--- Log opened Sun Dec 28 00:00:53 2014
20141228 00:08:24< gfgtdf> Necrosporus: yes it works the same way.
20141228 00:08:39< gfgtdf> Necrosporus: it basicly loads the autosave
20141228 00:10:40-!- happygrue [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141228 00:20:35-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141228 00:21:34< Necrosporus> Also are you aware then when you command a unit to move and fight, it still doesn't fight if its movement uncovers some shroud or fog
20141228 00:21:47< Necrosporus> The old bug from 1.11
20141228 00:27:41< gfgtdf> Necrosporus: is there a bug report ?
20141228 00:28:39< Necrosporus> I do not know
20141228 00:29:21< Necrosporus> But it's quite obvious, so it probably must be
20141228 00:30:16< Necrosporus> Also I'm not completely sure about conditions when this behavior triggers, but seems like it's when you uncover shroud or fog
20141228 00:30:49-!- irker731 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20141228 00:32:37-!- sachith500 [~kvirc@101.189.46.81] has quit [Ping timeout: 240 seconds]
20141228 00:34:38< Necrosporus> Though it's not quite annoying like old bug was (where move and attack never worked)
20141228 00:35:50< gfgtdf> Necrosporus: this happens in minline campaigns ?
20141228 00:35:56< gfgtdf> mainline*
20141228 00:38:05< Necrosporus> not sure, I first encountered in umc one
20141228 00:39:26< Necrosporus> The problem of bug reporting is most people somehow assume the developers to be aware of their problems so they do not report
20141228 00:39:37< Necrosporus> I guess so at least
20141228 00:40:08< Necrosporus> It's often so also
20141228 00:41:22< gfgtdf> Necrosporus: usually developers alrady know that a bug exist igf there is already a bug report 
20141228 00:50:56< Necrosporus> Anyway, have you checked?
20141228 00:51:06< Necrosporus> I guess it must be easy to check if I'm right
20141228 00:52:05< Necrosporus> If not and there are no bugreport yet I might try to find exact condition where it could be reproduced
20141228 00:58:13< Necrosporus> Also there is a problem in addon dialog, as there is no button to choose which addons to upgrade and which not to, only 'update all' but if you already finished one addon you might want to avoid upgrading it
20141228 00:59:45< Ravana_> I don't see need for that, if you want to permanently not update some addon just delete its info file
20141228 01:06:32-!- TooLmaN__ [~TooLmaN@mx1.thomsonplastics.com] has joined #wesnoth-dev
20141228 01:08:36-!- TooLmaN_ [~TooLmaN@c-73-190-210-54.hsd1.ga.comcast.net] has quit [Read error: Connection reset by peer]
20141228 01:11:02< Necrosporus> So it would be playable?
20141228 01:11:36< Necrosporus> Ravana_, it's a bit hacky. If noone sees problems then there's no need to have addon manager at all
20141228 01:11:54< Necrosporus> One could browse, download and unpack addons from site
20141228 01:11:55-!- TooLmaN__ [~TooLmaN@mx1.thomsonplastics.com] has quit [Ping timeout: 252 seconds]
20141228 01:12:41< Ravana_> generally you are supposed to accept addon updates so there would be ideological reason for this inability
20141228 01:13:21< Necrosporus> Not necessarily
20141228 01:13:47< Necrosporus> For example I finished playing a campaign but do not want its icon with laurels to be removed from list
20141228 01:15:11< Ravana_> not sure what happens with laurels, I have only ever completed one campaign
20141228 01:15:33< Necrosporus> mainline?
20141228 01:15:45< Ravana_> no, To Lands Unknown
20141228 01:16:40< Necrosporus> Campaign icons disappear obviously
20141228 01:17:34< Ravana_> I thought that was supposed to be saved in preferences not in addon files
20141228 01:18:33< Necrosporus> I guess it's saved in preferences
20141228 01:18:55< Necrosporus> Also I guess it would reappear with laurels, but not sure
20141228 01:19:29< Ravana_> I believe that as long as campaign id stays same laurels will remain
20141228 01:21:00-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141228 01:25:14-!- cib_ [~cib@tmo-103-81.customers.d1-online.com] has quit [Ping timeout: 245 seconds]
20141228 01:42:58-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has joined #wesnoth-dev
20141228 02:07:49-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has quit [Remote host closed the connection]
20141228 02:35:24< Necrosporus> Is it possible to not make autosaves?
20141228 02:38:17< fabi> Necrosporus: Yes, I believe it is.
20141228 02:38:55< fabi> By setting the Maximum Number of Autosaves to zero.
20141228 02:39:38< fabi> Necrosporus: See in Prefernces -> General -> Maximum auto-saves:
20141228 02:39:48< Necrosporus> But I do not want to delete existing autosaves
20141228 02:40:00< Necrosporus> I just want to prevent making new for certaing campaign
20141228 02:40:10< Necrosporus> It generates 5 MiB saves
20141228 02:40:28< fabi> oh
20141228 02:40:55< fabi> Which campaign is it?
20141228 02:48:20-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20141228 02:50:25-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has quit [Quit: internet issues...]
20141228 02:52:03< Necrosporus> fabi, Strange Legacy
20141228 02:54:32< Ravana_> command :nosaves
20141228 02:59:23-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has joined #wesnoth-dev
20141228 03:00:11< fabi> Ravana_: Good to know. Is there a prefernces option?
20141228 03:01:16< Ravana_> not sure, either way preferences would affect all games, command is just for one session
20141228 03:12:56< fabi> Ravana_: I see. Thank you for giving the hint. Maybe the campaign "Strange Legacy" should mention the command.
20141228 03:13:59-!- Ivanovic_ [~ivanovic@frnk-5f74d56e.pool.mediaWays.net] has joined #wesnoth-dev
20141228 03:16:36-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 250 seconds]
20141228 03:16:51< shadowm> fabi: Hi. When could you work on that backpo[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[D[Drt?
20141228 03:17:00< shadowm> backport
20141228 03:17:53-!- Ivanovic_ is now known as Ivanovic
20141228 03:25:37-!- sachith500 [~kvirc@101.189.46.81] has joined #wesnoth-dev
20141228 03:33:00-!- sachith500 [~kvirc@101.189.46.81] has quit [Ping timeout: 264 seconds]
20141228 03:43:48-!- gfgtdf [~chatzilla@d231237.adsl.hansenet.de] has quit [Ping timeout: 264 seconds]
20141228 03:44:05-!- gfgtdf [~chatzilla@d231237.adsl.hansenet.de] has joined #wesnoth-dev
20141228 03:49:12-!- gfgtdf [~chatzilla@d231237.adsl.hansenet.de] has quit [Read error: Connection reset by peer]
20141228 03:49:14< mattsc> I have a dumb question …
20141228 03:49:46< mattsc> What would happen if I copied a symbol like this http://imgur.com/gXFjIXc into a Wesnoth message.  On other platforms, I mean, it shows just fine on mine.
20141228 03:49:59< mattsc> ☞
20141228 03:50:00< mattsc> WHITE RIGHT POINTING INDEX
20141228 03:50:00< mattsc> Unicode: U+261E, UTF-8: E2 98 9E
20141228 03:50:26< mattsc> (actually, what I really want is something roughly resembling breath marks)
20141228 03:52:57-!- sachith500 [~kvirc@101.189.46.81] has joined #wesnoth-dev
20141228 03:55:01< shadowm> You'd need to make sure DejaVu Sans includes that glyph.
20141228 03:56:01< shadowm> Though since [message] is implemented via GUI2, font rendering is subject to fontconfig and Cairo's whims, so it's probably going to be displayed in a fallback font in the worst case.
20141228 03:59:38< mattsc> I see, thanks.
20141228 04:00:25< mattsc> 'RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK' (U+00BB) is the closest I have found so far, which seems to be included.
20141228 04:00:48< vultraz> isn't human_ai removed now?
20141228 04:14:39< Necrosporus> Wesnoth isn't really suited to be a RPG engine
20141228 04:15:03< vultraz> what brought that up
20141228 04:17:48< Necrosporus> it's slow
20141228 04:18:10< Necrosporus> And seems too hard to implement features bugfree
20141228 04:19:25< vultraz> No
20141228 04:19:39< vultraz> more like a campaign designer is obliged to make sure his work is bug free
20141228 04:19:56< vultraz> and a majority of UMC authors do not
20141228 04:20:04< vultraz> I can think of a notable example...
20141228 04:31:28-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Remote host closed the connection]
20141228 04:40:40< Necrosporus> vultraz, it's quite hard to write bug-free wml, sometimes it's almost impossible
20141228 04:43:52< vultraz> why do you say that?
20141228 04:48:14-!- happygrue [~Laptop@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20141228 04:49:45< Necrosporus> vultraz, because I tried
20141228 04:50:15< Necrosporus> Though it depends of what you are trying to do, if something like An Orcish incursion it's easy
20141228 04:50:29< Necrosporus> But if it's something complex with lots of events and items it's harder
20141228 04:53:15-!- Coffee_irc [~david@ppp118-210-58-226.lns20.adl2.internode.on.net] has quit [Ping timeout: 244 seconds]
20141228 04:58:49< Necrosporus> I wonder why even load menu could lag
20141228 05:11:37-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has quit [Ping timeout: 240 seconds]
20141228 05:12:37-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Bye for now]
20141228 05:32:02-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20141228 05:33:05-!- new_one [~new_one@50.15.126.83] has quit [Quit: WeeChat 1.0.1]
20141228 05:53:37< Necrosporus> There is a good side though, this thing discourages saveloading as it's not very pleasant to wait few minutes
20141228 05:54:36< shadowm> Minutes? How large is your saves dir?
20141228 05:55:35< Necrosporus> About 5MiB
20141228 05:56:34< shadowm> Is it literally minutes or is that just hyperbole?
20141228 05:58:02< Necrosporus> I guess at least a half of minute
20141228 05:58:14< shadowm> Would you mind uploading your dir as an archive somewhere?
20141228 05:58:35< Necrosporus> I mean loading a save
20141228 05:58:40< Necrosporus> Not opening the dialog
20141228 05:58:42< shadowm> Oh.
20141228 05:58:59< shadowm> Is it the same add-on with ginormous saves you were referring to previously?
20141228 05:59:25< Necrosporus> Dialog takes about 10 seconds I guess but only sometimes, I guess it's due to attempt to parse the top savefile
20141228 05:59:31< Necrosporus> Yes
20141228 05:59:38< shadowm> Okay then.
20141228 05:59:39< Necrosporus> Strange Legacy
20141228 06:01:55< Necrosporus> Also 5mib one save, not all
20141228 06:02:10< Necrosporus> All would be 158 mib
20141228 06:02:18< shadowm> And here I thought my add-ons were resource-demanding.
20141228 06:02:37-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has quit [Ping timeout: 240 seconds]
20141228 06:02:56< Necrosporus> Also Danse Macabre takes about 3MiB per save
20141228 06:03:01< shadowm> RSS goes from 83.8 MiB to 873.1 MiB with this thing.
20141228 06:03:15< Necrosporus> RSS?
20141228 06:03:23< shadowm> Er, 69.7 to 873.1 MiB.
20141228 06:05:12< shadowm> The flat WML is 93 MiB.
20141228 06:05:39< shadowm> The unparsed WML is 557 MiB.
20141228 06:05:49< shadowm> Clearly someone is abusing WML preprocessor macros.
20141228 06:06:04< shadowm> And/or file inclusions, but it's most likely macros.
20141228 06:09:21< Necrosporus> What are you talking about? 5 MiB is compressed save
20141228 06:09:44< shadowm> I'm not talking about saves.
20141228 06:10:13< vultraz> 557 MB for WML?
20141228 06:10:15< vultraz> wtf
20141228 06:10:42< Necrosporus> Also uncompressed save would be something like 84771966 bytes
20141228 06:11:38-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has quit [Quit: I'm done for today. I'll be back later.]
20141228 06:15:45-!- oldlaptop [~quassel@static-50-108-31-113.mskg.mi.frontiernet.net] has joined #wesnoth-dev
20141228 06:18:25< Necrosporus> Does Wesnoth unpack entire savefile when it needs to show icon and other stuff?
20141228 06:18:37< Necrosporus> Might it be changed to unpack only header?
20141228 06:18:56< Necrosporus> So enormous savefile won't cause lags in load dialog
20141228 06:20:56< vultraz> what addon is this?
20141228 06:21:19< shadowm> 02:59:41 <Necrosporus> Strange Legacy
20141228 06:27:57< vultraz> from what he says, it sounds horribly unoptimized 
20141228 06:28:50< Necrosporus> But it's really complex
20141228 06:31:46< vultraz> so?
20141228 06:31:58< vultraz> After the Storm is complex, and optimized
20141228 06:32:11< vultraz> Lazy authors are more often the problem then wesnoth
20141228 06:32:58< Necrosporus> after the storm doesn't have inventory system, lockpicking, freemap and other stuff
20141228 06:33:08< shadowm> shadowm@nanacore:~/src/wesnoth-1.12% du -h pptest-AtS/_main.cfg.plain
20141228 06:33:10< shadowm> 31M     pptest-AtS/_main.cfg.plain
20141228 06:33:15< shadowm> shadowm@nanacore:~/src/wesnoth-1.12% du -h pptest-Trader/_main.cfg.plain
20141228 06:33:18< shadowm> 557M    pptest-Trader/_main.cfg.plain
20141228 06:33:41< shadowm> ~1.5s vs. ~12.5s load time.
20141228 06:35:08< vultraz> Lockpicking?
20141228 06:35:14< vultraz> Interested to see how they implemented that
20141228 06:36:02< shadowm> And RSS is 181.3 MiB vs. 873.1 MiB.
20141228 06:36:48< Necrosporus> What is RSS?
20141228 06:36:51< shadowm> I expect 75% of the first to be image cache (inc. builder-generated entries).
20141228 06:37:23< Necrosporus> Uncompressed wml?
20141228 06:37:28< Necrosporus> Why rss?
20141228 06:37:42< shadowm> Resident set size.
20141228 06:41:41-!- [Relic] [~Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141228 07:05:09-!- Ivanovic [~ivanovic@frnk-5f74d56e.pool.mediaWays.net] has quit [Changing host]
20141228 07:05:09-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20141228 07:09:52< shadowm> lol replay_controller is in charge of registering era events for MP.
20141228 07:10:11< shadowm> This ridiculous task delegation design.
20141228 07:18:29< vultraz> wat
20141228 07:19:17-!- iceiceice [~chris@2604:2000:1481:a028:a190:b15f:f4be:511f] has joined #wesnoth-dev
20141228 07:19:28-!- iceiceice [~chris@2604:2000:1481:a028:a190:b15f:f4be:511f] has quit [Changing host]
20141228 07:19:28-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141228 07:29:47< vultraz> hello iceiceice 
20141228 07:29:54< iceiceice> hi vultraz
20141228 07:41:06< shadowm> Whoever can tell me who registers era events outside of replay_controller earns a cookie.
20141228 07:44:03< iceiceice> shadowm: https://github.com/wesnoth/wesnoth/blob/master/src/saved_game.cpp#L252
20141228 07:53:51< shadowm> Seriously?
20141228 07:54:37< vultraz> iceiceice: may I stick string.gmatch(names, "[^%s,][^,]*") into a split() function?
20141228 07:55:21< iceiceice> yeah i think that's a good thing
20141228 07:55:32< iceiceice> it will promote consistency
20141228 08:07:54-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141228 08:10:21-!- Sulfur [~Miranda@p5B0089BA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141228 08:15:53< vultraz> IRKER
20141228 08:15:53-!- irker618 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20141228 08:15:53< irker618> wesnoth: Charles Dang wesnoth:master 9b6f357a2724 / data/lua/wml-tags.lua: Put split-string-at-comma code into a function http://git.io/tg1iaA
20141228 08:15:54< irker618> wesnoth: Charles Dang wesnoth:master ec985105ef41 / data/lua/wml-tags.lua: Eliminate local var from trim() http://git.io/Wud9kA
20141228 08:31:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141228 08:39:54< vultraz> -- execute [do] up to 65536 times
20141228 08:40:04 * vultraz wonders who pulled that number out of a hat
20141228 08:41:28< shadowm> That limit predates the addition of Lua support.
20141228 08:42:56< vultraz> That doesn't answer where it came from
20141228 08:42:59< shadowm> It's 2^16. Completely arbitrary, it's a half-hearted attempt to prevent infinite loops in WML. I advise you don't question it.
20141228 08:43:52< shadowm> Just like we don't question the fact that the game is called "Wesnoth" and not "Software Bug Creator 3000".
20141228 08:44:14< vultraz> heh
20141228 08:44:42< vultraz> 65536 might as well be infinite
20141228 08:45:21< vultraz> who is going to wait for 65000 cycles?
20141228 08:45:36< shadowm> That doesn't take as long as you think.
20141228 08:46:57< shadowm> int main() { for (int i = 0; i < 65536; ++i); return 0; }
20141228 08:47:07< shadowm> This C11 program takes 0.001 seconds to run.
20141228 08:47:19< shadowm> Unoptimized.
20141228 08:49:10< shadowm> Though I guess the compiler still does some basic optimization there so never mind.
20141228 08:49:40< shadowm> I stuck a usleep(1) call and now it takes 3.539 seconds.
20141228 08:51:04< shadowm> Of course Wesnoth may do a lot more things during a loop step but it's still likely to go unnoticed e.g. in prestart.
20141228 08:55:46-!- travis-ci [~travis-ci@ec2-54-81-49-92.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 08:55:46< travis-ci> wesnoth/wesnoth#5286 (master - ec98510 : Charles Dang): The build has errored.
20141228 08:55:46< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45265860
20141228 08:55:46-!- travis-ci [~travis-ci@ec2-54-81-49-92.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 08:56:53< shadowm> Actually, it's not optimized at all, it's just that the plain operation without yielding the processor is that fast.
20141228 08:57:31< shadowm> This is the code the compiler emits: https://gist.github.com/shikadilord/4c622ed822f0de54783d
20141228 08:58:38< shadowm> The relevant bit is line 14 through 19.
20141228 08:59:26< shadowm> It's literally "increment the value in RBP by 1 until it equals 65535".
20141228 08:59:31< shadowm> Fast.
20141228 08:59:42< shadowm> 0.001 seconds.
20141228 09:00:47< shadowm> (If I had optimizations enabled, though, it'd probably get supressed since the code has no side-effects and the result is not read.)
20141228 09:02:10< shadowm> -O3 replaces it with a NOP equivalent.
20141228 09:02:14< shadowm> Yep.
20141228 09:02:33< shadowm>         xorl    %eax, %eax
20141228 09:06:01< vultraz> huh
20141228 09:06:50< shadowm> Actually, that's not a NOP equivalent, it just zeroes EAX.
20141228 09:07:54< shadowm> Since nothing uses it AFAICT it's probably an ABI implementation detail.
20141228 09:09:12< shadowm> Ah, RAX holds the return value of a function, that's why.
20141228 09:09:37< vultraz> I have no idea what you're saying
20141228 09:09:39-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has joined #wesnoth-dev
20141228 09:09:40< shadowm> No idea why it's just zeroing the lower word though.
20141228 09:09:57< shadowm> So yeah, -O3 simply drops the loop entirely.
20141228 09:10:29< shadowm> And that is why people shouldn't use trivial loops to create delays in C or C++ programs.
20141228 09:11:00< vultraz> A. -O can discard bits of code?
20141228 09:11:07< shadowm> Yes.
20141228 09:11:08< vultraz> B. why would you want a code delay
20141228 09:11:13< shadowm> That's what I just showed.
20141228 09:11:47< shadowm> The correct way to insert delays is to issue a system call that causes one such, which is why nanosleep() etc. exist.
20141228 09:12:18< shadowm> On a modern multitasking OS those also yield the processor so it can go execute another process instead.
20141228 09:12:35< shadowm> *thread/process.
20141228 09:12:55< shadowm> And why? Why *not*?
20141228 09:13:15< shadowm> You might need to schedule a task and not use alarms for some reason.
20141228 09:15:13< shadowm> In C and C++ there are a lot more situations where compilers are allowed to remove code from the pipeline, undefined behavior being a notable example.
20141228 09:16:30< vultraz> huh
20141228 09:27:02< vultraz> iceiceice: is there any way to shut up the used before defined lua warnings?
20141228 09:29:50< vultraz> (besides fixing them :P )
20141228 09:30:06< vultraz> they're popping up in my campaign and annoying things
20141228 09:31:55< irker618> wesnoth: Charles Dang wesnoth:master 95c94bc19239 / data/lua/wml-tags.lua: Allowed a comma separated id list for [remove_event] and [event] remove=yes http://git.io/32N1CA
20141228 09:32:05 * vultraz goes to update wiki
20141228 09:32:25-!- tsunamistate [~tsunami@46.174.243.15] has joined #wesnoth-dev
20141228 09:34:09-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141228 09:36:05-!- tsunamistate [~tsunami@46.174.243.15] has quit [Client Quit]
20141228 09:45:26< shadowm> "[in long mode] 32-bit operands generate a 32-bit result, zero-extended to a 64-bit result in the destination general-purpose register."
20141228 09:45:32< shadowm> And that answers my question.
20141228 10:04:01-!- mjs-de [~mjs-de@p508C9DCE.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141228 10:15:17-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 10:15:17< travis-ci> wesnoth/wesnoth#5287 (master - 95c94bc : Charles Dang): The build has errored.
20141228 10:15:17< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45268205
20141228 10:15:17-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 10:19:53-!- sachith500 [~kvirc@101.189.46.81] has quit [Ping timeout: 240 seconds]
20141228 10:31:55-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20141228 10:32:11-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141228 10:55:31-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20141228 11:28:47-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 258 seconds]
20141228 12:13:18-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has joined #wesnoth-dev
20141228 12:32:13-!- irker618 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20141228 12:35:53-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has quit [Ping timeout: 240 seconds]
20141228 12:37:53-!- Sulfur [~Miranda@p5B0089BA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141228 12:39:51-!- sachith500 [~kvirc@101.189.46.81] has joined #wesnoth-dev
20141228 12:50:36-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20141228 12:51:51-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141228 12:52:05-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20141228 12:55:45-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20141228 12:56:11-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141228 12:57:52-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Read error: Connection reset by peer]
20141228 12:57:59-!- rayblade53 [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20141228 12:58:19-!- rayblade53 is now known as vultraz
20141228 13:18:56-!- sachith500 [~kvirc@101.189.46.81] has quit [Ping timeout: 250 seconds]
20141228 13:21:33-!- Sulfur [~Miranda@p5B0089BA.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141228 13:27:32< Necrosporus> Is it possible to disable autosaves but enable start of scenario saves?
20141228 13:27:40< Necrosporus> nosaves disables everything
20141228 13:28:13< Necrosporus> and scenario saves are impossible to create manuall
20141228 13:55:22< Ravana_> I can only think of reseting saves option for the start then
20141228 14:05:16< zookeeper> well then you'd need to pre-empt the previous scenario ending...
20141228 14:17:14-!- new_one [~new_one@50.15.126.83] has joined #wesnoth-dev
20141228 14:31:32-!- Nostromus [~chatzilla@g224147071.adsl.alicedsl.de] has joined #wesnoth-dev
20141228 14:35:22-!- happygrue [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141228 14:54:01< zookeeper> fabi, the playlist fix seems to work correctly
20141228 14:54:14< fabi> zookeeper: :-)
20141228 14:54:36< fabi> zookeeper: I thought so. The question is: Does it break anything else?
20141228 14:56:19< zookeeper> i don't think it breaks any existing functionality
20141228 14:56:36< zookeeper> of course it subtly changes the API, but i think the old behavior was clearly a bug, so...
20141228 14:57:44< shadowm> fabi: Hi.
20141228 14:57:50-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20141228 14:58:31< zookeeper> i wonder if we should allow play_once,immediate=yes,no to play a track once _after_ the current track has finished
20141228 14:58:45< fabi> shadowm: I don't want to talk to you.
20141228 14:58:48< zookeeper> in theory that might be useful somewhere, but... uh, i dunno
20141228 14:58:56< fabi> zookeeper: I have thougt about that as well.
20141228 14:58:59< shadowm> fabi: I'm sorry?
20141228 15:00:16< shadowm> If you don't want to do the backport then I'll just find a volunteer as I said in my first email, but I needed to know whether you could do it before deciding to go and do that.
20141228 15:00:39< vultraz> subtle lvl over 9000...
20141228 15:00:52-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 240 seconds]
20141228 15:02:09< shadowm> vultraz: Do you want to do it?
20141228 15:02:48< vultraz> backport the classic theme? I can try
20141228 15:02:52< shadowm> Inspect fabi's commits involving the classic theme in master and backport it to 1.12.
20141228 15:03:56< shadowm> fabi: I don't even know what's this about, and ISTR you had other things pending backport for 1.12.1 back when we were in beta/RC stage, so if you could help me out here...
20141228 15:06:58-!- gfgtdf [~chatzilla@f054130060.adsl.alicedsl.de] has joined #wesnoth-dev
20141228 15:19:28-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has quit [Ping timeout: 255 seconds]
20141228 15:38:40-!- gfgtdf [~chatzilla@f054130060.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 34.0.5/20141126041045]]
20141228 15:54:35-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141228 16:03:29-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141228 16:04:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 256 seconds]
20141228 16:13:19< shadowm> 13:12:29 <vultraz> should c8848813fd192d78ac843921afeb2cb317c68158 be backported too
20141228 16:13:22< shadowm> vultraz: Yes.
20141228 16:13:37< shadowm> I'm surprised that even exists.
20141228 16:14:30< vultraz> yay for conflicts...
20141228 16:16:28< vultraz> why is there a rect= conflict in the middle of this...
20141228 16:17:12< shadowm> New plan: create a pull request so I can review it at the end.
20141228 16:19:44-!- AI0867_ is now known as AI0867
20141228 16:23:21 * vultraz headbangs
20141228 16:24:14< shadowm> Or just give up and I take care of it as usual.
20141228 16:24:20< shadowm> It's up to you.
20141228 16:26:25-!- [Relic] [~Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141228 16:28:33-!- [Relic] [~Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Client Quit]
20141228 16:28:51-!- [Relic] [~Relic]@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141228 16:36:22< vultraz> shadowm: ok, pr pr-ed
20141228 16:55:58< shadowm> https://github.com/Vultraz/wesnoth/commit/b878c7265d4defada47fed39facc49f0c20f7921#diff-19e972050045f02e412506c5aa3bc554L41
20141228 16:56:01< shadowm> :\
20141228 16:56:55< vultraz> I don't know what happened there
20141228 16:57:17< shadowm> Well, now I see why fabi didn't want to handle this.
20141228 16:57:43< shadowm> 1.12 got stuck with some unfortunate file paths.
20141228 16:58:01< vultraz> hm?
20141228 16:58:12-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has joined #wesnoth-dev
20141228 16:58:13< shadowm> "classic" is not classic in 1.12.
20141228 16:58:25< shadowm> Ignore the PR, just look at unmodified 1.12.
20141228 16:58:40-!- kex [~kex@77.29.142.40] has joined #wesnoth-dev
20141228 16:59:04< shadowm> I guess it's considered "classic" when compared to the aborted unitbox theme, though.
20141228 16:59:57< shadowm> You really should've told me this before continuing. :p As it is, someone will have to get creative to put it all back together in a form that complies with maintenance branch policies.
20141228 17:00:47< vultraz> Derp
20141228 17:00:52< vultraz> I didn't really notice D:
20141228 17:01:59< shadowm> Sigh, I regret promising this backport in the forums.
20141228 17:02:16-!- Sulfur [~Miranda@p5B0089BA.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141228 17:05:57< shadowm> vultraz: In general, never assume a commit you didn't write yourself is plain cherry-pick-able between branches, even if it somehow doesn't yield conflicts.
20141228 17:06:05-!- irker782 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20141228 17:06:05< irker782> wesnoth: Charles Dang wesnoth:1.12 3cc1015d0cc1 / projectfiles/CodeBlocks/.gitignore: Ignore projectfiles/CodeBlocks/cb/ http://git.io/eTr6_A
20141228 17:06:33< vultraz> so what do I do now
20141228 17:06:51< shadowm> Yours too, really. I slipped a few misplaced changelog entries a while ago because I didn't pay attention.
20141228 17:07:03< shadowm> They were my own commits.
20141228 17:07:43< shadowm> vultraz: I'll work on recreating the commits with non-conflicting paths later.
20141228 17:08:03< shadowm> And without those extraneous Unicode-corrupting changelog entries.
20141228 17:08:31< shadowm> Do not delete the PR, though. I can use the reference.
20141228 17:10:19< vultraz> alright
20141228 17:13:24< vultraz> is the invalid unicode in master also?
20141228 17:13:49-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has joined #wesnoth-dev
20141228 17:15:39< shadowm> Someone deleted those entries, it seems.
20141228 17:15:57< shadowm> Uh. Or something.
20141228 17:16:06< shadowm> 1.11.19? What was the original commit again?
20141228 17:16:07-!- happygrue_ [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141228 17:17:28< shadowm> I'm not sure what's going on here, but the entries are missing in HEAD.
20141228 17:18:23-!- happygrue [~Laptop@wesnoth/developer/wintermute] has quit [Ping timeout: 258 seconds]
20141228 17:18:53< shadowm> vultraz, these entries are yours.
20141228 17:19:07< vultraz> eh?
20141228 17:19:12< shadowm> You are the one who corrupted the changelog entries, not the original committer.
20141228 17:19:53< shadowm> Here's the proof: there's literally no 1.11.19 block in the master players_changelog, since 1.12 diverged after 1.11.11.
20141228 17:20:26< vultraz> ok..
20141228 17:20:34< shadowm> Or 1.11.18 block, for that matter.
20141228 17:20:40< shadowm> Also note that the entry you actually added is under the 1.11.16 block.
20141228 17:21:07< shadowm> So yeah, fix your editor, and review your commits better next time. (Backporting someone else's commit counts as your own.)
20141228 17:21:36-!- gfgtdf [~chatzilla@f054053068.adsl.alicedsl.de] has joined #wesnoth-dev
20141228 17:21:59< vultraz> I was using the TGit editor
20141228 17:22:20< shadowm> Well, the TGit editor is BROKEN.
20141228 17:22:49< vultraz> ok
20141228 17:23:07< shadowm> The changelogs use UTF-8. Everything uses UTF-8, really. Don't let your editor use any other encoding.
20141228 17:33:39-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has quit [Ping timeout: 264 seconds]
20141228 17:51:29-!- SpoOkyMagician [~chatzilla@cpe-74-132-242-221.swo.res.rr.com] has quit [Quit: okay, i'm starting to get sleepy now... woke up too early again. bbl.]
20141228 18:07:36< gfgtdf> do we allow a negative village income and negative early finishing bonus ?
20141228 18:08:45< shadowm> If the code does then yes.
20141228 18:12:31< shadowm> Should we? I guess the former makes sense for certain kind of scenarios; the latter, not so much.
20141228 18:15:46< gfgtdf> shadowm: maybe in some type of survival scenario whewre you shoudl get more gold in teh next scneario the longer you last.
20141228 18:16:19< shadowm> Less you mean?
20141228 18:19:14-!- Necrosporus_ is now known as Necrosporus
20141228 18:19:40< Necrosporus> anyways, I don't see a reason to remove functionality
20141228 18:21:28< gfgtdf> shadowm: no
20141228 18:21:42< gfgtdf> Necrosporus: you tak about carryover gold ?
20141228 18:21:45< gfgtdf> talk*
20141228 18:24:29< Necrosporus> I talk in general. If it's possible to use negative numbers why to remove it even if it doesn't seem to be used often?
20141228 18:25:18< shadowm> gfgtdf: I don't understand. Of course normally you want more gold, which means you want a positive bonus.
20141228 18:30:46< zookeeper> i guess he means that on turn 19/20 you'd get an amount equal to 19 turns of early finish
20141228 18:32:10< shadowm> That sounds more like negative turns.
20141228 18:33:00< zookeeper> maybe. an easy to do in WML anyway
20141228 18:37:36< zookeeper> (literally 3-6 lines)
20141228 18:50:30-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20141228 18:51:32-!- happygrue_ [~Laptop@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20141228 19:06:13< irker782> wesnoth: Chris Beck wesnoth:master ac0385d666e2 / src/ (SConscript tests/test_recall_list.cpp): add a unit test for recall list manager object http://git.io/qdYTAg
20141228 19:06:15< irker782> wesnoth: Chris Beck wesnoth:master 8539cb77cd69 / src/dialogs.cpp: add some debugging output to recall dialog when dismissing units http://git.io/hcvmdg
20141228 19:06:17< irker782> wesnoth: Chris Beck wesnoth:master 924fb820588e / src/ (actions/attack.cpp actions/attack.hpp dialogs.cpp dialogs.hpp): partially fix 23076 (don't hold std::vector<unit> in recall dialog) http://git.io/7KgPZA
20141228 19:06:19< irker782> wesnoth: Chris Beck wesnoth:master 696bbfe2bffc / src/ (dialogs.cpp dialogs.hpp menu_events.cpp): fully fix bug #23076 http://git.io/Oh5xVg
20141228 19:06:21< irker782> wesnoth: Chris Beck wesnoth:master 4cd2f6b88bf9 / src/ (7 files in 3 dirs): Merge branch 'test_recall_list' http://git.io/MPsj5A
20141228 19:10:00< irker782> wesnoth: Chris Beck wesnoth:master faeab22934c1 / .travis.yml: travis: refactor script, try to fix utf8 check program http://git.io/g7Z1OA
20141228 19:15:42< irker782> wesnoth: Chris Beck wesnoth:master 6695365efb25 / .travis.yml utils/travis/test_wrapper.sh: travis: ignore error code 200 from test executable if repeats 10x http://git.io/VTVBVA
20141228 19:17:22< irker782> wesnoth: Chris Beck wesnoth:master 826c447595a4 / .travis.yml: travis: remove "time" utility, travis has a new feature for this http://git.io/CNwH3A
20141228 19:18:52< gfgtdf> iceiceice: so you fixed http://gna.org/bugs/?23076, do you know why it worked in 1.12 but not in mastzer ?
20141228 19:19:27< irker782> wesnoth: Chris Beck wesnoth:master 7a148154c389 / utils/travis/check_utf8.sh: travis: also check changelogs & RN for utf8 errors http://git.io/nCz2qQ
20141228 19:20:24-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141228 19:20:39< iceiceice> gfgtdf: the problem was introduced when I switched over to reference counting for units,
20141228 19:21:04< iceiceice> there was some wierd stuff where the recall / unit list dialogs are based on making copies of all the units and holding them in private std::vector<unit> 's...
20141228 19:21:12< iceiceice> i don't really know why they were doing that, that's sort of a bad idea
20141228 19:21:25< iceiceice> i thought it would be fine to leave them doing that though
20141228 19:21:42< iceiceice> somehow the gui elements were getting out of sync with eachother
20141228 19:22:03< iceiceice> i fixed it by just making one list of the units in the dialog, and passing smart pointer between the gui elements
20141228 19:22:35< gfgtdf> iceiceice: so do you know how it worked in 1.12 ?
20141228 19:22:53< iceiceice> only vaguely
20141228 19:23:09< iceiceice> i'm not sure exactly where the breakage occurred tbh
20141228 19:23:54< iceiceice> the design was pretty strange... its like,
20141228 19:24:04< iceiceice> when you bring up the "Recall" dialog in 1.12,
20141228 19:24:14< iceiceice> i think there are three different lists of units,
20141228 19:24:19< iceiceice> one for the actual list of units,
20141228 19:24:24< iceiceice> one for th unit "preview pane" at left
20141228 19:24:31< iceiceice> and one for the "delete button action"
20141228 19:24:46< iceiceice> some of them were vector<unit *> some of them were vector<unit>
20141228 19:25:22< iceiceice> now they are all basically shared_ptr<vector<unit_const_ptr> >, that seems to fix it and it makes more sense anyways
20141228 19:26:24< iceiceice> idk let me know if its still broken some how
20141228 19:26:34< iceiceice> i'm going to mark it ready for test, i wasn't able to break it myself
20141228 19:29:01< gfgtdf> iceiceice: i aksed becasue i also tried to fix it but had absolutely no idea why this code worked in 1.12.
20141228 19:29:07< gfgtdf> iceiceice: ok ill test it
20141228 19:29:11< iceiceice> ok thk you
20141228 19:36:11< gfgtdf> iceiceice: game_lua_kernel uses game_lua_kernel.cpp
20141228 19:36:22< gfgtdf> iceiceice: uses lua_toboolean*
20141228 19:37:10< gfgtdf> iceiceice: this causes warnings, use luaW_toboolean instead.
20141228 19:37:16< iceiceice> hmmm
20141228 19:38:48< iceiceice> they are almost the same...
20141228 19:38:52< iceiceice> i see
20141228 19:39:35< gfgtdf> iceiceice: they do teh same exept that lua_toboolean return a c-style bool and luaW_toboolean a c++ bool.
20141228 19:41:32-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 19:41:32< travis-ci> wesnoth/wesnoth#5287 (master - 95c94bc : Charles Dang): The build passed.
20141228 19:41:32< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45268205
20141228 19:41:32-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 19:48:57-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 240 seconds]
20141228 19:57:15< shadowm> > shared_ptr<vector<unit_const_ptr> >
20141228 19:57:20< shadowm> D:
20141228 19:57:23< irker782> wesnoth: Chris Beck wesnoth:master 50794d6164c0 / src/scripting/game_lua_kernel.cpp: lua: fix some warnings from implicit int->bool (lua_toboolean) http://git.io/3y9-pQ
20141228 20:01:08-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141228 20:02:06< gfgtdf> iceiceice: it seems liek the messageboxes liek from [objewct] dont do ways anaymore
20141228 20:02:48< iceiceice> what do you mean "don't do ways" ?
20141228 20:03:51< iceiceice> shadowm: whats D: ?
20141228 20:04:20< shadowm> Have you ever stepped on lego?
20141228 20:04:21< gfgtdf> iceiceice: no wait it's somethign different
20141228 20:04:37< shadowm> It's that same expression.
20141228 20:04:37< iceiceice> yes i have
20141228 20:04:48< iceiceice> yes i know what the emoticon means.
20141228 20:11:07-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 20:11:07< travis-ci> wesnoth/wesnoth#5290 (master - 4cd2f6b : Chris Beck): The build was broken.
20141228 20:11:07< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45295477
20141228 20:11:07-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 20:19:03-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20141228 20:19:41< irker782> wesnoth: Chris Beck wesnoth:master 61d93802e468 / src/tests/test_recall_list.cpp: try to fix clang c++11 compilation http://git.io/qwYBYw
20141228 20:22:03-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 265 seconds]
20141228 20:36:33-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 20:36:33< travis-ci> wesnoth/wesnoth#5291 (master - faeab22 : Chris Beck): The build was broken.
20141228 20:36:33< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45295744
20141228 20:36:33-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 20:40:48< gfgtdf> vultraz: did you test https://github.com/wesnoth/wesnoth/commit/ec985105ef41a01a7ab43bf7ed416fc3aa41d452 ?
20141228 20:42:21< gfgtdf> vultraz: actually it breaks clear_variable
20141228 20:44:01< shadowm> Do you know why?
20141228 20:45:42< gfgtdf> shadowm: yes i know why
20141228 20:46:29< shadowm> Well, since I'm the one who greenlit the commit in question, I want to know too.
20141228 20:48:00< gfgtdf> shadowm: gsub returns 2 values, the second is s number, if we pass those 2 to set_variable we actualy set teh variable to teh seond value rather than deletign it
20141228 20:48:38< shadowm> Ahhh, right. Feel free to revert it then.
20141228 20:50:53-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has joined #wesnoth-dev
20141228 21:03:03-!- kex [~kex@77.29.142.40] has quit [Remote host closed the connection]
20141228 21:03:18-!- kex [~kex@77.29.142.40] has joined #wesnoth-dev
20141228 21:03:24-!- kex [~kex@77.29.142.40] has quit [Remote host closed the connection]
20141228 21:06:55-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 21:06:55< travis-ci> wesnoth/wesnoth#5292 (master - 6695365 : Chris Beck): The build was broken.
20141228 21:06:55< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45296084
20141228 21:06:55-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 21:07:51-!- Kexoth [~kex@77.29.142.40] has joined #wesnoth-dev
20141228 21:13:26-!- Kexoth [~kex@77.29.142.40] has quit [Remote host closed the connection]
20141228 21:22:09< gfgtdf> iceiceice: you know whether i can pass a local string to luaL_argerror ?
20141228 21:23:58< gfgtdf> iceiceice: i mean like luaL_argerror(L, 1, std::string("somemessage") + my_string_returning_function()) ?
20141228 21:24:42< c74d> gfgtdf, vultraz: if `f(args)` returns multiple values, one can get solely the first value by using `(f(args))` instead.
20141228 21:25:54< shadowm> That doesn't seem very intuitive.
20141228 21:30:44-!- sachith500 [~kvirc@101.189.46.81] has joined #wesnoth-dev
20141228 21:35:37-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 21:35:37< travis-ci> wesnoth/wesnoth#5293 (master - 826c447 : Chris Beck): The build was broken.
20141228 21:35:37< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45296151
20141228 21:35:37-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 21:39:47< c74d> I might even say it seems rather hack-ish.
20141228 21:49:57-!- mjs-de [~mjs-de@p508C9DCE.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20141228 22:10:32-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 22:10:32< travis-ci> wesnoth/wesnoth#5294 (master - 7a14815 : Chris Beck): The build was broken.
20141228 22:10:32< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45296226
20141228 22:10:32-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 22:21:27< irker782> wesnoth: gfgtdf wesnoth:master 57227f4030f0 / src/carryover.cpp: fix gold carryover. http://git.io/q8bnPg
20141228 22:21:30< irker782> wesnoth: gfgtdf wesnoth:master 7f0e7c890d5d / src/scripting/game_lua_kernel.cpp: better error on invalid variablename in wesnoth.set_variable http://git.io/1n8lZA
20141228 22:40:17-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 22:40:17< travis-ci> wesnoth/wesnoth#5295 (master - 50794d6 : Chris Beck): The build was broken.
20141228 22:40:17< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45298268
20141228 22:40:17-!- travis-ci [~travis-ci@ec2-54-196-66-210.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 22:53:04-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141228 22:58:00-!- Nostromus [~chatzilla@g224147071.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 34.0/20141127110442]]
20141228 23:01:21< irker782> wesnoth: Chris Beck wesnoth:master 032d8c1a93c5 / utils/travis/test_wrapper.sh: fixup travis test_wrapper.sh http://git.io/IHJzGg
20141228 23:07:03-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141228 23:08:12-!- iceiceice [~chris@50.245.222.235] has joined #wesnoth-dev
20141228 23:08:12-!- iceiceice [~chris@50.245.222.235] has quit [Changing host]
20141228 23:08:12-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141228 23:08:18-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 23:08:18< travis-ci> wesnoth/wesnoth#5296 (master - 61d9380 : Chris Beck): The build is still failing.
20141228 23:08:18< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45299766
20141228 23:08:18-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 23:08:24< iceiceice> c74d: can you not just use "first_val = f(args)" ?
20141228 23:08:36< iceiceice> i was pretty sure that lua ignores extra return values if you don't put them somewhere
20141228 23:09:19< iceiceice> gfgtdf: i think usually lua makes private copies of strings
20141228 23:09:24< c74d> iceiceice: that should work too, but putting the value in a variable can be avoided too.
20141228 23:10:01< iceiceice> gfgtdf: if its not okay then probably other things are broken i think...
20141228 23:10:06< iceiceice> i guess can look ata reference
20141228 23:11:55< iceiceice> gfgtdf: it looks that luaL_argerror calls luaL_error, which pushes the string onto the stack, so i think its ok
20141228 23:17:45< gfgtdf> iceiceice: ok
20141228 23:18:18< gfgtdf> iceiceice: yes first_val = f(args)is what happens previously
20141228 23:18:40< gfgtdf> iceiceice: we were talking about this commit which break clear variable:  https://github.com/wesnoth/wesnoth/commit/ec985105ef41a01a7ab43bf7ed416fc3aa41d452 
20141228 23:19:03< gfgtdf> happened*
20141228 23:19:38< shadowm> fabi: Did you lose access to your forum account?
20141228 23:20:10< gfgtdf> iceiceice: is https://gna.org/bugs/?23115 already fixed ? Or is there the special trick needed to reproduce ?
20141228 23:20:14-!- cib0 [~cib@tmo-103-81.customers.d1-online.com] has quit [Ping timeout: 245 seconds]
20141228 23:21:24< iceiceice> afaik its still broken
20141228 23:21:42< iceiceice> i don't understand why, i would like to fix it, it was broken only a few days ago
20141228 23:22:13< iceiceice> to reproduce it, i download "pick your advancement" add-on and activate it in any local mp game
20141228 23:22:35< gfgtdf> iceiceice: so that only happens on mp?
20141228 23:22:58< iceiceice> i didn't test any mainline campaigns, actually i didn't know there were mainline campaigns with menu items
20141228 23:23:29< iceiceice> how did you try to test
20141228 23:23:50< iceiceice> its only broken for custom menu items
20141228 23:24:05< gfgtdf> iceiceice: ok i coudl prerpduce it
20141228 23:24:23< gfgtdf> iceiceice: i didnt happen for me in sp though
20141228 23:24:25< iceiceice> its wierd because i did test this when i committed,
20141228 23:24:32< iceiceice> but i don't remember how anymore
20141228 23:24:33< gfgtdf> iceiceice: maybe it somewithng form with playmp_controller ?
20141228 23:24:42< iceiceice> i bisected it to a specific commit,
20141228 23:24:55< iceiceice> in that commit i try to eliminate resource links from wml menu items
20141228 23:25:04< iceiceice> by giving them a pointer to a manager
20141228 23:25:15< iceiceice> i think most likely i misunderstood something about their lifecycle
20141228 23:26:45< gfgtdf> iceiceice: ok it's not related to mp
20141228 23:27:35< gfgtdf> iceiceice: seems liek it doesnt happen when you load a savagem form older scenario with wml menu items
20141228 23:27:41< gfgtdf> from*
20141228 23:30:42< iceiceice> i think the problem is that "update_command(" ends up being called by the ctor, before "init_handler" is called...
20141228 23:30:50< iceiceice> i didn't realize that that happened before it hink
20141228 23:31:11-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20141228 23:31:29< iceiceice> its a bit of a knot, since the way it works right now, all of the "game_data" is supposed to be constructed before the lua kernel, and the events manager i think...
20141228 23:32:18< iceiceice> it would be better if these guys did not need to be attached to a wml event queue when they are created, since they get stored in carryover data also and passed to different managers i guess
20141228 23:32:48-!- ToBeCloud [uid51591@gateway/web/irccloud.com/x-jpbetpsnqquwtraw] has joined #wesnoth-dev
20141228 23:33:42< iceiceice> hmmm actually,
20141228 23:33:47< iceiceice> i think maybe its just that i broke the copy ctor
20141228 23:33:58-!- travis-ci [~travis-ci@ec2-54-81-49-92.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 23:33:58< travis-ci> wesnoth/wesnoth#5297 (master - 7f0e7c8 : gfgtdf): The build is still failing.
20141228 23:33:58< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45307729
20141228 23:33:58-!- travis-ci [~travis-ci@ec2-54-81-49-92.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 23:34:41< gfgtdf> iceiceice: the travis fails
20141228 23:34:48< iceiceice> yeah i saw
20141228 23:34:55< iceiceice> i think the commit its building now should fix
20141228 23:35:34< gfgtdf> iceiceice: ok
20141228 23:38:29< gfgtdf> iceiceice: is there a reaso why wml_menu_items_ in part of game_data ?
20141228 23:38:56< iceiceice> idk i think its always been that way
20141228 23:39:10< iceiceice> i mean gamedata roughly corresponds to carryover, right?
20141228 23:42:00< gfgtdf> iceiceice: hm well wml_menus, varaibles, and next_scenario is what carryover and game_data share.
20141228 23:42:25< irker782> wesnoth: Chris Beck wesnoth:master 188e5c70b189 / src/game_events/menu_item.cpp: fixup menu_item copy ctor http://git.io/qcn8Ag
20141228 23:46:51-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 256 seconds]
20141228 23:47:07< irker782> wesnoth: Chris Beck wesnoth:master 978ae9ebc39e / .travis.yml: travis: fixup 02 vs O0 build http://git.io/GbG21A
20141228 23:47:14< gfgtdf> iceiceice: you know what https://github.com/wesnoth/wesnoth/blob/master/src/game_data.cpp#L230 is? do we allow asisgning game_data pointer tpo game_data object ? :s
20141228 23:47:42< gfgtdf> iceiceice: also i dont think game_data and carryover interace directly with eachother
20141228 23:48:16< iceiceice> i never looked at that line,
20141228 23:48:30< iceiceice> its not good because its not exception safe :(
20141228 23:48:34< iceiceice> it should use copy and swap instead
20141228 23:48:41< gfgtdf> iceiceice: yes also that
20141228 23:49:29< gfgtdf> iceiceice: or maybe teh default implementation would just rok ?
20141228 23:49:32< gfgtdf> work*
20141228 23:49:34< iceiceice> maybe
20141228 23:50:15< iceiceice> idk if mt_rng has a copy ctor...
20141228 23:50:26< iceiceice> or if the wmi container does either
20141228 23:50:34< iceiceice> if they did i guess it would work
20141228 23:51:15< gfgtdf> iceiceice: i think for mt_rng a teh default copy constructor works
20141228 23:51:15< iceiceice> and if variable_set also has one
20141228 23:54:23< gfgtdf> iceiceice: variable_set is just an iterface i think
20141228 23:54:30< iceiceice> y, i see now
20141228 23:54:33< gfgtdf> iceiceice: you know why wmi_container uses boost::shared_ptr<wml_menu_item>
20141228 23:56:06< iceiceice>  not sure
20141228 23:57:19< iceiceice> idk if wml_menu_items are really copyable... they sort of correspond to hotkey registration / event registration
20141228 23:58:05< iceiceice> maybe they are doing too many things at once
20141228 23:58:21-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has joined #wesnoth-dev
20141228 23:58:21< travis-ci> wesnoth/wesnoth#5298 (master - 032d8c1 : Chris Beck): The build was fixed.
20141228 23:58:21< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/45309998
20141228 23:58:21-!- travis-ci [~travis-ci@ec2-54-144-216-102.compute-1.amazonaws.com] has left #wesnoth-dev []
20141228 23:58:25< iceiceice> huzzah
--- Log closed Mon Dec 29 00:00:17 2014