--- Log opened Tue Apr 07 00:00:26 2015
--- Day changed Tue Apr 07 2015
20150407 00:00:26< shadowm> I think it's okay to disable it only there if it's not too much trouble since no other file triggers that warning atm.
20150407 00:01:06< iceiceice> i dont think memcheck can give spurious "uninitialized jump"
20150407 00:01:13< iceiceice> at least idk what would cause that
20150407 00:01:50< shadowm> I'm not saying it's spurious, just out of our control.
20150407 00:02:21< shadowm> I used to get warnings about strtod() calls from config.cpp too.
20150407 00:02:28-!- new_one [~new_one@nat-168-7-230-56.rice.edu] has joined #wesnoth-dev
20150407 00:02:41< iceiceice> i would feel much better if memcheck did not complain about the unit tests
20150407 00:02:53< iceiceice> how did you get memcheck to give you line numbers?
20150407 00:02:57< shadowm> I wasn't running unit tests though.
20150407 00:02:58< iceiceice> do you need debugging symbols for that?
20150407 00:03:06< iceiceice> i get it when running unit tests though
20150407 00:03:07< shadowm> Yeah, I assume you do.
20150407 00:03:28< shadowm> The command line I used was just `valgrind wesnoth -dt` with an -O3 build.
20150407 00:03:51< iceiceice> i get a similar error reported with `valgrind wesnoth -u feeding`
20150407 00:04:38< shadowm> The thing is that in that line you linked I don't see any variables that could be propagated to library code that are size 8, just that computed value.
20150407 00:05:36< shadowm> Which obviously can't be uninitialized because it's not a variable.
20150407 00:06:47< shadowm> The other candidate is random_seed_ which is size 4 and initialized by the constructor.
20150407 00:07:07< iceiceice> yeah but maybe its some voodoo about stringstreams that i am oblivious
20150407 00:07:18< iceiceice> idk maybe i'm not using std::hex or something correctly
20150407 00:07:27< iceiceice> i tried splitting that line into 4 lines and i will see which one it complains about
20150407 00:07:55-!- cib0 [~cib@p5DC743F0.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds]
20150407 00:08:03< iceiceice> it might be earlier int he call stack also i guess
20150407 00:09:31< shadowm> Why does it go from libstdc++ to Boost.locale, actually?
20150407 00:11:00< iceiceice> dont know, thats wierd
20150407 00:11:00< shadowm> And still in namespace std. I thought it wasn't legal for libraries to add to or otherwise alter std.
20150407 00:12:20< shadowm> Maybe it's just some template instantiation nuance involving the linker that escapes me.
20150407 00:17:00< gfgtdf> any opinion about https://github.com/wesnoth/wesnoth/pull/396 (enable tree_view gui2 widget for lua)?
20150407 00:19:11-!- travis-ci [~travis-ci@ec2-54-162-185-52.compute-1.amazonaws.com] has joined #wesnoth-dev
20150407 00:19:12< travis-ci> gfgtdf/wesnoth-old#451 (lua_tree_view - 56fdcfc : gfgtdf): The build has errored.
20150407 00:19:12< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/57414972
20150407 00:19:12-!- travis-ci [~travis-ci@ec2-54-162-185-52.compute-1.amazonaws.com] has left #wesnoth-dev []
20150407 00:19:25< shadowm> Is that widget stable?
20150407 00:19:44< gfgtdf> shadowm: what you mena by stable ?
20150407 00:19:58< shadowm> As in whether it crashes all the time like GUI2 listboxes.
20150407 00:20:02-!- ToBeCloud [uid51591@gateway/web/irccloud.com/x-rzzeymrgttmmsfyj] has joined #wesnoth-dev
20150407 00:21:03< gfgtdf> shadowm: i notices a crash when the treeview is too small for which i added an assert here: https://github.com/wesnoth/wesnoth/pull/396/files#diff-255ebc5ec0800339c02358939074318bR482
20150407 00:21:07< iceiceice> i think GUI2 is not really stable in general: http://forums.wesnoth.org/viewtopic.php?f=15&t=41028&start=15#p582453
20150407 00:21:08< gfgtdf> notices*
20150407 00:21:27< shadowm> But then it's still a crash of a different kind. :)
20150407 00:21:41< iceiceice> that grid.cpp line 529 error has occurred in many different contexts...
20150407 00:24:06< shadowm> I don't know, all the issues in the Lua GUI2 API stem from bugs in core GUI2 so exposing another widget probably won't hurt any more.
20150407 00:24:25< shadowm> Plus I just remembered that the Campaigns menu is a tree view (yes).
20150407 00:25:02< gfgtdf> shadowm: whcih campaogn menu is that?
20150407 00:25:19< shadowm> The only Campaigns menu that there is, accessed right from the title screen.
20150407 00:25:54< gfgtdf> shadowm: thought there coudl be another campaigns election screen from an experimental mp lobby/mp create
20150407 00:26:16< gfgtdf> shadowm: i actualy cannot se a tree structore there
20150407 00:26:22< shadowm> The fact that it's a tree view is very cleverly hidden from view but it is a tree view.
20150407 00:27:59< shadowm> Well, seems I misread. There are actually two campaign menus in the same dialog.
20150407 00:28:31< shadowm> The tree is made invisible when not using --new-widgets. With that it just plain crashes.
20150407 00:29:44< shadowm> So the only other application of GUI2 tree views I know of is in the GUI2 MP lobby, but I think the listboxes and scroll views were more problematic there.
20150407 00:34:05< gfgtdf> shadowm: hm after experimenting more i get an assertion failre when i fodl some items, have to investigate more
20150407 00:37:28< gfgtdf> shadowm: hm ok it seems leik when i have a subnode folded and then try to fodl the parent node in a treeview with scrollbar i get an assertion failure :s
20150407 00:39:31-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Read error: Connection reset by peer]
20150407 00:44:05-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 00:45:58< gfgtdf> ok i think i foudn teh erro but i dotn knwo whethe ri can fix
20150407 00:47:50-!- irker009 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150407 00:47:50< irker009> wesnoth: Chris Beck wesnoth:master 29c99fe8705a / src/scripting/game_lua_kernel.cpp: ignore "maybe uninitialized" warning in game lua kernel. http://git.io/vez0q
20150407 00:48:00< iceiceice> shadowm: ^
20150407 00:48:17< iceiceice> i am still figuring out about the memcheck
20150407 00:50:41-!- travis-ci [~travis-ci@ec2-54-162-185-52.compute-1.amazonaws.com] has joined #wesnoth-dev
20150407 00:50:42< travis-ci> gfgtdf/wesnoth-old#452 (lua_tree_view - 5abe369 : gfgtdf): The build passed.
20150407 00:50:42< travis-ci> Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/57415059
20150407 00:50:42-!- travis-ci [~travis-ci@ec2-54-162-185-52.compute-1.amazonaws.com] has left #wesnoth-dev []
20150407 00:51:58-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150407 01:02:51< iceiceice> hmm ok
20150407 01:03:43-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150407 01:04:06< iceiceice> gfgtdf: valrgind says that new_seed may still be uninitialized after this line:
20150407 01:04:07< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/mt_rng.cpp#L92
20150407 01:04:34< iceiceice> i think i did not realize that
20150407 01:05:43< gfgtdf> iceiceice: you mena when str is not a valid hex number ?
20150407 01:05:51< iceiceice> yes
20150407 01:12:58< iceiceice> valgrind found another problem also in the play controller initialization...
20150407 01:13:08< iceiceice> i'm going to test first fix though
20150407 01:14:15< iceiceice> i wonder what strings it is failing to parse
20150407 01:15:51< iceiceice> debug random: Failed to seed a random number generator using seed string '', it could not be parsed to hex. Seeding with 42.
20150407 01:16:01< iceiceice> (that's what it says now after my patch
20150407 01:16:03< iceiceice> )
20150407 01:29:09< irker009> wesnoth: Chris Beck wesnoth:master ab49e328b938 / src/mt_rng.cpp: fix uninitialized rng seed (reported by valgrind) http://git.io/vezK8
20150407 01:29:11< irker009> wesnoth: Chris Beck wesnoth:master c10ddf275e16 / src/ (game_display.cpp mt_rng.cpp): fix uninitialized failure http://git.io/vezK4
20150407 01:34:25< iceiceice> ok it didnt find any other errors before but i will run again to be sure...
20150407 01:34:31< iceiceice> im running leakcheck while we are at it 
20150407 01:37:05-!- kex [~kex@89.205.97.191] has quit [Ping timeout: 256 seconds]
20150407 01:50:32-!- gfgtdf [~chatzilla@x50abb1d6.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 37.0.1/20150402191859]]
20150407 01:59:45-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 256 seconds]
20150407 02:03:08-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 02:16:29< irker009> wesnoth: Chris Beck wesnoth:master 5421c73be459 / src/formula_function.cpp: suppress false positive memory leak reported by valgrind http://git.io/vezHC
20150407 02:18:13-!- groggy [~chatzilla@75-143-179-137.dhcp.gnvl.sc.charter.com] has joined #wesnoth-dev
20150407 02:19:44-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Read error: Connection reset by peer]
20150407 02:24:39-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 02:39:40-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 265 seconds]
20150407 02:41:07-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 02:44:47-!- SpoOkyMagician [~chatzilla@cpe-74-132-57-45.kya.res.rr.com] has joined #wesnoth-dev
20150407 02:48:19-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20150407 02:51:09-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 244 seconds]
20150407 02:55:11-!- new_one [~new_one@nat-168-7-230-56.rice.edu] has quit [Remote host closed the connection]
20150407 03:10:19-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150407 03:12:38-!- ToBeCloud [uid51591@gateway/web/irccloud.com/x-rzzeymrgttmmsfyj] has quit [Quit: Connection closed for inactivity]
20150407 03:12:50-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150407 03:15:24-!- new_one [~new_one@128.42.76.177] has joined #wesnoth-dev
20150407 03:19:20< shadowm> ancestral: So how did yesterday's packaging exercise go?
20150407 03:30:29-!- SpoOkyMagician [~chatzilla@cpe-74-132-57-45.kya.res.rr.com] has quit [Quit: later]
20150407 03:30:55< shadowm> Did somebody remove link underlines in the help system?
20150407 03:33:55< ancestral> shadowm: Failing when building translations
20150407 03:35:14< shadowm> How do you feel about 1.12.2?
20150407 03:35:34< ancestral> Aha, have to mkdir first
20150407 03:35:41< irker009> wesnoth: Ignacio R. Morelle wesnoth:master 74abdda87490 / data/core/help.cfg: Fix inconsistent help markup quotes http://git.io/vegJy
20150407 03:36:30< ancestral> shadowm: I’ve got instructions from mattsc with creating the dmg, I haven’t actually done that, but it’s not scary
20150407 03:36:41< ancestral> 1.12.2 builds fine and runs fine
20150407 03:36:53< ancestral> 1.13 mattsc and I are still investigating why it’s failing for me
20150407 03:37:10< ancestral> So I’ll package on Wed night/Thu morn
20150407 03:37:35< ancestral> Meanwhile we’ll figure out why 1.13 isn’t working for me
20150407 03:37:58< shadowm> Okay.
20150407 03:38:06< irker009> wesnoth: Ignacio R. Morelle wesnoth:master 0fc006f8f331 / data/core/help.cfg: Fix help topics in the Commands section not using Title Case http://git.io/vegUJ
20150407 03:38:48-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150407 03:39:45< irker009> wesnoth: Ignacio R. Morelle wesnoth:master 34d46b231377 / changelog players_changelog: Changelog spacing fixes http://git.io/vegUC
20150407 03:45:27-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 252 seconds]
20150407 03:47:14-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 04:07:28-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20150407 04:14:02-!- new_one [~new_one@128.42.76.177] has quit [Quit: WeeChat 1.1.1]
20150407 04:15:21< vultraz> groggy: what was that?
20150407 04:17:14< groggy> chatzilla right-click offered two options, direct chat and open private chat. wasn't sure of the difference, so i tried direct chat first
20150407 04:21:36< shadowm> vultraz: Perhaps you might be able to shed some light on this mystery.
20150407 04:21:46< shadowm> 00:30:56 <shadowm> Did somebody remove link underlines in the help system?
20150407 04:21:47< vultraz> mystery?
20150407 04:21:51< vultraz> ah yes
20150407 04:21:52< vultraz> me
20150407 04:21:55< shadowm> Why?
20150407 04:22:04< vultraz> I thought it looked better without it
20150407 04:22:23< shadowm> No way to bring them back on hover?
20150407 04:22:51< vultraz> I don't think so
20150407 04:22:58< shadowm> :\
20150407 04:23:21< vultraz> Does ttext have a hover state?
20150407 04:23:27< shadowm> ttext?
20150407 04:23:37< shadowm> Since when does the help system use ttext?
20150407 04:23:50< vultraz> It doesn't. I'm wondering if that's something we could use if it did
20150407 04:23:53< shadowm> Also, ttext is a renderer, not a widget.
20150407 04:24:01< shadowm> It doesn't have "states".
20150407 04:24:13< vultraz> I've been trying to make the help browser use ttext but have failed
20150407 04:26:09-!- ancestral [~ancestral@174-20-187-184.mpls.qwest.net] has joined #wesnoth-dev
20150407 04:26:46< vultraz> If we could make it gui2 completely, I think iceiceice added link awareness
20150407 04:27:05< vultraz> But IIRC what it uses now (what sdl_tff?) doesn't have states
20150407 04:27:26< shadowm> SDL_ttf is a renderer, not a widget. Same thing.
20150407 04:28:16< shadowm> Except SDL_ttf is garbage and ttext (which uses Cairo/Pango instead of directly doing things with FreeType) is not.
20150407 04:33:39< vultraz> Well, perhaps you could help me convert it to ttext
20150407 04:34:03< shadowm> I have way too many things on my plate right now.
20150407 04:35:03< vultraz> (I assume ttext doesn't require manual handling of bold and italic text?)
20150407 04:36:13< vultraz> shadowm: anyway, is there anything else I changed about the help system you'd like me to address
20150407 04:37:18< vultraz> ?
20150407 04:38:11-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150407 04:38:33< iceiceice> shadowm, vultraz : i think i'm going to try to rejigger the default theme.
20150407 04:38:42< vultraz> say what?
20150407 04:38:44< iceiceice> to fix these issues https://gna.org/bugs/?23002
20150407 04:38:50< iceiceice> https://gna.org/bugs/?22995
20150407 04:38:56< iceiceice> because... they were reported long ago
20150407 04:39:10< iceiceice> and i think GunChleoc for instance will be frustrated if its still not fixed in 1.12.2
20150407 04:39:11< shadowm> vultraz: I haven't had a proper look at it yet, too busy with ther things.
20150407 04:39:27< iceiceice> and fabi seems on wesbreak
20150407 04:39:36< vultraz> iceiceice: alright, if you want
20150407 04:39:36< shadowm> ttext uses Pango, Pango has the <b> and <i> tags, vultraz.
20150407 04:39:43< vultraz> good
20150407 04:39:51< shadowm> You are supposed to know this. :p
20150407 04:39:54< vultraz> I do
20150407 04:40:00< vultraz> Jjustc hecking
20150407 04:40:03< vultraz> just checking*
20150407 04:40:24< shadowm> The problem is that you'll probably need to get creative with help markup parsing before passing it to ttext.
20150407 04:41:24< shadowm> Apparently the Pango devs planned back in 2009 or so to add an API for custom tags without requiring users to reimplement the whole parsing pipeline but that plan appears to have fallen by the wayside.
20150407 04:41:31 * vultraz groans
20150407 04:43:40< vultraz> iceiceice: also, I get this now: C:\Users\Charles\Documents\wesnoth-git\src\scripting\game_lua_kernel.cpp|133|warning: unknown option after '#pragma GCC diagnostic' kind [-Wpragmas]|
20150407 04:43:53< iceiceice> damn
20150407 04:43:57< iceiceice> ok
20150407 04:44:15< vultraz> (keep in mind I'm on gcc 4.5.2)
20150407 04:44:19< vultraz> er, tdm gcc
20150407 04:44:24< shadowm> Worth checking which version introduced the warning then.
20150407 04:44:26< iceiceice> well idk what makes more sense
20150407 04:44:35< iceiceice> so one thing we could do is add like
20150407 04:44:43< iceiceice> gcc diagnostic ignored pragma
20150407 04:44:59< iceiceice> oh 
20150407 04:45:10< iceiceice> i guess the issue is that it doesnt know how to ignore diagnostics in that version
20150407 04:45:49< vultraz> I'm off again, bb in an hour or so
20150407 04:48:46< shadowm> iceiceice: Will you be around within the next hour or so?
20150407 04:48:58< iceiceice> ehhh
20150407 04:49:03< iceiceice> kinda iffy i might go to sleep
20150407 04:49:26< shadowm> Okay, no hurry, I'll post the link here anyway so you can check it later when you have some time.
20150407 04:49:28< iceiceice> getting to be that time
20150407 04:50:07< iceiceice> ok
20150407 04:50:24< iceiceice> i was going to just restrict the ignore to gcc 4.6.0 +
20150407 04:50:31< iceiceice> and see if that does it
20150407 04:51:25< iceiceice> its kind of wierd,
20150407 04:51:32< iceiceice> this page seems to be for 4.3.4 based on the url: https://gcc.gnu.org/onlinedocs/gcc-4.3.4/gcc/Diagnostic-Pragmas.html
20150407 04:56:11< irker009> wesnoth: Chris Beck wesnoth:master 03ab839f300f / src/scripting/game_lua_kernel.cpp: fixup a gcc pragma to exclude gcc 4.5.x and older http://git.io/vegnE
20150407 05:00:15-!- groggy [~chatzilla@75-143-179-137.dhcp.gnvl.sc.charter.com] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 36.0.4/20150320202338]]
20150407 05:14:16< shadowm> iceiceice: Revamped INSTALL: http://pastebin.com/hWpbcdnC
20150407 05:15:11< iceiceice> i think we maybe should remove the stuff about sdl mixer 1.2.0 also
20150407 05:15:15< iceiceice> in master
20150407 05:15:19< iceiceice> or we could anyways
20150407 05:15:48< iceiceice> its similar to the no boost file system option i think
20150407 05:17:19 * iceiceice notices section 5
20150407 05:20:21< iceiceice> well
20150407 05:20:49< iceiceice> i'm not sure if it makes that much sense to put the scons and cmake instructions mixed together like that
20150407 05:20:56< iceiceice> since most readers are only using one of them and only care about those lines
20150407 05:21:07< iceiceice> but it seems fine
20150407 05:21:20< ancestral> iceiceice: I get an error with scons translate if the translate directory doesn’t exist
20150407 05:21:47< iceiceice> whats the error?
20150407 05:22:01< shadowm> iceiceice: It makes sense because otherwise we'd need to have the same descriptions in two different places.
20150407 05:22:20< shadowm> That's the way it was one hour ago.
20150407 05:22:20< ancestral> scons: *** Do not know how to make File target `translations' (/Users/xxxxxx/wesnoth 1.12/translations).  Stop.
20150407 05:22:20< iceiceice> yeah ...
20150407 05:22:31< iceiceice> i guess that's optimizing it for document size rather than ease of reading
20150407 05:22:53< shadowm> Maintainability was more of a concern in this case than length.
20150407 05:23:11< iceiceice> yeah ok, thats reasonable
20150407 05:23:34< shadowm> I'm already curious how long this will stay nicely formatted before somebody decides to throw a 200 chars long line into it or break the pretty alignment in some parts. ;)
20150407 05:24:44< iceiceice> ancestral: i didnt reproduce that, if i delete my translations folder it just makes a new one
20150407 05:24:53< iceiceice> is there like a permissions thing or something
20150407 05:24:59< iceiceice> (oh i'm on 1.12 branch atm also)
20150407 05:25:52< shadowm> Oh man.
20150407 05:26:03< shadowm> This mostly works as Markdown. Mostly.
20150407 05:26:21< shadowm> Only section 5 breaks completely.
20150407 05:26:31< iceiceice> ancestral: do you have wierd permissions or something in that folder?
20150407 05:26:36< ancestral> iceiceice: Should I `sudo scons translate`?
20150407 05:26:45< iceiceice> hehe
20150407 05:26:48< ancestral> iceiceice: I shouldn’t
20150407 05:26:52< iceiceice> i think it might be better to do like
20150407 05:26:59< iceiceice> chmod 
20150407 05:27:12< ancestral> I’ll delete the folder and chmod 755 everything if you want
20150407 05:27:35< ancestral> Just to see
20150407 05:27:48< ancestral> Oooh, I could even log in to my computer as root
20150407 05:27:51< iceiceice> there should be some way like
20150407 05:27:53< ancestral> Oh the damage that could be done!
20150407 05:27:56< iceiceice> you could do like ls -a or something?
20150407 05:28:03< iceiceice> theres some way to get it to tell you which permission flags are on
20150407 05:28:16< iceiceice> i guess thats probably not it anyways
20150407 05:28:21< iceiceice> i guess it woudl tell you there was a permissions error probably
20150407 05:28:24 * shadowm raises eyebrow.
20150407 05:28:46< iceiceice> idk im just going to google your problem :lol:
20150407 05:29:05< shadowm> Hope nobody notices the rm -rf line I added to SConscript the other day.
20150407 05:29:15< shadowm> *rm -rf /
20150407 05:29:34< iceiceice> no idea
20150407 05:29:44< ancestral> drwxr-xr-x
20150407 05:29:53< iceiceice> whoops forgot to delete the "april fools" rm -rf
20150407 05:29:57< ancestral> Heh
20150407 05:31:46< ancestral> What are “NLS tools?”
20150407 05:32:09< iceiceice> nls is the gettext stuff
20150407 05:32:22< iceiceice> its what makes all the stuff in translations
20150407 05:32:30< ancestral> NLS tools are not present...
20150407 05:32:34< ancestral> NLS catalogue installation is disabled.
20150407 05:32:37< ancestral> ^ that bad?
20150407 05:32:41< iceiceice> its not really bad
20150407 05:32:46< shadowm> It's bad, for a packager.
20150407 05:32:51< iceiceice> yeah thats true
20150407 05:33:01< ancestral> Okay, so
20150407 05:33:06< shadowm> Unless you are compiling them with an external script or something, anyway.
20150407 05:33:07< ancestral> How do I get NLS tools?
20150407 05:33:13< iceiceice> i had to fix this with homebrew once 
20150407 05:33:19< iceiceice> so apparently, 
20150407 05:33:26< iceiceice> it is a pain in the ass beacuse apple ships some version of gettext,
20150407 05:33:31< iceiceice> but it is not well updated and mostly broken
20150407 05:33:33< ancestral> Oh right
20150407 05:33:34< iceiceice> or something
20150407 05:33:37< ancestral> I remember this
20150407 05:33:40< iceiceice> yeah
20150407 05:33:46< iceiceice> so in the homebrew script...
20150407 05:33:47< iceiceice> it like
20150407 05:33:54< iceiceice> installs gettext using homebrew
20150407 05:33:59< iceiceice> in some sort of secret homebrew dir
20150407 05:34:14< iceiceice> and it points scons to use it using gettext_dir
20150407 05:34:17< ancestral> Speakeasy directory
20150407 05:34:24< iceiceice> yes :)
20150407 05:34:38< iceiceice> come and get your translations from my bathtub, good sir
20150407 05:37:24< shadowm> scons doesn't honor CXX from the environment. :(
20150407 05:37:48< shadowm> Ah yes, I think that was another of the packagers' complaints about scons.
20150407 05:38:12< shadowm> It does with OS_ENV=true now, but this was introduced last cycle or so.
20150407 05:38:19< iceiceice> yeah i was just going to mention that
20150407 05:38:36< shadowm> Okay, no, actually it doesn't seem to honor that either.
20150407 05:38:49< shadowm> Otherwise it should be failing early with `CXX=/bin/false scons OS_ENV=true --config=force`
20150407 05:39:05< shadowm> Instead I'm getting a backtrace I don't understand.
20150407 05:39:12< shadowm> loonycyborg: ^ 
20150407 05:39:13< iceiceice> i think it respects CXXTOOL
20150407 05:39:26< iceiceice> and it seems to overwrite CXX with CXXTOOL
20150407 05:40:02< shadowm> cmake takes CXX, like autotools do.
20150407 05:41:01< shadowm> Tried deleting the scons options cache, it still goes ahead with its own guess instead of using CXX.
20150407 05:41:32< shadowm> At the very least, it's not using CXX for the configuration step.
20150407 05:42:03< shadowm> Which, you know, why would you ever want to perform configuration with a compiler other than the one you are going to build with?
20150407 05:42:29< shadowm> And now you know why packagers hate scons.
20150407 05:43:15< iceiceice> its easy to love things that you dont have to maintain
20150407 05:43:30< shadowm> Like Wesnoth.
20150407 05:43:34< iceiceice> hehe
20150407 05:46:37-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 255 seconds]
20150407 05:52:30< shadowm> Incidentally, vanilla SDL 1.2.15 causes problems on Linux with some (or all?) window managers.
20150407 05:53:08< shadowm> Upstream never released a new version, so Debian and others ship their own patched versions instead.
20150407 05:53:38< shadowm> I'm not sure if this should be mentioned in INSTALL.
20150407 05:56:10< shadowm> Bleh I somehow deleted some dependencies
20150407 06:05:15-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20150407 06:09:03< iceiceice> hmm does anyone know where the images associated to themes are?
20150407 06:09:19< shadowm> So how do I disable Boost.locale with cmake again?
20150407 06:10:38-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150407 06:11:53-!- ancestral [~ancestral@174-20-187-184.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150407 06:12:02< shadowm> Ivanovic, iceiceice, vultraz (proof-reading?), whoever else cares: This is the final draft of the revamped INSTALL for master, provided I don't find any new mistakes later: http://pastebin.com/ct3JBRih
20150407 06:12:38< shadowm> This is for master. Whether I'll port this to 1.12 with the list of dependencies tweaked accordingly is still up in the air right now.
20150407 06:13:14< shadowm> Also, numbers, you stay away from that link, thank you very much.
20150407 06:14:21< shadowm> Just spotted a mistake on line 133. s/build/the actual build/
20150407 06:15:46< shadowm> iceiceice: data/core/images/themes, for themable UI themes.
20150407 06:15:54< shadowm> (Think [theme].)
20150407 06:16:48< iceiceice> i wanted to check if some of the theme item positions are now effectively hard coded into the images or not
20150407 06:18:00< shadowm> IIRC that's only the case with the minimap and the right sidebar tool buttons in the editor but I'm not sure.
20150407 06:18:29< shadowm> The major areas' positions are presumably hardcoded as well.
20150407 06:19:11< shadowm> Unit info area, minimap area.
20150407 06:20:28< shadowm> Eh.
20150407 06:20:59< shadowm> I use xvJf .tar.xz in an example when that should be xvjf .tar.bz2 still unless I change my mind before Sunday.
20150407 06:22:35< vultraz> Looks good to me
20150407 06:22:46< vultraz> You don't mention anything about openmp, is that intentional
20150407 06:25:24< shadowm> It's largely untested and known to break things on Apple OS X.
20150407 06:25:36< vultraz> Gotcha 
20150407 06:29:28-!- [Relic] [~Relic]@2602:306:33a3:6d30:8ce9:ee44:3997:e303] has quit [Quit: Leaving]
20150407 06:34:53< vultraz> I still support scrapping it altogether, but oh well
20150407 06:35:02< vultraz> shikadibot: seen mordante
20150407 06:35:02< shikadibot> vultraz: The person with the nick mordante last spoke 8d 18h ago.  3d 17h ago they left with the message: Remote host closed the connection
20150407 06:42:28< iceiceice> ok i give up on theme wml
20150407 06:43:05< shadowm> It's not unlike euthanizing a pet. You put it off for as long as possible because you somehow still hope it may be able to recover on its own.
20150407 06:43:07< iceiceice> it really needs to.... report errors...
20150407 06:43:51< shadowm> But instead you are causing your pet and everyone more pain.
20150407 06:44:41< shadowm> Then there's Ana, which no-one wants to euthanize because it's a GSoC product.
20150407 06:44:59< iceiceice> what is ana
20150407 06:45:41< shadowm> It's that thing ancestral dug up from its grave yesterday, an attempt at writing a replacement for our dated and clunky and generally horrible SDL_net-based network layer, using Boost.Asio.
20150407 06:46:10< shadowm> The student didn't finish it and most functionality was horribly broken, at least for wesnothd.
20150407 06:46:32< iceiceice> oh that reminds
20150407 06:46:46< vultraz> speaking of didn't finish
20150407 06:46:54< vultraz> what happened to the husks of zaroth's project
20150407 06:47:05< vultraz> or did riftwalker clean that up
20150407 06:47:20< shadowm> Some people claim it was a misconceived project since Boost.Asio isn't meant to be layered behind more abstractions.
20150407 06:47:35< iceiceice> vultraz: can you build now?
20150407 06:47:55< shadowm> I'm barely able to set up a socket pair with Asio so I don't have an opinion about it.
20150407 06:48:55< vultraz> iceiceice: w/o warnings? yes
20150407 06:49:20< shadowm> And that's the history of Ana and that's how we wound up with a smaller Asio-based campaignd client existing alongside the SDL_net-based wesnothd client and server and campaignd itself.
20150407 06:49:58< shadowm> The person doing the Asio campaignd client intended to tackle the wesnothd server next but that never came to fruition.
20150407 06:50:20< shadowm> (The Asio campaignd client shares no code with Ana, it just happens to use Asio too.)
20150407 06:50:37< iceiceice> shadowm: ok, i decided to try to run the mp unit test
20150407 06:50:44< iceiceice> with all parties going thorugh valgrind
20150407 06:51:08< iceiceice> hmm actually
20150407 06:51:15< iceiceice> i guess i should make only the server valgrind
20150407 06:51:26-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150407 06:52:13< shadowm> I hope you ordered pizza and beer or something then.
20150407 06:55:36< iceiceice> hmmm
20150407 06:55:53< iceiceice> i get some legitimate server error messages about "illegal command"
20150407 06:56:02< shadowm> In a way, I guess we should be glad that ISPs are ridiculous and IPv6 still isn't mainstream anywhere in the world, beacuse otherwise we'd be in great trouble seeing as how SDL_net doesn't do IPv6 even though it's been around for about as long.
20150407 06:58:31< iceiceice> man that didnt find much useful 
20150407 07:11:46< vultraz> iceiceice: so what are you doing with the theme?
20150407 07:13:31< iceiceice> vultraz: the translators have complained alot about the new theme
20150407 07:14:07< iceiceice> the level and some other stuff is getting truncated i guess
20150407 07:14:14< iceiceice> i pasted some bug reports...
20150407 07:14:19< iceiceice> theres also several forum threads
20150407 07:14:34< iceiceice> i think that basically fabi tried to cram too much stuff in the sidebar
20150407 07:14:56< iceiceice> what i was going to do was
20150407 07:15:10< iceiceice> take the status boxes that contain terrain description 
20150407 07:15:13< iceiceice> currently right under minimap
20150407 07:15:23< iceiceice> and put them to the right of the income, countdown timer
20150407 07:15:30< iceiceice> sort of where the used to be
20150407 07:15:37< iceiceice> because we cant afford allt hat dead space
20150407 07:15:55< iceiceice> then use the vertical space to make room for all the stuff that is getting cramped
20150407 07:16:11< iceiceice> and probably get rid of the flag at the right
20150407 07:16:17< iceiceice> since there is already one at top left
20150407 07:16:39< vultraz> Sounds good
20150407 07:16:49< iceiceice> its hard right now though
20150407 07:16:50< shadowm> You know the theme revamp wasn't *only* fabi's baby, right?
20150407 07:17:18< iceiceice> well who owns the new one should fix the reported problems
20150407 07:17:32< iceiceice> right now i assume no one owns it
20150407 07:17:38< vultraz> No one does
20150407 07:17:48< shadowm> I meant the design itself, not the code.
20150407 07:18:08< iceiceice> im sure it wasnt only his
20150407 07:18:12< shadowm> fabi mostly exchanged ideas with LordBob and vultraz, I believe.
20150407 07:18:20< vultraz> Yes
20150407 07:18:32< vultraz> The whole UI revamp was a collab between the three of us
20150407 07:18:54< iceiceice> how do you figure out what is going on
20150407 07:18:59< iceiceice> when the refs are bad or whatever
20150407 07:19:10< iceiceice> i foudn that once i edited enough things everything ends up in the center of the screen
20150407 07:19:18< shadowm> So I think getting most of those three people to reflect on the feedback obtained since 1.11.10 would be a good idea.
20150407 07:19:19< iceiceice> is there a useful log channel?
20150407 07:21:01< vultraz> I haven't been keeping up with the feedback
20150407 07:27:28< shadowm> Rebuilding Wesnoth 1.11.0 through 1.11.14. Wish me luck.
20150407 07:28:08< vultraz> ...why?
20150407 07:31:23< vultraz> noted
20150407 07:32:05-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 250 seconds]
20150407 07:34:03-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 07:34:18< shadowm> Turns out I don't need to do that and that's even worse.
20150407 07:34:25< shadowm> :(
20150407 08:09:20-!- mjs-de [~mjs-de@f049173123.adsl.alicedsl.de] has joined #wesnoth-dev
20150407 08:09:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 08:10:48-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150407 08:11:10-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150407 08:13:03-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Ping timeout: 250 seconds]
20150407 08:31:33-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20150407 09:21:20< vultraz> loonycyborg, mattsc, ancestral: can you please compile future releases with png support enabled
20150407 09:23:08-!- cib0 [~cib@132.231.178.135] has joined #wesnoth-dev
20150407 09:44:40-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 255 seconds]
20150407 09:46:44-!- horrowind [~Icedove@2a02:810a:8b40:85c:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150407 09:49:08-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 244 seconds]
20150407 10:01:20-!- cib0 [~cib@132.231.178.135] has quit [Read error: Connection reset by peer]
20150407 10:05:21-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150407 10:15:20-!- cib0 [~cib@132.231.178.16] has joined #wesnoth-dev
20150407 10:35:19-!- Crendgrim_ is now known as Crendgrim
20150407 10:36:56-!- mjs-de [~mjs-de@f049173123.adsl.alicedsl.de] has quit [Remote host closed the connection]
20150407 10:47:49-!- cib0 [~cib@132.231.178.16] has quit [Ping timeout: 248 seconds]
20150407 11:07:10-!- irker009 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150407 12:02:48< loonycyborg> vultraz: I still have no idea what you're talking about. wesnoth releases are capable of using png images.
20150407 12:04:25-!- new_one [~new_one@128.42.90.179] has joined #wesnoth-dev
20150407 12:11:08< loonycyborg> shadowm: it's only affected by 'scons cxxtool=foo' 
20150407 12:12:52< loonycyborg> There's no reason to believe that CXX will affect any particular command
20150407 12:12:58< loonycyborg> all this stuff is non-standard
20150407 12:15:46-!- cib0 [~cib@132.231.178.91] has joined #wesnoth-dev
20150407 12:19:30-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150407 12:19:30-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150407 12:19:30-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150407 12:43:23-!- irker355 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150407 12:43:23< irker355> wesnoth: loonycyborg wesnoth:master a52cc353dfde / scons/ (pango.py pkgconfig.py): scons: Fix blowing up with OS_ENV=true http://git.io/vea2I
20150407 13:36:56< irker355> wesnoth: loonycyborg wesnoth:master ff80a64ebba2 / scons/sdl.py: Revert "scons: fix config checks that require running test programs for out-of-t http://git.io/veaFq
20150407 13:36:58< irker355> wesnoth: loonycyborg wesnoth:master 2fca959437e4 / scons/sdl.py: scons: fix out-of-tree builds without using python's operator % http://git.io/veaFm
20150407 13:48:02-!- gfgtdf [~chatzilla@x50ab6de4.dyn.telefonica.de] has joined #wesnoth-dev
20150407 13:48:13< gfgtdf> vultraz: i answered you o github pr discussion
20150407 13:56:55-!- mjs-de [~mjs-de@f049109107.adsl.alicedsl.de] has joined #wesnoth-dev
20150407 14:05:47-!- prkc [~prkc@4E5CD609.dsl.pool.telekom.hu] has joined #wesnoth-dev
20150407 14:15:38< vultraz> loonycyborg: I am referring to:  " * libpng:  PNG screenshots, otherwise only BMP is supported. "
20150407 14:19:04< vultraz> gfgtdf: alright
20150407 14:24:36-!- kex [~kex@92.53.39.44] has joined #wesnoth-dev
20150407 14:30:42-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 272 seconds]
20150407 14:30:56-!- kex [~kex@92.53.39.44] has quit [Ping timeout: 240 seconds]
20150407 14:31:19-!- cib0 [~cib@132.231.178.91] has quit [Ping timeout: 255 seconds]
20150407 14:31:51-!- kex [~kex@92.53.39.44] has joined #wesnoth-dev
20150407 14:45:38-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150407 15:01:12-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150407 15:03:46-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 15:08:04< gfgtdf> mordante: any opinion on https://github.com/wesnoth/wesnoth/pull/396 ?
20150407 15:10:49< vultraz> gfgtdf: do the node contain listboxes?
20150407 15:10:58< iceiceice> loonycyborg: i tried to get the unicode strings to work in python, but it seems there are still problems somewhere in scons: https://github.com/wesnoth/wesnoth/pull/397
20150407 15:11:19< iceiceice> (the most recent push didn't fix it for me either)
20150407 15:11:32< gfgtdf> vultraz: no it doesnt. I didnt test puttign listboxed into node definition
20150407 15:11:39< gfgtdf> listboxes*
20150407 15:12:10< iceiceice> loonycyborg: i think maybe it could be easier to just copy the datafiles out of tree?
20150407 15:12:54< loonycyborg> hmm yes, but not sure what's the best way to do that
20150407 15:13:36< gfgtdf> vultraz: you talk about my example dialog or about treeviews in general ?
20150407 15:14:06< vultraz> gfgtdf: the example. I was just wondering if that's how the nodes were formatted
20150407 15:14:13< vultraz> or if you added a listbox
20150407 15:14:33< iceiceice> so whats the idea with the out of tree builds?
20150407 15:14:50< iceiceice> is it like you call scons in the root of repo, but with some "target_dir=..." argument?
20150407 15:15:02< gfgtdf> vultraz: in my example there are 2 types of nodes in teh treeview, one for the elements and one for teh categories.
20150407 15:15:10< iceiceice> or its like cmake, and the scons root is somewhere far away?
20150407 15:15:57-!- [Relic] [~Relic]@2602:306:33a3:6d30:ad18:75b2:374e:f8f3] has joined #wesnoth-dev
20150407 15:16:13< gfgtdf> vultraz: my sample dialog is pretty much copied fr,om teh treeview here: https://github.com/wesnoth/wesnoth/blob/master/data/gui/default/window/campaign_dialog.cfg#L210
20150407 15:17:37< vultraz> I never understood why there's a tree view there
20150407 15:19:15< vultraz> gfgtdf: btw, what about calling the function just 'set_tree_value'
20150407 15:20:15< gfgtdf> vultraz: the treview in teh campaogn dialog is hidden.
20150407 15:20:29< vultraz> yes, but why is it there anyway
20150407 15:20:29< gfgtdf> vultraz: idk why it was deactivated, mayabe it didnt work innteh expected way
20150407 15:20:46< gfgtdf> vultraz: you can activate by uing new widgets or somethign i think
20150407 15:21:29< gfgtdf> vultraz: why you think thats a better name ?
20150407 15:21:45< vultraz> gfgtdf: shorter, also it's consistent with set_dialog_value 
20150407 15:22:19-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150407 15:22:25< gfgtdf> vultraz: but it does different , set_dialog_value sets a value, this function adds a something.
20150407 15:23:10< vultraz> add_tree_value?
20150407 15:24:13< vultraz> also, new widgets is broken atm
20150407 15:24:18< gfgtdf> vultraz: thats possible but i added the 'dialog_' becasue the other lua gui2 function also contains dialog_ you think we shoudl remove it ?
20150407 15:24:19< vultraz> mordante should know, I guess
20150407 15:25:19< vultraz> hm
20150407 15:26:01< vultraz> that is a point
20150407 15:26:21< vultraz> idk :P
20150407 15:27:03< vultraz> just leave it for now i guess
20150407 15:27:21< gfgtdf> vultraz: i think it woudl be best if we had teh gui2 function in their own table dialog. (dialog.show, dialog.set_value ..)  but that woudl break too much wml now.
20150407 15:27:45< vultraz> indeed
20150407 15:28:47< loonycyborg> iceiceice: I do it like this: mkdir /tmp/testbuild; cd /tmp/testbuild; scons -Y ~/vcs-checkouts/wesnoth
20150407 15:29:27-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Remote host closed the connection]
20150407 15:30:44-!- Kwandulin [~Miranda@p5B0084F6.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150407 15:35:22< gfgtdf> vultraz: this code: http://pastebin.com/tsyHJ9h5 prints 1,3,nil, you know whhat i have to change to amke it 1,2,3 ?
20150407 15:36:15< vultraz> uh.... {1,2,3}?
20150407 15:37:00< gfgtdf> vultraz: hm maybe that example was to easy, wait
20150407 15:41:18-!- Elvish_Hunter [~chatzilla@151.16.242.204] has joined #wesnoth-dev
20150407 15:41:30< gfgtdf> vultraz: this example: http://pastebin.com/5MhQtF6z
20150407 15:41:33< Elvish_Hunter> Hi all
20150407 15:41:46< gfgtdf> vultraz: how do i fix f2 to do what i want ?
20150407 15:41:52< gfgtdf> Elvish_Hunter: mabye you know ?
20150407 15:42:00< gfgtdf> Elvish_Hunter: hi
20150407 15:42:36< irker355> wesnoth: Elvish_Hunter wesnoth:1.12 506a9b36ca4a / data/tools/GUI.pyw: WML tools GUI: backported improvements from master http://git.io/veVbv
20150407 15:45:44< Elvish_Hunter> vultraz: we're using Lua 5.2, and here the unpack function doesn't exist any more. It's called table.unpack
20150407 15:46:00< Elvish_Hunter> It was simply unpack until 5.1
20150407 15:46:25< gfgtdf> Elvish_Hunter: irt still works in my 1.13-dev build
20150407 15:46:29< gfgtdf> unpack i mean
20150407 15:47:45< Elvish_Hunter> Huh, you're right. It still works, despite the Lua manual stating "Function unpack was moved into the table library and therefore must be called as table.unpack"
20150407 15:49:08-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150407 15:49:17< iceiceice> i think there is some sort of compatibility mode we have in our lua?
20150407 15:49:44< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/lua/luaconf.h#L19
20150407 15:49:51< iceiceice> that might have to do with it?
20150407 15:49:52< Elvish_Hunter> ice^3: no, I tested on Lua 5.2 on the Terminal
20150407 15:50:33< vultraz> "// todo: remove after 1.11.2"
20150407 15:52:18< iceiceice> Elvish_Hunter: from grepping, there is a line in the source
20150407 15:52:28< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/lua/luaconf.h#L263
20150407 15:54:24< Elvish_Hunter> Then the same define must be enabled in the version shipped in the Ubuntu repositories; this should explain my result
20150407 15:55:13-!- travis-ci [~travis-ci@ec2-54-198-53-153.compute-1.amazonaws.com] has joined #wesnoth-dev
20150407 15:55:14< travis-ci> cbeck88/wesnoth#163 (try_fix_scons - b1db1db : Chris Beck): The build passed.
20150407 15:55:14< travis-ci> Build details : http://travis-ci.org/cbeck88/wesnoth/builds/57495973
20150407 15:55:14-!- travis-ci [~travis-ci@ec2-54-198-53-153.compute-1.amazonaws.com] has left #wesnoth-dev []
20150407 15:55:54< iceiceice> hmm well
20150407 15:55:57< Elvish_Hunter> By the way, does anyone have a guess about why two Travis builds out of six failed here? https://github.com/wesnoth/wesnoth/pull/394
20150407 15:56:01< iceiceice> idk what is gained by turning off the compatiblity
20150407 15:56:36< iceiceice> Checking for Boost program_options library version >= 1.35.0... no
20150407 15:56:39< iceiceice> that is wierd
20150407 15:56:51< iceiceice> i never saw that happen before
20150407 15:57:46< iceiceice> the second one looks legitimate though
20150407 15:57:46< iceiceice> http://pastebin.com/WvWheppw
20150407 16:11:15-!- nickanc [~Nickanc@wikimedia/Nickanc] has joined #wesnoth-dev
20150407 16:16:02< Elvish_Hunter> vultraz: I don't know exactly how Lua works internally, but it seems like table.unpack returns an iterator, which counts as a single object
20150407 16:16:28< Elvish_Hunter> To solve the problem, I'd move the unpacking command inside the f function
20150407 16:16:48< Elvish_Hunter> Of course, you'll need to check the data type first
20150407 16:21:22< gfgtdf> Elvish_Hunter: i cannot change the function f, im my usecase f is wesnoth.set_dialog_value
20150407 16:21:40< gfgtdf> anyway have to go
20150407 16:21:42-!- gfgtdf [~chatzilla@x50ab6de4.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.91.1 [Firefox 37.0.1/20150402191859]]
20150407 16:28:16-!- Elvish_Hunter_ [~chatzilla@151.16.177.226] has joined #wesnoth-dev
20150407 16:30:00-!- Elvish_Hunter [~chatzilla@151.16.242.204] has quit [Ping timeout: 256 seconds]
20150407 16:30:14-!- Elvish_Hunter_ is now known as Elvish_Hunter
20150407 16:35:23-!- nickanc [~Nickanc@wikimedia/Nickanc] has quit [Quit: Sto andando via]
20150407 16:38:10< iceiceice> gfgtdf: do you know why we do all this shuffling of the scenario configs?
20150407 16:38:10< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/game_config_manager.cpp#L269
20150407 16:38:22< iceiceice> esp. line 260 also?
20150407 16:38:28< iceiceice> it looks like it would be very slow
20150407 16:39:02< iceiceice> oh its to rename all the configs?
20150407 16:39:37< iceiceice> i think there still should be a way to do that without copying them maybe
20150407 16:40:37< iceiceice> s/rename all the configs/change scenario to multiplayer/
20150407 16:47:03-!- nickanc [~Nickanc@wikimedia/Nickanc] has joined #wesnoth-dev
20150407 16:48:18-!- horrowind [~Icedove@2a02:810a:8b40:85c:21b:fcff:fee3:c3ff] has quit [Read error: No route to host]
20150407 16:49:00-!- horrowin2 [~Icedove@2a02:810a:8b40:85c:21b:fcff:fee3:c3ff] has joined #wesnoth-dev
20150407 16:54:32-!- horrowin2 [~Icedove@2a02:810a:8b40:85c:21b:fcff:fee3:c3ff] has quit [Quit: horrowin2]
20150407 17:11:53-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has quit [Ping timeout: 250 seconds]
20150407 17:22:35-!- Elvish_Hunter [~chatzilla@151.16.177.226] has quit [Quit: Ciao!]
20150407 17:50:04-!- Ravana_ [~SZ_Bot@unaffiliated/ravana/x-2327071] has quit [Ping timeout: 250 seconds]
20150407 17:50:38-!- SZ_Bot [SZ_Bot@27-83-235-80.dyn.estpak.ee] has joined #wesnoth-dev
20150407 17:50:47-!- SZ_Bot is now known as Ravana_
20150407 17:50:49-!- Ravana_ [SZ_Bot@27-83-235-80.dyn.estpak.ee] has quit [Changing host]
20150407 17:50:49-!- Ravana_ [SZ_Bot@unaffiliated/ravana/x-2327071] has joined #wesnoth-dev
20150407 17:58:48< irker355> wesnoth: Chris Beck wesnoth:master ef27b9b42c9e / scons/sdl.py: scons: for sdl checks, fall back to old mechanism if new fails http://git.io/vereN
20150407 17:58:58< iceiceice> loonycyborg: ^
20150407 18:04:57< loonycyborg> hmm ok
20150407 18:05:08-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 244 seconds]
20150407 18:13:10< iceiceice> would like to know what the right way to do it is i guess... but this probably will work for everyone who actually will use it
20150407 18:14:07-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150407 18:15:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20150407 18:16:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 18:18:49-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150407 18:23:34-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20150407 18:25:31-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150407 18:29:12-!- Kwandulin [~Miranda@p5B0084F6.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150407 18:31:01-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20150407 18:41:03-!- travis-ci [~travis-ci@ec2-107-20-22-205.compute-1.amazonaws.com] has joined #wesnoth-dev
20150407 18:41:04< travis-ci> wesnoth/wesnoth#6208 (master - ef27b9b : Chris Beck): The build was broken.
20150407 18:41:04< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/57520051
20150407 18:41:04-!- travis-ci [~travis-ci@ec2-107-20-22-205.compute-1.amazonaws.com] has left #wesnoth-dev []
20150407 18:43:37-!- kex [~kex@92.53.39.44] has quit [Remote host closed the connection]
20150407 18:47:42-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20150407 18:56:24-!- travis-ci [~travis-ci@ec2-54-198-53-153.compute-1.amazonaws.com] has joined #wesnoth-dev
20150407 18:56:25< travis-ci> cbeck88/wesnoth#164 (require_scenario - be28215 : Chris Beck): The build passed.
20150407 18:56:25< travis-ci> Build details : http://travis-ci.org/cbeck88/wesnoth/builds/57522298
20150407 18:56:25-!- travis-ci [~travis-ci@ec2-54-198-53-153.compute-1.amazonaws.com] has left #wesnoth-dev []
20150407 19:06:06< shadowm> gfgtdf: Until we can do as you suggested (wesnoth.dialog.*), I think it's a good idea to keep using the *_dialog_* naming pattern for the Lua GUI2 API.
20150407 19:07:27< shadowm> I wanted to do that for 1.13.0 some time ago (was going to name it wesnoth.gui2 though) -- alas, I've been busy with other things and/or unmotivated for most of this development period. :\
20150407 19:14:31-!- ancestral [~ancestral@63.92.240.233] has quit [Quit: i go nstuf kthxbai]
20150407 19:17:58-!- new_one [~new_one@128.42.90.179] has quit [Remote host closed the connection]
20150407 19:27:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150407 19:33:45-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 19:36:12-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150407 19:39:42< mattsc> Hi ancestral: FYI, I just edited the PM from Sunday in which I sent you the link to my executable.
20150407 19:40:45-!- kex [~kex@89.205.29.105] has joined #wesnoth-dev
20150407 19:40:54< mattsc> I put together a quick OS X package containing my entire build (but minus translations or any of the bells and whistles) and sent you the link to that too.
20150407 19:41:34< mattsc> My fear is that this won’t work for you either (since it uses the same libraries that apparently don’t work for you when you build yourself).
20150407 19:42:54< mattsc> If that’s the case, then there’s no point in me temporarily doing the 1.13.0 package on the weekend if we don’t get the problems figured out, since it would crash for at least some of the users.
20150407 20:25:39-!- cib0 [~cib@p5DD22009.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150407 20:38:37< shadowm> AI0867, ancestral, Elvish_Hunter, gfgtdf, iceiceice, Ivanovic, lipkab, loonycyborg, mattsc, Rhonda, Soliton, shadowm, vultraz, zookeeper, and anyone else who cares: I plan to tag 1.12.2 24:30 hours from now (21:00 UTC). If any of you have any commits pending for the 1.12 branch, you'd better push them *NOW* rather than wait until the last minute.
20150407 20:39:01-!- shadowm changed the topic of #wesnoth-dev to: String-freeze active for 1.12 branch | Stable version 1.12.2 scheduled for April 8th 21:00 UTC | Development version 1.13.0 scheduled for April 12th | Wesnoth Developers Channel | Logs: http://irclogs.wesnoth.org
20150407 20:43:44-!- knotwork [~markm@unaffiliated/knotwork] has quit [Read error: Connection reset by peer]
20150407 20:45:10< mattsc> shadowm: thanks, I’m all set
20150407 20:46:06-!- nickanc [~Nickanc@wikimedia/Nickanc] has quit [Quit: Sto andando via]
20150407 20:46:33-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Not here much atm. Will check logs for my nick and forum PMs.]
20150407 20:55:04-!- new_one [~new_one@128.42.79.81] has joined #wesnoth-dev
20150407 20:56:25-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150407 20:57:30-!- kex [~kex@89.205.29.105] has quit [Remote host closed the connection]
20150407 20:58:50-!- irker355 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150407 20:59:47-!- irker403 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150407 20:59:47< irker403> wesnoth: Nils Kneuper wesnoth:master c3cfa955437c / po/ (22 files in 22 dirs): updated Chinese (Simplified) translation http://git.io/veoWh
20150407 20:59:47< irker403> wesnoth: Nils Kneuper wesnoth:1.12 396f19227b56 / po/ (23 files in 23 dirs): updated Chinese (Simplified) translation http://git.io/veole
20150407 21:03:13< zookeeper> i don't think i really have anything to do for 1.12.2.
20150407 21:05:53-!- kex [~kex@89.205.29.105] has joined #wesnoth-dev
20150407 21:23:20-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20150407 21:24:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 21:24:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20150407 21:41:57-!- gfgtdf [~chatzilla@x55b1ad7e.dyn.telefonica.de] has joined #wesnoth-dev
20150407 21:42:52< gfgtdf> iceiceice: we convert all [scenario] to [multiplayer] i think this is part of teh implementation fo mp campaigns
20150407 21:44:18< gfgtdf> iceiceice: also we set 'campaign_id' to notice which campaign are paret of the current campaign, (for the entry point selection)
20150407 21:45:15< gfgtdf> iceiceice: the current implementation is indeed  alitttle slow especiayl it does a lot of cyopying where moves or swaps should be used.
20150407 21:49:46-!- Elvish_Hunter [~chatzilla@151.16.177.226] has joined #wesnoth-dev
20150407 21:50:01< Elvish_Hunter> Hi all
20150407 21:50:10< gfgtdf> Elvish_Hunter: hi
20150407 21:51:00-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 21:52:24< irker403> wesnoth: Elvish_Hunter wesnoth:master 0182985c7224 / data/core/units/undead/ (Corpse_Soulless.cfg Corpse_Walking.cfg): Removed # wmllint: ignore from Walking Corpse and Soulless units http://git.io/veo1V
20150407 21:52:26< irker403> wesnoth: Elvish_Hunter wesnoth:master 20d7d8b53edc / data/tools/wmllint: wmllint: handle variation_name translatable attribute. Fixes bug #23217 http://git.io/veo1w
20150407 21:52:28< irker403> wesnoth: Elvish_Hunter wesnoth:master 7a98677a05d2 / data/core/units/monsters/Wolf_Great.cfg: Fixed non translatable variation_name reported by wmllint http://git.io/veo1r
20150407 21:52:51< Elvish_Hunter> shadowm: ^^
20150407 21:53:40< shadowm> 1.12?
20150407 21:53:54< Elvish_Hunter> I guess that this can't be backported to 1.12.2 without breaking the string freeze...
20150407 21:53:56-!- Haudegen [~quassel@85.124.51.57] has quit [Ping timeout: 240 seconds]
20150407 21:54:11< shadowm> The Great Wolf doesn't need a variation_name.
20150407 21:54:40< shadowm> No technical/internal variations need one, only variations that are supposed to be displayed in the help browser.
20150407 21:55:13< shadowm> I.e. any variations that don't have show_help=no.
20150407 21:55:16< Elvish_Hunter> And in this case, the red variation isn't supposed to appear, right?
20150407 21:55:40< shadowm> *don't have hide_help=no (default).
20150407 21:55:48< shadowm> No, it isn't.
20150407 21:56:24< Elvish_Hunter> Well then, I'm on it
20150407 21:56:29-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150407 21:56:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20150407 21:56:35< shadowm> Don't forget to assign the bug to yourself and mark it as Fixed btw.
20150407 21:56:42-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20150407 21:56:52< Elvish_Hunter> Of course ;-)
20150407 21:57:22< iceiceice> shadowm: i have no blockers right now or things i plan to commit
20150407 21:57:53< iceiceice> i'm not going to try to do any mp server fixups right now, we'll just see how the carryover thing works i think
20150407 21:58:08< iceiceice> i think maybe we should put something in the announcement though like,
20150407 21:58:22< shadowm> Blargh I deleted my scons options cache.
20150407 21:58:24< iceiceice> "bug #23002 will be promoted to a blocker for 1.12.3"
20150407 21:58:50< shadowm> Good thing I have the contents in the console above.
20150407 22:01:04< shadowm> `scons --config=force -j8 cxxtool=clang++-3.4 lockfile=yes ccache=yes server_uid=1000 server_gid=1000 forum_user_handler=yes prefsdir=.wesnoth-1.13 extra_flags_config='-pipe -march=native' extra_flags_release=-O3 default_targets=campaignd,test,wesnoth,wesnothd build=release`
20150407 22:01:22< shadowm> This happens to be why I normally use scons for development, because it remembers all options.
20150407 22:03:00< shadowm> With cmake everything goes away if I decide to switch the compiler.
20150407 22:04:18< iceiceice> what i prefer to do is make little scripts
20150407 22:04:25< iceiceice> build_foo.sh
20150407 22:04:28< iceiceice> in the repo
20150407 22:04:32< iceiceice> that call scons with some set of options
20150407 22:05:03< iceiceice> usually i just use one of the scripts, but sometimes i call it, interrupt, and then pass extra args
20150407 22:05:16< iceiceice> and then i dont get confused about what options i'm using
20150407 22:05:28< iceiceice> i added some build*.sh git ignore rule a long time ago i think
20150407 22:05:35< shadowm> Sucks I found that big bug before deciding to tackle this: http://forums.wesnoth.org/viewtopic.php?p=580885#p580885
20150407 22:06:00< shadowm> Turns out I can't fix the issue described in the first paragraph without breaking the string freeze, which I'm obviously not going to do.
20150407 22:06:26< shadowm> Oh well.
20150407 22:06:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150407 22:08:22< irker403> wesnoth: Elvish_Hunter wesnoth:1.12 20adc76d6472 / data/core/units/undead/ (Corpse_Soulless.cfg Corpse_Walking.cfg): Removed # wmllint: ignore from Walking Corpse and Soulless units http://git.io/veoHd
20150407 22:08:24< irker403> wesnoth: Elvish_Hunter wesnoth:1.12 0964d448b004 / data/tools/wmllint: wmllint: handle variation_name translatable attribute. Fixes bug #23217 http://git.io/veoHF
20150407 22:08:26< irker403> wesnoth: Elvish_Hunter wesnoth:1.12 dc9e44c20d69 / data/core/units/monsters/Wolf_Great.cfg: Replaced a variation_name reported by wmllint with variation_id http://git.io/veoHb
20150407 22:11:59-!- mjs-de [~mjs-de@f049109107.adsl.alicedsl.de] has quit [Ping timeout: 250 seconds]
20150407 22:12:24-!- mjs-de [~mjs-de@f049109107.adsl.alicedsl.de] has joined #wesnoth-dev
20150407 22:15:07< Elvish_Hunter> OK, I'm done with this fix for 1.12. I'll double check as soon as my computer finishes recompiling
20150407 22:15:07-!- Haudegen [~quassel@85.124.51.57] has joined #wesnoth-dev
20150407 22:15:31< shadowm> Don't forget fixing up 7a98677a05d2 on master.
20150407 22:16:13< irker403> wesnoth: Ignacio R. Morelle wesnoth:1.12 a1106c392f6d / players_changelog: Sync players_changelog http://git.io/veo5Q
20150407 22:16:16< irker403> wesnoth: Ignacio R. Morelle wesnoth:1.12 a8b8dd80233d / src/theme.cpp: theme ui: Process translations of the "Hotkey(s): " string at runtime http://git.io/veo57
20150407 22:16:19< irker403> wesnoth: Ignacio R. Morelle wesnoth:master 15ac0224caa2 / INSTALL: Complete INSTALL revamp http://git.io/veo5F
20150407 22:16:22< irker403> wesnoth: Ignacio R. Morelle wesnoth:master 01e21487bad6 / src/hotkey/hotkey_preferences_display.cpp: Force xgettext to add the Add Hotkey and Clear Hotkey button tooltips http://git.io/veo5b
20150407 22:16:25< irker403> wesnoth: Ignacio R. Morelle wesnoth:master 9e7e08908c13 / src/theme.cpp: theme ui: Process translations of the "Hotkey(s): " string at runtime http://git.io/veo5N
20150407 22:17:06-!- new_one [~new_one@128.42.79.81] has quit [Quit: WeeChat 1.1.1]
20150407 22:20:04< shadowm> vultraz: Re 571999462f823df84538c252dcc61b669a80a98a, next time please make sure to backport this kind of thing to 1.12.
20150407 22:22:49< shadowm> iceiceice: Honestly, I wouldn't consider #23002 a blocker.
20150407 22:23:11-!- mjs-de [~mjs-de@f049109107.adsl.alicedsl.de] has quit [Remote host closed the connection]
20150407 22:23:36< iceiceice> i think we should elevate its priority then
20150407 22:23:42< shadowm> The reason is that this, for one reason or another, doesn't happen for me with the Spanish translation even though it also uses "lvl".
20150407 22:25:19< shadowm> The information that gets truncated can still be obtained by other means (reading the unit's entry in the help or paying attention when recruiting and leveling up) or is new to 1.12 (side flags and numbers are a new addition intended to make the game UI more accessible).
20150407 22:25:50< shadowm> Elevating a bug's priority IME doesn't help at all but I'll make it Important. Still odds are fendrin won't do much.
20150407 22:26:38-!- new_one [~new_one@128.42.81.71] has joined #wesnoth-dev
20150407 22:26:52-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has joined #wesnoth-dev
20150407 22:27:10< aquileia> shadowm: Doesn't Spanish use "nv" ?
20150407 22:27:25< shadowm> aquileia you are one of the people I keep forgetting to include in the highlight list.
20150407 22:28:01< shadowm> Same thing I said about 1.12 applies to you. Also I have the suspicion I wanted to contact you about something but I don't remember what it was anymore.
20150407 22:28:27< gfgtdf> i actualy am nreu what i shoudl do about http://gna.org/bugs/?23331 in 1.12, i fixed it in master and added a note in teh RL for 1.12 here https://github.com/wesnoth/wesnoth/commit/f5f0d5e4e952bd358a8cf0f1f1c39295275ebfbc becasue fixing it can cause OOS with previous versions, andone here thinks we should fix it in 1.12 regardless of possible OOS ?
20150407 22:28:39< aquileia> I think the proper way to fix truncating would be to align it on the right side, so that it's not the number being omitted...
20150407 22:29:02< gfgtdf> s/nreu what i shoudl/not shure what i should
20150407 22:29:15< Elvish_Hunter> Speaking of fendrin, does anyone know anything about Wings of Victory? Is it safe to consider that campaign abandoned?
20150407 22:29:16< aquileia> shadowm: Probably the PR you said 'nay' to?
20150407 22:29:17< shadowm> aquileia, iceiceice: Ah yes, "lvl" is the English string.
20150407 22:30:04< shadowm> And Spanish uses "nv", yes.
20150407 22:30:17< shadowm> I'm forgetting my own language.
20150407 22:30:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150407 22:30:47< aquileia> Because that's what I did for French (they had 'niv' which was too long)
20150407 22:31:23< iceiceice> haha
20150407 22:31:30< iceiceice> its ok i already forgot my language, and its my only one :p
20150407 22:32:10< shadowm> gfgtdf: If the patch goes in, then I guess it would only affect old clients when running time over events that alter the gamestate in some way?
20150407 22:32:48< iceiceice> does the game always end after a timeover event?
20150407 22:33:00< iceiceice> or it can continue
20150407 22:33:15< shadowm> I surmise people will usually alter the gamestate in a victory event triggered from there via [endlevel] instead.
20150407 22:33:18< gfgtdf> shadowm: yes but note that also things like [set_variable] rand= counts as gamestate change even if the variable is just used for a message.
20150407 22:33:41< gfgtdf> iceiceice: it can contnue
20150407 22:33:54< gfgtdf> iceiceice: when temover eent changes teh number of  turns
20150407 22:34:01< gfgtdf> event*
20150407 22:34:20< shadowm> In other words, how common is it for people to write time over event handlers that will cause OOS with old clients and probably cause OOS without the patch anyway?
20150407 22:34:57< iceiceice> idk it sounds like we are nontrivially breaking mp compatibility
20150407 22:35:09< gfgtdf> shadowm: quite uncommon, i think the most important case is whta i sadi above: a rand= used for message.
20150407 22:35:18< iceiceice> i guess we did that sort of with the carryover change already
20150407 22:35:25< iceiceice> but in that case they are pretty much screwed already anyawys
20150407 22:35:33< iceiceice> so theres not much lost
20150407 22:35:47< shadowm> And IIUC it's the same situation here.
20150407 22:36:12< aquileia> As to #386 you leaned to 'nay' when vultraz wanted to know what you'd do, may I ask why? Is dropping support for the ../addon.pbl  format really an issue considering gitignore & co. ?
20150407 22:36:49< shadowm> With or without the patch it's worth warning people against using time over in MP with 1.12.x I guess.
20150407 22:37:10< shadowm> Using it for stuff that would cause OOS, I mean.
20150407 22:38:07< Elvish_Hunter> shadowm: I can think of a possible situation, where a UMC author may want to use [modify_turns] in a time over event...
20150407 22:38:22< iceiceice> yeah i see
20150407 22:38:32< iceiceice> so because its not synced they really shouldnt use it in mp anyways
20150407 22:38:38< iceiceice> unless they are ending the level immediately
20150407 22:38:41< iceiceice> in which case its fine
20150407 22:38:57< iceiceice> yeah i have no problem to backport that if you want gfgtdf
20150407 22:39:00< iceiceice> if you think it wont break anything else
20150407 22:39:14< iceiceice> or anyways i see no reason against it
20150407 22:39:42< gfgtdf> iceiceice: meaning you would do the backport ?
20150407 22:41:30< iceiceice> no probably not
20150407 22:41:38< gfgtdf> iceiceice: ok
20150407 22:41:55< gfgtdf> iceiceice: was not a request, just a clarification on "i have no problem to backport that if you want gfgtdf"
20150407 22:42:49< Elvish_Hunter> shadowm: everything seems fine with my commits on 1.12. Now to fix 1.13...
20150407 22:42:51< iceiceice> yeah... i have other things i ned to do in the next 24 hours :p
20150407 22:46:21< shadowm> aquileia: Having a .pbl in the parent dir is just too convenient.
20150407 22:46:56< irker403> wesnoth: Elvish_Hunter wesnoth:master deac9515984d / data/core/units/monsters/Wolf_Great.cfg: Deleted variation_name from Great Wolf, as it isn't displayed http://git.io/veKeI
20150407 22:47:30< aquileia> Although I wouldn't agree, that's an easy fix if you require it
20150407 22:47:36< iceiceice> yeah i really dont agree
20150407 22:47:48< shadowm> I keep my .pbl files in the add-ons dir instead of having multiple copies or symbolic links for when I have to work with different copies of an add-on.
20150407 22:49:04< iceiceice> why is this more convenient than putting the pbl in the dir?
20150407 22:50:21< shadowm> Because I don't want to copy the same file to multiple copies of the same add-on.
20150407 22:51:44< Elvish_Hunter> I think that I found a way to enable filtering units by active statuses. Is anyone interested? http://pastebin.com/85yui6UN
20150407 22:53:12< gfgtdf> Elvish_Hunter: afaik taht si already possible
20150407 22:53:34< Elvish_Hunter> Yes, it is, but not without using [filter_wml]...
20150407 22:53:46< gfgtdf> Elvish_Hunter: it is also possible with foruma filter
20150407 22:53:52< Elvish_Hunter> Or a Lua function
20150407 22:54:44< Elvish_Hunter> This patch adds a status= key to [filter], which accepts a comma separated list of statuses
20150407 22:56:18-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150407 22:56:42< Elvish_Hunter> If there are no objections, I'll push it tomorrow
20150407 22:57:50< Elvish_Hunter> But for now... I really need to go and get some sleep. Bye!
20150407 22:58:55< gfgtdf> Elvish_Hunter: with formula it is [filter] formula = "states['slowed']" [/filter] (for slow status. Ii woudn't call that complicated.
20150407 22:59:17< gfgtdf> Elvish_Hunter: still you can push if you want.
20150407 23:00:42< Elvish_Hunter> Yes, I found quite a lot of places in mainline where the code (which uses [filter_wml]) will be simplified.
20150407 23:00:50< Elvish_Hunter> See you tomorrow!
20150407 23:00:54-!- Elvish_Hunter [~chatzilla@151.16.177.226] has quit [Quit: Ciao!]
20150407 23:02:27< aquileia> vultraz, AI0867, shadowm, iceiceice: As all other checks for the pbl file use get_pbl_file_path now, I can simply revert the change to that single function to reenable ../addon.pbl, and I already prepared that commit
20150407 23:03:05< aquileia> The question is whether it really is useful enough to make an exception for it
20150407 23:03:19< iceiceice> shadowm: so i guess the thing is
20150407 23:03:27< iceiceice> i dont understand why you have multiple folders for the same addon
20150407 23:03:33< iceiceice> why dont you just have one repo
20150407 23:03:38< iceiceice> and switch between them
20150407 23:04:01< iceiceice> if you have multiple versions on different branches,
20150407 23:04:08< iceiceice> and _server.pbl is in your gitignore,
20150407 23:04:14< iceiceice> then when you change branches you dont need to copy it will just be there
20150407 23:04:51< iceiceice> i think its not really a big deal to allow it in the add-on folder though
20150407 23:05:02< iceiceice> i just dont see the use
20150407 23:06:02-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20150407 23:07:32< shadowm> aquileia: I don't find adding comments to .ign files to be necessary but I don't have a problem with it.
20150407 23:10:43< gfgtdf> iceiceice: i just got a pm in forum about [variables] wrong for observers in networked mp after scenario transition in 1.12, you know anything about this?
20150407 23:10:56< iceiceice> no i dont know
20150407 23:11:17< iceiceice> we need to make a way to unit test the carryover
20150407 23:28:35< irker403> wesnoth: gfgtdf wesnoth:1.12 2a8abdf199df / src/ (synced_context.cpp synced_context.hpp): split set_scontext_synced class http://git.io/veKYT
20150407 23:28:44< gfgtdf> what
20150407 23:28:50< shadowm> What.
20150407 23:29:04< aquileia> shadowm: Happy? https://github.com/wesnoth/wesnoth/pull/386
20150407 23:29:44< gfgtdf> shadowm: this was accidently a forcepush i thought "git push -f" pushes to my fork and not to wesnoth/wesnoth
20150407 23:29:55< shadowm> What.
20150407 23:30:27< shadowm> gfgtdf: Are you absolutely sure it's a force-push? I can fix it immediately if so.
20150407 23:31:50< gfgtdf> shadowm: i definitely typed "git push -f" but i did a "git update-ref refs/heads/1.12 refs/remotes/upstream/1.12" shortly before
20150407 23:32:01< shadowm> git push (-f or no -f) pushes to whichever remote was the last one you pushed to.
20150407 23:32:45< shadowm> gfgtdf: No commits were lost then, you can just revert HEAD.
20150407 23:33:11< aquileia> shadowm: Ahem... I think it pushes to the tracked remote of the current branch
20150407 23:33:15< shadowm> At least I don't think anyone else committed anything after a8b8dd80233d2050f81acdc527b7aac448630d61 and irker somehow missed it.
20150407 23:33:41< shadowm> aquileia: Ah okay, that probably makes more sense.
20150407 23:34:49< gfgtdf> shadowm: most likely my 1.12 branch is somwhow set up wrongly, on feature branches and master it pushes to my tree if i use "git push" wthithout specyfying the remote
20150407 23:34:58-!- knotwork [~markm@unaffiliated/knotwork] has joined #wesnoth-dev
20150407 23:35:37< aquileia> That's how I avoid accidentally force-pushing: My tracked remote pulls from wesnoth/wesnoth and pushes to aquileia/wesnoth, so any commits go to my fork by default
20150407 23:35:49< vultraz> iceiceice: ya know, we're gonna get a large number of people on the forums asking about these strict lua warnings
20150407 23:36:08< vultraz> s/warnings/errors
20150407 23:36:14< aquileia> In any case, I'm off to bed
20150407 23:37:19< aquileia> shadowm: Feel free to merge or close the PR
20150407 23:37:27-!- aquileia [95acd0d3@gateway/web/freenode/ip.149.172.208.211] has quit [Quit: Page closed]
20150407 23:37:35< iceiceice> vultraz: i guess i need to document it on wiki
20150407 23:37:52< vultraz> or make it  warning not an error?
20150407 23:39:36< shadowm> aquileia: You are a developer too, you know. :\
20150407 23:39:47< shadowm> I'd prefer if the last two commits were one instead.
20150407 23:45:14< iceiceice> vultraz: we could make it a warning,
20150407 23:45:18< iceiceice> but if we do it makes the lua interpreter act strangely
20150407 23:45:37< iceiceice> maybe its fine, would have to think about it though
20150407 23:45:45< iceiceice> do you know how the lua interpreter works?
20150407 23:46:08< vultraz> no? 
20150407 23:46:15< iceiceice> like, how it decides what to do?
20150407 23:46:39< vultraz> no
20150407 23:46:43< iceiceice> so normally, if you just run lua
20150407 23:46:52< iceiceice> you can't just type "5+5" and expect to see 10
20150407 23:46:58< iceiceice> you have to type "print(5+5)"
20150407 23:47:15< iceiceice> lua expects that everything you ever give it can resolve to a command
20150407 23:47:30< iceiceice> thats how it will work if you install lua from ubuntu repositoy,
20150407 23:47:36< iceiceice> i imagine thats how lua for windows works also
20150407 23:47:45< iceiceice> but thats not very nice for an interpreter
20150407 23:47:54< iceiceice> usually you want to be able to just type some variable name, and be like "wtf is this"
20150407 23:48:32< iceiceice> so i read about what people do, what the interpreter actually does rihgt now is this
20150407 23:48:50< iceiceice> (1) there is a function called "_pretty_print" which prints most lua things in a nice way
20150407 23:49:14< iceiceice> (2) take whatever the user typed, and see what lua thinks of "_pretty_print(stuff)"
20150407 23:49:28< iceiceice> if that evaluates to a valid lua command, then run it
20150407 23:49:42< iceiceice> if lua thinks it is an error, then instead just try to run "stuff" as a lua command
20150407 23:49:50< iceiceice> and if thats an error then call it an error
20150407 23:50:20< iceiceice> this turns out to work almost exactly the way we want, and i dont really want to start screwing around with it if i can help it
20150407 23:50:45< iceiceice> if we make things just a warning instead of an error,
20150407 23:50:52-!- prkc [~prkc@4E5CD609.dsl.pool.telekom.hu] has quit [Remote host closed the connection]
20150407 23:51:15< iceiceice> it could be that it starts taking second option when it should have taken first or something
20150407 23:51:24< irker403> wesnoth: gfgtdf wesnoth:1.12 39775b2a2c3f / src/ (playsingle_controller.cpp replay_controller.cpp): sync "time over" event http://git.io/veKnU
20150407 23:51:34< iceiceice> idk i havent really thought about it
20150407 23:51:53< iceiceice> usually though, when the user types some random variable that doesnt exist,
20150407 23:52:02< iceiceice> you want it to say "hey that doesnt exist" and not start running a bunch of code
20150407 23:52:31< vultraz> problem is, this strict lua thing doesn't always work well, and it's a new thing from 1.12 that is definitely gonna raise outcry
20150407 23:52:33< vultraz> for example
20150407 23:52:41< iceiceice> its not in 1.12
20150407 23:52:44< iceiceice> its only in master
20150407 23:52:53< vultraz> i meant, something new that wasn't in 1.12
20150407 23:52:56< vultraz> cannottype
20150407 23:53:20< iceiceice> i think actually the people who are most likely to care are in this channel :p
20150407 23:53:30< vultraz> point being, if I do, say 'if not lp8 then', I want it to work specifically if the variable is either false or missing
20150407 23:54:01< iceiceice> yeah... so there are various workarounds
20150407 23:54:19< vultraz> I COULD initialize lp8 as a nil variable first, yes
20150407 23:54:36< vultraz> mostly what I do is just comment out the error in ilua :P
20150407 23:54:40< iceiceice> i'm not saying it doesnt sometimes inconveient you
20150407 23:54:49< iceiceice> but i think it also saves you a lot of time in other cases
20150407 23:55:08< vultraz> well, it would be nice to have a switch for it
20150407 23:55:09< iceiceice> like any time you make a typo
20150407 23:55:13< iceiceice> there is a switch
20150407 23:55:13< vultraz> you added one, but it doesn't appear to work
20150407 23:55:19< iceiceice> really?
20150407 23:55:26< vultraz> yeah
20150407 23:55:47< iceiceice> how were you using it
20150407 23:56:06< iceiceice> ohhh i remember now, i think i added a simpler switch later
20150407 23:56:17< iceiceice> hmmm
20150407 23:56:40< vultraz> setting ilua.strict to false in my addon 
20150407 23:57:27< iceiceice> well heres the code:
20150407 23:57:27< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/data/lua/ilua.lua
20150407 23:58:05< irker403> wesnoth: gfgtdf wesnoth:1.12 3ed7c59d5405 / RELEASE_NOTES: Update RELEASE_NOTES http://git.io/veKc6
20150407 23:58:30< iceiceice> did i misunderstand something about lua?
20150407 23:58:48< shadowm> gfgtdf, will you revert that commit?
20150407 23:58:54< iceiceice> i thought that when you make a closure which references an external var it will continuet o refer to the external var
20150407 23:59:15< gfgtdf> shadowm: since i wanted this commit to be in 1.12 and i didnt plan to do that
20150407 23:59:18< shadowm> The one you accidentally pushed upstream, that is. Unless you decided to actually commit that fix, I guess.
20150407 23:59:39< gfgtdf> s/and/anyway
--- Log closed Wed Apr 08 00:00:23 2015