--- Log opened Thu May 22 00:00:55 2014
20140522 00:04:43-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 252 seconds]
20140522 00:05:01-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 00:09:32-!- Youngoli [~Youngoli@97.65.103.250] has joined #wesnoth-dev
20140522 00:09:42-!- Youngoli [~Youngoli@97.65.103.250] has left #wesnoth-dev []
20140522 00:10:25-!- iceiceice_ [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140522 00:10:35< iceiceice_> SigurdFD: sry i dc'd earlier without realizing it
20140522 00:10:43< iceiceice_> and i had to run
20140522 00:11:05< iceiceice_> anyways i'm not sure if my subsequent comments actually added anything from looking at the log
20140522 00:11:17< iceiceice_> but y thx for reporting and following up on this
20140522 00:23:53-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20140522 00:27:37< SigurdFD> ok, np.
20140522 00:28:23< SigurdFD> iceiceice: I have posted new report for the no id error. it's #22068....
20140522 00:28:47< SigurdFD> ... much simple to test & explain, since it applies to all mp scenario start saves in mp campaigns.
20140522 00:32:42-!- prophile [~alynn@oftn/member/prophile] has quit [Quit: The Game]
20140522 00:35:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 00:39:21-!- knotwork [~markm@142.68.84.150] has joined #wesnoth-dev
20140522 00:39:21-!- knotwork [~markm@142.68.84.150] has quit [Changing host]
20140522 00:39:21-!- knotwork [~markm@unaffiliated/knotwork] has joined #wesnoth-dev
20140522 00:42:07-!- spoffy [~spoffy@152.78.175.8] has quit [Ping timeout: 252 seconds]
20140522 00:46:29-!- aquileia [2edf524c@gateway/web/freenode/ip.46.223.82.76] has quit [Quit: Page closed]
20140522 00:57:41< iceiceice_> SigurdFD: here's a new theory
20140522 00:58:06< iceiceice_> it looks like the engine isn't entirely sure what the values of [campaign] type= ... [/campaign] should be
20140522 00:58:11< iceiceice_> some parts check for "multiplayer" 
20140522 00:58:14< iceiceice_> some parts check for "mp"
20140522 00:59:49< iceiceice_> i'm not sure yet if its consequential...
20140522 01:02:44< SigurdFD> ok
20140522 01:03:28-!- ancestral [~ancestral@12.23.74.29] has joined #wesnoth-dev
20140522 01:03:40< iceiceice_> it looks like the string "hybrid" only appears in one place,
20140522 01:03:43< iceiceice_> in the mp_create menu...
20140522 01:06:34< SigurdFD> fwiw, when I made the new bug report, I retested in the 1.11.0-dev state, and the test didn't have the [campaign] or type=...
20140522 01:06:45< SigurdFD> ... it was just a string of scenarios
20140522 01:09:50< SigurdFD> Anyone here that uses codeblocks compile?
20140522 01:17:34< iceiceice_> what do you mean the test?
20140522 01:18:07< iceiceice_> so it doesnt affect only campaigns?
20140522 01:19:21< iceiceice_> i think by default the engine will assign something to campaign_type anyways...
20140522 01:23:23< SigurdFD> I mean that in 1.11.0, there was no 'mp campaign' as it is today
20140522 01:23:39< SigurdFD> mp campaigns were strings of mulitplayer scenarios at that point
20140522 01:23:51-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 240 seconds]
20140522 01:23:55< SigurdFD> they more or less functioned the same, but there was no [campaign] tag read
20140522 01:24:53-!- kex [~kex@78.157.29.205] has quit [Remote host closed the connection]
20140522 01:25:03< SigurdFD> ie, in current 1.12, this bug would occur under a 'proper' mp campaign in the mp campaign tab, or as a string of scenarios in the mp scenario tab
20140522 01:25:30-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 01:26:25-!- kex [~kex@78.157.29.205] has quit [Read error: Connection reset by peer]
20140522 01:26:39-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 01:30:02-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 01:30:43-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 240 seconds]
20140522 01:32:54< iceiceice_> i see
20140522 01:35:11< iceiceice_> gfgtdf:
20140522 01:35:22-!- groggy [~chatzilla@24-178-146-64.dhcp.spbg.sc.charter.com] has joined #wesnoth-dev
20140522 01:35:25< iceiceice_> do you know why the carryover_sides object has an rng?
20140522 01:37:11-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 01:37:50-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20140522 01:38:40< iceiceice_> hmm i guess you are linked into that one actually?
20140522 01:38:50< iceiceice_> i didnt realize this until now
20140522 01:46:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140522 01:48:02-!- gfgtdf [~chatzilla@e176191162.adsl.alicedsl.de] has joined #wesnoth-dev
20140522 01:49:36-!- kex [~kex@78.157.29.205] has quit [Remote host closed the connection]
20140522 01:51:04-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 01:55:39< gfgtdf> iceiceice_:  i'd say the carryover_sides dotn need a rng only the carryover_sides_start uses one
20140522 01:55:43< gfgtdf> don't
20140522 01:56:08< iceiceice_> y im just trying to figure out whats going on here
20140522 01:56:37< gfgtdf> iceiceice_: start of scenario saves never contain a snapshot or a replay_start
20140522 01:56:58< iceiceice_> y but code comments in some of the mp load code suggest that they should
20140522 01:57:06< gfgtdf> iceiceice_: no tehy shouldnt 
20140522 01:57:07< iceiceice_> it might not be code that is running?
20140522 01:57:10< iceiceice_> but have a look
20140522 01:57:37< gfgtdf> iceiceice_: they just contai the id of teh next scenario which should be loaded when the start of scenario save is loaded
20140522 01:57:39< gfgtdf> iceiceice_: https://github.com/wesnoth/wesnoth/blob/fdf750cbfd4466348d59c90a412e4fece2157a67/src/playcampaign.cpp#L427
20140522 01:57:46< iceiceice_> https://github.com/wesnoth/wesnoth/blob/master/src/savegame.cpp#L708
20140522 01:58:16< iceiceice_> gfgtdf: i think that function is only used when using title screen -> load button
20140522 01:58:20< gfgtdf> iceiceice_: no my above ling was wrong: https://github.com/wesnoth/wesnoth/blob/master/src/playcampaign.cpp#L435
20140522 01:58:38< iceiceice_> (the one you linked)
20140522 01:58:43< SigurdFD> iceiceice: fwiw, sp start of scenario saves don't have a snapshot or replay_start and they work fine.
20140522 01:58:45< gfgtdf> iceiceice_: ye but teh same shoudl happen in the mo seen lod button
20140522 01:58:49< iceiceice_> the whole function play_game
20140522 01:59:55< gfgtdf> was that a question ?
20140522 02:00:51< gfgtdf> iceiceice_: ^
20140522 02:01:11< iceiceice_> gfgtdf: did you click on my link?
20140522 02:01:11< iceiceice_> to savegame.cpp?
20140522 02:02:08< iceiceice_> gfgtdf: this is the mp -> load code path, you can follow it back by grepping...
20140522 02:02:10< iceiceice_> start here: https://github.com/wesnoth/wesnoth/blob/master/src/multiplayer_connect_engine.cpp#L89
20140522 02:02:17< gfgtdf> iceiceice_: hm ye that comment there is news to be but i dotn think we shoudl act differently in mp than in sp if we dont have to
20140522 02:02:33< iceiceice_> to here: https://github.com/wesnoth/wesnoth/blob/master/src/mp_game_utils.cpp#L56
20140522 02:02:43< iceiceice_> to here: https://github.com/wesnoth/wesnoth/blob/master/src/savegame.cpp#L708
20140522 02:04:02-!- sachith500 [~kvirc@112.134.88.136] has joined #wesnoth-dev
20140522 02:04:57< gfgtdf> iceiceice_: so my first try would be ignoring the  commen in your link ading teh line like here: https://github.com/wesnoth/wesnoth/blob/master/src/playcampaign.cpp#L437   somewhere here https://github.com/wesnoth/wesnoth/blob/master/src/mp_game_utils.cpp#L60 wrapped in an if
20140522 02:05:55-!- sachith500 [~kvirc@112.134.88.136] has quit [Read error: Connection reset by peer]
20140522 02:06:13< iceiceice_> y i think you are right
20140522 02:06:38< iceiceice_> i also think maybe we should change game_classification::campaign_type to an enum
20140522 02:06:45< iceiceice_> right now it will just silently fail :/
20140522 02:06:54< iceiceice_> if there is a mismatch
20140522 02:07:10-!- sachith500 [~kvirc@112.134.88.136] has joined #wesnoth-dev
20140522 02:07:38< iceiceice_> although i dont think theres necessarily a bug there right now it seems like it could be more reliable
20140522 02:07:44< gfgtdf> iceiceice_: hm i dotn really like c++ enums and as longs as teh code dealign woith it is very performnace uncritica i have no problem with strings
20140522 02:07:56< iceiceice_> y but i want it to validate that the user gave a correct value
20140522 02:08:34< iceiceice_> or that there isnt some stupid typo in the code that screws it all up in a way we cant detect
20140522 02:08:58< iceiceice_> if theres a typo in one of the main campaign types that should be like a compiler error, enum achieves that
20140522 02:09:16< iceiceice_> we have to check 'is multiplayer" like all over the place
20140522 02:09:50< gfgtdf> iceiceice_: we somehere we have to parse the string to an enum, and if we write teh function for that we coudl aswell just write teh a function taht cheks whether the strign is one of the accepted values
20140522 02:10:43< iceiceice_> thats true but i guess also i want to know that someone didnt mistype "hybrid / hybird" or something
20140522 02:10:50< iceiceice_> when writing the check
20140522 02:11:08< iceiceice_> because if its wrong we wont really be able to detect it
20140522 02:11:29< iceiceice_> and it will probabl ymean they cant load their game
20140522 02:11:41< iceiceice_> in some corner case
20140522 02:11:46< gfgtdf> iceiceice_: but i dont really care ybout strng or enum i am onyl annoyed by the fact that in c++ i have to write a enum to string function while other languages like java/C# give it to me for free. And that no problem for me as long i i dotn write it.
20140522 02:12:06< iceiceice_> ok i might write it
20140522 02:12:12< iceiceice_> but after we figure out what the actual bug is :)
20140522 02:13:34< iceiceice_> we should probably have some kind of ENUM_MAKER macro
20140522 02:13:38< gfgtdf> iceiceice_: do we stil have problem in mp taht realoging reloaded files give wrogn replay or somethign liek that ?
20140522 02:13:40< iceiceice_> or like a "smart_enum" or smth
20140522 02:13:49< gfgtdf> ye a n enum kaer marco woudl be nice
20140522 02:15:01< gfgtdf> iceiceice_: may a that also generated an overlaof of lecical_cast<sting, ENUMTYPE> automaticly
20140522 02:15:28< iceiceice_> oh, yeah that would be pretty good
20140522 02:16:19< gfgtdf> iceiceice_: problem mogth be that if we use teh enum in a class declaration we migth have to call it like class_type::lexical_cast<class_type::ENUM_TYPE>("somevalue")
20140522 02:17:02< iceiceice_> i mean i guess it doesnt even have to actually use enum under the hood
20140522 02:17:12< iceiceice_> it could be anything and use like "is_..." functions
20140522 02:17:14< iceiceice_> like we have in team
20140522 02:21:53-!- sachith500 [~kvirc@112.134.88.136] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140522 02:23:52< groggy> gambit: i mentioned a couple of days ago that janebot is not in the credits alongside the other bots. it looks like she's your baby. if i don't hear anything in a day or two, i'll add janebot myself, but in case you want to do the honors, i don't want to step on your toes.
20140522 02:24:20< Gambit> IRC bots go in the credits? :s
20140522 02:26:15< iceiceice_> gfgtdf:
20140522 02:26:19< iceiceice_> i think i'm going to add it in savegame.cpp
20140522 02:26:29< iceiceice_> beacuse,
20140522 02:26:34< iceiceice_> the else blcok you are pointing out
20140522 02:26:54< iceiceice_> is like "if [snapshot] ... else if [replay_start], else (handle start of scenario save)
20140522 02:26:55< iceiceice_> "
20140522 02:27:00< iceiceice_> and the corresponding logic is at exactly this line:
20140522 02:27:09< iceiceice_> https://github.com/wesnoth/wesnoth/blob/master/src/savegame.cpp#L708
20140522 02:27:21< iceiceice_> "if !snapshot.empty , use snapshot, else use replay start"
20140522 02:28:57< gfgtdf> iceiceice_: is "enum aa { a,b,h,};" valid ? normal yid leave out the last semicolon but my compiler doesnt complain
20140522 02:29:10< iceiceice_> idk
20140522 02:29:32< iceiceice_> http://en.cppreference.com/w/cpp/language/enum
20140522 02:29:35< iceiceice_> it might be only C++11
20140522 02:29:54< iceiceice_> idk i really have no idea
20140522 02:29:54-!- kex [~kex@78.157.29.205] has quit [Remote host closed the connection]
20140522 02:30:09< groggy> gambit: yes, at the bottom
20140522 02:31:12< gfgtdf> 	iceiceice_: to read teh scenario we have to refresh teh game_config first and we do that after teh savegae call in https://github.com/wesnoth/wesnoth/blob/master/src/mp_game_utils.cpp#L58
20140522 02:31:24< gfgtdf> iceiceice_: seem liek not :) http://stackoverflow.com/questions/5093460/how-to-convert-an-enum-type-variable-to-a-string
20140522 02:33:48< shadowm> "id leave out the last semicolon but my compiler doesnt complain"
20140522 02:33:58< shadowm> gfgtdf: Those are commas, not semicolons.
20140522 02:34:29< shadowm> I'm pretty sure the actual semicolon in that line is essential just like with other type declarations.
20140522 02:34:36< gfgtdf> iceiceice_: wrign link http://stackoverflow.com/questions/6372650/trailing-commas-and-c
20140522 02:34:44< gfgtdf> wrong
20140522 02:35:13< gfgtdf> shadowm: i meant teh semicolon.
20140522 02:35:21< gfgtdf> teh comma
20140522 02:35:46-!- Ivanovic_ [~ivanovic@frnk-5f7528ca.pool.mediaWays.net] has joined #wesnoth-dev
20140522 02:36:18< shadowm> ... Yes.
20140522 02:36:51< shadowm> After seeing the correct link it's pretty obvious.
20140522 02:38:54-!- Ivanovic [~ivanovic@frnk-4d0165aa.pool.mediaWays.net] has quit [Ping timeout: 276 seconds]
20140522 02:39:40-!- Ivanovic_ is now known as Ivanovic
20140522 02:54:33-!- mattsc_ [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140522 02:54:57-!- ancestral_ [~ancestral@12.23.74.29] has joined #wesnoth-dev
20140522 02:58:10-!- groggy [~chatzilla@24-178-146-64.dhcp.spbg.sc.charter.com] has quit [Remote host closed the connection]
20140522 02:58:31-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20140522 03:03:04-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Ping timeout: 258 seconds]
20140522 03:03:04-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 258 seconds]
20140522 03:03:05-!- mattsc_ is now known as mattsc
20140522 03:03:06-!- timotei [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20140522 03:03:56< iceiceice_> hmmm
20140522 03:04:02< iceiceice_> gfgtdf: i think it doesnt work because,
20140522 03:04:08< iceiceice_> in the game our campaign type is "multiplayer"
20140522 03:04:15< iceiceice_> but LoW has campaign type "hybrid"
20140522 03:04:26< iceiceice_> so im getting errors "cannot find scenario!"
20140522 03:04:47< iceiceice_> i will push a branch so you can see what i tried (more or less what you said)
20140522 03:07:35< iceiceice_> https://github.com/cbeck88/wesnoth/commit/309e9faf72ee5bb4722b24d92bf7e3f562434ffd
20140522 03:07:43< iceiceice_> http://pastebin.com/jdyeAASx
20140522 03:09:06-!- Netsplit *.net <-> *.split quits: Necrosporus, ancestral, knotwork
20140522 03:09:06-!- ancestral_ is now known as ancestral
20140522 03:10:34< iceiceice_> idk i'm going to take a break, i think we should ask thunderstruck for advice about this
20140522 03:14:24-!- knotwork [~markm@142.68.84.150] has joined #wesnoth-dev
20140522 03:16:01-!- irker910 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140522 03:16:01< irker910> wesnoth: sigurdfdragon wesnoth:master ba4557bc4c1e / projectfiles/CodeBlocks/wesnoth.cbp: Fix CB project file. http://git.io/ll_Khw
20140522 03:16:01< irker910> wesnoth: Chris Beck wesnoth:master 74c4ff9973f6 / projectfiles/CodeBlocks/wesnoth.cbp: Merge pull request #167 from sigurdfdragon/CB_Project_Fix_20140521 http://git.io/Rx2Bnw
20140522 03:17:10< iceiceice_> thunderstruck: ^ can you have a look at bug # 22068 when you get a chance?
20140522 03:17:26-!- iceiceice_ [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140522 03:17:39-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 03:20:45< gfgtdf> iceiceice_: http://pastebin.com/AwFxKcvd
20140522 03:21:57-!- gfgtdf [~chatzilla@e176191162.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 29.0.1/20140506152807]]
20140522 04:19:05-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 04:23:41-!- kex [~kex@78.157.29.205] has quit [Ping timeout: 252 seconds]
20140522 04:23:57-!- Samual_ [~dioteckte@xonotic/core-team/Samual] has quit [Ping timeout: 252 seconds]
20140522 04:42:33-!- Sulfur [~Miranda@p5B008640.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140522 04:55:22-!- wesbot changed the topic of #wesnoth-dev to: 1.11.14 (1.12 beta 5) is too buggy to be announced | string+feature freeze active on 1.12 | 240 bugs, 344 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140522 05:10:39-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20140522 05:13:21-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140522 05:15:21-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140522 05:46:55-!- Ivanovic [~ivanovic@frnk-5f7528ca.pool.mediaWays.net] has quit [Changing host]
20140522 05:46:55-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20140522 05:48:14-!- sachith500 [~kvirc@112.135.27.131] has joined #wesnoth-dev
20140522 05:49:34-!- cib [~cib@p5DD222D9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140522 05:49:58-!- cib is now known as Guest80850
20140522 05:50:47-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 05:57:52-!- sachith500 [~kvirc@112.135.27.131] has quit [Read error: Connection reset by peer]
20140522 06:06:14-!- sachith500 [~kvirc@112.135.27.131] has joined #wesnoth-dev
20140522 06:07:49-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 06:07:56-!- Amu is now known as Smar
20140522 06:12:45-!- kex [~kex@78.157.29.205] has quit [Ping timeout: 276 seconds]
20140522 06:14:34-!- SigurdFD [~SigurdFD@24.154.98.89] has quit []
20140522 06:30:34-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 06:32:23-!- Guest80850 [~cib@p5DD222D9.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140522 06:35:08-!- kex [~kex@78.157.29.205] has quit [Ping timeout: 265 seconds]
20140522 06:35:10-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140522 06:35:26-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20140522 06:40:43-!- Upthorn [~ogmar@108-85-89-233.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20140522 06:42:53-!- Upth [~ogmar@108-85-89-233.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 264 seconds]
20140522 06:58:52-!- irker910 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140522 07:11:10-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20140522 07:13:09-!- ancestral [~ancestral@12.23.74.29] has quit [Quit: End Transmission.]
20140522 07:25:48-!- cib [~cib@132.231.178.148] has joined #wesnoth-dev
20140522 07:26:12-!- cib is now known as Guest15401
20140522 07:41:49-!- irker204 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140522 07:41:49< irker204> wesnoth: Guorui Xi(Kevin) wesnoth:master d3529f86f5c7 / players_changelog: Add the missing single quote http://git.io/EDIaKQ
20140522 07:49:31-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 08:00:56-!- sachith500 [~kvirc@112.135.27.131] has quit [Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/]
20140522 08:04:38-!- Necrosporus_ is now known as Necrosporus
20140522 08:04:40< Necrosporus> So frogatto uses WML too?
20140522 08:05:02< zookeeper> no
20140522 08:05:21< Necrosporus> <shadowmaster> I'd love to see a WML library that could be used not only by Wesntoh and Frogatto. :(
20140522 08:06:00< Necrosporus> Some fortunes though seem quite strange, like "Meh.  -- Zookeeper"
20140522 08:07:29-!- Sulfur [~Miranda@p5B008640.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140522 08:13:51-!- Guest15401 [~cib@132.231.178.148] has quit [Ping timeout: 252 seconds]
20140522 08:18:57-!- kex [~kex@78.157.29.205] has quit [Remote host closed the connection]
20140522 08:23:51-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20140522 08:23:53-!- Coffee_irc [~david@ppp121-45-15-35.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20140522 08:24:53-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Ping timeout: 264 seconds]
20140522 08:31:54-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 08:35:30-!- Guest15401 [~cib@132.231.178.27] has joined #wesnoth-dev
20140522 08:36:12-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140522 08:36:14-!- sachith500 [~kvirc@112.135.27.131] has joined #wesnoth-dev
20140522 08:39:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140522 08:46:21< Necrosporus> I don't think cloning wesnoth was actually a good idea
20140522 08:46:52< Necrosporus> 1.12.zip is already two times smaller than I downloaded and it's just 15%
20140522 09:02:38-!- Sulfur [~Miranda@p5B008640.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140522 09:02:39-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has joined #wesnoth-dev
20140522 09:09:58-!- esr1 [~esr@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Quit: WeeChat 0.4.2]
20140522 09:35:01< Soliton> if disk space is your primary concern then that is true.
20140522 09:46:59-!- Guest15401 [~cib@132.231.178.27] has quit [Ping timeout: 252 seconds]
20140522 09:50:18-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140522 10:21:50< Necrosporus> Soliton, it turns out to not be as bad as I thought, actually it was just 1.7 GiB, not 5 as I expected
20140522 10:28:13-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140522 10:30:48-!- mjs-de [~mjs-de@f049237167.adsl.alicedsl.de] has joined #wesnoth-dev
20140522 10:33:22-!- kex [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140522 10:33:23-!- sachith500 [~kvirc@112.135.27.131] has quit [Read error: Connection reset by peer]
20140522 10:34:27< Necrosporus> cc1plus: warnings being treated as errors // ./src/attack_prediction.cpp: In function ‘void<unnamed>::conditional_levelup(std::vector<double>&, double)’: ./src/attack_prediction.cpp:1165:25: error: use of old-style cast
20140522 10:34:37< Necrosporus> I can't build latest 1.12 branch
20140522 10:34:52-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140522 10:37:47-!- sachith500 [~kvirc@112.135.27.131] has joined #wesnoth-dev
20140522 10:38:57-!- matth1askrgr is now known as matthiaskrgr
20140522 10:43:05< Soliton> turn strict compilation off.
20140522 10:45:32< Soliton> not that i can see a c-style cast anywhere in that code.
20140522 10:47:52< Soliton> maybe DBL_MIN involves a cast.
20140522 10:49:02-!- sachith500 [~kvirc@112.135.27.131] has quit [Read error: Connection reset by peer]
20140522 10:49:52-!- sachith500 [~kvirc@112.135.27.131] has joined #wesnoth-dev
20140522 11:00:56< Necrosporus> #define DBL_MIN         __DBL_MIN__
20140522 11:01:21-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 276 seconds]
20140522 11:01:22< Necrosporus> from gcc-4.5.2/include/float.h
20140522 11:01:53< Necrosporus> Soliton, I have quite old version of gcc, may it be cause?
20140522 11:02:37< Soliton> if __DBL_MIN__ is ultimately defined by gcc, sure.
20140522 11:02:58< Necrosporus> I can't see where is it defined
20140522 11:03:58< vultraz> Agh
20140522 11:04:11< vultraz> wmltools y u no work with python 3 >_<
20140522 11:05:03< Soliton> Necrosporus: looks like it was fixed for gcc in 2010. any newer gcc should work.
20140522 11:25:30< Necrosporus> I can't say I like new visuals better than 1.10'th
20140522 11:25:50< Necrosporus> Crown was better looking, also buttons
20140522 11:26:50< Necrosporus> Though terrain description is quite useful
20140522 11:28:06< vultraz> Necrosporus: what about the maps :/
20140522 11:31:30< Necrosporus> Quite complex controlling
20140522 11:31:42< Necrosporus> Though it's good to be able to see villages
20140522 11:33:22< Necrosporus> vultraz, terrain description doesn't show terrain codes anywhere
20140522 11:33:56< Necrosporus> I know :show_terrain_codes command but I'd prefer to see code directly from game
20140522 11:34:29< Necrosporus> Can you add an option to append terrain code to terrain description (maybe advanced option and opted out by default)?
20140522 11:37:33< Necrosporus> Is it OK it uses same cache and preferences directory as 1.10?
20140522 11:37:40< Necrosporus> Won't it broke my 1.10 installation?
20140522 11:38:04-!- Guest15401 [~cib@132.231.178.103] has joined #wesnoth-dev
20140522 11:41:59< Necrosporus> Compress saved games doesn't have option "xz", only gzip and bzip2, while xz is better than bzip2 both by ratio and speed
20140522 11:43:22< Necrosporus> Some advanced options do not explain options at all, just repeat their names
20140522 11:44:04< Necrosporus> disable notifications (which notifications?) I know it's perhaps DM notifications about turn change
20140522 11:46:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140522 11:46:15< Necrosporus> UnitBox theme seems unusable
20140522 11:47:22< Necrosporus> Also terrain description at the top is not redrawn, only new one is drawn above old so it's unreadable
20140522 11:51:07< Necrosporus> For new style maps: Yes, map on a wooden table is more stylish, but why not make path markers same style too (for example animated pins which are sticked into the map?)
20140522 11:54:47< Necrosporus> Also scenario title in black rectangle doesn't harmonized with this style well, I think, though it's rather knocks than real criticism
20140522 11:58:33< Necrosporus> http://tau.rghost.ru/55593614/image.png Widescreen theme is screwed up completely
20140522 12:00:46< Necrosporus> Pandora theme is screwed up some too, though not as badly as others, only default one is usable
20140522 12:01:50-!- kex [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140522 12:04:21< Necrosporus> Also leader ellipse overlaps the castle
20140522 12:08:12< Necrosporus> vultraz, so reverted back to default. There's a problem, tooltip overlaps the minimap 
20140522 12:08:56< Necrosporus> Can you make toolitip appear on the left of button so it would rather overlap main area?
20140522 12:13:03< Necrosporus> So village display shows "Village (Flat, Village)" currently, I would like a option to show rather something like "Village (Flat, Village) Gs^V"
20140522 12:15:57< Necrosporus> vultraz, thought there are positive changes too, Resistances and Movement Cost display is much more visible
20140522 12:17:21< Necrosporus> Also display shows advancements right away, basically negating need to look at unit description
20140522 12:19:10< Necrosporus> Oh, seem terrain description contain code
20140522 12:19:12< Necrosporus> That's good
20140522 12:20:04< Necrosporus> Though why not in terrain hint string too?
20140522 12:23:05-!- happygrue [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20140522 12:24:26< Aishiko> morning happygrue 
20140522 12:33:45< Necrosporus> with default theme I got several labels "Planning mode deactivated" overlapping each other
20140522 12:54:27< Necrosporus> Leader crown has strange blue pixel at left down edge
20140522 12:54:49< Necrosporus> And otherwise I like old one better, though maybe movement orbs are better
20140522 12:56:59-!- spoffy [~spoffy@152.78.175.8] has joined #wesnoth-dev
20140522 13:11:11-!- sachith500 [~kvirc@112.135.27.131] has quit [Ping timeout: 246 seconds]
20140522 13:11:50< Necrosporus> Spearman got a picture from facing north
20140522 13:12:24-!- sachith500 [~kvirc@112.134.203.18] has joined #wesnoth-dev
20140522 13:14:58-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 13:23:12< happygrue> hello Aishiko, how goes it?
20140522 13:25:45-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140522 13:28:11-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140522 13:28:23-!- kex [~kex@89.205.75.19] has quit [Read error: Connection reset by peer]
20140522 13:29:54-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20140522 13:40:15-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140522 13:41:46-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 13:51:20< Necrosporus> And why is not possible  to set defeat message text?
20140522 14:17:11< Necrosporus> Seems Urza Nalmath (second scenario of south guard repeats same message twice
20140522 14:20:29-!- sachith500 [~kvirc@112.134.203.18] has quit [Ping timeout: 246 seconds]
20140522 14:23:49-!- Guest15401 [~cib@132.231.178.103] has quit [Ping timeout: 252 seconds]
20140522 14:25:22< Necrosporus> error display: could not open image 'terrain/.png'
20140522 14:25:50< Necrosporus> Not sure if that's first or second scenario
20140522 14:28:56-!- sachith500 [~kvirc@112.134.203.18] has joined #wesnoth-dev
20140522 14:29:19< zookeeper> had the bandits torched any villages at that point?
20140522 14:29:34< zookeeper> eh, nevermind
20140522 14:29:50-!- irker204 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140522 14:31:13< Necrosporus> They did
20140522 14:32:41< zookeeper> i'll fix the message repeat in a minute
20140522 14:33:39-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 14:37:08< Necrosporus> The problem seems undead_spotted is not set to yes in prerecruit event
20140522 14:37:18-!- irker325 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140522 14:37:19< irker325> wesnoth: ln-zookeeper wesnoth:master 86fe11bb2488 / data/campaigns/The_South_Guard/scenarios/02_Proven_by_the_Sword.cfg: Fixed repeated dialogue http://git.io/oAJr7A
20140522 14:37:54< Necrosporus> I guess, this would work too
20140522 14:38:53< Necrosporus> zookeeper, do you think this code you removed was left by mistake when someone rewrote the code moving dialogs in other place?
20140522 14:39:10< zookeeper> yes, it was my mistake
20140522 14:45:33< irker325> wesnoth: ln-zookeeper wesnoth:1.12 3a0d7cdbaa49 / data/campaigns/The_South_Guard/scenarios/02_Proven_by_the_Sword.cfg: Fixed repeated dialogue http://git.io/SE1Qxg
20140522 14:47:05-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140522 14:47:23< iceiceice> AI0867: it looks like debian builds are fixed
20140522 14:48:16< iceiceice> idk whats wrong with gentoo builds, but based on console output i think they might not have xvfb or something?
20140522 14:48:21< iceiceice> http://baldras.wesnoth.org:8080/me/my-views/view/All/job/Wesnoth/148/branch=master,compiler=4.4,label=Gentoo-64/consoleText
20140522 14:56:27-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20140522 15:03:34-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Ping timeout: 258 seconds]
20140522 15:05:33-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 15:06:44< Necrosporus> error ai/actions: Return value of AI ACTION was not checked. This may cause bugs!
20140522 15:06:50< Necrosporus> Is it ok?
20140522 15:08:04< iceiceice> Necrosporus: i think it means, 
20140522 15:08:19< iceiceice> the ai didnt check if the unit was ambushed when it moved 
20140522 15:08:30< iceiceice> etc.
20140522 15:08:41< iceiceice> it just ordered the move and then forgot about it
20140522 15:10:07< Necrosporus> So it's probably a bug
20140522 15:10:14< iceiceice> it depends
20140522 15:10:32< iceiceice> so the ai can see all anyways iiuc
20140522 15:11:00< iceiceice> so it might be that who made the ai checked for this differently and didn't rely on this mechanism...
20140522 15:11:27< iceiceice> it would be a bug if the ai got confused about where its units were and got out of sync i guess
20140522 15:11:49< iceiceice> i dont know if that can actually happen, if that is the case then its definitely a bug
20140522 15:15:45< Necrosporus> iceiceice, but trying to open a non-existing file is definitly a bug
20140522 15:16:18< iceiceice> what file?
20140522 15:16:30< Necrosporus> > error display: could not open image 'terrain/.png'
20140522 15:17:56< iceiceice> where did you get that?
20140522 15:18:03< Necrosporus> South guard
20140522 15:18:11< Necrosporus> not sure if it was first or second scenario
20140522 15:18:22< Necrosporus> and I didn't see which image was missing
20140522 15:18:29-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140522 15:19:11< Necrosporus> Must be first
20140522 15:19:21< Necrosporus> Or maybe even before it
20140522 15:21:33-!- travis-ci [~travis-ci@ec2-54-82-123-254.compute-1.amazonaws.com] has joined #wesnoth-dev
20140522 15:21:33< travis-ci> [travis-ci] wesnoth/wesnoth#2750 (master - 86fe11b : ln-zookeeper): The build has errored.
20140522 15:21:33< travis-ci> [travis-ci] Build details : http://travis-ci.org/wesnoth/wesnoth/builds/25793899
20140522 15:21:33-!- travis-ci [~travis-ci@ec2-54-82-123-254.compute-1.amazonaws.com] has left #wesnoth-dev []
20140522 15:24:52-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has joined #wesnoth-dev
20140522 15:29:02< Necrosporus> So travis tries to rebuild wesnoth each time something is changed even if it's wml which doesn't affect building?
20140522 15:31:47< iceiceice> y
20140522 15:39:20< loonycyborg> And it errors sometimes too :P
20140522 15:39:48< loonycyborg> Because build stalls due to lack of cpu time or something..
20140522 15:56:00< iceiceice> loonycyborg: do you have permissions to edit our jenkins set up?
20140522 15:56:46< loonycyborg> I don't think you need particular permissions even. Just edit travis.yml file
20140522 15:56:56< iceiceice> not for travis thouh, just for jenkins
20140522 15:56:57< loonycyborg> beside committer access that is :P
20140522 15:57:11< loonycyborg> oh was distracted
20140522 15:57:12< iceiceice> baldras.wesnoth.org:8080
20140522 15:57:15-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 265 seconds]
20140522 15:57:30< loonycyborg> Yes, I do
20140522 15:57:43< iceiceice> i think our gentoo builds are messed up for a very silly reason
20140522 15:58:04< loonycyborg> What reason?
20140522 15:58:16< iceiceice> we had to include xvfb to get some kind of null display object, so that we can actually run wesnoth to run the wml unit tests
20140522 15:58:26< iceiceice> but gentoo doesnt actually have xvfb-run
20140522 15:58:32< iceiceice> http://forums.gentoo.org/viewtopic-t-973526-start-0.html
20140522 15:58:48< iceiceice> heres the console output: http://baldras.wesnoth.org:8080/me/my-views/view/All/job/Wesnoth/148/branch=master,compiler=4.4,label=Gentoo-64/consoleText
20140522 16:00:14< loonycyborg> AI0867 was working on xvfb stuff
20140522 16:01:53< loonycyborg> But according to console output it seems to have xcfb-run, it just doesn't work for some reason..
20140522 16:04:54< iceiceice> hmm ok i misinterpreted it then
20140522 16:07:49-!- Guest15401 [~cib@132.231.178.66] has joined #wesnoth-dev
20140522 16:07:58< irker325> wesnoth: Chris Beck wesnoth:master d6dfec6151fb / src/ (6 files in 3 dirs): fixup error loggins, catch parsing exceptions in editor, -p option http://git.io/gxofDg
20140522 16:08:00< irker325> wesnoth: Chris Beck wesnoth:master ca5ea804e905 / / (3 files in 3 dirs): Merge branch 'master' of git://github.com/wesnoth/wesnoth http://git.io/i42M6w
20140522 16:10:14-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has quit [Remote host closed the connection]
20140522 16:12:55-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140522 16:21:59-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 252 seconds]
20140522 16:25:24-!- iceiceice_ [~chris@207-237-132-91.ny.subnet.cable.rcn.com] has joined #wesnoth-dev
20140522 16:26:44-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Ping timeout: 265 seconds]
20140522 16:27:02-!- Sulfur [~Miranda@p5B008640.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140522 16:27:16-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20140522 16:30:04-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140522 16:32:14-!- Guest15401 [~cib@132.231.178.66] has quit [Ping timeout: 240 seconds]
20140522 16:32:43-!- sachith500 [~kvirc@112.134.203.18] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140522 16:34:35-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20140522 16:34:47< Necrosporus> zookeeper, I think final dialogue has a bug too
20140522 16:35:05< Necrosporus> Though I didn't encounter it
20140522 16:36:33< Necrosporus> No man has set foot in the elvish woods for many a year
20140522 16:36:50< Necrosporus> Is 'a' a mistake?
20140522 16:37:23-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20140522 16:37:47< Necrosporus> Or not
20140522 16:37:53< Necrosporus> I misread
20140522 16:41:25< Aishiko> I'd say not Necrosporus just because that sounds and looks ole tyme english to me
20140522 16:45:07-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20140522 16:45:33< Necrosporus> Difficulty is called normal
20140522 16:45:55< Necrosporus> But I have 269 gold in start of 3rd scenario and enemy only 70
20140522 16:46:32< Aishiko> hmmm transposed the amounts perchance?  yes yes, transposed it is
20140522 16:46:38< Aishiko> =P
20140522 16:47:15-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140522 16:51:22< iceiceice_> lol good catch Necrosporus :)
20140522 16:52:15-!- aquileia [863c5492@gateway/web/freenode/ip.134.60.84.146] has joined #wesnoth-dev
20140522 16:53:07-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140522 16:53:21< Aishiko> please note I have no real idea of why that is but that would be my guess either that or the 70 is missing a 0 at the end of it
20140522 16:53:32< thunderstruck> iceiceice_: Okay, I will look at #22068.
20140522 16:56:27< aquileia> zookeeper: Just a short question... do you think we <i>should<\i> change to 417 or did you just mean we <i>could<\i> to have less work on syncing with mainline? Rather than me an Wrathbow guessing what you meant, I may as well ask you.
20140522 16:56:47< aquileia> So, do you see 501 as a problem?
20140522 17:00:24-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Read error: Connection reset by peer]
20140522 17:03:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140522 17:05:40-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 17:08:18< zookeeper> aquileia, well i do see 501 as a bit contrived time to shoehorn in mercenaries from a faraway land, so yes, personally i think 417 is better because it has much less restrictions and need for retconning.
20140522 17:11:23< aquileia> hmm... and if we assume we wouldn't need any retconning but only restrictions?
20140522 17:11:45< aquileia> Sorry for that second question, but your answer leaves that part open
20140522 17:12:15-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140522 17:12:39< zookeeper> well i didn't really mean that they're two distinctly distinct things
20140522 17:12:48-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140522 17:13:35< zookeeper> so i don't know how i would answer that, really :p
20140522 17:15:28< aquileia> Restrictions I can live with - if I have to change details of our plot, no problem. I'd interpret retconning as creating problems for others by changing recorded history
20140522 17:16:49< Necrosporus> to find the source of this beings of darkness that we must destroy it
20140522 17:16:54< Necrosporus> Is it old english too?
20140522 17:17:00< zookeeper> well, for example the fact that eldred had contracted foreign mercenaries to do (some of) his fighting for him... is that a restriction or a retcon? i don't know.
20140522 17:17:58-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 265 seconds]
20140522 17:18:22< aquileia> zookeeper: Where's the historic document that he didn't? After all, any campaign on imperial level fills spots (of vastly different size, though)
20140522 17:18:28< zookeeper> what i meant by retcon in this case i suppose is that you have to change strong implicit assumptions about the history, like who was fighting for eldred
20140522 17:18:42< aquileia> ah, ok
20140522 17:18:59< zookeeper> no one's ever said that he _didn't_ contract mercenaries, but i think it's pretty implicit that his troops were wesnothians
20140522 17:20:15< aquileia> That's the point where I didn't see the existing implication. I just thought that somebody who backstabs his father and pays tribute to orcs will readily buy an army
20140522 17:20:16-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Read error: Connection reset by peer]
20140522 17:20:37< zookeeper> yeah, i guess
20140522 17:20:39< aquileia> So thanks for that info
20140522 17:21:26< zookeeper> but that said, i think if that became canon, it'd warrant a subtle change to HttT's intro saying that he had secretly hired mercenaries etc
20140522 17:21:28< aquileia> After all, you know several orders of magnitude more about these things than I do
20140522 17:21:50< zookeeper> so is it a retcon? i still don't know :p
20140522 17:22:18< zookeeper> but around 417 is mostly a blank slate so there wouldn't be issues like this
20140522 17:23:25< zookeeper> the only downside i see is that you can't feature characters and points of history we're familiar with from other campaigns
20140522 17:23:27< aquileia> I know... but the throne dispute, the leader of the winning army dying by the hands of his rival, ... it's just too tempting ;)
20140522 17:23:53< zookeeper> shuffle things around a bit and you can have the same thing in 417 :P
20140522 17:24:42< aquileia> history repeats itself... well, that's in fact not uncommon...
20140522 17:26:10-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 17:26:22< iceiceice_> Necrosporus: whats the full sentence/
20140522 17:26:32< iceiceice_> thunderstruck: ok thx
20140522 17:28:07< Necrosporus> error display: could not open image 'units/human-loyalists/cavalryman/cavalryman-attack.png'
20140522 17:28:42< zookeeper> aquileia, well, one familiar character you can always feature is kalenz... dunno if anyone really wants to see more of him though :P
20140522 17:29:00< aquileia> He won't be part of it, I promise
20140522 17:29:45-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20140522 17:29:45-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20140522 17:29:45-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20140522 17:30:01< mordante> servus
20140522 17:30:13< aquileia> We'll have loyalists, outlaws, trolls... perhaps some drakes, but that's it I think
20140522 17:30:27< Necrosporus> image="units/human-loyalists/cavalryman/cavalryman-attack.png~RC(magenta>red)"
20140522 17:30:27-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Read error: Connection reset by peer]
20140522 17:30:43< Necrosporus> scenario 03 of South Guard
20140522 17:30:51< aquileia> Except if we don't find a good route for the trip
20140522 17:31:18< aquileia> damn, I'm late for the choir... bye
20140522 17:31:24-!- aquileia [863c5492@gateway/web/freenode/ip.134.60.84.146] has quit [Quit: Page closed]
20140522 17:35:11-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Excess Flood]
20140522 17:35:27< Necrosporus> iceiceice_, I would lead my men south of the river to find the source of these beings of darkness that we might destroy it. For surely, ere that be done, we cannot stand against them for long.
20140522 17:36:12< iceiceice_> Necrosporus: y i think its gramatically correct
20140522 17:36:26< iceiceice_> its just archaic
20140522 17:36:32-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 17:36:50< Necrosporus> Can we say it's beta if I could find several bugs by just playing an official campaign and looking into console (at least two, one is fixed by zookeeper other is missing image)?
20140522 17:37:06< iceiceice_> Necrosporus: i couldnt reproduce the missing image
20140522 17:37:14< Necrosporus> iceiceice_, which one?
20140522 17:37:21< Aishiko> which might be the feeling that one is going for, I notice that some have a certain style to their language
20140522 17:37:26< iceiceice_> you posted about "terrain/.png"
20140522 17:37:33< iceiceice_> i tried to get that but i couldnt
20140522 17:37:38< Necrosporus> Another one here
20140522 17:37:48< Necrosporus> > error display: could not open image 'units/human-loyalists/cavalryman/cavalryman-attack.png'
20140522 17:38:18< Necrosporus> It's used in 03
20140522 17:38:44-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20140522 17:39:09< Necrosporus> iceiceice_, unfortunately wesnoth doesn't say which line was exectuted when missed image was found
20140522 17:39:29< Necrosporus> maybe there's an option to make it tell?
20140522 17:39:41< iceiceice_> what do you mean by which line 
20140522 17:39:48< iceiceice_> you want C++ output, or like, which WML line?
20140522 17:39:59< iceiceice_> you can activate more extensive logging with --log-debug,
20140522 17:40:25< Necrosporus> iceiceice_, like cannot display image 'path/to/image.png' requested by 'path/to/wml/somefile.cfg:555'
20140522 17:40:41< iceiceice_> y so wesnoth doesn't actually know those line numbers,
20140522 17:40:53< iceiceice_> it all gets squashed into one monster cfg when you start the level
20140522 17:40:54< Necrosporus> at least a WML fragment
20140522 17:41:08< iceiceice_> it might not correspond to a WML fragment, it might be in the C++
20140522 17:41:21< Necrosporus> possibly
20140522 17:41:51< iceiceice_> there's only one possible place its coming from:
20140522 17:41:52< iceiceice_> chris@chris-KLR650 ~/wesnoth-src/clone/wesnoth $ grep "could not open image" -r src/
20140522 17:41:52< iceiceice_> src/image.cpp:		ERR_DP << "could not open image '" << val_.filename_ << "'" << std::endl;
20140522 17:42:34< iceiceice_> are you playing off of the repo version?
20140522 17:42:53< Necrosporus> fc1ca38
20140522 17:43:04< iceiceice_> ok
20140522 17:45:12< zookeeper> well, whoever removed the old cavalryman frames clearly didn't check for existing usages
20140522 17:45:22< Necrosporus> i do not know where terrain/.png is coming from
20140522 17:45:31< iceiceice_> Necrosporus: one thing you can try is
20140522 17:45:50< Necrosporus> I' m not sure I could reproduce it, I met it just once
20140522 17:45:53-!- _8680_ [~8680@2002:4404:712c:0:4531:aa24:cfb2:8aba] has quit [Remote host closed the connection]
20140522 17:45:53< iceiceice_> (warning: this is fairly involved but its what i would actually do)
20140522 17:46:06< iceiceice_> build in debug mode so you get debugging symbols
20140522 17:46:09< iceiceice_> like,
20140522 17:46:11< iceiceice_> scons build=debug
20140522 17:46:15< iceiceice_> as opposed to scons build=release
20140522 17:46:26< Necrosporus> what about ccmake?
20140522 17:46:29< iceiceice_> (it will remember whcih mode you use and you dont want to play in debug mode normally)
20140522 17:46:31< iceiceice_> oh idk
20140522 17:46:38< iceiceice_> theres probably something equivalent
20140522 17:46:42< iceiceice_> anyways,
20140522 17:46:52< iceiceice_> you can go to the error message where it says "error: could not open image"
20140522 17:46:57< iceiceice_> add a line "assert(false)"
20140522 17:47:05< iceiceice_> compile that in debug,
20140522 17:47:05-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has quit [Quit: leaving]
20140522 17:47:14< iceiceice_> then do "gdb wesnoth-debug"
20140522 17:47:25< zookeeper> Jetrel, ^ the cavalryman-attack.png issue seems to be your fault :p campaigns might use whatever unit frames in some circumstances like that one, so when removing/renaming those need to be checked too.
20140522 17:47:35< iceiceice_> it will make the game crash when the error happens,
20140522 17:47:46< iceiceice_> and in gdb, you can type "bt" to get a backtrace
20140522 17:47:47< Jetrel> existing usages?
20140522 17:47:53< Jetrel> Where's it being used?
20140522 17:47:57< iceiceice_> that will tell you exactly which functions in C++ called it and got the crash
20140522 17:48:00< Necrosporus> I surmised that
20140522 17:48:07< iceiceice_> (but I guess zookeeper knew whats wrong)
20140522 17:48:30< zookeeper> i don't know anything about the terrain/.png
20140522 17:48:32-!- _8680_ [~8680@2002:4404:712c:0:a8d7:85ee:3fb2:639f] has joined #wesnoth-dev
20140522 17:48:39-!- cib [~cib@p5DD222D9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140522 17:49:07-!- cib is now known as Guest89373
20140522 17:50:45< Jetrel> zookeeper:  okay, I'll keep that in mind next time.
20140522 17:52:34< zookeeper> i'd fix it right away if 1) github for windows didn't suck and 2) tortoisegit didn't suck and 3) if the trial period for sourcetree, which i was going to really try out for the first time right now, hadn't expired :p
20140522 17:55:14< irker325> wesnoth: ln-zookeeper wesnoth:master 3a860d46571b / data/campaigns/The_South_Guard/scenarios/03_A_Desparate_Errand.cfg: Fixed broken image reference http://git.io/K8xM7Q
20140522 17:55:17 * zookeeper grumbles
20140522 17:57:10< iceiceice_> thx for reporting Necrosporus 
20140522 17:57:27< irker325> wesnoth: ln-zookeeper wesnoth:1.12 31e140d1fa09 / data/campaigns/The_South_Guard/scenarios/03_A_Desparate_Errand.cfg: Fixed broken image reference http://git.io/yU7AoA
20140522 17:59:42< Necrosporus> By the way, it's not hard to make a script grepping WML for image reference and checking if images are on place 
20140522 18:02:07< zookeeper> well someone had attempted to fix the cavalryman reference earlier because of a wmlscope report, but didn't get the fix right :P
20140522 18:02:49< Necrosporus> shouldn't it be a portrait anyway?
20140522 18:03:23 * zookeeper shrugs
20140522 18:03:39< Necrosporus> though it's narrator
20140522 18:03:40< zookeeper> when talking about a unit in the context of a gameplay hint, it seems that the sprite is more fitting
20140522 18:03:44< Necrosporus> so probably that's ok
20140522 18:04:34-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20140522 18:08:35< Necrosporus> zookeeper, why not to use just git? I'm almost sure it's available even for windows
20140522 18:08:43< Necrosporus> msysgit perhaps
20140522 18:10:58< mattsc> Does anybody else have double vision with the current master?
20140522 18:11:30< mattsc> If I open the campaign menu, I see everything twice.  Also get a lot of duplicate unit and terrain messages.
20140522 18:11:42< mattsc> This is with a clean build I just made a few minutes ago.
20140522 18:14:28< Necrosporus> mattsc, I got similar issues with 1.12, but less severe
20140522 18:15:31< mattsc> I don’t think I had them with 1.12, but I just cleaned and started rebuilding.  Will know once that is done.
20140522 18:16:03< Necrosporus> mattsc, only with non-default theme
20140522 18:16:32< Necrosporus> or with default message about planning mode was duplicated
20140522 18:16:39< mattsc> I’m using default-everything, AFAIK
20140522 18:17:42< mattsc> Anyways, I just want to know if I need to check out my user dir for corruption or if this is a common problem.  If the latter, there’s no rush as I am sure it will get done at some point.
20140522 18:18:20< iceiceice_> mattsc: i didnt notice anything wierd
20140522 18:18:25< iceiceice_> in campaign menu
20140522 18:18:50< iceiceice_> the only things ive noticed at al like this is that the 1.12 test executable sometimes complains about multiple inclusions
20140522 18:20:10< mattsc> iceiceice_: well, it’s not particularly hard to notice:  http://imgur.com/TOEU0VX
20140522 18:20:31< iceiceice_> yeah i didnt get that
20140522 18:21:19< iceiceice_> (ca5ea80 clean)
20140522 18:23:36< Necrosporus> btw there were multiple inclusion messages about my custom units in 1.11.13
20140522 18:23:39< mattsc> same here
20140522 18:24:15< mattsc> Okay, I’ll check out what happened some other time.  Have to write a conference paper right now  :P
20140522 18:25:57< mattsc> IMy 1.12 clean build just got done and I can confirm that it does not happen there.
20140522 18:29:18< mattsc> However, one more quick little test: if I remove my entire user data directory, it still happens.
20140522 18:30:19< mattsc> iceiceice_: I also get lots of these: “error config: Multiple [unit_type]s with id=Blood Bat encountered.”
20140522 18:30:39< iceiceice_> y i got alot of those also
20140522 18:30:44< iceiceice_> err
20140522 18:30:47< iceiceice_> i didnt get them on my machine,
20140522 18:30:53< iceiceice_> but i got them sometimes when i run test executable
20140522 18:30:59< mattsc> One for each mainline unit.  So, somewhere something is being duplicated, but I don’t have time to figure out what right now.
20140522 18:31:14< iceiceice_> what revision are you at
20140522 18:31:34< mattsc> same as you.  clean build of ca5ea80 made a few minutes ago.
20140522 18:31:37< iceiceice_> ok
20140522 18:31:43< iceiceice_> thats really wierd,
20140522 18:31:57< iceiceice_> suggests a filesystem issue? idk
20140522 18:31:58< mattsc> happens in debug mode and without it, and whether I start from the CL or by clicking on the app.
20140522 18:32:59< mattsc> iceiceice_: well, as I said, it’s with a clean build and a removed user data directory.  I don’t know how that could be a filesystem issue.
20140522 18:33:22< iceiceice_> what i mean is in our filesystem code
20140522 18:33:30< iceiceice_> why would it include directory twice on your machine and not mine?
20140522 18:33:31< mattsc> oh, I see.  Maybe.
20140522 18:33:55< mattsc> iceiceice_: well, when I have more time, I can do another refactoring.  It started a “few” days ago.
20140522 18:34:00< iceiceice_> ok
20140522 18:34:03< iceiceice_> so a few days ago,
20140522 18:34:11< iceiceice_> i had to restructure the test directory,
20140522 18:34:26< mattsc> I don’t remember if “a few days” means 3 days or 3 weeks though.
20140522 18:34:32< iceiceice_> ok
20140522 18:34:56< mattsc> Anyways, I really need to get on with non-Wesnoth work though, I’m already in enough trouble the way it is.  :P
20140522 18:35:00< iceiceice_> ok c u
20140522 18:35:09< mattsc> Maybe I can do the refactoring tonight.  Will let you know.
20140522 18:35:14< mattsc> cheers
20140522 18:37:42< Ivanovic> AI0867, Aishiko, elias, fabi__, iceiceice_, loonycyborg, mattsc, mordante, shadowm, Soliton, thunderstruck, everyone else who cares: what do you think of another try of a beta release this weekend?
20140522 18:37:49< Necrosporus> mattsc, there's also cache and config
20140522 18:38:02< Ivanovic> i would have done it last weekend but was visiting at home and internet was completely broken 95% of the time...
20140522 18:38:27< Necrosporus> mattsc, you could make a quick-check (just don't remove it, rename)
20140522 18:38:31< mordante> Ivanovic, I'm in favour I fixed a Mac issue, mattsc tested it, but can't hurt to get more testing
20140522 18:38:55< mattsc> Necrosporus: both of those are in my user directory, so by removing that (yes, rename is what I did), they should be taken care of as well.
20140522 18:39:01< Ivanovic> mordante: is there a changelog / release_notes entry about it explicitly asking people to verify the fix?
20140522 18:39:09< Aishiko> Ivanovic, sounds like my trip all the motel & the hotel had lousy wifi do to bad setup
20140522 18:39:15< Ivanovic> mordante: and any update on the "distorted minimap" for the pandora?
20140522 18:39:49< Ivanovic> Aishiko: there it was a huge amount of fun to talk to the ISP and make them understand that the error was on their end
20140522 18:40:02< Ivanovic> over the course of the weekend i spent about 6h talking with them and waiting in queue
20140522 18:40:23< Aishiko> Ivanovic, I know that pain, usually takes me 2 to 3 people til I get someone that can go, ohh your right
20140522 18:40:51< Ivanovic> 2 or 3 hops is nothing
20140522 18:41:08< Ivanovic> i think my record on the weekend were about 5 hops and the last one "having to ask his supervisor"
20140522 18:41:35< mordante> Ivanovic, it's (as usual) in the changelog not in the release notes, do you think it is release notes worthy? https://gna.org/bugs/index.php?22045
20140522 18:41:40< Aishiko> then it takes seconds to a couple minutes to get it fixed until then..... and its worse when its an intermittent thing but try convincing support that rebooting your router, network, and computers every 2-6 hours is not a "fix"
20140522 18:42:08< mordante> Ivanovic, no I can't reproduce the minimap issue :-(
20140522 18:42:10< mattsc> Ivanovic: I’m all for another release.  I need to fix up a few non-leader MAI test scenarios, but that’s easy if gfgtdf or iceiceice tell me that the defeat_condition syntax is now fixed.
20140522 18:42:45-!- Ivanovic changed the topic of #wesnoth-dev to: 1.11.15 (1.12 beta 5) planned for Saturday, May 24th | string+feature freeze active on 1.12 | 240 bugs, 344 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140522 18:43:04< mattsc> iceiceice_: ^  (did I actually type out your nick rather than using tab completion?  :O )
20140522 18:43:09< mordante> Ivanovic, I still hope fabi__ will organise a WESSB so I could test it with a Pandora
20140522 18:43:25< Ivanovic> :)
20140522 18:44:06< iceiceice_> mattsc: hehe, its not so hard to type is it?
20140522 18:44:10< Aishiko> mordante, maybe next year have a gsoc proposal of a pandora emulator for wesnoth =)
20140522 18:44:13< iceiceice_> mattsc: i'm writing a unit test for the defeat condition
20140522 18:44:32< iceiceice_> i dont remember if it got pushed to 1.12 already yet or not
20140522 18:44:36< mattsc> iceiceice_: as long as you don’t trip up your fingers while doing so ...
20140522 18:44:36< iceiceice_> probably it did
20140522 18:44:48< mordante> Aishiko, well I tested in qemu and that didn't show the error :-(
20140522 18:45:11< mattsc> iceiceice_: I meant “fixed” as in “fixated” (the syntax), rather than un-bugged.
20140522 18:45:21< Ivanovic> mordante: during the weekend i will probably try the updated/rebuilt libs that aquileia found
20140522 18:45:24< Ivanovic> maybe those just work...
20140522 18:45:33< Aishiko> mordante, ahh thats not good, we can't fix if we can't reproduce the error
20140522 18:45:41< Ivanovic> (or they fail completely when trying to start the application, we'll see...)
20140522 18:45:49< Ivanovic> Aishiko: oh, but i can see the error
20140522 18:45:55< Ivanovic> and it happens on every real hardware pandora
20140522 18:45:56< Ivanovic> ;)
20140522 18:47:30< mattsc> iceiceice_, Ivanovic: if the defeat_condition thing is not finished for/in 1.12, then that’s a blocker, IMO.
20140522 18:47:31< Aishiko> Ivanovic, can we get tracebacks/debug logs form someone that has one? or better yet pinpoint which commit added the error?  you know for those that could work on it but lack the hardware?
20140522 18:48:01< Ivanovic> Aishiko: honestly i have no idea for how long the error has been there
20140522 18:48:11< Ivanovic> it could have been for a year and we might not have noticed...
20140522 18:48:33< Aishiko> Ivanovic, ohhh I thought it was something newish
20140522 18:48:51< mordante> Aishiko, exactly fixing bugs you can't reproduce is often difficult
20140522 18:49:10< mordante> Aishiko, we of course can ask the poor Ivanovic to bisect the error
20140522 18:49:13 * mordante flees
20140522 18:51:14< Aishiko> Ivanovic, did you notice the error in 1.10 or a dev build?
20140522 18:51:41< Ivanovic> Aishiko: i don't remember!
20140522 18:52:45< Aishiko> Ivanovic, LOL, well then!  looks like we need more info before we go looking as if it was 1.10 then it might already be fixed!
20140522 18:52:55< mordante> I'm also not sure whether it's an ancient bug triggered by fabi__'s improvements by the editor or a new bug
20140522 18:53:25< Ivanovic> Aishiko: all i can say is that i know for sure that the issue exists in 1.11.14
20140522 18:53:40< Ivanovic> and the issue existed at the end of january
20140522 18:53:56< Ivanovic> i think sometime in january was the time when i first noticed the issue
20140522 18:54:00-!- boucman [~rosen@2a02-8428-034f-f800-8df1-95f6-fb03-0ba6.rev.sfr.net] has joined #wesnoth-dev
20140522 18:54:03< Ivanovic> but it might have been there a lot before already
20140522 18:54:07-!- boucman [~rosen@2a02-8428-034f-f800-8df1-95f6-fb03-0ba6.rev.sfr.net] has quit [Changing host]
20140522 18:54:07-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20140522 18:56:13< Ivanovic> okay, i am off now for today
20140522 18:56:15< Ivanovic> CU
20140522 18:56:22< mordante> bye Ivanovic 
20140522 18:58:04-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20140522 18:59:59-!- gfgtdf [~chatzilla@e176191162.adsl.alicedsl.de] has joined #wesnoth-dev
20140522 19:00:42< gfgtdf> Ivanovic: i stil have to backport ~4 commits to 1.12 before the next resease.
20140522 19:01:27< mordante> Aishiko, how are things going for GSoC?
20140522 19:03:27< Necrosporus> zookeeper, do I understand correctly, if you don't have your units killed, you will not get loyal elvish rangers?
20140522 19:03:35< Necrosporus> in 04
20140522 19:05:14< Aishiko> mordante, its going, I'm working on 2 things either creating a spritesheet name space or just doing it under the image namespace, and getting the WML tags in place.
20140522 19:07:22-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Ping timeout: 258 seconds]
20140522 19:09:04< mordante> Aishiko, ok is there a final specification for the WML tags?
20140522 19:09:06-!- happygrue [~happygrue@2601:6:4380:7df:854c:c904:2f9d:b438] has joined #wesnoth-dev
20140522 19:09:07-!- happygrue [~happygrue@2601:6:4380:7df:854c:c904:2f9d:b438] has quit [Changing host]
20140522 19:09:07-!- happygrue [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20140522 19:09:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 19:09:14-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has joined #wesnoth-dev
20140522 19:12:44< Aishiko> mordante, I'm thinking that the second example found at http://wiki.wesnoth.org/SpriteSheetApplicationSAB#WML would be the best only Corrdinates=[x,y,x,y] should be [x,y,h,w] basically returning the SDL_Rect with the ImageName field being optional
20140522 19:13:05< mattsc> gfgtdf: is the defeat_condition thing on of those backports?
20140522 19:13:19< gfgtdf> mattsc: yes
20140522 19:13:44< mattsc> gfgtdf: okay.
20140522 19:13:58< mordante> Aishiko, IMO it would be best to make sure what the specifications are before starting to implement them
20140522 19:14:04< Aishiko> err x,y.w,h
20140522 19:14:06< mattsc> Ivanovic: that means I have a fix to make _after_ gfgtdf does that.
20140522 19:14:09< mordante> Aishiko, just to avoid writing code twice
20140522 19:15:33< mordante> (not that I have a strong opinion on the 'perfect' format)
20140522 19:18:17-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140522 19:18:41< Aishiko> alright, however I feel that having it give back the Rect arguements in 1 line will make things easier for users, and the ImageID since we need to know which image is which, and the name being optional so that those that care can have each named for quick reference when looking at the WML config file
20140522 19:19:48< iceiceice_> gfgtdf: i looked at your enum maker macro, its pretty good
20140522 19:20:01< iceiceice_> i think one of the functions took an unnecessary argument though/
20140522 19:20:34< gfgtdf> iceiceice_: which one?
20140522 19:20:53< gfgtdf> iceiceice_: you mean marcos or the generated function ?
20140522 19:21:03< gfgtdf> macro*
20140522 19:21:13< mordante> Aishiko, as said I've no strong preference regarding the method used, however I think it is time to discuss it with your mentor to decide whether he agrees with your proposal and then decree this is they way you are going
20140522 19:21:41< iceiceice_> why does string_to_enum_name take second argument?
20140522 19:21:43< iceiceice_> that is the default?
20140522 19:21:54< gfgtdf> 	iceiceice_: ye thats teh default
20140522 19:21:57< iceiceice_> oh ok
20140522 19:21:58< mordante> Aishiko, just to avoid to have to rewrite your code
20140522 19:22:38< mordante> Aishiko, btw I really look forward to see the spritesheets in Wesnoth :-)
20140522 19:23:03< iceiceice_> gfgtdf: shouldnt it throw an exception if it fails to parse the string?
20140522 19:23:11< gfgtdf> iceiceice_: i tried make it lexical_cast<enumname> instead of string_to_enumnameut somehow i coudnt mae this working.
20140522 19:23:30< gfgtdf> iceiceice_: hm i think we have 2 version for normla lexical cast 1 thows and the other has a default argument
20140522 19:23:31< iceiceice_> gfgtdf: yeah i think that woudl be hard to macroize
20140522 19:23:37< Aishiko> mordante, me too, it'll be useful and I'll have learned a lot in the process
20140522 19:23:49< iceiceice_> because, what if you make the enum as a member of a class
20140522 19:24:02< iceiceice_> most of the time that's what you'll do i guess
20140522 19:24:25< iceiceice_> i guess you could have a second macro to go at the bottom fo the header, that takes the full identifier
20140522 19:24:52< mordante> Aishiko, :-)
20140522 19:25:29< gfgtdf> iceiceice_: hm ye or one just call classname::enumname_to_string(somevalue)
20140522 19:25:42< iceiceice_> y
20140522 19:26:35< iceiceice_> do you know what file the macro should go in?
20140522 19:26:41< iceiceice_> should we just put it in like util.hpp or smth?
20140522 19:29:01< Aishiko> mordante, the upside is happygrue and I have a meeting planed for this evening we can hammer out the details for the next week at that meeting I've got a small list of questions I've made as I've been moving along
20140522 19:29:43< mordante> good to hear
20140522 19:30:12< mordante> if there are questions you want me to answer, best send an e-mail since I'm about to leave
20140522 19:30:47< Aishiko> mordante, thank you I will
20140522 19:31:29-!- gfgtdf_ [~chatzilla@e176191162.adsl.alicedsl.de] has joined #wesnoth-dev
20140522 19:32:21< gfgtdf_> iceiceice_: updatd version: http://pastebin.com/60BKh34Z
20140522 19:33:50-!- gfgtdf [~chatzilla@e176191162.adsl.alicedsl.de] has quit [Ping timeout: 240 seconds]
20140522 19:33:55-!- gfgtdf_ is now known as gfgtdf
20140522 19:34:31< gfgtdf> iceiceice_: hm eigher in util.hpp or in in its own file enum_maker.hpp
20140522 19:35:34< iceiceice_> y
20140522 19:36:17-!- Fortescue [sid1929@gateway/web/irccloud.com/x-jmknbidhwxnaulyi] has quit [Ping timeout: 264 seconds]
20140522 19:37:42-!- Fortescue [sid1929@gateway/web/irccloud.com/x-tgvzafgxmkrxubpo] has joined #wesnoth-dev
20140522 19:38:33< Necrosporus> zookeeper, also is it OK cavalryman who got first is not loyal?
20140522 19:38:53< mordante> Aishiko, you're welcome
20140522 19:38:57< mordante> I'm off bye
20140522 19:39:36-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20140522 19:40:40< Necrosporus> it's recalled every time also
20140522 19:40:47< zookeeper> Necrosporus, yes
20140522 19:42:54< Necrosporus> I reached blackriver and game started to lag
20140522 19:43:45< Necrosporus> from the first turn before recruiting anything
20140522 19:46:31< happygrue> gfgtdf: I never actually asked, but I assume the spritesheet branch you created last month is for Aishiko's spritesheet, right?
20140522 19:46:55< gfgtdf> happygrue: i dont remeber having created a spritesheet branch
20140522 19:47:14< happygrue> oh, I'll check again.
20140522 19:48:35< happygrue> https://github.com/wesnoth/wesnoth/tree/spritesheet
20140522 19:49:04< happygrue> ah, that's just the last commit
20140522 19:50:00< happygrue> I haven't messed around with branches much yet
20140522 19:50:17-!- Fortescue [sid1929@gateway/web/irccloud.com/x-tgvzafgxmkrxubpo] has quit [Ping timeout: 245 seconds]
20140522 19:51:32-!- Fortescue [sid1929@gateway/web/irccloud.com/x-xjstgrqylvbslyto] has joined #wesnoth-dev
20140522 19:53:44< iceiceice_> hmm
20140522 19:53:57< iceiceice_> it looks like i cannot pass a list of arguments to "[kill] canrecrui=" ?
20140522 19:54:07< iceiceice_> i guess thats fine, who knew
20140522 19:54:13< zookeeper> well what would you pass? yes,no? :P
20140522 19:54:38< iceiceice_> yes :)
20140522 19:54:45< gfgtdf> iceiceice_: i dont exacly know what camapign_type does but m first thought was that specified whether to srearch for [multiplayer] or for [scenario] in teh gaem config so i'd was the bug might  be campagn_type="multiplayer" for these camoagns
20140522 19:54:51< iceiceice_> it makes the macro a bit easier
20140522 19:55:14< zookeeper> iceiceice_, right... well, i doubt binary keys like that tend to support lists
20140522 19:55:25< iceiceice_> ok makes sense
20140522 19:55:35< zookeeper> what macro would that be, anyway?
20140522 19:55:53< iceiceice_> it was just for this test of check_Victory
20140522 19:56:13< iceiceice_> i have a macro that builds like 20 scenarios with different configurations of "defeat_conditions" and different units vs. leaders being killed in various configruations...
20140522 19:56:18< iceiceice_> so i wanted to just be able to say 
20140522 19:56:24< iceiceice_> {KILL_ 2,3 yes,no}
20140522 19:56:25< zookeeper> you could see what happens if you just pass an empty string?
20140522 19:56:27< iceiceice_> to kill all of 2 and 3
20140522 19:56:30< iceiceice_> y empty string odesnt work
20140522 19:56:34< zookeeper> okay
20140522 19:56:35< iceiceice_> it gives some default value
20140522 19:56:43< iceiceice_> maybe empty string should just be "ignore this attribute" ?
20140522 19:57:21 * zookeeper shrugs
20140522 19:57:24< iceiceice_> y i have no idea
20140522 19:57:36< iceiceice_> probly backwards compat is more important
20140522 19:58:24< zookeeper> so... what does the y stand for? "yes"?
20140522 19:58:31< zookeeper> you seem to use that a lot :p
20140522 19:58:51< iceiceice_> yeah :)
20140522 19:59:17< zookeeper> all righty
20140522 20:00:00< zookeeper> it was funny once when in some game someone asked what the key for teamchat was
20140522 20:00:06< iceiceice_> haha
20140522 20:00:06< zookeeper> and i said y
20140522 20:00:10< zookeeper> (because that's what the key was)
20140522 20:00:25< zookeeper> and he was all "dude i just want to know what is it"
20140522 20:00:26< iceiceice_> oh i misread
20140522 20:00:33< iceiceice_> i thought you said "whats the key for chat"
20140522 20:00:34< zookeeper> and i just kept repeating y
20140522 20:00:53< zookeeper> and he was getting all pissed off
20140522 20:00:59< iceiceice_> i dont think i even knew there was a key for teamchat, i always just use that little checkbox
20140522 20:01:30< zookeeper> well this wasn't wesnoth anyway
20140522 20:01:49< zookeeper> and i'd imagine there must still be a key for teamchat even if the checkbox is there... i hope
20140522 20:03:02< zookeeper> i dunno how many times i had to tell him "y" before he got that i wasn't just annoying him by asking why he wants to know
20140522 20:03:37< iceiceice_> haha
20140522 20:05:41-!- Fortescue [sid1929@gateway/web/irccloud.com/x-xjstgrqylvbslyto] has quit [Ping timeout: 264 seconds]
20140522 20:08:07-!- Fortescue [sid1929@gateway/web/irccloud.com/x-gvcaeysdjssordwq] has joined #wesnoth-dev
20140522 20:16:13< zookeeper> mattsc, err, did we have some specific plans for what to do with the AI controller? http://forums.wesnoth.org/viewtopic.php?f=3&t=40477
20140522 20:16:29< zookeeper> all i remember is that we discussed it, but no idea what came of it
20140522 20:21:36< mattsc> zookeeper: umm, that thread is full of misconceptions …
20140522 20:22:08< happygrue> what!?  On our forums!!
20140522 20:22:13< mattsc> zookeeper: we (you, I think) made some minor changes to the AI controller, but nothing significant
20140522 20:22:30< mattsc> We did discuss making it work for sides without leaders, but decided against it.
20140522 20:23:19< mattsc> zookeeper: all I can say, the AI controller works just fine based on all tests I have done.  It’s just that people don’t understand what it actually does.
20140522 20:24:12< mattsc> (well, there were a couple things that didn’t work, but they were minor and are fixed now)
20140522 20:24:39< zookeeper> mattsc, but if his report is accurate then obviously there's a problem, no?
20140522 20:25:00< mattsc> zookeeper: no  :P
20140522 20:25:13< zookeeper> i mean, it seems rather reasonable to expect that "be defensive" means "don't suicide your leader"
20140522 20:25:20< mattsc> The instruction to the AI is to be _more_ defensive.  It is not: never attack.
20140522 20:25:58< zookeeper> sure, but his report describes the leader being completely reckless
20140522 20:26:11< mattsc> zookeeper: it sounds like a reasonable assumption, but nowhere does it say that the AI controller does that.
20140522 20:26:28< mattsc> assumptions are dangerous, in particular when it comes to assumptions about what the AI might do.
20140522 20:26:32< mattsc> :D
20140522 20:27:19< mattsc> zookeeper: I’d agree that what you say would be nice.  You go ahead and code that then.  :P
20140522 20:27:51< zookeeper> assumptions are dangerous but if you can't rely on reasonable assumptions then you might as well not play :p
20140522 20:28:03-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140522 20:28:44< happygrue> Is it possible to give just leaders the most defensive and timid settings available, like "always run away", or just make them not move perhaps?
20140522 20:28:57< happygrue> I wonder if the AI would do better if the leaders just didn't move
20140522 20:29:30< happygrue> though some small chance that they could come attack would be best, it also opens the door to them standing in water somehow...
20140522 20:29:44< Aishiko> well you could set the AI's unit to a movement of 0 so they can't move
20140522 20:30:01< happygrue> and many UMC things do this for that reason
20140522 20:30:13< mattsc> happygrue: it really depends.  In my Freelands AI, I very much rely on the leader coming out protecting villages or dealing lots of damage). 
20140522 20:30:16< mattsc> It
20140522 20:30:17< happygrue> but it's also not ideal in other wyas
20140522 20:30:21< happygrue> *ways
20140522 20:30:32< happygrue> yes, that is also true
20140522 20:30:33< mattsc> ’s all a balancing act and it’s anything but trivial.
20140522 20:30:49< happygrue> yeah
20140522 20:30:58< happygrue> I was just musing out loud
20140522 20:31:16< happygrue> in scenarios where I KNOW the leader won't move it is much easier to plan how to attack the leader
20140522 20:31:18< zookeeper> mattsc, i don't really understand why the leader would choose such a risky melee attack. if leader_aggression is -4 by default, then i'd imagine it'd have deterred it from attacking... i didn't do the math and calculate the potential max damage ratios though
20140522 20:31:21< happygrue> so that has major flaws too
20140522 20:31:23< mattsc> I don’t disagree at all that the behavior pointed out in that thread is very bad and something should be done about it.
20140522 20:31:51< mattsc> zookeeper: the AI does no counter attack calculation whatsoever.
20140522 20:32:18< zookeeper> mattsc, i mean just the immediate retaliation damage
20140522 20:32:20< mattsc> If it feels it can deal enough damage to make an attack worth it, it will do so no matter how bad a situation it will get itself into with that.
20140522 20:32:49< mattsc> zookeeper: yes, that would have to be calculated to see if anything is wrong with that.
20140522 20:33:09< zookeeper> hmmkay
20140522 20:33:11< mattsc> But CTK is much more highly valued than damage without a kill, so that might play a role here.
20140522 20:33:34< mattsc> But for all I have seen, these calculations work, they are just sometimes unintuitive.
20140522 20:34:03-!- trewe [~trewe@2001:8a0:d11c:7901:626c:66ff:fe92:9b7c] has joined #wesnoth-dev
20140522 20:34:50< zookeeper> yeah
20140522 20:35:05< zookeeper> you're not in the middle of writing a reply already, are you?
20140522 20:35:13< mattsc> zookeeper: actually, reading over that description again, I’d say that behavior is what I would expect.
20140522 20:35:19< mattsc> zookeeper: no
20140522 20:35:33< zookeeper> ok, i'll write one
20140522 20:35:42< mattsc> zookeeper: that specific ally in that specific scenario is definitely problematic.
20140522 20:35:57< mattsc> One of the challenges is to keep him alive.
20140522 20:36:11< mattsc> And I think using passive_leader here would be a reasonable thing to do.
20140522 20:36:35< mattsc> But I’d also claim that the behavior described is “correct” as far as the AI evaluations are concerned.
20140522 20:36:55< mattsc> … which is not the same as that it makes sense from a human player’s perspective at all
20140522 20:40:04< zookeeper> yep
20140522 20:40:09< Aishiko> hmmmm we need a human perspective layer to the AI's actions for high value units!
20140522 20:41:18< zookeeper> one of the most common problems seems to be that the AI still doesn't value retaliation-free attacks (such as ranged vs a melee-only unit) enough, but rather just tries to maximize damage dealt
20140522 20:42:17< Aishiko> so instead of doing say ranged at 4x5 it'll do 5x6 melee when the defender has 0 ranged and 10x4 melee?
20140522 20:42:44< zookeeper> i dunno the specifics
20140522 20:42:47< iceiceice_> hmm
20140522 20:42:56< iceiceice_> its looks like there is a maximum length of filenames that wesnoth will accept
20140522 20:42:59< iceiceice_> and then use ... afterwards
20140522 20:43:14< iceiceice_> are there actually platforms that limit it to 20 chars or whatever?
20140522 20:43:22< Aishiko> iceiceice_, is that an absolute path name or just the local filename
20140522 20:44:02< iceiceice_> hmm so its pretty wierd
20140522 20:44:25< iceiceice_> my scenario id is like this:
20140522 20:44:26< iceiceice_> wesnoth -u check_victory_one_no_units_fail_one
20140522 20:44:39< iceiceice_> when i save it, the save dialog shortens it to:
20140522 20:45:03< iceiceice_> "unit_test_check_victory_on..._replay.gz"
20140522 20:45:36< iceiceice_> inside the replay file, my scenario name is actually "Unit Test check_victory_one_no_units _fail_one"
20140522 20:45:41< iceiceice_> with a space after "units"
20140522 20:45:44< iceiceice_> but apparently thats fine... ?
20140522 20:46:06< mattsc> Aishiko: it’s not that easy and the answer is always: it depends
20140522 20:46:30< iceiceice_> idk its not really a problem and it all works fine
20140522 20:46:44< iceiceice_> my guess is that the save dialog is just shortening things, perhaps unnecessarily,
20140522 20:46:49< iceiceice_> perhaps there are legit gui reasons?
20140522 20:46:52< Aishiko> mattsc, that sounds like the answer to most any question asked in life!
20140522 20:46:58< mattsc> And by the way, anonymissimus is correct in that it’s the scenario designer’s responsibility to make the scenario work with what the AI does.  Play to its strengths, avoid it’s weaknesses.
20140522 20:47:03-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Quit: leaving]
20140522 20:47:20< mattsc> Aishiko: indeed
20140522 20:47:37< Aishiko> iceiceice_, it might be that there isn't space to display the full name so truncates it for display purposes?
20140522 20:47:45< zookeeper> mattsc, the problem with passive_leader=yes seems to be that the wiki says it also prevents it from attacking adjacent enemies even when no retaliation is possible... which is a limitation that you basically never want
20140522 20:47:59< iceiceice_> y but i would hope it would like, let you just hit the right arrow over to the side to see the rest...
20140522 20:48:00< zookeeper> but yes, i suppose it's more of a scenario balancing issue
20140522 20:48:01< iceiceice_> maybe thats too hard in gui2 
20140522 20:48:38< mattsc> zookeeper: indeed.  The problem is that anything but a black/white setting is not at all trivial to do right.
20140522 20:48:43< iceiceice_> Aishiko: i think its not really about human perspective regarding the AI...
20140522 20:48:53< iceiceice_> Aishiko: its just not feasible to estimate retaliatory damage
20140522 20:49:23< iceiceice_> Aishiko: its already very slow to figure out what all of your possible moves are and estimate their values, if you want to ALSO consider all possible moves the human could take after that
20140522 20:49:39< iceiceice_> we just dont have a sophisticated way to rule out the "stupid" possibiltiies
20140522 20:49:44< mattsc> zookeeper: there are quite a few scenarios in SotBE that I would have set up differently if I’d written the campaign in the first place.  I tried to do what I could with minor changes.
20140522 20:50:02< mattsc> And I think I managed to make this scenario challenging but workable on medium difficulty.
20140522 20:50:40< mattsc> iceiceice_: well, actually, Fred does, and for the most part with reasonable delays.
20140522 20:51:01< mattsc> But there’s still lots of problems…
20140522 20:51:16< iceiceice_> maybe there could be like a limited version of this that is done only for the leader?
20140522 20:51:38< Aishiko> makes me glad I'm not trying to anything AI with zero experience in AI!
20140522 20:51:39< mattsc> Anyways, I have now tried to write and AI that is not just :blindly aggressive” for 2 years and have still not succeeded.  So obviously I don’t know what I am talking about.  :P
20140522 20:52:27< mattsc> iceiceice_: the goal is to come up with an AI that might be useful in other setting as well.  But so far I have not even managed in a very specific setting.  So …
20140522 20:54:26< iceiceice_> idk i wrote an email a long time ago :)
20140522 20:54:43< iceiceice_> you know my thoughts on this
20140522 20:54:47< zookeeper> mattsc, "These ai parameters that ai_controller relies on are only in effect if no enemy units are in reach" <- umm, is that right or wrong?
20140522 20:55:03< zookeeper> (those would be aggression and caution)
20140522 20:55:37< iceiceice_> i think making a good datastructure for pathfinding is a good step also
20140522 20:55:38< mattsc> It is definitely not correct for aggression.
20140522 20:55:41< iceiceice_> it could speed up everything
20140522 20:55:43< iceiceice_> so we'll do that first :)
20140522 20:55:51< zookeeper> mattsc, ok, great, thought so
20140522 20:56:31< mattsc> zookeeper: not correct for caution either
20140522 20:57:05< mattsc> It _is_ true for some of the others (protect …, target…)
20140522 21:02:39-!- kex [~kex@95.180.213.88] has joined #wesnoth-dev
20140522 21:05:50< zookeeper> mattsc, not that i'm volunteering to do the work, but do you think it'd be a good idea to have a collection of micro AI's (or whatever) for leaders only?
20140522 21:06:38< zookeeper> i dunno, maybe you already have some of those
20140522 21:07:00-!- irker325 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140522 21:07:03< zookeeper> basically so you could conveniently pick the behavior of the leader independently of the rest of the side, from a few basic archetypes
20140522 21:07:21< mattsc> zookeeper: I don’t think that that’s necessary as lots of MAIs have SUFs, so they can be used on leaders or other units at will.
20140522 21:07:49< mattsc> The question is more what kind of behavior would make sense and is that doable with current means or not.
20140522 21:08:57< mattsc> Having a “don’t attack with the leader unless it is safe, but then definitely do it” actions is something both Fred and the Bottleneck AI already do.
20140522 21:09:07< zookeeper> hmh, okay
20140522 21:09:18< mattsc> As I said, my hope is that some of that will become more generally usable in time.  At the moment it is not.
20140522 21:10:14< mattsc> But if you have suggestions for specific behavior for MAIs (or AIs in general), I’m always interested.
20140522 21:10:26< mattsc> No promises that I will be able to do it though.
20140522 21:11:39< zookeeper> a semi-passive leader would probably be what many scenarios would want... that is, one which can move within one turn of the keep if (for example to capture a village) it's completely safe to do so, and which attacks an adjacent enemy if it can do so without retaliation
20140522 21:13:37< mattsc> zookeeper: I think I have something better than that.  But it is currently deeply interleaved with the rest of the AI, so it’s going to take some time to sort all of this out ...
20140522 21:14:39< zookeeper> on the other hand, our AI might do stupid things but it's still a lot better than the AI of a lot of other games :P
20140522 21:15:09< mattsc> zookeeper: the Wesnoth AI is really quite amazing, no matter how much people complain about it
20140522 21:15:22< zookeeper> it's not like any game's AI is really a strategic mastermind
20140522 21:16:34< mattsc> Actually, the concept of the Wesnoth AI is very elegant in its “simplicity”.  (The concept being simple, not the practical issues.)
20140522 21:17:08< mattsc> Write an AI that is aggressive (almost) without compromise, and then design a faction around it that does best with that sort of behavior.
20140522 21:17:18< mattsc> If you do that, there are really no “wrong” moves.
20140522 21:17:54< mattsc> By contrast, if you try to tell the AI to defend “sometimes” and attack at other times, it is bound to make one mistake after another.  :P
20140522 21:19:02< mattsc> zookeeper: you know what’s wrong here is our perception of how an orc clan(?) works.  Do we really think the orcs would stop fighting because their leader is killed in battle?
20140522 21:19:36< mattsc> If he’s not killed by the enemy in battle, he’d probably be assassinated by his second in command that same night.  ;)
20140522 21:20:02< mattsc> Anyways, I really need to be off.  I’ll read up on things later.
20140522 21:20:05< zookeeper> me too
20140522 21:21:28< Aishiko> mattsc, has a point perhaps we need the orcs to just promote the next highest ranking orc to leader if the leader is killed
20140522 21:22:50-!- irker396 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140522 21:22:50< irker396> wesnoth: Chris Beck wesnoth:master d129d80ebedd / / (3 files in 3 dirs): Add unit tests for check victory function http://git.io/pb_Zzw
20140522 21:22:50< irker396> wesnoth: Chris Beck wesnoth:master 49f97ee9befc / data/campaigns/The_South_Guard/scenarios/03_A_Desparate_Errand.cfg: Merge branch 'master' of git://github.com/wesnoth/wesnoth http://git.io/SfZHRA
20140522 21:22:57< Aishiko> but there is also merit to the fact that if the leader is killed the weaker ones might run because, hey their leader was killed, you know the leader they all feared in a straight up fight and now dead, who's going to punish them if they run and live to fight another day
20140522 21:23:32< iceiceice_> gfgtdf: ok i didnt find any bugs in defeat condition
20140522 21:23:55-!- iceiceice_ is now known as iceiceice
20140522 21:24:03< gfgtdf> iceiceice: will you backport it ?
20140522 21:24:14-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds]
20140522 21:25:16< gfgtdf> iceiceice: is there soem code that converts [scenario] into [multiplayer] in the game config or something liek that ?
20140522 21:25:40< iceiceice> i only skimmed that code, i dont know
20140522 21:26:03< iceiceice> about backport: you mean to backport the check_victory code/
20140522 21:26:12< iceiceice> cant backport the unit test
20140522 21:29:53< gfgtdf> iceiceice: https://github.com/wesnoth/wesnoth/blob/master/src/game_config_manager.cpp#L144
20140522 21:30:58< gfgtdf> iceiceice: no i just meant the commits about check_victory in play_contreoller.cpp, lua.cpp team.cp/hpp
20140522 21:32:48< iceiceice>  it looksl ike its like 5 or 6 commits
20140522 21:32:59< iceiceice> if you dont want to cherry-pick it yourself i will but you have to make sure i have the right list :)
20140522 21:34:22< gfgtdf> iceiceice: hmi guess i can do it myself since i also want the [wml_message] think in 1.11.15
20140522 21:34:55< iceiceice> ok
20140522 21:38:07< gfgtdf> iceiceice: have you tried this: 20140522 02:31:12< gfgtdf> 	iceiceice_: to read the scenario we have to refresh the game_config first and we do that after the savegame call in https://github.com/wesnoth/wesnoth/blob/master/src/mp_game_utils.cpp#L58
20140522 21:39:56< iceiceice> gfgtdf: i added it in fill_mp_level_config
20140522 21:40:23< iceiceice> https://github.com/cbeck88/wesnoth/commit/309e9faf72ee5bb4722b24d92bf7e3f562434ffd
20140522 21:40:36< iceiceice> that commit didnt fix it
20140522 21:41:04< iceiceice> gfgtdf: if you cherry-pick those, i can write release_notes entry if you want
20140522 21:41:19-!- gfgtdf_ [~chatzilla@e177123014.adsl.alicedsl.de] has joined #wesnoth-dev
20140522 21:42:11< gfgtdf_> iceiceice: ^
20140522 21:43:48< iceiceice> gfgtdf: i think you might have missed my reply when you dc'd 
20140522 21:44:15< gfgtdf_> i cannot find it here https://github.com/cbeck88/wesnoth/commit/309e9faf72ee5bb4722b24d92bf7e3f562434ffd
20140522 21:44:34< gfgtdf_> iceiceice: iceiceice: what i meant was that maybe you cannot call game_config_.find_child(gamestat...  before calling 	resources::config_manager->load_game_config_for_game(state.classification());  And in 309e9faf72ee5bb4722b24d92bf7e3f562434ffd we do that
20140522 21:44:46-!- gfgtdf [~chatzilla@e176191162.adsl.alicedsl.de] has quit [Ping timeout: 265 seconds]
20140522 21:44:59-!- gfgtdf_ is now known as gfgtdf
20140522 21:46:03< iceiceice> oh
20140522 21:46:14< iceiceice> that's worth a shot
20140522 21:50:03-!- Guest89373 [~cib@p5DD222D9.dip0.t-ipconnect.de] has quit [Ping timeout: 252 seconds]
20140522 21:50:27-!- Guest89373 [~cib@p5DD222D9.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140522 21:58:00< iceiceice> hmm i have a general question about wml
20140522 21:58:13< iceiceice> so as i understand, variables are untyped meaning that
20140522 21:58:16< iceiceice> if i type 
20140522 21:58:31< iceiceice> {VARIABLE x 0}
20140522 21:58:36< iceiceice> its not different from {VARIABLE x "0"}
20140522 21:58:41< iceiceice> is that right?
20140522 21:59:06< iceiceice> the config::attribute_value basically elides the distinction?
20140522 21:59:37< iceiceice> at the C++ level, if i write 
20140522 21:59:42< iceiceice> cfg["x"] = 0
20140522 21:59:51< iceiceice> i guess thats probably different from cfg["x"] = "0"
20140522 21:59:55< iceiceice> because it uses a variant that reembers,
20140522 22:00:07< iceiceice> but in general the parser may not notice the difference, is that true?
20140522 22:01:49< iceiceice> you know, its like i think i have a good guess about the answers, but i'd rather to hear it from someone who knows what they are talking about
20140522 22:04:34< gfgtdf> 	iceiceice: i think even in c++ its the same
20140522 22:04:51< gfgtdf> whether you call ="0" or =0
20140522 22:04:55< gfgtdf> = 0
20140522 22:05:10< iceiceice> so i was very concerned about this when i was testing the RNG seed stuff
20140522 22:05:18< iceiceice> because if you store the seeds as hex strings,
20140522 22:05:23< iceiceice> if there is a leading zero, and no letters
20140522 22:05:30< iceiceice> it could get truncated
20140522 22:05:36< iceiceice> if it its converted to an integer
20140522 22:05:43< iceiceice> but in my unit tests that never happened
20140522 22:06:10< gfgtdf> hm eve in hex it shouldnt matter whether the first 0 is there or no
20140522 22:06:11< iceiceice> i even tested sending seeds like that as configs over socket to local host in the test network layer stuff
20140522 22:06:40< iceiceice> gfgtdf: i guess thats true but i wanted equality as strings 
20140522 22:07:09< iceiceice> in case you want to parse an array of uint32's
20140522 22:07:23< iceiceice> so the padding will happen in the right int
20140522 22:07:45< iceiceice> idk it passed all my tests so i'm not concerned anymore but i never realyl figured out why it was working
20140522 22:08:03< iceiceice> if its actually can break then my plan was to add "0x" at the start
20140522 22:08:13< iceiceice> but dont want to if dont have to
20140522 22:16:01< iceiceice> idk its not that important but i just wanted to figure out what was going on there
20140522 22:16:11< iceiceice> does it only convert to int if "to_int" is called at some point
20140522 22:18:02< shadowm> No.
20140522 22:21:18-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 22:22:39-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: Ik ga weg]
20140522 22:26:27-!- Guest89373 [~cib@p5DD222D9.dip0.t-ipconnect.de] has quit [Ping timeout: 252 seconds]
20140522 22:26:42-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 258 seconds]
20140522 22:27:26< iceiceice> hmm well its easy to test and the tests say otherwise...
20140522 22:27:54< iceiceice> paste this at the end of your src/tests/test_utils.cpp
20140522 22:27:54-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 22:27:55< iceiceice> http://pastebin.com/UdtqNsVu
20140522 22:28:17< iceiceice> and put `include "config.hpp"` at the top
20140522 22:28:23< iceiceice> my output was this:
20140522 22:28:45< iceiceice> http://pastebin.com/nUh0LNuE
20140522 22:31:23< iceiceice> is that a bug?
20140522 22:34:43-!- kex [~kex@95.180.213.88] has quit [Remote host closed the connection]
20140522 22:35:19-!- kex [~kex@95.180.213.88] has joined #wesnoth-dev
20140522 22:37:10-!- trewe [~trewe@2001:8a0:d11c:7901:626c:66ff:fe92:9b7c] has quit [Quit: quit]
20140522 22:39:16-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20140522 22:39:33-!- kex [~kex@95.180.213.88] has quit [Ping timeout: 252 seconds]
20140522 22:43:36< gfgtdf> iceiceice: i have no idea why the last one prints 00001111
20140522 22:45:05< gfgtdf> iceiceice: hm ok i think thats teh uintented behaviour
20140522 22:45:45< gfgtdf> the code onyl parsed string to int if it doent change teh string
20140522 22:46:11< iceiceice> idk i didnt make it, is that actually intended? or just some obscure artifact
20140522 22:46:40< gfgtdf> iceiceice: there is explicit code that does that.
20140522 22:46:45< iceiceice> either way imo there should be a "test_config" unit test with a bunch of hard coded examples...
20140522 22:46:57< iceiceice> that would be much easier for someone like me to figure out wtf is supposed to happen quickly
20140522 22:47:11-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140522 22:47:22-!- kex [~kex@78.157.29.205] has joined #wesnoth-dev
20140522 22:48:15< iceiceice> imo, the right answer is, whatever the devs who have been here the longest think it is... its not documented, and who knows what assumptions are made throughout the code
20140522 22:48:32< iceiceice> and if no one knows then i guess just commit tests based on current behavior
20140522 22:48:55< iceiceice> if they are expecting it to behave exactly liek WML configs though thats clearly not the case
20140522 22:48:59< iceiceice> or stopped being the case at some point
20140522 22:51:13< iceiceice> gfgtdf: i tried "running load game config for game" earlier in the pathway
20140522 22:51:14< iceiceice> https://github.com/cbeck88/wesnoth/compare/wesnoth:1.12...fixup_mp_sos_saves
20140522 22:51:17< iceiceice> it didnt fix it thouh
20140522 22:51:26< iceiceice> i'm not sure if state.classification() has the right value at that time though
20140522 22:55:06-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Ciao]
20140522 23:00:08< iceiceice> i really dont understand why i'm getting 585 either
20140522 23:00:10< iceiceice> shouldnt i get 1111
20140522 23:01:28< shadowm> 0111 is octal, not binary.
20140522 23:01:34< shadowm> I mean 01111.
20140522 23:02:08-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140522 23:02:58< shadowm> What I mean with this is that 01111 in base 8 is 585 decimal.
20140522 23:06:58< iceiceice> hmm i see
20140522 23:07:04< iceiceice> i didnt realize C++ would do that, that's on me 
20140522 23:07:25< iceiceice> y even 00001111L becomes 585
20140522 23:08:05 * shadowm reads that as 'why'.
20140522 23:08:10< shadowm> Because strtod seems to do that.
20140522 23:08:34< iceiceice> ok, thats all the time i have for today
20140522 23:08:37< shadowm> Reading the config class implementation should prove enlightening.
20140522 23:08:44-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has joined #wesnoth-dev
20140522 23:08:44< travis-ci> [travis-ci] gfgtdf/wesnoth-old#106 (cv_in_112 - 386893d : gfgtdf): The build passed.
20140522 23:08:44< travis-ci> [travis-ci] Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/25829666
20140522 23:08:44-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has left #wesnoth-dev []
20140522 23:09:04< iceiceice> shadowm: i've read it but not recently
20140522 23:09:23< shadowm> The strtol manpage specifically explains that numbers preceded with a 0x are considered hexadecimal, and those preceded with a single zero are octal.
20140522 23:09:40< iceiceice> but you also seemed yourself to expect that cfg["x"] = "00001111" would cout as 1111
20140522 23:09:42< shadowm> But the strtod manpage doesn't say whether it interprets numbers the same way as strtol, and strtod is what we use.
20140522 23:09:47< gfgtdf> shadowm: no that part of c++ language not of strtod i think
20140522 23:10:03< shadowm> iceiceice: Really? When?
20140522 23:10:13< shadowm> gfgtdf: No.
20140522 23:10:13< iceiceice> <iceiceice> does it only convert to int if "to_int" is called at some point
20140522 23:10:14< iceiceice> <shadowm> No.
20140522 23:10:22< shadowm> We were talking about strings, weren't we?
20140522 23:10:32< iceiceice> yes
20140522 23:10:39< shadowm> iceiceice: What I said is still true.
20140522 23:10:47< gfgtdf> shadowm_: no read iceiceice s link above
20140522 23:11:28< shadowm> config::attribute_value::operator=() will try to convert its argument to a number regardless of what you do.
20140522 23:11:45< iceiceice> shadowm that's not true
20140522 23:11:54< iceiceice> my test data demonstrates it
20140522 23:12:56< shadowm> Okay, then it's not true and I spent days on a workaround for MP password storage in vain.
20140522 23:12:57< gfgtdf> shadowm: operatore = wil only convert to number if that doesnt chage its strign repersentation
20140522 23:13:15< gfgtdf> shadowm: so it wont change "0011"
20140522 23:13:19< gfgtdf> convert*
20140522 23:14:42< shadowm> I wasn't informed of commit b58d88c8 .
20140522 23:15:06< shadowm> That seems to be what added the reverse check.
20140522 23:16:22< shadowm> Okay, so strings are save, and no, C++ doesn't have a base 2 number literal before C++11.
20140522 23:16:26< shadowm> Er, asfe.
20140522 23:16:31< shadowm> safe
20140522 23:16:40< shadowm> The typos are contagious.
20140522 23:17:08< shadowm> In C++11 I believe you'd represent a base 2 number as 0b0100110 but I'm not sure.
20140522 23:17:31< iceiceice> ok well i'm out of time for today and my battery is going
20140522 23:17:40< shadowm> And in previous versions any number preceded with a 0 is indeed octal.
20140522 23:18:03< iceiceice> if there are critical uses of config that assume certain behavior, then maybe we should figure out what those are write some tests
20140522 23:18:09< shadowm> This is why you can write calls to chmod() and such in a rather natural fashion.
20140522 23:18:20-!- iceiceice [~chris@207-237-132-91.ny.subnet.cable.rcn.com] has quit [Quit: Leaving]
20140522 23:18:46< shadowm> iceiceice: And with all that said, my first answer is still correct.
20140522 23:18:58< shadowm> It will try to convert to int the moment you set the attribute.
20140522 23:19:06< shadowm> Exact words. :)
20140522 23:19:29< shadowm> (Whether it succeeds and goes with the result or not is a different matter.)
20140522 23:22:48< shadowm> As for the historical background behind my previous statement, see bug #16571, commit fc9e7ef2919ecbb5a6d842067a5f99d07c8f9bee, and commit b58d88c84abc9bb3d8d8ff056cd7d604f713adaa, which apparently went past my radar.
20140522 23:23:37< shadowm> Also commit 76c34a23d3b562274781ea1ab7cb107285a098c9, which started it all.
20140522 23:24:24< shadowm> If you are trying to write tests, you might find the commit message in the last one useful.
20140522 23:25:16-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has joined #wesnoth-dev
20140522 23:25:16< travis-ci> [travis-ci] gfgtdf/wesnoth-old#107 (cv_in_112 - 9f2eb63 : gfgtdf): The build passed.
20140522 23:25:16< travis-ci> [travis-ci] Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/25830678
20140522 23:25:16-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has left #wesnoth-dev []
20140522 23:26:51-!- kex [~kex@78.157.29.205] has quit [Remote host closed the connection]
20140522 23:27:39< gfgtdf> i thought we onyl see these message then fail/success changes ?
20140522 23:29:27< shadowm> Hm, binary literals are part of C++14, not C++11.
20140522 23:30:18< vultraz> C++14 is a thing now?
20140522 23:31:40< shadowm> It's been a thing since a while.
20140522 23:32:03< vultraz> And yet we still don't use C++11
20140522 23:32:12< shadowm> AFAIK there's no guarantee it'll be finalized this year though, so the name is a placeholder like C++0x was at its time.
20140522 23:32:50< irker396> wesnoth: gfgtdf wesnoth:1.12 aabeae68f905 / src/ (play_controller.cpp scripting/lua.cpp team.cpp team.hpp): replace fight_on_without_leader with defeat_condition http://git.io/F8chGQ
20140522 23:32:52< irker396> wesnoth: gfgtdf wesnoth:1.12 3d6c602b7eb5 / src/ (play_controller.cpp team.cpp team.hpp): add "always" defeat_condition http://git.io/pDUUdg
20140522 23:32:54< irker396> wesnoth: gfgtdf wesnoth:1.12 a365475b8c34 / src/ (play_controller.cpp play_controller.hpp playsingle_controller.cpp): rename remove_from_carryover_on_leaders_loss_ http://git.io/nWaHIw
20140522 23:32:56< irker396> wesnoth: gfgtdf wesnoth:1.12 386893d6dff7 / src/team.cpp: rename values of [side]defeat_condition= http://git.io/3jpIUA
20140522 23:32:58< irker396> wesnoth: gfgtdf wesnoth:1.12 0111ffbcd042 / src/game_events/ (action_wml.cpp pump.cpp pump.hpp): add "write_to" parameter to wml_message http://git.io/g3iFfQ
20140522 23:33:00< irker396> wesnoth: gfgtdf wesnoth:1.12 9f2eb63b5235 / src/game_events/ (action_wml.cpp pump.cpp pump.hpp): replace write_to with to_chat in [wml_message] http://git.io/mstCZA
20140522 23:33:02< irker396> wesnoth: gfgtdf wesnoth:1.12 5085495316be / src/minimap.hpp: #include <map.hpp> -> #include "map.hpp" http://git.io/k4nCfg
20140522 23:33:04< irker396> wesnoth: gfgtdf wesnoth:1.12 2744fa130d6e / src/hotkey/hotkey_item.cpp: don't load null hotkeys http://git.io/zcgGHA
20140522 23:33:11< shadowm> And because it's not finalized, compiler support is by definition incomplete and/or experimental.
20140522 23:46:54-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has joined #wesnoth-dev
20140522 23:46:54< travis-ci> [travis-ci] gfgtdf/wesnoth-old#108 (cv_in_112 - 2744fa1 : gfgtdf): The build passed.
20140522 23:46:54< travis-ci> [travis-ci] Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/25830884
20140522 23:46:54-!- travis-ci [~travis-ci@ec2-54-198-38-196.compute-1.amazonaws.com] has left #wesnoth-dev []
20140522 23:47:42-!- RiftWalker [~nathan@ip24-252-126-205.no.no.cox.net] has quit [Ping timeout: 276 seconds]
20140522 23:49:16-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140522 23:59:37-!- travis-ci [~travis-ci@ec2-54-83-115-127.compute-1.amazonaws.com] has joined #wesnoth-dev
20140522 23:59:37< travis-ci> [travis-ci] wesnoth/wesnoth#2756 (1.12 - 2744fa1 : gfgtdf): The build has errored.
20140522 23:59:37< travis-ci> [travis-ci] Build details : http://travis-ci.org/wesnoth/wesnoth/builds/25832465
20140522 23:59:37-!- travis-ci [~travis-ci@ec2-54-83-115-127.compute-1.amazonaws.com] has left #wesnoth-dev []
--- Log closed Fri May 23 00:00:58 2014