--- Log opened Wed Apr 09 00:00:53 2014
20140409 00:09:51-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has joined #wesnoth-dev
20140409 00:16:27-!- neXyon [~neXyon@85-127-47-28.dynamic.xdsl-line.inode.at] has quit [Quit: bye]
20140409 00:17:44< aquileia> Is there any reason why the linker might have problems with "game_config_manager"? http://pastebin.com/vGjZbNNH
20140409 00:19:50< aquileia> I have no idea why it compiled without the unit tests and errors when I try to link it to wesnothtest...
20140409 00:20:19-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140409 00:20:41< aquileia> I mean - why should a project error due to a change in another subproject?
20140409 00:28:24-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: End Transmission.]
20140409 00:41:39-!- vultraz [~chatzilla@124.109.10.167] has quit [Read error: Connection reset by peer]
20140409 00:41:42-!- vultraz_ [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20140409 00:41:56-!- vultraz_ is now known as vultraz
20140409 00:50:16< aquileia> oh, damn... one of the tests had the same name, and the obj files were stored in the same directory! But when I moved the tests to a subdir, it tried to find the lualib there as well and failed. I hope to figuer this out, let's see...
20140409 01:09:30-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 01:17:30-!- matthiaskrgr [matthiaskr@gateway/shell/panicbnc/x-kxvejgnurotgwlkj] has quit [Quit: PanicBNC - https://PanicBNC.net - currently sucks]
20140409 01:23:01-!- irker791 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140409 01:23:01< irker791> wesnoth: Chris Beck wesnoth:1.12 f12c835ceee7 / src/game.cpp: fix whitespace http://git.io/cbxkuw
20140409 01:23:01< irker791> wesnoth: Chris Beck wesnoth:1.12 d2d6a9d91558 / src/playsingle_controller.cpp: fixup idle loop http://git.io/8lNuTw
20140409 01:23:02< irker791> wesnoth: Chris Beck wesnoth:1.12 9ab917b61781 / src/ (playmp_controller.cpp playmp_controller.hpp playsingle_controller.hpp): fix networked idle loop http://git.io/Aju94A
20140409 01:23:03< irker791> wesnoth: Chris Beck wesnoth:1.12 348ba7779250 / src/ (team.cpp team.hpp): create "fight_on_without_leader" field of [side] http://git.io/A_L0Sw
20140409 01:23:04< irker791> wesnoth: Chris Beck wesnoth:1.12 04eaedfe9066 / src/play_controller.cpp: add debugging output http://git.io/rUuCww
20140409 01:23:05< irker791> wesnoth: Chris Beck wesnoth:1.12 69e5b15fe2d1 / src/ (ai/testing.cpp ai/testing.hpp play_controller.cpp): refactor check_victory, don't use [side] no_leader attribute http://git.io/gENE-w
20140409 01:23:07< irker791> wesnoth: Chris Beck wesnoth:1.12 c7e5b7373035 / src/playturn.cpp: [side_drop] sides go to idle, not human (fix bug 21459) http://git.io/zofezg
20140409 01:23:09< irker791> wesnoth: Chris Beck wesnoth:1.12 117c5b349271 / changelog: update changelog http://git.io/fcN-ig
20140409 01:24:29< irker791> wesnoth: Chris Beck wesnoth:master eae7ba9c5cbd / src/game.cpp: fix whitespace http://git.io/qBIoAQ
20140409 01:24:31< irker791> wesnoth: Chris Beck wesnoth:master 22c565282a93 / src/playsingle_controller.cpp: fixup idle loop http://git.io/fEMCfg
20140409 01:24:33< irker791> wesnoth: Chris Beck wesnoth:master 3f4879ace449 / src/ (playmp_controller.cpp playmp_controller.hpp playsingle_controller.hpp): fix networked idle loop http://git.io/f0JF9g
20140409 01:24:35< irker791> wesnoth: Chris Beck wesnoth:master ec379ce355bb / src/ (team.cpp team.hpp): create "fight_on_without_leader" field of [side] http://git.io/EH66Lw
20140409 01:24:37< irker791> wesnoth: Chris Beck wesnoth:master 8c69c7373918 / src/play_controller.cpp: add debugging output http://git.io/Dlurcw
20140409 01:24:39< irker791> wesnoth: Chris Beck wesnoth:master 81f9e29e403a / src/ (ai/testing.cpp ai/testing.hpp play_controller.cpp): refactor check_victory, don't use [side] no_leader attribute http://git.io/L9rURQ
20140409 01:24:41< irker791> wesnoth: Chris Beck wesnoth:master 55e716428de6 / src/playturn.cpp: [side_drop] sides go to idle, not human (fix bug 21459) http://git.io/7zFupg
20140409 01:24:43< irker791> wesnoth: Chris Beck wesnoth:master 84bad5a87501 / changelog: update changelog http://git.io/tt5ogg
20140409 01:25:03-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 01:27:12-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on 1.12 | 230 bugs, 352 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140409 01:27:31-!- matthiaskrgr [matthiaskr@gateway/shell/panicbnc/x-rujgexwqeagnilrb] has joined #wesnoth-dev
20140409 01:37:50-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 252 seconds]
20140409 01:46:47< iceiceice> hmm github is not as smart as i thought it was :O
20140409 01:47:04< shadowm> vultraz: I must say, this is the first time ever since I first found Wesnoth that I have seen the first screenshot in http://wiki.wesnoth.org/Play not be Delfador striking enemies with lightning in HttT scenario 1.
20140409 01:48:57< shadowm> But the gate looks cool too, so whatever.
20140409 01:51:58< iceiceice> so i'm a bit concerned now: i had a bunch of pull requests that i was slowly modifying and rebasing, collecting feedback etc.
20140409 01:52:08< iceiceice> i decided to make a branch "testing" and cherry-pick commits from them
20140409 01:52:17< iceiceice> then i tested it thoroughly
20140409 01:52:30< iceiceice> and finally cherry-picked it into 1.12 and merged into master
20140409 01:52:41< iceiceice> now i thought the pull requests that were subsumed would become blank
20140409 01:52:45< iceiceice> since their changes are vacuous
20140409 01:53:14< iceiceice> but github seems to think they are still just the same as always and ready to be automatically merged since they got the travis seal of approval
20140409 01:53:20< iceiceice> what will happen if i click the button?
20140409 01:55:34< iceiceice> i hope that it would suddenly decide to grey the button once it gets a merge conflict... or at least tell me that something unusual happened
20140409 01:56:00< iceiceice> i mean really i thought all the pull requests would get updated once master changed, i've seen that happen before
20140409 01:59:07-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140409 02:00:55-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 02:02:00-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has quit [Quit: Page closed]
20140409 02:08:01< vultraz> shadowm: exactly. I felt that HttT S1 was used too much
20140409 02:08:52-!- fendrin [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20140409 02:10:15< vultraz> Screenshots from other campaigns uploaded would be nice, too, for use other places in the wiki
20140409 02:16:09-!- Ivanovic_ [~ivanovic@x2f425a6.dyn.telefonica.de] has joined #wesnoth-dev
20140409 02:19:14-!- Ivanovic [~ivanovic@x2f3e032.dyn.telefonica.de] has quit [Ping timeout: 240 seconds]
20140409 02:20:02-!- Ivanovic_ is now known as Ivanovic
20140409 02:26:20-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Remote host closed the connection]
20140409 02:26:52-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140409 02:34:23< vultraz> "The BFW project team only officially releases the source code. Binary packages are only provided by community volunteers and hosted here." should this wording be changed? It's not really the case :/
20140409 02:34:33< vultraz> (w.w.o/Downloads)
20140409 02:37:07-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140409 02:37:44-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 02:41:29-!- gfgtdf [~chatzilla@f054050123.adsl.alicedsl.de] has joined #wesnoth-dev
20140409 02:41:47< gfgtdf> iceiceice: so you cherrypicked your commits from your pr into original master ?
20140409 02:41:54-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 02:52:03-!- gfgtdf [~chatzilla@f054050123.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]]
20140409 02:54:33-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 03:00:51-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 03:01:25< iceiceice> y, i thought it woudl be smart and realize they are the same commits
20140409 03:01:37< iceiceice> or at least realize that merging the copies into master now wouldn't do anything
20140409 03:01:45< iceiceice> but it was fixed at least after rebase
20140409 03:04:43-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has quit [Remote host closed the connection]
20140409 03:06:34-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has joined #wesnoth-dev
20140409 03:15:11-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 03:32:39-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 03:36:18-!- _8680_ [~8680@2002:4404:712c:0:1859:8244:bb20:9700] has quit [Ping timeout: 240 seconds]
20140409 03:36:55-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 03:37:29-!- _8680_ [~8680@2002:4404:712c:0:e8c3:616a:f90c:b9bb] has joined #wesnoth-dev
20140409 03:41:28-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 03:51:01-!- matthiaskrgr [matthiaskr@gateway/shell/panicbnc/x-rujgexwqeagnilrb] has quit [Changing host]
20140409 03:51:01-!- matthiaskrgr [matthiaskr@unaffiliated/matthiaskrgr] has joined #wesnoth-dev
20140409 03:51:01-!- matthiaskrgr [matthiaskr@unaffiliated/matthiaskrgr] has quit [Changing host]
20140409 03:51:01-!- matthiaskrgr [matthiaskr@gateway/shell/panicbnc/x-rujgexwqeagnilrb] has joined #wesnoth-dev
20140409 04:05:59-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 255 seconds]
20140409 04:07:47-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Ping timeout: 246 seconds]
20140409 04:11:36-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 04:18:26-!- _8680_ [~8680@2002:4404:712c:0:e8c3:616a:f90c:b9bb] has quit [Ping timeout: 246 seconds]
20140409 04:19:38-!- _8680_ [~8680@2002:4404:712c:0:8449:32eb:33fa:fe7f] has joined #wesnoth-dev
20140409 04:20:46-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 04:21:01-!- exciton_ [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140409 04:21:16-!- exciton [chuck-the-@77.51.168.182] has joined #wesnoth-dev
20140409 04:26:09-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20140409 04:29:53-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Ciao]
20140409 04:31:20-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 04:38:07-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 240 seconds]
20140409 04:39:28-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20140409 04:40:16-!- Necrosporus_ [~Necrospor@unaffiliated/necrosporus] has joined #wesnoth-dev
20140409 04:43:29-!- Necrosporus [~Necrospor@unaffiliated/necrosporus] has quit [Ping timeout: 264 seconds]
20140409 04:47:43-!- exciton [chuck-the-@77.51.168.182] has quit [Read error: Connection reset by peer]
20140409 04:47:57-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140409 05:03:09-!- irker791 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140409 05:18:37-!- cib0 [~cib@p5DD207EC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140409 05:20:51-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 05:25:19-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 05:39:22-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140409 06:01:09-!- vultraz [~chatzilla@124.109.10.167] has quit [Ping timeout: 252 seconds]
20140409 06:01:56-!- vultraz [~chatzilla@124.109.10.167] has joined #wesnoth-dev
20140409 06:17:38-!- irker472 [~irker@fehu.ai0867.net] has joined #wesnoth-dev
20140409 06:17:38< irker472> wesnoth: aquileia wesnoth:master edfe6e4a744a / / (5 files in 3 dirs): Update projectfiles and corresponding readme http://git.io/ORSEzQ
20140409 06:17:38< irker472> wesnoth: Alexander van Gessel wesnoth:master 45f52737c89e / / (5 files in 3 dirs): Merge pull request #139 from aquileia/VC_project_update http://git.io/NJoofA
20140409 06:22:10< AI0867> iceiceice: if you cherry-pick the commits, then while master now has the same contents, it does not contain the commits themselves (that is, the PR's branch is not an ancestor of master), so they're not automatically considered merged
20140409 06:24:43< iceiceice> i see
20140409 06:24:56< iceiceice> but shouldnt the git hub display of the PR show that the content was in there?
20140409 06:25:18< iceiceice> if i cherry pick a commit and then try to merge it after,
20140409 06:25:25< iceiceice> that shoudlnt display as new content...
20140409 06:25:30< iceiceice> it should either be a merge commit
20140409 06:25:31< iceiceice> or no change
20140409 06:25:47< iceiceice> s/merge commit/merge conflict
20140409 06:26:07< iceiceice> i still think the github behavior was bizarre 
20140409 06:27:34< AI0867> not really
20140409 06:27:53< AI0867> the "this has been merged" heuristic is based on history of branches
20140409 06:28:02< AI0867> it does *not* check for contents
20140409 06:28:17< AI0867> and cherry-picking creates new commits
20140409 06:28:39< AI0867> commits that usually have the same content as a different commit (unless you have to hand-merge things), but have a different ancestor
20140409 06:29:32< iceiceice> wait so you are telling me that when i review a pull request
20140409 06:29:42< iceiceice> the lines that github shows me have nothing to do with the diff that will happen if i click merge?
20140409 06:30:18< iceiceice> whenever i review a pull request, i *always* will expect that i am seeing the literal results of "git diff" for master and topic
20140409 06:30:26< iceiceice> anything else would be bizarre
20140409 06:31:10< AI0867> it's a diff between master OR AN ANCESTOR OF MASTER (wherever the branch was based off of) and the head of the branch
20140409 06:31:17< iceiceice> thats horrible
20140409 06:31:31< AI0867> no, it's the diff that you submit
20140409 06:31:42< iceiceice> fine but its not useful to the reviewer
20140409 06:31:55< AI0867> if things have changed locally, then that's relevant for merging, but not for reviewing the commits
20140409 06:32:30< iceiceice> i dont understand your last sentence
20140409 06:32:31< AI0867> s/locally/in master/
20140409 06:32:46< AI0867> say that master points to commit 1234
20140409 06:32:56< iceiceice> the purpose of a pull request is for the owner of a repo to decide whether to merge into upstream
20140409 06:33:10< iceiceice> not to advise the pull requestee whether they might merge into their local copy of master etc.
20140409 06:33:17< iceiceice> so the info should be relevant to that operation
20140409 06:33:17< AI0867> someone creates a feature branch at this point, and creates commits 23,24 and 25 based on 1234
20140409 06:33:19< iceiceice> the pul
20140409 06:33:21< iceiceice> the pull
20140409 06:33:27< iceiceice> i understand what you are saying the system is
20140409 06:33:34< AI0867> someone else pushes commit 98313 to master
20140409 06:33:34< iceiceice> but that is very poorly conceived himo
20140409 06:34:09< AI0867> if you create a pull request based on that feature branch, you're asking for the series of 1234..25 to be merged
20140409 06:34:15< AI0867> so that's what's shown for review
20140409 06:34:59< AI0867> the fact that in master, 98313 also exists, and may or may not cause merge conflicts, is irrelevant for the purpose of reviewing the patch.
20140409 06:37:34< AI0867> 08:32 < iceiceice> the purpose of a pull request is for the owner of a repo to decide whether to merge into upstream ← yes, and for that, the main consideration is what the changes are
20140409 06:37:41< AI0867> 08:33 < iceiceice> not to advise the pull requestee whether they might merge into their local copy of master etc. ← ???
20140409 06:37:45-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 06:38:23< iceiceice> AI: let me try to explain again
20140409 06:38:40< iceiceice> if i make a pull request based on fork topic to go into upstream master
20140409 06:38:49< iceiceice> the diff that is shown should be between fork topic and upstream master
20140409 06:38:53< AI0867> no
20140409 06:39:04< iceiceice> ok, what should it be instead
20140409 06:39:05< AI0867> that diff will constantly changed as people push to upstream master
20140409 06:39:10< iceiceice> yes
20140409 06:39:12< iceiceice> that is what i want
20140409 06:39:24< AI0867> that's very much not what you committed to your fork topic
20140409 06:39:25< iceiceice> i want it to show me what will happen if i click the button
20140409 06:39:36< iceiceice> if i want the difference between fork topic and fork master,
20140409 06:39:39< iceiceice> i will click "compare"
20140409 06:39:45< AI0867> that's not what will happen
20140409 06:39:49< iceiceice> ok well,
20140409 06:39:51< AI0867> diff master topic
20140409 06:39:58< iceiceice> the pull request defintiely updates when fork topic changes
20140409 06:40:03< iceiceice> and sometimes when master changes
20140409 06:40:06< iceiceice> just not at other timer
20140409 06:40:08< iceiceice> other times
20140409 06:40:08< AI0867> a merge is not the same thing as that
20140409 06:40:35< AI0867> the "merge this automatically" button can change when master updates
20140409 06:40:39< AI0867> but nothing else
20140409 06:41:15< AI0867> a diff between master and the topic is not useful, because a merge incorporates *both* sets of history
20140409 06:41:24< AI0867> such a diff would show you undoing other people's work
20140409 06:42:18-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 06:44:27< iceiceice> ok but when fast forward merge is possible
20140409 06:44:29-!- Gallaecio [~quassel@84.120.115.132.dyn.user.ono.com] has quit [Read error: Connection reset by peer]
20140409 06:44:36< iceiceice> ideally github would show that
20140409 06:44:46< iceiceice> it seems to be inconsistent about this
20140409 06:47:21< iceiceice> i see so i guess the cherry-picked commit cant be fastfowarded, is that the issue?
20140409 06:50:44< AI0867> 20140408 16:54:54 error filesystem: Failed to retrieve file status for /home/travis/.local: No such file or directory
20140409 06:50:47< AI0867> 20140408 16:54:54 error filesystem: Could not create parents to /home/travis/.local/share
20140409 06:50:57< AI0867> I somehow got that on infinite loop on travis
20140409 06:51:09< AI0867> iceiceice: yes
20140409 06:51:19< AI0867> a fast-forward is shown as already merged
20140409 06:51:31< iceiceice> hmm...
20140409 06:51:36< AI0867> but cherry-pick creates a new commit, even if it contains the exact same content
20140409 06:52:00< iceiceice> y i guess it makes sense that i have to rebase eveyrthign then
20140409 06:52:25< iceiceice> if i do this cherry pick random development material to "testing" and merge that, i'm writing a new history
20140409 06:56:16< AI0867> yes
20140409 06:57:19-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 240 seconds]
20140409 07:01:13-!- cib0 [~cib@p5DD207EC.dip0.t-ipconnect.de] has quit [Read error: Connection reset by peer]
20140409 07:05:34-!- cib0 [~cib@p5DD207EC.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140409 07:12:56-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140409 07:14:15-!- Necrosporus_ is now known as Necrosporus
20140409 07:17:35-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev
20140409 07:18:27-!- cib [~cib@p5DD20F9E.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140409 07:18:49-!- cib is now known as Guest63444
20140409 07:20:03-!- cib0 [~cib@p5DD207EC.dip0.t-ipconnect.de] has quit [Disconnected by services]
20140409 07:20:09-!- Guest63444 is now known as cib0
20140409 07:22:29-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140409 07:27:12-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on 1.12 | 228 bugs, 352 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140409 07:35:06-!- _8680_ [~8680@2002:4404:712c:0:8449:32eb:33fa:fe7f] has quit [Ping timeout: 240 seconds]
20140409 07:36:05-!- _8680_ [~8680@2002:4404:712c:0:b9c2:8150:9880:aeac] has joined #wesnoth-dev
20140409 07:40:01-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has joined #wesnoth-dev
20140409 07:42:58-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20140409 08:06:20-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140409 08:14:32-!- bagzie [~bag@85-76-87-133-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140409 08:16:37-!- EdB [~edb@85.69.242.6] has joined #wesnoth-dev
20140409 08:17:54-!- Octalot [~noct@27.74.208.46.dyn.plus.net] has joined #wesnoth-dev
20140409 08:18:23-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has quit [Ping timeout: 246 seconds]
20140409 08:25:00-!- EliDupree [~quassel@66-189-34-122.dhcp.oxfr.ma.charter.com] has joined #wesnoth-dev
20140409 08:27:38-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20140409 08:29:52-!- mjs-de [~mjs-de@f048013210.adsl.alicedsl.de] has joined #wesnoth-dev
20140409 09:29:44-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has joined #wesnoth-dev
20140409 09:32:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140409 09:34:08-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 10:12:49-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has joined #wesnoth-dev
20140409 10:30:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Quit: Konversation terminated!]
20140409 10:39:30-!- _8680_ [~8680@2002:4404:712c:0:b9c2:8150:9880:aeac] has quit [Ping timeout: 240 seconds]
20140409 10:40:18-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Ping timeout: 240 seconds]
20140409 10:40:29-!- _8680_ [~8680@2002:4404:712c:0:247c:c81e:7d91:4532] has joined #wesnoth-dev
20140409 10:55:43-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has quit [Quit: Page closed]
20140409 11:01:30-!- _8680_ [~8680@2002:4404:712c:0:247c:c81e:7d91:4532] has quit [Ping timeout: 240 seconds]
20140409 11:02:38-!- _8680_ [~8680@2002:4404:712c:0:d0c0:89f6:8a97:9089] has joined #wesnoth-dev
20140409 11:02:40-!- travis-ci [~travis-ci@ec2-54-221-190-167.compute-1.amazonaws.com] has joined #wesnoth-dev
20140409 11:02:40< travis-ci> [travis-ci] AI0867/wesnoth#36 (boost_filesystem - a5d5266 : Alexander van Gessel): The build passed.
20140409 11:02:40< travis-ci> [travis-ci] Build details : http://travis-ci.org/AI0867/wesnoth/builds/22598352
20140409 11:02:40-!- travis-ci [~travis-ci@ec2-54-221-190-167.compute-1.amazonaws.com] has left #wesnoth-dev []
20140409 11:23:44-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has joined #wesnoth-dev
20140409 11:23:50< AI0867> Necrosporus: this is a bit opaquely written, but it makes my point:
20140409 11:23:52< AI0867> Q: Why the library does not convert the string from Locale's encoding not UTF-8 and wise versa on POSIX systems
20140409 11:23:55< AI0867> A: It is inherently incorrect to convert strings to/from locale encodings on POSIX platforms.
20140409 11:23:59< AI0867> You can create a file named "\xFF\xFF.txt" (invalid UTF-8), remove it, pass its name as a parameter to program and it would work whether the current locale is UTF-8 locale or not. Also changing the locale from let's say en_US.UTF-8 to en_US.ISO-8859-1 would not magically change all files in OS or the strings a user may pass to the program (which is different on Windows)
20140409 11:24:04< AI0867> POSIX OSs treat strings as NUL terminated cookies.
20140409 11:24:36< AI0867> So altering their content according to the locale would actually lead to incorrect behavior.
20140409 11:25:00< AI0867> The meaning of a locale under POSIX and Windows paltforms is different and has very different effects.
20140409 11:25:10< AI0867> from http://cppcms.com/files/nowide/html/
20140409 11:27:39-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 11:29:49-!- EdB [~edb@85.69.242.6] has quit [Quit: Konversation terminated!]
20140409 11:37:36-!- DCW [~Thunderbi@cpc66863-finc15-2-0-cust393.4-2.cable.virginm.net] has quit [Quit: DCW]
20140409 11:44:41-!- travis-ci [~travis-ci@ec2-54-221-190-167.compute-1.amazonaws.com] has joined #wesnoth-dev
20140409 11:44:41< travis-ci> [travis-ci] AI0867/wesnoth#37 (boost_filesystem_win - 7e88eda : Alexander van Gessel): The build passed.
20140409 11:44:41< travis-ci> [travis-ci] Build details : http://travis-ci.org/AI0867/wesnoth/builds/22600612
20140409 11:44:41-!- travis-ci [~travis-ci@ec2-54-221-190-167.compute-1.amazonaws.com] has left #wesnoth-dev []
20140409 11:52:16-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140409 11:59:06-!- loonycyborg_ [~loonycybo@wesnoth/developer/loonycyborg] has joined #wesnoth-dev
20140409 11:59:33-!- loonycyborg [~sergey@wesnoth/developer/loonycyborg] has quit [Quit: I ATE'NT DEAD]
20140409 11:59:33-!- loonycyborg_ is now known as loonycyborg
20140409 12:15:49-!- sachith500 [~kvirc@112.134.80.158] has joined #wesnoth-dev
20140409 12:23:13-!- sachith500 [~kvirc@112.134.80.158] has quit [Read error: Connection reset by peer]
20140409 12:37:49-!- sachith500 [~kvirc@112.134.80.158] has joined #wesnoth-dev
20140409 12:45:52-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 12:46:42-!- sachith500 [~kvirc@112.134.80.158] has quit [Read error: Connection reset by peer]
20140409 12:48:44-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140409 12:49:33-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 250 seconds]
20140409 12:50:20-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140409 12:50:54-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 12:52:01-!- Kexoth [~kex@89.205.75.19] has quit [Read error: Connection reset by peer]
20140409 12:52:12-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 13:18:07-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has joined #wesnoth-dev
20140409 13:19:21-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140409 13:20:12< aquileia> AI0867: In filesystem_boost you use game_config::preferences_dir, but there is only game_config::default_preferences_path
20140409 13:22:34< aquileia> filesystem.cpp used PREFERENCES_DIR for that task, perhaps you need to revive that macro
20140409 13:23:07 * AI0867 checks
20140409 13:25:00< AI0867> aquileia: I got rid of that a while back because it's duplicate storage of another variable. I seem to have missed this when I updated my branch (because it's in a _WIN32 #ifdef)
20140409 13:25:45< aquileia> AI0867: that's the reason why you have someone to check these things for you ;)
20140409 13:27:41-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Ping timeout: 250 seconds]
20140409 13:28:33-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 265 seconds]
20140409 13:29:00-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20140409 13:29:11< AI0867> aquileia: pushed to boost_filesystem
20140409 13:30:05< aquileia> AI0867: would you push it to boost_filesystem_win as well or should I switch branches?
20140409 13:31:26-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140409 13:31:58< aquileia> mattsc: There's some feedback for you in an unusual subforum http://r.wesnoth.org/p569208
20140409 13:32:37-!- Kevin_Xi [~kevin@223.72.182.186] has joined #wesnoth-dev
20140409 13:33:11< aquileia> At least it could be interpreted as feedback
20140409 13:33:17< mattsc> aquileia: thanks, I had just seen that.
20140409 13:35:25< mattsc> aquileia: ha, fun.  Yeah, I think the strategy he’s using doesn’t work in the reworked scenario.  :)
20140409 13:36:15< mattsc> And not just on nightmare, it doesn’t work on the medium level either.
20140409 13:40:40< aquileia> AI0867: never mind, I pasted the fix to my local copy
20140409 13:55:13-!- Kevin_Xi [~kevin@223.72.182.186] has quit [Quit: restart]
20140409 14:01:56-!- molgrum [~molgrum@212.85.89.43] has quit [Quit: Lämnar]
20140409 14:02:58-!- Dugi [93fbd156@gateway/web/freenode/ip.147.251.209.86] has joined #wesnoth-dev
20140409 14:03:07< Dugi> wesbot: seen mordante
20140409 14:03:07< wesbot> Dugi: The person with the nick mordante last spoke 3d ago.  3d ago was here and on the channel #wesnoth-de with the message: Quit: Leaving
20140409 14:03:44-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Remote host closed the connection]
20140409 14:06:16-!- molgrum [~molgrum@212.85.89.43] has joined #wesnoth-dev
20140409 14:12:48-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 14:14:12-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20140409 14:23:43-!- cib0 [~cib@p5DD20F9E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20140409 14:24:09-!- cib0 [~cib@p5DD20F9E.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140409 14:28:19-!- Kevin_Xi [~kevin@223.72.182.186] has joined #wesnoth-dev
20140409 14:32:18-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140409 14:35:41-!- bagzie [~bag@85-76-87-133-nat.elisa-mobile.fi] has quit []
20140409 14:37:54-!- molgrum [~molgrum@212.85.89.43] has quit [Quit: Lämnar]
20140409 14:43:09< AI0867> gfgtdf: has someone who is actually familiar with that part of the code seen PR #141?
20140409 14:45:52-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 14:56:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140409 14:57:18-!- cib0 [~cib@p5DD20F9E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20140409 14:59:08< aquileia> AI0867: http://pastebin.com/E7y3x8yw
20140409 15:05:16< AI0867> aquileia: that's a rather unhelpful set of error messages
20140409 15:05:43< aquileia> what should I provide instead?
20140409 15:06:23< AI0867> no, it's probably the best you can deliver
20140409 15:06:27< AI0867> but MSVC is rather unhelpful =/
20140409 15:06:29< aquileia> It quits due to an unhandled exception originating from boost::filesystem::path_traits
20140409 15:06:59< irker472> wesnoth: Alexander van Gessel wesnoth:master 409744c449ec / src/editor/map/context_manager.cpp: Wrap some directory composition in get_dir http://git.io/tkzQXg
20140409 15:07:01< irker472> wesnoth: Alexander van Gessel wesnoth:master 364a756b9351 / / (60 files in 14 dirs): Merge branch 'master' of github.com:wesnoth/wesnoth http://git.io/TIbEZw
20140409 15:09:24< aquileia> my testcase was std::string c = "G:/GitHub/wesnoth-old\\userdata/saves/Wesnoth_Tutorial_ä.gz"
20140409 15:10:03< AI0867> aquileia: I see. I think this is an encoding issue that is hit before it ever tries to stream a file
20140409 15:10:14< AI0867> er, where did you put that?
20140409 15:10:47-!- cib0 [~cib@p5DD20F9E.dip0.t-ipconnect.de] has joined #wesnoth-dev
20140409 15:10:59< aquileia> I entered it in the save dialog (the c= was only to say how the variable looked)
20140409 15:11:32< aquileia> oh, and it's a const reference
20140409 15:14:53-!- sachith500 [~kvirc@112.134.80.158] has joined #wesnoth-dev
20140409 15:16:36< AI0867> aquileia: I'll have a patch for you soon
20140409 15:17:18< aquileia> AI0867: Ah, I was trying to find it as well, but you were faster, then
20140409 15:18:15< AI0867> (my attempt is to imbue boost::filesystem::path with a utf8 locale during static initialization)
20140409 15:19:17< AI0867> pushed to boost_filesystem_win
20140409 15:20:19-!- _8680_ [~8680@2002:4404:712c:0:d0c0:89f6:8a97:9089] has quit [Ping timeout: 240 seconds]
20140409 15:21:10-!- _8680_ [~8680@2002:4404:712c:0:f86c:65cc:abfd:7376] has joined #wesnoth-dev
20140409 15:22:24-!- gfgtdf [~chatzilla@f054050123.adsl.alicedsl.de] has joined #wesnoth-dev
20140409 15:23:03< gfgtdf> AI0867: 141 is net ready yet
20140409 15:23:08< gfgtdf> not*
20140409 15:24:51-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 15:32:21< Necrosporus> AI0867, this one seem to explain issue better: http://www.dwheeler.com/essays/fixing-unix-linux-filenames.html
20140409 15:33:22< Necrosporus> You can create a file named "\xFF\xFF.txt" (invalid UTF-8), remove it, pass its name as a parameter to program and it would work whether the current locale is UTF-8 locale or not.  < this is correct
20140409 15:34:07< Necrosporus> And next sentence is correct too
20140409 15:35:07< Necrosporus> > It would work with ANY locale and changing the strings would lead to incorrect behavior. < this is incorrect
20140409 15:35:30< Necrosporus> it would, yes, but changing the strings correctly will not lead to incorrect behavior
20140409 15:37:59< Necrosporus> imagine another example. User passes a string, your task is to create file, named _"hello $string.txt", where hello is to be written on currently selected language
20140409 15:38:25-!- neXyon [~neXyon@85-127-73-255.dynamic.xdsl-line.inode.at] has joined #wesnoth-dev
20140409 15:40:40< Necrosporus> If locale is ru_RU.UTF8, then string representing 'hello' in Russian should be taken from the dictionary as it is (assuming it's itself utf8), but if locale is ru_RU.koi8r, it should be first converted into KOI8-R
20140409 15:42:03< Necrosporus> Though the problem is many programs tend to assume the filesystem is always UTF-8, despite locale settings, so in modern GNU/Linux, non-utf8 locales tend to cause various problems with application, as some of them would throw files in locale encoding and others in utf8
20140409 15:42:25-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has joined #wesnoth-dev
20140409 15:42:43< AI0867> Necrosporus: yes, having opaque strings as filenames does have some issues
20140409 15:43:04< AI0867> but what were we arguing about at this point?
20140409 15:45:52< Necrosporus> AI0867, creating files in locale encoding is backward-compatible with old software, creating files in UTF-8 is compatible with transferring files onto other machines, which assume filenames to be utf-8
20140409 15:46:31< Necrosporus> So you are either way you are losing something
20140409 15:47:37< Necrosporus> if you pick UTF8, you losing compatibility with a lot of programs including ls
20140409 15:49:03< Necrosporus> If you pick locale encoding, you add to common problem of transferring files to other machines with nfs and tar
20140409 15:50:40< Necrosporus> I think there should a standard be made. Should programs like ls convert filenames before outputting them to terminal or not
20140409 15:51:23< elias> the standard is utf8 I guess, problem is just things from before the standard
20140409 15:52:43< Necrosporus> elias, or maybe things from after the standard, which do not respect locale settings and always throwing utf8 out?
20140409 15:53:03-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 15:54:21< elias> hm, yes, I guess what I meant is, I agree there should be a standard (which should be to only allow utf8)
20140409 15:58:42< Necrosporus> elias, then example on this page http://cppcms.com/files/nowide/html/ won't work
20140409 15:59:55< aquileia> AI0867: The new bug is fixed, I can save with non-ASCII again
20140409 16:00:21< aquileia> Next in line is to test it for the installation directory
20140409 16:01:36< Necrosporus> assume I call it with rm foo.txt and 'foo.txt' is encoded as locale suggest (e.g. koi8r), then if program passes it as it is to unlink system call, then the system will report 'file not found', because the call is assumed to be encoded as utf8
20140409 16:02:45< Necrosporus> elias, so such standard is going to break non-utf8 locales  unless a lot of programs are changed
20140409 16:05:48< aquileia> gfgtdf: By the way https://github.com/gfgtdf/wesnoth-old/commit/edd70edde545258b73d85ea2329b37a9b45aaf9d is already fixed in master
20140409 16:07:56< Necrosporus> elias, so you are basically proposing to throw out non-utf8 locales completely
20140409 16:08:19-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 16:10:25< Necrosporus> AI0867, what do you think is better, to throw out support fo non-utf8 locales or to fix programs which do not support them?
20140409 16:17:35< gfgtdf> aquileia: you know why that doesn't casue merge conflicts ? Or why teh merge button is stil green ?
20140409 16:17:59< aquileia> No idea...
20140409 16:22:14< iceiceice> gfgtdf: i guess its the same reason my PRs were green yesterday
20140409 16:22:29-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 16:22:34< gfgtdf> iceiceice: what happend when you pressed merge ?
20140409 16:22:45< iceiceice> i never did but i guess that it basically wouldnt have changed anything
20140409 16:22:55< iceiceice> its like...
20140409 16:22:58< iceiceice> you have a diamond shape
20140409 16:23:05< iceiceice> A is the past, theres two branches B and C
20140409 16:23:08< iceiceice> when you merge you wil have D
20140409 16:23:12< AI0867> Necrosporus: well, many programs right now can handle *any* encoding, because they don't inspect the contents of the filename, but that breaks down when you want to interact with the user through something other than commandline arguments
20140409 16:23:16< iceiceice> if the change got committed separetly on B and C
20140409 16:23:27< iceiceice> hten it will merge correctly but still show as "green" in the github pull request dialog
20140409 16:23:39< iceiceice> because the change you introduce is a different commit from in B and in C
20140409 16:23:44< iceiceice> your options are,
20140409 16:23:46< iceiceice> 1.) deal with it
20140409 16:23:52< iceiceice> 2.) rebase your PR onto master before merging
20140409 16:23:59-!- exciton [chuck-the-@89.208.170.132] has quit [Read error: Connection reset by peer]
20140409 16:24:06< AI0867> as for whether *nix should be retrofitted to obey locales for this, or we should just ditch everything other than utf-8...
20140409 16:24:11< gfgtdf> 	iceiceice: 3.) click merge and see what happends.
20140409 16:24:14-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140409 16:24:15< AI0867> utf-8 sounds easier ;)
20140409 16:24:18< iceiceice> yeah i guess thats 1.) actually :)
20140409 16:24:55< gfgtdf> ah it somehow sunded liek giving up
20140409 16:24:58< gfgtdf> sounded
20140409 16:25:26< gfgtdf> or i could checrrypick on master
20140409 16:26:10< iceiceice> yeah you could do that too i guess
20140409 16:26:43< mattsc> iceiceice: I’ve been on the road for the last 3 days, so only kind of followed the logs.  I saw that you fixed the leaderless-AI-sides bug.  Great!
20140409 16:27:21< iceiceice> i mean only you can tell me if it is fixed enough :
20140409 16:27:22< iceiceice> :)
20140409 16:27:24< iceiceice> but i think i fixed it
20140409 16:27:31< mattsc> Is there anything you want me to do?  I need to test all the MAIs anyway before 1.11.13 is released, so I’ll let you know if there’s a problem with any of those scenarios.
20140409 16:27:53< iceiceice> yeah i mean i tested animals and it worked with the new flag
20140409 16:28:26< iceiceice> but i guess we should rigorously test other scenarios as well
20140409 16:28:52-!- Gallaecio [~quassel@84.120.115.132.dyn.user.ono.com] has joined #wesnoth-dev
20140409 16:29:00< iceiceice> my feeling is that we can do that gradually and inconcert with testing for other things, i doubt that this check_victory thing is going to break anyone else's deveopment branch uniquely
20140409 16:29:41< mattsc> Yep - as I said, I need to retest all MAIs anyway because of some other changes I made.
20140409 16:30:23< iceiceice> gfgtdf: i think i am going to cherrypick the server side controller tweaks into 1.12 soon
20140409 16:30:31< iceiceice> because i want to try to fix server side replays for 1.12
20140409 16:31:04< Necrosporus> AI0867, maybe then Wesnoth should throw warning, if locale is not supported?
20140409 16:31:11< iceiceice> like i said before i think it will make a small merge conflict in replays.cpp should you ever put the synced context stuff into 1.12
20140409 16:31:15< gfgtdf> iceiceice: you fixed teh already in 1.13-dev ?
20140409 16:31:30< gfgtdf> srever sided replay
20140409 16:31:31< iceiceice> its been in 1.13-dev since shortly after your thing
20140409 16:31:32< gfgtdf> s
20140409 16:31:37< iceiceice> no i didnt fix it in 1.13 dev
20140409 16:31:54< iceiceice> ok maybe i will just work there i gues
20140409 16:32:03< iceiceice> but i would rather not have a giant backlog of things that go into 1.12 suddenly
20140409 16:32:26< iceiceice> if we like together decide one day "we are ready!" and cherrypick like 100 commits into 1.12 i feel like someone might get mad :)
20140409 16:32:29< gfgtdf> what do you mean  ?
20140409 16:32:37< gfgtdf> ah
20140409 16:33:09< iceiceice> the server tweaks branch was originally off of 1.12 iirc actually
20140409 16:33:21< iceiceice> or maybe not that, but it was based before the synced stuff
20140409 16:33:31< iceiceice> when i merged to master i had to figure out this merge conflict htat was kind of annoying
20140409 16:33:48< iceiceice> if i cherry-pick back to 1.12 i'll have to figure out how to undo the change
20140409 16:33:59< iceiceice> and then if you cherry-pick to 1.12 i guess you'll have to redo it, or maybe not...
20140409 16:34:03< iceiceice> i guess maybe the merge will be smart?
20140409 16:34:28< iceiceice> idk if it becomes a problem you can ask me for help but since this content is ready to go into 1.12 i dont want to wait anymore
20140409 16:37:25-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has joined #wesnoth-dev
20140409 16:38:20< Necrosporus> AI0867, if you drop support for non-utf8 locales then your program won't support filenames on tengwar, as tengwar is not present in UTF-8 currently. I think that's bad
20140409 16:41:23< gfgtdf> iceiceicewe want to fix server sides replays in all cases before 1.12 relased i think ?
20140409 16:42:17< iceiceice> it should at least work as well as the other replays :)
20140409 16:42:20< iceiceice> and no regressions
20140409 16:42:22< iceiceice> i hope we can manage that
20140409 16:42:27< iceiceice> ideally we would fix in all cases i agree
20140409 16:43:53< gfgtdf> iceiceice: with in all cases i mean especialy mp realoding. and you fixed it currently?
20140409 16:44:06< iceiceice> no i havent done anything yet
20140409 16:44:12< iceiceice> i'm just about to start i think :)
20140409 16:52:38-!- Kexoth [~kex@89.205.75.19] has quit [Remote host closed the connection]
20140409 16:53:18-!- cib0 [~cib@p5DD20F9E.dip0.t-ipconnect.de] has quit [Ping timeout: 240 seconds]
20140409 16:53:18-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 16:55:52-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Quit: Leaving]
20140409 16:55:54< mattsc> shadowm: unfortunately, the consecutive line feed problem is the only OS X issue the upgrade fixed  (except for those that essentially say “upgrade things so that the ‘you need to upgrade this’ warnings go away”).
20140409 16:57:16< mattsc> I updated the OS X specific issues page, adding the status on 1-13.0-dev after the Xcode upgrade.
20140409 16:57:26-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 246 seconds]
20140409 16:58:32< aquileia> AI0867: Your branch still shows the unfortunate behaviour when the installation directory contains non-ASCII characters: most images aren't loaded correctly and thus the bug persists
20140409 17:00:21< aquileia> Perhaps it isn't the filesystem but a SDL1.2 related bug or something?
20140409 17:09:38-!- bagzie [~bag@85-76-83-250-nat.elisa-mobile.fi] has joined #wesnoth-dev
20140409 17:11:07-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 17:11:17< iceiceice> gfgtdf: do you know anything about this?
20140409 17:11:18< iceiceice> https://gna.org/bugs/?20322
20140409 17:14:36< gfgtdf> iceiceice: no but i think is the "calling rng from a unsynced context like selecct or durign ai calculation causes " which i fixed.
20140409 17:14:52< gfgtdf> s/causes/causes OOS
20140409 17:17:10-!- sachith500 [~kvirc@112.134.80.158] has quit [Read error: Connection reset by peer]
20140409 17:22:28< mattsc> iceiceice: just to confirm: the no_leader=yes side attribute is still valid, it just is not used to check victory conditions any more?
20140409 17:22:45< iceiceice> yes
20140409 17:22:59< iceiceice> i'm not really sure waht the status of no_leader=yes is now,
20140409 17:23:07< iceiceice> according to AIs comments on my PR,
20140409 17:23:14< iceiceice> it is modified if you unstore a leader with can_recruit = yes
20140409 17:23:28< iceiceice> and it may have some determining role in figuring out if a leader is carried over
20140409 17:24:01< iceiceice> AI0867 wrote that there are other ways for a side to get a leader from WML besides unstoring
20140409 17:24:07< iceiceice> so there might still be a bug there?
20140409 17:24:09< iceiceice> i'm not really sure
20140409 17:24:20< iceiceice> but anyways yes it has no role to play in check_victory 
20140409 17:24:50< mattsc> But I do leave it in the [side] tags for the time being, right?
20140409 17:25:05< iceiceice> if you dont have a leader then i would leave it in i guess
20140409 17:25:05< mattsc> Just add the new tag?
20140409 17:25:12< iceiceice> yeah that was what i tested myself
20140409 17:25:14< mattsc> okay, thanks
20140409 17:28:38< gfgtdf> iceiceice: i made in comment in https://gna.org/bugs/?20322.
20140409 17:37:23-!- molgrum [~molgrum@212.85.89.43] has joined #wesnoth-dev
20140409 17:40:53-!- Duthlet [~Duthlet@wesnoth/mp-mod/Duthlet] has quit [Quit: leaving]
20140409 17:41:28< gfgtdf> is there a way to see 'my last posts' on gna ?
20140409 17:42:54-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has quit [Ping timeout: 240 seconds]
20140409 17:47:06-!- iceiceice [~chris@207-237-132-90.ny.subnet.cable.rcn.com] has joined #wesnoth-dev
20140409 17:56:55-!- TC01__ [~quassel@128.220.109.252] has joined #wesnoth-dev
20140409 17:57:38-!- TC01 [~quassel@128.220.109.252] has quit [Ping timeout: 246 seconds]
20140409 18:03:37< shadowm> mattsc: What were the other text rendering-related issues it didn't fix?
20140409 18:04:05-!- Kevin_Xi [~kevin@223.72.182.186] has quit [Quit: Leaving.]
20140409 18:04:58< mattsc> shadowm: there are no other text rendering issues.  But I updated the SDL libraries also and was hoping that maybe that would fix some of those issues (even though it is only to a more recent version of SDL 1.2).
20140409 18:05:13< shadowm> Dugi: Do you know what 'dork' means? You really should watch your spelling here: http://forums.wesnoth.org/viewtopic.php?p=569140#p569140
20140409 18:05:34< shadowm> Because the name in question ends with 'rn', not 'rk'.
20140409 18:07:05-!- irker472 [~irker@fehu.ai0867.net] has quit [Quit: transmission timeout]
20140409 18:07:14-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140409 18:08:51-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has quit [Quit: i go nstuf kthxbai]
20140409 18:10:54-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 18:11:58-!- lipkab [~the_new_l@host-91-147-212-189.biatv.hu] has joined #wesnoth-dev
20140409 18:14:11< Dugi> shadowm: I remembered his name wrongly, I really thought he was named GbDork. Wondered why he picked such a weird name. Corrected the spelling in that post.
20140409 18:16:06-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has quit [Ping timeout: 240 seconds]
20140409 18:16:27-!- stikonas_ [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev
20140409 18:16:46< shadowm> Well, this is far from the first time I've seen this happen, so you might want to be more careful in the future.
20140409 18:18:38< Dugi> shadowm: What, did I misspell other users' names wrongly also in the past?
20140409 18:19:08-!- gfgtdf [~chatzilla@f054050123.adsl.alicedsl.de] has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]]
20140409 18:26:04< shadowm> Yes.
20140409 18:28:20< Dugi> I remember mistaking gfgtdf for somebody with a similar buttonmash as name, but can't remember when it was replaced by something slightly offensive... Well, I'll try to take care of it in the future.
20140409 18:34:12< lipkab> Dugi: You called me libkap once which is absolutely unforgiveable.
20140409 18:38:20< Dugi> I could not find any sign that it's actually an insult. Your actual name might be an insult, though: http://www.urbandictionary.com/define.php?term=Lipka
20140409 18:40:07< lipkab> Dugi: It's an insult because it suggests that I'm a dynamically linked library.
20140409 18:40:38-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 18:43:43< Dugi> lipkab: Not getting what has 'Lipkap' have in common with '.dll'.
20140409 18:45:19-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 250 seconds]
20140409 18:45:49-!- esr [~esr@static-71-162-243-5.phlapa.fios.verizon.net] has joined #wesnoth-dev
20140409 18:45:49-!- esr [~esr@static-71-162-243-5.phlapa.fios.verizon.net] has quit [Changing host]
20140409 18:45:49-!- esr [~esr@wesnoth/developer/esr] has joined #wesnoth-dev
20140409 18:46:18-!- _8680_ [~8680@2002:4404:712c:0:f86c:65cc:abfd:7376] has quit [Ping timeout: 240 seconds]
20140409 18:46:48< lipkab> Dugi: "libkap".
20140409 18:47:13-!- _8680_ [~8680@2002:4404:712c:0:b186:941d:9294:a3d] has joined #wesnoth-dev
20140409 18:48:08< Dugi> Ah, so I misspelled also the misspelled version. My aplogies, you see that my ability to read what I really see isn't the best.
20140409 18:51:43< lipkab> No problem ;) Actually, every second person or two misspells my nick the same way.
20140409 18:51:50-!- ancestral [~ancestral@75-161-229-57.mpls.qwest.net] has joined #wesnoth-dev
20140409 18:53:07< lipkab> I guess it's consequence of psychological priming and the high number of techies on IRC.
20140409 18:53:51< vultraz> I always have to remember not to spell it lipik + ab
20140409 18:54:56< lipkab> Because that means what?
20140409 18:56:59-!- fendrin [~fabi@wesnoth/developer/fendrin] has quit [Quit: Konversation terminated!]
20140409 18:58:17-!- fabi_ [~fabi@91-67-44-108-dynip.superkabel.de] has joined #wesnoth-dev
20140409 18:58:17-!- fabi_ [~fabi@91-67-44-108-dynip.superkabel.de] has quit [Changing host]
20140409 18:58:17-!- fabi_ [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20140409 19:00:18< Dugi> lipkab: Looking at urbandictionary, there is no key for lipik, but lipika is a kind of a very selfless girl (no obscene meaning).
20140409 19:00:54-!- exciton_ [chuck-the-@95.73.184.17] has joined #wesnoth-dev
20140409 19:02:36< lipkab> Maybe it has some meaning in a half-dead Micronesian language.
20140409 19:02:36-!- exciton_ [chuck-the-@95.73.184.17] has quit [Read error: Connection reset by peer]
20140409 19:04:29-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Read error: Connection reset by peer]
20140409 19:04:38-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 255 seconds]
20140409 19:05:52-!- exciton [chuck-the-@89.208.170.132] has joined #wesnoth-dev
20140409 19:07:30-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has quit [Ping timeout: 240 seconds]
20140409 19:08:22-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20140409 19:12:19-!- iwaim [~iwaim@2001:2c0:40e:2002:0:4:14:80] has joined #wesnoth-dev
20140409 19:18:42-!- lipkab [~the_new_l@host-91-147-212-189.biatv.hu] has quit [Read error: Connection reset by peer]
20140409 19:18:52< Necrosporus> AI0867, suddenly http://stackoverflow.com/questions/1629437/is-codepage-65001-and-utf-8-the-same-thing
20140409 19:19:14< Necrosporus> Maybe UTF-8 names in Wesnoth could have worked all along without modifications?
20140409 19:24:12< Necrosporus> Though I'm not sure if that's usable at all
20140409 19:27:12-!- wesbot changed the topic of #wesnoth-dev to: string+feature freeze active on 1.12 | 227 bugs, 351 feature requests, 28 patches | Logs: http://irclogs.wesnoth.org | Alternate logs: http://wesnoth.debian.net | Don't paste on IRC! Use a pastebin: http://pastebin.com | http://imagebin.org
20140409 19:45:15< aquileia> Necrosporus: If you didn't notice - even if you use full utf8 in savefiles or content it'll work perfectly. It's only that Wesnoth isn't displayed correctly if the path to Wesnoth is incorrect
20140409 19:45:37< aquileia> incorrect meaning non-ASCII
20140409 19:46:49< iceiceice> huzzah i am registered
20140409 19:46:51< aquileia> as AI0867's branch couldn't fix this it might well be a limitation of SDL or something similar
20140409 19:47:10< iceiceice> Dugi if you are seeking registration one fo the staff stopped lurking in #freenode, his name is mist
20140409 19:47:13< aquileia> iceiceice: congratulations! And iceiceice3?
20140409 20:00:10< Necrosporus> aquileia, yes, and it is a problem, because names like "My Documents" are localized and Wesnoth asks if user wants to keep saves in "My Documents"
20140409 20:01:19< iceiceice> i didnt register the other ones on irc
20140409 20:01:53< aquileia> What I wanted to say is that the nature of the bug does not necessarily point to wesnoth itself, seeing thatt it couldn't be fixed via the filesystem.cpp
20140409 20:02:31< Dugi> iceiceice: Thanks, I sent him a message.
20140409 20:05:17< Dugi> iceiceice: Now, I am registered.
20140409 20:05:24< iceiceice> huzzah
20140409 20:06:31< iceiceice> aquileia idk i couldnt really follow this argument... is this something that affects all cross platform software or just wesnoth
20140409 20:07:18< aquileia> iceiceice: I responded to Necrosporus
20140409 20:07:36< iceiceice> i know but i am asking about the broader issue necrosporus and ai discussed earlier
20140409 20:07:59< iceiceice> theres something about some library we use does something with localized filenames that is not ideal?
20140409 20:08:10< iceiceice> or it is a compatibility issue with windows vs linux localization?
20140409 20:08:27< aquileia> The isue is that most images in Wesnoth aren't loaded if (their? the exe's?) path contains non-ASCII characters
20140409 20:08:34-!- gfgtdf [~chatzilla@f054050123.adsl.alicedsl.de] has joined #wesnoth-dev
20140409 20:08:58< iceiceice> ok so why is this a "won't fix" ?
20140409 20:09:00< iceiceice> or did i misunderstand
20140409 20:09:02< aquileia> e.g. the main screen will just show the button labels and other texts, but no images
20140409 20:09:45< iceiceice> is it just that we arent sure what exactly is misbehaving?
20140409 20:09:46< aquileia> you have to install Wesnoth to a directory without these characters and everything's fine
20140409 20:09:54-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 240 seconds]
20140409 20:10:19< aquileia> It might even be a SDL issue - we'll see if SDL 2 fixes it
20140409 20:10:44< iceiceice> hmm
20140409 20:10:51< iceiceice> how does linux normally handle things like this?
20140409 20:11:04< iceiceice> if i install with chinese language settings will thingsl ike "/home" appear in chinese?
20140409 20:11:07-!- timotei_ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev
20140409 20:11:24< aquileia> or alternatively it's just that SDL is passed the wrong locale or something (I'm no dev, I don't know too much abot these things)
20140409 20:11:35< shadowm> In Linux /home is always /home, never translated.
20140409 20:12:33< shadowm> UTF-8 is the standard for i18n nowadays, so your system locale will usually use the UTF-8 encoding of Unicode (even if it's English).
20140409 20:12:36< aquileia> or something completely different...
20140409 20:12:51< shadowm> But Windows wants UTF-16.
20140409 20:13:45< gfgtdf> i think since windows vista direcotrynames are teh same on all languages
20140409 20:13:50< shadowm> The Win32 API supports both UTF-16 (all U-suffixed functions) and a legacy encoding (all A-suffixed functions). I believe the legacy encoding is system-dependent.
20140409 20:14:13< shadowm> e.g. CP 1252 for English and Spanish, who-knows-what for Russian, etc.
20140409 20:14:14< Necrosporus> aquileia, their, even if exe is in program files, it couldn't load addon's unless it's in ascii-only path, which is not necessarily true if user selected to use "My Documents" as userdata
20140409 20:14:16< iceiceice> so windows just broke backwards compatibility basically/
20140409 20:14:26< shadowm> No.
20140409 20:14:38< shadowm> This convention goes as far back as NT 3.1.
20140409 20:15:22< iceiceice> what does SDL have to do with directory encodings? thats a bit strange
20140409 20:15:41< shadowm> Before Windows Vista, if you had e.g. a Spanish version of Windows, some (not all) system directory names had translated names on the filesystem itself.
20140409 20:16:00< shadowm> e.g. instead of C:\Program Files you had C:\Archivos de Programa.
20140409 20:16:03< aquileia> I just said it's a possibility from the symptomes I saw, not that it's necessarily connected
20140409 20:16:10< aquileia> iceiceice: ^
20140409 20:16:19< iceiceice> and now they all do, or now they stopped that?
20140409 20:16:26< Necrosporus> shadowm, cp1251 for russian, and cp866 in windows console
20140409 20:16:31< shadowm> Since Vista, Windows uses the English names on the filesystem I believe, and handles the native language names using symlinks (?).
20140409 20:16:54< shadowm> And/or shell aliases (?).
20140409 20:16:58< iceiceice> hmm
20140409 20:17:06 * shadowm boots Vista to check.
20140409 20:17:49< Necrosporus> I can remember Program Files were Program Files in Russian even on XP
20140409 20:18:12< shadowm> But this doesn't mean that if you concatenate "C:\ÁÁÁÁ" (CP-1252) with "\€€€" (UTF-8) you can expect the end result to be valid UTF-8.
20140409 20:18:16< shadowm> Because it's not.
20140409 20:18:57< Necrosporus> First one should be iconved
20140409 20:19:06< shadowm> Windows may throw a CP-1252 or whatever path at Wesnoth via the *A API and Wesnoth isn't prepared to deal with it.
20140409 20:20:11< shadowm> Personally, I made Wesnoth's DE functions use the *U API to avoid future issues from its users, but neither my users nor the DE function implementations can ascertain the validity of their input such as Wesnoth's user data directory path.
20140409 20:20:32< gfgtdf> you eman teh *W api ?
20140409 20:20:42< shadowm> Er.
20140409 20:20:51< shadowm> Yes, wherever I said 'U' you can assume that to be 'W'.
20140409 20:21:09< shadowm> The 'W' is for 'wide' (as in, not 8 bits), I had forgotten that.
20140409 20:21:47< Necrosporus> There's also a problem with even std::fstream, in windows it accepts wchar_t* instead of standard char*
20140409 20:22:12< shadowm> Well, apparently Windows Vista doesn't provide a symlink for "Archivos de Programa" -> "Program Files".
20140409 20:23:06< shadowm> What it does provide is a shell alias so "Program Files" appears as "Archivos de Programa" in file listings.
20140409 20:23:57< shadowm> There are APIs to get paths to system directories like this, so it makes sense they wouldn't provide backwards compatibility for broken software.
20140409 20:24:27< shadowm> But you might still get trouble if your user name isn't limited to ASCII.
20140409 20:24:32< Necrosporus> By the way http://www.utf8everywhere.org/ this site suggests to define _UNICODE but never depend on function which require this define
20140409 20:28:53-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 20:32:42-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Ping timeout: 240 seconds]
20140409 20:33:06-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 20:33:30-!- _8680_ [~8680@2002:4404:712c:0:b186:941d:9294:a3d] has quit [Remote host closed the connection]
20140409 20:36:06< iceiceice> necrosporus: thats an interesting link, i never knew about all this
20140409 20:36:12-!- _8680_ [~8680@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20140409 20:36:29-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140409 20:36:54< Necrosporus> There's also interesting link about GNU/Linux http://www.dwheeler.com/essays/fixing-unix-linux-filenames.html
20140409 20:37:34< Necrosporus> Unfortunately, GNU/Linux has problems with locales too, though they are not as severe
20140409 20:37:42-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has joined #wesnoth-dev
20140409 20:37:48-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20140409 20:40:09-!- trademark_ [~trademark@nsg93-8-88-175-59-164.fbx.proxad.net] has quit [Ping timeout: 250 seconds]
20140409 20:45:47< iceiceice> im pretty sure i learned in highschool "unicode is 2 bytes instead of 1 to support internationalization" and that was it
20140409 20:52:54-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has quit [Ping timeout: 240 seconds]
20140409 20:53:18-!- Kexoth [~kex@89.205.75.19] has joined #wesnoth-dev
20140409 20:57:30-!- Kexoth [~kex@89.205.75.19] has quit [Ping timeout: 240 seconds]
20140409 21:03:44-!- thunderstruck [~zaibotren@cpc13-sgyl31-2-0-cust696.18-2.cable.virginm.net] has quit [Quit: leaving]
20140409 21:22:33-!- Samual [diotecktec@xonotic/core-team/Samual] has joined #wesnoth-dev
20140409 21:27:07< iceiceice> gfgtdf: ok... it is way harder to cherry-pick this to 1.12 than it was to rebase it onto master...
20140409 21:27:12< iceiceice> theres way more merge conflicts now
20140409 21:27:24< iceiceice> i'm not sure if i can resolve them without creating bugs :/
20140409 21:27:48< gfgtdf> iceiceice: are they ralted to my pr ?
20140409 21:28:06< iceiceice> yes its all the changes to replay.cpp
20140409 21:28:45< iceiceice> http://pastebin.com/pNUP6pdB
20140409 21:29:43< iceiceice> i wish that i had merged mine to 1.12 before rebasing it onto your thing ....
20140409 21:30:23< gfgtdf> did you change somethign at get_user_choice ?
20140409 21:30:42< iceiceice> no it has nothing to do with that
20140409 21:30:47< iceiceice> it was written before i even saw PR 121
20140409 21:30:58< iceiceice> i'm just trying to support "record_change_controller"
20140409 21:31:04< gfgtdf> hm then git seems to be rather stupid in this case
20140409 21:32:21< iceiceice> https://github.com/wesnoth/wesnoth/commit/8e5572a239eb0b24e37135cc7825d7462599295c
20140409 21:32:28< iceiceice> i'm just trying to cherry-pick that into 1.12
20140409 21:33:13< iceiceice> i guess that i need to add all the code "recruit" "disband" etc.?
20140409 21:33:18< iceiceice> since you removed that in your PR
20140409 21:33:28< iceiceice> and i'm basically undoing that change to resolve this merge...
20140409 21:34:15< gfgtdf> do you think you could just merge prp 121 onto 1.12 ?
20140409 21:34:24< gfgtdf> im quiet happy with it's current state.
20140409 21:34:44< iceiceice> ok its up to you but i'm not going to merge your commits into 1.12, you gotta do that
20140409 21:34:53< gfgtdf> ok hwo do i do that ?
20140409 21:34:55< iceiceice> but it would solve my problem ig uess 
20140409 21:34:55< gfgtdf> how*
20140409 21:35:27< gfgtdf> i guess i have to mage local copy of 1.12 ?
20140409 21:35:34< iceiceice> i think you use cherry-pick with the .. range syntax
20140409 21:36:23< iceiceice> hmm i guess you have a nontrivial merge commit also?
20140409 21:36:30< iceiceice> i'm not sure if you are supposed to cherry-pick that...
20140409 21:36:40< iceiceice> tbh i really need to read how merge works in git again
20140409 21:37:08< gfgtdf> iceiceice: what do you mean by non trivial ?
20140409 21:37:25< iceiceice> did you have to resolve merge conflicts when you merged your branch?
20140409 21:37:28< iceiceice> so it wasnt just a fastforward?
20140409 21:37:33< gfgtdf> no i clicked teh green button
20140409 21:37:37< gfgtdf> the*
20140409 21:38:12< iceiceice> hmm so you have this commit at the end: https://github.com/wesnoth/wesnoth/commit/8617a02a88a38d8fe01ee0a3672d75fc5fe72924
20140409 21:38:34< iceiceice> idk i always religiously rebase things before merging them into master so i usualyl dont get a commit like that at the end
20140409 21:38:50< iceiceice> i guess most everybody else likes to use the merges .. 
20140409 21:38:51< gfgtdf> i did rebase, and there were no merge conflicts.
20140409 21:39:01< iceiceice> so the merge doesnt do anything but its still there?
20140409 21:39:18< gfgtdf> i tihnk yes
20140409 21:39:30< iceiceice> idk i think you arent supposed to get a commit like that at the end if it is fully fastfoward merge
20140409 21:39:50< iceiceice> i could be wrong though
20140409 21:39:55< iceiceice> maybe the web interface always makes one
20140409 21:40:24< iceiceice> it might be that there minor conflicts and it automatically resolved?
20140409 21:40:31< gfgtdf> iceiceice: how do i make a local copy of original wesnoth/wesnoth when i only have a copy of gfg/wesnoth ?
20140409 21:40:33< iceiceice> anyways i dont know how to cherry-pick it, if you are supposed to include that guy or not
20140409 21:41:07< iceiceice> idk, do you just pull upstream master and pull upstream 1.12?
20140409 21:42:00< iceiceice> you want your master and 1.12 to match upstream all the time, right? is that not what your workflow is?
20140409 21:43:43< gfgtdf> idk until now i never used upstream an and always  used 'pull https://github.com/wesnoth/wesnoth-old master' and 'push -f https://github.com/gfgtdf/wesnoth-old sync_fix'
20140409 21:44:23< gfgtdf> ah i know
20140409 21:44:30< gfgtdf> i can still so that i think
20140409 21:47:28< iceiceice> i guess that you can cherry-pick the merge if you use the -m option and specify the parent
20140409 21:48:04< iceiceice> maybe thats all you have to do and it will figure out the other commits from there
20140409 21:48:13< gfgtdf> iceiceice: nut i dont wan the whole pr especiyl i dont want the 'enaböe teh deterministic mode' and i also expect merge conflicts, do you know how do i select teh branch form which to cherrypick ?
20140409 21:48:18< iceiceice> oh yeah right
20140409 21:48:43< iceiceice> so i would not expect any merge conflicts, right? i think 1.12 hasnt much changed since the 121 was committed
20140409 21:49:14< gfgtdf> i expect at least one merge conlfict for teh utf change in race.cpp
20140409 21:49:32< iceiceice> i see
20140409 21:49:59< iceiceice> i think maybe best thing is to just make a "1.12-temp" branch
20140409 21:50:07< iceiceice> and try cherry-picking the commits of 121 in order
20140409 21:50:10< iceiceice> and leave out the ones you dont want
20140409 21:50:20< iceiceice> then at the end try to merge to 1.12 and push it out
20140409 21:51:10< gfgtdf> but how can i select the branch form which i want to cherrypick ? i mean i if i use cherrrypick range it might be that there are a different number of commits between the 2 commit on different brnaches?
20140409 21:51:34< iceiceice> oh i guess dont use cherry-pick range
20140409 21:51:40< iceiceice> probly easiest to just cherry pick individual commits
20140409 21:51:44< iceiceice> if you only want some of them
20140409 21:51:54< iceiceice> and then you can compile to check it didnt break in between each one ig uess
20140409 21:52:21< iceiceice> idk maybe theres some crazy invocation of rebase --interactive --onto but i would steer clear of it i think
20140409 21:52:48< iceiceice> you could rebase the 1.12-temp at the end i guess
20140409 21:52:51< shadowm> What are you trying to do in general terms?
20140409 21:53:13-!- shadowm_desktop [ignacio@wesnoth/developer/shadowmaster] has joined #wesnoth-dev
20140409 21:53:24< iceiceice> shadowm: there was a PR 121 that was developed around the same time that i developed a "server controller tweaks" branch
20140409 21:53:36< shadowm> Parallel branches, okay, and?
20140409 21:53:38< iceiceice> they both were tested for like a week
20140409 21:53:49< iceiceice> then 121 got merged to master but not 1.12 because still testing i guess
20140409 21:54:02< iceiceice> i rebased my branch so i could merge to master after
20140409 21:54:13< iceiceice> but i had to resolve an annoying merge conflict i remember that was difficult to understand
20140409 21:54:15< iceiceice> in replay.cpp
20140409 21:54:19< iceiceice> it is working fine on master
20140409 21:54:25< shadowm> And now your branch can't be trivially reconciled against 1.12.
20140409 21:54:37< iceiceice> so at the time i thought 121 would be merged to 1.12 soon
20140409 21:54:43< iceiceice> so i was like i'll just wait
20140409 21:54:51< iceiceice> but now this is like more than a week ago
20140409 21:55:02< iceiceice> and i want to start working on fixing replay server which is also destined for master and 1.12
20140409 21:55:07< iceiceice> and i'm just worried we will build up a backlog
20140409 21:55:16-!- bagzie [~bag@85-76-83-250-nat.elisa-mobile.fi] has quit []
20140409 21:55:31< iceiceice> so i'm encouraging gfgtdf to figure out what parts of 121 he wants to go on 1.12 so i can merge and we dont have a backlog of things to cherrypick
20140409 21:55:42< iceiceice> i can cherry-pick my thing now but it will make more work for us which might be avoided
20140409 21:55:46-!- demiurgos [51258a4a@gateway/web/freenode/ip.81.37.138.74] has quit [Quit: Page closed]
20140409 21:55:51< iceiceice> becuase then we'll get double stupid merge conflicts
20140409 21:56:36< shadowm> Do you changes really have to be based on top of 121?
20140409 21:56:43< iceiceice> no they dont
20140409 21:56:49< iceiceice> they really should be independent
20140409 21:57:02< iceiceice> but for some reason git is being stupid here
20140409 21:57:13< shadowm> Git is not a programmer.
20140409 21:57:38-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 240 seconds]
20140409 21:57:48< gfgtdf> shadowm: is there some rule like every 1.12 commit should also be on 1.13 ?
20140409 21:58:06< shadowm> That statement might not make sense at first glance, so here's a more constructive example: branch A adds an instance of the C++ type T named foo, and branch B renames the type T to U.
20140409 21:58:24< shadowm> Then branch B goes upstream, and the author of A tries to rebase against upstream (B).
20140409 21:58:35< iceiceice> i get what you are saying but this is more innocuous than that
20140409 21:59:00< iceiceice> its like adding and removing some case options in do_replay
20140409 21:59:07< iceiceice> and git just throws its hands up in despair
20140409 21:59:18< iceiceice> and points out large regions that seem to get intermingled with + / -
20140409 21:59:41< shadowm> gfgtdf: Common sense comes first. If bug #12345 only exists in 1.12 because master completely refactored the involved code, accidentally eliminating the cause of the bug, why should the commit fixing bug #12345 go to both branches?
20140409 22:00:19-!- exciton_ [chuck-the-@95.73.184.17] has joined #wesnoth-dev
20140409 22:00:24< shadowm> iceiceice: Whitespace changes involved too?
20140409 22:00:41< gfgtdf> shadowm: yes i forgot a newline at teh end of file.
20140409 22:01:20< gfgtdf> shadowm: how do you mean i think if no reason agains bugfixed are good and should be on both branches.
20140409 22:01:41< shadowm> (Whitespace: U+0020 SPACE, U+0009 (horizontal tab), U+000A (line feed), U+000B (vertical tab), etc.)
20140409 22:02:17< shadowm> gfgtdf: Read again. The bug does not exist in master.
20140409 22:02:40< iceiceice> ok i think your common sense example might not have anything to dow ith the case at hand then
20140409 22:02:50< gfgtdf> ah i thought your uestion was 'why shoudl teh bug on master that accidebntly fixes 12345 go  to 1.12'
20140409 22:02:56< shadowm> No, it was only intended to answer gfgtdf's specific question.
20140409 22:03:06< iceiceice> oh i see
20140409 22:03:23< shadowm> Now I'm going to reiterate what I said the other day about branches.
20140409 22:04:06< shadowm> It'd be foolish to expect from a project like Wesnoth for any given stable branche to be mergeable onto master. We are talking of a project where code keeps being refactored or otherwise rearranged in non-trivial ways all the time.
20140409 22:04:38-!- exciton [chuck-the-@89.208.170.132] has quit [Ping timeout: 255 seconds]
20140409 22:04:40< shadowm> The lack of strict coding conventions to dictate standards for indentation and such don't really help Git's merging algorithms either.
20140409 22:05:17< shadowm> So someone might take a line in one branch and keep it intact in another branch save for tab vs. spaces and boom, you get a merge conflict.
20140409 22:05:43< shadowm> At some point after a stable branch has been created, you'll have to embrace merge conflicts.
20140409 22:06:34< shadowm> That said, I think `git merge` accepts some options to pick a specific merge algorithm and change how it behaves, but I've never used them..
20140409 22:07:52< shadowm> In general, I've quite presumably used less than 1% of Git's functionality.
20140409 22:08:11< iceiceice> ok, well,
20140409 22:08:13< shadowm> Reading the man pages sometimes feels like reading a dictionary.
20140409 22:08:18-!- tomreyn [~tomreyn@megaglest/team/tomreyn] has quit [Quit: tomreyn]
20140409 22:08:34< shadowm> There are thousands and thousands of words you never use unless you want to impress your language teacher.
20140409 22:09:07< iceiceice> ok well the issue is not about fancy git options i tink
20140409 22:09:50< iceiceice> the issue is we have some content that changed alot of lines of code, and some of it is intended to go on 1.12 as discussed on irc, but hasnt got merged yet
20140409 22:09:51< shadowm> The general issue is about A|B|C branch divergence, so "embrace merge conflicts" still applies.
20140409 22:10:11< iceiceice> and so all the subsequent commits to master are quite difficult to cherry-pick to 1.12
20140409 22:10:19< iceiceice> and every time we do we risk creating strange errors
20140409 22:10:35< shadowm> So it is intended to go to 1.12, okay, so why did it go to master without the 1.12 queue also being prepared at the same time?
20140409 22:10:49< iceiceice> you'll have to ask gfgtdf
20140409 22:11:27< gfgtdf> shadowm: becasue i want sure whether i could fix all the possible bugs accociated with it in time.
20140409 22:11:35< shadowm> I'm almost afraid to check what #121 entails.
20140409 22:12:02< gfgtdf> why ?
20140409 22:12:10< shadowm> "this patch also introduces the optional deterministic mode for SP"
20140409 22:12:14< shadowm> Not 1.12 material.
20140409 22:12:29< gfgtdf> i want to skip that for 1.12
20140409 22:12:57< iceiceice> the synced context stuff as i understand fixes a bunch of oos problems
20140409 22:13:07< iceiceice> and was tested in great detail with eli dupree as you will see in logs
20140409 22:13:17< gfgtdf> shadowm: can i add 19cf6f5cd3f8dcd0913c6a3a755d2d8249218e88 in 1.12 ?
20140409 22:13:45< shadowm> Okay, at first glance I can't tell what's a pure bug fix (for preexisting bugs) and what is not in the commit list.
20140409 22:14:04< shadowm> But since you have push access, well, it's not like I need to review the PR.
20140409 22:14:23< shadowm> gfgtdf: No.
20140409 22:15:19< shadowm> Unless you are 100% positive that the ground will be torn asunder  without it.
20140409 22:15:46< shadowm> Otherwise we get into the slippery slope of granting freeze exceptions to people.
20140409 22:15:56< shadowm> "Hey, my feature is important too! :( :( :("
20140409 22:15:57< iceiceice> i'm not advocating to break feature freeze here... its up to gfgtdf to do it right
20140409 22:16:07< iceiceice> i basically support the infrastructure improvements that fix bugs though
20140409 22:16:31< iceiceice> and i would just like it to resolve sooner if possible
20140409 22:16:34< shadowm> I don't know why you'd need to change the Lua API to fix what's presumably an engine bug.
20140409 22:16:53< iceiceice> i guess its just a debugging command, idk you'll have to ask gfgtdf
20140409 22:16:55< shadowm> We promised our users to not change things behind their backs.
20140409 22:17:06< shadowm> In fact, I did so, with my username.
20140409 22:17:32< shadowm> I don't like breaking promises. :p
20140409 22:18:01< gfgtdf> shadowm: in this case it was more like i accidently put the deletion of the "rand_rng::simple_rng &" parameter in the commit during rebasing an am too lazy to split.
20140409 22:18:59< gfgtdf> shadowm: can i put that commit onto 1.12 and then remove the fature in a later commit ?
20140409 22:19:08< iceiceice> ok well the only thing i am complaining about is that i dont really have an idea of a time frame of when this will go on 1.12 and if
20140409 22:19:12< iceiceice> so i keep waiting to merge my content
20140409 22:19:14< shadowm> If you still have the original branch, use rebase -i onto 1.12 and exclude or edit the involved commits accordingly?
20140409 22:19:16< iceiceice> and then i will make more and more
20140409 22:19:21< iceiceice> and its a backlog
20140409 22:19:28< shadowm> Editing commit diffs and messages mid-rebase is a thing that can be done, in case that wasn't obvious.
20140409 22:19:34< iceiceice> and it will be bad if 2 weeks from now we merge like 100 commits to 1.12 that we wrote 3 weesk ago
20140409 22:20:10< gfgtdf> shadowm: i thought it would be good if the commits have the same ids on both branches.
20140409 22:20:13< iceiceice> so i just think we should figure out what of the code we have written should go on 1.12 and do this in a way that makes as few bugs as possible
20140409 22:20:31< shadowm> And if it's too conflictive, you can alternatively rebase the original branch onto its original starting point for the same purpose, exclude/edit, finish, rebase again onto 1.12.
20140409 22:20:35< iceiceice> the longer the wait the greater probability of making a bug in merge commits
20140409 22:20:49< iceiceice> i'm going out for now, be back later
20140409 22:20:50-!- iceiceice [~chris@207-237-132-90.ny.subnet.cable.rcn.com] has quit [Quit: Leaving]
20140409 22:20:52< shadowm> gfgtdf: ... Yeah no, the odds of that happening get increasingly low as 1.12 and master diverge further.
20140409 22:21:12< shadowm> A commit's hash is calculated from the commit message, author information, date, and ancestry, amongst other things.
20140409 22:21:35< shadowm> The easiest way to alter the hash is to change its ancestry.
20140409 22:22:34< shadowm> So hoping commit hashes to remain the same between two divergent branches... I might as well run for president and hope to be elected. :)
20140409 22:23:36< gfgtdf> hm ok
20140409 22:23:38< shadowm> It can be done when the ancestry is compatible with both branches, but again, increasingly unlikely with divergent history.
20140409 22:24:55< shadowm> And the moment you start editing or excluding commits in a chain like A->B->C->D  -->  A->B'->C'->E->D', you can safely expect B', C', and D' to have completely different hashes from their original versions.
20140409 22:29:44< shadowm> Incidentally, for programmer-types, I find that looking at Git branches as though they were linked lists helps make complicated concepts like commit ancestry and "why push -f on shared repositories is evil" more evident.
20140409 22:30:29< shadowm> But that's just an unimportant aside note.
20140409 22:36:16-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: Computer's napping]
20140409 22:36:22-!- exciton_ [chuck-the-@95.73.184.17] has quit [Read error: Connection reset by peer]
20140409 22:36:36-!- exciton [chuck-the-@95.73.184.17] has joined #wesnoth-dev
20140409 23:00:59-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-dev
20140409 23:02:25-!- mjs-de [~mjs-de@f048013210.adsl.alicedsl.de] has quit [Remote host closed the connection]
20140409 23:09:23-!- iceiceice [~chris@cpe-66-108-20-80.nyc.res.rr.com] has joined #wesnoth-dev
20140409 23:09:53< iceiceice> shadowm: why is it even possible to check usernames in wml?
20140409 23:09:58-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140409 23:10:33-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 23:12:18-!- Gambit [~derek@wesnoth/developer/grickit] has quit [Ping timeout: 240 seconds]
20140409 23:12:45-!- noy [~Noy@wesnoth/developer/noy] has quit [Client Quit]
20140409 23:12:57< gfgtdf> 	iceiceice: why should it not be ?
20140409 23:13:21< gfgtdf> iceiceice: i think you mean just by checking out teh leaders names
20140409 23:13:30< iceiceice> i guess sometimes unfortunately users make banlists in their add-on
20140409 23:13:43< iceiceice> theres not really any legitimate need to get the users name as a string
20140409 23:13:58< iceiceice> you're right, it would be hard to block i guess
20140409 23:14:07< gfgtdf> iceiceice: is there a second way ?
20140409 23:14:10< iceiceice> i was thinking just dont make it a wml accessible field of side
20140409 23:14:24< iceiceice> maybe you could make it so that unit names are not readable, only writeable?
20140409 23:14:26< iceiceice> also
20140409 23:14:54< iceiceice> i doubt if there are very many legitimate reasons to ever read a unit name into a wml string,
20140409 23:14:59< iceiceice> i think they are typically oos anyways
20140409 23:15:52< gfgtdf> iceiceice: no unit names areew synced, well it doesnt casue OOs unless you use it for other hing,s but for example somepne might want to implement their own unit recall list or something
20140409 23:16:01-!- fabi__ [~fabi@91-67-44-108-dynip.superkabel.de] has joined #wesnoth-dev
20140409 23:16:01-!- fabi__ [~fabi@91-67-44-108-dynip.superkabel.de] has quit [Changing host]
20140409 23:16:01-!- fabi__ [~fabi@wesnoth/developer/fendrin] has joined #wesnoth-dev
20140409 23:16:06< iceiceice> gfgtdf: are you sure?
20140409 23:16:20< iceiceice> i know that the code that handles side drops will make edits to the leader names
20140409 23:16:22< gfgtdf> they are synced but on different local languages the might stil have different names
20140409 23:16:25< iceiceice> and i dont know if that is synced
20140409 23:16:33< iceiceice> maybe you changed this?
20140409 23:16:57< gfgtdf> iceiceice: hm no i jst thought about the recruitment of unit or making unit with wml
20140409 23:17:10< iceiceice> y i think generally speaking the leader's name is oos usually
20140409 23:17:18< iceiceice> maybe not at turn start....
20140409 23:17:19< iceiceice> but still
20140409 23:17:19< shadowm> iceiceice: Because we assign player names to leaders.
20140409 23:18:34< shadowm> Why? Well, just like with every other MP thing, I have no idea.
20140409 23:19:11< shadowm> But if I had to guess, I'd say it's a convenience for scripted dialogues and other such things.
20140409 23:19:16-!- Gambit [~derek@wesnoth/developer/grickit] has joined #wesnoth-dev
20140409 23:19:18-!- fabi_ [~fabi@wesnoth/developer/fendrin] has quit [Ping timeout: 240 seconds]
20140409 23:19:19-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 23:19:25< iceiceice> i see
20140409 23:19:43< shadowm> As well as, uh, allowing players to find other players on the map easily without adding a third identification.
20140409 23:21:16< iceiceice> yeah i guess it would be too much work disable access to player names
20140409 23:26:28-!- exciton [chuck-the-@95.73.184.17] has quit [Read error: Connection reset by peer]
20140409 23:26:42-!- exciton [chuck-the-@95.73.184.17] has joined #wesnoth-dev
20140409 23:28:13-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Read error: Connection reset by peer]
20140409 23:30:01-!- TC01__ is now known as TC01
20140409 23:30:42-!- _8680_ [~8680@2002:4404:712c:0:76de:2bff:fed4:2766] has quit [Ping timeout: 240 seconds]
20140409 23:31:13-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has quit [Read error: Operation timed out]
20140409 23:32:10-!- noy [~Noy@wesnoth/developer/noy] has quit [Quit: noy]
20140409 23:34:00-!- noy [~Noy@wesnoth/developer/noy] has joined #wesnoth-dev
20140409 23:38:18-!- noy [~Noy@wesnoth/developer/noy] has quit [Ping timeout: 240 seconds]
20140409 23:39:48-!- Dugi [93fbd156@gateway/web/freenode/ip.147.251.209.86] has quit [Ping timeout: 240 seconds]
20140409 23:40:46-!- aquileia [6dc00d61@gateway/web/freenode/ip.109.192.13.97] has quit [Quit: Page closed]
20140409 23:45:49-!- yann [~dwitch@nan92-1-81-57-214-146.fbx.proxad.net] has joined #wesnoth-dev
20140409 23:47:19-!- _8680_ [~8680@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
20140409 23:49:27-!- exciton [chuck-the-@95.73.184.17] has quit [Read error: Connection reset by peer]
20140409 23:49:41-!- exciton [chuck-the-@95.73.184.17] has joined #wesnoth-dev
20140409 23:52:15-!- c74d [~c74d3a4eb@2002:4404:712c:0:76de:2bff:fed4:2766] has joined #wesnoth-dev
--- Log closed Thu Apr 10 00:00:34 2014