--- Log opened Tue Apr 18 00:00:16 2017
20170418 00:09:20-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 00:12:26< vultraz_iOS> oh, i remembered i wanted to move read_file to global function table
20170418 00:12:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170418 00:13:37-!- Greg-Bog_ [~greg_bogg@173.240.241.83] has quit [Ping timeout: 258 seconds]
20170418 00:23:47< vultraz_iOS> i wonder if we should indeed fix the bug where the ToD editor doesn't update map color in real time
20170418 00:24:08< vultraz_iOS> in fact it's impossible to change viewing ToD in the editor at all
20170418 00:24:10< vultraz_iOS> interesting
20170418 00:33:16< vultraz_iOS> odd
20170418 00:47:14-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170418 00:47:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170418 01:39:12-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170418 01:39:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170418 02:01:23-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170418 02:26:51-!- ChipmunkV[m] [chipmunkvm@gateway/shell/matrix.org/x-carmuixceyagqjux] has quit [Ping timeout: 255 seconds]
20170418 02:28:05-!- ChipmunkV[m] [chipmunkvm@gateway/shell/matrix.org/x-rcdzlxjtflpaofdd] has joined #wesnoth-dev
20170418 02:51:41-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170418 03:15:45-!- JyrkiVesterinen [~JyrkiVest@87-100-179-36.bb.dnainternet.fi] has joined #wesnoth-dev
20170418 03:27:48-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170418 03:40:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 03:41:43-!- irker628 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170418 03:45:32-!- Kwandulin [~Kwandulin@p200300760F6D800395F68B7638287E0A.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170418 03:47:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20170418 04:29:48< celticminstrel> Hey, was it intentional to make female orc names be the same as male orc names?
20170418 04:29:58< celticminstrel> I don't really mind, just wondering if it was intentional or accidental.
20170418 04:30:28< celticminstrel> Same for trolls and woses, though I suppose those might not even have genders...
20170418 04:32:23-!- irker204 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170418 04:32:23< irker204> wesnoth: Celtic Minstrel wesnoth:master f43469e3c2d2 / data/core/macros/names.cfg: Add comments to the name generators so that their purpose can be divined in the  https://github.com/wesnoth/wesnoth/commit/f43469e3c2d29490bd575887f264eeb8641e235d
20170418 04:39:00-!- Kwandulin [~Kwandulin@p200300760F6D800395F68B7638287E0A.dip0.t-ipconnect.de] has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
20170418 05:01:53-!- atarocch [~atarocch@31.159.251.208] has joined #wesnoth-dev
20170418 05:10:15< celticminstrel> gfgtdf: Any idea if there's a way to do this? https://forums.wesnoth.org/viewtopic.php?f=21&t=45229
20170418 05:28:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 05:33:36-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170418 05:37:19< celticminstrel> This seems like a good idea for a feature request: https://forums.wesnoth.org/viewtopic.php?f=21&t=45133
20170418 05:37:35< celticminstrel> And likely simple to implement, too.
20170418 05:58:55-!- JyrkiVesterinen [~JyrkiVest@87-100-179-36.bb.dnainternet.fi] has quit [Quit: .]
20170418 06:18:20-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170418 06:25:38< celticminstrel> Argh, I forgot what I wanted to say to you...
20170418 06:25:46< celticminstrel> Maybe it wasn't important...
20170418 06:26:08< celticminstrel> (Also, do look at this thread: https://forums.wesnoth.org/viewtopic.php?f=21&t=45133 )
20170418 06:26:16< celticminstrel> vultraz_iOS: ^
20170418 06:31:19< vultraz_iOS> hm
20170418 06:31:28< vultraz_iOS> what should we do
20170418 06:31:47< celticminstrel> GUI2 wikidoc cleanup is desperately needed.
20170418 06:33:09< vultraz_iOS> indeed
20170418 06:33:27< vultraz_iOS> TBH I'd rather we threw that out the window and started over with Markdown
20170418 06:33:44< celticminstrel> That's impossible if it's to be wikidoc.
20170418 06:34:07< celticminstrel> Though I don't know what you mean by "the window" here...
20170418 06:35:06< vultraz_iOS> why?
20170418 06:35:10< celticminstrel> I'll say that throwing out the auto-wikidoc for GUI2 is technically an option, but it's set up already and presumably works, so you might as well keep it set up?
20170418 06:35:29< celticminstrel> It's impossible to use Markdown for wikidoc because MediaWiki does not use Markdown. That should be obvious.
20170418 06:35:47< vultraz_iOS> the problem with the current syntax is it's custom, so one needs to learn it, and most importantly, it's very hard to read in source form, where it's most useful
20170418 06:36:16< celticminstrel> Well, it's basically MediaWiki syntax with some macro-y extensions.
20170418 06:36:40< celticminstrel> You could put raw MediaWiki table syntax in if you wanted, but then it would be even harder to read.
20170418 06:36:46< vultraz_iOS> https://www.mediawiki.org/wiki/Extension:MarkdownExtraParser
20170418 06:38:13< celticminstrel> Seems kinda pointless to me.
20170418 06:38:46< celticminstrel> Also, I don't see how it's most useful in source form.
20170418 06:39:07< celticminstrel> The whole point of it is that it gets parsed and updated to the wiki with some regularity, where it should be even more useful than in source form.
20170418 06:39:49< vultraz_iOS> when was the last time it was even parsed and updated
20170418 06:39:56< vultraz_iOS> on the wiki, that is
20170418 06:39:57< celticminstrel> Not sure.
20170418 06:40:06< celticminstrel> But it should be done again soonish.
20170418 06:40:40< vultraz_iOS> not to mention the page structure is so weird, everything's spread out over a whole bunch of seemingly-redundant pages
20170418 06:40:50< celticminstrel> What are you talking about?
20170418 06:42:45< vultraz_iOS> https://wiki.wesnoth.org/Category:GUI_WML_Reference
20170418 06:42:53< vultraz_iOS> not very easy to find exactly what you want here
20170418 06:43:16< celticminstrel> https://wiki.wesnoth.org/GUIToolkit
20170418 06:43:36< celticminstrel> Based on that list, it seems like a fairly logical way to split things up.
20170418 06:45:08< celticminstrel> Is linked_group described anywhere there?
20170418 06:45:09-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170418 06:45:25< celticminstrel> GUIWindowDefinitionWML seems slightly useless, too.
20170418 06:45:40< vultraz_iOS> indeed
20170418 06:46:12< celticminstrel> Though I guess if you really want people to make alternate themes, it's good to document exactly which widgets are optional and manadatory and so forth.
20170418 06:46:20< celticminstrel> ^mandatory
20170418 06:47:02< celticminstrel> The reason it seems useless though is more...
20170418 06:47:10< celticminstrel> It doesn't describe the base [window] tag?
20170418 06:47:14< celticminstrel> Something like that?
20170418 06:48:17< celticminstrel> BTW, you actually changed the behaviour of the [line] tag.
20170418 06:48:24< celticminstrel> "When drawing a line it doesn't get blended on the surface but replaces the pixels instead."
20170418 06:48:28< celticminstrel> That's no longer true.
20170418 06:48:40< celticminstrel> (I already edited the source documentation to remove that reference.)
20170418 06:49:22< celticminstrel> I kinda want to add a [polygon] tag, but polygons are surprisingly hard.
20170418 06:49:46< celticminstrel> Especially if you want a filled one. In particular, you'd have to decide what to do when the edges cross.
20170418 06:50:51< celticminstrel> And a generic fill for a non-convex polygon isn't as simple as just drawing a bunch of lines like I did with the circle...
20170418 06:51:30< vultraz_iOS> why do we want a polygon tag
20170418 06:51:39< celticminstrel> I figured someone might find a use for it.
20170418 06:51:56< celticminstrel> Like an addon dev or something.
20170418 06:52:06< vultraz_iOS> not worth it, i think
20170418 06:52:29< celticminstrel> Unless someone specifically requests it, yeah. Simply because polygons are so hard.
20170418 06:53:14< vultraz_iOS> let's not get *too* into implementing shape drawing before we switch to hw rendering
20170418 06:53:25< celticminstrel> I still dislike the lack of a background behind the storyscreen title...
20170418 06:53:29< vultraz_iOS> OGL probably has a lot of shapws to draw
20170418 06:53:37< celticminstrel> You said it had problems if the title contains a line break, right?
20170418 06:53:41< celticminstrel> OGL doesn't have any shapes.
20170418 06:53:56< vultraz_iOS> i fixed it if it contains a line break
20170418 06:54:02< celticminstrel> Or rather, it jsut has the most primitive of primitives.
20170418 06:54:04< celticminstrel> ^just
20170418 06:54:04< vultraz_iOS> not perfectly, it will probably break if it has two line breaks
20170418 06:54:08< celticminstrel> Lines, triangles, quads.
20170418 06:54:33< celticminstrel> However, it automagically handles all the fill stuff, so with OGL it would certainly be far easier to create a [polygon] tag.
20170418 06:55:04< celticminstrel> Maybe the [circle] could even be reimplemented as a triangle fan...
20170418 06:55:52< celticminstrel> (A circle can also be a line loop, but using triangle fan makes it trivial to modify it to produce a sector instead.)
20170418 06:56:19< celticminstrel> (An arbitrary sector, not just my hackish octant thing that I did for rouded rectangles... >_> )
20170418 06:56:57< celticminstrel> What was your fix for the line break?
20170418 06:57:16< celticminstrel> This? https://github.com/wesnoth/wesnoth/commit/120aa0e6caf6876afa9f44dea8c1b6cd9efb28c4
20170418 06:57:39< vultraz_iOS> yes
20170418 06:59:08< celticminstrel> Hmm. It really looks kinda dubious somehow...
20170418 06:59:23< celticminstrel> Anyway, did you try putting back the background?
20170418 06:59:45< vultraz_iOS> in what respect?
20170418 06:59:47< vultraz_iOS> i did not
20170418 06:59:49< celticminstrel> Speaking of the background, I was also supposed to fix some "lumpiness" huh...
20170418 07:00:01< celticminstrel> Eh, it just looks like a hack, I guess?
20170418 07:00:24< celticminstrel> I don't really understand how it works.
20170418 07:01:37< vultraz_iOS> hm?
20170418 07:03:41< vultraz_iOS> what are you speaking of
20170418 07:04:17< celticminstrel> Apart from the line about lumpiness, I was referring to the commit I just linked.
20170418 07:04:56< vultraz_iOS> ok, so before you had 50
20170418 07:04:58< vultraz_iOS> height
20170418 07:05:04< celticminstrel> Heh, your most recent commit does nothing at all, but that's fine, I guess.
20170418 07:05:04< vultraz_iOS> works for one centered line
20170418 07:05:09< vultraz_iOS> but imagine you have 2 lines
20170418 07:05:17< vultraz_iOS> it's squashing 2 lines into 50 space
20170418 07:05:24< vultraz_iOS> so it gets too close to the top of the screen
20170418 07:05:46< vultraz_iOS> i simply top-aligned it with 20 buffer space and gave it 50 extra height to accommodate 
20170418 07:05:50< celticminstrel> Um...
20170418 07:06:11< celticminstrel> Isn't h = "(text_height)" supposed to make it grow so that the text fits?
20170418 07:06:24< vultraz_iOS> no
20170418 07:06:38< vultraz_iOS> text_width/height are the most useless keys in existance
20170418 07:06:50< celticminstrel> Oh wait, that's the width of the text, not the whole widget.
20170418 07:06:56< vultraz_iOS> they are literally the dimensions of the text after calculations
20170418 07:07:05< celticminstrel> s/width/height/
20170418 07:07:24< vultraz_iOS> yes
20170418 07:07:38< celticminstrel> Somehow I think your fix could make it look bad with a single line...
20170418 07:08:00< celticminstrel> Particularly with the background...
20170418 07:09:09< vultraz_iOS> that's why there's no background :P
20170418 07:09:13< celticminstrel> :(
20170418 07:09:29< celticminstrel> I like it better with the background.
20170418 07:09:56< celticminstrel> What if you set a grow factor on the label?
20170418 07:10:07< vultraz_iOS> could work
20170418 07:10:10< vultraz_iOS> right now it's 0
20170418 07:10:12< vultraz_iOS> and text 1
20170418 07:10:16< vultraz_iOS> but that's kinda desired...
20170418 07:10:35< celticminstrel> Leave the text y what it is now, instead of vertically centered...
20170418 07:10:43< celticminstrel> Sorry, what's desired?
20170418 07:10:56< vultraz_iOS> dunno if we want the label invisibly filling up all the space
20170418 07:11:57< vultraz_iOS> i could restore the borders if we're not going to fill the background
20170418 07:12:16< vultraz_iOS> from edge to egde, that is
20170418 07:12:40< vultraz_iOS> (speaking of backgrounds, I noticed Windows 10 uses that same top-shading technique in its lockscreen images to therefor it is good design)
20170418 07:12:54 * vultraz_iOS says, slightly tongue-in-cheek
20170418 07:17:15-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 07:21:52-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170418 07:24:28< celticminstrel> Hmm, it does seem like there are a lot of problems to consider if we want a background like the one I implemented...
20170418 07:27:46-!- celticminstrel is now known as celmin|sleep
20170418 07:30:20-!- atarocch [~atarocch@31.159.251.208] has quit [Ping timeout: 260 seconds]
20170418 07:40:50-!- TC01 [~quassel@venus.arosser.com] has quit [Ping timeout: 260 seconds]
20170418 07:41:02-!- TC01 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20170418 07:41:32-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170418 07:50:50< vultraz_iOS> zookeeper: do you happen to remember if the custom tod dialog in the editor set the color to solid if any of the sliders were at the extreme in 1.12?
20170418 07:55:41< zookeeper> oh wow, the dialog actually works in 1.12.
20170418 07:55:43-!- atarocch [~atarocch@31.159.251.208] has joined #wesnoth-dev
20170418 07:56:09< vultraz_iOS> i just need to know the behavior if one of the color sliders is set to an extreme
20170418 07:56:34< zookeeper> fine, i'll check
20170418 07:56:50< vultraz_iOS> to make sure it's not broken in 1.13
20170418 07:57:43< zookeeper> except how do i even do that? you can't edit the ToD's in 1.12 in the GUI.
20170418 07:57:49< zookeeper> there are no sliders AFAICT
20170418 07:58:11< vultraz_iOS> oh, you might have to enable a scenario in 1.12
20170418 07:58:51< zookeeper> you need to do that to enable the ToD menu at all.
20170418 08:03:24< vultraz_iOS> yeah, that changed in 1.13
20170418 08:05:53< irker204> wesnoth: Charles Dang wesnoth:master 94c9e5e604a1 / src/ (editor/controller/editor_controller.cpp gui/dialogs/editor/custom_tod.cpp): Editor: fixed ToD setting having no visible effect https://github.com/wesnoth/wesnoth/commit/94c9e5e604a141ab053a52dff6941e32d56edb27
20170418 08:06:11< vultraz_iOS> zookeeper: ^ might not be consistent to 1.12 behavior
20170418 08:06:23< vultraz_iOS> oh, and that annoying-ass slider bug is still there
20170418 08:17:08< zookeeper> the changes made with the sliders don't seem to persist
20170418 08:17:59< zookeeper> i change the ToD colour with them, press ok, change to another ToD, change back to the ToD i changed, and it's back to the default
20170418 08:18:39< zookeeper> i don't even know if that's how it's supposed to work
20170418 08:19:24< zookeeper> also, what are the + and - buttons for? i can just keep pressing + forever
20170418 08:20:38< zookeeper> and when i open the dialog, it doesn't even open to the currently selected ToD, but the first one in the schedule
20170418 08:20:50< zookeeper> ...or second
20170418 08:28:11-!- atarocch [~atarocch@31.159.251.208] has quit [Ping timeout: 255 seconds]
20170418 08:29:02-!- atarocch [~atarocch@31.159.251.208] has joined #wesnoth-dev
20170418 08:31:55< vultraz_iOS>  
20170418 08:32:13< vultraz_iOS> Even as far back as 1.11 we never implemented saving of custom tods
20170418 08:32:24< vultraz_iOS> So yes, it's buggy as hell
20170418 08:32:35-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has quit [Quit: .]
20170418 08:37:16< vultraz_iOS> anyway, at least people can make makes with tod active again
20170418 08:37:40< vultraz_iOS> maps
20170418 09:05:33-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 09:05:42-!- Shiki [~Shiki@dslb-188-105-048-166.188.105.pools.vodafone-ip.de] has joined #wesnoth-dev
20170418 09:08:23-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has joined #wesnoth-dev
20170418 09:09:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170418 09:15:43-!- TC01 [~quassel@venus.arosser.com] has quit [Ping timeout: 240 seconds]
20170418 09:20:53-!- TC01 [~quassel@venus.arosser.com] has joined #wesnoth-dev
20170418 10:05:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 10:09:50-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 252 seconds]
20170418 10:48:12< vultraz_iOS> zookeeper: what do you think about being able to not have to include [time] in every scenario, and instead be able to "register" times and refer to them by id?
20170418 11:00:03-!- gfgtdf [~chatzilla@x4e369be6.dyn.telefonica.de] has joined #wesnoth-dev
20170418 11:00:24< gfgtdf> i think not stores a copy of the ToD in every scnario is a good idea, how exactly would your idea of registering look lik e?
20170418 11:01:00< gfgtdf> vultraz_iOS: but more importantly, the editor shodul also be able to stoe scenarios without ToDs
20170418 11:01:14< gfgtdf> vultraz_iOS: it currently always inserts the default ToD by default
20170418 11:01:46< vultraz_iOS> not sure
20170418 11:01:49< vultraz_iOS> i need to ponder
20170418 11:02:22< gfgtdf> vultraz_iOS: you even the scenario editor was never mean to create complete scnearios
20170418 11:03:38< gfgtdf> vultraz_iOS: somliar to old scnearios there are ment to be included in standrst scnearios but instead of [scneairo] map_data={../editor/map/scenario1.map} you do just [scneairo] {../editor/scenarios/scenario1.cfg} in your scenario file
20170418 11:04:08< zookeeper> vultraz_iOS, sounds like something that wouldn't really have much if any benefit
20170418 11:04:09< gfgtdf> vultraz_iOS: so that the editor/scenarios/scenario1.cfg} can optionally contains other informations like sides, tods or similar.
20170418 11:05:24< vultraz_iOS> zookeeper: reducing WML?
20170418 11:05:48< zookeeper> exactly. very marginal reduction.
20170418 11:05:58< zookeeper> one might even call it "a meaningless reduction" :p
20170418 11:06:04-!- irker204 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170418 11:07:41< vultraz_iOS> oh come on
20170418 11:08:02< gfgtdf> well but note that less useless code in the scnario file does not oynl redsuces sisk size it also makes the scneariofile easier to read
20170418 11:10:03< gfgtdf> s/oynl redsuces sisk size/only reduces disk usage
20170418 11:12:58< zookeeper> if you have a savefile of 5k-50k lines, and the default schedule takes about 150 lines of that, then reducing that isn't going to make any perceptible difference
20170418 11:16:43< gfgtdf> zookeeper: this discussion isnt abotu savefiles.... it is  abotu scneaio files.
20170418 11:18:35< gfgtdf> oh wait, at lest what i said was about scewnario files
20170418 11:18:36< zookeeper> the scenario files wouldn't be affected since everyone uses macros for schedules anyway, but sure the memory footprint would be slightly smaller... well, maybe. i'm sure one could write the feature so it'd actually be consume more memory than it saves.
20170418 11:18:51< gfgtdf> vultraz taks mogth be about save files not suire
20170418 11:19:50< zookeeper> (P.S. if you wanted to make some trivial reductions to savefile size, then make empty [filter_recall], [variables] etc tags not be written for each unit)
20170418 11:21:02< gfgtdf> ok, unrelated to what vultraz said, i'll just repeat what i meant just to bea clear: i think its annoying that the scenario edtitor generated 70 lines to tod infomration i each scneario file.
20170418 11:21:58< zookeeper> well that's the scenario editor's fault. isn't it garbage anyway?
20170418 11:23:38< gfgtdf> zookeeper: well currenty one could say that but i do liek the diea of beegina ble to write instead of [scneairo] map_data={../editor/map/scenario1.map} you do just [scneairo] {../editor/scenarios/scenario1.cfg} in your scenario file and then stepwise able more features i the scneairo editor (specialyl location/map related thigns like custom time zones)(.
20170418 11:25:31< gfgtdf> that said if you wanted to reduce savefile i think a good plan is to make an id/register system for traits/abilities
20170418 11:28:47< gfgtdf> oh and not storeign event in savefiles (see notsoeasycoiding page) woudl also aba  good idea
20170418 11:29:30< zookeeper> or just have a special value for [snapshot] map_data= that tells that the map is still the same as in [replay_start] map_data=.
20170418 11:30:29< zookeeper> not storing events might make sense, although i don't really remember all the pros of storing them
20170418 11:32:41< zookeeper> anyway, if you wanted to make schedules work by id references, then sure i don't see any specific problem with that... or at least not many. you'll still break some rare things.
20170418 11:33:52< vultraz_iOS> they're rare
20170418 11:35:01< zookeeper> yes
20170418 11:44:04< gfgtdf> i wonder, do yu want to ids to complete sceduled to just to times ?
20170418 11:48:32-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 11:52:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170418 11:55:24< zookeeper> gfgtdf, uh... what?
20170418 11:55:41< gfgtdf> s/ids/give ids
20170418 11:55:59< zookeeper> i'm afraid that's not enough clarification :P
20170418 11:56:09< zookeeper> oh
20170418 11:56:15< zookeeper> to complete schedules or just to individual times
20170418 11:56:36< zookeeper> i dunno
20170418 11:58:20< zookeeper> doing it for individual times would at least allow the old schedule macros to work
20170418 12:02:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 12:05:50-!- Shiki [~Shiki@dslb-188-105-048-166.188.105.pools.vodafone-ip.de] has quit [Remote host closed the connection]
20170418 12:10:22< gfgtdf> i just looked into how to move the bugtracker to gituhub and it seems like github does not support addong images/attachments via api, 
20170418 12:11:01< gfgtdf> particular this woudl mena we cannot move out bugtacker to github issues withotu loosing all our images/attachments :(
20170418 12:17:55< vultraz_iOS> is that a huge problem?
20170418 12:19:58< zookeeper> any possibility of moving at least text type attachments into the actual report content?
20170418 12:28:02-!- irker221 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170418 12:28:02< irker221> wesnoth: Charles Dang wesnoth:master 72e9dee40d26 / src/units/unit.cpp: Unit: don't write empty subtags to config https://github.com/wesnoth/wesnoth/commit/72e9dee40d2616986d1f52447a1633f7e11a64e4
20170418 12:28:05< vultraz_iOS> zookeeper: ^
20170418 12:28:24< zookeeper> that was fast. are you sure you didn't break anything? :p
20170418 12:28:29< vultraz_iOS> no :P
20170418 12:28:49< vultraz_iOS> but there's no reason it should unless something relied on empty tag
20170418 12:28:50< vultraz_iOS> s
20170418 12:32:05< vultraz_iOS> zookeeper: the change applies to [variables], [filter_recall], [status], [modifications], [abilities], and [advancement]
20170418 12:34:05< irker221> wesnoth: ln-zookeeper wesnoth:master 18f866ebc938 / data/core/ (12 files in 2 dirs): Added new Walking Corpse variation portraits by doofus-01 https://github.com/wesnoth/wesnoth/commit/18f866ebc938e236ee30ec379e197ad31ccd9cee
20170418 12:34:08< zookeeper> well, i'd imagine it depends on whether any of the code using the contents of those tags assumes the existence of the tag. for example the code for recall filtering using the unit's [filter_recall] as-is and failing if none exists. unlikely, i guess, but just saying.
20170418 12:34:51< vultraz_iOS> ah, :inspect already looks much nicer
20170418 12:35:03< vultraz_iOS> now when you browse the unit configs you aren't inundated by empty tags 
20170418 12:35:19< vultraz_iOS> good suggestion, this was
20170418 12:38:08< vultraz_iOS> zookeeper: anywhere else this can also be applied?
20170418 12:38:51< zookeeper> i don't think so
20170418 12:59:28< irker221> wesnoth: Charles Dang wesnoth:master d2648d8f75d9 / src/team.cpp: Team: don't write empty [variables] tag https://github.com/wesnoth/wesnoth/commit/d2648d8f75d93fe8e2b0d86efa299e08ac8f2e25
20170418 13:06:02< irker221> wesnoth: ln-zookeeper wesnoth:master e228f8683a0b / data/campaigns/Under_the_Burning_Suns/ (8 files in 3 dirs): UtBS: New sprite for Human Commander by Vyncyn https://github.com/wesnoth/wesnoth/commit/e228f8683a0b0a598568391106b33c04181daad1
20170418 13:08:06< gfgtdf> it might break lua/wml coe though tries to use those tags though, e.g assumes that it doesnt manually ahve to add them
20170418 13:08:52< gfgtdf> vultraz_iOS: please add it to complability issues in RN.
20170418 13:46:47-!- JyrkiVesterinen [~JyrkiVest@85-76-70-126-nat.elisa-mobile.fi] has quit [Quit: JyrkiVesterinen]
20170418 13:48:21-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 13:50:43-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 268 seconds]
20170418 13:54:07-!- Kwandulin [~Kwandulin@p200300760F6D805120C7711FE77E4FF1.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170418 14:02:27-!- stikonas_ is now known as stikonas
20170418 15:10:46< celmin|sleep> gfgtdf: Any idea on this? https://forums.wesnoth.org/viewtopic.php?f=21&t=45229
20170418 15:11:56-!- celmin|sleep is now known as celticminstrel
20170418 15:12:14< gfgtdf> celmin|sleep: well no, i also don't know why why he'd want that. But what he can do is add a custom unit type to his faction and then use that unit as the rnadom slot (replace with another random unit is prestart event)
20170418 15:12:44< gfgtdf> s/is/in
20170418 15:13:36< celticminstrel> I don't know why he'd want it, either.
20170418 15:14:16< gfgtdf> celticminstrel: afaik we were discussing jire when we takled about moving the tracker right? you remember how the plan was to pay for it? It just looked at their hoepage and it doesn't seem to be free
20170418 15:14:42< celticminstrel> gfgtdf: Apparently there's an option to get a free license for open source projects.
20170418 15:16:01< gfgtdf> celticminstrel: hmm ok
20170418 15:16:25< gfgtdf> celticminstrel: did you see my notie eariler bug not beeing able to add attachements to github issue via api ?
20170418 15:16:37< celticminstrel> ???
20170418 15:17:01< gfgtdf> 	i just looked into how to move the bugtracker to gituhub and it seems like github does not support addong images/attachments via api, 
20170418 15:17:02< gfgtdf> 	gfgtdf	particular this woudl mena we cannot move out bugtacker to github issues withotu loosing all our images/attachments
20170418 15:18:39< celticminstrel> Losing attachments does seem like it'd be a bit of a problem.
20170418 15:19:17< celticminstrel> I'm not clear on whether JIRA is something that provides hosting for you or something you need to install on your own webhost...
20170418 15:20:00< gfgtdf> celticminstrel: i thinkjire offers both, not sure if the free opensource licent also applies to both
20170418 15:20:38< celticminstrel> I see.
20170418 15:23:19< gfgtdf> celticminstrel: searching for jire opensource free license puts me to this page: https://de.atlassian.com/software/views/open-source-license-request
20170418 15:23:41< gfgtdf> celticminstrel: when i click on the link there is get rediceted to some japenese page.
20170418 15:25:10< gfgtdf> celticminstrel: actuall im confused: one th one had it says: "If you have not done so already, download trial versions ..." which sounds liek i need to self host
20170418 15:25:37< gfgtdf> celticminstrel: bot on the other han dit says below "For cloud products, please allow 2 weeks for your open source license application to be processed" which sounds liek we can also store it in their cloud
20170418 15:28:51-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 15:31:08< celticminstrel> gfgtdf: I think that depends on which problem you're applying for...
20170418 15:31:29< celticminstrel> That application is not solely for JIRA.
20170418 15:34:42< celticminstrel> ^which program
20170418 15:35:14< celticminstrel> Maybe some programs can be either cloud or not.
20170418 15:46:01< irker221> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting b1a2d5910657 / / (3 files in 2 dirs): Properly port [kill] to Lua https://github.com/wesnoth/wesnoth/commit/b1a2d5910657def01a983583264d93a343e7b8e5
20170418 15:46:03< irker221> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting e48d1ae80523 / / (4 files in 3 dirs): Properly port [scroll] to Lua https://github.com/wesnoth/wesnoth/commit/e48d1ae80523ef8a806f0178255c6bd8c4ecde00
20170418 15:46:05< irker221> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting 36b0129fcd85 / / (14 files in 6 dirs): Properly port [music] to Lua https://github.com/wesnoth/wesnoth/commit/36b0129fcd850281beec5c2ccc2b2d4199d55723
20170418 15:46:07< irker221> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting 8d44cfd26837 / src/scripting/ (game_lua_kernel.cpp lua_common.hpp lua_unit.cpp): Lua: Add float mutator macros and some fixups/cleanup to other mutator/accessor  https://github.com/wesnoth/wesnoth/commit/8d44cfd268377841ad273f5d04d9e011fa727ba4
20170418 15:46:09< irker221> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting 98e68d659c09 / / (7 files in 5 dirs): Port [volume] to Lua https://github.com/wesnoth/wesnoth/commit/98e68d659c098ce031e395e65e991ffe414ea586
20170418 15:47:26< celticminstrel> Pentarctagon says [music] still doesn't quite work, but it works for me, so...
20170418 15:47:54< celticminstrel> I'm not sure what to do on the schedual front so I might just declare that PR ready to merge... though it might be prudent to test [kill] a little first...
20170418 15:48:03< celticminstrel> It also has changelog entries now.
20170418 15:49:40-!- gfgtdf [~chatzilla@x4e369be6.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20170418 15:53:45-!- gfgtdf [~chatzilla@x4e369be6.dyn.telefonica.de] has joined #wesnoth-dev
20170418 16:02:55< celticminstrel> ^schedule
20170418 16:12:03-!- JyrkiVesterinen [~JyrkiVest@87-100-253-69.bb.dnainternet.fi] has joined #wesnoth-dev
20170418 16:45:43-!- Kwandulin [~Kwandulin@p200300760F6D805120C7711FE77E4FF1.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170418 16:51:20< zookeeper> umm... esc doesn't skip dialogue
20170418 16:51:28< zookeeper> at least not always
20170418 16:55:26< zookeeper> now it does again. dunno what was up with that.
20170418 17:10:15-!- Kwandulin [~Kwandulin@p200300760F6D805120C7711FE77E4FF1.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170418 17:13:41< zookeeper> oh, handy. you can use the optional macro argument syntax to also kind of make local definitions. like if you need a local helper macro in another macro, you can just put it in as an optional argument instead, so it stays inside the parent macro.
20170418 17:14:27< irker221> wesnoth: Jyrki Vesterinen wesnoth:master a6719622b58c / / (6 files in 3 dirs): Execute a sensible action when the player double-clicks an add-on https://github.com/wesnoth/wesnoth/commit/a6719622b58c26feb44a93378b998c2bf380aad9
20170418 17:14:40< celticminstrel> Are you saying that things defined by #arg can also take arguments?
20170418 17:15:54< celticminstrel> JyrkiVesterinen: That looks good, but at the same time, it seems like it would be bad if you accidentally uninstalled an addon by double-clicking it. Maybe ask for confirmation for that case?
20170418 17:16:16< gfgtdf> JyrkiVesterinen: was about to say the same^^
20170418 17:16:23< JyrkiVesterinen> Why would it be a problem? The player can always simply double-click again to install the add-on back.
20170418 17:16:41< zookeeper> oh i didn't think of that. no, what i said won't of course apply if the helper macro needs arguments of its own... which i guess will be the case almost always :p meh
20170418 17:16:49< celticminstrel> Well, that's true, but that means they need to redownload.
20170418 17:17:05< celticminstrel> I'm not sure how serious a problem that is.
20170418 17:17:08< zookeeper> i just had this one case where i needed an argumentless helper macro and didn't stop to think further than that
20170418 17:17:42< JyrkiVesterinen> I think it's not a big problem. IMHO, it's better UI to just uninstall without prompting.
20170418 17:17:50< gfgtdf> also soe people make local changes to addon
20170418 17:17:54< gfgtdf> some*
20170418 17:18:12< gfgtdf> addons*
20170418 17:18:28< celticminstrel> They probably shouldn't be doing that, TBH. And if they do do it they should probably delete the info.cfg or something.
20170418 17:18:34< zookeeper> double-click where? the add-on browser or the "remove add-ons" dialog?
20170418 17:18:46< gfgtdf> zookeeper: the addon browser
20170418 17:19:15< gfgtdf> celticminstrel: does deleting info mean i cannot delete it via addonmanger lo longer ?
20170418 17:19:24< celticminstrel> Not sure?
20170418 17:19:33< zookeeper> uh-oh.
20170418 17:19:40< gfgtdf> celticminstrel: well you suggested that, so i iassumes you knew
20170418 17:20:34< gfgtdf> assumed*
20170418 17:22:01< zookeeper> JyrkiVesterinen, that... sounds very fishy to me. for installing sure that makes sense, but removal (and maybe update too?) should ask for confirmation. how is the user supposed to know that casually double-clicking on stuff is going to be a destructive operation?
20170418 17:22:43< JyrkiVesterinen> Any operation that you can undo (in this case, by reinstalling the add-on) isn't destructive.
20170418 17:23:18< zookeeper> well, substitute "destructive" for "really annoying" if you're on a bad connection and have to redownload a huge add-on
20170418 17:23:18< gfgtdf> JyrkiVesterinen: as i said above: you can't always undo it
20170418 17:24:00< JyrkiVesterinen> Hmm, okay, there seems to be enough resistance. I'll add a confirm prompt later.
20170418 17:26:12-!- atarocch [~atarocch@31.159.251.208] has quit [Ping timeout: 260 seconds]
20170418 17:26:54< zookeeper> can't say i'm a fan of UI systems where you (double)click on an item and there's completely different operations that may get performed based on state
20170418 17:27:21< celticminstrel> zookeeper: Honestly, it seems pretty intuitive in this case, possibly barring the uninstall possibility.
20170418 17:27:41< celticminstrel> If it's your own addon, it publishes. If it's not installed, it installed. If it has an update available, it updates.
20170418 17:28:02< zookeeper> yeah, install and update are intuitive, uninstall not really. i don't know what a publishable add-on looks like, so don't know about that.
20170418 17:28:54< zookeeper> but presumably publish brings up the server license dialog or something anyway to act as confirmation
20170418 17:29:07< celticminstrel> Oh yeah, I suppose it would...
20170418 17:29:13< celticminstrel> I haven't tried it yet.
20170418 17:32:21< zookeeper> i'd think the install/uninstall/upload/update/uploadupdate buttons should have colour-coding, they're all just blue now (or gray if disabled). red for uninstall, green for uploadupdate, etc.
20170418 17:32:57< celticminstrel> They're also not visible if the window is really small.
20170418 17:33:10< celticminstrel> (Well, if the window was really small when you opened the addon manager.)
20170418 17:33:33< celticminstrel> Colour-coding them does sound like a good idea though, and probably even easily done.
20170418 17:37:11< zookeeper> i don't dare test some buttons like "delete add-on from server" because now i'm not sure if there's sufficient confirmation
20170418 17:37:36< zookeeper> ...not that i would have dared to test it in the old interface either, i suppose :p
20170418 17:41:08-!- atarocch [~atarocch@31.159.251.208] has joined #wesnoth-dev
20170418 17:41:42-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20170418 17:47:42-!- atarocch [~atarocch@31.159.251.208] has quit [Ping timeout: 240 seconds]
20170418 17:50:10-!- atarocch [~atarocch@31.159.251.208] has joined #wesnoth-dev
20170418 17:52:11< JyrkiVesterinen> Publishing indeed shows a server licence dialog first, which acts as confirmation.
20170418 17:52:52< JyrkiVesterinen> I tested with a local campaignd instance. Being able to do that is the only reason why I even made it possible to compile campaignd with MSVC and run it on Windows.
20170418 17:53:42< JyrkiVesterinen> For already published add-ons which aren't outdated in the server, the only sensible double-click action would be "delete from server".
20170418 17:54:04< JyrkiVesterinen> However, that action is so dangerous that I didn't hook up double-clicking to it at all.
20170418 17:54:17< JyrkiVesterinen> The button on the right remains the only way to delete add-ons from server.
20170418 17:57:04< celticminstrel> That button is on the addon details page too, right?
20170418 17:57:48-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170418 17:57:50< JyrkiVesterinen> Yes, and I actually meant that button (and not the one in the list).
20170418 18:02:08< celticminstrel> Speaking of the buttons in the list, are they essential for the install dependency dialog to work?
20170418 18:03:45< JyrkiVesterinen> No. In fact, they are disabled in that dialog.
20170418 18:04:20< celticminstrel> I see.
20170418 18:19:57< gfgtdf> JyrkiVesterinen: were you the one who suggested jire bugtracker ?
20170418 18:20:18< JyrkiVesterinen> No.
20170418 18:20:32< celticminstrel> gfgtdf: It was Aginor IIRC
20170418 18:20:41< JyrkiVesterinen> However, I have used JIRA in an earlier job position, and I have written here that I like JIRA.
20170418 18:22:20< gfgtdf> 'used' as in 'maintained' ?
20170418 18:26:40< JyrkiVesterinen> No, I was merely a programmer who discussed with QA, marked bugs as fixed, etc.
20170418 18:26:49< irker221> wesnoth: Celtic Minstrel wesnoth:master 3802fda221e6 / src/scripting/lua_gui2.cpp: Rename some labels to be more explanatory https://github.com/wesnoth/wesnoth/commit/3802fda221e6ef5513677883186ed28149b088c6
20170418 18:26:51< irker221> wesnoth: Celtic Minstrel wesnoth:master 3c5a9aa57c7c / src/scripting/lua_gui2.cpp: Some style/formatting cleanup https://github.com/wesnoth/wesnoth/commit/3c5a9aa57c7cb0984f281f35de470395bbb2c3a8
20170418 18:26:53< irker221> wesnoth: Celtic Minstrel wesnoth:master 1954bb67b442 / src/ (9 files in 4 dirs): Lua GUI2: A few additions https://github.com/wesnoth/wesnoth/commit/1954bb67b4421fb763043d10d227842b6bb0b2f2
20170418 18:26:54< celticminstrel> gfgtdf: Thoughts? ^
20170418 18:34:12-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170418 18:36:56< gfgtdf> celticminstrel: in the lostbox lua api in wonder whether we  should just only return 1 argument which is a number or a array basted on whterh the list supports multi selection
20170418 18:37:34< celticminstrel> It'd be a possibility, I guess.
20170418 18:37:41< gfgtdf> celticminstrel: also i wonder whethe teh multimenu_button api that just takes an integer really makes sense.
20170418 18:38:30< celticminstrel> But the get_selected_row() function does return something meaningful even when multiselection is allowed - it's the most recently-selected row.
20170418 18:39:18< gfgtdf> really i'd have though it'D just retrn the topmost selecetd row
20170418 18:39:44< celticminstrel> Pretty sure it really is the most-recently-selected row.
20170418 18:39:58< celticminstrel> I haven't tested it, but the logic in the generator looks like that's what would happen.
20170418 18:40:24< gfgtdf> hm so what happens if the last selected item was deselected after selecting?
20170418 18:40:46< gfgtdf> hm ok i looked t the code
20170418 18:41:05< celticminstrel> In that case I think it does return the topmost selected item.
20170418 18:41:25< gfgtdf> ye just saw that in the code.
20170418 18:43:42< irker221> wesnoth: Charles Dang wesnoth:master 4e1df1cdd9fd / src/gui/widgets/listbox.cpp: Fixup signed comparison https://github.com/wesnoth/wesnoth/commit/4e1df1cdd9fd180c5cd0bbd15afee8ff4be78acf
20170418 18:44:05< celticminstrel> Ah, whoops.
20170418 18:48:26< gfgtdf> i wonder whether we should have some more general popup button, that can contain just any dialogs, for example with sliders etc.
20170418 18:48:42< celticminstrel> ???
20170418 18:49:40< celticminstrel> If you want a more complicated popup than menu_button or multimenu_button support, I think you should probably make the popup as its own dialog.
20170418 18:49:49< celticminstrel> ^than what
20170418 18:50:48< celticminstrel> I suppose you could make some kind of "popup_button" widget which automatically shows an arbitrary dialog (by ID probably? but that would make it useful for addons) as its popup.
20170418 18:51:14< celticminstrel> ^useless
20170418 18:52:07< gfgtdf> celticminstrel: well you ylso coudl put the childdialogs definitionswml directly in the popupbutton widget definitionwml
20170418 18:52:24< celticminstrel> I suppose...
20170418 18:52:35< celticminstrel> That could indeed work...
20170418 18:52:42< celticminstrel> Though it's somewhat unprecedented.
20170418 19:05:47-!- JyrkiVesterinen [~JyrkiVest@87-100-253-69.bb.dnainternet.fi] has quit [Quit: .]
20170418 19:27:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20170418 19:27:58-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 19:29:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Client Quit]
20170418 19:33:47< vultraz_iOS> ya know, the current tod tint in the editor really seems different..
20170418 19:33:55< vultraz_iOS> tint mechanism*
20170418 19:34:14< vultraz_iOS> i don't remember the colors having such an effect
20170418 19:36:19< Kwandulin> Yeah, midday looks really blue
20170418 19:36:32< celticminstrel> ...wait what? Why would midday be blue? o.O
20170418 19:37:25< Kwandulin> At least it looks like it on desert terrain
20170418 19:37:27< vultraz_iOS> we did adjust ToD colors recently
20170418 19:37:30< Kwandulin> totally unnatural
20170418 19:37:45< vultraz_iOS> but like...I think something got screwed with blending
20170418 19:37:49< Kwandulin> dawn/dust looks good to me, though
20170418 19:38:27< vultraz_iOS> esp with the ToD dialog
20170418 19:38:39< vultraz_iOS> the presets looks right
20170418 19:38:56< vultraz_iOS> but i dont remember the blending being so severe when they approached their limits
20170418 19:39:50< vultraz_iOS> dunno how to describe it
20170418 19:40:55< zookeeper> maybe it gets applied several times, or there's an off-by-one indexing problem, or something.
20170418 19:41:32< irker221> wesnoth: gfgtdf wesnoth:master eac013be8214 / src/generators/map_create.cpp: remove some outdated code in map_generator https://github.com/wesnoth/wesnoth/commit/eac013be82143ee6f86476e0e2aebedab4ac1711
20170418 19:41:59< vultraz_iOS> zookeeper: did you observe significant differences with 1.12 earlier?
20170418 19:42:11< zookeeper> i don't know
20170418 19:42:21< vultraz_iOS> like, I really don't remember mountains being *pink* when red was high/low
20170418 19:42:22< vultraz_iOS> :|
20170418 19:43:00< vultraz_iOS> celticminstrel: did we screw up when we touched the ToD blending code?
20170418 19:43:41< celticminstrel> Well first of all, is the problem everywhere or just in the editor?
20170418 19:43:57< celticminstrel> Where did we touch ToD blending code again?
20170418 19:45:00< vultraz_iOS> made it use color_t then reverted and you did some cleanup
20170418 19:45:04< vultraz_iOS> generally everything seems alright
20170418 19:45:08< vultraz_iOS> but...
20170418 19:45:10< vultraz_iOS> i dunno
20170418 19:46:30-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 19:49:03< vultraz_iOS> pink mountains :|
20170418 20:20:32< celticminstrel> Soooo, maybe PRs 962 and up can all be merged now? I dunno.
20170418 20:23:02-!- Shiki [~Shiki@p5B38C594.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170418 20:43:38-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170418 20:43:50-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 20:46:58-!- Kwandulin [~Kwandulin@p200300760F6D805120C7711FE77E4FF1.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170418 20:55:23< celticminstrel> vultraz_iOS: Why did you make the formula string constructor non-explicit?
20170418 20:56:33< vultraz_iOS> you mean variant?
20170418 20:56:39< celticminstrel> No. Formula.
20170418 20:56:57< vultraz_iOS> where
20170418 20:57:06< celticminstrel> formula.hpp
20170418 20:57:12< vultraz_iOS> ah
20170418 20:57:19< celticminstrel> Line 36
20170418 20:57:41< celticminstrel> Both those constructors were explicit originally, but the second one doesn't need it since it can't be called with one argument.
20170418 20:57:54< vultraz_iOS> well it takes two arguments and you told me only ctors with exactly one argument should be explicit and since I don't fully understand `explicit` I just go by that
20170418 20:58:15< celticminstrel> Constructors with one required argument should be explicit.
20170418 20:58:20< vultraz_iOS> oh
20170418 20:58:33< celticminstrel> This includes constructors with more than one argument where all but the first (or including the first) have a default.
20170418 20:58:58< vultraz_iOS> I see
20170418 20:59:09< celticminstrel> Any constructor that could potentially be called with one argument can benefit from explicit.
20170418 20:59:16< vultraz_iOS> well that changes things
20170418 20:59:47< celticminstrel> (Obviously such constructors shouldn't always be explicit, but generally if you're not sure you should probably make it explicit.)
20170418 21:04:22-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170418 21:04:58-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 21:07:31< celticminstrel> Wow, this branch is old.
20170418 21:07:39< celticminstrel> It has things like FOREACH and boost::shared_ptr.
20170418 21:09:16-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170418 21:09:20-!- Shiki [~Shiki@p5B38C594.dip0.t-ipconnect.de] has quit [Ping timeout: 260 seconds]
20170418 21:10:56< vultraz_iOS> what branch?
20170418 21:13:16< celticminstrel> lua_formula_bridge
20170418 21:18:11-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 21:18:22< celticminstrel> I suppose there's no reason to hold onto the commits that added x(n)or or short-circuiting and/or in WFL.
20170418 21:19:49< vultraz_iOS> ltter seems worthwhile
20170418 21:19:59< celticminstrel> ???
20170418 21:20:02< celticminstrel> Oh.
20170418 21:20:24< celticminstrel> It's probably not all that worthwhile; might be a minor performance boost at best.
20170418 21:20:41< vultraz_iOS> alright
20170418 21:20:43< celticminstrel> Short-circuiting isn't as useful when expressions can't have side-effects.
20170418 21:21:36< celticminstrel> I'm not sure if the possible performance boost would be worth it.
20170418 21:23:03< celticminstrel> At this point, if we want it, it's probably easier to reimplement it than update this old commit. :P
20170418 21:23:12< celticminstrel> It wasn't all that complicated IIRC.
20170418 21:23:16< gfgtdf> i think that anythign else than short-circuiting  is quite unexpected, and i'm quite surpised that forumla doesnt have it
20170418 21:23:42< celticminstrel> gfgtdf: It won't ever matter in practice that it doesn't short-circuit.
20170418 21:24:02< celticminstrel> Adding short-circuiting won't have any effect on the semantics of an and or or expression.
20170418 21:24:19< celticminstrel> At most it might have a small performance boost because the second argument need not always be evaluated.
20170418 21:24:39< gfgtdf> this can be a big perfoamcne boosed based on what teh second expresion does
20170418 21:24:56< celticminstrel> I don't think it's likely but I can't deny it's possible.
20170418 21:25:45< celticminstrel> If you want to implement it, it's just a fairly small change to the operator_expression class in formula.cpp.
20170418 21:25:46< gfgtdf> also even wirthout side effect it can be sematically differnt in pasrticular you can put expression in the second operand that are onyl allowed when the first is true like "x != 0 && 2/x == 1"
20170418 21:26:06< celticminstrel> Well, 2/0 is allowed in WFL. It evaluates to null.
20170418 21:27:08< gfgtdf> then use "arr.size() > 5 && arr[5] == 9"
20170418 21:27:31< celticminstrel> IIRC list[n] where n > length(list) also evaluates to null.
20170418 21:27:48< celticminstrel> In fact, generally expressions you'd expect to be invalid just evaluate to null.
20170418 21:38:51-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170418 21:39:26-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 21:45:23< irker221> wesnoth: Celtic Minstrel wesnoth:lua_formula_bridge 1f0e8c523843 / src/ (6 files in 2 dirs): WFL functions can now be defined in Lua https://github.com/wesnoth/wesnoth/commit/1f0e8c523843ea45e7721f5fc0976682d6a54079
20170418 21:45:35< celticminstrel> Now to test it...
20170418 21:54:03< celticminstrel> Whoops, missed an instance of boost::shared_ptr
20170418 21:56:14< irker221> wesnoth: Celtic Minstrel wesnoth:lua_formula_bridge 27a21421ab4a / src/scripting/lua_formula_bridge.cpp: fixup https://github.com/wesnoth/wesnoth/commit/27a21421ab4a098bb7b0c6f61a4feb9936f0e1fe
20170418 22:19:12-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20170418 22:31:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170418 22:32:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 22:47:20< vultraz_iOS> WFL functions in lua...?
20170418 22:47:20< vultraz_iOS> why
20170418 22:56:10-!- TC02 [~quassel@london.acm.jhu.edu] has quit [Ping timeout: 240 seconds]
20170418 22:57:44< gfgtdf> vultraz_iOS: i still wondering aout the slidre issue: in the mp staging gold silder: when the scenario specifies a gold value that is not a multiple of 25, shoudl the mp staging round it to the nearest multiple or or use it directly as default ?
20170418 22:58:13< vultraz_iOS> hmmm
20170418 22:58:21< vultraz_iOS> probably latter in this case
20170418 23:20:20< vultraz_iOS> is [time] mask= even useful for anything
20170418 23:21:45< celticminstrel> What does it do?
20170418 23:22:13< vultraz_iOS> AFAIK masks the map with an image when that ToD is active
20170418 23:22:31< celticminstrel> Sounds interesting.
20170418 23:23:15< zookeeper> it'd be useful for all sorts of things, yes
20170418 23:24:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170418 23:24:33< matthiaskrgr> anyone else got build failure on master? D: 
20170418 23:24:40-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170418 23:24:52< vultraz_iOS> nope
20170418 23:24:59< matthiaskrgr> https://pastebin.com/9dyWkwfF  
20170418 23:25:08< celticminstrel> Don't tell me it's a "conversion to bool" warning.
20170418 23:25:27< celticminstrel> Oh uh. That's really weird.
20170418 23:25:40< vultraz_iOS> definitely from eac013b 
20170418 23:25:48< celticminstrel> Has that been touched recently?
20170418 23:26:06< celticminstrel> Oh.
20170418 23:26:36< vultraz_iOS> simple fix
20170418 23:26:45< celticminstrel> Change it to name or something?
20170418 23:26:49< vultraz_iOS> yes
20170418 23:26:54< vultraz_iOS> busy tho
20170418 23:27:10< vultraz_iOS> matthiaskrgr: btw is this still relevant https://gna.org/bugs/index.php?25180
20170418 23:27:34< matthiaskrgr> hmmm
20170418 23:27:46< celticminstrel> What was even the point of that code...
20170418 23:29:44< celticminstrel> vultraz_iOS: Looks like it is still relevant.
20170418 23:29:53< celticminstrel> The snprintf still exists at that location.
20170418 23:29:54< vultraz_iOS> ok
20170418 23:30:07< vultraz_iOS> we should remove it
20170418 23:30:07< celticminstrel> What would be the stream way of doing that...
20170418 23:30:13< vultraz_iOS> and all other cases
20170418 23:30:33< gfgtdf> matthiaskrgr: did you see my message eareiler ?
20170418 23:31:35< celticminstrel> If I think hard about it, that definition of urlencode is actually wrong, isn't it?
20170418 23:32:01< vultraz_iOS> is it?
20170418 23:32:05< celticminstrel> To encode a multibyte UTF-8 character, you need multiple percents, right?
20170418 23:32:11< irker221> wesnoth: gfgtdf wesnoth:master cad5760da021 / src/generators/map_create.cpp: fixup eac013be82143 https://github.com/wesnoth/wesnoth/commit/cad5760da021a55e8f7cfe8f9dcbc63e2fa22a49
20170418 23:32:47< celticminstrel> For example, the multiplication sign is C3 97 in UTF-8, so you'd incode it as %C3%97
20170418 23:33:07-!- TC02 [~quassel@london.acm.jhu.edu] has joined #wesnoth-dev
20170418 23:33:17 * celticminstrel confirms this using Firefox.
20170418 23:33:53< matthiaskrgr> vultraz_iOS: I'm rebuilding with more recent gcc-7-dev now
20170418 23:34:01< matthiaskrgr> gfgtdf: mh what messages?  I'm not sure 
20170418 23:34:27< gfgtdf> matthiaskrgr: i asked you to do some check about code i found that looked liek memeoryleak
20170418 23:35:21< celticminstrel> Oh hey, urlencode is used in exactly one place.
20170418 23:35:28< celticminstrel> In the campaignd.
20170418 23:35:31< celticminstrel> I think?
20170418 23:35:57< celticminstrel> But anyway...
20170418 23:36:05< matthiaskrgr> gfgtdf: hmm I didn't find anything in my logs right now :/ (but they get pruned after couple of hundret lines)
20170418 23:36:21< gfgtdf> matthiaskrgr: was some days ago already iirc
20170418 23:36:39< matthiaskrgr> oh
20170418 23:36:41< celticminstrel> Apparently %D7 also works for the multiplication sign though, so I guess that implementation is technically acceptable.
20170418 23:36:54< vultraz_iOS> celticminstrel: most snprintf is in reports.cpp
20170418 23:37:00< vultraz_iOS> could you change it to streams or should i
20170418 23:37:10< celticminstrel> I'm on it.
20170418 23:37:55< gfgtdf> 20170414 12:10:33< gfgtdf> matthiaskrgr: could you please test the follwing cases for memeory leaks: 
20170418 23:38:00< gfgtdf> 20170414 12:10:42< gfgtdf> 1) upload an addonn that uses an .ign file
20170418 23:38:06< gfgtdf> 20170414 12:11:13< gfgtdf> 2) play in mp an usewr made scenario made via editor (from /editor/scenarios/)
20170418 23:38:11< gfgtdf> user-made*
20170418 23:39:04-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170418 23:39:12< matthiaskrgr> upload an addon??
20170418 23:39:29< gfgtdf> matthiaskrgr: yes in the addon manager
20170418 23:39:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170418 23:40:30< celticminstrel> gfgtdf: Speaking of .ign files, it's annoying how they don't have sane defaults.
20170418 23:41:22< gfgtdf> celticminstrel: really? i actualyl think for me the defaults were alywas sufficient
20170418 23:41:38< matthiaskrgr> ok I'm building right now 
20170418 23:41:45< celticminstrel> The problem is more that the ign file replaces the defaults rather than adding to them.
20170418 23:43:40-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170418 23:43:44< gfgtdf> celticminstrel: well not sure, mayb there are cases where one woudl want to add soem of the by default ignored filetypes
20170418 23:44:13< celticminstrel> Then there should be a syntax for explicitly overriding them.
20170418 23:45:09 * vultraz_iOS refactors the custom tod dialog
20170418 23:45:28< vultraz_iOS> this is actually probably the first GUI2 dialog I made for wesnoth...
20170418 23:45:36< gfgtdf> i think we it'd be good if we use the same syntax as git ignore files so that i (and othe who use git for thir addons) don't have to remember mutiple ignore formats.
20170418 23:45:38-!- Greg-Boggs [~greg_bogg@c-73-37-6-51.hsd1.or.comcast.net] has joined #wesnoth-dev
20170418 23:59:46< vultraz_iOS> celticminstrel: is this safe? static display* disp = display::get_singleton();
20170418 23:59:54< vultraz_iOS> ie, a static ptr and the singleton pattern
--- Log closed Wed Apr 19 00:00:12 2017