IRCloggy #git 2014-01-14

Provider of IRC logs since 2005.

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

	2014-01-14
	→ Kayra joined	00:00
	→ thomas_b joined	00:00
	← revyd left	00:02
	← thomas_b left	00:04
	← leeN left	00:05
	← SeptDePique left	00:05
	← becks__ left	00:05
osse	Just got an email from tip4commit. Is that a real thing? I got it due to a commit in git.git	00:06
	→ thomas_b joined	00:06
	← David left	00:06
	→ lazers joined	00:07
	← khmarbaise left	00:07
	→ psyl0n joined	00:07
rawtaz	i have a feeling, based on `git status` after doing `git add .`, that the `git add .` added files that are in a folder where i have a .gitignore specifying to ignore * and !.gitignore - should `git add .` really add files that should be ignored?	00:08
	← unreal left	00:09
	→ momomomomo joined	00:09
	→ teran joined	00:09
	← rednaks left	00:09
	← infynyxx left	00:10
rawtaz	"Ignored files reached by directory recursion or filename globbing performed by Git (quote your globs before the shell) will be silently ignored." - i interpret that as files that according to .gitignore should be ignored would not be added to the index by a `git add .`	00:10
	← momomomomo left	00:10
	← petey left	00:11
endiruna	how do i merge the history of two repositories. say that one of them has x->y and the second has y->z, I would like to have a unique repository that has x->y->z	00:11
	→ unreal joined	00:11
	→ gusblan joined	00:11
	← FortySix2 left	00:12
rawtaz	hm ok i am probably misreading which files were actually added..	00:12
	→ Vivekananda joined	00:12
	← Gavrisimo left	00:13
	← amcsi left	00:13
	← sb- left	00:13
	→ strictlyb joined	00:13
	→ LuxuryMode joined	00:13
	← thomas_b left	00:14
	→ becks__ joined	00:14
	→ jayallen_ joined	00:15
	← gnu-man_ left	00:15
	→ flagon joined	00:15
	← plasticle left	00:15
	→ plasticle joined	00:15
	← jayallen left	00:15
	jayallen_ → jayallen	00:15
	→ ciampix joined	00:16
	→ petey joined	00:16
	→ Watabou joined	00:16
	→ Noldorin joined	00:17
	← becks__ left	00:17
	← senj left	00:18
	← Destos left	00:18
	→ johnkevinmbasco joined	00:19
	← ketas left	00:19
	→ BlueOrg joined	00:19
	→ dyng joined	00:19
	← BlueOrg left	00:19
	← ShalokShalom left	00:19
	→ BlueOrg joined	00:20
	← BlueOrg left	00:20
	← decotii left	00:20
	← psyl0n left	00:20
	→ BlueOrg joined	00:20
	← ciampix left	00:21
	← Watabou left	00:21
	→ mklappstuhl joined	00:22
	← mitz left	00:22
	← BlueOrg left	00:23
	→ danecando joined	00:23
	→ mitz joined	00:23
	← chessguy left	00:23
	→ acharlieh joined	00:23
	← unreal left	00:23
	→ CKyle joined	00:24
	← mklappstuhl left	00:25
	← marcjs left	00:25
	← dyng left	00:25
	→ BlueOrg joined	00:26
	→ dyng joined	00:26
	→ dyng_ joined	00:27
	← dyng left	00:27
	→ guardianx joined	00:27
	← boombatower left	00:27
	← BlueOrg left	00:27
jrnieder	osse: probably perf pulled another batch of changes from git	00:28
	← johnkevinmbasco left	00:28
jrnieder	oh, no	00:28
	I was thinking of linux-tip	00:28
	→ BlueOrg joined	00:28
	← BlueOrg left	00:28
jrnieder	I don't know what tip4commit is. Filed it as spam :)	00:28
	→ explodes joined	00:28
	→ BlueOrg joined	00:28
	← leotello left	00:29
	→ becks__ joined	00:29
	← LuxuryMode left	00:30
	→ Destos joined	00:31
	← dyng_ left	00:31
	→ nZac joined	00:32
	→ dyng joined	00:32
	← becks__ left	00:32
	→ hays_ joined	00:33
	← hays left	00:33
	→ becks__ joined	00:35
	← dyng left	00:35
	→ dyng joined	00:35
	← nZac left	00:36
	← dyng left	00:36
	→ psyl0n joined	00:36
	← oskarkv left	00:36
	→ dyng joined	00:37
	← dyng left	00:37
	→ dyng joined	00:37
	← becks__ left	00:38
	← Argure left	00:38
	→ becks__ joined	00:39
	← gusblan left	00:39
	→ Argure joined	00:39
	← sohail left	00:40
	← King_Hual left	00:40
	← ErKa left	00:41
	→ marhaban joined	00:42
	← hyperair left	00:42
	→ harish joined	00:44
	→ revyd joined	00:45
	→ JerryT joined	00:45
	→ LucasKA joined	00:45
	← CKyle left	00:46
	← Destos left	00:46
	← FatDarrel left	00:46
	← becks__ left	00:46
	→ LuxuryMode joined	00:47
	→ becks__ joined	00:47
	→ fphilipe joined	00:47
	← hays_ left	00:48
	zz_karupanerura → karupanerura	00:49
	→ hays joined	00:49
	← Balliad left	00:49
	← dreiss left	00:50
	← blackmesa left	00:50
	→ Spudster joined	00:50
	← becks__ left	00:52
	→ ciampix joined	00:52
	← fphilipe left	00:52
	← psyl0n left	00:52
	← beastd left	00:52
	← rafaqueque left	00:53
	→ nutron joined	00:53
	← DrNick left	00:54
	← SocalNick left	00:54
	→ hays_ joined	00:54
	← hays left	00:55
	→ gnu-man joined	00:55
	→ gusblan joined	00:56
	← gusblan left	00:57
	← devoldmx_ left	00:57
	→ FatDarrel joined	00:57
	→ YoMomma_ joined	00:58
	→ David joined	00:58
	→ gusblan joined	00:58
jrnieder	osse: http://blog.anonymousads.com/2013/10/tip4commit-first-bitcoin-related.html	01:00
	← Argure left	01:00
	→ dosequis_ joined	01:01
	jrnieder is not thrilled with it. Creates a perverse incentive to break a change into as many commits as possible.	01:01
	→ Argure joined	01:01
	← tobias__ left	01:02
	← gusblan left	01:02
	← imachuchu left	01:03
	← harish left	01:04
	→ ojacobson joined	01:06
	← plasticle left	01:08
	→ skorgon_ joined	01:08
	← jbarker left	01:09
	← ciampix left	01:09
	← hays_ left	01:10
	← lyang0 left	01:11
	→ DrNick joined	01:11
	→ hays joined	01:11
	← endiruna left	01:11
	← mlimaloureiro left	01:11
	→ lyang0 joined	01:12
	← mizu_no_oto left	01:13
	← petey left	01:13
	→ Happzz joined	01:15
milki	one commit per character	01:15
	and an extra commit per commit that just says WIP	01:15
ojacobson	--allow-empty hides a multitude of sins	01:15
milki	:P	01:15
	→ Nizumzen joined	01:16
	→ hyperair joined	01:16
	→ mlimaloureiro joined	01:16
	← halu left	01:18
	← skorgon_ left	01:18
	→ psyl0n joined	01:19
	← psyl0n left	01:20
	→ psyl0n joined	01:20
jrnieder	also, gah, I wish I knew which mailing list the site I linked to scraped its content from :/	01:21
	← marhaban left	01:21
	→ tummy_ joined	01:21
rawtaz	which site?	01:22
jrnieder	the link I sent to osse	01:22
rawtaz	hmm	01:23
	← DrShoggoth left	01:23
jrnieder	anyway, never mind	01:23
	← FatDarrel left	01:23
jrnieder	I saw 'ads' and thought 'sleezy part of the internet' :)	01:23
	→ gusblan joined	01:23
jrnieder	from a little more searching I think I misread	01:24
	→ harish joined	01:24
	← gusblan left	01:25
	→ gusblan joined	01:26
	→ unreal joined	01:26
	→ Dougie187 joined	01:27
	→ simesy joined	01:28
	← YoMomma_ left	01:28
	← dyng left	01:29
	→ weiyang joined	01:30
	→ dyng joined	01:30
	← simesy left	01:30
	→ chrisTech joined	01:33
	← maestrojed left	01:33
chrisTech	How do I view/edit content of previously committed files? Or, how do I get the previous files back?	01:33
	→ Watabou joined	01:34
ojacobson	chrisTech: depends on why you want them :)	01:35
	← dyng left	01:35
	→ diegoviola joined	01:35
quup	chrisTech: git checkout a43hb13hashthingy	01:36
	chrisTech: http://git-scm.com/book/en/Git-Basics-Viewing-the-Commit-History	01:36
	→ dyng joined	01:38
chrisTech	ojacobson, I just want to be able to copy it.	01:38
	copy the content, that is	01:39
ojacobson	quup's basically covered it, then	01:39
	← MistuhKurtz left	01:40
	→ mizu_no_oto joined	01:40
	← mue_ left	01:41
	→ ewnd9 joined	01:42
chrisTech	That doesn't like reverse the updated file, does it?	01:42
	→ luc_ joined	01:42
chrisTech	Like update it to the previously committed one?	01:43
	← akiva left	01:43
	→ echo-area joined	01:44
	→ plasticle joined	01:45
	→ Annak joined	01:45
	gitinfo set mode: +v	01:45
	← Kayra left	01:45
	→ ErKa joined	01:45
	→ dohdoh joined	01:46
	← pen left	01:46
	← Diemuzi left	01:47
	→ jbarker joined	01:47
	Annak → Diemuzi	01:48
ojacobson	it updates the work tree to <sha1>, and points HEAD ("the current commit") at <sha1>	01:48
	← dohdoh left	01:48
	← chrisTech left	01:49
	→ moondoggy joined	01:49
	← moondoggy left	01:50
	→ moondoggy joined	01:50
	← namita left	01:50
	← bwreilly left	01:51
	← jbarker left	01:52
	← explodes left	01:52
	← Gamecubic left	01:52
	← phipes left	01:53
	← vipzrx left	01:54
	← flagon left	01:55
	← jayallen left	01:55
	→ Gamecubic joined	01:57
	→ ciampix joined	01:58
	← Diemuzi left	02:00
	→ phipes joined	02:00
	→ bwreilly joined	02:00
	→ Suprano joined	02:01
	← Marenz left	02:01
	→ simesy joined	02:03
	← ciampix left	02:03
rawtaz	ok soo.. when rebase has paused due to a conflict, and you edited the file with a conflict in it and then saved it, is there a way to revert that editing you just did, so that the file goest back to what rebase left it as when it asked you to resolve the conflict?	02:04
	← ruinze left	02:04
	← mizu_no_oto left	02:04
	← harish left	02:04
	→ pen joined	02:05
	← revyd left	02:05
	← ErKa left	02:05
	alpha123 → alpha123\|away	02:05
	← mlimaloureiro left	02:05
	→ revyd joined	02:05
	→ skorgon_ joined	02:06
rawtaz	must be one of the oddest questions today :)	02:06
	← jerius left	02:07
	→ daito joined	02:08
	→ Exancillatus joined	02:08
	→ ketas joined	02:08
	→ mizu_no_oto joined	02:09
	← gnu-man left	02:09
	← Gamecubic left	02:09
	← guardianx left	02:10
	← revyd left	02:10
	→ tmcmahon joined	02:10
	← Silmarilion left	02:10
	→ M4renz joined	02:10
	← dougsland left	02:11
	→ KeiKun_ joined	02:12
	← KeiKun left	02:12
	← ewnd9 left	02:12
	← daito left	02:13
	← Exancillatus left	02:13
	← Suprano left	02:14
	← gusblan left	02:15
	← skorgon_ left	02:16
	← tmcmahon left	02:17
	→ jbarker joined	02:18
	→ afhammad joined	02:18
	← makubi left	02:18
	→ Cephalostrum joined	02:18
	← Cephalostrum left	02:19
	← BlueOrg left	02:19
	→ Cephalostrum joined	02:19
	← Apocalisp left	02:20
	→ makubi joined	02:20
	← braj left	02:21
	→ daito_ joined	02:22
	← jbarker left	02:23
	→ Gamecubic joined	02:24
	→ Lewoco joined	02:24
	→ h0bbit_ joined	02:25
Lewoco	I'm trying to checkout the HEAD of a branch, but when I do 'git checkout HEAD' it seems to do nothing... why?	02:26
	← afhammad left	02:27
	→ marcjs joined	02:27
	← pen left	02:27
ojacobson	'HEAD' is the current commit	02:27
	→ pen joined	02:27
ojacobson	(and, if you're on a branch, the current branch)	02:27
Lewoco	ojacobson, Is there a special name to refer to the newest rev of a branch?	02:28
ojacobson	"the HEAD of a branch" doesn't strictly make any sense, but individual branches' most recent commits are co-identified by the branch name itself	02:28
	the most recent commit on master is named 'master'	02:28
	man gitrevisions	02:28
gitinfo	the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	02:28
	→ sr105 joined	02:29
	← simesy left	02:30
	← Nicholas_Rishel left	02:30
	← Targen left	02:30
	→ pencilcheck joined	02:32
	← marcjs left	02:32
	← h0bbit_ left	02:32
	← pen left	02:33
	→ xjiujiu joined	02:34
	← Nizumzen left	02:35
	→ fphilipe joined	02:36
	← psyl0n left	02:36
	← pheaver left	02:36
Lewoco	ojacobson, What do you mean by "most recent commits are co-identified by the branch name itself" ? If I do 'git checkout mybranchname' it doesn't seem to check out the most recent commit on that branch...	02:37
ojacobson	Lewoco: it absolutely does. What are you observing that convinces you otherwise? !repro	02:37
gitinfo	Lewoco: Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.	02:37
	← mizu_no_oto left	02:37
	→ damccull joined	02:38
	← darius93 left	02:39
	← mellernoia left	02:39
	→ darius93 joined	02:40
	← fphilipe left	02:41
	← lethjakman left	02:42
	→ iderik joined	02:42
	→ Goplat joined	02:42
	← Goplat left	02:43
	→ Goplat joined	02:43
	← lyang0 left	02:43
	← gwb3 left	02:43
	← shruggar left	02:44
Lewoco	ojacobson, Ok I think my problem is that somehow I created a branch on my local repo which is no longer following new revisions from the remote repo. Essentially when I checkout 'foo' it is an older revision than when I checkout 'origin/foo'.	02:44
	→ chadjuehring joined	02:44
	→ jbarker joined	02:44
	→ skorgon_ joined	02:45
	← beyondcreed left	02:45
	← chadjuehring left	02:45
	devin_ → dglambert	02:46
ojacobson	Local branches are always independent of same-named branches in remotes	02:46
	← dglambert left	02:46
	→ dglambert joined	02:46
	→ Destos joined	02:46
	scoT_T`zZz → scoT_T	02:46
ojacobson	If your workflow implies some relationship (which is common), set the remote branch as the upstream of your local branch and use eg. 'git pull' or 'git pull --rebase' to update your local branch, depending on whether you want to merge your local changes with upstream changes, or rebase your local changes on top of the upstream changes	02:46
	(there's no meaningful difference is there are no local changes :)	02:47
	← scoT_T left	02:47
	→ scoT_T joined	02:47
	← scoT_T left	02:47
	→ scoT_T joined	02:47
	← LucasKA left	02:48
	← jhulten left	02:48
Lewoco	ojacobson, Yay, 'git merge origin/foo' fast forwarded it to where I want. Thank you.	02:48
ojacobson	git merge can be told to default to upstream (and does by default in later git versions, IIRC), if that's going to com eup a lot :)	02:48
	man git-config	02:48
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	02:48
	← mribeirodantas left	02:50
	→ harish joined	02:50
	← pencilcheck left	02:53
	← Guest7625 left	02:55
	← skorgon_ left	02:55
	→ daito joined	02:56
	← daito_ left	02:57
	→ petey joined	02:58
	← bitkiller left	02:58
	→ kt_ joined	03:02
	← kt_ left	03:03
	→ mgolawala joined	03:03
	→ unomystEz joined	03:03
	← insejn left	03:04
	← David left	03:04
unomystEz	i cloned a repo from server1 and it's set as my origin (by default I suppose), I want to instead push to server2 on which I created a bare repo, but I can't execute git remote add origin server2:pathspec	03:04
	mgolawala → KT_	03:05
unomystEz	do I have to give it another name? I would prefer origin since I'm deleting the repo on server1 anyway	03:05
grawity	then remove the existing 'origin' remote that `git clone` created	03:05
ojacobson	unomystEz: man git-remote	03:05
gitinfo	unomystEz: the git-remote manpage is available at http://jk.gs/git-remote.html	03:05
grawity	or, just change its URL.	03:05
ojacobson	have a look at 'rm' or 'set-url'	03:05
	→ insejn joined	03:05
	→ griotspeak joined	03:05
unomystEz	ok	03:05
	set-url worked, thanks alot	03:06
	→ Etheco- joined	03:06
unomystEz	nice, and a git push -all pushed all my tags, branches, looks like i can delete the old repo now	03:08
	→ gusblan joined	03:09
	→ gwb3 joined	03:09
unomystEz	btw, I usually alias basically commands like co = checkout, etc.. do you guys usually scp your .gitconfig over to all your boxes you use git on?	03:10
	or how do you manage your git config across different machines?	03:10
	← Etheco left	03:10
	→ lyang0 joined	03:10
	← rylinaux left	03:11
grawity	unomystEz: I keep it in a Git repo	03:11
unomystEz	grawity, then pull changes on all the target machines?	03:12
grawity	sure	03:12
	unomystEz: search Github for people's repos named "dotfiles"	03:12
unomystEz	hmm,, i have a lot of boxes though... i have the same setup at home but just may adopt this at work	03:12
	thanks for the tip	03:12
grawity	I add something like `git pull --ff-only` in cron	03:13
	at work, you probably have NFS or at least sshfs? :P	03:13
	← KT_ left	03:13
unomystEz	grawity, i do but that could get messy (and potentially dangerous)	03:14
	i like the dotfiles approach, i'll read up on it more, will need it to be rock solid of course	03:14
	btw, why ff-only? to prevent clobbering local changes?	03:14
	→ axl_ joined	03:15
	← gwb3 left	03:15
ojacobson	or introducing merge commits (or, worse, conflicts) automatically	03:15
grawity	automated merges done from cronjobs? that way lies madness.	03:15
	→ carlos-huang joined	03:15
unomystEz	git pull --ff-only == git fetch && git merge --ff-only right?	03:16
	→ gwb3 joined	03:16
grawity	pretty much, yeah	03:16
	local changes are rarely needed if the program lets you include ~/.config/git/config-local or ~/.bashrc-$HOSTNAME or such	03:16
	← diegoviola left	03:16
grawity	my cronjob's https://github.com/grawity/code/blob/master/dist/pull	03:16
	← gwb3 left	03:16
	→ Guest80723 joined	03:17
	→ diegoviola joined	03:17
grawity	and yes, apparently I forgot --ff-only there	03:17
	→ gwb3 joined	03:18
ojacobson	corntab	03:18
unomystEz	if if ! git-on-master; then <-- two if keywordS?	03:18
	line 87	03:18
ojacobson	that's legit	03:19
	surprising, but legit	03:19
	have a look at line 93	03:19
	(I would not write a shell if that way, but it's permitted)	03:19
grawity	surprisingly, that turned out to be the easiest way to do what I wanted...	03:19
ojacobson	picking it apart, it looks like a verbose 'or'	03:20
	without having to deal with bash's \|\| operator's quirks	03:20
grawity	yep	03:20
	`if if` is confusing, but `if case ... esac` is nice for making filter lists like rm() in https://raw2.github.com/grawity/dotfiles/master/bashrc-rain	03:21
	← carlos-huang left	03:21
	← Guest80723 left	03:23
	→ skorgon_ joined	03:23
	→ cmpera joined	03:24
	← dyng left	03:24
	→ dyng joined	03:25
Vivekananda	hey everyone. I need to find out how to change folder structure iin a git repo. Right now my files are inside a/b/c/ and I want everything to get to a/	03:26
	← AaronMT left	03:26
Vivekananda	the .git is present in the same folder as a ie outside a/	03:26
grawity	so just `git mv` the files	03:26
	→ Targen joined	03:27
	alpha123\|away → alpha123	03:28
	← dyng left	03:29
	→ revyd joined	03:29
	→ berdario joined	03:30
	← dglambert left	03:30
	← bwreilly left	03:31
	→ rylinaux_ joined	03:32
Vivekananda	grawity: I did not know that	03:32
	I assumed I had to do a rebase or something	03:32
grawity	only if you want to move the files in all historical commits	03:33
	← gwb3 left	03:33
grawity	(in which case it'd probably be `git filter-branch` anyway)	03:33
	← shinnya left	03:33
	← skorgon_ left	03:33
grawity	`git rebase` is used for editing history and shuffling around commits – it has not much to do with shuffling around files.	03:34
	→ smwa joined	03:34
	→ chrisshattuck joined	03:35
Vivekananda	okay that confuses me a little. So is shuffling around commit not the same as shuffling around files ?	03:35
	also if I put all files inside a/ from a/b/c/ and then commit and push	03:36
	will I get this new structure in the remote repo ?	03:36
	→ SocalNick joined	03:36
grawity	yes	03:37
	← quup left	03:37
berdario	I want to do some whitespace cleanup... to avoid cluttering up the blame output, I was thinking of doing a root rebase, but upon reading this http://stackoverflow.com/questions/6088440/git-how-to-use-rebase-with-theirs-strategy it doesn't seem to be what I'm looking for (maybe filter-branch?)... I guess that since lot of new lines have been added since the first commit, rebasing everything unto a commit that already has	03:37
	the new lines doesn't make sense?	03:37
	→ quup joined	03:38
	← echo-area left	03:38
	→ herdingcat joined	03:38
	→ echo-area joined	03:39
berdario	(btw, "only" 65% of the lines in my git blame have been affected by the whitespace cleanup... maybe it's not even worth it to change the history?)	03:39
	→ LucasKA joined	03:39
	← angrytongan left	03:40
	← smwa left	03:42
	← joelm left	03:42
	→ bwreilly joined	03:42
	→ Celelibi joined	03:42
Celelibi	Hi	03:43
ojacobson	berdario: man git-blame and have a look at the -w option	03:43
gitinfo	berdario: the git-blame manpage is available at http://jk.gs/git-blame.html	03:43
Celelibi	How do I remove a change from the index and put it back as "unstaged" change?	03:43
	→ gwb3 joined	03:43
berdario	ojacobson thanks, awesome!	03:44
d0liver	Is there a way with git to remove all the tracked files from the cache that have been removed in the working directory?	03:44
Celelibi	git reset -p	03:45
ojacobson	d0liver: man git-add, read the description of --all	03:45
gitinfo	d0liver: the git-add manpage is available at http://jk.gs/git-add.html	03:45
	← chrisshattuck left	03:45
	← gwb3 left	03:45
	← acharlieh left	03:45
d0liver	ojacobson: I don't think that's what I'm after here	03:45
Vivekananda	I am reading the git cli help and it says I can do: git mv -v/r/f source ... destination to actually move files inside source to a destination and not just rename it	03:45
d0liver	that will add everything that's untracked and not ignored	03:45
Vivekananda	do I understand this correclty ?	03:45
	→ chrisshattuck joined	03:45
ojacobson	d0liver: that's not what --all does, read it again :)	03:45
	→ dreiss joined	03:46
ojacobson	'git add' will do that without passing options.	03:46
	→ b3tamax joined	03:46
	← petey left	03:46
	→ gwb3 joined	03:47
	→ Mars joined	03:48
d0liver	ojacobson: That's what it looks like to me from the description. Can you give me a counter-example?	03:48
	Mars → Guest30150	03:48
	← gwb3 left	03:48
ojacobson	-A, --all, --no-ignore-removal	03:48
	Update the index not only where the working tree has a file matching <pathspec> but also where the index already has an entry. This adds, modifies,	03:48
	and removes index entries to match the working tree.	03:48
	line break positioned to draw attention. :)	03:49
	→ gwb3 joined	03:49
	→ simesy joined	03:49
d0liver	I'm still not understanding what the difference is here.	03:49
	← gwb3 left	03:50
	← w0ng left	03:50
	→ w0ng joined	03:50
d0liver	adds, modifies, and removes index entries to match the working tree	03:50
	← b3tamax left	03:50
d0liver	that's what I was saying	03:50
ojacobson	'git add foo' will recurse through foo, adding not-ignored files and staging changes in tracked files	03:50
	'git add --all foo' will do all of that, and remove files that are absent in the work tree from the staging area, so that they will also be absent in the next commit	03:51
	→ b3tamax joined	03:52
d0liver	Okay, I see. It's still not really what I'm after though. I want to stage only the removals disregarding modifications, new files, etc.	03:52
	I was wondering if there was a way to essentially do a git commit -a but only with removals	03:53
	→ sohail joined	03:53
d0liver	or if the preferred thing is to just git rm them	03:53
	→ gwb3 joined	03:53
	← weiyang left	03:54
ojacobson	man git-rm outlines some alternatives for that	03:54
gitinfo	the git-rm manpage is available at http://jk.gs/git-rm.html	03:54
d0liver	okay	03:54
	thanks	03:54
	← pbrobst left	03:54
Celelibi	What is the rerere mechanism?	03:54
	← sr105 left	03:56
	← axl_ left	03:56
Celelibi	Can I merge everything but a commit from the other branch?	03:57
Vivekananda	any comments anyone ?	03:58
	← plasticle left	04:00
	→ ls_ joined	04:00
	→ pr0ton joined	04:00
ls_	Perhaps someone can tell me how to restore files deleted, with the 'rm' command. I forgot the command.	04:01
cehteh	checkout	04:01
	← Noldorin left	04:02
	→ skorgon_ joined	04:02
ls_	Meaning as long I have two things, the .git directory and the commit's sha1 sum, file are restorable?	04:03
	← veebers left	04:03
grawity	yes	04:03
	→ weiyang joined	04:07
	← ls_ left	04:08
	← cmpera left	04:08
	← hog left	04:08
	← CheckDavid left	04:09
	← ariscop left	04:09
Celelibi	So, what's the best way to deal with this: I have two set of binary files, one for 32 bits, and one for 64 bits.	04:10
	One set replace the other.	04:10
	They can't live together.	04:10
	But most of the code around is the same.	04:10
	→ cmpera joined	04:10
Celelibi	I thought making branches would be nice.	04:11
	But it's quite a pain to keep them in sync.	04:11
	→ JohnBat26 joined	04:11
ojacobson	Are these binary files as in "the output of some build system like Make" or binary files as in "don't lose them, we'd have to re-buy/re-draw/re-render/etc them, but they're not text"?	04:11
Vivekananda	do I add .gitignore itself to tracked files ?	04:11
grawity	Vivekananda: yes	04:11
Celelibi	ojacobson: more like "re-download by hand".	04:12
	→ gdogrady joined	04:12
Celelibi	These are library given as-is.	04:12
	libraries*	04:12
ojacobson	Use a dependency management tool, not git, to import libraries, ideally.	04:12
	Store the dependency info in git, rather than the dependencies themselves	04:13
	(eg. npm's package.json file, maven's pom.xml, python's setup.py, a Gemfile, whatever)	04:13
Celelibi	What is this?	04:13
	→ ls_ joined	04:13
	← Gamecubic left	04:13
	← skorgon_ left	04:14
ls_	Supprising I run the command sudo git checkout sha1 sum , yet now files are reverted?	04:14
Celelibi	And then, how do I build my project?	04:14
grawity	1) why did you use sudo, 2) reverted to /what/?	04:14
	→ preflex_ joined	04:15
	← preflex left	04:15
	← basdfasdf left	04:15
ls_	Oh I have fun deleting sys_conf files, rebooting and see how it effect the computer.	04:15
grawity	...ok	04:15
	preflex_ → preflex	04:15
ls_	Which explain why I thought I could use git, so I could revert when back to previous states.	04:16
Celelibi	ojacobson: these libs are kind of part of the source. They are .a (an archive full of .o) to be linked with.	04:16
ojacobson	Your other practical option is to restructure your project's directory layout so that you can have both sets of libs around at the same time, and arrange for your build script to pick the right set of libs automatically	04:16
	→ abvdfasdf joined	04:17
	→ takamichi joined	04:18
Celelibi	Hum... Yes, that's what I wanted to avoid.	04:18
	→ mizu_no_oto joined	04:18
ls_	Two, I reverting to the commit with the files.	04:19
	← daito left	04:19
	← Guest30150 left	04:20
	← ncthom91 left	04:22
	→ ariscop joined	04:23
	→ fphilipe joined	04:24
	→ dyng joined	04:24
	→ Nicholas_Rishel joined	04:24
	← obert left	04:25
ls_	Perhaps the 'git status' command showing deleted: is reversable?	04:26
	← fphilipe left	04:29
	← KeiKun_ left	04:29
	→ KT_ joined	04:29
	→ GT4066 joined	04:32
	→ Gamecubic joined	04:33
ls_	Interestingly... running 'sudo git checkout sha1 filename' worked?. How about wildcards in place of the filename?	04:33
	← nodiscc left	04:34
	→ ben_alman joined	04:34
	← m0viefreak left	04:35
	← Dougie187 left	04:36
	← ls_ left	04:38
	← mizu_no_oto left	04:39
	← ben_alman left	04:39
	→ David joined	04:39
	← ariscop left	04:39
	→ Mr_Grim1 joined	04:42
	→ ben_alman joined	04:45
	master_j → arch	04:47
	← SocalNick left	04:47
	← ben_alman left	04:50
	← kukks left	04:53
	← dreiss left	04:55
	← harish left	04:57
	← herdingcat left	04:57
	→ dreiss joined	04:58
	→ herdingcat joined	04:59
	← donmichelangelo left	05:00
	→ donmichelangelo joined	05:00
	→ guns joined	05:03
	← dyng left	05:03
	← sohail left	05:03
	← Spudster left	05:03
	→ mizu_no_oto joined	05:05
Mr_Grim1	im just reading up on some git and im wondering if someone could explain to me what this exactly does: git remote add upstream -m master git://github.com/openshift/rails-example.git	05:05
	git pull -s recursive -X theirs upstream master	05:05
	← n8chur left	05:07
	→ caasig joined	05:08
	→ myc joined	05:08
	← myc left	05:08
	→ ahmedelgabri joined	05:09
	← el3ctron left	05:10
	← pr0ton left	05:11
	→ Oobaew83 joined	05:11
	← gusblan left	05:11
Oobaew83	Hello.	05:12
gitinfo	Oobaew83: 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.	05:12
Oobaew83	Is there a nifty way to git log to xml? I saw I looked for about two hours and only for about 3 posts on SO.	05:12
ojacobson	Oobaew83: I'd be very surprised if there were: generating correct XML given all the permutations of possible inputs from Git is a reasonably hard proposition (you have to use a real XML library, you can't "just" use string manipulation to do it)	05:14
	man git-log is, of course, authoritative	05:14
gitinfo	the git-log manpage is available at http://jk.gs/git-log.html	05:14
	← KT_ left	05:14
ojacobson	What're you hoping to feed it to?	05:14
	There are more machine-friendly history formats available to git that might work just as well as XML	05:15
	← dreiss left	05:15
ojacobson	(eg. git rev-list and friends)	05:15
Oobaew83	I want to download the diff and commit messages from git repos	05:15
	http://stackoverflow.com/questions/6327276/parsing-git-log-output-preferably-as-xml/6327321#6327321	05:15
	→ KT_ joined	05:15
Oobaew83	That is one of the links I found, but similar to what you said - one of the comments mentions getting burned for all the xml format it doesn't cover, ojacobson	05:16
ojacobson	Yeah, and the proposed "fix" (of wrapping everything in <![CDATA[ blocks ]]> ) still runs afoul of the sequence ]]> if it's present in your git repo	05:16
	again, why XML specifically?	05:16
	← arch left	05:16
	→ master_j joined	05:17
Oobaew83	I will be working with other xml data, so it seemed conventional if it is possible in git.	05:17
ojacobson	mm	05:18
	Stock unix answer - "you could write something, I guess" - applies	05:18
Oobaew83	yeah...	05:18
	Is there another data format I could use instead of xml?	05:18
ojacobson	walking git-rev-list and running git-cat-file on each object as you go in a python/ruby/perl script that emits XML would work	05:18
	(note that this is more or less what git-log does :)	05:18
	← danecando left	05:19
	← berdario left	05:19
	→ daito joined	05:20
Oobaew83	ojacobson: I will look into this, thank you.	05:20
	← ahmedelgabri left	05:21
	→ h0bbit_ joined	05:21
	← daito left	05:21
	→ ahmedelgabri joined	05:21
ojacobson	parsing 'git cat-file commit <sha>' is much, much easier than parsing git-log's output, fwiw -- the format is designed to be machine-readable	05:21
	→ daito joined	05:21
	→ petey joined	05:22
	offby1 parses git cat-file in fact	05:22
frogonwheels	Oobaew83: Yeah- I've done a git log to html munger, and it wasn't pretty. Especially as it was a windzoe cmd script !	05:22
offby1	https://gist.github.com/8413511 <--- example	05:22
Oobaew83	my morale is declining	05:22
ojacobson	I've got one somewhere that turns git history into dot diagram files	05:23
	← mizu_no_oto left	05:23
	→ angrytongan joined	05:23
ojacobson	python, < 100 lines, only really depends on the subprocess module and on git being available	05:23
	it's not _actually_ hard, but you do have to do the toolsmithing if you want it done right	05:23
	if you can make brave assumptions about the contents of your log, you can cheat with custom formats and get pretty close to right	05:23
	offby1 hitches up his suspenders and spits a chaw of tobaccy	05:23
offby1	'at's raht.	05:24
Oobaew83	brave is right.	05:24
frogonwheels	.. and if I can do it using DOS cmd script - doing it in a real language should be a cinch.	05:25
	→ RofricTheBearded joined	05:25
ojacobson	Oobaew83: the line between bravery and stupidity... :)	05:25
Oobaew83	luck, sire.	05:25
ojacobson	on the other hand if your data does happen to be cooperating with your assumptions you can be done and in the pub pretty quick	05:26
	→ pen joined	05:26
	← ojacobson left	05:26
frogonwheels	ojacobson: are you insinuating that I crossed the line into stupidity?	05:26
	→ dyng joined	05:26
Oobaew83	how come git doesn't have xml style for pretty print...	05:26
	too much work?	05:26
	← chrisshattuck left	05:26
	← Nicholas_Rishel left	05:26
frogonwheels	Oobaew83: xml is not really a unixy thing	05:27
	Oobaew83: esp cli unix :)	05:27
Oobaew83	what data type is?	05:27
frogonwheels	um. text?	05:27
Oobaew83	how do you store a git commit and diffs as a text file?	05:28
	→ marcjs joined	05:28
Oobaew83	I am open to alternatives :)	05:28
	→ alezost joined	05:29
	← KT_ left	05:31
Oobaew83	a csv would be handy, but I don't think that will work	05:31
offby1	Oobaew83: you could store the commit by doing "git cat-file -p SHA > file"	05:32
	but I bet that's not what you meant	05:32
	you can capture the diff introduced by a non-merge commit with "git show SHA"	05:32
	← marcjs left	05:33
offby1	you can capture that diff, plus a bit of metadata that makes it easier to apply as a patch, with "git format-path"	05:33
	er	05:33
	git format-patch	05:33
	and you can apply that patch with "git am"	05:33
	→ ariscop joined	05:34
	← Tommy[D]_ left	05:34
	→ DolphinDream joined	05:37
	→ tgkokk joined	05:38
	→ Mars joined	05:40
	Mars → Guest57823	05:40
	→ hoschi joined	05:41
Oobaew83	I will look at this too, thanks offby1.	05:41
	→ KT_ joined	05:41
	← unomystEz left	05:41
	→ skorgon_ joined	05:42
	← Guest57823 left	05:45
	→ ben_alman joined	05:45
	← KT_ left	05:47
	→ beyondcreed joined	05:47
	← cmpera left	05:48
	→ Tommy[D]_ joined	05:48
	← beyondcreed left	05:48
	← ben_alman left	05:50
	→ akiva joined	05:52
	← skorgon_ left	05:52
	→ ArcTanSusan joined	05:55
	→ Macaveli joined	05:56
	← tgkokk left	05:57
	→ nZac joined	05:59
	emmajane → [emmajane]	05:59
	← tpollard left	05:59
	← LuxuryMode left	05:59
	→ kerrick joined	05:59
	→ KT_ joined	06:00
	← simesy left	06:00
	→ redrocket joined	06:00
	→ girishr joined	06:02
	→ GeertV joined	06:02
	← Joost left	06:02
	← nZac left	06:03
	→ harish joined	06:07
	→ pheaver joined	06:08
	[emmajane] → emmajane	06:09
	← petey left	06:09
	← jbarker left	06:09
	← akiva left	06:10
	→ ncthom91 joined	06:11
	→ Joost joined	06:11
	→ fphilipe joined	06:12
	→ freimatz_ joined	06:15
	→ EngierkO_ joined	06:15
	← cff left	06:16
	← EngierkO left	06:16
	← fphilipe left	06:17
	← Joost left	06:17
	← ahmedelgabri left	06:19
	← wilx left	06:20
	→ kcj joined	06:20
	→ dprophit joined	06:20
	← revyd left	06:23
	← tarkus left	06:23
	→ kbart joined	06:23
	→ revyd joined	06:23
	→ Joost joined	06:24
	← MissionCritical left	06:25
	→ ahmedelgabri joined	06:26
	← donmichelangelo left	06:27
	← DolphinDream left	06:27
	→ donmichelangelo joined	06:27
	← master_j left	06:27
	← revyd left	06:28
phipes	i just deleted a whole directory so i could add a submodule to that directory.	06:28
	→ marcjs joined	06:29
phipes	im sure there's a better way to do that.	06:29
	← freestyl3r left	06:29
milki	phipes: no, normally its the other way around, !submodule_rm	06:30
gitinfo	phipes: [!submodules_rm] You want to delete submodules? Excellent choice! All commands are in the superproject. Edit/delete .gitmodules to remove the submodule. Then `rm -rf submodulepath; git rm -f --cached submodulepath; git commit -am "Removed submodules!"` Inspect .git/config for "submodule" entries to remove. Inspect .git/modules for caches to remove "!gitslave" or "!subtree" (type them!) might be alternatives	06:30
	← harish left	06:30
	→ pantsman joined	06:33
	→ Mission-Critical joined	06:33
	← marcjs left	06:34
	→ arvind_khadri joined	06:34
	→ namita joined	06:35
	← ncthom91 left	06:35
	→ JohnB0 joined	06:35
	→ ThomasLocke joined	06:36
	→ teran_ joined	06:36
	← arvind_khadri left	06:39
	→ Draleth joined	06:39
	← teran left	06:39
	← guns left	06:40
	→ jargon joined	06:40
Draleth	My Google-fu has failed me; I can not find a hook script to echo commit messages just pulled from upstream that contain a tag like 'ANN' (for announcement, i.e. in client-side configuration INI changes).	06:40
	Any ideas how to accomplish this?	06:40
	→ teran joined	06:41
phipes	milki: git was complaining about some index when i was trying to add a submodule	06:41
	← starkhalo left	06:42
	← teran_ left	06:44
	→ jceb joined	06:44
	Mission-Critical → MissionCritical	06:45
	← cyphase left	06:46
	→ CKyle joined	06:47
	← geomyidae left	06:48
	→ geomyidae joined	06:49
	→ jbarker joined	06:50
	→ josealobato joined	06:50
	→ ehmatthes joined	06:51
	← kcj left	06:52
	→ kcj joined	06:53
	→ mlechner joined	06:54
	← jbarker left	06:55
	← takamichi left	06:56
	→ ben_alman joined	06:56
	→ lite_ joined	06:56
milki	Draleth: theres no hook fro something like that	06:56
	→ harish joined	06:57
Draleth	That's kinda lame. :/	06:58
	→ xjiujiu_ joined	06:59
	→ LeeXiaolan__ joined	06:59
	← the_cybersphinx left	06:59
	← Draleth left	06:59
	← JesseC left	06:59
	→ marcjs joined	07:00
	→ JesseC joined	07:00
	← pantsman left	07:00
	← spudz_ left	07:00
	→ spudz joined	07:01
	← ben_alman left	07:02
	← LeeXiaolan_ left	07:02
	← xjiujiu left	07:02
	← jceb left	07:02
josealobato	This is a rebase (changing the history) question. I know that changing the history once the changes are in the server is bad but I can not imagine what will happen. In my case the server creates the remote branch from master for me (Stash) and I check it out locally to work. If sometime later I rebase locally onto master where there was new changes. will i be able to push? what will be the problem?	07:05
	← Destos left	07:05
	→ soee joined	07:06
	← EngierkO_ left	07:06
	→ Misan joined	07:06
	→ EngierkO joined	07:08
	→ hahuang65 joined	07:08
	→ ncthom91 joined	07:08
	→ nnyk_ joined	07:09
	→ papegaaij joined	07:10
	→ takamichi joined	07:10
	→ cbreak-work joined	07:12
	→ Kicer86 joined	07:13
	← xjiujiu_ left	07:14
	→ xjiujiu joined	07:14
	← Macaveli left	07:14
	→ enr joined	07:16
	← oojacoboo left	07:16
	← chepre` left	07:17
	← KT_ left	07:18
	→ KT_ joined	07:19
	→ m0ltar joined	07:20
	→ lb1a joined	07:20
	→ unomystEz joined	07:20
	→ leeN joined	07:20
	← leeN left	07:20
	← namita left	07:21
	→ Macaveli joined	07:21
	← enr left	07:21
frogonwheels	josealobato: !rewrite	07:22
gitinfo	josealobato: [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	07:22
	← xjiujiu left	07:22
	→ ferreum joined	07:22
	→ xjiujiu joined	07:22
	→ FatDarrel joined	07:23
josealobato	thanks!	07:23
	← KT_ left	07:23
frogonwheels	josealobato: Hmm.. it is not clear that the scenario you are describing actually rewrites public history	07:23
	← ifdm_ left	07:23
	→ introom joined	07:23
	← CKyle left	07:24
frogonwheels	josealobato: !repro	07:24
gitinfo	josealobato: Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.	07:24
introom	can git track different files in different branches?	07:24
frogonwheels	introom: sure	07:24
introom	frogonwheels: wow. that's good. Wait a moment. Let me write down my problem in detail.	07:25
frogonwheels	introom: really what git cares about when you checkout a branch is the difference between the commit you are at now, and the commit you are going to, and the difference between them.	07:25
josealobato	@frofonwheels not sure either. The fact that the branch is a remote branch from the user perspective and the user change its parent using rebase looks to me like change in the history.	07:26
	frongonwheels.	07:26
frogonwheels	introom: git can actually track two totally unrelated git trees in the same repo, and while that sounds totally bizarre, I have actually used this to significant advantage.	07:26
	← marcjs left	07:26
frogonwheels	josealobato: frog<tab>	07:26
	;)	07:26
	→ j0ran joined	07:27
	→ rath joined	07:27
	→ fphilipe joined	07:27
	→ xequence joined	07:28
introom	frogonwheels: hi. https://gist.github.com/introom/3a642be58030d4ebe7f6	07:28
lb1a	frogonwheels, it works \o/	07:29
	→ jceb joined	07:29
	→ alansaul-laptop joined	07:30
	→ slyfox joined	07:30
	← slyfox left	07:30
	→ slyfox joined	07:30
	→ simesy joined	07:30
	← xjiujiu left	07:31
	→ xjiujiu joined	07:31
introom	frogonwheels: hi. do you have any idea on that?	07:31
	← thiago left	07:32
josealobato	frogonwheels: thanks!	07:32
	→ matanya joined	07:33
	← Bass10 left	07:33
	← ncthom91 left	07:34
	← Goplat left	07:35
frogonwheels	introom: yeah - use sparse checkout	07:36
	← angrytongan left	07:36
	← nnyk_ left	07:36
	→ CKyle joined	07:37
	→ Poly-C_atwork joined	07:37
	→ ifdm_ joined	07:37
	→ nilg joined	07:38
introom	frogonwheels: thanks. I will have a read on that.	07:38
frogonwheels	introom: man git-read-tree look for 'sparse checkout'	07:38
gitinfo	introom: the git-read-tree manpage is available at http://jk.gs/git-read-tree.html	07:38
frogonwheels	!dotfile	07:39
gitinfo	[!dotfiles] There are various issues associated with keeping dot-files in git (permissions, avoiding having a .git in ~/ or /). One attempt to manage this is http://dotfiles.github.com/	07:39
	→ nnyk_ joined	07:39
	← pen left	07:40
	→ ayaz joined	07:41
	← ayaz left	07:41
	→ ayaz joined	07:41
	→ cff joined	07:41
	← unomystEz left	07:41
	→ rob_w joined	07:41
	→ ajunior joined	07:41
	→ fRAiLtY- joined	07:42
	← xjiujiu left	07:42
	→ xjiujiu joined	07:42
	→ gain joined	07:44
	→ chepre` joined	07:44
	← chepre` left	07:44
	← cff left	07:45
	→ cff joined	07:46
	← dvide left	07:47
	← dik_dak left	07:47
	→ KT_ joined	07:51
	→ jbarker joined	07:51
	→ ahmedelg_ joined	07:52
	← David left	07:54
	← fphilipe left	07:55
	← Norm_ left	07:55
	→ fphilipe joined	07:55
	← KT_ left	07:55
	← jbarker left	07:55
	← ahmedelgabri left	07:55
	→ mishok13 joined	07:57
	→ arvind_khadri joined	07:58
	← mitz left	07:59
	← fphilipe left	08:00
	← jargon left	08:00
	→ eijk joined	08:01
	→ Raging_Hog joined	08:01
	→ jargon joined	08:02
	← jargon left	08:02
	→ jargon joined	08:02
	← FatDarrel left	08:02
	→ h_kon joined	08:04
	→ jnavila joined	08:04
	← phipes left	08:05
	→ xlinkz0 joined	08:06
	← zackiv31 left	08:06
	← nnyk_ left	08:07
	→ nnyk_ joined	08:07
	→ zackiv31 joined	08:07
	← Noctis_Lux left	08:08
	→ Noctis_Lux joined	08:09
	← CKyle left	08:09
	← nilg left	08:09
	→ YoMomma__ joined	08:09
	→ jnavila_ joined	08:09
	← JohnB0 left	08:09
	← jnavila left	08:10
	→ petey joined	08:10
	→ fire joined	08:11
	← mattcen left	08:12
	→ cmyers joined	08:12
	→ oskarkv joined	08:13
	→ Exancillatus joined	08:13
cmyers	does anyone know the plumbing, off the top of their head, to perform a merge of two hashes in a bare repo / without impacting the working tree (and just fail if there are any conflicts)	08:13
	or does such plumbing exist?	08:13
	lite_ → litebot_	08:14
	← joki left	08:14
	← litebot_ left	08:15
	← petey left	08:15
	→ litebot_ joined	08:15
	→ Sonderblade joined	08:16
osse	cmyers: merging requires a work tree	08:16
cmyers	osse: ballz.	08:17
osse	(afaik)	08:17
	→ luc___ joined	08:17
	cmyers wonders how the hell atlassian stash does it then	08:17
	← Sonderblade left	08:17
	→ macben joined	08:17
cmyers	it seems like merge expects HEAD to be one of the arguments	08:17
	← luc_ left	08:17
	← Exancillatus left	08:17
	← daito left	08:17
	→ Sonderblade joined	08:17
	← macben left	08:17
osse	cmyers: cd /tmp/foo; clone; merge; push; rm -r /tmp/foo perhaps?	08:17
cmyers	so if you create two pull requests at the same time, it has to test the merge to know if it merges cleanly...clearly it doesn't have races somehow	08:18
	→ harobed_ joined	08:18
	→ Exancillatus joined	08:18
cmyers	yeah, it probably does that	08:18
	→ macben joined	08:18
cmyers	but it is completely fucking opaque to the API they present to plugins	08:18
	litebot_ → lite_	08:18
cmyers	you get a GitScmCommandBuilder and call .builder(repo).merge().normal().argument(...)	08:19
frogonwheels	cmyers: the other thing is - how likely is that Stash uses jgit, and therefore they use some internal jgit voodoo to do it.	08:19
canton7	cmyers, github has some metrics it uses to decide whether a merge will conflict, but they're not bullet-proof. a merge it thinks will conflict might, in fact, not	08:19
jast	there are some classes of merges you can test without actually doing them	08:19
	but they're pretty limited	08:19
	→ joki joined	08:19
cmyers	but I have no clue how to set HEAD in this instance and their docs are shit	08:19
	← Gamecubic left	08:19
cmyers	frogonwheels: actually, it shells out to git, it doesn't use jgit	08:19
frogonwheels	cmyers: Is it just exposing jgit?	08:19
cmyers	whcih is why it is fast and doesn't suck	08:19
frogonwheels	cmyers: oh. wow	08:19
cmyers	wow indeed =)	08:19
frogonwheels	cmyers: yeah - that makes sense.	08:19
	→ rendar joined	08:19
jast	we should stop developing git and focus on jgit	08:20
	because everybody loves java	08:20
cmyers	it is amazing that so many java ninnies all decided to just bit the bullet and shell out all the fucking time	08:20
	frogonwheels kicks jast.	08:20
jast	aww	08:20
	where's the java AbstractLoveFactory	08:20
frogonwheels	:)	08:20
	→ gua_ould joined	08:20
cmyers	but don't worry, they did it as java-y as possible, with command-builder-factories and shit	08:20
jast	what a relief	08:21
cmyers	yeah, dodged a bullet thar	08:21
	← Olipro left	08:21
	→ phipes joined	08:21
cmyers	man, this shit is seriously making my life sad right now =( They do the merge, but I can't use their merge because it happens "later" for some definition of later (appears to happen synchronously, but after my plugin code's entry point)	08:22
	so if I want to have a merge ref, I need to perform the merge myself	08:22
	but their API is eye-bleeding obtuse	08:23
	← cads left	08:23
cmyers	why can't I just fucking do pullRequest.getMergeSha() or something? whine bitch	08:23
	→ mattcen joined	08:24
	→ intellix joined	08:24
	← Bosox20051 left	08:24
jast	so, what I'm taking away from your monologue is that atlassian is 125% awesome	08:25
cmyers	stash is the best product they've ever made, and it sucks a whole lot less than their other ones	08:25
	but yeah, it'll still make you cry	08:25
lb1a	ahh now i catched up.	08:26
jast	at some point, someone suggested that git use jira for tracking bugs	08:26
	← alansaul-laptop left	08:26
	← ehmatthes left	08:26
lb1a	all this java love distracted me	08:26
	→ swaechter joined	08:26
cmyers	jira isn't terrible from a user perspective, but running a large enterprise deployment is a PITA. If they will give the git project a hosted instance, it would be very low-effort	08:26
	of course, it's sorta weird for the premier OSS SCM to use a closed source bug tracker.	08:27
jast	they did hand out a hosted instance	08:27
	← girishr left	08:27
jast	it ended up not being used by anybody	08:27
	just like the previous time someone set up a bug tracker	08:27
cmyers	yeah, I guess that's not suprising.	08:27
jast	it's mostly outsiders who ask for a bug tracker for git	08:27
cmyers	our company runs on jira like it's our fucking business.	08:27
jast	our company is all redmine	08:27
cmyers	but we are a bunch of java ninnies to begin with	08:28
jast	which has its own special brand of PITAness	08:28
cmyers	we're super locked into several atlassian solutions	08:28
	← Happzz left	08:28
lb1a	our company is redmine, but we're all python lovers. so ruby sucks less than java but it still sucks	08:28
jast	I like ruby	08:28
	but rails (and everything inspired by it) is cancer	08:28
cmyers	python could have had me...but...whitespace sensitive => puke	08:28
jast	I don't care about whitespace sensitivity	08:29
cmyers	so ruby 4evar	08:29
	→ aChatir joined	08:29
jast	but I don't like python's TIOOWTDI approach	08:29
cmyers	yeah, I cut my teeth on perl, so ruby was a natural path for me too.	08:29
jast	I write perl at work	08:29
cmyers	I hear the newest major release of python is still a bit of a mess too, lots of broken libs and whatnot.	08:30
	scoT_T → scoT_T`zZz	08:30
jast	I hear python 2.x and 3.x are pretty much both used separately	08:30
cmyers	ruby 2.0 breaks some things but is generally not a disaster.	08:30
Poly-C_atwork	So the release notes say there's git-1.8.5.3 but no tarballs available.	08:30
	← aChatir left	08:30
cmyers	oh god, how am I going to do this back to the grind	08:31
	cmyers is grepping the stash source code for examples of doing merges with his fingers crossed	08:31
	→ afhammad joined	08:31
	→ jkremser joined	08:32
	← jkremser left	08:32
	→ jkremser joined	08:32
jast	.version	08:32
	gitinfo changed the topic to: Welcome to #git, the place for git help and seared salmon \| Current stable version: 1.8.5.3 \| Start here: http://jk.gs/git \| Seeing "Cannot send to channel" or unable to change nick? /msg gitinfo .voice \| git-bisect: No, not like that. Grow up.	08:32
jast	yay new version	08:33
	→ dreiss joined	08:33
Poly-C_atwork	Yeah but no tarballs yet :-/	08:33
	→ vtunka joined	08:33
jast	Poly-C_atwork: https://github.com/git/git/archive/v1.8.5.3.tar.gz	08:33
	→ tobias__ joined	08:33
jast	fortunately github autogenerates 'em	08:34
Poly-C_atwork	jast: meh... I was still looking at http://git-core.googlecode.com/files/ :-/	08:34
	← fire left	08:34
jast	I think the stuff on google code is updated manually	08:34
	and AFAIK it's not the official location anyway	08:34
	← YoMomma__ left	08:35
jast	no idea why that's the one linked to on git-scm.com	08:35
Poly-C_atwork	jast: exactly the URL I posted above ;)	08:35
	← lite_ left	08:35
jast	so apparently it _is_ official	08:35
	oh well	08:36
Poly-C_atwork	jast: it's not on the kernel.org mirrors as well.	08:36
	← Exancillatus left	08:37
	→ noname001__ joined	08:38
	→ daito joined	08:39
jast	yeah, all that stuff is cut manually, I think	08:39
	→ Happzz joined	08:39
	→ fphilipe joined	08:41
	→ jackneill joined	08:42
	← jackneill left	08:42
	→ jackneill joined	08:42
	← richwestcoast left	08:42
	← m0ltar left	08:44
	→ davipt joined	08:46
	← phipes left	08:46
	→ m0ltar joined	08:47
	→ konadr joined	08:47
	← rath left	08:48
	← konadr left	08:48
	→ konadr joined	08:48
	→ makubi1 joined	08:48
	zz_yuvipanda → yuvipanda	08:49
	← pheaver left	08:50
	← m0ltar left	08:50
	→ obert joined	08:50
	→ jbarker joined	08:51
makubi1	hi! I read that I can use a .gitattributes-file to get rid of the clrf-problem. I also think setting * text=auto or * text is the best setting for me. where is the difference between * text=auto and * text ? is there any difference?	08:51
	→ juril joined	08:52
	→ lavong joined	08:52
	→ marcjs joined	08:53
	← bwreilly left	08:54
	← ahmedelg_ left	08:54
	→ m0ltar joined	08:55
jast	text=auto enables git's automatic detection of textness	08:55
	← luc___ left	08:56
	← jbarker left	08:56
makubi1	jast: and how is that different to 'text'?	08:56
jast	it's not 100% reliable, of course. if you know some type of file is always text, explicitly setting text is the best idea. similarly, if you know some type of file is never text, expicitly setting -text is the best idea	08:56
	→ Balliad joined	08:56
jast	'text' forces git to assume that the file contains text, and so line ending conversion is enabled	08:56
makubi1	jast: alright, I understand, so binary files e.g. would be handled similar	08:56
	if I use '*'	08:57
	← Happzz left	08:57
	→ ben_alman joined	08:57
jast	yeah. if you use '* text', all files will be converted, even binary files (so that'll break them)	08:57
	→ luc___ joined	08:57
makubi1	jast: alright, thank you very much :)	08:57
jast	if you use '* text=auto', you get behaviour similar to enabling core.autocrlf	08:57
makubi1	jast: ok, so I may go with explicitly enabling it	08:58
	or enabling auto and explicitly disabling some file endings	08:58
	gonna have to think about it, thank you very much!	08:59
	← xlinkz0 left	08:59
	→ lite_ joined	08:59
	→ adsc joined	09:00
	→ jalmeida joined	09:00
	← ben_alman left	09:01
adsc	someone merged a branch into master too early...what is the general procedure to fix this? checkout the previous commit of master and then?	09:01
GyrosGeier	that's it	09:01
	or rather, use reset	09:02
	basically, you have to make master point at the previous commit	09:02
	→ Exancillatus joined	09:02
	← gdogrady left	09:03
	← lifelike left	09:03
	→ gdogrady joined	09:03
	→ KeiKun joined	09:03
	→ goTAN joined	09:04
	→ valtih joined	09:04
	← MattMaker left	09:04
	→ thevishy joined	09:04
	→ ahmedelgabri joined	09:04
adsc	and the dude can then push into the correct branch and all should be well?	09:04
	→ shruggar joined	09:04
	→ Norm_ joined	09:05
	← Exancillatus left	09:07
	→ MattMaker joined	09:07
	→ ben_alman joined	09:08
	→ psoo joined	09:09
	emmajane → [emmajane]	09:09
	→ elbeardmorez joined	09:11
	← luc___ left	09:11
makubi1	GyrosGeier: does this really help? I think if you would have to revert all commits that were merged from branchX --> master	09:11
	forget the 'if'	09:12
	→ luc___ joined	09:12
	← ben_alman left	09:12
	→ mitz joined	09:13
	← davipt left	09:14
	← asedeno left	09:15
	← nnyk_ left	09:17
	← takamichi left	09:17
	← ariscop left	09:18
	→ takamichi joined	09:19
	← ArcTanSusan left	09:19
	← lavong left	09:19
	→ irqq joined	09:20
jast	once the merge has been pushed, things get difficult	09:20
	!revert_merge	09:20
gitinfo	[!faulty_merge] If you have published a merge commit that turned out to be faulty and you'd like to get rid of it, you're probably not going to have a good time. Here's the full story: https://raw.github.com/git/git/master/Documentation/howto/revert-a-faulty-merge.txt	09:20
	← psoo left	09:20
	→ psoo joined	09:21
	← dreiss left	09:22
	→ jmux joined	09:22
	← goTAN left	09:23
	→ dobalina joined	09:23
	→ marhaban joined	09:23
	← davividal left	09:23
	→ asedeno joined	09:24
	[emmajane] → emmajane	09:24
	→ luc____ joined	09:24
	→ Vortex35 joined	09:25
	→ arand joined	09:25
	→ goTAN joined	09:26
	→ saorge joined	09:26
	← luc___ left	09:26
adsc	i just read up on reset, and it doesn't seem to be that scary	09:27
	← Vortex34 left	09:28
jmux	Is there a way to show the parent branch for a commit? My actual problem is the gerrit review process, which fetches and checks out the change (FETCH_HEAD).	09:28
arand	adsc: presumably, if you read up on it first, nothing is "scary" ;)	09:28
jmux	SO basically I want to find the first branch in my fetched change history and check, if its different from the current working branch.	09:29
	→ ciampix joined	09:30
	← ajunior left	09:30
	← saorge_ left	09:30
	← simesy left	09:31
jmux	I could write a script, but I wanted to ask, if there already exists a command.	09:31
adsc	holy crap, the link jast posted is scary, though, hella scary	09:32
	← teran left	09:32
	→ mellernoia joined	09:35
jast	adsc: the problem is that if only one person fixes the merge in their repo, but other people have already obtained a copy of it, they'll reintroduce the merge the next time they push	09:35
adsc	ow, that sucks	09:35
jast	if nobody else has the merge yet, you can just obliterate it with 'git reset' or something like that	09:35
adsc	they already have it	09:35
jast	or you can talk to everyone who has a copy of the merge and walk all of them through removing it in exactly the same way	09:36
kalleth	jast: good three way merge tool for osx go	09:36
jast	kalleth: I don't use osx	09:36
kalleth	humbug.	09:36
jast	my heart weeps for your suffering	09:36
kalleth	somehow i doubt that	09:37
	← lite_ left	09:37
	→ lite__ joined	09:37
	→ krosmen32 joined	09:41
	← goTAN left	09:42
	← intellix left	09:42
	→ joelrobichaud joined	09:43
	← lite__ left	09:44
	← MattMaker left	09:44
	→ Happzz joined	09:44
	→ MattMaker joined	09:44
	→ simesy joined	09:46
	← jrobichaud left	09:47
	← KeiKun left	09:47
	→ KeiKun_ joined	09:47
	← MattMaker left	09:47
	→ psyl0n joined	09:47
	→ Advocation joined	09:47
	→ MattMaker joined	09:48
	→ JTAAZQO5 joined	09:48
	→ vince_prignano joined	09:48
	→ goTAN joined	09:48
	← Macaveli left	09:49
	→ tvw joined	09:49
	→ surfperch joined	09:49
	→ nnyk_ joined	09:49
	→ Olipro joined	09:49
	→ Nizumzen joined	09:49
	→ jbarker joined	09:51
	→ endiruna joined	09:51
	→ Macaveli joined	09:52
	← kossy left	09:52
	→ Silmarilion joined	09:53
	→ mikecmpbll joined	09:54
	→ ChrisAnn joined	09:54
	→ eletuchy_ joined	09:54
	← goTAN left	09:54
	→ intellix joined	09:55
	← zackiv31 left	09:55
	← jbarker left	09:56
	← endiruna left	09:58
	← eletuchy left	09:58
	→ kossy joined	09:59
	← eletuchy_ left	09:59
	← diegoviola left	10:00
	← kossy left	10:01
	← macben left	10:01
	← surfperch left	10:01
adsc	alright, my master now points to the commit before the merge, how can I get that into the remote without having to merge?	10:01
	← takamichi left	10:02
	← aspiers left	10:02
	→ aspiers joined	10:03
	→ teran joined	10:03
	← dyng left	10:04
	→ dyng joined	10:04
	→ macben joined	10:04
	→ takamichi joined	10:04
	← tvw left	10:05
	← xjiujiu left	10:05
	→ albertgrala joined	10:05
	← simesy left	10:06
	→ poulson joined	10:06
	← bitkid left	10:06
	→ bitkid joined	10:06
	→ xjiujiu joined	10:07
	← tomek left	10:07
	← spacepluk left	10:08
	← teran left	10:09
	→ blinkdesign joined	10:09
	→ kossy joined	10:09
	← moondoggy left	10:09
	← dyng left	10:09
	← alezost left	10:10
	→ Gavrisimo joined	10:10
	← xjiujiu left	10:10
	← afhammad left	10:10
	→ alezost joined	10:10
	← kossy left	10:10
	→ dobalina1 joined	10:11
	← Eridius left	10:11
	→ Eridius joined	10:11
	→ petey joined	10:11
	→ moondoggy joined	10:11
	→ tomek joined	10:11
	→ zackiv31 joined	10:11
	→ xjiujiu joined	10:12
	← gtmanfred left	10:12
	→ martinjlowm joined	10:12
	→ kossy joined	10:12
	→ spacepluk joined	10:13
	→ gtmanfred joined	10:13
	→ angelsl joined	10:13
	← dobalina left	10:13
	← psyl0n left	10:14
	← cff left	10:15
	→ goTAN joined	10:15
	← fphilipe left	10:16
	← petey left	10:16
	→ LuxuryMode joined	10:17
	← hyperair left	10:17
	← kossy left	10:17
	→ kossy joined	10:19
	← amcsi_work_ left	10:19
	→ amcsi_work joined	10:19
	← weiyang left	10:21
	← martinjlowm left	10:21
	→ Bosox20051 joined	10:21
	→ dean[h] joined	10:22
dean[h]	Hi. I have a change in staging: renamed: web/assets/javascripts/lib/jQueryRotateCompressed.js -> web/assets/javascripts/lib/jQueryRotate.js <-- how do i undo that change	10:22
	← xjiujiu left	10:22
	→ xjiujiu joined	10:22
dean[h]	rename the file myself, or is there something i can do using git	10:23
	← dprophit left	10:25
	→ ewnd9 joined	10:26
moritz	dean[h]: a rename is just a git add and a git rm. 'git status' tells you how to undo both	10:26
	or if you want to discard all uncommitted changes, you can say 'git checkout .'	10:27
	← daito left	10:27
	→ diegoviola joined	10:29
	→ daito joined	10:30
	← xjiujiu left	10:31
	→ xjiujiu joined	10:32
	← goTAN left	10:33
	← krosmen32 left	10:33
	→ halu joined	10:33
	← introom left	10:34
	→ introom joined	10:35
	← jaeckel left	10:35
	← echo-area left	10:35
	phryk → whatvoiceplease	10:36
	whatvoiceplease → phryk	10:36
	→ davipt joined	10:36
	→ lite_ joined	10:36
	← xjiujiu left	10:38
	← luc____ left	10:38
	→ xjiujiu joined	10:38
	→ mklappstuhl joined	10:39
	← codehotter left	10:40
	→ codehotter joined	10:40
	→ gelardi_ joined	10:40
	← kossy left	10:41
	← marhaban left	10:41
	→ BlueOrg joined	10:42
	← BlueOrg left	10:42
	← Nizumzen left	10:42
	→ BlueOrg joined	10:42
	← BlueOrg left	10:42
	→ BlueOrg joined	10:43
	← BlueOrg left	10:43
	→ BlueOrg joined	10:43
	← halu left	10:43
	← macben left	10:44
	→ WaReZ joined	10:46
	→ dyng joined	10:47
	→ LBV_User joined	10:47
	← swaechter left	10:47
	← blinkdesign left	10:48
	→ swaechter joined	10:50
	→ jbarker joined	10:51
adsc	alright, i have now pushed --force my local reset master into origin, looks good so far	10:51
	if I do the reset on all workstations that have already pulled the faulty merge, I should be fine, right?	10:52
	→ lavong joined	10:52
adsc	as if it never happened	10:52
	← aspiers left	10:53
	← m0ltar left	10:54
	← jbarker left	10:56
	→ m0ltar joined	10:56
	← diegoviola left	10:56
	← Argure left	10:56
	→ psyl0n joined	10:58
	→ vurma joined	10:59
	→ dobalina joined	11:00
	→ DoniRS joined	11:00
	→ jaeckel joined	11:01
	→ PeterBooker joined	11:02
	→ Argure joined	11:02
	← dobalina1 left	11:02
	→ jbarker joined	11:02
	← daito left	11:03
	→ _5kg joined	11:04
	← mikecmpbll left	11:05
	← m0ltar left	11:05
	← vtunka left	11:05
	← jkremser left	11:05
	← MattMaker left	11:07
	→ MattMaker joined	11:07
	← jbarker left	11:07
	→ matteo joined	11:07
	→ mikecmpbll joined	11:09
osse	adsc: c'rec	11:11
	rewrite history soviet style	11:11
	→ SeptDePique joined	11:12
	← takamichi left	11:12
	→ tvw joined	11:13
	→ Kayra joined	11:13
	→ eletuchy joined	11:14
	→ m0ltar joined	11:14
	→ Nizumzen joined	11:15
	← Nizumzen left	11:15
	→ aspiers joined	11:15
	→ Arun joined	11:15
SeptDePique	hello all! pardon my stupid questions... if i set up a git repository and then commit with eclipse/egit...., is it possible to access this state of project at a future date?	11:15
GyrosGeier	sure	11:16
	that's the point of git	11:16
	← herdingcat left	11:16
SeptDePique	ok thx... i just wasn't sure, if i have to consider some configuration settings or so...	11:16
	← Gavrisimo left	11:17
	→ ben_alman joined	11:18
	← mklappstuhl left	11:18
	← Noctis_Lux left	11:20
GyrosGeier	SeptDePique, everytime you do a commit, that state is recorded, regardless of which tool you use on the higher level	11:20
SeptDePique	thx again	11:21
	→ mklappstuhl joined	11:22
	← ewnd9 left	11:23
	→ teran joined	11:23
	← ben_alman left	11:23
	→ takamichi joined	11:26
	→ goTAN joined	11:29
	← lite_ left	11:29
	→ skorgon_ joined	11:29
	→ jbarker joined	11:30
	→ MindfulMonk joined	11:30
	← adsc left	11:32
	→ canton7-mac joined	11:33
	← jalmeida left	11:34
	→ dmin joined	11:34
dmin	Hi, I'm trying to push a commit to the server (apache) on a windows machine in Git Bash... but I'm getting "http-push is not a git command"	11:35
	← dobalina left	11:36
dmin	does anyone know what could be the culprit?	11:36
	karupanerura → zz_karupanerura	11:36
	← josealobato left	11:36
	→ blinkdesign joined	11:36
	→ dougsland joined	11:40
	← mklappstuhl left	11:40
	→ pfFredd joined	11:41
	← BlueOrg left	11:41
	← vurma left	11:42
	← skorgon_ left	11:42
	→ Gavrisimo joined	11:43
	→ macben joined	11:43
	→ jkremser joined	11:43
	← jkremser left	11:43
	→ jkremser joined	11:43
	→ mklappstuhl joined	11:44
	← Watabou left	11:44
	← intellix left	11:45
	← dyng left	11:45
	→ dyng joined	11:46
arand	dmin: Seems to indicate that git isn't installed correctly, "git --exec-path" should point to a directory where the different git binaries are available.	11:46
	→ eijk_ joined	11:46
	← dbolser left	11:47
dmin	arand: git --version gives me : git version 1.8.5.2.msysgit.0	11:47
	issueing 'git http-push' gives the error saying its not a command	11:47
	is it not supported on windows?	11:47
	← mklappstuhl left	11:48
	→ eletuchy_ joined	11:48
	← moondoggy left	11:48
	← eletuchy left	11:48
	← eijk left	11:49
	← dyng left	11:50
	→ moondoggy joined	11:50
	← moondoggy left	11:51
	→ moondoggy joined	11:51
	← introom left	11:52
	→ adsc joined	11:52
	← psyl0n left	11:52
	→ fluxdude joined	11:54
lb1a	dmin, man git http-push say something about temporarily disabled	11:55
gitinfo	dmin: the git-http-push manpage is available at http://jk.gs/git-http-push.html	11:55
	← paulweb515 left	11:55
	→ psyl0n joined	11:58
	← psyl0n left	11:58
	→ psyl0n joined	11:58
dmin	lb1a: yes I read about the <7.16 version	11:58
	← LuxuryMode left	11:58
dmin	but the libcurl shipped with the git Windows download is 7.30	11:58
	→ ewet joined	11:59
	← marcjs left	12:00
dmin	on a Linux box with a low curl version git nicely says http-push is disabled ... but on the latest windows version, it just says "http-push is not a git command"	12:00
	the curl version is ok, so I'm puzzled	12:00
	← ewet left	12:01
	→ mklappstuhl joined	12:01
arand	dmin: And you have you checked the --exec-path?	12:02
	→ ewnd9 joined	12:02
	→ renatow joined	12:03
dmin	$ git --exec-path	12:03
	C:\Program Files (x86)\Git/libexec/git-core	12:03
arand	Chekc if http-push exists there.	12:03
	→ NET\|\|abuse joined	12:04
	→ simesy joined	12:04
NET\|\|abuse	hi folks, i'm joining a project and the other devs have the code for the site in the root of the git repository, so i checkout to ~/projectname and i have index.php directly in that directory.	12:04
dmin	hm, no git-http-push.exe	12:05
NET\|\|abuse	I need to add various assets and build scripts to automate deployment but I dont want to disrupt toher devs workflows untill i have to. how can i push the code into a subdirectory without modifying where they checkout from	12:05
	can i wrap this as a submodule?	12:05
	I've never done that before, so I'm not sure if that's the correct approach. if there's a better way please let me know?	12:06
	→ cmpera joined	12:06
dmin	should the latest Windows download include a git-http-push.exe ?	12:06
	← davipt left	12:07
	→ kriks joined	12:07
	← ferreum left	12:07
	← rvsjoen left	12:08
rawtaz	NET\|\|abuse: perhaps you can create a new branch, work on that, and puch it to the server for your deployment testing needs?	12:09
	→ tarkus joined	12:09
NET\|\|abuse	could i continue to merge changes from master even though the paths would be different?	12:10
	can you map a moved directory between branches like that?	12:10
	→ introom joined	12:10
	→ dmin_ joined	12:11
	← jargon left	12:11
dmin_	sorry, got disconnected. did I miss anything?	12:11
	← Kayra left	12:11
NET\|\|abuse	dmin_: omg, you would not believe what happend while you were gone!!!	12:11
	→ tgkokk joined	12:11
dmin_	lol	12:12
arand	dmin: no	12:12
	→ petey joined	12:12
dmin_	so is the git-http-push.exe not in the windows release deliberately ?	12:12
	← mklappstuhl left	12:12
rawtaz	NET\|\|abuse: yes you can merge changes from master to your custom branch	12:12
adsc	damn, it's exactly like described in the link jast posted...i went back to a commit before the merge, then pushed that with --force and thought all is well, but on the next workstation, when I did the same to the local repo, then pulled, it pulled in the initial commit of the branch that was erroneously merged	12:13
arand	(I would guess that the binary is supposed to be there, but I don't really know)	12:13
rawtaz	NET\|\|abuse: not sure what you mean by map a moved directory between branches	12:13
	→ Kayra joined	12:13
adsc	i guess the branches remain merged....	12:13
dmin_	hmm	12:13
adsc	this will be a nightmare to fix	12:13
NET\|\|abuse	ok, so master ->branch move all files into one subdirectory merge <- from master	12:13
	→ mklappstuhl joined	12:13
	← GeertV left	12:14
	→ rvsjoen joined	12:14
	→ GeertV joined	12:14
NET\|\|abuse	I dont like the idea of submodules for this, build and puppet scripts and things i want to add should be part of the code base, not some other project.	12:14
rawtaz	NET\|\|abuse: why do you need to move the files into a subdirectory? by creating a branch you get the isolation from others' work that you want for starters. the others wont see your work unless they look for it	12:14
	→ freeed joined	12:15
	← dmin left	12:15
	dmin_ → dmin	12:15
NET\|\|abuse	yeh, but it might take me 2 weeks to get this setup done, by then the changes on master will be substantial, i'd love to be able to keep up with changes on master	12:15
	→ afhammad joined	12:16
	← petey left	12:17
	→ Exancillatus joined	12:18
	→ ben_alman joined	12:19
rawtaz	NET\|\|abuse: you can regularly merge master to your branch to keep your branch/changes up to date with the rest of the team's work	12:20
	← Exancillatus left	12:20
rawtaz	consider this a feature branch or so	12:20
NET\|\|abuse	oh yeh, a feature branch for sure, that's what i mean, can i merge against the moved files though, that's where i'm worried	12:20
	→ ferreum joined	12:21
	→ dyng joined	12:21
rawtaz	i dont understand the question	12:22
NET\|\|abuse	sorry :)	12:22
	→ aprx joined	12:22
rawtaz	you can merge whatever you want, and if there's a conflict you will have to resolve that as a merge conflict	12:22
	→ jargon joined	12:23
rawtaz	if someone else made a change in a file that you have moved i guess that will result in a conflict. but im not sure, just a newbie like you :p	12:23
	hopefully someone else can verify/correct that	12:23
dmin	Ok... I've set up a special useraccount for pushing over ssh... apparently trying to use http dav push is asking too much ;-)	12:24
	works over ssh	12:24
	→ goTAN_ joined	12:24
	← goTAN_ left	12:24
	← ben_alman left	12:24
dmin	thanks anyways ;0)	12:24
	← goTAN left	12:24
	emmajane → [emmajane]	12:24
NET\|\|abuse	yeh, my question is what you said above. when they change the file in master but i've moved in branch and i try to merge, a conflict or will the move be tracked appropriately, is git that smart :)	12:24
	I guess i'll just try. #	12:25
	← dmin left	12:25
	→ skorgon_ joined	12:26
	← saustin left	12:26
rawtaz	i honestly dont know. but you can easily try it. make a dummy repo, make a file, create a branch, move the file in it and commit, checkout the other/first branch, edit the (not moved) file and commit, checkout the second branch again and try to merge from the first branch	12:26
	→ Nizumzen joined	12:27
	→ AzizLight joined	12:29
AzizLight	Hi everybody	12:29
	← ciampix left	12:29
	← cebor left	12:29
	← Raging_Hog left	12:29
	← afhammad left	12:30
	→ rethus joined	12:30
	→ the_cybersphinx joined	12:30
	← xintron left	12:30
	← jbarker left	12:31
	→ cebor joined	12:31
AzizLight	I think I found a bug in git, or it's a misunderstanding of how git-commit works: When I have no files staged, and I do `git commit ""` it stages all my tracked files before the commit. I would expect git to tell me that there are not files to commit...	12:31
	is that a known bug, and is there a workaround?	12:31
	← Nizumzen left	12:31
	→ xintron joined	12:31
	→ ben_alman joined	12:32
	← tgkokk left	12:33
	← JrCs left	12:33
	→ JrCs joined	12:34
	→ ariscop joined	12:34
	rawtaz &	12:34
	→ zequence joined	12:35
	← kcj left	12:35
	→ Noldorin joined	12:35
bremner	AzizLight: ooc, why are you passing "" to git commit?	12:36
	→ ewet joined	12:36
AzizLight	bremner: I am creating a wrapper bash function around git commit and sometimes it happens that git commit is called with ""	12:37
	← slyfox left	12:37
	→ bretzel joined	12:37
	← JohnBat26 left	12:38
	← nutron left	12:38
bremner	I guess it might be a bug. But I think correct behaviour would be to stop with an error, so maybe you should try to avoid calling with "".	12:39
	→ vurma joined	12:39
bremner	"" is not a valid filename, afaik	12:39
	[emmajane] → emmajane	12:39
	→ slyfox joined	12:39
	← slyfox left	12:39
	→ slyfox joined	12:39
	← Silmarilion left	12:39
	→ intellix joined	12:40
bretzel	Hi, I have a root-dir projects as one repo( remote ). but I have screwed one of my subdir ( sub-project ). I need 1: commit and push the other dirs, and then rollback back to the last-push for the screwed subdir(project)... ;-) cough!	12:40
	→ lite_ joined	12:40
	→ ph88 joined	12:41
ph88	how can i see which commit is tagged ?	12:41
jast	many commits may be tagged	12:42
	← Kayra left	12:42
ph88	yes how can i view the tags on a commit ?	12:42
	or rather a list of commits with their tags	12:42
AzizLight	bremner: I know that "" is not a valid filename, but the point is: an error should be thrown, files should not be staged... looks like a bug..	12:42
bremner	AzizLight: right. But your script also has a bug in it, and it might be faster to fix that...	12:43
	← swaechter left	12:43
	→ Droolio joined	12:44
	→ Silmarilion joined	12:44
AzizLight	bremner: I know. And I will fix it. Just wanted to report what I found. And it would be nice if that "bug" could be fixed.	12:44
	← skorgon_ left	12:44
bremner	AzizLight: sure, but if you really want to report a bug, send a mail to !list	12:44
gitinfo	AzizLight: [!mailing_list] The mailing list can be reached via [email@hidden.address] You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist	12:44
AzizLight	bremner: ok. Will do. Thanks	12:45
	← lite_ left	12:45
	← kriks left	12:46
	→ Exancillatus joined	12:46
	← MindfulMonk left	12:47
	→ nutron joined	12:51
	→ plasticle joined	12:51
	→ jasonroyle joined	12:51
arand	ph88: git log --decorate, or maybe something like "git tag \| while read -r tag; do git show --decorate "$tag"; done"	12:51
	← psoo left	12:52
	← jasonroyle left	12:52
	→ jasonroyle joined	12:53
	→ freestyl3r joined	12:53
	← simesy left	12:53
	→ doev joined	12:54
	→ marcjs joined	12:54
	← jasonroyle left	12:54
	→ jasonroyle joined	12:55
Agen_Terminator	Does anyone know how to use git commit on a specific file (as opposed to the last 2 commits for example)? Thanks	12:55
	← jasonroyle left	12:57
shruggar	Agen_Terminator, can you rephrase that?	12:57
	Agen_Terminator, I don't understand what you mean by "as opposed to the last 2 commits"	12:57
	← ewnd9 left	12:58
	→ dtrott joined	12:58
	→ sohail joined	12:59
	← sohail left	12:59
	→ sohail joined	12:59
	→ bristleback joined	12:59
bristleback	hi!	12:59
gitinfo	bristleback: 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.	12:59
bristleback	!backup	12:59
gitinfo	Worried about your data while trying out stuff in your repo? To back up commit history on all branches/tags: `git clone --mirror`. To backup everything, including work tree and staging area: `tar cf repo-backup.tar repodir`. Or do your experiment in a throwaway clone instead. See also http://sethrobertson.github.com/GitBestPractices/#backups	12:59
	→ MindfulMonk joined	12:59
bristleback	if I have made a commit into a public repository on github and now I have to modify that commit, how can I do it without forking the project locally?	13:00
	→ josealobato joined	13:00
shruggar	why do you need to modify the commit?	13:00
	← josealobato left	13:00
shruggar	and define "forking" in this context, as github tends to make people abuse the term	13:01
	→ cannonball joined	13:01
	→ jbarker joined	13:01
	← marcjs left	13:02
	← gypsymau1o left	13:02
	→ marhaban joined	13:04
lb1a	bristleback, !xy	13:04
gitinfo	bristleback: Woah, slow down for a bit. Are you sure that you need to jump to that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...	13:04
	→ smashwilson joined	13:06
bristleback	I have made a change in the documentation of cakephp and commited into their repository on github	13:06
	← jbarker left	13:06
bristleback	they made some comments and I need to add something to that commit	13:06
	← macben left	13:06
bristleback	how can I do it?	13:06
	← lazers left	13:06
	→ macben joined	13:07
ph88	thx arand	13:07
	← dsdeiz left	13:08
	← MindfulMonk left	13:08
	← adsc left	13:09
	← mikecmpbll left	13:09
	← cagedwisdom left	13:09
	→ mikecmpbll joined	13:09
ph88	what's the best way to take parts of a commit and move them to a new branch ?	13:09
arand	ph88: cherry-pick --no-commit and then edit as needed.	13:11
	← caasig left	13:11
ph88	cherry-pick xD	13:11
	→ rabbite joined	13:12
	→ vtunka joined	13:12
	→ Kayra joined	13:13
	→ yacks joined	13:15
	← JerryT left	13:16
	→ marcjs joined	13:17
	→ tgkokk joined	13:17
	← nutron left	13:17
	← Kayra left	13:18
	→ niatish joined	13:19
	← vtunka left	13:20
	→ nutron joined	13:20
	← facundor left	13:22
	← _5kg left	13:22
	← marhaban left	13:23
	→ thomas_b joined	13:23
bristleback	anyone any idea about my problem?	13:23
	→ carbon joined	13:24
	← carbon left	13:25
	← tgkokk left	13:26
	→ xlinkz0 joined	13:26
	← rfoust left	13:27
	→ josealobato joined	13:28
	← sohail left	13:28
	→ danicampa901 joined	13:29
	→ vtunka joined	13:30
	→ rudisimo joined	13:31
	← mitz left	13:31
	← bgerber left	13:32
	→ Sly1986 joined	13:32
	→ JohnBat26 joined	13:34
	→ ewnd9 joined	13:35
	← flaguy48 left	13:37
	→ flaguy48 joined	13:38
	→ llaskin joined	13:38
	→ bgerber joined	13:39
	← calcifea left	13:39
	→ swaechter joined	13:40
	← ahmedelgabri left	13:40
	→ calcifea joined	13:40
	→ Hounddog joined	13:40
	← cdleonard1 left	13:41
	→ cdleonard joined	13:41
	→ tgkokk joined	13:42
llaskin	https://gist.github.com/llaskin/4a17dcb4fa68fc8af1e7 I wrote a question up in my gist	13:42
	can anyoen help?	13:42
	→ ahmedelgabri joined	13:43
	← m0ltar left	13:43
	← ItSANgo left	13:44
	← iderik left	13:45
	→ m0ltar joined	13:46
MacGyver	If experimental only builds on master, i.e. if there's no changes being merged back, you can simply rebase experimental on top of the new master.	13:46
	← josealobato left	13:47
	→ fire joined	13:48
	→ ffio joined	13:48
llaskin	well how would I put experimental on top of the existing master?	13:49
	I am not "rebuilding my entire repo" here….trying to modify existing structure right?	13:49
	→ ItSANgo joined	13:49
	← Noldorin left	13:50
	→ diegoviola joined	13:50
	→ mizu_no_oto joined	13:50
	→ jbarker joined	13:51
	→ breakingthings joined	13:51
	← fire left	13:53
	→ Noldorin joined	13:53
	← ffio left	13:53
	→ tmcmahon joined	13:53
	→ dfanjul joined	13:54
	→ shinnya joined	13:55
	← _Vi left	13:55
	← Exancillatus left	13:55
	← tmcmahon left	13:55
	→ jerius joined	13:55
	← jbarker left	13:55
	← thrakcattack left	13:56
	→ autopilot joined	13:56
	→ wendellM joined	13:57
autopilot	hi, I've installed git with `make config ; ./configure ; make ; make install`, there doesn't appear to be an uninstall target... please could someone tell me how I remove git?	13:57
	→ thrakcattack joined	13:58
	← mizu_no_oto left	13:59
	→ skorgon_ joined	13:59
	← cmpera left	14:00
lb1a	bristleback, make a new commit?!	14:00
bristleback	no, not make a new commit, but modify the previous commit on github	14:01
	→ cmpera joined	14:01
lb1a	bristleback, a commit is not changeable	14:01
	→ sohail joined	14:02
	← sohail left	14:02
	→ sohail joined	14:02
	← psyl0n left	14:02
	← gelardi_ left	14:02
lb1a	bristleback, published commits should also never be rewritten (!rewrite).	14:02
gitinfo	bristleback: [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	14:02
	→ decotii joined	14:03
	← konadr left	14:03
	→ simesy joined	14:04
	→ konadr joined	14:04
	← konadr left	14:04
bristleback	but not rewriting public history because the commit was never merged with the repository	14:04
	but anyway	14:04
	I get the idea	14:04
	→ goTAN joined	14:04
	→ konadr joined	14:04
lb1a	bristleback, if you created a commit in a repo hosted on github, wheter due a push or in their webfrontend, this commit is public	14:05
	it's not a matter of merge	14:05
bristleback	ok	14:05
	thanks	14:05
lb1a	bristleback, create a commit that does the right thing, or maybe revert your 'wrong' commit previous to a brand new commit...	14:06
	→ hellopat joined	14:06
	← b3tamax left	14:06
bristleback	and how can I do that, revert the commit?	14:07
	that is actually what I'm trying to do	14:07
	← niatish left	14:07
lb1a	i doubt you can do it via the webgui. you'll have to clone it locally and use `git revert`	14:07
	man git revert	14:07
gitinfo	the git-revert manpage is available at http://jk.gs/git-revert.html	14:07
bristleback	ok	14:07
lb1a	maybe also !revert helps	14:07
gitinfo	That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?	14:07
lb1a	option a	14:08
bristleback	I now know what I have to do	14:08
	← Noldorin left	14:08
	← simesy left	14:08
	← ahmedelgabri left	14:09
MacGyver	llaskin: Look into git rebase and what it does. Also, you may want to look into hooks if you actually mean you want to automate the process.	14:10
	← ben_alman left	14:10
	→ JerryT joined	14:10
llaskin	MacGyver: nah not worried about automating this process	14:12
	← rethus left	14:12
	→ nilg joined	14:13
	← gwb3 left	14:13
	→ rurufufuss joined	14:14
cannonball	Hi all. I had what I think is a misconfiguration in my .git/config. Let me describe it, then I'll ask the question.	14:15
	→ Kayra joined	14:15
cannonball	Through some previous commandline incantations, I had a remote which had a non-default fetch setting. The default is:	14:16
	→ rfoust joined	14:16
	→ gwb3 joined	14:16
	→ _tpavel joined	14:16
cannonball	fetch = +refs/heads/:refs/remotes/NAME/	14:16
	But I had ended up with an actual branch name instead of the *.	14:16
	At some point, I did a git push --delete REMOTE BRANCH, and that branch no longer existed on the remote, but the .git/config still referenced it.	14:17
	← rurufufuss left	14:17
	→ shingshang joined	14:17
	→ mizu_no_oto joined	14:17
	→ Luciano joined	14:17
Luciano	hi	14:17
gitinfo	Luciano: 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.	14:17
Luciano	list	14:18
	→ juliohm joined	14:18
Luciano	I need a list film	14:18
	help me	14:18
	please	14:18
cannonball	Calling get fetch or git remote update ended up with a ssh hangup. I had to manually change the .git/config for it to work right. Was there a way I could have done this with the git cli instead of manually editing the .git/config?	14:18
jast	a what?	14:18
Luciano	tnx	14:19
	← Luciano left	14:19
	→ jbarker joined	14:22
	← gdogrady left	14:22
	→ tma996 joined	14:25
	→ str joined	14:25
	→ livingstn joined	14:26
	← goTAN left	14:26
	← Misan left	14:27
cmn	cannonball: you can use git config to remove particular lines, though I'm not sure it would have been more confortable	14:28
	→ zoig joined	14:28
cmn	but the ref being in a fetch refspec shouldn't have anything to do with an ssh hangup	14:28
	→ aChatir joined	14:29
cannonball	cmn: It complained that the remote branch didn't exist and just closed the connection.	14:29
	→ Destos joined	14:29
cannonball	So I guess my description wasn't exactly accurate.	14:29
	→ petey joined	14:29
cmn	that's something completely different from a hangup	14:29
	→ LuxuryMode joined	14:31
	→ GODDOG joined	14:31
	→ paulweb515 joined	14:31
	→ _Vi joined	14:32
	→ tonythomas01 joined	14:33
	→ psyl0n joined	14:33
	← GeertV left	14:33
	→ AaronMT joined	14:33
	← irqq left	14:36
	← skorgon_ left	14:37
	← dtrott left	14:37
	← LBV_User left	14:37
	→ Diemuzi joined	14:37
	gitinfo set mode: +v	14:37
	← paulweb515 left	14:38
	→ ehmatthes joined	14:38
	← GODDOG left	14:38
	→ GODDOG joined	14:39
	→ nZac joined	14:40
	← dyng left	14:40
	← nZac left	14:40
	← kermit left	14:40
	→ dyng joined	14:40
	← tonythomas01 left	14:42
	→ IrishPrime joined	14:42
	→ CheckDavid joined	14:44
	→ FortySix2 joined	14:45
	← mizu_no_oto left	14:46
	→ HelperW joined	14:47
	→ b3tamax joined	14:47
rawtaz	is there a way a move 0f4b138,6214afe,3c53773,acece25 (lines 11-8) to after/top of head other than cherry-picking? http://pastebin.mozilla.org/4012332	14:48
	← Cephalostrum left	14:48
	→ phile joined	14:48
rawtaz	i rebased master on line 20, but only want to grab the last four commits from the old "tree"	14:48
	← dyng left	14:48
	← ewet left	14:49
	→ dyng joined	14:49
	→ matheus23 joined	14:50
	← soee left	14:50
	← FortySix2 left	14:50
	→ paulweb515 joined	14:50
	→ FortySix2 joined	14:51
	→ Hounddog_ joined	14:51
	→ hyperair joined	14:51
	← llaskin left	14:52
	← Poly-C_atwork left	14:52
	← shinnya left	14:52
	← Hounddog left	14:52
	→ acharlieh joined	14:53
	→ rednaks joined	14:53
offby1	sure sounds like cherry pickign	14:54
rawtaz	yeah	14:54
	→ Bass10 joined	14:54
	→ kai_ joined	14:54
rawtaz	just curious if there's another way that is equally "simple"	14:54
	→ bhldev__ joined	14:54
offby1	probably not.	14:54
rawtaz	ok :)	14:54
	I WILL CHERRY PICK	14:55
	← nnyk_ left	14:55
rawtaz	thanks offby1	14:55
offby1	you could interactively rebase that branch onto the other, then delete the commits you don't want to keep, but that's probably harder	14:55
	← bhldev_ left	14:55
rawtaz	right. yeah sounds harder for sure	14:55
	← xjiujiu left	14:56
	→ dglambert joined	14:57
	← donmichelangelo left	15:00
	→ thiago joined	15:00
	← hyperair left	15:00
	→ donmichelangelo joined	15:01
	← sohail left	15:01
	← Macaveli left	15:01
	→ goTAN joined	15:02
	← makubi left	15:02
	→ nZac joined	15:02
	← swaechter left	15:02
	→ axl_ joined	15:03
	→ dsdeiz joined	15:04
	→ deanclkclk joined	15:04
deanclkclk	question	15:04
	← vince_prignano left	15:04
	← arvind_khadri left	15:04
deanclkclk	I am geting an error when I try to clone -> could not read username for "https://github.com": No such file or directory	15:05
	→ vince_prignano joined	15:05
deanclkclk	how can I change from git where the ssh directory is?	15:05
	← dean[h] left	15:05
cmn	you don't, you tell ssh, but first you would have to use the ssh transport	15:06
	→ makubi joined	15:06
deanclkclk	huh?	15:06
cmn	look at your url, what does it say to use?	15:06
	← WaReZ left	15:07
deanclkclk	I don't understand	15:07
	→ WaReZ joined	15:07
cmn	the url you just told us, it starts with https://	15:08
	← diegoviola left	15:08
	← autopilot left	15:08
	← axl_ left	15:08
	← vtunka left	15:09
	← mikecmpbll left	15:09
	← eletuchy_ left	15:09
	→ Cephalostrum joined	15:09
cmn	thus it's using the http(s) transport, not ssh	15:09
	→ orion joined	15:09
	→ Biopandemic2 joined	15:10
	← Biopandemic2 left	15:10
	→ Biopandemic2 joined	15:10
	→ irqq joined	15:10
	→ eletuchy joined	15:10
orion	Hi. I want to merge cmake_refactoring in to master~1. Is this possible?	15:10
	cmake_refactoring is a branch of mine, but the master branch changed a lot in one commit.	15:11
	However, if I do git diff master~ cmake_refactoring, the diff is clean and nice.	15:12
	→ ahmedelgabri joined	15:12
	← Kayra left	15:12
	→ ben_alman joined	15:12
orion	And I want to marge THOSE changes in. I don't want to resolve 1,000 merge conflicts introduced between master~ and master.	15:12
	→ xjiujiu joined	15:13
	→ Dougie187 joined	15:13
	← goTAN left	15:13
	← acharlieh left	15:14
	← eletuchy left	15:15
deanclkclk	so how would I be able to fix the problem?	15:15
	→ VictorCL joined	15:15
	Biopandemic2 → Biopandemic	15:15
	→ crazydiamond joined	15:15
cmn	either fix the issue with the credentials cache or curl or whatever is throwing that error, or use a ssh url if what you want is to use ssh to authenticate	15:16
	← dsdeiz left	15:16
	← jmux left	15:17
	→ mrbubbles joined	15:17
	→ acharlieh joined	15:18
	→ Not_Whatever joined	15:18
	← GT4066 left	15:18
	← Tajha left	15:19
	→ Kayra joined	15:19
	→ Foldager joined	15:19
	← BeerSerc_ left	15:20
	→ vtunka joined	15:21
	← jnavila_ left	15:21
	← charon left	15:21
	→ BeerSerc joined	15:22
	← papegaaij left	15:23
	→ ErKa joined	15:23
	→ hyperair joined	15:23
	→ sec^nd joined	15:24
sec^nd	I'm working on a rather medium sized projected for a company so I can't give many details but when I try to merge I always have merge conflicts. How do I fix this?	15:25
	← Diemuzi left	15:25
	→ jnavila joined	15:27
bremner	sec^nd: come up with a minimal example that doesn't contain proprietery information.	15:27
	or hire a consultant under NDA ;)	15:27
	→ skorgon_ joined	15:27
	← b3tamax left	15:28
sec^nd	bremner: I pull master, branch from that, they do a bunch of stuff elsewhere in the code, I try to do a merge with master before making a pull request.	15:28
	bremner: not sure what happened but everytime I get a merge conflict with their code not my new stuff which is fine, just stuff they pushed into master	15:28
	bremner: if master has been updated why do I get merge issues / conflicts?	15:28
	→ dtrott joined	15:28
	→ Exancillatus joined	15:28
bremner	that is pretty odd. Do you get the same problem if you rebase on master?	15:28
	← Exancillatus left	15:29
	← mklappstuhl left	15:29
	→ el3ctron_ joined	15:29
	→ bwreilly joined	15:30
	→ bambams_ joined	15:30
	→ _5kg joined	15:30
	← jnavila left	15:30
	→ mangoMan joined	15:30
mangoMan	EVERYONE SEND A MESSAGE TO THAT FAG METALEER HE IDLES IN ##PHYSICS AND #DEFOCUS HE IS A CANCER ON THIS FUCKING NETWORK shgh	15:31
	EVERYONE SEND A MESSAGE TO THAT FAG METALEER HE IDLES IN ##PHYSICS AND #DEFOCUS HE IS A CANCER ON THIS FUCKING NETWORK xdhdjg	15:31
	EVERYONE SEND A MESSAGE TO THAT FAG METALEER HE IDLES IN ##PHYSICS AND #DEFOCUS HE IS A CANCER ON THIS FUCKING NETWORK srfhdtj	15:31
	EVERYONE SEND A MESSAGE TO THAT FAG METALEER HE IDLES IN ##PHYSICS AND #DEFOCUS HE IS A CANCER ON THIS FUCKING NETWORK asryyut	15:31
	← mangoMan left	15:31
deanclkclk	cmn: I think the problem is git trying to access -> user/.ssh/? for pub key	15:31
	→ iderik joined	15:31
deanclkclk	the problem is	15:31
cmn	it's not	15:31
	ChanServ set mode: +o	15:31
	← dyng left	15:31
sec^nd	...	15:31
	bremner set mode: +b	15:31
cmn	git isn't trying to do anything with ssh	15:32
sec^nd	bremner: what is the difference between rebase and merge?	15:32
	→ Tajha joined	15:32
deanclkclk	the problem is "usr" directory has security restrictions	15:32
	→ dyng joined	15:32
	← skorgon_ left	15:32
bremner	sec^nd: explained in chapter 3 of !book	15:32
gitinfo	sec^nd: 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	15:32
	← tgkokk left	15:33
bremner	sec^nd: in fact your problem could be caused by too much rebasing either by you or by the others.	15:33
	← h0bbit_ left	15:33
	→ skorgon_ joined	15:33
sec^nd	bremner: what book?	15:34
bremner	the one that fsbot mentions just above	15:34
sec^nd	oh	15:34
	→ tgkokk joined	15:34
bremner	err, gitinfo. wrong bot	15:34
	→ becks__ joined	15:34
	ChanServ set mode: -o	15:35
	→ Swayz joined	15:35
	← zoig left	15:35
	→ goTAN joined	15:35
	← xequence left	15:36
bremner	sec^nd: is there any chance that your collaborators are force pushing the repo after rewriting history?	15:37
	← dyng left	15:37
	← bwreilly left	15:37
	← rob_w left	15:37
	→ DolphinDream joined	15:38
	→ mizu_no_oto joined	15:38
	← Exdaix left	15:38
	← LuxuryMode left	15:38
	→ Exdaix joined	15:38
	← lb1a left	15:38
	→ tplaner joined	15:39
	← kai_ left	15:40
	← hyperair left	15:40
	→ eletuchy joined	15:40
	→ kermit joined	15:41
	→ fire joined	15:41
	→ hyperair joined	15:41
	→ ojacobson joined	15:42
sec^nd	bremner: I'm not sure, how would I know if they are doing this?	15:42
	→ sr105 joined	15:42
bremner	sec^nd: when you fetch, it should mutter about "forced update"	15:43
	← ben_alman left	15:43
sec^nd	bremner: I'm not getting those issues, just a bunch of merge conflicts	15:43
	← zackiv31 left	15:43
	← kbart left	15:43
	→ ben_alman joined	15:44
	→ rgarrigue joined	15:44
	← psyl0n left	15:44
rgarrigue	Hi guys	15:44
gitinfo	rgarrigue: 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.	15:44
	→ swaechter joined	15:44
	← nilg left	15:45
	← Cyb3r-Assassin left	15:45
	→ zackiv31 joined	15:45
	← macben left	15:45
rgarrigue	I want to get a github repo at a specific state, knowing I got the commit SHA1	15:45
bremner	sec^nd: check that the hashes match on "common" commits. If not, your collegues have been naughty.	15:45
ojacobson	rgarrigue: goals unclear, can you back up a bit and tell us the bigger picture?	15:47
	→ Cyb3r-Assassin joined	15:47
	← Targen left	15:47
rgarrigue	Well, I made a pull request for a small feature on a github repo, the guys say "try again", I want the source to resume work	15:48
	Here is the pull req: https://github.com/ansible/ansible/commit/1251327c7327ee771f6543a0c14e11b4ee4beff4	15:48
	→ Exancillatus joined	15:49
	→ Nizumzen joined	15:49
	→ sohail joined	15:49
	← sohail left	15:49
	→ sohail joined	15:49
	← dalias left	15:49
	→ dalias joined	15:49
rgarrigue	ojacobson: I confess I'm a bit lost in git, I read stuff about fetch, checkout etc... I'ld like to understand the "normal" way to go if there's one	15:50
ojacobson	rgarrigue: I still don't understand your goals	15:50
	← skorgon_ left	15:50
sec^nd	bremner: will check	15:50
ojacobson	so I can't responsibly offer any suggestions :)	15:50
	You've got some changes (in a PR, and on a branch in your own repo), and ansible has its own set of branches in their repo; what do you want to do with that information?	15:50
	"I want the source to resume work" doesn't tell me much; the ansible team presumably does things on their own schedule	15:51
	→ gnu-man joined	15:52
	← vtunka left	15:53
	→ zatan joined	15:54
zatan	hey how can I check from where my current branch was branch of ?	15:54
	← Exancillatus left	15:54
	← ahmedelgabri left	15:54
ojacobson	Nil.	15:54
	Git doesn't have any concept of "branched from" internally.	15:54
	← FortySix2 left	15:55
	← nwertman left	15:55
	→ bwreilly joined	15:55
ojacobson	You can use the "upstream"/"tracking" info to track that locally (and automate things like merging/rebasing with upstream)	15:55
	→ ahmedelgabri joined	15:55
zatan	hmm	15:56
	→ Macaveli joined	15:56
rgarrigue	ojacobson: hm... I used github webui to edit one file from the ansible devel. It created a fork in my own home repo. Ok so far. I made my modification and a pull request. It resulted in the commit above, which is refused by mdehaan. Ok. If I want to rework my modification, do I have to clone ansible:devel, copy/paste again then work, or can I clone/checkout/whatever ansible:devel + my initial modification ?	15:56
	→ macben joined	15:57
ojacobson	You can continue making changes on the branch you already created in your fork	15:57
rgarrigue	ojacobson: the pull request removed that	15:57
ojacobson	Github will automatically update the pull request whenever the source branch changes	15:57
	← hyperair left	15:57
	→ lite_ joined	15:57
rgarrigue	I mean, I can't find it in my home repo i the WebUI	15:58
	→ hyperair joined	15:58
rgarrigue	oh, hell	15:58
	← VictorCL left	15:58
	← becks__ left	15:58
rgarrigue	didn't saw that, there's a branch:patch-1	15:59
	→ becks__ joined	15:59
	← aChatir left	15:59
	← deanclkclk left	16:00
	← M00sL0gB0t left	16:01
	→ M00sL0gB0t joined	16:01
mrbubbles	hey guys - i've been asked to use GIT on a wordpress project. I've cloned the existing repo and into a local branch called development (eventually I'll be pushing to a remote branch called development too). now in there is the file system to the existing site. I need to replace it with my new site file system (a newer version, updated config etc). What's the correct way of doing this? Do I create a new local branch and delete everything from the folder and	16:02
	my files?	16:02
	← Norm_ left	16:03
	→ lifelike joined	16:03
	← matheus23 left	16:04
	← goTAN left	16:04
	→ FortySix2 joined	16:04
	← dougsland left	16:05
	→ tholmes joined	16:05
	→ kukks joined	16:05
	→ mklappstuhl joined	16:06
	→ chrisshattuck joined	16:07
mrbubbles	apologies for the newb question btw	16:07
	← ThomasLocke left	16:07
ojacobson	mrbubbles: git is, fundamentally, snapshot based	16:07
mrbubbles	okay	16:08
	just got to that bit in my git book	16:08
	← cbreak-work left	16:08
ojacobson	so if you have a complete tree that you want to check in as the "successor" to some existing commit, you can definitely do that by nuking the work tree and replacing it with your own tree	16:08
	it's unusual to do it that way, but there are good uses for it	16:08
	(importing vendor tarballs is the usual case :)	16:08
mrbubbles	is there a better/different way?	16:08
	← introom left	16:09
	→ LuxuryMode joined	16:09
ojacobson	"Better" is subjective	16:09
mrbubbles	hehe true	16:09
	→ senj joined	16:10
ojacobson	For existing projects it's a lot more common to make your changes directly on the work tree git creates, though, rather than importing some modified tree	16:10
mrbubbles	okay	16:10
	← swaechter left	16:10
mrbubbles	so in my development branch I could just delete the contents of the folder, unzip the wordpress zip and stage the changed files	16:10
	← h_kon left	16:10
	← ferreum left	16:11
	→ ngladitz joined	16:11
ojacobson	So long as you leave .git untouched, yes	16:11
	man git-add and have a look at the --all option	16:11
gitinfo	the git-add manpage is available at http://jk.gs/git-add.html	16:11
	→ Gamecubic joined	16:11
mrbubbles	yup I know to stage all the new files	16:11
	→ b3tamax joined	16:12
	← macben left	16:12
mrbubbles	I guess I was just unsure what to do with the old ones. but, as you say, it's snapshot based so if I rm them then when I add the new ones it will only stage the new	16:12
	← jceb left	16:12
	← hyperair left	16:13
	← sr105 left	16:13
YoungFrog	don't forget to "git status" to check what you're going to commit, because the index most likely contains the older version, so you'll have to remove those.	16:13
	→ Gonzo89 joined	16:14
mrbubbles	why would i need to remove them? If I am clearing the whole folder first?	16:14
	→ kotk joined	16:14
	→ lethjakman joined	16:15
	← Baluse left	16:16
	→ m8 joined	16:16
	→ FortySix_ joined	16:16
	→ Baluse joined	16:17
	← thomas_b left	16:18
	→ thomas_b joined	16:18
	→ dougsland joined	16:19
	← juliohm left	16:19
	→ saustin joined	16:20
	← FortySix2 left	16:20
YoungFrog	I mean remove from the index. Clean the folder, import your stuff and see what git status says.	16:20
	→ KeiKun joined	16:22
	← b3tamax left	16:22
mrbubbles	ah will do, cheers :D	16:22
	→ infynyxx joined	16:22
ojacobson	'git add --all :/' will make the index exactly match the work tree, other than untracked ignored files	16:23
	→ lavaman joined	16:23
ojacobson	easier to do that than to try to pick out the right sequence of adds and removes, if all you want is to stage a tree verbatim :)	16:23
	← FortySix_ left	16:23
	→ b3tamax joined	16:23
	→ FortySix2 joined	16:24
	→ Exancillatus joined	16:24
	→ Pullphinger joined	16:25
mrbubbles	okay so I checked out remote branch master to local master, and development to local development	16:25
	← Etheco- left	16:25
mrbubbles	while on master all the files are there, i switch to development using git checkout, delete all the files and directories, switch back to development and all of the files are gone	16:26
	shouldn't they still be there if I've branched it ?	16:26
	← KeiKun_ left	16:26
ToxicFrog	mrbubbles: I think you mean "switched back to master"	16:27
	And git will try to "float" uncommitted changes in the working copy as you switch branches.	16:27
	← Macaveli left	16:27
mrbubbles	sorry yes I meant back to mater	16:27
ToxicFrog	A "git status" should show all the stuff you removed as uncommitted deletes.,	16:27
mrbubbles	Ah so i should commit that branch before I switch back	16:27
ToxicFrog	Yeah, if you actually want to delete them from development, delete them and commit the deletes	16:28
	If you just deleted them for fun and want them back, git reset to restore them before or after switching branches.	16:28
	← ahmedelgabri left	16:29
	→ macrover joined	16:29
	← dglambert left	16:30
	← mlechner left	16:30
	→ joelm joined	16:31
	← noname001__ left	16:32
	← konadr left	16:32
mrbubbles	ToxicFrog: okay think I'm getting my head around it, thank you sir	16:32
	→ psyl0n joined	16:33
	← eletuchy left	16:33
	→ eletuchy joined	16:33
	→ deanclkclk joined	16:34
	→ delboy1978uk joined	16:34
	← rgarrigue left	16:34
	← HelperW left	16:35
	← ErKa left	16:35
	→ introom joined	16:35
delboy1978uk	I'd like to clone a git repo from the first commit - is this possible or do i need to clone then reset?	16:35
	→ madewokherd joined	16:35
	← crazydiamond left	16:36
	→ deanclkclk_ joined	16:36
EugeneKay	clone has a --depth parameter(which you should never, ever use because) which does the opposite of what you're asking for.	16:36
	But usually it's the latest commit you're interested in, not the first. !xy ?	16:37
gitinfo	Woah, slow down for a bit. Are you sure that you need to jump to that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...	16:37
	← DoniRS left	16:37
	← gua_ould left	16:38
	← deanclkclk left	16:38
	deanclkclk_ → deanclkclk	16:39
	← kukks left	16:39
	← introom left	16:39
	← hahuang65 left	16:40
	→ koen_serry joined	16:41
mrbubbles	if I do git commit -a shouldn't that add any unstated files?	16:41
grawity	no.	16:41
ToxicFrog	mrbubbles: no.	16:41
mrbubbles	ah no, just stages tracked files but doesn't track new files	16:41
Dougie187	Only tracked unstaged files.	16:41
mrbubbles	GOT IT	16:41
	→ MistuhKurtz joined	16:41
	← Balliad left	16:44
	← saustin left	16:44
	← smef left	16:45
	← jkremser left	16:45
	← gnu-man left	16:46
	← Kayra left	16:46
	← ehmatthes left	16:46
	→ gnu-man joined	16:47
	← Sonderblade left	16:48
	→ King_Hual joined	16:48
	← arand left	16:48
	← makubi1 left	16:50
	→ doener joined	16:50
shruggar	is there a command for "don't actually switch to this revision, just tell me if any files in my work-tree would be overwritten if I did"?	16:53
ojacobson	git diff <commit-ish>	16:53
	--stat, --name-only, whatever	16:53
	→ iceblurgt1 joined	16:53
	→ Targen joined	16:54
	← Mamoru left	16:54
	← senj left	16:54
	← petey left	16:54
shruggar	that only tells me about tracked files, afaik	16:54
	→ petey joined	16:54
	← bwreilly left	16:55
	→ ahmedelgabri joined	16:56
	→ lazers joined	16:56
	← m8 left	16:56
	→ petey_ joined	16:57
	← lethjakman left	16:57
	← petey left	16:57
	← mishok13 left	16:57
	→ macben joined	16:57
	← Gonzo89 left	16:57
	→ Targen_ joined	16:58
	← Targen left	16:59
	→ mikecmpbll joined	16:59
	→ Kayra joined	16:59
	← tma996 left	16:59
	→ m8 joined	16:59
	← eletuchy left	16:59
	→ eletuchy joined	17:00
	← kermit left	17:00
	← macben left	17:02
	→ revyd joined	17:02
	→ kermit joined	17:02
mewph	Hello	17:03
	Anybody have any problems with creating a repo with git-gui on windows? It doesn't show up in the windows explorer at all	17:03
CareBear\	shruggar : I don't think so	17:03
	← Cephalostrum left	17:03
	→ SocalNick joined	17:03
CareBear\	mewph : I think you might need to create the directory yourself first	17:03
mewph	CareBear: I created the directory, and it shows up in git-bash and whatnot. but it's not showing in windows explorer and I can't seem to do much with it.	17:04
	→ dreiss joined	17:05
shruggar	is the status code output by --name-status always a single character?	17:05
	→ ErKa joined	17:05
	← ewnd9 left	17:06
lov	mewph: see if it's marked as hidden or something silly	17:06
	← Silmarilion left	17:06
mewph	lov: How?	17:06
lov	try navigating to the expected file path with explorer, see what happens	17:06
	→ Silmarilion joined	17:06
lov	uh, explorer settings, view hidden files?	17:06
	do a google search for how to see hidden files/folders on windows	17:07
	on your OS version in particular	17:07
mewph	lov: I did that, it says path does not exist	17:07
	← infynyxx left	17:07
ToxicFrog	mewph: are you sure you created it where you think you did?	17:08
	← Silmarilion left	17:08
	← MattMaker left	17:08
	→ kukks joined	17:08
mewph	It's quite bizarre	17:08
	→ pauld_ joined	17:08
mewph	ToxicFrog: It shows up in GIT-GUI, but not in windows explorer.	17:08
	→ Silmarilion joined	17:08
	→ ewnd9 joined	17:08
mewph	Git-gui reports it in C:\. When I use properties on it from the file selection in git-gui, it shows up in C:\. However, I can't change anything about it. It'll say file not found.	17:08
ToxicFrog	...as in, you've versioned C:\? Or as in you have C:\subdirectory\.git?	17:09
	← decotii left	17:09
mewph	shows up in git-bash when I navigate to /c and ls -la	17:09
	ToxicFrog: C:\subdir\.git yes.	17:09
	← ahmedelgabri left	17:09
	→ MattMaker joined	17:10
mewph	Why?	17:10
	← intellix left	17:10
mrbubbles	if I have a folder set to ignore in git ignore, then I use git checkout to switch folders, then switch back again, if the folder that i've ignored isn't in the branch that i switch to will it be deleted?	17:10
	← fire left	17:11
	← tplaner left	17:11
	← lavaman left	17:13
ToxicFrog	mewph: just wondering. So "ls c:/subdir/" works in gitbash but "dir c:\subdir\" doesn't in cmd?	17:14
mrbubbles	okay thats a yes by the looks of it	17:14
	← dreiss left	17:14
mewph	but dir C;\subdir\ does not in cmd	17:14
	ls /c/subdir works in git-bash	17:14
	← eletuchy left	17:14
	→ eletuchy joined	17:15
ToxicFrog	I assume the ; was a typo.	17:15
	That is weird.	17:15
	→ lavaman joined	17:15
	← gnu-man left	17:15
ToxicFrog	And c:/ in gitbash shows up normally, it hasn't gotten /c symlinked to something other than windows C:\ or the like?	17:15
mewph	yes	17:16
	→ kevwil joined	17:16
ToxicFrog	I've got nothing. Good luck.	17:16
mewph	Nope, /c and C:\ show up properly both places	17:16
	It's really quite bizarre.	17:16
	Okay thanks	17:16
	← jbarker left	17:16
ToxicFrog	(I haven't used gitbash in ages, I just have cygwin installed on the windows machine)	17:16
	→ gnu-man joined	17:17
	← xjiujiu left	17:18
	→ goTAN joined	17:20
CareBear\	don't use windows	17:20
	→ senj joined	17:21
	→ m0viefreak joined	17:21
	→ Panicg joined	17:21
	→ jbarker joined	17:21
	← jbarker left	17:22
	→ macben joined	17:22
	→ bwreilly joined	17:24
	→ bbloom joined	17:24
	← el3ctron_ left	17:24
	← psyl0n left	17:25
	→ rethus joined	17:26
	→ el3ctron_ joined	17:26
	→ imachuchu joined	17:27
	← koen_serry left	17:27
	← macben left	17:28
	← juril left	17:28
	→ Norm_ joined	17:29
	→ mishok13 joined	17:29
	← donmichelangelo left	17:30
	← Octalot left	17:30
	→ shuaib joined	17:30
	← fRAiLtY- left	17:30
	→ donmichelangelo joined	17:30
	← mikecmpbll left	17:31
	→ Hanumaan joined	17:31
	← zatan left	17:31
	→ DrShoggoth joined	17:32
Hanumaan	when I am trying to push getting following error "PUT error: curl result=22, HTTP code=403" . how to resolve ?	17:32
	→ pheaver joined	17:32
	→ petertretyakov joined	17:33
	← petey_ left	17:34
	→ starkhalo joined	17:34
EugeneKay	That HTTP code corresponds to Forbidden; I suspect you don't have permissions to push to that repo / authentication is not set up correctly.	17:34
	But it quickly becomes a problem of "what are you trying to push to?"	17:34
	→ petey joined	17:34
	← petey left	17:35
	→ petey joined	17:35
	→ h0bbit_ joined	17:36
	← MattMaker left	17:36
	→ HelperW joined	17:36
	← kpreid left	17:36
	← Dougie187 left	17:36
	→ MattMaker joined	17:37
	→ hahuang65 joined	17:37
	← mizu_no_oto left	17:38
	→ infynyxx joined	17:38
	← goTAN left	17:38
Hanumaan	EugenKay, got it on the repo Limit was not set properly ..	17:38
JosephFerano	I'm having an issue with git subtree, I make a change in the parent project, push it up to the sub-project, but then the parent project doesn't have that commit in its history when I try to do a split, so what ends up happening is that the remote rejects my changes because it says I'm behind	17:38
Hanumaan	I mean in the apache file ..	17:38
	← shruggar left	17:38
	→ namita joined	17:39
	← Sly1986 left	17:39
	← gain left	17:39
	→ jceb joined	17:39
	← gnu-man left	17:40
	← freimatz_ left	17:40
	→ Shusshu joined	17:40
	→ gnu-man joined	17:41
	← eletuchy left	17:41
JosephFerano	anyone here familiar with git subtree?	17:41
	→ lethjakman joined	17:42
	→ eletuchy joined	17:42
	← rethus left	17:42
	← petey left	17:42
_ikke_	nope, sorry	17:42
milki	why would the sub-project need to know about the parent project?	17:43
	→ renatow_ joined	17:43
	→ petey joined	17:43
	← p4denis left	17:43
	→ veebers joined	17:43
JosephFerano	milki: the parent project is not keeping track of when it has pushed something itself to the sub-project by using git subtree split	17:43
	→ wilx joined	17:44
	→ p4denis joined	17:44
	→ Apocalisp joined	17:44
	→ ZaeDrake joined	17:45
	← shingshang left	17:46
	→ petey_ joined	17:46
	← renatow left	17:46
	← petey left	17:46
	← rabbite left	17:46
	→ gdogrady joined	17:47
	← infynyxx left	17:47
	→ mmcdaris joined	17:47
	← bristleback left	17:47
JosephFerano	it really sounds like I should just switch to submodules...	17:47
	→ FatDarrel joined	17:48
	← shuaib left	17:48
	→ skorgon_ joined	17:48
	emmajane → ej-webinar	17:48
	← p4denis left	17:49
	→ yhager joined	17:50
	→ JesseCW joined	17:50
	JesseC → Guest94198	17:50
	← JesseCW left	17:50
	← KeiKun left	17:50
	← macrover left	17:50
	→ KeiKun joined	17:51
	← tvw left	17:52
	→ goTAN joined	17:52
	→ hyperair joined	17:52
	→ dreiss joined	17:53
	← thiago left	17:54
	→ jbarker joined	17:54
	← petey_ left	17:55
	→ phipes joined	17:56
milki	JosephFerano: that decision should be better informed before you try	17:57
	JosephFerano: you need to find out if the _workflow_ of subtree and submodules work for you	17:57
	← WaReZ left	17:57
milki	if you dont follow the workflow, those two things arent for you	17:57
JosephFerano	milki: I found out that there's a bug with subtree with the version of git I was using, so I'm removing and installing it through brew	17:57
milki	JosephFerano: !subproject	17:57
gitinfo	JosephFerano: [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.	17:57
JosephFerano	I'll look into git slave, since I haven't heard of that option	17:58
	I've researched submodule and subtree and subtree seems like the ideal way to handle this, but it was just giving me trouble	17:59
	→ shuaib joined	17:59
	← shuaib left	18:00
	→ aard_ joined	18:00
	← vurma left	18:00
	← phipes left	18:00
	→ ferreum joined	18:00
	→ phipes joined	18:01
	→ veebers_ joined	18:01
	← veebers left	18:01
	← veebers_ left	18:01
	← kermit left	18:02
	→ infynyxx joined	18:03
	← skorgon_ left	18:03
	← cmpera left	18:03
	← Nizumzen left	18:03
	← Shusshu left	18:03
	← Swayz left	18:04
	← mmcdaris left	18:04
	→ dav joined	18:04
	← Norm_ left	18:05
	← gnu-man left	18:05
	→ mmcdaris joined	18:06
gmac	!gitslave	18:06
gitinfo	gitslave (http://gitslave.sf.net) is useful to add subsidiary git repositories to a git superproject when you control and develop on the subprojects at more or less the same time as the superproject, and furthermore when you typically want to tag, branch, push, pull, etc. all repositories at the same time.	18:06
	→ cmpera joined	18:06
	→ veebers joined	18:06
dav	is there a simpler way of doing: git diff `git rev-list br1..br2 \| tail -1`~ br2	18:07
	in other word, diff between the last common commit of br1/br2 and br2's HEAD	18:07
	→ gnu-man joined	18:07
	← FortySix2 left	18:08
	← Kayra left	18:08
ojacobson	'git diff $(git merge-base a b) b'	18:09
	← xlinkz0 left	18:09
	ojacobson checks if diff accepts the ... notation	18:09
	← takamichi left	18:09
	← jbarker left	18:09
ojacobson	ah, it does	18:09
	'git diff a...b'	18:09
	(which is, per man git-diff, equivalent to what I said.)	18:09
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	18:09
	→ jbarker joined	18:09
	→ eletuchy_ joined	18:09
	← revyd left	18:09
	→ roymiloh joined	18:10
	→ revyd joined	18:10
	← yacks left	18:10
	← eletuchy left	18:10
	→ yacks joined	18:11
	← canton7-mac left	18:12
Hello71	!dots	18:12
gitinfo	A..B = stuff that happened between A and B (if A and B are related; otherwise refer to the more technical definition in "man gitrevisions"), A...B = (a) in history viewers: stuff that isn't in both A and B yet; (b) in "git diff": stuff that happened in B since the two diverged. "master.." is the same as "master..HEAD" and "..master" is the same as "HEAD..master", and so forth.	18:12
	→ WaReZ joined	18:12
	← jbarker left	18:12
	← SocalNick left	18:13
	→ bwreilly_ joined	18:14
	← eletuchy_ left	18:14
	← revyd left	18:15
	→ ecky--ptang-zoob joined	18:15
	← bwreilly_ left	18:15
	← dreiss left	18:15
	→ Norm_ joined	18:15
	← ZaeDrake left	18:15
	→ bwreilly_ joined	18:15
	← kevwil left	18:16
	← bwreilly left	18:16
ecky--ptang-zoob	I just learned plain GIT and learned how to commit stuff. But where does it commit?! Shouldn't there be an ftp server or some server where changes and new files get stored?!	18:16
	← lavong left	18:16
GyrosGeier	no	18:16
ojacobson	Git commits to the repository you run 'git commit' in.	18:16
GyrosGeier	it's all local	18:16
	→ gasbakid joined	18:17
	← ShooterMG left	18:17
ojacobson	If you want to share those commits with others, you have to arrange for others to be able to fetch them, either from your local repo or from some shared "server" repo you push to	18:17
GyrosGeier	which is great when working on a plane	18:17
ecky--ptang-zoob	if it's local, why is it needed? I need multiple to work on it obviously	18:17
ToxicFrog	ecky--ptang-zoob: each git repo is self-contained. When you "git commit" the commits are stored in the .git directory in your current repo.	18:17
	← hyperair left	18:17
ToxicFrog	to share those commits with others, you use "push" and "fetch" to sync them with other repos.	18:17
ecky--ptang-zoob	so if I co-program with someone on the same project, but he's on another PC, how do I get my changes there?	18:18
	→ SocalNick joined	18:18
	→ primitive joined	18:18
_ikke_	ecky--ptang-zoob: usually through some shared repo	18:18
	← deanclkclk left	18:18
ecky--ptang-zoob	on github?	18:18
ojacobson	Either he fetches from you, you push to him, or you push to a "shared" repo somewhere (eg. github, your own git server, whatever) and he fetches from it.	18:18
	!workflow	18:18
gitinfo	Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	18:18
	→ Swayz joined	18:19
ojacobson	!gcs	18:19
gitinfo	[!concepts] "Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs.html	18:19
	← dfanjul left	18:19
	→ kermit joined	18:19
ecky--ptang-zoob	On github it costs money to make a private repository. So it would be possible for me to create a repository locally on my own pc where other people have access to?	18:20
EugeneKay	ecky--ptang-zoob - !gitolite works great on a VPS; or Bitbucket has free private repos(with caveats)	18:20
gitinfo	ecky--ptang-zoob: 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/	18:20
	→ macben joined	18:21
	→ sealion joined	18:21
ecky--ptang-zoob	ok thx for the links. I'll read them. So github is maybe not so ideal?	18:21
	→ roland_ joined	18:23
	→ blinkdes_ joined	18:23
ecky--ptang-zoob	This link is fuxored though: http://gitolite.com/gcs.html	18:23
	← ben_alman left	18:23
	→ takamichi joined	18:23
grawity	Works fine, it just flips to "slideshow" mode when loaded.	18:23
	← fluxdude left	18:23
	→ nwertman joined	18:23
	← infynyxx left	18:23
grawity	Use ←/→ or PageUp/PageDown	18:24
	→ charon joined	18:24
	← kukks left	18:25
	→ rabbite joined	18:25
relipse	what does (modified content) mean and why can't i add it to staging (it is a submodule)	18:25
	→ dreiss joined	18:25
grawity	the submodule has uncommited changes inside	18:25
ecky--ptang-zoob	oh right, got it	18:25
relipse	it is a configuration file, is it better to move that outside the submodule directory?	18:25
grawity	so you'll have to commit them inside the submodule, and add the new commit to the parent	18:25
	if it's a config file, better don't track it at all	18:26
	→ Cephalostrum joined	18:26
grawity	track a .example or .dist copy of it instead	18:26
relipse	oh so add it to the ignore	18:26
	← kermit left	18:26
	← cmyers left	18:26
	→ eletuchy joined	18:26
	← Foldager left	18:26
	← lite_ left	18:26
	← Cephalostrum left	18:27
	→ hyperair joined	18:27
	← blinkdesign left	18:27
	← m8 left	18:27
	← ecky--ptang-zoob left	18:28
	→ cads joined	18:28
	→ beyondcreed joined	18:28
	→ Mars joined	18:28
	Mars → Guest54044	18:29
	← Advocation left	18:29
	← Guest54044 left	18:30
	← jaeckel left	18:30
	→ xok joined	18:30
xok	hello all...	18:30
gitinfo	xok: 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.	18:30
	→ eletuchy_ joined	18:31
xok	I have a bare repo and need to rebase it...	18:32
	can anyone help me do it?..	18:32
_ikke_	xok: step 1. clone it	18:32
	→ khmarbaise joined	18:32
	→ dyng joined	18:33
ojacobson	rebase (and, really, most history-manipulating options that aren't 'push') absolutely requires a work tree	18:33
	so, make a non-bare clone, make your changes there, and (force?) push them back	18:33
xok	_ikke_: thanks for the answer, I will be more specific... I have a server where the main repo resides, I am pushing the changes from the laptop and "fetch" those changes for the gitweb thing...	18:33
cbreak	there are quite a few that don't require a working dir	18:33
	→ ben_alman joined	18:33
xok	during set up I've made lots of commits and now wish to clean those things from the log...	18:34
	← Exancillatus left	18:34
	← gnu-man left	18:34
	→ dobalina joined	18:34
cbreak	the rawest probably being update-ref, and co	18:34
	← eletuchy left	18:34
xok	for that purpose I have a bare repo in the gitweb place, I want it to show the clean title, without any logs inside it...	18:34
	→ ZaeDrake joined	18:34
	→ kermit joined	18:34
	→ gnu-man joined	18:35
cbreak	I wonder if all filter-branch commands require a tree...	18:35
	commit-filter or the message ones might not	18:35
	→ pinage404 joined	18:36
	→ marhaban joined	18:37
	← doev left	18:37
	→ Noldorin joined	18:37
	→ skorgon_ joined	18:37
	← dyng left	18:37
	→ Exancillatus joined	18:38
	← mrbubbles left	18:39
	→ thiago joined	18:39
	← Exancillatus left	18:39
xok	cbreak: how do i clean the all logs inside the cloned repo for example?..	18:39
	I can clone the repo, clean the log and push it back to the server..	18:39
	← ChrisAnn left	18:39
cbreak	xok: depends on your definition of clean	18:40
	→ sandorjanssen joined	18:40
cbreak	you know how to use filter-branch?	18:40
	and git grafts?	18:40
xok	cbreak: just to leave files uncahged and remove all the "log" things from the repo...	18:40
cbreak	grafts allow you to cut history off	18:40
	← blinkdes_ left	18:40
cbreak	oh.	18:40
	completely kill history	18:40
	that's easy	18:40
	→ tjbiddle joined	18:41
cbreak	git checkout --orphan death && git commit -m "A new Life"	18:41
	→ frankstallone joined	18:41
xok	cbreak: thank you, I will test it out now...	18:41
cbreak	then push where ever you want	18:41
	→ TomyWork joined	18:41
TomyWork	hi	18:41
	can i somehow instruct git checkout <commit> <filename> not to touch the index?	18:41
	→ ncthom91 joined	18:42
_ikke_	TomyWork: nope	18:42
	→ khmarbaise_ joined	18:42
TomyWork	then i'll just give it a fake index file :P	18:42
cbreak	TomyWork: that's the path history takes to reach the working dir with checkout	18:42
	TomyWork: just use cat-file instead	18:42
xok	cbreak: that removed all the files in it...	18:42
TomyWork	it's not just a single file	18:42
cbreak	use cat-file multiple times	18:42
	← matanya left	18:42
xok	oh, sorry, it is..	18:42
	← khmarbaise left	18:42
	khmarbaise_ → khmarbaise	18:42
TomyWork	it's a multi-level folder structure	18:42
cbreak	sounds enjoyable.	18:43
	TomyWork: stash, checkout? :)	18:43
TomyWork	nah	18:43
	← petertretyakov left	18:43
cbreak	TomyWork: or commit reset --soft	18:43
	← marhaban left	18:43
	← charon left	18:43
	→ charon joined	18:43
cbreak	that'll save your index as commit	18:43
	← tgkokk left	18:43
TomyWork	i'll just hand it /dev/null as the index :)	18:43
cbreak	and with reset --soft HEAD~1 you can kill the commit	18:44
	and keep the index	18:44
TomyWork	or some temporary file	18:44
cbreak	dev/null sounds braindead.	18:44
TomyWork	yeah it probably wont work	18:44
cbreak	a different file might work...	18:44
TomyWork	i'm already doing that for the opposite direction (committing to a branch that isnt checked out), so it's no big deal :)	18:45
	→ looper joined	18:45
cbreak	you could just use new-workdir contrib script :)	18:45
	→ deanclkclk joined	18:45
	← ewnd9 left	18:46
TomyWork	too much porcelain :)	18:46
	→ FortySix2 joined	18:46
	← macben left	18:47
	→ lite_ joined	18:47
	← phipes left	18:48
	→ alexlindenlevy joined	18:48
	→ pepper_chico joined	18:49
osse	printf '%s\n' '#!/bin/sh' 'Usage: git get <commit> <filename>' '[ $# = 2 ] \|\| exit 1' 'git show "$1:$2" > "$2"' > "${PATH%%:*}"/git-get	18:50
_ikke_	still works only on one file	18:50
	→ revyd joined	18:51
osse	No it doesn't do even that. I forgot to comment the usage text :(	18:51
_ikke_	I was trying to parse it but wondered how that would work	18:51
	→ petertretyakov joined	18:51
	← jargon left	18:52
xok	cbreak: thank you very, very much, you taught me a great thing...	18:52
	← ncthom91 left	18:52
cbreak	xok: with great power come great responsibilities.	18:52
xok	cbreak: I promise I will be responsible... :-)	18:53
	← pinage404 left	18:53
cbreak	if you ever change history in an observable way, you could split the time stream, creating two alternate paths.	18:53
	ah well, Public history rewriting... :)	18:53
	← marcjs left	18:53
_ikke_	* Doc explaining marty about paralel timelines	18:55
	→ mmcdaris_ joined	18:55
xok	cbreak: what happens if I now push these changes to the main repo?...	18:55
	cbreak: will the repo hold the same thing as this cloned version?..	18:55
	← skorgon_ left	18:55
	← pepper_chico left	18:56
	→ maestrojed joined	18:56
	→ eletuchy joined	18:56
	← eletuchy left	18:56
	ej-webinar → emmajane	18:56
	→ eletuchy joined	18:57
_ikke_	https://www.youtube.com/watch?v=5OPkqY9doGE	18:57
	→ cored joined	18:57
cored	hi	18:57
	← mmcdaris left	18:57
	mmcdaris_ → mmcdaris	18:57
cored	I'm trying to remove a set of duplicate commits from my history	18:57
	after doing an interactive rebase and remove those commits	18:57
	← gnu-man left	18:57
tieTYT	I'm not sure that I can rebase -i HEAD~5, because I think the commits I want to squash may not be together. is there a way I can just specify the commit ids like 45684e2 and b4a1622?	18:58
cored	when I tried to push the branch to my remote, I get a difference between the two tips even when I just pull the branch locally, apart from that when I pull again after the rebase I get all the duplicated commits all over again	18:58
grawity	tieTYT: `rebase -i` lets you reorder commits	18:58
ToxicFrog	tieTYT: you are allowed to reorder comments when rebase -i'ing	18:58
_ikke_	tieTYT: no	18:58
cored	tieTYT: I'm doing something similar on mine	18:58
ToxicFrog	Er, s/comments/commits	18:58
	→ scelis joined	18:59
grawity	tieTYT: however, you always start with a range between HEAD and the given commit (whether by ID or by tag or by whatever) – you cannot specify separate commits at once	18:59
tieTYT	well I want to rebase the commits mentioned in this log: git log --oneline remotes/trunk..master The problem is I'm not sure if they're continuous or if there's commits in between. I'm not sure how to get the # for the HEAD~#	18:59
	→ ZujkisNx_ joined	19:00
Hanumaan	I have a git on apache server and I access it with http now I want to move the location on the server to the new location on the same server keeping the history of git so I just changed in apache document root but still not working what am i doing wrong ?	19:00
	← griotspeak left	19:00
	← eletuchy_ left	19:00
tieTYT	(btw, my understanding is that that log will show me local changes that haven't been pushed yet)	19:00
cored	any suggestion?	19:01
	→ CKyle joined	19:01
tieTYT	i suppose I could just say ~x. x being the number of lines in the log output and just make sure the ids match up 1:1	19:01
	← scelis left	19:02
cbreak	xok: you can push it as new branch, then the new history gets added on the side	19:02
	→ scelis joined	19:02
cbreak	xok: or you can force push it to other branches to replace their history	19:02
	← scelis left	19:02
_ikke_	tieTYT: Just pass git rebase the parent of the last commit you want to change	19:02
	← nicedice left	19:02
	→ scelis joined	19:03
_ikke_	tieTYT: git rebase -i <last_commit>~1	19:03
	→ ncthom91 joined	19:03
	→ nicedice joined	19:03
_ikke_	But make sure you are on master	19:03
tieTYT	doesn't that ~1 mean "the one before it"?	19:03
_ikke_	\yes	19:03
	yes	19:03
	tieTYT: you basically need to pass git rebase the base of the commits	19:03
	→ cmpera_ joined	19:03
	← mellernoia left	19:04
tieTYT	wouldn't that only squash 2 commits instead of all of them?	19:04
_ikke_	tieTYT: It won't do anything yet	19:04
tieTYT	right but my point is it gives me the option to interactively work on 2, instead of all	19:04
	← h0bbit_ left	19:04
tieTYT	I guess I'm confused about the ~1	19:04
	and why it's 1 instead of x. x being the number of unpushed commits	19:05
	← dougsland left	19:05
	→ SocalNick_ joined	19:05
	← rabbite left	19:05
	← sealion left	19:05
_ikke_	.digraph E->D->C->B->A	19:05
gitinfo	.digraph: http://g.jk.gs/fm.png	19:05
	→ jbarker joined	19:05
	← SocalNick left	19:05
	SocalNick_ → SocalNick	19:05
	← cmpera left	19:05
	cmpera_ → cmpera	19:05
_ikke_	imagine B, C, D and E haven't been pushed yet	19:05
	E is where master is	19:06
	A is where remotes/trunk is	19:06
tieTYT	ok	19:06
_ikke_	Does that fit your situation?	19:06
	→ tgkokk joined	19:06
tieTYT	I wasn't sure that it did at first, but now I think so	19:07
_ikke_	then git rebase -i remotes/trunk should list all the commits from B to E	19:07
tieTYT	I mean I committed stuff locally, then did an svn rebase. But I think that put all my local changes in front of trunk	19:07
	oh ok, that's easy	19:07
	→ rabbite joined	19:07
tieTYT	now if this was a feature branch and I wanted to squash the whole branch into one commit, could I do checkout feature; git rebase -i master ?	19:08
	→ Octalot joined	19:08
	← mklappstuhl left	19:08
	→ phipes joined	19:08
_ikke_	yes	19:08
	or even shorter, git rebase -i master feature	19:08
tieTYT	nice, ok thanks a lot	19:08
	← _tpavel left	19:08
_ikke_	(the last argument is the branch to checkout)	19:09
	← jbarker left	19:09
	→ jaeckel joined	19:09
	→ matanya joined	19:09
	→ jbarker joined	19:10
	← CKyle left	19:10
	← dav left	19:10
	← kermit left	19:10
	→ Exancillatus joined	19:10
	← goTAN left	19:10
tieTYT	oh that kind of confused me. Because I do that all the time without the -i. I guess by default rebase just "picks" everything?	19:10
	← lazers left	19:10
_ikke_	tieTYT: It basically picks everything between <upstream:the first argument> and HEAD	19:11
	← bwreilly_ left	19:11
tieTYT	i see	19:11
	← EngierkO left	19:11
	→ p4denis joined	19:11
	→ griotspeak joined	19:11
tieTYT	thanks. now I want to figure out how I can open this interactive editor in my ide instead of vim. thanks for the help	19:11
	← phipes left	19:12
ojacobson	tieTYT: $EDITOR, $VISUAL, $GIT_EDITOR, core.editor config	19:12
	if your IDE has a command-line tool that obeys the unix editor convention, then you can tell git to use it	19:12
_ikke_	tieTYT: Does your editor allow you to open files by passing it an argument	19:12
	← dtrott left	19:12
	→ amcsi joined	19:12
xok	cbreak: ok, now I have this local version, how do I put it on the server now?...	19:12
	→ pbrobst joined	19:12
ojacobson	(remain running until the file is closed, exit with status indicating failure/success)	19:12
cbreak	xok: push it	19:12
xok	cbreak: it says everything is up to date...	19:13
ojacobson	(accept filenames in argv)	19:13
cbreak	yeah. did you tell it what to push where?	19:13
	← teran left	19:13
_ikke_	remain running is probably the biggest problem	19:13
xok	cbreak: I am not sure I even know what do you mean...	19:13
cbreak	xok: git push origin foobar	19:13
_ikke_	I guess most of the time, the process communicates to the open instance to open the file, and then quits	19:13
cbreak	this pushes your local foobar branch to origin, as its local foobar branch	19:14
xok	cbreak: git push origin master << says everything is up to date...	19:14
cbreak	xok: yeah... because you don't want to push master	19:14
	xok: type git branch	19:14
	← doener left	19:14
ojacobson	_ikke_: OS X LaunchServices (the platform API that supports that program structure) has special flags for "don't reuse existing process" and "wait for process to exit" specifically for $EDITOR support :)	19:14
	→ marhaban joined	19:14
	← pbrobst left	19:14
	← jbarker left	19:14
xok	cbreak: yeah, I needed to push origin main (I've named it main)...	19:14
cbreak	you could rename it to master if you want.	19:15
	→ ben_alma_ joined	19:15
	→ bwreilly joined	19:15
cbreak	(with git branch -M)	19:15
xok	cbreak: the main repo contains the same stuff as its previous version...	19:16
	→ petey joined	19:16
	→ corni joined	19:16
	← Swayz left	19:16
tieTYT	ojacobson _ikke_ I'm not sure, I'll have to research that	19:16
xok	cbreak: it seems nothing has changed for the origin repo...	19:16
cbreak	yeah. that's normal.	19:16
tieTYT	I'm on windows	19:16
cbreak	xok: type git branch -a	19:16
	you will see that you created a new branch, main	19:16
xok	cbreak: yeah, it shows up...	19:17
cbreak	so, check it out	19:17
	→ doug joined	19:17
xok	but this new branch contains all the logs from old branch...	19:17
	← ben_alman left	19:17
doug	what's wrong ("wrong") with rebasing other than getting dates out of order?	19:17
Hello71	!rebaes	19:17
cbreak	xok: why do you think that?	19:17
Hello71	!rebase	19:17
gitinfo	[!merge_or_rebase] For a detailed discussion of the dis-/advantages of merge and rebase, and when to best use which, see: https://coderwall.com/p/ny1hia -- and here's how Linus explains which one to use: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html	19:17
Hello71	!rewrite	19:17
gitinfo	[!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	19:17
	← jason__ left	19:17
cbreak	xok: it contains exactly what you pushed.	19:17
_ikke_	Hello71: !private	19:18
	Hello71: !priv	19:18
gitinfo	Hello71: [!privmsg] You can get gitinfo keyword expansion (e.g. !doc or this one) sent directly to you by sending a private message to me: /msg gitinfo !doc. This avoids spamming the channel. Also see http://jk.gs/git/bot/trigger.php	19:18
xok	cbreak: can we have private conversation?..	19:18
Hello71	_ikke_: !itwasatypo	19:18
cbreak	no	19:18
	→ enr joined	19:18
	← KeiKun left	19:18
	← enr left	19:18
xok	cbreak: I just want to send you the link...	19:18
	cbreak: the link of the gitweb...	19:19
	← tarkus left	19:19
	→ KeiKun joined	19:19
	← HelperW left	19:19
cbreak	xok: don't rely on gitweb	19:19
xok	cbreak: I do not...	19:19
cbreak	use git log main on the command line	19:19
	→ enr joined	19:19
	→ HelperW joined	19:19
cbreak	if it shows more than git log main on your local repo, then something's wrong.	19:19
	you should then compare the output of git ls-remote origin and git show-ref	19:20
xok	cbreak: git log main shows only one commit...	19:20
cbreak	(from the local repo)	19:20
	good	19:20
	as expected	19:20
xok	cbreak: that's a great thing, but now, I want this to be the primary...	19:20
cbreak	as I said above	19:20
	you can rename it to master	19:20
	or you can check it out	19:20
	your choice	19:20
xok	how do I rename it?..	19:20
cbreak	as I said above	19:21
	git branch -M	19:21
	← Exancillatus left	19:21
xok	cbreak: thank you very much, you were very helpful...	19:22
	→ pbrobst joined	19:22
	← poulson left	19:22
cbreak	np :)	19:23
	← enr left	19:23
cbreak	for future changes, consider cherry-pick or similar	19:23
	→ infynyxx joined	19:24
	← HelperW left	19:24
xok	cbreak: git branch shows the asterisk on the master branch, I want it to be on the main instead, how do I do that?..	19:24
	→ fingerprint joined	19:25
cbreak	xok: you renamed main	19:25
	it's now called master	19:25
	→ Advocation joined	19:25
	← vince_prignano left	19:25
cbreak	do you say there still is a main?	19:25
	→ dougsland joined	19:26
	← Argure left	19:26
xok	cbreak: no, there is, but the asterisk sits on the old branch (I've renamed master to old just to be sure everything goes well)..	19:26
cbreak	alright. then just use checkout	19:27
	if you're in a bare, use git symbolic-ref	19:27
	→ JerryT_ joined	19:27
	← marhaban left	19:27
	→ tplaner joined	19:27
cbreak	it's used like: symbolic-ref HEAD main	19:27
	← JerryT left	19:27
xok	no, I am in the cloned repo....	19:28
	→ bvi joined	19:28
	→ Mars joined	19:28
	← redrocket left	19:28
	Mars → Guest42408	19:28
	→ Cephalostrum joined	19:28
	← Guest42408 left	19:28
	← Cephalostrum left	19:29
	→ kermit joined	19:29
	← infynyxx left	19:29
	→ Cephalostrum joined	19:30
xok	cbreak: I've got now three branches, how do I make one of them the main one?..	19:30
	→ Argure joined	19:31
	→ jbarker joined	19:32
cbreak	xok: check it out.	19:32
	xok: it's not main, it's current.	19:32
	← lite_ left	19:32
cbreak	the current branch is the one you have checked out	19:32
	← kermit left	19:32
cbreak	(it's the one HEAD points at)	19:32
xok	oh, I see...	19:32
	← mmcdaris left	19:32
	→ macben joined	19:32
	→ mmcdaris joined	19:33
	← Biopandemic left	19:33
	← Cephalostrum left	19:34
	→ Cephalostrum joined	19:35
	← paulweb515 left	19:35
	→ juliohm joined	19:37
	← juliohm left	19:37
	→ juliohm joined	19:37
	← ncthom91 left	19:37
	→ jhulten joined	19:37
xok	cbreak: if I am in the cloned repo I do "git checkout new_branch" and it switches to it, what do I do if I am in the bare repo?...	19:38
	→ kukks joined	19:38
	← Cephalostrum left	19:41
	→ vurma joined	19:41
	→ Cephalostrum joined	19:41
	← corni left	19:41
cbreak	xok: what I said above... git symbolic-ref HEAD new_branch	19:42
	→ place joined	19:42
cbreak	you need to work on your short term memory :)	19:43
	← Cephalostrum left	19:43
	→ mribeirodantas joined	19:43
	← Droolio left	19:43
	→ Cephalostrum joined	19:43
	← Cephalostrum left	19:44
	→ taylanub joined	19:44
	← salamanderrake left	19:44
	→ Cephalostrum joined	19:44
	← rabbite left	19:44
	emmajane → ejwalk	19:45
xok	cbreak: yeah, you are absolutely right, thanks again, I've finally did it... :-)	19:45
	cbreak: can you suggest anything for memory?...	19:46
	I am really struggling from it...	19:46
	I am not that old yet... :D	19:46
cbreak	forget about useless stuff like real life.	19:46
	works for me :)	19:46
	→ rabbite joined	19:46
xok	cbreak: :-D	19:46
	→ mfulz joined	19:46
	→ ncthom91 joined	19:46
taylanub	Not a git question but, is there any salvation to this nasty diff output ?: http://sprunge.us/DhEN (Change in function's prologue gets munged together with newly added function above it.)	19:47
	(--patience didn't help BTW.)	19:48
ojacobson	man git-diff, have a look at the options leading up to --diff-algorithm	19:48
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	19:48
	→ SWakey joined	19:48
	← _Vi left	19:48
	← AzizLight left	19:51
	→ infynyxx joined	19:51
	← tplaner left	19:52
	← SWakey left	19:52
	→ tplaner joined	19:52
	← tplaner left	19:53
	← cads left	19:53
	← ZaeDrake left	19:53
tieTYT	how does the git editor work? Does git create a temp file and use that as the last argument for your git editor command?	19:53
grawity	yes	19:54
tieTYT	is that explained in detail anywhere?	19:54
	→ skorgon_ joined	19:55
	← deanclkclk left	19:55
place	If I have a local clone of a fork, how do I fetch a branch on another fork of the same upstream? I tried 'git fetch other-rempte-addr' branchname and got a single line response with branchname -> FETCH_HEAD, but I can't do 'git checkout branchname'	19:55
tieTYT	git seems to be opening my editor with a file that only contains the path to the editor. Really confusing. Don't know how to debug that	19:55
taylanub	ojacobson: Am I out of luck if none of the 4 algorithms help ?	19:55
	taylanub wants s-expression diff. :(	19:56
	← tgkokk left	19:56
	← veebers left	19:56
	← hyperair left	19:57
	← aried3r left	19:58
	→ kermit joined	19:58
	← FortySix2 left	19:59
	→ tinti joined	19:59
	← DrShoggoth left	19:59
	← takamichi left	19:59
	← matteo left	20:00
	← skorgon_ left	20:00
imachuchu	place: what are you trying to do? Fetch a remote branch again to a new name? Fetch a different branch from the same remote repo?	20:00
	→ deanclkclk joined	20:00
mmathis	tieTYT: it looks like git creates a file in the repository's .git directory	20:00
	at least for me	20:00
	→ DrShoggoth joined	20:01
place	imachuchu: I'm trying to fetch a branch from a remote repository that is not my clones origin	20:01
mmathis	when I commit it opens vim .git/COMMIT_EDITMSG	20:01
	→ skorgon_ joined	20:01
	→ FortySix2 joined	20:02
mmathis	place: maybe try git checkout other_remote/branch	20:02
	→ rabbite1 joined	20:02
	← asakura left	20:03
	→ rgr joined	20:03
ojacobson	taylanub: well, you can always use your own difftool :)	20:04
	← rabbite left	20:04
	→ beastd joined	20:04
	← RogerWS left	20:04
	← bwreilly left	20:04
	→ sq-one joined	20:04
	→ mizu_no_oto joined	20:05
osse	taylanub: try --histogram	20:05
tieTYT	mmathis: is there any way to debug what command git is sending to my editor? I'd like to figure out why it's not working the way it should	20:05
	→ cavascript joined	20:06
cavascript	hello :)	20:06
gitinfo	cavascript: 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.	20:06
imachuchu	place: git remote add the repository, fetch the repository, then use it just like your other remote	20:06
taylanub	osse: doesn't help	20:06
place	imachuchu: thx I'll try it	20:06
osse	taylanub: then I don't know :/ That usually does the trick for C code	20:07
	← FortySix2 left	20:07
mmathis	tieTYT: I used htop	20:08
	→ gnu-man joined	20:08
	→ cyphase joined	20:08
	→ Mars___ joined	20:09
	→ SWakey_ joined	20:10
place	imachuchu: thanks that got me what I wanted	20:10
	→ m8 joined	20:11
	→ goTAN joined	20:13
	← kukks left	20:13
	← goTAN left	20:14
	← rgr left	20:14
	→ Mars____ joined	20:14
	→ goTAN joined	20:14
	← goTAN left	20:14
	→ halu joined	20:15
	← SWakey_ left	20:15
	← Mars___ left	20:16
	← revyd left	20:17
tieTYT	"Cannot 'squash' without a previous commit" Is there a way to go back into my interactive mode to fix my mistake?	20:17
	→ shawndumas joined	20:17
	← shawndumas left	20:17
	→ Mars___ joined	20:17
	→ revyd joined	20:17
	← Mars____ left	20:17
	→ godgodgodgo joined	20:17
	→ veebers joined	20:18
	→ blinkdesign joined	20:18
imachuchu	place: glad it worked out for you	20:19
	← albertgrala left	20:19
	← skorgon_ left	20:19
	← vurma left	20:19
	← Silmarilion left	20:20
mmathis	tieTYT: git rebase --abort and then redo your rebase	20:20
	→ rethus joined	20:21
	← revyd left	20:22
	← Mars___ left	20:22
	→ rq60 joined	20:22
	← petey left	20:22
	← SocalNick left	20:22
	→ dtrott joined	20:23
	→ SWakey joined	20:23
	→ kevwil joined	20:24
	→ aChatir joined	20:25
	← aChatir left	20:25
	→ ahmedelgabri joined	20:26
	← joelrobichaud left	20:27
	→ machouinard joined	20:28
	← rq60 left	20:28
	→ nZac_ joined	20:28
	→ rq60 joined	20:28
	→ mellernoia joined	20:28
	← ivan` left	20:28
	→ hwit joined	20:29
	→ daveytrain joined	20:30
	← hwit left	20:30
	← gdrc left	20:30
	← nZac left	20:30
	→ HelperW joined	20:30
	→ g3orge joined	20:30
	← doug left	20:31
	→ petey joined	20:31
	→ jason__ joined	20:31
	← rq60 left	20:32
	→ renatow__ joined	20:33
	→ paulweb515 joined	20:34
	← deanclkclk left	20:34
	nZac_ → nZac	20:34
	→ daveytra1n joined	20:34
	→ s0beit joined	20:34
	frobware → zz_frobware	20:34
	← freeed left	20:35
	← HelperW left	20:35
	→ rq60 joined	20:35
	← renatow_ left	20:35
	→ _Vi joined	20:35
	→ shruggar joined	20:35
	→ cedeon joined	20:36
	→ bwreilly joined	20:36
	← rq60 left	20:36
	← whaley left	20:37
	← daveytrain left	20:37
	→ psyl0n joined	20:37
	← psyl0n left	20:37
	→ psyl0n joined	20:37
	→ saustin joined	20:38
	→ rafaqueque joined	20:40
	← hoschi left	20:40
	← MistuhKurtz left	20:41
	→ RogerWS joined	20:41
	→ SocalNick joined	20:41
	← rabbite1 left	20:42
	→ renatow_ joined	20:42
	→ lite_ joined	20:43
	← renatow__ left	20:44
	→ Mars joined	20:44
	← ben_alma_ left	20:44
	Mars → Guest7218	20:44
	← jceb left	20:45
	← rafaqueque left	20:45
	→ rabbite joined	20:45
	→ tgkokk joined	20:45
	← Cyb3r-Assassin left	20:45
	→ nnyk_ joined	20:46
	← cavascript left	20:46
	← LuxuryMode left	20:46
	← lite_ left	20:48
	→ kukks joined	20:49
	→ rq60_ joined	20:49
	← ahmedelgabri left	20:49
	← gnu-man left	20:50
	← rq60_ left	20:50
	← Argure left	20:50
	→ schemanic joined	20:50
	← blinkdesign left	20:50
	→ ivan` joined	20:51
	→ rq60 joined	20:51
	← rq60 left	20:51
	→ rq60 joined	20:51
	← rq60 left	20:51
	← dtrott left	20:52
	→ VictorCL joined	20:52
	→ rq60 joined	20:52
	← rq60 left	20:52
	← renatow_ left	20:53
	← Gavrisimo left	20:54
	→ rq60 joined	20:54
	→ Argure joined	20:54
	→ dunpeal joined	20:54
	← rq60 left	20:54
	← psyl0n left	20:54
	→ ben_alman joined	20:54
	→ Gavrisimooo joined	20:55
	← tgkokk left	20:55
dunpeal	Hi. I have a tracked file which is really a template. Occassionally folks change it, mostly by accident. Is there a neat way to prevent any changes from being tracked, and/or reset the file whenever someone makes a commit while it is modified?	20:55
	→ tgkokk joined	20:55
	← VictorCL left	20:56
canton7	dunpeal, server-side hook?	20:56
	← ZujkisNx_ left	20:56
dunpeal	canton7: our model is distributed, so I'd rather a client-side solution	20:56
	→ rq60 joined	20:56
	← zackiv31 left	20:56
canton7	client-side hook?	20:56
	→ ZujkisNx_ joined	20:56
	← rq60 left	20:56
canton7	(each client will have to install it, which is the painful bit)	20:57
	→ rq60 joined	20:57
dunpeal	canton7: that's an idea. is there a way to set a mode on a file that is basically "stay the same until further notice"?	20:57
	some kind of a mode or a flag	20:57
	→ zackiv31 joined	20:57
canton7	there's a flag which can be abused, but lots of things reset it	20:57
	← zackiv31 left	20:57
	← b3tamax left	20:58
canton7	but no, ignores and tracked files don't go together	20:58
	← ZujkisNx_ left	20:58
	→ zackiv31 joined	20:58
dunpeal	canton7: what's the flag? just curious	20:58
	→ ZujkisNx_ joined	20:58
canton7	update-index --assume-unchanged iirc	20:59
	← petertretyakov left	20:59
	→ b3tamax joined	20:59
	← tgkokk left	21:00
	← rethus left	21:00
	→ VictorCL joined	21:00
	← angelsl left	21:01
dunpeal	yup, thanks.	21:01
	I'll probably publish a client side hook that will reset the file prior to committing.	21:01
	→ tgkokk joined	21:01
dunpeal	That's the behavior most devs would like anyway.	21:02
	← saustin left	21:02
	← sohail left	21:02
	→ dtrott joined	21:02
	← sandorjanssen left	21:02
	→ rafaqueque joined	21:02
	← VictorCL left	21:02
	→ VictorCL joined	21:03
dunpeal	anyone recall off-hand the name of the proper hook to create?	21:03
	← str left	21:03
	→ LuxuryMode joined	21:03
moritz	pre-commit?	21:03
	→ cagedwisdom joined	21:03
	← ncthom91 left	21:04
	← tgkokk left	21:04
	→ tgkokk joined	21:04
	← RofricTheBearded left	21:04
	shadowvi1e → shadowvice	21:05
	← Guest7218 left	21:05
	→ valgrind joined	21:05
	← JohnBat26 left	21:06
	← maestrojed left	21:06
	← GODDOG left	21:06
	← raniere left	21:07
canton7	dunpeal, man githooks	21:07
gitinfo	dunpeal: the githooks manpage is available at http://jk.gs/githooks.html	21:07
tieTYT	mmathis: ok :T thanks	21:08
	→ dsdeiz joined	21:09
	← TomyWork left	21:09
	← dsdeiz left	21:09
	→ MistuhKurtz joined	21:10
	← slyfox left	21:12
	→ slyfox joined	21:14
	← slyfox left	21:14
	→ slyfox joined	21:14
	← Advocation left	21:15
	→ angrytongan joined	21:15
	→ nilg joined	21:15
	← dobalina left	21:17
	← macben left	21:17
	→ matheus23 joined	21:17
	← rq60 left	21:20
	← jackneill left	21:20
	→ tpollard joined	21:21
	→ Silmarilion joined	21:22
	← LucasKA left	21:22
	→ scarabx joined	21:22
	→ raniere joined	21:23
	← dtrott left	21:23
	← ZujkisNx_ left	21:23
	← helgikrs left	21:23
	ejwalk → emmajane	21:24
	→ helgikrs joined	21:25
	→ Nizumzen joined	21:25
	← petey left	21:25
	→ pinage404 joined	21:25
	→ petey joined	21:26
	← dunpeal left	21:26
	← schemanic left	21:27
	← irqq left	21:27
	← kukks left	21:27
	← scarabx left	21:28
	→ oojacoboo joined	21:28
	→ dtrott joined	21:28
	→ dobalina joined	21:28
	→ Exancillatus joined	21:28
	→ bartman` joined	21:28
	← dobalina left	21:29
	→ scarabx joined	21:29
	← scarabx left	21:29
	→ scarabx joined	21:30
	← petey left	21:30
	→ FortySix2 joined	21:31
	← Kicer86 left	21:31
	→ petey joined	21:31
	← Pullphinger left	21:32
	← scarabx left	21:33
	→ maqr joined	21:33
	→ scarabx joined	21:33
	← tgkokk left	21:34
	← xok left	21:36
	← jbarker left	21:37
guardian	given an ambiguous short sha1 d8eff, how can I find the first ancestor that starts with d8eff?	21:37
ojacobson	"first" isn't terribly well-defined	21:38
	← looper left	21:38
cbreak	guardian: git rev-list , search in that thing	21:38
	maybe with grep	21:38
	→ tgkokk joined	21:38
	← scarabx left	21:38
cbreak	grep supports regexps or so I heard :)	21:38
	← roymiloh left	21:38
guardian	:)	21:38
	← pinage404 left	21:39
guardian	I was seeking advice on e.g. git log vs git rev-list	21:39
	or else	21:39
cbreak	rev-list is machine parsable by design	21:39
	← dreiss left	21:39
cbreak	log is for humans and their humanly needs	21:39
	→ Targen joined	21:41
guardian	hmm the git rev-list HEAD ^origin example in git rev-list man confuses me	21:41
	← Targen_ left	21:41
guardian	I would have thought of origin as a remote	21:41
canton7	guardian, "Specifying Revisions" in man git rev-parse	21:42
gitinfo	guardian: the git-rev-parse manpage is available at http://jk.gs/git-rev-parse.html	21:42
	→ deanclkclk joined	21:42
	← jizanthapus left	21:43
	→ scarabx joined	21:43
	→ teran joined	21:43
	← Panicg left	21:43
	→ deanclkclk_ joined	21:44
Kartagis	this is a user question: I need to go back and get a copy of one file at a specific commit. if I do "git checkout hash filename.module" can I get back to the most current commit when I'm done? if I'm not mistaken, they need cherry-pick, no?	21:44
	← nZac left	21:44
	→ AzizLight joined	21:45
	→ maestrojed joined	21:45
	→ skorgon_ joined	21:45
	→ travisstaton joined	21:45
	→ Mars___ joined	21:45
	→ Watabou joined	21:46
zackiv31	how do i name a detached HEAD branch ?	21:46
	← raniere left	21:46
	→ paul424 joined	21:46
ojacobson	!reattach	21:46
gitinfo	Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue	21:46
paul424	what git fetch origin branch_name does ?	21:46
	it seems to screw up my repo .. ?>"	21:46
	← KeiKun left	21:47
	→ KeiKun_ joined	21:47
	← deanclkclk left	21:47
	deanclkclk_ → deanclkclk	21:47
	← cedeon left	21:47
ojacobson	copies refs/heads/branch_name from the remote named origin to FETCH_HEAD locally	21:47
	like it says in the man page :)	21:47
Kartagis	this is a user question: I need to go back and get a copy of one file at a specific commit. if I do "git checkout hash filename.module" can I get back to the most current commit when I'm done? if I'm not mistaken, they need cherry-pick, no?	21:48
ojacobson	(downloading any necessary objects for that ref t make sense)	21:48
	→ btree joined	21:48
ojacobson	Kartagis: 'git checkout <commit-ish> <paths>' does something a bit surprising	21:48
	← aard_ left	21:48
ojacobson	it copies the version of <paths> from the commit <commit-ish> into your work tree, and stages them for commit	21:48
paul424	Merge branch 'development' of ssh://git.code.sf.net/p/opendungeons/git into	21:48
	development	21:48
ojacobson	it doesn't move the work tree to any commit at all :)	21:48
paul424	does it make sense att all ?	21:48
	ojacobson: sure	21:49
	← thevishy left	21:49
paul424	ojacobson: I am in fear that it screwed something or I screwed	21:49
Kartagis	ojacobson: what about git cherry-pick hash?	21:49
	← skorgon_ left	21:50
	← b3tamax left	21:50
	→ Biopandemic joined	21:50
juliohm	I forgot the command to see the diff for the changes to be commited?	21:50
	I added the changes with git add, how to see all?	21:50
	all added	21:51
canton7	git diff --cached for staged change only	21:51
ojacobson	'git diff --staged' -- see man git-diffed	21:51
	'git diff --staged' -- see man git-diff	21:51
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	21:51
canton7	git diff HEAD for everything	21:51
	← matheus23 left	21:51
juliohm	canton7, --cached and --staged are exactly the same?	21:51
ojacobson	eyup	21:52
	→ b3tamax joined	21:52
	→ vince_prignano joined	21:52
	→ skorgon_ joined	21:52
	→ dreiss joined	21:52
	← mizu_no_oto left	21:52
juliohm	thanks	21:52
	← ferreum left	21:52
	→ infynyxx_ joined	21:53
	← rfoust left	21:55
	← tgkokk left	21:56
	← gdogrady left	21:56
	← infynyxx left	21:57
	→ rainbyte joined	21:57
paul424	Merge branch 'development' of ssh://git.code.sf.net/p/opendungeons/git into development	21:57
	what does that mean .. which is remote ,which is local ?	21:58
	← cmpera left	21:58
	← DolphinDream left	21:58
	→ nZac joined	21:59
guardian	git rev-parse --verify 8274 tells me short SHA1 8274 is ambiguous. But git rev-list --all \| grep ^8274 only outputs a single long SHA1. So? Where does the ambiguity come from please?	21:59
	← tobias__ left	22:00
paul424	well wehbn I do pull I do fetch and auto-merge	22:00
	why the merge message is in my git log ?	22:00
	→ raniere joined	22:00
	→ gunmuc1 joined	22:00
	← weltall left	22:00
	→ BlueOrg joined	22:01
	→ simesy joined	22:02
	← gunmuc left	22:02
	← quup left	22:02
	→ quup joined	22:02
	→ weltall joined	22:02
	← cannonball left	22:03
	← JerryT_ left	22:04
	← sq-one left	22:04
	← godgodgodgo left	22:05
paul424	I haven't gave a git merge command :(	22:06
ojacobson	paul424: did you mean "pull" when you said "fetch"?	22:06
	← deanclkclk left	22:07
canton7	pull = fetch + merge. If you did a 'git pull', you did a merge	22:07
paul424	sure ... but why in my git log it prints explicitly about it	22:07
	→ LucasKA joined	22:07
	→ deanclkclk joined	22:07
ojacobson	git pull is a merge command.	22:07
	You asked for it, you got it.	22:07
	← bbloom left	22:08
paul424	ojacobson: naah I rarely pull that might be the case	22:08
	← infynyxx_ left	22:08
	← petey left	22:08
paul424	donno how git ack that ... Is it old feature of doing that , riht	22:09
	ri9ght ?	22:09
	→ petey joined	22:09
	→ Diemuzi joined	22:09
	gitinfo set mode: +v	22:09
paul424	ack == logs :)	22:10
	qq	22:10
	← slyfox left	22:10
	← ben_alman left	22:10
	← PeterBooker left	22:11
	→ pepper_chico joined	22:11
	→ salamanderrake joined	22:11
	← SWakey left	22:11
	→ marcjs joined	22:11
paul424	Your branch is ahead of 'origin/development' by 13 commits.	22:12
	← BlueOrg left	22:12
	← FortySix2 left	22:12
paul424	not true ... only 3 commits	22:12
	→ rq60 joined	22:13
paul424	why is that so ... may I post my git logs ?	22:13
ojacobson	Have you 'git fetch origin' recently?	22:13
paul424	yes	22:13
	→ btree_ joined	22:13
	→ kpreid joined	22:13
ojacobson	(If not, does it mention 'development' when you run 'git fetch origin' now? :)	22:13
	← petey left	22:13
ojacobson	Then your local branch is 13 commits ahead, i.e., there are 13 commits in git log origin/master..HEAD	22:14
PerlJam	paul424: was there a recent merge?	22:14
ojacobson	er origin/develop	22:14
	!lol may be enlightening	22:14
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	22:14
	← kevwil left	22:14
	→ petey joined	22:14
	→ infynyxx joined	22:15
	← btree left	22:15
	← nZac left	22:16
	→ dsdeiz joined	22:17
	← pepper_chico left	22:17
paul424	http://pastebin.com/jvXuucgV	22:17
	no Idea what happened to this repo	22:17
	← el3ctron_ left	22:17
ojacobson	Your team (mostly 'faichele@primusnetz.de') appear to use 'git pull' for integrating upstream changes into local work	22:18
	which works fine, but it generates these staggeringly merge-heavy histories	22:18
	← IrishPrime left	22:18
ojacobson	!workflow !pull_vs_rebase are both worthwhile reading; you'll need a grounding in git's basic principles to make sense of them	22:19
gitinfo	Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	22:19
ojacobson	!rebase	22:19
gitinfo	[!merge_or_rebase] For a detailed discussion of the dis-/advantages of merge and rebase, and when to best use which, see: https://coderwall.com/p/ny1hia -- and here's how Linus explains which one to use: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html	22:19
ojacobson	that	22:19
paul424	I was never aware of the problem ... I always used git pull	22:20
ojacobson	It's not a problem, exactly	22:20
	how it affects your workflow & expectations might be, but what it does is simple and predictable	22:20
	← alezost left	22:20
	← ItSANgo left	22:21
ojacobson	the "problem" in as much as there is one is idiot tutorials (and git's own advice, frustratingly) suggesting "git pull" without explaining what it does or when it's appropriate or what the workflow implications are	22:21
	→ hellopat_ joined	22:21
	← SocalNick left	22:21
paul424	http://pastebin.com/x33cLDe5 http://pastebin.com/uCSuAfv0	22:22
	can you compare local git log vs remote git log	22:22
	and why it claims it's 13 commits not 3 ?	22:22
	← hellopat_ left	22:22
	← rendar left	22:22
	← Mars___ left	22:23
	→ Mars joined	22:23
	Mars → Guest86704	22:23
	← LuxuryMode left	22:23
	← hellopat left	22:24
	← mishok13 left	22:26
	→ b4rbz joined	22:26
paul424	There was sayinh that 'git fetch origin' or 'git fetch origin branch_name' does not do what it does ..	22:28
	→ macben joined	22:28
paul424	Is there a diffrence more than one fetches one branch the other all ?	22:28
nevyn	!4fetch	22:29
	!four	22:29
gitinfo	[!fetchfour] We recommend against using 'git fetch/pull <remote> <refspec>' (i.e. with branch argument), because it doesn't update the <remote>/<branch> ref. The easy way to fetch things properly is to get everything: 'git fetch' or 'git pull' are sufficient if you have one remote; otherwise we recommend 'git fetch <remote>' (plus 'git merge <remote>/<branch>' if you wanted to pull/merge).	22:29
ojacobson	!fetch4	22:29
gitinfo	[!fetchfour] We recommend against using 'git fetch/pull <remote> <refspec>' (i.e. with branch argument), because it doesn't update the <remote>/<branch> ref. The easy way to fetch things properly is to get everything: 'git fetch' or 'git pull' are sufficient if you have one remote; otherwise we recommend 'git fetch <remote>' (plus 'git merge <remote>/<branch>' if you wanted to pull/merge).	22:29
nevyn	that one !.	22:29
ojacobson	but fetch does not, under any circumstance I can think of, create merge commits	22:29
	→ cedeon joined	22:29
nevyn	git fetch origin is fine.	22:29
ojacobson	pull does, and the presence of all those URLs in your log makes me believe that, in spite of your protestations, you run 'git pull' quite frequently	22:29
	← jason__ left	22:29
	← donmichelangelo left	22:30
	← Not_Whatever left	22:30
	→ FortySix2 joined	22:30
	→ donmichelangelo joined	22:30
	← skorgon_ left	22:30
	← Diemuzi left	22:31
	← beastd left	22:31
paul424	ok many thanks	22:32
	← infynyxx left	22:32
	← macben left	22:32
mmathis	fetchfour?	22:33
	!fetch3	22:33
	where does the four come from	22:33
nevyn	git fetch origin branch	22:34
	^^^ the behaviour of this is not obvious ^^^	22:34
	← Biopandemic left	22:35
	→ ItSANgo joined	22:35
grawity	in latest git.git it actually does update the ref	22:35
	→ axeLR joined	22:36
	← bwreilly left	22:36
paul424	http://fc08.deviantart.net/fs50/f/2009/301/6/f/Farfetch__d_by_Xous54.png you mean farfetch ?	22:36
	scoT_T`zZz → scoT_T	22:36
	← danicampa901 left	22:37
	→ gdogrady joined	22:38
	← EugeneKay left	22:38
	→ SocalNick joined	22:38
	→ EugeneKay joined	22:38
	← AaronMT left	22:39
	→ cads joined	22:39
	→ AaronMT joined	22:39
	ChanServ set mode: +v	22:39
	← g3orge left	22:40
paul424	nevyn: seems this cmd might be the pun on this pokemon	22:40
EugeneKay	No, it's "four word invocation"	22:40
	← mfulz left	22:41
	← iderik left	22:42
	→ ncthom91 joined	22:42
	→ kukks joined	22:43
	← kukks left	22:43
	→ kukks joined	22:43
paul424	It's funny when you get programming tools terms translated to your native language	22:43
	nowI know that staged area == poczekalnia ( like a waiting room before the GP ) :)	22:44
	GP's office	22:44
	in polish :P more worth than a paragraph of description :P	22:45
	→ endiruna joined	22:45
	← cads left	22:46
	← ojacobson left	22:46
	← ngladitz left	22:46
	← scelis left	22:48
	← bambams_ left	22:51
	← gasbakid left	22:52
	→ prawnsalad joined	22:52
prawnsalad	oh awesome, git people all around	22:53
	← Nizumzen left	22:53
prawnsalad	though i just realised my mistake	22:53
	good git people.	22:53
	← prawnsalad left	22:53
	→ saustin joined	22:54
	← Gamecubic left	22:55
	← m0ltar left	22:55
	← cedeon left	22:56
	→ m0ltar joined	22:57
	← vince_prignano left	22:57
	← gdogrady left	22:59
	→ mrbubbles joined	22:59
	→ gdogrady joined	22:59
	→ Gamecubic joined	23:00
	← m8 left	23:00
	→ jbarker joined	23:01
tieTYT	i'm trying to sell git to my coworkers. We use svn. In Svn, if I change line 1, then do an update and the remote repository also has a change to line 1, that will be a conflict. Is that also true in git? Assuming I use git svn rebase, I think not?	23:02
	← daveytra1n left	23:02
	← travisstaton left	23:02
	← btree_ left	23:03
tieTYT	I think not because that will essentially say, "take the remote repository change and apply it before your change"	23:03
imachuchu	tieTYT: it will be a conflict because git will try to reapply your changes ontop of the new remote head, but since the important parts of it have changed (line 1) it will throw a conflict.	23:04
	← VictorCL left	23:04
	→ HelloShitty joined	23:05
	← Noldorin left	23:05
	→ bwreilly joined	23:05
	← helgikrs left	23:05
	← phile left	23:05
tieTYT	oh	23:05
	← dreiss left	23:05
imachuchu	tieTYT: but the advantage of git is that you don't have to deal with it until you want to (when you rebase/merge)	23:06
	← matanya left	23:07
	→ matanya joined	23:07
	← petey left	23:09
	← CaveJohnson left	23:09
	→ petey joined	23:10
	→ VictorCL joined	23:11
tieTYT	isn't that true with svn? I can choose not to upgrade	23:11
	← m0ltar left	23:11
	← petey left	23:11
	← calcifea left	23:12
	→ petey joined	23:12
tieTYT	*update	23:12
	← matanya left	23:12
	← Soliton left	23:13
	← Exancillatus left	23:13
	→ Vampire0 joined	23:13
imachuchu	tieTYT: absolutely, but you can't use svn until you update. With git you can still use it all you want locally, you can even fix the conflicts in stages (a whole branch to track your commits to reconcile branches) instead hof having a hard stop until the conflict is resolved	23:14
	→ m0ltar joined	23:14
	← DolpheenDream left	23:16
	→ skorgon_ joined	23:16
tieTYT	i see	23:16
	← smashwilson left	23:17
	← shauna left	23:17
	→ DolpheenDream joined	23:17
	→ KaaK_ joined	23:17
	→ Matrixiumn joined	23:18
	← harobed_ left	23:18
	← kukks left	23:18
	→ calcifea joined	23:19
	← noecc left	23:19
	← m0ltar left	23:20
	← KaaK left	23:20
	→ Noldorin joined	23:21
	← khmarbaise left	23:21
	→ helgikrs joined	23:22
	← MistuhKurtz left	23:22
	← acharlieh left	23:23
	← insejn left	23:23
	→ dreiss joined	23:24
	← kermit left	23:24
	→ m0ltar joined	23:25
	→ hyperair joined	23:25
	→ marhaban joined	23:25
	→ MistuhKurtz joined	23:26
	→ insejn joined	23:26
	→ matanya joined	23:26
	→ kermit1 joined	23:27
	← Destos left	23:27
	← petey left	23:29
	← Gavrisimooo left	23:30
	← Vampire0 left	23:30
	← bhldev__ left	23:31
dtrott	imachuchu: To be fair thats somewhat of a client issue, you can personally use the git-svn integration and see most of the advantages of a git-git setup without having to sell anyone else....	23:31
	← FortySix2 left	23:31
	poly → polyzen	23:32
	→ FortySix2 joined	23:32
	→ bhldev joined	23:32
	← alexlindenlevy left	23:32
	→ FortySix_ joined	23:33
	← skorgon_ left	23:33
	→ alvaro_o joined	23:34
	← FortySix_ left	23:34
	→ infynyxx_ joined	23:34
	← DolpheenDream left	23:34
	← m0ltar left	23:35
	← hyperair left	23:35
	→ DolpheenDream joined	23:35
	→ m0ltar joined	23:36
	← marhaban left	23:36
	→ Gavrisimo joined	23:36
	→ revyd joined	23:37
	← FortySix2 left	23:38
	polyzen → poly	23:38
	→ Mamoru joined	23:38
	→ unomystEz joined	23:38
	← Mamoru left	23:38
	→ Mamoru joined	23:38
tieTYT	dtrott: that's my first step to selling it	23:41
	that's all I'm using	23:41
	← marcjs left	23:42
	← Mamoru left	23:43
	→ Mamoru joined	23:43
	→ Destos joined	23:44
	← ariscop left	23:45
	→ shauna joined	23:45
	→ daivyk joined	23:49
	← scarabx left	23:51
	← dtrott left	23:51
	→ blinkdesign joined	23:51
	← rednaks left	23:51
	← eijk_ left	23:52
	← rudisimo left	23:52
	← elbeardmorez left	23:52
	← becks__ left	23:52
	→ huahax joined	23:53
	← chrisshattuck left	23:53
	→ Vampire0 joined	23:54
	← valgrind left	23:54
	← Guest86704 left	23:55
	→ becks___ joined	23:55
	→ Mars joined	23:56
	← Olipro left	23:56
	Mars → Guest15306	23:56
	← blinkdesign left	23:56
	→ scarabx joined	23:57
	← endiruna left	23:57
	← Norm_ left	23:57
	→ Norm_ joined	23:58
	→ alexlindenlevy joined	23:58
	→ Olipro joined	23:59
	→ JerryT joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation