--- Log opened Sat Apr 29 00:00:10 2017
20170429 00:00:46< Aginor> try rebinding an already existing hotkey
20170429 00:00:57< Aginor> multiple times
20170429 00:01:23< Aginor> or rebinding a previously unbound hotkey
20170429 00:02:08< vultraz_iOS> in master (not your branch), if i clear a hotkey, bind it, clear it, and try to rebind the same one, it tells me it's already bound to Unrecognized Hotkey
20170429 00:02:19< Aginor> yes
20170429 00:02:24< Aginor> that's one of the regressions
20170429 00:02:36< Aginor> it's also got problems when rebinding multiple previously bound hotkeys
20170429 00:02:56< Aginor> and I'm not making any attempts to fix that in my PR
20170429 00:03:02< Aginor> that's out of scope
20170429 00:03:57< vultraz_iOS> yeah
20170429 00:03:57< vultraz_iOS> we'll deal with that
20170429 00:04:21< Aginor> I just change the hotkey system and fix the GUI2 binding
20170429 00:04:46< vultraz_iOS> yes
20170429 00:05:53< Aginor> hmm
20170429 00:06:01< Aginor> travis isn't running the build
20170429 00:06:05< vultraz_iOS> Aginor: do you know how to deal witht he damn label undrawing when dialogs are open?
20170429 00:06:08< vultraz_iOS> with the*
20170429 00:06:14< vultraz_iOS> celticminstrel fixed the theme buttons disappearing
20170429 00:06:18< vultraz_iOS> but we still need to deal with labels
20170429 00:06:19< Aginor> vultraz_iOS: I don't
20170429 00:06:22< vultraz_iOS> ok
20170429 00:06:58< Aginor> labels are drawn as volatiles, so undrawn and redrawn each frame
20170429 00:07:11< Aginor> which is a waste of cpu cycles
20170429 00:07:31< Aginor> they also don't have a concept of where they belong
20170429 00:07:51< Aginor> and will draw over the dialog if you don't hide them
20170429 00:08:01< Aginor> as the dialogs aren't redrawn every frame
20170429 00:09:43-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170429 00:10:17< vultraz_iOS> curses
20170429 00:22:19< Aginor> so if there's no objections I'll merge the PR once travis has completed
20170429 00:22:43< vultraz_iOS> go ahead
20170429 00:22:44< celticminstrel> vultraz_iOS: Speaking of theme buttons, they seem to still disappear in some cases (eg, select New Scenario).
20170429 00:23:18< vultraz_iOS> weird
20170429 00:23:19< celticminstrel> vultraz_iOS: As for labels, I had a vague idea.
20170429 00:24:17< vultraz_iOS> hmm
20170429 00:24:27< celticminstrel> How can you test if a GUI2 dialog is onscreen?
20170429 00:24:32< vultraz_iOS> ok i seem to recall the editor had something that stopped the buttons from redrawing sometime
20170429 00:24:53< vultraz_iOS> celticminstrel: gui::is_in_dialog
20170429 00:24:54< Aginor> celticminstrel: there's code in the display code to check if there's either a gui1 or gui2 dialog shown
20170429 00:25:00< vultraz_iOS> it's a flag shared between gui1 and 2
20170429 00:25:20< celticminstrel> (The odd thing about the buttons is that it's only action buttons missing, not menu buttons.)
20170429 00:25:27< celticminstrel> Oh, that covers GUI2 too?
20170429 00:25:28< Aginor> if it is, the labels are suppressed
20170429 00:25:36< celticminstrel> Hmm, then my vague idea might be useless after all.
20170429 00:25:39< Aginor> it queries gui2 internally
20170429 00:25:53< Aginor> unless vultraz_iOS changed things, it's two separate things
20170429 00:26:11< vultraz_iOS> i seem to recall the editor did not cause the buttons to vanish as much as the game did before
20170429 00:26:20< vultraz_iOS> there's something in the code that differs between game and editor
20170429 00:26:32< vultraz_iOS> obv adding the draw() calls had the inverse effect here
20170429 00:26:46< celticminstrel> That doesn't seem obvious to me?
20170429 00:28:58< vultraz_iOS> the question is what
20170429 00:30:16< celticminstrel> So what was wrong with labels again and how can I reproduce it?
20170429 00:31:07< vultraz_iOS> open a dialog and they vanish
20170429 00:31:16< celticminstrel> Oh, the exact opposite of what I thought.
20170429 00:34:54< celticminstrel> GUI2 windows know their rect, right?
20170429 00:35:20< Aginor> that'd work, exclude labels in the rect(s)
20170429 00:35:47< celticminstrel> I was thinking of making a clipping region, but I guess labels probably also know where they are too somehow... so that might work too...
20170429 00:36:03< celticminstrel> (Problem with a clipping region is that AFAIK SDL doesn't support arbitrary ones.)
20170429 00:36:44< Aginor> don't make it a proper one, make items aware of if they are obscured or not
20170429 00:36:50< Aginor> don't ever draw obscured things
20170429 00:37:02< celticminstrel> Hmm.
20170429 00:37:08< Aginor> if they're partially obscured, draw them before the items obscuring them
20170429 00:37:59-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 00:39:34< celticminstrel> I'm confused, where in terrain_label::draw does anything actually get drawn? :S
20170429 00:39:55< Aginor> I don't think so, but I don't really remember
20170429 00:43:46< celticminstrel> I don't suppose there's a way to temporarily disable double-buffering so that I can see things that get drawn as soon as they're drawn.
20170429 00:44:17< Aginor> nope
20170429 00:44:48< celticminstrel> (When stepping through in the debugger, I mean.)
20170429 00:45:02< Aginor> add more render calls
20170429 00:45:12-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Remote host closed the connection]
20170429 00:45:19< celticminstrel> Hmm?
20170429 00:45:31< Aginor> make the blit function internally actually render to screen
20170429 00:55:12-!- RatArmy_ [~ratarmy@om126234118053.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 00:57:52< celticminstrel> Oh right, vultraz_iOS, did you see my answer to your question about --give--?
20170429 00:58:04< vultraz_iOS> ye
20170429 00:58:04< vultraz_iOS> s
20170429 00:58:23< celticminstrel> Also, if it's not replaced, it should at least not be selectable.
20170429 01:07:13< celticminstrel> That doesn't appear to work, Aginor.
20170429 01:07:27< celticminstrel> Though I suppose it might be because I'm doing it wrong.
20170429 01:13:30< celticminstrel> So I have no idea when labels are actually drawn.
20170429 01:13:40< vultraz_iOS> draw_floating_labels?
20170429 01:14:03< vultraz_iOS> in display::flip
20170429 01:14:34< celticminstrel> Hmm, okay, that looks promising.
20170429 01:14:34< vultraz_iOS> draws them to back buffer, renders, then "undraws"
20170429 01:14:53< vultraz_iOS> actullly, i shouldn't say '?' since i know that's the code that handles them
20170429 01:15:02< vultraz_iOS> those functions are called in the gui2 window manegement
20170429 01:15:06< vultraz_iOS> see gui2::window::draw
20170429 01:15:30< vultraz_iOS> you should have asked me
20170429 01:16:16-!- Appleman1234_ [~Appleman1@pl12919.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170429 01:16:19< celticminstrel> Where in window::draw?
20170429 01:16:47-!- Appleman1234 [~Appleman1@pl44893.ag1212.nttpc.ne.jp] has quit [Read error: Connection reset by peer]
20170429 01:17:48< vultraz_iOS> just look :;
20170429 01:17:49< vultraz_iOS> :/
20170429 01:18:07< celticminstrel> Also in show...
20170429 01:18:13< vultraz_iOS> wait
20170429 01:18:15< vultraz_iOS> i meant show
20170429 01:18:16< vultraz_iOS> sorry
20170429 01:18:17< vultraz_iOS> >_>
20170429 01:18:29< celticminstrel> It's actually called from both.
20170429 01:18:45< celticminstrel> Though the case in draw might never be reached, I guess...
20170429 01:19:24< celticminstrel> Oh wait, show only calls undraw, draw calls draw.
20170429 01:19:53< vultraz_iOS> gui2 has its own call to flip the buffer
20170429 01:19:57< vultraz_iOS> keep that in mind
20170429 01:27:15< vultraz_iOS> i can't figure out this button issue
20170429 01:27:16< vultraz_iOS> :/
20170429 01:32:48-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 01:37:20-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 255 seconds]
20170429 01:46:04< celticminstrel> vultraz_iOS: I partially fixed it, but they still disappear after you close a dialog.
20170429 01:46:28-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170429 01:46:32< celticminstrel> For example, open the attack dialog, and the labels are still there; open damage calculations, and they're still there; close damage calculations, and they disappear.
20170429 01:47:24< irker166> wesnoth: Celtic Minstrel wesnoth:master e7cf8730023c / src/gui/widgets/ (window.cpp window.hpp): Partial fix for labels not being drawn while a dialog is open https://github.com/wesnoth/wesnoth/commit/e7cf8730023c7dd2d049e903a2ee3b5081fb7206
20170429 01:50:19< celticminstrel> Also, PR999 can probably be merged if no-one else has objections to it.
20170429 01:51:25-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170429 02:22:15< Aginor> celticminstrel: I'd suggest pushing that kind of logic out into a separate class that instead tracks windows
20170429 02:22:33< Aginor> celticminstrel: the cvideo class has the beginnings of that kind of logic already
20170429 02:26:24< celticminstrel> :|
20170429 02:26:55-!- atarocch [~atarocch@107.19.188.59] has joined #wesnoth-dev
20170429 02:52:26< vultraz_iOS> celticminstrel: actually, i noticed something weird about dialog closing
20170429 02:52:46< irker166> wesnoth: Andreas Löf wesnoth:master 92ca7fdb5eb1 / / (6 files in 3 dirs): Fix #24681: Use composed hotkeys where appropriate and revert help to be ':' https://github.com/wesnoth/wesnoth/commit/92ca7fdb5eb11c04b191359892517ea22d9f4491
20170429 02:52:48< irker166> wesnoth: Andreas Löf wesnoth:master daef8cd22ca9 / src/ (hotkey/hotkey_item.cpp hotkey/hotkey_item.hpp key.cpp key.hpp): Fix 24681: Proper filtering of modifiers when handling hotkeys https://github.com/wesnoth/wesnoth/commit/daef8cd22ca9ba689a0d5b6db1cbbbdb84285570
20170429 02:52:50< irker166> wesnoth: Andreas Löf wesnoth:master 46f16e1a8af2 / src/ (9 files in 3 dirs): Pipe raw SDL events throught the GUI2 event system for hotkeys https://github.com/wesnoth/wesnoth/commit/46f16e1a8af25f352624edbb6d8f199337d2fc62
20170429 02:52:52< irker166> wesnoth: Andreas wesnoth:master af1326d19bc9 / / (15 files in 5 dirs): Merge pull request #1003 from wesnoth/composed_hotkeys https://github.com/wesnoth/wesnoth/commit/af1326d19bc995671e6430cb232d1192f8b4c37b
20170429 02:52:53< vultraz_iOS> celticminstrel: if a dialog closes and another dialog os under it, the animations on the underlying app will briefly play
20170429 02:53:17< celticminstrel> Interesting.
20170429 02:53:33< vultraz_iOS> celticminstrel: you can observe this by opening httt and Create Unit while a flag on avillage is visible
20170429 02:53:39< vultraz_iOS> repeatedly trigger a tooltip
20170429 02:53:43< vultraz_iOS> and watch the flag
20170429 02:53:59< vultraz_iOS> Aginor: sweet :)
20170429 02:54:14< vultraz_iOS> (though in future please do use the Rebase and Merge option on GH)
20170429 02:55:01< celticminstrel> I generally use merge too though.
20170429 02:55:16< celticminstrel> Or squash.
20170429 03:02:28-!- RatArmy_ [~ratarmy@om126234118053.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 03:02:37< vultraz_iOS> Aginor: so this should be fixed now? https://gna.org/bugs/?25621
20170429 03:03:45< celticminstrel> From the sounds of it, he didn't do anything to fix something like that?
20170429 03:04:53< celticminstrel> FTR, I slightly prefer merge commits too, unless it's a fast-forward.
20170429 03:05:05< vultraz_iOS> seems to be fixed
20170429 03:05:48< celticminstrel> Since you don't have a German keyboard, you should ask gfgtfg or shiki to verify too.
20170429 03:05:52< celticminstrel> ^gfgtdf
20170429 03:06:07< vultraz_iOS> marked as RFT
20170429 03:07:05< celticminstrel> Firefox's form autofill is problematic when reloading an issue...
20170429 03:08:26< vultraz_iOS> we need to fix the other issues though
20170429 03:08:37< vultraz_iOS> Aginor: could you open a bug report for the hotkey issues you observed?
20170429 03:08:43< vultraz_iOS> Aginor: in prefs, that is
20170429 03:09:55< celticminstrel> Ugh, github's use of history API keeps breaking too.
20170429 03:11:44< irker166> wesnoth: Celtic Minstrel wesnoth:master b1d4be2f5241 / doc/manual/manual.txt: Update manual for #1003 https://github.com/wesnoth/wesnoth/commit/b1d4be2f5241a1f0a8d68fe684613fa264b75915
20170429 03:12:54< vultraz_iOS> should have mrked [ci skip]
20170429 03:12:56< vultraz_iOS> marked
20170429 03:13:02< celticminstrel> Oh right.
20170429 03:13:07< celticminstrel> I keep forgetting about that. >_>
20170429 03:13:12< celticminstrel> Sometimes I remember and sometimes I don't.
20170429 03:13:21< vultraz_iOS> likewise
20170429 03:13:24< celticminstrel> Heh
20170429 03:16:46< irker166> wesnoth: Charles Dang wesnoth:master 9742a1504603 / src/gui/core/event/dispatcher.cpp: GUI2: reformat dispatcher code https://github.com/wesnoth/wesnoth/commit/9742a1504603249e46690dd28dc1216a738eff82
20170429 03:16:58< vultraz_iOS> *feelsgood*
20170429 03:17:46< celticminstrel> ...
20170429 03:18:37< vultraz_iOS> hm?
20170429 03:18:42< celticminstrel> Nothing.
20170429 03:23:31< celticminstrel> Hmm...
20170429 03:23:45< celticminstrel> vultraz_iOS: Please comment on PR992 already.
20170429 03:24:05< vultraz_iOS> I have no comment!
20170429 03:24:12< vultraz_iOS> i agree with zookeeper that dvs is fine
20170429 03:24:15< vultraz_iOS> just merge it!
20170429 03:24:58< celticminstrel> Did you read the PR description?
20170429 03:25:57< vultraz_iOS> i have no idea which is better
20170429 03:26:30< celticminstrel> Well, you could, for example, comment on whether you think [insert_tag] should be reparsed within [objectives] whenever they're shown.
20170429 03:27:15< vultraz_iOS> well
20170429 03:27:25< vultraz_iOS> why the hell woud you use insert tag there
20170429 03:27:44< celticminstrel> I dunno, maybe the number of objectives actually depends on a variable?
20170429 03:29:33< vultraz_iOS> don't they support if
20170429 03:29:36< vultraz_iOS> wait
20170429 03:29:40< vultraz_iOS> [show if]
20170429 03:29:41< vultraz_iOS> is a thing
20170429 03:29:52< celticminstrel> True.
20170429 03:30:17< celticminstrel> But maybe there are cases where [insert_tag] would be easier.
20170429 03:30:37< celticminstrel> For example, suppose you have twenty objectives that can be completed in any order.
20170429 03:30:59< celticminstrel> You could add a show_if tag to each one, sure... or you could put them all in a variable, use [insert_tag], and just remove each one as it's completed.
20170429 03:33:04< vultraz_iOS> might as well use dvs for consistency
20170429 03:33:07< vultraz_iOS> and support insert tag
20170429 03:33:10-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit [Ping timeout: 245 seconds]
20170429 03:33:13< vultraz_iOS> for whatever weird crap dugi does
20170429 03:34:24-!- atarocch [~atarocch@107.19.188.59] has quit [Remote host closed the connection]
20170429 03:34:30-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20170429 03:38:15< wedge009> celticminstrel: I think PR 999 is okay.
20170429 03:38:56< celticminstrel> Well, I'll ask zookeeper what he things, and then merge it.
20170429 03:41:10< wedge009> things
20170429 03:41:26< celticminstrel> ^thinks
20170429 03:41:37< wedge009> Sorry, I knew what you meant, it just made me laugh.
20170429 03:41:57< celticminstrel> I was intending to do the correction anyway, but got a bit distracted.
20170429 03:43:37< vultraz_iOS> celticminstrel: good start on the labels
20170429 03:56:44< celticminstrel> vultraz_iOS: Should wesnoth.music_list.all return a config or an array of configs?
20170429 03:56:57< celticminstrel> The latter means you can index it with eg wesnoth.music_list.all[1]
20170429 03:57:43< vultraz_iOS> Latter seems more useful, but not sure what for
20170429 03:58:12< celticminstrel> Seems like Pentarctagon wants to be able to do helper.set_wml_variable_array("something", wesnoth.music_list.all)
20170429 04:01:33< vultraz_iOS> I see
20170429 04:01:57< celticminstrel> So I guess you can merge once I push this next commit?
20170429 04:02:12< vultraz_iOS> I assume child range wouldn't work as well for whim
20170429 04:02:13< vultraz_iOS> For him
20170429 04:02:43< celticminstrel> helper.child_range is an iterator, so it can't be passed to the helper function, I think?
20170429 04:02:59< celticminstrel> set_wml_variable_array or whatever it's called.
20170429 04:05:19< vultraz_iOS> Then array is acceptable
20170429 04:06:46< irker166> wesnoth: Celtic Minstrel wesnoth:wml_tag_porting a694201c0279 / changelog src/scripting/lua_audio.cpp src/sound.cpp src/sound.hpp: A few more improvements to the music API https://github.com/wesnoth/wesnoth/commit/a694201c0279bf58717523556aa7e731a25bdd14
20170429 04:09:57< celticminstrel> GN24739 might be doable...
20170429 04:10:09< celticminstrel> ^GN24239
20170429 04:10:23< celticminstrel> Wait no, it was right the first time.
20170429 04:20:04< vultraz_iOS> Btw, did we ever fix the bug where playing an mp game would eventually start spamming the console with invalid output and freeze
20170429 04:20:14< celticminstrel> The label on the location palette buttons weirdly shifts when you add a location.
20170429 04:20:34< celticminstrel> I don't remember any such bug,
20170429 04:21:19< celticminstrel> Can you add "publishable" to the addon status dropdown in the addons manager?
20170429 04:21:26< celticminstrel> Or something similar.
20170429 04:21:36< vultraz_iOS> blah
20170429 04:21:38< vultraz_iOS> i suppose
20170429 04:21:45< celticminstrel> I guess it'd combine outdated and local?
20170429 04:22:10< celticminstrel> I don't know how gfgtdf / matthiaskrgr manage to find their publishable addons at the bottom, unless they're coincidentally collated last.
20170429 04:22:21< celticminstrel> I just checked now, and my publishable addons are definitely not at the bottom.
20170429 04:22:49< vultraz_iOS> they used to be
20170429 04:22:51< vultraz_iOS> we did away with that
20170429 04:23:06< celticminstrel> But gfgtdf was talking about it being that way just recently?
20170429 04:25:26< celticminstrel> Can I use [size_lock] to constrain width but not height?
20170429 04:26:53< celticminstrel> I can't tell that the filter field in MP Create (1024 x 768 at least) is a filter field.
20170429 04:32:30-!- SigurdFD [~SigurdFD@dynamic-acs-72-23-110-196.zoominternet.net] has quit []
20170429 04:32:50< vultraz_iOS> Then set the definition
20170429 04:33:05< vultraz_iOS> Dunno about size lock
20170429 04:49:25< celticminstrel> Whyyyy are these layout exceptions not getting caught!? ><
20170429 04:49:25-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170429 04:50:39-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has joined #wesnoth-dev
20170429 04:53:30< celticminstrel> vultraz_iOS: It seems the definition is already set to filter.
20170429 04:54:03< celticminstrel> vultraz_iOS: Ah, the problem seems to be that it's initially selected for some reason.
20170429 04:54:37< vultraz_iOS> Why would it not be
20170429 04:55:08< celticminstrel> I'd say the name of game field should always be initially selected. It's the thing a user is most likely to want to change, right?
20170429 04:55:31< celticminstrel> Also, we really need to implement tabbing between fields...
20170429 04:55:41< vultraz_iOS> Both are selected
20170429 04:56:28< celticminstrel> That's impossible.
20170429 04:58:23< vultraz_iOS> Keyboard chain
20170429 04:59:40< celticminstrel> It's impossible to have two text fields selected at the same time.
20170429 04:59:48< celticminstrel> (Or even if it's possible, don't ever do it.)
20170429 05:02:38< vultraz_iOS> Oh
20170429 05:02:40< vultraz_iOS> Misread
20170429 05:02:54< celticminstrel> Anyway, you can't tell that the filter field is a filter field when it's selected.
20170429 05:04:11< celticminstrel> Okay, I managed to get rid of the scrollbar on 1024x768 by forcing the right panel to take 3/5 of the space.
20170429 05:05:21< vultraz_iOS> Are you redoing the layout
20170429 05:05:34< celticminstrel> I'm not entirely happy with that, but if I try to force it to take half of the space, it crashes.
20170429 05:07:51< celticminstrel> I'm not redoing the layout.
20170429 05:09:00< celticminstrel> Tried removing the spacer too, helps a bit but still not satisfied...
20170429 05:09:11< celticminstrel> I think maybe there's something wrong with how sliders claim space...
20170429 05:09:32< celticminstrel> It's probably either that or the menu button.
20170429 05:11:37-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 05:12:56< celticminstrel> Oh hey! Attempting to size-lock the menu button causes a crash!
20170429 05:15:57-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170429 05:16:34< celticminstrel> Okay, so I think the ideal solution here is probably to size-lock the grid on line 106 of mp_create_game.cfg.
20170429 05:16:44< celticminstrel> Maybe to a width of screen_height/4 or so,
20170429 05:17:02< celticminstrel> The only problem is that that causes a crash due to an uncaught layout exception.
20170429 05:17:38< celticminstrel> ^screen_width
20170429 05:20:46< celticminstrel> Oh, I should test size-locking the game options pane.
20170429 05:23:45< Kwandulin> Is it possible to inactivate a UMC for a certain version of wesnoth?
20170429 05:23:46< celticminstrel> Also crashes.
20170429 05:24:50< vultraz_iOS> god fucking dammit https://gna.org/bugs/?25680
20170429 05:26:36< celticminstrel> What about it?
20170429 05:26:40< vultraz_iOS> #0 0x2c05f74 in map_labels::enable(bool) /home/matthias/vcs/github/wesnoth/build/../src/map/label.cpp:260:20
20170429 05:26:42< vultraz_iOS> the fuck
20170429 05:27:50< vultraz_iOS> oh, hmmm
20170429 05:28:25< vultraz_iOS> ok
20170429 05:35:09< irker166> wesnoth: Charles Dang wesnoth:master bac460baef5b / src/editor/map/context_manager.cpp: Editor: second attempt to fix crash when closing map (bug #25680) https://github.com/wesnoth/wesnoth/commit/bac460baef5b67ae568b66bde9a674b7194ecb3f
20170429 05:36:06< vultraz_iOS> Dunno why that didn't crash for me
20170429 05:56:36-!- RatArmy_ [~ratarmy@126.234.114.89] has joined #wesnoth-dev
20170429 06:00:30< vultraz_iOS> celticminstrel: should create_map_generator return a shared or unique pt
20170429 06:00:30< vultraz_iOS> r
20170429 06:00:47< celticminstrel> No idea.
20170429 06:03:34-!- JyrkiVesterinen [~JyrkiVest@87-92-44-141.bb.dnainternet.fi] has joined #wesnoth-dev
20170429 06:06:04< vultraz_iOS> hmmmmmmmmmmmmm
20170429 06:06:11< vultraz_iOS> ok ill leave that aside
20170429 06:06:17 * vultraz_iOS works on something else
20170429 06:10:37 * vultraz_iOS has an idea how to clean up the dispatcher
20170429 06:16:57< celticminstrel> I guess I'll merge wml_tag_porting now?
20170429 06:18:06< celticminstrel> vultraz_iOS?
20170429 06:18:15 * celticminstrel was expecting you to do it, but whatever.
20170429 06:24:36< celticminstrel> Meh.
20170429 06:25:03-!- celticminstrel is now known as celmin|sleep
20170429 06:25:13< irker166> wesnoth: Celtic Minstrel wesnoth:master 339587ee2b64 / / (4 files in 3 dirs): Properly port [kill] to Lua https://github.com/wesnoth/wesnoth/commit/339587ee2b644ad700851379835700de480e18b6
20170429 06:25:15< irker166> wesnoth: Celtic Minstrel wesnoth:master 0a591cd4242e / / (4 files in 3 dirs): Properly port [scroll] to Lua https://github.com/wesnoth/wesnoth/commit/0a591cd4242e148f68bf6e734ad834b84e0f3ad9
20170429 06:25:17< irker166> wesnoth: Celtic Minstrel wesnoth:master 6906ced4e732 / / (15 files in 6 dirs): Properly port [music] to Lua https://github.com/wesnoth/wesnoth/commit/6906ced4e7321758c219cca50faaa2fd6bc38b13
20170429 06:25:19< irker166> wesnoth: Celtic Minstrel wesnoth:master 5b4dbe1c6108 / / (7 files in 5 dirs): Port [volume] to Lua https://github.com/wesnoth/wesnoth/commit/5b4dbe1c61086af398eb1e20f66a082c7b702614
20170429 06:25:21< irker166> wesnoth: Celtic Minstrel wesnoth:master eb0947c863b0 / / (6 files in 5 dirs): Some improvements to the new Lua music API https://github.com/wesnoth/wesnoth/commit/eb0947c863b08a33667f204f0885b4818101988f
20170429 06:25:23< irker166> wesnoth: Celtic Minstrel wesnoth:master c04456dec151 / src/ (scripting/lua_audio.cpp sound.cpp sound_music_track.hpp): Store music tracks as shared_ptrs https://github.com/wesnoth/wesnoth/commit/c04456dec151d907de58493b9abb0494c8fc91be
20170429 06:25:25< irker166> wesnoth: Celtic Minstrel wesnoth:master a694201c0279 / changelog src/scripting/lua_audio.cpp src/sound.cpp src/sound.hpp: A few more improvements to the music API https://github.com/wesnoth/wesnoth/commit/a694201c0279bf58717523556aa7e731a25bdd14
20170429 06:25:27< irker166> wesnoth: Celtic Minstrel wesnoth:master cdca6b79d83a / / (19 files in 8 dirs): Merge pull request #962 from wesnoth/wml_tag_porting https://github.com/wesnoth/wesnoth/commit/cdca6b79d83a3f1521824351f53d2ced6f2722b9
20170429 06:29:10< JyrkiVesterinen> Wedge009: Are you aware that the game doesn't compile with aquileia/external VS15 branch?
20170429 06:29:21< JyrkiVesterinen> Visual Studio complains about missing files.
20170429 06:29:23< JyrkiVesterinen> https://gist.github.com/jyrkive/ae72caa92da2b3cb362cd786276515ea
20170429 06:49:07-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170429 06:49:13-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170429 06:57:18< JyrkiVesterinen> 20170429 04:25:26< celticminstrel> Can I use [size_lock] to constrain width but not height?
20170429 06:57:20< JyrkiVesterinen> Maybe.
20170429 06:57:37< JyrkiVesterinen> It forces the wrapped widget to *request* certain size in both dimensions.
20170429 06:58:09< JyrkiVesterinen> However, the grid that contains the widget can give it more space in either direction, if there happens to be enough free space.
20170429 06:59:53-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 07:02:55< matthiaskrgr> ../src/scripting/lua_audio.cpp:140:39: warning: unused parameter 'L' [-Wunused-parameter]
20170429 07:02:58< matthiaskrgr> static int intf_music_next(lua_State* L) {
20170429 07:04:39-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 268 seconds]
20170429 07:24:36-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170429 07:33:11< JyrkiVesterinen> celmin|sleep: For future reference, German, Swedish and Finnish keyboard layouts are highly similar. I was able to verify the fix for bug 25621.
20170429 07:35:18< irker166> wesnoth: Jyrki Vesterinen wesnoth:master 9edf49e3af6b / src/scripting/lua_audio.cpp: Remove an unused variable https://github.com/wesnoth/wesnoth/commit/9edf49e3af6bc87bb2c6ab21cacdfdc4c1570e4e
20170429 07:35:27< JyrkiVesterinen> matthiaskrgr: ^
20170429 07:35:47< wedge009> JyrkiVesterinen: I use my own set, I don't use what's in external. There's probably some headers missing, but I just followed what aquileia specified. The instructions may need to be updated.
20170429 07:37:07< JyrkiVesterinen> Yeah. When I initially tested compiling with VS2017 and set up the externals with Boost develop, I also noticed that a lot of headers were missing.
20170429 07:37:36< JyrkiVesterinen> I think I can look into that next week.
20170429 07:37:54-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20170429 08:00:34-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170429 08:03:13-!- deathisundead [~quassel@unaffiliated/the-unforgiven/x-8713611] has quit [Ping timeout: 260 seconds]
20170429 08:24:05-!- Appleman1234_ [~Appleman1@pl12919.ag1212.nttpc.ne.jp] has quit [Quit: Leaving]
20170429 08:29:31-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170429 08:36:18-!- zookeeper changed the topic of #wesnoth-dev to: 1.13.8 (1.14. Beta 1) scheduled for Sunday, May 14th 00:01 UTC | Wesnoth Developers Channel | >>> Want to help? Go here: http://r.wesnoth.org/t42911 (and thanks!) <<< | Discord Server: https://discord.gg/tSmJS2E | Logs: http://irclogs.wesnoth.org | Bug tracker: http://bugs.wesnoth.org
20170429 08:36:19-!- RatArmy_ [~ratarmy@126.234.114.89] has quit [Read error: Connection reset by peer]
20170429 08:37:12< JyrkiVesterinen> Great, that additional minute will surely be of big help. :D
20170429 08:37:32< zookeeper> agreed
20170429 08:37:50-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170429 08:38:20-!- Appleman1234 [~Appleman1@pl8682.ag1212.nttpc.ne.jp] has joined #wesnoth-dev
20170429 08:48:14-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 08:52:35-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 240 seconds]
20170429 09:07:01-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 09:20:14-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 09:23:40-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 09:28:26-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 09:29:30-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170429 09:29:30-!- horrowind1 is now known as horrowind
20170429 09:42:46-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 09:44:47-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 09:45:26-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170429 09:45:26-!- horrowind1 is now known as horrowind
20170429 09:49:08< Aginor> https://gna.org/bugs/?25682 requires more of a horrible hack in the to-string function to treat colons as special things
20170429 09:52:37< Aginor> JyrkiVesterinen: do you know how similar Finnish keyboard layout is to Swedish_
20170429 09:52:57< Aginor> I'd imagine identical, but I don't actually know
20170429 09:53:05< JyrkiVesterinen> I believe it affects every character which has a dedicated key in the US layout but requires multiple keys in a different layout.
20170429 09:53:12< JyrkiVesterinen> Almost 100 % identical.
20170429 09:53:29< Aginor> JyrkiVesterinen: no, it's colons that are special
20170429 09:53:39< Aginor> colons have an ugly horrid hack associated with them
20170429 09:53:50< JyrkiVesterinen> https://en.wikipedia.org/wiki/QWERTY#Finnish_multilingual
20170429 09:55:27< Aginor> JyrkiVesterinen: looks like they're the same
20170429 09:55:53< Aginor> I ended up having to look at wikipedia, it's been way to long since I saw a physical swedish keyboard 
20170429 09:56:37< zookeeper> Aginor, did you happen to have any kind of guesses regarding the gamemap background rendering bug i've been mentioning a few times recently?
20170429 09:57:18< Aginor> zookeeper: I havneät seen you mention it, sorry
20170429 09:59:13< zookeeper> Aginor, ok, so basically if you open the editor, downscale the map to 10x10 or whatever so that the outside-the-map background is showing, then that background is drawn very slightly differently than it should. if you press F5, it fixes itself. this also happens outside the editor, where AFAICT there is no workaround.
20170429 10:01:07< zookeeper> the background gets rendered slightly lighter, with some seemingly random pixels having red value go from 35 to 37 and blue value from 31 to 32, etc. it's subtle and weird.
20170429 10:02:29< Aginor> the color shifts makes me wonder if it's a blending issue
20170429 10:02:37< Aginor> what's the background initialised to?
20170429 10:03:14< zookeeper> you can see the difference in screenshots, there doesn't seem to be any consistent pattern to it WRT which pixels are affected.
20170429 10:03:31< zookeeper> what do you mean? it's a bitmap, core/images/terrain/off-map/background.png
20170429 10:04:40< zookeeper> for a more easy comparison one can fill that image with some solid color
20170429 10:04:42< Aginor> zookeeper: what screenshots?
20170429 10:05:07< zookeeper> the ones you take before and after pressing F5 if you want to see the bug :P
20170429 10:05:26< Aginor> ah, those
20170429 10:05:39< zookeeper> ohhhh wait
20170429 10:05:58< zookeeper> i hadn't checked one specific thing since i changed the image to a solid color
20170429 10:06:10< zookeeper> the pattern is grass
20170429 10:06:16< Aginor> so blending
20170429 10:06:31< Aginor> render that background with blending disabled
20170429 10:07:31< Aginor> I think it might be a generally good idea with the background not being rendered with any blending
20170429 10:08:43< zookeeper> the problem is basically that when you resize the map, whatever was previously drawn on the parts of the screen that are now off the map still show through the background
20170429 10:08:56< Aginor> yes
20170429 10:09:01< Aginor> re+render the background
20170429 10:10:19< zookeeper> i don't know what blending there supposedly is, though
20170429 10:10:38< zookeeper> or why would the background be drawn in a manner that causes what's underneath it to show through at all
20170429 10:11:13< Aginor> it depends on how it's drawn and what the layering is
20170429 10:11:57< zookeeper> it just gets the image and sdl_blits it
20170429 10:12:01< Aginor> but if the background is an image with an alpha channel, or the previously drawn graphics that it's being drawn over has a non-zero alpha, there'll be blending
20170429 10:12:14< Aginor> sdl_blits does alpha blending
20170429 10:13:19< zookeeper> righty. changing the image to be 24-bit does fix it.
20170429 10:13:29-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170429 10:15:55< zookeeper> but doesn't this still indicate some kind of bug? i mean, if you draw a bitmap which has 0 transparency, even if it does have an alpha channel and you draw it with alpha blending or whatever, then it should completely cover whatever is underneath it.
20170429 10:16:23< zookeeper> could it be an off-by-1 or a rounding error in some alpha-related stuff, either in sdl or how we use it?
20170429 10:16:25-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 10:16:39< Aginor> I've checked the maths in SDL and it's pretty solid
20170429 10:17:58-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20170429 10:17:58-!- horrowind1 is now known as horrowind
20170429 10:18:25< Aginor> wesnoth, before SDL2, didn't have a concept of, or ever took alpha into account except for in gui2
20170429 10:19:00< Aginor> so there's been a few alpha related bugs where there's had to be blending disabled in order to make it work
20170429 10:19:29< Aginor> the draw/undraw pattern in particular gave me grief
20170429 10:20:36< zookeeper> uh. i dunno what you mean by "ever took alpha into account".
20170429 10:20:57< Aginor> any call to sdl1.2's blit function discarded alpha
20170429 10:21:10< Aginor> colours didn't have alpha components
20170429 10:21:24-!- JyrkiVesterinen [~JyrkiVest@87-92-44-141.bb.dnainternet.fi] has quit [Quit: Going offline for an hour or two]
20170429 10:21:28< Aginor> the custom-written blit functions were used on the instances that alpha was wanted
20170429 10:21:40< zookeeper> right, okay
20170429 10:21:56< zookeeper> which i presume were "most instances" of blitting :p
20170429 10:22:34-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20170429 10:22:51< Aginor> no, it was actually a rather small subset
20170429 10:23:04< Aginor> I've since stomped them all out except for one
20170429 10:23:53< zookeeper> i just mean that obviously we've always been able to use alpha-channeled images virtually everywhere
20170429 10:24:18< Aginor> yeah, sure
20170429 10:24:19< zookeeper> which means all that rendering must have gone through that custom-written blit function one way or another then
20170429 10:26:37-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170429 10:26:54< zookeeper> ok, so, anyway: i suppose that maybe i can fix the background bug by for example filling with 0,0,0 before blitting. but that still doesn't fix the underlying blending problem that caused the issue in the first place, whatever that may be. of course it might not cause a perceptible issue anywhere else, but still.
20170429 10:27:28< vultraz_iOS> imnot sure why it's really an issue at all
20170429 10:28:22< Aginor> zookeeper: disabling blending is not the answer. Ensuring that you start the draw-cycle with a sensible canvas is
20170429 10:28:23-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 10:31:48-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 10:32:25< zookeeper> Aginor, but we don't get a sensible canvas when we start drawing units on terrain either. yet the opaque parts of the unit shouldn't be affected by what terrain's pixels behind them are. why should the canvas matter in this case, if it doesn't with terrains and units? or are you saying that it's just some inherent imprecision in SDL and i _should_ expect terrain to show through units very 
20170429 10:32:25< zookeeper> very slightly?
20170429 10:34:12-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 260 seconds]
20170429 10:34:12-!- horrowind1 is now known as horrowind
20170429 10:34:13< zookeeper> or is the difference caused by there being a black canvas behind the big stack of terrain and unit images?
20170429 10:35:47-!- irker166 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170429 10:36:30-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 10:38:38< zookeeper> if that's the difference then that does make sense to me (kinda). gamemap drawing starts with a sensible canvas, and that means all subsequent blending works as you'd expect. background drawing doesn't start with a sensible canvas, so blending doesn't work as you'd expect (or at least not for the first layer of blitting).
20170429 10:41:09-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 260 seconds]
20170429 10:42:37-!- irker086 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170429 10:42:37< irker086> wesnoth: Andreas Löf wesnoth:master 73faf496a9f6 / src/hotkey/hotkey_item.cpp: Fix #25682: Ignore "shift" when the hotkey is ":" https://github.com/wesnoth/wesnoth/commit/73faf496a9f6070b974c10daa5ac2595d706b79e
20170429 10:44:57< Aginor> zookeeper: indeed
20170429 10:47:14-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 10:48:27-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170429 10:48:27-!- horrowind1 is now known as horrowind
20170429 10:50:08< zookeeper> great! i'll try the fill solution tonight
20170429 10:50:38-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 10:51:25-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170429 10:53:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 10:53:35< Aginor> vultraz_iOS: I've given you https://gna.org/bugs/index.php?23897 
20170429 10:53:43< Aginor> you need to follow up with the fedora packager
20170429 10:53:53< Aginor> or someone at least
20170429 10:54:00< vultraz_iOS> alright...
20170429 10:54:39< Aginor> otherwise any rpm-based system will probably never get a newer version of wesnoth
20170429 10:54:57< vultraz_iOS> rpm?
20170429 10:55:45< Aginor> https://en.wikipedia.org/wiki/RPM_Package_Manager
20170429 10:55:54< Aginor> any redhat/fedora/centos based system
20170429 11:00:18 * vultraz_iOS ponders the gui2 dispatch system
20170429 11:00:45< vultraz_iOS> i can get rid of all the functor wrappers...
20170429 11:00:52< vultraz_iOS> using a parameter pack
20170429 11:00:58< vultraz_iOS> but for some reason doing so doesn't fully work
20170429 11:03:18-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 11:05:15-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 245 seconds]
20170429 11:05:15-!- horrowind1 is now known as horrowind
20170429 11:08:07-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 11:12:56< vultraz_iOS> ahhh
20170429 11:12:58< vultraz_iOS> std::forward
20170429 11:13:01< vultraz_iOS> now we're getting somewhere
20170429 11:15:00< vultraz_iOS> hmm
20170429 11:15:03< vultraz_iOS> ..\..\src\gui\core\event\dispatcher_private.hpp|429|error: invalid initialization of non-const reference of type 'gui2::event::message&' from an rvalue of type 'gui2::event::message'|
20170429 11:15:07< vultraz_iOS> what... does that mean
20170429 11:16:45< vultraz_iOS> hmmmmmm
20170429 11:18:22< vultraz_iOS> maybe i don't need forward..?
20170429 11:19:26-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 11:22:32-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170429 11:34:50-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 11:34:51-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 11:37:05-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170429 11:37:05-!- horrowind1 is now known as horrowind
20170429 11:42:05-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection]
20170429 11:42:09-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170429 11:42:11-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev
20170429 11:43:37-!- mkdroid [~null@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20170429 11:43:46-!- atarocch [~atarocch@107.19.188.59] has joined #wesnoth-dev
20170429 11:48:37-!- Kwandulin [~Kwandulin@p200300760F6D809EFC70C9B757DC4C53.dip0.t-ipconnect.de] has quit [Ping timeout: 246 seconds]
20170429 11:50:04-!- mkdroid [~null@unaffiliated/matthiaskrgr] has quit [Quit: I'll be back!]
20170429 11:50:24-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170429 11:50:35-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 11:50:56-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 11:54:02-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 11:55:58-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 246 seconds]
20170429 11:55:58-!- horrowind1 is now known as horrowind
20170429 12:01:14-!- heirecka [~heirecka@exherbo/developer/heirecka] has quit []
20170429 12:08:39-!- JyrkiVesterinen [~JyrkiVest@87-92-44-141.bb.dnainternet.fi] has joined #wesnoth-dev
20170429 12:08:53-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 12:11:05-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20170429 12:11:05-!- horrowind1 is now known as horrowind
20170429 12:18:06-!- horrowind1 [~Thunderbi@x4e33de10.dyn.telefonica.de] has joined #wesnoth-dev
20170429 12:19:45-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Ping timeout: 258 seconds]
20170429 12:19:45-!- horrowind1 is now known as horrowind
20170429 12:19:49-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 12:24:00-!- horrowind [~Thunderbi@x4e33de10.dyn.telefonica.de] has quit [Quit: horrowind]
20170429 12:24:22-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 12:24:45-!- Greg-Boggs [~greg_bogg@173.240.241.83] has joined #wesnoth-dev
20170429 12:29:38-!- Greg-Boggs [~greg_bogg@173.240.241.83] has quit [Ping timeout: 268 seconds]
20170429 12:42:27-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170429 12:42:40-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 12:46:56-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 12:49:41-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has joined #wesnoth-dev
20170429 12:54:10-!- Kwandulin [~Kwandulin@p200300760F6D80DFC912DB11B98971C7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170429 13:04:20-!- travis-ci [~travis-ci@54.226.73.33] has joined #wesnoth-dev
20170429 13:04:21< travis-ci> gfgtdf/wesnoth#804 (unitmap_replace - 80781fc : gfgtdf): The build has errored.
20170429 13:04:21< travis-ci> Build details : https://travis-ci.org/gfgtdf/wesnoth/builds/227113255
20170429 13:04:21-!- travis-ci [~travis-ci@54.226.73.33] has left #wesnoth-dev []
20170429 13:14:01-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev
20170429 13:23:21-!- RatArmy_ [~ratarmy@om126234114089.16.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 13:26:49-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20170429 13:27:01-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 13:36:02-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170429 13:36:34-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 13:42:43-!- irker086 [~irker@uruz.ai0867.net] has quit [Quit: transmission timeout]
20170429 14:03:09-!- atarocch [~atarocch@107.19.188.59] has quit [Remote host closed the connection]
20170429 14:06:52-!- irker568 [~irker@uruz.ai0867.net] has joined #wesnoth-dev
20170429 14:06:52< irker568> wesnoth: gfgtdf wesnoth:master f71e64e3644d / src/ (6 files in 5 dirs): bring back unit_map::replace https://github.com/wesnoth/wesnoth/commit/f71e64e3644dece5d4d5a6a87235e880cab45253
20170429 14:07:50-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has quit [Quit: Connection closed for inactivity]
20170429 14:08:47< irker568> wesnoth: gfgtdf wesnoth:master 12f9215da24e / src/ (7 files in 3 dirs): deployed some range for loops on gameboard::teams() (#1007) https://github.com/wesnoth/wesnoth/commit/12f9215da24e5bee58ff268cba5893b183cd349d
20170429 14:21:24-!- heirecka [~heirecka@exherbo/developer/heirecka] has joined #wesnoth-dev
20170429 14:21:54-!- Kwandulin [~Kwandulin@p200300760F6D80DFC912DB11B98971C7.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20170429 14:28:36-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection]
20170429 14:29:03-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20170429 14:56:36-!- celmin|sleep is now known as celticminstrel
20170429 15:01:58-!- gfgtdf [~chatzilla@x4e36a612.dyn.telefonica.de] has joined #wesnoth-dev
20170429 15:06:50-!- Kwandulin [~Kwandulin@p200300760F6D80DFC912DB11B98971C7.dip0.t-ipconnect.de] has joined #wesnoth-dev
20170429 15:07:12< irker568> wesnoth: gfgtdf wesnoth:gfgtdf-patch-2 3f592ffb6cf4 / .travis.yml: remove 'allow_failures' from c++14 build https://github.com/wesnoth/wesnoth/commit/3f592ffb6cf4ed22aaf6d7259cd444cb4473b665
20170429 15:07:29< gfgtdf> any opionons on that ^ (pr #1008) ?
20170429 15:08:32< JyrkiVesterinen> It sounds good to me.
20170429 15:09:56< celticminstrel> gfgtdf: Any idea why this commit doesn't work? https://github.com/wesnoth/wesnoth/tree/lua_multi_require
20170429 15:13:14< celticminstrel> zookeeper: Could you please comment on this? https://github.com/wesnoth/wesnoth/pull/992
20170429 15:15:36< zookeeper> celticminstrel, uh, was "The way to do this is not parse the" supposed to be "The way to do this is to parse the"?
20170429 15:16:36< celticminstrel> No. The [objective] tag currently parses the config as its very first action.
20170429 15:17:10< zookeeper> okay
20170429 15:17:24< celticminstrel> As you can see in the middle commit.
20170429 15:18:41< gfgtdf> celticminstrel: mayb th eproblsm is that you dont expand relative paths ?
20170429 15:19:09< gfgtdf> celticminstrel: actuall that dpesn which strings did you use for testing ?
20170429 15:19:31< celticminstrel> gfgtdf: My test was 'wesnoth.require "lua/wml"'
20170429 15:19:41< celticminstrel> gfgtdf: The result was a table of {items = {...}}
20170429 15:19:57< celticminstrel> gfgtdf: Expected result would've had that plus other keys as well.
20170429 15:20:40< celticminstrel> gfgtdf: I think relative paths were expanded already by intf_require.
20170429 15:20:41< gfgtdf> celticminstrel: ok rhat this doens tseem to be the problem, still i'T be nice if it wokred with  relative paths aswell liek the single itme version
20170429 15:21:08< gfgtdf> celticminstrel: ah right
20170429 15:21:25-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity]
20170429 15:21:26< gfgtdf> celticminstrel: didnt assuigmed the wor thins in the code that the diff didn't show
20170429 15:21:38< celticminstrel> Sorry?
20170429 15:22:23< gfgtdf> celticminstrel: assmed the wrong  things in the code that the diff didn't show
20170429 15:24:01< celticminstrel> gfgtdf: So, any ideas?
20170429 15:24:07< gfgtdf> thinking
20170429 15:24:39< zookeeper> celticminstrel, i'll check more after i've munched on some pizza ->
20170429 15:32:53< gfgtdf> celticminstrel: hmm no i don't see it
20170429 15:33:42< gfgtdf> celticminstrel: but i think that it'd be better if that __call function stored the key somehow instead of using next, just in casr that the reult does depends on the order in which they are called.
20170429 15:34:05< celticminstrel> Hm?
20170429 15:34:42< gfgtdf> s/stored/sorted
20170429 15:34:53< gfgtdf> s/casr/case
20170429 15:35:10< celticminstrel> I'm not quite sure what you mean...
20170429 15:35:45< gfgtdf> celticminstrel: the __call function uses lua_next to iterator over all files.
20170429 15:35:59< celticminstrel> Essentially, yes.
20170429 15:36:21< gfgtdf> celticminstrel: which means undefined order. I think it'd be better to use a defin4ed order (aphabetical) just in ccase the result depends on the order.
20170429 15:36:41< celticminstrel> Oh, I think I get it.
20170429 15:36:56< celticminstrel> Load the modules in alphabetical order instead of iteration order.
20170429 15:37:03< celticminstrel> That does make sense.
20170429 15:37:18< celticminstrel> Well, they're loaded in alphabetical order already, I think; Evaluate in alphabetical order too though.
20170429 15:38:28< gfgtdf> celticminstrel: i _think_ that filesystem::get_files_in_dir does use alphabetical order, but lua_next iterates over the vector  in undefined order.
20170429 15:38:42< celticminstrel> ...hmm, I wonder if it could be because most of those files don't actually have a return statement, rather than an error in my code...
20170429 15:39:00< gfgtdf> celticminstrel: oh yes that shodul be it
20170429 15:39:22< gfgtdf> celticminstrel: i mean {val=nil} is afaik the same as {} form lua perspecive at lest
20170429 15:40:01< celticminstrel> Okay, so now I just need to make sure that requiring "dir/file.lua" and requiring "dir" doesn't load file.lua twice.
20170429 15:40:10< celticminstrel> Regardless of the order in which you require them.
20170429 15:40:23< celticminstrel> And also the thing you said about arbitrary ordering.
20170429 15:41:11< celticminstrel> gfgtdf: Okay to merge PR999?
20170429 15:41:53< gfgtdf> i'd it's fine. 'prevLabel' variabelname doesnt dexactly match out naming convetion though.
20170429 15:42:12< celticminstrel> I didn't think we had a strict naming convention.
20170429 15:42:34< irker568> wesnoth: Alekzander Bollig wesnoth:master ceba081542a4 / changelog data/core/about.cfg src/display.cpp: Fix for 21634 (#999) https://github.com/wesnoth/wesnoth/commit/ceba081542a448c92638d54b02a5a40c4c0d5f0b
20170429 15:43:44< celticminstrel> gfgtdf: What do you think of PR 1002?
20170429 15:44:24< celticminstrel> zookeeper: Is it okay to merge PR 1004?
20170429 15:45:23< gfgtdf> celticminstrel: i have no idea how #1002 work
20170429 15:45:30< celticminstrel> Nor do I.
20170429 15:45:48< gfgtdf> celticminstrel: i also think that the underlying bug could be fixed anyway as soon as we move help dialog to gui2 (if that happens)
20170429 15:46:36< gfgtdf> celticminstrel: we coudl try to merge before next 1.13 release and see whethe it gives problems and revert otherwise.
20170429 15:46:37< celticminstrel> Well, we don't know how long it'll take to finish the GUI2 help, so if there's a bug it's probably good to fix it in case GUI2 help isn't ready for 1.14.
20170429 15:48:07< celticminstrel> gfgtdf: Also, do you have any idea what could be wrong with the last commit here? https://github.com/wesnoth/wesnoth/pull/993
20170429 15:49:02< gfgtdf> 'wrong' as in 'breaks existen mapgens' or just 'doesn't wokr as intended'
20170429 15:49:30< celticminstrel> I guess the latter.
20170429 15:50:41< gfgtdf> have to go afk now will think about it later
20170429 15:50:48< celticminstrel> Okay.
20170429 15:50:55 * celticminstrel is heading out for a bit too, actually.
20170429 15:56:38< zookeeper> celticminstrel, it's okay to merge if it works
20170429 15:56:57< zookeeper> (which, of course, i'd imagine it would)
20170429 16:01:44< zookeeper> i hope there's some progress being made WRT the migration away from gna seeing how it could go down pretty much anytime from now on.
20170429 16:01:46-!- gfgtdf [~chatzilla@x4e36a612.dyn.telefonica.de] has quit [Ping timeout: 268 seconds]
20170429 16:14:29< zookeeper> celticminstrel, was there some problem with making $|var substitution happen at display time in objectives?
20170429 16:52:38< celticminstrel> zookeeper: Probably not? IIRC that's what the first commit does.
20170429 16:53:30< zookeeper> right
20170429 16:54:41< zookeeper> so basically it substitutes variables when [objectives] is encountered (just like now), but it also substitutes them when displaying objectives, which then catches $|var (unlike now)?
20170429 16:55:11< celticminstrel> That sounds about right.
20170429 16:55:34< zookeeper> well, that's what i think the sensible behavior is
20170429 16:56:08< celticminstrel> So you don't think there's any point in re-substituting any [insert_tag] when the user requests the objectives?
20170429 16:56:15< zookeeper> that [insert_tag] is limited to the former only doesn't seem like a problem (worth mentioning it in the wiki though)
20170429 16:57:56< zookeeper> i mean sure, ideally if you had an [insert_tag] which inserts an [insert_tag], then the latter would be handled at display time, but... seems like an extremely rare case, and one that probably wouldn't be hard to workaround anyway
20170429 16:58:45< celticminstrel> Objectives are just a string in the engine.
20170429 16:59:12< celticminstrel> The config is just saved by the [objectives] tag so that [show_objective] can re-parse it to a string as needed.
20170429 16:59:13< zookeeper> right, yeah. not worth fretting over that limitation then, i'd say.
20170429 16:59:36< celticminstrel> So IOW... in your opinion the third commit is not needed.
20170429 17:00:34< zookeeper> i guess
20170429 17:01:24< zookeeper> although if the second commit makes delayed_variable_substitution=yes be the default then that shouldn't be in either
20170429 17:01:27< celticminstrel> So with PR 1004, the bats still have a third animation inherited from the base, but with the direction changed to null so it won't match...
20170429 17:01:53< celticminstrel> ...ooh.
20170429 17:01:55< zookeeper> (i mean, the default shouldn't be yes if you add the delayed_variable_substitution= key)
20170429 17:05:38< celticminstrel> Okay.
20170429 17:21:52-!- gfgtdf [~chatzilla@x4e36a612.dyn.telefonica.de] has joined #wesnoth-dev
20170429 17:31:51< celticminstrel> When using github conflict resolution, is squashing still an option?
20170429 17:37:26< gfgtdf> celticminstrel: hmm i din't even knoe there github conflict reolutions existed.
20170429 17:38:44< gfgtdf> celticminstrel: i wonder why config::clear() works iterativeley instead of the default recurion. the commit says 'to avoid a crash on very large invalid wml data' but i really don'T see how a recursive function is less likeley to crash.
20170429 17:42:08< irker568> wesnoth: gfgtdf wesnoth:master d405ec1c9d10 / src/ (config.cpp config.hpp): use boost::string_view in config class https://github.com/wesnoth/wesnoth/commit/d405ec1c9d108ba7baf469bdedede03fcd4c0c46
20170429 17:59:00< celticminstrel> gfgtdf: I'm not even quite sure what config::clear() is doing...
20170429 18:00:11< celticminstrel> Is it equivalent to calling clear() on all child configs?
20170429 18:01:50< celticminstrel> It does look like there's a stack there, implying a recursive solution would be possible and perhaps more readable.
20170429 18:02:05< gfgtdf> celticminstrel: config clear is what the dtor does, meaning is equivalent to c = config();
20170429 18:02:47< gfgtdf> will just use unique_ptr adn replace this stuff with children_.cclear()
20170429 18:03:10< celticminstrel> Yeah, it'd probably good to not use raw pointers in config anyway...
20170429 18:03:26< gfgtdf> hmm seems like the rtraivs c++14 buidl doesnt use my the heteogenous lookup optimisations icne it used gcc4.8
20170429 18:03:36< gfgtdf> maybe we coudl make a the c++14 buidl use clang insead ?
20170429 18:04:15< celticminstrel> Apparently there are no functions that return a config pointer.
20170429 18:05:06< celticminstrel> There's one that returns an attribute_value pointer. I think maybe it'd be better for it to return a reference though.
20170429 18:05:52< celticminstrel> No opinion on the C++14 build.
20170429 18:05:52< gfgtdf> celticminstrel: the point of that function is that it returns null when there is no such attribute
20170429 18:06:16< gfgtdf> celticminstrel: so you can write if(attribute_value* val = cfg.get("key")) { ..code ..}
20170429 18:06:17< celticminstrel> Sure, but there should be no difference between "attribute does not exist" and "attribute is set to boost::none", right?
20170429 18:06:35< celticminstrel> Oh, I see, that won't work in that idiom though.
20170429 18:06:52< celticminstrel> Wait, but that means that get() should probably return nullptr if the attribute exists and is none?
20170429 18:07:04< gfgtdf> celticminstrel: ye jut thought the same
20170429 18:07:05< celticminstrel> Or maybe it does already. I haven't looked at it.
20170429 18:09:36< irker568> wesnoth: gfgtdf wesnoth:gfgtdf-patch-3 3a2925f3bfbb / .travis.yml: use clang for c++14 tests https://github.com/wesnoth/wesnoth/commit/3a2925f3bfbba332217486f270b2ed4fa018824e
20170429 18:29:20< celticminstrel> Ugh, what's with this crash in PR990... :|
20170429 18:29:41< celticminstrel> It seems like something is calling p.native() and somehow the "this" pointer is passed wrong.
20170429 18:29:56< celticminstrel> In the caller, p clearly has the correct value.
20170429 18:30:04< celticminstrel> Step in, and "this" is set to all 1's.
20170429 18:30:20< celticminstrel> ie, 0xffffffff
20170429 18:54:36< celticminstrel> Somehow seems like the cause might've been that the file being loaded doesn't exist.
20170429 19:07:59< celticminstrel> I wonder if we could draw the Wesnoth logo to the screen while the GUI2 configs are loaded, so that Wesnoth doesn't just stay at a black screen for a few seconds... though maybe it's only noticeable in a debug build?
20170429 20:09:47-!- vultraz_iOS [uid24821@wesnoth/developer/vultraz] has joined #wesnoth-dev
20170429 20:12:18< zookeeper> vultraz_iOS, mind making me a project admin on gna? i might as well stash whatever information admins can access on mailing lists, since probably no one else has.
20170429 20:13:08< zookeeper> or, well, i dunno if i only need the list administrator password.
20170429 20:14:15< vultraz_iOS> done
20170429 20:15:14< celticminstrel> So we need to set up a new mailing list, right?
20170429 20:15:23< zookeeper> great. although... i guess being a project admin didn't help me
20170429 20:15:48< vultraz_iOS> i don't know
20170429 20:16:02< celticminstrel> Maybe Google Groups would work?
20170429 20:17:10< vultraz_iOS> what
20170429 20:17:15< vultraz_iOS> that still exists?
20170429 20:17:34< celticminstrel> I just looked at the Wikipedia article and the initial blurb didn't say anything about it not existing.
20170429 20:18:10< celticminstrel> Ah, but they seem to be no longer maintaining it, so maybe something else is better.
20170429 20:18:22< zookeeper> i just want to make sure that someone has the subscriber lists safely stored somewhere, because if/when gna goes down they need to be informed.
20170429 20:18:23< zookeeper> if we knew exactly when it's going down or if we knew how/where/when we're gonna move to something else, we could obviously use the mailing lists themselves to inform all subscribers, but more likely the lists will just be gone one morning.
20170429 20:18:24< vultraz_iOS> facebook? :P
20170429 20:18:47< celticminstrel> Ewww, no.
20170429 20:18:52< celticminstrel> Facebook is the devil.
20170429 20:19:00< zookeeper> agreed
20170429 20:19:30< celticminstrel> Wikipedia lists seven options for the mailing lists.
20170429 20:20:00< celticminstrel> Plus LISTSERV.
20170429 20:22:22< vultraz_iOS> do we *need* a ml?
20170429 20:22:34< vultraz_iOS> I feel MLs are antiquated 
20170429 20:22:37< celticminstrel> AFAIK the translators rely heavily on it, so...
20170429 20:22:43< celticminstrel> We can't just drop it arbitrarily.
20170429 20:24:23< zookeeper> vultraz_iOS, whenever you find yourself typing "i feel [...] is antiquated", you need to stop, hold down backspace with one hand and slap yourself silly with the other hand
20170429 20:24:43< vultraz_iOS> is it not antiquated?
20170429 20:24:53< zookeeper> now!
20170429 20:25:27< celticminstrel> Mailing lists do have at least one advantage that the alternatives don't.
20170429 20:25:42< celticminstrel> Which is that you can read the discussion without an internet connection.
20170429 20:26:59< celticminstrel> I'm guessing GNA was using GNU Mailman.
20170429 20:27:36< zookeeper> vultraz_iOS, ok you can stop slapping now
20170429 20:27:47< vultraz_iOS> hm? i was watching FNAF theories
20170429 20:27:52< zookeeper> :|
20170429 20:29:09< celticminstrel> Besides Mailman, DadaMail and Sympa look like decentl possibilities.
20170429 20:29:31< celticminstrel> There's also this: http://mlmmj.org/
20170429 20:30:43-!- RatArmy_ [~ratarmy@om126212245133.14.openmobile.ne.jp] has joined #wesnoth-dev
20170429 20:33:16< celticminstrel> zookeeper: Did you manage to get the subscriber lists then?
20170429 20:35:00< celticminstrel> ^decent
20170429 20:38:05< zookeeper> not yet. i'll ask ivanovic for the password(s) since he probably has them.
20170429 20:53:01-!- RatArmy_ [~ratarmy@om126212245133.14.openmobile.ne.jp] has quit [Ping timeout: 255 seconds]
20170429 21:02:47< gfgtdf> appveyor faisl becasue of outdatd prjectfiles it seems
20170429 21:02:50< gfgtdf> fails*
20170429 21:05:07< gfgtdf> Cannot open source file: '..\..\src\hotkey\hotkey_preferences_display.cpp': No such file or directory 
20170429 21:07:06< celticminstrel> gfgtdf: Which build is failing? I'm building on MSVC 2013 and the debug build works fine. Is it just the release build faiing?
20170429 21:07:18< celticminstrel> ^failing
20170429 21:07:22< celticminstrel> Oh, wait.
20170429 21:07:32< celticminstrel> Was that before or after Aginor's stuff was merged?
20170429 21:07:32< gfgtdf> just looked at the 2015  release build
20170429 21:07:42< celticminstrel> It could have been Aginor's PR that broke it.
20170429 21:08:32< celticminstrel> Hmm, doesn't look like his PR touched that file though...
20170429 21:08:33< gfgtdf> oh wait it seem i was wrong
20170429 21:08:48< gfgtdf> it says 'one month ago'
20170429 21:09:12< celticminstrel> Wait, does that mean Appveyor hasn't been building at all recently?
20170429 21:09:16< gfgtdf> guess that what you get form jstu lciking at the first entry of 'wesnoth appveyor' at gogle
20170429 21:09:29< celticminstrel> Or you were just looking at an old build?
20170429 21:09:36< JyrkiVesterinen> Just where are you looking? The latest VS2015 release build has succeeded.
20170429 21:09:38< JyrkiVesterinen> https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/job/u1l089wa7da8tbrl
20170429 21:09:48< gfgtdf> ye didnt realise i wa slooking at and old build
20170429 21:09:56< gfgtdf> JyrkiVesterinen: how can a i trigger a new build ?
20170429 21:10:13< celticminstrel> gfgtdf: Shall I try merging PR 1002? Or should we ask someone (maybe wedge009?) to test it out first?
20170429 21:10:46< JyrkiVesterinen> gfgtdf: You can't. Only I can trigger AppVeyor builds manually.
20170429 21:11:01< gfgtdf> JyrkiVesterinen: can't you enable everyone form the dev team to do it ?
20170429 21:11:01< JyrkiVesterinen> Want new builds? :)
20170429 21:11:21< JyrkiVesterinen> Maybe I can, but I don't know how to do it.
20170429 21:11:35< gfgtdf> JyrkiVesterinen: ye i want to know if it still cpompiles after https://github.com/wesnoth/wesnoth/commit/d405ec1c9d108ba7baf469bdedede03fcd4c0c46
20170429 21:11:41< JyrkiVesterinen> (It's past midnight here, so I'm not willing to learn right now either...)
20170429 21:11:43 * celticminstrel pokes gfgtdf 
20170429 21:12:07< gfgtdf> celticminstrel: if someone is willing to text it that woudl be nice
20170429 21:12:11< celticminstrel> Oh, vultraz_iOS might know.
20170429 21:12:27< celticminstrel> vultraz_iOS: When using github conflict resolution, do you know if you can still squash the result?
20170429 21:12:31< JyrkiVesterinen> All right, I triggered new builds.
20170429 21:12:40< vultraz_iOS> I don't know
20170429 21:12:49< celticminstrel> vultraz_iOS: I thought you'd used it before?
20170429 21:12:55< vultraz_iOS> no
20170429 21:12:58< celticminstrel> Oh.
20170429 21:14:35< gfgtdf> JyrkiVesterinen: it seems liek here https://ci.appveyor.com/team you can create a new 'role' that has access to the wensoth appveyor projects
20170429 21:14:48< gfgtdf> JyrkiVesterinen: and then somehow add the wesnoth memebers to it
20170429 21:15:33< celticminstrel> Does anyone know if PR1002 theoretically could have any other side-effects?
20170429 21:15:35-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170429 21:15:35< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Release gfgtdf d405ec1: use boost::string_view in config class Failed
20170429 21:15:35< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-321
20170429 21:15:39-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170429 21:15:52< JyrkiVesterinen> Okay, that didn't take long.
20170429 21:17:42< gfgtdf> ye i suspected it woudl fail but i didn't that it'D fail that early
20170429 21:17:48< gfgtdf> it'd*
20170429 21:19:02< gfgtdf> JyrkiVesterinen: you know which boost version it uses ?
20170429 21:19:13-!- Appveyor [~Appveyor@74.205.54.20] has joined #wesnoth-dev
20170429 21:19:13< Appveyor> The Battle for Wesnoth (Visual Studio 2015) - Debug gfgtdf d405ec1: use boost::string_view in config class Failed
20170429 21:19:13< Appveyor> Details: https://ci.appveyor.com/project/wesnoth/wesnoth-7lnpw/build/Wesnoth-VS2015-master-321
20170429 21:19:18-!- Appveyor [~Appveyor@74.205.54.20] has left #wesnoth-dev []
20170429 21:19:43< JyrkiVesterinen> It's 1.62.
20170429 21:20:13< celticminstrel> Missing header errors usually do appear early (if that's what it was, I didn't look).
20170429 21:22:43-!- Kwandulin [~Kwandulin@p200300760F6D80DFC912DB11B98971C7.dip0.t-ipconnect.de] has quit [Quit: [endlevel] result=novictorynodefeatjustquit [/endlevel]]
20170429 21:22:57< celticminstrel> gfgtdf: In 553d4e7 you should also delete the config_clear_state struct, I think.
20170429 21:24:44< gfgtdf> hmm thae one shodul have string_viw.hpp  http://www.boost.org/doc/libs/1_62_0/boost/utility/string_view.hpp 
20170429 21:24:47< gfgtdf> celticminstrel: ye
20170429 21:25:14< celticminstrel> gfgtdf: And in e004b4fe, shouldn't the USE_HETEROGENOUS_LOOKUPS define be moved back to config.hpp?
20170429 21:25:26< celticminstrel> Lines 51-81 of config_attribute_value.hpp
20170429 21:25:52< gfgtdf> ye
20170429 21:26:56< JyrkiVesterinen> gfgtdf: aquileia/external doesn't contain the entire Boost library.
20170429 21:27:08< JyrkiVesterinen> Only the parts which were found necessary to build Wesnoth.
20170429 21:28:01< celticminstrel> gfgtdf: You should be able to add it to aquileia/external, right?
20170429 21:28:12< celticminstrel> You seem to have write access there.
20170429 21:28:46< celticminstrel> (Maybe just update the entire VC15 branch to Boost 1.62 or something.)
20170429 21:29:27< gfgtdf> celticminstrel: its moslielly  header oynl lib so i oynl need to add some hpp files
20170429 21:30:38< celticminstrel> gfgtdf: Well, that depends.
20170429 21:30:52 * celticminstrel goes to check
20170429 21:31:05< celticminstrel> Wait, is it VC14 or VC15?
20170429 21:31:20< gfgtdf> celticminstrel: i'd think both? not sure though
20170429 21:31:37< celticminstrel> Okay, BOOST_VERSION is 106200
20170429 21:31:42< JyrkiVesterinen> AppVeyor build with VC12 (2013) and VC14 (2015).
20170429 21:31:52< celticminstrel> So as long as you add the header files from that same exact Boost version, there should be no problems.
20170429 21:31:54< JyrkiVesterinen> *builds
20170429 21:32:12< celticminstrel> I guess that's 1.62.0.
20170429 21:34:03-!- JyrkiVesterinen [~JyrkiVest@87-92-44-141.bb.dnainternet.fi] has quit [Quit: Going to bed]
20170429 21:56:28< zookeeper> okay... doesn't look like the mailing list subscriber lists are easily accessible even for list admins. the user list is split alphabetically, so one would have to browse through 27 pages to get all of them... per list. -.-
20170429 21:57:08< celticminstrel> You could use curl? >_>
20170429 21:57:33< zookeeper> well, sure
20170429 21:57:46< zookeeper> but that's not "easily" either
20170429 21:58:06< celticminstrel> I guess, but it might be a little easier than browsing to each page in the browser.
20170429 22:00:29< vultraz_iOS> Just GNA things
20170429 22:00:54< celticminstrel> Well, not quite sure this is GNA's software though. Might be GNU Mailman or something.
20170429 22:08:13-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has joined #wesnoth-dev
20170429 22:08:27< vultraz_iOS> We can definitely get rid of the commits ml
20170429 22:11:29< celticminstrel> I wonder if it has any subscribers.
20170429 22:12:33< celticminstrel> In particular, if it has any subscribers who are actually active or semi-active.
20170429 22:12:38-!- clavi [~clavi@v22017034422546657.goodsrv.de] has quit [Quit: ZNC - http://znc.in]
20170429 22:12:45< celticminstrel> (Including as packagers, translators, etc)
20170429 22:12:56< celticminstrel> (So, checking IRC and github activity is not enough.)
20170429 22:14:16< celticminstrel> I think the best way to handle it would be to send a message to that ML saying that, due to GNA going down, the ML may be discontinued, and tell people to contact us if they've been relying on it.
20170429 22:14:34< celticminstrel> If no-one contacts us by the time GNA goes down, we'll assume no-one cared.
20170429 22:15:07< celticminstrel> Of course, if someone contacts us later, it would probably be relatively easy to set it up again.
20170429 22:16:43-!- clavi [~clavi@v22017034422546657.goodsrv.de] has joined #wesnoth-dev
20170429 22:27:02-!- nickanc [5d33ac22@wikimedia/nickanc] has joined #wesnoth-dev
20170429 22:27:49< nickanc> question: how does the wml change between [combo] and 1.13 [choice]? havent found documentation for [choice].
20170429 22:28:13< celticminstrel> IIRC it's the same for both?
20170429 22:28:29< celticminstrel> This is for scenario options, right?
20170429 22:29:27< nickanc> yes
20170429 22:29:45< celticminstrel> The code treats [choice] and [combo[ identically.
20170429 22:30:06< celticminstrel> It contains [item] tags which contain name and value keys.
20170429 22:30:26 * celticminstrel just checked.
20170429 22:30:33< nickanc> ok
20170429 22:31:19< celticminstrel> In the future it's theoretically possible that [combo] will be reused with a different meaning.
20170429 22:31:49< nickanc> thank you. I am just asking things, because I am going through tons of errors so understanding things without testing everything is useful
20170429 22:33:33-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 22:38:26-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has joined #wesnoth-dev
20170429 22:46:16-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 246 seconds]
20170429 22:48:45< irker568> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 90fe59a63985 / / (28 files in 8 dirs): Don't store the user's password in plaintext in the prefs file https://github.com/wesnoth/wesnoth/commit/90fe59a63985b00a3d82550e7bb75333b53141e7
20170429 22:48:47< irker568> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 8548065b4043 / src/ (4 files in 2 dirs): Address specific feedback from Jyrki https://github.com/wesnoth/wesnoth/commit/8548065b404370e854b2b0e98255ed95ad2c7932
20170429 22:48:49< irker568> wesnoth: Celtic Minstrel wesnoth:login_save_crypt 53e425d7184d / src/credentials.cpp: Fix formatting to make Vultraz happy https://github.com/wesnoth/wesnoth/commit/53e425d7184d6ca82d9a11dd3e8e8556fe9502de
20170429 22:48:51< irker568> wesnoth: Celtic Minstrel wesnoth:login_save_crypt d6c834774022 / src/credentials.cpp: Address the rest of Jyrki's feedback - don't store the passwords all in one stri https://github.com/wesnoth/wesnoth/commit/d6c83477402211623135e989fe60b7a254bafa29
20170429 22:54:05-!- atarocch [~atarocch@c-24-61-135-102.hsd1.ma.comcast.net] has joined #wesnoth-dev
20170429 22:56:18< irker568> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables 57471ed68b4c / src/play_controller.cpp: Objectives: Always interpolate variables before showing https://github.com/wesnoth/wesnoth/commit/57471ed68b4c987ede637e815e82bedd8efb4ead
20170429 22:56:20< irker568> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables 8d547245a634 / data/lua/wml/objectives.lua: Objectives: Support delayed_variable_substitution (default false) https://github.com/wesnoth/wesnoth/commit/8d547245a63418b45fc45325eb0101da5f06931d
20170429 22:56:22< irker568> wesnoth: Celtic Minstrel wesnoth:objectives_with_variables fbd6fb028a7a / data/lua/wml/objectives.lua: Objectives: Interpolate variables in [show_objectives] if delayed_variable_subst https://github.com/wesnoth/wesnoth/commit/fbd6fb028a7a65f8105613b07b84ebae7f6b78d2
20170429 23:01:43-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
20170429 23:08:08< Aginor> any other hotkey related problems?
20170429 23:08:33 * celticminstrel hasn't even looked yet. >_>
20170429 23:11:26< Aginor> ok :D
20170429 23:12:04< nickanc> suppose to be inside a [lua] tag and inside code one does not << >> because he puts WML macroes and say a $variable . why the game executes before substituting $variable inside, i.e. lua cries for unexpected symblo near $?
20170429 23:12:16< nickanc> *why the code
20170429 23:13:25-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has joined #wesnoth-dev
20170429 23:13:49< nickanc> suppose code=function({X},{Y},$side_number)
20170429 23:14:07< gfgtdf> nickanc: that doesnt work
20170429 23:14:17< gfgtdf> code= does not support varoiable subsitution
20170429 23:14:43< gfgtdf> nickanc: you hvae to write: code=function({X},{Y},wesnoth.get_variable("side_number"))
20170429 23:15:10< gfgtdf> nickanc: if you setup your lua as descibed in the wiki you cal also write code=function({X},{Y},V.side_number)
20170429 23:15:26< gfgtdf> nickanc: do you work on 1.13 or 1.12 =
20170429 23:15:28< gfgtdf> ?
20170429 23:15:41< nickanc> 1.13
20170429 23:16:28< gfgtdf> its really reccomened to use code << >>for lua code
20170429 23:16:39< gfgtdf> you can then use [args] to pass variables
20170429 23:18:21< gfgtdf> like https://pastebin.com/dQr0ztME
20170429 23:18:38< gfgtdf> not exactly usre on the syntax it migth brackets {} aorund the ...
20170429 23:19:04-!- atarocch [~atarocch@c-24-61-135-102.hsd1.ma.comcast.net] has quit [Ping timeout: 260 seconds]
20170429 23:19:28< gfgtdf> nore: or even better you can creat a custom wml tag in a that you then call later with paramters
20170429 23:20:49< nickanc> gfgtdf, thanks!
20170429 23:21:01< nickanc> I am gonna try, maybe it solves my problem
20170429 23:37:26< nickanc> gfgtdf it works, thank you! now 1.13 can have a 6p Orocia as 1.8 did.
20170429 23:45:19< gfgtdf> no problem
20170429 23:57:36-!- RatArmy_ [~ratarmy@om126200120136.15.openmobile.ne.jp] has quit [Read error: Connection reset by peer]
--- Log closed Sun Apr 30 00:00:47 2017