--- Log opened Mon Nov 17 00:00:47 2014
20141117 00:05:43< vultraz> shadowm: just got the missing race icons from LB https://www.dropbox.com/s/hx8o3wzinsfltuw/missing%20icons.7z?dl=0
20141117 00:05:48< vultraz> can you commit them
20141117 00:07:22< vultraz> (not the Khalifate, we already have that, he just included it in case he hadn't)
20141117 00:07:35< shadowm> vultraz: Why can't you do it?
20141117 00:08:01< vultraz> Because I hadn't bothered timotei_ to BTS me a clone yet
20141117 00:08:03< vultraz> haven't*
20141117 00:08:38< shadowm> Why haven't you asked me to update the tarball on wesnoth.org instead?
20141117 00:09:31< vultraz> Because I assumed I've asked you to do enough recently?
20141117 00:09:55< shadowm> I don't know how to add these icons.
20141117 00:10:22< shadowm> Also, what you just said contradicts what you asked me above.
20141117 00:12:03< vultraz> Oh, hell.
20141117 00:12:13-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 00:12:35< vultraz> Well, it appears LB did commit a falcon icon awhile back, but named it wrong.
20141117 00:13:07< vultraz> (and where did all the @2x images go?)
20141117 00:13:12< gfgtdf> shadowm: did you see my message above about server update ?
20141117 00:13:50< shadowm> Yes, I rebuilt the server about half an hour ago, forgot to mention that.
20141117 00:14:25< shadowm> master, I assume.
20141117 00:15:02< gfgtdf> shadowm: yes master
20141117 00:16:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 00:16:03< gfgtdf> shadowm: ty
20141117 00:16:30-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20141117 00:17:29< vultraz> shadowm: so, yes, I am an idiot. Seems he did it before but named it wrong. Just rename the damn things.
20141117 00:23:00< irker839> wesnoth: Ignacio R. Morelle wesnoth:master 53833dd0c3b0 / src/game_config.cpp: Remove pointless debug output, part of the team color implementation http://git.io/65oh-w
20141117 00:23:03< irker839> wesnoth: Ignacio R. Morelle wesnoth:master a36bc41e3655 / / (5 files in 2 dirs): Rename falcon race icon so the editor can actually find it http://git.io/i85dcA
20141117 00:23:06< irker839> wesnoth: Ignacio R. Morelle wesnoth:1.12 5aa1e9c40a1e / / (5 files in 2 dirs): Rename falcon race icon so the editor can actually find it http://git.io/gvmQFQ
20141117 00:24:36< shadowm> fabi: Could we discuss some time how that joystick configuration dialog that doesn't exist yet should work? A mock-up (even hand-drawn) would also help.
20141117 00:24:53< vultraz> Thanks
20141117 00:25:05< fabi> shadowm: yes, with pleasure. I have some minutes right now if you like.
20141117 00:25:29< shadowm> Well, I'm going afk right now, but that doesn't stop you from giving me the infodump in the meantime.
20141117 00:25:48< fabi> okay
20141117 00:31:43< fabi> There needs to be a boolean option "joystick support on/off".
20141117 00:32:16< fabi> Can be a checkbox. Maybe even an item in the advanced preferences is enoug.
20141117 00:33:41< fabi> The buttons of the joystick can already be assigned with the current hotkey preferences dialog. I don't see any problems there.
20141117 00:35:51< fabi> Then it comes to the joystick axes.
20141117 00:40:20< shadowm> Since the Hotkeys button that gives access to the hotkey preferences is in General, that checkbox should be there as well.
20141117 00:40:44< fabi> yes
20141117 00:41:14< fabi> To controll the axes we need to define a hotkey equivalent.
20141117 00:42:11< fabi> a control item that does not consume a "have been triggered" event like the normal hotkeys do but one that consumes a axis value.
20141117 00:43:00< fabi> The hotkey preferences dialog could be modified to configure them as well.
20141117 00:43:20< fabi> But I don't know if it is a good idea to merge the two types of input methods.
20141117 00:45:11< shadowm> Okay, that's the part where I get lost, even if I look at the commented-out advanced preference entries. What needs to be configured about the axes?
20141117 00:46:16< shadowm> I see options for an X axis, Y axis, "mouse X axis", and "mouse Y axis", each ranging from -1 to 7.
20141117 00:46:31< shadowm> Oh also a "thrust axis".
20141117 00:46:34< fabi> Example: One of the new items would be: scroll North/Sourth
20141117 00:46:43-!- happygrue [~Laptop@wesnoth/developer/wintermute] has joined #wesnoth-dev
20141117 00:46:54< fabi> another one would be scroll East/West
20141117 00:47:43< fabi> A pair for the mouse emulation.
20141117 00:47:59< fabi> Another pair for controllung the cursor directly. 
20141117 00:49:23< fabi> Like you assign keyboard/mouse/joypad buttons to one of our conventional hotkey actions the axis/control item thing works. 
20141117 00:49:39< fabi> -1 means disabled.
20141117 00:49:51< fabi> 0 to 7 is the number of a joysticks axis.
20141117 00:50:36< fabi> The hotkey dialog would not ask for a key to be pressed for assignment. 
20141117 00:50:43< fabi> But ask for an axis to be moved.
20141117 00:55:25< shadowm> Hm.
20141117 00:55:42< shadowm> I think I'm getting the general idea.
20141117 00:56:10< shadowm> fabi: How do multiple joysticks factor into this, though? (At least the commented-out preferences suggest they do.)
20141117 00:57:00< fabi> Every modern gamepad consists of 6 axes.
20141117 00:57:26< fabi> 4 of them come with the 2 sticks. The other 2 are thrusters.
20141117 00:58:23< fabi> Some systems/drivers/whatever handle them as one joystick with 6 axes. Some handle them as 3 joysticks with 2 axes each.
20141117 00:58:44< fabi> But you also want to attack more than one joystick when you play wesnoth in hotseat mode.
20141117 00:58:57< fabi> And every player might want to have a different control scheme. 
20141117 00:59:09< fabi> Thus the system needs to handle multiple joysticks.
20141117 00:59:37< vultraz> I still fail to see how you could play wesnoth with a joystick at all
20141117 00:59:41< vultraz> Even theoretically
20141117 00:59:46< shadowm> faib: So for example, four axes from two different joysticks may be used for mouse emulation?
20141117 00:59:50< shadowm> fabi:
20141117 01:00:01< shadowm> (Two for each joystick.)
20141117 01:00:07< fabi> yes
20141117 01:01:04< iceiceice> vultraz: i think the joystick just works as a mouse
20141117 01:01:19< vultraz> How so?
20141117 01:01:39< fabi> One of the sticks is used to scroll.
20141117 01:01:49< fabi> This is already implemented and work pretty well.
20141117 01:02:09< iceiceice> vultraz: it would work just like any console game when you are looking at a menu
20141117 01:02:28< shadowm> fabi: So, sticking to the mouse emulation case in particular, I guess one axis scrolls vertically and the other horizontally. Should it be possible for the player to assign, say, axis 1 from joystick 1 for vertical motion and axis 1 from joystick 2 for horizontal?
20141117 01:02:29< fabi> I think it is a good default to give one axes pair the scrolling function.
20141117 01:02:48< vultraz> iceiceice: ah, I see
20141117 01:03:52< fabi> shadowm: Well, most games I know allow for free axis assignment, at least if they come from the same joystick. You can't exactly know which axis is which number. 
20141117 01:04:21< fabi> Thus building logical axes pairs is difficult.
20141117 01:05:27< fabi> But I think in case of Wesnoth it does not make much sense to allow one of the function dimension on one and one on the other joystick device.
20141117 01:05:35< fabi> This is not true for all games. 
20141117 01:05:49< fabi> Especially simulations benefit from less strict rules.
20141117 01:07:04< fabi> vultraz: iceiceice is only partly right here. You can play Wesnoth with mouse emulation. Then the joystick just moves the mouse cursor around.
20141117 01:07:18< shadowm> But you said above that multiple sticks from the same joystick may actually be implemented as separate joysticks.
20141117 01:07:45< fabi> Yes, but then they are not on the same stick.
20141117 01:07:56< shadowm> So I guess at the code level every assignment should be handled as a value consisting of a (joystick, axis) value pair?
20141117 01:08:12< fabi> I mean you don't group the north south axis of stick a on joy dev 0 and the west/east axis of stick a on joy dev 1.
20141117 01:08:28< fabi> yes
20141117 01:08:39< shadowm> Right.
20141117 01:09:53< fabi> vultraz, iceiceice: But I also coded a control mode where the joystick directly moves the highlighted hex around.
20141117 01:10:48< iceiceice> fabi: hmm yeah its worth experimenting
20141117 01:11:02< iceiceice> i guess you could have a mode where joy1 pans the screen around, and joy2 directly controls hex position (not mouse emulation)
20141117 01:11:10< fabi> yes
20141117 01:11:12< iceiceice> for appropriate zoom levels that might be very convenient
20141117 01:11:20< fabi> That is already implemented.
20141117 01:11:22< iceiceice> nice :)
20141117 01:11:25< fabi> The only problem is: 
20141117 01:11:43< fabi> You can't play Wesnoth right know because the attack dialog is only mouse aware.
20141117 01:12:25< iceiceice> ah
20141117 01:12:35< iceiceice> so you need to make the "show_dialog" functions cause it to fall back to mouse mode
20141117 01:13:31< fabi> iceiceice, vultraz: BattleIsle1 is an example of a hex/turn based game controlled by joystick (although digital ones). BattleIsle2 is a good example of a hex/turn based game controlled best with keyboard.
20141117 01:14:34< fabi> iceiceice: I need gui1 and gui2 widgets (mostly buttons) which are joystick axes and hotkey aware.
20141117 01:15:17< fabi> The coding should be universal. I want to implement both, gamepad and keyboard support with it.
20141117 01:17:04-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 01:20:21-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20141117 01:29:47-!- stikonas__ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 01:30:30-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20141117 01:31:32-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 01:34:57-!- stikonas__ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20141117 01:35:09-!- stikonas__ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 01:41:52-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 240 seconds]
20141117 01:42:07< shadowm> fabi: What's the difference between cursor movement and mouse emulation in this context?
20141117 01:47:04-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 01:52:33-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 01:53:28-!- stikonas__ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20141117 01:58:14-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 02:02:06-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20141117 02:07:37-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 02:10:57-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141117 02:13:04-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 244 seconds]
20141117 02:14:25-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 02:14:30-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 02:14:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20141117 02:14:31< gfgtdf> iceiceice: you know where (or whether) we send the [wait_global] which we wait for in here: https://github.com/wesnoth/wesnoth/blob/master/src/server/server.cpp#L2670 ?
20141117 02:14:40-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141117 02:16:01< gfgtdf> iceiceice: forget it, i foudn out
20141117 02:27:48< irker839> wesnoth: gfgtdf wesnoth:master 59860ca9b2bc / src/server/ (game.cpp game.hpp server.cpp): remove unused wesnothd code http://git.io/5jRLog
20141117 02:41:41-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 02:59:39-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 03:00:19-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141117 03:01:50-!- Ivanovic_ [~ivanovic@frnk-5f74f730.pool.mediaWays.net] has joined #wesnoth-dev
20141117 03:01:50-!- Ivanovic_ [~ivanovic@frnk-5f74f730.pool.mediaWays.net] has quit [Changing host]
20141117 03:01:50-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20141117 03:04:02< iceiceice> does anyone know if the 1.12 add-on server is available yet?
20141117 03:04:34 * shadowm points iceiceice to the trailing section of the 1.11.19 announcement.
20141117 03:05:35-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 265 seconds]
20141117 03:05:46-!- Ivanovic_ is now known as Ivanovic
20141117 03:06:28 * iceiceice thanks shadowm
20141117 03:13:09< iceiceice> shadowm: hmm for some reason, my 1.11 can see the 1.12 version
20141117 03:13:40< shadowm> What specific version does "1.11" stand for in that sentence?
20141117 03:14:26< iceiceice> hmm i'll check what commit, it was shortly before 1.12 was tagged
20141117 03:14:34< iceiceice> i guess that could be the problem
20141117 03:14:39< shadowm> "The add-ons server for 1.12.x is already running. It was started for 1.11.10 and it will serve all future beta and RC releases from this series, as well as all releases in the stable 1.12.x series."
20141117 03:14:44< shadowm> 1.11.10 = beta 1
20141117 03:14:49< shadowm> 1.11.16 = RC 1
20141117 03:15:19< iceiceice> i see
20141117 03:16:01< iceiceice> okay i didn't understand how it worked, i thought after hte formal release everyone would reupload all the add-ons
20141117 03:16:35< shadowm> If it worked that way then I wouldn't have been so adamant about API stability in beta and RC phase.
20141117 03:17:00< iceiceice> its also a bit confusing because it means that some versions use my 1.11 data folder but the 1.12 add-on server
20141117 03:22:30-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 258 seconds]
20141117 03:26:14< iceiceice> shadowm: do you know who Dugi is talking about in this thread?
20141117 03:26:16< iceiceice> http://forums.wesnoth.org/viewtopic.php?f=4&t=41129
20141117 03:26:44< shadowm> Why would I?
20141117 03:27:03< iceiceice> im just asking, because i know the number of people who are pushing C++ these days is pretty small
20141117 03:27:21< iceiceice> gfgtdf: do you know?
20141117 03:27:24< shadowm> What are you asking me about? He said "translator", not "developer".
20141117 03:27:57< iceiceice> Dugi: "I was told that somebody (maybe zookeeper, I wasn't told this exactly in that PM) is looking at it already, but I have no more information how is that going"
20141117 03:28:09< shadowm> Ah, not the OP then.
20141117 03:28:16< iceiceice> yeah, sorry for being unclear
20141117 03:28:37< shadowm> I have no idea.
20141117 03:28:42< gfgtdf> iceiceice: me neigher
20141117 03:29:16< shadowm> I'm sure you could ask him directly, though.
20141117 03:29:33< iceiceice> well apparently he doesn't know either
20141117 03:29:58< shadowm> He was told *something* in a PM. Surely he knows who PM'd him.
20141117 03:30:06< iceiceice> i think maybe it refers to, at some point "figure out if Dugi's bug report is real and fix it" was on the must fix 1.12.0 list
20141117 03:30:16< shadowm> Then we can ask that person if they know who is working on it, assuming they even exist.
20141117 03:30:32< shadowm> Was it?
20141117 03:30:37< iceiceice> in an email
20141117 03:31:23< shadowm> "Verify if textdomain bug reported by Dugi is reproducible, and fix if it is (volunteers?)" -- mattsc, 2014/10/28
20141117 03:32:59< iceiceice> anyways i'm reasonably convinced something weird is going on, i think i will try to figure out for 1.12.1
20141117 03:33:14< shadowm> I'd like to handle it.
20141117 03:33:20< iceiceice> ok, be my guest
20141117 03:33:30< shadowm> But first I want to see a proper test case and bug report.
20141117 03:34:36< iceiceice> y that's what i said to dugi basically, maybe he will make one
20141117 03:34:56< iceiceice> but even if he doesn't it might be easy to spot the problem in the preprocessor code
20141117 03:35:23< iceiceice> there is a reasonbly simple way to reproduce using the swamplings add-on
20141117 03:35:35< iceiceice> (on 1.12)
20141117 03:35:43< iceiceice> anyways, i am out for now, i will be back later maybe
20141117 03:35:45-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141117 03:36:29< shadowm> I'm not quite willing to download a whole add-on to test a punctual bug.
20141117 03:40:23-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 03:41:15-!- kex [~kex@78.157.29.160] has quit []
20141117 03:48:26-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 03:50:15< happygrue> hmmm, getting a turn one lua error and then says the game is over when I am trying to spec a UMC game on the server.
20141117 03:50:37< happygrue> but anyway, is anyone around to come on the 1.11.19 server for a moment and test something for about 2 minutes?
20141117 03:50:40< gfgtdf> happygrue: on master ?
20141117 03:50:55< happygrue> I'm on the 1.11.19 server
20141117 03:51:40< gfgtdf> happygrue: what does the lua error say ?
20141117 03:51:59< gfgtdf> happygrue: does that UMC use a lot of lua ?
20141117 03:52:04< shadowm> happygrue: What game causes it?
20141117 03:54:16< happygrue> the WC II game caused it, but they reloaded it
20141117 03:54:23< happygrue> so the game that did is no longer up
20141117 03:55:02< shadowm> happygrue: It requires an era, did you have it installed? (Admins can bypass the era requirement and join and potentially crash like that.)
20141117 03:55:06< happygrue> stderr doesn't actually say anything about lua, but looks like just OOS stuff
20141117 03:55:37-!- Appleman1234 [~Appleman1@rrcs-97-79-164-178.sw.biz.rr.com] has joined #wesnoth-dev
20141117 03:55:47-!- kex [~kex@78.157.29.160] has quit [Remote host closed the connection]
20141117 03:55:54< happygrue> just checking to see if this is something they know about
20141117 03:56:39< happygrue> apparently it is, after map1 in mp campaigns, observers always oos
20141117 03:56:50< happygrue> so they reload if someone wants to watch I guess
20141117 03:57:10< happygrue> though I think I watched one fine just the other day
20141117 03:57:17< happygrue> maybe it was the first map still..., hmmm.
20141117 03:57:29< gfgtdf> happygrue: they start observing in the next scenario or they advance to the next scenario liek the players ?
20141117 03:58:19< gfgtdf> happygrue: i recently talked with iceiceice about a bug that might bugs of observer starts obsering a non-first scenario.
20141117 03:58:52< happygrue> this was a game going on, the 3rd or 4th map of an mp campaign, about turn 7.  I tried to observe and got oos but I can't find what it said about lua yet
20141117 03:59:18< happygrue> can someone else come observe the game where I am?  I want to test if spectator chat is broken
20141117 03:59:23< gfgtdf> happygrue: did you get teh OOS suring start or prestart wml event ?
20141117 03:59:29-!- Ivanovic_ [~ivanovic@frnk-d9333302.pool.mediaways.net] has joined #wesnoth-dev
20141117 03:59:29-!- Ivanovic_ [~ivanovic@frnk-d9333302.pool.mediaways.net] has quit [Changing host]
20141117 03:59:29-!- Ivanovic_ [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20141117 04:00:00< happygrue> gfgtdf: it was saying turn 1, but I'm not sure if it was prestart. Let me see if I can't get it again.
20141117 04:01:40< gfgtdf> happygrue: there is a known bug when one advances to the next scnario teh serve rmight have a wrong data about controllers in the first turn of a new scenario, We currently dont know whether it only effects games where the host is not player on or all games.
20141117 04:02:46< happygrue> it seems pre start, according to those who are getting the bug regularly (UMC author)
20141117 04:03:34-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 255 seconds]
20141117 04:05:14-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 256 seconds]
20141117 04:05:25-!- Ivanovic_ is now known as Ivanovic
20141117 04:05:31< happygrue> shadowm: regarding the eras, WC II is an era the just mashes up default units, so there is no problem there, but I do have it in any case.
20141117 04:07:53-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141117 04:08:54< gfgtdf> i go to bed now
20141117 04:08:55-!- gfgtdf [~chatzilla@f054144148.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.91 [Firefox 33.1/20141106120505]]
20141117 04:10:24-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141117 04:12:03-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 04:19:10< shadowm> Blargh fine I'll download this stupid add-on.
20141117 04:19:34< shadowm> It's one of *those* bugs.
20141117 04:21:55< shadowm> iceiceice: WHy do you include ostream explicitly in src/gui/dialogs/lua_interpreter.cpp when you don't use its definitions directly and basic_ostream is an implementation prerequisite of basic_ostringstream (from sstream)?
20141117 04:22:56< shadowm> (Meaning there is no way sstream will expose basic_ostringstream without basic_ostream unless you are dealing with a completely broken standard library implementation.)
20141117 04:24:40-!- happygrue [~Laptop@wesnoth/developer/wintermute] has quit [Remote host closed the connection]
20141117 04:25:42< shadowm> Also, GUI2's design privileges WML-specified property values over C++-mandated ones. Your id=msg scroll_label has its vertical_scrollbar_mode property set in C++ when it's supposed to be settable by WML too: http://wiki.wesnoth.org/GUIWidgetInstanceWML#Scroll_label
20141117 04:26:00< shadowm> Also, that wiki documentation block is completely broken.
20141117 04:33:13-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 04:42:38 * shadowm gives the raw preprocessor output a weird look.
20141117 04:48:04-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 04:56:33-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: Ik ga weg]
20141117 05:44:16-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 05:49:08-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 264 seconds]
20141117 05:53:26< fabi> shadowm: I call "cursor" the highlighted hex. 
20141117 05:54:05< fabi> Meaning you can move the hex field that is the "active" one with the gamepad without moving the mouse pointer.
20141117 06:06:47-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141117 06:07:24< iceiceice> shadowm: re: ostream: that is a good question
20141117 06:07:30< iceiceice> re: wiki block,
20141117 06:07:57< iceiceice> pretty much all the comments are broken right now in that file, i mean for another thing i attributed the .cpp file copyright to crab which is clearly not accurate
20141117 06:08:25< iceiceice> i plan to refactor and move all the implementation details into a private subclass and out of the header.
20141117 06:08:42< iceiceice> re gui2-design
20141117 06:08:52< iceiceice> first of all where was i supposed to learn this
20141117 06:09:07< iceiceice> second of all, it doesn't make that much sense to me in this particular case because the lua interpreter is a window and not a widget
20141117 06:09:27< iceiceice> i don't intend that someone will instantiate it with different scroll_label properties, because frankly it won't work as well
20141117 06:09:55< iceiceice> it currently still has some annoying bugs caused by bad scroll_label behavior that i have been unable to solve
20141117 06:10:47< iceiceice> if the window ever has a resize event (because of a new longest line in the console log), it forces the vertical scroll bar to be reset to the top, ignoring my "move to the bottom" command.
20141117 06:10:53< shadowm> "first of all where was i supposed to learn this" By getting told off.
20141117 06:11:04< shadowm> That's how I learned, anyway.
20141117 06:11:18< iceiceice> okay, fair enoguh
20141117 06:11:48< iceiceice> when i mentioned some of this stuff in channel before, gfgtdf said you also had been complaining about messed up scroll label behavior,
20141117 06:11:50< shadowm> You do reaise a valid point.
20141117 06:11:52< shadowm> *raise
20141117 06:12:09< iceiceice> what is the probelm that you were having? i'm curious if its basically the same thing
20141117 06:12:31< shadowm> Uh.
20141117 06:12:39< iceiceice> in this particular case it's really annoyning, if the console ever has an extremely long line printed,
20141117 06:13:00< iceiceice> then for the rest of time the scroll label will scroll to the top instead of the bottom after every command you make
20141117 06:13:13< iceiceice> that's the only reason i'm manually wrapping the tab completion results to 80 chars
20141117 06:13:46< fabi> iceiceice: Hello. Do you have a few minutes?
20141117 06:13:47< shadowm> I don't really recall voicing any complaints about scroll_label in particular, just about the experimental MP lobby's bug with the chat log getting scrolled back to top. I didn't really stop to think whether it uses a scroll_label or not.
20141117 06:14:10< shadowm> Though that also happens to other scrollable widgets there, IIRC.
20141117 06:14:10< iceiceice> ok, it sounds like the same thing though
20141117 06:14:38< iceiceice> it seems to me that if the scroll label ever requests more horizontal space and gets denied,
20141117 06:14:46< iceiceice> it has a hissyfit and scrolls itself to the top
20141117 06:15:16< shadowm> I do vaguely remember asking for a public interface to scroll a widget (may have been a listbox) and getting a negative response.
20141117 06:16:13< shadowm> Not sure if it was about the interface not existing (it clearly doesn't) or about its hypothetical existence being a violation of some nebulous GUI2 design guideline I'm unaware of.
20141117 06:16:39< iceiceice> shadowm: there is some way to do it that i found,
20141117 06:17:34< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/gui/widgets/scrollbar_container.hpp#L183
20141117 06:17:56< iceiceice> that's a public method that lets you position the scroll bar as you like
20141117 06:18:18< iceiceice> and it seems to work, except apparently in connection with resize events
20141117 06:20:33< shadowm> Can't you make sure somehow that the layout is recalculated *before* calling that?
20141117 06:20:59< shadowm> I seem to recall finding a way to force a GUI2 dialog to do that for some other purpose.
20141117 06:24:13-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141117 06:27:45< zookeeper> iceiceice, i know dugi couldn't have been talking about me
20141117 06:28:00< iceiceice> shadowm: i really have no idea
20141117 06:30:27< iceiceice> zookeeper: ok
20141117 06:32:36-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141117 06:33:46< fabi> shadowm: The wmlindent run is causing a huge merge party for me. I hate you.
20141117 06:33:54< fabi> :-)
20141117 06:34:18< shadowm> That's why you should configure your editor to indent WML correctly.
20141117 06:34:51< fabi> It is not an editor problem.
20141117 06:35:32< fabi> A gsoc student removed the additional indentation layer caused by the MP hack. He forgot to correct the indentation. 
20141117 06:35:56< shadowm> The diff for 39e2d710bddfefb2bce2842b0aefaaf6a171b935 shows the original had tabs and trailing spaces.
20141117 06:37:11< shadowm> I assume that's what's conflicting for you, anyway, because it's my most recent pass.
20141117 06:37:36< fabi> Well, the point is: Ask before you touch LoW, please.
20141117 06:38:52< shadowm> Yes, but make sure to configure your editor correctly as well.
20141117 06:39:05< fabi> Yes. I will have a look at it.
20141117 06:39:38< fabi> Since I am using our emacs mode it must be a bug in it.
20141117 06:40:41< shadowm> That also happened less than an hour before 1.12.0 was tagged for the express purpose of having 1.12.0 run wmllint and wmlindent-clean. I intend to do the same for future stable releases as long as I remember.
20141117 06:41:30< shadowm> (Which is improbable.)
20141117 06:42:20< fabi> I thought about using the tool myself but stepped away from it because I new it would cause me troubles.
20141117 06:42:24< fabi> s/new/knew
20141117 06:44:31-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 07:06:54-!- mjs-de [~mjs-de@f048067162.adsl.alicedsl.de] has joined #wesnoth-dev
20141117 07:33:21-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 07:36:45< shadowm> iceiceice: I *think* I see the problem (also got a much more manageable test case).
20141117 07:37:00< shadowm> Eh. The problem... with macros and textdomains.
20141117 07:38:15-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 272 seconds]
20141117 07:38:36< shadowm> https://gist.github.com/shikadilord/6052c74cd7f4c200fbcb <-- Guess what textdomain applies to the string literal in line 21?
20141117 07:48:02-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141117 07:53:27-!- boucman_work [~jrosen@193.56.60.161] has joined #wesnoth-dev
20141117 07:53:27-!- boucman_work [~jrosen@193.56.60.161] has quit [Changing host]
20141117 07:53:27-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141117 08:05:34-!- Jetrel [~Jetrel@c-73-164-34-22.hsd1.mn.comcast.net] has quit [Quit: "The highest possible stage in moral culture is when we recognize that we ought to control our thoughts." - Charles Darwin]
20141117 08:06:21-!- Jetrel [~Jetrel@c-73-164-34-22.hsd1.mn.comcast.net] has joined #wesnoth-dev
20141117 08:11:06< iceiceice> shadowm: link is dead, is it private or something?
20141117 08:15:06< shadowm> No, I just went and reduced it a bit more.
20141117 08:15:16< shadowm> https://gist.github.com/shikadilord/7b31fefb0aca49434e52 <-- Line 17. Guess the textdomain.
20141117 08:16:25< shadowm> Whoever guesses it right gets a bag of cookies.
20141117 08:18:10< iceiceice> hmm well i guess 17 is in __CORE_TEXT_DOMAIN
20141117 08:18:23< iceiceice> based on having observed the bug in practice...
20141117 08:18:45 * shadowm shows iceiceice the bag of cookies.
20141117 08:19:07< shadowm> Nope.
20141117 08:19:14< iceiceice> are either of them in ZZZZZ?
20141117 08:19:29< shadowm> I said line 17.
20141117 08:19:42< shadowm> It's not ZZZZZZZZ.
20141117 08:20:54< shadowm> And here's the shocker: it's not __CORE_TEXT_DOMAIN__ either.
20141117 08:21:19< iceiceice> its #wesnoth or something?
20141117 08:21:26< shadowm> Neither it is __CORE_TEXTDOMAIN__, which is the actual thing.
20141117 08:21:49< shadowm> It's not #wesnoth either.
20141117 08:21:55< shadowm> It's wesnoth.
20141117 08:22:34< iceiceice> poor little preprocessor
20141117 08:22:56< shadowm> Comment out lines 13 and 19 and you get ZZZZZZZZ on line 17.
20141117 08:23:45< shadowm> I have no idea where that 'wesnoth' textdomain is coming from, because the first textdomain seen is __CORE_TEXTDOMAIN__.
20141117 08:25:32< iceiceice> okay, i will be back in a bit, i'm glad you looked into this
20141117 08:25:53< shadowm> Later when I have energy to read C++ I'll add some probes to figure out what's going through its head, but in any case the problem appears to be specific nested macros.
20141117 08:26:10< shadowm> For example, if you comment out line 15, you also get the correct textdomain.
20141117 08:26:28< shadowm> (Line 15 commented out with lines 13 and 19 not commented.)
20141117 08:26:52< shadowm> So I guess nested macros corrupt the preprocessor context in some way.
20141117 08:27:12< shadowm> I imagine 'wesnoth' is a default-initialized value coming from somewhere.
20141117 08:28:27< shadowm> In general, every time you substitute a macro or macro argument, the preprocessor has to leave <FE>textdomain directives for the parser both at the start and end of the substitution to set/reset the textdomain for the context.
20141117 08:29:30< shadowm> Also, apparently, #define directives turn into waste.
20141117 08:30:15< shadowm> A #define block until its matching #enddef gets turned into a block full of empty lines instead of skipping it altogether and adding a trailing <FE>line directive.
20141117 08:30:55< shadowm> There's probably a small optimization opportunity there I could look into for 1.14.
20141117 08:31:42< shadowm> (Sure, it doesn't usually matter, but then you have whole directories consisting of nothing but macro definitions in data/core/macros and data/core/terrain-graphics.)
20141117 08:34:11< shadowm> OTOH I guess <FE>line would have a larger footprint in the average case.
20141117 08:36:00< shadowm> Yeah, it's a bad idea, actually. Detecting void blocks to coalesce would be a waste of time.
20141117 08:37:54< shadowm> Still, we have 13 KiB of emptiness with only a few interspersed file-start <FE>lines at the start of the data/core WML document.
20141117 08:48:35< iceiceice> shadowm: does the preprocessor expand the tree of macros in pre-order?
20141117 08:48:48< iceiceice> maybe it would be better if it operated in post-order
20141117 08:49:00< shadowm> I don't know what either thing means.
20141117 08:49:23< iceiceice> right so,
20141117 08:49:45< iceiceice> "pre order" means my recursive algorithm has the structure "do something at the current node, then do each of its children"
20141117 08:49:53< iceiceice> "post order" means "do each of my children, then do me"
20141117 08:52:28< iceiceice> i guess i need to read the code to figure out if this question makes sense
20141117 08:52:43< shadowm> I think it substitutes crap in-place.
20141117 08:53:13< iceiceice> so i think that ideally, it wouldn't have to leave text domain directives around
20141117 08:53:17< shadowm> The preprocessor doesn't really have a notion of a tree AFAIK, it just substitutes whatever needs to be substituted. And that's why #define incurs in no substitutions on its own.
20141117 08:53:53< shadowm> That's also why you can refer to nonexistent macros in your #define block and everything is fine as long as they are defined later before your #define is required to be substituted for the first time.
20141117 08:53:57< iceiceice> right so, the easiest way to figure out what text domain a string is supposed to be in, is if you look at the file where it originally appeared
20141117 08:54:24< shadowm> The parser usually doesn't read files.
20141117 08:54:33< iceiceice> y but
20141117 08:54:37< shadowm> (Saved games and .pbl files are the most notable exceptions.)
20141117 08:55:01< iceiceice> i think if the preprocessor were arranged to always try to fully expand macro arguments before trying to expand the macro at hand,
20141117 08:55:08< shadowm> Instead, there is a pipeline starting with the preprocessor reading a file or directory into a massive plain text document that is fed to the parser.
20141117 08:55:18< iceiceice> then whenever you have these tstrings in nested macros, you will always finally encounter them in the original context
20141117 08:55:26< shadowm> The parser then returns a monolithic WML document (config) back.
20141117 08:55:31< iceiceice> but that may be antithetical to the idea of a preprocessor
20141117 08:56:02< shadowm> The preprocessor does fully expand everything in-place.
20141117 08:56:12< shadowm> It will even traverse your filesystem if necessary.
20141117 08:57:27< shadowm> Really, the whole thing could be optimized even further, but that'd blur the distinction between the preprocessor and parser even further than it already is.
20141117 08:57:57< shadowm> People who worship the Unix Way™ would then fry you alive.
20141117 08:58:10< iceiceice> does the preprocessor give you a config or just another string?
20141117 08:58:34< shadowm> "[...] into a massive plain text documentthat is fed to the parser."
20141117 08:58:42< iceiceice> hmm
20141117 08:58:47< iceiceice> its too bad it can't hold tstrings
20141117 08:59:37< shadowm> Well, actually, I remembered things wrong, it's not a massive plain text document.
20141117 09:00:45< shadowm> Which is good. Probably.
20141117 09:01:40< shadowm> Okay, no.
20141117 09:02:14< shadowm> What happens when you toss a streambuf into a stringstream?
20141117 09:02:26< shadowm> More specifically, by using operator<<().
20141117 09:03:35< shadowm> See, I shouldn't be reading C++ right now.
20141117 09:04:03< shadowm> Okay, so the preprocessor gives you a *thing* that is a subclass of std::basic_istream<char>.
20141117 09:04:30< shadowm> You get to take full responsibility for the thing, including releasing it at a later time using operator delete.
20141117 09:05:07< shadowm> The parser reads from the thing by feeding the thing into the tokenizer and reading from it.
20141117 09:07:18< shadowm> Well, the good thing is that the parser and preprocessor run simultaneously, not in separate steps.
20141117 09:10:05< shadowm> I'm sure this will make more sense to me later when I have actually eaten.
20141117 09:11:02< vultraz> FOOOOOD
20141117 09:11:46< shadowm> In any case, this is a massive distraction.
20141117 09:12:14< shadowm> I don't care what the design alternatives are, my focus is on determining the actual cause of the bug and fixing it for 1.12.1.
20141117 09:13:14< shadowm> The whole byzantine iostreams-as-a-framework mess is part of the front-end implementation of the preprocessor, and the issue is quite clearly in the back-end.
20141117 09:14:05< shadowm> (And as I'm sure you can tell by now, I barely understand iostreams.)
20141117 09:22:07-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 09:23:38-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 09:26:33-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 244 seconds]
20141117 09:26:40-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 258 seconds]
20141117 09:41:43< irker839> wesnoth: Chris Beck wesnoth:master 9eb72f5cb535 / src/gui/dialogs/ (lua_interpreter.cpp lua_interpreter.hpp): clean up gui2 dialog "lua interpreter" http://git.io/NCfu9A
20141117 09:46:12-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has joined #wesnoth-dev
20141117 09:58:20-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141117 10:09:59-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 10:20:32-!- Anakonda [Anakonda@87-92-221-69.bb.dnainternet.fi] has joined #wesnoth-dev
20141117 10:35:45-!- travis-ci [~travis-ci@ec2-23-20-198-120.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 10:35:45< travis-ci> wesnoth/wesnoth#4808 (master - 9eb72f5 : Chris Beck): The build has errored.
20141117 10:35:45< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41225927
20141117 10:35:45-!- travis-ci [~travis-ci@ec2-23-20-198-120.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 10:54:52-!- enchilado [~enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 255 seconds]
20141117 11:26:20-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20141117 11:35:46-!- EdB [~edb@89-158-11-138.rev.numericable.fr] has quit [Quit: Konversation terminated!]
20141117 11:36:23-!- boucman_2 [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has joined #wesnoth-dev
20141117 11:37:27-!- fabi [~quassel@p20030051AA25B93574CC88744CAA1BCB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141117 11:37:27-!- fabi [~quassel@p20030051AA25B93574CC88744CAA1BCB.dip0.t-ipconnect.de] has quit [Changing host]
20141117 11:37:27-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141117 11:40:04-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 258 seconds]
20141117 11:40:50-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20141117 11:41:24-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141117 11:43:15-!- boucman_2 is now known as boucman_work
20141117 11:43:30-!- boucman_work [~jrosen@bob75-2-81-56-46-209.fbx.proxad.net] has quit [Changing host]
20141117 11:43:30-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141117 11:58:44-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 12:04:22< Soliton> so i can now create a mp game with 4 billion sides, right?
20141117 12:10:20< zookeeper> wesnoth can finally be a proper MMO \o/
20141117 12:10:44< zookeeper> just remember to have turn bell on
20141117 12:21:38-!- mjs-de [~mjs-de@f048067162.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141117 12:41:44-!- irker839 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20141117 12:50:36< vultraz> Wesnoth is the new World of Warcraft
20141117 12:55:25-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has joined #wesnoth-dev
20141117 13:00:46-!- kex [~kex@78.157.29.160] has quit [Remote host closed the connection]
20141117 13:14:11-!- mjs-de [~mjs-de@vpn40145.itmc.tu-dortmund.de] has joined #wesnoth-dev
20141117 13:35:38-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20141117 14:11:26-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141117 14:27:30-!- enchilado [~enchilado@130.102.73.156] has joined #wesnoth-dev
20141117 14:28:06-!- enchilado [~enchilado@130.102.73.156] has quit [Changing host]
20141117 14:28:06-!- enchilado [~enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20141117 14:32:22-!- cib0 [~cib@132.231.178.137] has joined #wesnoth-dev
20141117 14:41:20-!- cib0 [~cib@132.231.178.137] has quit [Ping timeout: 250 seconds]
20141117 14:46:49-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 14:49:14-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 14:54:32-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 264 seconds]
20141117 15:07:37-!- mjs-de [~mjs-de@vpn40145.itmc.tu-dortmund.de] has quit [Remote host closed the connection]
20141117 15:09:14-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20141117 15:14:41-!- cib0 [~cib@132.231.178.79] has joined #wesnoth-dev
20141117 15:28:25-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141117 15:34:21-!- gfgtdf [~chatzilla@d107142.adsl.hansenet.de] has joined #wesnoth-dev
20141117 15:53:49-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 15:58:47-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 16:00:32-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 16:01:27-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141117 16:01:30-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20141117 16:08:42-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has quit [Remote host closed the connection]
20141117 16:23:50-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 16:28:24-!- cib0 [~cib@132.231.178.79] has quit [Ping timeout: 265 seconds]
20141117 16:34:26-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Read error: Connection reset by peer]
20141117 16:34:56-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141117 16:38:17-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 16:44:36-!- mjs-de [~mjs-de@vpn40145.itmc.tu-dortmund.de] has joined #wesnoth-dev
20141117 17:03:12-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Ping timeout: 265 seconds]
20141117 17:27:37-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141117 17:53:35-!- irker816 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141117 17:53:35< irker816> wesnoth: gfgtdf wesnoth:master 24359765758e / src/replay.cpp: don't create blank attributes in replay.cpp http://git.io/klhWDA
20141117 17:53:35< irker816> wesnoth: gfgtdf wesnoth:master 7651676c3c0c / src/server/game.cpp: add comment http://git.io/2g_ykw
20141117 17:53:35< irker816> wesnoth: gfgtdf wesnoth:master 2944051f8304 / src/server/game.cpp: remove old server compability code http://git.io/aWFPTQ
20141117 17:53:36< irker816> wesnoth: gfgtdf wesnoth:master db31880012b9 / src/server/ (game.cpp game.hpp): only allow "dependent" commands for players http://git.io/PfL1YA
20141117 17:53:38< irker816> wesnoth: gfgtdf wesnoth:master 2ea7c1e09da9 / src/server/game.cpp: refactor wesnothd::game::is_legal_command http://git.io/2AZT0A
20141117 17:53:40< irker816> wesnoth: gfgtdf wesnoth:master b3d74c482fca / src/server/game.cpp: check "from_side" in wesnothd::game::is_legal_command http://git.io/nH_7ng
20141117 17:53:42< irker816> wesnoth: gfgtdf wesnoth:master 80ca2ec06a11 / src/server/game.cpp: fixup a comment http://git.io/etRaNw
20141117 17:55:31-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20141117 17:56:26-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 18:09:12-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has joined #wesnoth-dev
20141117 18:10:42-!- DCW [~Thunderbi@cpc66866-finc15-2-0-cust47.4-2.cable.virginm.net] has quit [Client Quit]
20141117 18:12:13-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141117 18:13:55-!- Sulfur [~Miranda@p5B32782F.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141117 18:16:01-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 18:17:45-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 18:42:16-!- gfgtdf [~chatzilla@d107142.adsl.hansenet.de] has quit [Quit: ChatZilla 0.9.91 [Firefox 33.1/20141106120505]]
20141117 18:51:49-!- mjs-de [~mjs-de@vpn40145.itmc.tu-dortmund.de] has quit [Remote host closed the connection]
20141117 18:55:13-!- travis-ci [~travis-ci@ec2-54-162-92-239.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 18:55:13< travis-ci> wesnoth/wesnoth#4809 (master - 80ca2ec : gfgtdf): The build has errored.
20141117 18:55:13< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41271756
20141117 18:55:13-!- travis-ci [~travis-ci@ec2-54-162-92-239.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 18:58:28< irker816> wesnoth: Chris Beck wesnoth:master f621233cc2e4 / src/gui/dialogs/lua_interpreter.cpp: tlua_interpreter: add some asserts for null pointers http://git.io/w5UTqw
20141117 19:09:14< irker816> wesnoth: Chris Beck wesnoth:master 9f69a5196ed0 / src/tests/gui/test_gui2.cpp: add gui2 unit tests for advanced_graphics_options, mp_alerts dialogs http://git.io/aldGQQ
20141117 19:20:10-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20141117 19:34:27-!- Dugi [93fbd29f@gateway/web/freenode/ip.147.251.210.159] has joined #wesnoth-dev
20141117 19:36:07< irker816> wesnoth: Chris Beck wesnoth:master 968d8a900dc2 / src/sdl/utils.cpp: add an assert to detect possible division by zero http://git.io/NlYBeg
20141117 19:36:07< Dugi> Hi.
20141117 19:39:44-!- cib0 [~cib@p5DD21EDB.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141117 19:48:32-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 19:50:09-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141117 19:56:05-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 19:56:05< travis-ci> wesnoth/wesnoth#4810 (master - f621233 : Chris Beck): The build failed.
20141117 19:56:05< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41278491
20141117 19:56:05-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 20:03:37-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20141117 20:12:50-!- Dugi [93fbd29f@gateway/web/freenode/ip.147.251.210.159] has quit [Quit: Page closed]
20141117 20:18:43-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 255 seconds]
20141117 20:26:37-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 20:28:59-!- travis-ci [~travis-ci@ec2-54-162-92-239.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 20:28:59< travis-ci> wesnoth/wesnoth#4811 (master - 9f69a51 : Chris Beck): The build has errored.
20141117 20:28:59< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41279576
20141117 20:28:59-!- travis-ci [~travis-ci@ec2-54-162-92-239.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 20:32:40-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141117 20:43:02-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 20:50:44-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 20:50:44< travis-ci> wesnoth/wesnoth#4812 (master - 968d8a9 : Chris Beck): The build has errored.
20141117 20:50:44< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41282439
20141117 20:50:44-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 20:51:00-!- fkhodkov [~user@2a01:d0:ffff:10ea::2] has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
20141117 20:52:38-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 21:13:48-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 21:18:13-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20141117 21:19:05-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 264 seconds]
20141117 21:20:38-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 21:31:35-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Bye for now]
20141117 21:32:34< Ivanovic> noy: please check for the mail from carols and act about it
20141117 21:32:35< Ivanovic> thanks!
20141117 21:44:27-!- gfgtdf [~chatzilla@d107142.adsl.hansenet.de] has joined #wesnoth-dev
20141117 21:44:41-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 21:44:41< travis-ci> wesnoth/wesnoth#4812 (master - 968d8a9 : Chris Beck): The build passed.
20141117 21:44:41< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41282439
20141117 21:44:41-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 21:48:58-!- Ard0nik [~user@adsl-75-28-101-131.dsl.irvnca.sbcglobal.net] has quit [Read error: Connection reset by peer]
20141117 21:50:27-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20141117 22:01:44-!- Sulfur [~Miranda@p5B32782F.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141117 22:02:51-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has joined #wesnoth-dev
20141117 22:10:36-!- shikadibot_ is now known as shikadibot
20141117 22:11:50-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141117 22:11:58< iceiceice> shadowm: the path in that guy's error message "C:/Program Files (x86)/Battle for Wesnoth 1.12.0\userdata/data/add-ons\After_the_Storm/_main.cfg"
20141117 22:12:04< iceiceice> is that normal on windows?
20141117 22:12:12< shadowm> Yes.
20141117 22:12:29< iceiceice> that is truly bizarre
20141117 22:12:34< shadowm> No, it isn't.
20141117 22:12:51< iceiceice> i've never known any program to format paths with mixed separators like that
20141117 22:13:19< shadowm> Don't worry about it.
20141117 22:13:25< iceiceice> ok, i won't :)
20141117 22:15:48< irker816> wesnoth: Chris Beck wesnoth:master caeaa6ba67ed / src/ (17 files in 4 dirs): refactor lua show_dialog http://git.io/cBS_Fg
20141117 22:16:16< shadowm> loonycyborg: Is there any way I can get the 1.12.0 binary + libraries for Windows without downloading the whole massive thing including data files, btw?
20141117 22:17:19< loonycyborg> shadowm: hmm I don't think so
20141117 22:17:30< iceiceice> shadowm: is there any particular reason that "torrents are unofficial" ?
20141117 22:17:33< loonycyborg> our files.wesnoth.org thing was on old server
20141117 22:17:48< iceiceice> i think if you were using a torrent you can indeed select a subset of the files
20141117 22:17:56< shadowm> loonycyborg: What is that even supposed to mean? http://files.wesnoth.org/ is also on the new server.
20141117 22:17:59< loonycyborg> if it's archine then no
20141117 22:18:14< loonycyborg> I mean torrents
20141117 22:18:27< loonycyborg> ahh it was moved
20141117 22:18:29< shadowm> Huh?
20141117 22:18:31-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141117 22:18:42< loonycyborg> but don't remember how to upload files there anymore
20141117 22:18:50< loonycyborg> and it would require me to do it manually
20141117 22:19:24< shadowm> Too bad you weren't paying attention at the time I did the migration.
20141117 22:19:29< loonycyborg> there's a way to get trunk binaries
20141117 22:19:32< loonycyborg> from jenkins
20141117 22:19:38< loonycyborg> and we have 1.12 branch there
20141117 22:19:40< shadowm> Yes, but I want the real deal.
20141117 22:20:23< shadowm> loonycyborg: Even if you remembered how you did it, that method wouldn't work anymore because we have a completely different configuration for everything.
20141117 22:20:34< shadowm> Plus I didn't migrate your account for reasons I don't remember.
20141117 22:20:53< shadowm> If you want access to files.wesnoth.org you'll have to send me your SSH public key again.
20141117 22:21:18< loonycyborg> kinda busy atm
20141117 22:21:48< loonycyborg> sec
20141117 22:23:54< loonycyborg> iceiceice: I think the only reason why torrents aren't official is that we aren't doing them :P
20141117 22:24:25< iceiceice> that's a pretty good reason :)
20141117 22:25:28< loonycyborg> how would we be doing official ones anyway?
20141117 22:25:37< loonycyborg> upload a torrent to tpb? :P
20141117 22:29:23< iceiceice> i think we just make a tracker or whatever
20141117 22:29:32< iceiceice> and then someone has to point the client at it
20141117 22:29:34< iceiceice> and just leave it open
20141117 22:29:43< iceiceice> and we put a link to the tracker on the website
20141117 22:30:17< iceiceice> *tracker being like, a text file
20141117 22:30:31< iceiceice> i might be confused how this works though, i don't remember if i ever did it or not
20141117 22:30:59-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141117 22:32:39< shadowm> And then what? People will download from us if nobody else is seeding at a given time?
20141117 22:32:58< iceiceice> there has to be at least one seed active,
20141117 22:33:04< iceiceice> or when you connect the tracker it will say "cannot find any seeds"
20141117 22:33:24< shadowm> So it's not our responsibility to seed I take it?
20141117 22:33:34< iceiceice> i mean someone has to get the ball rolling
20141117 22:34:05< iceiceice> actually i think its even okay if not everyone has it,
20141117 22:34:15< iceiceice> as long as every bit of the file is known by at least someone who is seeding
20141117 22:35:44-!- gfgtdf_ [~chatzilla@f050183003.adsl.alicedsl.de] has joined #wesnoth-dev
20141117 22:37:39< iceiceice> loonycyborg: maybe you are right actually, maybe its better to to use some existing public tracker service (I don't think it needs to be tpb though :) )
20141117 22:38:31< iceiceice> but i think even if you do that you can just put a link to the .torrent file on the wiki, and people's clients will connect to it automatically
20141117 22:38:57-!- gfgtdf [~chatzilla@d107142.adsl.hansenet.de] has quit [Ping timeout: 264 seconds]
20141117 22:39:08-!- gfgtdf_ is now known as gfgtdf
20141117 22:49:02-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 245 seconds]
20141117 22:51:42-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 22:52:33-!- mjs-de [~mjs-de@wh.Uni-Dortmund.DE] has quit [Remote host closed the connection]
20141117 22:59:17-!- cib0 [~cib@p5DD21EDB.dip0.t-ipconnect.de] has quit [Ping timeout: 264 seconds]
20141117 23:01:09-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20141117 23:04:56-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has joined #wesnoth-dev
20141117 23:04:56< travis-ci> wesnoth/wesnoth#4813 (master - caeaa6b : Chris Beck): The build has errored.
20141117 23:04:56< travis-ci> Build details : http://travis-ci.org/wesnoth/wesnoth/builds/41299164
20141117 23:04:56-!- travis-ci [~travis-ci@ec2-54-89-55-158.compute-1.amazonaws.com] has left #wesnoth-dev []
20141117 23:05:18-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 255 seconds]
20141117 23:14:15-!- kex [~kex@78.157.29.160] has quit [Remote host closed the connection]
20141117 23:14:31-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 23:15:22< gfgtdf> iceiceice: im thinking about whether it woudl make sense to seperate the gamestate part from that mpsnc part in [change_controller] more, E.g instead of having [change_controller] having 2 different tags [change_controller] which does not effect wml at all and a another tag [rename_side] which then sended by the server too.
20141117 23:16:06< iceiceice> sure that seems pretty reasonable
20141117 23:16:39< iceiceice> i actually don't realy like that side_drop etc. has all these special name side effects enacted by some clients and not others
20141117 23:22:11-!- kex [~kex@78.157.29.160] has quit []
20141117 23:23:20< gfgtdf> iceiceice: also do you know why we dont use response.clone() in here: https://github.com/wesnoth/wesnoth/blob/master/src/server/game.cpp#L561 ?
20141117 23:23:50< iceiceice> did i write that line?
20141117 23:24:00< gfgtdf> iceiceice: idk
20141117 23:24:03< iceiceice> if so it's probably because i didn't know about .clone()
20141117 23:24:43< iceiceice> yeah i did write that line :)
20141117 23:25:14< iceiceice> gfgtdf: do you have access to the coverity scans of wesnoth?
20141117 23:25:28< gfgtdf> iceiceice: no, ot i lest im nor aare of
20141117 23:25:31< gfgtdf> or*
20141117 23:25:34< gfgtdf> aware*
20141117 23:25:40< iceiceice> ok, i looked at it again today
20141117 23:25:58< iceiceice> i think there were some new server issues it found but i have to look again to remember
20141117 23:28:32< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/server/game.cpp#L503
20141117 23:28:36-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141117 23:28:46< iceiceice> coverity thinks that on some code path, oldplayer can be used when it is past the end of its iterator
20141117 23:28:49< iceiceice> *its container
20141117 23:30:01< iceiceice> also it points out that we test for this possibility earlier: https://github.com/wesnoth/wesnoth/blob/master/src/server/game.cpp#L451
20141117 23:30:09< iceiceice> but we don't return or anything in that case
20141117 23:31:09< gfgtdf> iceiceice: hm but i dodnt change game::transfer_side_control at all
20141117 23:32:02< iceiceice> yeah i think this is not as new as your changes
20141117 23:32:08< iceiceice> its just new since the last time i looked at this
20141117 23:32:16< gfgtdf> iceiceice: hm ok, maybe we shoudl add a return after teh missing_user in the if
20141117 23:32:37-!- enchilado [~enchilado@defocus/yummy/enchilado] has quit [Ping timeout: 240 seconds]
20141117 23:35:06< iceiceice> yeah maybe
20141117 23:35:44< iceiceice> gfgtdf: another one is that it doesn't like that the server never catches utf8::invalid_utf8_exception
20141117 23:35:58< iceiceice> for instance it think that if it is thrown here, it will crash the server:
20141117 23:35:58< iceiceice> https://github.com/wesnoth/wesnoth/blob/master/src/server/server.cpp#L1380
20141117 23:35:59< gfgtdf> iceiceice: really=
20141117 23:36:08< gfgtdf> iceiceice: iditn thought teh server woudl try pasring utf8
20141117 23:36:45< gfgtdf> iceiceice: y that looks bad indeed
20141117 23:37:53< shadowm> Hi don't mind me I'm just passing by.
20141117 23:39:02< iceiceice> shadowm: yeah, probably we should patch this immediately
20141117 23:39:29< shadowm> Commits welcome.
20141117 23:39:53< iceiceice> actually i didnt look at 1.12 though
20141117 23:40:12< shadowm> I'd expect that to be the case in 1.12 too from the big unicode refactoring (sans namespace changes).
20141117 23:40:38< shadowm> Not in 1.10 though, which IIRC used libc tolower to implement that function.
20141117 23:41:30-!- ancestral [~ancestral@63.92.240.233] has joined #wesnoth-dev
20141117 23:41:51< shadowm> Okay no, that last part is false.
20141117 23:42:30< shadowm> Yeah, the cat's out of the bag now, this needs to be patched for 1.10, 1.12, and 1.13.
20141117 23:42:40< iceiceice> hmm... does it really make sense for utf8::lowercase to throw an exception anyways?
20141117 23:42:44< iceiceice> why can't it just return the original string.
20141117 23:43:07< shadowm> (By all means, keep telling people in a public channel how to crash the server.)
20141117 23:43:18< iceiceice> T_T sorry
20141117 23:43:34-!- enchilado [~enchilado@130.102.75.254] has joined #wesnoth-dev
20141117 23:44:36< irker816> wesnoth: Chris Beck wesnoth:master 67eb086736e7 / src/generators/lua_map_generator.cpp: fixup caeaa6ba67ed0e05220e189c12df768479320e01 http://git.io/R_wuUg
20141117 23:46:01< gfgtdf> iceiceice: with the new changes to lua_kernel_bse one can not use wesnoth.require/dofile in the lua map generator ?
20141117 23:46:02-!- mjs-de [~mjs-de@f048067162.adsl.alicedsl.de] has joined #wesnoth-dev
20141117 23:46:12< gfgtdf> now*
20141117 23:46:16< iceiceice> i think that shold be possible
20141117 23:46:50< iceiceice> i didnt test but if its broken it means i made a bug somehow
20141117 23:47:10< iceiceice> actually if it is broken it means probably its broken everywhere, so the passage of many unit tests would be surprising
20141117 23:47:32< iceiceice> i tried to move over the gui2 show dialog stuff
20141117 23:47:55-!- enchilado [~enchilado@130.102.75.254] has quit [Changing host]
20141117 23:47:55-!- enchilado [~enchilado@defocus/yummy/enchilado] has joined #wesnoth-dev
20141117 23:50:41< irker816> wesnoth: Chris Beck wesnoth:master 1a0d9a09851c / src/server/player.cpp: fix an uncaught bad lexical cast exception http://git.io/rRWJBA
20141117 23:51:13< iceiceice> hmm actually that commit message is inaccurate.. oh well
20141117 23:52:26< gfgtdf> iceiceice: dont you need to pasa  default value to lexial_cat_default ?
20141117 23:52:31< gfgtdf> pass*
20141117 23:52:38< iceiceice> gfgtdf: i think it will just default construct if i dont
20141117 23:52:48< gfgtdf> iceiceice: hm ok
20141117 23:53:01-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141117 23:53:13< shadowm> template<typename To, typename From> To lexical_cast_default(From a, To def=To()) { ... }
20141117 23:53:19< shadowm> Yes.
20141117 23:54:21< shadowm> Wow, my Internet sucks today.
20141117 23:55:19< shadowm> I had asked AI0867 about uploading a newer build to coverity  but it seems like that hasn't happened yet.
20141117 23:55:59< irker816> wesnoth: Chris Beck wesnoth:master f6bdc94ccda4 / src/server/ (game.cpp room_manager.cpp server.cpp): replace many lexical_cast with lexical_cast_default http://git.io/oUa6bg
20141117 23:57:13< gfgtdf> iceiceice: i noticed a prblem with the "split [change_controller]" approach: actually [change_controller] does three things: change teh controlling client for that side, change teh controller tpye (ai/human) and renames the side from the mp sync point of wiew you'd put 'rename' and 'controller type' together, but from playcontrollers logiy´c point of view dou'd put 'controlling  client' and...
20141117 23:57:14< gfgtdf> ...'contorller type' together
20141117 23:58:49< iceiceice> hmm so, i guess should decide if you want the names to be synced or not
20141117 23:59:04< iceiceice> since player can also just rename them in the game i think, and i doubt if that is synced
--- Log closed Tue Nov 18 00:00:05 2014