--- Log opened Wed Dec 27 00:00:17 2017
20171227 00:01:57< Ravana_> when I now manually deleted last byte from credentials file, I can get it
20171227 00:02:33< Ravana_> once
20171227 00:03:25< vultraz> an empty file always has 1 character in ir
20171227 00:03:27< vultraz> it
20171227 00:03:30< vultraz> for some reason
20171227 00:11:42-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20171227 00:12:28-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20171227 00:12:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Read error: Connection reset by peer]
20171227 00:14:00-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20171227 00:18:17-!- stikonas_ is now known as stikonas
20171227 00:24:30< Ravana_> seems 1.13.10 client doesn't receive message that user has been banned, it says waiting for server
20171227 00:25:03-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has quit [Ping timeout: 248 seconds]
20171227 00:27:00-!- aeth [~Michael@wesnoth/umc-dev/developer/aethaeryn] has joined #wesnoth-dev
20171227 00:30:05-!- Greg-Boggs [~greg_bogg@c-73-11-32-127.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20171227 00:30:54< Ravana_> now I tried logging in with different name - and now the situation is that each login attempt appends ~20 more bytes to credentials file, and password prompt starts as empty
20171227 00:31:04-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has joined #wesnoth-dev
20171227 01:06:01< Ravana_> https://github.com/wesnoth/wesnoth/issues/2327
20171227 01:07:21< Ravana_> 1.13 bans command also shows that it has 2 automatic "login attempts exceeded" bans still remaining with end time in 2016, so looks like it has been broken for a while
20171227 01:11:19-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 272 seconds]
20171227 01:12:51-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has quit [Remote host closed the connection]
20171227 01:14:02-!- Greg-Boggs [~greg_bogg@c-73-11-32-127.hsd1.or.comcast.net] has joined #wesnoth-dev
20171227 01:21:59-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20171227 01:33:22-!- Greg-Boggs [~greg_bogg@c-73-11-32-127.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20171227 01:36:21-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 264 seconds]
20171227 01:40:42-!- Greg-Boggs [~greg_bogg@c-73-11-32-127.hsd1.or.comcast.net] has joined #wesnoth-dev
20171227 01:46:57-!- Greg-Boggs [~greg_bogg@c-73-11-32-127.hsd1.or.comcast.net] has quit [Remote host closed the connection]
20171227 01:48:43-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has joined #wesnoth-dev
20171227 01:52:29-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has quit [Remote host closed the connection]
20171227 01:55:20-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20171227 02:04:23-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has joined #wesnoth-dev
20171227 02:34:46-!- gfgtdf [~chatzilla@x4e3681c3.dyn.telefonica.de] has quit [Quit: ChatZilla 0.9.93 [Firefox 52.5.2/20171206101620]]
20171227 02:44:29-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has quit [Remote host closed the connection]
20171227 02:49:35-!- irker761 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20171227 03:10:52-!- vultraz [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20171227 03:45:14-!- Bonobo [~Bonobo@60-241-159-43.tpgi.com.au] has joined #wesnoth-dev
20171227 04:00:43-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20171227 04:00:54-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20171227 04:15:14-!- celmin|StarWars is now known as celticminstrel
20171227 04:40:42-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20171227 04:40:53-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20171227 04:46:22-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has joined #wesnoth-dev
20171227 04:54:14-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has quit [Remote host closed the connection]
20171227 05:01:09-!- celticminstrel is now known as celmin|sleep
20171227 05:19:01-!- vultraz [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20171227 07:02:07-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20171227 07:22:09-!- irker386 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20171227 07:22:09< irker386> wesnoth/wesnoth:master Gregory A Lundberg 5b5e7ed4ac Issue 2327 - Fix kban syntax error AppVeyor: All builds passed
20171227 08:02:16-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has joined #wesnoth-dev
20171227 08:27:10-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20171227 08:27:17-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20171227 09:04:44-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20171227 09:09:04-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20171227 10:07:05-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has quit [Ping timeout: 240 seconds]
20171227 10:17:41-!- vultraz [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20171227 10:22:14-!- irker386 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20171227 10:22:40-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has joined #wesnoth-dev
20171227 10:37:05-!- mjs-de [~mjs-de@185.22.140.20] has joined #wesnoth-dev
20171227 10:59:23-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20171227 11:01:05-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20171227 11:01:05-!- wedge010 is now known as wedge009
20171227 11:13:55-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has quit [Ping timeout: 256 seconds]
20171227 11:32:10-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20171227 11:57:23-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has joined #wesnoth-dev
20171227 13:10:04-!- Oebele [~quassel@92-110-163-84.cable.dynamic.v4.ziggo.nl] has quit [Remote host closed the connection]
20171227 13:50:22< vn971> Hm... Now I have a weird question. Can I access a WML macro from Lua? For example, I want to use WEAPON_SPECIAL_BACKSTAB macro inside Lua. How do I do it?
20171227 13:53:57< vn971> Or.. What really is a WEAPON_SPECIAL_BACKSTAB?  I would say it's a structured text. Like WML. So, it could not only be shared as a WML macro expansion, but maybe somehow else? Like a variable?
20171227 13:53:57< vn971> I wonder if anyone proposed to make well-formatted correct macros be variables.
20171227 13:55:33< DeFender1031> macros could probably be reimplemented as lua functions, but due to the nature of preprocessing, it isn't really possible to use WML macros from lua
20171227 13:56:06< Ravana_> it might be possible to create recursive function that would encode any wml to Lua table
20171227 13:56:07< vn971> DeFender1031: well yes, using directly is not possible.
20171227 13:56:52< DeFender1031> Ravana_, sure, I can think of a number of hacky ways to write a macro-to-lua converter function or tag or something
20171227 13:57:03< vn971> but maybe the WEAPON_SPECIAL_BACKSTAB should, conceptually, be a variable, not a macro?
20171227 13:57:18< vn971> the macro could then be defined as something that just expands the variable.
20171227 13:57:24< DeFender1031> vn971, how would you envision that as working?
20171227 13:58:57< vn971> DeFender1031: well what's the problem?
20171227 13:58:57< vn971> local backstab_special = T.something {***}
20171227 13:58:57< vn971> wesnoth.set_variable(***)
20171227 13:58:57< vn971> #define WEAPON_SPECIAL_BACKSTAB  \n $backstab_special #enddef
20171227 13:59:38-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20171227 13:59:52< vn971> (first 2 lines can be written as one assignment, if desired)
20171227 14:04:35< DeFender1031> table variables can't be expanded like that
20171227 14:04:42< DeFender1031> you could probably do it with [insert_tag] though
20171227 14:04:45< vn971> I am not sure about the memory impact. WML Macros can probably be thrown away (from memory) once everything is expanded. Variables should probably stay (and eat memory).
20171227 14:05:02< zookeeper> "accessing a macro from Lua" doesn't really make much sense to me.
20171227 14:05:15< JyrkiVesterinen> Memory usage is dominated by graphics anyway. I wouldn't worry about memory usage of WML and Lua.
20171227 14:05:33< vn971> zookeeper: of course not. But this is not the real goal. I'm just trying to understand what "WEAPON_SPECIAL_BACKSTAB" should be, conceptually.
20171227 14:05:44< vn971> zookeeper: I want to used the backstab itself in Lua, that's all.
20171227 14:07:14< zookeeper> mmkay. i suppose currently if your lua code resides in a .cfg file then you can simply have it preprocessed (just make sure the part of lua code you call the macro from isn't in <<>> or whatever), but you can't have .lua files be preprocessed?
20171227 14:19:28< vn971> zookeeper: lua files can be preprocessed in the following way:
20171227 14:19:28< vn971> -- << file_name.lua  it's just a textual comment for backtrace
20171227 14:19:28< vn971> print("lua")
20171227 14:19:28< vn971> -- >>
20171227 14:20:56< vn971> zookeeper: but even if lua would be inside cfg, I don't see how a macro would help. Macro is a text expansion, not code. So I'd just have square brackets like `[ability]` in Lua code, which won't help.
20171227 14:21:20< vn971> (if I get it correctly at least.)
20171227 14:21:53< zookeeper> ok... in that case i guess i don't see why you'd be interested in being able to access macro definitions from lua in the first place. what would you do with them?
20171227 14:24:19< vn971> zookeeper: I don't want macro themselves. I want backstab specifically. How can I use the default "backstab" implementation in Lua?
20171227 14:25:35< vn971> zookeeper: if we make WEAPON_SPECIAL_BACKSTAB be conceptually a variable, I would be able to do that. Otherwise I have to copy-paste code and transform WML to Lua manually.
20171227 14:27:23< zookeeper> i don't know what you mean by "using WML (the backstab implementation) in lua" in the first place.
20171227 14:27:35< vn971> zookeeper: sorry for my confusing first question (topic starter). Please read from the second onwards, where I ask if current macros should be defined as variables instead.
20171227 14:28:05< vn971> zookeeper: disregard that first question please.
20171227 14:28:36< zookeeper> okay
20171227 14:29:45< vn971> So.. Should backstab, as well as other simple WML things be re-defined as variables instead? When needed.
20171227 14:29:50< zookeeper> so would you basically want some way, from lua, to ask the WML preprocessor for the contents (either plain, or recursively preprocessed) of a given macro as a variable/table/whatever?
20171227 14:30:57< vn971> zookeeper: no. I want to change backstab itself. So the base would be a variable, and macros would only left out for compatibility reasons. A macro would be defined from the variable then.
20171227 14:31:06< celmin|sleep> vn971: [lua]code=<<local backstab = helper.get_child((...), "damage")>> [args]{WEAPON_SPECIAL_BACKSTAB}[/args][/lua]
20171227 14:31:52< vn971> celmin|sleep: nice.
20171227 14:32:55< zookeeper> well okay, if you want to suggest changing something utterly fundamental as how WML macros work or are in practise used, then i think you need to come up with a pretty good rationale beyond wanting to easily do something with backstab in lua :P
20171227 14:33:07< vn971> celmin|sleep: since it's easy enough, does that mean that we better keep those specials (and other things) be macros?
20171227 14:33:16< celmin|sleep> Particularly when it's something that can already be done. :P
20171227 14:34:19< Ravana_> so helper.get_child accepts string in place of wml object as well?
20171227 14:34:21< vn971> zookeeper: I don't propose to change anything except mindset. Not a single C++ LoC to be done at least. Wait a couple of minutes please, I'll write.
20171227 14:34:21< celmin|sleep> I do think the specials and abilities should be turned into something different from macros, but it's not really a priority.
20171227 14:34:25< AI0867> vn971: from what I gather, you want to access "backstab itself"
20171227 14:34:35< celmin|sleep> Ravana_: Uh, (...) in this case is a WML object. :P
20171227 14:34:48< Ravana_> how?
20171227 14:34:49< vn971> AI0867: yes, it's definition as a table to be used in Lua.
20171227 14:34:53< AI0867> vn971: what you need to understand is that "backstab" is just a piece of text (in the form of WML) that is essentially pasted into every unit that uses it
20171227 14:35:00< AI0867> there is no global "backstab object"
20171227 14:35:04< celmin|sleep> Because that's what's passed into the function.
20171227 14:35:16< celmin|sleep> The code from the code attribute is compiled into a function.
20171227 14:35:21-!- celmin|sleep is now known as celticminstrel
20171227 14:35:31< celticminstrel> And that function is called with the [args] WML table as its only parameter.
20171227 14:35:37< JyrkiVesterinen> Specifically https://github.com/wesnoth/wesnoth/blob/a31858124c4fe4d3e83e733c4f614c76af6b78e7/data/core/macros/abilities.cfg#L280-L308
20171227 14:35:38< AI0867> in that case, what you should probably do is use [set_variables] to store it into a WML var and use that from lua
20171227 14:35:39< vn971> AI0867: yes, exactly. And that is the problem, too. Would it be better (or would it not to) for backstabe to be a correct WML table instead?
20171227 14:35:39< celticminstrel> That's what the [lua] tag does.
20171227 14:35:58< vn971> AI0867: you can keep WML tables as variables, and accessing it would be easy from both lua and WML.
20171227 14:36:04< celticminstrel> AI0867: Yeah that would work, but IMO the way I posted is easier.
20171227 14:36:22< celticminstrel> Unless you need to access it both from WML and Lua.
20171227 14:36:45< vn971> AI0867: set_variables seem a bit hard because we can't write  `variable= [tag]`  in one line.
20171227 14:36:56< celticminstrel> ???
20171227 14:37:07< vn971> or can I ?
20171227 14:37:38< celticminstrel> Something like... [set_variables]variable=backstab [value]{WEAPON_SPECIAL_BACKSTAB}[/value][/set_variables]?
20171227 14:37:56< celticminstrel> What would variable=[tag] even mean?
20171227 14:38:13< vn971> celticminstrel: ah, I didn't know that value can be a tag [value].
20171227 14:38:23< zookeeper> either you're using unusual terminology for referring to things, or you're confused about how WML works. i don't know which :p
20171227 14:38:27< vn971> I thought it's mandatory to write `value=`
20171227 14:38:41< celticminstrel> vn971: No, uh...
20171227 14:38:47< celticminstrel> [set_variable] requires value=
20171227 14:38:53< celticminstrel> [set_variables] requires [value]
20171227 14:39:17< celticminstrel> At least I think that's right?
20171227 14:39:35< celticminstrel> AFAIK, [set_variables] is what you need when setting a container variable, even if it's not an array.
20171227 14:41:41< zookeeper> yes
20171227 14:46:31< vn971> Here is a gist of what I propose/ask:  https://gist.github.com/vn971/75475f93fc3f65d35fc0cca5d8b47cb1
20171227 14:49:44< celticminstrel> What I wanted to do was something more like...
20171227 14:50:23< celticminstrel> 1. Instead of an [abilities] tag in units, you have an abilities= attribute, which is a comma-separated list of ability IDs. For example, abilities=heals,cures,leadership
20171227 14:50:52< celticminstrel> Similarly, weapons have a specials= attribute instead of a [specials] tag.
20171227 14:51:51< celticminstrel> 2. Each ability or weapon special is stored in a single central location, so if you wanted backstab in Lua all you'd need to do is write something like "wesnoth.weapon_specials.backstab" which would be either a WML table or some weapon-special-specific data type.
20171227 14:52:00< DeFender1031> celticminstrel, I've always been in support of that idea, but it also only solves the specific case of ability/special macros, but doesn't more generally make macros which are more function- or data-like available to lua.
20171227 14:52:19< celticminstrel> Yeah, it's not a general solution...
20171227 14:53:06< vn971> celticminstrel: that is a separate proposal I think, is it? 99% of currently defined WML macros could as well be variables, because they are well-formatted correct WML "table".
20171227 14:53:20< celticminstrel> I dunno, maybe?
20171227 14:53:32< zookeeper> vn971, not that i understand the details of your suggestion, but you do actually have some kind of rationale for why the current extremely simple method should be changed into something completely new and alien?
20171227 14:53:38< vn971> so my question is about macros VS tables generally.
20171227 14:53:47< celticminstrel> I mean, I suppose it is theoretically possible for the preprocessor to keep its list of defined macros to be later injected into the Lua engine.
20171227 14:54:01< DeFender1031> vn971, except not, since they've got parameters and all...
20171227 14:54:08< celticminstrel> So I still need to get this thing done... https://github.com/wesnoth/wesnoth/pull/2297
20171227 14:54:16< DeFender1031> vn971, more like functions that output tables.
20171227 14:54:25< celticminstrel> True, true.
20171227 14:54:51< celticminstrel> And sometimes the input isn't necessarily a valid WML chunk...
20171227 14:55:08< DeFender1031> also true
20171227 14:55:45< DeFender1031> basically, though some macros may just be simple, well-formed tables, most are not
20171227 14:56:49< vn971> zookeeper: 1. to allow using those pre-defined abilities/units/objectives/events/whatever in Lua  2. Ultimately, to make Lua coding as easy as possible and move away from WML text macros as much as possible.  I know the "2." scares away, sorry about that. But still..
20171227 14:58:20< vn971> DeFender1031: which macros cannot be made a Lua function (that returns WML table) ?
20171227 14:58:37< celticminstrel> IIRC most of the core macros are one of the following: ActionWML, AnimationWML, weapon specials, abilities, traits, advancements, objects.
20171227 14:58:48-!- gfgtdf [~chatzilla@x4e3632ce.dyn.telefonica.de] has joined #wesnoth-dev
20171227 14:58:51< vn971> DeFender1031: I've seen in "Creep Wars" addon some weird stuff like half-open tags etc, but most normal usages outside I've seen are sane.
20171227 14:59:22< celticminstrel> (In the game config context, at least. The GUI context also contains tons of macros, but those are completely separate.)
20171227 14:59:29< DeFender1031> vn971, any that generate partial tags, any that take input parameters that are partial tags, any that generate something other than a whole table. There are lots.
20171227 14:59:43< DeFender1031> Sure, they could be re-worked as such
20171227 14:59:49< DeFender1031> but it's very much not a one-to-one
20171227 14:59:58< celticminstrel> Oh right, there's also all the terrain graphics macros.
20171227 15:00:11< vn971> DeFender1031: sure they are lots. Should we consider them as "bad", and easy Lua coding as much as we can for those "definitions" that are, in fact, valid WML tables?
20171227 15:00:20< vn971> * and ease
20171227 15:00:33< DeFender1031> no. they are not bad
20171227 15:01:05< vn971> ok, forget about bad. They are good. But should we make Lua coding easier for those "definitions" that are BOTH good and correct WML tables?
20171227 15:02:32< celticminstrel> It seems pretty easy right now TBH.
20171227 15:02:45< fabi> It is
20171227 15:02:52< fabi> my transcompiler does exactly that
20171227 15:03:02< celticminstrel> Just expand them into [args] and then access them from the Lua.
20171227 15:03:02< vn971> celticminstrel: have you used those "definitions" in Lua or in WML?
20171227 15:03:30< celticminstrel> Or if you want to use them in WML as well, expand them into a [set_variables][value] and access them from wherever.
20171227 15:03:34< vn971> celticminstrel: so I should have a separate [lua] tag which for each macro->variable transformation, right?
20171227 15:03:54< celticminstrel> Well, you could, but you don't have to. You can expand as many as you want into a single [lua][args].
20171227 15:03:56< gfgtdf> if you just want abilitites/weapon specials you can also write a alua scrips tzhat scans all unit types and stores all found abilitites/specials in a table indexed by their id (do this once at startup)
20171227 15:04:06< gfgtdf> a lua script that*
20171227 15:04:16< celticminstrel> Well, that's true.
20171227 15:04:18< vn971> celticminstrel: how can I expand two macros with a singlu [lua]  ?
20171227 15:04:34< DeFender1031> celticminstrel, yes, those are two of the "hacky ways to write a macro-to-lua converter" I mentioned above.
20171227 15:04:52< celticminstrel> Well, everything inside [args] is passed to the interpreter, so just expand multiple macros there.
20171227 15:04:57< DeFender1031> vn971, pass each macro as separate args
20171227 15:05:17< celticminstrel> Oh wait, can you have multiple [args] tags?
20171227 15:05:32< DeFender1031> also, note that that only works for parameterless macros.
20171227 15:05:33< vn971> celticminstrel: I don't think so.
20171227 15:05:39< DeFender1031> no, that's not what I meant
20171227 15:05:43< vn971> IDK how would Lua read `...` multiple times.
20171227 15:05:46< DeFender1031> args can contain subtags, right?
20171227 15:05:58< celticminstrel> Yeah.
20171227 15:06:10< DeFender1031> so you'd have each macro you're passing in in a separate subtag
20171227 15:06:20< DeFender1031> and then disentangle them from the lua end
20171227 15:06:25< celticminstrel> If you have two macros that produce the same tag you may need some get_nth_child instead of get_child.
20171227 15:06:39< vn971> [args] [backstab]{macro}[/backstab]  [teleport]{macro}[/teleport] [/args]
20171227 15:06:42< celticminstrel> (...) is basically just a special variable.
20171227 15:06:49< DeFender1031> not if you wrap them in tags you name yourself...
20171227 15:06:55< celticminstrel> vn971: Well you could do that too, sure.
20171227 15:07:09< vn971> and in Lua   local t = ...;   t.backstab;  t.teleport
20171227 15:07:26< celticminstrel> Yeah.
20171227 15:07:48< gfgtdf> the mainline tepeoprt macdo does already create the container tags, soit's liekely just be 	[args] {macrobackstab} {macro_telepert} [/args]
20171227 15:08:20< vn971> *sigh*
20171227 15:08:22< gfgtdf> and no, you cannot acces tags with t.backstab etc.
20171227 15:08:24< celticminstrel> But not all the ability/special macros produce a unique container tag.
20171227 15:08:32< celticminstrel> ...oh right.
20171227 15:09:16< celticminstrel> You know, it wouldn't be too much work to actually set this up for everyone, would it?
20171227 15:09:20< vn971> gfgtdf: indeed I cannot.
20171227 15:09:29-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has joined #wesnoth-dev
20171227 15:09:48-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20171227 15:10:01< celticminstrel> Adding another lua tag in the core _main.cfg that stores all abilities, specials, and traits in some central place.
20171227 15:10:15< celticminstrel> Might be a bit of a maintenance nightmare though...
20171227 15:10:44< celticminstrel> Then again... adding new traits or abilities or specials is pretty uncommon.
20171227 15:10:55< celticminstrel> (In core, that is.)
20171227 15:12:52 * celticminstrel checked and discovered that ActionWML [lua] only takes the first [args] tag. This could be changed though... and what about the global [lua] tag, isn't that parsed in C++ somewhere?
20171227 15:12:55< vn971> cat ./*.cfg | grep '#define' | wc      545 macro definitions in core/macros
20171227 15:12:56< gfgtdf> really, jsut use here is somehow how one could put abilitites macros and store them ina  lua bariable that acesses them by id (uuntested) https://pastebin.com/gE2wDEYQ
20171227 15:13:13< celticminstrel> And that's not even including all the terrain graphics macros.
20171227 15:13:42< gfgtdf> celticminstrel: since 1.13 the global args also suports [args], and since args can use subtags i see no advantage of suporting multiple args.
20171227 15:13:49< gfgtdf> global [lua] i meant
20171227 15:14:24< gfgtdf> ok my post oabove somehow merged 2 sentences together ina unifitting way.
20171227 15:14:37< gfgtdf> the '	really, jsut use ' belonged to another post
20171227 15:15:12< celticminstrel> Hehe, TELEPOET.
20171227 15:15:47< vn971> gfgtdf: so you think re-defining those things in Lua is not worth it? (And make macros be defined through variables.)
20171227 15:16:09< celticminstrel> Well, how many of those 545 variables are things you might need from Lua?
20171227 15:16:16< gfgtdf> vn971: i am not sure what you mean (i wasn't here form the strt of the diuscussion)
20171227 15:16:20< celticminstrel> I think you can probably eliminate all the ActionWML and AnimationWML macros, at least.
20171227 15:16:21< vn971> celticminstrel: 1-5 ))
20171227 15:16:27< celticminstrel> ?
20171227 15:16:35< gfgtdf> start*
20171227 15:17:48< vn971> celticminstrel: well if such a PR would (could) be accepted, I can try write something along the lines of what was proposed. [lua] code that would define some of the stuff as variable, under the same name.
20171227 15:17:58< vn971> similar to this I mean:  https://pastebin.com/gE2wDEYQ
20171227 15:19:37< gfgtdf> vn971: so you mean putting that code in mainline ?
20171227 15:19:43< vn971> gfgtdf: yes.
20171227 15:20:43< vn971> gfgtdf: for my specific purposes I can just use the approach celticminstrel proposed, and I'd be done the hacky coding relatively fast of course. But if 10-100 other people will have to do it, it'd be much slower than defining upstream of course.
20171227 15:22:13< vn971> I have an even more hacky+easier idea BTW. Something along the following lines:
20171227 15:22:14< vn971> #define MACRO_TAG NAME
20171227 15:22:14< vn971> [{NAME}]
20171227 15:22:14< vn971>   {{NAME}}
20171227 15:22:14< vn971> [/{NAME}]
20171227 15:23:01< vn971> then, use it as {MACRO_TAG WEAPON_SPECIAL_BACKSTAB} {WEAPON_SPECIAL_BACKSTAB  another_macro_name}  to generate list of tags those name reflects their contents.
20171227 15:23:25< gfgtdf> if i'd do it for maline i'd prpably use the 'search though all unittypes' apprach that ui eantioned above, the advntages are that 1) it also finds abilitits in umc, 2) it creates 0 overhead if itis not used.
20171227 15:23:28< vn971> and uniformly parse it all with 5-10 lines of Lua code, no matter of the number of inner tags.
20171227 15:24:16< vn971> gfgtdf: then it won't work for an ability not used by mainline Era, but still defined in core ?
20171227 15:24:18< gfgtdf> well the args appraoch is also use a few lines indpendent of the number of arguments, just add more abilitites in the codenippet i poste eariler
20171227 15:24:41< gfgtdf> yes it won't (are there any such abilitites?)
20171227 15:24:59< Ravana_> plague should be such
20171227 15:25:28< vn971> gfgtdf: I address the "2)" specifically
20171227 15:25:33< gfgtdf> lich and copses have plague uurc ?
20171227 15:25:50< gfgtdf> iirc* corpses*
20171227 15:25:52< Ravana_> only corpse plague, not that general version
20171227 15:26:15< gfgtdf> Ravana_: te gerneral version has a macor parameter, which none of our porposes suluition here woudl cover
20171227 15:26:34< gfgtdf> execpt ,aby fabis one (use an external scipt to generate lua code)
20171227 15:27:13< vn971> gfgtdf: I think we just desire different things. I want my thing to work on any WML-table-compatible macro, no matter if it has `id` field defined etc. You wrote a solution to address weapon specials specifically.
20171227 15:27:39< vn971> also, I'd use the macro name itself as variable name.
20171227 15:27:59< vn971> especially since users might have their own variables with some of those id-s.
20171227 15:28:11< gfgtdf> hmm ye
20171227 15:29:27< vn971> Should I try and create a PR then?
20171227 15:29:36< gfgtdf> for what exactly ?
20171227 15:31:13< vn971> for defining variables with those WML tags contents.
20171227 15:31:52< vn971> Like a "WEAPON_SPECIAL_BACKSTAB" variable that would hold the "backstab" WML table, etc.
20171227 15:32:10< gfgtdf> it's still unclar how it'd look liek i mean my code above was just about a few mainline abilitites, you want to do the same about basicially eevery macro?
20171227 15:32:51< vn971> gfgtdf: it won't live upstream if I'll only do it for 1-2 variables, I think.
20171227 15:33:00< vn971> either I do it massively or not at all.
20171227 15:33:38< vn971> I personally would happily only do that for 1-10 of the macros I really need, but I guess you just wouldn't accept such PR. I wouldn't, be I a wesnoth maintainer.
20171227 15:33:51< vn971> * if I'd be
20171227 15:33:52< celticminstrel> gfgtdf: There's the aged trait I think?
20171227 15:34:10< celticminstrel> Also, you mean "were I a ..."
20171227 15:35:14< celticminstrel> If it's not too much effort, feel free to submit that PR and we can take a look at it.
20171227 15:35:27-!- Bonobo [~Bonobo@60-241-159-43.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20171227 15:36:16< vn971> cool, will try to do then.)
20171227 15:36:20< celticminstrel> Obviously one covering all abilities and specials and traits.
20171227 15:36:28< celticminstrel> Not just the ones you're interested in. ;)
20171227 15:37:01-!- Greg-Boggs [~greg_bogg@2601:1c2:1a80:1c20:5421:392e:5cd6:dd6f] has quit [Remote host closed the connection]
20171227 15:37:27< vn971> celticminstrel: right (about "were I"), thanks for the correction. Yeah sure.
20171227 15:38:04< celticminstrel> (Or if you want to use "if" it'd be "if I were".)
20171227 15:38:10< gfgtdf> hmm if you are not oynl about abilitites etc.. don't really like the idea how parsing a lt of macros only for a small change that they'll be used. what we coudl so make  add a mainline macro to make ist easier to put them in on your own liek https://pastebin.com/0aqUtihV
20171227 15:39:41< zookeeper> gfgtdf, yeah but wouldn't that break if used with an unbalanced macro?
20171227 15:39:52< mattsc> gfgtdf: the AI items on the EasyCoding page are still all valid and (at least theoretically) useful
20171227 15:40:18< mattsc> I have no idea if anybody but myself uses them …
20171227 15:41:07< gfgtdf> mattsc: we somehow want to rmeove the easycoding age and move those thigns to the tracker, maybe you could do that for these ai issues.
20171227 15:41:38< mattsc> gfgtdf: sure; as long as it does not have to happen today
20171227 15:41:39< gfgtdf> zookeeper: well yes, but i don't think we hve unbalances macros, at lest in mainline that one woudl want to use ina  lua always ?
20171227 15:41:52< gfgtdf> mattsc: ok thx
20171227 15:43:34< celticminstrel> gfgtdf: I don't think parsing each of the ability macros once more would make that much of a difference?
20171227 15:44:30< gfgtdf> well my worry was mostly about when he said 'any WML-table-compatible macro'
20171227 15:45:05< vn971> zookeeper: actually, the code from gfgtdf does not iterate through all the macros. You still have to write {LUA_MACRO macro_name}  for any macro_name you want, avoiding those that are "unbalanced".
20171227 15:46:33< zookeeper> gfgtdf, well by "unbalanced" i don't only mean the weird rare stuff, but also macros that don't contain any tags/keys but are just a a text string, a terrain code, etc etc.
20171227 15:46:38< zookeeper> and we have tons of those.
20171227 15:47:42< zookeeper> vn971, sure, but that means LUA_MACRO comes with a bunch of limitations on what sort of macros it can be called on, so it can be kind of confusing.
20171227 15:53:09< celticminstrel> If my plan is implemented, all (or at least most) of the SPECIAL_NOTES macros won't be needed anymore.
20171227 15:53:33< vn971> So here is the thing I initially proposed: https://gist.github.com/vn971/39c44873ed510ab8aaaeb606bb2669f3   Code from gfgtdf is above. Also, we could maybe use [set_variables], which could be easier, but IDK how to do it (I never used [set_variables])
20171227 15:53:38< celticminstrel> I guess that's not all the ones that are just data without an attribtue though...
20171227 15:54:27< celticminstrel> ^attribute
20171227 15:54:55< vn971> the benefit of gfgtdf-s approach, as I see it, is that we can place "to-variable" wrapping near original macro definitions themselves, which seems better from maintenance perspective.
20171227 15:55:28< celticminstrel> You're setting them in WML variables rather than Lua variables, I see.
20171227 15:59:04< zookeeper> i'd like to remind that all this relates to something pretty weird that has only been brought up once by one person for reasons unknown, meaning that so far there's next to no reason to rush some random helper stuff into core.
20171227 16:00:43< celticminstrel> Well, opening a PR doesn't mean it'd get merged right away.
20171227 16:00:58< zookeeper> it's weird and complicated, so if it's really the best way to solve whatever problem vn971 wants to solve with it and it turns out there's actually a more common need for the same mechanism, then sure.
20171227 16:01:06< celticminstrel> Or it might not even get merged at all.
20171227 16:01:25< celticminstrel> Not sure. Need to see it and talk about it first of all.
20171227 16:02:43< vn971> any thoughts on which approach is better? Current alternatives: https://gist.github.com/vn971/39c44873ed510ab8aaaeb606bb2669f3    https://pastebin.com/0aqUtihV    https://gist.github.com/vn971/2e079af9b234003aed494a37388129ce
20171227 16:02:50< zookeeper> but as long as it's some weird hack and no one even knows how necessary in the end it even is, i'd hold off on putting that sort of hacky helpers into core and instead let them be tried out in whatever content vn971 is making, first.
20171227 16:04:38< celticminstrel> Well, first of all, is the goal to make them available as data to WML, or is the goal to make them available to Lua?
20171227 16:04:52< celticminstrel> If the latter, then I'd definitely pick the pastebin option.
20171227 16:05:24< celticminstrel> Though maybe splitting things up a bit more, I suppose.
20171227 16:05:56< celticminstrel> Also, do we want all macros that could possibly be interpreted as data?
20171227 16:06:36< vn971> celticminstrel: isn't it the same? All approaches bisically assign some variables, and that's it, no?
20171227 16:07:00< celticminstrel> I guess?
20171227 16:08:04< vn971> celticminstrel: I would go for those that can really be needed by future addon writers. Like specials, traits. IDK what else.
20171227 16:08:36< vn971> * that can realistically be needed
20171227 16:09:38< celticminstrel> Well, I had a plan to collect traits, abilities, and specials into a central location, and then in Lua you could for example access them via wesnoth.abilities.backstab.
20171227 16:09:42-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20171227 16:10:16< celticminstrel> And that would have the keys of the WML table defining backstab, ie the [damage] tag.
20171227 16:11:00< vn971> there are 289 macros that take no arguments    cat ./*.cfg | grep '^#define [^ ]*$' | wc
20171227 16:11:02< celticminstrel> It's been abandoned for awhile now, not sure if it could still be cleanly merged.
20171227 16:11:39-!- Sigyn [sigyn@freenode/utility-bot/sigyn] has quit [Read error: Connection reset by peer]
20171227 16:11:50< vn971> celticminstrel: yes, I've re-read your earlier messages, and I think it does make sense. Wouldn't it break existing addon code though?
20171227 16:12:06-!- Sigyn [sigyn@freenode/utility-bot/sigyn] has joined #wesnoth-dev
20171227 16:12:21< vn971> celticminstrel: I mean, should wesnoth parse abilities and collect everything written in the legacy format along the way?
20171227 16:12:56< celticminstrel> It would be possible to support the new abilities= attribute and also at the same time support the old [abilities] tag. It's possible in WML to have an attribute and a tag with the same name.
20171227 16:13:01< celticminstrel> So yes.
20171227 16:14:04< celticminstrel> Doesn't look like my branch ever got pushed to github.
20171227 16:14:18< celticminstrel> It's so old that I'm not sure there's much point now.
20171227 16:14:26< vn971> * rememebering won't actually be needed, sorry, my mistake. Only parse both formats.
20171227 16:14:33-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20171227 16:14:44< celticminstrel> (Last commit Aug 1 2016)
20171227 16:14:56< celticminstrel> Yeah, parsing both formats is definitely possible.
20171227 16:15:18< celticminstrel> (Wow, I've really been here more than a year now?)
20171227 16:18:30< vn971> if anybody hasn't `grep` by his hand, here's a list of all those 289 macros defined in core that have no arguments:  https://gist.github.com/vn971/94a9e05c00abc2c58c0c1680bc0e7881
20171227 16:18:52< celticminstrel> Oh, lots of AI stuff too.
20171227 16:19:28< celticminstrel> Okay, after AMLA_DEFAULT it gets into AnimationWML, which probably isn't needed...
20171227 16:19:42< celticminstrel> Then there's the deprecated stuff...
20171227 16:19:54< celticminstrel> Oh, names.
20171227 16:20:26< vn971> the "day time" macros is the hugest, it seems. I wouldn't touch it as it's unrealistically Lua writers will really use it.
20171227 16:20:31< celticminstrel> Times and schedules could be centralized too actually. Well, they sort of are already, but only for the editor's benefit.
20171227 16:20:42< celticminstrel> Yeah sure.
20171227 16:20:48< vn971> basically, it leaves only abilities, traits and weapon specials. :-/
20171227 16:21:15< celticminstrel> And maybe AMLA_DEFAULT I guess.
20171227 16:21:31< vn971> NP with that.
20171227 16:46:35-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20171227 16:46:44-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20171227 16:57:34-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20171227 17:01:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20171227 17:08:06< vn971> somewhy all of the code above doesn't work outside of [event] BTW. Had to re-wrap in [event] to get it working.
20171227 17:09:17-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20171227 17:13:59< gfgtdf> vn971: you did use wesntoh 1.13 ?
20171227 17:14:30< vn971> gfgtdf: hm, yes, I tested on 1.12. Might be the reason. Thanks for the remark (question).
20171227 17:27:33-!- irker785 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20171227 17:27:33< irker785> wesnoth/wesnoth:master gfgtdf 391c73ab65 allow spwcifying the next player number  AppVeyor: vs2017/Release Failed
20171227 17:27:33< irker785> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-679
20171227 17:34:26-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has joined #wesnoth-dev
20171227 17:45:02< irker785> wesnoth/wesnoth:master gfgtdf 391c73ab65 allow spwcifying the next player number  AppVeyor: 3/6 builds failed
20171227 17:45:03< irker785> Details vs2017/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-679
20171227 17:45:04< irker785> Details vs2015/Release: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-978
20171227 17:45:05< irker785> Details vs2013/Release: https://ci.appveyor.com/project/wesnoth/wesnoth/build/Wesnoth-VS2013-master-986
20171227 18:00:47-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20171227 18:06:10< vn971> and here it's done.. https://github.com/wesnoth/wesnoth/pull/2332
20171227 18:08:40< vn971> Sigyn gave me a warning to stop automated messages.  > (notice) Your actions in #wesnoth-dev tripped automated anti-spam measures, but were ignored based on your time in channel; stop now, or automated action will still be taken. If you have any questions, please don't hesitate to contact a member of staff
20171227 18:09:19< JyrkiVesterinen> It might be configured too aggressively.
20171227 18:10:04< JyrkiVesterinen> The spambot uses to start spamming immediately when it joins the channel, and send nothing except the spam messages.
20171227 18:10:17< vn971> well, since I'm still not banned, I'm OK with how it is now. After all, spammers were indeed stopped quickly. But if I'll get kicked, lower the aggression and bring me back please.xDD
20171227 18:11:18< DeFender1031> how DOES sigyn decide what's spam and what isn't?
20171227 18:11:34< JyrkiVesterinen> Presence of links or channel mentions, I guess.
20171227 18:11:43< vn971> the bot also probably doesn't distinguish links to github/wesnoth/wesnoth and arbitrary links.
20171227 18:12:10< DeFender1031> but vn971 hasn't really done either of those frequently
20171227 18:12:15< JyrkiVesterinen> Weren't most of your links to Gist anyway?
20171227 18:12:50< JyrkiVesterinen> DeFender1031: Indeed. I guess that Sigyn doesn't care about frequency, only the number of links.
20171227 18:13:04< DeFender1031> odd
20171227 18:13:05< vn971> and mentions.
20171227 18:13:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20171227 18:13:53< DeFender1031> also, IIRC, sigyn doesn't just boot you from the channel if you trip it, it'll ban you off freenode.
20171227 18:14:55< vn971> Ouch!
20171227 18:19:03-!- travis-ci [~travis-ci@ec2-54-205-27-188.compute-1.amazonaws.com] has joined #wesnoth-dev
20171227 18:19:04< travis-ci> gfgtdf/wesnoth#1004 (endnextside - f197513 : gfgtdf): The build failed.
20171227 18:19:04< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth/builds/322258323
20171227 18:19:04-!- travis-ci [~travis-ci@ec2-54-205-27-188.compute-1.amazonaws.com] has left #wesnoth-dev []
20171227 18:38:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20171227 18:38:33-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20171227 20:03:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20171227 20:04:20-!- vultraz [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20171227 20:09:16-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20171227 20:12:58-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20171227 20:13:31-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 240 seconds]
20171227 20:13:32-!- wedge010 is now known as wedge009
20171227 20:43:51-!- JyrkiVesterinen [~JyrkiVest@85-23-197-3.bb.dnainternet.fi] has quit [Quit: .]
20171227 20:45:18-!- irker785 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20171227 21:14:35-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 256 seconds]
20171227 21:20:08-!- travis-ci [~travis-ci@ec2-54-204-244-121.compute-1.amazonaws.com] has joined #wesnoth-dev
20171227 21:20:09< travis-ci> gfgtdf/wesnoth#1005 (endnextside - fe68330 : gfgtdf): The build is still failing.
20171227 21:20:09< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth/builds/322317115
20171227 21:20:09-!- travis-ci [~travis-ci@ec2-54-204-244-121.compute-1.amazonaws.com] has left #wesnoth-dev []
20171227 21:20:18-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20171227 21:25:12-!- mjs-de [~mjs-de@185.22.140.20] has quit [Quit: On the road again]
20171227 21:43:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20171227 21:56:45-!- atarocch [~atarocch@93.56.164.28] has quit [Remote host closed the connection]
20171227 22:02:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20171227 22:15:07-!- gfgtdf [~chatzilla@x4e3632ce.dyn.telefonica.de] has quit [Read error: Connection reset by peer]
20171227 22:17:15-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20171227 22:19:11-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20171227 22:19:12-!- wedge010 is now known as wedge009
20171227 22:25:14-!- wedge010 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has joined #wesnoth-dev
20171227 22:25:35-!- wedge009 [~Thunderbi@60-241-236-92.static.tpgi.com.au] has quit [Ping timeout: 248 seconds]
20171227 22:25:36-!- wedge010 is now known as wedge009
20171227 22:27:25-!- Bonobo [~Bonobo@60-241-159-43.tpgi.com.au] has joined #wesnoth-dev
20171227 23:07:26-!- irker345 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20171227 23:07:26< irker345> wesnoth/wesnoth:master gfgtdf da21eeea83 hgjfffj AppVeyor: vs2017/Release Failed
20171227 23:07:26< irker345> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-605wt/build/Wesnoth-VS2017-master-683
20171227 23:08:52-!- esr [~esr@wesnoth/developer/esr] has quit [Read error: Connection reset by peer]
20171227 23:53:39-!- travis-ci [~travis-ci@ec2-54-205-27-188.compute-1.amazonaws.com] has joined #wesnoth-dev
20171227 23:53:40< travis-ci> gfgtdf/wesnoth#1006 (endnextside - 583dbb8 : gfgtdf): The build has errored.
20171227 23:53:40< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth/builds/322351066
20171227 23:53:40-!- travis-ci [~travis-ci@ec2-54-205-27-188.compute-1.amazonaws.com] has left #wesnoth-dev []
--- Log closed Thu Dec 28 00:00:18 2017