--- Log opened Wed Mar 27 00:00:46 2013
20130327 00:01:38< shadowm> mordante: Last time I checked, the topic said "waiting on mordante for 1.11.2" or something to that effect for days.
20130327 00:05:32-!- exciton_ [chuck-the-@89.208.169.104] has quit [Ping timeout: 252 seconds]
20130327 00:07:17-!- exciton [chuck-the-@89.208.169.104] has joined #wesnoth-dev
20130327 00:10:17< shadowm> Ivanovic: And finally restricted the development servers to accept only 1.11.2.
20130327 00:18:39-!- _trewe_ [~trewe@87.196.159.196] has quit [Quit: Quit]
20130327 00:25:49< anonymissimus> esr: that error seems to come at random (on Linux less than on windows I think); strategies other than trying again ?
20130327 00:26:57< anonymissimus> although I should have way enough memory available in the page file/swap space
20130327 00:31:23< Gambit> A call for opinions and feedback from those who develop and maintain mainline campaigns: http://forums.wesnoth.org/viewtopic.php?f=2&t=38574
20130327 00:31:25< Gambit> zookeeper^
20130327 00:31:33< Gambit> esr, Espreon^
20130327 00:34:05< anonymissimus> esr: some people suggest to reduce the number of threads http://git.661346.n2.nabble.com/git-gc-exit-with-out-of-memory-malloc-failed-error-td6795172.html
20130327 00:35:29< janebot_> Wesnoth Forums | Developers’ Discussions • A question for mainline campaign developers and maintainers by Gambit [ 03-26-2013 23:30 ] [ http://r.wesnoth.org/p550390 ]
20130327 00:36:19-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130327 00:36:45-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bye]
20130327 00:40:04< anonymissimus> zookeeper: just to be sure, *pls* don't remove the data\core\images\maps\wesnoth.png since you would break UMC :)
20130327 00:46:58-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection]
20130327 00:46:59< esr> anonymissimus: I had threads set to 1.
20130327 00:47:10< AI0867> what was the window size?
20130327 00:47:41< AI0867> I've successfully run it on baldras with a window of 1250 and a depth of 250
20130327 00:49:42< AI0867> esr: I notice that you've checked out master on the repo. The .git dir itself is now 1.8 GiB
20130327 00:51:16< esr> I used 1250/250 as well.
20130327 00:51:45< esr> Hm, how did you measure the size without the checkout files?
20130327 00:52:38< AI0867> du -hs wesnoth-git/.git
20130327 00:52:55< AI0867> or you can just delete the checkout files
20130327 00:53:01< shadowm> Someone apparently needs coffee.
20130327 01:01:10-!- mattsc [~mattsc@fw.hia.nrc.ca] has quit [Quit: bye]
20130327 01:01:53< esr> AI0867: Thanks, now I can be sure our figures are consistent.
20130327 01:05:37-!- _8680a [~androirc@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 01:08:05< shadowm> esr: And talking about editing commit messages, we should probably get rid of the $Id$ lines in source code files, or did you take that into account already?
20130327 01:08:41< shadowm> The alternative is using some git feature whose name escapes me at the moment, but I am not convinced $Id$ lines were useful to begin with.
20130327 01:08:55< esr> shadowm: I think those are already gone, but I'll check.
20130327 01:09:10< shadowm> Ah right, gitattributes, that's the feature.
20130327 01:09:14< esr> I agree, they're not very useful.
20130327 01:12:02-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 01:12:57< esr> There are a few left.
20130327 01:30:56< esr> Actually, using a different search, there are a crapload of them.
20130327 01:31:05< esr> I'll get rid of them.
20130327 01:33:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20130327 01:34:00-!- Samual [diotecktec@xonotic/core-team/Samual] has quit [Quit: The person who said nothing is impossible obviously never tried to slam a revolving door.]
20130327 01:55:08-!- anonymissimus [~chatzilla@HSI-KBW-078-042-163-136.hsi3.kabel-badenwuerttemberg.de] has quit [Quit: ChatZilla 0.9.90 [Firefox 11.0/20120312181643]]
20130327 01:58:42-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has joined #wesnoth-dev
20130327 02:35:25< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by beetlenaut [ 03-27-2013 01:28 ] [ http://r.wesnoth.org/p550397 ]
20130327 02:37:05-!- Gallaecio_ [~quassel@84.120.218.224.dyn.user.ono.com] has quit [Ping timeout: 255 seconds]
20130327 02:42:13-!- Gallaecio [~quassel@84.120.218.224.dyn.user.ono.com] has joined #wesnoth-dev
20130327 02:50:06-!- mattsc [~mattsc@d154-20-34-165.bchsia.telus.net] has joined #wesnoth-dev
20130327 02:55:21< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by shadowmaster [ 03-27-2013 01:47 ] [ http://r.wesnoth.org/p550401 ]
20130327 02:55:22< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by Gambit [ 03-27-2013 01:53 ] [ http://r.wesnoth.org/p550402 ]
20130327 02:59:21-!- Octalot [~noct@host86-146-156-9.range86-146.btcentralplus.com] has quit []
20130327 03:07:15< esr> I'm beginning the git repo upload.
20130327 03:07:58-!- Samual [diotecktec@xonotic/core-team/Samual] has joined #wesnoth-dev
20130327 03:08:44< shadowm> \o/
20130327 03:09:29< vultraz> wheee!
20130327 03:10:08-!- _8680a [~androirc@ip68-225-249-82.oc.oc.cox.net] has quit [Remote host closed the connection]
20130327 03:10:36-!- _8680a [~androirc@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 03:19:24< esr> SourceForge's web interface is indexing the repo.
20130327 03:19:42-!- loonybot [~loonybot@wesnoth/bot/loonybot] has quit [Remote host closed the connection]
20130327 03:19:55< shadowm> esr: Did you already finish the upload?
20130327 03:20:41< esr> Yes.  IU want to do a sanity check before announcing it.
20130327 03:20:51-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130327 03:21:25< esr> Things are looking promising.
20130327 03:22:25< shadowm> I want a sanity check before dealing with the consequences, but everyone here is as crazy as I am, or worse.
20130327 03:22:34< shadowm> esr: What will be done about irker?
20130327 03:23:16-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20130327 03:23:25< esr> shadowm: Good question.  First thing I'll try is to beat SF into officially supporting it.
20130327 03:23:56< shadowm> esr: How about we don't hold our breath on that?
20130327 03:24:31< esr> shadowm: OK.  What alternative do you suggest?
20130327 03:24:34< shadowm> esr: Would it be possible for you to install a hook to prod ai0867.net? IIRC the instance we run here is public.
20130327 03:24:40< shadowm> May want to ask AI0867 about it.
20130327 03:25:01< vultraz> What's the problem with irker?
20130327 03:25:04< esr> Yes, that probably would be possible.  
20130327 03:25:48< esr> vultraz: It's inviting for IRC spammers. Ideally you swant it to only be accessible to hook scriots running inside your firewall.
20130327 03:27:54-!- _8680a_ [~androirc@mdf0436d0.tmodns.net] has joined #wesnoth-dev
20130327 03:27:56< shadowm> So, ask AI0867 about making an arragement for ai0867.net. He has root access there (duh).
20130327 03:28:08< shadowm> Or more specifically here.
20130327 03:28:38-!- _8680_ [~quassel@ip68-225-249-82.oc.oc.cox.net] has quit [Ping timeout: 256 seconds]
20130327 03:28:39< shadowm> Alternatively we might be able to do something on baldras.
20130327 03:29:22< esr> Both baldras and ai0867.net have the same issue.  We have to pray spammers don't find them.
20130327 03:29:36< vultraz> What is this baldras you two keep speaking of
20130327 03:29:59< shadowm> vultraz: Nothing of your concern yet.
20130327 03:30:15< shadowm> It really only concerns the wesnoth.org staff atm.
20130327 03:30:21< shadowm> That is, including me.
20130327 03:30:36< shadowm> esr: Didn't you write irker-filter?
20130327 03:30:54< esr> No.  What is it?
20130327 03:30:59< shadowm> That's a thing that runs on ai0867.net and IIRC it has something to do with public irkers.
20130327 03:31:09< shadowm> Okay, great, we are both clueless.
20130327 03:31:19-!- _8680a [~androirc@ip68-225-249-82.oc.oc.cox.net] has quit [Ping timeout: 246 seconds]
20130327 03:31:41-!- _8680a_ is now known as _8680a
20130327 03:33:59-!- Gambit [~gambit@wesnoth/developer/grickit] has quit [Quit: Balls]
20130327 03:35:02< esr> Still waiting for SF to finish the repo analysis.
20130327 03:45:11-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20130327 03:51:23-!- _8680_ [~quassel@ip68-4-203-69.oc.oc.cox.net] has joined #wesnoth-dev
20130327 03:55:24-!- Gallaecio [~quassel@84.120.218.224.dyn.user.ono.com] has quit [Ping timeout: 264 seconds]
20130327 04:00:03-!- Gallaecio [~quassel@84.120.218.224.dyn.user.ono.com] has joined #wesnoth-dev
20130327 04:05:34-!- wesbot changed the topic of #wesnoth-dev to: git conversion started, please keep svn commits to a minimum! | 171 bugs, 332 feature requests, 20 patches | Logs: http://irclogs.wesnoth.org | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20130327 04:06:27-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Read error: Connection reset by peer]
20130327 04:11:07-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130327 04:53:19-!- Samual [diotecktec@xonotic/core-team/Samual] has quit [Ping timeout: 264 seconds]
20130327 04:54:48-!- Guest44736 [~eli@66-189-34-122.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 264 seconds]
20130327 05:07:05-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20130327 05:09:50-!- Samual [diotecktec@xonotic/core-team/Samual] has joined #wesnoth-dev
20130327 05:11:10< esr> *** ANNOUNCEMENT:  THE GIT REPO IS OPEN ****
20130327 05:11:27< esr> git clone ssh://esr@git.code.sf.net/p/wesnoth/code wesnoth-code
20130327 05:11:35< esr> No.
20130327 05:12:35< esr> git clone git://git.code.sf.net/p/wesnoth/code wesnoth-code
20130327 05:12:40-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has joined #wesnoth-dev
20130327 05:15:01< esr> Looks like what SF wants is git clone ssh://FOO@git.code.sf.net/p/wesnoth/code wesnoth-code where FOO is your user ID.
20130327 05:15:24< esr> I'm going to try a read-only clone now.
20130327 05:35:22< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by beetlenaut [ 03-27-2013 04:32 ] [ http://r.wesnoth.org/p550404 ]
20130327 05:35:51-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 05:40:49-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has quit [Remote host closed the connection]
20130327 05:41:15-!- mgla [~shishir@unaffiliated/noobjoe] has joined #wesnoth-dev
20130327 05:42:50< esr> Commits to the new repo are working.
20130327 05:46:15< vultraz> yay!
20130327 05:46:57< vultraz> now to begin the long task of cloning and hoping that my internet doesn't get interrupted by phone calls, power outages/flickers, or accidental unplugs
20130327 05:52:37< shadowm> vultraz: Be patint.
20130327 05:52:38< shadowm> Patient.
20130327 05:52:42< shadowm> We need to figure this out first.
20130327 05:57:24< vultraz> Though, why do I want a clone since I can't bloody compile :S
20130327 05:58:06< vultraz> (well, I could find another way besides msvc, but it's slow as a jello-encased snail)
20130327 05:58:21< vultraz> (who's crawling over sticky paper)
20130327 05:58:40< vultraz> So maybe I should just wait until I get a new laptop
20130327 06:00:27-!- _8680_ [~quassel@ip68-4-203-69.oc.oc.cox.net] has quit []
20130327 06:08:18-!- ancestral [~ancestral@75-168-48-55.mpls.qwest.net] has quit [Quit: And that’s the end of THAT chapter.]
20130327 06:26:19-!- _8680a_ [~androirc@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 06:27:01-!- _8680a [~androirc@mdf0436d0.tmodns.net] has quit [Ping timeout: 246 seconds]
20130327 06:30:08-!- _8680_ [~quassel@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 06:30:48-!- Blueblaze [~Blueblaze@adsl-99-158-47-208.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20130327 06:39:15-!- Blueblaze [~Blueblaze@adsl-99-158-47-208.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20130327 06:46:25-!- mattsc [~mattsc@d154-20-34-165.bchsia.telus.net] has quit [Quit: bye]
20130327 06:53:14< esr> And it's done.  I think.
20130327 06:53:27< fabi> esr: hmmm
20130327 06:53:44< esr> fabi: ?
20130327 06:55:10< fabi> esr: It works for me on the commandline, at least it starts the process.
20130327 06:55:36< fabi> But it's not working with my eclipse git connector.
20130327 06:55:53< fabi> I first thought it wouldn't work at all.
20130327 06:56:33< shadowm> esr: How large is the clone?
20130327 06:57:16< esr> 1.5GB.
20130327 06:57:36< shadowm> 14:05:49 <Ivanovic> shadowm, esr, vultraz: why not host the tarball of git on sf.net in the files section?
20130327 06:57:59< shadowm> My only other options are abusing ai0867.net, wesnoth.org, or B's bandwidth twice.
20130327 06:58:24< shadowm> Once for cloning on one of those three servers.
20130327 06:58:55< shadowm> Twice for downloading to my machine with multiple rsync or wget invocations over the course of at least 4 hours.
20130327 06:59:32-!- Blueblaze [~Blueblaze@adsl-99-158-47-208.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20130327 07:01:18< esr> I'm going to sleep.  It's been a long couple of days.
20130327 07:01:46< shadowm> Ugh.
20130327 07:02:03-!- Blueblaze [~Blueblaze@adsl-99-158-47-208.dsl.hstntx.sbcglobal.net] has joined #wesnoth-dev
20130327 07:03:00< shadowm> AI0867, Ivanovic: Should I abuse ai0867.net or wesnoth.org or B?
20130327 07:03:16< shadowm> Since it looks like esr doesn't want to give me an answer.
20130327 07:04:33< esr> shadowm: It's not that I don't want to give you an answer, I just don't have one.  Don't know what the tradeoffs are.
20130327 07:04:59< shadowm> esr: The tradeoff is that you may lose the only person who both hates and cares for wmllint atm.
20130327 07:05:03< shadowm> :)
20130327 07:06:13< shadowm> According to my calculations I've still got 5 GiB before exceeding my 20 GiB quota.
20130327 07:06:26< shadowm> Sooo... maybe I could try the direct method.
20130327 07:07:26-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Quit: free all the bandwidth]
20130327 07:09:01< fabi> shadowm: My internet is fast and rock solid but I already got a "remote connection closed" after 3% of the clone.
20130327 07:09:17< shadowm> Well, that could be SourceForge.net at work.
20130327 07:09:33< shadowm> Its unreliability has never stopped being a thing that is true and annoying as hell.
20130327 07:09:50< shadowm> Actually, if that's the case I simply won't bother.
20130327 07:09:53-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130327 07:09:59< fabi> Well, If anybody had asked me I would have chosen github.
20130327 07:10:18< shadowm> I would feel tempted to murder esr if I got to 99%, the connection failed, and then had to try again.
20130327 07:10:29< fabi> I had a wesnoth clone on github for over a year and they never complained beside the fact that it was beyond the soft limit.
20130327 07:10:30< shadowm> No, not just tempted.
20130327 07:10:33< shadowm> Obliged.
20130327 07:23:32-!- Blueblaze [~Blueblaze@adsl-99-158-47-208.dsl.hstntx.sbcglobal.net] has quit [Quit: Blueblaze]
20130327 07:25:30< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by 8680 [ 03-27-2013 06:18 ] [ http://r.wesnoth.org/p550406 ]
20130327 07:32:50-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20130327 07:52:21< zookeeper> anonymissimus, yeah, i don't see any reason to remove wesnoth.png until 1.12.
20130327 07:55:30< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by zookeeper [ 03-27-2013 06:49 ] [ http://r.wesnoth.org/p550407 ]
20130327 08:11:28-!- stikonas [~gentoo@5.20.200.50] has joined #wesnoth-dev
20130327 08:11:28-!- stikonas [~gentoo@5.20.200.50] has quit [Changing host]
20130327 08:11:29-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20130327 08:15:56-!- mgla [~shishir@unaffiliated/noobjoe] has quit [Read error: No route to host]
20130327 08:33:55-!- mgla [~shishir@unaffiliated/noobjoe] has joined #wesnoth-dev
20130327 08:36:23-!- _Coffee [~david@ppp118-210-51-209.lns20.adl2.internode.on.net] has joined #wesnoth-dev
20130327 08:44:09-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130327 08:51:16-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 08:56:04< janebot_> Wesnoth Forums | Developers’ Discussions • Re: A question for mainline campaign developers and maintain by fabi [ 03-27-2013 07:46 ] [ http://r.wesnoth.org/p550410 ]
20130327 09:06:29-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20130327 09:07:26< fabi> The git cloning takes ages. It's doing 18 KiB/s.
20130327 09:09:57-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Read error: Operation timed out]
20130327 09:09:58< _Coffee> fabi: that's about 3 times as quick as Internet in Australia after it has rained :)
20130327 09:10:19< fabi> _Coffee: It rains in Australia?
20130327 09:10:34< _Coffee> of course :P
20130327 09:10:44< shadowm> Australia isn't all desert, you know.
20130327 09:11:42< bumbadadabum> shadowm: You're funny
20130327 09:12:11< fabi> I remembered the cloning from github was just a matter of a few minutes. Something is wrong.
20130327 09:12:37< shadowm> bumbadadabum: http://en.wikipedia.org/wiki/File:Australia-climate-map_MJC01.png
20130327 09:13:13< fabi> At least in those green regions it needs to rain every day.
20130327 09:13:20< shadowm> fabi: The thing that sucks is that SF.net's file release system is awesome because it has worldwide mirrors and stuff.
20130327 09:13:20< _Coffee> we all live on the coast because the big bit in the middle is all desert
20130327 09:13:41< shadowm> fabi: But code repositories aren't mirrored.
20130327 09:14:36< shadowm> And apparently they are all served from the same host.
20130327 09:14:56< fabi> The download will take that long that my internet connection gets reset meanwhile (happens every 12 hours). Thus I won't be able to clone the thing as well.
20130327 09:14:57< shadowm> It probably uses anycast internally or something, but it's still a bottleneck from our point of view.
20130327 09:15:23< _Coffee> don't copy from KDE and git: http://jefferai.org/2013/03/24/too-perfect-a-mirror/
20130327 09:21:16-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has left #wesnoth-dev []
20130327 09:26:15-!- _8680a_ [~androirc@ip68-225-249-82.oc.oc.cox.net] has quit [Remote host closed the connection]
20130327 09:26:33-!- _8680a_ [~androirc@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 09:28:00-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130327 09:51:26< fabi> _Coffee: interesting story
20130327 09:51:27< stikonas> Wesnoth won't have this problem.  KDE has lots of repositories, some of them  might not have even a single local clone. Wesnoth has just one  repository with many local clones
20130327 09:51:35< stikonas> _Coffee: ^^
20130327 09:52:53< fabi> It's up to 35 KiB/s now.
20130327 09:53:28< stikonas> fabi: 4.5 MB/s for me...
20130327 09:53:46< fabi> stikonas: Where are you?
20130327 09:53:57< stikonas> Lithuania. We have a very fast internet...
20130327 09:54:22< fabi> Yeah, it's not that bad here either. 
20130327 09:54:44< fabi> I can have 4 MB/s as well when the server is fast.
20130327 09:55:38< stikonas> According to http://netindex.com/ average speed of the Internet is the fastest in the world in my city (Vilnius) :)
20130327 09:56:45< fabi> :-)
20130327 10:01:34< fabi> Now it's 1.5 MiB/s. That seems right.
20130327 10:02:11< stikonas> I finished downloading :)...
20130327 10:02:37< stikonas> sorry. I did not think that I would slow down everybody else
20130327 10:02:54< fabi> rofl
20130327 10:03:10< stikonas> maybe I should allow people to clone from me...
20130327 10:05:02< fabi> stikonas: Have you worked with git before?
20130327 10:05:22< stikonas> fabi: yes.
20130327 10:05:51< stikonas> I actually have gitolite installed on my laptop. So I can even allow other people to clone from me
20130327 10:07:04< fabi> I have read some theoretic stuff about git and watched some videos where torwalds explains the concept.
20130327 10:07:29< fabi> Still no clue how the actual usage is like.
20130327 10:07:33< stikonas> have you seen this: http://git.or.cz/course/svn.html
20130327 10:07:48< fabi> Yes
20130327 10:07:50< stikonas> you make some changes first
20130327 10:08:08< stikonas> then you git add files you want to commit
20130327 10:08:27< stikonas> and then you do git commit -m "message"
20130327 10:08:34< fabi> pretty much like svn
20130327 10:08:41< stikonas> or equivallently, you can do git commi -a -m "message"
20130327 10:08:48< stikonas> yeah, pretty much
20130327 10:08:58< fabi> -a ?
20130327 10:09:05< fabi> auto add?
20130327 10:09:07< stikonas> add all midified files
20130327 10:09:30< stikonas> but after you commit, the changes are still in you local reposiroty, so you have to push them to sf.net with git push
20130327 10:09:50< stikonas> you might have to do "git push origin master" the first time you do it
20130327 10:10:30< stikonas> and similarly with git pull you fetch remote changes
20130327 10:11:09< fabi> I think I can follow this path. 
20130327 10:11:16< stikonas> you can also do "git pull --rebase"  vs "git pull"
20130327 10:11:26< fabi> Let's discuss a concrete working example that bothers me more.
20130327 10:11:29< fabi> Please.
20130327 10:11:33< stikonas> ok
20130327 10:12:03< fabi> Say I want to show shadowm what I recently did to the editor's gui. Because he is the gui guru and we somehow are a mad team on the purpose.
20130327 10:12:38< fabi> I clone from SF. That is in progress right now.
20130327 10:12:51< stikonas> yes
20130327 10:12:59< fabi> After I finished fetching I am going to do a branch for the new feature first, right?
20130327 10:13:12< stikonas> right
20130327 10:13:24< fabi> Let's say I call it wesnoth-editor
20130327 10:13:50< stikonas> then you do some coding there
20130327 10:13:54< fabi> Then I do my development and 
20130327 10:14:12< fabi> there the questions arrive:
20130327 10:14:32< fabi> Do I merge it localy back into master?
20130327 10:14:49< fabi> or do I push the branch to SF and merge it there?
20130327 10:15:00< stikonas> you can only merge locally
20130327 10:15:15< stikonas> there is no remote merging in git
20130327 10:15:37< stikonas> but you don't have to merge your code in order to show it to shadowm
20130327 10:15:49< stikonas> you can first push the branch to SF
20130327 10:16:02-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 252 seconds]
20130327 10:16:08< stikonas> and shadowm will pull it from SF
20130327 10:16:50< fabi> I should do that early on because git does otherwise no longer mean that my work is mirrored.
20130327 10:16:53< stikonas> or alternatively (probably more difficult) you pusblish your repository somewhere else and shadowm pulls changes from it
20130327 10:17:00-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 10:17:36< stikonas> fabi: it probably depends on you workflow...
20130327 10:17:50< stikonas> although it's probably a good idea to publish it
20130327 10:18:05< stikonas> unless it is an experiment and you are not sure if it would succeed
20130327 10:19:33< stikonas> you and other Wesnoth developers should decide which specific workflow you want to use. git allows different workflows
20130327 10:20:19< stikonas> e.g. I often use git for keeping track of LaTeX documents where I am the only person working on them...
20130327 10:20:50< stikonas> kernel people do changes and ask subsystem maintainers to pull from them (they don't have push righs)
20130327 10:21:01< stikonas> eventually Linus pulls from everybody else
20130327 10:21:23< lipkab> The 1.10 repo stays on Gna, right?
20130327 10:22:17< stikonas> strange, I can't see any version branches in git repository
20130327 10:22:29< stikonas> neither 1.10, not 1.8.
20130327 10:22:47< stikonas> that's very strange indeed 
20130327 10:22:50< fabi> Yes, I saw that as well.
20130327 10:23:56< stikonas> esr: is it intentional that there are no branches (1.10 1.8 1.6 ...) in git repository ?
20130327 10:27:21< shadowm> Hi.
20130327 10:27:32< shadowm> How about we avoid making topic branches upstream?
20130327 10:28:14< shadowm> (X.Y branches are maintenance branches, not topic branches.)
20130327 10:28:39< shadowm> stikonas: What does git branch -r show?
20130327 10:29:06< shadowm> fabi, stikonas : I would prefer it if people worked in their own private branches and published series of patches for me to see.
20130327 10:29:17< stikonas> shadowm: http://paste.kde.org/~stikonas/710372/
20130327 10:29:36< stikonas> I have checked it already with git branch -a
20130327 10:29:40< shadowm> stikonas: esr screwed up then.
20130327 10:29:53< stikonas> indeed
20130327 10:30:13< stikonas> it might be possible to redo the conversion and add those branches
20130327 10:30:15< shadowm> It's not like I wasn't expecting such a thing to happen anyway.
20130327 10:30:37< shadowm> fabi: Before anyone pushes anywhere I think we should really discuss what our branching policy will be with Ivanovic and esr.
20130327 10:30:38< stikonas> you can easilly add new branches without having to force push
20130327 10:31:07< shadowm> My opinion is that topic branches should not exist upstream unless they are really really really important for everyone to see and work together on them.
20130327 10:31:20< shadowm> e.g. Hypothetical GL branch.
20130327 10:31:29< shadowm> I mean a third one.
20130327 10:32:20< shadowm> People can use git-format-patch or what have you to publish their work.
20130327 10:32:38< shadowm> (Or alternatively, github repositories.)
20130327 10:32:45< stikonas> or you can install something like gitolite and publish your local clone
20130327 10:32:55< stikonas> github wouldn't accept large repositories
20130327 10:32:57-!- DCW [~Thunderbi@cpc1-finc14-2-0-cust12.4-2.cable.virginmedia.com] has joined #wesnoth-dev
20130327 10:33:23< shadowm> People have published Wesnoth on Github before.
20130327 10:33:23< fabi> stikonas: I had a wesnoth clone for over a year on github.
20130327 10:33:40< stikonas> ok, then.
20130327 10:33:57< shadowm> It's just that people decided to introduce FUD to force us to use SF.net.
20130327 10:34:11-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Ping timeout: 245 seconds]
20130327 10:34:18< fabi> shadowm: That was my idea as well.
20130327 10:34:22< shadowm> SourceForge.net may be open but they have quite a history of screwing up since 2008.
20130327 10:34:28< shadowm> *open source
20130327 10:34:46-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has quit [Read error: Connection reset by peer]
20130327 10:34:51< fabi> I mean the FUD thing, not the branching policy.
20130327 10:34:53-!- thunderstruck [~thunderst@cpc5-sgyl29-2-0-cust174.sgyl.cable.virginmedia.com] has joined #wesnoth-dev
20130327 10:35:51 * shadowm reads on gitolite.
20130327 10:36:34< shadowm> Sounds cool (from the Debian package description, anyway).
20130327 10:38:20-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130327 10:38:49< shadowm> I'll give it a try once I figure out how to keep track of people downloading from me so they don't accidentally cause the ultimate DoS scenario for me.
20130327 10:39:10< shadowm> *the amount of data people are downloading from me
20130327 10:39:24-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 10:41:00< fabi> hmmm, the process of getting a working eclipse git wesnoth configuration is not as easy as I thought.
20130327 10:45:45< AI0867> have 2 cookies involving the CIA bot been excised from the fortunes during the transition?
20130327 10:46:33-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 10:46:41< fabi> AI0867: Do you know about ircer? Does he handle git already?
20130327 10:47:29-!- mjs-de [~mjs-de@g224183147.adsl.alicedsl.de] has joined #wesnoth-dev
20130327 10:48:39< AI0867> esr, shadowm: irker-filter is something I wrote to deal with forges with irker hooks that don't have a local irkerd. Its main advantage over a firewall entry is that it's content-aware
20130327 10:49:48< stikonas> AI0867: the transition seems to be incomplete anyway...
20130327 10:49:53< AI0867> regarding wesnoth, with gna we used shadowm's mail listener/svnpoller, but if we can retarget sourceforge's CIA hook, we can also use nenolod's CIA proxy
20130327 10:50:09< AI0867> shadowm: B has way more bandwidth
20130327 10:51:46< shadowm> AI0867: You will probably need to repeat the line about irker-filter for esr later.
20130327 10:51:52< shadowm> He tends to miss his scrollback.
20130327 10:52:44< shadowm> AI0867: I don't know if SourceForge has a working Git CIA hook or whether we can install our own..
20130327 10:52:52< shadowm> That's stuff that esr should know.
20130327 10:56:37< lipkab> fatal: git upload-pack: aborting due to possible repository corruption on the remote side.
20130327 10:56:41< lipkab> Cool.
20130327 10:57:54< shadowm> Peopple: Post in the mailing list in reply to esr's message.
20130327 10:58:05< shadowm> He is about 72% guaranteed to miss everything you say here.
20130327 10:58:13< shadowm> At least whenever he is not around.
20130327 11:00:32-!- _8680_ [~quassel@ip68-225-249-82.oc.oc.cox.net] has quit []
20130327 11:01:04< stikonas> can anybody post about these maintenance branches there. I'm not a wesnoth developer, so I'm not on wesnoth-dev mailing list
20130327 11:03:44-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130327 11:03:48< AI0867> fabi: it does not care what provides the input, but the sourceforge repo currently isn't set up to do so
20130327 11:04:02-!- irker423 [~irker@ai0867.net] has joined #wesnoth-dev
20130327 11:04:02< irker423> see?
20130327 11:04:02< shadowm> Exaclty how the hell do I see a list of branches on the web interface.
20130327 11:04:45< shadowm> This code browser interface is worse than the SVN one.
20130327 11:04:47< AI0867> I have no idea
20130327 11:05:14< AI0867> expected features of sourceforge:
20130327 11:05:21< AI0867> 1) crappy interface. Check
20130327 11:05:53< stikonas> it's still not too late to switch :)
20130327 11:06:18< shadowm> It's never too late to dump esr overboard and switch to GitHub, yeah.
20130327 11:06:40< shadowm> :)
20130327 11:08:16< shadowm> Sent the email.
20130327 11:09:22-!- rei4dan [~reinis@46.109.19.10] has quit [Quit: WeeChat 0.4.0]
20130327 11:10:13-!- mgla [~shishir@unaffiliated/noobjoe] has quit [Ping timeout: 248 seconds]
20130327 11:17:27< shadowm> AI0867: Explain this for us maybe? http://forums.wesnoth.org/viewtopic.php?p=550417#p550417
20130327 11:17:51< shadowm> ALthough I don't understand why they are mentioning dutch when their IP is from _Korea_.
20130327 11:19:43< AI0867> I think he's trying to talk about the various conventions regarding names
20130327 11:20:19< lipkab> He wants the credits list to be sorted by second name.
20130327 11:20:27< AI0867> that is, western names usually have a surname and a given name, and should be sorted by the surname
20130327 11:20:51< AI0867> but in dutch, the surname can be prefixed various particles that should not affect the sort order
20130327 11:21:11< AI0867> that is, my name is Alexander van Gessel, with the surname being "van Gessel", but I should still be sorted under G
20130327 11:21:56< shadowm> I think that doing this would overcomplicate everything and make it harder to (human-)parse the credits.
20130327 11:22:10< AI0867> korea (like various countries around it) has a completely different system
20130327 11:22:19< shadowm> I am aware.
20130327 11:22:27< AI0867> and then there's the fact that half of korea has the same family name
20130327 11:22:49< shadowm> I think our current solution is the lowest common denominator and suits us fine.
20130327 11:23:14< AI0867> well, the dutch rule is pretty simple: the prefixes aren't capitalized, so just sort by the first capital letter
20130327 11:24:07< AI0867> good luck finding a program that wasn't written specifically for the dutch market that implements that though
20130327 11:26:59 * AI0867 is currently cloning at 2.5 MiB/s
20130327 11:27:26< AI0867> though the speed moves around a lot
20130327 11:29:23< AI0867> shadowm: though I could do sequences-of-patches, I'd really rather like to have topic branches
20130327 11:31:08-!- DCW [~Thunderbi@cpc1-finc14-2-0-cust12.4-2.cable.virginmedia.com] has quit [Remote host closed the connection]
20130327 11:38:11< shadowm> fabi: You do realize that in order to have one forum for each scenario we'd need approximately 246 forums?
20130327 11:38:39< fabi> and that hurts why?
20130327 11:38:47< shadowm> Because that's 246 forums.
20130327 11:39:03< fabi> a forum is just a node in a database
20130327 11:39:13< fabi> nothing will explode
20130327 11:39:25< shadowm> I imagine the search index will become more complicated that way.
20130327 11:39:38< shadowm> Fuck, the search page itself will become irritatingly cramped.
20130327 11:39:46< AI0867> 11:39 < fabi> nothing will explode <-- I'm not so sure
20130327 11:40:02< AI0867> just think of the forum jump listbox
20130327 11:40:02< fabi> people use google for search because the forum search sucks anyway
20130327 11:40:12< shadowm> You are speaking for yourself there.
20130327 11:41:07< AI0867> you're trying to shoehorn what is not a bad idea in itself into a forum
20130327 11:41:43< AI0867> if you just want a Q&A format that is sorted by campaign and scenario, a simple form somewhere might be a better solution
20130327 11:41:51< AI0867> s/sorted/indexed/
20130327 11:42:10< shadowm> Anyway, later. I will veto that change (again, don't think I've forgotten you proposed this before) regardless of Gambit's opinion anyway.
20130327 11:42:19< AI0867> I can confirm that there are no maintenance branches on the sourceforge repo
20130327 11:42:24< AI0867> there are also no tags
20130327 11:42:33< AI0867> there are a few topic branches though
20130327 11:42:44< AI0867> esr: ^
20130327 11:43:06< shadowm> AI0867: emaaaaaaaaaaail.
20130327 11:43:33< shadowm> He is 85% guaranteed to miss all this if he doesn't see it on emails.
20130327 11:45:24< shadowm> In fact, that might be just as well.
20130327 11:45:46< shadowm> My suggestion to dump him overboard might not have been properly formulated in hindsight.
20130327 11:45:56< shadowm> I should write a whole essay on the matter.
20130327 11:46:28< fabi> I think you should watch your words more carefully.
20130327 11:46:55< fabi> This isn't funny anymore.
20130327 11:47:06< shadowm> I actually find it pretty funny.
20130327 11:47:42< fabi> Don't make me angry.
20130327 11:48:15< shadowm> Okay, sure, getting angry would not make anything funny.
20130327 11:48:24< shadowm> You need a sense of humor.
20130327 11:48:29< fabi> I am glad esr did the work, no matter if it takes one more run.
20130327 11:49:14< fabi> That is exactly what I would characterize your weakest point as well, lack of a sense of humor. But I do not want to start a flamewar right now.
20130327 11:49:15< shadowm> I was not being serious. When I am serious I generally make a point of listing my arguments one by one.
20130327 11:49:37-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 11:50:09-!- witness [uid10044@gateway/web/irccloud.com/x-kmepvsuuoqzwxsmb] has joined #wesnoth-dev
20130327 11:51:18< AI0867> sourceforge still only talks about shell, CVS and SVN on the SSH key page...
20130327 11:51:39< lipkab> Receiving objects: 100% (697384/697384), 1.46 GiB | 618 KiB/s, done.
20130327 11:51:56< lipkab> Hoorray!
20130327 11:52:04< lipkab> In mere three tries.
20130327 11:52:07 * shadowm steals lipkab's Internet.
20130327 11:52:16< AI0867> don't get too excited. You may yet have to delete it and start over
20130327 11:52:58< shadowm> Hopefully not? It's probably just that esr did a weird git push instead of rsyncing the repository over (assuming that's even possible).
20130327 11:53:38< AI0867> well, the tags being missing can be explained by that
20130327 11:53:51< AI0867> but the maintenance branches being missing, while various topic branches are present?
20130327 11:54:07< fabi> One strange thing is that eclipse insists in calling the project "code". I wonder if this information is from the git repository.
20130327 11:55:24< AI0867> if the data is actually present somewhere, a push can fix it, but if it's not, then the entire repo will have to be redone...
20130327 11:55:37 * AI0867 checks the repo on baldras
20130327 11:55:51< AI0867> it's gone...
20130327 11:56:47< shadowm> Is that where the one on SF.net came from?
20130327 11:57:03< shadowm> Or more specifically in this case, s/Is/Was/.
20130327 11:57:13-!- rei4dan [~reinis@46.109.19.10] has joined #wesnoth-dev
20130327 11:57:35< AI0867> it's the place where the repack was (re)done
20130327 11:57:43< fabi> And failed, again.
20130327 11:57:52< AI0867> as esr's machine didn't have enough RAM to repack efficiently
20130327 12:00:08-!- Crab_ [Crab_@wesnoth/developer/crab] has left #wesnoth-dev []
20130327 12:00:36< shadowm> 4 GiB, IIRC.
20130327 12:00:59< shadowm> Is the overzealous repack actually helpful though?
20130327 12:01:19< shadowm> For example, what will happen as we start pushing commits?
20130327 12:02:51< fabi> I guess it will get more and more packed with every commit. Like when you do more clothes in your Rucksack.
20130327 12:03:38< shadowm> I was actually expecting it to be the other way around, more and more unpacked.
20130327 12:03:53-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Ping timeout: 240 seconds]
20130327 12:03:53< shadowm> Otherwise why do local git repositories decide to repack every thousand commits or so?
20130327 12:04:45< zookeeper> _8680a_, btw, the guy is can barely scribble down a few words of english so i wouldn't make assumptions about whether his intent is to demand or politely suggest the credits change.
20130327 12:06:25< shadowm> zookeeper: I suggested that to him in the admin channel even before realizing that that person can't, in fact, write coherent English.
20130327 12:08:53-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 240 seconds]
20130327 12:10:36-!- loonybot [~loonybot@ppp158-255-133-34.pppoe.spdop.ru] has joined #wesnoth-dev
20130327 12:10:37-!- loonybot [~loonybot@ppp158-255-133-34.pppoe.spdop.ru] has quit [Changing host]
20130327 12:10:37-!- loonybot [~loonybot@wesnoth/bot/loonybot] has joined #wesnoth-dev
20130327 12:11:04-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 12:15:07< zookeeper> shadowm, right.
20130327 12:15:16-!- shadowm is now known as Pasta
20130327 12:15:25< esr> There are a bunch of topic branches in the repo
20130327 12:15:34-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Ping timeout: 256 seconds]
20130327 12:15:38-!- bumbadadabum is now known as Lasagna
20130327 12:15:46-!- Lasagna is now known as Anonymous
20130327 12:16:00-!- Anonymous is now known as bumbadadabum
20130327 12:16:21< Pasta> esr: The point is that the maintenance branches disappeared, as well as the tags.
20130327 12:17:15< Pasta> The maintenance branches are far more important than the topic branches, especially for our favorite release manager, and us wesnoth.org admins.
20130327 12:17:15< elias> git ls-remote git://git.code.sf.net/p/wesnoth/code
20130327 12:17:19-!- Pasta is now known as shadowm
20130327 12:17:28< elias> ^ that should list every existing branch and tag
20130327 12:17:55< AI0867> $ git ls-remote git://git.code.sf.net/p/wesnoth/code|wc -l
20130327 12:17:56< AI0867> 10
20130327 12:18:01< shadowm> http://pastebin.com/ezHPvPM5
20130327 12:18:04< esr> Pasta: They'r all there.  Holds on, I'll telll you how to materialize them all in a moment.
20130327 12:18:34< shadowm> I am curious how they can be there and yet not appear in ls-remote.
20130327 12:19:43< esr> Wait. I forgot some pushes.  Shall fix.
20130327 12:20:07< shadowm> Was rsyncing the repository over not an option?
20130327 12:20:40< shadowm> Originally, that is. Obviously it'd be a terrible idea right now.
20130327 12:20:52< shadowm> Also, how can I make the web interface not suck=
20130327 12:21:14< AI0867> by moving to github
20130327 12:21:15< shadowm> (Is this what I'm signing in for if I transfer wesnoth-umc-dev to Allura too?)
20130327 12:21:27< AI0867> afraid so
20130327 12:21:59< elias> i thought github refused to host wesnoth
20130327 12:22:07< fabi> It did not.
20130327 12:22:09< shadowm> AI0867: Okay, then you can tell Espreon that I'll not approve the opt-in upgrade until it becomes absolutely unavoidable for XYZ reason.
20130327 12:22:24< shadowm> Let's give SF.net a while to sort out their mess.
20130327 12:22:37< shadowm> Like they did with SVN repository UUIDs in January, apparently.
20130327 12:22:50< elias> fabi: well, then it should be easy for anyone to keep a clone or a few over there :)
20130327 12:23:06< fabi> elias: I had one for a year, deleted it yesterday.
20130327 12:23:17< elias> i see
20130327 12:23:22< AI0867> there's also all the mobile ports
20130327 12:23:26< AI0867> they're *all* on github
20130327 12:23:36< shadowm> Except for Kyle's.
20130327 12:23:50< AI0867> of course, I also very much doubt that they're up-to-date with the released binaries
20130327 12:23:51< shadowm> (It has not stopped being a thing that exists or anything.)
20130327 12:23:52< fabi> github already learned that wesnoth is that big and they do not even contact you, with a wesnoth fork.
20130327 12:24:35< esr> fabi: github *did* refuse.
20130327 12:25:04< shadowm> No, github was vague about the issue and we didn't pursue that path further.
20130327 12:25:28< esr> shadowm: Did you read the mail?  I saw a refusal there.
20130327 12:25:46< fabi> I have read it, it was posted on the mailing list, right?
20130327 12:25:48< shadowm> I am the forum administrator, I read the copy of the email posted there.
20130327 12:25:49< AI0867> shadowm: github.com/dodikk/iWesnoth is a mirror of kyle's repo
20130327 12:25:53< shadowm> fabi: No, Moderators'.
20130327 12:26:04< elias> anyway, the important step was the git switch - changing the repo URL in the future would be very easy
20130327 12:26:34< fabi> It sounded to my like: Please stay below 1G if you can, if not we will just tolerate it like we do already.
20130327 12:26:46< fabi> esr: ^
20130327 12:27:19< shadowm> http://forums.wesnoth.org/viewtopic.php?f=16&t=38361
20130327 12:27:30< fabi> esr: But elias is very right. I like to give sourceforge a try. Let's see how the bug and patch thingies work, those have to make me happy.
20130327 12:27:37< AI0867> there is also https://github.com/mirrors
20130327 12:27:37< shadowm> And TBH I can see why they'd be concerned.
20130327 12:27:53< shadowm> I mean, look, I am concerned by my inability to clone this monster on my own.
20130327 12:28:31< shadowm> It also seems that whenever I go to the web interface, it gets stuck doing some "analyzing repository" thing forever.
20130327 12:29:05< AI0867> same here
20130327 12:29:12< fabi> and here
20130327 12:29:39< esr> OK, All the mauntainance branches are up now.  I was very tired last night and forgot to use  --all.
20130327 12:30:52< shadowm> What did you use then?
20130327 12:31:28< shadowm> Yep, I see the tags too.
20130327 12:31:36< fabi> "Thanks Mark! If you try to get it as close as you can to around 1GB, we'd appreciate it!" That sounds like: 1.5G is okay, if you try everything and we did, didn't we?
20130327 12:34:01< AI0867> esr: why did the fortunes get edited in the "git transition" commit?
20130327 12:35:28< esr> AI0867: Look at the diff - it will be clear.
20130327 12:36:44< AI0867> they're two cookies about the CIA bot
20130327 12:37:13< esr> AI0867: That's right - it's deader than the pharaohs.
20130327 12:37:40< AI0867> so?
20130327 12:38:02< shadowm> They were edited in the "git transition" commit?
20130327 12:38:15< AI0867> they were removed
20130327 12:38:17< shadowm> And here I thought you were going to police every commit for commit message<->diff accuracy.
20130327 12:38:22-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 246 seconds]
20130327 12:38:30< esr> Oh.  *blink* that's right, none of you were the ciabot maintainer.  
20130327 12:39:33< esr> Your right, maybe I shouldn't have lumpred that one in there but I was used to thinking of git and CIA in the same breath.  Becquse I maintain the hooks.
20130327 12:40:48< fabi> esr: Is the new git repository open? I mean, can I commit my stuff now?
20130327 12:41:06< esr> If that's the worst thing you can find to criticize I'll figure I got off lightly.
20130327 12:41:11< shadowm> esr, AI0867 hey you two are around right now at the same time!
20130327 12:41:17< shadowm> Figure out the irker mess. :p
20130327 12:41:29< esr> fabi: Yes, it's open.
20130327 12:41:30< AI0867> what kind of hooks do we have on sourceforge?
20130327 12:41:44< AI0867> can we retarget CIA's XML-RPC stuff?
20130327 12:42:14< elias> SF hooks run on a server with no network
20130327 12:42:18< esr> Project admins have write access to git hooks via shell.
20130327 12:42:23< shadowm> AI0867: I'd rather not use esr's hook unless he got around to implement verbose commit messages.
20130327 12:42:46< esr> shadowm: What do you mean by "verbose"?
20130327 12:42:51< AI0867> more than one line
20130327 12:42:57< shadowm> Commit messages with more than just the subject line.
20130327 12:43:08< shadowm> We need those here.
20130327 12:43:31< shadowm> esr: The active project admins right now are you and Ivanovic.
20130327 12:43:49< shadowm> And Ivanovic is unavailable.
20130327 12:44:01< esr> No, you don't.  If yo're writing commits that can't by summartized in one line, your git practice is faulty  and needs to be fixed.
20130327 12:44:16< shadowm> esr: Sigh.
20130327 12:44:25< shadowm> Ever heard of commit message bodies?
20130327 12:44:47< shadowm> You told me about them too even though I kind of started policying that here even before you jumped onto the git bandwagon.
20130327 12:45:08< shadowm> My point is that for IRC we want to see more than just the subject line.
20130327 12:45:30< esr> AI0867: I repeat: If you're writing commits that can't be *summarized* in one line, your git practice is faulty and needs to be fixed.
20130327 12:45:47< esr> Soory, I meant that for shadowm.
20130327 12:45:52< shadowm> No, I get that.
20130327 12:46:05< shadowm> I am saying that for IRC we want to see more than just the subject line.
20130327 12:46:47< esr> AI0867: Perhaps it would be simplest if we set your admin bit so you can edit the hooks?
20130327 12:46:49< shadowm> Especially with the limits you imposed (which you'll have a hard time policying, btw), the subject line is nigh useless for IRC notifications.
20130327 12:47:23< esr> shadowm: No it isn't.  Fix your practice - learn to write good summaries.
20130327 12:47:36< shadowm> No, I don't need to fix my practice.
20130327 12:47:44< shadowm> I will repeat, for IRC we want to see more than just the subject line.
20130327 12:48:03< shadowm> You can't fit all the details in 72 characters in a single line.
20130327 12:48:38< shadowm> Given our workflow, we want to see what the details are most of the time, and it's not always practical to go and fetch and check the new commit.
20130327 12:49:08< esr> You don't *want* all the details.  Just enough to tell IRC watchers whether they are interested and should pull it to look at git log.
20130327 12:49:14< shadowm> No, we do.
20130327 12:49:32< esr> Sorry, I won't budge on this.
20130327 12:49:49< shadowm> It's not always practical to go and fetch and check the new commit.
20130327 12:50:01< esr> You're trying to use the tool in an inappropriate way.
20130327 12:50:10< shadowm> I for one would rather see the commit messages in full here like we've been doing for YEARS already.
20130327 12:50:29< _Coffee> you could just block all april 1st pushes ;)
20130327 12:50:45< shadowm> Where by "in full" I mean "in full except when they are over 8 lines-or-so long".
20130327 12:50:49< _Coffee> or live the jumping bunnies for a day
20130327 12:52:10< esr> AI0867: I can install irkerhook, or I can sert your admin bit so you can do it.  Does your filter need any specialization in the hook?
20130327 12:52:20< AI0867> esr: that would help. Does that actually mean shell access to the repo itself?
20130327 12:52:31< esr> AI0867: Yes it does.
20130327 12:52:53< shadowm> Does the irkerhook also limit the output to just the subject line?
20130327 12:52:54< AI0867> esr: it does not, a list of IP-masks/hosts and targets will lock everything down nicely
20130327 12:53:03-!- mgla [~shishir@unaffiliated/noobjoe] has joined #wesnoth-dev
20130327 12:53:41< AI0867> of course, it's theoretically quite possible for someone else to start a sourceforge project, point their hook at me and then start spamming #wesnoth-dev, but...
20130327 12:53:59< esr> AI0867: Ahhhh, you firewalled it properly.  Good, that's how the probnlem *should* be solved.  Did you read my security analysis document?
20130327 12:54:12< AI0867> yes
20130327 12:54:24< esr> That was fun to write.
20130327 12:56:05< esr> AI0867: OK, your admin but is set.
20130327 12:57:18< esr> I wonder how long that repo analysis is going to take to finish?  It's been about 9 hours now.
20130327 12:57:27< shadowm> 08:52:52 <shadowm> Does the irkerhook also limit the output to just the subject line?
20130327 12:57:54< esr> shadowm: Yes, it does, because under git that;s the right thing to do.
20130327 12:58:24< AI0867> I would hardly be surprised is the answer is "it's going to take until we finish writing allura"
20130327 12:58:24< esr> It wasn't under Subversion where commebting style is different.
20130327 12:58:41< shadowm> AI0867: Will you accept a patch to make it so it shows the body too?
20130327 12:59:16< esr> shadowm: I will back tanty such patch out.  *Use the tool correctly.*
20130327 12:59:16< AI0867> shikadibot: log master
20130327 12:59:17< shikadibot> AI0867: "master" is not a valid commit specification!
20130327 12:59:33-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20130327 12:59:45< shadowm> esr: This is not a negotiable thing.
20130327 13:00:10< shadowm> You come here after a long time of inactivity and suddenly demand that everyone sticks to conventions that were never suggested or enforced here before.
20130327 13:00:19< esr> shadowm: You're right, it isn't.
20130327 13:00:20< shadowm> I know how Git works, don't try to lecture me.
20130327 13:00:23< AI0867> esr: I don't have a "project admin" menu. I do have such a menu under wesnoth-umc-dev
20130327 13:00:42< shadowm> This is more of a matter of how #wesnoth-dev (the IRC channel) works, which is completely different.
20130327 13:00:47< esr> AI0867: Hm, try refreshing?
20130327 13:00:56< AI0867> that's what I've been doing
20130327 13:01:51< esr> shadowm: Igf you know how git works, why have I had to correct wrapped multiline comments from you after you knew wwe were moving? 
20130327 13:02:04< shadowm> esr: Timestamps?
20130327 13:02:16< stikonas> shadowm: have you managed to clone a repository?
20130327 13:02:35< shadowm> stikonas: No, and I never will.
20130327 13:02:49< shadowm> I will have to clone on B and rsync a tarball from there.
20130327 13:02:49< stikonas> Why not?
20130327 13:02:58< stikonas> oh, ok
20130327 13:03:17< stikonas> btw, if somebody wants to clone from me, I can give an URL
20130327 13:03:21< stikonas> I have a fast connection
20130327 13:03:44< shadowm> esr: Oh right, you mean those > 72 characters long summary lines? Pragmatism.
20130327 13:03:48< shadowm> It's as simple as that.
20130327 13:04:11< esr> shadowm: I'm not going to help you be sloppy.
20130327 13:04:12< shadowm> My 1280x800 screen here does 157 columns.
20130327 13:04:20< shadowm> esr: I am not asking for your help.
20130327 13:04:27< shadowm> I know you are very unhelpful most of the time.
20130327 13:04:52< shadowm> At least when it comes to not being stuck with an ancient terminal.
20130327 13:04:57< esr> Fuck you.  I just spend three days of hard work nursemaiding this conversion.
20130327 13:05:08< shadowm> Yes, and I can appreciate that.
20130327 13:05:20< esr> You do not get to give me any kip about being unhelpful.
20130327 13:05:22< shadowm> It's your sudden spurt of political changes out of nowhere that I do not appreciate.
20130327 13:06:30< shadowm> I try to keep my summary lines within reasonable limits. It turns out that in order to make them as helpful and concise as possible at the same time, 72 characters is not always enough.
20130327 13:07:16< shadowm> You would probably throw a fit if you saw a few certain other developers' opinions on the matter.
20130327 13:07:28< esr> shadowm: Long lines create readability problems for people browsing the history with giit log and gitk.  You should stop doing that.
20130327 13:07:43< shadowm> I use git log, thank you very much.
20130327 13:07:59< esr> I can't force you to change, but I can refuse to cooperate with or enable sloppy practice.
20130327 13:08:16< shadowm> I am not cooperating or enabling sloppy practice.
20130327 13:08:23< shadowm> Let me get this through your head again.
20130327 13:08:37< shadowm> I want to see the message bodies on IRC.
20130327 13:08:42< shadowm> Not just the summary lines.
20130327 13:09:04< shadowm> You can stick with 72 character summary lines if you want; this is completely independent of that.
20130327 13:10:15< shadowm> And this is completely independent of message formatting standards. I want to be able to know more than just the bare minimum about a commit without having to go find my desktop and fetch the commit, or interrupt whatever download I've got going on, or any other such scenarios.
20130327 13:11:03< shadowm> So as I said before: IRC, not Git.
20130327 13:14:08-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20130327 13:14:37< elias> and more specifically #wesnoth-dev - some/many IRC channels may prefer just the subject... definitely not this one :)
20130327 13:15:01< shadowm> Hence I marked it in bold and stuff a page above.
20130327 13:15:40< shadowm> In fact, this is exactly why irker-svnpoller has an internal option to choose between displaying bodies and only displaying the summary.
20130327 13:17:14< esr> shadowm: And I'm refusing to implement that because devs (includuing you) shoul stop being mentally lazy and spend the effort required to write a good short summary, for the benefit of future readers.
20130327 13:17:43< shadowm> esr: 1) Nobody is asking you to implement anything.
20130327 13:17:51< shadowm> 08:58:40 <shadowm> AI0867: Will you accept a patch to make it so it shows the body too?
20130327 13:17:59< shadowm> ^ The implied author is me.
20130327 13:18:14< shadowm> esr: 2) This is not about just writing a good and short summary.
20130327 13:18:17< esr> I will back such a patch out because I think it encourages sloppy practice.
20130327 13:18:34< shadowm> This is about getting as much information as possible on our IRC channel as we've always done for years.
20130327 13:18:52< esr> I know your position.  I reject it.
20130327 13:18:55-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has left #wesnoth-dev []
20130327 13:19:16< shadowm> No, you don't know my position, as it is evident by the way you keep rephrasing it.
20130327 13:19:30< AI0867> "to upload files, use web.sourceforge.net"
20130327 13:19:37< AI0867> port 80 on web.sourceforge.net is closed...
20130327 13:19:58< shadowm> You are obsessed with proper summary practice whereas I am obsessed with seeing more than just the summary without having to have the ability to fetch the commit mself all the time.
20130327 13:20:51< shadowm> Two parallel subjects which you are forcing to converge as though we traced them on a sphere.
20130327 13:21:14< zookeeper> does all this mean that the new irc commit messages won't contain urls to a web interface?
20130327 13:21:19< esr> shadowm: Yes, that means one of us has to win and as the person who is senior in this area I am pulling rank.  Just as you would on the forums.
20130327 13:21:48-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 13:21:48< shadowm> esr: Sorry, but I do not consider this vague "senior" title thing to be a replacement for good judgment.
20130327 13:22:05< shadowm> Especially when it is self-conferred around here.
20130327 13:22:14< esr> shadowm: I agree.  I think you are exhibiting poor judgment on this issue.
20130327 13:22:34< shadowm> I think you are being inflexible about what would otherwise be a minor matter.
20130327 13:22:48< zookeeper> shadowm is right, i decree so
20130327 13:23:14< shadowm> Nobody is asking you to relinquish your newly-attained control over commit format policies.
20130327 13:23:40< shadowm> Nobody is asking you to change those policies (yet) either.
20130327 13:23:52< shadowm> And finally, nobody is asking you to lift a single finger for this simple feature.
20130327 13:24:32< AI0867> okay, I can work around this using scp...
20130327 13:25:18< esr> I am reasoning from the larger context here.  If irkkerhook dissplays the entire message, as signifgicant part of the incentive for people to write good summary lines will disappear.
20130327 13:26:08< AI0867> esr: which commit-hook did irkerhook.py need to be run from again? The manpage on catb.org says post-commit, but that's for local things
20130327 13:26:10< esr> I want devs (all devs, not just Wesnoth ones) to have incentives to use git correctly.  So I wrote irkerhook as I did.
20130327 13:26:13< shadowm> The inverse is just as likely to happen under such a constrained design.
20130327 13:26:33< shadowm> For example, I have seen people in #frogatto write really long single-line commit messages for that reason.
20130327 13:26:36< shadowm> And yes, they use Git.
20130327 13:26:39< fabi> esr: Can we have it like this: irker reports only the head line automatically. If ask he gives the rest of the message as well. I would prefer that solution as it keeps the irc traffic small.  
20130327 13:26:59< shadowm> fabi: irker is a write-only service, it cannot get "asked" for things.
20130327 13:27:00< AI0867> irker can't do that
20130327 13:27:05< AI0867> wesbot and shikadibot could
20130327 13:27:11< AI0867> like they do for GNA right now
20130327 13:27:19< fabi> Okay, that is fine for me.
20130327 13:27:29< zookeeper> i'd rather see an occasional non-conformist commit message than not being able to see the details of a commit without fetching or whatever the git word for update is.
20130327 13:27:32< fabi> low traffic plus extra info when ask
20130327 13:27:46< shadowm> Unfortunately, I do not know yet how to fetch a remote commit without a local clone.
20130327 13:27:56< shadowm> So it will take a while for shikadibot to gain that ability (if ever).
20130327 13:27:58-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum]
20130327 13:28:04< shadowm> *remote Git commit
20130327 13:28:04< esr> fabi: That's how it is now.
20130327 13:28:28< shadowm> I am not sure about wesbot, I assume Rh0nda doesn't have much time to figure that out atm.
20130327 13:28:54< shadowm> I imagine as a very cheap alternative one could fetch from the web interface URL and parse that, but that's tacky as hell.
20130327 13:29:09< esr> AI0867: It can be tun fom the pack update hook as well.
20130327 13:29:17< shadowm> Not to mention that it'd be very likely to break given SourceForge.net's habit of moving things around every few months.
20130327 13:30:35-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20130327 13:31:06< shadowm> The thing about being able to see more than just the summary is that I can determine the importance of the commit without having to look at the diff or require people to stick "[IMPORTANT!]" to the subject line.
20130327 13:31:15< shadowm> s/summary/subject line/
20130327 13:31:38< shadowm> For example, if I commit something with a long message, odds are that it is extremely important.
20130327 13:31:39< AI0867> esr: the manpage contradicts itself somewhat on this matter
20130327 13:31:46< zookeeper> people could stick a tinyurl in the subject line which takes you to the full commit message :>
20130327 13:31:47 * zookeeper ducks
20130327 13:32:07< AI0867> it first speaks of post-commit hooks, then buried in the git section, there's something about the update hook in parentheses
20130327 13:32:20< zookeeper> s/tinyurl/pastebin, whatever
20130327 13:32:28< AI0867> github's irker hook already has that
20130327 13:32:52< esr> AI0867: Yes, look in the header comment - I thinkI dedcribe the update setup in there.
20130327 13:33:50< shadowm> Oops.
20130327 13:34:04< shadowm> AI0867: I think I closed a vi session you were using on ai0867.net?
20130327 13:34:57< shadowm> I didn't realize you had attached with -x while I was inspecting the irker docs.
20130327 13:36:30< AI0867> esr: sourceforge runs python 2.4.3
20130327 13:36:47< shadowm> By the way, if anyone can tell me how to fetch a Git commit message from a remote repository without having to parse some tacky HTML page I'd appreciate some pointers.
20130327 13:37:00< shadowm> I remember searching information about that a couple of years ago to no avail.
20130327 13:37:24< esr> AI0867: Ugh.  What do I need to backport?
20130327 13:37:39< AI0867> well, the json module is missing, not sure what else
20130327 13:37:46< vultraz> AI0867: isn't the latest python version 3.something?
20130327 13:37:46< AI0867> I can get simplejson for that
20130327 13:38:03< AI0867> vultraz: yes, though for python2, version 2.7 is the standard nowadays
20130327 13:38:30< elias> what does the hook do on sourceforge?
20130327 13:38:31< vultraz> so how come sf is so behind :S
20130327 13:38:38< AI0867> esr: also, nenolod's github account has moved to 'mogri'
20130327 13:38:47< AI0867> so the link to irker-cia-proxy no longer works
20130327 13:38:51< elias> without network, it might not be all that useful - unless they allow it now
20130327 13:39:35< shadowm> AI0867: Any idea why he changed his account name bte?
20130327 13:39:37< shadowm> +btw
20130327 13:39:39< AI0867> no
20130327 13:39:59< AI0867> elias: no network on the shell, but I assume the server the hooks get run on do have network?
20130327 13:40:57< elias> AI0867: the answer was no 4 months or so ago - all they allowed was sending emails back then
20130327 13:41:07< AI0867> gah, simplejson only goes back to python2.5
20130327 13:41:19< AI0867> okay, then we're hosed
20130327 13:41:32< shadowm> What.
20130327 13:41:34< AI0867> we could go back to GNA-style reporting
20130327 13:41:40< AI0867> but yeah, that's pretty much it
20130327 13:41:42< elias> you could ask in #sourceforge, their staff at least is very helpful
20130327 13:41:47< shadowm> elias: Was that under allura or the old platform?
20130327 13:41:54< elias> allura
20130327 13:43:58< shadowm> Anyway, later, have fun.
20130327 13:44:00 * stikonas thinks that even if self-hosting had brought other problems, it would still have been easier...
20130327 13:44:19< shadowm> I am sure you will have lots of fun figuring out this (hopefully not real) conundrum.
20130327 13:49:23-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130327 13:49:26< AI0867> esr: looks like the only way to for hooks on sourceforge to get data out is by sending email. We could adapt the svnpoller for this, but this sounds like a terrible mess in the making
20130327 13:50:13-!- mgla [~shishir@unaffiliated/noobjoe] has quit [Ping timeout: 248 seconds]
20130327 13:50:15< AI0867> I'm especially happy with how their new system uses 10 (okay 9) year old software
20130327 13:50:27< esr> AI0867: Yuck.  You mean the host where the hooks run really doesn't have network?
20130327 13:50:30< AI0867> yes
20130327 13:50:40< AI0867> that's what elias is saying
20130327 13:50:47< esr> That's awful.
20130327 13:50:48< AI0867> and the shell I have doesn't have network either
20130327 13:51:03< AI0867> I've had to find how to ssh into the same shell and then abuse scp
20130327 13:51:06< esr> Dammit.
20130327 13:51:20< _Coffee> shadowm: to get a particular file out of a git repository maybe you want "git show <treeish>:<file>"?
20130327 13:51:27< shadowm> If you are completely sure that emailing is the only option, then I may be able to come up with an alternative from svnmail-filter that parses the commit mailing list emails.
20130327 13:51:34< shadowm> _Coffee: Remote?
20130327 13:51:49< AI0867> shadowm: I'm sure that we can, but do we really want to get into that?
20130327 13:51:56 * esr really needs to spend a year solving the forges-are-so-fucking-broken problem.
20130327 13:52:07< shadowm> _Coffee: I want to fetch commit messages from a remote repoisotry, not a local repository.
20130327 13:52:10< shadowm> repository.
20130327 13:52:52< shadowm> Also, just commit messages, not files. Files can be quite arbitrarily large and I'd rather not have shikadibot be able to fetch battle.ogg on request.
20130327 13:52:52< _Coffee> shadowm: I see what you mean, your copy on PULL will have a different master
20130327 13:53:15< shadowm> _Coffee: The point is that I want to be able to this without a local repository.
20130327 13:53:51< shadowm> AI0867: If there is no other alternative...
20130327 13:54:47< shadowm> I could alternatively implement YET another layer to the irker cia proxy thing that handles emails with procmails, then you could use a CIA emailing hook on the upstream repository.
20130327 13:55:15< shadowm> IIRC CIA emails were basically the same thing as the XML-RPC method, just encapsulated in an email message.
20130327 13:55:29< shadowm> *procmail
20130327 13:56:05-!- Guest44736 [~eli@66-189-34-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20130327 13:57:19< shadowm> AI0867: Is there a working commits mailing list already?
20130327 13:57:33< shadowm> Wait, there isn't even a mailing lists menu.
20130327 13:58:08< shadowm> You will have to set up that at some point, regardless of whether we will have irker notifications or not, so maybe you could do that now?
20130327 13:58:14< shadowm> *set that up
20130327 13:58:37< shadowm> Then I can take a look at how parseable the emails are.
20130327 13:59:21< shadowm> Or alternatively, refer me to an existing SF.net project under Allura that uses Git with a commits mailing list.
20130327 13:59:50< shadowm> In any case, see you in 12 hours or so.
20130327 14:00:39< shadowm> lipkab: http://forums.wesnoth.org/viewtopic.php?p=550427#p550427 
20130327 14:00:59< shadowm> Oh also, what's the plan for the trackers? I don't think I heard about that again.
20130327 14:02:21< shadowm> I don't have a problem with continuing to work with Gna.org for that, it's less trouble for me since I will apparently need to update an indeterminate amount of forum post rows because of the [bug] and [patch] bbcodes, but I am sure we all agree that having some Git+tracker integration would be nice.
20130327 14:02:52< AI0867> shadowm: I still don't have a project admin menu
20130327 14:02:54< shadowm> Also, I guess I will need to add a [commit] bbcode and hide [rev] from plain view.
20130327 14:03:06< shadowm> AI0867: Boo. :(
20130327 14:04:11-!- mgla [~shishir@unaffiliated/noobjoe] has joined #wesnoth-dev
20130327 14:14:15-!- artisticdude [~artisticd@126.sub-70-192-193.myvzw.com] has joined #wesnoth-dev
20130327 14:14:56-!- Octalot [~noct@host86-146-156-9.range86-146.btcentralplus.com] has joined #wesnoth-dev
20130327 14:19:27< AI0867> shadowm: http://stackoverflow.com/questions/4288598/git-remote-log-and-changes-request
20130327 14:20:25< AI0867> so yeah, no way for shikadibot to do the job without cloning or scraping
20130327 14:20:47-!- artisticdude [~artisticd@126.sub-70-192-193.myvzw.com] has quit [Ping timeout: 252 seconds]
20130327 14:21:53< AI0867> and --depth doesn't look like it'll help much
20130327 14:26:29-!- _8680a_ [~androirc@ip68-225-249-82.oc.oc.cox.net] has quit [Remote host closed the connection]
20130327 14:26:55-!- artisticdude [~artisticd@32.sub-70-192-201.myvzw.com] has joined #wesnoth-dev
20130327 14:27:05-!- _8680a_ [~androirc@68.225.249.82] has joined #wesnoth-dev
20130327 14:45:29-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 14:49:48< AI0867> okay, I've been looking at the stuff irkerhook uses to find which commits to send, and it doesn't look like they'll run there
20130327 14:50:27< AI0867> it doesn't seem to want to find any objects
20130327 14:51:37-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 14:53:36-!- mattsc [~mattsc@207.230.251.234] has joined #wesnoth-dev
20130327 14:54:19-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20130327 14:54:26< shadowm> Back.
20130327 14:54:55< shadowm> Let's pretend it was 12 hours from my perspective.
20130327 14:57:04< shadowm> AI0867: Scraping?
20130327 14:57:16< lipkab> shadowm: The background thing is messed up in HttT's WML, not in C++.
20130327 14:57:40< lipkab> The map isn't scaled.
20130327 14:57:44< AI0867> shadowm: from the webinterface
20130327 14:57:46< shadowm> lipkab: Okay, then zookeeper I guess?
20130327 14:57:56< lipkab> Yeah.
20130327 14:57:58< zookeeper> hmh?
20130327 14:58:06< shadowm> zookeeper: http://forums.wesnoth.org/viewtopic.php?p=550427#p550427
20130327 14:58:26< shadowm> AI0867: Tacky... HTML...
20130327 14:58:30< shadowm> No.
20130327 14:58:46< shadowm> There has to be some other way, come on.
20130327 14:59:25< AI0867> without having a clone? not really
20130327 14:59:27-!- vultraz is now known as artisticdudee
20130327 14:59:29-!- artisticdude [~artisticd@32.sub-70-192-201.myvzw.com] has left #wesnoth-dev []
20130327 14:59:32< zookeeper> oops. yeah.
20130327 14:59:38< AI0867> you can list objects, but you need to grab them to get the commit message within
20130327 14:59:40< shadowm> So we will have to host a clone in ai0867.net?
20130327 14:59:43< shadowm> For shikadibot?
20130327 14:59:49< AI0867> ugh
20130327 14:59:50< shadowm> That's simply preposterous.
20130327 15:00:16< AI0867> well, we have more storage now, so we *can*, but do we want to?
20130327 15:00:22< shadowm> I don't.
20130327 15:00:30< shadowm> Because every request will have to involve a git pull.
20130327 15:00:31-!- artisticdudee is now known as Hulavutaa
20130327 15:00:38< shadowm> etc etc DoS etc etc etc
20130327 15:00:54< AI0867> nah, just fetch using cron and reply "I don't know about that commit" otherwise
20130327 15:01:12< shadowm> Let's pretend I am irker
20130327 15:01:15-!- Hulavutaa is now known as vultraz
20130327 15:01:25< AI0867> that was for shikadibot
20130327 15:01:34< AI0867> if you're irker, then we can pipe the data over mail
20130327 15:01:40< shadowm> Wesnoth: SHA1SHA1SHA1 /src/foo.cpp (1 file in 1 dir): Increased suckiness.
20130327 15:01:45< shadowm> Now I am zookeeper.
20130327 15:01:53< shadowm> shikadibot: gitlog SHA1SHA1SHA1
20130327 15:01:57< shadowm> Now I am shikadibot.
20130327 15:02:04< shadowm> zookeeper: I don't know about that commit.
20130327 15:02:09< shadowm> How will zookeeper feel?
20130327 15:02:19< shadowm> We can't have a cronjob running every 10 seconds.
20130327 15:02:31< shadowm> I guess I will have to go the HTML way.
20130327 15:02:33 * zookeeper blinks
20130327 15:02:50< AI0867> or we can have a "keep git repos synced" process running...
20130327 15:02:52< shadowm> Has anyone mirrored a tarball of that abomination yet?
20130327 15:03:03< shadowm> Or should I host one on baldras using which facility?
20130327 15:03:09< shadowm> I have root access to the httpd facility.
20130327 15:03:10< AI0867> stikonas offered to
20130327 15:03:32< AI0867> I can drop it on ai0867.net, a certain VM on baldras, or on asheviere
20130327 15:03:55< shadowm> If I use the httpd facility then vultraz would be able to benefit from it too.
20130327 15:04:16< shadowm> As well as anyone else who can't possibly download 1.6 GiB in an atomic step.
20130327 15:04:48< shadowm> You said baldras is better suited for this because of bandwidth and quotas.
20130327 15:05:15< AI0867> yes, but I only have access to the one VM
20130327 15:05:26< AI0867> and it doesn't have an httpd on there
20130327 15:05:30< shadowm> I have root access to the httpd VM.
20130327 15:05:35< AI0867> okay
20130327 15:05:37-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has joined #wesnoth-dev
20130327 15:05:42< AI0867> gimme an account and I'll try uploading
20130327 15:06:11< shadowm> Hm, is there any reason you'd prefer to upload instead of me just cloning on baldras and tarballing it?
20130327 15:06:25< AI0867> uh, not really
20130327 15:06:28< AI0867> so go ahead and do that
20130327 15:06:35< stikonas> AI0867: who wants to clone from me?
20130327 15:06:45< AI0867> stikonas: shadowm
20130327 15:06:52< shadowm> rsync -P or wget -c, not git clone.
20130327 15:06:52< stikonas> git clone git://stikonas.homelinux.org/wesnoth.git
20130327 15:07:04< stikonas> but please change remote to SF after cloning...
20130327 15:07:27< shadowm> git clone could only result in me wanting to fling a chair SF.net-wards if my connection is interrupted around 98%.
20130327 15:07:32< shadowm> Or anywhere after 10%, really.
20130327 15:07:35< Octalot> There's a git-bundle that seems more suited that tarballs.  Just looking up FlightGear's documentation: http://wiki.flightgear.org/FlightGear_and_Git#Special_for_fgdata:
20130327 15:08:27< shadowm> Uh.
20130327 15:08:28< stikonas> shadowm: you can probably wget as well
20130327 15:08:40< stikonas> one moment please
20130327 15:09:43< shadowm> Octalot: It sounds like it's intended to be used with a list of refspecs?
20130327 15:10:05< vultraz> ermegard you're referencing fg
20130327 15:10:17< shadowm> I am referencing what?
20130327 15:11:05< shadowm> I am really unsure how this git bundle create thing can be more efficient than a tarball... considering I will have to list a bunch of refspecs anyway.
20130327 15:11:18< shadowm> Or create a bunch of bundles, each one for a refspec? Confusing.
20130327 15:11:23< vultraz> Octalot liked to a FlgithGear page
20130327 15:11:49< shadowm> vultraz: Is this somehow important?
20130327 15:11:54< bumbadadabum> Flgith
20130327 15:11:55< vultraz> that happens to be the project whose 2.5 gb repo I successfully cloned
20130327 15:12:08< stikonas> shadowm: http://stikonas.homelinux.org/files/wesnoth-bare.tar
20130327 15:12:17< shadowm> stikonas: I love you.
20130327 15:12:29< stikonas> sorry
20130327 15:12:29< AI0867> ah, so the shell doesn't actually have access to the repo, just the copied config files
20130327 15:12:34< shadowm> stikonas: Wait, how much is your transfer rate upstream?
20130327 15:12:34< Octalot> shadowm: IIUC, it's another format for a git archive, so it includes the master (trunk) branch, and all other branches that you tell bundle to include.
20130327 15:12:35< AI0867> ...
20130327 15:12:35< stikonas> I need to rerun tar
20130327 15:12:53< stikonas> shadowm: ~5MB/s
20130327 15:12:55< shadowm> Er, upstream? I mean upload rate.
20130327 15:13:13< stikonas> but I need to rerun tar... So please wait a few minutes
20130327 15:13:19< shadowm> Okay, I can wait.
20130327 15:13:24< shadowm> This will probably take a few days anyway.
20130327 15:13:37< shadowm> I cannot afford the download between 18:00 and 00:01 UTC-03:00.
20130327 15:13:50< stikonas> shadowm: what is your download speed?
20130327 15:14:02< shadowm> It's 11:13 right now, but depending on RF quality I may spend all the time from here until 18:00 downloading.
20130327 15:14:02< AI0867> this of course makes it impossible to actually TEST the hook
20130327 15:14:14< shadowm> stikonas: Theoretically up to 350 KiB/sec or so.
20130327 15:14:21< shadowm> It varies according to RF interference and other factors.
20130327 15:14:29-!- horon [~horon@nttkyo320074.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has joined #wesnoth-dev
20130327 15:14:43< stikonas> I usually turn off my laptop for the night, so please download with resume capability...
20130327 15:14:53< stikonas> but wget is supposed to do that...
20130327 15:15:04< shadowm> Yeah, obviously, the whole point of this exercise is having resume capability.
20130327 15:15:07< stikonas> shadowm: ok, you can start downloading
20130327 15:15:11< shadowm> Are you sure you wouldn't prefer to- okay.
20130327 15:15:40< shadowm> Hm.
20130327 15:15:53< shadowm> Less than 30 KiB/sec down.
20130327 15:16:05< stikonas> hmm, that would take too long then
20130327 15:16:14< shadowm> Yep, hence I said days.
20130327 15:16:18< vultraz> it would take 36 hours
20130327 15:16:25< vultraz> Trust me, I know
20130327 15:16:26< shadowm> wget says 26.
20130327 15:16:38< shadowm> Maybe I'll just try making a clone at baldras, tarballing it, and...
20130327 15:16:47 * shadowm presses CTRL+C.
20130327 15:16:58< vultraz> You trust that your internet will hold at a constant speed for a whole day?
20130327 15:17:13< shadowm> vultraz: Shoooooooosh.
20130327 15:17:14< shadowm> You are being noisy.
20130327 15:19:47< shadowm> Okay, actually everything is being slow.
20130327 15:20:11< shadowm> I am barely able to write a thing on A (Wesnoth.org live).
20130327 15:20:18-!- horon [~horon@nttkyo320074.tkyo.nt.ngn2.ppp.infoweb.ne.jp] has quit [Quit: Leaving...]
20130327 15:20:48< bumbadadabum> Hey guise
20130327 15:20:59< bumbadadabum> 1.11.2 took me 3 minutes to dl
20130327 15:21:03< bumbadadabum> maybe even less
20130327 15:21:22< elias> yeah, SF has good b/w
20130327 15:21:30< shadowm> No.
20130327 15:21:36< bumbadadabum> elias: It used to be bad
20130327 15:21:39< shadowm> SF.net's file release system mirrors do.
20130327 15:21:51< bumbadadabum> but last few versions downloaded very wuickly
20130327 15:21:56< bumbadadabum> *quickly
20130327 15:21:58< elias> also git, took me <3 mins to clone as well
20130327 15:22:33< shadowm> And here I am struggling with SSH.
20130327 15:22:40< shadowm> How can I possibly hope to download this monstrosity?
20130327 15:22:50< stikonas> shadowm: DVD :D
20130327 15:22:52< bumbadadabum> move to europe
20130327 15:22:56< bumbadadabum> get internet
20130327 15:23:10< elias> since there's ssh access on SF, maybe could just tar it on there?
20130327 15:23:35< stikonas> there is already tar at http://stikonas.homelinux.org/files/wesnoth-bare.tar the problem is downloading it with slow connection
20130327 15:24:18< AI0867> let's see how fast I can pull that
20130327 15:24:31< shadowm> I need to create me a user on the httpd facility.
20130327 15:24:41< AI0867> 700 KiB/s on my laptop
20130327 15:24:45< shadowm> Using root access with an unreliable SSH link is making me nervous.
20130327 15:25:01< AI0867> tmux
20130327 15:25:13< shadowm> You mean GNU screen.
20130327 15:25:22< AI0867> have you seen the sources of screen?
20130327 15:25:24< shadowm> I haven't tried tmux yet and I'm too busy right now to learn something new.
20130327 15:25:25< shadowm> No.
20130327 15:25:34< AI0867> that's why I started looking for an alternative
20130327 15:25:35< bumbadadabum> http://stikonas.homelinux.org/files/wesnoth-bare.tar <- 500 kb/s
20130327 15:25:40< AI0867> also, there's a screenlike config for tmux
20130327 15:26:10< shadowm> Hang on, I'll try something.
20130327 15:26:12-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit []
20130327 15:26:21< loonycyborg> I've managed to clone&build new repo. Everything seems to work fine.
20130327 15:26:33< loonycyborg> Good thing I have a good connection here :P
20130327 15:26:43< vultraz> don't listen to bumbadadabum, he lives in dutchland, where internet is so goddamn fast he can download League of Legends in half an hour
20130327 15:27:00< vultraz> he is not a fair comparison
20130327 15:27:33< shadowm> Okay, reconnected my modem with an extender cable.
20130327 15:27:36< stikonas> vultraz: I'm using the cheapest ISP plan (50 Mbit/s). There are also 100 and 300 Mbps plans...
20130327 15:27:48< shadowm> Doing 280 KiB/sec instead of 30 KiB/sec all of a sudden.
20130327 15:28:08< vultraz> wow apparently you all live in super internet lands o_o
20130327 15:28:21 * vultraz sits here with his 30 Kib/s
20130327 15:28:25< lipkab>  vultraz: Where do you live?
20130327 15:28:31< vultraz> Micronesia
20130327 15:28:37< lipkab> Ah.
20130327 15:28:39< shadowm> ugjhkdflgjdf
20130327 15:28:41< bumbadadabum> KiB/s?
20130327 15:28:50 * shadowm flings chair vultraz-wards.
20130327 15:29:01< bumbadadabum> is that your download speed
20130327 15:29:09< bumbadadabum> or the speed you're downloading at?
20130327 15:29:30< lipkab> vultraz: I'd gladly trade my internet for some palm trees.
20130327 15:29:49< bumbadadabum> lipkab: You don't
20130327 15:30:11< shadowm> Okay, stabilized around 140 KiB/sec.
20130327 15:30:16< lipkab> bumbadadabum: ?
20130327 15:30:23< bumbadadabum> lipkab: Stay in Europe
20130327 15:30:31< elias> 140 KiB/sec means you should have the 1.5GB in a few hours
20130327 15:30:32< shadowm> stikonas: So, around what time (UTC) does this machine go sto sleep?
20130327 15:30:51-!- mattsc [~mattsc@207.230.251.234] has quit [Quit: I'm asleep]
20130327 15:30:52< shadowm> elias: Except it isn't stable because of RF shenanigans.
20130327 15:31:04< stikonas> shadowm: maybe ~ 22 UTC
20130327 15:31:20< lipkab> bumbadadabum: Are palm trees and Europe mutually exclusive things?
20130327 15:31:22< stikonas> you should have at least 7h
20130327 15:31:36< bumbadadabum> lipkab: Hmm...
20130327 15:31:43< bumbadadabum> I think
20130327 15:31:51< shadowm> stioknas: Okay, thanks.
20130327 15:32:01< stikonas> shadowm: also, please download http://stikonas.homelinux.org/files/wesnoth-bare.tar.sha256
20130327 15:32:03< shadowm> AI0867: Right now it is paramount to set up a new commit mailing lists.
20130327 15:32:18< vultraz> lipkab: go chill at a beach in the south of france
20130327 15:32:28< shadowm> I was about to ask you about that, stikonas.
20130327 15:32:32< bumbadadabum> lipkab: Italy
20130327 15:32:33-!- nephro [50e9e70c@gateway/web/freenode/ip.80.233.231.12] has joined #wesnoth-dev
20130327 15:32:36< bumbadadabum> Greece
20130327 15:32:40< lipkab> bumbadadabum: Amazingly, they're not! There're palm trees in Span and Portugal!
20130327 15:32:46< lipkab> *Spain
20130327 15:32:46< bumbadadabum> there are plenty of good beaches close to you
20130327 15:32:52< bumbadadabum> realy?
20130327 15:32:55< bumbadadabum> *really
20130327 15:32:56< bumbadadabum> kewl
20130327 15:33:00< lipkab> Yeah.
20130327 15:33:14< lipkab> At least my sister told me so.
20130327 15:33:50< lipkab> There're no good beaches here. I'm sitting in a country with a damn seashore.
20130327 15:33:52< bumbadadabum> I haven't been to spain in the last 10 years
20130327 15:34:12< lipkab> ...and I have never been in Spain ever.
20130327 15:34:13< shadowm> I accidentally cat'd the portion of the tarball that's already downloaded and that apparently broke my screen session's status bar.
20130327 15:34:18< vultraz> I've never been to Europe
20130327 15:34:20< bumbadadabum> I don't like beacher
20130327 15:34:22< bumbadadabum> *beaches
20130327 15:34:27< bumbadadabum> salt water
20130327 15:34:29< bumbadadabum> sand
20130327 15:34:32< bumbadadabum> meh
20130327 15:35:19< vultraz> bumbadadabum: but, the hot chicks :P (also why are we discussing this here)
20130327 15:35:52< bumbadadabum> vultraz: You mean 50-year old women and young children?
20130327 15:35:57< bumbadadabum> I bet you love those
20130327 15:36:12< shadowm> This is the developers channel, you two.
20130327 15:36:18< lipkab> bumbadadabum: In fact, I don't like beaches either. Just palm trees and cacti. And fish.
20130327 15:36:30< AI0867> I'm currently effectively rewriting irkerhook.py in shell, to dump into mail
20130327 15:36:39< AI0867> shadowm: what do you want the irker messages to look like?
20130327 15:36:44< bumbadadabum> #wesnoth-offtopic
20130327 15:36:53< vultraz> lipkab: well, we have a lot of palm trees here :P
20130327 15:36:58< bumbadadabum> yup, that's dead
20130327 15:37:00< vultraz> ##somedutchguy
20130327 15:37:10< bumbadadabum> vultraz: SSSSHHHHH!
20130327 15:37:15< shadowm> AI0867: Whatever they look like at the moment, including up to 6 lines of the message body after the summary line (excluding blank lines).
20130327 15:37:24< bumbadadabum> vultraz: That's my sacred sanctuary
20130327 15:37:38< shadowm> Or whatever the config I am using on ai0867.net for irker-svnpoller looks like.
20130327 15:38:12< shadowm> Sorry, I can't be very responsive with this download going on.
20130327 15:39:21< shadowm> I am sure my carrier hates me even more than I do hate them.
20130327 15:39:34< shadowm> Long, massive downloads on 3.5G. Preposterous.
20130327 15:40:28< shadowm> That must be why they made sure to cut off this thing as soon as the bill period expired.
20130327 15:41:40< shadowm> AI0867: Five non-summary body lines maximum, a "Log message trimmed." line is added if the message is longer than that.
20130327 15:41:43-!- Unnheulu [56ba860d@gateway/web/freenode/ip.86.186.134.13] has joined #wesnoth-dev
20130327 15:41:59< shadowm> In parentheses.
20130327 15:42:15< shadowm> Also, there's some pretty formatting going on, and blank lines are skipped entirely.
20130327 15:42:39< shadowm> You can take a look at ~irker/irker-svnpoller/irker-svnpoller for the deets.
20130327 15:42:51< shadowm> Assuming you can understand my Python, that is.
20130327 15:44:36< shadowm> Okay, since I am clearly useless with my remaining downstream bandwidth, I'm leaving for now while wget does its thing.
20130327 15:44:53< shadowm> I have already downloaded 135 MiB.
20130327 15:45:24< shadowm> If something really really really urgent crops up, poke me in the admins channel, because I cannot even keep my mail client running like this.
20130327 15:45:48< bumbadadabum> #wesnoth-totallynottheadminchannel ?
20130327 15:46:04< shadowm> Good thing that I will only have to do this once in my entire life.
20130327 15:46:31< shadowm> Unless someone decides to rewrite the upstream history, that is.
20130327 15:47:16-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Quit: Konversation terminated!]
20130327 15:47:57< shadowm> AI0867: But yeah, right now we need a commits mailing list.
20130327 15:50:52-!- mgla [~shishir@unaffiliated/noobjoe] has quit [Ping timeout: 272 seconds]
20130327 15:51:09-!- mattsc [~mattsc@fw.hia.nrc.ca] has joined #wesnoth-dev
20130327 15:52:21-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 15:53:28< Octalot> I've tried my resumable clone trick, and think the tarball is probably the better option.  This is very ugly working draft at the moment, but a resumable clone by using tags: http://pastebin.com/bRQKmhBQ    It's still running, and some of the downloads are still too big (272MB from 1.3.1 to 1.4).
20130327 16:00:09-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum]
20130327 16:01:31-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 16:07:46-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum]
20130327 16:20:09-!- irker423 [~irker@ai0867.net] has quit [Quit: transmission timeout]
20130327 16:30:14-!- Captain_Wrathbow [~Wrathbow@adsl-71-150-251-229.dsl.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20130327 16:55:28-!- Unnheulu [56ba860d@gateway/web/freenode/ip.86.186.134.13] has quit [Ping timeout: 245 seconds]
20130327 16:55:47-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 16:57:22-!- Unnheulu [56ba860d@gateway/web/freenode/ip.86.186.134.13] has joined #wesnoth-dev
20130327 17:02:33-!- skyfaller [~skyfaller@ool-43551edd.dyn.optonline.net] has joined #wesnoth-dev
20130327 17:02:34-!- skyfaller [~skyfaller@ool-43551edd.dyn.optonline.net] has quit [Changing host]
20130327 17:02:34-!- skyfaller [~skyfaller@wikipedia/Skyfaller] has joined #wesnoth-dev
20130327 17:02:36-!- Captain_Wrathbow is now known as Cpt_Wrathbow
20130327 17:02:50-!- Cpt_Wrathbow is now known as Captain_Wrathbow
20130327 17:04:50-!- mattsc [~mattsc@fw.hia.nrc.ca] has quit [Quit: bye]
20130327 17:13:10-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Read error: Connection reset by peer]
20130327 17:13:42-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 17:13:43-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Remote host closed the connection]
20130327 17:13:57-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 17:17:47-!- mattsc [~mattsc@fw.hia.nrc.ca] has joined #wesnoth-dev
20130327 17:17:52-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 17:18:32-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Read error: Connection reset by peer]
20130327 17:19:02-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 17:19:43-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has quit [Quit: bumbadadabum]
20130327 17:20:28-!- Unnheulu [56ba860d@gateway/web/freenode/ip.86.186.134.13] has quit [Ping timeout: 245 seconds]
20130327 17:20:28-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Read error: Connection reset by peer]
20130327 17:20:54-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 17:21:25-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 248 seconds]
20130327 17:23:44-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Read error: Connection reset by peer]
20130327 17:32:52-!- Captain_Wrathbow [~Wrathbow@adsl-71-150-251-229.dsl.mdsnwi.sbcglobal.net] has quit [Quit: Soli Deo Gloria]
20130327 17:36:36-!- boucman_work [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130327 17:38:04-!- irker369 [~irker@ai0867.net] has joined #wesnoth-dev
20130327 17:38:07< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-3-g3f5e874 / file foo: contents
20130327 17:38:21< AI0867> okay, that's starting to look like something
20130327 17:58:23-!- nephro [50e9e70c@gateway/web/freenode/ip.80.233.231.12] has quit [Ping timeout: 245 seconds]
20130327 17:58:59< stikonas> shadowm: how is your downlaod going?
20130327 17:58:59-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has quit [Read error: Connection reset by peer]
20130327 17:58:59-!- Upthorn [~ogmar@108-85-91-228.lightspeed.frokca.sbcglobal.net] has joined #wesnoth-dev
20130327 18:01:29< shadowm> stikonas: Poorly.
20130327 18:01:50< shadowm> 33% done (851 MiB).
20130327 18:01:55< shadowm> I mean 44%.
20130327 18:02:05< stikonas> that's not too bad...
20130327 18:02:26< shadowm> It will get worse soon.
20130327 18:02:43< shadowm> My connection gets increasingly unstable after noon.
20130327 18:03:22< shadowm> I blame that stupid ball of super-hot plasma that people call "sun".
20130327 18:03:59< stikonas> I doubt that connection in Pluto is better...
20130327 18:04:54< shadowm> In all seriousness, it's probably related to other people's connections since this shares infrastructure with cell phone lines.
20130327 18:05:48< shadowm> So it's generally more stable between 01:00 am and 12:00 pm.
20130327 18:07:53-!- oldtopman [~oldtopman@unaffiliated/oldtopman] has joined #wesnoth-dev
20130327 18:07:55< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-3-g3f5e874 / file foo: contents
20130327 18:12:29< shadowm> Later again.
20130327 18:17:21-!- Octalot [~noct@host86-146-156-9.range86-146.btcentralplus.com] has quit []
20130327 18:23:14-!- bumbadadabum [~bumbadada@d155109.upc-d.chello.nl] has joined #wesnoth-dev
20130327 18:39:20-!- boucman_work [~rosen@wesnoth/developer/boucman] has quit [Ping timeout: 272 seconds]
20130327 18:44:45-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has joined #wesnoth-dev
20130327 18:53:01-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20130327 18:56:58-!- Captain_Wrathbow [~Wrathbow@adsl-71-150-251-229.dsl.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20130327 19:00:13-!- ejls_ is now known as ejls
20130327 19:01:06-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev
20130327 19:10:47-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 19:17:08-!- trewe [~trewe@87-196-129-238.net.novis.pt] has joined #wesnoth-dev
20130327 19:26:45-!- _8680a_ [~androirc@68.225.249.82] has quit [Remote host closed the connection]
20130327 19:27:05-!- _8680a_ [~androirc@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 19:42:52< mattsc> skyfaller, Alarantalara: here's one of those things that makes perfect sense in retrospect, but that I just never thought about before...  In this code: http://pastebin.com/5suDC7m7
20130327 19:43:12< mattsc> execution of the first loop takes 2.2s on my computer, and of the second 0.01s.
20130327 19:43:28< skyfaller> what? why?
20130327 19:44:15< mattsc> I believe it's because in the first case we need to search all variables to see if one of them has the right name, whereas in the second, we just access the value of the given local variable.
20130327 19:44:51< vultraz> deja vu...
20130327 19:45:42< mattsc> Moral of the story is that we should make sure to declare all the config variables in AI code that gets executed over and over again, even if we set them to false (or nil)
20130327 19:46:19< mattsc> Mostly I am writing this down here so that (maybe) I'll remember it next time.
20130327 19:46:59< mattsc> And secondarily so that those here who actually know anything about coding can tell me that I am an idiot (and give me the correct explanation).  :)
20130327 19:47:17< skyfaller> interesting...
20130327 19:52:17< mattsc> skyfaller: I am working on a method that will let us do (somewhat accurate) counter attack calculations for pretty much anything.
20130327 19:52:35< skyfaller> ooh, tasty. How will we do that?
20130327 19:52:54< mattsc> And I found this when I was trying to figure out wth checking my cutoff criterion made everything so much slower.
20130327 19:53:25< mattsc> By not checking all possible combinations ...
20130327 19:53:43< mattsc> Give me until tomorrow or so and we can discuss this some more.
20130327 19:53:47< skyfaller> ok :)
20130327 19:53:59< mattsc> I first want to see if it actually works  :)
20130327 20:00:18-!- _8680_ [~quassel@ip68-225-249-82.oc.oc.cox.net] has joined #wesnoth-dev
20130327 20:00:35-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20130327 20:09:03-!- trewe_ [~trewe@87-196-129-238.net.novis.pt] has joined #wesnoth-dev
20130327 20:12:14-!- trewe [~trewe@87-196-129-238.net.novis.pt] has quit [Ping timeout: 252 seconds]
20130327 20:14:49-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Quit: Leaving]
20130327 20:37:03-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20130327 20:38:16< timotei> wow
20130327 20:38:25< timotei> You guys have to see this: http://vps2.etotheipiplusone.com:30176/redmine/attachments/48
20130327 20:38:28< timotei> Kate in browser :D
20130327 20:38:54< bumbadadabum> That's pretty cool
20130327 20:40:06< timotei> Yup.
20130327 20:53:08-!- Captain_Wrathbow [~Wrathbow@adsl-71-150-251-229.dsl.mdsnwi.sbcglobal.net] has quit [Quit: Soli Deo Gloria]
20130327 21:04:45< esr> Wzcellent.  The SF web interface has completed its meditation.
20130327 21:05:06< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-1-g7ea91f7 / bar: bar
20130327 21:05:08< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-2-gf353a8b / foo: foo
20130327 21:05:10< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-3-g3f5e874 / file foo: contents
20130327 21:05:40< AI0867> =)
20130327 21:06:06-!- witness [uid10044@gateway/web/irccloud.com/x-kmepvsuuoqzwxsmb] has quit [Ping timeout: 264 seconds]
20130327 21:08:13-!- happygrue_ [~happygrue@wesnoth/developer/wintermute] has joined #wesnoth-dev
20130327 21:09:42-!- happygrue [~happygrue@wesnoth/developer/wintermute] has quit [Ping timeout: 264 seconds]
20130327 21:11:58 * esr applauds AI0867
20130327 21:23:24-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 21:27:40-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Client Quit]
20130327 21:42:56-!- Octalot [~noct@host86-146-156-9.range86-146.btcentralplus.com] has joined #wesnoth-dev
20130327 21:49:11< Octalot> mattsc: in http://pastebin.com/5suDC7m7 surely the second loop exits immediately because you're testing a double-negative   "if (not false) then end"
20130327 21:50:20< mattsc> Octalot: yes, but so does the first.
20130327 21:50:45< mattsc> time_up is not defined at that point, so (not nil) also exits right away.
20130327 21:51:45< Espreon> Hmm, it seems my commits got associated with my account to some degree, but my name doesn't have a link to my profile.
20130327 21:51:54< Espreon> ... oh well.
20130327 21:52:04< Espreon> https://sourceforge.net/p/wesnoth/code/ci/043c4f9fd33ac5abac1e9df311d2ad3edf70db47/tree/
20130327 21:52:04-!- trewe_ [~trewe@87-196-129-238.net.novis.pt] has quit [Ping timeout: 245 seconds]
20130327 21:52:57< Octalot> mattsc: ah, ok.
20130327 21:53:45< mattsc> Octalot: sorry, I should have mentioned that.
20130327 21:53:55-!- prkc [~negusnyul@2E6B931C.dsl.pool.telekom.hu] has quit [Quit: Konversation terminated!]
20130327 21:57:26-!- trewe_ [~trewe@87.196.129.238] has joined #wesnoth-dev
20130327 22:01:28< AI0867> esr: it's completed its meditation, but we still can't see any branches other than master
20130327 22:02:09< esr> That's odd.  Let me check something.
20130327 22:03:23< AI0867> well, my pipeline currently consists of 3 parts:
20130327 22:04:16< AI0867> 1) small reimplementation of irkerhook.py (probably should be merged back in) that does absolutely no input massaging after retrieving it and mails it to me (also, everything is hardcoded)
20130327 22:04:26< AI0867> 2) a procmail entry on my server
20130327 22:04:27< esr> You can see the tracking branches in your local clone, yes?
20130327 22:04:31< AI0867> yes
20130327 22:05:06< AI0867> 3) A tiny (and hardcoded) filtering program that pipes the contents of the mail to irkerd
20130327 22:05:14-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has quit [Remote host closed the connection]
20130327 22:05:22< esr> Hm.  So this is strictly a problem with their web interface.  Maybe there's some option we need to set?
20130327 22:05:36< AI0867> I can't even access any options
20130327 22:05:46< AI0867> are you sure you didn't just set the "shell access" bit?
20130327 22:09:00-!- [Relic] [~relic@99-58-54-211.lightspeed.milwwi.sbcglobal.net] has joined #wesnoth-dev
20130327 22:10:30< esr> I set the only admin bit I have. :-)
20130327 22:11:13< AI0867> maybe it'll take 24 hours or something
20130327 22:11:27< AI0867> I know what sourceforge admin rights look like from wesnoth-umc-dev
20130327 22:11:37< AI0867> and I still don't have the same menus on wesnoth
20130327 22:12:11< esr> I just did "browse commits" which is taking long enough that it mabeforcing a re-index that will make the branches visible.
20130327 22:12:24-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has joined #wesnoth-dev
20130327 22:18:48< AI0867> gah, 2.4 is old enough that it doesn't even have python's version of the ternary operator yet
20130327 22:19:33< AI0867> you can replace the sha1 hash in the url with 'master' and it'll work, but not with '1.10' or the like
20130327 22:27:52< esr> AI0867: Sounds like you're backporting irkerhook.py.  I'll take that patch.
20130327 22:28:47-!- Gambit [~gambit@wesnoth/developer/grickit] has joined #wesnoth-dev
20130327 22:28:48< AI0867> that's what it might eventually turn into
20130327 22:29:07< AI0867> right now I'm removing all traces of modern python out of a much smaller script
20130327 22:29:22< AI0867> and trying to figure out why git rev-parse isn't working here
20130327 22:29:34< AI0867> 'here' = on the sf shell
20130327 22:30:22< AI0867> well, it's not that it doesn't work, it doesn't work from the python scripts
20130327 22:31:12< esr> Yesw, it looks like a full reindex is under way.
20130327 22:41:29-!- lipkab [~lipk@host-91-147-212-174.biatv.hu] has quit [Quit: Konversation terminated!]
20130327 22:44:11-!- _Coffee [~david@ppp118-210-51-209.lns20.adl2.internode.on.net] has quit [Ping timeout: 245 seconds]
20130327 22:46:09-!- Captain_Wrathbow [~Wrathbow@adsl-71-150-251-229.dsl.mdsnwi.sbcglobal.net] has joined #wesnoth-dev
20130327 22:48:45< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-1-g7ea91f7 / bar: bar
20130327 22:48:46< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-2-gf353a8b / foo: foo
20130327 22:48:48< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-3-g3f5e874 / file foo: contents
20130327 22:49:37< irker369> Alexander van Gessel (AI0867) wesnoth:wesnoth * 1.11.2-7-g2561c5b / data/tools/terrain2wiki.py: Use sourceforge rather than GNA for terrain image URLs
20130327 22:50:36< irker369> Alexander van Gessel (AI0867) wesnoth:master * 1.11.2-7-g2561c5b / data/tools/terrain2wiki.py: Use sourceforge rather than GNA for terrain image URLs
20130327 22:51:51< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-1-g7ea91f7 / bar: bar
20130327 22:51:52< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-2-gf353a8b / foo: foo
20130327 22:51:54< irker369> Alexander van Gessel (AI0867) wesnoth:master * original-3-g3f5e874 / file foo: contents
20130327 22:55:55-!- Alarantalara [~Adium@CPEc0c1c09e8055-CM00252eac6d62.cpe.net.cable.rogers.com] has quit [Quit: Leaving.]
20130327 22:57:02< AI0867> does anyone have a ready commit lying around?
20130327 23:02:31-!- witness [uid10044@gateway/web/irccloud.com/x-qrspdignwhkoyxrf] has joined #wesnoth-dev
20130327 23:04:16< AI0867> whatever
20130327 23:08:28< irker369> Alexander van Gessel (AI0867) wesnoth:master * 1.11.2-8-gf76cd24 / misc/fortunes/wesnoth: Restore some fortune cookies
20130327 23:08:42< AI0867> okay, so I still had to do that by hand
20130327 23:08:51< AI0867> not sure why
20130327 23:10:32< vultraz> what are the fortune cookies anyway?
20130327 23:10:38< bumbadadabum> cookies
20130327 23:10:43< bumbadadabum> that have fortunes in them
20130327 23:12:24< vultraz> ....Obvious bumbadadabum is obvious
20130327 23:28:32< shadowm> stikonas:  93% done.
20130327 23:28:59< shadowm> 1.7 GiB.
20130327 23:29:02< AI0867> and it looks like sourceforge only provides raw files as attachments
20130327 23:29:19< AI0867> so you can host images off the webinterface, but you can't show the changelog
20130327 23:29:24< shadowm> Downloading a less than 30 KiB/sec right now because quotas.
20130327 23:29:32< shadowm> *ar
20130327 23:29:39< shadowm> *arrrrrrr
20130327 23:29:42< shadowm> *at
20130327 23:30:47< irker369> AzAzQWxleGFuZGVyIHZhbiBHZXNzZWwgKEFJMDg2NykPIAJ3ZXNub3RoDzoDMDdtYXN0ZXIP
20130327 23:30:47< irker369> ICogAjEuMTEuMi05LWc4NjA5MTY2DyAvIAJwbGF5ZXJzX2NoYW5nZWxvZw86IFRlc3QgY29t
20130327 23:30:48< irker369> bWl0OiBSZXN0b3JlIGxpbmsgdG8gbWFpbiBjaGFuZ2Vsb2c=
20130327 23:30:59< AI0867> well, at least my hook is getting run...
20130327 23:31:10< stikonas> shadowm: nice :)
20130327 23:31:23< stikonas> dows wget show estimated time?
20130327 23:31:57< shadowm> With all this RF unstability, it's not very reliable, but it says 95 minutes.
20130327 23:32:42< stikonas> I'm not sure if I will stay that late...
20130327 23:33:25< shadowm> I can continue tomorrow.
20130327 23:33:50< stikonas> ok. It's already half past midnight here
20130327 23:34:07< stikonas> but you can still download for now
20130327 23:35:35-!- noy [~Noy@wesnoth/developer/noy] has quit [Read error: No route to host]
20130327 23:35:44-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20130327 23:36:33< fabi> master: master [rejected - non-fast-forward]
20130327 23:36:51< fabi> Can't do a commit&push.
20130327 23:37:23< AI0867> yeah, I did that
20130327 23:37:30-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 245 seconds]
20130327 23:37:33< AI0867> er, no, not that part
20130327 23:37:37< AI0867> non-fast-forward
20130327 23:37:42< AI0867> that mean you have to pull first
20130327 23:37:51< fabi> esr: ping
20130327 23:37:53< AI0867> your master is not a child of the origin's master
20130327 23:38:12< fabi> how can that be?
20130327 23:38:15< AI0867> but I *am* making the repo as read-only right now to test my script =P
20130327 23:38:25< AI0867> fabi: because stuff has happened on the origin since you last pulled
20130327 23:38:27< AI0867> so pull again
20130327 23:38:51< AI0867> then try to push again
20130327 23:38:59< AI0867> and then you should see a different error ;0
20130327 23:39:03< AI0867> s/0/)/
20130327 23:39:23< fabi> AI0867: What is the different between "fetch from upstream" and "pull from upstream"?
20130327 23:39:54< AI0867> fetch gets the commits
20130327 23:40:02< AI0867> pull fetches and then merges
20130327 23:40:14< AI0867> fetch does not affect your branch at all
20130327 23:43:50< stikonas> fabi: btw, if you want to avoid lots of merge commits, you can do git pull --rebase
20130327 23:44:31< stikonas> it will rebase your commits on top of remote commits instead of merging them
20130327 23:44:39< irker369> AzAzZmVuZHJpbg8gAndlc25vdGgPOgMwN21hc3Rlcg8gKiACMS4xMS4yLTgtZzc0ZjQ5NzIP
20130327 23:44:39< irker369> IC8gAnNyYy9lZGl0b3IvdG9vbGtpdC9lZGl0b3JfdG9vbGtpdC5jcHAPOiBFZGl0b3IgVG9v
20130327 23:44:40< irker369> bGtpdCBidXR0b25zIGJlaGF2ZSBsaWtlIHJhZGlvIGJ1dHRvbnMu
20130327 23:44:44< irker369> AzAzZmVuZHJpbg8gAndlc25vdGgPOgMwN21hc3Rlcg8gKiACMS4xMS4yLTEwLWc5OWE4NDFm
20130327 23:44:45< irker369> DyAvIAIPOiBNZXJnZSBicmFuY2ggJ21hc3Rlcicgb2Ygc3NoOi8vZmVuZHJpbkBnaXQuY29k
20130327 23:44:45< irker369> ZS5zZi5uZXQvcC93ZXNub3RoL2NvZGU=
20130327 23:45:13-!- Unnheulu [56ba860d@gateway/web/freenode/ip.86.186.134.13] has joined #wesnoth-dev
20130327 23:45:18< fabi> stikonas: no idea what you are talking about :-)
20130327 23:46:43< fabi> AI0867: remote rejected. How long are you going to take the lock?
20130327 23:47:08< fabi> s/take/keep
20130327 23:47:15< stikonas> fabi: git can't merge remote commits
20130327 23:47:39< stikonas> if somebody pushed before you, you must get remote commits first
20130327 23:47:56< stikonas> like I said, try git pull --rebase
20130327 23:49:03< stikonas> if you don't do --rebase then git will create a merge commit (e.g.see http://stikonas.homelinux.org/files/1.png)
20130327 23:49:29< irker369> AzAzQWxleGFuZGVyIHZhbiBHZXNzZWwgKEFJMDg2NykPIAJ3ZXNub3RoDzoDMDdtYXN0ZXIP
20130327 23:49:30< irker369> ICogAjEuMTEuMi05LWc4NjA5MTY2DyAvIAJwbGF5ZXJzX2NoYW5nZWxvZw86IFRlc3QgY29t
20130327 23:49:30< irker369> bWl0OiBSZXN0b3JlIGxpbmsgdG8gbWFpbiBjaGFuZ2Vsb2c=
20130327 23:49:37< AI0867> fabi: can you give me your entire response?
20130327 23:49:55< shadowm> stikonas: I tried teaching the Frogatto people about that (and the autorebase option as a last resort), but they didn't heed my advice for long.
20130327 23:50:28< AI0867> ugh
20130327 23:50:30< shadowm> They prefer to see merge commits to complicating (???) their workflow with rebase.
20130327 23:50:54< stikonas> I am not sure why people find it difficult. I'm not even a programmer and I don't find it difficult
20130327 23:51:33< irker369> AzAzZmVuZHJpbg8gAndlc25vdGgPOgMwN21hc3Rlcg8gKiACMS4xMS4yLTgtZzc0ZjQ5NzIP
20130327 23:51:33< stikonas> and you can do git pull --rebase even after git pull, it just have to be done before git push
20130327 23:51:34< irker369> IC8gAnNyYy9lZGl0b3IvdG9vbGtpdC9lZGl0b3JfdG9vbGtpdC5jcHAPOiBFZGl0b3IgVG9v
20130327 23:51:34< irker369> bGtpdCBidXR0b25zIGJlaGF2ZSBsaWtlIHJhZGlvIGJ1dHRvbnMu
20130327 23:51:38< irker369> AzAzZmVuZHJpbg8gAndlc25vdGgPOgMwN21hc3Rlcg8gKiACMS4xMS4yLTEwLWc5OWE4NDFm
20130327 23:51:39< irker369> DyAvIAIPOiBNZXJnZSBicmFuY2ggJ21hc3Rlcicgb2Ygc3NoOi8vZmVuZHJpbkBnaXQuY29k
20130327 23:51:39< irker369> ZS5zZi5uZXQvcC93ZXNub3RoL2NvZGU=
20130327 23:51:48< irker369> AzAzQWxleGFuZGVyIHZhbiBHZXNzZWwgKEFJMDg2NykPIAJ3ZXNub3RoDzoDMDdtYXN0ZXIP
20130327 23:51:49< irker369> ICogAjEuMTEuMi05LWc4NjA5MTY2DyAvIAJwbGF5ZXJzX2NoYW5nZWxvZw86IFRlc3QgY29t
20130327 23:51:49< irker369> bWl0OiBSZXN0b3JlIGxpbmsgdG8gbWFpbiBjaGFuZ2Vsb2c=
20130327 23:51:50< shadowm> stikonas: I think it's called laziness.
20130327 23:52:08< shadowm> AI0867: Is.... using a test channel other than this one not an option?
20130327 23:52:14< AI0867> possibly
20130327 23:52:14< stikonas> Everybody knows that wesnoth developers are not lazy :)
20130327 23:52:23< fabi> AI0867: http://pastebin.com/3tYYG6Ca
20130327 23:52:24< shadowm> Or will we have to see the Base64 spam until you are done?
20130327 23:52:32< AI0867> I don't get why it's not decoding it
20130327 23:52:49< AI0867> is sf's mail system silently base64 encoding everything but not setting the appropriate headers?
20130327 23:53:05< shadowm> stikonas: That's the thing, three of those people (i.e. everyone who weren't me or DDR at the time) were Wesnoth developers.
20130327 23:53:58< stikonas> that's strange, it seems that git hook does not accept commit as valid
20130327 23:56:33< shadowm> Byte 1,799,417,252 of 1,898,168,320.
20130327 23:56:58< shadowm> Now at 1,800,603,992.
20130327 23:57:00< stikonas> 100 MiB left
20130327 23:57:02< irker369> Alexander van Gessel (AI0867) wesnoth:master * 1.11.2-9-g8609166 / players_changelog: Test commit: Restore link to main changelog
20130327 23:57:11< AI0867> okay, getting there
20130327 23:58:15< AI0867> fabi: I've unlocked it
20130327 23:58:19< fabi> stikonas: I guess it is because AI0867 locked the repository.
20130327 23:58:36< stikonas> oh, I see
20130327 23:58:45< irker369> fendrin wesnoth:master * 1.11.2-8-g74f4972 / src/editor/toolkit/editor_toolkit.cpp: Editor Toolkit buttons behave like radio buttons.
20130327 23:58:46< AI0867> yeah, I wanted to not have to make up a new commit every test
20130327 23:58:47< irker369> fendrin wesnoth:master * 1.11.2-10-g99a841f / : Merge branch 'master' of ssh://fendrin@git.code.sf.net/p/wesnoth/code
20130327 23:59:39< shadowm> Ugly merge commit.
20130327 23:59:48 * shadowm is not impressed.
--- Log closed Thu Mar 28 00:00:06 2013