IRCloggy #git 2013-03-28

Provider of IRC logs since 2005.

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

	2013-03-28
	← jacoluta left	00:00
	← yann\|work left	00:00
	← infynyxx left	00:01
	→ ScottE_ joined	00:01
	→ yann\|work joined	00:01
	← jeffreylevesque left	00:01
	← ScottE left	00:02
	→ danishkhan joined	00:02
	→ BSaboia joined	00:03
	→ engkur joined	00:03
	→ infynyxx joined	00:03
	→ coco89\|away joined	00:04
	← etcetera left	00:04
coco89\|away	I'm getting rejected when pushing to github!!	00:04
	→ x82_nicole joined	00:04
coco89\|away	error: failed to push some refs to 'https://github'...	00:05
	← jfelchner1 left	00:05
kini	EugeneKay: thanks	00:05
	← MistuhKurtz left	00:06
milki	coco89\|away: any more errors?	00:06
coco89\|away	milki: hint: updates were rejected because the tip of your current branch is behind its remote counterpart. Merge the remote changes (e.g. 'git pull') before pushing again.	00:06
milki	coco89\|away: what dont you understand about that?	00:06
coco89\|away	milki: everything	00:06
	tip of current branch what	00:06
	remote counterpart what	00:07
	merge what	00:07
milki	ok	00:07
	← miso1337 left	00:07
milki	so what is your current branch?	00:07
coco89\|away	what	00:07
	:P	00:07
	as in path?	00:07
milki	coco89\|away: whats a branch?	00:07
coco89\|away	a path?	00:07
	← yann\|work left	00:07
milki	awesome	00:07
	so that is not correct	00:07
	:P	00:07
coco89\|away	~/Project/GoldTag	00:07
milki	um	00:07
	coco89\|away: please read i think either the 1st or 2nd chapter of !book	00:08
gitinfo	coco89\|away: 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	00:08
milki	o	00:08
	or maybe 3	00:08
	since thats about branches	00:08
coco89\|away	milki: Okay I will when i get enough time	00:08
	← danishkhan left	00:08
	→ mezner joined	00:08
coco89\|away	milki: what's it complaining about anyway	00:08
	← ScottE_ left	00:08
coco89\|away	what am i doing wrong	00:08
milki	your remote has new commits that you dont have	00:09
	← jjore left	00:09
	→ diegoviola joined	00:09
milki	if you push, you will lose those commits	00:09
coco89\|away	milki: my remote repository has new files you mean?	00:09
milki	no	00:09
	commits	00:09
coco89\|away	milki: what do you mean new commits	00:09
milki	do you know what a commit is?	00:09
RandalSchwartz	it's what your girlfriend is looking for. :)	00:10
	→ CCB2013 joined	00:11
	← francisfish left	00:11
	← dukedave left	00:12
	← Adys left	00:12
milki	pybokeh: please dont pm me	00:12
	→ francisfish joined	00:12
	→ Adys joined	00:12
coco89\|away	milki: so there's a snapshot in my remote repository how does that work	00:12
	→ danishkhan joined	00:13
milki	coco89\|away: this is going to require a lot of babying. your time will be better spent going through the chapter in the book first	00:13
coco89\|away	milki: but i understand what a snapshot is but I don't get how there's a comit in the remote repo	00:13
	← chisler left	00:13
milki	coco89\|away: !gitt uses the snapshot terminology	00:14
gitinfo	coco89\|away: A book designed to teach people about Git in a real world usage model. The book follows a fictional company as they implement and learn about Git. Covers all 21 standard Git commands. Available for free at http://cbx33.github.com/gitt/	00:14
milki	or not...	00:14
	coco89\|away: maybe its !parable	00:14
gitinfo	coco89\|away: 'The git parable' provides some good reasoning behind git. http://tom.preston-werner.com/2009/05/19/the-git-parable.html	00:14
	→ elplatt joined	00:15
	→ momomomomo joined	00:17
	→ josefig joined	00:18
	← josefig left	00:18
	→ josefig joined	00:18
	← taylorlapeyre left	00:18
	← josefig left	00:18
	→ josefig joined	00:19
	← daikichi left	00:19
	← theotherguy left	00:21
	→ chisler joined	00:21
	→ sraf joined	00:22
	← CCB2013 left	00:22
	← breck7 left	00:22
	← digitalfiz left	00:23
	→ CCB2013 joined	00:23
	← arnihr__ left	00:23
	← CCB2013 left	00:23
	→ arnihr__ joined	00:23
	dwmw2 → dwmw2_gone	00:24
	→ digitalfiz joined	00:24
	← jiangenj left	00:24
	→ xpen joined	00:25
	← variable left	00:25
	← Ziber left	00:25
	← Mez left	00:25
	→ yann\|work joined	00:26
	→ Davey joined	00:26
	→ Mez joined	00:26
	← Mez left	00:26
	→ Mez joined	00:26
	→ huoxito joined	00:27
	← xpen_ left	00:28
	← elplatt left	00:28
coco89\|away	milki: i dont understand cos i never pulled anything	00:30
	← hyperair left	00:30
	→ elplatt joined	00:30
	← elplatt left	00:30
	← yshh left	00:30
milki	coco89\|away: right. the hint suggests that you do pull	00:30
	→ primitive joined	00:31
	→ variable joined	00:31
	← gerard0 left	00:33
	← goshawk left	00:33
	← yann\|work left	00:33
	← lacrymology left	00:34
	→ aartist joined	00:35
	→ ojacobson_ joined	00:35
coco89\|away	milki: why?	00:36
	milki: how can something on github be more recent than whatevers on my local machine?	00:36
	← a-l-e left	00:36
	→ yann\|work joined	00:36
	→ gerard0 joined	00:36
	← pastime left	00:37
milki	coco89\|away: because someone else updated it?	00:37
	← pybokeh left	00:37
coco89\|away	milki: it's just me here	00:37
	→ haaja joined	00:38
	← jMCg left	00:38
	→ Ziber joined	00:38
coco89\|away	milki: unless this is what happens when you change a file on the GUI version on github	00:38
milki	that definitely changes the remote repository	00:38
	did you not think it would?	00:38
	← ojacobson left	00:38
coco89\|away	milki: I see, now that I think of it, make sense	00:38
	lol	00:38
milki	o.O	00:39
	→ chuckharmston joined	00:39
	→ lacrymology joined	00:39
	← segher left	00:39
coco89\|away	milki: so what do i write just 'git pull' ?	00:40
	or git pull origin master	00:40
	so from origin -> master	00:40
	← bitkiller left	00:40
milki	no, origin is the name of your remote	00:40
	← barneybook\|2 left	00:40
milki	coco89\|away: pull = fetch + merge	00:40
	← jonshea left	00:40
milki	coco89\|away: so that would fetch stuff from the origin repo	00:40
	and attempt to automerge your current branch	00:40
coco89\|away	yeh that's what we wanted right?	00:41
milki	thats what the hint suggests	00:41
coco89\|away	so we need to do	00:41
	git pull origin master ?	00:41
EugeneKay	!pull4	00:41
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	00:41
milki	o, and that	00:41
coco89\|away	so what do i do	00:42
	just 'git pull'?	00:42
	← marcdel left	00:42
	← huoxito left	00:42
	→ marcdel joined	00:43
	← Esa\|iPhone6 left	00:43
milki	that will attempt to do the default stuff	00:43
	coco89\|away: !tas	00:43
gitinfo	coco89\|away: [!just_ask] You can just ask your question. If anybody knows the answer, they will answer soon (most of the time)	00:43
milki	coco89\|away: !tias	00:43
gitinfo	coco89\|away: Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://gitolite.com/1-basic-usage/tias.html may help with git-specific TIAS.	00:43
milki	why in the world did tas complete tot just_ask o.O	00:44
	← knightshade left	00:44
	← kmels left	00:44
	← chuckharmston left	00:44
	→ huoxito joined	00:44
EugeneKay	t_as	00:44
	← chisler left	00:45
	← infynyxx left	00:45
milki	ignores non-alpha characters?	00:46
coco89\|away	milki: git pull origin master	00:46
	and then git push origin master	00:46
milki	coco89\|away: no	00:46
coco89\|away	oh?	00:46
	well something happened	00:46
milki	yes, something happened and probably put you in a strange state	00:47
coco89\|away	what state?	00:47
	← oriba left	00:47
milki	which i dont actually know how to fix o.O	00:47
coco89\|away	sigh	00:47
milki	coco89\|away: something about FETCH_HEAD	00:47
	← ozette left	00:47
coco89\|away	:/	00:47
	→ segher joined	00:47
coco89\|away	see this is what happens	00:48
	when you don't tell me exactly what to do	00:48
	lol	00:48
milki	yup	00:48
	arent you having fun	00:48
	:P	00:48
coco89\|away	no	00:48
milki	well i am	00:48
	-.-	00:48
coco89\|away	_ikke_: hi	00:48
	Merge branch 'master' of https://..	00:49
	that's what it says now	00:49
	instead of a commit name	00:49
	→ xiukun joined	00:50
	← xiukun left	00:50
	→ xiukun joined	00:50
	← xiukun left	00:52
	→ jMCg joined	00:52
	← jMCg left	00:52
	→ jMCg joined	00:52
	← terryjfundak left	00:53
	← simesy left	00:54
	← mikef left	00:54
	→ simesy joined	00:54
coco89\|away	well, i've learnt nothing here	00:55
	→ pastime joined	00:55
coco89\|away	other than i screwed something up	00:55
	lol	00:55
	→ bitkiller joined	00:56
	→ knightshade joined	00:56
	← nioncode left	00:57
	→ breck7 joined	00:57
	← dreiss left	00:57
	← jaequery left	00:58
	← dmarkham_ left	00:58
	← simesy left	00:58
	→ jaequery joined	01:00
	→ simesy joined	01:01
	→ deleff joined	01:01
	← coco89\|away left	01:04
milki	coco89\|away: told you you should have read the book	01:04
	← eletuchy left	01:05
	← rrxtns left	01:05
	← redrocket left	01:06
	← shennyg__ left	01:06
	← czpython left	01:07
	→ dukedave joined	01:07
	→ shane joined	01:07
	shane → Guest9419	01:08
	← elixir_ left	01:10
	→ weiyang joined	01:10
	→ kerrick joined	01:10
	→ yshh joined	01:11
	← DonRichie left	01:11
	← Noldorin left	01:11
	← lacrymology left	01:12
	← alvaro_o left	01:12
	← momomomomo left	01:12
	← heroux left	01:13
	→ momomomomo joined	01:13
	→ R1KRDO joined	01:13
	→ heroux joined	01:14
	→ Targen joined	01:15
	→ czpython joined	01:20
	→ natechan joined	01:21
	← JStoker left	01:22
	→ jonshea joined	01:23
	← dukedave left	01:23
	→ m0ltar joined	01:23
	← mariusso left	01:25
	→ maxmanders joined	01:26
	← m0ltar left	01:26
	← penguin1_ left	01:26
	→ Esa_ joined	01:27
	← darnir left	01:28
	← Guest9419 left	01:29
	→ maxmande_ joined	01:29
	→ mitz joined	01:29
	→ Goplat joined	01:29
	→ eletuchy joined	01:29
	← primitive left	01:30
	← batrick left	01:30
	← maxmanders left	01:30
	← rakl left	01:31
	→ hyperair joined	01:34
	← kmels__ left	01:35
	→ batrick joined	01:36
	← titusjose__ left	01:36
	← hwrd\|work left	01:37
	← R1KRDO left	01:38
	← deleff left	01:38
	← marcdel left	01:38
	→ deleff joined	01:39
	← mgaunard_ left	01:40
	← cannonball left	01:40
	← claudiu_____ left	01:41
	→ acrocity_ joined	01:41
	← deleff left	01:41
	← maxmande_ left	01:42
	→ maestrojed joined	01:42
	← eletuchy left	01:43
	← acrocity left	01:43
	acrocity_ → acrocity	01:43
	→ barneybook joined	01:43
	→ rakl joined	01:44
	→ eliasp joined	01:47
	→ jason237 joined	01:47
	→ JStoker joined	01:48
	JStoker → Guest23108	01:48
	→ coco89\|away joined	01:48
coco89\|away	can i add two repositories	01:48
	← josefig left	01:49
imMute	uh?	01:49
	like mathematically add them together??	01:49
	← BSaboia left	01:50
RandalSchwartz	1 + 1 = 2 :)	01:50
coco89\|away	yes	01:50
RandalSchwartz	how are you talking when you're "\|away"?	01:50
	is that why you're so quiet? :)	01:50
coco89\|away	hueheuhuheue	01:51
	← Chaser_ left	01:51
	→ juvenal joined	01:51
coco89\|away	no really, how do i add another repository	01:51
	I've done git init on a different directory	01:52
	so now I have two masters	01:52
	if I do git add .	01:52
	→ dhruvasagar joined	01:52
coco89\|away	and then git push how does it know where to push it to	01:52
milki	coco89\|away: see chapter 2.5	01:53
	← xpen left	01:53
	← FatDarrel left	01:53
coco89\|away	noooo	01:54
	milki: you're going to break everything again by telling me to read stuff	01:54
	lol	01:54
	how do i unstage some files	01:55
	→ mithro joined	01:55
imMute	coco89\|away: 'git status' tells you exactly how to	01:55
	← danishkhan left	01:57
	← ojacobson_ left	01:57
	← bitkiller left	01:57
	← phillijw left	01:57
coco89\|away	thanks	01:57
RandalSchwartz	a man with two masters can never be happy	01:58
coco89\|away	it gives an error when i put	01:58
	git reset HEAD .	01:58
	fatal ambiguous argument HEAD	01:58
RandalSchwartz	you have too much HEAD	01:59
	← maestrojed left	01:59
imMute	coco89\|away: did you forget the "--" ?	01:59
coco89\|away	: unknown revision or path	01:59
RandalSchwartz	why are you resetting?	01:59
	what script are you following?	01:59
coco89\|away	cos i got the LF CRLF warnings again	01:59
	→ tinti joined	02:00
	← FernandoBasso left	02:00
RandalSchwartz	are you on windows?	02:00
coco89\|away	yes	02:00
	← engkur left	02:00
coco89\|away	so what do i do git reset --head ?	02:00
imMute	coco89\|away: read what 'git status' says	02:01
	← mezner left	02:01
RandalSchwartz	what are you TRYING to do	02:01
	→ Transcended joined	02:01
coco89\|away	RandalSchwartz: I did git add .	02:01
	and it was like 'warning: the file LF will be replaced by CRLF'	02:01
	to a load of files	02:01
	← scorphus left	02:01
coco89\|away	The file will have its original line endings in the working directory	02:02
	so i want to take them away from the staging area	02:02
	which apparently	02:02
	is meant to be git reset HEAD .	02:02
	right??	02:02
Transcended	Hey all, I am having some serious difficulty cloning a git repository that I have just been given permission to... But when I try git clone repoaddress I'm getting fatal: https://github.com/PlanningSight/PlanningSight.git/info/refs not found: did you run git update-server-info on the server?	02:02
	→ engkur joined	02:03
Transcended	tried googling this to no avail	02:03
coco89\|away	RandalSchwartz?	02:03
	← Guest23108 left	02:03
	→ Guest23108 joined	02:03
	→ R1KRDO joined	02:04
RandalSchwartz	Transcended - are you sure you're supposed to be using http[s] protocol?	02:04
	and not git?	02:04
coco89\|away	RandalSchwartz dude?	02:05
RandalSchwartz	and if so, did the repo owner run update-server-info?	02:05
	→ marcdel joined	02:05
	← thiago left	02:05
Transcended	RandalSchwartz: That is the address straight from github	02:05
RandalSchwartz	coco89\|away - I'm not a windows user. I'm not sure why you are doing that.	02:05
coco89\|away	RandalSchwartz: because this http://youtu.be/3nSoTC3GHn8?t=7m6s	02:05
RandalSchwartz	wait - why are you cloning from github via https	02:05
Transcended	RandalSchwartz: how does the repo owner do that?	02:05
RandalSchwartz	use git protocol!	02:06
Transcended	RandalSchwartz: I'll try	02:06
RandalSchwartz	get yourself a keypair, and become a real person on github :)	02:06
	RandalSchwartz almost said "get yourself a pair"...	02:06
	offby1 conses up a pair	02:06
Transcended	bah	02:06
coco89\|away	why is everyone so clueless about git reset HEAD <file>?	02:06
Transcended	RandalSchwartz: that worked hah	02:06
	→ tmcmahon joined	02:06
RandalSchwartz	there ya go	02:06
coco89\|away	http://youtu.be/3nSoTC3GHn8?t=7m6s	02:07
	http://youtu.be/3nSoTC3GHn8?t=7m6s	02:07
RandalSchwartz	case CLOSED!	02:07
	coco89\|away - I never have your problems, so I can't help you	02:07
	please stop presuming I'm just ignoring you	02:07
	→ luc_ joined	02:07
RandalSchwartz	sometimes, there are things I don't know	02:07
coco89\|away	Ah I see	02:07
	→ BSaboia joined	02:07
coco89\|away	sure thanks !	02:08
	→ infynyxx joined	02:08
	Guest23108 → JStoker	02:09
	→ sonicblud joined	02:09
	→ erkules_ joined	02:09
	← sonicblud left	02:09
	→ jayrobot joined	02:10
coco89\|away	someone help?	02:10
	← whowantstolivefo left	02:10
	gitinfo set mode: +v	02:10
coco89\|away	jayrobot hi i need help	02:11
	← Layke left	02:11
imMute	coco89\|away: did you happen to start a rebase?	02:12
	← erkules left	02:12
imMute	or checkout a version by SHA?	02:12
coco89\|away	imMute: i dont think so	02:12
	→ hwrd\|work joined	02:12
coco89\|away	what's that	02:12
	→ jimmy_lo joined	02:13
Transcended	Is there a command I can use to entirely overwrite the current state of a repo with my current working copy?	02:14
	rather than individually removing files?	02:14
	← tinti left	02:14
	← matthewpucc left	02:14
	← breck7 left	02:14
RandalSchwartz	Transcended - you mean make the workdir the same as your HEAD commit?	02:15
	→ brent joined	02:15
RandalSchwartz	because you've made mistakes in editing that you want to forget?	02:15
Transcended	no sir	02:16
	I want to overwrite the remote repo with my working dir	02:16
RandalSchwartz	wait - are you pushing to a non-bare repo?	02:16
Transcended	yes	02:16
RandalSchwartz	don't	02:16
	!non-bare	02:16
gitinfo	[!bare] an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/	02:16
Transcended	wait	02:16
RandalSchwartz	so don't	02:16
	just don't	02:16
Transcended	maybe i dont know what you mean by non-bare	02:16
RandalSchwartz	if it has a workdir, it's non-bare	02:16
	go read that URL	02:16
Transcended	I mean my local working copy	02:17
	I just moved some files around	02:17
RandalSchwartz	"overwrite the remote repo"	02:17
	as in "push a commit to a bare repo"?	02:17
Transcended	ughh	02:17
RandalSchwartz	or "change some checked out files"	02:17
	if there are checked out files, it's a non-bare repo	02:17
	and in that case, DON'T	02:17
Transcended	its a fresh repo	02:18
	that was just created tonight	02:18
RandalSchwartz	fresh doesn't matter	02:18
Transcended	that link isnt even loading fyi	02:18
RandalSchwartz	does it have a workdir or not?	02:18
Remram	ah, so it's not my wifi breaking up!	02:18
RandalSchwartz	are there files checked out or not	02:18
	← jimmy_lo left	02:18
Transcended	i just cloned the repo	02:18
RandalSchwartz	you're being nonresposnive	02:18
	I'm about to stop helping	02:18
	answer my friggin question	02:18
Remram	RandalSchwartz: this guy doesn't no, so it's obviously non-bare	02:18
RandalSchwartz	yeah, likely non bare	02:19
Transcended	if I just cloned the repo, that would mean files are checked out, no?	02:19
Remram	but I think the main problem here is like you have no idea what git is	02:19
RandalSchwartz	it doesn't meant that	02:19
	that's why you're not answering my question	02:19
	ARE THERE INDIVIDUAL FILES IN THE REMOTE REPO	02:19
	or just the equivalent of the .git dir	02:19
	← mmlac left	02:19
Transcended	yes, there are files	02:19
RandalSchwartz	yeah, non-bare	02:19
Transcended	it's not empty	02:19
RandalSchwartz	so DON'T	02:19
	that's broken	02:19
	always push to bare repos	02:19
	→ abackstrom joined	02:20
RandalSchwartz	never non-bare	02:20
	→ SinnerNyx joined	02:20
Transcended	i honestly have no idea what you mean by that	02:20
RandalSchwartz	for example, when you push to github, there aren't "files" that get updated	02:20
	just the .git dir equivalent	02:20
	you need to set up something similar	02:20
coco89\|away	Okay Guys what do i do in order to create another repository	02:20
RandalSchwartz	if you intend to push	02:20
coco89\|away	I've already done it in github	02:20
	what do i do	02:20
Transcended	hmmm	02:21
coco89\|away	in	02:21
	git	02:21
	bash	02:21
	← gitinfo left	02:21
RandalSchwartz	whoa... gitinfo... come back!	02:21
coco89\|away	anyone??/	02:22
Transcended	So Randal... could you propose the best solution for this. Someone made an initial commit with a bunch of files that I want to move into a single directory	02:22
	I cloned the repo	02:22
	moved the files in my local copy	02:22
SinnerNyx	hello, I hope no one minds if I ask an msysgit question here... Can I change the location for the ~/.git/.gitignore_global file? I have tortoisegit installed and right now it looks for the file at ~/.git, but when I create a ~/.git folder it adds the untracked file icon overlay to all files in my user directory...	02:22
	← Guest58866 left	02:22
	← jkremser left	02:22
coco89\|away	RandalSchwartz hi	02:22
	← heytrav left	02:22
	→ Layke joined	02:23
garbagecollectio	hey	02:23
Transcended	Or anyone?	02:23
garbagecollectio	I'm on a commit	02:23
	on a shaw id	02:23
	i made all these changes	02:23
	→ xiukun joined	02:24
RandalSchwartz	push to a branch other than "master" in that repo	02:24
	→ heytrav joined	02:24
coco89\|away	RandalSchwartz: who are you talking to?	02:24
RandalSchwartz	for example, I have used "push origin master:from-laptop/new-master"	02:24
	← RoboTeddy left	02:25
RandalSchwartz	then go to that repo, and "git checkout master; git merge from-laptop/new-master"	02:25
	that'll update it with those commits	02:25
coco89\|away	oh my fuck.	02:25
RandalSchwartz	just don't push to the checked out branch	02:25
	→ rideh joined	02:25
RandalSchwartz	I'm talking to Transcended	02:25
coco89\|away	ok	02:25
RandalSchwartz	because he makes sense :)	02:25
	→ danishkhan joined	02:25
coco89\|away	i'm making perfect sense.	02:25
RandalSchwartz	you are on windows, I have no clue	02:25
	← rakl left	02:26
SinnerNyx	RandalSchwartz: I'm in windows :P. but to make a git repo it's just git init no?	02:26
coco89\|away	SinnerNyx: yes but apparently it goes schitzo when i try to amke two	02:26
	← vedant left	02:27
SinnerNyx	what is the exact error? Or does it say "Error: going shitzo"?	02:27
	RandalSchwartz chuckles at "shitzo"	02:28
	← gwty left	02:28
	→ ehsan joined	02:29
	offby1 recalls the sainted George Carlin	02:29
coco89\|away	SinnerNyx: just complains about amibuguous argument HEAD	02:29
	← b1rkh0ff left	02:29
offby1	"He had a s*** fit? Glad I wasn't there"	02:29
	→ Vampire0 joined	02:29
cmn	what's the actual message?	02:29
coco89\|away	cmn: ambiguous argument HEAD. unknown revision or path not in the working tree	02:30
cmn	does HEAD point to a commit?	02:30
	→ redrocket joined	02:31
coco89\|away	cmn: not sure	02:31
	I guess?	02:31
SinnerNyx	coco89\|away: are you in a folder that is a descendant folder of an existing git repository	02:31
	?	02:31
	→ stliu joined	02:31
RandalSchwartz	git branch -a \| grep HEAD	02:31
cmn	don't guess, check	02:32
RandalSchwartz	if there's more than one, that's odd	02:32
cmn	rev-parse HEAD, does it complain?	02:32
RandalSchwartz	^^^^	02:32
	← pastime left	02:32
SinnerNyx	does git stand for something?	02:32
cmn	no	02:32
RandalSchwartz	it's like perl... it doesn't stand for anything	02:32
	← BSaboia left	02:33
	SeoZ-work[AWAY] → SeoZ	02:33
coco89\|away	rev-parse command not found	02:33
	← Vampire0_ left	02:33
	→ gitinfo joined	02:33
	ChanServ set mode: +o	02:33
SinnerNyx	coco89\|away: git rev-parse HEAD	02:33
	← ISF_ec09 left	02:33
SinnerNyx	true. my nephew just asked	02:33
	→ jast joined	02:34
	← ehsan left	02:34
	→ ehsan joined	02:34
	← tvw left	02:35
	← SBoudrias left	02:35
	← tmcmahon left	02:36
	← czpython left	02:36
	→ gwty joined	02:40
	← ehsan left	02:41
	← kpreid left	02:42
	→ leeN joined	02:42
	→ pastime joined	02:42
	→ kpreid joined	02:43
	← garbagecollectio left	02:43
	→ war2mac joined	02:44
	← rryoumaa left	02:44
	← hyperair left	02:45
	→ hyperair joined	02:46
	→ normanrichards joined	02:47
	→ Gentlecat joined	02:48
	← his_shadow left	02:50
	← engkur left	02:50
	→ ISF_ec09 joined	02:51
	← Underbyte left	02:51
	← coco89\|away left	02:53
	→ Dougie187 joined	02:53
	→ eletuchy joined	02:53
	→ garbagecollectio joined	02:53
	→ thiago joined	02:54
	→ eletuchy_ joined	02:57
	← nigelb left	02:58
	← eletuchy left	02:58
	→ averiso joined	02:58
	← danishkhan left	02:59
	→ danishkhan joined	02:59
	← bartman left	03:00
	← knightshade left	03:02
	→ bartman joined	03:02
	← nathanr left	03:03
	← R1KRDO left	03:03
	← danishkhan left	03:04
	→ danishkhan joined	03:04
	← orafu left	03:05
	→ nigelb joined	03:06
	→ whowantstolivefo joined	03:07
	← danishkhan left	03:07
	→ danishkhan joined	03:07
	→ xpen joined	03:08
	→ ehsan joined	03:08
	→ orafu joined	03:08
	← paddymahoney left	03:10
	→ Froker joined	03:11
	→ pigeon joined	03:12
	← momomomomo left	03:12
	← Transcended left	03:12
	← Imaginativeone left	03:12
	→ JeremyWei joined	03:13
	→ momomomomo joined	03:13
	→ undecim joined	03:15
	← hyperair left	03:15
	→ preflex_ joined	03:16
	→ PanPan joined	03:17
	→ hyperair joined	03:17
	← preflex left	03:17
	preflex_ → preflex	03:17
	← gwty left	03:18
	→ maestrojed joined	03:18
	← Kje10L left	03:20
Dougie187	What use do tags have in git?	03:21
	← ehsan left	03:22
	→ Bombe joined	03:22
Dougie187	Is it just a way to name a specific commit with a version number, or something similar to that?	03:22
	→ candlejackson joined	03:22
Dougie187	because you can't actually checkout the tag, right?	03:22
imMute	you can checkout the tag, but then you're in a detached head state (with all those caveats)	03:23
	but yeah, they're just tagging a specific commit with some extra data	03:23
	← dvide left	03:24
Dougie187	So, to really use them, you would want to do something like "git show tagname; git checkout hash_from_tag" ?	03:24
	→ R1KRDO joined	03:24
imMute	no, you can checkout the tag directly	03:24
Dougie187	Oh ok	03:24
	but it's detached as you said.	03:24
imMute	so is checking out the hash	03:25
Dougie187	I get a weird thing that says (no_branch) when I do git branch	03:25
	Oh really?	03:25
	ok	03:25
	← Esa_ left	03:25
cmn	!detached	03:27
gitinfo	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	03:27
	← x82_nicole left	03:28
Dougie187	!reattach	03:28
gitinfo	Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue	03:28
Dougie187	Thanks	03:29
	← Dougie187 left	03:29
	→ sgambino joined	03:30
	← danishkhan left	03:32
	→ gwty joined	03:33
	→ security joined	03:34
	← megha left	03:35
	→ engkur joined	03:36
	→ x82_nicole joined	03:37
	tashi → tashi_	03:39
	← war2mac left	03:39
	← gwty left	03:40
	→ rchavik joined	03:43
	← acharlieh left	03:45
	→ pppaul joined	03:47
	→ luc__ joined	03:47
	← luc_ left	03:47
	→ jimmy_lo_chien_f joined	03:47
jimmy_lo_chien_f	Hi, does git format-patch create patches by date?	03:49
	← x82_nicole left	03:49
jimmy_lo_chien_f	I find that older one may has small number than newer	03:49
	For example, 2012/01/02 may be listed after 2013/01/02	03:50
	← fly9 left	03:50
SamB	it does it by commit order	03:51
	(which doesn't always match the dates)	03:51
	← madewokherd left	03:51
	→ sangi joined	03:52
jimmy_lo_chien_f	So I have to sort them before I patch them.	03:52
garbagecollectio	how do i split a string on newline or space	03:52
	it finds a space or a newline	03:52
	→ x82_nicole joined	03:53
	← Wulf left	03:53
	→ abonstu joined	03:54
	→ gwty joined	03:54
	← causasui left	03:56
	→ jazz264 joined	03:56
	gitinfo set mode: +v	03:57
	→ revyd joined	03:58
	→ causasui joined	03:59
	← causasui left	03:59
	→ causasui joined	03:59
	← Bombe left	04:01
	← Froker left	04:01
	→ Froker joined	04:01
	→ Joe5 joined	04:01
	← eletuchy_ left	04:02
	→ sgambino_ joined	04:02
	→ acharlieh joined	04:02
	← dhruvasagar left	04:02
	← chrisgeorge left	04:02
	→ Error404NotFound joined	04:02
	← sgambino left	04:02
	sgambino_ → sgambino	04:02
	→ danishkhan joined	04:03
	→ dhruvasagar joined	04:04
	← abonstu left	04:04
	→ eletuchy joined	04:05
	→ sgambino_ joined	04:05
	← dous left	04:05
	← lahwran left	04:06
	→ lyddonb joined	04:06
	← sgambino left	04:07
	sgambino_ → sgambino	04:07
	→ dous joined	04:07
	→ Bombe joined	04:08
	← eletuchy left	04:10
	→ war2mac joined	04:10
	← war2mac left	04:11
	→ war2mac joined	04:11
	← Saby left	04:11
	← danishkhan left	04:11
	← dhruvasagar left	04:11
	→ lahwran- joined	04:13
	← R1KRDO left	04:13
	→ mezner joined	04:14
	→ tharvey_home joined	04:14
	→ beautifulmind joined	04:15
	→ cschneid joined	04:16
tharvey_home	I've created a topic branch for a project and now I want to rebase against a specific tag in that projects upstream master - how do I rebase to a tagged revision of master?	04:17
	→ tjbiddle joined	04:18
cmn	tell rebase where you want to rebase on top of	04:19
tharvey_home	is 'where' a tag or do I need to give it the hash the tag represents?	04:19
cmn	either works, just like every other time	04:20
	← acharlieh left	04:20
tharvey_home	so I pulled master a few weeks back, created a topic branch and worked on it, now I want to pull master again to a specific tag and rebase on it. So I would 'git checkout topic', then 'git rebase master' but how do specify the tag?	04:21
	→ squisher joined	04:21
cmn	what do you mean pull to a specific tag?	04:22
nevyn	tharvey_home: none of that has fetched.	04:22
	→ rakl joined	04:22
nevyn	cmn: he wants to rebase to a particular tag and update master to that tag I think	04:22
cmn	you specify the tag the same way you specify any other reference	04:22
tharvey_home	oh... so once in my 'topic' branch I would 'git rebase <tagname>' ?	04:23
nevyn	cmn: however the way tharvey_home used the word "pull" seems to indicate a missunderstanding..	04:23
	tharvey_home: you can do that.	04:23
	→ Cassiel joined	04:23
tharvey_home	yes, quite likely a misunderstanding on my end	04:23
	→ jsallis joined	04:23
nevyn	!pull	04:23
gitinfo	pull=fetch+merge (or with flags/config also fetch+rebase). It is thus not the opposite of push in any sense. A good article that explains the difference between fetch and pull: http://longair.net/blog/2009/04/16/git-fetch-and-merge/	04:23
cmn	rebasing on a tag is exactly the same as rebasing on a branch	04:24
	tell rebase where it should start	04:24
tharvey_home	I don't think I used the term fetch - I started out with a 'git clone <upstream_project>', then 'git checkout mytopic', then made edits, now I want to rebase on the upstream project's more recent activity, but to a specific tag (which didn't exist when I cloned)	04:26
	I believe this is a common workflow but I still don't quite understand it	04:27
	← x82_nicole left	04:27
	← the_cybersphinx left	04:27
	← vidplace7 left	04:27
	← squentin left	04:27
	→ vidplace7 joined	04:29
nevyn	tharvey_home: I understand what you think you want to do but you've got it just a little wrong.	04:29
	→ squentin joined	04:30
cmn	fetch; the tag will most likely appear locally	04:30
nevyn	tharvey_home: so you want to fetch the upstream repo then rebase your work onto a specific tag.	04:30
cmn	if it doesn't ask explicitly, fetch origin --tags e.g.	04:30
tharvey_home	yes... from reading that link it seems pull is 'not' what I wanted to do on master	04:30
cmn	you don't want to do anything on master from what you're saying	04:31
nevyn	tharvey_home: fetching only upto some point in history is a bad idea and poorly supported.	04:31
	you fetch everything then just work with whatever tags/branches you care about.	04:32
	→ eletuchy joined	04:32
	→ JohnBat26 joined	04:33
	← jsallis left	04:34
tharvey_home	so leave master alone, and in my topic branch just 'git fetch <tag>; git rebase'?	04:35
	→ longbeach joined	04:35
cmn	no, just run git fetch	04:35
	the new tags will be downloaded	04:36
	then rebase as usual	04:36
	→ curvedmark joined	04:36
	→ danishkhan joined	04:38
tharvey_home	i still don't understand where/how to specify the tag then?	04:38
cmn	to rebase	04:38
tharvey_home	sorry I'm very confused...	04:38
	← tjbiddle left	04:39
	← jonshea left	04:40
	← Villadelfia left	04:40
	→ russt joined	04:40
tharvey_home	ok... so workflow is: git clone <remote>; git checkout mytopic; edit files and days/weeks later, git fetch origin; git rebase <tag>	04:40
	I 'think' that worked... hopefully I understood	04:41
	← heroux left	04:41
	← SeySayux left	04:41
cmn	yes	04:41
	but you shouldn't need to rebase in most cases, anyway	04:42
tharvey_home	cmn, nevyn thanks so much!	04:42
	← Error404NotFound left	04:42
nevyn	tharvey_home: I strongly reccomend !book or !gcs	04:42
gitinfo	tharvey_home: 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	04:42
	tharvey_home: [!concepts] "Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs/	04:42
nevyn	or the git for ages 4 and up video from this year's linux conf...	04:43
	http://mirror.linux.org.au/pub/linux.conf.au/2013/mp4/Git_For_Ages_4_And_Up.mp4	04:43
tharvey_home	ya... I've been fumbling around for a while, time to re-read and look for some new books to fix some of my misunderstandings - those look good	04:43
nevyn	the video is actually quite good.	04:44
	→ pitzips joined	04:44
	← while1eq1 left	04:44
	→ Saby joined	04:44
	← Saby left	04:44
	→ Saby joined	04:44
	→ jbr joined	04:44
milki	nevyn: perhaps add that to the talk trigger?	04:44
	← jbr left	04:44
nevyn	milki: I was just thinking that..	04:45
	→ heroux joined	04:46
	→ Villadelfia joined	04:46
	← danishkhan left	04:46
	→ pur3forlyphe joined	04:47
	→ SeySayux joined	04:47
	← undecim left	04:49
	← war2mac left	04:50
	← gwty left	04:50
	→ paddymahoney joined	04:50
	→ war2mac joined	04:51
	← garbagecollectio left	04:51
	→ Error404NotFound joined	04:56
	→ pretty_function joined	04:57
	→ NimeshNeema joined	04:57
	← engkur left	04:58
	→ engkur joined	04:58
	→ Raging_Hog joined	05:01
	← m0viefreak left	05:01
	← natechan left	05:01
	← momomomomo left	05:02
	→ gwty joined	05:02
	← mezner left	05:04
	→ rikko joined	05:05
	→ jdolitsky joined	05:05
	← squisher left	05:06
	→ g0bl1n joined	05:07
	← g0bl1n left	05:07
	→ g0bl1n joined	05:07
	← war2mac left	05:07
	← pretty_function left	05:08
	→ vedant joined	05:08
	→ cutaway joined	05:17
	→ yacks joined	05:17
	→ breck7 joined	05:17
	← yacks left	05:18
	→ shreya joined	05:18
	→ yacks joined	05:18
	← macbr left	05:19
	← Gentlecat left	05:21
	← DEac- left	05:22
	→ DEac- joined	05:23
	→ dsdeiz joined	05:23
	lahwran- → lahwran	05:24
	→ hoschi joined	05:25
	→ o] joined	05:26
	→ thufir_ joined	05:27
o]	hello	05:27
nevyn	!hi	05:27
gitinfo	[!welcome] Welcome to #git, a place full of helpful gits. If you have a question, feel free to just go ahead and ask—somebody should answer shortly. For more info on this channel, see http://jk.gs/git/ Take backups (type !backup to learn how) before taking advice.	05:27
o]	not sure if it is the right place to ask, but I just have set up a new git hub repository, but I am getting this when I try to push:	05:27
	Permission denied (publickey).	05:27
	fatal: The remote end hung up unexpectedly	05:27
milki	o]: you dont have ssh keys setup properly	05:28
o]	bah	05:28
nevyn	#github may be more appropriate...	05:28
milki	except theres something ranting there right now	05:28
o]	I can commit to other remote repositories (dreamhost and googlecode) without any hassle	05:28
nevyn	did you copy your pubkey into the github ui?	05:28
	double check it's not broken	05:29
milki	o]: double check by sshing into github	05:29
o]	ok	05:29
nevyn	that's a easier check...	05:29
o]	same error	05:29
milki	ya, so try adding your ssh key again	05:29
o]	how?	05:29
	(I am using git bash on windows)	05:30
milki	if you ask how...that means you didnt do it before?	05:30
	← vedant left	05:30
	→ shreya_ joined	05:30
	← shreya_ left	05:30
	→ rurufufuss joined	05:30
o]	I just added a remote repo to my local repo. when I tried to push, it asked me to accept a RSA fingerprint...	05:30
	and after that, I got this damned error	05:30
milki	o]: https://help.github.com/articles/generating-ssh-keys	05:31
	step 3	05:31
	← Davey left	05:31
o]	thank you :)	05:31
	→ nkambo joined	05:32
	← heroux left	05:33
o]	Hi thinlizzy! You've successfully authenticated, but GitHub does not provide she	05:35
	ll access.	05:35
	:))))))))))))))))))))))))	05:35
	→ heroux joined	05:35
	← dsdeiz left	05:35
milki	yay	05:35
o]	https://github.com/thinlizzy/fileutils	05:36
	wheeeee	05:36
	→ curvedma_ joined	05:36
	← curvedma_ left	05:36
	→ curvedma_ joined	05:36
	← curvedma_ left	05:36
	← Layke left	05:37
	→ curvedma_ joined	05:37
	← curvedma_ left	05:37
	← rikko left	05:39
	← Saby left	05:39
	← curvedmark left	05:39
	← wrouesnel left	05:41
	→ wrouesnel joined	05:41
	← heroux left	05:43
	→ Davey joined	05:43
	← Tommy[D]_ left	05:46
	→ heroux joined	05:46
	← aartist left	05:47
	→ Cache_Money joined	05:49
	→ artnez joined	05:50
Cache_Money	I'm on a new branch, made some changes I like, committed them, made some changes I don't like... how do I get back to the last commit? git revert?	05:51
SinnerNyx	Cache_Money: I think git reset along with a refex	05:51
	→ mnathani joined	05:51
nevyn	did you commit the changes you don't like?	05:51
	→ rjc joined	05:51
mnathani	!deploy	05:52
gitinfo	Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html	05:52
Cache_Money	yes, i committed the changes i liked	05:52
nevyn	did you commit the changes you don't like?	05:52
SinnerNyx	o ya. if you didnt then git reset --hard HEAD i think will do the trick	05:52
nevyn	it will delete all changes to the working directory.	05:53
Cache_Money	no, i haven't committed the changes I don't like	05:53
SinnerNyx	Cache_Money: that's what you want right? to delete the changes to the working directory?	05:53
nevyn	git reset --hard HEAD will delete all the changes in the working directory and reset to your last commit	05:54
Cache_Money	nevyn: that's what I'm looking for. Thanks	05:54
nevyn	note I said delete.	05:54
	→ Layke joined	05:54
nevyn	Cache_Money: beware reset is one of the commands whcih actually delets in git...	05:55
	so the changes you don't like will be gone.	05:55
Cache_Money	that's what I was looking to do	05:55
	thanks	05:55
	→ buibex joined	05:55
	nevyn sometimes thinks that's what he wants too. then discovers that it was the right idea 3 days later..	05:56
	→ [dee] joined	05:57
	→ Wulf joined	05:59
	→ system64 joined	05:59
	→ Tommy[D]_ joined	05:59
	← [dee] left	05:59
	← system64 left	05:59
	← Error404NotFound left	06:03
	→ vedant joined	06:03
	→ Err404NotFound joined	06:04
	← donmichelangelo left	06:04
	← artnez left	06:05
	→ Dr4k3 joined	06:05
	→ donmichelangelo joined	06:06
	← cagedwisdom left	06:06
	← Dr4k3 left	06:07
	← kadoban left	06:12
	→ tonko joined	06:13
	← tonko left	06:13
	← toothrot left	06:15
	→ mark_ joined	06:15
	← Cassiel left	06:17
	← thufir_ left	06:17
	gitinfo set mode: +v	06:18
mark_	I hav a github account can someone please tell me how an i sync it with git command line?	06:18
	*can	06:19
nevyn	mark_: configure ssh.	06:23
	mark_: configure ssh keys more specifically	06:23
	configure github with those keys.	06:23
	← mikehaas763 left	06:23
nevyn	git clone <privateurlfromgithub>	06:23
	→ jceb joined	06:24
	→ dhruvasagar joined	06:24
	← skyxxx left	06:24
mark_	ohk tht sounds complicated but interesting , i'll try doing that :)	06:25
nevyn	mark_: it's probably better explained umm... here:	06:26
	https://help.github.com/articles/generating-ssh-keys	06:27
Wulf	Hi	06:27
mark_	:) thks	06:28
	← jceb left	06:28
Wulf	I found in the git repository of my work a commit which claims that I am the author, but I'm not. It's even my time zone in the time stamp. On github I see that one of my workmates pushed this one. How might this have happened? I don't assume that he faked as being me	06:29
	→ toothr joined	06:30
	← paddymahoney left	06:30
	→ jjore joined	06:30
	← rurufufuss left	06:31
	← Goplat left	06:31
nevyn	Wulf: you sure you're not looking at the associated merge?	06:33
	in your local history?	06:33
	← maestrojed left	06:33
Wulf	nevyn: there's a merge shortly after by my colleague, but the "wrong" commit is not a merge	06:34
	← huoxito left	06:35
	← vedant left	06:35
	→ vedant joined	06:36
	→ a-l-e joined	06:37
	→ elixir_ joined	06:37
Wulf	what's the difference between "author" and "committer"? I'm the author, he's the commmitter	06:39
rjc	Wulf: the author is the person (account/email to be precise) who writes the code in question	06:41
	Wulf: committer is the person who commits the changes	06:41
	Wulf: this is, e.g. in a situation when one receives a patch via email	06:42
Wulf	ah, makes sense.	06:42
	rjc: any idea why I'm listed as the author, when in fact I'm not?	06:42
	← normanrichards left	06:42
	← Katafalkas left	06:43
	→ maestrojed joined	06:43
rjc	Wulf: how have your received the changes?	06:43
	Wulf: btw, I've give a wrong egzample - in that scenario author an a committer would be the same	06:44
	← dhruvasagar left	06:44
Wulf	rjc: I received them via github	06:45
	← heroux left	06:45
rjc	Wulf: if committer pulls changes from another public repo and then commits the changes locally and then pushes them, others will see different author and commiter	06:45
	→ soee joined	06:46
	→ dhruvasagar joined	06:46
rjc	Wulf: have you pulled and merged?	06:46
Wulf	rjc: yes	06:47
rjc	Wulf: fetched and merged rather ;)	06:47
Wulf	pull without --rebase	06:47
	← pitzips left	06:47
Wulf	or is it possible that github magically changes the author?	06:48
rjc	Wulf: can you post a link? - hard to tell witout looking	06:48
Wulf	rjc: sorry, not allowed	06:48
nevyn	mark_: so what was your question ?	06:48
Wulf	my and my workmate's names both start with "J". I found another commit where both names start with "D"	06:49
mark_	have added the ssh key wanted to know how could i get the control of all the repositries(branches) in my github through command line :P	06:50
	→ heroux joined	06:50
	→ mihamina joined	06:50
	→ fRAiLtY- joined	06:51
	← vedant left	06:53
	← pur3forlyphe left	06:53
	→ k_sze[work] joined	06:53
	→ Wulf4 joined	06:53
	← Wulf left	06:54
	Wulf4 → Wulf	06:54
	← gwty left	06:54
nevyn	mark_: what do you mean by "control"	06:54
	→ pretty_function joined	06:55
	→ _zoom_ joined	06:55
	→ pantsman joined	06:55
	← pantsman left	06:55
	→ pantsman joined	06:55
	← _zoom_ left	06:55
mark_	i mean how do i use ssh (github does not give shell access , right ?) so how do i start branching and merging from my git shell	06:56
	→ gwty joined	06:56
nevyn	git shell?	06:56
	mark_: so you work with github as a remote.	06:56
	you make a local clone	06:56
	← pppaul left	06:56
	← Adlai left	06:57
	← beautifulmind left	06:58
	security → fire	06:59
mark_	what i meant was, would it add my github user as the author for all the further commits i make using the git command line ?	06:59
	→ pitzips joined	07:00
	← buibex left	07:01
	→ tPl0ch joined	07:01
	← donmichelangelo left	07:03
	← gwty left	07:03
	→ donmichelangelo joined	07:03
	← jjore left	07:03
	→ papegaaij joined	07:05
	← SamB left	07:05
	→ SamB joined	07:06
	← dharrigan left	07:06
	→ dharrigan joined	07:07
	← dharrigan left	07:07
	→ dharrigan joined	07:07
	→ arvind_khadri joined	07:08
	→ jceb joined	07:10
	→ Katafalkas joined	07:10
	← sangi left	07:11
	→ Adlai joined	07:12
	→ cbreak-work joined	07:12
	→ sangi joined	07:12
	→ rath joined	07:13
	← Targen left	07:13
	← jasom left	07:14
	→ Kronuz joined	07:15
	→ swiety_ joined	07:15
	→ gwty joined	07:16
	→ Squarism2 joined	07:20
	→ Squarism joined	07:20
	→ Gentlecat joined	07:21
	← mishok13 left	07:22
	← jceb left	07:23
	→ theotherguy joined	07:24
	→ BadDesign joined	07:26
	← BadDesign left	07:26
	→ BadDesign joined	07:26
	← gyxt_ left	07:26
	← breck7 left	07:27
	← gwty left	07:27
	→ SamB_ joined	07:27
	→ _Froker_ joined	07:28
	→ mabrand joined	07:28
	→ Noldorin joined	07:29
	← _Froker_ left	07:29
	← Froker left	07:29
	→ Froker joined	07:29
	← Cache_Money left	07:30
	← arand left	07:30
	→ rob_w joined	07:30
	← mabrand left	07:30
	← eletuchy left	07:30
	← SamB left	07:31
	← leeN left	07:31
nevyn	no.	07:31
	you'd need to configure your local git correctly with your name and email	07:32
	← swiety_ left	07:32
	← nilg` left	07:32
	← nilg left	07:32
	← Froker left	07:33
mark_	hmm , n suppose i have made some changes to master and create a new branch (which i want to commit) how do i avoid recommiting the changes made in master ?	07:33
	→ goshawk joined	07:34
nevyn	master is a branch you can't make a new branch in a branh EPARSEFAIL	07:34
	→ Froker joined	07:34
fuzzybunny69y	hey guys how do I add a file to a previous commit that isn't the most recent one?	07:35
mark_	ya i am nt making a new branch is master , i have aother branch , wheni try to commit this branch it also shows the files modified in master	07:35
	though this is not a problem	07:35
	but it increases the size of the commit unnecessarily	07:36
weiyang	I get a problem like this, for example, tag1 works, while current HEAD failed.	07:37
	while tag1 is merged into current branch	07:38
	← maestrojed left	07:38
	→ giallu joined	07:38
weiyang	so, in time line, tag1 is HEAD~5,	07:38
	while many commits before HEAD~5 is not on tag1 branch	07:38
	→ gwty joined	07:39
	→ war2mac joined	07:39
	→ mastro joined	07:39
	← alexwhitman left	07:39
	← mastro left	07:40
	← tharvey_home left	07:40
	→ mastro joined	07:41
	← thiago left	07:41
	→ alexwhitman joined	07:41
	→ netiul joined	07:41
mark_	how can we solve this ?	07:42
	→ thehandler joined	07:42
weiyang	fuzzybunny69y, git rebase -i	07:42
	→ skyxxx joined	07:43
mark_	<weiyang> do u have any idea how can i do that using the github interface itself ?	07:47
	→ flijten joined	07:47
weiyang	mark_, do what? I didn't understand your situation	07:47
	← xpen left	07:48
weiyang	mark_, btw, I am not familiar with github...	07:48
	speakman_ → speakman	07:48
	← unreal left	07:48
	← nkambo left	07:48
	→ anderse joined	07:49
	erkules_ → erkules	07:50
	→ Yougi joined	07:51
	← Noldorin left	07:53
	→ cebor joined	07:55
	← littlebearz left	07:56
	→ littlebearz joined	07:57
	← anderse left	07:58
	← weiyang left	07:58
	→ _Qman joined	07:59
	← elixir_ left	07:59
	→ unreal joined	07:59
	→ elixir_ joined	08:01
	← xiangfu left	08:02
	→ nkambo joined	08:03
	← hwrd\|work left	08:03
	→ jasom joined	08:03
	→ suy joined	08:03
	→ psoo joined	08:03
	→ xiangfu joined	08:04
	← shreya left	08:05
	→ tbosua joined	08:06
	← pitzips left	08:06
	← gwty left	08:07
	→ mishok13 joined	08:07
	→ YamakasY2 joined	08:07
	← mark_ left	08:08
	← _Qman left	08:08
	← Davey left	08:08
	← jdolitsky left	08:09
	← threesome left	08:11
	→ Xethron joined	08:11
	→ Sigma joined	08:11
	→ nilg joined	08:11
	→ _Qman joined	08:12
	→ thufir_ joined	08:12
	→ smirn0v joined	08:13
	← rideh left	08:13
	→ Cromulent joined	08:14
	← JeremyWei left	08:16
	→ pitzips joined	08:18
	→ LeMike joined	08:20
	→ alorence joined	08:22
	→ cibegtd joined	08:22
cibegtd	hi	08:22
gitinfo	cibegtd: 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.	08:22
	← goshawk left	08:23
	← pitzips left	08:23
	← g0bl1n left	08:23
cibegtd	is it possible to edit (change) a file with git diff branch0:afile branch1:afile	08:23
	this gives met readonly message	08:24
	this gives me readonly message	08:24
	← Cromulent left	08:24
cibegtd	my difftool is vimdiff	08:25
	→ ropo joined	08:25
	→ vy joined	08:25
vy	I have forked a repo, created a new branch, modified some files, committed changes and sent a pull request. The author of the upstream told me that some files are updated, and hence I need to update my repo and re-issue my changes. How can I do that? Do I need to create a new branch and repeat all my modifications from scratch? Isn't there a shortcut for that purpose?	08:26
cmn	cibegtd: a diff is a diff, you can change it if your editor allows, but it doesn't change the files	08:26
cibegtd	cmn: I need to compare the file in a branch in other branch and then change the file	08:27
cmn	do that, but the diff has nothing to do with changing the file afterwards	08:27
	→ tsdh joined	08:27
cmn	if you want to change a file in a branch, checkout that branch and commit a change	08:27
cibegtd	cmn: thanks	08:28
	→ sheerun joined	08:28
	→ maxmanders joined	08:31
mgedmin	I keep doing git branch newbranch; git commit -a -m "fixes that should be on newbranch"	08:33
	lb → bl	08:33
mgedmin	how can I teach myself to not do that any more?	08:33
	also, how do I swap the newbranch and oldbranch refs with minimum amount of pain?	08:33
	bl → Guest82959	08:33
	← pantsman left	08:34
	Guest82959 → lb	08:34
	← babilen left	08:34
	→ Cromulent joined	08:34
	← maxmanders left	08:36
osse	mgedmin: use git checkout -b newbranch to create and swtich to the new branch at the same time	08:39
	→ AccessGranted joined	08:40
	→ jargon joined	08:40
	→ maxmanders joined	08:42
	→ threesome joined	08:42
	← w0rmdrink left	08:45
	← simesy left	08:46
	→ simesy joined	08:46
	→ anderse joined	08:46
	→ JosefAssad joined	08:47
	→ buibex joined	08:48
	→ nicolash joined	08:49
	→ gwty joined	08:51
	← simesy left	08:51
	→ Dr4k3 joined	08:52
	→ _iron joined	08:55
	→ mastroWork joined	08:55
	← maxmanders left	08:55
	→ maxmanders joined	08:56
	← psoo left	08:57
	← maxmanders left	08:58
	→ mgaunard_ joined	08:58
	→ maxmanders joined	08:58
	← mastro left	08:58
	adhawkins-away → adhawkins	09:01
	← mgaunard_ left	09:03
	→ ferdev joined	09:04
	← anderse left	09:04
	→ psoo joined	09:04
	← fuzzybunny69y left	09:07
	← SicoAnimal left	09:07
	← Err404NotFound left	09:08
	→ YamakasY3 joined	09:08
	← ferdev left	09:08
	→ TheFuzzball joined	09:09
	← cibegtd left	09:09
	→ marhaban joined	09:09
	← juvenal left	09:10
	← marhaban left	09:10
	← psoo left	09:11
	← YamakasY2 left	09:11
	→ psoo joined	09:11
	→ spuz joined	09:13
	← mastroWork left	09:13
	← psoo left	09:13
	→ FernandoBasso joined	09:14
	← jdunck left	09:14
	→ cilly joined	09:14
	→ ferdev joined	09:16
	→ psoo joined	09:16
	→ simesy joined	09:17
	→ AdrienBrault joined	09:18
	← Phlogistique left	09:19
	→ SicoAnimal joined	09:19
	→ AdrienBr_ joined	09:19
	→ vedant joined	09:21
	← d2dchat left	09:21
	→ Err404NotFound joined	09:21
	← ferdev left	09:22
	← AdrienBrault left	09:23
	← Cromulent left	09:23
	→ babilen joined	09:24
	← kcj left	09:24
	→ bartzy joined	09:24
bartzy	Hey	09:24
	is it the best practice to have just one .gitignore file in the project's root ?	09:24
	→ ferdev joined	09:24
	← simesy left	09:25
	→ fuzzybunny69y joined	09:26
	→ mastroWork joined	09:26
	→ duosrx joined	09:26
	→ Imaginativeone joined	09:26
	ChanServ set mode: +v	09:27
	→ Transcended joined	09:28
	← ferdev left	09:29
cmn	no	09:29
	← sheerun left	09:29
	← nathancahill left	09:29
	← elixir_ left	09:30
	← LeMike left	09:30
	→ elixir_ joined	09:30
	← cilly left	09:30
	→ Anders_J joined	09:30
	→ ferdev joined	09:31
	← gwty left	09:32
	→ narfnarf joined	09:32
	← ferdev left	09:34
	→ sheerun joined	09:34
	→ nathancahill joined	09:36
	→ ferdev joined	09:36
cbreak-work	do what ever you want	09:36
	→ zoresvit joined	09:36
	← Transcended left	09:39
	→ tevio joined	09:39
	← vedant left	09:39
	→ w0rmdrink joined	09:40
	→ vedant joined	09:40
	→ mastro__ joined	09:40
	→ valgrind joined	09:42
	← nathancahill left	09:42
	← mastroWork left	09:43
	→ nathancahill joined	09:44
	← rakl left	09:44
	→ gwty joined	09:46
	← fuzzybunny69y left	09:46
	← psoo left	09:46
	→ psoo joined	09:47
	→ rakl joined	09:47
	← maxmanders left	09:47
	← ferdev left	09:47
	→ LeMike joined	09:47
	→ rahulprasad joined	09:49
	→ fuzzybunny69y joined	09:49
	→ ferdev joined	09:50
fuzzybunny69y	hey guys I have some files that I would like to put into older commits. Someone mentioned earlier that this can be done using a git rebase -i but I am not sure how to do it. Would someone be able to guide me through it so I can hopefully understand it.	09:50
	→ maxmanders joined	09:51
fuzzybunny69y	I am reading the help on rebase but I don't know what it is talking about	09:51
	← ferdev left	09:52
	→ karupanerura joined	09:52
	→ pretty_f_ joined	09:52
	→ achadwick joined	09:52
	← pretty_function left	09:53
	→ ferdev joined	09:54
	→ exoon joined	09:55
	→ puchrojo joined	09:56
	→ mgaunard_ joined	09:56
jast	fuzzybunny69y: !fixit might help	09:56
gitinfo	fuzzybunny69y: [!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!	09:56
	← stliu left	09:57
	← nathancahill left	09:57
	← jazz264 left	09:59
	← thufir_ left	09:59
	pretty_f_ → pretty_funcion	09:59
	pretty_funcion → pretty_function	09:59
	→ marhaban joined	10:01
	→ starlays joined	10:02
	→ doflip joined	10:02
	← doflip left	10:02
	← rahulprasad left	10:04
	→ ozette joined	10:04
	→ nathancahill joined	10:05
	← mastro__ left	10:05
	← babilen left	10:06
	→ chisler joined	10:07
	→ babilen joined	10:07
	← vedant left	10:08
	← yshh left	10:09
	← nathancahill left	10:09
	→ yshh joined	10:09
	← yshh left	10:10
	→ jkremser joined	10:11
	← marlun left	10:11
	→ rosco_y joined	10:14
	← rosco_y left	10:15
	← fuzzybunny69y left	10:15
	← FernandoBasso left	10:15
	→ JSharpe__ joined	10:15
	→ csmrfx joined	10:16
	← maxmanders left	10:16
	→ Sicp joined	10:16
	→ the_cybersphinx joined	10:17
	→ rahulprasad joined	10:17
	→ elixir__ joined	10:17
	← elixir_ left	10:17
	→ mastro__ joined	10:18
	← JSharpe_ left	10:19
	← Mannequin left	10:19
csmrfx	At my dev repo, with single master branch, git log says last commit is bc96f. At gitorious, log says last commit is bc96f. At public, where gitorious is the remote origin master, git reset --hard origin/master gets the previous commit 5c4c0. Ideas?	10:20
	→ beautifulmind joined	10:21
csmrfx	Ok, git pull solved the problem.	10:22
	→ tinti joined	10:23
	→ rideh joined	10:23
	dwmw2_gone → dwmw2	10:25
	← sangi left	10:27
	→ sangi joined	10:27
	→ maxmanders joined	10:28
	← Dr4k3 left	10:29
	← ferdev left	10:30
	→ ferdev joined	10:32
cbreak-work	csmrfx: mrfx, you have to fetch	10:32
	git operates purely local, with only few exceptions	10:32
	← zoresvit left	10:32
csmrfx	cbreak-work: git reset --hard origin/master is 'bad'?	10:34
	← grebo_ left	10:34
cbreak-work	depends.	10:34
	do you want to kill the working dir and the current branch?	10:34
csmrfx	well we dont have branches	10:35
	← gwty left	10:35
csmrfx	but yes, wipe working dir with whatever is on the origin/master	10:35
	→ m0viefreak joined	10:35
	← tinti left	10:35
	← m0viefreak left	10:35
	→ m0viefreak joined	10:35
csmrfx	Weird how that didn't work with git reset --hard	10:35
	← xerxas left	10:35
	← HardDisk_WP left	10:36
cbreak-work	you have branches, at least one	10:36
csmrfx	only one branch	10:36
	→ HardDisk_WP joined	10:36
	← HardDisk_WP left	10:36
	→ HardDisk_WP joined	10:36
cbreak-work	git reset --hard will work	10:36
	← Natureshadow left	10:36
	→ angelsl joined	10:36
cbreak-work	it will reset to the old master from origin you had from before	10:36
	you have to fetch if you want new history from other repositories	10:37
	→ RaceCondition joined	10:37
csmrfx	perhaps gitorious has some kind of cache that has delay	10:37
	only one origin, with one branch, ftr	10:37
	← Adys left	10:37
	→ xerxas joined	10:38
	← HeN left	10:38
	← xiangfu left	10:38
	→ mulinux joined	10:39
	→ Adys joined	10:39
	← maxmanders left	10:39
	→ xiangfu joined	10:40
	← mtlatif left	10:42
	← Gentlecat left	10:43
	→ Mannequin joined	10:43
	→ Natureshadow joined	10:44
	→ acrocity_ joined	10:44
	← acrocity left	10:46
	acrocity_ → acrocity	10:46
	→ gwty joined	10:46
	→ maxmanders joined	10:46
	→ kristofers joined	10:46
	← yann\|work left	10:47
	→ yann\|work joined	10:47
	← luc__ left	10:48
	← AdrienBr_ left	10:48
	→ Pieplay joined	10:48
	→ DoniRS joined	10:49
	→ jaseemabid joined	10:49
	→ tinti joined	10:49
Pieplay	So how does git handle corruption,	10:49
	← ISF_ec09 left	10:49
Pieplay	if a blob is corrupted but the file in the checkout is good against the SHA1	10:50
	→ mastroWork joined	10:50
Pieplay	does it use the file in the chechout??	10:50
grawity	When git tries to read the blob, it uses only the blob... or complains "blob XXXXXXX is corrupt"	10:51
	← hyperair left	10:51
Pieplay	okay, what about corruption on the origin and good version of the file on the clone	10:51
	2 blobs in this case	10:51
	→ zoresvit joined	10:52
	→ ISF_ec09 joined	10:52
	← mastro__ left	10:53
Pieplay	It probably complains, gotcha	10:54
	← rideh left	10:55
	← xiukun left	10:56
	← spuz left	10:56
	→ gusnan joined	10:59
	← berndj left	10:59
Nevik	Pieplay: a corrupted git object can (nearly) alwats be detected	11:00
	because the SHA in its name wont match the SHA of its contents	11:00
	← engkur left	11:02
	← ferdev left	11:02
	← PanPan left	11:02
	→ tvw joined	11:02
	→ scorphus joined	11:03
	→ mastro__ joined	11:03
	← sgambino left	11:04
	→ ferdev joined	11:04
	← maxmanders left	11:05
	← mastroWork left	11:05
	← Err404NotFound left	11:05
Pieplay	If the file is still uncorrupted in my checkout and I commit it, hence making a blob with the correct SHA, will that fix it?	11:06
	← Milossh left	11:06
	→ DonRichie joined	11:06
	→ mklappstuhl joined	11:06
	← keller left	11:07
	→ keller joined	11:07
grawity	You might need to remove the corrupted blob first	11:07
	→ Err404NotFound joined	11:07
grawity	and instead of committing, I would try `git hash-object -w yourfile`	11:07
	→ maxmanders joined	11:08
grawity	(which just makes a blob out of the file)	11:08
Pieplay	oh okay	11:08
	← ferdev left	11:09
	← keller left	11:10
	→ keller joined	11:10
Pieplay	i'm kinda abusing git to make a x2 redundant storage for files/disk ("time capsule"). a script that sha's every blob and checkout file finds and replaces the bad apples by it's mirror	11:11
	→ ferdev joined	11:12
Pieplay	would be nice	11:12
	← Natureshadow left	11:12
	← Katafalkas left	11:13
grawity	Start with `git fsck`	11:13
	also, that really sounds like abuse of git :P try something like ZFS instead	11:14
	← heroux left	11:14
	← NiteRain left	11:14
	← gitinfo left	11:14
Pieplay	well I know git for my dev projects, so why not. Will look into it	11:14
	← durnik left	11:15
	→ Natureshadow joined	11:15
	← xiangfu left	11:15
	→ fnoyanisi joined	11:16
	→ gitinfo joined	11:16
	ChanServ set mode: +o	11:16
	→ durnik joined	11:16
	→ xiangfu joined	11:17
fnoyanisi	hi, I am quite new to version control systems. Started with git, and reading the book online. Just confused; when we "push" data, the "committed" data is uploaded to repository. And any file to be committed, it has to be "added" first. right?	11:17
Seveas	fnoyanisi: correct	11:17
csmrfx	uploaded to a remote repo	11:18
	→ heroux joined	11:18
fnoyanisi	so, where is "staging" step	11:18
Seveas	'git add' adds thing to the commit staging area. 'git commit' commits that staging area to the local repository. 'git push' pushes to a remote repo	11:18
fnoyanisi	hmm.... ok	11:19
	thanks, much appreciated!	11:19
	→ Katafalkas joined	11:19
	→ WhereIsMySpoon joined	11:21
	→ frogonwheels_ joined	11:21
	→ RickXy joined	11:21
	→ txwikinger2 joined	11:22
	→ jaeckel_ joined	11:22
	→ edenc joined	11:22
	→ Sargun_ joined	11:22
	→ bremner` joined	11:22
	← kennethreitz left	11:23
	← canton7 left	11:23
	← adhawkins left	11:23
	← vy left	11:23
	← ishi left	11:23
	← mh` left	11:23
	← netj left	11:23
	← ivan` left	11:23
	← hash left	11:23
	← ivan left	11:23
	← bremner left	11:23
	← drue left	11:23
	← tystr left	11:23
	← jast left	11:23
	← specing left	11:23
	← ocharles left	11:23
	← asedeno left	11:23
	← errr left	11:23
	← yx left	11:23
	← nevyn left	11:23
	← mmattice left	11:23
	← Sargun left	11:23
	← spyroboy left	11:23
	← gcmalloc left	11:23
	← tr4656 left	11:23
	← quuxman left	11:23
	← DrNick left	11:23
	← rawtaz left	11:23
	← frogonwheels left	11:23
	← vmiklos left	11:23
	← andres left	11:23
	← reikalus1kka left	11:23
	← majoh left	11:23
	← PaulePanter left	11:23
	← kortti left	11:23
	← tango_ left	11:23
	← speakman left	11:23
	← muep left	11:23
	← skoop left	11:23
	← _W_ left	11:23
	← no_maam left	11:23
	← jonmasters left	11:23
	← robinsmidsrod left	11:23
	← s4muel left	11:23
	← edenc_ left	11:23
	← JonasF_ left	11:23
	← MissionCritical left	11:23
	← mattp left	11:23
	← AAA_awright left	11:23
	← Nikoli left	11:23
	← rrs left	11:23
	← KaiSforza left	11:23
	← cehteh left	11:23
	← Moult left	11:23
	← 16SAA520N left	11:23
	← avalarion left	11:23
	← multi_io_ left	11:23
	← WhereIsMySpoon_ left	11:23
	← xxtjaxx_ left	11:23
	← sqz_ left	11:23
	← SomeoneWeird left	11:23
	← mitz left	11:23
	→ Edgan_ joined	11:23
	→ mjessome_ joined	11:23
	→ errr_ joined	11:23
	← errr_ left	11:23
	→ errr_ joined	11:23
	→ brocken_ joined	11:23
	← ferdev left	11:23
	← brocken left	11:23
	→ specing_ joined	11:23
	→ mh` joined	11:23
	→ cehteh joined	11:23
	→ mastroWork joined	11:23
	→ ivan` joined	11:23
	→ mmattice joined	11:23
	→ t_hash joined	11:23
csmrfx	Suppose your "pub" was out of sync with origin, and pub was critical. How would you granularly diff changes between pub and origin? File by file?	11:23
cmn	you use log, because you've been committing changes that explain what they do and why they're needed	11:23
	← smirn0v left	11:23
csmrfx	ok, thats a good start?	11:23
	(after stashing=	11:23
	*)	11:23
cmn	if you haven't been doing that, then it depends on what you've been doing	11:23
	and you need to adjust how you look at changes according to how you've been making them	11:23
	← ryao left	11:23
	→ nevyn joined	11:23
	→ tystr_ joined	11:23
	← txwikinger2 left	11:23
	→ txwikinger2 joined	11:23
	← specing_ left	11:23
	→ specing_ joined	11:23
	← bremner` left	11:23
	→ bremner` joined	11:23
	→ cthuluh joined	11:23
	← ivan` left	11:23
	→ ivan` joined	11:23
	← jaeckel_ left	11:23
	→ jaeckel_ joined	11:23
	→ s4muel joined	11:23
	→ ivan joined	11:23
	→ rawtaz joined	11:23
	→ reikalus1kka joined	11:23
	→ kortti joined	11:23
	→ codon joined	11:23
	→ muep joined	11:23
	→ AAA_awright_ joined	11:23
	→ roger_raymond joined	11:23
	→ murosai joined	11:23
	→ RichardBronosky joined	11:23
	→ ryao joined	11:23
	→ no_maam joined	11:23
	→ speakman joined	11:23
	→ tango_ joined	11:23
	→ peper joined	11:23
	→ majoh joined	11:23
	← kloeri left	11:24
	→ linpc joined	11:24
csmrfx	fnoyanisi: google visual git guide, perhaps the illustrations there help form mental map	11:24
	→ spacepluk joined	11:24
	→ multi_io joined	11:24
	→ JonasF_ joined	11:24
	specing_ → specing	11:24
	→ MagnoBooter joined	11:24
	→ Ferris_ joined	11:24
	← mulinux left	11:24
	→ thnee joined	11:24
	→ pdelvo joined	11:24
	→ mattp joined	11:24
	→ xxtjaxx joined	11:24
	→ kloeri joined	11:24
	← xiangfu left	11:24
grawity	fnoyanisi: with distributed VCSes like Git or Hg, the data is added to the local repository when you 'commit', and to the remote repository when you 'push'	11:24
	→ drue_ joined	11:24
	→ jargan joined	11:24
	→ corsair__ joined	11:24
	→ kandinsk1 joined	11:24
	→ spyroboy_ joined	11:24
	→ quuxman_ joined	11:24
	→ adhawkins_ joined	11:24
	→ gcmalloc_ joined	11:24
	→ PaulePan1er joined	11:24
	← k_sze[work] left	11:24
	→ asedeno_ joined	11:24
	→ vmiklos_ joined	11:24
	← asedeno_ left	11:24
	→ asedeno_ joined	11:24
	← spyroboy_ left	11:24
	→ spyroboy_ joined	11:24
	→ netj joined	11:24
	→ rupert joined	11:24
	→ jonmasters_ joined	11:24
	→ j416_ joined	11:24
	→ yshh joined	11:24
	→ Moult_ joined	11:24
	→ Johanna_Meszoros joined	11:24
	→ tswett_ joined	11:24
	→ eregon joined	11:24
	→ andres_f joined	11:24
	→ kennethreitz_ joined	11:24
	→ _W joined	11:24
	← kennethreitz_ left	11:24
	→ kennethreitz_ joined	11:24
	→ tr4656_ joined	11:24
	→ Juri_Agarin joined	11:24
	→ dpwright joined	11:24
	→ BooK joined	11:24
	→ sqz joined	11:24
	→ ndim joined	11:24
	→ juvenal joined	11:24
	→ DrNick joined	11:24
	→ SomeoneWeird joined	11:24
	← Pieplay left	11:24
	→ SpecialEmily joined	11:24
	→ KaiSforza joined	11:24
	SomeoneWeird → Guest3815	11:24
	j416_ → Guest45369	11:25
	← yshh left	11:25
	→ ferdev joined	11:25
	← vmiklos_ left	11:25
	→ Nikoli joined	11:25
	← Katafalkas left	11:25
	← techlife left	11:25
csmrfx	Is it possible to check remote changes into "non-active" branch, and then diff/merge them into current branch from there?	11:25
	jaeckel_ → jaeckel	11:25
	→ vmiklos joined	11:25
	← adhawkins_ left	11:25
	→ robinsmidsrod joined	11:25
	→ g3orge joined	11:25
	→ adhawkins joined	11:26
csmrfx	(even if both remote == origin and local branches are called 'master')	11:26
	← Guest3815 left	11:26
	→ Guest3815 joined	11:26
	← mastro__ left	11:26
	→ xiangfu joined	11:26
	→ ocharles joined	11:26
	Guest3815 → SomeoneWeird	11:26
grawity	csmrfx: it is, but how would it be different from just `git diff master..origin/master` and such?	11:26
	→ skoopty joined	11:26
	→ sonda joined	11:26
	← durnik left	11:27
	→ durnik joined	11:27
	→ BSaboia joined	11:28
	→ yx joined	11:28
	← RobertLaptop left	11:28
	← Etheco left	11:28
	← totte left	11:28
	← madduck left	11:28
	← Conjuro left	11:28
	← seanius left	11:28
	← billy_ran_away left	11:28
	← ToBeFree left	11:28
	← Timbo left	11:28
	← JPohlmann left	11:28
	← cdeszaq left	11:28
	← marienz left	11:28
	← nanun left	11:28
	← muddo left	11:28
	← palyboy left	11:28
	← NoXzema left	11:28
	← abcds223 left	11:28
	← ketas left	11:28
	← i7c left	11:28
	← Tectu left	11:28
	← JPT left	11:28
	← dsop left	11:28
	← yess left	11:28
	← brandonw left	11:28
	← phiwo left	11:28
	← PerlJam left	11:28
	← ozzloy left	11:28
	← dv_ left	11:28
	← spb left	11:28
	← TheMoonMaster left	11:28
	← m4rcu5 left	11:28
	← GodEater left	11:28
	← Happzz left	11:28
	← plasticboy left	11:28
	← eddd_ left	11:28
	← gtmanfred left	11:28
	← Fissure left	11:28
	← BrianR left	11:28
	← Sevein left	11:28
	← xeon-enouf left	11:28
	← beautifulmind left	11:28
	→ techlife joined	11:29
	← marhaban left	11:29
	→ Goles joined	11:29
	→ Katafalkas joined	11:31
	← Goles left	11:31
	→ RobertLaptop_ joined	11:31
	→ kirin`_ joined	11:31
	← kirin`_ left	11:31
	→ kirin`_ joined	11:31
	← mishok13 left	11:31
	→ GodEater joined	11:31
	→ Timbo joined	11:31
	→ xeon-enouf joined	11:31
	→ Etheco joined	11:31
	→ totte joined	11:31
	→ dv_ joined	11:31
	→ madduck joined	11:31
	→ cdeszaq joined	11:31
	→ Conjuro joined	11:31
	→ phiwo joined	11:31
	→ seanius joined	11:31
	→ billy_ran_away joined	11:31
	→ ToBeFree joined	11:31
	→ JPohlmann joined	11:31
	→ marienz joined	11:31
	→ nanun joined	11:31
	→ muddo joined	11:31
	→ palyboy joined	11:31
	→ NoXzema joined	11:31
	→ abcds223 joined	11:31
	→ i7c joined	11:31
	→ ketas joined	11:31
	→ Tectu joined	11:31
	→ JPT joined	11:31
	→ dsop joined	11:31
	→ yess joined	11:31
	→ brandonw joined	11:31
	→ PerlJam joined	11:31
	→ ozzloy joined	11:31
	→ TheMoonMaster joined	11:31
	→ spb joined	11:31
	→ m4rcu5 joined	11:31
	→ Happzz joined	11:31
	→ plasticboy joined	11:31
	→ eddd_ joined	11:31
	→ gtmanfred joined	11:31
	→ Fissure joined	11:31
	→ BrianR joined	11:31
	→ Sevein joined	11:31
	→ mgedmin joined	11:31
	→ tom[] joined	11:31
	→ ejnahc joined	11:31
	→ tokkee joined	11:31
	→ VadtecWk joined	11:31
	→ crater2150 joined	11:31
	→ tgummerer joined	11:31
	→ wagle joined	11:31
	→ docwhat joined	11:31
	→ acidjazz joined	11:31
	→ tpope joined	11:31
	→ ben_alman joined	11:31
	→ Byan joined	11:31
	→ pasky_ joined	11:31
	→ frew joined	11:31
	→ adamm joined	11:31
	→ soren joined	11:31
	→ dean joined	11:31
	→ justinfung joined	11:31
	→ davidh joined	11:31
	→ jrockway joined	11:31
	→ TML joined	11:31
	→ Gizmokid2005 joined	11:31
	→ torjus joined	11:31
	cameron.freenode.net set mode: +vv	11:31
	← LeMike left	11:31
	← puchrojo left	11:31
	← nicolash left	11:31
	← YamakasY3 left	11:31
	→ Goles joined	11:31
	→ nicolash1 joined	11:31
	→ puchrojo joined	11:32
	→ LeMike joined	11:32
csmrfx	grawity: plain diff only shows the changes	11:32
	grawity: I am trying to figure out the strategy on how to get these changes from remote into mission critical local, safely, in a controlled manner	11:32
	→ YamakasY2 joined	11:32
	→ smirn0v joined	11:32
	← Goles left	11:32
	← durnik left	11:32
	→ rabellamy joined	11:32
grawity	`git log -p master..origin/master` or something like that...	11:33
	`git branch changes-from-master origin/master`	11:33
	← Samual left	11:34
	← tinti left	11:34
	→ baba joined	11:34
	→ Mission-Critical joined	11:34
	← theotherguy left	11:34
	→ Samual joined	11:34
	→ xlq joined	11:34
csmrfx	grawity: hm, is that last example supposed to rename a branch?	11:35
xlq	How can I get a stash back that I popped?	11:35
	← Squarism left	11:35
	← fire left	11:35
Seveas	csmrfx: no, it creates a new branch, with origin/master as base	11:35
	← Squarism2 left	11:35
grawity	csmrfx: no, it's supposed to create a new branch "changes-from-master" pointing to the same commit as "origin/master"	11:35
cbreak-work	if you want to rename a branch, use git branch -m	11:35
grawity	csmrfx: in other words, copy, not move	11:35
	← RaceCondition left	11:35
cbreak-work	xlq: popped stashes are gone	11:35
	xlq: it's tricky to get them back	11:36
	xlq: see man git-fsck --lost-found	11:36
gitinfo	xlq: the git-fsck manpage is available at http://jk.gs/git-fsck.html	11:36
xlq	Thanks cbreak-work, found it. :)	11:36
	→ ak5 joined	11:36
	← ocharles left	11:37
	→ ocharles joined	11:37
	← brent left	11:38
	→ durnik joined	11:39
csmrfx	grawity: does that do something like a fetch?	11:40
	← nicolash1 left	11:40
grawity	csmrfx: no	11:40
	csmrfx: you already fetched	11:40
	← skoopty left	11:40
grawity	origin/master already points to a local branch that you fetched from the remote.	11:40
	→ skoop joined	11:40
	← ferdev left	11:41
grawity	(and if you haven't already fetched, then do it)	11:41
	← exoon left	11:41
csmrfx	Scary idea	11:41
	→ RaceCondition joined	11:41
	← kirin`_ left	11:41
grawity	…	11:41
csmrfx	It's a bit of a mess.	11:41
	→ milStaja joined	11:42
csmrfx	Seems the changes need to be pushed, actually.	11:42
	→ kirin` joined	11:43
	→ mastro__ joined	11:43
	→ rideh joined	11:43
	→ ferdev joined	11:43
	← durnik left	11:44
csmrfx	git stash save --keep-index would leave the working directory as it is, correct?	11:44
	csmrfx just doesn't want anything* in the working dir files to change*	11:45
grawity	I think it only leaves those changes that you have `git add`ed	11:45
	so no	11:45
	→ Squarism joined	11:45
	→ Squarism2 joined	11:45
	← mastroWork left	11:46
csmrfx	Ok, perhaps stash isn't the right tool for 'saving snapshot's?	11:46
	← ISF_ec09 left	11:46
	PaulePan1er → PaulePanter	11:46
grawity	what, exactly, are you trying to do?	11:46
	← kirin` left	11:46
csmrfx	Now, I want easily recoverable snapshot of the current working dir state	11:47
	→ rurufufuss joined	11:47
	→ kirin` joined	11:48
	→ kryl99 joined	11:48
	Mission-Critical → MissionCritical	11:49
	← ferdev left	11:49
	→ mishok13 joined	11:49
osse	sounds like git stash to me :O	11:49
	→ marhaban joined	11:50
	← heroux left	11:50
	← zoresvit left	11:50
csmrfx	Heh, I am just paranoid	11:50
	→ nihilista joined	11:51
csmrfx	The last commit to local is ok. No non-committed changes to local dir. Is git stash safe in this scenario?	11:51
	← juvenal left	11:51
	← Imaginativeone left	11:51
	→ ferdev joined	11:51
osse	in that case git stash has nothing to do	11:51
grawity	If there aren't any non-committed changes, then `git stash` will not do anything	11:51
osse	If I understand you correctly	11:51
	→ heroux joined	11:52
grawity	It looks like the current working dir state is equal to the latest commit	11:52
csmrfx	yes, it's kinda silly	11:52
grawity	So it's already recoverable in that you can just checkout that commit	11:52
csmrfx	yes	11:52
	← jason237 left	11:53
csmrfx	Suppose you're going on a holiday and want the person X who shows up to fix potential problems to be able to know "this is the good version"?	11:53
osse	you can tag it	11:53
	aka. label	11:53
	← xlq left	11:54
	ChanServ set mode: +v	11:54
osse	git tag -a -m 'Hey, Brian, this is the good version!'	11:54
	jargan → jast	11:54
	→ BillCriswell joined	11:54
osse	oh, I forgot the tag name itself	11:54
	← kryl99 left	11:54
	→ durnik joined	11:54
csmrfx	yes, tag seems like the idea	11:55
	→ beautifulmind joined	11:55
osse	csmrfx: if you are paranoid for realsies you can sign the tag with a key, so that Brian knows it was actually you who tagged it and not an imposter :P	11:55
csmrfx	ok	11:56
	not paranoid in that way	11:56
	← ferdev left	11:56
osse	:)	11:56
csmrfx	imposter would have hard time getting to this compu	11:56
	hm, is it possible to get tags to show in "git log"?	11:57
grawity	log --decorate	11:57
	← blami_orcl left	11:57
	→ blami_orcl joined	11:57
	→ mastroWork joined	11:58
	→ Goles joined	11:58
grawity	Although it won't show tag messages, just the names.	11:58
	→ ferdev joined	11:58
	→ dermoth_ joined	11:58
	← Goles left	11:59
	→ Goles joined	11:59
	← Goles left	11:59
	← gwty left	11:59
csmrfx	Does doing git pull on a local that is ahead of the remote origin do interactive merge?	12:00
	← mastro__ left	12:00
	← dermoth\|home left	12:00
	→ kryl99 joined	12:01
osse	csmrfx: if it's strictly ahead I don't think it will do anything.	12:01
	Because there is nothing to merge	12:01
	→ dsdeiz joined	12:01
	← Natureshadow left	12:02
Seveas	osse: and if it's not strictly ahead, but the remote has some commits too, the merge will not be interactive until it fails. At that point you get to resolve the conflicts.	12:03
csmrfx	Is it possible to get git to drop me into difftool from CONFLICTs, then?	12:03
	← maxmanders left	12:03
osse	csmrfx: yes. 'git mergetool'	12:03
	← narfnarf left	12:03
csmrfx	ahh 8)	12:03
	← ferdev left	12:03
csmrfx	this looks familiar 8)	12:04
cmn	well, yes and no, your difftool may also be a mergetool, but they're for different things, asking for a difftool won't get you the tool	12:04
	→ Dr4k3 joined	12:04
	← jxpx777 left	12:04
	fnoyanisi → fnoyanisi_away	12:05
	→ dirte joined	12:05
	→ ferdev joined	12:05
	gitinfo set mode: +v	12:06
	frogonwheels_ → frogonwheels	12:06
csmrfx	hm, it looks familiar but is not	12:06
	→ zoresvit joined	12:06
cmn	it may be the same program, but they're for different things	12:06
csmrfx	What to google for mergetool help? Is that emacs by default, guess	12:06
cmn	is what emacs?	12:06
	csmrfx hits the vim commands, but it is futile	12:07
cmn	mergetool and difftool are vimdiff unless you tell git otherwise	12:07
csmrfx	what "git mergetool" opens	12:07
	→ nathanr joined	12:07
osse	csmrfx: if not configured it tried to open your difftool	12:07
	csmrfx: so if you use vimdiff it'll use vimdiff	12:07
	*it tries	12:07
csmrfx	which is vimdiff, but it didn't open vimdiff	12:07
osse	oh.	12:07
	← durnik left	12:08
	→ maxmanders joined	12:08
osse	csmrfx: run 'git config merge.tool'	12:09
	→ mastro__ joined	12:10
	→ mulinux joined	12:10
csmrfx	oh man I made a mess	12:11
	← nathanr left	12:12
	← mastroWork left	12:13
	← BadDesign left	12:13
csmrfx	osse that does nothing, git config -l doesnt list diff or mergetools	12:14
	→ Danic11 joined	12:14
	→ dvide joined	12:14
	→ fuzzybunny69y joined	12:14
	→ Element9 joined	12:14
	← ferdev left	12:15
	gitinfo set mode: +v	12:15
Danic11	Hi there. I need to exclude a file in subdir from polluting my git status. (in dont want it in origin, and I dont want it updated from origin). So I edit .git/info/exclude and add the line mysite/settings.py But that file is still being detected as modified by git stauts.	12:15
	→ gwty joined	12:15
	→ ignas joined	12:16
	→ ferdev joined	12:16
	← milStaja left	12:16
	→ tmcmahon joined	12:16
osse	csmrfx: then I don't know	12:17
	← ferdev left	12:17
osse	csmrfx: I guess you have to configure it explcitly then: 'git config --global merge.tool vimdiff'	12:17
csmrfx	I did done: git mergetool -t vimdiff	12:17
	→ garbagecollectio joined	12:17
	← yx left	12:17
	→ engkur joined	12:18
cmn	Danic11: is it tracked?	12:18
	← Xethron left	12:18
osse	csmrfx: yeah that should work	12:18
Danic11	yes	12:19
csmrfx	Ok, should I remove the "<<<<<<< HEAD" from the 3-way diff before saving?	12:19
Danic11	ahh. hmm	12:19
	→ ferdev joined	12:19
csmrfx	3-way diff-view, in "git mergetool"	12:19
cmn	Danic11: then that's why git shows it, you told it you care about the file	12:19
osse	csmrfx: those <<<<<< ======== >>>>>>> indicate the conflict. you must clean all that mess.	12:19
csmrfx	Ok, git doesn't do anything with those?	12:19
	← Froker left	12:20
grawity	git adds them, you (or your mergetool) remove them	12:20
osse	csmrfx: git puts them there to show you the conflict and expects you to remove them before commiting	12:20
	← thehandler left	12:20
csmrfx	ok, thanks	12:20
	← marhaban left	12:20
csmrfx	really simple q's but better ask than be sorry. I appreciate it 8)	12:20
Danic11	cmn, I did "git rm mysite/settings.py --cached" and now ints tracked as deleted :facepalm:	12:20
	→ Xethron joined	12:21
cmn	right	12:21
	← sangi left	12:21
cmn	git is doing what ask it to	12:21
grawity	now when you commit, it won't be tracked anymore	12:21
osse	csmrfx: and with that in mind: for simple conflicts maybe it's just easier to clean it up by hand :)	12:21
	csmrfx: the markers are in the actual file, so you can open it with any editor directly.	12:21
	← tmcmahon left	12:21
csmrfx	It's difficult, when you're ahead ;)	12:21
grawity	...so why are you pulling if you're ahead?	12:22
	→ yx joined	12:22
csmrfx	because of CONFLICT	12:22
osse	if he actually has conflicts then I guess he cannot be ahead	12:22
	except in the sense that he has made more commits than the other guy	12:22
csmrfx	well I am.	12:22
rking	How granular do you like your commits?	12:23
csmrfx	depends	12:23
rking	I've always said, "Make them atomic. One short subject = completely describes one commit, where possible."	12:23
	But I'm trying to merge a long-running branch, and teammates are saying they want squashing.	12:24
	← cagedwisdom_ left	12:24
rking	Which, IMO sounds cute on paper, like you can wave a magic wand over 350 commits and make them all be conceptually organized without losing much useful information, but IRL I don't think it's anywhere near worth the effort.	12:24
cmn	tell them that doesn't solve any problems	12:24
rking	It's weird because I'm in the sharp minority here: 1 vs 4	12:25
cmn	unless the history is confusing or full of 'oops' commits	12:25
rking	But zero of the 4 have actually read the commits or could even point to one single pair of commits that should be squashed.	12:25
csmrfx	vimdiff isn't a good tool for merges it seems. Colorscheme doesn't show all chars	12:25
	csmrfx is having a bad git experience	12:26
cmn	are they asking for a squash-merge instead or what?	12:26
rking	There are almost no oops commits. There are maybe 7-10 commits that are labeled clearly, "This is taking a step back to take a few steps forward in the next few commits"	12:26
cmn	csmrfx: use a mergetool you like; this is not a git thing	12:26
csmrfx	I know	12:26
grawity	heh, I didn't even know vim can work as a mergetool	12:26
	csmrfx doesn't know emacs keybindings	12:26
rking	csmrfx: I have a theme that works for me, vim-detailed, but YMMV	12:26
	cmn: I guess so. One suggestion was to squash everything but keep the branch as it is	12:27
	← Dr4k3 left	12:27
cmn	that is contradictory	12:27
csmrfx	Whew, what a huge mess from being 1 commit ahead,	12:27
grawity	csmrfx: don't use emacs then, try kdiff3 or meld	12:27
csmrfx	but it is solved now.	12:28
osse	grawity: depends on your qualifications for what a mergetool is obviously, but vimdiff as a mergetool just shows a four-way diff between the usual suspects, there is no different functionality between that and when you use it as a difftool.	12:28
csmrfx	grawity: can't, neither installed	12:28
rking	cmn: I was honestly stunned that they were bringing this up just now. It's been 6 weeks (which is too long, far too long), but they've seen the granular commits coming in every day.	12:28
cmn	then ignore them	12:28
	← mulinux left	12:29
osse	what are the arguments for squashing? :O	12:29
rking	cmn: I mean the idea was that 'master' would contain one giant commit (with a useless history, IMO), and then leave the branch sitting there for reference.	12:29
cmn	lemon squash is tasty	12:29
	that is the worst idea they could have	12:29
rking	osse: "I don't want to have to page through this when looking back through the history."	12:29
	"I can't understand this pull request. There's too much."	12:29
osse	rking: even with --oneline? :P	12:30
csmrfx	I always try to think of the others. "Does the guy X coming from out of nowhere to this have the ability to work with this?"	12:30
rking	Maybe what I need to do is figure out how to say, --without-author rking	12:30
	← Schmoozerd_ left	12:30
cmn	the first one can be legitimate in some circumstances, where the individual commits are less important than the fact the branch was merged	12:30
	but that's what --first-parent and sensible merging come in	12:30
	→ Schmoozerd_ joined	12:30
rking	cmn: Please elaborate? I have the sense that a few basic tools skills could solve their issues.	12:31
	← babilen left	12:31
cmn	--first-parent only shows the first parent, so the commits that got merged aren't shown	12:31
rking	Hrm, that sounds exactly like it.	12:32
cmn	but that only works if you've been sensible and did real merges into the main branch	12:32
rking	So that's a git log or git diff option?	12:32
cmn	log	12:32
	or any history-traversal, really	12:32
	← dsdeiz left	12:33
rking	Interesting.	12:33
	→ itatitat joined	12:33
cmn	as for the second; squashing isn't going to change anything, but you might want to consider merging in work more often	12:34
rking	OK, this could definitely help	12:34
	Yeah, it was completely against my will that it went that long without merge.	12:34
cmn	even if only the refactoring for some later features have been done	12:34
	→ memoryleak joined	12:34
rking	I'm of the "Feature toggles, not feature branches" bent	12:34
csmrfx	I removed a tag, amended a commit message. Boom! somehow I need to merge remote changes again	12:34
	oh, nevermind	12:35
	git pull && git push "fixed	12:35
	→ marhaban joined	12:36
cmn	not somehow, "amended a commit message"	12:36
	look at the history to make sure you've actually done something sensible there	12:36
	← marhaban left	12:37
csmrfx	Howcome that caused (1.) A new commit with new hash, and (2.) another new commit with "Merge branch 'master' of yaddayadda..."	12:37
	→ shreya joined	12:37
	→ JaredMiller joined	12:37
osse	csmrfx: 1. the hash is different because the commit's contents are different	12:37
	so that's unavoidable	12:37
csmrfx	I just wanted to clean up the commit message and move the tag forward one commit.	12:37
osse	csmrfx: since the hash is different you have basically forked the branch.	12:38
	← garbagecollectio left	12:38
	→ HeN joined	12:38
	csmrfx still* doesn't have that one commit that is tagged so that anyone who can read git logs knows which one is the commit*	12:38
osse	the old commit is lost in the ether	12:38
csmrfx	no, its there	12:38
	with the "wrong" commit message	12:39
grawity	delete the old tag, then re-tag	12:39
osse	csmrfx: the trick is to not push until you are certain you don't have to use --amend or -d or stuff like that.	12:39
	← Danic11 left	12:39
	csmrfx needs more coffee or something	12:39
	mgedmin wants to say "git pull -r" helps avoid useless merge commits, but is afraid of jumping into the middle of a conversation while missing all context	12:39
	→ freeed joined	12:40
	← ferdev left	12:40
	← rahulprasad left	12:40
rking	Ha! I didn't now '-r' worked as '--rebase'. Every day I learn I'm near zero with my git skills.	12:40
osse	csmrfx: you can use push --force to clean up the mess, but then you have to let everybody else know, or else they get the same "What the hell is this?" merge experience yuo just did	12:40
	→ mtlatif joined	12:41
rking	And worse, they could end up re-pushing things that you meant to get rid of.	12:41
	→ nathanr joined	12:41
csmrfx	interestingly, git diff shows no output for the last 2 commits.	12:41
	→ amcsi_work joined	12:41
	← giallu left	12:41
	grawity is still trying to understand the part where origin/master is merged into master, which is already ahead O.o	12:42
csmrfx	me too	12:42
	→ ferdev joined	12:42
csmrfx	Although I've mostly moved on to typing stuff and not caring if it seemingly works. (D	12:43
	← RaceCondition left	12:43
cmn	-r works as --rebase currently, but you can't rely on that	12:44
	← engkur left	12:44
grawity	I don't know much Git, but it almost looks like the merge should have been done in the other direction	12:44
	→ RaceCondition joined	12:44
	→ jxpx777 joined	12:45
	← Timbo left	12:45
mgedmin	(when did -r got deprecated? I must've been napping)	12:46
	→ nicolash joined	12:46
csmrfx	I seem to forget all these gitentials too fast	12:47
	SeoZ → SeoZ-work[AWAY]	12:47
	← jaequery left	12:48
	← ferdev left	12:48
	← RaceCondition left	12:48
	→ Ryback_ joined	12:48
csmrfx	it seems vim as a tool for "git mergetool" creates file.BACKUP.<version>, .BASE, .LOCAL, .REMOTE temp files	12:50
	→ ferdev joined	12:50
csmrfx	any ideas? do other merge tools clean up after?	12:50
grawity	that's what `git mergetool` itself does – and cleans up afterwards	12:50
csmrfx	ok, no it didn't	12:50
grawity	since most tools cannot read directly from git, they have to be given actual files to merge	12:50
cmn	mgedmin: it was never official	12:50
	← BillCriswell left	12:51
	← chisler left	12:51
	→ sunya7a joined	12:52
csmrfx	Do you just remove those files by hand?	12:52
	← nathanr left	12:52
	← kryl99 left	12:52
csmrfx	and whats with the #merge#<ver># -files litter? that sounds like recipe for "rm -disaster"	12:53
	← ferdev left	12:53
	← Schmoozerd_ left	12:54
	→ Schmoozerd_ joined	12:55
jast	I've never seen any files like that	12:55
	→ trungqtran joined	12:55
	→ ferdev joined	12:55
	← itatitat left	12:55
	→ nathanr joined	12:55
cmn	sounds like the mergetool crapping its pants	12:55
	← ferdev left	12:56
cmn	git-mergetool shouldn't be creating anything like that	12:56
csmrfx	then again, git created two commits with absolutely no diff either	12:56
	→ theotherguy joined	12:56
	← nathanr left	12:57
	→ babilen joined	12:57
	csmrfx probably did something wrong, again	12:58
cmn	if you tell it to, then it will	12:58
	or you're looking at a merge commit	12:58
rking	cmn: Why can't you rely on -r ?	12:58
	→ nathanr joined	12:58
rking	It's in the manpage… they'd be pretty odd to remove it now, no?	12:58
cmn	because that may not mean --rebase next week	12:58
	did that actually make it there?	12:58
csmrfx	well, the thing is, I thought the merge was the one before the two 'null diff'-commits.	12:58
	← tevio left	12:59
	→ sgambino joined	12:59
cmn	up to very recently, -r only worked because --rebase happened to be the only option that started with r	12:59
csmrfx	So, can I get git to "do a merge" and tell me which ones changed, and let me edit those by hand (instead of mergetool)?	13:00
	→ R1KRDO joined	13:00
	← mklappstuhl left	13:00
cmn	hm, I might have been thinking of --r	13:00
csmrfx	IOW, can I bypass this painful and messy procedure somehow	13:00
	→ mmitchell joined	13:00
cmn	csmrfx: ask status	13:00
	← duosrx left	13:00
osse	csmrfx: yes you can edit by hand. open the file yourself instead of using mergetool.	13:00
	→ lacrymology joined	13:01
osse	I think the "root cause" is the lack of understanding the whole thing about rewriting history	13:01
csmrfx	I dont.	13:01
cmn	it's been in the code for a while now, but there was much more recent activity in the ML surrounding whether you can rely ont hat	13:01
	← littlebearz left	13:01
csmrfx	Is merge 'rewriting' history?	13:01
	→ marhaban joined	13:01
	← marhaban left	13:01
csmrfx	isn't it 'creating new' history, actually	13:02
cmn	it is	13:02
	← JaredMiller left	13:02
rking	Hrm. That isn't what I'd expect, either.	13:02
cmn	but you rewrote some history and then merged both	13:02
	→ tinti joined	13:02
rking	Oh, "it is ⇒ creating new history"	13:02
	→ R1KRDO1 joined	13:02
rking	That is what I'd expect, not that it's rewriting history. =)	13:02
	→ littlebearz joined	13:03
	← pastime left	13:03
csmrfx	Hm, tell me, how exactly did I "rewrite history"?	13:03
	→ chuckharmston joined	13:03
osse	csmrfx: by amending a commit	13:03
csmrfx	Ok.	13:03
	BTW, it did not rewrite history, it created a new commit!	13:04
	← maxmanders left	13:04
osse	csmrfx: indeed.	13:04
cmn	those are not linked	13:04
rking	If the SHA1s in a branch changed, then history got rewritten.	13:04
osse	csmrfx: but that commit has the same parent as the other one. So you have effectively "forked off"	13:04
cmn	you did rewrite history by creating a new commit and replacing the tip of the branch	13:04
rking	So if I git push a commit 1234, but then git commit --amend, HEAD no longer points to 1234	13:04
csmrfx	New commit is just plain writing history. Rewriting is when you change that.	13:04
	← R1KRDO left	13:05
	→ maxmanders joined	13:05
bremner`	in that sense, there is no rewriting in git	13:05
csmrfx	or?	13:05
bremner`	everything is just making new commits	13:05
	→ frenchlaf joined	13:05
csmrfx	ok, so on the same semantic page now.	13:06
rking	Yeah, which isn't a useful sense. =)	13:06
	It's a senseless sense.	13:06
	← maxmanders left	13:06
cmn	the same page which doesn't help you do anything, brilliant	13:06
	rewriting history means making a branch point to a commit which isn't a descendant of where it used to point	13:07
	→ thehandler joined	13:07
cmn	this is what you did	13:07
	← memoryleak left	13:07
csmrfx	At least I have some whiff what is it you're talking about.	13:07
	→ memoryleak joined	13:07
	csmrfx at least I dont have to use RCS or CVS anymore	13:08
	→ yshh joined	13:09
	→ Tech163 joined	13:09
	→ breakingthings joined	13:10
	→ mitz joined	13:11
	← yshh left	13:12
	→ yshh joined	13:12
	← Tech163 left	13:13
osse	running 'git log --graph --oneline --decorate master origin/master' would be instructive, I think	13:13
	→ jonshea joined	13:14
	→ jonatasnona joined	13:14
	→ chrismar035 joined	13:15
	← memoryleak left	13:16
	→ hyperair joined	13:16
	→ mmlac joined	13:17
	→ maxmanders joined	13:18
	← elixir__ left	13:18
	→ rahulprasad joined	13:19
	→ m0ltar joined	13:19
csmrfx	osse: lol I made a little christmas tree!	13:20
	And howcome that isn't a command in git in itself	13:20
osse	csmrfx: isn't it?	13:20
csmrfx	like "git graph"	13:21
osse	csmrfx: the first word there says 'git' here :O Maybe I have dead pixels	13:21
	→ mezner joined	13:21
	ChanServ set mode: +o	13:21
osse	csmrfx: many people make an alias for that, including myself	13:21
	→ fooey joined	13:21
	EugeneKay changed the topic to: Current stable version: 1.8.2 \| Start here: http://jk.gs/git/ \| Getting "Cannot send to channel" or unable to change nick? /msg gitinfo .voice \| Be careful when rewriting history. It may push you to use the dark side of the --force	13:21
	→ blz joined	13:21
osse	csmrfx: git config --global alias.graph 'log --graph --decorate -oneline' :)	13:22
	now you can do 'git graph'	13:22
	EugeneKay set mode: -o	13:22
blz	Hello! I want to go back to an earlier commit and throw out all the commits after that one. What command should I be using?	13:22
_ikke_	blz: !fixup	13:22
gitinfo	blz: 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!	13:22
	→ simesy joined	13:23
blz	!rewrite	13:23
gitinfo	[!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to). If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstream_rebase	13:23
Seveas	blz: git reset --hard sha1_of_the_old_commit will throw out all commits and uncommitted changes after that sha1	13:23
EugeneKay	blz - man git-reset, but beware that time-travel is not advised.	13:23
gitinfo	blz: the git-reset manpage is available at http://jk.gs/git-reset.html	13:23
	→ elixir__ joined	13:23
blz	What's the recommended way of dealing with this, then? My issue is that I've cloned a github repository, implemented some changes, and issued a pull request. The pull request has __not__ been merged yet, but it turns out I have to undo the changes I did to function A and actually put them in function B	13:24
Seveas	is there a bot trigger that has http://justinhileman.info/article/git-pretty/git-pretty.png ? found it quite useful	13:25
blz	You guys make it sound like reverting to a previous commit and rewriting public history is a bad idea	13:25
cmn	it is	13:26
_ikke_	blz: Have you pushed those commits?	13:26
EugeneKay	push -f should never be used. See the /topic :v	13:26
	← maxmanders left	13:26
Seveas	blz: it is generally considered 'know what you're doing' territory. But in your case it looks like you'll jsut issue a new pull request when you're done. Should be fine	13:26
	← g3orge left	13:27
Seveas	blz: also, it makes sense to create new branches for each pull request	13:27
blz	_ikke_, Yes, I pushed them but they haven't been merged	13:27
	← simesy left	13:27
blz	Seveas, ah yeah I guess that makes sense	13:27
	I'll keep that in mind for next time	13:27
	→ maxmanders joined	13:27
	← frenchlaf left	13:28
	→ jacobbednarz joined	13:28
	→ mastroWork joined	13:28
	← mezner left	13:29
	→ khushildep joined	13:29
	→ mezner joined	13:29
	← valgrind left	13:30
	→ xckpd7 joined	13:31
	← mastro__ left	13:32
	→ jamesez joined	13:32
xckpd7	2 questions. One, let's say I rebased a branch off of another at some point a long time ago. Is there anyway of going back and finding out if the rebase happened automatically, or if I had to manually do a rebase?	13:32
	→ jason237 joined	13:32
cmn	how would a rebase happen automatically?	13:33
EugeneKay	Do you mean "if there were conflicts" ?	13:33
	And not within the data structure, no. If you left a note in the commit message then you can	13:34
	→ normanrichards joined	13:34
	→ revagomes joined	13:34
cmn	the reflog might still have a clue to it	13:34
xckpd7	thanks. yeah I don't have the nomenclature	13:34
	← maxmanders left	13:35
xckpd7	one more question... I know git tracks differences in code on a per line basis... is there any type of program / process where I could better diff, not just per line?	13:35
	→ duosrx joined	13:35
osse	if a conflcit was solved manually you might get a hint by checking if two adjacent commit dates are more than a couple of seconds apart :)	13:35
	→ spuz joined	13:35
EugeneKay	It tracks snapshots; commits are usually viewed as being a diff relative to the parent(s).	13:35
xckpd7	so if I have a 500 character line, and I change one character...	13:35
EugeneKay	man git-difftool	13:36
gitinfo	the git-difftool manpage is available at http://jk.gs/git-difftool.html	13:36
	→ hellopat joined	13:36
cmn	then what?	13:36
osse	xckpd7: the whole file will be stored, yes	13:36
cmn	do you want to see --word-diff?	13:36
	the whole file will become an object, it won't always be a full object, though	13:37
	don't mix the data model with the storage	13:37
	→ alunduil joined	13:37
xckpd7	I am so confused by what you're saying right now	13:37
osse	cmn: what do you mean by full object?	13:37
cmn	osse: it will be eventually deltified	13:37
	xckpd7: don't worry about it; what came after the ellipsis?	13:38
	← mastroWork left	13:38
EugeneKay	Each git commit is an independent snapshot of the working tree. Internally, git dedupes objects that compose the tree using diffs to save disk space.	13:38
osse	xckpd7: git doesn't track differences. It tracks complete versions, and usually shows you differences. The diff is computed "at run-time" so to speak.	13:38
xckpd7	osse: ok makes sense	13:38
	← _iron left	13:39
EugeneKay	Common commands such as 'git cherry-pick' and 'git show' work on the diff of a commit relative to the one before it	13:39
xckpd7	cmn: let's say I have a 1000 character line. I have a diff where just one character was changed. What is the best way visually to represent changes on a smaller scale than "oh line 54 changed"	13:39
EugeneKay	But that diff is not stored on-disk; it is calculated.	13:39
cmn	--word-diff, for example	13:39
xckpd7	cmn: I want to see within the lines, what specifically changed	13:39
cmn	or some tool that deals with your insanely long lines	13:39
xckpd7	cmn: I don't see that in the difftool documentation	13:39
EugeneKay	man git-diff	13:40
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	13:40
cmn	that's not a difftool thing, it's a diff thing	13:40
	diftool runs whatever external problem you want	13:40
EugeneKay	And I would bitchslap you until you made those 80-char lines.	13:40
csmrfx	Failboatus Maximus: commit some files, push to remote. Forget to change file a, change it and commit --amend. "fatal: you are in the middle of a merge -- cannot amend."	13:40
	Should just commit?	13:40
xckpd7	EugeneKay: I'm just making an example. The reason this is just coming up, is that I find that I keep incorrectly rebasing + merging things, and I need to do a better job of it	13:41
EugeneKay	--amend oughta work o.O. !repro ?	13:41
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session. This will help immensely with troubleshooting.	13:41
osse	csmrfx: after push, --amend is verboten! :)	13:41
xckpd7	EugeneKay: and just seeing the diffs of the lines just isn't sufficient enough	13:41
_ikke_	osse: Is that new?	13:41
	fnoyanisi_away → fnoyanisi	13:41
osse	_ikke_: Not sure what you're referring to, but I'm gonna go with "yes".	13:42
csmrfx	Thanks!	13:42
	→ macabre joined	13:42
	→ decotii joined	13:42
EugeneKay	xckpd7 - I find that a 3-way merge/diff tool in a GUI works better than the standard -+. man git-config, see diff.tool. kdiff3 or something; I use SmartGit myself.	13:42
gitinfo	xckpd7: the git-config manpage is available at http://jk.gs/git-config.html	13:42
	→ prjkt joined	13:42
	→ acharlieh joined	13:42
prjkt	hi, on Windows git (called tortoise) I added a new folder. how do I commit this change?	13:42
	I mean inside the folder that was already in it	13:42
	I created a new folder	13:42
	how do I commit this change	13:42
csmrfx	add it, and commit	13:43
_ikke_	prjkt: git doesn't track directories (so no empty directories)	13:43
	← nitrix left	13:43
osse	csmrfx: but I suspect somewhere between "amend" and "fatal" you did git pull. Is that right?	13:43
csmrfx	prjkt: see if git status tells you if there are differences	13:43
prjkt	how can it not track directories? That can't possibly be true	13:43
	the directory name is very meaningful for me	13:43
xckpd7	EugeneKay: it light specifically highlights the issues... that is good	13:43
	→ nitrix joined	13:43
csmrfx	osse: dont ask me, I just press the keys on the keyboard!	13:43
prjkt	you're saying that if I were to roll back to when the directories had different names or were empty, I wouldn't even see them?	13:43
osse	csmrfx: lol	13:44
xckpd7	EugeneKay: that is probably going to be infinitely helpful to me. so GUI is the way to go rather than terminal	13:44
	→ maxmanders joined	13:44
EugeneKay	xckpd7 - you can set up git to use a third-party tool ;-)	13:44
	← mihamina left	13:44
osse	prjkt: no problem if they had different names. But as long as you have something to roll back to they cannot have ever been empty.	13:44
prjkt	I guess i should just keep everything in one file and do directories like this /* . / and later / LIB */ and include any source that should be under the direcotry"lib" right in the source code	13:44
	→ guardadigital joined	13:44
prjkt	why can't they be empty?	13:44
	← hyperair left	13:44
	← nitrix left	13:45
	→ nitrix joined	13:45
osse	prjkt: I'm not sure of the technicalities of it. But since git tracks content by checksumming stuff I guess it needs something to checksum.	13:45
	→ catphish joined	13:45
	← nitrix left	13:45
	← Nibato left	13:45
	→ prjkt_ joined	13:45
prjkt_	sorry I disconnected myself	13:45
	so why can't they be empty	13:45
	isn't this a bug?	13:45
csmrfx	no	13:45
_ikke_	prjkt_: nope	13:45
	prjkt_: git is a content tracker	13:45
csmrfx	empty directories cannot be compare	13:45
	*compared	13:46
	← Mannequin left	13:46
cmn	it's due to the current index format not representing dirs	13:46
	you could make it track empty dirs, but it's not worth the effort	13:46
EugeneKay	And would be backwards incompatible	13:46
cmn	if you need an empty dir, create it	13:46
	← mezner left	13:46
	→ sysadminAli joined	13:46
_ikke_	And usually, it's to the deploy tool to create such things	13:46
prjkt_	cmn, right	13:46
	how can I get it to track empty directories?	13:47
	← mmitchell left	13:47
	→ xpen joined	13:47
_ikke_	prjkt_: add an empty file to it	13:47
cmn	you dont	13:47
_ikke_	.gitkeep	13:47
cmn	you make it non-empty	13:47
_ikke_	for example	13:47
	→ mezner joined	13:47
prjkt_	now wait a minute	13:47
	← jonatasnona left	13:47
prjkt_	are you saying that Git doesn't even keep track of my directory structure?	13:47
	csmrfx waits	13:47
cmn	git can't stage dirs	13:47
_ikke_	prjkt_: it does	13:47
	→ timini joined	13:47
cmn	git stages files	13:47
prjkt_	or if i wanted to use it this way is the only 'bug' the fact that it won't track empty directories?	13:47
cmn	it's a design decision	13:47
	if you need a magic empty dir, create it	13:47
	← wrouesnel left	13:47
prjkt_	or are there any "bugs" (from the point of view of wanting it to track my directory structure) that I should know about?	13:48
	I did create it	13:48
cmn	like what? what issues are you encountering?	13:48
prjkt_	can you tell me about what other surprises I will run into?	13:48
osse	prjkt_: git perfectly tracks directory structures provided that the structure has files in every directory	13:48
_ikke_	prjkt_: We can't see into the future	13:48
prjkt_	I just told you the issue. I can't commit my changes, since my changes consisted of making a skeleton of empty directories.	13:48
timini	If im on a branch called testing and I do "git merge master" will that pull all the changes from master into my current branch?	13:48
prjkt_	okay	13:48
	← elixir__ left	13:48
osse	timini: yes	13:48
cmn	then make them non-empty; git only tracks files, track something inside them	13:49
EugeneKay	Except for hte fact that "pull" is also a git command, yes.	13:49
timini	thanks osse	13:49
	← prjkt left	13:49
_ikke_	prjkt_: !next	13:49
gitinfo	prjkt_: Another satisfied customer. NEXT!	13:49
prjkt_	let me ask this way. Suppose that I did a "commit", typed a lot of bash commands like moving rename creating etc files and folders. If I have to roll back to that commit, what other "breakage" would I have when trying to replay those bash commands, other than the fact that what bash works with would include empty directories whereas what I got from the repository wouldn't?	13:50
	any other surprises in store for me?	13:50
	→ elixir__ joined	13:50
cmn	again, we can't know unless we know what you're going to do	13:50
	if you need a particular empty dir, create it	13:50
	that's all you've told us	13:50
	→ FernandoBasso joined	13:50
prjkt_	well I'd like to use 'git' like a 'time machine' for my project directory	13:50
	I make some small changes, test everything, and commit the current version	13:51
osse	prjkt_: moved and created files: no problem. Git will restore them if you roll back.	13:51
prjkt_	as though it were a zip file or something	13:51
cmn	then you're probably looking for another tool	13:51
prjkt_	but it's my source code?	13:51
	which is why I thought I wanted Git?	13:51
_ikke_	prjkt_: git creates snapshots of the working tree, but empty dirs are not tracked	13:51
xckpd7	what would be a good way to find all Mac gui's that showed git diffs?	13:51
cmn	time-based backups isn't what version control is about	13:51
xckpd7	I know there is smart git, kailedoscope	13:51
_ikke_	prjkt_: renames aren't stored in git, but infered when needed	13:51
prjkt_	is that the ONLY exception to the definitino of "snapshots of the working tree"?	13:51
	ahhhhh	13:51
	that is a really really big one as well	13:52
cmn	big what?	13:52
prjkt_	that renames aren't stored	13:52
	a big exception that I could run into	13:52
cmn	are you looking for a backup tool or what?	13:52
osse	prjkt_: I don't think you'd ever have a problem with it.	13:52
cmn	have you experienced a problem?	13:52
_ikke_	prjkt_: git can follow renames quite well	13:52
	SeoZ-work[AWAY] → SeoZ	13:52
prjkt_	maybe I should do it like this: as part of my "testing" phase I zip my tree and place the zip file into the root of the tree. Then I can be 100% sure if committing it that Git really does have an absolute current version.	13:52
cmn	what are you actually trying to do? do version control on your project or keep timed backups?	13:52
prjkt_	Version control on my "project" but my project is very esoteric. it's more like system configuration. so you see...	13:53
	← beautifulmind left	13:53
	→ nitrix joined	13:53
	← nitrix left	13:53
	→ AaronMT joined	13:53
prjkt_	empty directories can be QUITE important, if I'm doing replay attacks of my own bash scripts....	13:53
	→ thl0 joined	13:53
cmn	then look into puppet, chef, or etckeeper	13:53
	← dhruvasagar left	13:53
prjkt_	No, because i'm developing an application	13:53
	← theotherguy left	13:53
prjkt_	git is great for that	13:53
	→ natechan joined	13:53
prjkt_	anyway just to be on the safe side, for now I'm going to run a script that zips up my repository and put it in the root before each commit.	13:54
	→ mastroWork joined	13:54
cmn	safe side of what?	13:54
osse	prjkt_: then I would use write a script to create the directories you need and version the script itself. So whenver you do a replay you run that script beforehand to set things up	13:54
lov	o_O	13:54
	→ szx joined	13:54
osse	prjkt_: basically a "Makefile"	13:54
	→ TeckniX joined	13:55
prjkt_	osse, that's a good idea but you see I'm doing things manually. it doesn't make sense to make, delete, rename, make another one, delete all of it and start over... all replayed from a script	13:55
	it makes sense to just store the end result	13:55
	just like you wouldn't store backspaces if you were replaying typing into notepad	13:55
	→ AnimeCYC joined	13:55
	→ Nibato joined	13:56
osse	prjkt_: then you're in a niche where git can't help you unfortunately :/	13:56
lov	prjkt_: perhaps SVN would be more to your liking.	13:56
cmn	it sounds like you really would want to do that	13:56
	→ rbmk joined	13:56
prjkt_	anyway later when I'm more comfortable and the project is larger (image files, whatever) can I can go back and retroactively delete the 'zip' file from all of the versions?	13:56
	I really like git	13:56
cmn	that's not how version control works, no	13:56
prjkt_	I just want to know its limitations and understand it	13:56
	cmn: it's impossible to meddle with the history of the version control?	13:56
	→ nitrix joined	13:56
lov	prjkt_: the limitation is that git is a mechanism for tracking files; folders are just an abstraction that the filesystem provides for you.	13:56
cmn	no, but it causes pain	13:56
	the point is not to	13:56
prjkt_	for example if I've been versioning my credit card number I can't remove it from all versions before pushing it to github and publishing it for the world??	13:56
cmn	if you're looking for backups, take backups	13:57
lov	prjkt_: meddling with history is fine unless anyone else has that history.	13:57
prjkt_	okay	13:57
cmn	you can do that, but you don't store data to begin with	13:57
	← maglinvinnn left	13:57
cmn	and that is a different case altogether, in any case	13:57
osse	prjkt_: if the end result is a bunch of empty directories, I would consider storing a list of those directories in result.txt instead of the directories themselves	13:57
prjkt_	lov, just to be super-clear. Imagine taht I have a single file I'm now versioning "my_credentials.txt" that has my credit card information and which I'm testing my banking software with. Will I be able to have a normal git history I can share with someoene else, that is as though that file never exisetd?	13:58
	→ x82_nicole joined	13:58
	→ Goles joined	13:58
cmn	you can rewrite it	13:58
	but it will be different history	13:58
lov	prjkt_: you can run commands that will rewrite history. It is important to note, however, that this will cause entirely new commits to be created, and your tree will be fundamentally different from others.	13:58
prjkt_	osse, let me tell you what I'm doing. I've created a bunch of empty skeleton directories. Then I push commit. Then I start working on one. This involves creating more skeleton directories. Then I push commit, etc.	13:58
cmn	simply don't store that data	13:58
prjkt_	it's the same as creating dummy code, only dummy directories.	13:58
	← mmlac left	13:58
cmn	this is for extraordinary cases	13:58
_ikke_	prjkt_: Just to be clear, you can rewrite the entire git history if you want	13:59
lov	yeah, to be honest, it's probably best to just not store the data. You can use .gitignore to ignore the file.	13:59
	← xpen left	13:59
prjkt_	lov: different from other trees?	13:59
	okay	13:59
	← nitrix left	13:59
	→ JaredMiller joined	13:59
_ikke_	prjkt_: Why are these skelleton directories so important?	13:59
lov	prjkt_: Lets say that you have A -> B -> C -> D. If you rewrite history to scrub a file that you introduced in B, your tree will be A -> B' -> C' -> D'. If anyone else has a copy of your repository, they'd have A -> B -> C -> D.	13:59
	→ jonatasnona joined	13:59
	← JaredMiller left	13:59
prjkt_	I mean it's just how I code. You would consider it pretty weird if photoshop didn't save any light pencil marks that you will later use to know where you want to actually draw, on the theory that these are "empty"	13:59
lov	This causes problems when you want to work together, because you're both working against fundamentally different HEADs.	14:00
prjkt_	right	14:00
	lov, but if nobody else has a copy already, then it's like it never existred, right?	14:00
cmn	what does photoshop have to do with this?	14:00
lov	that's correct, though you'll still need to futz with github to point to the new head.	14:00
osse	cmn: analogy	14:00
prjkt_	It's not actualyl on github now	14:00
cmn	but it doesn't apply	14:00
prjkt_	I'm just trying to find an alternative to using no source control	14:00
	→ nitrix joined	14:01
prjkt_	anyway thanks for the help	14:01
	one final thing.	14:01
	I heard some people do manage their config files with git. do you have any thoughts on this?	14:01
	← mtk left	14:01
prjkt_	by manage I mean version	14:01
	← gwty left	14:01
cmn	you can store files, it's no different than any other bunch of files you want to version	14:02
	but you'd have to symlink into your config repo	14:02
prjkt_	ok	14:02
	thanks everyone	14:02
	great feedback. talk to you later.	14:02
	*answers	14:02
_ikke_	prjkt_: photoshop doesn't store arbitrary data you want	14:02
	→ FilipeMaia joined	14:02
	← prjkt_ left	14:02
FilipeMaia	If I did git rm to a bunch of files and commited, how can I get back one of those files?	14:03
	Is it possible to revert the removal of one of those files?	14:03
EugeneKay	git checkout $SHA -- file.txt	14:03
cmn	see what status tells you	14:03
	← utopia_euphoria left	14:03
FilipeMaia	EugeneKay: was that an answer to me?	14:03
	→ PDL joined	14:03
EugeneKay	Yes	14:03
	→ Error404NotFound joined	14:03
	→ rednaks joined	14:04
FilipeMaia	EugeneKay: so i just checkout an old version and commit it?	14:04
	← nitrix left	14:04
	→ nitrix joined	14:04
EugeneKay	What you do with it is up to you	14:04
	← Err404NotFound left	14:04
FilipeMaia	EugeneKay: ok, thanks	14:04
	→ utopia_euphoria joined	14:04
	gitinfo set mode: +v	14:05
PDL	If there are 2 remote branches	14:05
	how do I know which branch I'm pulling the master from	14:05
	→ valgrind joined	14:05
	← fuzzybunny69y left	14:05
PDL	which remote*	14:05
	→ jcromartie joined	14:05
cmn	you've told git (or it was set up when you cloned)	14:05
PDL	Ok	14:06
	→ huoxito joined	14:06
	← guardadigital left	14:06
PDL	Is there any way to see?	14:06
cmn	though you presumably mean remotes, not remote branches	14:06
	← arvind_khadri left	14:06
cmn	git branch -vv	14:06
	→ mikecmpbll joined	14:06
PDL	Yeah remotes	14:06
	→ chrisgeorge joined	14:07
	→ mastro__ joined	14:07
	→ mklappstuhl joined	14:07
	→ kadoban joined	14:08
	→ monod joined	14:09
FernandoBasso	How to push only the master branch?	14:09
	→ arvind_khadri joined	14:09
PerlJam	FernandoBasso: git push origin master	14:10
	← mastroWork left	14:10
	← bartzy left	14:11
	→ BillCriswell joined	14:11
FernandoBasso	H'm, then I did something wrong...	14:12
	← bzyx left	14:13
	→ bartzy joined	14:13
	→ mtk joined	14:14
	→ bzyx joined	14:14
	→ raycloud joined	14:14
	→ axl_ joined	14:15
	→ mmitchell joined	14:15
	→ jaequery joined	14:15
	→ Tech163 joined	14:16
	→ aartist joined	14:16
	→ v0n joined	14:16
	← mmitchell left	14:17
	AAA_awright_ → AAA_awright	14:18
	→ Davey joined	14:18
	← mitz left	14:18
	→ mmitchell joined	14:19
	← buibex left	14:19
	→ mitz joined	14:19
	← mmitchell left	14:20
	← Tech163 left	14:20
	→ mmitchell joined	14:21
	← axl_ left	14:21
	← barneybook left	14:21
_ikke_	What might this be? http://stackoverflow.com/questions/15684197	14:22
	→ aisbaa joined	14:22
	← raycloud left	14:23
	→ simesy joined	14:23
	← arvind_khadri left	14:24
	→ Dr4k3 joined	14:24
	→ berndj joined	14:24
	← starlays left	14:24
	← netiul left	14:25
	→ netiul joined	14:26
	← Dr4k3 left	14:26
FernandoBasso	How can I fetch, and merge to say, master, without creating a commit message (when there is no conflict)?	14:27
	← puchrojo left	14:27
osse	FernandoBasso: git merge --no-commit	14:27
EugeneKay	You mean without typing one in as newer versions of git require?	14:27
	← simesy left	14:27
FernandoBasso	Thanks.	14:27
_ikke_	osse: Doesn't that just leave out the commit part, which you then still have to do yourself?	14:29
EugeneKay	You're probably looking for --no-edit	14:29
osse	_ikke_: indeed it does. Wasn't that the question?	14:29
	ohhhhh	14:29
	→ axl_ joined	14:30
	→ tevio joined	14:30
	← netiul left	14:32
	→ Will_Tampa joined	14:32
jphillips	Anyone have a good, detailed explanation of merge commits? I want to know a lot more about them	14:32
EugeneKay	!concepts is helpful	14:33
gitinfo	"Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs/	14:33
	→ dsdeiz joined	14:33
EugeneKay	Basically, it's a commit with two parents. The process of combining the two parents is the more interesting bit, and quickly devolves into merge resolution strategies	14:33
	tashi_ → tashi	14:33
	← jaseemabid left	14:33
	← Schmoozerd_ left	14:34
jphillips	so basically a commit with 2 parents	14:34
EugeneKay	That's what I said, yes.	14:35
	→ mastroWork joined	14:35
	→ rikko joined	14:35
_ikke_	jphillips: and potentially, even more than two, although, you should never want that	14:35
	← maxmanders left	14:36
	← rakl left	14:37
	→ mmitchel_ joined	14:37
	← PDL left	14:37
EugeneKay	Meh, octomerges aren't as bad as they're madeo ut to be	14:37
	→ Schmoozerd joined	14:37
	→ maxmanders joined	14:37
	← rob_w left	14:38
jphillips	!octomerge	14:38
	← nihilista left	14:38
	← mastro__ left	14:38
	→ raycloud joined	14:38
	← mastroWork left	14:38
	← normanrichards left	14:39
	← csmrfx left	14:39
	→ netiul joined	14:39
EugeneKay	ctrl-f for octopus in man git-merge	14:40
gitinfo	the git-merge manpage is available at http://jk.gs/git-merge.html	14:40
	→ nihilista joined	14:40
	← roger_raymond left	14:40
	→ ojacobson joined	14:40
	← mmitchell left	14:41
	→ roger_raymond joined	14:41
	→ chrisdev joined	14:42
osse	I like octomerges	14:43
	they look cool	14:43
	← sysadminAli left	14:43
osse	What's wrong with them?	14:43
	→ stliu joined	14:43
	← zoresvit left	14:43
	EugeneKay shrugs	14:43
Seveas	They make peoples brain explode I think. I use them only for one project	14:44
	→ juvenal joined	14:44
Remram	is there a way to see which commits have not been pushed to a specific remote?	14:44
	← DonRichie left	14:44
	SeoZ → SeoZ-work[AWAY]	14:45
EugeneKay	Remram - !lol is good for visualizing that sort of thing	14:45
gitinfo	Remram: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	14:45
Seveas	3rd party tool where I have per-feature branches that have not yet been accepted upstream. And a 'camel' branch which I constantly rebuild as an octomerge between all these branches	14:45
	→ hwrd\|work joined	14:45
Remram	I use log --branches --not --remote but it uses all remotes	14:45
EugeneKay	Remram - `git branch -vv` too	14:46
rking	I wonder if there's a way to 'git blame' where the commit subject is at the end of the line	14:47
	I bet it would take a bit of scripting, shouldn't be too hard, though.	14:47
	← rabellamy left	14:47
Remram	branch -vv is good enough, thanks!	14:47
	→ Milossh joined	14:48
	← Milossh left	14:48
	→ Milossh joined	14:48
	→ danishkhan joined	14:49
	← Milossh left	14:49
	→ Dr4k3 joined	14:50
	← FernandoBasso left	14:50
	← bgerber left	14:50
	← soee left	14:51
	← LeMike left	14:51
_ikke_	RandalSchwartz: https://github.com/cstar/heroku-buildpack-zotonic	14:51
Remram	rking: you can use awk...	14:52
	→ sysadminAli joined	14:52
Remram	git blame ... \| awk -F"\t" '{print $4 "\t" $2}'	14:53
	← elixir__ left	14:53
	→ mastroWork joined	14:54
	→ thiago joined	14:54
	→ marhaban joined	14:54
	← maxmanders left	14:56
	→ momomomomo joined	14:56
	← bzyx left	14:56
	→ maxmanders joined	14:56
	→ mastro__ joined	14:57
	← huoxito left	14:58
Remram	EugeneKay: hmm, turns out this is only useful for branches set to track remote branch for the remote I'm interested in	14:58
	→ giallu joined	14:58
	→ bzyx joined	14:58
EugeneKay	Hence the other suggestion first	14:58
	→ bgerber joined	14:58
	→ DonRichie joined	14:58
	→ jjore joined	14:59
	→ barneybook joined	14:59
	→ fly9 joined	14:59
	← rchavik left	14:59
Remram	well the other doesn't do what I want, I have to look around for every remote branch from that repo and check whether it's up to date	14:59
	→ simesy joined	15:00
Remram	I found git log --branches --not --remotes=<myremote>, does what I want. Thanks anyway!	15:00
	→ zoresvit joined	15:00
	→ memoryleak joined	15:00
	→ mastro joined	15:00
	← mastroWork left	15:01
	← maxmanders left	15:01
	← simesy left	15:01
	← jacobbednarz left	15:01
	→ simesy joined	15:02
	← sheerun left	15:03
	→ goshawk joined	15:03
	← mastro__ left	15:03
rking	Remram: That's not quite it. Thanks though.	15:04
	→ nioncode joined	15:04
rking	Give me a few more minutes and I'll push the version I came up with.	15:04
	→ mastroWork joined	15:04
Remram	some kind of regexp would probably give you more flexibility	15:04
	→ mustmodify joined	15:05
	← chrismar035 left	15:06
	← simesy left	15:06
	→ mastro__ joined	15:07
	→ LeMike joined	15:07
	← mastro left	15:07
	→ vedant joined	15:07
	← fly9 left	15:07
	→ sheerun joined	15:08
	→ chrismar035 joined	15:08
	→ Dougie187 joined	15:08
	→ mastro joined	15:10
	fnoyanisi → fnoyanisi_away	15:10
rking	https://raw.github.com/sharpsaw/git-dots/d936f70efef783b82f13114f4e83321fa8aeb694/bin/gblamec	15:10
	← hwrd\|work left	15:10
	← mastroWork left	15:10
rking	A regex isn't enough. I want the commit message itself.	15:10
	← fnoyanisi_away left	15:11
rking	If you get that script, then do like: gblamec foo.rb \| vim - +ft\ ruby	15:11
	You'll see the gootness.	15:11
	→ dhruvasagar joined	15:12
	← bartzy left	15:13
	← mastro__ left	15:13
	→ bartzy joined	15:13
	→ mastroWork joined	15:13
	→ agu10^ joined	15:13
agu10^	Hello! I have a VPS running ubuntu server, and i develop on windows 8. Is there anyway to upload my website to my server using git?	15:14
_ikke_	!deploy	15:14
gitinfo	Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html	15:14
agu10^	such as git push server origin ?	15:14
	→ buibex joined	15:15
_ikke_	agu10^: also read !bare	15:15
gitinfo	agu10^: an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/	15:15
	← pretty_function left	15:16
agu10^	"Deploying on a push to a bare repo should be easy. Just add a post-receive hook that contains this code:"	15:16
	← skyxxx left	15:16
agu10^	sure but am i supposed to know what such a hook is?	15:16
	← mastro left	15:16
cmn	man git hooks	15:17
gitinfo	the git manpage is available at http://jk.gs/git.html	15:17
grawity	man githooks	15:17
gitinfo	the githooks manpage is available at http://jk.gs/githooks.html	15:17
cmn	man githooks	15:17
gitinfo	the githooks manpage is available at http://jk.gs/githooks.html	15:17
	→ manearrior joined	15:17
	→ JaredMiller joined	15:17
	← duosrx left	15:18
	← averiso left	15:18
	→ maxmanders joined	15:18
	← JaredMiller left	15:18
	← mastroWork left	15:18
	→ Natureshadow joined	15:19
	→ FernandoBasso joined	15:19
	→ jaseemabid joined	15:20
	→ averiso joined	15:20
	→ joshsmith joined	15:20
	→ jeffreylevesque joined	15:21
	→ w0m joined	15:21
	← papegaaij left	15:22
	→ darnir joined	15:23
	← rahulprasad left	15:23
	← timini left	15:24
	← jjore left	15:24
	→ howlymowly joined	15:24
	→ oakdog8 joined	15:25
	← tvw left	15:26
	← valgrind left	15:26
	← danishkhan left	15:27
	← maletor left	15:27
	← thyseus left	15:28
	→ mviitanen joined	15:29
	← dsdeiz left	15:29
	→ puchrojo joined	15:30
	← memoryleak left	15:31
	← _Qman left	15:31
	→ AdrienBrault joined	15:31
xckpd7	what gui do you guys use to merge? because someone suggested that a GUI is better for identifying changes within a line rather than a terminal which generally will only show if a line changed	15:31
	my problem is that I need a mac merge tool, but kaleidoscope doesn't work because it doesn't support merging by hand	15:32
	→ thyseus joined	15:32
_ikke_	kdiff3	15:32
	← manearrior left	15:32
_ikke_	Is what I use	15:32
	It's cross platform	15:32
	→ memoryleak joined	15:33
	← sunya7a left	15:33
	← dhruvasagar left	15:33
	→ mastroWork joined	15:33
	← thiago left	15:34
	→ Esa_ joined	15:38
	gitinfo set mode: +v	15:38
	→ mastro__ joined	15:38
mustmodify	the auto-generated diff doesn't do a good job of expressing the intent of my change... can I either provide git with some hints as to where changes start and end, or create my own diff file that would be used as the commit?	15:38
	← natechan left	15:38
_ikke_	mustmodify: No	15:39
	mustmodify: diffs aren't stored	15:39
xckpd7	_ikke_: thanks	15:39
_ikke_	mustmodify: They are calculated when needed	15:39
	mustmodify: And git doesn't provide anything to influence that	15:39
mustmodify	_ikke_: ok. thanks-ish. :)	15:40
	← mastroWork left	15:41
	← Davey left	15:41
_ikke_	Can't help it	15:42
	→ mastroWork joined	15:42
	→ btree joined	15:42
	→ Underbyte joined	15:43
	→ maletor joined	15:43
	← bzyx left	15:43
	← btree left	15:44
	→ bzyx joined	15:44
osse	isn't it possible to provide a seperate "differ" ?	15:44
	with gitattributes or somesuch?	15:44
	→ jordanm joined	15:45
	→ mastro joined	15:45
	← jonatasnona left	15:45
_ikke_	Yeah, you can replace the diff engine	15:45
	← mastro__ left	15:45
	← jonshea left	15:46
	→ kryl99 joined	15:47
FernandoBasso	I push a change to a remote repository, my coworker fetch it (it end in remote/origin/master). Then he runs "git checkout -b tmp origin/master", and merge into master, but the changes are not in master.	15:47
	And sometimes, git fetch says everything is up to date even though they are not.	15:48
jordanm	We have a git structure with a master, release branches, and feature branches. If a feature gets merged into a release branch and then the merge is reverted to pull it out of a particular release, it becomes no longer possible to merge those feature branches into future releases. The original release branch would have been merged into master and a new release branch created off of the new master	15:48
	is there a way to work around or avoid this type of issue?	15:48
	→ dusted joined	15:48
	← mastroWork left	15:48
	← marcdel left	15:49
	→ danishkhan joined	15:50
	→ mastroWork joined	15:50
osse	FernandoBasso: you have to checkout master to be able to merge into it	15:50
FernandoBasso	osse: Yes, we did that.	15:50
milki	jordanm: if you rebase the release branch, it will be mergeable again	15:50
	jordanm: rather the feature branch	15:51
osse	FernandoBasso: then I don't know what's wrong. But there is no need to checkout a temp branch, unless you want to.	15:51
	→ rcu joined	15:51
jordanm	milki, rebase onto master, or the new release branch?	15:51
FernandoBasso	osse: Well, thanks for helping. Perhaps I did something wrong.	15:52
milki	jordanm: sounds like it should be master	15:52
jordanm	milian, thanks for the help	15:52
milki	jordanm: after the original release was merged into master	15:52
	← milian left	15:53
	← mastro left	15:53
	→ mastro__ joined	15:53
osse	FernandoBasso: if your coworker does this (the two first commands in any order): git checkout master; git fetch; git merge origin/master - then everything should be dandy.	15:53
	I don't know why they aren't	15:53
	← DoniRS left	15:54
	→ dhruvasagar joined	15:54
	← Asenar left	15:54
	→ natechan joined	15:56
	← mastroWork left	15:56
	→ Davey joined	15:57
	→ NiteRain joined	15:57
	→ mastroWork joined	15:58
	← goshawk left	15:58
	→ rakl joined	15:58
	← mmitchel_ left	15:58
	→ mmitchell joined	15:58
	← joshsmith left	15:59
	→ joshsmith joined	16:00
	← chrismar035 left	16:00
	→ gwty joined	16:01
	← mastro__ left	16:01
	← m0viefreak left	16:01
	→ hyperair joined	16:01
	→ mastro__ joined	16:02
	← trungqtran left	16:02
jordanm	milki, that was perfect, thanks again	16:03
	→ chrismar035 joined	16:03
FernandoBasso	osse: The strangest parte is when I pushed, and he tried to fetch but nothing was fetched. Very strange.	16:03
	← mezner left	16:03
	← mustmodify left	16:03
cbreak-work	what would you want to fetch?	16:04
osse	FernandoBasso: maybe some configuration stops git from fetching origin/master unless you explicitly name it	16:04
	→ Cromulent joined	16:04
FernandoBasso	Actually, we have a basic plain config. user, email, core.editor...	16:04
	← mastroWork left	16:05
	← dhruvasagar left	16:05
	← netj left	16:06
	← mikecmpbll left	16:06
	← thl0 left	16:06
	→ rahulprasad joined	16:06
	→ ScottE joined	16:06
	← zoresvit left	16:06
	← maletor left	16:07
	← decotii left	16:07
	← netiul left	16:08
	← tinti left	16:08
	← vedant left	16:08
	→ skyxxx joined	16:08
	← hyperair left	16:09
	→ netj joined	16:09
	← LeMike left	16:09
	→ Gentlecat joined	16:09
	→ dmarkham_ joined	16:09
	→ vedant joined	16:10
	→ johnflux joined	16:10
	← cbreak-work left	16:10
johnflux	I'm trying to do: git clone git://blah	16:11
	← heytrav left	16:11
johnflux	and it says "fatal: I don't handle protocol 'git'"	16:11
	← fooey left	16:11
FernandoBasso	What does git --version say?	16:11
	→ nezZario joined	16:12
nezZario	more git confusion.. what would be the rough syntax to have different fetch and pull addresses for origin?	16:12
	→ Parhs joined	16:12
	← Katafalkas left	16:13
johnflux	nezZario: instead of origin, have "originA", "originB" etc or whatever oyu want to call them?	16:13
	→ ccolorado joined	16:13
johnflux	FernandoBasso: 1.8.1.2	16:13
	→ Targen joined	16:13
johnflux	FernandoBasso: specifically I'm doing: git clone git://github.com/CyanogenMod/android_kernel_samsung_smdk4412.git	16:13
	→ Katafalkas joined	16:13
FernandoBasso	johnflux: it is generally git@ not git://	16:13
_ikke_	nezZario: I believe git has an option to do what you want\	16:13
	← x82_nicole left	16:13
jast	nezZario: since pull just runs 'fetch', there's no such thing, though of course you can define two different remotes and use a different one for each operation	16:13
_ikke_	oh, I need to read better	16:14
	← threesome left	16:14
nezZario	sorry, crap, I meant fetch and push	16:14
	bitbucket is a bit annoying in that it doesn't allow write access via keys, although I sync servers with a automated git pull	16:14
_ikke_	git remote show -n shows fetch and pull urls	16:15
	so they can be different	16:15
	← Raging_Hog left	16:15
jast	nezZario: man git-config, search for 'pushurl'	16:15
gitinfo	nezZario: the git-config manpage is available at http://jk.gs/git-config.html	16:15
nezZario	yeah, i see that, I just don't know how to say "this one is fetch" and not replace both	16:15
	k, that's what i needed	16:15
	→ martinFTW joined	16:15
FernandoBasso	I wonder, can one push branch_x into the remote branch master?	16:15
johnflux	git clone [email@hidden.address]	16:15
	fatal: repository 'git@github.com/CyanogenMod/android_kernel_samsung_smdk4412.git' does not exist	16:15
	FernandoBasso: ^^	16:15
jast	FernandoBasso: sure... git push yourremote branch_x:master	16:15
_ikke_	git config remote.<remote>.pushurl <url>	16:16
FernandoBasso	jast: I am confused with this : syntax.	16:16
jast	FernandoBasso: it's a refspec, documented in man git-fetch/push	16:16
gitinfo	FernandoBasso: the git-fetch manpage is available at http://jk.gs/git-fetch.html	16:16
FernandoBasso	johnflux: Now it says the repository does not exist.	16:16
	→ Cache_Money joined	16:16
johnflux	FernandoBasso: okay using the https protocol works - wonder why it failed with the git:// version though?	16:16
	→ LeMike joined	16:16
	← nielsbusch left	16:16
jast	johnflux: the original URL was correct, and that it didn't work suggests that your git is broken/incomplete	16:16
johnflux	jast: strange :)	16:17
jast	yes	16:17
	← babilen left	16:17
_ikke_	nezZario: with git config you can do it	16:17
	or git remote set-url <remote> --push <url>	16:17
	→ nielsbusch joined	16:17
	← szx left	16:17
	→ babilen joined	16:18
	← babilen left	16:18
	← martinFTW left	16:18
_ikke_	btw, I question yesterday which I couldn't precicely answer, what is the difference between git branch -t and git branch -u	16:18
	→ etcetera joined	16:19
	← psoo left	16:19
	← tsdh left	16:19
	← aartist left	16:19
	→ hwrd\|work joined	16:20
	dwmw2 → dwmw2_gone	16:20
	→ cannonball joined	16:20
	spyroboy_ → spyroboy	16:20
	← rath left	16:21
jast	_ikke_: (1) git branch -t doesn't exist; (2) the difference between --set-upstream and -u is the order of arguments	16:21
	→ mastroWork joined	16:21
jast	(-u defaults to changing the current branch and you just need to specify the upstream)	16:21
	→ aartist joined	16:21
_ikke_	jast: my manual mentions -t	16:21
	man git branch	16:22
	→ heytrav joined	16:22
jast	ah, yeah, now it does	16:22
	-t works for newly created branches only	16:22
_ikke_	right, that's what I thought	16:22
jast	otherwise it's equivalent ot --set-upstream which is different from -u	16:22
	(which is shorthand for --set-upstream-to)	16:22
	it's a mess :)	16:22
_ikke_	yeah	16:23
	the latter is newly added to fix the former	16:23
jast	yeah	16:23
	← LeMike left	16:23
ccolorado	Hello, I am tring to do git mv vim home/vim I get the following error fatal: source directory is empty, source=vim/bundle/ctrlp.vim, destination=home/vim/bundle/ctrlp.vim ( git 1.8.2 ). The only relevant thing I can find about it referes to a pacth for v1.4.4 back on 2006. this directory is not empty "vim/bundle/ctrlp.vim". Any ideas ?	16:23
	→ jonshea joined	16:24
	← mastro__ left	16:24
	→ paddymahoney joined	16:25
	→ Copter2 joined	16:25
	← catphish left	16:25
	← eliasp left	16:25
	← FernandoBasso left	16:25
	→ catphish joined	16:25
	← gitinfo left	16:25
	→ eliasp joined	16:26
	← alexander__b left	16:26
	← JStoker left	16:27
	→ maletor joined	16:27
	→ x82_nicole joined	16:27
	→ alexander__b joined	16:28
	← Cromulent left	16:28
	→ decotii joined	16:28
	→ dhruvasagar joined	16:28
jast	aww, he's gone	16:28
	→ otters joined	16:28
	→ ofcan joined	16:28
otters	how do I move working changes from one branch to another?	16:28
jast	I made a picture to explain refspecs: http://qkme.me/3tke7r	16:28
	→ JStoker joined	16:29
ofcan	when I fork a repo from github, what does default origin point to?	16:29
jast	ofcan: origin always points to whatever URL you use when cloning	16:29
grawity	when you fork, or clone?	16:29
jast	otters: uncommitted changes, you mean	16:29
	?	16:29
	← Bombe left	16:30
	→ Bombe joined	16:30
_ikke_	jast: haha :D	16:30
	jast: genius	16:30
otters	jast: yes	16:30
jast	otters: !float	16:30
otters	right now I do git stash; git checkout correct-branch; git stash pop	16:30
jast	hey, the bot is dead	16:31
	WHY :(	16:31
ofcan	jast: what use do I have of it?	16:31
	→ fooey joined	16:31
_ikke_	ofcan: use of what?	16:31
	← rahulprasad left	16:31
	→ Chillance joined	16:31
jast	oh, it's having trouble reconnecting	16:31
ofcan	jast: Can I push to that origin?	16:31
_ikke_	ofcan: yes	16:31
jast	ofcan: if you cloned your fork via a https:// or ssh:// or git@ type URL, yes	16:31
	← xckpd7 left	16:31
	→ gitinfo joined	16:31
	ChanServ set mode: +o	16:31
	← Katafalkas left	16:31
	→ theotherguy joined	16:31
jast	otters: !float	16:32
	wait for it... :)	16:32
_ikke_	!botsnack	16:32
jast	wait for it...	16:32
	← smirn0v left	16:32
jast	just a few minutes to go...	16:32
gitinfo	otters: If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there. If the files you changed differ between branches, the checkout will fail. In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.	16:32
spyroboy	waiting ..	16:32
jast	there!	16:32
gitinfo	Om nom nom	16:32
ofcan	_ikke_: but when I push to it I get requested for username and password and when I enter mine it fails	16:32
jast	ofcan: then you probably cloned a repository to which you don't have push access	16:32
	be sure to clone your fork and not the original!	16:32
_ikke_	ofcan: Did you clone your fork, or the original repository?	16:32
jast	do we have !refspec	16:33
	← aartist left	16:33
jast	we don't, yay	16:33
ofcan	_ikke_: ah, I cloned the original. I can't push because I'm not the owner of that repo, right?	16:34
otters	oh cool	16:34
jast	.trigger_edit refspecs Refspecs are used by fetch/push to specify which refs to transmit where. They have the form "source:destination". More info: http://jk.gs/git-fetch.html or http://i.qkme.me/3tke7r.jpg	16:34
gitinfo	jast: Okay.	16:34
	→ aartist joined	16:34
_ikke_	jast: (+)source:destination, right?	16:34
	← stliu left	16:34
	→ mmlac joined	16:35
	→ Saby joined	16:35
	← Saby left	16:35
	→ Saby joined	16:35
jast	.trigger_edit refspecs Refspecs are used by fetch/push to specify which refs to transmit where. They have the form "source:destination". They can be prefixed with a "+" to force the update, possibly displacing existing history. More info: http://jk.gs/git-fetch.html or http://i.qkme.me/3tke7r.jpg	16:35
gitinfo	jast: Okay.	16:35
	← LekeFly left	16:35
	→ fobban joined	16:36
	← lavaman left	16:36
jast	... and justice is restored	16:36
	← Vinnie_win left	16:36
	→ thl0 joined	16:36
	← thl0 left	16:36
	→ pen joined	16:37
	→ thl0 joined	16:37
	← tbosua left	16:37
	→ cyberbob joined	16:38
cyberbob	hi all I am new to git, and want to install some good gitweb interface on a Linux macine for my personal git repository	16:38
jast	if it's just for looking at your commit history and files and such, gitweb and cgit are the classics	16:39
	anything else comes with a lot of bells and whistles and is correspondingly harder to set up	16:39
	← Cache_Money left	16:39
cyberbob	Can someone guide me to a better guide not find enough information on the web no sure If I am too lazy in that :(	16:39
	← otters left	16:39
jast	gitweb is the most 'official' thing	16:39
	it's basically just a CGI script	16:39
	works with anything	16:39
	on debian you can simply install it as a package; other distros might have similar things	16:40
	← joshsmith left	16:40
	← rurufufuss left	16:40
	← jxpx777 left	16:40
cyberbob	jast: I want to have a classic interface like this as well http://repo.or.cz/w/vlc.git	16:40
	← Mez left	16:40
	→ swiety_ joined	16:40
cyberbob	I have a centos machine	16:40
	→ Mez joined	16:40
	← Mez left	16:40
	→ Mez joined	16:40
grawity	AFAIK that is the default gitweb interface...	16:41
jast	cyberbob: that's gitweb	16:41
	it's part of the git source distribution	16:41
	→ zymaster joined	16:42
jast	centos might have it too, no idea	16:42
zymaster	Hey quick question. So I installed gitosis on my server. Now when I try to ssh into my server as the git user I do not get blocked. How do I fix that? Not every git user on my server should have ssh access	16:42
EugeneKay	cyberbob - !host_gui	16:43
gitinfo	cyberbob: If you want to mange your hosted git repos through a web GUI, you have a few options: !gitlab !gitorious !rhode_code and !gerrit. Keep in mind that all of these are less flexible and more resource-intensive than !gitolite	16:43
EugeneKay	Gitlab and Gitorious are the most github-like.	16:43
jast	zymaster: !gitosis	16:43
gitinfo	zymaster: gitosis is no longer maintained and supported by the author; we usually recommend gitolite instead which has much better documentation and more features: http://github.com/sitaramc/gitolite -- if you're already stuck with gitosis we'll try to help, but no promises!	16:43
EugeneKay	If you're not dead-set on hosting it yourself you can just use a free Bitbucket private repo(or pay Github $7/mo for a Basic plan)	16:43
jast	zymaster: chances are you accidentally left your own SSH key from before the gitosis setup in .ssh/authorized_keys for the gitosis user	16:44
zymaster	Ok then, does anyone know how to get rid of gitosis?	16:44
jast	gitolite's docs have a migration guide	16:44
	← buibex left	16:44
	→ rahulprasad joined	16:44
sitaram	zymaster: http://gitolite.com/gitolite/gsmigr.html	16:44
zymaster	thanks all	16:45
jast	sitaram: do you have a highlight on gitolite or something? :)	16:45
	← mastroWork left	16:45
	← LimeBlast left	16:45
	← thl0 left	16:45
jast	or worse, on "gitosis" :)	16:45
	adhawkins → adhawkins-away	16:47
Dougie187	How do you merge branches from two separate repositories, if they don't have a common commit?	16:47
EugeneKay	git doesn't care. It just merges them.	16:47
Dougie187	I get this. Unable to find common commit with master	16:48
	Automatic merge failed; fix conflicts and then commit the result.	16:48
	← dhruvasagar left	16:48
charon	only works conflict-free if the file sets are disjoint though	16:48
_ikke_	Dougie187: Fix the conflicts	16:48
charon	Dougie187: what is the actual issue? do you have two disjoint histories for "the same" project?	16:48
	← rcu left	16:48
Dougie187	charon: no. Two independent repositories that were never connected.	16:48
EugeneKay	That's just the normal conflict message	16:48
Dougie187	How do I fix the conflicts? (or at least get a list of them?)	16:48
_ikke_	git status	16:49
EugeneKay	`git status`	16:49
_ikke_	!eek	16:49
gitinfo	[!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, don't panic. Read "How to resolve conflicts" in man git-merge. Carefully go through the conflicts. Picking one side verbatim is not always the right choice!	16:49
_ikke_	Dougie187: I can strongly recommend a mergetool	16:49
jast	of course, with no common history there's no merge base, so you don't get 3-way merging	16:49
	← spuz left	16:49
jast	right? :)	16:49
	← ropo left	16:49
zymaster	Hey guys I don't actually have any gitosis repositories and migration looks a little too complex for no purpose so is there a way to just uninstall gitosis completely?	16:49
	← Copter2 left	16:50
PerlJam	zymaster: just ignore its existence and there's no need to uninstall :)	16:50
jast	zymaster: remove its files (if you can find them), remove the gitosis user, delete its home dir	16:50
charon	jast: technically i suppose it does use a base: the empty tree	16:50
	→ breck7 joined	16:50
jast	charon: point	16:50
charon	(which is different from a 2-way merge)	16:51
Dougie187	Ok, thanks	16:51
	:D	16:51
	→ Noldorin joined	16:51
jast	zymaster: if you can't find the files (i.e. gitosis's libs and such), it won't kill you if you just leave them be... just a bit of dead weight	16:51
	← averiso left	16:51
jast	the only way in which gitosis "hooks into" your system is in the .ssh/authorized_keys flie	16:51
_ikke_	charon: In what way is it different?	16:51
	→ canton7 joined	16:52
	→ BadDesign joined	16:52
	← ofcan left	16:52
charon	_ikke_: that three-way-with-empty-base merge fails if two files exist on both sides and have different contents	16:52
	→ averiso joined	16:52
charon	whereas a read-tree-style two-way merge's job is to update files ;)	16:52
	← alexander__b left	16:54
	→ alexander__b joined	16:54
	← JohnBat26 left	16:54
	→ rath joined	16:55
sitaram	zymaster: as far as I can recall, uninstalling gitosis is (a) removing its lines from ~/.ssh/authorized_keys (2) removing ~/repositories/gitosis-admin. I don't remember anything else	16:55
zymaster	sitaram: its alright I figured it out. Thx all switching to gitolite	16:56
	← shreya left	16:56
	← alorence left	16:57
	→ etcetera_ joined	16:57
	→ tvw joined	16:58
	← gwty left	16:58
	→ ehsan joined	16:58
	← x82_nicole left	16:58
	→ dhruvasagar joined	16:58
	← axl_ left	16:59
	← vedant left	16:59
	← etcetera left	16:59
	etcetera_ → etcetera	16:59
	→ axl_ joined	16:59
	← maxmanders left	16:59
	→ nilg` joined	17:00
	← lindenle left	17:00
	→ hellopat_ joined	17:00
	→ x82_nicole joined	17:00
	→ dreiss joined	17:00
	→ lindenle joined	17:02
	← nilg left	17:02
	← lindenle left	17:02
	→ trungqtran joined	17:02
	→ spuz_ joined	17:03
	→ simesy joined	17:03
	→ thiago joined	17:03
	← hellopat left	17:03
	← spuz_ left	17:04
	→ itatitat joined	17:04
	→ spuz joined	17:04
	→ kit joined	17:05
	← flijten left	17:05
	→ ch3sh1r joined	17:05
	← AdrienBrault left	17:06
	→ LeMike joined	17:06
	← YamakasY2 left	17:06
	→ thl0 joined	17:07
	→ AdrienBrault joined	17:07
	→ tirengarfio_ joined	17:07
	→ dukedave joined	17:07
itatitat	guys I downloaded some files from a server by ftp. Those files were from a git repostiroy in github. The problem now I created a git in the same folder, and look that they fight.....so I removed everything, git folder and files, to start my working copy like if I never downloaded files from ftp. Now I want to get them from github. But the problem!!!!, is that everytime I remove those files and I create a new working copy, they appear like	17:07
	"deleted"	17:07
	how can I forget everything about those files?	17:07
	← simesy left	17:08
	→ mikef joined	17:08
	← hellopat_ left	17:08
spuz	itatitat: sounds like you downloaded a .git folder from this ftp site	17:08
itatitat	yes, exactly	17:08
spuz	make sure you remove that - make sure you know what you're doing though as you could lose lots of work if you aren't careful	17:09
	→ hellopat joined	17:09
spuz	(standard disclaimer)	17:09
itatitat	yes of course, I didnt push anything	17:09
	can be a disaster, you're right	17:09
	but how can I remove everything?, I created a new folder, with a new working copy inside, everything clean, but later, if I make "git status", I see many files like "deleted"	17:10
	its suppose to be a new working copy...	17:10
	← mgaunard_ left	17:10
	← mmlac left	17:10
spuz	itatitat: if you create a new folder and there is a .git directory somewhere in the directory tree above it, git will assume you are in that repository	17:11
	→ threesome joined	17:11
spuz	so look for the .git repo in your tree	17:11
itatitat	jummm.........I havent git in a superior level.....	17:12
	ok, I'll make that	17:12
	→ gwty joined	17:13
	→ jjore joined	17:13
	→ blackjack joined	17:14
	→ marcdel joined	17:14
spuz	itatitat: you must do, otherwise when you run 'git' it would say 'fatal: Not a git repository (or any of the parent directories): .git'	17:14
	→ elixir joined	17:15
	→ vedant joined	17:15
grawity	git rev-parse --git-dir	17:15
	← Guest45369 left	17:16
	→ j416 joined	17:16
	← agu10^ left	17:16
itatitat	I did it again everything. I made git pull, git status and they appear again "deleted"	17:16
	was an empty folder on the top	17:16
	→ jonshea_ joined	17:17
itatitat	in a different folder	17:17
	→ tinti joined	17:17
	→ josefig joined	17:18
	← jonshea left	17:18
	jonshea_ → jonshea	17:18
	← josefig left	17:18
	← revyd left	17:19
	← Squarism left	17:19
	← memoryleak left	17:19
	← Squarism2 left	17:19
	→ archdori_ joined	17:19
	← twisted` left	17:20
	← ChrisAnn left	17:20
	→ FatDarrel joined	17:21
	← robbyoconnor left	17:21
	← alunduil left	17:21
	→ robbyoconnor joined	17:21
	← beawesomeinstead left	17:22
	← vedant left	17:22
	→ sunya7a joined	17:22
	← jab416171 left	17:24
	← dytrivedi left	17:24
	→ hyperair joined	17:24
	→ prjkt joined	17:25
	← fRAiLtY- left	17:25
	← sheerun left	17:25
prjkt	hi, I'd like to know - does one repository only incldue one .git folder at the top level? Or can subfolders also have .git folders that are generated automatically when I commit etc at the top level?	17:26
	→ dytrivedi joined	17:26
ojacobson	prjkt: a .git folder is the repository.	17:26
	← breakingthings left	17:26
prjkt	ojacobson, so there's only one.	17:26
	← jjardon left	17:26
	← nielsbusch left	17:27
	→ jdunck joined	17:27
prjkt	Do you think it's safe to move the .git file and then move it back	17:27
	I can't find an easy way to ask winrar to ignore the .git file, but I want it to archive the whole repository.	17:27
	→ dmsuperman joined	17:27
ojacobson	man git-archive	17:27
gitinfo	the git-archive manpage is available at http://jk.gs/git-archive.html	17:27
prjkt	so the batch file would: move the .git folder . if it was moved successfully create the archive. move the .git folder back	17:27
	will that foul anything up?	17:27
dmsuperman	Using git log, how can I get --cherry-pick in conjunction with --all? It doesnt' seem to have any impact when using them both	17:28
prjkt	ok that's pretty good	17:28
	but I do like winrar	17:28
dmsuperman	I want to list all my commits for a week excluding any cherry picks	17:28
	← kit left	17:28
	← infynyxx left	17:28
prjkt	actually on second thought given that there is windows explorer integration it's probably better not to move the.git file around	17:28
dmsuperman	whereas --cherr-pick seems to expect you to be logging the differences between two branches	17:28
j416	prjkt: do you have to use rar? otherwise there's git-archive	17:30
	prjkt: man git-archive	17:30
gitinfo	prjkt: the git-archive manpage is available at http://jk.gs/git-archive.html	17:30
	← dytrivedi left	17:30
j416	prjkt: and yes, you can move the .git dir someplace else and then move it back, no problem.	17:30
	→ vedant joined	17:30
prjkt	okay, thank you	17:30
	→ sheerun joined	17:30
prjkt	I will see how this does	17:30
dmsuperman	right now I'm resorting to taking every single item from git log, fetching a git patch-id from it, and comparing those, but that runs pretty slowly for just 10-20 commits	17:31
	since it grows exponentially with each commit	17:31
	← ehsan left	17:32
	→ kit joined	17:33
	→ huoxito joined	17:33
	← FilipeMaia left	17:33
	← Saby left	17:34
	← mitz left	17:34
	← hellopat left	17:35
	→ hellopat joined	17:35
	→ alvaro_o joined	17:35
	→ Saby joined	17:36
	← Saby left	17:36
	→ Saby joined	17:36
	→ normanrichards joined	17:36
	← ignas left	17:36
	→ diimdeep joined	17:37
	← dreiss left	17:37
	← catphish left	17:37
	← bmcorser_ left	17:38
	→ maestrojed joined	17:38
	← dvide left	17:39
	→ diimdeep_ joined	17:40
	← Goles left	17:40
zymaster	Hey in gitolite what is the difference between RW and RW+	17:40
	← TheFuzzball left	17:40
	← itatitat left	17:40
	← achadwick left	17:41
	← darnir left	17:42
	→ TheFuzzball joined	17:42
	gitinfo set mode: +v	17:43
diimdeep_	hello, I have repo with two commits on master, i want leave recent but delete all that comes with previous, how do that ?	17:43
	← suy left	17:45
	← AccessGranted left	17:46
	← fooey left	17:46
	← zymaster left	17:47
prjkt	hi. why would the "commit" button be greyed out in tortoisegit?	17:49
	what causes this?	17:49
	it shows changed files to files being tracked....	17:49
	← Edgan_ left	17:49
	→ Edgan joined	17:49
	← ozette left	17:49
	← blz left	17:50
	→ tenmilestereo joined	17:50
	gitinfo set mode: +v	17:50
	← axl_ left	17:50
	→ jab416171 joined	17:50
	← mishok13 left	17:50
	→ dytrivedi joined	17:51
	→ twisted` joined	17:51
	→ beawesomeinstead joined	17:51
	→ bmcorser_ joined	17:51
	← spuz left	17:51
milki	prjkt: !gui	17:51
gitinfo	prjkt: Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.	17:51
	→ jjardon joined	17:51
	← johnflux left	17:54
	→ johnflux joined	17:54
	← johnflux left	17:54
	→ johnflux joined	17:54
Synthead	is there a way to view a whole file from a certain commit?	17:55
	preferably with vim?	17:55
	← tPl0ch left	17:56
	→ maxmanders joined	17:56
	← mikef left	17:56
milki	Synthead: git show \| vim -	17:57
	might need a -R or somethign	17:57
vlt	Synthead: `git show <commit>:<file> \| vim -`	17:58
	← etcetera left	17:58
	→ jjido joined	17:59
	→ zakora joined	17:59
	← zakora left	17:59
	→ zakora joined	18:00
	← zakora left	18:00
	→ etcetera joined	18:00
Synthead	milki, vlt: that's awesome. thanks!	18:00
	→ zakora joined	18:00
	← zakora left	18:00
	← brocken_ left	18:00
	→ Noel joined	18:01
	→ BSaboia__ joined	18:01
	Noel → Guest40640	18:01
	← Guest40640 left	18:02
	→ nu11space joined	18:02
lov	How would I re-checkout all of the files in a directory? I've got files changed in multiple directories, but I want to reset one of them to a 'clean' state.	18:02
	← swiety_ left	18:02
	gitinfo set mode: +v	18:02
	← maxmanders left	18:03
nu11space	q: I am getting bad merges (no conflicts) in one file with very many similar c entry points. the merge jumbles the new code to wrong functions. anyone know why this is? is there a way to use more context in git internally so it doesn't lost track of which function it's merging to?	18:03
	← Sigma left	18:03
	→ dc5ala joined	18:04
	← BSaboia left	18:04
nu11space	lov, you can checkout individual files, probably directories too, using checkout	18:04
	from any previous revision	18:04
	→ hiroyuki joined	18:05
lov	I know I can check out files. I want to know about doing entire directories, which appears to be a no-op.	18:05
	← m0ltar left	18:05
	← hiroyuki left	18:05
	→ Tommy[D] joined	18:05
	→ zakora joined	18:05
	← Tommy[D]_ left	18:06
	← Error404NotFound left	18:06
	→ anonymuse joined	18:06
lov	ah, whoops, typo on my end ._.	18:06
	→ Err404NotFound joined	18:06
	lov tugs at his collar	18:06
lov	thanks anyway	18:06
	→ Katafalkas joined	18:06
	bremner` → bremner	18:07
	← rednaks left	18:08
	← ch3sh1r left	18:08
prjkt	does Git know about the zip format? (meaning, for example, if you zip a folder and it tracks the zip as well, then it doesn't include it as binary since it has all the contents already)	18:09
	→ Goles joined	18:09
	→ rednaks joined	18:10
prjkt	?	18:10
lov	git knows about files.	18:10
	it doesn't know or care what those files actually ARE.	18:10
prjkt	ok	18:10
lov	(at least this is true enough for our purposes)	18:11
prjkt	but for example if there are three copies of the same file, 1.txt, 2.txt, 3.txt does it recognize that they're the same?	18:11
lov	yes.	18:11
prjkt	ok	18:11
lov	to be more accurate, git knows about content which has certain SHA hashes.	18:11
prjkt	are there any conditions that mean you can't commit a file?	18:11
	low, that happens on a file basis or a block basis?	18:11
dr_lepper	death, for example	18:11
prjkt	death has not occurred here	18:12
grawity	prjkt: git hashes the entire file	18:12
lov	yeah if you're dead you probably can't commit.	18:12
prjkt	what other reasons would git have for not wanting to commit?	18:12
lov	known issue.	18:12
	I don't know, why don't you tell me what problem your actually having?	18:12
prjkt	"Fixed upstream"	18:12
lov	?	18:12
	← x82_nicole left	18:12
prjkt	The reason why, low, is that then you wouldn't answer my qeustion	18:12
grawity	more precisely, sha1("blob " + length + "\0" + file_contents)	18:12
	← tfnico___ left	18:12
	← jmcantrell left	18:12
prjkt	due to the problem being with a gui. so instead sneakily i am just asking under what conditions git wouldn't commit	18:12
grawity	It will commit any file	18:13
	→ breakingthings joined	18:13
lov	prjkt: I can't enumerate all of the failure conditions of commit because I don't actually know them.	18:13
prjkt	so there aren't any common ones	18:13
lov	also, it's lov with a v not w	18:13
prjkt	like a file is already open or something	18:13
	oh sorry	18:13
lov	that shouldn't matter.	18:13
prjkt	I have a weird font	18:13
	→ jxpx777 joined	18:13
grawity	prjkt: That depends on the OS.	18:13
prjkt	I won't even try to address you, lol	18:13
grawity	prjkt: some operating systems lock files by default, like Windows often does	18:13
prjkt	but the question is, if a file is locked, does that mean that git can't commit it?	18:14
	(yes this is windows btw)	18:14
grawity	there is nothing git- or Git-specific about this; if a file is locked in such a way that other programs cannot access it, then git cannot either	18:14
	← twisted` left	18:15
	← tevio left	18:15
lov	^^	18:15
	→ x82_nicole joined	18:15
	← johnlcox left	18:15
lov	if git cannot READ a file, it shouldn't be able to commit it.	18:15
	→ AnimeCYC_ joined	18:15
lov	similarly, if it cannot write to a file then checkouts will fail, etc.	18:15
	← digitalfiz left	18:15
lov	You should figure out who has a read/write lock on the file.	18:15
	← kit left	18:16
milki	arent locks just suggestions?	18:16
	→ alunduil joined	18:16
lov	to thieves and politicians, yes.	18:16
grawity	milki: compare advisory and mandatory locking	18:16
	← dmsuperman left	18:16
prjkt	I figured out the problem apparently the gui wouldnt let me commit without a message. sorry!	18:16
	← pl0sh left	18:16
	→ nielsbusch joined	18:16
grawity	milki: Linux flock() and fcntl-locking are advisory, but it's possible to enable mandatory locking	18:16
lov	facepalm	18:17
milki	how do you do mandatory locking?	18:17
grawity	milki: in contrast, Windows does the latter by default	18:17
	← normanrichards left	18:17
milki	oo	18:17
	EugeneKay introduces lov to /me	18:17
	← GameGamer43 left	18:17
grawity	milki: and applications must explicitly set "share flags" to avoid a file from being locked – if I remember correctly	18:17
	← jjardon left	18:17
	lov sighs	18:17
milki	so complicated -.-	18:17
grawity	milki: for Linux check Documentation/filesystems/mandatory-locking.txt	18:17
	← jab416171 left	18:17
	← xerxas left	18:17
	← AnimeCYC left	18:18
	← dc5ala left	18:18
milki	171 lines	18:18
	hmm	18:18
	← bmcorser_ left	18:18
	→ mikef joined	18:18
	← mtlatif left	18:18
	→ jceb joined	18:18
milki	ah, kernel level	18:18
	← arnihr__ left	18:19
	← MetaCosm left	18:19
	← zakora left	18:19
	← nicolash left	18:19
	→ Cromulent joined	18:19
grawity	milki: ah, looks like there's one more thing it doesn't mention: the filesystem has to be mounted (or remounted) with -o mand	18:19
	← HeN left	18:20
milki	whoa byte level granularity	18:20
	→ dvide joined	18:20
	← war2mac left	18:20
	← sysadminAli left	18:20
	← jjido left	18:21
milki	cool, this is nice to know	18:21
grawity	also note how the file says "avoid mandatory locking" at the very top. :P	18:22
	← j416 left	18:22
	→ j416 joined	18:22
	← trungqtran left	18:23
	→ eletuchy joined	18:23
	← jjore left	18:23
	→ Sigma joined	18:23
	→ BadProphet joined	18:24
	← kryl99 left	18:24
	→ FilipeMaia joined	18:25
	← mklappstuhl left	18:25
milki	aha	18:25
	→ sysadminAli joined	18:26
	→ leeN joined	18:26
	→ infynyxx joined	18:26
	← jason237 left	18:26
	← cyberbob left	18:26
	→ trungqtran joined	18:27
	← eletuchy left	18:27
	← dytrivedi left	18:28
	→ sluther joined	18:28
	→ pitzips joined	18:28
	← breck7 left	18:28
	→ scwizard joined	18:29
scwizard	I get error: unknown option `orphan'	18:29
	how do I create an orphaned branch?	18:29
	← beawesomeinstead left	18:29
	← chrismar035 left	18:29
	→ primitive joined	18:30
cbreak	scwizard: man git-checkout	18:30
gitinfo	scwizard: the git-checkout manpage is available at http://jk.gs/git-checkout.html	18:30
cbreak	use --orphan	18:30
	← NimeshNeema left	18:30
	← momomomomo left	18:30
cbreak	scwizard: unless you have a really prehistoric version of git, that'll work	18:31
	→ chrismar035 joined	18:31
cbreak	IF you really have a prehistoric version and can't bring yourself to upgrade, ask again :)	18:31
	← dhruvasagar left	18:32
	→ fooey joined	18:32
	← ccolorado left	18:33
	→ maxmanders joined	18:34
	← averiso left	18:35
	→ dreiss joined	18:35
	→ szx joined	18:36
	← maxmanders left	18:37
	→ knightshade joined	18:37
	← jkremser left	18:37
	→ jjido joined	18:38
	→ jjore joined	18:39
	→ ErKa joined	18:39
	← jjido left	18:40
	→ rabellamy joined	18:40
	→ jjido joined	18:40
	→ m0ltar joined	18:40
	→ ThePawnBreak joined	18:42
	→ smirn0v joined	18:42
	→ GameGamer43 joined	18:42
	← infynyxx left	18:44
	← khushildep left	18:45
	← bartzy left	18:46
	→ maxmanders joined	18:46
	→ maxmande_ joined	18:47
	→ erichynds joined	18:48
	← mtk left	18:48
	→ mtk joined	18:48
	← jjido left	18:49
	← maxmanders left	18:50
	← maxmande_ left	18:51
	→ arnihr__ joined	18:51
	← rikko left	18:52
	← WinstonSmith left	18:52
	→ WinstonSmith joined	18:53
	← marcdel left	18:53
	← rgngl left	18:53
	→ momomomomo joined	18:54
	→ kevwil joined	18:56
	← aartist left	18:56
	→ jjardon joined	18:57
	→ dhruvasagar joined	18:59
Adys	i was cleaning up my tree and ran git filter-branch --force --index-filter 'git rm --cached --ignore-unmatch anthill/static/css/global.css' --prune-empty --tag-name-filter cat -- --all	18:59
	i forgot i wanted to keep the file in a couple of commits from a while back	18:59
	→ nathancahill joined	18:59
Adys	it's not massively important but is there a way to revert it?	18:59
	← nathancahill left	18:59
	→ netiul joined	19:00
	→ trungqtr_ joined	19:00
	← trungqtran left	19:01
	trungqtr_ → trungqtran	19:01
	← Esa_ left	19:01
	← Yougi left	19:01
cbreak	Adys: there should be an original namespace	19:04
	search in .git/refs	19:04
	Adys: also, check out the reflogs	19:04
Adys	cbreak: i see one; what do i do with it?	19:05
	← paddymahoney left	19:05
cbreak	those are the original refs	19:05
	← dhruvasagar left	19:05
cbreak	you can reset to them	19:05
Adys	how?	19:05
	cbreak: also, i ran the command multiple times (on different files); will original just restore the last one or the one from the first time i ran it?	19:06
	← sheerun left	19:06
	→ chuckhar_ joined	19:06
cbreak	the latest afaik	19:06
	I never used it	19:06
	as for how to use it, look which sha the ref points at	19:07
	check the branch out, and git reset --hard to that sha	19:07
	→ infynyxx joined	19:07
cbreak	you can also get the sha out of the reflog	19:07
	that should contain older versions too	19:07
	Adys: man git-reflog for details	19:07
gitinfo	Adys: the git-reflog manpage is available at http://jk.gs/git-reflog.html	19:07
	← Watabou left	19:07
cbreak	for example git reset --hard HEAD@{1.hour.ago}	19:08
Adys	cbreak: .git/refs/original/refs/heads/master points to a sha	19:08
	i guess thats the one i want?	19:08
cbreak	yes.	19:08
	← chuckharmston left	19:09
cbreak	it's one that's older than the one you have now	19:09
	but remember: you rewrote --all and not just master	19:09
Adys	yeah thats not an issue with that file	19:10
	← danishkhan left	19:10
cbreak	oh, don't say that until you look at the history graph	19:10
Adys	cbreak: auxiliary question, how do I apply filter-branch only on commits more recent than say 3 days?	19:10
	→ mitz joined	19:10
cbreak	you sure filter-branch is appropriate for that?	19:10
	maybe consider git rebase -i	19:11
	(if no merges are involved)	19:11
Adys	cbreak: i basically need to get completely rid of css/global.css from the history, but only the recent ones	19:11
	it's a compiled file	19:11
	→ sheerun joined	19:11
	← elixir left	19:12
	← trungqtran left	19:12
	→ mloskot joined	19:12
cbreak	maybe you can put a check into the index filter	19:13
	→ eletuchy joined	19:13
	← tureba left	19:13
cbreak	or just pass a different range of refs to rewrite	19:13
	git filter-branch ... -- a..b	19:13
	a..b means ^a b	19:14
	← lacrymology left	19:14
cbreak	so "reachable from b but not a"	19:14
	you can make quite complex expressions if needed.	19:14
	but I've not tried any of that... :)	19:14
	← marhaban left	19:14
	← nielsbusch left	19:15
	← etcetera left	19:16
	← x82_nicole left	19:16
	→ elixir joined	19:17
	→ nathancahill joined	19:18
	← Cromulent left	19:19
	→ war2mac joined	19:19
	→ trungqtran joined	19:19
	→ paddymahoney joined	19:20
	→ buibex joined	19:21
	← war2mac left	19:22
	→ tplaner joined	19:22
	→ sq-one joined	19:22
	← yacks left	19:23
	← fooey left	19:23
	← mitz left	19:24
	→ mikehaas763 joined	19:24
	← cebor left	19:25
	← rabellamy left	19:26
	← jMCg left	19:26
	→ jMCg joined	19:27
	→ bartzy joined	19:27
	← dreiss left	19:27
	← tianon left	19:29
	← candlejackson left	19:30
	← vedant left	19:30
	→ tureba joined	19:30
	→ normanrichards joined	19:30
	→ candlejackson joined	19:30
	← rakl left	19:30
	← diimdeep_ left	19:31
	→ fooey joined	19:31
	→ goshawk joined	19:31
	→ johnflux_ joined	19:32
	← johnflux_ left	19:32
	→ johnflux_ joined	19:32
	← johnflux left	19:32
	← monod left	19:32
	→ mmlac joined	19:33
	→ revyd joined	19:33
	→ m0viefreak joined	19:34
	← m0viefreak left	19:34
	→ m0viefreak joined	19:34
	→ megha joined	19:35
szx	why would git say everything is up-to-date while it's clearly not?	19:35
	→ etcetera joined	19:35
	← baba left	19:35
	← etcetera left	19:37
	→ x82_nicole joined	19:37
rjc	szx: can you elaborate, please?	19:37
	← rideh left	19:39
	→ dreiss joined	19:39
szx	something weird is going on with my repo, I pushed to two remotes and one of them accepted the push (bitbucket) and another said "Everything is up-to-date" but it can't be up-to-date! I pushed that commit for the first time	19:40
	→ danishkhan joined	19:40
szx	and the strageest thing is that afterwards git log rendered bitbucket as not updated and github as up-to-date	19:40
	i.e. vice versa	19:41
	oh btw the second one is github	19:41
	also I can't see any updates on github	19:42
rjc	are you sure you've got both remotes set up corectly?	19:42
	i.e. not the other way around?	19:42
szx	sure	19:42
	I double checked	19:42
	now each time I push to github it says that everything is up-to-date and both remotes are at HEAD	19:43
	→ jdolitsky joined	19:43
szx	but github is still 1 commit behind	19:43
charon	!repro of 'git remote -v', 'git config push.default', how you push and how you check that it's not up to date	19:43
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session. This will help immensely with troubleshooting.	19:43
	← juvenal left	19:44
	→ maxmanders joined	19:44
szx	I use git log and gitk	19:45
	→ pr0ton joined	19:45
szx	not precisely git log, I have an alias which expands to : log --graph --decorate --pretty=oneline --abbrev-commit --all	19:45
	I tried fetching from both	19:46
	and switching branches around	19:46
	nothing works	19:46
	← hoschi left	19:47
	→ szx` joined	19:47
	→ Octalot joined	19:47
	← danishkhan left	19:48
cbreak	szx: irrelevant	19:48
	git remote -v	19:49
	← jaequery left	19:49
Nugget	I wish git log had a --fix-my-typos option.	19:50
	→ marcdel joined	19:50
	← prjkt left	19:50
	→ Watabou joined	19:50
	→ creese joined	19:50
	← szx left	19:51
szx`	cbreak: git remote -v looks fine	19:51
cbreak	pastebin your git push output and git ls-remote output	19:52
creese	Is it reasonable for 'git pull' to result in a merge commit?	19:52
cbreak	creese: of course	19:52
	creese: pull is fetch + merge	19:52
	the merge part makes merge commits	19:52
	→ nielsbusch joined	19:52
creese	I know that, I've just never seen it create the extra commit before	19:52
rjc	szx`: it really is usefull for us to have a look so please use gist or other pastebin	19:52
	← sonda left	19:52
creese	usually, it just grab the ones I'm missing	19:53
cbreak	creese: fast forward merging is only possible if there's no divergent history	19:53
rjc	szx`: I understand that to you it all might look OK but then again the more eyes look at it ... ;^)	19:53
creese	ok, thanks, I guess I've only seen ff merging when pulling up to now	19:54
	← oojacoboo left	19:55
	→ rakl joined	19:56
	← momomomomo left	19:57
szx`	rjc: http://pastebin.com/uRzcKgb9	19:57
	→ corni joined	19:58
	← corni left	19:58
	→ corni joined	19:58
	→ cagedwisdom joined	19:58
	← robbyoconnor left	19:58
szx`	do you see anything wrong?	19:58
	← creese left	19:59
	← maxmanders left	19:59
cbreak	lots of origin. how did you push?	19:59
szx`	git push origin master	20:00
	→ rabellamy joined	20:00
rjc	szx`: another thing apart from origin is the fact that it is duplicated in both github and bitbucket remotes - any reason for that?	20:00
	→ aberrant joined	20:01
szx`	it's a trick to push to both remotes with one command	20:01
	→ Parhs8 joined	20:01
	→ Gaddel joined	20:01
szx`	learned it somewhere on stackoverflow	20:01
	← pr0ton left	20:01
	→ Shusshu joined	20:01
	← buibex left	20:02
szx`	it worked perfectly since the beginning	20:02
Gaddel	hi, i have a development branch and a master branch. i want to merge things from master into the development branch	20:02
	← AdrienBrault left	20:02
Gaddel	i am a bit confused with when to use pull/checkout/rebase/merge	20:02
	← Shusshu left	20:02
	→ jaequery joined	20:02
cbreak	szx`: you have to push to the remote you want to push	20:02
	and not the broken origin	20:03
	→ robbyoconnor joined	20:03
szx`	believe me I tried	20:03
	← nathancahill left	20:03
szx`	maybe I should just re-add all the remotes	20:03
	← Parhs left	20:04
szx`	?	20:04
	← dreiss left	20:04
cbreak	what for?	20:04
szx`	don't know	20:04
	← normanrichards left	20:04
rjc	szx`: what I'd suggest is fix origin to point only to master repo	20:04
	← AnimeCYC_ left	20:04
	→ jason237 joined	20:04
rjc	then you can set up a remote with two remote git repositories	20:04
cbreak	did git ls-remote show the expected values?	20:05
	→ tianon joined	20:05
	→ Parhs joined	20:05
	→ AnimeCYC joined	20:05
	→ _iron joined	20:05
	← xiangfu left	20:05
	← Parhs8 left	20:06
	→ normanrichards joined	20:06
	→ thinkdevcode joined	20:07
	← AnimeCYC left	20:07
rjc	szx`: btw, that trick with two remote repositories under one git remote only works for push	20:07
	← j416 left	20:07
	← robbyoconnor left	20:07
rjc	szx`: so simply don't fetch from that remote	20:07
	→ j416 joined	20:07
	→ adurity joined	20:07
	→ AnimeCYC joined	20:07
	→ momomomomo joined	20:08
	→ pl0sh joined	20:08
	→ jab416171 joined	20:08
	← chuckhar_ left	20:09
	→ Parhs8 joined	20:09
	→ milian joined	20:09
szx`	well one remote looks ok in ls-remote, but the other looks exactly the same even though I remotved some branches	20:09
	← revyd left	20:10
	→ bmcorser_ joined	20:10
szx`	not even removed, I've never pushed them to the second remote, so it's clearly not ok	20:10
	→ tfnico___ joined	20:10
	→ jmcantrell joined	20:10
	→ mojzu joined	20:10
szx`	could it be a bug in git?	20:10
	← Noldorin left	20:11
	← fooey left	20:11
rjc	szx`: if you've 'git push origin' it would have pushed to both repos at the same time since you have them both defined under origin	20:12
	← Parhs left	20:12
	→ johnlcox joined	20:12
	→ etcetera joined	20:12
	→ Horstinator joined	20:13
	→ twisted` joined	20:13
	→ robbyoconnor joined	20:13
szx`	but I didn't do push origin, I always do push origin master	20:13
	→ Parhs joined	20:13
	← nu11space left	20:13
	← infynyxx left	20:13
szx`	also the phantom branches are not show on github	20:13
	(the broken remote)	20:13
	← paddymahoney left	20:13
Gaddel	hi, i'm trying to merge/rebase from master into my development branch. "git merge master" only says "already up to date", but "diff master..branch" shows there are many differences	20:13
	→ digitalfiz joined	20:14
	→ infynyxx joined	20:14
	← szx` left	20:14
	← sysadminAli left	20:15
	→ danishkhan joined	20:15
	← momomomomo left	20:15
	← Parhs8 left	20:16
	→ mtlatif joined	20:16
	→ Parhs8 joined	20:17
	→ HeN joined	20:17
	← robbyoconnor left	20:17
	→ MetaCosm joined	20:17
	← Parhs left	20:18
	→ DoniRS joined	20:18
	→ Parhs joined	20:18
	→ dreiss joined	20:18
osse	Gaddel: that means your branch already contains all of master	20:20
	Gaddel: the tip of the master branch is found in the history of your branch	20:20
	← danishkhan left	20:20
	→ pr0ton joined	20:20
Gaddel	osse: i must've messed something up then, because master definitely has things that are ahead of my branch	20:20
osse	then I suppse the merge went sour	20:21
rjc	Gaddel: just to make sure, you are on your devel branch when you merge, right?	20:21
Gaddel	well, i have not yet merged successfully. every time i've tried it simply says "already up to date"	20:21
	rjc: correct	20:21
	← dukedave left	20:22
	← Parhs8 left	20:22
	→ Parhs8 joined	20:22
	→ MEch0z_ joined	20:22
	← angelsl left	20:22
rjc	well, is it possible that you simply have commits which are in devel but not in master?	20:23
	← Parhs left	20:23
Gaddel	rjc: all of my commits are in the development branch i think.	20:23
	→ Parhs joined	20:23
	← Parhs left	20:23
	→ Parhs joined	20:23
	→ Mech0z joined	20:23
Mech0z	Is it possible to add a file to gitignore with powershell that is unstaged	20:24
osse	Gaddel: git log --oneline --decorate --graph master branch	20:24
Mech0z	I have added the file manually, to the file, but nothing seems to matter, its still unstaged (will be added if I do git add .)	20:24
rjc	Gaddel: that's what I mean, if you're mergin master into devel and devel is ahead, that is expected output	20:24
Gaddel	rjc: okay...how do i make devel not ahead, then?	20:25
	rjc: i am very new to git	20:25
	← jceb left	20:25
rjc	Gaddel: work on devel, checkout master and merge devel into master	20:26
	→ dytrivedi joined	20:26
	← Horstinator left	20:26
	→ Schrostfutz joined	20:26
Schrostfutz	hi, how can i display the changes made by a commit	20:26
	→ nathancahill joined	20:26
	→ mitz joined	20:27
gtmanfred	git show <commit rev> ?	20:27
	→ xerxas joined	20:27
osse	Gaddel: the git log command I have you should help clear things up a little bit	20:27
	→ justache joined	20:27
	→ beawesomeinstead joined	20:27
	← Parhs8 left	20:27
Gaddel	osse: i ran it but unfortunately i am having difficulty understanding it	20:27
osse	Gaddel: can you put the output up somewhere and show us?	20:28
	→ Parhs8 joined	20:28
osse	Gaddel: not everything but say 20 lines or so	20:28
	← Parhs left	20:28
Gaddel	osse: http://bpaste.net/show/Ng5SFvJ79SVLbpNBEWFs/	20:29
cbreak	Mech0z: use a text editor.	20:29
	like vim.	20:29
osse	Gaddel: 'socks-proxy' is "your branch" right?	20:29
	← ojacobson left	20:30
Gaddel	osse: correct	20:30
osse	Gaddel: the other one you were referring to?	20:30
cbreak	Gaddel: looks like you have merged master	20:30
osse	ok	20:30
	← rahulprasad left	20:30
cbreak	so no problem.	20:30
Gaddel	cbreak: it does not appear so	20:30
cbreak	it does.	20:30
	look at line 2	20:30
osse	Gaddel: the second-to-last on the socks-proxy branch is the master branch.	20:30
	→ vedant joined	20:31
	→ rgngl joined	20:31
	← bartek left	20:31
osse	I fully understand it can be confusing, but I don't think anything is wrong :)	20:31
Gaddel	well the issue is	20:31
	i am not seeing various changes made in master	20:31
	→ NimeshNeema joined	20:31
Gaddel	when i for example read the files in the socks-proxy branch in vim	20:31
cbreak	Gaddel: then you undid the changes in your only commit in the socks-proxy branch	20:32
Gaddel	cbreak: i must have then	20:32
	cbreak: how do i fix that?	20:32
cbreak	you can see the changes you did in the last commit with git show HEAD	20:32
	and the changes you did not yet commit with git diff HEAD	20:32
osse	Gaddel: does 'git diff master socks-proxy' make sense? OR does it contain unintended changes?	20:33
	← mitz left	20:33
cbreak	you can get those changes with git checkout -p HEAD~1 -- filename for example	20:33
Gaddel	okay, i think i see now that	20:33
	when i view the repo on github it looks alright, but my local copy does not seem to be up to date	20:33
	→ jjido joined	20:34
	← Watabou left	20:34
	→ Watabou joined	20:34
	← keller left	20:34
	→ szx joined	20:34
Gaddel	okay i see	20:35
	my local copy of master now has all the changes of socks-proxy, which i didn't want	20:35
osse	Gaddel: I see orogin/master is behind master, so I think you need to 'git fetch'	20:35
Gaddel	i think i might just start from scratch or something	20:35
szx	cbreak: could it be something wrong with ssh? because ls-remote shows different things when I clone with ssh vs git://	20:35
	← ijuma left	20:35
	_W → _W_	20:35
Gaddel	osse: ran git fetch; othing appears to have happened	20:36
szx	cbreak: and I've removed and re-added all remotes, but that didn't help either, now hen I push to github git pushes to bitbucket lol	20:36
cbreak	szx: different URLs? different remotes?	20:36
Gaddel	nothing*	20:36
	← kandinsk1 left	20:36
cbreak	szx: stupid settings in .ssh's config?	20:37
	← s00pcan left	20:37
	→ chuckharmston joined	20:37
osse	Gaddel: then you're up-to-date with respect to github	20:37
	→ zakora joined	20:37
szx	what could potentially be wrong in .ssh/config?	20:37
osse	Gaddel: if the origin/master thing hasn't moved around in the output of git log	20:37
szx	I have a couple of Hostname/IdentityFile pairs	20:38
	← MEch0z_ left	20:38
szx	(they point to different ssh keys)	20:38
Gaddel	osse: it would appear i am not. my copy of master is different from what's in my github repo on the master branch, or the master origin	20:38
	"Your branch is ahead of 'origin/master' by 2 commits."	20:38
	→ Dooom joined	20:38
cbreak	szx: ssh configuration can change host names	20:39
	→ paddymahoney joined	20:39
rjc	Gaddel: which means you had not pushed your local changes to that remote	20:39
	→ etcetera_ joined	20:39
Gaddel	rjc: it looks my local copy of master is the same as my local of socks-proxy though; i don't want master to have any of socks-proxy's changes though, only the other way around	20:40
szx	cbreak: this is what I have: http://pastebin.com/kTqR6P1E	20:40
	→ calamari joined	20:40
szx	looks ok?	20:40
Gaddel	rjc: basically i want to get rid of my copy of master and replace it with the remote master, how do i do that?	20:41
cbreak	szx: nothing that looks off to me	20:41
	Gaddel: first fetch it	20:41
	← etcetera_ left	20:41
cbreak	Gaddel: then reset --hard to it	20:41
	make sure you are on master before the reset	20:41
Gaddel	cbreak to fetch it do i just do "git fetch"?	20:42
	or git fetch master?	20:42
	← etcetera left	20:42
calamari	does git provide any mechanism for indicating "I plan to work on these files"? I don't want to use checkout, because that would suggest enforcement, this would be more like a hint	20:42
milki	calamari: no	20:42
	← erichynds left	20:42
calamari	milki: thank you	20:42
	→ rahulprasad joined	20:42
	→ fooey joined	20:43
cbreak	Gaddel: git fetch master will not do anything sane	20:43
milki	calamari: at the basic level, a distributed system has no such concept of locking because that defeats the purpose	20:43
cbreak	Gaddel: you will probably want git fetch origin	20:43
	← nathancahill left	20:43
	← w0rmdrink left	20:43
cbreak	calamari: checkout is on a branch level	20:43
calamari	milki: agreed. this wouldn't be locking, it'd be a hint	20:44
milki	calamari: same thing	20:44
cbreak	calamari: and it's local only	20:44
	no one knows if someone else checks out something	20:44
	← BillCriswell left	20:44
Gaddel	cbreak: okay. "HEAD is now at 873e086 socks proxy" looks like my master still isn't what it should be	20:44
cbreak	checkout does not suggest enforcement in any way	20:44
	→ Parhs joined	20:44
	← Parhs left	20:44
	→ Parhs joined	20:44
	→ nathancahill joined	20:44
cbreak	Gaddel: as I said: first check out master, and fetch origin	20:45
milki	calamari: the real question then is, why do you want this idea?	20:45
cbreak	then reset --hard to origin/master	20:45
Gaddel	cbreak: okay thanks. i had the reset syntax off	20:46
osse	calamari: checkout doesn't suggest enforcement. not in git.	20:46
cbreak	Gaddel: it's all totally intuitive if you know how it works already... :)	20:46
calamari	milki: we are currently using visual source safe, which of course blows chunks. this idea of locking is ingrained. I was hoping there could be some way of indicating a person is working on a file, that way it initiates conversation.	20:46
	← szx left	20:46
_ikke_	calamari: Not within git	20:46
	calamari: git is distributes	20:46
cbreak	calamari: there is	20:46
	it's called email	20:46
osse	lol	20:47
milki	calamari: sounds like you are looking for a document management system rather than source control	20:47
cbreak	or shouting-across-the-room	20:47
	I prefer the latter	20:47
	→ FernandoBasso joined	20:47
calamari	milki: in some respects that is very true	20:47
osse	calamari: as long as you're not working on the very same lines within the same file there is usually no problem	20:47
	← jaequery left	20:47
	→ danishkhan joined	20:47
	→ momomomomo joined	20:47
osse	calamari: git is pretty good with this "merge" malarkey	20:47
calamari	agreed	20:47
	← Parhs8 left	20:48
	← rath left	20:48
	← jjido left	20:48
osse	in my experience that kind of checkout is more of an inconvenience than a convenience	20:48
	but to each his own, of course	20:48
	← zakora left	20:49
	→ buibex joined	20:49
calamari	there is also a really cool gui, "git extensions" which goes a long way	20:49
	← Parhs left	20:49
calamari	I prefer command line but most of my coworkers hate it	20:49
cbreak	someone should make a real git GUI	20:49
	using OpenGL. and railguns.	20:50
calamari	yes, actually that is part of the problem. since checkout is enforced then people just make changes on their local coppies and once the file is available they don't merge, because thats too much trouble, they just check in and wipe out the other persons changes	20:50
cbreak	and LASERs!	20:50
	→ Parhs joined	20:50
	← Parhs left	20:50
	→ Parhs joined	20:50
rjc	calamari: hate is quite a strong word, any reason why they dislike it?	20:50
	→ oinkon joined	20:50
cbreak	calamari: are those idiots allowed to do that?	20:50
	→ Esa_ joined	20:50
oinkon	I somehow ended up with a branch named origin/master and i get this ambiguous branch error. how do i delete this branch?	20:50
cbreak	delete other people's work?	20:50
	→ taylorlapeyre joined	20:50
	← natechan left	20:50
cbreak	I'd get fired if I'd do that where I work.	20:50
oinkon	I mean a local branch named origin/master	20:51
_ikke_	oinkon: git branch -d refs/heads/origin/master	20:51
oinkon	thanks!	20:51
calamari	cbreak: it's kinda free for all, it's unlike any place I've ever worked before	20:51
	→ frenchlaf joined	20:51
calamari	cbreak: with git, I know it's going to enforce the merge .. they'd have to go out of their way to overwrite my changes	20:52
_ikke_	git branch -f origin/master master :D	20:52
	git push -f origin master	20:52
calamari	shh!	20:52
cbreak	git push -fu --all	20:52
_ikke_	that first line doesn't make sense	20:52
	← ThePawnBreak left	20:53
	→ lokus joined	20:53
_ikke_	:D	20:53
lokus	hi folks.. i did a 'git bisect good' by mistake when it should have been 'git bisect bad' .. how can i easily correct it without starting again?	20:53
	← fobban left	20:53
cbreak	that's probably the only git push syntax that sounds like it is.	20:53
Gaddel	okay, so i have master and socks-proxy reset. now, i have a separate directory containing development files. there's no git metadata in this folder. how do i merge that folder with my socks-proxy branch?	20:54
	← taylorlapeyre left	20:54
charon	lokus: git bisect log, edit that	20:54
cbreak	lokus: git bisect bad blahblah	20:54
	← rbmk left	20:54
cbreak	it takes a re as argument	20:54
	rev	20:54
	← Parhs left	20:54
cbreak	Gaddel: pointless. if it's not in git, then git doesn't care	20:54
	Gaddel: just make a new branch for it, copy the files over and commit	20:54
	then the stuff is in git, and git can merge	20:55
lokus	cbreak: rev? not commit? i'm not sure what the last commit was that bisect had used. i'll try log as charon suggested	20:55
Gaddel	cbreak: just so i can understand the semantics for git checkout	20:55
	BadDesign → OptimizationNazi	20:55
cbreak	lokus: a commit will do	20:55
Gaddel	cbreak: if i checkout a branch, it's as if it's a totally different directory or what?	20:55
calamari	thanks everyone for your ideas and feedback	20:55
cbreak	Gaddel: no.	20:55
	← Dooom left	20:55
	← scwizard left	20:55
cbreak	directories are not involved	20:55
	→ whowantstolivef1 joined	20:55
Gaddel	cbreak: okay, i need to read up on how the branches correspond to directories and such	20:55
cbreak	if you switch to a different branch, then all files get changed to be like on that branch	20:55
	Gaddel: branches do not have any relation with directories	20:55
osse	calamari: heh, your (plural) way of working can only work properly if no more than one person edits a file at a time, so you have no need for locking anyway :P	20:56
	→ Parhs joined	20:56
	← Parhs left	20:56
	→ Parhs joined	20:56
Gaddel	cbreak: okay, so git checkout like...swaps out the files after you run it, or what?	20:56
cbreak	no, during	20:56
Gaddel	okay, that's what i meant	20:56
	→ Parhs8 joined	20:56
Gaddel	so if i'm currently in the socks-proxy branch and cp the files over here	20:56
	and commit them	20:56
	← Schrostfutz left	20:56
Gaddel	they'll stay in only that branch	20:56
cbreak	commits are bound to a branch usually, yes	20:57
Gaddel	ok	20:57
	→ s00pcan joined	20:57
lokus	charon: i found the SHA in git bisect log which should have been 'bad'. when i issue git bisect bad <sha> it won't work though.. "some good revs are not ancestor of the bad rev. cannot work properly in this case"	20:57
charon	lokus: no i mean, save the log, edit that single command, 'git bisect reset', then run the log again	20:58
lokus	i see. thanks.	20:58
	andres_f → andres	20:58
rjc	Gaddel: treat it like switching between filesystem snapshots	20:58
	← smirn0v left	20:58
	← buibex left	20:58
	→ ruckerz joined	20:59
osse	calamari: btw, I feel your pain.	20:59
	not right now, cause I'm on holiday. But usually	20:59
	← Parhs left	21:00
calamari	osse: in a perfect world, you'd check out code, change it, and check it back in, and everyone is happy. but of course there are bugs and other people need to modify the code and fix something "quick". that's when the trouble starts	21:00
	or someone leaves on vacation or gets sick and has things checked out	21:01
	← freeed left	21:01
	→ Parhs joined	21:01
osse	calamari: that last thing is why I don't like the checkout-and-lock thing.	21:01
	← dmarkham_ left	21:01
calamari	best I can tell, git avoids all those problems. but there is the increased possibility of two people working on a file because they don't realize someone else is too	21:01
EugeneKay	That's okay, it makes it easier to resolve problems	21:02
osse	calamari: in an as close to perfect world as you're gonna get: you'd check out code, change it, and check it back in, and everyone is happy. but of course there are bugs and other people need to modify the code and fix something "quick". that's when you get rejected when you try to check in, so you resolve the conflict in one of several different ways, and then check in. no trouble starts.	21:02
	← x82_nicole left	21:03
calamari	osse: is that subversion?	21:03
	← Parhs8 left	21:03
osse	calamari: no that's git, except I used some terminology that doesn't fit, such as "check in". By that I mean 'push' in git-speak	21:03
	← mojzu left	21:03
calamari	nice	21:04
	← tirengarfio left	21:04
	→ theother_ joined	21:04
	← tirengarfio_ left	21:04
	← alunduil left	21:04
osse	calamari: what would happens is that the guy with the quicke pushes. you try to push and you get rejected. then you download his changes first. make them fit in with your own (either merge or rebase), then you push.	21:05
calamari	the merge happens on the pull, correct?	21:05
	← s0ber left	21:05
osse	calamari: yes. or you can fetch + merge and do the steps seperately	21:06
	→ crocket joined	21:06
crocket	Can anyone clone git://snowberry.me/LuceneTest.git ?	21:06
	← theotherguy left	21:06
crocket	I just want to make sure my git-daemon works.	21:06
osse	crocket: works	21:06
_ikke_	osse: According to the irc log, you have been seen last 9 days ago :P	21:06
	→ s0ber joined	21:06
_ikke_	http://itvends.com/irc/git.html	21:07
	→ pt joined	21:07
milki	ooo, i moved up to 5!	21:07
osse	_ikke_: by "on holiday" I mean "off work", not "in a foreign country getting a tan" :)	21:07
crocket	osse, thanks	21:08
_ikke_	osse: Oh, that wasn't even a response to that	21:08
	osse: I was just looking at the stats when noticing that	21:08
osse	_ikke_: what was it then? :O	21:08
_ikke_	See the link	21:08
osse	cooool	21:08
_ikke_	Jast has been seen 15 days ago to that log	21:08
	→ pantsman joined	21:08
	← pantsman left	21:08
	→ pantsman joined	21:08
	← tenmilestereo left	21:09
osse	huh? is my nick osse_ ??	21:09
	→ rtjure joined	21:09
osse	see the random quote on osse_. I just said that five minutes ago!	21:09
_ikke_	hmm, odd	21:10
	→ looper joined	21:10
osse	well it's not even, that's for sure	21:10
cbreak	hmm... apparently I talk to myself a lot...	21:11
	maybe people are just speechlessly impressed by my typing speed...	21:11
_ikke_	or you hit enter a lot ;-)	21:11
	→ x82_nicole joined	21:12
osse	the stats are obviously flawed! :P	21:12
_ikke_	osse: You are flawed :P\	21:12
	← quuxman_ left	21:12
osse	indeed I am	21:12
	which is weird	21:13
	because a long time ago I was told by some dude in a robe that I was made in someone's image	21:13
cbreak	osse: you should be careful	21:13
	maybe that guy comes back to harvest the organs for your template	21:14
milki	osse: wizard hat and robe?	21:14
osse	cbreak: he will eventually, anyway.	21:14
	→ Houl joined	21:14
	← Sigma left	21:15
	OptimizationNazi → BadDesign	21:15
	→ natechan joined	21:15
	← s00pcan left	21:15
osse	milki: the dude? no wizard; just the robe. the whatever-it-is whose image I'm allegedly made in sounded like one, if you were to believe the dude in the robe.	21:16
	→ jlg13 joined	21:16
milki	o.O	21:16
kini	git seems to pack refs on `git clone --mirror` - are there other situations when refs may become packed?	21:17
	for example would garbage collection make this happen?	21:17
	→ Error404NotFound joined	21:17
	→ Phrogz_ joined	21:18
kini	I have some code which depends on branches sitting in .git/refs/heads/ which would break if refs were moved somewhere else	21:18
_ikke_	kini: git garbage collect can indeed do that	21:18
osse	kini: packing doesn't move refs in that sense	21:18
charon	kini: so your code is broken. fix it?	21:18
_ikke_	don't rely on anything in the .git dir	21:18
	use git's interface for that	21:19
charon	kini: script around for-each-ref and update-ref instead of doing everything manually	21:19
Phrogz_	Gerrit confuses me. Given that I'm on a local non-tracking branch with all the changes I want to push to the master, but I know that there have been remote changes that I need to first merge, how do I pull changes from a specific remote branch into my current branch, forcing the merge?	21:19
	(And then I'll just git push origin HEAD:refs/for/branch-name like normal)	21:19
	← Err404NotFound left	21:19
	→ oriba joined	21:20
	Phrogz_ → Phrogz	21:20
	→ dukedave joined	21:20
cbreak	Phrogz: fetch + merge = pull	21:22
	← R1KRDO1 left	21:23
	← normanrichards left	21:23
Phrogz	cbreak: So if I already did git fetch origin then I just need to merge with some change on the branch I want that I now know about locally?	21:24
	← TeckniX left	21:24
	→ TeckniX_ joined	21:24
	→ Sigma joined	21:24
cbreak	if you fetched origin, then you have origin's branches in the remote tracking branches	21:24
	you can git merge origin/foo or what ever. see git branch -r	21:24
Phrogz	OK. Will read. Thank you.	21:25
Gaddel	okay, i am trying to (once again) merge from master into develop (socks-proxy)	21:25
osse	woohoo!	21:25
Gaddel	i tried git checkout socks-proxy; git rebase master	21:25
	it says socks-proxy is already up to date	21:26
	→ R1KRDO joined	21:26
	← hellopat left	21:26
osse	then it probably is	21:26
Gaddel	i just want all the recent diffs in master to be used by socks-proxy.	21:26
	→ hellopat joined	21:26
rjc	Gaddel: since you've rebased, they should already be there	21:27
Gaddel	rjc: they aren't. not in the socks-proxy branch.	21:27
	unless i am misunderstanding what a merge does	21:27
	→ Chaser joined	21:28
rjc	merge != rebase	21:28
	← jamesez left	21:28
cbreak	Gaddel: your socks proxy branch was like master	21:28
	because you reset it	21:28
	then you made a new commit	21:28
rjc	Gaddel: btw, if you're collaborating with anyone I'd advise agains rebase	21:29
cbreak	so the only thing that changed was the new commit	21:29
	rebasing wouldn't change anything	21:29
Gaddel	okay. the person in charge of the github project always tells people to rebase to use the newest master, so that's what i was going by	21:29
	i just want to merge my branch with the latest master. what is the best way of doing that?	21:30
cbreak	use git merge	21:30
	but he obviously doesn't want you to merge	21:30
	so why do you want to merge?	21:30
	← Houl left	21:30
	→ Parhs8 joined	21:31
	→ psoo joined	21:31
	← jdolitsky left	21:32
	← whowantstolivef1 left	21:32
	← rtjure left	21:32
	← rabellamy left	21:33
Gaddel	cbreak: i do not know. i just want to update my branch so it has master's recent changes.	21:33
calamari	one thing our team does is push temporary files, database snapshots and such, and then when the next one is released, delete the old version. if this is done in git, will a clone operation take longer and longer each time?	21:34
rjc	Gaddel: ok, that makes more sense now - socks-proxy is local-only branch	21:34
osse	Gaddel: does 'git diff master spcks-proxy' only show your changes?	21:34
	← Parhs left	21:34
Gaddel	osse: yes it does	21:34
cbreak	Gaddel: so, why do you think you don't have the recent changes?	21:34
	Gaddel: does git log not show them?	21:34
osse	Gaddel: then everything is already good.	21:34
	← dirte left	21:34
	← pr0ton left	21:35
Gaddel	my branch is a revision of an older version of master though, so the diff shows, for example, removal of lines that the new master has added	21:35
	→ Parhs joined	21:35
cbreak	no	21:35
	your branch is a revision of the newest master	21:35
Gaddel	but	21:36
	← fooey left	21:36
	← AaronMT left	21:36
	→ pr0ton joined	21:36
cbreak	because you MADE it by modifying the newest master	21:36
	← dreiss left	21:36
Gaddel	no, i made it by copying it recursively from another directory	21:36
	→ AaronMT joined	21:36
cbreak	-> modifying newest master	21:36
	doesn't matter how you modifyed it	21:36
	you removed the lines	21:36
Gaddel	right	21:36
	so how do i	21:37
	← Parhs8 left	21:37
Gaddel	i understand why that step would do so	21:37
	→ dmarkham_ joined	21:37
Gaddel	but how do i then	21:37
	← netiul left	21:37
cbreak	do you know which commit your changes were ripped out of?	21:37
Gaddel	i could find it	21:37
	→ _delirium joined	21:37
cbreak	then git checkout -b tmp thatcommit	21:37
	and commit your changed files on that branch	21:38
osse	Gaddel: sounds to me like the commit(s) that contain(s) your intended changes also contain other changes that you don't want.	21:38
	If so, then that's the source of the problem	21:38
cbreak	osse: because he messed around outside git	21:38
_delirium	hey there guys I have a Git repo but its NOT a working tree - how can I push master, all branches and all tags upto my new GitHub repo?	21:38
osse	yup	21:38
	cbreak: used add, add -p was appropriate :)	21:38
cbreak	_delirium: git push --all and then git push --tags	21:38
	master is a branch like any other	21:39
Gaddel	this is getting so complicated i'm considering just changing the things line by line	21:39
cbreak	Gaddel: it's easy.	21:39
	Gaddel: but you pay the price for messing around outside git	21:39
Gaddel	yeah, i see that now	21:39
cbreak	things get complicated without git helping :)	21:39
Gaddel	"and commit your changed files on that branch"	21:39
cbreak	that's a 5 second job	21:40
Gaddel	do you mean	21:40
	→ w0rmdrink joined	21:40
Gaddel	an older branch of master?	21:40
	an older commit	21:40
	of master	21:40
cbreak	no	21:40
	find the commit your changes are based on	21:40
Gaddel	ok	21:40
cbreak	make a new branch off of that commit	21:40
	commit your stuff there	21:40
alip	master, master, where are the dreams that i've been after?	21:40
	← chrisdev left	21:40
cbreak	(I'd use commit-tree but you can just copy over your files like you did before)	21:40
Gaddel	okay, and then?	21:41
	→ thasmo joined	21:41
PerlJam	alip: promised only lies.	21:41
osse	Gaddel: do you have a copy of the code with your changes in another directory? If so you can checkout master, make a new branch, then copy in stuff and use git add -p to only add the stuff that makes sense to add.	21:41
cbreak	Gaddel: then you're done. you can now rebase or merge	21:41
	← Dougie187 left	21:41
	← vedant left	21:41
	→ s00pcan joined	21:41
Gaddel	osse: i might do that. how does git add -p work?	21:42
_ikke_	Gaddel: It iterates each hunk and gives you the choice to add it or not	21:42
	→ lavaman joined	21:42
Gaddel	ok	21:42
	thanks	21:42
	← oakdog8 left	21:42
	← longbeach left	21:43
osse	Gaddel: yup. I think it would be a good tool in this case, if I understand your predicament correctly.	21:43
	→ kevinfish joined	21:44
cbreak	my solution would be quicker	21:44
	and automatic :)	21:44
	→ jdolitsky joined	21:44
	← theother_ left	21:45
osse	cbreak: I don't see how your way would not make him end up in the same situation	21:45
	→ john_anderson joined	21:45
	← huoxito left	21:45
	← Dr4k3 left	21:46
ruckerz	git push origin <branch> seems kind of strange, i'd rather have git push check which branch i'm working in and use that as the default branch	21:46
	to psuh	21:46
osse	ruckerz: that's configurable	21:46
	← _iron left	21:46
	← Xethron left	21:47
cbreak	osse: by using git's facilities	21:47
	osse: his problem is not that he has unrelated changes	21:47
	← mikef left	21:47
	→ dreiss joined	21:47
cbreak	osse: his problem is that he has an old version, so committing on top of a new would undo all the new changes	21:47
	ruckerz: man git-config	21:47
gitinfo	ruckerz: the git-config manpage is available at http://jk.gs/git-config.html	21:47
cbreak	ruckerz: look for push.default	21:47
	→ mojzu joined	21:48
cbreak	you probably want upstream or similar	21:48
osse	cbreak: ahh, now I see what you mean.	21:48
ruckerz	oh im sure it's configurable.. but seems like it should be the default behavior	21:48
osse	cbreak: if he can identity which version of the code he started to work off of, that could work well indeed	21:48
cbreak	ruckerz: the old default is to push all branches that match names	21:48
osse	ruckerz: it's going to change in Git 2	21:48
cbreak	ruckerz: the new default will be something similar to upstream	21:48
	osse: yes, that's the key	21:49
	→ vedant joined	21:49
	← Orbitrix left	21:49
	→ Gentlecat_ joined	21:49
	← Gentlecat left	21:49
osse	I think it was supposed to change in Git 1.8. The Git 1.7.x release notes kept saying so. I guess it could pushed back	21:49
	gitinfo set mode: +v	21:50
	→ rideh joined	21:50
osse	I mean I guess it was pushed back	21:50
	← puchrojo left	21:50
	← Sigma left	21:50
	← AaronMT left	21:50
	← Bree left	21:51
	→ scwizard joined	21:52
scwizard	I have unmerged files	21:52
	so I can't checkout and I can't stash	21:52
	how do I continue doing stuff in other branches and then try the merge again later?	21:52
cbreak	that's normal.	21:52
	not at all.	21:52
	conclude the merge or abort	21:52
_ikke_	abort the merge, and retry the merge later	21:52
scwizard	how do I abort?	21:52
cbreak	git merge --abort	21:52
	surprise!	21:52
	→ Orbitrix joined	21:52
scwizard	cbreak: I tried that, it said merge was not possible because I have unmerged files	21:53
cbreak	then you have a prehistoric git without --abort	21:53
scwizard	https://gist.github.com/anonymous/5267149	21:53
cbreak	do a git reset --hard HEAD or similar	21:53
scwizard	ok	21:53
cbreak	(remember that this will nuke your working dir clean)	21:53
	→ pib1904 joined	21:53
osse	mushroom cloud	21:54
	← tswett_ left	21:54
scwizard	cbreak: that is fine	21:54
	git reset hard head worked	21:54
Gaddel	cbreak: okay, i tried to do what you suggested. i now have a problem in that my branch is based off a fork of another github branch of master. and that repo's master branch is out of date.	21:54
	← blackjack left	21:55
	← jlg13 left	21:55
cbreak	git rebase master	21:55
	while on your new branch	21:55
	that will rebase y our new branch onto your master	21:55
Gaddel	okay	21:56
crocket	git rebase slave	21:56
	→ penguin1 joined	21:56
osse	git rebase cable select	21:56
Gaddel	my branch is already up to date with this local copy of master	21:56
	← frenchlaf left	21:57
	← sgambino left	21:57
Gaddel	so i don't even see the point of git rebase master	21:57
	i'm working with a different git clone entirely at the moment	21:57
	there are 3 repos; the main one with the up to date master, a fork with an out of date master, and my fork of that fork	21:58
cbreak	Gaddel: why the hell?	21:58
Gaddel	i am not too good with git.	21:58
cbreak	do you want to make your life more complicated intentionally?	21:58
	Gaddel: git remote add somenamehere path/to/your/real/git/repo	21:58
	then git rebase somenamehere/master	21:58
	← trungqtran left	21:58
cbreak	after git fetch of course	21:59
	← breakingthings left	21:59
Gaddel	ok	21:59
	→ etcetera joined	21:59
	← pitzips left	22:00
	← sunya7a left	22:00
kini	_ikke_, charon, osse: thanks	22:00
	<charon> kini: so your code is broken. fix it?	22:00
	yeah I guess I will	22:00
	← mojzu left	22:00
	← nielsbusch left	22:01
kini	I wanted to have a "branch alias", which I was implementing as a symbolic link in .git/refs/heads	22:01
	→ juvenal joined	22:01
	→ mojzu joined	22:01
	← etcetera left	22:01
kini	it's probably easier to just copy the branch the first time and then synchronize the "alias" branch with the original branch in a post-receive hook	22:01
	← lavaman left	22:01
osse	may I ask why?	22:02
cbreak	kini: man git-symbolic-ref	22:02
gitinfo	kini: the git-symbolic-ref manpage is available at http://jk.gs/git-symbolic-ref.html	22:02
	→ rtjure joined	22:02
kini	cbreak: you can't pull those, can you?	22:03
cbreak	pull?	22:03
kini	yes	22:03
diegoviola	i'd like to compare the changes from one branch to another, and show a diffstat for the changes, how can i do that please?	22:03
	← adurity left	22:03
cbreak	from the outside they look like normal refs	22:03
kini	oh, ok - that's handy	22:03
cbreak	kini: git ls-remote .	22:03
	try it	22:03
	diegoviola: man git-diff	22:04
gitinfo	diegoviola: the git-diff manpage is available at http://jk.gs/git-diff.html	22:04
diegoviola	ty	22:04
	← m0ltar left	22:04
kini	<osse> may I ask why?	22:04
	long story... trying to integrate some git functionality into a bug tracker	22:04
cbreak	look at --stat, --numstat and so on	22:04
diegoviola	thanks	22:04
kini	since I can attach a branch to an issue, I want to be able to pull a generic branch name corresponding to the ticket number without knowing the name of the branch that's attached to the ticket	22:04
	but never mind, it's not that important anyway	22:05
	← Gaddel left	22:06
	← f8l left	22:06
	→ kcj joined	22:06
	→ etcetera joined	22:07
	→ weichkaes joined	22:07
	→ maxmanders joined	22:08
	← kcj left	22:09
	← nezZario left	22:09
	→ Lefont joined	22:09
	→ shamoon joined	22:10
	← infynyxx left	22:11
	→ maxmande_ joined	22:11
	→ apok joined	22:11
	← shamoon left	22:12
	← maxmanders left	22:12
	← mmitchell left	22:13
	← macabre left	22:14
	→ ozette joined	22:14
kevinfish	anyone know how to do git conflict resolution using xemacs emerge?	22:15
	i followed the instructions here:	22:15
	http://emacswiki.org/emacs/EmergeDiff	22:15
	← acharlieh left	22:15
kevinfish	but used gnuclient, dropped the -t and now I get the error message: file '-f' does not exist or is not readable	22:15
	← _br_ left	22:16
	→ fuzzybunny69y joined	22:17
	← tinti left	22:17
	← TeckniX_ left	22:18
	← elixir left	22:19
	→ trungqtran joined	22:19
	→ _br_ joined	22:20
	← trungqtran left	22:20
	← _br_ left	22:22
	→ kcj joined	22:22
	← mviitanen left	22:22
	→ _br_ joined	22:23
	→ f8l joined	22:23
	← _br_ left	22:23
	→ _br_ joined	22:24
	← mojzu left	22:25
	→ mojzu joined	22:26
	→ lresende joined	22:27
	→ keller joined	22:27
	← chrismar035 left	22:27
	← thl0 left	22:28
	← etcetera left	22:29
	← russt left	22:29
	← _delirium left	22:30
	← chuckharmston left	22:30
	→ l_r joined	22:30
l_r	hello	22:30
gitinfo	l_r: 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.	22:30
l_r	what is the status of git++?	22:30
	→ meoblast001 joined	22:30
l_r	i am curious about it since one time ago i read some bullshit from linus being against C++ for implementing git	22:31
	→ Transcended joined	22:31
	→ daniel_- joined	22:31
	← daniel_- left	22:31
	→ daniel_- joined	22:31
meoblast001	hello. i have a file .classpath which uses NT line endings, but my version of Eclipse converted it to UNIX line endings. now Git shows it under git status as a changed, unstaged file... naturally i do git checkout -- .classpath to revert this change, but it doesn't seem to work, the file is still listed as changed	22:32
	→ etcetera joined	22:32
meoblast001	is there some way to get this file to go back to the way it's supposed to be at head?	22:32
	→ elixir joined	22:32
	→ Noldorin joined	22:34
	→ kmels joined	22:36
	← rakl left	22:36
	→ mitz joined	22:37
	← pr0ton left	22:38
	← pt left	22:38
osse	l_r: git has grown since then and it's all C.	22:38
	l_r: with some circular stuff involving shell and perl	22:38
	← Lefont left	22:38
	← thinkdevcode left	22:39
	← rtjure left	22:39
	→ pr0ton joined	22:39
	→ lavaman joined	22:39
	→ happyface joined	22:39
	→ Lefont joined	22:41
jphillips	What should the BASE file be when merging?	22:41
	When cherry picking	22:41
	→ infynyxx joined	22:41
	← infynyxx left	22:42
	← jdunck left	22:42
	→ infynyxx joined	22:42
	← revagomes left	22:43
	→ taylorlapeyre joined	22:44
	← mikehaas763 left	22:44
	→ Bitwise joined	22:44
Bitwise	Hello.	22:44
gitinfo	Bitwise: 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.	22:44
	→ jdunck joined	22:45
ruckerz	!backup	22:45
gitinfo	Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	22:45
Bitwise	Has anyone ever used Git in IntelliJ? I need a little quick start guide, a little push in the right direction. (No pun intended)	22:45
	→ theotherguy joined	22:45
	← kevwil left	22:46
	→ Sigma joined	22:49
johnw	i have, but have forgotten it all	22:49
	← mitz left	22:49
	← Lefont left	22:49
	→ pt joined	22:50
	→ \|Freeman\| joined	22:50
	← gwty left	22:51
Bitwise	:-\	22:51
	← v0n left	22:52
Bitwise	johnw, Do you remember anything?	22:52
	← etcetera left	22:52
johnw	i remember it worked	22:52
	← Phrogz left	22:53
johnw	then Java melted my brain away	22:53
	← psoo left	22:53
xrq	sad, sad times. I switched to a team that isn't using git	22:53
	maybe it's time to play technology advocate	22:53
	→ etcetera joined	22:54
	← Goles left	22:55
	← BadDesign left	22:55
	→ gwty joined	22:55
Bitwise	xrq, Can you help me really quick? I'm trying to clone a repo so I can edit and submit my contributions.	22:56
	← f8l left	22:56
Bitwise	I'm doing it from within my IDE. (IntelliJ)	22:56
	← redrocket left	22:57
	→ rtjure joined	22:57
	→ mikehaas763 joined	22:58
	→ monod joined	22:59
	→ sgambino joined	22:59
	← leeN left	23:00
	← looper left	23:00
	← rtjure left	23:01
	← nathancahill left	23:01
	← mojzu left	23:01
	← dusted left	23:01
	→ krasnozer joined	23:02
	← oinkon left	23:02
	← AnimeCYC left	23:03
	← infynyxx left	23:04
	← thiago left	23:04
	→ rtjure joined	23:04
	→ rakl joined	23:05
	← krasnozer left	23:05
	← anonymuse left	23:06
	← monod left	23:06
	← cutaway left	23:06
xrq	Bitwise: I've never used IntelliJ - if you want to clone a repo though just use git clone	23:07
	→ thiago joined	23:08
	→ teratorn joined	23:08
	→ robbyoconnor joined	23:08
	← R1KRDO left	23:08
teratorn	hi, how can I git add a file while ignoring mode changes?	23:09
	← x82_nicole left	23:09
	← setmeaway2 left	23:10
justache	teratorn, did you try googling that question?	23:11
	← fuzzybunny69y left	23:12
	← teratorn left	23:12
justache	git config core.filemode false	23:12
	git add <your file>	23:12
	git config core.filemode true	23:12
	← robbyoconnor left	23:12
	← etcetera left	23:13
Bitwise	justache, Do you have a link to a quick start guide?	23:13
	← john_anderson left	23:13
justache	like a cheat sheet?	23:13
	← rtjure left	23:14
Bitwise	There is a project I'd like to contribute to but I can't figure out how to import it into my IDE	23:14
	← vedant left	23:14
justache	I'm afraid I don't	23:15
Bitwise	Darn.	23:15
	→ alunduil joined	23:15
	← alvaro_o left	23:15
	← decotii left	23:16
justache	Bitwise: http://www.sbf5.com/~cduan/technical/git/	23:16
	also http://try.github.com/levels/1/challenges/1	23:16
kini	What can I expect to do with a read-only gitdir?	23:17
	Can I assume that git commands that don't actually "do anything" to the state of the repo don't need write access?	23:17
	← Transcended left	23:18
kini	(such as `git log`, `git show`, etc.)	23:18
Bitwise	Thank you justache	23:18
	→ mmitchell joined	23:19
	← jjore left	23:20
	← jdolitsky left	23:20
	→ infynyxx joined	23:21
	→ f8l joined	23:21
	← pantsman left	23:21
	→ AdrienBrault joined	23:21
	→ etcetera joined	23:23
	→ cramm joined	23:23
	→ trungqtran joined	23:23
	gitinfo set mode: +v	23:23
	gitinfo set mode: +v	23:24
	← mmitchell left	23:24
	← marcdel left	23:24
	← Element9 left	23:26
	→ fly9 joined	23:26
	→ Jojomofoman joined	23:27
kevinfish	so I had a conflict and I installed meld and did git config --global merge.tool meld but now when I go git mergetool it says "no files need merging." The file is not fixed. How do I get git to recognize that?	23:27
	← thasmo left	23:28
	→ oinkon joined	23:29
	← gusnan left	23:30
	→ kmels_ joined	23:30
	← l_r left	23:30
	→ normanrichards joined	23:32
	← maxmande_ left	23:33
	← infynyxx left	23:33
	← kmels left	23:33
	→ jdolitsky joined	23:33
	→ maxmanders joined	23:34
	← lokus left	23:34
	← NiteRain left	23:34
	← jargon left	23:35
	→ pastime joined	23:36
	→ maxmande_ joined	23:37
	→ sysadminAli joined	23:37
	→ thl0 joined	23:38
	→ landypro joined	23:38
	← landypro left	23:38
	← maxmanders left	23:38
	← bzyx left	23:39
	← danishkhan left	23:40
	→ maxmanders joined	23:40
	← gwty left	23:41
	← maxmande_ left	23:41
	→ daed joined	23:42
	→ gwty joined	23:42
	→ maxmande_ joined	23:42
	← thl0 left	23:43
	→ AdobeArtist joined	23:43
	→ wrouesnel joined	23:43
	← AdobeArtist left	23:43
	← aberrant left	23:43
	← cramm left	23:43
	→ AdobeArtist joined	23:44
	← fly9 left	23:44
	← AdobeArtist left	23:44
	← maxmanders left	23:45
	← frewsxcv94040 left	23:45
	← Davey left	23:45
	← jonshea left	23:46
	← JosefAssad left	23:46
	← maxmande_ left	23:47
	← ozette left	23:47
	← sheerun left	23:48
	← FernandoBasso left	23:48
	→ x82_nicole joined	23:48
	→ mojzu joined	23:49
	→ FernandoBasso joined	23:49
	→ jonshea joined	23:50
	→ aberrant joined	23:50
	← lresende left	23:51
	← jonshea left	23:51
	→ mitz joined	23:53
	← etcetera left	23:53
	→ mklappstuhl joined	23:54
	← trungqtran left	23:54
	→ etcetera joined	23:54
	← bremner left	23:55
	← Ryback_ left	23:55
	← jdunck left	23:55
	← BadProphet left	23:57
	← mitz left	23:57
	→ ozette joined	23:57
	→ Davey joined	23:58
	→ AaronMT joined	23:58
	← mloskot left	23:59
	← sq-one left	23:59
	← tvw left	23:59
	→ bremner joined	23:59
	← bremner left	23:59
	→ bremner joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation