IRCloggy #git 2013-03-15

Provider of IRC logs since 2005.

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

	2013-03-15
greymalkin	Anyone know how to manually check a commit/tag's signature (without using verify-tag, I want to use my own GPG options)	00:00
	← shawnj left	00:00
	← erty left	00:01
milki	you can probably either configure a gpg program or pass it in as an option	00:01
Adys	I have a bunch of commits that begin with "project_name: ". I want to remove that bit. any idea how?	00:02
	→ huoxito joined	00:02
	← p3rror left	00:02
	← pastime left	00:03
milki	Adys: you can read up on man git filter-branch	00:03
gitinfo	Adys: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	00:03
Adys	I'm there right now I've never been good with it :P	00:04
	I figured i need msg-filter	00:04
milki	sounds reasonable	00:04
	← Sonderblade left	00:05
	→ v0n joined	00:05
andres	greymalkin: yea, otherwise you would need to parse the tags content, split at the gpg part and then verify the first part against the second	00:06
	doesn't sound like fun to me.	00:06
Seveas	bazz: you disappeared for 6 hours :)	00:06
	← b1rkh0ff left	00:07
	← jn_ left	00:07
Seveas	bazz: anyway, I said: then do it differently: checkout their new branch, create a new branch of your own based on that and cherry-pick your patch on your new branch	00:07
	← m0viefreak left	00:08
grawity	greymalkin: take a look at contrib/examples/git-verify-tag.sh and contrib/examples/git-tag.sh	00:08
spyro	what's the ETA on a release for 2.0?	00:08
Seveas	spyro: no ETA, no plan afaik	00:09
greymalkin	grawity: What is this... you're in almost every channel I've asked a question on for the last few months :)	00:09
milki	spyro: over 9000	00:09
	← dsdeiz left	00:09
spyro	har har har	00:09
	→ kenperkins joined	00:09
	→ madewokherd joined	00:09
spyro	Just curious how much longer I'm going to be spammed with a warning about push.default	00:09
milki	until you choose a default?	00:10
	← kenperkins left	00:10
	← mario-goulart left	00:10
	→ kenperkins joined	00:10
	← hwrd\|work left	00:11
	← kmels__ left	00:14
greymalkin	grawity: But, as usual, that's exactly what I was looking for	00:15
	→ ErKa joined	00:15
	← marhaban left	00:15
rawtaz	is there any specific reason that the --staged alias which git diff has isnt available for git rm as well?	00:16
	it would kinda make sense	00:16
	if it was, i mean	00:16
milki	rawtaz: either no one cares, its actually there, someone forgot, its already patched, etc	00:17
rawtaz	milki: all that would fall under no specific reason imo, but i see what you mean. afayk there's no super specific reason i take it	00:18
	like, it hasnt been decided NOT to be done	00:18
	milki shrugs	00:18
	← jjore left	00:18
	→ bitkiller joined	00:18
rawtaz	from a usability perspective its a valid question :)	00:19
	→ pastime joined	00:19
	→ savid joined	00:19
	→ tplaner joined	00:19
	→ xpen_ joined	00:19
milki	provide a patch to the git developer mailing list	00:19
rawtaz	i was just asking dude	00:20
	;)	00:20
	milki falls under doesnt really care	00:20
rawtaz	yeah, the tone of your answers have made that clear already, thanks :)	00:20
	milki uses --cached	00:20
rawtaz	thanks though	00:21
milki	np	00:21
	git isnt really designed	00:21
rawtaz	hehe yeah	00:21
	i like that summary :D	00:21
milki	lol	00:22
	→ tmcmahon joined	00:22
	← xpen left	00:22
	← chuckharmston left	00:25
	→ boombatower joined	00:27
	→ m0viefreak joined	00:29
	→ hwrd\|work joined	00:29
Adys	i just split with filter-branch a bit of my project into its own project. but Im left with an old empty merge as the first commit, like this: http://paste2.org/p/3157531	00:29
	i want to get rid of it, any idea?	00:29
	→ Cakey joined	00:29
	← tplaner left	00:29
cbreak	Adys: you can use grafts to remove that parent, and then filter-branch to bake the graft	00:30
	or you can use a parent filter	00:30
Adys	grafts?	00:30
	← hwrd\|work left	00:31
cbreak	https://git.wiki.kernel.org/index.php/GraftPoint	00:31
	you'd make it look like the parent you would like to have has no parent	00:32
	→ infynyxx joined	00:32
Adys	thats neat, though i have no idea how to use that	00:32
	← savid left	00:32
cbreak	just put the sha you want to graft into that file	00:33
rking	So, I think it's a pretty bad practice to mash the giant green "Merge This Without Verifying It" button on Github. What is a more correct process, though? Should I merge it locally, verify it, then do the 'signed-off-by' when I commit and go back to the Github UI to close the Pull Req?	00:33
	← _FrnchFrgg_ left	00:33
	← hyperair left	00:33
Adys	cbreak: so if i understand this correctly, i put the 2nd commit into the file with no parent and it gets rid of the first?	00:34
cbreak	yes	00:34
	it should look as if it was in effect immediately	00:34
	and if you like it, you can filter-branch to bake it	00:34
Adys	very neat	00:34
rawtaz	the index == the staging area, right?	00:34
Adys	cbreak: how?	00:34
cbreak	(note that I only did that once, ages ago)	00:34
	just filter	00:34
Adys	oh cool	00:35
cbreak	maybe with a tag name filter if you have tags	00:35
	rawtaz: yes.	00:35
rawtaz	thanks	00:35
	← giallu left	00:35
	← TheFuzzball left	00:35
Adys	thanks a lot cbreak, that's the whole project cleaned up :)	00:35
	← yshh left	00:35
cbreak	np :)	00:36
	as said above, using --parent-filter would be an alternative to grafts	00:37
	but I think it's harder to use	00:37
	(well, I suppose in this case it'd be an inverse grep ...)	00:38
	← samholmes left	00:39
	← dmarkham_ left	00:39
	→ drewbert joined	00:40
	→ tplaner joined	00:40
	← arkady91 left	00:41
	← rakl left	00:42
	← miso1337 left	00:42
	← jdunck left	00:42
	← Bass10 left	00:44
	→ Jojomofoman joined	00:44
	← devth left	00:45
rawtaz	theres one thing i dont understand regarding file moving, reading http://git-scm.com/book/en/Git-Basics-Recording-Changes-to-the-Repository - if git tracks files by content, then why cant i just do a regular `mv foo bar` in the OS and git will know that this is the same file? is it cuz of the possibility that another file with identical content was added and this first one removed, that it doesnt assume things?	00:45
	← bitkiller left	00:45
	← ozette left	00:46
rawtaz	having read about git tracking contents makes you surprised when you read that you still need to tell git about renames/moves	00:46
	just noting	00:46
	→ bitkiller joined	00:48
	← bitkiller left	00:48
	← aberrant left	00:49
	→ bitkiller joined	00:49
rawtaz	also, is it impossible in git to remove file A and add file B and NOT have it be seen as a rename by git, when the files have the same content?	00:50
spyro	rawtaz: Because git needs to be informed of the change in index contents	00:51
	← tplaner left	00:51
haarg	rawtaz: you can just do a mv foo bar. but you still need to update the index with that information.	00:52
rawtaz	haarg: yeah, as in the book	00:52
haarg	mv foo bar; git rm foo; git add bar # git will see this the same as git mv foo bar	00:52
	← dreiss left	00:52
rawtaz	spyro: i see	00:52
spyro	so unless you "git add" the new file, git doesn't even know the new file even exists	00:53
	← apok left	00:53
rawtaz	spyro: do you mean regarding my first question now, manually doing a regular OS-level `mv` ?	00:53
	→ wrouesnel joined	00:53
spyro	if you play with files and don't tell git, git won't feel safe making assumptions	00:54
	← jcromartie left	00:54
rawtaz	yeah	00:54
	→ tplaner joined	00:54
rawtaz	but what about the second question? that one is just curiousity	00:54
	→ tongcx joined	00:54
rawtaz	i.e. whether it is possible to delete file A and add file B in the same commit, them having the same contents, without git seeing that as a rename	00:55
	seeing/treating	00:55
haarg	git doesn't actually record renames anywhere	00:55
	→ fire__ joined	00:55
rawtaz	exactly. it goes by content	00:55
	but if you do git rm a; git add b then git will treat that as a rename, from what i can tell	00:55
haarg	it's all in the tools that are looking at the history afterwards. git blame you can tell if it should track changes across renames.	00:55
	← fire left	00:56
rawtaz	so im wondering if one can make git not treat that as a rename, and instead see it for what it is (in this example), a regular adding of one file and removing another	00:56
haarg	same with git log. it has --follow if you want it to try to follow history across renames.	00:56
	→ JeremyWei joined	00:57
	← mitz left	00:57
haarg	as i said, it depends on what tool you are using to look at the history.	00:57
	git log by default won't track across renames	00:57
	← scelis left	00:57
rawtaz	haarg: right.. but at this point i am only at the commit stage (in my reasoning). git status says it is a rename of that/those files. im not seeing a parameter to `git status` that controls how to treat (potential) renames	00:58
haarg	i'm not aware of any option to status to control that	00:59
rawtaz	ok	00:59
haarg	but it sounds pretty useless	00:59
rawtaz	but i do see your point. the perception from git that this is a rename is cosmetic.	00:59
haarg	yes	01:00
	← pastime left	01:00
	→ pastime joined	01:01
Adys	cbreak: and done https://github.com/Adys/django-push-notifications/commits/master	01:01
	→ jcromartie joined	01:01
	→ R1KRDO joined	01:01
	← maletor left	01:01
	← tmcmahon left	01:01
	→ lavaman joined	01:02
	→ private joined	01:02
	private → Guest71029	01:02
	← pastime left	01:04
	← tplaner left	01:04
	← Thra11 left	01:05
	→ dreiss joined	01:05
	← pxdr left	01:06
	→ hwrd\|work joined	01:06
	→ pxdr joined	01:06
	← undecim_ left	01:07
	→ undecim_ joined	01:07
	→ kaylee joined	01:07
	→ pastime_ joined	01:09
	→ cwinebrinner joined	01:09
	← Jahm left	01:09
	← dreiss left	01:10
	← kaylee left	01:10
cwinebrinner	I added some files and committed. Very stupidly, I accidentally deleted the files and amended them to the previous commit. Is there any way to recover them?	01:10
	→ mloskot joined	01:11
	← mloskot left	01:11
	→ mloskot joined	01:11
RandalSchwartz	git checkout HEAD@{1} or something	01:11
	← pastime_ left	01:11
RandalSchwartz	git reset --hard HEAD@{1}	01:11
	← v0n left	01:11
RandalSchwartz	might do it	01:11
	→ nbari joined	01:11
	→ pastime joined	01:11
cwinebrinner	RandalSchwartz: I'll give it a try. Thanks.	01:12
rawtaz	cwinebrinner: i think it might be helpful to `git reflog` to see the things RandalSchwartz is referencing	01:12
RandalSchwartz	do a git reflog	01:12
	yeah	01:12
	rawtaz is speculating	01:12
RandalSchwartz	git reflog HEAD	01:12
rawtaz	oh :)	01:12
	rawtaz claps	01:12
RandalSchwartz	see which commit you want	01:12
	→ yshh joined	01:12
	→ Jahm joined	01:12
RandalSchwartz	then git reset --hard that	01:12
	→ v0n joined	01:12
cwinebrinner	My saviors =)	01:13
rawtaz	♥	01:13
	← R1KRDO left	01:14
cwinebrinner	Thank you both so very much.	01:15
	RandalSchwartz bows	01:15
	RandalSchwartz hits head on desk	01:15
RandalSchwartz	ouch!	01:15
	← penguin1 left	01:15
rawtaz	lol	01:16
ari-_-e	really the savior is git :)	01:16
rawtaz	ari-_-e: shh, dont tell!	01:16
	actually we cant have that rationale. with it linus would be the savior, or his mom!	01:16
	→ hyperair joined	01:16
ari-_-e	really the savior is the big bang	01:17
	← pr0ton left	01:17
	→ sgronblom joined	01:17
	← tongcx left	01:17
rawtaz	word.	01:17
ari-_-e	taking that to its logical conclusion	01:17
	← flavius left	01:17
	→ stliu joined	01:18
	→ barq joined	01:18
	→ beakman joined	01:18
	← etcetera left	01:19
	← NiteRain left	01:19
	← Guest50399 left	01:20
	→ akiress joined	01:21
	akiress → Guest7404	01:21
	← Octalot left	01:22
	← nbari left	01:22
	← Guest7404 left	01:23
	← elixir left	01:24
	→ akiress joined	01:24
	→ miso1337 joined	01:26
	→ etcetera joined	01:26
	→ dhruvasagar joined	01:27
	← hwrd\|work left	01:28
relipse	Jesus is the Savior	01:30
	→ hwrd\|work joined	01:30
	→ jaequery joined	01:31
mikehaas763	Anyone know about when git got tab completion? I just installed git in my dev environment on debian 6 squeeze from the standard apt-get repo and there is not tab completion. Super bothersome	01:32
imMute	mikehaas763: looong time ago	01:32
RandalSchwartz	depends on the shell too	01:32
imMute	make sure you source /etc/bash_completion.d/git (or thereabouts) in your .bashrc	01:32
	← AaronMT left	01:32
RandalSchwartz	I'm hoping bash is never my default shell. :)	01:32
imMute	O.o	01:33
mikehaas763	I love bash :(	01:33
RandalSchwartz	zsh for the win	01:33
mikehaas763	Anyways, you're right. Surprising that it didn't do that automatically. It's a nice touch on Ubuntu	01:33
	← drewbert left	01:34
	→ Goplat joined	01:34
	→ macrover joined	01:34
mikehaas763	I've barely played with zsh. What does it offer that bash doesn't?	01:34
	→ rakl joined	01:34
gtmanfred	mikehaas763: http://ix.io/2O8	01:35
	→ rchavik joined	01:36
	← josefig left	01:36
	← mastro left	01:36
gtmanfred	mikehaas763: http://ompldr.org/vaHJoaA	01:36
	plus massively expanded parameter expansion	01:37
mikehaas763	that font hurts my eyes	01:37
gtmanfred	http://ompldr.org/vaHJoag	01:37
mikehaas763	..I'm sure its customizable	01:37
	→ dsdeiz joined	01:37
gtmanfred	and tab completion with descriptions	01:37
	← boombatower left	01:38
	← xpen_ left	01:39
gtmanfred	and glob expansion http://ompldr.org/vaHJobA	01:39
	→ deepinskin joined	01:40
	→ boombatower joined	01:40
	→ aberrant joined	01:40
	→ SBoudrias joined	01:40
	→ Progster joined	01:41
	← aberrant left	01:41
gtmanfred	http://code.gtmanfred.com/cgit/dotfiles.git/tree/zsh and http://code.gtmanfred.com/cgit/dotfiles.git/tree/zshrc	01:41
	→ rurufufuss joined	01:42
gtmanfred	mikehaas763: i also really like hash, hash -d /srv/git/pacman.git=pacman.git; ... then i can just cd ~pacman.git from anywhere	01:42
	← barq left	01:42
	← boombatower left	01:42
	← quazimodo left	01:43
gtmanfred	i probably spent way to much time setting up my shell...	01:43
	→ boombatower joined	01:43
	← wrouesnel left	01:44
	→ Vert joined	01:44
gtmanfred	i particularly like the table flips on return values of anything besides 0	01:44
	← kcj left	01:45
	← notVert left	01:45
	→ quazimodo joined	01:45
	→ thiago joined	01:45
	→ pishposh joined	01:46
	← lavaman left	01:46
	← hwrd\|work left	01:46
	← d2dchat left	01:46
pishposh	what up blood! what does master in 'git push origin master' mean? where does my file go?	01:47
	→ dreiss joined	01:47
mikehaas763	it means you are pushing the 'master' branch to the remote setup and name origin	01:47
	master and origin are basically the default names you get	01:47
	but you can set up other remotes and new branches	01:47
	← milleja46 left	01:48
mikehaas763	*pushing it to the remote named origin	01:48
pishposh	mikehaas763: master being the path to my file on my machine?	01:48
mikehaas763	nope, git has a concept called branches	01:48
	type git branch and you will probably just see master. That's because you don't have other branches set up.	01:49
	→ jpcamara joined	01:49
	→ milleja46 joined	01:49
mikehaas763	http://git-scm.com/book/en/Git-Branching-Basic-Branching-and-Merging	01:49
	That's a good introduction to branching	01:49
	← infynyxx left	01:49
pishposh	mikehaas763: k lemme see what you have here	01:50
	← soypirate left	01:50
	→ Chryyz joined	01:50
	← rakl left	01:52
	← etcetera left	01:53
	→ mastro joined	01:53
	→ luc_ joined	01:53
	→ ojacobson_ joined	01:54
	← ojacobson left	01:55
	ojacobson_ → ojacobson	01:55
	← dhruvasagar left	01:56
	← oriba left	01:56
	→ pr0ton joined	01:57
	→ jpcamara_ joined	02:00
	← goshawk left	02:01
	→ dydave joined	02:01
	← jpcamara left	02:01
	← Cakey left	02:02
	→ Cakey joined	02:02
	← pr0ton left	02:02
	← rurufufuss left	02:03
	→ cryosphere joined	02:04
	→ Watabou joined	02:05
	→ rakl joined	02:05
	← mloskot left	02:05
	← jwmann left	02:05
	→ kadoban joined	02:07
	→ jwmann joined	02:08
	← v0n left	02:09
	← daniel_-_ left	02:10
	→ luc__ joined	02:10
	→ angelsl joined	02:10
	← luc__ left	02:10
	← luc_ left	02:11
	→ pr0ton joined	02:11
	→ luc__ joined	02:11
	← yankeefan04 left	02:12
	→ yankeefan04 joined	02:12
	→ intripoon_ joined	02:12
	← pr0ton left	02:14
	← intripoon left	02:15
	← p3lim left	02:16
	← ErKa left	02:18
	→ hwrd\|work joined	02:18
	← Asenar left	02:20
	← ehsan left	02:21
	→ parallel21 joined	02:22
	→ Asenar joined	02:22
	← Jojomofoman left	02:22
parallel21	Just converted a repo from svn to git and I'll still have a branch title "remotes/trunk"	02:23
	I'm trying to delete it, but nothing seems to find it	02:23
	← yankeefan04 left	02:23
thiago	git update-ref -d refs/remotes/trunk	02:24
	← fuzzybunny69y left	02:25
	← jaequery left	02:25
parallel21	Yes!	02:25
	thiago: Thank you!	02:25
cmn	that's one way, but branch -r should be able to find it; what's this "nothing" that can't find it	02:25
parallel21	branch -r listed it	02:26
	→ erkules_ joined	02:26
parallel21	But how does one remove it with -r	02:26
thiago	branch -r -d didn't do it?	02:26
cmn	the same way as without, you add -d	02:26
parallel21	I didn't try that… but am making note of both commands. Thanks for the tip	02:27
	→ xiangfu joined	02:27
cmn	so what did you mean when you said nothing could find it?	02:28
	← erkules left	02:28
	← jwmann left	02:29
parallel21	I kept getting errors saying error: error: refname refs/heads/trunk not found	02:29
cmn	because that branch doesn't exist, but you still haven't said what you did	02:30
	→ Vampire0 joined	02:30
	→ solusfaran joined	02:31
	← Vampire0_ left	02:33
	← kpreid left	02:33
	→ kpreid joined	02:33
	→ jpcamara joined	02:34
	← jpcamara_ left	02:35
parallel21	So that error was from git branch -m trunk master	02:35
	I thought I'd rename it if I could not get rid of it	02:36
	→ normanrichards joined	02:38
	→ infynyxx joined	02:38
	→ hafeez3000 joined	02:39
	→ NiteRain joined	02:41
	← kadoban left	02:41
parallel21	None of the commands I was using could find it	02:45
	← parallel21 left	02:45
	→ luc_ joined	02:47
	→ dhruvasagar joined	02:47
	← tvw left	02:48
	← luc_ left	02:48
	← tbmiller left	02:49
	→ jaequery joined	02:49
	→ luc_ joined	02:49
	← luc__ left	02:49
	← pastime left	02:50
	← bitkiller left	02:50
	← infynyxx left	02:52
	→ raspy_freenode joined	02:53
	→ theotherguy joined	02:54
	← Noldorin left	02:55
	→ ehsan joined	02:56
	← Guest71449 left	02:56
	→ Pigeon joined	02:56
	→ xpen joined	02:56
	Pigeon → Guest43706	02:56
	← acl_ left	02:57
	← SonikBoom left	02:57
	→ Slush- joined	02:59
	← pppaul left	02:59
	→ jonkratz joined	02:59
	→ kadoban joined	03:00
	→ infynyxx joined	03:00
	← xpen left	03:01
	→ luc__ joined	03:02
	→ JDuke128 joined	03:02
	→ Noldorin joined	03:02
	→ bitkiller joined	03:02
	← SBoudrias left	03:03
	← mastro left	03:04
	← hyperair left	03:05
	← luc_ left	03:05
	→ frenchlaf joined	03:05
	→ hyperair joined	03:05
	→ pastime joined	03:05
	← macrover left	03:05
	→ brundlfly joined	03:06
	← orafu left	03:07
	← kadoban left	03:07
	→ orafu joined	03:07
	→ averiso joined	03:07
brundlfly	what is meant by the error REJECTED_NONFASTFORWARD occurs?	03:08
	← xiangfu left	03:08
	→ Orbitrix joined	03:08
EugeneKay	!ff	03:09
gitinfo	A fast-forward merge occurs when you merge a commit which is a descendant of !HEAD. No new commit is created, instead the branch is simply moved forward. See http://sandofsky.com/images/fast_forward.pdf	03:09
offby1	that means that the remote has "diverged" from you -- or vice versa	03:09
	that too	03:09
EugeneKay	It means it isn't one of those ^. You need to create a merge.	03:09
	offby1 hasn't gotten used to these here new-fangled robots	03:09
EugeneKay	Try `git pull` then push again	03:09
	→ xiangfu joined	03:10
	← infynyxx left	03:11
	→ mnb joined	03:11
	← raspy_freenode left	03:13
	← hwrd\|work left	03:16
	← mnb left	03:17
	← lindenle left	03:18
	→ mastro joined	03:18
brundlfly	I cant push my master branch to my github account, any other branch works fine. Is there anything else I can try, Sorry I am new to git.	03:19
	← lacrymology left	03:19
	← dhruvasagar left	03:19
	← ojacobson left	03:19
	← frenchlaf left	03:19
	← eletuchy left	03:20
SamB	brundlfly: what happens when you try?	03:20
brundlfly	same issue REJECTED_NONFASTFORWARD (Using git in Netbeans 7.3)	03:21
	created a branch and it works fine	03:21
	but not with the master branch	03:21
	← miso1337 left	03:22
	← rakl left	03:23
	→ engkur joined	03:23
	← rektide left	03:24
	← chrisgeorge left	03:26
	→ josefig joined	03:27
	→ mikeym_ joined	03:27
	← ktcsoz left	03:27
	← Progster left	03:28
	→ dmarkham_ joined	03:28
	→ ktcsoz joined	03:28
	← sunny256 left	03:29
	→ sunny256 joined	03:30
	→ yacks joined	03:31
	← apol left	03:32
	→ lindenle joined	03:32
	← itatitat left	03:33
cmn	brundlfly: see what man git push says about non fast-forwards	03:34
gitinfo	brundlfly: the git-push manpage is available at http://jk.gs/git-push.html	03:34
	→ Davey joined	03:35
	→ taylorlapeyre joined	03:35
brundlfly	I appreciate your help.	03:35
	← akiress left	03:35
	← Anders_J left	03:36
	← Orbitrix left	03:38
	← solusfaran left	03:38
	← hafeez3000 left	03:39
	→ Parhs8 joined	03:44
	→ miso1337 joined	03:45
	→ hwrd\|work joined	03:46
	← Bankzilla left	03:47
	→ mihamina1 joined	03:47
	← JDuke128 left	03:48
	← Noldorin left	03:49
	→ JDuke128 joined	03:50
	← mihamina left	03:51
offby1	SamB: God bless you for asking "what happens when you try"	03:52
	→ Noldorin joined	03:52
	← jonkratz left	03:52
SamB	I thought that might be it ...	03:53
	← Noldorin left	03:53
	→ Noldorin joined	03:54
	← dpac\|away left	03:54
	← ktcsoz left	03:55
	← Noldorin left	03:55
	→ Noldorin joined	03:56
	→ dpac joined	03:56
brundlfly	I pushed my branch but there is a different user name displayed as author. This user has not even branched the repository. What should I do?	03:56
	← Noldorin left	03:57
	← jcromartie left	03:57
	→ Noldorin joined	03:58
	← jpcamara left	03:58
	← bitkiller left	03:59
	→ maqr joined	03:59
	← Noldorin left	04:00
	→ Noldorin joined	04:00
	→ chuckharmston joined	04:00
	→ Error404NotFound joined	04:01
	→ m0ltar joined	04:01
	dpac → dpac\|away	04:01
	← Noldorin left	04:02
	← mihamina1 left	04:02
	← Cakey left	04:02
	→ theother_ joined	04:02
	→ Noldorin joined	04:02
	→ Cakey joined	04:03
	← Noldorin left	04:04
	→ luc_ joined	04:04
	→ beneggett joined	04:04
	→ Noldorin joined	04:05
	→ kishorebabu joined	04:05
	← luc__ left	04:05
offby1	fix your commit, then re-push with -f. Or else: live with it.	04:06
	!rewrite	04:06
gitinfo	[!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to). If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstream_rebase	04:06
	← Noldorin left	04:06
offby1	brundlfly: the user name displayed is part of the commit you pushed; it has nothing to do with the process of pushing. If you examine the commit on your local disk, you'll see the same user name.	04:06
	→ Noldorin joined	04:07
	← Guest43706 left	04:07
	→ beautifulmind joined	04:07
	→ Pigeon joined	04:07
	→ OOPMan joined	04:07
	→ jdyin joined	04:08
	Pigeon → Guest54594	04:08
	← Noldorin left	04:08
	← fire__ left	04:08
	→ Noldorin joined	04:09
	← huoxito left	04:09
	← miso1337 left	04:09
	→ hatake joined	04:09
	→ shawnj joined	04:09
	← hatake left	04:10
	← Noldorin left	04:10
	→ Noldorin joined	04:11
	← Noldorin left	04:12
	→ danishkhan joined	04:13
rking	I could use a little help with the '--signoff' usage. I have a commit that I (laboriously) reviewed, and I want to set it apart from the sloppy Pull Requests that get merged.	04:13
	I can easily merge it onto 'master', but now I'm confused about how to sign it.	04:13
	→ Noldorin joined	04:13
rking	I was going down the road of 'git format-patch' with the intention of 'git am --signoff', but I was having a bit of trouble and also it seems like a weird roundabout process.	04:14
	← ISF_ left	04:14
	← Noldorin left	04:15
	← chuckharmston left	04:15
	→ marcdel joined	04:15
	→ Noldorin joined	04:15
	→ eletuchy joined	04:16
	shawnj → shawnj\|away	04:16
	← Noldorin left	04:17
	← hwrd\|work left	04:18
rking	Aha, I think I see how it's supposed to go. This page had some clues: http://wiki.koha-community.org/wiki/Sign_off_on_patches	04:19
	→ lresende joined	04:19
	→ irqq joined	04:19
rking	I'll try it from the top	04:19
	→ mezner joined	04:19
	← Wulf left	04:19
	→ jeffsi joined	04:20
	→ Tao joined	04:21
	← cryosphere left	04:21
	← danishkhan left	04:22
	shawnj\|away → shawnj	04:22
	← shawnj left	04:22
	← marcdel left	04:23
	→ marcdel joined	04:23
	→ Noldorin joined	04:24
	→ miso1337 joined	04:25
	→ huoxito joined	04:25
	→ lresende_ joined	04:25
	← lresende left	04:26
	lresende_ → lresende	04:26
	→ sushant1 joined	04:26
	← L-Chymera left	04:27
	← Noldorin left	04:29
	← eletuchy left	04:29
	← taylorlapeyre left	04:31
	→ yankeefan04 joined	04:31
	← paddymahoney left	04:32
rking	OK, so this is halfway decent. Still has room for streamlining, but:	04:32
	<verify branch foo is good>; git checkout master && git format-patch -s ..foo && git am 00*.patch	04:33
	→ jday_ joined	04:34
rking	Then I edited it to say "Merged #123" hoping to make Github respond to like I had clicked "Merge this Pull Request"	04:34
	→ hwrd\|work joined	04:36
	← markalanevans left	04:36
	→ v0n joined	04:37
	← NiteRain left	04:38
	← beakman left	04:38
	← JeremyWei left	04:38
	→ harshpb joined	04:38
	→ eletuchy joined	04:39
	← dsdeiz left	04:42
	← boombatower left	04:42
	→ dsdeiz joined	04:42
	→ pppaul joined	04:43
	← eletuchy left	04:43
	→ vedant joined	04:44
	← hwrd\|work left	04:45
	← v0n left	04:45
	→ preflex_ joined	04:46
	← preflex left	04:47
	preflex_ → preflex	04:47
	→ shezbaig_wk joined	04:47
	← shezbaig_wk left	04:47
	→ arvind_khadri joined	04:48
	→ gamja9e joined	04:49
	← gamja9e left	04:51
	→ Raging_Hog joined	04:51
	→ gamja9e joined	04:51
	← gamja9e left	04:51
	← jday_ left	04:52
	← pastime left	04:52
	→ hafeez3000 joined	04:54
	← Davey left	04:54
	← jeffsi left	04:54
	→ dhruvasagar joined	04:57
	← miso1337 left	04:58
	→ s00pcan joined	04:58
	← Raging_Hog left	04:58
	→ Turicas joined	04:58
	← mezner left	04:59
	→ JeremyWei joined	04:59
	→ eletuchy joined	05:00
	→ chuckharmston joined	05:02
	→ robbyoconnor joined	05:03
	← chuckharmston left	05:05
	→ BryanRuiz1_ joined	05:06
BryanRuiz1_	how do i say last 3 commits? HEAD^^^ ?	05:06
	← angelsl left	05:08
	→ jave joined	05:10
	← dmarkham_ left	05:12
	← mikehaas763 left	05:12
	→ pretty_function joined	05:13
	← Mannequin left	05:14
	→ Noldorin joined	05:14
	← oste left	05:14
	→ boombatower joined	05:15
	← Noldorin left	05:15
	→ darnir joined	05:16
	← krazyj left	05:16
	→ Noldorin joined	05:16
	← Noldorin left	05:18
	→ xaocon\|away joined	05:18
darnir	How to ignore all files except those that are currently tracked?	05:18
	→ Noldorin joined	05:18
	→ jjore joined	05:20
	← Noldorin left	05:20
	→ hoschi joined	05:21
	→ Noldorin joined	05:21
	← Noldorin left	05:22
	→ Noldorin joined	05:23
	← Noldorin left	05:24
	→ Noldorin joined	05:25
	← irqq left	05:25
	← mastro left	05:26
	← Cakey left	05:26
	← ehsan left	05:27
Blaster	hey how could I just temporarily check out a file from a previous commit and then change back to the new one later?	05:27
skorgon	Blaster: git checkout <sha1> -- path/to/file	05:28
darnir	How to ignore all files except those that are currently tracked?	05:29
	→ miso1337 joined	05:29
skorgon	or use 'git show <commit>:path/to/file'	05:29
	← Noldorin left	05:31
Blaster	skorgon: how do I get back to the latest commit version of that file after I do that?	05:31
skorgon	Blaster: read the output of git status	05:31
	← stliu left	05:31
	← vedant left	05:31
Blaster	i did	05:32
	and I run git reset HEAD <file> and its still at the old version	05:32
	→ danishkhan joined	05:32
skorgon	git checkout -- <file>, IIRC	05:32
	← nihilista left	05:33
Blaster	the old version got staged for some reason after checking it out from an older commit	05:33
	→ Noldorin joined	05:33
skorgon	reset HEAD should unstage it	05:34
	← Noldorin left	05:34
Blaster	yeah I had to do that, and then checkout head	05:35
	→ Noldorin joined	05:35
	→ kerrick joined	05:36
	← Noldorin left	05:36
	← gerard0 left	05:36
	→ Noldorin joined	05:36
	← danishkhan left	05:37
	← Noldorin left	05:37
	→ mastro joined	05:38
	→ p3rror joined	05:39
	← marcdel left	05:39
	← xiangfu left	05:40
	← p3rror left	05:40
	→ xiangfu joined	05:41
	→ lresende_ joined	05:41
	→ marcdel joined	05:42
	← yankeefan04 left	05:42
	→ longbeach joined	05:43
	← lresende left	05:43
	lresende_ → lresende	05:43
	← hafeez3000 left	05:44
	← s00pcan left	05:44
	← marcdel left	05:44
	→ Tommy[D]_ joined	05:44
	→ hafeez3000 joined	05:46
	→ snearch joined	05:46
	← snearch left	05:47
	→ snearch joined	05:47
	← Tommy[D] left	05:48
	→ s00pcan joined	05:48
	→ p3rror joined	05:49
	← brundlfly left	05:49
	← JDuke128 left	05:49
	← deepinskin left	05:50
	→ Kulrak joined	05:51
	→ Raging_Hog joined	05:53
	→ system64 joined	05:54
	← lresende left	05:54
	→ Noldorin joined	05:54
	← Noldorin left	05:55
	→ Noldorin joined	05:56
	→ Xethron joined	05:57
	← Noldorin left	05:57
	→ Noldorin joined	05:58
	← Raging_Hog left	05:58
	← netiul left	05:58
	← Noldorin left	05:59
	→ Noldorin joined	05:59
	← Noldorin left	06:00
	→ lethjakman joined	06:00
	← boombatower left	06:03
	← BryanRuiz1_ left	06:05
	→ eijk joined	06:06
	← s00pcan left	06:07
	→ tewecske joined	06:07
	→ s00pcan joined	06:07
	← cwinebrinner left	06:08
	← Error404NotFound left	06:08
	→ Wulf joined	06:08
	← system64 left	06:08
	→ Err404NotFound joined	06:09
	← PaulCapestany left	06:10
	→ etcetera joined	06:11
	→ vdv joined	06:12
	→ borillion joined	06:12
	← harshpb left	06:14
	→ fuzzybunny69y joined	06:15
	← m0viefreak left	06:15
	→ harshpb joined	06:15
	← etcetera left	06:16
borillion	Um I guess Im missing something but how to you grab a directory or a subpart of a tree ?	06:16
	→ etcetera joined	06:16
EugeneKay	borillion - man git-archive	06:20
gitinfo	borillion: the git-archive manpage is available at http://jk.gs/git-archive.html	06:20
	→ kcj joined	06:20
borillion	:P kk	06:20
EugeneKay	If you just want a subdir of it you can look in man gitrevisions for the syntax to use	06:20
gitinfo	the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	06:20
borillion	forgot about that man command lol	06:21
	Thanks EugeneKay, been a while seen I saw you last on here lol	06:21
	← normanrichards left	06:23
	→ mikepack joined	06:24
	← Goplat left	06:25
	← Turicas left	06:26
	← kerrick left	06:26
	← darnir left	06:28
	← Watabou left	06:28
	→ kerrick joined	06:28
EugeneKay	I'm usually around	06:28
	→ lresende joined	06:29
	→ nevyn joined	06:30
	→ subhojit777 joined	06:31
	← averiso left	06:33
	→ averiso joined	06:35
	← Dreamer3 left	06:36
	→ jceb joined	06:36
	← borillion left	06:36
	→ borillion joined	06:37
	→ iksik joined	06:37
	← iksik left	06:37
	→ iksik joined	06:37
	← iksik left	06:37
	← causasui_ left	06:37
	→ Dreamer3 joined	06:38
	→ ThePawnBreak joined	06:41
	← huoxito left	06:41
	← lresende left	06:42
	← OOPMan left	06:44
	→ vedant joined	06:47
	← dydave left	06:48
	← hafeez3000 left	06:49
	← borillion left	06:49
	← jceb left	06:50
	← ThePawnBreak left	06:52
lyang0	I'll probably make most of my 1:1 calls, what does it mean in english ?	06:53
	→ Raging_Hog joined	06:55
	→ sigkill joined	06:58
	← pppaul left	06:58
	→ infynyxx joined	06:59
	→ hmsimha joined	06:59
	→ barneybook joined	06:59
	← unknowndna left	07:00
	→ hafeez3000 joined	07:01
	→ marcdel joined	07:04
	← marcdel left	07:04
	→ lowrezz joined	07:06
	→ ThomasLocke joined	07:06
	→ chuckharmston joined	07:07
	← lowrezz left	07:07
	← infynyxx left	07:08
	→ xpen joined	07:08
	→ ThePawnBreak joined	07:09
	→ teearri joined	07:10
	← teear left	07:10
	← jjore left	07:11
	← mikepack left	07:12
	← etcetera left	07:13
	→ kadoban joined	07:13
	→ wrouesnel joined	07:14
	→ Katafalkas joined	07:15
	← chuckharmston left	07:16
	→ jargon joined	07:16
	→ newbie999 joined	07:16
	→ maqr_ joined	07:18
	← maqr left	07:21
	← isomorph left	07:21
	→ ryez joined	07:22
	→ dsdeiz_ joined	07:23
	← dsdeiz left	07:23
	← scorphus left	07:23
	← Yakko left	07:27
	← wrouesnel left	07:27
	← Katafalkas left	07:28
	→ stliu joined	07:30
	→ tsdh joined	07:30
	→ ddv joined	07:31
	→ aweorit joined	07:33
	← xiangfu left	07:35
	→ xiangfu joined	07:35
	← vedant left	07:37
	→ rob_w joined	07:38
	→ flijten joined	07:40
	→ Anders_J joined	07:40
	→ chuckharmston joined	07:41
	→ rath joined	07:42
	← beautifulmind left	07:42
	→ rurufufuss joined	07:43
	→ tPl0ch joined	07:43
	← vdv left	07:43
	→ dsdeiz joined	07:44
	→ vedant joined	07:44
	→ k_sze[work] joined	07:44
k_sze[work]	grrrrr	07:45
Nevik	sorry	07:45
	dogs have to stay outside	07:45
k_sze[work]	why did git stash pop automatically stage the files in the stash?	07:45
Nevik	because stash also saves your staging area	07:45
	it stages what was staged when you stashed	07:45
	← fuzzybunny69y left	07:46
	← dsdeiz_ left	07:46
k_sze[work]	Nevik: but I am certain that I did not stage them before stashing.	07:46
	I even still have the scrollback log of my terminal to prove it.	07:46
Nevik	you can probably tell stash to not stage somehow	07:46
	man git stash will know	07:46
gitinfo	the git-stash manpage is available at http://jk.gs/git-stash.html	07:46
	erkules_ → erkules	07:49
	→ marthinal joined	07:51
	← chuckharmston left	07:51
	← dreiss left	07:53
	→ pen joined	07:54
	→ Sigma joined	07:55
	← harshpb left	07:56
	→ giallu joined	07:56
	← guampa left	07:57
	→ tbosua joined	07:57
	→ harshpb joined	07:57
	→ pencilcheck joined	07:58
	← vedant left	08:00
	→ vedant joined	08:01
	← pen left	08:02
	→ suy joined	08:03
	→ elixir joined	08:03
	← elixir left	08:04
	→ kewah joined	08:05
	→ gain joined	08:05
	← relipse left	08:06
	← vedant left	08:07
	→ douglaslassance joined	08:07
	→ leeN joined	08:08
	← leeN left	08:08
	→ vedant joined	08:08
	→ LeMike joined	08:08
	← thehandler left	08:09
	← vedant left	08:13
	← hoschi left	08:16
	→ chuckharmston joined	08:17
	→ francisfish joined	08:18
	→ netiul joined	08:19
	← thiago left	08:19
	→ mklappstuhl joined	08:19
	→ Sonderblade joined	08:21
	→ relipse joined	08:22
	→ psoo joined	08:23
	→ mihamina joined	08:23
	← dhruvasagar left	08:24
	← chuckharmston left	08:25
	→ danishkhan joined	08:26
	→ dydave joined	08:27
	← eletuchy left	08:27
	→ eletuchy_ joined	08:27
	← hafeez3000 left	08:28
	→ AccessGranted joined	08:28
	→ timj joined	08:29
	→ hoschi joined	08:30
	← danishkhan left	08:30
	→ smirn0v joined	08:30
	→ ak5 joined	08:31
	← timj1 left	08:32
	← eletuchy_ left	08:32
	← LekeFly left	08:32
	← lethjakman left	08:33
	→ OOPMan joined	08:34
	→ lowrezz_ joined	08:35
lowrezz_	hi	08:35
	how to delete a revision in a remote repo	08:35
	← _Vi left	08:35
	← madewokherd left	08:37
	→ thierryp joined	08:37
Seveas	lowrezz_: you don't. you could force a push to the relevant ref of a local ref where that commit is not a parent of and then gc will take care of it after a while	08:38
lowrezz_	ok	08:38
	and how can i change an old commit message	08:38
Seveas	if it's at the tip of a branch: git commit --amend. If not, git rebase --interactive	08:39
	→ irqq joined	08:39
	→ elixir joined	08:40
	→ TheFuzzball joined	08:40
	← xpen left	08:41
	→ didlixForHire joined	08:41
	Tao → Conjuro	08:41
	→ maxmanders joined	08:43
	→ zoran119_ joined	08:45
	← zoran119 left	08:46
	→ BadDesign joined	08:46
	← BadDesign left	08:46
	→ BadDesign joined	08:46
	→ marcdel joined	08:47
	→ dhruvasagar joined	08:49
	→ arkady91 joined	08:51
	← averiso left	08:51
	← ferdev left	08:52
	→ b1rkh0ff joined	08:52
	→ chuckharmston joined	08:52
	→ marhaban joined	08:53
	← didlixForHire left	08:53
	← Xethron left	08:54
	→ Xethron joined	08:55
	→ alexwhitman joined	08:56
	→ LimeBlast joined	08:57
	← arkady91 left	08:58
	→ koltroll joined	08:58
	← fp_ left	08:59
	← spyro left	09:00
	← douglaslassance left	09:00
	← Jahm left	09:00
	← harshpb left	09:00
	← Guest54594 left	09:00
	→ chichou_ joined	09:00
	→ faceleg joined	09:00
	← pencilcheck left	09:01
	→ harshpb joined	09:01
	← chichou_ left	09:01
	← chuckharmston left	09:01
	← faceleg left	09:02
	→ spyro joined	09:02
	← harshpb left	09:02
	← marcdel left	09:02
	→ harshpb joined	09:02
	← elixir left	09:03
	→ elixir joined	09:03
	→ daniel_-_ joined	09:03
	← threesome left	09:04
	→ threesome joined	09:05
	→ Error404NotFound joined	09:05
	← Error404NotFound left	09:06
	← Err404NotFound left	09:07
	→ douglaslassance joined	09:07
	→ infynyxx joined	09:08
	→ LekeFly joined	09:09
	← infynyxx left	09:11
	← Vile left	09:12
	→ ferdev joined	09:12
	→ shadok_ joined	09:14
	← shadok_ left	09:14
	→ shadok_ joined	09:14
	→ infynyxx joined	09:15
	← zoran119_ left	09:15
	→ zoran119 joined	09:15
	← ThePawnBreak left	09:18
	← pretty_function left	09:19
	→ exoon joined	09:20
	→ hafeez3000 joined	09:21
	← BadDesign left	09:22
	← zoran119 left	09:23
	→ sie joined	09:23
	→ zoran119 joined	09:24
	← marhaban left	09:26
charon	lowrezz_: read about !rewriting though	09:27
gitinfo	lowrezz_: Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to jump through hoops (and you have to tell them to), so it's infinitely better to just move on without rewriting. That said, you can use `git push -f` to force your new history through (may get rejected, though). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstrea	09:27
lowrezz_	thanks	09:28
	i didn't change it anyway	09:28
	→ chuckharmston joined	09:28
lowrezz_	slowly i begin to like git	09:28
	ppl on this channel helped me a lot especially cbreak	09:29
charon	lowrezz_: removing a commit from a public repo (other than by simply deleting the only branch it's on) is also rewriting	09:29
	→ zoresvit joined	09:29
	→ tvw joined	09:29
charon	(i don't know your reasons, so i'm not _yet_ saying you shouldn't ... but you better have good reasons ;) )	09:30
	← daniel_-_ left	09:30
	← Asenar left	09:30
	← yacks left	09:31
	← sie left	09:32
	→ alkino joined	09:33
	← j0ran left	09:34
	→ j0ran joined	09:34
	← j0ran left	09:35
	→ j0ran joined	09:35
	← rurufufuss left	09:35
	← zoresvit left	09:36
	← chuckharmston left	09:37
	→ fire joined	09:38
	← mihamina left	09:38
	→ marhaban joined	09:38
	→ yacks joined	09:39
	← infynyxx left	09:39
	← dpwrussell left	09:40
	→ Vile joined	09:41
	→ dpwrussell joined	09:41
	→ Asenar joined	09:42
	← kcj left	09:42
	→ pantsman joined	09:44
	← marhaban left	09:44
	→ jkremser joined	09:46
	→ mikecmpbll joined	09:46
	→ eletuchy joined	09:47
	→ pen joined	09:47
	← eletuchy left	09:47
	→ eletuchy joined	09:47
	→ axrfnu joined	09:47
	→ rurufufuss joined	09:48
	→ marhaban joined	09:50
	→ cilly joined	09:51
	→ zoresvit joined	09:51
	→ pen_ joined	09:52
	← josefig left	09:52
	← pen left	09:53
	→ daniel_-_ joined	09:53
	← stliu left	09:53
	→ ozette joined	09:54
	→ sveinse joined	09:54
	← miso1337 left	09:55
sveinse	I'm a git n00b, and I accidentally commited wrongfully and pushed into gitorious. How can I undo that? I.e. I'd like to delete the commit, not just recommit	09:56
dr_lepper	sveinse: use git reset and then git push -f	09:57
	dwmw2_gone → dwmw2	09:57
	→ obiwahn joined	09:57
obiwahn	hi	09:58
sveinse	My local repo wasnt updated, so my commit and push created a branch. When I pulled it got merged and when I pushed it the branch&merge appeared in remote	09:58
obiwahn	how do i rebase a single commit onto my curretn branch?	09:58
	current	09:58
	← JeremyWei left	09:59
	→ xpen joined	09:59
sveinse	dr_lepper: I have run git rebase so the local repo looks good. But when I run git push -f origin +master I get error: src refspec master does match any	10:00
	What is master in this setting?	10:00
obiwahn	should i checkout the branch and git rebase -i --onto target?	10:01
	← hyperair left	10:01
dr_lepper	sveinse: do you have local branch master?	10:01
	→ alorence joined	10:02
	← m0ltar left	10:03
	→ didlixForHire joined	10:04
	← shadok_ left	10:04
	→ chuckharmston joined	10:04
	→ m0ltar joined	10:05
	← zoresvit left	10:05
	← chichou left	10:05
	← didlixForHire left	10:09
sveinse	Is there a caching involved on gitorious? I've pushed my repo (after rebasing), but I'm not cloning the correct data	10:10
	→ bil` joined	10:10
cbreak	should not happen	10:10
	how did you push? did it succeed?	10:10
	→ FernandoBasso joined	10:11
sveinse	Yes, it sais so	10:11
	← bil` left	10:11
charon	sveinse: pastebin it, including the command line you used	10:11
	there are a few opportunities for mistakes that might not be obvious	10:11
	→ marlun_ joined	10:12
	← Raging_Hog left	10:13
	← chuckharmston left	10:13
marlun_	To view the message I've created with `git branch --edit-description` am I supposed to edit it again or can I view it in some other way?	10:13
charon	marlun_: git config branch.foo.description	10:15
	i suppose you could make an alias	10:15
sveinse	Is this channel logged?	10:16
	→ chichou joined	10:17
	→ PeterB joined	10:18
charon	yes	10:19
PeterB	Morning, I have "*.yml text eol=lf" in my .gitattributes file - this should convert eol to lf always?	10:19
	Have endless EOL issues at the moment, and files are getting checked in (somehow) with CRLF	10:19
	→ Chikore joined	10:20
	← pen_ left	10:21
marlun_	charon: How would I get the name of the current branch into that alias?	10:21
charon	PeterB: perhaps by a client that doesn't respect the attribute; eol=lf is only supported starting with 1.7.2 afaics	10:21
	→ zoresvit joined	10:22
sveinse	Ok, here it is: http://paste.ubuntu.com/5616090/ obfuscating the names for discression	10:22
	→ sbodo joined	10:22
	→ twb joined	10:23
twb	emacs just trashed some segments of my .emacs, and I also had uncommitted changes	10:23
sveinse	If I clone the repo in new location, these changes are not present	10:23
twb	Is there something like git add -p, but to reset changes back to HEAD ?	10:23
charon	twb: git checkout -p HEAD ?	10:23
	also git reset -p	10:23
	depending on whether you want to overwrite the worktree	10:24
	marlun_: git config alias.desc '!f () { git config branch."$1".description; }; f "$@"'	10:24
	← jdyin left	10:24
twb	Hm, it's sitting there thinking an awful long time	10:25
charon	sveinse: what's HEAD in the remote repo? what does 'git branch -a' say after cloning? i suspect you simply need to check out acme-trunk, e.g. using git checkout -t origin/acme-trunk	10:25
	→ Yougi joined	10:25
sveinse	charon: It sais * acme-trunk. Running git checkout -t origin/acme-trunk returns fatal: A branch named 'acme-trunk' already exists	10:27
	→ beautifulmind joined	10:27
marlun_	charon: thanks but that seem to require that I send in the branch name, would want it to figure that out itself?	10:27
PeterB	charon: that seems ok	10:28
twb	Ah OK, works now	10:28
	I think I just can't spell	10:28
	marlun_ → marlun	10:28
charon	marlun_: well, script around git-symbolic-ref to determine what HEAD points to	10:28
sveinse	the log on gitorious also does not include my rebase, so perhaps gitorious silently rejected my forced push	10:28
	← marlun left	10:28
	→ marlun joined	10:28
marlun	charon: thanks, I'll look into it!	10:29
cbreak	sveinse: you changed laerdal-trunk but pushed acme	10:29
	→ kmels__ joined	10:29
charon	cbreak: oh, nice catch.	10:29
sveinse	Yeah ;) that slipped through :) obfuscation you know ;)	10:29
	→ marhaban1 joined	10:30
cbreak	sveinse: and if you clone, and switch to the right branch	10:30
	you don't get 159494f?	10:30
charon	actually	10:31
	if you look at the commit, and at the hash in the log, they're the same	10:31
	so what's the problem?	10:31
	← marhaban left	10:31
	→ pen joined	10:32
charon	sveinse: that being said, i fail to see how the first log adds up with git-reset's output	10:32
sveinse	look at the first log. commit 06a56185 contains unfortunate log entry. It contains info which should not be present on gitorious, so I need to delete it	10:32
	(It does not show in my pastebin, but there are some private info there)	10:33
charon	!secret	10:33
gitinfo	[!topsecret] If your situation is so secret that you can't tell us how to !reproduce it, there's really not much we can do. Perhaps you need to find or buy support you can trust? There are people with clearances even on #git but since clearance levels are classified and #git is not a secure channel, this may not help you. However, you are protected by the biggest secret of all: We just don't care.	10:33
charon	sigh	10:33
	charon stops caring	10:33
cbreak	sveinse: so you git reset --hard HEAD~1?	10:33
alkino	!reproduce	10:33
gitinfo	[!repro] Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session. This will help immensely with troubleshooting.	10:33
cbreak	or HEAD^2?	10:33
sveinse	First this is no secret. It's just this commit message which is secret. So I'd like to remove it before showing url. The code is linux kernel, so its public per def	10:35
	→ marhaban joined	10:35
sveinse	I fiddled around with some git rebase and git reset's without too much knowledge to why and what. At some point I ended up with the latter log	10:36
	← zoresvit left	10:36
	← marhaban1 left	10:37
charon	sveinse: what i'm getting annoyed at is that it doesn't add up in several places. we already established that you censored some branchnames, but not all. now i notice that line 2, 27 and 28 cannot possibly have happened like this.	10:37
	how are we supposed to help with the actual problem if you create more problems in trying to obfuscate?	10:37
	← k_sze[work] left	10:37
charon	in any case, what was pushed (159494f in line 44) matches what is shown in the log in line 50	10:37
	so i'm not sure there's any real problem at all	10:37
cbreak	yeah, as I asked above	10:38
	"you don't get 159494f?"	10:38
sveinse	OK, let me try to redo my own steps, to see if I can recreate. obfuscate in this context is just use name "acme" instead of our real name	10:38
charon	sveinse: anyway, don't feel too bad, i'm mostly ranting because it goes like this all the time. people try to obfuscate and/or snip and cause lots more problems, wasting everyone's time.	10:39
	of course the top-scorer in that area is still the guy who snipped out data in a length-prefixed format (git fast-export dumps)	10:40
cbreak	well, it looks like the problem is solved anyway	10:40
	→ chuckharmston joined	10:40
sveinse	And I did mess with a lot of git rebase, git reset stuff, so the point is valid	10:40
cbreak	charon: smart...	10:40
jaequery	how can i just pull from a branch and completely replace my local ?	10:40
	i keep getting Conflicts	10:41
	→ engkur_ joined	10:41
	→ pencilcheck joined	10:41
jaequery	i just wanna get whatever is on the remote branch and overwrite everythig on my local branch	10:41
	how to dO?	10:41
sveinse	Well my issue is that my local repo sais acme-trunk head is 15949... while a freshly cloned repo shows the old layout with 06a561 at the end	10:41
cbreak	jaequery: then you do not want pull	10:41
	pull is fetch + merge	10:41
	→ sdeboey joined	10:41
jaequery	right	10:41
cbreak	you sound like you want fetch + reset --hard	10:41
	you probably already fetched, so just git reset --hard @{u}	10:42
bremner	but see also !backup	10:42
gitinfo	Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	10:42
cbreak	bremner: he asked for what ever he'll get :D	10:42
jaequery	not work	10:42
	fatal: ambiguous argument '@{u}': unknown revision or path not in the working tree.	10:42
cbreak	sveinse: run !lol on both	10:42
gitinfo	sveinse: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	10:42
cbreak	jaequery: sounds like you have no upstream. explicitly name the branch you want to reset to	10:43
	jaequery: probably one of git branch -r	10:43
	← engkur left	10:43
jaequery	how do i specify it?	10:43
	→ apol joined	10:43
	← OOPMan left	10:43
cbreak	git reset --hard BLAHHERE	10:43
	it's a command line argument	10:43
	→ Raging_Hog joined	10:44
	← elixir left	10:44
cbreak	you might decide to use something like origin/master	10:44
	← Archdave left	10:44
cbreak	but refer to the output of branch -r	10:44
	← pen left	10:44
jaequery	@{cool	10:44
	so what is @{u}	10:45
	default upstream branch?	10:45
	→ Archdave joined	10:45
cbreak	no	10:46
	→ wrouesnel joined	10:46
cbreak	the upstream branch that you set	10:46
	sounds like you have none set	10:46
	(it is set automatically sometimes, or when you git push -u, or set it with git branch --set-upstream or similar)	10:46
jaequery	yea ok i just did origin/bla and it worked	10:46
	thx	10:46
sveinse	cbreak: http://paste.ubuntu.com/5616154/	10:47
charon	cbreak: i made !obfuscate	10:47
gitinfo	cbreak: Please do not elide, obfuscate, or otherwise hide information from !repro pastes. This invariably leads to pastes that hide the actual issue or (worse) are not consistent, sending the investigation down false paths. See !secret.	10:47
cbreak	sveinse: compare git remote -vv	10:48
	charon: interesting :)	10:48
	I am not sure if I am against obfuscation, but people who do it have to live with being called liars if they screw it up :)	10:49
charon	i've had enough, after three screwups in two days	10:49
cbreak	:/	10:49
	← chuckharmston left	10:49
	→ aqabiz joined	10:50
charon	seriously, this is unpaid git support	10:50
	→ zoresvit joined	10:51
charon	if your my-stuff-is-oh-so-secret attitude makes our lives hard, go find some paid support with an NDA	10:51
	</rant>	10:51
sveinse	cbreak: Haha. My "freshly cloned repo" was pulling from our mirror, not gitorious. So I'm very embarrassed and it seems to be correct now	10:51
cbreak	paid git support. Now that sounds like no fun.	10:51
	→ itatitat joined	10:51
cbreak	sveinse: you didn't notice that when entering the URL for cloning? :/	10:52
	← beautifulmind left	10:52
	← mastro left	10:52
	→ uman joined	10:52
cbreak	but good that it works now :)	10:52
sveinse	cbreak: That is exactly my embarrasement....	10:52
uman	Is there a way to add a set of files to a list to prevent them from showing up as untracked in git-status	10:52
cbreak	yes	10:52
	uman: man gitignore	10:53
gitinfo	uman: the gitignore manpage is available at http://jk.gs/gitignore.html	10:53
uman	Even better if I can use wildcards. The dozens of .pyc get annoying	10:53
	cbreak: thanks	10:53
cbreak	do something like echo "*.pyc" >> .gitignore	10:53
	→ elixir joined	10:53
cbreak	and they should vanish	10:53
	← maxmanders left	10:53
sveinse	So to disclose everyting, since I'm almost being offended by being called secretive: This is the repo: https://gitorious.org/lm-linux	10:53
PeterB	Can anyone see why this .gitattributes file would let some yml and php files in with CRLF? All in deep directory trees (a limit to depth?)	10:53
sveinse	And thank you very much all of you	10:53
cbreak	np :)	10:53
	← jargon left	10:53
	cbreak is more secretive than most	10:54
	← chichou left	10:54
	→ jargon joined	10:55
	→ maxmanders joined	10:55
	← jaequery left	10:55
	→ chichou joined	10:56
	→ RaceCondition joined	10:56
	→ mulinux joined	10:57
charon	PeterB: !repro if nobody has any ideas	10:57
gitinfo	PeterB: Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session. This will help immensely with troubleshooting.	10:57
	← zoresvit left	10:58
charon	i.e. show commands showing the .gitattributes and how you determine that things go wrong	10:58
	→ nihilista joined	11:00
	→ fisted_ joined	11:00
	← RaceCondition left	11:01
	→ jn_ joined	11:01
	← timj left	11:01
	← sdeboey left	11:02
	← Sigma left	11:04
	← fisted left	11:04
	fisted_ → fisted	11:04
charon	PeterB: a quick test also shows that it's not that hard to manually get crlf files into the repo even if their attributes say that they should be lf	11:04
	https://gist.github.com/trast/8430c2806fc4295cfb7d	11:05
	← luc_ left	11:06
charon	hrm, made a mistake in resetting the attributes	11:06
	→ mastro joined	11:06
	← lowrezz_ left	11:06
	→ kcj joined	11:07
charon	anyway, afaics it does show a few warnings but probably not enough	11:07
	after that you can "live" with the crlf file without noticing	11:07
	→ gusnan joined	11:08
PeterB	charon: http://pastebin.com/dw6tW4qA	11:08
	→ sonda joined	11:08
PeterB	I don't know how to show which files have CRLF, so I've added a comment that they're there	11:08
	→ beautifulmind joined	11:09
PeterB	I did this line ending normalisation 3 weeks ago when I imported from SVN=>Git, but some are still there	11:09
	This is now causing massive problems when merging new branches (all created off the same Git master), due to line ending conflicts	11:10
	← yshh left	11:13
	← maxmanders left	11:13
cbreak	find out where the line endings changed	11:14
Bombe	And kill that person.	11:14
	← elixir left	11:14
	→ elixir joined	11:14
	→ maxmanders joined	11:14
	← twb left	11:14
PeterB	Bombe: the person will be me	11:14
	and the problem probably working on Windows over SAMBA and committing on Linux	11:15
	cbreak: how do I find out where they changed?	11:15
	← j0ran left	11:15
	← FernandoBasso left	11:15
	← aweorit left	11:15
	→ j0ran1 joined	11:15
cbreak	git log -- filename	11:15
	→ zoresvit joined	11:15
cbreak	if there's a commit where all lines changed, then that's it	11:16
	→ chuckharmston joined	11:16
	← jargon left	11:17
PeterB	the commits they changed in were old SVN ones	11:18
	← kcj left	11:18
PeterB	I followed the instructions in "Re-normalizing a repo" on https://help.github.com/articles/dealing-with-line-endings when I first imported	11:19
	and I've tried that again now	11:19
	but it doesn't normalize the line endings	11:19
	← elixir left	11:19
	→ pastime joined	11:20
	← harshpb left	11:20
	← daniel_-_ left	11:21
	→ harshpb joined	11:21
	← eletuchy left	11:21
	→ eletuchy joined	11:21
	← harshpb left	11:22
	← dhruvasagar left	11:22
	→ kristofers joined	11:22
	← xpen left	11:22
	→ harshpb joined	11:22
	→ barq joined	11:22
	gitinfo set mode: +v	11:23
	→ dhruvasagar joined	11:24
	→ gommo joined	11:25
	← chuckharmston left	11:25
	→ fuzzybunny69y joined	11:25
	→ timj joined	11:26
	← s0undt3ch left	11:26
	→ Terminus- joined	11:29
	→ s0undt3ch joined	11:29
Terminus-	hello. i'm trying to get webdav running on httpd. what could possibly cause this when doing an http-push? http://dpaste.org/vUKLF/	11:29
	→ RaceCondition joined	11:30
Terminus-	it looks like it's failing on heads/master which doesn't exist yet coz the http-push i'm attempting to do is supposed to populate the bare repo.	11:30
	← itatitat left	11:31
	← sgronblom left	11:32
cbreak	PeterB: the way I'd normalize line endings is by simply converting them and committing them.	11:34
PeterB	cbreak: thanks	11:34
	once I've merged in these broken branches (by hand, copying changes over) I'll do that	11:34
grawity	Terminus-: i'm not sure one can push to a repository that doesn't exist yet	11:34
	→ frenchlaf joined	11:34
	← dhruvasagar left	11:34
grawity	Terminus-: nevermind me	11:35
cbreak	PeterB: sounds braindead	11:35
	why not fix line endings now?	11:35
	← mulinux left	11:35
PeterB	cbreak: they are wrong in one branch but not the other	11:35
Terminus-	grawity: it's a repo created with git init --bare.	11:35
PeterB	and some files in both	11:35
cbreak	yeah, fix them in all branches...	11:35
PeterB	so I have to fix both - rebase won't complete without large errors due to EOL	11:36
cbreak	PeterB: or use a mergetool that can ignore line endings	11:36
	← alorence left	11:36
PeterB	cbreak: ah hmm, yes	11:36
	← zoresvit left	11:36
	→ zoresvit joined	11:36
PeterB	yet to use a mergetool, have been hand correcting merges	11:36
cbreak	PeterB: man git-merge	11:36
gitinfo	PeterB: the git-merge manpage is available at http://jk.gs/git-merge.html	11:36
cbreak	look at the options for renormalize	11:37
	→ elixir joined	11:37
cbreak	git can do that on its own	11:37
Schmoozerd	why not rewrite the history to only have correct line-endings? (i think fast-export, convert, fast-import might already work)	11:38
cbreak	filter-branch would probably be cleaner.	11:38
PeterB	was git written with cross-platform development in mind? I understand it wants me to be responsible with line endings, but it makes it complex to work with	11:39
	← Parhs8 left	11:39
	→ Parhs joined	11:40
	← Parhs left	11:40
	→ Parhs joined	11:40
grawity	I think it was written with Linux kernel development in mind	11:40
cbreak	PeterB: line endings are your responsibility	11:41
	unless you turn on line ending conversion	11:41
PeterB	cbreak: which is what gitattributes should do?	11:42
Terminus-	PeterB: what grawity said. it started when bitkeeper decided to withdraw licenses from kernel development.	11:42
cbreak	PeterB: and it does	11:42
	but it won't fix past mistakes	11:42
PeterB	Schmoozerd: cbreak: rewriting history 0- hadn't realised gitattributes and a new checkout/commit wouldn't make future commits have the correct line endings	11:42
	← sveinse left	11:43
	← engkur_ left	11:43
cbreak	they should	11:43
	→ engkur joined	11:43
Schmoozerd	they will (if set up correctly) -- but if you attempt to merge something, you get old stuff which causes the problems	11:44
	← hafeez3000 left	11:44
	→ geekbri joined	11:45
PeterB	Schmoozerd: the branches I'm merging have all been created since adding the .gitattributes file and supposedly normalising	11:45
	→ shalok joined	11:46
	← mikecmpbll left	11:46
	← netiul left	11:46
	← Chikore left	11:47
	→ mikecmpbll joined	11:48
	← aqabiz left	11:48
totte	I've got a .gitmodules file from one repository and would like to use it in another repository - can I somehow add it in one go or do I have to run "git submodule add foo" for each submodule all over again?	11:50
	→ wasanzy joined	11:50
	→ netiul joined	11:50
	← harshpb left	11:51
	← timj left	11:51
totte	Simply copying the file doesn't seem to work. I ran "git submodule update --init --recursive" which had no effect, seemingly.	11:51
	→ harshpb joined	11:51
	← tbosua left	11:52
	→ chuckharmston joined	11:52
	← dpwrussell left	11:52
	→ dhruvasagar joined	11:52
	→ dpwrussell joined	11:53
	→ Cromulent joined	11:53
	→ SonikBoom joined	11:54
	← nkryptic left	11:56
	→ milStaja joined	11:57
	→ _goTAN joined	11:58
	→ hafeez3000 joined	12:01
	← chuckharmston left	12:02
	← Cromulent left	12:03
	← engkur left	12:03
	→ peff joined	12:04
	← rchavik left	12:05
	→ tbosua joined	12:07
	← apol left	12:08
	← RaceCondition left	12:09
	→ jargon joined	12:10
	← harshpb left	12:10
	→ harshpb joined	12:11
	→ timj joined	12:12
	← shalok left	12:12
	→ antix_ joined	12:12
	← giallu left	12:15
	← elixir left	12:15
	→ pandeiro joined	12:17
	→ mmitchell joined	12:17
	→ mulinux joined	12:19
	← antix_ left	12:19
	← dydave left	12:19
	→ RaceCondition joined	12:20
	SeoZ → SeoZ-work[AWAY]	12:20
	← RaceCondition left	12:20
	← gommo left	12:20
	→ antix_ joined	12:21
cbreak	totte: you have to add the submodules	12:21
	totte: the .gitmodules file is not relevant for submodules you have, it's only used to find URLs in the init step	12:21
	← kadoban left	12:21
	← pencilcheck left	12:22
	→ pen joined	12:23
	← hays left	12:23
	← f8l left	12:24
	← harshpb left	12:24
	→ RaceCondition joined	12:24
	← RaceCondition left	12:26
	← thyseus left	12:26
	→ f8l joined	12:26
	→ _Vi joined	12:27
	→ thyseus joined	12:28
	→ chuckharmston joined	12:28
	← eletuchy left	12:30
	→ eletuchy joined	12:30
	← antix_ left	12:30
	→ antix_ joined	12:32
	→ hays joined	12:32
	→ theothe__ joined	12:32
	→ theoth___ joined	12:32
	← theother_ left	12:33
	← theotherguy left	12:34
	→ erichynds joined	12:35
	← p3rror left	12:35
	← maxmanders left	12:35
	→ maxmanders joined	12:36
	← chuckharmston left	12:38
	→ chichouw joined	12:38
	← chichou left	12:39
	← darklajid left	12:39
	← peff left	12:41
	→ Mannequin joined	12:41
	← antix_ left	12:41
	← hays left	12:41
	→ chichou joined	12:42
	← pastime left	12:42
	→ antix_ joined	12:42
	← chichouw left	12:43
	→ engkur joined	12:44
	→ baba joined	12:44
	→ RaceCondition joined	12:46
	← fire left	12:46
	← frenchlaf left	12:46
	→ chichouw joined	12:47
	← marhaban left	12:47
totte	cbreak: Alright, thanks.	12:47
	← chichou left	12:48
	→ sgambino joined	12:48
	→ Sigma joined	12:50
	→ jpcamara joined	12:51
	← antix_ left	12:52
	→ chichou joined	12:53
	→ antix_ joined	12:53
	← chichouw left	12:53
	→ gerard0 joined	12:53
	→ vv0rm joined	12:54
	← ddelrio1986 left	12:55
vv0rm	Hello, I would like to gitignore a modified file and delete it from the remote without removing it from the local copy, is that possible ?	12:55
	→ xpen joined	12:55
	→ jcromartie joined	12:56
Bombe	vv0rm, no. Everything you commit to the repository will happen to other copies of the repository. You git-rm it locally, it will be deleted in other working trees, too.	12:57
	→ chichouw joined	12:57
Nevik	he's already gone	12:57
	gotta love that	12:57
	← engkur left	12:57
vv0rm	Why ?	12:57
	I'm here :)	12:57
Nevik	oh stupid font	12:58
	also dont start your name with vv :P	12:58
	← chichou left	12:58
Nevik	vv0rm: `git rm --cached <that file>`; `git commit -m "your message"`; `git push`	12:58
	that will remove it from the repo but leave it in your working tree	12:59
	if it's added to gitignore already, it will then be ignored from further adding	12:59
	→ nkr joined	12:59
	← jpcamara left	12:59
	→ freeed joined	12:59
	→ jpcamara joined	13:00
Bombe	Oh, local copy only. I should read more carefully.	13:00
Nevik	indeed	13:00
	← yacks left	13:00
	→ PaulCapestany joined	13:01
	← antix_ left	13:03
	→ chuckharmston joined	13:04
	→ _iron joined	13:05
	← mastro left	13:05
	→ Thra11 joined	13:05
	← chuckharmston left	13:06
	→ chuckhar_ joined	13:06
vv0rm	Nevik: perfect, thank you	13:06
	vv0rm → worm	13:07
	worm → worm_	13:07
worm_	just for you Nevik :P	13:07
Nevik	\o/	13:08
	worm_: the fact of the matter is that in my font, vv and w are exactly one pixel different	13:08
jast	so your font sucks	13:08
Nevik	yes	13:08
	i said so already	13:08
	← j0ran1 left	13:08
worm_	yeah it confused other people sometimes too	13:08
Nevik	but honestly, cba to change it	13:08
jast	I hear it's possible to switch fonts	13:08
Nevik	at least not right now	13:09
jast	and it takes less time than what I had to do the past two evenings to migrate my windows setup to a new disk	13:09
Nevik	maebee laeter :P	13:09
	→ ddelrio1986 joined	13:09
	→ mastro joined	13:09
	← maxmanders left	13:09
	← chuckhar_ left	13:10
jast	fixing bootloaders by hex-editing boot sectors is tons of fun	13:10
	→ angelsl joined	13:11
	← ddelrio1986 left	13:11
	→ ddelrio1986 joined	13:12
	→ maxmanders joined	13:12
cbreak	can't you just clone the disk?	13:14
	← marlun left	13:15
	→ antix_ joined	13:16
	← CodeShark left	13:16
	→ scelis joined	13:16
	→ trungqtran joined	13:17
	→ cdeszaq joined	13:18
	← exoon left	13:18
	→ pdurbin joined	13:19
	→ thatguycraig joined	13:20
	→ tbmiller joined	13:20
	→ chuckharmston joined	13:20
pdurbin	osse: more force pushing feature branches. having the other guy reset hard to the rebased branch you pushed: http://irclog.greptilian.com/sourcefu/2013-03-15	13:21
	milki: you might be interested in this too ^^	13:21
	→ R1KRDO joined	13:21
	→ breakingthings joined	13:21
	← antix_ left	13:22
	← chichouw left	13:23
	→ DoniRS joined	13:23
	→ engkur joined	13:23
	→ miso1337 joined	13:23
	← brandonw left	13:25
	→ antix_ joined	13:26
	→ bitkiller joined	13:27
	→ GenTAR joined	13:27
	→ chichou joined	13:27
	→ mavcunha joined	13:28
	→ AaronMT joined	13:30
	→ brandonw joined	13:30
	→ tureba joined	13:31
	← subhojit777 left	13:32
	→ normanrichards joined	13:33
	← chichou left	13:33
	← tureba_ left	13:33
	← hafeez3000 left	13:35
	→ slaren joined	13:36
	→ yacks joined	13:36
	→ JaredMiller joined	13:36
	→ ojacobson joined	13:37
	← brandonw left	13:38
	→ chichou joined	13:38
	→ maxloe joined	13:38
	← antix_ left	13:38
	← JaredMiller left	13:38
	→ ovidiu-florin joined	13:39
	← normanrichards left	13:39
	→ marhaban joined	13:40
	← tureba left	13:40
	← engkur left	13:41
	→ TeckniX joined	13:41
	→ engkur joined	13:41
	→ brandonw joined	13:42
	← mavcunha left	13:42
	→ antix_ joined	13:42
	→ mavcunha joined	13:42
	→ fisted_ joined	13:42
	← tvw left	13:42
	← fisted left	13:42
	fisted_ → fisted	13:43
	← chichou left	13:43
	← uman left	13:43
	→ frenchlaf joined	13:43
	→ hafeez3000 joined	13:43
	← bitkiller left	13:44
	→ hyperair joined	13:44
	← frenchlaf left	13:45
	→ bitkiller joined	13:45
	→ v0n joined	13:45
	→ rbmk joined	13:45
	→ thl0 joined	13:45
	→ rakl joined	13:46
	← jn_ left	13:47
	→ stliu joined	13:47
	← mikecmpbll left	13:48
	→ oste joined	13:48
	→ mikecmpbll joined	13:48
	→ worstadmin joined	13:49
	← fatninja left	13:49
	→ jn_ joined	13:49
	← rob_w left	13:50
	→ guampa joined	13:50
	→ crocket joined	13:51
	→ chichou joined	13:51
crocket	Is it possible to fetch remote branches during clone?	13:51
canton7	it should be default	13:52
grawity	clone always does that	13:52
jast	you mean those in refs/remotes/ within the remote repo?	13:52
	or just all the normal branches in the remote repo?	13:52
	→ fatninja joined	13:52
	← Fandekasp left	13:52
	← antix_ left	13:54
	← pdurbin left	13:54
	→ Fandekasp joined	13:54
	← grn left	13:56
	→ sdeboey joined	13:56
	← miso1337 left	13:57
	→ chichouw joined	13:57
	→ miso1337 joined	13:57
	← chichou left	13:58
	→ OOPMan joined	13:59
	← miso1337 left	13:59
	→ antix_ joined	14:01
	→ dc5ala joined	14:01
	→ miso1337 joined	14:01
	→ Gentlecat_ joined	14:02
	← Gentlecat left	14:02
	→ jamesez joined	14:02
	→ chichou joined	14:02
	→ JPeterson joined	14:02
	← chichouw left	14:03
	→ les_sylvains joined	14:03
	← Gentlecat_ left	14:03
	← chichou left	14:03
	→ grn joined	14:04
	← PeterB left	14:04
	→ Gentlecat_ joined	14:04
	→ compojoom joined	14:04
	← PaulCapestany left	14:04
	→ chilversc_ joined	14:04
	← miso1337 left	14:04
compojoom	I trying to merge my master with another branch. When I do it - I get some conflicts and when I try to resolve them git complains of "fatal unable to stat path_name... no such file or directory"	14:05
	← kristofers left	14:05
compojoom	any ideas what to do?	14:05
	→ kristofers joined	14:05
	→ Jrobbins joined	14:05
chilversc_	is there any good model for handling a scenario where you've developed several new features that are all on a next branch, but then you need to release just one of those features earlier than expected?	14:06
	my only thought at the moment is delete everything that isn't quite ready yet, branch from there for the release, then revert the delete on the next branch?	14:07
	← maxloe left	14:07
	→ chichou joined	14:07
	← engkur left	14:08
	← pen left	14:08
	→ natechan joined	14:09
	→ maxloe joined	14:10
	gitinfo set mode: +v	14:10
Jrobbins	Anyone know how to denydeletes for specific branches of a central repository? My company is useing 2 main branches for pushing to production and I want to prevent them from being deleted while allowing developers to delete their own branches if they choose	14:11
	→ Grey89 joined	14:11
Jrobbins	I have only been only successful in preventing deletion of all branches	14:11
	← antix_ left	14:11
	→ engkur joined	14:11
	← Grey89 left	14:11
	← Raging_Hog left	14:12
	→ chichouw joined	14:12
	→ jwmann joined	14:12
	← engkur left	14:13
	← chichou left	14:13
	← kewah left	14:13
	→ zoned joined	14:13
chilversc_	Jrobbins: you could use a separate repository for pushing to production, and control access to that repository?	14:14
	→ yankeefan04 joined	14:14
ojacobson	Jrobbins: !gitolite	14:16
gitinfo	Jrobbins: Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/	14:16
	← sushant1 left	14:16
Jrobbins	chilversc_ that would kind of be counter productive to our use of git. This is something I would really like to impliment but would not be worth restructuring our file structure	14:16
	we are currently using Stash by atlassian	14:16
	→ blami_orcl joined	14:17
ojacobson	Stash generally gives you access control by repository, AIUI	14:17
	if you're on the latest version you may! be able to write a hook plugin that prevents branch deletion	14:18
	otherwise I think you'll need to ask atlassian's support folls	14:18
	→ nkryptic joined	14:18
	→ giallu joined	14:18
	→ miso1337 joined	14:19
	→ slap_stick joined	14:19
	→ axl_ joined	14:19
Jrobbins	gotcha, I was really hoping there would be something nice and easy i could do in the config file like assign [recieve] to a branch but it appears nothing like that works	14:19
	→ JaredMiller1 joined	14:19
	← slap_stick left	14:19
Jrobbins	thats for the input though! hooks were next on my list to check out	14:20
	← arvind_khadri left	14:20
	← JaredMiller1 left	14:21
	← ESphynx left	14:22
	→ antix_ joined	14:22
	→ dherzog joined	14:23
	← nihilista left	14:23
	→ mkjackson joined	14:23
	→ Davey joined	14:23
mkjackson	is it standard to commit the .gitmodules files to the repo?	14:24
	→ itatitat joined	14:24
	→ huoxito joined	14:24
mkjackson	while some dot files (IE .gitignore) might make sense to commit, I feel like that one is not one of them	14:24
	→ nihilista joined	14:24
	← geekbri left	14:25
ojacobson	Yes, you generally check in .gitmodules if you're using submodules.	14:25
	It's used to give new clones of the outer repository their initial configuration for each submodule	14:25
	→ p3lim joined	14:27
	→ jonkratz joined	14:29
	→ malesca joined	14:30
	← nihilista left	14:30
	← antix_ left	14:31
	→ nihilista joined	14:31
	→ sebuba joined	14:32
	→ chichou joined	14:32
	→ antix_ joined	14:32
	→ mkjackson1 joined	14:32
	← SomeoneWeird left	14:32
	→ engkur joined	14:33
	← chichouw left	14:33
	← fuzzybunny69y left	14:33
	← maxloe left	14:33
	→ Guest16785 joined	14:33
	← mkjackson left	14:33
	→ mezner joined	14:34
	→ fuzzybunny69y joined	14:34
	→ geekbri joined	14:35
	→ maxloe joined	14:36
	← antix_ left	14:36
	→ chichouw joined	14:37
	→ beakman joined	14:38
	→ pen joined	14:38
	→ cannonball joined	14:38
	← chichou left	14:38
	← dhruvasagar left	14:39
cbreak	Jrobbins: sounds like a job for gitolite	14:40
	← Fandekasp left	14:40
	← oste left	14:41
	→ ThePawnBreak joined	14:41
	→ AnimeCYC joined	14:42
johnw	gitolite, to the rescue!	14:42
	→ Fandekasp joined	14:42
	→ chichou joined	14:42
	← chichouw left	14:43
	→ antix_ joined	14:44
	← beneggett left	14:44
	→ dhruvasagar joined	14:45
mkjackson1	ojacobson: damn… I want that to work but some of our team have situations where they've had to modify .gitmodules to specify which user name to use when authenticating over ssh… :-(	14:46
grawity	Then do not put user names in submodule URLs	14:46
	→ yshh joined	14:47
	→ chichouw joined	14:47
	→ mitz joined	14:47
cbreak	mkjackson1: don't modify that file	14:47
	mkjackson1: it MUST contain unpersonalized urls	14:47
	← mulinux left	14:47
	← chichou left	14:48
cbreak	remember, that file is used by default	14:48
	← mitz left	14:48
	→ notaro1997 joined	14:48
cbreak	if someone wants to override the paths, then he can do that in his own .git/config	14:48
ojacobson	mkjackson1: 'git submodule init' and its moral equivalents copy .gitmodules into .git/config	14:48
	← zoresvit left	14:48
	→ zoresvit1 joined	14:48
ojacobson	you can then modify .git/config in repo-specific ways	14:48
	gitinfo set mode: +v	14:48
ojacobson	it'll stay modified, unless you run 'git submodule sync'	14:48
notaro1997	Can someone explain what forking is?	14:48
	→ BtbN joined	14:48
cbreak	notaro1997: cloning	14:48
	← tbmiller left	14:49
csmrfx	forking? or branching	14:49
	→ sonOfRa joined	14:49
	→ thiago joined	14:49
notaro1997	so it clones the code to your account, so you can like, edit it and stuff	14:49
cbreak	forking is just an other name for cloning	14:49
csmrfx	perhaps forking? ask github	14:49
cbreak	usually drama is involved	14:49
csmrfx	perhaps branching? would be on topic	14:50
	← crocket left	14:50
	→ crocket joined	14:50
	← les_sylvains left	14:50
crocket	jast : all remote branches in a repo	14:50
	→ pencilcheck joined	14:50
crocket	Can they be cloned, too?	14:50
	← eletuchy left	14:51
	→ tbmiller joined	14:51
	→ eletuchy joined	14:51
grawity	You mean those which already exist at refs/remotes/ in the original repo?	14:51
	← beautifulmind left	14:52
	→ chichou joined	14:52
	→ Jahm joined	14:52
	→ jeffsi joined	14:52
cbreak	crocket: you'll have to use --mirror or change your fetch refspec yourself	14:52
	← jeffsi left	14:52
cbreak	note that this is usually a bad idea.	14:52
	(both)	14:52
	← Terminus- left	14:52
	→ OliverJAsh joined	14:52
crocket	cbreak: What is a bad idea?	14:52
	--mirror and fetch?	14:53
cbreak	it's a bad idea because it gets you refs/remotes too, and because it implies forcing	14:53
	← chichouw left	14:53
	← pen left	14:54
	→ penguin1 joined	14:54
crocket	cbreak: Why is getting refs/remotes bad?	14:54
cbreak	because that namespace is reserved for remotes	14:55
	← antix_ left	14:55
	→ Elvanor joined	14:56
Elvanor	I don't understand why Merge messages do not list the impacted files	14:56
grawity	git fetch origin refs/remotes/:refs/from-origin/	14:56
Elvanor	Can someone explain me what's going on there?	14:56
	← nihilista left	14:56
ojacobson	Elvanor: it would be redundant	14:57
	← kenperkins left	14:57
	→ chichouw joined	14:57
Elvanor	ojacobson: what do you mean?	14:57
ojacobson	you can list the affected files wrt any of the parents using git diff --stat or --names-only	14:57
Elvanor	ojacobson: I have a concrete case there where I really dont understand what happened	14:57
ojacobson	same as any other commit	14:57
	← mkjackson1 left	14:57
	→ kenperkins joined	14:57
ojacobson	git show foo; git diff foo^1..foo; git diff foo^2..foo will all be informative	14:58
Elvanor	I made a commit to 4 files. Later one of my colleague makes a merge. Unfortunately after the merge, none of my changes were visible	14:58
	eg, they were somehow discarded	14:58
	← chichou left	14:58
Elvanor	I have no way to see precisely what made the merge???	14:58
	→ lresende joined	14:58
	→ valgrind joined	14:59
	← _Vi left	14:59
ojacobson	I'm not sure I understand the question. "Some git-merge command" made the merge, generally. No, git doesn't record the exact options used. You can reconstruct what it did by looking at the merge, and at the diffs between the merge and its various parents	15:00
	→ antix_ joined	15:00
ojacobson	for example, if there's no diff between the merge and its first parent, then it's very likely whoever created the merge asked git to discard the second parent's changes with --ours	15:00
charon	Elvanor: note that of your cow orker simply discarded your changes (e.g. did an 'ours' merge) then 'git show $merge' will not show any conflicts	15:01
	→ m0viefreak joined	15:01
ovidiu-florin	hello world :D	15:01
	I have a remote repository, how can I pun in there two projects? each project has it's own independent .git.	15:01
cbreak	Elvanor: your coworker probably killed your changes.	15:01
ovidiu-florin	can I just place the two in the other one?	15:01
	← sonda left	15:01
cbreak	Elvanor: diff the merge with its parents	15:01
	← m0viefreak left	15:01
	→ m0viefreak joined	15:01
grawity	ovidiu-florin: no, why do you want this?	15:01
cbreak	if you diff it with your side of the merge, then you see what he did	15:01
	→ _Vi joined	15:01
ovidiu-florin	grawity: because my project is composed of two parts: client and server. but I have only one repo available, and I have to put them both over there	15:02
	→ chichou joined	15:02
Elvanor	cbreak: how can I see that?	15:02
cbreak	ovidiu-florin: git init a new one	15:02
	Elvanor: man git-diff	15:02
gitinfo	Elvanor: the git-diff manpage is available at http://jk.gs/git-diff.html	15:02
	← jargon left	15:02
	→ mkjackson joined	15:02
cbreak	git diff mergecommit^1..mergecommit	15:03
	replace 1 with 2 to see the diff with the other parent	15:03
	2 is probably your side	15:03
	← mkjackson left	15:03
ovidiu-florin	cbreak: how will that solve my problem?	15:03
	← chichouw left	15:03
	← AaronMT left	15:04
cbreak	ovidiu-florin: you'll have two remotes after initing a second	15:04
ovidiu-florin	ah...	15:04
Elvanor	I dont have the git-diff command	15:04
	is that git diff maybe?	15:04
	→ AaronMT joined	15:04
ovidiu-florin	that's the problem I have only one remote. I can't get annother one	15:04
	← marthinal left	15:04
	← OliverJAsh left	15:04
cbreak	ovidiu-florin: if you really want to shove two histories into one git repository, then you'll have to deal with manual target branch specifications	15:04
	it's possible	15:04
	but it's confusing :)	15:05
	→ sonda joined	15:05
cbreak	Elvanor: it is git diff	15:05
	Elvanor: but you have to write git-diff for the man page because man pages are just for one word things	15:05
	→ les_sylvains joined	15:05
	→ OliverJAsh joined	15:06
grawity	cbreak: `man` has added support for two-word invocations long ago	15:06
Elvanor	ha ok	15:06
grawity	you can do 'man git diff' now	15:06
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	15:06
	← chisler left	15:06
cbreak	grawity: I just get the man pages for git and diff when I do that	15:06
grawity	old man version?	15:07
	→ marthinal joined	15:07
grawity	2.6.3 here, and `man git diff` opens git-diff(1)	15:07
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	15:07
	→ chichouw joined	15:07
cbreak	1.6c	15:07
grawity	shaddup gitinfo	15:07
	→ chisler joined	15:07
	→ mitz joined	15:07
cbreak	man man says it's about 8 years old	15:07
	← chichou left	15:07
grawity	Centos? :D	15:08
compojoom	can someone tell me what is the reason for this? My master branch had in the past a directory called administrator. I've moved the files around, commited the changes. Then I've created a new branch. I've been working on the new branch for around a year and today when I try to merge - suddenly git complains about the files in administrator. Neither master, nor my new branch has this folder....	15:08
	...It has existed for a period of time... but it is not there anymore	15:08
	← sbodo left	15:08
cbreak	grawity: mac os x	15:08
	← lresende left	15:09
cbreak	compojoom: complains?	15:09
	→ JaredMiller joined	15:09
compojoom	cbreak: merge conflict - it shows the files that were in the administrator folder, but when I try to resolve it I get a fatal no stat error. The folder/files are not there anymore	15:10
	→ lethjakman joined	15:10
	← barq left	15:10
	← chisler left	15:10
	→ NiteRain joined	15:10
cbreak	pastebin git status output	15:10
	→ chisler joined	15:11
	→ Goles joined	15:11
	→ chichou joined	15:12
	→ kkimlabs joined	15:12
	← antix_ left	15:12
ovidiu-florin	I've asked my coworkers, and I've been advised to create a new empty repository (I did) and to merge the two rpojects in this one.	15:12
	and then push	15:13
	→ markalanevans joined	15:13
	← chichouw left	15:13
	→ cryosphere joined	15:13
kkimlabs	Q: Can I cherry-pick another' branch's head just by using the branch name? something like git cherry-pick some-branch:HEAD	15:13
	→ sbodo joined	15:13
	← jpcamara left	15:13
	→ sanguisdex joined	15:14
	← obiwahn left	15:14
	← JaredMiller left	15:14
	← sdeboey left	15:14
	→ jargon joined	15:15
	→ sdeboey joined	15:16
m0viefreak	kkimlabs: git cherry-pick some-branch	15:16
	→ normanrichards joined	15:16
	→ chichouw joined	15:17
cbreak	ovidiu-florin: sounds like a stupid idea.	15:17
	→ jpcamara joined	15:17
cbreak	ovidiu-florin: that'd remove your separation	15:17
	just create a new remote for the second project	15:17
sanguisdex	so I just deleted a submodule and now my main parrent repo is locked in master and git status is returning "fatal: bad default revision 'HEAD'" I bet there is a quick fix for this but I don't know what it is	15:17
	ideas?	15:17
	documentation?	15:17
ovidiu-florin	cbreak: I don't have the permission to create a new repo	15:17
cbreak	then tell who ever has to make one for you...	15:18
	→ jaequery joined	15:18
	← chichou left	15:18
bob2	submodules always sound like such a good idea	15:19
	← rakl left	15:19
Nevik	sanguisdex: how did you delete the submod?	15:20
	← dhruvasagar left	15:21
	→ dhruvasagar joined	15:21
	← thierryp left	15:21
	← _iron left	15:21
	→ tureba joined	15:21
	→ nihilista joined	15:21
charon	bob2: how about: some people, when they have a project structure problem, think: i know, i'll use submodules. now they have an arbitrary number of subproblems	15:21
jast	aww. I like submodules.	15:22
	then again I know how to use them. :}	15:22
bob2	charon, ahahahaha	15:23
sanguisdex	Nevik: deleted the ref in .gitmodules and deleted the .git folder in the submodule folder, as it was time for the submodule code to become part of the rest of the repo	15:23
	→ eijk_ joined	15:23
Nevik	sanguisdex: then your index is still referencing the submod's commit	15:24
	though git status should notice that	15:24
	sanguisdex: is there a HEAD file in your .git folder?	15:24
cbreak	sanguisdex: git rm --cached, you still have to remove the submodule	15:25
ovidiu-florin	cbreak: I've been told that the separation is not important and that I should merge the two	15:25
Nevik	ah that was how	15:25
worstadmin	I have a very odd problem where I can clone a repo on one server - and the other tells me on an intial clone that the refernces for master are pointing to an invalid object - has anyone encountered this?	15:25
sanguisdex	Nevik: there is a HEAD file	15:25
cbreak	ovidiu-florin: hmm. :/	15:25
	→ kevwil joined	15:26
ovidiu-florin	I've cloned an empty remote repository, How can I merge it with a local one?	15:26
cbreak	ovidiu-florin: no need to clone it	15:26
grawity	hmm, actually, you can have multiple separate repos in a single remote .git repo	15:26
cbreak	ovidiu-florin: if you want to merge the histories, go into one of your two repos	15:26
	add the other as remote	15:26
	and fetch	15:26
grawity	if the local git version supports namespaces	15:26
cbreak	then you can git merge	15:26
	→ alunduil joined	15:26
	← eijk left	15:27
	← flijten left	15:27
	← hafeez3000 left	15:27
	← Davey left	15:28
	← Guest16785 left	15:28
	→ Guest16785 joined	15:28
	→ antix_ joined	15:28
	Guest16785 → SomeoneWeird	15:28
	← netiul left	15:28
	→ j^2 joined	15:29
	← zoresvit1 left	15:30
	→ chichou joined	15:32
ovidiu-florin	cbreak: git remote add remotename RemoteURL/blah.git ?	15:32
	is this right?	15:32
sanguisdex	Nevik: cbreak for some reson I opened a new terminal window and it just had started working agin	15:32
cbreak	ovidiu-florin: yes.	15:32
sanguisdex	thanks for your attempted help	15:33
	exit	15:33
	← sanguisdex left	15:33
Nevik	exit!	15:33
	← antix_ left	15:33
	← b1rkh0ff left	15:33
	← chichouw left	15:33
	→ dmarkham_ joined	15:34
	← mastro left	15:35
ovidiu-florin	cbreak: after git fetch remotename should I have a new branch ?	15:35
	git branch -a returns only master	15:35
	← csmrfx left	15:36
	→ primitive joined	15:36
cbreak	you should have all branches that are on the remote after git fetch	15:36
ovidiu-florin	the remote was empty	15:36
	still is actually	15:36
	→ chichouw joined	15:37
ovidiu-florin	how do I merge them?	15:37
	actually what do I merge?	15:37
	→ gustavnils joined	15:37
ovidiu-florin	FETCH_HEAD?	15:37
	→ rakl joined	15:37
	← nihilista left	15:38
	← chichou left	15:38
	→ Davey joined	15:39
	→ JaredMiller1 joined	15:39
	→ nihilista joined	15:40
	← nihilista left	15:40
	← sbodo left	15:40
	← JaredMiller1 left	15:40
	→ nihilista joined	15:40
	→ PaulCapestany joined	15:41
	← dhruvasagar left	15:41
	← sdeboey left	15:41
	→ onerope joined	15:41
	→ ehsan joined	15:41
	→ sdeboey joined	15:42
	→ chichou joined	15:42
	← onerope left	15:42
Nevik	the fuck, has subtree still not been moved into git mainline?	15:42
	← mitz left	15:42
Nevik	.version	15:42
	gitinfo changed the topic to: Latest version: 1.8.1.3 \| http://git-scm.com \| Help with git and #git: http://jk.gs/git/ \| Getting "Cannot send to channel" or unable to change nick? /msg gitinfo .voice \| The git-jokes program needs your support! /msg EugeneKay with patches	15:42
charon	Nevik: cd contrib/subtree && make install	15:42
	→ maletor joined	15:42
Nevik	arg	15:42
	stupid bot	15:43
	→ dhruvasagar joined	15:43
	← Davey left	15:43
Nevik	charon: i dont compile it myself, that's why it's such a pain	15:43
	← chichouw left	15:43
charon	are you on windows? otherwise i don't think there's an excuse for that	15:43
	it takes like 5 minutes	15:43
	:p	15:43
	← xiangfu left	15:43
Nevik	charon: thats a lot longer than just installing the newest package :P	15:43
	→ sbodo joined	15:44
Nevik	but i guess i'll have to do that until that idiot decides to provide a contrib package	15:44
charon	but you get to use the latest and greatest! and find and report bugs, of course.	15:44
Nevik	fuck bugs	15:44
	and bug reports	15:44
	why dont they move subtree into mainline anyway	15:45
	isnt it stable enough by now?	15:45
charon	you have it backwards. where's fucking bugs and fixing women supposed to get you?	15:45
	→ pieces029 joined	15:45
	← sdeboey left	15:45
Nevik	charon: while i have no woman, that problem doesnt present itself	15:45
	→ zoresvit joined	15:45
Nevik	(i also have no bugs so the former is not posing any room for mishap)	15:45
	→ b1rkh0ff joined	15:46
charon	anyway, this seems rather derailed. if you want to argue for moving it out of contrib, i suggest you wait until the latest patch series is picked up (maybe it's been already) and then send a followup to that	15:46
Nevik	the mailing list is a pain to follow	15:46
	someone needs to write some proper online interface for mailing lists	15:46
	→ chichouw joined	15:47
	→ Tobbi joined	15:47
	→ mastro joined	15:48
cbreak	ovidiu-florin: no	15:48
	ovidiu-florin: just fetch from the remote	15:48
	then merge the branches you want	15:48
	← guampa left	15:48
	← chichou left	15:48
	← mikecmpbll left	15:48
cbreak	ovidiu-florin: the remote is not empty, it's your second repository	15:48
ovidiu-florin	cbreak: there are no branches on the remote	15:48
cbreak	then you picked the wrong remote	15:49
	→ guampa joined	15:49
cbreak	as I said ages ago	15:49
	fetch your second repository into the first	15:49
	your second repository is the remote you have to add	15:49
	→ antix_ joined	15:49
	→ madewokherd joined	15:50
	→ exoon joined	15:50
	→ chichou joined	15:52
	← dmarkham_ left	15:52
	→ mulinux joined	15:52
	← chichouw left	15:53
	← tbmiller left	15:53
	→ gregsmith joined	15:53
	→ d2dchat joined	15:54
	← mmitchell left	15:54
	eijk_ → eijk	15:54
	→ mmitchel_ joined	15:54
	← suy left	15:54
	← ThePawnBreak left	15:54
	→ ThePawnBreak joined	15:55
	← milStaja left	15:55
	→ didlixForHire joined	15:55
	← baneM left	15:55
	→ mikecmpbll joined	15:56
	→ sluther joined	15:56
	← pandeiro left	15:57
	→ baneM joined	15:57
	→ suy joined	15:59
	← dhruvasagar left	16:00
	← antix_ left	16:01
	← Sigma left	16:01
	← chichou left	16:02
	→ dhruvasagar joined	16:02
	→ pepellou joined	16:03
	← mmitchel_ left	16:03
	← trungqtran left	16:05
	← maletor left	16:06
	→ dreiss joined	16:06
	← zoned left	16:06
	← jargon left	16:06
esc	anyone have experience with http://rhodecode.org/	16:07
	→ mrwonko joined	16:08
	→ sfalcon joined	16:08
	← mikeym_ left	16:08
	→ Davey joined	16:09
	← OOPMan left	16:10
jonkratz	Looks like a decent option if you want to host your own server. Otherwise, why not bitbucket?	16:10
	← sluther left	16:10
tango_	why not zoidberg?	16:10
	← eletuchy left	16:10
	→ p3rror joined	16:11
	← kishorebabu left	16:11
	→ eletuchy joined	16:11
esc	tango_: have a url for that?	16:11
	← tinti left	16:11
jast	sure	16:11
	http://en.wikipedia.org/wiki/Doctor_Zoidberg	16:11
	← les_sylvains left	16:11
	← thiago left	16:11
jonkratz	http://knowyourmeme.com/memes/futurama-zoidberg-why-not-zoidberg	16:11
esc	jonkratz: dunno, i just discovered it	16:12
	http://www.tumblr.com/tagged/zoidberg%20gif	16:12
jonkratz	Are you wanting to host your own server?	16:12
esc	jonkratz: my goal is just to know what is out there	16:12
	→ kishorebabu joined	16:12
	← meLon left	16:12
esc	i have been playing with gerrit recently, and it is bit heavyweight for my liking	16:13
	→ mikeym_ joined	16:13
	→ meLon joined	16:13
	→ etcetera joined	16:13
esc	i mean, the space of git webapps is rapidly increasing	16:13
	← etcetera left	16:13
esc	some have more features than others	16:13
jonkratz	esc: sure, not a bad deal. I'm curious if anyone has experience. If you're looking for just a place to host your code, bitbucket is great for private repos, and github is good for public	16:13
esc	jonkratz: i have access to a couple of machines with gitolite on them too	16:14
	→ antix_ joined	16:14
	← sigkill left	16:14
esc	mainly I am evaluating potential solutions to setup at customer sites	16:14
	ldap/ad integration and single-sign on are kinda important	16:15
	→ soypirate joined	16:15
jonkratz	esc: sure, how do you like gitolite?	16:15
esc	it is great, but it is only one part of the solution	16:15
	← itatitat left	16:16
mrwonko	Hi. I need some git help: A couple of months ago, before they used git, we forked ioquake3, which was still using svn at the time, and have since used git for version control. So the first commit was all of ioquake3 in its state back then. Occasionally new patches were manually applied. Ioquake3 has moved to git since and undergone a lot of changes (including renamed files etc.). We'd like to get back in sync so we can add the official repo	16:16
	as a remote and just fetch updates that way from now on. I've started by cloning the current state of ioquake3 and now I'd basically like to apply the entire history (about 300 commits) of our fork, safe for the parts that contained ioquake3 patches. How would I do that? I'll likely have to resolve a bazillion conflicts, but I'm okay with spending some time on it if it means less hassle in the future.	16:16
	→ sluther joined	16:16
esc	i have a setup that queries LDAP/AD instead of using ssh-keys	16:16
	→ rasmusto joined	16:16
esc	mainly by doing crazy shit with kerberos	16:16
	→ maletor joined	16:16
esc	my old sysadmin set it up, I am still sifting through the files, to get a handle on it	16:17
	→ axl__ joined	16:17
	← jpcamara left	16:17
esc	the main problem with repo-browsers like goblet, cgit, gitiles and gitweb is that they don't really support authn/authz	16:18
mrwonko	I guess I could just create a diff of the entire history and then manually apply it, filtering out the important parts?	16:18
esc	you can do the authn in your front-end webserver, but the authz remains	16:18
grawity	my todo list still has "send feature request for openssh: export kerberos principal as environment variable" from two years ago... if that were implemented, it'd make gitolite+kerberos very easy.	16:18
	← axl_ left	16:19
	axl__ → axl_	16:19
charon	mrwonko: is your own history of 300 patches linear?	16:19
esc	grawity: we have some kind of weird config setup for sshd	16:19
charon	mrwonko: or more git-like and all tangled	16:19
	← Xethron left	16:19
mrwonko	charon: as in free of branches? not quite, there were some, but they've since been merged back	16:19
	→ chichou joined	16:20
	→ Xethron joined	16:20
	← smirn0v left	16:21
	← alunduil left	16:21
mrwonko	the repo in question is https://github.com/dmead/jkaq3 (we're trying to recreate Jedi Academy MP, which uses a modified quake 3 engine, based on ioq3)	16:21
	← ferdev left	16:21
	→ alunduil joined	16:21
charon	mrwonko: tricky. i'd probably try to rebase your work over on their git in little chunks. for the most part big chunks should also work, but with too much parallel work you may have to redo some of the branching	16:21
	when you hit a patch from upstream, --skip it	16:21
	(or exclude up-front in the interactive rebase instructions)	16:22
	← sluther left	16:22
	→ Noldorin joined	16:22
	→ zoned joined	16:22
	← LeMike left	16:22
charon	mrwonko: another option you can investigate is using git-replace to "identify" (in the "make the same object" sense) your initial commit with the corresponding version in upstream	16:22
	then merge, either in one go or more likely in little steops	16:22
	that does have the disadvantage that even if you get through all the conflicts, you'll have somewhat ugly redundant imports	16:22
	← meLon left	16:23
mrwonko	I see. I might opt for the rebase then.	16:23
	→ meLon joined	16:23
	→ lavaman joined	16:23
	← antix_ left	16:23
mrwonko	not quite sure how I'd do that, but I'll do some reading on the topic	16:23
charon	third option is to go and make a rebase --preserve-merges that actually works	16:23
	then use that	16:23
	:)	16:24
	← psoo left	16:24
charon	mrwonko: well, if you only had a small linear run of modifications, using rebase would be easy, right?	16:24
	(otherwise read a tutorial on rebase)	16:24
	note that it's absolutely no problem to fetch both histories into the same repo, which is somewhat of a requirement here	16:24
	← yacks left	16:24
charon	mrwonko: it starts getting annoying when the rebase (with implied linearization) starts getting conflicts among your patches, hence the suggestion to do it in small chunks and (where needed) actually reconstruct side branches.	16:25
mrwonko	yeah, I'll have quite some reading to do... I've mostly just used git for the history's sake, with occasional branching, so far.	16:25
	some tagging and stashing and that's about it	16:26
charon	oh, btw, option 2(b) is as follows: identify the initial commit, and also identify a newly-made "fake" merge wherever you imported from upstream	16:26
	→ danman joined	16:26
	← stliu left	16:26
	← j^2 left	16:26
charon	if you actually faithfully and completely integrated their changes, this should "just work"	16:26
	→ j^2 joined	16:27
charon	otoh it'll suck at detecting that you did not really merge their changes completely.	16:27
	→ sfvivek joined	16:28
	← tbosua left	16:28
	← marhaban left	16:28
	→ hafeez3000 joined	16:29
	→ yacks joined	16:29
	← LekeFly left	16:30
	→ marhaban joined	16:30
	→ EF0 joined	16:30
	→ kadoban joined	16:30
	→ s00pcan_ joined	16:30
	← s00pcan_ left	16:30
	← ThomasLocke left	16:30
	→ ferdev joined	16:31
charon	mrwonko: note that you should really grok !cs and what it means that in git, every commit is a "snapshot", before attempting a solution that involves git-replace. otherwise you'll cause a big mess.	16:31
gitinfo	mrwonko: "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup	16:31
	← Davey left	16:31
	← jonkratz left	16:32
	→ antix_ joined	16:32
	← madewokherd left	16:34
mrwonko	I'm fine with directed acyclic graphs, I'll take a look at that, thanks	16:34
	so I can just enter !concepcts or !bottomup and gitinfo will tell me stuff?	16:34
gitinfo	'Git from the bottom up' starts with explaining the building blocks of git and proceeds to tell you how they fit together. http://ftp.newartisans.com/pub/git.from.bottom.up.pdf	16:34
mrwonko	* !concepts	16:35
gitinfo	"Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs/	16:35
	→ mmitchell joined	16:36
	← valgrind left	16:36
	→ spaceonline joined	16:36
	← wrouesnel left	16:36
	→ Ishango joined	16:37
	← chichou left	16:37
	← antix_ left	16:38
charon	actually on 2(b), i think the following would be a reasonably safe approach: progressing from older history, stop at every commit P that imported from upstream by applying a patch. P^ is obviously your history before that "merge". identify the commit U corresponding to the revision that you imported. merge P^ with U (using git-merge as usual). if any conflicts arise, fix them exactly the same way as in P. call this merge M. use 'git diff P M' to ensure	16:38
	that you got the "re-merging" right (this provides a level of safety). once you did so (and note that you can amend merges, too), use 'git replace P M' to tell git that it should use M instead of P in the history.	16:38
	← mikecmpbll left	16:38
	→ jargon joined	16:38
charon	that's all very conditional on the imports being clean, full "merges" of what upstream did in the meantime, but at least doing these diffs tells you if something is wrong	16:38
	→ sluther joined	16:39
charon	while you're reading, also get comfortable with the consequences of !rewriting since you'll inevitably hit that	16:39
gitinfo	Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to jump through hoops (and you have to tell them to), so it's infinitely better to just move on without rewriting. That said, you can use `git push -f` to force your new history through (may get rejected, though). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstrea	16:39
	→ goshawk joined	16:39
	→ netiul joined	16:39
charon	it will mean a flag day for your project, but for the benefits of linking with usptream it's probably worth it	16:39
	→ xxHarlequin joined	16:40
charon	(the other option would be to provide a set of replace refs that people can fetch to achieve the linking)	16:40
	→ nfroidure joined	16:40
	← Iszak left	16:40
	→ lioux joined	16:40
	→ infynyxx joined	16:40
	gitinfo set mode: +v	16:41
	← MistuhKurtz left	16:41
	← markalanevans left	16:41
	← rath left	16:41
	→ Davey joined	16:41
	← Sonderblade left	16:42
	← xHarlequin left	16:43
	→ jonkratz joined	16:43
	← dsdeiz left	16:43
lioux	I need a piece of include/linux/tcp.h -> http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/log/include/linux/tcp.h. I need tcp: report ECN_SEEN in tcp_info. However, I want to know if this commit touched more files. Can I find out through the web git interface. I don't want to import a partial fix. This is for an Android AK Kernel for Galaxy Nexus	16:43
	← gregsmith left	16:43
	← hyperair left	16:43
	← infynyxx left	16:43
	→ marhaban1 joined	16:44
	→ infynyxx joined	16:44
	← sluther left	16:44
charon	lioux: so you'd like to "copy" a fix from upstream to your own kernel repo? why aren't you using git-cherry-pick?	16:44
	→ milStaja joined	16:45
	← RaceCondition left	16:45
	← JesseC left	16:46
lioux	The thing is: I used cherry pick to get all 3 parts of TCP Early Retransmit. Then to find out that it needs a define in include/linux/tcp.h from a previous patch to THAT file. So I want to cherry pick everything from the latest version version of tcp.h I have to the version that has the code I need: tcp: Replace constants with #define macros	16:46
	However, I want to "reduce" the mess and import the all files touched by the 2 commits on tcp.h.	16:47
	→ scorphus joined	16:47
	→ noam_ joined	16:47
lioux	That's why I want to if there is a way to find out if any other files were part of a larger commit when tcp.h was changed	16:47
	it's a mess... I know	16:47
charon	sounds like this will quickly devolve into a whole chain of things you need to import	16:47
	anyway, use git-blame to find out what the commit was, then git-show to see what it touched	16:47
	→ mezgani joined	16:48
andres	with --stat for a quicker overview	16:48
lioux	I found out the change needed through the web interface ... touch a while: ab56222a32b9dbaae19c1d37f07b0ac4fc3c27ec.	16:48
	Use git-show with that?	16:48
	← yankeefan04 left	16:48
charon	yes. or paste it into the webinterface's "show me a commit" url, which amounts to the same	16:49
	→ RaceCondition joined	16:49
	→ JaredMiller joined	16:50
	← Fullma left	16:50
	→ dogarrhea joined	16:50
	← JaredMiller left	16:50
	← p3rror left	16:51
	← milStaja left	16:51
	← zoned left	16:51
	← fatninja left	16:52
	→ didlixFo_ joined	16:52
	← marhaban1 left	16:52
	→ fatninja joined	16:52
	← didlixForHire left	16:55
	→ War2 joined	16:58
	← War2 left	16:58
	→ War2 joined	16:58
mrwonko	Couldn't I just cherry-pick all the relevant commits from our fork into the most recent version of ioquake3?	16:58
	(although that will take forever with 200+ commits, but I'll have to do a lot of work either way)	16:59
	→ maxibyte joined	17:00
	← rue left	17:00
	→ MistuhKurtz joined	17:00
	→ zoned joined	17:00
	→ rue joined	17:01
	→ markalanevans joined	17:01
	← breakingthings left	17:01
	← engkur left	17:01
	← markalanevans left	17:01
chilversc_	is there a nice way to handle rolling a branch back to a specific point, and moving those changes to another branch?	17:02
	→ jjore joined	17:02
	← gain left	17:02
	→ markalanevans joined	17:03
Schmoozerd	mrwonko - read the manpage for rebase, especially the --interactive part, this should help you a lot (at least with the management overhead)	17:03
grawity	chilversc_: to a new branch?	17:03
charon	chilversc_: git checkout curbranch; git branch newname; git reset --hard oldpoint	17:03
	← dreiss left	17:03
	← spaceonline left	17:03
chilversc_	all I can think of is creating a single reverse merge back to where I want to be, branching the head, and reverting the reverse merge on the new branch	17:03
charon	chilversc_: warnings apply as to --hard	17:03
	← mulinux left	17:03
charon	chilversc_: also see !rewriting	17:03
gitinfo	chilversc_: Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to jump through hoops (and you have to tell them to), so it's infinitely better to just move on without rewriting. That said, you can use `git push -f` to force your new history through (may get rejected, though). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstrea	17:03
chilversc_	pushed history though	17:03
charon	ah well, then you can't really do this. if anything you'd have to revert the work that has happened since the old point	17:04
	→ beneggett joined	17:05
	→ tmcmahon joined	17:05
	→ dmarkham_ joined	17:06
chilversc_	something like git revert HEAD~3..HEAD, but only creating a single commit	17:07
charon	git read-tree HEAD~3; git commit	17:07
	only works if you're reverting up to HEAD though	17:07
	→ jebenexer joined	17:07
charon	otherwise something like git diff HEAD~3..HEAD \| git apply; git commit	17:08
	no wait	17:08
	git diff HEAD~3..HEAD \| git apply -R; git commit	17:08
chilversc_	nm, git revert --no-commit rolls all commits in to one (without actually committing)	17:09
	so that seems to be the best way, revert on the main branch, create a new branch, and revert the revert	17:10
	→ engkur joined	17:12
	← j^2 left	17:13
	← dhruvasagar left	17:13
	← tmcmahon left	17:13
mrwonko	this is a lot to take in... okay, so git rebase -i gives me a list of commits, by default marked to be applied ("pick"), that I can replace with "edit" if I feel I need to change something (I really just want to edit when there's a conflict, which I guess should automatically cause a pause?) or "fixup" to merge them into previous commits. I guess if I don't want them at all I can just omit the line? And I can apply them onto a different code	17:13
	base as well?	17:13
	← LimeBlast left	17:13
mrwonko	I should do some simple tests on this before trying to do the whole thing	17:13
	→ j^2 joined	17:14
	→ psoo joined	17:14
	→ ScottE joined	17:14
chilversc_	omitting the line will drop that commit as if it never happened	17:14
	← PaulCapestany left	17:14
	← dpac\|away left	17:14
chilversc_	mrwonko: tip, make a temp branch before you begin rebasing so you can always undo it	17:14
	→ Yuuhi joined	17:14
	→ dpac joined	17:14
mrwonko	yes, I'm definitely working on a new branch	17:14
chilversc_	mrwonko: copying over to somewhere else is a little trickier, if its only a few commits I think cherry-pick is simpler	17:15
mrwonko	we're talking about roughly 300 commits, including some to be dropped	17:15
chilversc_	mrwonko: I mean if its only 3~4 you want on the other branch	17:15
	← dc5ala left	17:15
chilversc_	mrwonko: though, ~300 seems like you probably shouldn't be rebasing	17:15
	← suy left	17:15
	→ les_sylvains joined	17:17
mrwonko	hmm	17:17
	okay, let's keep it simple for now	17:17
	← Noldorin left	17:18
	→ mgedmin_ joined	17:19
mrwonko	I have one remote, let's call it ioq, and another, jka. I want to apply the first commit after the initial one from jka onto ioq. (and then the second one, the third one, maybe omit one, and so on). git cherry-pick is my friend?	17:19
	or should I just git diff and then patch?	17:19
	← mgedmin_ left	17:19
	→ opus joined	17:20
	← R1KRDO left	17:21
	→ R1KRDO joined	17:21
	→ gregsmith joined	17:21
	← didlixFo_ left	17:21
	→ Chillance joined	17:21
chilversc_	mrwonko: why not merge?	17:22
	← jn_ left	17:22
lethjakman	hey, if I git reset it'll --hard it'll remove all commits after the commit I'm sitting on right?	17:22
	→ aweorit joined	17:22
lethjakman	git reset --hard	17:22
mrwonko	chilversc_, if I can, sure. Have you read the description I previously posted?	17:22
	the thing is that git is not aware of the point where we branched of since we forked ioq when it was still using svn	17:23
	and some of its patches have been manually applied	17:23
	← infynyxx left	17:23
mrwonko	I don't think all of them, and possibly not cleanly	17:24
	I'm only just joining the project and trying to get it in a stage where merging is possible in the future	17:24
	→ robinmholt joined	17:25
	dwmw2 → dwmw2_gone	17:26
	← GenTAR left	17:26
	← skorgon left	17:26
chilversc_	mrwonko: the rebase --onto would let you move a whole set of commits from one branch to another, but that's a lot of history to rewrite	17:26
mrwonko	it is.	17:27
chilversc_	mrwonko: which is fine if you have the only clone, but if anyone else has cloned your repository, they will have to repeat this, or somehow rebase their changes on to your new version	17:27
	so I guess the first thing that matters is how public is this, ie will re-writing the history like this cause lots of problems	17:27
	→ jluc_ joined	17:27
chilversc_	as its esentually, throwing away the whole repository, and creating a new but very similar looking repository	17:28
mrwonko	yeah, but I think the cost of everybody having to rebase is outweighted by the benefit of being able to merge in from upstream in the future. It's not very public, I think we have <10 branches	17:28
chilversc_	true, but if you can sort out some merges to bring it in to line rather than rebasing that might also work	17:29
Schmoozerd	though as you have no shared history, you will eventually get problems with the upstream - so making the (maybe required) cut as soon as possible might be the less bad choice	17:29
	gitinfo set mode: +v	17:29
mrwonko	sounds like a nice solution, how would that work?	17:29
	← oojacoboo left	17:29
	→ dhruvasagar joined	17:29
chilversc_	mrwonko: it might not, that's the problem, as Schmoozerd said, with no common history its always going to be a pain	17:29
mrwonko	so starting over via rebase might just be the best thing to do	17:30
chilversc_	mrwonko: the merge basically works by treating it as if it were 2 separate projects, with separate history that at some point were merged together	17:30
	and it does take some fiddling to make it work	17:30
	← Gentlecat_ left	17:31
	raek_ → raek	17:31
	← Xethron left	17:31
	gitinfo set mode: +v	17:31
mrwonko	okay, say I were to rebase. I suppose I can do that one commit at a time to get the hang of it?	17:31
	← zoresvit left	17:32
mrwonko	and to be able to test it continuously	17:32
	→ macrover joined	17:32
mrwonko	and the command for that would be rebase --onto	17:32
	I'll need to fetch the second remote first, I guess, and put that in a separate branch	17:33
	← douglaslassance left	17:34
	→ looper joined	17:34
	← jonkratz left	17:34
	→ whowantstolivef1 joined	17:35
	← tallest_redd left	17:36
	← mezgani left	17:37
	→ p3rror joined	17:37
	→ s00pcan_ joined	17:37
	← s00pcan_ left	17:37
	→ tallest_red joined	17:38
	← Tobbi left	17:38
	← Guest71029 left	17:38
	→ trungqtran joined	17:39
	← dpac left	17:39
	→ rideh joined	17:39
	→ s00pcan_ joined	17:40
	← s00pcan_ left	17:40
	→ dpac joined	17:40
	→ s00pcan_ joined	17:41
	→ undecim__ joined	17:41
	← gregsmith left	17:41
	→ marcdel joined	17:42
	← macrover left	17:42
	← fuzzybunny69y left	17:42
	← s00pcan_ left	17:43
	← maqr_ left	17:44
	← undecim_ left	17:45
	← blami_orcl left	17:46
	→ blami_orcl joined	17:46
	→ thierryp joined	17:46
	→ hwrd\|work joined	17:48
	→ douglaslassance joined	17:48
	← psoo left	17:49
	← angelsl left	17:50
	← normanrichards left	17:50
	→ vdv joined	17:50
	← brocken left	17:53
	→ didlixForHire joined	17:53
	→ mikef joined	17:55
	→ jjido joined	17:55
	← penguin1 left	17:55
	← sfalcon left	17:56
mrwonko	so if I want to rebase a branch a couple of commits at a time, I need to keep creating new copies of it, then rebase from that, right? something like git checkout rebaseMe && git checkout -b tmp && git rebase --onto rebaseTarget -i <firstCommit^> repeatedly? and that apparently makes tmp == rebaseTarget, so I can delete tmp?	17:57
	← geekbri left	17:58
	← _goTAN left	17:58
mrwonko	or do git merge rebaseTarget in the end?	17:58
	← douglaslassance left	17:58
	← zoned left	17:59
	← j^2 left	17:59
	→ alvaro_o joined	18:00
	→ j^2 joined	18:00
	→ penguin1 joined	18:00
	← maxmanders left	18:01
	→ geekbri joined	18:01
	→ pfefferz joined	18:01
	← marcdel left	18:02
	← AccessGranted left	18:02
	← mastro left	18:03
jluc_	hello	18:03
gitinfo	jluc_: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	18:03
	← GodEater left	18:03
	→ GodEater joined	18:03
	← Thra11 left	18:03
	→ mviitanen joined	18:03
jluc_	i'm a basic git user : i pull changes, compile and test the new version	18:03
	i'm used to gitg to view the changes and the change lanes	18:04
	→ PaulCapestany joined	18:04
jluc_	i have 2 repo and a couple of branches on each	18:04
	right now i have a problem with gitg :	18:04
	when i open one of the repo, i access the other one instead	18:04
	Is there a way to cure this ?	18:05
lov	try eating some pills	18:05
	→ sluther joined	18:05
lov	it doesn't really matter which pills, just pick some at random	18:05
	→ douglaslassance joined	18:05
lov	I find that's the best way to cure my problems	18:05
mrwonko	on rebase: "error: addinfo_cache failed for path '...'" - I guess that's just the kind of error I should expect when trying to rebase on top of a changed file? how do I resolve it?	18:05
jluc_	thanx love	18:06
	→ mastro joined	18:06
	← lindenle left	18:06
jluc_	i've tried other tools : gitk, qgit, giggle	18:06
	→ tenmilestereo joined	18:06
	→ cebor joined	18:06
jluc_	but it seems all branches are allways mixed together in the display of these guis	18:07
	← trungqtran left	18:07
jluc_	when in gitg i could choose which branch to watch	18:07
	→ lindenle joined	18:07
	→ apok joined	18:07
	→ trungqtran joined	18:08
jluc_	is there something i'm missing ?	18:08
	(à part les pillules d'amour)	18:08
	← mikef left	18:08
	→ etcetera joined	18:09
	← alunduil left	18:09
	← sluther left	18:09
	← thierryp left	18:11
	← EF0 left	18:11
_ikke_	jluc_: gitk by default only shows the current branch	18:12
	jluc_: unless you specify --all	18:12
	(and any branch reachable by current)	18:12
	→ system64 joined	18:12
	→ s00pcan_ joined	18:13
	← alkino left	18:13
	→ beaky joined	18:14
beaky	hello	18:14
	how do I rename a branch?	18:14
	→ ferdev_ joined	18:15
beaky	I tried git branch -m and it seemed to work	18:15
	← ferdev left	18:15
	ferdev_ → ferdev	18:15
	← exoon left	18:15
_ikke_	So problem solved?	18:16
	← mastro left	18:16
	→ rath joined	18:16
	→ exoon joined	18:16
	← s00pcan_ left	18:16
	← opus left	18:16
	← jjido left	18:17
	→ jjido joined	18:18
	← Asenar left	18:18
	→ lacrymology joined	18:18
	→ Thra11 joined	18:18
	← wasanzy left	18:19
	← ferdev left	18:19
	← markalanevans left	18:20
	→ ferdev joined	18:20
	→ skorgon joined	18:20
	→ vaanz joined	18:20
	← alvaro_o left	18:21
	→ alunduil joined	18:22
	← lioux left	18:22
	← rurufufuss left	18:24
	← thatguycraig left	18:24
	→ Sigma joined	18:25
	→ marcdel joined	18:25
	→ Octalot joined	18:25
	← ak5 left	18:25
	→ opus joined	18:27
	← R1KRDO left	18:28
Blaster	hey how come I would get	18:28
	! [rejected] master -> master (non-fast-forward)	18:28
	→ R1KRDO joined	18:28
Blaster	when I do a git push	18:28
	but when I git pull it says already up to date	18:28
	→ mastro joined	18:28
j416	!non-ff	18:28
jluc_	wait i've not gitk installed anymore so i try again _ikke_	18:28
j416	hm	18:28
Blaster	and git status is clean	18:28
j416	!faq non-ff	18:28
gitinfo	Your push would lose changes on the remote; more details available at http://jk.gs/gitfaq#non-ff	18:28
j416	Blaster: ↑	18:28
charon	Blaster: are you on the master branch? is the upstream set correctly?	18:28
Blaster	ahh	18:29
	I wasnt on master	18:29
	→ ni291187 joined	18:29
charon	see push.default in man git-config to see why push tried to push master anyway	18:30
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	18:30
	← Derega left	18:30
	→ ironm joined	18:30
	← ni291187 left	18:31
jluc_	i'm afraid i'm lost without gitg but i'll have to sort it out some other time	18:31
	← jluc_ left	18:31
	→ breakingthings joined	18:32
ironm	Hello. Please allow me one question. How can I rename a name of an existing git project using git commands?	18:32
	Or should I delete this project and create a new one with new name?	18:32
	thank you in advance for any hints	18:32
	← marthinal left	18:32
cmn	git doesn't have the concept of a project, what are you trying to do? rename the dir?	18:33
lucs	ironm: You can just rename the directory holding the project, and that's it, nothing else to change.	18:33
	→ ni291187 joined	18:33
ironm	cmn, yes ..	18:34
cmn	then just rename it	18:34
ironm	lucs, the git repo is on the server	18:34
cmn	then rename it there	18:34
ironm	and I use gitolite	18:34
cmn	ok, start with that	18:34
	← ni291187 left	18:35
ironm	cmn, sorry .. I didn't mention it before	18:35
	→ mezgani joined	18:35
cmn	http://gitolite.com/gitolite/repos.html	18:36
lucs	ironm: The point is, the repo itself doesn't care about the name of the directory holding it, but yeah, if stuff is pointing to that directory and you change its name, you have to adjust.	18:36
	→ marthinal joined	18:36
cmn	the gitolite documentation will generally have anything you're looking for	18:36
	← jjido left	18:36
	→ jjido joined	18:37
	← jjido left	18:37
	→ jjido joined	18:37
ironm	lucs, I have to change gitolite.conf .. and probably rename manually the dir .. but it is maybe not clean way	18:37
	→ madewokherd joined	18:38
ironm	cmn, I try to find it there	18:38
	← p3rror left	18:38
ironm	thanks a lot lucs and cmn	18:38
grawity	it's at the very bottom of that page	18:38
ironm	grawity, what page ?	18:38
	→ fobban joined	18:39
ironm	the single one ?	18:39
	→ jn_ joined	18:39
	← tallest_red left	18:39
	← netiul left	18:40
	← jjido left	18:40
cmn	the documentation I just pointed you to	18:40
	→ jjido joined	18:40
ironm	cmn, thank you :)	18:41
	→ gregsmith joined	18:41
beaky	how do I abort a merge? I got a merge failure	18:42
	→ ni291187 joined	18:42
grawity	git merge --abort	18:42
beaky	ah thanks	18:42
	← ni291187 left	18:42
	→ tallest_red joined	18:43
	← jjido left	18:43
	← kishorebabu left	18:43
	← jargon left	18:44
	← Jrobbins left	18:44
	→ sukima joined	18:44
sukima	How do I move a commit in the middle of a branch into it's own branch: https://gist.github.com/sukima/465c64e348da665bc818	18:45
ironm	any hint what keyword I should look for? (to rename or delete the project with gitolite) ... didn't find it in the docu yet	18:45
robinmholt	sukima: git cherry-pick?	18:45
m0viefreak	sukima: git branch newbranch A; git checkout newbranch; git cherry-pick C	18:45
	← hafeez3000 left	18:45
robinmholt	ironm: http://gitolite.com/gitolite/repos.html	18:45
sukima	But won't that merge the commit not move it?	18:45
robinmholt	sukima: Define move? Do you want to get rid of in the original? Can you 'git rebase -i' on that branch?	18:46
	→ buibex joined	18:46
ironm	robinmholt, great. Thank you very much!	18:46
robinmholt	ironm: thank cmn. I just reposted it.	18:47
ironm	right .. thank you very much ... I didn't notice it: <cmn> http://gitolite.com/gitolite/repos.html	18:48
	← tallest_red left	18:48
robinmholt	sukima: OK. Looked at your diagram.	18:48
	git checkout -b other A; git cherry-pick C; git checkout E; git rebase -i B # Remove C from the list	18:48
sukima	perfect thank you	18:49
scorphus	!fixup	18:49
gitinfo	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!	18:49
	gitinfo set mode: +v	18:50
	→ tallest_red joined	18:50
	← didlixForHire left	18:50
	→ pandeiro joined	18:50
	← WinstonSmith left	18:51
	→ didlixForHire joined	18:51
ironm	cmn, It looks I was blind before as I didn't noticed your kind hint. Thanks again :)	18:52
	→ rymate1234 joined	18:52
rymate1234	hi	18:52
gitinfo	rymate1234: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	18:52
rymate1234	so	18:52
	kenperkins → kenperkins\|away	18:52
rymate1234	I did git stash before I git pulled some changes	18:52
	where do my changes get stashed?	18:52
grawity	`git stash list`	18:53
	technically they're kept in the reflog of refs/stash, but it's best not to mess with that directly	18:53
	← kevwil left	18:54
Nevik	my stash contains: 3 cuban cigars, 20g of marihuana and 3kg of sugar	18:54
	ahw	18:54
rymate1234	;p;	18:54
Nevik	that wouldve been funnier if you hadnt said anything meanwhile	18:54
rymate1234	lol	18:54
	thanks grawity	18:55
	→ kevwil joined	18:55
	← didlixForHire left	18:56
	← yacks left	18:56
	→ WinstonSmith joined	18:57
	← Elvanor left	18:57
	← ferdev left	18:58
	→ _iron joined	18:59
	← bitkiller left	19:00
	← mezgani left	19:01
	← tsdh left	19:01
	→ x82_nicole joined	19:01
	gitinfo set mode: +v	19:01
	← mklappstuhl left	19:02
	→ p3rror joined	19:02
x82_nicole	If I have a production and staging repo, and use GitHub as a central repo (two branches there, prod/staging). If I push the prod branch to the staging repo, does that overwrite the branch? Like, how would I revert that? Can I switch remote branches? :/	19:03
	→ jarr0dsz joined	19:04
	← War2 left	19:05
	→ bitkiller joined	19:06
	← Sigma left	19:06
	← thegorn left	19:07
	← babilen left	19:07
	→ thegorn joined	19:07
	← m0viefreak left	19:07
	← PaulCapestany left	19:08
	← _Vi left	19:08
	← alunduil left	19:08
	→ babilen joined	19:09
	→ infynyxx joined	19:09
	→ zoned joined	19:10
	← infynyxx left	19:11
	→ fuzzybunny69y joined	19:12
	→ alunduil joined	19:12
	← cebor left	19:12
	← tenmilestereo left	19:12
	← Goles left	19:13
	← opus left	19:13
	← johnkpaul left	19:14
	→ Goles joined	19:14
	→ johnkpaul joined	19:14
	← Thra11 left	19:15
	→ Thra11 joined	19:16
_ikke_	x82_nicole: git by default blocks non fast forward pusehs	19:16
	pushes	19:16
	x82_nicole: But you can overwrite a branch	19:17
	x82_nicole: But what do you want to achieve?	19:17
x82_nicole	correct my mistake. Can I just push the production branch? It should just overwrite everything?	19:17
	→ alexbligh1 joined	19:17
_ikke_	x82_nicole: What do you mean with overwrite?	19:18
x82_nicole	I pushed the wrong branch, so if I push the right branch, does the remote repo switch branches? merge them?	19:18
_ikke_	x82_nicole: The remote by default does othing	19:18
x82_nicole	Does it just deploy whatever branch you push last?	19:18
alexbligh1	with git diff, I can show the diff effect of a single git with git diff ffbfc93{^,}. With git format-patch I need to write git format-patch ffbfc93^..ffbfc93 apparrently. Writing the former does not cause format-patch to parse the two revisions as a range. Is there an easier way?	19:18
_ikke_	x82_nicole: !deploy	19:18
gitinfo	x82_nicole: Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html	19:18
_ikke_	x82_nicole: git doesn't update files on push	19:19
alexbligh1	s/of a single diff/of a single commit/	19:19
	← jaequery left	19:20
	← Goles left	19:22
	← breakingthings left	19:22
	→ Goles joined	19:22
	← sonda left	19:24
	← SonikBoom left	19:24
	→ tinti joined	19:25
	← Vile left	19:26
	← _iron left	19:26
	← huoxito left	19:28
	← marcdel left	19:30
	→ ferdev joined	19:31
	← fobban left	19:35
	← Thra11 left	19:35
	← chisler left	19:37
	← douglaslassance left	19:37
	→ tagava joined	19:38
	→ s00pcan_ joined	19:38
	→ oojacoboo joined	19:38
	← s00pcan_ left	19:39
	→ s00pcan_ joined	19:40
	← s00pcan_ left	19:40
	→ breakingthings joined	19:40
	→ huoxito joined	19:41
	← relipse left	19:43
	← gregsmith left	19:45
	← jkremser left	19:45
	→ puchrojo joined	19:45
	→ fobban joined	19:46
	→ otc joined	19:46
	→ tmcmahon joined	19:46
	← tmcmahon left	19:47
	← OliverJAsh left	19:47
_ikke_	.version	19:48
	→ hays joined	19:48
_ikke_	1.8.2 has been announce, but probably not yet propagated	19:48
hays	I have changed a file in a git repo, and added it to the commit.. I was thinking that 'git diff' would show me the differences between the what is committed and what is not... I think I am wrong about this. How can I do what I want to do?	19:49
cmn	_ikke_: the data source for that is wrong	19:49
	→ OliverJAsh joined	19:49
	→ jalama joined	19:49
_ikke_	cmn: It takes a while for the source to get updated	19:49
BinGOs	it scrapes git-scm.com to get the version	19:49
_ikke_	hays: git diff --cached	19:49
cmn	_ikke_: I know, but right now that source is not updating at all	19:49
_ikke_	define at all?	19:50
hays	_ikke_: thanks	19:50
	→ jjore_ joined	19:50
cmn	it stopped updating after 1.8.1.3 was released	19:50
	it completey missed 1.8.1.4 and .5	19:50
	→ Thra11 joined	19:51
_ikke_	Ah ok	19:51
	!seen chacon	19:52
	.seen chacon	19:52
	hmm,	19:52
cmn	I fixed the code (sort of), but I don't think it's been deployed	19:52
charon	wasn't his nick schacon	19:53
	?	19:53
_ikke_	tyes	19:53
	But not sure if the but even supports that	19:53
	→ relipse joined	19:53
	→ Praise joined	19:54
	→ Kamal_ joined	19:54
	← RaceCondition left	19:55
BinGOs	its a simple parse of the html from git-scm.com to find the version.	19:55
	← kmels__ left	19:55
cmn	BinGOs: which is wrong	19:56
	→ drizzd joined	19:56
	→ anonymuse joined	19:56
	→ didlixForHire joined	19:56
cmn	how often does one have to say it: the data source is wrong	19:56
_ikke_	git-scm is wrong	19:56
	outdated	19:56
	cmn: Where does git-scm even get it's data?	19:57
BinGOs	I hate parsing html full stop.	19:58
cmn	the github API	19:58
_ikke_	Who doesn't?	19:58
jphillips	how do I list my remotes	19:58
	→ Synthead joined	19:58
_ikke_	git remote show	19:58
cmn	for it was broken for a bit for the git repo; and after it came back, the lib that reads from it started returning nil for some tags	19:58
jphillips	how do I list my remote with URLs?	19:59
cmn	jphillips: git remote -v	19:59
Synthead	does git stash remember changes between branches? as an example, if I do a git stash, then switch branches and do another git stash, can I go to the first branch and get the changes back?	19:59
_ikke_	Synthead: By default, git applies the diff of the staff to the current branch	20:00
cmn	the cause of that should be fixed, but I put in some code to ignore those values, but I don't think that's been deployed on the site yet (or the issue on heroku has a diferent problem that my test env)	20:00
	→ tenmilestereo joined	20:00
	← ojacobson left	20:00
	← theothe__ left	20:01
	→ staticshock joined	20:01
	← sukima left	20:01
	← Nibato left	20:01
	← OliverJAsh left	20:02
	← mikeym_ left	20:02
	← mutex left	20:02
	← Davey left	20:03
	← fuzzybunny69y left	20:03
	dpac → dpac\|away	20:04
	← exoon left	20:05
	→ acidjazz joined	20:05
acidjazz	hello everybody	20:06
	so .. i wnat to use git.. but this gig i got demands svn.. there any kind of solution for me?	20:06
greymalkin	acidjazz: git-svn	20:06
acidjazz	does it go back and forth?	20:07
	→ fulld joined	20:07
_ikke_	yes	20:07
	← theoth___ left	20:07
greymalkin	Yes, both ways;	20:07
acidjazz	can i mirror my checkins to the svn repo but then also grab anything new out of it?	20:07
	→ Nibato joined	20:07
fulld	Can I get the aha for remotes/origin/atc:style.css	20:07
_ikke_	But you have to follow the rules	20:07
acidjazz	i like both ways	20:07
fulld	*sha	20:07
	→ PaulCapestany joined	20:08
greymalkin	acidjazz: it's as simple as `git-svn pull`	20:08
_ikke_	git scn rebase	20:08
	svn*	20:08
greymalkin	Ah... yeah, sorry.	20:08
	It's been a while	20:08
	← nkr left	20:08
_ikke_	git svn dcommit and git svn rebase are basically the commands you need	20:09
	Don't create local non-ff merges	20:09
acidjazz	i see	20:09
	thanks	20:09
staticshock	man git-show has no mention of -s	20:10
gitinfo	the git-show manpage is available at http://jk.gs/git-show.html	20:10
	← DoniRS left	20:11
staticshock	how would i know where to look for it?	20:11
	→ OliverJAsh joined	20:13
	→ gasbakid joined	20:14
	← gasbakid left	20:14
	← jalama left	20:14
	← OliverJAsh left	20:15
	→ lioux joined	20:16
_ikke_	What would it do?	20:16
Nevik	it does mention it	20:16
	git show -s --format=%s v1.0.0^{commit}	20:16
	Shows the subject of the commit pointed to by the tag v1.0.0.	20:16
_ikke_	yeah, but that's only an example	20:16
Nevik	indeed	20:16
	gitinfo set mode: +v	20:16
_ikke_	It doesn't explain exactly what the -s does	20:16
	→ arietis joined	20:16
Nevik	possibly an artifact from old option?	20:17
charon	-s suppresses the patch	20:17
Nevik	yeah, it's the git-log option	20:17
	→ oakdog8 joined	20:17
	→ w09x joined	20:17
lioux	Hi, just found out that the best non-messy thing I could do was import 2 years of commits for net/ipv4/tcp.c..... Any way I can do range with cherry-pick? It's for Android AK Kernel.... I really hope the users appreciate this.....	20:17
	I don't mean cherry pick exactly. I mean what's the beast approach in such a situation	20:18
	← mmitchell left	20:18
	→ leehambley joined	20:18
Nevik	staticshock: btw, from the manpage: The command takes options applicable to the git diff-tree command to control how the changes the commit introduces are shown. This manual page describes only the most frequently used options.	20:18
	so the git-diff-tree manpage is where to look	20:19
lioux	All this to get "tcp: early retransmit"...	20:19
	→ mmitchell joined	20:19
lioux	I am dreading when I turn to "TCP Small Queues"	20:19
	lioux hates bufferbloat with a vengeance	20:19
	← etcetera left	20:19
	→ Sky[x] joined	20:20
	→ thiago joined	20:20
	← drizzd left	20:21
staticshock	Nevik: thanks, i didn't catch that	20:21
charon	lioux: newer cherry-pick supports ranges	20:21
	but it'll linearize so it might not be what you need	20:21
Nevik	staticshock: some git commands, especially porcelain ones, are basically wrappers for others, so often times they pass through many options	20:22
	← rideh left	20:22
Nevik	if there are many such options, the wrapper command's manpage doesnt re-list them all	20:22
	← hoschi left	20:22
Nevik	so if you think options are missing, look for such a reference	20:22
	← jfinn left	20:22
staticshock	thanks	20:22
Nevik	no problem ;)	20:22
	→ OliverJAsh joined	20:23
charon	in particular most commands that do something log-like support the same options rev-list options, and most commands that do something diff-like support the same git-diff options	20:23
	so e.g. git-log supports both sets :)	20:24
	and show is really a log with --no-walk -p by default	20:24
	→ daniel_- joined	20:24
lioux	Let me summarize my pain (hopes for joy l8): The last import for root/net/ipv4/tcp.c was on 2010-08-30. I need to bring it up to date to 2012-05-03. So I need all changesets relating to that file from then to now	20:25
	piece of cake	20:25
_ikke_	lioux: Do you have direct access to that repo?	20:25
	lioux: The git protocol doesn't let you download history per file	20:25
	→ etcetera joined	20:25
lioux	heh, figures	20:25
	I can write a script to grab the changeset based on date range.... that will take a while though since I don't know the git interface for perl/python/etc	20:26
_ikke_	git bundle allows you to create a file that you can import locally again	20:26
	← huoxito left	20:26
	← lavaman left	20:26
_ikke_	lioux: git doesn't really work with changesets.	20:26
lioux	_ikke_ I want to import from linux mainline	20:26
	maxibyte → zack6849\|Offline	20:27
lioux	_ikke_: sorry for the incorrect lingo, I'm new to git	20:27
_ikke_	lioux: Ok	20:27
	← Yuuhi left	20:27
_ikke_	lioux: I guess you refer to commits	20:27
lioux	okay, so a commit refer to a specific change, not a specific file then CHECK, that's what I refer	20:27
_ikke_	Do you have the linux repository local?	20:29
lioux	I cloned the android one and added the linux as remote....	20:30
_ikke_	Ok, and then?	20:30
	lioux is really lost, has only been trying git a couple days.... promisses to read a book on it	20:30
_ikke_	!book for some suggestions	20:31
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	20:31
_ikke_	lioux: What do you want to achieve now?	20:31
	→ tyler__ joined	20:31
	← HardDisk_WP left	20:31
_ikke_	ok	20:31
	I read it	20:31
	You added linux as a remote. Did you do anything after that?	20:31
	→ marcdel joined	20:32
	→ _Vi joined	20:32
lioux	I want to grab all commits containing root/net/ipv4/tcp.c from 2010-08-30 to 2012-05-03.	20:32
	_ikke_: did lots of things. :) THat's the current problem	20:32
_ikke_	define grab	20:33
lioux	I want to get clean code for Early Retransmission onto Android AK Kernel	20:33
	We've already added Codel	20:33
_ikke_	git log -- root/net/ipv/tcp.c give you the history of all commits touch that file	20:33
	But you have to be on the linux branch for that	20:34
	→ srayner joined	20:34
lioux	linux-git is the alias I gave to the remove linux mainline	20:35
_ikke_	ok	20:35
leehambley	can I somehow completely disable the prompt for username and password when using https? I'm trying to script a credentials check, but Git is doing some weirdness forcing reopening of the teminal or something, I've tried every redirection mechanism I could think of	20:35
_ikke_	have you fetched / pulled from that remote already?	20:35
	leehambley: Have you looked at credential helpers?	20:36
lioux	yup, but for other patches. This current git clone is all clean. I only have what I cloned from origin	20:36
_ikke_	lioux: ok	20:36
	Before you can do anything, you first have to fetch the history	20:36
	lioux: git fetch linux-git	20:37
	← Count_Niedar left	20:37
	← w09x left	20:37
lioux	git remote update ? is that the same?	20:37
	→ bzyx joined	20:37
leehambley	_ikke_; as far as they can find it looks like I need to use git config to set one, but I can't do anything more than a line or two	20:37
	i'm just trying to ls-remote and see what git's exit status is	20:37
	→ Bass10 joined	20:37
_ikke_	leehambley: about the same	20:37
lioux	done	20:38
_ikke_	lioux: ok	20:38
	waht branch are you currently on?	20:38
	what*	20:38
	→ Grum joined	20:38
lioux	# On branch ak-cylon-4.2.2-experimental10	20:39
	→ Niedar joined	20:39
	→ cozby joined	20:39
cozby	hi, on my remote box, I have cron job that does a git pull every 2 mins	20:40
_ikke_	That is the branch you are currently working on?	20:40
cozby	this works, but I believe avoiding a merge would be best	20:40
lioux	yes	20:40
cozby	so just a clean fetch really	20:40
	→ douglaslassance joined	20:40
cozby	however is it possible to just fetch the latest?	20:41
	no merge, I'm trying git fetch && git reset --hard && git checkout -f branch_name	20:41
leehambley	_ikke_: can't find a way to make credential helpers work	20:41
_ikke_	fetch the latest?	20:41
	cozby: checkout is not necessary after reset --hard	20:41
	← pandeiro left	20:41
cozby	_ikke_: hm I see, but the reset hard doesn't seem to have the latest commit	20:42
	→ HardDisk_WP joined	20:42
	← HardDisk_WP left	20:42
	→ HardDisk_WP joined	20:42
	← sbodo left	20:42
rawtaz	hm, seems like branch deletions are not recorded in the reflog	20:42
cozby	I have been testing this. It doesn't quite work.	20:42
_ikke_	git reset --hard <remote>/<branch_name>	20:43
	lioux: Do you know from what branch you want to get the changes?	20:43
cozby	_ikke_: hrmm that works	20:43
	thanks	20:43
	← les_sylvains left	20:45
cozby	_ikke_: so calling git reset --hard will only reset the local branch to the first commit?	20:45
	→ megha joined	20:45
_ikke_	cozby: Yeah, to HEAD to be exactly	20:45
cozby	I see.. OK thanks	20:45
	← douglaslassance left	20:45
	→ sbodo joined	20:45
	← leehambley left	20:45
cozby	_ikke_: so no fetch is required?	20:45
	→ brson joined	20:46
	← Yougi left	20:46
	← baba left	20:46
lioux	_ikke_: still here. I'm trying something like 'git log --remotes=linux-git --since=2010-08-30 --until=2012-05-03 net/ipv4/tcp.c'	20:46
_ikke_	cozby: fetch is required to download the history	20:47
	lioux: And does that give you what you want?	20:48
cozby	_ikke_: ah, so its not necessary if you're just reseting the head to the latest	20:48
_ikke_	yes	20:48
	← guampa left	20:48
	← geekbri left	20:49
	→ guampa joined	20:50
lioux	_ikke_: it seems to. Regarding cherry-pick, it refers to mainline: is it the origin or the destination?	20:50
	→ infynyxx joined	20:51
_ikke_	lioux: mainline is 1	20:52
	→ mikef joined	20:52
_ikke_	(first parent)	20:52
	→ sq-one joined	20:54
	← berserkr left	20:54
	← jebenexer left	20:54
	← vdv left	20:55
	→ macrover joined	20:55
	→ Xethron joined	20:55
	← redrocket left	20:55
	← freeed left	20:56
	← marcdel left	20:56
	← p3rror left	20:57
	← jcromartie left	20:57
	→ p3rror joined	20:57
	← jjore_ left	20:57
	→ Vile joined	20:59
	→ kogent joined	20:59
	→ w09x joined	20:59
	← Goles left	20:59
	kenperkins\|away → kenperkins	20:59
	← OliverJAsh left	21:00
	→ cebor joined	21:02
	← ddelrio1986 left	21:03
	→ arkady91 joined	21:03
	← cozby left	21:04
lioux	_ikke_: Thanks, now to bite the bullet. I'll feed the commits to cherry pick and go play with the hundreds of insuing conflicts	21:04
_ikke_	heh	21:05
lioux	Only 54 commits spanning 2 years.... what could go wrong? ;)	21:05
_ikke_	Not very much I guess :P	21:06
	→ lavaman joined	21:06
_ikke_	Good luck	21:06
	← relipse left	21:06
	→ marcdel joined	21:06
	→ relipse joined	21:07
	→ jeffreylevesque joined	21:08
	→ ErKa joined	21:09
	← relipse left	21:09
	→ relipse joined	21:09
	← Vile left	21:12
	← fulld left	21:12
	← mrwonko left	21:12
_ikke_	lioux: I'm curious to how it ends	21:13
	← d2dchat left	21:13
	← natechan left	21:13
	→ ferne joined	21:13
	← rath left	21:13
	→ RaceCondition joined	21:14
	→ jjore_ joined	21:14
	← macrover left	21:14
	→ thierryp joined	21:14
	→ Corren joined	21:15
	→ gommo joined	21:15
	← kenperkins left	21:16
	→ Watabou joined	21:16
	← Corren left	21:16
	← otc left	21:16
	← RaceCondition left	21:16
	← maxloe left	21:16
	← ferne left	21:18
	← v0n left	21:18
	← relipse left	21:20
	→ relipse joined	21:21
	← etcetera left	21:22
	← slaren left	21:23
	→ etcetera joined	21:23
rawtaz	question: in http://grab.by/kKKq (im happy to provide console log instead upon request), how come the three commits on the yellow branch is shown on that branch instead of on the red branch? i ask because those commits were done on the upstream/master branch (by another guy), not on the origin's master branch (which is my fork)?	21:23
	what determines on which branch they are shown in the log, so to speak. the fact that it shows them on the yellow branch instead of where they were done confuse me	21:24
	→ kenperkins joined	21:24
	← mezner left	21:24
	→ mezner joined	21:25
	← kevwil left	21:25
_ikke_	rawtaz: Where do you expect them to be	21:27
	?	21:27
rawtaz	is it simply that when building the graph, commits are shown farthest to the right (though i think that's tweakable), and this is the reason theyre shown on the yellow branch (since it has them, too)?	21:27
	→ StrongOrder joined	21:27
rawtaz	_ikke_: thats a very good question. ive been thinking about it. i thikn the most definite info would be to show them on each of the red and yellow branch. cuz both of them have those commits/the contents of the commits	21:27
	i guess theyre simply put on one side when theire in multiple branches, to make things less messy	21:28
	s/on/to/	21:29
	→ Rix joined	21:29
	← thierryp left	21:29
_ikke_	rawtaz: It has been a different branch	21:30
rawtaz	this would make sense regarding the 4685b5e commit which is shown on the red branch, since less-origin/master hasnt been updated to that commit yet	21:30
_ikke_	rawtaz: Because of the merge commit, it sees that it was a seperate branch	21:30
	← danman left	21:30
_ikke_	and showing it accordingly	21:30
	← kadoban left	21:30
rawtaz	_ikke_: which merge commit, the 7a75 one or the 11a1 commit?	21:30
_ikke_	b1bc5c1	21:30
	→ jaseemabid joined	21:31
rawtaz	ah	21:31
	← p3rror left	21:31
_ikke_	You updated your origin/master branch to that branch, but upstream/master has moved on in the mean time	21:32
rawtaz	hmm	21:33
	→ kevwil joined	21:34
rawtaz	sorry. not getting this shit :(	21:34
_ikke_	rawtaz: In git, a branch is just a pointer to a commit	21:34
rawtaz	i have never updated my origin/master. the only thing i have done with it is forking the upstream repository. that is all	21:34
	yeah	21:34
	→ BadDesign joined	21:35
rawtaz	now that im looking at this im even more confused over that the in the merge commit you mentioned (b1bc) it looks like my origin/master (the fork) has been merge into upstream/master :D	21:36
_ikke_	rawtaz: Note that this is merely a representation	21:36
	← rbmk left	21:36
rawtaz	yeah..	21:36
_ikke_	You don't have a local master branch?	21:37
rawtaz	i do	21:38
	its in the blue line, at the very bottom	21:38
_ikke_	Have you called your remotes less-origin and less-upstream?	21:38
rawtaz	here's the entire history: http://grab.by/kKNe	21:39
	if you are asking if i named them that, then yes. i added them manually	21:39
	→ Vile joined	21:40
_ikke_	Ok	21:40
rawtaz	this is my local git repo, with the master. then i created a yii-less branch, checked it out, and then used git subtree to add in my fork less-origin/master	21:40
	← gommo left	21:40
_ikke_	rawtaz: You have merged two seperate repos together"?	21:40
rawtaz	the add was done in 11a1. then i also pulled from upstream in 7a75	21:40
_ikke_	I see two root commits	21:40
rawtaz	i guess technically the answer is yes. moment	21:41
	→ gommo joined	21:41
_ikke_	What I see:	21:41
	Upstream made a whole set of commits	21:42
rawtaz	here are the commands i have run: http://pastebin.com/27cNqp0y	21:42
	← fobban left	21:42
rawtaz	except for the very last test commit/change i did on yii-less	21:42
_ikke_	Ahh, subtrees	21:42
rawtaz	yea, the contrib	21:42
_ikke_	Ok, that explains	21:42
rawtaz	i believe i have done that part right though :)	21:42
	← oakdog8 left	21:42
_ikke_	I believe you (haven	21:42
rawtaz	yeah. i can understand you were slightly puzzled about what was going on	21:42
_ikke_	haven't used them yet)	21:43
rawtaz	havent used what? :)	21:43
	→ redrocket joined	21:43
_ikke_	subtree merges	21:43
	Well, the contrib subtree command	21:43
rawtaz	yes. i have not used subtree merge, i have only used the contrib "git subtree"	21:43
	to add in my form from github	21:43
	← anonymuse left	21:44
rawtaz	fork*	21:44
_ikke_	The history looks a bit messed up	21:45
	I wonder why you used subtrees for this	21:45
	← gommo left	21:45
	→ p3rror joined	21:45
rawtaz	basically because i have red both pros and cons about submodules and got the feeling that subtree was the slightly newer kid on the block and that many people were moving to using that instead of submodules	21:45
_ikke_	You basically forked the project, and then upstream made 4 more commits	21:45
	→ leehambley joined	21:46
_ikke_	rawtaz: Yes, but I don't really see how this would be a submodule / subproject	21:46
	→ Radegast joined	21:46
rawtaz	exactly. after my fork in fc46, upstream made four commits. the first three are shown on my fork in this history, and i am curious why its shown there instead of in the upstream tree/line. but i guess itsjust a cosmetic design choice (putting it on the far side)	21:46
	← jaseemabid left	21:47
_ikke_	Yeah, don't attach too much value on how it gets rendered	21:47
	Note that it doesn't really know what the main branch is	21:47
	← buibex left	21:47
rawtaz	_ikke_: i think my naming the remote fork "less-origin" might have added confusion here. the local repository has nothing to do with the fork on github. the fork on github is of the upstream project, and is meant to be a third party lib that i want to have in a subfolder in my local project/repo	21:48
_ikke_	oh ok	21:48
rawtaz	does that make sense?	21:48
	sorry for the confusion :<	21:48
_ikke_	I find 11a1fab an odd commit	21:49
Radegast	How can I get the most recent tag from all branches? "git describe --all" returns "heads/master"	21:49
	→ buibex joined	21:49
_ikke_	though, that can be due to my lack of understanding what subtree actually does	21:49
rawtaz	_ikke_: i think that subtree is simpler than you might think, but indeed one has to read its man page (https://github.com/git/git/blob/master/contrib/subtree/git-subtree.txt) to get the bigger picture	21:50
	← leehambley left	21:50
rawtaz	_ikke_: that commit, 11a1, is the one where subtree is used to add in the other repo	21:50
	just FYI	21:50
_ikke_	ok	21:50
	← Vile left	21:50
_ikke_	That makes sense then	21:51
	→ gommo joined	21:51
	_ikke_ wonders how usefull this approach is	21:51
_ikke_	(not regarding to you, but subtrees in general)	21:51
rawtaz	yeah	21:52
	← StrongOrder left	21:52
rawtaz	well its like everything else, has its pros and cons	21:52
	one con, but potentially pro for some i guess, is that the other repos history will be part of your repos history	21:52
_ikke_	Yeah, but I wonder why one would want to merge the histories of two unrelated projects together	21:52
rawtaz	mm	21:52
	→ sluther joined	21:53
	← irqq left	21:53
_ikke_	7a75407 is where you did another merge?	21:53
	Yeah, that's the last pull, rihgt?	21:53
	← phiwo left	21:54
	← sluther left	21:54
	→ phiwo joined	21:54
rawtaz	yeah. in http://pastebin.com/27cNqp0y line 2 is 11a1 and line 6 is 7a75	21:54
	→ mezgani joined	21:54
	← j^2 left	21:54
	→ pastime joined	21:54
rawtaz	(beats me how can be arsed to write the "full short" has =D)	21:54
	→ Corren joined	21:54
	← Corren left	21:54
	→ sluther joined	21:54
_ikke_	The last sentence does not compute :P	21:55
rawtaz	oh i meant hash at the end :D	21:55
_ikke_	But I can see how you can be confused about that branch	21:55
	← breakingthings left	21:55
	← ScottE left	21:55
rawtaz	_ikke_: to just sum things up; do you see what i mean about "the commit is in the yellow branch instead of the red one!"?	21:56
_ikke_	It's probably because less-origin/master is still pointing to it	21:56
rawtaz	with the consideration that the commit in question (the three ew talked about) were not done in my fork but in upstream's repo	21:56
	← p3rror left	21:56
rawtaz	aha	21:56
_ikke_	This has just to do with how those histories are rendered	21:56
rawtaz	so you mean that if origin/master wouldnt point to b4f8 anymore this might very well not look like this	21:57
	yeah	21:57
_ikke_	Well, like I said earlier, it doesn't really know what the main branch is	21:57
	It sees history as diverged, and it choose to show those 3 commits as seperate	21:57
rawtaz	can you elaborate on what you mean by "main" branch?	21:57
	← mezner left	21:58
	← kenperkins left	21:58
_ikke_	Well, for example, in that part of history, it looks like the purple line is the main branch, and the yellow line diverged and got merged again	21:58
rawtaz	yeah	21:58
	exactly	21:58
_ikke_	But that doesn't really mean anything	21:58
rawtaz	that puzzles me a bit as well, why the yellow one looks merged to the red..	21:58
_ikke_	rawtaz: What I suspect is that upstream was on the purple branch, when he merged the commits from the yellow branch in	21:59
	← xaocon\|away left	21:59
_ikke_	making the purple branch the first parent, and the yellow branch second	21:59
	→ jaseemabid joined	22:00
_ikke_	git log -1 --format=fuller b1bc5c1	22:00
rawtaz	_ikke_: but the yellow branch is my origin/master, isnt it? if yes, then the only relation that and the upstream master has is that the yellow is the fork of the red one, but that was way before the three commits	22:00
	does that make sense?	22:01
_ikke_	rawtaz: As it looks from here, those three commits were already there when you forked the repo	22:01
rawtaz	by the three commits you mean 98, 4a and b4, right?	22:01
_ikke_	yes	22:01
rawtaz	dang	22:01
	yes	22:01
	that is accurate	22:01
	← xxHarlequin left	22:01
rawtaz	i think i did not include that fact in my thinking. you are right. i only forked once and havent touched it since	22:01
	i was for some reasin thinking i forked on fc46	22:02
	← koltroll left	22:02
rawtaz	but i guess it just starts from that point due to the tag. perhaps i forked off that tag or something (not sure how that works, YET)	22:02
_ikke_	rawtaz: What program do you use btw?	22:02
rawtaz	_ikke_: the screenshot is from SourceTree	22:02
_ikke_	ok	22:02
rawtaz	but i am paralleling with the console	22:02
_ikke_	right	22:03
rawtaz	this is just for kicks :-)=	22:03
_ikke_	Can you show the output of that command I gave?	22:03
	→ anjen joined	22:03
rawtaz	indeed. terribly sorry i missed it	22:03
	→ koltroll joined	22:04
	→ JSharpe joined	22:04
	← TeckniX left	22:04
	→ JesseC joined	22:04
	← vaanz left	22:05
rawtaz	_ikke_: updated http://pastebin.com/Ztv3dhZh	22:05
	← mviitanen left	22:05
_ikke_	yeah, confirmed	22:05
	← etcetera left	22:05
_ikke_	The purple branch is indeed mainline	22:05
	Those three commits happened on some other branch that got merged in again	22:06
rawtaz	when you say merged in again, to where do you mean?	22:06
_ikke_	b1bc5c1	22:06
	well, the merge commits says it merged master back in	22:07
	← chuckharmston left	22:07
rawtaz	ah right. since i was on the yii-less branch	22:07
	no, i take that back	22:07
	← PaulCapestany left	22:07
_ikke_	Is Crisu83 upstream, or another branch?	22:07
	repo*	22:07
rawtaz	yeah Crisu83 is the guy with the upstream repo, it is a repo	22:08
_ikke_	Is it the repo you forked from/	22:08
	?	22:08
rawtaz	yes. forked on github	22:09
	https://github.com/rawtaz/yii-less is my fork	22:09
	→ b_jonas joined	22:09
_ikke_	I wonder how that branch got there	22:09
b_jonas	stupid question. are mercurial's pull and fetch commands similar to git fetch and git pull in that order?	22:10
rawtaz	so, are you saying that when i forked, i somehow forked another branch from crisu83's repo than the master one?	22:10
_ikke_	b_jonas: Yes	22:10
b_jonas	crazy	22:10
_ikke_	Yeah	22:10
b_jonas	thanks	22:10
	← hwrd\|work left	22:11
rawtaz	_ikke_: was that Yeah for me	22:11
	+?	22:11
	← cebor left	22:11
_ikke_	no, for b_jonas	22:11
rawtaz	ok :)	22:11
	→ arand joined	22:11
	→ PaulCapestany joined	22:11
	→ shawnj\|away joined	22:12
_ikke_	rawtaz: What I guess what happened is this: You forked the repo at commit b4 (the master branch).	22:12
	→ hwrd\|work joined	22:12
rawtaz	agreed	22:12
	at that time, was there just one branch in the upstream?	22:12
_ikke_	upstream created another branch based on the tag	22:12
rawtaz	after i forked?	22:13
_ikke_	rawtaz: I don't know	22:13
	← AnimeCYC left	22:13
_ikke_	It doesn't really matter	22:13
	← infynyxx left	22:13
_ikke_	But then, instead of merging that branch back to master, it merged master into that branch	22:13
	That's why the branch is merged like that	22:13
	→ cebor joined	22:14
rawtaz	aha. i think i see what you are saying	22:14
	→ tplaner joined	22:14
rawtaz	lemme turn the wheels another rotation :)	22:14
	← cannonball left	22:15
JesseC	How do you specify a username for a remote clone?	22:15
_ikke_	JesseC: What protocol?	22:15
	:?	22:15
JesseC	I'm using beanstalkapp, can't figure it out for the life of me.	22:15
rawtaz	depends on the transport youre using	22:15
JesseC	I would assume just https, I don't think they offer ssh	22:15
_ikke_	JesseC: Doesn't it ask for a username?	22:16
	← shawnj\|away left	22:16
	← tPl0ch left	22:16
	→ shawnj joined	22:16
	← shawnj left	22:16
JesseC	nope, just says that unauthorized if I put nothing, and I tried username:/repository.path and it prompted for a password, but then said it couldn't "resolve" the name	22:17
	← ThePawnBreak left	22:17
JesseC	https://username:/repo.path	22:17
arand	When I'm in the process of filter-branch --tree-filter what is the status of the index? does it simply stay at the state of the ref I'm rewriting (e.g. master)? If I use --index-filter as well, does that happen in parallel, and can the other refer to the other, somehow?	22:17
	→ shawnj joined	22:17
_ikke_	JesseC: What version of git are you using?	22:17
JesseC	_ikke_ 1.7.1	22:18
_ikke_	arand: I'm not sure, but I would expect that the index is updated before the work tree	22:18
	JesseC: You'd need to update to a newer version, where things got fixed	22:18
rawtaz	JesseC: for a HTTML url the format is usually http://username@domain.com/...	22:18
	→ Vile joined	22:18
	→ vdv joined	22:19
	← worstadmin left	22:19
	← mmitchell left	22:20
	→ akresic joined	22:20
rawtaz	_ikke_: well, what you said makes sense. it looks possible that this is what has been done and why it looks like this. the b1bc would reference the parents 4685 and b4f8, and since the latter is in my fork it is shown as connected in this graph	22:21
	→ mmitchell joined	22:21
	← mezgani left	22:21
	← akresic left	22:22
	→ jlg13 joined	22:22
rawtaz	_ikke_: i was looking to see if one coule see branch changes on github but i cant find anything like that. the upstream currently has one branch and it is named master, but i guess it is possible that he renamed that red branch from something to master	22:22
_ikke_	rawtaz: And since 46 is mentioned first, it tells us that the author was on that branch when her merged master in (+ the commit message saying he merged master)	22:22
	rawtaz: He probably had a branch, merged master later in, and removed that branch again	22:22
	rawtaz: Basically the opposite of what one would expect normally	22:23
rawtaz	_ikke_: isnt the reason that 46 is before b1 in the list (i mean below) simply that 46 contains a commit that was made after b4 but before b1 ?	22:24
	← shawnj left	22:24
	→ p3rror joined	22:24
	← thl0 left	22:24
	← p3rror left	22:24
	→ thl0 joined	22:25
_ikke_	rawtaz: no	22:25
	← jn_ left	22:25
	← mmitchell left	22:25
_ikke_	rawtaz: when merging, git puts the current branch as the first parent, and the branch getting merged in as seccond	22:26
	second*	22:26
rawtaz	makes sense	22:26
JesseC	Anyone familiar with beanstalkapp as a git repository host?	22:27
_ikke_	nope	22:27
	← nkryptic left	22:27
Radegast	what does it mean when "git describe --all" returns "heads/master" instead of a tag?	22:27
	← gommo left	22:27
	← thl0 left	22:27
JesseC	I think I managed to get the syntax correct for logging in, but now it says "did you run git update-server-info" on the server	22:27
rawtaz	_ikke_: can you clarify; 46 is is first of <what collection of stuff>?	22:28
	→ thl0 joined	22:28
	→ gommo joined	22:28
rawtaz	_ikke_: sorry, i take that back. nvm	22:28
cmn	Radegast: it means you're at the tip of master	22:28
	→ jfinn joined	22:29
cmn	Radegast: what's confusing/wrong about that output?	22:29
rawtaz	_ikke_: i just dont see how 46 would ever NOT be before b1, since 46 is a regular commit and was made BEFORE the next action, being the merging, took place. why would it not come before the merging in the history, as long as the merging happened after the 46 commit?	22:30
	← Vile left	22:30
_ikke_	rawtaz: Merge doesn't look at the order of when the commits happend	22:30
Radegast	cmn: it should return the latest tag if I understand docs correctly. I just want to get the latest tag from all branches	22:31
_ikke_	rawtaz: It looks at what commit you are when merging, and what branch you are merging in	22:31
rawtaz	hm i thought it did. in the sense that it "replays" them, which presumably would be done in order	22:31
	← arietis left	22:31
	← sgambino left	22:31
_ikke_	rawtaz: Yes, but that has nothing to do with the order of which the parents appear in the commit	22:31
	← R1KRDO left	22:32
rawtaz	i guess given that this is a recursive merge	22:32
_ikke_	rawtaz: If he was on master, and merged the ohter branch in b1 would appear first	22:32
	← thl0 left	22:32
	← gommo left	22:32
cmn	Radegast: read what --all does again	22:32
	it looks through all refs	22:32
	→ raspy_freenode joined	22:32
cmn	and that branch is a better fit	22:32
	what do you mean latest tag from all branches?	22:33
	→ kenperkins joined	22:33
rawtaz	_ikke_: im sorry im not getting this.. surely the commit we now see in 46, which is a regular file edit and no merge in any way, would not come after merging the red and yellow branch, except if the author did that file edit commit (46) after the merging one?	22:34
	i just dont understand how it could	22:34
	probably misunderstood that	22:34
	← Grum left	22:35
	← cebor left	22:36
_ikke_	rawtaz: I think we are talking about different things here	22:36
Radegast	cmn: I have a tag in the master (2.70) and a newer tag in another branch (2.71). git describe returns 2.70 while I want it to return 2.71	22:37
_ikke_	rawtaz: I'm talking about http://pastebin.com/Ztv3dhZh line 48	22:37
cmn	first of all, tags don't belong to branches	22:37
	how are you calling describe? what are you asking it to tell you?	22:37
	→ Grum joined	22:38
	← koltroll left	22:38
	← Xethron left	22:39
	→ pppaul joined	22:39
rawtaz	_ikke_: right! you are just discussing a key part that helps us identify what the author has done. in this case that the first commit named in a merge message like that one is an indicator of which branch the author was on when doing the merge. in this case he was on the branch that has that commit.	22:39
	_ikke_: that makes it all clear	22:39
	← srayner left	22:39
	← kevwil left	22:39
_ikke_	rawtaz: correct	22:39
	← axl_ left	22:39
rawtaz	awesome	22:40
	smart	22:40
	ok	22:40
	scrolls up to make sure he understands the two other things you said	22:40
	→ shalokshalom joined	22:40
	← cenuij left	22:41
	← jlg13 left	22:41
	→ gommo joined	22:42
	← gommo left	22:43
	← jjore_ left	22:43
	← jjore left	22:43
	→ gommo joined	22:43
rawtaz	_ikke_: looking at the commit messages it looks like the branch he created was for adding composer. makes sense.	22:43
Radegast	cmn: for me it was logical to do the opposite of what everyone else does. I kept a stable branch and continued with unstable version in master. Now I see it was not a good idea. Anyway, I found this, which does what I want: git describe --tags $(git rev-list --tags --max-count=1)	22:43
rawtaz	_ikke_: also, after having done that merge, both master and the other branch was pointing to the same commit (b1), correct?	22:44
cmn	Radegast: what are you actually trying to achieve?	22:44
	list all tags?	22:44
Radegast	cmn: find the latest tag	22:45
	← relipse left	22:45
cmn	why with describe then?	22:46
	that's meant to give you a version to refer to	22:46
	→ kuyan joined	22:47
Radegast	cmn: what's the right way to do it?	22:47
	← WhereIsMySpoon_ left	22:47
cmn	first you need to define what latest means	22:47
	← buibex left	22:47
	← gommo left	22:48
Radegast	cmn: chronologically	22:48
	← staticshock left	22:49
	→ samholmes joined	22:49
samholmes	How do I just pull the contents of the latest comment from a repository?	22:49
cmn	chronologically of what?	22:50
JesseC	_ikke_: compiling the latest version of git 1.8.2 fixed all of the issues I was having, thanks.	22:50
samholmes	I'm trying to just get these files from github https://github.com/jethrolarson/placeholder-augment	22:50
cmn	the annotated tag, the tagged object...?	22:50
samholmes	Into my project's public lib scripts directory.	22:50
	So, I just need to be able to git clone without cloning the repo.	22:50
cmn	are you going to have later tags that don't fit into version numbers that you'll want it to show?	22:50
	→ netiul joined	22:50
	← sluther left	22:50
	← undecim__ left	22:52
	→ undecim__ joined	22:52
_ikke_	JesseC: yw	22:53
kuyan	Hi guys, I need some help. So, I have files in my repository that I for some reason cannot delete.	22:53
	This is a newly-cloned repository.	22:53
	← lifelike left	22:53
kuyan	Uh, instead of delete, I mean remove from git.	22:54
	I tried `git checkout -- <file>`, `git clean -df`, `git checkout <file>`, `git reset --hard`, `git reset --hard origin/master`, all that.	22:54
	It won't go away.	22:54
cmn	man git rm --cached	22:54
gitinfo	the git-rm manpage is available at http://jk.gs/git-rm.html	22:54
cmn	why would any of those delete a file?	22:54
kuyan	I meant discard the changes in the working directory. Sorry >_>	22:55
	Not thinking today.	22:55
	I just cloned the repository and the files are showing as modified, even though I cloned it less than a second ago.	22:55
	← BadDesign left	22:56
	→ taylorlapeyre joined	22:56
cmn	on OSX there used to be some bug that caused that	22:56
kuyan	Yeah, I'm on OS X.	22:56
	→ jason237 joined	22:56
cmn	or it could be core.autocrlf set to true on unix ssytems	22:56
kuyan	The repositories in question are github.com/kennethreitz/python-guide and github.com/kuyan/python-guide	22:57
	→ Vile joined	22:57
kuyan	both core.autocrlf and core.safecrlf are unset	22:57
rawtaz	_ikke_: i think that everything i was wondering about this (starting with why the three commits were shown on the yellow instead of red branch) have been answered. thank you so much ♥	22:57
	→ jjore_ joined	22:57
	→ jjore joined	22:57
kuyan	It's probably OS X - I had an issue completely like this with Homebrew just a few days ago: https://github.com/mxcl/homebrew/issues/13373	22:57
	← bzyx left	22:58
	→ fuzzybunny69y joined	22:59
JesseC	I have a folder that I renamed, but git shows the old folder and the new folder when I do a clone. Will it hurt anything to do "git rm -r old_directory" and then commit my change?	22:59
	← didlixForHire left	23:00
	→ fisted_ joined	23:01
	→ gommo joined	23:01
	← compojoom left	23:02
	→ kmels__ joined	23:03
	→ etcetera joined	23:03
	← puchrojo left	23:04
	← fisted left	23:04
	fisted_ → fisted	23:04
kuyan	cmn: do you have anything that you could link me about that bug on osx?	23:04
	← whowantstolivef1 left	23:05
cmn	I don't know more about the bug; just that it was a race condition; the git list has more information about it; or upgrade	23:07
kuyan	cmn: okay, thanks!	23:07
	→ Sigma joined	23:08
	← Vile left	23:08
	← Rix left	23:08
	← lonewulf` left	23:09
	→ lonewulf` joined	23:09
	← marhaban left	23:11
	← trungqtran left	23:12
	→ codon joined	23:13
	← sebuba left	23:13
	← sq-one left	23:14
	→ jcromartie joined	23:14
	← jcromartie left	23:14
	← fuzzybunny69y left	23:15
	← dfr\|work left	23:15
	← kuyan left	23:17
	→ tanath joined	23:17
	← longbeach left	23:19
	← lacrymology left	23:20
	← vdv left	23:21
	← pppaul left	23:21
	→ dreiss joined	23:21
	← worm_ left	23:21
	→ phantomcircuit joined	23:21
	← kristofers left	23:21
	→ fuzzybunny69y joined	23:21
	→ lacrymology joined	23:22
	→ infynyxx joined	23:23
	→ cebor joined	23:23
	→ cenuij joined	23:23
	→ kadoban joined	23:23
	← m0ltar left	23:24
	→ monk3y joined	23:25
	← scelis left	23:26
	← sbodo left	23:27
	→ kmels_ joined	23:28
	→ pppaul joined	23:30
	← pppaul left	23:30
	→ pppaul joined	23:30
	← jason237 left	23:31
	← kmels__ left	23:32
	← primitive left	23:33
	→ Vile joined	23:34
	← anjen left	23:35
	← TheFuzzball left	23:37
	→ Jojomofoman joined	23:37
	→ warthog9 joined	23:40
	→ JaredMiller joined	23:40
	← Joe5 left	23:40
	← pantsman left	23:43
	→ h0bbit joined	23:45
	← Wulf left	23:45
	← Vile left	23:46
	← thiago left	23:46
	→ leehambley joined	23:46
	← jaseemabid left	23:47
	← warthog9 left	23:47
	→ jalama joined	23:49
	→ vooor joined	23:51
	← leehambley left	23:51
	← vooor left	23:51
	→ SBoudrias joined	23:51
	→ huoxito joined	23:51
	← JSharpe left	23:52
	← jalama left	23:53
	← NiteRain left	23:53
	← erichynds left	23:54
	→ PlantMan joined	23:55
	← h0bbit left	23:55
jasom	I'm importing a subversion repository to git; any suggestion for how to represent deleted branches?	23:55
	jasom is leaning towards something like refs/deleted-branches/branch-name	23:56
jasom	maybe with an @revision-number for good measure, as there might be duplicate branch names otherwise	23:57
	← gustavnils left	23:57
	→ macrover joined	23:57
	← rakl left	23:59

Logs Search ←Prev date Next date→

Channels Documentation