IRCloggy #git 2013-12-09

Provider of IRC logs since 2005.

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

	2013-12-09
	← robbyoconnor left	00:00
	← WaReZ left	00:00
	→ LuxuryMode joined	00:00
epta	osse: "dir" here is undefined. Question is how to know where git will clone specific repo. I don't wanna split url and retrieve potentially dir name from it, than check if that dir already exist, then add 1 (or what git actually does) and so on. Git already do that during "git clone" so question is how to get dir name before cloning or make git return it after clone.	00:01
	← irqq left	00:01
osse	epta: it sounds easier to specify the dir yourself	00:02
	but from what I can tell the "algorithm" to find the dir name is the text after the last slash but removing .git it's there	00:02
	← valgrind left	00:03
epta	osse: it is not all "algorithm"	00:03
	→ cyphase joined	00:03
imMute	osse: why not just tell git where to put it? you can do that you know	00:03
Hello71	!xyproblem ?	00:03
	!xy ?	00:03
gitinfo	Woah, slow down for a bit. Are you sure that you need to jump to that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...	00:03
	← rjhunter left	00:04
osse	imMute: you're preaching to the choir, buddy	00:04
imMute	osse: whups. silly 4 letter nicks everywhere.	00:04
	→ andschwa joined	00:05
epta	I don't udnerstand why should I reinvent this "choosing-potential-directory-to-clone" algorithm if git already did it great	00:06
imMute	epta: you don't. you just tell git where to put the clone.	00:06
Tantris	offby1: yes this sounds great except that it gives me no output: git show master README.md	00:06
offby1	Tantris: I've probably got the syntax wrong. Hold on	00:06
epta	imMute: how?	00:06
osse	epta: git clone url dir	00:07
epta	Oh lord	00:07
	Again, dir is undefined here	00:07
Tantris	offby1: the other thing is that as I understand this would just "copy" the version from master to my branch, but I want to merge it to my branch	00:07
offby1	Tantris: I think you omitted a colon. ``git show master:.pythonrc`` works for me	00:07
imMute	no its not	00:07
	epta: no, its defined by you, to be whatever makes sense to your application.	00:07
osse	epta: you asked how to tell git-clone where to put the clone. Well, that's how you do it	00:07
offby1	Tantris: about merging versus copying: you're confused.	00:07
epta	osse: why are you think that I'm asking about it? Read my question again, please	00:08
imMute	epta: counter question: why do you have to let Git pick where to put the clone?	00:08
Tantris	offby1: well as I understand it the show will give me the exact contents of the commit and that would replace my current version	00:09
osse	epta: git-clone doesn't have the flag you're looking for. I think the easiest way for you to do what you want is to decide on a name beforehand.	00:09
	On the other hand, parsing that "Cloning into 'foo'" line shouldn't be too difficult	00:09
offby1	Tantris: it won't replace anything unless you redirect the output, but I recommend that because I thought that's what you wanted.	00:09
osse	Another (ugly) option is to compare which directories exist before and after	00:09
Tantris	offby1: What I want is to merge / rebase my changes to a single file on top of a branch	00:09
epta	imMute: because process that calls git clone thinks that git should know about where to clone this repo better than that process	00:10
offby1	Tantris: you need to realize that you don't actually want to merge or to rebase.	00:10
	merging and rebasing affect an entire project, but you've said you only want to affect a single file.	00:10
	← andschwa left	00:10
offby1	Therefore you don't want to merge or to rebase.	00:10
imMute	epta: I would disagree with that, but I don't know your situation. therfore you have to parse the output of 'git clone'	00:10
epta	osse: yep, parsing seems to be the working solution here	00:10
	→ andschwa` joined	00:10
offby1	QED	00:10
	→ FunkyELF joined	00:11
osse	epta: are you not in control of the process that does the cloning? Isn't it scriptable/tunable?	00:11
	← ErKa left	00:11
Tantris	offby1: yeah person A and B made changes to a repository, As changes are in master and Bs changes are in a branch, I now checked out a new branch based on Bs and want to pull all changes from master from a single file and forget all other changes in master	00:11
FunkyELF	Trying to follow this. On step 4 I get "fatal: Couldn't find remote ref master" http://stackoverflow.com/questions/4658606/import-existing-source-code-to-github	00:12
	← mike_af left	00:12
	→ TN_Jonesboro joined	00:12
Tantris	offby1: but both A an B made changes to this specific file and I want to keep all of them	00:12
	→ mike_af joined	00:12
offby1	Tantris: are you saying that you want to preserve the history of this file as well as its contents? If so, I don't know how to do that.	00:12
	oh, you want to "merge a single file". I see	00:12
	I'd just do "git merge --no-commit", then unstage all the other files before committing.	00:13
	Never done that, but it sounds reasonable.	00:13
osse	FunkyELF: what's the exact command you're trying to execute?	00:13
epta	osse: I am. But I wanna make it maximum windows/linux portable, so manipulating with dirs and filenames is the last thing what I'll do here	00:13
	→ scientes joined	00:13
Tantris	offby1: ahaha yeah you are right, just merge everything and after that unstage all except the specific file	00:14
FunkyELF	osse: git pull origin master	00:14
	→ normaliy joined	00:14
osse	epta: hmm, there's is no manipulating going on. It's just a matter of deciding on a name and giving it to git-clone	00:14
FunkyELF	osse: I think I'm good now. I did "git push -u origin master"	00:15
osse	FunkyELF: how is that step 4 ?	00:15
	FunkyELF: ahh, yeah the -u is necessary	00:15
FunkyELF	osse: sorry... it was step 5	00:15
	I never ran step 5 and just want to the git push origin master and it worked.	00:15
	→ mike_af_ joined	00:16
	← mike_af left	00:16
	→ john2x joined	00:17
	← blast_hardcheese left	00:18
	→ hellopat joined	00:19
	→ madewokherd joined	00:19
	← koda left	00:19
	← halu left	00:21
	→ maestrojed joined	00:23
	← akheron left	00:23
	← andschwa` left	00:24
	→ akheron joined	00:24
	← dkpi left	00:24
	→ blast_hardcheese joined	00:25
	→ jim joined	00:25
	→ dsdeiz joined	00:26
jim	hi, what happens if I try to commit with nothing added?	00:26
Cathy	well, you can also commit the removal of files	00:27
	or typechanges, renames, etc.	00:27
	if there is nothing at all to commit then you cannot commit	00:27
jim	and it doesn't alter the repo in that case?	00:28
Cathy	because it won't happen, yes	00:28
jim	got it, thank you	00:28
	→ gittish joined	00:29
Cathy	jim: looking at the man page there's an option to forcibly commit en empty commit, namely --allow-empty	00:29
	the default is not to allow it however	00:30
	→ lc0dd0cl joined	00:30
jim	oh, interesting... for the purpose of sticking an arbitrary commit message?	00:30
Cathy	the man page says it is "primarily for use by foreign SCM interface scripts. "	00:31
	← LuxuryMode left	00:32
Hello71	e.g. you want to poke the post-update hook to deploy	00:32
	← TN_Jonesboro left	00:32
	← lc0dd0cl2 left	00:32
jim	cathy, ahh, whose developers asked for that?	00:32
	→ andschwa` joined	00:33
	→ dhruvasagar joined	00:33
	← DolphinDream left	00:33
	→ TN_Jonesboro joined	00:33
	← maestrojed left	00:33
Cathy	jim: i wouldn't know	00:33
jim	ok, well at least thanks for your time so far. appreciated.	00:34
	→ LuxuryMode joined	00:34
	→ tmcmahon joined	00:35
	← dxdt left	00:37
	→ PaulCapestany joined	00:38
	← Tantris left	00:39
	→ St_Marx joined	00:41
Celelaptop	How do I undo a merge currently conflicting?	00:44
SamB	"git merge --abort"	00:46
	← KK4MGV left	00:46
Celelaptop	"Entry 'quad.c' not uptodate. Cannot merge."	00:46
	→ treaki_ joined	00:47
Celelaptop	I tried to resolve the conflicts before I quit merging.	00:47
	← LuxuryMode left	00:47
offby1	Celelaptop: "git reset --hard"	00:48
	throws it away, goes back to where you were	00:48
Celelaptop	thanks.	00:48
	← treaki__ left	00:48
	← lc0dd0cl left	00:49
	→ rjhunter joined	00:49
Cathy	git reset --hard might undo more than the failed merge has done	00:50
	so long as you realise that before using it...	00:50
	← rjhunter left	00:50
SamB	hmm, it can be hard to keep track of what commands have an --abort flag :-(	00:50
	→ yhager joined	00:51
	→ rjhunter joined	00:51
offby1	Cathy: details, details	00:51
	honestly I can't think of any other way. Maybe "git status" has a suggestion	00:52
	← rjhunter left	00:52
	→ hyperair joined	00:52
	→ h0rrorvacui joined	00:52
	← tureba_ left	00:52
Cathy	git reset --hard will discard in loose changes in the working directory	00:52
	which might have been kicking around before the attempted merge	00:52
	that's what i meant	00:52
	→ tureba joined	00:52
	→ rjhunter joined	00:52
	→ hugo_dc joined	00:52
offby1	Yeah, I know	00:53
	→ KK4MGV joined	00:53
Celelaptop	<Cathy> git reset --hard might undo more than the failed merge has done <-- I know. But I had everything committed before I try to merge.	00:53
	→ rhunter joined	00:54
	← scientes left	00:55
	← zhost left	00:55
	← rhunter left	00:55
SamB	Celelaptop: good :-)	00:55
	→ pweaver_ joined	00:56
	→ rhunter joined	00:57
	← rjhunter left	00:57
	→ robbyoconnor joined	00:58
	→ ewnd9 joined	01:00
	← rhunter left	01:01
	→ mk_qi joined	01:02
	→ dyng joined	01:04
	→ naifas joined	01:05
	← dsdeiz left	01:05
	→ sohail joined	01:06
	← sohail left	01:06
	→ sohail joined	01:06
	← Kayra left	01:06
	→ dxdt joined	01:06
	→ codeman joined	01:06
	← gnoel left	01:07
	← harish left	01:07
	→ rjhunter joined	01:08
	→ SonicX joined	01:09
	→ JPT joined	01:09
	← CheckDavid left	01:10
	→ iliketu__ joined	01:11
	← Zarthus left	01:12
	→ vipzrx joined	01:12
	← HelloShitty left	01:14
	→ scientes joined	01:15
	→ fuzzybun1y69y joined	01:15
	← mdiers left	01:16
	← ewnd9 left	01:17
	← phipes left	01:18
	→ SocalNick joined	01:21
	← naifas left	01:22
scientes	there is no way to fetch tags only on one branch?	01:23
	they arn't indexed like that?	01:23
	→ nwertman joined	01:24
scientes	i'm having to pull down the entire linux-stable to run one git tag --contains	01:24
	trout → variable	01:24
	← yhager left	01:26
	← rafaqueque left	01:28
	← echo-area left	01:29
offby1	scientes: it's common to have to download the whole repository in order to do _anything_	01:29
	there are probably exceptions, but they're narrow	01:29
	← Guest32973 left	01:29
	→ Droolio joined	01:29
	→ marlboromoo joined	01:29
scientes	yeah but all the branches	01:30
	→ pmahent joined	01:30
	→ lindenle joined	01:30
	← kermit left	01:31
offby1	what about them?	01:31
	→ nwertman_ joined	01:31
	→ gnoel joined	01:31
scientes	anyways its now downloaded over slow starbucks internet	01:31
offby1	heh	01:31
	→ kermit joined	01:32
	← codeman left	01:33
	← nwertman left	01:34
scientes	now i feel like i have to run gc --aggressive	01:34
	← gnoel left	01:35
	← tmcmahon left	01:36
	← dxdt left	01:39
	← lifelike left	01:39
	→ momomomomo joined	01:39
	→ nettoweb joined	01:39
	← blast_hardcheese left	01:39
	← M00sL0gB0t_ left	01:41
	→ M00sL0gB0t_ joined	01:41
	→ weiyang joined	01:42
	→ phipes joined	01:43
	← M00sL0gB0t_ left	01:43
	← SocalNick left	01:43
	→ M00sL0gB0t joined	01:44
	← pweaver_ left	01:44
	← sam113101 left	01:46
	← dyng left	01:46
	→ yourbestfriend joined	01:47
	← marlboromoo left	01:47
	← M00sL0gB0t left	01:47
	→ daito joined	01:49
	→ M00sL0gB0t joined	01:49
	← yourbestfriend left	01:49
	→ mmcdaris joined	01:50
	→ marlboromoo joined	01:50
	→ yourbestfriend joined	01:51
	← AzizLight left	01:52
	→ LuxuryMode joined	01:52
	→ dsdeiz joined	01:52
	← nettoweb left	01:52
	← momomomomo left	01:54
	← yourbestfriend left	01:54
	← luxgeek left	01:55
	→ luxgeek joined	01:55
	→ dyng joined	01:56
	→ yourbestfriend joined	01:57
	→ blast_hardcheese joined	01:57
	→ luc_ joined	01:58
	→ hellome joined	01:58
	→ rchavik joined	01:58
	← andschwa` left	01:58
	← MetalMatze left	01:59
	← SHyx0rmZ left	01:59
	→ dxdt joined	02:00
	← yourbestfriend left	02:00
	→ bunbury joined	02:01
	← robbyoconnor left	02:01
	→ yourbestfriend joined	02:01
	→ yuwang_ joined	02:01
	→ DolphinDream joined	02:02
	← Mez_ left	02:03
	← h0rrorvacui left	02:03
	→ Mez joined	02:03
	← dhruvasagar left	02:04
	← blast_hardcheese left	02:05
	← hellome left	02:05
	→ hellome joined	02:05
	→ saorge_ joined	02:07
	→ andschwa` joined	02:08
	← svm_invictvs left	02:09
	← saorge left	02:10
	→ svm_invictvs joined	02:11
	→ kara_sweets joined	02:15
	→ harish joined	02:15
	← Mez left	02:15
	→ Mez joined	02:15
	← gittish left	02:16
	← KillYourTV left	02:16
	← kara_sweets left	02:17
	← hellome left	02:17
	→ KillYourTV joined	02:18
	→ hellome joined	02:18
	← Vampire0 left	02:19
	← hugo_dc left	02:21
	s0beit\|ZNC → s0beit	02:21
	→ dkpi joined	02:21
	← Mez left	02:23
	→ Mez joined	02:23
	→ fandikurnia01 joined	02:23
	→ echo-area joined	02:24
	← dekiss left	02:26
	→ yhager joined	02:28
	← Mez left	02:28
	→ d8bhatta joined	02:29
	→ Mez joined	02:29
	→ michael_lee joined	02:29
	→ joelrobichaud joined	02:30
	← michael_lee left	02:32
	← Mez left	02:34
	→ Mez joined	02:34
	→ michael_lee joined	02:35
	→ el3ctron joined	02:36
	← fandikurnia01 left	02:36
	← Gamecubic_ left	02:37
	→ zophy joined	02:37
	→ fandikurnia01 joined	02:37
	← iliketu__ left	02:37
	← chrisss404 left	02:39
	→ dhruvasagar joined	02:39
	← dhruvasagar left	02:40
	← el3ctron left	02:41
emma	Hey what does this mean? Cloning into plugapi...	02:41
	error: SSL certificate problem, verify that the CA cert is OK. Details:	02:41
	error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed while accessing https://github.com/atomjack/plugapi.git/info/refs	02:41
	→ Kayra joined	02:41
	→ el3ctron joined	02:41
emma	do i have to have a git key?	02:41
alindeman	Seems like you might have an outdated CA bundle	02:41
	← teran left	02:42
	→ chrisss404 joined	02:42
alindeman	emma: https://help.github.com/articles/error-ssl-certificate-problem-verify-that-the-ca-cert-is-ok	02:42
	← kpreid left	02:42
	→ kpreid joined	02:42
	→ HawkTracker joined	02:42
emma	i can use ssh instead	02:43
	but if im on a different server than i usually use	02:43
	isn't that a problem if it asks for my key?	02:43
	← fuzzybun1y69y left	02:43
HawkTracker	hello how can I use git to copy contents of latest commit from a certain branch to another directory	02:43
alindeman	emma: I'm not sure what you mean, sorry	02:43
emma	im not either	02:43
	when i git clone from my home server it asks me for a password each time	02:44
	im now on a different server	02:44
	← normaliy left	02:44
emma	i imagine it will ask me for a password but i figure i dont have my key on this server so how is that going to work	02:44
skorgon	HawkTracker: man git archive	02:44
gitinfo	HawkTracker: the git-archive manpage is available at http://jk.gs/git-archive.html	02:44
HawkTracker	can i git clone locally?	02:44
skorgon	HawkTracker: sure you can	02:45
alindeman	emma: I don't know. Solving the CA issue is usually just a matter of updating an OS package	02:45
HawkTracker	git archive forces .zip though...	02:45
Hello71	wat	02:45
	man git archive	02:46
gitinfo	the git-archive manpage is available at http://jk.gs/git-archive.html	02:46
Hello71	man git archive	02:46
gitinfo	the git-archive manpage is available at http://jk.gs/git-archive.html	02:46
Hello71	man git archive	02:46
gitinfo	the git-archive manpage is available at http://jk.gs/git-archive.html	02:46
	← pbujok left	02:46
Hello71	strictly speaking you could use git checkout	02:46
	← Mez left	02:46
	← tidu12 left	02:47
	→ Mez joined	02:47
	→ Gamecubic_ joined	02:47
	→ AzizLight joined	02:48
	← LuxuryMode left	02:49
	← hellopat left	02:51
	→ g0bl1n joined	02:52
	← Mez left	02:52
rjhunter	emma: If you're using git on Windows, you might need to set up the initial trusted root certificates	02:52
	→ Than joined	02:52
	→ Mez joined	02:52
	← AzizLight left	02:53
	← Kayra left	02:53
	→ robbyoconnor joined	02:54
	← PaulCapestany left	02:54
	← echo-area left	02:54
	→ SamB_ joined	02:56
	→ hellopat joined	02:57
	← SamB_ left	02:57
scientes	anyone have a script that bisects the kernel with kvm/qemu and a test program?	02:59
	← andschwa` left	03:00
	→ silviof1 joined	03:01
	← mike_af_ left	03:01
	→ echo-area joined	03:01
	→ kevwil joined	03:02
skorgon	scientes: i don't know if there is anything useful or even working, but tools/testing/ktest might be helpful. there seems to be a kvm example configuration	03:03
	← petey left	03:03
	← grep0r left	03:03
	→ oojacoboo joined	03:03
	→ petey joined	03:03
	← silviof left	03:04
emma	rjhunter: no im using it on linux	03:04
	← weiyang left	03:04
rjhunter	scientes: that seems like the sort of thing that would have come up a few times in the history of kernel development -- if you discover a hidden gem, it's probably worth a blog post for future Web searchers	03:04
	→ andschwa` joined	03:05
	← Mez left	03:05
	→ Mez joined	03:05
emma	rjhunter: this is the message i get when i try git clone with ssh:	03:05
	Permission denied (publickey)	03:05
	hey Mez i haven't seen you around in ages	03:05
	→ dsockwell joined	03:06
emma	isn't there some git ssh key or something you need?	03:06
	← bdavenport left	03:06
skorgon	there is no 'git ssh key' just your usual ssh key(s).	03:07
	→ weiyang joined	03:07
	← hahuang65 left	03:07
	← petey left	03:08
	→ bdavenport joined	03:08
	→ mizu_no_oto joined	03:08
rjhunter	emma: If you haven't used key-based authentication with SSH before, you'll need to generate a keypair first. Then give your public key to whoever's on the other side (GitHub account preferences, or ~/.ssh/authorized_keys)	03:09
emma	oh my	03:09
	i see	03:09
	i have to generate a key with ssh..	03:09
	← SonicX left	03:09
IsoLinCHiP	Is there a way to split up .gitrc into multiple files that get sourced from the main file?	03:09
	← Mez left	03:09
skorgon	IsoLinCHiP: man git config ther is a whole section regarding includes	03:11
gitinfo	IsoLinCHiP: the git-config manpage is available at http://jk.gs/git-config.html	03:11
	→ Mez joined	03:11
	→ wrouesnel joined	03:11
	→ guns joined	03:11
IsoLinCHiP	Sorry, my mistake was searching for .gitrc clearly is the wrong filename :(	03:11
	but thx skorgon :)	03:12
	→ teran joined	03:13
	← dyng left	03:15
	← regreddit left	03:15
	→ grep0r joined	03:16
	← yhager left	03:16
	→ dyng joined	03:17
	← harv left	03:17
	→ batrick joined	03:18
	→ fairuz joined	03:19
	← Mez left	03:20
	← wrouesnel left	03:20
	→ Mez joined	03:21
	← mitz left	03:21
	← dyng left	03:21
	← andschwa` left	03:23
	→ harv joined	03:23
	← karavanjo__ left	03:23
	→ mitz joined	03:24
	← dxdt left	03:24
	← phipes left	03:24
	→ phipes joined	03:25
	← Mez left	03:26
	← mizu_no_oto left	03:26
	→ Mez joined	03:26
	← psyl0n left	03:26
	→ acharlieh joined	03:28
	← dvide_ left	03:29
	→ ariscop joined	03:29
	← grep0r left	03:29
	→ petey joined	03:29
	→ andschwa` joined	03:30
	→ asamat joined	03:30
	→ grep0r joined	03:32
	← phipes left	03:33
	→ phipes joined	03:33
	← phipes left	03:34
	← asamat left	03:38
	← guns left	03:39
	← kevwil left	03:42
	→ PaulCapestany joined	03:42
	← Mez left	03:43
	→ Mez joined	03:44
	← ecube left	03:45
	→ danecando joined	03:45
	← go1111111 left	03:46
	← teran left	03:46
	← danecando left	03:47
	→ ecube joined	03:47
	← HawkTracker left	03:48
	→ arvind_khadri joined	03:48
	← Mez left	03:49
	→ AzizLight joined	03:49
	← daito left	03:52
	→ aberrant joined	03:52
	← dkpi left	03:53
	← lindenle left	03:53
	← AzizLight left	03:53
	← Wulf left	03:54
	→ Mez joined	03:55
	→ Wulf joined	03:56
	→ go1111111 joined	03:58
	s0beit → s0beit\|ZNC	03:59
	s0beit\|ZNC → s0beit	03:59
	→ daito joined	04:00
	→ decimad2 joined	04:02
	→ chernjie joined	04:02
	← decimad left	04:04
	← scientes left	04:04
	← Noldorin left	04:04
	→ Kayra joined	04:04
	→ phipes joined	04:04
	← Mez left	04:05
	→ Mez joined	04:06
	→ Noldorin joined	04:06
	← Kayra left	04:09
	← mmcdaris left	04:10
	← Mez left	04:11
kent\n	if you're on an older git that doesn't have `git tag --points-at sha1`, how would you go about getting that?	04:11
	← Noldorin left	04:11
	→ Mez joined	04:12
rjhunter	kent\n: probably start with `git rev-list --tags`	04:12
	← bgerber left	04:12
	← fandikurnia01 left	04:13
SamB	or "git ls-remote ."	04:13
	→ mmcdaris joined	04:13
	→ fandikurnia01 joined	04:13
	← Dougie187 left	04:14
	→ bgerber joined	04:14
SamB	rjhunter: that doesn't include the names of the tags ...	04:14
kent\n	and isn't ls-remote also a newer tool?	04:14
	kent\n is having trouble knowing what is and isnt' available on stuff that he may be supporting	04:15
	← d8bhatta left	04:15
rjhunter	SamB: good point	04:15
mattcen	kent\n: Check the changelog? ls-remote appears to have been there since before version 1.0	04:16
rjhunter	kent\n: how far back do you want to support? git.git has all the details if you really want to check	04:16
SamB	mattcen: indeed it does	04:17
	I was just going to say the same thing	04:17
	it looks like it was there since before they started the changelog	04:17
kent\n	Well, I'm not going to support back further than I have to, I'm just reacting in responses I see to public automated testing of code, I see a failure, I'll add support.	04:18
	→ Imagineer2 joined	04:18
	← preflex left	04:18
	→ preflex_ joined	04:18
kent\n	hrm, git ls-remote is a good start	04:19
	just it gives lots of lines	04:19
	somehow I see 2 sets of the same tag	04:19
	tag , and tag^{}	04:19
	preflex_ → preflex	04:19
	← oojacoboo left	04:19
	← Guest83652 left	04:20
	← Imagineer left	04:21
	→ ewnd9 joined	04:21
mattcen	kent\n: The latter points to a tree rather than a commit	04:22
kent\n	mattcen: weird, git cat-file -t says without ^{} is a tag, and with ^{} is a commit	04:23
	mattcen looks again	04:23
	← Mez left	04:24
mattcen	Sorry, you're right.	04:24
	→ xjiujiu joined	04:25
	→ lifelike joined	04:25
	← go1111111 left	04:26
mattcen	That's only the case for heavy-weight tags; if you have a heavy-weight tag, tagname points to the tag which contains the description, date, creator etc., and tagname^{} points to the relevant commit	04:26
	For a light-weight tag, tagname points to the commit itself, and there is no tagname^{}	04:26
	← freestyl3r left	04:26
	→ psyl0n joined	04:27
kent\n	aha. I was wondering how to disambiguate annotated tags too =)	04:27
	← HicksD left	04:28
	→ HicksD joined	04:30
	← visy left	04:30
	← daito left	04:30
	← adiga left	04:30
	→ dyng joined	04:31
	→ Mez joined	04:31
	← psyl0n left	04:31
	← kermit left	04:31
mattcen	kent\n: git tag \| while read -r name; do if [[ $(git cat-file -t "$name") = tag ]]; then weight=Heavy; else weight=Light; fi; echo "${weight}-weight: $name"; done	04:32
	→ kermit joined	04:33
kent\n	I'm guessing for-each-ref is the "new" form of ls-remote?	04:33
	→ ethlor joined	04:35
	→ devians joined	04:35
mattcen	Oh, that would work too apparently; I have't used that before.	04:35
devians	hey all. do submodules only exist on the branch that they are created?	04:35
mattcen	devians: Yes	04:35
	Until you merge that branch into another branch	04:35
	Their existence in a given branch is determined by the .gitmodule file at the root of the repo.	04:36
	← Mez left	04:36
	← ewnd9 left	04:36
mattcen	Sorry, .gitmodules, plural	04:36
	→ Mez joined	04:36
	→ geomyidae joined	04:36
	→ teran joined	04:37
mattcen	Actually, that last part may not strictly be true; I actually think there needs to be some other metadata stored in a commit for it to reference a commit of a submodule, but I'm not sure of the details	04:37
	← JSharpe_ left	04:38
	← dyng left	04:38
	→ ewnd9 joined	04:38
devians	mattcen I'm on an old version of git on this box, 1.7.1, so hopefully thats still correct	04:39
	← ewnd9 left	04:39
mattcen	I expect that if that sort of thing changed, it'd break repos for a lot of people.	04:40
	→ ewnd9 joined	04:40
devians	I just know at some point the submodule handling changed	04:40
mattcen	Then you perhaps know more than I; I've not seen it change to my knowledge	04:41
	→ chuckharmston joined	04:42
	← SebastianFlyte left	04:42
	→ oojacoboo joined	04:43
	← mmcdaris left	04:43
	→ dreiss joined	04:45
	← ewnd9 left	04:45
	← Mez left	04:46
	→ dyng joined	04:46
	→ ewnd9 joined	04:46
	→ hoyang joined	04:47
	→ Mez joined	04:47
	→ AzizLight joined	04:50
	→ mike_af joined	04:50
	→ kevwil joined	04:50
	→ Vampire0 joined	04:52
	→ dsdeiz_ joined	04:52
	← dsdeiz left	04:53
	← ewnd9 left	04:53
	← AzizLight left	04:54
	→ ewnd9 joined	04:55
	← mike_af left	04:55
	← dyng left	04:56
	← kevwil left	04:57
	← ewnd9 left	04:57
	← oojacoboo left	04:57
	→ saurabh joined	04:57
	→ dyng joined	04:57
	→ ewnd9 joined	04:58
	saurabh → Guest15100	04:58
	→ jxpx777 joined	04:58
	→ SebastianFlyte joined	05:00
	← hyperboreean left	05:01
	→ greggroth joined	05:02
	→ redrocket joined	05:02
	← Mez left	05:03
	→ Mez joined	05:04
	→ dsdeiz joined	05:05
	← dsdeiz_ left	05:05
	← ewnd9 left	05:06
	→ mmcdaris joined	05:06
	← ethlor left	05:08
	← phipes left	05:09
	→ yhager joined	05:09
	← Watabou left	05:09
	→ phipes joined	05:09
	← teran left	05:09
	← chuckharmston left	05:10
	→ alezost joined	05:10
	→ chuckharmston joined	05:10
	← cilly left	05:12
	← Mez left	05:12
SamB	!broken reference	05:12
gitinfo	[!doesntwork] Sorry to hear it doesn't work, but how about some details? At this point we don't even know whether it did nothing, spit out an error message, or abducted your pet unicorn in a flying saucer. That makes it hard to troubleshoot – unicorn abduction requires a completely different solution than an error message, for instance. So, please be specific...	05:12
SamB	hmm, that is not what I was looking for	05:12
	→ Mez joined	05:13
	→ cilly joined	05:13
mattcen	!fix broken repo	05:14
gitinfo	[!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!	05:14
mattcen	SamB: That one?	05:14
	← phipes left	05:14
	← chuckharmston left	05:14
SamB	I ran into a mention of something called a "broken" reference in refs.c	05:15
	← decimad2 left	05:15
	← dreiss left	05:15
	← mk_qi left	05:16
	SamB grumbles about C's weak typing, and how that's NOT helping heap profiling at all ...	05:16
	← Demego left	05:19
	← Tommy[D] left	05:19
	← petey left	05:21
	← nicedice_ left	05:21
	← devians left	05:23
	→ johnflux__ joined	05:23
	→ revyd joined	05:23
	← enape_101 left	05:24
	→ dhruvasagar joined	05:25
	← Mez left	05:27
	→ Mez joined	05:28
	→ daito joined	05:28
	→ mk_qi joined	05:31
	→ Tommy[D] joined	05:32
	→ irqq joined	05:32
	← Mez left	05:33
	→ Mez joined	05:34
	← ColKurtz left	05:34
	← ManneW left	05:34
	← nephyrin left	05:34
	→ nephyrin joined	05:35
	← lun_ left	05:37
	→ beautifulmind joined	05:37
	← yhager left	05:38
	← Mez left	05:38
	→ Nicholas_Rishel joined	05:38
	← g0bl1n left	05:39
	→ Mez joined	05:39
	← m0viefreak left	05:40
	← ariscop left	05:40
	→ lun_ joined	05:40
	← kukks left	05:42
	← DolphinDream left	05:42
	→ longbeach joined	05:43
	← thiago left	05:43
	← dhruvasagar left	05:44
	← Mez left	05:44
	← dyng left	05:46
	→ karavanjo joined	05:46
	→ ManneW joined	05:48
	→ dmarkham joined	05:48
	← arvind_khadri left	05:49
	→ Mez joined	05:50
	→ Watabou joined	05:50
	← longbeach left	05:50
	→ AzizLight joined	05:50
	→ zeroprofit joined	05:51
	← zeroprofit left	05:51
	→ zequence joined	05:52
	→ dhruvasagar joined	05:52
	→ hahuang65 joined	05:53
	← AzizLight left	05:55
	← greggroth left	05:56
	← madewokherd left	05:57
	← dhruvasagar left	05:58
	→ dhruvasagar joined	05:58
	← Mez left	06:00
	→ h0bbit_ joined	06:00
	→ Mez joined	06:01
	→ ir7466 joined	06:01
	← tpollard left	06:02
	← bbloom_ left	06:05
	← Mez left	06:06
	→ Mez joined	06:06
	← dhruvasagar left	06:09
	← echo-area left	06:10
	→ dhruvasagar joined	06:10
	→ GT4066 joined	06:12
	→ echo-area joined	06:12
	→ psyl0n joined	06:15
	← johnflux__ left	06:15
	← echo-area left	06:17
	← Mez left	06:18
	→ echo-area joined	06:18
	← matanya left	06:19
	→ dik_dak joined	06:19
	← tjbiddle left	06:19
	→ Mez joined	06:19
	← psyl0n left	06:19
	← Watabou left	06:21
	→ trusktr joined	06:21
	→ maleknet joined	06:22
	← maleknet left	06:22
	→ JEEB joined	06:23
	← Mez left	06:24
	→ Mez joined	06:25
	→ dhruvasa1ar joined	06:25
	→ chuckharmston joined	06:25
	← dhruvasagar left	06:27
	← PeterBooker left	06:28
	← trusktr left	06:28
	← misterpink left	06:28
	→ wrouesnel joined	06:29
	→ jargon joined	06:29
	→ yhager joined	06:30
	→ thiago joined	06:31
	→ decimad joined	06:31
	← Mez left	06:32
	← lbf1 left	06:32
	→ Mez joined	06:33
	← JEEB left	06:33
	→ cff joined	06:33
	→ Watabou joined	06:33
	→ trusktr joined	06:36
	← yhager left	06:38
	← Mez left	06:38
	→ Mez joined	06:39
	← irqq left	06:39
	→ mike_af joined	06:39
	→ dyng joined	06:39
	→ ThomasLocke joined	06:40
	← bwreilly left	06:41
	→ iliketu__ joined	06:42
	← zophy left	06:43
	← asdfasdf left	06:43
	→ JEEB joined	06:43
	→ mishok13 joined	06:44
	← mike_af left	06:44
	→ asdfasdf joined	06:45
nitrix	Yeah yeah, always pick on the language when it's actually the developer's fault.	06:46
	← cybersphinx left	06:46
	← Mez left	06:46
	→ bbloom joined	06:47
	→ Mez joined	06:47
	← thiago left	06:48
	→ thiago joined	06:49
	→ pantsman joined	06:50
SamB	nitrix: hey, I didn't write git	06:50
nitrix	:>	06:50
	→ Zarthus joined	06:50
	→ AzizLight joined	06:51
	← etcetera left	06:53
	← ValicekB left	06:53
	→ gittish joined	06:54
	← gittish left	06:55
	→ johnkevinmbasco joined	06:55
	→ subhojit777 joined	06:55
	← AzizLight left	06:56
	← Mez left	06:57
	→ Mez joined	06:58
	→ arvind_khadri joined	06:58
	→ babonk joined	06:59
babonk	how do I copy the files from a repo without regard to history/cloning?	06:59
	like I just want the files	06:59
	at the head of the branch	06:59
	→ ValicekB joined	07:00
rjhunter	babonk: `git archive` if you want to put things outside the repo	07:01
babonk	rjhunter: is this what people traditionally use if they want to start a project from a template in a repo?	07:01
SamB	no, people don't do that	07:01
babonk	im not trying to get a zip or a tar	07:01
	just want the starting setup for a new project, pulled from a repo	07:01
	→ Raging_Hog joined	07:01
	← trusktr left	07:02
	← glebihan left	07:03
	← Nicholas_Rishel left	07:03
	→ stodan joined	07:05
	← PaulCapestany left	07:06
	→ GODDOG joined	07:07
	→ ewnd9 joined	07:08
	← daito left	07:10
	→ xrq` joined	07:12
	→ marthinal joined	07:14
	← pantsman left	07:15
	← Mez left	07:15
	← dyng left	07:18
	→ twittig joined	07:18
	→ dyng joined	07:19
	← GODDOG left	07:20
	← mishok13 left	07:21
	→ daito joined	07:21
	→ petey joined	07:22
	← Gamecubic_ left	07:23
	← chuckharmston left	07:23
	← joelrobichaud left	07:25
	← revyd left	07:26
	→ matanya joined	07:26
	→ Mez joined	07:27
	← petey left	07:27
	→ user258467 joined	07:27
	→ Vinnie1991 joined	07:27
	→ HelperW joined	07:29
	← Vinnie1991 left	07:30
	← thiago left	07:33
	→ j0ran joined	07:33
	← dyng left	07:34
	→ etcetera joined	07:36
	← Mez left	07:36
	→ dyng joined	07:37
	→ Hoksilato joined	07:40
	→ Mez joined	07:43
	← Mez left	07:43
	→ Mez joined	07:43
	→ nnyk_ joined	07:43
	← dyng left	07:43
	← yuwang_ left	07:46
	← j0ran left	07:46
	← bunbury left	07:46
	→ dyng joined	07:47
nnyk_	Hi, could someone give some pointers on a way to remove a blob with <hash> from my object database?	07:47
	← dyng left	07:47
	← etcetera left	07:47
dr_lepper	nnyk_: is it contained in a commit?	07:47
	→ yhager joined	07:49
	→ dyng joined	07:49
	→ lc0dd0cl joined	07:50
	→ arjun_ joined	07:50
	→ GeertV joined	07:51
	→ Tomylobo joined	07:51
	→ AzizLight joined	07:52
_ikke_	nnyk_: !xy	07:52
gitinfo	nnyk_: Woah, slow down for a bit. Are you sure that you need to jump to that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...	07:52
	→ cbreak-work joined	07:53
	→ mishok13 joined	07:53
	→ nnyk joined	07:54
	← nnyk_ left	07:54
	← Goplat left	07:54
	← Mez left	07:55
nnyk	dr_lepper: yes its committed already	07:55
	→ papegaaij joined	07:56
	← AzizLight left	07:56
dr_lepper	nnyk: then you'll have to remove it from the commit it has been introduced in and all the subsequent commits	07:56
	→ JStalin joined	07:57
	→ Cornelio_ joined	07:57
	→ xjiujiu_ joined	07:58
nnyk	dr_lepper: if i have the hash of the blob on the commits, what command removes it?	07:58
JStalin	hi, how can I remove one file from repository, so its not cloned everywhere (config file, with db access different on each machine)	07:58
	→ lbf joined	07:58
_ikke_	!config	07:59
gitinfo	[!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas	07:59
nnyk	JStalin: .gitignore	07:59
_ikke_	nnyk: You need to rewrite history	07:59
	→ jnavila joined	07:59
nnyk	?	07:59
	← endiruna left	07:59
	← Cornelio left	07:59
dr_lepper	nnyk: you would need some for of git filter-branch	07:59
	index-filter I believe	08:00
_ikke_	filter-branch	08:00
	→ teran joined	08:00
_ikke_	man git filter-branch	08:00
gitinfo	the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	08:00
	← xjiujiu left	08:01
	→ j0ran joined	08:01
	← teran left	08:01
	→ teran joined	08:01
nnyk	_ikke_: tried rewriting with rebase -i... didnt change anything	08:01
	→ sabgenton joined	08:01
	nnyk goes off to read on filter-branch	08:01
JStalin	nnyk: hey, thx!	08:01
	← JStalin left	08:01
nnyk	thanks dr_lepper _ikke_	08:01
	np	08:02
user258467	hi, I have a testing branch and a master, I checkout the testing and I want the testing reflect exactly the master so I do git merge master but here it tells there are conflict how do I tell git to merge so testing is overwrite?	08:02
	→ rendar joined	08:03
	→ psyl0n joined	08:03
sabgenton	speaking of : git filter-branch	08:04
	Cannot rewrite branch(es) with a dirty working directory.	08:04
	Git 1.7.1 says this after cp -r original/ copy/	08:04
	← Norm_ left	08:04
moritz	user258467: why have separate branches if they are supposed to look exactly he same?	08:04
sabgenton	do older versions of git detect inode change or something?	08:04
	→ h_kon joined	08:04
user258467	moritz, that is not the question, I would reset my testing to the master and when I want to test some feature I go to testing but for the moment the branch needs to reflect the master	08:05
sabgenton	a simple checkout correct the problem but what an earth would cause that? even rsync -a still fails!	08:05
moritz	user258467: then man git-reset	08:05
gitinfo	user258467: the git-reset manpage is available at http://jk.gs/git-reset.html	08:05
	→ Mez joined	08:06
	← Mez left	08:06
	→ Mez joined	08:06
	→ Vinnie1991 joined	08:07
ir7466	hello.... I have an android project. the bin/ folder is on github	08:07
	→ kjellski joined	08:07
ir7466	How do i remove the bin/ folder from github?	08:07
	← qknight left	08:07
	← psyl0n left	08:08
moritz	git rm bin/; git commit 'removed bin/'; git push	08:08
	→ pbrobst joined	08:08
	→ jceb_ joined	08:08
	← Norrin left	08:09
	← whaley left	08:09
sabgenton	basicly in git 1.7.1 for me copying the repo always results in a dirty copy , strangely	08:09
	← dserodio left	08:10
user258467	moritz ok thanks	08:10
	← beautifulmind left	08:10
ir7466	moritz, and then if I add bin/ to .gitignore it won't get re-added?	08:10
	← pekster left	08:10
	← Mapley left	08:10
moritz	ir7466: correct	08:10
	← milli left	08:11
	→ alorence joined	08:11
	→ harobed_ joined	08:11
ir7466	i'm sorry for the basic questions (as i've been using git mainly through UIs to date), but in git bash... in order to run the command you mentioned, how do I specify the repo?	08:11
	the repo that I want to remove bin/ from I mean	08:11
	→ thierryp joined	08:11
	→ pekster joined	08:12
	← parshap left	08:12
	→ lb1a joined	08:12
	→ parshap joined	08:12
	← lun_ left	08:12
	← variousnefarious left	08:12
	→ variousnefarious joined	08:12
	→ rob_w joined	08:13
	→ yuwang_ joined	08:13
	→ _christoph_ joined	08:13
	→ whaley joined	08:15
	→ dserodio joined	08:15
	→ Mapley joined	08:15
moritz	ir7466: first you clone the repo, then you cd into the repo	08:15
_christoph_	Hi, I am handling a svn directory via git-svn. Now the directory URL changed. I updated the URL in the svn-remote section of .git/config	08:15
	← Mez left	08:16
	→ milli joined	08:16
_christoph_	unfortunately I get the following error when executing git svn rebase now on master(synced to trunk before...):	08:16
ir7466	moritz; in order to commit this in one hit, can I just do git rm bin/; ........... then edit .gitignore in a text editor	08:16
	then commit and push?	08:16
_christoph_	Unable to determine upstream SVN information from working tree history	08:16
ir7466	without touching eclipse at all?	08:16
canton7	_christoph_, all sorts of pain. you can do a lot of history rewriting iirc, or clone afresh	08:16
	_christoph_, you may have noticed the git-svn-id: line in each commit message. well, yeah, that's causing the pain	08:17
	→ Mez joined	08:17
_christoph_	canton7: If I clone I lose all my local branches ... :/	08:17
canton7	_christoph_, do another git svn clone. add the old git-svn repo as a remote to the new one, and fetch your branches	08:17
ir7466	moritz: git rm bin/ ........ "not removing bin recursively without -r"	08:18
	is it safe to do that?	08:18
canton7	_christoph_, see https://git.wiki.kernel.org/index.php/GitSvnSwitch too	08:18
	← thierryp left	08:18
_christoph_	but they will then not be connected to the new trunk branch as you mention, that these svn-ids will change or?	08:18
ir7466	git rm -r bin/	08:18
	?	08:18
canton7	ir7466, that's a safety net to stop you accidentally deleting a whole folder when you only meant to delete a file	08:19
ir7466	is it safe to do though?	08:19
	← Zarthus left	08:19
	→ gua_ould joined	08:19
canton7	_christoph_, then you rebase those fetched branches back onto the new git-svn trunk, and everything should just work fine. iirc the git-svn-id lines only appear on commits which have been dcommitted	08:19
_christoph_	canton7: THX for the link	08:19
	→ Misan joined	08:20
	→ Gonzo89 joined	08:21
	← gua_ould left	08:22
	← xjiujiu_ left	08:22
ir7466	what is the git command where you can see the remove repo you are working on?	08:22
	is it git remote -v?	08:22
	→ gua_ould joined	08:22
moritz	ir7466: !tias	08:22
gitinfo	ir7466: 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/tias.html may help with git-specific TIAS.	08:22
	→ xjiujiu joined	08:23
	← Tomylobo left	08:23
Gonzo89	Is it normal that Git Bash (Windows) show a different Time then cmd.exe/local time? Git Bash date returns "Mon Dec 9 08:18:05 GMT 2013" instead 09:23:30 UTC+1 which is my local time.	08:24
	← pbrobst left	08:24
	→ chuckharmston joined	08:24
_ikke_	Gonzo89: It means that no timezone has been set aparently	08:25
canton7	a timezone which adds 5 minutes?	08:26
	hmm, that or those were copied 5 mins apart :P	08:26
_ikke_	I was merely commenting on the GMT vs UTC+1	08:26
Gonzo89	Ah sorry minutes are correct .. just my fault problem is just the our different	08:26
	didn't find a way to setup the timezone	08:27
	for git bash	08:27
	← kermit left	08:28
	→ mike_af joined	08:28
	→ lindenle joined	08:29
	← chuckharmston left	08:29
	← ir7466 left	08:29
	← Mez left	08:30
	→ dyng_ joined	08:30
	← dyng_ left	08:30
	← dyng left	08:30
	→ dyng_ joined	08:30
	→ Mez joined	08:31
	← jakob left	08:32
	→ jakob joined	08:32
	← jakob left	08:32
	→ jakob joined	08:33
	← jakob left	08:33
	→ johnflux__ joined	08:33
	→ jakob joined	08:33
	← jakob left	08:33
	← mike_af left	08:33
	→ SHyx0rmZ joined	08:33
	→ jakob joined	08:33
	← jakob left	08:33
	→ nettoweb joined	08:33
	→ jakob joined	08:33
	← jakob left	08:33
	→ jakob joined	08:34
	← jakob left	08:34
	→ jakob joined	08:34
	← jakob left	08:34
	→ jakob joined	08:34
	← jakob left	08:34
	→ jakob joined	08:34
	← jakob left	08:34
	← sohail left	08:34
	→ jakob joined	08:35
	← jakob left	08:35
	→ jakob joined	08:35
	← jakob left	08:35
	→ jakob joined	08:35
	← jakob left	08:35
	→ jakob joined	08:35
	← jakob left	08:35
	→ jakob joined	08:35
	← jakob left	08:35
	← Mez left	08:36
	→ jakob joined	08:36
	← jakob left	08:36
	→ Mez joined	08:36
	→ jakob joined	08:36
_ikke_	try export TZ='Europe/Amsterdam'	08:37
	Gonzo89: ^	08:37
Gonzo89	_ikke_: not working :(	08:38
	→ mlechner joined	08:38
	← nettoweb left	08:38
	→ gittish joined	08:39
_ikke_	http://code.google.com/p/msysgit/issues/detail?id=322	08:39
	→ mikecmpbll joined	08:39
_ikke_	aparently an msys issue	08:40
	← xjiujiu left	08:40
	← lindenle left	08:41
Gonzo89	_ikke_: thx	08:42
	_ikke_: good keyword found solution here: http://stackoverflow.com/questions/2563044/changing-timezone-in-msys	08:42
	→ xjiujiu joined	08:42
	← SeySayux_ left	08:42
Gonzo89	export TZ=CET-1CEST	08:42
	→ nettoweb joined	08:42
	← _christoph_ left	08:43
	← Imagineer2 left	08:43
	→ Plummer joined	08:44
	← SHyx0rmZ left	08:45
	← iliketu__ left	08:46
	→ dreiss joined	08:46
	→ dhruvasagar joined	08:46
_ikke_	Gonzo89: nice	08:47
	→ m0ltar joined	08:47
_ikke_	You have to place it in .bashrc or something to make it permanent	08:48
	← dhruvasa1ar left	08:48
	→ SeySayux joined	08:48
	← mikecmpbll left	08:49
	→ blast_hardcheese joined	08:49
	← mattcen left	08:49
	→ juril joined	08:49
Gonzo89	_ikke_: yes already did :)	08:50
	← deryni left	08:51
_ikke_	right	08:51
	→ delasteve joined	08:51
	→ VinceThePrince joined	08:52
	→ mattcen joined	08:52
	→ lavong joined	08:53
VinceThePrince	I want to upgrade oh my zsh but it keeps saying: annot pull with rebase: Your index contains uncommitted changes. So I did : git add . and git commit -a and still receiving the same message	08:53
	VinceThePrince → Macaveli	08:55
	→ Silmarilion joined	08:55
	→ timthelion joined	08:56
	← Macaveli left	08:57
	← kjellski left	08:57
	→ thierryp joined	08:58
	← nettoweb left	08:58
	→ aChatir joined	08:59
	← thierryp left	08:59
	← aChatir left	09:00
	→ koda joined	09:01
	→ mikecmpbll joined	09:02
	→ netiul joined	09:02
	← timthelion left	09:02
	→ dfanjul joined	09:03
	→ shruggar joined	09:03
	→ psoo joined	09:06
	→ iliketu__ joined	09:06
	← iliketu__ left	09:07
	→ dkpi joined	09:07
lbf	so could someone please try http://pastebin.com/aT4K7mv6 ? i get "fatal: Unable to create '.../index.lock.lock' No such file or directory" and i don't know if it's a bug in git or i'm just doing something wrong	09:08
	→ Octalot joined	09:08
	← hans_henrik left	09:10
	→ Macaveli joined	09:10
lbf	* i get ".../sub/.git/index.lock.lock", and indeed it doesn't exist, but why should it ? it's a submodule, .git is a file, not a directory	09:11
	← Macaveli left	09:11
_ikke_	lbf: Hmm, might be a bug where it doesn't follow the link	09:12
johnflux__	Hey all	09:12
_ikke_	but the double .lock.lock also doesn't seem correct	09:12
	→ hans_henrik joined	09:12
johnflux__	when I "git rebase origin/next" it says Failed to merge in the changes.	09:13
	etc	09:13
	but "git mergetool" says that there are no files to merge	09:13
	but git status says that I have an unmerged file	09:13
	→ TomyWork joined	09:13
lbf	_ikke_: it does look quite strange indeed ..	09:15
_ikke_	might want to take it to the !mailing_list	09:15
gitinfo	The mailing list can be reached via [email@hidden.address] You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist	09:15
lbf	_ikke_: yeah, i was thinking of doing that, just wanted to see if others can reproduce the problem first	09:16
	← harish left	09:17
	→ konadr joined	09:18
_ikke_	lbf: I don't get any error	09:18
lbf	_ikke_: k, git version ?	09:18
	→ somedude11101 joined	09:19
	→ halu joined	09:19
somedude11101	so my git has stopped working again...	09:19
johnflux__	okay, if I can't use git mergetool, how can I do this manually? git status tells me "both modified" some file - how can I see both modifications?	09:19
_ikke_	lbf: https://gist.github.com/Ikke/095f561302946247db59	09:19
	johnflux__: If you check the file, you'll see conflict markers like <<<<<<<<<<<	09:19
	← dmarkham left	09:20
lbf	_ikke_: thanks ! .. i'm guessing it's either msys specific, or only specific to certain git versions, or maybe just me then ..	09:20
somedude11101	there will also be HEAD	09:20
	and the branch names	09:21
	→ _Qman joined	09:21
somedude11101	search the file for head	09:21
	also if you use lets say an IDE like Eclipse that does Syntax checking you'll see errors as its well a syntax error	09:21
_ikke_	I have 1.8.4.2	09:21
	→ Kayra joined	09:22
	← nnyk left	09:22
	← veebers left	09:22
	← dkpi left	09:22
	→ matteo joined	09:22
lbf	_ikke_: yeah mine was 1.8.4 so it seems msys specific	09:22
	← fairuz left	09:22
	→ beautifulmind joined	09:22
	→ kewah joined	09:23
	→ petey joined	09:23
	→ WaReZ joined	09:23
somedude11101	so if I run git I get Git Has stopped working for Windows	09:24
	→ rethus joined	09:24
somedude11101	this has been happening several times recently	09:24
	I recently to fix it but I would really like to know why this happens	09:24
	it stops randomly	09:24
	→ nettoweb joined	09:25
somedude11101	I thought maybe AVG but I couldn't find the bitguard.dll that allegedly breaks msysgit	09:25
_ikke_	What version of git are you using for windows?	09:25
	→ chuckharmston joined	09:25
somedude11101	Also seems its broken cygwin too	09:25
_ikke_	cygwin	09:25
somedude11101	msysgit	09:25
	→ TLGreg joined	09:26
somedude11101	I installed cygwin in case it broke again because I have a huge fucking project and I don't have time for this shit	09:26
	→ hoschi joined	09:26
somedude11101	broke in powershell	09:26
	and cygwin is all kinds of fucked	09:26
	I'd reinstall Windows, but like I said, all kinds of fucked on this project and finals	09:26
	→ bsdbeard joined	09:27
	→ timthelion joined	09:27
	← petey left	09:28
	← somedude11101 left	09:28
bsdbeard	I want to backup some big local git repos from one HDD to another, any idea if it's faster to clone them or use rsync?	09:28
	← Kayra left	09:29
	← chuckharmston left	09:29
johnflux__	_ikke_: I don't have any conflict markers	09:30
	← dreiss left	09:31
	→ Sendoushi joined	09:31
_ikke_	bsdbeard: How big is the repo?	09:32
bsdbeard	cd src	09:32
	oops	09:32
	← hoschi left	09:32
	← Than left	09:33
bsdbeard	_ikke_, 870 MB	09:34
_ikke_	rsync just transfers each file as they currently are, git clone will create a packfile and transfer that.	09:35
	I think git clone will be faster (even if it has to create a packfile)	09:35
bsdbeard	_ikke_, alright, thanks for the info :)	09:36
osse	so equivalent to git gc; rsync? :)	09:37
_ikke_	yes, I guess so, though, rsync might transfer more (loose objects)	09:37
Nevik	would it be possible for git to tell rsync what to transfer? (and if so, would it then gain anything from it)	09:42
	hmmm	09:42
	→ threesome joined	09:43
_ikke_	Why not use git clone then?	09:43
Nevik	git could possibly create a bundle/archive to give rsync, on which it then could do resumable file transfer	09:43
	_ikke_: ^	09:43
	resumability	09:43
_ikke_	okay	09:43
	→ mf_ joined	09:43
_ikke_	You can use git bundle to create a bundle	09:43
bsdbeard	Nevik, arent clones resumable?	09:44
_ikke_	nope	09:44
	packfiles are created in memory	09:44
Nevik	resumable transfer (cloning/fetching/pushing) is something that would be useful for many users (esp when combining big repos with unreliable remotes)	09:44
_ikke_	There is a thread going on at the mailing list about that	09:44
Nevik	ah	09:44
	thought it might	09:45
bsdbeard	I'm kind of surprised it's not already implemented :o	09:45
Nevik	it's been brought up in #github	09:45
	basically whenever their fs cluster takes a crap	09:45
	bsdbeard: the majority of transfers seems to be reliable enough :P	09:45
	← sabgenton left	09:45
Nevik	i.e. atm cost outweighs (perceived) benefit	09:45
	of course, there's also details to straighten out about what/how to implement it in the first place	09:46
_ikke_	git creates a packfile on-the-fly with only the objects the remote is still missing	09:46
Nevik	indeed	09:46
_ikke_	And with objects that are reachable	09:46
	karupanerura → zz_karupanerura	09:46
	← jnavila left	09:47
Nevik	i dont know too much about how git interacts with its transport protocols, but would something like this work: git communicates with remote (e.g. via ssh) to negotiate what objects to pack, then create that pack file (writing it to disk) and invokes rsync (signaling the remote to do the same)	09:47
	then rsync takes over the transfer, using the open ssh pipe	09:47
moritz	sounds complicated	09:48
Nevik	just thinking out loud though :P	09:48
bsdbeard	Would you even need to write it to disk? You could use some RAM filesystem	09:48
moritz	first, the "pack file" could be just in-memory	09:48
Nevik	bsdbeard: but then it might be lost if the sending end dies	09:48
_ikke_	bsdbeard: yes, but that would be transparent to git	09:48
Nevik	(it might also be lost on the fs, but that's less likely)	09:48
moritz	second, since pack files are already compressed, there's not much use in using rsync	09:48
bsdbeard	Nevik, Keep track of how much content was sent every 100kb or so, then it won't matter	09:48
	like a journaling system	09:49
Nevik	moritz: we've already established that the main advantage of rsync is that it makes the transfer resumable	09:49
	bsdbeard: true enough, i guess	09:49
_ikke_	moritz: resumability I guess.	09:49
moritz	I'm not sure if that warrants the extra complexity	09:49
Nevik	as weve also mentioned	09:49
_ikke_	moritz: To some people it might be	09:49
bsdbeard	Guys you'll get so much street cred if you implement this	09:50
Nevik	atm enough people have reliable transfers, so it's not worth the effort (yet)	09:50
	moritz should read the backlog more thoroughly	09:50
Nevik	indeed ;)	09:50
	← daito left	09:51
	→ psyl0n joined	09:51
	→ daito joined	09:52
	→ FerreusR joined	09:52
	← tty3 left	09:52
	→ johnflux joined	09:52
	→ tty3 joined	09:52
	→ jkremser joined	09:52
	← jkremser left	09:52
	→ jkremser joined	09:52
	→ Dudalus joined	09:53
	← Silmarilion left	09:55
	→ irqq joined	09:55
	← daito left	09:55
	← johnflux__ left	09:56
	← psyl0n left	09:56
	→ jnavila joined	09:56
	← timthelion left	09:57
	→ achadwick joined	09:57
	← achadwick left	09:57
	← netj_ left	09:58
	→ netj joined	09:59
	← robinsmidsrod left	09:59
	→ robinsmidsrod joined	09:59
	← robinsmidsrod left	09:59
	← hyperair left	10:00
	→ Norm_ joined	10:01
	→ Norrin joined	10:01
	← Norrin left	10:01
	→ Norrin joined	10:01
	← Norrin left	10:01
	→ robinsmidsrod joined	10:01
	← weiyang left	10:02
	→ daito joined	10:04
	← zmansiv left	10:04
	← rjc left	10:05
	→ xlinkz0 joined	10:05
	← nick_h left	10:06
	→ nick_h joined	10:06
	← Spyro left	10:06
	← gua_ould left	10:06
	← drbean left	10:06
	→ achadwick joined	10:07
	→ jeffhung joined	10:08
	→ Huevos joined	10:08
	← hellome left	10:08
	← daito left	10:08
	→ eijk_ joined	10:08
	← Orbitrix left	10:09
	← el3ctron left	10:10
	→ Spyro joined	10:10
	← monokrome left	10:10
	→ rjc joined	10:10
	← so left	10:10
	→ gmac joined	10:11
	→ zmansiv joined	10:11
	→ mdiers joined	10:11
	→ raonyguimaraes joined	10:12
	→ lun_ joined	10:13
	→ drbean joined	10:13
johnflux	when I do "git log", can I show the full path or something? The trouble is that I'm recursively doing "git log" on lots of git repos, and there's no way to see which repo i'm currently on	10:13
	← johnkevinmbasco left	10:15
	→ timthelion joined	10:16
	→ psyl0n joined	10:16
_ikke_	johnflux: my PS1 shows the PWD on the right side	10:17
	← ewnd9 left	10:17
	→ mike_af joined	10:17
	← lun_ left	10:17
	← johnflux left	10:18
	→ johnflux joined	10:19
	← johnflux left	10:19
	→ johnflux joined	10:19
	← nettoweb left	10:19
	Keridos → Keridos_off	10:19
	→ kennon joined	10:20
	→ Chaser joined	10:20
	→ so joined	10:20
	← dyng_ left	10:21
	← FerreusR left	10:21
	← mike_af left	10:22
	→ monokrome joined	10:22
	← bsdbeard left	10:22
	→ jaseemabid joined	10:23
	→ c0ded joined	10:24
	→ mk_io joined	10:24
	→ FerreusR joined	10:25
	→ chuckharmston joined	10:25
	→ rafeca joined	10:26
	→ hyperair joined	10:26
	← Hoksilato left	10:27
	→ Silmarilion joined	10:27
	← mk_qi left	10:28
	→ abelsiqueira joined	10:28
xlinkz0	msysgit is pretty sweet, it even has support for vi mode	10:29
	← johnflux left	10:29
xlinkz0	but is it possible to configure it so that the cursor changes when in normal mode? :D	10:29
	→ johnflux_ joined	10:29
	← chuckharmston left	10:30
	→ leakybuckit joined	10:30
	← SebastianFlyte left	10:30
	→ bob14 joined	10:30
	→ pengunix joined	10:31
	→ thierryp joined	10:31
	← arjun_ left	10:33
	→ harish joined	10:34
	→ johnkevinmbasco joined	10:34
	→ YoungFrog joined	10:34
	← andschwa` left	10:37
moritz	isn't that a vim setting?	10:37
	or wait, does it use windows' dumb terminal?	10:37
	← Sylario_ left	10:39
	→ Sylario joined	10:39
	→ Kayra joined	10:40
	→ mk_qi joined	10:41
koda	sigh	10:43
	← Pullphinger left	10:43
koda	i did a mess with the line handling	10:43
	i have autocrlf = true	10:43
	in .gitconfig	10:43
	(which i didn't have before	10:44
	→ calmdev joined	10:44
koda	and after a commit	10:44
	all my unix script have ^M at the end	10:44
	and they do not work anymore	10:44
	→ trusktr joined	10:44
	← jaseemabid left	10:44
	← echo-area left	10:44
	→ SebastianFlyte joined	10:44
koda	what can i do to rectify this?	10:44
	→ vldmld joined	10:45
xlinkz0	moritz: it does use windows's default terminal emulator however that does support atleast two cursor types	10:45
	← Kayra left	10:45
	← Spyro left	10:45
	→ qdb joined	10:45
qdb	hi	10:45
xlinkz0	in vim atleas there's special ones for normal/insert	10:45
	Keridos_off → Keridos	10:45
	← luc_ left	10:46
	→ arjun_ joined	10:46
koda	also if i change it to 'false' now all files will be prompted to change	10:47
	→ Vampire0_ joined	10:48
qdb	i have made wrong commit. then i have rolled it back with windows github program. then i pressed sync. but it cannot do it , says open shell to fix it. then i have searched in google "git rollback commit" and opened http://stackoverflow.com/questions/927358/how-to-undo-the-last-git-commit and see "git revert HEAD" and try it in shell , it says to me :	10:48
	→ fairuz1 joined	10:48
	← fairuz1 left	10:48
	← bob14 left	10:49
qdb	error: Your local changes to the following files would be overwritten by merge: index2.php Please, commit your changes or stash them before you can merge. Aborting	10:49
	← harobed_ left	10:49
	← h_kon left	10:49
moritz	qdb: so, it already gives you instructions what to do. Follow them.	10:50
	← yhager left	10:50
	← Vampire0 left	10:50
qdb	moritz ok i ll try	10:50
	→ dyng joined	10:51
qdb	moritz: done , but my editor says that my file is changed . that is not big problem.	10:52
	→ harobed joined	10:53
	→ h_kon joined	10:54
	← dyng left	10:55
	→ dyng joined	10:55
qdb	it has been synced without errors	10:56
	← fandikurnia01 left	10:57
qdb	but i recommited rolled back commit with other name : "test", and revert has appeared in github site as "revert "test" "	10:57
	← subhojit777 left	10:58
qdb	are commits in github site are undeletable? only can be reverted ? not possible to silently delete ?	11:00
	← timthelion left	11:00
	→ sohail joined	11:00
	→ adiga joined	11:01
qdb	moritz	11:01
	→ nbrosnahan joined	11:01
c0ded	oh yea deff gonna need a crash course to remember all this shit	11:01
	been over a yrs ince i was dev/code working on projects with github	11:01
	← Vampire0_ left	11:02
qdb	and thank you	11:02
c0ded	lol forgot all the little stuff to do for commits and setting up reops and stuff	11:02
	← mikecmpbll left	11:02
c0ded	when its time im sure ill be asking for help too guys hehe :)	11:02
	→ mikecmpbll joined	11:02
qdb	i have created near 4 repos and every time i had read instruction , c0ded	11:03
	oh .. just "read" ... past simple ...	11:04
	nobody answers to me	11:04
	→ marhaban joined	11:04
qdb	i will not wait, will leave this commit and revert	11:05
	→ gain joined	11:05
	→ LimeBlast joined	11:05
nbrosnahan	qdb, what's the question?	11:05
	f8ld → f8l	11:06
	→ Vampire0 joined	11:06
qdb	i will search that in google ..	11:06
	nbrosnahan:	11:06
gain	hi all, when I checkout to a hash, how can I return to the last commit?	11:06
qdb	are commits in github site are undeletable? only can be reverted ? not possible to silently delete ?	11:06
gain	now I've scrolled my bash finding the last hash and git checkout <hash>... but if I haven't the last commit hash, how can I see it?	11:07
	→ davividal joined	11:07
nbrosnahan	Sorry, don't know.	11:07
	Might try a git rebase -i and see what you can expunge. do it on a test repo though.	11:08
	← SebastianFlyte left	11:09
	← nbrosnahan left	11:09
gain	maybe I've foud: git checkout branchname... it was so simple... :D	11:09
	→ qdb_ joined	11:09
	← trusktr left	11:10
	← h_kon left	11:10
gain	'cause git checkout <lasthash> say I'm not on any branch... so the correct way is git checkout branchname, I think..	11:10
	→ subhojit777 joined	11:11
	→ h_kon joined	11:11
	← qdb left	11:11
	qdb_ → qdb	11:11
qdb	thank you	11:11
	← dik_dak left	11:11
	→ AzizLight joined	11:16
qdb	are commits in github site undeletable? only can be reverted ? not possible to silently delete ?	11:16
	→ ewnd9 joined	11:18
	→ ChrisAnn joined	11:19
moritz	you can push with --force to get rid of the remote commits	11:19
qdb	thank you . i searched what is remote commit and found http://christoph.ruegg.name/blog/git-howto-revert-a-commit-already-pushed-to-a-remote-reposit.html	11:21
	← gittish left	11:22
	→ daito joined	11:24
	→ petey joined	11:24
	← abelsiqueira left	11:24
	→ andschwa` joined	11:25
	→ chuckharmston joined	11:25
	← petey left	11:28
	← daito left	11:28
	← _5kg left	11:28
	→ jackneill joined	11:28
	← chuckharmston left	11:30
	→ Advocation joined	11:30
	→ the_cybersphinx joined	11:31
	← marhaban left	11:31
	→ d8bhatta joined	11:31
	← Watabou left	11:32
	← dhruvasagar left	11:34
	← d8bhatta left	11:36
	→ dhruvasagar joined	11:37
	← impaktor left	11:39
	← subhojit777 left	11:39
	→ gittish joined	11:39
	→ zophy joined	11:40
	← h_kon left	11:40
	→ lun_ joined	11:40
	← lavong left	11:41
	← teran left	11:42
	← gittish left	11:42
	→ teran joined	11:42
	← xlinkz0 left	11:44
	← calmdev left	11:45
	← mmcdaris left	11:47
	→ mmcdaris joined	11:48
	← dhruvasagar left	11:50
	→ hs366 joined	11:50
	→ subhojit777 joined	11:50
qdb	i have managed to delete them	11:51
	rolled back in gui then git push origin master -f	11:51
	→ dhruvasagar joined	11:51
	← qdb left	11:52
	→ carbon joined	11:52
	→ oneeyerows joined	11:53
	→ zhost joined	11:53
	→ thevishy joined	11:54
	← andschwa` left	11:57
	→ Guest563 joined	11:57
	← Guest563 left	11:58
	→ SicoWork joined	11:58
	→ dort joined	11:59
Gonzo89	Looking for a solution to disable "LF will be replaced..." warnings I noticed there is no clarity of using "core.autocrlf=true" or not. Should I rather use ".gitattributes"? (Windows/Linux environment)	11:59
_ikke_	Gonzo89: !crlf	12:01
gitinfo	Gonzo89: to fix problems with line endings on different platforms, check out http://line-endings.gitrecipes.de/	12:01
	→ nettoweb joined	12:01
	→ timthelion joined	12:02
	→ tma996 joined	12:02
	← dort left	12:02
	← TomyWork left	12:02
	→ dort joined	12:03
Gonzo89	_ikke_: okay thx	12:03
	→ abelsiqueira joined	12:05
	→ mike_af joined	12:06
	← AzizLight left	12:06
	→ lavong joined	12:07
	← dort left	12:07
	← JEEB left	12:07
	← eletuchy_ left	12:07
	→ User458764 joined	12:08
	→ eletuchy joined	12:08
	→ xlinkz0 joined	12:08
	→ dort joined	12:08
	← eletuchy left	12:08
	→ eletuchy joined	12:08
	← nettoweb left	12:09
	→ JEEB joined	12:09
	← mike_af left	12:10
	← kewah left	12:11
	← xjiujiu left	12:13
	→ xjiujiu joined	12:14
	→ Guest563 joined	12:15
	→ fairuz joined	12:16
	→ TomyWork joined	12:16
	→ kewah joined	12:16
	← dort left	12:17
	→ vtunka joined	12:17
	← davividal left	12:18
	← cilly left	12:18
	→ dsdeiz_ joined	12:20
	← dsdeiz left	12:22
	← TomyWork left	12:22
	→ cilly joined	12:23
	→ joel_falcou joined	12:23
	← beautifulmind left	12:23
	→ blinkdesign joined	12:23
	→ dort joined	12:24
	← timthelion left	12:24
	→ Gamecubic joined	12:24
	← Guest15100 left	12:24
	→ chuckharmston joined	12:25
	← dsdeiz_ left	12:25
	→ davividal joined	12:26
	← arjun_ left	12:26
	← User458764 left	12:27
	← psyl0n left	12:28
	→ hellopat_ joined	12:29
	→ psyl0n joined	12:29
	→ DoniRS joined	12:30
	← chuckharmston left	12:30
	→ tvw joined	12:31
	← psyl0n left	12:31
	← cilly left	12:31
	← dort left	12:31
	→ psyl0n joined	12:31
	← hellopat left	12:32
	← xlinkz0 left	12:32
	→ ariscop joined	12:32
	← Fandekasp left	12:33
	→ Phillemann joined	12:34
	← hellopat_ left	12:34
Phillemann	I'd like to merge branch A into B, but ignore all changes in a specific file and keep B's version of the file.	12:34
	Any ideas on how to do that?	12:35
_ikke_	Phillemann: That's called an evil merge	12:36
	It's possible, but not really recommended	12:36
Phillemann	I understand.	12:37
	_ikke_ suspects !config is applicable	12:37
_ikke_	if gitinfo is cooperative	12:37
Phillemann	So the simplest thing would be to copy the file to a temporary location and copy it back. ;)	12:37
	→ xlinkz0 joined	12:37
YoungFrog	probably !config in PRIVMSG and not in ACTION.	12:37
gitinfo	[!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas	12:37
Phillemann	Luckily, it's just one file...	12:37
YoungFrog	Phillemann: I would merge as usual, then checkout B's version of the file and amend. Probably inefficient, though, but I'm too ignorant to know better.	12:38
_ikke_	YoungFrog: Nah, gitinfo just has lag	12:38
	Phillemann: You don't have to copy it, you can just checkout the file	12:38
YoungFrog	_ikke_: ok	12:38
_ikke_	YoungFrog: Or even make a separate commit for it, so everybody knows what happened	12:38
Phillemann	Hmmmmm	12:38
_ikke_	Phillemann: ^	12:38
	← teran left	12:38
YoungFrog	yeah, that makes more sense	12:39
	Keridos → Keridos_off	12:39
Phillemann	I suppose that's the best, yeah. Thanks. :)	12:39
YoungFrog	I can even s/more//	12:39
	→ nettoweb joined	12:39
	→ LionKing joined	12:39
	← sohail left	12:39
	→ johnflux__ joined	12:40
	← johnflux_ left	12:40
	← wrouesnel left	12:40
	→ TomyWork joined	12:42
	→ b3tamax joined	12:42
	→ xjiujiu_ joined	12:44
	← kewah left	12:45
	← vldmld left	12:45
	→ dorohoro joined	12:45
	→ arjun_ joined	12:47
	← xjiujiu left	12:47
	← dorohoro left	12:47
	→ kewah joined	12:47
	→ luser0 joined	12:49
	→ qknight joined	12:50
	→ dcpc007 joined	12:50
	← decimad left	12:51
	→ Fandekasp joined	12:51
	← luser0 left	12:52
	→ fRAiLtY- joined	12:52
	← ewnd9 left	12:52
	→ ewnd9 joined	12:54
	→ Spyro joined	12:55
	→ el3ctron_ joined	12:55
	→ zsolti joined	12:55
	→ decimad joined	12:56
	→ h_kon joined	12:57
	→ smashwilson joined	12:58
	← dyng left	12:59
	→ mintuz joined	13:00
	→ Etheco joined	13:00
	← ewnd9 left	13:01
fahadash	How do I add something to an existing repository which is already cloned at a local directory other than the directory that has stuff to add ?	13:01
	← mikecmpbll left	13:02
	← mintuz left	13:03
	→ Pullphinger joined	13:03
	→ mintuz joined	13:03
	← Cornelio_ left	13:04
	← halu left	13:04
	← TomyWork left	13:05
	→ txomon\|home joined	13:05
	← xlinkz0 left	13:05
	← HelperW left	13:06
_ikke_	fahadash: I don't understand your question	13:06
	← zumba_addict left	13:06
	→ tomreyn joined	13:07
fahadash	Me neither, I am hung over around this time :)	13:07
	→ xpe joined	13:07
fahadash	I have a repo cloned at my local drive at this location /home/fahadash/somelocation, now I have some files that I want to add to that repo- at /home/fahadash/path/to/newfiles	13:08
	← carbon left	13:08
	→ Gavrisimo joined	13:08
	← mintuz left	13:08
	→ carbon joined	13:09
LionKing	fahadash: and path/to/newfiles is not a subdirectory of somelocation?	13:09
	→ g3orge joined	13:09
	← YoungFrog left	13:09
fahadash	no	13:09
_ikke_	Just move those files there?	13:10
fahadash	We can't keep files at two places ?	13:10
_ikke_	no	13:10
fahadash	ok	13:10
LionKing	fahadash: If they do not have a common (grand/great) parent, no	13:11
	← xjiujiu_ left	13:12
	← markovh left	13:12
	← carbon left	13:12
	← Gamecubic left	13:13
	← GT4066 left	13:13
	← Guest563 left	13:13
	→ xjiujiu joined	13:13
	→ HelperW joined	13:13
	→ Cromulent joined	13:14
	→ fightback joined	13:14
	→ Gamecubic joined	13:14
Happzz	hi	13:15
	i have this in my post-receive: git log --pretty=%h -n 1 > /usr/local/www/.gitversion	13:16
	if i do it manually while being in the bare git repo, it works	13:16
	the post-receive hook doesn't seem to be doing it. how can i debug that?	13:16
	→ mintuz joined	13:16
	→ marhaban joined	13:17
	→ ewnd9 joined	13:17
	→ GODDOG joined	13:17
	→ Sonderblade joined	13:17
	→ echo-area joined	13:17
	→ cannonball joined	13:17
	→ carbon joined	13:17
	← Sonderblade left	13:17
	← Fandekasp left	13:17
	→ Sonderblade joined	13:18
	← _Vi left	13:18
	← Advocation left	13:20
	→ mudmaster joined	13:21
	→ xlinkz0 joined	13:21
	→ babonk_ joined	13:22
	← babonk_ left	13:22
	→ babonk_ joined	13:22
	← calcifea left	13:22
	→ breakingthings joined	13:22
	→ calcifea_ joined	13:22
	→ Stygia joined	13:22
Happzz	nvm it wasn't +x	13:23
	← Plummer left	13:23
	← babonk left	13:23
	← edwardly left	13:23
	← mudmaste_ left	13:23
	← h34d3r left	13:23
	→ dr_lepper joined	13:23
	→ midnightmagic_ joined	13:23
	babonk_ → babonk	13:23
	← mlechner left	13:23
	← arvind_khadri left	13:23
	← cehteh left	13:23
	→ mlechner joined	13:23
	→ cehteh joined	13:23
	→ arvind_khadri joined	13:23
	→ h34d3r- joined	13:24
	→ kent\n_ joined	13:24
	→ edwardly joined	13:24
	→ Someguy123 joined	13:25
	→ petey joined	13:25
	← Cromulent left	13:25
	→ chuckharmston joined	13:25
	← kent\n left	13:26
	→ D-Boy joined	13:27
	← GODDOG left	13:28
	← calcifea_ left	13:28
	← petey left	13:29
	→ smolyn joined	13:30
	→ markovh joined	13:30
	← chuckharmston left	13:30
	→ xlinkz0_ joined	13:30
	→ bunbury joined	13:31
	→ richgilbank joined	13:31
	← raonyguimaraes left	13:32
	← xlinkz0 left	13:33
smolyn	hello git experts-- i have a question about rebasing. I'm doing a "git rebase --onto newbase newbase topic". topic has been alive a long time. Say I have a file A in topic that does NOT exist in newbase (or the common base)... is there a way to have rebase just sort of ignore all the intermediate conflicts/steps? I find I'm having to spend a long time fixing conflicts I already fixed long ago. I just want it to ffwd to the version in HEAD of	13:33
	topic. There's no chance it's colliding with newbase since it's not in newbase.	13:33
	→ scientes joined	13:33
	→ YoungFrog joined	13:33
	← Acidnerd1 left	13:34
	→ Advocation joined	13:35
	→ xlinkz0 joined	13:35
	→ _5kg joined	13:36
	← subhojit777 left	13:36
	← acharlieh left	13:37
	← richgilbank left	13:37
	← xlinkz0_ left	13:38
	→ tmcmahon joined	13:38
	→ StarFire_ joined	13:39
	→ kevwil joined	13:39
	← Pyrogerg left	13:40
	← Gonzo89 left	13:41
	← LionKing left	13:41
	→ hyperboreean joined	13:41
	← johnkevinmbasco left	13:41
	→ Pyrogerg joined	13:42
	→ chuckharmston joined	13:43
	→ Guest26621 joined	13:43
	→ mikecmpbll joined	13:44
	→ austincheney_ joined	13:45
	→ maveneagle joined	13:48
	→ DolphinDream joined	13:48
	→ Joshy joined	13:49
Joshy	Guys, my names Josh. I released a MMORPG on the 5th called Helbreath. I would love your feedback on the website and the game its self >> http://hbserver.org	13:49
	← Joshy left	13:49
	→ _Vi joined	13:49
	← ewnd9 left	13:49
	→ mizu_no_oto joined	13:52
	← spudz_ left	13:52
	← fightback left	13:54
	→ gittish joined	13:54
	← Gavrisimo left	13:55
	→ mike_af joined	13:55
	← oneeyerows left	13:55
	← gittish left	13:55
	← karavanjo left	13:57
	→ d8bhatta joined	13:58
	← HelperW left	13:58
	← Advocation left	13:58
	→ spudz joined	13:58
	→ neo1691 joined	13:59
	← mike_af left	14:00
	→ dsdeiz joined	14:01
	→ renatow joined	14:01
neo1691	Hey guys i have got three videos of git. 1) talk by Scott Chacon 2) Talk by Jessica Kerr 3) Randall Schwartz which one should a beginner to git watch first.	14:01
	→ PaulCapestany joined	14:01
	→ hellopat joined	14:02
	→ decotii joined	14:03
	→ krasnus joined	14:03
	→ scampbell joined	14:04
	→ m0viefreak joined	14:06
	← nettoweb left	14:06
	→ dyng joined	14:06
	→ g105b joined	14:08
	→ tjbiddle joined	14:10
	→ gittish joined	14:11
	→ Cromulent joined	14:11
	→ Acidnerd joined	14:14
	Guest92191 → Caius	14:15
	→ waruii joined	14:15
	→ dAnjou joined	14:16
g105b	What command can I use to lose any files that shouldn't be present in the repository? As if I remove the whole directory and clone it again	14:16
spb	git clean	14:17
ciaranm	git clean	14:17
	→ jerius joined	14:17
	→ ustunozgur joined	14:17
	→ erichynds joined	14:17
	← mk_qi left	14:17
g105b	Nice one - was wrestling with git reset, didn't know clean	14:17
	← maveneagle left	14:18
waruii	I try to partition a repo with some 30k commits and planned to clone it and delete files from the respective partition with git-filter-branch ... however nothing seems to have changed ....	14:18
g105b	Hmm, I'm still getting `The following untracked working tree files would be overwritten by checkout`	14:18
dAnjou	hi, i have a string in my code that doesn't seem to be used anywhere but maybe it was used some time ago. can i search in the git "history" somehow?	14:19
	← Dudalus left	14:19
	← rafeca left	14:19
	→ swaechter joined	14:21
waruii	dAnjou: try `git log -S string`	14:21
	→ timthelion joined	14:21
waruii	anyone here has removed a lot of files from a repo?`	14:21
	→ sohail joined	14:21
	← neo1691 left	14:22
	→ rafeca joined	14:22
	← kevwil left	14:22
dAnjou	waruii: that gives only a log	14:22
waruii	dAnjou: what do you want instead?	14:23
dAnjou	the actual line and file	14:23
	→ liamkeily joined	14:24
	→ pzl joined	14:24
	→ j416_ joined	14:24
dAnjou	i guess this is it: http://stackoverflow.com/questions/9522852/how-can-i-search-a-word-in-git-repo-over-the-whole-history	14:24
	← j416 left	14:25
	→ FortySix2 joined	14:26
	← linduxed left	14:26
	→ dort joined	14:26
waruii	dAnjou: for me that outputs every tree containing that word. so if the string lived through five commits you'll get five lines of output	14:27
	← dhruvasagar left	14:27
	← dort left	14:28
	→ dort joined	14:28
	← d8bhatta left	14:28
dAnjou	that's alright	14:28
	→ Bitgod joined	14:28
waruii	in that case, there you go ;)	14:28
Bitgod	question (w/ bitbucket)	14:28
	i have a repo named qoffice_lite	14:29
	→ nettoweb joined	14:29
Bitgod	i want two versions, x86 and x64	14:29
	would those be considered branches?	14:29
	PerlPilot → PerlJam	14:29
	← m0ltar left	14:29
	→ gnoel joined	14:30
	← KillYourTV left	14:30
sqz	if I expect merge conflicts, how can I forcefully pull in a newer branch (bruteforce) ?	14:32
	as in it overwrites always with the newer branch files	14:32
	→ m0ltar joined	14:33
	← j0ran left	14:33
	→ KillYourTV joined	14:33
	→ HelperW joined	14:34
	← g3orge left	14:34
	→ Dougie187 joined	14:34
Bitgod	anyone???	14:36
	← Phillemann left	14:36
	→ dhruvasagar joined	14:36
	→ jasonroyle joined	14:37
moritz	Bitgod: the usually approach would be to keep them in the same branch	14:37
	→ calcifea_ joined	14:37
moritz	Bitgod: and decide at configure time what platform to build for	14:37
	because you'll probably have a lot of logic that is platform independent, and you want changes to that logic to go into both versions by default	14:38
	and branches aren't well suited for that	14:38
	→ ColKurtz joined	14:39
	→ Roots47 joined	14:41
	→ iamjarvo_ joined	14:41
grawity	sqz: merge with -X theirs	14:41
	sqz: or, if you want to discard your local history, reset --hard to the desired commit instead of merging	14:41
	← xlinkz0 left	14:42
	← b3tamax left	14:43
	→ Gavrisimo joined	14:43
	← Pyrogerg left	14:43
Bitgod	okay	14:43
	well i only do Windows	14:44
	← xjiujiu left	14:44
	Xiti` → Xiti	14:44
	← Gavrisimo left	14:44
	→ Kayra joined	14:44
	→ DolpheenDream joined	14:45
	← mk_io left	14:45
	← davidfic left	14:45
	→ acharlieh joined	14:45
	← g105b left	14:46
	→ g105b joined	14:47
	→ xlinkz0 joined	14:47
	→ bwreilly joined	14:47
	← iamjarvo_ left	14:48
	→ brotspinne joined	14:48
	← tmcmahon left	14:48
brotspinne	hello. is there a way to call `git for-each-ref` on a remote repo?	14:48
_ikke_	Nope	14:49
	brotspinne: There is ls-remote though	14:49
	calcifea_ → calcifea	14:49
brotspinne	_ikke_: I want to do something like this: git for-each-ref --sort=-taggerdate --format='%(refname:short)' 'refs/tags/someprefix-*' --count 1	14:50
	→ b3tamax joined	14:50
	→ dxdt joined	14:50
	→ neo1691 joined	14:50
	← nettoweb left	14:50
	→ brewmast1r joined	14:50
	← ColKurtz left	14:51
brotspinne	_ikke_: I want the tag or the commit the tag is pointing to with a tag prefix but sorted by tagger date	14:51
_ikke_	brotspinne: Like I said, for-each-ref is not possible over the git protocol, only ls-remote. If you want that, you need to use ssh	14:51
brotspinne	_ikke_: is this possible with ls-remote?	14:51
_ikke_	Nope, it can only show what refs there are	14:51
	← bwreilly left	14:52
brotspinne	ok	14:52
	→ MJaoune joined	14:52
brotspinne	thanks!	14:52
MJaoune	anyone familiar with this problem? fatal: unable to access '<repo url>': SSL certificate problem: self signed certificate in certificate chain	14:53
	→ mklappstuhl joined	14:53
brewmast1r	i just got some git weirdness: i had a clean working tree, then i checked out another branch, and the working tree is dirty, full of changes	14:53
	← neo1691 left	14:53
	→ g0bl1n joined	14:53
	→ snuffeluffegus joined	14:53
brewmast1r	i would expect the tree to be clean after a git checkout to a different branch	14:53
	← rchavik left	14:54
waruii	was the branch new?	14:54
brewmast1r	no	14:54
	← scampbell left	14:54
_ikke_	MJaoune: Isn't the message quite clear?	14:54
	→ dhruvasa1ar joined	14:54
MJaoune	_ikke_: umm no... not for me at least	14:55
	← epta left	14:55
MJaoune	_ikki_: I am using repo.or.cz btw	14:55
waruii	brewmast1r: i guess the files were ignored in your other branch and in the "new old" one they are not	14:55
_ikke_	MJaoune: They have a self-signed certificate	14:55
	→ diegoviola joined	14:56
MJaoune	_ikke_: i was able to push before, but after setting up a new git repo i now get this message	14:56
	i added my self to the admin list	14:56
brewmast1r	waruii: i wish that were the case	14:56
	← brotspinne left	14:56
MJaoune	to be able to push	14:56
_ikke_	MJaoune: This is not a git issue	14:56
	← dhruvasagar left	14:56
_ikke_	MJaoune: It's an https issue	14:56
	MJaoune: Did you use an https:// url?	14:57
MJaoune	_ikke_: Yes	14:57
	_ikke_: they made the https the read and write push url	14:57
brewmast1r	waruii: files that were on the branch that I was on are now showing up in the "new old" branch as "deleted:"	14:57
MJaoune	_ikke_: if i use http its read only	14:57
_ikke_	MJaoune: But they don't provide a good ssl certificate	14:58
	→ neo1691 joined	14:58
_ikke_	MJaoune: Not one automatically trusted by your system	14:58
MJaoune	_ikke_: ah i see	14:58
	_ikke_: how can i trust it then	14:58
	?	14:58
_ikke_	You have to add their root certificate to your list of trusted certificates, or you tell curl to not verify certificates comming from repo.or.cz	14:59
	← g105b left	15:00
MJaoune	here's their root cert http://repo.or.cz/rorcz_root_cert.pem	15:00
	→ cn28h joined	15:00
_ikke_	MJaoune: But the latter means someone can easily pretend they're repo.or.cz	15:00
MJaoune	i configured it correctly	15:00
_ikke_	How?	15:00
	← rafeca left	15:00
	→ g105b joined	15:00
MJaoune	git config --global http.https://repo.or.cz.sslCAInfo \	15:00
	$HOME/.ssh/rorcz_root_cert.pem	15:00
	→ aro_ joined	15:00
	← Vinnie1991 left	15:01
	← mika left	15:01
	← flippo left	15:01
	← aartist left	15:01
	← jkyle left	15:01
	← brainproxy left	15:01
	← Zeelot left	15:01
	← vocifer left	15:01
	← lacrosse left	15:01
	← awx_ left	15:01
	← sneakret_ left	15:01
	← da5id left	15:01
	← ESphynx left	15:01
	← shennyg left	15:01
	← tchan left	15:01
	← potatoe left	15:01
	← cods left	15:01
	← frogonwheels left	15:01
	← Gizmokid2005 left	15:01
	← iAsterisk left	15:01
	← jon- left	15:01
	← brewmast1r left	15:01
	← thegorn_ left	15:01
	← ribasushi left	15:01
	← burzum` left	15:01
	← ToothlessGear left	15:01
	← austincheney left	15:01
	← aro left	15:01
	← Kulrak left	15:01
	← Exodist left	15:01
	← maxorator left	15:01
	← dxdt left	15:01
	→ ESphynx joined	15:01
	← PaulCapestany left	15:01
	→ ToothlessGear joined	15:02
	→ thiago joined	15:02
	← ToothlessGear left	15:02
	→ ToothlessGear joined	15:02
MJaoune	_ikke_: still here?	15:02
	← dhruvasa1ar left	15:02
	← skorgon left	15:02
_ikke_	MJaoune: Yes, hold on	15:02
	→ lacrosse joined	15:03
	→ brainproxy joined	15:03
	→ mike_af joined	15:03
	→ potatoe joined	15:03
	→ rafeca joined	15:03
	← mizu_no_oto left	15:03
	← snuffeluffegus left	15:03
	→ calmdev joined	15:03
	→ austincheney joined	15:03
	→ jkyle joined	15:03
	→ frogonwheels joined	15:03
	→ cods joined	15:04
_ikke_	MJaoune: You have to convert the PEM file to CRT aparently	15:04
	→ tomgabbi joined	15:04
	← calmdev left	15:04
Hello71	wat	15:04
	→ Zeelot joined	15:04
_ikke_	openssl x509 -in rorcz_root_cert.pem -inform PEM -out rorcz_root_cert.crt -outform DER	15:04
	← shruggar left	15:05
MJaoune	_ikke_: Just fixed it	15:05
	_ikke_: I used the git config http.sslCert $HOME/certs/rorcz_test_user_1.pem	15:05
	→ mika joined	15:05
	→ brewmaster joined	15:05
MJaoune	instead of using git config --global http.https://repo.or.cz.sslCert \	15:06
	$HOME/certs/rorcz_test_user_1.pem	15:06
	_ikke_: Thanks for trying to help :) Really appreciated	15:06
	← g105b left	15:06
	→ staticshock joined	15:06
	← h_kon left	15:06
staticshock	anyone know how full-screen a diff in tig?	15:06
MJaoune	_ikke_: you actually gave me the hint that the problem was with the certificate fiels	15:06
	files*	15:06
	← apetresc left	15:06
	→ g105b joined	15:07
	→ Gizmokid2005 joined	15:07
grawity	staticshock: the 'maximize' key is O	15:07
	staticshock: see tigmanual(7)	15:07
staticshock	grawity: thanks!	15:07
_ikke_	MJaoune: Nope, can't really help you with that	15:07
	→ maxorator joined	15:07
	→ vocifer joined	15:07
	→ sneakret_ joined	15:07
	→ da5id joined	15:07
	→ iAsterisk joined	15:07
	→ jon- joined	15:07
	→ thegorn_ joined	15:07
	→ burzum` joined	15:07
	hubbard.freenode.net set mode: +v	15:07
	← mdiers left	15:07
M00sL0gB0t	ValueError: need more than 1 value to unpack (file "/usr/local/lib/python2.7/dist-packages/willie/coretasks.py", line 132, in track_modes)	15:07
	→ mdiers joined	15:07
staticshock	grawity: it doesn't actually show up in tigmanual	15:08
	grawity: but it works, so thanks!	15:08
	→ dvide joined	15:08
grawity	it does, right under "View Manipulation" (5th item)	15:08
	→ shennyg_ joined	15:08
	→ tchan joined	15:08
	← neo1691 left	15:08
staticshock	grawity: right, it does, but mine says that the key is "M"	15:08
	grawity: which does not work. "O" works.	15:09
	← robbyoconnor left	15:09
grawity	older version with a typo in the manual, maybe?	15:09
staticshock	i guess	15:09
	→ iamjarvo_ joined	15:09
	→ Kulrak joined	15:09
	→ neo1691 joined	15:09
	→ ribasushi joined	15:09
	← stodan left	15:09
	→ Norrin joined	15:09
	→ shruggar joined	15:09
	← hs366 left	15:10
	→ dragon3_ joined	15:10
	← dsdeiz left	15:10
	→ Exodist joined	15:10
brewmaster	waruii: "git diff mybranch origin/mybranch" returns nothing, but my local working tree has hundreds of changes	15:10
	i've never seen this before	15:11
	← chuckharmston left	15:11
	← tvw left	15:11
	← ShooterMG left	15:12
	→ quizzi joined	15:13
	→ chuckharmston joined	15:13
jast	brewmaster: you compared one committed version with another committed version. uncommitted changes don't play into that...	15:13
	→ ShooterMG joined	15:13
brewmaster	jast: OK, I also did a git clone to a separate directory	15:13
	→ bwreilly joined	15:14
	→ kenperkins joined	15:14
brewmaster	jast: and I did a diff -r -x .git on the clean checkout of my branch (with a clean working copy) vs. the filthy one i'm talking about	15:14
	and there's no difference	15:15
waruii	brewmaster: had a look at http://stackoverflow.com/questions/2691618/git-is-deleting-an-ignored-file-when-i-switch-branches ?	15:15
	← quizzi left	15:16
	← g0bl1n left	15:16
	← dyng left	15:16
	→ rubatdub joined	15:16
	→ dyng joined	15:16
	→ quizzi joined	15:16
	← ShooterMG left	15:16
brewmaster	waruii: there's no problem with ignored files	15:16
	→ flippo joined	15:16
	→ ErKa joined	15:16
	→ richgilbank joined	15:16
brewmaster	waruii: git status right now shows hundreds of new / deleted / modified files that aren't ignored in any way	15:17
	← takamichi left	15:17
	→ awx_ joined	15:17
brewmaster	waruii: about half are "to be committed", the other half "changed but not updated", and then a couple untracked files (that I don't ignore anywhere)	15:18
	→ Feelsenburg joined	15:18
	← Norm_ left	15:18
	→ krasnus_ joined	15:18
	← krasnus_ left	15:18
	→ maleknet joined	15:19
	← MJaoune left	15:19
brewmaster	waruii: you think it's safe to just wipe this local branch out and check out a fresh copy?	15:19
	i have no local changes for this branch	15:19
	→ takamichi joined	15:19
brewmaster	but i have other local branches that i need	15:19
waruii	if those changes don't mean anything to you, you can also just git reset --hard	15:20
	→ LuxuryMode joined	15:20
	← rob_w left	15:20
	← pzl left	15:21
	← krasnus left	15:22
	← neo1691 left	15:22
	← cff left	15:23
	→ neo1691 joined	15:23
	← dccc left	15:23
	→ fandikurnia01 joined	15:24
	→ shawndumas joined	15:24
	→ Advocation joined	15:24
	→ bamj0rb joined	15:24
	gitinfo set mode: +v	15:24
	→ Zarthus joined	15:25
	← papegaaij left	15:25
brewmaster	waruii: thanks	15:26
	← Guest26621 left	15:27
	→ danman joined	15:27
	← mdiers left	15:28
	→ johnnyfuchs joined	15:30
	← amcsi_work_ left	15:30
	→ amcsi_work joined	15:31
	← johnnyfuchs left	15:31
	← psyl0n left	15:31
	→ dxdt joined	15:31
	→ johnnyfuchs joined	15:32
	→ hholtmann joined	15:34
	← neo1691 left	15:35
	→ ojacobson joined	15:35
	→ lindenle joined	15:36
	→ dccc joined	15:36
	← fandikurnia01 left	15:36
	← Silmarilion left	15:36
	→ dhruvasagar joined	15:36
	→ oojacoboo joined	15:37
	← Kayra left	15:37
	→ Xiti` joined	15:37
	← b3tamax left	15:38
	← Xiti left	15:39
	→ greggroth joined	15:41
	← dhruvasagar left	15:43
	→ neo1691 joined	15:43
	← lindenle left	15:43
	← aberrant left	15:43
	→ saurabh joined	15:43
	saurabh → Guest5840	15:44
	→ phile joined	15:44
	← mikecmpbll left	15:44
	← graphitemaster left	15:44
	→ graphitemaster joined	15:44
	← mricon left	15:44
	→ Feelsenburg1 joined	15:47
	→ mmitchell joined	15:47
	← Feelsenburg left	15:47
	→ johnmilton joined	15:47
	← thierryp left	15:48
	→ Xiti joined	15:49
	← Xiti` left	15:49
	← neo1691 left	15:49
	← irqq left	15:51
	← hyperair left	15:51
	→ vak joined	15:52
vak	hi all	15:52
	← Entomo left	15:52
	← netsroht left	15:53
	→ mizu_no_oto joined	15:53
	← bbloom left	15:53
	→ bbloom joined	15:54
	← Xiti left	15:54
vak	'git branch -avv' gives me [origin/master: ahead 1] ... for my tracked branch, However 'git push origin' says there is nothing to push. Why do I see then this 'ahead 1' ?	15:55
ojacobson	'git push origin' doesn't do what you think it does :)	15:55
waruii	git pull might	15:55
	;)	15:55
ojacobson	the actual behaviour depends on the push.default config value, but the default will be to pairwise match branches locally and on origin, by name (not by tracking info)	15:55
	and push the matches	15:56
	which is very unlikely to be what you want	15:56
vak	ojacobson: waruii: i did 'git pull' as well	15:56
	→ netsroht joined	15:56
	← FortySix2 left	15:56
ojacobson	anyways, the formal answer is "you see 'ahead 1' because your local branch is set to track origin/master, and contains one commit not contained in origin/master"	15:57
	which is maybe not helpful	15:58
	but is at least true	15:58
vak	git pull ==> Already up-to-date. git push ==> Everything up-to-date. and still 'ahead 1'...	15:58
ojacobson	back up a step and talk about your goals	15:58
	rather than about what commands you're running	15:58
vak	ojacobson: thank you, got it ))	15:58
	← twittig left	15:59
	→ King_Hual joined	15:59
	← rjhunter left	16:00
	→ nettoweb joined	16:00
	← ariscop left	16:00
vak	ojacobson: the formal answer did answer my question )	16:00
	← chuckharmston left	16:00
	← waruii left	16:00
	→ ariscop joined	16:01
	→ dkipnis joined	16:01
	← takamichi left	16:01
	→ chuckharmston joined	16:01
dkipnis	quit	16:02
	← dkipnis left	16:02
	← dyng left	16:02
	→ takamichi joined	16:03
	→ rjhunter joined	16:03
	→ str joined	16:03
	← Raging_Hog left	16:04
	← tomgabbi left	16:04
	← swaechter left	16:04
	← jargon left	16:04
	← leakybuckit left	16:04
	← marlboromoo left	16:05
	← chuckharmston left	16:05
	← YoungFrog left	16:06
	← netj left	16:06
	→ netj joined	16:06
	→ FortySix2 joined	16:06
	→ aberrant joined	16:06
	← FortySix2 left	16:07
	→ ngw joined	16:07
	← netiul left	16:08
	→ Kayra joined	16:08
	← timthelion left	16:08
	→ g0bl1n joined	16:08
	→ ajm203 joined	16:09
	→ dkipnis joined	16:09
	← dkipnis left	16:10
	← ThomasLocke left	16:10
	→ koda_ joined	16:10
	→ chrisgeorge joined	16:10
	→ m8 joined	16:11
	← koda_ left	16:11
	← Kayra left	16:13
	← abelsiqueira left	16:13
	→ asedeno joined	16:13
	→ VictorCL joined	16:13
	← cbreak-work left	16:14
	→ dhruvasagar joined	16:15
	← mlechner left	16:15
	→ robbyoconnor joined	16:15
	→ shadok joined	16:18
	← shadok left	16:18
	→ shadok joined	16:18
	→ guampa joined	16:18
	← carbon left	16:18
	→ PeterBooker joined	16:18
	← tma996 left	16:19
	→ mikecmpbll joined	16:20
	→ carbon joined	16:20
	yourbestfriend → FlashCode`	16:22
	← arvind_khadri left	16:23
	→ FortySix2 joined	16:23
	FlashCode` → sam113101	16:23
	sam113101 → yourbestfriend	16:23
	← GeertV left	16:23
	yourbestfriend → yourbestfriend`	16:24
	→ GeertV joined	16:24
	→ zumba_addict joined	16:24
	← nwertman_ left	16:25
	← GodEater left	16:25
	← liamkeily left	16:25
	→ shinnya joined	16:26
	← h0bbit_ left	16:26
	← dxdt left	16:26
	yourbestfriend` → FlashCode`	16:27
	→ primitive joined	16:28
	→ CKyle_ joined	16:28
	FlashCode` → yourbestfriend	16:28
	← GeertV left	16:29
	→ GodEater joined	16:29
	← GodEater left	16:29
	→ GodEater joined	16:29
	→ lavaman joined	16:30
	← Cromulent left	16:30
	← DoniRS left	16:30
	← maleknet left	16:30
	→ etcetera joined	16:30
	← dragon3_ left	16:31
	← psoo left	16:31
	→ dxdt joined	16:31
	← jnavila left	16:32
	← juril left	16:32
	← HelperW left	16:34
	← StarFire_ left	16:34
	← jceb_ left	16:35
	→ ethlor joined	16:36
	← robbyoconnor left	16:37
	→ robbyoconnor joined	16:38
	→ b3tamax joined	16:38
	→ jnewt joined	16:39
	→ knuth joined	16:39
	→ mmitchel_ joined	16:41
	→ nwertman joined	16:42
jnewt	git stash pop -> conflict, both modified. how to get the stash version of the file in working directory?	16:43
	← eletuchy left	16:43
	→ replyjava joined	16:43
xlinkz0	any windows users?	16:44
	→ eletuchy joined	16:44
dr_lepper	jnewt: git checkout --theirs -- path/to/file	16:44
bamj0rb	jnewt: Perhaps you could do git checkout stash@{0} -- path/to/file ?	16:44
dr_lepper	or --ours	16:44
lov	jnewt: I'm not sure if there's a correct way, but you could always do git checkout stash@{0} -- pa- yeah what bamj0rb said	16:44
	← b3tamax left	16:44
lov	xlinkz0: there are probably a few, just ask the question.	16:45
	→ hyperair joined	16:45
	← mmitchell left	16:45
	← chrisgeorge left	16:45
	← mmitchel_ left	16:46
	→ eletuchy_ joined	16:46
	→ mmitchell joined	16:46
	→ Tomylobo joined	16:46
xlinkz0	i wanted to ask what should i put in my windows path to make other programs be able to use git	16:47
	but i figured out how to open other programs that need git from msysgit	16:47
Silex	hello, I'm using the classic gitflow develop/master paradigm, and I bumped into an issue I'm not sure how to solve: I have an old maintainance branch for some tagged release on master. I'm currently doing a topic branch on develop which changes would theorically also apply to this old branch	16:47
	← eletuchy left	16:47
	← Guest5840 left	16:47
jnewt	thanks	16:47
Silex	am I supposed to cherry-pick in taht situation? how could I structure it to avoid this?	16:47
	← replyjava left	16:47
	← john2x left	16:48
	→ chrisgeorge joined	16:48
	→ cff joined	16:48
	→ Search112 joined	16:48
	→ john2x joined	16:48
	→ abelsiqueira joined	16:49
dr_lepper	Silex: you could start developing it on top of your old maintenance branch	16:49
	then you could've just merged it into master	16:49
	→ tang^ joined	16:49
Silex	dr_lepper: well, that means developping on an old base, which is kind annoying	16:49
dr_lepper	if you want to backport something, you have to cherry-pick or rebase	16:49
Silex	but yeah, I guess I could do the topic branch on delelop and just rebase before mergin	16:49
	develop*	16:50
dr_lepper	I think those are the only two options	16:50
Silex	Alright. Thanks	16:50
dr_lepper	develop on master and backport, or develop on old code and merge upstream	16:50
	← Search112 left	16:50
	→ kesha joined	16:50
	← guampa left	16:50
Silex	"backport" could be cherry-pick or a crafty rebase --onto right before mergin to old branch/master	16:51
	to it looked like I developped on old code	16:51
	so it*	16:51
	CKyle_ → CKyle	16:51
Silex	I guess I'll just cherry-pick, this is really old code	16:51
	← maxorator left	16:51
	Silex doens't like cherry-picking a lot	16:51
dr_lepper	Silex: rebase is nothing more than an automated cherry-pick	16:52
	← JonnyNomad left	16:52
	pixelgremlins_ba → pixelgremlins	16:52
	← joel_falcou left	16:52
	← jon- left	16:53
	← john2x left	16:53
dr_lepper	Silex: generally, I would say that there is nothing wrong in rebasing/cherrypicking something to backport a fix or a feature to an old release	16:53
	← reg_ left	16:53
	← xlinkz0 left	16:53
	→ jon- joined	16:53
dr_lepper	it's not as if the history could be useful to you in the future	16:53
	qsusOFFcs → qsuscs	16:53
	→ reg_ joined	16:53
	← kewah left	16:54
	→ Vinnie19_ joined	16:54
	← robbyoconnor left	16:54
Silex	rebasing avoids the commit duplications	16:55
	but at the same time my tip is so far away from from this old branch it doens't make much sense to tie them	16:55
	→ PeterBoo joined	16:55
dr_lepper	exactly my point	16:56
	← ehmatthes left	16:56
	← kesha left	16:56
	→ b3tamax joined	16:56
	→ irqq joined	16:56
Silex	I'm brain damaged about duplication, I always find it wrong :)	16:57
	→ PaulCapestany joined	16:57
	→ v0n joined	16:57
	← CKyle left	16:58
	→ dkpi joined	16:58
	← Misan left	16:58
	← PeterBooker left	16:59
	← Pullphinger left	16:59
	→ kesha joined	16:59
	PeterBoo → PeterBooker	16:59
	→ lisak joined	16:59
	→ maxorator joined	16:59
lisak	hey, is there any problem with deleting corrupted (diverged) branch from upstream and pushing new one that I restored from other branch locally	17:01
ojacobson	lisak: it's equivalent to force-pushing your branch	17:01
	so !rewrite applies	17:01
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), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	17:01
lisak	and telling my colleagues that the have to reset --hard origin/thebranch	17:02
	← carbon left	17:02
	→ yaymukund joined	17:02
	← Gamecubic left	17:02
yaymukund	how do I checkout a local [branch] if it exists. if it doesn't, check out origin/[branch] and track it. is there a command that does that? git checkout -b seems to complain if there's no such local branch.	17:03
ojacobson	'git checkout branch'	17:04
	provided 'branch' is unique across remotes	17:04
	otherwise you have to spell out which remote you want	17:04
yaymukund	ah yeah, I'm going off that assumption for now.	17:04
	hmm, maybe i shouldn't.. heh.	17:04
	← ustunozgur left	17:06
Bitgod	what is the purpose of a branch?	17:07
	← alorence left	17:07
Bitgod	i thought it was for if you had multple versions of main repo	17:07
lisak	ojacobson, so deleting remote branch and pushing local is the same as force pushing ?	17:07
Bitgod	like 1.0, 2.0 etc	17:07
	→ Lekensteyn joined	17:07
yaymukund	To crush your enemies, see them driven before you, and to hear the lamentation of their...er, commits.	17:07
lisak	ojacobson, so if I want to restore remote branches test & production from my local production branch, I force push it to the remote production and test branch ?	17:08
	→ ehmatthes joined	17:08
lisak	and I'm done ?	17:08
	→ Kayra joined	17:08
yaymukund	Bitgod: you should read up on common git workflows. git is a tool. It's better to ask what branches can do rather than what they're for, because everyone uses it slightly differently. To answer your question-- that is one valid way to use branching, but not the only one.	17:09
	← kesha left	17:09
	← gain left	17:09
	← dhruvasagar left	17:09
Lekensteyn	Hi, I just lost an untracked file by running `git stash save -u somename` (git 1.8.5.1). Running `find .git -mmin -5 -ls` showed some blobs in .git/objects/xx/yyyyy..., can I look in one of these blobs? `git show xxyyyy....` says that it is a bad object	17:09
Bitgod	yaymukund	17:09
	what if i have 32-bit and 64-bit versions of my code	17:10
	say 1.0 has 32 and 64, so does 1.01 etc	17:10
	< new	17:10
	→ CKyle joined	17:10
moritz	Lekensteyn: how about just applying the stash?	17:10
	← jasonroyle left	17:10
moritz	Bitgod: don't use separate branches for that	17:10
	→ SocalNick joined	17:10
moritz	work with #ifdef's or whatever tools your language has for such jobs	17:11
Lekensteyn	moritz: `git stash list` does not show a file, so I am afraid to lose the file. I will copy the repo and test, thank you for the suggestion.	17:11
Bitgod	moritz, then whats a good way?	17:11
	→ clarkfischer joined	17:11
	→ teran joined	17:11
moritz	Bitgod: 18:11 < moritz> work with #ifdef's or whatever tools your language has for such jobs	17:11
	→ kesha joined	17:11
Lekensteyn	moritz: o wow, it returned! I almost got a heart attack	17:12
	→ maleknet joined	17:12
moritz	Lekensteyn: git stash list lists stashes, not files. Which explains why you didn't see the file in the list :-)	17:12
	→ EGreg joined	17:12
Bitgod	moritz, #ifdef for 32 vs 64?	17:12
Lekensteyn	oh, I meant `git stash show`	17:12
Bitgod	and if its 64bit, do different dlls?	17:12
Lekensteyn	or even `git show stash@{{0}`	17:12
	→ DrShoggoth joined	17:13
Lekensteyn	typo, but you get the point	17:13
moritz	Bitgod: for example, yes	17:13
	← Stygia left	17:13
	← Kayra left	17:13
	→ hoyang_ joined	17:13
Bitgod	so i have to include a 32bit dll and a 64bit dll of a lib?	17:13
	like sqlite	17:13
	in my install heh	17:13
	← arjun_ left	17:13
	← _Qman left	17:13
Lekensteyn	moritz: is it a bug that the untracked files are not shown or an unimplemented feature?	17:13
	← johnflux__ left	17:14
moritz	Lekensteyn: I have no idea	17:14
	→ Feelsenburg joined	17:14
	→ johnflux__ joined	17:14
	→ imachuchu joined	17:14
yaymukund	Bitgod: It sounds like you have two separate problems here -- 1. maintaining two code branches. ifdefs might be the better approach, idk. 2. versioning. I would use tags for this, because they shouldn't change once you've pushed them.	17:14
	← Feelsenburg1 left	17:14
ojacobson	"target platform" isn't something I'd diverge development over	17:15
	I don't think it justifies recording every change twice (once per platform :)	17:15
	→ lindenle joined	17:15
ojacobson	Using branches in git is completely unavoidable; it's how git handles communication between repositories	17:16
	your local 'master' is actually completely independent of origin's 'master', other than by convention	17:16
	there's nothing in git that makes them "the same branch"	17:16
	← kesha left	17:16
	← hoyang left	17:16
yaymukund	okay, lunch time for me!	17:17
lisak	ojacobson, pls can I just do : (healthy branch) $git push origin production --force (healthy branch) $git push origin test --force	17:18
	→ daito joined	17:18
lisak	to recover remote test & production from my local healthy branch ?	17:18
moritz	Lekensteyn: you can	17:18
	sorry, meant lisak	17:18
	→ dhruvasagar joined	17:18
	← calcifea left	17:19
	← maxorator left	17:19
	→ ncthom91 joined	17:19
	← [emmajane] left	17:19
	← LuxuryMode left	17:19
lisak	scared shitless	17:20
	→ calcifea joined	17:20
	→ rath joined	17:20
ojacobson	lisak: I can tell you what it'll do (so can the man page), but not whether it's the right thing to do	17:20
	← oojacoboo left	17:20
ojacobson	fortunately, if you know what git-push does, it's pretty clear how to recover from a bad-idea push	17:20
	→ ncthom91_ joined	17:20
ojacobson	also, keep in mind that rewriting history nearly _always_ means working with the other collaborators to recover their local changes	17:20
	('git reset --hard' is a terrible approach, since that obliterates any changes they may have been working on)	17:21
	→ CheckDavid joined	17:22
	→ LuxuryMode joined	17:22
	← thevishy left	17:22
	← daito left	17:22
lisak	I guess that push results in finding shared ancestor commit object and retrieving and appending the rest of the all objects on it	17:23
	← g105b left	17:23
	← fRAiLtY- left	17:23
	← koda left	17:24
	→ g105b joined	17:24
	← ncthom91 left	17:24
	→ [emmajane] joined	17:24
lisak	the remote branch is broken because there is a commit with the same content 200 times under a different HASH, because somebody was rebasing instead of merging	17:24
ojacobson	simpler	17:24
	push copies refs (branches) verbatim, uploading any objects necessary for the branch to make sense.	17:24
	By default it'll only copy a branch over top of an existing branch if the new branch has the old branch as an ancestor; --force disables that check and blindly copies.	17:25
	→ Pullphinger joined	17:25
	← KK4MGV left	17:25
	→ Nazca__ joined	17:25
	← scientes left	17:25
	← hahuang65 left	17:25
	→ maxorator joined	17:25
	→ LuxuryMo_ joined	17:26
lisak	meaning that if there is a --force flag, it copies right away	17:26
	I like it :-)	17:26
	→ KK4MGV joined	17:26
	→ kermit joined	17:26
	← gittish left	17:27
	← blinkdesign left	17:28
	← Nazzy left	17:28
	← LuxuryMode left	17:28
	→ blinkdesign joined	17:28
	→ User458764 joined	17:28
	Xack → Cave_Johnson	17:29
	→ BoyanYordanov joined	17:30
	← BoyanYordanov left	17:30
	← thiago left	17:30
	← dfanjul left	17:31
	← iamjarvo_ left	17:31
	← harobed left	17:31
	→ r00tWearsRedHats joined	17:31
	Cave_Johnson → CaveJohnson	17:31
	← etcetera left	17:31
	→ iamjarvo_ joined	17:31
	← mikecmpbll left	17:32
	→ blindingdawn joined	17:32
	← mishok13 left	17:32
	← FerreusR left	17:32
	← g0bl1n left	17:33
	← blinkdesign left	17:33
	← shruggar left	17:33
	→ etcetera joined	17:34
Baluse	Hello	17:34
	← Roots47 left	17:34
Baluse	i accidentaly commit and pushed to remote wihout staging any file apart one that had only whitespace changes.. Then i commited the correct one	17:34
	is it possible to remove this wrong commit ?	17:35
	← iamjarvo_ left	17:35
lisak	guys I got "Everything up-to-date" message	17:36
Lekensteyn	Baluse: so you now have a commit "base", a commit "whitespace" and a commit "correct" in your history?	17:36
	← shawndumas left	17:36
Baluse	yes something like that	17:37
Lekensteyn	Baluse: and "whitespace" is pushed to remote, and now you want to remove all changes made by "whitespace"? (or squash the changes with "correct"?)	17:37
lisak	ojacobson, so it obviously doesn't work that way	17:37
	→ chuckharmston joined	17:37
	← rath left	17:37
	→ Roots47 joined	17:37
Baluse	Lekensteyn dont know its complicated now i got confused :S	17:37
	← nwertman left	17:38
	Keridos_off → Keridos	17:38
	→ Kayra joined	17:38
	→ dfanjul joined	17:38
	→ infynyxx joined	17:38
Celelaptop	Once I resolved the merging conflicts, what should I do?	17:39
	I just add and commit?	17:39
Lekensteyn	Baluse: by "remove a commit", do you mean (1) remove ALL changes introduced by that commit, or (2) merge the whitespace commit in your "correct" commit	17:40
	→ nwertman joined	17:40
Baluse	i want to remove it as never happened	17:40
Lekensteyn	Celelaptop: `git status` should give you a hint.	17:40
Baluse	2 seems more safe	17:40
	→ _iron joined	17:40
Lekensteyn	Baluse: if you do not mind rewriting the history (i.e. you are the only user of the remote repo), then `git rebase -i HEAD~2` presents you two commits. Remove the line to drop the changes or use "squash" to merge the changes in one commit	17:41
	← CKyle left	17:41
	← User458764 left	17:41
Baluse	how to use squash ?	17:42
Lekensteyn	Baluse: read the instructions in the editor that shows up when you run that comman	17:42
	→ g0bl1n joined	17:42
Lekensteyn	command* (or look in the manual page of git-rebase)	17:42
	← Vinnie19_ left	17:42
Baluse	ok thanks..	17:43
	→ petey joined	17:43
	← shinnya left	17:43
	→ Bass10 joined	17:44
	← dfanjul left	17:44
Lekensteyn	moritz: the untracked files are available in the very non-obvious history: http://stackoverflow.com/a/12681856/427545. I think that I should pay attention to the third "Merge" commit in `git show stash@{0}`	17:44
	← FortySix2 left	17:44
	→ leeN joined	17:44
	→ AWinterman joined	17:44
	← mmcdaris left	17:45
	→ jvns joined	17:45
	→ FortySix2 joined	17:46
Celelaptop	Hum... I added a remote repository, I merged it into mine, and now I'd like to push it onto that other repository.	17:46
jvns	I'm trying to update the upstream URL of a submodule I have. I've changed it in .gitmodules and .git/config and run 'git submodule --sync' and 'git submodule update', but it isn't pulling my new changes	17:46
	→ jonatasnona joined	17:46
	→ rcu joined	17:46
Celelaptop	I don't understand what <repository> and <refspec> should be on the push command.	17:47
jvns	I'm using git version 1.8.4.4	17:47
	← Sonderblade left	17:47
	← SocalNick left	17:48
	← Octalot left	17:48
OMGOMG	or version "great disappointment" as it's commonly known	17:49
	→ Shusshu joined	17:49
Lekensteyn	Celelaptop: the <repository> is shown in `git remote show`. The <refspec> is the branch you want to push to the remote (with the same name).	17:49
	Celelaptop: if the name differs, you need to use: git push remote-name local-branch-name:remote-branch-name	17:50
	→ dmarkham joined	17:50
	← FortySix2 left	17:50
	← [emmajane] left	17:50
bamj0rb	Celelaptop: So we're sure that you understand what you did why don't you tell us how you did it. :)	17:52
	← Shusshu left	17:52
	→ SocalNick joined	17:52
jvns	OMGOMG: what do you mean?	17:53
	← mizu_no_oto left	17:53
	→ Stalkr joined	17:53
	← mueslo left	17:53
	← blindingdawn left	17:53
	← dAnjou left	17:53
	→ joelm joined	17:53
	← silviof1 left	17:54
	→ yhager joined	17:54
Stalkr	Are there any general guidelines for when you are supposed to use rebase over merge?	17:54
OMGOMG	http://en.wikipedia.org/wiki/Great_Disappointment	17:55
	→ [emmajane] joined	17:55
	← CheckDavid left	17:55
	→ dfanjul joined	17:55
	← dfanjul left	17:56
ojacobson	Stalkr: !rebase_vs_merge	17:57
	damn.	17:57
	Stalkr: !rebase	17:57
gitinfo	Stalkr: [!merge_or_rebase] For a detailed discussion of the dis-/advantages of merge and rebase, and when to best use which, see: https://coderwall.com/p/ny1hia -- and here's how Linus explains which one to use: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html	17:57
ojacobson	I was close!	17:57
Stalkr	:-P	17:57
	→ iamjarvo_ joined	17:57
	← maleknet left	17:57
Stalkr	So far I have been using merge for everything, but I want to take advantage of the powers of rebase	17:57
	→ radixo joined	17:57
radixo	hello guys.. we are migrating from svn to git.. and the people are very confortable in to use svn update without commit anything	17:58
	← iamjarvo_ left	17:58
radixo	someone knows how could we get the same effect.	17:58
Baluse	i did squash	17:58
	and then pulled and got problems	17:58
	now i have 4	17:58
	commits	17:59
	← PeterBooker left	17:59
radixo	or how to pull without commit all changes.	17:59
	→ PeterBooker joined	17:59
	→ iamjarvo_ joined	17:59
radixo	and how ..	17:59
	→ jhulten_ joined	17:59
	← eijk_ left	17:59
	→ aard_ joined	17:59
imachuchu	radixo: you can just do that (git's will !float the changes over), but why are you so commit adverse?	17:59
gitinfo	radixo: 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.	17:59
	← staticshock left	18:00
	→ mueslo joined	18:00
Celelaptop	Lekensteyn: thanks.	18:00
axrfnu	radixo: either float as abofe or stash away your changes, or even create a separate branch and afterwardss pull	18:00
	→ arietis joined	18:00
	← g105b left	18:00
axrfnu	s/abofe/above/	18:00
Celelaptop	bamj0rb: Why do you want to know how I did what I did?	18:01
radixo	the problem is that the file is different in every developer machine.	18:01
imachuchu	radixo: I'm confused, is this 4 commits past the merge, or 4 merge commits, or 4 commits without a merge?	18:01
Celelaptop	git remote add someone-gh https://someurl/foo.git	18:01
radixo	cause it has the conf from local paths..	18:01
Celelaptop	git fetch someone-gh	18:01
	← lun_ left	18:02
Celelaptop	git merge someone-gh/master	18:02
	→ devbug joined	18:02
	← lavong left	18:02
imachuchu	radixo: ahh, so what you really want is for git to ignore that file, man git-ignore	18:02
	→ revyd joined	18:02
axrfnu	radixo: second imachuchu. sounds like gitignore is the hammer to your nail ;)	18:03
	→ koda joined	18:03
	← petey left	18:03
	→ petey joined	18:04
	← petey left	18:04
radixo	i will read the manual to see..	18:05
	but i think it is not..	18:05
	let us see	18:05
	→ petey joined	18:05
	← KK4MGV left	18:05
	← abelsiqueira left	18:05
	← vocifer left	18:05
	→ hahuang65 joined	18:05
	← Vampire0 left	18:06
	→ KK4MGV joined	18:06
	→ infynyxx_ joined	18:07
	→ Vampire0 joined	18:07
koda	how do you reset origin?	18:07
	like when i do git rebase -i origin i'd like that git didn't pick stuff from the ancient stuff	18:07
	→ mmcdaris joined	18:07
	← petey left	18:08
osse	koda: specify another base: git rebase -i HEAD~5 or so	18:08
imachuchu	koda: is origin just a branch?	18:09
bamj0rb	Celelaptop: OK, then someone-gh would be your <remote>.	18:09
	← mintuz left	18:09
bamj0rb	Celelaptop: And most likely, master would be your <refspec> (but it is really up to you and the upstream team).	18:10
	← infynyxx left	18:10
	→ freestyl3r joined	18:10
radixo	guys..	18:10
	i think what i need is	18:10
	→ hoschi joined	18:10
radixo	git update-index --assume-unchanged file	18:10
	but it always returns fatal error..	18:10
	→ scientes joined	18:12
	← scientes left	18:12
	→ scientes joined	18:12
	→ SeptDePique joined	18:12
bamj0rb	radixo: There are different ways to handle that. I am not sure about that assume-unchanged bit, but I would guess that is local only and doesn't synchronize between repos.	18:13
	radixo: You could have a config.example file that is tracked, and force every developer to copy it to config and edit it.	18:13
koda	isn't origin just the first ref that you mark as start of the branch?	18:13
bamj0rb	koda: origin is just the default name for a remote repository.	18:14
osse	koda: there is no marker that indicates the start of a branch	18:14
	xrq` → xrq	18:14
	→ Noldorin joined	18:14
Stalkr	In a team of three we're working off a develop branch (using gitflow). When we introduce a new feature, we branch off develop and make our changes locally. We don't push those branches. Is this a case were you would rebase the new feature into develop instead of merge?	18:15
imachuchu	osse: there is, it's the root of the repo. Branches just share commits until they diverge and an interactive rebase just changes commits in the range you specify	18:15
Stalkr	develop is public, pushed to the repo. feature/something is local	18:15
	→ boblee joined	18:15
	→ thiago joined	18:15
	← mmcdaris left	18:16
osse	imachuchu: technically correct, the best kind of correct	18:16
imachuchu	sorry osse, I meant that for koda (you already know all that!)	18:17
axrfnu	radixo: i'd rather explitly ignore that file (maybe via .git/info/exclude), because after a year you'll be asking yourself which files you assumed to have not changed	18:17
osse	imachuchu: hehe. carry on.	18:17
	← Sendoushi left	18:17
	← jonatasnona left	18:17
	→ dbuch joined	18:18
	→ jonatasnona joined	18:18
	← nwertman left	18:18
	→ murtas joined	18:19
	→ dreiss joined	18:20
	→ nwertman joined	18:21
	→ rainbyte16 joined	18:21
	← EGreg left	18:22
	→ osmosis joined	18:22
	← rcu left	18:23
	→ pbrobst joined	18:24
	→ fostertheweb joined	18:24
	← smolyn left	18:25
	→ lite_ joined	18:25
	→ cebor joined	18:26
	→ pr0ton joined	18:26
bamj0rb	Oh, I see, he meant <upstream> in rebase, not origin..	18:26
	← bwreilly left	18:27
	← pbrobst left	18:28
	→ EGreg joined	18:28
	→ Imagineer joined	18:29
	→ SocalNick_ joined	18:29
	← SocalNick left	18:30
	→ LBV_User joined	18:30
	→ anjen joined	18:31
	← LuxuryMo_ left	18:31
	→ eijk_ joined	18:31
	← anjen left	18:32
	← Etheco left	18:32
	→ akiva joined	18:32
	→ SocalNick joined	18:33
	← fostertheweb left	18:33
	← SocalNick_ left	18:33
	← achadwick left	18:34
	← lite_ left	18:34
	→ thevishy joined	18:34
	← Roots47 left	18:35
	← murtas left	18:35
	→ murtas joined	18:36
	← str left	18:36
	← TLGreg left	18:36
	← murtas left	18:36
	→ Gamecubic joined	18:37
	→ boombatower joined	18:37
	→ Nicholas_Rishel joined	18:37
	→ TLGreg joined	18:38
	→ murtas joined	18:38
	→ mdhopkins joined	18:38
	← hoschi left	18:39
	→ nicedice joined	18:40
	← infynyxx_ left	18:40
	← mklappstuhl left	18:40
	← murtas left	18:41
	← cebor left	18:41
	→ infynyxx joined	18:41
	→ murtas joined	18:42
	← EGreg left	18:42
	← murtas left	18:42
	→ schemanic joined	18:42
	← infynyxx left	18:43
	→ cebor joined	18:43
	→ EGreg joined	18:44
	→ murtas joined	18:45
	← murtas left	18:45
	← rafeca left	18:45
	→ murtas joined	18:46
	← marhaban left	18:48
	← b3tamax left	18:48
	← greggroth left	18:48
	→ Roots47 joined	18:49
	→ psyl0n joined	18:49
	→ shawndumas joined	18:50
	← arietis left	18:50
	← Kayra left	18:52
	→ infynyxx joined	18:53
	← thevishy left	18:53
	← EGreg left	18:53
	→ Watabou joined	18:54
	→ zackiv31 joined	18:54
	→ pbrobst joined	18:54
	→ PeterBoo joined	18:55
	← PeterBoo left	18:55
	→ Silmarilion joined	18:55
	→ LuxuryMode joined	18:56
	→ sq-one joined	18:56
	→ b3tamax joined	18:56
	→ ebertx joined	18:57
	← scientes left	18:57
imachuchu	3	18:58
	→ hoschi joined	18:58
imachuchu	grr, sorry	18:58
	← ethlor left	18:58
	→ EGreg joined	18:58
	← lisak left	18:59
	← PeterBooker left	18:59
	→ ethlor joined	19:00
	→ DoniRS joined	19:01
	→ bwreilly joined	19:01
	← dreiss left	19:01
	→ austincheney__ joined	19:01
koda	found it, i had to you set-head!	19:02
	← PaulCapestany left	19:02
	→ dyng joined	19:03
	→ PaulCapestany joined	19:03
	→ marhaban joined	19:03
	→ greggroth joined	19:03
	← sohail left	19:04
	← austincheney_ left	19:04
	→ HG` joined	19:04
	→ lc0dd0cl2 joined	19:04
	← lc0dd0cl left	19:04
	← luxgeek left	19:05
koda	do not you	19:06
	→ Cromulent joined	19:06
	→ luxgeek joined	19:06
	→ sohail joined	19:06
	lc0dd0cl2 → lc0dd0cl	19:07
	→ blindingdawn joined	19:07
bamj0rb	o_O	19:07
osse	ø_Ø	19:07
	← dyng left	19:07
	← jkremser left	19:08
	→ WaffleCat joined	19:09
	← Gamecubic left	19:09
	← yaymukund left	19:10
	→ jkremser joined	19:11
	← jkremser left	19:11
	→ jkremser joined	19:11
	→ petey joined	19:11
	← b3tamax left	19:11
	← yhager left	19:12
	→ yaymukund joined	19:12
	→ oojacoboo joined	19:13
	→ yhager joined	19:14
	→ mizu_no_oto joined	19:14
	← dhruvasagar left	19:14
	→ thevishy joined	19:15
	→ b3tamax joined	19:15
	→ Gamecubic joined	19:16
	→ FortySix2 joined	19:16
	moritz tries to build the mirror image using U+20E5 COMBINING REVERSE SOLIDUS OVERLAY and fails	19:16
	← radixo left	19:16
	← matanya left	19:17
	← HG` left	19:18
	→ dreiss joined	19:18
	→ lite_ joined	19:18
	← Roots47 left	19:18
	→ Roots47 joined	19:20
	→ PeterBooker joined	19:21
	← gnoel left	19:21
	→ rafaqueque joined	19:21
	→ valgrind joined	19:21
	→ CKyle joined	19:21
bamj0rb	I think we should use up any unallocated code points in Unicode for lolcats.	19:21
	→ nnyk_ joined	19:22
	→ Kayra joined	19:23
	← mizu_no_oto left	19:23
	→ scientes joined	19:24
	← ethlor left	19:25
	← dmarkham left	19:25
	→ marhaban1 joined	19:26
	← haarg left	19:26
	← marhaban left	19:26
	← lite_ left	19:26
cbreak	😸	19:27
	😼🙀😿😻😽😺😾😹	19:27
comps	..	19:27
	← Kayra left	19:27
	→ dmarkham joined	19:27
cbreak	all hail apple's color emoticon font!	19:28
comps	a bunch of black question marks on white hexagon	19:28
cbreak	your font is insufficient :)	19:29
	← LuxuryMode left	19:30
comps	�	19:30
	← Droolio left	19:30
	→ carbon joined	19:30
	→ Pyrogerg joined	19:30
cbreak	But, U+1F63A to U+1F63E are all cats :P	19:30
	← etcetera left	19:30
	← xpe left	19:30
	← marhaban1 left	19:30
comps	hexdump says bfef 00bd	19:30
cbreak	that's utf8	19:31
	→ mizu_no_oto joined	19:31
	→ haarg joined	19:31
moritz	cbreak: bless you :-)	19:31
	→ marhaban joined	19:31
	→ obozhdi joined	19:31
comps	though I agree, DejaVu fonts are by no means complete ..	19:32
	← atomx left	19:32
	→ atomx joined	19:33
	→ LuxuryMode joined	19:33
	← nnyk_ left	19:34
	← Roots47 left	19:34
	← obozhdi left	19:35
	→ pantsman joined	19:35
	← pantsman left	19:35
	→ pantsman joined	19:35
cbreak	real fonts are quite expensive	19:36
bamj0rb	Ooooh... http://codepoints.net/emoticons	19:37
	I am planning to write a message board software... I wonder if I could just use those for emoticons.	19:37
	→ Roots47 joined	19:37
cbreak	yep :)	19:37
	you can.	19:37
	← davividal left	19:37
cbreak	but you rely on the User having an appropriate font	19:38
	(you can assume OS X to do that of course)	19:38
	→ alvaro_o joined	19:38
	→ nutron joined	19:38
bamj0rb	I suppose if there are any free ones then the site could host them.	19:39
	→ davividal joined	19:39
	← davividal left	19:39
bamj0rb	Though I plan to also interface gracefully with mail clients, and that will be more of a challenge. Could always be configurable whether to render as ASCII, Unicode, or graphical.	19:39
cbreak	it's simple text.	19:40
	← tirengarfio left	19:41
	→ halu joined	19:41
bamj0rb	Sure, but I wouldn't want users with insufficient fonts/terminals/etc. so be confused.	19:41
	← jerius left	19:41
	← Noldorin left	19:42
cbreak	they deserve it for not using a mac...	19:43
bamj0rb	(I'm rather annoyed that consoles in Linux do not support Unicode fully... At least, AFAIK)	19:43
Hello71	define console	19:43
	→ MacWinner joined	19:43
	→ veebers joined	19:44
	← b3tamax left	19:44
	← oojacoboo left	19:44
	→ Norm_ joined	19:44
	← marhaban left	19:44
	→ tirengarfio joined	19:44
bamj0rb	Well I mean the virtual consoles.	19:45
	→ marhaban joined	19:45
	← bbloom left	19:45
	← marhaban left	19:46
	→ uu1101 joined	19:46
	← psyl0n left	19:48
	→ mmitchel_ joined	19:49
	← lindenle left	19:49
	→ naomi_ joined	19:49
	→ b3tamax joined	19:49
	← vak left	19:50
	← koda left	19:50
	→ maestrojed joined	19:50
cbreak	my echo can echo emoticon cats. :)	19:50
	→ timthelion joined	19:50
	← jonatasnona left	19:50
cbreak	(in color even)	19:50
	← naomi left	19:50
cbreak	I did think unicode would be widely supported on other unixes too	19:51
	← teran left	19:51
	← maestrojed left	19:51
	→ jonatasnona joined	19:51
	→ nnyk_ joined	19:52
	← mmitchell left	19:52
	← nutron left	19:52
	← LimeBlast left	19:52
	← pixelgremlins left	19:53
	← harish left	19:53
	← mizu_no_oto left	19:54
	→ joshbetz joined	19:55
	← iamjarvo_ left	19:55
	← joshbetz left	19:56
	← Cromulent left	19:56
	→ iamjarvo_ joined	19:56
	← iamjarvo_ left	19:56
	← joki left	19:56
	→ iamjarvo_ joined	19:56
bamj0rb	Apparently it's possible, but I think it requires framebuffers to be enabled, and basically have to configure things yourself.	19:56
	← jkremser left	19:56
bamj0rb	Since Linux aims to be much more than a highly restricted proprietary desktop OS it obviously is not so easy for these things to Just Work(tm).	19:57
	← CKyle left	19:57
	← hahuang65 left	19:58
SamB	bamj0rb: yeah, textmode is only capable of displaying 512 distinct glyphs	19:58
	← PeterBooker left	19:59
	→ etcetera joined	19:59
SamB	and even THAT is only possible at the cost of having only 8 background colors available	19:59
	← matteo left	20:00
	→ PeterBooker joined	20:00
	← sohail left	20:00
	→ joki joined	20:01
SamB	though seemingly it's possible to fake textmode quite well in a raster mode; NO$GMB for DOS seems to, anyway	20:01
	← tirengarfio left	20:02
cbreak	funny that OS X doesn't have that problem	20:02
	→ Sound joined	20:02
	→ tirengarfio joined	20:02
SamB	textmode is a legacy hardware mode	20:02
	← JPohlman1 left	20:03
cbreak	funny that macintosh hardware doesn't have that problem then :D	20:03
SamB	the restrictions can be avoided by not using that mode	20:04
bamj0rb	It's not that it doesn't have that problem. It's that the OS doesn't let you access that mode. :P	20:04
	If you ever need that mode you will just have to send it to Apple with a blank cheque attached. ;)	20:04
cbreak	need... to be restricted?	20:05
	← PaulCapestany left	20:05
	→ beastd joined	20:05
bamj0rb	It's not about being restricted. It's about lower-level access. :)	20:06
	→ sohail joined	20:06
	← sohail left	20:06
	→ sohail joined	20:06
	→ PaulCapestany joined	20:06
	→ gnoel joined	20:06
SamB	like NO$GMB for DOS does: because textmode only supports displaying a 1bpp glyph in any given cell, but the gameboy needs 2bpp and the color gameboy needs even more, NO$GMB for DOS does a reasonable immitation of textmode in a raster mode ...	20:06
	→ andschwa` joined	20:06
	← DolpheenDream left	20:06
	← petey left	20:06
	→ impaktor joined	20:06
	→ LucasKA joined	20:07
cbreak	I tried low level access for my mac once	20:07
	→ petey joined	20:07
	→ maestrojed joined	20:07
cbreak	it required a different computer to attach KDB to the running kernel on the first computer.	20:07
SamB	cbreak: I suspect OS X just only has a framebuffer-based console (well, maybe serial too, but no textmode), however it's exposed to userspace	20:07
	cbreak: that's standard whenever you want to use a debugger on a kernel	20:08
	→ liamkeily joined	20:08
cbreak	the only serial ports macs have are USB, SATA, Ethernet, Firewire (sometimes) and HDMI. And of course Thunderbolt	20:08
	← rethus left	20:08
cbreak	SamB: yeah. It's very low level.	20:08
	it was not very productive	20:08
	I'd have had to use a debug kernel for more symbol information, which was too bothersome	20:09
	← devbug left	20:09
	Keridos → Keridos_off	20:09
SamB	there are some alternatives if you don't mind observing the kernel in-motion without being able to change anything, but these are obviously pretty unreliable in general since races ...	20:09
	← Keridos_off left	20:09
cbreak	my problem was OpenGL related kernel panics	20:10
	← diegoviola left	20:10
cbreak	I hoped to find out some details, but the only thing I did find was that it was probably a bug on apple's side...	20:10
SamB	I guess the main kernel debugging I've done was actually either under QEMU, or of NT	20:11
	→ oojacoboo joined	20:11
	← petey left	20:11
	→ troyswanson joined	20:11
	← infynyxx left	20:12
troyswanson	quick question, git friends. i need to reset the master branch back to before a pull request was merged, but i dont want to modify the branch that the pull request came from	20:12
	→ BBBThunda joined	20:12
	midnightmagic_ → midnightmagic	20:12
ToxicFrog	troyswanson: merges don't change the source branch, only the destination branch	20:13
troyswanson	i basically just want to undo the pull request merge	20:13
	ok cool, so i can do a git reset --hard [whatever]	20:13
	← Nicholas_Rishel left	20:13
bamj0rb	git checkout master && git reset <commit>.	20:13
ToxicFrog	The source branch was unaffected by the merge, and will continue to be unaffected if you rewind master.	20:13
SamB	(in the latter case, not only does one get annoyed about the sparse debugging symbols from MS, but also at the non-existant debugging symbols for 3rd-party (not-MS-compiled) drivers ...)	20:13
	→ teran joined	20:13
	→ petey joined	20:13
troyswanson	will i need to git push --force on master	20:14
	since history is being rewritten	20:14
bamj0rb	If you have already pushed the merge, yes.	20:14
troyswanson	ok cool.	20:14
	gonna try it	20:14
	very scared	20:14
bamj0rb	!backup and !rewrite	20:14
	gitinfo: Wake up. >:(	20:14
phy1729	I have my dotfiles at bitbucket currently; I want to move them to github. I want to keep the history but entirely purge the .gist-vim file first. How can I purge the file?	20:15
bamj0rb	git filter-branch?	20:15
	bamj0rb has never heard of a .gist-vim file.	20:15
gitinfo	Worried about your data while trying out stuff in your repo? To back up commit history on all branches/tags: `git clone --mirror`. To backup everything, including work tree and staging area: `tar cf repo-backup.tar repodir`. Or do your experiment in a throwaway clone instead. See also http://sethrobertson.github.com/GitBestPractices/#backups	20:16
	[!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), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	20:16
bamj0rb	Did somebody just poke him behind the scenes, or is he just lagging like a mofo? :)	20:16
	troyswanson: ^	20:17
troyswanson	bamj0rb: thanks	20:17
	← hoyang_ left	20:17
	← scientes left	20:18
phy1729	bamj0rb: it's for vim's Gist plugin; it has a token to paste gists to githun automatically under my username	20:18
	→ scientes joined	20:18
	← scientes left	20:18
	→ scientes joined	20:18
	→ jeffszusz joined	20:18
jeffszusz	how would i git stash changes to all files -except- two specific ones?	20:19
	← BBBThunda left	20:19
	← marthinal left	20:19
troyswanson	actually, it might make more sense to revert the individual commits present in the pull request. i won't be able to simply reset to a certain commit in the log	20:19
	← schemanic left	20:19
	→ BBBThunda joined	20:19
PerlJam	jeffszusz: -p maybe ?	20:19
	→ schemanic joined	20:20
jrnieder	jeffszusz: when I've wanted that kind of thing, I'd stage the changes to those two files for a new commit	20:21
	jeffszusz: and possibly even commit it, depending on the context	20:21
	jeffszusz: i.e., "git add file1 file2; git commit"	20:21
	jeffszusz: if I didn't want to commit for whatever reason, I'd use "git stash -k"	20:21
jeffszusz	oh keep index?	20:22
	nice	20:22
	thanks jrnieder	20:23
	← oojacoboo left	20:23
	← pr0ton left	20:23
	→ Kayra joined	20:23
jrnieder	sure thing	20:24
	I learned that trick from the "Testing partial commits" example in man git-stash	20:25
gitinfo	the git-stash manpage is available at http://jk.gs/git-stash.html	20:25
	→ anjen joined	20:25
	→ babylon9 joined	20:25
	→ oojacoboo joined	20:27
	→ joaoh82 joined	20:27
	→ lindenle joined	20:27
	← Kayra left	20:28
	← hoschi left	20:30
	→ pr0ton joined	20:30
	← anjen left	20:30
	← cebor left	20:30
majuscule	How can I checkout svn externals using with git-svn?	20:31
_ikke_	I think they aren't supported	20:31
	← lindenle left	20:32
	← schemanic left	20:32
	→ cebor joined	20:32
	← _Vi left	20:32
	← pr0ton left	20:32
	→ schemanic joined	20:33
majuscule	i have a dev sitting next to me who's done it but forgotten how, haha. i think they pretend to be submodules somehow?	20:33
jrnieder	majuscule: http://bugs.debian.org/619584	20:33
	→ pr0ton joined	20:33
	← jizanthapus left	20:33
jrnieder	majuscule: there's a 'git svn show-externals' script	20:33
	s/script/command/	20:34
	→ Nicholas_Rishel joined	20:34
	→ endiruna joined	20:34
majuscule	hmmmm, that's a step. thanks	20:35
	← teran left	20:35
	→ infynyxx joined	20:35
	→ aChatir joined	20:36
jrnieder	majuscule: here's the blog post linked to from there (since the bugreport suffered from linkrot): http://algorithm.com.au/blog/files/git-svn-svn-externals.html	20:36
	← quizzi left	20:36
	← joaoh82 left	20:37
	← cebor left	20:37
	→ CKyle joined	20:37
	← m0ltar left	20:37
	→ mdiers joined	20:37
	← mf_ left	20:37
	→ lite_ joined	20:37
	→ m0ltar joined	20:37
	← aChatir left	20:38
	← m0ltar left	20:38
majuscule	ahh, thank you very much!	20:38
	→ cebor joined	20:38
jeffszusz	lol	20:39
	why's the topic say "git-hg: don't you know that's poison?" :P	20:39
	← endiruna left	20:39
	← schemanic left	20:39
	→ endiruna joined	20:39
	→ mizu_no_oto joined	20:40
	← dreiss left	20:40
	→ Mrono joined	20:40
Mrono	In a multi user environment how do you control/enforce the names people commit with to prevent masquarding	20:41
lov	if that's an actual concern, signatures.	20:41
	→ mf_ joined	20:41
SamB	yeah, signatures are basically the only way	20:41
	← oojacoboo left	20:42
	← lite_ left	20:42
	← rasto left	20:42
bamj0rb	Would there be any way to enforce that with Gitolite/hooks?	20:42
	s,/,+,	20:42
	← petey left	20:42
SamB	I'm thinking you'd have a GPG keyring that had the keys to be accepted, and in one of the hooks you'd check all new commits against that keyring	20:43
	→ petey joined	20:43
	← johnnyfuchs left	20:44
SamB	and, if the commiter ID didn't match any identity listed in the keyring for a key that had been used to sign that commit, reject those commits	20:44
	of course this can't possibly prevent innacurate Author IDs	20:45
ojacobson	bamj0rb: of course hooks can forbid pushing new commits that don't claim they were authored by the pushing user	20:45
	but it's hard to do right and it kneecaps one of the reasons for using DSCM in the first place	20:45
	Since SCM is _not_ a replacement for communication I think that's probably the wrong place to "solve" authenticity problems	20:45
	if you need a mathematical proof of authenticity, insist on signed commits	20:46
	but you probably only need social proof, so fucking talk to the purported author	20:46
	← erichynds left	20:46
	← infynyxx left	20:46
	→ Noldorin joined	20:47
	→ chrisledet joined	20:47
SamB	ojacobson: yeah, the GPG approach suggested here is at least supposed to permit whatever distributed development model the approved committers want to use, as long as they don't try to merge commits from non-approved committers ...	20:47
	and anyway you can't do anything about Author because rebase, cherry-pick, outside contributors, etc.	20:47
	← r00tWearsRedHats left	20:48
	← petey left	20:48
SamB	of course, the keys used shouldn't be used for more important communication	20:48
	→ abelsiqueira joined	20:48
SamB	because they will be vulnerable to compromise due to the constant use	20:48
	→ hahuang65 joined	20:49
	← shawndumas left	20:49
	← jonatasnona left	20:50
	→ bartman` joined	20:50
	→ jonatasnona joined	20:50
	← _iron left	20:50
	→ harish joined	20:50
	← SocalNick left	20:51
	→ shawndumas joined	20:52
	← rainbyte16 left	20:52
	→ phipes joined	20:52
	← chrisledet left	20:53
	→ ethlor joined	20:53
	→ chrisledet joined	20:53
	← mmitchel_ left	20:53
	→ SHyx0rmZ joined	20:54
	→ atomx_ joined	20:54
	→ Fandekasp joined	20:54
	→ jerius joined	20:54
	← phipes left	20:54
	→ mmitchell joined	20:54
	→ phipes joined	20:54
	← phipes left	20:55
	← atomx left	20:55
	→ rasto joined	20:55
	→ phipes joined	20:55
	← cebor left	20:56
	→ cebor joined	20:56
	→ robtow1 joined	20:56
	← PeterBooker left	20:57
Mrono	good news guys, due to a regime change my plans to shift us from svn to git are making headway once again	20:57
	→ mmitchel_ joined	20:58
	→ guns joined	20:58
bamj0rb	Mostly good for you, but yes.	20:58
	→ r00tWearsRedHats joined	21:00
	→ tpollard joined	21:00
	← LBV_User left	21:00
	← thegorn_ left	21:01
	→ thegorn joined	21:01
	← impaktor left	21:01
	← jonatasnona left	21:01
	← troyswanson left	21:01
	← mmitchell left	21:02
	→ jonatasnona joined	21:02
	← mmitchel_ left	21:02
	← scientes left	21:02
	← Sound left	21:02
	← jonatasnona left	21:03
	→ jonatasnona joined	21:04
	→ pixelgremlins joined	21:04
	← pr0ton left	21:04
SamB	hmm, given a list of refs, is there a convenient way to find out the sizes of the objects they refer to?	21:04
	← blindingdawn left	21:04
	← johnflux__ left	21:04
	→ blindingdawn joined	21:05
	← robtow1 left	21:05
SamB	(would it make a difference if this list of refs was, in fact, "all of them"?)	21:05
	→ Norm__ joined	21:06
	→ CheckDavid joined	21:06
	← jonatasnona left	21:07
	← tdmackey left	21:07
	→ jonatasnona joined	21:07
	← chrisledet left	21:07
	→ raonyguimaraes joined	21:07
	→ pr0ton joined	21:07
	← Norm_ left	21:08
	← jonatasnona left	21:09
	← dcpc007 left	21:09
	→ diegoviola joined	21:09
	← pr0ton left	21:09
	→ jonatasnona joined	21:09
	→ pr0ton joined	21:10
	→ dcpc007 joined	21:10
	← rubatdub left	21:11
	← jonatasnona left	21:11
	→ SocalNick joined	21:11
	→ jonatasnona joined	21:11
	← jonatasnona left	21:13
	→ jonatasnona joined	21:13
	← thegorn left	21:14
	→ jarr0dsz joined	21:14
	→ thegorn joined	21:14
	← jonatasnona left	21:15
	→ jonatasnona joined	21:15
	→ johnnyfuchs joined	21:15
	← johnnyfuchs left	21:15
	← aard_ left	21:16
	← jonatasnona left	21:16
	→ jonatasnona joined	21:17
	← jonatasnona left	21:18
	→ _Vi joined	21:20
	← AWinterman left	21:20
	→ ewnd9 joined	21:21
	← FortySix2 left	21:22
	→ FortySix2 joined	21:23
	→ Kayra joined	21:23
	← greggroth left	21:23
	→ infynyxx joined	21:23
	→ skorgon joined	21:24
	→ lindenle joined	21:26
	→ bbloom joined	21:26
	→ scientes joined	21:26
	← scientes left	21:26
	→ scientes joined	21:26
	← harish left	21:27
EugeneKay	Convenient? No.	21:27
	As for Author IDs stuff above(I'm too lazy to actually read through it, but hey....) Gerrit has a (annoying) system which tries to make sure youre using the right email address for submitted patches	21:27
	← babylon9 left	21:28
	← Kayra left	21:28
	← cannonball left	21:29
	← Imagineer left	21:30
	← cn28h left	21:30
	← andschwa` left	21:31
	→ PeterBooker joined	21:31
	← b3tamax left	21:31
	← jackneill left	21:32
	← abelsiqueira left	21:32
	← eijk_ left	21:34
	← tjbiddle left	21:35
	← irqq left	21:35
	→ Naoe-Kanno joined	21:36
	→ fostertheweb joined	21:36
	← SHyx0rmZ left	21:38
	→ schemanic joined	21:39
	← pr0ton left	21:40
	← SocalNick left	21:41
	→ greggroth joined	21:42
	← vtunka left	21:42
	→ tidu12 joined	21:43
	→ cannonball joined	21:43
	→ SocalNick joined	21:43
	← Advocation left	21:44
	← phipes left	21:44
	→ cff_ joined	21:44
	→ AWinterman joined	21:44
	← sq-one left	21:44
	← tidu12 left	21:45
	→ fluxdude joined	21:45
	→ tidu12 joined	21:48
	← fostertheweb left	21:50
	← FortySix2 left	21:50
	← rendar left	21:50
	→ pr0ton joined	21:51
	→ FortySix2 joined	21:51
	← EGreg left	21:51
	← gnoel left	21:52
	← CKyle left	21:53
	yourbestfriend → YOURBESTFRIEND	21:54
	← alezost left	21:55
	→ CKyle joined	21:55
	← adiga left	21:56
	→ SocalNick_ joined	21:57
	→ neoian2 joined	21:58
	← iamjarvo_ left	21:59
	← PeterBooker left	21:59
	← SocalNick_ left	21:59
	→ iamjarvo_ joined	21:59
	→ dreiss joined	21:59
	← SocalNick left	21:59
jgornick	Hey guys, if I'm in the master branch, is it possible to merge into a develop branch without having to checkout the develop branch and then merge the master branch in?	22:00
	← pr0ton left	22:00
	→ SocalNick joined	22:00
	→ petey joined	22:01
	→ PeterBooker joined	22:01
	← SocalNick left	22:01
bamj0rb	jgornick: No. Merging always happens to the working tree.	22:01
jgornick	bamj0rb: Ok very cool :)	22:02
	→ SocalNick joined	22:02
bamj0rb	jgornick: You could clone elsewhere, merge, push, and rm -fR, but it's probably not worth it.	22:02
jgornick	Yeah, I'll checkout instead :)	22:02
	← leprechau left	22:02
bamj0rb	jgornick: You could perhaps use GIT_WORK_TREE to convince Git to do the merge elsewhere, but again probably isn't worth the mess and cleanup.	22:03
jgornick	Yeah, doing git checkout develop && git merge master is just fine :)	22:03
	← iamjarvo_ left	22:03
	← mizu_no_oto left	22:04
	→ leprechau joined	22:04
	← atomx_ left	22:04
	← CKyle left	22:04
	→ psyl0n joined	22:05
	→ atomx joined	22:05
	← shawndumas left	22:05
	← psyl0n left	22:06
	← atomx left	22:07
	→ atomx joined	22:08
	jcarouth` → jcarouth	22:09
	← leprechau left	22:09
	→ pr0ton joined	22:09
	← Tomylobo left	22:09
	→ aantix joined	22:10
	→ fobius joined	22:10
	← m8 left	22:10
	← dort left	22:10
	← CodeShark left	22:11
	← thevishy left	22:11
	→ schemanic_ joined	22:11
	→ marhaban joined	22:12
	→ schemanic__ joined	22:13
	← schemanic left	22:13
	→ psyl0n joined	22:14
	← schemanic_ left	22:16
	→ Feelsenburg1 joined	22:16
	← Pullphinger left	22:17
	→ CKyle joined	22:17
	← Feelsenburg left	22:18
	→ gittish joined	22:19
	← cannonball left	22:19
	← hellopat left	22:21
	← Nicholas_Rishel left	22:21
	← LuxuryMode left	22:23
	← jeffszusz left	22:24
	← mdiers left	22:26
	s0beit → s0beit\|ZNC	22:29
	→ gunmuc joined	22:29
	→ Kayra joined	22:31
	← Kayra left	22:31
	→ Kayra joined	22:32
	← psyl0n left	22:32
	← petey left	22:32
	→ johnflux__ joined	22:32
	← smashwilson left	22:32
	← johnflux__ left	22:32
	→ johnflux__ joined	22:32
	→ AaronMT joined	22:33
	← Roots47 left	22:33
	→ Technodrome joined	22:35
Technodrome	i've been stuck in svn land for so long	22:35
	→ decimad2 joined	22:35
	→ BlakeRG joined	22:35
	→ AWinterm_ joined	22:36
BlakeRG	i am having an issue i've never seen happen with git before. As soon as i check out a branch two files appear to be modified and i try to reset them nothing happens. If i remove them from the filesystem and resotre via git they still appear to be modified. When i look at the diff the entire file is modified.. any ideas?	22:37
	← FortySix2 left	22:37
	→ Roots47 joined	22:37
bamj0rb	Technodrome: Whe one of your "bros" says to check out some girl your response is "I don't have time for that!" ?	22:37
	← decimad left	22:37
bamj0rb	BlakeRG: EOL issues. !eol	22:37
gitinfo	BlakeRG: The option/attribute to use when you need to care about different line-endings in your file: http://timclem.wordpress.com/2012/03/01/mind-the-end-of-your-line/	22:37
	→ Nicholas_Rishel joined	22:38
bamj0rb	Hmmm, what about !crlf...	22:38
gitinfo	to fix problems with line endings on different platforms, check out http://line-endings.gitrecipes.de/	22:38
	→ mizu_no_oto joined	22:38
	← AWinterman left	22:40
	→ andrei_aru joined	22:40
	→ cromestant joined	22:40
	→ karavanjo joined	22:41
	← andrei_aru left	22:41
	→ Strato joined	22:41
	← Strato left	22:41
	→ FortySix2 joined	22:41
fobius	Is there any difference between running git merge --squash and git merge --squash --no-commit?	22:41
tang^	yes	22:42
	← nettoweb left	22:42
fobius	What?	22:42
tang^	the latter doesn't commit automatically	22:42
fobius	Neither does an automatic commit for me	22:42
	→ Strato joined	22:43
tang^	well, I stand corrected	22:43
	the documentation does say that --squash does not commit the merge	22:44
fobius	I think they do the same thing. I was just wondering because pro git has a section where they do git merge -s subtree --squash --no-commit somebranch	22:44
cromestant	hello everyone, I have a repo based on the atalassian git worlflow we have a master branch and a develop branch, aswel as other feature branches, I want to merge a feature branch into develop, but am getting weird behavior. Just to be clear the origin repo was created as a bare repo. so on a new directory I clone the repo (git clone user@server:path/to/repo.git) then I checkout the develop branch (git checkout -b develop), then git status shows nothing	22:44
	be updated and git log shows a month old commit. so I try and do a git pull origin develop and I immediately get a vi prompt as a commit message editor. the first line states: Merge branch 'develop' of server:/path/to/repo.git develop into develop. what am I doing wrong here?	22:44
	→ andschwa` joined	22:44
	← Dougie187 left	22:45
fobius	But I'm finding pro git to be increasingly unreliable since diff-tree does not do what they said it does either.	22:45
	← danman left	22:46
fobius	cromestant: What are you confused about? It sounds like it's doing an automatic merge and asking you to commit the result.	22:48
	← dark_splinter left	22:48
bamj0rb	cromestant: git pull is actually git fetch and git merge. So it is trying to merge the branch.	22:48
	→ rhombus joined	22:48
bamj0rb	cromestant: If you are on master then git checkout -b develop will create a local branch based on origin/master, NOT origin/develop.	22:49
	→ dlam joined	22:49
cromestant	thank you, but I have not done any changes I'm just trying to get the changes from the server	22:49
	the "last commit"	22:49
bamj0rb	(Technically based on just refs/heads/master, but since you just cloned that is based on origin/master)	22:49
dlam	hmm im doing a 'git bisect' and the first bad commit is a merge: how do i inspect what went wrong?	22:49
bamj0rb	cromestant: Check git log origin/develop.	22:49
cromestant	bamj0rb: ok so that is probably what is happening, how do I switch to origin develop	22:49
	→ nettoweb joined	22:50
	← greggroth left	22:50
fobius	cromestant: git checkout -b develop origin/develop will setup develop to track origin/develop	22:50
bamj0rb	cromestant: If you are currently on your develop then you probably want to do git reset origin/develop.	22:50
	→ gnoel joined	22:50
cromestant	ok great!	22:51
	let me try this	22:51
bamj0rb	dlam: A merge commit is not particularly special. It introduces changes. The only real difference is that it has multiple parents.	22:52
	dlam: You can access the different parents of a commit like this: <commit>^1 = 1st parent, <commit>^2 = 2nd parent, etc.	22:52
	← WaReZ left	22:52
	→ harobed joined	22:52
dlam	ohhh cool thanks, didnt know about the concept of 'parents'	22:53
	← zumba_addict left	22:53
bamj0rb	dlam: man gitrevisions	22:53
gitinfo	dlam: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	22:53
	← JEEB left	22:53
	← AaronMT left	22:53
	← richgilbank left	22:54
	→ JEEBcz joined	22:54
	→ DolpheenDream joined	22:54
	→ PeterBoo joined	22:55
	→ AaronMT joined	22:56
	← yaymukund left	22:56
	← lavaman left	22:56
	← jerius left	22:56
	→ mdiers joined	22:57
	← PeterBoo left	22:57
	→ dark_splinter joined	22:57
	→ psyl0n joined	22:57
	← Technodrome left	22:57
	← nettoweb left	22:58
	← BBBThunda left	22:58
	← cff_ left	22:59
	← PeterBooker left	22:59
	← chuckharmston left	22:59
fobius	Is there a sane way to check against all subtrees in a project to see if they are different from their origins?	23:00
	→ chuckharmston joined	23:00
bamj0rb	Define "subtrees". :-/	23:00
	I'm going to say no, but that isn't definitive.	23:00
	→ Gavrisimo joined	23:00
	← johnmilton left	23:01
	→ WaReZ joined	23:01
bamj0rb	I suppose if you just mean a set of paths then you should be able to list them and loop diffing them each. Of course, then you have to define "origin".	23:01
cromestant	that was it thank you all	23:02
	← jave left	23:03
bamj0rb	cromestant: IIRC, git checkout -b origin/develop would have created a local branch named develop based on origin/develop. Or the most clear way is to use the 2 argument git checkout.	23:03
	← rhombus left	23:03
bamj0rb	Actually, I think the last time I tried git checkout -b origin/develop I got a branch literally called origin/develop AKA refs/heads/origin/develop. So I'd probably stick with 2 args.	23:03
	← infynyxx left	23:04
cromestant	yes, exactly what just happened, detached my head :)	23:04
	← chuckharmston left	23:04
	→ jave joined	23:05
	→ sassyapril joined	23:05
	← ewnd9 left	23:05
	← CKyle left	23:06
	→ Technodrome joined	23:07
	← ojacobson left	23:07
	→ shawndumas joined	23:07
	← pengunix left	23:08
fobius	bamj0rb: I'm talking about something like git diff HEAD:/someBranch someBranch except for every someBranch	23:09
	I guess I could just do like ls -d ./bundle \| xargs -I branch git diff HEAD:/branch branch	23:10
	← tufflax left	23:10
	← jarr0dsz left	23:10
	← bamj0rb left	23:10
	→ jarr0dsz joined	23:11
	→ tomnewmann joined	23:11
	← valgrind left	23:12
	→ CKyle joined	23:12
	← tomnewmann left	23:13
	← sassyapril left	23:13
osse	fobius: what does HEAD:/branch mean?	23:13
	← Feelsenburg1 left	23:15
	← el3ctron_ left	23:15
	← esto left	23:15
	→ esto joined	23:15
fobius	osse: Should have been HEAD:branch	23:16
	osse: Just means directory branch on HEAD	23:16
	→ hwrd\|work joined	23:16
bremner	ooc, why is there a directory named branch?	23:16
	→ AWinterman joined	23:16
fobius	bremner: It's not named branch - that's just the placeholder I was using with xargs	23:17
	← halu left	23:17
	← ianmcorvidae left	23:18
	← Roots47 left	23:18
	→ ianmcorvidae joined	23:19
	qsuscs → qsusOFFcs	23:19
	← BlakeRG left	23:19
	← AWinterm_ left	23:20
osse	fobius: git for-each-ref --format='%(refname:short)' 'refs/heads/*' \| while read -r branch; do git diff HEAD "$branch" -- some-dir; done	23:20
	or something like that	23:20
	→ valgrind joined	23:20
	← SocalNick left	23:20
osse	?	23:20
	→ Roots47 joined	23:21
	← chrisgeorge left	23:21
	→ mklappstuhl joined	23:21
	→ SocalNick joined	23:21
fobius	osse: I don't think that's quite the same. git diff HEAD branch -- some-dir implies that the two branches are rooted in the same place and that I want to restrict the changes that I see to that dir.	23:22
	← FortySix2 left	23:22
fobius	In this case the two branches have different roots - the root of one is a subdirectory of the ohter	23:22
osse	I thought that was what you meant	23:22
	oh	23:22
	→ mastro joined	23:23
fobius	I think I can get it to work this way though	23:23
	→ chrisgeorge joined	23:23
fobius	or that would work but it would be do git diff HEAD:branch "$branch"	23:23
	← aantix left	23:23
	SamB mumbles something about man git diff-trees	23:24
	← pantsman left	23:24
osse	fobius: find the sha1s of the trees ("directories") you want to compare and use git diff-tree	23:24
	SamB: heh	23:24
	← JEEBcz left	23:24
	← timthelion left	23:24
osse	you beet mae 2 itt	23:24
SamB	but spelled it wrong I think	23:24
osse	you did	23:24
	← Spyro left	23:25
	→ joaoh82 joined	23:25
	← DoniRS left	23:26
fobius	diff-trees will work too. It's just a little harder I think.	23:26
	← schemanic__ left	23:26
	→ eijk_ joined	23:26
	→ schemanic joined	23:26
	← andschwa` left	23:27
	← liamkeily left	23:27
	← fluxdude left	23:27
	j416_ → j416	23:29
	← jvns left	23:29
	← mbrevda left	23:30
	→ madewokherd joined	23:31
	← mklappstuhl left	23:31
	← Strato left	23:31
	← zhost left	23:32
	← Technodrome left	23:32
	→ Brmm joined	23:32
	→ tufflax joined	23:33
	ELLIOTTCABLE_ → ELLIOTTCABLE	23:35
	→ ajm203_ joined	23:35
	→ JonnyNomad joined	23:35
	← ajm203 left	23:36
	ajm203_ → ajm203	23:36
	← King_Hual left	23:36
	→ teran joined	23:37
	→ nettoweb joined	23:38
	→ Spyro joined	23:38
	← threesome left	23:38
	→ hellome joined	23:39
BlaDe^	is gitolite recommended?	23:40
	← hellome left	23:40
	→ hellome joined	23:41
	← CKyle left	23:41
	→ greggroth joined	23:41
	← madewokherd left	23:42
osse	BlaDe^: for what? Certainly not for cleaning your car! :O	23:43
	← albel727 left	23:43
	→ Technodrome joined	23:43
	← shawndumas left	23:43
	← fairuz left	23:43
	← teran left	23:44
	← cromestant left	23:44
	← DolpheenDream left	23:44
	← mastro left	23:46
	← harobed left	23:46
	← Technodrome left	23:46
	← leeN left	23:46
	← netj left	23:47
	← schemanic left	23:47
	→ netj joined	23:47
	→ schemanic joined	23:48
	← AWinterman left	23:50
BlaDe^	osse: hosting our git repos heh	23:50
	osse: currently on beanstalk, wondering if it'll be a headache to host ourselves or just to go with gitjub	23:50
	github*	23:50
	i basically want to start implementing auto linting and unit testing and such with githooks	23:50
	→ AWinterman joined	23:50
	→ lavaman joined	23:51
osse	BlaDe^: I have no personal experience with gitolite from the admin perspective, but yeah it is generally the tool that is recommended	23:51
BlaDe^	i would probably still use github in read-only mode	23:51
	← AWinterman left	23:51
	→ platzhirsch joined	23:52
	→ AWinterman joined	23:52
platzhirsch	When I enter git rebase -i HEAD~2 vim starts to ask what actions I want to perform, but when I move to the commit and press <e> simply the vim command is executed and not the edit command	23:52
skorgon	platzhirsch, you have to choose the action for each commit and then write the file and close the editor. then git goes and starts the actual rebase	23:53
	← Silmarilion left	23:53
	← eijk_ left	23:53
platzhirsch	skorgon: do I have to insert the e?	23:53
	ah nevermind, now I see it	23:54
	there is already pick, which I replace with edit	23:54
skorgon	if you want to edit a commit, you have to replace the 'pick' with e or edit	23:54
platzhirsch	Didn't quite realize it's there :) thanks	23:54
skorgon	np	23:54
	← breakingthings left	23:54
platzhirsch	Sweet, I always wanted to time travel	23:55
	Devastator_ → Devastator	23:55
	← DrShoggoth left	23:55
skorgon	platzhirsch, just be aware of !rewrite	23:56
gitinfo	platzhirsch: [!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), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	23:56
	← AWinterman left	23:56
platzhirsch	oh good point. No problems since I am the only one working on it, but otherwise...	23:56
	→ AWinterman joined	23:57
	← nettoweb left	23:57
	→ el3ctron joined	23:57
bambams	sitaram: I'm not sure how to create a user account in Linux that can ONLY grant shell access via su -.	23:58
	→ shruggar joined	23:58
	← Stalkr left	23:58
bambams	Hmmm, perhaps passwd --lock. I will investigate...	23:58
	→ johnkevinmbasco joined	23:58
skorgon	bambams, set the login shell to /bin/false?	23:59
	← hholtmann left	23:59
bambams	skorgon: I think that would prevent su - from working, no?	23:59
	← kermit left	23:59

Logs Search ←Prev date Next date→

Channels Documentation