--- Log opened Sat Sep 19 00:00:43 2015
20150919 00:02:28< celtiminstrel> Oh, attribute_value...
20150919 00:04:49< celtiminstrel> Never mind.
20150919 00:07:17-!- MurifoX [bd058569@gateway/web/freenode/ip.189.5.133.105] has joined #wesnoth-dev
20150919 00:09:26< celtiminstrel> Okay, so $other_unit is not being expanded.
20150919 00:13:18< celtiminstrel> I changed one of them to [filter]level=$this_unit.level[/filter].
20150919 00:13:37< celtiminstrel> (For the leadership definition.)
20150919 00:13:52< celtiminstrel> Shouldn't that mean that any unit next to the leadership unit receives the benefit?
20150919 00:14:34-!- Jozrael [~Jozrael@72.43.14.98] has quit [Read error: Connection reset by peer]
20150919 00:14:52-!- MurifoX [bd058569@gateway/web/freenode/ip.189.5.133.105] has quit [Quit: Page closed]
20150919 00:16:49-!- Jozrael [~Jozrael@rrcs-72-43-14-98.nys.biz.rr.com] has joined #wesnoth-dev
20150919 00:19:34-!- Jetrel [~Jetrel@c-73-228-139-39.hsd1.mn.comcast.net] has joined #wesnoth-dev
20150919 00:21:01-!- Jozrael [~Jozrael@rrcs-72-43-14-98.nys.biz.rr.com] has quit [Ping timeout: 246 seconds]
20150919 00:29:44 * celtiminstrel suspects that commit c9d05ff might have broken SUFs that use $this_unit...
20150919 00:35:44-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150919 00:45:16-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Read error: Connection reset by peer]
20150919 00:51:21-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150919 00:51:35-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has joined #wesnoth-dev
20150919 01:10:08< mattsc> Hmm…  I just upgraded to Xcode 7 and Wesnoth still compiles and runs.  First time that’s happened after I upgraded to a new major version.
20150919 01:10:44< mattsc> Well, almost.  I now get a gazillion “Unused typedef 'boost_static_assert_typedef_238’” warnings  (with a variety of numbers at the end).
20150919 01:11:44< mattsc> So I had to turn -Werror=unused off for now.
20150919 01:20:56-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150919 01:20:57-!- louis94 [~~louis94@109.129.229.117] has quit [Quit: Konversation terminated!]
20150919 01:35:11< celtiminstrel> I don't really have a problem with that.
20150919 01:35:24< celtiminstrel> Except that it means I'll have to rebuild the entire project once it's committed. >_>
20150919 01:39:26-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20150919 01:48:59< celtiminstrel> I'm going to try doing a testcase for $this_unit.
20150919 01:49:45-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150919 01:50:41< celtiminstrel> I would expect something like hitpoints=$($this_unit.max_hitpoints / 2), but it looks like hitpoints is not a valid key in SUF.
20150919 01:50:50< celtiminstrel> Why not?
20150919 01:50:57< celtiminstrel> Because it falls under [filter_wml]?
20150919 01:51:49< shadowm> [filter_wml] does value matches on the serialized config, that's why it'll work in [filter_wml].
20150919 01:52:09< celtiminstrel> Where "it" is... $this_unit?
20150919 01:52:12< celtiminstrel> Or my example?
20150919 01:52:35< shadowm> "It" being "trying to match any field not directly supported in SUFs".
20150919 01:52:41< celtiminstrel> Oh, I missed the part where I was going to say that I expect the example to be typical.
20150919 01:53:04< shadowm> This page lists the attributes and tags supported in a SUF: http://wiki.wesnoth.org/StandardUnitFilter
20150919 01:53:13< celtiminstrel> Yes, I have the page open.
20150919 01:54:23< mattsc> celtiminstrel: I’d rather not have 108 warnings pop up every time I compile.
20150919 01:55:09< celtiminstrel> I think $this_unit probably already works in [filter_wml]...
20150919 01:59:23< mattsc> celtiminstrel: I remember using this_unit in my campaigns, but I just checked and it’s all either with [modify_unit] or formula= , so that doesn’t help.
20150919 02:00:37< shadowm> The same page in question says $this_unit can be used there.
20150919 02:00:41-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150919 02:00:44< celtiminstrel> Yes.
20150919 02:09:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 250 seconds]
20150919 02:27:36-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150919 02:27:43< celtiminstrel> These test cases both fail: http://pastebin.com/yGBTeC2M
20150919 02:27:54< celtiminstrel> Do they look like they're coded correctly?
20150919 02:28:35< celtiminstrel> Either $this_unit is broken in SUFs, or I made a mistake in the test cases.
20150919 02:30:44< celtiminstrel> mattsc: I don't see why it would work with formula either, though... but formula variables should work, I guess...
20150919 02:31:06< mattsc> celtiminstrel: for the compiling to work, I actually need to remove -Werrer=unused _and_ add -Wno-unused-local-typedefs to the C++ compiler flags.  Is that okay with you?
20150919 02:31:23< mattsc> celtiminstrel: sorry, had this mostly typed already when you asked the other question ...
20150919 02:31:26< celtiminstrel> Sure.
20150919 02:31:56-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Quit: wedge009]
20150919 02:32:07< mattsc> I don’t know what you mean with it should work with formula, but formula variables should work …
20150919 02:32:21< mattsc> *should not work
20150919 02:32:30< celtiminstrel> I mean if you put "$this_unit" in your formula, does it work?
20150919 02:32:52< mattsc> I use things like this:
20150919 02:32:54< mattsc> formula = '$this_unit.moves > 0'
20150919 02:33:00< mattsc> all over the place and yes, that works
20150919 02:33:02< celtiminstrel> Hmm.
20150919 02:33:37< celtiminstrel> By "formula variables" I mean that formula="moves > 0" should theoretically work too. I think.
20150919 02:33:38< mattsc> I do that both in WML and in Lua (AI) code.
20150919 02:33:46< mattsc> Oh, I see.
20150919 02:34:18< mattsc> As I side note, 
20150919 02:34:40< mattsc> I actually do not use it in AI code any more, becaise it is much slower than just getting all units and then running a loop over them.
20150919 02:37:04< celtiminstrel> I'll try a test case with formula= too...
20150919 02:38:09< celtiminstrel> But can you see anything wrong with the test cases I posted?
20150919 02:39:59< celtiminstrel> ...oh. I should've used max_hitpoints instead of hitpoints in one place, at least.
20150919 02:41:11< celtiminstrel> As written the filter in the first scenario can never pass.
20150919 02:43:05< celtiminstrel> formula="$this_unit.moves < $this_unit.max_moves" yields a formula error "expected another token", which is about what I'd expect if the variables expand to nothing.
20150919 02:43:52< celtiminstrel> But the [filter_wml] test does pass now that I've fixed the condition on it.
20150919 02:44:14< celtiminstrel> So, in other words, unless there are more errors in my test cases, $this_unit currently only works within [filter_wml].
20150919 02:46:26< mattsc> I looked at them, I didn’t see anything wrong, but yes, you’re right about the max_hitpoints
20150919 02:46:54< celtiminstrel> I changed the second one to use max_hitpoints as well. Shouldn't matter in that case, but makes more sense to me.
20150919 02:46:54< mattsc> sounds good then
20150919 02:47:30< mattsc> ok
20150919 02:49:16< mattsc> Well, I have to be off again (as I said, I’m traveling).  I’ll check logs tomorrow and commit the warning flag changes sometime.
20150919 02:51:55-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20150919 02:57:59-!- oldlaptop [~quassel@192.183.51.75] has quit [Ping timeout: 240 seconds]
20150919 03:00:27-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has quit [Ping timeout: 244 seconds]
20150919 03:11:51-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 252 seconds]
20150919 03:13:48-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150919 03:58:42 * celtiminstrel suspects vconfig::child might have slightly unexpected results in the presence of [insert_tag].
20150919 04:20:22-!- ancestral [~ancestral@97-116-185-165.mpls.qwest.net] has joined #wesnoth-dev
20150919 04:20:46-!- irker316 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150919 04:20:46< irker316> wesnoth: Celtic Minstrel wesnoth:master edce9ac64ae4 / run_wml_tests: run_wml_tests: Add some fallback cases and a few more options http://git.io/vnOpF
20150919 04:20:48< irker316> wesnoth: Celtic Minstrel wesnoth:master a5184181df15 / / (5 files in 3 dirs): Fix SUF $this_unit only working within [filter_wml] http://git.io/vnOpb
20150919 04:20:57< celtiminstrel> Should that be pushed to 1.12? ^
20150919 04:21:11< celtiminstrel> Dunno if it'd be easy/hard.
20150919 04:21:27-!- Shackra [~Jorge@186.177.2.148] has quit [Ping timeout: 240 seconds]
20150919 04:24:12< elias> yay, units.wesnoth.org generation worked two weeks in a row... I think it's finally getting somewhat robust
20150919 04:24:23< celtiminstrel> Yay!
20150919 04:24:40< celtiminstrel> Maybe I'll see if I can get it to work locally again later.
20150919 04:24:54< elias> are you who wanted to run it in Windows?
20150919 04:24:58< celtiminstrel> Mac.
20150919 04:25:01< elias> hm
20150919 04:25:10< celtiminstrel> That was years ago, I think.
20150919 04:25:19< elias> well, I didn't change anything about running it local
20150919 04:25:26< elias> so it's stll hard at best :/
20150919 04:26:33< elias> the server runs data/tools/unit_tree/update-wmlunits
20150919 04:26:49< elias> so that's basically the starting point
20150919 04:27:44< elias> not sure why that part is in bash and not python... moving that into the main python script would probably be a good idea
20150919 04:28:34< celtiminstrel> I dunno.
20150919 04:28:42< shadowm> 01:20:59 <celtiminstrel> Should that be pushed to 1.12? ^
20150919 04:28:53< shadowm> No.
20150919 04:28:57< celtiminstrel> 'kay
20150919 04:29:11< celtiminstrel> Do you have time for PR492 yet?
20150919 04:29:56< shadowm> Possibly.
20150919 04:30:36< shadowm> ~30 mins.
20150919 04:33:35< ancestral> elias: If I have time one day I’d love to go through and copy the advancement tree parsing
20150919 04:34:12< ancestral> elias: Have I showed you my latest iteration? http://wesnoth.io/
20150919 04:34:37< elias> ohh, nice
20150919 04:34:51< ancestral> (It’s only the mainline stuff, it doesn’t do add-ons, but eventually)
20150919 04:35:14< ancestral> The terrain table columns are sortable, btw
20150919 04:35:26< elias> hard to say how important the add-ons are anyway... and they were about 90% of the work *sigh*
20150919 04:35:32< ancestral> I bet!
20150919 04:35:44-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20150919 04:35:45< ancestral> Yeah, the advancement trees are my bane right now, with missing icons
20150919 04:35:46< elias> mm, I see
20150919 04:35:47< celtiminstrel> Why's that?
20150919 04:35:54< celtiminstrel> About them being most of the work.
20150919 04:36:05< fabi> hello
20150919 04:36:05< ancestral> Broken WML?
20150919 04:36:26< elias> anyway, I have no plans on doing much work on units.wesnoth.org myself
20150919 04:36:52< elias> the unit reports are your CSS from... dunno, 5 or more years ago
20150919 04:37:04< ancestral> Yes, I saw that, I was humbled :-)
20150919 04:37:07< fabi> elias: Haven't you been the maintainer?
20150919 04:37:20< ancestral> I have to retool my scripts that generate my pages. It’s an embarrassment how I got what I got.
20150919 04:37:34< elias> fabi: yes, that's why I keep going back every half year and spend a few days getting it running again :P
20150919 04:37:57< elias> ancestral: heh, same with mine... so many things I'd have to do differently to do them properly
20150919 04:38:12< ancestral> celtiminstrel: I’m guessing bad WML in the addons cause problems parsing and pulling out unit data
20150919 04:38:16< elias> partly because stuff changed, e.g. there was no working dependencies list for addons
20150919 04:38:36< ancestral> I know when I posted my own era sometimes units didn’t show if I omitted some data the parser wanted
20150919 04:38:49< celtiminstrel> I seem to recall noticing recently (before elias's latest changes) that some units were listed for add-ons that they didn't exist in.
20150919 04:39:00< elias> now that there is, wouldn't have to duplicate all era-of-myths units in every single addon using them... making the output probably 75% smaller :P
20150919 04:39:30< celtiminstrel> Ah, perhaps that's why...
20150919 04:40:43< ancestral> elias: Well, one of these days, if I get around to updating my stuff and working with addons, I’ll seek your input, but I can’t say it’ll happen soon.
20150919 04:40:43< elias> the script parses the entire WML and includes all [unit] it encounters - it can't know if 1 or all EoM units are used in an era/campaign
20150919 04:40:44< ancestral>  Your scripts are doing more than anything anyone else has atm
20150919 04:40:47< elias> well, in theory it could know...
20150919 04:40:56< elias> it just doesn't parse those parts
20150919 04:41:57< ancestral> (fabi: my less featured http://wesnoth.io)
20150919 04:43:05< fabi> ancestral: Yeah, I know your Beastury and I like it.
20150919 04:44:07< ancestral> Fortunately (er, at least for me I guess?) there have been no default era unit stat changes in the last 1½ years
20150919 04:44:33< celtiminstrel> Didn't one of the dwarves change recently?
20150919 04:44:43< fabi> Well, Wesnoth seems pretty shut down recently.
20150919 04:44:56< fabi> But I promise this won't last for long.
20150919 04:45:24< fabi> I bet half of the former coders will come back at some time.
20150919 04:45:55< ancestral> Good stuff is happening thanks to folks like celtiminstrel :)
20150919 04:46:04< celtiminstrel> And Aginor.
20150919 04:46:16< ancestral> Yes, especially!
20150919 04:47:22< fabi> Yes.
20150919 04:48:09< fabi> Let's keep up some good mood in here.
20150919 04:48:39< fabi> I pretty much feel that 1.12 has some features that will lead to a lot nice addons durings the times.
20150919 04:48:54< celtiminstrel> I have another campaign in progress.
20150919 04:50:34< fabi> I thought about another multiplayer game mode.
20150919 04:50:48< fabi> One that allows every turn to happen simultaniously.
20150919 04:51:04< celtiminstrel> That sounds like a nightmare.
20150919 04:51:05< fabi> Thus you can play it with many players.
20150919 04:51:38< Aginor> celtiminstrel: thanks :)
20150919 04:51:38< fabi> It will be quite different to current wesnoth in many ways.
20150919 04:52:24< ancestral> fabi: Make it happen I will be so excited
20150919 04:52:30< ancestral> aeth: ^
20150919 04:53:18< celtiminstrel> Oh, I forgot the tail on the member variable in that commit... oh well. >_>
20150919 04:53:24< fabi> ancestral: Well, it will be certainly off balance. I bet that change will kill multiplayer. Would you volounteer to balance it?
20150919 04:53:48< ancestral> I get the feeling this is not a one man job ;-)
20150919 04:54:03< fabi> ancestral: Yes. Indeed. 
20150919 04:54:16< ancestral> But you know… maybe a little overhaul isn’t a bad idea…
20150919 04:54:22< fabi> ancestral: But coding the feature is not that hard.
20150919 04:54:40< ancestral> You’ll have to share with me more about this
20150919 04:54:47< fabi> ancestral: So I could give you a playground to experiment a little...
20150919 04:58:22< shadowm> celtiminstrel: Is there something specific about your PR you want me t comment on?
20150919 04:59:22< shadowm> I don't think I'll be of much help with things like spotting regressions.
20150919 04:59:48-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150919 05:01:00< celtiminstrel> There was nothing specific I had in mind but... maybe if you had any comments on style and/or on the Lua API additions?
20150919 05:03:22-!- higgins [~higgins@105.ip-167-114-152.net] has quit [Quit: Leaving]
20150919 05:03:27< shadowm> I can't comment on the Lua, at least.
20150919 05:04:03< celtiminstrel> By which you mean the Lua code, not the API, right?
20150919 05:04:20< shadowm> I can say that wesnoth.deselect_hex was long overdue, though, which is my fault.
20150919 05:05:35< shadowm> But...
20150919 05:06:50< shadowm> Eh, I guess it works because you select an offmap location.
20150919 05:07:23< celtiminstrel> That's how it was done in the original [message] implementation.
20150919 05:08:14< shadowm> The fact that this wesnoth.skipping_replay function exists seems fishy.
20150919 05:08:43< celtiminstrel> Yeah, maybe...
20150919 05:09:46< shadowm> Basically, the logic for whether to skip a message or not shouldn't be in Lua IMO.
20150919 05:10:19< celtiminstrel> I could build it into the show_message_dialog function instead.
20150919 05:10:22< shadowm> I'd expect it to be handled on the engine side without WML/Lua having to be aware that it even is a thing.
20150919 05:10:49< celtiminstrel> But that does mean it would be handled a fair bit later in the [message] logic.
20150919 05:11:10< celtiminstrel> Not sure if that's a problem.
20150919 05:11:27< Aginor> I am tempted to go through the entire source and introduce macros for constucting/deconstructing colours and defining the bitmasks properly for colours
20150919 05:11:39< Aginor> does anyone have opinions on this?
20150919 05:11:42< celtiminstrel> Why macros instead of functions?
20150919 05:11:42< shadowm> It's not a "problem", but it's a waste of time to do the whole unit matching thing if the message isn't going to be displayed.
20150919 05:12:30< celtiminstrel> Yeah.
20150919 05:13:05< Aginor> celtiminstrel: because they'd be "(x & ASMASK) >>  24" or such, making me feel that a function is overkill
20150919 05:13:23< Aginor> where AMASK would be defined to be 0xFF000000
20150919 05:13:25< celtiminstrel> They'd be inlined.
20150919 05:13:39< celtiminstrel> Probably, even if you didn't explicitly mark them inline.
20150919 05:13:43< shadowm> An inline function would work just as well with the added bonus of type verification.
20150919 05:13:44< Aginor> yeah
20150919 05:13:57< Aginor> I could just as easily have functions :)
20150919 05:14:05< shadowm> C++ coders usually refrain from using macros because they want their types to be checked by the compiler.
20150919 05:14:10< celtiminstrel> shadowm: Would you like me to put the replay-skipping logic into show_message_dialog instead?
20150919 05:14:16< Aginor> :D
20150919 05:14:36< shadowm> celtiminstrel: I'm not sure.
20150919 05:14:38< celtiminstrel> If there's a third option I can't think of it.
20150919 05:16:11< shadowm> Firstly, I think it's not a good idea to let WML/Lua be able to tell the difference between replays and not replays.
20150919 05:16:45< shadowm> But here we need to be able to know when to skip messages, and I guess some specific replay action causes the engine to skip messages too?
20150919 05:17:27< shadowm> But there is also a user action that causes the game to skip messages, which is pressing Escape on an input-less [message].
20150919 05:18:02< celtiminstrel> No, those two are separate.
20150919 05:18:15< shadowm> If I had to choose between informing WML/Lua of the existence of replay mode, or the existence of message-skipping mode, I'd choose the latter.
20150919 05:18:56< celtiminstrel> My implementation of the message-skipping mode uses a Lua local variable, but I suppose I could replace it with an API function instead.
20150919 05:19:32< celtiminstrel> Then replay can be rolled into it.
20150919 05:19:44-!- higgins [~higgins@105.ip-167-114-152.net] has joined #wesnoth-dev
20150919 05:23:11< Aginor> so there's at least two different colour formats in the game, both represented as Uint32... there's scope for improvement
20150919 05:23:23< Aginor> pango vs rest of rendering pilepine
20150919 05:25:16< shadowm> Ultimately, I'm afraid I can't be very helpful here because my knowledge of the target area is largely outdated, I'm not an authority on Lua "best practices" in Wesnoth (I've probably done 80% of my own Lua code completely wrong), and the reimplementation of [message] in Lua in particular is a task I *did* consider at one point and then gave up on it due to its sheer complexity.
20150919 05:32:55-!- Nobun [~nobun@5.170.83.191] has joined #wesnoth-dev
20150919 05:36:23-!- Nobun1 [~nobun@5.170.80.123] has joined #wesnoth-dev
20150919 05:36:28-!- Nobun1 is now known as Nobun_1
20150919 05:37:51-!- Nobun [~nobun@5.170.83.191] has quit [Ping timeout: 240 seconds]
20150919 05:37:58-!- Nobun_1 is now known as Nobun
20150919 05:40:08< Nobun> Hi. I'd like to talk with you developers about wespo (an alternate program to create a .po file from a wesnoth add-on) and its future development. But since it is not an official wesnoth utility (even if does the work of an official utility) I don't know if this channel is the right place I can discuss with you
20150919 05:42:05< Nobun> (a note: some questions I have are also about the target I should do to allow wespo to replace / be added to wmlxgettext as a tool officially recognized for obtaining .po files)
20150919 05:42:28< celtiminstrel> wmlxgettext is old and written in Perl.
20150919 05:42:36< celtiminstrel> What's wespo written in?
20150919 05:43:00< Nobun> currently it is still in development and written in Qt
20150919 05:43:05< celtiminstrel> Eh.
20150919 05:43:06< Nobun> (qt - c++)
20150919 05:43:10< celtiminstrel> I know.
20150919 05:43:30< ancestral> It has cross-platform potential
20150919 05:44:28< Nobun> well... I cannot provide an official port to Mac, since I don't have a Mac pc, but I use cross-platform code so it should be builded also on mac
20150919 05:45:13< ancestral> Nobun: Most of our developers are using Linux or BSD
20150919 05:45:29< Nobun> (celtiminstrel... I specifyed c++ afeter qt becouse qt could be used also with other languages and not only c++ ;) to be not too vague ;) )
20150919 05:45:47< celtiminstrel> It's a C++ library, I'd assume C++ unless otherwise specified.
20150919 05:46:04< Nobun> I have Ubuntu 14.04 LTS as my own system
20150919 05:46:52< Nobun> celtiminstrel: you are right... I was feared too much I not answered correctly to you :P
20150919 05:47:49< Nobun> however now I try to explain why I try to contact you and to speak about what should I do with wespo
20150919 05:48:53< Nobun> before all: currently wespo is not a good candidate to be recognized as an official tool.... there are still some bugs into the application
20150919 05:49:14< Nobun> I am working in my spare time to fix them
20150919 05:49:52< celtiminstrel> I don't think it's a good candidate in general to be recognized as an official tool (eg being included in the wesnoth repo).
20150919 05:49:58< Nobun> but I am in a point I have to do some important choices, so I'd like to hear your opinion and what it is expected by a tool officially recognized
20150919 05:50:44< celtiminstrel> This is just my personal opinion, but I think tools like this are better written in a language like Python or Perl.
20150919 05:51:12< Nobun> ok I see your point
20150919 05:51:44< Nobun> I choosed Qt becouse it is the platform I know better (wespo started several years ago, mainly for personal use)
20150919 05:51:48< ancestral> Nobun: There are some devs who do more translations that may have strong opinions or suggestions for you
20150919 05:52:15< ancestral> If you don’t get that much response here right now, feel free to bring it up on the forums
20150919 05:53:10< Nobun> I have already a forum topic for wespo, but I used it only to speak to people that could find wespo useful for them
20150919 05:53:50< Nobun> simply becouse wespo (in its current config) tries to be as simple as possible for final user (a very simple gui... click a button and all is done)
20150919 05:54:31< Nobun> but now I see some issues / decisions that could be better speaked with developers
20150919 05:55:17< Nobun> (for example... in its current state wespo does not work on command line, but only on this simply gui)
20150919 05:55:55< shadowm> 02:45:15 <ancestral> Nobun: Most of our developers are using Linux or BSD
20150919 05:56:19< Nobun> yeah, shadowm: but I don't see the pont
20150919 05:56:21< Nobun> point
20150919 05:56:21< shadowm> ancestral: Really? Last time I checked we had exactly zero developers on *BSD.
20150919 05:56:31< ancestral> Okay
20150919 05:56:41< shadowm> And we had more developers on Windows than on Linux.
20150919 05:57:19< Nobun> I am impressed on this second information, shadowm: compiling wesnoth on windows is a huge pain
20150919 05:57:47< celtiminstrel> I think Mac (or Darwin, at least) is actually partly based on a BSD.
20150919 05:57:54< shadowm> Sure, but this channel is full of masochists anyway.
20150919 05:58:07< shadowm> Otherwise we wouldn't use C++.
20150919 05:58:09< Nobun> I did in past when there was the issue on 1.8.0 with the lobby gui... before the official exe released... I needed a week to figure out to compile it on win natively
20150919 05:58:11< ancestral> (celtiminstrel: Heh, true, OS X is a BSD)
20150919 05:58:21< ancestral> At least sort of
20150919 05:58:35< shadowm> Branding-wise OS X is not a *BSD.
20150919 05:59:07< shadowm> The packaging process is also fairly different from what you'd have on a branded BSD derivative.
20150919 06:00:31< shadowm> Nobun: To answer your question, for a tool to be merged into mainline we expect it: 1) to be useful to a large number of people; 2) to have an active maintainer (i.e. you).
20150919 06:01:12< Nobun> yeah... exactly what I am trying to do
20150919 06:01:16< shadowm> Also 3) not introduce a large number of new dependencies.
20150919 06:01:40< shadowm> Wesnoth does not depend on Qt, so that would be a problem.
20150919 06:01:43< Nobun> about 3)... my idea is not to include wespo into main core, but left it as separate application
20150919 06:02:16< Nobun> becouse it would be totally crazy to add qt dependencies for wespo on wesnoth
20150919 06:03:18< shadowm> More to the point, what would be the benefit of having it in the mainline repository?
20150919 06:03:46< Nobun> well... I think the benefit are mainly 2:
20150919 06:04:02< Nobun> 1) visibility for end-users
20150919 06:04:49< Nobun> 2) let the devs know that also wespo exist, and if it will work better, they can use it to create .po without too much headache on using wmlxgettext
20150919 06:05:20< shadowm> What kind of devs are we talking about?
20150919 06:05:21< Nobun> since .po file in source package are already created
20150919 06:05:40< Nobun> devs (maintainers) of main core add-ons
20150919 06:05:43< Nobun> obliouvsly
20150919 06:06:01< shadowm> There are no main core add-ons, though.
20150919 06:06:18-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150919 06:06:35< Nobun> well... I am speaking about official campaigns, for example (Heir to the throne, etc)
20150919 06:07:00< shadowm> The mainline campaigns are not considered add-ons. :p
20150919 06:07:28< celtiminstrel> Obliviously? :P
20150919 06:07:45< Nobun> yeah I know... but I said add-ons improperly to mark the fact I am speaking not about the application core, but about wesnoth data files
20150919 06:07:56< Nobun> ;)
20150919 06:08:01< shadowm> It's true that Debian and derivatives ship them in optional packages, but as far as the game is concerned they are part of the game.
20150919 06:08:36< celtiminstrel> ...why do they ship them in optional packages?
20150919 06:08:41< shadowm> As far as the Wesnoth development processs is concerned, too, they are part of the game, which means that campaign maintainers never have to touch the translation toolchain.
20150919 06:09:23< ancestral> There is a tools directory in the Wesnoth binary packages
20150919 06:09:31< Nobun> shadowm: can you explain your last sentence?
20150919 06:09:31< shadowm> Instead, the translation catalogues are created by the release team at specifically coordinated checkpoints.
20150919 06:09:50< shadowm> Campaign maintainers are not supposed to generate translations.
20150919 06:10:30< Nobun> well but the .po file should be created somehow... or not?
20150919 06:10:40< Nobun> who creates the .po file there? how?
20150919 06:11:18< shadowm> The translation catalogues are generated by the release team at specifically scheduled points around the time a release is supposed to happen. During a string freeze (beta/RC/stable releases), this is done around a week before releasing; when not in a string freeze, this is done immediately before releasing (alpha development versions).
20150919 06:11:19< celtiminstrel> He just said. The release team does it.
20150919 06:11:53< Nobun> yes... the team does it.... but how?
20150919 06:11:57< shadowm> This used to be Ivanovic's task, and now it's mine.
20150919 06:12:14< ancestral> Perhaps a better question is, is this a problem that needs solving?
20150919 06:12:23< celtiminstrel> Isn't it done with scons?
20150919 06:12:28< Nobun> nope
20150919 06:12:30< shadowm> We both use cmake to build the game binaries on Linux. The cmake recipe has an option to generate the translation catalogue templates.
20150919 06:12:40< celtiminstrel> Or is that just generation .mo from .po?
20150919 06:12:49< celtiminstrel> ^generating
20150919 06:12:54< shadowm> -DENABLE_POT_UPDATE_TARGET=ON
20150919 06:13:18< Nobun> as far as I can see... when you download a source tree you already have the .po
20150919 06:13:20< shadowm> celtiminstrel: Yes, both scons and cmake are able to compile translation catalogues and they are supposed to do that as part of the usual build process.
20150919 06:13:26< shadowm> Nobun: Yes.
20150919 06:14:12< Nobun> but what I am trying to understand (sorry if I can be pedantic) is how the official .po of official campaigns are actually generated
20150919 06:14:27< shadowm> cmake does it for us using wmlxgettext.
20150919 06:14:55< Nobun> ok so you do it using cmake that calls wmlxgettext
20150919 06:15:01< Nobun> now I understood
20150919 06:15:32< Nobun> ok so the reason 2 we can say "does not exist"
20150919 06:16:00< Nobun> since the official .po can be generated using build toolchain
20150919 06:16:15< Nobun> did I understood correctly?
20150919 06:16:23< shadowm> Yes.
20150919 06:17:38< Nobun> well... I tried to use cmake in past to compile wesnoth, but without success.... I have more luck using scons (that is the way I build wesnoth when a new release is out, on my Ubuntu)
20150919 06:17:59< shadowm> The scons recipe does have a pot-update target as well, but nobody uses it AFAIK.
20150919 06:18:32< Nobun> so... coming to the point...
20150919 06:19:13< Nobun> the only reason of wespo to be "official" is for final users (mantainers / developers of add-ons)
20150919 06:19:25< Nobun> that can be not enough
20150919 06:19:33< Nobun> so the solution could be:
20150919 06:19:45< shadowm> Regarding visibility, I think that is a challenge that can only be properly overcome by taking advantage of the wiki and word-of-mouth.
20150919 06:20:30< shadowm> For example, Wesnoth has a team-colorizing script that's used in the process of building units.wesnoth.org, but nobody knows about it.
20150919 06:21:06< shadowm> I didn't back then either, so I wrote my own team-colorizing CLI app, and then its GUI counterpart that is slightly better known.
20150919 06:21:15< Aginor> wedge009: the bug seems to be happening on classes that has/relies on a undraw() method
20150919 06:21:41< Aginor> wedge009: where a saved part of the original map is blitted ontop of the current map
20150919 06:22:08< shadowm> (It's still not very known because although I did bother making sure to provide binaries for Windows and OS X, I didn't work on promoting it around the forums myself.)
20150919 06:22:42 * Aginor is making progress on the SDL2 alpha blending issue
20150919 06:22:43< Nobun> shadowm: don't get me wrong. I am not concerned by visibility itself
20150919 06:23:04-!- [Relic] [~Relic]@2602:306:33a3:6d30:3dc0:8552:1a83:7c5] has quit [Quit: I press the magic X and all the weirdos go away!]
20150919 06:23:14< Nobun> the reason of all is... I am trying to do a wmlxgettext replacement that can be useful to all
20150919 06:24:19< Nobun> since, as a user, I experienced to never understand how to create a .po for a campaign I developed (never released since it is incomplete)... so I created a tool for myself to make the work
20150919 06:25:45< Nobun> the idea is to do now a tool that is not useful to myself only (and give it to everyone that can find it useful) but to do an actual useful tool...
20150919 06:26:21< Nobun> this means I could also freeze wespo development and try to develop an alternate version of wmlxgettext trying to rewrite it using python
20150919 06:26:37< Nobun> (even I never used python in my life)
20150919 06:27:09< Nobun> rewrinting wmlxgettext, in this case, could mean:
20150919 06:27:22< Nobun> - change command line and make it more understandable to users
20150919 06:27:50< Nobun> - give documentations and perhaps develop a gui with tk
20150919 06:28:22< celtiminstrel> Someone apparently already tried to rewrite wmlxgettext in Python.
20150919 06:28:32< Nobun> yeah I know
20150919 06:28:39< celtiminstrel> Really?
20150919 06:29:08< celtiminstrel> Anyway, from what I understand, your tool might well be very useful to the average add-ons developer.
20150919 06:29:12< Nobun> since there is two versions of wmlxgettext (one in perl one in python) in official core....
20150919 06:29:24< celtiminstrel> Oh, really? I actually didn't know that.
20150919 06:29:54< Nobun> yeah but the python one actually doesn't work well, so it is untrusty
20150919 06:31:15< Nobun> the only advantage of wespo (even if bugged) alongside wmlxgettext is the chance to redirect the translation comment
20150919 06:31:18< shadowm> The Python version was rejected but it was only finally removed in version 1.12.3.
20150919 06:32:18< Nobun> this is very useful on lua side... since wespo doesn't parse it (only captures string) but it allows you to let wespo know what to write in string comment
20150919 06:32:41< celtiminstrel> Chance to do what now?
20150919 06:33:09< Nobun> I make an example on pastebin... wait a moment
20150919 06:43:28-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150919 06:45:06< Nobun> sorry for taking so much time
20150919 06:45:23< Nobun> here is the link: http://pastebin.com/zfP2eAjx
20150919 06:45:33< Nobun> I hope I was clear
20150919 06:46:15< Nobun> (the lua code is bad... I am not a wesnoth lua expert... but I hope I could explain the point)
20150919 06:47:54< Nobun> (the wespo comment redirection could be done also on WML since it is a WML comment introduced by #, that is interpreted by wespo as a special command)
20150919 06:48:44< Nobun> (more than redirection... I should say "comment overriding"... it is more correct.... my english is not so good)
20150919 06:48:54< celtiminstrel> wesnoth.message is more analogous to [chat] than [message]. Just FYI.
20150919 06:48:58< celtiminstrel> But I see your point.
20150919 06:49:42< Nobun> yeah but for wespo, every lua instruction is the same (since it extracts infos only on WML)
20150919 06:50:02< Nobun> it cares only by translatable strings and textdomains
20150919 06:50:15< celtiminstrel> Why is it [lua]?
20150919 06:50:52< Nobun> to make a default comment... [lua]: 
20150919 06:50:57< celtiminstrel> When Lua is placed in a separate file, I though it use usually included via wesnoth.dofile rather than by macro inclusion (the latter is kind of weird to make work, anyway).
20150919 06:51:03< celtiminstrel> ^thought
20150919 06:51:09< Nobun> yeah
20150919 06:51:20< Nobun> but I make the decision for this reason
20150919 06:51:21< celtiminstrel> I guess it doesn't actually parse the WML though, just scans all the files...
20150919 06:51:34< Nobun> yeah but do a sort of parse
20150919 06:51:57< Nobun> it can extract useful infos (like message speaker on tag message and so on)
20150919 06:52:15< Nobun> since the syntax on .po file is
20150919 06:52:40< Nobun> [tagname]: list of useful infos; divided by commas;
20150919 06:52:58< Nobun> (the syntax used by wmlxgettext)
20150919 06:52:58< celtiminstrel> That's just Wesnoth convention though, isn't it?
20150919 06:53:03< Nobun> yep
20150919 06:53:33< Nobun> according to the convention, I marked [lua] as default to say "hey it is a lua code"
20150919 06:53:56< Nobun> regardless the fact the lua code is written in a separate file or in a [lua] tag inside a WML file
20150919 06:55:29< Nobun> however... coming to the point... to shadowm:
20150919 06:55:51< Nobun> in case I would try to rewrite by scratch a python version of wmlxgettext
20150919 06:56:41< Nobun> 1) is there a command line syntax I should follow, or am I free to choose also different params and command line structure?
20150919 06:57:29< Nobun> 2) and... if I should follow a certain structure for the command line... what is exactly the command line structure I should follow?
20150919 06:58:01< shadowm> For the record the author of the Python wmlxgettext port closed the original bug report with Won't Fix.
20150919 06:58:05< Nobun> (by command line syntax I mean command line parameters required and optional by wmlxgettext)
20150919 06:58:22< shadowm> https://gna.org/bugs/index.php?11294#comment6 -- 'Marking as "Wont Fix" because Python was too slow for Ivanovic. Until he changes his mind, we're stuck with the Perl implementation.'
20150919 06:59:07< celtiminstrel> Was it too slow because it was badly coded or because it was written in Python?
20150919 06:59:35< shadowm> I kind of don't want to know.
20150919 06:59:51 * vultraz facepalms
20150919 06:59:52< celtiminstrel> Personally, I don't care if it's written in Perl or Python, but I think it would be good for it to be updated a bit.
20150919 07:00:00< shadowm> I'd like to think it wasn't badly coded given who the author was.
20150919 07:00:12< celtiminstrel> I have no idea who the author was.
20150919 07:00:19< shadowm> The same person who closed that bug report.
20150919 07:00:21 * celtiminstrel didn't click the buglink.
20150919 07:00:25< vultraz> Read what shadowm said
20150919 07:01:24< shadowm> And also, knowing *who* spearheaded that effort, I'd like to think if he knew there was a solution he'd have insisted on it.
20150919 07:01:54< shadowm> He's certainly been much pushier about far more controversial things before.
20150919 07:02:20< shadowm> But to answer Nobon's questions...
20150919 07:02:27< celtiminstrel> I don't actually know who that is.
20150919 07:02:40< shadowm> Wow really?
20150919 07:02:49< shadowm> I'd expect every single OSS developer in existence to know that name. :p
20150919 07:03:10< vultraz> Even I know 
20150919 07:03:12< shadowm> Nobun: If you want it to be considered for mainline you will have to make it a drop-in replacement for the current wmlxgettext replacement, meaning: exact same command line invocation.
20150919 07:03:39< Nobun> ok so can I have the exact same command invocation?
20150919 07:03:52< shadowm> Must, rather than can, but yes.
20150919 07:04:22< Nobun> perhaps I could consider to rewrite again the program from scratch, using standard c++ this time and the c++ libraries used by wesnoth core
20150919 07:04:29-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150919 07:04:35< celtiminstrel> Well, sorry for not knowing. >_>
20150919 07:04:47< shadowm> celtiminstrel: https://en.wikipedia.org/wiki/Eric_S._Raymond
20150919 07:04:50 * Aginor is disappointed in people not knowing who esr is
20150919 07:05:24< shadowm> But we'd rather prefer if you didn't say that name or his initialism.
20150919 07:05:42< shadowm> In the channel, at least.
20150919 07:05:53< celtiminstrel> Presumably because it could ping him.
20150919 07:06:04< Aginor> ok
20150919 07:07:00< Aginor> anyhow, my 2c is that python has come a long way since 2008 and may have better performance
20150919 07:07:06< Aginor> computers sure have if nothing else
20150919 07:10:33< Nobun> however
20150919 07:11:32< Nobun> if a an expert developer stopped python for this reason, I'm not supposed to try to do better since I am only a person who know programming only a bit becouse of personal interest
20150919 07:12:11< Nobun> I'm not an expert, so I can only listen what an expert sayd
20150919 07:12:23< Nobun> the only chance is to do a c++ version of wmlxgettext
20150919 07:12:49< Nobun> that is surely quicker as performance (even if bad written)
20150919 07:13:12< celtiminstrel> I imagine badly-written C++ can also perform poorly.
20150919 07:13:18< Nobun> however I am stuck to the question asked before.... what is the exact command line invocation of wmlxgettext?
20150919 07:13:22< celtiminstrel> The problem with C++ in general though is that it needs to be compiled.
20150919 07:13:22< shadowm> celtiminstrel: Certainly.
20150919 07:13:39< shadowm> That is also a problem, since it introduces a prerequisite for the pot-update target that didn't exist before.
20150919 07:13:44< celtiminstrel> Nobun: You can probably find it by digging through the cmake or scons files.
20150919 07:14:08< Nobun> I will try with cmake one, than
20150919 07:14:20< Nobun> becouse scons is too much complex to me
20150919 07:14:22< shadowm> It is certainly feasible to make cmake use a tool it must built first, but it seems a bit awkward.
20150919 07:14:47< celtiminstrel> I was trying to figure out how to make scons do that a few days ago (for test cases).
20150919 07:14:47< Nobun> not so much
20150919 07:15:19< Aginor> it's outside my carved-off scope, but I think that adding wmlgettext++ as a pre-requisite will complicate the build process further. Will it have enough benefit to be worth the complication?
20150919 07:15:36< Nobun> I had developed (in past) a program that does different compiles in a order
20150919 07:15:47< Nobun> using cmake, so it is possible
20150919 07:15:54< celtiminstrel> I don't think it would have enough benefit to be worth it.
20150919 07:15:55< Aginor> Nobun: mkdir mybuild && cd mybuild && cmake .. && make VERBOSE=1
20150919 07:16:07< celtiminstrel> But I'm not the one who deals with translations.
20150919 07:16:20 * Aginor knows nothing of the translation
20150919 07:16:40< shadowm> The current wmlxgettext implementation takes --directory=<STARTING DIRECTORY> --domain=<TEXTDOMAIN> and optionally --initialdomain=<DEFAULT TEXTDOMAIN>, writes the generated catalogue to stdout, and errors and warnings to stderr.
20150919 07:16:43< Aginor> and I need to sneak off or I'll be in trouble
20150919 07:16:54 * celtiminstrel brings the updated PR488 to the attention of zookeeper.
20150919 07:17:06< Nobun> thank shadow
20150919 07:17:10-!- celtiminstrel is now known as celmin|sleep
20150919 07:17:41< shadowm> The exact invocation for mainline is unclear (add-ons have a single textdomain, mainline doesn't).
20150919 07:18:06< shadowm> So figuring that out will require a bit of digging around in the CMake or SCons recipes.
20150919 07:18:08< celmin|sleep> Theoretically, couldn't an add-on have multiple textdomains?
20150919 07:18:13< celmin|sleep> Though that would be a little silly.
20150919 07:18:24< shadowm> Yes, but WesCamp-i18n wasn't designed for that.
20150919 07:18:35< celmin|sleep> That still exists?
20150919 07:18:42-!- ancestral [~ancestral@97-116-185-165.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150919 07:19:07< shadowm> As for --initialdomain, it's an optional parameter and its argument is the default textdomain used before the first #textdomain binding for each input file.
20150919 07:19:54< shadowm> Also, wmlxgettext can apparently also take paths to individual input files as the remainder of the command line.
20150919 07:20:50< shadowm> Also, the default value for --initialdomain is supposed to match the PACKAGE macro in src/wesconfig.h, so it's currently "wesnoth" (for the wesnoth textdomain).
20150919 07:22:35< shadowm> Looking at how I run wmlxgettext on my add-on, I'm not really sure what the purpose of --directory is, but it's certainly used (I'm fairly sure I modelled my invocation after the WesCamp-i18n scripts).
20150919 07:22:44< shadowm>         @find . \( -name '*.cfg' -o -name '*.lua' \) -type f -print0 | xargs -0 wmlxgettext --directory=$(targetdir) --domain $(textdomain) > $*.pot
20150919 07:26:04< Nobun> thank for the informations, shadowm
20150919 07:26:25< Nobun> I'll check what to do... when I will have time
20150919 07:27:18< Nobun> perhaps I could continue with wespo as a separate application for the use ONLY of campaign developers (for easier approach) and let wmlxgettext as it is
20150919 07:28:23< Nobun> or perhaps I will try to make a c++ version of wmlxgettext
20150919 07:28:32< Nobun> or perhaps both
20150919 07:29:20-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150919 07:30:37-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20150919 07:31:35< Nobun> only a final question shadowm:
20150919 07:32:22< Nobun> but it is actually required that wmlxgettext is "ready to use" BEFORE building wesnoth?
20150919 07:32:40< Nobun> trying to explain my question
20150919 07:33:33< Nobun> currently when you build wesnoth from source you don't require to generate .pot files, since they were already generated by developers and inserted in source tree by them
20150919 07:34:20< Nobun> so we can say that the .pot generation follows a different "compile process" that is conceptually separated by the actual wesnoth build
20150919 07:35:17< shadowm> I already answered this I think.
20150919 07:35:24< shadowm> 04:13:39 <shadowm> That is also a problem, since it introduces a prerequisite for the pot-update target that didn't exist before.
20150919 07:35:28< shadowm> 04:14:22 <shadowm> It is certainly feasible to make cmake use a tool it must built first, but it seems a bit awkward.
20150919 07:35:32< shadowm> *build
20150919 07:36:22< shadowm> Like Aginor said, it's a matter of whether the benefits outweigh the disadvantages.
20150919 07:37:05< Nobun> ok... so if a c++ replacement will be done, I should also find a solution to solve this issue into build generation
20150919 07:37:06< shadowm> If you asked me, I think there's no point in moving this kind of tool to C++ land unless it's to use Wesnoth's WML parser.
20150919 07:37:39< shadowm> Which may be a good idea but it also means we lose things like the ability to have 1:1 line mappings and po file comments.
20150919 07:37:57< shadowm> In other words, I wouldn't do it.
20150919 07:38:10< Nobun> ok
20150919 07:39:35< Nobun> I am only trying to contribute as I can... since I can try to see how to make a tool that is available for end users and that it is simplier to use than wmlxgettext
20150919 07:40:00< Nobun> (end users = campaign - add-on maintainers)
20150919 07:40:13< Nobun> so coming to a last step
20150919 07:40:27< shadowm> The tool was supposed to be WesCamp-i18n. :p
20150919 07:40:43< shadowm> But then that kind of fell through thanks to sloppy administration practices.
20150919 07:40:44< Nobun> but it was abandoned, as far as I know
20150919 07:40:56< shadowm> Not officially, it's just that the admin is AWOL.
20150919 07:41:18< Nobun> however... if I remember well.... WesCamp requires you upload your add-on
20150919 07:41:31< shadowm> I am supposed to look into bringing it back but I'd really prefer to hand the stupid thing to someone else.
20150919 07:41:58< Nobun> I'd like to allow users to create .po file locally on their pcs 
20150919 07:42:41< Nobun> yeah... I am trying to contribute somehow, not to add duties to you ;)
20150919 07:44:42< zookeeper> celmin|sleep, API-wise it looks good to me now.
20150919 07:44:43< zookeeper> but of course you need to just deprecate, not remove the old leadership macros (personally, i'd just rewrite them as wrappers for the new macro as to not even pester people with deprecation warnings). also, reading the auto-stored variable section of the changelog was extremely confusing and i couldn't really tell anything about what it was saying.
20150919 07:45:17< zookeeper> it says that there's a new auto-stored variable available, but then it never says what it is
20150919 07:45:46< zookeeper> then it gives tags which already exist and says they "refer to the unit [...]"
20150919 07:48:00< zookeeper> i just can't make heads or tails of that section
20150919 07:57:07-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150919 08:02:27-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 256 seconds]
20150919 08:02:53-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20150919 08:30:48-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150919 08:39:04-!- mjs-de [~mjs-de@f049239245.adsl.alicedsl.de] has joined #wesnoth-dev
20150919 08:50:59-!- jinblack [~jinblack@128.111.48.6] has quit [Ping timeout: 246 seconds]
20150919 08:53:09-!- Nobun [~nobun@5.170.80.123] has quit [Quit: Salve a tutti]
20150919 09:04:49-!- jinblack [~jinblack@jinblack.it] has joined #wesnoth-dev
20150919 09:05:05-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 255 seconds]
20150919 10:08:10-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150919 10:20:01-!- mjs-de [~mjs-de@f049239245.adsl.alicedsl.de] has quit [Remote host closed the connection]
20150919 10:22:24< Aginor> shadowm: I just had a thought; What's your cpu?
20150919 10:23:01< Aginor> (what if the bug is nvidia independant and the problem is instead with one of SDL's optimised blit operations?)
20150919 10:49:27-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has joined #wesnoth-dev
20150919 10:53:32< Aginor> ok, I think I have a solution for the alpha-blending issue;
20150919 10:54:06< Aginor> The root cause is that a number of places draws a copy of the screen onto it self, this seems to be broken on some hardware
20150919 10:54:48< Aginor> I propose the simpleish fix of not drawing a copy back onto the screen, but instead invalidating that portion of the screen so it is redrawn properly
20150919 10:55:38< Aginor> this ought to be more robust, there's no checks if something has changed before the copy is put onto the screen and it will simplify some of the rendering logic
20150919 10:56:02< Aginor> it is also a step in the right direction for moving to proper hardware acceleration
20150919 10:56:26< Aginor> shadowm, wedge009, lipkab: do you have any thoughts on the above?
20150919 10:57:27< Aginor> My plan is to make a PR for these changes, starting with the functionality that I've seen problems in, but ultimately I'd like to replace the undraw() functions with a simple invalidation of the bounding rectangles
20150919 11:02:27-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20150919 11:02:33-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20150919 11:05:23-!- gen_battle [~genbattle@182.16.153.90] has joined #wesnoth-dev
20150919 11:07:58-!- genbattle_ [~genbattle@182.16.153.90] has joined #wesnoth-dev
20150919 11:08:54-!- genbattle [~genbattle@182.16.153.90] has quit [Ping timeout: 260 seconds]
20150919 11:10:50-!- gen_battle [~genbattle@182.16.153.90] has quit [Ping timeout: 240 seconds]
20150919 11:16:28-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150919 12:09:37-!- louis94 [~~louis94@109.129.229.117] has joined #wesnoth-dev
20150919 12:12:45-!- ancestral [~ancestral@97-116-185-165.mpls.qwest.net] has joined #wesnoth-dev
20150919 13:01:52-!- mjs-de [~mjs-de@f049239245.adsl.alicedsl.de] has joined #wesnoth-dev
20150919 13:05:14-!- louis94 [~~louis94@109.129.229.117] has quit [Ping timeout: 246 seconds]
20150919 13:13:01-!- ancestral [~ancestral@97-116-185-165.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20150919 13:18:10-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150919 13:22:03< celmin|sleep> zookeeper: Regarding leadership, I moved the old macros into "deprecated-utils.cfg" but was unsure if I needed to d anything else (there's a DEPRECATE macro that's never used anywhere).
20150919 13:22:38< celmin|sleep> zookeeper: Also any suggestions on the .cw/.ccw syntax specifically?
20150919 13:23:18< celmin|sleep> ^do
20150919 13:24:17< celmin|sleep> (I guess the DEPRECATE macro wouldn't be very useful here anyway.)
20150919 13:24:36< celmin|sleep> (I did add translation rules to wmllint.)
20150919 13:30:07< Ravana_> could I see how you did that leadership?
20150919 13:30:46< celmin|sleep> https://github.com/CelticMinstrel/wesnoth/blob/f54d6b5d28cb15ff9adbb452331358032339af92/data/core/macros/abilities.cfg#L99
20150919 13:31:17-!- irker316 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20150919 13:31:30< Ravana_> ok, I thought that it would be good if the value= could refer to $other_unit
20150919 13:32:04< celmin|sleep> Unforunately the value= is not in the filter, so it can't.
20150919 13:32:38< Ravana_> could +tag way reach to that value= ?
20150919 13:32:50< celmin|sleep> I was thinking about an additional change at some point that allows formulas for the value, or something like that, but I think I'll leave that for later.
20150919 13:33:04< celmin|sleep> ...I'm not quite sure what you mean, but I highly doubt it would work.
20150919 13:34:41< celmin|sleep> Using $other_unit in the backstab special might actually be confusing.
20150919 13:35:06< Ravana_> like http://pastebin.com/XiSkrYn4
20150919 13:35:38< celmin|sleep> Yeah, no, that won't work.
20150919 13:36:37< celmin|sleep> It might look like that's inside the filter, but it's not.
20150919 13:49:54-!- celmin|sleep is now known as celticminstrel
20150919 13:50:53-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150919 13:51:24-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150919 13:57:08-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has quit [Remote host closed the connection]
20150919 14:00:13-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has joined #wesnoth-dev
20150919 14:02:38< celticminstrel> mattsc: By the way, would you mind trying run_wml_tests to see if it works for you? You'll probably need to pass -pxcode for it to work.
20150919 14:03:12< celticminstrel> vultraz: What's your opinion on merging PR492? Should it be tested more extensively first? (I can show you the tests I did, if you want.)
20150919 14:03:20-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has joined #wesnoth-dev
20150919 14:04:25< vultraz> I still think we should deprecate get_user_choice since it calls message. Players have a whole dev cycle to deal with converting, and can synch the results themselves, but if it's really an issue just leave it. As for testing, I trust you did all the relevant tests. If an issue is found later it can be fixed
20150919 14:05:32< celticminstrel> There are a few things I didn't test, but I did all the basic tests (one with options, one with text input, also the escape key).
20150919 14:06:41< celticminstrel> I think something does need to be done about get_user_choice (deprecating it or otherwise), but I guess that can wait as well.
20150919 14:07:04< celticminstrel> Another option rather than deprecating it would be to rewrite it to simple sync a call to show_message_dialog.
20150919 14:08:03< celticminstrel> But then it would have fewer features.
20150919 14:08:13< celticminstrel> eg, no automatic speaker detection.
20150919 14:08:26-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has quit [Read error: Connection reset by peer]
20150919 14:12:13-!- Shackra [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150919 14:14:57-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20150919 14:17:12< mattsc> celticminstrel: it runs, but it gives only fails  (with or without that argument)
20150919 14:17:35-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has joined #wesnoth-dev
20150919 14:18:37< celticminstrel> mattsc: If you pass -v, does it print a correct path?
20150919 14:18:55< celticminstrel> I suppose it should be set to not even attempt the tests if the Wesnoth binary cannot be found.
20150919 14:19:18< celticminstrel> Correct path to Wesnoth, I mean.
20150919 14:19:18< mattsc> celticminstrel: let me ask a different question first.  Do I need to compile unit tests for this first?
20150919 14:19:31< celticminstrel> No, only the main Wesnoth app.
20150919 14:19:34< mattsc> ok
20150919 14:19:51< celticminstrel> (Using XCode.)
20150919 14:20:09< mattsc> well, the error.log file says “command not found”, so my guess is it does not find the binary
20150919 14:20:13< mattsc> let me check
20150919 14:20:31< celticminstrel> If it's really fast, that's another indication it didn't find the binary.
20150919 14:20:39< mattsc> yes, I was just going to say that
20150919 14:21:59< mattsc> yes, it says ./wesnoth, which is neither the path nor the correct spelling (capital W needed) of the Wesnoth binary
20150919 14:22:04-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has quit [Read error: Connection reset by peer]
20150919 14:22:27< celticminstrel> Normally capitalization isn't a problem on OSX.
20150919 14:22:38< celticminstrel> Though I suppose some people might've set their filesystem case-sensitive.
20150919 14:23:02< celticminstrel> Did you pull from master before trying this?
20150919 14:23:45< zookeeper> celticminstrel, oh right, i just missed the changes to deprecated-utils.cfg.
20150919 14:23:46< mattsc> it is if you run it from the commandline, isn’t it?
20150919 14:24:13< mattsc> anyways, it’s not a problem for testing, give me a moment
20150919 14:24:14< celticminstrel> No. The tab-completion is case-sensitive, though, which is kind of silly.
20150919 14:24:30< celticminstrel> I think there's an option somewhere in the shell to make tab-completion case-insensitive.
20150919 14:24:35< zookeeper> celticminstrel, so basically the new cw/ccw thing works like this: ne.cw is se, n.ccw is nw, etc?
20150919 14:24:49< celticminstrel> zookeeper: Yes.
20150919 14:25:02< mattsc> Hmm, you’re right (I just tried).  I never knew that, only that the tab completion is.
20150919 14:25:16< zookeeper> celticminstrel, seems pretty straightforward and nice to me, and i think cw and ccw are the logical names to use
20150919 14:25:30< celticminstrel> I was just recently thinking maybe replace the . with :
20150919 14:26:02< zookeeper> yeah that wouldn't hurt
20150919 14:27:10< mattsc> and no, I had pulled last night, hadn’t seen yet that you had made changes since.
20150919 14:27:12< celticminstrel> Alright then.
20150919 14:27:39< celticminstrel> mattsc: Sorry for not being clear. >_>
20150919 14:27:58< mattsc> No worries…  But it still uses ./wesnoth.
20150919 14:28:14< celticminstrel> Even with -pxcode?
20150919 14:29:39< mattsc> oh, right ...
20150919 14:30:33< mattsc> If I use that, it finds the binary, but still fails
20150919 14:30:40< celticminstrel> Excellent.
20150919 14:30:53< celticminstrel> Does the command-line include "timeout"?
20150919 14:31:03< mattsc> no
20150919 14:31:07< mattsc> I need to take care of some other stuff now, I’ll be back later.
20150919 14:31:10< celticminstrel> Hmm, okay.
20150919 14:31:15< celticminstrel> Well, it's a start, I guess.
20150919 14:31:53< mattsc> or what do you mean with “does it include timeout”?
20150919 14:32:00< mattsc> Does it say that somewhere?  No.
20150919 14:32:09< mattsc> anyways, I’ll be back.
20150919 14:32:12< celticminstrel> timeout 10 /path/to/Wesnoth
20150919 14:32:30< zookeeper> celticminstrel, it looks like you can't use multiple rotations, like n:cw:cw:cw? if so then that'd be good to remember to mention in the wiki.
20150919 14:32:55< celticminstrel> zookeeper: You can if you use parentheses, actually.
20150919 14:33:17< celticminstrel> I did that so that if someone set a variable to something like "n.cw", then "$var|.cw" would still work.
20150919 14:33:27< celticminstrel> Well. "($var|).cw".
20150919 14:33:38< celticminstrel> ...I guess the | isn't needed in that case.
20150919 14:33:47< celticminstrel> Mind you, I haven't tested the parentheses...
20150919 14:34:15< zookeeper> err, so... "((n:cw):cw):cw"? O.o
20150919 14:34:20< celticminstrel> Yeah.
20150919 14:34:24< zookeeper> okay
20150919 14:34:52< celticminstrel> I didn't think multiple rotations would be something people want, because every possibly relative direction is obtainable without them.
20150919 14:35:08< celticminstrel> I only did it in case someone set a variable to a direction that already had a rotation.
20150919 14:35:50< zookeeper> yeah, if you keep track of rotations in a variable, you might just keep tacking on rotations to it instead of re-calculating the final direction every time you do a rotation
20150919 14:36:01< zookeeper> ...if that makes sense
20150919 14:36:08< celticminstrel> (ie, n, n:cw, n:ccw, -n, -n:cw, -n:ccw covers all six possible directions.)
20150919 14:36:10< celticminstrel> Yeah.
20150919 14:36:57< celticminstrel> Might be easier if the parentheses were optional...
20150919 14:38:40< celticminstrel> I dunno, it can always be changed again later, too.
20150919 14:41:45< zookeeper> sure, it doesn't really matter if you can't chain them, as long as the documentation is clear on it
20150919 14:41:55< zookeeper> or if you can only chain with parentheses
20150919 14:45:21< celticminstrel> Where do you think is a good place to document it? Maybe the StandardLocationFilter page?
20150919 14:46:14< zookeeper> will that feature work anywhere outside abilities/specials?
20150919 14:46:24< zookeeper> in adjacent filtering in general?
20150919 14:46:32< celticminstrel> It works pretty much anywhere a direction is needed.
20150919 14:47:25< mattsc> celticminstrel: I still don’t have time to be back, but I just noticed that the script finds the 1.12 binary.
20150919 14:47:57< celticminstrel> mattsc: Ah, interesting.
20150919 14:48:47< celticminstrel> It takes the first matching binary it can find, so I guess there's still a little room for improvement.
20150919 14:48:55-!- jinblack [~jinblack@jinblack.it] has quit [Ping timeout: 250 seconds]
20150919 14:50:40< mattsc> yeah, I figured
20150919 14:51:06< mattsc> When I get around to it later today I’ll force it to use the correct one and will tell you what happens
20150919 14:51:27< celticminstrel> You can do -p/path/to/wesnoth for that.
20150919 14:51:43< zookeeper> celticminstrel, ok, IIRC that means only the adjacent filtering tags in SUF, SLF and abilities/specials, so i guess it would really need to be mentioned in all of them. it's a bit awkward, but a dedicated page just for adjacency filtering doesn't sound very nice either.
20150919 14:52:46< zookeeper> i don't like the idea of a dedicated page, but i guess it might be warranted
20150919 14:53:10< celticminstrel> zookeeper: A quick grep shows a total of 25 locations in the source that call parse_direction.
20150919 14:53:34< celticminstrel> You could use these in [modify_unit] to set a unit's facing direction, I believe.
20150919 14:53:59< celticminstrel> You can certainly do that using the Lua API with a proxy unit.
20150919 14:54:20< celticminstrel> Oh, animations too.
20150919 14:54:49< celticminstrel> Admittedly, they're probably pretty useless in animations. But they will work there.
20150919 15:01:56< celticminstrel> For [filter_adjacent] on AbilitiesWML I was actually going to change it to say "This is a shorthand for..." referring you to the StandardUnitFilter page.
20150919 15:05:26-!- jinblack [~jinblack@169-231-98-254.wireless.ucsb.edu] has joined #wesnoth-dev
20150919 15:09:16-!- Shackra [~Jorge@186.177.2.148] has quit [Ping timeout: 265 seconds]
20150919 15:09:22-!- Shackra_ [~Jorge@186.177.2.148] has joined #wesnoth-dev
20150919 15:18:05-!- [Relic] [~Relic]@2602:306:33a3:6d30:dde2:558f:db7e:fdf9] has joined #wesnoth-dev
20150919 15:29:03-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150919 15:29:59< mattsc> celticminstrel: takes _much_ longer now with result: “115 out of 120 tests were correct.”
20150919 15:30:08< celticminstrel> Okay.
20150919 15:30:54< celticminstrel> Probably the other five just timed out. When I run it I have to increase the timeout for any of the tests to pass (I use -t30).
20150919 15:35:36< mattsc> running again with that setting ...
20150919 15:38:50-!- louis94 [~~louis94@109.129.229.117] has joined #wesnoth-dev
20150919 16:00:26-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has joined #wesnoth-dev
20150919 16:01:30-!- louis94 [~~louis94@109.129.229.117] has quit [Ping timeout: 260 seconds]
20150919 16:05:30-!- jinblack [~jinblack@169-231-98-254.wireless.ucsb.edu] has quit [Ping timeout: 268 seconds]
20150919 16:08:38-!- Kwandulin [~Miranda@p5B008173.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20150919 16:14:28-!- irker392 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20150919 16:14:28< irker392> wesnoth: Celtic Minstrel wesnoth:master 91ac90db111e / / (5 files in 2 dirs): Lua API: Add wesnoth.show_message_dialog() http://git.io/vnGUl
20150919 16:14:28< irker392> wesnoth: Celtic Minstrel wesnoth:master 6200f60c79ea / data/lua/ (wml-tags.lua wml-utils.lua): Split some utility functions from wml-tags.lua into a new file http://git.io/vnGU8
20150919 16:14:29< irker392> wesnoth: Celtic Minstrel wesnoth:master d4835b0157bc / / (7 files in 4 dirs): Port [message] ActionWML tag from C++ to Lua http://git.io/vnGU4
20150919 16:14:29< irker392> wesnoth: Celtic Minstrel wesnoth:master 376020097fbc / / (10 files in 4 dirs): Port [object] ActionWML tag from C++ to Lua http://git.io/vnGUB
20150919 16:14:30< irker392> wesnoth: Celtic Minstrel wesnoth:master e3fc258ce9f4 / changelog src/scripting/lua_gui2.cpp: Simplify wesnoth.show_message_dialog implementation http://git.io/vnGUR
20150919 16:14:32< irker392> wesnoth: Celtic Minstrel wesnoth:master 65256c61aa41 / src/savegame.cpp: Convert used items from old saves http://git.io/vnGU0
20150919 16:14:34< irker392> wesnoth: Celtic Minstrel wesnoth:master a93ce122ff88 / changelog data/lua/wml/object.lua: Add [found_item] ConditionalWML that checks if an [object] ActionWML has been ta http://git.io/vnGUu
20150919 16:14:36< irker392> wesnoth: Celtic Minstrel wesnoth:master 03580ed9a9ca / data/lua/ (wml-utils.lua wml/object.lua): Allow spaces in [object]id= and show error if invalid characters are present http://git.io/vnGUz
20150919 16:14:38< irker392> wesnoth: Celtic Minstrel wesnoth:master 907d527cc38f / / (6 files in 4 dirs): Lua API: Replace wesnoth.skipping_replay() with is_skipping_messages() and skip_ http://git.io/vnGUg
20150919 16:14:40< irker392> wesnoth: CelticMinstrel wesnoth:master e841252f19e6 / / (15 files in 6 dirs): Merge pull request #492 from CelticMinstrel/lua-message-object http://git.io/vnGU2
20150919 16:15:09-!- louis94 [~~louis94@109.129.229.117] has joined #wesnoth-dev
20150919 16:19:36-!- jinblack [~jinblack@ip184-187-174-107.sb.sd.cox.net] has joined #wesnoth-dev
20150919 16:24:15-!- jinblack [~jinblack@ip184-187-174-107.sb.sd.cox.net] has quit [Ping timeout: 255 seconds]
20150919 16:37:02< irker392> wesnoth: Charles Dang wesnoth:master e96f1c1bc11e / data/lua/backwards-compatibility.lua: Cleaned out implementation of legacy function http://git.io/vnGtD
20150919 16:55:05-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20150919 16:58:30-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 265 seconds]
20150919 16:58:30-!- wedge010 is now known as wedge009
20150919 17:11:46-!- gfgtdf [~chatzilla@f054130198.adsl.alicedsl.de] has joined #wesnoth-dev
20150919 17:13:36-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20150919 17:14:09< gfgtdf> celticminstrel: why did you change all those things in unit_filter.cpp? Afaik there are changed to read the vconfig variable only once to make filters perform better if they were applied to multiple units.
20150919 17:20:02-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20150919 17:22:59-!- louis94 [~~louis94@109.129.229.117] has quit [Quit: Konversation terminated!]
20150919 17:23:08-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20150919 17:23:23-!- louis94 [~~louis94@109.129.229.117] has joined #wesnoth-dev
20150919 17:28:58-!- louis94 [~~louis94@109.129.229.117] has quit [Ping timeout: 244 seconds]
20150919 17:30:55-!- oldlaptop [~quassel@50.36.238.180] has joined #wesnoth-dev
20150919 17:31:57-!- gfgtdf [~chatzilla@f054130198.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.92 [Firefox 40.0.3/20150826023504]]
20150919 17:34:16-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20150919 17:38:14-!- subliun [~quassel@unaffiliated/subliun] has quit [Ping timeout: 250 seconds]
20150919 17:41:09-!- subliun [~quassel@unaffiliated/subliun] has joined #wesnoth-dev
20150919 17:47:13< irker392> wesnoth: Charles Dang wesnoth:master 660647f7d8ce / data/lua/wml/ (message.lua object.lua): Strip newline whitespace http://git.io/vnGBH
20150919 18:11:57< celticminstrel> gfgtdf: Yes, I thought that was the reasoning for the unit filter change. Unfortunately, that optimization breaks the use of the $this_unit variable.
20150919 18:18:30< celticminstrel> vultraz: Since you're around, is there any particular reason against merging PR488 now?
20150919 18:35:04< celticminstrel> vultraz?
20150919 18:41:00-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has joined #wesnoth-dev
20150919 18:50:57< vultraz> celticminstrel: sorry was playing amnesia
20150919 18:52:00< celticminstrel> Ah.
20150919 18:52:39< vultraz> celticminstrel: well for one, you have diff artifacts in the changelog
20150919 18:53:18< celticminstrel> Aside from that.
20150919 18:53:23< celticminstrel> I already fixed that.
20150919 18:53:51< vultraz> tab spacing in abilities.cfg
20150919 18:53:55< celticminstrel> Eh?
20150919 18:54:10< vultraz> you have tab spacing there
20150919 18:54:24< vultraz> Also, is it not possible to have level take a comma separated list?
20150919 18:54:33< celticminstrel> Which level?
20150919 18:54:51< celticminstrel> For leadership?
20150919 18:54:58< celticminstrel> That would give the wrong result, I think.
20150919 18:55:30< vultraz> or at least an [or] block?
20150919 18:55:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20150919 18:55:38< vultraz> it seems weird to duplicate the code 5 times over
20150919 18:55:44< celticminstrel> No, each of those is a separate [leadership] because the value= is different.
20150919 18:56:17< vultraz> oh, ok
20150919 18:56:37-!- trewe [~trewe@2001:8a0:d101:7b01:626c:66ff:fe92:9b7c] has joined #wesnoth-dev
20150919 18:56:43-!- ancestral [~ancestral@203.sub-70-197-198.myvzw.com] has quit [Read error: Connection reset by peer]
20150919 18:56:53< vultraz> unit_abilities.cpp:380, is that index supposed to have quotes?
20150919 18:57:48< celticminstrel> Yes, as you can see at 363, i is a config.
20150919 18:58:56< vultraz> oh, ok
20150919 18:59:26< vultraz> Your changelog entry is also unclear what the auto stored wml variable is
20150919 18:59:49< celticminstrel> Oh right, zookeeper mentioned that too.
20150919 18:59:59< celticminstrel> Forgot to fix that.
20150919 19:01:47< vultraz> Is it deliberate that all the other levels of leadership have no description?
20150919 19:03:27< vultraz> Also, your math
20150919 19:03:38< vultraz> level=$($other_unit.level - 5) 4 3 2 1
20150919 19:03:47< vultraz> oh nvm
20150919 19:03:53< vultraz> it's units of a lower level
20150919 19:03:56< vultraz> so that's 0 - 4, ok
20150919 19:04:01 * vultraz cannot think right now >_>
20150919 19:07:50< vultraz> what does your change to attack.cpp do?
20150919 19:07:59< vultraz> make people face each other on attack?
20150919 19:14:39-!- Turuk [~Turuk@cpe-174-97-3-219.cinci.res.rr.com] has joined #wesnoth-dev
20150919 19:14:39-!- Turuk [~Turuk@cpe-174-97-3-219.cinci.res.rr.com] has quit [Changing host]
20150919 19:14:39-!- Turuk [~Turuk@wesnoth/forum-moderator/turuk] has joined #wesnoth-dev
20150919 19:19:22< celticminstrel> Yes vultraz, that's deliberate. In fact, that implementation is exactly the former {ABILITY_LEADERSHIP_LEVEL_5} except with the $other_unit stuff.
20150919 19:19:32< vultraz> ok
20150919 19:20:38< celticminstrel> The change to attack.cpp makes people face each other while checking an attack and before the attack actually starts. They were already set to face each other when the attack animation began, but unfortunately that was after the damage was calculated, so too late for the new backstab implementation.
20150919 19:21:26< vultraz> the backstab implementation check facing?
20150919 19:21:36< celticminstrel> Yes.
20150919 19:21:54< vultraz> that seems...a bit over-complicated
20150919 19:22:08< celticminstrel> If you're facing north, backstab is active when your ally is north of the enemy.
20150919 19:22:15< celticminstrel> I don't think it's complicated at all.
20150919 19:22:30< celticminstrel> You can see the impementation there.
20150919 19:22:36< celticminstrel> It's... five lines>
20150919 19:22:38< celticminstrel> ^?
20150919 19:22:43< vultraz> No I meant gameplay wise
20150919 19:22:56< vultraz> Won't this mean you now have to coordinate your backstab around enemy facing?
20150919 19:23:19< vultraz> Like, if you have a unit s of you and an ally s of that and you have the backstab, but the enemy is facing N
20150919 19:23:24< vultraz> will it work?
20150919 19:23:32< celticminstrel> Sorry, nine.
20150919 19:24:00< celticminstrel> No, the enemy's facing direction is irrelevant. Only the attacker's direction is checked.
20150919 19:24:22< celticminstrel> But even if it was relevant, the enemy is guaranteed to be facing you when the calculations take place.
20150919 19:24:32< vultraz> Ok, so no gameplay change?
20150919 19:24:36< celticminstrel> The enemy's facing is updated at the same time as yours, so that you're facing each other.
20150919 19:24:40< celticminstrel> Yeah, no gameplay change.
20150919 19:24:44< vultraz> Ok
20150919 19:25:38< celticminstrel> Okay, so, to recap, the only thing I need to fix is the changelog, right?
20150919 19:26:00< vultraz> And the tab indent in abilities.cfg
20150919 19:26:11< celticminstrel> Does that even matter?
20150919 19:26:25< celticminstrel> You're complaining that I indented with tabs, I suppose?
20150919 19:26:48< vultraz> Yeah. It doesn't exactly matter since it'll be caught with wmlindent later but I was just pointing it out
20150919 19:27:20< celticminstrel> Should I not bother fixing it then?
20150919 19:27:33< celticminstrel> It'll probably happen again whenever I edit config files.
20150919 19:28:00< celticminstrel> I don't think my editor has an option to auto-convert tabs to spaces, and even if it does, I wouldn't want to use it for general stuff.
20150919 19:28:16< celticminstrel> (I mean when pressing the tab key. It does have an option to convert them en masse.)
20150919 19:31:21< vultraz> Once you fix the changelog it should be ok to merge, I think
20150919 19:33:02< irker392> wesnoth: aquileia wesnoth:master 6dd554c97904 / projectfiles/VC9/ (wesnoth.vcproj wesnothd.vcproj): Update vcproj for 7875501, 20fafde & 0abde38 http://git.io/vnGFI
20150919 19:36:32< celticminstrel> vultraz: Regarding the changelog, is it better now?
20150919 19:37:44-!- Appleman1234 [~Appleman1@KD111239017006.au-net.ne.jp] has quit [Ping timeout: 272 seconds]
20150919 19:37:46< vultraz> Better, yes
20150919 19:37:56< vultraz> You forgot to update the directional syntax though
20150919 19:38:03< vultraz> I saw you changed it to dir:cw/ccw
20150919 19:38:19< vultraz> Changelog still says dir.cw/ccw
20150919 19:39:23< vultraz> And there's a stray dash
20150919 19:39:28< vultraz> "existing - operator"
20150919 19:39:54< celticminstrel> That's meant to be read as "existing minus operator".
20150919 19:40:29< vultraz> Quote it then
20150919 19:40:36< vultraz> "-" operator
20150919 19:40:49< celticminstrel> 'kay
20150919 19:41:46< celticminstrel> So with that fixed, it seems okay to merge?
20150919 19:43:43< vultraz> yes
20150919 19:44:14< celticminstrel> ...whoops, pushed to wrong repo. XD
20150919 19:44:33< irker392> wesnoth: Celtic Minstrel wesnoth:master 12c96a94d24d / / (7 files in 4 dirs): New autostored WML variable $other_unit for [filter_adjacent] http://git.io/vnGAk
20150919 19:44:35< irker392> wesnoth: Celtic Minstrel wesnoth:master 9940ee403667 / changelog src/map_location.cpp: New rotate operators for directions http://git.io/vnGAI
20150919 19:44:37< irker392> wesnoth: Celtic Minstrel wesnoth:master f54d6b5d28cb / data/ (47 files in 18 dirs): Rewrite leadership ability as a single macro using $other_unit http://git.io/vnGAL
20150919 19:44:39< irker392> wesnoth: Celtic Minstrel wesnoth:master 421b34d04a8e / src/unit_abilities.cpp: Add deprecation warning for backstab= http://git.io/vnGAt
20150919 19:44:41< irker392> wesnoth: Celtic Minstrel wesnoth:master fb1a84594ad5 / changelog src/unit_abilities.cpp: Enhance ability/weapon special [filter_adjacent] and [filter_adjacent_location] http://git.io/vnGAq
20150919 19:44:43< irker392> wesnoth: Celtic Minstrel wesnoth:master 4ca0aeb50656 / src/scripting/game_lua_kernel.cpp: Expose the extra filter unit to the Lua API http://git.io/vnGAm
20150919 19:44:45< irker392> wesnoth: Celtic Minstrel wesnoth:master 34b0a9ca6f4a / src/map_location.cpp: Change direction rotation syntax to not be confused with variable substitution http://git.io/vnGAY
20150919 19:44:47< irker392> wesnoth: Celtic Minstrel wesnoth:master 7ca65c819ecd / changelog data/core/macros/abilities.cfg: Clarify changelog http://git.io/vnGAO
20150919 19:44:49< irker392> wesnoth: Celtic Minstrel wesnoth:master e74c119b9b2b / / (55 files in 22 dirs): Merge pull request #488 from CelticMinstrel/backstab http://git.io/vnGA3
20150919 19:52:50< vultraz> Now, just the looping tags left
20150919 19:54:16-!- jinblack [~jinblack@ip184-187-174-106.sb.sd.cox.net] has joined #wesnoth-dev
20150919 20:01:16< celticminstrel> Yup.
20150919 20:01:32< celticminstrel> And break/continue, if you weren't counting those as looping tags.
20150919 20:07:04< irker392> wesnoth: Wedge009 wesnoth:master 9bcfb8dff583 / data/core/encyclopedia/geography.cfg: Correct then/than mix-up. http://git.io/vnZeM
20150919 20:07:06< irker392> wesnoth: Wedge009 wesnoth:master 27f6aea2150f / data/campaigns/Liberty/ (scenarios/03_A_Strategy_Of_Hope.cfg utils/utils.cfg): Correcting born -> borne http://git.io/vnZeD
20150919 20:07:08< irker392> wesnoth: Wedge009 wesnoth:master f74796809aa4 / data/campaigns/Sceptre_of_Fire/scenarios/6_Towards_the_Caves.cfg: Correct its vs it's http://git.io/vnZey
20150919 20:07:10< irker392> wesnoth: Wedge009 wesnoth:master a5aed635eb65 / data/campaigns/The_Rise_Of_Wesnoth/scenarios/21_The_Plan.cfg: Add missing full stop. http://git.io/vnZeS
20150919 20:07:12< irker392> wesnoth: Wedge009 wesnoth:master f2eadeb5511c / data/campaigns/Legend_of_Wesmere/scenarios/chapter2/07_Elves_Last_Stand.cfg: Grammatical corrections in Elves' Last Stand. http://git.io/vnZe9
20150919 20:07:14< irker392> wesnoth: Wedge009 wesnoth:master 48c2098902a7 / data/campaigns/Legend_of_Wesmere/scenarios/chapter3/09_Bounty_Hunters.cfg: maybe = perhaps vs may be = might be http://git.io/vnZeH
20150919 20:07:16< irker392> wesnoth: Wedge009 wesnoth:master f3d23218a8e6 / data/campaigns/Legend_of_Wesmere/scenarios/chapter4/15_The_Treaty.cfg: Separated 'can not' is more common where 'not' forms part of a construction with http://git.io/vnZeQ
20150919 20:07:18< irker392> wesnoth: Wedge009 wesnoth:master 2db2e81e7c3f / data/campaigns/Legend_of_Wesmere/scenarios/chapter5/19_Costly_Revenge.cfg: In this context, referring to the city forming the Saurians' central city, capit http://git.io/vnZe7
20150919 20:07:20< irker392> wesnoth: Wedge009 wesnoth:master 73b10f315a02 / data/ (3 files in 3 dirs): Separated 'can not' is more common where 'not' forms part of a construction with http://git.io/vnZe5
20150919 20:07:22< irker392> wesnoth: Wedge009 wesnoth:master 52f8080a9285 / doc/manual/ (manual.en.html manual.en_GB.html): Some grammatical corrections. http://git.io/vnZed
20150919 20:07:24< irker392> wesnoth: Wedge009 wesnoth:master 531c1d9dbd0e / data/core/units/humans/Loyalist_Royal_Guard.cfg: Surely 'capital' is referring to Weldyn, not some government building within it, http://git.io/vnZeF
20150919 20:07:26< irker392> wesnoth: Wedge009 wesnoth:master 3a0dd3ed2442 / data/campaigns/ (4 files in 3 dirs): Add missing apostrophe in contraction of 'let us'. http://git.io/vnZeb
20150919 20:07:28< irker392> wesnoth: Wedge009 wesnoth:master 496e5b5437a0 / data/campaigns/Sceptre_of_Fire/scenarios/6_Towards_the_Caves.cfg: Repositioned 'south' to make more sense. http://git.io/vnZeN
20150919 20:07:30< irker392> wesnoth: Wedge009 wesnoth:master c9fbe1315607 / data/campaigns/Son_Of_The_Black_Eye/scenarios/01_End_of_Peace.cfg: Make hyphenation consistent http://git.io/vnZeA
20150919 20:07:32< irker392> wesnoth: Wedge009 wesnoth:master 0b9fbb9678bf / data/campaigns/Son_Of_The_Black_Eye/scenarios/05_To_the_Harbor_of_Tirigaz.cfg: Punctuation, for clarity on who is heading to Tirigaz. http://git.io/vnZex
20150919 20:07:34< irker392> wesnoth: Wedge009 wesnoth:master 45dcb6893563 / data/campaigns/Son_Of_The_Black_Eye/scenarios/09_Shan_Taum_the_Smug.cfg: Grammatical correction. http://git.io/vnZep
20150919 20:07:36< irker392> wesnoth: Wedge009 wesnoth:master c355cd95032f / data/campaigns/Son_Of_The_Black_Eye/scenarios/16_The_Coward.cfg: Add missing 'he' to make more sense of the sentence. http://git.io/vnZeh
20150919 20:07:38< irker392> wesnoth: Wedge009 wesnoth:master 210ba16c6e76 / data/campaigns/The_Rise_Of_Wesnoth/scenarios/ (04a_The_Swamp_of_Esten.cfg 04b_The_Midlands.cfg): Grammar correction: 'have taken' vs 'have took' http://git.io/vnZej
20150919 20:07:40< irker392> wesnoth: Wedge009 wesnoth:master d0006da05d3d / data/campaigns/ (Dead_Water/utils/items.cfg Northern_Rebirth/scenarios/07a_Settling_Disputes.cfg): Remove redundant wording. http://git.io/vnZve
20150919 20:07:42< irker392> wesnoth: Wedge009 wesnoth:master 529bb982a77e / doc/manual/manual.en.html: Official name is just OS X now. http://git.io/vnZvv
20150919 20:07:44< irker392> wesnoth: Wedge009 wesnoth:master 971243975bb0 / data/campaigns/Northern_Rebirth/scenarios/04_Clearing_the_Mines.cfg: 'Most vile' vs 'most vilest' + other minor changes to make more sense of sentenc http://git.io/vnZvf
20150919 20:07:46< irker392> wesnoth: Wedge009 wesnoth:master c55cd29297bd / data/campaigns/Northern_Rebirth/scenarios/05a_01_The_Pursuit.cfg: Added missing comma. http://git.io/vnZvJ
20150919 20:07:48< irker392> wesnoth: Wedge009 wesnoth:master e14333bad0df / data/campaigns/Northern_Rebirth/scenarios/05a_02_Dealings.cfg: Correcting a typo. http://git.io/vnZvU
20150919 20:07:50< irker392> wesnoth: Wedge009 wesnoth:master 53191251fc11 / data/campaigns/ (3 files in 2 dirs): Remove redundant wording. http://git.io/vnZvT
20150919 20:07:52< irker392> wesnoth: Wedge009 wesnoth:master 6043550999de / data/campaigns/Northern_Rebirth/scenarios/08b_Ray_of_Hope.cfg: Remove duplicate word. http://git.io/vnZvk
20150919 20:07:54< irker392> wesnoth: Wedge009 wesnoth:master bcece91e9a2c / data/campaigns/Northern_Rebirth/scenarios/09a_Introductions.cfg: 'guys' seems excessively colloquial here. http://git.io/vnZvI
20150919 20:07:56< irker392> wesnoth: Wedge009 wesnoth:master 464646978d87 / data/campaigns/Northern_Rebirth/scenarios/09b_Judgment.cfg: Correct nonsensical text. http://git.io/vnZvt
20150919 20:07:58< irker392> wesnoth: Wedge009 wesnoth:master 473548cc89ed / data/campaigns/Northern_Rebirth/scenarios/10a_Stolen_Gold.cfg: Sentence structure may be a bit confusing but it seems to follow the pattern of  http://git.io/vnZvq
20150919 20:08:00< irker392> wesnoth: Wedge009 wesnoth:master 87acd9362669 / data/campaigns/Northern_Rebirth/scenarios/13a_Showdown.cfg: Added missing comma. http://git.io/vnZvm
20150919 20:08:02< irker392> wesnoth: Wedge009 wesnoth:master 643d55e46704 / data/campaigns/ (3 files in 2 dirs): 'x and I' vs 'me and x' http://git.io/vnZvY
20150919 20:08:04< irker392> wesnoth: Wedge009 wesnoth:master e1c1b5c4befa / data/campaigns/Northern_Rebirth/scenarios/14a_Epilogue.cfg: 'gonna' sounds rather informal for Eryssa + 'Father' is Morvin's title, not his  http://git.io/vnZvO
20150919 20:08:06< irker392> wesnoth: Wedge009 wesnoth:master f42e386d364e / data/campaigns/Northern_Rebirth/scenarios/10a_Stolen_Gold.cfg: Slight re-arrangement of 'than' after discussion with vultraz and shadowm. http://git.io/vnZv3
20150919 20:08:08< irker392> wesnoth: Wedge009 wesnoth:master a6fc91090e87 / data/core/encyclopedia/geography.cfg: Turns out it wasn't a then/than mix-up. Still needed correcting, though. http://git.io/vnZvs
20150919 20:08:10< irker392> wesnoth: Charles Dang wesnoth:master bec40a3c3922 / / (37 files in 19 dirs): Merge pull request #483 from Wedge009/Spelling_Corrections http://git.io/vnZvG
20150919 20:08:21< vultraz> sooo many corrections
20150919 20:11:43-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: mattsc]
20150919 20:17:30< celticminstrel> wedge009: Unguessable is totally a word. Also, southwest, gameplay, username, setup, hotkeys, hitpoints, onto are all more correct as compounds. I think hotseat is too, but less certain on that. I also think "goto-mode" is better than "go-to-mode". Then you misspelled "dextrous" (it was correct to begin with). I don't think the change of "build up an army" to "build an army" was necessary (though both ways work for me).
20150919 20:18:47< vultraz> He said the non-compounding was for the en_GB version only
20150919 20:19:26< vultraz> As for unguessable, I think his change sounds better
20150919 20:19:46< celticminstrel> Hmm, maybe, not sure on that, but I think most of them are better as compounds even for en_GB.
20150919 20:20:21< celticminstrel> The main possible exceptions I can see are southwest and onto; the others are all game/computing terms that are normally compound as far as I know.
20150919 20:20:40< celticminstrel> I don't mind the unguessable change, just objecting to the justification.
20150919 20:20:40 * vultraz shrugs
20150919 20:20:51< celticminstrel> Mind if I go through this and make line comments?
20150919 20:20:54< vultraz> I don't know what's used in en_Gb, so I trust his judgement 
20150919 20:20:58< vultraz> But sure
20150919 20:41:24-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20150919 20:44:02-!- Appleman1234 [~Appleman1@KD118156254118.au-net.ne.jp] has joined #wesnoth-dev
20150919 20:51:30-!- louis94 [~~louis94@109.129.229.117] has joined #wesnoth-dev
20150919 20:52:51< celticminstrel> Hm, looks like most of those compounds I mentioned were later reverted.
20150919 20:53:06< celticminstrel> With the exception of southwest (which I can believe to be a real GB difference).
20150919 20:53:09< shadowm> 07:54:06 <Aginor> The root cause is that a number of places draws a copy of the screen onto it self, this seems to be broken on some hardware
20150919 20:53:19< shadowm> Aginor: Really? That doesn't sound right.
20150919 20:54:08< shadowm> Is the source and destination data supposed to be identical in this case?
20150919 20:54:27< shadowm> Also, for reference:
20150919 20:54:29< shadowm>     model name      : Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz
20150919 20:54:37< celticminstrel> I spoke too soon.
20150919 20:55:21-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20150919 21:00:20-!- agentcarr16 [ae59818c@gateway/web/freenode/ip.174.89.129.140] has joined #wesnoth-dev
20150919 21:00:49< celticminstrel> Well, I've finished my notes, anyway.
20150919 21:02:14-!- agentcarr16 [ae59818c@gateway/web/freenode/ip.174.89.129.140] has quit [Client Quit]
20150919 21:12:44< zookeeper> Turuk, i hope no one's seriously suggesting renaming chantal because there's cookware by the same name somewhere in the world.
20150919 21:13:33< shadowm> Hahaha what.
20150919 21:14:30< vultraz> wat
20150919 21:15:33< zookeeper> i couldn't tell from the thread whether it was a joke or not
20150919 21:25:51< irker392> wesnoth: Charles Dang wesnoth:master 205b82da697c / data/gui/default/window/title_screen.cfg: Pluralize campaigns main menu item http://git.io/vnZ3G
20150919 21:43:59< celticminstrel> wedge009: I'd like responses to my comments on http://git.io/vnZvG
20150919 21:54:18< mattsc> celticminstrel: running the script with -t30 still only gives me 115 out of 120 successes
20150919 21:55:53< mattsc> errors.log contains 3 errors messages, they are all ‘unknown scenario’ errors.
20150919 21:56:51< mattsc> 'filter_this_unit_wml’, 'filter_this_unit_tl' and 'filter_this_unit_fai'
20150919 21:57:54< mattsc> I assume you have not added those yet?
20150919 21:59:01< celticminstrel> They should be at data/core/test/filter_this_unit.cfg...
20150919 21:59:22< shadowm> data/test/test
20150919 21:59:23< mattsc> celticminstrel: oh, argh!
20150919 21:59:45< shadowm> I mean data/test/scenarios
20150919 21:59:56< celticminstrel> shadowm: Yeah, that.
20150919 22:02:13< celticminstrel> zookeeper: Can I ask you to look over PR493? (Mainly break/continue.)
20150919 22:04:05-!- mjs-de [~mjs-de@f049239245.adsl.alicedsl.de] has quit [Remote host closed the connection]
20150919 22:04:37< mattsc> celticminstrel: I’m not going to tell you what I did  :P  but I just restarted the tests; I bet they will work now
20150919 22:04:48< vultraz> What's with the crazy indent in some of the loops?
20150919 22:05:03< celticminstrel> vultraz: Huh?
20150919 22:06:20< vultraz> 391, 426, etc
20150919 22:08:19< zookeeper> celticminstrel, well i can't tell you anything regarding the code. if they according to the spec we discussed, then it's all good as far as i'm concerned.
20150919 22:08:25< zookeeper> +work
20150919 22:08:46< zookeeper> that is, if break/continue work as one would expect them to work
20150919 22:08:54< vultraz> hm
20150919 22:09:02< vultraz> does handle_event_commands now get a default scope?
20150919 22:10:11< celticminstrel>  What do you mean?
20150919 22:10:47< vultraz> Just wondering if we always have to call it with a scope from now on
20150919 22:11:00< celticminstrel> I can add that.
20150919 22:11:12< celticminstrel> Default shoud be "plain".
20150919 22:11:16< celticminstrel> ^"l
20150919 22:17:27< celticminstrel> vultraz: Very good question on the extra indent. I have no idea why it's there.
20150919 22:17:44< mattsc> celticminstrel: yes, all tests working now
20150919 22:18:05< celticminstrel> zookeeper: I suggest looking at this in particular: https://github.com/wesnoth/wesnoth/pull/493/files#diff-f926d696584c12ac59237a45606e23f3R1
20150919 22:18:59< celticminstrel> That's the unit tests, so it'll show you how the tags are used.
20150919 22:21:50-!- travis-ci [~travis-ci@ec2-54-145-151-113.compute-1.amazonaws.com] has joined #wesnoth-dev
20150919 22:21:51< travis-ci> wesnoth/wesnoth#7418 (master - 205b82d : Charles Dang): The build has errored.
20150919 22:21:51< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/81201282
20150919 22:21:51-!- travis-ci [~travis-ci@ec2-54-145-151-113.compute-1.amazonaws.com] has left #wesnoth-dev []
20150919 22:27:24< irker392> wesnoth: mattsc wesnoth:master 6a3a346e4194 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: Xcode project: do not warn about unused local typedefs http://git.io/vnZRY
20150919 22:30:27-!- Catoptromancy [~Cato@unaffiliated/catoptromancy] has quit [Remote host closed the connection]
20150919 22:52:38-!- [Relic] [~Relic]@2602:306:33a3:6d30:dde2:558f:db7e:fdf9] has quit [Quit: I press the magic X and all the weirdos go away!]
20150919 22:53:01-!- trewe [~trewe@2001:8a0:d101:7b01:626c:66ff:fe92:9b7c] has quit [Quit: quit]
20150919 22:58:00-!- jinblack [~jinblack@ip184-187-174-106.sb.sd.cox.net] has quit [Remote host closed the connection]
20150919 23:05:43< vultraz> celticminstrel: I think I might have noticed a bug in your object impl
20150919 23:05:57< celticminstrel> vultraz: Okay, what is it?
20150919 23:05:59< vultraz> If the filter doesnt match, won't it still look for the unit at x1y1?
20150919 23:06:16< celticminstrel> I think so?
20150919 23:06:18< celticminstrel> Oh wait.
20150919 23:06:22< celticminstrel> I don't know.
20150919 23:06:27-!- travis-ci [~travis-ci@ec2-54-147-48-205.compute-1.amazonaws.com] has joined #wesnoth-dev
20150919 23:06:28< travis-ci> wesnoth/wesnoth#7419 (master - 6a3a346 : mattsc): The build passed.
20150919 23:06:28< travis-ci> Build details : https://travis-ci.org/wesnoth/wesnoth/builds/81207491
20150919 23:06:28-!- travis-ci [~travis-ci@ec2-54-147-48-205.compute-1.amazonaws.com] has left #wesnoth-dev []
20150919 23:06:28-!- iceiceice [~chris@ext-74.ias.edu] has joined #wesnoth-dev
20150919 23:06:28-!- iceiceice [~chris@ext-74.ias.edu] has quit [Changing host]
20150919 23:06:28-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20150919 23:06:29< vultraz> line 21 says if not unit, and IIRC unit will still be nil if get_units fails
20150919 23:08:39< vultraz> meaning if a filter is specified and won't match, it will always match to x1,y1... and since this will usually match (moveto events and such), the object will still be applied even if it's the wrong unit
20150919 23:09:56< mattsc> vultraz: could you point me to that line, please?
20150919 23:10:29< vultraz> 16 - 26 in data/lua/wml/object.lua
20150919 23:11:39< mattsc> wesnoth.get_units() returns an empty table when no units matches the filer.
20150919 23:11:59< mattsc> Thus, wesnoth.get_units(filter)[1] is nil when no matching unit is found.
20150919 23:12:15< mattsc> Is that what you meant?
20150919 23:12:18< vultraz> Yes
20150919 23:12:26< vultraz> Meaning line 21 then passes
20150919 23:12:47< vultraz> I'm thinking merging to two blocks into an if:else statement would fix
20150919 23:12:49< vultraz> this
20150919 23:13:19< mattsc> oh, I see what you are saying.
20150919 23:13:26< mattsc> Yes, that’s correct.
20150919 23:13:44< mattsc> I had misunderstood what you meant ...
20150919 23:14:23< mattsc> you could fix it as easily as:
20150919 23:14:33< mattsc> if filter and (not unit) then …
20150919 23:14:38< mattsc> in l.21
20150919 23:14:56< mattsc> but there certainly are other options
20150919 23:15:45< celticminstrel> Well, if you know how to fix it, go ahead. :)
20150919 23:15:55< celticminstrel> I only tested it with filters, sorry.
20150919 23:16:08< celticminstrel> (With the Staff of Speed mainline macro, to be specific.)
20150919 23:16:22< vultraz> I'll use the if else, I think
20150919 23:16:56< mattsc> celticminstrel: on a different topic, is there a reason why your removal of -Werror should not be ported to 1.12?
20150919 23:17:34< mattsc> As in, did you avoid doing it, or did you just not do it yet?
20150919 23:18:12< celticminstrel> mattsc: I have no plans to compile 1.12, so I don't really care either way.
20150919 23:18:15< mattsc> I’ve made that change to the projectfile and it compiles fine with that in Xcode 7, so I’ll commit it unless you tell me that I should not.
20150919 23:18:22< mattsc> ok, thanks
20150919 23:19:16< irker392> wesnoth: mattsc wesnoth:1.12 8e7b90dc4369 / projectfiles/Xcode/Wesnoth.xcodeproj/project.pbxproj: XCode project: remove -Werror setting http://git.io/vnZgV
20150919 23:19:30< irker392> wesnoth: Charles Dang wesnoth:master 9ea9c88356fe / data/lua/wml/object.lua: Fixed bug where unit at x1,y1 would be considered for object even if filter fail http://git.io/vnZgo
20150919 23:21:03< mattsc> vultraz: that won’t work …
20150919 23:21:15< celticminstrel> You sure?
20150919 23:21:22< mattsc> filter in line 19 is not defined
20150919 23:21:38< mattsc> so that line will always return all units
20150919 23:21:42< vultraz> oh, derp
20150919 23:21:53< celticminstrel> Oh.
20150919 23:21:59< celticminstrel> Heh, I see.
20150919 23:22:07< vultraz> Yeah I guess it should be in a variable after all :|
20150919 23:23:52< irker392> wesnoth: Charles Dang wesnoth:master 45aa0d4ae84e / data/lua/wml/object.lua: Fixup 9ea9c88356fe http://git.io/vnZgA
20150919 23:41:21-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20150919 23:53:15< vultraz> I wonder if we could port the core conditional tags to Lua
20150919 23:53:22< vultraz> [have_unit] etc
20150919 23:54:59-!- jinblack [~jinblack@128.111.48.6] has joined #wesnoth-dev
20150919 23:55:53< shadowm> What would the purpose be?
20150919 23:56:34< zookeeper> to port them to lua! haven't you heard that that's good? :]
20150919 23:56:41< celticminstrel> Pfft.
20150919 23:58:30< vultraz> Basically, yeah
20150919 23:58:51< vultraz> I thought the goal was to move everything to lua anyway
20150919 23:58:58< celticminstrel> I would say there's not much point to port things just for the sake of porting things though.
20150919 23:59:03< celticminstrel> ^+ing
20150919 23:59:09< shadowm> ^
20150919 23:59:15< vultraz> but why did you do message and object, then :P
20150919 23:59:27< celticminstrel> So that I could implement break and continue more easily. :P
--- Log closed Sun Sep 20 00:00:28 2015