IRCloggy #git 2012-08-15

Provider of IRC logs since 2005.

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

	2012-08-15
	← opus left	00:00
whomp_	how can i make git remember my username and password for each remote change?	00:00
	→ AaronMT joined	00:01
	← lustucru left	00:02
	→ Tabstar joined	00:04
	Tabstar → Tabmow	00:04
	→ spork985 joined	00:05
	← Superlokkus left	00:07
	← dwmw2_gone left	00:08
	→ dwmw2_gone joined	00:08
	→ ovnicraft joined	00:08
	→ w09x joined	00:10
	← spork985 left	00:11
	← praba left	00:11
	← whomp_ left	00:12
	→ lustucru joined	00:12
	← barq left	00:12
	← lustucru left	00:12
	← Octalot left	00:13
	← thelazydeveloper left	00:13
	← aruiz left	00:14
	→ eletuchy joined	00:15
	← landypro_ left	00:16
	→ lustucru joined	00:18
	← ShadeTornado left	00:18
	← SeySayux left	00:19
	→ Dimitr joined	00:19
	← fireh left	00:19
	← jargon- left	00:19
	→ luxurymode joined	00:20
	← eletuchy left	00:20
	→ lustucru_ joined	00:21
	→ vDubG joined	00:21
Sargun_Screen	Where do I get all the graphics used in the git-scm documentation?	00:21
	← lustucru left	00:22
Sargun_Screen	i.e. http://git-scm.com/figures/18333fig0303-tn.png	00:23
	← luxurymode left	00:23
	→ kaizoku joined	00:24
milki	there arent any	00:24
	those are just for illustration purposes	00:24
kaizoku	can you turn off rename detection for git add/rm?	00:24
	I am trying to rm the old version of something and add a new version, but it just shows up as a rename since the filenames are similar	00:24
milki	probably a git config setting	00:25
	→ praba joined	00:26
	← _Vi left	00:27
	→ adam_g joined	00:27
	→ _Vi joined	00:28
	→ opus joined	00:28
kaizoku	setting diff.renames didn't help :/	00:29
Sargun_Screen	milki: How were they generates	00:29
	*generated.	00:29
kaizoku	just gonna split it up into two commits	00:29
milki	Sargun_Screen: maybe graphviz	00:29
	← Dimitr left	00:30
	← hyperair left	00:30
	→ SeySayux joined	00:30
	→ Eridius_ joined	00:30
	→ Dimitr joined	00:30
	← Speed left	00:31
	→ eletuchy joined	00:33
	← Eridius left	00:33
	← afuentes left	00:34
	← fly9 left	00:34
	← ashirus left	00:35
	→ jjbohn joined	00:36
	← nveid left	00:37
	← praba left	00:37
	← maestrojed left	00:37
	→ aspotashev\| joined	00:39
	→ towski joined	00:39
	← btanaka left	00:40
	← e0je left	00:40
	← aspotashev left	00:40
	← ISF left	00:41
	← john_anderson left	00:41
	← neurodrone left	00:43
	→ chriskon149 joined	00:45
	→ praba joined	00:46
sitaram	cbreak: just reading the backlog... was all that intentional? It was perfectly clear to me what he wanted to do	00:48
	→ luxurymode joined	00:48
	← cmullican left	00:49
	→ mritz_ joined	00:51
	→ iori joined	00:53
	← asakura left	00:54
	← linguini left	00:55
	← hwrd\|work left	00:56
	← opus left	00:57
	← Destos left	00:58
	→ hwrd\|work joined	00:58
	← dinesh6654 left	00:59
adam_g	i'm trying to do a bisect between two tags in a repo, but 'git bisect start <newer, bad tag> <older, good tag>' results in 'Bisecting: a merge base must be tested' instead of a nice report of how many revisions left. what does this message mean?	00:59
	← AaronMT left	01:01
	← codeFiend left	01:01
	← maletor left	01:01
	← jjbohn left	01:03
	→ opus joined	01:04
	→ ehsan joined	01:05
	← opus left	01:07
	→ opus joined	01:08
	→ asakura joined	01:08
	← etcetera left	01:10
	→ carpii joined	01:10
carpii	how can i show all log entries between two commits, for a given file, but also show the diff they created ?	01:11
	← kaizoku left	01:12
	→ codeFiend joined	01:12
carpii	i can git diff start..end file but this seems to generate an overall diff and doesnt show which commit did eac hbit	01:12
EugeneKay	carpii - man git-log	01:12
gitinfo	carpii: the git-log manpage is available at http://git-scm.com/docs/git-log.html	01:12
	← hwrd\|work left	01:12
	← iori left	01:13
	→ vitor-br joined	01:13
EugeneKay	I won't spoon feed you the options you want, but log has a bunch of diff-related ones.	01:13
carpii	ive tried --full-diff from the manpage, but it doesnt seem to be changing the output	01:14
	→ hyperair joined	01:14
	→ Gizmokid2005 joined	01:15
	→ shawngo joined	01:15
	→ iori joined	01:16
carpii	ah got it, thx/ seems it wanted -p too	01:16
	or instead of, rather	01:17
	← buhman left	01:19
	← ehsan left	01:19
	← palm_vue left	01:20
	← bizhanMona left	01:20
	← Lgb left	01:20
	→ mike007 joined	01:20
	← towski left	01:20
	→ troyt joined	01:20
	→ cooloney joined	01:21
	→ intripoon joined	01:22
	← TomyLobo left	01:23
	→ buhman joined	01:23
	→ Lgb joined	01:23
	→ palm_vue joined	01:23
	→ hays joined	01:23
	→ bizhanMona joined	01:23
	← tinti_ left	01:23
	← aspotashev\| left	01:23
	← iliketurtles left	01:25
	← intripoon_ left	01:26
sitaram	EugeneKay: got a minute? How does http://dpaste.org/n3FxF/ read to you? Context is that question Jon-- had earlier, and this would fit just after the TOC in http://sitaramc.github.com/gitolite/why.html	01:26
EugeneKay	Was Jon the one who didn't understand what deploy meant?	01:26
sitaram	EugeneKay: if he did, I missed that part in the backlog. he just wanted "how do I let someone fetch but not push", essentially	01:27
EugeneKay	Yeah, that guy.	01:27
	← _Vi left	01:27
EugeneKay	He really wanted to deploy, but he didn't know it.	01:27
	→ _Vi joined	01:28
	→ jjbohn joined	01:28
sitaram	what is deploy anyway? Regardless of how you do it, it has to start with someone/something fetching the latest and greatest from somewhere. And -- usually -- not being able to push back.	01:28
EugeneKay	Anyway.... as for the patch.... I would add a sentence, "Gitolite does better, and more."	01:28
sitaram	euuwill do	01:29
	oops	01:29
	tab fail	01:29
	EugeneKay: will do	01:29
	→ harish joined	01:29
	→ agumonkey joined	01:31
	← glennpratt left	01:31
agumonkey	hi guys	01:31
	→ rohan32 joined	01:31
	← ovnicraft left	01:32
agumonkey	I mistyped a rebase, now the master is on top of a topic. I read about resetting using reflog except I don't know which commit was the last pre-rebase one in master. any idea?	01:32
	← zdk left	01:32
	← danmackay left	01:33
	→ thufir_ joined	01:33
	← thufir_ left	01:33
carpii	you dont need to know, just git reset --hard to the commit shown in reflog	01:34
	git reflog on its own shows a list, so youll soon see what i mean	01:34
rudi_s	agumonkey: And use gitk HEAD@{X} to view the commits if you're unsure which one is the right one.	01:34
carpii	tho best to do it without --hard until youre sure youve done it right	01:34
	→ tqmz__ joined	01:36
	→ Vampire0_ joined	01:36
agumonkey	reflog output after rebase confuses me, I've used it previously without too much troubles, but this time I'm all fuzzy. I'll try gitk to get my feet back on the ground	01:36
	→ zdk joined	01:36
	→ danmackay joined	01:36
	← amcsi left	01:37
grawity	agumonkey: just reset master to what it was before rebase; most likely "master@{1}" – check with `git reflog master`	01:37
	← dhruvasagar left	01:37
agumonkey	ah reflog can delta branch names too.. git's too powerful :p	01:38
	→ private joined	01:38
agumonkey	indeed it's a lot clearer now. Thanks, you rock	01:38
	private → Guest5123	01:39
	← tqmz left	01:39
	← Vampire0 left	01:39
	→ dinesh6654 joined	01:39
agumonkey	do i need to `gc` to prune the rebasecommits ?	01:39
	← Dimitr left	01:40
	→ jaylevitt joined	01:40
	→ jso_ joined	01:41
	← dvide left	01:42
	→ lng joined	01:42
	→ xiangfu joined	01:42
grawity	no, don't bother with it	01:43
	I think git cleans up unreferenced commits and stuff every now and then?	01:43
imMute	yes	01:44
grawity	hmm, actually, I wonder what happens with the old commit if I `commit --amend`	01:44
	← Jay_Levitt left	01:44
grawity	it just remains there for the occassional cleanup?	01:44
rudi_s	Yeah.	01:44
imMute	yep	01:44
	← jaylevitt left	01:44
	→ lustucru joined	01:44
carpii	think its time based, around 30 days	01:45
	→ Milossh joined	01:45
lng	Hi! Git-flow creates branches hierarchy in Git: feature/api, feature/login. How can I acheve the same with Git commands?	01:45
grawity	lng: the same way you create any other branch	01:45
	← mystMan left	01:45
grawity	git checkout -b feature/api, git branch feature/login.	01:45
lng	grawity: does it mean I prepend feature/?	01:46
	← carpii left	01:46
lng	grawity: I see	01:46
	grawity: thanks	01:46
agumonkey	anyone had problems using '/' in branch names ? with 2 '/' git complains about not finding a file, is it confusing name with path ?	01:46
	← avpx left	01:46
	← lustucru_ left	01:47
	← praba left	01:47
rudi_s	carpii: The auto gc depends on the number of loose objects and packs, see man git-gc - the reflog is a separate part. (Not sure if it's handled by git gc --auto though.)	01:47
gitinfo	the git-gc manpage is available at http://git-scm.com/docs/git-gc.html	01:47
	→ dhruvasagar joined	01:47
	→ jackh joined	01:47
lng	how to get git-gc?	01:47
	→ boombatower joined	01:47
lng	which git version?	01:48
imMute	don't worry about running git gc. git will take care of that itself	01:48
	← chriskon149 left	01:48
lng	I have `git gc`, but not `git-gc`	01:48
rudi_s	imMute: Depends on what you do, if you perform many rebases/amends a git gc can save quite some space.	01:48
	lng: It's just the name of the man page.	01:49
	man git-gc documents git gc.	01:49
gitinfo	the git-gc manpage is available at http://git-scm.com/docs/git-gc.html	01:49
	← quazimodo left	01:49
lng	rudi_s: thanks	01:49
rudi_s	(In the past git-gc was the real command name, but that was changed some time ago.)	01:49
	np	01:49
	→ etcetera joined	01:50
	← Tommy[D] left	01:50
	→ praba joined	01:50
	→ Tommy[D] joined	01:50
	→ hellopat joined	01:50
	← cenuij left	01:51
	← flippo left	01:51
	← Sym3tri left	01:51
	← arosen left	01:52
	→ Kulrak joined	01:52
	← Kulrak left	01:52
	→ Kulrak joined	01:52
	← xiangfu left	01:52
	→ dbinoj joined	01:53
dbinoj	can anyone point me to a good resource where i can learn how to use git? i tried many docs, youtube videos and still not sure how people use git in day-to-day development. i understand it is great to keep track of file versions, changes, etc but i'm not getting the idea of how it is done.	01:53
	← darren left	01:53
	← cakehero left	01:53
lng	dbinoj: http://git-scm.com/doc	01:53
dbinoj	lng, thanks. i didnt see those videos until now :)	01:54
lng	dbinoj: read text	01:54
	← LekeFly left	01:55
	← jjbohn left	01:56
dbinoj	lng, i didn't understand what they r trying to say in text documentation. i needed some one to explain. hence i thought video would be good for understanding.	01:56
lng	dbinoj: oh	01:56
	→ Chib joined	01:56
	← jso_ left	01:57
lng	dbinoj: if you understand how to connect to Freenode, I think you can understand Git Doc	01:57
dbinoj	lng, :) i'll give it a shot again today. maybe i was high when i read it earlier and didnt understand a word ;)	01:58
	lng, thanks btw	01:58
	→ iliketurtles joined	01:58
lng	dbinoj: was hight?	01:58
	drunk?	01:58
	or under drugs?	01:58
	→ dinesh6654_ joined	01:59
dbinoj	lng, i'd prefer 'distracted' ;)'	01:59
	→ jjbohn joined	01:59
lng	when you are distracted, do soemthing else	02:00
	:)	02:00
	← apok left	02:00
lng	don't touch git	02:00
	← dinesh6654 left	02:00
	← ferdna left	02:00
dbinoj	lng, sure :)	02:00
Kindari	alias git="echo You are drunk. Go away."	02:01
lng	haha	02:01
	→ ology joined	02:01
	→ xiangfu joined	02:01
	→ Turicas joined	02:02
	← dbinoj left	02:03
	→ Entomo joined	02:04
	← jjbohn left	02:04
EugeneKay	I think you mean gti	02:04
	Entomo → Guest45417	02:05
mh`	hah	02:05
	→ hwrd\|work joined	02:07
	→ centipedefarmer joined	02:08
	← Sky[x] left	02:08
	← archdori left	02:08
	← Fandekasp left	02:08
	→ chriskon149 joined	02:09
	→ iamjarvo joined	02:10
	→ arosen joined	02:10
	← eletuchy left	02:11
	← Marbug left	02:11
	→ cakehero joined	02:11
	→ weiyang joined	02:12
	← Eridius_ left	02:14
	← lustucru left	02:15
	← agumonkey left	02:16
	→ SegFaultAX joined	02:16
	← RaceCondition left	02:16
	→ __machine joined	02:17
	← Guest45417 left	02:17
	→ ehsan joined	02:17
	← HyperNerdV2 left	02:17
__machine	is it possible to clone a remote git repo into an existing non-empty folder, without losing any local files? i get a fatal error destination path already exists and is not empty... can i force it without losing local files? e.g. preserve local file as having modifications if there is a same named file in the git repo?	02:18
	← artnez left	02:18
	→ Purveyor joined	02:18
	← TheJH left	02:18
	→ StubbornTurtle joined	02:19
imMute	IIRC there's a no-checkout or somesuch flag	02:20
EugeneKay	__machine - the sane answer is don't do that. Instead put the repo in a subfolder, then cp the two together	02:20
StubbornTurtle	with a repo, can you check out a repo at a certain tag at any point in time?	02:20
	← mike007 left	02:20
	← pdtpatrick left	02:20
__machine	or maybe i can clone it into another folder and then just move the .git folder ... then i will have local modifications to make the repo identical to my local folder... and i will discard or commit those differences?	02:21
imMute	StubbornTurtle: yep. IIRC, you have to make a branch, though. so `git checkout -b newBranchName tagName`	02:21
	← Xizor left	02:21
StubbornTurtle	will tags slow down if we have say, 1000 of them?	02:22
imMute	nope	02:22
StubbornTurtle	I was going to create a tag for every release by date and release number that day	02:23
	so we can roll back at any point if necessary	02:23
	is that the best approach?	02:23
imMute	don't use tags for that. use tags to make certain commits stand out above others. you can rollback to any commit, tagged or not	02:23
StubbornTurtle	what about lightweight tags?	02:24
	→ Marbug joined	02:24
imMute	"lightweight" ?	02:24
StubbornTurtle	http://learn.github.com/p/tagging.html#lightweight_tags	02:24
imMute	doen't get much lighter than git tags	02:24
StubbornTurtle	"This is basically just the commit checksum stored in a file - no other information is kept."	02:25
imMute	IIRC, they're the same under the hood	02:25
	could be wrong though.	02:26
StubbornTurtle	well, I wouldn't tag every commit	02:26
	→ andrewhl joined	02:26
StubbornTurtle	just any time we roll out from our dev server to main servers	02:26
imMute	sure, if it helps you guys identify things	02:27
	← cakehero left	02:27
StubbornTurtle	imMute: fair enough. Just making sure I'm following best practices; at least generally	02:28
	← _Vi left	02:28
	→ _Vi joined	02:28
	← chriskon149 left	02:30
	→ wolftankk joined	02:32
	← andrewhl left	02:32
	← wolftankk left	02:36
	← hwrd\|work left	02:36
	→ gavin_huang joined	02:37
	← codeFiend left	02:38
	← dhruvasagar left	02:38
	← ZyX-I left	02:38
	← etcetera left	02:38
	→ ZyX-I joined	02:38
	→ hwrd\|work joined	02:39
	← ddelrio1986 left	02:39
	→ dhruvasagar joined	02:40
	→ noyb joined	02:40
	→ krazyj joined	02:40
	→ andrewhl joined	02:41
krazyj	could anyone help me figure out the proper sequence of adding a git submodule while applying a gitignore to the pull?	02:41
	it seems as soon as you `git submodule add` it creates the directory and pulls it	02:41
	I'm confused how to apply a gitignore to that first pull	02:41
	← osmosis left	02:41
	← centipedefarmer left	02:43
	← nor5trom left	02:44
	← orafu left	02:44
	→ orafu joined	02:44
	→ ISF joined	02:44
lng	If master branch is has version tags, is it easy to switch between them to do hotfixes?	02:45
	→ Escherial joined	02:46
Escherial	so, i've committed the common error of pushing a commit onto a branch that i later realize should be in its own branch...i'm not too good with reverting commits and keeping their contents :\	02:47
	← graphitemaster left	02:47
Escherial	how can i migrate it to its own branch now?	02:47
	→ scant joined	02:48
	← luxurymode left	02:48
	→ eletuchy joined	02:48
	← highway900 left	02:49
scant	i am replacing git 1.7.4 with the latest one using an installer (git-osx-installer for my mac os x 10.7) and i was wondering, do i need to uninstall the old git program manually or will the program handle it for me?	02:49
	→ eletuchy_ joined	02:50
	→ cakehero joined	02:50
	← hwrd\|work left	02:50
Escherial	scant: almost certain that you can install over the old one	02:51
	→ Fandekasp joined	02:52
	→ archdori joined	02:52
	← eletuchy left	02:52
scant	kk thx	02:52
	→ chriskon149 joined	02:52
	→ echo-area joined	02:53
	← ehsan left	02:53
	→ darren joined	02:54
	← iliketurtles left	02:55
	← oogatta left	02:56
	← scant left	02:56
	← andrewhl left	02:57
	← dinesh6654_ left	02:58
	→ oogatta joined	02:59
	→ jargon- joined	02:59
	→ hwrd\|work joined	03:01
	← oogatta left	03:02
	← chriskon149 left	03:03
	→ dinesh6654 joined	03:04
	→ oogatta joined	03:05
	← ISF left	03:05
	→ codeFiend joined	03:05
	mpenz → Mpenzi	03:06
	← goshawk left	03:06
	→ artnez joined	03:06
	→ ISF joined	03:06
	→ andrewhl joined	03:07
	→ neurodrone joined	03:07
	→ Nikez joined	03:08
Nikez	Hi,	03:09
gitinfo	Nikez: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	03:09
Nikez	I seem to have a weird issue, whenever I cherry-pick and I encounter a merge conflict it resets the original author. I am on ubuntu 10.04 running 1.7.0.4	03:09
	→ HyperNerdV2 joined	03:10
	→ etcetera joined	03:11
	← dstorrs left	03:11
	← simesy left	03:11
	← rehat left	03:11
	→ v0n joined	03:12
Nikez	Wow, nevermind. Perhaps I should try reading first.	03:12
	→ wolftankk joined	03:13
	→ chrisgeorge joined	03:13
	← StubbornTurtle left	03:15
	← opus left	03:15
	→ opus joined	03:15
	→ graphitemaster joined	03:16
	← Escherial left	03:16
	SeoZ-work[AWAY] → SeoZ	03:18
	← etcetera left	03:18
	→ jso_ joined	03:20
	← jso_ left	03:20
	→ dleonardi joined	03:20
	→ etcetera joined	03:24
	← opus left	03:25
	→ ReekenX joined	03:26
	→ jpalmer joined	03:27
	← praba left	03:28
	← etcetera left	03:29
	→ dsdeiz joined	03:29
	→ mike007 joined	03:30
dsdeiz	hi, is there something special to do for updating a submodule's submodule? right now when i want to pull the latest updates i do git foreach git pull origin master; but it doesn't seem to go through a submodule's submodule	03:31
	→ josh__ joined	03:31
	→ airportyh joined	03:32
	→ CannedCorn joined	03:32
	→ duckxx joined	03:32
airportyh	Hello all, how do I push my tags to github?	03:32
	git tag gives me a list of tags, but don't see that list on github	03:33
	← TheLZA left	03:33
	→ highway900 joined	03:33
	→ aeontech joined	03:33
	← duckx left	03:33
dsdeiz	there's git push --tags	03:33
	← cakehero left	03:34
	← duckxx left	03:34
airportyh	dsdeiz: ah, thanks!	03:34
	→ aFeijo joined	03:34
aFeijo	hi folks	03:34
gitinfo	aFeijo: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	03:34
aFeijo	I need to generate this: git diff .module, but it returns an error "ambiguous argument '.module'"	03:35
	← codeFiend left	03:35
	aeontech → codeFiend	03:35
	→ Goplat joined	03:35
	→ towski joined	03:35
milki	hmm	03:36
	→ fireh joined	03:36
	← hwrd\|work left	03:36
milki	try adding --	03:36
	← josh__ left	03:37
	← fireh left	03:37
	→ Eridius joined	03:37
	→ Xilent_ joined	03:37
	→ itmustbejj joined	03:37
itmustbejj	I was wondering if you guys could help me with a semantics issue I'm having with a coworker….If I have Feature A that is branched off master and the two have diverged and I want to bring masters commits to feature A via rebase do I say "I am rebasing master INTO feature A" or "I am rebasing feature a ONTO master" or are both correct?	03:38
aFeijo	milki, -- before *.module returns nothing, after returns: fatal bad revision	03:38
	→ fireh joined	03:38
	→ luxurymode joined	03:39
	← jargon- left	03:40
wereHamster	itmustbejj: you are rebasing feature onto (the new) master	03:41
	← Xilent left	03:41
	← airportyh left	03:41
wereHamster	itmustbejj: you take commits on the feature branch and apply them onto the new master	03:41
itmustbejj	wereHamster so if I have feature checked out and say git rebase master, you are saying feature becomes "new master"?	03:42
frogonwheels	itmustbejj: if you were rebasing feature A INTO master - then I would probably conclude that you were rebasing Feature A ONTO master and updating master to include feature A	03:42
wereHamster	no	03:42
frogonwheels	itmustbejj: feature checked out.. so 'master' doesn't change. only feature.	03:43
	← dsdeiz left	03:43
itmustbejj	frogonwheels I understand that part, just to be in my use case I am "unrolling feature A's divergent changes, applying master's divergent changes into feature then reapplying feature's changes back on top	03:44
	or at least that is contextually how I see it	03:44
	← Chib left	03:44
	→ dsdeiz joined	03:45
	→ praba joined	03:45
itmustbejj	that the rebase unrolls the divergent changes on the branch you are on, fast forward merges the other (masters) changes into yours then reapplies your original changes on top of the new head of feature (which does match the head of master at that point)	03:45
frogonwheels	itmustbejj: a better way of seeing it is what wereHamster said. which is re-applying feature A's changes on top of the latest master	03:45
	re-applying or replaying	03:45
dsdeiz	so um anyone? do i need to like enter a submodule's submodule directory and do git pull to update it? or is there a better way?	03:45
itmustbejj	frogonwheels right that is exactly what I understand to happen. To me that is putting master's changes "into" feature branch	03:45
	→ rehat joined	03:45
	← HyperNerdV2 left	03:46
itmustbejj	but I could be mistaken in my terminology	03:46
wereHamster	itmustbejj: well, yes, after you 'git rebase master' that feature branch will include all the commits in master	03:46
frogonwheels	itmustbejj: yeah. probably a slightly misleading way of viewing it, but sure.	03:46
itmustbejj	to me if I rebase "onto" a branch that branch is the one that has changed	03:46
	but in actuality feature is the one whose history has changed	03:47
	→ diegoviola joined	03:47
itmustbejj	hence my saying "rebase master into feature a" because feature a's history is the one that is altered	03:47
SethRobertson	I think of it as `git merge` and `git rebase` do the same job, just in different ways	03:47
wereHamster	itmustbejj: git checkout feature; git rebase master; does not touch master.	03:47
itmustbejj	wereHamster exactly	03:47
	→ cenuij joined	03:47
	← cenuij left	03:47
	→ cenuij joined	03:47
	→ centipedefarmer joined	03:47
	→ igotnolegs joined	03:47
wereHamster	itmustbejj: how does 'onto X' suggest that X changes?	03:48
itmustbejj	wereHamster to me "rebase feature onto x's" implies that something is the one being altered	03:49
wereHamster	yeah, feature is.	03:50
itmustbejj	errr sorry I meant to say that master is the one being altered	03:50
wereHamster	I'm sorry you feel that way. But that's the terminology that git uses.	03:50
	← brentratliff left	03:51
itmustbejj	wereHamster I'm perfectly fine adopting the correct nomenclature, I just was unsure which was correct	03:51
	but it sounds like I got my answer	03:51
	onto not into	03:51
	→ lldong joined	03:51
itmustbejj	which sucks for me cuz my coworker is ungracious in victory :P	03:52
	← Ryback_ left	03:52
itmustbejj	wereHamster frogonwheels thanks for the help	03:52
	← praba left	03:54
	← piotrj left	03:55
	← codeFiend left	03:55
	← centipedefarmer left	03:57
	→ duckx joined	03:57
	→ simesy joined	03:57
	→ brentratliff joined	03:58
	← CannedCorn left	03:59
	→ caseymcg joined	04:01
	→ Bosox20051 joined	04:01
	→ arooni-mobile joined	04:03
arooni-mobile	hi folks. there's an external api that is a separate git project that id like to pull into my existing git project. whats the best way of doing that?	04:04
frogonwheels	arooni-mobile: !subproject	04:04
gitinfo	arooni-mobile: [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.	04:04
	→ lirt joined	04:04
frogonwheels	arooni-mobile: I'd say this would be an ideal !submodule candidate.	04:05
gitinfo	arooni-mobile: git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html	04:05
	→ CareBear\ joined	04:06
	← kukks left	04:07
	← rohan32 left	04:07
	← dleonardi left	04:08
	← helgikrs left	04:08
	← luke-jr left	04:08
	← kpreid left	04:08
	→ luke-jr joined	04:09
arooni-mobile	im trying to run: git submodule add url/to/gitproject .... but im getting You need to run this command from the toplevel of the working tree. ... i want the lib to be stored in lib/apiname	04:09
	→ kpreid joined	04:09
arooni-mobile	how can i make this happen	04:09
	← iamjarvo left	04:10
	← mike007 left	04:10
	→ apok joined	04:10
arooni-mobile	suggestions frogonwheels ?	04:10
frogonwheels	arooni-mobile: so start in the top level of your project .. and git submodule add urltoproject lib/apiname	04:10
arooni-mobile	ok silly me	04:10
	← ISF left	04:11
frogonwheels	arooni-mobile: and don't forget to commit the change	04:11
	← Guest69455 left	04:11
	→ chriskon149 joined	04:11
	→ rchavik joined	04:12
	← caseymcg left	04:12
	← gavin_huang left	04:13
arooni-mobile	frogonwheels, thanks!	04:13
frogonwheels	arooni-mobile: so remember now that a submodule is referenced by COMMIT, not branch	04:14
arooni-mobile	frogonwheels, if i want to change the api however; this wont allow me do it right	04:14
	i mean if i want to change the file itself	04:14
frogonwheels	arooni-mobile: you can.. just remember !detached	04:15
gitinfo	arooni-mobile: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	04:15
	→ _huoxito joined	04:15
	← matthiaskrgr left	04:15
frogonwheels	arooni-mobile: you can cd lib/apiname && git checkout master <make some changes> git commit -m 'some changes' && git push origin master && cd .. && git add apiname && git commit -m 'some changes in apiname'	04:16
	arooni-mobile: you should understand that process, and what it all does ^^	04:16
arooni-mobile	so i have to basically do two commits	04:16
	one in the submodule; and one in the main project?	04:16
frogonwheels	arooni-mobile: exactly	04:16
	← bgerber left	04:17
arooni-mobile	if im not sure if the api author will ever update this file	04:17
	and im going to rock'n'roll on it and adapt it as i like	04:17
	← vitor-br left	04:17
arooni-mobile	is it better just to grab the file and keep it as part of my project	04:17
frogonwheels	arooni-mobile: If you plan on updating it yourself, you may then need to change (eventually) the submodule url to your own copy of that interface repo	04:18
arooni-mobile	i guess i could always remove the submodule reference and incorporate it as part of my project right	04:18
	i dont lose anything by sticking with the submodule approach for now	04:18
frogonwheels	arooni-mobile: that m ight be a !subtree then	04:18
gitinfo	arooni-mobile: The git subtree merge method is ideal to incorporate a subsidiary git repositories directly in to single git repository with "unified" git history, where you only need to pull changes in from external sources not contribute your own changes back (which if technically possible is at least difficult). See http://progit.org/book/ch6-7.html Type "!subtree_alt" for more options	04:18
frogonwheels	arooni-mobile: but no, you don't lose anything.	04:18
arooni-mobile	frogonwheels, ive already set up submodule; would you recommend swtiching to subtree now?	04:20
frogonwheels	arooni-mobile: meh, submodule will be fine. I haven't used subtree, so can't really advise.	04:20
	→ hwrd\|work joined	04:21
	→ chriskon_ joined	04:21
	→ matthiaskrgr joined	04:21
	← chriskon149 left	04:22
	← gigamonkey left	04:22
arooni-mobile	frogonwheels, im trying to follow your outline for how to commit changes from submodule; ran git checkout master (from lib directory), says i'm already on master. then i added some comments to the api, the ni did git commit -m "some changes" but it didnt commit	04:23
	https://gist.github.com/9d2d328a6a51c8ac04f5	04:23
	→ helgikrs joined	04:24
	→ bgerber joined	04:24
frogonwheels	arooni-mobile: didyou add the changes? sorry, forgot that. . git add <somefile> you need to look at some basic git !doc howtos.	04:24
gitinfo	arooni-mobile: A list of useful documentation of all kinds is here: http://git-scm.com/documentation -- or try the wiki at http://git.wiki.kernel.org/. Also try typing "!book" "!cs" "!bottomup" "!parable" "!best_practices" or "!vcbe" or "!designers" here in IRC. !book is probably the most helpful.	04:24
	← chriskon_ left	04:25
arooni-mobile	forgot i had to do the full part of two commits	04:26
	← chrisgeorge left	04:27
	← _W_ left	04:27
	→ maestrojed joined	04:27
	→ gigamonkey joined	04:27
	→ _W_ joined	04:27
	← _Vi left	04:28
	→ _Vi joined	04:28
	→ Sigma joined	04:29
	← Nikez left	04:30
	→ alanp joined	04:31
	→ etcetera joined	04:31
	← Davey left	04:33
	→ glennpratt joined	04:34
	→ thirdknife joined	04:35
	← hwrd\|work left	04:35
	→ beautifulmind joined	04:36
	← graphitemaster left	04:38
	← itmustbejj left	04:38
	← diegoviola left	04:38
	→ adlwalrus joined	04:41
	→ chriskon149 joined	04:41
adlwalrus	How can i deploy via FTP by just doing a push?	04:41
milki	adlwalrus: !deploy	04:42
gitinfo	adlwalrus: Git is not a deployment tool. You can build one around it for simple environments. http://sitaramc.github.com/the-list-and-irc/deploy.html	04:42
	→ toabctl joined	04:42
	→ Chibby_ joined	04:43
	→ gavin_huang joined	04:44
	→ graphitemaster joined	04:45
adlwalrus	milki: ah, thanks. :)	04:48
	← adlwalrus left	04:48
	← w09x left	04:48
	← mcluskyd_ left	04:48
	→ codetroll joined	04:48
	→ jargon- joined	04:49
	← Sigma left	04:49
	← etcetera left	04:50
	← andrewhl left	04:50
	← ZyX-I left	04:51
	← lldong left	04:53
	← AlexWaters left	04:56
	← chriskon149 left	04:56
	→ olaf_ joined	04:58
	← apok left	04:59
	← aFeijo left	04:59
	← arooni-mobile left	05:00
	→ subbyyy joined	05:01
	← Esa left	05:02
	← oakdog8_ left	05:02
	← luxurymode left	05:03
	→ etcetera joined	05:04
	→ j416 joined	05:05
	→ samy_ joined	05:05
	→ shruggar joined	05:05
lng	I have tagged master with vesrsions. After certain version checkout, how can I commit changes to similar version of master?	05:05
	← codetroll left	05:06
j416	lng: you might be interested in reading this http://nvie.com/posts/a-successful-git-branching-model/	05:06
lng	j416: this article has no info on this question	05:07
j416	lng: normally you might merge into the latest master. It is not common to incorporate changes into something that is not a branch.	05:07
lng	j416: we need to maintain multiple versions of master	05:07
j416	lng: then make multiple branches!	05:07
	or: git branch mymasterversion <thetag>	05:08
	to start a new branch "mymasterversion" from <thetag>	05:08
lng	j416: how would you name them? matser-v1.0.0?	05:08
	← schemula left	05:09
j416	I would use a naming scheme similar to what is suggested in the link I gave you	05:09
	→ schemula joined	05:09
j416	you might want to adapt it to your environment.	05:09
lng	well, do you use Git-flow?	05:09
j416	I do not	05:09
	← cwang left	05:09
lng	j416: try it	05:10
j416	I have no reason to	05:10
	sorry	05:10
lng	http://jeffkreeftmeijer.com/2010/why-arent-you-using-git-flow/	05:10
j416	I know what it is yes	05:10
lng	j416: ah	05:10
j416	I prefer using git directly, but thanks. :)	05:10
	gives me more control	05:10
lng	I see	05:10
	j416: if mster is tagged by versoins	05:11
	is it possible to merge hotfix branch to master of certian version only?	05:11
j416	if those versions apply only to what is in the master branch, well, maybe tag them as you said	05:11
	a tag is immutable	05:12
	← jonshea left	05:12
j416	but you can create a branch from the tag, merge into that, and then create a new tag	05:12
	(or even merge with detached head, but having a branch to work on tends to make life a little easier)	05:12
lng	ah, yes	05:12
	what does 'detached head' mean?	05:13
	← samy_ left	05:13
lng	HEAD does not point anywhere	05:13
j416	this might not be what you need at all, but about naming your versions -- if you haven't read it, it's a good read http://semver.org/	05:14
	HEAD always points to somewhere	05:14
lng	sometimes head is missing	05:14
j416	detached HEAD means that HEAD is pointing to something that is not a branch head	05:14
	if you check out a tag, then HEAD will be detached	05:14
lng	ah	05:14
	→ Kyzz joined	05:14
j416	because there is nothing to move along when you make changes (branch heads are automatically updated when you commit)	05:15
lng	but when you checkout tag, you create new branch, right?	05:15
j416	no	05:16
	it is the same as checking out a previous commit	05:16
lng	I see	05:16
j416	you can do: git checkout -b branchfromtag tag	05:16
	to create a new branch directly on checkout	05:16
frogonwheels	lng: !detached	05:17
gitinfo	lng: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	05:17
j416	you can do most things while HEAD is detached, the only difference is that when you move away from it (by checking out something else), there will be no pointer to where you were	05:17
	right, thanks frogonwheels	05:17
	← shruggar left	05:17
lng	thanks	05:17
	→ shruggar joined	05:17
frogonwheels	j416: 'no pointer' except the reflog which will eventually release it.	05:18
j416	lng: you can easily create a branch from where you are with your detached HEAD, simply 'git checkout -b mynewbranch' (or git branch mynewbranch)	05:18
	frogonwheels: indeed	05:18
lng	if I work in detached HEAD, how do I commit to certain branch?	05:18
	← Turicas left	05:18
j416	lng: you do not	05:18
lng	oh	05:19
j416	lng: you will have to be on that branch for the branch head to be updated. You can however reset a branch pointer to anything you like, should you have the need.	05:19
	(git reset)	05:19
lng	for instance - if I worked in detached HEAD, how can I save changes?	05:19
frogonwheels	lng: !reattach	05:20
gitinfo	lng: Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue	05:20
j416	git branch mysavedchanges	05:20
frogonwheels	lng: you should really read the gitinfo links supplied	05:20
j416	frogonwheels: I should just save myself some typing haha. thanks	05:20
lng	I do	05:20
frogonwheels	lng: It says to see ! reattach	05:20
j416	lng: when I was new with git, I spent a good few days to read what was then called the 'git community book'; I believe it has been replaced by 'pro git' -- it helps a lot to gain an understanding of the basic underpinnings: http://git-scm.com/book/	05:21
frogonwheels	lng: !doc	05:21
gitinfo	lng: A list of useful documentation of all kinds is here: http://git-scm.com/documentation -- or try the wiki at http://git.wiki.kernel.org/. Also try typing "!book" "!cs" "!bottomup" "!parable" "!best_practices" or "!vcbe" or "!designers" here in IRC. !book is probably the most helpful.	05:21
j416	frogonwheels: I'll just shut up. :D	05:22
lng	thanks	05:22
j416	!book	05:22
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	05:22
j416	hah.	05:22
frogonwheels	j416: no, not at all. That was also for your benefit so you know the manyuseful links from gitinfo	05:22
	→ iamjarvo joined	05:22
lng	I know about http://git-scm.com/doc	05:22
j416	frogonwheels: :)	05:22
	→ fornext joined	05:22
	→ jameshyde_ joined	05:23
frogonwheels	j416: it's worth having a read of the available triggers (fllow the link http://jk.gs/git/ in the channel message)	05:23
	← shruggar left	05:23
jameshyde_	After cloning via "git clone http:...git", how to update the cloned copy with latest updates?	05:24
	← Arokh left	05:25
	→ arvind_khadri joined	05:25
	→ kleist joined	05:26
frogonwheels	jameshyde_: that depends on whether you are comitting stuff yourself to a degree..	05:26
	← iamjarvo left	05:27
frogonwheels	jameshyde_: if you're not, just git pull but if you are, you can still do that, but you should probably understand what !pull does	05:27
gitinfo	jameshyde_: pull=fetch+merge (or with flags/config also fetch+rebase). It is thus not the opposite of push in any sense. A good article that explains the difference between fetch and pull: http://longair.net/blog/2009/04/16/git-fetch-and-merge/	05:27
	→ nickpresta joined	05:27
	← bubbles10101 left	05:27
j416	frogonwheels: thanks. I did in fact read them a while ago -- but I forget :)	05:27
jameshyde_	frogonwheels, yes I have to commit stuff locally, not remotely back to the repo.	05:27
frogonwheels	jameshyde_: ok .. well then decide whether you want to keep on MERGING the upstream stuff into your version of the branch, or REBASING your local changes on top of the latest upstream commits	05:28
	jameshyde_: a git fetch will fetch the upstream stuff into its 'remote tracking branch'.. you can then rebase your changes ontop of that with git rebase @{u}	05:29
	→ gtuckerkellogg joined	05:29
frogonwheels	jameshyde_: alternatively, you can git pull -r	05:29
	→ dstorrs joined	05:29
frogonwheels	jameshyde_: many ways to skin this cat, depending on what you want.	05:29
jameshyde_	frogonwheels, I have to go with REBASING.	05:30
	← SeoZ left	05:30
	← nickpresta left	05:30
lng	say my master ragged like this: v1.0, v2.0, v3.0	05:31
frogonwheels	jameshyde_: :) ok. rebasing it is. git pull -r or the git fetch && git rebase @{u}	05:31
lng	suddenly, I need to hot fix v1.0	05:31
frogonwheels	jameshyde_: effectively the same. you can also set the option in git-config to default git-pull to rebase for that branch.	05:31
lng	I do checkout v1.0	05:31
	→ SeoZ joined	05:31
jameshyde_	frogonwheels, what's @{u}?	05:32
frogonwheels	jameshyde_: 'upstream branch' man git-rev-parse	05:32
gitinfo	jameshyde_: the git-rev-parse manpage is available at http://git-scm.com/docs/git-rev-parse.html	05:32
lng	do fixing and, finaly, I merge it to master creating v.1.1 tag	05:32
	what would be the sequence of tags? v1.0, v2.0, v3.0, v.1.1? Or the order doesn't matter?	05:33
jameshyde_	frogonwheels, if I accidentally deleted one file from the clone directory, how to pull it back from upstream repo? git pull does not seem to work in this case.	05:33
	← SeoZ left	05:33
	← subbyyy left	05:34
frogonwheels	jameshyde_: for a start, everything is cloned into the .git directory, all history &c.. so no need to worry about 'remote' operations.	05:34
	jameshyde_: you can just git checkout <Revision> -- filanem	05:34
	→ SeoZ joined	05:34
frogonwheels	jameshyde_: where <revision> could be @{u} or HEAD or master - anyu !commitish reference that the file existed in	05:34
	← oakdog8 left	05:35
frogonwheels	s/filanem/filename/	05:35
jameshyde_	frogonwheels, thanks, it works!	05:35
	→ banghouse2 joined	05:35
	← banghouse2 left	05:35
	→ banghouse2 joined	05:36
	← intripoon left	05:37
	← marc_kdab_ left	05:38
	← banghouse2 left	05:38
	→ marc_kdab_ joined	05:38
	→ corburn joined	05:38
	→ pantsman joined	05:38
	→ intripoon joined	05:39
	← gerard0 left	05:39
	← russt left	05:40
	→ orkz joined	05:40
	→ BadDesign joined	05:41
	← duckx left	05:42
	→ bubbles10101 joined	05:42
	← bubbles10101 left	05:42
	→ bubbles10101 joined	05:42
	← toabctl left	05:43
	→ gerard0 joined	05:43
	← bubbles10101 left	05:43
	→ bubbles10101 joined	05:43
	← bubbles10101 left	05:43
	← dinesh6654 left	05:43
	→ bubbles10101 joined	05:44
	← hellopat left	05:44
	← bubbles10101 left	05:44
	→ bubbles10101 joined	05:45
	← bubbles10101 left	05:47
lng	How to execute hook script if _only_ master branch was updated?	05:47
	→ bubbles10101 joined	05:47
	← bubbles10101 left	05:47
	→ bubbles10101 joined	05:47
	→ sqlbyme joined	05:47
	← StefanM left	05:47
sharpobject	!dream	05:49
	→ tekbwainz joined	05:50
frogonwheels	lng: !xy	05:50
gitinfo	lng: This sounds like an "XY Problem" http://mywiki.wooledge.org/XyProblem So let's step back for a minute. What are you actually trying to achieve? Why are you doing it this way?	05:50
	← tekbwainz left	05:50
	→ tekbwainz joined	05:50
	tekbwainz → phreaky	05:50
	→ shruggar joined	05:51
	← sqlbyme left	05:51
_ikke_	Nice, that trigger is much better than it used to be	05:52
	→ Fullmoon joined	05:52
	← bubbles10101 left	05:52
frogonwheels	_ikke_: thankyou. :)	05:52
	→ bubbles10101 joined	05:52
	← Chappo left	05:53
	← bubbles10101 left	05:53
	→ bubbles10101 joined	05:53
_ikke_	The earlier version said the user was doing it wrong, which isn't necessarily always the case	05:54
frogonwheels	_ikke_: Yes. Exactly.	05:55
	→ HyperNerdV2 joined	05:55
frogonwheels	_ikke_: !doinitwrong	05:55
_ikke_	yeah, indeed	05:55
frogonwheels	I think that one got updated as well.	05:55
_ikke_	Yeah, it's now an alias for xy	05:55
	→ maleknet joined	05:55
frogonwheels	ah yes. no w	05:56
_ikke_	lol, !zombie	05:56
gitinfo	Brainzz! More brainzzzz! For information about detached heads, see !detached	05:56
frogonwheels	_ikke_: :) thanks :) Must have been in a strange frame of mind when I wrote that one	05:57
_ikke_	Oh, didn't see you wrote that one :D	05:58
	→ ovnicraft joined	05:58
	→ Nirvanko joined	05:59
	← graphitemaster left	06:00
	→ diegoviola joined	06:01
	→ yareally joined	06:01
	← arvind_khadri left	06:02
	← ovnicraft left	06:03
	→ jonshea joined	06:04
	→ apok joined	06:07
	→ josh__ joined	06:09
	← josh__ left	06:10
	← towski left	06:10
	← shruggar left	06:11
	→ unmanbearpig joined	06:12
	→ dinesh6654 joined	06:12
	→ lustucru joined	06:12
	← boombatower left	06:13
	→ boombatower_ joined	06:13
	← jackh left	06:13
	→ ovnicraft joined	06:15
	→ senny joined	06:16
	→ papegaaij joined	06:16
	→ arvind_khadri joined	06:18
	← maestrojed left	06:18
	← etcetera left	06:19
	→ agumonkey joined	06:19
	← corburn left	06:20
	→ etcetera joined	06:20
	← neurodrone left	06:20
	← boombatower_ left	06:20
	← gtuckerkellogg left	06:21
	→ rctay joined	06:21
	→ corburn joined	06:21
	← lustucru left	06:21
	← rctay left	06:22
lng	frogonwheels: XY? I just want to deploy to Prod when master branch is updated	06:22
	→ rctay joined	06:22
_ikke_	lng: define updated?	06:22
	lng: The hook script is always run. So you have to check inside the hook if the conditions you want are met	06:23
lng	_ikke_: that's what I thought of actually	06:23
_ikke_	lng: Are you talking about a post-recieve hook?	06:24
lng	yes	06:24
	← v0n left	06:24
_ikke_	You'd have to read from stdin	06:24
	← thirdknife left	06:25
_ikke_	see man githooks, the format is described in the pre-receive hook	06:25
gitinfo	the githooks manpage is available at http://git-scm.com/docs/githooks.html	06:25
_ikke_	<old-value> SP <new-value> SP <ref-name> LF	06:25
	← Error404NotFound left	06:25
lng	thank you for the info	06:25
	→ thirdknife joined	06:26
frogonwheels	lng: exactly. your x was 'deploy prod when master branch is updated' and your y was 'How to execute hook script if _only_ master branch was updated?'	06:27
	→ tsdh joined	06:27
	→ oakdog8 joined	06:27
lng	frogonwheels: I'm learning by doing	06:27
	and it is always successful	06:27
frogonwheels	lng: sure. which is good. buf if you'd asked your x, I'd have said !deploy	06:28
gitinfo	lng: Git is not a deployment tool. You can build one around it for simple environments. http://sitaramc.github.com/the-list-and-irc/deploy.html	06:28
	← mritz_ left	06:28
frogonwheels	lng: assuming Prod is the production server, and master is the branch you're updating?	06:29
	← _Vi left	06:29
lng	frogonwheels: yes	06:29
frogonwheels	lng: anyway, what _ikke_ said is relevent	06:29
lng	actually merging	06:29
	← BadDesign left	06:30
	→ _Vi joined	06:30
lng	for now I use Ant	06:30
	← rctay left	06:30
lng	and it works pretty good	06:30
	← tqmz__ left	06:30
frogonwheels	lng: merging? that's a whole different kettle of poisson. you shouldn't push to a !non-bare repo - which you'd need if you're going to merge.	06:30
gitinfo	lng: [!bare] an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/	06:30
lng	but we need to add versioning support to the API	06:30
	→ spronk joined	06:30
	→ tqmz joined	06:30
_ikke_	lng: and?	06:31
	lng: That's not something git is concerned about	06:31
lng	yes	06:31
	I know	06:31
_ikke_	frogonwheels: How was what I said irrelevant?	06:32
lng	I just need it to trigger build script...	06:32
frogonwheels	_ikke_: _is_ _relevent_ is what I said!	06:32
lng	anyway	06:33
	← Goplat left	06:33
lng	I have enough info now	06:33
	I will read first	06:33
_ikke_	frogonwheels: Oh, read it incorrectly ;)	06:33
lng	thanks a lot guys	06:33
	→ rctay joined	06:33
	← fisted left	06:34
	← rctay left	06:34
	→ fisted joined	06:34
	→ robbyoconnor joined	06:35
	← noyb left	06:36
	← igotnolegs left	06:37
	← yoklov left	06:39
	← hyperair left	06:40
	→ beepbeep_ joined	06:40
	← harish left	06:40
	→ maestrojed joined	06:40
	→ ferdna joined	06:41
	← _Vi left	06:41
	→ cbreak-work joined	06:41
	→ hyperair joined	06:42
	→ jnewt joined	06:42
	→ quazimodo joined	06:43
jnewt	anyone uset odt2txt for git diff of open doc files? it seems so easy to setup, yet git diff still shows the binary files differ message.	06:43
	→ fayimora joined	06:44
	← JrCs left	06:44
frogonwheels	jnewt: !doesntwork	06:44
gitinfo	jnewt: Sorry to hear it doesn't work. What happened? Did it fall asleep on your couch, with your sister? Eat your homework? What did you want it to do? What happened instead? Please be specific!	06:44
	← dhruvasagar left	06:44
	→ flijten joined	06:45
	→ JrCs joined	06:45
Nirvanko	I've read about git submodules and now I am pondering about whether to use its functionality or not, it seems to be a little bit complicated.	06:45
	git clone might be my solution.	06:45
	→ dhruvasagar joined	06:46
_ikke_	Nirvanko: Well, what problem do you want to solve	06:46
	→ lb joined	06:46
Nirvanko	_ikke_, I have to integrate some large projects in my project.	06:46
	← mishok13 left	06:47
	← darren left	06:47
_ikke_	Nirvanko: Is that project something you develop yourself, or is someone / something else responsible for that?	06:47
	→ jceb joined	06:47
_ikke_	(the large projects you want to integrate)	06:47
jnewt	frogonwheels: git diff still shows the standard "Binary files a/docs/test.odt and b/docs/test.odt differ" command was git diff docs/test.odt	06:47
	← wolftankk left	06:48
_ikke_	jnewt: How have you setup your gitattributes?	06:48
Nirvanko	_ikke_, no. That's all third party libraries.	06:48
	→ darren joined	06:49
	← jonshea left	06:49
_ikke_	Nirvanko: afaik, you still have to setup gitattributes to tell git to use it	06:49
jnewt	.ods diff=odf .odt diff=odf *.odp diff=odf each on one line _ikke_	06:49
_ikke_	Nirvanko: wait, two conversations mixed up	06:49
AAA_awright	Nirvanko: Git submodules let you point to a particular repository at a particular commit, while letting the submodule be independently developed and committed to. Initalizing/updating and maybe symlinking to files within the repository could be done in a Makefile.	06:49
_ikke_	Nirvanko: Is your project depending on a certain version of those projects?	06:50
	Nirvanko: If so, submodules are the best way, because as AAA_awright says, it keeps track of the version of a project, so that everyone has the same version	06:50
	→ Alvo joined	06:51
	← fayimora left	06:51
	← HyperNerdV2 left	06:52
	→ harish joined	06:52
jnewt	when i just do odt2txt /docs/test.odt odt2txt outputs a text representation of the file. this leads me to believe that the issue is that git is not using it.	06:53
	→ ph^_ joined	06:53
Nirvanko	_ikke_, changing branches produces the following - warning: unable to rmdir vendor/pagerfanta: Directory not empty.	06:54
	What's wrong.	06:54
	→ jonshea joined	06:54
	← ovnicraft left	06:54
Nirvanko	vendor/pagerfant is the submodule I've added.	06:55
_ikke_	Nirvanko: Are you on an older version of git?	06:55
	→ shruggar joined	06:55
Nirvanko	1.7.1	06:55
	← maestrojed left	06:56
_ikke_	Nirvanko: Newer versions of git deal with that a lot better	06:56
	Nirvanko: the problem with older versions of git is that the .git directory is inside the submodule folder, and git can't remove that folder. That's why it's complaining	06:56
	← jameshyde_ left	06:57
	← jonshea left	06:57
Nirvanko	The latest stable version is 1.7.14	06:58
jnewt	maybe it's the file location of .gitattributes or .gitconfig? I expect them to work in the same way that .gitignore does. there are other options such as having the files in $HOME or $GIT_DIR, but i still think it should work as is	06:58
	→ opus joined	06:58
Nirvanko	err, 1.7.11.4	06:58
_ikke_	Nirvanko: yes. It got in somewhere after 1.7.5	06:58
	→ leeN joined	06:59
_ikke_	Nirvanko: Ah, found it. It landed in 1.7.8	07:00
	→ nasso joined	07:01
	→ Araxia joined	07:01
	← darren left	07:01
_ikke_	https://gist.github.com/3357263	07:01
	← sh4rm4 left	07:03
oojacoboo	wow, that was fun, I just cherry-picked all the commits and merged them and fixed god knows how many thousands of lines of conflicts	07:04
	I love git	07:04
	→ charon joined	07:04
oojacoboo	it only took 14 hours	07:04
	→ arturaz joined	07:04
_ikke_	oojacoboo: Note sure if that was meant sarcastically or not	07:05
	← opus left	07:05
oojacoboo	14 hours of conflict merging sound good to you?	07:05
	→ opus joined	07:05
jnewt	no other way i'd rather spend a day	07:06
_ikke_	oojacoboo: No clue, don't know how complex the commits were. But I don't think there is any tool that could do it better than that	07:06
oojacoboo	jnewt, exactly… my eyeballs burn	07:06
Nirvanko	_ikke_, what is the official site of the Git project? I would like to update Git.	07:06
_ikke_	oojacoboo: Obviously, history was not compattible, so you have to merge one way or another	07:07
Nirvanko	Upto 1.7.8 I guess	07:07
_ikke_	Nirvanko: What system are you on?	07:07
Nirvanko	Linux, Slackware	07:07
jnewt	oojacoboo, be glad it's done, go to sleep. tomorrow will be a better day.	07:07
oojacoboo	_ikke_, right… I love employees	07:07
	jnewt, that part of it, now it's gotta be tested like mad, there is no telling what I missed in 14 hours of conflict merging...	07:08
_ikke_	git tries to make it as easy as possible, but there's only so much git can do	07:08
oojacoboo	just thinking about it makes me sick	07:08
jnewt	you missed a lot. anything else, and you win.	07:08
_ikke_	Nirvanko: git-scm.org	07:08
	.com I meant	07:08
oojacoboo	well, the app loads and no errors are being thrown	07:08
	so that's a decent start	07:08
_ikke_	oojacoboo: I've hear about merges that took a lot longer than 14 hours under svn	07:10
	head*	07:10
	→ mishok13 joined	07:10
Nirvanko	_ikke_, but they claim that the latest stable version is 1.7.11.4	07:10
oojacoboo	I guess :/	07:10
_ikke_	Nirvanko: Yes, and?	07:10
oojacoboo	Nirvanko, get a package manager for the love of everyone	07:10
Nirvanko	So where should I get 1.7.8 )	07:10
_ikke_	oojacoboo: Merging code is not a computer problem, it's a developer problem.	07:10
	→ shreya joined	07:11
_ikke_	Nirvanko: Under the tarballs section	07:11
oojacoboo	_ikke_, yes, I concur	07:11
_ikke_	But why not use 1.7.11.4 directly	07:11
oojacoboo	I just need something to complain about	07:11
_ikke_	oojacoboo: yeah, solving that much conflicts is a huge pain	07:11
	← artnez left	07:12
oojacoboo	what does "local branch should track remote branch" mean?	07:12
	when doing a checkout	07:12
_ikke_	oojacoboo: !repro?	07:12
gitinfo	oojacoboo: Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session, or at least explain exactly what you did that led up to the problem. This will help immensely with troubleshooting.	07:12
	→ dc5ala joined	07:13
oojacoboo	SourceTree checkbox when checking out a remote branch :/	07:13
_ikke_	oojacoboo: I guess it	07:13
	← _huoxito left	07:13
_ikke_	What command did you execute?	07:14
oojacoboo	nothing yet	07:14
	git branch --set-upstream foo upstream/foo	07:14
	apparently it's something like that	07:14
_ikke_	yup, that's how it's don	07:14
	done	07:14
	→ bogwonch joined	07:14
oojacoboo	yea, but what does that DO?	07:14
_ikke_	Just wondering what triggered that message	07:14
	← juvenal left	07:15
_ikke_	oojacoboo: It adds some config options to .git/config to tell what local branches belong to what remote branches	07:15
oojacoboo	as opposed to it not being sent to remote and instead just being local?	07:15
	← ConstantineXVI left	07:15
_ikke_	oojacoboo: It helps when doing git push (without any parameters), and when doing git status (your branch is behind ..)	07:16
	→ wolftankk joined	07:16
oojacoboo	gotcha	07:16
	SeoZ → SeoZ-work[AWAY]	07:16
	← arvind_khadri left	07:17
	← bizhanMona left	07:20
	← palm_vue left	07:20
	← Lgb left	07:20
	← eletuchy_ left	07:22
	→ palm_vue joined	07:23
	→ Lgb joined	07:23
	→ bizhanMona joined	07:23
	← thirdknife left	07:23
	→ nor5trom joined	07:24
	→ arvind_khadri joined	07:25
	→ tiglog joined	07:25
oojacoboo	git stage hunk --no-homo	07:26
	← beautifulmind left	07:26
	→ dsirijus joined	07:28
	← beepbeep_ left	07:28
	← apok left	07:29
	← localhost left	07:29
frogonwheels	oojacoboo: it also gives you @{u} man git-rev-parse	07:29
gitinfo	oojacoboo: the git-rev-parse manpage is available at http://git-scm.com/docs/git-rev-parse.html	07:29
	→ _Vi joined	07:30
	→ d3vic3 joined	07:30
oojacoboo	huh?	07:30
_ikke_	oojacoboo: @{u} is a reference to the remote tracking branch	07:31
	→ kurrataW joined	07:31
oojacoboo	ah	07:31
frogonwheels	oojacoboo: which you can find on the git-rev-parse manpage :)	07:31
oojacoboo	I feel like learning the git api is harder than most languages	07:32
	→ hwrd\|work joined	07:32
oojacoboo	frankly feels like there are too many options	07:32
frogonwheels	oojacoboo: !simple	07:32
gitinfo	oojacoboo: At its heart git is made up of many concepts that are individually simple. Getting the whole picture right is often tricky, and it is usually about breaking up the complex concept into its simple, individual parts and grokking those. Both !bottomup and !cs will help with that.	07:32
_ikke_	oojacoboo: most things you don't need to operate git	07:33
	oojacoboo: some things are shortcuts you'll learn as you go	07:33
oojacoboo	yea	07:33
frogonwheels	oojacoboo: .. and you learn what options are right for your workflow.	07:33
oojacoboo	I just find it confusing, I kinda wish they'd just remove and consolidate half the stuff	07:33
	→ jkremser joined	07:34
frogonwheels	oojacoboo: I don't think they should remove anything. If you're getting lost, you can always use one of the many UI wrappers	07:34
wereHamster	oojacoboo: but then git wouldn't be much different from cp -r and diff	07:34
	← dinesh6654 left	07:35
_ikke_	oojacoboo: git takes some time to get used to, but in return you get a very flexible system that supports many different workflows	07:35
frogonwheels	oojacoboo: I mean that honestly. A good UI - even just gitk and git gui or git-cola or smartgit (commercial x-platform) can help you understand - the cli can come later	07:35
	→ drizzd joined	07:35
oojacoboo	frogonwheels, I'm already using Atlassian's SourceTree	07:36
	I find a gui pretty paramount for keeping things organized	07:36
	the git daemon is nice, but there is only so much you can get out of a cli	07:36
	→ ConstantineXVI joined	07:37
_ikke_	git daemon?	07:38
oojacoboo	script	07:38
_ikke_	..	07:38
oojacoboo	?	07:38
	← dhruvasagar left	07:38
	← archdori left	07:39
_ikke_	git deamon is "A really simple server for git repositories"	07:39
cbreak-work	oojacoboo: don't waste your time with git daemon	07:39
jnewt	how do i stage all deletes that are showing in the "not staged for commit"? git rm -r --cached . removes everything. git rm -r doesn't work if there are other staged files.	07:39
cbreak-work	it's for read only unauthenticated access only	07:40
	← shruggar left	07:40
cbreak-work	(or writable unauthenticated if you're so promiscuous...)	07:40
_ikke_	jnewt: man git rm	07:40
	← weiyang left	07:40
	_ikke_ pokes gitinfo	07:40
cbreak-work	jnewt: and man git-add -u	07:40
	→ archdori joined	07:40
oojacoboo	I guess I mean the git client daemon	07:40
cbreak-work	oojacoboo: the only git daemon is the git server	07:40
_ikke_	jnewt: git diff --name-only --diff-filter=D -z \| xargs -0 git rm --cached	07:41
cbreak-work	but it's not used much, ssh:// is a much much better transport	07:41
	no daemon needed	07:41
_ikke_	oojacoboo: there is no client daemon	07:41
oojacoboo	ok, so the shell script?	07:41
gitinfo	jnewt: the git-rm manpage is available at http://git-scm.com/docs/git-rm.html	07:41
	jnewt: the git-add manpage is available at http://git-scm.com/docs/git-add.html	07:41
	← ConstantineXVI left	07:41
cbreak-work	oojacoboo: what shell script?	07:41
	← hwrd\|work left	07:41
oojacoboo	what is the "git" command in my terminal :/	07:41
jnewt	_ikke_ wth is that? i'm looking for the easy way, and i know how to rtfm, i still don't know how to do it.	07:41
_ikke_	oojacoboo: A program	07:41
cbreak-work	oojacoboo: that's git	07:41
oojacoboo	lol	07:41
	a program?	07:41
cbreak-work	a bunch of them.	07:42
oojacoboo	are those not daemons?	07:42
_ikke_	jnewt: git has no option: check what files don't exist anymore on the filesystem and stage them to be removed	07:42
cbreak-work	no.	07:42
	they only run when you start them	07:42
	if you do git log, you start git	07:42
oojacoboo	ah	07:42
cbreak-work	it will print the log, and then it terminates	07:42
oojacoboo	what's it written in, perl?	07:42
cbreak-work	C	07:42
	mostly	07:42
oojacoboo	fun	07:43
	→ hwrd\|work joined	07:43
	→ arex\ joined	07:43
jnewt	so wait a minute. i asked a question, got told to rtfm, then told git doesn't do it? good thing i was told to rtfm.	07:43
	→ eletuchy joined	07:43
cbreak-work	jnewt: git doesn't do what?	07:44
_ikke_	jnewt: That snippet was from the manual	07:44
cbreak-work	jnewt: just read what I wrote	07:44
_ikke_	under other ways to do it	07:44
	← rehat left	07:44
	gitinfo set mode: +v	07:44
_ikke_	cbreak-work: But that also adds unstaged changes	07:44
arex\	Is there a concencus on what is the "best" Git solution for enterprise (in-house, active directory, many teams/projects)?	07:44
cbreak-work	_ikke_: that's what he wants	07:44
	unstaged deletes.	07:44
_ikke_	cbreak-work: Yeah, but unstaged modifies would be added too	07:45
cbreak-work	yeah, and?	07:45
_ikke_	cbreak-work: It might be what he wants, but it also might not be what he wants	07:45
	git add -u add's everythin	07:45
	→ timthelion joined	07:45
cbreak-work	if he doesn't want that, he can use -i too	07:45
	→ ncd\|leeN joined	07:45
	← ncd\|leeN left	07:45
_ikke_	Of course	07:45
	or he can execute	07:46
	git diff --name-only --diff-filter=D -z \| xargs -0 git rm --cached	07:46
	copied from the man git rm page	07:46
gitinfo	the git-rm manpage is available at http://git-scm.com/docs/git-rm.html	07:46
cbreak-work	nah. that would involve reading the manual	07:46
	← simesy left	07:46
	← pantsman left	07:47
_ikke_	:D	07:47
oojacoboo	anyone use SourceTree?	07:47
	if you change the remote branch being tracked, is that like rebasing?	07:48
cbreak-work	oojacoboo: no.	07:48
oojacoboo	k	07:48
cbreak-work	rebasing is history rewriting. changing tracking branches is just a config change	07:48
	← hwrd\|work left	07:48
oojacoboo	yea, I get it now, thanks	07:48
jnewt	to add new modifications "git add ." if that modification happens to be a delete: "git diff --name-only --diff-filter=D -z \| xargs -0 git rm --cached" .. no one see anything odd about that?	07:49
	← ferdna left	07:49
cbreak-work	jnewt: no	07:49
	to add new modifications, git add -u	07:50
	git add . means "add the current directory and its contained files"	07:50
	jnewt: are you slow or something?	07:50
	I told you about -u ages ago...	07:50
_ikke_	jnewt: The difference is that you don't want to rm all, you want to rm just a part	07:50
	→ Element9 joined	07:51
	→ weiyang joined	07:53
blathijs	intransit: But you can still pass a path to git add -u, right?	07:54
	s/intransit/jnewt/	07:54
cbreak-work	of course you can.	07:54
	→ ConstantineXVI joined	07:54
cbreak-work	default is .	07:54
arex\	http://stackoverflow.com/questions/11965787/software-for-git-in-the-enterprise	07:54
	→ Chappo joined	07:54
cbreak-work	enterprise?	07:54
	git works on all kinds of starships.	07:55
	if you have some kind of unix os at least :)	07:55
	← highway900 left	07:55
cbreak-work	there's no additional software needed	07:55
	→ Weiss joined	07:57
	← Textmode left	07:58
torbjorn	I did git init in a directory, how can I make it so that others can access it, pull from it, and push back into it?	08:00
_ikke_	torbjorn: !bare	08:01
gitinfo	torbjorn: an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/	08:01
	→ dzonder joined	08:01
frogonwheels	torbjorn: you might want to read some !doc there are some good simple examples.	08:01
gitinfo	torbjorn: A list of useful documentation of all kinds is here: http://git-scm.com/documentation -- or try the wiki at http://git.wiki.kernel.org/. Also try typing "!book" "!cs" "!bottomup" "!parable" "!best_practices" or "!vcbe" or "!designers" here in IRC. !book is probably the most helpful.	08:01
frogonwheels	torbjorn: !gitolite	08:01
gitinfo	torbjorn: Want to host as many git repos (and users!) as you like, on your own server, with fine-grained access control? You want gitolite: https://github.com/sitaramc/gitolite - Documentation: http://sitaramc.github.com/gitolite/master-toc.html	08:01
torbjorn	thanks!	08:01
	→ Kaisyu joined	08:01
	← Bosox20051 left	08:02
	→ whitman joined	08:02
	→ LimeBlast joined	08:03
	→ jjido joined	08:05
	← Kaisyu left	08:05
	← Weiss left	08:05
	→ Weiss joined	08:06
	← blami left	08:07
	→ Textmode joined	08:08
	→ aspotashev joined	08:11
	← etcetera left	08:12
	→ jtri joined	08:13
	→ psoo joined	08:13
	← jtri left	08:14
	→ polymar joined	08:16
	→ tatsuya_o joined	08:16
	← sawdisk left	08:17
	← jazper- left	08:17
	← tweedledumber left	08:18
torbjorn	is there a way to add an empty directory to a git repo?	08:19
	→ tweedledumber joined	08:19
	→ dr_lepper joined	08:19
	← dr_lepper left	08:19
	→ dr_lepper joined	08:19
muep	torbjorn: no	08:20
	← borwar left	08:20
m1sc	torbjorn: no -- you can achieve this by putting an empty .gitignore in it for example	08:21
torbjorn	thanks	08:21
_ikke_	!empty	08:21
gitinfo	[!orphan_null] To create an orphan branch with the null/empty commit, use: `git checkout -b orphan $(echo "empty" \| git commit-tree 4b825dc642cb6eb9a060e54bf8d69288fbee4904)`	08:21
_ikke_	Nope	08:21
	ignore that	08:21
	→ _iron joined	08:22
	→ tinti_ joined	08:22
	← Textmode left	08:23
	→ dipsi joined	08:23
	← dipsi left	08:23
	→ dipsi joined	08:24
	→ thirdknife joined	08:24
	← dipsi left	08:24
	← opus left	08:24
	→ dipsi joined	08:25
	← dipsi left	08:25
lng	when I do git checkout, should it update local source code from remote?	08:25
cbreak-work	no	08:25
	→ shruggar joined	08:25
	→ dipsi joined	08:25
cbreak-work	checkout is local only	08:25
lng	or it just swithing the brunch	08:25
	← hyperair left	08:25
lng	I see	08:25
	← dipsi left	08:25
arex\	cbreak: i realize no other software is needed, but couldn't it be useful?	08:25
	→ dipsi joined	08:26
cbreak-work	arex\: useful?	08:26
	← dipsi left	08:26
lng	cbreak-work: I am asking because in this article [http://sitaramc.github.com/the-list-and-irc/deploy.html], `git checkout -f master` is used for deployment	08:26
	and I don't understand how it works	08:27
arex\	cbreak-work: Do you think it's best for 10 dev teams to only use git from the command line (or an IDE plugin)? (genuine question)	08:27
cbreak-work	that's used after getting a push.	08:27
	it's not a clean strategy	08:27
	lng: you should use git fetch and then git archive for deployment	08:28
jnewt	arex\ in an enterprise setting i'd think you'd have your own server already, so why would you have someone else host it?	08:28
	or maybe you don't	08:28
cbreak-work	arex\: what ever suits them best.	08:28
lng	cbreak-work: I use Ant anyway ;-)	08:28
cbreak-work	I use git on the command line, + a few visual tools	08:28
arex\	jnewt: no, i'm up for hosting it myself, just wondering what software to use	08:28
jnewt	gitolite / gitweb is what i'm using. but i'm pretty new still	08:29
lng	cbreak-work: okay	08:29
arex\	as per http://stackoverflow.com/questions/11965787/software-for-git-in-the-enterprise	08:29
	← LittleFool left	08:29
cbreak-work	gitolite.	08:29
	no competition.	08:29
	← jceb left	08:29
arex\	cbreak-work: what is gitolite's advantages? i'll take a look at it	08:30
	← _Vi left	08:30
	← jdav_gon` left	08:30
cbreak-work	arex\: advantage: it does its job	08:30
jnewt	and for free	08:30
	→ _Vi joined	08:30
	← Adys left	08:31
arex\	Have you tried Atlassian Stash? Seems sweet on the surface	08:31
	→ Adys joined	08:31
	← jjido left	08:31
	→ jjido joined	08:31
	→ Vert joined	08:33
	→ LittleFool joined	08:33
	→ hyperair joined	08:33
	→ toabctl joined	08:33
	← Vile left	08:34
operhiem1	Am I recalling correctly that there's server software which offers git and Mercurial access to the same repository transparently? I thought I recalled such a thing but I'm having trouble finding it.	08:34
	→ borwar joined	08:34
	→ oehman joined	08:35
	→ graphitemaster joined	08:35
jnewt	arex\ looks like gitweb and gitolite. not sure whats so sweet / special about it.	08:36
cbreak-work	operhiem1: doesn't sound plausible	08:36
	git's model is quite different from hg	08:37
	← charon left	08:37
	→ charon joined	08:37
	→ opus joined	08:37
	→ Arokh joined	08:37
arex\	jnewt: integration with all the other atlassian tools and slick web interface :P	08:38
	← jjido left	08:38
jnewt	gitweb = slick web interface. i don't know much about atlassian anything, so maybe.	08:38
operhiem1	cbreak-work: Ah, alright. Hm. Maybe I was just misremembering hg-git.	08:39
arex\	jnewt: guess i'll have to look at gitweb as well then	08:39
charon	operhiem1: srabbelier's hg-git project should still work, though i'm not sure it was ever released for hg. (it's definitely not in git mainline.) but it's client-side	08:39
	→ gtuckerkellogg joined	08:39
cbreak-work	from what I heard from trolls who wandered in here	08:40
	hg has branches that can have multiple heads without names	08:40
	and commits are always bound to a branch	08:40
	you can't merge them into other branches	08:40
	also, branches seem to be global and not local	08:40
	(which is weird, considering that hg is a distributed scm as well)	08:41
	→ Textmode joined	08:41
	← C8H10N4O2 left	08:41
	← jkremser left	08:42
charon	cbreak-work: well, i don't have a clue either, but i have in fact tried hg-git at one point because i needed to interface with hg. it worked. so i'm not sure what the problem would be, other than lack of time/motivation to move it forward.	08:42
lb	!pull	08:42
gitinfo	pull=fetch+merge (or with flags/config also fetch+rebase). It is thus not the opposite of push in any sense. A good article that explains the difference between fetch and pull: http://longair.net/blog/2009/04/16/git-fetch-and-merge/	08:42
	← toabctl left	08:42
	→ toabctl joined	08:43
	→ jceb joined	08:44
	← tatsuya_o left	08:45
	← oakdog8 left	08:45
	→ Kaisyu joined	08:49
	← eedeep left	08:54
	← lb left	08:56
	→ lb joined	08:56
	← oehman left	08:57
	→ oehman joined	08:57
	← unmanbearpig left	08:58
	→ blami joined	08:59
	→ C8H10N4O2 joined	09:00
	→ oakdog8 joined	09:02
	← oojacoboo left	09:03
	→ oojacoboo joined	09:05
	→ Sky[x] joined	09:05
	→ ashirus joined	09:07
	→ LekeFly joined	09:07
	→ thiago joined	09:08
	← lirt left	09:08
	← opus left	09:09
	← LekeFly left	09:09
	→ Vile joined	09:10
	← wolftankk left	09:12
	← oehman left	09:12
	→ oehman joined	09:13
	← polymar left	09:13
	← gtuckerkellogg left	09:13
	→ angelsl joined	09:13
	← thirdknife left	09:14
	→ oznt joined	09:15
oznt	hi everyone	09:16
gitinfo	oznt: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	09:16
	← aspotashev left	09:17
oznt	i wonder why all my commits show always in the same branch ... how do i get a git log like here? http://gitready.com/intermediate/2009/01/26/text-based-graph.html?	09:17
	← dsirijus left	09:17
	→ wolftankk joined	09:18
_ikke_	git log --graph	09:18
cbreak-work	oznt: you'll need multiple branches for that	09:18
	then you have to commit in them, and later merge	09:19
_ikke_	!lol	09:19
gitinfo	git config --global alias.lol "log --oneline --graph --decorate"	09:19
cbreak-work	git log won't help you if they are in the same branch.	09:19
_ikke_	Nope	09:19
oznt	cbreak-work, i have two branches	09:19
_ikke_	and you probably want to add --all too	09:20
oznt	cbreak-work, and i have commited into both of them but git log --graph is just flat	09:20
cbreak-work	it shows history as it is	09:20
_ikke_	oznt: git log by default only shows commits reachable by HEAD	09:20
oznt	wait ... git log --graph --all is not flat	09:21
	that is a good start already	09:21
cbreak-work	all makes it show all branches/tags	09:21
oznt	so now if I merge the branches i will see that kind of graphs?	09:21
cbreak-work	maybe.	09:21
	← spyro left	09:21
	→ opus joined	09:22
oznt	yes ! it works :-)	09:22
	i love git , the more i use it i can become a better programmer	09:22
	this is really cool, thanks guys	09:22
	← eletuchy left	09:22
	→ spyro joined	09:24
	→ marthinal joined	09:25
	→ berserkr joined	09:25
	← oehman left	09:27
	→ tatsuya_o joined	09:27
	→ tvw joined	09:28
	← Fandekasp left	09:30
	← archdori left	09:30
	→ samtron1412 joined	09:31
	← samtron1412 left	09:32
	← TheHaven left	09:33
	← ijuma left	09:35
	→ gusnan joined	09:35
dsdeiz	hi kinda having a small issue with git submodules. i cloned a repo of mine which has a directory .oh-my-zsh. i moved .oh-my-zsh to oh-my-zsh, committed and pushed. right now the directory oh-my-zsh is empty. any ideas how to update the submodule? i have already modified .gitmodules file	09:36
	→ rvl joined	09:36
	← opus left	09:36
cbreak-work	git submodule update --init	09:36
dsdeiz	did that as well	09:36
	← angelsl left	09:40
	→ ahayweh joined	09:40
dsdeiz	i tried git submodule sync; nothing happened	09:40
FauxFaux	Is there a way to get gitk / gitg / git log --graph / etc. to hilight cherry-picks (i.e. common commits (by message would be fine)) between two parallel branches?	09:42
	→ angelsl joined	09:42
_ikke_	FauxFaux: You might fiddle with git cherry	09:43
	although, that's the opposite	09:43
	Well, both actually	09:44
	← thiago left	09:44
_ikke_	You want the commits prefixed with a -	09:44
	← palm_vue left	09:45
	← _iron left	09:45
_ikke_	But not sure if that's going to help	09:45
	← jargon- left	09:46
	→ palm_vue joined	09:46
FauxFaux	..and distracted by other work, sorry. I'll have a look later, thanks _ikke_.	09:47
	→ Rajesh joined	09:47
	Rajesh → Guest99058	09:47
	→ arietis joined	09:47
	← tiglog left	09:48
	→ babilen joined	09:48
	→ OOPMan joined	09:51
	← weiyang left	09:52
	← Nirvanko left	09:53
	→ Nirvanko joined	09:53
	→ rcu joined	09:54
	→ jargon- joined	09:55
	← jargon- left	09:55
	→ jargon- joined	09:55
Nirvanko	Who can explain how the rebase command is being performed? I don't understand why after rebasing the commit C3 becomes C3' like in the very first example http://git-scm.com/book/en/Git-Branching-Rebasing	09:55
charon	Nirvanko: it builds a new commit C3' that "looks like" C3, but in fact is a completely new commit. "looks like" is defined as having the same diff (as far as possible), the same message, etc.	09:56
	jargon- → jargon	09:56
charon	(it does not have the same parents!)	09:57
Nirvanko	charon, is that C3' is a difference between C3 and C4 commits?	09:57
	← jargon left	09:57
	→ highway900 joined	09:57
charon	Nirvanko: can you rephrase that? i'm not sure what you're asking	09:57
	→ jargon- joined	09:57
	← fireh left	09:58
_ikke_	Nirvanko: git doesn't store diffs, it stores the whole content of each object	09:58
Nirvanko	We rebase commit C3 onto C4 and get C3', so what is C3' again?	09:58
	← Tommy[D] left	09:58
_ikke_	Nirvanko: C3, but based on a new parent	09:58
	→ juvenal joined	09:59
charon	Nirvanko: let diff(C) be the diff of a commit C w.r.t. its (sole) parent. then rebase attempts to build a new commit C3' such that diff(C3') = diff(C3), thus making a commit that -- in human terms -- "is the same" as the old one, since it makes "the same change"	09:59
Sargun_Screen	How were the figures on git-scm.com generated?	09:59
charon	Nirvanko: like _ikke_ says, the diffs are all only a way of looking at commits, and not stored. they are merely a way of thinking about commits. so for git, C3' is something entirely different from C3	10:00
_ikke_	Nirvanko: git basically calculates a diff between C3 and C3^, applies that diff to the new parrent, and creates a commit with the same message and author data as C3	10:00
	→ Tommy[D] joined	10:01
	→ achadwick joined	10:03
	→ eijk joined	10:04
	→ jkremser joined	10:04
	← Guest99058 left	10:04
	← d3vic3 left	10:05
	→ mjt joined	10:05
	→ marhaban joined	10:05
	→ jkremser_ joined	10:05
	← Eising left	10:06
	→ d3vic3 joined	10:06
	← harish left	10:07
mjt	Hello. A question about workflow, sort of. Suppose there's a large project (like linux kernel for example) using git. I've a clone of Linus's repository. But I need _two_ or more "work trees" locally - to work on two separate branches and to be able to bisect something else without stopping my work in half-commit. How to keep just one (large) repository locally (and to sync it from remote only once) and allow multiple local working copies?	10:07
shruggar	mjt: I don't understand, why would you need separate work-trees?	10:07
charon	mjt: you can have a local "read-only" mirror repository of linus's, and then use the alternates mechanism	10:08
mjt	say, i'm developing something and hit a bug which I need to bisect. I don't want to commit or rollback my local changes which are in half-undefined-state	10:08
charon	mjt: note that alternates work even without such a read-only repo, but in this case they are very dangerous. you must never gc within such a repo	10:08
mjt	alternates?	10:09
	← rvl left	10:09
shruggar	mjt: you will not get any benefit from using git until you learn to commit on a whim	10:09
	mjt: history is mutable until it's public	10:09
charon	shruggar: well, i would want to do the same, and in fact i have occasionally made a quick clone in a temp location to bisect while i can hack in another terminal...	10:09
	i mean, it's not like bisecting is a 5-second issue	10:10
	mjt: i think it's in man gitrepository-layout	10:10
gitinfo	mjt: the gitrepository-layout manpage is available at http://git-scm.com/docs/gitrepository-layout.html	10:10
mjt	shruggar: it is not about committing often, there may be other reasons	10:10
shruggar	I just am always weary of anything that includes "I don't want to commit yet"	10:10
	→ lirt joined	10:10
charon	mjt: it corresponds to the --shared option to clone	10:10
	mjt: most of the caveats are spelled out in man git-clone, referring to --shared, but they apply to alternates in general	10:11
gitinfo	mjt: the git-clone manpage is available at http://git-scm.com/docs/git-clone.html	10:11
mjt	shruggar: say, i've all the .o files of a large project for MY branch/feature locally, but need to do a git bisect without recompiling everything with rigth options	10:11
	→ mklappstuhl joined	10:11
mjt	hmm. --shared.	10:11
	← timthelion left	10:12
charon	mjt: as i said, be VERY careful, and best stick to using a single, read-only mirror of upstream as the alternates repo	10:12
mjt	yeah. that might work	10:12
	← Guest5123 left	10:13
charon	of course, it'll only save you transferring the upstream work, you'll still have to shuffle your own stuff around by some other means	10:13
	→ ahayweh_ joined	10:13
charon	like perhaps another, not-read-only bare repo, that you are not using as an alternate	10:13
mjt	right now i've 6 separate/independent clones of qemu.org git repo here and 4 copies of linux git repo here, and i wanted to get rid of them somehow...	10:13
	thank you, very good suggestions indeed!	10:15
	← ahayweh left	10:16
	← hyperair left	10:16
	→ Fandekasp joined	10:17
	→ archdori joined	10:17
	→ rvl joined	10:17
	← marhaban left	10:18
	→ beautifulmind joined	10:21
	→ default joined	10:22
	default → Guest13529	10:22
	← gerard0 left	10:23
	→ LekeFly joined	10:23
	→ cybersphinx joined	10:24
	← ahayweh_ left	10:25
	← Guest53490 left	10:26
	→ gerard0 joined	10:27
	→ thiago joined	10:27
	← jkremser_ left	10:28
	→ spaceonline joined	10:29
	← wolftankk left	10:30
	← _Vi left	10:30
	→ rmr joined	10:30
	→ _Vi joined	10:31
	← mklappstuhl left	10:32
	→ eletuchy joined	10:33
	← gerard0 left	10:33
	→ jonnu joined	10:33
	→ codecowboy joined	10:33
	← echo-area left	10:34
codecowboy	hi. i am having a problem with git - i have just cloned a repo and changed 1 file. git status is showing that many more files are modified. what would cause this?	10:34
charon	autocrlf issues	10:35
	maybe somebody more competent than me knows the solution, but the problem is usually autocrlf :)	10:35
adhawkins	codecowboy: What does git diff show?	10:35
	→ wolftankk joined	10:35
cbreak-work	codecowboy: what does git config -l show? what does git diff -w show? and what OS are you using?	10:36
	→ gerard0 joined	10:37
	← eletuchy left	10:37
	→ Superlokkus joined	10:37
	← lausi left	10:39
	→ lausi joined	10:39
	→ cedeon joined	10:41
	→ RaceCondition joined	10:41
	← rchavik left	10:42
	← ReekenX left	10:42
codecowboy	cbreak-work, http://pastie.org/4493011	10:42
	← OOPMan left	10:43
codecowboy	cbreak-work, i'm using an Ubuntu VM. There are some devs using Windows (tortoise and git bash)	10:43
_ikke_	You probably have mixed case files inside the repo	10:44
codecowboy	cbreak-work, should i be doing gitt diff -w on an individual file?	10:44
	_ikke_, I have core.ignorecase=true in config	10:44
cbreak-work	pastebin git status	10:45
	← Araxia left	10:45
cbreak-work	git diff -w ignores whitespace changes	10:45
	if it shows diffs, then it's not whitespace related.	10:45
codecowboy	cbreak it shows diffs	10:45
	cbreak-work, ^^	10:45
cbreak-work	I mean the same.	10:46
_ikke_	codecowboy: mixed line ending I mean	10:46
cbreak-work	what kind of change does it show?	10:46
codecowboy	_ikke_, can i swtch that off in config	10:46
_ikke_	codecowboy: best way is to do that in a .gitattributes file in the repository	10:47
codecowboy	cbreak-work, i think git diff-w is showing the real changes i.e what i would expect from git status	10:47
cbreak-work	then it's whitespace related	10:48
codecowboy	_ikke_, sure but i dont control that	10:48
_ikke_	codecowboy: Don't control what?	10:48
	→ klj613 joined	10:48
codecowboy	_ikke_, the git repo. i cant make a change like that without getting it approved etc	10:48
	→ jjido joined	10:48
klj613	how do i git diff which would ignore pulled changes from master? (merge commits)	10:48
	← babilen left	10:49
codecowboy	_ikke_, cbreak-work is there a way to resolve this locally for now?	10:49
	← rvl left	10:49
_ikke_	codecowboy: git config core.autocrlf false	10:49
	← dsdeiz left	10:49
	← angelsl left	10:51
codecowboy	_ikke_, done but git status is still showing loads of changed files	10:51
	← d3vic3 left	10:51
	→ PigeonFriend joined	10:51
	← shreya left	10:51
	→ harish joined	10:53
	→ d3vic3 joined	10:53
	→ Lord_DeathMatch joined	10:55
	← oznt left	10:56
	← gerard0 left	10:56
	→ webguynow joined	10:57
	→ rvl joined	10:58
	→ angelsl joined	10:58
	→ gerard0 joined	10:59
	← juvenal left	11:00
	← madewokherd left	11:00
	← spaceonline left	11:01
	← quazimodo left	11:03
codecowboy	_ikke_, think you meant true	11:04
	← Chappo left	11:04
	→ quazimodo joined	11:05
	→ marhaban joined	11:05
	← lng left	11:06
	→ Chappo joined	11:06
	→ joaquin46 joined	11:07
	→ ZorgHCS joined	11:07
axrfnu	hi all. i have a local and a remote branch pointing to the same commit. how can i make the local branch automatically push to its remote counterpart on origin without editing .git/config?	11:07
	← joaquin46 left	11:07
_ikke_	axrfnu: You mean without yourself editing .git/config?	11:08
axrfnu	_ikke_: yes, exactly	11:08
_ikke_	"git push -u origin master" would do that for the master branch for example	11:09
	-u -> --set-upstream	11:09
	Which causes git to record that information into .git/config	11:10
axrfnu	d'oh, that was too obvious. i already had -u in mind, but just had used it to set up a fresh local tracking branch. _ikke_: thank you for the hint	11:10
	→ shreya joined	11:12
_ikke_	axrfnu: git checkout has -t which does the same when you want to create a new local branch based on a remote branch	11:12
	← wolftankk left	11:12
	← gerard0 left	11:14
codecowboy	_ikke_, cbreak-work seems to be ok now. thanks for your help!	11:14
	→ longbeach joined	11:15
	→ dsirijus joined	11:16
	← rmr left	11:17
	→ rwestphal joined	11:17
	→ iamjarvo joined	11:17
	→ wolftankk joined	11:17
	← wolftankk left	11:17
	→ nicoulaj joined	11:17
	→ gerard0 joined	11:18
	← tinti_ left	11:18
	→ chester_ joined	11:18
	→ thirdknife joined	11:19
	← rvl left	11:19
	→ TomyLobo joined	11:19
	→ rmr joined	11:19
	→ paolooo joined	11:20
paolooo	hello @all, how do i update my local copy? Is it $ git pull master?	11:21
	→ Geralt joined	11:21
wereHamster	paolooo: git pull.	11:22
_ikke_	paolooo: The second parameter for pull is the remote where to pull from	11:22
	← xiangfu left	11:23
	→ fusion27 joined	11:24
fusion27	First time trying to work with submodules, not even sure that submodules are what I want	11:24
_ikke_	!submodule	11:24
gitinfo	git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html	11:24
fusion27	I'm using a JS lib that I got from github	11:24
	← thirdknife left	11:24
fusion27	I'm still so new to source control	11:24
	and of course, git	11:25
	tried using another lib I got from git a while ago, but thought I 'un-did' it as a submodule	11:25
	was coming around with the new one to take another stab at it	11:25
	git submodule init	11:25
	No submodule mapping found in .gitmodules for path 'bootbox'	11:25
	bootbox is the old one I thought I undid	11:25
_ikke_	fusion27: If you want to add a submodule, git submodule add it	11:26
fusion27	bootbox has no records in .gitmodules	11:26
	yeah _ikke_ I did that in our development repo	11:27
	we then pull that to prod	11:27
paolooo	wereHamster: _ikke_ Thanks :)	11:27
fusion27	but when I got to prod, it looked like it pulled the directory over but not the source	11:27
paolooo	wereHamster: _ikke_ do I need to add master? -> $ git pull master?	11:27
_ikke_	wereHamster: no	11:27
	paolooo: *	11:27
fusion27	so that's why I ran git submodule init	11:28
paolooo	$ git pull <-- would be fine?	11:28
	→ rvl joined	11:28
fusion27	I suppse I need to be doing this in our development repo?	11:28
	seems like the more I learn with git, the less I know	11:29
	:D	11:29
_ikke_	fusion27: What is the difference between the production repo and the development repo?	11:29
	paolooo: Yes, but it relies on some settings to be set	11:29
paolooo	_ikke_: ah ok thanks	11:30
_ikke_	(for master it already works)	11:30
	But not that it merges into the current branch	11:30
	note*	11:30
paolooo	_ikke_: Thanks	11:30
	← zdk left	11:30
paolooo	_ikke_: How do I check if there's a changes on the git files?	11:31
fusion27	_ikke_: you're talking to a total noobie under the thumb of a micromanager who doesn't like us 'messing' with git, the development repo is firewalled, production is not	11:31
	the other developer and I realized that running with no source control was stupid	11:31
	that was about 99% of the reason we chose git	11:32
	I suppose pretty much anything would've worked	11:32
	just heard a lot of great things about git	11:32
_ikke_	Who added the submodule to the repository?	11:32
	→ arkiver joined	11:33
	→ unmanbearpig joined	11:33
	→ Jippi joined	11:33
fusion27	I did	11:33
	bootbox, I think I did a git clone	11:33
	← _sundar_ left	11:34
	← rvl left	11:34
	→ asafdav2 joined	11:36
	gitinfo set mode: +v	11:37
	← Chappo left	11:38
	← berserkr left	11:38
	→ timthelion joined	11:38
	→ Chappo joined	11:40
_ikke_	fusion27: To what repository did you add that submodule?	11:40
	→ thirdknife joined	11:41
	→ FreeArtMan joined	11:41
	← chrisward left	11:41
	→ subbyyy joined	11:42
	← subbyyy left	11:42
	← rcu left	11:42
	→ Ryback_ joined	11:43
	← Ryback_ left	11:43
	→ Ryback_ joined	11:43
	→ chuckharmston joined	11:43
	← Psycho_pr left	11:43
	→ Psycho_pr1 joined	11:43
	← leeN left	11:44
	→ chrisward joined	11:44
	← dsirijus left	11:45
	→ jjbohn joined	11:46
	→ leeN joined	11:48
	← toabctl left	11:49
	← ology left	11:52
	← gerard0 left	11:53
	→ thirdknife_ joined	11:55
	→ rurufufuss joined	11:56
	← thirdknife left	11:56
	→ gerard0 joined	11:57
	→ barq joined	11:57
	← C8H10N4O2 left	11:59
	→ toabctl joined	12:01
	→ hyperair joined	12:02
fusion27	Hi _ikke_ I added it to our development repo	12:03
	→ jackh joined	12:03
fusion27	then pulled to prod	12:03
_ikke_	fusion27: Have you commited that in the development repo?	12:03
	← gerard0 left	12:04
	← Guest13529 left	12:05
	← jaclinuxhelp left	12:05
	← maleknet left	12:06
	→ geekbri joined	12:07
	→ gerard0 joined	12:08
	→ TRUPPP joined	12:08
	← beautifulmind left	12:09
	← archdori left	12:12
	← Fandekasp left	12:12
	→ beautifulmind joined	12:12
	→ guampa joined	12:12
fusion27	_ikke_: I think one of us did a few days ago, but we 'rewinded' that but checking out back to a point in history that kept our repos a mirrors	12:13
	← beautifulmind left	12:13
	← TRUPPP left	12:14
	← gerard0 left	12:14
	← rmr left	12:14
	← Lord_DeathMatch left	12:14
	← highway900 left	12:14
	← hyperair left	12:16
	← jjido left	12:16
	→ oehman joined	12:17
	→ gerard0 joined	12:18
	← thirdknife_ left	12:18
	→ t0rc joined	12:19
	← t0rc left	12:21
	← landypro left	12:23
	→ skyf joined	12:23
	→ ChadStudmuffin joined	12:23
	→ landypro joined	12:23
	← gerard0 left	12:24
	← jjbohn left	12:25
	→ gerard0 joined	12:25
	→ lupo7 joined	12:26
	→ tom[] joined	12:26
	→ AaronMT joined	12:27
	→ juvenal joined	12:27
	→ freeed joined	12:29
	← _W_ left	12:30
	← Chibby_ left	12:31
	← _Vi left	12:31
	→ thirdknife_ joined	12:31
	→ ddelrio1986 joined	12:31
	← gerard0 left	12:31
	→ _Vi joined	12:32
	→ gerard0 joined	12:33
	← paolooo left	12:34
	→ Fandekasp joined	12:35
	→ archdori joined	12:36
	→ babilen joined	12:36
	← cagedwisdom left	12:38
	← gerard0 left	12:39
	→ Lord_DeathMatch joined	12:39
	→ sh4rm4 joined	12:40
	→ joephelius joined	12:40
	→ gerard0 joined	12:40
klj613	how do i git diff which would ignore pulled changes from master? (merge commits)	12:41
	→ alanmac_ joined	12:41
_ikke_	klj613: care to elaborate?	12:41
klj613	`git diff master feature-branch` shows the code from the master branch pulled in	12:41
	← shruggar left	12:41
	← alanmac_ left	12:41
cmn	it shows the differences between master and feature-branch, yes	12:42
	← babilen left	12:42
	→ shruggar joined	12:42
cmn	what are you trying to see?	12:42
	→ alanmac joined	12:42
klj613	and let say half way through feature-branch i merged master in. to keep up to date	12:42
	it shows the master code changes aswell	12:42
alanmac	How do I get git to always annotate a tag? tag = tag -a in .gitconfig doesn't work	12:42
	← tom[] left	12:42
cmn	it really shouldn't show them	12:43
	← marthinal left	12:43
klj613	alanmac, wouldnt that cause a loop? lol	12:43
cmn	alanmac: you can't alias builtin commands	12:43
alanmac	pull = pull --ff-only works but?	12:43
cmn	it really shouldn't work	12:43
	← schemula left	12:43
cmn	have you ever seen it refuse to do a non-ff update?	12:43
	merge, rather	12:43
alanmac	yep	12:44
	→ schemula joined	12:44
alanmac	that's the point though I want to know when it's not ff	12:44
	← Superlokkus left	12:44
cmn	actually, nevermind, that's right	12:44
	pull isn't a builtin	12:44
alanmac	I see	12:44
klj613	so git diff 'should' ignore the code merged in from master?	12:44
cmn	if the code is the same in both branches, then the diff shouldn't show them	12:44
	there isn't a way to show it	12:44
	→ t0rc joined	12:46
	→ C8H10N4O2 joined	12:46
	→ hyperair joined	12:46
	→ Guest31840 joined	12:46
	← sh4rm4 left	12:47
	→ beautifulmind joined	12:48
	→ TheJH joined	12:48
klj613	okay, thanks.	12:48
	→ asafdav21 joined	12:50
	← arvind_khadri left	12:51
	→ weiyang joined	12:51
	→ om joined	12:51
	→ tommyvyo joined	12:51
	← asafdav2 left	12:52
	→ sh4rm4 joined	12:52
	→ andref_dias joined	12:52
	← fusion27 left	12:53
	gitinfo set mode: +v	12:54
	← beautifulmind left	12:54
om	hello	12:54
gitinfo	om: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	12:54
om	git beginner here, i sent a pull request without setting my name and email correctly, how can i update the pull request so it shows my proper name and email?	12:55
	someone sent me an email saying this: "After re-commit, you can force push to this branch to update this pull request." what does this exactly mean and what should I do?	12:55
	i've updated my name and email using git config, but now what?	12:55
FauxFaux	om: First, fix the commits so that git log looks right. !postpro	12:56
gitinfo	om: [!postproduction] So, you want to make your commit history look pretty before pushing? http://sethrobertson.github.com/GitPostProduction talks you through how to use 'rebase -i' to do this.	12:56
wereHamster	om: push your new commits into th same branch that the pull request references	12:56
FauxFaux	Mmm, that doesn't mention resetting --author.	12:56
shruggar	if it's only one commit, then git commit --amend --reset-author would do it	12:56
	→ beautifulmind joined	12:57
om	ok i did that shruggar, now i should push the stuff to my fork?	12:58
	and what did the person who sent me the email meant by 'force push'?	12:59
	← beautifulmind left	12:59
	← iori left	12:59
_ikke_	om: git push -f	12:59
	om: git push -f <remote> <branch>	13:00
om	and i should do this after i've pushed the stuff to the fork from my repository?	13:00
	→ iori joined	13:00
	→ nwest joined	13:00
	→ Pugnax joined	13:00
_ikke_	om: No, you should do this while you are pushing the stuff to the fork of your repository	13:01
	← iori left	13:01
_ikke_	om: Because you have rewritten your commits, git won't accept that immediately	13:02
	← ChadStudmuffin left	13:02
	→ elux joined	13:02
	← skyf left	13:03
shruggar	and just so you know, om, these instructions shouldn't be taken as general advice. Everyone here is making assumptions about the simplicity of your situation, as your question implies that you probably haven't been using git for long enough to have a complicated one :)	13:03
	→ ChadStudmuffin joined	13:04
	→ chris2 joined	13:05
	← sh4rm4 left	13:05
	→ sh4rm4 joined	13:06
	→ jatt joined	13:06
	→ gfontenot joined	13:06
	← landypro left	13:06
	← gfontenot left	13:06
	→ ehsan joined	13:06
	← psoo left	13:07
	→ arvind_khadri joined	13:07
	← spyro left	13:08
	→ paolooo joined	13:08
paolooo	hi @all, I tried to $ git pull <-- but I get an error: " Please, commit your changes or stash them before you can merge.	13:08
	→ giallu joined	13:08
	← weiyang left	13:08
	← tjadc left	13:09
_ikke_	paolooo: Well, what does that message say?	13:09
	→ macabre joined	13:09
	→ centipedefarmer joined	13:09
	→ pmdarrow joined	13:09
paolooo	_ikke_: "Your local changes to the following files would be overwrittern by merge:... "	13:10
_ikke_	yes	13:10
	You have a dirty work tree	13:10
FauxFaux	(hurr)	13:10
	→ Chib joined	13:11
paolooo	_ikke_: yes I want to update my local files.	13:11
	→ spyro joined	13:12
	→ Dimitr joined	13:12
	→ chrisdev joined	13:12
FauxFaux	paolooo: That's great. You still have to do what it says before doing that, though.	13:12
canton7	paolooo, git really doesn't like overriting uncommitted stuff, as it's a situation which can't be recovered from, and can very easily lead to lost work	13:12
	→ macIvy joined	13:12
canton7	so either commit your changes, stash them, or if you really don't want them, nuke them	13:13
	→ skyf joined	13:13
canton7	s/overriting/overwriting	13:13
_ikke_	paolooo: note that git pull is not necessarily meant for just getting the latest changes	13:13
	paolooo: It would create a merge commit everytime you do, which would make history quite messy in time	13:13
om	_ikke_: i've pushed the files to my repository, now, how do i update the pull request?	13:14
canton7	om, if you force-push to the branch on which a pull request has been built, the pull request gets updated	13:14
	← ehsan left	13:15
paolooo	_ikke_: so, I should commit the files first, then pull to overwrite my files?	13:15
	→ neurodrone joined	13:15
_ikke_	paolooo: You don't want to overwrite your files, do you?	13:15
	← hyperair left	13:15
	→ fb6667 joined	13:16
	→ ehsan joined	13:16
canton7	pull won't overwrite them. it will take your version of the history with your changes to those files, and the remote version, and merge them together	13:16
	← Twey left	13:16
paolooo	_ikke_: Yes, I want to overwrite because somebody modified the file, and I want his changes to appear on my local files.	13:16
	← fb6667 left	13:16
	→ cannonball joined	13:17
	→ hyperair joined	13:17
	← iamjarvo left	13:17
	← borwar left	13:17
	→ linoj joined	13:17
	→ afuentes joined	13:18
	→ iamjarvo joined	13:18
_ikke_	paolooo: But do you want to loose you're changes?	13:18
	→ beautifulmind joined	13:18
paolooo	_ikke_: yes I want to loose the changes	13:18
	_ikke_: or perhaps I'll just delete the file and then pull	13:19
canton7	paolooo, 'git status' will tell you how to undo your changes	13:19
_ikke_	paolooo: That won't help either	13:19
	← palm_vue left	13:20
	← bizhanMona left	13:20
	← Lgb left	13:20
	← ehsan left	13:20
charon	i conjecture that "X and then pull" is the wrong solution to any problem, for all X	13:21
	→ _oakdog8 joined	13:21
	oakdog8 → Guest71934	13:21
	← _oakdog8 left	13:22
	→ _oakdog8 joined	13:22
	← _oakdog8 left	13:22
	→ singingwolfboy joined	13:22
	→ Twey joined	13:22
	→ palm_vue joined	13:23
	→ Lgb joined	13:23
singingwolfboy	On GitHub, I have a pull request which has since been merged, but I can still view the commits that were in that pull request. How do I view an equivalent list of commits using git on the command line?	13:23
	→ bizhanMona joined	13:23
	→ wolftankk joined	13:24
FauxFaux	git ls-remote origin; git fetch origin refs/pull/528937/merge; git log FETCH_HEAD	13:24
	→ dv310p3r joined	13:24
grawity	wouldn't that just be `git log origin/pull/528937/merge`	13:24
canton7	the orgin/pull stuff isn't fetched by default, so you need to explicitely fetch it	13:24
FauxFaux	What he said. grawity: !fetch4	13:24
gitinfo	[!fetchfour] Never use the four-word version of git-fetch or git-pull (e.g. git fetch remote refspec). It always ends in tears. Yes, if you understand the implications of FETCH_HEAD it can technically be done, but really it is easier to just fetch the whole remote (or perhaps edit the fetchspec if you never want other bits). If you must, see !fetch4why	13:24
grawity	FauxFaux: and I used the four-word version when?	13:25
FauxFaux	grawity: I did, because, as canton7 said, they're not fetched by the safe version, because github are Mean.	13:25
_ikke_	because they're in a different namespace, right?	13:27
	← macIvy left	13:27
	→ gfonteno_ joined	13:27
	nischayn22 → nischayn22\|Away	13:28
canton7	'git fetch' fetches (by default) refs/heads/ . pull/x is indeed a different namespace entirely	13:28
	← gfonteno_ left	13:28
	→ gfontenot joined	13:28
	→ babilen joined	13:28
	→ Kun joined	13:28
canton7	oops, refs/pull/x not pull/x	13:28
	← singingwolfboy left	13:28
	← Kun left	13:29
	→ orkz_ joined	13:29
	→ andrewhl joined	13:30
	← thirdknife_ left	13:30
	← shawngo left	13:31
	← _Vi left	13:32
	← kermit left	13:32
	← orkz left	13:32
	orkz_ → orkz	13:32
	→ OOPMan joined	13:32
	→ _Vi joined	13:32
	→ ShadeTornado joined	13:32
	→ kermit joined	13:33
	→ denysonique joined	13:33
	→ chrisgeorge joined	13:33
	← wolftankk left	13:33
	→ nicklas_ joined	13:34
	→ weiyang joined	13:35
	→ mmc1 joined	13:35
	← nicklas_ left	13:36
	→ mikewintermute joined	13:36
	→ nicklasbj joined	13:36
	gitinfo set mode: +v	13:37
	→ _oakdog8 joined	13:38
	→ wolftankk joined	13:38
	→ john2496 joined	13:38
	← _oakdog8 left	13:39
	→ _oakdog8 joined	13:39
	_oakdog8 → oakdog8	13:39
	← oakdog8 left	13:39
	→ TeckniX joined	13:39
	→ yko joined	13:39
	→ dranov joined	13:39
	→ oakdog8 joined	13:39
	← rurufufuss left	13:40
	→ ZyX-I joined	13:40
	gitinfo set mode: +v	13:41
nicklasbj	Hi All. Is it possible to run git outside of a git repo and specify the path? 'git checkout XXXX path' returns fatal: Not a git repository.	13:41
	→ Davey joined	13:41
shruggar	nicklasbj: if you weren't trying to check-out, I'd say yet	13:41
	*yes	13:41
	← barq left	13:41
shruggar	it's a bit more convoluted if you want to check out. What are you actually trying to do?	13:41
	← dranov left	13:42
shruggar	Do you actually want "git archive", perhaps?	13:42
FauxFaux	nicklasbj: man git GIT_WORK_TREE and GIT_DIR, and lots of caution. You're probably trying to write !deploy, though.	13:42
gitinfo	nicklasbj: Git is not a deployment tool. You can build one around it for simple environments. http://sitaramc.github.com/the-list-and-irc/deploy.html	13:42
	nicklasbj: the git manpage is available at http://git-scm.com/docs/git.html	13:42
nicklasbj	shruggar: I'm new to git, so maybe :) I'm trying to do something similar to svn update -rXX	13:42
FauxFaux	nicklasbj: That doesn't work outside of an svn repository..	13:42
	s/repository/working copy/	13:43
	← Guest31840 left	13:43
	← yko left	13:43
nicklasbj	FauxFaux: It is a git repo, I just want to update it to a specific hash without having to cd into the folder(I'm calling the command from an external place cd is difficult)	13:44
shruggar	nicklasbj: "trying to do something similar to svn <….anything…>" is not likely to be sane ;)	13:44
canton7	(cd repo && checkout whatever) maybe?	13:44
FauxFaux	nicklasbj: You know that you can run (cd /foo/bar && git checkout blah), with the brackets, and the shell will cd back for you afterwards, right?	13:44
shruggar	I don't actually know what that command does, though. What, again, are you actually trying to do?	13:44
FauxFaux	Bah!	13:44
canton7	victory! :)	13:45
	→ fireh joined	13:45
	→ revagomes joined	13:45
	FauxFaux dispatches the assasins.	13:45
nicklasbj	hehe.. Sorry for any confusion.. I'm calling the git command from a java program and I can't seem to get cd to work there	13:45
shruggar	canton7: I hereby award you 0.03 internet points	13:45
canton7	yours was more informative, I swear, I don't deserve to die!	13:45
cmn	then just do the same, cd into wherever and call git, then cd back up	13:46
FauxFaux	nicklasbj: Have you considered using jgit?	13:46
cmn	jgit may also be able to use this	13:46
	→ Superlokkus joined	13:46
	→ v0n joined	13:46
nicklasbj	FauxFaux: looks great	13:46
charon	nicklasbj: note that checking out an old version results in !detached	13:46
gitinfo	nicklasbj: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	13:46
	← tommyvyo left	13:47
	→ dozymoe joined	13:47
nicklasbj	FauxFaux: thanks!	13:47
FauxFaux	That's exactly what svn update -r does, fwiw, except of course you can only ¡float yourself back into sanity. Such a sad version control system.	13:47
mkj_wk	nicklasbj: WRT the "cd command" not working, cd isn't actually a command. You need to change the working directory from within your Java program before invoking the Git command	13:48
nicklasbj	It's ok if it is detached, I'm only crawling the version system now.	13:48
	mkj_wk: ah ok, makes sense	13:48
mkj_wk	if you want to use jgit be aware of its limitations, like not honoring (most of?) .gitattributes	13:48
	→ Destos joined	13:48
cmn	crawling?	13:48
mkj_wk	jgit should work fine for inspecting what's there	13:48
	← dozymoe left	13:49
nicklasbj	cmn: I'm calculating metrics for each commit for visualizing it	13:49
cmn	there is no need to call checkout	13:49
mkj_wk	it's when you come to committing that we've had, um, challenges with some people committing with egit (on top of jgit) and others with command line git (directly or under ever other ui)	13:49
	← fireh left	13:49
cmn	if you just want to look through commits, use jgit to look at what's there	13:49
	← AaronMT left	13:50
nicklasbj	Yes, jgit should be great for my purpose	13:50
cmn	there is absolutely no need to mess with the working directory, or even to have one	13:50
	→ psoo joined	13:50
mkj_wk	agreed	13:50
nicklasbj	Thanks all for your help	13:50
	→ jjbohn joined	13:51
	← tatsuya_o left	13:51
	→ nifty_ joined	13:51
	→ tatsuya_o joined	13:52
	← weiyang left	13:54
	← jjbohn left	13:55
	→ sawdisk joined	13:55
	→ xMYTHICx joined	13:56
	← nicklasbj left	13:58
	→ alexisabril joined	13:59
	← mikewintermute left	13:59
	→ Dimitr\|2 joined	13:59
	→ mikewintermute joined	13:59
	→ aldodelgado joined	13:59
	→ ology joined	14:00
	→ ehsan joined	14:01
	← Dimitr left	14:03
	→ Ecesis joined	14:03
	→ Ecesis1 joined	14:04
	← Dimitr\|2 left	14:04
	← Ecesis left	14:05
	→ Dimitr joined	14:05
	→ zackiv31 joined	14:05
	← mikewintermute left	14:06
	→ mikewintermute joined	14:06
	→ ahayweh joined	14:06
	← spronk left	14:06
	→ tukkol joined	14:07
	→ dozymoe joined	14:07
	gitinfo set mode: +v	14:07
	→ spronk joined	14:07
	← spronk left	14:07
	← mikewintermute left	14:08
	← gavin_huang left	14:08
	← ph^_ left	14:08
	→ mikewintermute joined	14:08
	← beautifulmind left	14:09
	← dozymoe left	14:10
	← kcj left	14:10
	← v0n left	14:10
	→ dozymoe joined	14:11
	→ jjbohn joined	14:13
	← cmn left	14:16
	→ marthinal joined	14:17
	→ weiyang joined	14:17
	→ cmn joined	14:17
	← marthinal left	14:19
	← jackh left	14:19
	→ rchavik joined	14:19
	→ marthinal joined	14:19
	← lb left	14:19
	→ blexim joined	14:21
	→ fly9 joined	14:21
	← weiyang left	14:21
	← kanru_ left	14:21
	← psoo left	14:21
	gitinfo set mode: +v	14:22
	→ v0n joined	14:23
blexim	hi, is this a reasonable channel to ask really basic git questions?	14:23
corburn	yup	14:23
blexim	sweet	14:23
	so, I have a master branch & a couple of branches hanging off of it	14:24
	in one of the non-master branches I've made some changes to a directory	14:24
	→ dhruvasagar joined	14:24
blexim	but now I'd like to revert just those changes to look like master and leave all the other directories alone	14:24
Nirvanko	Hoe to tell git to remove untracked directories?	14:25
blexim	should I be reading the checkout manpages, or reset or what?	14:25
Nirvanko	How*	14:25
cmn	if they're untracked, git doesn't care about them; use /bin/rm	14:25
	if there's a lot, you can use git clean	14:25
	→ kanru_ joined	14:25
	→ Dimitr\|2 joined	14:26
	← Ecesis1 left	14:26
	→ cwang joined	14:26
PerlJam	blexim: you can use "git checkout" to make that dir look as it did in a specific commit	14:26
	← jonnu left	14:26
blexim	ok & that'll keep me in the same branch as I started in?	14:26
PerlJam	blexim: git checkout master -- path/you/want/to/change	14:27
blexim	awesome, thanks	14:27
	→ Rajesh joined	14:28
	Rajesh → Guest86229	14:28
Nirvanko	cmn, git clean does not remove directories. If it had done this I wouldn't ask.	14:28
cmn	yes, it does, see man git clean	14:28
gitinfo	the git-clean manpage is available at http://git-scm.com/docs/git-clean.html	14:28
	← Dimitr left	14:29
PerlJam	Nirvanko: git clean -dxf # caveat emptor	14:29
tukkol	hi. is there a way to see merge commits in an individual file log? i.e. in branchA file X is changed and I merge it into branchB then I can only see the merge commit when using 'git log' not with 'git log X'	14:29
PerlJam	well, I guess you aren't buying anything so ... caveat lector	14:29
	→ shawngo joined	14:29
	→ ovnicraft joined	14:29
	← rchavik left	14:30
	→ Jay_Levitt joined	14:30
	→ rchavik joined	14:31
	→ luxurymode joined	14:31
	← ShadeTornado left	14:32
	← _Vi left	14:32
	→ ShadeTornado joined	14:32
	→ _Vi joined	14:33
	← sh4rm4 left	14:33
	← Tommy[D] left	14:33
	→ Tommy[D] joined	14:33
	→ sh4rm4 joined	14:34
	→ jmd joined	14:34
	→ nickpresta joined	14:34
	← d3vic3 left	14:35
cmn	tukkol: --full-history might help	14:35
	← TheJH left	14:35
	← Vert left	14:36
	→ d3vic3 joined	14:36
	← ShooterMG left	14:37
Nirvanko	PerlJam, I think your Latin isn't that good as your knowledge of Git.	14:37
	← skyf left	14:37
	→ skyf_ joined	14:37
	← Kulrak left	14:38
	→ axl_ joined	14:39
	→ flavius joined	14:39
	← gxk left	14:39
PerlJam	Nirvanko: you're right, I should have said "hic sunt leones" or something :)	14:40
	← Guest86229 left	14:40
	→ mikecmpbll joined	14:40
mikecmpbll	hi chaps.	14:40
gitinfo	mikecmpbll: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	14:40
	← rchavik left	14:41
	← Pugnax left	14:41
mikecmpbll	can i just run this by someone just for peace of mind. I want to push a local branch that i've been working on to remote repo, should i do: git push -u <branch-name>	14:41
	→ mattalexx joined	14:42
	← mattalexx left	14:42
	→ mattalexx joined	14:42
jast	mikecmpbll: git push -u <remote name> <branch name>	14:42
mikecmpbll	so remote-name might be origin, for example?	14:42
jast	yes	14:42
mikecmpbll	and branch name is the name that I want it to be on remote? or the name of my local branch?	14:43
	← tvw left	14:43
	→ weiyang joined	14:44
cmn	both, it's a refspec	14:44
	if you only specify the local name, git assumes you want it to be called the same	14:44
mikecmpbll	gotcha	14:44
	→ rchavik joined	14:44
	← freeed left	14:45
	→ Kulrak joined	14:45
mikecmpbll	i'll give it a shot, git makes me silly nervous.	14:45
	← papegaaij left	14:45
	← toabctl left	14:46
tukkol	cmn: unfortunately --full-history does not help :( .. the merge is still not visible for an individual file...	14:47
	← hyperair left	14:47
	← xMYTHICx left	14:47
mikecmpbll	phew, success, thanks guys :).	14:47
	→ spiffytech joined	14:47
	← vdv left	14:48
cmn	then there's probably not way of doing it, other than by hand	14:48
	← LekeFly left	14:48
	→ freeed joined	14:48
cmn	if the file wasn't modified in the merge, there isn't any reason for git to show it	14:48
	← phreaky left	14:50
spiffytech	I remember seeing a git config option that would print intra-line diffs on one line, with the old and new substrings denoted inside the single line, instead of one entire line for the old version of the line and a second line showing the entire new version, but I can't remember the option. What is it?	14:50
	← Jay_Levitt left	14:50
	→ macIvy joined	14:50
cmn	--word-diff	14:50
	→ AaronMT joined	14:50
jast	sitaram: in our SCM stuff I'm currently planning to add a button to a web interface to create a new gitolite-controlled repository. is there a straightforward way to run a command on the server to make gitolite create a new repo? or am I stuck creating it manually and then running "gitolite setup"?	14:51
cbreak-work	--color-words is also interesting	14:51
	← ddelrio1986 left	14:51
	← rchavik left	14:51
spiffytech	Ah, cool, thanks!	14:51
cbreak-work	jast: how about changing the config file and then commit & push?	14:52
jast	changing what config file?	14:52
	→ Rajesh joined	14:53
cmn	the one that tells gitolite to create it?	14:53
jast	there is no config file that creates it	14:53
tukkol	cmn: but there has to be a way to determine when a change for a particular file became active in a given branch or not?	14:53
	Rajesh → Guest35519	14:54
	← oehman left	14:54
cmn	what about gitolite-admin/conf/gitolite.conf ?	14:54
	tukkol: yes, and that's shown by log, the merge isn't that commit	14:54
jast	gitolite.conf doesn't list the repository	14:55
cmn	oh, wild repos	14:55
jast	yes	14:55
	→ mritz joined	14:55
	→ maestrojed joined	14:55
	← AaronMT left	14:55
	→ Synthead joined	14:56
	→ ddelrio1986 joined	14:56
cmn	tukkol: if you want to pretend that side-branches didn't happen, --first-parent can help	14:56
	but only if you've made sure to make the merges be significant	14:57
	→ jackh joined	14:57
cmn	tukkol: basically, your definition of branch clashes with git's	14:58
	← Guest35519 left	14:58
	← archdori left	14:58
	← Fandekasp left	14:58
mikecmpbll	now that I've done git push -u <remote name> <branch name>, and the branch now exists on remote, can I use git push and git pull?	15:00
	→ TheLZA joined	15:01
	← TheLZA left	15:01
	← mritz left	15:01
	→ TheLZA joined	15:02
	← asafdav21 left	15:02
	← LimeBlast left	15:02
charon	mikecmpbll: pull yes. push's behavior depends also on push.default, see man git-config	15:02
gitinfo	mikecmpbll: the git-config manpage is available at http://git-scm.com/docs/git-config.html	15:02
mikecmpbll	charon: ta, i'll take a look.	15:03
	← Lord_DeathMatch left	15:04
	← schemula left	15:04
	→ john_anderson joined	15:05
	→ codetroll joined	15:05
	→ LimeBlast joined	15:05
	← andref_dias left	15:05
	← flijten left	15:05
	→ russt joined	15:06
	→ Rajesh_ joined	15:06
	← Rajesh_ left	15:07
	→ Rajesh_ joined	15:07
	→ calvinx joined	15:07
	→ Fangorn_ joined	15:07
	← codetroll left	15:07
	← Rajesh_ left	15:08
	→ codetroll joined	15:08
calvinx	Is anyone here familiar with git flow? I would like to find out how git flow release finish "my_tag_name" works because I am expected a permanent tag to be added into the master branch but this isn't so.	15:08
	→ yoklov joined	15:08
	→ Rajesh_ joined	15:08
calvinx	how do I ensure that when I finish a release, a tag gets added to my "master" and "develop" branches?	15:09
	← jatt left	15:09
	← freeed left	15:10
	← Chappo left	15:10
canton7	You should get a tag added to the master branch	15:11
cbreak-work	tags can't be added to branches	15:11
	→ Fandekasp joined	15:11
	→ Chappo joined	15:11
cbreak-work	just git tag something if you want to tag it	15:12
	→ Araxia joined	15:12
	← cbreak-work left	15:12
canton7	sorry, to the merge commit on the master branch	15:12
	← proradii left	15:13
	← Fangorn_ left	15:13
	→ freeed joined	15:13
	→ duckxx joined	15:14
	← arkiver left	15:14
	← Rajesh_ left	15:14
	← Kaisyu left	15:15
	→ jasmina87SA__ joined	15:16
	→ ISF joined	15:16
	← OOPMan left	15:16
	← jasmina87SA__ left	15:16
	← iamjarvo left	15:18
tukkol	cmn: hm. i don't see where the conflict in my interpretation of a branch is yet :( . i guess it's actually more about how a merge is defined. in my interpretation a merge (no-ff) is the topological point when two (or more) branches are joined and thus the commits of the other branches becomes active.	15:18
	← Dimitr\|2 left	15:18
	→ eijk_ joined	15:18
	← eijk left	15:18
	→ iamjarvo joined	15:18
cmn	active isn't something that exists	15:18
	you assume that the other branch is separate	15:18
	when you merge, it becomes part of the branch	15:18
	→ codeM0nK3Y joined	15:18
cmn	thus the merge commits isn't where anything becomes "active"	15:19
	you're simply splicing two histories; the changes happened whenever, not in the merge	15:19
tukkol	well it becomes "visibile" from the point of view of the branch you merge into.	15:19
	→ rgsteele joined	15:20
	← Araxia left	15:20
cmn	it becomes reachable, sure	15:20
	→ private joined	15:20
cmn	but it doesn't affect where things happened	15:20
	→ tplaner_ joined	15:20
	private → Guest5287	15:20
	← centipedefarmer left	15:20
	← unmanbearpig left	15:21
	← john2496 left	15:22
	→ john2496 joined	15:22
	→ localhost joined	15:22
	→ thelazydeveloper joined	15:22
	→ zastern joined	15:23
tukkol	still it's part of the topology of the branch and thus relevant for the history and as it affects all files since the last common anchestor it would make sense to see it somehow.. ?!	15:23
cmn	it doesn't affect all files	15:23
tukkol	...no all filles changed since the last...	15:24
cmn	no	15:24
	← ZyX-I left	15:24
cmn	it affects files with merge conflicts	15:24
	→ hellopat joined	15:24
	→ kevwil joined	15:24
	→ Jay_Levitt joined	15:25
	→ Araxia joined	15:26
	→ unmanbearpig joined	15:26
	← robbyoconnor left	15:26
	← Araxia left	15:27
tukkol	cmn: that might be correct but still it is a relevant information. don't you agree?	15:27
cmn	maybe	15:27
	→ thk\|work joined	15:27
	← barneybook left	15:27
tukkol	at least from a user perspective	15:27
cmn	what's relevant depends completely on the situation	15:27
tukkol	that the information is not necessary for the merge commit itself is clear to me.	15:27
cmn	huh?	15:27
tukkol	(the information which files where purely affected by previous commits)	15:28
	from the other branch	15:28
	→ LekeFly joined	15:28
	← nickpresta left	15:28
	← localhost left	15:28
cmn	if you want that, you can ask for that	15:28
	← corburn left	15:28
tukkol	so.. back to my original question.. how? ;)	15:28
cmn	git diff --name-only merge^1 merge^2	15:29
	→ gavin_huang joined	15:30
sitaram	jast: hi!	15:30
tukkol	no. i have file X and since "today" it does not compile. now I look at the log of the file and see a change that was made a year ago only, where actually a branch was merged today with a change made to that file a year ago.	15:30
	← Jippi left	15:30
tukkol	now i want to see when the change became part of my branch.	15:31
cmn	and you've checked that --first-parent doesn't work?	15:31
tukkol	let me check again	15:31
sitaram	jast: if I understand you correctly, you want a server side command to create a "wild-repos" repo	15:31
	→ psoo_ joined	15:32
sitaram	jast: can I assume latest v3 in use?	15:32
	→ drive-by-advice joined	15:32
	→ leila joined	15:33
	→ cagedwisdom joined	15:33
tukkol	mkdir mergetest && cd mergetest && git init . && echo x > X && git add X && git commit -m "init" && git branch test && echo y > X && git commit -a -m "changed X" && git checkout test && git merge --no-ff master && git log --first-parent X	15:33
	→ SirCmpwn_ joined	15:33
tukkol	doesn't show the merge :\|	15:33
	← ology left	15:33
	→ dinesh6654 joined	15:34
tukkol	(that is git 1.7.8)	15:34
	SirCmpwn_ → SirCmpwn	15:34
	→ tlocalhos joined	15:34
	← om left	15:35
	← tlocalhos left	15:35
	→ bigmeow joined	15:35
bigmeow	hi all	15:35
cmn	then --first-parent isn't that smart	15:35
bigmeow	how to ignore files that end with ~?	15:35
	vim and emacs generate lots of such files:(	15:35
canton7	man gitignore	15:36
gitinfo	the gitignore manpage is available at http://git-scm.com/docs/gitignore.html	15:36
	← Jay_Levitt left	15:36
	→ localhost joined	15:36
bigmeow	canton7: dude, does .gitignore support wildchar?	15:36
agumonkey	bigmeow: beside gitignore, you can tell emacs/vim to store the temp files somewhere and/or not create them at all	15:36
tukkol	cmn: hehe ;).. any other advice?	15:37
	(oh and thanks for your time and support btw!)	15:37
	← localhost left	15:37
bigmeow	agumonkey: dude, but filename~ maybe useful sometimes:)	15:37
cmn	it's still not clear what you want; you seem to know where the problem originated	15:37
bigmeow	cmn seems blind :)	15:37
canton7	bigmeow, read "pattern format" in that man page. it's basically shell globs	15:37
cmn	bigmeow: wtf are you talking about?	15:37
tukkol	cmn: yeah well but it's a temporal problem not a topological...	15:37
bigmeow	cmn: sorry:(	15:37
agumonkey	bigmeow: true	15:38
cmn	temporal?	15:38
	git sorts things topologically	15:38
	← tsdh left	15:38
	← shreya left	15:39
tukkol	cmn: as my example with the exeggerated year. why would i assume that the problem from a change a year ago would affect me when I always was 'up to date' and it worked until today.	15:39
cmn	the timing is irrelevant	15:39
	→ jdav_gone joined	15:40
cmn	you merged something, and it broke	15:40
tukkol	that's why i said "it's a temporal problem not a topological" ;)	15:40
	time IS relevant for a user.	15:40
cmn	you're just saying the opposite follows?	15:40
sitaram	jast: if so, here's how: GL_USER=some-user-because-wild-repos-need-one gitolite perms -c reponame < /dev/null	15:40
	→ psoo joined	15:40
	← fornext left	15:41
	← Fandekasp left	15:41
	→ ShooterMG joined	15:41
	← psoo_ left	15:42
tukkol	no. i just ask how would a merge done by person X make a person Y realize that a change for an arbitrary file made a year ago would be the cause of a problem?	15:42
	especially when person Y was not aware of a merge	15:42
	← psoo left	15:42
	→ psoo joined	15:42
cmn	that's a person and communication problem	15:42
	→ kukks joined	15:42
	← kukks left	15:42
	→ kukks joined	15:42
	→ kvothe joined	15:42
	← kvothe left	15:42
	→ kvothe joined	15:42
cmn	if you're merging old branches without testing or telling people, git can't help you	15:42
	→ artnez joined	15:42
tukkol	cmn: so blame the person not the tool for not providing sufficient info? ;)	15:42
	→ Fandekasp joined	15:42
	← marhaban left	15:43
bigmeow	how to remove all the files which are not version controled by git?	15:43
	→ snearch joined	15:43
cmn	man git clean	15:43
gitinfo	the git-clean manpage is available at http://git-scm.com/docs/git-clean.html	15:43
bigmeow	especially the files which are in .gitignore?	15:43
	← ChadStudmuffin left	15:43
bigmeow	cmn: git clen not works for the files which are in .gitignore:(	15:43
tukkol	i don't see why you can't see that this information might be helpful. regardless of how it can be circumvented by layer 8...	15:43
	anyhow..	15:44
cmn	bigmeow: yes, it does, see the man page	15:44
canton7	bigmeow, please read the man page.	15:44
	→ rohan32 joined	15:44
joshc	bigmeow: see the -X or -x flags	15:44
canton7	joshc, !fish :P	15:44
gitinfo	joshc: In the long run, it is much better to give pointers to answers than the answers themselves.	15:44
cmn	tukkol: I never said that	15:44
	← rohan32 left	15:44
cmn	but you're asking for information that you already have	15:45
	← ovnicraft left	15:45
tukkol	so, git --first-parent is not smart and it's not possible to see if a merge recently "brought" a commit to a specific file of a given branch.	15:45
cmn	look at gitk	15:45
	→ artem joined	15:45
bigmeow	how to push to git repo which is hosted on google code?	15:45
	← arturaz left	15:45
cmn	your situation doesn't make much sense; you already know what went wrong	15:45
	← kevwil left	15:46
tukkol	but only after layer 8 communicated and not by looking at the place where the error occured (i.e. a file)	15:46
	← dinesh6654 left	15:46
cmn	what on earth is this layer8?	15:47
tukkol	users :p	15:47
	humans	15:47
cmn	use bisect if you want to find out where an error was introduced	15:47
jast	sitaram: thanks	15:47
tukkol	http://en.wikipedia.org/wiki/Layer_8 ;)	15:47
	→ toabctl joined	15:47
cmn	why would you bring the network layer model into this?	15:47
sitaram	cmn: why do we bring furniture into this when we say PEBCAK?	15:48
tukkol	it's internet jargon not network layer model.. sorry for that.	15:48
jast	new channel rule: don't explain jokes ;)	15:48
sitaram	:)	15:48
	→ shreya joined	15:48
cmn	sitaram: to hit people with?	15:48
	← artnez left	15:49
	→ ovnicraft joined	15:49
	← artem left	15:49
cmn	tukkol: it's only jargon as it relates to the OSI model	15:50
	→ Shusshu joined	15:50
	← whitman left	15:50
	→ gxk joined	15:50
	← nor5trom left	15:50
sitaram	cmn: I'd go with tukkol on this -- I've heard it often enough that it's kinda become independent of the OSI model (to me)	15:50
	← giallu left	15:51
cmn	fine, but the only reason it exists is because of the model	15:51
sitaram	yes, but now it's grown up and moved out	15:51
	it might get married and change its last name too	15:52
bigmeow	anybody here who have tried to use git service provided by google code?	15:52
	it is kind of bull shit:(	15:53
cmn	change its last name? how traditional	15:53
	→ imachuchu joined	15:54
	→ artnez joined	15:54
	→ snappy joined	15:54
	gitinfo set mode: +v	15:54
	→ Rajesh joined	15:54
	Rajesh → Guest87497	15:55
	→ micah joined	15:55
	← jmd left	15:55
cagedwisdom	does anyone use git-annex? Is it good?	15:56
	→ MrRoot joined	15:57
milki	cagedwisdom: good for what?	15:57
micah	what happens when a hook has an error in one of its operations? does it error out the whole hook?	15:57
milki	cagedwisdom: and what metrics are we using to evaluate goodness	15:57
	← FreeArtMan left	15:57
cagedwisdom	milki, what it claims to do. manage large files with git and stuff.	15:57
milki	micah: that depens on your hook. its just a normal shell script.	15:57
canton7	cagedwisdom, people do use it, and they seem to like it. git-media is also considered good	15:57
	→ kenperkins joined	15:58
bigmeow	fk google code;(	15:58
cagedwisdom	milki, a normal shell script? It's written in haskell	15:58
milki	cagedwisdom: wut?	15:58
bigmeow	i still cannot push to that fk google's git repo:(	15:58
	← ISF left	15:58
micah	milki: yeah, the hook is just a normal bash script, should i set it up so every command that is run has the exit code tested and then to exit the hook if it doesn't work right?	15:59
snappy	Hi, I'm relatively new to git. I made some changes, and I want to roll back to an earlier commit, and push it to our master (central) repository. I did a git reset to the earlier commit, but when I try to git push --force, the remote master rjects it, since it's non-fast-forward. If I pull, I will receive the commits I want to rollback.	15:59
	I'm not sure how to resolve this.	15:59
	← mishok13 left	15:59
	← maestrojed left	15:59
	← jkremser left	15:59
cjb	snappy: you can configure the remote to accept non-fast-forward pushes, and it sounds like you have to do that since you've already pushed the changes you want to destroy.	15:59
micah	milki: or just a 'set -e' at the top	15:59
canton7	bigmeow, this is a fairly large channel: there are 955 people in here currently. If you have a specific question, please ask by all means. But please don't just rant against services	15:59
bigmeow	failed to push to git repo which is hosted google code:(	16:00
	i have followed the introduction, failed:(	16:00
	why?	16:00
milki	bigmeow: show us your error messages	16:00
	bigmeow: and what you did	16:00
bigmeow	http://iyanwu.com/pastebin/forumdisplay.php?fid=9	16:00
	← leila left	16:00
	← shruggar left	16:00
	→ shruggar joined	16:00
bigmeow	milki: ^ this link:) ↑	16:00
	← Guest87497 left	16:01
milki	uh, well looks like its a google specific thing	16:01
cmn	that seems to show that you failed to input the right username and password	16:01
milki	so ask google	16:01
	← toabctl left	16:01
cmn	have you seen what the error tells you to?	16:01
	← hellopat left	16:01
bigmeow	milki: dude, google lies:(	16:01
	→ darren__ joined	16:02
bigmeow	cmn: i have create that file which named .netrc:(	16:02
	← skyf_ left	16:03
	→ hellopat joined	16:03
cmn	if you don't trust google with your data, don't use their service, it's as simple as that	16:03
	← micah left	16:03
bigmeow	cmn: dude, google is asshole:(	16:04
	← joephelius left	16:04
bigmeow	cmn: it lies:(	16:04
	← jargon- left	16:04
canton7	bigmeow, cat ~/.netrc ?	16:05
imachuchu	snappy: you may also want to look at git revert, it's not the same as reset (as it creates a new commit), but it may be what you want. !rewrite	16:06
gitinfo	snappy: [!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	16:06
	→ leila joined	16:06
	← leila left	16:06
imachuchu	snappy: (sorry I thought that trigger had a link to a revert/reset explanation)	16:07
	→ boombatower joined	16:07
	← Shusshu left	16:08
	← cjb left	16:09
	← hellopat left	16:09
	← Superlokkus left	16:10
	→ etcetera joined	16:10
	← codetroll left	16:10
bigmeow	canton7: cat ~/.netrc, http://iyanwu.com/pastebin/showthread.php?tid=1726	16:11
	→ Survone joined	16:11
	← calvinx left	16:11
	← macIvy left	16:11
	← jackh left	16:12
canton7	bigmeow, and "xxxxfwe" is what appears at the top of https://code.google.com/hosting/settings ?	16:12
	← angelsl left	16:12
bigmeow	canton7: dude, it's the password generated by google:(	16:13
	← Survone left	16:13
canton7	and the output of git remote -v ?	16:13
milki	my googlecode password is longer than yours	16:13
	← ipalaus left	16:13
cmn	does that mean you're more manly?	16:14
milki	:P	16:14
	→ jkremser joined	16:15
	→ leila joined	16:16
	← leila left	16:16
	← nicoulaj left	16:16
bigmeow	canton7: works now.. if .netrc is used, the username cannot be shown in the url...	16:17
	← mikewintermute left	16:18
	→ mikewintermute joined	16:18
cmn	who'd've thunk; following the instructions makes it work	16:19
	← arosen left	16:19
	→ kevwil joined	16:19
canton7	yeah, shocking that. maybe they're not liars after all	16:20
	← Davey left	16:21
	→ micah joined	16:21
micah	is there a post-push hook that can run locally after I've pushed?	16:21
	→ drizzd_ joined	16:22
	← senny left	16:22
	→ Davey joined	16:22
	→ TheJH joined	16:22
	→ mishok13 joined	16:24
	← drizzd left	16:25
	→ hyperair joined	16:26
	← babilen left	16:27
	→ jonshea joined	16:28
	← gigamonkey left	16:28
	← mmc1 left	16:29
	→ Yuuhi joined	16:30
	← mikecmpbll left	16:30
	→ gigamonkey joined	16:31
	→ pdtpatrick joined	16:31
	→ raar joined	16:31
	← klj613 left	16:31
bigmeow	canton7: google's git system is quite different from github's:)	16:31
	how often should if commit , and how only should i push to remote repo?	16:32
	raar → Guest81831	16:32
bigmeow	is there any suggestion on this?	16:32
	how to write meaningful comments?	16:32
	← _Vi left	16:32
bigmeow	too much comments is bullshit or not?	16:33
	→ codetroll joined	16:33
canton7	bigmeow, how you commit shouldn't depend on where you intend on pushing. still follow http://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html	16:33
bigmeow	-m "add a funciont"	16:33
	-m "small modification 1"	16:33
	→ _Vi joined	16:33
bigmeow	-m "small modification 2"	16:33
fly9	anyone using gitlab?	16:33
canton7	both examples of terrible commit messages	16:33
bigmeow	-m "small modification 3"	16:33
	like this?	16:33
	fly9: what is git lab?	16:34
	→ macIvy joined	16:34
fly9	bigmeow: private github	16:34
bigmeow	canton7: why it is terrible?	16:34
	fly9: dude, why not github:)	16:34
fly9	bigmeow: open source	16:34
canton7	bigmeow, read the page I linked to	16:35
fly9	bigmeow: saves company $15k a year	16:35
bigmeow	fly9: which git service are you using now?	16:35
fly9	bigmeow: gitlab	16:35
bigmeow	fly9: is gitlab kind of open source git platform?	16:35
canton7	bigmeow, http://gitlabhq.com/	16:36
fly9	bigmeow: it would cost $15k to use github and host it internally	16:36
micah	15k?!	16:36
fly9	it doesn't have everything but its pretty great	16:36
micah	gitlabhq is... not fun to setup	16:37
fly9	micah: yeah its licensed by users, we have a lot of users	16:37
	micah: I've got it down to a science, but yeah getting there is a pain	16:37
micah	fly9: oh you mean if you pay for github's thing	16:37
mkj_wk	also last I checked github enterprise was still limited to running only on a virtual machine	16:37
fly9	micah: lots of bad or incomplete tutorials out there	16:38
	← pdtpatrick left	16:38
fly9	micah: exactly	16:38
	← arvind_khadri left	16:38
fly9	mkj_wk: yeah good point	16:38
	→ pdtpatrick joined	16:38
fly9	mkj_wk: although we run gitlab on a vm haha	16:39
	→ duckx joined	16:39
mkj_wk	fly9: then a lack of distinction without a difference :)	16:40
	← unmanbearpig left	16:41
mkj_wk	we wanted to have a deployment on big fast physical hardware with failover which made the license issue a moot point here	16:41
	(for one particular implementation, that is; we also have a github enterprise where that wasn't a concern)	16:42
	→ drodger joined	16:42
	← duckxx left	16:42
	→ maestrojed joined	16:43
	→ methods joined	16:43
bigmeow	micah: how much time have you spent to setup gitlab?	16:43
methods	is it possible to unstage my commits so i can commit them differently ?	16:43
snappy	imachuchu: thank you for that info	16:43
fly9	mkj_wk: yeah we get by with periodic snapshots and git handles the repo backups easily	16:43
	← pdtpatrick left	16:44
	← dr_lepper left	16:44
bigmeow	fly9: why gitlab need licence?	16:44
fly9	bigmeow: no its open source	16:44
tystr	hmm	16:44
micah	bigmeow: more than I wanted to, and then I found out it destroyed my gitolite config because I didn't know it had to start its own from scratch and only use that	16:44
	so I got frustrated and gave up	16:44
tystr	why when I type "git rebase -i HEAD~3" do I see 13 commits?	16:44
bigmeow	micah: why not use svn + git solution?	16:45
	→ _68x joined	16:45
canton7	tystr, merges?	16:45
micah	bigmeow: what?	16:45
	← LimeBlast left	16:45
fly9	micah: it can point to any gitolite but yeah that'd be frustrating if you tried that way	16:45
micah	bigmeow: i'm not using svn at all	16:45
tystr	canton7 what do you mean?	16:45
	← _68x left	16:45
tystr	I'm just trying to modify a commit a made a couple of commits ago	16:45
bigmeow	micah: do you use eclipse?	16:45
canton7	tystr, is one of the commits in the last 3 a merge commit?	16:45
micah	bigmeow: no	16:45
tystr	oh, not in the last 3	16:46
	:q!	16:46
	oops	16:46
imachuchu	methods: no (but there may be a terminology issue here). Do you want to remove previous commits or unstage changes to be added to a commit?	16:46
methods	i already committed without doing a push but i want to uncommit it so i can commit against differently	16:46
tystr	canton7 interesting…it has merge errors if I just exit vi leaving pick for every commit	16:46
	→ AaronMT joined	16:46
	← Twey left	16:47
canton7	tystr, are you absolutely sure about your assertion that there aren't any merges?	16:47
	→ Twey joined	16:47
	← weiyang left	16:47
imachuchu	methods: "rebase -i" is the general purpose command, "commit --amend" allows you to change your previous commit. Read !fixup and definetly think about !backup before you do something possibly dumb	16:48
gitinfo	methods: 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!	16:48
	methods: Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	16:48
methods	hm `git reset —soft HEAD^` worked	16:48
tystr	canton7 oh checking github there is indeed a merge commit in there	16:48
methods	then `git reset`	16:48
canton7	tystr, I knew it :) rebase and merge don't play well together. read --preserve-merges in man git-rebase, and also "bugs"	16:49
gitinfo	tystr: the git-rebase manpage is available at http://git-scm.com/docs/git-rebase.html	16:49
tystr	canton7 ok thanks :)	16:49
imachuchu	methods: git reset moves around refs, so yes it can be used to remove commits, though you lose those changes too	16:49
	← Pupeno_W left	16:50
	→ dvide joined	16:50
	→ Pupeno_W joined	16:50
tystr	canton7 worked like a charm…thanks!	16:51
canton7	tystr, sweet!	16:51
tystr	i didn't know about the --preserve-merges option	16:51
	← ovnicraft left	16:51
	→ HyperNerdV2 joined	16:51
	→ hwrd\|work joined	16:51
methods	thanks	16:52
	← methods left	16:52
	→ stkrzysiak joined	16:53
	← AaronMT left	16:53
	→ nevyn joined	16:54
	→ towski joined	16:55
	← john_anderson left	16:55
	← thiago left	16:55
	← luxurymode left	16:55
	→ apok joined	16:56
	→ jkremser_ joined	16:56
	← jkremser left	16:57
	→ benbergman joined	16:58
	→ thirdknife joined	16:59
	← PigeonFriend left	16:59
	← macIvy left	16:59
	→ rapha joined	16:59
rapha	hi!	17:00
	I'm trying to switch to the master branch but it tells me "fatal: A branch named 'master' already exists" - what am I doing wrong?	17:00
	→ opus joined	17:00
canton7	tell me, are you trying to switch using 'git checkout -b' ?	17:01
cbreak	rapha: stop trying to create a new master branch then :D	17:01
	← tatsuya_o left	17:01
	→ ovnicraft joined	17:02
	← Textmode left	17:03
	← cedeon left	17:03
	← nightcrawler786 left	17:03
	← russt left	17:03
	→ nightcrawler786 joined	17:03
imachuchu	rapha: assuming that canton7 is correct, just drop the -b option	17:04
	← diegoviola left	17:04
canton7	and read up on what the -b actually does. !book is good	17:05
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	17:05
	→ Textmode joined	17:05
	← Chib left	17:06
	→ russt joined	17:07
	← t0rc left	17:07
	← ovnicraft left	17:08
	← apok left	17:09
	← dstorrs left	17:09
	→ apok joined	17:09
rapha	canton7: cbreak: imachuchu: "git checkout master" indeed helped. thank you guys!	17:10
	← shruggar left	17:12
	→ Harley joined	17:12
	→ iwoj joined	17:12
	gitinfo set mode: +v	17:12
	→ mklappstuhl joined	17:13
	← Harley left	17:13
iwoj	i've made changes to some files, but they haven't been staged yet.	17:13
	how do I delete those changes?	17:13
mjt	git checkout <filename>	17:13
	→ corburn joined	17:13
iwoj	If I git checkout the last commit, it creates a new branch	17:13
	← geekbri left	17:13
iwoj	is there a way to go all the files in one go?	17:14
	→ _W_ joined	17:14
	→ dinesh6654 joined	17:14
	→ pdtpatrick joined	17:14
	→ osmosis joined	17:15
	← Fandekasp left	17:15
	→ skyf joined	17:16
	→ dr_lepper joined	17:16
	← dr_lepper left	17:16
	→ dr_lepper joined	17:16
	→ john_anderson joined	17:16
	→ Fandekasp joined	17:16
	→ hellopat joined	17:17
	← kleist left	17:18
	← iamjarvo left	17:18
	→ lacrymology joined	17:18
	→ localhost joined	17:19
	→ jceb_ joined	17:20
	← v0n left	17:20
canton7	iwoj, git checkout -- .	17:20
	← jez9999 left	17:21
	→ StefanM joined	17:21
	→ macer1 joined	17:22
	→ macIvy joined	17:22
	← jceb_ left	17:23
	gitinfo set mode: +v	17:23
macer1	Hi guys	17:24
gitinfo	macer1: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	17:24
macer1	how can I delete file from all commits?	17:24
	I mean a file	17:24
	delete a file	17:24
cmn	man git filter-branch	17:24
gitinfo	the git-filter-branch manpage is available at http://git-scm.com/docs/git-filter-branch.html	17:24
cmn	take into account that it's !rewriting	17:24
gitinfo	[!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	17:24
rapha	now i want to do "git submodule add ssh://blahblah blah" and i get "you are on a branch yet to be born" ... am i doing something wrong again?	17:24
macer1	I don't care about that, nobody is contributing to my repo atm	17:25
	or even forking it	17:25
	← Alvo left	17:25
cmn	it's about people working on their repos	17:25
	not about giving you changes	17:25
	← jceb left	17:25
macer1	nobody download my repo >.>	17:25
SirCmpwn	is this channel now +m?	17:25
	→ jceb joined	17:25
SirCmpwn	no...	17:25
	O_o	17:26
macer1	SirCmpwn: /topic	17:26
	Getting "Cannot send to channel"? /msg gitinfo .voice"	17:26
SirCmpwn	macer1: right, right	17:26
	macer1: it's +n, which I'm pretty sure is the "identify with nickserv" mode	17:26
	← macer1 left	17:26
	→ macer1 joined	17:26
	lindbohm.freenode.net set mode: +v	17:26
	ChanServ set mode: -v	17:26
macer1	.	17:27
SirCmpwn	no, it isn't. I dunno	17:27
macer1	ohh	17:27
	I thinked I was indetified	17:27
	gitinfo set mode: +v	17:27
canton7	Identified people can speak fine. Non-identified people need to be voiced, which the bot can do	17:27
	→ ology joined	17:27
rapha	canton7: do you think the problem is in the supermodule or in the repo i want to add as a submodule?	17:28
macer1	so, how to delete file from all commits?	17:28
canton7	rapha, make a commit, then try again	17:28
	← revagomes left	17:28
cmn	macer1: again, man git filter-branch	17:28
gitinfo	macer1: the git-filter-branch manpage is available at http://git-scm.com/docs/git-filter-branch.html	17:28
canton7	rapha, (use --allow-empty if you want to just make an empty commit)	17:29
macer1	oh	17:29
imachuchu	macer1: as cmn said, man git filter branch	17:29
gitinfo	macer1: the git manpage is available at http://git-scm.com/docs/git.html	17:29
macer1	git filter-branch --tree-filter 'rm randomunneededcrap' HEAD	17:29
rapha	canton7: committing seems to work in both and they're both up to date.	17:29
	→ api joined	17:30
canton7	macer1, need a bit more than that iirc. But I think there's an example	17:30
	macer1, yeah second example	17:30
	← codetroll left	17:30
	→ fayimora joined	17:31
	← opus left	17:31
	← john_anderson left	17:31
canton7	rapha, the error means, as far as I know, that you're on an orphan branch (one which isn't based on anything) which doesn't yet have any commits. For example, the error is raised when trying to do something in a new repo which you haven't yet committed to, or if you did 'checkout --orphan' and haven't yet committed	17:31
	← codecowboy left	17:32
	← localhost left	17:32
	→ v0n joined	17:32
	→ ChadStudmuffin joined	17:33
	← timthelion left	17:33
	← _Vi left	17:33
	→ joephelius joined	17:33
	← jceb left	17:33
rapha	canton7: that's very curious then, as the supermodule is quite old already, has around 15 submodules and has been committed to a couple dozen times. makes me quite clueless...	17:34
	← ashirus left	17:34
	→ _Vi joined	17:34
	→ jceb joined	17:34
canton7	rapha, maybe it's a problem with the submodule?	17:34
	← achadwick left	17:35
rapha	canton7: could be, i did rewrite it yesterday to correct the authors on some commits and then "git push -f"'d the changes. do you think that might be causing the problem?	17:35
	→ opus joined	17:36
canton7	rapha, I would think it unlikely..	17:38
	→ williamcotton joined	17:38
cmn	does git rev-parse HEAD error out?	17:38
canton7	(but that's not to say it can't be)	17:38
rapha	canton7: when i do a normal "git clone" of the intended submodule repo, it works just fine ...	17:39
	← TomyLobo left	17:39
	← fayimora left	17:39
	→ skorgon joined	17:39
	← LekeFly left	17:40
	← Nirvanko left	17:41
canton7	try cmn's suggestion. He knows more about this sort of stuff than me	17:41
	→ revagomes joined	17:41
	→ dstorrs joined	17:42
rapha	cmn: sorry didn't see you at first.	17:42
	→ velikasha joined	17:43
	← velikasha left	17:43
rapha	cmn: works for both, the supermodule and the intended new submodule.	17:43
	← mjt left	17:43
cmn	then that means that you're not on an unborn branch	17:43
	← kevwil left	17:44
	← hellopat left	17:44
	← juvenal left	17:45
	← russt left	17:45
	→ thiago joined	17:45
rapha	cmn: that doesn't disturb git in thinking so :-)	17:45
cmn	git is generally right, which means there's something odd there	17:45
	try removing any trace of the submodule	17:46
	← jceb left	17:46
rapha	cmn: funny, when I do the "git submodule add" a second time, it says "Adding existing repo at 'website' to the index", but it will be empty and updating won't fill it either.	17:46
	→ kevwil joined	17:46
cmn	the repo is there, clearly	17:47
rapha	cmn: already did that a couple of times. "rm -rf" for the submodule dir, removed it from .gitmodules and .git/config - that should be all traces iirc.	17:47
cmn	no	17:47
rapha	oh	17:47
cmn	git rm --cached dir	17:47
rapha	where else?	17:47
	ah	17:47
	← iwoj left	17:47
cmn	it shouldn't affect it, really, but this is an odd situation	17:47
rapha	cmn: git rm --cached website -> fatal: pathspec 'website' did not match any files (it is present, physically)	17:48
cmn	ok, then it never got that far in adding the submodule	17:53
	→ juvenal joined	17:53
	→ LekeFly joined	17:53
	→ jceb joined	17:53
	→ FreeArtMan joined	17:53
	→ Sym3tri joined	17:53
	← jceb left	17:53
	→ jceb joined	17:53
	→ maletor joined	17:53
	→ xrq joined	17:54
	→ cakehero joined	17:54
	← maletor left	17:54
rapha	cmn: is there any way to debug this?	17:55
	← jceb left	17:55
	← bogwonch left	17:55
	→ jceb joined	17:55
	← chuckharmston left	17:56
cmn	I believe there is a tool that will tell you which child printed out which line to the terminal, but I can't remember the name	17:56
xrq	why doesn't git stash allow you to make a comment? I'm trying to find an item to pop, trying to find it is like a needle in a haystack without relevant comments about what they were	17:56
cmn	xrq: it does	17:56
	→ lresende joined	17:56
cmn	put it at the end of the stash line	17:56
rapha	hmmm	17:57
xrq	that doesn't work. doing that prints out a git usage list	17:57
	← jceb left	17:57
rapha	maybe i should just re-do the supermodule from scratch. it only serves as a collection of submodules anyways.	17:57
imachuchu	xrq: use commits instead, then rebase them into something prettier later	17:57
cmn	xrq: see man git stash save	17:57
gitinfo	xrq: the git-stash manpage is available at http://git-scm.com/docs/git-stash.html	17:57
	← kevwil left	18:00
	← Vile left	18:01
	→ GNUcifer joined	18:02
xrq	that works, thanks	18:04
	→ jcao219 joined	18:05
	→ Sigma joined	18:05
jcao219	How do I undo my last commit? I did "git reset HEAD^" but that gives me a prompt "more? "	18:06
EugeneKay	jcao219 - !fixup	18:06
	EugeneKay pokes the bot with a stick	18:06
aboudreault	!fixup	18:06
	aboudreault - !fixup	18:07
EugeneKay	He'll catch up	18:07
	← ConstantineXVI left	18:07
jcao219	!fixup	18:07
gitinfo	jcao219: 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!	18:07
	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!	18:07
	aboudreault: 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!	18:07
	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!	18:07
EugeneKay	There we go.	18:07
	→ ConstantineXVI joined	18:07
aboudreault	:)	18:07
jcao219	ah	18:07
cbreak	jcao219: windows.?	18:08
jcao219	yes	18:08
cbreak	ha. ha.	18:08
	^ has special meaning there afaik	18:08
	← mutex left	18:08
jcao219	darn!	18:08
cbreak	use ~1 instead	18:08
jcao219	thanks	18:08
cbreak	I think the special meaning is "continue on next line" but I am not sure.	18:09
	→ pantsman joined	18:10
	← pantsman left	18:10
	→ pantsman joined	18:10
	→ berserkr joined	18:10
	→ ashirus joined	18:12
	→ tatsuya_o joined	18:12
	→ jceb joined	18:12
	← olaf_ left	18:12
	→ mutex joined	18:14
	→ maletor joined	18:14
	← Chappo left	18:14
	→ Chappo joined	18:15
	← tatsuya_o left	18:16
	→ iamjarvo joined	18:16
	→ sq-one joined	18:16
	← ashirus left	18:17
grawity	it's a general escape char in cmd.exe, much like \ in sh/bash	18:17
	← mikewintermute left	18:18
	→ flijten joined	18:18
	→ kevwil joined	18:18
grawity	so yes, "continue on next line" if it's at the end of the line.	18:18
	← kevwil left	18:18
	→ ashirus joined	18:19
	→ luxurymode joined	18:19
	← Fandekasp left	18:19
	← nhooey\|mini__ left	18:20
grawity	(though cmd.exe's parser is quite ugly, and it runs twice since git is usually invoked via git.cmd, so you never know if it's HEAD^^ or HEAD^^^^ that equals HEAD~1 ...)	18:20
	→ rehat joined	18:21
	→ ctdabomb joined	18:23
	gitinfo set mode: +v	18:24
	← magcius left	18:24
ctdabomb	is it possible to commit a file with spaces in the title? i get the error ctdabomb@ctdabomb-laptop:~/supertux$ git add data/levels/world1/01- Welcome to Antarctica.stl fatal: pathspec 'data/levels/world1/01-' did not match any files	18:25
	→ magcius joined	18:25
	→ ph^ joined	18:25
corburn	Use "\ " escapes	18:25
	→ chuckharmston joined	18:25
ctdabomb	er never mind, I got it	18:25
	commited the "whole folder"	18:26
cbreak	git doesn't care about spaces	18:26
	your shell might though	18:26
	← ashirus left	18:26
cbreak	just quote or escape	18:27
	→ codeFiend joined	18:28
	← magcius left	18:29
	→ fayimora joined	18:30
	← maletor left	18:30
	→ magcius joined	18:31
	← dstorrs left	18:33
	← ology left	18:33
	→ ashirus joined	18:33
	← _Vi left	18:33
	← luxurymode left	18:34
	→ luxurymode joined	18:34
	→ _Vi joined	18:34
	← luxurymode left	18:34
	→ dstorrs joined	18:35
	← ctdabomb left	18:35
	→ gdmfsob joined	18:37
	→ Fandekasp joined	18:37
	← joephelius left	18:37
rapha	cmn: so the problem seems to be in the working copy of the supermodule. i'm rebuilding it regardless as some other things about it were uncomfy, too.	18:37
	→ chuckhar_ joined	18:37
	→ Zenopus_ joined	18:37
	→ nyxx joined	18:37
	→ savid joined	18:38
	→ sh0ne joined	18:38
	← wolftankk left	18:38
	→ joephelius joined	18:38
	← jceb left	18:38
	→ Orphis_ joined	18:38
	→ Moult_ joined	18:38
	→ jiunec joined	18:38
	→ RickXy joined	18:38
	→ JamesB joined	18:38
	→ hide_ joined	18:38
	→ frogonwheels_ joined	18:39
savid	Does git have a way to automatically convert dos-style newlines to unix-style newlines on commit? My associate is on windows, and he keeps committing CRLF line endings.	18:39
	JamesB → Guest57735	18:39
	→ sinful_ joined	18:39
canton7	!crlf	18:39
gitinfo	to fix problems with line endings on different platforms, check out http://line-endings.gitrecipes.de/	18:39
	→ hexa2k9- joined	18:39
	→ luke-jr_ joined	18:39
savid	thanks!	18:39
	→ shibayu36_ joined	18:40
	→ ryland_ joined	18:40
	→ lu_zero_ joined	18:40
	← lu_zero_ left	18:40
	→ lu_zero_ joined	18:40
	→ myme_ joined	18:40
	→ tizzo- joined	18:40
	→ wanders_ joined	18:40
	→ cooloney_ joined	18:40
canton7	savid, also might be worth looking at core.autocrlf in man git-config	18:40
gitinfo	savid: the git-config manpage is available at http://git-scm.com/docs/git-config.html	18:40
	→ tureba_ joined	18:40
	→ twem2_ joined	18:40
	→ TML__ joined	18:40
	→ d3vic3_ joined	18:40
	→ w0m_ joined	18:40
	→ elliottc1ble joined	18:40
	→ xxtjaxx_ joined	18:40
	→ pasky_ joined	18:40
	→ styx__ joined	18:40
	→ FabTG_ joined	18:40
	→ chuckf_ joined	18:40
	→ wereHams1er joined	18:40
	→ WakiMiko_ joined	18:40
	→ opus__ joined	18:41
	→ Remram_ joined	18:41
	→ bartman` joined	18:41
	→ KBme_ joined	18:41
	→ antix_ joined	18:43
	→ burzum` joined	18:43
	→ luxurymode joined	18:43
	← jcao219 left	18:43
	→ lustucru joined	18:43
	→ jcao219 joined	18:44
	→ barrbrain_ joined	18:44
	→ ArchDeacon joined	18:44
	→ ivan\_ joined	18:44
	← TML__ left	18:44
	→ schlumpi joined	18:44
	→ tg` joined	18:44
	→ NfNitLoo` joined	18:44
	→ TML_ joined	18:44
	← TML_ left	18:44
	→ SirFunk_ joined	18:45
	→ milli` joined	18:45
	→ lamont` joined	18:45
	← ashirus left	18:45
	→ Mez__ joined	18:45
	→ BinGOs_ joined	18:45
	← chuckharmston left	18:45
	← sq-one left	18:45
	← berserkr left	18:45
	← Sym3tri left	18:45
	← opus left	18:45
	← mishok13 left	18:45
	← d3vic3 left	18:45
	← luke-jr left	18:45
	← cenuij left	18:45
	← cooloney left	18:45
	← barrbrain left	18:45
	← bartman left	18:45
	← ivan\ left	18:45
	← Zenopus left	18:45
	← tureba left	18:45
	← Remram left	18:45
	← wereHamster left	18:45
	← hpa left	18:45
	← ro0 left	18:45
	← milli left	18:45
	← w0m left	18:45
	← lamont left	18:45
	← chuckf left	18:45
	← NfNitLoop left	18:45
	← nyx left	18:45
	← mbreslin left	18:45
	← Mez left	18:45
	← KBme left	18:45
	← arex\ left	18:45
	← schlumpi_ left	18:45
	← BinGOs left	18:45
	← ryland left	18:45
	← burzum left	18:45
	← cespare left	18:45
	← tizzo-afk left	18:45
	→ no_maam_ joined	18:45
	opus__ → opus	18:45
	Zenopus_ → Zenopus	18:45
rapha	i slightly changed the location (just the path bit) of my origin repository to which I push ... can I just change it in the working copy's .git/config or do you need to do more than that?	18:46
	barrbrain_ → barrbrain	18:46
	tg` → tg	18:46
	ivan\_ → ivan\	18:46
	→ arex\ joined	18:46
cbreak	rapha: man git-remote set-url	18:46
gitinfo	rapha: the git-remote manpage is available at http://git-scm.com/docs/git-remote.html	18:46
m1sc	rapha: either using set-url (man git-remote) or by hand, yes	18:46
gitinfo	rapha: the git-remote manpage is available at http://git-scm.com/docs/git-remote.html	18:46
	→ hpa_ joined	18:46
	← magcius left	18:48
	← ArchDeacon left	18:49
	→ Sym3tri joined	18:50
	→ magcius joined	18:51
	→ AaronMT joined	18:51
	→ sq-one joined	18:51
	← blast_hardcheese left	18:51
	→ cespare joined	18:51
	← AaronMT left	18:51
	→ babilen joined	18:51
	← jiunec left	18:52
	→ jiunec joined	18:52
	jiunec → cenuij	18:53
	→ markatto joined	18:53
	→ rking joined	18:53
	→ berserkr joined	18:54
	← dc5ala left	18:55
	→ chester_1 joined	18:55
	← hexa2k9- left	18:56
	→ _iron joined	18:57
	→ hexa2k9 joined	18:57
	← intransit left	18:57
	← chester_ left	18:58
	← cakehero left	18:59
	→ bogwonch joined	18:59
	← mklappstuhl left	18:59
	← gigamonkey left	19:00
	→ TML joined	19:01
	→ mikewintermute joined	19:02
	← jkremser_ left	19:02
	→ ro0 joined	19:02
	→ timthelion joined	19:03
	← thirdknife left	19:06
	→ cmullican joined	19:07
	→ blast_hardcheese joined	19:07
rapha	cbreak: m1sc: thanks, worked like a charm! :)	19:08
	→ ReekenX joined	19:11
	→ martinjlowm joined	19:11
	→ ipalaus joined	19:11
	← Sky[x] left	19:12
	← Eridius left	19:13
	→ jceb__ joined	19:14
	← joephelius left	19:14
	← dv310p3r left	19:15
	→ joephelius joined	19:15
	← mikewintermute left	19:16
	→ pandeiro joined	19:16
	← shawngo left	19:16
	← palm_vue left	19:20
	← bizhanMona left	19:20
	← Lgb left	19:20
	← jceb__ left	19:21
	→ tatsuya_o joined	19:21
	← Nibato left	19:22
	← cwang left	19:22
	→ Lgb joined	19:23
	→ palm_vue joined	19:23
	→ tjholowaychuk joined	19:23
	← ph^ left	19:23
	→ bizhanMona joined	19:23
	→ rcu joined	19:24
	→ Nibato joined	19:25
	← fisted left	19:26
	← Khisanth left	19:26
	← Sigma left	19:27
	→ ph^ joined	19:27
	→ fisted joined	19:27
	→ john_anderson joined	19:29
	← TheLZA left	19:29
	← marthinal left	19:29
mikemol	So, I've got a folder which contains three files, A, B and C. I've modified B. I need to remove A and C (and commit their removal) before I commit B.	19:30
	← rgsteele left	19:31
	← bogwonch left	19:31
Tommy[D]	man git-stash	19:32
gitinfo	the git-stash manpage is available at http://git-scm.com/docs/git-stash.html	19:32
mikemol	ty	19:32
FauxFaux	mikemol: You can just do that with git add / git rm...	19:32
rapha	bye all!	19:33
	← rapha left	19:33
mikemol	I'll check out git-stash, thanks.	19:33
	→ sec_goat joined	19:33
	← gtrak left	19:33
	← ska-fan left	19:33
	← _Vi left	19:34
	→ _Vi joined	19:34
mikemol	FauxFaux: The trouble is that I can't seem to stage the commit for rm A and rm C without staging the modification for B. But as it happens, I just realized I need to regenerate B after I commit the rms for A and C, anyway, and the regenerated B is the sensible thing to commit.	19:34
	← BinGOs_ left	19:34
	→ qknight joined	19:35
FauxFaux	mikemol: git rm /does/ stage the rms?	19:35
qknight	is a 'git clone remote...' the only thing i can do after 'git reset --hard HEAD'? i just want to get to the state of the remote repo again	19:35
	← snearch left	19:35
	→ BinGOs joined	19:35
FauxFaux	qknight: git reset --hard origin/master # is what you want.	19:36
	→ shawngo joined	19:36
	→ TheLZA joined	19:36
	← TheLZA left	19:36
	→ TheLZA joined	19:36
	← TheLZA left	19:36
mikemol	FauxFaux: No, but I'm unfamiliar how to stage something without 'git add', and this particular case wouldn't let me 'add' unless I 'add'ed the folder that contained A, B and C, which meant scooping up B.	19:36
	← sec_goat left	19:37
	→ TheLZA joined	19:37
	← TheLZA left	19:37
FauxFaux	git rm /does do git add/.	19:37
	← etcetera left	19:37
	→ TheLZA joined	19:37
	← TheLZA left	19:37
	→ beneggett joined	19:37
	→ archdori joined	19:37
	← Ad0 left	19:38
mikemol	FauxFaux: Except git status puts "deleted: folder/A" into the "Changes not staged for commit" section.	19:38
	→ ijuma joined	19:38
	← Fandekasp left	19:38
mikemol	Ah, I see what happened.	19:38
	← opus left	19:38
	→ agileadam joined	19:38
mikemol	I must have run a simple 'rm' instead of a 'git rm' at some point.	19:39
	← micah left	19:39
	→ ska-fan joined	19:39
mikemol	git checkout path/file;git rm path/file;git status # shows what I was hoping to see.	19:39
canton7	you could just 'git rm path/file' without the preceding checkout	19:39
	→ jobin joined	19:39
	→ arturaz joined	19:40
	← shreya left	19:40
cmn	with --cached	19:40
	← nifty_ left	19:40
	← alanmac left	19:41
	→ Khisanth joined	19:41
	→ TheLZA joined	19:43
	→ dv310p3r joined	19:43
	→ txdv joined	19:44
	← spiffytech left	19:46
	← Khisanth left	19:46
	→ t0rc joined	19:48
	← maestrojed left	19:50
	→ Speed joined	19:50
	← aldodelgado left	19:51
	→ aldodelgado joined	19:51
	← Chappo left	19:51
	→ Chappo joined	19:52
	→ etcetera joined	19:53
	→ Fangorn_ joined	19:53
	→ bogwonch joined	19:53
	← etcetera left	19:54
	← aldodelgado left	19:57
	→ aldodelgado joined	19:57
	← lresende left	19:57
	→ lresende_ joined	19:57
	← flijten left	19:57
	← fayimora left	19:58
	→ aaronj_ joined	19:58
	→ Khisanth joined	20:00
	→ heinrich5991 joined	20:02
	← bubbles10101 left	20:02
	← longbeach left	20:02
	← joephelius left	20:03
	← graphitemaster left	20:03
	← tatsuya_o left	20:04
	→ Bosox20051 joined	20:04
	← aaronj_ left	20:05
	→ tatsuya_o joined	20:05
	→ aaronj joined	20:06
	→ fayimora joined	20:06
	→ opus joined	20:07
	← paolooo left	20:07
aaronj	I'm trying to use git to streamline development with wordpress and having trouble with rebasing and/or merging local customizations with new versions from the origin repo	20:08
	→ bubbles10101 joined	20:08
aaronj	https://gist.github.com/3363167 is an example of what I'm trying	20:08
	I don't understand why I'm getting merge errors or why a rebase would need to merge since I'm not actually modifying any files on my branch, just adding new ones	20:09
	← ReekenX left	20:09
	← chrisgeorge left	20:10
PerlJam	aaronj: I think perhaps you want "git merge -s theirs mysite-3.3.1"	20:11
	→ specialGuest joined	20:12
	← specialGuest left	20:12
	→ specialGuest joined	20:12
aaronj	PerlJam: I'll have a look at -s, thanks	20:12
	→ rtjure joined	20:12
	→ etcetera joined	20:12
imachuchu	umm, at least according to mine, there isn't a "theirs" strategy (with good reason really)	20:12
	→ chrisgeorge joined	20:12
imachuchu	there is a "-x theirs" but that can be completely different and so should be read up on	20:13
	← flavius left	20:13
	→ civilordergone joined	20:13
jast	plus it's actually -Xtheirs ;)	20:14
civilordergone	hey, i'm trying to find a script that allows a commit / push to a master branch to deploy to a particular directory	20:14
imachuchu	yes, it's -X not -x	20:14
aaronj	theirs isn't an option for me, I must have an older version of git	20:15
jast	-X is tricky anyway	20:16
	← iamjarvo left	20:16
	← artnez left	20:16
imachuchu	aaronj: try this (assuming the link you posted) "git pull origin", "git checkout -b rebaseBranch <whateverYourWorkBranchIsNamed>", "git rebase master"	20:16
	→ artnez joined	20:16
aaronj	would it be easier to just make a patch on my branch and apply it on a new branch i make based on the new version?	20:17
	← pmdarrow left	20:17
aaronj	imachuchu: I don't want to rebase to master, but to a specific tagged version	20:18
imachuchu	and "-s theirs" would render your history moot (the "ours" option really shouldn't be used that often, it's for blowing away useless obsolete branches)	20:18
aaronj	imachuchu: unless I'm missunderstanding what rebase master does	20:18
	basically, I just have like two new directories (submodules) + a couple new files, no actual modifications of the source tree apart from those additions	20:18
	← arturaz left	20:19
aaronj	I just want a simple way to keep those changes but update everything else to the next tagged release	20:19
	→ vervic joined	20:19
	← HyperNerdV2 left	20:20
imachuchu	aaronj: actually try the rebase listed above, it might actually be exactly what you want	20:20
aaronj	imachuchu: will do	20:20
imachuchu	aaronj: a rebase (like this one at least) takes your changes and reapplies them to the new root	20:20
aaronj	imachuchu: actually, isn't that the same as what I posted the last three lines on my link?	20:21
	imachuchu: see last three lines on https://gist.github.com/3363167	20:21
	→ hellopat joined	20:21
	→ AaronMT joined	20:21
imachuchu	aaronj: nope, you are trying to rebase ontop of your branch instead of the new master	20:21
	aaronj: "git rebase mysite-3.3.1" instead of "git rebase master"	20:22
aaronj	imachuchu: if I rebase to a new master, and I then checkout a tag 3.4.1, it won't have the stuff I added to master, correct?	20:22
imachuchu	aaronj: though make sure the new branch (before rebase) = your development branch	20:22
	NfNitLoo` → NfNitLoop	20:22
	← skorgon left	20:22
	← aldodelgado left	20:22
imachuchu	aaronj: before I say anything incorrect, are you changing master or is it a mirror of the upstream?	20:23
aaronj	the thing is, I don't want to keep up with master, that's an unstable place, I want to use a versioned tag	20:23
imachuchu	aaronj: changing = applying your commits to	20:23
	← corburn left	20:23
imachuchu	aaronj: ahh, then "git rebase <whateverYouLike>" where whateverYouLike can be a tag	20:23
aaronj	I'm not touching master, I checkout tag 3.3.1, in my example, then make a new branch git checkout -b mysite-3.3.1	20:24
	hmmm	20:25
imachuchu	aaronj: ok then "git pull", "git checkout -b mergeBranch mysite-3.3.1", "git rebase 3.3.1"	20:25
	→ intransit joined	20:25
imachuchu	aaronj: that should make mergeBranch = your changes applied ontop of tag 3.3.1	20:25
aaronj	ok, I see	20:26
	lemme try that	20:26
	← fly9 left	20:27
imachuchu	aaronj: rebase is (IMO) the most useful command in git (well it and cherry-pick, which it is built out of). It's what seperates it from the savages (i.e. bazaar, mercurial)	20:28
	aaronj: ooh, and !rewrite if it applies	20:28
gitinfo	aaronj: [!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	20:28
	← fayimora left	20:29
	← Chappo left	20:29
imachuchu	aaronj: (if somebody else interacts with this repo you can be changing shared history, which everyone then needs to go back and tell git to accept)	20:29
	→ Chappo joined	20:30
	→ ashirus joined	20:30
	← skyf left	20:30
	← Geralt left	20:31
	← rcu left	20:31
	→ rcu joined	20:33
	→ shreya joined	20:34
	← _Vi left	20:34
aaronj	imachuchu: fail: https://gist.github.com/3363372	20:35
	→ _Vi joined	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
aaronj	imachuchu: those are the exact steps I took so should be easy to reproduce	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
	← _iron left	20:35
	→ Harzilein joined	20:35
	← Harzilein left	20:35
	← specialGuest left	20:35
	← archdori left	20:37
	← ChadStudmuffin left	20:37
	← TheLZA left	20:37
	→ joephelius joined	20:38
	← GNUcifer left	20:38
imachuchu	aaronj: just for shiggles, what does a "git status" show as the conflicting file/files?	20:39
	→ Sky[x] joined	20:39
aaronj	imachuchu: common.js and script-loader.js	20:39
	imachuchu: full output of git status: https://gist.github.com/3363420	20:40
	→ TheLZA joined	20:40
	← TheLZA left	20:40
	→ aspotashev joined	20:40
	→ TheLZA joined	20:40
	← TheLZA left	20:40
	→ Eridius joined	20:41
imachuchu	aaronj: and you didn't change them at all (I don't know WP at all, does it automatically modify them when you run it or something)? Either way give me a sec	20:41
aaronj	would it just be easier for me to create a patch against 3.3.1, then checkout 3.4.1 and apply that patch?	20:41
imachuchu	aaronj: yes/no because then you lose your history (if you've made 5 commits it will now only show 1)	20:42
aaronj	imachuchu: haven't touched any files, I did the exact commands here: https://gist.github.com/3363372	20:42
	ahh, yea that's not ideal	20:42
	→ mritz_ joined	20:42
	← sq-one left	20:42
	→ cillybabe joined	20:43
aaronj	imachuchu: wp isn't running off this repo, it's just a sandbox in my homedir	20:43
	→ russt joined	20:44
	← opus left	20:44
	→ TheLZA joined	20:44
	← TheLZA left	20:44
imachuchu	aaronj: ooooh! !float I bet is what's causing this	20:44
gitinfo	aaronj: 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.	20:44
	← jonshea left	20:44
aaronj	imachuchu: I don't see how since I've committed everything	20:45
	→ jonshea joined	20:45
	← Anubisss_ left	20:45
aaronj	imachuchu: see: https://gist.github.com/3363372	20:45
	don't want to flood the channel with commands	20:45
	→ Anubisss joined	20:45
	← yoklov left	20:45
imachuchu	aaronj: yeah maybe not, whatever, replace "git rebase" with "git cherry-pick <firstCommitThatYouChanged>..<lastCommitYouChanged>"	20:46
	aaronj: crude, but that would still conflict	20:46
	→ nor5trom joined	20:47
imachuchu	aaronj: "git cherry-pick -X theirs <firstCommitThatYouChanged>..<lastCommitYouChanged>"	20:47
	cillybabe → cilly	20:47
	← hellopat left	20:48
	→ robottinosino joined	20:48
aaronj	imachuchu: there's only one commit, so just the hash of that then as it appears in git log?	20:49
	can't I do 3.3.1..mysite-3.3.1 ?	20:49
	those should be pointers to those hashes, right?	20:49
imachuchu	aaronj: yeah should be (i've always done subranges but yeah)	20:50
	→ Lulzon joined	20:50
imachuchu	aaronj: what it's saying though is that one of your commits changes the common.js and the script-loader.php file and that it can't easily reconsile the differences	20:51
aaronj	imachuchu: https://gist.github.com/3363543	20:51
	← FreeArtMan left	20:51
aaronj	imachuchu: that's impossible	20:51
imachuchu	aaronj: so somehow you are commiting changes to those two files (why I dislike the "commit -a" option)	20:51
aaronj	hmmm	20:52
	what does -a do?	20:52
	none of my commits are changing anything, all I do is add an empty file	20:52
imachuchu	aaronj: "commit all tracked files that are different from the tracked ones in the repo"	20:52
	aaronj: try again, this time with "git add newfile.txt" and """git commit -m "Adding a file" """	20:53
aaronj	k	20:53
	← AaronMT left	20:54
	→ Sigma joined	20:54
	← Sigma left	20:54
	→ Sigma joined	20:54
	← linoj left	20:54
	← freeed left	20:54
aaronj	imachuchu: the git rebase 3.4.1	20:55
	imachuchu: failed, same merge errors	20:55
	← lirt left	20:55
aaronj	I can't do the chrry pick because I don't have 'theirs'	20:55
	← Chappo left	20:55
	hpa_ → hpa	20:55
	→ GNUcifer joined	20:56
imachuchu	aaronj: the test should work just fine (clone; checkout; add; commit; checkout -b; rebase)	20:56
aaronj	git clone https://github.com/WordPress/WordPress.git && cd WordPress && git checkout 3.3.1 && git checkout -b mysite-3.3.1 && touch newfile.txt && git add newfile.txt && git commit -m "Adding a file" && git checkout -b mysite-3.3.1--3.4.1-rebase && git rebase 3.4.1	20:56
	→ artem joined	20:56
	→ lirt joined	20:56
aaronj	imachuchu: can you try running the above?	20:56
	imachuchu: see if you get a merge error?	20:56
imachuchu	aaronj: give me a sec to boot up a vm	20:56
	← rwestphal left	20:56
	→ Chappo joined	20:57
aaronj	imachuchu++	20:57
	← chuckhar_ left	20:57
	→ ology joined	20:57
	← lirt left	20:57
	→ opus joined	20:57
	← markatto left	20:58
imachuchu	aaronj: (and it will be a bit, this comps a bit of a clunker)	20:58
	← msmithng left	20:58
aaronj	heh	20:58
	← artnez left	20:58
	→ lirt joined	20:58
	→ jsec joined	20:58
	jsec → Guest27074	20:58
	→ Vile joined	20:58
aaronj	I wonder if there's something funky about the WP git repo	20:58
	← macer1 left	20:59
	← lirt left	20:59
	→ cirenyc joined	20:59
aaronj	it's synced via SVN every 15 minutes, would that make a difference?	20:59
	← chrisgeorge left	21:00
imachuchu	aaronj: I know nothing about how their repo is set up	21:00
	→ lirt joined	21:00
	→ keystr0k joined	21:00
	→ chrisgeorge joined	21:00
	→ markatto joined	21:01
	← lirt left	21:01
	→ leila_ joined	21:01
	→ lirt joined	21:02
	← lirt left	21:03
	→ Dimitr\|2 joined	21:03
	← joephelius left	21:03
	← gfontenot left	21:03
	→ lirt joined	21:03
	← agileadam left	21:04
	keystr0k → agileadam	21:04
	architekt → architekt_	21:04
	← lirt left	21:04
	← nor5trom left	21:05
	← tjholowaychuk left	21:05
	→ lirt joined	21:05
	architekt_ → architekt	21:05
	← lirt left	21:06
	→ specialGuest joined	21:06
	← specialGuest left	21:06
	→ specialGuest joined	21:06
	← chester_1 left	21:06
	← LekeFly left	21:07
	→ lirt joined	21:07
	← pdtpatrick left	21:07
aaronj	anyone else have cycles to help troubleshoot this? Is there a better way of managing local customizations of an upstream repository, and merging periodically with upstream version releases? I would think this is a pretty common use case for DVCS, no?	21:07
	← cdash004 left	21:07
	← Sigma left	21:07
	← lirt left	21:08
	→ lirt joined	21:09
	← lirt left	21:09
	→ robbyoconnor joined	21:10
imachuchu	aaronj: yeah, ok, that repo looks like crap	21:10
	→ lirt joined	21:10
aaronj	imachuchu: :(	21:11
	← phishy left	21:11
	← lirt left	21:11
	→ bogwonch1 joined	21:12
imachuchu	the reason for the problems, if I understand it correctly, is that they aren't actually decendents of eachother	21:12
aaronj	imachuchu: so maybe I should just download each release, unzip it into a git repo i maintain with a tag for each release?	21:12
	→ HyperNerdV2 joined	21:12
	→ fly9 joined	21:12
aaronj	imachuchu: and have a branch for my stuff, which I merge in with my own tagged releases?	21:12
	→ lirt joined	21:12
aaronj	lot of extra steps but beats dealing with merge comflicts	21:12
imachuchu	svn handles branches much differently then git, so git can't see those tags as directly inheriting from eachother = no rebase (but we can still cherry-pick)	21:12
	← opus left	21:12
	← leeN left	21:13
imachuchu	aaronj: well give me a sec, really there shouldn't be any merge conflicts, just a bit funcky of a command to move the commits	21:13
	← lirt left	21:13
	← bogwonch left	21:13
	→ lirt joined	21:14
	← artem left	21:14
	← lirt left	21:15
	← Chappo left	21:16
	→ lirt joined	21:16
	← pandeiro left	21:16
	← cirenyc left	21:16
imachuchu	aaronj: try "git cherry-pick --strategy-option=theirs <beginningCommit>..<endingCommit>", or just "git cherry-pick --strategy-option=theirs <testCommit>"	21:16
	← HyperNerdV2 left	21:17
imachuchu	aaronj: try commit hashes too instead of a a ref, I'm not sure if it might think of them as their whole history (I haven't tested)	21:17
	← lirt left	21:17
	→ matled joined	21:17
matled	how can I check if a certain commit is an ancestor of another commit (for a script)?	21:17
	→ lirt joined	21:18
	oojacoboo → ichi	21:18
	→ thinice joined	21:18
	← afuentes left	21:18
aaronj	imachuchu: I think I got it working with cherry-pick	21:18
	ichi → oojacoboo	21:18
	← lirt left	21:18
	→ lirt joined	21:19
aaronj	imachuchu: https://gist.github.com/3363767	21:20
	← andrewhl left	21:20
aaronj	imachuchu: I think it gives me an upgrade path from 3.4.1 while preserving the updates I made previously	21:20
	→ olaf_ joined	21:21
matled	ah, git rev-list A --not B should work	21:21
aaronj	imachuchu: but honestly, why is it this hard? :)	21:21
	← lirt left	21:21
	→ lirt joined	21:22
	← j416 left	21:23
	→ mike007 joined	21:23
	← lirt left	21:23
	→ lirt joined	21:24
	← ashirus left	21:24
	→ joephelius joined	21:24
	← bgerber left	21:24
	← lirt left	21:25
	→ lirt joined	21:25
	→ victor` joined	21:26
	← lirt left	21:26
imachuchu	aaronj: because git uses the history between commits to figure out how the underlying code has changed	21:27
	→ lirt joined	21:27
	← rcu left	21:28
imachuchu	aaronj: it makes merges in git much easier/faster/simpler then those in SVN. The catch is it needs to know how the code has been changed, i.e. the lineage between the two points (even if they split in the past)	21:28
	← lirt left	21:28
	→ bgerber joined	21:28
aaronj	imachuchu: thx for your help	21:29
imachuchu	aaronj: since this repo is really bad (you can go take a look at it, there are many dead branches in the past that don't merge anywhere) git's confused on how to reconsile the differences	21:29
	← olaf_ left	21:29
imachuchu	aaronj: np, just remember that command, or more precisely "I am taking these commits one by one, reapplying them over here, and taking the bases changes wherever they conflict (which should be never)"	21:30
	aaronj: I guess "bases changes" should be "bases code"	21:30
	← Guest57735 left	21:30
	← hoschi left	21:31
	← rtjure left	21:31
	← agileadam left	21:31
	→ ashirus joined	21:31
qknight	FauxFaux: thanks	21:32
	→ sjefen6 joined	21:32
	← rehat left	21:32
	← api left	21:33
aaronj	imachuchu: nod	21:33
	← _Vi left	21:35
	→ _Vi joined	21:35
	← tplaner_ left	21:36
	← nwest left	21:37
	← eijk_ left	21:38
	← martinjlowm left	21:40
	→ pdtpatrick joined	21:40
mh`	can anyone confirm if 'github for mac' (the gui thing) respects local repository hooks?	21:42
	don't have a mac to test on..	21:42
	← juvenal left	21:43
	→ Chappo joined	21:43
cmn	it shells out for most thing	21:43
	→ w41 joined	21:44
	← luxurymode left	21:44
mh`	ok, so it should be fine i presume	21:44
	not doing anything tricky.. just concerned with pre-commit mostly	21:44
	← Element9 left	21:44
	→ yoklov joined	21:46
	← w41_ left	21:46
	← john2496 left	21:47
	→ rehat joined	21:48
	→ marhaban joined	21:48
m1sc	aaronj: why the second feature branch?	21:48
	← shreya left	21:48
	← Guest5287 left	21:49
	→ madewokherd joined	21:49
	← MrRoot left	21:49
	→ amcsi joined	21:49
m1sc	aaronj: it looks like you're terribly mixing up things	21:49
	→ kcj joined	21:50
	→ AaronMT joined	21:51
	→ maestrojed joined	21:51
	→ ChadStudmuffin joined	21:51
	→ opus joined	21:51
	← webguynow left	21:52
	→ diminoten joined	21:56
	← shawngo left	21:58
	→ FernandoBasso joined	22:00
	→ alegen joined	22:01
	→ alanp_ joined	22:01
	→ dfr\|mac joined	22:01
	← alanp left	22:01
	→ schemula joined	22:01
	oakdog8 → oakdog8_	22:02
	← victor` left	22:02
	→ gfontenot joined	22:03
	→ TheLZA joined	22:04
	← TheLZA left	22:04
	← ashirus left	22:04
	→ TheLZA joined	22:04
	← TheLZA left	22:04
	← sh0ne left	22:04
	← drizzd_ left	22:07
	→ graphitemaster joined	22:07
	← timthelion left	22:07
	← v0n left	22:08
	← Eridius left	22:08
	← gdmfsob left	22:08
mattalexx	I feel like I'm going crazy. I just pushed a branch to github and it's not there!	22:08
	I've triple checked that I'm pushing to the correct remote	22:08
	← pantsman left	22:08
	← revagomes left	22:09
m1sc	mattalexx: what's git ls-remote telling you?	22:09
cmn	and that you're pushing the right branch?	22:09
	← maestrojed left	22:09
gfontenot	looks like github has a backlog right now	22:10
	https://status.github.com	22:10
	← ddelrio1986 left	22:10
gfontenot	saw someone mentioning it on twitter, too	22:10
mattalexx	m1sc, The branch is in `git ls-remote`	22:11
	gfontenot, Thanks	22:11
	→ dfr\|mac_ joined	22:11
mattalexx	cmn, Yup	22:11
cmn	then it's just the webpage being slow	22:12
	← civilordergone left	22:12
	→ civilordergone joined	22:12
	→ TheLZA joined	22:12
	← TheLZA left	22:12
	→ TheLZA joined	22:13
	← TheLZA left	22:13
	→ TheLZA joined	22:13
	← TheLZA left	22:13
	← dfr\|mac left	22:14
	→ kfogel joined	22:15
	← kfogel left	22:15
	← ShadeTornado left	22:15
	→ PatF joined	22:16
	→ gigamonkey joined	22:16
	→ HyperNerdV2 joined	22:16
	→ tinti_ joined	22:16
	← yoklov left	22:18
	← alexisabril left	22:19
	← Destos left	22:19
	← kanru_ left	22:19
	← PatF left	22:21
	→ v0n joined	22:21
	← marhaban left	22:23
	→ SkAzZ joined	22:24
	→ giallu joined	22:24
	← bogwonch1 left	22:25
	→ bogwonch joined	22:25
	luke-jr_ → Luke-Jr	22:26
	← specialGuest left	22:27
	← etcetera left	22:27
	→ SkAzZ1 joined	22:27
	← macabre left	22:28
	→ lustucru_ joined	22:28
	← dv310p3r left	22:28
	← v0n left	22:28
	→ specialGuest joined	22:29
	→ nglorioso joined	22:29
	← tatsuya_o left	22:29
	→ Anubisss_ joined	22:29
	← SkAzZ left	22:29
	← lustucru left	22:29
	→ etcetera joined	22:30
	→ gordian joined	22:30
	← bogwonch left	22:30
	→ Anubisss__ joined	22:30
	gitinfo set mode: +v	22:30
	← gigamonkey left	22:32
	← Anubisss left	22:32
	→ mmc1 joined	22:33
	→ quotemstr joined	22:34
	← Anubisss_ left	22:34
quotemstr	Say I have a remote tracking branch trunk/master, but no local branch that tracks it. If I checkout trunk/master, I get into detached head mode. How do I create a local branch that tracks trunk/head?	22:34
FauxFaux	git checkout -t -b foo trunk/master, or, if "master" is unambigious, git checkout master.	22:35
	→ kanru_ joined	22:35
quotemstr	Thanks.	22:36
	→ ashirus joined	22:36
	← TeckniX left	22:36
	→ jagginess joined	22:38
	→ raykyri joined	22:38
	← lustucru_ left	22:38
jagginess	howdy	22:39
	← Purveyor left	22:39
	← cannonball left	22:39
jagginess	can someone explain to me what "subversion bridging command" does? < does this imply git<->git, I'm interested in cloning a local git	22:41
	← mike007 left	22:41
cmn	why would a subversion bridge imply git-to-git? what are you trying to do?	22:42
	→ lustucru joined	22:43
	→ Carsekyx joined	22:43
quotemstr	cmn: I think he means round-trip.	22:43
	← Synthead left	22:43
	← arietis left	22:44
	→ artnez joined	22:45
jagginess	I'm reading section 8.1.1 of the noobie manual i downloaded from the git site	22:45
	"The base command in Git for all the Subversion bridging commands is git svn."	22:45
	→ dsirijus joined	22:45
jagginess	thought a local git can be made to replicate/clone to a remote one..	22:45
	← Orbitrix left	22:45
cmn	the guide probably doesn't say "local git"	22:46
jagginess	nmind i found it..	22:46
cmn	do you want to grab info from a subversion repo or a git one?	22:46
	← jagginess left	22:46
	← specialGuest left	22:47
	→ specialGuest joined	22:47
	← tqmz left	22:47
	← dzonder left	22:48
	→ amalloy joined	22:49
amalloy	can i customize the initial/template contents of COMMIT_EDITMSG? i'd like another blank line before the "hints" at the bottom	22:50
	→ btanaka joined	22:50
	→ v0n joined	22:52
	→ simesy joined	22:53
	← hwrd\|work left	22:54
	← lacrymology left	22:54
	← SegFaultAX left	22:55
	→ artem joined	22:56
	← Carsekyx left	22:56
	→ jey joined	22:56
	→ bogwonch joined	22:56
	← tinti_ left	22:56
	→ tinti_ joined	22:56
jey	after I do a subtree merge, how can I see the full history for files that are in the merged subtree? by default I only see the merge commit when I do "git log -- subtree/filename"	22:56
	← dfr\|mac_ left	22:57
	← HyperNerdV2 left	22:57
	→ mbreslin joined	22:57
	← axl_ left	22:57
gordian	anyone have experience using git with Magento?	22:58
	← mritz_ left	22:58
FauxFaux	!anyone	22:58
	→ Lord_DeathMatch joined	22:58
jey	if I add "--follow" it gets worse -- I see no output. If I try "--full-history" I get the same output as when I omit it	22:58
	(i.e. only see the merge commit)	22:58
	← lupo7 left	22:59
FauxFaux	jey: git log -M -C #?	22:59
	← artnez left	23:00
jey	no luck. I'm starting to wonder whether I screwed up the merge commit in some way	23:00
	← artem left	23:00
	← nglorioso left	23:00
jey	s/merge commit/subtree merge/	23:00
	→ artnez joined	23:00
	← savid left	23:00
	← tweedledumber left	23:01
	→ Destos joined	23:01
	← tinti left	23:02
	→ SegFaultAX joined	23:03
	← opus left	23:03
krazyj	hey everyone… i'm looking for a workflow where my commits that are made in between github pushes are saved to a remote server… then, when I'm ready to push for github, i can squash it all and push to github	23:03
	← thinice left	23:03
krazyj	any ideas how that would be architected?	23:03
	← berserkr left	23:04
	→ gitinfo joined	23:05
	ChanServ set mode: +o	23:05
	← bogwonch left	23:05
imachuchu	krazyj: have an external mirror server (for backup purposes) then push to github as normal?	23:06
krazyj	imachuchu: yeah	23:06
	← v0n left	23:06
krazyj	and, preferably, squash all the commits, when pushing to github	23:06
	→ nglorioso joined	23:07
FauxFaux	You can just maintain a branch which you push to github from that you just merge --squash your real branch into.	23:07
	Or just dirtymerge it with commit-tree. </a FauxFaux exclusive solution>	23:07
	← amalloy left	23:07
imachuchu	krazyj: rebase -i	23:07
	← sinful_ left	23:08
	← buhman left	23:09
	← jjbohn left	23:10
	← ehsan left	23:10
	← stkrzysiak left	23:11
	← zastern left	23:13
	SeoZ-work[AWAY] → SeoZ	23:14
	← robottinosino left	23:14
	→ Fernando1asso joined	23:15
	← FernandoBasso left	23:15
	← darren__ left	23:15
	← chrisgeorge left	23:16
	← raykyri left	23:17
	← chrisdev left	23:17
	← giallu left	23:18
	→ jjang joined	23:19
	→ v0n joined	23:19
	← lresende_ left	23:20
	→ jjang_ joined	23:22
	→ eletuchy joined	23:22
	← xrq left	23:22
	← jjang left	23:22
	→ Anders_J joined	23:25
	← Sky[x] left	23:25
	→ Jay_Levitt joined	23:26
	← mmc1 left	23:28
	← flaguy48 left	23:28
	→ sitaram joined	23:28
	← leila_ left	23:29
	→ Esa joined	23:30
	← fly9 left	23:30
	← Textmode left	23:31
	← SirCmpwn left	23:31
	→ iliketurtles joined	23:31
	← etcetera left	23:31
	← vervic left	23:32
	← beneggett left	23:33
	→ opus joined	23:34
	← Jay_Levitt left	23:34
	← gordian left	23:36
	← jh3 left	23:36
	→ toky joined	23:36
toky	hello	23:36
gitinfo	toky: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	23:36
toky	Is there how-to somewhere to install git on a new centos ami?	23:37
	→ BiggFREE joined	23:37
toky	i do not mean create a git repo	23:37
	→ yoklov joined	23:38
	→ HyperNerdV2 joined	23:38
	← gavin_huang left	23:38
cmn	what's wrong with yum install git?	23:38
toky	not sure if im asking the right question...	23:39
	we have git account and need to standup our git server (new one)	23:39
	→ cwang joined	23:39
cmn	a git account meaning what?	23:39
	git doesn't deal with accounts	23:39
toky	i've been given access to some kind of gihub blessed AMI in which im supposed to install a ghp package	23:40
cmn	then you'll want to ask github	23:40
EugeneKay	Github enterprise? You'll need to talk to their suppor.	23:40
toky	there is no one here from github?	23:40
	← jobin left	23:40
EugeneKay	No. We deal with the open source git.	23:41
thiago	they usually are in #github	23:41
milki	github doesnt use irc as an official support channel	23:41
cmn	github is off-topic here, this is git	23:41
	← specialGuest left	23:41
imachuchu	git and github are two seperate things	23:41
EugeneKay	#github is usually not staffed by any githubbers proper; just users.	23:41
imachuchu	but I think there is a github channel	23:41
cmn	you can try #github or sending them an e-mail	23:41
EugeneKay	Contact support.	23:41
	← dzan left	23:41
cmn	they probably have documentation on this, though	23:41
	→ SyouJyoA joined	23:41
imachuchu	toky: sorry, it's just a lot/most of us don't have expieriance with github, just the use of the software git	23:42
	← thelazydeveloper left	23:42
	← ninegrid left	23:42
toky	i thought more ppl would have setup the enterprise one...	23:43
	→ cakehero joined	23:43
	← Psycho_pr1 left	23:43
	→ boombatower_ joined	23:43
	→ Psycho_pr joined	23:43
	→ thursday joined	23:43
	thiago doesn't even have an account on github	23:43
	→ dzan joined	23:43
toky	i'll keep dissecting their AMI...I just hoped someone would've done it before	23:43
thiago	much less enterprise	23:43
imachuchu	toky: it's not an issue of enterprise or consumer, it's that (for me at least) I've never really used github	23:44
thiago	I don't even know what you mean by AMI	23:44
	thiago is thinking of whoami	23:44
jey	where does the stdout of git commands go? I'm looking at the source code to apenwarr's git-subtree.sh, and it has "echo" commands that I don't see on my screen	23:44
	← rehat left	23:45
thiago	jey: usually, stdout goes to stdout	23:45
	are you sure the echoes were executed?	23:45
toky	thiago: an AWS AMI	23:45
	← zackiv31 left	23:45
thiago	toky: Amazon Web Service?	23:46
	← Destos left	23:46
toky	yes	23:46
thiago	well, I know AWS. I have no idea what AMI is.	23:46
	→ leila_ joined	23:46
	→ metcalfc joined	23:46
	→ etcetera joined	23:46
toky	thiago: AMIs are pretty basic part of AWS	23:47
	← boombatower_ left	23:47
	→ Destos joined	23:48
	→ Fangorn__ joined	23:48
	← Fangorn_ left	23:49
cmn	thiago: it's a pre-build VM image	23:50
	pre-built	23:50
	→ Textmode joined	23:51
	← thursday left	23:52
	← imachuchu left	23:55
	→ MrRoot joined	23:56
	→ eltigre joined	23:56
	← gusnan left	23:57
eltigre	hey, I need some help. I tried to setup a ssh key for github, but git is still asking me for my username and password when interacting with the remote repository	23:58
	any ideas?	23:58
	and ssh [email@hidden.address] correctly identifies me	23:59

Logs Search ←Prev date Next date→

Channels Documentation