No change to the wml api anymore please. Do something that works with current wml please.
20081225 16:03:58< mordante> fendrin_ impossible
20081225 16:04:06< Ivanovic> mordante: your first commit today was 31716
20081225 16:04:08< mordante> and I did mention that before
20081225 16:04:35< mordante> Ivanovic 31715 should also be save, missed that message
20081225 16:05:15< mordante> loonycyborg hmm, I'll have to look into it, but works properly for me :-(
20081225 16:07:09< loonycyborg> mordante: Try it in a small window.
20081225 16:07:17< Ivanovic> no crash in r31715
20081225 16:07:24< loonycyborg> Mine is 956x709
20081225 16:07:41< mordante> I use 800x600 for testing
20081225 16:08:14< loonycyborg> Then in larger :)
20081225 16:09:42< Ivanovic> this is a plain test with 100% default settings for the game (just music is turned off)
20081225 16:09:49< loonycyborg> Multiple messages dismissed doesn't happen in IftU
20081225 16:09:54< Ivanovic> so it is windowed in 1024x768 in german translation
20081225 16:10:08< loonycyborg> But they still have scrollbars.
20081225 16:11:17< Ivanovic> no crash in 31720 either
20081225 16:13:57< mordante> windowed 1024x768 and German translation works here as well
20081225 16:14:13< mordante> loonycyborg I now also see some dialogs missing :-/
20081225 16:15:09< loonycyborg> Do you see scrollbars?
20081225 16:15:20< mordante> yes but that's a known problem
20081225 16:15:46< loonycyborg> And you wanted this problem go in the release? :)
20081225 16:15:53< Ivanovic> crash in 31723: http://pastebin.com/m15c8fa1
20081225 16:15:56< Ivanovic> starting with this revision
20081225 16:16:20< mordante> loonycyborg well I don't see it as a blocker
20081225 16:16:56< loonycyborg> But if you click on the scrollbar it won't be dismissable anymore.
20081225 16:17:07< loonycyborg> With a click that is.
20081225 16:17:18< mordante> no but that's kind of intended, otherwise you can't scroll
20081225 16:17:37< mordante> I've to see whether I want to show a button to click the dialog away or not
20081225 16:17:44< loonycyborg> May cause a *lot* of annoyance for people who are used to dismissing with a click :)
20081225 16:18:12< loonycyborg> Button is a good idea.
20081225 16:18:27< mordante> well the goal is to show/hide the scrollbars as needed
20081225 16:18:43< loonycyborg> This is a worthy goal :)
20081225 16:18:50< mordante> but at the moment I'm still looking what the best solution is
20081225 16:19:15< Ivanovic> mordante: so yes, your changes *do* break the game
20081225 16:19:28< mordante> btw loonycyborg what platform are you on?
20081225 16:19:39< loonycyborg> Currently on Gentoo.
20081225 16:19:46< mordante> 32 or 64 bit?
20081225 16:19:52< loonycyborg> 32 bit.
20081225 16:20:06< Ivanovic> unstable or stable?
20081225 16:20:26< loonycyborg> Mostly stable :)
20081225 16:20:52 * Ivanovic is using ~amd64
20081225 16:21:43< loonycyborg> Pango and cairo are from unstable.
20081225 16:22:05< Ivanovic> okay, then it should be the same base for what matters
20081225 16:22:12< Ivanovic> though: which gcc are you using?
20081225 16:22:28< mordante> it would be nice if we have some 64 bit testers around
20081225 16:22:31< Ivanovic> this *might* create a difference, too
20081225 16:22:36< Ivanovic> mordante: i am on 64bit
20081225 16:22:43< Ivanovic> mordante: and here it crashes!
20081225 16:22:45< Ivanovic> ;)
20081225 16:22:49< mordante> sorry I meant some more
20081225 16:23:00< fendrin_> only 32bit, no problems here
20081225 16:23:04< loonycyborg> gcc 4.1.2
20081225 16:23:28< loonycyborg> Shadow_Master is on 64bit system :)
20081225 16:24:26< Espreon> I am not. :)
20081225 16:24:39< loonycyborg> Ivanovic: Remind me how do you repro those crashes?
20081225 16:25:03< Ivanovic> loonycyborg: start wesnoth, start two brothers
20081225 16:25:16< Ivanovic> click through the story screens, click through the dialog
20081225 16:25:21< Ivanovic> :debug
20081225 16:25:34< Ivanovic> create some "mage killer" and kill the undead leader
20081225 16:25:47< Ivanovic> in the closing dialog the game exists
20081225 16:25:52< mordante> Ivanovic can you revert r31716 and test again?
20081225 16:26:14< Ivanovic> mordante: starting in which revision?
20081225 16:26:22< Ivanovic> current head and just revert 31716?
20081225 16:26:30< mordante> yep
20081225 16:27:33< Ivanovic> nils@rechner1 ~/wesnoth_up $ svn merge -r31716:31715 svn+ssh://ivanovic@svn.gna.org/svn/wesnoth/trunk .
20081225 16:27:34< Ivanovic> -- Rückwärtiges Zusammenführen von r31716 in ».«:
20081225 16:27:36< loonycyborg> Ivanovic: It successfully enters next scenario for me.
20081225 16:27:36< Ivanovic> U    src/text.cpp
20081225 16:27:38< Ivanovic> U    changelog
20081225 16:27:39< Ivanovic> somthing like this?
20081225 16:28:00< mordante> yep
20081225 16:29:22< Ivanovic> still the crash
20081225 16:30:29< mordante> not too surprised, but wanted to make sure that that commit didn't cause the problem
20081225 16:30:58< Ivanovic> it is in line 387 in data/campaigns/Two_Brothers/scenarios/1*.cfg
20081225 16:31:18< Ivanovic> that is when it switches from the evil one (after dying) to arne and back again
20081225 16:31:27< Ivanovic> then the game is dead...
20081225 16:31:50< mordante> you see the evil dude once?
20081225 16:32:16< loonycyborg> mordante: Also, portraits of Galas and Mal-Keshar aren't used in IftU.
20081225 16:32:26< loonycyborg> Generic portraits are used instead.
20081225 16:32:44-!- ABCD [n=ABCD@wikipedia/ABCD] has quit [Remote closed the connection]
20081225 16:33:06< Ivanovic> loonycyborg: which version of dev-libs/glib are you using?
20081225 16:33:08< mordante> yes known problem, can't be fixed before the portrait API is stable
20081225 16:33:12< Ivanovic> dev-libs/glib-2.18.3 over here
20081225 16:33:22< Ivanovic> mordante: yes, i see him once
20081225 16:33:27< Ivanovic> and i see him the 2nd time
20081225 16:33:36< Ivanovic> just when i click a button it crashes
20081225 16:33:53< mordante> what kind of button?
20081225 16:34:20< loonycyborg> Ivanovic: 2.18.1
20081225 16:35:01< Ivanovic> mordante: *any*
20081225 16:35:09< Ivanovic> like eg the left mousebutton to go on
20081225 16:35:30< loonycyborg> Ivanovic: Try pressing keyboard instead.
20081225 16:35:50< Ivanovic> which one, space, enter, what else?
20081225 16:35:51< loonycyborg> It misses messages with clicks sometimes :)
20081225 16:36:01< loonycyborg> I use enter
20081225 16:36:48< Ivanovic> the same
20081225 16:43:00< mordante> Ivanovic can you try this patch http://paste.debian.net/24405
20081225 16:43:13< Espreon> See you guys later.
20081225 16:43:18-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20081225 16:44:40< mordante> not that I expect too much of it, but I'm running out of ideas
20081225 16:47:53< Ivanovic> mordante: now it was working a little longer
20081225 16:48:07< Ivanovic> that is: it went back to arne again and then it froze
20081225 16:48:49-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081225 16:48:51< Ivanovic> http://pastebin.com/m452d3d9c
20081225 16:49:30< Ivanovic> next would have been the portrait of a spearman, bowman or horseman
20081225 16:51:12< Ivanovic> argh
20081225 16:51:21< mordante> can you try to run under valgrind, see whether that has something to complain about
20081225 16:51:22< Ivanovic> 2nd try and crash at exactly the same (old) place
20081225 16:51:41< Ivanovic> mordante: commands for running it there with params?
20081225 16:51:50< Ivanovic> i got valgrind installed, so just give me the commands
20081225 16:53:05< mordante> valgrind --tool=memcheck -v --leak-check=full --show-reachable=yes wesnoth parameters
20081225 16:53:24< mordante> not sure what's needed when crashing ;-)
20081225 16:55:38< Ivanovic> okay, thanks to valgrind (and -O0) wesnoth is really fast as hell
20081225 16:56:32< loonycyborg> Make it extra fast with -DGLIBCXX_DEBUG :)
20081225 16:57:17< loonycyborg> *-D_GLIBCXX_DEBU
20081225 16:57:25< loonycyborg> *-D_GLIBCXX_DEBUG
20081225 16:59:22< loonycyborg> Btw I'd test Ivanovic's crash with libstdc++ debug mode but I'm afraid dummy unit UB will get me first..
20081225 16:59:54< mordante> loonycyborg cib0 did commit a fix for the dummy unit
20081225 17:00:30< loonycyborg> Great! Revision?
20081225 17:02:01< mordante> 31715
20081225 17:07:01< loonycyborg> Still not getting the crash after glib update.
20081225 17:08:21< Ivanovic> mordante: http://pastebin.com/m5905e2dd
20081225 17:11:06< Ivanovic> loonycyborg: maybe your version of libpng?
20081225 17:11:12< Ivanovic> i am using 1.2.34
20081225 17:13:53< mordante> I really start to think it might be a bug in cairo/pango/glib
20081225 17:14:13< Ivanovic> mordante: this might be as well
20081225 17:14:27< Ivanovic> but i have no probs (noticed) in other progs
20081225 17:14:35< Ivanovic> and so i can just blame wesnoth for the moment
20081225 17:16:26< loonycyborg> Ivanovic: 1.2.22 and it will upgrade to 1.2.33
20081225 17:16:41< loonycyborg> There's no way I'll unkeyword it :)
20081225 17:16:42< Ivanovic> okay, i will download cairo, pango and glib to the last version marked as stable in gentoo
20081225 17:16:49< Ivanovic> that is: =glib-2.16.5 =cairo-1.6.4-r1 =pango-1.20.5
20081225 17:17:19< Ivanovic> loonycyborg: uhm, you should already have 1.2.33
20081225 17:17:23< Ivanovic> http://gentoo-portage.com/media-libs/libpng
20081225 17:17:58< loonycyborg> I have more interesting thing to do than to constantly run emerge world :)
20081225 17:18:04< loonycyborg> *things
20081225 17:18:32< Ivanovic> uhm, no need for world
20081225 17:18:44< Ivanovic> emerge --sync && emerge -uNDa world
20081225 17:19:01< Ivanovic> just a small potion gets updated (when doing so daily ;) )
20081225 17:21:07 * loonycyborg has some packages emerged with command-line overrides of USE and ACCEPT_KEYWORDS
20081225 17:21:22< loonycyborg> And too lazy to fix that :)
20081225 17:21:28< mordante> but I'd like to see more 64 bit tests so we can see whether that might be the problem
20081225 17:21:35< cib0> =)
20081225 17:21:39< mordante> I looked at the code and I see no odd things 
20081225 17:22:18< mordante> cib0 do you run on a 64 bit machine?
20081225 17:22:27< cib0> nope, sorry
20081225 17:22:49< fendrin_> mordante: I hava a 64 bit debian system. Shall I reboot?
20081225 17:22:59< mordante> yes please
20081225 17:23:03< Ivanovic> hehe, now there was no crash in the first try
20081225 17:23:25< Ivanovic> in the 2nd there was one at a slightly different place (with the older versions of those libs)
20081225 17:24:16< Ivanovic> http://pastebin.com/m3fb4c994
20081225 17:24:24-!- fendrin_ [n=fabi@f051072099.adsl.alicedsl.de] has quit [Remote closed the connection]
20081225 17:25:04< Ivanovic> okay, going back to the normal version of my libs since this seems to not have changed anything
20081225 17:25:16< cib0> anyway, i got the bug tracked down.. i already thought i had caused it, but i had no idea how, but it turns out i just made the bug worse so you actually notice it now..
20081225 17:25:54< cib0>  formerly, if you had put an [endlevel]result=victory in a prestart event, it would have crashed the game, with my tweak to endlevel this now also is true for result=continue, so it screws some of the mainline campaigns
20081225 17:26:36< Ivanovic> loonycyborg: uhm, run: emerge -uNDpv world and note down which ones it wants to change to older versions, those are unkeyworded on your side
20081225 17:26:57< Ivanovic> those where it wants to change the useflags you can just add to /etc/portage/package.keyword
20081225 17:29:55-!- fendrin [n=fabi@f051072099.adsl.alicedsl.de] has joined #wesnoth-dev
20081225 17:30:04-!- Rrenys [n=rrenys@81-20-159-197.levira.ee] has quit [Remote closed the connection]
20081225 17:30:07< Ivanovic> wb fendrin
20081225 17:30:18< Ivanovic> so lets see if you can reproduce the issue on your 64bit debian system
20081225 17:30:56< fendrin> grub failed to start it. give me time.
20081225 17:30:58< loonycyborg> Bah! I have packages from no longer existing ebuilds so emerge world fails for me :)
20081225 17:31:15< Ivanovic> loonycyborg: lovely...
20081225 17:31:50< Ivanovic> hey sirp got an android phone....
20081225 17:31:53< Ivanovic> http://www.golem.de/0812/64298.html
20081225 17:32:10< loonycyborg> Ivanovic: http://rafb.net/p/cGF6vU14.html <- That's lovely :)
20081225 17:32:44 * loonycyborg tried lots of kde overlays :)
20081225 17:32:49< Ivanovic> ugh
20081225 17:33:44< Ivanovic> my first reaction would be "emerge -C kde-base/automoc media-sound/phonon app-office/akonadi-server"
20081225 17:33:54< Ivanovic> and with the update, the correct ones should get in again
20081225 17:34:27< loonycyborg> In fact I seems to have svn version of automoc installed currently :)
20081225 17:34:34< loonycyborg> *seem
20081225 17:35:47< mordante> yes I read it on /. as well yesterday
20081225 17:36:08< loonycyborg> mordante: Wrong window? :)
20081225 17:36:21< mordante> no reply to Ivanovic 
20081225 17:38:29< fendrin> The debian64 is history. The grub option is all that is left of it. 
20081225 17:39:06< Ivanovic> fendrin: :(
20081225 17:39:58< mordante> :-(
20081225 17:41:31< zookeeper> cib0, well, [endlevel]result=victory in a prestart event wouldn't really make any sense so it's not used anywhere...
20081225 17:41:37< zookeeper> (i think)
20081225 17:42:09< Ivanovic> mordante: the good news is that i got no crash in the first scenario of httt (okay, only tried once)
20081225 17:42:26< Ivanovic> but in two brothers i do get constant crashes, though the position might vary a little
20081225 17:42:37< fendrin> zookeeper: It's used in LoW all the time.
20081225 17:44:00< zookeeper> fendrin, yeah, well, except for LoW i meant :P
20081225 17:44:16< mordante> Ivanovic well I hate random crashes
20081225 17:44:22< zookeeper> fendrin, btw in those cases i'd suggest using the new endlevel options anyway
20081225 17:44:24< cib0> zookeeper: with the feature you suggested and which is implemented now, you can basically only choose between victory and defeat(with continue being a shortcut for victory), so.. =)
20081225 17:44:28< Ivanovic> mordante: it is not tooo random
20081225 17:44:35< Ivanovic> since the backtrace looks comparable each time
20081225 17:44:42< Ivanovic> and it is in the same mission every time
20081225 17:44:51< mordante> without me being able to reproduce it crashing will also be fix for me
20081225 17:44:55< zookeeper> cib0, yeah, i meant it didn't make sense in the old way
20081225 17:45:08< mordante> yes the trace looks the same, but when it happens is random
20081225 17:45:27< Ivanovic> lets boot my tablet and see if i also get it on this 32bit machine
20081225 17:45:39< Ivanovic> this will take a while since i got to make sure that the packages are recent first
20081225 17:46:07< mordante> no problem, I just work on other things in the meantime
20081225 17:46:33< Ivanovic> and no, i will not recompile all of my system with debug info so that you got info about what exactly it calls where it crashes
20081225 17:47:31< mordante> sometimes you have such good ideas :-P
20081225 17:48:02< mordante> the thought crossed my mind, but I want to see more 64 bit systems before
20081225 17:48:54< Ivanovic> mordante: you got the number for the closest lunatic asylum ?
20081225 17:48:58< Ivanovic> want me to call them for you?
20081225 17:48:59< cib0> when looking through all that meta-scenario code, it really doesn't surprise me that there are so many bugs with it, especially with MP campaigns.. i think it could well use a semantical re-write
20081225 17:48:59< Ivanovic> ;)
20081225 17:49:52< mordante> no I just fear you get it cold in your room, so want the CPU to generate some extra heat ;-)
20081225 17:50:04< mordante> turning up the heat is so boring
20081225 17:50:28< Ivanovic> hmm, i could just turn on my xbox360
20081225 17:50:37< Ivanovic> that is better for heating than just using my computer
20081225 17:50:56< Ivanovic> (the q9300 does not produce this much heat)
20081225 17:51:48< fendrin> zookeeper cib0: I go and move that to an start event or what is the solution?
20081225 17:53:01< CIA-54> cib * r31733 /trunk/src/playsingle_controller.cpp: Fix a bug where calling endlevel in a prestart or start event would cause unexpected behaviour.
20081225 17:53:28< cib0> fendrin: hope that answers your question :p
20081225 17:54:10< Ivanovic> mordante: okay, this system update will take a while, it will also update gcc
20081225 17:54:15< cib0> geez, after not having slept in 24 hours my spelling really isn't that good anymore
20081225 17:54:22< Ivanovic> so i will probably not be able to test before 8pm
20081225 17:54:31< Ivanovic> (since we will be having dinner soon, too)
20081225 17:56:18< mordante> ok
20081225 17:58:26< mordante> Ivanovic can you try to see whether it works on your desktop whether HEAD still works after reverting 31721 (this should restore to old drawing behaviour)
20081225 18:02:49< loonycyborg> Still not getting Ivanovic's crash with -D_GLIBCXX_DEBUG
20081225 18:09:49< ilor> mordante: getting a large amount of warnings and errors in trunk/msvc
20081225 18:10:12< mordante> ilor interesting, what warnings/errors?
20081225 18:10:29< ilor> msvc doesnt like int f() { assert(false); }
20081225 18:10:41< ilor> errors with a "function must return a value"
20081225 18:10:57< loonycyborg> Hm.. gcc merely warns about that..
20081225 18:11:01< ilor> and there's a large amount of http://msdn.microsoft.com/en-us/library/6b3sy7ae(VS.80).aspx
20081225 18:11:36< ilor> loonycyborg: msvc also has a warning for "not all control paths return a value"
20081225 18:12:00< loonycyborg> Why it's an error in this case?
20081225 18:12:13< ilor> no control paths return a value
20081225 18:12:23< mordante> IIRC gcc reconizes assert and determines a return value is not needed
20081225 18:12:56< ilor> asserts don't work in non-debug mode
20081225 18:13:23< mordante> in gcc they do ;-)
20081225 18:13:30< ilor> what?
20081225 18:13:34< loonycyborg> Asserts work unless -DNDEBUG is passed.
20081225 18:13:45< mordante> but obviously it's a bit odd
20081225 18:13:48< ilor> that's my definition of non-debug mode ;)
20081225 18:14:00< loonycyborg> And why anyone would do that with wesnoth?
20081225 18:14:50< mordante> but I'm quite sure I don't want diamond inheritance, can you paste the errors?
20081225 18:15:41< ilor> loonycyborg: do what?
20081225 18:15:58< loonycyborg> Compile with -DNDEBUG
20081225 18:16:02< ilor> mordante: Warning	4	warning C4250: 'gui2::tgenerator<minimum_selection,maximum_selection,placement,select_action>' : inherits 'gui2::policy::placement::thorizontal_list::gui2::policy::placement::thorizontal_list::find_widget' via dominance wesnoth\src\gui\widgets\generator_private.hpp	704
20081225 18:16:25-!- Sirp [n=me@222-154-40-244.jetstream.xtra.co.nz] has joined #wesnoth-dev
20081225 18:17:13< ilor> mordante: also Warning	8	warning C4250: 'gui2::tgenerator<minimum_selection,maximum_selection,placement,select_action>' : inherits 'gui2::policy::placement::tmatrix::gui2::policy::placement::tmatrix::find_widget' via dominance \wesnoth\src\gui\widgets\generator_private.hpp	704
20081225 18:17:59< ilor> loonycyborg: because checking asserts in some tight-loop situations can be useful for debug but harmful for performance?
20081225 18:18:29< loonycyborg> ilor: This is irrelevant if no such assert exist in inner loops.
20081225 18:19:11< ilor> it's somewhat relevant if we use assert(false) as a fatal_error() and /some/ ;) compilers complain 
20081225 18:20:07< ilor> I can add a fatal_error(char* reason) to asserts.hpp that'll get recognized properly by msvc, I think it would be better anyway than an assert
20081225 18:20:27< loonycyborg> Let's fix the compiler then. This should be warning, not error :D
20081225 18:20:40< mordante> we have a function for that in asserts.hpp ;-)
20081225 18:21:14< mordante> not returning a value, when promissing to do that is UB
20081225 18:22:25< loonycyborg> No. Checking return value of that function is UB :)
20081225 18:22:57< ilor> loonycyborg: actually I think it's UB because of some compiler stuff that might end up in stack corruption
20081225 18:23:57< loonycyborg> Anyway, UB isn't supposed to cause compile errors :)
20081225 18:24:32< ilor> mordante: why have a function that does an assert instead of a pure virtual?
20081225 18:25:21< Ivanovic> [17:58:25] <mordante> Ivanovic can you try to see whether it works on your desktop whether HEAD still works after reverting 31721 (this should restore to old drawing behaviour)
20081225 18:25:25< Ivanovic> okay, trying to do so now
20081225 18:26:01< fendrin> cib0: Thank you, that fixed it.
20081225 18:26:04< Ivanovic> my laptop is still at 9/25 packages (running emerge -uND world)
20081225 18:26:12< Ivanovic> that is, curently gcc is built there
20081225 18:26:23< mordante> ilor since the class is instanciated in the macro functions tgenerator_::build
20081225 18:27:24< mordante> loonycyborg afaik UB can do everything and I'm quite sure the not return is UB
20081225 18:27:52< ilor> msvc doesn't recognize abort() as a "won't return
20081225 18:27:55< ilor> grea
20081225 18:27:56< ilor> t
20081225 18:28:37< Ivanovic> mordante: uhm, reverting 31721 makes it still use the new dialogs
20081225 18:28:41< Ivanovic> and i got a crash (again)
20081225 18:29:39< mordante> yes but the old drawing engine
20081225 18:29:56< Ivanovic> but there, there are no problems with the scrollbars
20081225 18:29:58< Ivanovic> ;)
20081225 18:30:01< mordante> you should have less scrollbars with it
20081225 18:30:19< Ivanovic> (as in none are shown compared to a vertical and horizontal bar for every dialog)
20081225 18:30:42< mordante> yes that's expected/wanted
20081225 18:33:00< Ivanovic> ah, okay
20081225 18:33:07< Ivanovic> at least it does not help with the crash
20081225 18:34:33< mordante> it still crashes?
20081225 18:35:05< Ivanovic> yes
20081225 18:35:36< mordante> interesting
20081225 18:37:55< mordante> ilor can you test something? gui/widgets/generator.hpp the tgenerator_ class, let it virtual inherit twidget? (no idea why I didn't put a virtual there in the first place :-/ )
20081225 18:38:07< ilor> mordante: sure
20081225 18:39:19< ilor> still getting the same warnings about find_widget
20081225 18:39:20< mordante> Ivanovic this means to old code was also buggy, only not used that often
20081225 18:41:55< ilor> mordante: anywawy, I disabled these errors and got a build, let's see if it works
20081225 18:45:05< ilor> yay, language dialog popped up
20081225 18:45:35< ilor> it takes an awful lot of time for it to show in debug mode though
20081225 18:46:05< ilor> and I can't click OK because when I mouseover the button the languages list becomec longer and covers the button
20081225 18:46:42< ilor> (awful lot of time is around 4 seconds)
20081225 18:47:28-!- mib_t4mrj9 [i=d4b743e6@gateway/web/ajax/mibbit.com/x-d95fc8fb93a9efce] has joined #wesnoth-dev
20081225 18:47:29< mordante> which works for me :-/
20081225 18:48:06 * mib_t4mrj9 waves shortly, just wanted to check the topic. ;)
20081225 18:48:22< mib_t4mrj9> urks, why didn't it give me my nick ....
20081225 18:48:29-!- mib_t4mrj9 is now known as Rh0nda
20081225 18:48:52< mordante> ah yes that helps, hi Rh0nda 
20081225 18:49:40< Rh0nda> Hi, and bye again. Don't want to get "used" to mibbit or this eepc windows system. :)
20081225 18:50:12< mordante> bye Rh0nda 
20081225 18:52:08< Rh0nda> happy xmas, and if you understand german, see http://a1fie.deviantart.com/ for my christmas wishes :)
20081225 18:54:04< CIA-54> loonycyborg * r31734 /trunk/src/dialogs.cpp: 
20081225 18:54:04< CIA-54> Fixed comparsion of iterators from different sequences UB.
20081225 18:54:04< CIA-54> Used function added in 31715 instead of comparsion.
20081225 18:54:51-!- Rh0nda [i=d4b743e6@gateway/web/ajax/mibbit.com/x-d95fc8fb93a9efce] has quit ["http://www.mibbit.com ajax IRC Client"]
20081225 18:54:53< loonycyborg> Looks like cib0 missed one :)
20081225 18:57:20< loonycyborg> Hm.. Does wesnoth.org timeout for anyone other than me?
20081225 18:57:49< loonycyborg> nwm it works again.
20081225 18:58:10< mordante> ilor does this fix the compiler warnings for you? http://paste.debian.net/24417
20081225 19:00:36-!- ilor_ [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20081225 19:00:48< mordante> ilor_ does this fix the compiler warnings for you? http://paste.debian.net/24417
20081225 19:00:52< ilor_> mordante: seems to
20081225 19:01:00< cib0> loonycyborg: yeah, i lack experience with the right tools so i might have missed quite a few of those
20081225 19:04:03< loonycyborg> cib0: Did I use that function correctly or some other approach was warranted?
20081225 19:08:44< cib0> loonycyborg: well, you used it as i intended it to be used, but im sort of a newb to C++ so i should rather be asking you, i guess :p
20081225 19:09:04< CIA-54> mordante * r31735 /trunk/src/gui/widgets/generator_private.hpp: Make the find_widget calls explicit.
20081225 19:09:12< CIA-54> mordante * r31736 /trunk/src/gui/widgets/generator.hpp: Let tgenerator_ inherit twidget virtual.
20081225 19:09:29< loonycyborg> But you know wesnoth's code better :)
20081225 19:11:50< cib0> i still only know little bits
20081225 19:14:21 * loonycyborg is not very good at wading through other people's code and those dummy unit types made him run screaming :)
20081225 19:15:01< CIA-54> ilor * r31737 /trunk/src/gui/widgets/listbox.hpp: fix struct/class naming inconsistency
20081225 19:18:10-!- ilor [n=user@unaffiliated/ilor] has quit [Read error: 110 (Connection timed out)]
20081225 19:30:59-!- Psyche^ [n=Psyche@e177236082.adsl.alicedsl.de] has joined #wesnoth-dev
20081225 19:32:18-!- Patterner [n=Psyche@e177113157.adsl.alicedsl.de] has quit [Read error: 60 (Operation timed out)]
20081225 19:32:19-!- Psyche^ is now known as Patterner
20081225 19:34:56< CIA-54> zookeeper * r31738 /trunk/data/core/help.cfg: Added a simple terrain description for the coastal reef.
20081225 19:39:13-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has joined #wesnoth-dev
20081225 19:40:13< [Relic]> Hello :)
20081225 19:40:26-!- loonycyborg is now known as [loonycyborg]
20081225 19:40:31< [loonycyborg]> Hello :)
20081225 19:40:35-!- [loonycyborg] is now known as loonycyborg
20081225 19:40:42< [Relic]> :)
20081225 19:41:53< mordante> hi [Relic] 
20081225 19:42:48< zookeeper> fendrin, so, was that paste of mine of any use?
20081225 19:44:08< zookeeper> (i do intend on documenting it somewhat properly in ai_controller.cfg later)
20081225 19:45:10< [Relic]> :)
20081225 19:45:22< fendrin> zookeeper: no, not realy.
20081225 19:46:57-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has quit [Remote closed the connection]
20081225 19:51:28< zookeeper> fendrin, oh :] what exactly do you want to know?
20081225 19:54:13< fendrin> zookeeper: I need to know how to specify a default/fallback ai that takes over when the player clears the orders.
20081225 19:54:44< zookeeper> the default is the [side]'s normal [ai] block(s)
20081225 19:54:58< zookeeper> when clearing orders, those are used
20081225 19:56:51< zookeeper> so when enabling the ai controller in some scenario, you just write the default [ai]'s for the sides normally, and then define the [ai] blocks to use with certain orders in the ai controller, like in my paste
20081225 19:58:02< zookeeper> i did realize that actually i could allow you to write the order-specific [ai] blocks directly in the [side] as well (like [ai] order=aggression ... [/ai]) instead of when calling the controller macro, so i might change it to work like that later..
20081225 20:00:10-!- voris [n=voris@c-24-16-107-149.hsd1.wa.comcast.net] has joined #wesnoth-dev
20081225 20:01:02< ilor_> mordante: did a release build, the language dialog is still annoyingy slow (around 0,5 sec)
20081225 20:04:18< Ivanovic> hi Sirp
20081225 20:04:33< Ivanovic> Sirp: so you got an android phone from google as christmas gratification?
20081225 20:04:44< Ivanovic> how do you like it?
20081225 20:04:58< Ivanovic> mordante: i am currently building latest trunk on my table
20081225 20:05:05< Sirp> Ivanovic: I'm in New Zealand at the moment..... :)
20081225 20:05:13< Ivanovic> in ~5 to 10mins i should know if it works or not
20081225 20:05:28< Ivanovic> Sirp: hey, i don't know when you would have gotten it, might as well have been before you left
20081225 20:06:10< CIA-54> ilor * r31739 /trunk/src/ (5 files in 2 dirs): 
20081225 20:06:10< CIA-54> rewrite aserts.hpp to use a helper function that calls abort(), which, coupled
20081225 20:06:10< CIA-54> with a #ifdef and a __cdecl, makes msvc happier with regards to
20081225 20:06:10< CIA-54> function-must-return-a-value warnings/errors when said function does an
20081225 20:06:10< CIA-54> unconditional ERROR_LOG instead of a return foo;. Also make some functions use
20081225 20:06:12< CIA-54> that ERROR_LOG instead of an assert(false);
20081225 20:06:22< ilor_> mordante: curiously, editor dialogs are fast
20081225 20:06:42< Sirp> Ivanovic: I think I'll get it on Jan 7.
20081225 20:07:05< Ivanovic> ah, okay
20081225 20:08:35< mordante> ilor_ does it still draw over the button?
20081225 20:08:49< ilor_> yes
20081225 20:09:23< mordante> then I think the problem of the slowness is that the dialog tries to draw too much
20081225 20:09:28< CIA-54> ilor * r31740 /trunk/src/ (game.cpp network.cpp playsingle_controller.cpp sound.cpp): fix some unused-variable warnings in catch() blocks msvc complained about
20081225 20:12:45< loonycyborg> ilor_: That dialog is slow with gcc too.
20081225 20:13:27< mordante> odd the language is not really slow for me on -O0 and an old system
20081225 20:13:31 * loonycyborg wonders whether he should fix icc's warnings too :)
20081225 20:14:02< loonycyborg> Define old :)
20081225 20:14:37< mordante> over 5 years
20081225 20:15:23< loonycyborg> My system is 10 years old, though I did some upgrades since then :)
20081225 20:15:49< mordante> no upgrades only a new cpu cooler
20081225 20:15:52< Ivanovic> *WOW*
20081225 20:15:59< Ivanovic> okay, on my tablet i got crashes, too
20081225 20:16:19< Ivanovic> that is there i play through the mission once, quit the campaign, and start it again
20081225 20:16:27< Ivanovic> in the opening dialog it crashes
20081225 20:17:42< Ivanovic> building with debug infos on that box now
20081225 20:18:01< Ivanovic> but yes, it *might* be related to the versions of some libs
20081225 20:19:01< mordante> I would love to get a backtrace of that crash
20081225 20:19:10< mordante> but scenario 1 works as expected?
20081225 20:19:14< Ivanovic> once
20081225 20:19:24< Ivanovic> i had it running once (same test as before)
20081225 20:19:37< Ivanovic> then i quit the campaign to the main menu
20081225 20:19:47< Ivanovic> and started the campaign again
20081225 20:20:26< Ivanovic> then i got a crash in the opening dialog (instead of the "normal" place, the end dialog, on my desktop system)
20081225 20:21:03-!- noy [n=Noy@d75-157-52-251.bchsia.telus.net] has joined #wesnoth-dev
20081225 20:21:11< Ivanovic> mordante: any wishes about the BT?
20081225 20:21:25< mordante> with debug info
20081225 20:21:39< Ivanovic> plain starting in gdb with the debug info compiled in and that's it?
20081225 20:21:52< mordante> yes, no other wishes
20081225 20:22:54< Ivanovic> so basically the same way as i did on my main box
20081225 20:23:00< mordante> indeed
20081225 20:23:42 * Ivanovic is currently rebuilding on his laptop
20081225 20:26:47-!- ilor_ [n=user@unaffiliated/ilor] has quit []
20081225 20:26:52< mordante> I'm afk now, but will be back later
20081225 20:27:03< Ivanovic> mordante: starting in gdb now
20081225 20:28:35-!- Elvish_Pillager [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20081225 20:29:58-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 20:32:24-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has quit [Nick collision from services.]
20081225 20:32:27-!- Blueblaze3 [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 20:33:03< Ivanovic> mordante: http://pastebin.com/m19ef4a71
20081225 20:33:19< Ivanovic> looks strange to me, as in "man, this time it really is a different one..."
20081225 20:36:19 * loonycyborg envies Ivanovic's ability to run into bugs :)
20081225 20:36:50-!- stikonas [n=and@ctv-79-132-180-249.vinita.lt] has joined #wesnoth-dev
20081225 20:39:31-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20081225 20:39:42-!- fendrin [n=fabi@f051072099.adsl.alicedsl.de] has quit [Read error: 104 (Connection reset by peer)]
20081225 20:40:34< Ivanovic> somehow reproducable (on my box): http://pastebin.com/m128b5f3
20081225 20:40:53-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 20:40:58< Ivanovic> that is: both time i had to get into two brothers and finish the first scenario twice
20081225 20:41:27< Ivanovic> at the third time it freezes when the beginning dialog switches to bjarn the first time (the first "new" dialog there)
20081225 20:43:06-!- fendrin [n=fabi@g228074222.adsl.alicedsl.de] has joined #wesnoth-dev
20081225 20:44:47-!- Elvish_Pillage2 [n=eli@66-189-15-130.dhcp.oxfr.ma.charter.com] has quit [Read error: 110 (Connection timed out)]
20081225 20:45:53< ilor> Ivanovic: sound-related stuff in that backtrace... maybe try reverting commit 31642
20081225 20:46:05< ilor> wesbot: log 31642
20081225 20:46:08< wesbot> ilor * r31642 : fix a C++ typo: audio_lock lock(); does not create a variable, it's a function declaration that does nothing. What was most likely intended was audio_lock lock;
20081225 20:46:13< wesbot> URL: http://svn.gna.org/viewcvs/wesnoth?view=rev&rev=31642
20081225 20:46:25-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081225 20:47:00< Ivanovic> ilor: uhm, the crash only starts to happen with one rev from mordante from today
20081225 20:47:04< Ivanovic> without this, there is no crash
20081225 20:47:34< ilor> ah, okay, the sound bits confused me and that commit of mine was iirc the only change related to sound
20081225 20:49:00< Ivanovic> ilor: i just made a backtrace of all threads to be sure
20081225 20:49:22-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081225 20:49:48< Ivanovic> cf this one on my 64bit gentoo box: http://pastebin.com/m3fb4c994
20081225 20:50:24-!- loonycyborg [n=sergey@79.139.247.111] has quit [Read error: 104 (Connection reset by peer)]
20081225 20:54:54-!- Blueblaze3 [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has quit [Read error: 110 (Connection timed out)]
20081225 20:55:45< Ivanovic> asserts.hpp: In function 'void wesnoth_fatal()':
20081225 20:55:45< Ivanovic> asserts.hpp:26: error: 'abort' was not declared in this scope
20081225 20:56:19< Ivanovic> ilor: looks like your changes broke some things...
20081225 20:56:54< Ivanovic> that is with current head on my desktop box using autotools buidl
20081225 20:56:54< ilor> which cpp was that from?
20081225 20:57:13< Ivanovic> In file included from unit_types.cpp:22:
20081225 20:57:31< ilor> asserts.cpp needs to include cstdlib
20081225 20:57:36< ilor> s/cpp/hpp/
20081225 21:01:32-!- Netsplit kubrick.freenode.net <-> irc.freenode.net quits: cib0, Rhonda, stikonas, Sirp, ESR__, crimson_penguin, zookeeper, [Relic], fendrin, noy,  (+25 more, use /NETSPLIT to show all of them)
20081225 21:02:20-!- Netsplit over, joins: fendrin, Rhonda, noy, suokko_, stikonas, Elvish_Pillager, voris, [Relic], isaac, AI0867
20081225 21:02:30-!- Netsplit over, joins: cib0, BenUrban, ESR__, ESR_
20081225 21:03:18-!- Netsplit over, joins: Espreon, Patterner
20081225 21:05:53< Espreon> ...
20081225 21:06:54< voris> wow
20081225 21:08:03< Espreon> Poor CIA-54... poor shikadibot...
20081225 21:10:14-!- busfahrer [n=busfahre@unixboard/user/busfahrer] has joined #wesnoth-dev
20081225 21:10:14-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 21:10:14-!- ilor [n=user@unaffiliated/ilor] has joined #wesnoth-dev
20081225 21:10:14-!- Sirp [n=me@222-154-40-244.jetstream.xtra.co.nz] has joined #wesnoth-dev
20081225 21:10:14-!- crimson_penguin [n=ben@64.201.60.217] has joined #wesnoth-dev
20081225 21:10:14-!- zookeeper [n=l@dsl-tkubrasgw1-fe9ff800-215.dhcp.inet.fi] has joined #wesnoth-dev
20081225 21:10:14-!- Shadow_Master [n=shadowm@unaffiliated/shadowmaster/x-275924] has joined #wesnoth-dev
20081225 21:10:14-!- Ivanovic [n=ivanovic@dtmd-4db261d6.pool.einsundeins.de] has joined #wesnoth-dev
20081225 21:10:14-!- ettin [n=jorda@166.207.219.87.dynamic.jazztel.es] has joined #wesnoth-dev
20081225 21:10:14-!- CIA-54 [n=CIA@208.69.182.149] has joined #wesnoth-dev
20081225 21:10:14-!- Tigge [n=tigge@bacchus.olf.sgsnet.se] has joined #wesnoth-dev
20081225 21:10:14-!- IRSeekBot3 [n=IRSeekLo@irseek/log-bot/IRSeekBot] has joined #wesnoth-dev
20081225 21:10:14-!- wesbot [n=wesbot@asteria.debian.or.at] has joined #wesnoth-dev
20081225 21:10:14-!- freim [n=hogne@feathers.os-tr.net] has joined #wesnoth-dev
20081225 21:10:14-!- shikadibot [n=shikadib@van-gessel.demon.nl] has joined #wesnoth-dev
20081225 21:10:14-!- Smar [n=smar@freenet/translator/finnish/Smar] has joined #wesnoth-dev
20081225 21:10:14-!- Soliton [n=Soliton@rzdspc10.informatik.uni-hamburg.de] has joined #wesnoth-dev
20081225 21:10:14-!- erl [n=erl@vetinte.spelaroll.se] has joined #wesnoth-dev
20081225 21:10:20< voris> Did someone cut the atlantic cable or something? ;)
20081225 21:10:59< Ivanovic> [21:08:55] [Notice] -tomaw- [Global Notice] Hi all, one of our hubs just disappeared taking services with it.  We're looking in to the cause now.
20081225 21:11:37< voris> ahhh
20081225 21:13:26< Espreon> At least shikadibot is back.
20081225 21:13:36< Ivanovic> ilor: still broken
20081225 21:13:48< Ivanovic> let me build with just 1 thread to more easily find where it exactly breaks
20081225 21:14:55< ilor> add an abort() after the calls to wesnoth_fatal() in asserts.hpp to make it quiet
20081225 21:14:55< Ivanovic> http://pastebin.com/m23117510
20081225 21:15:57< Ivanovic> http://pastebin.com/m1f75f01e
20081225 21:17:08< ilor> change the wesnoth_fatal() calls in asserts.hpp to abort()
20081225 21:17:51< ilor> if that doesn't help revert my entire commit and I'll get back to it when I have gcc to test
20081225 21:18:24< ilor> that's r31739
20081225 21:19:20< Ivanovic> http://pastebin.com/me9610ba
20081225 21:19:28< Ivanovic> okay, with this change it does build
20081225 21:20:25< ilor> sorry for the mess, I thought only msvc was dumb enough to require special guidance for a function that does abort() and nothing else
20081225 21:23:35< Ivanovic> okay, i will commit this change
20081225 21:25:18< CIA-54> ivanovic * r31742 /trunk/src/asserts.hpp: compilation fix for gcc 4.3.x
20081225 21:35:51-!- [Relic] [n=[Relic]@adsl-76-229-202-137.dsl.milwwi.sbcglobal.net] has quit ["Leaving"]
20081225 21:39:08-!- loonycyborg [n=sergey@79.139.247.111] has joined #wesnoth-dev
20081225 21:39:24-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has quit [Read error: 60 (Operation timed out)]
20081225 21:42:47-!- Blueblaze [n=Blueblaz@c-98-196-208-51.hsd1.tx.comcast.net] has joined #wesnoth-dev
20081225 21:46:40-!- Sirp [n=me@222-154-40-244.jetstream.xtra.co.nz] has quit [Read error: 110 (Connection timed out)]
20081225 21:53:29< CIA-54> zookeeper * r31743 /trunk/data/core/macros/ai_controller.cfg: Removed a silly forgotten piece of debug code and prevented some unfortunate line-wrapping in menu option text.
20081225 22:01:16-!- stikonas [n=and@ctv-79-132-180-249.vinita.lt] has quit [Read error: 110 (Connection timed out)]
20081225 22:06:09-!- cib0 [n=cib@p4FD0E4BF.dip.t-dialin.net] has quit ["Leaving."]
20081225 22:08:38-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has quit [Remote closed the connection]
20081225 22:25:44< Ivanovic> mordante: ping me when you got anything to test regarding the crash bug
20081225 23:02:43-!- Espreon [n=espreon@c-76-117-239-33.hsd1.pa.comcast.net] has joined #wesnoth-dev
20081225 23:05:12-!- crimson_penguin [n=ben@64.201.60.217] has quit ["Leaving"]
20081225 23:06:51< loonycyborg> Ivanovic: You might also try disabling custom alloc.
20081225 23:07:37< loonycyborg> Maybe mordante's changes exposed a new bug in it..
20081225 23:12:15< loonycyborg> Or it interferes with glib in bogus ways on 64bit systems..
20081225 23:12:48< mordante> loonycyborg stole my thoughts ;-)
20081225 23:15:19< mordante> only I wonder why that breaks on 32 bit as well
20081225 23:21:35< mordante> Ivanovic what did you exactly do to trigger the problem on your thinkpad?
20081225 23:28:57< Ivanovic> start two brothers on easy
20081225 23:29:12< Ivanovic> go through the story screen, click through the starting dialog
20081225 23:29:47< Ivanovic> switch to debug mode and create a berserk close the the enemy leader, kill him, click through the end dialog, when in "post scenario mode" exit campaign to main menu
20081225 23:29:50< Ivanovic> repeat once
20081225 23:30:12< Ivanovic> in the third try, it will exit when the first new dialog (by bjarn) comes up
20081225 23:30:52< Ivanovic> that is: always trying to do things exactly the same way, in the third try it does quit the game with the message "Abgebrochen" (aborted) in the terminal
20081225 23:31:26< mordante> how evil to use a beserker ;-) I'll try to reproduce it now
20081225 23:37:05< Ivanovic> ok
20081225 23:37:06< mordante> oke I get a crash the second time
20081225 23:37:47-!- stikonas [n=and@ctv-79-132-188-113.vinita.lt] has joined #wesnoth-dev
20081225 23:38:27< mordante> I haven't compiled in debug mode so can't look at it further, but at least I can now reproduce it :-)
20081225 23:38:44< mordante> did you already try with disabling the custom malloc?
20081225 23:39:14< loonycyborg> I can't :(
20081225 23:39:24< loonycyborg> Tried four times.
20081225 23:52:13< Ivanovic> no, i have not tried to disable it
20081225 23:52:20< Ivanovic> but at least you do get it now, too
20081225 23:52:24< Ivanovic> afk again
20081225 23:52:32< Ivanovic> (basically already lying in bed...)
20081225 23:52:58< mordante> well I'm about to do that, guess we'll have to look further tomorrow
20081225 23:53:18< mordante> at least I'm sure I won't fix it today (local time ;-) )
20081225 23:54:11< Shadow_Master> have youu managed to fix the remaining issues?
20081225 23:54:35< Shadow_Master> I saw theat LoW was not working last night
20081225 23:55:13< Shadow_Master> merry xmas btw :) 
20081225 23:55:48< mordante> merry xmas to you too
20081225 23:55:58< mordante> I didn't do anything with LoW
20081225 23:56:29< loonycyborg> Yes. It's zookeeper's doing :)
20081225 23:56:32< Shadow_Master> I know you aren't guilty ;) 
20081225 23:56:34< zookeeper> what
20081225 23:56:40 * Shadow_Master hugs mordante 
20081225 23:56:40< zookeeper> oh, that.
20081225 23:56:56< mordante> thanks Shadow_Master :-)
20081225 23:57:28< Shadow_Master> remember what I told you about the spacebar and the new characterd ialogs ;)
20081225 23:57:32< Shadow_Master>  
20081225 23:58:00< Espreon> Hello Shadow_Master.
20081225 23:58:05< mordante> yes it's on the place which shall not be mentioned ;-)
--- Log closed Fri Dec 26 00:00:45 2008