--- Log opened Mon Jul 19 00:00:26 2010
20100719 00:01:08< CIA-87> ivanovic * r44323 /trunk/utils/pofix.py: remove conversion rule that does not work since several translations have the valid string "Epilog"
20100719 00:03:15< CIA-87> ivanovic * r44324 /trunk/utils/pofix.py: removed one line too much...
20100719 00:18:36< Ivanovic> esr: is there a way to add some kind of paralization to utils/pofix.py?
20100719 00:19:02< esr> "paralization"?
20100719 00:19:04< Ivanovic> esr: that is: when several files are used and there are several cpu cores, it would be really great if those could somehow be used automated
20100719 00:19:14< Espreon> Indeed...
20100719 00:19:14< esr> Oh.
20100719 00:19:25< Espreon> Poor Espreon here has to run pofix all the time...
20100719 00:19:36< Ivanovic> esr: eg i do call pofix this way: cd po/wesnoth/ && ../../utils/pofix.py *po wesnoth.pot
20100719 00:20:04< Ivanovic> so there are about 40 files per textdomain that are to be processed, but only one of my 4 cores is used
20100719 00:20:32< esr> I don't know. The underlying question is probably whether Python's threading library can exploit multicore CPUs. If so, I could probably rewite it to spawn worker threads.
20100719 00:20:35-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100719 00:20:53< Ivanovic> some plain --jobs parameter would already help a lot...
20100719 00:21:12< Ivanovic> so that i can supply it with the amount of worker threads i want
20100719 00:21:34< esr> Not unless I thread it.  There's problem with what Pythonistas call the GIl, the Great Interpreter Lock.
20100719 00:21:55-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100719 00:22:17< Ivanovic> since it is supposed to work on a per file basis it should be possible to thread it just based on the files
20100719 00:22:28< Ivanovic> sure, if you only supply one file, only one thread can be used
20100719 00:22:31< loonycyborg> esr: There's some new python threading api that is based on processes and so isn't affected by GIL.
20100719 00:22:47< Ivanovic> but with more files, like the 52 currently in trunk, it would be nice to be able to use more cores
20100719 00:23:36< loonycyborg> http://docs.python.org/library/multiprocessing.html
20100719 00:24:29< esr> loonycyborg: Good point.  That's probably be the right thing here.
20100719 00:24:32< CIA-87> ivanovic * r44325 /trunk/po/ (322 files in 9 dirs): ran utils/pofix on all po/wesnoth*/*po* files
20100719 00:30:30-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100719 00:31:10-!- eleazar [~eleazar@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has quit [Remote host closed the connection]
20100719 00:42:29-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100719 00:46:40-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Remote host closed the connection]
20100719 00:49:57-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20100719 00:58:12-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 245 seconds]
20100719 00:58:24-!- fendrin [~fabi@88-134-103-91-dynip.superkabel.de] has joined #wesnoth-dev
20100719 00:58:24-!- fendrin [~fabi@88-134-103-91-dynip.superkabel.de] has quit [Changing host]
20100719 00:58:24-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20100719 01:06:58-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20100719 01:09:52-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100719 01:10:26-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100719 01:11:13-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 258 seconds]
20100719 01:13:31< CIA-87> ivanovic * r44326 /trunk/data/campaigns/Sceptre_of_Fire/scenarios/6_Towards_the_Caves.cfg: a conversion added in pofix.py and missing in the source file...
20100719 01:14:23< CIA-87> ivanovic * r44327 /trunk/ (870 files in 49 dirs): 
20100719 01:14:23< CIA-87> pot-update
20100719 01:14:23< CIA-87> regenerated doc files (removed too incomplete (<80% completed) manuals!)
20100719 01:14:23< CIA-87> ivanovic * r44328 /branches/1.8/po/ (wesnoth/vi.po wesnoth-aoi/vi.po wesnoth-dm/vi.po): updated Vietnamese translation
20100719 01:14:47< CIA-87> ivanovic * r44329 /trunk/po/ (17 files in 17 dirs): updated Vietnamese translation
20100719 01:14:55< Espreon> Sigh.......
20100719 01:21:54-!- Yoyobuae [~gerald@182.Red-213-98-251.dynamicIP.rima-tde.net] has quit [Ping timeout: 260 seconds]
20100719 01:28:16-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100719 01:37:16-!- alink [~alink@wesnoth/developer/alink] has joined #wesnoth-dev
20100719 01:49:54-!- phlaem [~a@e178091174.adsl.alicedsl.de] has quit [Quit: Leaving]
20100719 02:14:10-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100719 02:17:44-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: Zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz]
20100719 02:18:29-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20100719 02:40:33< CIA-87> alink * r44330 /trunk/src/ (5 files): 
20100719 02:40:33< CIA-87> New imagepath function ~L(lightmap.png) to do ToD Light color-shift per pixel.
20100719 02:40:33< CIA-87> Possibly temporary, since added for the prototype of local ToD light transitions.
20100719 02:40:33< CIA-87> But I suppose it could be used for some special effect on units.
20100719 02:40:35< CIA-87> alink * r44331 /trunk/src/display.cpp: 
20100719 02:40:35< CIA-87> Improve local ToD prototype by adding light transitions.
20100719 02:40:35< CIA-87> Code don't even try to be optimized (even if it seems to run ok)
20100719 02:40:35< CIA-87> So, still disabled by game_config key 'local_light'=no.
20100719 02:40:51< alink> eleazar: done ^
20100719 02:41:15< alink> old :http://img696.imageshack.us/img696/1399/todlight1.png
20100719 02:41:29< alink> local ToD : http://img193.imageshack.us/img193/7905/localtod.png
20100719 02:41:50< alink> and now with light transition : http://img839.imageshack.us/img839/3309/todlight2.png
20100719 02:42:14< alink> light transition quickly Gimped by me feel free to tweak them
20100719 02:42:43< alink> I tried to not make them overlap but maybe not required anymore
20100719 02:43:39< alink> but I was glad that it looked good enough at the first try :-)
20100719 02:44:05-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Remote host closed the connection]
20100719 02:44:29< alink> and I love/hate when you don't see anything special in the final result, because you precisely fixed what was weird :-/
20100719 02:46:09< alink> and for coding, I discovered the power of ~BLIT(), it almost like a new rendering engine in one string. Very useful to try dirty prototype just by changing a string somewhere
20100719 02:48:40< alink> in fact, in theory, we could replace the whole engine just by assigning a string to each hex like grass.png~BLIT(transition.png)~CS(tod.rgb)~BLIT(grid.png)~BLIT(unit.png~TC(red)~BLIT(ellipse.png)) etc..
20100719 02:50:15< alink> or create terrain variant like grass with colored flowers : grass.png~BLIT(flower.png~CS(100,200,150), 30,40)~BLIT(flower.png~CS(100,200,150), 60,35) etc...
20100719 02:51:02-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20100719 02:53:22-!- esr [~chatzilla@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20100719 02:53:22-!- esr [~chatzilla@wesnoth/developer/esr] has joined #wesnoth-dev
20100719 02:53:58< alink> ~IF_PIXEL_OPAQUE(x, y, ~MOD1, ~MOD2)  <:o)
20100719 02:54:54< alink> and you can compute with image and string
20100719 02:57:00-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 02:57:13< happygrue> is bruno wolff on IRC at all?
20100719 02:57:45< alink> maybe ask wesbot
20100719 02:58:08< alink> if you know his IRC nick ;p
20100719 02:59:54< happygrue> I do not
20100719 03:00:02< happygrue> I can't find any alias for him at all
20100719 03:00:35< alink> he used brunowolff at [Mon May 28 2007] [19:23:09]
20100719 03:00:48< happygrue> wesbot: seen brunowolff
20100719 03:00:48< wesbot> happygrue: Sorry, I don't know of brunowolff.
20100719 03:00:55< happygrue> wesbot: seen bruno*
20100719 03:01:08-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 03:01:27< happygrue> ah, well thanks.
20100719 03:01:44-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 03:05:10< alink> make wesbot supporting wildcard in 'seen' seems a bad idea
20100719 03:06:45-!- jbjerk [~jbjerk@ppp-70-226-197-86.dsl.spfdil.ameritech.net] has joined #wesnoth-dev
20100719 03:06:54-!- jbjerk is now known as eleazzaar
20100719 03:10:42< alink> eleazzaar: just committed and posted light transitions first prototype
20100719 03:11:17< eleazzaar> cool. looking at it now...
20100719 03:13:27< eleazzaar> how bad is the performance hit?
20100719 03:13:38< eleazzaar> knowing that it's not optimized yet
20100719 03:14:38< alink> the cpu post processing cost is probably ok and too small for my box to be perceived 
20100719 03:15:13< alink> it's mainly a memory usage problem, but i should create a complex map alternating various ToD areas to test it
20100719 03:15:47< alink> and once cached all is ok (scrolling etc..)
20100719 03:15:56-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 03:16:21< eleazzaar> where's the mask files?
20100719 03:16:42-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 03:16:47 * eleazzaar is excited
20100719 03:17:00< alink> in terrain/light-?.png
20100719 03:17:47< alink> nothing definitive, but it uses 128 as 0 to allow light addition and substraction
20100719 03:18:33< eleazzaar> are you sure you commited those?
20100719 03:18:52< alink> oh :-/
20100719 03:19:02< eleazzaar> just one file in your last commit
20100719 03:20:14< CIA-87> alink * r44332 /trunk/data/core/images/terrain/ (7 files): forget images files in previous commit
20100719 03:20:15< alink> eleazzaar: done, thanks :-)
20100719 03:20:31< eleazzaar> ;D
20100719 03:24:02< eleazzaar> so lighter/brighter terrain "overlaps" onto darker terrain?
20100719 03:25:26< alink> yes, the ToD of each hex overlap a bit on adjacent
20100719 03:25:36< alink> but it's blended
20100719 03:26:01< alink> the thickness and alpha of light-?.png allows to control it
20100719 03:26:33< eleazzaar> Ok, i understand how it's working
20100719 03:26:35< alink> but I believe that the separation line should use 50% alpha to allow seamless matche
20100719 03:26:56< eleazzaar> yes, like my fancy grass transition
20100719 03:28:17< alink> maybe but in this case each light transition is adjacent to another different one
20100719 03:28:39< alink> well maybe the fancy grass do that too
20100719 03:28:58< eleazzaar> yeah, i was refering to a double sided grass-to-grass transition
20100719 03:29:16< eleazzaar> not that you could read that from my mind ;)
20100719 03:29:31< alink> ok then it's probably the same with more alpha
20100719 03:29:51< eleazzaar> this is more naturalistic than i had hoped we would be able to do
20100719 03:31:26< eleazzaar> alink: this will understand additional normal transitions like light-n-ne.png ?
20100719 03:31:31-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 03:31:33-!- King_Elendil_ [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 03:31:33< alink> yes I was not sure how it would look, but that's seems not bad for a first try
20100719 03:31:57< alink> eleazzaar: mmh no and not sure yet if we can allow it
20100719 03:33:13< alink> it's a bit complex since light-n-ne.png may be adjacent to 2 different ToD
20100719 03:33:58< alink> to be complete it creates a lightmap from the opaque light.png colored to the tod of the hex and blit all half-transparent transitions light-?.png colored using tod from adjacent hexes
20100719 03:34:14-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20100719 03:36:54< alink> mmh yeah supporting light-n-ne.png may be useful for nice corner
20100719 03:36:58< eleazzaar> so it's a giant overlay?
20100719 03:37:14< alink> no it uses a "lightmap"
20100719 03:37:37< eleazzaar> is that an SDL function?
20100719 03:37:51< alink> no a rendering concept, I ll explain
20100719 03:38:32< alink> it creates a image with only color light pixels (so using light.png and light-?.png with ToD coloring)
20100719 03:39:11< alink> then to color-shift the terrain it look each pixel of the lightmap and use that RGB to colorshift the pixel of the terrain image
20100719 03:39:44< eleazzaar> gotcha
20100719 03:40:14< eleazzaar> is this how it was done before?  with a lightmap only one without any details?
20100719 03:41:22< alink> no, because before the RGB used was the same for the whole hex, so it was just one RGB value not a whole 72x72 image
20100719 03:41:48< alink> it's probably slower
20100719 03:42:30< eleazzaar> it's also cooler
20100719 03:42:31< alink> but it's not very different from what we already do for the hex mask cutting, where we use the alpha instead of the RGB
20100719 03:42:50< alink> in fact i copied and adapted the mask function
20100719 03:45:42< eleazzaar> 1.10 is gonna look extremely better!
20100719 03:45:46< alink> and the new imagepath function ~L(lightmap.png) could be used for light effect
20100719 03:45:54-!- King_Elendil_ [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 03:45:58< Espreon> eleazzaar: Hell yeah!
20100719 03:46:18< eleazzaar> light effects as in... MoL?
20100719 03:46:22-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 03:46:43< Espreon> Terrain improvements, good typography, this... and that'll only be a little taste...
20100719 03:47:10< alink> eleazzaar: my only fear now is to code this cleanly, which will be a lot of work, mainly because underlying main bases (ToD areas and image cache) are in a bad shape and I obviously need them more now
20100719 03:47:24< eleazzaar> Espreon: who is working on typography?
20100719 03:47:32< Espreon> I am!
20100719 03:47:56< eleazzaar> Espreon: cool
20100719 03:48:03< alink> eleazzaar: no effect on static images like units or terrain images
20100719 03:48:10< Espreon> eleazzaar: Why thank you.
20100719 03:48:11< alink> *no, effect
20100719 03:48:31< alink> eleazzaar: but I am less pessimistic about MoL
20100719 03:48:58< eleazzaar> so like the cave "beam of light"?
20100719 03:49:57< alink> yeah any modification of image, but with color shift instead of overlay
20100719 03:50:01< alink> shadows too
20100719 03:50:16< Espreon> Let's not forget about deluminates...
20100719 03:50:40< alink> and for the MoL moves, I realized that even static light like ToD, are not really static, they change at each turn.
20100719 03:51:03< alink> So depending of the caching strategy, few MoL moves may be acceptable too
20100719 03:54:11< eleazzaar> MoL is IMHO less important than getting illuminated terrain (or darkened) like lava and cave beam to work right
20100719 03:54:52-!- ancestral [~ancestral@97-116-105-226.mpls.qwest.net] has joined #wesnoth-dev
20100719 03:54:58< eleazzaar> but what you've done now is a really big improvement as it is
20100719 03:55:25< alink> yeah but they have the same WML problem as MoL, they just define a lawful bonus, no color info
20100719 03:55:41< alink> eleazzaar: don't forget that it's just an experimental prototype
20100719 03:56:14< eleazzaar> could we make the terrain add to a ToD area instead?
20100719 03:57:18< alink> we could add ToD area in scenario, but add ToD info into terrain seems weird
20100719 03:58:19< alink> but i didn't looked that code yet
20100719 03:59:36< eleazzaar> i just have the general perception that there are too many different  ways currently of effecting the game concept of light
20100719 04:00:42< alink> I believe it's because you associate too much lawful bonus to light. The engine currently allow to give such bonus without light thing
20100719 04:00:58< alink> except the overlay in sidebar
20100719 04:01:07-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 04:01:18-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 04:02:26< alink> for WML, MoL could be considered to just have a nice staff boosting morale of adjacent lawful units
20100719 04:03:04< eleazzaar> well, a bonus could be given without light
20100719 04:03:24< eleazzaar> but IMHO light should aways give the lawful bonus
20100719 04:03:39< eleazzaar> or in lesser amounts remove the chaotic bonus
20100719 04:04:37 * eleazzaar is testing
20100719 04:04:52< alink> we could add a new key for light, and always associate light with lawful bonus in mainline, but still allow UMC to separate them
20100719 04:05:36< alink> to have illuminated terrain without bonus (like lava maybe), and terrain with bonus but not illuminated ('holy' area ?)
20100719 04:05:48< eleazzaar> UMC could use a different schedual.cfg
20100719 04:06:33< eleazzaar> i think the intention of illuminated lava is to correspond to dawn/dusk
20100719 04:06:40< eleazzaar> in a cave
20100719 04:06:59< eleazzaar> of course that shouldn't override daylight if it is outside
20100719 04:07:11< eleazzaar> i'm not sure how it actually works
20100719 04:08:35< alink> it's complicated and things are possibly weird when you start to use non-standard bonus (other than 25%)
20100719 04:13:14< alink> eleazzaar: the lightmap used in my previous screenshots (test-scenario): http://img828.imageshack.us/img828/4668/lightmap.png
20100719 04:13:38< eleazzaar> kinda trippy
20100719 04:13:43< alink> I didn't realized that it created these little ellipses
20100719 04:14:17< alink> maybe try to make transitions overlap a little to see if it helps for the corners
20100719 04:14:27< alink> not sure how they will add up
20100719 04:14:36< CIA-87> espreon * r44333 /trunk/po/wesnoth-ei/ (7 files): Reverted the harmful effects of r43685.
20100719 04:14:53< eleazzaar> you would get different "artifacts" then
20100719 04:15:06< alink> once you start to blit several image, with the alpha blending, order matters
20100719 04:15:48-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 04:16:24-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 04:19:56< alink> Perhaps I should try to this light-n-ne.png. It's somehow similar to fog and shroud. I feared that it could multiply the number of possible lightmap but if light-n-ne always replace light-n + light-ne, then it should be the same
20100719 04:20:16< eleazzaar> yeah that's how it works
20100719 04:21:57< alink> well in terrain rules, but it this case, I will need to auto-generate them and sometime split n-ne-se-sw or things like that
20100719 04:22:34< alink> but yeah that seems doable
20100719 04:22:40< eleazzaar> why would you need to split?
20100719 04:23:38< alink> I assumed that I wouldn't have a image file for each case, that's what fog do I think
20100719 04:24:18< eleazzaar> a) i can make all combinations if neccesary
20100719 04:24:56< eleazzaar> b) if N+NE+SE doesnt' exist use N+NE  and SE like terrain does
20100719 04:25:54< alink> b) is what I call a split 
20100719 04:26:18< alink> anyway I should check that after having realistic caching for lightmap
20100719 04:26:51< eleazzaar> iit's not urgent-- 6 trans look reasonably good
20100719 04:28:18< alink> maybe try to make them a bit more fuzzy (less round) to hide the pattern near corner
20100719 04:29:09< alink> if you can ever see it with real terrain images. Maybe I keep my lightmap view in mind too much
20100719 04:30:44-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 04:31:09-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 04:31:47-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Quit: Leaving]
20100719 04:35:01-!- Ivanovic_ [~ivanovic@dtmd-4db2a6db.pool.mediaWays.net] has joined #wesnoth-dev
20100719 04:35:34-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Client Quit]
20100719 04:38:46-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 276 seconds]
20100719 04:38:59-!- Ivanovic_ is now known as Ivanovic
20100719 04:42:07-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has joined #wesnoth-dev
20100719 04:43:56< CIA-87> espreon * r44334 /trunk/data/core/images/terrain/ (7 files): Ran umcpropfix.
20100719 04:55:59-!- Gambit [~gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has quit [Quit: An ordinary boy. An ordinary name. But ordinary's just not good enough today.]
20100719 04:59:02-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has quit [Read error: Connection reset by peer]
20100719 04:59:21-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has joined #wesnoth-dev
20100719 05:04:20< CIA-87> espreon * r44335 /trunk/data/campaigns/Eastern_Invasion/scenarios/ (44 files): Standardized EI's scenario filenames; fixed the numbering.
20100719 05:15:44< CIA-87> alink * r44336 /trunk/src/ (display.cpp mouse_events.cpp): 
20100719 05:15:45< CIA-87> Almost revert r44309 (passing by value was wanted there)
20100719 05:15:45< CIA-87> But rewrite and improve comments to make it more clear.
20100719 05:18:06< alink> ^mordante: I guess I wrote my comments at a bad place (few lines below)
20100719 05:18:29< CIA-87> alink * r44337 /trunk/src/display.cpp: remove debug code from r44309 (used to visualize lightmap)
20100719 05:19:08< alink> this was a tricky area, these changing "const reference" are unintuitive 
20100719 05:20:04< alink> in fact, I suspect that some other functions of mouse code should use more passing by value
20100719 05:21:34-!- ancestral [~ancestral@97-116-105-226.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100719 05:22:30< alink> mmh sorry my commit msg of r44337 is wrong, it should be "remove debug code from r44336"
20100719 05:23:39< alink> and in 107 commits, we will reach r44444 :-)
20100719 05:24:42-!- Upthorn [ogmar@adsl-75-26-196-123.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 265 seconds]
20100719 05:26:09-!- alink [~alink@wesnoth/developer/alink] has quit [Remote host closed the connection]
20100719 05:49:19-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100719 05:54:19-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has quit [Quit: have to go to sleep]
20100719 05:56:07-!- Upth [ogmar@adsl-75-26-196-123.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100719 05:56:07-!- Upth is now known as Upthorn
20100719 05:59:21< CIA-87> espreon * r44338 /trunk/data/campaigns/Eastern_Invasion/ (23 files in 2 dirs): Standardized EI's scenario ids.
20100719 05:59:59< CIA-87> espreon * r44339 /trunk/data/campaigns/Eastern_Invasion/scenarios/ (16_Weldyn_Under_Attack.cfg 16_Weldyn_under_Attack.cfg): Synchronized a filename with its id.
20100719 06:32:08-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 276 seconds]
20100719 06:33:19< CIA-87> espreon * r44340 /trunk/data/campaigns/Eastern_Invasion/ (60 files in 2 dirs): Standardized the filenames of EI's map files.
20100719 06:34:01-!- silene [~plouf@AMarseille-251-1-59-126.w83-197.abo.wanadoo.fr] has joined #wesnoth-dev
20100719 06:34:02-!- silene [~plouf@AMarseille-251-1-59-126.w83-197.abo.wanadoo.fr] has quit [Changing host]
20100719 06:34:02-!- silene [~plouf@wesnoth/developer/silene] has joined #wesnoth-dev
20100719 06:37:23-!- Aethaeryn [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Quit: ...]
20100719 06:40:06-!- ancestral [~ancestral@97-116-105-226.mpls.qwest.net] has joined #wesnoth-dev
20100719 06:43:11< CIA-87> espreon * r44341 /trunk/data/campaigns/An_Orcish_Incursion/scenarios/ (3 files): Smote some unnecessary quotation marks for the sake of consistency.
20100719 06:44:23< silene> Ivanovic: if you are still looking for parallelization, you should consider using "xargs -P"
20100719 06:47:12< CIA-87> espreon * r44342 /trunk/data/campaigns/An_Orcish_Incursion/scenarios/ (14 files): Standardized the filenames of AOI's scenario files.
20100719 06:52:33< CIA-87> espreon * r44343 /trunk/data/campaigns/An_Orcish_Incursion/ (21 files in 2 dirs): Standardized the filenames of AOI's map files.
20100719 06:54:52< CIA-87> espreon * r44344 /trunk/data/campaigns/An_Orcish_Incursion/ (8 files in 2 dirs): Standardized the ids of AOI's scenarios.
20100719 07:08:30< CIA-87> espreon * r44345 /trunk/data/campaigns/Dead_Water/ (14 files in 2 dirs): Standardized the ids of DW's scenarios.
20100719 07:10:33< CIA-87> espreon * r44346 /trunk/data/campaigns/Dead_Water/scenarios/ (08_Talking_To_Tyegea.cfg 08_Talking_to_Tyegea.cfg): Standardized a filename.
20100719 07:10:41-!- dtiger [~dtiger@dynamic-vpdn-93-125-14-150.telecom.by] has joined #wesnoth-dev
20100719 07:25:53-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 264 seconds]
20100719 07:33:13-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20100719 07:36:13-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has quit [Quit: crimson_penguin]
20100719 07:42:30-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100719 08:08:40-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 08:08:50< timotei> morning
20100719 08:11:09< gabba> morning timotei
20100719 08:11:17< Crab_> hi, timotei
20100719 08:11:29< timotei> hi Crab_ , gabba 
20100719 08:13:29< timotei> Crab_: I managed to fix yesterday the bugs (I hope all) with the wmltools
20100719 08:13:37< Crab_> that's good :)
20100719 08:13:39< timotei> Crab_: but I can't reproduce the bug with campaign start:(
20100719 08:13:49< timotei> Crab_: can you send me your test campaign/addon?
20100719 08:14:00< Crab_> timotei: have you tried with existing campaign ?
20100719 08:14:20< Crab_> e.g., I've used Wings of Victory, from addon server
20100719 08:14:36< timotei> well, if I want to test it with that, I have to "modify" things so they will match data/add-ons dir, instead of campaign
20100719 08:14:39< timotei> one sec
20100719 08:14:59< timotei> (stupid mouse, it's a microsoft one. it was supposed to WORK with microsoft windows, isn't it?)
20100719 08:19:09-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: suspend to RAM]
20100719 08:22:33-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100719 08:33:58-!- Yoyobuae [~gerald@182.Red-213-98-251.dynamicIP.rima-tde.net] has joined #wesnoth-dev
20100719 08:36:33-!- Yoyobuae [~gerald@182.Red-213-98-251.dynamicIP.rima-tde.net] has quit [Client Quit]
20100719 08:38:51-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100719 08:44:36-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 09:00:01< CIA-87> silene * r44347 /trunk/ (data/lua/helper.lua src/scripting/lua.cpp): 
20100719 09:00:01< CIA-87> Added __shallow_literal and __shallow_parsed partial converters for optimizing code.
20100719 09:00:01< CIA-87> Added corresponding helpers.
20100719 09:02:32-!- zookeeper [~l@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20100719 09:08:28-!- ancestral [~ancestral@97-116-105-226.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20100719 09:40:51-!- Ivanovic [~ivanovic@dtmd-4db2a6db.pool.mediaWays.net] has quit [Changing host]
20100719 09:40:51-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20100719 09:43:28-!- timotei21 [~timotei@193.34.191.4] has joined #wesnoth-dev
20100719 09:44:24< Ivanovic> moin
20100719 09:44:30-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 260 seconds]
20100719 09:44:36< Crab_> hi, Ivanovic
20100719 09:49:43-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has quit []
20100719 10:01:47< timotei21> Crab_: from what addon
20100719 10:01:58< timotei21> Crab_: addon's server you downloaded Wings of Victory?
20100719 10:02:03< timotei21> Crab_: I can't find it here (trunk)
20100719 10:02:56< Crab_> well, get any campaign from addon server and try with it
20100719 10:32:01< timotei21> Crab_: ok, now I think I got your error. so, you right click on a scneario file (inside /scenarios) and it didn't start right
20100719 10:32:42< Crab_> yes, this, too
20100719 10:32:56< Crab_> I've tried to start the campaign by right-clicking _main.cfg, too
20100719 10:33:03< timotei21> that didn't work?
20100719 10:33:10< Crab_> no
20100719 10:33:12< timotei21> let me read again the logs
20100719 10:33:23< Crab_> ok. note that there were two separate errors in the command-line
20100719 10:33:30< timotei21> yes
20100719 10:33:50< Crab_> one with user data dir, another with scenario id being used (we don't need it when we want to start a campaign, afair)
20100719 10:34:22< timotei21> yes
20100719 10:34:38< timotei21> I've fixed that part (with non-needed scenario id) - forgot an if :P
20100719 10:35:05< timotei21> can you tell me if it works for you (from terminal): wesnoth --config-dir /tmp/ ?
20100719 10:35:14< timotei21> I want to see if on linux is the same behaviour
20100719 10:35:46< timotei21> on windows: wesnoth --config-dir C:/tmp overrides the "whole" config dir path (not just: "mydocuments"/tmp)
20100719 10:39:12< Crab_> User configuration directory: /home/crab//tmp/
20100719 10:39:21< timotei21> ah
20100719 10:39:30< Crab_> so, the windows behavior is buggy )
20100719 10:39:39< timotei21> damn, why does windows break things?:|
20100719 10:40:17< timotei21> maybe the code for windows is buggy?
20100719 10:40:19< timotei21> I'll check it
20100719 10:40:59< Crab_> filesystem.cpp:497
20100719 10:41:15< timotei21> thanks
20100719 10:41:22< Crab_> it has an undocumented feature - if it sees : in a path (e.g., c:\foo\bar ), then it treats it as a full path
20100719 10:42:47< Crab_> so, `relative` paths should work both on windows and gnu/linux
20100719 10:43:24< timotei21> yeah, I think I should use that... but what if the user uses another custom userdata path?:|
20100719 10:44:57< Crab_> I'd say `well, let it work with userdata inside home directory, it's not too important atm to extend *nix code to make it work with userdata outside home`
20100719 10:45:32< timotei21> ok
20100719 10:47:53< timotei21> hmm, if I choose 'open scenario'. how should I proceed i the following case: we have: myaddon/scenarios/my_scenario.cfg ? Should I 'recursively' fallback until find an _main.cfg in a directory?
20100719 10:48:09< timotei21> and pick up that as a 'campaign' file (to get campaign id's)?
20100719 10:48:33< Crab_> I think that you can pick '_main.cfg' in the root folder of that particular project
20100719 10:48:40< timotei21> or I should treat just the case myaddon/scenarios/my_scenario.cfg, so falling back 1 directory, and check only that
20100719 10:49:09< timotei21> Crab_: yes, that should work in case we are working on a project. but if we work on "user addons" project, where are multiple campaigns...
20100719 10:49:35< Crab_> maybe add all them as separate projects ?
20100719 10:50:05< timotei21> well, I still need that "user addons' project. since we can make a simple file (era/faction)
20100719 10:50:55< timotei21> and we won't create a new project just for that, and will create the file(s) directly in our addons' directory
20100719 10:51:17< Crab_> why not create a new project just for that ?
20100719 10:51:37< timotei21> hmmm
20100719 10:52:16-!- timotei21 is now known as timotei
20100719 10:52:23-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20100719 10:52:24-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 10:54:51-!- gabba [~gabba@wesnoth/developer/gabba] has left #wesnoth-dev []
20100719 10:59:52-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100719 11:04:27< CIA-87> gabba * r44348 /trunk/src/whiteboard/manager.cpp: Whiteboard: prevent executing moves when it's not your turn. You can still delete/reorder actions out-of-turn.
20100719 11:04:32< CIA-87> gabba * r44349 /trunk/src/whiteboard/validate_visitor.cpp: Whiteboard: Fix crash when a unit had a planned move to a hex which suddenly isn't available anymore because of a change in the planned actions.
20100719 11:04:39< CIA-87> gabba * r44350 /trunk/src/ (mouse_events.cpp whiteboard/move.cpp): Whiteboard: more progress toward planning outside your turn.
20100719 11:04:45< CIA-87> gabba * r44351 /trunk/src/ (5 files in 2 dirs): whiteboard: improved ctrl key behavior: now instantly switches between footsteps and arrow
20100719 11:04:54< CIA-87> gabba * r44352 /trunk/src/whiteboard/ (10 files): Whiteboard: reworked stream operators to work in a semi-polymorphic way.
20100719 11:04:58< CIA-87> gabba * r44353 /trunk/src/whiteboard/validate_visitor.cpp: Whiteboard: Fix deletion of attacks with a zero-hex move.
20100719 11:05:06< CIA-87> gabba * r44354 /trunk/src/whiteboard/ (manager.cpp manager.hpp): Whiteboard: scoped_planned_unit_map now only removes the planned unit map if it wasn't set when it was created.
20100719 11:19:56-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100719 11:26:28-!- loonybot [~loonybot@ppp79-139-141-141.pppoe.spdop.ru] has joined #wesnoth-dev
20100719 11:26:28-!- loonybot [~loonybot@ppp79-139-141-141.pppoe.spdop.ru] has quit [Changing host]
20100719 11:26:28-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20100719 11:27:26-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20100719 11:51:38-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100719 11:55:20< timotei> hi fendrin 
20100719 11:55:36< timotei> Crab_: ok, I'll use the "get project's _main.cfg" idea.
20100719 11:55:44< Crab_> ok
20100719 11:56:02< timotei> but I'll need to know fendrin's opinion regarding the "user addons" project
20100719 11:56:13< Crab_> ok
20100719 11:56:29< Unnheulu> timotei, will it have an upload script, or must that be done ingame?
20100719 11:56:45< timotei> Unnheulu: it has already-> right click on project->Upload addon :)
20100719 11:56:56< Unnheulu> Ooh, cool
20100719 11:57:10< timotei> have you managed to test the plugin?
20100719 11:57:17< Unnheulu> How does it upload the addon (copy/paste of ingame code?)
20100719 11:57:32< timotei> now, using the script 'wesnoth_add_manager' from data/tools
20100719 11:57:38< timotei> s/now/no*
20100719 11:57:40< Unnheulu> Ah
20100719 11:57:56< Unnheulu> How do I try it?
20100719 11:59:09< timotei> well, if you didn't let's wait until it's release for testing (that is this weeks I hope)
20100719 11:59:32< Unnheulu> Kay
20100719 11:59:52< Unnheulu> Eww, conflicts...
20100719 12:00:16< timotei> :P
20100719 12:00:35-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100719 12:01:26< Unnheulu> Finally...
20100719 12:01:44< Unnheulu> I just checked out the latest revision...how do I test it! :P
20100719 12:01:58< Unnheulu> s/!// >.>
20100719 12:02:27< timotei> just checkout the utils/java/eclipse_plugin/readme
20100719 12:02:33< timotei> (need to 'make' first)
20100719 12:02:54< timotei> Unnheulu: or here: http://eclipse.wesnoth.org/README.pdf
20100719 12:04:26< Unnheulu> Thanks
20100719 12:04:35< timotei> np
20100719 12:04:38< Unnheulu> Out of curiosity, what does the xyz.xyz.xyz mean?
20100719 12:04:51< timotei> ha?
20100719 12:04:54< timotei> in what context?
20100719 12:04:58< Unnheulu> Despite my best efforts to learn java I never understodd :P
20100719 12:05:03< Unnheulu> Like org.wesnoth.wml.ui
20100719 12:05:14< timotei> well, usually a package
20100719 12:05:22< timotei> just like an url but with '.' instead of '/'
20100719 12:05:25< timotei> org/wesnoth/wml/ui
20100719 12:05:31< timotei> a path also
20100719 12:05:36< Unnheulu> But C++ and python don't need it?
20100719 12:05:47< timotei> C++ has namespaces
20100719 12:05:51< timotei> pythong I don't know
20100719 12:06:07< timotei> package(java) = namespace(C#,C++,other) 
20100719 12:06:09< timotei> almost equal
20100719 12:06:28< Unnheulu> Ah, namespace is like std::
20100719 12:06:33< timotei> yes
20100719 12:06:34< Unnheulu> I don't think python has that
20100719 12:06:48< timotei> they have modules instead iirc
20100719 12:07:04< Unnheulu> Yeah
20100719 12:07:26< Unnheulu> But you can ignore the <module>. by just importing *
20100719 12:08:05< Unnheulu> Thanks
20100719 12:08:52< timotei> np
20100719 12:10:03-!- EdB [~edb@tss37-1-89-84-18-220.dsl.club-internet.fr] has joined #wesnoth-dev
20100719 12:17:48< Unnheulu> timotei, I'm not sure I understand 2.1?
20100719 12:18:41< timotei> entire 2.1? if you have trunk, then you can look in wesnoth/utils/java
20100719 12:18:59< timotei> there are the 3 requested projcets
20100719 12:18:59< Unnheulu> Nevermind, I get it now
20100719 12:19:07< timotei> :P
20100719 12:19:32< Unnheulu> So for 2.3 then I press browse next to select root directory?
20100719 12:19:58< Unnheulu> s/3/1.3/
20100719 12:23:29< Unnheulu> Man, if I had just followed 3 instead of 2 things would be simpler >.<
20100719 12:26:26< timotei> well... you wanted to be a developer:))
20100719 12:26:42< Unnheulu> :P
20100719 12:27:12< Unnheulu> I don't think I'd be a very good developer with my current knowledge of Java (most of it learnt on an airoplane in febuary :P)
20100719 12:28:01< timotei> :)
20100719 12:28:04< timotei> ok I have to go
20100719 12:28:04< timotei> bye
20100719 12:28:12< Unnheulu> Bye
20100719 12:28:13-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
20100719 12:30:43-!- dtiger [~dtiger@dynamic-vpdn-93-125-14-150.telecom.by] has quit [Remote host closed the connection]
20100719 12:58:20-!- dtiger [~dtiger@dynamic-vpdn-93-125-14-150.telecom.by] has joined #wesnoth-dev
20100719 13:01:02-!- Gambit [~Gambit@pa-67-234-73-7.dhcp.embarqhsd.net] has joined #wesnoth-dev
20100719 13:02:55-!- dtiger [~dtiger@dynamic-vpdn-93-125-14-150.telecom.by] has quit [Remote host closed the connection]
20100719 13:14:09< CIA-87> thespaceinvader * r44355 /trunk/ (10 files in 5 dirs): Add and wire 2 new Orc portraits - Leader/Ruler and Sovereign. Update changelogs, portrait credits.
20100719 13:14:53-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has joined #wesnoth-dev
20100719 13:27:56-!- pokhbocee [~baris@95.10.29.193] has joined #wesnoth-dev
20100719 13:28:41< pokhbocee> hey Crab_
20100719 13:28:46< Crab_> hi, pokhbocee
20100719 13:28:47< pokhbocee> hey everbody
20100719 13:29:54-!- DesertPanther_ [~Khalid@41.234.232.138] has joined #wesnoth-dev
20100719 13:32:35-!- DesertPanther [~Khalid@unaffiliated/desertpanther] has quit [Ping timeout: 258 seconds]
20100719 13:36:38< pokhbocee> Crab_ : if wesnoth.get_locations()  wesnoth.get_teams() are not implemented yet, i would like to work on it
20100719 13:45:27< Crab_> pokhbocee: it's better to discuss any potential lua enhancements with silene, he knows better what things are good to do and what are not...
20100719 13:46:32< pokhbocee> oh ok.
20100719 13:47:04< Crab_> pokhbocee: he reads the logs, so you can leave a message for him here.
20100719 13:48:52< pokhbocee> i ask about those, since last time we discuss u told me about those
20100719 13:50:22< pokhbocee> are those related to ai?
20100719 13:52:25< Crab_> `getting access to a list of locations which match a particular criteria` is important for writing AI snippets, yes
20100719 13:54:52-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100719 13:57:35< pokhbocee> ok ill leave a message to silene
20100719 14:00:01-!- Johannes13 [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100719 14:00:51-!- pokhbocee [~baris@95.10.29.193] has quit [Quit: Leaving]
20100719 14:01:56-!- pokhbocee [~baris@95.10.29.193] has joined #wesnoth-dev
20100719 14:07:01< pokhbocee> silene: hello, i wanted to ask that if  wesnoth.get_locations() and  wesnoth.get_teams() are not implemented yet, i would like to work on it or smth else according to ur suggestions
20100719 14:10:12-!- EdB [~edb@tss37-1-89-84-18-220.dsl.club-internet.fr] has quit [Remote host closed the connection]
20100719 14:10:37< pokhbocee> Crab_: is silene around, coz last time he was gone for some time
20100719 14:11:11< Crab_> yes, he's around. for example, committed today :)
20100719 14:11:30< pokhbocee> :D ok thank you
20100719 14:19:05-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 14:22:54< timotei> fendrin: ping
20100719 14:34:28< silene> pokhbocee: get_teams doesn't make much sense (?), but there is a use for get_locations; i suggest working on trunk, since the code is evolving fast
20100719 14:39:26< pokhbocee> silene: those were Crab_'s suggestions so i really am not sure.
20100719 14:39:37< pokhbocee> so im listening to you
20100719 14:41:48< CIA-87> silene * r44356 /trunk/data/ (3 files in 2 dirs): Updated LoW to new Lua idioms.
20100719 14:42:06< CIA-87> silene * r44357 /trunk/data/lua/wml-tags.lua: Simplified code a bit.
20100719 14:42:13< CIA-87> silene * r44358 /trunk/src/sound.cpp: Cleaned code a bit.
20100719 14:42:14< Crab_> about get_teams - I was thinking about a helper function which will do something like 'get me the list of all teams which are my enemies', sort of 'standard team filter'.  about 'get locations' - just allow to access the SLF from lua
20100719 14:42:18< CIA-87> silene * r44359 /trunk/src/scripting/lua.cpp: Added set_music function.
20100719 14:42:21< CIA-87> silene * r44360 /trunk/ (3 files in 3 dirs): Moved [music] implementation to Lua and added test.
20100719 14:42:26< CIA-87> silene * r44361 /trunk/ (data/lua/wml-tags.lua src/game_events.cpp): Moved [command] implementation to Lua.
20100719 14:42:28< CIA-87> silene * r44362 /trunk/ (data/lua/wml-tags.lua src/game_events.cpp): Moved [if] and [while] implementations to Lua.
20100719 14:42:33< CIA-87> silene * r44363 /trunk/data/lua/wml-tags.lua: Added error message on missing tag.
20100719 14:44:13< silene> Crab_: i feel like get_teams doesn't bring much wrt is_enemy; do you have a particular use in mind?
20100719 14:46:12-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 14:47:22< Crab_> silene: no, is_enemy is enough for me, now.
20100719 14:56:09-!- EdB [~edb@4.103.89-79.rev.gaoland.net] has joined #wesnoth-dev
20100719 14:56:24-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 14:56:57-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 14:58:39< fendrin> timotei: pong
20100719 14:58:47< timotei> fendrin: hi
20100719 14:59:07< timotei> fendrin: I was asking myself if the 'user addons' is good to be there.
20100719 14:59:19< timotei> fendrin: but in the meantime I got the answer myself: yes
20100719 14:59:29< timotei> fendrin: in case the user edits directly the addon
20100719 14:59:47< timotei> (so the addon is not outside the 'userdir/data/addons'
20100719 15:01:28< timotei> fendrin: am I right?
20100719 15:01:51< fendrin> I don't really get what you mean with "is good to be there".
20100719 15:02:19< timotei> well, there are/were some issues when trying to start a scenario for example, from the user addon's project
20100719 15:02:37< timotei> and I was wondering where to fallback for searching the '_main.cfg' file for campaign's id
20100719 15:03:14< timotei> and as Crab_ suggested to search in project's root, but since the 'user addons' project contains more than 1 project, I need to refine the search
20100719 15:03:22-!- Appleman1234 [~Appleman1@CPE-60-226-176-19.qld.bigpond.net.au] has quit [Ping timeout: 265 seconds]
20100719 15:03:33< timotei> so I was thinking to remove the 'user addons' project, but I don't think anymore it's the case
20100719 15:04:18< fendrin> user addons (guess you refer to the .wesnoth/data/addons directory) should contain subdirectories with a _main.cfg in each, right?
20100719 15:04:39< timotei> yes
20100719 15:04:40< timotei> :)
20100719 15:04:50-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Quit: Got to go, I'm in a hurry so bye.]
20100719 15:05:05< timotei> so, I'll fallback to the first subdirectory of the selected scenario
20100719 15:05:32< timotei> useraddons/mycampaign/customs/scenarios/myscenario.cfg will search in: useraddons/ and as a fallback in useraddons/mycampaign
20100719 15:06:06< timotei> or I should start start searching upwards from the scenario file? (which is in reallity more the case?)
20100719 15:06:40< timotei> in my example: useraddons/mycampaign/customs/scenarios/ , useraddons/mycampaign/customs/ , useraddons/mycamapign/ an finally useraddons/
20100719 15:08:26< timotei> fendrin: what do you say about this search?
20100719 15:09:58< fendrin> timotei: The first subdirectory of the selected scenario means it's root?
20100719 15:10:23< timotei> it's: project/firstdir
20100719 15:10:34< timotei> where project is the project where the scenario file resides
20100719 15:11:11< fendrin> I don't get that, how do you sort the subdirectory so you can say : This is the first, that is the second and so on.
20100719 15:11:25< timotei> that was stupid way of telling this, sorry
20100719 15:11:31< Crab_> timotei: the 'search upwards' methods makes sense, if you don't want the 1:1 relation between projects and campaigns.
20100719 15:12:04< timotei> fendrin: but you can think this way: /home/wesnoth/data: /home is first, wesnoth is 2nd, data is 3rd.. etc
20100719 15:13:07< fendrin> timotei: okay, but normally a campaign will have many subdirectories under it's root.
20100719 15:13:47< timotei> well, yeah, but directly. I'm talking about the scenario file being 'burried' in too much directories
20100719 15:13:55< fendrin> timotei: you can search for _main.cfg with the campaign tag only in two places. The addons directory itself or the root of the campaign containded in addons.
20100719 15:14:17< fendrin> The first situation should be very rarely since it is depreciated.
20100719 15:14:17< timotei> ok
20100719 15:14:26< timotei> good
20100719 15:14:42< fendrin> So I am not even sure if searching at that place is a good idea anymore.
20100719 15:19:39< CIA-87> silene * r44364 /trunk/ (data/lua/wml-tags.lua src/game_events.cpp): Moved [switch] implementation to Lua.
20100719 15:19:44< CIA-87> silene * r44365 /trunk/data/lua/wml-tags.lua: Removed [filter*] tags when handling sequence of commands.
20100719 15:19:49< CIA-87> silene * r44366 /trunk/src/scripting/lua.cpp: Fixed unbalanced stack.
20100719 15:19:52< CIA-87> silene * r44367 /trunk/src/game_events.cpp: Moved handling of command sequences to Lua.
20100719 15:20:37-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625231939]]
20100719 15:23:35-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 15:28:48-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Ping timeout: 240 seconds]
20100719 15:36:57-!- pokhbocee [~baris@95.10.29.193] has quit [Quit: Leaving]
20100719 15:46:47-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20100719 15:58:19< CIA-87> silene * r44368 /trunk/src/scripting/lua.cpp: Added scroll_to_tile function.
20100719 15:58:21< CIA-87> silene * r44369 /trunk/src/scripting/lua.cpp: Added missing storage specifier.
20100719 15:58:31< CIA-87> silene * r44370 /trunk/ (data/lua/wml-tags.lua src/game_events.cpp): Moved [scroll_to] and [scroll_to_unit] implementations to Lua.
20100719 15:58:46-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 16:10:55-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 16:11:26-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 16:11:41-!- Crab_ [~Crab_@wesnoth/developer/crab] has quit [Quit: Leaving.]
20100719 16:22:45< zookeeper> uh. why the heck is there a file named orc_grunt_6.png in portraits/orcs/transparent/?
20100719 16:23:35< zookeeper> since there's no reasoning for that anywhere that i can see, i guess i'll just rename it...
20100719 16:25:52-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 16:26:15-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 16:27:33< CIA-87> zookeeper * r44371 /trunk/data/core/images/portraits/orcs/transparent/ (grunt-6.png orc_grunt_6.png): Renamed orc_grunt_6.png to grunt-6.png to keep the naming consistent.
20100719 16:37:56-!- mjs-de [~mjs-de@vpw.wh.uni-dortmund.de] has joined #wesnoth-dev
20100719 16:41:13-!- King_Elendil_ [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 16:41:13-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 16:41:16-!- King_Elendil_ is now known as King_Elendil
20100719 16:41:50< timotei> esr: can you also add on wmlscope the "-p"/"--progress" command please? You added only to wmllint. (or, if you allow, I can try do it myself)
20100719 16:49:16< esr> timotei: It does two passes over the tree, so you'll get each filename as a progress message twice.
20100719 16:50:50< timotei> esr: can't we know when it's the first and the second pass?
20100719 16:55:47-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 16:56:11-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 16:56:20< esr> Do want a message for each?
20100719 16:56:52< timotei> esr: well, I need some sort of "progress", so I can inform the user: hey, this is still running
20100719 16:57:09< timotei> esr: I don't think is any problem if they are outputed twice
20100719 17:01:49< timotei> esr: and ideally I would need only a known multiple (so I can split in my program to get only 1) info per file
20100719 17:02:09< timotei> esr: cause with -w1 and -w2 usually I get different things
20100719 17:04:21< CIA-87> esr * r44372 /trunk/data/tools/ (wesnoth/wmltools.py wmlscope): Progress option for wmlscope.
20100719 17:04:31< timotei> esr: thank you!
20100719 17:10:49-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 17:11:02-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 17:16:35< King_Elendil> who was the author of LOW?
20100719 17:18:39< timotei> King_Elendil: http://wiki.wesnoth.org/Credits#Legend_of_Wesmere
20100719 17:18:53< King_Elendil> thanks!
20100719 17:18:57< timotei> np
20100719 17:25:51-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 17:26:01-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 17:39:34-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has joined #wesnoth-dev
20100719 17:40:08-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 17:40:50-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 17:41:59< CIA-87> timotei * r44373 /trunk/utils/java/org.wesnoth.wml.ui/src/org/wesnoth/ui/syntax/bracketmatching/WMLBracketMatching.java: eclipse plugin: fix highlighting start/end tags - now it won't highlight spaces
20100719 17:42:18< CIA-87> timotei * r44374 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/ (3 files in 2 dirs): eclipse plugin: use a '.wesnoth' file than '.ignore' so we can put more information in the file
20100719 17:42:29< timotei> Espreon: god, you almost screw up my commits with yout filename change :P
20100719 17:42:35< CIA-87> timotei * r44375 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/WorkspaceUtils.java: eclipse plugin: add filter for hiding workspace's projects in 'user addon's project
20100719 17:42:37< timotei> Espreon: windows is a case-insensitive bastard
20100719 17:42:48< CIA-87> timotei * r44376 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/ (GameUtils.java ProjectUtils.java WorkspaceUtils.java): eclipse plugin: fix starting scenario/campaign
20100719 17:43:04< CIA-87> timotei * r44377 /trunk/utils/java/eclipse_plugin/plugin.xml: eclipse plugin: enable campaign starting only on containers
20100719 17:43:18< CIA-87> timotei * r44378 /trunk/utils/java/eclipse_plugin/ (7 files in 5 dirs): eclipse plugin: make the wmltools menu entry available for folders too
20100719 17:43:34< CIA-87> timotei * r44379 /trunk/utils/java/eclipse_plugin/src/wesnoth_eclipse_plugin/utils/WMLTools.java: eclipse plugin: add a better progress reporting for the wmltools jobs
20100719 17:45:25-!- Elvish_Pillager [~eli@71-10-224-192.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20100719 17:48:39-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20100719 17:49:34-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 17:54:05< timotei> esr: wow, runned: python wmlscope --progress /wesnoth/data/core and for 98% of output it stayed on /wesnoth/data/core/units/wose/Wose.cfg
20100719 17:54:20< timotei> esr: any idea why?
20100719 17:54:45< timotei> fendrin: what would be the amount of scenario gold based on difficulty?
20100719 17:55:32-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 17:55:48-!- King_Elendil [~King_Elen@75.143.235.104] has joined #wesnoth-dev
20100719 17:56:28< timotei> fendrin: also, would one use GOLD EASY_AMOUNT NORMAL_AMOUNT HARD_AMOUNT
20100719 17:56:28< timotei>     Macro which will let you say {GOLD x y z} to set starting gold depending on easy/medium/hard - x/y/z
20100719 17:56:28< timotei>  ?
20100719 17:58:20< esr> timotei: No idea.  You should perhaps loom into this.
20100719 17:58:26< timotei> ok
20100719 17:58:28< esr> s/loom/look/
20100719 18:05:20-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has joined #wesnoth-dev
20100719 18:06:00-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has quit [Read error: Connection reset by peer]
20100719 18:06:11-!- elvish_sovereign [~elvish_so@pool-173-59-71-113.phlapa.east.verizon.net] has joined #wesnoth-dev
20100719 18:08:21< silene> eleazzaar: in case you are not aware of it, shallow water and coastal reefs don't mix well (e.g. Low 3)
20100719 18:09:11< timotei> silene: what exactly does the '@' character mean? I can't find that in wiki
20100719 18:09:32< silene> timotei: in which context?
20100719 18:09:36-!- crimson_penguin [~ben@wesnoth/developer/crimsonpenguin] has joined #wesnoth-dev
20100719 18:09:42< timotei> silene: it has different meanings?
20100719 18:09:52< timotei> silene: well, when specifying map_data for example
20100719 18:09:52-!- King_Elendil [~King_Elen@75.143.235.104] has quit [Read error: Connection reset by peer]
20100719 18:10:06< timotei> silene: fendrin told me to put an @ before maps name/path
20100719 18:10:09< fendrin> timotei: The amount of scenario gold based on difficulty? I don't get this question.
20100719 18:10:30< timotei> fendrin: fendrin> I know that the difficult levels are set in the campaign wizard.
20100719 18:10:30< timotei> 20100718 15:07:05< timotei> yes
20100719 18:10:30< timotei> 20100718 15:07:27< fendrin> Can that information be reused to offer a starting gold definement per difficult level in the new scenario wizard?
20100719 18:10:57-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100719 18:11:02< silene> timotei: don't; it's the 1.6 syntax for specifying file locations
20100719 18:11:19< fendrin> timotei: That means that you should offer an input field per difficult level of the campaign for the starting gold.
20100719 18:11:32< silene> timotei: starting from 1.8, location of user files starts by ~ (e.g. ~add-ons/...)
20100719 18:11:38-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has quit [Client Quit]
20100719 18:12:12< timotei> fendrin: should that input be in campaign wizard (under difficulty levels?). 
20100719 18:12:18< timotei> silene: ok
20100719 18:12:47< silene> timotei: and to answer your earlier, yes it has several meanings
20100719 18:13:43< timotei> silene: thanks
20100719 18:14:12< CIA-87> silene * r44380 /trunk/src/ (generate_report.cpp reports.hpp): Avoided costly copy.
20100719 18:14:30< CIA-87> silene * r44381 /trunk/src/ (7 files in 2 dirs): 
20100719 18:14:30< CIA-87> Simplified code.
20100719 18:14:30< CIA-87> Fixed internationalization of selected terrain report.
20100719 18:18:30< fendrin> timotei: no, in the campaign wizard you specify the number and names of the difficult levels, in the scenario wizard you can set the starting gold per difficult level.
20100719 18:18:39< timotei> ok
20100719 18:20:09< Elvish_Pillager> silene: when I call string.match(unit.name, "some pattern") - with "unit" being the cfg of an anonymous local player during prestart - it tells me "bad argument #1 to 'match' (string expected, got userdata)". What am I doing wrong here? (It works fine with a named player.)
20100719 18:20:25< Elvish_Pillager> * an anonymous local player's leader unit, that is.
20100719 18:36:51-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100719 18:45:39< CIA-87> ivanovic * r44382 /branches/1.8/ (30 files in 13 dirs): updated Chinese (simplified) translation
20100719 18:46:01< CIA-87> ivanovic * r44383 /trunk/ (32 files in 15 dirs): updated Chinese (simplified) translation
20100719 19:00:57< timotei> fendrin: what should I do with an outer-worksace imported map?
20100719 19:01:06< timotei> copy in current directory?
20100719 19:13:17-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20100719 19:17:31-!- anonymissimus [~chatzilla@HSI-KBW-078-042-069-017.hsi3.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20100719 19:18:48-!- anonymissimus [~chatzilla@HSI-KBW-078-042-069-017.hsi3.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819]]
20100719 19:19:58-!- EdB [~edb@4.103.89-79.rev.gaoland.net] has quit [Remote host closed the connection]
20100719 19:19:59-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 260 seconds]
20100719 19:27:16-!- Upthorn [ogmar@adsl-75-26-196-123.dsl.scrm01.sbcglobal.net] has quit [Read error: Connection reset by peer]
20100719 19:27:46-!- Upth [ogmar@adsl-75-26-189-133.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100719 19:27:46-!- Upth is now known as Upthorn
20100719 19:32:07< silene> Elvish_Pillager: name is a translatable string; never do anything on them (this is true for plain wml too); are you sure you shouldn't be matching unit.id instead?
20100719 19:32:14-!- billynux [~billy@wesnoth/developer/billynux] has joined #wesnoth-dev
20100719 19:33:48< silene> if you really need to use the name, then use tostring(unit.name), but again it seems like a bad idea
20100719 19:41:39-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 19:41:46-!- shadowm_laptop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20100719 19:50:15-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20100719 19:50:16< fendrin> timotei: yes, copy it in the map folder of the campaign.
20100719 19:51:37< timotei> ok
20100719 19:54:54< CIA-87> silene * r44384 /branches/1.8/src/playcampaign.cpp: 
20100719 19:54:54< CIA-87> Fixed 1.9 syntax creeping on 1.8 and causing segfaults. (Fix for bug #16283.)
20100719 19:54:54< CIA-87> (Where is the compiler typechecker when you need it?!)
20100719 20:10:08-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20100719 20:14:04-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20100719 20:14:12< CIA-87> mordante * r44385 /trunk/src/scripting/lua.cpp: Fix a compiler warning.
20100719 20:14:16< mordante> servus
20100719 20:14:19< timotei> hi mordante :D
20100719 20:14:26< mordante> hi timotei 
20100719 20:14:29< timotei> entering the chat with a commit?:)
20100719 20:14:36< mordante> yup
20100719 20:14:51< timotei> yo billynux. didn't see you online :D
20100719 20:15:17< timotei> (I think now, when seeing a mentor/student I get my mind to his student/mentor)
20100719 20:15:29< billynux> hi timotei, mordante 
20100719 20:15:37< mordante> hi billynux 
20100719 20:15:41< billynux> timotei, :)
20100719 20:25:43< billynux> mordante, I managed to get a smaller version of the out-of-sync error
20100719 20:28:17-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Ping timeout: 264 seconds]
20100719 20:28:23< CIA-87> zookeeper * r44386 /trunk/data/campaigns/Under_the_Burning_Suns/scenarios/08_Out_of_the_Frying_Pan.cfg: Rewrote the rising water event to be much shorter and more efficient, and cleaned up some of the human explorers' conditional dialogue.
20100719 20:29:09< mordante> billynux, nice also reliable to reproduce?
20100719 20:30:10< billynux> mordante, I don't think so. My guess is that sometimes packets are just sent wrong and the error is not picked up, but this is not predictable
20100719 20:31:15< mordante> :-(
20100719 20:31:37< mordante> guess we need to look what's exactly send and what received
20100719 20:33:04< billynux> yes, I wanted to postpone that until I get the connection/disconnection between ana server/client working
20100719 20:34:07< mordante> ok and how does it go with the connect/disconnect?
20100719 20:34:35< billynux> ATM I'm having trouble simply connecting an ana client to an ana server
20100719 20:36:17< billynux> mordante, and... if I connect one SDL client and then disconnect it, it still shows up as connected in the lobby
20100719 20:36:51< billynux> mordante, i.e. if I reconnect with the same nick it'll ask to use a different one, and after logging in it will show it as connected (the old disconnected one)
20100719 20:38:49 * boucman might have a working bridge macro at last
20100719 20:40:48-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100719 20:40:58-!- timotei21 [~timotei@193.34.191.4] has joined #wesnoth-dev
20100719 20:41:34< timotei21> .
20100719 20:43:53-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20100719 20:44:05< CIA-87> espreon * r44387 /trunk/data/campaigns/ (5 files in 5 dirs): Ran umcpropfix.
20100719 20:45:31< mordante> billynux, and you properly close the socket when closing the connection?
20100719 20:46:08< CIA-87> espreon * r44388 /trunk/data/core/images/ (9 files in 5 dirs): Ran umcpropfix.
20100719 20:47:04< billynux> mordante, When a client gets disconnected from the server? yes, let me recheck that
20100719 20:47:11< CIA-87> espreon * r44389 /trunk/images/misc/l10n/zh_CN/logo.png: Ran umcpropfix.
20100719 20:53:30< CIA-87> espreon * r44390 /trunk/ (l10n-spec l10n-track): Added Chinese Simplified's language code to l10n-spec; updated the track file.
20100719 20:54:06< billynux> mordante, well, yes, the socket cancels all pending operations ( asio_server.cpp:369 ) and then gets destroyed ( first removing the client from the server list in ana )
20100719 21:01:12< eleazzaar> silene: yes, thanks, i'm aware of the problem with reef and water.
20100719 21:02:50< mordante> billynux, odd
20100719 21:04:12< billynux> mordante, my guess is that the implementation of handle_disconnect (network_manager_ana.cpp:1211) is not implemented right
20100719 21:05:27< billynux> all it does (in the else clause) is remove the client from the corresponding server_manager
20100719 21:05:27-!- timotei21 is now known as timotei
20100719 21:05:36-!- timotei [~timotei@193.34.191.4] has quit [Changing host]
20100719 21:05:36-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20100719 21:06:03< Unnheulu> timotei, since when have you been cloaked?
20100719 21:06:07< mordante> billynux, maybe also explicitly call the socket::close
20100719 21:06:20< timotei> Unnheulu: well, don't know:-??
20100719 21:06:23 * Unnheulu never noticed it before :P
20100719 21:06:40< mordante> timotei, since you asked for it probably :-P
20100719 21:06:47< timotei> mordante: yeah
20100719 21:06:47 * boucman just spent ~1/2 h fixing an non-existent bug
20100719 21:06:54-!- gabba [~gabba@wesnoth/developer/gabba] has joined #wesnoth-dev
20100719 21:06:59< billynux> mordante, good point, in the destructor? (I'll try this)
20100719 21:06:59< gabba> bonjour
20100719 21:07:00< mordante> joy boucman :-(
20100719 21:07:05< boucman> salut gabba
20100719 21:07:08< mordante> salut gabba 
20100719 21:07:13< gabba> salut boucman et mordante
20100719 21:07:21< Espreon> gabba: Ic grete þe!
20100719 21:07:32< boucman> mordante: the single hex stone bridge looked too much like the straight bridge, and I was searching why it didn't display properly
20100719 21:07:33< gabba> hi Espreon
20100719 21:07:38< timotei> hi gabba 
20100719 21:07:39< boucman> now for some deep testing
20100719 21:07:45< gabba> hey timotei
20100719 21:07:49< boucman> gabba: want a rematch ? :)
20100719 21:07:50< CIA-87> caslav_ilic * r44391 /branches/1.8/ (l10n-spec l10n-track): Updated localized images for Chinese (Simplified).
20100719 21:08:09< boucman> hmm
20100719 21:08:14< boucman> eleazzaar: around ?
20100719 21:08:18< eleazzaar> yeah
20100719 21:08:25< Espreon> Hmmmm... I forgot about 1.8... meh... oh well...
20100719 21:08:30< mordante> boucman, and the problem was?
20100719 21:08:40< gabba> boucman: why not :), let me just check if my latest changes haven't broken anything critical
20100719 21:08:45< boucman> if you have 5' could you do the gluing job of yesterday for single hex bridges ?
20100719 21:08:55< boucman> sure, i'm ready to test whenever you are
20100719 21:09:09< mordante> btw boucman did you still have questions regarding haloes?
20100719 21:09:26< boucman> everything worked correctly, but I was not able to differentiate the images
20100719 21:09:32< eleazzaar> boucman: that's 4 different version per direction, eyt
20100719 21:09:35< boucman> mordante: yes, did you see my last commit in that area ?
20100719 21:09:44< eleazzaar> s/eyt/ yes
20100719 21:09:52< mordante> boucman, not yet catching up with the commits right now
20100719 21:09:56< CIA-87> espreon * r44392 /trunk/doc/manual/ (20 files in 2 dirs): Ran umcpropfix.
20100719 21:10:00< boucman> no, just three images, -ns -nesw -senw
20100719 21:10:09< boucman> you have to glue at both end for these
20100719 21:10:30< mordante> btw billynux you saw my commit that really fixed devdocs?
20100719 21:10:38< eleazzaar> yeah, but what if one end is by a castle and the other isn't?
20100719 21:10:52< eleazzaar> of *both* ends could be by castles
20100719 21:11:08< billynux> mordante, yes... good call, it was a doxygen mainpage issue
20100719 21:11:16< boucman> daaaaamit
20100719 21:11:18< billynux> mordante, closing the socket doesn't help
20100719 21:11:20< boucman> hadn't thought of that
20100719 21:11:25< mordante> :-(
20100719 21:11:41< mordante> alink regarding r44336 is the hardcoded filename intended?
20100719 21:11:42< billynux> mordante, if you look at server.cpp the "has logged off" message is never shown
20100719 21:12:04< mordante> billynux, do you cancel before or after closing it?
20100719 21:12:27< mordante> let me check I think I have some paranoia shutdown code for asio
20100719 21:12:29< boucman> eleazzaar: you're right, we need all these variations and a naming scheme???
20100719 21:12:32< gabba> boucman: yesterday was tough... I spent a lot of time converting planned moves to store only hex positions instead of direct references to units (since units can disappear or whatever through wml events), but then I realized that I needed to apply the temp unit map every time a move/attack wants to find its unit... I stashed the whole patch, and now I'm not sure if I should finish the job or remove any trace of that idea and just validate as ofte
20100719 21:12:35< billynux> mordante,  I cancel before
20100719 21:12:45< billynux> mordante, The point is, I have to issue a network::error exception
20100719 21:12:56< billynux> mordante, for the server code to catch it
20100719 21:12:59< eleazzaar> 4 variants per direction
20100719 21:13:12< Unnheulu> gabba, if I were to play a game in svn (assuming I find someone with the same revision) your code would work?
20100719 21:13:15< Unnheulu> Or is it in a branch
20100719 21:14:01< boucman> eleazzaar: if you count three directions, then yes
20100719 21:14:17< gabba> Unnheulu: no, it's in svn head: just type :wb to activate the whiteboard or use the ctrl key to just plan one move, but.... expect crashes. The whole thing is still undergoing heavy refactorings.
20100719 21:14:46< eleazzaar> boucman: i'll just go make them and you can rename them to whatever scheme you come up with
20100719 21:14:47< Unnheulu> Kay
20100719 21:14:53< mordante> billynux, here some code I use which works for me(tm) http://paste.debian.net/80927/
20100719 21:15:02< boucman> eleazzaar: that's what I was going to say :)
20100719 21:15:29< mordante> alink nevermind I just read r44337
20100719 21:15:31< eleazzaar> that's what happens when you work together with someone for a long time
20100719 21:15:42-!- phlaem [~a@e178100145.adsl.alicedsl.de] has joined #wesnoth-dev
20100719 21:15:48< gabba> Unnheulu: I'm gonna play such a game with boucman pretty soon
20100719 21:16:18< billynux> mordante, but I'm not having trouble closing the socket
20100719 21:16:32< billynux> mordante, the problem is, the server never finds out it was closed
20100719 21:16:43< Unnheulu> gabba, ooh, if its in my timezone, mind me observing?
20100719 21:16:47< Unnheulu> I like watching good players play
20100719 21:17:11< billynux> mordante, and looking at server.cpp:788 you can see it only figures out a client has disconnected from a trown exception
20100719 21:17:13< mordante> billynux, true but maybe you don't close the socket the way the server expects
20100719 21:17:13< boucman> Unnheulu: neither of us are good players :P
20100719 21:17:19< Unnheulu> Heh
20100719 21:17:38< gabba> Unnheulu: I'm not a good player by any stretch of the imagination (well, maybe in campaigns...) and the game might crash, but you're welcome
20100719 21:17:50< boucman> gabba: I can't really answer your question, I'm not sure how messy your code is at this point, it's your call
20100719 21:18:14< boucman> Unnheulu: since we're playtesting a feature you might even join in a triangle game if you want
20100719 21:18:36< timotei> Unnheulu: you want to be the 'tester'?:P
20100719 21:18:42< billynux> mordante, I don't think so... the server doesn't handle sockets really
20100719 21:18:56< billynux> mordante, socket is a typedef to a unsigned int for them
20100719 21:19:07< billynux> an ana::net_id to me
20100719 21:19:13< mordante> ok
20100719 21:19:31< mordante> and what if the server and the clients all run ANA, does it then work as expected 
20100719 21:19:32< billynux> I'm guessing I'll have to throw network::error( " ... " , client_id ); somewhere
20100719 21:19:38< mordante> (my compiler still runs)
20100719 21:20:07< billynux> mordante, I think they would
20100719 21:20:21-!- sevis [~sevis@188.91.77.218] has joined #wesnoth-dev
20100719 21:20:23< billynux> mordante, the problem is the way server.cpp expects disconnections to be informed
20100719 21:20:57< billynux> so... I'm guessing my handle_disconnect should put the socket descriptor (client_id) in some queue somewhere
20100719 21:21:13< billynux> and throw an exception next time some other method is called from server.cpp
20100719 21:21:19< billynux> (like receive_data)
20100719 21:21:45< billynux> mordante, needless to say: I disapprove :)
20100719 21:22:50< billynux> a client disconnection is not a program exception, attempting to read/write to a non existent client is
20100719 21:23:13-!- Upthorn [ogmar@adsl-75-26-189-133.dsl.scrm01.sbcglobal.net] has quit [Ping timeout: 240 seconds]
20100719 21:23:38< mordante> depends how you look at exceptions ;-) you can use them as message system as well
20100719 21:24:09< billynux> I know you can, that doesn't mean I approve that usage :P
20100719 21:24:37< mordante> I use them for that in the gui system to pass up a resize request up the callstack
20100719 21:25:00< billynux> mordante, you can see what I think proper usage should look like at src/ana/apps/chat/server.cpp:85
20100719 21:25:09< mordante> and I think it's one of the places where they're better as the alternative; moving up error codes
20100719 21:25:30< billynux> mordante, yes... some cases will look cleaner that way
20100719 21:26:10< mordante> it's a tool, just like goto's wonderful when used properly, but easy to turn things into a mess
20100719 21:26:34< billynux> mordante,  in this case, the network.hpp API should provide an accept_disconnection method (the same way it has an accept_connection)
20100719 21:27:08< billynux> mordante, that will look a little cleaner
20100719 21:27:20< CIA-87> gabba * r44393 /trunk/src/whiteboard/manager.cpp: Whiteboard: highlight planned actions even if whiteboard is inactive. Allows reordering at all times.
20100719 21:28:22< boucman> gabba: at this point, activating WB should be more or less equivalent to switching the effect of the ctrl key, shouldn't it ?
20100719 21:29:00< boucman> eleazzaar: ok, macros have need rearenged to handle that case... just need to test it :)
20100719 21:29:04< gabba> boucman: ok, we can play a game with this revision^, but by all means avoid giving a second order to a unit that already has one planned action. The ctrl key works well now, so I suggest we play with the wb "off". Turn it back off if it turns on by itself. Also, run wesnoth under debugger with the --log-info=whiteboard option. Unnheulu you're welcome to join.
20100719 21:29:13< CIA-87> eleazar * r44394 /trunk/data/core/images/terrain/bridge/ (9 files): more sliced and diced bridges for boucman.
20100719 21:29:38< Unnheulu> Ahm, I'm compiling r44392
20100719 21:29:42< boucman> eleazzaar: great, i'll wire them during the other's turns :)
20100719 21:30:39< Unnheulu> So watching'll be difficult ;)
20100719 21:30:54< boucman> gabba: you sure you don't want us to have WB activated ?
20100719 21:31:00< gabba> Unnheulu: bah there's only one source file different, you could quickly compile the next revision
20100719 21:31:09< boucman> Unnheulu: the feature we test won't change anything visually for observers...
20100719 21:31:30< Unnheulu> boucman, I was under the impression you could only see games on the same revision as you?
20100719 21:31:34< gabba> boucman: yes, I want to test gameplay with the ctrl key... you think it's a bad idea?
20100719 21:32:02< boucman> gabba: no, it's just annoying since I planned to use WB more or less all the time
20100719 21:32:10< boucman> but that's the burden of beta-testers :P
20100719 21:32:26< gabba> Unnheulu: true, you won't go out-of-sync
20100719 21:32:26< boucman> Unnheulu: no, you can use another rev, it might cause problems, but you can
20100719 21:32:41< Unnheulu> aethaeryn lied :/
20100719 21:32:43< gabba> boucman: your call, then play with wb and only use ctrl for direct execution
20100719 21:32:50< gabba> I'll test the opposite
20100719 21:32:52< Unnheulu> I've been ritually compiling new revisions just for him
20100719 21:33:06< boucman> gabba: nah, i'll play a couple of turns wo WB then switch back
20100719 21:33:41< boucman> ok, i'm there
20100719 21:33:51< boucman> Unnheulu: do we wait for you, or will you observe ?
20100719 21:33:57< Unnheulu> boucman, I'll observe
20100719 21:34:02< gabba> boucman: err it doesn't build here
20100719 21:34:18< boucman> in that case I seriously advise going to stable, there are much better players there...
20100719 21:34:36-!- dtiger [~dtiger@dynamic-vpdn-91-149-132-232.telecom.by] has joined #wesnoth-dev
20100719 21:34:39< gabba> undefined reference to 'editor::left_button_function', and similar errors
20100719 21:34:39< boucman> you are welcome of course, but it won't be interesting
20100719 21:34:56< Unnheulu> Hah
20100719 21:34:57-!- Crab_ [~Crab_@wesnoth/developer/crab] has joined #wesnoth-dev
20100719 21:35:11< Unnheulu> I'm on stable at the moment
20100719 21:35:11< boucman> it compiles fine for me... @44394, no local changes
20100719 21:35:16< Unnheulu> Playing peteporty
20100719 21:36:23< mordante> boucman, regarding the haloes maybe the halo code should keep track of the render/unrender state and when unrendered directly delete the halo
20100719 21:36:44< mordante> not ideal but at least keeps the quirks in the halo code
20100719 21:37:04< boucman> mordante: that was too big a change for me... silene seems also to think that my hack should be fixed but I don't know the area enough to do it myself
20100719 21:37:13< gabba> boucman: same rev, no local changes, but doesn't link... wtf?
20100719 21:37:17< CIA-87> espreon * r44395 /branches/1.8/po/wesnoth-low/CMakeLists.txt: Ran umcpropfix.
20100719 21:37:30< boucman> gabba: erase all objects and try again, maybe...
20100719 21:37:36< silene> mordante: that was my suggestion too
20100719 21:37:41< Unnheulu> Ok, make installing latest rev
20100719 21:37:44< gabba> oh noes
20100719 21:37:49< mordante> silene, ok I missed that 
20100719 21:37:53< gabba> it's gonna be loooonnng...
20100719 21:38:16 * gabba does 'make clean' and cries
20100719 21:38:40 * Unnheulu feels bad for gabba
20100719 21:38:48< billynux> mordante, "error: ‘_’ was not declared in this scope" ->  when trying to throw network::error(_("Client disconnected"),id);
20100719 21:39:25< CIA-87> espreon * r44396 /branches/1.8/doc/manual/ (38 files in 7 dirs): Ran umcpropfix.
20100719 21:39:30< billynux> ah... the dreaded complete recompilation ...
20100719 21:39:43-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20100719 21:39:45< mordante> the _ is in the gettext.hpp header
20100719 21:40:11< CIA-87> espreon * r44397 /branches/1.8/doc/design/gui2.tex: Ran umcpropfix.
20100719 21:42:01< CIA-87> espreon * r44398 /branches/1.8/data/campaigns/ (9 files in 7 dirs): Ran umcpropfix.
20100719 21:43:02< gabba> boucman: exact same error after a full recompile... grrmmbl
20100719 21:43:20< boucman> ach
20100719 21:43:28< boucman> that's really weird...
20100719 21:43:29< gabba> you use cmake too?
20100719 21:43:32< boucman> scons
20100719 21:45:30< CIA-87> espreon * r44399 /branches/1.8/data/core/images/ (5 files in 3 dirs): Ran umcpropfix.
20100719 21:46:29< gabba> boucman: ah, I build with editor off, I bet someone broke that
20100719 21:46:42< mordante> billynux, do you have local changes?
20100719 21:47:06< CIA-87> espreon * r44400 /branches/1.8/images/misc/l10n/zh_CN/logo.png: Ran umcpropfix.
20100719 21:47:06< billynux> mordante, yes (but where you talking to gabba?)
20100719 21:47:37< mordante> billynux, no to you ;-) I try to reproduce your problem, but not sure how to reproduce it
20100719 21:47:44< Unnheulu> Bah, I just make install 44399 and 44400 is committed
20100719 21:47:59< Unnheulu> One day /me will try and code wesnoth
20100719 21:48:10< gabba> boucman: in generate_report.cpp, there are lines such as "editor::selected_terrain.empty()", I bet they introduce a dependency on the editor in the game, so building w/o editor is now impossible I think
20100719 21:48:25< mordante> Unnheulu, today is a good day to try ;-)
20100719 21:48:34-!- Johannes13_ [~Johannes@unaffiliated/johannes13] has joined #wesnoth-dev
20100719 21:48:44< mordante> gabba, you build without the editor?
20100719 21:48:49< gabba> mordante: yes
20100719 21:48:54< Unnheulu> mordante, not really, only 3 of my 5 patches were committed into xmoto (my c++ code is still kinda ugly)
20100719 21:48:59< gabba> and without translations, too
20100719 21:49:01< billynux> mordante, the disconnection issue, as follows: Start an ana wesnothd, connect an sdl client and then quit (from the lobby) on it, reconnect with it and you'll see the old client
20100719 21:49:15< mordante> gabba, I guess indeed that that's not much tested
20100719 21:49:34< billynux> Unnheulu, you coded on xmoto? I played that a lot
20100719 21:49:34< gabba> mordante: well, it's the first time it breaks
20100719 21:49:40< mordante> billynux, ah ok I tried with the ana client on the SDL server
20100719 21:49:45< Unnheulu> billynux, nothing in the latest release
20100719 21:50:08< gabba> boucman: anyways, rebuilding with the editor for now, so we can play
20100719 21:51:13-!- Johannes13 [~Johannes@unaffiliated/johannes13] has quit [Ping timeout: 240 seconds]
20100719 21:52:40< boucman> afk 10'
20100719 21:53:09< billynux> mordante, ok, I fixed it
20100719 21:53:34-!- Blueblaze [~Blueblaze@adsl-76-202-21-28.dsl.hstntx.sbcglobal.net] has quit [Ping timeout: 276 seconds]
20100719 21:54:53< mordante> billynux, ok cool, what was the problem?
20100719 21:55:47< CIA-87> billynux * r44401 /trunk/src/ (4 files in 2 dirs): Added a disconnected ids queue to the ana manager to inform disconnections to the server implementation.
20100719 21:55:48< billynux> mordante, what I said. No problem, just that I wasn't throwing the exception that tells the server implementation that a client has disconnected
20100719 21:55:53< billynux> mordante, ^ :)
20100719 21:56:04< gabba> silene: looks like you broke the no-editor build with r44381, it would be nice if you could get it working again, it saves me some building time...
20100719 21:56:20< billynux> mordante, and my disconnection handler is not called from the server itself, so I had to create a disconnected ids queue
20100719 21:56:29< silene> gabba: will do; i didn't even know it was possible to compile without the editor
20100719 21:57:00< billynux> mordante, to which I push an id every time a client disconnects and then added a method that would check for something there and throw the exception from some of the methods called from the server implementation
20100719 21:57:01< gabba> silene: I found the option in ccmake, I guess it's just a define...
20100719 21:58:03< billynux> mordante, next bug: I can't connect an ana client to an ana server
20100719 21:58:06< billynux> :S
20100719 21:58:33-!- Unnheulu_ [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100719 21:58:37< boucman> back...
20100719 21:58:54< gabba> boucman: I'm starting a game
20100719 21:58:59< boucman> k
20100719 21:59:25 * Unnheulu_ is make installing
20100719 21:59:28< mordante> billynux, :-)
20100719 21:59:39< gabba> boucman: don't forget to run with --log-info=whiteboard
20100719 21:59:46 * boucman never make install
20100719 21:59:52< CIA-87> silene * r44402 /trunk/src/generate_report.cpp: Hopefully fixed linking when the editor is disabled.
20100719 21:59:52-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Disconnected by services]
20100719 21:59:56< boucman> oops, restarting :P
20100719 22:00:00 * gabba never make install either
20100719 22:00:05-!- Unnheulu_ is now known as Unnheulu
20100719 22:00:16< Unnheulu> boucman, where's the executable saved?
20100719 22:00:26< boucman> root of your checkout
20100719 22:00:47< Unnheulu> ./we<tab> isn't showing me anything...
20100719 22:01:06 * Unnheulu investigates
20100719 22:01:09< boucman> weird... what build system do you use ?
20100719 22:01:20< Unnheulu> Autotools
20100719 22:01:29< boucman> hmm
20100719 22:01:39 * boucman thought that one didn't work anymore :P
20100719 22:01:47< Unnheulu> Heh
20100719 22:01:57< Unnheulu> It doesn't in 1.8branch, but appears to in trunk
20100719 22:02:05< mordante> it does silene uses it
20100719 22:02:07< gabba> so, 2p map?
20100719 22:02:12< boucman> yes
20100719 22:02:22< silene> Unnheulu: in src/
20100719 22:02:25< Unnheulu> Anyhow, make installing 44402
20100719 22:02:28< Unnheulu> silene, thanks
20100719 22:02:30< loonycyborg> Unnheulu: It works. Though you'd need to use 'src/wesnoth .' instead.
20100719 22:02:32< billynux> mordante, can you repoduce this bug: An ana client won't connect to the ana server, it stays on "Reading from Server..." at 0%
20100719 22:03:02< gabba> boucman: game's up
20100719 22:03:57 * Unnheulu can't connect to the server
20100719 22:04:15< Unnheulu> Ah connected now
20100719 22:05:03< mordante> billynux, seems to work here
20100719 22:06:14< billynux> mordante, :( both ana binaries work fine?
20100719 22:06:25< billynux> I hate it when errors are not consistent
20100719 22:07:58< mordante> but I run a slightly older binary so might be a recent change, but then I need to recompile
20100719 22:08:17< mordante> (just discovered I forgot to update properly the last time)
20100719 22:08:38< billynux> mordante, ok, but do tell my to which revision does your binary belong to :P
20100719 22:09:23< mordante> too late I just updated the checkout ;-)
20100719 22:09:35< mordante> and no version numbers for a git checkout
20100719 22:10:49-!- rhyging5 [553a2ad3@gateway/web/freenode/ip.85.58.42.211] has joined #wesnoth-dev
20100719 22:10:55< billynux> mordante,  ok, np, anyway I just tried connecting the ana client to the sdl server and I got a "failed to uncompress WML" error, the second time worked, the last thing it sent me was the last thing the ana server sent me (but somehow I didn't get to the lobby there)
20100719 22:11:07-!- Upth [ogmar@adsl-75-26-189-133.dsl.scrm01.sbcglobal.net] has joined #wesnoth-dev
20100719 22:11:07-!- Upth is now known as Upthorn
20100719 22:11:56< rhyging5> hhiii
20100719 22:12:54< Unnheulu> Hiah
20100719 22:13:01< boucman> hey rhyging5
20100719 22:13:06< Unnheulu> rhyging5, has anyone told you your nick sounds welsh?
20100719 22:13:35< billynux> mordante, and then, when joining a game created by another ana client (on my local SDL server) I got an "invalid id: random" error
20100719 22:16:02< billynux> mordante, and after I got a game started (two ana clients on a local sdl server) I got the out-of-sync error again
20100719 22:18:12< mordante> hmm odd
20100719 22:18:55< billynux> mordante, anyway, I want to fix the ana client + ana server connection error
20100719 22:19:04< mordante> I understand
20100719 22:19:39< billynux> mordante, The client receives the lobby data fine, but sometimes it shows the lobby and sometimes it just stays there looking like its still trying to read
20100719 22:20:01< mordante> so the problem seems to be random
20100719 22:20:09< mordante> might explain why it works for me
20100719 22:21:15< CIA-87> ivanovic * r44403 /trunk/data/core/about.cfg: updated noys credits update as requested
20100719 22:21:18< CIA-87> ivanovic * r44404 /branches/1.8/data/core/about.cfg: updated noys credits update as requested
20100719 22:28:04< billynux> mordante, How do you propose I debug that? I receive exactly what I'm supposed to receive... but I don't show the lobby :S
20100719 22:28:21< CIA-87> eleazar * r44405 /trunk/data/core/ (52 files in 3 dirs): Made interior of cave wall pure black. This is a bit of a hack-job so there are glitches, but it will work untill i redo the whole cave wall.
20100719 22:30:39< mordante> billynux, hmm that's really odd, not sure what the best approach would be
20100719 22:30:54< billynux> indeed :(
20100719 22:32:47< billynux> hm... it might just be that I didn't receive a [join_lobby] message before the [gamelist][/gamelist][user]...
20100719 22:33:25< billynux> mordante, it appears this is the case: ^
20100719 22:33:53< Unnheulu> OOS :)
20100719 22:35:41< Unnheulu> In the notification box when someone talks, I assume that its a bug that /me doesn't work?
20100719 22:35:53< timotei> fendrin: around?
20100719 22:36:24< mordante> billynux, might be but I would expect message to arrive in the same order as being send
20100719 22:38:19< billynux> mordante, yes, it's a buffer queue I use, but maybe it wasn't sent
20100719 22:39:41< boucman> gabba: segfault
20100719 22:39:48< boucman> no idea what I did this time...
20100719 22:39:52< boucman> backtracing now
20100719 22:40:15< gabba> ok
20100719 22:40:45< mordante> billynux, puzzling
20100719 22:40:48< boucman> http://wesnoth.pastebin.com/ckyXv0y8
20100719 22:40:55< mordante> but I'm about to leave
20100719 22:40:59< boucman> I have gdb running if you want any extra info
20100719 22:41:17< gabba> can I have your log as well?
20100719 22:41:21-!- sevis [~sevis@188.91.77.218] has quit [Quit: leaving]
20100719 22:41:51< billynux> mordante, ok, I'll keep debugging it, it's hard to figure out where/what to test
20100719 22:42:08< boucman> http://wesnoth.pastebin.com/w7zAyapw
20100719 22:42:19< billynux> mordante, if there is a problem in the ana server implementation, it should be the case that the sdl server sometimes doesn't join either
20100719 22:42:25< Unnheulu> boucman, I seem to recall having a segfault at the end of a mp game, where I'm the only person left, although that was in stable and not using the whiteboard ;)
20100719 22:42:32< mordante> I understand, I hope to be able to test again tomorrow
20100719 22:42:55< gabba> boucman: I can see it happened when you executed a move, at least
20100719 22:43:15< billynux> ok
20100719 22:43:29< Unnheulu> gabba, I executed a move towards the end to...
20100719 22:43:53< mordante> good luck bug hunting
20100719 22:44:35< boucman> yes, but nothing special about it afaict, I was moving to a village I owned
20100719 22:44:48< boucman> I don't think it would have changed fog or anything
20100719 22:45:05< gabba> boucman: do you have the console output/log file?
20100719 22:45:25< boucman> isn't that my second pastebin ?
20100719 22:45:39< mordante> night
20100719 22:45:40< gabba> boucman: sorry, didn't see that link
20100719 22:45:48-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Quit: Ex-Chat]
20100719 22:46:09-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20100719 22:47:26-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has joined #wesnoth-dev
20100719 22:48:38< gabba> boucman: ok, I spotted a small weirdness with a return value thanks to this, but no obvious cause for the crash... weird
20100719 22:49:24< boucman> gabba: I still have gdb opened, I can check variable values and stuff like that if you want
20100719 22:49:47-!- Unnheulu [~ieuan@cpc5-pnth2-0-0-cust800.5-2.cable.virginmedia.com] has quit [Client Quit]
20100719 22:51:37-!- dtiger [~dtiger@dynamic-vpdn-91-149-132-232.telecom.by] has quit [Remote host closed the connection]
20100719 22:51:44< gabba> I never use gdb from command line, but can you check the value of viewer_actions()->actions()->size() from within manager.cpp
20100719 22:51:50< gabba> ?
20100719 22:52:27< boucman> that's frame 6 ?
20100719 22:52:30< CIA-87> silene * r44406 /trunk/src/scripting/lua.cpp: Discarded math.random, as it is an OoS-magnet.
20100719 22:52:45< gabba> boucman: yes
20100719 22:54:45< boucman> hmm, my gdb jutsu isn't good enough it seems
20100719 22:55:29< gabba> boucman: gdb 7+ supports reverse debugging, we should probably look into it for next time... just move back a few instructions, and reproduce the crash ;)
20100719 22:56:01< gabba> boucman: ok never mind
20100719 22:56:07< boucman> haha, getting there
20100719 22:56:38< boucman> (gdb) p (*viewer_actions()).actions_.size()
20100719 22:56:38< boucman> $8 = 18446181126268953992
20100719 22:56:50< boucman> is that a clue ? :P
20100719 22:56:51< Espreon> Heh, gdbjutsu...
20100719 22:57:20< gabba> eh ? interesting and unexpected value
20100719 22:57:49< boucman> gabba: it looks like something got negative here...
20100719 22:58:10< gabba> looks like my deque got corrupted
20100719 22:59:30< boucman> I don't remember having a "ennemy spoted" action at that point, but it could be related
20100719 23:00:42< gabba> maybe faulty code in side_actions::execute, but it's good as far as I can tell. A few unit tests would help there.
20100719 23:01:52< gabba> boucman: try again, or enough fooling around for tonight?
20100719 23:02:41< boucman> I'd say enough for tonight, you have all sorts of small issues around that could have corrupted things (in particular that ennemy spoted stuff) 
20100719 23:03:03< boucman> I won't be around for long tomorow evening, so feel free to test with someone else too
20100719 23:04:08< Crab_> gabba: I recommend looking not for 'what' has corrupted things, but, in general, what has the potential to corrupt things in there. the code should be fool-proof, if possible. 
20100719 23:04:12-!- silene [~plouf@wesnoth/developer/silene] has quit [Quit: Leaving.]
20100719 23:04:29< timotei> boucman: naruto/ninja?:P
20100719 23:04:46< gabba> boucman: yeah, I have lots of stuff to fix, and that's without the whole hex vs unit reference conundrum
20100719 23:04:55< timotei> gabba: why not trying to debug with MSVC?:-"
20100719 23:05:42< gabba> timotei: you pronounced the name of the IDE whose name you should not pronounce
20100719 23:06:10< boucman> "he who shall not be named"
20100719 23:06:11< gabba> timotei: kidding aside, I'm happy with Eclipse Helios + gdb 7, no command line in sight
20100719 23:06:21< timotei> gabba: well... I didn't pronoune VS :P
20100719 23:06:39-!- grzywacz [~grzywacz@wesnoth/developer/grzywacz] has joined #wesnoth-dev
20100719 23:09:24< gabba> Crab_: I do try hard to have fool-proof code. In this particular case a smart pointer freaks out on deletion though, and I really have no clue why.
20100719 23:09:35-!- rhyging5 [553a2ad3@gateway/web/freenode/ip.85.58.42.211] has quit [Quit: Page closed]
20100719 23:09:47< gabba> Crab_: btw maybe you have an opinion on my current dilemma
20100719 23:09:56< boucman> daaamit :P
20100719 23:10:00< boucman> eleazzaar: still around ?
20100719 23:10:44< boucman> gabba: you could try to add some heavy logging in the faulty area to have more clues next tile we crash
20100719 23:11:24< gabba> boucman: good idea too
20100719 23:12:33< boucman> gabba: btw I still have gdb opened if there are other things you want to know
20100719 23:12:56< gabba> Crab_: I tried to convert my whole wb code to use map_location's + unit_map instead of direct references to units, but I realized this means I need to apply the future unit map every time a move/attack needs to get a reference to its acting unit. This can pose problems since I can't modify the unit map while the future state is applied. Any advice?
20100719 23:13:14< gabba> boucman: hmmm...
20100719 23:13:28-!- DesertPanther_ [~Khalid@41.234.232.138] has left #wesnoth-dev ["Leaving"]
20100719 23:15:02< billynux> I'm off, night
20100719 23:15:03-!- billynux [~billy@wesnoth/developer/billynux] has quit [Quit: Leaving]
20100719 23:15:30< Crab_> gabba: in general, many places of the code work with locations instead of units because wml events tend to change units upon actions, but those are 'expected' changes.
20100719 23:16:32< eleazzaar> boucman: i'm back
20100719 23:17:46< boucman> eleazzaar: we still miss some hexes :(
20100719 23:17:56< Crab_> gabba: and you situation is quite different, I think...
20100719 23:18:08< boucman> single hex: normal on one end, broken on the other
20100719 23:18:29< boucman> and small on one end, broken on the other
20100719 23:18:58< eleazzaar> boucman: you mean for the 1hex bridges?
20100719 23:19:03< boucman> eleazzaar: Ill finish testing and Ill commit what I have as is, we'll see about adding these once I'm done, it will make less renamings
20100719 23:19:05< boucman> eleazzaar: yes
20100719 23:19:54< gabba> Crab_: I have the same potential problem with WML events, but I need to validate everything everytime they happen, so I guess I can rely on validation to make sure my pointer or ref is valid.
20100719 23:20:53< timotei> gabba: doesn't it work with writing std::cerr<<(testcondition) on several places?
20100719 23:21:11< CIA-87> thespaceinvader * r44407 /trunk/ (4 files in 3 dirs): Add new Orc Leader alternate portrait. Update changelogs, portrait credits.
20100719 23:21:12< timotei> gabba: that is, blind debugging at first, until you 'locate' the error
20100719 23:22:04< eleazzaar> boucman: no, we have all 12 possible variantions of the 1 hex bridge
20100719 23:22:31< Crab_> gabba: that is hard problem in general, since unit identity can be changed by wml events (think about dehydration/turn-into-undead/etc stuff )
20100719 23:23:08< gabba> timotei: wesnoth has a much better log system than std::cerr, and yes I use it quite a bit... just not much in the area where the crash happened
20100719 23:23:22< eleazzaar> i.e.: stonebridge-senw-both.png, stonebridge-senw-nw.png, stonebridge-senw-se.png, and stonebridge-senw.png
20100719 23:23:35< Crab_> gabba: the question is, do you 'lose' too much actions if you discard them on unit_map::iterator being invalid ?
20100719 23:24:00< boucman> eleazzaar: there are three possible endings for stonebridge : normal, small, broken
20100719 23:24:45< zookeeper> Crab_, gabba, i haven't been following, but are you talking about what kind of changes to a unit should make the WB no longer consider it the same unit?
20100719 23:24:47< eleazzaar> broken is not relevant to 1 hex bridges
20100719 23:24:57< Crab_> zookeeper: yes, that's related
20100719 23:25:14< boucman> eleazzaar: well, if you put one hex between a castle and water, you have the single hex bridge that goes down into watter
20100719 23:25:20< Crab_> zookeeper: we're talking about a slightly different aspect of it, yes. but the true solution requires knowing the answer to this question.
20100719 23:25:40< gabba> Crab_: Right now I just delete invalid actions right away, but I do have a system in place to give them "invalid visuals"
20100719 23:26:05< zookeeper> Crab_, ok...well, my hasty opinion would be that it might be a good idea to have a new unit attribute wb_id=<unique id> very similar to id=
20100719 23:26:29< gabba> The problem is precisely that I have to adapt the code for an action that's still displayed but has an invalid unit, without the game segfaulting all over the place
20100719 23:26:35< eleazzaar> boucman: broken edges are not in trunk
20100719 23:26:54< Crab_> gabba: so, why you'd thought about switching to map locations ? can't you just drop the action ?
20100719 23:27:00< Crab_> gabba: or it drops too much ?
20100719 23:27:06< eleazzaar> broken ends i mean
20100719 23:27:10< Crab_> gabba: btw, your newest segfault, it's reproducable ?
20100719 23:27:15< zookeeper> so if the scenario author wanted to change the unit just a bit but invalidate further WB plans for it, they could wipe or change the wb_id. also if they wanted to for example replace the unit with a completely, radically different one, they could copy over the wb_id from the old unit to the new, in order to keep the planned moves.
20100719 23:27:15< eleazzaar> i'll be back in a few minutes
20100719 23:27:20< boucman> eleazzaar: good point, I have them but they arn't commited yet :P
20100719 23:28:03< gabba> Crab_: not reproductible, the conditions are too vague
20100719 23:28:34< zookeeper> but until you want to start making a fancy solution like that, i'd suggest just using the unit's id as the identifier...but maybe you already do that, i don't know
20100719 23:28:52< Crab_> gabba: what are they ?
20100719 23:28:59< zookeeper> and as i said, i haven't been following your discussion so maybe i'm just talking about something irrelevant ;)
20100719 23:29:16< boucman> Crab_: I executed a move :P
20100719 23:29:24< Crab_> zookeeper: thanks )
20100719 23:30:08< Crab_> gabba: ^ see the zookeeper's proposal, use id= for tracking. it's documented to be unique, and that is even enforced in some places.
20100719 23:30:21< gabba> Crab_: call side_actions::execute_next() basically, which works fine 99.999% of the time it seems
20100719 23:30:27< boucman> no, seriously... I didn't do anything weird afaict, I was using y to execute a planned move, nothing special about it, the target hex was free, and a unit near it means very unlikely to remove fog/discover ennemy unit
20100719 23:30:59< Crab_> ok.
20100719 23:31:30< gabba> Crab_: noted about the id() (or underlying id I guess), that was in a TODO in my code anyways
20100719 23:32:03< gabba> Crab_: well, my first motivation for switching to map_locations was to "do things correctly" and several people advised me to do so.
20100719 23:32:07< Crab_> gabba: those are different
20100719 23:32:33< Crab_> gabba: id (string,set by wml or autogenerated) is not underlying_id (size_t, present only for units on unit_map)
20100719 23:32:48< gabba> Crab_: ok, understood
20100719 23:33:20< Crab_> gabba: map locations are for situations 'we want the wml hell to break loose, and still refer to the units at their original places'
20100719 23:33:35< boucman> Crab_: and is there an easy way, given an id, to know if a unit still exists ?
20100719 23:33:57< CIA-87> espreon * r44408 /trunk/data/core/images/portraits/orcs/transparent/leader-2.png: Ran umcpropfix.
20100719 23:33:58< Crab_> gabba: you can, of course, get units by keeping the ref to the old units map and finding in it by location....
20100719 23:34:41< Crab_> gabba: that might work, too.
20100719 23:35:32< Crab_> boucman: by underlying_id, 'still exists on unit map' - yes, o(ln(units));       by  id - no fast way, O(units)
20100719 23:36:22< gabba> well I think that as long as I'm not facing the same unit in the same hex, it can kiss its planned move goodbye; what I'm not sure is about deleting said move vs marking it invalid: the question is impossible to solve w/o playtesting, and I already have the framework in place to give them "invalid" visuals
20100719 23:36:45< boucman> gabba: tricky case we havn't tested, but you might test alone if you have 5' : player A's turn, player B moves a unit to a hex, player A moves a unit to the same hex, player A executes, player A ends turn, thus invalidating player B's move...
20100719 23:38:00< gabba> boucman: already tested in my games vs the AI, action validation will detect B's move as invalid (and as of now, delete it)
20100719 23:38:51-!- Appleman1234 [~Appleman1@131.181.101.30] has joined #wesnoth-dev
20100719 23:38:53< boucman> k, good
20100719 23:39:59< eleazzaar> boucman: i'm back now
20100719 23:40:50< boucman> eleazzaar: i'm about to commit my big bridge change, it will probably require some testing...
20100719 23:41:04< eleazzaar> ok
20100719 23:43:09-!- zookeeper [~l@wesnoth/developer/zookeeper] has quit []
20100719 23:46:47< CIA-87> boucman * r44409 /trunk/ (107 files in 7 dirs): new track layout system, works for tracks, stonebridge, woodbridge, and allows easier plugin of images within a bridge structure. The basic macros are here but all the random variations etc are not.
20100719 23:47:50< boucman> eleazzaar: done
20100719 23:47:50< CIA-87> boucman * r44410 /trunk/data/core/units/humans/Mage_of_Light.cfg: revert accidental test code in MoL unit
20100719 23:52:12< boucman> eleazzaar: I'll document the macros once everything is in order...
20100719 23:52:24-!- thespaceinvader [~chatzilla@wesnoth/artist/thespaceinvader] has quit [Quit: night all]
20100719 23:57:15< CIA-87> espreon * r44411 /trunk/data/core/images/terrain/bridge/ (6 files): Ran umcpropfix.
20100719 23:57:22< timotei> ok, good night guys:)
20100719 23:57:34-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: Leaving]
--- Log closed Tue Jul 20 00:00:32 2010