--- Log opened Sun Sep 19 00:00:06 2010
--- Day changed Sun Sep 19 2010
20100919 00:00:06< eleazzaar> three color variations seems about right
20100919 00:00:19< eleazzaar> we have 3 colors of grass, 3 colors of dirt
20100919 00:01:46< CIA-35> zookeeper * r46586 /trunk/data/campaigns/Liberty/ (33 files in 2 dirs): Replaced the old rogue mage line sprites with new (unanimated) ones by Blarumyrran. Animation WML and spell effects still need cleaning up.
20100919 00:01:47< zookeeper> hmh, well, if you want it that way...
20100919 00:02:17< eleazzaar> as long as it isn't gonna hurt performance
20100919 00:02:41< alink> eleazzaar: with the current disabled animation for water transitions (which I think looks ok) it has no cost
20100919 00:02:49-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100919 00:03:30< alink> if enabled again, then we need to fix the sync issue, and if done right it may sill have no cost
20100919 00:03:58< eleazzaar> alink: zookeeper: i saw a mod that color-shifted the water to create a new terrain
20100919 00:03:58< alink> no cost from framerate, but I am a bit more concerned about memory
20100919 00:04:55< eleazzaar> nearly 300 PNGs per complete water terrain
20100919 00:05:53< alink> with the current 15(!) frames used by water, each water variant cost as much as 15 terrains variant, so need to decide how to spend the (limited somehow) ressources
20100919 00:05:55< zookeeper> alink, any idea if it'd hurt performance (after initial caching, or whatever) if imagepathfunctions could be used in terrain rules?
20100919 00:06:15< zookeeper> something like [image] image_mod=<imagepathfunctions>
20100919 00:06:52< eleazzaar> alink: generally i wouldn't expect all water terrains to be used on the same map
20100919 00:07:16< eleazzaar> does that make a difference or all even unused terrains loaded into memory?
20100919 00:07:19< alink> zookeeper: imagepathfunctions is just like using a new image file, a bit postprocessing time (depending of the effect) but maybe even less than read a true image on disk
20100919 00:08:09< SpoOkyMagician> anonymissimus: thanks for #16720
20100919 00:08:21< alink> eleazzaar: currently we don't clean things well, so if used on a map, some parts stay used during the next scenario but we can fix that
20100919 00:08:25< anonymissimus> I wanted thet severel times myself
20100919 00:08:34< SpoOkyMagician> heh
20100919 00:09:19< alink> eleazzaar: and yes maybe just need that the map creator may use terrain complexity wisely
20100919 00:09:19< anonymissimus> for id it's more important than for type, since you other wise may need quite some [or]s
20100919 00:09:35< SpoOkyMagician> indeed
20100919 00:10:25< zookeeper> alink, ok, so you think it'd be a good idea to do things like different-colour variations of water using such a feature, if we had it?
20100919 00:10:37< alink> zookeeper: actually if terrain images used an image_mod instead of~FOO(), that could probably help parsing
20100919 00:11:29< alink> zookeeper: yes it could even be good for perfomance (less disk space and I/O)
20100919 00:12:09< eleazzaar> i can't think of any other terrain i would want to color shift (besides water)
20100919 00:12:15< alink> (for loading)
20100919 00:12:41< eleazzaar> 1 ocean terrain is 2.1 MB
20100919 00:13:09< alink> eleazzaar: not the same in memory
20100919 00:13:40< alink> nvm, you probably mean for disk space
20100919 00:13:45< eleazzaar> yeah
20100919 00:14:59< zookeeper> right, i forgot that you could (i think) currently already use IPF's for terrain
20100919 00:15:57< zookeeper> so i suppose it'd be possible to make a new builder macro which would allow you do specify IPF's to use
20100919 00:16:51< zookeeper> eleazzaar, so unless i'm missing something, i think i could add what i just described without big difficulties
20100919 00:17:17< alink> one last thing worrying me a little about the (15frames * types of water) is the local ToD light transitions. I assumed that the number of hexes at the interface between ToD areas should be low. But now I realize that any water hex count as 15. Hopefully you will not mix too much  light variations and water wariations
20100919 00:17:34< zookeeper> would a color shift using ~CS be good enough for creating your water variant?
20100919 00:18:22< zookeeper> http://wiki.wesnoth.org/ImagePathFunctionWML#Color-shift_function
20100919 00:19:40< eleazzaar> zookeeper: i'm interested in trying, if it is just as fast
20100919 00:21:39-!- SpoOkyMagician is now known as SpoOkyMagician_
20100919 00:22:00< alink> mmmh sorry, I was maybe too quick about memory usage. I assumed that the source image was already in cache
20100919 00:22:37< alink> if not, it needs to load source and source~IPF
20100919 00:22:56< eleazzaar> zookeeper: can ~GS() be used to *partially* desaturate?
20100919 00:23:13< alink> but source shouldn't need ToD versions and hex-cutting, so that should be lighter, I need to check
20100919 00:23:14< zookeeper> alink, what determines whether or not it is? is it just a matter of which terrain the game would try to render first?
20100919 00:23:23< zookeeper> eleazzaar, nope
20100919 00:24:17-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100919 00:26:06< alink> zookeeper: if both are used on the map then for memory it costs 2 full water types. If only source~IPF is used on map, memory will holds 1 full source~IPF and 1 light source (no ToD, hex-cut)
20100919 00:26:22< alink> but we may optimize this later
20100919 00:27:18< eleazzaar> so auto-color-shifted water would be either better, or the same?
20100919 00:27:28< alink> the only reason to keep source in memory is to avoid needing to reload the png when another source~IPF2 is needed
20100919 00:28:14< alink> eleazzaar: better when both are used on map, a bit less good when only source~IPF is used
20100919 00:28:51< eleazzaar> "source~IPF" = color shifted water?
20100919 00:28:52< alink> but as I said we may be able to optimize that.
20100919 00:28:58< alink> eleazzaar: yes
20100919 00:29:35< alink> but the gain of disk-space and I/O is a clear win
20100919 00:29:50< eleazzaar> OK, so how do i do this?
20100919 00:30:21< eleazzaar> assuming i've figured out how i want to change the RGB
20100919 00:30:57< alink> without terrain macro you just use "water.png~CS(5,5,5)" but maybe zookeeper can provide you a macro
20100919 00:31:29< alink> "water.png~CS(5,5,5)" <-- of course that ignore all the complexity of animation and transition ;-P
20100919 00:32:44< alink> the small horror is that it may require to copy-paste the ugly animation macros
20100919 00:33:56< alink> anyway, I should check how to optimize that cache for IPF on terrains, that's a powerfull way to generate terrain images
20100919 00:34:32< alink> IIRC I already mentioned "grass.png~BLIT(flower.png)"
20100919 00:34:41< zookeeper> eleazzaar, i can (try to) write it for you, but i'm going to bed pretty soon so it'll have to wait until tomorrow
20100919 00:34:51< eleazzaar> zookeeper: that sounds good
20100919 00:35:25< alink> and we could use "grass.png~MASK(transition-mask-n.png)" to autogenerate transition 
20100919 00:35:39< alink> (~MASK() is not implemented yet)
20100919 00:37:04< zookeeper> eleazzaar, ok, i'll let you know when i've got something
20100919 00:37:21< eleazzaar> zookeeper: thanks
20100919 00:37:59< eleazzaar> alink: several terrains do use what are essentially the same masks
20100919 00:39:54< alink> eleazzaar: so that "grass.png~MASK(transition-mask-n.png)" could decently replace Gimp work by runtime postprocessing ?
20100919 00:40:32< eleazzaar> in some circumstances
20100919 00:41:06< alink> yeah I assume that we can't do castles like that :P
20100919 00:41:11< eleazzaar> heh
20100919 00:42:13< eleazzaar> hills, mountains = no
20100919 00:42:22< eleazzaar> sand, beach = yes
20100919 00:43:22< eleazzaar> dirt  = no, roads = yes
20100919 00:44:19< alink> (poor) animated water by blitting waves at different locations and alpha on a blue tile?
20100919 00:44:44< eleazzaar> random locations?
20100919 00:44:53< alink> kidding, that would be an horrible ~IPF
20100919 00:45:13< alink> eleazzaar: well still using 15 frames, so no, need to specify them
20100919 00:46:20< alink> but to get my previous example, "grass.png~BLIT(flower.png~CS(100,0,0),15,15)~BLIT(flower.png~CS(0,100,0),60,15) " to blit red and blue flowers
20100919 00:49:23< eleazzaar> what's the advantage of BLIT over just using an overlay?
20100919 00:50:32< alink> that's a new idea, but that should be better for framerate, the long stuff above, is just one image to blit
20100919 00:51:03< alink> with overlay it's 3 images to blit each time, and we lose time to blit the grass under flowers
20100919 00:52:25< alink> and half-transparent overlay need alpha-blending each times
20100919 00:52:58< eleazzaar> that may make sense in a different instance, but with flowers work best as an overlay since they often go over grass-to-grass transitions
20100919 00:53:17< eleazzaar> or other transitions
20100919 00:53:58< alink> eleazzaar: yes, I suppose, my example is not for arts, I just pick it to show the syntax 
20100919 00:54:27-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100919 00:54:42-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100919 00:56:54< alink> anyway the point was if you have 2 similar terrains, it's maybe not a bad idea to use such trick to generate them (as the suggested water trick)
20100919 00:57:46< alink> ~CS() to change color ~BLIT() to add an overlay on it, maybe also ~FLIP() etc..
20100919 00:58:49< alink> but I should probably optimize that part of the cache before advertising such methods :)
20100919 01:03:08< alink> FYI, the current local ToD prototype simply use these IPF with something like grass.png~LIGHT(base-light.png~CS(15,5,5)~BLIT(light-trans-n.png~CS(50,20,20)))
20100919 01:03:55< alink> which shows the power of IPF (and the ugliness of this prototype :-/)
20100919 01:07:05-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has left #wesnoth-dev []
20100919 01:07:42-!- SpoOkyMagician_ [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: be back in a bit]
20100919 01:18:53-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100919 01:20:19-!- johndh [~johndh@adsl-85-37-244.mco.bellsouth.net] has quit [Ping timeout: 240 seconds]
20100919 01:28:01< zookeeper> eleazzaar, have you noticed that the ford transitions have a slightly different colour/opacity/something than the ford base tiles themselves?
20100919 01:28:18< zookeeper> pretty clearly visible for example in DiD's first scenario
20100919 01:29:09< eleazzaar> it's because of the way the transparent transitions overlap
20100919 01:29:34< eleazzaar> i'd like to redo the ford from scratch
20100919 01:29:51< eleazzaar> so i haven't bothered with making lots of transitions
20100919 01:30:05-!- Upth [ogmar@adsl-75-26-208-67.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100919 01:31:12< zookeeper> okay
20100919 01:33:50-!- Upthorn [ogmar@adsl-75-26-191-169.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 272 seconds]
20100919 01:36:39-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100919 01:54:31-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100919 01:55:14-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100919 01:58:11-!- Greywhind [~Greywhind@96.238.16.162] has joined #wesnoth-dev
20100919 02:04:03-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100919 02:04:25-!- ancestral [~ancestral@166.137.143.60] has joined #wesnoth-dev
20100919 02:09:35-!- PeterPorty [~Pete@pc-253-249-120-200.cm.vtr.net] has joined #wesnoth-dev
20100919 02:15:37-!- wesbot changed the topic of #wesnoth-dev to: 159 bugs, 296 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100919 02:17:13-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 02:25:03-!- PeterPorty [~Pete@pc-253-249-120-200.cm.vtr.net] has left #wesnoth-dev ["Saliendo"]
20100919 02:31:25-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: im going out again for a bit. be back in about 1-2 hours. later]
20100919 02:52:44-!- shadowm_laptop [~ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 255 seconds]
20100919 03:00:09-!- anonymissimus [~chatzilla@HSI-KBW-095-208-007-247.hsi5.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819]]
20100919 03:18:55-!- johndh [84aa0b84@gateway/web/freenode/ip.132.170.11.132] has joined #wesnoth-dev
20100919 03:26:51-!- ancestral [~ancestral@166.137.143.60] has quit [Quit: Smell ya later!]
20100919 03:33:33-!- happygrue [~George@wesnoth/developer/wintermute] has joined #wesnoth-dev
20100919 03:50:01-!- loonybot [~loonybot@ppp95-165-187-6.pppoe.spdop.ru] has joined #wesnoth-dev
20100919 03:50:01-!- loonybot [~loonybot@ppp95-165-187-6.pppoe.spdop.ru] has quit [Changing host]
20100919 03:50:01-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100919 03:50:39-!- shadowm_A [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 03:50:48-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100919 03:55:33-!- shadowm_A is now known as shadowm_laptop
20100919 03:55:40-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100919 03:56:04-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100919 03:56:35-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100919 03:59:25-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has joined #wesnoth-dev
20100919 04:05:56-!- Ivanovic_ [~ivanovic@dtmd-4db2c6c0.pool.mediaWays.net] has joined #wesnoth-dev
20100919 04:08:08-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has quit [Quit: And that's the end of THAT chapter]
20100919 04:09:28-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 272 seconds]
20100919 04:09:52-!- Ivanovic_ is now known as Ivanovic
20100919 04:12:20< CIA-35> eleazar * r46587 /trunk/data/ (5 files in 3 dirs): converted volcanoes from 'Qv' to 'Mv' for saner and easier wildcarding.Think i got all the maps that used it too.
20100919 04:20:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100919 04:34:53-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 255 seconds]
20100919 05:04:15-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has joined #wesnoth-dev
20100919 05:06:02-!- Blueblaze [~Blueblaze@adsl-99-185-8-197.dsl.hstntx.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100919 05:07:10-!- Blueblaze [~Blueblaze@adsl-99-185-8-197.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100919 05:33:53-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Disconnected by services]
20100919 05:34:18-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 05:44:56-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20100919 05:44:59-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: GO, GET TO THE CHOPPAH!!!]
20100919 05:45:01-!- Gambit [~Gambit@unaffiliated/gambit] has quit [Read error: Connection reset by peer]
20100919 05:45:24-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100919 05:54:03-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20100919 06:09:02-!- fangism1 [~davidfang@gilbert.achronix.com] has joined #wesnoth-dev
20100919 06:10:26-!- fangism2 [~davidfang@gilbert.achronix.com] has quit [Ping timeout: 272 seconds]
20100919 06:49:04-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 272 seconds]
20100919 06:49:36-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100919 07:25:32-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 276 seconds]
20100919 07:30:17-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100919 07:30:29-!- shadowm_unstable [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 07:32:08-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: taking a break]
20100919 07:36:18-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100919 07:44:12-!- shadowm_unstable is now known as shadowm_laptop
20100919 07:46:19< shadowmaster> Rhonda: can I have a copy of the custom version of seen.pl you use for wesbot?
20100919 07:46:53< shadowmaster> since the irssi-scripts package only has this  mess that talks in Polish, and I don't really speak Polish.
20100919 07:50:22-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has joined #wesnoth-dev
20100919 07:50:22-!- silene [~plouf@bau91-1-82-239-244-109.fbx.proxad.net] has quit [Changing host]
20100919 07:50:22-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100919 07:51:43-!- Upthorn [~ogmar@adsl-75-26-183-62.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100919 07:54:37-!- Upth [ogmar@adsl-75-26-208-67.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 252 seconds]
20100919 07:56:52-!- Upth [ogmar@75.26.163.0] has joined #wesnoth-dev
20100919 07:57:20-!- Upthorn [~ogmar@adsl-75-26-183-62.dsl.scrm01.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100919 07:57:21-!- dtiger [~dtiger@dynamic-vpdn-93-125-12-3.telecom.by] has joined #wesnoth-dev
20100919 08:03:26-!- Upthorn [ogmar@adsl-75-26-163-0.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100919 08:04:18-!- freim [~hogne@feathers.os-tr.net] has quit [Ping timeout: 245 seconds]
20100919 08:04:44-!- Upth [ogmar@75.26.163.0] has quit [Ping timeout: 240 seconds]
20100919 08:07:02-!- freim [~hogne@feathers.os-tr.net] has joined #wesnoth-dev
20100919 08:24:57-!- joo is now known as joo|sleep
20100919 08:25:32-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100919 08:27:58-!- fangism2 [~davidfang@gilbert.achronix.com] has joined #wesnoth-dev
20100919 08:28:28-!- fangism1 [~davidfang@gilbert.achronix.com] has quit [Ping timeout: 245 seconds]
20100919 08:48:54-!- Dalton is now known as Unaffiliated
20100919 08:49:11-!- Unaffiliated [~~@unaffiliated/op] has quit []
20100919 08:49:32-!- Dalton [~~@unaffiliated/op] has joined #wesnoth-dev
20100919 09:32:14-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100919 09:46:19-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100919 10:00:06-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100919 10:04:53-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20100919 10:05:19-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 10:05:47-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Client Quit]
20100919 10:24:06-!- Blueblaze [~Blueblaze@adsl-99-185-8-197.dsl.hstntx.sbcglobal.net] has left #wesnoth-dev []
20100919 10:28:59-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100919 10:29:07-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100919 10:29:07-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100919 10:32:27-!- Ivanovic [~ivanovic@dtmd-4db2c6c0.pool.mediaWays.net] has quit [Changing host]
20100919 10:32:27-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100919 10:33:53< Ivanovic> moin
20100919 10:35:52-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100919 10:36:29-!- ancestral [~ancestral@97-116-168-17.mpls.qwest.net] has quit [Quit: And that's the end of THAT chapter]
20100919 10:42:20< zookeeper> eleazzaar, did you mean to change the default base of the snowy human city village back to Rrc in your last commit?
20100919 10:42:30< zookeeper> (looks likely accidental to me)
20100919 10:47:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 245 seconds]
20100919 10:56:07< Ivanovic> eleazzaar: have you asked esr for a wmllint conversion of the volcano?
20100919 11:07:43-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has joined #wesnoth-dev
20100919 11:12:29< zookeeper> silene, if i want to check how the preprocessor expands a certain terrain graphics macro mess, what's the easiest way to get to display or output the expanded form?
20100919 11:13:50< silene> zookeeper: just open the cache file
20100919 11:14:08< Ivanovic> silene: you saw that i assigned two bugs to you some days ago?
20100919 11:14:12< silene> yes
20100919 11:14:38< Ivanovic> okay
20100919 11:15:50< zookeeper> silene, hmh, ok, i didn't realize those actually contain readable WML. any way to tell which cache file is which?
20100919 11:16:25< silene> zookeeper: no, it's a bit complicated; i usually empty the cache first and then open the biggest file
20100919 11:20:27< zookeeper> right
20100919 11:24:16-!- EdB [~edb@79.89.103.70] has joined #wesnoth-dev
20100919 11:29:22-!- johndh [84aa0b84@gateway/web/freenode/ip.132.170.11.132] has quit []
20100919 11:29:31-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100919 11:36:45< Soliton> Simons_Mith: you can just report it in here anyone can do such a change though if noone reacts you should better file a bug report or it gets forgotten.
20100919 11:50:48-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20100919 11:53:29-!- Soliton [~Soliton@wesnoth/developer/soliton] has quit [Ping timeout: 276 seconds]
20100919 11:54:17-!- Soliton [~Soliton@wesnoth/developer/soliton] has joined #wesnoth-dev
20100919 12:02:03-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20100919 12:03:07-!- mjs-de [~mjs-de@p3EE236D8.dip.t-dialin.net] has joined #wesnoth-dev
20100919 12:05:41-!- fangism [~davidfang@gilbert.achronix.com] has joined #wesnoth-dev
20100919 12:07:41< eleazzaar> zookeeper: The default base of snowy human village has been Rrc for some time now
20100919 12:07:51-!- fangism2 [~davidfang@gilbert.achronix.com] has quit [Ping timeout: 265 seconds]
20100919 12:09:39-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100919 12:09:46-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Client Quit]
20100919 12:09:52< zookeeper> eleazzaar, yeah, but i changed it to Aa a few days ago
20100919 12:11:02< zookeeper> btw, i'm almost done with the water thing, still struggling with transitions a bit
20100919 12:11:29< Ivanovic> water thing?
20100919 12:11:39< Ivanovic> is that the beach animation?
20100919 12:11:43< Ivanovic> *dreaming*
20100919 12:15:31< zookeeper> no
20100919 12:16:42< Ivanovic> :(
20100919 12:16:59< CIA-35> eleazar * r46588 /trunk/data/tools/wmllint: added the volcano terrain code change to WMLLINT.
20100919 12:17:49-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20100919 12:17:49-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20100919 12:17:49-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100919 12:17:53< zookeeper> ohh, cool, i think i got the transitions too now
20100919 12:18:00< mordante> servus
20100919 12:18:07< Ivanovic> hi mordante
20100919 12:18:10< mordante> hi Ivanovic 
20100919 12:18:34< Ivanovic> eleazzaar, zookeeper: i think there is something a little broken with water when animations are turned off
20100919 12:18:47< Ivanovic> i think i saw some black edges then
20100919 12:19:44< CIA-35> mordante * r46589 /trunk/ (changelog src/gui/dialogs/title_screen.cpp): 
20100919 12:19:44< CIA-35> Fullscreen hotkey works again in the title screen.
20100919 12:19:44< CIA-35> Since switching to fullscreen mode on my system the code hasn't been
20100919 12:19:44< CIA-35> fully tested, so some testing would be nice.
20100919 12:20:04< mordante> would be nice if somebody tests it^
20100919 12:20:19< eleazzaar> Ivanovic: should not be any black edges
20100919 12:20:27< Ivanovic> mordante: you saw the comments about 1.9.1?
20100919 12:20:33< Ivanovic> about the new main menu
20100919 12:20:44< mordante> Ivanovic, no, still catching up in the log
20100919 12:20:46< Ivanovic> personally i really prefer it in 800x480, looks a lot better there now
20100919 12:21:11< Ivanovic> and yeah, the logo fade in animation makes things crawl
20100919 12:21:23< mordante> really? works nice for me
20100919 12:21:33< mordante> I could tweak some parameters
20100919 12:21:35< Ivanovic> (on the pandora in 1.9.1 that is, have switched it off in the default config)
20100919 12:21:56< Ivanovic> alink improved some stuff, cf https://gna.org/bugs/index.php?16724
20100919 12:22:08< Ivanovic> about the logo: https://gna.org/bugs/index.php?16722
20100919 12:23:07< Ivanovic> something more about menu response time: http://forums.wesnoth.org/viewtopic.php?p=456891#p456891
20100919 12:23:07< mordante> I've an idea about the resize issue, working on a fix/work-around now
20100919 12:23:47< Ivanovic> some more comment regarding the main menu: http://forums.wesnoth.org/viewtopic.php?p=456885#p456885
20100919 12:24:45< zookeeper> eleazzaar, my water IPF thing also changes all the existing water transitions to use my new little macro...if i commit that, can you check that nothing's broken?
20100919 12:25:19< mordante> to be honest I think the thing stuck in the middle looked silly, so this is as wanted
20100919 12:26:40< CIA-35> caslav_ilic * r46590 /branches/1.8/po/wesnoth-utbs/ (sr.po sr@ijekavian.po sr@ijekavianlatin.po sr@latin.po): Updated translations for Serbian (intentionally not applying to trunk too).
20100919 12:27:50< Ivanovic> mordante: huh? what do you mean with "the thing stuck in the middle"?
20100919 12:28:07< eleazzaar> zookeeper: yes, i'll check it once committed
20100919 12:28:07< mordante> the button row in the middle 
20100919 12:28:45-!- loonybot [~loonybot@ppp95-165-187-6.pppoe.spdop.ru] has joined #wesnoth-dev
20100919 12:28:45-!- loonybot [~loonybot@ppp95-165-187-6.pppoe.spdop.ru] has quit [Changing host]
20100919 12:28:45-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100919 12:28:45< Ivanovic> i think the main problem is that it looks really strange and *huge* when switching to eg 1920x1200
20100919 12:29:09< CIA-35> zookeeper * r46591 /trunk/data/core/ (3 files in 2 dirs): Added two new macros to internal-generic.cfg for creating IPF-capable animated water terrains and transitions, changed all water transitions to use them and added two new colour-shifted water variant terrains.
20100919 12:29:12< mordante> maybe the tips area can have a maximum size
20100919 12:29:22< zookeeper> eleazzaar, ^ there you go. you just need to add editor icons for the new terrains, since i can't do those using ~CS
20100919 12:29:38-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100919 12:30:09< eleazzaar> mordante: Ivanovic generally speaking a good maximum line lenght is 80-120 characters
20100919 12:30:11< zookeeper> i had to write new macros since i didn't find a way to abuse the builder macros after all
20100919 12:30:33< eleazzaar> readability suffers for a single line of text longer than that
20100919 12:30:44< Ivanovic> eleazzaar: personally i think that 80 is not really good and not enough, 120 is what i think works best
20100919 12:30:53< mordante> eleazzaar, any idea how many pixels that would be?
20100919 12:32:03< eleazzaar> i'd set it up and measure, but the whole screen turns black when i resize the screen via the handle in the corner
20100919 12:32:24< eleazzaar> build is a few days old
20100919 12:32:26< mordante> I could, just hoped you would know the number ;-)
20100919 12:32:36< Ivanovic> mordante: could you please try this: (use an older build before the commit from alink to optimize the logo blade in
20100919 12:32:44< Ivanovic> select the software mouse cursor
20100919 12:32:58< Ivanovic> restart wesnoth and try to move the mouse while the logo is fading in
20100919 12:33:21< Ivanovic> especially obvious in large resolutions
20100919 12:33:52< Ivanovic> and yeah, your commit with ctrl+f works nicely
20100919 12:34:13< mordante> I first want to look further at #16724 will test the other things later
20100919 12:34:17< mordante> thanks for testing
20100919 12:34:27< Ivanovic> i would welcome the logo blending in faster
20100919 12:34:35< Ivanovic> it really takes a while now, too much IMO
20100919 12:37:08< eleazzaar> mordante: i'd say 720 pixels is a about right for a max for that tip window
20100919 12:37:44< mordante> thanks eleazzaar 
20100919 12:40:57< ilor> hi all
20100919 12:41:02< Ivanovic> in 1024x768 when german is selected the whole box inluding borders has about 770px
20100919 12:41:08< Ivanovic> (tip of the day that is)
20100919 12:41:15< Ivanovic> i think this is about the perfect size
20100919 12:41:26< Ivanovic> hi ilor
20100919 12:41:34< ilor> I have the new addon server merged with trunk locally in git, suggestions on how I should proceed with pushing it to svn?
20100919 12:41:52< ilor> it was forked off 1.8.0 and my local git merge history shows that
20100919 12:42:08< eleazzaar> zookeeper: it seems to work
20100919 12:42:29< Ivanovic> ilor: some days ago Rhonda already observed that we got a *really* high number of connections that are never correctly closed and Soliton said that it was related to the addon server, really looking forward to see the replacement available
20100919 12:42:54< Ivanovic> ilor: it is completely written from scratch, correct?
20100919 12:43:13< ilor> Ivanovic: the addon server is a new bit of python code, yes
20100919 12:43:22< ilor> but there had to be some changes in game code 
20100919 12:44:36< zookeeper> eleazzaar, awesome
20100919 12:44:59< Ivanovic> ilor: hmm, not sure at all
20100919 12:45:10< eleazzaar> the colors aren't quite what i wanted, but this makes my job much easier
20100919 12:45:15< Ivanovic> ilor: maybe you want to talk to Soliton if we want to replace the 1.8 server
20100919 12:45:21< zookeeper> eleazzaar, hmh, actually i forgot to update the layer numbering when i added the new transitions in the middle, i wonder why that doesn't seem to affect anything.
20100919 12:45:29< Ivanovic> if we do want this, then commiting to branches/1.8 *and* trunk/ would make sense
20100919 12:45:40< zookeeper> eleazzaar, yeah, you can just tweak the ~CS values there
20100919 12:46:00-!- EdB [~edb@79.89.103.70] has quit [Remote host closed the connection]
20100919 12:46:09< ilor> Ivanovic: I'd prefer to first commit it to a separate testing branch
20100919 12:46:32< Ivanovic> ilor: the server is only used on a "by hand" basis anyway
20100919 12:46:40< Ivanovic> so the code of this seperate part can directly reside in trunk
20100919 12:46:50< Ivanovic> no need for an additional branch at all
20100919 12:47:14< ilor> Ivanovic: the game code was changed and does not work with the old server in this local branch of mine
20100919 12:47:15< Ivanovic> currently the addon server is in the c++ code, so it is compiled by hand every now and then, just like the mp server is
20100919 12:47:36< Ivanovic> ilor: so at the start your addon server will not work with the game, okay
20100919 12:47:50< Ivanovic> but i don't see why an additional branch for the server would be required
20100919 12:48:25< ilor> Ivanovic: because we have one repo and the in-game changes need to be worked on and tested as well
20100919 12:49:32-!- EdB [~edb@70.103.89-79.rev.gaoland.net] has joined #wesnoth-dev
20100919 12:49:38< Ivanovic> then just do so in branches/
20100919 12:49:56< eleazzaar> zookeeper: you realized those transitions had the animations disabled previously?
20100919 12:50:09< ilor> which brings me to the question of how do I make git create a branch in svn :)
20100919 12:50:20< Ivanovic> might make sense to coordinate with billynux to have the network style in sync (he wrote the new asynchronous network library)
20100919 12:50:55< ilor> Ivanovic: yeah the new addon server code currently uses curl, which brings us to three separate network systems
20100919 12:51:29< Ivanovic> ugh
20100919 12:51:53< eleazzaar> Currently there are 4 shallow water colors -- once this transition is complete there will be 3
20100919 12:52:11< eleazzaar> so don't use the new ones yet, cause one of them will go away
20100919 12:53:18< AI0867> src/gettext.cpp:33: error: no match for ‘operator<<’ in ‘lg::wml_error << "Encountered empty translatable string\012" <-- does anyone have any idea what could cause this error? it's a new line, but ostream& operator<<(ostream&, const char*) should be defined whenever an std::stringstream exists, which is what lg::wml_error is.
20100919 12:53:52< ilor> so.. any git-svn gurus around?
20100919 12:54:08< ilor> while in the new merged branch, I tried a " git svn branch -m "new addon server branch" new_addon_server
20100919 12:54:23< ilor> which resulted in lots of cpu used by git svn and not much going on
20100919 12:54:58< Soliton> i'd just do it with svn.
20100919 12:55:23< silene> ilor: does your git svn already know about wesnoth branches?
20100919 12:55:42< AI0867> ilor: yeah, you can then do git-svn fetch --fetch-all, and use git branch remotebranchname local
20100919 12:55:53< ilor> silene: it's the git-svn tarball so I think yes
20100919 12:56:29< ilor> AI0867: so I should wait more?
20100919 12:56:34< AI0867> er, reversed
20100919 12:56:46< silene> ilor: then the way you did it should have worked (that's what i use)
20100919 12:56:48< CIA-35> ivanovic * r46592 /trunk/po/ (15 files in 15 dirs): updated Japanese translation
20100919 12:56:51< CIA-35> ivanovic * r46593 /branches/1.8/po/ (15 files in 15 dirs): updated Japanese translation
20100919 12:57:12< ilor> silene: maybe the issue is the long merge history (originally branched off 1.8.0)
20100919 12:57:50< AI0867> long branches over git-svn tend to take a lot of CPU, yes
20100919 12:58:13< ilor> AI0867: and this is my netbook, so I'll wait till I get to my desktop in the evening I guess
20100919 12:58:29< silene> ilor: that would be strange, git svn doesn't do anything with branches, it just sends a remote request
20100919 12:58:53< silene> (that is, mkdir + svn commit)
20100919 12:59:10< ilor> silene: well, thi git-svn I've shown hangs with lots of cpu 
20100919 13:00:30< Rhonda> shadowmaster: It's not really custom, I got the patch from the author - but sure.
20100919 13:00:48 * ilor is leaning towards the "use svn" option ;)
20100919 13:00:48-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: must... get... some... sleep... 7 am.... -.- '  ...Zzz...]
20100919 13:01:10< silene> ilor: what happens if you use the -n option in your line?
20100919 13:01:52< mordante> wesbot, seen alink
20100919 13:01:53< wesbot> mordante: The person with the nick alink last spoke 11h 57m ago.  7h 7m ago was here and on the channel #wesnoth with the message: Remote host closed the connection
20100919 13:01:53< ilor> silene: it eats cpu like before
20100919 13:02:18< silene> ilor: then you have definitely hit a bug in git svn
20100919 13:03:09< ilor> silene: well this is a slow cpu and several 1000s of commits in between... maybe not something it was designed for
20100919 13:03:48< Ivanovic> ilor: so, uhm, maybe just wait longer?
20100919 13:03:49< silene> ilor: yes, but as i have said, nothing should happen for you locally, so i don't think it's a cpu issue
20100919 13:06:19-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20100919 13:07:05< Ivanovic> ilor: have you considered exporting it as one huge patch, creating a branch using plain svn and applying the patch?
20100919 13:08:01< silene> Ivanovic: i don't think that's not needed, i'm sure git rebase will work fine once the branch has been created by a plain svn
20100919 13:08:08< Ivanovic> svn copy svn+ssh://ilor@svn.gna.org/svn/wesnoth/trunk svn+ssh://ilor@svn.gna.org/svn/wesnoth/branches/new_addon_server
20100919 13:08:33< Ivanovic> silene: but it was not created so far
20100919 13:09:33< silene> Ivanovic: i'm talking about exporting the patch and applying the patch, that's the unneeded part
20100919 13:11:30-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20100919 13:13:22< CIA-35> ilor * r46594 /branches/new_addon_server/: Create a branch for the new addon server
20100919 13:15:39-!- fendrin [~fabi@88-134-102-144-dynip.superkabel.de] has joined #wesnoth-dev
20100919 13:15:39-!- fendrin [~fabi@88-134-102-144-dynip.superkabel.de] has quit [Changing host]
20100919 13:15:39-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100919 13:18:04-!- timotei [~timotei@193.34.191.4] has joined #wesnoth-dev
20100919 13:18:04-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20100919 13:18:04-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100919 13:18:15< timotei> hey there ilor 
20100919 13:18:17< timotei> :)
20100919 13:18:31< timotei> do you still use git-svn on windows?
20100919 13:18:33< timotei> or moved over on linux?
20100919 13:18:53< ilor> timotei: linux
20100919 13:18:57< timotei> oh
20100919 13:19:01< timotei> ok:D
20100919 13:19:14< timotei> but you used git-svn on windows?
20100919 13:19:15< ilor> I'm currently getting warnings about warning: refname 'new_addon_server' is ambiguous.
20100919 13:19:33< ilor> timotei: yeah and I still do if I happen to be on windows
20100919 13:20:00< timotei> ilor: did you by changes get a 'checksum mismatch' type errors?
20100919 13:20:04< timotei> chance*
20100919 13:21:50< silene> ilor: did you happen to give the exact same name to your local branch and the remote branch, then the warnings are expected (and fortunate)
20100919 13:22:33-!- fabi [~fabi@88-134-102-144-dynip.superkabel.de] has joined #wesnoth-dev
20100919 13:22:33-!- fabi [~fabi@88-134-102-144-dynip.superkabel.de] has quit [Changing host]
20100919 13:22:33-!- fabi [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100919 13:22:59 * Soliton prefixes local tracking branches with L.
20100919 13:23:23< silene> i use "my"
20100919 13:23:41< ilor> silene: yeah, now that I think about it same thing happens with 1.x local branches so this is harmless atm
20100919 13:24:09< ilor> ok here goes...
20100919 13:24:40< CIA-35> ilor * r46595 /branches/new_addon_server/ (51 files in 13 dirs): (log message trimmed)
20100919 13:24:40< CIA-35> Merge the new addon server code, originally forked from 1.8.0, into new_addon_server.
20100919 13:24:40< CIA-35> This is currently stale code that compiles, starts, but needs testing and polishing, and likely new features, before it will be usable.
20100919 13:24:40< CIA-35> Conflicts:
20100919 13:24:40< CIA-35> CMakeLists.txt
20100919 13:24:41< CIA-35> SConstruct
20100919 13:24:41< CIA-35> data/tools/wesnoth/campaignserver_client.py
20100919 13:24:43-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 240 seconds]
20100919 13:26:05< ilor> thanks for the help, my svn-fu is weak since I've been only using git at work for the past half a year ;)
20100919 13:28:27< ilor> I've to admit this code looks far uglier now, after these few months, than it did back when we were finishing the class project
20100919 13:33:56< ilor> I'm off now, I'll be checking the logs and hope to do some cleanups in the evening
20100919 13:37:44< CIA-35> eleazar * r46596 /trunk/data/core/ (4 files in 2 dirs): refinedments to the auto-color water variations. Note this is probably a speed regression due to previously disabled animations being re-enabled.
20100919 13:42:46< Ivanovic> mordante: you saw what alink mentioned yesterday regarding some possible libsdl bugs?
20100919 13:43:48< mordante> Ivanovic, yes that's why I pinged him
20100919 13:44:46< mordante> at least one of the reasons
20100919 13:44:52< Ivanovic> :)
20100919 13:49:05< mordante> off for dinner
20100919 13:50:55< loonycyborg> mordante: Perhaps you could place some other widget between tip of the day and the main menu to prevent it from being stretched like that :P
20100919 13:51:58< loonycyborg> it == tip of the day
20100919 13:52:25< zookeeper> eleazzaar, yes, i realized; the animations are disabled inside the transition macro
20100919 13:52:56< ilor> mordante: will you be around this evening (in about 4hrs)? I have a gui2 issue in the new addon server code with timers that I'd like to run by you, but haven't got the time now
20100919 13:57:28-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100919 13:57:28-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100919 13:57:28-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100919 13:59:49-!- Gambit [~Gambit@pa-67-234-74-246.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100919 13:59:50-!- Gambit [~Gambit@pa-67-234-74-246.dhcp.embarqhsd.net] has quit [Changing host]
20100919 13:59:50-!- Gambit [~Gambit@unaffiliated/gambit] has joined #wesnoth-dev
20100919 13:59:52< CIA-35> silene * r46597 /trunk/m4/boost.m4: Avoided usage of forward incompatible test. (Fix for bug #16693.)
20100919 13:59:57< CIA-35> silene * r46598 /trunk/configure.ac: Avoided usage of GNU extension. (Fix for bug #16696.)
20100919 14:00:02< CIA-35> silene * r46599 /trunk/src/gettext.hpp: Fixed compilation with gettext 0.18. (Fix for bug #16694.)
20100919 14:01:19< ilor> we're still using autotools?
20100919 14:02:02< Ivanovic> ilor: silene maintains the autotools recipe
20100919 14:02:13< Ivanovic> so in fact we got 3 build systems that are usable on linux
20100919 14:02:43-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has joined #wesnoth-dev
20100919 14:03:03< ilor> I see. FYI, the new addon server branch is mostly cmake-oriented, scons should work, autotools not up-to-date
20100919 14:04:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20100919 14:05:12< CIA-35> silene * r46600 /branches/1.8/m4/boost.m4: 
20100919 14:05:12< CIA-35> Avoided usage of forward incompatible test. (Fix for bug #16693.)
20100919 14:05:12< CIA-35> Backported from trunk r46597.
20100919 14:05:12< CIA-35> silene * r46601 /branches/1.8/configure.ac: 
20100919 14:05:12< CIA-35> Avoided usage of GNU extension. (Fix for bug #16696.)
20100919 14:05:13< CIA-35> Backported from trunk r46598.
20100919 14:05:13< CIA-35> silene * r46602 /branches/1.8/src/gettext.hpp: 
20100919 14:05:14< CIA-35> Fixed compilation with gettext 0.18. (Fix for bug #16694.)
20100919 14:05:15< CIA-35> Backported from trunk r46599.
20100919 14:08:14< timotei> what's the m4 tingy?
20100919 14:08:18< timotei> make version 4?
20100919 14:08:57< loonycyborg> It's a macro processor.
20100919 14:09:29< loonycyborg> http://en.wikipedia.org/wiki/M4_%28computer_language%29
20100919 14:09:35< timotei> thanks
20100919 14:15:37-!- wesbot changed the topic of #wesnoth-dev to: 158 bugs, 297 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100919 14:16:31-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100919 14:16:35< zookeeper> eleazzaar, seems a bit odd how the medium shallow water is now the most saturated one
20100919 14:16:44< zookeeper> i'd expect it to rather be in the middle :p
20100919 14:17:28< zookeeper> (most saturated when not counting the tropical water...maybe your intention was it to be in between shallow and tropical, and not the medium one of the normal shallow waters)
20100919 14:17:32< Ivanovic> Soliton: is there some good way to reproduce the server crash issues? https://gna.org/bugs/index.php?16618
20100919 14:17:40< Ivanovic> since that was billynux last question
20100919 14:17:44< Ivanovic> or do they "just happen"?
20100919 14:21:45< CIA-35> zookeeper * r46603 /trunk/data/core/terrain.cfg: Made the shallow water terrains appear in order of saturation (from most saturated to least saturated) in the editor.
20100919 14:26:53-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 276 seconds]
20100919 14:32:21< Soliton> no idea how to reproduce. if he's really waiting for a nice recipe to reproduce we can just as well role the network changes back on the server side...
20100919 14:33:31< Soliton> i mean there are backtraces in the report that point to exact source lines which throw TODO notes...
20100919 14:34:41< Soliton> and the issue with pure virtual function calls is very likely some race condition where a simple reproduction isn't possible.
20100919 14:35:59< CIA-35> zookeeper * r46604 /trunk/data/campaigns/Sceptre_of_Fire/images/units/heroes/ (6 files): New-style shadows for Haldric II (by SkyOne).
20100919 14:37:56< CIA-35> zookeeper * r46605 /trunk/data/campaigns/Heir_To_The_Throne/images/units/ (6 files): New-style shadows for Asheviere (by SkyOne).
20100919 14:45:04-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100919 14:54:43-!- homunkoloss [~homunkolo@chello084115138094.1.graz.surfer.at] has joined #wesnoth-dev
20100919 15:11:21-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100919 15:12:55< alink> mordante: I studied a bit more resize update yesterday(when manually resizing with mouse grab). I found a possible source of problem and tried to code a workaround
20100919 15:13:24< alink> it helps but now I see that it's not the proper way to fix it
20100919 15:15:06-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100919 15:15:14< alink> (use some hardcoded delay/value, and but the the bug seems to depend of general speed of system. So probably impossible to tune it right)
20100919 15:17:50< alink> what I saw is that in handler.cpp, when handling DRAW_EVENT, we miss following SDL_VIDEORESIZE events. So the first resize triggers a redraw and we miss most of the next resize
20100919 15:18:20< alink> I am not sure why draw() eats/block these SDL events
20100919 15:19:01-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 252 seconds]
20100919 15:19:44< alink> my idea was to wait a little after the first resize before redrawing, that is, start to redraw only when there is no recent resize (wait here is a fraction of second)
20100919 15:20:51< alink> I even have a patch (as I said, probably not the good way, but maybe will give you a better idea)
20100919 15:22:57< alink> http://wesnoth.pastebin.com/9MjNJ9Xz
20100919 15:23:09< alink> (with some debug code)
20100919 15:24:12< alink> btw SDL video surface is still resized in event.cpp
20100919 15:25:32< silene> alink: why not simply remove all the extra VIDEORESIZE from the current queue and keep only the last one?
20100919 15:26:45< silene> (we already do that for input events)
20100919 15:27:30< alink> silene: that's sort of already what my patch does. It ignore size info of the event, and later query screen size (which was correctly updated by event.cpp)
20100919 15:28:18< alink> later = few frames later (more if user constantly resize)
20100919 15:28:52< alink> the problem is to identify the last one
20100919 15:29:24< alink> or better find why draw() eats resize events
20100919 15:29:30< silene> it's easy, it's the last one in the queue; perhaps you are not looking at the proper files? take a look at src/events.cpp
20100919 15:30:36< alink> as I said, what I saw here, is that during a busy draw(), they don't enter in the queue.
20100919 15:30:50< alink> which is weird
20100919 15:32:27< mordante> loonycyborg, something like that might be needed, but need to check how to do it :-P
20100919 15:32:48< mordante> ilor, probably yes, but maybe a bit later
20100919 15:33:36< alink> silene: the test was: disable draw() in handler.cpp(see above) and notice how we stopped missing resize events in event.cpp
20100919 15:33:59< mordante> alink, is it an SDL or Wesnoth bug?
20100919 15:35:06< alink> not sure now, in theory SDL should not lose events, but maybe it does if too busy. Or there is something affecting events in draw()
20100919 15:35:45< mordante> I think silene's idea is better, not sure how easy to implement
20100919 15:36:41< alink> well, if some events don't enter the queue, taking the last one will not help
20100919 15:37:35< mordante> agreed, that's why I asked whether it's an SDL or Wesnoth bug
20100919 15:37:50< mordante> I expect SDL since I also have seen this kind of behaviour with the old gui
20100919 15:38:15< alink> I need to go now, I will redo my test later, and post my debug code to illustrate the weird thing I mentioned (so we can also check if it's specific to my system)
20100919 15:38:19< alink> bbl
20100919 15:38:27< mordante> ok see you later
20100919 15:40:24< zookeeper> is there anymore a way to enable the normal undo and use that instead of the WB?
20100919 15:44:55< silene> zookeeper: undo has never been disabled, only delayed shroud
20100919 15:47:02< zookeeper> gah. right. silly me.
20100919 15:50:48< zookeeper> awesome idea, any takers? http://gna.org/bugs/?16729
20100919 15:53:16-!- Blueblaze2 [~Blueblaze@adsl-99-171-163-136.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100919 15:54:53< Ivanovic> my idea is "won't do"
20100919 15:55:15< Ivanovic> or "can already be done in WML"
20100919 15:59:47< CIA-35> ilor * r46606 /branches/new_addon_server/src/ (7 files in 2 dirs): Move src/addon_client.[ch]pp to src/addon/client.[ch]pp in spirit of the move of the other addon_ files
20100919 16:01:05< ilor> BTW, I'm open to suggestions as to where in the tree should thenew addon server live. Currently it's data/tools/addon_server.
20100919 16:02:48< Soliton> since it's rather an external tool and not something that works on our data/WML it should go into utils/.
20100919 16:05:28< ilor> thing is though, it uses the wml processing classes from data/utils/wesnoth
20100919 16:05:57< Soliton> s/utils/tools/?
20100919 16:06:33< Soliton> then i guess it's fine where it is.
20100919 16:07:11< Soliton> the addon manager is there too now that i think about it so it fits.
20100919 16:07:32< Soliton> well, no idea if we still need it.
20100919 16:07:47< silene> i would rather prefer that the whole data/utils move away (in plain utils/ for instance), since the game doesn't need it at all
20100919 16:08:51< ilor> Soliton: yes, I meant tools
20100919 16:09:02< ilor> internet's getting wonky, see you in a bit
20100919 16:09:59-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has quit [Ping timeout: 265 seconds]
20100919 16:24:03-!- dtiger [~dtiger@dynamic-vpdn-93-125-12-3.telecom.by] has quit [Remote host closed the connection]
20100919 16:27:36-!- Blueblaze [~Blueblaze@adsl-76-202-18-174.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100919 16:29:42-!- dtiger [~dtiger@dynamic-vpdn-93-125-12-3.telecom.by] has joined #wesnoth-dev
20100919 16:31:05-!- homunkoloss [~homunkolo@chello084115138094.1.graz.surfer.at] has left #wesnoth-dev []
20100919 16:32:08-!- Blueblaze2 [~Blueblaze@adsl-99-171-163-136.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
20100919 16:32:32-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100919 16:35:02-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100919 16:40:21-!- Elvish_Pillager [~eli@dhip-164.foss.residences.colby.edu] has joined #wesnoth-dev
20100919 16:56:16< eleazzaar> zookeeper: eleazzaar: Currently there are 4 shallow water colors -- once this transition is complete there will be 3
20100919 16:56:16< eleazzaar> [05:52am] eleazzaar: so don't use the new ones yet, cause one of them will go away
20100919 16:56:46-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Quit: moving to a tty]
20100919 16:57:24-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20100919 17:01:53< alink> back
20100919 17:03:13-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100919 17:03:45-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100919 17:07:30-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20100919 17:08:00-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20100919 17:08:00-!- esr [~chatzilla@wesnoth/developer/esr] has joined #wesnoth-dev
20100919 17:08:26-!- timotei21 [~timotei@188.24.6.167] has joined #wesnoth-dev
20100919 17:08:37< esr> shadowmaster: You were looking for me?
20100919 17:08:59< alink> mordante:here is my debug test http://wesnoth.pastebin.com/s92YyjZD
20100919 17:09:19< stikonas> is gna.org down at the moment?
20100919 17:09:29< alink> and I do quick resize grab with the window to stress test it
20100919 17:10:04< alink> btw, sometimes the window stay black (missed draw event ? but not studied that yet)
20100919 17:10:44-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20100919 17:11:23< timotei21> stikonas: it doesn't work for me either
20100919 17:11:42< alink> mordante: if it's a SDL bug, we can maybe found a workaround by changing some SDL calls since 1.8 titlescreen doesn't seem to have this bug
20100919 17:15:18-!- eleazzaar [~jbjerk@ppp-70-226-224-2.dsl.spfdil.ameritech.net] has quit [Ping timeout: 240 seconds]
20100919 17:16:39-!- eleazzaar [~jbjerk@ppp-70-226-224-2.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20100919 17:23:31-!- timotei-temp [~timotei@193.34.191.4] has joined #wesnoth-dev
20100919 17:26:12-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Ping timeout: 272 seconds]
20100919 17:26:26-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100919 17:27:43-!- timotei21 [~timotei@188.24.6.167] has quit [Ping timeout: 252 seconds]
20100919 17:29:41-!- fangism1 [~davidfang@gilbert.achronix.com] has joined #wesnoth-dev
20100919 17:30:40-!- timotei-temp [~timotei@193.34.191.4] has quit [Read error: Connection reset by peer]
20100919 17:31:48-!- fangism [~davidfang@gilbert.achronix.com] has quit [Ping timeout: 240 seconds]
20100919 17:33:21-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20100919 17:33:48-!- Blueblaze [~Blueblaze@adsl-76-202-18-174.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 272 seconds]
20100919 17:37:56-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has joined #wesnoth-dev
20100919 17:40:21< zookeeper> eleazzaar, right, i forgot that for a moment, but luckily my reordering shouldn't affect that ;)
20100919 17:42:34< eleazzaar> is GNA messed up?
20100919 17:42:37< eleazzaar> i can't commit
20100919 17:42:44< alink> yes it is
20100919 17:44:27< alink> mordante: draw_timer and timer_sdl_draw_event seems to do the same thing
20100919 17:46:13-!- Greywhind [~Greywhind@96.238.16.162] has quit [Quit: leaving]
20100919 17:48:21-!- joo|sleep is now known as joo
20100919 17:55:17< alink> gna.org is back
20100919 17:59:10< CIA-35> eleazar * r46607 /trunk/data/core/ (terrain-graphics.cfg terrain.cfg): Making the standard shallow water color-shifted from tropical. Removed extra water terrain.
20100919 18:07:34< fabi> eleazzaar: Can the same technique be used to define a even darker underground water terrain that is a mixture between cave and water? 
20100919 18:07:35-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has joined #wesnoth-dev
20100919 18:08:37< mordante> alink, I also have some rezise/redraw issues in 1.8 (and even further back)
20100919 18:08:52< alink> mordante: in titlescreen ?
20100919 18:08:56< mordante> resize*
20100919 18:09:15< mordante> alink, yes sometimes there the screen also doesn't get drawn at the new size
20100919 18:09:20< eleazzaar> fabi: it could make darker water, but not mix in cave
20100919 18:09:59< alink> mordante: btw I have a possible fix, at least some info, but I need a function to mark everything dirty
20100919 18:10:47-!- SpoOkyMagician is now known as SpoOkyMagician-
20100919 18:11:19< mordante> if you want everything dirty setting the window dirty should do the trick
20100919 18:11:26< alink> mordante: http://wesnoth.pastebin.com/KrDEZfcJ  <-- this mostly fix it, just sometimes something weird with the window manager
20100919 18:11:52< mordante> speaking about tricks r46584 breaks on small resolutions, where the logo is party drawn over the buttons
20100919 18:13:12< alink> mordante: I suppose, but as I mentioned, it crash before it could happen (even without r46584 ).
20100919 18:13:22< alink> It doesn't crash on yours ?
20100919 18:13:40< mordante> I know but r46584 has nothing to do with the crash
20100919 18:14:07< mordante> whether it crashes for me depends on the checkout... one the one with a fix it doesn't ;-)
20100919 18:14:23-!- timooootei [c122bf04@gateway/web/freenode/ip.193.34.191.4] has joined #wesnoth-dev
20100919 18:14:38< timooootei> Soliton: hey there
20100919 18:14:51< alink> I know that's why I said "Depending how that is fixed (allow it or prevent it), you may need to revert this"
20100919 18:15:18< timooootei> Soliton: I've downloaded (again) the last tarbal with git. Put the .git/config file (with my current commit user) in the extracted tarball
20100919 18:15:33-!- SpoOkyMagician- is now known as SpoOkyMagician
20100919 18:15:46< timooootei> Soliton: however git svn rebase doesn't work. As you said last time you need to fetch first something. 
20100919 18:15:50< timooootei> I did: git svn fetch
20100919 18:15:52< mordante> alink, I read that, but wanted to let you know upfront
20100919 18:15:56< timooootei> but that fetches from branches/1.8
20100919 18:16:10< alink> mordante: I am also not sure if logo under buttons will look nice, even if it's nice if GUI allows it
20100919 18:16:15< alink> mordante: ok
20100919 18:16:20< fabi> eleazzaar: Mixing wasn't meant art wise, but gameplay wise. This would solve the problem I have with flying units in underground scenarios that feature water. The minimum of the movement would count for flying and the maximum for non flying units. I think that can be coded with some filtering.
20100919 18:16:51< timooootei> any clue about this?
20100919 18:16:58< timooootei> mordante: ^
20100919 18:18:02< alink> mordante: feel free to revert it
20100919 18:18:20< eleazzaar> fabi: i'm not interested in making confusing terrains like that.  Our combo terrains are bad enough
20100919 18:19:02< mordante> alink, I like it better as the alternatives and only happens in rare cases
20100919 18:19:17< mordante> maybe we could scale down the logo on smaller screens instead
20100919 18:20:45< alink> mordante: btw I had an idea about logo speed for slow box, instead of fixed increment, specificy a start time and duration, and then update percentage as (current_time - start_time) / duration
20100919 18:21:22< alink> that way it will just be less smooth on slow system
20100919 18:21:25< mordante> timooootei, git svn fetch should fetch branches and the trunk
20100919 18:21:36-!- timooootei_ [c122bf04@gateway/web/freenode/ip.193.34.191.4] has joined #wesnoth-dev
20100919 18:21:41< mordante> timooootei_, git svn fetch should fetch branches and the trunk
20100919 18:21:56< timooootei> mordante: yep. but rebase didn't work 
20100919 18:22:06< timooootei> I mean, it doesn't work. It just stays...
20100919 18:22:11< timooootei> like it's waiting for something
20100919 18:22:25< mordante> timooootei, now or a while ago?
20100919 18:22:34-!- timooootei_ [c122bf04@gateway/web/freenode/ip.193.34.191.4] has quit [Client Quit]
20100919 18:22:42< timooootei> well... even now
20100919 18:22:42< mordante> while as in an hour or something
20100919 18:22:49< timooootei> not really so much
20100919 18:22:52< timooootei> but 4-5 mins?
20100919 18:22:53< timooootei> or so
20100919 18:22:55< alink> mordante: note that logo size problem is maybe also related to the bug I submitted about size depending of localized image
20100919 18:24:49< CIA-35> mordante * r46608 /trunk/ (changelog src/gui/widgets/window.cpp): 
20100919 18:24:49< CIA-35> Honour the minimum game window sizes in gui2.
20100919 18:24:49< CIA-35> Did some testing and didn't note regressions, but there might be some
20100919 18:24:49< CIA-35> issues still lurking, this area of interaction between gui1 and gui2 is
20100919 18:24:49< CIA-35> tricky. (Fixes bug #16724.)
20100919 18:24:51< mordante> alink, yeah making the steps bigger might fix the issue on slower boxes
20100919 18:25:03< mordante> and yes still need to look at the localized issues
20100919 18:25:05< timooootei> mordante: I remember I had some kind of issues like that before, but maybe I lost something
20100919 18:25:14< timooootei> mordante: what's the good workflow of using the tarbal?
20100919 18:25:26< fabi> eleazzaar: Confusion is a good point. I think I may code a workaround.
20100919 18:25:56< timooootei> fabi: hehe, I managed to hide the non-wesnoth wizards and some unnecessary menus, so people won't be confused :D
20100919 18:26:07< mordante> timooootei, can't remember exactly but IIRC I first do $ git reset --hard HEAD ; git svn rebase 
20100919 18:26:13< timooootei> hmm
20100919 18:26:19< timooootei> I did just git reset --hard
20100919 18:26:26< timooootei> let me check with HEAD
20100919 18:26:27< fabi> timooootei: I am really looking forward to the standalone version.
20100919 18:26:38< timooootei> fabi: hope will be up soon
20100919 18:29:39< timooootei> mordante: nope. the same result
20100919 18:33:05< timooootei> mordante: actually it seems git svn is using about 10-12% processor... but nothing else
20100919 18:33:15< mordante> timooootei, what does git svn fetch do, since that should work
20100919 18:33:32< mordante> 10-12% processor and how much IO?
20100919 18:34:02< timooootei> timo@sabayon /media/work/work/wesnoth-git $ git svn fetch Enter passphrase for key '/home/timo/.ssh/id_rsa':          M       src/gui/widgets/window.cpp         M       changelog r46608 = ded696a7eadd70111b7e6a089c24bf1c790020d9 (refs/remotes/trunk)
20100919 18:34:07< timooootei> and that's all
20100919 18:34:13< timooootei> mordante: I don't know how to read IO :D
20100919 18:34:53< timooootei> ok, found the IO thingy
20100919 18:34:55< timooootei> nothing
20100919 18:35:02< timooootei> it doesn't show any IO read/write
20100919 18:35:23< CIA-35> mordante * r46609 /trunk/src/gui/dialogs/title_screen.cpp: 
20100919 18:35:23< CIA-35> Fixed overlapping logo rendering in the title screen.
20100919 18:35:23< CIA-35> Reverts r46584.
20100919 18:36:07< mordante> timooootei, and no harddisk activity?
20100919 18:36:13< timooootei> there are 3 git related processes: 
20100919 18:36:32< timooootei> git svn rebase
20100919 18:36:37< timooootei> and git log ...
20100919 18:36:45< timooootei> the git log has from time to time IO read
20100919 18:36:52< timooootei> but that's all
20100919 18:36:53< mordante> alink, for me with patch http://wesnoth.pastebin.com/KrDEZfcJ the updates of the title screen look worse
20100919 18:36:59< mordante> more often an area that's black
20100919 18:37:35< mordante> also not sure why you add the pump in during the drawing of a grid
20100919 18:38:07< alink> that's because, here I miss events during the drawing
20100919 18:38:59< mordante> and the change in the handler.cpp?
20100919 18:39:21< alink> but the pump break the drawing too (that's why i needed the "set all dirty" too), so the "fix" may just be a side effect
20100919 18:39:35< alink> but both changes are needed (here)
20100919 18:40:42< alink> "but the pump break the drawing too" note that it's weird because DL_PumpEvents should not have such effect, I think
20100919 18:40:54< alink> *SDL_PumpEvents
20100919 18:41:32< alink> but anyway that's not a real fix, just more clues
20100919 18:41:48< mordante> hmm yeah playing a bit more I see the difference
20100919 18:42:00< mordante> which SDL version do you use?
20100919 18:42:09< alink> 1.2.13
20100919 18:43:13< alink> (1.2.13-2 of debian stable)
20100919 18:43:29< mordante> good saves me looking up my version ;-)
20100919 18:46:12< mordante> alink, the version with your patch seems also much more cpu intensive
20100919 18:46:37< timooootei> woot
20100919 18:46:39< timooootei> it works now:))
20100919 18:46:47< timooootei> mordante: it seems it wanted to wait :))
20100919 18:46:49< mordante> what did you change/do
20100919 18:46:56< mordante> ah ok
20100919 18:47:00< alink> mordante: probably since I always mark everything dirty, instead of just after resize
20100919 18:47:05< mordante> patience is a virtue ;-)
20100919 18:47:21< mordante> alink, probably
20100919 18:47:43< alink> I try with the window.set_dirty that you mentioned
20100919 18:48:01< mordante> well that will also force a full redraw
20100919 18:48:35< alink> yes but i can call it only when resizing (where full redraw is needed anyway)
20100919 18:48:42< mordante> ah ok
20100919 18:49:03< mordante> but I guess it's already done by the invalidate_layout call
20100919 18:50:15< alink> if it was, then my (true||..) change would be not needed and it is needed (here)
20100919 18:51:26< alink> it is also possible that some weird bad order in events triggers the draw before the videomode change
20100919 18:54:14< mordante> I can't see a differrence between the true or the !first version (except for CPU usage(
20100919 18:55:16< alink> here i have black window (with briefly visible titlescreen). That is when adding only SDL_PumpEvents
20100919 18:56:54< alink> updating svn after your last commit ...
20100919 18:57:03< mordante> I had a look at the SDL sources and in the new version there seem to be a maximum of 128 events in the queue
20100919 18:57:31< mordante> so it might be your change happens to put events in the queue that didn't make it otherwise...
20100919 18:57:45< alink> interesting, that's why I added the pump, I suspected some overload like that
20100919 18:57:50< alink> but 128 seems small
20100919 18:59:01< alink> if it's that we can pump more often, or find if something push too much events
20100919 19:00:33< silene> if the queue is only 128 and you pump without handling events, you will cause events to be lost
20100919 19:00:33< mordante> we could try to build SDL ourselves and make the queue large or at least add an indication on queue overflow
20100919 19:00:49< silene> you should not pump unless you are ready to handle events
20100919 19:01:55< mordante> what happens if more than 128 events are gathered and then you pump? I also assume you lose them
20100919 19:03:19< silene> yes, so you should pump and handle events
20100919 19:04:15< silene> for instance, you could merge mouse motion events, no need to keep 128 of them
20100919 19:05:18< CIA-35> timotei * r46610 /trunk/utils/java/org.wesnoth/plugin.xml: eclipse plugin: remove non-needed wizards
20100919 19:06:22< alink> I am not sure that I see crazy number of events here, maybe ~20 by sec (probably draw event). But maybe I test it wrong
20100919 19:08:30< alink> afk
20100919 19:08:39< mordante> me too afk
20100919 19:09:30-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has quit [Read error: Operation timed out]
20100919 19:11:40-!- Gallaecio [~quassel@249.243.165.83.dynamic.mundo-r.com] has joined #wesnoth-dev
20100919 19:27:07-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100919 19:37:20-!- Blueblaze [~Blueblaze@adsl-76-202-20-49.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100919 19:40:07-!- Greywhind [~Greywhind@harkness-49.resnet.brown.edu] has joined #wesnoth-dev
20100919 19:42:52-!- crimson_ipad [~crimsonip@24.114.232.32] has joined #wesnoth-dev
20100919 19:43:52-!- crimson_ipad [~crimsonip@24.114.232.32] has quit [Remote host closed the connection]
20100919 19:44:49-!- timooootei [c122bf04@gateway/web/freenode/ip.193.34.191.4] has quit [Quit: Page closed]
20100919 19:51:41-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100919 19:57:40-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100919 20:02:52-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100919 20:11:06-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Quit: janebot]
20100919 20:13:10-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20100919 20:15:38-!- wesbot changed the topic of #wesnoth-dev to: 159 bugs, 296 feature requests, 15 patches | logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://wesnoth.pastebin.com | http://imagebin.org
20100919 20:18:12-!- Octalot [~noct@cpc3-hitc1-0-0-cust867.9-2.cable.virginmedia.com] has quit [Remote host closed the connection]
20100919 20:24:55-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20100919 20:28:13-!- PeterPorty [~Pete@pc-253-249-120-200.cm.vtr.net] has joined #wesnoth-dev
20100919 20:30:08-!- mjs-de [~mjs-de@p3EE236D8.dip.t-dialin.net] has quit [Ping timeout: 245 seconds]
20100919 20:36:01-!- PeterPorty [~Pete@pc-253-249-120-200.cm.vtr.net] has quit [Quit: Saliendo]
20100919 20:57:44-!- EdB [~edb@70.103.89-79.rev.gaoland.net] has quit [Remote host closed the connection]
20100919 20:59:10-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100919 21:10:51-!- dtiger [~dtiger@dynamic-vpdn-93-125-12-3.telecom.by] has quit [Remote host closed the connection]
20100919 21:17:22-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 272 seconds]
20100919 21:29:48-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100919 21:30:18-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
20100919 21:34:56-!- knotwork_ [~markm@142.177.178.49] has joined #wesnoth-dev
20100919 21:38:31-!- knotwork [~markm@142.177.234.182] has quit [Ping timeout: 252 seconds]
20100919 21:47:39-!- Vetinari [~lukjad@unaffiliated/lukjad] has quit [Remote host closed the connection]
20100919 21:47:45-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100919 21:48:29-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100919 21:50:22-!- Vetinari [~lukjad@unaffiliated/lukjad] has joined #wesnoth-dev
20100919 21:52:33-!- anonymissimus [~chatzilla@HSI-KBW-095-208-007-247.hsi5.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20100919 22:04:24-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100919 22:06:46-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100919 22:16:45< mordante> alink, I looked a bit further at your patch and only the SDL_PumpEvents part seems to work better for me
20100919 22:16:55< mordante> still feels like black magic however
20100919 22:17:17< mordante> I only get 1 resize event in the queue and most of the time less then 50 events in the queue
20100919 22:17:22-!- Mickcy [~mickcy@ip13-170-173-82.adsl2.static.versatel.nl] has left #wesnoth-dev []
20100919 22:17:25< mordante> so we don't lose events
20100919 22:17:43< mordante> so I think I'll commit with some comment
20100919 22:17:52< mordante> what do you think?
20100919 22:18:30-!- Bocom [~Bocom@c-65cee255.013-31-6b736412.cust.bredbandsbolaget.se] has quit [Quit: Lämnar]
20100919 22:19:05-!- mjs-de [~mjs-de@p3EE236D8.dip.t-dialin.net] has joined #wesnoth-dev
20100919 22:27:59< alink> mordante: a bit busy, but I quickly recheck
20100919 22:28:27< mordante> ok
20100919 22:30:29< alink> mmh weird, now it works fine with only the SDL_PumpEvents.
20100919 22:31:42< alink> but without it, I still have often broken resize, so it definitively helps here
20100919 22:32:02< mordante> here too it seems better, but still a bit weird
20100919 22:33:15-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100919 22:33:29< alink> with it, all is ok
20100919 22:33:57< mordante> ok then I commit it like that
20100919 22:34:12< alink> I wonder why I needed the redraw before, maybe my debug code
20100919 22:36:10< alink> mordante: btw I picked the place without knowing the code, so don't assume that it was carefully chosen, maybe there is a better one
20100919 22:36:23< mordante> no idea but the entire thing is a bit black magic
20100919 22:36:51< mordante> I also added the call in the pump after handling all events but it had no effect there..
20100919 22:36:55< alink> agreed :-/
20100919 22:38:23< mordante> I guess for some reason only one event can be retrieved from X11 not sure
20100919 22:38:47< mordante> but the issue seems old since the old gui has the same problem
20100919 22:38:56< alink> no, i can get all of them when disabling draw()
20100919 22:40:44< CIA-35> mordante * r46611 /trunk/ (changelog src/gui/widgets/grid.cpp): 
20100919 22:40:44< CIA-35> Fixed resize events being lost.
20100919 22:40:44< CIA-35> The change is a bit black magic, but seems to work. Based on a patch
20100919 22:40:44< CIA-35> by alink.
20100919 22:41:16< alink> btw I use KDE, if relevant
20100919 22:41:27< mordante> no idea I use fvwm
20100919 22:41:46< mordante> what do you mean with disabling draw?
20100919 22:41:48< alink> also (un)maximize always worked fine
20100919 22:42:14< alink> in handler.cpp   commenting off the draw() few lines above the resize code
20100919 22:43:04< alink> as I tried to explain here  http://wesnoth.pastebin.com/s92YyjZD
20100919 22:44:06-!- teaser [~tsr@h-37-9.A254.priv.bahnhof.se] has quit [Quit: Saliendo]
20100919 22:44:15< mordante> but is the screen then drawn?
20100919 22:44:37< alink> no :-)  
20100919 22:44:57< alink> but the SDL surface is resized and ready for draw
20100919 22:45:17< alink> by the other codepath in event.cpp
20100919 22:45:56-!- stikonas [~and@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20100919 22:46:17< mordante> yeah and I get many more resize events (eg smaller granuality)
20100919 22:46:28< mordante> almost think events get lost while blitting...
20100919 22:47:02< alink> me too and accurate (I make KDE display window geometry and compare with debug output)
20100919 22:48:07< alink> when broken the last event is rarely accurate, even if a few pixel error is not very visible in titlescreen
20100919 22:48:28< mordante> I leave it as is for now seems to be an SDL oddity
20100919 22:48:37< mordante> maybe 1.2.14 fixes it...
20100919 22:49:44< alink> ok, and not very important after all
20100919 22:49:55< alink> but annoying
20100919 22:51:18< alink> maybe more debug code will help. Something like quit a dump event queue on output, and info like that
20100919 22:52:21< alink> but not today, afk
20100919 22:52:48< mordante> maybe might help, but not sure
20100919 22:52:53< mordante> see you later
20100919 22:52:56< mordante> I'm off now
20100919 22:54:08-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20100919 22:54:24-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100919 22:54:30-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100919 22:55:58-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has joined #wesnoth-dev
20100919 22:55:58-!- stikonas [~and@ctv-213-164-99-190.vinita.lt] has quit [Changing host]
20100919 22:55:58-!- stikonas [~and@wesnoth/translator/stikonas] has joined #wesnoth-dev
20100919 22:55:59< eleazzaar> zookeeper: it's rather clunky and error prone to put in the same ~CS modification in 3 places.  Would it be possible to define it once somewhere?
20100919 22:57:23< zookeeper> eleazzaar, well, sure you can make a macro for it
20100919 22:57:50< boucman> eleazzaar: i'll see what I can do for bridge vs forest, probably fixable easily by using a different flag here...
20100919 22:57:51-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100914125854]]
20100919 22:57:55< zookeeper> i'm not sure where would be the best place, i guess you could just put it somewhere near the top of terrain-graphics.cfg for now
20100919 22:58:33< boucman> zookeeper: i don't like the way you did your water macro, but I like the idea, I'll see if I can rewrite it differently (using a new BUILDER macro probably)
20100919 22:58:55< boucman> zookeeper: you could do it in one place by defining a new builder :P
20100919 23:00:48< zookeeper> boucman, i knew you wouldn't like it :P
20100919 23:00:54< boucman> hehe
20100919 23:01:06< boucman> we coder introduce ugly graphics to push artist to replace them
20100919 23:01:13< zookeeper> boucman, i tried to do it with a custom builder already, but it would have been too messy so i just did it the simple way
20100919 23:01:14< boucman> are you using the same trick on me ? ;)
20100919 23:01:38-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Read error: Connection reset by peer]
20100919 23:01:47< boucman> k... I'm a bit it's that complicated to do with a builder, but i'll have a look soon (hopefully)
20100919 23:02:23< zookeeper> you can't do it with a custom builder only, you'd have to duplicate or redefine all the macros in builder.cfg which would get pretty ridiculously verbose
20100919 23:02:28< boucman> i'm a bit suprised
20100919 23:02:43< zookeeper> maybe there's some trick i missed, maybe not
20100919 23:03:26< boucman> zookeeper: I do it with lots of macros so I can easily do all sorts of variations, but if we do a special builder for water, (which would make sense here) we don't need that many macros I guess
20100919 23:09:02< eleazzaar> has someone compared the performance of a map like BoP before and after?
20100919 23:09:30< CIA-35> eleazar * r46612 /trunk/data/core/terrain-graphics.cfg: refinedments to the auto-colored water.
20100919 23:09:52-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100919 23:10:33< zookeeper> boucman, anyway, if you rewrite the water macros then please don't hide it behind more layers of macros unless there's a really good reason for it
20100919 23:10:45-!- johndh [84aa3333@gateway/web/freenode/ip.132.170.51.51] has joined #wesnoth-dev
20100919 23:10:55< boucman> zookeeper: not sure what you mean heare
20100919 23:10:56< boucman> here
20100919 23:11:14< zookeeper> i'd really like to make the whole system more comprehensible and i've already rewritten (not committed) a couple of macros which would utterly simplify forests, base tiles and most of castles
20100919 23:11:59< boucman> i'm not sure how you can simplify base tile more than they currently are... you just use the COMPLETE macro, and everything is done for you
20100919 23:12:06< boucman> forest should be more or less the same
20100919 23:12:12< boucman> I havn't worked on castles yet
20100919 23:12:17< johndh> esr: Might I bug you for a moment?
20100919 23:12:46< esr> johndh: I saw you were looking for me.  What's up?
20100919 23:12:48< zookeeper> boucman, and the COMPLETE macro calls a macro calls a macro calls a macro and it takes me forever to really find out what it actually does
20100919 23:13:02< johndh> You're in charge of writing-related things, right?
20100919 23:13:10< boucman> zookeeper: the easy way to know what it does is to look at the wiki documentation
20100919 23:13:12< boucman> :P
20100919 23:13:28< esr> johndh: Nore or less, yess.
20100919 23:13:36< boucman> but yes, it's a macro that is supposed to be simple to use, not to understand... it's hard to do both
20100919 23:14:11< johndh> I noticed when playing that the in-game help doesn't have descriptions for some of the races, so I went about writing some
20100919 23:14:52< johndh> and I've been polishing them up to try and get them commit-worthy, but thusfar I've gotten very little dev attention
20100919 23:15:00< esr> Hm?  I thought we had descriptions for all of them.
20100919 23:15:04< johndh> so I'm not really sure if I'm on the right track
20100919 23:15:25< johndh> Mermen, nagas, and a couple others are blank for me.
20100919 23:15:32< johndh> or are they in 1.9?
20100919 23:15:53< esr> johndh: Looking...
20100919 23:17:04-!- Gallaecio [~quassel@249.243.165.83.dynamic.mundo-r.com] has quit [Ping timeout: 272 seconds]
20100919 23:18:07< esr> johndh: This may take a few minutes, I have to recompile.
20100919 23:18:16< johndh> Okay
20100919 23:18:25-!- MeccaGod [~majs@h14n4fls31o279.telia.com] has quit []
20100919 23:19:01< johndh> For some of the minor races, like bats and mechanical, I just wrote a couple sentences to have *something* there so it doesn't look like they were overlooked, but for nagas and mermen I've written quite a bit.
20100919 23:21:39< CIA-35> eleazar * r46613 /trunk/data/core/terrain-graphics.cfg: adds important missing comma to previous commit.
20100919 23:22:51< Ivanovic> mordante, alink: i think that you really should test your stuff using libsdl 1.2.14
20100919 23:23:01< Ivanovic> it seems to do a hell of a lot of things at least slightly different
20100919 23:23:19< Ivanovic> (eg the ~millions of missed down events when clicking
20100919 23:25:18-!- Gallaecio [~quassel@225.189.60.213.dynamic.mundo-r.com] has joined #wesnoth-dev
20100919 23:45:49-!- Upth [ogmar@adsl-75-26-165-190.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100919 23:48:52-!- Upthorn [ogmar@adsl-75-26-163-0.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 252 seconds]
20100919 23:51:49< CIA-35> soliton * r46614 /trunk/src/about.cpp: fixed doubling of non-campaign about entries
20100919 23:51:57-!- Upthorn [ogmar@adsl-75-26-194-192.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100919 23:52:55-!- SpoOkyMagician [~chatzilla@74-138-202-211.dhcp.insightbb.com] has quit [Quit: meh...]
20100919 23:53:55-!- Upth [ogmar@adsl-75-26-165-190.dsl.scrm01.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100919 23:56:26-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Quit: Leaving.]
20100919 23:57:53< zookeeper> esr, i haven't looked at the stuff johnd and co have been doing in detail, but i've been glancing at it every now and then and it looks mostly pretty good
20100919 23:58:10< esr> Noted
20100919 23:58:31< zookeeper> i guess i'll have another look now...
--- Log closed Mon Sep 20 00:00:00 2010