--- Log opened Sat Mar 22 00:00:09 2014
20140322 00:00:19< zookeeper> yes, i said nothing about not avoiding it
20140322 00:02:12< zookeeper> anyways, i gotta get some sleep
20140322 00:02:12< aquileia> possible syntax for maximum compatibility: http://pastebin.com/bsxVqgK8 --> http://pastebin.com/cZbH74zv
20140322 00:03:13< aquileia> oh, the grunt-attack- will obviously be only grunt-
20140322 00:03:43< Coffee_irc> aquileia/Aishiko: the spitesheets can be a good reason to fix another minor problem in the WML
20140322 00:03:44< aquileia> if no number is given (grunt.png) it would default to 1
20140322 00:04:00< Aishiko> Coffee_irc, what is that?
20140322 00:04:07< Coffee_irc> the "unit/type/name.png" doesn't really need be specified in each type of animation
20140322 00:04:19< Coffee_irc> not if they all refer back to one sheet
20140322 00:04:46< Coffee_irc> you could have a [spritesheet] with a "name" that you can refer back to instead or something
20140322 00:04:57< Aishiko> so basically a path could be units/orc-grunt.png?
20140322 00:05:11< Coffee_irc> Aishiko: that's what I am thinking
20140322 00:05:20< Aishiko> thats not a bad idea
20140322 00:05:29< Coffee_irc> it is unneccessary duplication at the moment
20140322 00:05:31-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 246 seconds]
20140322 00:05:32< gfgtdf> can a side that lost his leader in a 3p scenario still move with his remaining units (asuming the other 2 sides still have their units)?
20140322 00:05:42< aquileia> hmm... nice idea, but I think that it should at least be grunt-n for the frame
20140322 00:06:03< aquileia> Else you can't easily grab halos from other animations
20140322 00:06:07< Coffee_irc> if you have several spritesheets you need to merge, then you could give each one a "name" or "id"
20140322 00:06:15< iceiceice> gfgtdf: yes
20140322 00:06:15< Coffee_irc> just something to consider
20140322 00:06:18-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds]
20140322 00:06:26< iceiceice> at least i'm pretty sure
20140322 00:06:35< iceiceice> actually i'm definitely sure
20140322 00:06:45< aquileia> and the name would default to the name of the .png? That sounds great!
20140322 00:06:45< gfgtdf> iceiceice: ok ty, i already thought that was a bug
20140322 00:07:01< iceiceice> idk if it's "intended" but that's currently how FFA's work in 1.10
20140322 00:07:16< gfgtdf> 'FFA' ?
20140322 00:07:20< iceiceice> free for all
20140322 00:07:58-!- RiftWalker [~nathan@c-76-22-135-46.hsd1.tn.comcast.net] has joined #wesnoth-dev
20140322 00:10:40< aquileia> Coffee_irc: Like this? http://pastebin.com/kF4AS1hW 
20140322 00:12:18< Coffee_irc> aquileia: I figure it is really for Aishiko to spend time considering all the possibilities and implications of the syntax if it is accepted as a GSoC project
20140322 00:12:39< Coffee_irc> but you could use angle brackets or other mechanisms too
20140322 00:13:16< Coffee_irc> those 3 individual frame blocks could be merged into one with a spritesheet
20140322 00:13:40< Coffee_irc> most units would only need one [frame] block
20140322 00:13:45< Coffee_irc> for each type of animation
20140322 00:13:59< Aishiko> like sub components of the spritesheet block
20140322 00:17:41< Coffee_irc> Aishiko: whatever syntax you pick I suggest shuld be easy to mix with the image path functions we currently have (http://wiki.wesnoth.org/ImagePathFunctionWML)
20140322 00:19:15< Aishiko> Coffee_irc, I'm using the current syntax as a starting point, as it must remain usable until such time as everything (if it happens) is converted to a sprite sheet
20140322 00:19:41< Coffee_irc> Aishiko/aquileia: http://pastebin.com/vB7jKkC6
20140322 00:19:55< Coffee_irc> this is another potential way to do the same thing
20140322 00:20:34< Coffee_irc> *ignore the death animation
20140322 00:21:57< Aishiko> I like the name being in <>
20140322 00:22:09< Coffee_irc> Aishiko: I got the idea from the terrain WML
20140322 00:22:19< Coffee_irc> which is worth studying
20140322 00:22:59< aquileia> Ivanovic, shadowm: I'd like to provide the build dependencies for MSVC2013 at the same place as e.g. http://sourceforge.net/projects/wesnoth/files/unofficial/Windows%20Compile%20Stuff/externalVC9.zip/download
20140322 00:23:21< aquileia> with VC9 substituted by VC12 of course
20140322 00:23:25< Coffee_irc> Aishiko: if implemented right you might be able to use it for animated terrain also
20140322 00:23:38< happygrue> So I have some scenario tweaks to THoT I'd like to commit (I had been waiting to see if I could get through the last scenario, but I just don't have the jucie to play it for debugging)
20140322 00:23:43< Coffee_irc> but using angle brackets like that might clash, I don't know
20140322 00:23:53< happygrue> but I'm a bit scared with the branching and I don't want to mess stuff up with a commit
20140322 00:24:13< aquileia> 1. Would you trust me and upload the file there or should I provide a link to something else?
20140322 00:24:17-!- neXyon [~neXyon@85-127-227-111.dynamic.xdsl-line.inode.at] has quit [Quit: bye]
20140322 00:24:37< aquileia> 2. Do you wnat the file via github, dropbox, ... ?
20140322 00:24:41< aquileia> *want
20140322 00:25:10< happygrue> I am using GitHub at the moment, and I have up to now been keeping the rebase option checked, but that's the part that scares me now
20140322 00:25:22< _8680_> happygrue, are you asking for help with Git?
20140322 00:25:48< happygrue> yes
20140322 00:26:05< happygrue> maybe I should just do a pull request?
20140322 00:26:20< _8680_> What do you want to do?
20140322 00:26:57< _8680_> You want to commit your changes on a topic branch?
20140322 00:27:16< happygrue> I don't really know what I want to do, that is the problem.  :D
20140322 00:27:42< happygrue> I have some changes, I would like to apply them to both branches (scenario tweaks are bug fixes?  Or no?)
20140322 00:27:51< _8680_> Which branches?
20140322 00:27:54< happygrue> but I also don't know the best way to do that
20140322 00:28:05< aquileia> I guess 1.12 and master
20140322 00:28:06< happygrue> er, 1.12 and 1.13
20140322 00:28:11< happygrue> yes
20140322 00:28:34-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140322 00:29:11< happygrue> but so far I have been using the gui client with the "use rebase" option.  That is the part that I fear might be trouble now
20140322 00:29:14< _8680_> I believe the standard procedure is to commit to `master` and cherry-pick to `1.12`.
20140322 00:29:20< happygrue> I see
20140322 00:33:08< Aishiko> Coffee_irc, well I was hoping that whatever gets used to implement the unit spritesheets would be compatible with other parts of the code like terrain and attacks
20140322 00:33:08< iceiceice> happygrue: its usually best to branch before making changes, or it can be a little harder to recover if you make a mistake later
20140322 00:33:19< iceiceice> if you've already changed files and are now just trying to think about committing them,
20140322 00:33:34< iceiceice> you can "checkout -b THoT_Tweaks" and then commit
20140322 00:34:30< happygrue> oaky, that does sound like what I want to do
20140322 00:38:16< Coffee_irc> Aishiko: we've still got a lot of very old animation syntax ('begin' 'end' tags)
20140322 00:38:29-!- Kevin_Xi [~kevin@223.72.182.158] has joined #wesnoth-dev
20140322 00:39:06< Coffee_irc> just for your information I'm planning on putting a warning with 1.13 that those very old tags will be removed for 1.13+1
20140322 00:39:32< Coffee_irc> not that it will impact you much for spritesheets
20140322 00:40:33< Aishiko> Coffee_irc, let me guess they were kept around for backwards compatibility until now
20140322 00:40:50< Coffee_irc> Aishiko: yes, and they have been depreciated on the wiki for some time
20140322 00:41:42< Coffee_irc> still a lot of UMCs use them and I think they will be happiest to move over to a complete spritesheet system from the very old syntax
20140322 00:43:55< Aishiko> Basically update once and not have to do it again next year, right
20140322 00:44:20< Coffee_irc> Aishiko: I would hope that if a spritesheet system is implemented it would be done once and right
20140322 00:45:10< Coffee_irc> this would solve the WML complexity problem (previously each frame required a separate [frame] tag) and the image loading issues
20140322 00:45:13< Aishiko> Coffee_irc, what I mean is not update to what was used in 1.12 and then next year have a new system come out that and they need to update yet again
20140322 00:45:47-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has quit [Quit: leaving]
20140322 00:46:43< Coffee_irc> Aishiko: well, if you extend the existing syntax [] (which I am biased about because I wrote most of it) then it might be easy to port to spritesheets
20140322 00:46:57-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20140322 00:49:01< Coffee_irc> the spritesheets would solve the image loading issues, take less memory, and possibly remove unneccessary "unit/type/..."s in the code and combine the [frames] to make them silmpler
20140322 00:49:42< Coffee_irc> or so I believe are the reasons for making them work
20140322 00:51:07< Coffee_irc> I suppose the artists might like them for working on as well
20140322 00:53:46< Coffee_irc> personally I can't really see anything after spritesheets to replace them
20140322 00:53:50< Aishiko> that is something I have to keep in mind, performance
20140322 00:55:01< Coffee_irc> Aishiko: you should get better performance from spritesheets I would think just by making them work
20140322 00:55:09< Coffee_irc> not really doing anything special
20140322 00:56:21< Coffee_irc> but mordante and others might know more about which libraries to use to facilitate this best
20140322 01:00:25< Aishiko> I'll ask him when I see him this weekend
20140322 01:01:52< RiftWalker> http://wiki.wesnoth.org/SoC2014_RiftWalker_Unify_SP_MP
20140322 01:01:55< RiftWalker> thoughts?
20140322 01:08:57< irker031> wesnoth: happygrue wesnoth:master 3265ad3b3f3e / / (5 files in 2 dirs): Rebalanced three THoT scenarios. http://git.io/4WZRjw
20140322 01:10:06< happygrue> iceiceice: so that is fine right?  ^
20140322 01:11:33< iceiceice> it looks like whitespace might be a bit funny in changelog? not sure. otherwise i dont see any issue
20140322 01:12:03< happygrue> in the players changelog you mean?
20140322 01:12:04< iceiceice> like 8680 said, when you are done you would change to 1.12 and cherry-pick this commit
20140322 01:12:08< iceiceice> no in the changelog file
20140322 01:12:35< iceiceice> should "Rebalanced scenarios 'Invaders', Mages and Drakes' and 'Fear'." be at the same depth as "fixed missing objective"?
20140322 01:13:01< iceiceice> idk, at a glance that's my guess
20140322 01:13:31< happygrue> yep, you're right
20140322 01:14:08-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 01:15:24-!- RiftWalker [~nathan@c-76-22-135-46.hsd1.tn.comcast.net] has quit [Quit: Leaving.]
20140322 01:18:18-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140322 01:23:34< happygrue> Hmmm, notepad++ is failing me, as it is showing perfectly there.
20140322 01:24:35< mattsc> happygrue: spaces vs. tabs?
20140322 01:25:04< vultraz> N++ is funky with spaces vs tabs
20140322 01:25:12< vultraz> IIRC a tab is default 4 width
20140322 01:25:18< happygrue> I've tried deleting it all and spacing out but it doesn't seem to matter, but I guess I should try again
20140322 01:29:00< happygrue> there we go
20140322 01:29:05< aquileia> happygrue: There's a plugin TabIndentSpaceAlign that helps a little, but unfortunately you have to activate it every time you open N++
20140322 01:29:43< irker031> wesnoth: happygrue wesnoth:master cd1a8ea6c803 / changelog players_changelog: Fixed whitespace error in changelogs http://git.io/uacROQ
20140322 01:30:01< happygrue> I think it was autocompleting the spacing for me when I hit enter and I trusted it
20140322 01:30:52 * happygrue makes a note to check for whitespace while previewing the commit
20140322 01:31:14< happygrue> so if I were to cherry pick that now, it is two commits?
20140322 01:31:26< happygrue> first one, then the other?
20140322 01:32:12< mattsc> ooo, ooo, I should get my changelog entry in first, so that happygrue will encounter a conflict  ;)
20140322 01:32:37< vultraz> mattroll
20140322 01:32:56 * happygrue happygrue may just say "screw it" to the fine users of the upcoming stable version
20140322 01:33:45 * mattsc remembers that he still might want happygrue to test Fred at some point in the (distant) future
20140322 01:34:02< happygrue> I want to go through SotBE too
20140322 01:34:12< mattsc> yay
20140322 01:35:04< happygrue> but seriously, I think trying to cherry pick stuff to 1.12 now is beyond what my brain can handle tonight.  I'm still enough of a noob with git (only git, I'm great at everything else...) that I don't want to mess anything up
20140322 01:35:42< happygrue> if some whitespace can defeate me for 15 minutes then I better not do anything dangerous ;)
20140322 01:35:44< mattsc> happygrue: if it's small commits, I often simply make them to both repositories
20140322 01:36:11< happygrue> I've only been using the gui client, and so I have a hazy idea about under the hood stuff
20140322 01:36:51< mattsc> I explain that I have two local copies of the repository, one of which is set to master, the other to 1.12
20140322 01:36:58< mattsc> i *should explain
20140322 01:37:13< happygrue> yeah, I don't and looking at it made me think maybe I should 
20140322 01:37:34-!- iceiceice [~chris@207-237-132-90.ny.subnet.cable.rcn.com] has quit [Remote host closed the connection]
20140322 01:39:29< mattsc> happygrue: you don't actually need to check it out twice, you just make a local copy of it, and then you tell the github app that you have another local repository.
20140322 01:39:47-!- sachith500 [~kvirc@112.134.146.132] has joined #wesnoth-dev
20140322 01:40:01< happygrue> ah right, that would be easy
20140322 01:40:30< _8680_> You’re copying the whole thing, all ~1.8 GB of it, not using new-workdir?
20140322 01:41:15< mattsc> _8680_: yes - I don't know anything about new-workdir and disk space is cheap these days :P
20140322 01:41:50< mattsc> _8680_: if you read this channel, you will have noticed that I never claim that I actually know what I am doing.  :)
20140322 01:42:23-!- ejls [~ejls@etiennesimon.eu] has joined #wesnoth-dev
20140322 01:42:40< _8680_> happygrue: Cherry-picking shouldn’t be more complex than `git checkout 1.12 && git cherry-pick master~ master` (for the second-to-last and last commits on `master`).
20140322 01:42:58< mattsc> ... which doesn't mean that I am not willing to learn, just that I usually go by TBITEOTGE
20140322 01:43:11< _8680_> I don’t know how much more complicated your Git GUI makes it, though.
20140322 01:43:36< _8680_> ‘TBITEOTGE’?
20140322 01:43:46< mattsc> the better is the enemy of the good enough
20140322 01:44:59< _8680_> Ah.
20140322 01:45:35< _8680_> Well then, I guess that makes ‘good enough’ the enemy of ‘better’.
20140322 01:46:06< mattsc> _8680_: no, but seriously, as you know, I really don't know much about coding, how it should be done and the like, so once I find something that works, I usually just stick with it as long as it works.  I'm always happy to be told that I'm an idiot and that there are better ways to do it though.
20140322 01:46:20< _8680_> …I meant no offense, to be clear.
20140322 01:46:49< _8680_> (Not that you seem offended.)
20140322 01:46:59< mattsc> _8680_: I figured (hoped) that that my last line made it clear that no offense was taken.  :)
20140322 01:47:02< mattsc> exactly
20140322 01:47:55< Aishiko> mattsc, I'm similar though I would hope it would be something more like "why are you using/doing X when Y would be easier?"
20140322 01:48:16< Aishiko> which lets me respond "Y? what is that?" and hopefully learn something
20140322 01:49:33-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140322 01:50:02-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140322 01:50:17-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140322 01:50:35< mattsc> _8680_: As I have said before, I really have no coding background and am just making things up as I go along (you should hear the comments of the people who have to work with the robotic control system I wrote for out site testing telescopes).
20140322 01:51:24< mattsc> But the thing is, the code worked for 5 years on remote sites which had the closest person who could take care of it 1000 miles away.
20140322 01:51:52< Aishiko> it worked even if it wasn't pretty/elegant I bet
20140322 01:52:00< mattsc> So I know how to make things work, but the method by which I do so is ... well, "quick and dirty" is a compliment.  :)
20140322 01:52:01< _8680_> That sounds impressive. What was it written in, Erlang?
20140322 01:52:29< mattsc> Hehe.  No, a mix of C, IDL and CL scripts.
20140322 01:53:00-!- sachith500 [~kvirc@112.134.146.132] has quit [Read error: Connection reset by peer]
20140322 01:53:06-!- ejls [~ejls@etiennesimon.eu] has quit [Quit: ejls]
20140322 01:53:18-!- sachith500 [~kvirc@112.134.146.132] has joined #wesnoth-dev
20140322 01:53:52< mattsc> Anyways, I'll have to be afk again.  Not that I am saying anything important anyway.  TTYL.
20140322 01:53:55< _8680_> There are two ‘better’ ways of copying a local repository to have multiple branches checked-out simultaneously: (1) cloning the repository locally, which is more efficient (in terms of speed of copying and space used) but can be inconvenient, and (2) `git-new-workdir`, which allows having multiple working directories backed by the same repository.
20140322 01:53:59-!- ejls [~ejls@etiennesimon.eu] has joined #wesnoth-dev
20140322 01:54:48< mattsc> _8680_: (1) is what I did - I think (will check on this later, really need to be off now).
20140322 01:55:04< _8680_> (If one clones into `foodir`, the repository is in `foodir/.git` and the working tree(/directory) is everything else in `foodir`.)
20140322 01:55:20 * _8680_ leaves too.
20140322 02:03:35-!- sachith500 [~kvirc@112.134.146.132] has quit [Read error: Connection reset by peer]
20140322 02:04:08-!- sachith500 [~kvirc@112.134.146.132] has joined #wesnoth-dev
20140322 02:09:15< gfgtdf> mattsc: online ?
20140322 02:09:30< aquileia> anonymissimus: Feel free to discard some of my changes to http://wiki.wesnoth.org/CompilingWesnothOnWindows
20140322 02:11:54< Aishiko> gfgtdf, he just went afk
20140322 02:11:56< mattsc> gfgtdf: not really - I'm here, but don't have time to sit at the computer right now
20140322 02:12:19< mattsc> Aishiko: yeah (thanks).  And sorry, didn't mean to ignore you earlier, but I really need to run off.
20140322 02:12:32< aquileia> I want to add my excludeVC12.zip for those on VC2013, but I have no access to the sourceforge repo to add it
20140322 02:12:43< Aishiko> mattsc, its OK I'm used to not getting responses =) life happens
20140322 02:12:44< gfgtdf> mattsc: ok i just remembered that i asked you the question i wanted to ask you already :)
20140322 02:12:54< gfgtdf> do we have another AI expert ?
20140322 02:14:04< gfgtdf> wesbot: seen crab_irc
20140322 02:14:04< wesbot> gfgtdf: Sorry, I don't know of crab_irc.
20140322 02:14:12< gfgtdf> wesbot: seen crab
20140322 02:14:12< wesbot> gfgtdf: Sorry, I don't know of crab.
20140322 02:14:24< gfgtdf> wesbot: seen Crab
20140322 02:14:24< wesbot> gfgtdf: Sorry, I don't know of Crab.
20140322 02:14:32< aquileia> wesbot: seen Crab_
20140322 02:14:32< wesbot> aquileia: The person with the nick Crab_ last spoke 1d 6h ago.  18h 16m ago they left with the message: Quit: Crab_
20140322 02:14:43< gfgtdf> ty
20140322 02:16:14< mattsc> gfgtdf: you need to do an s/another/an up there.  :)  But you found him.
20140322 02:16:35< aquileia> anonymissimus: I mean externalVC12 of course... The link is already there but commented out as I don't know where the file will end up.
20140322 02:18:13< aquileia> I could provide it on github or dropbox, but sourceforge would be nice as the others are already there
20140322 02:23:50-!- sachith500 [~kvirc@112.134.146.132] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140322 02:23:58-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140322 02:32:00-!- aquileia [4e2ad392@gateway/web/freenode/ip.78.42.211.146] has quit [Quit: Page closed]
20140322 02:49:38-!- aquileia [4e2ad392@gateway/web/freenode/ip.78.42.211.146] has joined #wesnoth-dev
20140322 02:50:44< aquileia> Ivanovic, shadowm, anonymissimus: I uploaded the archive here: https://github.com/aquileia/wesnoth-files/blob/master/externalVC12.zip
20140322 02:51:39< aquileia> It'd be nice if someone would transfer it to sourceforge, but I can understand as well if you prefer not to do so.
20140322 02:52:47 * aquileia is off
20140322 02:52:51-!- aquileia [4e2ad392@gateway/web/freenode/ip.78.42.211.146] has quit [Client Quit]
20140322 03:02:24-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 03:06:49-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 246 seconds]
20140322 03:10:04-!- Ivanovic_ [~ivanovic@x2f4f68a.dyn.telefonica.de] has joined #wesnoth-dev
20140322 03:12:54-!- gfgtdf [~chatzilla@f054135125.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]]
20140322 03:13:52-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has quit [Ping timeout: 265 seconds]
20140322 03:13:58-!- Ivanovic_ is now known as Ivanovic
20140322 03:22:30-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 252 seconds]
20140322 03:24:30-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 265 seconds]
20140322 03:26:19< irker031> wesnoth: mattsc wesnoth:1.12 e1e4dc1b3941 / RELEASE_NOTES changelog players_changelog: Update changelogs and release notes http://git.io/ZWzJCQ
20140322 03:34:04< irker031> wesnoth: mattsc wesnoth:master feafb1dadf14 / RELEASE_NOTES changelog players_changelog: Update changelogs and release notes http://git.io/CVRrVA
20140322 03:38:39< iceiceice> gfgtdf: if you remember we were discussing "get_const" like you were saying earlier for config attributes -- turns out that what i said there was wrong. i think what you were proposing was a good idea
20140322 03:40:04-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has joined #wesnoth-dev
20140322 04:01:19-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has quit [Ping timeout: 245 seconds]
20140322 04:10:54-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140322 04:19:03-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Ciao]
20140322 04:20:31-!- goblinThing [44bd8c2c@gateway/web/freenode/ip.68.189.140.44] has quit [Quit: Page closed]
20140322 04:28:23-!- Kevin_Xi [~kevin@223.72.182.158] has left #wesnoth-dev []
20140322 04:36:38-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140322 04:47:50-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20140322 04:50:36-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 04:54:58-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 246 seconds]
20140322 04:56:38-!- qwebirc [75c41fe9@gateway/web/freenode/ip.117.196.31.233] has joined #wesnoth-dev
20140322 05:56:03-!- qwebirc [75c41fe9@gateway/web/freenode/ip.117.196.31.233] has quit [Ping timeout: 245 seconds]
20140322 06:09:26-!- Kevin_Xi [~kevin@223.72.182.158] has joined #wesnoth-dev
20140322 06:13:07-!- qwebirc [75c41fe9@gateway/web/freenode/ip.117.196.31.233] has joined #wesnoth-dev
20140322 06:27:50-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20140322 06:34:07-!- irker031 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20140322 06:36:40-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has quit [Quit: Smell ya later!]
20140322 06:57:33< qwebirc> can i get password for multiplayer battles??
20140322 06:57:44< qwebirc> on the game server???
20140322 06:59:43-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20140322 07:04:15-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140322 07:04:51-!- ancestral [~ancestral@108-249-32-209.lightspeed.mdsnwi.sbcglobal.net] has quit [Quit: Smell ya later!]
20140322 07:09:43-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 252 seconds]
20140322 07:22:37-!- vorobeez [558e940c@gateway/web/freenode/ip.85.142.148.12] has joined #wesnoth-dev
20140322 07:37:04< qwebirc> on the game server???
20140322 07:37:07< qwebirc> can i get password for multiplayer battles??
20140322 07:45:55-!- sachith500 [~kvirc@112.135.71.53] has joined #wesnoth-dev
20140322 07:57:14-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20140322 08:02:56-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140322 08:05:37< zookeeper> Coffee_irc, Aishiko, i don't think the names/id's for ss's are a good idea. i mean, sure, they'd eliminate some very slight duplication that hasn't ever really bothered at least me, but at the cost of making image paths harder to understand. the current way is a very simple and consistent way of referring to images.
20140322 08:08:57-!- Ivanovic [~ivanovic@x2f4f68a.dyn.telefonica.de] has quit [Changing host]
20140322 08:08:57-!- Ivanovic [~ivanovic@wesnoth/developer/ivanovic] has joined #wesnoth-dev
20140322 08:30:54-!- irker064 [~irker@ai0867.net] has joined #wesnoth-dev
20140322 08:30:54< irker064> wesnoth: Nils Kneuper wesnoth:1.12 bc91c86866b7 / data/core/about.cfg po/wesnoth-ai/el.po po/wesnoth-editor/el.po: updated Greek translation http://git.io/2IIAZQ
20140322 08:30:54< irker064> wesnoth: Nils Kneuper wesnoth:1.12 0f7e2241c600 / changelog players_changelog: changelog update for Greek translation update http://git.io/P9T-yQ
20140322 08:30:55< irker064> wesnoth: Nils Kneuper wesnoth:1.12 f50432bf8cf8 / po/ (wesnoth-sotbe/de.po wesnoth-thot/de.po): updated German translation http://git.io/kft1Jg
20140322 08:30:56< irker064> wesnoth: Nils Kneuper wesnoth:master bbdf3a16de8e / data/core/about.cfg po/wesnoth-ai/el.po po/wesnoth-editor/el.po: updated Greek translation http://git.io/kCZhyQ
20140322 08:30:57< irker064> wesnoth: Nils Kneuper wesnoth:master 025342283bf1 / changelog players_changelog: changelog update for Greek translation update http://git.io/hXZtFw
20140322 08:30:58< irker064> wesnoth: Nils Kneuper wesnoth:master 80feaee81880 / po/ (wesnoth-sotbe/de.po wesnoth-thot/de.po): updated German translation http://git.io/qDRNVQ
20140322 08:33:06-!- Ivanovic changed the topic of #wesnoth-dev to: 1.11.12 planned once bug #21808 is fixed | string+feature freeze active on 1.12 | 228 bugs, 353 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140322 08:44:23-!- sachith500 [~kvirc@112.135.71.53] has quit [Read error: Connection reset by peer]
20140322 09:07:38-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Ping timeout: 240 seconds]
20140322 09:08:07-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140322 09:17:52-!- sachith500 [~kvirc@112.135.71.53] has joined #wesnoth-dev
20140322 09:19:17-!- Octalot [~noct@31.185.149.167] has quit [Ping timeout: 264 seconds]
20140322 09:52:43-!- qwebirc [75c41fe9@gateway/web/freenode/ip.117.196.31.233] has quit [Ping timeout: 245 seconds]
20140322 10:08:24-!- sachith500 [~kvirc@112.135.71.53] has quit [Ping timeout: 245 seconds]
20140322 10:15:08-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 10:19:37-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140322 10:22:13-!- stikonas [~gentoo@cpc18-sgyl27-2-0-cust35.18-2.cable.virginm.net] has joined #wesnoth-dev
20140322 10:22:13-!- stikonas [~gentoo@cpc18-sgyl27-2-0-cust35.18-2.cable.virginm.net] has quit [Changing host]
20140322 10:22:13-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140322 10:26:58< AI0867> wesbot: bug 21808
20140322 10:26:58< wesbot> Bug #21808  Assigned to: Andrius Silinskas <thunderstruck>  Status: None  Priority: 7 - High
20140322 10:27:01< wesbot> Summary: cannot join a reloaded game as an observer
20140322 10:27:04< wesbot> Original submission: Affects 1.13.0-dev, 1.12 branch,and 1.11.7 checkout (wi
20140322 10:27:07< wesbot> th localhost 1.11.7 server).If you make a reloaded game which had observers 
20140322 10:27:10< wesbot> URL: http://gna.org/bugs/?21808
20140322 10:58:56-!- markus_ [~mjs-de@f048182147.adsl.alicedsl.de] has joined #wesnoth-dev
20140322 11:02:25-!- mjs-de [~mjs-de@f048241023.adsl.alicedsl.de] has quit [Ping timeout: 240 seconds]
20140322 11:10:06-!- bagzie [~bag@85-76-44-253-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140322 11:30:49-!- sachith500 [~kvirc@112.135.71.53] has joined #wesnoth-dev
20140322 11:31:01-!- irker064 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20140322 11:48:40< thunderstruck> iceiceice, Soliton: I was looking into bug 21808. The problem is that mp::configure is responsible for some part of mp_game_settings initialisation.
20140322 11:49:17< thunderstruck> And so, if it is skipped, these values will be replaced by defaults.
20140322 11:49:53< thunderstruck> The solution is probably to get these values from save game config for reloaded games.
20140322 11:58:53-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has joined #wesnoth-dev
20140322 12:01:31< Coffee_irc> zookeeper: you could implement the spritesheets several ways, but I think it makes sense to define the size, etc. in an 'id' mapped collection once for a unit
20140322 12:02:05< Coffee_irc> otherwise you need to define size, etc. for each animation type for a unit at each point
20140322 12:03:29-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 12:06:01-!- Kevin_Xi [~kevin@223.72.182.158] has left #wesnoth-dev []
20140322 12:06:19-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140322 12:07:59-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 245 seconds]
20140322 12:14:01-!- sachith500 [~kvirc@112.135.71.53] has quit [Read error: Connection reset by peer]
20140322 12:15:15-!- sachith500 [~kvirc@112.135.71.53] has joined #wesnoth-dev
20140322 12:16:06-!- sachith500 [~kvirc@112.135.71.53] has quit [Read error: Connection reset by peer]
20140322 12:20:36-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20140322 12:33:37-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140322 12:36:10-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140322 12:41:20-!- mordante [~mordante@wesnoth/developer/mordante] has joined #wesnoth-dev
20140322 12:41:34< mordante> servus
20140322 12:43:10< mordante> riftwalker std::swap will probably be differently implemented in C++11 since there we have move support
20140322 12:44:08< mordante> RiftWalker, actually I haven't seen std::swap used in the assignment operator, I usually see a destructor and placement new
20140322 12:48:16< mordante> Aishiko, I think caps on the number of animations in a sprite-sheet row is a bad idea
20140322 12:48:58< mordante> Aishiko, a limit that looks sane now, might bit us in the future unless it's really huge (1024 images per row for example)
20140322 12:49:31< mordante> Aishiko, also not sure whether we need to enforce rows, some animations are used in multiple animations
20140322 12:52:29< mordante> _8680_, 124 was closed for two reasons: the main one adding functions without a use case
20140322 12:53:26< mordante> _8680_, the second once somebody in the future would need to know the number of bits s/he would probably not know about this function and write sizeof(…) * CHAR_BIT in that place
20140322 12:53:48< thunderstruck> What is a convenient way to push a commit to both master and 1.12?
20140322 12:54:56< mordante> what do you exactly mean? you have one patch for both master and 1.12?
20140322 12:55:03< thunderstruck> mordante: yes
20140322 12:56:19< mordante> what I do is commit the patch for 1.12 -> push 1.12 -> pull/fetch master -> cherry-pick the committed SHA1 id from 1.12 -> maybe fix the changelog -> push master
20140322 12:56:40< mordante> or if I have a patch sometimes just apply the patch from a diff directly to both branches
20140322 12:58:19< mordante> _8680_, aquileia if the two of you together see a usecase for that patch in a branch of aquileia then aquileia should pull your branch is his branch
20140322 12:58:56< mordante> _8680_, aquileia then when aquileia's pull request comes the function does have a usecase
20140322 12:59:09-!- adityapande [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #wesnoth-dev
20140322 13:00:05< mordante> _8680_, aquileia and before merging please change the numeric_limits to CHAR_BIT, the latter is much easier to understand
20140322 13:00:23< thunderstruck> mordante: Ok. I guess I do the same, but first push master, because my commit is already there.
20140322 13:01:39< mordante> that's also possible, when I have a commit for both I tend to fix it in 1.12 first, hence the order
20140322 13:04:06< thunderstruck> iceiceice: https://github.com/thunderstruck/wesnoth/commit/fa5b916bf1240c20d3faccb4f3be59c00003433b
20140322 13:04:14< thunderstruck> In case you're interested
20140322 13:05:06< mordante> Aishiko, Coffee_irc indeed the spritesheet idea is to use them, which should improve performance, but nothing more is to be done
20140322 13:05:26-!- gfgtdf [~chatzilla@f054135125.adsl.alicedsl.de] has joined #wesnoth-dev
20140322 13:05:35< gfgtdf> wesbot: sen crab_
20140322 13:05:40< gfgtdf> wesbot: seen crab_
20140322 13:05:40< wesbot> gfgtdf: The person with the nick Crab_ last spoke 1d 17h ago.  1d 5h ago they left with the message: Quit: Crab_
20140322 13:06:26-!- aquileia [4e2ad392@gateway/web/freenode/ip.78.42.211.146] has joined #wesnoth-dev
20140322 13:06:29 * mordante 's somewhat AFK looking at GSoC proposals
20140322 13:06:40-!- irker004 [~irker@ai0867.net] has joined #wesnoth-dev
20140322 13:06:40< irker004> wesnoth: Andrius Silinskas wesnoth:master fa5b916bf124 / changelog src/mp_game_utils.cpp: Add missing MP settings for reloaded games. Fixes #21808. http://git.io/eyrz2g
20140322 13:11:47< aquileia> mordante: Ok, I think I'll do a separate pull request for byte_size_from_utf8_first() with _8680_'s branches merged before.
20140322 13:11:54-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20140322 13:13:11< mordante> aquileia, yes please adding unused functions is not something I enjoy, especially when completely trivial
20140322 13:14:03< aquileia> The current state of my working branch includes too many other changes, so I think it's better to separate it
20140322 13:14:59< aquileia> at least the last time I did a PR shadowm said I should split it...
20140322 13:16:30< aquileia> mordante: Did you consider the 'tcontent' solution?
20140322 13:21:09< irker004> wesnoth: Andrius Silinskas wesnoth:1.12 52e7d96ba26d / changelog src/mp_game_utils.cpp: Add missing MP settings for reloaded games. Fixes #21808. http://git.io/Bu4aJQ
20140322 13:21:15< thunderstruck> Ivanovic: ^
20140322 13:21:21< thunderstruck> #21808 should be fixed
20140322 13:21:40< mordante> aquileia, best try to make one pull request per topic, that makes reviewing and merging them easier
20140322 13:21:43< thunderstruck> and sorry for the release delay
20140322 13:22:17< aquileia> mordante: Ok, then I'll:
20140322 13:22:33< mordante> aquileia, haven't had time to look at it, and today I will spend a lot of time at reviewing all proposals so not sure whether I have time today
20140322 13:22:35< aquileia> 1. Do a pull request for the u8 function family
20140322 13:23:15< aquileia> 2. Create a PR for byte_size_from_utf8_first() with _8680_'s branches merged in
20140322 13:23:40< aquileia> 3. Create a PR for numbox without tcontent
20140322 13:23:55< aquileia> 4. Create a PR for tcontent
20140322 13:24:18< aquileia> But don't say I didn't warn you ;)
20140322 13:25:05< mordante> I don't mind you don't know how many topic branches I have locally :-P maybe 3 and 4 can be merged
20140322 13:27:11-!- wesbot changed the topic of #wesnoth-dev to: 1.11.12 planned once bug #21808 is fixed | string+feature freeze active on 1.12 | 227 bugs, 353 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140322 13:28:38< aquileia> Ivanovic: "once bug #21808 is fixed" ... that seems to be fulfilled
20140322 13:28:44< Necrosporus> First scenario in "Invasion from the Unknown" has string [gold_carryover] bonus=no carryover_percentage=80 [/gold_carryover] but when I finish it seem there's early finish bonus anyways
20140322 13:28:52-!- happygrue [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20140322 13:36:34< Necrosporus> I'm not sure if that's problem of wesnoth 1.10.7 or addon
20140322 13:38:43-!- Kevin_Xi [~kevin@223.72.182.158] has joined #wesnoth-dev
20140322 13:47:19< irker004> wesnoth: loonycyborg wesnoth:master 7116ccf9efe5 / SConstruct: scons: Made out-of-tree builds with -Y option work http://git.io/x57vRQ
20140322 13:48:09< loonycyborg> It just told me the reporitory has moved when I made a commit
20140322 13:48:38< loonycyborg> What exactly should I do about this?
20140322 13:49:19-!- adityapande [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Quit: Page closed]
20140322 13:49:50< Ivanovic> AI0867, Coffee_irc, fabi, iceiceice, loonycyborg, mordante, shadowm, Soliton, trademark, zookeeper, everyone else who cares: ready for 1.11.12?
20140322 13:50:12< loonycyborg> This depends on the answer on my above question :P
20140322 13:51:32-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 13:51:39-!- markus_ is now known as mjs-de
20140322 13:51:49< loonycyborg> hmm
20140322 13:51:50< Ivanovic> uhm, the repository is at git@github.com:wesnoth/wesnoth.git
20140322 13:52:06< Ivanovic> did you still use the wesnoth-old.git?
20140322 13:52:10< loonycyborg> yes
20140322 13:52:19< loonycyborg> It still seems to have redirection
20140322 13:52:34< loonycyborg> So my commit made it to right repository right?
20140322 13:52:36< Ivanovic> it still has but you should move to the "new" location anyway
20140322 13:52:39< Ivanovic> it should have
20140322 13:53:12< Ivanovic> since when going to https://github.com/wesnoth/wesnoth-old i am redirected correctly
20140322 13:53:35< loonycyborg> Will editing .git/config and fixing the url there do the right thing?
20140322 13:53:56< Ivanovic> that should be good enough
20140322 13:54:03< loonycyborg> ok
20140322 13:54:27< Ivanovic> hmm, anyway, i don't have the old tarball here and my upload speed here is only a third of what i got in my flat in neu-isenburg
20140322 13:54:35< Ivanovic> i guess i will wait until tomorrow with releasing...
20140322 13:54:59-!- Ivanovic changed the topic of #wesnoth-dev to: 1.11.12 planned for 23.03.2014, 17:00 UTC | string+feature freeze active on 1.12 | 227 bugs, 353 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140322 13:55:17< Ivanovic> folks, you got a little more than 24h left to get your stuff in!
20140322 13:55:52< Ivanovic> and i think i should really order this FTTH access
20140322 13:56:02< Ivanovic> those 200 mbit down 100 up sound so great...
20140322 13:56:05-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 246 seconds]
20140322 13:56:27< Ivanovic> but is this really worth a premium of 25€?
20140322 13:58:35-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140322 14:00:01< shadowm> happygrue: I saw some balancing changes to THoT in master but not 1.12?
20140322 14:00:18< shadowm> aquileia: I can't help you with the dependencies package.
20140322 14:00:45< aquileia> ok
20140322 14:01:43-!- lobby [~wesnoth@wesnoth/bot/lobby] has joined #wesnoth-dev
20140322 14:01:43-!- Topic for #wesnoth-dev: 1.11.12 planned for 23.03.2014, 17:00 UTC | string+feature freeze active on 1.12 | 227 bugs, 353 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140322 14:01:43-!- Topic set by Ivanovic [~ivanovic@wesnoth/developer/ivanovic] [Sat Mar 22 13:54:59 2014]
20140322 14:01:43[Users #wesnoth-dev]
20140322 14:01:43[ _8680_         ] [ Duthlet  ] [ Gallaecio ] [ Kevin_Xi    ] [ Samual         ] [ trademark] 
20140322 14:01:43[ AI0867         ] [ EdB      ] [ Gambit    ] [ knotwork    ] [ shadowm        ] [ Upth     ] 
20140322 14:01:43[ Aishiko        ] [ ejls     ] [ gfgtdf    ] [ lobby       ] [ shadowm_desktop] [ vorobeez ] 
20140322 14:01:43[ apoi           ] [ elias    ] [ happygrue ] [ Marqin      ] [ shikadibot     ] [ vultraz  ] 
20140322 14:01:43[ aquileia       ] [ EliDupree] [ iceiceice ] [ matthiaskrgr] [ Smar           ] [ wesbot   ] 
20140322 14:01:43[ bagzie         ] [ enchilado] [ irker004  ] [ melinath    ] [ Soliton        ] [ yann     ] 
20140322 14:01:43[ Coffee_irc     ] [ Espreon  ] [ Ivanovic  ] [ mjs-de      ] [ stikonas       ] [ zookeeper] 
20140322 14:01:43[ Crendgrim      ] [ esr      ] [ iwaim     ] [ molgrum     ] [ thunderstruck  ] [ {V}      ] 
20140322 14:01:43[ crimson_penguin] [ exciton  ] [ janebot   ] [ mordante    ] [ timotei        ] 
20140322 14:01:43[ DDR            ] [ fabi     ] [ Jetrel    ] [ Necrosporus ] [ ToBeFree       ] 
20140322 14:01:43[ DHost          ] [ Fortescue] [ justinzane] [ Rh0nda      ] [ tomreyn        ] 
20140322 14:01:43-!- Irssi: #wesnoth-dev: Total of 63 nicks [0 ops, 0 halfops, 0 voices, 63 normal]
20140322 14:01:44< shadowm> Ivanovic, loonycyborg The correct way to reassign the repository URL is included in my email from that day, but hand-editing the config file also works as a last resort solution.
20140322 14:01:45-!- nurupo [~nurupo.ga@unaffiliated/nurupo] has joined #wesnoth-dev
20140322 14:01:49-!- Channel #wesnoth-dev created Tue Jan 27 05:28:41 2009
20140322 14:01:57-!- Ardonik [~user@adsl-75-28-99-109.dsl.irvnca.sbcglobal.net] has joined #wesnoth-dev
20140322 14:01:57< shadowm> Bah.
20140322 14:02:13< shadowm> I just logged in and the network is already falling apart.
20140322 14:02:55-!- Irssi: Join to #wesnoth-dev was synced in 85 secs
20140322 14:02:59-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140322 14:03:07-!- loonycyborg [~loonycybo@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20140322 14:03:23< aquileia> Ivanovic: Would you upload https://github.com/aquileia/wesnoth-files/blob/master/externalVC12.zip at http://sourceforge.net/projects/wesnoth/files/unofficial/Windows%20Compile%20Stuff/ when it suits you?
20140322 14:03:24< shadowm> loonycyborg: The correct way to reassign the repository URL is included in my email from that day, but hand-editing the config file also works as a last resort solution.
20140322 14:03:38-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 240 seconds]
20140322 14:04:59< loonycyborg> What e-mail? What's the topic? I only remember seeing some e-mails about gsoc :P
20140322 14:05:13 * shadowm hits loonycyborg with the squeaky rubber mallet.
20140322 14:05:16-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140322 14:05:38< shadowm> loonycyborg: https://mail.gna.org/public/wesnoth-dev/2014-02/msg00125.html
20140322 14:06:01-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140322 14:06:05< shadowm> But since it's a mailing list and email as a medium sucks, the archive there shows the command censored since it looks like an email address.
20140322 14:06:31< Ivanovic> ah, what a smart idea, spaces in filenames on a server...
20140322 14:07:08< shadowm> *part of it looks like
20140322 14:07:30< aquileia> I'll change the wiki page refering to it anyway, so... a new path would be no problem
20140322 14:08:40< aquileia> It's just that the VC9 and VC10 versions are already there, so it'd be nice to have all three at one spot
20140322 14:09:19< aquileia> And - if you want to do it at all - take your time, there's no need to hurry
20140322 14:09:23< Ivanovic> aquileia: i am uploading there right now
20140322 14:09:29< Ivanovic> the upload just takes a little...
20140322 14:09:44< aquileia> I thought your connection isn't that good today?
20140322 14:10:16-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140322 14:10:38< Ivanovic> yeah, just 1mbit upstream instead of 3.5mbit
20140322 14:10:49-!- thunders1ruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140322 14:10:52< shadowm> Pfft.
20140322 14:10:53< Ivanovic> which is annoying for a 350MB package
20140322 14:11:33< shadowm> Here I have 2 mbps down and 1 mbps up or so -- maximum, not average.
20140322 14:14:09< happygrue> shadowm: yes, I was having some git issues and got into a time crunch so I left it.  I have been using a GUI client so I didn't want to mess with it and cause problems because I am still only comfortable doing very basic things with git.
20140322 14:15:42-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140322 14:16:35< mordante> Ivanovic, I'm ready
20140322 14:17:48< mattsc> Ivanovic: me too
20140322 14:19:01< zookeeper> Coffee_irc, sure, but i always advocate not needing to define those things in the first place :p
20140322 14:19:55-!- bagzie [~bag@85-76-44-253-nat.elisa-mobile.fi] has quit []
20140322 14:20:59-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Ping timeout: 269 seconds]
20140322 14:21:57-!- bagzie [~bag@85-76-67-13-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140322 14:22:52< vorobeez> hello, mattsc, i download and install AI-demos addon to my wesnoth. Now i change some code in test scenario and lua-test. How i can execute this scenario in game?
20140322 14:25:30< happygrue> shadowm: I'm busy for a bit but I will look again later this afternoon. Does it cause problems to cherry pick those two commits (one tweak, one whitespace fix) over to 1.12 now?
20140322 14:25:55< mattsc> vorobeez: you select the campaign in the campaign menu and then move to the test scenario sign post in the first scenario.
20140322 14:26:25< mattsc> vorobeez: alternatively from the commandline: wesnoth -t aid_test
20140322 14:26:49-!- sachith500 [~kvirc@112.135.71.53] has joined #wesnoth-dev
20140322 14:27:09< shadowm> happygrue: You can expect them to conflict over the changelog files.
20140322 14:28:06< shadowm> happygrue: Good news, the first one doesn't conflict: http://pastebin.com/Tz9NrmMe
20140322 14:28:20< vorobeez> mattsc: thank you)
20140322 14:28:53< shadowm> happygrue: Wait, the second one probably wouldn't conflict either, it's just that I haven't applied the first one which it relies upon.
20140322 14:29:31< mattsc> vorobeez: I also usually add the -d (—debug) option (or start debug mode in-game with :debug), as being able to add/kill/modify units is really useful for a lot of the tests.
20140322 14:30:29< shadowm> Uh. Wow, I picked the wrong second commit somehow.
20140322 14:30:47< shadowm> happygrue: They don't conflict, actually. :) <http://pastebin.com/fSwFjLvy>
20140322 14:35:58-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 246 seconds]
20140322 14:36:57-!- thunders1ruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Quit: leaving]
20140322 14:37:12-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140322 14:41:27< Ivanovic> okay, i really love this f***ing webinterface from sf.net
20140322 14:42:52< Ivanovic> and which genius had the idea that spaces in filenames make sense?
20140322 14:44:49< happygrue> \o/
20140322 14:44:50< happygrue> thanks shadowm
20140322 14:45:12< Duthlet> If I make changes after a pull request, should I just make a new PR?
20140322 14:45:53< AI0867> Ivanovic: what about spaces? I recently disallowed them
20140322 14:46:40< aquileia> AI0867: It's a subdir on sourceforge, not in Wesnoth itself
20140322 14:47:34< aquileia> AI0867: As you were the most involved in it, could you review https://github.com/wesnoth/wesnoth/pull/126 ?
20140322 14:49:13< aquileia> After that I'd do another PR to substitute the current byte_size_... by _8680_'s function
20140322 14:53:28< AI0867> I've already looked at it
20140322 14:54:24< aquileia> 4 minutes... that's fast
20140322 14:54:56< happygrue> shadowm: so is a pull request on 1.12 the safest way for me to move those two commits over to 1.12?
20140322 14:55:20< happygrue> redunant redundancy ftw!
20140322 14:55:43< Ivanovic> AI0867: spaces in the SF.net interface
20140322 14:58:07< aquileia> Ivanovic: Thanks
20140322 15:00:21-!- timotei [~timotei@wesnoth/developer/timotei] has quit [Quit: leaving]
20140322 15:00:41-!- timotei_ [~timotei@79.119.106.168] has joined #wesnoth-dev
20140322 15:00:41-!- timotei_ [~timotei@79.119.106.168] has quit [Changing host]
20140322 15:00:41-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20140322 15:07:29< vorobeez> mattsc: i can use [avoid] like this: [ai][avoid] terrain = V [/avoid][/ai]?
20140322 15:07:51< vorobeez> [ai][avoid] x,y = 20,4 [/avoid][/ai]?
20140322 15:11:19< aquileia> Hmm... I tried to pass a string instead of a reference to it
20140322 15:11:29< mattsc> vorobeez: assuming V is a placeholder for the real terrain code, then yes
20140322 15:12:36< vorobeez> mattsc: i put in test_lua.lua "DBG.dbms(ai.get_avoid())", but it return empty table =\
20140322 15:12:54< vorobeez> in test.cfg i put [avoid] tag
20140322 15:14:28< vorobeez> mattsc: do i need [filter] tag?
20140322 15:17:09< mattsc> no - you shouldn’t
20140322 15:18:14< vorobeez> mattsc: i don't understand what's wrong
20140322 15:18:42< mattsc> vorobeez: I need to be afk for ~30min; if you can’t figure it out, pastebin me your scenario and lua code and I’ll have a look when I get back
20140322 15:18:50< mordante> aquileia, just had a quick glimpse at your utf8 patch, maybe make a namespace utf8 and move everything in there
20140322 15:19:00< vorobeez> mattsc: ok, thank you
20140322 15:19:13< mordante> aquileia, that gives utf8::string, which looks nicer to me
20140322 15:19:33< aquileia> mordante: But utf8_string already exists
20140322 15:20:14< aquileia> So I could possibly break existing code
20140322 15:20:29< mordante> it can't if you convert everything
20140322 15:20:41< mordante> imo u8erase looks rather ugly as function name
20140322 15:21:04< aquileia> oh well...
20140322 15:21:20< mordante> btw I really like the ideas in that patch :-)
20140322 15:21:26< aquileia> Thanks
20140322 15:22:16< vorobeez> mattsc: http://pastebin.com/iXqN96m5    http://pastebin.com/xy6x1VUH
20140322 15:22:32< mordante> I hope that at some point we have and utf.cpp and utf.hpp and have all utf code in that file and then can drop the wide_string stuff
20140322 15:23:30< AI0867> mordante: I'm doing just that
20140322 15:23:40< mordante> AI0867, :-)
20140322 15:23:40< aquileia> wanted to write that...
20140322 15:23:43< AI0867> though with a utf16 string for windows compat
20140322 15:24:04< shadowm> happygrue: Depends on whether you can or cannot switch branches easily there.
20140322 15:24:09< mordante> too bad we still need that
20140322 15:24:11< aquileia> AI0867: What does that imply for my patch?
20140322 15:24:24< AI0867> aquileia: I can do my stuff on top of yours
20140322 15:24:27< aquileia> ok
20140322 15:24:41< AI0867> it all compiles and works locally, but I haven't actually committed anything yet
20140322 15:24:50< AI0867> (nor can I test the windows API stuff)
20140322 15:25:02< aquileia> perheps it's really time for utf.cpp then (for both my UTF-8 and your UTF-16 additions)
20140322 15:25:12< aquileia> *perhaps
20140322 15:25:20< mordante> aquileia, regarding u8index, maybe return std::string::npos if the position is invalid
20140322 15:26:05-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140322 15:26:07< mordante> (and let it return a std::size_t)
20140322 15:27:01< aquileia> That's possible, but the current implementation has its advantage as well: u8truncate automatically fetches invalid input because u8index corrects it
20140322 15:27:03< mordante> also for the index, these are the common size types for strings (actually they are std::string::size_type)
20140322 15:28:11< aquileia> What's bad with size_t ?
20140322 15:28:28< mordante> nothing but you use unsigned
20140322 15:28:42< aquileia> in one case
20140322 15:29:05< aquileia> I probably overlooked it when switching to size_t
20140322 15:29:26< mordante> I hoped that was the reason ;-)
20140322 15:30:21< mordante> why is truncate happier with the length instead of std::string::npos?
20140322 15:31:28< aquileia> Because if you want to truncate at a non-existing position, it simply does nothing
20140322 15:33:01-!- travis-ci [~travis-ci@ec2-23-22-178-67.compute-1.amazonaws.com] has joined #wesnoth-dev
20140322 15:33:01< travis-ci> [travis-ci] gfgtdf/wesnoth-old#51 (sync_2 - e1fde8b : gfgtdf): The build was broken.
20140322 15:33:01< travis-ci> [travis-ci] Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/21317762
20140322 15:33:01-!- travis-ci [~travis-ci@ec2-23-22-178-67.compute-1.amazonaws.com] has left #wesnoth-dev []
20140322 15:33:15< aquileia> That's how the old truncate_as_wstring works as well (also the reason a bug where it would have truncated at a wrong position never had consequences)
20140322 15:33:38< mordante> but std::string::npos is also an invalid position
20140322 15:33:47< aquileia> But if you want -1, then I'll just catch that separately
20140322 15:35:03< aquileia> mordante: I wanted to say that the current u8index, by returning the length, catches this without a separate check
20140322 15:35:19< aquileia> But if you want -1, no problem
20140322 15:35:44 * aquileia sees he repeated himself
20140322 15:36:09< Ivanovic> aquileia: file should be online
20140322 15:36:32< aquileia> Ivanovic: Yes, the wiki is already updated. Thanks again
20140322 15:36:39< Ivanovic> cool, thanks!
20140322 15:39:43-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 15:43:26< aquileia> mordante, AI0867, as to the assert... can I delete it?
20140322 15:43:28< mattsc> vorobeez: ah, yes, there is something that you couldn’t really have known
20140322 15:43:39< mordante> aquileia, which line
20140322 15:43:58< aquileia> text.cpp line 189 in my patch
20140322 15:43:58-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140322 15:44:50< mattsc> vorobeez: there is a bug (well, actually, Crab_ says it used to be an intentional feature, but these days it really is a bug), that aspects (such as aggression or [avoid]) set in the [side] definition are not recognized if a Lua engine is also defined.
20140322 15:45:29< mattsc> vorobeez: what you need to do instead is to put this into an event (e.g. prestart): http://pastebin.com/Zkatq1jR
20140322 15:46:48< mattsc> vorobeez: also, for help with trouble shooting things like this, you can do the following: type :inspect once the scenario has started, then click on ‘Team 1’, and then on ‘ai component structure’ and/or ‘ai config full’.
20140322 15:47:13< mattsc> That will show you if your code is not reading the [avoid] content correctly, or if it did not get set correctly for the side.
20140322 15:47:22< mordante> aquileia, maybe I missed something what happens if you do »truncate("Spaß", 5);« with the current code
20140322 15:48:04< aquileia> currently truncate_as_widestring will do nothing, which seems fine
20140322 15:48:14< aquileia> utf8::truncate will do so as well
20140322 15:48:34-!- Spoffy [~chatzilla@152.78.175.8] has joined #wesnoth-dev
20140322 15:48:43< aquileia> std::truncate... errors, I think?
20140322 15:49:03< aquileia> It won't handle it correctly
20140322 15:49:58< aquileia> oh, no... for std::truncate "Spaß" would have 5 chars, so it'd work as well
20140322 15:50:03-!- bagzie [~bag@85-76-67-13-nat.elisa-mobile.fi] has quit []
20140322 15:52:42-!- MaraJade [goossenm@shell.onid.oregonstate.edu] has joined #wesnoth-dev
20140322 15:54:21< mordante> from the comment it's not clear whether it returns 4 or 5 for the index function
20140322 15:54:33< vorobeez> mattsc: thank you
20140322 15:55:16< mordante> afk
20140322 15:55:30< aquileia> mordante: codepoint index corresponing to the ...th character in an UTF-8 encoded string
20140322 15:55:34< aquileia> better?
20140322 16:16:10-!- neXyon [~neXyon@85-127-251-230.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20140322 16:22:33-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140322 16:23:02< aquileia> AI0867, mordante, after I renamed all instances of utf8_string to utf8::string, I now get an error "qualified name is not allowed" in the typedef
20140322 16:24:14< aquileia> Is 'string' somehow protected in C++ ?
20140322 16:25:09< aquileia> It's no keyword I think, so...
20140322 16:27:25< shadowm> Is there a namespace utf8 visible in that unit?
20140322 16:27:31-!- anonymissimus [~chatzilla@HSI-KBW-149-172-228-192.hsi13.kabel-badenwuerttemberg.de] has joined #wesnoth-dev
20140322 16:28:07< aquileia> shadowm: It is defined later on in that file
20140322 16:28:21< shadowm> The namespace needs to be defined first.
20140322 16:28:55< aquileia> So I'll have to move it in the namespace away from the other string typedef's?
20140322 16:29:47< aquileia> anonymissimus: I updated http://wiki.wesnoth.org/CompilingWesnothOnWindows, feel free to revert some of the changes if you disagree
20140322 16:29:55< shadowm> I said that the namespace needs to be defined first, not that the typedef needs to be in the namespace -- that would depend on what you are actually trying to do.
20140322 16:29:59< anonymissimus> mordante: is the cmake system supposed to require the project root folder to be named wesnoth-code ? if no, could you pls make a quick check by renaming your root folder to something else ?
20140322 16:30:19< anonymissimus> mordante: perhaps also delete CMakeCache.txt to be sure
20140322 16:31:37< anonymissimus> since it happened to me that it wrote the build files into the wrong folder when I have 2 folders with the wesnoth source parallel
20140322 16:31:48< aquileia> shadowm: It works if I move it, but I'd prefer to have it where it was before... How could I define the namespace early?
20140322 16:33:00< aquileia> shadowm: http://pastebin.com/YVR4gqFj
20140322 16:33:09< aquileia> That's the current situation
20140322 16:33:25< shadowm> Why not have the typedefs at the end?
20140322 16:33:59< aquileia> They are needed all the time, I don't know if that'd work
20140322 16:33:59< shadowm> All you need for the typedef to work is to have declared the types it deals with somewhere before in the same unit (including any headers it depends upon), so... namespace utf8 { class string; } or (IIRC) class utf8::string;, but the latter probably won't work without the compiler knowing beforehand that utf8 names a namespace and not a class/struct/union.
20140322 16:34:22< shadowm> Those two proposals are forward-declarations of utf8::string.
20140322 16:34:33< shadowm> Oh wait.
20140322 16:34:42< shadowm> It's in the right-hand side of the typedef.
20140322 16:35:06< shadowm> namespace utf8 { typedef std::string string; }
20140322 16:35:15< aquileia> Thanks!
20140322 16:35:37< shadowm> Or something with the using directive may work too.
20140322 16:36:09< aquileia> It works with defining the namespace twice
20140322 16:36:28 * shadowm is not too familiarized with using beyond the using namespace form...
20140322 16:37:31< shadowm> After a quick read, it seems namespace utf8 { using std::string; } should work too.
20140322 16:39:17< anonymissimus> aquileia: i trust you on the matter; i need to be very thankful for anyone updating it
20140322 16:39:52< anonymissimus> aquileia: if you're using 2008 version you can update the project file for new files as well
20140322 16:40:06< anonymissimus> most of the time that's all about build maintenance
20140322 16:40:08< aquileia> anonymissimus: I use 2013
20140322 16:40:12< shadowm> I hate these "last check" things.
20140322 16:40:54-!- travis-ci [~travis-ci@ec2-23-22-178-67.compute-1.amazonaws.com] has joined #wesnoth-dev
20140322 16:40:54< travis-ci> [travis-ci] gfgtdf/wesnoth-old#52 (sync_2 - c7afb25 : gfgtdf): The build was fixed.
20140322 16:40:54< travis-ci> [travis-ci] Build details : http://travis-ci.org/gfgtdf/wesnoth-old/builds/21320011
20140322 16:40:54-!- travis-ci [~travis-ci@ec2-23-22-178-67.compute-1.amazonaws.com] has left #wesnoth-dev []
20140322 16:41:01< shadowm> Most of the time they only discourage people from even trying when the specified commit/date is older than a couple of weeks. It'd be more productive to include instructions for updating file lists as necessary.
20140322 16:41:22< aquileia> I splitted 2008 because it seems it has to be handled separately, I left it as is
20140322 16:41:27< shadowm> Also, there's absolutely no reason to include those ancient guides in the page.
20140322 16:42:35< shadowm> If they are old and no longer useful, then they should be removed, or at the very least moved to a separate page. If they may still be useful, then someone should take care of bringing them up to date.
20140322 16:44:41< anonymissimus> shadowm: I'd rather say there's absolutely no reason to delete old outdated guides for as long as the main reference points to the newest guide
20140322 16:44:59< aquileia> shadowm: Some of them are useful because they involve different toolchains
20140322 16:45:09< vultraz> I cannot see the point of keeping old outdated guides
20140322 16:45:19< shadowm> anonymissimus: Which is why I said "at the very least moved to a separate page".
20140322 16:45:48< anonymissimus> some of those guides include info on how to build the prerequisites for instance
20140322 16:45:53< shadowm> Otherwise it's just bothersome visual cruft that may even confuse people who don't realize they don't need to/should not read it.
20140322 16:46:12< vultraz> If there's anything useful in them those bits should be copied
20140322 16:47:10< shadowm> The introduction to this section is absolutely unacceptable as documentation: http://wiki.wesnoth.org/CompilingWesnothOnWindows#Old_guides
20140322 16:47:14-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has quit [Quit: leaving]
20140322 16:47:19< aquileia> vultraz: It's not that easy... MSVC and minGW are completely different
20140322 16:47:52< gfgtdf> iceiceice: online ?
20140322 16:49:06< aquileia> shadowm, vultraz, anonymissimus: If we agree that wesnoth.sln always is up-to-date, some of those parts could be moved/obsoleted
20140322 16:49:11< shadowm> This is so ridiculously outdated it's absurd: http://wiki.wesnoth.org/CompilingWesnothOnWindows#.5BXan.27s_guide_to_Compiling_the_Battle_for_Wesnoth_on_Windows.5D
20140322 16:49:50-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20140322 16:49:56< shadowm> aquileia: No, we will never agree on that, hence "It'd be more productive to include instructions for updating file lists as necessary.".
20140322 16:51:04< vultraz> Why does this page not include shadowm's great CB documentation
20140322 16:53:07< anonymissimus> aquileia: in reality, over the last 2 years or even more, the wesnoth.sln is as up to date as active *I* am
20140322 16:53:32< anonymissimus> since none else updates is sort of regularly
20140322 16:53:56< anonymissimus> so it's not certainly up to date
20140322 16:54:12< aquileia> anonymissimus: Oh, one thing... it'd be nice if the default build configuration would be ReleaseDEBUG instead of Debug
20140322 16:54:56< Necrosporus> vultraz, the point is one might want to complie and use old version of wesnoth
20140322 16:55:06< aquileia> For now I included how to change that manually in the guide, but...
20140322 16:55:07< anonymissimus> aquileia: you can easily chance that...but mixtures of release and debug like that will fool you
20140322 16:55:21< anonymissimus> they dont display values correctly
20140322 16:55:36< aquileia> oh... I wasn't aware of that
20140322 16:55:39< anonymissimus> what a variable is and such
20140322 16:55:59< anonymissimus> perhaps its better in newer MSVC versions though
20140322 16:56:03< shadowm> Necrosporus: Older versions are usually available prebuilt on SF.net, and if someone needs to build something between releases they should look at older versions of the page.
20140322 16:56:11< vultraz> ^
20140322 16:56:14< shadowm> It's a wiki. Past versions of each page are preserved.
20140322 16:56:27< anonymissimus> for 2010 at least they apparently only "improved" the gui
20140322 16:56:51< vultraz> And in any case, the chances someone would want to build an ancient wesnoth version are pretty slim
20140322 16:57:34< shadowm> Especially on Windows and especially with Microsoft "I don't know how to do parallel incremental builds" Visual C++.
20140322 16:57:59< aquileia> incremental yes, parallel no
20140322 16:58:14< shadowm> No, it can do either thing, I am aware.
20140322 16:58:23< shadowm> It just can't do _both_.
20140322 16:58:51< shadowm> At least not the last version I tried before I decided Microsoft clearly didn't want me to develop software for their platform.
20140322 16:59:23< anonymissimus> parallel would be multiple processes, and incremental minimum rebuild ?
20140322 16:59:41< shadowm> Yes.
20140322 17:00:24< anonymissimus> hm yeah i think thats still true and one reason I could never build wesnoth as fast as I want
20140322 17:01:10< aquileia> I think it would be nice to figure out which flags clang needs to run with MSVS
20140322 17:01:10< anonymissimus> though when compiling with MinGw in parallel, its using full CPU anyway
20140322 17:01:20-!- sachith500|2 [~kvirc@112.134.181.96] has joined #wesnoth-dev
20140322 17:02:09< shadowm> I'd expect clang on Windows to require not many additional flags besides those that are required for it on Linux. For building Wesnoth in particular, that is.
20140322 17:02:12< aquileia> Clang includes nicely into Visual Studio, but fails at interpreting the specific flags
20140322 17:03:27< anonymissimus> the problem could be whether Clang can compile the system headers
20140322 17:03:46< anonymissimus> or is it using MinGw headers on windows ?
20140322 17:03:55< anonymissimus> or its own ones
20140322 17:04:01-!- sachith500 [~kvirc@112.135.71.53] has quit [Ping timeout: 240 seconds]
20140322 17:04:41< aquileia> As it includes into MSVS, it is passed the same flags/headers as VC++ I think
20140322 17:04:50< vultraz> If the old guides get moved to a new page, be sure to include the Obsolete template
20140322 17:05:00< aquileia> That is?
20140322 17:05:55< vultraz> aquileia: was that directed at me?
20140322 17:06:02< aquileia> yes
20140322 17:08:56< vultraz> look at w.w.o/Template:Obsolete
20140322 17:09:08< vultraz> it includes that
20140322 17:10:16< mordante> aquileia, then the index would return 4 u8index for Spaß?
20140322 17:11:27< aquileia> mordante: utf8::size("Spaß") returns 4, utf8::index("Spaß",5) returns 5
20140322 17:11:34< mordante> aquileia, namespaces are 'open' 'objects' so for example all code in gui2 uses »namespace gui« being placed in the same namespace
20140322 17:11:36< aquileia> if you could pass strings, that is
20140322 17:12:08< mordante> aquileia, only anonymous namespaces have slightly different rules, just don't put them in a header and you're fine
20140322 17:12:18< aquileia> utf8::index("ünicode",5) returns 6
20140322 17:13:13< aquileia> it just shifts the index backwards and returns the last codepoint if the string ends prematurely
20140322 17:13:52< mordante> aquileia, oke, since that was not entirely clear from the description of the function
20140322 17:14:18< aquileia> I'll update the pull request later today
20140322 17:14:25-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Ping timeout: 240 seconds]
20140322 17:15:07< mordante> anonymissimus, I'm not entirely sure what you mean but I've several names for my project root folder, including wesnoth, master, 1.10, wesnoth-old and never had any issues with cmake regarding it
20140322 17:15:32< aquileia> vultraz: {{Obsolete}} or [[Category:Obsolete]] ?
20140322 17:15:41< vultraz> former
20140322 17:15:48< vultraz> it includes the latter
20140322 17:15:52< aquileia> ah, ok
20140322 17:15:54< vultraz> as well
20140322 17:27:58-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 17:32:18-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140322 17:40:09< happygrue> shadowm: will you please cherry pick those two commits of mine over to 1.12?  I am still just not confident enough in what I'm doing to dare and I just ran out of free time again.  By the time I figure it out maybe they will be conflicting.
20140322 17:50:25< happygrue> Either that or just leave it out, that's fine too.  I just can't get a grip today and I'd rather not slip up playing with something I don't understand yet.
20140322 17:51:15< happygrue> Apparently, more coffee was needed this morning.  :(
20140322 17:51:35 * aquileia ducks
20140322 17:51:44< sachith500|2> bed time for me. good night everyone :D
20140322 17:51:46 * sachith500|2 waves
20140322 17:51:49< happygrue> o/
20140322 17:51:54-!- sachith500|2 [~kvirc@112.134.181.96] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20140322 17:51:59< aquileia> "be bold" - but don't let others hit you!
20140322 17:57:31< aquileia> anonymissimus: Like I said, you can revert whatever you want
20140322 18:00:57-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140322 18:06:18< AI0867> aquileia: ETA on fixing #126 ?
20140322 18:08:13-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140322 18:11:21< aquileia> AI0867: Sorry, but what is ETA?
20140322 18:12:13< gfgtdf> wesbot seen: iceiceice
20140322 18:12:19< gfgtdf> wesbot: seen iceiceice
20140322 18:12:19< wesbot> gfgtdf: The person with the nick iceiceice last spoke 14h 33m ago.  57m 54s ago they left with the message: Ping timeout: 240 seconds
20140322 18:12:37< mordante> aquileia, Estimated Time of Arrival
20140322 18:13:32-!- Kevin_Xi [~kevin@223.72.182.158] has left #wesnoth-dev []
20140322 18:14:04< aquileia> AI0867: Sorry, I'm at a phone call right now, about 1 hour
20140322 18:17:47< aquileia> mordante: thanks
20140322 18:18:31< aquileia> AI0867: Most of the time goes into moving to a separate namespace
20140322 18:19:44< aquileia> I'm nearly finished with that, but.... bug fixing can take quite some time, and I'll probably have created one or two with the renaming
20140322 18:21:41< AI0867> aquileia: I'd rather wait with the namespace stuff, I just want the compilation fix
20140322 18:22:15< aquileia> Well, I'm in midst the renaming, so... I can't really go back
20140322 18:23:40< aquileia> Most of the time I'll wait for Travis - my compiler accepts many things that error in strict mode
20140322 18:24:38< AI0867> your renaming will probably be really annoying to merge with my own changes =/
20140322 18:24:47< AI0867> which is why I'd like to get the existing PR in now
20140322 18:26:41< aquileia> Sorry for that...
20140322 18:28:31< aquileia> Hmm... but else I'd have that merge problem...
20140322 18:30:11< aquileia> I will try to do the renaming in a separate commit - then it's possible to revert them if there is a merge conflict
20140322 18:30:30-!- irker004 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20140322 18:30:46< _8680_> mordante: I did use `CHAR_BIT` originally, but Soliton told me to change it to `numeric_limits`.
20140322 18:31:05< AI0867> what are you planning to do anyway? All I've really heard so far is utf8::string
20140322 18:31:30< aquileia> utf8::string, utf8::erase, utf8::insert, ...
20140322 18:32:16< aquileia> mordante said the u8 prefix was ugly and nobody intervened, so I just changed it
20140322 18:32:20< shadowm> happygrue: Sure.
20140322 18:32:40< AI0867> hmm, okay
20140322 18:33:45< aquileia> I hope it doesn't clash too much... it doesn't affect that much code other than my own functions
20140322 18:34:50< AI0867> okay, I may have overestimated the scope of what you were renaming
20140322 18:38:37< shadowm> happygrue: Okay, for future reference, all but the "Miscellaneous and bug fixes" sections of the changelogs must go in alphabetical order.
20140322 18:41:03-!- irker840 [~irker@ai0867.net] has joined #wesnoth-dev
20140322 18:41:03< irker840> wesnoth: happygrue wesnoth:1.12 bb0a0c95ca62 / / (5 files in 2 dirs): Rebalanced three THoT scenarios. http://git.io/ReVDYA
20140322 18:41:03< irker840> wesnoth: happygrue wesnoth:1.12 fe86d5e429d8 / changelog players_changelog: Fixed whitespace error in changelogs http://git.io/-Gquzw
20140322 18:41:03< irker840> wesnoth: Ignacio R. Morelle wesnoth:1.12 a47ea57ae2d1 / players_changelog: Sort players changelog categories by alphabetical order http://git.io/fOZtaQ
20140322 18:41:05< irker840> wesnoth: Ignacio R. Morelle wesnoth:master e96bd0f1e102 / players_changelog: Sort players changelog categories by alphabetical order http://git.io/jgYOvg
20140322 18:42:38< happygrue> Ah right
20140322 18:43:03< happygrue> somehow I had missed the language catigory entirely
20140322 18:43:19< happygrue> thanks shadowm
20140322 18:48:41< mordante> _8680_, you asked and Soliton gave his preference I gave mine. Also curious why do we need to know or care about the number of bits in a value?
20140322 18:49:22-!- bagzie [~bag@85-76-86-147-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140322 18:49:35< _8680_> mordante: For generic functions that operate on those bits.
20140322 18:50:13< mordante> _8680_, obviously, but why do we need to know that in the context of Wesnoth
20140322 18:50:39< mordante> even on projects where I use bit manipulation I never used that
20140322 18:51:36< mordante> I used 12-bit sign magnitude numbers without caring about the number of bits, just used the proper uintxx_t
20140322 18:51:55< mordante> (or better said look at the CPU specs to define those types)
20140322 18:52:16< mordante> so that makes me really curious why we would need something like it in Wesnoth
20140322 18:53:11< shadowm> happygrue: np.
20140322 18:55:04< _8680_> The `count_leading_zeros` function I wrote takes a generic parameter `n` and passes it to (where available) `__builtin_clz` or a variant thereof. `__builtin_clz` takes an `unsigned int`, and its variants take `unsigned long int` and `unsigned long long int`. Thus, I use `bit_width(n)` to adjust the return value of `__builtin_clz` to account for the potential widening of `n` when `n` is passed to `__builtin_clz`.
20140322 18:56:47< _8680_> Otherwise, one might have `count_leading_zeros((std::uint8_t) 1) == 31`.
20140322 18:57:43-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has quit [Quit: Page closed]
20140322 19:02:08< mordante> which `count_leading_zeros` function and why would we need that in Wesnoth?
20140322 19:02:21-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140322 19:02:39< mordante> sorry if I sound harsh, but I really wonder why we need to care about such things in Wesnoth
20140322 19:03:56< aquileia> mordante: Perhaps it's best if I make the pull request together with the application - but currently I'm at the rest of the utf8 stuff, so I have no time to do it now
20140322 19:04:29< _8680_> I wrote it for aquileia to use in some function dealing with UTF-8.
20140322 19:04:46< mordante> aquileia, can you at least shed a light why it's needed, because before you pull it there is a `bug' in that code
20140322 19:05:22< demiurgos> hello, im here for the gsoc, and im a bit lost with the IRC
20140322 19:05:33< mordante> hi demiurgos 
20140322 19:06:08< mordante> also _8680_ if you wrote it for aquileia it would have been a good idea to write that in your pull request
20140322 19:06:13< aquileia> No bug, but AI0867, _8680_ an I agreed that it would improve readability and speed of byte_size_from_utf8_first
20140322 19:06:55< aquileia> at that moment I was not aware how much work it would be for _8680_ though... sorry again
20140322 19:07:13< mordante> because having a pull request without a usecase looks odd
20140322 19:07:32< mordante> aquileia, there is a `bug' in _8680_'s code
20140322 19:07:42< _8680_> Hm?
20140322 19:07:58< mordante> aquileia, _8680_ what's wrong with this line? »inline std::size_t bit_width(T x)«
20140322 19:08:42< happygrue> wesbot: bug 21759
20140322 19:08:42< wesbot> Bug #21759  Assigned to: None  Status: None  Priority: 5 - Normal
20140322 19:08:42< wesbot> Summary: timer refreshed too often when time runs out
20140322 19:08:42< wesbot> Original submission: When time runs out the timer bonus is given two times; ev
20140322 19:08:45< wesbot> en worse when there's still a random number coming from the server, timer bonus 
20140322 19:08:48< wesbot> URL: http://gna.org/bugs/?21759
20140322 19:08:55< _8680_> I don’t see what’s wrong with that line.
20140322 19:09:32< aquileia> mordante: You know I have no chance to see an error if there was one...
20140322 19:10:07< aquileia> I have no idea of template handling
20140322 19:10:13< mordante> aquileia, still you can learn by observing
20140322 19:10:26< aquileia> Yes, of course
20140322 19:10:31< mordante> and the issue is not with the template itself
20140322 19:10:43-!- inc_ [6ad4fcaa@gateway/web/freenode/ip.106.212.252.170] has joined #wesnoth-dev
20140322 19:11:01< _8680_> GCC is okay with it, Clang is okay with it even with -Weverything, and the unit tests pass. Is it a project-specific style issue that I’m not aware of?
20140322 19:11:23< mordante> but to make it concrete, what does this function do »inline std::size_t bit_width(unit x)«
20140322 19:13:28< _8680_> …It returns the bit width of `x`?
20140322 19:14:00-!- vorobeez [558e940c@gateway/web/freenode/ip.85.142.148.12] has quit [Quit: Page closed]
20140322 19:14:04< _8680_> I don’t know what you mean by `unit`, though.
20140322 19:14:56-!- inc_ [6ad4fcaa@gateway/web/freenode/ip.106.212.252.170] has quit [Client Quit]
20140322 19:15:53< mordante> the unit class in Wesnoth
20140322 19:16:04-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 19:17:21< _8680_> Oh. Okay, why would it be called with a `unit` instance as argument?
20140322 19:17:40< mordante> why not it is a template
20140322 19:17:52< mordante> so it can eat everything
20140322 19:18:07-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140322 19:18:14< happygrue> like a grue!
20140322 19:18:42< _8680_> Okay. It’s valid; is that a problem?
20140322 19:18:54< mordante> hmm now I know what a grue really is, is that dangerous?
20140322 19:19:22< _8680_> It’s just `sizeof` multiplied by the quantity of bits per byte.
20140322 19:19:40< mordante> _8680_, oh no, but what happens if I get curious and want to know the number of bits in an object of the type CVideo?
20140322 19:19:43< happygrue> only if the lighting is poor, mordante.  ;)
20140322 19:20:11< _8680_> The same thing that would happen if you wrote `sizeof(CVideo) * CHAR_BIT`.
20140322 19:20:14 * mordante turns on some candles, just in case the electricity fails
20140322 19:20:53-!- kex [~kex@89.205.75.19] has quit [Ping timeout: 252 seconds]
20140322 19:21:26-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140322 19:21:27< mordante> _8680_, what does this code do? »CVideo video; std::cerr << bit_width(video) << '\n';«
20140322 19:22:13< _8680_> It writes `sizeof(video) * CHAR_BIT` and a newline to stderr, without buffering.
20140322 19:22:25< _8680_> I guess?
20140322 19:22:51< _8680_> Unless CVideo does something odd in its copy constructor…?
20140322 19:22:51< mordante> nice guess, now try it
20140322 19:23:29< _8680_> Now I’m guessing that it does.
20140322 19:25:23< mordante> just ask your compiler what it thinks about it
20140322 19:25:37< gfgtdf> iceiceice: does loading replays from server works on current master ?
20140322 19:25:46< iceiceice> i'm not sure
20140322 19:26:06< gfgtdf> iceiceice: i think in 1.11.11 it didn't work
20140322 19:26:12< iceiceice> i just started trying to fix it on my branch, i am investigating further
20140322 19:26:35< iceiceice> so if you are concerned about the "missing mandatory child",
20140322 19:26:53< iceiceice> i ran a backtrace and found the line, although i'm not sure what to do about it
20140322 19:27:03< iceiceice> if you want a work around for further testing purposes i could probably make one
20140322 19:28:38< gfgtdf> iceiceice: hm i still don't really understand  teh difference beteween game_data don'game_state objects, 
20140322 19:30:18< iceiceice> yeah game_data hasn't been important for any code that i have written
20140322 19:31:07< AI0867> _8680_, aquileia: To see what's wrong, try comparing it with numeric_limits
20140322 19:32:08< demiurgos> happygrue: i made a proposal for the gsoc about random maps and i want to know about them
20140322 19:32:55< AI0867> that is, how do you ask it about the number of bits in an int?
20140322 19:33:40< mordante> _8680_, any luck compiling the code?
20140322 19:34:36< _8680_> If Travis were made to use `-std=c++11`, I could use `std::enable_if` with `std::is_trivially_copy_constructible`. I could also add a note in the Doxygen documentation that `x` will be copied.
20140322 19:35:08-!- Spoffy [~chatzilla@152.78.175.8] has quit [Read error: Operation timed out]
20140322 19:35:14< mordante> _8680_, but why do you need to copy the object?
20140322 19:36:58< aquileia> AI0867: It took longer than I expected, sorry. I hope it'll compile now...
20140322 19:37:12< _8680_> So what you’re getting at is that it should take a reference?
20140322 19:38:03< mordante> almost there, do you want to modify the object?
20140322 19:38:03< AI0867> nope
20140322 19:39:09< aquileia> ... const reference?
20140322 19:39:17-!- ykanarev [~ykanarev@78.81.70.234] has joined #wesnoth-dev
20140322 19:39:18< _8680_> aquileia: Yes.
20140322 19:39:22 * shadowm stares blankly.
20140322 19:39:27< mordante> see aquileia you know something about c++ ;-)
20140322 19:39:34< aquileia> not really
20140322 19:40:05< mordante> in general with templated code don't start to copy objects you never know what the user puts in it
20140322 19:40:08< aquileia> It's just luck that I got this point in the big mordante quiz ;)
20140322 19:40:08-!- kex [~kex@89.205.75.19] has joined #wesnoth-dev
20140322 19:40:37< mordante> of course you can use std::enable_if or boost::enable_if to restrict the types
20140322 19:40:57< mordante> or even static_assert or BOOST_STATIC_ASSERT
20140322 19:41:21< mordante> else the function can become a performance issue over the handcrafted code
20140322 19:41:49< mordante> since the function is inline I expect the compiler to avoid the copy, but still
20140322 19:44:22< AI0867> aquileia: ttext_::get_value() returns a temporary, not a reference
20140322 19:45:38-!- demiurgos_ [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140322 19:45:46-!- demiurgos_ [51258a4a@gateway/web/freenode/ip.81.37.138.74] has quit [Client Quit]
20140322 19:46:00-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has quit [Quit: Page closed]
20140322 19:46:29-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140322 19:47:50< aquileia> AI0867: But isn't it referenced by tmp?
20140322 19:48:51< AI0867> the version in master is just a copy
20140322 19:49:40< AI0867> you made a non-const reference
20140322 19:49:44< aquileia> and just now, clang errored on this...
20140322 19:49:58< AI0867> you can bind a const reference to a temporary, but not a non-const one
20140322 19:50:54< aquileia> If I make it a utf8_string instead of utf8_string&, will it work?
20140322 19:50:57< AI0867> yes
20140322 19:51:28< aquileia> ok, thanks
20140322 19:52:12< AI0867> the reason I looked at this in the first place was that it looked like you had a reference to a value that you were then setting. That can cause very annoying and subtle bugs if the object isn't designed properly
20140322 19:54:10-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Quit: leaving]
20140322 20:00:48-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit []
20140322 20:02:38< aquileia> AI0867: The pull request is updated accordingly, thanks.
20140322 20:03:11< aquileia> References are very handy, but... I obviously have to learn a lot more about how they are passed
20140322 20:04:06-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140322 20:04:16< happygrue> hi demiurgos, I'll go look for the proposal.  Do you have any specific questions right now?
20140322 20:04:18-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20140322 20:04:42-!- anonymissimus [~chatzilla@HSI-KBW-149-172-228-192.hsi13.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 12.0/20120423122928]]
20140322 20:05:14< demiurgos> happygrue: yeah, mordante told me to talk to you
20140322 20:05:16-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20140322 20:05:55< happygrue> yep, I found it
20140322 20:06:38< vultraz> hm, damn, RiftWalker isn't around
20140322 20:08:13< happygrue> demiurgos: so random maps could surely be improved.  I do not use them much and I haven't tested them recently, but one typical problem is that too much similar terrian is clumped together, so eg dwarves will just stay on the hills and everything else is too far away, so it is not possible to play a "real" game there
20140322 20:08:35< happygrue> have you played the WC (there are several types) add-ons?
20140322 20:08:52< aquileia> happygrue: Are you aware of YAMG?
20140322 20:09:03< happygrue> nope
20140322 20:09:58< happygrue> if it it is even better than yams then it sounds interesting!
20140322 20:11:35< aquileia> It's the ramdom map generator by pyrophorous that is already in mainline, but its interface isn't complete
20140322 20:12:07< aquileia> It works as a standalone tool, but as soon as you want to call it from inside Wesnoth... boom
20140322 20:12:14< happygrue> If I remember right, the WC mods and perhaps others some WML on top of a random map to make them more playable, is it similar?
20140322 20:12:43< demiurgos> happygrue: so thats what seemed to me, not only that is "irreal" maps, also all of them seems to similar
20140322 20:13:04< aquileia> It only generates random maps, it's a rival to the standard generator
20140322 20:13:24< happygrue> aquileia: I'm not seeing it on the development addon server, is it only for stable?
20140322 20:14:00< aquileia> happygrue: https://github.com/wesnoth/wesnoth/tree/master/src/generators/yamg
20140322 20:14:16< happygrue> ah
20140322 20:15:00< aquileia> it's part of the project, but doesn't work at the moment... I'm working on it, but extremely slowly
20140322 20:15:10< happygrue> okay so... how oh.
20140322 20:15:16< happygrue> was about to ask how I can make that work :D
20140322 20:16:58< aquileia> The standalone version works perfectly http://r.wesnoth.org/t36466
20140322 20:17:20< happygrue> afk for a bit.  demiurgos I'd look at WC for now, or you can look at what aquileia is mentioning.  Feel free to ask any questions and I'll get back to you.
20140322 20:18:45< happygrue> just because those are fairly good for a battle against the AI.  They are still not anywhere near balanced enough to really play a fair game against humans on, but that is a very very high bar
20140322 20:23:27< aquileia> happygrue, demiurgos: YAMG was designed as a standalone tool, pyrophorous began the conversion to a WML interface but probably lost interest. Its results are nice and it has one big advantage: If you like a map, you can produce variations of it
20140322 20:24:47< aquileia> If you like the overall design of a map but need a pond at the SW corner, YAMG will do it but the standard generator can't
20140322 20:25:12-!- nurupo is now known as \-\-\-\-\-\-\-\-
20140322 20:25:45-!- \-\-\-\-\-\-\-\- is now known as |-]-|-[-|-]-|-[-
20140322 20:26:14-!- |-]-|-[-|-]-|-[- is now known as |-]-|-[-]-|-[-|
20140322 20:26:15< demiurgos> aquileia: seems interesting
20140322 20:27:08-!- |-]-|-[-]-|-[-| is now known as nurupo
20140322 20:30:48< aquileia> AI0867: the linker found an error between the function signature of utf8::index in definition and declaration
20140322 20:30:56< aquileia> it's fixed now
20140322 20:31:14< aquileia> Why doesn't Visual Studio complain?
20140322 20:31:30 * aquileia asks himself
20140322 20:31:39< gfgtdf> wesbot: seen jamit
20140322 20:31:39< wesbot> gfgtdf: The person with the nick jamit last spoke 19d 19h ago.  19d 19h ago they left with the message: Quit: Leaving.
20140322 20:41:07< gfgtdf> in actions/move.cpp we currently have an argumtent is_replay, do you think i can remove that to make movements in replay behave just like normal movements ?
20140322 20:48:07< AI0867> aquileia: int and long are different types, but on various systems, they may be identical
20140322 20:48:19< AI0867> win32 and win64 are among those
20140322 20:48:47< aquileia> great... so I'll have to rely on travis for these things...
20140322 20:49:32< aquileia> But the last error (passing a temporary object) was accepted by my compiler as well
20140322 20:50:13< aquileia> So it's really unfortunate that Travis doesn't build incrementally...
20140322 20:50:26-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140322 20:57:07< mordante> I'm off bye
20140322 20:57:36-!- mordante [~mordante@wesnoth/developer/mordante] has quit [Quit: Leaving]
20140322 20:59:54< demiurgos> happygrue: so, i should  check src/generators
20140322 21:02:18< _8680_> aquileia: Why can’t you build with Clang yourself? It doesn’t work on Microsoft Windows, or is your computer not powerful enough?
20140322 21:02:54< aquileia> _8680_: It would work if I could pass the correct compiler flags, I think
20140322 21:03:16< _8680_> Why can’t you?
20140322 21:04:18< aquileia> I neither know where to set it nor what flags would be needed. I guess I'll figure it out later, but currently...
20140322 21:04:47< aquileia> currently I just use VC++ because it works
20140322 21:07:27-!- trewe [~trewe@2001:8a0:d13e:ee01:626c:66ff:fe92:9b7c] has joined #wesnoth-dev
20140322 21:12:43-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has quit [Ping timeout: 245 seconds]
20140322 21:19:42< aquileia> AI0867: Somehow, the old tests in test_serialization error...
20140322 21:21:39-!- Spoffy [~chatzilla@152.78.175.8] has joined #wesnoth-dev
20140322 21:22:14< aquileia> Argh... a simple copy-paste error!
20140322 21:30:25-!- trewe [~trewe@2001:8a0:d13e:ee01:626c:66ff:fe92:9b7c] has quit [Ping timeout: 240 seconds]
20140322 21:35:53< aquileia> AI0867: It errored during the unit test and this is fixed now... it has to work...
20140322 21:36:21< aquileia> would you merge it or do you want to wait for Travis in case I'm wrong?
20140322 21:40:20-!- neXyon [~neXyon@85-127-251-230.dynamic.xdsl-line.inode.at] has quit [Quit: bye]
20140322 21:41:31-!- neXyon [~neXyon@85-127-251-230.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20140322 21:50:32-!- adityapande [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has joined #wesnoth-dev
20140322 21:53:42-!- adityapande [ca4eaca2@gateway/web/freenode/ip.202.78.172.162] has quit [Client Quit]
20140322 22:03:46-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140322 22:11:22< aquileia> AI0867: I got an "error: illegal character encoding" in my testcase... are our compile tools UTF-8 ready?
20140322 22:11:47< aquileia> it errors ue to a 'ü' character which is absolutely fine!
20140322 22:11:58< aquileia> s/ue/due
20140322 22:18:46< aquileia> AI0867: GCC accepted my pull request, clang errored
20140322 22:19:00< aquileia> shadowm: Is there a policy in this case?
20140322 22:19:19< _8680_> Fix Clang’s error?
20140322 22:19:34-!- ykanarev [~ykanarev@78.81.70.234] has quit [Quit: Leaving]
20140322 22:21:56< aquileia> _8680_: Nice one
20140322 22:22:31< aquileia> but I guess Travis wouldn't update that fast
20140322 22:23:04< _8680_> I mean, fix the issue in BfW that causes Clang to emit an error message.
20140322 22:23:49< aquileia> There IS no issue, it's really an issue with clang
20140322 22:23:57< _8680_> Oh?
20140322 22:24:58< aquileia> clang errored because  used an UTF-8 character in a string to test my function
20140322 22:25:43< aquileia> It doesn't error on code, but on "ünicod€ check"
20140322 22:26:05< aquileia> the two symbols I mixed in are enough to stop it...
20140322 22:26:53< _8680_> I’m not sure that’s supposed to work in pre-C++11 C++.
20140322 22:27:13< _8680_> Are you sure the file is encoded as UTF-8?
20140322 22:27:32< aquileia> but GCC accepted it without C++11
20140322 22:28:01-!- trademark [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
20140322 22:28:03< _8680_> GCC accepting it doesn’t necessarily mean that it’s okay.
20140322 22:28:04< aquileia> It is
20140322 22:28:15< _8680_> As far as I know, BfW is intended to be compatible with both GCC and Clang, among other compilers.
20140322 22:28:30< _8680_> Thus, Clang needs to accept it as well.
20140322 22:28:37< aquileia> it's 'UTF-8 without BOM', the format Wesnoth wants
20140322 22:30:30< aquileia> it seems to be a missing feature / error of clang: GitHub, GCC, VC++, every other program accepts it
20140322 22:30:32< _8680_> I recommend using "\uNNNN" escape sequences.
20140322 22:30:48< _8680_> Though I’m not entirely sure those work pre-C++11 either.
20140322 22:30:52< _8680_> GitHub?
20140322 22:31:01< _8680_> GitHub compiles C++?
20140322 22:31:24< aquileia> No, I mean GitHub displays it correctly
20140322 22:31:50< _8680_> Oh, okay. That doesn’t mean that it’s allowed in C++, though.
20140322 22:32:39< aquileia> The mere presence of UTF-8 in a string?
20140322 22:33:18< _8680_> Of non-US-ASCII characters, yes.
20140322 22:33:30< _8680_> (It’s all UTF-8, if you’ve encoded the file as such.)
20140322 22:34:02< aquileia> yes, I know
20140322 22:34:18< aquileia> but what can I do then?
20140322 22:34:56< aquileia> How can I test whether Wesnoth copes with ü if clang doesn't ?
20140322 22:35:04< _8680_> Try "\uNNNN", where NNNN is the four-radix-16-digit number of the code-point.
20140322 22:37:16< aquileia> \u00FC is the code corresponing to ü, it seems
20140322 22:37:36< mattsc> shadowm: just to confirm that IRC, the list of known OS X issues is for the 1.12.0 announcement, right?  So I should test things in the 1.12 branch, not master?
20140322 22:37:44< _8680_> Alternatively, you could turn off that diagnostic check. Were I in charge, that would require a week of discussion and at least one vote by the technical direction committe, but I’m not (I’m not even a developer), so I expect you could just ask some developer to turn it off.
20140322 22:37:51< aquileia> but clang displayed it as <FC> in the error mesage, so...
20140322 22:38:24< _8680_> s/committe/&e/
20140322 22:38:59< _8680_> “FC” would be a valid (in many contexts) shortening of 00FC.
20140322 22:39:47< aquileia> So I _really_ think it's an issue with clang...
20140322 22:41:33< _8680_> Clang knows what you mean, it’s just being more strict than GCC about (I guess) making you follow the rules.
20140322 22:42:09< aquileia> The error message is http://pastebin.com/X7eCpqSg
20140322 22:42:25< _8680_> I believe that if Clang were told to use C++11 rather than whatever old verison BfW uses, it would compile it as you wish.
20140322 22:42:32< aquileia> and this really is an issue?
20140322 22:44:35< aquileia> I see two options...
20140322 22:44:44< aquileia> deactivate -Winvalid-source-encoding
20140322 22:44:58< aquileia> or load that string from a file
20140322 22:45:31< aquileia> but even the latter might fail... it's in the unit test after all, so it'll be read at compile time...
20140322 22:47:16< aquileia> oh, you already stated the first option... I forgot
20140322 22:47:21< aquileia> sorry
20140322 22:48:10< _8680_> (3) Use “\xC3\xBCnicod\xE2\x82\xAC” rather than “ünicod€” (and have “// "ünicod€" in UTF-8” above it for clarity).
20140322 22:49:08< _8680_> That’s “ünicod€”, escaped, as UTF-8.
20140322 22:49:38< aquileia> I'll try, but it'll take half an hour for clang to report back
20140322 22:52:00< aquileia> clang shouldn't see the comment as the precompiler erases it, right?
20140322 22:54:50< _8680_> Precompiler? If you mean the preprocessor, then I don’t think that the preprocessor deletes comments, but regardless, the compiler will ignore the text of comments — that’s why they exist.
20140322 22:56:15< aquileia> oh, yes...
20140322 22:56:41 * aquileia holes up after that embarassing mistake
20140322 22:57:23< aquileia> Thanks
20140322 23:00:09-!- gfgtdf_ [~chatzilla@d112102.adsl.hansenet.de] has joined #wesnoth-dev
20140322 23:00:25< shadowm> mattsc: Yes.
20140322 23:01:33< mattsc> shadowm: okay, thanks.  I’m back now, so I’ll (slowly) work through those issues.
20140322 23:02:25-!- gfgtdf [~chatzilla@f054135125.adsl.alicedsl.de] has quit [Ping timeout: 265 seconds]
20140322 23:02:33-!- gfgtdf_ is now known as gfgtdf
20140322 23:14:40< gfgtdf> EliDupree: online ?
20140322 23:14:50< EliDupree> yes
20140322 23:15:21< gfgtdf> EliDupree: do you think you can get a (c++) callstack from inside a lua function fron a unit ability filter ?
20140322 23:15:42< EliDupree> hmm
20140322 23:16:07< EliDupree> Well, if it segfaults then yes! ^_^
20140322 23:16:17< EliDupree> otherwise, I don't know a way to do it.
20140322 23:19:08< gfgtdf> EliDupree: maybe you can set a breakpoint at scriptin/lua.cpp line 4152 (run_filter function) or something like that
20140322 23:21:15< gfgtdf> s/scriptin/scripting
20140322 23:37:50< aquileia> _8680_: Thanks, it works!
20140322 23:38:25-!- ZubinKapadia [~shubham-p@14.139.122.114] has joined #wesnoth-dev
20140322 23:38:55< aquileia> AI0867: clang compiles now thanks to _8680_
20140322 23:39:02< aquileia> the PR is ready
20140322 23:39:24-!- cib0 [~cib@p20030067CE5CD701267703FFFEE75B84.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20140322 23:41:40< aquileia> GCC will pass as it already passed last time and the changes are only inside of strings
20140322 23:42:03< aquileia> so: PR 126 can finally be merged!
20140322 23:52:12-!- ZubinKapadia [~shubham-p@14.139.122.114] has quit [Ping timeout: 265 seconds]
20140322 23:54:47< aquileia> GCC passed like I predicted
--- Log closed Sun Mar 23 00:00:28 2014