IRCloggy #git 2013-02-07

Provider of IRC logs since 2005.

WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

	2013-02-07
	← mmitchell left	00:00
	→ mithro- joined	00:04
	← Chillance left	00:04
	→ mmitchell joined	00:05
	← russt left	00:05
	← leeN left	00:05
	← rakl left	00:06
	← x82_nicole left	00:06
	→ aartist joined	00:08
	→ soypirate joined	00:08
	← Orbitrix left	00:09
	→ pppaul joined	00:09
	← mmitchell left	00:09
	← LeMike left	00:10
	← kevwil left	00:10
	← macabre left	00:11
	← Ksg89 left	00:14
	→ rakl joined	00:14
	← rakl left	00:14
	← gyre007 left	00:15
	→ gamja9e joined	00:15
	← snql left	00:16
	→ rakl joined	00:17
	← eletuchy left	00:17
	← gamja9e left	00:18
	← natechan left	00:18
	← normanrichards left	00:19
	→ main joined	00:20
	← sluther left	00:20
	→ dvide joined	00:21
	← lavaman left	00:24
	→ HumanSky joined	00:25
	← kogent left	00:26
	← jcromartie left	00:27
	→ Liir joined	00:28
	← Guest66543 left	00:30
	← AnimeCYC left	00:32
	← hyperair left	00:32
	→ Fandekasp joined	00:32
	← yshh left	00:34
	→ Fels joined	00:34
	← bmac_ left	00:35
	← boombatower left	00:36
	→ gyre007 joined	00:36
	→ natechan joined	00:36
	→ timj joined	00:37
	← Raydiation left	00:39
	← FernandoBasso left	00:39
	← karupanerura left	00:40
	→ mattt_ joined	00:40
	→ eletuchy joined	00:40
mattt_	How can I have git diff pipe to "vim -" by default, rather than less?	00:40
	→ pity joined	00:41
	→ JoeHazzers joined	00:41
	→ karupanerura joined	00:42
JoeHazzers	none of you would happen to know of any fairly "flashy" frontend for git, would you?	00:42
	← SBoudrias left	00:43
	→ SBoudrias joined	00:44
	← svm_invictvs left	00:44
	→ trungqtran joined	00:45
	← AaronMT left	00:46
	→ nickpresta joined	00:47
	← natechan left	00:47
	→ alunduil joined	00:48
	→ jjang_ joined	00:49
	→ AaronMT joined	00:49
	← jjang_ left	00:50
	← jjang left	00:50
	→ rchavik joined	00:50
	→ jjang joined	00:50
	← infynyxx left	00:50
	→ pgokeeffe joined	00:51
	→ gamja9e joined	00:52
	← gamja9e left	00:52
	← lethjakman left	00:52
	→ kaen joined	00:54
	← main left	00:55
bremner	zsh?	00:56
mattt_	snap	00:56
	← opus left	00:57
mattt_	Is it possible to delete one file and rename another as the deleted file in a single commit?	00:57
bremner	mattt_: sure, just do the appropriate git rm and git add	00:57
	it won't be detected as a rename though	00:57
mattt_	shazam	00:57
	:O	00:58
	hm.. what if i did it in two commits?	00:58
bremner	then the rename should be detected. I'm not actually 100% about the first case.	00:58
mattt_	it won't be detected even if i did a git mv?	00:58
	they're both tracked files	00:59
	hrm.. jeeze, this is what branching is for. duh.	00:59
	→ normanrichards joined	00:59
	→ cmmagid joined	01:00
mattt_	sort of, I guess. Didn't wanna refactor this class, I just rewrote it.	01:00
	← gyre007 left	01:01
	→ apok joined	01:03
	← fuzzybunny69y left	01:04
	→ fuzzybunny69y joined	01:06
	← dreiss left	01:06
	← bootstrapper left	01:06
	← pxdr left	01:07
	→ pxdr joined	01:07
	← btanaka left	01:07
	← decotii left	01:07
	→ infynyxx joined	01:08
	→ boombatower joined	01:08
	← nickpresta left	01:09
	← mattt_ left	01:10
	→ hyperair joined	01:12
	← aantix left	01:12
	→ tjbiddle_ joined	01:13
	→ yshh joined	01:14
	→ BiggFREE joined	01:16
	← hwrd\|work left	01:17
	← tjbiddle left	01:17
	← tjbiddle_ left	01:18
	← trungqtran left	01:18
	← Liir left	01:19
	← juril left	01:19
	← pgokeeffe left	01:23
	← jcao219 left	01:24
	→ pgokeeffe joined	01:26
	← _br_ left	01:26
	← Guest43828 left	01:26
	← fuzzybunny69y left	01:29
	→ _br_ joined	01:29
	← amccloud left	01:30
	← apok left	01:31
	← abf left	01:31
	← _br_ left	01:31
	→ fuzzybunny69y joined	01:32
	→ R1KRDO joined	01:33
	← tinti_ left	01:33
	→ dhruvasagar joined	01:34
	→ Yuuhi` joined	01:34
	→ _br_ joined	01:35
	→ opus joined	01:35
	← Yuuhi left	01:36
	→ JDuke128 joined	01:36
	← breck_ left	01:36
	← fuzzybunny69y left	01:36
	→ pseut joined	01:36
	← pgokeeffe left	01:37
	→ Imaginativeone joined	01:37
	→ pgokeeffe joined	01:37
	← R1KRDO left	01:39
	→ R1KRDO joined	01:39
	← iSaleK left	01:40
	← Anders_J left	01:40
	← yann\|work left	01:40
	→ cods joined	01:41
	→ v0n joined	01:41
	→ iSaleK joined	01:42
	← normanrichards left	01:42
	← iSaleK left	01:42
	← jxpx777 left	01:44
	← infynyxx left	01:44
	← pseut left	01:45
	← NiteRain left	01:45
	← Fandekasp left	01:46
	← soypirate left	01:46
	→ jcao219 joined	01:47
	→ Targen joined	01:48
	← glebd left	01:48
	→ trungqtran joined	01:51
	→ Orbitrix joined	01:51
	→ yann\|work joined	01:53
	→ jxpx777 joined	01:56
	→ stliu joined	01:56
	← Yuuhi` left	01:56
	→ pseut joined	01:57
	→ Davey joined	01:58
	← thiago left	01:59
	→ DrHouse\|WKSTN1 joined	02:00
	← kpreid left	02:00
	→ kpreid joined	02:00
	← jaequery left	02:03
	← pseut left	02:03
	→ luc_ joined	02:04
	← luc_ left	02:05
	→ luc_ joined	02:06
	→ hafeez3000 joined	02:07
	→ hwrd\|work joined	02:08
	dwmw2 → dwmw2_gone	02:08
	→ pseut joined	02:09
	← jalama left	02:10
	→ tjbiddle joined	02:10
	→ rideh joined	02:11
	→ breck joined	02:12
	→ erkules_ joined	02:12
	← trungqtran left	02:14
	← danmackay left	02:15
	← peppyheppy left	02:15
	← erkules left	02:15
	← tjbiddle left	02:16
	← chuckharmston left	02:17
	← cannonball left	02:17
	← rideh left	02:17
	→ bootstrapper joined	02:17
	→ Vampire0 joined	02:18
	→ jdyin joined	02:18
	← daniel_- left	02:19
	← bootstrapper left	02:19
	→ danmackay joined	02:20
	← cmmagid left	02:20
	← Vampire0_ left	02:20
	→ bmac_ joined	02:21
	← pfefferz left	02:21
	← goshawk left	02:22
	← khmarbaise left	02:22
	→ R1KRDO1 joined	02:23
	→ eydaimon joined	02:23
eydaimon	how can I easily diff a file in one branch against the head of another?	02:23
	← aberrant left	02:24
eydaimon	or just viewing the same file in a different branch will suffice for what I'm trying	02:24
	← R1KRDO left	02:25
	← R1KRDO1 left	02:27
	→ pfefferz joined	02:27
engla	eydaimon: the syntax is branch:path/to/file.c	02:29
	→ jcromartie joined	02:29
engla	you can use it with git show or with diff	02:29
eydaimon	engla: thank you	02:30
engla	to be honest, I use path limiting more often.. git diff branch -- path/to/file.c	02:31
eydaimon	cool. the latter one worked	02:31
	→ fuzzybunny69y joined	02:31
eydaimon	couldn't get the first syntax to work	02:31
	git diff routes/index.coffee master:routes/index.coffee	02:31
	fatal: master:routes/index.coffee: no such path in the working tree.	02:32
engla	maybe it still needs the path from the root of the repo	02:32
	anyway, that's not the convenient way anyway	02:32
	unless you want to diff two different files with each other	02:33
eydaimon	yeah, the git diff branch -- worked well	02:33
	now I'll just need to forget it and have to dig through my IRC log a few times to remember it :P	02:34
	→ philsturgeon joined	02:34
philsturgeon	hey, im trying to use git svn to import this repo: https://svn.php.net/repository/phpdoc/en/trunk/	02:34
	← jdunck left	02:34
philsturgeon	I've tried this git svn clone https://svn.php.net/repository/phpdoc/modules/doc-en --no-metadata -t phpdoc/en/tags -b phpdoc/en/branches -T phpdoc/en/trunk phpdoc-en	02:34
engla	eydaimon: all the commands (ok maybe not all) work the same way, you can path limit on git log for example or even git commit	02:35
	← dhruvasagar left	02:35
engla	git log -p path/to/file.c	02:35
philsturgeon	sorry i mean this git svn clone https://svn.php.net/repository/phpdoc/modules/doc-en --no-metadata -t phpdoc/modules/doc-en/tags -b phpdoc/modules/doc-en/branches -T phpdoc/modules/doc-en/trunk phpdoc-en	02:36
	http://d.pr/i/RWYF	02:36
	← bmac_ left	02:36
philsturgeon	can anybody tell me what im getting wrong?	02:36
	→ dhruvasagar joined	02:37
eydaimon	engla: good to know. thank you :)	02:37
	← looper left	02:37
	→ deadman1 joined	02:38
eydaimon	engla: actually I knew you could path limit, but not that you could do it like that across branches for diff	02:38
	i suppose it didn't occur to diff a whole branch	02:38
	← deadman1 left	02:38
	← sysadminAli left	02:38
	→ ddman joined	02:39
ddman	how do i connect to github (https) through a proxy ?	02:39
	git config https.proxy isn't working ?	02:39
	← ErKa left	02:39
	→ looper joined	02:40
	→ wrouesnel joined	02:40
bob2	what does "isn't working" mean?	02:41
	→ cgcardona joined	02:41
	← philsturgeon left	02:42
bob2	http.proxy does work	02:42
	→ Goplat joined	02:42
cgcardona	I'm getting the following error when I try to `git fetch origin master` 'error: insufficient permission for adding an object to repository database .git/objects' - The permissions on the .git/object directory are 755. Do I need to make them 775, 757, or 777? Is that unsafe in any way? thanks	02:44
	→ megha joined	02:44
	← opus left	02:45
bremner	cgcardona: if you have to ask if 777 permissions are unsafe...	02:46
	← jonshea left	02:46
	→ NiteRain joined	02:48
	→ glennpratt joined	02:49
bremner	cgcardona: are you trying to share a git repo between multiple users?	02:49
ddman	bob2, it is trying to connect to port 443 of the host instead of proxy	02:50
	← fission6 left	02:52
	← megha left	02:52
	← jxpx777 left	02:52
	← pseut left	02:55
	← AaronMT left	02:55
	→ leeN joined	02:57
	→ lresende joined	02:57
	← breck left	02:58
	→ psoo joined	02:58
ddman	git clone https://github.com//openstack/cinder /opt/stack/cinder	02:58
	Cloning into /opt/stack/cinder...	02:58
	error: Failed connect to github.com:443; Connection refused while accessing https://github.com//openstack/cinder/info/refs	02:58
bob2	we can't see your config	02:59
	→ pseut joined	02:59
	→ breck joined	03:00
	→ cryosphere joined	03:01
ddman	http.proxy=http://proxy.com:8088/	03:01
	https.proxy=http://proxy.com:8088/	03:01
	core.repositoryformatversion=0	03:01
	core.filemode=true	03:01
	core.bare=false	03:01
	core.logallrefupdates=true	03:01
	remote.origin.fetch=+refs/heads/:refs/remotes/origin/	03:01
	remote.origin.url=http://github.com/openstack-dev/devstack.git	03:01
	branch.master.remote=origin	03:01
	branch.master.merge=refs/heads/master	03:02
	i changed the proxy.com to my proxy	03:02
cgcardona	bremner: I've got a git repo on github and I'm trying to push and pull to/from it. so `git push origin master` works but `git pull origin master` gives me the error that I pasted above	03:03
milki	cgcardona: similar to !fetch4 probably	03:03
gitinfo	cgcardona: [!fetchfour] Never use the four-word version of git-fetch or git-pull (e.g. git fetch <remote> <refspec>). It always ends in tears. Yes, if you understand the implications of FETCH_HEAD it can technically be done, but really it is easier to just fetch the whole remote (or perhaps edit the fetchspec if you never want other bits). If you must, see !fetch4why	03:03
cgcardona	'error: insufficient permission for adding an object to repository database .git/objects'	03:03
	← rakl left	03:03
milki	o, but that error isnt related to the blurb	03:04
cgcardona	also `git fetch` gives me the same error	03:04
	it's related to the permissions of .git/objects.	03:04
	I'm googling around of course	03:04
milki	who made the repo?	03:04
cgcardona	i just thought something might jump out to someone in here.	03:04
	i did	03:04
milki	huh	03:04
cgcardona	i created it on github and then cloned it	03:05
milki	shouldnt have problems on fresh clones	03:05
	← Dreamer3 left	03:05
	← yakko left	03:05
	milki doesnt know	03:05
cgcardona	ok no prob	03:06
	milki blames filesystem or os	03:06
cgcardona	i ran the clone without sudo. perhaps thats it.	03:06
milki	which is fine. you shouldnt need root to clone...	03:06
cgcardona	I can of course just check out a fresh copy. I don't have any local changes. But I really need to figure it out.	03:06
	thanks anyway	03:06
	milki: Yea I didn't think I did	03:06
	what should the permissions on the .git/objects directory be?	03:06
	← Milossh left	03:07
cgcardona	i'm checking google	03:07
	← glennpratt left	03:07
milki	depends. default is probably 755 if you have umask 022	03:07
bob2	very likely you used sudo	03:08
	← breck left	03:09
	→ Noldorin joined	03:09
	→ d2dchat joined	03:09
	→ Milossh joined	03:11
cgcardona	bob2: I didn't use sudo to do the initial clone.	03:12
	i can bring up the command in my history from last night	03:13
	→ jonshea joined	03:14
	→ jaequery joined	03:14
	← jaequery left	03:15
bob2	how on earth did you end up with the perms broken then?	03:15
	← d2dchat left	03:15
	→ bootstrapper joined	03:17
	← bootstrapper left	03:18
	→ bootstrapper joined	03:18
cgcardona	i'm unclear on that	03:19
	i just created a repo from my github account. went to terminal and 'git clone [email@hidden.address] which pulled it down to my local machine. I then made local changed and committed and pushed back to github. That part worked. But then I merged in some changes from a friend and attempted to `git fetch` them to my local machine and got the fail	03:20
	← Jahm left	03:20
cgcardona	Even though I don't need to use sudo to push or pull do i need to be using it to switch branches, add/remove, and commit?	03:21
	→ Jahm joined	03:21
	→ trungqtran joined	03:21
	← adamben left	03:22
bob2	unless you're on some crappy filesystem or something	03:22
	→ Progster joined	03:23
	← pgokeeffe left	03:23
cgcardona	so yes I do need to be using sudo to switch branches, add/remove, and commit but don't use sudo to push/pull? Also FWIW I'm on OSX	03:23
	→ eli joined	03:24
	→ infynyxx joined	03:24
	← aartist left	03:25
	→ svetlyak40wt joined	03:25
	← infynyxx left	03:25
	→ thiago joined	03:26
	→ weiyang joined	03:26
	→ jalama joined	03:26
bob2	no of course you don't need sudo for any of that	03:28
cgcardona	interesting. So I probably don't need sudo to use vim either (I realize it's off topic.) Yea I need to change my user's status on this machine.	03:28
	→ jaequery joined	03:28
cgcardona	thanks bob2	03:28
	← trungqtran left	03:30
	→ tavist0 joined	03:30
	→ virako joined	03:31
bob2	what	03:31
	← BiggFREE left	03:32
cgcardona	ok i found the issue. I've been using sudo to use vim and to switch between branches, add/remove, and commit. This is what the issue was. When I make a fresh checkout and don't use sudo for any of those it works fine.	03:33
	sweet	03:33
	→ hafeez3000_ joined	03:33
	← pppaul left	03:34
bob2	yes don't use sudo if you don't know why you'd want to use sudo	03:35
	← hafeez3000 left	03:35
	hafeez3000_ → hafeez3000	03:35
	→ ajc_ joined	03:35
	← JDuke128 left	03:36
	→ amigojapan joined	03:37
cgcardona	the situation is I just left my job of a year and a half and am migrating back on to my personal mac. I haven't really touched the box in about a year and a half and I vaguely remember needing to use sudo to use vim. Anyway it doesn't really matter I've got it figured out now.	03:38
	→ R1KRDO joined	03:40
	← hwrd\|work left	03:41
	→ jaseemabid joined	03:41
	← Milossh left	03:43
	← pfefferz left	03:44
	← Progster left	03:45
	← wrouesnel left	03:45
	← bootstrapper left	03:46
	→ dmarkham joined	03:46
	→ khmarbaise joined	03:47
	← jaseemabid left	03:49
	→ sgambino joined	03:49
	→ ErickLan_ joined	03:50
	→ pfefferz joined	03:51
	← engla left	03:51
	→ mdszy joined	03:53
	← tureba left	03:53
	→ tjbiddle joined	03:53
mdszy	Whenever I have this git post-receive hook run a git command in another directory, I keep getting "Not a git repostitory '.' errors".	03:53
	Anyone have any idea why?	03:53
	→ tureba joined	03:54
mdszy	I tried cding into the directory through the script, and setting the GIT_WORK_TREE variable.	03:54
	← cgcardona left	03:54
	← fuzzybunny69y left	03:54
bob2	bpaste.net the script	03:54
	← virako left	03:54
mdszy	http://sprunge.us/cDjF?sh	03:55
	→ fission6 joined	03:56
bob2	yeah	03:57
	those commands require a repository	03:57
	you probably don't want merge either	03:57
mdszy	bob2, I know that, and whenever I run them by hand inside of the repo they work fine. But not in the script.	03:57
bob2	sine it'll preserve local mods isntead of nuking them	03:57
mdszy	what do you suggest I do, then?	03:57
	→ fuzzybunny69y joined	03:57
bob2	'ls -a /home/pi/statsapp'	03:58
mdszy	.git is in there.	03:58
	← ErickLan_ left	03:58
mdszy	Like I said, if the commands are run by hand -- outside of the script -- they work	03:58
	if I run the script by hand, they work	03:58
	→ tinocos joined	03:58
mdszy	but if the script is run as a hook, they don't	03:58
cryosphere	if I have a merge conflict - how do I keep my (local, current branch) version	03:58
	rather than resolve line by line	03:58
	← Fels left	04:00
	← tchan left	04:00
	← jcromartie left	04:01
EugeneKay	cryosphere - git checkout HEAD -- file.txt	04:01
	cryosphere - you may also be interested in the -Xours option in man git-merge	04:01
gitinfo	cryosphere: the git-merge manpage is available at http://jk.gs/git-merge.html	04:01
	→ dreiss joined	04:04
	← myx left	04:04
	← jiku left	04:05
	→ shakes808 joined	04:05
	gitinfo set mode: +v	04:06
	→ bootstrapper` joined	04:06
	→ jcromartie joined	04:06
	→ tchan joined	04:06
	← tchan left	04:06
	→ tchan joined	04:06
mdszy	any ideas as to what might be my issue?	04:09
	← weiyang left	04:09
	← Bankzilla left	04:12
	→ beautifulmind joined	04:13
	← bootstrapper` left	04:13
	← HumanSky left	04:14
	← m0ltar left	04:14
	→ lavaman joined	04:15
	→ guns joined	04:15
	→ breck joined	04:16
	→ rakl joined	04:17
	← SBoudrias left	04:17
	→ __class__ joined	04:17
	← prudnikov left	04:18
	← skorgon left	04:18
	← Praise left	04:18
	SeoZ-work[AWAY] → SeoZ	04:19
	← mdszy left	04:19
	← dmarkham left	04:19
	← thesisus left	04:20
	→ skorgon joined	04:20
	→ herdingcat joined	04:21
	→ normanrichards joined	04:22
	→ Praise joined	04:22
	→ pierceray joined	04:25
	← the_cybersphinx left	04:25
	→ R1KRDO1 joined	04:28
	→ glennpratt joined	04:28
	→ sangi joined	04:29
	← R1KRDO left	04:30
	→ diegoviola joined	04:31
	← Noldorin left	04:31
	← arvind_khadri left	04:32
	→ goshawk joined	04:33
	← Goles left	04:33
	← hafeez3000 left	04:35
	← shakes808 left	04:37
	→ wrouesnel joined	04:38
	← breck left	04:38
	← R1KRDO1 left	04:39
	← glennpratt left	04:41
	← hatake left	04:41
	← fuzzybunny69y left	04:41
	→ Fandekasp joined	04:41
	→ hatake joined	04:41
	→ glennpratt joined	04:41
	← amigojapan left	04:43
	→ mikewintermute joined	04:45
	→ hinst joined	04:45
	← pierceray left	04:45
	→ pppaul joined	04:45
	← tinocos left	04:47
	← eletuchy left	04:47
	gitinfo set mode: +v	04:47
	→ eletuchy joined	04:48
	← Davey left	04:49
	← rakl left	04:50
	← aspiers left	04:50
	→ fuzzybunny69y joined	04:52
	← jaequery left	04:52
	← lavaman left	04:52
	→ Zombie__ joined	04:53
	→ gmc_ joined	04:55
	→ aspiers joined	04:55
	→ infynyxx joined	04:55
	← schoos left	04:56
	→ jaequery joined	04:57
	← gmc_ left	04:57
	← pppaul left	04:57
	← glennpratt left	05:00
	→ wroathe joined	05:00
	← preflex left	05:01
	→ preflex_ joined	05:01
	→ lavaman joined	05:02
	← hinst left	05:02
	preflex_ → preflex	05:02
	→ cagedwisdom_ joined	05:02
	← tjbiddle left	05:04
	← ping-pong left	05:05
	→ rastar joined	05:05
	← wroathe left	05:05
	→ gmc_ joined	05:07
	← gmc_ left	05:08
	← leeN left	05:11
	→ Noldorin joined	05:11
	← dhruvasagar left	05:12
	← ISF left	05:12
	→ wrouesnel1 joined	05:12
	→ dhruvasagar joined	05:13
	← fuzzybunny69y left	05:15
	→ jaseemabid joined	05:17
	→ citizen-stig joined	05:17
	← goshawk left	05:17
	← marc_kdab_ left	05:18
	← DrHouse\|WKSTN1 left	05:18
	→ marc_kdab_ joined	05:18
	→ pretty_function joined	05:21
	← Zombie__ left	05:22
	← monk3y left	05:24
	→ a-l-e joined	05:25
	→ hwrd\|work joined	05:25
	← Saby left	05:26
	← wrouesnel1 left	05:26
	← OOPMan left	05:26
	← hellopat left	05:27
	→ OOPMan joined	05:27
	← kill-9_ left	05:29
	← svetlyak40wt left	05:30
Caelum	can I change the comment at the bottom of commit messages?	05:31
	I want it to just show the modified: file lines, without the instructions	05:31
	→ pgokeeffe joined	05:31
	← OOPMan left	05:32
	→ ping-pong joined	05:34
	→ ISF joined	05:34
	← jaseemabid left	05:35
	→ private joined	05:36
	private → Guest36822	05:36
thiago	Caelum: why?	05:39
	← Noldorin left	05:40
	→ OOPMan joined	05:41
	→ sluther joined	05:42
	→ tjbiddle joined	05:43
	→ nafg_ joined	05:45
	← Err404NotFound left	05:46
	← zivester left	05:46
	→ fuzzybunny69y joined	05:46
	→ Err404NotFound joined	05:47
	→ zivester joined	05:49
Caelum	thiago: because I email the commit message	05:50
	← lavaman left	05:50
Caelum	thiago: along with the comment	05:50
	← herdingcat left	05:50
	→ breck joined	05:50
	→ lun_ joined	05:51
	→ herdingcat joined	05:52
	→ archdori joined	05:53
	← marc_kdab_ left	05:53
	→ TheJH joined	05:53
	→ marc_kdab_ joined	05:53
	← breck left	05:55
	→ breck joined	05:55
	← douglarek left	05:55
	→ maleknet joined	05:56
	← Fandekasp left	05:56
	← lun_ left	05:58
	← huoxito left	05:58
	← dhruvasagar left	05:58
	→ dmarkham joined	05:58
	→ lun_ joined	05:59
	→ lpapp joined	05:59
	← nigelb left	05:59
	← jsec left	05:59
	← lpapp left	05:59
	→ lpapp joined	05:59
	← hwrd\|work left	05:59
	→ nigelb joined	05:59
lpapp	hi, how can one modify the commit creating time/date? Used to have an option because an acquaintance implemented it few years ago, but cannot recall anymore.	05:59
	creation*	06:00
	→ hwrd\|work joined	06:00
	← eletuchy left	06:00
	← resmo_ left	06:00
	→ ivan` joined	06:01
	← OOPMan left	06:02
	→ jsec joined	06:02
	← infynyxx left	06:02
	← maleknet left	06:03
	← CareBear\ left	06:04
	→ jaseemabid joined	06:04
	← babilen left	06:04
	→ babilen joined	06:05
	← lpapp left	06:06
	→ ryao joined	06:06
	→ noyb joined	06:07
	← lamarus left	06:07
	→ lamarus joined	06:07
	← TheJH left	06:09
	← WinstonSmith left	06:10
	← ryao left	06:11
	→ WinstonSmith joined	06:12
	→ ryao joined	06:12
	← sluther left	06:13
	← ISF left	06:15
	→ subhojit777 joined	06:15
	→ maleknet joined	06:15
	→ amesha joined	06:16
	← asaf left	06:17
	← bgerber left	06:17
	hatake → daenglinuxx	06:17
	← ryao left	06:17
	→ ryao joined	06:18
amesha	hi everyone, i have a repository which has a set of files in it that i would like to track with version control for my own use, but i would also like to publish the repository publicly without those files	06:18
	← lun_ left	06:19
amesha	could anyone point me in the direction of a good strategy for accomplishing something like that?	06:19
thiago	either you track or you don't	06:19
amesha	so it's not possible in one repository?	06:19
thiago	so my recommendation is that you keep two branches	06:19
	one without the files and one with	06:19
amesha	ah okay	06:19
	→ amigojapan joined	06:19
amesha	thanks, thiago	06:20
	← madewokherd left	06:21
frogonwheels	amesha: it also depends on the relationship. you could have just those files in a sep repository (design it to be in a sep dir)	06:21
	← fphilipe left	06:21
frogonwheels	amesha: .. or if they are config files that need to be in the same dir, symlink them in from a separate repo.	06:21
	→ shruggar joined	06:22
	← fuzzybunny69y left	06:22
	← ryao left	06:23
	← Tritania left	06:24
	→ ryao joined	06:24
	→ Fretta joined	06:24
	→ etcetera joined	06:24
	→ Tritania joined	06:25
	← Err404NotFound left	06:25
	→ fuzzybunny69y joined	06:25
	→ lun_ joined	06:25
	← jcao219 left	06:26
	← citizen-stig left	06:28
	← ryao left	06:29
	→ jjang_ joined	06:30
	← v0n left	06:30
	→ ryao joined	06:30
	← jjang_ left	06:33
	← jjang left	06:34
	← tjbiddle left	06:34
	← jaseemabid left	06:35
	← ryao left	06:35
	→ ISF joined	06:35
	→ Error404NotFound joined	06:36
nevyn	so first perforce now visualstudio...	06:36
	win.	06:36
	→ aberrant joined	06:36
	→ ryao joined	06:36
	← amesha left	06:37
	← breck left	06:37
	→ amesha joined	06:37
	→ asaf joined	06:39
	← lun_ left	06:40
	→ LeMike joined	06:41
	→ Cromulent joined	06:41
	← ryao left	06:41
	→ ryao joined	06:42
	→ lun_ joined	06:44
	← ryao left	06:44
	→ breck joined	06:45
	← Fretta left	06:45
milki	dows	06:45
	→ ruskie joined	06:45
	← grummund left	06:46
	← Goplat left	06:47
	← lun_ left	06:49
	← marc_kdab_ left	06:49
	→ soee joined	06:49
	← etcetera left	06:50
	← herdingcat left	06:51
Caelum	you mean the VCS ms wrote?	06:52
	how is it	06:52
	→ Sigma joined	06:53
	← skorgon left	06:54
	→ jjang joined	06:54
nevyn	http://blogs.msdn.com/b/bharry/archive/2013/01/30/git-init-vs.aspx	06:55
	Caelum: it's now git ;)	06:55
	→ infynyxx joined	06:56
	← shruggar left	06:56
	← tewecske left	06:56
	← nafg_ left	06:56
	← fission6 left	06:57
	← breck left	06:57
	→ herdingcat joined	06:58
	→ breck_ joined	06:58
	← fbdystang left	06:58
	→ lirt joined	06:58
	← amesha left	06:58
	→ amesha joined	06:59
	← Bass10 left	06:59
	→ skorgon joined	07:00
	→ pantsman joined	07:01
	→ lun_ joined	07:02
	← sajan left	07:02
	→ nafg joined	07:03
	→ JohnBat26 joined	07:03
	→ system64 joined	07:04
	→ papegaaij joined	07:05
	→ jaseemabid joined	07:07
	→ Fandekasp joined	07:07
	← jaseemabid left	07:08
	← psoo left	07:09
	← lun_ left	07:10
	→ lun__ joined	07:10
	← khmarbaise left	07:10
	← pseut left	07:10
	← archdori left	07:10
	→ jaseemabid joined	07:12
	← infynyxx left	07:12
	→ arvind_khadri joined	07:12
	← jcromartie left	07:14
	← PaulCapestany left	07:16
	← robustus left	07:17
	→ thehandler joined	07:18
	→ cbreak-work joined	07:18
	← breck_ left	07:19
	→ smirn0v joined	07:19
	← thehandler left	07:19
	→ ThomasLocke joined	07:19
	→ breck joined	07:19
	→ beliveyourdream joined	07:20
	→ robustus joined	07:20
	→ skyxxx joined	07:20
	← amesha left	07:20
	→ thehandler joined	07:20
	← Sigma left	07:21
	← mishok13 left	07:21
	→ dhruvasagar joined	07:21
	→ amesha joined	07:21
	→ OOPMan joined	07:23
	→ rath joined	07:24
	← Ertac left	07:24
	← Cromulent left	07:25
	→ WorkeD joined	07:27
	→ Milossh joined	07:27
	→ jkremser joined	07:29
	→ CareBear\ joined	07:30
	chrisgward → chrisward	07:30
	← jjang left	07:31
	→ nilg joined	07:32
	→ fRAiLtY- joined	07:32
	← beautifulmind left	07:32
	→ marthinal joined	07:33
	→ shruggar joined	07:34
	← toothrot left	07:34
	← noyb left	07:34
	→ tsdh joined	07:36
	← revyd left	07:36
	→ toothr joined	07:37
	→ GeertV joined	07:38
	← cheater left	07:39
	← Sebo left	07:39
	← lifelike left	07:40
	→ lifelike joined	07:40
	← clone1018 left	07:41
	← amesha left	07:42
	→ amesha joined	07:43
	← segher left	07:44
	← hyperboreean left	07:44
	→ fphilipe joined	07:45
	← jaseemabid left	07:46
	→ hoschi joined	07:46
	← normanrichards left	07:47
	→ TonyM joined	07:48
	→ main joined	07:48
	→ mishok13 joined	07:50
	→ Sigma joined	07:50
	→ btree joined	07:51
	→ gerryvdm_mbp joined	07:52
	← gerryvdm_mbp left	07:53
	← btree left	07:53
	← arvind_khadri left	07:54
	→ arvind_khadri joined	07:54
	← lresende left	07:55
	← lun__ left	07:55
	→ dc5ala joined	07:55
	← amir_ left	07:56
	→ jjang joined	07:58
	→ lun_ joined	07:58
	→ tbosua joined	07:58
	← fphilipe left	07:59
	→ GarethAdams joined	07:59
	→ jaseemabid joined	07:59
	→ tjadc joined	08:01
	← amesha left	08:02
	← sangi left	08:03
	→ amesha joined	08:03
	← smirn0v left	08:03
	→ tPl0ch joined	08:04
	← thiago left	08:04
	← LeMike left	08:04
	← yewton left	08:04
	← jaseemabid left	08:05
	→ yewton joined	08:05
	→ smirn0v joined	08:07
	→ pppaul joined	08:07
	→ jargon joined	08:08
	→ flijten joined	08:09
	→ rzs joined	08:10
	← Milossh left	08:12
rzs	I have a git repo with mutiple subdirectories. I want to migrate two subdirectories into a git repo of their own. Im able to migrate a single subdirectory by following the instructions given at http://www.geekaholic.org/2010/02/splitting-git-repo.html	08:12
	Can i migrate more than one subdirectory into a separate git repo ?	08:12
	→ WillMarshall joined	08:13
	→ lng joined	08:14
lng	Hi! After `git submodule add`, I add it to .gitignore, right?	08:15
	← alexwhitman left	08:15
	→ timj1 joined	08:15
lng	because now I see it as new file	08:15
	→ horatio_cromwell joined	08:16
	→ alexwhitman joined	08:16
	← yshh left	08:16
jast	lng: nah, that's the plan. you can 'git add' the submodule to record a changed revision in the submodule (remember that submodules are versioned, i.e. there's always a specific revision associated with any given state of the main project)	08:16
	→ Zer000 joined	08:16
	← shruggar left	08:17
lng	jast: I see thanks!	08:17
	← guns left	08:17
jast	rzs: you can do the same procedure multiple times	08:17
	← bauruine left	08:17
	← letharion_ left	08:17
jast	rzs: side note, the --index-filter approach in the last step is indeed _much_ faster than the --tree-filter approach	08:17
Zer000	So I committed and pushed something I regret - some rather large media files that shouldn't even be in the project. How do I undo the last commit locally and on the remote?	08:17
jast	also, don't bother doing gc/prune etc. before you're all done	08:17
	← timj left	08:18
jast	the steps for that aren't quite right in that post, anyway	08:18
	→ Sonderblade joined	08:18
	← system64 left	08:18
	→ yshh joined	08:18
jast	Zer000: !fixit	08:18
gitinfo	Zer000: [!fixup] So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	08:18
	→ Milossh joined	08:19
	← main left	08:19
	→ lethjakman joined	08:20
	← dmarkham left	08:20
	→ cheater joined	08:21
Zer000	jast that helps thank you!	08:22
	← dreiss left	08:22
	→ lxsameer joined	08:22
	← amesha left	08:22
lxsameer	how can i fetch all the remote refs?	08:22
	→ amesha joined	08:23
rzs	jast: but if i do those steps then the new repo contains the contents of the project i filtered rather than the filtered project directory	08:23
	→ jceb_ joined	08:23
Zer000	There's something I don't get though, If my local repo = my remote repo at the moment, I do "git reset HEAD^" then make my changes, commit, then push, what happens to remote?	08:23
rzs	jast: so if i run the steps again for another project wont it mix the contents of that project with the previous one ?	08:23
	← Milossh left	08:23
Zer000	ah, it says I cannot push	08:24
	→ shruggar joined	08:24
	← tavist0 left	08:25
	← shruggar left	08:25
	→ dougmencken joined	08:25
dougmencken	hi guys!	08:25
	← jjang left	08:25
dougmencken	after 'git reset', there're no "renamed --->", but "deleted" and "untracked"	08:26
	how to turn them back to "renamed"?	08:26
	→ Milossh joined	08:26
GarethAdams	dougmencken: from what I can tell, the rename detection only happens on `add`ed files	08:28
	→ acrocity_ joined	08:28
	← JoeHazzers left	08:28
	→ swair joined	08:28
Nevik	well, it must be a matching delete and add	08:28
	→ adamben joined	08:28
dougmencken	hmm, it was a rename, then I modified	08:28
Nevik	matching as in git's content compare says "yep, that's probably a rename"	08:29
	dougmencken: dont modify and rename in the same commit	08:29
dougmencken	new file: and deleted: now :(	08:29
Nevik	split those actions up into two commits	08:29
	← acrocity left	08:29
	acrocity_ → acrocity	08:29
dougmencken	Nevik: but other files won't find renamed header then	08:29
	← yshh left	08:29
	→ yshh_ joined	08:29
dougmencken	Nevik: and this will be broken un-buildable commit	08:29
Nevik	so you have two choices now:	08:30
	1) live with it (have the working build and forget about the rename detection)	08:30
	2) revert either the rename or the content changes manually, make an intermediate commit, then make the reverted change again and commit	08:30
	stash can come in helpful in some of those cases	08:31
	in choice 2, you can do it easy and produce one broken build for the intermediate commit, or if you'Re really pedantic, you can go and fix the code for it	08:32
jast	Zer000: there's a --force option for push	08:32
Nevik	that's all up to you, dougmencken	08:32
	← lirt left	08:32
Zer000	jast, I think I got it to work!	08:32
	good night	08:32
	← Zer000 left	08:32
	→ lirt joined	08:33
jast	rzs: 1) create copy of original; run filter-branch to reduce to subdir a. 2) create another copy of original; run filter-branch to reduce to subdir b. and so on. finally, remove all those subdirs from original.	08:33
	→ LeMike joined	08:33
	← wolfmitchell left	08:34
	→ BadDesign joined	08:34
	← BadDesign left	08:34
	→ BadDesign joined	08:34
	SeoZ → SeoZ-work[AWAY]	08:34
	→ AdrienBrault joined	08:35
	← boombatower left	08:35
	← amesha left	08:35
	→ amesha joined	08:36
	→ wolfmitchell joined	08:36
	→ Fels joined	08:37
rzs	jast: thanks for reply. i want to keep my original repo as it is. If i understand u correctly, i create a copy of original repo and do filter branch to reduce to a. Then create another copy of original and run filter branch to reduce to b. So, now i have two copies of original A and B respectively. Can i push them to my new repo such that the new repo contains A and B subdirectories separate instead of the contents of both mixed ?	08:38
	← pppaul left	08:38
	→ francisfish joined	08:38
jast	rzs: you _could_, but I don't see the big advantage over having two separate repositories	08:39
rzs	jast: A and B wont be repos in themselves they would just be part of the new repository	08:39
Caelum	figured out my prepare-commit-msg hook: perl -ni -e 'print unless /^#.+/ && !/^#\t[^:]+:/' $1	08:41
jast	oh, so you want to have a/ and b/ in there?	08:41
rzs	jast: yeah	08:41
jast	might not be possible with the standard subdirectory-filter, then	08:41
	this is going to be trickier	08:41
	← smirn0v left	08:42
rzs	jast:ok, plz let me know if u have any ideas	08:42
jast	I think it involves a more complex --index-filter, one in which you delete everything _but_ a and b	08:42
	that'll require a certain amount of shell scripting skills	08:42
	→ smirn0v joined	08:42
dougmencken	Nevik: okay, let's choose 1: rename detection is just useful thing, not a must :)	08:42
Nevik	jast: wouldnt merging the two (A and B) repos work though?	08:42
	after extracting them separately, jast	08:42
	dougmencken: okay :D	08:43
jast	I suppose... if you don't mind the weird history	08:43
dougmencken	thanks for help, see you	08:43
	← dougmencken left	08:43
Nevik	i dont know what history filter-branch for subdirs produces. will it keep the original hashes? hmmmm no, it would produce new commits, right?	08:43
	→ maxmanders joined	08:43
	← jaequery left	08:44
grawity	Any kind of filtering will produce new commits	08:44
Nevik	damn, then no, merging will be ugly	08:44
jast	also how do you tell merge to move the files back into subdirs	08:45
	subdirectory-filter promotes them to the top level	08:45
	← heroux left	08:45
Nevik	ah, see, didnt know that. kay, i'll shut up now :P	08:45
	→ x82_nicole joined	08:46
	← weltall left	08:46
	→ weltall joined	08:46
	← pantsman left	08:47
	→ svetlyak40wt joined	08:48
	→ sobhan joined	08:48
	→ anderse joined	08:49
	→ dreiss joined	08:50
	← hwrd\|work left	08:50
	→ main joined	08:51
	← WorkeD left	08:53
	← francisfish left	08:54
lng	I'm on master branch and have 'remotes/origin/HEAD -> origin/master', but I also have 'remotes/origin/2.0' - I need to checkout this branch. How do I do it? I forgot...	08:54
	→ hwrd\|work joined	08:55
lng	do I git checkout -b remotes/origin/2.0?	08:55
cmn	you can create a local branch of the same name, git checkout -t origin/2.0	08:55
	and that starting point	08:55
	→ letharion_ joined	08:56
jast	with recent gits, 'git checkout 2.0' will do the trick	08:56
cmn	as long as the name is unique enough	08:56
jast	sure	08:57
	→ francisfish joined	08:57
lng	yes!!	08:57
	git checkout 2.0!	08:57
	jast: thanks!	08:57
	→ memoryleak joined	08:57
lng	cmn: why origin/2.0? locally I had only master	08:57
cmn	precisely because of that	08:58
lng	ah	08:58
cmn	it's a remote-tracking branch	08:58
lng	yes	08:58
rzs	When i run git subtree split -P subdirname -b foo i get 'git subtree command not found'. Do i need to install subtree separately ?	08:58
cmn	what I don't like about the magic jast showed is that it gives the impression of checking out the remote-tracking branch, which is not quite what's happening	08:58
	→ xxc_michael joined	08:58
cmn	rzs: it's in contrib/ and many packagers don't install it by default	08:59
rzs	cmn: ok, im on a ubuntu machine. how can i install it ?	08:59
cmn	make install under contrib/ in the git source	09:00
	→ giallu joined	09:00
	→ spuz joined	09:01
	← xiangfu left	09:01
	→ leeN joined	09:02
	→ FernandoBasso joined	09:02
cmn	rather, contrib/subtree/ or whatever its subdir is	09:02
	← amesha left	09:03
rzs	cmn: alright thanks	09:03
	→ amesha joined	09:03
	→ jlopez joined	09:04
	← anderse left	09:04
	→ DeviantPeer joined	09:04
	→ hyperboreean joined	09:05
	→ valgrind joined	09:05
	→ blackmat` joined	09:06
	← swair left	09:06
	← TeamColtra left	09:06
	→ swair_ joined	09:06
	← jdav_gone left	09:06
	← swair_ left	09:06
	gitinfo set mode: +v	09:06
blackmat`	helo	09:06
	← frsk left	09:06
	→ Xethron joined	09:07
	→ LimeBlast joined	09:08
	→ cheater_ joined	09:08
	← Iszak left	09:09
	← blubberdiblub left	09:09
	← dreiss left	09:10
	← cheater left	09:11
	← Fels left	09:13
	adhawkins-away → adhawkins	09:13
	← lifelike left	09:14
DeviantPeer	if I have a branch (master) with a long history of commits, is it possible to create a new branch (publish, or public) where most of the commits from master would show up as squashed?	09:14
	→ alkino joined	09:14
DeviantPeer	I looked into checkout --orphan but it's too looks like it's too drastic (as if starting a completly new commit history).	09:15
	← Vile left	09:15
	→ acrocity_ joined	09:15
	← acrocity left	09:16
	acrocity_ → acrocity	09:16
DeviantPeer	So it would be something like master:a-B-C-D-E-F-G... ; and in publish:D-F...	09:16
	→ laurav joined	09:17
GarethAdams	DeviantPeer: do you know about `rebase -i`?	09:17
	→ Anders_J joined	09:18
DeviantPeer	GarethAdams: I though I did. :) So I would start with a "publish" branch created from the first commit, and then rebase -i from there?	09:18
	← x82_nicole left	09:19
	→ segher joined	09:19
DeviantPeer	GarethAdams: or branch publish from master:HEAD and rebase in the publish branch?	09:19
	→ lifelike joined	09:19
	→ Iszak joined	09:20
	→ Cromulent joined	09:21
	→ fphilipe joined	09:22
	→ koskoz joined	09:23
koskoz	hi	09:23
	is is possible to show all the modifications made to a file on all branches, not just a specific one?	09:23
	← pgokeeffe left	09:23
	→ jdav_gone joined	09:25
	→ ozette joined	09:25
	→ howlymowly joined	09:25
	→ buibex joined	09:25
	← hwrd\|work left	09:27
	→ marhaban joined	09:27
	→ pgokeeffe joined	09:28
cmn	--all and filter by that path	09:28
	← b1rkh0ff left	09:28
	← amesha left	09:30
	→ amesha joined	09:31
	→ Fels joined	09:31
	→ Sebo joined	09:33
koskoz	cmn: with which command?	09:35
	← Imaginativeone left	09:35
cmn	log	09:35
	the one that walks through history	09:35
pxdr	ping cmn	09:35
m1au_	hi,	09:37
	does anyone use gitlab with puppet modules?	09:37
	and have problems of files directory not displaying the content in gitlab?	09:37
	← nigelb left	09:38
	← PaulePanter left	09:38
	← amesha left	09:39
cmn	you should ask the gitlab people	09:40
	← smirn0v left	09:41
	→ myx joined	09:41
	→ smirn0v joined	09:42
koskoz	thanks cmn, now I have to find what to do with my commit sha1	09:43
	← WillMarshall left	09:43
cmn	what to do with it?	09:43
m1au_	cmn: do they have official channel?	09:44
cmn	I believe so	09:44
	← sobhan left	09:44
m1au_	there are 24 ppl on #gitlab with no topic.	09:44
cmn	http://gitlabhq.com/community/	09:44
m1au_	does not look like one.	09:44
cmn	we don't know about gitlab-specific issues	09:44
	→ Vile joined	09:45
m1au_	yes, saw forum, wiki but no irc on that page	09:45
cmn	then they don't offer IRC support	09:45
m1au_	oh well, gotta write in their forum then	09:45
	← netiul left	09:45
	→ zoresvit joined	09:46
	← diegoviola left	09:47
koskoz	cmn: I wanna show the file in this commit	09:47
cmn	then use show or similar	09:47
koskoz	yeah, just saw that on google	09:47
	appart from add, commit, checkout, branch, pull, push and reset I'm noob	09:47
	← alexO left	09:48
cmn	read !book	09:48
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	09:48
	→ dreixel joined	09:48
	← pgokeeffe left	09:49
	← marhaban left	09:50
	→ snql joined	09:50
dreixel	my git merge failed. As in, completely: if I edit the (very minor) conflicts and try to commit, it's just a new commit. I don't want this because it'll make future merges impossible... what can I do?	09:51
	→ bauruine joined	09:51
cmn	a merge will always be a new commit	09:51
	← pity left	09:52
cmn	and why would this make future merges impossible?	09:52
	← Gentlecat_ left	09:52
	→ Ksg89 joined	09:52
dreixel	well, it becomes a single new commit. So it doesn't preserve the fact that it's actually hundreds of old commits	09:52
	→ BiggFREE joined	09:52
dreixel	my git merge ends with "Merge with strategy octopus failed."	09:53
	and not with "Automatic merge failed; fix conflicts and then commit the result."	09:53
	→ jatt joined	09:53
cmn	this is not true	09:54
dreixel	well, I've never seen this behaviour before. perhaps I've done something wrong, but I don't know what.	09:54
	→ chisler joined	09:54
cmn	first of all, don't do octopus merges	09:55
	merge whatever branches are appropriate one by one and fix whatever conflicts	09:55
	and read up on what merges are	09:55
	← Sebo left	09:55
	→ swair joined	09:56
	→ marhaban joined	09:56
dreixel	ah, nice and friendly.	09:56
jast	in any case, don't use octopus merges	09:58
	they're really more of a curiosity and, as you noticed, they don't work nearly as reliably as normal merges	09:59
dreixel	it's not like I chose to. I just did "git merge origin master"	09:59
jast	yeah, so now you know to not make that mistake again :}	09:59
	→ netiul joined	10:00
dreixel	but what's wrong with that?	10:00
cmn	you're asking it to merge two different branches	10:00
dreixel	oh, maybe that doesn't mean what I think it does...	10:00
	right.	10:00
	so git merge master is what I want, I guess.	10:01
	← marhaban left	10:01
	→ georg joined	10:01
jast	dreixel: or git merge origin/master	10:01
	'origin master' is only valid in push/fetch/pull, because those accept a remote and a refspec as arguments. merge doesn't. it wants a ref.	10:02
dreixel	jast: alright, that must be where I got confused...	10:02
	that clears it up, thanks.	10:02
	→ Chiku\|dc joined	10:03
	→ Xeno joined	10:03
	→ PaulePanter joined	10:04
	← xxthink left	10:04
	→ anderse joined	10:05
	← dreixel left	10:05
	← _br_ left	10:05
	→ nigelb joined	10:06
	← danielfilho left	10:07
	→ daniel_- joined	10:07
	→ marhaban joined	10:08
	→ jmd joined	10:08
	← marhaban left	10:08
jmd	how can I do git <something> on all files whose names match a pattern?	10:09
	→ pgokeeffe joined	10:09
	→ _br_ joined	10:09
	→ marhaban joined	10:10
jast	unless you want to use shell script, that would depend on the <something>	10:10
	← BiggFREE left	10:11
	→ kristofers joined	10:11
	← Guest36822 left	10:12
	← Xeno left	10:13
	← anderse left	10:13
	daenglinuxx → terlalu	10:13
	→ anderse joined	10:14
	→ wasanzy joined	10:14
	→ Goles joined	10:15
	→ suy joined	10:15
	→ arkiver joined	10:15
	← _br_ left	10:16
	→ _br_ joined	10:18
	← swair left	10:20
	→ beautifulmind joined	10:21
	← FernandoBasso left	10:21
	→ exoon joined	10:21
	← iflista_ left	10:21
	← _br_ left	10:22
	← aberrant left	10:22
	← anderse left	10:22
	→ _br_ joined	10:23
	→ mklappstuhl joined	10:24
Chiku\|dc	does env GIT_PROXY_COMMAND just for git:// ? what about https:// ?	10:25
	SeoZ-work[AWAY] → SeoZ	10:25
	→ clone1018 joined	10:25
	→ angelsl joined	10:26
jast	I don't know GIT_PROXY_COMMAND, but chances are http(s) uses the standard http(s)_proxy env var	10:26
	← ISF left	10:28
	terlalu → blackjack	10:29
	→ e66 joined	10:29
Chiku\|dc	I use this for SOCKS	10:31
	← Fels left	10:31
	dwmw2_gone → dwmw2	10:32
	← luc_ left	10:32
	← herdingcat left	10:33
	→ shruggar joined	10:33
	← Fandekasp left	10:33
	← guampa left	10:36
	→ Hounddog joined	10:37
	→ Element9 joined	10:37
	← rastar left	10:37
	→ gain joined	10:39
	← babilen left	10:39
	← stliu left	10:39
	→ Goles_ joined	10:39
	← Orphis left	10:40
	→ Fels joined	10:40
	← Goles left	10:40
	← maleknet left	10:42
	← rootcoma left	10:42
	→ Orphis joined	10:43
	→ txomon\|home joined	10:44
	← asakura left	10:46
	← tystr left	10:49
	→ psoo joined	10:49
	← AdrienBrault left	10:50
	→ ankurankan joined	10:50
	→ wrouesnel1 joined	10:50
	← djsecilla left	10:51
	← jdyin left	10:51
	→ AdrienBrault joined	10:52
	→ asakura joined	10:52
	→ tystr joined	10:52
	→ pity joined	10:52
	← m0viefreak left	10:55
	← Xethron left	10:56
	→ avjpl joined	10:56
	→ Xethron joined	10:56
	← buibex left	10:56
	← avjpl left	10:57
	← txomon\|home left	10:57
	→ h0bbit joined	10:58
	→ txomon\|home joined	10:59
	→ gyre007 joined	10:59
	← alkino left	10:59
	← fisted left	11:01
	→ Sebo joined	11:01
	← jargon left	11:01
	← Ksg89 left	11:02
	→ fisted joined	11:02
	← joeytwiddle left	11:03
	→ maleknet joined	11:03
	→ rastar joined	11:04
	← yshh_ left	11:05
	→ buibex joined	11:05
	← jatt left	11:07
	→ joeytwiddle joined	11:07
	→ babilen joined	11:08
	← ostree left	11:09
	← Tommy[D] left	11:09
	← asakura left	11:09
	→ FernandoBasso joined	11:09
	→ Tommy[D] joined	11:10
	→ bgerber joined	11:10
	→ eletuchy joined	11:11
	→ eletuchy_ joined	11:12
	← eletuchy left	11:12
	→ alkino joined	11:13
	← smirn0v left	11:14
	← maleknet left	11:15
	→ zerba joined	11:15
Nevik	osse: Trying SHA 00000000000000000000000000000000000f4241; covered 0.0000000000000000000000000000000000000001% of search space; step: 0.540s; ETA: 25025712967994912512904150712367479521280.000000 years	11:16
	→ smirn0v joined	11:16
	← hyperair left	11:16
	→ iSaleK joined	11:16
Nevik	osse: i dont think we're gonna see the completion of that algorithm :P	11:16
	← fphilipe left	11:17
	← Infiltrator left	11:18
Nevik	oh that wasnt quite correct	11:18
	osse: it's only ~2.5e34 years	11:19
	← Merwin left	11:20
	→ the_cybersphinx joined	11:20
	← chisler left	11:20
	→ Merwin joined	11:20
	← wrouesnel1 left	11:21
	→ asakura joined	11:21
	→ Fandekasp joined	11:21
	→ Progster joined	11:22
	→ milStaja joined	11:23
	← FabTG left	11:23
	← lethjakman left	11:24
	← dermoth\|home left	11:25
	→ aberrant joined	11:25
	→ Infiltrator joined	11:26
	→ dermoth\|home joined	11:26
	→ anderse joined	11:26
	→ shwaiil joined	11:27
shwaiil	hi	11:27
	← Xethron left	11:27
	← JohnBat26 left	11:27
	→ maleknet joined	11:28
shwaiil	Q: I had to git checkout some files to an older version. Wondering if I could do a git reset HEAD to make sure everything is sync with master. I've tried reset head but didn't worked ?! :T	11:29
	Thanks for looking!	11:29
	→ bmac_ joined	11:29
	← bauruine left	11:29
	← marhaban left	11:30
	→ warthog9 joined	11:30
	→ bauruine joined	11:30
	→ apeiros_ joined	11:30
	→ quazimodo joined	11:31
apeiros_	hi there. `git diff --summary A..B` only shows created, deleted and mode-changed files - is there a way to also get modified files?	11:32
	(git version is 1.8.1)	11:32
shwaiil	btw, I did git reset HEAD --hard	11:33
	erkules_ → erkules	11:33
whowantstolivefo	https://github.com/refinery/refinerycms/pull/2176#issuecomment-13225888 <<< what can i do ? for right put my files to git ?	11:33
	← smirn0v left	11:34
	→ gavin_huang joined	11:35
	← sallesao left	11:35
	← avtobiff left	11:35
sveajobb	shwaiil, then the changes should have been removed, are you sure you didn't change where HEAD i.e. checked out another commit instead	11:36
	→ smirn0v joined	11:36
	← Sonderblade left	11:36
	→ rastar1 joined	11:36
shwaiil	sveajobb: tks for looking! I'm making sure I've merged the branches I want to see, and cleaning the cache, because I think everything was done properly :Z just a sec	11:37
	← rastar left	11:37
	← zoresvit left	11:38
	← gavin_huang left	11:38
sveajobb	shwaiil, well if you did a merge, then you actually moved the HEAD to a new commit which is the merge, try to see this visualized in gitk for example. in order to remove the changes you can do git reset --hard THE_OLD_BRANCH_OR_REF	11:38
	→ gavin_huang joined	11:39
	← ajc_ left	11:40
	← gavin_huang left	11:40
	→ curvedmark joined	11:40
	→ osa1 joined	11:40
shwaiil	sveajobb: In local I merged several branches to master. in a different server I pulled master (everything fine). there was two files that I had to revert back, so I did git checkout hash path/file. Everything was up to date, but not this files I reverted.	11:40
	I had to merge more local changes to the master. In the other server, I've pulled, but it's not like in the local master, the files I've reverted previously remain in that state.	11:41
	← mattcen left	11:42
	→ mattcen joined	11:42
sveajobb	and the changes you made, are they commited?	11:43
	what does git status say on the servers ? do they differ?	11:43
shwaiil	sveajobb: nice ehh just a sec	11:43
	→ JohnBat26 joined	11:43
shwaiil	only some untracked files are present	11:44
	← adamben left	11:44
shwaiil	I'l just do it manually it's fine I guess, just wondering if git reset HEAD --hard should have worked	11:44
	← osa1 left	11:44
	← looper left	11:44
sveajobb	yeah well, if nothing shows up in git status, then git reset --hard won't really do anything	11:45
	→ roland joined	11:45
sveajobb	buy yeah, if you checkout an old version of a file and later on git reset --hard, that should work	11:45
	→ engla joined	11:45
shwaiil	ok thanks : )	11:46
	← nafg left	11:47
	→ robotmay joined	11:48
	→ marhaban joined	11:48
	← marhaban left	11:48
	→ marhaban joined	11:49
	← ankurankan left	11:49
	← lirt left	11:49
	→ Sonderblade joined	11:49
	← lxsameer left	11:50
	→ lirt joined	11:50
	← marhaban left	11:53
shwaiil	sveajobb: I did git checkout hash path/file(s) and its fine now	11:54
	← main left	11:54
	→ rwestphal joined	11:55
	→ zoresvit joined	11:55
	← weltall left	11:56
	→ huoxito joined	11:56
	→ weltall joined	11:58
sveajobb	shwaiil, nice :)	12:02
	→ untrothy joined	12:02
	← jalama left	12:04
	← horatio_cromwell left	12:05
	SeoZ → SeoZ-work[AWAY]	12:07
	← alkino left	12:07
	← Error404NotFound left	12:08
	→ gusnan joined	12:12
	← giallu left	12:12
	→ Asenar joined	12:12
	→ fphilipe joined	12:14
wrouesnel	ok running git mergetool with meld gives me local, base and remote files --- which one do I have to actually merge into ?	12:15
	← Putr[alt] left	12:15
	← aberrant left	12:17
	→ Error404NotFound joined	12:21
	→ alkino joined	12:21
	→ ostree joined	12:22
	← amigojapan left	12:22
	→ csmrfx joined	12:22
	→ adamben joined	12:23
	← rwestphal left	12:23
csmrfx	Is there a flag for "dry-run" with git pull?	12:23
	← untrothy left	12:23
	→ KevinSjoberg joined	12:23
csmrfx	ie. not really doing anything, just to see what might change	12:23
	→ rwestphal joined	12:23
	← pity left	12:28
	← lirt left	12:28
	← BadDesign left	12:28
	← smirn0v left	12:29
	← valgrind left	12:29
	← bmac_ left	12:29
	→ lirt joined	12:29
	← francisfish left	12:30
	← keller left	12:30
	→ smirn0v joined	12:30
	→ STenyaK2 joined	12:30
	← Error404NotFound left	12:31
	→ megha joined	12:32
bremner	csmrfx: you can use git fetch or git remote update, but predicting the outcome of merges is harder in most casees.	12:32
	gitinfo set mode: +v	12:32
STenyaK2	How can I view a summary of all changes between two specific commits, in similar format as a "git status" does for changes in working copy? Thanks!	12:32
bremner	STenyaK2: man git diff	12:32
gitinfo	STenyaK2: the git-diff manpage is available at http://jk.gs/git-diff.html	12:32
	→ jaseemabid joined	12:33
	← acrocity left	12:36
	→ keller joined	12:38
	← fuzzybunny69y left	12:38
	→ yshh joined	12:39
	→ acrocity joined	12:41
	← fphilipe left	12:41
	← OOPMan left	12:42
	→ Ksg89 joined	12:42
	← Fels left	12:43
	→ xiangfu joined	12:43
	→ Error404NotFound joined	12:44
	→ fphilipe joined	12:45
	← warthog9 left	12:45
	→ Tommy[D]_ joined	12:48
	← Tommy[D] left	12:48
	← fphilipe left	12:48
	→ picoguy joined	12:49
	← n4irc left	12:50
	← Goles_ left	12:51
	→ marc_kdab_ joined	12:51
	→ Goles joined	12:51
	→ aldskf3 joined	12:51
	← Targen left	12:53
	→ Noldorin joined	12:53
cheater_	hi	12:54
	gitinfo set mode: +v	12:55
	← alkino left	12:55
cheater_	cmn: hi, are you around?	12:56
Goles	test	12:56
	Oh I have voice now :)	12:56
cheater_	well done	12:56
Goles	I was wondering how can I download a submodule for my git project … I added a repo as submodule, commited to my master branch and pushed. If I then clone my repo, the submodule dir is (of course) empty. How can I fetch the submodule?	12:56
	→ guampa joined	12:56
	→ blue-0x joined	12:56
	→ jcao219 joined	12:56
	← buibex left	12:56
	→ dhruvasa1ar joined	12:56
Nevik	Goles: see http://git-scm.com/book/en/Git-Tools-Submodules	12:57
	Goles: and after that, man git submodule for the technical details	12:57
gitinfo	Goles: the git-submodule manpage is available at http://jk.gs/git-submodule.html	12:57
	→ cannonball joined	12:58
Goles	thanks guys	12:58
	Nevik that's blog is neat	12:58
Nevik	Goles: that's the Pro Git book	12:58
	it's generally a recommended read for all git users	12:59
Goles	Nevik cool! I've been exploring git more these days… have been using it in "basic" mode for a couple of years… just discovered 'rebase'	12:59
	→ thyseus_ joined	12:59
Nevik	:)	12:59
	→ samuel02 joined	12:59
	← thyseus left	13:01
	← mihamina left	13:01
	→ kill-9_ joined	13:02
	← jcao219 left	13:03
	← Progster left	13:03
	→ ncd\|leeN joined	13:04
	← yshh left	13:04
	← smirn0v left	13:04
	← Cromulent left	13:04
	→ looper joined	13:04
	← leeN left	13:04
	← Noldorin left	13:05
	→ Mannequin joined	13:05
	← daniel_- left	13:05
	← NiteRain left	13:05
	← adamben left	13:06
	→ jcromartie joined	13:06
	→ m0ltar joined	13:06
	→ smirn0v joined	13:07
	→ alkino joined	13:08
	← sgambino left	13:10
	← keller left	13:10
	← txomon\|home left	13:10
	→ adamben joined	13:11
cheater_	hey guys, i did some history rewriting on a page. first i tried filter-branch with --tree-filter, then (on a separate copy) with --index-filter, and i ran git reflog to expire caches and i ran gc --aggressive.. the files were removed from the checkouts, but "git verify-pack -v .git/objects/pack/pack-*.idx \| grep -v chain" still shows those files. why does it show them? i am afraid that the files didn't get "really removed"	13:11
	toothr → toothrot	13:13
	→ Shurakai joined	13:13
	→ geekbri joined	13:13
	← Ksg89 left	13:14
	← eliasp left	13:14
	→ millun joined	13:14
jast	cheater_: there's a checklist for shrinking repositories in man git-filter-branch... (by far the easiest way to get rid of them is to clone the filtered repo into a new one)	13:14
gitinfo	cheater_: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	13:14
whowantstolivefo	hey guys anyone online!?!>?	13:14
	← blue-0x left	13:15
whowantstolivefo	when i type " git rebase -i HEAD^^ " i get the following error " /usr/lib/git-core/git-rebase: 1: eval: sublime1: not found Could not execute editor " but i already typed " git config --global core.editor "sublime1 --wait" " it dosnt help	13:16
	→ guy joined	13:16
	← jcromartie left	13:16
guy	These are my firewall rules, though I am not able to establish connection with Git over SSH.	13:16
	https://gist.github.com/gajus/de69403b579fb3ca0a63	13:16
	any clue what is missing?	13:16
	← adamben left	13:17
jast	whowantstolivefo: it's complaining that sublime1 doesn't exist or can't be found, so if you configure it to use that, it's not a huge surprise that it fails. you may need to give it a full path to the editor (if it's installed).	13:18
whowantstolivefo	just i made it as my first text editor. when i do in console " sublime1 . " and it is opening all file folder as IDE	13:18
	← arkiver left	13:18
whowantstolivefo	i try to configure with that command >>>>>>>>>>>> git config --global core.editor "sublime1 --wait" <<<<<<<<<<<<<<	13:19
jast	try to give it the full path to the editor	13:19
	→ juvenal joined	13:19
jast	guy: when you try it, what happens? timeout?	13:20
	→ roger_padactor joined	13:20
guy	Yes	13:20
	ssh: connect to host github.com port 22: Connection timed outfatal: The remote end hung up unexpectedly	13:20
	→ adamben joined	13:20
jast	iptables -L -v might help you see where the packets get stuck	13:20
roger_padactor	slightly confused by git lingo… I have a developer branch that is currently checkout… I'm using sourceTree app. when I want to merge dev into master I'm confused because it says merge master into developer which doesn't sound right.	13:21
jast	oh hey, you didn't include port 22 in the "outgoing" rules	13:21
	roger_padactor: you can only merge into the checked out branch	13:22
roger_padactor	ok thanks	13:22
guy	Thank you jast	13:23
	← guy left	13:24
	→ jsoa joined	13:25
	→ tmcmahon joined	13:25
	→ prudnikov joined	13:25
	← Exio left	13:25
	← alunduil left	13:27
	→ sint joined	13:27
	← KevinSjoberg left	13:27
	← h0bbit left	13:30
	→ OOPMan joined	13:30
	→ thesisus joined	13:30
	← dhruvasagar left	13:31
	← dhruvasa1ar left	13:31
	→ LORDs_diakonos joined	13:32
	→ jxpx777 joined	13:32
	→ heroux joined	13:33
	→ amir joined	13:33
	→ kurkale6ka joined	13:33
	← rzs left	13:33
kurkale6ka	Hi, when using git log can I grep for a word appearing inside the patches?	13:34
	← juvenal left	13:35
charon	kukks: -G	13:35
	→ netzmonster joined	13:35
	→ chaw joined	13:35
charon	or -S if you know that the word was introduced or removed (-G is much slower than -S)	13:36
kurkale6ka	charon: cheers	13:36
	→ fphilipe joined	13:37
	→ Exio joined	13:37
	← blackmat` left	13:38
chaw	hey.. good day/night.. I have a Q regarding using Git + SVN	13:39
	My coworker set up the initial Git Repo (hosted on github, private) from our internal SVN server	13:39
	→ chisler joined	13:39
chaw	so he set up the initial git repo from SVN.. and has all of that set up on his machine	13:39
	meaning if he dcommits, it will commit to SVN	13:40
	→ buibex joined	13:40
chaw	and if he rebases it will pull from SVN to the GH Repo (which he can then sync upstream to GitHub)	13:40
	I'm trying to set up the same for myself here.. and it seems I find like 1000 different ways of going about it.. and all of them seem messy	13:40
	← tmcmahon left	13:41
	→ decotii joined	13:41
	← xiangfu left	13:42
chaw	I did find one guide, which looks promising.. my only issue is that I am not setting this stuff up from scratch	13:43
	→ glennpratt joined	13:43
	← pretty_function left	13:43
	→ juvenal joined	13:45
chaw	I guess I'll give it another try and report back with specific issues :)	13:47
	→ sajan joined	13:47
	→ giallu joined	13:48
	→ Dreamer3 joined	13:49
	← subhojit777 left	13:49
	→ itatitat joined	13:50
	← curvedmark left	13:50
	→ cheater joined	13:50
	← itatitat left	13:50
	→ sgambino joined	13:51
	← prudnikov left	13:52
	→ xiangfu joined	13:52
	← cheater_ left	13:53
	→ freeed joined	13:53
	→ HumanSky joined	13:54
	← lifelike left	13:54
	→ lifelike joined	13:55
	← lifelike left	13:56
	→ Targen joined	13:56
	→ lifelike joined	13:57
	← Hounddog left	13:57
	→ Hounddog joined	13:57
	← lifelike left	13:58
	→ sixcorners joined	13:58
	→ prudnikov joined	13:58
	→ lifelike joined	13:58
	← wolog left	13:59
	← lifelike left	13:59
	→ normanrichards joined	13:59
	← megha left	13:59
	→ lifelike joined	13:59
	→ arkiver joined	14:00
	← georg left	14:00
	← hoschi left	14:00
	→ megha joined	14:01
	← lng left	14:01
	→ erichynds joined	14:01
	← relipse left	14:01
	→ __BigO__ joined	14:02
	← normanrichards left	14:03
	→ x4rlos joined	14:04
	→ AaronMT joined	14:04
	← beautifulmind left	14:04
	→ fphilipe_ joined	14:05
	← fphilipe_ left	14:05
	← fphilipe left	14:05
	← LORDs_diakonos left	14:06
x4rlos	Hi all. I am looking to set up a git repository for internal use for our software. I have checked online for methods of creating a repository, and it seems people use all different methods. Do i go along the git-daemon route? or just the git-init --bare route? Gitweb is mentioned a few times, but i think that the commandline is enough.	14:06
	→ RaceCondition joined	14:06
	→ fphilipe joined	14:06
x4rlos	Anyone got any practical advice?	14:06
	→ trungqtran joined	14:06
	→ wolog joined	14:06
	← blackjack left	14:07
	← thesisus left	14:08
cheater	jast: thanks, looking	14:08
	adhawkins → adhawkins-away	14:09
	← Sebo left	14:09
	→ breakingthings joined	14:09
	→ Sebo joined	14:09
	← xxc_michael left	14:10
	→ rideh joined	14:11
	← lirt left	14:11
paulweb515	x4rlos: how will the repo be shared by your team? Does everybody have a userid on the hosting server?	14:11
	← netzmonster left	14:12
	→ lirt joined	14:12
	→ siriuz joined	14:12
x4rlos	I could certainly make it that way if this changes things.	14:13
	→ thatguycraig joined	14:13
	→ rcu joined	14:13
	← alkino left	14:14
cheater	jast: hmm ok	14:14
	jast: i must be doing something wrong...	14:14
paulweb515	x4rlos: you need to do something to allow your team to write to the shared repo.	14:14
	→ koderde joined	14:14
	→ scampbell joined	14:14
	→ yshh joined	14:14
	→ Zombie__ joined	14:15
	← Targen left	14:15
paulweb515	x4rlos: some projects use one of the git* daemons to do it. Our project uses SSH and userids to manage it, as well as Gerrit (a review system that manages the userids, and only gerrit writes to the repo)	14:15
x4rlos	paulweb515: then yes :-) Sorry, i interpreted the way you was asking like there was an option :-) Yes, i can create user account (with or without login ability)	14:15
paulweb515	x4rlos: it just depends on which methods is 1) easiest to set up, 2) complies with whatever security guidelines your company or orginization supports and 3) meets your team's workflow	14:16
	← scorphus left	14:16
	→ kevwil joined	14:17
	→ macabre joined	14:17
	→ Chappo joined	14:17
	← milStaja left	14:17
	← maleknet left	14:18
x4rlos	Gerrit i have come across before, and it is pretty good (from a user perspective). Security-wise isn't a large concern as it will be internal only (backup & version control) and workflow, it's mainly 1 programmer and a couple of people who will help out from time to time.	14:20
	← yshh left	14:20
	→ Noldorin joined	14:21
x4rlos	So given this, i guess i'm looking at the git init --bare, setting up the accounts, and then integrating gerrit. That simple? :-)_	14:21
	→ Jackneill joined	14:24
	← Jackneill left	14:24
	→ Jackneill joined	14:24
	→ mmitchell joined	14:24
	→ Targen joined	14:24
	← dharrigan left	14:25
paulweb515	x4rlos: that will give you a decent setup, especially for part time contributors	14:26
	→ dharrigan joined	14:26
	→ farciarz84 joined	14:26
	← dharrigan left	14:26
	→ alunduil joined	14:26
x4rlos	cool. I shall have a look. What ports does git need open to operate? Given its over ssh 22 is obvious one, as is gerrit on 80 - but any others for uploading? Or all done over the ssh protocol?	14:26
	→ AnimeCYC joined	14:27
farciarz84	hi, I'm new to git. I've done commit which was bad, how to restore to previous version?	14:27
	→ alkino joined	14:27
	→ dharrigan joined	14:27
	← dharrigan left	14:27
	→ dharrigan joined	14:27
csmrfx	farciarz84: checkout the previous commit?	14:28
	→ yshh joined	14:28
csmrfx	http://stackoverflow.com/questions/927358/undo-last-git-commit	14:28
farciarz84	csmrfx: what are the commits names?	14:28
csmrfx	farciarz84: in git, each commit is given a hash	14:29
jast	cheater: to be fair, it's pretty easy to get wrong	14:29
	← jackyyll left	14:29
csmrfx	farciarz84: you can see the list with git log	14:29
farciarz84	csmrfx: ok I got this hash,	14:30
csmrfx	farciarz84: You really want to go through git basics. Search google for: git visual reference	14:30
	← dharrigan left	14:30
	← lamarus left	14:31
	→ dharrigan joined	14:31
	→ jperry2 joined	14:31
	→ R1KRDO joined	14:31
csmrfx	or, youtube for introduction to git - or git power workflow	14:32
	farciarz84: ^	14:32
	← FernandoBasso left	14:32
farciarz84	ok so in my instance I need to git checkout hash, where hash one before last hash in log hist	14:33
	→ security joined	14:33
farciarz84	csmrfx: am I right?	14:34
	← megha left	14:34
csmrfx	http://stackoverflow.com/questions/2007662/rollback-to-an-old-commit-using-git	14:34
	→ tvw joined	14:34
csmrfx	farciarz84: there are several ways to your end...	14:34
	→ amigojapan joined	14:34
	→ NiteRain joined	14:34
	← tvw left	14:36
	→ tvw joined	14:36
	→ thl0 joined	14:37
	← thl0 left	14:37
	← tvw left	14:37
	→ rabellamy joined	14:37
	→ thl0 joined	14:37
	← RaceCondition left	14:38
	→ RaceCondition joined	14:39
farciarz84	csmrfx: it works, you've helped me a lot thx	14:39
	→ danielfilho joined	14:39
	→ TeckniX joined	14:39
	← rabellamy left	14:40
	→ tvw joined	14:41
csmrfx	farciarz84: gl!	14:41
	← NiteRain left	14:41
paulweb515	x4rlos: if you use SSH, then you just use the defaults. If you use Gerrit, it supports SSH and https on a higher port, ex, ours is configured: ssh://git.eclipse.org:29418/platform/eclipse.platform.ua. They'll discuss configuration and ports in the Gerrit docs, I'm sure	14:42
	→ dhruvasagar joined	14:42
	← thl0 left	14:43
	→ dhruvasa1ar joined	14:43
	→ thl0 joined	14:43
	← dsantiago left	14:43
	← tvw left	14:44
	→ tvw joined	14:44
	→ ojacobson joined	14:45
	→ Guest86070 joined	14:45
	← Guest86070 left	14:46
	→ dsantiago joined	14:46
	← lb left	14:47
	← adamben left	14:47
	← yshh left	14:47
x4rlos	paulweb515: cool, thanks. You been a good help :-)	14:47
	→ jkeiper joined	14:48
	→ axl_ joined	14:48
paulweb515	x4rlos: we're just digging ourselves out from our own transition to git now :-)	14:48
x4rlos	good luck :-)	14:48
jkeiper	hey folks. i merged my fork with upstream master but ended up with a bunch of baloney in a resulting pull request. i know rebasing is the answer, but can someone tell me how to rebase without the ridiculous pile of files in the previous merge? will they just go away?	14:49
	→ aberrant joined	14:49
	→ cmmagid joined	14:49
	→ TheJH joined	14:49
	← HumanSky left	14:50
	→ Guest98299 joined	14:51
	→ swair joined	14:51
	← swair left	14:51
	← sint left	14:52
	→ hinst joined	14:53
	→ v0n joined	14:54
	← thl0 left	14:55
	→ Ksg89 joined	14:56
	← Chappo left	14:57
	← breck left	14:57
	→ natechan joined	14:57
	← soee left	14:57
whowantstolivefo	when i type " git rebase -i HEAD^^ " i get the following error " /usr/lib/git-core/git-rebase: 1: eval: sublime1: not found Could not execute editor " but i already typed " git config --global core.editor "sublime1 --wait" " it dosnt help i try to configure with that command >>>>>>>>>>>> git config --global core.editor "sublime1 --wait" <<<<<<<<<<<<<<	14:57
	→ breck joined	14:57
jast	whowantstolivefo: so you ignored my previous suggestion. good to know.	14:57
	← marthinal left	14:57
whowantstolivefo	jast: i did how u say also	14:58
	→ marthinal joined	14:58
whowantstolivefo	jast: https://gist.github.com/whowantstolivefo/4731432	14:58
	← jakob left	14:59
	← thehandler left	14:59
jast	are you sure that's the full path to the executable? it looks more like a directory	15:00
	→ asker joined	15:00
	→ banister`gone joined	15:00
	← banister`gone left	15:00
	← axl_ left	15:00
	→ banisterfiend joined	15:00
banisterfiend	hi what does: git cherry-pick A..B do?	15:00
whowantstolivefo	jast: http://www.technoreply.com/how-to-install-sublime-text-2-on-ubuntu-12-04-unity/ <<< i installed sublimetext2 as it says here	15:01
banisterfiend	it doesnt appear to be cherry-picking that rnage of commits	15:01
	gitinfo set mode: +v	15:01
asker	hey guys, im new to git and im trying to save my login credential	15:01
	git config --global credential.helper cache <-- done	15:01
	← rastar1 left	15:01
	← Guest98299 left	15:01
asker	it shows me that credential-cache is not a git command, what im doing wrong?	15:02
	→ JDuke128 joined	15:02
	← m1sc left	15:02
	← eletuchy_ left	15:02
	→ eletuchy joined	15:03
	→ normanrichards joined	15:03
	→ caitp joined	15:03
grawity	asker: if you're using msysgit, the "cache" helper hasn't been ported to Windows yet	15:04
asker	ya i do	15:04
	so i installed winstore and tried it with winstore	15:05
	→ Bass10 joined	15:05
asker	the same error	15:05
grawity	Where did you install the winstore helper?	15:05
asker	just run the exe and it asked me whether i want to install it, i submited	15:05
	→ txomon\|home joined	15:06
asker	where does it install it in? where the setup exe is?	15:06
	→ srcerer_ joined	15:06
	→ lb_ joined	15:07
	→ jakob joined	15:07
	→ fr33domlover joined	15:07
fr33domlover	Question to GitHub users: TRUE or FALSE: for a free software activist/supporter/enthusiast, using github would count as hypocricy because the github software isn't free (unlike the gitorius software, for example)	15:07
	← dsantiago left	15:08
	← danielfilho left	15:08
asker	grawity: you lead me to the right trace, i had to install it in the git/cmd dir, now it works fine, ty	15:08
	→ axl_ joined	15:08
	← srcerer left	15:08
	→ danielfilho joined	15:08
	→ m1sc joined	15:08
	→ ISF joined	15:08
grawity	fr33domlover: FileNotFound	15:08
	← alunduil left	15:08
Bombe	FILE_NOT_FOUND, please.	15:08
	Also, the question totally sucks.	15:09
fr33domlover	I'm not criticizing	15:09
	just want to understand	15:09
Bombe	No, you don’t. Please go away.	15:09
	→ ciphergoth joined	15:10
caitp	anyone used sed to replace a git commit sha in a text file?	15:10
ciphergoth	A Google search for "git meld" finds lots of ways of using Git with GNU Meld, some of them pretty old. What's the current best way to do it? Is it this? https://github.com/wmanley/git-meld	15:10
caitp	i think [a-z0-9]\{40\} should be good enough, but something I'm doing is wrong .v.	15:10
fr33domlover	@Bombe someobe must have a better answer	15:11
	*someone	15:11
Bombe	caitp, what are you doing, what happens, what did you expect?	15:11
	fr33domlover, why? The question is outright bad.	15:11
caitp	I'm writing an update shell script for a 3rd party library in mozilla, which should update the README file to reference the correct commit id	15:11
	security → megha	15:12
caitp	so, nothing happens because the search in my sed expression is not matched (which I think it should be)	15:12
	← breck left	15:12
	← prudnikov left	15:12
	← jonshea left	15:12
Bombe	caitp, test your regular expression with grep -c, it helps you narrow down why it doesn’t work.	15:13
ciphergoth	Or, what's your favourite tool for viewing diffs under Unix in git?	15:13
fr33domlover	@Bombe the question "is it okay to kill people is the name of peace" has a lot to think about.. so it my question about github	15:13
whowantstolivefo	jast: i just went to my ~.gitconfig file and change sublimetext with the nano and it is work now	15:13
Bombe	fr33domlover, no. Your question mixes up totally different groups of people and then ignores the fact that essentially all hardware is non-free.	15:14
caitp	https://gist.github.com/caitp/3532692a8cbc3f705777 it looks like so, currently, but lets see	15:14
	oh right, I keep forgetting gist doesn't do GFM	15:15
	derp	15:16
ciphergoth	Wow, normally it's really easy to get opinions on IRC. It looks like it's easier here to get an opinion on whether one should kill for peace than how to diff with git :)	15:16
	ChanServ set mode: +o	15:16
caitp	git diff oldcommit newcomment <specific files if desired>?	15:16
	what are you trying to do?	15:16
ciphergoth	caitp: I was hoping to get a GUI diff. As I say, Google gives me lots of answers, but I was hoping to get opinions on what's currently preferred	15:17
	under Unix	15:17
	EugeneKay changed the topic to: Latest version: 1.8.1.2 \| http://git-scm.com \| Help with git and #git: http://jk.gs/git/ \| git-chainsaw: just remember not to remove the branch you're on \| Getting "Cannot send to channel" or unable to change nick? /msg gitinfo .voice	15:17
	→ BadDesign joined	15:17
	← BadDesign left	15:17
	→ BadDesign joined	15:17
	EugeneKay set mode: +v	15:17
	EugeneKay set mode: -o	15:17
caitp	oh, not sure	15:17
	bugzilla and github are the diff uis i tend to use	15:17
fr33domlover	@Bombe non-free hardware don't justify anything	15:17
	→ mikecmpbll joined	15:18
Bombe	fr33domlover, but non-free software does? How so?	15:18
ciphergoth	caitp OK thanks	15:18
	→ acuzio joined	15:18
	← shashin left	15:18
	→ thl0 joined	15:18
fr33domlover	@Bombe I didn't say it does, I don't use github. I'm neutral, just asking an interesting question	15:18
	→ HumanSky joined	15:19
	→ jcp_ joined	15:19
jcp_	i did commit . without thinking and need to uncommit two specific files. is it just git checkout HEAD -- file	15:20
	→ prudnikov joined	15:21
	→ machty joined	15:21
_ikke_	jcp_: commit . or add .?	15:21
machty	is there something fundamentally different between github/bitbucket where bitbucket doesn't automatically fetch remote branches for you?	15:22
jcp_	commit .	15:22
	i added the specific files i wanted to commit then did a commit .	15:22
_ikke_	jcp_: And what is it that you want to achieve?	15:22
Bombe	fr33domlover, that question is not interesting.	15:22
jcp_	_ikke_: i want to uncommit two files that i didn't mean to commit	15:23
joephilly	machty, auto fetch remote branches? Why would it do that?	15:23
	← rcu left	15:23
machty	joephilly: maybe i'm being a foolish clown but i feel like when i do a git fetch on a github repo i see all the diffferent little branches that have been created since my last fetch	15:24
	not the case for bitbucket, but i could be confused	15:24
	gitinfo set mode: +v	15:24
protagoras271	@ciphergoth: use tig	15:24
joephilly	machty, are you doing "git fetch origin" or specifying a branch?	15:24
	→ todd_dsm joined	15:25
	→ shashin joined	15:25
	← psoo left	15:25
ciphergoth	protagoras271, this tig? http://jonas.nitro.dk/tig/	15:25
	← Sonderblade left	15:25
	← banisterfiend left	15:25
ciphergoth	I was hoping for something more GUIish, like meld	15:25
	or kdiff3	15:25
protagoras271	@ciphergoth: thats the one, Its useful when you need something a bit more powerful than the native commands. If you are looking for things less powerful and moar pretty check out gitk, qgit, gitg, and git-gui	15:27
	I think git also ships with a native gui	15:27
jast	fr33domlover: I'll go with false	15:27
	→ alunduil joined	15:28
ciphergoth	protagoras271, I've got gitg, I quite like it, but the diff viewer isn't all that compared to something like GNU meld http://upload.wikimedia.org/wikipedia/commons/3/30/Meld_file1.png	15:29
protagoras271	@ciphergoth: I do like how the sides flow into eachother... I don't think I have seen a git gui client that does that.	15:30
_ikke_	jcp_: Easiest way is to do git reset --soft HEAD^, git reset -- file1 file2; git commit	15:31
ciphergoth	protagoras271, I don't think it's the job of the git client. Instead, git can defer to an external merge tool. So eg git-meld makes meld work with git https://github.com/wmanley/git-meld	15:31
	← thl0 left	15:31
ciphergoth	But since the integration looks a bit complex, I wanted to check whether eg the latest versions of git make git-meld unnecessary	15:31
	← axl_ left	15:33
	→ thl0 joined	15:33
	← Mannequin left	15:33
acuzio	ladies.. - and gents : I have a branch that contains some changes and now i move back to the "master" ., when doing so , git asks me to specifically commit the change or risk losing it , why does this happen ? I thought you could simply go from one branch to the other and git would maintain "state"	15:34
	→ axl_ joined	15:34
caitp	if it's not staged, and you checkout something else, it can be lost	15:34
	you can stage it without commiting	15:34
acuzio	caitp: hmm , interesting - i am assuming i can also stash it then	15:35
caitp	stash is what i meant to say actually	15:35
	i haven't had my coffee yet ;p	15:35
acuzio	caitp: :-) - i haven;t seen this before - is this new ?	15:35
caitp	it was new at some point	15:35
acuzio	caitp: thanks - i am going to go and check what version that happened - i had an old version of git and don;t remember seeing this feature -	15:36
	→ skul joined	15:36
	→ dsantiago joined	15:40
jcp_	_ikke_: thanks	15:40
	← fr33domlover left	15:40
	→ Xeno joined	15:41
	→ hellopat joined	15:41
	← hellopat left	15:41
	→ Goplat joined	15:41
ciphergoth	top tip: type "git mergetool"	15:41
	→ hellopat joined	15:42
	→ artagnon joined	15:42
artagnon	How do I configure a repository to always push to remote "mine", but merge from "origin" when pulling? This is a very common fork case.	15:42
	I can only see branch.<branchname>.remote, which is for both pushing and pulling.	15:43
	← thl0 left	15:44
	← shashin left	15:44
joephilly	artagnon, I think its a default setting	15:44
	← BadDesign left	15:45
PerlJam	artagnon: Just always specify the remote when you push. "git push mine"	15:45
	← jcp_ left	15:45
	→ chuckharmston joined	15:45
artagnon	Ugh. I can't believe such a simple configuration option doesn't exist.	15:45
	joephilly: Pardon?	15:45
	→ thiago joined	15:46
lov	artagnon: you can set the configuration for this, though I don't recall how offhand.	15:46
	(at least I think you can!)	15:46
	← dhruvasa1ar left	15:46
artagnon	Could you look it up? I've been going through the manpage and trying to figure this out for sometime now.	15:46
lov	artagnon: I'm pretty new to git, so I'm not the person to ask :<	15:47
	→ jaequery joined	15:47
_ikke_	artagnon: You can set a different fetch and push url for a remote	15:47
EugeneKay	artagnon - man git-config; branch.<name>.merge, in particular the last sentence.	15:47
gitinfo	artagnon: the git-config manpage is available at http://jk.gs/git-config.html	15:47
	← mklappstuhl left	15:48
PerlJam	artagnon: git config remote.origin.pushurl <URL> # maybe that'll work for you?	15:48
	← lirt left	15:48
	← Noldorin left	15:48
	← dhruvasagar left	15:48
PerlJam	artagnon: same remote, but pushes and pulls are to different repos	15:48
EugeneKay	That's another ay to do it	15:48
	→ jackyyll joined	15:48
artagnon	Ugh, pushurl is gross imo.	15:49
EugeneKay	But it can(will) give you problems when somebody else has pushed to your fork	15:49
	artagnon looks at .merge	15:49
	→ jonshea joined	15:49
	→ lirt joined	15:49
	→ keller joined	15:49
	← TheJH left	15:49
	← chisler left	15:50
	→ padhah joined	15:50
	→ bootstrapper joined	15:51
	← dvide left	15:51
	→ breck joined	15:51
artagnon	If you wish to setup 'git pull' so that it merges into <name> from another branch in the local repository, you can point branch.<name>.merge to the desired branch, and use the special setting `.` (a period) for branch.<name>.remote.	15:51
	How is this relevant?	15:51
	I don't want to merge one branch into another local branch.	15:52
	← Goles left	15:52
	← padhah left	15:52
	→ padhah joined	15:52
	← jkeiper left	15:53
	→ jkeiper joined	15:53
EugeneKay	You can abuse your fetchspec to do just that	15:53
	← padhah left	15:53
	→ PaulCapestany joined	15:53
	→ jragon joined	15:53
	← jragon left	15:53
	→ jragon joined	15:53
	← jaequery left	15:53
	→ padhah joined	15:54
artagnon	I think pushurl is useful if I have a server that's configured to have one high-availability read-only anongit.* url, and another less-available committer-only url.	15:54
EugeneKay	Set remote.upstream.fetch to something like +refs/heads/:refs/heads/upstream/	15:54
	← skorgon left	15:54
EugeneKay	Instead of refs/remotes/upstream/*	15:54
	→ jdunck joined	15:54
	← cmmagid left	15:54
EugeneKay	Then your branch.foo.merge to upstream/foo ;-)	15:54
	→ cmmagid joined	15:55
	→ adamben joined	15:55
EugeneKay	It craps up your 'git branch' output a bit, but chances are you want to see the upstream branches anyway, so it really saves you using -a	15:55
	→ thl0 joined	15:55
	← JohnBat26 left	15:55
	← jmd left	15:55
	→ ErKa joined	15:55
	→ skorgon joined	15:56
	→ anish[1] joined	15:56
artagnon	! This is a really bad hack.	15:57
	→ yshh joined	15:57
artagnon	The refs/remotes namespace is for remote branches, from which to merge. refs/heads is for local branches only.	15:58
	So, you're proposing creating a local branch called "upstream/<branchname>" and fetching into that.	15:58
	← kojul left	15:58
	← skyxxx left	15:59
	← tvw left	15:59
	→ kojul joined	15:59
artagnon	And you're merging from that local branch.	15:59
	→ tvw joined	15:59
	← buibex left	15:59
	→ shashin joined	16:00
artagnon	So far so good. Won't pushes go to "upstream/foo" (as opposed to "foo" on the remote?	16:00
	← CodeShark left	16:00
	→ buibex joined	16:00
artagnon	... and where in this entire process do I push to the remote "mine"?	16:00
	→ pluto joined	16:01
	← juvenal left	16:01
	→ LuxuryMode joined	16:01
artagnon	EugeneKay: What you said doesn't make sense.	16:01
	← bauruine left	16:02
	artagnon starts writing a remotes.<branchname>.remotepush configuration option	16:02
	→ polymar joined	16:02
	← yshh left	16:03
	← netiul left	16:03
	← iSaleK left	16:03
	→ jsoa` joined	16:03
	← normanrichards left	16:03
	→ iSaleK joined	16:03
	→ Noldorin joined	16:03
	← apeiros_ left	16:04
	← aldskf3 left	16:04
	← kevwil left	16:04
pluto	hi all. i have a question about removing from bare git repo a 'unnamed danling history'. please look for example at http://pastebin.com/76CwMvCU	16:04
	→ infynyxx joined	16:04
pluto	i'd like to drop commits fedb87e, 7c604de in simple way	16:04
	← Exio left	16:05
	← cmmagid left	16:05
	← todd_dsm left	16:05
artagnon	pluto: You could try `rebase -i -p` as a preliminary step.	16:05
	If it doesn't work, you'll have to rewind history, rewrite, and recreate the merge yourself.	16:05
	← tvw left	16:05
cheater	jast: yeah :\	16:06
	→ thesisus joined	16:06
	← jsoa left	16:06
	→ tvw joined	16:06
	→ cmmagid joined	16:07
	→ kevwil joined	16:07
	← smirn0v left	16:07
	← tvw left	16:07
	← mikewintermute left	16:07
	← fRAiLtY- left	16:08
	← kojul left	16:08
	← thl0 left	16:08
	→ tvw joined	16:08
	← exoon left	16:08
	→ todd_dsm joined	16:08
	→ Exio joined	16:08
	← picoguy left	16:08
	→ thl0 joined	16:09
	← alunduil left	16:09
	← tvw left	16:09
pluto	artagnon: 1. rebase doesnt work in bare repo. 2. i just want to only drop two dangling commits.	16:09
	→ TheJH joined	16:09
	← dharrigan left	16:09
	→ tvw joined	16:09
	→ dharrigan joined	16:09
	← dharrigan left	16:09
	→ dharrigan joined	16:09
artagnon	pluto: Attach a worktree to it!	16:09
	→ ojacobson_ joined	16:10
jast	if they're dangling, how is rebase going to make any difference?	16:10
	← ojacobson left	16:11
	→ juvenal joined	16:11
	← lirt left	16:12
	← flijten left	16:13
	← kevwil left	16:13
	← Xeno left	16:13
	→ lirt joined	16:13
	← jdunck left	16:13
	← asker left	16:14
	→ hyperair joined	16:14
	← padhah left	16:14
	← ThomasLocke left	16:15
	→ chisler joined	16:15
	→ alunduil joined	16:16
	→ dreiss joined	16:16
	→ Saby joined	16:16
	← Saby left	16:16
	→ Saby joined	16:16
	← rath left	16:16
	← millun left	16:17
	→ kevwil joined	16:18
	← alunduil left	16:18
	← thl0 left	16:19
	→ m0viefreak joined	16:19
	← wasanzy left	16:19
	← m0viefreak left	16:20
	→ m0viefreak joined	16:20
	→ alunduil joined	16:20
	← cagedwisdom_ left	16:20
	← rideh left	16:21
	→ lethjakman joined	16:21
	→ picoguy joined	16:22
	← sirdancealot7 left	16:22
	← dc5ala left	16:22
	→ kenperkins joined	16:22
	← aberrant left	16:22
	← bootstrapper left	16:23
	→ hatake joined	16:23
	→ aberrant joined	16:23
	← cheater left	16:23
	← samuel02 left	16:23
	← Noldorin left	16:24
	← infynyxx left	16:24
	→ main joined	16:24
	→ infynyxx joined	16:25
	← thesisus left	16:26
	← lun_ left	16:27
	→ gustavnils joined	16:27
	→ sluther joined	16:29
	← beliveyourdream left	16:29
	← todd_dsm left	16:30
	← thiago left	16:31
	← shashin left	16:32
	← ckoch786 left	16:32
	→ rastar joined	16:32
	→ lun_ joined	16:33
	→ normanrichards joined	16:33
	← ciphergoth left	16:34
	→ shashin joined	16:35
	← Jackneill left	16:35
	→ pongeboll joined	16:35
	← jaseemabid left	16:35
	← pongeboll left	16:35
	→ sirdancealot7 joined	16:36
	→ berserkr joined	16:36
	← rchavik left	16:36
	← marthinal left	16:37
	← tvw left	16:38
	→ NiteRain joined	16:38
	→ tvw joined	16:38
	→ bootstrapper joined	16:39
	← kermit left	16:39
	→ sobhan joined	16:40
	→ kermit joined	16:40
	→ fission6 joined	16:40
	→ rideh joined	16:41
	← tbosua left	16:41
	← cbreak-work left	16:41
	← Goplat left	16:41
	→ marthinal joined	16:41
	→ amccloud joined	16:41
	← xiangfu left	16:42
	← corimarisol left	16:43
	← Sebo left	16:43
	→ ckoch786 joined	16:45
	← lirt left	16:46
	→ sniffells joined	16:46
	→ lirt joined	16:47
	← tPl0ch left	16:47
	→ lavaman joined	16:48
	→ kojul joined	16:49
	← Targen left	16:49
	→ fibsifan joined	16:49
	← hatake left	16:49
	← svetlyak40wt left	16:49
	← a-l-e left	16:50
	→ thl0 joined	16:50
	→ jaequery joined	16:50
	← fibsifan left	16:50
	← hyperair left	16:50
	→ Cromulent joined	16:50
	→ skyxxx joined	16:52
	← clone1018 left	16:53
	→ mikef_ joined	16:53
	→ clone1018 joined	16:53
	← LeMike left	16:54
	← sixcorners left	16:54
	→ aartist joined	16:55
	→ cogitos joined	16:55
	← skul left	16:56
	← PaulCapestany left	16:57
	→ fobban joined	16:57
	→ sysadminAli joined	16:58
	← tsdh left	16:58
	→ exoon joined	16:58
	← thl0 left	16:58
	→ LeMike joined	16:58
	→ chrisgeorge joined	16:58
	→ BradLilley joined	16:59
	→ relipse joined	16:59
	→ samuel02 joined	16:59
	← arkiver left	17:00
	← samuel02 left	17:00
	→ dmarkham joined	17:01
	← JDuke128 left	17:01
	← Ksg89 left	17:01
	← koskoz left	17:01
	← cryosphere left	17:02
	← lirt left	17:02
	← RaceCondition left	17:02
	← GeertV left	17:03
	← arvind_khadri left	17:03
	→ lirt joined	17:03
	→ GeertV joined	17:03
	→ daniel_- joined	17:04
	← chisler left	17:04
	→ gumnos joined	17:04
	← AdrienBrault left	17:04
	← jxpx777 left	17:04
	→ abcdefg_ joined	17:05
gumnos	Is there an easy way to mark a branch as "a merge should never come from this branch"? Perhaps some hook/script place?	17:05
	← LeMike left	17:05
	← BradLilley left	17:05
	→ chisler joined	17:05
	→ soypirate joined	17:05
	→ boombatower joined	17:06
	← artagnon left	17:07
	← sobhan left	17:07
	→ BradLilley joined	17:07
	← GeertV left	17:08
	→ thesisus joined	17:09
	→ Fretta joined	17:09
	← Hounddog left	17:10
	→ jaseemabid joined	17:10
	→ TZer0 joined	17:10
	← pluto left	17:10
	← kevwil left	17:11
	srcerer_ → srcerer	17:11
	→ vansonhk_ joined	17:11
	→ kevwil joined	17:11
	→ lresende joined	17:11
	gitinfo set mode: +v	17:12
TZer0	Quick question... how do I refer to a variable in my git config within a config file or file in a project?	17:12
	← jceb_ left	17:12
	← memoryleak left	17:12
	→ tjbiddle joined	17:13
	← spuz left	17:13
	→ hwrd\|work joined	17:13
ojacobson_	however you want to	17:14
sitaram	LOL!	17:14
	→ thl0 joined	17:14
TZer0	ojacobson_: well, say I have a config value user.something in my config.	17:14
	And I'd like to refer to that value in .gitsubmodules	17:15
	← jaequery left	17:15
TZer0	oops, meant a key associated with a value	17:15
	→ samuel02 joined	17:15
	← Cromulent left	17:16
m0viefreak	thats not possible	17:16
	why would you want to do that, anyways?	17:16
	→ smirn0v joined	17:17
	← soypirate left	17:17
TZer0	m0viefreak: we have a server where we have sperate users, but access to the same folder.	17:17
	← fphilipe left	17:17
	← mikecmpbll left	17:18
	→ mikewintermute joined	17:18
m0viefreak	?	17:18
	→ aantix joined	17:19
TZer0	so I'd like to be able to run git config --global user.whatever myusernameonthatserver	17:19
	→ soypirate joined	17:19
TZer0	and then refer to user.whatever in the config files.	17:19
m0viefreak	why would you want to refer to that in config files	17:19
	the user.name / user.email is used for committing	17:19
	not for some server related stuff	17:19
TZer0	This is for pulling a submodule.	17:19
	← Sigma left	17:19
TZer0	Look. It is simple	17:20
ojacobson_	Give the submodule a single, stable canonical URL.	17:20
TZer0	we're three people	17:20
	we are using a stable URL	17:20
	we have three different users on the server	17:20
ojacobson_	Use URL rewriting, SSH host aliases, or whatever else you like to do per-user customization of the URL.	17:20
TZer0	same path, whatever	17:20
ojacobson_	(eg. to inject different usernames into it)	17:20
TZer0	mh, true	17:20
	← polymar left	17:21
	← eli left	17:22
	← angelsl left	17:22
TZer0	ojacobson_: thanks, that works	17:22
	→ cryosphere joined	17:24
	→ bryanstedman joined	17:24
	→ tjbiddle_ joined	17:25
	← thl0 left	17:25
	→ j416 joined	17:27
	← tjbiddle left	17:27
	tjbiddle_ → tjbiddle	17:27
	→ goshawk joined	17:27
	← cryosphere left	17:28
	→ juril joined	17:29
	← infynyxx left	17:29
	← buibex left	17:29
	← kpreid left	17:29
	→ thl0 joined	17:29
	← smirn0v left	17:30
	→ rakl joined	17:30
	← Dreamer3 left	17:30
	→ jaequery joined	17:31
	← jaequery left	17:31
	→ g0bl1n joined	17:32
	← LimeBlast left	17:33
	→ maletor joined	17:33
	→ apeiros_ joined	17:33
	← dreiss left	17:33
	← kurkale6ka left	17:34
	→ swiety joined	17:35
	← juvenal left	17:35
	← mikef_ left	17:35
	→ asker joined	17:35
	→ JohnBat26 joined	17:35
	gitinfo set mode: +v	17:35
	← bgerber left	17:35
	→ private joined	17:36
asker	hey guys, im wondering whether i can remove files from git when they r missing on my workspace	17:36
	private → Guest25067	17:36
	← jkeiper left	17:36
asker	lets say i delete like 20 random files in my project, do i have to list them all in git rm??	17:37
osse	you can list them with git ls-files --deleted	17:37
m0viefreak	git update -u	17:38
	should work, too	17:38
	uhm	17:38
	git add -u	17:38
	← rideh left	17:38
	← mishok13 left	17:39
asker	yah that worked out, thank you ;)	17:39
	any downside of using git add -u instead of listing all files manually?	17:39
	beside you staging all deleted files?	17:40
m0viefreak	well, it would stage everything else, too (like other modified files)	17:40
	but usually this is what you want	17:40
	and its probably easier to add -u everything	17:41
	and then unstage 1 file you changed, if you dont want that	17:41
	← STenyaK2 left	17:41
asker	oh ofc thats ok ;) ty	17:41
	mb one more thing, now i rollback to the lastest commit?	17:41
	without listing all commitS?	17:42
	→ trypwire joined	17:42
joephilly	head~1 ?	17:42
	nm, didnt read the whole thing	17:42
m0viefreak	git revert HEAD or git reset --hard HEAD~1 ?	17:42
	→ Noldorin joined	17:42
m0viefreak	or what do you mean?	17:42
joephilly	throw away changes?	17:43
asker	joephilly: yap	17:43
trypwire	hey guys. I have a pretty situational question: suppose I fork a repo, make a new branch, write some bug-fixin' code, and push that code up to my branch on my forked repo. I'd like to issue a pull request (github), but stuff has changed on the main repository… I would like to rebase, but I already pushed my code to my forked repo… what can I do?	17:43
	→ PaulCapestany joined	17:43
	← pgokeeffe left	17:43
	→ juvenal joined	17:44
trypwire	ideally i would have done `git pull --rebase` before pushing any of my new commits. but I didn't :(	17:44
	→ RonScagz joined	17:44
joephilly	asker, want to keep staged changes or not?	17:44
asker	no	17:44
	everything to the last commit	17:44
	empty stage	17:44
	← RonScagz left	17:45
joephilly	git reset --hard head <-- something like that?	17:45
asker	HEAD is poiting to the last commit?	17:45
osse	yes	17:45
joephilly	head is the tip of the branch I think?	17:46
	← kermit left	17:46
osse	most of the time. It points to what will be the parent if you make a new one	17:46
	→ bgerber joined	17:46
	← lirt left	17:46
asker	git reset --hard head worked out ty :D	17:46
osse	you could also probably do 'git checkout .' which may make more sense	17:47
	checkout the directory (again) instead of this weird "reset" monkey business	17:47
	→ warez joined	17:47
ojacobson_	trypwire: if you're reasonably sure nobody has pulled your changes yet, you can rebase them and re-push them (git push --force to the same github remote and branch)	17:47
asker	doesnt work	17:47
	← aberrant left	17:47
asker	git checkout head...	17:47
ojacobson_	or leave it as-is	17:47
m0viefreak	git checkout -- .	17:48
asker	no	17:48
	← eletuchy left	17:48
osse	asker: the dot was part of the command	17:48
	← ThePeach left	17:48
asker	oh k ;)	17:48
	→ lirt joined	17:48
asker	yap thats worked out, nice, thank you guys	17:48
trypwire	ojacobson_: i'm positive nobody pulled my changes. but will rebasing now recognize the point at which I started making changes? If my commits are already pushed?	17:48
asker	make one more think: how to get the commit where head is poiting to	17:49
	← lirt left	17:49
osse	asker: 'git show'	17:49
	asker: well, 'git show HEAD' but it defaults to HEAD if you don't specify a commit	17:49
	you can 'show' any commit, btw	17:50
	→ thiago joined	17:50
asker	aaaaight, rly nice, thanks (new to this version managment stuff, thats why kind of silly quastions)	17:50
osse	new git or to version control in general?	17:50
	→ lirt joined	17:50
	← g0bl1n left	17:50
m0viefreak	if you want HEAD's sha1: git rev-parse HEAD	17:50
asker	version control	17:50
	never used it	17:50
osse	well, you picked the right tool to learn it with	17:50
	← lirt left	17:51
asker	yap seems rly reseanable to use it	17:51
osse	(not really, because any other tool will frustrate you)	17:51
	← sdboyer left	17:51
asker	and the doc on the git side with them grafics is rly nice to understand the whole idea of that	17:51
	← lun_ left	17:51
	→ Liir joined	17:51
	→ Sigma joined	17:52
	← Asenar left	17:52
asker	i have read about this rebasing stuff, what are you using more often, merge or rebase?	17:52
trypwire	ojacobson_: ah, yea it worked :) cheers	17:52
	→ Cromulent joined	17:52
asker	aaah i have something more i want to know that i didnt understand:	17:53
	← Iszak left	17:53
joephilly	I rebase a lot	17:53
asker	tags and branches... why to use tags if you have branches?	17:53
	→ lun_ joined	17:53
joephilly	I hate automatic merge commits	17:53
m0viefreak	branches move, tags are fixed	17:53
asker	ok, so i will learn more about rebasing	17:53
joephilly	tag never moves	17:53
asker	thats the diff?	17:53
m0viefreak	yes	17:53
cmn	asker: branches is where you develop	17:54
	a tag is saying you're sticking a name to a particular version	17:54
asker	but you can go to a tag and dev there too	17:54
cmn	no, you can't	17:54
	→ cilly joined	17:54
asker	oh i see, if i go to a tag and dev, then commit -> tag wont showing on the lastest commit	17:54
	→ mikef_ joined	17:55
	→ jaequery joined	17:55
joephilly	tag doesn't automatically move to the next commit like a branch does	17:55
cmn	what do you mean showing?	17:55
	if you checkout a tag, you'll detach HEAD	17:55
	← trypwire left	17:55
asker	if i checkout a tag	17:55
	the commit is loaded	17:55
	megha → baba	17:55
	← zoresvit left	17:56
	→ zoresvit joined	17:56
	← alkino left	17:56
joephilly	tags are for marking a specific spot in the timeline	17:56
	that spot wont change	17:56
	ojacobson_ → ojacobson	17:56
cmn	loaded? when you do the checkout, that commit is checked out, yes	17:57
	but you detach HEAD	17:57
asker	ok, i will probably get it all better when i actually use it in real dev and not just in example projects	17:57
	detach HEAD? what does that mean?	17:57
cmn	!detach	17:58
gitinfo	[!detached] A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	17:58
	← maxmanders left	17:58
	→ cheater joined	17:58
	→ rideh joined	17:58
asker	ok i think i got it ;)	17:59
	← jlopez left	17:59
asker	i will just use git to get it better, theory is good but practise is better	17:59
	← breakingthings left	17:59
	→ edoceo joined	18:00
	← _br_ left	18:01
	baba → gia	18:01
	→ CodeShark joined	18:01
asker	thank you have a nice evening!	18:01
	← asker left	18:01
	gia → baba	18:01
	→ aberrant joined	18:02
	→ lacrymology joined	18:02
	← JohnBat26 left	18:02
	← j416 left	18:02
	Nevik pokes osse	18:03
osse	je suis	18:03
	← anderse left	18:03
Nevik	osse: check out my changes :>	18:03
	← NiteRain left	18:03
osse	:O	18:04
	→ _br_ joined	18:04
	← chaw left	18:04
osse	nice work	18:04
	→ JohnBat26 joined	18:04
apeiros_	is `git symbolic-ref --short HEAD` the proper way to get the currently checked out branch?	18:05
	← CodeShark left	18:05
	← _br_ left	18:06
	← alunduil left	18:06
osse	apeiros_: is it perl?	18:06
apeiros_	osse: hu? this is plain shell	18:06
	→ _br_ joined	18:06
	← _br_ left	18:06
osse	apeiros_: if it's interactive then the shortest way I know is 'git branch' and look for the line with the asterisk	18:07
milki	apeiros_: looks right	18:07
osse	Nevik: about bytes vs ints and so on. CPUus typically work fastest when working with word-sized data (e.g. ints)	18:07
apeiros_	osse: ah, well, yes, I will use it to shell out	18:08
	→ _br_ joined	18:08
edoceo	I've pushed a branch to a repo that has a lot of noise in it, sauasge lets say and I need to clean it	18:08
milki	edoceo: !sausage	18:08
apeiros_	and no, it's ruby. I wanted to use grit, but grit doesn't seem to have commands for `git checkout`.	18:08
gitinfo	edoceo: [!sausage_making] Some developers like to "hide the sausage making", transforming their commits before presenting them to the outside world. See http://sethrobertson.github.com/GitBestPractices/#sausage and !perfect	18:08
osse	apeiros_: then $(git symbolic-ref --short HEAD) is better :P	18:08
	← shruggar left	18:08
	milki liked grit	18:08
milki	apeiros_: if you want python, go with dulwich	18:08
edoceo	milki: yes, I've read that numerous times and followed it but I still don't "get it"	18:08
	→ alunduil joined	18:08
apeiros_	osse: not sure I understand - I used `` to delimit the command. it isn't meant to be part of the command	18:09
edoceo	When I make a new branch from my sploppy one - do I need to rebase?	18:09
Nevik	osse: i hope that the compiler does a bit of magic on that	18:09
apeiros_	osse: or in what way is surrounding the command with $() better?	18:09
osse	Nevik: the compiler can't do anything about it.	18:09
milki	edoceo: depends on what you want to do	18:09
Nevik	osse: this is what the output currently looks like: http://pastie.org/private/eldhslpor9vf5tm7ym2lpa (there's a button on the top right that make the display wider)	18:09
milki	edoceo: is the new branch going to be the cleaned up one?	18:09
osse	apeiros_: is it meant for a script?	18:09
	← alunduil left	18:09
milki	edoceo: while you keep the sloppy one just for an old reference?	18:09
	→ Lutt joined	18:09
edoceo	yes, I basically need to hide two comits where I did stupid things	18:09
apeiros_	osse: yes. I will use it within a ruby script.	18:10
edoceo	Eventtaully we'll kill the slop	18:10
Nevik	osse: is there a specific part of the code you are referring to with that?	18:10
	→ alunduil joined	18:10
milki	edoceo: on the new branch then, you can use interactive rebase to clean the new branch up	18:10
	edoceo: and you shouldnt touch the sloppy branch anymore	18:10
osse	Nevik: not really. Afaik manipulating ints is just generally faster than manipulating data types smaller than int	18:10
	on any modern CPU	18:11
edoceo	Do I checkout the new from the old? That is `git checkout -b user/edoceo/12345-clean` while on user/edoceo/12345` ?	18:11
osse	apeiros_: ahh ok. Never mind me then.	18:11
edoceo	Then `git rebase master` interactively ?	18:11
	← ozette left	18:11
osse	apeiros_: I thought it was a shell script	18:11
	→ whowantstolivef1 joined	18:11
milki	that would create a new branch using HEAD as starting point, so it will work	18:11
Nevik	osse: we can do comparison tests at some point	18:11
apeiros_	osse: ok	18:11
milki	edoceo: no, just git rebase -i to mess with the current branch	18:11
Nevik	squeezing more efficiency out of that is only in our favour	18:11
	← Zombie__ left	18:12
	← swiety left	18:12
edoceo	Huh!? In user/edoceo/12345 I just say `git rebase -i`	18:12
	← tvw left	18:12
milki	edoceo: no, git rebase -i in the branch you want to clean	18:12
	edoceo: o, and the number of commits you want to change	18:12
	→ lirt joined	18:12
edoceo	Ok, I'll repeat this to make sure I'm getting it right...	18:13
milki	edoceo: something like git rebase -i HEAD~5 to mess with the last 5 commits	18:13
	← Lutt left	18:13
	← thl0 left	18:13
edoceo	Oh, I need to mess with more, it took me like two weeks to get this right	18:13
	→ jxpx777 joined	18:13
milki	edoceo: maybe HEAD@{2 weeks ago} will work too	18:13
apeiros_	is there a nicer way to extract a directory at a specific revision from a repository than this? https://gist.github.com/apeiros/4732913	18:13
	→ thl0 joined	18:13
	← danielfilho left	18:13
edoceo	So, I'm on the old branch user/edoceo/12345; and I'll `git checkout -b user/edoceo/12345-new` - step one	18:14
	Then on user/edoceo/12345-new is when I `git rebase -i HEAD@{2 weeks ago}` ?	18:14
milki	apeiros_: could also check it out to ta different working tree	18:14
osse	Nevik: http://stackoverflow.com/a/1148505/1249680	18:15
milki	edoceo: ya, though i dont know if @{2 weeks ago} will work or is sufficient	18:15
edoceo	Could I say `master` with out being to dangerous?	18:15
apeiros_	milki: oooh, thats now possible? won't that be a lot slower for cases where the changeset is small?	18:15
milki	no, that would be a different meaning	18:15
edoceo	I want it to clean merge with the current master at the end of this	18:15
	← babilen left	18:15
	→ thl0_ joined	18:15
milki	apeiros_: its always been possible to checkout to a different worktree	18:15
	apeiros_: GIT_WORK_TREE stuff	18:16
	← thl0 left	18:16
	← satiricon left	18:16
milki	edoceo: no, a rebase with a different branch name means something different	18:16
edoceo	So, I should find the reference hash for when I split out user/edoceo/12345 - and rebase to that one?	18:16
apeiros_	milki: so, `GIT_WORK_TREE /to/some/other/dir git checkout SOME_REVISION`, like that?	18:16
	→ d2dchat joined	18:16
milki	apeiros_: depends on your shell, bash probably needs an = there	18:16
	← bgerber left	18:17
apeiros_	ah right, yes	18:17
	← cilly left	18:17
apeiros_	got to try that	18:17
	← thl0_ left	18:17
	→ thl0 joined	18:17
	← caitp left	18:17
Nevik	osse: it's on the to-do list. but i have finals on monday and tuesday and just worked on it a bit today during study breaks	18:18
edoceo	!perfect	18:18
gitinfo	[!postproduction] So, you want to make your commit history look pretty before pushing? http://sethrobertson.github.com/GitPostProduction talks you through how to use 'rebase -i' to do this.	18:18
	→ balls2thewall joined	18:18
	→ jcromartie joined	18:18
osse	Nevik: So I can just revert the commit where you changed size_t to uint8_t? :D	18:18
	Nah	18:18
	→ eletuchy joined	18:18
apeiros_	milki: funny, but it still changes the current working directory (deletes all files in the directory I check out)	18:19
	← Noldorin left	18:19
Nevik	osse: you could, but i think that is one place where it doesnt matter so much, since we only use our own loops in the reporting/printing functions	18:20
	← GarethAdams left	18:20
Nevik	osse: what happens often is the memcmp and the main loop, so that's where we gotta tweak	18:20
	← m0viefreak left	18:21
apeiros_	I did: `GIT_WORK_TREE=/foo git checkout MY_TAG -- dir/in/repo`, that deleted all files in dir/in/repo, and created dir/in/repo with the files of the specified revision in /foo. that's almost what I want. if I could get git to not modify the current workdir, that'd be awesome…	18:21
osse	Nevik: true	18:21
	← Cromulent left	18:21
	← thl0 left	18:22
	→ dreiss joined	18:22
	→ blackmat` joined	18:22
	← zoresvit left	18:22
	← balls2thewall left	18:22
	← d2dchat left	18:23
	→ Desproges joined	18:24
	→ svetlyak40wt joined	18:24
EugeneKay	apeiros_ - have ou read !deploy ?	18:24
gitinfo	apeiros_: Git is not a deployment tool. You can build one around it for simple environments. http://sitaramc.github.com/the-list-and-irc/deploy.html	18:24
apeiros_	EugeneKay: nice of you to draw conclusions of what I intend to do. No, I don't try to use git to deploy.	18:25
	← kini left	18:25
	← sirdancealot7 left	18:25
	→ kini joined	18:25
EugeneKay	I assumed no such thing. It just gives good info about GIT_WORK_TREE ;-)	18:26
cmn	it is extremely rare that such a question isn't about deployment	18:26
	← tjbiddle left	18:26
apeiros_	I use git to do what it was designed to do - get an older revision and compare it to a newer revision. it's serialized data. I try to load the data as efficiently as possible without affecting the working directory.	18:26
EugeneKay	Which is very close to doing a deployment, if yu think about it	18:27
	← kini left	18:27
cmn	it's the same, just under a different name, really	18:27
EugeneKay	You may be interested in man git-archive for obtaining a copy of path/to/dir/ in a sane format	18:27
gitinfo	the git-archive manpage is available at http://jk.gs/git-archive.html	18:27
osse	apeiros_: if you want a complete copy of a partical commit then perhaps you want git archive	18:27
EugeneKay	(which is also covered under deployment)	18:27
osse	damn	18:27
milki	apeiros_: that i dont know o.O, certainly not a problem when its a bare repo	18:27
cmn	but if you want to compare, then diff seems like a much simpler way	18:27
	→ smirn0v joined	18:28
cmn	also, if you override the worktree, you also should override the gitdir	18:28
milki	o?	18:28
	← eydaimon left	18:28
	← lun_ left	18:28
	→ bauruine joined	18:28
	→ zortnac joined	18:28
	→ grizwako joined	18:29
cmn	they have odd interactions if you don't set them both	18:29
	← smirn0v left	18:29
	→ tplaner joined	18:29
	← bootstrapper left	18:29
	← lirt left	18:29
	→ NiteRain joined	18:30
	→ rath joined	18:30
	→ jdunck joined	18:30
	→ lun_ joined	18:31
	→ Vert joined	18:31
	→ rabellamy_ joined	18:32
	→ kini joined	18:32
	← __BigO__ left	18:32
	← anish[1] left	18:32
	← Vile left	18:32
	→ bootstrapper joined	18:33
	→ infynyxx joined	18:33
	→ TheComrade joined	18:33
	→ bgerber joined	18:34
	← marthinal left	18:34
	→ marthinal joined	18:34
	→ padhah joined	18:35
	→ cworth` joined	18:35
	← TheComrade left	18:35
	← jaseemabid left	18:36
	→ tjbiddle joined	18:36
	← Liir left	18:37
	← erkules left	18:37
	→ yakko joined	18:37
	→ smirn0v joined	18:37
	→ erkules joined	18:38
	→ sirdancealot7 joined	18:38
	← cworth left	18:38
	→ thl0 joined	18:40
	→ Archdude joined	18:40
	← cheater left	18:41
	→ shesek joined	18:41
	→ __BigO__ joined	18:41
shesek	what could be causing that: http://pastie.org/6089048 ?	18:41
	← bootstrapper left	18:42
Archdude	If me and a coworker are working on the same 'master' branch and he merges a file into master that I'm working on in my branch - what's the proper way to sync up?	18:42
	→ MistuhKurtz joined	18:42
	→ Chillance joined	18:42
Archdude	it's saying there are conflicts	18:43
	→ hatake joined	18:43
osse	merge your two branches	18:43
apeiros_	osse, milki: sorry, had to answer the door, forgot to write here that I'm afk	18:43
	← vansonhk_ left	18:44
	← dreiss left	18:44
gumnos	I tried asking earlier but didn't get a response. Is there an easy way to mark a branch as "a merge should never come from this branch"? Perhaps some hook/script place? I want to prevent merging changes from a customer-specific branch (yeah, I know it's somewhat bad practice to do that, but this is a whatever-pays-the-bills case) back into other dev/master branches.	18:44
Archdude	osse: merge my work branch into master? or can I merge master into mine for now?	18:44
apeiros_	osse, milki: the point is, I only need one directory of the repo. i.e. the repo might be rather big. the serialized data chunk is rather small. so if I can somehow only check out the specific directory at that revision, that'd be awesome.	18:45
	I'm more worried about not affecting the working directory, though	18:45
	i.e., I want to change revision, load data, restore working dir	18:46
	← smirn0v left	18:46
apeiros_	but that process already requires a clean working dir, which sucks :-/	18:46
osse	Archdude: yuo can do both if you want. What makes sense is up to you. Are you done with whatever task you're working on in your branch? If yes you can merge into master and be done. If no it's probably a good idea to merge from master into yours so that you're "up-to-date" on what happens in master.	18:46
	← cworth` left	18:46
Archdude	osse: thanks	18:46
	← laurav left	18:47
apeiros_	EugeneKay: ok @ good info. sorry if I got jumpy	18:47
osse	If it's something you're gonna be working on for a longer time I recommend you merge from master to your branch from time to time	18:47
	← exoon left	18:47
apeiros_	EugeneKay: git-archive without the archive bit would be awesome… :-S	18:48
osse	apeiros_: git archive \| tar -xf - :)	18:48
	→ daniel_-_ joined	18:48
	→ HG` joined	18:48
apeiros_	osse: that hurts me physically wrt efficiency :)	18:48
osse	why?	18:48
	it's a plain tar. there is no compression involved	18:49
apeiros_	taring the whole thing just to untar it again?	18:49
	← papegaaij left	18:49
apeiros_	ah, true, at least it's not compressed	18:49
osse	a plain tar file is basically just all files concatenated	18:49
	there is very little metadata	18:49
	← daniel_- left	18:49
osse	as far as I know	18:49
	← shwaiil left	18:49
osse	To quote Wikipedia on the issue: "Tar is modified pitch produced primarily from the wood and roots of pine by destructive distillation under pyrolysis."	18:50
	hmm	18:50
	← adamben left	18:50
	← robbyoconnor left	18:50
	← robotmay left	18:51
	← mikef_ left	18:51
	← daniel_-_ left	18:52
	← svetlyak40wt left	18:53
	→ Liir joined	18:53
EugeneKay	I think you may have the wrong page	18:53
osse	:P	18:53
	→ apineda_ joined	18:54
	→ babilen joined	18:54
	→ apok joined	18:54
osse	Lookinigg at the correct article, it looks like there is a 512 byte header for each file	18:54
	Looking	18:54
	I say "correct", which is not to say that the other article is wrong :P	18:54
	→ Sebo joined	18:55
	← apineda left	18:55
	→ jjore joined	18:55
	← sirdancealot7 left	18:56
	← suy left	18:56
	← mikewintermute left	18:56
MistuhKurtz	hey guys I have a little something I'm confused about. "git status" is showing that 1 clone of a repo is ahead of origin/master by 14 commits. Another clone is showing it's ahead by 2 commits. But git log seems to show that they all have the same commits	18:57
	→ tewecske joined	18:57
apeiros_	osse, EugeneKay: ok, the following seems to be relatively efficient and leave the working directory unaffected:	18:57
	git archive --format=tar SOME_REVISION DIR_IN_REPO \| tar -xf EXTERNAL_DIR	18:57
	← amccloud left	18:57
	← shashin left	18:57
apeiros_	thanks for the help :)	18:57
MistuhKurtz	how can I determine what's going on here?	18:58
canton7	MistuhKurtz, with the same commit hashes?	18:58
osse	MistuhKurtz: git log starts listing the current branch. Run git log --all or git log origin/master to see the rest	18:58
	→ drizzd joined	18:58
	→ jmzc joined	18:58
jmzc	hi	18:58
gitinfo	jmzc: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	18:58
osse	apeiros_: you can leave out --format=tar if you want to :)	18:58
apeiros_	I prefer to be explicit	18:59
	not that some change in default renders it not working	18:59
EugeneKay	It's a good policy. Defaults should never be trusted	18:59
osse	apeiros_: and do you mean tar -C EXTERNAL_DIR -xf - ?	18:59
jmzc	i'm newbie with git and I'd like to know what is the difference between master and origin/master branch , when is printed in logs	18:59
apeiros_	`-f -` is actually a default too	18:59
	← prudnikov left	18:59
MistuhKurtz	osse, they're both in master. canton7 can I just read the last 4 digits to compare or is there a tool I can use to definitively compare the two?	18:59
apeiros_	I think tar's defaults are stable, though :)	18:59
EugeneKay	jmzc - master is your local branch; origin/master is your copy of the master branch as it exists on the remote named origin.	18:59
	→ rabellamy joined	19:00
osse	apeiros_: sure, but EXTERNAL_DIR comes right after -f :O	19:00
apeiros_	osse: no, I have no -f in there?	19:00
	← thl0 left	19:00
apeiros_	ah, whoops	19:00
	copy & paste mistake	19:00
	→ smirn0v joined	19:00
	→ thl0 joined	19:00
apeiros_	git archive --format=tar SOME_REVISION DIR_IN_REPO \| tar -xC EXTERNAL_DIR	19:00
canton7	MistuhKurtz, last few are fine	19:00
osse	Woohoo!	19:00
	← zortnac left	19:00
jmzc	EugeneKay: ok, thanks ...but they don't have to store the same contents, right ?	19:01
MistuhKurtz	yep they're the same canton7	19:01
	→ pantsman joined	19:01
	← thl0 left	19:02
osse	MistuhKurtz: you might want to give git log --all -decorate --oneline --graph a spin. It gives a better overview of what is going on	19:02
jmzc	EugeneKay: my doubt is cause of when i make a 'git pull' , the next log is printed : "master -> origin/master"	19:02
	→ thl0 joined	19:02
MistuhKurtz	okay git log origin/master is showing differences now	19:02
	→ mikef_ joined	19:02
jmzc	EugeneKay: but i think that it should be in the reverse way	19:02
	→ netzmonster joined	19:03
	← tewecske left	19:03
	→ amccloud joined	19:04
canton7	jmzc, git pull first does a fetch. this stores the remote's 'master' branch in 'origin/master'	19:04
	so the 'master' in that output is the remote's master, not your local one	19:04
jmzc	canton7: so, remote brach is 'master' ? EugeneKay told me that " master is your local branch; origin/master is your copy of the master branch as it exists on the remote named origin."ç	19:05
canton7	he's right	19:05
	← normanrichards left	19:05
	→ rondon joined	19:05
	← rabellamy left	19:05
canton7	a repo exists on the remote machine, known here as 'origin', which has a master branch	19:06
MistuhKurtz	osse, so one of them shows a commit with (origin/master, origin/HEAD) in front of the message, and the other doesn't	19:06
canton7	origin/master is a local record of the state of that remote branch. so when you 'git fetch', you update origin/master to match the master branch in the remote repo	19:06
	→ francisfish joined	19:06
osse	MistuhKurtz: and another commit later on shows (master, HEAD) right?	19:06
	→ kogent joined	19:06
canton7	and then you've got a local branch 'master'	19:07
MistuhKurtz	(HEAD, master), yep	19:07
canton7	MistuhKurtz, 'origin/HEAD' has very little meaning	19:07
	it's the default branch that's checked out when you clone a repo	19:07
	→ prudnikov joined	19:07
canton7	(most of the time)	19:07
osse	MistuhKurtz: if you do a plain 'git log' it will show the commit log from HEAD and backwards	19:07
	MistuhKurtz: which explains why you didn't see the commits you where looking for	19:07
	were	19:08
	→ maxmanders joined	19:09
MistuhKurtz	sorry I'm not fully following the implications	19:09
jmzc	canton7: what a mess :-P	19:09
canton7	jmzc, it makes an awful lot of sense :P	19:09
	→ sirdancealot7 joined	19:09
canton7	every time you want to compare your local master and the remote's master, you probably don't want to go fetching data from the remote repo	19:10
osse	MistuhKurtz: techincally the master branch you have on your computer and the master branch on wherever origin is are different.	19:10
jmzc	canton7: you said " git pull first does a fetch. this stores the remote's 'master' branch in 'origin/master'" , but isnt origin the remote repo?	19:10
osse	MistuhKurtz: the implication is that the master branch at origin is ahead of yours. Work has been done that you have not incorporated into your own master branch (yet)	19:11
	→ svetlyak40wt joined	19:11
canton7	jmzc, origin is the name you've given to the remote repo, yes. 'git pull' is shorthand for 'git fetch && git merge' (with appropriate flags and arguments)	19:11
	→ Plouj- joined	19:11
MistuhKurtz	even though it says the two repos are ahead of origin/master osse ?	19:11
EugeneKay	jmzc - your repo keeps a local copy of the branches as they exist on the remote repo; this is so you can do operations "offline"	19:11
	← howlymowly left	19:11
	← iSaleK left	19:12
	→ anderse joined	19:12
EugeneKay	jmzc - these "remote-tracking" branches are stored in your local repo, and can be viewed with 'git branch -a'	19:12
	→ JaredMiller joined	19:12
osse	MistuhKurtz: I lost oversight. What did the git status message say again?	19:12
	← Plouj- left	19:12
	→ Plouj joined	19:12
EugeneKay	jmzc - when you do a 'git fetch, the status of the remote-tracking branches	19:12
Plouj	hi	19:12
	does anyone know what's the format of a remote git notes ref given a remote called 'origin' and a notes namespace called 'mine'? Eg: what's the remote equivalent or refs/notes/mine ?	19:12
EugeneKay	is synchronized	19:12
	← thesisus left	19:12
	← JaredMiller left	19:12
	→ dreiss joined	19:12
MistuhKurtz	osse, the one with the (origin/master, origin/HEAD) in front of the commit says it's ahead of origin/master by 2 commits	19:12
	→ kermit joined	19:12
MistuhKurtz	the other one says it's ahead by 14 commits	19:12
grawity	Plouj: I don't think git fetches refs/notes/* by default	19:13
canton7	jmzc, you've grocked that 'origin/master' is the name of a local branch? if you look in .git/refs/remotes you'll see it	19:13
	← thl0 left	19:13
grawity	Plouj: See remote.origin.fetch in .git/config	19:13
Plouj	grawity: What if I've already fetched the notes refs?	19:13
osse	MistuhKurtz: I'm not following. Can you pastebin the output of git status and git log --all --etc ?	19:13
MistuhKurtz	ya 1 sec	19:13
	← AnimeCYC left	19:13
canton7	Plouj, where did you fetch it to?	19:13
jmzc	EugeneKay, canton7: ok, thanks	19:14
grawity	Plouj: If you don't specify a local ref, then FETCH_HEAD is used	19:14
canton7	jmzc, !book offers from good reading	19:14
gitinfo	jmzc: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	19:14
Plouj	I just did 'git fetch origin refs/notes/mine:refs/notes/mine'	19:14
	→ scorphus joined	19:14
grawity	Plouj: well	19:14
canton7	particularly cs and bottomup	19:14
ojacobson	Then they're in refs/notes/mine	19:14
grawity	Plouj: then it fetched refs/notes/mine to refs/notes/mine	19:14
	→ opus joined	19:14
	← mikef_ left	19:14
	← Nevik left	19:14
Plouj	oh, then there's no separate 'origin' ref, like origin/branch vs branch?	19:14
canton7	you could fetch to origin/refs/notes/mine	19:15
Plouj	yeah	19:15
grawity	Plouj: you explicitly specified the target ref	19:15
osse	MistuhKurtz: oh you're dealing with two different clones? That makes more sense. Run 'git fetch' in both and now origin/master should point at the same commit in both	19:15
canton7	your 'normal' fetchspecs do that. e.g. remote.origin.fetch = +refs/heads/:refs/remotes/origin/	19:15
	→ j416 joined	19:15
Plouj	grawity: Ok, I see what happened.	19:15
grawity	Plouj: refs/remotes/origin/branch is created only becau—————what canton7 said	19:15
	→ CodeShark joined	19:15
	→ archonix joined	19:16
	← rondon left	19:16
MistuhKurtz	osse, http://pastebin.com/krxcx8gs	19:16
Plouj	would something like +refs/notes/:refs/remotes/origin/notes/ work for fetching notes?	19:16
canton7	MistuhKurtz, go into staging, 'git fetch'	19:16
MistuhKurtz	no args?	19:16
	← zerba left	19:17
canton7	'git fetch origin' if you want	19:17
	→ n3m joined	19:17
ojacobson	Plouj: from git-fetch's point of view, notes are no different from any other constellation of objects	19:17
canton7	can't remember what the default behaviour of 'git fetch' is if there's more than one remote...	19:17
ojacobson	the name transformations and whatnot are as given in man git-fetch	19:17
gitinfo	the git-fetch manpage is available at http://jk.gs/git-fetch.html	19:17
osse	MistuhKurtz: perhaps the staging repo doesn't have a remote configured.	19:17
MistuhKurtz	remotes are configured correctly	19:17
canton7	Plouj, sure	19:17
	← jcromartie left	19:17
MistuhKurtz	git remote -v matches	19:17
	→ thesisus joined	19:17
canton7	aha	19:17
	← R1KRDO left	19:17
	→ breakingthings joined	19:17
canton7	MistuhKurtz, 'git config remote.origin.fetch' in both?	19:18
	→ R1KRDO joined	19:18
Plouj	thanks!	19:18
	for the sanity checks :)	19:18
canton7	it's all so simple when you know it :P	19:18
MistuhKurtz	canton7, they both match	19:18
	+refs/heads/:refs/remotes/origin/	19:19
	← francisfish left	19:19
	→ jcromartie joined	19:19
MistuhKurtz	after running 'git fetch' in staging I don't see the "ahead of origin by x commits" message anymore	19:19
Plouj	canton7: indeed	19:19
canton7	MistuhKurtz, cool. so the problem is that in staging origin/master is way behind, or that it doesn't exist at all?	19:19
osse	MistuhKurtz: git fetch apparently hasn't been run in staging for a long time, which means the information about the status of origin/master there is outdated	19:20
MistuhKurtz	it's on github and their front-end log looks like it matches too	19:20
	← WinstonSmith left	19:21
MistuhKurtz	yep same last commit with the same hash	19:21
	← gain left	19:21
	→ khmarbaise joined	19:22
osse	run git fetch often. it can't hurt	19:23
MistuhKurtz	okay, and that's run along with 'pull' too? should I just stick to fetch -> merge instead of 'pull' so I can make sure everything's alright beforehand?	19:23
osse	MistuhKurtz: stick to fetch and merge unless you know what the result of the merge will be beforehand	19:24
grawity	would `git pull --ff-only` be safe in this case?	19:25
osse	MistuhKurtz: if you make a "pull mistake" it's fixable, so in practice it's no big deal. I think it's easier to not lose track of what's going on if you fetch first though	19:25
	especially because git status will tell you	19:25
MistuhKurtz	osse, thanks for the advice. I think I agree it's best. what's beautiful about git is how confident I can (usually) be	19:26
	so anything that helps improve confidence is good practice in my book	19:26
	→ h0bbit joined	19:26
osse	grawity: yeah I think so if what you were going to do was to fast-forward if possible anyway.	19:27
MistuhKurtz	osse, so I can safely go ahead and do a 'git fetch' on my production as well?	19:27
	→ WinstonSmith joined	19:27
osse	MistuhKurtz: git fetch only updates the <remote>/* branches. So yes.	19:27
	MistuhKurtz: what it casically does is update the information about what the status is at the remote	19:28
	→ daniel_- joined	19:28
grawity	hmm	19:29
canton7	(unless you configured your repo as a mirror, or fiddled with the fetchspecs)	19:29
MistuhKurtz	alright, git status says working dir is clean	19:29
	I'll take it everything is settled?	19:29
	← marc_kdab_ left	19:29
MistuhKurtz	one last git log..	19:30
osse	It can say that even if master and origin/master are different.	19:30
	But if it ONLY says that then master and origin/master are probably equal	19:30
	← rath left	19:30
MistuhKurtz	yep everything matches up	19:30
	both clones say this for the last commit:	19:30
	← Chinorro left	19:30
MistuhKurtz	* 5c8a332 (HEAD, origin/master, origin/HEAD, master) Changed Media Coverage design to reflect new specs	19:31
	what does that all mean?	19:31
canton7	the currently-checkout-out branch, the local copy of the remote master branch, the default branch on origin, and the local master all point to the same commit	19:31
	(in that order)	19:31
	← skyxxx left	19:31
	← thatguycraig left	19:32
	→ Nevik joined	19:32
	→ shashin joined	19:32
	→ wlonk joined	19:32
	→ hobodave joined	19:33
MistuhKurtz	HEAD represents the state at the last commit of the currently checked out branch, right?	19:34
osse	Also the media coverage design has been chagned	19:34
MistuhKurtz	I thought origin/master was the default branch on origin canton7 ?	19:34
osse	MistuhKurtz: not last commit; any commit	19:34
canton7	MistuhKurtz, origin/master is the local record of the 'master' branch on origin	19:35
	origin/HEAD is the branch that's checked out by default when you clone the remote repo	19:35
	HEAD is the commit that's currently checked out. so if you 'git commit', the commit pointed ot by HEAD will be the new commit's parent	19:35
	well, HEAD is strictly the branch which is currently checked out. cat .git/HEAD and you'll see	19:37
	→ rabellamy joined	19:37
	← spyro left	19:37
canton7	similarly cat .git/refs/heads/master while you're at it ;)	19:37
MistuhKurtz	trying to wrap my head around origin/HEAD still	19:37
canton7	don't worry about it	19:37
	→ spyro joined	19:37
canton7	all git repos have a HEAD. when the repo has a working tree, i.e. it's not bare, HEAD has a lot of meaning	19:38
osse	MistuhKurtz: if you make a new clone, that commit will be checked out in the new clone. That's about all it means.	19:38
	→ fphilipe joined	19:38
canton7	in a bare repo, it doesn't mean much, since there's no working tree. it just indicates to git which commit should be checked out when you clone the repo	19:38
	s/which commit/which branch	19:38
	→ Cromulent joined	19:38
canton7	(although 'which commit' is strictly more accurate I guess....)	19:38
	← berserkr left	19:38
MistuhKurtz	haha	19:38
	→ bootstrapper joined	19:39
osse	canton7: "well, HEAD is strictly the branch which is currently checked out" No I think you were right the first time that HEAD is which commit. Hence detached heads	19:39
	← rastar left	19:39
canton7	side point: if 'master' and 'develop' point to the same commit, and HEAD points to 'develop', you might get 'master' checked out when you clone the repo	19:39
	due to a limitation in the git protocol	19:39
	→ melee joined	19:39
MistuhKurtz	so origin/HEAD just means that what's currently checked out matches the HEAD of origin?	19:40
canton7	that's what I was referring to with by 'strictly' comment	19:40
	→ thl0 joined	19:40
canton7	MistuhKurtz, origin/HEAD is a pointer to a commit, just like any branch	19:40
	it's shown up in 'git log' because you passed --all	19:40
MistuhKurtz	it's a pointer to a commit on the remote origin?	19:41
	← eletuchy left	19:41
	→ eletuchy joined	19:41
	← cmmagid left	19:41
	→ mikef_ joined	19:41
	→ cmmagid joined	19:41
	← n3m left	19:41
osse	MistuhKurtz: "so origin/HEAD just means that what's currently checked out matches the HEAD of origin?" No. If you check something else out, 'origin/HEAD' will still appear next to the same commit.	19:41
	← cmmagid left	19:42
	→ cmmagid_ joined	19:42
	← rabellamy_ left	19:43
	← lun_ left	19:43
	← dreiss left	19:44
	→ lun_ joined	19:44
	→ dreiss joined	19:44
	← breck left	19:45
	← rabellamy left	19:45
	→ shruggar joined	19:45
	← amccloud left	19:45
	→ rabellamy joined	19:45
	← NiteRain left	19:46
MistuhKurtz	oh whoops osse sorry that was a bad question	19:46
	adhawkins-away → adhawkins	19:46
	← lun_ left	19:46
	→ lun_ joined	19:46
	← netzmonster left	19:46
osse	Not really. The answer just happened to be no :)	19:46
MistuhKurtz	haha	19:47
	so origin/HEAD is the default branch on origin. What's the 'default' branch?	19:47
	← _br_ left	19:47
canton7	the branch which is checked out by default when you clone the repo	19:47
	(yet again...)	19:47
	← bootstrapper left	19:48
MistuhKurtz	Think I got it	19:48
	sorry for the repetitiveness, thanks for beating it into my brain canton7 :)	19:48
	and you too osse, really appreciate the help.	19:48
canton7	heh, no worries :)	19:48
	→ marc_kdab_ joined	19:48
	← dogarrhea left	19:48
osse	MistuhKurtz: make a commit on master. Then check out some old commit at random, then run the log command again. Now you'll see these labels at different commits	19:49
	→ _br_ joined	19:49
MistuhKurtz	ahhhhhhhhh	19:50
	AHHHHHHH	19:50
osse	:O:O:O:O	19:50
MistuhKurtz	:D	19:50
	← hinst left	19:51
MistuhKurtz	my new coworker and I are diving into git together and are thoroughly enjoying it. before hiring him as my new team member I had pretty much just developed alone. we're having a blast. I look forward to being able to be as helpful as you two someday!	19:51
osse	did you just have a a-ha moment?	19:52
MistuhKurtz	oh yes	19:52
osse	do you need to... clean up?	19:52
MistuhKurtz	hahaha	19:52
	← kevwil left	19:52
	← thl0 left	19:52
	→ sq-one joined	19:52
	→ howlymowly joined	19:53
MistuhKurtz	yeah but only in my brain. I'll get the janitor up there right on it	19:53
	→ thelodger joined	19:53
	← _br_ left	19:53
	→ d2dchat joined	19:54
	→ _br_ joined	19:54
	← _br_ left	19:54
cbreak	!cs	19:54
gitinfo	"Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup	19:54
	← svetlyak40wt left	19:55
	← juril left	19:56
	→ francisfish joined	19:57
	← rabellamy left	19:57
	← scorphus left	19:58
	→ _br_ joined	19:58
	← looper left	19:58
	→ rabellamy joined	19:58
	← rabellamy left	19:58
	→ looper joined	20:00
fission6	git pull, will pull and merge respective branches correct	20:01
MistuhKurtz	yep	20:01
	fetch & merge, rather	20:01
	→ kevwil joined	20:02
	→ scorphus joined	20:02
	← jkremser left	20:02
	→ odwalla joined	20:02
	→ juril joined	20:02
	← d2dchat left	20:04
fission6	if i have a local branch which i have pushed upstream but doesn't seem to track it how od i set that up MistuhKurtz	20:04
paulweb515	fission6: will fetch from the remote, and then merge the checked out branch only	20:04
fission6	i get this when i do git pull http://dpaste.org/SFX7I/	20:04
paulweb515	fission6: use a git push -u	20:04
	← philips_ left	20:04
canton7	fission6, or git branch --set-upstream	20:05
	→ Mayq joined	20:05
canton7	or --set-upstream-to in newer git	20:05
paulweb515	fission6: i.e. git push -u origin myBranch	20:05
	→ danielfilho joined	20:05
fission6	paulweb515: i did that and get To prevent you from losing history, non-fast-forward updates were rejected	20:05
paulweb515	fission6: you've made changes to your local branch and someone has pushed updates to the remote branch?	20:06
canton7	and the message goes on to tell you waht to do?	20:06
fission6	ok i think i got it	20:07
	so how do i update and track all remote branches	20:07
	→ svetlyak40wt joined	20:07
	← maletor left	20:07
paulweb515	fission6: git fetch && git merge --rebase origin/myBranch; Then you can push and -u	20:07
	← Milossh left	20:07
fission6	how do i pull down and track all remote branches	20:08
PerlJam	fission6: "git fetch" will get them all.	20:08
	→ exoon joined	20:08
fission6	i seem them listed with git branch -a	20:08
	but just the remote refs	20:08
PerlJam	fission6: tracking them all locally is weird unless you're going to contribute back to every one of them on a regular basis.	20:09
paulweb515	fission6: you would only track them if you checked out a local branch of the same name	20:09
fission6	well basically i have a repo on unfuddle, i want to move it over to github, so i want to have the entire repo then push it to github right	20:09
	← opus left	20:09
	→ d2dchat joined	20:10
	→ AnimeCYC joined	20:10
grawity	hmm, I wonder if you could push github refs/remotes/origin/:refs/heads/	20:10
	← Chillance left	20:10
	→ philips_ joined	20:10
	← philips_ left	20:10
	→ Chillance joined	20:10
	← d2dchat left	20:10
PerlJam	fission6: you could "git clone --mirror" and then switch remotes and "git push --all" (or something like that)	20:11
	← jsoa` left	20:11
fission6	hmmm	20:12
	→ avjpl joined	20:12
	← avjpl left	20:12
grawity	I just tested my `git push` line, and yes, it works	20:14
	→ WorkeD joined	20:15
	→ MrSaku joined	20:16
	← shruggar left	20:16
	→ sdboyer joined	20:16
grawity	I wish I remembered it last week :/	20:16
MrSaku	How to repeat holocaust, fast	20:16
	Whats difference between nigger and barrel of shit? its the barrel hhaahaha@!	20:16
	← sdboyer left	20:16
	← grizwako left	20:17
	← amigojapan left	20:17
	← MrSaku left	20:17
	→ esreskkesketske joined	20:17
	→ LeMike joined	20:17
	→ philips_ joined	20:18
	← machty left	20:18
osse	What a nice man	20:18
	→ etcetera joined	20:18
	← Mayq left	20:18
	← JohnBat26 left	20:18
canton7	he jumped into a load of other channels too. ah well, I'll ban him next time	20:18
	← sysadminAli left	20:18
	→ jbroome joined	20:19
	ChanServ set mode: +o	20:19
osse	I've seen worse	20:19
	← jbroome left	20:19
osse	I've seen thing you people wouldn't believe	20:19
	Attack ships on fire off the shoulder of Orion	20:20
	→ sdboyer joined	20:20
cbreak	I've seen blade runner too.	20:20
osse	Many people have, I suppose	20:20
	→ lahwran joined	20:20
cbreak	hmm... I forgot how to ban people...	20:21
	ah well	20:21
	cbreak set mode: -o	20:21
	→ sysadminAli joined	20:21
	→ bootstrapper joined	20:21
osse	I feel safer now	20:22
	→ normanrichards joined	20:22
	→ cmmagid joined	20:22
	→ machty joined	20:23
	← cmmagid left	20:23
	→ cmmagid joined	20:24
	← cmmagid left	20:24
	← cmmagid_ left	20:24
	→ cmmagid joined	20:24
	→ hyperair joined	20:24
	→ robbyoconnor joined	20:25
	← Cromulent left	20:26
	→ razieliyo joined	20:26
razieliyo	#github	20:26
	huh	20:26
	fail	20:26
	lol	20:26
	← jcromartie left	20:26
	→ beepbeep_ joined	20:27
	← scorphus left	20:27
	→ opus joined	20:27
beepbeep_	.random	20:27
gitinfo	Using rebase -i a lot to shape history? Look at rebase.autosquash and git commit --fixup. [11]	20:27
beepbeep_	.random	20:28
gitinfo	<cbreak> password generator: <cbreak> dd if=/dev/random bs=512 count=8 \| openssl base64 [2]	20:28
razieliyo	I want to keep backup of my config files, and I've started a git repo in ~/, just to push the config files/folders I want, do you think this may be a problem in the future?	20:28
	← Natureshadow left	20:29
_ikke_	look at !etc	20:29
gitinfo	[!fetchfour] Never use the four-word version of git-fetch or git-pull (e.g. git fetch <remote> <refspec>). It always ends in tears. Yes, if you understand the implications of FETCH_HEAD it can technically be done, but really it is easier to just fetch the whole remote (or perhaps edit the fetchspec if you never want other bits). If you must, see !fetch4why	20:29
_ikke_	arg	20:29
	!etckeeper	20:29
gitinfo	etckeeper is a collection of tools to let /etc be stored in a git, mercurial, darcs, or bzr repository. It hooks into various package managers. http://kitenet.net/~joey/code/etckeeper/	20:29
	→ Natureshadow joined	20:29
_ikke_	razieliyo: sorry, not really relevant	20:29
	razieliyo: Although git can be used for it, it's not really an ideal solution	20:30
	→ AdrienBrault joined	20:30
	← \etc\ left	20:31
	← infynyxx left	20:33
grawity	cbreak: openssl rand 100 -base64	20:34
	← asaf left	20:35
	← sysadminAli left	20:35
	← bgerber left	20:35
	→ tsdh joined	20:36
beepbeep_	.random	20:36
gitinfo	Things aren't working or working badly? Blaming Java, PHP or Rails is a good default. [20]	20:36
beepbeep_	.random	20:37
gitinfo	DOUGH -- 200 g saw dust, 800 g flour, 590 millichopins water, 40 g yeast, 1 pinch of salt, 1 tbsp olive oil, 1 tbsp camphor. SAUCE -- 1 can minced tomatoes, 118 millichopins petroleum ether, 50 g laburnum (dried), 50 g repulsion gel, salt/pepper. TOPPINGS -- fish-shaped sediment, 100 g zombie flesh, 20 g rhubarb, 200 g cheese. Bake at 130 °Rø for 103.97 Sothic nanocycles. Enjoy! [1]	20:37
beepbeep_	.random	20:37
gitinfo	Things aren't working or working badly? Blaming Java, PHP or Rails is a good default. [20]	20:37
	← alunduil left	20:39
	→ sysadminAli joined	20:41
	← aberrant left	20:42
	→ Milossh joined	20:42
	→ shruggar joined	20:42
	← whowantstolivef1 left	20:43
	→ n3m joined	20:43
cbreak	beepbeep_: you could try to whisper to gitinfo	20:44
_ikke_	But then we wouldn't share the fun	20:44
	← WorkeD left	20:44
	← Milossh left	20:46
	← wlonk left	20:46
	→ rabellamy joined	20:46
	→ na3r joined	20:47
	→ bgerber joined	20:48
	→ jakemp joined	20:48
	→ Milossh joined	20:49
	← Milossh left	20:49
jakemp	is there an email frontend for git?	20:49
	→ seanh_ joined	20:50
	→ Milossh joined	20:50
	← fphilipe left	20:50
	← Milossh left	20:50
	← shruggar left	20:50
	→ Shusshu joined	20:50
jakemp	so repos are made for each email address, a directory for each subject, and allowing versioning using the attachments.	20:51
	→ Milossh joined	20:51
	← Milossh left	20:52
	← Shusshu left	20:52
	← timj1 left	20:52
	← gajop left	20:52
	→ thl0 joined	20:53
	→ Milossh joined	20:53
	← seanh_ left	20:53
	→ laurav joined	20:53
	→ esreskkesketske_ joined	20:54
	← oojacoboo left	20:55
Nevik	jakemp: ive not heard of any such thing, but it's probably not too hard to implement	20:56
	once you work out the details	20:56
	← esreskkesketske left	20:56
	← svetlyak40wt left	20:57
	← thl0 left	20:57
	→ shruggar joined	20:57
	← maxmanders left	20:57
jakemp	yeah, I can script it pretty easily, just wanted to make sure I wasn't reinventing the wheel.	20:57
	→ amccloud joined	20:57
jakemp	thanks for the input.	20:58
	→ wget joined	20:58
	← na3r left	20:59
	→ oojacoboo joined	20:59
	→ nicekiwi joined	21:00
	← smirn0v left	21:00
nicekiwi	there are files on my remote repo that are not in the repo on github, how can I get rid of them without messing up stuff?	21:00
	→ Gentlecat_ joined	21:00
nicekiwi	ive been doing pull requests from git hub on my remote server	21:01
	→ maxmanders joined	21:01
	← shashin left	21:02
	→ thl0 joined	21:02
	→ dmatt joined	21:02
milki	nicekiwi: my basic guess is your remote repo is a non-bare repo. you could reset --hard or clean	21:02
	adhawkins → adhawkins-away	21:02
	← blackmat` left	21:02
	← opus left	21:03
nicekiwi	milki: hmm ok..	21:03
	→ aantix_ joined	21:06
	→ wlonk joined	21:07
	→ Noldorin joined	21:08
	→ gajop joined	21:08
	← gustavnils left	21:09
	← aantix left	21:09
	aantix_ → aantix	21:09
	→ Ahti333 joined	21:09
	→ Goles joined	21:10
	← rabellamy left	21:10
kanzure	is .mailmap a git feature?	21:10
Ahti333	i have a merge conflict with a pretty large file, is there any way i can get a copy of both versions in two different files (eg original file is a, i want to have something like a.mine and a.theirs)	21:10
	← lresende left	21:11
kanzure	Ahti333: you can check out those files separately with "git checkout" iirc	21:11
Ahti333	using --mine etc? wouldnt that check them out overwriting the conflicted file and marking the merge as resolved?	21:12
	Tommy[D]_ → Tommy[D]	21:14
grawity	kanzure: yes, see `man git shortlog`	21:14
gitinfo	kanzure: the git-shortlog manpage is available at http://jk.gs/git-shortlog.html	21:14
Ahti333	ah okay, it does not mark the merge resolved, so i can first checkout --ours myfile, then copy that aside and then checkout --theirs	21:14
	← n3m left	21:15
	← beepbeep_ left	21:15
	← juvenal left	21:16
	← chuckharmston left	21:16
	→ revyd joined	21:18
	← marthinal left	21:18
	← sysadminAli left	21:19
	← ncd\|leeN left	21:19
	← j416 left	21:19
	← picoguy left	21:19
	← mikef_ left	21:20
	← ErKa left	21:20
	← fobban left	21:21
	→ ryanplas joined	21:21
	← tsdh left	21:22
	→ marhaban joined	21:22
wget	Hello guys. I'm looking if there is a way to display an old version of a file under version control, that has been committed and published for ages, but without checking it out or reverting the repo to the previous version of the file. I tried git diff but it displays nothing.	21:22
PerlJam	wget: git help show	21:22
	wget: git show <commit>:path/to/file # IIRC, but I rarely do this so it doesn't stick in my brain too well	21:23
	← thl0 left	21:23
	← dmatt left	21:23
	→ mishok13 joined	21:23
wget	PerlJam: Thanks a lot! ;-)	21:24
	← p4tux left	21:24
	← eletuchy left	21:25
	← francisfish left	21:25
	→ chuckharmston joined	21:26
wget	PerlJam: Remembering the command name is always a tricky process compared to checking out its description with the man pages.	21:26
	← etcetera left	21:26
	→ etcetera joined	21:27
	← shruggar left	21:27
jmzc	hi	21:27
	← roger_padactor left	21:28
	← etcetera left	21:29
jmzc	when I talk about 'local repository' , is this repository a real copy of sources in my machine? I don't mean my workspace, but the 'local repository'	21:29
	or 'local repository' only has got control data ?	21:29
	→ ajacom joined	21:30
ajacom	is there a command to automatically fetch a branch and merge it with the current ?	21:30
bremner	jmzc: can you give some context?	21:30
PerlJam	jmzc: I'm not quite sure what you're asking. The local repo has everything needed to recreate a working copy.	21:30
	← snql left	21:30
jmzc	bremner: first, the context is I'm newbie :-)	21:30
Seveas	local repository is the result of 'git clone'. A bare repo (somename.git) is a local repository. A non-bare repo (somename/.git) is a repo with a checkedout working directory	21:31
	ajacom, git pull <remote> <branch>	21:31
	→ _iron joined	21:31
jmzc	I want to know how many copies of my source files there are	21:31
PerlJam	jmzc: why?	21:32
ajacom	Seveas, git pull heroku master ... or git pull heroku/master current-branch ?	21:32
Seveas	ajacom, with an origin named heroku and a remote branch named master: git pull heroku master	21:33
jmzc	if I delete all my wokspace and drop remote repo, do ? have lost all my files ?	21:33
ajacom	thanks	21:33
Seveas	jmzc, unless someone else has a copy	21:33
cbreak	jmzc: drop?	21:33
	→ marthinal joined	21:33
jmzc	drop o become unaccessible	21:34
	desapear	21:34
	← ajacom left	21:34
cbreak	if you delete a repository, then that repository is gone	21:34
	the history it had might be in other repositories too	21:34
jmzc	Seveas: so, administrative area is only meta-data of my workspace	21:35
cbreak	for example if you fetched from it, or pushed the history somewhere else	21:35
	← bgerber left	21:35
cbreak	the working directory is not part of history	21:35
	→ glebd joined	21:35
bremner	what "administrative area"?	21:35
Seveas	jmzc, no, if you have a .git, you have your entire repo.	21:35
cbreak	you can delete it as much as you want, git won't care	21:35
	← danielfilho left	21:35
cbreak	(until you commit :)	21:35
jmzc	Seveas: so, I've got my workspace copy, my .git copy and my remote repo copy ?	21:36
	← ryanplas left	21:36
	← e66 left	21:36
jmzc	I talk about FILES	21:36
	physically	21:36
cbreak	git doesn't store files	21:36
	← chuckharmston left	21:36
Seveas	jmzc, err.... the .git dir contains files :-)	21:36
	← Remram left	21:36
jmzc	Seveas: source files ?	21:36
	→ e66 joined	21:36
cbreak	git just stores snapshots of history	21:36
jmzc	cbreak: ok	21:37
cbreak	i.E. the content of files	21:37
	← siriuz left	21:37
Seveas	jmzc, let's take a step back. Why don't you tell us what the actual problem is you're having instead of being vague?	21:37
	← marthinal left	21:37
	→ ankurankan joined	21:37
Seveas	Did you do rm -rf in your workdir but want your files back?	21:37
cbreak	git can recreate files from history	21:37
jmzc	i'm being vague because I'm newbie in git and I don't speak English very well ...2 problems in 1	21:38
Seveas	jmzc, both those problems are not easily solved. So let's focus on the easier problems :)	21:38
	← laurav left	21:38
	← lacrymology left	21:38
	→ siriuz joined	21:38
	→ thl0 joined	21:38
cbreak	well, english isn't that hard if you already know a germanic language	21:38
	and it is quite simple overall	21:38
	→ tjbiddle_ joined	21:39
Seveas	spanish isn't germanic though :-)	21:39
jmzc	i don't have any specific problem ...I only want to know where are my files , and what is .git ...for my, .git folder is a black box	21:39
Seveas	jmzc, ok. Start with reading !cs and maybe pro git then	21:40
gitinfo	jmzc: "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup	21:40
jmzc	Spanish is easer ...i swear :-P	21:40
	→ recurrence joined	21:40
jmzc	ok, thanks	21:40
	← skywalkerhere left	21:41
Seveas	jmzc, basically, in .git/objects, you'll see a lot of files. The contents of these represent all states of all files in your repository. That link above explains it in some more detail, pro git (it's a book) explains a lot more even	21:41
	← aantix left	21:41
	→ laurav joined	21:42
	→ aantix joined	21:42
	→ Remram joined	21:42
jmzc	ok, i'll read ... I don't know if i'll learn English or git on first	21:42
	← tjbiddle left	21:43
jmzc	thanks git-ers !!!	21:43
	tjbiddle_ → tjbiddle	21:43
	→ bgerber joined	21:45
	→ Chinorro joined	21:46
cbreak	jmzc: objects might not be there	21:46
	there are also pack files, and if you're playing with fire, alternates	21:46
xiong	Hi, I seem to have done stupid, then compounded my mistake: http://imgur.com/ATcPGNG	21:47
	I merged in error; release-1.0.1 was not quite okay. I wanted to undo the merge, leaving the branches in the states with the green circles; and dropping entirely what is now the top three commits.	21:49
	← padhah left	21:49
	← ISF left	21:49
xiong	Now when I try to switch out of trunk back into the release-1.0.1, git complains of uncommitted changes.	21:49
cbreak	xiong: did you push already?	21:49
	← tjadc left	21:50
	← thl0 left	21:50
cbreak	and which branch do you want to change?	21:50
	← laurav left	21:51
	→ etcetera joined	21:51
	→ lun__ joined	21:52
	← _iron left	21:52
	→ chuckharmston joined	21:52
	← lun_ left	21:53
xiong	I did not push since the pushes shown. So GitHub is at 1.0.0 on devel and trunk.	21:53
	← abcdefg_ left	21:53
	← geekbri left	21:54
	← simesy left	21:54
	← lun__ left	21:54
	→ lun_ joined	21:54
	→ abf joined	21:54
xiong	I don't want to retain anything after the branch heads shown. That is, devel should be where devel is, release-1.0.1 should be where that is, trunk should be where that is. I just want to back entirely out of the merge and be on the release branch.	21:55
	As a bonus, I'd like to know how to back out of a merge before I add to the cock-up with git-reset.	21:55
	← freeed left	21:55
cbreak	xiong: so you ... want that everything is as it is?	21:56
xiong	No; I want not to have those top three commits.	21:57
cbreak	you don't have them in history	21:57
	they are just in stashes	21:57
	← tjbiddle left	21:57
cbreak	the stash will vanish if you do git stash drop or so	21:57
xiong	Actually, I realize now it's worse than shown. Notice that 'trunk' itself is no longer at 'remotes/origin/trunk'.	21:57
	← jashsu left	21:58
cbreak	yeah, and? maybe your GUI doesn't show it	21:58
	I like !lol more	21:58
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	21:58
xiong	Let's try to break the problem up into sub-issues. Can I get trunk to point where the remote is?	21:58
cbreak	where is it now?	21:59
kanzure	grawity: thank you	21:59
cbreak	pastebin !lol output or so	21:59
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	22:00
xiong	If you like, I can do the decorated log instead. This is clearer in my mind.	22:00
	'trunk' is now way back at 0.1.4	22:00
	→ opus joined	22:00
cbreak	maybe you reset it back.	22:00
	you can probably git checkout it and then just git pull	22:00
	→ tjbiddle joined	22:00
xiong	I did indeed. I got into this mess by trying to back out of the merge using git reset and git stash.	22:00
cbreak	that will work.	22:00
	your merge is no longer referenced by it	22:01
xiong	Ah, but it did not... the way I did it.	22:01
cbreak	you just see stash because your GUI shows it	22:01
	but it's no longer history (in trunk)	22:01
xiong	Yes, I'm hoping that git gc will make the unreferenced commits go away.	22:01
	← dijonyummy left	22:01
xiong	Please, one issue at a time.	22:01
	← __BigO__ left	22:01
	→ infynyxx joined	22:01
	← amccloud left	22:01
	→ mklappstuhl joined	22:02
xiong	First issue: I want to move 'trunk' up to where it is on the remote. I realize that since it is the remote branch, I could pull. But I'd rather not rely on that as a backup.	22:02
	→ yshh joined	22:03
xiong	The commit labeled 'remotes/origin/trunk' is not missing from my local repo yet, right?	22:03
cbreak	you can also reset --hard if you don't like pull	22:03
	or you can just merge	22:03
	since you don't need the fetch part of pull	22:03
xiong	Whoa.	22:03
cbreak	gitk shows it, so it's not missing	22:03
xiong	One thing, please. I want to move 'trunk' up to where it belongs.	22:03
cbreak	then do it.	22:03
	→ thl0 joined	22:03
cbreak	plenty of ways to do it :)	22:03
xiong	Don't understand.	22:03
	← jmzc left	22:04
	← yshh left	22:04
xiong	One way, please, without involving the remote.	22:04
	← bootstrapper left	22:04
	→ yshh joined	22:04
cbreak	git checkout trunk && git merge origin/trunk	22:04
xiong	I don't want to do another merge.	22:05
cbreak	do it anyway.	22:05
	← thesisus left	22:05
cbreak	you can add --ff-only to merge too, if you want to be sure that it will not make a merge commit, but it should not create one in such a situation.	22:05
	since it should be fast forwardable	22:05
xiong	I don't want to create a new commit under any circumstances.	22:05
	← dmarkham left	22:05
xiong	I want to label the commit as 'trunk'. I want trunk to be there.	22:05
	→ amccloud joined	22:06
xiong	Um, git-reset but forward, not backward.	22:06
	→ thesisus joined	22:06
	→ Vile joined	22:06
	← revyd left	22:06
	→ dmarkham joined	22:06
csmrfx	isn't "trunk" technically used for what git knows as "branches"	22:06
	+?	22:06
xiong	'trunk' = 'remotes/origin/trunk'	22:06
cbreak	xiong: do what I said	22:06
	xiong: all will do the same thing	22:06
xiong	Sorry csmrfx; not sure I quite understand.	22:06
cbreak	csmrfx: trunk is just the name of a branch	22:07
xiong	cbreak, I'm already checked out on trunk.	22:07
	I do not want to involve the remote at all.	22:07
cbreak	yeah, don't care	22:07
	git merge origin/master	22:07
	origin/trunk of course	22:08
xiong	I can't depend on having the commit I fumbled already pushed to GitHub.	22:08
cbreak	that will do what you want... as I told you many times already	22:08
xiong	Call me fussy.	22:08
axl_	can i undo a commit from my log that hasn't been pushed to the repo ? I don't want to revert it, just remove the commit and undo those changes, as though they were never done	22:08
cbreak	axl_: yes.	22:08
xiong	I will have this problem again and I don't want to waste anybody's time asking for the solution again because I don't have a remote to save my butt.	22:09
cbreak	axl_: git reset --soft HEAD~1	22:09
	xiong: so, do it...	22:09
xiong	Do what?	22:09
cbreak	xiong: merges are local only	22:09
	do what I said... duh...	22:09
axl_	cbreak: should i pass the SHA instead of the HEAD~1 ?	22:09
	← cannonball left	22:09
cbreak	axl_: what ever you want. both will work	22:09
	→ tjbiddle_ joined	22:09
axl_	cbreak: awesome! thanks	22:09
xiong	Let's try again. How do I arbitrarily set the branch tip to an arbitrary sha?	22:10
csmrfx	IMO "label" is known in git as a "tag"	22:10
cbreak	axl_: (the sha1 of the commit you want to be at)	22:10
	xiong: git reset, as I told you already...	22:10
	→ yshh_ joined	22:10
cbreak	xiong: but you don't need arbitrary.	22:10
csmrfx	axl_: yes you can, be our guest	22:10
cbreak	you just need fast forward	22:10
	← yshh left	22:10
	← MistuhKurtz left	22:10
	→ francisfish joined	22:11
axl_	cbreak: wait… i don't want to uncommit all commits until that commit… just that one commit	22:11
csmrfx	xiong: your question doesn't make sense to me	22:11
xiong	The most general solution, yes, probably, git-reset. I have only ever used it to go backwards in time with HEAD^.	22:11
axl_	csmrfx: ?	22:11
	→ laurav joined	22:11
cbreak	axl_: you can only remove the newest commit that way	22:11
	← breakingthings left	22:11
cbreak	axl_: if it's not the newest, then ask again :)	22:11
	→ jcao219 joined	22:11
csmrfx	how about googling "how to revert to previous commit git"	22:11
axl_	cbreak: even though the commits on top of it aren't dependent on it?	22:12
cbreak	axl_: all commits are dependent on their parent	22:12
	← francisfish left	22:12
axl_	csmrfx: thats, not what i am asking	22:12
xiong	csmrfx, I showed a shot of gitk in its current state.	22:12
csmrfx	axl_: depends on what you want to do	22:12
axl_	cbreak: all right… i guess that makes sense.	22:12
	→ drizzd_ joined	22:12
cbreak	axl_: if you want to remove an earlier commit, then that's history rewriting	22:12
	← erichynds left	22:12
	← thesisus left	22:13
cbreak	axl_: see man git-rebase -i	22:13
gitinfo	axl_: the git-rebase manpage is available at http://jk.gs/git-rebase.html	22:13
	← tjbiddle left	22:13
	tjbiddle_ → tjbiddle	22:13
axl_	cool	22:13
cbreak	it can do that easily, but it will rewrite history	22:13
axl_	will look into it	22:13
cbreak	which will cause a huge pile of trouble if you pushed already	22:13
axl_	haven't pusehd	22:13
	pushed*	22:13
cbreak	it's also harder to use.	22:14
axl_	thanks a bunch guys	22:14
	← thl0 left	22:14
cbreak	axl_: see !cs for the details of what it does	22:14
gitinfo	axl_: "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup	22:14
axl_	interesting	22:15
	← machty left	22:15
	← drizzd left	22:15
xiong	Okay; I have used $ git reset --hard <sha> to get the tip of the trunk pointing to the correct commit. Now local and remote trunk are at the same place.	22:15
	← razieliyo left	22:15
cbreak	xiong: you don't know where the real remote trunk is at until you fetch from the remote	22:16
xiong	But I cannot switch out of the trunk branch back to the release branch because there are uncommitted changes.	22:16
cbreak	xiong: your remotes/origin/trunk is just a tracking branch	22:16
xiong	I'm the only author.	22:16
cbreak	it'll be where you fetched last	22:16
	xiong: there should not be any uncommitted changes	22:16
	since you just did a git reset --hard	22:16
xiong	Ah, but there are.	22:16
cbreak	can you look at git status/git diff to get the details?	22:16
	maybe you use windows	22:16
	← archonix left	22:17
xiong	Three files that were untracked, created in the work done in the devel branch.	22:17
	← jrex left	22:17
csmrfx	well, either check out the version overwriting or commit?	22:17
xiong	No, it's linux.	22:17
cbreak	if they are untracked, then just delete them.	22:17
csmrfx	(or gitignore, or mv files to safety, or merge, or...)	22:17
	→ yshh joined	22:18
xiong	I understand that I can manually rm those files and presumably, switch to the release branch and find them back where they were.	22:18
	← yshh_ left	22:18
xiong	I'm trying to learn how to do things cleanly, though.	22:18
cbreak	delete them...	22:18
	→ user12 joined	22:19
cbreak	you can use git clean if you want git to do it for you. but read the man page of that command...	22:19
	just plain rm is faster though	22:19
	xiong reads	22:19
	→ milStaja joined	22:19
	← axl_ left	22:19
csmrfx	well cleaning up mess is rm	22:19
	← sq-one left	22:19
xiong	Ah. That's much better.	22:19
	→ john_anderson joined	22:20
osse	Doing things cleanly doesn't necessarily mean that everything you type in the sheel should start with 'git ' :P	22:20
	git cd ..	22:20
	git make install	22:20
	← R1KRDO left	22:20
	→ n4irc joined	22:21
	← Element9 left	22:21
	← jragon left	22:21
OMGOMG	there should be a git zsnes though	22:21
	→ breck joined	22:21
	← user12 left	22:21
	← drizzd_ left	22:22
	← lamont left	22:22
OMGOMG	sometimes 100 savestates isn't enough	22:22
	← mishok13 left	22:22
	← jonshea left	22:22
	← yshh left	22:22
	← perlstein left	22:22
	← breck left	22:23
	→ yshh joined	22:23
	← soypirate left	22:23
	← exoon left	22:24
	← yshh left	22:24
	→ yshh_ joined	22:25
	→ Dooom joined	22:25
	→ x82_nicole joined	22:25
	← Shurakai left	22:25
	→ shruggar joined	22:26
xiong	Okay: http://imgur.com/mpSWSyW	22:26
	So here's what I did:	22:27
	$ git reset --hard <sha> # to fix the tip of 'trunk'	22:27
	$ git clean -f # to drop the untracked files	22:27
	→ \etc\ joined	22:27
	← LuxuryMode left	22:28
xiong	-- then I checked out the 'release' branch --	22:28
	→ yshh joined	22:28
xiong	$ git stash drop	22:28
	$ git gc	22:28
csmrfx	xiong: should that be "head" instead of "tip"?	22:28
xiong	Now everything seems okay... in git. The underlying issue with my code is another thing, of course.	22:28
	→ lamont joined	22:29
xiong	How many HEADs are there, csmrfx? I'm a bit confused about that myself.	22:29
	I thought there was only ever one HEAD.	22:29
csmrfx	one single head for each branch	22:29
	its just a "pointer"	22:29
bremner	no, just one HEAD	22:29
xiong	That's why I say 'tip'.	22:29
	→ TeamColtra joined	22:30
csmrfx	http://stackoverflow.com/questions/2304087/what-is-git-head-exactly	22:30
	← yshh_ left	22:30
csmrfx	if you say "tip" on #git, people wont get shit	22:30
	→ mishok13 joined	22:30
	← HumanSky left	22:30
xiong	I don't know exactly what to call them. I think of them as pseudo-tags or branch tags. As you see, I rarely make explicit tags.	22:30
csmrfx	I always thought one is supposed to call it (them) head	22:31
xiong	If anybody has a canonical word for these, please let me know. It doesn't work for me to call them 'branches', since I think of a branch as being all the commits in that line.	22:31
	← shruggar left	22:32
	← esreskkesketske_ left	22:32
bremner	well, git disagrees ;)	22:32
	← ojacobson left	22:32
	← yshh left	22:32
xiong	Okay, bremner; so you call a 'branch' only the most recent commit? What I've been calling the tip of the branch?	22:32
	→ yshh joined	22:33
bremner	I guess "heads" is semi-blessed by the implementation, since they all live unfer refs/heads	22:33
charon	xiong: that's a bit of a dangerous mindset to get into, because the set of commits in a branch can change at any time, commits may be part of no branches, etc. commits don't have an intrinsic branch membership	22:33
bremner	xiong: a branch is a symbolic-ref (pointer to a commit)	22:33
	← bgerber left	22:33
charon	(not that there's anything wrong with referring to a "topic branch" but really meaning the three commits not in master but on that branch)	22:33
xiong	Maybe not, charon; but logically, they're connected. If you have a better word for that, fine; I'll try to adopt that.	22:33
csmrfx	that SO thread says "When you switch branches with git checkout, the HEAD revision changes to point to the tip of the new branch.	22:33
	+" - so they use 'tip' too, but the thing is -- HEAD points to the tip	22:33
	ack?	22:33
	csmrfx beware the detached head	22:33
	→ megha joined	22:34
	→ asaf joined	22:34
	← aartist left	22:34
charon	HEAD points to the branch. symbolically.	22:34
	← gyre007 left	22:34
charon	that is, HEAD points to the ref that constitutes a branch, by full name (refs/heads/X usually)	22:34
csmrfx	http://marklodato.github.com/visual-git-guide/commit-master.svg.png	22:34
bremner	oh, right, I am muddying the water, sorry. Branches are not symbolic refs, just refs.	22:34
	← Goles left	22:34
	← baba left	22:35
	← cmmagid left	22:35
xiong	I realize history isn't always this neat. But consider the sequence of 6 commits from Now that's peculiar to Fix POD coverage on Cookbook. The most recent commit of those 6 is feature-cover.	22:35
csmrfx	xiong: in git, "branches" mean the branches, so you call branches "branches"	22:35
	→ eletuchy joined	22:35
csmrfx	singular: "branch"	22:35
	→ breck joined	22:36
xiong	I consider all those 6 commits as belonging to the feature-cover branch and the most recent its tip. If that's poor jargon, what should I call these?	22:36
	dwmw2 → dwmw2_gone	22:36
EugeneKay	In high-level terms, a "branch" is a line of development.	22:36
csmrfx	well if you made branches in your repo, then you would have named them	22:36
charon	xiong: you don't have to defend yourself :) "dangerous mindset" was a bit of a strong choice of words. we'd also call that "the feature-cover branch". it's just important that you are aware that there's nothing in the commits _per se_ that ties them to that branch	22:36
	→ mint joined	22:36
	← mint left	22:36
EugeneKay	And the "head"(or "tip") is the last commit which was made on a particular branch.	22:36
charon	xiong: so that you're not surprised if you wake up in the morning and the branch is something completely new.	22:36
	→ pgokeeffe joined	22:36
charon	(the same branch, that is)	22:37
xiong	charon, the only thing that ties them together is the history that one commit followed another. I understand that.	22:37
	→ bgerber joined	22:37
csmrfx	"$ git branch" lists the branches	22:37
xiong	EugeneKay, you seem to be endorsing my usage of branch and tip.	22:37
	I didn't think it was all that wacky. -	22:37
EugeneKay	That's because it isn't ;-)	22:37
	→ mint joined	22:37
charon	agreed	22:37
	→ nafg joined	22:37
	→ [Araelium]Seth joined	22:37
EugeneKay	The terminology is just very, very messy and there is no man gitdictionary	22:37
xiong	But if there's a better term for "sequence of commits leading to a branch" then I'll use it.	22:38
EugeneKay	Actually	22:38
	man gitglossary	22:38
gitinfo	the gitglossary manpage is available at http://jk.gs/gitglossary.html	22:38
bremner	xiong: but it isn't necc. a sequence	22:38
xiong	And I'll use "branch" to apply exclusively to what I've been calling a "tip".	22:38
EugeneKay	It even has an entry on branch which is remarkably close to hat I just said	22:38
	So there ;-)	22:38
	→ aartist joined	22:38
charon	i suppose the real issue is that in speaking of "foo branch", there's an implicit base commit assumed to be known to everyone	22:38
xiong	A "branch" is an active line of development. The most recent commit on a branch is referred to as the tip of that branch.	22:38
csmrfx	git concepts chapter of the user-manual file:///usr/share/doc/git-doc/user-manual.html#git-concepts	22:39
charon	people are usually surprised when they find out that a) git doesn't know anything about base commits and b) they didn't know either	22:39
	or some such.	22:39
xiong	Thanks, EugeneKay.	22:39
	← DeviantPeer left	22:39
	← alegen_ left	22:39
csmrfx	$ man git	22:39
gitinfo	the git manpage is available at http://jk.gs/git.html	22:39
xiong	Perhaps if I run into this quibble again I can switch from "branch" and "it's tip" to "chain" and "branch".	22:39
	s/it's/its/	22:40
	← juril left	22:40
bremner	well, it isn't necessarily a chain either	22:40
xiong	Fascinating as this is, guys; I still have to go and fix my code.	22:40
EugeneKay	Code is for the weak of mind	22:40
xiong	Thanks for all the help sorting out my repo.	22:40
bremner	non-linear history. get used to it.	22:40
	;)	22:40
EugeneKay	Real men will the computer to do its bidding	22:40
	csmrfx only fixes other peoples code	22:40
	← pgokeeffe left	22:41
	gitinfo set mode: +v	22:41
	← eletuchy left	22:41
	← HG` left	22:42
[Araelium]Seth	[newb] I have git repo for my website. At the top level I have two directories which we'll call /resources and /site. Resources contains all of the raw photoshop files, and big bulky assets. /site contains my actual website's (rails) application.	22:43
	When deploying I want to "checkout" (in svn terms) only the /site directory so I don't have to download all of the contents of /resources. According to Internet Wisdom this is what a "sparse checkout" is in git, yet when I try it, it still downloads all of /resources and /site into the .git/objects directory, but in the working copy only /site is shown. That kind of misses the whole point. Can I prevent it?	22:43
	→ eletuchy joined	22:43
	→ dijonyummy joined	22:43
cbreak	[Araelium]Seth: git doesn't work that way	22:43
	→ lresende joined	22:43
cbreak	a clone contains the full history, including every version of every file ever committed	22:43
[Araelium]Seth	The Internet has lied to me. :(	22:43
cbreak	you can do sparse checkouts, but you still will have every version of every file ever committed in git	22:44
	and sparse checkouts are not really supported	22:44
[Araelium]Seth	And seem useless	22:44
	→ thl0 joined	22:44
[Araelium]Seth	I don't understand the point then, if they don't prevent the download of the other directories	22:44
cbreak	in general, it is a bad idea to commit big binary files	22:44
	← mishok13 left	22:45
cbreak	[Araelium]Seth: someone probably requested the feature	22:45
[Araelium]Seth	They're not gigantic, but it's stuff I do want versioned. Hmm.	22:45
bremner	[Araelium]Seth: for source code, the .git directory is typically much smaller than the checkout	22:45
	well, sometimes anyway.	22:45
[Araelium]Seth	I guess the only way then is two separate repositories.	22:45
	← dijonyummy left	22:45
[Araelium]Seth	+submodules, which I can read up on now.	22:45
cbreak	those are one way	22:46
	but submodules are usually used for external things	22:46
charon	!annex	22:46
gitinfo	git-annex and git-media are two solutions to the !binary problem. They work by keeping the blobs outside of the repo, storing a reference to the blob in the repo instead. See http://git-annex.branchable.com/ and https://github.com/schacon/git-media	22:46
cbreak	maybe git slave is better for your situation	22:46
charon	!binary	22:46
gitinfo	Storing binary files in git causes repo balloon, because they do not compress/diff well. In other words, each time you change a file the repo will grow by the size of the file. See !annex for some solutions	22:46
charon	hum, no other solutions? bah	22:47
bremner	svn ;)	22:47
[Araelium]Seth	heh.	22:47
	Thanks. I'll read about all of this.	22:47
cbreak	charon: well... if you tell git to track them, then git will track them.	22:47
	→ dijonyummy joined	22:48
csmrfx	[Araelium]Seth: deploy post-commit	22:48
	+in a hook	22:48
	→ shruggar joined	22:49
	← revagomes left	22:50
	→ __BigO__ joined	22:50
	→ Fels joined	22:51
	→ banisterfiend joined	22:51
banisterfiend	hi, i dont quite understand the help description of "git log -S"	22:51
	can someone fill me in?	22:51
	← samir left	22:52
	→ thl0_ joined	22:52
	← farciarz84 left	22:52
	← ankurankan left	22:52
[Araelium]Seth	csmrfx: can you elaborate?	22:53
csmrfx	banisterfiend: stop trolling	22:53
	real people need real help here	22:53
	→ adamben joined	22:54
	← hobodave left	22:54
	← bronsen left	22:54
csmrfx	banisterfiend: lol. did you see the man page already?	22:55
	→ maletor joined	22:55
csmrfx	suppose it iterates from earliest commit grepping for the string argument	22:55
	← cwillu_at_work left	22:55
	← yshh left	22:56
	← thl0 left	22:56
xiong	-S or -s?	22:56
banisterfiend	xiong: -S	22:56
	→ samir joined	22:56
	← FUZxxl left	22:56
	→ FUZxxl joined	22:56
banisterfiend	csmrfx: of course, i still dont quite get it	22:56
xiong	I admit, I can't find that parameter in the manpage. It must be a parameter because when issued by itself it demands a value.	22:56
banisterfiend	-S<string>	22:57
	Look for differences that introduce or remove an instance of <string>.	22:57
	← mint left	22:57
	→ cwillu_at_work joined	22:57
banisterfiend	ah	22:58
	i get it	22:58
	nm	22:58
csmrfx	your man page looks different from mine	22:58
banisterfiend	read "differences" as "diff" and it makes sense	22:58
	→ pppaul joined	22:58
csmrfx	-S<string> Look for differences that introduce or remove an instance of <string>. Note that this is different than the string simply appearing in diff output; see the pickaxe entry in gitdiffcore(7) for more details	22:58
xiong	I think -S should be listed under the 'Commit Limiting' section, then; but doesn't seem to be.	22:58
csmrfx	has flag for string-match, iterates from earliest commit of the branch (I guess) grepping all files for string, makes log every time string-match state toggles	22:59
	← eletuchy left	23:00
banisterfiend	does -S also look at the commit message?	23:00
	or just the diffs?	23:00
xiong	Perhaps -S is undocumented in the manpage?	23:00
csmrfx	See my paste on the line above starting "-S<str"	23:00
	← myx left	23:00
	← natechan left	23:00
	← fisted left	23:01
	← jwmann left	23:01
	← mmitchell left	23:01
xiong	Ah. Finally saw it. Listed under 'COMMON DIFF OPTIONS'.	23:01
csmrfx	It matches or greps each commit, each file, for the string argument	23:01
	→ fisted joined	23:02
csmrfx	not the diff	23:02
	→ alegen joined	23:02
csmrfx	hm, should have written "each commits every file"	23:03
	← Guest25067 left	23:03
	← lavaman left	23:03
	→ natechan joined	23:03
	← dmarkham left	23:03
xiong	In other news, Error-Base-v1.0.0 is merged, pushed, uploaded to CPAN, and awaiting fresh smoke. Thanks guys, couldn't have done it without you.	23:04
csmrfx	banisterfiend: it seems rather pointless as both removing a string instance in a file or making new both cause commit to appear	23:05
banisterfiend	csmrfx: no idea what you're talking about homie	23:05
	→ dmarkham joined	23:05
csmrfx	git log -S output	23:05
	← trungqtran left	23:06
	→ eletuchy joined	23:06
	← TheJH left	23:06
	← samuel02 left	23:08
	→ revyd joined	23:08
	→ pgokeeffe joined	23:08
	→ axl_ joined	23:09
	→ maqr joined	23:10
	SeoZ-work[AWAY] → SeoZ	23:10
	← scampbell left	23:11
	← mklappstuhl left	23:12
	← thl0_ left	23:13
	← pgokeeffe left	23:14
	→ GSpotAssassin joined	23:15
	→ ISF joined	23:15
GSpotAssassin	How do I search the commit history of 1 file to see when a line matching some grep was removed?	23:15
	← milStaja left	23:15
	→ n3m joined	23:16
shesek	anyone know what could be causing that: http://pastie.org/6089048 ?	23:16
	← howlymowly left	23:18
grawity	GSpotAssassin: git log -G"line" -- file	23:19
csmrfx	GSpotAssassin: maybe --S and --pickaxe-all	23:19
	→ pgokeeffe joined	23:19
	← jdav_gone left	23:19
grawity	shesek: the first command does not pull origin/master to your own master branch	23:19
csmrfx	banisterfiend: me neither, btw	23:20
grawity	shesek: it only fetches it to FETCH_HEAD, I think	23:20
csmrfx	A pull is the same as fetch and .. ?	23:20
grawity	I think !fetch4 explains it somewhat	23:21
gitinfo	[!fetchfour] Never use the four-word version of git-fetch or git-pull (e.g. git fetch <remote> <refspec>). It always ends in tears. Yes, if you understand the implications of FETCH_HEAD it can technically be done, but really it is easier to just fetch the whole remote (or perhaps edit the fetchspec if you never want other bits). If you must, see !fetch4why	23:21
	→ SBoudrias joined	23:21
	← wlonk left	23:21
shesek	!fetch4why	23:21
gitinfo	The four-word version of git-fetch doesn't update remote-tracking branches and tags. It fetches into FETCH_HEAD only; you probably don't want to have to deal with that. Pull in four-word form automatically uses FETCH_HEAD, but your remote-tracking will still be outdated. Even updating them manually with 'git fetch origin master:remotes/origin/master' (clunky) still doesn't update tags.	23:21
shesek	grawity, thanks. and how should I do that?	23:22
grawity	Use "git pull <remote>" or "git pull"	23:22
glebd	I have a mirror repository. Can I turn it into a normal, non-mirror, repository? Is it enough just to write mirror=false into my local clone and push to the mirror, or is anything more needed?	23:24
	← laurav left	23:24
grawity	You need to also fix the refspec in remote.origin.fetch	23:25
	→ danielfilho joined	23:25
	→ ojacobson joined	23:25
	← etcetera left	23:25
	→ mklappstuhl joined	23:26
shesek	grawity, I'll look into that. thanks	23:26
	← alegen left	23:27
	← lun_ left	23:27
	→ Xethron joined	23:28
	← mutoso left	23:28
	← h0bbit left	23:29
	lb_ → lb	23:30
	← [Araelium]Seth left	23:30
	→ mutoso joined	23:30
	← infynyxx left	23:31
	→ jdav_gone joined	23:31
	→ thasmo joined	23:31
	← apeiros_ left	23:31
	→ machty joined	23:32
	→ alegen joined	23:35
	← bgerber left	23:35
	← SeoZ left	23:36
	← Fels left	23:37
	→ m0viefreak joined	23:37
	→ rabellamy joined	23:37
	← m0viefreak left	23:38
	→ m0viefreak joined	23:38
	→ SeoZ joined	23:39
	← n3m left	23:39
	← hwrd\|work left	23:40
	← amccloud left	23:42
	← normanrichards left	23:43
	← __BigO__ left	23:43
	→ Mannequin joined	23:43
	← looper left	23:44
	← chisler left	23:46
	← maxmanders left	23:46
	→ infynyxx joined	23:46
	→ cworth joined	23:46
	→ bgerber joined	23:46
	→ fuzzybunny69y joined	23:47
	→ tjbiddle_ joined	23:47
	← decotii left	23:48
	← thasmo left	23:48
	← tjbiddle left	23:48
	tjbiddle_ → tjbiddle	23:48
	← machty left	23:48
	← fuzzybunny69y left	23:48
	→ hwrd\|work joined	23:49
	→ dunpeal joined	23:51
dunpeal	Hi. What's the best way to have git just print the name of the current branch, and nothing else?	23:51
banisterfiend	dunpeal: git name-rev HEAD	23:51
dunpeal	ah, right, thanks.	23:52
banisterfiend	dunpeal: not quite, it also shows HEAD	23:52
	dunpeal: try this: git name-rev --name-only HEAD	23:53
dunpeal	yup, thanks.	23:53
ojacobson	There's also 'git symbolic-ref --short HEAD'	23:53
	→ Ksg89 joined	23:53
	← segher left	23:53
ojacobson	(which will puke in predictable ways if HEAD is not a symbolic ref, i.e., when you have a detatched HEAD)	23:54
	← natechan left	23:54
	← darklajid left	23:55
	→ skyxxx joined	23:55
	← chuckharmston left	23:57
	← kevwil left	23:58
	← hwrd\|work left	23:58
	← TonyM left	23:58
	→ segher joined	23:59
	→ hwrd\|work joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation