--- Log opened Mon Mar 07 00:00:24 2016
20160307 00:00:45-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 276 seconds]
20160307 00:25:57-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160307 00:32:26< celticminstrel> I can never seem to find the GUI2 reference on the wiki.
20160307 00:36:09-!- louis94 [~~louis94@91.178.241.37] has quit [Quit: Konversation terminated!]
20160307 02:21:40< celticminstrel> I almost wonder if I should /x these regexes for better readability, because it's getting harder and harder to understand them.
20160307 02:49:18-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160307 03:15:19< celticminstrel> I don't really understand how gettext uses backslashes.
20160307 03:15:38< celticminstrel> But I'm pretty sure it's wrong for a string to end with \"
20160307 03:16:00< celticminstrel> And I would assume that a string containing \\n is not going to be interpreted as containing a newline.
20160307 03:31:09< celticminstrel> Okay, so, both with and without the wmlxgettext change I did there, it doesn't seem to actually work properly on Lua files.
20160307 03:32:25< celticminstrel> The patch does at least seem to do what it says, even if that doesn't help much.
20160307 03:32:53< celticminstrel> Basically there seems to be something fundamentally broken about how it matches the Lua strings in the first place.
20160307 03:36:08< celticminstrel> And it seems to work for the most basic of cases which are probably the majority.
20160307 03:44:37< celticminstrel> Sooo in light of that I'm going to not bother trying to fix it. I kinda hope that guy who was working on a Python version comes back at some point. (Though I forget his name, so maybe he's actually lurking or something.)
20160307 03:47:54-!- danni_ [~quassel@203-206-161-81.perm.iinet.net.au] has quit [Remote host closed the connection]
20160307 03:48:00-!- danni [~quassel@203-206-161-81.perm.iinet.net.au] has quit [Remote host closed the connection]
20160307 03:49:06-!- danni [~quassel@203-206-161-81.perm.iinet.net.au] has joined #wesnoth-dev
20160307 04:02:57-!- danni [~quassel@203-206-161-81.perm.iinet.net.au] has quit [Remote host closed the connection]
20160307 04:13:43-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160307 04:14:15-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20160307 04:43:21-!- irker065 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20160307 04:48:28< Aginor> good job everyone on the 1.13.4 hotfix release
20160307 04:57:02-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160307 05:14:04-!- janebot_ [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20160307 05:14:10-!- janebot_ [~Gambot@grickit.us] has joined #wesnoth-dev
20160307 05:22:28< vultraz> shadowm: I've gotten the files for the widget set up, but I'm not sure how to actually populate it
20160307 05:23:23< vultraz> does find_widget work inside widgets?
20160307 05:23:41< vultraz> or else, am I supposed to use [text] draw routines?
20160307 05:24:32< vultraz> listboxes use other widgets inside it but that uses tgenerator
20160307 05:25:09< vultraz> am I supposed to have a content grid and use grid.find
20160307 05:25:31< vultraz> none of this is self-evident :|
20160307 05:29:28< celticminstrel> What do you mean by "populate it"?
20160307 05:30:01< vultraz> fill it in
20160307 05:30:17< celticminstrel> With what exactly?
20160307 05:30:20< celticminstrel> Data?
20160307 05:30:21< vultraz> text
20160307 05:30:23< vultraz> yes
20160307 05:30:34< vultraz> the stats of the unit
20160307 05:30:37< celticminstrel> And you mean from the perspective of the class implementing the widget, I presume?
20160307 05:30:44< vultraz> though in these cases, the unit type
20160307 05:30:47< vultraz> yes
20160307 05:32:39< vultraz> right now I'm thinking a display_stats(unit_type*) function or something in the widget class that gets the content grid and uses grid.find to get the widgets and fill them in
20160307 05:33:24< vultraz> but I'm not 100% sure how content_grid() works
20160307 05:34:22< vultraz> hmmm
20160307 05:34:30< vultraz> there IS precedent for find_widget in widgets
20160307 05:34:59< vultraz> wait
20160307 05:35:11< vultraz> content_grid seems to be unique to scrollbar containers 
20160307 05:35:21< vultraz> ok, I guess I can just use a named grid
20160307 05:35:36< vultraz> or...wait, why do I need to do that?
20160307 05:37:20< vultraz> ok .. I think I'll try keeping a unit type pointer in the class
20160307 05:37:50< vultraz> then just changing the pointer and calling the draw function to operate on that
20160307 05:41:06 * vultraz grits teeth
20160307 05:41:18< celticminstrel> You'll want to have some sort of set_value() type function with two overloads, one taking a unit_type and the other taking a unit.
20160307 05:41:42< celticminstrel> (Though you can put off the implementation of the second for now, since it's not needed for recruit and create unit.)
20160307 05:42:08< celticminstrel> Eh?
20160307 05:42:28< celticminstrel> I think it would be better to store the actual stats in the class rather than the pointer to whatever.
20160307 05:43:01< celticminstrel> ...but none of that is actually working towards your problem of how to work with widgets within widgets.
20160307 05:43:10< celticminstrel> I suggest that you start by trying find_widget.
20160307 05:43:28< celticminstrel> Maybe you can even pass "this" as the first argument.
20160307 05:44:41< vultraz> gui2 makes me want to scream sometimes
20160307 05:49:30< vultraz> I need to get a grid somehow...
20160307 05:50:04< vultraz> shadowm: help needed
20160307 05:50:10< vultraz> how do I get a grid
20160307 05:50:16< vultraz> to find the other widgets
20160307 05:55:38< vultraz> what do you mean config is not a type: :|
20160307 06:02:53< celticminstrel> You probably need to include config.hpp?
20160307 06:03:09< vultraz> I did
20160307 06:03:24< vultraz> didn't work so I changed it to FOREACH( AUTO&
20160307 06:07:16-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Read error: Connection reset by peer]
20160307 06:07:50-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev
20160307 06:08:29< vultraz> GAH
20160307 06:08:31< vultraz> const
20160307 06:08:34< vultraz> damn const
20160307 06:10:29< vultraz> is this safe
20160307 06:10:32< celticminstrel> ?
20160307 06:10:33< vultraz> 	find_widget<tunit_preview_pane>(&window, "unit_details", false)
20160307 06:10:34< vultraz> 		.set_displayed_type(const_cast<unit_type*>(units_[selected_row]));
20160307 06:10:54< celticminstrel> Why the const_cast?
20160307 06:11:31< vultraz> because units_ is a vector of const pointers and set_displayed_type has to take a non-const pointer since the class pointer has to be non-const to be reassigned
20160307 06:11:36< celticminstrel> I assume you defined set_displayed_type, so you can change it to take a const unit_type pointer (or better still, reference) rather than adding the const_cast.
20160307 06:12:00< celticminstrel> Uh.
20160307 06:12:07< celticminstrel> "reassigned"?
20160307 06:12:13< vultraz> if I make it a reference I need to define it in the constructor 
20160307 06:12:23< vultraz> celticminstrel: make it point to different types, I mean
20160307 06:12:31< celticminstrel> Uh. What.
20160307 06:12:36< celticminstrel> Oh wait.
20160307 06:12:45< vultraz> .set_displayed_type sets the pointer to the specified unit type
20160307 06:12:47< celticminstrel> You mean "different unit types", not "different classes".
20160307 06:12:47< vultraz> that
20160307 06:12:51< vultraz> 's then used to fill out the list
20160307 06:12:52< celticminstrel> Right.
20160307 06:12:56< vultraz> er, the panel
20160307 06:13:09< celticminstrel> But wait, what are you going to do when you need to make it show a specific unit rather than a type?
20160307 06:13:20< celticminstrel> Why does it need a reference back to the unit[_type]?
20160307 06:14:01< vultraz> I don't know yet how to handle a specific unit
20160307 06:14:19< celticminstrel> [Mar 07@01:12:13am] vultraz: if I make it a reference I need to define it in the constructor
20160307 06:14:20< celticminstrel> That's if you make the member variable a reference. You can still make the function parameter a reference without that problem.
20160307 06:14:31< vultraz> ohh, I seeeee
20160307 06:14:40< celticminstrel> So why does your widget need a reference back to the unit_type.
20160307 06:15:10< celticminstrel> I'm asking because doing it that way will probably make your life harder when you get to adding support for showing a specific unit.
20160307 06:15:27< vultraz> because it need the data for filling out the panel!
20160307 06:15:47< celticminstrel> I don't see why that requires it to keep track of a reference to the unit_type.
20160307 06:15:53< celticminstrel> The data is static, isn't it?
20160307 06:15:54< vultraz> grrrrr... error: binding 'const unit_type' to reference of type 'unit_type&' discards qualifiers|
20160307 06:16:17< celticminstrel> It's not actually going to change while the widget is displayed, is it?
20160307 06:16:26< celticminstrel> (Except by changing it to a completely different unit_type.)
20160307 06:16:33< vultraz> the data is static yes
20160307 06:16:48< celticminstrel> So, why do you need to store a reference back to the unit_type?
20160307 06:17:06< vultraz> because I need to fill it out when a type changes!
20160307 06:17:26< celticminstrel> But when the type changes, you're given a new unit_type, so you're not even using that reference.
20160307 06:18:09< celticminstrel> Do you need to be able to query "which unit is this pane showing?"?
20160307 06:18:15< vultraz> no
20160307 06:18:46< celticminstrel> Is there some other reason I haven't thought of that requires you to store a reference to the unit_type?
20160307 06:18:59< vultraz> are you saying set_displayed_type should be the function doing the filling out directly?
20160307 06:19:11< vultraz> right now it sets the pointer and then calls display_stats
20160307 06:19:42< celticminstrel> So display_stats() takes no arguments?
20160307 06:19:49< vultraz> no
20160307 06:20:09< celticminstrel> Does it need to be called at any time other than when the unit changes?
20160307 06:20:09< vultraz> I guess I could try your suggestion
20160307 06:20:14< vultraz> no
20160307 06:20:53< celticminstrel> Well, if you really need a reference to the unit_type, you'll need to eventually learn about unions (or variants), but I suspect you don't need that reference.
20160307 06:21:15< vultraz> you're probably right
20160307 06:21:18< celticminstrel> [Mar 07@01:18:59am] vultraz: are you saying set_displayed_type should be the function doing the filling out directly?
20160307 06:21:19< celticminstrel> And yes, that's pretty much what I was suggesting.
20160307 06:22:18< vultraz> ok
20160307 06:22:23< vultraz> I still have the problem of the grid, though
20160307 06:23:13< vultraz> without the grid I can't find any widgets
20160307 06:25:17< vultraz> (it does build now)
20160307 06:25:29< shadowm> The first rule of const_cast club is you don't use const_cast.
20160307 06:25:57-!- Kwandulin [~Miranda@p200300760F0BD491046E263E91486EBF.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160307 06:26:04< celticminstrel> I used const_cast somewhere to implement my operator[]()const in terms of my operator()[].
20160307 06:26:24< celticminstrel> It's safe there, I'm pretty sure.
20160307 06:26:41< celticminstrel> But yeah, generally don't.
20160307 06:26:55< shadowm> More seriously, if you feel the need to use const_cast, it means that either 1) you are doing something wrong (so shame on you); 2) dealing with a poorly designed API (so shame on the author); 3) you are doing something wrong (so shame on you).
20160307 06:27:57< vultraz> I removed it
20160307 06:28:32< vultraz> More importantly, do you know how to get a widget's grid from within that widget without using scrollbar containers. Is tgenerator needed.
20160307 06:29:04< shadowm> I don't see any question marks so I'm assuming that's not a question.
20160307 06:29:30< shadowm> (Besides, I've never written a GUI2 widget before, so I obviously have no idea.)
20160307 06:29:37< vultraz> More importantly, do you know how to get a widget's grid from within that widget without using scrollbar containers? Is tgenerator needed?*
20160307 06:40:36-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160307 06:43:14< celticminstrel> I just discovered a bizarre formula quirk - all functions compare equal.
20160307 06:43:58< celticminstrel> Or at least, it looks like that.
20160307 06:44:07< celticminstrel> I imagine the parser is actually treating them as undefined variables.
20160307 06:46:43-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has joined #wesnoth-dev
20160307 06:56:30-!- janebot_ [~Gambot@grickit.us] has quit [Remote host closed the connection]
20160307 06:56:36-!- janebot_ [~Gambot@grickit.us] has joined #wesnoth-dev
20160307 07:04:38< vultraz> looks like I might need a... swap_grid method?
20160307 07:04:49< vultraz> wait no
20160307 07:04:54< vultraz> even that's in a scrollbar panel
20160307 07:05:02< vultraz> maybe I should put the widget in a scrollbar panel
20160307 07:05:36< vultraz> no...
20160307 07:05:52-!- ancestral [~ancestral@97-116-184-84.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20160307 07:13:47< vultraz> god dammit, I can't think of anything
20160307 07:13:59-!- Kwandulin [~Miranda@p200300760F0BD491046E263E91486EBF.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160307 08:00:28-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160307 08:04:46-!- Kwandulin [~Miranda@p200300760F0BD4915C89C0A240173820.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160307 08:33:25< vultraz> I don't think its possible
20160307 08:34:56< vultraz> how is such a simple thing not possible
20160307 08:36:02-!- Kwandulin [~Miranda@p200300760F0BD4915C89C0A240173820.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160307 08:41:13-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20160307 08:42:02< celticminstrel> I'm sure it's possible, but sorry I can't help you figure out how.
20160307 08:42:07< celticminstrel> At least not right now.
20160307 08:42:10-!- celticminstrel is now known as celmin|sleep
20160307 08:42:18< vultraz> zookeeper: doofus posted chasm transitions 
20160307 08:57:02 * vultraz is experimenting on contrast boost for forest
20160307 09:10:09< wedge009> Regarding https://gna.org/bugs/?24456, does anyone recall if the damage indicator in the side-bar should include the effects of charge?
20160307 09:11:57< vultraz> hm
20160307 09:11:59< vultraz> not sure..
20160307 09:12:10< zookeeper> i think that was intentional
20160307 09:12:45< zookeeper> it's not dependent on what you attack, after all
20160307 09:14:04< zookeeper> vultraz, no, not forests, they're good as-is
20160307 09:14:38< wedge009> Thanks, that's what I thought.
20160307 09:15:29< vultraz> zookeeper: I think Pine Forest at least could use it
20160307 09:16:30< vultraz> but im not sure
20160307 09:17:02< zookeeper> good
20160307 09:22:21-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 246 seconds]
20160307 09:23:39-!- aeonchild [enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 276 seconds]
20160307 09:24:53-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160307 09:27:30-!- aeonchild [enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20160307 09:42:32-!- Appleman1234 [~Appleman1@KD106161146204.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20160307 10:07:02< zookeeper> celmin|sleep, yeah, backstab seems to be displayed right now
20160307 10:11:46-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160307 10:14:05-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 244 seconds]
20160307 10:14:06-!- wedge010 is now known as wedge009
20160307 10:39:13-!- Appleman1234 [~Appleman1@KD106161149214.au-net.ne.jp] has joined #wesnoth-dev
20160307 10:52:26-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has joined #wesnoth-dev
20160307 10:53:43< Elvish_Hunter> Hi all
20160307 10:54:14< vultraz> hey
20160307 10:54:35< Elvish_Hunter> loonycyborg: thanks for your reports. I already sent those informations to Nobun.
20160307 10:55:12< loonycyborg> kk
20160307 10:55:37< Elvish_Hunter> celmin|sleep: the Python 3 version of wmlxgettext is available here: https://github.com/AncientLich/wmlxgettext-unoff/
20160307 10:57:37-!- Appleman1234 [~Appleman1@KD106161149214.au-net.ne.jp] has quit [Remote host closed the connection]
20160307 10:58:59-!- Appleman1234 [~Appleman1@KD106161149214.au-net.ne.jp] has joined #wesnoth-dev
20160307 11:01:13< Elvish_Hunter> loonycyborg: the POT-Creation-Date is different, because the strftime %z directive isn't portable across platforms.
20160307 11:02:18< Elvish_Hunter> For that reason, Nobun decided (after my suggestion) to make the time UTC, because the script must be able to run on Windows as well.
20160307 11:12:45< Elvish_Hunter> Sorry if I have to leave so soon, but today I'm quite busy...
20160307 11:13:05-!- Elvish_Hunter [~elvish_hu@wesnoth/developer/elvish-hunter] has left #wesnoth-dev ["Ciao!"]
20160307 11:22:49-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160307 11:38:26-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160307 11:42:34-!- mjs-de [~mjs-de@x5ce33779.dyn.telefonica.de] has joined #wesnoth-dev
20160307 11:52:43-!- louis94 [~~louis94@91.178.240.61] has joined #wesnoth-dev
20160307 11:53:28-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 250 seconds]
20160307 12:01:18-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160307 12:06:12-!- louis94 [~~louis94@91.178.240.61] has quit [Quit: Konversation terminated!]
20160307 12:52:11-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 244 seconds]
20160307 12:54:52-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20160307 12:54:52-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20160307 12:54:52-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20160307 12:55:28-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160307 13:02:01-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20160307 13:17:29-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160307 13:23:17-!- travis-ci [~travis-ci@ec2-54-166-112-204.compute-1.amazonaws.com] has joined #wesnoth-dev
20160307 13:23:18< travis-ci> dreamer-88/wesnoth#7 (master - c7de7ce : Celtic Minstrel): The build has errored.
20160307 13:23:18< travis-ci> Build details : https://travis-ci.org/dreamer-88/wesnoth/builds/114230662
20160307 13:23:18-!- travis-ci [~travis-ci@ec2-54-166-112-204.compute-1.amazonaws.com] has left #wesnoth-dev []
20160307 13:47:45-!- gfgtdf [~chatzilla@x50ab6ea7.dyn.telefonica.de] has joined #wesnoth-dev
20160307 13:49:42< gfgtdf> vultraz: i think a unit panel widget shoudl supprot both untis types and unit, so that it can be used in rectuit dialogs in recall dialogs in debug create dialogs and also in the ingame right side panel if we ever port that to gui2#+
20160307 13:49:58< vultraz> yes, I agree
20160307 13:50:16< vultraz> but can you help me figure out how to actually access the internals?
20160307 13:50:24< vultraz> I need the widget's internal grid
20160307 13:50:27< vultraz> for find_widget
20160307 13:50:34< vultraz> or do I need to use [draw] and [text]
20160307 13:52:24< vultraz> at this point I'm thinking it might just be better to make it a non-widget helper class
20160307 14:02:15< vultraz> gfgtdf: do you know anything about getting the internal grid of a widget
20160307 14:13:10< gfgtdf> vultraz: does your widget inherit form tcontrol ?
20160307 14:13:16< vultraz> yes
20160307 14:14:44< gfgtdf> vultraz: for tcontrol you can usually define the 'child widgets' in the .cfg file liek its done in listbox_default.cfg
20160307 14:15:40< gfgtdf> vultraz: then if you gave them an id you can usuualy call find_widget(this, "<widget_id>") from the widgets methods to find whild widgets
20160307 14:17:55< gfgtdf> vultraz: for that you of coruse have to give those child widgets ids in the .cfg file
20160307 14:32:03< vultraz> hmm
20160307 14:32:09< vultraz> that's not working for some reason :|
20160307 14:54:06-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160307 15:07:30< vultraz> I don't understand
20160307 15:10:31< gfgtdf> vultraz: hmm mabye you can show code ?
20160307 15:10:38< vultraz> OK
20160307 15:10:42< vultraz> lemme make a commit in my fork
20160307 15:19:39< vultraz> gfgtdf: https://github.com/Vultraz/wesnoth/commit/08e660c83bf84124b8ddd910c50f05d675ef706b
20160307 15:42:00< vultraz> gfgtdf: it's late, im off, but if you figure out something just leave a message and ill read it in the morning
20160307 16:12:24-!- Kwandulin [~Miranda@p200300760F7D73CB49D2A469E7978ECD.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160307 16:18:05< gfgtdf> vultraz: hmm i'm wonderweing whetehr it shoudl maybe inherit form tcontriner_
20160307 16:18:33< gfgtdf> vultraz: " A container is a class build with multiple items either acting as one widget"
20160307 16:18:39< gfgtdf> vultraz: this seems to e what we want
20160307 16:19:09< gfgtdf> be*
20160307 16:24:14< gfgtdf> vultraz: y the more i think about it it makes sense. I'd try to copy code from te scrollbar_panel widget
20160307 16:42:19-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 16:42:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 16:42:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 16:54:11-!- aquileia [~androirc@176.2.34.132] has joined #wesnoth-dev
20160307 16:58:23< aquileia> shadowm, vultraz: Can't find it in the logs, but you recently contemplated that only Galician and Spanish are fully translated. Ever looked up why? Look at the stats and content of wesnoth-ai and it should be obvious
20160307 16:59:40< aquileia> We can't demand of the translation teams to work on that just to get 100% in the statistics
20160307 17:01:04< aquileia> 400 translatable strings from the micro-ai test scenarios which are only accessible via command line
20160307 17:02:04-!- fendrin [~quassel@176.7.53.120] has joined #wesnoth-dev
20160307 17:02:04-!- fendrin [~quassel@176.7.53.120] has quit [Changing host]
20160307 17:02:04-!- fendrin [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160307 17:03:37-!- aquileia [~androirc@176.2.34.132] has quit [Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )]
20160307 17:05:37-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 240 seconds]
20160307 17:12:45< celmin|sleep> zookeeper: Huh?
20160307 17:17:27< zookeeper> ?
20160307 17:19:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160307 17:19:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 17:20:06-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has quit [Ping timeout: 276 seconds]
20160307 17:22:19< celmin|sleep> zookeeper: You said something about backstab.
20160307 17:22:53< zookeeper> but you don't know what it is? :P
20160307 17:23:50< celmin|sleep> I can't tell if it was good or bad.
20160307 17:24:15< zookeeper> "right" was as in "correctly", so good
20160307 17:24:24< celmin|sleep> Ah, okay.
20160307 17:24:44-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160307 17:24:54-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160307 17:25:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 17:25:16< celmin|sleep> For some reason I bound "right now" together as a phrase.
20160307 17:25:43< zookeeper> fair enough
20160307 17:33:24-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160307 17:34:58< celmin|sleep> Nobun's English is pretty bad...
20160307 17:42:20< celmin|sleep> I'm pretty sure WML has no such thing as [-tag], heh.
20160307 17:43:51< celmin|sleep> Also, I think he's mishandling [+tag].
20160307 17:53:17< celmin|sleep> His Lua parsing might be worse than current wmlxgettext in some ways.
20160307 17:55:26-!- louis94 [~~louis94@91.178.240.61] has joined #wesnoth-dev
20160307 17:58:19< celmin|sleep> In fact, based on the documentation, his WML file parsing might be worse too in some ways...
20160307 18:00:37-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160307 18:01:03-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 18:03:06-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160307 18:05:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Read error: Connection reset by peer]
20160307 18:05:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 18:07:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 18:09:44-!- bumbadadabum [~bumbadada@wesnoth/developer/bumbadadabum] has joined #wesnoth-dev
20160307 18:12:58-!- celmin|sleep is now known as celticminstrel
20160307 18:25:53-!- mjs-de [~mjs-de@x5ce33779.dyn.telefonica.de] has quit [Remote host closed the connection]
20160307 18:30:41-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 18:34:08-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 18:34:15-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 18:37:44-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has quit [Quit: Leaving]
20160307 18:38:04-!- prkc [~prkc@gateway/vpn/privateinternetaccess/prkc] has joined #wesnoth-dev
20160307 18:43:52-!- Kwandulin_2 [~Miranda@p200300760F7D73CB49D2A469E7978ECD.dip0.t-ipconnect.de] has joined #wesnoth-dev
20160307 18:44:45-!- Kwandulin [~Miranda@p200300760F7D73CB49D2A469E7978ECD.dip0.t-ipconnect.de] has quit [Ping timeout: 248 seconds]
20160307 18:56:37< celticminstrel> There's something wrong with the debugging information from the XCode project.
20160307 18:57:09< celticminstrel> Some of the functions seem to have symbols but not line information, and thus show up greyed out in the stack trace and as assembly if you step into them.
20160307 18:57:54< celticminstrel> (Normally, greyed-out entries in the stack trace are stack frames not belonging to your application, such as API calls.)
20160307 19:15:01-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20160307 19:15:42-!- gfgtdf [~chatzilla@x50ab6ea7.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20160307 19:16:29-!- louis94 [~~louis94@91.178.240.61] has quit [Quit: Konversation terminated!]
20160307 19:23:43-!- louis94 [~~louis94@91.178.240.61] has joined #wesnoth-dev
20160307 19:27:18< celticminstrel> Several times now I've considered adding a function or operator to the formula engine, only to suddenly realize it can already be done with the tools the engine provides.
20160307 19:30:12-!- louis94 [~~louis94@91.178.240.61] has quit [Quit: Konversation terminated!]
20160307 19:30:31-!- louis94 [~~louis94@91.178.240.61] has joined #wesnoth-dev
20160307 19:32:42< celticminstrel> It turns out that the formula engine has had string concatenation since 1.9!
20160307 19:38:17-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has joined #wesnoth-dev
20160307 19:50:09-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20160307 19:56:59-!- gfgtdf [~chatzilla@x50ab6ea7.dyn.telefonica.de] has joined #wesnoth-dev
20160307 19:57:09< gfgtdf> celticminstrel: do you intend to use formule in your add ons ?
20160307 19:57:35< celticminstrel> Probably not FormulaAI, at least.
20160307 19:57:41< celticminstrel> But for other things, possibly.
20160307 19:59:08< celticminstrel> Why do you ask?
20160307 19:59:58-!- louis94 [~~louis94@91.178.240.61] has quit [Ping timeout: 244 seconds]
20160307 20:00:25< gfgtdf> celticminstrel: i just wondered where one woudl use foruma in addon but then i rmembered that even the $() things are actualyl formula
20160307 20:00:32< celticminstrel> Yeah.
20160307 20:00:43< celticminstrel> Also formula= in unit filters.
20160307 20:02:19< gfgtdf> celticminstrel: y i know, some things can be easiyl done with formula filters that are quiet hard to do in wml like 'all units that have a arcane attack that does at least 10 damage'. But usually i think lua is more powerful
20160307 20:02:47< celticminstrel> I'm not sure Lua is more powerful; it might actually be equal.
20160307 20:03:22< celticminstrel> Though Lua is procedural while formula is functional, so the way to do certain things may be radically different.
20160307 20:04:10< celticminstrel> There's no direct looping construct in WFL - you loop using higher-order functions such as map, reduce, and filter.
20160307 20:06:38< gfgtdf> celticminstrel: hmm im gettign linker errors relted to template memeber fucntions and i doner why. I thought i can do that as long as the are defined in teh same file that uses them
20160307 20:08:13< celticminstrel> I think I would need more information to help.
20160307 20:13:42< gfgtdf> celticminstrel: hmm seems liek some pot or image updates were done since i pushed to my branch coudl take a while
20160307 20:15:29< gfgtdf> celticminstrel: hmm found it, i wrote handle_up_down_arrow() instead of ttree_view::handle_up_down_arrow()
20160307 20:26:59-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 20:27:49-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 20:32:18-!- travis-ci [~travis-ci@ec2-54-166-112-204.compute-1.amazonaws.com] has joined #wesnoth-dev
20160307 20:32:18< travis-ci> gfgtdf/wesnoth-old#607 (master - bf54c5c : gfgtdf): The build is still failing.
20160307 20:32:18< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth-old/builds/114341108
20160307 20:32:18-!- travis-ci [~travis-ci@ec2-54-166-112-204.compute-1.amazonaws.com] has left #wesnoth-dev []
20160307 20:32:19< zookeeper> vultraz, the filter bug with the debug unit create dialog still exists
20160307 20:32:27< zookeeper> urgh, poor timing...
20160307 20:32:39< gfgtdf> ?
20160307 20:33:30< vultraz> zookeeper: I fixed that!
20160307 20:33:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 20:33:44< celticminstrel> What was the bug, again?
20160307 20:33:46-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 20:33:52< celticminstrel> Crash if the filter matches nothing?
20160307 20:33:56< zookeeper> vultraz, it still happens!
20160307 20:33:56< vultraz> yes
20160307 20:34:07< zookeeper> no
20160307 20:36:41< vultraz> wait, you mean a different bug?
20160307 20:36:55< zookeeper> yes
20160307 20:38:42< vultraz> what is this bug
20160307 20:39:55< zookeeper> the one where it doesn't work right. trying to reliably reproduce atm.
20160307 20:40:21< vultraz> well I can't help atm
20160307 20:40:42< vultraz> I'm working on a bunch of stuff in my branch 
20160307 20:40:55< celticminstrel> Doesn't work right how?
20160307 20:42:40< zookeeper> create a dark adept. open the dialog again, type "cu" to narrow the list down to cuttlefish, try to create it. no go.
20160307 20:43:02< celticminstrel> No unit appears?
20160307 20:43:33< zookeeper> yes
20160307 20:43:37< celticminstrel> I mysteriously got this once when trying to create a thief; trying a second time worked, but I have no idea what went wrong the first time.
20160307 20:44:04< vultraz> ahh
20160307 20:44:05< zookeeper> it happens constantly to me, but i use that dialog a lot too
20160307 20:44:05< vultraz> ok
20160307 20:44:11< vultraz> so not the crash
20160307 20:57:33< vultraz> gfgtdf: did you find anything about making tunit_preview_pane use tcontainer?
20160307 20:57:55< gfgtdf> vultraz: hmm not exactly sure what you mean
20160307 20:58:10< gfgtdf> vultraz: i think tunit_preview_pane  shoudl inhertid from tcontainer
20160307 20:58:19< vultraz> [03:24:13]	gfgtdf	vultraz: y the more i think about it it makes sense. I'd try to copy code from te scrollbar_panel widget
20160307 20:58:40< gfgtdf> vultraz: (istead of from twidget becasue tcontainer already inhertis form twidget)
20160307 20:58:55< vultraz> hm....
20160307 20:59:33< gfgtdf> vultraz: y maybe i express myself unclear. tunit_preview_pane  shoudl inherit from tcontainer and to find out how to do this i reccomend to look at other widgets that so that like tscrollbar_container
20160307 21:00:10< vultraz> ok so class tunit_preview_pane : public tcotainer_ instead of class tunit_preview_pane : public tcontrol?
20160307 21:00:42< gfgtdf> vultraz: y thhats teh easy part
20160307 21:05:13< vultraz> ok
20160307 21:05:17< vultraz> I'll see what I can do : |
20160307 21:05:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20160307 21:06:00< vultraz> ahh, yes, you're right
20160307 21:06:04< vultraz> const config& child = cfg.child("grid");
20160307 21:06:10< vultraz> (from the scrollbar panel)
20160307 21:06:14< vultraz> exactly what I need
20160307 21:13:23< vultraz> er..
20160307 21:13:44< vultraz> gfgtdf: did you mean scrollbar panel or scrollbar container
20160307 21:14:25< gfgtdf> vultraz: scrollbar panel i think not wure thought
20160307 21:14:39< gfgtdf> vultraz: maybe there is also an easier exac´mply out there than scrollbar_poanels
20160307 21:14:42< gfgtdf> sure*
20160307 21:18:26< vultraz> hmmmm... ok
20160307 21:18:36< vultraz> scrollbar panels can take grids from wml
20160307 21:18:43< vultraz> I don't need to do that here
20160307 21:19:27< vultraz> I guess window_builder/ handles building the widget from WML?
20160307 21:20:55< gfgtdf> vultraz: scrollbar panels actualyl have 2 grids, one that defeines the 'viewport' with its scrollbars (defeined in the widget definition) and another one the for the 'content' defeined by teh dialog so they are a little more complicated then what you need
20160307 21:21:12< gfgtdf> vultraz: you ony need teht part that reds from the widget definition
20160307 21:21:39< vultraz> I assume that's in uh.. widget_defintion/?
20160307 21:21:44< vultraz> definition*
20160307 21:22:01< gfgtdf> vultraz: y i think so
20160307 21:22:30-!- Kwandulin_2 [~Miranda@p200300760F7D73CB49D2A469E7978ECD.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20160307 21:33:31-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 21:34:02< fendrin> vultraz: https://forums.wesnoth.org/viewtopic.php?f=9&t=12787&view=unread#unread
20160307 21:34:46< fendrin> I think that thread needs some attention. I guess the artist is waiting for Jetrel or another authority. 
20160307 21:35:01< vultraz> oh, yeah
20160307 21:35:06< vultraz> I was supposed to commit that
20160307 21:35:16< fendrin> But not only the finished work.
20160307 21:35:21< celticminstrel> What is it?
20160307 21:35:28< fendrin> animation for units
20160307 21:35:50< vultraz> im not actually sure which of the two to commit
20160307 21:35:56< vultraz> and i think the ne one needs work
20160307 21:36:09< fendrin> vultraz: runemaster.zip
20160307 21:36:20< vultraz> I did commit that
20160307 21:36:24< fendrin> cool
20160307 21:36:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 21:36:45< celticminstrel> Ooh, runemaster animations?
20160307 21:36:58< fendrin> celticminstrel: standing
20160307 21:37:13< shadowm> aquileia: You should tell mattsc to make them untranslatable or do it yourself then.
20160307 21:37:54< shadowm> vultraz: How goes the announcement?
20160307 21:38:21< vultraz> I can't do that right now
20160307 21:38:28< vultraz> But I've been meaning to ask, is there a template somewhere
20160307 21:38:32< shadowm> You've been doing coding stuff though.
20160307 21:38:33< vultraz> Or do I just copy the last one
20160307 21:38:49< shadowm> (Also, _you_ should be the one asking me "shadowm: How goes the announcement's screenshots?")
20160307 21:38:54< shadowm> Copy the last one, obviously.
20160307 21:39:20< vultraz> I've been doing coding stuff since you always wait a few days so I figured it doesn't matter
20160307 21:39:31< shadowm> You didn't read the wiki page.
20160307 21:39:42< shadowm> See why it's frustrating to work with you?
20160307 21:39:56< celticminstrel> I think you're supposed to post a preview in the secret mods forum, vultraz.
20160307 21:39:57-!- irker516 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20160307 21:39:57< irker516> wesnoth: gfgtdf wesnoth:master 21001bcb3201 / src/gui/widgets/ (tree_view_node.cpp tree_view_node.hpp): add ttree_view_node::unfolded_ memeber variable https://github.com/wesnoth/wesnoth/commit/21001bcb3201b46f4c4b15de1388d4bb843a2403
20160307 21:39:57< irker516> wesnoth: gfgtdf wesnoth:master d26c2b604323 / src/gui/ (3 files in 2 dirs): add unfolded=yes/no key in gui2 treeview node wml https://github.com/wesnoth/wesnoth/commit/d26c2b604323a5db2cd9de5ff0e843fd122f18da
20160307 21:39:58< irker516> wesnoth: gfgtdf wesnoth:master 33b8d66b3441 / src/gui/widgets/ (tree_view.cpp tree_view.hpp): implement right/left key handlers for treeviews https://github.com/wesnoth/wesnoth/commit/33b8d66b3441bf83f63d24e6a734cdcdf960ac5d
20160307 21:39:59< irker516> wesnoth: gfgtdf wesnoth:master c357d91a4383 / src/gui/widgets/ (tree_view.cpp tree_view.hpp tree_view_node.cpp tree_view_node.hpp): implement up/down key handlers for treeviews https://github.com/wesnoth/wesnoth/commit/c357d91a438310ff4e74d26a1fcff7b4249bd9da
20160307 21:40:00< irker516> wesnoth: gfgtdf wesnoth:master 90643a63365e / data/gui/default/window/campaign_dialog.cfg: have modifications in campaign dialog visible by default. https://github.com/wesnoth/wesnoth/commit/90643a63365ed72405f1136de9886b3347d7689f
20160307 21:40:12< vultraz> celticminstrel: I know
20160307 21:40:14< celticminstrel> gfgtdf: Yay! \o/
20160307 21:40:17< vultraz> I'll try to deal with it today
20160307 21:40:23< celticminstrel> I was thinking about doing that,actually.
20160307 21:40:24< shadowm> "vultraz: here have some helpful guide with detailed instructions"
20160307 21:40:29< celticminstrel> So thanks.
20160307 21:40:35< shadowm> "*ignores the guide, has someone else do stuff for him*"
20160307 21:40:58< shadowm> https://wiki.wesnoth.org/ReleasingWesnoth#The_real_announcement
20160307 21:41:03< shadowm> "Wait until the announcement conditions are met (waited at least 24 hours and OS X and Windows builds are ready OR 72 hours passed). "
20160307 21:41:09< celticminstrel> By "that" I specifically mean the key handlers for treeviews.
20160307 21:41:28< shadowm> If ancestral and loonycyborg had posted their builds within 5 hours after tagging, you'd have only 19 hours to get the announcement done.
20160307 21:41:44< vultraz> I thought ancestral was still pending
20160307 21:41:48< shadowm> I think. You know I don't know maths.
20160307 21:42:00< celticminstrel> I think he said he'd get to it today.
20160307 21:42:04< shadowm> You are again gambling on life, vultraz.
20160307 21:42:10< celticminstrel> Or was that yesterday.
20160307 21:42:12< gfgtdf> celticminstrel: y specailyl in the camapign dialog this was quite a regression in the way key up/down were handeled
20160307 21:42:38< gfgtdf> celticminstrel: another thigns that bothers me is how keys are handleed in 'list with search field' dialogs
20160307 21:42:45< celticminstrel> gfgtdf: Does right enter a selected tree node and left move to the parent of the current node?
20160307 21:42:55< shadowm> It doesn't matter whether the odds that they won't be immediately available are close to 100%, you do your work ASAP instead of anything else that isn't important for you or the project.
20160307 21:43:37< gfgtdf> celticminstrel: i think that the dialog shoudl automatically send arrow keys to be handled by the listbox and character keys to be handled  by teh serach bot.
20160307 21:43:54< celticminstrel> Makes sense.
20160307 21:43:54< vultraz> I'll deal with it today
20160307 21:43:56< gfgtdf> celticminstrel: 'right' only epands teh node without entering it
20160307 21:44:00< gfgtdf> expands
20160307 21:44:06< gfgtdf> celticminstrel: you can enter it my 'down'
20160307 21:44:10< celticminstrel> What if it's already expanded?
20160307 21:44:14< vultraz> Just not right now because it's 9 AM and I have an appointment in an hour
20160307 21:44:25< shadowm> That's not an excuse because 12 hours ago it wasn't 9 AM.
20160307 21:44:45< gfgtdf> celticminstrel: does nothign currently
20160307 21:44:50< gfgtdf> celticminstrel: maybe it shoudl enter it
20160307 21:44:56-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has joined #wesnoth-dev
20160307 21:45:03< gfgtdf> celticminstrel: if its already expanded
20160307 21:45:05< shadowm> You can't do it right now, but you still should've started sooner.
20160307 21:45:20< celticminstrel> Maybe, but I guess the current behaviour is fine. And left moves to the parent?
20160307 21:45:37< shadowm> And if you really aren't interested in doing community management stuff, get a replacement ASAP.
20160307 21:46:12< gfgtdf> celticminstrel: hmm no it only fold the currently selected node.
20160307 21:46:18< celticminstrel> I see.
20160307 21:46:24< gfgtdf> celticminstrel: can also be changed
20160307 21:46:48< loonycyborg> actually I should have started uploading tarballs sooner
20160307 21:46:56< gfgtdf> celticminstrel: but, specialyl since we dont have any dialogs taht really use treeview in that way i'm not that motivated to do it right now
20160307 21:46:57< loonycyborg> I have a lot better bandwithd than shadowm
20160307 21:47:06< celticminstrel> Fair enough.
20160307 21:47:31< shadowm> loonycyborg: The trick is that I've never uploaded the source tarballs from my desktop.
20160307 21:47:35< vultraz> look, alright, I'll deal with it right now
20160307 21:47:40< gfgtdf> celticminstrel: in the campaign treeview for example the 'modifications' node currently cannot be expanded to its also impossible to unfodl/fold it using keys
20160307 21:47:56< loonycyborg> I guess you can make them right on baldras too
20160307 21:47:58< gfgtdf> celticminstrel: but the campaigns dialog treeview isnt optimal anyway
20160307 21:48:06< shadowm> vultraz: No, you don't need to. If that's the only conclusion you can draw from what I said then there's a problem with you.
20160307 21:48:32< shadowm> I didn't say "DO IT IMMEDIATELY", I just said "why weren't you doing this much sooner instead of messing around with GUI2 as usual".
20160307 21:48:55< vultraz> That comes across as passive aggressively implying i should DO IT IMMEDIATELY 
20160307 21:49:13< shadowm> Nope, you should've started sooner.
20160307 21:49:22< loonycyborg> you misunderstood him, he just expessed his mild disappointment with your task prioretization
20160307 21:49:32< shadowm> Especially since it's at least a billion times easier than looking at GUI2's innards.
20160307 21:50:05< loonycyborg> Do it as soon as you can so nothing will weigh on you when you work on gui2
20160307 21:50:13< loonycyborg> that's the idea
20160307 21:50:16< shadowm> Exactly!
20160307 21:51:04< shadowm> And in the future, if someone hands you a guide, you are supposed to read and study it.
20160307 21:57:26-!- SpoOkyMagician [~chatzilla@2607:fcc8:be59:b00:d5fb:e8aa:3e47:6b94] has quit [Quit: later]
20160307 22:03:17-!- fendrin [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20160307 22:04:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 22:05:40-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20160307 22:11:42-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20160307 22:11:49< shadowm> vultraz: Why is the resolution combobox in Preferences → Display disabled when in full screen mode?
20160307 22:12:25< vultraz> Because we no longer support setting resolution when in fullscreen
20160307 22:13:06-!- travis-ci [~travis-ci@ec2-54-234-212-188.compute-1.amazonaws.com] has joined #wesnoth-dev
20160307 22:13:07< shadowm> Why?
20160307 22:13:07< travis-ci> wesnoth/wesnoth#8792 (master - 90643a6 : gfgtdf): The build was broken.
20160307 22:13:07< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114362404
20160307 22:13:07-!- travis-ci [~travis-ci@ec2-54-234-212-188.compute-1.amazonaws.com] has left #wesnoth-dev []
20160307 22:13:28< celticminstrel> I think you need to ask Aginor for that one.
20160307 22:13:29< vultraz> There's been extensive discussion here about it, but Aginor could give you the best summary
20160307 22:14:01< celticminstrel> But basically, fullscreen mode currently always uses your current screen resolution.
20160307 22:14:11< loonycyborg> it's the way things are moving
20160307 22:14:27< loonycyborg> I heard wayland won't let you change resolution via xrandr
20160307 22:15:13-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20160307 22:15:13-!- wedge010 is now known as wedge009
20160307 22:15:27< shadowm> Wel, xrandr uses the randr protocol which is an extension to X11. :p
20160307 22:15:43< shadowm> So obviously it doesn't belong in non-X11 land.
20160307 22:15:58< loonycyborg> For xwayland apps
20160307 22:16:20< shadowm> SDL 2.0.4 is the first version with (experimental) native Wayland support, isn't it?
20160307 22:16:30< irker516> wesnoth: gfgtdf wesnoth:master 1efd5c750e71 / data/gui/schema.cfg: Update schema.cfg https://github.com/wesnoth/wesnoth/commit/1efd5c750e716daa8cb7984b4ee474131f833d58
20160307 22:16:39< loonycyborg> iirc it's 2.0
20160307 22:16:45-!- louis94 [~~louis94@91.178.240.61] has joined #wesnoth-dev
20160307 22:16:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 22:17:01< loonycyborg> I recently installed fedora to try wayland
20160307 22:17:25< loonycyborg> running wesnoth with sdl-wayland there makes gnome session crash
20160307 22:17:34< vultraz> shadowm: why would you want to change resolution in fullscreen anyway
20160307 22:18:05< loonycyborg> I really hate when apps change resolution
20160307 22:18:07< shadowm> 2.0.2 adds Wayland support, 2.0.4 "enables it by default", according to /WhatsNew.txt.
20160307 22:18:28< loonycyborg> it tended to mess up desktop icons for earlier versions of kde
20160307 22:18:34< celticminstrel> I also hate when apps change resolution, for similar reasons.
20160307 22:18:35< shadowm> vultraz: Because applications sometimes choose the wrong resolution for any number of reasons (including driver issues).
20160307 22:18:35< loonycyborg> and even when it doesn't
20160307 22:18:41< loonycyborg> switching to other apps takes time
20160307 22:18:51-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20160307 22:18:51-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20160307 22:19:00< shadowm> vultraz: Also, for low-end GPUs, it's sometimes advantageous to use a lower resolution than the native resolution.
20160307 22:19:23< loonycyborg> in that case you can change desktop res before running the app
20160307 22:19:35< loonycyborg> or even run it in another VT
20160307 22:19:44< shadowm> Yes, because that's obviously not a chore.
20160307 22:20:14< shadowm> (Plus you're expecting me to know how to use a command line and/or assuming I'm using Linux.)
20160307 22:22:37< loonycyborg> any desktop can change resolution, it's mostly right click->screen settings or something like that
20160307 22:23:00< celticminstrel> No, it's not in the right-click menu on Macs.
20160307 22:23:07< celticminstrel> You need to go into System Preferences.
20160307 22:23:15< celticminstrel> Which is in the Apple menu, but still.
20160307 22:23:23< shadowm> loonycyborg: But you must do it by hand as opposed to having the application where you need it do it automatically.
20160307 22:23:30< celticminstrel> (I don't care about the change personally since I don't use fullscreen.)
20160307 22:24:31< shadowm> Yeah, developers don't care much about gameplay breaking when it does either since they don't play the game. :p
20160307 22:24:48< celticminstrel> I do play the game... sometimes...
20160307 22:24:48< loonycyborg> I play wesnoth in maximized window
20160307 22:25:00< loonycyborg> fullscreen would be absolutely unplayable
20160307 22:25:11< loonycyborg> because I also need to use irc
20160307 22:25:16< loonycyborg> and other stuff
20160307 22:25:38< shadowm> That's not the game being unplayable, that's IRC being unusable.
20160307 22:26:56< shadowm> If you must multi-task while playing the game, well, then obviously fullscreen isn't for you. But many people would rather focus on the game without any distractions.
20160307 22:27:20< loonycyborg> they can just make taskbar auto-hide
20160307 22:27:49< shadowm> So you are advocating for the removal of the fullscreen option or something? I'm confused.
20160307 22:29:16< vultraz> same as loonycyborg 
20160307 22:29:26< vultraz> I'm always tabbing out
20160307 22:29:33< loonycyborg> I wouldn't care about full screen
20160307 22:29:42< celticminstrel> The fullscreen thing could easily be changed back.
20160307 22:29:43< vultraz> just wesnoth though
20160307 22:29:48< vultraz> I always play terraria fullscreen
20160307 22:29:55< shadowm> My point is that it doesn't mean that other people wouldn't.
20160307 22:29:59< vultraz> celticminstrel: if so, I'll add a setting
20160307 22:30:08< loonycyborg> we were just supporting resolution change because it was easy enough
20160307 22:30:11< shadowm> I don't care about MP either, but I've never (seriously) proposed removing MP support.
20160307 22:30:16< vultraz> celticminstrel: Fullscreen/Windowed/Windowed Fullscreen
20160307 22:30:46< celticminstrel> vultraz: No, revert fullscrenn checkbox to old behaviour.
20160307 22:30:53< vultraz> no
20160307 22:30:55< celticminstrel> ^screen
20160307 22:31:03< loonycyborg> but I expect wayland to make it harder
20160307 22:31:24< celticminstrel> Then if you want the native fullscreen option still, add it to the combobox.
20160307 22:31:34< shadowm> Anyway, all I'm saying here is that someone, sooner or later, will ask about the removal of fullscreen resolution options.
20160307 22:31:53< vultraz> We'll see
20160307 22:31:53< shadowm> That someone might not understand all the technical jargon that Aginor will surely dump on me.
20160307 22:32:22< shadowm> It's you people's job to make sure decisions don't seem completely arbitrary and opaque to users.
20160307 22:32:23< gfgtdf> i still dotn see why we need the resolutojn option anyway fi we dont support it in fullscreen anymore, i mean what are teh advantage ofer resizing the window the normal way ?
20160307 22:32:44< shadowm> gfgtdf: Exact sizes.
20160307 22:32:58< gfgtdf> shadowm: when would you want to have exact sizes ??
20160307 22:33:07< shadowm> Because some people want to do that.
20160307 22:33:15< gfgtdf> shadowm: why?
20160307 22:33:24< shadowm> Because they like it that way.
20160307 22:33:34< shadowm> (Also, screenshots.)
20160307 22:33:45< gfgtdf> shadowm: i dotn really thign tehy notive a differetn of 10 pixels less or mpore
20160307 22:33:59< gfgtdf> notice*
20160307 22:34:02< shadowm> 1 pixel tends to make a difference when you are taking screenshots for a website.
20160307 22:34:43< gfgtdf> shadowm: so ity mainly there for screenshots ?
20160307 22:34:58< shadowm> No. It's there because some people want to have exact sizes.
20160307 22:35:32< zookeeper> uhh. yeah, obviously you need to be able to set a game window to a specific size.
20160307 22:35:32-!- ChipmunkV [~vova@d0017-2-88-172-31-68.fbx.proxad.net] has quit [Quit: ChipmunkV]
20160307 22:35:34< gfgtdf> shadowm: you know someone that wants that?
20160307 22:36:17< shadowm> gfgtdf: Are you trying to pull this "i dont need this so i dont see why we should have it" crap on me too? :|
20160307 22:36:25< zookeeper> gfgtdf, are you seriously suggesting that the resolution option shouldn't be there and people should just grab the corner of the window and resize it to the size they want?
20160307 22:36:36< gfgtdf> zookeeper: yes
20160307 22:36:39< zookeeper> holy crap
20160307 22:36:40< shadowm> I don't know anyone who needs to play Wesnoth on OS X.
20160307 22:36:41< zookeeper> that's nuts :p
20160307 22:36:46< gfgtdf> zookeeper: its much eaiser than oping teh preferences menu
20160307 22:36:57< shadowm> (Okay, actually, I do now. celmin.)
20160307 22:36:59< vultraz> gfgtdf has a point
20160307 22:37:10< vultraz> resolution is mostly useful in fullscreen mode
20160307 22:37:17< vultraz> and since we don't support that, it has minimal use
20160307 22:37:21< vultraz> (not *none*, minimal)
20160307 22:37:24< vultraz> shadowm: ancestral?
20160307 22:37:33< shadowm> Doesn't play Wesnoth nowadays, I don't think so.
20160307 22:37:36< zookeeper> this is absurd
20160307 22:37:39< shadowm> vultraz: I'm a journalist.
20160307 22:37:40< loonycyborg> we still would support fullscreen, no?
20160307 22:37:48< shadowm> I need screenshots of a very specific size.
20160307 22:37:58< shadowm> My desktop environment (Windows) doesn't display window sizes when resizing.
20160307 22:38:08< shadowm> I don't know how to use a command line and I'm on a deadline.
20160307 22:38:14< shadowm> What do you suggest I do?
20160307 22:38:30< vultraz> what kind of OS displays window sizes when resizing
20160307 22:38:34< gfgtdf> shadowm: i actualyl exptected you to know how to use a comandline 
20160307 22:38:47< gfgtdf> loonycyborg: y we shodul definitlexy keep fullscreen mode
20160307 22:38:47< celticminstrel> gfgtdf: He's giving a hypothetical situation here.
20160307 22:38:47< shadowm> vultraz: Answer my question instead of going off on an absurd tangent.
20160307 22:38:52< gfgtdf> loonycyborg: i mean i personalyl dont use it
20160307 22:39:00< celticminstrel> He's not actually a journalist, I don't think.
20160307 22:39:01< shadowm> gfgtdf: For the love of god.
20160307 22:39:09< vultraz> gfgtdf: he's speaking theoretically 
20160307 22:39:29< celticminstrel> Okay people, let's not just remove stuff willy-nilly, okay?
20160307 22:39:31< vultraz> shadowm: I said *minimal* use not *no* use
20160307 22:39:39< gfgtdf> celticminstrel: the theoretica solution is beeing a reease manager who doesnt know how to use commandlines?
20160307 22:39:41< vultraz> shadowm: therefor your case is valid
20160307 22:39:48< gfgtdf> situation*
20160307 22:39:50< loonycyborg> but anyway
20160307 22:40:01< celticminstrel> gfgtdf: The theoretical situation was a journalist who doesn't know how to use commandlines.
20160307 22:40:10< loonycyborg> aren't there ways to set window sizes to particular dimensions?
20160307 22:40:17< loonycyborg> that works for all resizable windows
20160307 22:40:23< celticminstrel> Not that I know of.
20160307 22:40:31< loonycyborg> I'm sure people who care about it would know a way
20160307 22:40:37< shadowm> -r
20160307 22:40:42< zookeeper> i don't know which trendy apple design school vultraz and gfgtdf went to, but... come on now :p that's a completely silly idea for a game. for most apps, yeah sure, but games are very much a special case in that regard.
20160307 22:40:44< shadowm> Command line switch.
20160307 22:40:54< shadowm> Oh, you meant for any application's windows.
20160307 22:40:55< loonycyborg> yes but it's wesnoth only
20160307 22:41:01< shadowm> Not on Windows, not built-in anyway.
20160307 22:41:10< celticminstrel> Not on Mac either, as far as I know.
20160307 22:41:44< vultraz> zookeeper: I've seen games that don't even support anything other than native fullscreen
20160307 22:41:48< vultraz> (recent games)
20160307 22:41:52< vultraz> I bought one
20160307 22:41:53< celticminstrel> That's terrible.
20160307 22:41:57< shadowm> I don't even know what it is on Linux and you'd probably think I would know.
20160307 22:42:07< celticminstrel> I will do my best to never buy it.
20160307 22:42:24< shadowm> *Linux+X11
20160307 22:42:45< celticminstrel> Mind you, when I'm playing on Windows I do tend to use fullscreen, though... because anything else I'm doing involves switching to Mac.
20160307 22:43:28< gfgtdf> zookeeper: how's that silly? You cannot tell me that as a player you'd notice a bug differene beween some pixels in size more or less. you notice a big different 
20160307 22:43:28< loonycyborg> I think in e 17 when resizing windows it showed exact dimensions in number
20160307 22:43:30< loonycyborg> s
20160307 22:44:08< gfgtdf> big*
20160307 22:44:25< shadowm> loonycyborg: Well, yes, KDE's window manager has a similar option, but you still must have some manual precision to achieve the desired results.
20160307 22:44:47< shadowm> Or use its complicated (but occasionally handy) special window settings mdoule.
20160307 22:45:16< gfgtdf> and i ia slo think that jornalists woudl usualyl have their won tools for changign window soze to an exact value
20160307 22:45:25< shadowm> Also, did you know? Not everyone has the required manual precision to make a window a specific size.
20160307 22:45:40 * shadowm glares at gfgtdf.
20160307 22:45:50< gfgtdf> i mean most of the simple games like solirair  usualyl don't have a  resolution combobox
20160307 22:45:51< loonycyborg> I have the manual precision, but I cba :P
20160307 22:46:05< gfgtdf> solitair
20160307 22:46:14< shadowm> Solitaire.
20160307 22:46:30< vultraz> heh
20160307 22:46:31< vultraz> I'm out
20160307 22:46:39< shadowm> Yeah, I am too.
20160307 22:46:44< vultraz> gonna be late for my appointment if I stick around 
20160307 22:46:57< vultraz> hopefully I'll finish the unit preview pane widget later
20160307 22:47:05< vultraz> and get the announcement draft up
20160307 22:47:10< shadowm> vultraz: NO.
20160307 22:47:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 22:47:15< shadowm> You get the announcement draft up _first_
20160307 22:47:28< vultraz> yes
20160307 22:48:45< shadowm> The most important UI design skill is being able to put on someone else's shoes.
20160307 22:48:56-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 22:49:02< shadowm> And that's all there really is to say on the matter.
20160307 22:49:32< shadowm> "but shadowm how is it any different from putting on my own shoes in the morning" No.
20160307 22:50:47< gfgtdf> vultraz: another issue: when i maximize the window, restart the game adn teh unmaximize the window, it is too big, it shodul have teh size that it had in before maximizing i'd say
20160307 22:50:55-!- travis-ci [~travis-ci@ec2-54-234-212-188.compute-1.amazonaws.com] has joined #wesnoth-dev
20160307 22:50:56< travis-ci> wesnoth/wesnoth#8793 (master - 1efd5c7 : gfgtdf): The build was fixed.
20160307 22:50:56< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/114374902
20160307 22:50:56-!- travis-ci [~travis-ci@ec2-54-234-212-188.compute-1.amazonaws.com] has left #wesnoth-dev []
20160307 22:51:29< gfgtdf> vultraz: the problem is that the new big size is usually to beig to fit on the screen because it is basicalyl the size of the maximized window
20160307 22:54:00-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20160307 22:57:29< loonycyborg> oh yes I know the real reason for removing support of changing resolution
20160307 22:57:51< loonycyborg> because on modern screens all resolution except screen native look like crap
20160307 22:57:53-!- Nobun [~nobun@host208-90-dynamic.50-82-r.retail.telecomitalia.it] has joined #wesnoth-dev
20160307 22:58:52< Nobun> Hi... starting from tomorrow I will keep my pc logged to #wesnoth-irc for any report regarding wmlxgettext
20160307 22:59:16< celticminstrel> Oh hi.
20160307 22:59:22< Nobun> hi :)
20160307 22:59:28< celticminstrel> I was just mentioning it a little while ago.
20160307 22:59:45< Nobun> yeah... I readed a message from Elvish Hunter so why I comed here :)
20160307 22:59:52< celticminstrel> Ah.
20160307 23:00:11< Nobun> can you tell me more about the bug on lua script that occurs on perl version=
20160307 23:00:39< Nobun> so I will see if the same bug happen in my script and, if so, how to solve it
20160307 23:00:46< celticminstrel> Nobun: What's with this [-tag] thing your documentation mentioned?
20160307 23:01:05< celticminstrel> I have never heard of such a thing before.
20160307 23:01:20< Nobun> well it was a "safety measure" since I wasn't sure about WML
20160307 23:01:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20160307 23:01:27< celticminstrel> I see...
20160307 23:01:36< Nobun> I am sure that [+tag] exists
20160307 23:01:42< celticminstrel> I'm pretty sure it doesn't exist, but I guess it doesn't break anything to handle something that doesn't exist.
20160307 23:01:44< Nobun> (like [+units]
20160307 23:01:48< celticminstrel> [+tag] does indeed exist.
20160307 23:01:52< Nobun> yep
20160307 23:02:03< celticminstrel> [+tag] is an extension to the previous tag with the same name.
20160307 23:02:13< Nobun> I added -tag becouse, as +tag exist, maybe could exist -tag (in the moment or in a future)
20160307 23:02:25< Nobun> however the - sign can be easily removed
20160307 23:02:27< celticminstrel> I kinda doubt it, but whatever.
20160307 23:02:30< celticminstrel> So about Lua, uhh...
20160307 23:02:42< celticminstrel> The Perl version is actually very, very bad at handling strings in Lua.
20160307 23:02:58< celticminstrel> It can handle the most basic form of string only, _ "single-quoted text"
20160307 23:03:34< celticminstrel> Theoretically it can handle multiline quoted strings, but I think it didn't handle them correctly.
20160307 23:03:42< celticminstrel> It can't handle single-quoted strings.
20160307 23:04:20< Nobun> can you make me an example?
20160307 23:04:37< celticminstrel> I have an example, but I don't recall specifically which strings failed in Perl.
20160307 23:04:43< Nobun> (you could use pastebin or something similar for sample code)
20160307 23:05:04< Nobun> if you remember at least one...
20160307 23:05:11< celticminstrel> It's a very artificial example.
20160307 23:05:43< loonycyborg> there's another issue with wesnoth+wayland: it won't get to have any window decorations
20160307 23:05:52< celticminstrel> And doesn't cover the issue of Lua functions.
20160307 23:06:20< loonycyborg> it's currently a matter of dispute: what better: client-side decorations or server-side decorations?
20160307 23:06:21< Nobun> yep... the lua functions problem is a problem also in my script :(
20160307 23:06:38< loonycyborg> perhaps wesnoth will even have to draw own decorations :P
20160307 23:10:46< celticminstrel> I'm testing all the cases I made, so give me a few minutes.
20160307 23:11:13< Nobun> don't worry... I will wait patiently :)
20160307 23:11:23< shadowm> loonycyborg: We all know what the answer to that is, don't be silly.
20160307 23:11:29< Nobun> thank for the hard work you are making :)
20160307 23:13:07-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20160307 23:15:35< zookeeper> vultraz, what games would those be?
20160307 23:16:49< zookeeper> gfgtdf, it's silly because people have a habit of running games in a specific resolution, want to make screenshots with a specific resolution, and need to tell you what their resolution is when reporting bugs or whatever.
20160307 23:18:14< loonycyborg> shadowm: but I fear that's exactly how it will end up. Perhaps there will be a third party decoration lib though
20160307 23:19:25< shadowm> loonycyborg: To elaborate, there _is_ some value to client-side decorations as an _opt-in_ thing (as it is on Windows, for example), but it should by no means be a requirement unless you want to have desktop environments become even more fragmented and unusable than they already are thanks to the coexistence of applications written for different toolkits.
20160307 23:20:06< loonycyborg> you have a point here but
20160307 23:20:12< Nobun> loonycyborg: later (or tomorrow) I'd like to ask you some details about the tests you did on my wmlxgettext python version
20160307 23:20:15< celticminstrel> Nobun: I have two general suggestions about the Lua side.
20160307 23:20:39< celticminstrel> Consider making the # in special comments optional - I think the Perl version doesn't require it.
20160307 23:20:40< Nobun> celticminstrel: feel free to say them :)
20160307 23:20:44< loonycyborg> Nobun: I'll do more tests when the issue with escaping " will be fixed
20160307 23:21:01< loonycyborg> shadowm: they could go other route
20160307 23:21:13< loonycyborg> like always expect csd
20160307 23:21:25< Nobun> tell me more about exaping " problem
20160307 23:21:30< loonycyborg> but hope that gui toolkit is configured to use system theme
20160307 23:21:38< celticminstrel> Also, from the perspective of the Lua code, the textdomain is set only once in the file, and adding a textdomain comment will do nothing to change that.
20160307 23:21:42< Nobun> I am not sure I understood what happened
20160307 23:21:45< loonycyborg> perhaps they'll add freedesktop standard or something
20160307 23:21:58< loonycyborg> for toolkits to get proper theme
20160307 23:22:06< shadowm> loonycyborg: Toolkits cooperating with each other is a bit of a pipe dream, especially given Gtk 3's psychotic tendencies.
20160307 23:22:18< loonycyborg> it's not that bad
20160307 23:22:43< loonycyborg> in fact it's possible to make kde and gnome apps look very close
20160307 23:22:58< loonycyborg> using appropriate theme engines
20160307 23:23:10< loonycyborg> or themes with both kde and gtk support
20160307 23:23:13< loonycyborg> like qtcurve
20160307 23:23:25< loonycyborg> perhaps they'll generalize this way to decorations too
20160307 23:23:35< shadowm> There is no QtCurve for Gtk 3 because Gtk 3.
20160307 23:24:06< celticminstrel> Nobun: http://pastebin.com/C5BWsKCy
20160307 23:24:08< shadowm> Oxygen-gtk 3 was completely broken circa Gtk 3.10 or so because they decided to remove some features it needs.
20160307 23:25:13< Nobun> loonycyborg: the issue with escaping "  you mentioned what happened exactly? what is the WML/Lua string that worked bad? what is the desired result? what is instead the result occurred on my script? (with this details I can understand where is the problem and how to fix it)
20160307 23:25:27< Nobun> celticminstrel: thank... now I will read
20160307 23:25:50< loonycyborg> Nobun: msgcat was barfing because of unescaped ""
20160307 23:26:01< loonycyborg> that was in pastebins I shared
20160307 23:26:12< shadowm> 20:21:39 <celticminstrel> Also, from the perspective of the Lua code, the textdomain is set only once in the file
20160307 23:26:13< celticminstrel> Nobun: It'd also be great if it could parse Lua scope to determine which textdomain is used, but that may be more work than it's worth. However, one thing that might be simpler towards that end would be supporting a one-time textdomain override.
20160307 23:26:18< shadowm> celticminstrel: Huh?
20160307 23:26:30< celticminstrel> shadowm: Sorry, I phrased that badly, didn't I.
20160307 23:26:51< shadowm> Oh, but it seems you do realize you can instantiate as many textdomain query objects as you want, so never mind.
20160307 23:27:00< celticminstrel> I mean that once you define _ to be textdomain X, it's textdomain X no matter what... unless of course you later redefine it, which would just be crazy.
20160307 23:27:05< celticminstrel> Yeah, I realize that.
20160307 23:27:12< Nobun> celticminstrel: for textdomain, it works exactly as perl (since it is a parser based on regular expression like the perl one)
20160307 23:27:29< Nobun> so it requires -- # textdomain <domain> to switch domain
20160307 23:27:46< Nobun> -- #textdomain <domain> (sorry)
20160307 23:28:05< celticminstrel> Nobun: That's the problem. I'm thinking perhaps it should not recognize those in Lua files.
20160307 23:28:23< celticminstrel> Nobun: Because, in WML, that line actually changes the active textdomain, but in Lua, it does nothing whatsoever.
20160307 23:28:39< celticminstrel> Nobun: It can, admittedly, be used with care to get correct results.
20160307 23:29:22< Nobun> celticminstrel; this is why in lua you use the _ =  textdomain_function
20160307 23:29:25< shadowm> The old Perl wmlxgettext uses that trick only because silene discovered it was a thing (nobody at the time could read wmlxgettext's very "concise" Perl).
20160307 23:29:47< shadowm> (It was improved later by AI0867. Much later.)
20160307 23:29:52< Nobun> but putting the comment before let the parser to understand it
20160307 23:30:00< celticminstrel> Nobun: Anyway, about the one-time textdomain override (setting a different textdomain for a single string) - it would look something like this: wesnoth.textdomain(<domain>)(<string>) - no _ is used.
20160307 23:30:03< Nobun> (the parser of wmlxgettext)
20160307 23:30:27< celticminstrel> Also Nobun, as far as I can tell, the current Perl parser actually parses your "_ = wesnoth.textdomain <domain>" line.
20160307 23:30:53< Nobun> perhaps also mine... I think
20160307 23:31:12< Nobun> since the domain is currently given by the comment
20160307 23:31:42< Nobun> but I could change the rule... to do this I need to know how the textdomain is changed in lua code on wesnoth
20160307 23:32:00< celticminstrel> Nobun: I'm guessing you do understand how bracketed strings work in Lua?
20160307 23:32:08< Nobun> yep
20160307 23:32:16< celticminstrel> 'kay, just checking.
20160307 23:32:17< Nobun> more or less
20160307 23:32:43< celticminstrel> Making it work absolutely correctly in Lua would be quite a lot of work.
20160307 23:33:01< Nobun> well... I think "absoultely correctly" is impossible
20160307 23:33:24< celticminstrel> It would mean tracking the variable it was assigned to, tracking nested Lua scopes, etc. And yeah, even that wouldn't be absolutely correct since the variable could be reassigned to at some point by the code.
20160307 23:33:33< Nobun> since it would require a real parser, wich is out-of-my-ability
20160307 23:33:44< celticminstrel> So obviously you need to at least assume some kind of sane organization on the part of the coder.
20160307 23:34:37< Nobun> at the moment my parser works on lua like the perl wmlxgettext with two differences
20160307 23:34:42< celticminstrel> To match a textdomain change in Lua, I think I would recommend something that matches the full variable assignment.
20160307 23:35:01< Nobun> I am afraid that could not work
20160307 23:35:07< Nobun> I try to explain why
20160307 23:35:22< celticminstrel> I'll give you a paste to show what I mean, but go ahead and try to explain why you don't think it can work, too.
20160307 23:35:33< Nobun> the idea around the combination of -- # texdomain comment and the following line _ = function
20160307 23:36:10< Nobun> is "let me identify when a string is translatable or not", thinking also about wmlxgettext (perl version)
20160307 23:37:16< Nobun> so why... even if the actual textdomain change is performed by the function assigned to _, the wmlxgettext parser (both mine and perl one) will behave a little different
20160307 23:37:19< Nobun> so:
20160307 23:37:47< Nobun> when _ is found it means "ok... this is translatable... if _ is not present the string is not translatable"
20160307 23:38:17< Nobun> this rule is somewhat require to distinguish in a unique way the translatable strings (regadless of textdomain, in this moment)
20160307 23:38:51< Nobun> take mind that wmlxgettext use regexpes... so the parsing is somehow limited
20160307 23:39:40< Nobun> of you want temporary regexpes you could use # -- textdomain <something> and then reassinging the _ function
20160307 23:40:03< Nobun> and, then come back to the previous textdomain in the same way.
20160307 23:40:35< Nobun> Unluckly it is not confortable, but honestly I cannot see another solution... (perhaps I am not able to see another solution)
20160307 23:41:21< Nobun> the actual difference is that my script recognize both _ "string", and _ ' string' and _ [==[ string ]==] 
20160307 23:42:08< Nobun> (my script only accepts brackets string marked with exactly two equal signs)
20160307 23:42:32< Nobun> (sorry for the long reply, celticminstrel :) )
20160307 23:43:08< celticminstrel> Nobun: It's no harder to accept arbitrary numbers of equals signs than it is to accept both double and single quotes with a single pattern.
20160307 23:44:27< Nobun> they are actually a lot of different pattnerns... a single pattern is not enough
20160307 23:44:44< celticminstrel> You can match both single and double quotes with one pattern.
20160307 23:45:04< Nobun> not properly if you consider escape quotes
20160307 23:45:34< celticminstrel> Hmm, you may be right, I suppose.
20160307 23:45:37< Nobun> since the '' quote escape \' while " escape \"
20160307 23:45:44< celticminstrel> However that issue doesn't apply to the bracketed strings.
20160307 23:45:59< celticminstrel> You can't escape the closer of the bracketed strings.
20160307 23:46:23< Nobun> the problem there is not the escape, infact
20160307 23:46:27< celticminstrel> If you write ]==\] the string doesn't close, but you also get a literal \
20160307 23:46:57< Nobun> yeah you are right
20160307 23:47:24< Nobun> but, as I mentioned, I tried different solutions to support bracket strings on my parser
20160307 23:47:40< Nobun> the problem there was another
20160307 23:48:03< Nobun> when you use bracket strings you don't have the ]==] in the same line as the [==[ started
20160307 23:48:14< Nobun> (now I am speaking about ==)
20160307 23:48:29< celticminstrel> Well, you can, but yeah, normally you don't.
20160307 23:48:34< Nobun> yep
20160307 23:48:55< Nobun> but let me explain taking using == as an example
20160307 23:49:17< Nobun> after that I will try to explain where is the problem on supporting aribitrary equals
20160307 23:49:20-!- louis94 [~~louis94@91.178.240.61] has quit [Ping timeout: 250 seconds]
20160307 23:49:39< Nobun> well... taking the multiline example with two equals
20160307 23:50:14< Nobun> if I know that two equals are expected I can be sure to identify in an unique way the line where the string ends
20160307 23:50:34< Nobun> on my script, infact, any line is parsed atomisticly
20160307 23:50:36< Nobun> so
20160307 23:50:47< Nobun> I cannot use a regexp for a multiline string
20160307 23:51:05< celticminstrel> I realize.
20160307 23:51:29< celticminstrel> But bracketed strings aren't the only ones with a multiline case, you know.
20160307 23:51:36< Nobun> yeah
20160307 23:51:53< Nobun> but it wasn't a problem in " and ' cases
20160307 23:52:05< celticminstrel> Why not?
20160307 23:52:23< celticminstrel> For any multiline string, don't you need to store how the string should be ended?
20160307 23:52:29< Nobun> becouse I know that in a multiline ' string the end-marked is an unescaped '
20160307 23:52:37< Nobun> the same for " (unescaped ")
20160307 23:52:53< celticminstrel> The same for [[, and [=[, and [==[, and [===[.
20160307 23:53:11< celticminstrel> Except that escaping is irrelevant.
20160307 23:53:50< Nobun> you could be true... but if you see the sourcecode you could notice why it is not so
20160307 23:54:18< celticminstrel> Feel free to link me to the specific relevant line number on your repo.
20160307 23:54:20< Nobun> the lua " and ' quoters required 4 states in total (2 states for " and two states for ')
20160307 23:55:05< Nobun> the regexp used by lua multiline string [==[ ]==] required 4 states alone
20160307 23:55:34< Nobun> celticminstrel: see my sourcedoc documentation, chapter 4,
20160307 23:55:47< Nobun> now I will point you the paragraphs
20160307 23:56:01< Nobun> the regexpes used are instead explained on chapter 6
20160307 23:57:28< Nobun> waching directly the source could be not enough since the regexp are located on regexp variable, that could be completely disorienting without a global sight of the context
20160307 23:58:35< Nobun> See paragraph 4.2.2.2 of my sourcedoc (Capture String: Lua States)
--- Log closed Tue Mar 08 00:00:06 2016