IRCloggy #git 2011-07-17

Provider of IRC logs since 2005.

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

	2011-07-17
aLinux	i dont have ~/.ssh folder, http://help.github.com/linux-set-up-git/	00:00
	cd ~/.ssh	00:00
	← caseymcg left	00:00
aLinux	what to do ?	00:00
hsb	mkdir .ssh :)	00:00
	← _iron left	00:01
	→ rideh joined	00:01
	← Xanax left	00:01
	→ Pathin joined	00:01
BinGOs	go to Step 3.	00:02
	← jutaro left	00:03
	← spearce left	00:03
aLinux	i don't have $ cp id_rsa* key_backupCopies	00:03
	id_rsa*	00:03
	not id_rsa* in .ssh	00:04
	← Corren left	00:04
	← chrisf left	00:04
	← Pathin left	00:04
BinGOs	cd ~/.ssh "If it says “No such file or directory“ skip to step 3. Otherwise continue to step 2."	00:05
	← thews left	00:05
	→ thews joined	00:07
	← Yuuhi left	00:07
	→ spearce joined	00:07
	← j0llyr0t2en left	00:07
	→ j0llyr0tten joined	00:07
	→ Pathin joined	00:07
	← Pathin left	00:08
	→ Pathin joined	00:08
	← leavittx left	00:09
	→ _pr0t0type_ joined	00:12
	← brian_g left	00:13
	← v4n_ left	00:14
Adys	I haven't seen any announcement or anyone talking about it, but for those still interested, (since a couple of days?) google code now supports Git.	00:14
	← j416 left	00:15
	← p3rror left	00:15
	← aLinux left	00:17
fp1	Where's 1.7.6 for Mac OS X?	00:19
offby1	aliceinwire: you could also try "ssh localhost true"; that will probably create the .ssh directory for you. The benefit of doing it that way is that the permissions will be correct.	00:19
	oops, that was meant for alinux	00:19
	→ joshsmith joined	00:20
	← nibalizer left	00:23
	→ phretor joined	00:23
	← cesario left	00:25
	← Dreamer3 left	00:26
	→ codatory joined	00:27
	← sword left	00:34
	← canton7 left	00:34
	→ leavittx joined	00:38
	→ snitko joined	00:38
	→ banisterfiend joined	00:42
	→ jsjgruber_ joined	00:42
	← airstrike left	00:46
	→ worstadmin joined	00:47
	← snitko left	00:47
	→ tty1 joined	00:47
worstadmin	If I want to start a new git repo based off an existing, is the best way to just git init and cp -R the old one?	00:47
	→ snitko joined	00:48
	← jsjgruber_ left	00:48
worstadmin	Ie, I need to form a branch but I dont have write access to the repo	00:49
	In essence it's a fork	00:50
	→ dwmw2_gone__ joined	00:50
	← dwmw2_gone_ left	00:50
	→ sword joined	00:52
bremner	worstadmin: just clone it?	00:53
	what is the problem exactly?	00:53
	← Pathin left	00:53
worstadmin	bremner, Im start a repo based off an existing repo that wont be commit changes upstream	00:54
bremner	so?	00:54
worstadmin	*starting	00:54
bremner	nothing says you have to push to repo you clone from	00:54
worstadmin	Er will when they push its going to the origin right?	00:55
	*well	00:55
	Im going to need people to clone and push	00:55
bremner	you can change the origin after you clone	00:55
worstadmin	Okay let me rephrase this: Theres a repo that I will be forking - and this new repo based off it will have different code. So when I clone the existing I now need to put it under new management /new devs/ etc	00:56
	Isnt the best way to do that to simply create a new git repo and copy in the old code?	00:56
	Since it will have new origin/new logs/etc?	00:56
	← rtyler left	00:57
worstadmin	I mean I would have to, even setting a new 'git remote add myorigin' when they clone that will they recieve 'myorigin' >	00:58
	→ p3rror joined	00:58
bremner	worstadmin: are you saying you want to discart the old history? why?	00:59
	→ psoo joined	00:59
bremner	s/discart/discard/ ?	00:59
	→ nibalizer joined	01:00
worstadmin	bremner, Im cloning a repo with no write access. I need people to be able to edit/track/push code based on that codebase.	01:00
bremner	worstadmin: I do this all the time. clone. then git remote set-url origin	01:01
worstadmin	And I point the new remote at a fresh git repo?	01:02
bremner	sure	01:02
	and git push --all	01:02
worstadmin	Sure or yes?	01:02
	Sorry but sure could mean like "well yeah but thats dumb" hah	01:02
bremner	yes	01:02
worstadmin	Ah okay, ty	01:02
	Thanks that worked perfectly	01:04
bremner	great.	01:04
	→ BlindOracle joined	01:04
	→ Pathin joined	01:06
	← rvsjoen left	01:08
	→ rvsjoen joined	01:08
	← Dave^\| left	01:08
	← snitko left	01:09
	→ snitko joined	01:10
sitaram	necromancer: I just woke up but I have to go out for an hour or two. My quick diagnosis is that the compile failed because you added a comma to the username (gitolite uses spaces to separate tokens)	01:12
	bremner: ^	01:12
	necromancer: so you ought to be able to fix that, re commit, and push. Should work (if "info" is still showing you "R W ...")	01:12
	← spearce left	01:13
	→ agilo joined	01:14
	← p3rror left	01:14
agilo	is it a security risk to develop from a server (and have a private key stored there)? If so, what's the common way to go about developping from a server?	01:14
sitaram	bremner: in latest git (not even tagged yet, but pushed to pu) dont-panic has been replaced by even more generic "gl-admin-push". This is because some guys made changes that even gl-dont-panic could not fix!	01:15
bremner	sitaram: users never cease to impress me ;)	01:17
	→ zamabe joined	01:17
	← gusnan left	01:17
	← jimjastajay left	01:18
lahwran	for someone who has never used git before, should I recommend git-cheetah or tortoisegit?	01:18
sitaram	me too!	01:18
	→ Davey_ joined	01:18
ProLoser\|Mobile	what's dont-panic?	01:18
bremner	lahwran: people into gui's seem to like git-extensions	01:18
ProLoser\|Mobile	lahwran: are you on win?	01:18
sitaram	ProLoser\|Mobile: gl-dont-panic; a program that comes with gitolite to help if you managed to lock yourself out	01:18
ProLoser\|Mobile	lahwran: turtoisegit ain't bad but i really liked learning the shell+git-gui combo	01:19
lahwran	I am at the moment, but not normally. I also am used to cli git.	01:19
	I'm saying, what should I recommend to a team of people whom I will be setting up with git	01:19
	→ cakehero joined	01:19
ProLoser\|Mobile	i taught a team of people using cli+gui	01:19
	up to you honestly	01:20
	i've tried lots of different solutions and i have mixed opinions and jump from one to another a lot	01:20
	→ ISF joined	01:20
lahwran	well, git-cheetah appears to be a separate implementation of what tortoisegit does	01:20
	← BlindOracle left	01:20
lahwran	git-cheetah is built into msysgit by the looks of it, but also appears to be less mature (I'm saying all this without having installed either yet)	01:21
ProLoser\|Mobile	i've never used cheetah	01:21
	i'm on a mac, there are a few solutions here like git tower	01:21
	and now github came out with a mac app too	01:21
lahwran	alright	01:22
ProLoser\|Mobile	but the gui's are all really slow	01:22
	i git gui to stage and commit changes, but i do everything else via cli	01:22
	namely because it's easier to review changes using the gui	01:22
	→ kerrick joined	01:22
lahwran	alright, I'll recommend tortoisegit then	01:22
	hopefully it won't turn out to be a horribly wrong choice	01:22
	← psoo left	01:22
codatory	I do complex commits with GitX on the Mac, everything else I use terminal commands and bash aliases / functions :-)	01:22
bremner	lahwran: any reason you are not considering git extensions?	01:23
lahwran	git extensions?	01:23
codatory	I usually recommend git-gui to people because then they don't use it as a crutch too long ;-)	01:23
lahwran	what is git-gui?	01:23
bremner	lahwran: it's another gui	01:23
	→ v4n_ joined	01:23
lahwran	is that the other option under "windows explorer integration" in the msysgit installer?	01:23
	← cakehero left	01:24
	← _pr0t0type_ left	01:25
	← phretor left	01:28
	← snitko left	01:30
	→ snitko joined	01:30
	→ caseymcg joined	01:32
	→ Corren joined	01:33
	→ andrewpthorp joined	01:38
	← Fudge left	01:39
	← snitko left	01:40
	→ Fudge joined	01:40
	→ dr_win joined	01:50
	← Anubisss left	01:51
	← caseymcg left	01:52
	← Vampire0 left	01:54
	→ tuv joined	01:55
	→ Dreamer3 joined	01:55
	← tuv left	01:55
	→ snitko joined	01:56
	→ glennpratt joined	01:57
	← andrewpthorp left	01:59
	→ fbass joined	01:59
	→ spearce joined	01:59
	← chrijon left	02:01
	→ Danny_Joris joined	02:02
fowlmouth	getting this error trying to clone	02:04
	error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed while accessing https://github.com/Mon-Ouie/ray.git/info/refs	02:04
	→ Vampire0 joined	02:07
	← kennethreitz left	02:08
Danny_Joris	hi all, I'm wondering: is it possible to hotlink to github projects? Kind of what google does for jquery for example? There is this open source project CodeMirror, I'd like to see hosted somewhere (preferably on google). Would this be possible with github as well? https://github.com/marijnh/CodeMirror2 https://raw.github.com/marijnh/CodeMirror2/master/lib/codemirror.js	02:10
	← Chillance left	02:12
	← fbass left	02:12
	→ fbass joined	02:13
	← lunks left	02:14
	← sadmac left	02:14
	← snitko left	02:15
	→ snitko joined	02:15
	→ sadmac joined	02:15
	← ProLoser\|Mobile left	02:16
	→ DrDogwelder joined	02:17
	← Danny_Joris left	02:17
	→ Scott0 joined	02:18
	→ hyperair joined	02:19
	→ caseymcg joined	02:19
	← v4n_ left	02:20
	→ angelsl joined	02:22
	→ v4n_ joined	02:23
	← caseymcg left	02:23
	← hyperair left	02:25
	→ nikhgupta joined	02:29
	← v4n_ left	02:30
	→ caseymcg joined	02:30
	→ v4n_ joined	02:30
	← leavittx left	02:31
	← akosikeno left	02:32
	→ BiggFREE joined	02:32
	← phaedrix left	02:33
	← BiggFREE left	02:34
	← TomyLobo left	02:34
	→ BiggFREE joined	02:34
	→ hyperair joined	02:34
	← spearce left	02:34
	→ chrisf joined	02:35
	→ phaedrix joined	02:36
	← guns left	02:38
	← snitko left	02:40
	→ bblfish_ joined	02:40
	→ snitko joined	02:40
	← fp1 left	02:41
	→ Xanax joined	02:41
	← bblfish left	02:42
	bblfish_ → bblfish	02:42
	← BiggFREE left	02:44
	→ cherro joined	02:48
	→ macabre joined	02:50
	← towski left	02:54
	→ arunce joined	02:54
	→ shiba_yu36 joined	02:55
	← Xanax left	02:56
	← DrDogwelder left	02:57
	← msieradzki left	02:58
	← shiba_yu36 left	02:59
	→ shiba_yu36 joined	03:01
	← snitko left	03:02
	→ snitko joined	03:03
	← zamabe left	03:06
	← cherro left	03:07
	← pheaver left	03:12
	→ tomas_w joined	03:13
	← Fudge left	03:20
	→ pheaver joined	03:20
	← pheaver left	03:20
	→ Fudge joined	03:22
	← snitko left	03:22
	→ n8o-mba joined	03:23
	→ muneeb joined	03:26
	← muneeb left	03:26
	← ivan left	03:26
	→ ivan joined	03:26
	→ muneeb joined	03:26
	← Corren left	03:27
	← kermit left	03:31
	← v4n_ left	03:32
	→ kermit joined	03:36
	← kukks left	03:36
	→ snitko joined	03:38
	← snitko left	03:42
	→ snitko joined	03:43
lahwran	do passphrases work in tortoisegit with openssh?	03:44
	→ MetaMucil joined	03:45
	→ krz joined	03:48
	→ djbpython joined	03:50
	← EricInBNE left	03:52
	→ anzime joined	03:54
	→ EricInBNE joined	03:54
AAA_awright	lahwran: They do iirc, I haven't used it in a while	03:55
	→ zamabe joined	03:56
	→ rchavik joined	03:57
	← snitko left	04:02
	→ snitko joined	04:02
	← glennpratt left	04:03
	← shedammit left	04:05
	→ Corren joined	04:06
	← Chibby_ left	04:08
	→ Chibby_ joined	04:09
	← Pathin left	04:09
	← sechrist left	04:10
	→ towski joined	04:10
	← rideh left	04:10
	← n8o-mba left	04:11
	→ Pathin joined	04:12
	← n8o left	04:12
	→ n8o joined	04:13
	→ rideh joined	04:15
	← rchavik left	04:18
	→ TML joined	04:21
	← Corren left	04:21
	→ mikew3456 joined	04:22
	← Chibby_ left	04:22
	← nikhgupta left	04:23
	→ Chib joined	04:23
	ChanServ set mode: +v	04:23
mikew3456	how can i check the diff's of all files between two commits	04:23
	← mDraze left	04:24
	→ cytrinox_ joined	04:26
	→ shiba_yu_ joined	04:27
	← shiba_yu36 left	04:27
	→ rchavik joined	04:28
	→ DarkHouse joined	04:29
	→ glennpratt joined	04:29
	← agilo left	04:29
	← cytrinox left	04:29
	cytrinox_ → cytrinox	04:29
	← seivan left	04:30
	→ quentusrex joined	04:30
	← n8o left	04:31
	→ n8o_ joined	04:31
	← diegoviola left	04:32
codatory	mikew3456: git diff commit1..commit2	04:34
	→ sechrist joined	04:34
mikew3456	thx	04:35
	→ diegoviola joined	04:37
	← mikew3456 left	04:40
	← snitko left	04:42
	→ snitko joined	04:43
	← Chib left	04:44
	→ andrewpthorp joined	04:45
	→ Chib joined	04:45
	← codatory left	04:46
	→ seivan joined	04:47
	→ mandric joined	04:49
	→ bitri joined	04:50
	→ akosikeno joined	04:53
	← madewokherd left	04:53
	→ BlindOracle joined	04:54
	← alberto56 left	04:55
	← glennpratt left	04:56
	← Bass10 left	04:58
	→ nikhgupta joined	04:59
	← Dreamer3 left	05:01
	← snitko left	05:02
	→ Dreamer3 joined	05:02
	← Davey_ left	05:04
	→ snitko joined	05:04
	→ accel joined	05:05
accel	is it possible to store a git repo in "chunks?" for example, suppose I am encryting my repo + backing it up to a remote server. Instead of encrypting the entire repo historye very time, I want to do thigns like: "encrypt commits 1-100; backup; encrypt commits 101-200; backup; encrypt commits 201-300; backup; ... " i.e. access the git repo in chunks ... and then, somehow, magically ahve a way to assemble the chunks back together	05:06
	→ max_ joined	05:06
	← max_ left	05:07
	→ Synthead joined	05:08
Synthead	is there a way I can host a folder from a git repo via apache (httpd) for development purposes?	05:08
	→ dr0id joined	05:08
	← diegoviola left	05:08
	→ aLinux joined	05:09
aLinux	hi	05:10
	← thiago left	05:12
aLinux	how can i remove a repository ?	05:12
	https://github.com/repositories/new	05:13
	← Kreative` left	05:13
	← arooni left	05:13
	← ilteris left	05:15
	→ ilteris joined	05:15
	← n8o_ left	05:15
	→ n8o joined	05:16
aLinux	after forking a project from github i get the original contributor name at top	05:19
	how can i remove/undo forking that project ?	05:19
	→ Prasad joined	05:21
	→ sou_ joined	05:21
	← rideh left	05:22
	→ n8o_ joined	05:23
	← bitri left	05:23
	→ diegoviola joined	05:23
	← n8o_ left	05:23
	→ n8o_ joined	05:24
	← n8o left	05:24
	← n8o_ left	05:24
	→ n8o joined	05:24
	← sou_ left	05:31
aLinux	how can i check the list of repository i have made ?	05:31
	→ leavittx joined	05:32
	→ avinashhm joined	05:33
	→ dfr\|mac joined	05:35
	← aLinux left	05:41
	→ sou_ joined	05:45
	→ SimonNa joined	05:47
	→ sonnym joined	05:48
	← caseymcg left	05:49
	→ Saumil joined	05:49
	← Chib left	05:51
	→ Chibby_ joined	05:51
	← accel left	05:52
	→ cccaldas_ joined	05:52
	← bgerber left	05:52
	cccaldas_ → cccaldas	05:53
	← cccaldas left	05:53
	→ cccaldas joined	05:54
	kjdavies_ → kjdavies	05:55
	← Prasad left	05:58
	← MetaMucil left	06:02
	→ pheaver joined	06:02
	→ bluj joined	06:04
	→ rvsjoen_ joined	06:06
	→ aminpy joined	06:08
	← rvsjoen left	06:08
	→ jutaro joined	06:09
	→ preflex_ joined	06:11
	← Fullmoon left	06:11
	← Synthead left	06:12
	← preflex left	06:12
	→ beautifulmind joined	06:12
	← n8o left	06:13
	→ Fullmoon joined	06:13
necromancer	sitaram: i was able to fix it, thanks!	06:18
	← anzime left	06:19
	← cccaldas left	06:19
	← Fullmoon left	06:20
	← zivester left	06:20
	← mandric left	06:21
	→ fp1 joined	06:27
	← catsup left	06:28
	→ catsup joined	06:28
	← d0k left	06:28
	← nikhgupta left	06:31
	→ deepthinker82 joined	06:35
	← optixx left	06:36
	→ optixx joined	06:37
	← jutaro left	06:37
	→ cccaldas joined	06:39
	← Saumil left	06:39
	→ bn_ joined	06:40
	← MikhX left	06:40
	← dfr\|mac left	06:40
	← chrisf\|laptop left	06:42
	→ NetRoY joined	06:43
	← MavJS left	06:44
	→ jutaro joined	06:45
	→ anzime joined	06:46
	← anzime left	06:47
	→ kriegerod joined	06:49
	→ JigABoot joined	06:49
JigABoot	I spent three nights on Neverland Ranch when I was 13! Michael Jackson and I had so much fun cuddling in bed, climing trees, and riding gokarts. One day when we were climbing a tree, he yanked down my shorts and underoos! Then he proceeded to tap his hard 8 inch cock on my cheeks, spit in my crack, and slowly slip it in. It hurt at first, but then I got this raging woody!	06:49
	← jutaro left	06:50
	← seivan left	06:52
fowlmouth	lol	06:53
	→ suzzy joined	06:53
	→ airborn joined	06:54
bn_	hi. I'm using git v1.7.4.4 and have created a remote tracking branch A off master, pulled it down, made some some commits & pushes, then made a new local branch A' off A to test if the buildmaster's updates to master would merge in cleanly to A. After safely merging in updates from master to A', I switched back to A and merged that merge in. Several commits & pushes later in A, I switch back to A' to attempt to merge in the new changes I made	06:55
	→ MavJS joined	06:56
	→ chrisf\|laptop joined	06:56
	← beautifulmind left	06:57
	→ airtonix_ joined	06:58
	← airtonix left	06:59
bn_	note, FWIW, master is also a remote tracking branch	07:00
	← snitko left	07:01
	→ lunks joined	07:04
lunks	hey, any operators around?	07:04
	→ lstarnes joined	07:05
	→ tewecske joined	07:05
	→ mikeym joined	07:05
	fowlmouth → linux-cry-babies	07:06
	← lstarnes left	07:06
	← linux-cry-babies left	07:06
	→ mst joined	07:06
	ChanServ set mode: +o	07:06
	mst set mode: +b	07:06
	mst kicked suzzy (suzzy)	07:06
	mst set mode: -o	07:06
	← pheaver left	07:06
lunks	mst: thank you	07:06
mst	lunks: do make sure to explain to the ops what happened and why when they turn up so nobody comes and yells at me :)	07:07
lunks	mst: I surely will. :)	07:07
	→ jutaro joined	07:07
mst	good. because otherwise I may cry. and I'll bill you for the kleenex :)	07:07
	have fun.	07:07
	← mst left	07:07
	← sechrist left	07:07
	← NetRoY left	07:07
bn_	took a look at the man page for "git merge" too, but it also didn't seem to help :/	07:08
	→ seivan joined	07:09
	← drev1 left	07:12
	← jutaro left	07:13
	→ scyx joined	07:13
	← seivan left	07:16
bn_	anyone?	07:16
	← maestrojed left	07:16
	← JigABoot left	07:16
	→ maestrojed joined	07:17
	→ cesario joined	07:18
	← f0i left	07:19
	→ drev1 joined	07:20
bn_	ack! figured out my mistake, disregard my question	07:20
	→ haydenmuhl joined	07:22
	← cccaldas left	07:22
	← mikeym left	07:23
	← drev1 left	07:24
	→ drizzd joined	07:25
	← airborn left	07:28
	→ f0i joined	07:29
	← fbass left	07:30
	← iXeno left	07:33
	→ mikeym joined	07:37
	← mikeym left	07:37
	→ mikeym joined	07:37
	← mikeym left	07:37
	→ iXeno joined	07:38
	← flexd left	07:39
	→ flexd joined	07:39
	→ brainbox1100 joined	07:44
	← dr0id left	07:45
brainbox1100	I just did a pull and have an unmerged path "both modified: ProjectDetailView.xib" is there a quick way to use my version and throw away the one that was pulled? Since I know mine is right?	07:46
	→ seivan joined	07:46
	← seivan left	07:48
	→ Phoul joined	07:48
	← fp1 left	07:50
	→ razvand joined	07:52
	→ vmil86 joined	07:53
razvand	Hi! I have a local Git repository clone that I want updated each time I do a commit from a remote clone. I used a post-receive hook but it doesn't work. Is there a proper way/hook to handle this?	07:55
	← Adaptee left	07:57
	→ dwmw2_gone___ joined	07:57
	← jast left	07:58
	← dwmw2_gone__ left	07:58
	→ seivan joined	08:00
	→ Adaptee joined	08:00
	→ rfx joined	08:01
	← joshsmith left	08:04
bn_	brainbox1100: I think there may be a shortcut mentioned in the man page for git merge. In general though you should avoid doing a merge if you have local changes. If you do, you should stash or commit first...	08:05
	hth, night!	08:05
	← bn_ left	08:05
	→ kane77 joined	08:05
	→ jutaro joined	08:06
	← seivan left	08:09
	← muneeb left	08:10
	← towski left	08:11
	→ phretor joined	08:11
	→ jast joined	08:14
	→ paissad joined	08:14
	→ bblfish_ joined	08:15
	→ seivan joined	08:15
paissad	hello guys, i would like to see the differences of contents between the commit & the HEAD in the remote repository	08:15
	is it possible ?	08:15
	→ airborn joined	08:16
	← Phoul left	08:17
	← bblfish left	08:17
	bblfish_ → bblfish	08:17
	→ bgerber joined	08:20
	→ delian66 joined	08:21
	→ muneeb joined	08:24
rfx	When I use git status I see a file being marked modified, but git diff doesn't show me anything. How can I find out what git means exactly with modified?	08:25
	→ namemon joined	08:27
haydenmuhl	rfx: The changes are probably in the index. Does it also say "Changes to be committed:"?	08:28
	← bauruine left	08:30
	← bgerber left	08:30
	→ bauruine joined	08:31
rfx	haydenmuhl: yes. I've ran git add --all because I swapped a complete source tree	08:32
	← DarkHouse left	08:32
haydenmuhl	rfx: You want "git diff --cached"	08:33
	That shows you what is in the index, and what will be committed if you run git commit immediately.	08:34
	→ bgerber joined	08:34
	→ ProLoser\|Mobile joined	08:35
rfx	oh god, I just hit me from above what you mean. Sorry, not enough coffee, of course :)	08:37
	← seivan left	08:37
rfx	s/I/it/	08:37
	→ hachque joined	08:37
	← tansell-laptop left	08:38
rfx	Is it possible to have git ignore whitespace changes completely? I've to sync two big trees with thousands of files. git diff -b spares me the whitespaces. But git add doesn't know about it. I.e. I don't want to add files whose content only differ in whitespace, is that possible?	08:40
	I.e. when git diff -b does not show anything, don't add the file to the index?	08:42
	→ caseymcg joined	08:42
haydenmuhl	I don't see anything in the git add man page. You could git diff --name-only to get the list you want, and only add those.	08:44
	You might be able to pipe the output of git diff into git add...	08:44
	← muneeb left	08:45
	← jfkw left	08:45
	→ muneeb joined	08:46
	→ cbx33 joined	08:47
haydenmuhl	Nevermind. That doesn't seem to work.	08:47
cbx33	is git gui included in the msysgit package?	08:47
drizzd	yes	08:47
	→ j416 joined	08:48
cbx33	sweet	08:48
	thanks	08:48
	Mac OS too?	08:48
	hey j416	08:48
j416	o/	08:48
drizzd	cbx33: I would be surprised if it was not	08:48
rfx	haydenmuhl: I see, thanks for your hint. I'll try it in bash itself then.	08:49
cbx33	actually j416 should be able to answer	08:49
drizzd	were not	08:49
j416	?	08:49
cbx33	is git gui provided in MacOS git?	08:49
j416	yes	08:49
	well	08:49
	I'm not sure what you mean by mac os git	08:49
	is there a difference?	08:49
haydenmuhl	rfx: I tried it out, and the -b and --name-only flags don't play well together.	08:49
cbx33	ok	08:49
	how do you get Git for MacOS how is it packaged?	08:50
haydenmuhl	rfx: --name-only will always give you the names of files which only differ by white space.	08:50
j416	I'm not sure if the bundle they make for download on the site is any different	08:50
rfx	haydenmuhl: yes, I tried it to. I meant I'll do "for each file call diff -b" and based on the result git add it or not ...	08:50
j416	I compile it from source.	08:50
cbx33	is that the only choice for MacOS users?	08:50
j416	cbx33: 10:50 j416: I'm not sure if the bundle they make for download on the site is any different	08:50
haydenmuhl	rfx: Ah, I see. That's probably a good idea.	08:50
j416	s/make/make available/	08:50
cbx33	ahh	08:51
	← airborn left	08:52
rfx	haydenmuhl: unfortunately that status code of git diff -b also doesn't change :-( So I really have to look at the output off it.	08:54
haydenmuhl	rfx: Not difficult. Just check for zero length output.	08:55
	← caseymcg left	08:55
scyx	can anyone help me out with understanding something about remote branches? i've added a few remote branches to a git repo, but whenever i go git clone on it, the new repo doesn't have those remote branches. am i doing something wrong?	08:55
rfx	haydenmuhl: true. Came up with this now; slow as hell but seems to work: git status -s\|grep '^ M'\|sed 's/^ M //'\|while read file; do test -z "$(git diff -b $file)" ; if [ "$?" != "0" ]; then echo "git add $file"; fi; done	08:56
	Ignoring things like problematic characters and whitespace in filenames for now ...	08:56
	← [intra]lanman left	08:57
haydenmuhl	I'm no bash guru, but I find quoting all your variables to be helpful.	08:57
	→ tore_ joined	08:58
j416	bash trims whitespace and splits values up into words unless you quote when you assign.	08:58
	s/values/the value/	08:58
rfx	So in my case this should probably be sufficient: test -z "$(git diff -b "$file")"	08:58
mjt	that's true for any shell (incl. csh and derivates)	08:59
j416	in that case you have to escape the inner "s	08:59
	→ munichlinux joined	08:59
rfx	zomg. git is so great. I figured it automatically (!) quotes filenames with special chars and such. wow	09:00
	never saw any other tool doing that so far	09:00
mjt	rfx: for this very test - verifying that there's no changes - you may use git diff --exit-code	09:00
	← munichlinux left	09:01
j416	rfx: another way to go is commit everything, diff that, then amend without the whitespace differences	09:01
mjt	and write the test like if git diff --exit-code ... ; then echo changed; else echo unchanged; fi	09:01
	→ munichlinux joined	09:01
rfx	mjt: wow, very cool	09:01
j416	mjt: nice :)	09:02
mjt	speaking of quoting filenames -- ls(1) does that too for several decades.	09:02
j416	thanks	09:02
rfx	mjt: but not by default? when I "ls" over a dir with a file name with a space in it, it does nothing special afaics.	09:04
	$ ls\|grep ' ' # => foo bar ; without quotes or anything	09:04
	← optixx left	09:04
j416	rfx: so, what is it that you want to achive? You want to add only lines that were actually changed?	09:05
	(avoiding whitespace differences?)	09:05
	→ optixx joined	09:05
	→ juangiordana joined	09:05
	← cbx33 left	09:06
mjt	rfx: for ls it's historical, so that all ls implementations work the same for basic operations. ls -b, ls -Q or ls --quoting-style= (from memory, manpage should have more details)	09:06
	→ SimonN joined	09:07
rfx	mjt: I see. yeah well, my real point actually was that it works expected usable out of the box. No one thinks (ok, bold statement, but...) about adding this (as default) options to ls, imho	09:07
	→ nmemon joined	09:07
	← SimonN left	09:07
	← SimonNa left	09:07
j416	rfx: what is the task?	09:08
mjt	i think diff does some quoting by default (in diff -r)	09:08
	→ SimonNa joined	09:08
	← juangiordana left	09:09
	← namemon left	09:11
	nmemon → namemon	09:11
	← resmo_ left	09:11
	→ airborn joined	09:12
	← muneeb left	09:12
	→ juangiordana joined	09:14
	← mjessome left	09:14
	j416 gives up solving rfx's problem	09:15
	→ dr0id joined	09:17
	→ Pupeno joined	09:21
	← juangiordana left	09:23
	→ juangiordana joined	09:25
	← Amorphous left	09:27
	→ brian_g joined	09:27
	→ seivan joined	09:29
	← Nuck left	09:30
	← kerrick left	09:30
	→ feefree joined	09:32
	→ chrischambers joined	09:33
	← feefree left	09:38
	← diegoviola left	09:39
rfx	j416: I thought I had solved the problem, sorry was away tinkering with it	09:40
j416	rfx: so what is the problem?	09:40
	← haydenmuhl left	09:40
rfx	but I realized I still have troubles with files with problematic characters in it. Takes for example: git status -s\|grep '^??'\|grep /fm # => ?? "http/images_neu/startpage/fm\t.jpg" . Shows a file with a, yes, a tab in it. I thought I can use output of git status to simple feed it to git add if I want to, but it doesn't work: git add "http/images_neu/startpage/fm\t.jpg" gives: fatal: pathspec	09:42
	'http/images_neu/startpage/fm\t.jpg' did not match any files	09:42
j416	..I still don't know what you want to achieve	09:42
	→ Amorphous joined	09:43
rfx	I go through the output of git status currently, apply a pattern with grep on it and want to add these files to the index. But I can't get it working for files with special characters, because the file name produces by git status is not accepted for git add	09:43
j416	no	09:44
	I want to know what you want to achive	09:44
	not what you are failing to do	09:45
	end result	09:45
	← angelsl left	09:45
rfx	only add files matching a certain pattern to the index	09:45
j416	and why do you want to do this?	09:46
rfx	(the files are not known yet at all)	09:46
	need them commited to the branch, so I can later diff two branches	09:46
j416	what is the certain pattern?	09:46
rfx	j416: good question. now that you ask: all untracked files.	09:47
	→ angelsl joined	09:47
j416	rfx: git add . && git commit	09:47
	=	09:47
	?	09:47
	adds everything	09:47
rfx	forget that pattern, I mis-worded that completely because I was so focused on git status output	09:47
	Only untracked files, not everything. I've other files I don't want to add.	09:48
j416	I thought you were saying something about whitespace at first	09:48
rfx	that has been solved already	09:48
	I just hit the next riddle	09:48
j416	ok..	09:48
BinGOs	isn't that what .gitignore is for	09:48
rfx	j416: I apologize for not making that clear, I see now. sorry	09:49
j416	rfx: if your solution was what you had above, it's not very elegant :)	09:49
	but you seem to be happy with it	09:49
	about untracked, I dunno	09:49
rfx	j416: I'd be interested to know a more elegant solution in case you want to.	09:52
j416	rfx: well I don't know the problem, so.	09:53
rfx	j416: I thought you referred to the whitespace problem above when you said "not very elegant"?	09:53
j416	but n/m, if you are using that and it works, why change it :)	09:53
	rfx: indeed	09:53
rfx	I'm confused now	09:53
	→ feefree joined	09:53
j416	whitespace problem	09:53
	I don't know what you wanted to achieve	09:53
rfx	j416: from all the modified files, excludes files from being added to the index when their whole content only differs in whitespace (i.e. git diff -b has no output)	09:54
j416	(and for the other problem, I assume you are scripting this? otherwise you can use 'git add -i' and get a nice list of untracked files, and add them)	09:55
BinGOs	o.O	09:55
	← feefree left	09:56
j416	rfx: but you want to add whitespace diffs if they are in the same file, if there is only a single unrelated change in that file?	09:56
rfx	j416: sorry, parse error .. could you rephrase that? But if you're asking about the same file: yes, only for a file itself.	09:57
j416	i.e. you only removed whitespace from line 1, but you actually changed the real content of line 5	09:58
	do you still want to add the whitespace-only change in line 1?	09:58
rfx	j416: you asking because of hunks? That doesn't matter in that case. If any real content of a file changes, just add the whole file; no concern about individual hunks.	09:59
j416	rfx: is that the desired result, or is it just because you think it's easier?	10:00
rfx	j416: desired	10:00
j416	hm ok	10:00
	→ nenopera joined	10:00
j416	my solution would be adding only actual changed lines	10:01
	sorry for taking your time. :)	10:01
	← TTilus left	10:01
rfx	oh	10:01
	← seivan left	10:01
	← rchavik left	10:02
	← bla left	10:02
	→ bla joined	10:02
rfx	well .. you're question missed an option "don't care" would actual. I didn't though in terms of "easier", just I didn't care really. So I think I see where you're heading and that would be certainly interesting how to do this.	10:02
	← leavittx left	10:02
j416	ok	10:02
	→ seivan joined	10:03
j416	so I guess you want the effect of having all extraneous whitespace removed first, then commit the result, without actually removing the whitespace, just finding the changed content?	10:04
rfx	That sounds ok ... (except that "removed first" sounds dangerous, i.e. modifying the files before commiting...)	10:05
	ChanServ set mode: +v	10:05
j416	'effect of'	10:05
rfx	ack	10:05
j416	sounds ok or sounds like what you want?	10:05
	sigh	10:06
rfx	err, both. sounds ok what I want	10:06
nenopera	hi everyone	10:06
	I'm using https://github.com/wmanley/git-meld.git to make branch directory comparisons and found something extrange	10:06
	sorry about my english	10:07
	I clone git-meld repo in ~/bin , to keep tracks on changes to the project, but found that git think git-meld directory is an alias (but It is a directory)	10:08
	← Rejinka left	10:08
ioerror__	nenopera: I guess ~/bin is in your PATH and your directory is called git-meld. that's expected..	10:09
	→ Rejinka joined	10:09
nenopera	but it is a directory, git must not try to execute it	10:10
ioerror__	nenopera: ~/bin is usually not the right place to put source code	10:10
	→ nocturnal joined	10:10
nenopera	ioerror: I know	10:10
ioerror__	nenopera: if it's in your path it's a place to put executables	10:10
j416	rfx: I did something similar a while ago for someone here, that's why	10:10
	found it, but there should be a more elegant way of doing it still: http://pastie.org/2173001	10:11
ioerror__	nenopera: why do you want it to sit there?	10:11
j416	rfx: slightly different problem though, he had already committed everything and wanted to split it	10:11
nenopera	Ioerror: to keep track on changes easily. I use git config --global alias.meld \!~/bin/git-meld/git-melp.pl	10:13
	→ kumbayo joined	10:13
	→ TTilus joined	10:14
nenopera	ioerror: my poor man solution is to call git-meld meld-git and it works, but I think git shouldn't consider git-meld directory a executable	10:14
rfx	j416: interesting solution ... basically you use the output of diff -b, pipe into apply and thus the whitespacs changes are ignored	10:14
ioerror__	nenopera: what you should rather do: put it in ~/src/meld for exampled and softlink ~/bin/meld to ~/src/meld/meld.bin whatever, then configure meld as git merge tool.	10:14
j416	rfx: yeah	10:14
	rfx: not ignored, but revese-applied	10:15
	reverse*	10:15
ioerror__	nenopera: it must because this is how the whole external command thing works	10:15
rfx	j416: thanks for that, I'll see how I can replace it with my current solution to see them compared	10:15
	→ kmox83 joined	10:15
	→ Chib joined	10:16
	← Chib left	10:17
ioerror__	nenopera: see man git-config merge.tool	10:17
jast	nenopera: automatic message the 'git-config' manpage can be found at http://www.kernel.org/pub/software/scm/git/docs/git-config.html	10:17
rfx	on to the next problem :) back to the "only add untracked files" problem ...	10:17
	→ Chib joined	10:17
	→ Yuuhi joined	10:17
j416	rfx: I was about to make an example for you, but since I don't know exactly what you have and how you want to apply it, it's a bit useless.	10:17
	rfx: so I'll not do that	10:17
	rfx: I gave you a possible solution to the add untracked files	10:18
	was it not sufficient?	10:18
rfx	j416: perfectly fine, I think the basic concept is important (piping diff/apply) and I'll toy around with it	10:18
j416	I assume you need to do this in a script then?	10:18
rfx	j416: wait, you did? mom	10:18
j416	rfx: you should be able to diff it without committing even, if that was what I did in the example	10:18
	(diff the index and modify it)	10:19
	("git apply --cached" I think)	10:20
	→ phretor_ joined	10:20
	← phretor left	10:20
rfx	j416: about the untracked file issue: you suggested git add . , but that will add all changes, I only want to add the untracked files	10:20
	← kmox83 left	10:20
	← Rejinka left	10:20
nenopera	ioerror: I will use ln -s for sure. But Has it sense to execute a directory? bash says when doing ~/bin/git-meld (git-meld: is a directory) git says (fatal: cannot exec 'git-meld': Permission denied) The anoying part is I think the permission denied was for my declared alias	10:20
j416	rfx: I suggested 'git add -i'	10:20
rfx	j416: my bad. there was a problem with it, let me scroll back	10:21
	→ Rejinka joined	10:21
nenopera	ioerror: I know is up to me to change git code, but think that git can verify if some file is a directory, prior to execute it	10:21
rfx	j416: got it working now. I didn't realize how to use it, after I choose "a" (for add untracked) that I was required to select all files with "*". seems I have that now.	10:23
	← quentusrex left	10:23
j416	rfx: and by the way,	10:23
	rfx: git ls-files -o \| git update-index --add --stdin	10:23
	should do what you want without using git add -i. :)	10:24
	looks good and should be a stable way of doing it; quoting: http://kerneltrap.org/mailarchive/git/2007/8/5/254068	10:24
rfx	impressive!	10:26
j416	thank google	10:26
rfx	thank you :)	10:26
j416	http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=git+add+only+untracked+files < third hit :)	10:26
	← nenopera left	10:27
rfx	I had a similar search with different words and didn't have this result unfortunately	10:27
j416	gotta use the right words... :P	10:27
	← shiba_yu_ left	10:30
	→ kmox83 joined	10:31
	preflex_ → preflex	10:32
	← workmad3 left	10:34
	← seivan left	10:35
	← kmox83 left	10:36
	← namemon left	10:36
	← sebastorama left	10:38
ioerror__	nenopera: directories in PATH is just not the place to put non-executables. put it somewhere else and link the binary.	10:39
rfx	the ls-files -o solution didn't work in my case because I've other "other" files besides untracked I don't want to add to the index. Your git add -i solution is the working one	10:40
j416	rfx: so you didn't try it?	10:44
	rfx: adds only the untracked for me: http://pastie.org/2226203	10:45
	→ N0va` joined	10:45
	← Zariel left	10:46
rfx	j416: I did but it added also the other other files (I don't remember which state they where exactly where, I'm already past that point now)	10:48
j416	o_O	10:49
	→ seivan joined	10:50
	← razvand left	10:55
	→ berserkr joined	10:56
	→ Textmode joined	10:56
	← seivan left	10:56
rfx	j416: btw, I can't thank you enough for your patience and help!	10:58
	→ seivan joined	10:58
	→ mragab joined	10:59
	← Adys left	11:03
	← ProLoser\|Mobile left	11:04
	→ ProLoser\|Mobile joined	11:05
	← ProLoser\|Mobile left	11:06
	→ beautifulmind joined	11:08
	→ ikkio joined	11:10
	→ Agos joined	11:11
	← ikkio left	11:12
j416	rfx: :)	11:14
	rfx: hope things worked out well despite my failing to understand what you were after.	11:15
	← dermoth_ left	11:16
	→ viranch joined	11:17
	← kriegerod left	11:17
viranch	hi, i'm migrating all my mercurial repos to git. in one of them, i'm getting the error: "repository has at least one unnamed head: hg r25", any ideas?	11:17
j416	viranch: no, but I'm interested in the argument(s) you have for converting from mercurial to git, if you feel like sharing	11:19
viranch	j416: well, i felt mercurial to be very slow after i started using git	11:20
	→ Anubisss joined	11:20
j416	reason enough I suppose :)	11:20
viranch	also, the disk usage increases for a mercurial repo as code gets committed	11:21
	;)	11:21
	→ TomyLobo joined	11:21
j416	viranch: by the way, the size of the git repo also increases if you add things to it. :P	11:23
	→ Adys joined	11:23
tango_	who would have guess	11:23
viranch	j416: sure! one of the mercurial repo rose to 12mb after certain ~70 commits!	11:23
tango_	+ed	11:23
j416	(well, it compresses better and has better delta stuff, but still, it won't shrink.. :P)	11:23
viranch	so if someone clones the mercurial repo, he has to download freaking 12mbs!!	11:24
j416	viranch: that doesn't say much though, don't know what those ~70 commits contained :) But I guess your point is git can store things more efficiently. mhm.	11:24
viranch	which was rather very very small project	11:25
	j416: hmm yes :)	11:25
j416	iirc, mercurial stores deltas between revisions?	11:25
	I mean, that's how it compresses?	11:25
viranch	yes	11:25
j416	but it uses a similar object db to that of git, apart from the packing?	11:25
	in that it stores whole objects	11:25
	not the diffs	11:26
	right?	11:26
viranch	i guess, i've read that at least	11:26
j416	ok	11:26
	ty	11:26
	← jutaro left	11:26
viranch	;)	11:26
j416	viranch: I googled a bit about your problem btw, it seems people use 'hg strip' (?) to remove unnamed heads	11:26
	not confirmed	11:27
viranch	j416: oh alright, thanks for the heads up :)	11:27
	→ Murr_ joined	11:27
	→ rchavik joined	11:28
j416	viranch: https://lists.canonical.com/archives/bazaar/2008q2/041698.html	11:28
	I don't know what an 'unnamed head' is though, maybe it's something you want to keep	11:29
	right: https://lists.canonical.com/archives/bazaar/2008q2/041711.html	11:29
	← phretor_ left	11:29
viranch	j416: i guess i found the perfect one: https://github.com/offbytwo/git-hg/issues/12 :)	11:30
j416	viranch: right, that's what the first link suggested	11:30
	good or not, I can't tell	11:30
viranch	oh it worked ;)	11:31
	→ jutaro joined	11:32
j416	viranch: if I understood the mercurial documentation correctly, another solution would be to first clone the repo to a new mercurial repo, and then import that; unnamed heads should be gone	11:33
	← dwmw2_gone___ left	11:35
	→ dwmw2_gone___ joined	11:35
	← beautifulmind left	11:36
	→ kriegerod joined	11:37
viranch	j416: the heads are in the remote repo, so cloning it else where wouldn't make any difference	11:38
j416	viranch: oh, ok	11:38
	viranch: then I'm out of ideas	11:38
	→ niblha joined	11:42
rfx	I'm back without Q :)	11:46
niblha	I'm getting different diff paths for certain files in a commit when using git format-patch vs using a git diff. E.g >git format-patch -k master gives a path like ROOT/../../thefile.txt when I'm expecting ROOT/subdir1/subdir2/thefile.txt. (ie most other files in the commit are having the subdir path prepended correctly. any ideas what I'm doing wrong?	11:47
rfx	no wait, back with Q actually	11:47
	→ sikao_lfs1 joined	11:48
	← akosikeno left	11:49
	← dr0id left	11:50
	← chrisf\|laptop left	11:50
	← viranch left	11:52
rfx	How do I diff files between branches, but only the ones changed (modified), not adds/deleted? Answer: git diff --diff-filter=M branch1..branch2	11:52
niblha	actually now it hit me that it only affects the paths of files that are in dirs that I've added since last svn origin head (I'm interacting with an svn repo through git-svn) .. guess that has something do with it..	11:53
	← bauruine left	11:54
	→ w0bni joined	11:55
	→ urbanmonk joined	11:56
	→ dr0id joined	11:56
	← seivan left	12:00
	→ seivan joined	12:02
	→ akosikeno joined	12:03
	← rfx left	12:04
	← avinashhm left	12:06
	→ Chibby joined	12:07
	← Chib left	12:08
	→ Prasad joined	12:11
	← Murr_ left	12:14
	→ __name__ joined	12:15
	← __name__ left	12:15
	→ __name__ joined	12:15
	← krz left	12:15
	← __name__ left	12:16
nocturnal	if I've got a master branch locally and no files remotely, how can I upload my local branch to the remote repo when the repo is empty?	12:18
	← kumbayo left	12:18
	→ Murr_ joined	12:19
	→ namemon joined	12:21
	→ glennpratt joined	12:21
	→ dr_lepper joined	12:22
	← dr_lepper left	12:22
	→ dr_lepper joined	12:22
	→ eijk joined	12:25
bremner	nocturnal: it sounds like you might find !bare helpful	12:27
jast	nocturnal: automatic message an explanation of bare and non-bare repositories can be found here: http://bare-vs-nonbare.gitrecipes.de/	12:27
	← phaedrix left	12:29
	→ arunce_ joined	12:34
	← EricInBNE left	12:34
	← arunce left	12:35
	→ arunce joined	12:37
	→ dekebalus joined	12:37
	→ herdingcat joined	12:37
	→ whitman joined	12:38
	← arunce_ left	12:38
	→ Chillance joined	12:42
	← glennpratt left	12:42
sitaram	necromancer: would be nice if you confirmed my diagnosis also, in addition to saying your fixed it :)	12:43
	← djbpython left	12:43
	→ arunce_ joined	12:43
bremner	you sound like Dr. House ;)	12:44
	← akosikeno left	12:44
	← arunce left	12:44
	→ jp42 joined	12:44
	→ nd___ joined	12:46
	← dekebalus left	12:49
	← nd__ left	12:49
bremner	during a rebase, if I haven't run rebase --continue yet, how can I re-resolve the conflicts I just messed up? Do I have to start the rebase again?	12:49
	oh, well. I realized I need to redo the rebase anyway.	12:50
	elliottcable → ec\|fkn_away_nick	12:51
	ec\|fkn_away_nick → elliottcable	12:51
	← chrischambers left	12:51
	→ EricInBNE joined	12:51
	← Adaptee left	12:54
	← babilen left	12:54
	→ workmad3 joined	12:55
	→ Error404NotFound joined	12:56
	← seivan left	12:56
	→ akosikeno joined	12:57
	→ Dave^\| joined	12:58
	← SimonNa left	12:58
	← andrewpthorp left	13:00
	← flaguy48 left	13:02
	→ nowhereman joined	13:02
	← herdingcat left	13:03
	airtonix_ → airtonix	13:04
	→ apocalyptiq joined	13:04
	→ herdingcat joined	13:04
	← nowhere_man left	13:05
	← thews left	13:06
	← lm-brodo left	13:07
	← sagex_ left	13:08
	rvsjoen_ → rvsjoen	13:08
	← Chillance left	13:10
nocturnal	thanks bremner	13:10
	→ sagex_ joined	13:10
	← dr0id left	13:11
	→ djbpython joined	13:12
	← Error404NotFound left	13:12
	→ SimonNa joined	13:12
	← herdingcat left	13:12
	→ herdingcat joined	13:13
	→ aspotashev joined	13:17
	→ babilen joined	13:17
	→ Bass10 joined	13:18
	→ thiago joined	13:19
	← namemon left	13:20
	→ namemon joined	13:21
	← whitman left	13:21
	← sbell left	13:22
	→ JDuke128 joined	13:25
JDuke128	hi , i was searching some git security , i want to put password to my repo but i didnt see any doc for that ?	13:25
	→ Vikash joined	13:25
	→ Chillance joined	13:25
	→ Error404NotFound joined	13:26
j416	JDuke128: git is a source code manager, not an encryption utility	13:26
	JDuke128: you will have to use other tools for that	13:26
	← rchavik left	13:26
JDuke128	j416 , tool ?	13:27
	which one?	13:27
j416	I can't say, it depends on what your goal is.	13:27
JDuke128	everyone can checkout my repo on git ? :D	13:27
j416	if they have access to it, yes.	13:27
drizzd	I think he wants ssh	13:27
j416	JDuke128: you might want to check out gitolite	13:27
drizzd	or https + .htaccess	13:27
j416	JDuke128: and also read up a bit on how distributed version control systems work :)	13:28
JDuke128	but svn also has password protection...	13:28
	why git doesnt have :S	13:28
	j416 afk	13:29
wereHamster	JDuke128: git doesn't have what?	13:29
JDuke128	password on read and write some project	13:29
	like svn doesnt allow read on some projects	13:30
	privileges	13:30
	svn --user xxx --pass yyy co hxxp://xxx.com	13:30
	so with user and pass allowed...	13:30
wereHamster	of course it has. if you use ssh then only users which have ssh access can access the repository	13:30
	if you use http you can use whatever http protection the web server provides	13:30
	← Chibby left	13:30
	→ Chib joined	13:31
	→ bauruine joined	13:31
wereHamster	.htaccess was mentioned earlier, pki is another option	13:31
JDuke128	wereHamster , how ? git daemon ? any tutorial for that?	13:31
wereHamster	no, git daemon is for read-only anonymous access, it provides no authentication or encryption	13:32
	JDuke128: google 'gitolite'	13:32
sitaram	JDuke128: let's be clear about one thing. Are you intending to allow other people access to your repos or are they meant to be only for you?	13:32
JDuke128	okay , i want to ask 1+ question is , I want encyrpted repo , when my repo is stolen by someone , they cant read any source all will be AES 1024 bit like encyrpted , so my stolen project will be secure again.On Client side source code decrpts	13:33
	→ rchavik joined	13:33
JDuke128	i ve key on my usb	13:33
	and only its decoding sources	13:33
	my IDE will decode source at runtime	13:34
sitaram	JDuke128: use an encrypted file system. Luks on Linux works great. I've heard there are good tools for WIndows also	13:34
JDuke128	possible?	13:34
bremner	JDuke128: http://xkcd.com/538/	13:34
JDuke128	ok i m using Solaris	13:34
sitaram	anything else is impossible, IMP	13:34
	IMO	13:34
	JDuke128: and you didn't answer my previous question	13:35
JDuke128	i need encyrpted file system on Solaris so... ZFS not enough (:	13:35
drizzd	JDuke128: what kind of IDE decrypts source code for you?	13:35
JDuke128	IntelliJ	13:35
	IntelliJ IDEA	13:35
jast	there is no AES1024	13:36
	in general, the easiest way is to just make an encrypted volume and put your data on it	13:36
JDuke128	256	13:36
	← Adys left	13:36
jast	I don't know solaris well enough to be able to recommend anything, though	13:36
JDuke128	oh	13:37
	http://blogs.oracle.com/darren/entry/encrypting_files_in_solaris_10	13:37
jast	according to wikipedia, ZFS supports transparent encryption	13:37
bremner	if you encrypt, then check in to git, your git performance will suck	13:37
JDuke128	is really good :)	13:37
	my server is super fast doesnt matter	13:38
bremner	no, it does matter, trust me.	13:38
JDuke128	hmm	13:38
	← Vikash left	13:38
JDuke128	what can i do so ?	13:38
	← Chillance left	13:38
jast	http://download.oracle.com/docs/cd/E19963-01/html/821-1448/gkkih.html	13:38
bremner	I'm talking about things basically not working well because you check binary blobs into git; that is different than running git on top of encrypted file system, which is fine.	13:39
sitaram	bremner: is that comment only about ZFS? I have luks on all my systems and haven't really noticed much of a problem (though I agree my usage is light)	13:39
	← hachque left	13:39
sitaram	bremner: scratch what I said then; I misunderstood	13:39
jast	sitaram: he was talking about encrypting files prior to adding them to git	13:39
sitaram	yup	13:39
JDuke128	maybe client side encyrption better...	13:39
	but i think this time it makes search problem	13:39
	i cant make serverside search :)	13:39
	→ Chillance joined	13:39
jast	correct	13:39
JDuke128	do i really need search ? i think no...	13:40
bremner	if you have a remote server, then an encrypted file system doesn't really help, since things are unencrypted for transmission	13:40
JDuke128	Base64 + AES	13:40
drizzd	I don't see the problem. Use encrypted file systems everywhere, use https/ssh always. You're done.	13:40
bremner	heck, if your remote server is compromised, the intruder has things nicely decrypted already in a typical luks-like setup	13:41
	← Prasad left	13:41
BinGOs	The only safe option is to nuke from orbit.	13:41
	→ jfkw joined	13:41
drizzd	bremner: yeah, I hate that about cryptsetup	13:42
bremner	well, paranoia is fine. but security-theatre is a pet peeve.	13:42
JDuke128	but	13:42
	if someone get my user ssh access , i m done too :S	13:43
	← Error404NotFound left	13:43
jast	if your remote server is compromised, the attacker can simply read the encryption key from memory	13:43
	→ Andy_S joined	13:44
bremner	the private key never goes there.	13:44
	oh, still about luks. dunno	13:44
	← aspotashev left	13:44
jast	never goes where?	13:45
bremner	to the remote server	13:45
	(ssh)	13:45
jast	I was talking about the remote server on which the filesystem is encrypted	13:45
bremner	right. (belatedly) understood.	13:45
JDuke128	mm	13:46
	can i implement finger vain authentication to ssh ? http://www.hitachi.eu/veinid/	13:47
BinGOs	fs encryption is a safe-guard against physical theft of the encrypted media.	13:47
	→ [intra]lanman joined	13:48
bremner	which is typically not the main issue with a colocated server ;)	13:48
JDuke128	http://www.zdnet.com/blog/btl/forget-passwords-try-finger-veins/12082	13:49
bremner	JDuke128: veinid isn't really related to git.	13:49
	← delian66 left	13:49
	← cesario left	13:50
	← BlindOracle left	13:50
JDuke128	rite	13:50
bremner	git just uses ssh.	13:50
	← banisterfiend left	13:51
namemon	hi, is it possible get back master if one deletes it anyhow?	13:51
	→ glennpratt joined	13:52
	→ banisterfiend joined	13:52
	→ nightshift joined	13:53
	→ BlindOracle joined	13:54
bremner	namemon: probably man git-reflog can help	13:54
jast	namemon: automatic message the 'git-reflog' manpage can be found at http://www.kernel.org/pub/software/scm/git/docs/git-reflog.html	13:54
	→ PigeonFriend joined	13:55
	→ andrewpthorp joined	13:56
	← mragab left	13:56
	→ Error404NotFound joined	13:56
	← [intra]lanman left	13:56
	← airborn left	13:59
	→ madewokherd joined	14:00
	← glennpratt left	14:01
	→ aspotashev joined	14:03
	→ glennpratt joined	14:04
	→ awallin joined	14:06
	→ Corren joined	14:06
	→ phretor joined	14:07
	← airtonix left	14:09
	→ airborn joined	14:10
	→ airtonix joined	14:10
	→ kephas joined	14:10
	→ Doc__ joined	14:11
	← herdingcat left	14:11
	← Andy_S left	14:12
	← Doc__ left	14:12
	← nowhereman left	14:13
	← namemon left	14:14
	← airborn left	14:14
	← aspotashev left	14:14
madewokherd	hey, I forgot to configure my name in a new repo, and it has some commits now	14:17
	→ lanthan_ joined	14:17
madewokherd	is there an easy way I can go back and change the name in the existing commits?	14:17
	← lanthan left	14:18
	← jeff-h left	14:18
	→ jeff-h joined	14:19
	→ bpg_in joined	14:22
	→ replore_ joined	14:22
	→ whitman joined	14:23
	← nightshift left	14:23
jast	madewokherd: yep, but keep in mind it's generally a bad idea to do that if someone else already has the current history	14:23
	you'll want to do something involving "git filter-branch --env-filter"	14:24
	← jp42 left	14:24
madewokherd	odd, the env-filter documentation says to see git-commit(1) for details on the environment variables, yet the git-commit man page lacks an environment variable section	14:26
	but I found a thing in google that works >_>	14:26
	mrchrisadams_ → mrchrisadams	14:26
	→ Adys joined	14:28
j416	madewokherd: man git-commit	14:28
jast	madewokherd: automatic message the 'git-commit' manpage can be found at http://www.kernel.org/pub/software/scm/git/docs/git-commit.html	14:28
j416	ops	14:28
madewokherd	only thing in "ENVIRONMENT AND CONFIGURATION VARIABLES" is GIT_EDITOR	14:28
j416	madewokherd: man git-commit-tree	14:29
jast	madewokherd: automatic message the 'git-commit-tree' manpage can be found at http://www.kernel.org/pub/software/scm/git/docs/git-commit-tree.html	14:29
j416	there.	14:29
madewokherd	oh	14:29
	then I think the man page for filter-branch needs to be updated	14:30
	→ alberto56 joined	14:30
j416	nnn no?	14:31
	it clearly says git-commit-tree	14:31
	← Agos left	14:31
	→ masando joined	14:32
	← Octalot left	14:32
madewokherd	only under --parent-filter	14:32
	or maybe it was fixed since 1.7.2.5	14:33
	→ zivester joined	14:34
madewokherd	yeah, the on the website it says commit-tree, so I guess it was fixed	14:34
	← ilteris left	14:35
	→ variable joined	14:35
	→ Vikash joined	14:35
variable	I just commited a changed and pulled from master which resulted in a merge conflict	14:35
	how can I revert to my last commit ?	14:35
	← airtonix left	14:38
	← Corren left	14:38
	→ airtonix joined	14:38
	→ liquidproof joined	14:42
	← BlindOracle left	14:42
	← apocalyptiq left	14:43
	→ giallu joined	14:43
	← giallu left	14:43
	→ giallu joined	14:43
	← liquid-silence left	14:44
	→ nowhere_man joined	14:49
	← giallu left	14:49
	← kephas left	14:52
	→ szihs joined	14:53
	← phretor left	14:57
j416	variable: git reset --hard	14:58
	→ v4n_ joined	14:58
j416	to get to where you were before you started the merge	14:58
	→ phretor joined	14:59
	→ tobago joined	15:00
	→ rideh joined	15:00
	→ Morpf joined	15:00
	← sou_ left	15:00
variable	j416: thanks	15:00
	→ djbpython1 joined	15:01
	← djbpython left	15:01
	→ giallu joined	15:02
	← giallu left	15:02
	→ giallu joined	15:02
	→ BlindOracle joined	15:03
	← djbpython1 left	15:06
	→ thews joined	15:06
	→ codatory joined	15:07
	→ seivan joined	15:08
	→ sou_ joined	15:08
	→ droolingmnky joined	15:08
	→ shedammit joined	15:09
	← shedammit left	15:09
	→ shedammit joined	15:09
	← phretor left	15:10
	← giallu left	15:10
	→ phretor joined	15:12
	← szihs left	15:13
	→ bagheera joined	15:13
	→ systix joined	15:16
	→ szihs joined	15:16
	← replore_ left	15:17
	← optixx left	15:18
	→ optixx joined	15:18
	← zivester left	15:20
	← LongBeach left	15:20
	← szihs left	15:21
	→ ziro` joined	15:22
	← droolingmnky left	15:22
	← Murr_ left	15:24
	→ Prasad joined	15:25
	→ Murr_ joined	15:26
	← Vikash left	15:26
	→ sternocera joined	15:28
sternocera	Hello. I'm paranoid about tampering of my github hosted repository. Please reassure me.	15:28
	→ pandeiro joined	15:28
j416	sternocera: if you have the sha1 of the head of your branch, don't worry	15:28
	→ djbpython joined	15:29
	← systix left	15:29
rane_	sternocera: what's your concern	15:30
sternocera	rane_: Malicious parties tampering with my code	15:30
	← angelsl left	15:30
	← SirFunk left	15:30
sternocera	I don't believe that's a very well-founded concern, but I'd like to know exactly how it can be addressed.	15:30
rane_	host your own git repository	15:30
sternocera	I have reasons for not wanting to do that.	15:31
rane_	github staff has access to your code, i don't think there's any way around that	15:31
sternocera	rane_: I don't care about that. I care about their being able to modify it without me noticing.	15:32
rane_	ah, that i don't think is possible	15:32
	→ jjlee joined	15:32
codatory	You can't (really) edit a commit after its been published	15:33
jjlee	How do I find out the type of a git object given its hash?	15:33
sternocera	codatory: Right, because that would entail changing the SHA-1 digest of tip, because the changes would have to cascade.	15:34
	→ gusnan joined	15:34
sternocera	What I'd like to get is an authoritative document or webpage that explains that in a nutshell.	15:34
jast	jjlee: git cat-file -t <ID>	15:34
jjlee	jast, thanks	15:34
codatory	sternocera: Well, if you try it you'll note that to pull somewhere else you have to do a git pull --force and see a bunch of errors	15:35
sternocera	codatory: Right...the only way it could be done is if someone modified the file while not changing the SHA-1 digest...they'd have to find a collision, which is infeasible.	15:36
	← phretor left	15:37
j416	sternocera: read this: http://book.git-scm.com/1_the_git_object_model.html	15:38
	sternocera: if you have the sha1 of a commit which you have verified to be the correct one (i.e. your version), you can be sure that if the remote also has that same sha1, there has been no tampering	15:40
	(all the way up to that commit, throughout the whole commit chain	15:40
	)	15:40
	sternocera: http://www.youtube.com/watch?v=4XpnKHJAok8#t=56m	15:42
	watch that for 2 minutes or so from there	15:43
	:)	15:43
	→ phretor joined	15:43
sternocera	j416: Thanjks	15:44
	← tomas_w left	15:44
sternocera	j416: This is exactly what I was looking for. Awesome. Thanks again.	15:45
j416	:)	15:45
	you're welcome	15:45
	← nowhere_man left	15:48
	→ nowhereman joined	15:49
	← rchavik left	15:50
	← Error404NotFound left	15:51
	→ tomas_w joined	15:51
	← avar left	15:52
	→ avar joined	15:52
	← avar left	15:52
	→ avar joined	15:52
	← SimonNa left	15:52
	→ bennigraf joined	15:53
	← sou_ left	15:54
	→ DarkHouse joined	15:58
	→ SirFunk joined	15:58
	← niblha left	15:59
	→ droolingmnky joined	16:04
	→ Octalot joined	16:04
	← sternocera left	16:05
	← codatory left	16:05
	← jjlee left	16:05
	→ Error404NotFound joined	16:06
	← rvsjoen left	16:07
	→ rvsjoen joined	16:08
	→ MikhX joined	16:08
	← f0i left	16:08
	→ codatory joined	16:12
	→ Abbaduxo joined	16:14
	← Abbaduxo left	16:14
	← EricInBNE left	16:16
	→ EricInBNE joined	16:17
	→ liquid-silence joined	16:18
	← eijk left	16:19
	← codatory left	16:23
	→ f0i joined	16:24
	← glennpratt left	16:26
	→ codatory joined	16:26
	→ glennpratt joined	16:27
	← glennpratt left	16:28
	→ zivester joined	16:28
	→ NetRoY joined	16:29
	→ mandric joined	16:29
	← NetRoY left	16:29
	← freaky[t] left	16:30
	← sikao_lfs1 left	16:31
	← brainbox1100 left	16:32
	← codatory left	16:32
	→ _james_b joined	16:32
	→ Corren joined	16:33
	← Corren left	16:36
	→ dfr\|mac joined	16:37
	→ freaky[t] joined	16:38
	← mandric left	16:39
	→ phaedrix joined	16:44
	→ towski joined	16:44
	→ kephas joined	16:44
	← Error404NotFound left	16:45
	→ anzime joined	16:45
	← droolingmnky left	16:47
	← nowhereman left	16:48
	← tobago left	16:48
	→ bitplane joined	16:48
	→ mandric joined	16:48
	← jeff-h left	16:48
	elliottcable → d57bfa72c2e4b4a4	16:49
	→ Kreative` joined	16:49
	← bitplane left	16:49
	→ bitplane joined	16:49
	→ _psychic_ joined	16:49
bitplane	Hi, I have a quick question. I've got a file that I want to include in my project, it's made by someone else. Can I commit the file as me, but set the author as someone else?	16:49
	d57bfa72c2e4b4a4 → LZ05qOyiJ0CXpo00	16:49
	→ Corren joined	16:50
	→ jeff-h joined	16:50
	→ glennpratt joined	16:50
bitplane	and if I do that, is it a recommended thing to do? I'd like to preserve authorship even though it's public domain	16:50
	LZ05qOyiJ0CXpo00 → twtsqZOWqHlDutmM	16:52
	→ Myth17 joined	16:53
	← Myth17 left	16:53
	→ Myth17 joined	16:53
	→ dr_lepper_ joined	16:53
	← urbanmonk left	16:54
	← f0i left	16:54
	← dr_lepper left	16:54
	→ ceej joined	16:54
	→ f0i joined	16:55
j416	bitplane: yes you can, but .. it's your project, you made the decision to add that code	16:55
bitplane	so the author is the author of the code change, not the code itself?	16:55
j416	would make more sense to me if you set the author to yourself and wrote in the message that you imported stuff from somewhere	16:55
	the author of the change, yes, that is what I would feel be most logical	16:56
	→ nitish joined	16:56
	cehteh usually puts a note directly into the file when he borrows some code	16:56
	→ amstan joined	16:56
j416	yeah	16:56
	it's not that the person who wrote that code is adding it to your repo	16:56
	I'd be rather confused to find myself as author of some commit in someone elses repo that I never touched	16:56
cehteh	maybe with reference from where i got the code (upstream web page)	16:57
amstan	i have random backup files showing up in my repo, they look like: HyperTriangle.l#1, where HyperTriangle and the digit changes, how can i git ignore it?	16:57
j416	wouldn't want to be responsible for it either	16:57
	cehteh: yeah	16:57
bitplane	thanks guys, I'll just add it to the commit message then	16:57
amstan	i tried .l# but i have a feeling that the # comments the rest of the line	16:57
zorun	maybe \# ?	16:58
amstan	zorun: seems to work, thanks	16:58
	← mandric left	16:58
	← amstan left	16:58
	→ nitish_mythology joined	16:59
	← Myth17 left	16:59
j416	strange though, man page of gitignore doesn't state that you have to escape # in the middle of a line	16:59
	maybe it's implied.	16:59
	→ cherro joined	17:01
mmattice	it says a line _starting_ with # is a comment	17:02
	← phretor left	17:02
j416	yeah	17:02
	← nitish left	17:02
j416	should reword that if it actually treats inline # as comment marker	17:03
scyx	hi, i'm new to git, could anyone tell me how i could do this right: i want to continue developing a github-hosted program on my local gitosis server while still keeping track of the original development on github and also another fork of the program which is hosted on github. the original github project probably won't see any more development, but i'd like to occasionaly partially merge stuff from the forked github project into my local gitosis repo..	17:03
rane_	local gitosis repo, why does it have to be gitosis, why can't it be a normal repo	17:04
	you have more than one developer?	17:04
	→ JoeMaverickSett joined	17:04
	← MavJS left	17:04
	JoeMaverickSett → MavJS	17:04
j416	scyx: clone the repo, set up remotes for all the places you want to track.	17:04
	← nitish_mythology left	17:05
scyx	rane_: well i guess it wouldn't make much of a difference if i'd skip gitosis, i guess. but gitosis doesn't seem to be much additional work, setting it up seemsed to be pretty straightforward	17:05
	→ phretor joined	17:06
j416	scyx: !gitosis	17:08
jast	scyx: automatic message gitosis is no longer maintained and supported by the author; we usually recommend gitolite instead which has much better documentation and more features: http://github.com/sitaramc/gitolite	17:08
	← Corren left	17:08
j416	scyx: also, if you're the only dev, .. this is overkill.	17:09
jast	I kind of prefer it anyway :)	17:09
j416	jast: gitolite even though you are the only dev?	17:09
jast	yeah	17:09
j416	interesting	17:09
jast	though I actually only have an old gitosis instance (don't tell anyone)	17:09
j416	heh	17:09
	what advantage do you have?	17:09
jast	just on the off chance that other people want to join in later on	17:10
j416	ah	17:10
	handy.	17:10
	→ OOPMan joined	17:10
jast	though in fact lately I've just uploaded stuff to github instead	17:10
	and non-OSS solo projects I don't upload in the first place	17:11
j416	ok :)	17:12
scyx	hm okay, might just skip gitosis, but atm i'm trying to figure out the remote tracking bit of git.. what i did until now was: git clone original.project.from.github.git; git remote add origin local.server.git; git remote add upstream original.github.git; git remote add fork forked.github.git	17:12
j416	yeah github is what I use for stuff that I make that I don't care if other people steal	17:12
	good for offsite backup	17:12
	→ sonnym1 joined	17:13
j416	erm, steal, I mean, it's even better because other people _can_ get copies. handy.	17:13
	← sonnym left	17:13
	← Prasad left	17:13
	← munichlinux left	17:14
	→ pantsman joined	17:14
	← pantsman left	17:14
	→ pantsman joined	17:14
	← Kreative` left	17:16
	→ mnaser joined	17:23
	← kriegerod left	17:24
	← OOPMan left	17:25
	→ patrick99e99 joined	17:25
	→ d0k joined	17:25
	→ [intra]lanman joined	17:25
	← [intra]lanman left	17:25
	→ [intra]lanman joined	17:25
scyx	so where do remote tracking branches come into play in my scenario? can't i just work on my local master branch, then occasionally git fetch from both repos and git merge it in?	17:27
	→ Davey_ joined	17:28
	← Davey_ left	17:28
	→ Davey_ joined	17:28
	← andrewpthorp left	17:29
scyx	also, is there something like selective merge? the github fork has some changes i'd not want to merge into my project but others which i want. should i just make a patch before mergeing it all in and then undo the changes which i don't want with the patch afterwards? :/	17:29
banisterfiend	scyx: cherry-pick	17:30
	→ aspotashev joined	17:30
	→ Kreative` joined	17:30
	→ sebastorama joined	17:31
scyx	banisterfiend: can it cherry-pick ignore recurring changes in single files? such as a different package names (java)?	17:31
banisterfiend	i dont know what you mean, cherry-pick essentially just applies a patch	17:32
bremner	scyx: cherry-pick -n, perhaps	17:32
	→ hendeby joined	17:33
	← DarkHouse left	17:33
	→ Tavoris joined	17:33
	← hendeby left	17:33
bremner	oh, read too quickly. No idea about that.	17:33
	→ Err404NotFound joined	17:34
	← EricInBNE left	17:34
	← sebastorama left	17:34
	← pantsman left	17:36
	→ pantsman joined	17:37
	→ Corren joined	17:37
Tavoris	finger bremner	17:38
	← Tavoris left	17:38
j416	lol	17:38
	← zamabe left	17:39
	← Corren left	17:43
	→ Haza` joined	17:45
	→ kerrick joined	17:46
	→ xxneolithicxx joined	17:50
	← [intra]lanman left	17:50
	← anzime left	17:52
	→ flaguy48 joined	17:52
	← envi left	17:52
	← MikhX left	17:52
	← Murr_ left	17:53
	→ Murr_ joined	17:54
	→ envi joined	17:55
	→ leavittx joined	17:55
	→ fp1 joined	17:57
	← Murr_ left	17:58
	→ doubletap joined	17:59
	→ fourcolors joined	17:59
	← _psychic_ left	18:01
	→ zamabe joined	18:03
	→ DarkHouse joined	18:04
	← xxneolithicxx left	18:04
	→ mDraze joined	18:06
	→ fbass joined	18:06
	← akosikeno left	18:06
	← PigeonFriend left	18:07
	← thews left	18:08
	twtsqZOWqHlDutmM → Du07okDjKfe0X4qF	18:09
	→ avinashhm joined	18:11
	← paissad left	18:11
	← mcc left	18:12
	→ kukks joined	18:14
	← awallin left	18:14
	→ mandric joined	18:15
	→ coldboot\|home joined	18:16
	← catsup left	18:18
	→ catsup joined	18:19
	← catsup left	18:19
	→ beiter joined	18:19
	→ MikhX joined	18:19
	→ mcc joined	18:19
	→ catsup joined	18:19
	→ cbx33 joined	18:19
	← catsup left	18:19
	→ catsup joined	18:19
	→ akosikeno joined	18:20
cbx33	hey all	18:20
	what (besides Git) does msysgit come installed with	18:21
	does it actually install cygwin, and bash?	18:21
	← jeff-h left	18:24
	→ jeff-h joined	18:24
	→ drev1 joined	18:25
	→ thews joined	18:25
	→ sbell joined	18:25
	→ cirwin joined	18:26
cirwin	Given a SHA1 of a file, can I find out which commit it is in?	18:26
	→ phretor_ joined	18:26
cirwin	*commits	18:26
banisterfiend	cirwin: good question	18:27
	→ jalama joined	18:27
cirwin	ah, long time no see banisterfiend :)	18:27
banisterfiend	;)	18:27
	← __iron left	18:28
cirwin	so the issue I have is that I want to update the yajl-ruby library, using the latest code from the yajl library. Unfortunately the maintainer of yajl-ruby has copied some random files from yajl into yajl-ruby. I'd like to see if they (ever) corresponded to a revision of yajl	18:28
	← phretor left	18:29
j416	cirwin: https://github.com/jwiegley/git-scripts/blob/master/git-find-blob	18:30
	something like that perhaps.	18:30
cirwin	I suppose so	18:30
j416	not sure git has anything builtin for it	18:30
	given gits object model, you would have to search for it like this in either case, since the blob doesn't have a reference to its parent	18:31
	s/parent/tree/	18:31
cirwin	sure — I was hoping to be lazy :p	18:32
	← catsup left	18:32
	→ catsup joined	18:32
	→ canton7 joined	18:32
	← ceej left	18:34
	← ramusara left	18:34
	← MikhX left	18:35
	→ mvaladas joined	18:37
	← mvaladas left	18:38
	← zamabe left	18:39
	→ mvaladas joined	18:39
	→ MikhX joined	18:39
cirwin	hrm	18:40
	this is fail	18:40
	← bblfish left	18:40
	cirwin goes to read about subtree merging instead...	18:41
	← bitplane left	18:41
	← _james_b left	18:42
	← mvaladas left	18:43
	→ mvaladas joined	18:43
	← jalama left	18:44
	→ mjessome joined	18:44
	← cyphase left	18:45
	→ bblfish joined	18:45
	← Err404NotFound left	18:50
	→ dannas joined	18:51
	→ rpg joined	18:52
	→ _james_b joined	18:54
	← _james_b left	18:54
	Du07okDjKfe0X4qF → G0EleimMoNF4WDUO	18:54
	→ Murr_ joined	18:55
	← dannas left	18:55
	← mvaladas left	18:56
	→ cyphase joined	18:57
	← fp1 left	18:58
	← aminpy left	19:00
	← akosikeno left	19:01
	← Octalot left	19:01
	← dr_lepper_ left	19:01
	→ RaceCondition joined	19:01
	← phretor_ left	19:02
	→ Err404NotFound joined	19:02
	← Modius left	19:02
	→ phretor joined	19:03
	→ nowhere_man joined	19:08
	→ zwick joined	19:08
	→ chrisf\|laptop joined	19:09
	← kephas left	19:11
	→ kumbayo joined	19:16
	← fourcolors left	19:17
	→ mvaladas joined	19:18
	→ joshsmith joined	19:20
	← dwmw2_gone___ left	19:23
	→ dwmw2_gone_ joined	19:25
cbx33	do i remember correctly that people in here said not to use git instaweb?	19:27
jast	dunno	19:27
	→ HG` joined	19:28
cbx33	it's weird if I just run git instaweb	19:28
	it works	19:28
jast	I never use it, but I don't know of any strong reasons to avoid it if you just need a quick something	19:28
cbx33	it I run git instaweb --start	19:28
	it doesn't	19:28
	← chrisf\|laptop left	19:28
scyx	weird. how come that when i have a repo with several remote branches, when i clone that repo, the remote branches don't appear in the newly cloned repo? i guess i'm misunderstanding something about git here..	19:30
jast	scyx: yeah, the missing piece is the distinction between local branches and remote-tracking ("mirror") branches	19:30
	← pantsman left	19:30
jast	the branches are cloned, but you don't automatically get local branches created for them	19:31
	→ phretor_ joined	19:31
jast	you can see the mirror branches by using 'git branch -r' (r = 'remote')	19:31
	to create a local branch for one, usually you can just use 'git checkout <name of branch>'	19:31
	→ nimrod10` joined	19:31
	← Err404NotFound left	19:31
	→ [intra]lanman joined	19:32
	← mvaladas left	19:33
	← phretor left	19:33
scyx	hm. when i do git branch -a on my original repo, i get "local-branch, master, remotes/repo1/master, remotes/repo2/master, remotes/origin/master" (origin being my gitosis server)	19:33
	← nimrod10 left	19:34
	← tomas_w left	19:34
scyx	when i do git clone original-repo and do git branch -a, i get: "local-branch, master, remotes/origin/HEAD -> origin/local-branch, remotes/origin/local-branch, remotes/origin/master"	19:35
	→ LiohAu joined	19:36
scyx	bah, i guess i should read up on git fundamentals again. what confuses me is that when i push my repo to gitosis and then clone it from there again, the remote-tracking branches are lost, only origin is there	19:37
	→ OOPMan joined	19:37
	→ jpeazy joined	19:38
	← jpeazy left	19:39
	→ tomas_w joined	19:39
	G0EleimMoNF4WDUO → lx5daVLzXAYWRFzQ	19:39
	→ jpeazy joined	19:39
jpeazy	If I messed up the file for my interactive rebase, how do I go back to editing it?	19:39
	← tomas_w left	19:39
	→ belkiss joined	19:41
	← mnaser left	19:41
	← Scott0 left	19:42
	← mandric left	19:43
	← kimico left	19:44
cbx33	scyx: when you clone you get the references	19:44
	→ Err404NotFound joined	19:44
cbx33	but the branches are not setup automatically	19:44
	→ kimico joined	19:44
jpeazy	git rebase -i --abort I guess	19:45
	:0	19:45
	← jpeazy left	19:45
jast	scyx: clone only clones local branches, i.e. even if you push the remote-tracking branches somewhere, clone won't get them	19:47
	→ Juo joined	19:49
	→ ThomasLocke joined	19:50
	← aspotashev left	19:50
	→ Scott0 joined	19:52
	← zwick left	19:53
	← OOPMan left	19:54
	→ Opettaja joined	19:54
	← lukeman left	19:54
	← banisterfiend left	19:55
	→ franklin joined	19:55
	→ ceej joined	19:56
	← MikhX left	19:56
	→ banisterfiend joined	19:56
	→ MikhX joined	19:58
	→ mandric joined	19:59
	→ LongBeach joined	19:59
	← jason237 left	20:00
	→ mnaser joined	20:00
	→ phretor joined	20:01
	← phretor_ left	20:02
	→ johnw joined	20:09
	→ OOPMan joined	20:10
	→ dreiss joined	20:10
	→ msieradzki joined	20:11
	→ caseymcg joined	20:12
	→ jason237 joined	20:13
	← marvin2 left	20:19
scyx	jast,cbx33: found something.. https://git.wiki.kernel.org/index.php/GitFaq#How_do_I_clone_a_repository_with_all_remotely_tracked_branches.3F seems to be a workaround for the not cloning remote tracking branches, but its not pretty	20:19
cbx33	yeh	20:19
	i remember seeing something like that	20:20
	→ marvin2 joined	20:21
	→ awallin joined	20:22
	← RaceCondition left	20:24
	← ceej left	20:25
	→ trufa joined	20:29
	← tore_ left	20:30
	→ Dave^\|\| joined	20:30
	← LiohAu left	20:30
	← dreiss left	20:31
	← Dave^\| left	20:33
	← babilen left	20:34
	→ Dave^\| joined	20:34
	← Dave^\|\| left	20:34
	← trufa left	20:35
	← bittin left	20:37
	lx5daVLzXAYWRFzQ → r3wPYEWNrKkAPAVO	20:38
	← Pupeno left	20:40
	→ Pupeno joined	20:41
	← Pupeno left	20:41
	→ RaceCondition joined	20:42
	← bblfish left	20:43
	→ LiohAu joined	20:44
	→ ningia joined	20:45
	→ bblfish joined	20:46
	→ disappearedng joined	20:47
	r3wPYEWNrKkAPAVO → elliottcable	20:48
	← glennpratt left	20:48
	→ cccaldas joined	20:48
	→ bblfish_ joined	20:49
	← kerrick left	20:50
	← deepthinker82 left	20:50
	→ glennpratt joined	20:50
	→ FunnyMan3595 joined	20:50
	← Phlogistique left	20:51
	→ babilen joined	20:51
	← bblfish left	20:52
	bblfish_ → bblfish	20:52
	← mjessome left	20:53
	← msieradzki left	20:54
	→ diegoviola joined	20:55
	← LiohAu left	20:55
	← kumbayo left	20:55
	→ Dave^\|\| joined	20:55
	← Scott0 left	20:57
	← kst` left	20:57
	→ kst joined	20:58
	← mandric left	20:58
	← Dave^\| left	20:59
	→ giallu joined	21:01
	← giallu left	21:01
	→ giallu joined	21:01
	→ trufa joined	21:03
	← MikhX left	21:05
	← disappearedng left	21:06
	← Dave^\|\| left	21:08
	→ Dave^\|\| joined	21:08
	← macabre left	21:09
	← cherro left	21:10
	→ MikhX joined	21:12
	← spaceonline left	21:12
	← ThomasLocke left	21:13
	← awallin left	21:13
	→ mandric joined	21:15
	→ diegovio1a joined	21:19
	← diegovio1a left	21:20
	← RaceCondition left	21:20
	← diegoviola left	21:20
	← [intra]lanman left	21:21
worstadmin	How can I set my staging and working directories to match a certain tag? git reset v2.1.0 --hard ?	21:24
rane_	--hard after reset	21:24
	but depending on your goal, you might want to use checkout instead	21:25
	← CSdread_ left	21:25
	← cbx33 left	21:25
	→ airborn joined	21:26
worstadmin	rane_, If I checkout that means my working directory stays the same though correct?	21:28
	→ msieradzki joined	21:28
offby1	no	21:29
	"git reset --hard" moves your branch. This can be somewhat dangerous if that results in a bunch of commits becoming unreferenced.	21:29
	← JEEB left	21:29
offby1	"git checkout", on the other hand, has no such risk	21:29
	→ dudz_ joined	21:29
offby1	_its_ risk is that you might wind up with a "detached head", which itself isn't a risk; but if you then make some commits, and then check out some other branch, those new commits will be unreferenced.	21:30
rane_	git checkout is more suitable if you just want to go check what the project looked like at specific version	21:30
offby1	Git is risky!	21:30
	offby1 wears armor	21:30
rane_	reset would be more suitable if you wanted to fork the project beginning from v2.1.0	21:31
	← glennpratt left	21:31
worstadmin	Yes that's what I want to do, but I want it to update my working direction as well	21:31
	← Murr_ left	21:31
	← johnw left	21:31
worstadmin	In this case revert the working directory to the exact state of tag v2.1.0	21:32
	→ Murr_ joined	21:32
rane_	git reset --hard v2.1.0	21:32
offby1	worstadmin: general-purpose git advice: 1) Read "Git For Computer Scientists" (http://eagain.net/articles/git-for-computer-scientists/); 2) Run "gitk --all"; 3) Learn about "git reflog"	21:33
	← LongBeach left	21:33
canton7	worstadmin if you want ot keep the work on your current branch since v2.1.0, don't reset --hard	21:33
worstadmin	I understand	21:33
	Moving the head pointer around - I think I see the difference now	21:34
	Operates like a linked list	21:34
canton7	exactly	21:34
	← mandric left	21:34
worstadmin	It scares me linus just hacked this up	21:34
canton7	checkout moves HEAD, reset moves the branch that HEAD points to, if it point to one, otherwise HEAD, rebase effectively moves the pointers from a commit to its parent	21:34
	← mnaser left	21:37
	→ aspotashev joined	21:37
	→ Scott0 joined	21:37
	→ joshteam joined	21:38
	→ s0ber_ joined	21:39
	← s0ber left	21:41
	s0ber_ → s0ber	21:41
FunnyMan3595	Hey, folks. git push is confusing me at the moment by saying that everything's up to date even though I think it should be tracking a remote branch that's a step behind.	21:42
	→ lanthan__ joined	21:42
FunnyMan3595	Pastebinning the config now.	21:43
canton7	FunnyMan3595, look at man git-config, push.default	21:43
jast	FunnyMan3595: automatic message the 'git-config' manpage can be found at http://www.kernel.org/pub/software/scm/git/docs/git-config.html	21:43
canton7	it's a little odd	21:43
	the default is, anyway	21:43
drizzd	FunnyMan3595: git push <remote> <branch>	21:44
	→ sol0 joined	21:44
	← lanthan_ left	21:44
canton7	his point was it was tracked, so should be no need to do that	21:44
	→ macabre joined	21:44
	→ dinsdale joined	21:45
drizzd	canton7: it's not clear at this point what exactly did happen, so instead of assuming I gave him the command that will definitely work	21:46
canton7	fair enough. my bet is he already knows that :)	21:47
	← coldboot\|home left	21:47
drizzd	otoh, if your assumption is correct, then my command would not work either	21:47
	→ mjessome joined	21:47
canton7	by default, "git push" pushes all "matching" branches, where matching means the local branch name matches the remote one	21:48
	← peterhil left	21:48
canton7	iirc, tracking set up, or no	21:48
	→ peterhil joined	21:48
canton7	i did some experimenting: https://gist.github.com/1053846	21:48
	← peterhil left	21:50
	→ peterhil joined	21:50
	← beiter left	21:50
	← fbass left	21:50
	→ sechrist joined	21:50
	← peterhil left	21:52
	→ peterhil joined	21:52
	← peterhil left	21:53
	→ peterhil joined	21:53
	→ rascal999 joined	21:53
rascal999	is git pull recommended over push?	21:54
	← ThiefMaster\|work left	21:54
	← peterhil left	21:54
	→ ThiefMaster\|work joined	21:55
drizzd	rascal999: the two are not interchangeable, so your question makes no sense	21:55
	→ tizzo-afk joined	21:55
cirwin	rascal999: a pull-only workflow can be clunky compared to pushing to a central server	21:56
canton7	rascal999, hmm? what context? for sharing changes, different situations have different preferred approaches. the most comment i've seen is push-based, where people push to, and pull from, a central bare repo	21:56
	→ CoinOpeBoy joined	21:56
	drizzd marvels at the mind reading skills in here	21:57
zorun	drizzd: you're not the only one :)	21:57
rascal999	http://hans.fugal.net/blog/2008/11/10/git-push-is-worse-than-worthless/	21:57
jast	that's practically ancient	21:58
	also the page is taking ages to load for me	21:58
canton7	drizzd, experience. you get to know the usual questions, then could make a good guess at which of those this one was	21:58
drizzd	canton7: and to what end?	21:58
canton7	the appearance of mind-reading :)	21:59
	← joel_falcou left	21:59
jast	rascal999: !bare @	21:59
	rascal999: automatic message an explanation of bare and non-bare repositories can be found here: http://bare-vs-nonbare.gitrecipes.de/	21:59
	← Juo left	21:59
jast	this might help	22:00
FunnyMan3595	canton7, drizzd: Thanks; got distracted by other worries right after asking. I'll take a look at that now.	22:00
	← Dave^\|\| left	22:00
	→ Dave^\|\| joined	22:00
	← whitman left	22:01
FunnyMan3595	Aha, looks like push.default = upstream will do what I was expecting.	22:01
	← HG` left	22:01
	→ cschneid joined	22:01
canton7	FunnyMan3595 only sensible option imo. only pushes the current branch though	22:02
	← kpreid left	22:02
canton7	I'd love to see a setting that pushed all tracking branches	22:02
FunnyMan3595	Brilliant, that's exactly what I want. I can even drop the remote name now.	22:02
	Many thanks!	22:02
	→ kpreid joined	22:03
	← FunnyMan3595 left	22:03
	← jeff-h left	22:03
	← drizzd left	22:03
jast	canton7: patches welcome, I'm sure :)	22:04
	→ jeff-h joined	22:04
canton7	jast yeah i know :) it's a time thing, as usual. which is why i'm not complaining	22:05
	→ Unode\|Work joined	22:05
jast	:)	22:05
	← joshsmith left	22:06
Unode\|Work	Is it possible to use git submodules in a way that the submodules always point to HEAD instead of a specific commit?	22:06
jast	no	22:06
canton7	Unode\|Work nope	22:06
	look at git-slave, might be more to your tastes	22:06
Unode\|Work	canton7, will check thanks	22:06
	→ Octalot joined	22:07
squeakytoy	Japan beats usa and wins womens fifa world cup, and an american girl twitters "We will always have WWII" sigh	22:07
	← jstemmer left	22:08
	→ joshsmith joined	22:08
	← SirFunk left	22:09
	← kane77 left	22:09
jast	I don't even know what that is supposed to mean	22:09
	→ gfto joined	22:10
	→ ProLoser\|Mobile joined	22:10
bremner	it means next time the want to use nukes	22:11
Unode\|Work	canton7, seems like a solution but also a different tool... I'm not sure it will work with all the plugins on our system that call "git" since from what I read you should use "gits". Thanks for the pointer though, I'll need to digest this further	22:11
	← N0va` left	22:12
canton7	Unode\|Work, yeah it might well be that it's not what you want -- there's no "ideal" solution. i was just throwing it out there in case it happened to be what you were after	22:12
Unode\|Work	canton7, no actually the tool seems to do what I wanted but not in the intended way :)	22:12
canton7	hehe, well it's a start!	22:12
Unode\|Work	yup	22:13
	← sol0 left	22:13
Unode\|Work	thanks again. now to get things working	22:13
	← mastroDani left	22:18
	→ fbass joined	22:18
	→ disappearedng joined	22:19
	← disappearedng left	22:19
	→ disappearedng joined	22:19
	← bblfish left	22:19
	← joshsmith left	22:19
	→ bblfish joined	22:20
	← ningia left	22:20
	→ glennpratt joined	22:22
	→ RaceCondition joined	22:22
	← disappearedng left	22:24
	← phretor left	22:24
	← Dave^\|\| left	22:24
	← joshteam left	22:25
	→ angelsl joined	22:28
	← canton7 left	22:29
	← jonanin left	22:30
	← gfto left	22:33
	→ AdmiralSausage joined	22:34
	→ phretor joined	22:36
	→ [intra]lanman joined	22:36
	← macabre left	22:38
	→ ScottG joined	22:39
	← j416 left	22:39
	→ jonanin joined	22:39
ScottG	What would I put in my .gitignore to ignore everything but files that end in .py?	22:39
	I tried !*.py but that didn't work.	22:39
	← neuro_damage left	22:40
	→ mnaser joined	22:41
	← tiago left	22:42
	→ jbw` joined	22:44
	→ tlvb joined	22:45
	→ trenthauck joined	22:45
tlvb	Is there a way to have gitolite block gitweb from showing a particular repo? right now I have my own user as RW+ and no other access specifications.	22:46
	← fbass left	22:46
	→ phretor_ joined	22:48
	← jutaro left	22:48
	← phretor left	22:50
	← giallu left	22:50
	← trenthauck left	22:54
	← cschneid left	22:55
	→ dreiss joined	22:56
	← scyx left	22:56
	← ScottG left	22:58
	→ tiago joined	22:58
bremner	tlvb: do you have a description for that repo?	22:59
	← vmil86 left	22:59
	← Sonderblade left	23:01
	→ nazgul102 joined	23:01
	← v4n_ left	23:02
	→ fbass joined	23:02
	→ tansell-laptop joined	23:02
	← Vile left	23:03
	→ chrisf\|laptop joined	23:04
	← nazgul101 left	23:04
	→ diegoviola joined	23:04
	← Kreative` left	23:05
	← kipras left	23:06
	← belkiss left	23:09
	→ p3rror joined	23:10
	← p3rror left	23:10
	← tewecske left	23:12
	→ p3rror joined	23:12
	← alnewkirk left	23:13
	← oc80z left	23:13
	→ oc80z joined	23:14
	→ alnewkirk joined	23:15
	← phretor_ left	23:17
	← RaceCondition left	23:17
	← AdmiralSausage left	23:18
	← Davey_ left	23:19
	→ apocalyptiq joined	23:20
	→ Corren joined	23:24
	→ ketas joined	23:25
	← mrtadis_ left	23:28
	← orafu left	23:29
	→ orafu joined	23:29
	→ jpmelos joined	23:30
	← jeff-h left	23:30
	← rideh left	23:31
	ziro` → Guest76450	23:36
	→ spearce joined	23:37
	← w0bni left	23:40
	← jpmelos left	23:40
	← workmad3 left	23:43
	→ EricInBNE joined	23:45
	→ mrtadis joined	23:45
	← Unode\|Work left	23:46
	← ProLoser\|Mobile left	23:47
	← apocalyptiq left	23:49
	← airborn left	23:52
	← ketas left	23:52
	→ Kreative` joined	23:53
	← diegoviola left	23:58
	← rascal999 left	23:59

Logs Search ←Prev date Next date→

Channels Documentation