--- Log opened Thu Oct 30 00:00:55 2014
20141030 00:22:23< iceiceice> gfgtdf: re "current player" and "name"
20141030 00:22:43< iceiceice> i know thunderstruck is thinking to add a third one that owuld be like "playerid" or something, and be used only by connect engine
20141030 00:23:38< iceiceice> i think it makes sense to have an "initial player name" (used only by mp connect, and mp server description generator), and a "current player name"
20141030 00:24:22< gfgtdf> iceiceice: hm y that sounds right but maybe we can make "current_player" and "name" one then ? Meaning removing one of them. Note that "name" is used by none of teh wesnoth dialogs
20141030 00:24:39< gfgtdf> iceiceice: its used by debug messages and it is exposed to wml
20141030 00:24:50< gfgtdf> iceiceice: btu all c++ dialogs use current_player""
20141030 00:25:33< iceiceice> yeah i think gettign rid of name woudl reduce ambiguity
20141030 00:25:43< iceiceice> maybe we can deprecate it
20141030 00:25:47< iceiceice> and remove eventually
20141030 00:26:04< iceiceice> or make it an alias for "current_player" or something, although that might be confusing
20141030 00:27:43< gfgtdf> iceiceice: i also noticed that "name" is sotred in carryover but appearently never read
20141030 00:28:42< iceiceice> another thing thats worht noting is,
20141030 00:28:52< iceiceice> i think "name" might be an OOS source
20141030 00:29:07< iceiceice> since when a player drops it gets renamed to "ai player X" for host but i dont think it does on clients
20141030 00:29:19< iceiceice> not htat its too important
20141030 00:29:45< iceiceice> anyways i dont think anyone can be using it in too important a way or their scenario woudln't work at all
20141030 00:30:25< gfgtdf> iceiceice: similar like "controller" is a OOS source, i thougth abotu makign [store_side]controller always returning "ai" or "human" (network-> human, ntework_ai -> ai)
20141030 00:31:18< iceiceice> maybe we can make it accessible only in a synced context?
20141030 00:31:23< gfgtdf> iceiceice: maybe than add another method for quering whether its remote or local
20141030 00:32:54< gfgtdf> iceiceice: synced context wouldn't save you. What you need is the local_choice function. but there are also uses  where  you want to call functions that give differnet results of sidderent clients
20141030 00:33:29< iceiceice> y
20141030 00:33:36< iceiceice> i wanted to ask you one like that,
20141030 00:33:50< iceiceice> do you know a way to ask lua which side is the viewing side?
20141030 00:34:26< iceiceice> like, basically i want to get the result of display::viewing_side()
20141030 00:35:25< iceiceice> there's all this infrastructure to make custom reports and such on the top bar and sidebar, 
20141030 00:35:37< gfgtdf> iceiceice: i think the viewing side is always the last local human side (maybe not in replays?) ? I currently dont know how to ask for viewing side but getting teh last local human side shouldnt be too hard.
20141030 00:35:38< iceiceice> but if you look in src/reports.cpp half of that stuff is like "return viewing_side().gold"
20141030 00:36:23< iceiceice> i see, i guess thats a good way tod o it, just update on turn start events
20141030 00:36:36< iceiceice> it will break though if control changes
20141030 00:37:39< iceiceice> i thought of a hilariously complicated hack to determine if the current player is also the viewing player,
20141030 00:37:49< iceiceice> by getting the text assigned to the countdown timer and checking what color code it has
20141030 00:37:56< iceiceice> but it only works if you are playing with a timer
20141030 00:38:30< gfgtdf> iceiceice: i just thougth about how making such a hack but i didnt remember the mp timer
20141030 00:39:13< iceiceice> ok, i will be back later
20141030 00:39:15-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141030 01:11:21-!- sachith500 [~kvirc@112.134.175.200] has joined #wesnoth-dev
20141030 01:16:16-!- sachith500 [~kvirc@112.134.175.200] has quit [Ping timeout: 256 seconds]
20141030 01:25:19-!- mjs-de [~mjs-de@f049161087.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141030 01:35:33-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141030 01:38:54-!- sachith500 [~kvirc@112.134.175.200] has joined #wesnoth-dev
20141030 01:49:28-!- Appleman1234 [~Appleman1@pool-173-74-87-52.dllstx.fios.verizon.net] has quit [Ping timeout: 255 seconds]
20141030 01:56:15-!- ancestral [~ancestral@174-20-180-46.mpls.qwest.net] has joined #wesnoth-dev
20141030 02:09:28-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141030 02:17:23-!- gfgtdf [~chatzilla@f054141011.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.91 [Firefox 32.0.3/20140923175406]]
20141030 02:25:01-!- knotwork [~markm@unaffiliated/knotwork] has quit [Ping timeout: 255 seconds]
20141030 02:26:14< mattsc> argh ...
20141030 02:30:12-!- sachith500 [~kvirc@112.134.175.200] has quit [Ping timeout: 264 seconds]
20141030 02:48:22-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141030 03:24:33-!- Ivanovic_ [~ivanovic@frnk-5f74e564.pool.mediaWays.net] has joined #wesnoth-dev
20141030 03:27:51-!- Ivanovic [~ivanovic@frnk-5f751ccf.pool.mediaWays.net] has quit [Ping timeout: 244 seconds]
20141030 03:28:27-!- Ivanovic_ is now known as Ivanovic
20141030 03:31:37< mattsc> sigh:  http://imgur.com/A2kiRqb
20141030 03:32:02< mattsc> Am I allowed to release 1.12.0 looking like that?
20141030 03:32:04< shadowm> That's why drugs are bad.
20141030 03:32:55< mattsc> You should see how that looks with the drugs …
20141030 03:33:30< mattsc> My guess is that this is caused by two separate versions of lippng being used simultaneously.
20141030 03:33:42< shadowm> Wait what how?
20141030 03:34:29< shadowm> You mean you compiled against a different libpng than the version loaded into the Wesnoth process at runtime?
20141030 03:34:54< shadowm> *compiled SDL_image against
20141030 03:35:28< mattsc> something along those lines, yes
20141030 03:36:26< mattsc> the good news, is, however, that I can now get Wesnoth to build and start on both OS X 10.9 and 10.5
20141030 03:36:45< mattsc> the rest are details (even if a PITA)
20141030 03:37:52< iceiceice> mattsc: when i compiled with different libpng version ijust got segfault on start of game
20141030 03:37:53< iceiceice> every time
20141030 03:37:53< shadowm> Do we have people on obsolete OS X versions still?
20141030 03:38:10< mattsc> shadowm: yes
20141030 03:38:31< shadowm> That's annoying.
20141030 03:38:32< iceiceice> it was consistent, its because when it tried to draw a button, it has to ask libpng to do something, and it goes into the library at an invalid location
20141030 03:38:46< mattsc> iceiceice: yes, I’ve had a lot of segfaults along the way.  So this is actually progress.
20141030 03:38:52< iceiceice> so ... if you dont crash then i guess you must have nearly identical png version?
20141030 03:39:14< shadowm> A bus error or segmentation fault is really the second-best possible outcome in these situations.
20141030 03:39:48< mattsc> iceiceice, shadowm: yes …
20141030 03:39:58< iceiceice> shadowm: did you see the thing about, china has banned its goverment offices from ever buying windows again, in retaliation for microsoft dropping support for XP?
20141030 03:40:07< mattsc> All of this is progress and it’s really not that hard to figure out the rest.  I just need to do the work.
20141030 03:40:32< shadowm> iceiceice: No, I remember reading they still relied on XP (for SOME reason) but didn't hear anything like that. Sounds stupid.
20141030 03:40:55< iceiceice> apparently 45% of the personal computers are on xp still as well in china
20141030 03:41:14< mattsc> The main problem is that I need to build the libs on the 10.5 machine; then compile Wesnoth on 10.9, then transfer things back to 10.5 to test.  It’s not a problem, just slow, with a lot of dead time in between.
20141030 03:41:29< mattsc> *not a fundamental problem
20141030 03:41:56< iceiceice> so wait, macports didnt' work out?
20141030 03:41:58< shadowm> It sounds stupid because I imagine Microsoft made improvements to i18n in later versions.
20141030 03:42:13< iceiceice> maybe but they probably broke everything else
20141030 03:42:19< mattsc> iceiceice: yes it did; me doing shortcuts around just doing everything with Macports did not ...
20141030 03:42:21< shadowm> Plus they had *years* to achieve some deal with MS.
20141030 03:42:36< mattsc> plus me actually having to do work during the day is a major annoyance  ;)
20141030 03:42:47< iceiceice> yeah i hear you :)
20141030 03:43:41< iceiceice> mattsc: i actually have no idea how to properly compile sdl_image right now,
20141030 03:43:58< iceiceice> the only thing i've managed to do is statically link lib png inside it, which leavaes everything working perfectly
20141030 03:44:17< ancestral> mattsc: How many people are using 10.5?
20141030 03:44:20< iceiceice> its just that i dont think homebrew will approve of this, also it requires me to edit other packages besides the wesnoth package
20141030 03:44:24< ancestral> How many people are using non-Intel?
20141030 03:44:48< ancestral> It’s been 8 years since Macs went full over to Intel
20141030 03:45:30< iceiceice> ancestral: i think old computers become like old cars,
20141030 03:45:40< ancestral> iceiceice: I don’t think so
20141030 03:45:41< iceiceice> people will just keep running them, until they cant
20141030 03:45:45< ancestral> They don’t become classic
20141030 03:45:49< iceiceice> no thats not what i meant
20141030 03:45:52< ancestral> People don’t spend tons of money fixing up old computers
20141030 03:46:16< ancestral> iceiceice: That may be true, but those same people typically aren’t playing them in order to play turn-based strategy games
20141030 03:46:32< ancestral> If they play any games, it’s solitaire or free online games
20141030 03:46:41< iceiceice> i mean that all the school buses in america when i was a kid, became the public transportation in guatemala when i was a backpacker... and they will keep running and getting shuffled from country to country until ... idk engine block cracks? severe structural failure?
20141030 03:46:50< ancestral> In fact, most of those same people probably have more powerful phones and game on them instead
20141030 03:47:19< ancestral> mattsc: I’m very curious who, if anyone, is running a Mac with10.5
20141030 03:48:02< iceiceice> ancestral: i think any place where consumer culture is not so strong, its like if it aint broke why replace it
20141030 03:48:09< ancestral> A $500 Mac mini is more powerful than any Mac running 10.5
20141030 03:48:31< ancestral> If mattsc has to spend all this time supporting 2 players, is it worth it?
20141030 03:48:48< ancestral> What if these same people upgraded their computers recently?
20141030 03:48:52< iceiceice> its true, we could just tell them "get macports"
20141030 03:49:07< iceiceice> and then let macports and them sort out the drivers
20141030 03:49:10< shadowm> A car will get you to places regardless of its age as long as it works.
20141030 03:49:24< iceiceice> *libraries
20141030 03:49:26< shadowm> A computer might be unable to run newer software you suddenly need.
20141030 03:49:42< iceiceice> you can always switch to linux :)
20141030 03:50:04< shadowm> I don't expect Linux to be great on PPC.
20141030 03:50:12< iceiceice> yeah i really have no idea about that
20141030 03:50:17< ancestral> Making a PowerPC binary is not worth the time
20141030 03:50:28< ancestral> brb
20141030 03:50:28< shadowm> I also don't know if there are any popular distributions built for PPC besides Debian (certainly not Ubuntu).
20141030 03:51:50< shadowm> Why I don't expect Linux to be great on PPC: because it's far too easy to write C or C++ software that doesn't account for endianess differences between platforms.
20141030 03:54:20< shadowm> *Linux-based OSes
20141030 03:54:53< shadowm> Just to clarify, because the kernel supports a number of architectures most garden-variety coders have barely heard of.
20141030 03:55:12< mattsc> ancestral: AFAIC Wesnoth has not been working on PPC for quite some time.
20141030 03:55:24-!- ancestral [~ancestral@174-20-180-46.mpls.qwest.net] has quit [Ping timeout: 264 seconds]
20141030 03:55:35< mattsc> However, there are several people who still use 10.5.
20141030 03:55:50-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 03:55:51< mattsc> My main playtester on 10.5 is one of the Forum Regulars, but I know that there are others.
20141030 03:56:21< mattsc> I’d hazard guess that there are more than 2; whether there are more than 5, I don’t know ...
20141030 03:57:03< iceiceice> btw: ancestral, did you see my link before?
20141030 03:57:13< ancestral> I’ll check the logs
20141030 03:57:25< ancestral> iceiceice: Last couple of minutes?
20141030 03:57:28< iceiceice> i'll give you a fresh one, its changed a bit
20141030 03:57:43< vultraz> I thought we dropped PPC support years ago
20141030 03:57:44< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/9627acdb8d73859e41be4ef39dd4be9f61211506/wesnoth-new.rb
20141030 03:57:56< mattsc> iceiceice: I got sdl_image to build without problems with Macports.  I just haven’t updated all the other libs yet.
20141030 03:58:07< ancestral> I see it
20141030 03:58:28< ancestral> vultraz: We ought to
20141030 03:59:16< vultraz> ancestral: personally, I think we should also stop supporting Leopard. It seems it's becoming a PITA for just 5 people
20141030 03:59:18< mattsc> vultraz: yes; some of the libraries I am currently replacing are still i386 and ppc fat libraries, but the rest of ppc support has not been around in quite some time
20141030 03:59:20< iceiceice> ancestral: if you could tell me if "brew install --head --with-ccache --verbose --debug [url]" works,
20141030 03:59:23< iceiceice> i would appreciate
20141030 03:59:27< ancestral> vultraz: Yes
20141030 03:59:38< iceiceice> so right now its pointed at my fork instead of main wesntoh master,
20141030 03:59:46< iceiceice> i find that if i push updates to my fork, and run this,
20141030 03:59:49< ancestral> Especially because all Intel 10.5 users can/could have/did upgrade to 10.6
20141030 03:59:52< iceiceice> it fetches and rebuilds master in < 3 minutes
20141030 03:59:53< iceiceice> consistenyl
20141030 04:00:11< ancestral> Every Intel 10.5 machine can be upgraded to 10.6
20141030 04:00:28< ancestral> iceiceice: I’ll give it a shot now
20141030 04:02:02< iceiceice> is tiger before or after leopard?
20141030 04:02:39< vultraz> iceiceice: tiger is 10.4
20141030 04:02:47< vultraz> iceiceice: Leopard is 10.5
20141030 04:03:03< iceiceice> this is crazy then:
20141030 04:03:04< iceiceice> https://github.com/mistydemeo/tigerbrew
20141030 04:03:15< iceiceice> experimental fork of homebrew (1-2 years old?)
20141030 04:03:29< iceiceice> for OS and platform that is ???
20141030 04:04:12-!- Appleman1234 [~Appleman1@pool-173-74-87-52.dllstx.fios.verizon.net] has joined #wesnoth-dev
20141030 04:04:51< mattsc> Well, my take is we continue to support 10.5 through 1.12 and drop it for 1.13 …
20141030 04:04:54< ancestral> iceiceice: There was a brief period of time where businesses invested in PowerMac G5s and didn’t have a Mac alternative to PPC until very late ’06 when the Mac Pro was released
20141030 04:04:58< mattsc> Anyways, I need to be off.  TTYL.
20141030 04:05:19< ancestral> mattsc: Cool, appreciate your efforts
20141030 04:05:35< mattsc> ancestral: thanks; and I yours.
20141030 04:06:20< ancestral> vultraz: Almost everyone I know mispronounces the ‘X’ in OS X
20141030 04:06:30< ancestral> (Hint: It’s a roman numeral)
20141030 04:06:48< vultraz> ancestral: guilty 
20141030 04:06:51< iceiceice> wait you mean its not OS X-treme ??
20141030 04:06:58< ancestral>  /facepalm
20141030 04:06:59< iceiceice> D:
20141030 04:07:10< ancestral> Hehehe
20141030 04:07:18< shadowm> I do know it's a roman numeral but I still pronounce it as if it were a regular x.
20141030 04:07:33< vultraz> OS Ten just seems trite
20141030 04:07:39< ancestral> And yes, that means Yosemite is Oh Ess Ten, Ten Point Ten
20141030 04:07:47< shadowm> If it wasn't predeced by an acronym I'd be less compelled to mispronounce it.
20141030 04:07:48< ancestral> (As ridiculous as that might sound…)
20141030 04:08:18< ancestral> Not sure what will happen if iOS reaches 10
20141030 04:08:25< ancestral> The universe may explode
20141030 04:08:47< shadowm> They'll pull a Microsoft and call it iOS 50.
20141030 04:09:04< iceiceice> they have to call it OS 2000
20141030 04:09:12< vultraz> iOS X
20141030 04:09:23< shadowm> Or do it better and unify both OSes.
20141030 04:09:30< iceiceice> did anyone watch that film "her" ?
20141030 04:09:33< shadowm> Then it'll be OS X something something Phone Edition.
20141030 04:09:42< ancestral> Start using years? Kinda like Adobe did briefly (I owned a copy of Adobe Illustrator’88 back in the day)
20141030 04:09:54< vultraz> One thing Microsoft did right is having one OS for tablets and laptops/desktops
20141030 04:10:05< shadowm> They did it wrong.
20141030 04:10:05< ancestral> That’s very debatable
20141030 04:10:08< iceiceice> vultraz: i actually really don't like that idea
20141030 04:10:11< shadowm> Seriously.
20141030 04:10:34< iceiceice> i also really didn't like unity
20141030 04:10:43< vultraz> ancestral: even worse is the fact that iPhone and iPad apps are not compatible
20141030 04:10:48< iceiceice> iirc that was a large factor in switching to mint
20141030 04:10:51< iceiceice> for me
20141030 04:10:52< shadowm> The idea isn't particularly sound from an engineering standpoint, and the results in the UX side were clearly a disaster.
20141030 04:10:54< vultraz> they need specific versions for each
20141030 04:10:55< iceiceice> and thats kind of the same issue
20141030 04:11:25< shadowm> And now MS is scrambling to amend their mistake without deunifying the OS.
20141030 04:11:41< ancestral> vultraz: Imagine trying to touch a close button in a window with your finger… or trying to emulate pinch and zoom without a touch screen or touch pad
20141030 04:11:43< shadowm> So their engineering cost will keep increasing but hopefully at least users will be happy.
20141030 04:11:47< shadowm> *costs
20141030 04:11:50< ancestral> These things have very different input methods
20141030 04:12:17< iceiceice> random question: can android phones run Xmonad?
20141030 04:12:49< ancestral> iceiceice: For fun, I mapped your gist here: http://wesnoth.io/brew/head
20141030 04:13:11< vultraz> ancestral: the windows 8 UI is pretty good for both touch and mouse
20141030 04:13:59< shadowm> No.
20141030 04:14:02< ancestral> I don’t have any experience with 8 myself
20141030 04:14:32< shadowm> I mean, I wouldn't rule out that maybe Windows 8 works better with the younger regeneration, but.
20141030 04:14:37< shadowm> *generation
20141030 04:14:38< iceiceice> i remember being really frustrated when i got a new computer
20141030 04:14:42< iceiceice> and it had windows 8
20141030 04:14:49< iceiceice> there was some damn thing that kept closing my windows
20141030 04:14:53< iceiceice> and i couldnt turn it off
20141030 04:15:05< iceiceice> i ended up googling for hacks to disable
20141030 04:15:08< iceiceice> i cant remember what it is now
20141030 04:15:42< iceiceice> i remember now,
20141030 04:15:48< iceiceice> it was like some crazy gesture system
20141030 04:16:00< iceiceice> if i moved my mouse left then right then it swiped everthing off the screen and went to tablet mode
20141030 04:16:06< iceiceice> it drove me crazy
20141030 04:16:33< iceiceice> btw, does anyone know if you still have to "defragment your hard drive" on windows?
20141030 04:17:13< iceiceice> wow they still haven't fixed that???
20141030 04:17:16< iceiceice> wow...
20141030 04:17:17< shadowm> With an HDD, sure, but there's a scheduled task by default.
20141030 04:17:41< ancestral> OS X defrags on the fly
20141030 04:18:28< shadowm> And with an SSD defragmentation is irrelevant and even counterproductive.
20141030 04:18:42< vultraz> iceiceice: the first time I tried 8, I was extremely frustrated for about two days. Now I realize it's generally superior to 7
20141030 04:19:05< ancestral> vultraz: By your logic, 10 is 2 times superior to 8!
20141030 04:19:10< iceiceice> whatever, i put mint on this machine, and now the windows partition doesn't even work anymore
20141030 04:19:17< ancestral> :-P
20141030 04:19:23< vultraz> ancestral: not necessarily :P
20141030 04:19:39< vultraz> I don't intend to use 10 until it comes out fully
20141030 04:19:51< vultraz> But I'm sure shadowm has the dev preview tucked away on a VM somewhere
20141030 04:19:59< ancestral> iceiceice: It’s still going
20141030 04:20:08< shadowm> I don't, but I had planned to get it next month or so.
20141030 04:20:14< iceiceice> ancestral: the first build should be slow,
20141030 04:20:21< shadowm> Depends on how much Microsoft will pester me for feedback.
20141030 04:20:25< iceiceice> it has to copy the repo at depth 1, and then build everything
20141030 04:20:28< iceiceice> but it will all be in ccache
20141030 04:20:40< iceiceice> so even when the build folder gets deleted,
20141030 04:20:47< iceiceice> any files that didn't change will build instantly the next time
20141030 04:20:59< ancestral> Pulling down ~1MiB/s here, not too bad
20141030 04:21:10< shadowm> vultraz: Unfortunately for you, everyone else and their dogs sees 8 as this generation's Millenium Edition.
20141030 04:21:42< iceiceice> i really should put a chunk of retirement money in negative microsoft stock :p
20141030 04:21:46< vultraz> shadowm: not everyone. I've seen many favorable comments that agree withme
20141030 04:21:51< shadowm> vultraz: Everyone.
20141030 04:21:52< ancestral> (Amended my shortcut, needs to have the file name in the url)
20141030 04:22:22< ancestral> iceiceice: I would have invested in negative RIMM years ago
20141030 04:22:37< ancestral> Fortunately, I do have a good chunk of AAPL
20141030 04:23:07< vultraz> Oh?
20141030 04:23:53< iceiceice> ancestral = steve jobbs
20141030 04:23:56< iceiceice> = elvis
20141030 04:24:01< ancestral> vultraz, how many shares of Wesnoth stock would you buy if there was an IPO? :-P
20141030 04:24:32< shadowm> All of them ofc.
20141030 04:24:35< vultraz> ancestral: hm...depends
20141030 04:24:40< iceiceice> ancestral: it depends what the price is clearly :p
20141030 04:24:42< ancestral> Take over the board of directors
20141030 04:24:59< ancestral> Proabably wouldn’t be worth much
20141030 04:25:03< iceiceice> wow
20141030 04:25:09< iceiceice> we crossed 100k commits recently i think
20141030 04:25:14< ancestral> Not exactly making monies here
20141030 04:25:25< vultraz> Let's put dugi in charge and set up an IPO 
20141030 04:25:39< shadowm> I thought you'd be all for becoming my boss.
20141030 04:25:49< vultraz> (joking)
20141030 04:25:55< ancestral> iceiceice: Okay I got an error
20141030 04:26:09< iceiceice> ok what happened?
20141030 04:26:11< vultraz> I'd totally buy a lot if there was financial future in this project
20141030 04:26:18< vultraz> Right now there isn't
20141030 04:26:27< iceiceice> vultraz: we have 11 million dollars of future:
20141030 04:26:27< iceiceice> https://www.openhub.net/p/wesnoth/estimated_cost
20141030 04:26:27< shadowm> There never was.
20141030 04:26:38< ancestral> “wesnoth, cutter and exploder cannot be built”
20141030 04:26:51< iceiceice> what test failed?
20141030 04:26:57< ancestral> https://paste.ee/p/v0zcq
20141030 04:27:15< ancestral> Unit tests?
20141030 04:27:18< vultraz> iceiceice: p_P
20141030 04:27:26< ancestral> Oh yeah the xcb-shm
20141030 04:27:37< ancestral> We ran into this before, right?
20141030 04:27:50< iceiceice> yeah i cant remember what happened
20141030 04:27:51< vultraz> Is $55k/year good?
20141030 04:28:16< iceiceice> vultraz: a significant fraction of code base is WML, if someone is getting paid 55k / year to write wml i consider that pretty good :)
20141030 04:28:50< iceiceice> its not exactly a highly marketable skill
20141030 04:28:51< ancestral> vultraz: I’ll take that pay
20141030 04:28:59< ancestral> It depends where you live of course
20141030 04:29:28< ancestral> Higher costs of living in Silicon Valley and large cities
20141030 04:30:24< ancestral> vultraz: Not to be confused with http://en.wikipedia.org/wiki/Wireless_Markup_Language
20141030 04:30:31< ancestral> or http://en.wikipedia.org/wiki/Website_Meta_Language
20141030 04:31:43< ancestral> iceiceice: Found that thing
20141030 04:31:52< iceiceice> ok you just beat me
20141030 04:31:52< ancestral> http://pastebin.com/C4WhsvTH
20141030 04:31:55< ancestral> :)
20141030 04:32:19< ancestral> iceiceice: Could/should that be added to the .rb file?
20141030 04:32:55< iceiceice> maybe
20141030 04:33:13< iceiceice> its a little hard cause i guess you dont want to add ti the second time also
20141030 04:33:37< iceiceice> ideally it should be in their installer script
20141030 04:34:15-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 255 seconds]
20141030 04:34:29< iceiceice> hmmmm
20141030 04:34:38< iceiceice> im reading the tracker actually,
20141030 04:34:50< iceiceice> maybe what it means is we are suppsoed to disable x-support for os x
20141030 04:35:15< shadowm> We had X11 support enabled on OS X builds?
20141030 04:35:31< iceiceice> no,
20141030 04:35:41< iceiceice> i just didnt add instructions to disbale it to the homebrew installer
20141030 04:35:48< iceiceice> when it is building SDL or whatever i guess
20141030 04:35:53< iceiceice> *cairo
20141030 04:35:54-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141030 04:35:56< shadowm> Yeah, I think we dropped X11 support on OS X ages ago?
20141030 04:36:23< iceiceice> ancestral: i'm going to change the script to account for this
20141030 04:40:25< iceiceice> ancestral: try this one: https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/a09c792e516f7bfdba846b06e6838c36b71f32fc/wesnoth-new.rb
20141030 04:45:31-!- Sulfur [~Miranda@p5B3279A8.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141030 04:51:32-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 245 seconds]
20141030 04:53:16-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141030 04:57:23-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Bye for now]
20141030 05:12:58-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 05:18:53-!- ancestral_ [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 05:19:20< ancestral_> shadowm: Dropping X11 on Mac long ago — must have
20141030 05:20:04< ancestral_> (I’d have gotten requests to install XQuartz if it relied on it)
20141030 05:20:32< ancestral_> iceiceice: Finished, testing to make sure it runs. I haven’t tried the new script, I can if you want
20141030 05:20:47-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Ping timeout: 265 seconds]
20141030 05:20:47-!- ancestral_ is now known as ancestral
20141030 05:20:54< ancestral> Yay music and sound!
20141030 05:21:00< iceiceice> :)
20141030 05:21:27< ancestral> shadowm: I can see some carefully bolded text in the hints dialog box. Nice touch!
20141030 05:22:05< ancestral> Is there supposed to be a close box for the hints or not really?
20141030 05:23:06< ancestral> I suppose closing the hints isn’t really that effective… “let me see just a little more of a… background image”
20141030 05:23:26-!- Gallaecio [~quassel@84.120.180.152.dyn.user.ono.com] has joined #wesnoth-dev
20141030 05:23:27< ancestral> Trying the tutorial
20141030 05:25:20-!- kex [~kex@78.157.29.160] has quit [Remote host closed the connection]
20141030 05:27:32< ancestral> No crashes. Looks good.
20141030 05:28:10< ancestral> `Data directory: /usr/local/Cellar/wesnoth-new/HEAD/share/wesnoth`
20141030 05:28:23< ancestral> `User configuration directory: /Users/ancestral/.wesnoth1.13`
20141030 05:28:33< ancestral> `User data directory: /Users/martinproud/.wesnoth1.13`
20141030 05:29:11< ancestral>  /Users/ancestral/.wesnoth1.13
20141030 05:29:17< ancestral> `Cache directory: /Users/ancestral/.wesnoth1.13/cache`
20141030 05:29:36< iceiceice> yeah so one thing  i didn't figure out, 
20141030 05:29:44< ancestral> Curious
20141030 05:29:47< iceiceice> i noticed that at the very beginning of scons run it says "error: no repo"
20141030 05:29:48< ancestral> `20141030 00:25:50 error display: could not open image 'buttons/misc/orb-active.png’`
20141030 05:29:59< iceiceice> yeah i think thats just messed up right now
20141030 05:30:03< iceiceice> i get that also on linux i think
20141030 05:30:21< ancestral> iceiceice: Lemme check
20141030 05:30:22< iceiceice> whenever you open the preferences dialog
20141030 05:30:48< ancestral> Let me check about the repo message
20141030 05:31:01< ancestral> That might have occured before I did the export
20141030 05:31:46< iceiceice> its happening for me right now
20141030 05:31:52< ancestral> Oh
20141030 05:32:17< ancestral> I’ll run it again since you updated the script, right?
20141030 05:35:04< iceiceice> yeah sure
20141030 05:35:13< iceiceice> if you exported that thing though then it wont matter
20141030 05:35:23< iceiceice> heres current script:
20141030 05:35:24< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/23a5320a75e0ee4b7059e2632d6b408e43620c0c/wesnoth-new.rb
20141030 05:35:46< iceiceice> maybe if you close the terminal and open again i guess it will reset the export environment?
20141030 05:35:54< iceiceice> i'm not sure 
20141030 05:36:56< ancestral> It would
20141030 05:38:44< ancestral> iceiceice: Heh, to uninstall I had to brew rm thelonglink
20141030 05:39:00< iceiceice> oh
20141030 05:39:05< iceiceice> i think you can brew uninstall wesnoth-new
20141030 05:39:17< ancestral> Oh, probably
20141030 05:39:19< iceiceice> that works for me anyways
20141030 05:40:32-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 265 seconds]
20141030 05:41:11< ancestral> `brew install --HEAD --with-ccache --verbose--debug http://wesnoth.io/brew/head/wesnoth-new.rb`
20141030 05:41:23< ancestral> it’s downloading xz, cairo, pango, sdl
20141030 05:41:31< ancestral> deps
20141030 05:41:51< ancestral> Interesting, xz-5.0.7.yosemite.bottle.tar.gz
20141030 05:41:52< iceiceice> im trying to make sure that it builds cairo and pango without x11
20141030 05:42:02< ancestral> Good
20141030 05:42:07< iceiceice> bottle is a binary package i think
20141030 05:42:24< ancestral> I have XQuartz installed but it’s optional and third-party
20141030 05:42:42< ancestral> (Continuation of the X11 project for Darwin/OS X)
20141030 05:43:37< ancestral> So basically, the only thing missing is actually the packaging into an app, which you said there is info on in sconstruct documentation, which I can certainly look at
20141030 05:44:07< iceiceice> is it installing basically correctly though?
20141030 05:44:15< iceiceice> oh, the other thing,
20141030 05:44:23< iceiceice> i couldn't figure out how to build wesnothd without getting errors
20141030 05:44:27< ancestral> It did install correctly (to /usr/local/bin/)
20141030 05:44:37< iceiceice> ok that's good
20141030 05:44:50< iceiceice> homebrew is set up so that its supposed to do soem of that work,
20141030 05:44:53< ancestral> Is wesnothd a different build script, or part of this one?
20141030 05:45:00< ancestral> iceiceice: Got an error
20141030 05:45:03< iceiceice> wesnothd is an entirely differente executable
20141030 05:45:17< iceiceice> its for if you want to LAN games
20141030 05:45:25< ancestral> Right, for hosting
20141030 05:45:44< iceiceice> yeah, i guess you could make like an unofficial mirror also
20141030 05:46:14< ancestral> iceiceice: https://paste.ee/p/KdhbP
20141030 05:46:58< iceiceice> hmm
20141030 05:47:08< iceiceice> can you try again, but put a space between --verbose and --debug?
20141030 05:47:22< ancestral> Man you’re good
20141030 05:48:02< ancestral> Didn’t realize I mistyped it, the linebreak was between the two
20141030 05:48:17< ancestral> Hmm, pango not building
20141030 05:48:51< ancestral> brew rm pango
20141030 05:49:19< ancestral> Maybe because of my botched thing before
20141030 05:49:49< iceiceice> it should give some kind of compiler error or something
20141030 05:49:57< iceiceice> or a ruby error caused by me somehow
20141030 05:49:59< ancestral> bad interpreter
20141030 05:50:17< iceiceice> crap
20141030 05:50:40< ancestral> `/bin/sh: /usr/local/Cellar/gobject-introspection/1.42.0/bin/g-ir-scanner: /usr/local/opt  \n python/bin/python: bad interpreter: No such file or directory  \n
20141030 05:50:40< ancestral> make[4]: *** [Pango-1.0.gir] Error 126`
20141030 05:51:57< ancestral> Going to brew update
20141030 05:52:03< iceiceice> y good idea
20141030 05:52:29< iceiceice> the linux brew guy said that gobject-introspection didn't build for him either
20141030 05:52:39< iceiceice> i never had a problem though
20141030 05:52:51< ancestral> ==> Installing dependencies for wesnoth-new: readline, sqlite, gdbm, openssl, python, pango, sdl
20141030 05:53:57< ancestral> Sucess I think!
20141030 05:54:20< iceiceice> wow it needs sqlite?
20141030 05:54:23< iceiceice> thats wierd
20141030 05:54:28< iceiceice> i must have done something wrong somehow
20141030 05:54:41< iceiceice> but hopefully it all builds anyways :)
20141030 05:55:26< ancestral> What’s weird is I didn’t think that was dl’d the first time
20141030 05:56:00< ancestral> Or maybe you edited your gist recently
20141030 05:56:25< ancestral> iceiceice: Okay so I wonder
20141030 05:56:50< ancestral> Mayyyybe the wesnoth binary I ran wasn’t from this script
20141030 05:56:59< ancestral> Nah, it must have been
20141030 05:57:08< iceiceice> did ti have messed up title bar at the top?
20141030 05:57:11< iceiceice> i.e. with no version info?
20141030 05:57:27< iceiceice> if so it had to be mine
20141030 05:57:29< ancestral> -bash: wesnoth: command not found
20141030 05:57:41< ancestral> 🍺  /usr/local/Cellar/wesnoth-new/HEAD: 5 files, 652K, built in 17 seconds
20141030 05:57:58< ancestral> I tried wesnoth-new also
20141030 05:58:51< iceiceice> maybe the gist link became stale or something?
20141030 05:58:52< ancestral> Well, I had been running a wesnoth in /usr/local/bin, I just don’t know from when and how
20141030 05:59:05< iceiceice> does that happen?
20141030 05:59:12< ancestral> iceiceice: Let me try not using my redirect
20141030 05:59:21< ancestral> Trying to be too clever
20141030 05:59:41< iceiceice> heres script again: https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/23a5320a75e0ee4b7059e2632d6b408e43620c0c/wesnoth-new.rb
20141030 06:00:01< iceiceice> i dont think i changed anything substantial, maybe some spacing
20141030 06:00:10< ancestral> https://gist.githubusercontent.com/cbeck88/f5898a5cbdfe99a8d427/raw/23a5320a75e0ee4b7059e2632d6b408e43620c0c/wesnoth-new.rb
20141030 06:00:43< ancestral> Different domains, but looks like they’re both ok
20141030 06:00:56< ancestral> Warning: wesnoth-new-HEAD already installed
20141030 06:00:57< ancestral> Hmmm
20141030 06:01:26< iceiceice> hm is `brew rm ...` the same as `brew uninstall ..` ?
20141030 06:01:33< iceiceice> i have been using `brew uninstall ...`
20141030 06:01:38< ancestral> You might be right
20141030 06:01:51< ancestral> I just did `brew uninstall wesnoth-new
20141030 06:01:54< ancestral> Alright
20141030 06:02:08< iceiceice> man page says they are the same
20141030 06:02:09< ancestral> Rebuilding
20141030 06:03:52-!- Sulfur [~Miranda@p5B3279A8.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20141030 06:04:12< ancestral> `git checkout-index -a -f --prefix=/private/tmp/wesnoth-new-CebSaz/`
20141030 06:04:49< ancestral> Yeah it’s building into my temp directory and then deleting it
20141030 06:05:06< iceiceice> damn
20141030 06:05:58< iceiceice> so,
20141030 06:06:06< iceiceice> i ran brew audit on the formula, and it told me to change a line
20141030 06:06:11< iceiceice> but it hink maybe that thing was lies
20141030 06:06:19< iceiceice> i'm going to change it back now i guess
20141030 06:06:46< iceiceice> it wanted me to change `system "scons", *args` to "scons *args"
20141030 06:07:01< iceiceice> maybe i read the instructions wrong
20141030 06:07:13< iceiceice> oh i think i know what it might be
20141030 06:07:43< shadowm> ancestral: What hints dialog?
20141030 06:08:06< ancestral> I may be dreaming…
20141030 06:08:07< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/2cbcb3d4312435d94a70c121c445a55a4b18ba72/wesnoth-new.rb
20141030 06:08:26< ancestral> shadowm: The hints show up in a box. Somehow I thought there was a way to dismiss them
20141030 06:08:38< shadowm> What hints box?
20141030 06:09:06< ancestral> When you open Wesnoth and are at the main menu
20141030 06:09:30< shadowm> Ah, the tips of the day.
20141030 06:09:35< ancestral> That thing, right
20141030 06:13:51< ancestral> Did there use to be a way to dismiss them, or did I just imagine it? (I don’t care, just curious)
20141030 06:14:31< ancestral> iceiceice: Good, my link is working. http://wesnoth.io/brew/head/wesnoth-new.rb redirects to https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/ which is the latest version of wesnoth-new-rb
20141030 06:15:00< shadowm> With the GUI1 titlescreen they'd disappear temporarily when bringing up GUI1 dialogs but that's it.
20141030 06:15:38< ancestral> iceiceice: Anyway, same problem: https://paste.ee/p/LuBzN
20141030 06:16:08< ancestral> What’s that —prefix for? where is it being set?
20141030 06:17:15< iceiceice> ancestral: i think for some reaosn its not treating the scons line as a shell command now,
20141030 06:17:22< iceiceice> i think maybe using "system" is ncessary
20141030 06:17:39< iceiceice> maybe the brew audit for linux brew only works for linux, although that would be a bit bizarre
20141030 06:17:51< iceiceice> it could just buggy or something
20141030 06:18:35< iceiceice> heres how it was before: https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/e5fef395b21ff248f4251d5d9eac0c6a9a9cd7b2/wesnoth-new.rb
20141030 06:19:02< ancestral> Want me to test with that version?
20141030 06:19:30< ancestral> Or you’re just showing me the `system “scoons”, *args` line
20141030 06:19:40< ancestral> *scons
20141030 06:19:44< iceiceice> yeah see if that line works
20141030 06:19:56< ancestral> I’ll test with this version
20141030 06:20:28< iceiceice> it might help if i actually knew ruby instead of pretending i do :p
20141030 06:21:11< ancestral> Yeah I see
20141030 06:21:14< ancestral> This works
20141030 06:21:21< ancestral> It actually runs scons
20141030 06:22:01< ancestral> I get it. I had the steps wrong in my head.
20141030 06:24:00-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20141030 06:24:03-!- vultraz [~chatzilla@124.109.10.167] has quit [Changing host]
20141030 06:24:03-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has joined #wesnoth-dev
20141030 06:33:09< ancestral> src/whiteboard
20141030 06:33:18< ancestral> Must be ending soon
20141030 06:34:54< iceiceice> yeah thats a good indicator
20141030 06:36:25< ancestral> Translations now
20141030 06:36:42< ancestral> scons: done building targets.
20141030 06:37:46< iceiceice> cool
20141030 06:37:47< ancestral> iceiceice: Works, installs as ‘wesnoth’
20141030 06:38:06< ancestral> ln -s ../Cellar/wesnoth-new/HEAD/share/wesnoth wesnoth
20141030 06:38:09< iceiceice> ok, well thats great
20141030 06:38:40< iceiceice> so, whats left is (1) os x bundles
20141030 06:38:45< iceiceice> (2) get the version number thingie
20141030 06:38:53< iceiceice> (3) wesnothd (maybe...)
20141030 06:39:06< iceiceice> i dont think wesnothd is really essential but its probably fixable
20141030 06:39:19< iceiceice> the os x bundle thing:
20141030 06:39:36< ancestral> Command W and Q seem to work correctly, that’s good
20141030 06:39:39< iceiceice> there's some code that says, if you do "scons wesnoth-bundle", its supposed to make a Battle for Wesnoth.app
20141030 06:39:47< ancestral> Alright, let me try
20141030 06:39:52< iceiceice> i dont know what you are supposed to do with that though, where is that suppsoed to be installed?
20141030 06:40:11< iceiceice> you probably have to copy it somewhere after the scons runs
20141030 06:42:01< shadowm> wesnothd is needed for hosting LAN games.
20141030 06:42:10< ancestral> We know
20141030 06:43:40< ancestral> iceiceice: I’ll copy your gist and change s/wesnoth/wesnoth-bundle
20141030 06:44:08< ancestral> What is ccache for?
20141030 06:47:04< ancestral> iceiceice: `Error: uninitialized constant WesnothBundle`
20141030 06:47:18< ancestral> Oh wait
20141030 06:47:24< ancestral> My mistake hang on
20141030 06:47:48< shadowm> Speeding up compilation for files whose generated code didn't change (e.g. after someone alters a comment block in a frequently-included header).
20141030 06:48:46< shadowm> Or also after switching tags/branches causing some other factor scons doesn't account for to change.
20141030 06:49:06< ancestral> Thanks
20141030 06:53:34-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Remote host closed the connection]
20141030 07:10:03-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20141030 07:13:47-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141030 07:18:38-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 250 seconds]
20141030 07:42:24-!- Gallaecio [~quassel@84.120.180.152.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20141030 07:47:26-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20141030 07:50:15-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20141030 07:51:46-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141030 08:06:43< shadowm> Yes, hi.
20141030 08:06:55< shadowm> I'm sorry I didn't notice earlier, but the irker issue was actually at our end. It should be fixed now.
20141030 08:07:11< shadowm> ("our end" = not wesnoth.org)
20141030 08:12:03-!- shadowm [~ignacio@wesnoth/developer/shadowm] has quit []
20141030 08:12:04-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141030 08:14:03-!- shadowm [~ignacio@fehu.ai0867.net] has joined #wesnoth-dev
20141030 08:14:04-!- shadowm [~ignacio@fehu.ai0867.net] has quit [Changing host]
20141030 08:14:04-!- shadowm [~ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141030 08:41:38-!- Coffee_irc [~david@ppp118-210-90-49.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20141030 08:57:56< shadowm> Yay.
20141030 08:58:02< shadowm> I can't build wesnoth with clang 3.5.
20141030 09:00:15-!- irker765 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141030 09:00:15< irker765> wesnoth: Ignacio R. Morelle wesnoth:1.12 4946dc6bd79f / src/campaign_server/campaign_server.cpp: campaignd: Emit debug messages when writing configuration to disk http://git.io/2mgpxQ
20141030 09:00:18< irker765> wesnoth: Ignacio R. Morelle wesnoth:1.12 5fa2b70bcfcb / src/campaign_server/campaign_server.cpp: campaignd: Write config to disk even if the clock is running backwards http://git.io/3-n9KQ
20141030 09:00:21< irker765> wesnoth: Ignacio R. Morelle wesnoth:1.12 5e80fb1a3820 / src/campaign_server/campaign_server.cpp: campaignd: Fix abs() value truncation warning on clang 3.5 http://git.io/zNk0Hg
20141030 09:00:24< irker765> wesnoth: Ignacio R. Morelle wesnoth:1.12 64c71161f1fb / src/campaign_server/campaign_server.cpp: campaignd: Use a POSIX monotonic clock for scheduling config commits http://git.io/osYxLw
20141030 09:00:33< irker765> wesnoth: Ignacio R. Morelle wesnoth:master 58f7622db747 / src/campaign_server/campaign_server.cpp: campaignd: Fix abs() value truncation warning on clang 3.5 http://git.io/hWUOiQ
20141030 09:01:58< shadowm> iceiceice: What do you think about this? http://pastebin.com/DDebE63Q
20141030 09:02:46-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141030 09:02:55< shadowm> To be honest I had never heard of anything like that. Considering that there's nothing enforcing the non-null pointer requirement at runtime...
20141030 09:07:28-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 265 seconds]
20141030 09:08:04< shadowm> But yeah, apparently the standard says that this cannot be null.
20141030 09:08:08< shadowm> Yay for UB optimizations.
20141030 09:12:54< shadowm> AFAICT after disabling -Werror, those three warnings (which come from a header) are all that needs to be corrected for clang 3.5.
20141030 09:22:14< shadowm> Oh crap.
20141030 09:22:28< shadowm> src/lua/lgc.cpp:128:34: error: address of '((o)->value_).gc->ts' will always evaluate to 'true' [-Werror,-Wpointer-bool-conversion]
20141030 09:22:35< shadowm> :\
20141030 09:23:38< shadowm> And on 1.12 I'm getting a few like this: src/floating_point_emulation.hpp:437:21: error: using integer absolute value function 'abs' when argument is of floating point type [-Werror,-Wabsolute-value]
20141030 09:24:56< irker765> wesnoth: Ignacio R. Morelle wesnoth:master e5744562cc33 / src/game_data.hpp: Remove `this` null pointer comparisons http://git.io/kvIhQg
20141030 09:25:04< shadowm> I don't know how to deal with the lua/lgc.cpp case because it's coming from a macro.
20141030 09:25:59-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141030 09:27:38 * shadowm falls back to clang 3.4 on 1.12 for now.
20141030 09:30:48-!- kex [~kex@78.157.29.160] has quit [Ping timeout: 264 seconds]
20141030 09:35:45-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has quit [Ping timeout: 255 seconds]
20141030 10:12:31-!- kex [~kex@78.157.29.160] has joined #wesnoth-dev
20141030 10:23:00-!- sachith500 [~kvirc@112.135.142.87] has joined #wesnoth-dev
20141030 10:58:45-!- Anakonda [Anakonda@87-92-148-120.bb.dnainternet.fi] has joined #wesnoth-dev
20141030 11:00:17-!- sachith500 [~kvirc@112.135.142.87] has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
20141030 11:24:09-!- prkc [~prkc@4E5C2C2A.dsl.pool.telekom.hu] has joined #wesnoth-dev
20141030 11:37:53-!- thunderstruck [~zaibotren@cpc8-sgyl29-2-0-cust37.sgyl.cable.virginm.net] has joined #wesnoth-dev
20141030 11:41:55-!- kex [~kex@78.157.29.160] has quit [Remote host closed the connection]
20141030 12:19:07-!- kex [~kex@77.29.84.206] has joined #wesnoth-dev
20141030 12:25:00-!- irker765 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20141030 12:32:21-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20141030 12:36:41-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141030 12:56:51-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Read error: Connection reset by peer]
20141030 12:57:03-!- matthiaskrgr [matthiaskr@unaffiliated/matthiaskrgr] has quit [Ping timeout: 265 seconds]
20141030 12:58:13-!- matthiaskrgr [matthiaskr@I.Eat.Babies.PanicBNC.nl] has joined #wesnoth-dev
20141030 12:58:37-!- matthiaskrgr is now known as Guest75272
20141030 12:59:39-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141030 13:00:48-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141030 13:11:53-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Ping timeout: 264 seconds]
20141030 13:19:45-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 13:25:08-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20141030 13:27:25-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141030 13:32:03-!- Coffee_irc [~david@ppp118-210-90-49.lns20.adl2.internode.on.net] has quit [Quit: Konversation terminated!]
20141030 13:32:25< fendrin> hi mattsc
20141030 13:32:52< mattsc> hi fendrin
20141030 13:42:34-!- mjs-de [~mjs-de@f049176157.adsl.alicedsl.de] has joined #wesnoth-dev
20141030 13:55:12-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20141030 13:57:31-!- lipkab [~the_new_l@host-91-147-211-128.biatv.hu] has joined #wesnoth-dev
20141030 13:59:00-!- fendrin [~quassel@wesnoth/developer/fendrin] has quit [Ping timeout: 244 seconds]
20141030 13:59:27-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141030 14:06:41-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Ping timeout: 272 seconds]
20141030 14:09:54-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 14:10:55< ancestral> Anyone very familiar with scons?
20141030 14:17:30-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141030 14:26:51-!- Guest75272 [matthiaskr@I.Eat.Babies.PanicBNC.nl] has quit [Changing host]
20141030 14:26:51-!- Guest75272 [matthiaskr@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20141030 14:27:06-!- Guest75272 is now known as matthiaskrgr
20141030 14:47:23-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20141030 14:57:58< mattsc> gfgtdf, iceiceice, ancestral: I am going to go back to the old filesystem code for the OS X 1.12-branch releases.
20141030 15:00:35< mattsc> I believe that I’d have to update pango/cairo and all the other libraries too in order to get it to work correctly for OS X 10.5, and that brings with it a whole rat’s tail of other problems (I’ve done it in master, I know) that I am not comfortable with so close to the next stable release.
20141030 15:00:59< mattsc> And since there appear to be no problems at all with the old filesystem code in OS X, I think that’s the best solution.
20141030 15:01:53< mattsc> And just to reiterate this, this is only for backward compatibility with OS X 10.5, there’s no problem at all with more recent versions (so it won’t be a problem in master).
20141030 15:09:07< lipkab> Can someone with an up-to-date 1.12 build spare an instance to help me with testing?
20141030 15:14:50< mattsc> lipkab: I can’t right now because my builds are all messed up due to trying to get the 10.5 version to work  :(
20141030 15:15:09< lipkab> :(
20141030 15:16:37< mattsc> I’ll be back up and running by tonight at the latest, if you haven’t found somebody else to help you by then.
20141030 15:28:48-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Remote host closed the connection]
20141030 15:31:40-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141030 15:38:28-!- kex [~kex@77.29.84.206] has quit []
20141030 15:41:51-!- kex [~kex@77.29.84.206] has joined #wesnoth-dev
20141030 15:59:45-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 16:03:51-!- cjhopman_ [cjhopman@nat/google/x-umkfgtyovlvjuoes] has joined #wesnoth-dev
20141030 16:05:00-!- cjhopman [cjhopman@nat/google/x-iqwhtawstobpnqfp] has quit [Ping timeout: 265 seconds]
20141030 16:05:46-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141030 16:06:23< iceiceice> ancestral: sorry i left last night without saying anything, i got distracted
20141030 16:06:32< ancestral> Oh that’s quite alright!
20141030 16:06:55< ancestral> So I was able to try wesnoth-bundle and I can see the commands it ran
20141030 16:06:57< iceiceice> i made some edits to the formula, i realized we were missing some minor deps
20141030 16:07:07< iceiceice> ok good
20141030 16:07:59< ancestral> Well
20141030 16:08:06< ancestral> Except I think the bundle is getting deleted
20141030 16:08:11< ancestral> https://paste.ee/p/bBjgw
20141030 16:08:34< iceiceice> ok
20141030 16:08:42< iceiceice> yeah so i think the script should install it somewhere
20141030 16:08:50< iceiceice> or put it somewhere for packager to fin
20141030 16:08:51< iceiceice> *find
20141030 16:08:58< iceiceice> where is a good place for that?
20141030 16:09:19< ancestral> Well, maybe ~/Applications/
20141030 16:09:24< ancestral> or just ~/
20141030 16:09:45< ancestral> OS X puts Applications in /Applications so for an actual formal install that’s what you’d want
20141030 16:10:18< ancestral> Typically in OS X you’d only put stuff in /Applications until it’s done done
20141030 16:10:23< ancestral> Like a release
20141030 16:10:27< loonycyborg> ancestral: what did you want to ask about scons?
20141030 16:10:34< ancestral> So I’d say ~/ is probably better
20141030 16:10:59< ancestral> loonycyborg: iceiceice might be able to assist, I am doing a scons wesnoth-bundle but I’m having a problem finding the app
20141030 16:11:17< ancestral> it creates; I think it’s getting put into the tmp directory and then getting deleted before it’s copied anywhere
20141030 16:11:30< iceiceice> yeah so homebrew writes something about the app bundles:
20141030 16:11:31< iceiceice> https://github.com/Homebrew/homebrew/blob/master/share/doc/homebrew/Acceptable-Formulae.md
20141030 16:11:31< loonycyborg> I can't test that target due to not having a mac
20141030 16:11:34< ancestral> loonycyborg: https://paste.ee/p/bBjgw
20141030 16:11:38< iceiceice> https://github.com/Homebrew/homebrew/blob/master/share/doc/homebrew/Acceptable-Formulae.md#stuff-that-builds-a-app
20141030 16:11:38< ancestral> Right
20141030 16:11:50< ancestral> I was reading up scons Mkdir and Copy
20141030 16:12:15< iceiceice> ancestral: i think in this case it might be easier to make the homebrew script copy it out
20141030 16:12:26< ancestral> Sure
20141030 16:12:29< ancestral> the .rb file?
20141030 16:12:32< iceiceice> y
20141030 16:12:42< mattsc> ancestral: if you use Spotlight and search for ‘Battle for Wesnoth’, does it show you anything?
20141030 16:13:19< ancestral> mattsc: Sadly no
20141030 16:13:33< ancestral> I did a find / -iname “Battle for Wesnoth” and didn’t find anything
20141030 16:13:51< mattsc> ancestral: I had a problem with Macports recently where it put an intermediate product into a hidden (dotted) directory …
20141030 16:15:20< ancestral> iceiceice: To clarify, with the rest of the homebrew stuff, homebrew puts things in /usr/local/bin, like if I do a regular wesnoth scons build, so maybe wesnoth-bundle should go there, but ultimately, if making a bundle app (wrapper) it probably should be copied out like you said
20141030 16:15:33< mattsc> ancestral: you might need find / -iname “Battle for Wesnoth*”
20141030 16:15:46< ancestral> mattsc: Right, UNIX find should look through hidden directories
20141030 16:15:54< ancestral> I’ll try it
20141030 16:16:16< ancestral> mattsc: Is there a way to ignore permissions error output?
20141030 16:16:30< ancestral> on `find`?
20141030 16:16:36< mattsc> you could send error output to  file instead of console
20141030 16:16:54< mattsc> append:  2>~/Desktop/error.txt
20141030 16:16:56< mattsc> or something
20141030 16:17:07< ancestral> or 2> /dev/null ?
20141030 16:17:21< mattsc> if you don’t want it for reference later, sure
20141030 16:18:03< iceiceice> hmm i'm not sure right now to make homebrew put stuff in the home folder,
20141030 16:18:12< iceiceice> all the examples are like "bin.install [...]"
20141030 16:18:13< ancestral> But just to clarify, `Copy("Battle For Wesnoth.app/Contents/MacOS/wesnoth", "wesnoth”)` means it’s still called “Battle for Wesnoth” right?
20141030 16:18:41< iceiceice> ancestral: i think that is copying the wesnoth folder from the build, into the bundle
20141030 16:18:55< ancestral> That’s what I was thinking
20141030 16:18:57< iceiceice> so it should still be called Battle For Wesnoth.app
20141030 16:19:08< iceiceice> but its in the temporary dir so it got deleted i guess
20141030 16:19:16< ancestral> I jusyt get this feeling it’s never leaving the tmp directory
20141030 16:19:30< ancestral> So after the scons command I can just add something to it?
20141030 16:19:59< ancestral> e.g. `mv ??? ~/Wesnoth.app`
20141030 16:20:20< ancestral> How do I reference the built app?
20141030 16:20:34< iceiceice> i thik its just a directory, so that should work?
20141030 16:20:51< iceiceice> but its better not to hardcode ~ stuff in the scons
20141030 16:20:58< ancestral> Or just `mv “Battle For Wesnoth.app” ~/Wesnoth.app` wouldu be fine? (no curly quotes)
20141030 16:21:08< mattsc> ancestral: another option, if you just want to ignore the permission error lines but keep the others is to pipe into grep:
20141030 16:21:11< iceiceice> if you were going to do it that way it should be like an argument "bundledir" or something
20141030 16:21:15< mattsc> … | grep -iv permission
20141030 16:21:21< ancestral> I see
20141030 16:21:51< ancestral> mattsc: Searching still, I have results, but they’re other files I have with the name in it
20141030 16:22:08< ancestral> (Like a download of a web page with “Battle for Wesnoth” in the file name)
20141030 16:23:45< mattsc> ancestral, iceiceice: Btw, did you see what I said about the OS X 1.12 releases?
20141030 16:23:55< iceiceice> yeah i saw
20141030 16:24:05< iceiceice> ancestral: try this script
20141030 16:24:06< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/fb392e24c0fb3ddad1df9bea9549100e2bbe8343/wesnoth-new.rb
20141030 16:24:11< mattsc> I have now reverted to the old libs and it works just fine (including map localization) on both 10.9 and 10.5
20141030 16:24:14< iceiceice> with argument "--build-os-x-bundle"
20141030 16:24:21< ancestral> OK
20141030 16:24:41< iceiceice> it shold put it in /usr/local/etc, thats probably not ideal but lets just see if it works at all
20141030 16:24:53< mattsc> I need two minor changes to the C++ code, but I would say this is the way to go for 1.12.0.
20141030 16:26:32< mattsc> I’ll stop bugging you for now then, I consider my problem solved and will ask a couple follow-up questions later  :)
20141030 16:28:40< ancestral> Okay
20141030 16:28:41< ancestral> iceiceice: running it now
20141030 16:29:48< ancestral> Might need to add `brew link gettext --force` to the .rb too
20141030 16:30:12< iceiceice> ancestral: actually at least the new versions don't use gettext
20141030 16:30:17< ancestral> I tried unlinking gettext and brewing just for the heck of it but didn’t make a difference
20141030 16:30:18< ancestral> oh
20141030 16:30:25< ancestral> (tried last night)
20141030 16:30:54< iceiceice> i dont know righ tnow how to set up different dependencies for stable vs development
20141030 16:30:54< ancestral> ---[applying configuration]---
20141030 16:30:54< ancestral> error: No repo or cache detected.
20141030 16:30:54< ancestral> Not a problem?
20141030 16:31:06< iceiceice> i think thats okay actually
20141030 16:31:15< iceiceice> the scons runs something called "autorevision" that looks for the git tag number and writes it somewhere
20141030 16:31:19< iceiceice> i think its making that message
20141030 16:32:40< iceiceice> ancestral: what i meant is, new versions of wesnoth don't use gettext because it doesn't support unicode on all platforms, we're using boost now instead
20141030 16:32:58< iceiceice> *libintl, i guess is what i mean
20141030 16:33:14< ancestral> Oh yeah? Is that a major change?
20141030 16:33:20< iceiceice> its part of the boost filesystem thing
20141030 16:35:00-!- boucman_work [~jrosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141030 16:35:13< ancestral> I forgot to change the flag
20141030 16:35:44< mattsc> ancestral, iceiceice: btw, I don’t remember if I have said that yet, I think it’s great that you are working on this!
20141030 16:36:16< ancestral> mattsc: Thanks (I’m just a guinea pig; you should really be thanking iceiceice)
20141030 16:37:00< mattsc> Well, you can divide the thanks between the two of you as you like, I’ll leave that up to you  ;)
20141030 16:37:52< mattsc> I just think that it will be a huge addition to the project; and I'm hopeful that it will make my own life easier as well.
20141030 16:37:57< iceiceice> :)
20141030 16:38:05< mattsc> So, I’m actually being quite selfish here
20141030 16:38:09< ancestral> mattsc: I think this is a good way to build it, but it’s not a good delivery method for average users
20141030 16:38:12< iceiceice> i'm not sure if this bundle code will actually work, it looks like last person to touch it was ESR in 2008...
20141030 16:38:37< iceiceice> hopefully no one at apple had an epiphany and changed the format
20141030 16:39:00< mattsc> ancestral: I assume that you’re right about that, but if anything, it can’t make things easier.
20141030 16:39:10< ancestral> harder?
20141030 16:39:21< mattsc> umm, yes  :P
20141030 16:39:23< mattsc> sorry
20141030 16:39:25< ancestral> :-D
20141030 16:39:42< mattsc> I do build intermediate dev versions for people to test occasionally, so at the very least that wouldn’t have to happen
20141030 16:40:24< ancestral> Right!
20141030 16:41:10-!- AI0867 [~ai@wesnoth/developer/ai0867] has joined #wesnoth-dev
20141030 16:42:00< ancestral> OK
20141030 16:43:14< ancestral> I see the app
20141030 16:43:31< ancestral> in /usr/local/etc and I have the wesnoth executable in /usr/local/bin
20141030 16:43:43< iceiceice> does the app work?
20141030 16:43:46< ancestral> Okay the executable works, the command wesnoth is functional
20141030 16:43:54< ancestral> Let’s see
20141030 16:44:02< ancestral> No
20141030 16:44:20< ancestral> You can’t open the application “Battle For Wesnoth” because it may be damaged or incomplete.”
20141030 16:44:35< ancestral> Browsing the package
20141030 16:44:48< ancestral> No resources
20141030 16:45:00< ancestral> Probably need to copy those into the packaage
20141030 16:45:56< ancestral> iceiceice: I think it’s because there’s no PkgInfo and/or Info.plist
20141030 16:46:29< ancestral> Odd because `echo "APPL????" > ""Battle For Wesnoth.app/Contents/PkgInfo””` is in the output
20141030 16:46:49< iceiceice> what does "echo APPL????" do for you?
20141030 16:49:07< ancestral> iceiceice: https://paste.ee/p/pzYGY
20141030 16:49:35< ancestral> It should be creating a file with “APPL????” in it
20141030 16:49:48< ancestral> That’s the app indicator
20141030 16:50:12< iceiceice> oh i see
20141030 16:50:17< ancestral> I think if there were documents which would open to it a different indicator could be used
20141030 16:52:15< ancestral> Or it might be used for app registration (like, if Wesnoth were to be on the Mac App Store)
20141030 16:52:54< iceiceice> ok, i think whatever this plist thing is might be required now,
20141030 16:53:00< iceiceice> and the pkginfo is from os x 10.5
20141030 16:53:16< iceiceice> unfortunately i cant really fix the app bundle code in the scons script :/
20141030 16:53:29< ancestral> APPL???? is the default thing
20141030 16:53:32< ancestral> And that’s fine
20141030 16:53:44< ancestral> Just the file isn’t in the bundle which is odd
20141030 16:53:54< ancestral> Since the output seems to suggest it’s being created unless 
20141030 16:54:04< iceiceice> does it fix it if its in there?
20141030 16:54:06< ancestral> it’s echoing the echo statement or something
20141030 16:54:10< ancestral> I’ll try
20141030 16:54:56< ancestral> iceiceice: Okay so it works if that file and the Info.plist are there
20141030 16:55:27< ancestral> The resources are loading probably from somewhere else in /usr/local/ I guess
20141030 16:56:11< iceiceice> thats bad though right?
20141030 16:56:23< iceiceice> it means that if you copy the bundle and give it to someone it wont work anymore
20141030 16:56:28< ancestral> Correct
20141030 16:56:47< ancestral> One second
20141030 16:57:17< ancestral> Right okay, the executable must have a link to the resources specified there
20141030 16:57:38< ancestral> (for kicks I just dragged the app to /Applications and it runs)
20141030 17:08:05< ancestral> iceiceice: gettext is still listed as a dependency, right?
20141030 17:08:26< iceiceice> it is, and it is for 1.10 also
20141030 17:08:56< ancestral> What I mean is, I uninstalled, unlinked gettext, and uninstalled gettext
20141030 17:09:03< ancestral> Then re-installed
20141030 17:09:09< ancestral> wesnoth-new
20141030 17:09:33< ancestral> And it’s getting gettext still. So that’s still necessary atm?
20141030 17:09:38< iceiceice> yeah i didnt change it
20141030 17:09:55< iceiceice> because i dont know how to make the .rb file give different deps for stable vs dev vs head
20141030 17:10:02< ancestral> Ah ok
20141030 17:10:06< ancestral> You did say that
20141030 17:10:17< iceiceice> i guess i can try something
20141030 17:10:55< ancestral> No problem
20141030 17:11:35< iceiceice> hmm scons needs to change also maybe?
20141030 17:13:13< iceiceice> oh
20141030 17:14:31-!- boucman_work [~jrosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141030 17:15:51< iceiceice> ok try this maybe
20141030 17:15:57< iceiceice> with --head option i dont think it should require gettext
20141030 17:16:28< iceiceice> hmmm wait
20141030 17:16:33< iceiceice> when i remove gettext,
20141030 17:16:41< iceiceice> cairo and pango become unhappy
20141030 17:16:52< iceiceice> so i dont think this is going to work
20141030 17:18:49-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Read error: Connection reset by peer]
20141030 17:20:55-!- kex [~kex@77.29.84.206] has quit [Remote host closed the connection]
20141030 17:20:57< ancestral> Sure
20141030 17:21:27< ancestral> So then we might need brew link gettext —force
20141030 17:21:46< iceiceice> why do we have to do that agian?
20141030 17:22:08-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20141030 17:22:21< ancestral> I get an error otherwise
20141030 17:22:37< iceiceice> at which step?
20141030 17:22:38< ancestral> error_action(["wesnoth"], [])
20141030 17:22:38< ancestral> scons: *** [wesnoth] Target disabled because its prerequisites are not met
20141030 17:22:39< ancestral> 
20141030 17:22:43< iceiceice> ohhh i see
20141030 17:22:51< iceiceice> yeah then i guess the sconstruct fix might fix it
20141030 17:23:32< iceiceice> does your system come with msgfmt though?
20141030 17:23:41< iceiceice> the translations might not work if gettext isnt linked, i guess we'll see
20141030 17:24:00< ancestral> msgfmt? I don’t know
20141030 17:24:04-!- Gallaecio [~quassel@84.120.180.152.dyn.user.ono.com] has joined #wesnoth-dev
20141030 17:24:15< iceiceice> so when its listing thousands of lines of translations blahblah
20141030 17:24:28< iceiceice> thats when its running msgfmt on every translator file i think
20141030 17:29:39< iceiceice> ok i'm going for lunch, be back later
20141030 17:29:41-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141030 17:33:09-!- gfgtdf [~chatzilla@f054141011.adsl.alicedsl.de] has joined #wesnoth-dev
20141030 17:36:22-!- mjs-de [~mjs-de@f049176157.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141030 17:42:11< ancestral> Looks like Command+Q doesn’t quit, Command+W does. I remember that change, I’m not sure why. Maybe in the Xcode project they re-mapped/double-mapped some of this stuff
20141030 17:43:57-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20141030 17:44:19< mattsc> ancestral: for me command+Q does quit (always has and still does)
20141030 17:44:38< mattsc> gfgtdf: hi
20141030 17:45:07< gfgtdf> mattsc: hi
20141030 17:45:31< mattsc> gfgtdf: so I am going back to the old filesystem code for 1.12 for compatibility with OS X 10.5
20141030 17:45:44< mattsc> I need two little changes to the C++ code for that
20141030 17:46:06< mattsc> They are easy enough that I could figure them out myself, but you will know without having to look up anything, so would you do them for me  :)
20141030 17:46:38< gfgtdf> mattsc: which changes do you mean ?
20141030 17:46:57< mattsc> gfgtdf: sorry, wasn’t prepared, looking them up ...
20141030 17:48:11< mattsc> gfgtdf: https://github.com/wesnoth/wesnoth/blob/1.12/src/sound.cpp#L493
20141030 17:48:44< ancestral> mattsc: Then there must be some Xcode glue to make that work
20141030 17:48:46< mattsc> that needs to be an #ifndef for both pandora and apple, but I don’t know the syntax how to do that off the top off my head
20141030 17:48:54< gfgtdf> mattsc: so you want to use teh odl SDL_mixer code too ?
20141030 17:49:27< mattsc> I want to use the code in l.497  (if tested that that works)
20141030 17:50:01< mattsc> the sdl_mixer version I have is too old (1.12.8 or so)
20141030 17:50:03< gfgtdf> mattsc; i'd try be replaing teh ifdef with #if defined(PANDORA) || defined(__APPLE__)
20141030 17:50:21< mattsc> gfgtdf: okay, will do
20141030 17:50:50< mattsc> gfgtdf: the other one is here: https://github.com/wesnoth/wesnoth/blob/1.12/src/gettext.cpp#L135
20141030 17:51:06< mattsc> that gives me an ‘errno not defined' error message
20141030 17:52:42< gfgtdf> mattsc: for strerror or errno ?
20141030 17:52:43< mattsc> gfgtdf: btw, the #ifdef line is an #ifndef ; so do I just add ! in front of the defined()’s
20141030 17:53:03< mattsc> I think it’s for errno; let me check (I commented it out for testing)
20141030 17:53:19< gfgtdf> mattsc: ok than include <cerrno>
20141030 17:53:59< gfgtdf> mattsc: y but you need to to replce the || with a && then too
20141030 17:54:14< gfgtdf> #if !defined(PANDORA) && !defined(__APPLE__)
20141030 17:55:41< mattsc> gfgtdf: right …
20141030 17:55:57< mattsc> gfgtdf: yes: src/gettext.cpp:135:47: error: use of undeclared identifier 'errno’
20141030 17:56:24< gfgtdf> mattsc: y add a #include <cerrno> where tehother includes are
20141030 17:56:25< mattsc> so where do I add that include?  Inside the #ifdef?
20141030 17:56:40< mattsc> for all systems?
20141030 17:56:48< mattsc> or guarded by #ifdef also?
20141030 17:57:57< mattsc> gfgtdf: sorry for the dumb questions, I just don’t want to violate any coding standard which I am not aware of  :P  (I do know what the commands do)
20141030 17:59:09< gfgtdf> mattsc: I dont really care whether you put an ifdef around it, maybe someone else does.. . the include shoudl go somewhere here https://github.com/wesnoth/wesnoth/blob/1.12/src/gettext.cpp#L19
20141030 17:59:59< mattsc> gfgtdf: okay - well, the Windows specific include in l.24 is inside #ifdef, so I’ll do the same
20141030 18:00:07< mattsc> gfgtdf: thanks, that’s all  :)
20141030 18:03:24< ancestral> iceiceice: Okay, maybe these flags will help defeat that gettext error and not require —force
20141030 18:03:29< ancestral>     LDFLAGS:  -L/usr/local/opt/gettext/lib
20141030 18:03:30< ancestral>     CPPFLAGS: -I/usr/local/opt/gettext/include
20141030 18:05:31< ancestral> Aha — wesnothd built for me
20141030 18:22:49< ancestral> mattsc, would this kind of flag go into https://github.com/wesnoth/wesnoth/blob/master/scons/gettext.py?
20141030 18:23:32< ancestral> (I can wait for iceiceice to get back)
20141030 18:24:41< mattsc> ancestral: uh, yeah, better wait since I’d just be guessing.  You might be able to set those as environment variables, but I don’t know if you want to apply them at all times or not.
20141030 18:25:17< ancestral> Sure
20141030 18:34:30-!- Sulfur [~Miranda@p5B3279A8.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141030 18:41:22-!- iceiceice [~chris@192.16.204.74] has joined #wesnoth-dev
20141030 18:41:22-!- iceiceice [~chris@192.16.204.74] has quit [Changing host]
20141030 18:41:22-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141030 18:42:28< iceiceice> ancestral: i think actually, what it should be is 
20141030 18:42:33< iceiceice> scons should get "gettextdir=/usr/local/opt/gettext/"
20141030 18:42:53< iceiceice> in the ruby script
20141030 18:43:11< iceiceice> because otherwise on every other build for every other system it will be looking in /usr/local/opt for things
20141030 18:43:25< ancestral> Right
20141030 18:43:43< iceiceice> but also, maybe it hsold be like, .../cellar/gettext/ something something?
20141030 18:43:54< iceiceice> do you have to link gettext to find it in /usr/local/opt?
20141030 18:44:38< iceiceice> gfgtdf: does the server still require lib intl? or does it use gettext_boost
20141030 18:47:12< gfgtdf> iceiceice: actual i would be uprosed if the server prints a translated message even once. 
20141030 18:47:28< iceiceice> hmm ok
20141030 18:47:40< iceiceice> for some reasons scons lists libintl as a server dependency
20141030 18:47:50< iceiceice> but it might just be an oversight
20141030 18:48:27< gfgtdf> iceiceice: well at least on the msvc projectfile teh server lings against the tstrng, gettext(-boost) objects but idk why
20141030 18:49:06< gfgtdf> iceiceice: building wesnothd currently
20141030 18:50:36< gfgtdf> iceiceice: on my msvc build wesnothd doesnt need libintl
20141030 18:50:59< gfgtdf> iceiceice: all dlls that are loaded are SDL.dll and SDL_net.dll
20141030 18:51:02< ancestral> iceiceice: Let me check
20141030 18:51:29< ancestral> Yes, /usr/local/opt/
20141030 18:51:46< gfgtdf> iceiceice: note that all boost ist loked staticly so it doesnt appear in "loaded dls"
20141030 18:52:04< gfgtdf> linked*
20141030 18:52:25< iceiceice> ancestral: if you unlink gettext is it still in usr/local/opt
20141030 18:52:26< iceiceice> ?
20141030 18:52:29< iceiceice> my guess is no
20141030 18:52:35< ancestral> Le’t see
20141030 18:52:39< ancestral> *let’s
20141030 18:52:54< ancestral> Still there
20141030 18:53:31< ancestral> But if I uninstall it, then it’s gone
20141030 18:54:08< iceiceice> what if you install but leave it keg only?
20141030 18:54:14< iceiceice> (i.e. dont link --force)
20141030 18:54:21< ancestral> Then I get the error
20141030 18:54:29< iceiceice> and its not in usr/local/opt?
20141030 18:54:36< ancestral> Hang on
20141030 18:54:41< ancestral> I’ve uninstalled gettext
20141030 18:54:54< ancestral> I’ll uninstall wesnoth, and rebuild
20141030 18:59:24< ancestral> iceiceice: Maybe you made a change to the script
20141030 18:59:30< ancestral> http://wesnoth.io/brew/head/wesnoth-new.rb
20141030 18:59:45< ancestral> You did
20141030 18:59:51< ancestral> So this might be working then
20141030 19:00:10< ancestral> It’s cloning now from your github
20141030 19:01:27< ancestral> gettext got put into /usr/local/Cellar and it’s in /usr/local/opt
20141030 19:02:48< iceiceice> ok
20141030 19:02:56< iceiceice> i changed my formula again, i think it should work now without linking gettext
20141030 19:02:57< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/fccd7a365a2479e6e3db3788c22d98ec21335d66/wesnoth-new.rb
20141030 19:03:00< ancestral> Changing install name in /usr/local/Cellar/gettext/0.19.3/bin/envsubst
20141030 19:03:01< ancestral>   from @@HOMEBREW_CELLAR@@/gettext/0.19.3/lib/libintl.8.dylib
20141030 19:03:01< ancestral>     to /usr/local/Cellar/gettext/0.19.3/lib/libintl.8.dylib
20141030 19:03:02< ancestral> etc.
20141030 19:03:33< ancestral> Oh nm that
20141030 19:03:40< ancestral> Okay
20141030 19:05:27< iceiceice> ok im testing now also
20141030 19:05:48< ancestral> It’s still asking for gettext but
20141030 19:06:13< ancestral> It doesn’t error as it did yesterday
20141030 19:08:04< ancestral> I’m deleting my caches folder (/Library/Caches/Homebrew)
20141030 19:08:24< iceiceice> ok
20141030 19:08:28< iceiceice> i thnk that will delete your ccache also
20141030 19:08:32< iceiceice> so it will take you like an hour to build
20141030 19:08:38< iceiceice> but its good for a test
20141030 19:08:43< ancestral> That’s alright
20141030 19:08:50< ancestral> I have the day off anyway :-P
20141030 19:09:13< ancestral> I took the ccache argument off the command too
20141030 19:12:30< lipkab> Hm... I'm getting a lot of Russian spam lately.
20141030 19:13:06< lipkab> Too bad I can't understand it.
20141030 19:14:48< lipkab> Apparently, they are targeting some SUSE workers alongside with me.
20141030 19:15:26< lipkab> Which is strange because I have absolutely no affiliation with SUSE. Besides being registered on the openSUSE forums, that is.
20141030 19:18:08< iceiceice> ancestral: it seems to work for me
20141030 19:18:23< iceiceice> but linuxbrew is a little bit leaky, its hard to know if its working becasue it finds brew gettext or system gettext
20141030 19:18:29< ancestral> Building, at src/ai/*
20141030 19:18:36< iceiceice> ah so it found gettext/
20141030 19:18:37< iceiceice> ?
20141030 19:18:44< iceiceice> thats good
20141030 19:18:55< ancestral> iceiceice: It downloaded gettext
20141030 19:19:06< iceiceice> and passed the libintl check?
20141030 19:19:07< ancestral> But didn’t complain about it not linked or what have you
20141030 19:19:15< ancestral> I believe so, I’ll look for that line
20141030 19:19:36< iceiceice> oh hmm
20141030 19:19:48< iceiceice> i think i didnt push a change to my fork,
20141030 19:19:58< iceiceice> if it will fail, i think it will fail now when you are trying to translate stuff
20141030 19:20:20< ancestral> Does that happen after building src?
20141030 19:20:44< iceiceice> acutally i think it will just break
20141030 19:20:50< iceiceice> you should probably abort, theres no way it will find gettext
20141030 19:20:53< ancestral> `HEAD is now at ee3304d scons: libintl not required for BFS builds (boost locale replaces)`
20141030 19:20:55< iceiceice> if it doesnt run that check
20141030 19:21:23< iceiceice> the check is the only place that it looks at the "gettextdir" variable i think
20141030 19:21:33< iceiceice> idk maybe OS X provides msgfmt?
20141030 19:21:36< iceiceice> i guess we will find out
20141030 19:22:09< iceiceice> i have to go soon
20141030 19:22:14< ancestral> Sure thing
20141030 19:22:33< ancestral> At src/gui/*
20141030 19:23:39< iceiceice> ancestral: yeah the translations start after building src
20141030 19:24:02< iceiceice> a bit unfortunate... i think we can control it actually
20141030 19:24:14< iceiceice> if we put "wesnoth" before "install" i think it will be opposite
20141030 19:24:21< iceiceice> but it probably doesnt really matter
20141030 19:28:09< iceiceice> ancestral: i think we're doing it right,
20141030 19:28:15< iceiceice> there's some other formulas that use gettext this way: https://github.com/Homebrew/homebrew/blob/01cf6061a12616536e88a31ab91da03fb7568fb6/Library/Formula/glib-networking.rb#L47
20141030 19:32:44< iceiceice> actually i think theres an even better way, updated: https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/a7c22ebce6eb5a803498f7feeeee764d6b614975/wesnoth-new.rb
20141030 19:33:05< ancestral> What did you change?
20141030 19:33:21< ancestral> I can ask github that question
20141030 19:33:57< iceiceice> apparently in homebrew language, you can just ask the gettext formula for its own prefix
20141030 19:34:27< iceiceice> so if they change how gettext is working, the wesnoth formula might keep working
20141030 19:35:11< ancestral>  /usr/local/opt/gettext/bin/msgfmt
20141030 19:35:15< ancestral> Translations worked
20141030 19:35:23< ancestral> Oh wait now
20141030 19:35:30< ancestral> Oh hahaha
20141030 19:35:34< ancestral> Error: File exists - /usr/local/etc/Battle For Wesnoth.app
20141030 19:36:16< ancestral> Maybe should add a line to rm it beforehand or otherwise replace
20141030 19:36:55< iceiceice> yeah i just changed that line actually
20141030 19:36:55< iceiceice> https://gist.github.com/cbeck88/f5898a5cbdfe99a8d427/raw/d00da673a033456fdcbdf32e3f9fcbf483f25861/wesnoth-new.rb
20141030 19:37:10< iceiceice> it shold just forcibly move it to ~/ now
20141030 19:37:20< ancestral> I’ll give it a shot
20141030 19:38:13-!- lipkab [~the_new_l@host-91-147-211-128.biatv.hu] has quit [Ping timeout: 265 seconds]
20141030 19:42:51< ancestral> Done
20141030 19:43:25< gfgtdf> iceiceice: do you think we can now assume that all windows builds use the boost fileystem code ?
20141030 19:43:47< iceiceice> i think so
20141030 19:43:57< iceiceice> why?
20141030 19:44:20< ancestral> ==> mv -f Battle For Wesnoth.app ~/
20141030 19:44:28< ancestral> iceiceice: Need quotes?
20141030 19:44:34< ancestral> “Battle For Wesnoth.app”
20141030 19:44:47< iceiceice> it should do that i think,
20141030 19:44:52< iceiceice> idk did it fail?
20141030 19:44:56< ancestral> Well
20141030 19:45:01< ancestral> I don’t see it in ~/
20141030 19:45:07< iceiceice> the ruby line was: `system "mv", "-f", "Battle For Wesnoth.app", "~/"`
20141030 19:45:11< ancestral> I do see one in /usr/local/etc but that’s my old one?
20141030 19:45:15< ancestral> Ahh
20141030 19:45:24< ancestral> let me trash my old app
20141030 19:45:53< iceiceice> ancestral: i noticed also in the manpages,
20141030 19:46:17< iceiceice> theres something called 'linkapps" ?
20141030 19:46:28< iceiceice> i guess we don't need to do that
20141030 19:47:34< iceiceice> idk i have to go actually
20141030 19:47:35-!- iceiceice [~chris@wesnoth/developer/iceiceice] has quit [Quit: Leaving]
20141030 19:49:37< ancestral> np, thanks
20141030 19:55:46-!- Sulfur [~Miranda@p5B3279A8.dip0.t-ipconnect.de] has quit [Ping timeout: 255 seconds]
20141030 20:00:33-!- Dugi [93fbd29f@gateway/web/freenode/ip.147.251.210.159] has joined #wesnoth-dev
20141030 20:00:46< Dugi> Hello.
20141030 20:03:10< Dugi> I have a weird problem, I rebased the branch, fixed the merge conflicts, git tells me that I successfully rebased and updated, but this https://github.com/wesnoth/wesnoth/pull/134 tells me that there are some merge conflicts. Do I have to commit the rebase somehow?
20141030 20:08:52< gfgtdf> Dugi: you need to forcepush to your branch
20141030 20:14:03< Dugi> That is git push --force ?
20141030 20:18:12< Dugi> gfgtdf: That is git push --force ?
20141030 20:34:06-!- shadowm_desktop [ignacio@wesnoth/developer/shadowm] has joined #wesnoth-dev
20141030 20:34:31< gfgtdf> Dugi: yes
20141030 20:34:45< Dugi> gfgtdf: Thaniks.
20141030 20:45:57-!- irker657 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20141030 20:45:58< irker657> wesnoth: gfgtdf wesnoth:master 3d7cd48e6514 / src/ (desktop/clipboard.cpp desktop/open.cpp font.cpp game_preferences.cpp): simplyfy win32 specific code http://git.io/lQTE4w
20141030 20:55:44< Dugi> gfgtdf: Well, travis failed, a symptom that I did the intended changes.
20141030 20:59:40< Dugi> What do I do with this strange compiling problem? First it reported problems with finding boost_filesystem and boost_locale when configuring, so I installed these. But then, the compilation crashed with a load of undefined references to boost::filesystem something and boost::locale something. I am pretty sure this wasn't created by problems in my changes, any ideas how could this have happened?
20141030 21:02:11< Dugi> I have found the headers for boost locale and boost filesystem.
20141030 21:09:26< gfgtdf> Dugi: the depdnedencies hvae changed, new dependencies are boost filesystem and boost locale. IT seems like your linker fails o find these libs
20141030 21:09:41< gfgtdf> Dugi: but i dont know how compilation/installing works on linux sry.
20141030 21:09:44< Dugi> gfgtdf: Ah, that explains why this crap happened.
20141030 21:10:22< Dugi> It seems that I got the compilation to run through it, but I wonder if it will link and execute properly.
20141030 21:10:44< Dugi> Strangely, these undefined references are reported to be in the boost functions, not wesnoth functions.
20141030 21:17:36-!- iceiceice [~chris@wesnoth/developer/iceiceice] has joined #wesnoth-dev
20141030 21:17:49< iceiceice> ancestral: i just saw forum post,
20141030 21:18:00< iceiceice> i guess i should change the recipe so that it points at official wesnoth head and not my github fork?
20141030 21:18:10< iceiceice> i was testing with my fork because i wanted to see what happens if i push a small commit
20141030 21:18:19< ancestral> Oh right
20141030 21:18:38< iceiceice> the post can also be slightly clearer, that the first line installs brew, and the second is wesnoth
20141030 21:18:46< ancestral> Sure, my link points to the latest of your gist, but if you prefer something else, I can change it
20141030 21:18:54< iceiceice> yeah i mean i have to change my gist
20141030 21:18:55< ancestral> True
20141030 21:19:00< iceiceice> i think its fine to point at the gist
20141030 21:19:23< iceiceice> i am thinking to make a pull request to homebrew, maybe they will have some feedback, or can explain how to make .app's
20141030 21:19:37< iceiceice> surely theres some other project that knows how
20141030 21:20:46< iceiceice> Dugi: what kind of linux are you on?
20141030 21:21:04< Dugi> iceiceice: Ubuntu 14.04, nothing unusual.
20141030 21:21:24< shadowm> iceiceice: So what do you think of clang 3.5?
20141030 21:21:52< iceiceice> idk
20141030 21:21:56< iceiceice> i tried to google this thing about lua
20141030 21:22:00< iceiceice> i didnt find anything
20141030 21:22:01< ancestral> shadowm: Odd thing I’ve noticed, in forum posts, is [tt] seems to print noticeably smaller than regular text
20141030 21:22:06< ancestral> I think it has something to do with inheriting a style which has an undefined font size where regular body type has a size defined elsewhere
20141030 21:22:08< shadowm> ancestral: Known.
20141030 21:22:21< iceiceice> is our own brand of lua so different from everyone else?
20141030 21:22:33< iceiceice> i find it troublesome
20141030 21:22:42< shadowm> iceiceice: No, it's probably just that nobody has tried to build lua with clang 3.5 before.
20141030 21:22:51< iceiceice> no one on the whole internet?
20141030 21:23:07< shadowm> Yeah.
20141030 21:23:27< iceiceice> is it possible for me even to install clang 3.5 or is it really bleeding edge
20141030 21:23:39< shadowm> It was released on September 4.
20141030 21:24:02< shadowm> I don't know what distribution you are using or whether clang 3.5 is available there, so I can't help you with that.
20141030 21:25:09< shadowm> I wonder if it'd be more productive to disable these warnings for now or not. I know I for one don't want to deal with cleaning up macro-layered code from a third party.
20141030 21:25:41< shadowm> Hence I fell back to clang 3.4 for 1.12, since I need a working 1.12 build for playtesting.
20141030 21:25:56-!- vultraz [~chatzilla@wesnoth/developer/vultraz] has quit [Ping timeout: 265 seconds]
20141030 21:26:05< iceiceice> shadowm: i didnt see any other complaints on lua mailing list
20141030 21:26:06< iceiceice> http://comments.gmane.org/gmane.comp.lang.lua.general/112725
20141030 21:26:15< iceiceice> i think maybe should run git blame and see if we added this line at some point
20141030 21:26:32< shadowm> Sigh.
20141030 21:27:26< shadowm> iceiceice: I forgot to mention that not everyone builds 3rd party libraries with -Werror.
20141030 21:27:58< iceiceice> yeah but i would bet some of the people who do might complain
20141030 21:28:18< shadowm> I wouldn't bother reporting upstream, I'd just assume they are aware of the warnings.
20141030 21:29:15< shadowm> iceiceice: So as I expected, the line in question was introduced by commit 03d625a08ebbbe53a22b2b0cf3fa4a86931cf135.
20141030 21:30:44< iceiceice> hmm
20141030 21:31:22< iceiceice> how many warnings do you get? i don't remember now
20141030 21:31:23< shadowm> Again, it involves a macro or two. 
20141030 21:33:04-!- fabi [~quassel@wesnoth/developer/fendrin] has quit [Remote host closed the connection]
20141030 21:33:08< shadowm> iceiceice: Just that one.
20141030 21:33:17< shadowm> On master, not 1.12. 1.12 is a mess.
20141030 21:33:32< iceiceice> i can no longer find your original link
20141030 21:34:07< shadowm> src/lua/lgc.cpp:128:34: warning: address of '((o)->value_).gc->ts' will always evaluate to 'true' [-Wpointer-bool-conversion]
20141030 21:34:56-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141030 21:39:24< shadowm> Soooooooo.
20141030 21:39:34< shadowm> We are not even using the latest Lua 5.2.x version.
20141030 21:40:04< shadowm> Great job.
20141030 21:40:31< shadowm> ಠ_ಠ
20141030 21:40:57< shadowm> See, this is why I'm reluctant to bring in SDL_ttf. Because nobody ever remembers to integrate upstream bug fixes later!
20141030 21:41:42< shadowm> We are using Lua 5.2.0 and in the meantime 5.2.3 and its predecessors have been released.
20141030 21:42:10< shadowm> http://www.lua.org/bugs.html#5.2.1  http://www.lua.org/bugs.html#5.2.2  http://www.lua.org/bugs.html#5.2.3
20141030 21:42:57< shadowm> AI0867 and Crab, the last people who dealt with Lua library integration, are both inactive.
20141030 21:44:50-!- kex [~kex@92.53.30.12] has joined #wesnoth-dev
20141030 21:45:25< shadowm> So who wants to deal with this mess now?
20141030 21:50:23< iceiceice> why cant we use lua as a library?
20141030 21:50:49< iceiceice> (did i fail to understand something)
20141030 21:51:02< shadowm> Yes, C++ compilation for exception safety.
20141030 21:51:17< iceiceice> gah
20141030 21:51:19< shadowm> Distributors compile Lua as C, and that's a problem for us.
20141030 21:52:29< shadowm> From the amount of stuff listed in the upstream changelogs, it may actually be manageable for me, but I won't get around to it today and I.
20141030 21:52:38< shadowm> *to it today.
20141030 21:53:25< shadowm> It should be, in theory, as simple as getting hold of a diff between 5.2.0 and 5.2.3, fixing filename references, and trying to apply it on top or below our own patches.
20141030 21:53:29< iceiceice> this should still be an automated process somehow, am i wrong?
20141030 21:53:34< iceiceice> i guess you could apply patches,
20141030 21:53:44< iceiceice> why dont we just replace the whole /lua folder ?
20141030 21:53:46< shadowm> No.
20141030 21:53:48< iceiceice> is there no C++ exception safety version?
20141030 21:54:19< shadowm> `git log --format=oneline 92da79025de882d0fb474d9384ff18cfcce6ed42..HEAD -- src/lua `
20141030 21:55:01< shadowm> Okay, just... don't worry about it. Don't do anything. I'll try to get it done in master tomorrow.
20141030 21:56:07< shadowm> (And that's actually `git log --format=oneline 03d625a08ebbbe53a22b2b0cf3fa4a86931cf135..HEAD -- src/lua`.)
20141030 22:03:09< shadowm> mattsc: Here you have a release blocker to add to your list: https://gna.org/bugs/?22893
20141030 22:10:43< iceiceice> mattsc: did you already change the pandora lines to include apple?
20141030 22:11:10< mattsc> shadowm: okay, will do
20141030 22:11:20< mattsc> iceiceice: only locally
20141030 22:11:28< iceiceice> i think it might be better to add a new define,
20141030 22:11:36< iceiceice> because at least homebrew should be able to build with the new sdl mixer
20141030 22:11:47< iceiceice> i havent fully figured out what to do though
20141030 22:12:22< mattsc> iceiceice: okay; what do you want me to do in that case?
20141030 22:12:56< mattsc> I was going to do one more test on my old dusty 10.5 system at home tonight and then I have a whole bunch of changes to commit
20141030 22:13:17< iceiceice> so i guess we have to decide if we want it to be a general scons option not to require SDL mixer 1.2.12
20141030 22:13:29< iceiceice> i'm inclined to say, no
20141030 22:13:55< iceiceice> and then what i would suggest is, instead of if !defined(PANDORA) && !defined(__APPL__)
20141030 22:14:10< iceiceice> put like if !defined(PANDORA) && !defined(OLD_SDL_MIXER)
20141030 22:14:19< iceiceice> or even  if !defined(PANDORA) && !defined(MATTSC)
20141030 22:14:32< iceiceice> and then put define whatever on that file int he projct file
20141030 22:14:41< iceiceice> idk
20141030 22:14:51< iceiceice> is there some way to detect if you are compiling on OSX 10.5?
20141030 22:14:55 * mattsc is never particularly defined, at any time or location 
20141030 22:15:03< iceiceice> hehe
20141030 22:16:01< iceiceice> another idea is make it "ifndef OLD_SDL_MIXER" and then i change the cmake recipe to do this in case of pandora
20141030 22:16:01< mattsc> iceiceice: probably, but I am not going to set up two different sets of libraries for Xcode for different versions of OS X
20141030 22:16:15< iceiceice> do you compile wesnoth on 10.5 also?
20141030 22:16:20< mattsc> no
20141030 22:16:22< iceiceice> or do you compile on 10.9? i dont remember what you said now
20141030 22:16:23< iceiceice> yeah ok
20141030 22:16:52< mattsc> I always compile on 10.9, I just use the 10.5 SDKs for release builds
20141030 22:17:03< Dugi> Guys, have any of you tried to compile these new versions on Ubuntu? When I tried to google the issues I had, these were the results http://stackoverflow.com/questions/7972314/c-boostfilesystem-undefined-reference-to-boostfilesystem3pathroot-nam and http://stackoverflow.com/questions/21750547/undefined-reference-to-boostlocaleconvbetween Looks like if somebody left a linker error somewhere.
20141030 22:17:41< iceiceice> Dugi: i use linux mint, which is derived from ubuntu
20141030 22:17:56< Dugi> iceiceice: And it compiles without issues?
20141030 22:18:05< iceiceice> yup
20141030 22:18:09< bumbadadabum> I compiled it just fine on mint as well
20141030 22:18:14< iceiceice> well wait are you using scons or cmake
20141030 22:18:26< iceiceice> for cmake i have linker errors atm,
20141030 22:18:46< iceiceice> cmake is a pita and often not maintained very well
20141030 22:18:52< bumbadadabum> ^
20141030 22:18:57< Dugi> iceiceice: Any idea what might be causing this? http://pastebin.com/gTaYPWas
20141030 22:19:03< shadowm> iceiceice: cmake is what Ivanovic uses, so you'd better maintain it well.
20141030 22:19:15< shadowm> ;)
20141030 22:19:56< Dugi> iceiceice: So somebody neglected cmake? The problem is that if I fixed it, it would be fixed on master shortly after and it would mess up my branch,
20141030 22:20:09< iceiceice> if you fixed it i would merge your pull request
20141030 22:20:29< iceiceice> one thing that pisses me off about cmake is, i have no idea how to make it use build parallelism
20141030 22:20:30< iceiceice> or cmake
20141030 22:20:36< iceiceice> so it always takes an hour
20141030 22:20:38< shadowm> make -j8
20141030 22:20:50< iceiceice> ok, i'm glad to know that now,
20141030 22:20:54< Dugi> iceiceice: So you mean making ANOTHER pull request for this fix?
20141030 22:21:01< iceiceice> Dugi: sure
20141030 22:21:12< bumbadadabum> how many -j's do you use?
20141030 22:21:14< iceiceice> you see how many pull requests I have open right now? XD
20141030 22:21:18< bumbadadabum> I personally compile with -j3
20141030 22:21:21< iceiceice> i use j3 also
20141030 22:21:31< iceiceice> shadowm: i still feel the documentation for cmake is horrific
20141030 22:21:42< iceiceice> and scons is so simple that i didnt require docs for the first 6 months
20141030 22:21:44< shadowm> iceiceice: You had a <comma> above, what was the rest you wanted to tell me? :p
20141030 22:22:07< Dugi> I used to use -j3 too, but then I got a machine with a bigger RAM and I can use -j4 (no need to have more jobs on this i7 laptop).
20141030 22:22:09< iceiceice> also when i finally did look at cmake docs, they mention all this stuff about "undefined behavior may occur if..."
20141030 22:22:14< iceiceice> i mean wtf are they thinking
20141030 22:22:16< shadowm> iceiceice: The -j switch being a make thing is common knowledge, actually. scons just borrowed the syntax, and it's also known that cmake is just a makefile generator.
20141030 22:22:22< iceiceice> why can the cmake language have undefined behavior.
20141030 22:22:39< iceiceice> these people just have vastly different goals in mind from what i would want from a build tool
20141030 22:22:40 * shadowm uses up to -j16.
20141030 22:23:00< bumbadadabum> shadowm: how much ram did you download for that
20141030 22:23:13< shadowm> Like 1 TiB.
20141030 22:23:22< iceiceice> i thought its not supposed to be more than your number of cores?
20141030 22:23:23< shadowm> No, I have 16 GiB installed.
20141030 22:23:34< shadowm> iceiceice: Yeah, it's not. Don't try this at home.
20141030 22:23:43< bumbadadabum> iceiceice: obviously
20141030 22:23:43< iceiceice> wont it just be slower?
20141030 22:23:48< iceiceice> whats the point
20141030 22:23:53< bumbadadabum> yeah, because it has to thread everything
20141030 22:23:55< bumbadadabum> IIRC
20141030 22:24:09< bumbadadabum> (if you do too many -j's)
20141030 22:24:12< shadowm> I haven't objectively perceived any advantages or disadvantages to using more than 8.
20141030 22:24:22< bumbadadabum> it just stalls more if you use more
20141030 22:24:32< shadowm> Not within that range and on this machine. Obviously any of my laptops would go comatose with that.
20141030 22:24:42< bumbadadabum> yeah I tried -j4 once
20141030 22:24:49< shadowm> Unless I was compiling e.g. Linux, which is more CPU-friendly during builds.
20141030 22:24:56< bumbadadabum> and it made my computer completely freeze
20141030 22:25:25< Dugi> Why the hell I can't even find a mention of boost in cmake files?
20141030 22:25:27< shadowm> Wesnoth instead turns the compiler into a resource hog with a myriad template instantiations and convoluted cookie-cutter code (BOOST_FOREACH etc.).
20141030 22:27:10< iceiceice> Dugi: https://github.com/wesnoth/wesnoth/blob/master/CMakeLists.txt#L24
20141030 22:27:15< Dugi> Ah, found it, I was looking at the wrong folder.
20141030 22:28:38< mattsc> iceiceice: so how about I use the not-Apple define right now, and you can change it to not-Xcode or something if/when you get the homebrew stuff figured out?
20141030 22:28:58< iceiceice> sure, but the homebrew stuff is "live" now
20141030 22:29:13< iceiceice> in the sense that ancestral made a forum post, and i was going to make a PR to homebrew
20141030 22:29:21< mattsc> didn’t you just say that you still have things to figure out?
20141030 22:29:28< iceiceice> yeah we could still figure out the bundle crap
20141030 22:29:35< iceiceice> but i dont think its strictly necessary
20141030 22:29:45< iceiceice> its building fine with no errors, 
20141030 22:29:53< iceiceice> just install and "brew install [wesnoth-link]
20141030 22:29:53< iceiceice> "
20141030 22:31:56< mattsc> Okay, well then just tell me what to use.  Or you could do it yourself if you prefer.  (I have no idea where to set up a define like that, but I bet it’s not all that hard to figure out.)
20141030 22:32:14-!- kex [~kex@92.53.30.12] has quit [Remote host closed the connection]
20141030 22:33:08< iceiceice> ok well actually, as a hack,
20141030 22:33:13< iceiceice> if you dont want to do any source code changes,
20141030 22:33:19< iceiceice> you could just put "define PANDORA" on sound.cpp
20141030 22:33:23< iceiceice> in your project
20141030 22:33:25< iceiceice> and see if it works
20141030 22:33:46-!- kex [~kex@92.53.30.12] has joined #wesnoth-dev
20141030 22:34:05< iceiceice> i dont think thers other pandora stuff there, let me chekc though
20141030 22:34:41< mattsc> okay, I’ll try that later just to check that, but I’ll do a ‘define Xcode’ or something for the real version.
20141030 22:35:28< iceiceice> oh
20141030 22:35:30< iceiceice> i just realized,
20141030 22:35:38< iceiceice> later in sound.cpp theres some other code to spit out the SDL versions
20141030 22:36:02< iceiceice> i guess maybe should just write a check "if SDL mixer 1.2.12 ... (do the 1.2.12 thing) ... otherwise ... (do the old thing)"
20141030 22:36:09< iceiceice> and drop the ifdef pandora in code
20141030 22:36:25< shadowm> You can check SDL_mixer versions at compile time.
20141030 22:37:24< iceiceice> mattsc: so wait are you using the old header or the new header for SDL_mixer ?
20141030 22:38:21< mattsc> iceiceice: the old one
20141030 22:38:27< iceiceice> ok, 
20141030 22:38:32-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20141030 22:38:34< iceiceice> i think this is probably what i should have done in the first place anyways
20141030 22:38:41< mattsc> SDL_mixer was the problem for 10.5, the boost libs were pretty easy to figure out
20141030 22:39:35< mattsc> iceiceice: okay, are you going to change that then?
20141030 22:39:44< iceiceice> yeah, give me a sec
20141030 22:39:50< mattsc> or do you want me to?
20141030 22:39:56< iceiceice> if you want to you can,
20141030 22:40:02< iceiceice> you'll have to look up some crap on the sdl website
20141030 22:40:28< iceiceice> i already looked at it earlier for unrelated reasons
20141030 22:40:41< mattsc> okay, then go ahead, I won’t fight you for it.  :)
20141030 22:41:59< mattsc> should I commit my Xcode changes first?
20141030 22:42:16< mattsc> It shouldn’t really matter, should it?
20141030 22:42:34-!- gfgtdf_ [~chatzilla@d108222.adsl.hansenet.de] has joined #wesnoth-dev
20141030 22:43:41< mattsc> iceiceice: btw, I figured out how to set up preprocessor defines in Xcode, so if we ever want to distinguish between Xcode builds and homebrew builds, that’s no problem at all.
20141030 22:44:41-!- gfgtdf [~chatzilla@f054141011.adsl.alicedsl.de] has quit [Ping timeout: 260 seconds]
20141030 22:44:43-!- gfgtdf_ is now known as gfgtdf
20141030 22:57:18< iceiceice> mattsc: yeah do whatever you like :)
20141030 23:00:07-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has joined #wesnoth-dev
20141030 23:05:04-!- fabi [~quassel@p20030051AA26F7073003F9D556AD7749.dip0.t-ipconnect.de] has joined #wesnoth-dev
20141030 23:05:04-!- fabi [~quassel@p20030051AA26F7073003F9D556AD7749.dip0.t-ipconnect.de] has quit [Changing host]
20141030 23:05:04-!- fabi [~quassel@wesnoth/developer/fendrin] has joined #wesnoth-dev
20141030 23:16:05-!- zookeeper [zookeeper@wesnoth/developer/zookeeper] has quit [Ping timeout: 264 seconds]
20141030 23:18:42< Dugi> Trying to compile it using scons, remembered why I was using cmake. scons have no progress bar.
20141030 23:19:18< shadowm> cmake's progress indicators are only marginally more reliable than conspiracy theorists.
20141030 23:19:53< Dugi> Cool wording.
20141030 23:21:08< Dugi> Still, better than nothing. WIth cmake I can observe that it gets to 70% quickly and then it will continue with a constant speed, on scons this is like... no idea if it's gonna end in a few seconds or in an hour.
20141030 23:21:23< Dugi> How do I enable parallelisation there?
20141030 23:21:34< shadowm> scons -j<N>
20141030 23:21:59< Dugi> Same as in cmake? Good to know.
20141030 23:22:03< Dugi> Thanks.
20141030 23:23:34< fabi> hi Dugi
20141030 23:23:48< fabi> hi shadowm, ancestral, iceiceice
20141030 23:24:18-!- mjs-de [~mjs-de@f048232042.adsl.alicedsl.de] has joined #wesnoth-dev
20141030 23:25:14< shadowm> Hi fabi.
20141030 23:25:23< enchilado> Hi shadowm.
20141030 23:25:46< shadowm> Bye enchilado.
20141030 23:25:48-!- Anakonda [Anakonda@87-92-148-120.bb.dnainternet.fi] has quit [Read error: Connection reset by peer]
20141030 23:25:59< iceiceice> salutations fabi
20141030 23:26:17< enchilado> :(
20141030 23:27:33< fabi> mattsc: I have a fix for the "upkeep bug". But it is not assigned to me. Do you have a link?
20141030 23:27:46< gfgtdf> fabi: please mark http://gna.org/bugs/?22881 as fixed when you have tested it on 1.12
20141030 23:28:32< fabi> gfgtdf: Yes, I have a TODO folder in my browser's bookmark and there is already one pointing to that bug.
20141030 23:28:59< mattsc> fabi: great!  https://gna.org/bugs/?22806  (it appears assigned to you)
20141030 23:29:07< fabi> mattsc: Oh sorry.
20141030 23:29:24< mattsc> no worries  :)
20141030 23:29:37< iceiceice> ancestral: i broke the script somehow i think
20141030 23:29:50< iceiceice> err
20141030 23:29:54< iceiceice> i guess i didnt change the script
20141030 23:30:03< iceiceice> but i'm nwo testing my pull request and it doesnt work
20141030 23:31:04< Dugi> Hi, fabi.
20141030 23:33:33< irker657> wesnoth: Fabian Müller wesnoth:1.12 08d02dda3d73 / changelog data/themes/default.cfg: * Fix bug #22806: missing mouseover info: upkeep http://git.io/w2w6qQ
20141030 23:35:01< mattsc> fabi: one more thing I keep forgetting: could you retroactively add the LoW changes to the 1.11.18 changelog?
20141030 23:35:21< mattsc> If you don’t have time, I could do it too, but you did them, so you’re more familiar with the details.
20141030 23:36:10< fabi> mattsc: Yes, let me do it.
20141030 23:36:16< mattsc> thanks
20141030 23:36:40-!- ancestral [~ancestral@71-34-14-121.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20141030 23:39:46-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20141030 23:40:18< shadowm> fabi: No bullet points in commit message summaries, please.
20141030 23:40:42< fabi> shadowm: Why not?
20141030 23:41:04< shadowm> Because you don't want to be chastised for it next time the repository gets rewritten in a decade or so.
20141030 23:41:30< shadowm> If you absolutely need a bulleted list it must be in the message body, not the summary.
20141030 23:42:15< fabi> okay
20141030 23:43:08< fabi> shadowm: I did not want it. Copy and Paste Error.
20141030 23:46:30< fabi> mattsc: https://gna.org/bugs/?16600
20141030 23:46:52< fabi> mattsc: I don't get it. He is talking about which Scenario? 
20141030 23:47:10< mattsc> fabi:  “Ok, so I managed to (buggily) win the level, and it wasn't the final level.”
20141030 23:47:23< mattsc> He’s talking about Human Alliance
20141030 23:47:23< shadowm> GAH.
20141030 23:47:33< shadowm> AtS E1S6.2 crashes the game at the end.
20141030 23:47:51< fabi> mattsc: Oh yes. The save seems to be from 7.
20141030 23:48:21< fabi> mattsc: Are you sure it is about 14?
20141030 23:48:51< fabi> mattsc: Poor boy. 7 is exactly 1/3 of the campaign. Far away from final.
20141030 23:49:11-!- mjs-de [~mjs-de@f048232042.adsl.alicedsl.de] has quit [Remote host closed the connection]
20141030 23:49:29< shadowm> Whoever caused this will pay with their blood.
20141030 23:49:41< fabi> shadowm: Are you awake?
20141030 23:49:48< mattsc> Hmm, fabi, you’re right ...
20141030 23:49:53< shadowm> Nope, I'm dead sleep right now.
20141030 23:49:58< shadowm> *asleep
20141030 23:50:19< mattsc> I’m confused now, he probably does me S7.  But fortunately the comment I made applies there too.  :)
20141030 23:50:32< fabi> Dreaming about greed and violence.
20141030 23:50:45< fabi> mattsc: rofl
20141030 23:52:24< fabi> mattsc: Are there fixes made by you with still missing changelog entries?
20141030 23:52:44< mattsc> fabi: yes, there are, but not for 1.11.18
20141030 23:52:53< fabi> okay
20141030 23:53:33< mattsc> I’ll add them to the changelog later, don’t worry about those
20141030 23:54:15< shadowm> Well, great, it's a heisenbug.
20141030 23:54:56-!- Dugi [93fbd29f@gateway/web/freenode/ip.147.251.210.159] has quit [Quit: Page closed]
20141030 23:55:14< shadowm> Why are libogg/libvorbis involved.
20141030 23:55:31< shadowm> :\
20141030 23:56:07< shadowm> http://pastebin.com/BckcABSv -- No, I don't have a better backtrace yet because it doesn't affect -O0 builds.
20141030 23:57:30< mattsc> that looks like it might be caused by the new filesystem code ...
20141030 23:57:48< shadowm> But it happens on 1.11.18 too, which doesn't any of those patches AFAIR.
20141030 23:58:13< mattsc> true  (I think)
20141030 23:58:33< shadowm> I'm running this through valgrind before checking with 1.11.17.
20141030 23:59:53-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
--- Log closed Fri Oct 31 00:00:34 2014