--- Log opened Sat Apr 12 00:00:01 2014
20140412 00:02:59-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Ping timeout: 255 seconds]
20140412 00:03:44< irker807> wesnoth: David Mikos wesnoth:master 59e95ff30488 / / (3 files in 2 dirs): Add 'elemental' trait to mudcrawler line. http://git.io/1lOHvw
20140412 00:17:04-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140412 00:22:03< irker807> wesnoth: mattsc wesnoth:master bb8c08a31421 / data/ai/micro_ais/cas/ (7 files): Micro AIs: standardize eval/exec function table setup http://git.io/uTZ0fw
20140412 00:23:19< irker807> wesnoth: mattsc wesnoth:1.12 19462f3a153d / data/ai/micro_ais/cas/ (7 files): Micro AIs: standardize eval/exec function table setup http://git.io/c4F7YA
20140412 00:27:09< irker807> wesnoth: mattsc wesnoth:master 122a1c852f04 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Update Xcode project http://git.io/8EDzyw
20140412 00:29:21< mattsc> gfgtdf: yep, that bug is fixed
20140412 00:29:32< gfgtdf> mattsc: :)
20140412 00:29:32< mattsc> gfgtdf: however, I somehow manage to trigger another assert  :P
20140412 00:29:41< gfgtdf> mattsc: :(
20140412 00:29:50< mattsc> Give me a second to try to figure out how I did that
20140412 00:29:51< gfgtdf> which assert and how ?
20140412 00:30:42-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 240 seconds]
20140412 00:31:14< mattsc> gfgtdf: so … now if you press the “stop” button while the side is recruiting in the replay, you get this one:
20140412 00:31:16< mattsc> Assertion failed: (num_iters()==0 && o.num_iters() == 0), function swap, file /mats/misc/Wesnoth/wesnoth/src/unit_map.cpp, line 55.
20140412 00:31:46< gfgtdf> the square "replay reset" button ?
20140412 00:31:59< mattsc> jawohl
20140412 00:32:46< gfgtdf> mattsc: do you think we shoudl disallow that button also while another action is executed?
20140412 00:33:14< mattsc> Well, that one’s the reset to beginning button, right?
20140412 00:33:31< mattsc> I think it’d be fine to not allow that while things are going on.  That’s what the pause button is for.
20140412 00:33:39< mattsc> Let me check if that one works.
20140412 00:34:34< mattsc> Yeah, the pause button seems to work.  It has no effect during a side’s turn, but stops “continuous replay” once the current side is done.
20140412 00:35:31< mattsc> … and just for completeness, the square stop button also triggers the assertion fault during continuous replay.
20140412 00:35:40< gfgtdf> so that people would click 'pause' and then wait and only then could click 'reset' would be no problem ?
20140412 00:36:00< mattsc> I think that’s perfectly acceptable.
20140412 00:36:09< iceiceice> i mean it depends on the circumstance right?
20140412 00:36:17< iceiceice> i guess if there is a bugged scenario that is stuck you might want to hit the square
20140412 00:36:26< iceiceice> maybe would just have to reload then though i guess
20140412 00:36:32< iceiceice> meh its up to you
20140412 00:36:44< mattsc> iceiceice: what does “bugged scenario” means?
20140412 00:37:27< iceiceice> so what i understand is that when you hit the square it stops what is happening and replaces the game_state object with the original 
20140412 00:37:56< irker807> wesnoth: gfgtdf wesnoth:master 0cda27dd2d30 / src/replay_controller.cpp: disable RESET_REPLAY during actions http://git.io/tYMW9g
20140412 00:38:01< iceiceice> i dont have a particular bad case in mind
20140412 00:38:05< gfgtdf> mattsc: try this ^
20140412 00:38:11< mattsc> iceiceice: AFAIK, the square button reset the replay to the beginning.  Is that what you’re saying?
20140412 00:38:16< iceiceice> yes
20140412 00:38:56< mattsc> I think it’s fine if you have to wait until the end of the current side’s turn until you can do that
20140412 00:38:59< iceiceice> idk gfgtdf has been "fixing" a lot of bugs lately by just putting "commands_disabled" everywhere or disabling the users ability to do things
20140412 00:39:21< gfgtdf> iceiceice: not a lot
20140412 00:39:34< iceiceice> ok, i wont say anymore, as long as you guys thought about it
20140412 00:39:48< mattsc> I have to admit that I have not been following too closely what the two of you have been doing/discussing.  Too much and too far above my head.  :)
20140412 00:41:42< mattsc> gfgtdf: yes, that does what you said it would no.  No more crashes, and no access to the big blue square button until the replay is stopped.
20140412 00:41:49< mattsc> s /no/do
20140412 00:43:25< mattsc> Axully, not quite true.  If you do a quick staccato on that button, you can trigger it in between side turns during continuous replay.
20140412 00:43:28< gfgtdf> mattsc: i tihnk there were plans to implement a replay csingle action button, especaily when i tihnk of 'pause replay' i think it woudl be nice it teh replays woudl pause immediately, intead of playing the turn to end
20140412 00:43:40< mattsc> But then it just resets to the beginning, so that’s what I’d expect it to do.
20140412 00:44:25< mattsc> gfgtdf: agrred, it would be _nice_.  But I think what it is right now is perfectly good enough.
20140412 00:44:51< mattsc> gfgtdf: so remind me, what’s the status about fixing this in 1.12?
20140412 00:45:41< gfgtdf> mattsc: i made a comment here: https://gna.org/bugs/?21905
20140412 00:46:19< mattsc> Oh, right, that’s where I read something about it.
20140412 00:46:21< iceiceice> ah i see
20140412 00:46:25< gfgtdf> s/sues/uses
20140412 00:46:32< iceiceice> gfgtdf: are oyu going to backport as soon as i fix blindfold replays/
20140412 00:46:44< iceiceice> or is it waiting on other things
20140412 00:46:45< iceiceice> too
20140412 00:47:19< gfgtdf> iceiceice: there are currently no bugs that im aware of.
20140412 00:47:28< iceiceice> ok
20140412 00:47:45< gfgtdf> iceiceice: did you inspect teh 'wrign controller in mp campaigns' issue ?
20140412 00:48:03< iceiceice> which
20140412 00:48:22< gfgtdf> the side 2 has 'controlled=netwrok' on both cients
20140412 00:48:28< iceiceice> oh right
20140412 00:48:30< iceiceice> ok so 
20140412 00:48:54< iceiceice> i'm not really sure what to do about this, right now, there were some underlying bugs in mp campaign transitions
20140412 00:49:00< iceiceice> that i found when trying to figure out what i should do about it
20140412 00:49:09< iceiceice> those bugs existed at least back to 1.11.5
20140412 00:49:15< iceiceice> and maybe even in 1.10
20140412 00:49:47< iceiceice> i reported on the bug tracker and thunderstruck assigned to himself so i guess i will see what he does, 
20140412 00:50:01< iceiceice> if he makes something that fixes 1.12 then i will figure out how to make it compatible with the server controller tweaks i think
20140412 00:50:12< mattsc> gfgtdf: I’ll leave it to you to mark that bug as fixed then whenever you have backported things to 1.12.
20140412 00:50:29< iceiceice> but i'm not going to try to fix it until i get a sense of broadly how he thinks the system should work
20140412 00:50:47< gfgtdf> mattsc: ok
20140412 00:50:50< gfgtdf> iceiceice: ok
20140412 00:51:10< mattsc> gfgtdf: thanks.  :)
20140412 00:51:41< iceiceice> gfgtdf: if you want to backport to 1.12, i dont think you necessarily have to wait for blindfold replays to be working,
20140412 00:51:56< iceiceice> fixing the sync stuff is probably more important than that feature and i think it will be easy to fix anyways
20140412 00:52:05< gfgtdf> iceiceice: ok
20140412 00:52:07< iceiceice> if i really cant figure it out and we have to release then i guess we will disable blindfold for 1.12 release
20140412 00:52:14< iceiceice> which is pretty easy
20140412 00:53:30< iceiceice> btw i really think someone should look into 21866
20140412 00:53:37< iceiceice> i'm not sure who but someone who knows about unit underlying id system
20140412 00:53:55< iceiceice> the bug as reported isnt awful, but it might indicate a more serious problem
20140412 00:54:05< iceiceice> wesbot: bug #21866
20140412 00:54:06< wesbot> Bug #21866  Assigned to: None  Status: None  Priority: 5 - Normal
20140412 00:54:06< wesbot> Summary: MP campaign underyling_id collision
20140412 00:54:06< wesbot> Original submission: In an MP campaign, this bug can cause units to not appear
20140412 00:54:09< wesbot>  on the right-click recall list, though they will still be present on the menu b
20140412 00:54:12< wesbot> URL: http://gna.org/bugs/?21866
20140412 00:55:10< mattsc> iceiceice: that one in my mind is related to a discussion of whether underlying_if should be exposed to Lua
20140412 00:55:59< mattsc> ugh …  am I allowed to make a silly comment in this channel?  Or should I do this in a private channel?
20140412 00:56:00< iceiceice> hmm
20140412 00:56:46< gfgtdf> iceiceice: there was an idea about using UUID as unit sunderlaying id's i liek that but i dont knwo enough about how to gte a good random number.
20140412 00:57:09< gfgtdf> get*
20140412 00:57:25< iceiceice> what is UUID?
20140412 00:58:04< Aishiko> Universially Unigue Identifeier (sp?)
20140412 00:58:38< gfgtdf> generating a very large rnadom  number for unit id's with the intention that if the numbers are large enough the proability of collision can be assumed to be null
20140412 00:58:41< iceiceice> ah i never knew about this, i read UUID and i thought it had something to do with my system coming out of hibernation :p
20140412 00:58:43< gfgtdf> random*
20140412 00:58:46< Aishiko> if your running *nix do a "ls -lF /dev/disk/by-uuid" to see some on your system
20140412 00:59:11< shadowm_desktop> "Universally Unique Identifier".
20140412 00:59:22< iceiceice> wait why can we not just systematically assign unique identifiers?
20140412 00:59:51< gfgtdf> to units ?
20140412 00:59:53< Aishiko> nothing to say we can't iceiceice 
20140412 01:00:34< Aishiko> though what would be the point, unless you wanted to do something like an all stars battle where you recruit in the top 20 powerhouse units you've ever had...
20140412 01:00:56< iceiceice> i mean they only have to be unique within a scenario / campaign... ?
20140412 01:01:08< gfgtdf> we currently assig units ids with an increating int counter. 'per camaign' .
20140412 01:01:14< iceiceice> yea whats wrong with that?
20140412 01:02:06< mattsc> Aishiko: Identifeier would be the German spelling - but it refers to the party you have when you get your first ID.
20140412 01:03:51< gfgtdf> are the units uids exposed by lua/wml ? meansing whenn i[store_unit]  and then [unstore] it, will it have teh sae uid ?
20140412 01:03:57< gfgtdf> meaning*
20140412 01:04:26< mattsc> gfgtdf: yes, they are
20140412 01:04:42< mattsc> However, if you reload, the won’t be the same as they were before.
20140412 01:04:44< Aishiko> mattsc, thanks for the spelling
20140412 01:04:55< iceiceice> hmm
20140412 01:05:06< iceiceice> is there any legitimate reason to read the uid?
20140412 01:05:21< gfgtdf> i also dont know what happends when you transfer a unit from one campaign to another.
20140412 01:05:49< gfgtdf> via global variables.
20140412 01:07:07< mattsc> iceiceice: it’s the only unique identifier of the unit
20140412 01:07:32< EliDupree> AFAICT id= is also unique
20140412 01:07:34< iceiceice> but if you read it it can cause OOS?
20140412 01:07:42< iceiceice> when you reload?
20140412 01:07:44< EliDupree> creating units with identical id= caused buggy savefiles
20140412 01:08:04< mattsc> iceiceice: id= should be unique but is not forced to be
20140412 01:08:34< iceiceice> but still it sounds like its not a good idea for WML to make use of underlying ids, right?
20140412 01:08:48< iceiceice> i mean a scenario could assign their own ids if they want and store them with the units
20140412 01:09:00< Aishiko> iceiceice,  that might be a good thing to patch, making id= forced to be unique
20140412 01:09:13< iceiceice> if WML can write uids it sounds like it will cause engine bugs
20140412 01:09:19< mattsc> I believe the intention is to give WML/Lua other ways of doing unit comparison
20140412 01:09:21< iceiceice> and anyways apparently there are already engine bugs with it in campaigns
20140412 01:09:45< gfgtdf> iceiceice: underlaying ids are not the same in replays. And using uids in wml is consideredbad practics.
20140412 01:09:57< iceiceice> so why dont we just make it unavailable to umc?
20140412 01:10:24< mattsc> for example, in Lua you can now do (thanks to anonymissimus): if unit1 == unit2 then … - this uses underlying_id, but that’s hidden from the user
20140412 01:11:14< iceiceice> it is really a pretty strange set up we have where units have "id" and "underlying_id"
20140412 01:11:28< mattsc> iceiceice: I think that’s a good idea - even though it would make a dirty trick I am using in Grnk not work any more.   :\ 
20140412 01:11:30< iceiceice> its like "wait these are both just numbers that identify the unit... but one of them is `the bugged one`"
20140412 01:11:54< iceiceice> i think id should bascailly be the WML one and underlying id the engine one
20140412 01:12:58< iceiceice> maybe allow to read uid in debug mode only or something
20140412 01:14:34< mattsc> iceiceice: tbh, why would you even need that?  (speaking as somebody who has requested having access to it before)
20140412 01:15:14< iceiceice> idk its probably less effort to make that change, and precludes the possibility that some dev will complain later :p
20140412 01:17:16-!- TC01 [~quassel@128.220.109.252] has quit [Remote host closed the connection]
20140412 01:23:17-!- un214 [~un214@2602:306:cccf:a499:224:8cff:fed2:ef57] has joined #wesnoth-dev
20140412 01:23:31-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20140412 01:57:47-!- un214 [~un214@2602:306:cccf:a499:224:8cff:fed2:ef57] has quit [Read error: Connection reset by peer]
20140412 02:13:18-!- Ivanovic_ [~ivanovic@x2f50bf1.dyn.telefonica.de] has joined #wesnoth-dev
20140412 02:16:00-!- Ivanovic [~ivanovic@x2f3be66.dyn.telefonica.de] has quit [Ping timeout: 276 seconds]
20140412 02:17:12-!- Ivanovic_ is now known as Ivanovic
20140412 02:36:30-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: Ik ga weg]
20140412 02:36:40-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Remote host closed the connection]
20140412 02:37:03-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140412 02:44:52-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on 1.12 | 232 bugs, 351 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
20140412 02:56:54< gfgtdf> iceiceice: i have a bug n 1.11.12+dev : when i host a mp game with wesnothd, but play a game only vs myself (with both sides run on the same client) on a scenario that contains an ai side, i get from teh server the the server things teh ai side is owned by 'unknown'
20140412 02:57:05-!- Jetrel [~Jetrel@c-75-73-180-126.hsd1.mn.comcast.net] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin]
20140412 02:57:36< iceiceice> hmm i dont have that bug 
20140412 02:57:40< iceiceice> it must be new on 1.12?
20140412 02:58:00-!- Jetrel [~Jetrel@c-75-73-180-126.hsd1.mn.comcast.net] has joined #wesnoth-dev
20140412 02:59:01< gfgtdf> iceiceice: you dont have thjat bug on which version ?
20140412 02:59:22< iceiceice> i have been running many many tests like that
20140412 02:59:26< iceiceice> i didnt get any bugs like that ever
20140412 02:59:34< iceiceice> on all versions
20140412 02:59:38< iceiceice> on master, on 1.12, 
20140412 02:59:48< iceiceice> on pre PR 121, on post PR 121,
20140412 03:00:04< iceiceice> maybe you have a special setting?
20140412 03:01:29< gfgtdf> iceiceice	: try 2p basilisk caves, it seems liek i dint get teh bug oins siver 
20140412 03:01:58< gfgtdf> iceiceice: it has controller=null
20140412 03:02:01< gfgtdf> not ai
20140412 03:02:19< iceiceice> can you reproduce on master?
20140412 03:02:42< iceiceice> most likely this has been fixed by the controller type things, this was tested quite a bit
20140412 03:03:00< iceiceice> i dont think it really makes sense to make a different fix for what is currently on 1.12
20140412 03:05:03< gfgtdf> iceiceice: i test on offical server ?
20140412 03:05:13< iceiceice> no always test with local server
20140412 03:05:35< gfgtdf> iceiceice: you can aks shadowm to rebuild offical server from master
20140412 03:05:51< gfgtdf> ask*
20140412 03:05:56< iceiceice> theres no need for this kind of stuff
20140412 03:06:05< iceiceice> it will be faster and more certain if you make a local server
20140412 03:06:10< iceiceice> actually i just use a script when i do these tests,
20140412 03:06:18< iceiceice> wiki.wesnoth.org/DebuggingNetworkedMP
20140412 03:06:20< iceiceice> it makes it very easey
20140412 03:06:31< gfgtdf> iceiceice: faster ? compiling c++ is never fast i think.
20140412 03:06:33< iceiceice> and i dont have to worry about the offical version
20140412 03:06:41< iceiceice> you dont compile the server?
20140412 03:06:58< iceiceice> i find the server builds with clang in < 2 minutes
20140412 03:08:02< gfgtdf> iceiceice: only when i have to, yes the server builds faser then the normal game, but even the time that the compiler needs fo find out that the server is up to data is time i'd like to save.
20140412 03:09:26< iceiceice> ok but if you ever tell me you have a networked mp issue i am going to ask you, are you using exactly the same server as associated to that commit?
20140412 03:09:33< iceiceice> because it will cause problems and you reported a problem like this to me before once
20140412 03:09:55< gfgtdf> iceiceice: i also have teh problem with 1.13.-dev
20140412 03:10:29< iceiceice> using the current server build?
20140412 03:10:37< iceiceice> because i have no idea when anyone reset the server build
20140412 03:10:42< iceiceice> on the official server
20140412 03:11:11< gfgtdf> I used a local server , i think from 1.13-dev 100%  but im not completeley sure.
20140412 03:12:41< gfgtdf> iceiceice: shadowm will can recompile the server, but i think 1.11.12 and 1.13-dev currentl use teh same offical server
20140412 03:13:00< iceiceice> no that is not true!
20140412 03:13:04< iceiceice> the server controller tweaks 
20140412 03:13:09< iceiceice> make crucial server side changes
20140412 03:13:15< iceiceice> and they are only on master right now
20140412 03:13:55< gfgtdf> iceiceice: yes thay are only on master, so ?
20140412 03:14:16< iceiceice> so they are not on 1.12
20140412 03:14:21-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 258 seconds]
20140412 03:14:21< gfgtdf> iceiceice: i mean when you click 'connect offical server' they bith connect to teh same server
20140412 03:14:25< iceiceice> oh sure
20140412 03:14:34< gfgtdf> unlike 1.10
20140412 03:14:40< gfgtdf> which uses a different srver
20140412 03:14:41< iceiceice> look, if you are having a controller problem,
20140412 03:14:51< iceiceice> its not really enough to just test 1.12 and master,
20140412 03:14:59< iceiceice> you should test individual commits that made server changes
20140412 03:15:04< iceiceice> like post PR 121,
20140412 03:15:06< iceiceice> post controller tweaks,
20140412 03:15:08< iceiceice> post PR 141 
20140412 03:15:09< iceiceice> etc
20140412 03:15:16< iceiceice> and use the right server for each build or the test is useless
20140412 03:15:20< gfgtdf> pr 141 didt make controller twaeks
20140412 03:15:54< iceiceice> yeah but it makes some changes to the sync mechanism
20140412 03:15:56< iceiceice> right?
20140412 03:15:59< gfgtdf> server changes*
20140412 03:16:04< gfgtdf> not on srver side
20140412 03:18:31-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140412 03:18:48< iceiceice> ok 
20140412 03:19:01< iceiceice> so what is your bug, how do i reproduce?
20140412 03:19:07-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 03:19:08< iceiceice> just play a netowrked game on CotB
20140412 03:19:12< iceiceice> with 1 ai side?
20140412 03:20:09< gfgtdf> with 1 null controlled (it's a hidden side), teh other 2 by the same player
20140412 03:23:41-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140412 03:38:24< gfgtdf> iceiceice: when i start teh game with 2 differenr players and then 1 player leaves an d teh other assigns to local player i dont get this bug.
20140412 03:38:38< iceiceice> i just finished build, will test now
20140412 03:57:09-!- gfgtdf [~chatzilla@f054167165.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]]
20140412 04:05:07-!- sachith500 [~kvirc@112.134.141.28] has joined #wesnoth-dev
20140412 04:08:29< iceiceice> ok i get the but,
20140412 04:08:34< iceiceice> *bug
20140412 04:08:37< iceiceice> what other builds does this happen on?
20140412 04:23:59-!- gfgtdf [~chatzilla@f054167165.adsl.alicedsl.de] has joined #wesnoth-dev
20140412 04:25:58< gfgtdf> iceiceice: teh sourc eof teh bug is, that in fb726e0610af42fa815cf340fa68c570b8bef0d3, in game.cpp line 377(after teh changes) the case (*user == owner_) && ((**side)["controller"] == "null") isn't handles properly becasue the side_controllers_[side_num] = (**side)["controller"].to_string(); isn't reached
20140412 04:26:14-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 04:26:38< iceiceice> hmm
20140412 04:26:51< iceiceice> so you think null sides are not owned by the owner on the server?
20140412 04:26:57< iceiceice> or this is verified from debugging output?
20140412 04:27:20< gfgtdf> 	iceiceice: null sides are skipped, and not owned by anyone
20140412 04:28:04< iceiceice> i will check but the way the server seems to work, the variable "user" there is a network connection type object
20140412 04:28:14< gfgtdf> iceiceice: that doesnt matter if there are at least 2 sides becase in taht case (*user == owner_) if false for at least one side
20140412 04:28:16< iceiceice> and i thought that empty sides are owned by the owner but i might ahve it wrong
20140412 04:29:03< gfgtdf> no we need controller = "null" for null controlled sides, becasue otherwise https://github.com/wesnoth/wesnoth/blob/master/src/server/game.cpp#L1060 win't work
20140412 04:29:10< gfgtdf> won't
20140412 04:30:42-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140412 04:31:18-!- gfgtdf [~chatzilla@f054167165.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]]
20140412 04:31:49< iceiceice> ok well lets try your proposed fix and test it
20140412 04:33:11< iceiceice> for what it is worth the bug does not appear just before the controller tweaks changes
20140412 04:33:17< iceiceice> so you are probably right i think
20140412 04:34:05-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140412 04:35:21< iceiceice> ohhh i see now what you mean
20140412 05:02:21-!- irker807 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140412 05:09:26-!- exciton_ [chuck-the-@95.72.225.147] has quit [Ping timeout: 255 seconds]
20140412 05:11:29-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 05:24:07-!- irker325 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140412 05:24:07< irker325> wesnoth: Chris Beck wesnoth:master 84722f567dd9 / src/server/game.cpp: fixup server controller tweaks http://git.io/Sx3ZpQ
20140412 05:24:43-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140412 05:25:28< iceiceice> gfgtdf: i have pushed a fix along the lines you mentioned, thanks for spotting this
20140412 05:49:59-!- Sulfur [~Miranda@p5B0091C4.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140412 05:54:56-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140412 06:03:52-!- Gallaecio [~quassel@84.120.115.132.dyn.user.ono.com] has quit [Remote host closed the connection]
20140412 06:06:29-!- sachith500 [~kvirc@112.134.141.28] has quit [Read error: Connection reset by peer]
20140412 06:06:46-!- sachith500 [~kvirc@112.134.141.28] has joined #wesnoth-dev
20140412 06:07:32-!- exciton_ [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 06:11:05-!- exciton [chuck-the-@95.72.225.147] has quit [Ping timeout: 255 seconds]
20140412 06:14:26-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 06:16:11-!- sachith500 [~kvirc@112.134.141.28] has quit [Read error: Connection reset by peer]
20140412 06:19:10-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140412 06:29:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 06:35:12-!- exciton_ [chuck-the-@95.72.225.147] has quit [Read error: Connection reset by peer]
20140412 06:35:25-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 06:55:28-!- exciton [chuck-the-@95.72.225.147] has quit [Read error: Connection reset by peer]
20140412 06:55:43-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 07:04:33-!- exciton [chuck-the-@95.72.225.147] has quit [Read error: Connection reset by peer]
20140412 07:04:47-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 07:08:22-!- bagzie [~bag@85-76-73-199-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140412 07:19:10-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 07:24:01-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 250 seconds]
20140412 07:32:11-!- exciton [chuck-the-@95.72.225.147] has quit [Read error: Connection reset by peer]
20140412 07:32:25-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 07:37:47-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140412 07:41:21-!- exciton [chuck-the-@95.72.225.147] has quit [Read error: Connection reset by peer]
20140412 07:41:35-!- exciton [chuck-the-@95.72.225.147] has joined #wesnoth-dev
20140412 07:45:37-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140412 07:47:03-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140412 07:47:41-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20140412 08:09:41-!- DHost [~Pcy@vps.plok.fr] has quit [Ping timeout: 246 seconds]
20140412 08:19:07-!- Elvish_Hunter [6d378ad1@gateway/web/freenode/ip.109.55.138.209] has joined #wesnoth-dev
20140412 08:19:28< Elvish_Hunter> Hi all
20140412 08:19:47-!- exciton [chuck-the-@95.72.225.147] has quit [Ping timeout: 255 seconds]
20140412 08:20:45< Coffee_irc> hi Elvish_Hunter
20140412 08:21:13< Elvish_Hunter> Hi Coffee
20140412 08:21:31-!- exciton [chuck-the-@95.72.132.71] has joined #wesnoth-dev
20140412 08:23:34< Coffee_irc> I suppose I should have another go at fixing the sometimes jerky movement bug
20140412 08:24:05-!- irker325 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140412 08:24:39< vultraz> Elvish_Hunter: are you the WLP guy?
20140412 08:25:26< Elvish_Hunter> vultraz: yep. And I already answererd to your PM.
20140412 08:26:39< vultraz> ah, I didn't see the notif
20140412 08:27:20< Elvish_Hunter> No problem ;-)
20140412 08:28:01-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140412 08:28:45-!- exciton [chuck-the-@95.72.132.71] has quit [Read error: Connection reset by peer]
20140412 08:29:00-!- exciton [chuck-the-@95.72.132.71] has joined #wesnoth-dev
20140412 08:29:23 * vultraz deletes page
20140412 08:33:03-!- mjs-de [~mjs-de@f048242163.adsl.alicedsl.de] has joined #wesnoth-dev
20140412 08:43:15-!- neXyon [~neXyon@85-127-246-188.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20140412 08:44:37-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140412 08:44:49< janebot> Wesnoth Forums | Developers’ Discussions • Re: Suggestion: keep backwards compatibility by Elvish_Hunter [ 04-12-2014 08:39 ] [ http://r.wesnoth.org/p569343 ]
20140412 08:50:34-!- irker375 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140412 08:50:34< irker375> wesnoth: Elvish_Hunter wesnoth:master c35e08dbfd07 / data/lua/wml-tags.lua: Removed sanity check for missing [do] in [while] http://git.io/Vs637A
20140412 08:50:34< irker375> wesnoth: Elvish_Hunter wesnoth:master 082992beed0c / data/lua/wml-tags.lua src/storyscreen/controller.cpp src/storyscreen/part.cpp: Removed sanity checks for missing variable=, [case] and value= in [switch] http://git.io/NwxAOQ
20140412 08:50:34< irker375> wesnoth: Elvish_Hunter wesnoth:master fc31b5651aa3 / changelog: Changelog entry http://git.io/RJ1HVQ
20140412 08:52:08-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140412 09:03:14-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20140412 09:19:21< janebot> Wesnoth Forums | Developers’ Discussions • Re: Suggestion: keep backwards compatibility by zookeeper [ 04-12-2014 09:18 ] [ http://r.wesnoth.org/p569345 ]
20140412 09:38:19< Elvish_Hunter> OK, now I have to go. Bye!
20140412 09:38:27-!- Elvish_Hunter [6d378ad1@gateway/web/freenode/ip.109.55.138.209] has quit [Quit: Ciao!]
20140412 09:39:18< janebot> Wesnoth Forums | Developers’ Discussions • Re: Suggestion: keep backwards compatibility by Elvish_Hunter [ 04-12-2014 09:37 ] [ http://r.wesnoth.org/p569349 ]
20140412 09:51:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 09:56:55-!- exciton [chuck-the-@95.72.132.71] has quit [Read error: Connection reset by peer]
20140412 09:57:09-!- exciton [chuck-the-@95.72.132.71] has joined #wesnoth-dev
20140412 10:02:00-!- exciton [chuck-the-@95.72.132.71] has quit [Read error: Connection reset by peer]
20140412 10:02:16-!- exciton [chuck-the-@95.72.132.71] has joined #wesnoth-dev
20140412 10:02:31-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has joined #wesnoth-dev
20140412 10:02:49-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 252 seconds]
20140412 10:21:20-!- exciton [chuck-the-@95.72.132.71] has quit [Remote host closed the connection]
20140412 10:26:06-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has quit [Ping timeout: 276 seconds]
20140412 10:30:12-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 10:30:45-!- DCW1 [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has joined #wesnoth-dev
20140412 10:30:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 258 seconds]
20140412 10:32:29-!- Sulfur [~Miranda@p5B0091C4.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140412 10:43:39-!- DCW1 [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has quit [Ping timeout: 276 seconds]
20140412 10:43:48-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has joined #wesnoth-dev
20140412 10:53:49-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has joined #wesnoth-dev
20140412 10:56:18-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20140412 10:56:44-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 11:03:19-!- Sulfur [~Miranda@p5B0091C4.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140412 11:14:48-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has quit [Remote host closed the connection]
20140412 11:33:54-!- Gambit [~derek@pa-67-234-76-186.dhcp.embarqhsd.net] has joined #wesnoth-dev
20140412 11:33:59-!- Gambit [~derek@pa-67-234-76-186.dhcp.embarqhsd.net] has quit [Changing host]
20140412 11:33:59-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140412 11:34:06-!- molgrum [~molgrum@212.85.89.43] has quit [Quit: Lämnar]
20140412 11:41:14-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140412 11:41:15-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140412 11:51:21-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140412 12:07:30-!- sachith500 [~kvirc@112.134.81.225] has joined #wesnoth-dev
20140412 12:25:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 12:32:55-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140412 12:44:20-!- mordante [~mordante@roadie.xs4all.nl] has joined #wesnoth-dev
20140412 12:44:20-!- mordante [~mordante@roadie.xs4all.nl] has quit [Changing host]
20140412 12:44:20-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20140412 12:44:35< mordante> servus
20140412 12:45:06-!- sachith500 [~kvirc@112.134.81.225] has quit [Read error: Connection reset by peer]
20140412 12:51:37< mordante> AI0867, ah I see where it's used now, thanks
20140412 12:51:48< mordante> AI0867, (the bit_width stuff)
20140412 12:53:47< mordante> lipkab re 1ad329c114de, I see it like std::string's c_str()
20140412 12:54:18< mordante> lipkab We could use the surface class, but then we should not call the 'free' function on it
20140412 12:54:46< mordante> lipkab I like to use how the class evolves before we make our final choose
20140412 12:55:14< mordante> lipkab however the current implementation no longer walks the two paths it did before
20140412 12:55:58< mordante> AI0867, btw I somehow have the feeling the solution might be a bit overcomplicated
20140412 12:59:31< mordante> Dugi well I read your questions and conversation with happygrue, he gave good information
20140412 13:00:03< mordante> Dugi, please don't make any assumptions regarding who and who's not going to be accepted
20140412 13:02:31< mordante> Dugi also told you before, you too often give too little information and then assume it is clear to everybody
20140412 13:02:50-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140412 13:04:56< mordante> Dugi please be more verbose in general. Just have a good look at your application and ask yourself what will be done in one summer and what not
20140412 13:05:26< mordante> dugi then write that list down, next put up the weeks when the parts are done
20140412 13:09:58-!- irker375 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140412 13:15:57< mordante> Aishiko, around?
20140412 13:33:37-!- happygrue [~happygrue@2601:6:4380:7df:59cd:e1f:3ffb:4034] has joined #wesnoth-dev
20140412 13:33:37-!- happygrue [~happygrue@2601:6:4380:7df:59cd:e1f:3ffb:4034] has quit [Changing host]
20140412 13:33:37-!- happygrue [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20140412 13:36:26-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 13:39:48< AI0867> mordante: yeah, it might be a bit overengineered.
20140412 13:56:00-!- Octalot [~noct@27.74.208.46.dyn.plus.net] has quit []
20140412 14:19:12-!- bagzie [~bag@85-76-73-199-nat.elisa-mobile.fi] has quit []
20140412 14:27:33< AI0867> mordante: btw, I've updated my boost_filesystem branch. By itself it doesn't fumctionally change anything, other than adding the dependency and bumping our min boost version to 1.44
20140412 14:30:21< AI0867> there's also my boost_filesystem_win branch which adds boost-locale and further increases the version requirement. It seems to fix unicode path handling on windows for WML files.
20140412 14:32:37< mordante> AI0867, I think Ivanovic has problems upgrading boost on the Pandora, Debian Wheezy has boost 1.49
20140412 14:32:44< AI0867> images are loaded using SDL, which in 1.2 uses the stdlib functions, so they still fail. but I have an idea foe a manual loader that could work
20140412 14:33:18< AI0867> I know, but it'a  imdependant reimplemwntation of filesystem.hpp
20140412 14:33:27< AI0867> they are intervhangable
20140412 14:33:30< mordante> did you also look at SDL 2.0 for image loading?
20140412 14:33:47< AI0867> so pandora couls continue using the old one.
20140412 14:34:00< AI0867> that would also fix things
20140412 14:34:26< AI0867> it aeems to he proper unicode handling
20140412 14:34:37 * AI0867 is typing on a phone
20140412 14:35:27< AI0867> the same thing for images probably goes for music/sound
20140412 14:36:07< mordante> we will move to SDL2 in the near future (I still like to get that project done)
20140412 14:37:17< AI0867> true.
20140412 14:38:14< AI0867> but if I turn this off by default and rebase it to 1.12, maybe we can finally have proper unicode handling on windows.
20140412 14:39:19< AI0867> without fecting other platforms
20140412 14:40:23< AI0867> +af
20140412 14:40:24< mordante> I think it might be a bit late for 1.12, seems quite intrusive 
20140412 14:40:59< AI0867> well, if I turn it off by default, it would only really to8ch qindows
20140412 14:42:28< AI0867> like my unicode refactor, besides that it's basically just renames
20140412 14:42:39< AI0867> if it compiles, it's good
20140412 14:44:38< mordante> I tend to be a bit more conservative ;-)
20140412 14:45:08< AI0867> I will weite a mail out rhis, but I'm waiting for some test results fiest
20140412 14:45:40< happygrue> Bold words, indeed.
20140412 14:46:09< happygrue> I know, I know, you're on a phone. ;)
20140412 14:47:42< AI0867> with a full keyboard so I can actuly control irssi and tmux propeely
20140412 14:48:06< AI0867> so the keys are even tinier than usual
20140412 14:49:29< AI0867> at least I'm in landscape mode. in portrait mode you pretty much need a stylus
20140412 14:49:30< mordante> but if for 1.12 it is only enabled on Windows I would be less afraid, I expect Windows to be the best tested platform by our userbase
20140412 14:53:38-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 14:54:52-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140412 14:58:41-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 15:04:20-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140412 15:04:35-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 15:14:23-!- Gallaecio [~quassel@84.120.115.132.dyn.user.ono.com] has joined #wesnoth-dev
20140412 15:15:00-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140412 15:15:52-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140412 15:21:08-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 15:36:44< AI0867> I'm looking at compiling wesnoth on windows on the wiki, and I'm missing information on the codeblocks and codelite methods
20140412 15:37:21< AI0867> is MSVC express sufficient fo the MSVC methods? should I get express 2013 for windows or express 2013 for windows desktop?
20140412 15:37:43< AI0867> desktop, it seems
20140412 15:41:00-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has joined #wesnoth-dev
20140412 15:43:14< aquileia> Hi AI0867, I just read in the logs that you want to compile on Windows yourself? The CodeBlocks description is in the readme file of the corresponding dir of the repository
20140412 15:43:49< AI0867> ah, okay
20140412 15:44:01< AI0867> I also have a tiny test program for you to, well, test
20140412 15:44:09< AI0867> to see if it might help with the image issues
20140412 15:45:12< aquileia> I'm currently trying to compile your first test file, but MSVS somehow can't resolve the sdl_rwops dependency...
20140412 15:46:08< AI0867> huh?
20140412 15:46:10< AI0867> that's basic SDL
20140412 15:46:25-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140412 15:46:27< AI0867> it's failing at link-time?
20140412 15:46:33< aquileia> Therefore I'll try it with g++ again (I thought VC might be more simple as it finds the header files itself, but obviously not the lib)
20140412 15:46:44< aquileia> yes, it's thee linker thatfails 
20140412 15:48:37< AI0867> could you also confirm whether sound/music also fails with unicode paths, like images?
20140412 15:48:45< AI0867> I suspect that it does
20140412 15:49:12< aquileia> I'll do that immediately, wait a sec...
20140412 15:50:35-!- TC01 [~quassel@128.220.109.252] has joined #wesnoth-dev
20140412 15:51:13< aquileia> AI0867: You are right, no sound
20140412 15:52:02< aquileia> all three of - error display: could not open image - error font: Failed opening font - error audio: Could not load music file
20140412 15:52:33< aquileia> as soon as the path contains non-ASCII
20140412 16:02:15< aquileia> AI0867: g++ refuses to find the SDL_image lib even if I pass their full path... back to VC, next try
20140412 16:04:07< zookeeper> Coffee_irc, https://github.com/wesnoth/wesnoth/blob/1c1022e07a8c684a78503187da5e135b2ae23f3f/data/core/units/humans/Loyalist_Spearman.cfg
20140412 16:04:15< zookeeper> the javelin hit sound was moved from -100 to 0
20140412 16:04:30< zookeeper> it's too late, even the defense anim hit sounds play at -25
20140412 16:08:10-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit []
20140412 16:12:07-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140412 16:12:31-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20140412 16:19:03< zookeeper> it seems that you can no longer use left/right arrow keys to move sliders (at least in preferences), whereas it used to work at least in 1.10.
20140412 16:27:28-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140412 16:28:03-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 16:29:18-!- kex_ [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 16:29:26-!- Kexoth [~kex@89.205.75.19] has quit [Read error: Connection reset by peer]
20140412 16:31:25-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Remote host closed the connection]
20140412 16:34:02-!- kex_ [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140412 16:39:46-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140412 16:53:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20140412 16:54:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140412 16:55:48< AI0867> "do you want to run code blocks now?" "ja" "nee"
20140412 16:55:58< AI0867> great. Partial translations that I didn't ask for
20140412 16:58:12< aquileia> AI0867: I got it to compile: I needed to list the SDL libs under "additional libraries" although I put their path in the "library dependencies"... typical Microsoft
20140412 16:58:32-!- bagzie [~bag@85-76-73-149-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140412 16:58:38< aquileia> now I wonder what exactly I should do with your test
20140412 16:59:19< aquileia> Do I pass 4 image file paths?
20140412 17:03:35-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 255 seconds]
20140412 17:05:16-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 17:08:15-!- timotei__ is now known as timotei_
20140412 17:08:18-!- timotei_ is now known as timotei
20140412 17:12:07-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140412 17:32:04< AI0867> aquileia: you pass one or more image file paths
20140412 17:32:12< AI0867> it attempts to load them in 4 ways
20140412 17:32:50< AI0867> directly with SDL_image, through SDL_RWops's own file read, which should have the same effect, manually, and manually using boost
20140412 17:33:03< AI0867> I suspect that only the 4th will work for unicode-containing paths
20140412 17:36:00< mordante> AI0867, wonderful how often Windows is convinced you want to use certain locales :-(
20140412 17:41:46-!- redcloud [5feab663@gateway/web/freenode/ip.95.234.182.99] has joined #wesnoth-dev
20140412 17:42:48< redcloud> Hi all! Is there any site where I can follow the porting to SDL 2 status?
20140412 17:49:08< loonycyborg> github?
20140412 17:50:31< redcloud> which branch?
20140412 17:50:45< loonycyborg> master
20140412 17:50:57< loonycyborg> it's currently an optional feature
20140412 17:52:23< redcloud> thanks, what's the outlook for a first playable drop?
20140412 17:53:46< mordante> redcloud, no there is no site for it, do you have any questions regarding it?
20140412 17:53:54< aquileia> AI0867: load_rw_file_boost calls convert_aux which throws an exception (the same function that caused the bug in your boost_filesystem_branch), you might have to set the locale again
20140412 17:55:05< redcloud> thanks mordante, nothing specific, just to know what's the outlook for a first playable drop
20140412 17:56:31< mordante> redcloud, not sure yet, also depends on whether that project gets a GSoC student
20140412 17:57:17-!- irker571 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140412 17:57:17< irker571> wesnoth: gfgtdf wesnoth:master f941c7008101 / src/actions/create.cpp: enable undo recruits/recalls if no rng was used http://git.io/7sWP_w
20140412 17:58:31-!- gfgtdf [~chatzilla@f054167165.adsl.alicedsl.de] has joined #wesnoth-dev
20140412 18:00:47< redcloud> ok many thanks!
20140412 18:03:57< aquileia> AI0867: I need to add std::locale loc = boost::locale::generator().generate(""); boost::filesystem::path::imbue(loc); right?
20140412 18:04:51< aquileia> oh, you already do that...
20140412 18:05:39-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Ping timeout: 276 seconds]
20140412 18:06:45-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140412 18:06:57-!- redcloud [5feab663@gateway/web/freenode/ip.95.234.182.99] has quit [Quit: Page closed]
20140412 18:12:47 * mordante tries to use a spoon to remove his eyes
20140412 18:13:58< loonycyborg> I think you're supposed to use utf8 codecvt facet there :P
20140412 18:14:00< aquileia> mordante: Could you recommend me a wiki page / forum post / existing dialog to learn how [tooltip] is used, please?
20140412 18:14:21< aquileia> loonycyborg: Like this? https://groups.google.com/forum/#!msg/boost-list/ZkB_pPpYgoA/wBLfai5xel0J
20140412 18:14:57< loonycyborg> yes, something like that
20140412 18:15:42< loonycyborg> imbue boost::filesystem with it
20140412 18:15:54< loonycyborg> iirc it has a static global imbue() method
20140412 18:16:46< aquileia> loonycyborg: AI0867 already uses imbue, but without codecvt AFAICT
20140412 18:17:26-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 18:18:23< aquileia> I'm rather new to real coding (read C++ & libraries other than std), so I might have overlooked it, though...
20140412 18:18:49< loonycyborg> locale stuff is kinda dense and rarely used
20140412 18:19:33< loonycyborg> At least not many people know how to use it properly :P
20140412 18:20:08< mordante> loonycyborg, great :-/
20140412 18:20:34< loonycyborg> I mean facets and locales in C++
20140412 18:21:08< loonycyborg> iirc there was a post in boost ml
20140412 18:21:12< mordante> aquileia, not really, what do you want to know?
20140412 18:21:21< loonycyborg> on how you use utf8 with boost.filesystem v3
20140412 18:22:05-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140412 18:22:50< aquileia> mordante: If I want to do the dialog for YAMG I'll need tooltips and I thought there might be a good example of how to usse them - and if there is, you'd probably know
20140412 18:23:18< aquileia> loonycyborg: I've two questions in parallel ;)
20140412 18:27:28< mordante> aquileia, doesn't setting the tooltip values in the gui elements work?
20140412 18:29:16< aquileia> I didn't try anything yet - using GUI2 isn't obvious, so I thought I'd ask before scanning all existing dialogs
20140412 18:29:56< mordante> gfgtdf, regarding 632ef2d6 is adding bool is_determisic_mode_; to gamestatus.hpp intended?
20140412 18:30:50< mordante> aquileia, well that part of the tooltips should be in the GUI2 wiki, just look at, for example, the title_screen.cfg
20140412 18:30:59< gfgtdf> mordante: no it's not
20140412 18:31:48< gfgtdf> mordante: shall i remove  it immedialtely ?
20140412 18:32:02< aquileia> mordante: Thanks
20140412 18:33:17< mordante> gfgtdf, yes please, cppcheck complained about its lacking in the constructor
20140412 18:34:46< irker571> wesnoth: gfgtdf wesnoth:master 4c1c487fec3d / src/gamestatus.hpp: removes unused variable http://git.io/7zQC5Q
20140412 18:35:40< gfgtdf> mordante: i removed it ^
20140412 18:37:33< mordante> thanks gfgtdf 
20140412 18:38:13< mordante> gfgtdf, why do you rebase?
20140412 18:39:15< gfgtdf> mordante: i rebased my pr to merge small bugfixes with the commits that instroduced it
20140412 18:40:25< mordante> just merging didn't work?
20140412 18:41:24< gfgtdf> mordante: no think i oversee that line in git add -p <filename> commands, when 
20140412 18:48:54-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140412 19:07:38< mordante> AI0867, are you compiling on Windows at the moment?
20140412 19:09:28< gfgtdf> mordante: do you now about the unit name generation ?
20140412 19:10:37< gfgtdf> know*
20140412 19:11:19< mordante> a bit
20140412 19:12:17< mordante> gfgtdf, ^
20140412 19:12:46< gfgtdf> mordante: i have the problme that unit name reation for undead units like ghosts still invokes (unused) random calls, which makes undoing impossible
20140412 19:12:50< gfgtdf> problem*
20140412 19:12:53< gfgtdf> creation*
20140412 19:13:14< mordante> can somebody on Windows give this patch a spin? http://paste.debian.net/93406/
20140412 19:14:45< mordante> what are »(unused) random calls« ?
20140412 19:15:14< gfgtdf> mordante: mean random is callesd, but the return value is never used,
20140412 19:16:24< gfgtdf> mordante: here: https://github.com/wesnoth/wesnoth/blob/master/src/race.cpp#L88 we call random, and we use it later, we do that because we don't want to rand a different time on different local languages.
20140412 19:17:12< gfgtdf> mordante: because on order to stay in sync we need to call rand the exact same time on all languages.
20140412 19:17:17< irker571> wesnoth: Mark de Wever wesnoth:master 851ef445ab16 / src/gui/dialogs/campaign_selection.hpp: Fix comment typos. http://git.io/v_UGKw
20140412 19:17:19< irker571> wesnoth: Mark de Wever wesnoth:master ef1e0e199b71 / src/gui/dialogs/campaign_selection.hpp: Initialise all members. http://git.io/ivAGjA
20140412 19:17:21< irker571> wesnoth: Mark de Wever wesnoth:master 5acb30af5fe6 / src/game_preferences_display.cpp: Pre instead of post increment a variable. http://git.io/8ZiZCg
20140412 19:17:23< irker571> wesnoth: Mark de Wever wesnoth:master 2f780b96bebb / src/playturn_network_adapter.cpp: Replace a increment loop with std::advance. http://git.io/q2-OuQ
20140412 19:17:25< irker571> wesnoth: Mark de Wever wesnoth:master e8116089a271 / src/playturn_network_adapter.cpp: Pre instead of post increment a variable. http://git.io/8ugTXQ
20140412 19:17:27< irker571> wesnoth: Mark de Wever wesnoth:master b90fd14075a0 / src/playturn_network_adapter.cpp: Pre instead of post increment a variable. http://git.io/mEoCig
20140412 19:17:29< irker571> wesnoth: Mark de Wever wesnoth:master e381f564c5c7 / src/playturn_network_adapter.cpp: Pre instead of post increment a variable. http://git.io/KWu89w
20140412 19:17:31< irker571> wesnoth: Mark de Wever wesnoth:master 4ca1afc41312 / src/playturn_network_adapter.cpp: Use empty() instead of comparing size() with 0. http://git.io/YgJ1ew
20140412 19:17:33< irker571> wesnoth: Mark de Wever wesnoth:master 35f0d7245469 / src/playturn_network_adapter.cpp: Use empty() instead of comparing size() with 0. http://git.io/kiDFYQ
20140412 19:17:35< irker571> wesnoth: Mark de Wever wesnoth:master b099e7c752d2 / src/map_label.cpp: Use dtor + placement new in assignment operator. http://git.io/phCZhA
20140412 19:17:37< irker571> wesnoth: Mark de Wever wesnoth:master 80d49c6011d3 / src/help.cpp: Use a variable as const ref instead of const. http://git.io/WMKYeA
20140412 19:17:39< irker571> wesnoth: Mark de Wever wesnoth:master 0840f7627d05 / src/race.cpp: Limit variable scope. http://git.io/9i64Wg
20140412 19:17:41< irker571> wesnoth: Mark de Wever wesnoth:master a3c51d2f81ae / src/replay_controller.cpp: Remove an unneeded helper variable. http://git.io/uSd12w
20140412 19:17:43< irker571> wesnoth: Mark de Wever wesnoth:master 2b943e2967b4 / src/serialization/unicode.cpp: Limit variable scope. http://git.io/o8cryA
20140412 19:17:45< irker571> wesnoth: Mark de Wever wesnoth:master 734f1be224c9 / src/ (8 files in 3 dirs): Merge branch 'cppcheck' http://git.io/aLrKMA
20140412 19:18:53< mordante> gfgtdf, only names of undead or of all units?
20140412 19:19:19< gfgtdf> mordante: undead have no names, for normal units we use teh rnadom values liekwe should
20140412 19:19:31< gfgtdf> idk if there are other units that have no names.
20140412 19:20:51< gfgtdf> the random*
20140412 19:25:27< mordante> gfgtdf, so you like to make it possible to undo the recruiting of undeads since they don't need the RNG
20140412 19:25:39< gfgtdf> yes
20140412 19:25:59< gfgtdf> that was also possible on a sooner verion of wenoth i tink
20140412 19:27:03< mordante> no idea, I can't recall feeling the need to undo a recruitment
20140412 19:27:55< gfgtdf> mordante 
20140412 19:28:56< gfgtdf> i just had 'mordante' in the chatbox left and accidently pressed enter
20140412 19:29:57< aquileia> mordante: patch for what? Or is it enough to test whether it  breaks sth.?
20140412 19:33:02< mordante> aquileia, compile testing, quite sure it works but don't like to commit it before it was compiled
20140412 19:33:31< aquileia> mordante: No problem, it'll be done
20140412 19:36:16< AI0867> mordante: I was trying to quickly set up an environment, but I won't be near that computer again for at least 2 weeks
20140412 19:37:40< AI0867> 19:35 < mordante> AI0867, wonderful how often Windows is convinced you want to use certain locales :-( ← yes, and utf-8 is not a valid locale for many purposes because it can use more than 2 characters
20140412 19:38:37< aquileia> AI0867: It seems the boost path isn't imbued correctly in your testcase... or at least it errors at the same place where your branch errored due to missing utf8 locale
20140412 19:38:53< AI0867> aquileia: could you pastebin the output of the program for various inputs? (valid and invalid) If it returns the same as boost_filesystem_win used to, I know I'm on the right track
20140412 19:39:50< mordante> thanks aquileia 
20140412 19:40:41< mordante> gfgtdf, testing a patch at the moment
20140412 19:41:13< aquileia> AI0867: You'll have to wait about half an hour, I'm currently compiling Wesnoth (usually it'd build incrementally, but I reset my solution for the unit tests)
20140412 19:41:17< gfgtdf> mordante: can i assume the undead have no names on all labguages ?
20140412 19:41:22< mordante> AI0867, ok aquileia is already testing, and git blamed you for the code ;-)
20140412 19:43:38< happygrue> special units in campaigns could be undead and have names IIRC
20140412 19:44:38< mordante> gfgtdf, the good news is, the random generator is no longer called for the undead, but still no undo
20140412 19:45:09-!- Sulfur [~Miranda@p5B0091C4.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140412 19:46:02< mordante> hmm or the hotkeys are broken
20140412 19:47:09< gfgtdf> mordante: do you have f941c700810199a585f446272d184fbabf0ed879 on your local ?
20140412 19:47:29< mordante> well they are, but that doesn't matter for this case
20140412 19:47:38< mordante> fabi__, what's the status of the hotkeys?
20140412 19:48:24< mordante> gfgtdf, no, will test my patch with that commit
20140412 19:55:30< mordante> gfgtdf, with that commit in, the undoing of recruiting works
20140412 19:55:50< mordante> will polish it and then commit it
20140412 19:56:25< mordante> btw I assume you do _not_ want to port this to 1.12
20140412 20:00:13< gfgtdf> mordante: why not ? this was also possible in 10.6
20140412 20:00:28< gfgtdf> 1.10.6*
20140412 20:00:45-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140412 20:01:37< mordante> gfgtdf, since it breaks compatibility, 1.12 should be compatible with its betas and rcs
20140412 20:02:01< gfgtdf> mordante: how does it brak compability ?
20140412 20:03:42< mordante> gfgtdf, simple create an undead undo it and create a non-undead unit, this causes an OOS against a beta player
20140412 20:04:40< gfgtdf> mordante: i was told that i can brak such copability in 1.12, especialy i was told that i can backport pr 121, which brakes even more compability.
20140412 20:05:05-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has quit [Read error: Connection reset by peer]
20140412 20:05:33< irker571> wesnoth: Mark de Wever wesnoth:master 6ec8d8026088 / changelog src/race.cpp: Test whether there is a prefix for a markov name. http://git.io/FBsPFw
20140412 20:05:44-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 20:05:51< mordante> gfgtdf, told by whom? By Ivanovic ?
20140412 20:06:06< gfgtdf> im ure it was not invanovic, but idk by who
20140412 20:06:09< gfgtdf> sure*
20140412 20:06:43-!- neXyon [~neXyon@85-127-246-188.dynamic.xdsl-line.inode.at] has quit [Quit: bye]
20140412 20:07:20-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140412 20:07:20< lipkab> mordante: Hello. I'm not sure if we've talked since I updated my proposal; do you want me to add anything else?
20140412 20:07:51< mordante> gfgtdf, then best ask Ivanovic, he's the release manager, so if he gives green light go ahead. (Btw I just committed to code)
20140412 20:10:09-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 250 seconds]
20140412 20:10:31< mordante> lipkab, btw I left some notes in the log this morning
20140412 20:11:03-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140412 20:12:31< mordante> lipkab, one major thing week 4 and week 8 I think these tasks will cost more than a week
20140412 20:13:02< mordante> these are the main areas where I expect surprises
20140412 20:13:54< mordante> (I would take a flashlight and lots of batteries with me, the surprises are probably grues)
20140412 20:15:01< mordante> btw I still have some things pending and want to look at that next week
20140412 20:15:12< mordante> but I've some other things to work on as well
20140412 20:17:34< mordante> I'm off now, bye
20140412 20:17:40< lipkab> mordante: Those are both two week tasks - week 3 and 4 and 8 and 9 go together.
20140412 20:17:41< lipkab> Bye.
20140412 20:18:23< lipkab> I think that should be enough. If not there's still two weeks padding time at the end.
20140412 20:18:40-!- molgrum [~molgrum@212.85.89.43] has joined #wesnoth-dev
20140412 20:18:41< lipkab> Re notes: fair enough.
20140412 20:20:32< mordante> maybe we have some more time to look at the schedule later this week
20140412 20:20:41< mordante> bye
20140412 20:20:46-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20140412 20:21:21< aquileia> mordante: Just one minute too late...
20140412 20:21:55< aquileia> it compiles, runs and shows the tray icon
20140412 20:28:06-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140412 20:33:11-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 276 seconds]
20140412 20:35:11-!- lipkab_ [~lipkab@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140412 20:35:18-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has quit [Ping timeout: 240 seconds]
20140412 20:38:35-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Quit: leaving]
20140412 20:48:43-!- trewe [~trewe@244.163.54.77.rev.vodafone.pt] has joined #wesnoth-dev
20140412 20:50:06-!- gfgtdf [~chatzilla@f054167165.adsl.alicedsl.de] has quit [Ping timeout: 245 seconds]
20140412 20:52:46< aquileia> AI0867: http://pastebin.com/jJDFZ8HJ
20140412 20:56:59< AI0867> aquileia: could you replace all the cout with cerr and retry that?
20140412 20:57:10< aquileia> ok
20140412 21:06:39< aquileia> I can't see an difference between stdout and stderr output, but here it is http://pastebin.com/ufRiGTU2
20140412 21:06:49< aquileia> *any
20140412 21:09:03< aquileia> AI0867: ^
20140412 21:15:45< AI0867> aquileia: reconsidering what I wrote, I am unsurprised
20140412 21:15:55< AI0867> could you interleave the load and print statements?
20140412 21:16:14< AI0867> sorry I didn't think this through
20140412 21:16:38< aquileia> I'll have to google what you said... but I probably can
20140412 21:17:02< AI0867> I mean, file1 = foo; stuff(file1); file2 = bar; stuff(file2);
20140412 21:17:21< AI0867> if the 4th load crashes it, then of couse it never reaches the printing function
20140412 21:17:58< aquileia> ok
20140412 21:21:23< aquileia> NULL NULL 2592x1936: 370795886
20140412 21:21:44< aquileia> so the third attempt seems to suffice, no need for the fourth
20140412 21:22:02< aquileia> AI0867: ^
20140412 21:24:28< aquileia> and... no need to find the error source  of the fourth
20140412 21:26:33< aquileia> AI0867: That means that the Wesnoth bug can be circumvented by loading manually, right?
20140412 21:26:40< AI0867> yes
20140412 21:27:37< AI0867> aquileia: does the third one still work if you move the locale initing between the third and the fourth?
20140412 21:28:18< aquileia> I'll try that, but before that I thought that I will omit the chcp - just in case...
20140412 21:28:45< AI0867> that would be interesting, if you can fix the std functions by imbuing boost::filesystem with a boost::locale generated std::locale
20140412 21:28:59< AI0867> right
20140412 21:29:10< AI0867> your commandline arguments might fail then though
20140412 21:29:21< AI0867> because they're fed into the program using the same codepage
20140412 21:29:35< AI0867> so you may have to hardcode them
20140412 21:32:00-!- exciton_ [chuck-the-@95.73.184.123] has joined #wesnoth-dev
20140412 21:33:12< aquileia> I was definitely in another codepage as it displays C:\Users\Sebastian\Pictures\Ausfl³ge\schwan.jpg
20140412 21:33:36< aquileia> ³ instead of ü
20140412 21:33:51< aquileia> but it still works: NULL NULL 2592x1936: 370795886
20140412 21:33:53-!- lipkab_ [~lipkab@host-91-147-212-189.biatv.hu] has quit [Ping timeout: 252 seconds]
20140412 21:35:23-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 255 seconds]
20140412 21:37:32< AI0867> yes, because the codepages match
20140412 21:37:41< AI0867> that doesn't tell us very much, unfortunately
20140412 21:37:50< AI0867> no, actually it does
20140412 21:37:58-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140412 21:38:11< AI0867> it tells us that just using std stuff won't work in anything other than codepage 65001
20140412 21:38:24< aquileia> AI0867: the locale isn't necessary for 3 - I moved it without consequences
20140412 21:38:51< AI0867> and if you hardcode in the path, using proper utf-8?
20140412 21:38:58< AI0867> I'm guessing it will fail
20140412 21:39:05< aquileia> I'll try...
20140412 21:41:42< aquileia> As Visual Studio ruined my utf-8 encoding the last time I tried (you probably remember the unit test you fixed)... could you tell me the utf-8 code for 'ü' ?
20140412 21:42:04< aquileia> If I hardcode it, it's better to be sure with that
20140412 21:42:19< AI0867> googling it told me U+00FC
20140412 21:42:33< AI0867> which is 0xC3BC in utf-8
20140412 21:43:51-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has quit [Ping timeout: 245 seconds]
20140412 21:44:36< aquileia> AI0867: Do I need "0xC3BC" or "\uC3BC", then?
20140412 21:46:22< AI0867> \u00FC
20140412 21:46:31-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140412 21:46:42< AI0867> \u describes 4-nibble codepoints
20140412 21:46:47< AI0867> \U describes 8-nibble codepoints
20140412 21:47:09< AI0867> \x describes raw hex
20140412 21:47:39< aquileia> sorry to  bother you with these things, but  don't want to mess that up...
20140412 21:47:45< AI0867> no problem
20140412 21:47:57< AI0867> I'm asking you to test stuff for me, so…
20140412 21:52:11< aquileia> AI0867: http://pastebin.com/bLxgM98W
20140412 21:52:42-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has quit [Ping timeout: 240 seconds]
20140412 21:54:47< AI0867> and that's with a random codepage?
20140412 21:54:55< aquileia> AI0867: and the current state of main: http://pastebin.com/erkLCAUk
20140412 21:54:59< AI0867> same result if you lead with chcp 1252?
20140412 21:55:37< aquileia> yes
20140412 21:55:42< AI0867> interesting
20140412 21:55:48< AI0867> I have no idea how this is happening
20140412 21:56:07< AI0867> especially since #3 is essentially how #2 is implemented by SDL internally
20140412 21:57:02< aquileia> perhaps my compiler handles it differently?
20140412 21:57:18< AI0867> I have no idea
20140412 21:57:27< AI0867> it's not the same of course. SDL is written in C, not C++
20140412 21:57:48< AI0867> so it uses fopen, not std::ifstream
20140412 21:58:13< aquileia> Shall I do a last test by removing the hardcoded one and passing it under cp 1252?
20140412 21:58:34< AI0867> okay
20140412 21:58:42< AI0867> I'd have to think about what the results might tell us though
20140412 21:59:26-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 258 seconds]
20140412 22:01:34< aquileia> it works as well... I thought  it might error, but no
20140412 22:01:51< AI0867> I'm just confused now
20140412 22:02:37< aquileia> so std::ifstream is probably a lot more advanced than fopen? Or am I the confused one?
20140412 22:04:18< aquileia> AI0867: In any case, it seems you might have a relatively easy fix for Wesnoth!
20140412 22:05:41 * aquileia is probably too ignorant to see all the implications...
20140412 22:09:05-!- lipkab [~lipkab@host-91-147-212-189.biatv.hu] has quit [Ping timeout: 252 seconds]
20140412 22:12:34< aquileia> AI0867: In case you want to apply your fix, may I suggest to try sound.cpp line 485-500 first? It should be relatively easy to fix the background music as it's only 15 lines, and it's easy to test as well. A perfect testcase IMHO.
20140412 22:19:47< aquileia> Except if MixLoadMUS (the SDL call for music) is complicated, of course... I don't have looked into it
20140412 22:24:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140412 22:30:20< AI0867> Mix_LoadMUS_RW seems to be undocumented for a reason
20140412 22:30:55< AI0867> that is, while Mix_LoadWAV is implemented simply as Mix_LoadWAV_RW(SDL_RWopsFromFile), MIX_LoadMUS has lots of extra logic
20140412 22:31:16< AI0867> and it's causing me segfaults
20140412 22:31:43-!- bagzie [~bag@85-76-73-149-nat.elisa-mobile.fi] has quit []
20140412 22:32:56< AI0867> in fact, it doesn't use MIX_LoadMUS_RW, it uses MIX_LoadMUSType_RW
20140412 22:33:17< AI0867> and Mix_LoadMUS_RW is implemented as that function with default arguments
20140412 22:34:24< aquileia> I can't comment on that, I only found information on the backend but not the code
20140412 22:34:53< AI0867> if I ditch that particular one though, it seems to work fine
20140412 22:34:58< AI0867> I'll commit this so you can test
20140412 22:39:07< aquileia> AI0867: damn, Mix_LoadMUSType_RW uses fopen as well... so you might have to emulate that as well
20140412 22:39:21< aquileia> perhaps another testacse is more appropriate
20140412 22:40:27< aquileia> on the other hand, it's the easiest one on Wesnoth's side - just one single place where it interacts with SDL loading
20140412 22:40:51< aquileia> it's your choice of course - your code, your decision
20140412 22:41:18< aquileia> I'll happily test whaatever you throw at me
20140412 22:42:15< AI0867> aquileia: pushed to boost_filesystem_win
20140412 22:42:23< aquileia> ok
20140412 22:42:28< AI0867> you'll need to add a new file to your project
20140412 22:42:34< AI0867> src/filesystem_sdl.cpp
20140412 22:44:20< AI0867> aquileia: it uses fopen? where?
20140412 22:45:17< aquileia> If http://lists.libsdl.org/pipermail/commits-libsdl.org/2012-January/005148.html is up to date, it uses it directly
20140412 22:45:33< aquileia> I found no other source
20140412 22:45:36-!- haudegen_ [~quassel@91.141.3.152.wireless.dyn.drei.com] has joined #wesnoth-dev
20140412 22:46:05< aquileia> below the comment "/* Figure out what kind of file this is */"
20140412 22:46:07 * AI0867 reads
20140412 22:46:54< AI0867> right. I have sources from *after* that diff
20140412 22:47:01< aquileia> ah, ok
20140412 22:47:12< AI0867> but given that this is 2012, my debian libraries probably predate the diff
20140412 22:47:23< AI0867> which certainly explains a few things
20140412 22:48:12< aquileia> oh... my lib version is from 2009 I  think (it's the standard one packed for MSVS by anonymissimus)
20140412 22:48:49< aquileia> So I'll have to update SDL before, and I certainly won't have the time for that tonight, sorry
20140412 22:48:50< AI0867> yeah, that's probably going to fail hard if you uncomment the commented code I added in that commit
20140412 22:48:58< AI0867> no, I commented the music bit out
20140412 22:49:01< AI0867> so that's not an issue
20140412 22:49:33< aquileia> I guess it's better to update, though?
20140412 22:50:53< AI0867> meh, that would only tell us that we'd either have to require everyone else to update too, or we'd still be screwed
20140412 22:51:20< AI0867> s/or we'd/or that we'd/
20140412 22:53:12< aquileia> Well, I can let Ivanovic update the Windows library package provided on the wiki, and with an #ifdef it would only affect us few windows compilers
20140412 22:53:55< AI0867> true
20140412 22:54:04< aquileia> but I can try before updating, of course
20140412 22:54:05< AI0867> but I'd first like to check if this actually helps =P
20140412 22:54:20< aquileia> yes, true
20140412 22:54:40< AI0867> it should work for images and sounds, just not music
20140412 22:55:00< AI0867> 'should' being the operative word, of course
20140412 22:55:01< aquileia> oh, so you already fixed those?
20140412 22:55:18< AI0867> https://github.com/AI0867/wesnoth/commit/b502746ec997d5d835c9bc5b382c700210eb2f66
20140412 22:56:11-!- mjs-de [~mjs-de@f048242163.adsl.alicedsl.de] has quit [Remote host closed the connection]
20140412 23:02:10< aquileia> AI0867: One thing I should have told you before... it's
20140412 23:02:11< aquileia> #include <SDL_rwops.h>
20140412 23:03:09< aquileia> there's no SDL subdir in the include path
20140412 23:08:55-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has joined #wesnoth-dev
20140412 23:09:09< aquileia> While it builds I can already paste you the warnings it gave me: http://pastebin.com/iuGHs5Z2
20140412 23:10:11< aquileia> by the way, the second one (third line) was already there in your sdl.cpp, but it didn't matter
20140412 23:11:30< aquileia> The simple solution would be to ignore them ;)
20140412 23:15:36-!- travis-ci [~travis-ci@ec2-54-237-114-110.compute-1.amazonaws.com] has joined #wesnoth-dev
20140412 23:15:36< travis-ci> [travis-ci] AI0867/wesnoth#41 (boost_filesystem_win - b502746 : Alexander van Gessel): The build was broken.
20140412 23:15:36< travis-ci> [travis-ci] Build details : http://travis-ci.org/AI0867/wesnoth/builds/22862275
20140412 23:15:36-!- travis-ci [~travis-ci@ec2-54-237-114-110.compute-1.amazonaws.com] has left #wesnoth-dev []
20140412 23:18:36< aquileia> AI0867: Ok, you can't ignore the first one - clang is a lot stricter than gcc
20140412 23:19:05-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 250 seconds]
20140412 23:19:25-!- TC01 [~quassel@128.220.109.252] has quit [Remote host closed the connection]
20140412 23:19:27-!- haudegen_ [~quassel@91.141.3.152.wireless.dyn.drei.com] has quit [Remote host closed the connection]
20140412 23:20:48< aquileia> AI0867: Out of curiosity, why do you need that line (class SDL_RWops) at all?
20140412 23:21:09-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has quit [Remote host closed the connection]
20140412 23:21:43< AI0867> so you don't have to include the header defining it in that header
20140412 23:22:16< aquileia> should I change it to struct, then?
20140412 23:22:54< AI0867> yes
20140412 23:24:17< aquileia> the warning is gone, so travis should be fine as well with that tiny chaange
20140412 23:24:22< aquileia> *change
20140412 23:24:37< AI0867> yeah, I'm also fixing the other warning, then force-pushing
20140412 23:25:08< AI0867> done
20140412 23:26:20< aquileia> I won't pull it though, I think it currently does a complete rebuild, sigh...
20140412 23:26:42< AI0867> MSVC?
20140412 23:27:09< aquileia> Yes, I did a single misclick and bang
20140412 23:27:10< AI0867> I think it *always* does that. It doesn't really support partial builds
20140412 23:28:25< aquileia> No, it does partial builds, it's my fault
20140412 23:28:29< AI0867> anyway, I'm about to leave for probably the rest of the weekend
20140412 23:28:44< AI0867> I'd heard otherwise, but I've hardly used it myself
20140412 23:29:05< AI0867> "about to leave" probably means somewhere in the next 5-30 minutes
20140412 23:29:06< aquileia> ok, you might want to check the logs, then
20140412 23:29:08< AI0867> sure
20140412 23:29:18< AI0867> not that my clients is likely to go anywhere
20140412 23:29:35< AI0867> s/clients/client
20140412 23:30:25-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140412 23:30:34< aquileia> if it's 20-30 min it will probably build in that time, but it's no pressing matter, so feel free to leave before that
20140412 23:32:54< aquileia> I can guarantee that it tries its best on a minimal build. shadowm said that MSVC doesn't use multiple cores if rebuilding, though
20140412 23:33:08< aquileia> so it has its quirks, but it's ok
20140412 23:37:18-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140412 23:39:25-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has quit [Quit: leaving]
20140412 23:42:19< aquileia> AI0867: Thee build succeded, let's see if it loads correctly...
20140412 23:42:23< aquileia> *The
20140412 23:44:18< aquileia> AI0867: Unfortunately the error is still there
20140412 23:47:09< AI0867> as in, it doesn't load the stuff, or it crashes like the test program?
20140412 23:47:42< aquileia> it doesn't load it
20140412 23:48:07< AI0867> neither images nor sounds?
20140412 23:48:09< AI0867> okay =/
20140412 23:48:15< AI0867> yet the WML works?
20140412 23:48:18< aquileia> yep, neither
20140412 23:49:00-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140412 23:49:50< aquileia> I'll have to test that, but I think it doesn't work correctly with a unicode path as well, but I'll have to debug first
20140412 23:51:38-!- exciton_ [chuck-the-@95.73.184.123] has quit [Read error: Connection reset by peer]
20140412 23:51:52-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140412 23:52:06< aquileia> I first started it normally to be a little bit faster
20140412 23:52:20-!- travis-ci [~travis-ci@ec2-54-204-141-91.compute-1.amazonaws.com] has joined #wesnoth-dev
20140412 23:52:20< travis-ci> [travis-ci] AI0867/wesnoth#42 (boost_filesystem_win - 33561aa : Alexander van Gessel): The build was fixed.
20140412 23:52:20< travis-ci> [travis-ci] Build details : http://travis-ci.org/AI0867/wesnoth/builds/22863684
20140412 23:52:20-!- travis-ci [~travis-ci@ec2-54-204-141-91.compute-1.amazonaws.com] has left #wesnoth-dev []
20140412 23:53:28< aquileia> ahm... what? 
20140412 23:54:05< aquileia> Ah... I forgot that the Debug build is in another directory!
20140412 23:54:42< aquileia> AI0867: I withdraw my report: Images are fixed!
20140412 23:55:20< aquileia> sounds are fixed
20140412 23:55:22-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140412 23:56:32< aquileia> the GUI of the preferenced dialog is screwed as the text doesn't load, but you didn't try to fix the font loading I think, so that's all right
20140412 23:57:35< AI0867> =)
20140412 23:57:53< AI0867> put the font loading next on my TODO list
20140412 23:58:24< aquileia> AI0867: you completely fixed images and sound, and I'm currently loading thee tutorial in debug speed (that is - for the next thre minutes)
20140412 23:59:16< AI0867> well, I'll be off now. It's good to hear that this seems to work. Maybe we can manage to squeeze proper unicode support into 1.12 after all
--- Log closed Sun Apr 13 00:00:07 2014