IRCloggy #git 2013-11-28

Provider of IRC logs since 2005.

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

	2013-11-28
	→ lite_ joined	00:00
	← Terabyte left	00:00
	← rednaks left	00:01
grawity	hmm, Git requires one to commit before merging, not the other way around	00:01
	anyway, you could teach them to rebase or pull --rebase	00:01
	← bartzy left	00:02
	← jon_r left	00:02
	→ kris joined	00:03
aaronds	sorry grawity, got that wrong. It's after committing but before pushing.	00:03
	→ johnkevinmbasco joined	00:03
	← Noldorin left	00:03
	← lite_ left	00:04
	→ simesy joined	00:04
	← kris left	00:04
	→ MattMaker joined	00:05
	← kriks left	00:05
	← alvaro_o left	00:06
	← alp82 left	00:07
	→ alvaro_o joined	00:07
	← kirin` left	00:07
majuscule	aaronds: rebasing is what you're looking for i think	00:07
	aaronds: you can also try to always use --fast-forward when merging	00:08
	→ bwreilly joined	00:08
majuscule	aaronds: but rebasing'd probably be preferred	00:08
	← amcsi left	00:08
	← hyperair left	00:09
aaronds	ok cool - so is this generally what happens in most organisations? Needing to rebase when pulling changes? Or do people just remember to pull before commiting lol	00:09
	← adac left	00:09
	→ ericy joined	00:09
	→ rjhunter joined	00:09
	→ crazydiamond_ joined	00:11
jrnieder	aaronds: yeah, I think lots of people rebase, unless they're working on a long-term topic branch where a rebase might be disruptive	00:11
aaronds	ok cool	00:12
	thanks everyone	00:12
jrnieder	aaronds: or log --no-merges :)	00:12
	← crazydiamond_ left	00:12
	← boombatower left	00:13
	← zhost left	00:14
	← rjhunter left	00:14
	← dnlrn left	00:15
	→ luxgeek joined	00:15
	→ Noldorin joined	00:15
	← maestrojed left	00:17
	→ kh0d joined	00:17
	→ dxdt joined	00:19
	← HappyLappy left	00:19
	→ jan1337z joined	00:20
	← PeterBooker left	00:20
	← etcetera left	00:21
	→ madewokherd joined	00:22
	→ etcetera joined	00:22
	← etcetera left	00:23
	→ jon_r joined	00:23
	→ etcetera joined	00:23
	← pfFredd left	00:24
	← looper left	00:25
	← asQuirreL left	00:25
	← Kayra left	00:26
	→ sdx23 joined	00:27
	→ wpaulson joined	00:27
	→ Cerise joined	00:28
	← wpaulson left	00:28
	← ericy left	00:30
	→ specing joined	00:32
	→ m_ke joined	00:32
	→ ehmatthes joined	00:32
	→ AlecTaylor joined	00:32
	← xobb left	00:34
	← el3ctron_ left	00:34
	→ dravidan joined	00:34
	→ farn joined	00:35
	→ undecim joined	00:35
	→ treaki_ joined	00:36
	← treaki__ left	00:38
	→ refried_ joined	00:38
	→ HappyLappy joined	00:39
	← jarr0dsz left	00:39
	→ el3ctron_ joined	00:40
	← ehmatthes left	00:40
	→ codeman joined	00:40
	← maqr left	00:41
	← dhackner left	00:42
	← m_ke left	00:42
	→ dhackner joined	00:42
	← Ltk left	00:43
	← el3ctron_ left	00:43
	→ el3ctron_ joined	00:43
	← el3ctron_ left	00:43
	→ DABVCW6 joined	00:44
	→ relipse joined	00:45
	← dsdeiz left	00:45
relipse	Ziber: hey you there?	00:45
	→ dreiss joined	00:45
	→ ericy joined	00:46
	← dhackner left	00:47
	← simesy left	00:47
	→ jcardinal joined	00:48
	← SynthesizerPatel left	00:48
	← echo-area left	00:48
	→ ezekielnoob3 joined	00:49
	→ dhackner joined	00:49
	← dxdt left	00:49
	→ echo-area joined	00:50
	→ kirin` joined	00:50
	HappyLappy → HiramHibbard	00:50
	→ groglogic joined	00:51
	← mdiers__ left	00:51
	← devanblue left	00:52
jcardinal	Git noob here. I have a repo that I've done a handful of commits on from a few different machines, all pushed to the origin server. Some of the machines I was committing from had the wrong author info, which I wanted to correct. Searching suggested I should do a rebase and edit each commit that had the wrong author info and amend those commits. I did that, but now git status says my branch and the one on the origin have	00:52
	diverged. The local copy is the corrected one, and I want to push those corrections to the origin. How do I do that?	00:52
	→ b_rare joined	00:52
	→ el3ctron_ joined	00:52
	← el3ctron_ left	00:53
	→ gregorg joined	00:53
	→ el3ctron_ joined	00:53
	← ezekielnoob3 left	00:53
	← el3ctron_ left	00:54
	← smashwilson left	00:54
dalias	jcardinal, you should wait for someone more experienced to give you a better answer, but the short answer is that this is "changing history" and considered a bad thing if the repo is already published and others may have cloned it	00:55
	you probably need to force the push if you actually want to do it	00:55
	← Skrypter left	00:55
	← etcetera left	00:55
jcardinal	dalias: Thanks. In this case, the origin is on a server owned by me and no one else has worked on with this repo yet, so I feel safe "changing history". That said, how would I force that change through?	00:56
	→ PeterBooker joined	00:56
	← PeterBooker left	00:56
	← kh0d left	00:57
jcardinal	Is it literally as simple as just adding the --force option to my push?	00:57
	→ shruggar joined	00:57
	← freestyl3r left	00:57
	→ SoftwareMechanic joined	00:57
dalias	i suspect so, but i'm also a git newbie, so i don't want to tell you something wrong and break your repo	00:58
	→ pantsman joined	00:58
jcardinal	dalias: Thanks for being cautious with your advice. My gut says that's all there is to it, but like you said...I'd don't want to risk destroying the repo.	00:59
	I guess I'll just clone it elsewhere for safety, force the push, and if things get broken, I'll just have to delete the rep on the origin server and recreate it there from the backup.	01:00
	← HiramHibbard left	01:00
dalias	that sounds right to me	01:00
	← groglogic left	01:00
	→ HappyLappy joined	01:00
OMGOMG	i thought destroying the repo was what you wanted	01:00
	← HappyLappy left	01:01
dalias	heh that's one way of looking at it...	01:01
	→ ewnd9 joined	01:01
	→ caseydriscoll joined	01:01
	→ dhruvasagar joined	01:02
	← b_rare left	01:02
	→ mk_qi joined	01:03
	→ hholtmann joined	01:03
	→ mdiers__ joined	01:03
jcardinal	Well, I just forced the push, and it worked. The only side effect, which is understandable, is that all of the commits after the first one I edited had their timestamps updated.	01:03
	→ mribeirodantas joined	01:03
	→ HappyLappy joined	01:04
jcardinal	OMGOMG: I guess I wanted to rewrite history on the repo; some may consider that destroying it :P	01:04
OMGOMG	absolutely	01:05
	← robbyoconnor left	01:05
OMGOMG	look into filter-branch as well	01:05
	← HappyLappy left	01:05
	→ el3ctron_ joined	01:05
	→ rjhunter joined	01:06
jcardinal	"Otherwise, all information (including original commit times or merge information) will be preserved." Looks like I should have waited a few minutes longer for a better answer!	01:06
	→ hyperair joined	01:06
	→ weiyang joined	01:06
	← hahuang65 left	01:07
OMGOMG	you'd still have to force push	01:07
jcardinal	I'm ok with that; at least I'd maintain my timestamps	01:08
	(To be clear, the message is well-taken that I should never rewrite history on a public / shared repo.)	01:08
	← pantsman left	01:09
	→ Watabou joined	01:10
	→ sohail joined	01:11
	← el3ctron_ left	01:11
	→ lindenle joined	01:11
	← knighthawk left	01:11
	→ knighthawk joined	01:12
	← SHyx0rmZ left	01:13
	→ nnyk_ joined	01:13
	← nnyk_ left	01:14
	→ nnyk_ joined	01:15
	← bwreilly left	01:16
	← sohail left	01:16
	→ sohail joined	01:16
	→ bbloom joined	01:16
	← knighthawk left	01:16
	← nbrosnahan left	01:17
	← sohail left	01:18
	← rjhunter left	01:19
	← wrouesnel1 left	01:21
	← lindenle left	01:22
	← pr0ton left	01:23
	← ewnd9 left	01:24
	← kiba left	01:24
jcardinal	Ok, history correctly rewritten thanks to git-filter-branch and a handy script from github: https://help.github.com/articles/changing-author-info	01:25
	← Zarthus left	01:25
jcardinal	Timestamps preserved. Thanks again OMGOMG!	01:25
dalias	:)	01:26
	← crazydiamond left	01:27
	← M00sL0gB0t_ left	01:27
	→ M00sL0gB0t_ joined	01:27
	← King_Hual left	01:28
	→ fairuz joined	01:29
	← aaronds left	01:31
	← Guest25228 left	01:32
	→ Guest25228 joined	01:32
	← echo-area left	01:33
	← apipkin left	01:35
	→ Balliad joined	01:39
	→ xobb joined	01:39
	→ simesy joined	01:41
	→ \|Starcraftmazter joined	01:41
	← mloskot left	01:43
	← Ardit left	01:44
	← Starcraftmazter left	01:44
	→ rchavik joined	01:44
	← AlecTaylor left	01:45
	← sgu left	01:47
	← penguin42 left	01:47
	← shruggar left	01:48
	← joshbetz left	01:51
	← dhruvasagar left	01:52
	← dhackner left	01:53
	← joaoh82 left	01:53
	→ joaoh82 joined	01:54
	← xobb left	01:56
	← nnyk_ left	01:56
	← akiva left	01:57
	→ whitby joined	01:57
	→ john2x joined	01:57
	→ Imagineer joined	01:58
	← joaoh82 left	01:58
	→ joaoh82 joined	01:59
	→ dsdeiz joined	02:00
	← ericy left	02:01
	→ psyl0n joined	02:03
	← psyl0n left	02:03
	→ psyl0n joined	02:03
	← Meistarin left	02:03
	← lu_zero left	02:05
	→ lu_zero joined	02:05
	→ luc_ joined	02:06
	→ ni_ joined	02:08
	← codeman left	02:08
	→ boombatower joined	02:08
	← jonatasnona left	02:08
	← chrisss404 left	02:09
	← kirin` left	02:09
	→ chrisss404 joined	02:09
	← Devastator left	02:10
	← thiago left	02:10
	→ jshawl joined	02:11
	→ Devastator joined	02:11
	→ groglogic joined	02:11
	→ Goplat joined	02:12
	→ codeman joined	02:13
	← tobiasBora left	02:14
	← Balliad left	02:15
	← groglogic left	02:16
	→ Skrypter joined	02:16
	→ jmaya joined	02:17
	→ akiress joined	02:18
	→ ericy joined	02:18
	← kukks left	02:18
	← luxgeek left	02:18
	← b3tamax_ left	02:19
	← b3tamax left	02:19
	← endiruna left	02:19
	→ daito joined	02:20
	← jmaya left	02:20
	← Dougie2187 left	02:21
	→ harish joined	02:21
	← john2x left	02:21
	← Devastator left	02:21
	→ Devastator joined	02:21
	→ Dougie2187 joined	02:21
	→ lindenle joined	02:23
	→ Meistarin joined	02:23
	→ Devastator_ joined	02:24
	← GT4066 left	02:24
	← pbrobst left	02:25
	← Devastator left	02:25
	→ kukks joined	02:26
	← kukks left	02:26
	→ kukks joined	02:26
	← carbon2 left	02:28
	← mmcdaris left	02:28
	← nemesit\|znc left	02:29
	→ rjhunter joined	02:30
	→ s_e joined	02:32
	→ echo-area joined	02:32
	→ buzzybron joined	02:32
	← ericy left	02:35
	← rjhunter left	02:35
	→ nemesit\|znc joined	02:36
	← Davey left	02:36
	→ Curtisbeef_ joined	02:37
	← Curtisbeef left	02:38
jshawl	I'm so about git right now	02:38
	ask me any question	02:38
	→ dhruvasagar joined	02:38
	→ nerdfiles joined	02:39
Celelaptop	What happen if I rebase a branch on which another branch is based?	02:40
	← unkmar left	02:41
jshawl	updates the original branch with the work done with the other	02:41
	do you have branch names or relevant git commands?	02:42
	→ knighthawk joined	02:43
	→ stany joined	02:43
	→ john2x joined	02:43
Celelaptop	Imagine an fix_foo branch from which I make a fix_bar branch.	02:43
	And suddenly I see a typo that I introduced in the fix_foo branch.	02:44
	← caspingus left	02:44
	→ git-newbie joined	02:44
Celelaptop	Then, I'd like to rebase -i branch fix_foo to fix what happened in the past.	02:44
	→ Barrayar joined	02:44
	← Barrayar left	02:44
	← john2x left	02:45
Celelaptop	It's just an hypothetic use case.	02:45
	← nemesit\|znc left	02:45
	→ k4rtik joined	02:46
Celelaptop	That I would have encountered if I hadn't been too lazy to create that new branch tonight. :p	02:46
	← git-newbie left	02:47
	← JustinMac left	02:47
	→ lavaman joined	02:47
	← lavaman left	02:47
Hello71	man git-rebase	02:48
gitinfo	the git-rebase manpage is available at http://jk.gs/git-rebase.html	02:48
	← simesy left	02:48
	← jcardinal left	02:48
	→ simesy joined	02:48
	→ Vampire0 joined	02:49
	← Vampire0 left	02:49
	→ Vampire0 joined	02:49
jshawl	I undid a rebase yesterday and now i feel like a boss	02:49
	← jshawl left	02:50
	← echo-area left	02:50
	→ el3ctron_ joined	02:50
	→ juvenal joined	02:51
	← el3ctron_ left	02:51
	→ k4rtik_ joined	02:51
	← k4rtik left	02:52
	→ Davey joined	02:52
	→ simesy_ joined	02:52
	→ saurabh joined	02:52
	← ni_ left	02:52
	→ echo-area joined	02:52
	saurabh → Guest96996	02:52
	→ ericy joined	02:53
	← Vampire0_ left	02:53
	← simesy left	02:53
	→ nemesit\|znc joined	02:55
	← CheckDavid left	02:57
	→ el3ctron joined	02:57
	← guampa left	02:58
	← nemesit\|znc left	03:00
	← liamkeily left	03:00
	→ silviof1 joined	03:01
	→ SoftwareMechani1 joined	03:02
	← refried_ left	03:03
	← juvenal left	03:04
	← silviof left	03:04
	→ Nicholas_Rishel joined	03:04
	← Nicholas_Rishel left	03:05
	← SoftwareMechanic left	03:05
	→ Etheco- joined	03:08
	→ thiago joined	03:10
	← thiago left	03:10
	→ thiago joined	03:10
	← petey left	03:10
	← pandeiro left	03:10
	→ petey joined	03:11
	← el3ctron left	03:11
	← codeman left	03:11
	→ CKyle joined	03:11
	→ el3ctron_ joined	03:12
	→ nemesit\|znc joined	03:12
	→ caspingus joined	03:14
	→ rubix joined	03:14
	← el3ctron_ left	03:14
rubix	I have a question.	03:15
	→ btiernay joined	03:15
	← petey left	03:16
btiernay	Is there a way to push a tag to a remote repository without first cloning?	03:16
	I'm trying to tag a remote repository from a CI server without having to checkout the repository (no need)	03:16
	So I just want to tag this "configuration repository" with a tag	03:17
	Any pointers would be greatly appreciated :)	03:17
	← LBV_User left	03:19
	← caspingus left	03:19
rubix	I use command: git svn clone -s https://someurl/svn/project to clone an svn repository	03:20
	which dirs and files layout like this:	03:20
	\|--aaa/ #empty	03:20
	\|--branches/ #r35, updating...	03:20
	\|--\|--file1	03:20
	\|--\|--file2	03:20
	\|--\|--file3	03:20
	\|--\|--dir1/	03:20
	\|--\|--dir2/	03:20
	\|--\|--dir3/	03:20
	\|--tags/ #empty	03:20
	\|--trunk/ #empty	03:20
	git auto checkout master to track 'trunk r1', and It's empty.	03:20
	now I get a directory 'project' which contains only '.git/'	03:20
	→ iamjarvo joined	03:20
rubix	git branch -ra outputs:	03:20
	*master	03:20
	remotes/dir1	03:20
	remotes/dir2	03:20
	remotes/dir3	03:20
	remotes/trunk	03:20
	→ john2x joined	03:21
rubix	how can I corporate with others?(they use svn)	03:21
	← nemesit\|znc left	03:21
rubix	part of ./git/config:	03:22
	[svn-remote "svn"]	03:22
	url = https://someurl/svn/project	03:22
	fetch = trunk:refs/remotes/trunk	03:22
	branches = branches/:refs/remotes/	03:22
	tags = tags/:refs/remotes/tags/	03:22
	← john2x left	03:23
rubix	Is there anyone who can answer me?	03:24
	OR... Does anyone can tell me some forums to ask this question(use git to corporate with svn)?	03:24
	→ angelsl joined	03:24
	→ nemesit\|znc joined	03:30
	→ liamkeily joined	03:31
	→ Ltk joined	03:31
rubix	anyone could answer my question? thanks	03:31
	← Ltk left	03:31
	← dhruvasagar left	03:31
Celelaptop	Sorry, I never used git svn.	03:32
rubix	all right, but thank u	03:34
hyperair	rubix: try git svn clone --stdlayout	03:35
	← ericy left	03:36
	← liamkeily left	03:36
rubix	that is the problem.	03:37
	I have done this: git svn clone -s https://url/svn/project	03:37
	the I get an empty directory: 'project'	03:37
	← bobbobbins left	03:37
rubix	git branch -ra outputs:	03:38
	*master #which points to trunk r1	03:38
	remotes/dir1	03:38
	remotes/dir2	03:38
	remotes/dir3	03:38
	remotes/trunk	03:38
	→ maestrojed joined	03:38
rubix	svn/project layout:	03:39
	\|--aaa/ #empty	03:39
	\|--branches/ #r35, updating...	03:39
	\|--\|--file1	03:39
	\|--\|--file2	03:39
	\|--\|--file3	03:39
	\|--\|--dir1/	03:39
	\|--\|--dir2/	03:39
	\|--\|--dir3/	03:39
	\|--tags/ #empty	03:39
	\|--trunk/ #empty	03:39
	← lindenle left	03:40
	→ Ltk joined	03:41
	← Ltk left	03:41
rubix	i can checkout remotes/dir1, remotes/dir2, remotes/dir3	03:41
	but how to checkout file1, file2, file3, etc...	03:41
	also, how to chekcout dir1, dir2, dir3, file1, file2, file3, etc... as a whole branch?	03:41
	→ caspingus joined	03:42
rubix	svn sucks	03:42
	→ petey joined	03:43
	→ sohail joined	03:43
	← sohail left	03:43
	→ sohail joined	03:43
	← mdiers left	03:44
	→ sabgenton joined	03:45
	← sabgenton left	03:46
	→ dhruvasagar joined	03:47
	→ sabgenton joined	03:47
	← caseydriscoll left	03:48
	← abelsiqueira left	03:48
	← knighthawk left	03:48
	→ caseydriscoll joined	03:48
fairuz	rubix \| !paste	03:49
gitinfo	Please paste snippets longer than one line at a pastebin site, such as https://gist.github.com/ rather than in-channel.	03:49
sabgenton	is there any danger or different in using git rebase master foobranch vs actual checking out foobranch then runing git rebase master ?	03:49
	does rebase master foobranch care about where you checked out in anyway?	03:50
rubix	sorry. I apologize for my unknowns.	03:51
	← daito left	03:53
	→ ericy joined	03:53
	← nemesit\|znc left	03:54
	→ mdiers joined	03:55
	→ akiva joined	03:55
	← dhruvasagar left	03:55
	→ CodeShark joined	03:56
	→ Ltk joined	03:57
	← undecim left	03:58
	← Davey left	03:58
	← Noldorin left	03:59
	→ k-man joined	03:59
	← psyl0n left	03:59
	→ Davey joined	03:59
	→ WABSRX3 joined	03:59
k-man	if i have an untracked file on a branch, what happens to it when i switch branches?	04:00
	→ davo joined	04:01
	← k4rtik_ left	04:02
	→ master_j joined	04:04
	→ rurufufuss joined	04:05
	→ liamkeily joined	04:07
	← Davey left	04:07
	→ john2x joined	04:08
	→ Davey_ joined	04:08
	← Davey_ left	04:09
	→ SoftwareMechanic joined	04:09
	→ Davey joined	04:09
	→ el3ctron joined	04:09
	→ el3ctron1 joined	04:11
	→ frogonwheels joined	04:11
	← SoftwareMechani1 left	04:12
	← john2x left	04:12
	→ el3ctron3 joined	04:14
	← Davey left	04:14
	→ el3ctron5 joined	04:14
	→ Davey joined	04:14
	→ WABSR0L joined	04:14
	← ericy left	04:16
	→ subhojit777 joined	04:17
	← CKyle left	04:17
	← liamkeily left	04:18
	← dravidan left	04:18
	→ el3ctron7 joined	04:21
	← Davey left	04:21
	→ nemesit\|znc joined	04:22
	→ Davey joined	04:22
	→ john2x joined	04:24
	← oojacoboo left	04:24
	→ Davey_ joined	04:27
	← Davey left	04:27
	s0beit → s0beit\|ZNC	04:27
	→ el3ctro10 joined	04:27
	← el3ctron7 left	04:28
	← el3ctron5 left	04:28
	← el3ctron1 left	04:28
	← el3ctron left	04:28
	← master_j left	04:28
	→ master_j joined	04:29
	← el3ctron3 left	04:29
	← Davey_ left	04:29
	→ Davey joined	04:29
rubix	:quit	04:30
	← rubix left	04:30
	→ davetoo joined	04:30
	← bitkiller left	04:30
davetoo	git-scm.com b0rked :(	04:31
	"Sorry, no Host Found"	04:31
	or... maybe it's my firefox :(	04:32
	→ ericy joined	04:32
	→ groglogic joined	04:32
davetoo	wow, in fact it is. n/mind.	04:33
	→ daito joined	04:35
	→ genehack joined	04:35
sabgenton	davetoo: you had me worried there :)	04:35
	← thiago left	04:36
davetoo	well I still don't understand what's going on.	04:36
	I can get it in Chrome, but not firefox, nor wget	04:36
	← kukks left	04:37
	← john2x left	04:37
	→ thiago joined	04:37
	→ soiitaire joined	04:38
davetoo	still debugging	04:38
	→ cads joined	04:38
davetoo	I blame New Relic	04:40
	← whitby left	04:41
davetoo	interesting.	04:42
	$ w3m -dump http://www.git-scm.com/	04:42
	Sorry, no Host found	04:42
	→ closures999 joined	04:45
davetoo	Something is giving firefox a redirect from http://git-scm.com/ to www.git-scm.com, and the server isn't configured to serve that hostname	04:46
	← daito left	04:47
	← soiitaire left	04:47
	← Guest25228 left	04:48
	→ marlboromoo joined	04:49
	← caspingus left	04:50
	→ LuxuryMode joined	04:53
	← PaulCapestany left	04:53
	← mribeirodantas left	04:54
davetoo	One of my firefox addons is sending stuff to domain-robot.org, and somehow in the process, rewriting the url to www.git-scm.com.	04:55
	→ caspingus joined	04:55
	← caspingus left	04:55
	← yhager left	04:55
	→ Davey_ joined	04:56
	→ diegoviola joined	04:56
	← groglogic left	04:56
thiago	the server is doing that	04:56
davetoo	oh?	04:56
	→ yhager joined	04:57
thiago	$ curl -s -I http://git-scm.com \| sed -n '1p;/Location/p'	04:57
	HTTP/1.1 301 Moved Permanently	04:57
	Location: http://www.git-scm.com	04:57
	← preflex left	04:57
	→ preflex_ joined	04:58
	preflex_ → preflex	04:58
davetoo	interesting; once I disabled my NoScript plugin I stopped having issues, and .. in chrome, the location bar is showing 'git-scm.com' rather than www...	04:58
	though if I add the www in chrome, I get the 'No host found' error	04:59
	→ beautifulmind joined	04:59
	← donmichelangelo left	05:00
davetoo	anyway, what I was after was... looking for the features added/changed between git 1.7.9 and 1.8.current. Cygwin's still on 1.7.9	05:00
	→ donmichelangelo joined	05:01
	← veebers left	05:01
thiago	interesting, I see the same	05:01
	the browser doesn't work, but I do get a page in curl	05:01
pekster	Smells like botched server-side code based on the user-agent, but I dunno	05:02
thiago	checking	05:02
pekster	I never put back my UA switcher plugin :)	05:02
davetoo	I think I'm looking at several interactions on my end. Wacky firefox plugin behavior among them. I don't understand why I don't get a redirect if I disable my NoScript plugin.	05:02
	regardless, this same bookmark in this same browser worked earlier today :)	05:02
thiago	huh, works now	05:03
pekster	Yup	05:03
davetoo	any of you in here had the "pleasure" of interacting with git-fusion yet?	05:03
pekster	I had to do a few refreshes	05:03
davetoo	or been forced to? :)	05:03
	that is, Perforce Git-Fusion.	05:03
	← nerdfiles left	05:04
thiago	I am getting redirected, but the page does work	05:04
davetoo	Maybe some kind of "sticky" cache and I'm still on a broken node. It'll sort itself.	05:05
Sigma	davetoo: not yet, but will have too pretty soon. Surely it must be better than git-p4 though, no? :)	05:05
davetoo	I haven't tried git-p4.	05:06
	When I walked into this new job five weeks ago they said "here, make this work" :)	05:06
	→ john2x joined	05:06
davetoo	except that .. that version didn't.	05:06
	← Watabou left	05:07
	→ rudi_s joined	05:07
davetoo	Sigma: read my travails here: http://forums.perforce.com/index.php?/forum/27-git-fusion/	05:07
	I'm 'dcarmean'	05:08
	the last thread makes me think that... well,	05:08
	← ericy left	05:08
davetoo	that my git references are going to explode within weeks as soon as they do the usual willy-nilly perforce-side cherrypicking that they like to do	05:09
	→ Watabou joined	05:10
	← jon_r left	05:10
	← griotspeak left	05:11
	← kermit left	05:12
	→ rjhunter joined	05:15
	← the_cybersphinx left	05:15
	→ jon_r joined	05:16
	→ kermit joined	05:19
	→ ryez joined	05:20
	← Bass10 left	05:20
	→ Qo027 joined	05:21
	→ CKyle joined	05:21
	← nicedice left	05:22
	→ dhruvasagar joined	05:23
	← Qo027 left	05:24
	→ Qo027 joined	05:25
	← hobodave left	05:25
	→ ericy joined	05:25
	← Qo027 left	05:25
	→ noyb joined	05:25
	→ ivar_ joined	05:28
	← ivar_ left	05:28
	← dhruvasagar left	05:28
	→ robbyoconnor joined	05:28
	→ dhruvasagar joined	05:29
	→ forrest joined	05:29
	→ daito joined	05:29
	gitinfo set mode: +v	05:30
	→ robinsmidsrod joined	05:30
forrest	Hi guys, I've screwed up a commit where I've somehow pulled in the previous commit (5e45f11, and 315b83): https://github.com/gravyboat/salt/commits/develop . I've tried several ways to fix it, but it seems to create issues because of the other file that was part of the initial commit. How can I fix this previous commit, while leaving the file associated with the commit I somehow grabbed?	05:31
	← buzzybron left	05:31
	→ etcetera joined	05:33
	← Tommy[D]_ left	05:33
	→ johnnyfuchs joined	05:34
	← johnnyfuchs left	05:34
thiago	forrest: sorry, I didn't get it	05:34
	start with what you had, what you did, and what the result was	05:34
skorgon	forrest: !fixup	05:36
gitinfo	forrest: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	05:36
forrest	Yea I was trying to avoid doing that for the sake of screwing up that previous commit	05:36
	← jon_r left	05:37
	← john2x left	05:38
	← Davey left	05:38
	Davey_ → Davey	05:38
forrest	thaigo, basically I somehow merged my commit, with a previous commit, I'll take a look at the links that were provided here.	05:38
	→ wrouesnel joined	05:38
forrest	thanks guys	05:38
rjhunter	forrest: you had two commits, you squashed them into one commit, you pushed, and now you want to separate out the commits again?	05:38
forrest	I never squashed them rjhunter	05:39
	I made changes to a file, and made my commit, which somehow lead me to creating what appears to be a 'dup' of the previous commit. I'm pretty sure this was caused by pulling down the upstream, and somehow either screwing up my merge, or making my commit, then merging the changes, which attached it to that other commit	05:40
	← Davey left	05:41
	← CKyle left	05:41
	→ Davey joined	05:41
	← Davey left	05:42
	← simesy_ left	05:42
	→ simesy joined	05:42
	→ Davey joined	05:43
	← ErKa left	05:43
	→ Dreadnaught joined	05:45
	→ Tommy[D]_ joined	05:46
dsdeiz	hey, so say this commit is a part of a merge commit. is there a way to know that? not sure if this makes sense tho :D	05:46
	say i have only the commit hash. how do i know what the commit hash of the merge commit it belongs to	05:46
rjhunter	dsdeiz: `git show` shows the parents of a merge commit	05:47
	(among other things)	05:47
	dsdeiz: oh wait, you want to go the other way? you want to find the merge commit?	05:48
	dsdeiz: that's harder (though still possible) because commits only list their parents, not their children	05:48
dsdeiz	yeah, it's the other way around	05:49
	→ marthinal joined	05:49
	← etcetera left	05:50
rjhunter	dsdeiz: if you know a branch that has the merge commit on it somewhere, you can use `git log mybranch` and look for lines that look like Merge: 315b83 a0324c	05:50
	dsdeiz: you can narrow it down with tooling, but if you're just casually looking for it that's probably the easiest	05:51
	← maestrojed left	05:51
dsdeiz	ah cool thx	05:51
rjhunter	dsdeiz: you can also say `git log mycommit..mybranch`	05:52
	→ quejajajajaja joined	05:52
quejajajajaja	Hello	05:52
gitinfo	quejajajajaja: 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.	05:52
	← petey left	05:52
quejajajajaja	On Gihub, if I am forking a programming and creating a setup file, what version do I specify if there is no version given in the program I am forking?	05:52
	→ dunpeal joined	05:52
	→ jaimef joined	05:52
	← dhruvasagar left	05:52
dunpeal	What's the difference between `--all` and `--mirror` for git push?	05:53
	→ dhruvasagar joined	05:53
thiago	quejajajajaja: version?	05:53
quejajajajaja	thiago, the version of the program	05:53
thiago	dhruvasagar: --all pushes refs/heads/*	05:53
quejajajajaja	in setup.py I need to include a version	05:53
thiago	dhruvasagar: --mirror pushes refs/* and also deletes	05:53
	quejajajajaja: non-Git question. Git does not know anything about versions.	05:54
dhruvasagar	thiago: ?	05:54
dunpeal	thiago: so --mirror will also push tags and remote branch references?	05:54
quejajajajaja	well thiago, it is closely related	05:54
thiago	dhruvasagar: oops, wrong person :-)	05:54
	dunpeal: yes	05:54
dhruvasagar	thiago: np	05:54
thiago	quejajajajaja: no, it isn't. The version is something you define. Write whatever you want.	05:54
dunpeal	thiago: is there anything else significant that --mirror will push and --all will not?	05:54
quejajajajaja	ok	05:55
davetoo	wow, I'm impressed. I got git to compile under cygwin, and it's been passing tests for a couple of minutes now.	05:55
rjhunter	davetoo: congratulations :-)	05:55
thiago	dunpeal: tags, remote branches, anything else under refs/ that usually isn't exposed (backup filter-branch, etc.)	05:55
davetoo	..to the git devs :)	05:55
thiago	dunpeal: what for? There are binaries for msys/mingw.	05:55
dunpeal	thiago: so --all, followed by a --tags push, should be sufficient in most cases... thanks.	05:56
thiago	dunpeal: yes	05:56
	dunpeal: note that --mirror deletes branches that don't exist locally. --all has no such behaviour.	05:56
	← yhager left	05:57
	← madewokherd left	05:57
dunpeal	nod, thanks.	05:57
davetoo	sweet! even gitk is working!	05:58
thiago	quejajajajaja: do you want to extract something from git that might be used as a version?	05:58
quejajajajaja	thiago, yes	05:58
thiago	quejajajajaja: see the output of git describe	05:58
quejajajajaja	thank you	05:59
	→ maestrojed joined	05:59
	← jaimef left	05:59
forrest	Thanks for your help guys.	06:00
	← forrest left	06:00
	→ jaimef joined	06:00
	← aartist left	06:00
davetoo	wow. Looking at git's repo in gitk is.. impressive.	06:01
	← dhruvasagar left	06:02
	← tpollard left	06:02
	→ dhruvasagar joined	06:03
	→ kukks joined	06:03
davetoo	.. the github network diagram is equally impressive	06:05
	→ purpleidea joined	06:06
	→ HELLC joined	06:07
	→ WaReZ joined	06:08
	← jAyenGreen left	06:08
	→ aartist joined	06:08
purpleidea	hi there, i have a question about branches: i'm currently working on git master and i've made some changes to the code, some of which i've added (with git add -p) and some changes are still unmodified. I decided i want to start a new feature branch from the last commit. I've tried stashing, and then branching in an attempt to get a clean tree to work on. I'd like the current WIP changes to be a separate feature branch too if possible. What are the c	06:09
	← Dougie187 left	06:09
	← daito left	06:14
	← ericy left	06:14
	→ jaseemabid joined	06:16
	→ sgo11 joined	06:16
	→ a-l-e joined	06:16
	← closures999 left	06:16
sgo11	why git does not sync my empty directories? they are important to hold temp files.	06:16
purpleidea	sgo11: git doesn't track empty dirs	06:16
	→ closures999 joined	06:17
purpleidea	sgo11: you can put a .foo dot file inside and add it if you need a dir. also maybe a README file or similar.	06:17
	→ k4rtik joined	06:18
	← jaseemabid left	06:18
sgo11	purpleidea, I just don't understand why git does not do this. track file all permissions and empty dirs should be very easy. but git just ignore all these important changes.	06:18
	→ jaseemabid joined	06:18
	← jaseemabid left	06:18
purpleidea	sgo11: what are you actually trying to do?	06:18
	→ rouf joined	06:18
	→ ewnd9 joined	06:19
purpleidea	sgo11: this is part of git design. I can't change that. I've offered you a work around. If you describe the actual problem you're trying to solve, I might be able to help. See: "x y problem".	06:19
	→ JustinMac joined	06:19
haarg	git tracks content, not filesystems	06:20
	← Imagineer left	06:21
sgo11	purpleidea, ok. thanks a lot. I will use your workaround. I asked file permissions problem the day before yesterday. git ignores my file permissions change. I have to write a script to update those permission change after git pull. anyway, I have to accept the current design anyway.	06:21
	← closures999 left	06:21
	← noyb left	06:21
	→ grahamsavage joined	06:22
	← \|Starcraftmazter left	06:22
rjhunter	sgo11: you may be using git for something that git isn't very good at.	06:22
moritz	!deploy	06:22
gitinfo	Git is not a deployment tool, but you can build one around it(in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/the-list-and-irc/deploy.html	06:22
grahamsavage	hey. i'm looking for a graphical UI for git for ubuntu.. i've got git cola but it's nowhere near as good as osx's tower	06:23
	git's awesome for deployment	06:24
	don't let the nay-sayers tell you otherwise	06:24
	rjhunter says "nay"	06:24
sgo11	well. I understand git is not a deployment tool. but I just don't understand why git can not be a deployment tool. file permission track should be very easy to handle. I am very happy with git for my current work except not tracking empty dirs and file permissions. but anyway, sorry about the complain. I will accept the current design. thanks a lot again.	06:24
	← davetoo left	06:24
	← dreiss left	06:25
grahamsavage	sgoll: it's very easy to use as a deployment tool.. we've got it setup so we can deploy multiple apps to multiple servers	06:25
	sgoll: file permissions you build on top off git	06:25
	← Gamecubic left	06:26
grahamsavage	ok so basic approach i like to go with.. is on the server create a cache diretory which has like /srv/urcompany/repositories/appname and set up your deployment to push to that	06:26
rjhunter	purpleidea: did you figure out your branch/WIP situation?	06:26
purpleidea	rjhunter: i did not	06:26
	← LuxuryMode left	06:26
	← Goplat left	06:27
	← iamjarvo left	06:27
	← akurilin left	06:27
grahamsavage	sgoll: then when you deploy.. basically do a cp -R into a new directory excluding the .git directory... run a script to change all the file permissions (or even better use setfacl to set the perms on the parent directory and just have the majority of them inherit).	06:27
	→ iamjarvo joined	06:27
	← threesome left	06:27
grahamsavage	run any scripts you need too (processing js files / whatever).. then update the symlink that points at the current working copy	06:28
	→ wirrbel joined	06:28
sgo11	grahamsavage, ok. thanks. I am currently using a script to do these jobs. update permission and create empty dirs.	06:28
grahamsavage	sgoll: empty directories.. you have two options.. either you can just stick a .GIT_KEEP file in each dir	06:28
	and commit that	06:29
	or create them manually.. we symlink directories we don't want to get blown away each release	06:29
purpleidea	rjhunter: do you have any tips?	06:29
	(or can you describe the method)	06:29
	→ daito joined	06:29
grahamsavage	sgo11: https://gist.github.com/anonymous/550fcf27c6341eeb5540 << that's what we use for setting the permissions for the apps	06:30
rjhunter	purpleidea: i'm trying to get my head around your exact situation, but here's one option that might be illustrative:	06:30
sgo11	grahamsavage, ok. thanks a lot. finally, you type my nick correct. :)	06:30
	→ Search12 joined	06:30
grahamsavage	sgo11: lol.. tab tab	06:30
purpleidea	rjhunter: (it's sort of like: i've been hacking on master instead of doing feature branches, but now i want to work on different features in parallel, and the current work-in-progress should turn into a feature branch)	06:31
	← kukks left	06:31
sgo11	grahamsavage, ^_^	06:31
	← iamjarvo left	06:32
rjhunter	purpleidea: commit your work (to master is fine). make a new branch "feature1". cherry-pick the relevant commits for "feature1" from master. do the same for feature2. reset "master" to origin/master.	06:32
	← dhruvasagar left	06:32
	← wirrbel left	06:32
	→ dhruvasagar joined	06:32
	→ ericy joined	06:32
	→ User458764 joined	06:32
rjhunter	purpleidea: does that make sense? does it apply to your situation?	06:32
	→ JohnBat26 joined	06:33
purpleidea	rjhunter: cool idea, but i don't think it does because i'm not ready to 1) commit my changes, and 2) i still want to keep some of the code staged, and some unstaged... it was my mistake for hacking in place, instead of working on a feature branch.	06:33
rjhunter	purpleidea: seriously, don't worry about "ready to commit". just make sure the commit messages say "WIP" on them so you know not to push them.	06:34
	← JohnBat26 left	06:34
purpleidea	i tried to git stash, and then feature branch, and other things, but nothing worked as expected :P i hit the maximum of my git skills, and now i need to learn more.	06:34
	rjhunter: but i want to unstage them all later, and i'd like to keep the index correct...	06:34
	→ maleknet joined	06:36
	→ JohnBat26 joined	06:36
	→ rob_w joined	06:37
rjhunter	purpleidea: turn the index you want to keep into a commit. turn the work tree you want to keep into another commit. label them appropriately so you can find them again later. when you're ready, use git checkout and git reset to put the working tree and index back to the state you want	06:38
purpleidea	rjhunter: okay, i'm going to try this...	06:38
rjhunter	purpleidea: you can throw away those WIP commits once you're ready to replace them with "real" ready-to-push commits	06:38
purpleidea	(cp -a git-thing git-thing.backup) ;)	06:38
	→ john2x joined	06:39
rjhunter	purpleidea: yay for safe learning :-)	06:39
	← joelrobichaud left	06:39
	← john2x left	06:40
	→ ThomasLocke joined	06:40
	← aartist left	06:40
	→ john2x joined	06:40
purpleidea	rjhunter: two commits made. now i branch, right? then what was the next step sorry? (i've never cherry-picked)	06:40
	(i feel like a git virgin saying that)	06:41
rjhunter	purpleidea: cherry-pick is really just the same as you looking at some change and doing the same thing yourself, committing the result.	06:41
	→ noyb joined	06:41
rjhunter	purpleidea: like applying a patch + committing with the same message	06:41
	purpleidea: so branch feature1 from origin/master, check it out, then look through `git log master` (or wherever your mixture was) to find the relevant commits	06:42
purpleidea	rjhunter: done...	06:43
	→ lite_ joined	06:43
	← rob_w left	06:43
	← simesy left	06:44
rjhunter	purpleidea: and I guess you can do the same for feature2	06:44
purpleidea	rjhunter: (reading cherry pick docs...)	06:44
	rjhunter: feature i can do after i sort this all out... it hasn't been started yet which is good.	06:44
	→ simesy joined	06:44
	← john2x left	06:44
purpleidea	rjhunter: so according to cherry-pick it applies git commits... however on my new feature branch (since it's branched from master) it already has the two XXX WIP commits applied ...)	06:45
	did i need to reset the branch somewhere first? or?	06:45
	or branch from earlier on ?	06:46
rjhunter	purpleidea: it may not have been clear, but when I suggested branching feature1 from origin/master, that's actually different from master	06:46
	→ hoschi joined	06:46
purpleidea	rjhunter: oh. so i branch from the commit BEFORE my two XXX WIP commits, right?	06:47
rjhunter	purpleidea: you can literally say "origin/master"	06:47
purpleidea	(which is where origin/master points of course	06:47
rjhunter	purpleidea: `git checkout -b feature1 origin/master`	06:47
	← sgo11 left	06:47
	← User458764 left	06:47
rjhunter	purpleidea: you can change it after the fact, of course (with git reset)	06:48
	purpleidea: but i figure we'll walk through doing it "cleanly"	06:48
	← simesy left	06:48
	→ jargon joined	06:49
	← dsdeiz left	06:49
purpleidea	rjhunter: okay, i've figure it out (that part is done) and i understand what's going on. Pedantic question: "is it more technically correct to branch from some sha1 than origin/master (in case i hadn't pushed some commits to origin?)"	06:49
	→ rob_w joined	06:50
	→ kukks joined	06:50
	→ dsdeiz joined	06:50
rjhunter	purpleidea: git branch (or git checkout -b) always resolves whatever you give it to a sha1	06:50
	→ closures999 joined	06:51
rjhunter	purpleidea: there are some options around "tracking upstream" that like to know what the branch was called if you want to know things like your branch is ahead of upstream by 3 commits	06:51
purpleidea	rjhunter: okay. so i'm ready to cherry pick, and i know the sha1's of the two commits. How do I get them to be the unstaged, and staged indexes respectively? (instead of just committing them on)	06:51
	rjhunter: got it.	06:51
	← Search12 left	06:52
	→ matanya joined	06:53
	→ jon_r joined	06:53
	→ lindenle joined	06:53
rjhunter	purpleidea: so just to clarify, you have just the two commits (one destined for the work tree, and one for the index)?	06:54
	purpleidea: in that case, cherry-pick won't be necessary	06:54
purpleidea	rjhunter: correct.	06:54
	→ Raging_Hog joined	06:54
rjhunter	purpleidea: sorry, i thought you had a few "real" commits already	06:54
	purpleidea: cherry-pick is good to bring those in	06:55
SamB	that sounds suspiciously like "git stash"	06:55
purpleidea	(terminology check: index means "things that are staged?)	06:55
	SamB: i thought so too, but I was never able to "make it work".	06:55
rjhunter	purpleidea: yes, the index is "the staging area"	06:56
purpleidea	rjhunter: cool.	06:56
	→ pantsman joined	06:57
	← WaReZ left	06:57
	→ WaReZ joined	06:58
	→ l0cust joined	06:58
	← harish left	06:59
purpleidea	rjhunter: so what should i do now?	06:59
l0cust	Hey, y'all. Friend and I are having an argument as to who has done more work on our git-managed project. Is there a git command that would give relevant statistics?	06:59
purpleidea	l0cust: git log	07:00
	→ iamjarvo joined	07:00
purpleidea	l0cust: or maybe: http://gitstats.sourceforge.net/	07:00
	→ yhager joined	07:00
rjhunter	purpleidea: so you can use `git checkout --patch 315b83` to check out the contents of commit 315b83 to your working tree. I suggest starting with what you eventually want in your index.	07:01
	→ eijk joined	07:01
purpleidea	rjhunter: don't i want to start with what should be staged first?	07:02
grahamsavage	holy crap	07:02
	smartgit is awesome	07:02
	i just tested out 8 git clients.. smartgit really takes the cake	07:02
	← ewnd9 left	07:02
	→ pen joined	07:03
l0cust	purpleidea: thank you, gitstats seems to be what I was looking for	07:03
	→ nbrosnahan joined	07:03
purpleidea	l0cust: np	07:03
	l0cust: (i've never tried it, so ymmv)	07:03
quejajajajaja	hey guys, i cloned a project from my github page and then I made changes to the clone on my computer	07:03
	← nbrosnahan left	07:03
rjhunter	purpleidea: yes. you can put it straight in your index (skipping the working tree) with `git reset --soft`, but that also updates HEAD (which you may or may not want)	07:03
quejajajajaja	and tried pushing it, but no changes were made	07:03
	git push origin master	07:03
	is the command i used	07:03
	I created a new directory structure and a set up file	07:03
	why isnt it showing up on my github?	07:04
	← jon_r left	07:04
grahamsavage	quejajajajaja: you done a commit?	07:04
	type git status.. do you see any files?	07:04
quejajajajaja	grahamsavage, how do I do a commit?	07:04
	git commit ?	07:04
grahamsavage	yep	07:04
rjhunter	purpleidea: you can use `git checkout --patch desiredindex` and `git add`, then another `git checkout --patch desiredworkingtree`	07:04
	← h0rrorvacui left	07:04
grahamsavage	find an intro tutorial for git somewhere... some of the things can be confusing at first.. but a 15min tute will show you branching merging pushing pulling etc	07:05
	→ nbrosnahan joined	07:05
	→ Cromulent joined	07:05
	← iamjarvo left	07:05
	→ soee joined	07:06
rjhunter	quejajajajaja: the git main workflow is: repeat ( change files in your work tree, stage the changes for commit, commit ) and then once you're happy you push those commits to someone else (like github)	07:07
purpleidea	rjhunter: just working through this... the first one prompted me with do you want to add this, sort of like git add --patch... y,y,y,y,y,y... all went well...	07:07
rjhunter	purpleidea: you can specify a path (like .) instead of --patch if you'd rather not check the changes as you go	07:08
	→ rath joined	07:09
purpleidea	rjhunter: oh that's okay. SO: git d --cached in the original git backup folder, matches the git d --cached in the branch folder.	07:09
	→ LionKing joined	07:09
purpleidea	GREAT news so far.	07:09
	rjhunter: now i do the SECOND checkout, but i don't want to add any of the files to the index. do i just do n,n,n,n,n ?	07:09
	← ericy left	07:10
	← sohail left	07:10
purpleidea	rjhunter: so doing n,n,n,n is as if nothing happened.	07:11
	Apply this hunk to index and worktree [y,n,q,a,d,/,K,g,e,?]? n	07:12
	(i just want to apply to worktree and not index actually)	07:12
	→ ewnd9 joined	07:12
	← noyb left	07:12
	→ cbreak-work joined	07:13
purpleidea	rjhunter: Maybe I could just copy the files from git master (in a different folder) to this git branch	07:13
rjhunter	purpleidea: sorry, i might have misled you again -- i forgot that `git checkout` (and everything, actually) always modifies the index on the way to the worktree	07:13
purpleidea	rjhunter: being mislead here is okay. i have backups, and i'm learning.	07:13
	rjhunter: also, i think this works. I just now have to copy the files over from the master...	07:13
	← robbyoconnor left	07:14
rjhunter	purpleidea: but if you throw away the current changes, then you should get what you want with something like `git checkout worktreesha .` `git reset indexsha .`	07:14
	→ robbyoconnor joined	07:14
purpleidea	rjhunter: you lost me there...	07:14
	and actually, it won't let me checkout master because: "	07:15
	error: Your local changes to the following files would be overwritten by checkout:	07:15
	← Dreadnaught left	07:15
	← Cromulent left	07:16
rjhunter	purpleidea: yup, so let's clean up first -- `git reset .` to put the index back to HEAD (unstage everything). Then `git checkout .` to put the worktree back to the index.	07:16
	← steveeJ left	07:16
	← pen left	07:17
rjhunter	(those two commands are a handy pair, and are useful quite often)	07:17
purpleidea	rjhunter: actually it's not back to the clean branch... i can do this though.	07:18
rjhunter	purpleidea: `git clean` is sometimes a handy third item in that 'cleanup' set	07:19
	→ pen joined	07:19
	← robbyoconnor left	07:20
	→ robbyoconnor joined	07:20
purpleidea	rjhunter: okay i restored my backup :P should i still do the two inital WIP commits ?	07:21
	← Grant484 left	07:21
	→ Grant484 joined	07:22
	← pantsman left	07:22
rjhunter	purpleidea: yup, sure -- the sequence should look something like `git commit -m 'WIP: the index for feature1' && git add . && git commit -m 'WIP: the work tree for feature1' && git checkout -b feature1 origin/master && git checkout master:/'work tree' && git reset master:/index .`	07:24
purpleidea	git checkout master:/'work tree' <- master:/sha1_of_work_tree right?	07:25
rjhunter	the :/ syntax is one of the many ways you can specify a sha	07:26
purpleidea	rjhunter: ah	07:26
	okay let me try this	07:26
rjhunter	it means the first commit which has a commit message matching this pattern	07:26
purpleidea	TIL	07:26
	→ ericy joined	07:26
purpleidea	makes sense. / is search	07:27
rjhunter	exactly	07:27
purpleidea	OH. it clicked why you did that. FANCY	07:27
	rjhunter: you	07:27
	're spoiling me	07:27
rjhunter	:-)	07:28
	→ dreiss joined	07:28
	→ harish joined	07:28
rjhunter	I need to head off, but I'll join from my phone in case I've sent you in a horribly wrong direction or something :-)	07:29
	→ jarr0dsz joined	07:30
	→ juril joined	07:30
purpleidea	rjhunter: actually the search didn't work, but i can do it manually...	07:30
	→ rhunter joined	07:30
	→ netiul joined	07:31
	← netiul left	07:31
	→ netiul joined	07:31
	← rjhunter left	07:32
Nevik	hmmmm	07:32
	does anyone know who maintains git-scm.com?	07:33
	← stany left	07:33
	→ arjun_ joined	07:33
Nevik	the site was unreachable for me (timeout) last night, and now i'm getting a simple html response saying "Sorry, no Host found"	07:33
	← dhruvasagar left	07:33
	← _W_ left	07:33
	→ stany joined	07:33
	→ FrodoTheHobbit joined	07:34
j416	Nevik: http://downforeveryoneorjustme.com/git-scm.com	07:34
	→ dhruvasagar joined	07:34
	← jarr0dsz left	07:34
Nevik	j416: yes, that's why i want to investigate it :P	07:34
	← nbrosnahan left	07:34
Nevik	j416: can you do a ping real quick and see what ip you get for it?, just to double check	07:34
purpleidea	rhunter: is this rjhunter?	07:35
j416	Nevik: https://github.com/github/gitscm-next	07:35
	→ kjellski joined	07:35
j416	Nevik: if I read the "Contributing" part right, apparently git-scm.com is written by... me	07:35
	quite a surprise	07:36
Kartagis	anyone know how repos can be created via web? bash commands?	07:36
purpleidea	@later tell rjhunter the first checkout part didn't work, but thank you for all the help!	07:36
	→ rjhunter joined	07:36
purpleidea	rjhunter: the first checkout part didn't work, but thank you for all the help!	07:37
rjhunter	purpleidea: yes, this is rjhunter but I'm on a phone now so it's harder to type effectively	07:38
j416	Nevik: 75.101.145.87	07:38
purpleidea	rjhunter: nobody should have to do git on a phone! i appreciate all the help though	07:38
rjhunter	... Or stay connected, for that matter	07:38
	← stany left	07:38
Nevik	hm	07:38
j416	looks like it's heroku	07:39
	→ stodan joined	07:39
Nevik	i get a different (single) ip from my local cache and google dns	07:39
	← beawesomeinstead left	07:39
Nevik	opendns has the one you mentioned	07:39
	that is rather.... weird	07:39
j416	so use that one? :)	07:39
Nevik	the thing is, my router is not set up to use google dns, but the one my ISP provides	07:39
j416	strange	07:40
Nevik	sooo somehow faulty DNS entries go spread around, it seems	07:40
j416	thanks for reminding me though to use google DNS	07:40
	← rhunter left	07:40
purpleidea	rjhunter: i'm going to read a bunch of git stuff to learn a bit more and come up with a solution!	07:40
Nevik	lol	07:40
	→ beawesomeinstead joined	07:40
j416	Nevik: I've been meaning to change the dns settings for a while but I keep forgetting	07:41
	← arjun_ left	07:41
	→ cff joined	07:41
j416	the DNS my ISP provides isn't very stable	07:41
Nevik	j416: i recommend not using google dns	07:41
j416	really	07:41
	why?	07:41
rjhunter	purpleidea: awesome :)	07:41
Nevik	ive heard it's somewhat unreliable	07:41
j416	tss	07:41
Nevik	j416: most especially that they dont honor dns timeouts	07:41
j416	Imma use it anyway	07:41
Nevik	lol sure :P	07:41
	why not use opendns :P	07:41
j416	better?	07:41
	hah, just found it	07:42
	their site says "The safest, smartest, fastest and most reliable DNS on the planet."	07:42
purpleidea	rjhunter: i owe you some puppet help if you need it!	07:42
j416	I'm sold	07:42
	talk about falling for marketing	07:42
	lol	07:42
	brb	07:42
	← diosmiux left	07:43
	→ j416_ joined	07:44
	→ h_kon joined	07:44
	j416_ → Guest2667	07:44
	j416 → Guest21318	07:44
	← ericy left	07:45
rjhunter	purpleidea I'm not using much Puppet right now but who knows what the future might bring :)	07:45
	Guest2667 → j416	07:45
	← TomyLobo left	07:45
	← j416 left	07:45
	→ j416 joined	07:45
purpleidea	rjhunter: indeed! thanks again	07:45
j416	sigh, silly nick implementation	07:45
	Nevik: thanks	07:45
	Nevik++	07:45
	→ mishok13 joined	07:45
	← pheaver left	07:45
	j416 off to work o/	07:45
Nevik	lol	07:46
	j416: uhhh glad i could help? xD	07:46
Kartagis	http://www.git-scm.com/docs/gitweb gives me Sorry, no host found	07:46
Nevik	Kartagis: same here. there's probably something up at their host	07:47
	← Guest21318 left	07:47
Nevik	it's a problem with getting the wrong ip from dns, so probably heroku changed IPs or something	07:47
	← dreiss left	07:50
	→ pdtpatrick joined	07:52
	← pdtpatri1k left	07:53
	→ Sylario joined	07:54
	→ jnavila joined	07:54
	→ alorence joined	07:54
	← beautifulmind left	07:54
	← a-l-e left	07:54
	→ simesy joined	07:55
	robbyoconnor → r0bby	07:56
	r0bby → r0bby_	07:56
	r0bby_ → robbyoconnor	07:56
	← kukks left	07:57
	→ _W_ joined	07:59
	→ katterstrophe joined	07:59
	← Grant484 left	07:59
	→ Xethron joined	07:59
	← simesy left	07:59
	← Xethron left	07:59
	→ Xethron joined	07:59
	← harish left	08:02
	→ ns5 joined	08:02
ns5	how to change author of all commits?	08:03
Kartagis	anyone know how repos can be created via web? bash commands?	08:04
	← Norm_ left	08:04
	← k4rtik left	08:04
	→ Grant484 joined	08:05
	→ nbrosnahan joined	08:05
	→ harish joined	08:05
	→ threesome joined	08:06
	← purpleidea left	08:06
	→ bvi joined	08:06
	→ purpleidea joined	08:08
	← purpleidea left	08:08
	→ purpleidea joined	08:08
canton7	Kartagis, local repos? repos on github?	08:08
Kartagis	canton7: repos on my server	08:08
canton7	Kartagis, gitolite can be set to create a repo the first time you push to a particular address	08:09
	← lacrosse_ left	08:09
Kartagis	canton7: hrm, do you know of drupal?	08:09
canton7	I do. python framework right?	08:10
Kartagis	no, php	08:10
canton7	ah yeah, getting it mixed with django	08:10
	→ jon_r joined	08:10
	gitinfo set mode: +v	08:10
	→ gua_ould joined	08:10
Kartagis	I was looking to create a similar functionality. on their website, you create a project, hence a repo	08:11
	want a screenshot?	08:11
	← gua_ould left	08:11
canton7	they're probably calling 'git init --bare' somehow, either via system() of via one of the many git wrappers for php	08:12
	→ gua_ould joined	08:13
	→ GeertV joined	08:13
	← Exdaix left	08:15
	← tirengarfio_ left	08:15
	→ tirengarfio joined	08:16
	→ thierryp joined	08:16
	← katterstrophe left	08:18
	→ armin_w joined	08:18
	→ harobed_ joined	08:18
	← harobed_ left	08:18
	→ harobed_ joined	08:19
	→ h0bbit_ joined	08:20
	→ ntio joined	08:20
	← joaoh82 left	08:20
	→ Asenar joined	08:20
	→ irqq joined	08:21
	→ joaoh82 joined	08:21
	← mitz left	08:23
	→ simesy joined	08:23
	→ mitz joined	08:23
	→ bkhan joined	08:24
	→ Gonzo89 joined	08:25
	→ Sonderblade joined	08:25
	← Sonderblade left	08:26
	→ Sonderblade joined	08:26
	← thiago left	08:27
	→ TomyWork joined	08:30
	→ stany joined	08:33
	← stany left	08:33
	→ TaylanUB joined	08:34
	→ stany joined	08:34
	← stany left	08:34
	→ freimatz_ joined	08:34
	freimatz_ → freimatz	08:36
	→ tvw joined	08:36
	← maestrojed left	08:38
	→ jarr0dsz joined	08:38
	← freimatz left	08:38
	← master_j left	08:39
	→ nafg___ joined	08:39
jarr0dsz	got disconnected, having some troubles getting git repo to work correctly after removing the .git from my local project	08:40
milki	jarr0dsz: the .git is pretty important	08:40
jarr0dsz	i removed it because there where some issues on an older project and wanted to start fresh	08:40
milki	jarr0dsz: hope you have a !backup	08:41
gitinfo	jarr0dsz: 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	08:41
	→ gasbakid joined	08:41
milki	jarr0dsz: do you have a remote you can just clone again?	08:41
jarr0dsz	no need for the commit history so removed .git and copied the config from an backup now try to deploy again but this fails with several errors	08:41
milki	o	08:41
	jarr0dsz: you can just git init then	08:41
	→ adac joined	08:41
milki	wait, what does deploy do here?	08:41
jarr0dsz	yes i have done git init locally and copied the config from a backup ( which hold the remote git repo info )	08:41
	→ slassh joined	08:41
jarr0dsz	git push origin master goes okay but let me gist the error trace	08:42
milki	can you step back and explain what you are trying to do	08:42
Kartagis	I have /var/www/git/my-project.git on the server owned by git:apache and of which perms is 755. I try to push to it, and I get fatal: 'my-project.git' does not appear to be a git repository	08:42
milki	Kartagis: what is your git url?	08:42
Kartagis	milki: origin?	08:43
milki	Kartagis: yes, git remote -vv	08:43
jarr0dsz	anyone can help me fix these? a bit lost on this https://gist.github.com/anonymous/7688900	08:43
	← thierryp left	08:43
	→ rhunter joined	08:43
jarr0dsz	i have the remote repo still there but not with latest changes	08:43
milki	jarr0dsz: git init doesnt give you a remote	08:43
	→ thierryp joined	08:43
milki	jarr0dsz: git remote -vv?	08:43
jarr0dsz	preferrable i would like to clearout the remote repo so i can do just git push origin master clean	08:44
Kartagis	origin [email@hidden.address] (fetch)	08:44
	origin [email@hidden.address] (push)	08:44
	milki: ^	08:44
jarr0dsz	git remote -v yes i already checked that is correct 2 entry's "origin fetch and push "	08:44
milki	Kartagis: and how is that supposed to map to /var/www/git/my-project.git?	08:44
jarr0dsz	with correct path to my remote .git repo	08:44
	→ jeff joined	08:44
	→ Eridius joined	08:44
	→ he_re joined	08:44
	← Eridius left	08:44
	→ Eridius joined	08:44
	← jeff left	08:44
	→ jeff joined	08:44
	→ lllama joined	08:44
jarr0dsz	so i changed git reset --hard origin/master to origin only but fails also	08:44
	→ kenyon joined	08:45
jarr0dsz	assuming it will take the master branch automatically	08:45
milki	jarr0dsz: did you fetch?	08:45
	→ intellix joined	08:45
jarr0dsz	milki i dont want to fetch i want to write my local changes to remote repo, i don't want the remote repo data anymore	08:45
milki	jarr0dsz: you cant refer to origin/master until you have fetched	08:45
	→ Hamled joined	08:45
Kartagis	milki: it should be [email@hidden.address] ?	08:45
jarr0dsz	hmz okay, can i just reset to origin? instead of origin/mater	08:46
	or would it be better removing the remote repo and start from scratch?	08:46
milki	jarr0dsz: no that wont work	08:46
	Kartagis: /var/www/git/my-project.git would definitely work	08:46
jarr0dsz	fetching is not an option there is an older version on the remote git repo, is there any shortcut i can do to fix and start cleanly?	08:46
	→ Misan joined	08:47
milki	jarr0dsz: why do you need to reset?	08:47
jarr0dsz	perhaps remove the remote git repo init bare it again then push all my local changes?	08:47
milki	jarr0dsz: you can just git push -f probably	08:47
jarr0dsz	already did that too milki pushing is not an issue that works	08:47
	← rhunter left	08:47
milki	jarr0dsz: then i totally dont understand your situation	08:47
jarr0dsz	i need to reset cause i have some issues i could not track back, so i copied an working older backup to my working station	08:47
	← harish left	08:48
jarr0dsz	i can push.. but not deploy it fails with https://gist.github.com/anonymous/7688900	08:48
milki	what in the world is deploy	08:48
jarr0dsz	because locally i have a new git init, as you stated this reference to origin/master would not work without a fetch which is non wanted	08:48
	so not quite sure how i best would fix now	08:48
milki	if you cant get someone else to understand, starting over from scratch would probably solve your problems	08:49
Kartagis	milki: that worked, thanks	08:49
jarr0dsz	never mind the deploy i have capistrano non git related the problem is it tries to git reset --hard origin/master	08:49
	→ halu joined	08:49
milki	well, wherever its doing that, it doesnt know what your remotes are	08:49
Kartagis	milki: can I set a path so that git would be relative?	08:50
	← purpleidea left	08:50
Kartagis	git/*	08:50
milki	Kartagis: it defaults to the relative home of your remote user if you dont put a leading /	08:50
jarr0dsz	git branch shows me *master	08:50
milki	Kartagis: jarr0dsz that has nothing to do with your remote	08:50
	er	08:50
	mixing up people here now	08:50
jarr0dsz	locally but argh don't know what the issue is even sa bit hard to look for a solution then ;p	08:51
	→ hahuang65 joined	08:51
	→ lb1a joined	08:51
milki	do you even know where these errors are?	08:51
	are they remote? are they local?	08:51
	→ purpleidea joined	08:51
Kartagis	milki: so [email@hidden.address] would work as well?	08:51
	← ianmcorvidae left	08:52
milki	Kartagis: no, you have a leading / now	08:52
	→ ianmcorvidae joined	08:53
	→ ericy joined	08:53
Kartagis	milki: in another project, they tell me to set the url [email@hidden.address] I want a similar structure. how should my fs structure be?	08:54
	→ mastro joined	08:54
milki	Kartagis: you need to find the homedir of your git user	08:54
	then put my-project.git there	08:54
Kartagis	milki: in such a structure, are these users real?	08:55
	← Grant484 left	08:55
Kartagis	or can they be virtual?	08:55
milki	must be a real user	08:55
	you are using ssh	08:55
	← johnkevinmbasco left	08:55
milki	no such thing as a virtual user with ssh	08:55
Kartagis	hrm	08:56
milki	Kartagis: you might be looking at gitolite, but you dont seem to know linux enough to use that	08:56
	→ Grant484 joined	08:56
	→ psoo joined	08:56
Kartagis	so, a $HOME is essential	08:56
	→ Balliad joined	08:56
Kartagis	milki: I know linux enough (I think it is enough)	08:56
	just, I didn't know git pushed over ssh	08:57
	← psoo left	08:57
Kartagis	then again, I should have gotten that from [email@hidden.address]	08:57
	damn it	08:57
	→ psoo joined	08:57
Kartagis	(not you, me)	08:57
	← lllama left	08:57
milki	Kartagis: well, if you can, check out !gitolite if you want those nice paths	08:58
gitinfo	Kartagis: 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/	08:58
milki	im going to bed -.-	08:59
	← hahuang65 left	08:59
	← angelsl left	09:02
	← Rob___ left	09:03
	← ns5 left	09:04
	← el3ctro10 left	09:04
	→ ns5 joined	09:06
	→ el3ctron joined	09:06
Cathy	[01:55:57] <milki> no such thing as a virtual user with ssh	09:06
	of course there is	09:06
	← yhager left	09:06
Cathy	ssh is commonly used to log into shell accounts on the system, but it obviously doesn't have to used that way	09:06
	it could be backed by an arbitrary authentication paradigm	09:06
dalias	well the sshd implementation depends on there being a unix account corresponding to the client-requested username	09:07
Cathy	you mean the OpenSSH implementation specifically	09:08
grawity	AFAIK, Launchpad has a custom sshd written in twisted-conch that does only virtual users	09:08
	← ns5 left	09:08
dalias	cathy, and dropbear	09:08
Cathy	nothing stops you from writing your own sshd anyway	09:08
dalias	altho dropbear may support some virtual user thing if you run the daemon as non-root	09:08
	i forget	09:08
grawity	so yes, most common implementations require this, but the protocol certainly does not...	09:09
Cathy	and i'm not actually sure it's true that OpenSSH requires shell accounts	09:09
	OpenSSH allows arbitrary PAM authenticaiton	09:09
dalias	it probably doesn't	09:09
Cathy	you could write a PAM module that is not backed by shell accounts	09:09
dalias	if you use pam creatively	09:09
	altho PAM is at the system level	09:09
grawity	Cathy: it does require that the account could be looked up with getpwnam() and similar interfaces, as far as I know	09:09
dalias	so the admin would be responsible for setting up the non-unix-account ssh setup	09:09
grawity	Cathy: because it needs to switch the server process to some UID	09:10
Cathy	you could replace those functions with your own shared library when you load sshd	09:10
	→ ns5 joined	09:10
	← ns5 left	09:10
grawity	Cathy: but then one could just write a NSS (libnss_*) module that returns all tohose 'virtual' users	09:10
Cathy	good idea	09:10
grawity	like the existing ones for LDAP, /etc/passwd, Hesiod...	09:10
	so the line between 'real' and 'virtual' users is kinda thin	09:10
Cathy	it's also possible to run an sshd from within a virtual machine	09:11
	we could call those virtual users	09:11
	← ericy left	09:11
dalias	:)	09:11
	moritz calls virtual users "bots"	09:13
	adhawkins-away → adhawkins	09:13
lite_	can i revert a specific file to whatever version it had before I branched out ?	09:15
Cathy	try git checkout	09:15
	← teran left	09:16
Cathy	well, it might be more useful to say try `man 1 git-checkout`	09:16
grawity	or `man git checkout`	09:16
gitinfo	the git-checkout manpage is available at http://jk.gs/git-checkout.html	09:16
Nevik	indeed	09:16
	→ teran joined	09:16
Nevik	lite_: git checkout <desired revision> -- path/to/file	09:17
lite_	i just did that but my working dir still appears to be clean	09:17
	wait. 1 moment	09:18
	← h0bbit_ left	09:18
Nevik	lite_: then the file probably hadnt changed	09:18
lite_	yeah sorry. i had the wrong revision.	09:19
	→ blinkdesign joined	09:20
	→ h0bbit_ joined	09:20
	← lindenle left	09:20
	← joaoh82 left	09:21
	→ FerreusR joined	09:21
	← teran left	09:21
	→ joaoh82 joined	09:21
Cathy	grawity: did you know that `man git checkout` apparently doesn't work on OS X?	09:23
gitinfo	grawity: the git-checkout manpage is available at http://jk.gs/git-checkout.html	09:23
LionKing	Cathy: try git help checkout	09:23
Cathy	it displays error message "No manual entry for checkout" and then shows the man page for git(1)	09:23
	LionKing: i know how to get to the man page	09:23
	i just thought grawity might find that fact interesting	09:23
	← fairuz left	09:23
LionKing	Cathy: OK	09:24
	→ pfFredd joined	09:24
	← closures999 left	09:24
	→ fairuz joined	09:25
	← fairuz left	09:25
	← shadok left	09:25
Nevik	well, macos is kinda messed up	09:25
	Cathy: it's probably just a question of how git is installed though	09:26
LionKing	Cathy: Stupid question: Shouln´t it be man git-checkout	09:26
gitinfo	Cathy: the git-checkout manpage is available at http://jk.gs/git-checkout.html	09:26
Cathy	git is installed out of the box on OS X	09:26
	→ freestyl3r joined	09:26
LionKing	with the dash between git and checkout?	09:26
Nevik	Cathy: it's not	09:26
	it comes with developer tools, if anything	09:26
	and that is usually an old-ass version	09:26
Cathy	well Nevik it's installed on my MacBook Pro I got three days ago	09:26
	and i took no steps to install it	09:26
Nevik	it probably had dev tools preinstalled	09:26
Cathy	looks like git version 1.8.3.4 (Apple Git-47)	09:27
Nevik	who knows what apple does with these things	09:27
	ha, that's even halfway decent	09:27
	LionKing: the dash should make no difference if git is installed properly	09:27
LionKing	Nevik: I think it is quite a difference	09:27
Nevik	but some versions of man might have a limitation with the two-name lookup	09:27
Cathy	i think it's just a different behaviour of `man` on OS X	09:27
	→ knighthawk joined	09:28
	→ ericy joined	09:28
LionKing	Without the dash you give the man two arguments: git and checkout	09:28
Nevik	yeah. i dont know if the alias lookup happens in the shell or in man	09:28
	LionKing: well, it works on all linuxes :P	09:28
	← gasbakid left	09:28
Nevik	LionKing: basically, linux man tries to look up the full list of arguments as a single command first, then falls back to multiple commands if it doesnt find anything	09:29
	behaviorally anyway. as i said, i dont know where that resolution happens	09:29
LionKing	Nevik: oh, thank you. I didn´t know that...	09:29
Nevik	i know that git subtree isnt in the default man list for git	09:29
Cathy	let's apply science to find out	09:29
	← bkhan left	09:29
Nevik	oh, it is now	09:30
	that's recent	09:30
	← Xethron left	09:30
Cathy	here's what i've discovered	09:30
	[cathyjf@cathy-vm ~]$ strace man git checkout	09:31
gitinfo	the git-checkout manpage is available at http://jk.gs/git-checkout.html	09:31
Cathy	execve("/usr/bin/man", ["man", "git", "checkout"], [/* 46 vars */]) = 0	09:31
	→ bartzy joined	09:31
	→ gasbakid joined	09:31
Cathy	looks like the `man` command handles the resolution	09:31
	→ harish joined	09:31
	→ ruby\|angry_hamst joined	09:31
Nevik	cool	09:31
	so macos man is dumber	09:31
	← grahamsavage left	09:33
	← eletuchy left	09:33
	→ lite__ joined	09:33
Nevik	man's manpage probably describes how it tries to resolve the arguments. but im too lazy to sift trough it :D	09:33
	→ eletuchy joined	09:34
Cathy	it's only 721 lines	09:34
Nevik	but my laziness is over 9000!	09:35
Cathy	i just read it and found nothing relevant	09:35
	heh	09:36
	← ericy left	09:36
Cathy	man --version returns 1.6c on OS X	09:36
	← harish left	09:36
Cathy	compared to 2.6.5 on my Fedora installation	09:36
	→ ericy joined	09:36
	← lite_ left	09:36
	→ lite_ joined	09:37
	← cads left	09:37
	→ MetalMatze joined	09:37
Nevik	that might do it :D	09:38
	→ micw joined	09:38
	← karavanjo left	09:38
	→ karavanjo_ joined	09:38
	← lite__ left	09:40
	→ zhost joined	09:41
	→ achadwick joined	09:42
	→ Sendoushi joined	09:44
	→ grahamsavage joined	09:45
	← svector left	09:45
	← boombatower left	09:46
	→ mihamina joined	09:47
	← m0viefreak left	09:47
	← sabgenton left	09:51
	→ fRAiLtY- joined	09:52
	→ Guest75350 joined	09:52
	→ GT4066 joined	09:52
	→ leakybuckit joined	09:53
	← user258467 left	09:54
	← hyperair left	09:54
	← ericy left	09:54
	→ raonyguimaraes joined	09:55
	← purpleidea left	09:55
	→ harish joined	09:55
	→ jkremser joined	09:56
	← jkremser left	09:56
	→ jkremser joined	09:56
	← ewnd9 left	09:56
	→ teran joined	09:56
	→ blahdodo joined	09:56
	→ purpleidea joined	09:57
	← rfree left	09:57
	← gua_ould left	09:57
	→ user258467 joined	09:57
	→ rfree joined	09:57
	← l0cust left	09:58
	→ Norm_ joined	09:58
	← weiyang left	09:59
	→ svector joined	09:59
	→ mlechner joined	09:59
	→ gua_ould joined	10:00
	→ alezost joined	10:01
	← pfFredd left	10:01
	← Skrypter left	10:02
	→ xobb joined	10:03
	← tvw left	10:04
	→ oo_miguel joined	10:04
alezost	http://www.git-scm.com gives me "Sorry, no Host found". Does anyone know what the problem is?	10:04
bremner	alezost: probably local to your network	10:06
	→ lacrosse_ joined	10:06
	→ eijk_ joined	10:06
	← soee left	10:07
	← JustinMac left	10:07
	→ soee joined	10:08
	→ xpe joined	10:08
	→ sabgenton joined	10:08
	→ rendar joined	10:10
	← eijk left	10:10
	← jon_r left	10:11
	← psoo left	10:11
	→ psoo joined	10:11
	← psoo left	10:11
	→ psoo joined	10:12
	→ ericy joined	10:12
	→ eijk joined	10:13
alezost	bremner: no idea what i can do with that, but thanks for the reply	10:13
	→ closures999 joined	10:14
bremner	well, I was just being http://downforeveryoneorjustme.com/git-scm.com	10:15
	→ beautifulmind joined	10:15
	→ angelsl joined	10:16
	← eijk_ left	10:17
	← harish left	10:18
	→ JustinMac joined	10:18
Nevik	no, it's a DNS problem	10:18
	ive had it since yesterday	10:18
	some NSs return an IP that results in the above message	10:19
	alezost: it's something to do with git-scm's hosting, so unless someone knows who does technical adminning on it, we wont be able to ring any bells :D	10:19
	← Sendoushi left	10:21
	← JustinMac left	10:21
	← mk_qi left	10:23
	→ eletuchy_ joined	10:23
	← Guest75350 left	10:24
	← eletuchy left	10:24
Nevik	alezost: since it doesnt seem to be an isolated problem, i have opened an issue: https://github.com/github/gitscm-next/issues/326	10:24
	though i dont know if any techies/webmasters will notice that	10:24
	→ singham joined	10:25
singham	lo all	10:25
	what does "git --al" do?	10:25
	i forgot the last 'L'	10:25
	← yacks left	10:26
Nevik	singham: nothing	10:26
	the command "git" without any subcommand doesnt do anything	10:26
singham	Nevik: well it's added the files to staging using the 'a' ... i'm not sure about the l	10:26
Nevik	if anything, it should show you a usage message	10:26
	singham: you possibly mean "git add --all" or something	10:26
singham	Nevik: yes exactly...but I typed the above and it added files to staging. I was wondering if the 'L' paremter did naynthing	10:27
	→ PeterBooker joined	10:27
Nevik	$ git --al	10:27
	Unknown option: --al	10:27
	^	10:27
	--all does the same	10:27
singham	whoops	10:27
	sorry i meant	10:27
	← grahamsavage left	10:27
Nevik	so please clarify what command exactly you entered	10:27
singham	$ git add --al	10:27
	→ Sendoushi joined	10:28
Nevik	my guess is that git augments that argument to --all	10:28
	since it's probably unambiguous	10:28
singham	right, thanks for your help	10:28
Nevik	singham: see man git add for details on what --all does	10:28
gitinfo	singham: the git-add manpage is available at http://jk.gs/git-add.html	10:28
	→ grahamsavage joined	10:28
singham	cheers!	10:28
	→ yacks joined	10:29
oo_miguel	Can I somehow see which merges had conflicts and who resolved them?	10:29
Nevik	no	10:29
	← ericy left	10:29
	→ leandrosansilva joined	10:30
Nevik	there would be a trial-and-error method of finding out which merges must have produced conflicts; but there's no telling what happened between the conflict and the resolution	10:30
	it's usually safe to assume that whoever created the merge commit did resolve the conflicts though	10:30
	unless you have extreme pair/group programming in your project	10:31
oo_miguel	I just wonder, because some of my code is gone, and I suppose, that a collegue removed it during a merge	10:31
	just overwriting my version with his one	10:32
Nevik	oo_miguel: the diff on the merge commit will show what changed	10:32
YoungFrog	perhaps "git log -S..." will help you find when it has disappeard, too	10:32
Nevik	true	10:32
	→ kriks joined	10:32
oo_miguel	what does git show exaclty tell me?	10:32
	I have a git show on a commmit right AFTER the merge, where all my lines get removed	10:33
	:P	10:33
	← luc_ left	10:33
oo_miguel	does this mean its his fault ?	10:33
Nevik	that means that commit removed those lines, yes	10:34
	it's usually the point of commit messages to give reasons for why things change	10:34
	if that is not the case, time to go ask your colleague :P	10:34
	→ StarFire_ joined	10:34
oo_miguel	yes, he did it not by purpose, I just wonder where the lines get removed	10:35
	i thought it might be git, by doing some automerging	10:35
Nevik	in the commit whose diff shows them as removed :P	10:35
	← ruby\|angry_hamst left	10:36
	← LionKing left	10:38
alezost	Nevik: thank you very much, now i see	10:39
Nevik	alezost: sure, no problem :)	10:39
	→ ruby\|angry_hamst joined	10:39
YoungFrog	oo_miguel: git won't remove lines in a merge on its own for no reason : if it removes them, it means that they were present before in both branches, and removed in one of the commit.	10:39
oo_miguel	yes, thats makes sense	10:40
	← leandrosansilva left	10:40
oo_miguel	I will examine the log carefullly	10:40
	→ leandrosansilva joined	10:40
	← MetalMatze left	10:42
Nevik	oo_miguel: if `git show` on the commit after the merge shows them as removed, that's where they were removed	10:44
	→ penguin42 joined	10:44
	← h0bbit_ left	10:44
Nevik	not much reason to examine the log further, really	10:44
	→ papegaaij joined	10:46
	→ ericy joined	10:46
	→ bitkiller joined	10:49
	→ raf joined	10:51
	← nbrosnahan left	10:52
	→ arjun_ joined	10:52
	→ codiez joined	10:52
	→ akbar204 joined	10:55
akbar204	Hi	10:55
gitinfo	akbar204: 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.	10:55
	→ eijk_ joined	10:55
akbar204	anybody there?	10:55
	← eijk left	10:55
akbar204	what happends if i am on a different branch and i do a git push origin development??	10:56
	→ grahamsavage_ joined	10:57
	← grahamsavage left	10:59
	← akbar204 left	11:00
oo_miguel	Nevik: ok so I know what happened	11:00
micw	hi	11:00
Nevik	apparently akbar really didnt wanna know the answer to his question	11:01
oo_miguel	my collegue opened his file, pulled , git merged without conflicts, saved his file after doing some minor changes, overwriting my changes, pushed	11:01
	what can I do about it now?	11:01
Nevik	oo_miguel: first of all, you go to your boss and make him make your colleague get an editor which notices file changes on the filesystem	11:01
	alternatively, whack your colleague with a sledgehammer and get a new colleage	11:01
oo_miguel	hehehehe	11:01
	← rouf left	11:02
oo_miguel	I did this already	11:02
	;)	11:02
Nevik	oo_miguel: then you check out the old version of the file or copy those changes from the older version, depending on if new changes have been made	11:02
grawity	so use `git checkout` to get an older version of the file – specifically, the one from the merge	11:02
Nevik	grawity: lemme finish, man :<	11:02
oo_miguel	:)	11:02
	← grahamsavage_ left	11:02
micw	allows "subtree" do edit/commit/push code in sub repositories or are those read only?	11:02
Nevik	irc needs "is currently typing" notifications >.>	11:02
	→ rouf joined	11:03
	← joaoh82 left	11:03
oo_miguel	ok so I fix it manually in one word..	11:03
	→ joaoh82 joined	11:03
Nevik	oo_miguel: if there have been new changes, you need to yes. if there havent been, a checkout and commit fixes it	11:03
	there's nothing more automatic than that	11:03
	→ grahamsavage_ joined	11:05
oo_miguel	Nevik: ok makes sense	11:05
	thx	11:05
	← ericy left	11:10
	→ user125 joined	11:11
	→ adambrenecki joined	11:11
	← jarr0dsz left	11:12
user125	hello. am i supposed to do a git-rm -rf dirname before i do a filter-branch removal?	11:12
	→ timthelion joined	11:12
	← echo-area left	11:12
	→ z8z joined	11:16
	← z8z left	11:16
	→ z8z joined	11:16
	← mlechner left	11:17
	← timthelion left	11:17
	→ mlechner joined	11:18
	← closures999 left	11:19
	→ devanblue joined	11:22
	← mlechner left	11:24
	← dhruvasagar left	11:24
	→ dhruvasagar joined	11:25
	→ ericy joined	11:26
	← donmichelangelo left	11:27
	→ donmichelangelo joined	11:27
	→ LuxuryMode joined	11:28
	→ psyl0n joined	11:29
	← Watabou left	11:29
	← epta left	11:29
	→ asQuirreL joined	11:30
	← purpleidea left	11:31
	→ the_cybersphinx joined	11:31
	← LuxuryMode left	11:32
	→ purpleidea joined	11:32
	← purpleidea left	11:32
	→ purpleidea joined	11:32
	← joaoh82 left	11:33
	← mitz left	11:33
	→ joaoh82 joined	11:33
	→ mitz joined	11:34
	→ LuxuryMode joined	11:34
	→ harish joined	11:35
	← armin_w left	11:35
	→ SuRfDeMoN joined	11:36
	← joaoh82 left	11:38
	→ joaoh82 joined	11:38
	← joaoh82 left	11:38
	← pen left	11:39
	→ pen joined	11:40
	← ericy left	11:41
	→ m0ltar joined	11:42
	→ ewnd9 joined	11:42
	← asQuirreL left	11:42
	← eijk_ left	11:43
	purpleidea → purpleidea-afk	11:43
	purpleidea-afk → purpleidea	11:43
	← arjun_ left	11:43
	→ gwb3 joined	11:46
	→ konadr joined	11:46
	← z8z left	11:48
	← gua_ould left	11:48
	→ fairuz joined	11:48
user125	is there a way to change repacking tmp folder from .git/objects/pack/ ?	11:50
	→ mlechner joined	11:50
	← slassh left	11:51
lb1a	user125, git repack ?	11:51
user125	lb1a: my git gc --prune=now is failing because i dont have enough space for the temp pack files	11:53
lb1a	git cannot help you with your available hdd space Oo	11:54
user125	if i could change .git/object/pack/ to /tmp/	11:54
	← dsdeiz left	11:54
	← setmeaway left	11:55
	→ vals_ joined	11:56
	← Mattias left	11:57
	← tango_ left	11:57
	→ ericy joined	11:58
	← paulweb515 left	11:58
	→ joaoh82 joined	11:58
ayaka	I want to cherry-pick some commites but I got http://paste.debian.net/68220/	12:00
	→ Mattias joined	12:00
	→ joaoh82_ joined	12:01
	← nafg___ left	12:02
	← Guest96996 left	12:03
	← joaoh82 left	12:04
lb1a	ayaka, and your question is	12:07
	← mdiers__ left	12:08
ayaka	lb1a, how to fix it	12:09
_ikke_	ayaka: Your commit doesn't include new changes	12:09
	ayaka: git is telling you that	12:09
	← SicoWork left	12:09
lb1a	ayaka, if your read your paste, you will have every information to continure what you were doing	12:09
	→ SicoWork joined	12:09
	← codiez left	12:10
	→ jarr0dsz joined	12:10
	→ codiez joined	12:10
	← jarr0dsz left	12:11
	← ericy left	12:11
	→ jarr0dsz joined	12:11
	→ Plouj joined	12:12
ayaka	_ikke_, lb1a but it is a new branch, and the commit I refering in cherry-pick is not in the log of this branch(it is in the other branch)	12:12
	← xobb left	12:12
	← codiez left	12:12
	→ codiez joined	12:12
	→ VictorCL joined	12:13
	→ armin_w joined	12:13
	→ vtunka joined	12:13
	← knighthawk left	12:14
user125	uhh... git is so complicated.	12:15
	→ hs366 joined	12:15
	→ closures999 joined	12:16
	← kcj left	12:16
	→ gua_ould joined	12:16
user125	i removed a huge blob of data from the repo using filter-branch, yet the .git folder is still too large in comparison with the working tree	12:16
	→ timthelion joined	12:17
Celelaptop	What is the best way to contribute to an open source software using git? Given that most of the software requires patches to be sent via a mailing list and to be applyable on origin/master directly.	12:17
lb1a	Celelaptop, !best	12:17
gitinfo	Celelaptop: [!bestway] You want to know the best way to do X? If you can give us a proper definition of "best", we'll give you a proper way to do X ;)	12:17
	← teran left	12:17
Celelaptop	Should I patch my local master? Create a branch for all my modifications? Create one branch per patch serie? Should these branches be based on origin/master or on master? Should I update these branches with a pull a or rebase?	12:17
	→ teran joined	12:18
Celelaptop	lb1a: so let's rephrase the question as: how most people do?	12:18
	← oo_miguel left	12:18
lb1a	Celelaptop, i'd say it always depends on the project	12:19
	← mihamina left	12:20
	← dhruvasagar left	12:20
lb1a	Most projects document a proper way to contribute	12:20
	→ abelsiqueira joined	12:21
Celelaptop	What I've seen for syslinux and qemu, they only say "submit patch on the mailing list, and follow the guidelines of linux".	12:21
	→ kcj joined	12:21
lb1a	hm hmm git-scm.com seems to be down "Sorry, no Host found"	12:21
	Celelaptop, so follow the contribution guidelines of the linux kernel	12:21
Celelaptop	It doesn't talk much about git.	12:22
	→ johnkevinmbasco joined	12:22
Celelaptop	And I don't think they care about the way contributors use git on their own computers.	12:22
	← timthelion left	12:22
	→ mdiers__ joined	12:22
	→ xobb joined	12:23
	→ rainbyte joined	12:23
lb1a	Celelaptop, you asked how to contribute to a project, not how to setup your workflow at home	12:23
Celelaptop	Yes but some of the workflows do not match very well the projects requirement that all patches should apply on origin/master, for instance	12:24
lb1a	why not?	12:25
	← graphitemaster left	12:25
	→ graphitemaster joined	12:25
	← joaoh82_ left	12:25
Celelaptop	Doing several kind of midifications on the same branch (local master or all in one branch) is not very good because the patch you will produce may not apply on origin/master.	12:25
lb1a	why do you care about contributing when it comes to your workflow while development. you can always craft patches after all or rebase history according to the guidelines...	12:25
	→ Luke-Jr joined	12:26
	← closures999 left	12:26
	→ joaoh82 joined	12:26
Celelaptop	Well, rewriting history is not always a fun thing to do.	12:27
	→ ericy joined	12:28
Celelaptop	Is there a difference between basing a branch on origin/master or on master?	12:28
	← TheFuzzball left	12:28
	← mdiers__ left	12:29
canton7	Celelaptop, depends if origin/master points to the same commit as master ;)	12:30
	← hs366 left	12:32
	→ arjun_ joined	12:32
	→ TheFuzzball joined	12:32
	→ hs366 joined	12:33
	→ ciampix joined	12:33
	→ echo-area joined	12:33
	← intellix left	12:33
	← hs366 left	12:33
	→ hs366 joined	12:34
	← user125 left	12:35
	← thierryp left	12:35
	← lun___ left	12:36
YoungFrog	So the question is : what is the difference ? origin/master is a branch that you shouldn't ever checkout, it simply follows master on origin. your own master is whatever you do with it, but usually it's set up so that it is somehow related to origin/master (either an exact copy, or you merge regularly from origin/master to master)	12:37
	by "follows", I mean that git fetch will make it point to the same commits as "master" on "origin".	12:38
	→ carbon2 joined	12:38
canton7	another point to make it that origin/master is updated on every 'git fetch', but your local master is normally only updated by you when you check it out, presumably with the intention of working on it. therefore origin/master is always going to better reflect the state of the origin repo	12:38
	← raf left	12:39
	← hs366 left	12:40
	→ hs366 joined	12:40
	← grahamsavage_ left	12:40
Celelaptop	ok thx	12:40
	→ mdiers__ joined	12:41
	→ Gavrisimo joined	12:41
	← devanblue left	12:44
	← joaoh82 left	12:46
	→ joaoh82 joined	12:47
btiernay	Is it possible to push a tag to a repo for which you have not first cloned?	12:47
	← ericy left	12:47
_ikke_	btiernay: yes	12:47
	→ asQuirreL joined	12:47
btiernay	_ikke_, Interesting, please do expound sir!	12:48
_ikke_	btiernay: Add it as a remote, and then git push <remote> <tag>	12:48
	It will push every commit reachable by that tag though	12:48
	→ Xethron joined	12:48
btiernay	_ikke_, Is it possible even if you don't have the repo locally, but have git installed?	12:48
Celelaptop	And if I have a branch with unmegered commits while origin/master is updated. Should I merge or rebase it?	12:49
btiernay	I'm trying to tag a repo from CI	12:49
_ikke_	btiernay: No	12:49
btiernay	But not the main repo	12:49
	Aw shucks	12:49
_ikke_	btiernay: You have to have the history for that tag you want to push	12:49
btiernay	I can't just say "Hey remote, tag HEAD with this tag"?	12:50
_ikke_	nope	12:51
	The git protocol doesn't allow that	12:51
	← maleknet left	12:51
btiernay	Okay, fair enough!	12:51
	Thanks :)	12:52
_ikke_	It only allows you to say: "Hey remote, I have here some refs with this according history"	12:52
StarFire_	How can I completely remove a subdirectory from a converted SVN repo, if the instructions in ¨Prog Git¨, chapter 9.7, ¨Removing objects¨ does not work	12:52
	?	12:52
btiernay	One last thing, is it possible to force delete a remote tag? By this I mean, if it doesn't exist, don't complain	12:52
	→ ni_ joined	12:52
	← asQuirreL left	12:52
	← mdiers left	12:52
_ikke_	StarFire_: You need to use filter-branch fro that	12:52
canton7	StarFire_, define "does not work"	12:52
btiernay	_ikke_, Thanks for the explanation	12:53
StarFire_	_ikke_: It is described with filter-branch in said chapter	12:53
	canton7: count-objects shows be before and after filter-branch the same ¨size-pack¨	12:53
_ikke_	StarFire_: Then explain what command you executed, and what output you get	12:53
	→ mdiers joined	12:54
	← mdiers__ left	12:54
StarFire_	I did: git rm -r _dev/defaults/doc	12:55
	git commit -m ¨Removed huge documentation¨	12:55
canton7	cna we have a pastebin, please, with a terminal log of you executing the command and the resulting output?	12:55
btiernay	Basically I want do is delete a tag on the remote if it exists, otherwise do nothing	12:55
canton7	"insignificant" things tend to get lost otherwise	12:55
	← SicoWork left	12:55
	→ tma996 joined	12:56
StarFire_	git filter-branch --index-filter ´git rm -r --cached --ignore-unmatch _dev/defaults/doc´ HEAD	12:56
	→ teran_ joined	12:56
StarFire_	rm -Rf .git/refs/original	12:56
	rm -Rf .git/logs	12:56
	git gc	12:56
canton7	StarFire_, can we have a pastebin, please, with a terminal log of you executing the command and the resulting output?	12:56
StarFire_	canton7: It is quite a repo (21300 committs), I don´t have all the output...	12:57
canton7	iirc a filter-branch output isn't more than a few lines? my memory is that it continuously rewrites the same line to show progress. I could be wrong though	12:58
	anyway, output of 'git branch'?	12:58
StarFire_	canton7: Yes, it has output if he has something to do	12:58
	→ dsdeiz joined	12:58
canton7	fair enough	12:58
StarFire_	canton7: In my case it startes about commit 13000 (til the end)	12:59
	→ kbart joined	12:59
canton7	all looking very promising	12:59
	output of 'git branch'?	12:59
	← teran left	12:59
StarFire_	* master	12:59
canton7	good, was worried you were on a detached HEAD	13:00
	look at some point in history - does that dir still exist?	13:00
	→ _toordog joined	13:00
StarFire_	I am quite new to git. How do I look that up in some point in the history?	13:01
_ikke_	git log shows you the history	13:01
RichiH	Dougie2187: aye... and i am incredibly annoyed by the fact that	13:01
_ikke_	gitk would also be easy for that	13:01
canton7	I'm talking about looking at a commit in history	13:01
	not a diff	13:01
RichiH	a) the history is in the wrong order	13:01
canton7	StarFire_, easy way: git checkout -b tempbranch HEAD^^^^^	13:02
RichiH	b) the merge commit ist not marked as such	13:02
canton7	(you could detach your HEAD, but I'm not going to suggest that to someone who's new)	13:02
RichiH	c) 12 hours is not "dragging on"	13:02
	← pen left	13:02
	→ pen joined	13:03
StarFire_	canton7: It does not exist in HEAD^^^^^^	13:03
	→ ericy joined	13:04
canton7	StarFire_, then it looks like your method of determining whether filter-branch worked is flawed, rather than the filter-branch itself?	13:04
_ikke_	git log --all -- path/to/dir should show you if there are commits left that have that dir	13:04
canton7	^^ also worth doing	13:04
	← TheFuzzball left	13:05
	→ mdiers__ joined	13:06
	→ davividal joined	13:06
	→ thierryp joined	13:06
StarFire_	_ikke_: There are quite a lot of commits...	13:06
RichiH	!lolo	13:06
	!lol	13:06
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	13:06
Kartagis	if I checkout a branch (say 8.x-2.x) then git push origin HEAD, will it be pushed to that branch?	13:07
StarFire_	_ikke_: Actually it looks like the committs I wanted to remove	13:07
	→ k4rtik joined	13:07
_ikke_	Kartagis: no	13:08
	← daito left	13:08
_ikke_	Kartagis: You have to specify the name of the branch	13:08
	Kartagis: git push origin HEAD:8.x-2.x	13:08
	→ slassh joined	13:08
	→ TheFuzzball joined	13:08
_ikke_	Kartagis: (it also depends a lot on the setting of push.default, and if an upstream is defined for that branch)	13:08
	← TheFuzzball left	13:09
	vals_ → tango_	13:09
	→ TheFuzzball joined	13:09
canton7	StarFire_, do you have any branches other than master?	13:09
	← TheFuzzball left	13:10
	→ TheFuzzball joined	13:10
	← TheFuzzball left	13:10
StarFire_	canton7: Yes, 34 ¨remotes¨ (they are not real remotes, they are from converting) and 13 ¨remote tags¨ (also from conversion)	13:11
	→ TheFuzzball joined	13:11
canton7	StarFire_, so the log --all will be showing commits belonging to those remotes	13:11
	if you don't have any local branches other than master, 'git log -- path/to/dir'	13:11
	← mdiers__ left	13:11
	← davividal left	13:12
StarFire_	canton7: git log -- path/to/dir returned nothing...	13:13
	→ bgy joined	13:13
bgy	hi	13:13
canton7	StarFire_, good, then your local master branch has definitely been rewritten correctly	13:13
StarFire_	hmm	13:13
bgy	Is it possible to do something like /foo/{sub,other,directory} in .gitignore?	13:14
StarFire_	Should I checkout every branch and do the same procedure as in master?	13:14
canton7	StarFire_, I thought you only had one local branch/	13:14
	bgy, from man gitignore, "Otherwise, Git treats the pattern as a shell glob suitable for consumption by fnmatch(3) with the FNM_PATHNAME flag: wildcards in the pattern will not match a / in the pathname. For example, "Documentation/*.html" matches "Documentation/git.html" but not "Documentation/ppc/ppc.html" or "tools/perf/Documentation/perf.html""	13:15
gitinfo	bgy: the gitignore manpage is available at http://jk.gs/gitignore.html	13:15
	→ intellix joined	13:15
StarFire_	canton7: I don´t know actually. The other branches start all with remote/...	13:15
	canton7: but I converted that repo from SVN	13:15
canton7	StarFire_, so they're local read-only copies of branches which exist in those remote repos?	13:15
	→ asQuirreL joined	13:15
StarFire_	canton7: So they are kind of local too...	13:15
canton7	StarFire_, what are the exact remote branch names?	13:15
StarFire_	git branch -a shows me:	13:16
	* master	13:16
	remotes/BR_R1	13:16
	remotes/R1	13:16
canton7	ok, git-svn converted repo. all good	13:16
	→ hyperair joined	13:16
StarFire_	remotes/R2	13:16
	etc.	13:16
canton7	so there are a ton of remote branches, but you only created one local branch, master	13:16
	so create a new local branch for each of those remote branches	13:16
	then do the filter-branch again, but use '-- --all' instead of 'HEAD'	13:16
	← j0ran left	13:16
bgy	canton7, yeah I read it, but I doesn't seem to handle such pattern, that's why I asked here	13:16
StarFire_	canton7: OK, I will do this. It´ll take a while	13:17
	canton7: Thank you!	13:17
canton7	StarFire_, wait a sec	13:17
StarFire_	yes?	13:17
	→ koda joined	13:18
canton7	you can script that branch-creation process	13:18
StarFire_	Like described in Pro Git?	13:18
canton7	not sure. been a while since I read that book	13:18
Celelaptop	When I create a branch, with git branch newbranch startingbranch, will that make subsequent git pull to merge from startingbranch?	13:19
	→ davividal joined	13:19
StarFire_	I alread have. I also have removed the branches that I no longer need	13:19
	canton7 is confused	13:19
canton7	never mind, go and create all of your local branches manually	13:19
	← subhojit777 left	13:19
	→ WABL58H joined	13:19
canton7	Celelaptop, depends whether startingbranch is a remote-tracking branch iirc	13:20
StarFire_	canton7: I did that on another copy of that repo	13:20
	← thierryp left	13:20
canton7	ok cool	13:20
StarFire_	canton7: sorry to confuse you...	13:20
Celelaptop	what is a remote-tracking branch?	13:20
	→ thierryp joined	13:20
canton7	Celelaptop, one that starts with e.g. 'origin/'. They're local read-only records of the states of remote branches	13:20
Celelaptop	ok	13:21
	So it cannot be a local branch?	13:21
canton7	(to clarify, if the starting point is a remote-tracking branch, then tracking info is set up, and 'git pull' will just work (tm))	13:21
	→ Kayra joined	13:21
canton7	(if the starting point is a local branch, everything's the same exept that tracking info isn't set up)	13:22
	→ SicoWork joined	13:22
	→ buzzybron joined	13:22
	→ mdiers__ joined	13:23
	← buzzybron left	13:24
Celelaptop	Hum... But git pull will fetch and merge from the branch set with git branch --track, doesn't it?	13:25
	So pull wouldn't work if the tracking origin isn't set.	13:25
	→ paulweb515 joined	13:26
	→ guampa joined	13:28
	→ lun__ joined	13:28
	← mdiers__ left	13:29
	← leandrosansilva left	13:29
	→ leandrosansilva_ joined	13:30
	→ grahamsavage joined	13:30
	← grahamsavage left	13:30
	→ grahamsavage joined	13:30
	← paulweb515 left	13:32
	← rjhunter left	13:32
	→ snag joined	13:33
	← VictorCL left	13:34
	→ VictorCL joined	13:34
micw	recently i have seen a nice flow diagram how to do things with git (in the style "if you hate someone - rebase interactive on origin" ;-) ) but i cannot find it anymore. does anyone know this?	13:34
	← xobb left	13:35
micw	ah, remembered the name. it's "Git pretty". found ;)	13:36
	→ infynyxx joined	13:36
	← grahamsavage left	13:36
	→ paulweb515 joined	13:37
	→ grahamsavage joined	13:37
	→ wlemuel joined	13:37
	← edude03 left	13:37
Nevik	i love how the hate is so nonchalant in that	13:37
	→ edude03 joined	13:37
Nevik	do you hate them? rebase away!	13:37
	← TheFuzzball left	13:39
micw	;)	13:39
	← ruby\|angry_hamst left	13:40
	→ ruby\|angry_hamst joined	13:40
	← wlemuel left	13:40
	→ mdiers__ joined	13:41
	→ TheFuzzball joined	13:42
	→ xobb joined	13:42
canton7	Celelaptop, yup	13:44
	but the logic is, if you created the branch from another local branch, there isn't yet a remote branch for it to track	13:44
	if you created the branch from a remote branch, there is a remote branch to track, so tracking is set up	13:44
	← ericy left	13:45
Celelaptop	Ok.	13:46
	→ wlemuel joined	13:46
Celelaptop	So the usage is: if I track a remote branch I pull, if I derive a local branch into another local branch, i just merge or rebase.	13:46
	← infynyxx left	13:47
	← TheFuzzball left	13:47
canton7	Celelaptop, you can set up tracking when you first push the branch	13:48
	with git push -u	13:49
	or later, with git branch --set-upstream-to	13:49
	← grahamsavage left	13:50
	→ grahamsavage joined	13:51
	← grahamsavage left	13:51
	← Dougie2187 left	13:52
	→ grahamsavage joined	13:52
	→ Dougie2187 joined	13:52
	→ dhruvasagar joined	13:54
	← rchavik left	13:56
	→ rethus joined	13:57
	← RogerWS left	13:57
	→ liamkeil1 joined	13:58
	← hs366 left	13:59
	→ stany_ joined	14:00
	→ __Hassen__ joined	14:00
	→ St_Marx joined	14:00
__Hassen__	why isn't "man grep" working in git bash? isn't it supposed to be working like the linux bash?	14:01
	i use win7	14:01
	→ aartist joined	14:02
	← mdiers__ left	14:02
	→ infynyxx joined	14:02
	← codiez left	14:02
	→ ericy joined	14:03
canton7	__Hassen__, with some caveats. one is a lack of man pages	14:03
	← renatow_ left	14:03
	← kriks left	14:03
	← lun__ left	14:05
__Hassen__	canton7, ah so doesn't it contain the full list of man pages like linux OS's do?	14:05
	→ telmich joined	14:05
telmich	good day	14:05
	← stany_ left	14:05
canton7	__Hassen__, correct. google is your friend here - plenty of man pages online	14:05
	← dsdeiz left	14:05
telmich	how can I create a new branch that has no ancestors? i.e. like a fresh branch in an existing project?	14:05
	→ renatow joined	14:05
	→ codiez joined	14:05
canton7	telmich, man git checkout --orphan	14:05
gitinfo	telmich: the git-checkout manpage is available at http://jk.gs/git-checkout.html	14:05
	→ stany joined	14:05
__Hassen__	canton7, understood.	14:05
	← sonOfRa left	14:06
	→ sonOfRa joined	14:06
	← stany left	14:06
	__Hassen__ → Hassen	14:06
	→ lun__ joined	14:07
	← saorge left	14:08
	← flaguy48 left	14:08
telmich	canton7: thanks	14:08
	← snag left	14:08
	← Raging_Hog left	14:09
	→ sohail joined	14:09
	← rethus left	14:09
	→ saorge joined	14:09
	← arjun_ left	14:10
	← gasbakid left	14:13
	→ mdiers__ joined	14:13
	→ soon joined	14:13
	← fairuz left	14:13
	← johnkevinmbasco left	14:14
	→ Cromulent joined	14:14
	← lun__ left	14:15
	→ sebrock joined	14:16
	→ johnkevinmbasco joined	14:16
	→ fairuz joined	14:17
	← Kayra left	14:17
soon	Anyone have experience using git to synchronize a local folder and a webdav folder? (as an alternative to e.g. Dropobx)	14:17
Kartagis	how do I allow creation of repos on-the-fly?	14:18
	→ johnkevi_ joined	14:18
bremner	Kartagis: context?	14:18
Kartagis	bremner: like a repo is not created on the server beforehand, but it is created when the user pushes	14:19
	← adac left	14:19
	← infynyxx left	14:19
bremner	Kartagis: gitolite can do that with "wildrepos"	14:20
_ikke_	Kartagis: !gitolite has support for 'wild' repo's	14:20
gitinfo	Kartagis: Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/	14:20
	← mdiers__ left	14:20
	→ jonatasnona joined	14:20
Kartagis	cool	14:20
	thanks	14:20
	→ j0ran joined	14:20
	← johnkevinmbasco left	14:21
	→ vurma joined	14:22
	← ericy left	14:24
	→ aberrant joined	14:25
soon	everybody liked Kartagis' question and nobody liked mine .. <sulk> :-)	14:26
Kartagis	aww	14:26
	what was yours?	14:26
_ikke_	soon: Because you mentioned dropbox :P	14:26
	soon: Git is not really meant for synchronization	14:26
	It's a lot of overhead	14:27
	soon: even rsync in a cron is a better sollution imo	14:27
soon	I've considered rsync, but it's not quite up to the job	14:27
bremner	soon: git annex assistant is designed as a dropbox replacement	14:27
soon	ahh ... I can google that	14:28
_ikke_	soon: Why isn't rsync good enough?	14:28
	→ bluesm joined	14:28
bremner	of course there are several others that don't involve git.	14:28
	→ Ingmar joined	14:28
	← Ingmar left	14:28
	→ Ingmar joined	14:28
soon	_ikke_ rsync will not handle deletions very well	14:28
	← Ingmar left	14:28
_ikke_	soon: Can you explain?	14:29
soon	I could do a bi-lateral rsync, but if I decided to delete a file, it would be re-sync'ed	14:29
	← beautifulmind left	14:29
soon	thus I would have to manually delete the file from every copy/client	14:29
	→ Ingmar joined	14:30
	← Ingmar left	14:30
	→ Ingmar joined	14:30
_ikke_	Okay, you want two way synchronization	14:30
soon	yes	14:30
	→ arjun_ joined	14:30
moritz	unison does two-way syncing	14:30
	→ pandeiro joined	14:30
	→ lindenle joined	14:30
soon	I found unison ... looks promising, but apparently not really developed anymore, hence I thought I'd ask if Git was a better solution	14:31
	but as stated ... perhaps overkill	14:31
	I'll look at unison and git annex assis.	14:31
	→ adac joined	14:31
soon	If neither fits my problem-pants, I'll come back :-)	14:31
	→ mdiers__ joined	14:32
	→ osa1 joined	14:32
osa1	git book webpage is down?	14:32
_ikke_	osa1: progit?	14:32
osa1	http://www.git-scm.com/book	14:32
	this one	14:32
	→ lorenx joined	14:32
_ikke_	osa1: aparently, yes	14:33
osa1	is this correct URL?	14:33
	← ewnd9 left	14:33
_ikke_	yes, it is	14:33
	→ JonasF joined	14:33
	← wlemuel left	14:34
_ikke_	Even the website of the maintainer of that website is down aparently	14:34
	← JonasF left	14:34
	← thierryp left	14:34
	← mastro left	14:36
moritz	how meta	14:36
	→ edwardly joined	14:36
_ikke_	I'm so meta, even this acronym	14:36
moritz	I never met a meta I didn't like	14:37
	← simesy left	14:37
	→ simesy joined	14:38
	→ teran joined	14:39
	← jonatasnona left	14:39
	← soon left	14:39
	← mlechner left	14:40
	Mapley → Maplez	14:40
	Maplez → Mapley	14:40
lorenx	can i have a git repository ignoring inner/path/* and a second nested git repository in inner/path?	14:40
	← abelsiqueira left	14:40
_ikke_	lorenx: sure	14:41
	← takamichi left	14:41
	→ aChatir joined	14:41
	→ thierryp joined	14:41
	→ ericy joined	14:41
	→ akitada joined	14:41
	← dccc left	14:41
lorenx	_ikke_: m, ok... i tried but i'm having problems when i switch from a post-nested branch and a pre-nested one	14:42
	← joaoh82 left	14:42
	← k4rtik left	14:42
	→ mgedmin joined	14:42
	→ luxgeek joined	14:42
	← SeySayux left	14:42
	← teran_ left	14:42
	→ joaoh82 joined	14:42
mgedmin	how hard would it be to write a hook to stop pushes if I've got commits with 'WIP' at the start of the commit message?	14:42
	→ takamichi joined	14:42
	← hyperair left	14:43
	← simesy left	14:43
lorenx	let me explain... at some point of my history i decided to have a second nested repository so i moved away all the inner/path/* (to delete it from the main repository), i ignore inner/path/*, i create a second nested repository in inner/path/ and copy back here my previous files	14:43
	← kcj left	14:43
	→ jeek joined	14:43
	→ takamich_ joined	14:44
	← luxgeek left	14:44
	→ SeySayux joined	14:44
	→ nilg joined	14:44
_ikke_	lorenx: yes, that's a problem indeed	14:44
	→ tvw joined	14:44
lorenx	but now, when i checkout a previous branch, the main git says i have to delete those files	14:44
	what am i missing?	14:44
_ikke_	lorenx: You aren't missing anything, it's just a situation that is difficult to handle	14:44
	→ mr_red joined	14:45
lorenx	but it is a normal scenario with nested repos	14:45
	i guess	14:45
_ikke_	lorenx: Normally submodules are used	14:45
lorenx	excuse me?	14:45
	← ntio left	14:45
_ikke_	!submodules	14:45
gitinfo	[!submodule] git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html	14:45
	→ lun__ joined	14:46
_ikke_	lorenx: The problem is that git sees that there are files that that are not being tracked, and git refused to delete / overwrite files that aren't tracked, so it complains	14:46
moritz	you can delete them with git clean	14:46
lorenx	_ikke_: yeah but i thought that telling it to ignore that path would be enough	14:47
	← takamichi left	14:47
	→ ewnd9 joined	14:47
moritz	you were wrong :-)	14:47
	← angelsl left	14:47
_ikke_	lorenx: Would you think it's a good idea to delete / overwrite ignored files without question?	14:47
lorenx	moritz: git clean doesn't sound a great solution	14:47
	→ xyxy joined	14:48
lorenx	_ikke_: cause i thouhgt that if they are ignored git has to ignore them :P	14:48
_ikke_	lorenx: .gitignore only tells git to not show it as untracked in git status, and to warn you when you try to add it to git	14:48
	→ przemkovv joined	14:48
	← el3ctron left	14:48
lorenx	_ikke_: anyway... i'll read about git submodule	14:49
	→ King_Hual joined	14:49
	← edwardly left	14:49
_ikke_	(submodules have explicit support for this, where git places the .git dir in the parent repo, so the entire dir can be removed, but only when there aren't any uncomitted changes in the submodule)	14:49
	→ Kayra joined	14:50
	← soee left	14:51
	→ closures999 joined	14:51
	→ edwardly joined	14:51
	→ jonatasnona joined	14:52
	→ draknir joined	14:53
	→ dccc joined	14:53
	gitinfo set mode: +v	14:54
draknir	hello :) does anyone else have trouble accessing git-scm.com ?	14:54
Kartagis	all of us	14:55
	→ de-Vri-es joined	14:55
	← stodan left	14:55
draknir	ok, good to know	14:55
	← codiez left	14:55
draknir	is there anywhere else I can download git for windows?	14:56
	← Gavrisimo left	14:56
	→ codiez joined	14:56
	→ Gavrisimo joined	14:57
	→ Octalot joined	14:57
	→ Gavrisim_ joined	14:58
	← ericy left	14:58
	← arjun_ left	14:59
	← donmichelangelo left	15:00
	← eletuchy_ left	15:00
	← matanya left	15:00
	→ eletuchy joined	15:00
	← codiez left	15:00
	← de-Vri-es left	15:01
	→ donmichelangelo joined	15:01
	← dhruvasagar left	15:01
	← Gavrisimo left	15:02
	← leandrosansilva_ left	15:02
	← Gavrisim_ left	15:02
	→ phile joined	15:02
lorenx	_ikke_: i removed the second git repository (inner/path/.git and inner/path/.gitignore) and i removed the ignore path in the main git but now i cannot add my inner/path/* files back, is it expected?	15:03
	→ stany joined	15:04
	→ bamj0rb joined	15:04
	gitinfo set mode: +v	15:04
	← eletuchy left	15:05
	→ sebhoss joined	15:05
_ikke_	lorenx: What do you mean with that you cannot add the files back?	15:05
	← sebhoss left	15:05
lorenx	that if i got git status in the main git, it doesn't tell me it find the inner/path/* files	15:05
	everything is clean for it	15:05
	← hholtmann left	15:05
lb1a	hm where to determine what's the latest git version? topic says 1.8.4.4, website says 1.8.4.3	15:06
lorenx	even if i edit a inner/path/* file	15:06
_ikke_	lb1a: does git ls-tree HEAD -- inner/path return anything?	15:06
	→ hholtmann joined	15:06
lorenx	it seems it is still ignorring that path	15:06
	← _toordog left	15:06
moritz	lb1a: look at the newest tag in the git git repo?	15:06
_ikke_	lb1a: git mailing list	15:06
	lorenx: and git-scm is often behind	15:07
lb1a	hmkay	15:07
	← ewnd9 left	15:07
lorenx	it returns a line	15:07
	← zhost left	15:07
lb1a	git-scm seems a bit unstable today...	15:07
	anyway thanks	15:07
_ikke_	lb1a: it is	15:07
	1.8.4.4 has been released	15:07
	→ PaulCapestany joined	15:07
	← takamich_ left	15:07
lorenx	_ikke_: it returns a commit line but if i do git show it says "fatal: bad object"	15:08
	→ simesy joined	15:08
	← kpreid left	15:09
_ikke_	Can you paste that line here?	15:09
	← stany left	15:09
lorenx	_ikke_: i see know you were talking with lbla	15:09
	→ AaronMT joined	15:09
lorenx	anyway: 160000 commit 51a8c8e82dad63b14a205a7f9d33dc5b498c0a61 inner/path	15:10
	← AaronMT left	15:10
_ikke_	lorenx: That means git already sees it as a submodule	15:10
lorenx	then "git show 51a8c8e82dad63b14a205a7f9d33dc5b498c0a61" returns "fatal: bad object 51a8c8e82dad63b14a205a7f9d33dc5b498c0a61"	15:10
	→ AaronMT joined	15:10
_ikke_	git rm --cached inner/path and comitting that should solve it	15:10
lorenx	i haven't created any submodule yet	15:10
	→ takamichi joined	15:11
lorenx	_ikke_: git rm? but "git status" is empty	15:11
	← jonatasnona left	15:11
	← kbart left	15:11
_ikke_	lorenx: yes, because git doesn't look at the files in a submodule	15:11
	→ kukks joined	15:11
	← kukks left	15:11
	→ kukks joined	15:11
	← h_kon left	15:11
_ikke_	"160000 commit" is a submodule	15:11
	→ jonatasnona joined	15:11
_ikke_	51a8c8e82dad63b14a205a7f9d33dc5b498c0a61 is the commit that was checked out in the sub repository	15:12
	→ pfFredd joined	15:12
	→ h_kon joined	15:12
	→ TheBrainWork joined	15:12
lorenx	_ikke_: it seems it worked	15:12
	→ arjun_ joined	15:13
lorenx	now git tells me "deleted: inner/path" and "untracked: inner/path/" at the same time	15:13
	sounds weird :P	15:13
	← liamkeil1 left	15:14
	← cagedwisdom left	15:14
	→ ohmy joined	15:14
ohmy	Hello	15:14
lorenx	_ikke_: i haven't undertood clearly what happened but it worked, thanks a lot	15:15
	← tvw left	15:15
	← Kayra left	15:15
ohmy	i have a remote got repository, cointaining my first commit. I would like to reset it as if it was a clean and empty repo, how can i make this please ?	15:15
	→ ericy joined	15:15
_ikke_	try git push --delete origin master	15:16
	Not sure if it works though. Guess it will complain	15:16
	← h_kon left	15:17
	← simesy left	15:17
ohmy	_ikke_: errir unknown option delete :(	15:17
	→ kpreid joined	15:17
_ikke_	ohmy: Older git version?	15:17
	ohmy: try git push origin :master	15:17
moritz	or just delete the remote repo, and create a new one	15:17
	I guess a normal branch deletion can't delete the default branch	15:18
	→ eijk_ joined	15:18
ohmy	moritz: thanks	15:18
_ikke_	Poor repos, people deleting them for no reason :P	15:18
ohmy	_ikke_: thanks	15:18
	→ ttl117 joined	15:18
	→ jeffszusz joined	15:18
	→ _toordog joined	15:19
	→ Venix joined	15:19
Venix	just thought you'd like to see my response to www.godhatesfags.com http://pastebin.com/F9siPCbR	15:19
	ChanServ set mode: +o	15:20
	bremner set mode: +q	15:20
ttl117	Anyone using gitflow? I did a new repo(git flow init), switched to master, did 2 commits of production(master) versions. What's the correct way now to continue working on develop branch? It's empty and there's not git flow command for this situation	15:20
	ChanServ set mode: -o	15:21
moritz	git checkout develop; git merge master # I suppose	15:21
	→ Sadeukko joined	15:22
	← Venix left	15:22
_ikke_	Amount of spam is increasing lately	15:23
	← Xethron left	15:23
	← Norm_ left	15:24
	← papegaaij left	15:25
lorenx	_ikke_: do i need a repo url to init a submodule?	15:26
	← lite_ left	15:26
	→ de-Vri-es joined	15:26
	→ Gavrisimo joined	15:27
	← psoo left	15:27
lorenx	cause "git submodule init inner/path" does nothing	15:28
	← draknir left	15:28
	Plouj → msg	15:29
	msg → Plouj	15:29
	← jarr0dsz left	15:29
	→ Gavrisim_ joined	15:29
	→ cads joined	15:29
lorenx	it seems i can "submodule add" a remote repo only	15:31
	→ eletuchy joined	15:31
	← aberrant left	15:31
	← Gavrisimo left	15:32
	← ericy left	15:32
	← Gavrisim_ left	15:33
_ikke_	lorenx: Yes, a submodule is basically a reference to an external repository	15:33
lorenx	_ikke_: so it's not what i need	15:33
_ikke_	Ok, what do you need?	15:33
lorenx	i just need 2 nested locel reposotory	15:34
	*local	15:34
	→ Dougie187 joined	15:35
lorenx	but ignoring the subpath and creating a normal nested repo doesn't work perfectly, as i already told you	15:35
	→ Rhonda joined	15:35
lorenx	you told me "it's just a situation that is difficult to handle"	15:36
	what about subtree?	15:36
	← magnus^^p left	15:36
lorenx	or i should just move my second repo out of the main one and symlink them	15:36
_ikke_	!subprojects	15:37
gitinfo	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.	15:37
	← VictorCL left	15:37
	→ psavage joined	15:38
psavage	hey all a bit of a workflow question	15:38
	say I have a two features I'm developing in separate branches	15:39
	now all of a sudden one depends on the other	15:39
	but they still both need continued development	15:39
	→ shinnya joined	15:39
	→ rubatdub joined	15:40
psavage	how is that strategy best handled?	15:40
bamj0rb	Depends. You could continuously rebase one branch onto the other one. Or continuously merge one branch into the other.	15:40
psavage	right hte rebasing I get	15:41
	If I keep merging one branch into another	15:41
bamj0rb	That's probably what I would do, but you do have to be a bit more careful to not screw up conflict resolution.	15:41
	(The rebasing..)	15:41
psavage	yeh	15:41
	← ohmy left	15:41
	← xyxy left	15:42
	← LuxuryMode left	15:42
	→ VictorCL joined	15:42
psavage	bamj0rb, what if I have 3-4 branches that all need combining and developing on?	15:42
	nested rebasing?	15:42
bamj0rb	First of all I'd say your features shouldn't all be intertwined like that. It probably means you're doing it wrong. :)	15:42
	→ petey joined	15:43
psavage	bamj0rb, yeh - I was thinking the same - not that I have that many branches ;)	15:43
	→ omarek joined	15:43
bamj0rb	It might be useful to rebase in order to take advantage of refactoring, but it might complicate source control and make a mess.	15:43
omarek	Hi, how do I remove the last commit ? I realized I made a mistake. The entire commit is pointless, a proper solution will take more time and there's nothing to really --amend...	15:44
bamj0rb	It might be better to just develop the other features against the old code, and once it's considered "good" either rebase or merge then.	15:44
	If it doesn't actually /require/ the other branch then you can get away developing with the dependency.	15:44
_ikke_	omarek: git reset HEAD^	15:44
	That will undo the commit, but leave the changes there	15:44
psavage	bamj0rb, yeh that was my thinking	15:44
	ok	15:44
	ty	15:44
	← micw left	15:45
omarek	_ikke_: Then what's the point ? I want to cancel the changes.	15:45
_ikke_	omarek: Okay, then use reset --hard	15:45
omarek	The changes in commit.	15:45
	← eletuchy left	15:46
	→ rethus joined	15:46
_ikke_	omarek: (I though you just wanted to undo the commit, not remove the commit with its changes entirely)	15:46
	→ noyb joined	15:46
	→ eletuchy joined	15:46
omarek	_ikke_: is it possible to do to a remote bare repo ?	15:46
	Without logging in via ssh, that is.	15:46
_ikke_	nope	15:47
	← sebrock left	15:47
bamj0rb	omarek: It's not a good idea to edit published history. If you must, you can push --force. You should understand the implications of that first... It won't remove that commit for other people. They'll need to manually do it (or clone anew).	15:48
kalleth	i just did it on a published branch on github that other people have cloned	15:48
	they threw things at me across the office.	15:48
	keep that in mind ;(	15:48
psavage	kalleth, yeh not a good idea ;)	15:48
	→ simesy joined	15:49
Nevik	kalleth: at least you'll remember it :D	15:49
omarek	bamj0rb: Okay, so what's the best thing to do ? Commit --ammend -m "sorry about that" ?	15:49
kalleth	i'd do it again!	15:49
omarek	and push it ?	15:49
Nevik	it's like locking yourself out of your apartment in your underwear	15:49
	it is burnt into your memory	15:49
	→ ericy joined	15:49
kalleth	it was that or have 5 commits in a row with the title "Buggeration."	15:49
bamj0rb	omarek: commit --amend is still editing history. Same thing.	15:49
psavage	It's like the whole series of Steins;Gate ;)	15:49
Nevik	kalleth: why would you push them? o.O	15:49
psavage	kalleth, squash them?	15:50
	then you'd only have 1 :p	15:50
Nevik	psavage: i dont watch titles with semicola in them	15:50
omarek	The change I commited turns out to be harmful.	15:50
psavage	Nevik, our loss :p	15:50
kalleth	because they were to fix a weirdness with CI and i couldn't be arsed to branch off to a different branch to perform the fix and then adjust CI to trigger on that branch ;)	15:50
bamj0rb	omarek: You can use git revert to make a new commit that undoes an undesirable one. That is best if you want to undo it and still preserve history.	15:50
psavage	your :p	15:50
Nevik	psavage: hehe	15:50
	kalleth: then you deserve things to be throw at you	15:51
omarek	bamj0rb: Doable to a remote bare repo ?	15:51
Nevik	especially heavy, desnse objects	15:51
	at high velocities	15:51
bamj0rb	omarek: If you are in control of the project/remote/etc., then you could still edit history (I've edited history published to GitHub tons of times... Nobody is interested enough in my code to care :P).	15:51
psavage	sometimes people use it as part of their workflow	15:51
bamj0rb	omarek: You are just making a new commit. It is like manually undoing everything bad and committing again: "Ooops, this was stupid, undone..."	15:51
Nevik	bamj0rb: you could always edit history -- at the risk of the hate of everyone else in the world :P	15:52
	→ Goplat joined	15:52
bamj0rb	Nevik: Sure. Unless you actually get everybody to cooperate though your history editing will not work. They don't bother, they'll merge back in with you, etc. The history will still ultimately be there when you fetch again.	15:52
Nevik	that depends on who has push access	15:53
	in a stanard fork-PR workflow it's no problem	15:53
	because you can just reject everything that isnt right	15:53
	bamj0rb is a rebase whore. I am no stranger to editing history. I edit every commit like 20 times. :P	15:53
psavage	Depends on how the workflow goes, if you have someone who is in complete charge of a canonical repo, you may have contributors that clean the PRs ;)	15:53
	← armin_w left	15:53
Nevik	i rebase a lot too	15:53
	but i dont collaborate much	15:53
	psavage too	15:53
psavage	rebase --onto a lot too :)	15:53
	I love that command	15:53
Nevik	lol	15:53
	← simesy left	15:54
psavage	pick up a branch and move it else where :)	15:54
bamj0rb	Yes, I do that a lot as well.	15:54
Nevik	rebase --onto A^ A A^	15:54
	# keep branch moving so it doesnt go stale	15:54
bamj0rb	rebase --onto $new_base $thing~(git rev-list $new_base..$thing) $thing # ;)	15:55
Nevik	why use the revlist	15:55
bamj0rb	(Hmm, rev-list might be wrong.... I usually do log --oneline instead)	15:55
Nevik	but why	15:56
	rebase can take a difference	15:56
bamj0rb	Just saves you the trouble of counting revisions.	15:56
psavage	mine are usually branch tips anyway	15:56
	← Gonzo89 left	15:56
omarek	I just made a new commit.	15:57
bamj0rb	--onto A B C tells git to only take the revisions between B and C (if I'm thinking clearly this morning...).	15:57
Nevik	bamj0rb: using the new_base..thing is the same as not passing the third argument	15:57
bamj0rb	Nevik: The third argument checks out $thing (if you aren't already there).	15:57
	← Sonderblade left	15:57
Nevik	OH jessus	15:58
bamj0rb	If you're already on C then you can just do rebase $new_base.	15:58
	← rubatdub left	15:58
Nevik	youre using the list as a number indicator	15:58
bamj0rb	Only when I know that it's The Right Thing(tm).	15:58
Nevik	i'd jsut copy the shas from gitk which i have open anyway, lol	15:59
StarFire_	canton7: I am trying now with local branches and -- --all	15:59
psavage	hehehe ++ Nevik	15:59
bamj0rb	I do not use gitk. >:) I only use the command line. EVAR.	15:59
	← lindenle left	15:59
psavage	though these days it's magit in emacs ;)	15:59
StarFire_	canton7: It has now about 24000 commits instead of 23300	15:59
Nevik	bamj0rb: s/gitk/tig or git-log--graph/	15:59
	or what psavage said	15:59
	if youre an emacs kind of person	15:59
psavage	which i wasn't until 2-3 weeks ago	16:00
	← rethus left	16:00
psavage	now I'd never go back	16:00
	→ dreiss joined	16:01
Nevik	heh	16:01
	← TomyWork left	16:01
	bamj0rb uses Vim. >:)	16:01
psavage	getting flake8 python checking on the fly is really nice ;)	16:02
	← joaoh82 left	16:02
	→ joaoh82 joined	16:02
Nevik	im using vim for some stuff as well	16:03
canton7	StarFire_, sounds good	16:03
	← lorenx left	16:03
	bamj0rb uses Perl. >:)	16:03
	← omarek left	16:04
	→ LuxuryMode joined	16:05
	← LuxuryMode left	16:05
	← thierryp left	16:05
	← leakybuckit left	16:05
	→ LuxuryMode joined	16:05
	qsusOFFcs → qsuscs	16:06
Nevik	ive always wondered why perl has that extraneous vowel in its name	16:06
	that seems... unperlish to me	16:06
psavage	hehe	16:08
	→ sgambino joined	16:08
	→ CKyle joined	16:10
	← ericy left	16:11
	← tma996 left	16:12
	→ m0viefreak joined	16:12
	→ siavashserver joined	16:13
	← ttl117 left	16:14
	← bartzy left	16:14
siavashserver	Hello	16:14
gitinfo	siavashserver: 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.	16:14
	← cbreak-work left	16:14
	← carbon2 left	16:15
siavashserver	!backup send-email	16:15
gitinfo	Worried about your data while trying out stuff in your repo? To back up commit history on all branches/tags: `git clone --mirror`. To backup everything, including work tree and staging area: `tar cf repo-backup.tar repodir`. Or do your experiment in a throwaway clone instead. See also http://sethrobertson.github.com/GitBestPractices/#backups	16:15
	← threesome left	16:16
	← psavage left	16:16
	→ rubatdub joined	16:17
	← VictorCL left	16:17
	→ dsdeiz joined	16:18
	← wrouesnel left	16:18
	← eletuchy left	16:18
	→ eletuchy joined	16:19
	→ Orphis joined	16:19
	→ abelsiqueira joined	16:19
	→ pedda joined	16:19
	→ hyperair joined	16:20
	→ eruMelko_ joined	16:20
	← Ltk left	16:20
	← sabgenton left	16:21
	→ Jackneill joined	16:23
	← eletuchy left	16:23
	→ ngladitz joined	16:23
	← ngladitz left	16:23
	→ ngladitz joined	16:23
njbair	# this log format can be split per-virtual-host based on the first field	16:24
	LogFormat "%V %h %l %u %t \"%r\" %s %b" vcommon	16:24
	← rath left	16:25
	adhawkins → adhawkins-away	16:26
bamj0rb	(!backup is specifically help for backing up xD)	16:26
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	16:26
bamj0rb	Awww. >:( STFU gitinfo. Not now.	16:26
	← dsdeiz left	16:27
	→ Zarthus joined	16:27
	→ ericy joined	16:27
	→ Gavrisimo joined	16:30
StarFire_	bibi, cu	16:30
	→ lite_ joined	16:30
	← StarFire_ left	16:30
	← hyperair left	16:31
Nevik	bamj0rb: simple: use another punctuation in front of the exclamation mark	16:31
	like .!backup	16:31
bamj0rb	Ah.	16:32
	→ dhruvasagar joined	16:32
Nevik	i didnt know it would ignore the bracket though	16:32
	that's interesting	16:32
	→ TomyLobo joined	16:33
bamj0rb	It might be too complex, but in theory it could keep track of when it last said it, and if the target was joined at the time, and just tell the user to read the scrollback. :D	16:33
	Then if the user is really sure they want it repeated do !! instead. >:)	16:33
	← jargon left	16:34
	← dreiss left	16:34
bamj0rb	Nevik: Perl is not about consonants. It is about punctuation. :P	16:34
Nevik	btw that wouldnt be very complex	16:35
	but generally not necessary	16:35
	← Gavrisimo left	16:35
Nevik	if it bothers you, you can just write a reactive ignore script for your client	16:35
	← rubatdub left	16:35
	← pandeiro left	16:35
	← GeertV left	16:35
Nevik	shouldnt be a big deal for a perler :P	16:35
bamj0rb	It's a simple idea, but the implementation can certainly be finicky. It would be easy to do, but difficult to tune /well/.	16:35
	Like you said, it's probably not worth the trouble.	16:36
	← aChatir left	16:36
Nevik	well, tuning something "well" is never done anyway	16:36
	you always just make things good enough	16:36
	which in this case would be 5-10 minutes or 30-50 lines	16:36
	→ King_Hual` joined	16:37
	→ rjhunter joined	16:37
	← rob_w left	16:37
	→ endiruna joined	16:37
Nevik	s/or/and/	16:37
	← user258467 left	16:38
	← LuxuryMode left	16:38
bamj0rb	Are you sure about that? What about network interruptions, odd scrollback, IRC n00bs that don't know what you're talking about, etc. It might end up being more noise than just repeating the help. :)	16:38
	← pen left	16:38
bamj0rb	Who maintains gitinfo anyway?	16:39
	→ yhager joined	16:39
	→ user258467 joined	16:40
Nevik	jast	16:40
	← King_Hual left	16:41
	→ hobodave joined	16:41
	← King_Hual` left	16:41
Nevik	bamj0rb: https://github.com/jast/gitinfo/	16:41
	→ King_Hual joined	16:41
	→ aberrant joined	16:41
Nevik	youre gonna love those language stats	16:41
	← endiruna left	16:42
	← rjhunter left	16:42
	→ endiruna joined	16:42
	→ doev joined	16:42
	← jnavila left	16:43
	← siavashserver left	16:44
	← abelsiqueira left	16:45
	→ dsdeiz joined	16:45
	← ciampix left	16:45
	→ dsantiago joined	16:46
bamj0rb	The language stats seem to be removed from GitHub...	16:46
	← noyb left	16:47
	→ rubatdub joined	16:47
	→ LuxuryMode joined	16:47
	← dsantiago left	16:47
	→ hyperair joined	16:47
Nevik	nope	16:48
	theyre in the same place where they used to be, only now you have to click them to see the numbers	16:48
	it's a silly change that no one at github has yet seen fit to fix	16:48
	because fuck the users	16:48
bamj0rb	I don't remember where they used to be. :(	16:48
	→ Bass10 joined	16:49
	→ rcu joined	16:49
	→ eletuchy joined	16:49
	← Bass10 left	16:49
bamj0rb	Oh, there we go. Just a mysterous line. BRILANT UI GitHub.	16:49
	← kjellski left	16:49
	← dhruvasagar left	16:50
	← ericy left	16:50
	→ Bass10 joined	16:50
	← eruMelko_ left	16:50
	← rcu left	16:50
bamj0rb	I should probably study gitinfo's code. I am in need of learning an event framework in Perl. Bambot is in need of refactoring. >:)	16:51
	→ irqq_ joined	16:51
	← bbloom left	16:51
	← kukks left	16:52
Nevik	lul	16:52
	→ kriks joined	16:52
Nevik	bamj0rb: feel free to let github knokw how you feel about the language bar :P	16:52
	← johnkevi_ left	16:53
	← gua_ould left	16:53
	→ pheaver joined	16:54
	← irqq left	16:54
	irqq_ → irqq	16:54
bamj0rb	If I'm going to talk to a brick wall I want it to be one that I like.	16:54
	← Misan left	16:56
	Nevik hands bamj0rb finger paint	16:56
	→ abelsiqueira joined	16:57
	← arjun_ left	16:57
	← yhager left	16:58
	→ Norm_ joined	16:59
	← bvi left	16:59
	→ Noldorin joined	17:01
	← pheaver left	17:01
	← lite_ left	17:02
	→ VictorCL joined	17:03
	→ Abbas-- joined	17:03
	→ lite_ joined	17:03
Abbas--	guys, i keep getting this error. i just added my github repo as git origin master. http://i.imgur.com/OaDwvcS.jpg	17:03
	when i try to push from local folder, this is error i get ^	17:03
bamj0rb	Your remote is origin. origin/master is a remote tracking branch.	17:04
imMute	Abbas--: did you read the message? it's pretty detailed	17:04
	→ groglogic joined	17:04
Abbas--	imMute what is an upstream branch?	17:04
bamj0rb	See also man git-push.	17:04
gitinfo	the git-push manpage is available at http://jk.gs/git-push.html	17:04
	← CKyle left	17:04
imMute	Abbas--: the branch on the remote	17:04
	← LuxuryMode left	17:05
	← lb1a left	17:05
imMute	Abbas--: translation: "I have no idea where I should push this to. Here's how to tell me."	17:05
Abbas--	ah , thanks :)	17:05
	so it shud act like normal after i run that command?	17:05
	→ Macka joined	17:06
Abbas--	i see progress :D thanks http://i.imgur.com/nxEy4ce.jpg	17:06
	← Davey left	17:06
imMute	depends on your definition of "normal" (and the setting for ... push.default)	17:06
	→ ericy joined	17:06
	← ericy left	17:06
Abbas--	I'm the sole developer for my git repo. i just use it so i can have multiple time-frame backups	17:07
	→ drjd joined	17:07
	→ ericy joined	17:07
Abbas--	so as long as this allows me to push new code without losing previous history on github, it'll be great	17:07
	← alorence left	17:07
	→ stany_ joined	17:07
	← PaulCapestany left	17:09
	→ mikecmpbll joined	17:09
Freeaqingme	Hi; I need to move 20 submodules within my repository. How should I do so?	17:09
	→ LuxuryMode joined	17:09
Macka	Hi. I'm preparing to migrate a Perforce repo to git. I've used git-p4.py to clone the repo which produced a very large 598M git repo. There's one single file there that's a whopping 405M.	17:11
	.git/objects/pack/pack-93e056ddb68d22c2e0e515446c1837305e543033.pack	17:11
	It's not a blob	17:11
	← jwmann left	17:11
Macka	so don't think I can find it and prune it with a filter-branch	17:12
	← stany_ left	17:12
Freeaqingme	Macka, out of curiosity, why are you migrating?	17:12
	← mikecmpbll left	17:12
imMute	Hassen: that's a packfile. it's a bunch of objects collected into a single file (like a tarball, but extra compression and deltas)	17:12
Macka	is there a way to reduce the size of that, or get rid of it without loosing any of the commit history?	17:12
Hassen	imMute, at what?	17:13
	→ mikecmpbll joined	17:13
imMute	Macka: you can repack it, and tell it to compress more aggressively	17:13
	Hassen: sorry, wrong nick.	17:13
	← rurufufuss left	17:13
imMute	Macka: see my message to Hassen	17:13
Hassen	imMute, i see.	17:14
Macka	we're migrating because the company is splitting. Also we've moving to a new model that will use Gerrit & Jenkins.	17:14
	← aberrant left	17:14
	→ threesome joined	17:14
	← ThomasLocke left	17:15
	→ dhruvasagar joined	17:15
	→ shwaiil joined	17:16
shwaiil	HI	17:16
	oops! Hi: )	17:16
	→ TheFuzzball joined	17:17
shwaiil	Q: Can I prevent other members committing to the Master Branch ? I'd like them to create their own branch from Master but not being able to merge and push to Master branch - if this makes sense ? Thanks!	17:17
	→ jwmann joined	17:17
moritz	shwaiil: !gitolite has per-branch write access control	17:17
gitinfo	shwaiil: 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/	17:17
shwaiil	moritz: thanks :)	17:18
Macka	so I need to unpack the packfile to discover what's big in there, and clean up that way?	17:18
shwaiil	moritz: I've always worked solo, I guess Pull Requests are done in a different way correct and probably better for this cases when you want to keep control of your repo ?	17:18
imMute	Macka: no, "git repack" is the command you want.	17:18
	Macka: see https://www.kernel.org/pub/software/scm/git/docs/git-repack.html for more	17:19
	→ HappyLappy joined	17:19
	→ Gavrisimo joined	17:19
Macka	thanks very much. will read up on that and give it a go.	17:20
	← fRAiLtY- left	17:21
	← lun__ left	17:22
	← TheFuzzball left	17:22
	→ master_j joined	17:22
	→ yhager joined	17:22
	→ Kayra joined	17:22
	→ Kayra_ joined	17:24
moritz	shwaiil: yes, that's one approach	17:24
	← Kayra left	17:24
	→ PaulCapestany joined	17:24
moritz	I personally prefer trust, and haven't been burned yet	17:24
	← fairuz left	17:25
shwaiil	moritz: yeah I see. So, basically, I can just ask my colleges not to push to Master and do a Pull request from their branch. If this makes sense ?	17:25
moritz	shwaiil: or you can just allow them to push to master, and ask them get a review first if they push anything controversal	17:25
	s/if/before/	17:26
shwaiil	moritz: ok, so I can do that too I see. Thanks!	17:26
moritz	it's really a cultural thing	17:26
	← psyl0n left	17:26
moritz	you have to talk to them, and find out what works for everybody	17:26
	→ bobbobbins joined	17:26
shwaiil	yeah, I think I'll go with the pull request, first because I never done that as I always work solo	17:27
	second, because I need to send stuff for staging and the production: a shell script does this for me, that I created	17:27
	at least I could trigger that or do it manually if everything's ok, no conflicts and stuff.	17:27
	→ siavashserver joined	17:28
siavashserver	How can I send-email a specific git commit?	17:28
	← mishok13 left	17:28
	← ericy left	17:28
imMute	siavashserver: run the hook manually with whatever arguments it expects to send that commit.	17:29
	→ kukks joined	17:29
	← kukks left	17:29
	→ kukks joined	17:29
	→ lun___ joined	17:29
siavashserver	imMute: simply git send-email?	17:30
	← donmichelangelo left	17:30
	← vtunka left	17:30
imMute	oh, send-email is an actual git command. ignore everything I said.	17:31
	→ psyl0n joined	17:31
	← psyl0n left	17:31
	→ psyl0n joined	17:31
	→ donmichelangelo joined	17:31
	→ dxdt joined	17:31
	siavashserver erases	17:31
	← Goplat left	17:32
	→ fairuz joined	17:32
	← rainbyte left	17:33
siavashserver	Never mind got it	17:34
bremner	is it "-1" ?	17:34
	← dsdeiz left	17:34
siavashserver	bremner: No I'm not after last commit	17:34
	git format-patch --cover-letter -M origin -o outgoing/	17:34
	and manually git send email /outgoing/whatever	17:34
	← Noldorin left	17:34
bremner	but there is a -1 argument to git-format-patch that does a single patch	17:35
	← master_j left	17:35
	→ Noldorin joined	17:35
	← Balliad left	17:35
bremner	and those args can also be passed to git-send-email	17:35
	→ BiggFREE joined	17:35
	← intellix left	17:36
siavashserver	nice	17:37
	→ btree joined	17:38
	← teran left	17:38
	← pfFredd left	17:39
	→ master_j joined	17:39
	← btree left	17:40
	→ aberrant joined	17:40
	← koda left	17:41
	← Octalot left	17:41
	← Abbas-- left	17:41
	→ thiago joined	17:42
	← blinkdesign left	17:43
	← achadwick left	17:43
	← HappyLappy left	17:44
	→ kirin` joined	17:44
	→ HiramHibbard joined	17:45
	← St_Marx left	17:45
	→ pen joined	17:46
	← FerreusR left	17:46
	→ angelsl joined	17:46
	→ ericy joined	17:46
	→ mfulz joined	17:46
	→ pencilcheck joined	17:47
	→ St_Marx joined	17:47
	← pencilcheck left	17:47
	← pen left	17:47
	→ pen joined	17:48
	← juril left	17:48
	→ kallis joined	17:49
	← siavashserver left	17:49
	← quejajajajaja left	17:51
	← xobb left	17:51
	← groglogic left	17:51
	→ groglogic joined	17:52
	← tango_ left	17:52
	← kukks left	17:53
	→ kukks joined	17:53
	← MattMaker left	17:54
	← mikecmpbll left	17:54
	→ mikecmpbll joined	17:56
	→ User458764 joined	17:57
	← kallis left	17:59
	← dhruvasagar left	18:01
	← groglogic left	18:02
	← Vampire0 left	18:05
	← singham left	18:06
	← pen left	18:06
	→ pen joined	18:06
	→ noyb joined	18:06
	← xpe left	18:07
	→ looper joined	18:08
	→ diosmiux joined	18:09
	← spaceonline left	18:09
	← aberrant left	18:09
	→ GarethAdams joined	18:10
	← ericy left	18:10
	→ kjellski joined	18:14
	→ pheaver joined	18:14
	← Kayra_ left	18:15
	← Sendoushi left	18:16
	← bobbobbins left	18:16
	→ devbug joined	18:17
	→ Vampire0 joined	18:18
	← alip left	18:18
	→ lbod joined	18:19
	← asQuirreL left	18:19
	← Ingmar left	18:19
	← User458764 left	18:20
	→ asQuirreL joined	18:21
lbod	hi, this should be quick for you guys but it's workflow im constantly using at the moment. I'm working against an upstream and origin at the moment and my commits are added to upstream via GH pull requests. I want to blast everything in my local master, overwriting every file and any in progress commits, merges etc and then push that to my remote origin so it's always in sync with the remote upstream.. Whats the best way to do this?	18:21
	← edude03 left	18:21
	→ ErKa joined	18:22
	← joaoh82 left	18:22
bamj0rb	git checkout master && git reset --hard upstream/master # (assuming your upstream remote is named "upstream")	18:23
	→ joaoh82 joined	18:23
	← ChrisAnn left	18:23
lbod	bamj0rb so if i get "Your branch and 'upstream/master' have diverged" should i just use -f?	18:24
bamj0rb	lbod: Warning. That will happily destroy uncommitted state, and it will hide all of your "local master stuff".	18:24
lbod	yeh thats fine	18:24
bamj0rb	Depends if you want to keep what you have in your branch or not. If you just want to follow upstream the you probably want to reset --hard.	18:25
	Or, if you just want to push upstream/master to origin/master, you can do that directly: git push origin upstream/master:master	18:25
	Add a -f if necessary (and you're sure).	18:26
	→ ericy joined	18:26
lbod	i think i follow.. ive been using git for a while now, only at home and not at work but it still confuses the hell outa me, makes me feel very inexperienced	18:26
	→ codiez joined	18:27
bamj0rb	If you just want to avoid messy merging, but want to preserve your own work, the you could instead rebase onto upstream/master. That would effectively move your divergent commits onto the end of upstream/master (so no merge necessary).	18:27
lbod	after checking git reflog, that looks like it worked ok, even though it warned me about merges	18:27
bamj0rb	Probably means you had previously merged (or pulled, which is fetch + merge) with origin or something.	18:28
lbod	yeh, ive been trying to use rebase and ended up in a messy state too.. i guess i need to create some test remote repos and work on them	18:28
bamj0rb	Yes, you just need to practice it a bit and you'll figure it out. It's actually really easy, but if you haven't done it before then it's easy to get confused.	18:29
	← michael_lee left	18:29
lbod	reflog looks fine.. a side note, how long till you got properly acquainted with remote repo's?	18:29
	(and thanks btw)	18:29
	→ jAyenGreen joined	18:30
	gitinfo set mode: +v	18:30
	← codiez left	18:30
	→ l0cust joined	18:30
lbod	bamj0rb i think thats what's starting to annoy me is i feel ive been doing it for a while but with really basic workflow and all against my own remote repo	18:31
	→ codiez joined	18:31
	← shwaiil left	18:32
	→ edude03 joined	18:32
jAyenGreen	When I create a submodule, commit it to the base and then push the base, then clone the repo locally, it seems the fact that the local path is different from the remote path results in an error about the submodule path not being a repo	18:32
bamj0rb	lbod: I basically got curious about remotes pretty early. First I created a local one to push/pull with. The I created a GitHub account and started using that.	18:32
	lbod: If you're obsessive compulsive then that helps to encourage you to figure out how it works. :)	18:33
	← Hello71 left	18:33
	→ bobbobbins joined	18:33
lbod	you might find it amusing, we still use cvs at work, so i only ever use git from home on o/s projects.. anyway, thx for the help..	18:34
bamj0rb	lbod: (On the other had, I find the reflog hard to follow, so if it looks fine to you then that's something to be proud of :D)	18:34
lbod	ahah shit, mibbe im wrong then :D	18:34
	← diosmiux left	18:35
bamj0rb	lbod: It happens. You can be the force that triggers the switch to Git. There might also be a tool to interface Git with CVS (there is for Subversion). Worst case, you could always use both (secretly using Git). That's pretty common.	18:35
	→ Hello71 joined	18:35
	← codiez left	18:35
	← pedda left	18:36
bamj0rb	They predominantly use Mercurial at my company (which I am not happy about, but they're too stoopit for Git), and where I'm contracting right now they usually use nothing. D:	18:36
	→ matanya joined	18:36
	← HiramHibbard left	18:36
bamj0rb	I basic find(1) reveals like 120 copies of binaries... Let alone source copies...	18:36
Hello71	the correct solution is to name one file for each version in DVCS	18:36
	because then it's distributed	18:37
	→ dhruvasagar joined	18:38
	→ blinkdesign joined	18:38
	← blinkdesign left	18:38
	→ blinkdesign joined	18:39
	← m0ltar left	18:39
	← asQuirreL left	18:39
	→ joelrobichaud joined	18:39
	← l0cust left	18:40
	→ danecando joined	18:40
	→ speeder joined	18:40
speeder	can I sort of create a stash in a git repository, and apply it in ANOTHER repository?	18:40
	→ hs366 joined	18:41
jAyenGreen	If someone could look at the steps I followed and tell me where I went wrong I'd greatly appreciate it http://pastebin.com/tkkizPfJ	18:41
	→ richwestcoast joined	18:41
Hello71	speeder: you are looking for the "patch" function.	18:42
	← petey left	18:42
	← Gavrisimo left	18:42
speeder	thanks :)	18:42
bamj0rb	speeder: \o/	18:42
speeder	bamj0rb: hey!	18:42
	=D	18:42
	→ petey joined	18:42
canton7	speeder, or cherry-pick a commit	18:42
EugeneKay	speeder - Sure, a stash is just a (pair of) commit(s) stored with a funny ref. `git stash` in A, then in B `git fetch A refs/stash` to yank it over	18:42
	Or patch, sure.	18:43
lbod	thx again & bye, helpful ppl in here!	18:43
	← lbod left	18:43
speeder	I want to do that because I am creating several games that have the same source code (only assets change, and some scripts), and that is my best idea to update they all at once...	18:44
	← joaoh82 left	18:44
	→ joaoh82 joined	18:44
bamj0rb	The stash? Noo...	18:44
	→ Gavrisim_ joined	18:44
	→ BigBlueBacon joined	18:44
EugeneKay	Then it's a traditional patch scenario	18:44
	→ HiramHibbard joined	18:44
	jAyenGreen checks breath	18:44
	→ jnavila joined	18:45
	→ petey_ joined	18:45
	→ gua_ould joined	18:45
	← petey left	18:46
	← eletuchy left	18:46
	← bobbobbins left	18:47
	← richwestcoast left	18:47
	→ undecim joined	18:48
	→ richwestcoast joined	18:48
	→ bobbobbins joined	18:49
	← shinnya left	18:50
	← abelsiqueira left	18:51
	← ericy left	18:51
	← HiramHibbard left	18:51
	← noyb left	18:52
	→ ravigehlot joined	18:52
	→ fossrox joined	18:53
	← vurma left	18:53
	← mikecmpbll left	18:53
	→ Black_Prince joined	18:53
	← jnavila left	18:53
Nevik	speeder: git subtree might be interesting for you	18:53
	or submodules	18:53
	→ JustinMac joined	18:54
Black_Prince	what happened to git remote helpers python module in git 1.8.5?	18:54
ravigehlot	By mistake, I ended up including some large video files on my git index. The problem is that my hosting account has exceeded its quota. When I git push origin master, it exceeded the quota and hung up. So I git rm -r -cached videos/, re-committed. But when I re-run git push origin master, it tries to compress/commit the previous commit instead of deleting the videos.	18:54
cbreak	ravigehlot: git stores history	18:55
	it does not delete it	18:55
	→ jnavila joined	18:55
cbreak	ravigehlot: if you want to delete it, rewrite local history, so that the commit in which you added the files never happened	18:55
	ravigehlot: can you do a !lol and pastebin the first dozen lines or so?	18:55
gitinfo	ravigehlot: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	18:55
ravigehlot	cbreak: that makes sense	18:55
	← Sadeukko left	18:56
Nevik	ravigehlot: is your host github?	18:56
ravigehlot	Nevik: hawkhost	18:56
	Nevik: I got 3 GB total	18:56
Nevik	k, nvm :D	18:56
cbreak	github has a limit? :)	18:56
	→ liamkeil1 joined	18:56
Nevik	cbreak: yes	18:56
	but it's a limit, not a quote	18:56
	quota* even	18:56
ravigehlot	cbreak: how can I re-write history to have it forget the faulty commit?	18:56
Nevik	though depending on what youre thinking of as a quota, it might be called that too	18:56
cbreak	ravigehlot: in your situation, the easiest would probably be to reset the commit away	18:57
	→ codiez joined	18:57
cbreak	ravigehlot: git reset --soft commitbefore	18:57
	then git rm --cached -f filesyouwantgone && git commit -C oldcommit	18:57
	→ ciaranm joined	18:57
speeder	nevik too much confusing	18:57
cbreak	then git push -f	18:58
ravigehlot	cbreak: right. But I went back and made two other commits. One to delete the videos/ and another one to delete public/. So now I am ahead by 3 commits.	18:58
	cbreak: should I ran that 3 times?	18:58
Nevik	speeder: i didnt say you HAVE to use them. but using them is probably less hassle than patching everything manually all the time	18:58
	→ sq-one joined	18:58
speeder	nevik I tried before, did not worked, but I don't remember why	18:59
	← ereslibre left	18:59
Nevik	i wouldnt know :P	18:59
	← Noldorin left	18:59
pekster	I have a project and a local clone I'm using as a remote; on the remote is an 'expiry' branch and I'd like to cherry-pick 1 commit, but I need to change a couple lines as it contains info that "should not" go into the upstream. Is there a sane way to let git know I've "already merged" that commit since the final merge won't be the commit from my remote?	18:59
	→ abelsiqueira joined	18:59
	→ FerreusR joined	19:00
	← Black_Prince left	19:01
	← doev left	19:01
bamj0rb	speeder: Any way to make the program shared and just supply it different data (where scripts are also data)? You could track the assets/scripts without the program, and just combine them when installed.	19:01
	→ leeN joined	19:01
	→ HiramHibbard joined	19:01
	→ robertnix joined	19:01
	← HiramHibbard left	19:02
bamj0rb	speeder: i.e., the_game -C path/to/game/assets	19:02
EugeneKay	Huh	19:02
	Try that one again?	19:02
speeder	bamj0rb: the problem is that my stupid SDK (Corona) do not support that	19:03
Nevik	bamj0rb: the assets are probably part of a development environment	19:03
bamj0rb	Seems like a good opportunity to refactor. ;)	19:03
speeder	Corona works by packing everything into a directory as the game (and everything I mean EVERYTHING, any random garbage there, go with it...)	19:03
Nevik	bamj0rb: that doesnt help if you need them in your project configuration	19:04
speeder	and Corona does not support build scripts (on purpose)	19:04
	← abelsiqueira left	19:04
cbreak	ravigehlot: you should give me the !lol output	19:04
gitinfo	ravigehlot: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	19:04
speeder	(if you want to use automated build you have to pay them 1k USD)	19:04
	gitinfo set mode: +v	19:04
cbreak	ravigehlot: basically, you want to reset to the commit before the mess started	19:04
Nevik	speeder: oh btw, you're not on linux for that kind of development, eh?	19:04
	→ abelsiqueira joined	19:04
speeder	Nevik OSX	19:04
Nevik	symlinks might be an easy solution	19:04
bamj0rb	speeder: Ah, so you're using a third party proprietary framework...	19:04
speeder	because iOS...	19:04
	← AaronMT left	19:05
	← yhager left	19:05
Nevik	dunno how well macos's fs does symlinks	19:05
bamj0rb	It is UNIX. Should be fine.	19:05
speeder	Nevik OSX is a fork of BSD	19:05
Nevik	bamj0rb: not if the fs doesnt like symlinks	19:05
pekster	HFS+ does symlinks fine; what the proprietary frameworks does though..	19:05
	→ yhager joined	19:05
Nevik	thanks pekster ;)	19:05
speeder	Corona dislike symlinks and hardlinks	19:05
Nevik	lawl	19:05
	so fail	19:05
speeder	I already tried (and ended with a crazy mess)	19:06
ravigehlot	cbreak: give me a sec please	19:06
Nevik	speeder: why did you pick that :P	19:06
speeder	Nevik because it was really fast to develop cross-platform stuff	19:06
	now I use it for legacy stuff	19:06
cbreak	Qt.	19:06
speeder	I switched 3 months ago to a new one	19:06
Nevik	ah	19:06
cbreak	the one and only.	19:06
speeder	cbreak: games, not business apps	19:06
	← jnavila left	19:06
speeder	cbreak: also cross-platform I mean, Android + iOS	19:06
cbreak	OS X supports symlinks perfectly :)	19:07
Nevik	cbreak: you should read	19:07
cbreak	speeder: yeah, Qt can do that :)	19:07
	(somewhat :)	19:07
Nevik	lol	19:07
	Nevik gives cbreak a cookie	19:07
	→ ericy joined	19:07
	→ daito_ joined	19:07
	→ MattMaker joined	19:07
	cbreak has cookies disabled	19:07
speeder	Qt for making games is like trying to nail something using a spoon in place of a hammer	19:07
	yes, you can do it, but...	19:07
	→ AaronMT joined	19:08
	→ JustinMa_ joined	19:08
	← BigBlueBacon left	19:09
	← liamkeil1 left	19:09
	→ thierryp joined	19:10
	→ zhost joined	19:10
	→ ehmatthes joined	19:10
	→ Gamecubic joined	19:11
	← daito_ left	19:12
fossrox	hi guys, i don't have much experience with git yet and have such as issue: i converted my bzr repo to git, and then i updated the email information according to 1st answer http://stackoverflow.com/questions/750172/how-do-i-change-the-author-of-a-commit-in-git -- all looked ok, meaning that 'git log' was showing new email and complete list of commits	19:12
	→ tango_ joined	19:13
fossrox	but then i notices that with git log --all i can still see old email addresses, and when tried to play with python-git i could see imports and rewrites (the email rewrites), is there a way i can remove that completely from my repo? how much i should worry about this not going to published to public repo?	19:14
	← abelsiqueira left	19:14
	← ehmatthes left	19:15
	← JustinMa_ left	19:15
	→ Octalot joined	19:15
	→ abelsiqueira joined	19:16
fossrox	in next step i installed some gui tools to visualize the branches, i have there only master branch with correct commits and then some commits from import which do not belong to any branch at all, not sure how to get rid of that stuff	19:16
	→ valgrind joined	19:17
Nevik	fossrox: can you post a !repro or screenshot of that git log showing both old and new email address?	19:17
gitinfo	fossrox: 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.	19:17
Nevik	my guess is you have some branch or tag that still points to the non-rewritten history	19:17
	← kukks left	19:18
fossrox	ok, will start with paste	19:18
	← master_j left	19:19
	← dxdt left	19:20
kpreid	is there a convenient way to remove empty directories (only)?	19:20
	→ User458764 joined	19:20
Nevik	kpreid: reclone :P	19:20
	→ jnavila joined	19:21
kpreid	oh, hm, it had ignored files which is probably why it didn't get auto removed anyway	19:21
Nevik	otherwise, besides writing a script for it, no	19:21
moritz	find -type d -empty -delete	19:21
	doesn't need a script :-)	19:21
Nevik	ah	19:21
	forgot about find	19:21
	the magic all-cure	19:21
	→ aberrant joined	19:22
	← aberrant left	19:22
	→ Sendoushi joined	19:24
	← St_Marx left	19:24
	← jkremser left	19:25
	→ damccull joined	19:25
damccull	If someone submitted a pull request to my github project and I want to take the code from their request but not the changes to the version files...can I do that?	19:25
Nevik	damccull: sure	19:26
damccull	where should I look to learn how?	19:26
	→ lamont joined	19:26
	← ericy left	19:26
	← abelsiqueira left	19:26
Nevik	damccull: just trying it out. your friend is the checkout command in this case	19:27
	what exactly to do depends on both the situation and what you want it to look like afterward	19:27
damccull	Hmmm. Like there's 4 files changed and I want 3 of them	19:27
Nevik	damccull: are the version changes (that you dont want) in an extra commit or in the same commit as the other changes? (if the repo is public, feel free to link us)	19:29
	→ abelsiqueira joined	19:29
damccull	https://github.com/damccull/sublimetext-SolarizedToggle/pull/12/files	19:30
	I want all the files except the bottom two: messages.json, and update-*	19:30
	→ atomx joined	19:30
	← hs366 left	19:30
	← edude03 left	19:30
	→ hs366 joined	19:31
	→ deuti joined	19:31
Nevik	damccull: will it be a problem if the updated versions are in your history, but you remove them in another commit?	19:31
damccull	not really.	19:32
	can you back-date individual files?	19:32
Nevik	damccull: then you can merge the PR (manually or via the website) and then undo the changes in an additional commit	19:32
damccull	hmm	19:32
Nevik	what do you mean by back-date files? git does not keep track of individual files' dates	19:32
	it only keeps the dates of commits	19:32
	i can also tell you how to modify the second commit to remove the changes from history, if you want	19:33
cbreak	damccull: forget about pull/merge	19:33
	and do a git cherry-pick -n instead?	19:33
	or a normal cherry-pick and a commit --amend?	19:33
	I think I'd do the latter.	19:34
	→ nbrosnahan joined	19:34
	Nevik sighs	19:34
damccull	i shall google cherry pick	19:34
cbreak	there's man git-cherry-pick	19:34
gitinfo	the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html	19:34
	→ JustinMa_ joined	19:34
damccull	thanks. reading.	19:35
	← davo left	19:35
	← LuxuryMode left	19:35
	→ finspin joined	19:36
	← codiez left	19:36
	→ codiez joined	19:37
	→ LuxuryMode joined	19:38
fossrox	Nevik, gitinfo http://paste.debian.net/68309/ -- does it help?	19:38
finspin	i have cloned remote repository, created a new branch, made changes but haven't committed yet. remote master have changed in the meanwhile. what's the best way to proceed?	19:39
	← JustinMa_ left	19:39
	→ Nazca__ joined	19:39
davividal	We recently migrated from SVN on a (PHP) Symfony application. We have a lot of config files that we need to be versioned and a couple of times a year we need to change. Yet, we need to modify those same files for each developer (folders, URLs, emails, bla bla bla). So, we can't use .gitignore. What's the alternative?	19:39
	← pheaver left	19:39
	→ Ingmar joined	19:39
	← Ingmar left	19:39
	→ Ingmar joined	19:39
finspin	i thought of committing changes to local branch, switch to master, pull the latest version and merge my branch. there will be conflicts to resolve, right?	19:40
	→ cilly joined	19:40
finspin	oh yeah, i forgot to mention that i want to push changes back to remote master	19:40
	← codiez left	19:41
bamj0rb	finspin: Conflicts are nothing to worry about (!panic). Also, there's no guarantee there will be conflicts. Also, you can either merge or rebase (similar, but different).	19:41
gitinfo	finspin: [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, don't panic. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice!	19:41
	← zhost left	19:41
Nevik	fossrox: that ipython shows the rewrite need not concern you. my guess is that it includes events from the reflog (which git log doesnt)	19:41
	← joaoh82 left	19:41
	→ eletuchy joined	19:41
	← nbrosnahan left	19:42
	→ ncd\|leeN joined	19:42
	← diegoviola left	19:42
	← ncd\|leeN left	19:42
	→ joaoh82 joined	19:42
Nevik	fossrox: as for the duplicate history, the normal log doesnt show branch names so that output doesnt help so much. try !lol and see if that makes sense to you	19:42
gitinfo	fossrox: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	19:42
	← leeN left	19:42
finspin	bamj0rb: gitinfo: thanks guys. i haven't done this before so just wanted to make sure im not doing anything stupid.	19:42
gitinfo	finspin: you're welcome, but please note that I'm a bot. I'm not programmed to care.	19:42
Nevik	fossrox: if it doesnt, paste it again, i hope it doesnt need too much editing	19:42
	← Nazca left	19:42
Nevik	fossrox: if the names arent too confidential, you can also make the pastie private and pm me the link	19:43
	→ leeN joined	19:43
Nevik	i love how the bot auto-responds to thanks	19:43
	→ nicedice joined	19:44
	→ ericy joined	19:44
finspin	Nevik: love it too! it made me smile. for a second i imagined this super smart robot getting bored answering the same newbie questions all over again :)	19:44
Nevik	;)	19:44
	← takamichi left	19:44
fossrox	!lol	19:45
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	19:45
	→ zhost joined	19:45
offby1	!herp derp	19:45
EugeneKay	.trigger_edit herp derp	19:45
gitinfo	EugeneKay: Okay.	19:45
Nevik	LOL	19:45
	totally worth it	19:45
	→ nbrosnahan joined	19:46
	← leeN left	19:46
	→ leeN joined	19:47
	→ Noldorin joined	19:47
	→ davo joined	19:47
luto	!herp	19:48
gitinfo	derp	19:48
	← Hassen left	19:48
Nevik	!gitten	19:48
gitinfo	Here, have a cute and cuddly Gitten. Nyaa~	19:48
	← Sendoushi left	19:49
fossrox	Nevik: http://paste.debian.net/68315/ -- added the graph on the bottom	19:49
	→ petey joined	19:50
	→ Sendoushi joined	19:50
	→ stodan joined	19:50
davividal	We recently migrated from SVN on a (PHP) Symfony application. We have a lot of config files that we need to be versioned and a couple of times a year we need to change. Yet, we need to modify those same files for each developer (folders, URLs, emails, bla bla bla). So, we can't use .gitignore. What's the alternative?	19:51
fossrox	with gui the bottom/duplicated commits look like one line in a separated branch parallel to master	19:51
Nevik	fossrox: is [email@hidden.address] or <example.com> the name+email you want to have? (i.e. which is the old vs new one)	19:51
	→ codiez joined	19:51
Cathy	davividal: you could have the versioned config file called config.template or something	19:51
fossrox	Nevik: i want to have it 'User <example.com>' to avoid spam in future	19:51
moritz	davividal: the normal approach is to include your.conf.example in the git repo	19:51
Cathy	davividal: then each developer copies it to config	19:52
moritz	davividal: and .gitignore the actual .conf file	19:52
fossrox	but people should figure out how to contact still i think	19:52
Nevik	fossrox: heh. that's a bit weird of a measure but okay.	19:52
	fossrox: your history is fine	19:52
Cathy	i thought worrying about spam was passé in 2013	19:52
Nevik	fossrox: in the graph you can see the bottom strip to be origina/refs/heads/master, which is a backup ref created by filter-branch when you rewrote history	19:52
	fossrox: if the new history (pointed to by the branch master) is like you want it, you can delete that backup ref	19:53
	→ Skrypter joined	19:53
fossrox	Nevik: how i would go about the deletion?	19:53
	← joki left	19:53
davividal	Cathy: moritz yeah, the usual would be great, but I'm instructed to do small changes a time. So renaming is out for now :( I was thinking about update-index --assume-unchanged	19:53
	← petey_ left	19:54
moritz	davividal: renames are small changes in git land :-)	19:54
Nevik	fossrox: if you want to be absolutely safe, make a !backup of your repo; then remove the folder .git/refs/original/	19:54
gitinfo	fossrox: 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	19:54
Cathy	davividal: that's also very error prone	19:54
davividal	moritz: I know, but it's not my call... :(	19:54
fossrox	Nevik: thank you very much!	19:54
	→ pantsman joined	19:54
Nevik	fossrox: sure, no problem ;)	19:54
	← pantsman left	19:54
	→ pantsman joined	19:55
Cathy	well people can just avoid committing their changes through any number of methods	19:55
	they're all going to require people not to make a mistake though	19:55
fossrox	:)	19:55
	← Sendoushi left	19:55
	← leeN left	19:55
Cathy	and you'll deal with erroneous commits to this file being pushed all the time	19:55
davividal	Cathy: I'm already dealing... :(	19:55
	Cathy: some guy just removed the files using git rm --cached and add them into gitignore	19:56
fossrox	!thanks	19:56
gitinfo	You're very welcome!	19:56
fossrox	:D	19:56
Cathy	davividal: sounds like you need to rename the file	19:56
	however, i can imagine one highly obfuscated alternative	19:56
	you could implement a post-receive hook on the central repository that examines all the changesets being pushed for whether they touch this file	19:56
	and if so, shows a message saying the file should not normally be checked in	19:56
	and requires them to type something ot confirm they really want ot do it	19:56
moritz	you could patch the application to also look in app.local.config if it exists	19:57
	→ ojacobson joined	19:57
moritz	and so give developers a method to override settings locally without touching the "proper" config file	19:57
Cathy	having files in the repository which are intended to be locally modified is fraught with peril anyway	19:57
	← ruby\|angry_hamst left	19:58
Cathy	actually the guy who gitignored the file would have been fine so long as he doesn't push that commit	19:58
	he could set it up to rebase what he actually wants to push to remove that commit	19:58
	→ rafaqueque joined	19:59
Cathy	not a good approach though	19:59
	→ joki joined	19:59
	← joaoh82 left	20:00
	→ joaoh82 joined	20:00
	← angelsl left	20:00
	→ rednaks joined	20:02
davividal	well, new issue! Someone just pushed into master! I need to undo his commits. :( Could I undo, say, the last 3 commits and put those 3 commits into another branch?	20:03
	← ericy left	20:03
	→ St_Marx joined	20:04
Cathy	davividal: you can do that very easily	20:05
	but somebody else might push them back later, if they've already pulled them	20:05
	← jizanthapus left	20:05
Cathy	just fix it to be the way you would like in your local repository and then push both branches (with -f)	20:05
	→ veebers joined	20:05
Cathy	maybe include another commit to master	20:06
	that way somebody else trying to push those commits back will get an error	20:06
	→ bwreilly joined	20:06
Cathy	better act quickly to reduce problems	20:06
	davividal: also you probably don't want to get in the habit of doing such things. if it works for you, you can just revert the commits in a new commit	20:07
	there's no real risk of losing data when you do this kind of surgery, but it could confuse your team	20:07
	especially people who aren't that familiar with git	20:07
	like the poor guy who pushed these commits	20:08
	he'll have to strip them off his local repository in order to pull again	20:08
	← rednaks left	20:09
davividal	Cathy: I can strip his HEAD off, no worries :) (pun intended)	20:09
	→ karstensrage joined	20:09
Cathy	it might be warranted to send out an email about this operation	20:10
	or else somebody might not understand what is going on and do a push -f of their own	20:10
karstensrage	is git-scm.com down for everyone?	20:10
davividal	I know... we had to do this SVN -> git migration in a hurry this week and we're paying a high price for it :(	20:10
karstensrage	Sorry, no Host found	20:10
Cathy	well davividal this kind of surgeyr wouldn't exactly be clean in svn either	20:11
davividal	karstensrage: http://www.downforeveryoneorjustme.com/	20:11
	← ianmcorvidae left	20:11
	← St_Marx left	20:11
davividal	Cathy: we don't have SVN anymore :)	20:11
Cathy	in this case, that link is not that useful	20:11
	karstensrage: git-scm.com returns a number of A records	20:11
	some of the servers are down, some of htem aren't	20:11
	54.225.172.112 appears to be up	20:12
	→ St_Marx joined	20:12
Cathy	so you can see 54.225.172.112 git-scm.com to your /etc/hosts file	20:12
	*so you can add	20:12
	and then you should be able to use it	20:12
	→ rhombus joined	20:12
	← rhombus left	20:12
	→ \|rhombus\| joined	20:12
davividal	Cathy: what I mean is: we migrated from SVN this week without training. That's why I'm dealing with those problems	20:12
	→ cff_ joined	20:12
	← \|rhombus\| left	20:13
	→ hyperboreean joined	20:13
	→ rhombus joined	20:13
CareBear\	davividal : exciting times	20:15
	→ ereslibre joined	20:15
	→ pdtpatri1k joined	20:15
	← sohail left	20:15
	← pdtpatri1k left	20:16
	→ kloeri joined	20:16
	← hobodave left	20:17
	← Dougie2187 left	20:17
	→ gan joined	20:17
	→ Dougie2187 joined	20:17
	← speeder left	20:18
	→ takamichi joined	20:19
	← VictorCL left	20:19
	→ ZujkisNx joined	20:20
	→ ericy joined	20:20
	→ Sendoushi joined	20:21
	→ bartzy joined	20:22
	← bamj0rb left	20:23
	→ revyd joined	20:23
	← ecube left	20:24
	→ bamj0rb joined	20:24
	gitinfo set mode: +v	20:24
	→ ianmcorvidae joined	20:25
	← JustinMac left	20:25
Nikoli	hi, do git tests support only bash? because several tests fail with dash	20:28
	← Dougie2187 left	20:28
	→ kcj joined	20:28
	→ koda joined	20:28
	→ Watabou joined	20:29
offby1	I assume that the tests indeed demand a specific shell	20:29
	I also assume that there's a top-level makefile that ensures that the right shell gets used.	20:30
	← karavanjo_ left	20:30
offby1	I'd be further surprised if the tests didn't start with a "shebang line" that specifies the shell to use	20:30
ojacobson	It's entirely possible they specify sh but actually lean on bash features -- that's a regular failing for folks coming from systems where sh is bash (including redhat, and older debian)	20:30
offby1	in other words: why on Earth are you running bash scripts with some other shell?	20:30
ojacobson	specific examples would hlep	20:30
	→ m0ltar joined	20:30
	offby1 nods gravely	20:31
offby1	they sholy would	20:32
EugeneKay	All hail the glorious bash master race	20:32
	→ katterstrophe joined	20:34
	← ericy left	20:35
	→ aChatir joined	20:35
canton7	from what I've seen from the list, several people are very passionate about shell cross-compatibility, so I'd be surprised if basic oversights still lingered	20:35
	← aChatir left	20:36
	→ l0cust joined	20:36
	← kermit left	20:37
	← Skrypter left	20:38
	← paulweb515 left	20:39
	→ techkid6 joined	20:39
	→ pfFredd joined	20:40
	→ eletuchy_ joined	20:42
	→ tpollard joined	20:42
	← eletuchy left	20:42
	→ sohail joined	20:44
	← sohail left	20:44
	→ sohail joined	20:44
	← revyd left	20:44
	← codiez left	20:45
	→ shuaib joined	20:45
	→ simesy joined	20:46
	← LuxuryMode left	20:47
	← danecando left	20:48
	→ jhulten joined	20:49
	← katterstrophe left	20:50
	← simesy left	20:51
	→ ericy joined	20:51
	→ dxdt joined	20:52
	← Cromulent left	20:53
	← finspin left	20:54
	← Noldorin left	20:56
	← schribl left	20:56
	← St_Marx left	20:56
	← hoschi left	20:57
	→ St_Marx joined	20:57
techkid6	Is there a way to git rid of previous (pushed) commits and re-push them?	20:58
	← rubatdub left	20:58
moritz	only with force-push	20:59
	← eqyiel left	20:59
	→ eqyiel joined	20:59
	← damccull left	20:59
techkid6	force-push?	20:59
	← TheBrainWork left	21:00
	→ ecube joined	21:00
moritz	push --force	21:00
techkid6	ok	21:00
	← Gavrisim_ left	21:00
techkid6	thanks	21:00
	→ Gavrisimo joined	21:01
	← cilly left	21:03
	→ cillybabe joined	21:03
	← Gavrisimo left	21:04
techkid6	so do I just have to bring the head back to the commit before the (three) i messed up?	21:04
	then force a push?	21:04
	← gua_ould left	21:04
	← bobbobbins left	21:04
canton7	beware that you're !rewriting_public_history potentially	21:04
gitinfo	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	21:04
canton7	but yeah	21:04
	or git push -f origin HEAD^^^:master	21:04
techkid6	alright, thanks	21:05
	← stodan left	21:05
	cillybabe → cilly	21:05
	→ Goplat joined	21:07
	→ Noldorin joined	21:08
	→ LuxuryMode joined	21:08
	← sohail left	21:11
	← cilly left	21:15
	← TaylanUB left	21:16
	← lite_ left	21:16
	→ VictorCL joined	21:17
	→ lite_ joined	21:17
	← lite_ left	21:17
	← JohnBat26 left	21:17
	→ lite_ joined	21:17
	→ paulweb515 joined	21:19
	← telmich left	21:20
	← abelsiqueira left	21:20
	← kcj left	21:20
	← m0ltar left	21:21
	← davividal left	21:21
	→ CheckDavid joined	21:21
	← dhruvasagar left	21:21
	← lite_ left	21:21
	→ cfoellmann joined	21:22
	→ aChatir joined	21:22
	← ericy left	21:22
	→ abelsiqueira joined	21:22
	← SoftwareMechanic left	21:22
	→ simesy joined	21:23
	→ JustinMac joined	21:24
	→ srayner joined	21:25
	← joaoh82 left	21:25
	→ leeN joined	21:26
	← aChatir left	21:26
	← closures999 left	21:27
	← simesy left	21:28
	→ farmanimal joined	21:28
	← visy left	21:28
	← shuaib left	21:29
	← St_Marx left	21:31
	→ St_Marx joined	21:32
	← jeffszusz left	21:32
	→ setmeaway joined	21:32
	← leeN left	21:33
	← marthinal left	21:34
	→ ec\|ipse joined	21:36
ec\|ipse	im new to git and having some issues	21:37
	← petey left	21:37
ec\|ipse	so I did "git init test" to make a test repo	21:37
	im on windows btw	21:37
	← srayner left	21:37
	→ petey joined	21:37
	← St_Marx left	21:38
	→ St_Marx joined	21:38
	→ ericy joined	21:38
ec\|ipse	when i try to add a file to it, it says "fatal: pathspec 'test.html' did not match any files"	21:38
moritz	so, is there a test.html?	21:38
ec\|ipse	i dragged test.html into the test folder, and ran "git add test.html"	21:39
	← cfoellmann left	21:39
	→ lindenle joined	21:39
moritz	ec\|ipse: does a 'dir' or 'ls' command in the test/ directory show it?	21:39
	← jds left	21:40
ELLIOTTCABLE	ec\|ipse: oh gods	21:40
	your nick is the worst thing D:	21:40
ec\|ipse	moritz: yes it shows it	21:41
	hah	21:41
	is there some kind of git navigation im missing? I don't know how to "be" in the test repo, It says ~ (master)	21:42
	← petey left	21:42
ec\|ipse	before any of this I just typed git init and now I have a .git folder also, idk if that affects anything	21:42
	→ leeN joined	21:42
imMute	ec\|ipse: what does 'dir' or 'ls' tell you?	21:43
ec\|ipse	oh shit	21:44
	hah... ok. I realize my mistake. the git bash confused me because i didn't think of it as a terminal, just commands for git	21:44
	itd help if i were on linux. well anyways, i cd'd into the test repo and it worked	21:45
	Nazca__ → Nazca	21:45
	← Nazca left	21:45
	→ Nazca joined	21:45
	← St_Marx left	21:46
imMute	ec\|ipse: "git bash" brings bash to windows via cygwin. it still uses the shitty Command Prompt terminal emulator though.	21:46
	← lamont left	21:46
EugeneKay	cygwin supports a local SSH server, so you can putty into that	21:46
ec\|ipse	ok I see. Thanks!	21:47
	→ Gavrisimo joined	21:48
	→ acl_ joined	21:49
	→ Kayra joined	21:49
	← WaReZ left	21:50
	← matanya left	21:52
	← maroloccio left	21:52
	← jhulten left	21:53
	→ LNDN joined	21:53
	→ WaReZ joined	21:55
	→ quejajajajaja joined	21:55
	← abelsiqueira left	21:56
	→ simesy joined	21:56
	← VictorCL left	21:57
	→ unreal joined	21:57
	← dxdt left	22:00
	→ HiramHibbard joined	22:02
	← lindenle left	22:04
	← WaReZ left	22:04
	← ec\|ipse left	22:05
	← cff_ left	22:05
	→ madewokherd joined	22:05
	← nilg left	22:06
	← HiramHibbard left	22:06
	← kjellski left	22:06
	← User458764 left	22:06
	← alezost left	22:07
	→ Ardit joined	22:08
	→ matanya joined	22:11
	← King_Hual left	22:11
	← Jackneill left	22:13
	→ shruggar joined	22:13
	← polyzen left	22:13
	← sq-one left	22:15
	← irqq left	22:15
	→ noyb joined	22:15
	→ tmcmahon joined	22:15
	→ carbon joined	22:15
	→ joaoh82 joined	22:19
	→ snearch joined	22:19
	← atomx left	22:20
	→ DMA joined	22:20
DMA	So... git 1.8.5 is not even a day old	22:20
	But is it stable, I guess?	22:20
	← snearch left	22:21
	→ Su7 joined	22:22
Su7	Hi all !	22:23
gitinfo	Su7: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	22:23
Su7	!backup	22:23
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	22:23
Su7	Well, thanks for the backup stuff !	22:24
	← robertnix left	22:24
Su7	I need some help guys.	22:24
	I have a v2.0 tag that is ahead of master by several commits	22:24
	I'd like to move master to this tag (so it includes all commits between current master and v2.0)	22:25
	does anyone have a clue ?	22:25
	→ cagedwisdom joined	22:25
bamj0rb	Su7: The most straightforward way is to merge: git checkout master && git merge tags/v2.0	22:26
	← Kicer86 left	22:26
Su7	bamj0rb: that's what I tried in the first place, but I do have the dev branch that contains commits AFTER v2.0	22:27
	← slassh left	22:27
Su7	so I'll need to merge the merge (!) of v2.0 and master to dev and master ?	22:27
bamj0rb	Su7: It's up to you. You can rebase instead. But it's a bit odd to rebase master on a tag (depends on your workflow).	22:28
	Su7: I personally find it odd that you have a tag /after/ master, but again it all comes down to workflow.	22:29
Su7	bamj0rb: the idea is to write code on dev, then merge it to master, then tag	22:29
	but I forgot to merge to master for v2.0	22:29
	let me try the rebase thing	22:29
	← l0cust left	22:30
	← quejajajajaja left	22:31
	← Goplat left	22:33
	← simesy left	22:33
	qsuscs → qsusOFFcs	22:34
	→ mribeirodantas joined	22:35
	← richwestcoast left	22:35
	← adac left	22:35
bamj0rb	My workflow is still a work in progress, but I like to develop in a feature branch, and rebase onto master.	22:35
	I consider master the main "development" branch.	22:36
	→ lamont joined	22:37
	← mdiers__ left	22:37
	→ xobb joined	22:38
	← drjd left	22:38
	← eijk_ left	22:38
	← Gavrisimo left	22:38
	← rendar left	22:40
	← LuxuryMode left	22:41
	← Watabou left	22:42
Su7	bamj0rb: rebase worked like a charm. Thanks !	22:42
	← xobb left	22:42
	← Sendoushi left	22:44
	← jnavila left	22:44
	→ johnkevinmbasco joined	22:45
bamj0rb	Su7: Do note that editing published history would be bad. :)	22:45
	→ magu\|m joined	22:46
Su7	Why would I do that ? :(	22:46
bamj0rb	I am currently torn between making production/testing a branch or tag. Currently they are tags, but that makes it somewhat hard to manage for bug fixes, etc. Doable, but yeah.	22:46
	Su7: If master has already been pushed then it is "published". See !rewriting_public_history. You shouldn't do that, but you should understand what it means. :)	22:47
gitinfo	Su7: 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	22:47
	→ mdiers__ joined	22:48
	→ wrouesnel1 joined	22:48
Su7	Well, I'm the only one working on the repo at the moment so I don't really face this kind of issues :)	22:49
	← magu\|q left	22:49
	← ericy left	22:51
	→ flaguy48 joined	22:51
	→ jarr0dsz joined	22:51
	← looper left	22:52
	← Kayra left	22:52
	← endiruna left	22:53
	← fossrox left	22:54
	→ atomx joined	22:54
	→ schribl joined	22:55
	→ looper joined	22:55
	→ jhulten joined	22:55
	→ LuxuryMode joined	22:56
	← ngladitz left	22:56
	← DMA left	22:57
	→ grahamsavage_ joined	22:57
	← TomyLobo left	22:57
	← halu left	22:58
	→ petey joined	22:58
	← phile left	22:58
	→ VictorCL joined	22:59
	→ Kayra joined	22:59
	← LuxuryMode left	23:00
	← grahamsavage left	23:00
	→ oojacoboo joined	23:01
	← looper left	23:02
	→ Tacker joined	23:02
	← devbug left	23:02
	← fairuz left	23:04
	→ groglogic joined	23:06
	← groglogic left	23:07
	→ ericy joined	23:07
	← bamj0rb left	23:08
	→ HiramHibbard joined	23:09
	← HiramHibbard left	23:09
	→ HiramHibbard joined	23:10
	→ tvw joined	23:11
	← harobed_ left	23:12
	→ rainbyte joined	23:13
	← przemoc left	23:15
	→ abelsiqueira joined	23:16
	→ ehmatthes joined	23:16
	← HiramHibbard left	23:18
	→ xerxas joined	23:18
	→ richwestcoast joined	23:19
	← pantsman left	23:20
	← Kayra left	23:21
	← FerreusR left	23:24
	← ericy left	23:25
	→ Balliad joined	23:26
	→ kukks joined	23:26
	← kukks left	23:26
	→ kukks joined	23:26
	→ damccull joined	23:27
	→ harobed_ joined	23:28
	← harobed_ left	23:28
	← joaoh82 left	23:28
	← rhombus left	23:28
	→ harobed_ joined	23:28
	→ joaoh82 joined	23:28
	← joaoh82 left	23:29
	→ joaoh82 joined	23:29
	→ Watabou joined	23:29
	← jhulten left	23:30
	← harobed_ left	23:33
	← petey left	23:34
	→ petey joined	23:34
	← hs366 left	23:35
	→ dsdeiz joined	23:35
	← nbrosnahan left	23:35
	→ Goplat joined	23:36
	← Watabou left	23:37
	← mfulz left	23:39
	→ l0cust joined	23:40
	→ juliohm joined	23:41
	→ ericy joined	23:41
	→ el3ctron_ joined	23:43
	← mdiers__ left	23:43
	← VictorCL left	23:44
	→ trusktr joined	23:45
	← koda left	23:46
	← Zarthus left	23:47
	→ jhulten joined	23:48
	→ JBlazeCo_ joined	23:49
	← noyb left	23:50
	→ etcetera joined	23:50
	← etcetera left	23:50
	← nutron left	23:51
	← freestyl3r left	23:52
	→ simesy joined	23:53
	← leeN left	23:54
	→ mdiers__ joined	23:55
	← RDove left	23:56
	← raonyguimaraes left	23:56
	← mdiers left	23:56
	← tmcmahon left	23:58
	← valgrind left	23:59
	← tvw left	23:59
	← petey left	23:59

Logs Search ←Prev date Next date→

Channels Documentation