IRCloggy #git 2012-04-26

Provider of IRC logs since 2005.

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

	2012-04-26
	← kenperkins left	00:00
	← master-lincoln left	00:01
	→ DangerousDave joined	00:01
	← Araxia_ left	00:03
avu	Is there anything in the plumbing that has an exit code of 0 only if the current working directory is completely clean (ie no unstaged changes, no untracked files, 'git status' would turn absolutely nothing)? (I would want to run this in a pre-deploy script)	00:03
	→ Araxia_ joined	00:03
FauxFaux	exit $(git status -s \| wc -l)	00:03
	→ macmartine joined	00:03
avu	FauxFaux: yes, I am aware I can "parse" the output of 'git status', I was asking if there's something neater futher down in the plumbing	00:04
	← dreiss left	00:04
	← satyanash left	00:04
	← macmartine left	00:05
	← vinay left	00:05
	→ thursday joined	00:05
FauxFaux	I'm reasonably sure there isn't.	00:05
jrnieder	avu: "git update-index --refresh -q >/dev/null 2>&1"	00:05
	that's my favorite way to spell it	00:05
	→ AaronMT joined	00:06
jrnieder	others will probably curse at you because the exit code is just a side-effect	00:06
	← AaronMT left	00:06
jrnieder	that's doesn't check for untracked files though, I think	00:06
	← eletuchy left	00:06
jrnieder	you might also want to run "git ls-files -o"	00:06
avu	that also lists ignored files though. and the former also returns 0 if there are staged changed	00:07
	← jnxa left	00:07
jrnieder	if ignored files are ok, they you'll need --exclude-standard	00:07
	s/they/then/	00:08
	→ mikepack joined	00:08
	← Shishire left	00:08
avu	s/changed/changes/	00:08
jrnieder	avu: and you're right about it accepting staged changes	00:08
	→ aleperalta joined	00:09
jrnieder	so you might want to use "git diff-index --exit-code HEAD" instead	00:09
	→ nif joined	00:09
	← isao left	00:09
	← ericmuyser left	00:09
jrnieder	refreshing first is probably a good idea to avoid false positives after "touch"-ing a file, and people won't curse at you for it	00:09
aleperalta	Hello, can I cherry-pick and then edit the files? as I don't want a particular line to be merged.	00:09
	→ jfelchner1 joined	00:09
	→ Shishire joined	00:10
jrnieder	aleperalta: sure	00:10
FauxFaux	aleperalta: cherry-pick --no-commit	00:10
jrnieder	as FauxFaux said, or commit --amend when done	00:10
	timtowdi	00:10
aleperalta	FauxFaux, jrnieder, thanks a lot	00:10
avu	jrnieder: alright, seems that a combination of ls-files and diff-index can do what I want. Now I have to decide if I think that's better or worse than parsing the output of status :)	00:11
	← jfelchner left	00:12
avu	jrnieder, FauxFaux: thanks for your help	00:12
jrnieder	avu: status --porcelain :)	00:12
	→ krzkrzkrz joined	00:12
jrnieder	I just like the commands that express exact intent a little more	00:12
	I think the primary intended purpose of "git status --porcelain" is as input for a GUI that wants to show the same information as "git status" does	00:14
	← nif left	00:14
	← Shishire left	00:14
avu	jrnieder: yeah, sure, if I parse status' output then I will use it like this, otherwise who knows how the user's config alone might break the script :)	00:14
	← ehsan left	00:14
	← krz left	00:14
	krzkrzkrz → krz	00:14
avu	I just would have prefered something with a nice exit value I can just throw at fabric's local() ;)	00:15
	anyway, again, thanks for all the input!	00:15
	← avu left	00:15
jrnieder	cheers	00:15
	← davidbanham left	00:16
	→ master-lincoln joined	00:16
	← brngardner left	00:16
	tizzo → tizzo-afk	00:16
	→ Vert joined	00:16
	← ology left	00:16
	← jim0203 left	00:17
	→ jim0203 joined	00:17
	← berserkr left	00:17
	→ replore_ joined	00:17
	→ brngardner joined	00:18
	← jfelchner1 left	00:18
	← alberto56 left	00:18
	← Radium left	00:19
	→ victorbarna joined	00:20
	← tqmz left	00:20
	← victorbarna left	00:21
	← diegoviola left	00:21
	← jrnieder left	00:21
	← skaushik92 left	00:23
	← cccaldas left	00:23
	→ diegoviola joined	00:24
	→ krzkrzkrz joined	00:24
	→ jalama joined	00:25
	→ pun joined	00:26
	← krz left	00:27
	krzkrzkrz → krz	00:27
	→ davidbanham joined	00:27
	← cakehero left	00:28
	→ ehsan joined	00:30
	→ krzkrzkrz joined	00:32
	← master-lincoln left	00:34
	→ krzkrzkrz_ joined	00:35
	← krz left	00:35
	krzkrzkrz_ → krz	00:35
	← krzkrzkrz left	00:37
	→ edude03 joined	00:38
	← Infiltrator left	00:38
	→ Infiltrator joined	00:39
	→ w09x_ joined	00:40
	← w09x left	00:43
	→ fyolnish joined	00:43
	← Eridius left	00:44
	← streblo left	00:44
	← xsbeats left	00:45
	→ v4n_ joined	00:46
	← usmanity left	00:47
	← fir31ight left	00:47
	← avinashhm left	00:47
	→ dirkle joined	00:47
	→ fir31ight joined	00:47
	← edude03 left	00:48
	→ Eridius joined	00:50
	← thelazydeveloper left	00:51
	→ BrutalBob joined	00:52
	→ edude03 joined	00:55
	← DangerousDave left	00:55
	→ dreiss joined	00:55
	← brngardner left	00:56
	← enmand left	00:56
	← w09x_ left	00:57
	→ w09x joined	00:57
	← kukks left	00:58
	→ enmand joined	00:59
	→ nrichards joined	01:02
	← v4n_ left	01:02
	→ v4n_ joined	01:03
	→ alesan joined	01:04
alesan	hi	01:04
gitinfo	alesan: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	01:04
alesan	I am using git submodule that is very very tricky	01:04
	→ alberto56 joined	01:04
	→ dfr\|lghtninghall joined	01:04
alesan	at some point I get this:	01:04
	→ v4n__ joined	01:04
alesan	Initialized empty Git repository in /ephemeral/staging/build/wireless/.git/	01:05
	fatal: reference is not a tree: 97db460da18ebaf88eaaea8e8d13c0b0d0df86bb	01:05
	Unable to checkout '97db460da18ebaf88eaaea8e8d13c0b0d0df86bb' in submodule path 'wireless'	01:05
	what could it be?	01:05
	this is after git submodule update --init and while doing git pull	01:06
	no sorry	01:06
	this is while git submodule update --init	01:06
	→ cooloney joined	01:07
	→ Goplat joined	01:07
	→ w09x_ joined	01:07
	← w09x_ left	01:08
	→ w09x_ joined	01:08
	← w09x_ left	01:09
	→ w09x_ joined	01:09
	← mister_m left	01:10
	← w09x left	01:10
	← w09x_ left	01:11
	← wilmoore left	01:11
	← CannedCorn left	01:11
	→ w09x joined	01:11
	→ weiyang joined	01:11
	← claudiu__ left	01:13
	← lectrick left	01:13
	← bcardarella left	01:13
	← tfnico_ left	01:13
	← xerxas left	01:13
	← denysonique left	01:13
	← jaequery left	01:13
	← etcetera left	01:13
	← offbyone left	01:13
	→ claudiu__ joined	01:13
	→ bcardarella joined	01:14
	→ jaequery joined	01:14
	← bcardarella left	01:15
	← MetaCosm left	01:15
	← johtso left	01:15
	← claudiu__ left	01:15
	→ nobitanobi joined	01:15
	← jnewland left	01:15
	← pl0sh_ left	01:15
	← v4n_ left	01:16
	← alberto56 left	01:16
	← actel left	01:16
nobitanobi	When I do git diff I can see the changes that hasn't been commited yet. How can I see the changes that have been commited, and that are about to be pushed?	01:17
	← dreiss left	01:17
	← v4n__ left	01:17
	→ atdt joined	01:18
	→ v4n_ joined	01:19
	→ v4n__ joined	01:19
	→ CEnnis91 joined	01:20
	← v4n_ left	01:20
	→ denysonique joined	01:21
	→ pureluck joined	01:21
	→ bcardarella joined	01:21
	→ claudiu__ joined	01:21
	→ Mowsh joined	01:21
	→ jmcantrell joined	01:22
	← marhaban left	01:22
	← shirish left	01:23
	→ pl0sh_ joined	01:23
	→ marhaban joined	01:24
	← thebignoise left	01:24
	→ akraut joined	01:24
	→ GameGamer43 joined	01:24
	← kodie left	01:24
	→ jbergstroem joined	01:24
	→ kodie joined	01:25
jbergstroem	If I have a remote tag and branch with the same name, how do I remove the branch?	01:25
	← dstorrs left	01:25
	→ pikpik joined	01:26
	→ macmartine joined	01:26
	← sayd left	01:27
	← jameswarren left	01:27
	→ thebignoise joined	01:28
	→ ben225 joined	01:28
	← g00s left	01:28
	→ pandeiro joined	01:28
	→ arnihermann_ joined	01:28
	→ johtso joined	01:28
	→ tfnico_ joined	01:29
	→ jameswarren joined	01:29
	→ MetaCosm joined	01:30
	← Araxia_ left	01:30
	→ mbrochh joined	01:30
	← mbrochh left	01:30
	→ mbrochh joined	01:30
	← ben_alman left	01:30
	→ actel joined	01:30
	→ Vampire0 joined	01:31
	← Vampire0_ left	01:33
	← ColKurtz left	01:33
	← twisted` left	01:33
	← weiyang left	01:34
	→ shirish joined	01:34
	→ Targen joined	01:34
	← thearthur left	01:35
	← thebignoise left	01:35
	← ben225 left	01:35
	← alesan left	01:36
	→ twisted` joined	01:36
	← HyperNerdV2 left	01:37
	→ ben_alman joined	01:39
	← harish left	01:39
	← jalama left	01:42
	→ lectrick joined	01:43
	→ xerxas joined	01:43
	→ xjiujiu joined	01:44
	→ ben225 joined	01:45
	→ mister_m joined	01:46
	→ X99 joined	01:47
	← v4n__ left	01:47
	→ NimeshNeema joined	01:47
	← ben225 left	01:47
	← palyboy left	01:49
	← shirish left	01:49
	← nobitanobi left	01:49
	← xjiujiu left	01:50
	← SegFaultAX\|work left	01:50
	→ smw joined	01:50
	→ luke3321 joined	01:50
	← timoxley left	01:51
	← pandeiro left	01:52
	→ frogonwheels joined	01:53
	← X99 left	01:53
	→ xjiujiu joined	01:53
	→ cakehero joined	01:54
	→ w09x_ joined	01:54
	→ palyboy joined	01:56
	→ weiyang joined	01:56
	← mgpcoe left	01:56
	→ timoxley joined	01:56
	→ etcetera joined	01:56
	← w09x_ left	01:57
	→ w09x_ joined	01:57
	← jbergstroem left	01:57
	← w09x left	01:57
	→ intripoon_ joined	01:58
	← w09x_ left	01:58
	→ w09x joined	01:58
	→ zdk joined	01:59
	→ Slix` joined	01:59
	→ gavin_huang joined	02:00
	← Eridius left	02:00
	← intripoon left	02:01
	← matthiaskrgr left	02:02
	← darren_ left	02:03
	→ v4n__ joined	02:03
	→ darren_ joined	02:04
	→ ericmuyser joined	02:05
	← pun left	02:08
	← MissionCritical left	02:09
	← jcastelli left	02:09
	→ alexim joined	02:10
	← darren_ left	02:10
	← davidbanham left	02:12
	→ ben225 joined	02:15
	→ ChaoWang joined	02:16
	→ Eridius joined	02:16
	→ thebignoise joined	02:16
	→ xiangfu joined	02:16
	← mischat left	02:16
	← ben225 left	02:17
	→ DofLeFou joined	02:19
	← dcarper left	02:20
	→ krzkrzkrz joined	02:21
	← luke3321 left	02:21
	← marhaban left	02:23
	← krz left	02:23
	krzkrzkrz → krz	02:23
	← hyperair left	02:24
	→ urbanmonk joined	02:25
urbanmonk	how do I undo a commit .. I mean to start a new branch with my changes and commit there ... nothing has been pushed	02:26
SethRobertson	have you committed the changes?	02:27
	→ tshauck joined	02:27
urbanmonk	yes	02:27
	i've made one commit to master	02:27
	which i meant for a new branch	02:28
EugeneKay	urbanmonk - !fixup	02:28
gitinfo	urbanmonk: 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!	02:28
SethRobertson	Then !fixup will tell you, but `git branch newbranch; git reset --hard HEAD^` will probably work, subject to exclusions that fixup tells you	02:28
gitinfo	So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	02:28
	→ xpot-mobile joined	02:28
	→ dreiss joined	02:28
urbanmonk	can I do git reset --soft HEAD^1 ..	02:28
	wait won't --hard change all the files	02:28
	don't i want --soft	02:28
	I want to keep the changes	02:29
	just move them to a new branch	02:29
	→ BrainFreak__ joined	02:32
	→ fuzzybunny69y joined	02:33
	← dreiss left	02:34
	→ scruple joined	02:34
	→ Per4u3e joined	02:35
	← Yuuhi` left	02:36
	→ oogatta joined	02:36
	← Vortex34 left	02:38
	→ hwrd\|work joined	02:39
	← maestrojed left	02:40
	→ Vesperus joined	02:41
	← bcardarella left	02:43
	→ jackh joined	02:43
	→ tyatpi joined	02:45
	→ offbyone joined	02:46
	→ desaila joined	02:46
	→ krzkrzkrz joined	02:47
	← ericmuyser left	02:49
	← krz left	02:49
	krzkrzkrz → krz	02:49
	← orafu left	02:49
	→ orafu joined	02:50
	← Mowsh left	02:51
	← DofLeFou left	02:52
	→ matthiaskrgr joined	02:54
	→ RobW joined	02:56
	← JEntrep left	02:56
	← weiyang left	02:56
	→ hyperair joined	02:56
	← enmand left	02:56
	← dv310p3r left	02:57
	← todd_dsm left	02:58
	← Lulzon left	02:59
	← aleperalta left	03:00
	← guampa left	03:01
	← rockets left	03:02
	→ seoaqua joined	03:02
	→ jalama joined	03:03
	→ JEntrep joined	03:03
	→ jexmex_ joined	03:03
	→ ISF joined	03:04
jexmex_	I did a checkout of a branch, but now if I try to checkout master again, it complains files that were added by checking out brach will be overwritten	03:04
	how do I handle this?	03:04
	→ JD joined	03:05
	← JD left	03:05
	Adys_ → Adys	03:05
	→ JD joined	03:05
	← JD left	03:05
jexmex_	these files are part of a submodule on the new branch	03:05
	but I dont want them to be merged yet (the same files, essentially anyways, exist on the master. We deleted them in the branch and made the a submodule pointing to another repo)	03:06
	← kami8785 left	03:06
jexmex_	should I just force it?	03:06
	← tshauck left	03:06
	→ dstorrs joined	03:07
	← Vesperus left	03:08
	← jalama left	03:08
	← yoklov left	03:09
	← Davey left	03:10
	← jim0203 left	03:11
	→ todd_dsm joined	03:11
	← tyatpi left	03:12
	→ yoklov joined	03:12
	→ krzkrzkrz joined	03:13
	← krz left	03:15
	krzkrzkrz → krz	03:15
	← i10neorg left	03:15
	→ simplew joined	03:17
simplew	in git archive is possible to start from a subdir form the git repo?	03:18
	SethRobertson: ping	03:19
	→ jfelchner1 joined	03:24
	→ agile joined	03:26
	← mikepack left	03:26
	→ tyatpi joined	03:27
	→ jiggliemon joined	03:28
	← jameswarren left	03:28
	→ bodom joined	03:29
	← Twey left	03:29
jiggliemon	Prob a newb question. I have a key on my host machine, and I try moving it to my VM (ubuntu 11, w/ git installed) but I still get this: http://cl.ly/0K0J46264218213H3K3U What's the deal yo?	03:29
	→ Twey joined	03:29
	→ subho joined	03:29
milki	probably not using the key...	03:31
	← cakehero left	03:32
jiggliemon	milki: You got any suggestions ?	03:33
milki	ssh -i to make sure the key works	03:33
	if it does, you need to either put it in an agent or use .ssh/config	03:33
	→ weiyang joined	03:33
	gitinfo set mode: +v	03:36
RobW	hey I'm using a submodule I only have pull access to, and the owners have put two tags on the same commit.	03:36
	anyone know if there's a way to check out one specific tag?	03:36
	← maletor left	03:36
RobW	git seems to default to the tag that was added first, no matter which I check out. Wouldn't matter but the tags hook into an update system.	03:37
	→ maletor joined	03:38
	← Slix` left	03:39
	→ fission6 joined	03:42
	← marc_kdab left	03:45
	→ dreiss joined	03:46
	→ ericmuyser joined	03:49
	→ hagabaka joined	03:49
	← hagabaka left	03:49
	→ hagabaka joined	03:49
	← timoxley left	03:51
	→ gigamonkey joined	03:53
	← jiggliemon left	03:53
	← krz left	03:53
	→ codeshah joined	03:54
	→ akiress joined	03:54
	→ macmartine_ joined	03:55
bambams	Did anybody happen to skim over my blog post to make sure I got it right? :\	03:56
	← maletor left	03:56
	← macmartine left	03:57
	macmartine_ → macmartine	03:57
	→ derp joined	03:57
	→ g00s joined	03:58
	→ beautifulmind joined	03:58
	← mister_m left	03:58
	→ redrocket joined	03:58
	→ wilmoore joined	03:59
	← derp left	04:00
	→ leo2007 joined	04:00
	→ sayd joined	04:01
	← BrutalBob left	04:02
	→ jeby joined	04:02
leo2007	In fixing the broken vc-git-state in emacs, I need to find a way to get the status of a file? Ideas?	04:03
	the output of which is then parsed by emacs	04:03
	← jeby left	04:04
	→ jeby joined	04:05
	→ jceb joined	04:05
	← sam113101 left	04:05
	← RobW left	04:06
	→ brngardner joined	04:07
	← brngardner left	04:07
	→ satyanash joined	04:09
leo2007	git diff-index -p --raw -z HEAD -- 80python.el	04:10
	returns	04:10
	:100644 100644 d5b1daa9b80734dc4fabbcf8efc086da8faa4469 0000000000000000000000000000000000000000 M80python.eldiff --git a/80python.el b/80python.el	04:10
	but that file has no changes	04:10
	← dstorrs left	04:13
	← jeby left	04:13
	→ jeby joined	04:13
	← mccann left	04:16
	← igotnolegs left	04:16
	← akiress left	04:17
	→ h0bbit joined	04:19
leo2007	<sha1> is shown as all 0's if a file is new on the filesystem and it is out of sync with the index.	04:19
	← xpot-mobile left	04:20
	→ ehsan_ joined	04:21
	→ dcarper joined	04:21
	← smw left	04:21
	← gchristensen left	04:22
	← albel727 left	04:22
	← codeshah left	04:22
	→ maletor joined	04:23
	← camdenx left	04:23
	→ mritz joined	04:23
	← ehsan left	04:24
	← madewokherd left	04:24
	← dcarper left	04:25
	→ albel727 joined	04:25
	→ caseymcg joined	04:25
	← flaviocamilo left	04:26
	→ w09x_ joined	04:27
	← w09x_ left	04:27
	→ w09x_ joined	04:28
	← jeby left	04:28
	→ jameswarren joined	04:28
	← w09x_ left	04:28
	← w09x left	04:29
	→ jeby joined	04:29
	→ w09x joined	04:29
	→ ferdna joined	04:29
	→ joshsmith joined	04:29
	infogulch → infogulch\|afk	04:29
	→ timoxley joined	04:31
	→ jeby1 joined	04:32
	← jeby left	04:32
	→ subbyyy joined	04:32
	← subbyyy left	04:33
	← SeySayux left	04:36
	→ subbyyy joined	04:38
	→ akiress joined	04:43
	akiress → Guest38096	04:43
	→ SeySayux joined	04:43
	← g00s left	04:44
	→ igotnolegs joined	04:44
	→ davidbanham joined	04:44
	← dreiss left	04:45
	→ `Assassin joined	04:46
	→ maestrojed joined	04:48
	← Milossh left	04:48
	← ericmuyser left	04:48
	→ JohnyBaby joined	04:52
	← simplew left	04:52
	← _Vi left	04:53
	← dirkle left	04:53
	← yoklov left	04:54
	← hwrd\|work left	04:55
	← fuzzybunny69y left	04:55
	← lcpfnvcy left	04:55
	← ceej left	04:55
	→ lcpfnvcy joined	04:55
	← v4n__ left	04:56
leo2007	what does it mean by "out of sync"	04:58
bob2	ENOCONTEXT	04:58
	← JEntrep left	04:59
	← jexmex_ left	05:00
	← wilmoore left	05:01
hendry	hi, I have a shallow copy of git which I've commited too, and now I want to prune it so it's "depth 1" again. My general taxi is to do a "depth 1" clone the repo, but I can't clone for an already shallow copy...	05:01
	So I'm asking how to make a shallow clone, of an already shall clone, if that makes sense	05:02
wereHamster	why do you want to make it depth 1 again?	05:02
	→ giallu joined	05:03
	← codewrangler left	05:03
hendry	wereHamster: make this as small as possible for delivering to customer	05:04
	← hagabaka left	05:04
	→ xecycle joined	05:05
	← xecycle left	05:05
hendry	oh this might be the answer http://stackoverflow.com/a/7937916/4534	05:08
wereHamster	make a tarball	05:08
	→ leavittx joined	05:08
	← Bunbury left	05:08
	neersighted\|AFK → neersighted	05:09
	→ hagabaka joined	05:10
	← hagabaka left	05:10
	→ hagabaka joined	05:10
	← bodom left	05:11
	→ khmarbaise joined	05:13
	→ herdingcat joined	05:13
	→ fmPirate joined	05:13
	→ dreiss joined	05:13
	→ fuzzybunny69y joined	05:13
	← fuzzybunny69y left	05:14
	← jceb left	05:18
	→ g00s joined	05:19
	→ isao joined	05:21
	neersighted → neer\|tf2\|koth	05:21
	neer\|tf2\|koth → neersighted	05:22
	← subho left	05:24
	← diegoviola left	05:27
	→ neunon joined	05:27
	← neunon left	05:28
	→ neunon joined	05:28
	→ JohnyBaby_ joined	05:28
	← fmPirate left	05:30
	→ fmPirate joined	05:30
	← leavittx left	05:30
	← JohnyBaby left	05:31
	→ airborn joined	05:34
	→ krz joined	05:34
	→ jceb joined	05:35
	← slikts left	05:37
	← JohnyBaby_ left	05:37
	← dfr\|lghtninghall left	05:38
	→ dstorrs joined	05:38
	← macmartine left	05:39
	→ JohnyBaby joined	05:39
	← khmarbaise left	05:41
	→ dc5ala joined	05:43
	→ harshpb joined	05:45
	← maletor left	05:45
	→ friskd joined	05:45
	← xjiujiu left	05:46
	← jceb left	05:47
	← edude03 left	05:50
	→ mishok13 joined	05:51
	→ hwrd\|work joined	05:53
	→ Sigma joined	05:55
	→ shruggar joined	05:56
	→ bitslip_ joined	05:57
	→ BiggFREE joined	05:58
	← BiggFREE left	05:58
	← milki left	05:58
	→ BiggFREE joined	05:58
	← RandalSchwartz left	05:58
	← jpalmer left	05:59
	→ catsup joined	05:59
	→ pfFredd joined	05:59
	← crazed_ left	05:59
	→ machine1 joined	05:59
	→ khmarbaise joined	06:00
	→ up_the_irons joined	06:01
	→ JStoker joined	06:01
	→ crazed joined	06:03
	→ milki joined	06:03
	→ blorbx joined	06:03
	← ferdna left	06:04
	→ wilmoore joined	06:04
	→ maletor joined	06:05
	← JohnyBaby left	06:06
	→ smw joined	06:06
	→ JohnyBaby joined	06:07
	← pfFredd left	06:07
	→ sq-one joined	06:07
	→ ggherdov joined	06:08
	→ gmarseliss joined	06:08
	← gmarseliss left	06:08
	→ glcrazy joined	06:08
	→ arvind_khadri joined	06:08
	→ rurufufuss joined	06:09
	→ macrover joined	06:09
	← khmarbaise left	06:10
	← glcrazy left	06:10
	→ glcrazy joined	06:11
	← project2501b left	06:11
	→ tsdh joined	06:12
	← joshsmith left	06:13
	← wilmoore left	06:16
	← g00s left	06:16
	→ cbreak-work joined	06:16
	→ rurufufuss_ joined	06:16
	← cbreak-work left	06:16
	→ cbreak-work joined	06:17
	→ papegaaij joined	06:17
	→ Sitethief joined	06:18
	← JohnyBaby left	06:18
	← l4m6da left	06:19
	← macrover left	06:21
	→ harish joined	06:21
	→ dcarper joined	06:21
	← barrbrain left	06:22
	← OOPMan left	06:23
	→ JohnyBaby joined	06:23
	← hwrd\|work left	06:23
	← rurufufuss_ left	06:24
	← mishok13 left	06:24
	← shruggar left	06:25
	→ flijten joined	06:26
	← dcarper left	06:26
	→ DofLeFou joined	06:28
	→ jceb joined	06:29
	← Per4u3e left	06:32
	← crazed left	06:33
	→ tewecske joined	06:34
	→ chepre joined	06:34
	← scruple left	06:34
	← isao left	06:34
	→ Dave^\| joined	06:34
	→ crazed joined	06:34
	→ _sundar_ joined	06:35
	← h0bbit left	06:38
	→ Pugnax joined	06:41
	→ shruggar joined	06:41
	← airborn left	06:42
	→ dharrigan joined	06:42
	← dharrigan left	06:43
	→ dharrigan joined	06:43
	← Goplat left	06:43
	← aspotashev left	06:44
	← shruggar left	06:45
	← tsdh left	06:46
	→ JohnBat26 joined	06:46
	← igotnolegs left	06:46
	tizzo-afk → tizzo	06:48
	← flijten left	06:48
	→ tsdh joined	06:48
	→ ThomasLocke joined	06:49
	neersighted → neersighted\|AFK	06:49
	← neurodrone left	06:50
	← ggherdov left	06:51
	← ISF left	06:52
	→ mishok13 joined	06:52
	→ pantsman joined	06:52
	← crazed left	06:53
	→ dpino joined	06:53
	← mritz left	06:53
	→ crazed joined	06:54
	← krz left	06:55
	→ tqmz joined	06:55
	→ dekroning joined	06:56
	→ vals_ joined	06:56
	→ Radium joined	06:57
	← tango_ left	06:57
	vals_ → tango_	06:59
	→ SamB joined	06:59
	→ OOPMan joined	07:01
	← WillMars_ left	07:01
	← Eridius left	07:04
	→ marthinal joined	07:05
	← timoxley left	07:07
	→ na3r joined	07:08
	← papegaaij left	07:11
	→ papegaaij joined	07:11
	← arvind_khadri left	07:13
	→ dr_lepper joined	07:13
	← dr_lepper left	07:13
	→ dr_lepper joined	07:13
	← papegaaij left	07:13
	→ arvind_khadri joined	07:14
	← arvind_khadri left	07:14
	→ arvind_khadri joined	07:14
	← arvind_khadri left	07:14
	← subbyyy left	07:14
	→ pro_metedor joined	07:15
	→ Per4u3e joined	07:16
	← dropdrive left	07:17
	→ scientes joined	07:17
scientes	im using mozilla-central in git	07:17
	how do i look up by hg commit ids?	07:18
	← dr_lepper left	07:18
	→ papegaaij joined	07:18
	→ Element9 joined	07:18
	→ dr_lepper joined	07:18
	→ harish_ joined	07:20
	← harish left	07:20
	→ lb joined	07:21
	→ bondar joined	07:21
	→ ben225 joined	07:22
	← Adys left	07:22
	→ resmo_ joined	07:23
	→ shruggar joined	07:24
	→ dropdrive joined	07:24
	→ papegaaij_ joined	07:26
	← papegaaij left	07:26
	→ Chib joined	07:27
	→ ivan12347 joined	07:27
	→ ivan12348 joined	07:27
	→ QaDeS joined	07:28
	← ivan12345 left	07:30
	← ivan12346 left	07:31
	← Per4u3e left	07:33
cbreak-work	scientes: git has nothing to do with hg	07:34
	e-ndy\|afk → e-ndy	07:34
	← Radium left	07:36
	→ fuzzybunny69y joined	07:36
	→ fayaz joined	07:36
	← xiangfu left	07:36
	→ amcsi_work joined	07:37
scientes	its just a hg repo hosted as a git clone	07:37
	so i was wondering if it provided lossless conversion	07:37
	like hg-git does	07:37
	→ Radium joined	07:39
	← beautifulmind left	07:40
cbreak-work	define lossless	07:42
	I doubt commit messages or history would get lost	07:42
scientes	hg-git can take a git repo, and keep all the git commit sha1s	07:42
cbreak-work	of course...	07:42
	because SHAs are derived from content	07:43
scientes	and its reproducable, so the same edit will produce the same git commits	07:43
cbreak-work	and it'd not be much of an scm if it'd change content	07:43
scientes	even when done in mercurial	07:43
	← thursday left	07:45
	→ angelsl joined	07:47
	→ netiru joined	07:47
	← bondar left	07:48
	→ ipalaus_ joined	07:48
	→ Chappo joined	07:51
	← Bass10 left	07:52
	← rvsjoen left	07:53
	→ rvsjoen joined	07:53
	→ jdpacker joined	07:53
	← jdpacker left	07:54
	← glcrazy left	07:55
	← TTSDA left	07:56
	→ rtuin joined	07:56
	→ timoxley joined	07:56
	→ flijten joined	07:56
	→ arvind_khadri joined	07:57
	→ Shusshu joined	07:57
	← arvind_khadri left	07:58
	→ adhawkins joined	07:58
	→ glcrazy joined	07:58
	→ arvind_khadri joined	07:58
	← arvind_khadri left	07:58
	→ arvind_khadri joined	07:58
	→ xiangfu joined	08:00
	← machine1 left	08:00
	→ Pugnax_ joined	08:01
	← Pugnax left	08:03
	← sq-one left	08:03
	→ machine1 joined	08:04
	→ twb joined	08:04
	← sayd left	08:04
	e-ndy → e-ndy\|afk	08:05
	← Orbitrix left	08:05
	→ hagebake joined	08:05
	← hagebake left	08:05
	→ hagebake joined	08:05
twb	I have LESS=SR, but when using git I would prefer it to be LESS=SRF which is what git sets it to. Currently I do this by unsetting LESS if I happen to be doing a lot of git runs, but is there a way I can make that automagic? I doubt there's a GIT_LESS, but GIT_PAGER='env -u LESS less' seems pretty fugly.	08:06
	← hagabaka left	08:06
	e-ndy\|afk → e-ndy	08:06
	← Chappo left	08:06
twb	GIT_PAGER='env -u LESS less' wouldn't work anyway, come to think of it, because git sets LESS iff its not set when git runs, not when less runs. So I guess I'd have to alias git=env -u LESS git	08:06
	← odin_ left	08:07
	→ thiago joined	08:07
	→ robbyoconnor joined	08:08
	← robbyoconnor left	08:08
	→ robbyoconnor joined	08:08
	← netiru left	08:08
	← machine1 left	08:08
	→ Orbitrix joined	08:09
	→ LimeBlast joined	08:10
	→ sfiku joined	08:10
	tizzo → tizzo-afk	08:10
	← pro_metedor left	08:11
	→ eijk joined	08:12
wereHamster	twb: git sets LESS=FRSX	08:12
	→ filier joined	08:12
wereHamster	twb: man git-config	08:12
gitinfo	twb: the git-config manpage is available at http://jk.gs/git-config.html	08:12
twb	wereHamster: it only sets it if LESS is not already set	08:13
wereHamster	twb: the core.pager option	08:13
twb	That's the problem	08:13
	→ TomyLobo joined	08:13
	← tyatpi left	08:13
	← ben225 left	08:13
twb	Hm, I see the git-config manpage has some notes that might be related	08:13
	→ Zenopus_ joined	08:13
	→ kevinfagan joined	08:14
	→ Iota_ joined	08:14
	← maletor left	08:15
	→ danmackay_ joined	08:15
	→ charon_ joined	08:15
	← charon_ left	08:15
	→ charon_ joined	08:15
twb	It looks like I can do something like GIT_PAGER='less -F' to get the behaviour I want	08:15
	← charon left	08:15
	charon_ → charon	08:15
	→ ph^ joined	08:15
	→ dj_tjerk\| joined	08:15
	→ ramlev joined	08:16
	→ jabosh joined	08:16
	← Twey left	08:16
	→ comboy_ joined	08:16
twb	That would break if less isn't installed, but I'm unlikely to run into that.	08:17
	→ ed8 joined	08:17
	→ zenazn joined	08:17
	→ mitsuhiko_ joined	08:17
ed8	hi, I want to start hosting a git projects on my company network. Is there a tutorial for that ?	08:18
	→ \toothrot joined	08:18
	← \toothrot left	08:18
	→ \toothrot joined	08:18
	→ tureba_ joined	08:18
	→ hsb_ joined	08:18
	→ CellKill_ joined	08:18
	→ muep_ joined	08:18
	→ bbuss_ joined	08:18
	→ ChoiZ_ joined	08:18
	← kadoban left	08:18
	→ hgb_ joined	08:18
twb	ed8: you can just create --bare repos and put them in a group-editable area	08:18
	→ quuxman_ joined	08:18
et	ed8: looking at gitolite would probably be a good start	08:18
	→ Twey joined	08:18
	→ phillijw_ joined	08:18
	→ Nikoli_ joined	08:18
twb	Yeah or gitolite if you want a web UI thingo	08:18
	→ dzan_ joined	08:18
	→ luka12346 joined	08:18
	→ f0i joined	08:18
	→ tonyj joined	08:18
	→ helium-3 joined	08:19
	→ Rafusek_ joined	08:19
	→ cods_ joined	08:19
	→ legnaleurc_ joined	08:19
et	gitolite isn't a web-ui-thing ...	08:19
	→ berserkr joined	08:19
twb	Huh. So it provides what, the git:// smart server?	08:19
	← cods left	08:19
	→ selckin1 joined	08:19
	→ anildigital_ joined	08:19
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	→ armani__ joined	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	→ e-ndy- joined	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	→ Harzilein_ joined	08:20
	→ lake1 joined	08:20
	← Harzilein_ left	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	← danmackay left	08:20
	← shibayu36 left	08:20
	← dzan left	08:20
	← Zenopus left	08:20
	← macUzer left	08:20
	← bla left	08:20
	← ramlev` left	08:20
	← zeom left	08:20
	← lake left	08:20
	← Iota left	08:20
	← Poapfel left	08:20
	← hgb left	08:20
	← mitsuhiko left	08:20
	← phillijw left	08:20
	← __class__ left	08:20
	← bolt left	08:20
	← pa left	08:20
	← joeytwiddle left	08:20
	← tureba left	08:20
	← selckin left	08:20
	← tonyj_ left	08:20
	← markatto left	08:20
	← Evanlec left	08:20
	← e-ndy left	08:20
	← Rafusek left	08:20
	← [E]sc left	08:20
	← merkosh left	08:20
	← legnaleurc left	08:20
	← Nikoli left	08:20
	→ The-C0mpiler joined	08:20
	→ Harzilein_ joined	08:20
	← Harzilein_ left	08:20
	Zenopus_ → Zenopus	08:21
	← cods_ left	08:21
	→ Harzilein_ joined	08:21
	→ pro_metedor joined	08:21
	← Harzilein_ left	08:21
	→ Harzilein_ joined	08:21
	← Harzilein_ left	08:21
	→ _class_ joined	08:21
	→ Harzilein_ joined	08:21
	→ cods joined	08:21
	→ demi` joined	08:21
	→ batrick_ joined	08:21
	→ markatto joined	08:21
	→ Evanlec joined	08:21
	→ linduxed joined	08:21
	← armani__ left	08:21
	→ Poapfel joined	08:21
	→ jab416171\|Cloud joined	08:21
	→ zorzar_ joined	08:21
et	twb: it usually (but does not have to) works over ssh, and provides repository- and branch-level access control, also logs stuff so you can see who pushed what ... etc	08:21
	→ luke joined	08:22
	→ bolt joined	08:22
	→ dcarper joined	08:22
twb	Oh, OK	08:22
	→ joeytwiddle joined	08:22
	← dcarper left	08:22
	→ piti joined	08:22
	luke → Guest37316	08:22
	← ed8 left	08:22
	→ ivan joined	08:22
	→ dcarper joined	08:22
	→ Valodim joined	08:22
	→ shibayu36 joined	08:23
	→ [E]sc joined	08:23
twb	(As you might have guessed, I am too lazy to do anything fancier than POSIX DACs)	08:23
	← FACEFOX-DOT-COM left	08:23
	Guest37316 → crooksey	08:23
	→ monty joined	08:24
	← hagebake left	08:24
	→ camdenx joined	08:24
	← FACEFOX left	08:24
	_class_ → __class__	08:24
	→ mattp_ joined	08:25
	→ Milossh joined	08:25
	← Pugnax_ left	08:25
	→ h0bbit joined	08:25
	← tsdh left	08:27
	→ babilen joined	08:27
	← fission6 left	08:27
twb	Hm, I'm looking at gitolite's docs to see what it's special sauce is. Does it take the same approach as rrsync, where it adds advice via an authorized_keys forced command?	08:28
	→ marcogmonteiro joined	08:28
	← camdenx left	08:28
	→ bla joined	08:29
	→ hagabaka joined	08:29
	← hagabaka left	08:29
	→ hagabaka joined	08:29
	→ zeom joined	08:29
	zeom → Guest94016	08:29
	→ pa joined	08:30
	→ FACEFOX joined	08:32
	→ Lord_DeathMatch joined	08:32
	→ achadwick joined	08:32
	→ userToto joined	08:33
	← ipalaus_ left	08:33
	→ Geralt joined	08:33
	← Geralt left	08:33
	→ Geralt joined	08:33
	← delight left	08:34
	→ FACEFOX-DOT-COM joined	08:34
	→ ed8 joined	08:35
	← userToto left	08:35
	→ userToto joined	08:37
	← userToto left	08:37
	→ PigeonFriend joined	08:37
	→ kami8785 joined	08:38
	← Targen left	08:38
	← Vinnie_win left	08:39
sitaram	twb: lots of rum and coke went into the making of gitolite	08:39
	sauce, not so much	08:40
	→ userToto joined	08:40
cmn	so you're saying you were drunk when you wrote it?	08:41
	← maestrojed left	08:41
	← hagabaka left	08:41
	← The-C0mpiler left	08:41
	← PigeonFriend left	08:41
	→ Vinnie_win joined	08:42
	→ The-Compiler joined	08:42
	← userToto left	08:42
sitaram	cmn: you know I like to be precise. Lots of rum and coke does not always mean drunk :)	08:42
	Jacky → JackyAlcine	08:42
cmn	sure, it does generally imply it	08:42
sitaram	twb: this may be a good starting point: http://sitaramc.github.com/gitolite/glssh.html (how gitolite uses ssh)	08:43
cmn	but I appreciate that you left the time-scale open	08:43
twb	sitaram: thanks	08:43
	← papegaaij_ left	08:43
	→ PigeonFriend joined	08:43
	→ pmatthews joined	08:44
twb	cmn: there was still some blood in is alcohol stream	08:44
	s/is/his/	08:44
	→ userToto joined	08:44
	← Vile left	08:44
	← userToto left	08:44
twb	sitaram: yeah it looks like you take the same approach with gitolite-shell as the rrsync script does	08:45
	→ userToto joined	08:45
	→ theotherguy joined	08:45
	← theotherguy left	08:45
	→ papegaaij joined	08:45
	gitinfo set mode: +v	08:45
twb	rrsync is a little perl thing that prevents rsync from accessing files it shouldn't, e.g. if you want to use rsync to back up /etc as root, but don't want the backup user to have write access as root to all of /	08:45
sitaram	twb: rrsync? I thought you typo-d rsync. Hmm...	08:46
	sitaram googles	08:46
twb	Nope it's a separate thingo	08:46
	It should ship with rsync	08:46
	In Debian it's in /usr/share/doc/rsync/examples/rrsync.gz IIRC	08:46
	→ hagabaka joined	08:46
twb	s/examples/scripts/	08:46
	JackyAlcine → Jacky	08:46
	→ netiru joined	08:47
	→ knopkodav joined	08:47
userToto	hello I've got a problem with Git and I don't know where it comes from : when I want to fetch from my devhost I get upload-pack failed to find object... and when I got a git gc it works why do I have to do this everytime ?	08:47
	I explained a little more my situation here : http://pastebin.com/Vdi3Cyan	08:47
ed8	tg: you told about gitosis/gitolite, do you recommend a tutorial ? I would also be interested for one about git-web	08:48
twb	sitaram: example usage: http://paste.debian.net/165372/	08:48
	→ master-lincoln joined	08:49
cmn	userToto: your HEAD ref is messed up, make it point somwhere else	08:50
	but the real question is why are you merging on the server?	08:50
ed8	does gitosis still under development ?	08:50
twb	sitaram: I haven't worked out if forced command actually locks out stuff like the internal-sftp server, or having sshd to chroot into /srv/vcs first, though	08:50
	← scientes left	08:50
thiago	ed8: gitosis is no longer developed	08:50
twb	sitaram: kudos for your docs saying "if you deploy this, you ARE a sysadmin, bloody well go learn to use SSH properly" :-)	08:51
sitaram	twb: :)	08:52
userToto	cmn: why is it messed up ? as user 'developper1' I can see git log working the files don't change between 'commitUSER' and 'developper1'.	08:52
ed8	thiago: should I install gitweb to manage my projects ?	08:52
cmn	git said it was messed up	08:52
	→ g00s joined	08:52
userToto	cmn: as for the merge on SERVER it's not my policy I have to do so	08:52
	→ shame joined	08:53
cmn	then change it	08:53
thiago	ed8: you can install gitweb, but it can't manage projects	08:53
cmn	it's only going to cause problems	08:53
thiago	ed8: I recommend gitolite	08:53
ed8	thiago: is there a web UI with it ?	08:53
	← babilen left	08:53
userToto	cmn: but it's only messed for one user not the other and also how to do I make it point to something valid ?	08:53
sitaram	twb: this will blow your mind: https://github.com/sitaramc/gitolite/blob/g2/contrib/adc/rsync (warning: not yet ported to gitolite v3; only works on v2.x)	08:54
thiago	ed8: ask the channel. I've never personally installed it. (never needed)	08:54
ed8	is there a web UI with gitolite or should I isntall gitweb alongside ?	08:54
cmn	userToto: it's probably a permissions issue	08:54
sitaram	ed8: no web UI with gitolite but someone made one: google gitlab	08:54
	← pro_metedor left	08:55
cmn	since you're trying to log in to someone else's computer and access their personal files	08:55
	← agile left	08:55
	→ stodan joined	08:55
twb	sitaram: yeah that looks pretty similar, at least in intent if not execution	08:55
	← satyanash left	08:55
	→ LekeFly joined	08:56
	→ sangi joined	08:56
twb	The main advantage of rrsync is it's tiny so you can just drop it somewhere and have it work	08:56
sitaram	twb: agreed; gitolite is overkill if all you want is rsync. But even if you want different permissions to different paths for the same user, rrsync won't cut it	08:56
	AFAICT	08:56
userToto	cmn: Ok will see if my umask is correct	08:57
twb	sitaram: correct	08:57
	sitaram: unless you have one key per share per user	08:57
sitaram	which they will hate you for :)	08:57
twb	Or you bind-mount all the shares into a common dir and allow access to it	08:57
	sitaram: usually I run both ends for my backup (not git) case ;-)	08:57
	And hey I am >this< close to forcing them to use browser-side TLS certs	08:58
	:-)	08:58
	→ tatsuya_o joined	08:58
sitaram	twb: if you do use this let me know so I can up the priority on porting rsync, because it will be quite different in gitolite v3.0	09:00
	→ pro_metedor joined	09:00
	→ andreagelati joined	09:01
	← master-lincoln left	09:01
twb	Unlikely that I'll give a shit in the near future, sorry	09:01
	← OOPMan left	09:02
ed8	when installing gitolite on a server for the whole company, what public key should I use in : gitolite setup -pk your-name.pub ?	09:02
	→ Pugnax joined	09:02
ed8	If I use mine, will I be the only to have access to the repo ?	09:02
	← e-ndy- left	09:02
twb	sitaram: incidentally, have you seen monkeysphere?	09:02
ed8	I need other to get access to it	09:02
twb	sitaram: it uses gnupg to bolt a web of trust model on top of ssh authorized_keys	09:02
	→ e-ndy joined	09:03
twb	Including key revocation	09:03
	← glcrazy left	09:03
	→ oehman joined	09:03
	→ fisted_ joined	09:03
	← hsb_ left	09:05
	→ hsb joined	09:05
	→ _Qman joined	09:05
	→ satyanash_ joined	09:05
	→ noisebleed joined	09:06
	← noisebleed left	09:06
	→ noisebleed joined	09:06
	→ cedeon joined	09:06
	← fisted left	09:06
	← thiago left	09:08
	→ whitman joined	09:09
userToto	cmn: Thanks it worked using find and adding the +x right to directories under .git :)	09:09
	← bolt left	09:10
userToto	Bye	09:10
	← Tocacar left	09:10
	← userToto left	09:10
	← dreiss left	09:12
	← lindenle left	09:13
	← PapaSierra left	09:13
	→ Vile joined	09:14
	→ project2501b joined	09:15
	← project2501a left	09:17
	→ ssp_ joined	09:17
	← g00s left	09:19
	← Shusshu left	09:20
	→ mischat joined	09:21
	→ tsdh joined	09:22
	← ocharles left	09:23
	→ _iron joined	09:24
	← davidbanham left	09:25
	→ ocharles joined	09:26
	→ babilen joined	09:26
	← ssp_ left	09:27
	→ davidbanham joined	09:27
	← warthog9 left	09:29
	← fuzzybunny69y left	09:30
ed8	I'm trying to clone the gitolite-admin repo from my server with: git remote add origin git@moorea:repositories/gitolite-admin.git	09:31
	but I got the error:fatal: Not a git repository (or any parent up to mount parent /mnt)	09:31
	and : Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).	09:31
	does it ring a bell ?	09:31
	← dr_lepper left	09:32
	→ dr_lepper joined	09:33
wereHamster	ed8: use git@moorea:gitolite-admin.git	09:34
	→ martinjlowm joined	09:34
	← Jacky left	09:34
	← _iron left	09:36
	← xiangfu left	09:36
	→ ipalaus_ joined	09:37
sitaram	twb: monkeysphere? I've heard of it...	09:39
	→ Miika-- joined	09:39
Miika--	Hello, on --pretty=format, what switch adds filename? I mean like %H adds the commit hash...	09:40
twb	Miika--: more than one file may be edited	09:41
	Miika--: perhaps you want --stat or --numstat ?	09:41
	→ Davey joined	09:42
sitaram	twb: oh damn; I'm getting old... http://sitaramc.github.com/gitolite/g2/monkeysphere.html	09:42
	→ bolt joined	09:42
twb	haha	09:43
	→ _Vi joined	09:43
	→ Per4u3e joined	09:44
sitaram	well that's a contributed doc and I forgot :)	09:44
twb	ah	09:44
sitaram	twb: honestly, sometimes I forget stuff I wrote so this is not that bad :)	09:44
	← jackh left	09:46
	← Geralt left	09:49
	→ Kicer86 joined	09:49
	→ thiago joined	09:50
	→ jalcine joined	09:53
twb	Miika--: I am doing some very nasty things which you might find appalling and/or interesting:	09:53
	git rev-list HEAD \| while read i _; do git ls-tree -r $i; done \| while read _ _ _ i; do echo $i; done \| sort -u \| while read f; do echo "$(git rev-list --count HEAD -- "$f") $f"; done \| sort -nr > $PWD.total-manifest-by-count; git rev-list HEAD \| while read h; do git show --numstat --oneline $h \| { read h d; c=0; while read x y _; do c=$((c+x+y)); done; echo "$h $c $d"; }; done \| sort -nk2 > $PWD.commits-by-size	09:53
The-Compiler	wow	09:54
Miika--	twb: :)	09:54
	← jalcine left	09:54
Miika--	i just creaated an alias for log that produces xml-like output	09:54
twb	That generates two summaries in ../ -- one of the commits by their size (in number of lines added+removed), and one listing files by the number of commits made to them	09:54
	I'm doing those to find things to rebase + squash prior to publishing	09:55
	← hgb_ left	09:55
twb	And then this to find commits that are probably squashable, with the aforementioned sorting by size:	09:55
	git show $(grep 'B$l\(\(arg\\|e\(rg$?\)h\)\\|raino\)\\|F$ix\\|nord\\|oo\\|uck$\\|Meh\\|Oops\\|T$weak\\|ypo$\\|U$gh\\|ndo$\\|Whitespace\\|Yuk\\|b$l\(\(arg\\|e\(rg$?\)h\)\\|raino\)\\|f$ix\\|nord\\|oo\\|uck$\\|meh\\|oops\\|t$weak\\|ypo$\\|u$gh\\|ndo$\\|whitespace\\|yuk' $PWD.commits-by-size \| cut -d' ' -f1)	09:55
	→ beautifulmind joined	09:56
twb	You could pass the same regex to git log -p --grep	09:56
	← weiyang left	09:56
Miika--	There's a list of placeholders for --pretty=format: but cannot find one that outputs changed filenames	09:57
	← caseymcg left	09:58
twb	Well, good luck. It's pub time	09:58
	← twb left	09:58
	→ J0rd4n joined	09:58
	← Pugnax left	09:59
	← tr-808 left	10:00
cmn	Miika--: that doesn't fit in a placeholder	10:00
	← replore_ left	10:00
cmn	use --name-status or --name-changed	10:00
	--name-only sorry	10:00
ikrabbe	hi, is there a way to define an alias with arguments? git myAlias-CMD REVISION	10:00
	← martinjlowm left	10:01
cmn	what got mixed up in there is that maybe you want 'git whatchanged'	10:01
	→ martinjlowm joined	10:01
cmn	ikrabbe: they're like that already	10:01
	← fyolnish left	10:01
	→ mischat_ joined	10:02
	→ ehsan joined	10:02
	← f0i left	10:03
Miika--	I have it like this, but I would add changed filenames too	10:03
ikrabbe	cmn: so I can define “git config alias.myAlias '!'"echo \"Revision=$1\"; git show \"$1\""” i.e.?	10:03
Miika--	http://stackoverflow.com/questions/6327276/parsing-git-log-output-preferably-as-xml	10:03
	→ fyolnish joined	10:03
	← netiru left	10:03
	← mischat left	10:04
	mischat_ → mischat	10:04
	→ batmanian joined	10:04
	← andreagelati left	10:04
ikrabbe	my problem is, that I need the argument multiple times, not just as the last one, appended	10:04
	← fyolnish left	10:05
cmn	ikrabbe: either that or you might have to define a function to use $1	10:05
	ikrabbe: parse it	10:05
	Miika--: sorry, that first to ikrabbe was to you	10:05
ikrabbe	cmn?parse it?	10:05
cmn	yeah	10:06
ikrabbe	what does that mean?	10:06
cmn	the word parse?	10:06
	→ shuaib joined	10:06
ikrabbe	in this context	10:06
shuaib	how can I tell git on the command line which ssh key to use	10:06
	← DofLeFou left	10:06
	← ehsan_ left	10:06
	→ Kabaka joined	10:06
cmn	it means the same as in any other context, grab the output and figure out where stuff is	10:06
ikrabbe	you want to tell me to write a command that parses its arguments, right?	10:07
	← ehsan left	10:07
cmn	yes	10:07
Miika--	cmn: any links on that?	10:07
ikrabbe	hmm, that renders alias useless for that case, ok	10:07
cmn	ikrabbe: wait, I 'm getting my people confused	10:07
	ikrabbe: I mean function in the sense of a shell function	10:08
	→ c1de0x joined	10:08
cmn	'!f() { echo yada yada $1} f' is a typical construct	10:08
shuaib	I am getting permission denied saying public key issue, even though my keys are in ~/.ssh	10:09
	what could be the issue	10:09
cmn	I couldn't see any in the config manpage, but I'm sure the Internet is full of examples	10:09
shuaib	can I mention what key to use to git on the command line	10:09
cmn	shuaib: you need to configure ssh to use the right key	10:09
	→ ryannathans joined	10:09
shuaib	cmn, elaborate please?	10:09
ryannathans	git gui, best software out.	10:09
	for windows i couldn't get it working in comment prompt	10:10
	→ tr-808 joined	10:10
ryannathans	so i use the ui	10:10
	→ DofLeFou joined	10:11
cmn	shuaib: write the configuration in ~/.ssh/config to tell ssh which key you'd like it to use	10:11
pro_metedor	I need to update a submodule, but I don't see files I have added to the submodule	10:11
	what can I do to make them tracked?	10:12
	← martinjlowm left	10:12
	→ martinjlowm joined	10:12
	→ netiru joined	10:14
	→ aJacom joined	10:15
	← harish_ left	10:15
	→ master-lincoln joined	10:16
	← Textmode left	10:16
	← Radium left	10:17
	→ bobsleer joined	10:19
	← DofLeFou left	10:19
	→ f0i joined	10:21
	→ JackyAlcine joined	10:22
	← JackyAlcine left	10:22
	→ JackyAlcine joined	10:22
	← JackyAlcine left	10:22
	→ jalama joined	10:22
	→ marhaban joined	10:24
	→ camdenx joined	10:24
	← herdingcat left	10:24
	→ arietis joined	10:25
	← camdenx left	10:26
	← berserkr left	10:27
	→ DofLeFou joined	10:29
	→ berserkr joined	10:30
	← jalama left	10:30
	→ glcrazy joined	10:30
	→ thierryp joined	10:30
	← DofLeFou left	10:34
	← shuaib left	10:34
	← shruggar left	10:34
	→ iocor joined	10:35
	← Per4u3e left	10:36
	→ DofLeFou joined	10:36
	← marhaban left	10:37
	→ shruggar joined	10:40
	→ Per4u3e joined	10:43
	← DofLeFou left	10:44
	→ borwar joined	10:44
	→ DofLeFou joined	10:45
	→ Altreus joined	10:45
Altreus	When I do `git fetch origin master`, and it receives stuff, should I expect origin/master to have advanced beyond master?	10:46
	using git log with --branches shows them all to be in the same place	10:46
	→ SicoWork joined	10:47
FauxFaux	Altreus: !four	10:48
gitinfo	Altreus: [!fetchfour] Never use the four-word version of git-fetch or git-pull (e.g. git fetch remote refspec). It always ends in tears. Yes, if you understand the implications of FETCH_HEAD it can technically be done, but really it is easier to just fetch the whole remote (or perhaps edit the fetchspec if you never want other bits).	10:48
cmn	in other words, no, because you've just told git not to update it	10:49
	→ pro_metedor_ joined	10:49
	← Sitethief left	10:49
Altreus	Is there a doc I can read to get a better understanding of what fetch does?	10:49
tsdh	Altreus: git fetch --help ?	10:50
	← pro_metedor left	10:50
	pro_metedor_ → pro_metedor	10:50
Altreus	key word "better"	10:50
	I read that and my understanding is at level 0	10:50
	wello OK 1	10:50
tsdh	Altreus: If that's the case, you'll probably won't need it anyway. ;-)	10:51
cmn	have you read progit?	10:51
Altreus	I have not	10:51
cmn	then do	10:51
Altreus	ok	10:51
	oh brillpads it's in ebook formats	10:52
	now I just need an ebook reader	10:52
tsdh	Altreus: http://progit.org/book/ You can also read it as PDF or HTML.	10:53
Altreus	I found it :)	10:54
	I might get the dead tree version	10:54
	oh maybe work will pay for it	10:54
	→ John___ joined	10:54
	→ j0ran joined	10:54
	← f0i left	10:55
	→ jalcine- joined	10:56
tsdh	Altreus: Go for it. It's definitively worth it.	10:56
	← kcm1700 left	10:57
	→ kcm1700 joined	10:57
	→ Pugnax joined	10:58
	→ Peste_Bubonica joined	10:58
	← hyperboreean left	10:59
	→ hyperboreean joined	11:00
	← elnn left	11:00
	← j0ran left	11:01
	→ elnn joined	11:01
	→ thorpe_ joined	11:01
	→ enmand joined	11:02
	→ w09x_ joined	11:03
	→ jmd joined	11:05
	← John___ left	11:05
jmd	How can I stop git from setting everything to read-only on reset?	11:06
	← oogatta left	11:06
	← w09x left	11:07
	← ryannathans left	11:08
	← aJacom left	11:10
	← ChaoWang left	11:11
	→ fuzzybunny69y joined	11:11
	← zdk left	11:12
	→ Vile1 joined	11:14
	Valodim → s1n1st3r	11:14
	s1n1st3r → Valodim	11:14
	→ tvw joined	11:14
	→ popcorn joined	11:14
	← popcorn left	11:14
cmn	jmd: everything meaning what? the files in the worktree?	11:15
	that's probably a umask issue	11:15
	← rchavik left	11:15
	← Vile left	11:15
	→ FernandoBasso joined	11:19
	← chronos left	11:19
	← dc5ala left	11:19
	← dekroning left	11:19
	→ rgr joined	11:19
pro_metedor	I am trying to commit a submodule modified file, but it is untracked. Git add doesn't work!	11:20
	http://book.git-scm.com/5_submodules.html	11:20
	Lord_DeathMatch → Lord_Deathmatch-	11:20
rgr	Hi, I have a pack file which is 320M ... dont know how or way. I'm at a stage where I dont mind losing old stuff in my git repo. Can I somehow get it to remove all code from say before a month ago? Or better to delete the repo completely and recreate?	11:20
	Lord_Deathmatch- → L_DM-eating	11:20
rgr	(its failing to rsync across ssh to hostgator)	11:21
	→ n2diy joined	11:21
	→ marhaban joined	11:22
	→ w09x joined	11:22
	roflin_ → roflin	11:22
pro_metedor	git add a/a.txt	11:23
	fatal: Path 'a/a.txt' is in submodule 'a'	11:23
	← TomyLobo left	11:23
	← lapistan- left	11:23
	→ vervic joined	11:24
	→ lapistano joined	11:24
cmn	pro_metedor: why are you trying to add a submodule's file to the supermodule?	11:24
	→ skylamer` joined	11:25
	← master-lincoln left	11:25
pro_metedor	I am trying to commit modified submodule file	11:25
	I have modified it but don't know...	11:25
cmn	go into its repo	11:25
pro_metedor	maybe I shotld commit it from inside a submodule dir	11:25
cmn	the submodule is a different repo	11:26
	← w09x_ left	11:26
cmn	if you want to do stuff to it, do it there	11:26
ndim	Make sure that the submodule working dir has a branch checked out.	11:26
cmn	and that the branch makes sense	11:27
	or better yet, treat the submodule as what it is, which is a separate project	11:27
pro_metedor	you mean different than master?	11:27
cmn	if that's what makes sense	11:27
pro_metedor	yeah, I am getting it now	11:27
	yeah ok, I see	11:28
	→ RaceCondition joined	11:28
	→ subho joined	11:29
	← marhaban left	11:29
	← leo2007 left	11:29
	→ Burgov_ joined	11:30
	← kcj left	11:30
	← Burgov_ left	11:31
	→ burgov joined	11:31
	gitinfo set mode: +v	11:31
burgov	hi	11:31
gitinfo	burgov: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	11:31
burgov	I have one very large object in my repository (36 MB). How can I find out what it is and possibly get rid of it?	11:31
rgr	me too : a pack file thats 320M! Trying to find the same.	11:32
FauxFaux	There's a set of scripts called things like "git object sizes" that'll tell you what's huge.	11:32
wereHamster	burgov: google 'git find-large'	11:32
	→ marhaban joined	11:33
	← ipalaus_ left	11:33
FauxFaux	rgr: I believe the easiest way to throw away old history is to man git replace large blocks of your history with a reference to a much older commit, or a new empty commit (git checkout --orphan), but it's messy.	11:33
gitinfo	rgr: the git-replace manpage is available at http://jk.gs/git-replace.html	11:33
	→ dv310p3r joined	11:33
rgr	FauxFaux: thanks. messy sounds bad ;)	11:34
	I'll look into that	11:34
	→ ipalaus_ joined	11:34
FauxFaux	Or, if it's recent or the history is short, just delete the commits from rebase --interactive, or something.	11:34
	← Per4u3e left	11:34
burgov	that should be doable	11:34
	thanks for the tips	11:34
	← fayaz left	11:35
	→ Per4u3e joined	11:36
	→ rchavik joined	11:37
	← rchavik left	11:37
	→ rchavik joined	11:37
	← iocor left	11:37
	→ ccc joined	11:39
	→ ChrisLE joined	11:39
	gitinfo set mode: +v	11:41
	← lb left	11:41
ChrisLE	hi, is it possible to make commit or push rules like: your bracket is not set in the next line. You have not 3 spaces in befor your statement.	11:41
	→ lb joined	11:41
	← Miika-- left	11:41
	→ jexmex_ joined	11:42
	← Per4u3e left	11:42
FauxFaux	ChrisLE: Sure. You can run arbitrary code during hooks. Make them do whatever you want.	11:42
ChrisLE	ok, it calls git hooks?	11:42
	helium-3 → dilfridge	11:42
FauxFaux	Yes.	11:42
	→ Per4u3e joined	11:43
ChrisLE	ah ok, only code or maybe executeables?	11:43
	← ccc left	11:43
	← ketas left	11:44
FauxFaux	What?	11:45
ChrisLE	these rules are written in what? c, perl, python or what? And can I invoke a analyser.exe for example?	11:46
FauxFaux	It's a shell script (or maybe anything #!able? I have no idea.), so, yeah, you can just directly call analyser.exe.	11:47
ChrisLE	ah ok thx :)	11:47
	→ guampa joined	11:47
ChrisLE	can I bypass each hook or have I set it to my self?	11:47
	← J0rd4n left	11:49
	→ linusoleander joined	11:49
wereHamster	a hook can be anything that can be any executable	11:49
	git simply exec()s the file	11:50
ChrisLE	ok thx I will take a look	11:51
	→ bartek joined	11:51
	→ mischat_ joined	11:52
	← burgov left	11:53
	← dr_lepper left	11:54
	← mischat left	11:54
	mischat_ → mischat	11:54
	→ cannonball joined	11:55
	← dv310p3r left	11:56
	→ RandalSchwartz joined	11:56
	← RandalSchwartz left	11:56
	→ RandalSchwartz joined	11:56
	L_DM-eating → Lord_DeathMatch	11:58
	→ LmAt joined	11:58
	→ J0rd4n joined	11:58
	← J0rd4n left	11:58
	→ J0rd4n joined	11:58
	→ dr_leppe1 joined	11:59
	← davidbanham left	11:59
	→ replore_ joined	12:00
	← _Qman left	12:00
	→ n1x joined	12:00
	→ jackh joined	12:00
	→ _iron joined	12:01
	→ felgru joined	12:01
	→ mintux joined	12:02
	→ lihaitao joined	12:02
mintux	I want to push to remote over http but I got this error : error: unpack failed: unpack-objects abnormal exit	12:02
	→ master-lincoln joined	12:02
	→ burgov joined	12:02
	← ChrisLE left	12:03
	← Gentlecat left	12:03
	→ edude03 joined	12:05
	→ _Qman joined	12:05
	← thierryp left	12:05
	→ gusnan joined	12:06
	→ thierryp joined	12:07
	→ i10neorg joined	12:07
	← n1x left	12:07
	← tsdh left	12:07
	← ipalaus_ left	12:08
	← jackh left	12:09
	← jexmex_ left	12:09
	→ davidbanham joined	12:11
	→ leavittx joined	12:12
	→ jn_ joined	12:13
	→ keller joined	12:14
	→ marhaban1 joined	12:14
jn_	Hi me and my friend is resting out git... we have two branches A and B if I push to branch B and my mate switched to B and do pull, it tells him repo is already up to date? Im using linux and he's using windows if it matters.. repos are on github	12:14
	testing *	12:14
	← marhaban left	12:15
	← jalcine- left	12:15
	← thorpe_ left	12:16
adhawkins	jn, do you both have the branches set to track the remote branch (at github)?	12:17
	→ Briareos1 joined	12:17
Briareos1	is there a way to auto generate simple commit messages ? e.g. "hello.txt was deleted ; 4 lines changed in foo.txt ..."	12:17
cbreak-work	Briareos1: sounds stupid. Can't you just use --stat or --numstat?	12:18
	→ thorpe joined	12:18
	gitinfo set mode: +v	12:19
thorpe	Anyone know any good resources on svn -> git migrations? Ive read a few but am still researching.	12:19
	← bobsleer left	12:20
cbreak-work	man page for git-svn? :)	12:20
Briareos1	cbreak-work: never heard of those - how do you enter them (not in manpage)	12:21
thorpe	Yeah, ive been using git-svn as a client for 6 months or so.	12:21
jn_	<adhawkins> sorry can you tell me how I can check up on that ?	12:21
adhawkins	git remote -vv	12:21
thorpe	Just looking for some more info on a complete migration from svn to git.	12:21
adhawkins	It will list your branches, along with any remote branches they track	12:21
jn_	on branch B i do your command and i get origin for pull and fetch	12:22
adhawkins	Sorry, git branch -vv	12:22
	My mistake	12:22
	← marhaban1 left	12:22
jn_	ah I think they differ	12:23
adhawkins	For example, I get:	12:23
	* schema-change-2012-04 90fb1a5 [origin/schema-change-2012-04] Add Number	12:23
	That shows my local schema-change-xxx branch is tracking the same one on origin	12:23
jn_	my mate don't have origin when using that command, only me	12:23
adhawkins	If they're not, then you have two branches, one locally and one on origin	12:23
	→ bobsleer joined	12:23
adhawkins	Have him do	12:23
	→ chronos joined	12:23
adhawkins	git branch -u origin branchname	12:23
	← fuzzybunny69y left	12:24
adhawkins	In the branch	12:24
	I think that should resolve it?	12:24
	→ ketas joined	12:24
	← nrichards left	12:24
	→ g00s joined	12:24
	← chronos left	12:24
jn_	git complains about the -u i think	12:24
	think *	12:24
	error: unknown switch `u'	12:25
adhawkins	My head isn't working today	12:25
	→ Yuuhi joined	12:25
adhawkins	git push -u origin branch	12:25
jn_	hehe :)	12:25
	→ camdenx joined	12:25
	← Deck` left	12:25
bobsleer	why does git reports (for 1 of my 100 repo's) inconsistent tags list, between git clone and git fetch (from a remote)? I could provide more info with a public test repo.	12:25
	there's 1 tag not available after the git fetch (from remote).	12:26
jn_	my mate gets an : rejected non-fast-forward	12:26
wereHamster	bobsleer: git fetch --tags	12:26
	jn_: !non-ff	12:26
bobsleer	wereHamster: i try now	12:26
wereHamster	!nonff	12:26
	!ff	12:26
gitinfo	A fast-forward is a special type of merge where you have a revision and you are "merging" another branch's changes that happen to be a descendant of what you have. In such these cases, you do not make a new merge commit but instead just update to his revision. This will happen frequently on a remote-tracking branch of a remote repository. (From man gitglossary) See also:	12:26
adhawkins	jn_, he might need to explicitly merge the remote branch first	12:26
wereHamster	why is there no help for !non-ff ?	12:27
	→ marhaban joined	12:27
jn_	Can my mate just reset only that branch to head ?	12:28
	← oehman left	12:28
et	wereHamster: because no one created it - at least that's my guess	12:28
bobsleer	wereHamster: holy f*ck.. why without --tags, a part of tags are available?	12:29
	wereHamster: also thanx for your help	12:29
	← camdenx left	12:29
cbreak-work	only those that point to things you're fetching anyway afaik	12:30
	→ jsoa joined	12:30
	→ sentinal8473 joined	12:30
	→ d0k joined	12:30
	← mintux left	12:31
	← mischat left	12:31
	→ zealiod joined	12:31
zealiod	whats the phrase i use in gitolite.conf to output debug info?	12:31
bobsleer	wereHamster: never mind, I read the man now	12:31
	← tango_ left	12:32
	→ ehsan joined	12:33
	→ tango_ joined	12:34
	← marhaban left	12:35
	→ jas4711 joined	12:35
	← harshpb left	12:36
cbreak-work	Briareos1: they should be in man git-log	12:38
gitinfo	Briareos1: the git-log manpage is available at http://jk.gs/git-log.html	12:38
cbreak-work	or man git-diff	12:38
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	12:38
	← BiggFREE left	12:41
	← papegaaij left	12:42
	→ cchivers joined	12:42
	← ripero left	12:44
	→ TrahDivad joined	12:44
	→ crane joined	12:45
	→ Bass10 joined	12:45
	→ freeed joined	12:46
	→ ceej joined	12:46
	gitinfo set mode: +v	12:46
crane	hi, im currently working in a branch but now i noticed a bug in master and want to fix it. but i dont want to commit... how can i handle this?	12:46
	→ oriba joined	12:47
	← bobsleer left	12:47
	← g00s left	12:47
	→ ipalaus_ joined	12:47
	→ cebewee joined	12:48
cebewee	hi everyone	12:48
gitinfo	cebewee: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	12:48
et	crane: option a) you commit, but don't push that branch, and amend that commit later	12:48
cebewee	is there a way to fetch an unreferenced commit from another (local) repository	12:49
crane	et: i remember some kind of cache where this "pre-"commit could be saved...?	12:49
et	crane: or you stash your changes, fix your stuff, then apply the stuff from stash	12:49
crane	yes... that was what i meant :D	12:50
	thanks	12:50
	← leavittx left	12:50
crane	i forgott the stash ^^	12:50
	← eightyeight left	12:50
	→ jacky joined	12:51
	→ iocor joined	12:51
	← txomon left	12:52
	← zealiod left	12:53
cebewee	(the commit is not actually unreferenced, but in a non-standard location: refs/testboard/$NAME)	12:53
	← FernandoBasso left	12:53
	→ macabre joined	12:53
	→ eightyeight joined	12:56
	→ kukks joined	12:56
	→ harish_ joined	12:57
	← ipalaus_ left	12:57
	→ n1x joined	12:58
cebewee	sorry, it seems i had some typo in there before -- now it works in the expected way	13:00
	→ glennpratt joined	13:00
	→ warthog9 joined	13:01
	→ neurodrone joined	13:01
	← kaen left	13:02
	← burgov left	13:03
	← thierryp left	13:03
	← Pugnax left	13:04
	→ herdingcat joined	13:04
	← rtuin left	13:05
	← kpreid left	13:05
	→ thierryp joined	13:05
	← eerdogan left	13:06
	→ eerdogan joined	13:06
	← _iron left	13:07
	→ Deck` joined	13:07
	← bartek left	13:08
	← eerdogan left	13:08
	→ eerdogan joined	13:08
	→ dekroning joined	13:08
	→ jexmex_ joined	13:08
	← master-lincoln left	13:09
	← shruggar left	13:10
	→ bglusman joined	13:10
	← jexmex_ left	13:10
	→ shruggar joined	13:11
	→ revagomes joined	13:11
	← SicoWork left	13:11
	← herdingcat left	13:11
	→ the_cybersphinx joined	13:12
	→ isao joined	13:12
	→ cakehero joined	13:12
	← _Vi left	13:13
	→ OOPMan joined	13:13
	→ t4nkd joined	13:14
ishi	hi :)	13:14
	← cybersphinx left	13:15
ishi	I'm trying to fix something that was rejected in review in gerrit. I execuded the command gerrit suggested (the fetch && checkout).	13:16
	→ marhaban joined	13:16
	→ dv310p3r joined	13:16
ishi	next I did the checkout -b fix (as suggested by git)	13:16
	→ cybersphinx joined	13:16
ishi	I did the 'fixing'	13:16
	← jas4711 left	13:16
	← dekroning left	13:16
	← selckin1 left	13:16
ishi	next, I did the commit --amend -a	13:16
	so far so good, or at least I think so ;)	13:17
	→ selckin joined	13:17
ishi	now I'd like to push the changes so that they can be reviewed again...	13:17
	so I do: git push origin fix:/refs/for/master	13:17
	which doesn't work	13:18
	← shame left	13:18
	← cebewee left	13:18
cmn	take a look at the refspec again	13:18
ishi	I get: [remote rejected] fix_pom -> /refs/for/master (not valid ref)	13:18
cmn	it's not a path	13:18
ishi	yeah... now I only need to find out which bit is the 'path' and wich is 'refspec' ;)	13:19
cmn	there is no path	13:19
	← w09x left	13:19
	→ amigojapan joined	13:19
	← iocor left	13:20
	← the_cybersphinx left	13:20
	→ vectorshelve joined	13:20
ishi	I'm rading man git-push	13:20
gitinfo	the git-push manpage is available at http://jk.gs/git-push.html	13:20
	→ style joined	13:20
cmn	you're thinking too complicated, take a look at the refspec you gave push	13:21
	there's an extra slash	13:21
vectorshelve	I have been having changed files in my rep for a long time.. is it possible to list the files(changes) that have been done post a particular date ?	13:21
ishi	you mean the leading one?	13:21
style	Hi, any ideas why git clone works when I do it locally (git clone /var/git/asd) but when I do it over ssh (git clone ssh://user@server/var/git/asd) it says error: refs/remotes/origin/master does not point to a valid object!	13:21
ishi	wow... thanks	13:21
	now... I how I could 'get it' from the 'error message'... I don't know ;)	13:22
cmn	it tells you it's not a valid ref	13:23
vectorshelve	cmn: ^^ help	13:23
	→ master-lincoln joined	13:23
cmn	so you need to take a look at that part of the ref that gerrit would care about	13:23
ishi	so now I could go back to my work, by doing git checkout master, wait for the 'patch' to be reviewed, right?	13:23
	→ Geralt joined	13:23
ishi	cmn: well.. but if I put there foo/bar/goo, it'd just as well tell me that it's wrong	13:23
	in the same way	13:24
	→ codewrangler joined	13:24
cmn	right	13:24
	→ jrdavid joined	13:24
cmn	hm, does it have the same error string?	13:24
	I'd expect it to say that it doesn't exist	13:25
ishi	hmm...	13:25
	I'm affraid to check ;)	13:25
	I wouldn't be able to 'fix my workspace' if it, for some reason, worked ;)	13:26
	→ juanfer joined	13:26
	→ ericmuyser joined	13:26
	→ yoklov joined	13:26
	→ brngardner joined	13:27
	← ericmuyser left	13:27
cmn	style: the first clone isn't a real clone, it just copies some stuff over in the filesystem	13:27
	← kermit left	13:27
cmn	you should run git fsck on the repo	13:27
	→ nrichards joined	13:27
	→ r0bby joined	13:28
	← yoklov left	13:28
	gitinfo set mode: +v	13:28
vectorshelve	I have been having changed files in my rep for a long time.. is it possible to list the files(changes) that have been done post a particular date ?	13:28
	→ the_cybersphinx joined	13:29
cmn	you can tell log when it should stop listing things	13:29
	but dates aren't very important for git	13:29
	→ r0bby_ joined	13:29
	← dcarper left	13:30
	← beautifulmind left	13:30
cmn	once you get the commit you want from log, you can use git diff --name-only <thatcommit>	13:30
	← robbyoconnor left	13:31
	← pro_metedor left	13:31
	← jeby1 left	13:32
	← kami8785 left	13:32
	→ yizumi joined	13:32
	→ kermit joined	13:32
	→ tyatpi joined	13:33
	← cybersphinx left	13:33
	← r0bby left	13:33
jn_	Can I reset the branch Im standing on to master head, and only that branch ?	13:34
	→ AaronMT joined	13:34
style	cmn: fsck just says that threre is one dandling commit	13:34
	cmn: git clone over ssh works on windows btw, but not in other linux	13:35
cmn	that's really odd	13:35
style	indeed	13:35
cmn	maybe with GIT_TRACE=1	13:35
style	cmn: output is the same as previous	13:36
cmn	right	13:36
vectorshelve	cmn: no in my case the files have been in changes state for a long time and not commited.. now I need the recently edited files among them	13:36
cmn	but the important thing is the debug output	13:36
saml	hey, what's good branching stuff? isn't gitflow so complicated?	13:37
cmn	vectorshelve: then use some script around ls -l or stat or something	13:37
	that's really not a git problem	13:37
vectorshelve	cmn: any command for it	13:37
style	cmn: I can't get any debug output	13:37
cmn	there is no command, I'd use some combination of ls, sort and awk	13:37
vectorshelve	cmn: k thanks man	13:37
	→ friskd_ joined	13:37
	← cchivers left	13:38
cmn	style: can't meaning what?	13:39
	did you set GIT_TRACE=1?	13:39
style	in terminal yes, and after that I ran clone command	13:39
	← sfiku left	13:40
	→ iocor joined	13:40
cmn	after as in after pressing enter?	13:40
	→ mrwalker joined	13:40
	← friskd left	13:40
	friskd_ → friskd	13:40
style	oh sorry, I forgot to export it	13:40
	← yizumi left	13:40
	← linusoleander left	13:41
amcsi_work	hello	13:41
gitinfo	amcsi_work: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	13:41
	← Geralt left	13:42
saml	hey amcsi_work	13:42
amcsi_work	git diff shows no modifications in my file, but git status shows a bunch with line ending errors	13:42
ishi	bugger... say, someone -2-ed my changeset. I fixed and pushed it (thanks cmn). Now on gerrit he +2-ed it, but it doesn't seem to 'catch on'	13:42
amcsi_work	maybe because I didn't convert my local files to dos yet	13:42
	how do I diff --name-only with also listing the line-ending different files?	13:43
cmn	ishi: catch on as in what? did you use a Change-Id?	13:43
amcsi_work	because it isn'T listing them	13:43
saml	can i checkout a tag?	13:43
	← JohnyBaby left	13:44
cmn	you can checkout any commit	13:45
bremner	but beware !detached	13:45
gitinfo	A detached HEAD 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	13:45
cmn	if that tag points to a commit, you can also checkout that tag	13:45
	→ rtuin joined	13:45
saml	so, i'm trying to have one remote master branch that's stable enough for production release.. and i tag the branch whenever i build a release candidate. when i need to release a hotfix, i checkout a tag, patch, and release?	13:46
	→ cybersphinx joined	13:46
cmn	no	13:46
saml	or do i create master and release branch on remote?	13:46
cmn	you commit on top of the release branch you have for that version	13:46
saml	development pushes to master and when things are stable merge master to release?	13:46
	so, remote has two branches: master and release	13:47
	each release gets a branch?	13:47
ishi	the change didn't become 'MERGED'... but it seems to be 'OK' now... I don't know what happened ;)	13:47
cmn	saml: each supported release should get a branch	13:48
	→ karsten joined	13:48
saml	i release daily.. so create a branch daily?	13:48
	→ w09x joined	13:48
	← the_cybersphinx left	13:49
	→ ISF joined	13:49
	→ jackh joined	13:49
cmn	do you release a completely new version every day?	13:49
	→ dcarper joined	13:49
cmn	ok, so replace 'version' by 'minor version'	13:49
	→ JohnyBaby joined	13:50
	← vectorshelve left	13:50
saml	i mean.. it's a web application.. tickets come in.. developers close ticket and release	13:50
cmn	then you do hot-fixes either on the stable branch or a branch off of that one	13:50
	and merge into the stable branch and whatever main dev branch you have	13:51
saml	yah	13:51
	→ TheHaven joined	13:52
amcsi_work	I want to have all the files listed in git status as changed so I can pipe and xargs them into a unix2dos	13:52
	but git diff --name-only isn't showing the files with only name changes	13:53
cmn	what are name changes?	13:53
PerlJam	--name-only doesn't do what you think it does :)	13:53
cmn	the -only is the counterpart to -status	13:54
	→ maxamillion joined	13:54
ishi	ok, now, if I understand it, I created a branch 'fix', but now I merged it to my 'master'... why do I still see branch 'fix' when I do git branch?	13:54
	→ w09x_ joined	13:54
cmn	because you never deleted it?	13:54
maxamillion	anyone know why running 'git branch' and 'echo $(git branch)' offer up such different output? and is there a way to make them the same?	13:54
	→ sebrock joined	13:54
ishi	how do I get rid of it... it was only 'temporary' branch for fixing the commit...	13:54
cmn	man git branch -d	13:54
gitinfo	the git-branch manpage is available at http://jk.gs/git-branch.html	13:54
cmn	maxamillion: different in what way?	13:55
	→ SkAzZ joined	13:55
	← w09x_ left	13:55
maxamillion	cmn: did you run them?	13:55
cmn	yyes	13:55
ishi	cmn: do I want to do it? I mean... I don't want the changes from this branch to be lost, or something...	13:55
cmn	and the only difference is the one you're forcing on to it	13:55
	ishi: you merged it, no?	13:55
	then they're in the other branch as well	13:55
maxamillion	cmn: oh wait, nvm ... its the asterisk, the shell is globbing	13:56
ishi	I hope so, yes ;)	13:56
et	uh	13:56
ishi	ok, it seems to have worked ;)	13:56
amcsi_work	PerlJam: "--name-only	13:57
	Show only names of changed files."	13:57
et	maxamillion: switch to zsh, it doesn't do this madness ;) (at least not by default)	13:57
	ishi goes back to coding	13:57
	← seoaqua left	13:57
maxamillion	et: I've pondered it	13:57
	← DofLeFou left	13:57
maxamillion	anyhoo, thanks for entertaining my noob moment :D	13:57
	← maxamillion left	13:57
	→ seoaqua joined	13:57
	← satyanash_ left	13:57
cmn	maxamillion: why do you want to put branch through echo, anyway	13:57
	← OOPMan left	13:58
	→ BarryCrayon joined	13:58
	← seoaqua left	13:58
	gitinfo set mode: +v	13:58
	← w09x left	13:58
cmn	it's porcelain, and doesn't guarantee stable output, so it's not something you should use in a script	13:58
	→ seoaqua joined	13:59
	← seoaqua left	14:00
	→ seoaqua joined	14:00
	→ beautifulmind joined	14:00
	→ _Vi joined	14:01
	→ Gentlecat joined	14:01
style	cmn: problem was old version of git, it was 1.7.4 after upgrading to 1.7.10 it worked, thanks for your help!	14:01
	← master-lincoln left	14:02
	→ fission6 joined	14:03
	→ alansaul joined	14:03
	← ehsan left	14:03
	← edude03 left	14:03
	← marhaban left	14:04
cmn	style: that really shouldn't be a problem	14:04
	→ mritz joined	14:04
cmn	the wire protocol hasn't changed between them	14:04
	Amgine → quieropintardene	14:05
style	cmn: weird, but anyways upgrade helped so there is no problem anymore	14:05
cmn	is the repo public?	14:05
style	nope	14:05
cmn	ok	14:05
	← sentinal8473 left	14:06
	← eightyeight left	14:06
	← eerdogan left	14:06
	quieropintardene → qieropintarnegro	14:06
	→ pro_metedor joined	14:06
	← isao left	14:07
	→ marhaban joined	14:10
	→ CalimeroTeknik joined	14:10
	← iocor left	14:11
CalimeroTeknik	I was slowly browsing though folders to git cat-file a given file at a given commit, and wondered, is there a faster way?	14:11
	through*	14:12
	← oriba left	14:12
CalimeroTeknik	I had to run like, 15 git cat-file -p commands	14:12
cbreak-work	CalimeroTeknik: to do what?	14:12
CalimeroTeknik	it made me want to code a FUSE fs for git cat-file	14:12
	→ eightyeight joined	14:12
CalimeroTeknik	to cat a file but at its state at a given commit	14:13
	not at its current state	14:13
cmn	use show	14:13
	→ sagsousuke joined	14:13
	← replore_ left	14:13
cmn	git show commit:path	14:13
cbreak-work	git cat-file is the best tool for that	14:13
CalimeroTeknik	oh! thanks	14:13
	← ellisonch left	14:13
	← dr_leppe1 left	14:13
	← chepre left	14:13
cbreak-work	git show is the same with a similar syntax	14:13
	(for files)	14:13
	→ iocor joined	14:13
cmn	hm, yeah, nowadays, cat-file understands commit:path as well	14:13
kini	what's the point of having separate commands then	14:14
CalimeroTeknik	that's perfect, thanks	14:14
cbreak-work	cat-file is plumbing	14:14
	show is not	14:14
	they act differently for commits/trees	14:14
	and cat-file has more accessors	14:14
CalimeroTeknik	cat-file -p commit:file is like cat, show commit:file uses less	14:14
amcsi_work	please help me someone	14:14
	there has to be a simple way to do this	14:14
	← pro_metedor left	14:15
	← RaceCondition left	14:15
	→ RaceCondition joined	14:15
	→ dr_leppe1 joined	14:15
CalimeroTeknik	sorry amcsi_work but I didn't see a question; asking for help won't solve that	14:15
	← LmAt left	14:15
	→ satyanash joined	14:15
amcsi_work	how do I list all the files that are modified by name-only, and including the ones with line ending changes only	14:16
	becuase git diff --name-only doesn't list the files with only line-ending changes	14:16
	Ineed to change those files with piping to xargs unix2dos to fix the line ending issues	14:16
cmn	you'd need to inspect them	14:16
	← newbie999 left	14:17
cmn	or set up the automatic eof conversion	14:17
amcsi_work	there are like 100 files	14:17
cmn	or just pass all of them	14:17
canton7	git status -s \| grep M works for me...	14:17
amcsi_work	what automatic eof conversion? git is going to commit them in a different format which will show too many changes in the log	14:18
	→ pro_metedor joined	14:18
	→ beatak joined	14:18
amcsi_work	and how do I remove the M parts in them?	14:18
cmn	the automatic eof conversion git does if you tell it to	14:18
amcsi_work	autocrlf is on, if that's what you're saying	14:19
	← Vile1 left	14:19
cbreak-work	you don't need dos2unix if you have autocrlf	14:19
cmn	he's using unix2dos though	14:19
	for whatever reason	14:19
cbreak-work	that'll be useless too	14:20
	since git will convert it back	14:20
cmn	right	14:20
	→ Vile joined	14:20
	cbreak-work disables all this line ending change stuff	14:20
	→ edude03 joined	14:20
cmn	the real solution is to fix the editor	14:20
	→ xsbeats joined	14:20
	← ThomasLocke left	14:21
cmn	so yeah, disable the conversion, and pass your files through that filter	14:23
	→ airborn joined	14:23
cmn	you can probably convince grep to search for lack of ^M	14:23
cbreak-work	or just filter everything? :)	14:23
	→ syphar joined	14:23
cmn	that assumes that the filter is smart enough not to turn \r\n into \r\r\n	14:24
	← alansaul left	14:24
	→ Jackneill joined	14:24
	→ master-lincoln joined	14:25
amcsi_work	My editor is fine. The editor of the people who uploaded the files to the ftp suck.	14:25
	→ camdenx joined	14:25
cmn	so you do have a place where you can know which ones are wrong?	14:27
	→ martinjlowm-lptp joined	14:27
amcsi_work	I converted a unix file that was marked as modified to dos, and now git doesn'T show it as modified	14:28
	← camdenx left	14:28
amcsi_work	which is great	14:28
	now I want to do the same for the other 100 versioned, text files	14:28
	→ felgru_ joined	14:29
cbreak-work	files are supposed to be in unix format	14:29
wereHamster	for file in <all files>; do ...; done	14:29
	→ chrisgeorge joined	14:30
amcsi_work	yea, the <all files> is the part I'm stuck at. please help me with that	14:30
wereHamster	so.. all versioned files?	14:31
amcsi_work	yes	14:31
	← mritz left	14:31
cbreak-work	ls-files	14:32
	← felgru left	14:32
	← edude03 left	14:32
	→ dfr\|lghtninghall joined	14:34
	← marhaban left	14:34
	→ FernandoBasso joined	14:34
	→ ipalaus_ joined	14:36
	→ resmo__ joined	14:36
	← gusnan left	14:36
	← Lord_DeathMatch left	14:37
amcsi_work	thank you!	14:37
	← ben_alman left	14:38
	→ ehsan joined	14:38
amcsi_work	great, and now it only shows the real modified files as modified files now, cool	14:38
	→ newbie999 joined	14:38
	← dcarper left	14:39
amcsi_work	btw, does core.autocrlf = true put files as LF in my repository automatically?	14:39
	new files	14:39
	or does only core.autocrlf = input do that?	14:39
	→ harshpb joined	14:39
	← bgerber left	14:39
	← thebignoise left	14:39
cmn	the actual rules are a bit more complicated and you're referring to the old way of doing things	14:39
	← jmd left	14:39
	← ISF left	14:39
cmn	if you want conversion, you should use attributes	14:40
	← master-lincoln left	14:40
	← resmo left	14:40
	← JohnyBaby left	14:40
amcsi_work	could you show me a simple way of making it so automatic conversion would be made to all _newly_ added files and to make them be LF in the repo?	14:41
	← iocor left	14:41
	← lihaitao left	14:42
	→ leavittx joined	14:42
	← gigamonkey left	14:42
	← frogonwheels left	14:42
	→ JohnyBaby joined	14:43
	← _Qman left	14:44
	← vervic left	14:44
	← RaceCondition left	14:45
	→ jatt joined	14:45
	→ Fullmoon joined	14:45
amcsi_work	you say gitattributes? If so, do I have to give every single type of text file extension and tell them each to be LF?	14:46
cmn	new files aren't special	14:46
	→ ben_alman joined	14:46
	← Guest94016 left	14:46
	→ zeom joined	14:46
cmn	you can put '* text' and then say which ones aren't	14:46
amcsi_work	then I'd have to specify all the different kinds of binary files	14:47
	that is a lot of manual work, and is error prone	14:47
	I like git's built in binary detecting	14:47
	← arvind_khadri left	14:47
	→ bgerber joined	14:48
	← tuopppi_ left	14:49
	→ RaceCondition joined	14:49
	→ iocor joined	14:50
amcsi_work	do you disagree?	14:51
	← Altreus left	14:52
	← beautifulmind left	14:52
	→ pierrelechelle joined	14:53
	→ bartek joined	14:54
	→ tsdh joined	14:55
	→ eph3meral joined	14:55
	← borwar left	14:55
eph3meral	so it appears we've lost a fair bit of data somehow, neither I nor my co-worker are quite sure how it happened or who's responsible, but we think it's probably because one of us (likely me) did a rebase and a force push without first pulling his changes	14:55
	← juanfer left	14:56
	tizzo-afk → tizzo	14:56
	→ juanfer joined	14:56
	→ orspeeder joined	14:56
	← orspeeder left	14:56
	← crooksey left	14:57
canton7	that would probably be it... force pushes are generally a bad idea in a shared situation	14:57
	→ speeder joined	14:57
eph3meral	he has since deleted his branch and so doesn't appear to have the history	14:57
	and neither do i	14:58
speeder	I saw on some git sites, the expression "pull request". What it means?	14:58
	qieropintarnegro → Amgine	14:58
	← airborn left	14:58
eph3meral	progit.org says that unless you try really hard, git still has all your info	14:58
	is this true? and how would I go about trying to find this stuff? or is it in fact just "gone" ?	14:58
canton7	eph3meral, so he pushed some data and you overwrote it? The branch he deleted is probably in his reflog somewhere	14:58
	man git-reflog or !gka on his computer	14:58
gitinfo	the git-reflog manpage is available at http://jk.gs/git-reflog.html	14:58
	For a better way to view the reflog, try: gka() { gitk --all $(git log -g --format="%h" -50) "$@"; }; gka	14:58
eph3meral	ok thanks	14:59
	← JackWinter left	14:59
amcsi_work	what does autocrlf = input do? Does that automatically enforce LF in the repo? Because I'd rather use that in windows then	14:59
canton7	eph3meral, or you can try git fsck --lost-found on the server, look for dangling commits, and use git show / git cat-file -p to look at them. See if you can find the commit which was at the tip of the branch before it was rewritten	14:59
	← Per4u3e left	14:59
	→ Dragooon joined	14:59
canton7	speeder, conceptually, it's where I clone your repo, commit some changes, and send you a message saying "pull this repo, branch x, to get some awesome new changes"!	15:00
	speeder, sites like github formalise this, and allow you to send a message to someone else through github with all the required details filled in automatically	15:01
eph3meral	canton7, ok thanks I'll try that stuff	15:01
	canton7, oh and by the way "on the server" essentially means github.com	15:01
	→ JackWinter joined	15:02
canton7	eph3meral, ah, don't think there's a way to do any recovery server-side then -- at least I haven't come across a way to with github	15:02
	← dpino left	15:02
	→ slikts joined	15:02
eph3meral	canton7, ok, so I guess my only option is to check the reflog on his machine then eh?	15:03
	and $@ is the params list right?	15:03
	are there some other common params?	15:03
	as in, isn't $@ the argv array that gets passed to bash?	15:03
	(or rather that bash passes to a function?)	15:03
speeder	canton7: oh nice	15:04
canton7	eph3meral, yup. and yes, that's what the $@ is -- it defines a function gka, then passes any arguments given to gka to the underlying gitk command. Tbh you probably don't need to pass any args though	15:04
	→ theCole joined	15:05
	→ oriba joined	15:07
	← cooloney left	15:07
	← pmatthews left	15:07
	← neurodrone left	15:07
	← davidbanham left	15:08
	← ajpiano left	15:08
	← angelsl left	15:08
	→ jsoa_ joined	15:08
	← cbreak-work left	15:09
eph3meral	canton7, hmmm... so... git log = "the reflog" ?	15:10
	→ ajpiano joined	15:10
bremner	no	15:10
	← JohnyBaby left	15:10
bremner	eph3meral: man git reflog	15:10
gitinfo	eph3meral: the git-reflog manpage is available at http://jk.gs/git-reflog.html	15:10
eph3meral	bremner, yes, apparently you haven't been following our conversation	15:10
canton7	eph3meral, no. git-reflog = the reflog. git log can also get at the reflog, though, with -g	15:10
eph3meral	ahhh, -g is the reflog ok cool	15:10
bremner	eph3meral: well, missing context or not, git log != reflog	15:10
pierrelechelle	Hey guys, I've a little problem with Git (and Github), I'm trying to get directory on git without any files. I've create empty file in each directory, and Git still doesn't see directories. I can't get my gitignore to work (file : http://pastebin.com/ACp0sSwM)	15:11
	→ bumb joined	15:11
	← jsoa left	15:11
	jsoa_ → jsoa	15:11
	← thiago left	15:11
PerlJam	pierrelechelle: you want to put a non-empty file in each dir.	15:12
	← bumb left	15:12
PerlJam	pierrelechelle: usually that's .gitignore that ignores everything	15:12
	← luka12346 left	15:14
	→ luka12345 joined	15:14
	← tsdh left	15:14
	← resmo_ left	15:16
	→ bumb joined	15:16
	→ reels joined	15:17
	→ mikepack joined	15:18
	→ ToBeFree joined	15:18
	→ Mowsh joined	15:18
	→ ellisonch joined	15:18
	← Guest38096 left	15:18
	← bumb left	15:18
	→ hennekey joined	15:19
pierrelechelle	PerlJam: I've put an empty file in each directory and it's name is "empty", I've add a rule to the gitignore for that (!empty)	15:19
PerlJam	pierrelechelle: read what I wrote again :)	15:19
pierrelechelle	PerlJam: Can you explain it differently ? My poor english can't understand :D	15:20
	← crane left	15:20
	← style left	15:20
PerlJam	pierrelechelle: to store the directory structure in git, you need non-empty files that git is not ignoring to be in each directory.	15:21
	gitinfo set mode: +v	15:22
	← harshpb left	15:22
	→ jim0203 joined	15:22
	→ JohnyBaby joined	15:22
pierrelechelle	PerlJam: I understand now ! But Git is not ignoring my empty file in cache and logs directory :s	15:23
	← thierryp left	15:24
	→ kadoban joined	15:24
	→ LmAt joined	15:24
	→ isao joined	15:24
	→ TeckniX joined	15:25
	hennekey → _hennekey	15:26
	← ToBeFree left	15:26
	_hennekey → hennekey	15:27
amcsi_work	how do I delete the git status cache?	15:27
	← cakehero left	15:27
	→ ToBeFree joined	15:27
FauxFaux	amcsi_work: Is there a git status cache?	15:28
amcsi_work	I dunno, but diff shows no difference in the files, but status says there is	15:28
	→ rockets joined	15:28
	→ aldodelgado joined	15:28
FauxFaux	amcsi_work: Probably autocrlf being terrible.	15:29
	→ bumb joined	15:29
	→ TomyLobo joined	15:29
amcsi_work	well, I turned it off	15:29
	← bumb left	15:30
	← RaceCondition left	15:30
amcsi_work	I don't know what I have to change to make the file not be considered modified	15:30
	→ gigamonkey joined	15:30
amcsi_work	when it's unix, it there is no diff, but the file is considered modified. when I convert it to dos, it shows all lines changed, and the file is considered modified	15:31
	→ harshpb joined	15:31
	← oriba left	15:31
amcsi_work	what am I supposed to do (other than checkout) to fix this?	15:32
	← jackh left	15:32
	← hyperair left	15:32
	← legnaleurc_ left	15:32
	→ RaceCondition joined	15:33
	→ thearthur joined	15:33
	→ gitn00b joined	15:33
	→ drvanon joined	15:34
	→ legnaleurc joined	15:34
	gitinfo set mode: +v	15:34
	← amcsi_work left	15:34
pierrelechelle	PerlJam: Any idea ?	15:34
gitn00b	Is there a way to tell git that even though a file changed a lot across a rename, it's still the same file? The only way I can think of is to make two commits, one with all the changes, and a second one with the file rename.	15:34
	→ yoklov joined	15:35
	→ amcsi_work joined	15:35
	← dfr\|lghtninghall left	15:35
FauxFaux	gitn00b: Yep, that's the only way.	15:35
amcsi_work	I fsomeone answered me, I didn't receive anything, because I was disconnected	15:35
gitn00b	FauxFaux: Sigh, I wish you could explicitly mark renames. Oh well.	15:35
	← mishok13 left	15:36
	→ Radium joined	15:36
FauxFaux	gitn00b: Come up with a way to store that information that fits within the existing git model, all the code to get log / diff / etc. to honour it, then convince over half of the mailing list that what you want makes sense, then!	15:36
offby1	gitn00b: actually I like that git doesn't "know" a file's name.	15:36
	makes it simpler.	15:36
gitn00b	offby1: Two commits are better than one, I guess.	15:37
	← mbreit_ left	15:37
FauxFaux	I still maintain that the detection of splitting of files sucks, but I can't be bothered to discuss it again.	15:37
	→ hyperair joined	15:38
	← isao left	15:38
	→ Targen joined	15:39
	← harshpb left	15:39
gitn00b	Speaking of git models, this reminds me: why doesn't git store a global commit index?	15:40
yoklov	So, I committed some code in master which I didn't end up liking, checked out a previous commit, branched from it, how can I get this back into master without ruining everything?	15:40
gitn00b	An index of all commits?	15:40
	→ pmatthews joined	15:40
	← dvide left	15:40
	← rchavik left	15:40
	← xsbeats left	15:41
bremner	gitn00b: is this another mercurial troll?	15:41
gitn00b	This would eliminate: (1) detached HEADs (2) lost commits (3) slowdown to find a commit very far from a branch ref (4) bare repos	15:41
	→ irqq joined	15:41
gitn00b	bremner: I'm just wondering why does git not do this.	15:41
Menno	gitn00b: what's wrong with two commits?	15:42
	changes+renames are a nightmare in any VCS	15:42
gitn00b	Menno: No, not in any VCS. Only in some, like git.	15:42
	Menno: But two commits aren't too bad, agreed.	15:42
	← dr_leppe1 left	15:42
	→ neurodrone joined	15:42
Menno	The ability to do it in one commit comes at a high price.	15:43
	→ Targen_ joined	15:45
	→ thiago joined	15:46
	tizzo → tizzo-afk	15:47
	← Targen left	15:48
	← lb left	15:48
	→ mishok13 joined	15:49
	→ dr_lepper joined	15:49
	← dr_lepper left	15:49
	→ dr_lepper joined	15:49
	→ fyolnish joined	15:49
	→ maestrojed joined	15:50
	← `Assassin left	15:50
	← rtuin left	15:50
	e-ndy → e-ndy\|afk	15:51
	→ jpalmer joined	15:51
	← filier left	15:51
	← jceb left	15:51
yoklov	figured it out, git branch -f master was what i was looking for.	15:51
	→ ology joined	15:51
	← fling left	15:52
	← enmand left	15:52
	← jn_ left	15:52
	← Radium left	15:53
	→ hwrd\|work joined	15:53
	→ harshpb joined	15:54
	← RaceCondition left	15:54
	→ JEntrep joined	15:54
	→ enmand joined	15:54
	← dvgsvc left	15:55
	→ ronniy joined	15:55
	← brngardner left	15:56
	→ chrisshattuck joined	15:57
	← gitn00b left	15:57
	→ Icedream joined	15:57
	→ jeshwanth joined	15:57
	Icedream → Guest37186	15:58
	← Guest37186 left	15:58
	→ Guest37186 joined	15:58
	Guest37186 → Icedream	15:58
	→ master-lincoln joined	16:00
	→ raatiniemi joined	16:00
	← jeshwanth left	16:02
amcsi_work	By the way, core.autocrlf converts all new files' line endings to lf in the repo	16:02
	so it's pretty safe and good	16:03
	→ jeshwanth joined	16:03
	→ macmartine joined	16:03
	← JohnyBaby left	16:03
amcsi_work	what's great about it is that when I see there are line ending problems, I know that I just have to convert all my files to end with CRLF on windows	16:03
	← seoaqua left	16:03
amcsi_work	instead of checking per file which kind of line ending it should have	16:04
	→ wilmoore joined	16:04
Vinnie_win	What do I do to get git onto my mac?	16:04
amcsi_work	g2g bye	16:04
	← harshpb left	16:04
shruggar	Vinnie_win: "port install git" probably works	16:05
Vinnie_win	thanks I will try that	16:05
	← amcsi_work left	16:05
	← mishok13 left	16:05
Vinnie_win	"port: not found"	16:05
	← Davey left	16:05
	← hwrd\|work left	16:05
shruggar	install macports, then ;)	16:05
	→ harshpb joined	16:05
haarg	git is built in with newer versions of os x	16:05
Vinnie_win	strange because I'm on 10.7.3	16:05
	→ Araxia_ joined	16:06
haarg	/usr/bin/git	16:06
Mkaysi	Vinnie_win: from git homepage, http://code.google.com/p/git-osx-installer/downloads/list?can=3	16:06
	→ brngardner joined	16:06
haarg	or maybe it comes with xcode or something?	16:06
	← pierrelechelle left	16:06
	← yoklov left	16:06
cmn	Xcode bundles the dev tools, clang, git at least	16:07
	→ Davey joined	16:07
	→ dvgsvc joined	16:08
Vinnie_win	Mkaysi: I guess I will try that. I thought it was built in	16:08
	→ pandeiro joined	16:08
	← jeshwanth left	16:09
	← tchan left	16:09
	→ jeshwanth joined	16:09
	← irqq left	16:11
	← jeshwanth left	16:11
cmn	git isn't part of the standard packages of any OS	16:11
	hm, base packages might be a better way of saying it	16:11
bremner	it is in debian	16:11
	ok,	16:11
	← hennekey left	16:12
	← Davey left	16:12
bremner	but the distinction seems to be more important on MacOS than on a typical linux distro.	16:12
	← fyolnish left	16:13
FauxFaux	Really should talk some of the distros into shipping etckeeper, which depends on /a/ vcs, by some point.	16:15
	→ Davey joined	16:15
	→ Destos joined	16:16
	← harshpb left	16:16
	← leavittx left	16:16
	← aldodelgado left	16:16
bremner	FauxFaux: what do you mean by "shipping"? don't most distros make it available?	16:17
	← pro_metedor left	16:17
	→ pro_metedor joined	16:17
FauxFaux	Heh. I meant "enable".	16:17
haarg	i think it used to be included with os x by default. now it comes with xcode.	16:17
	← ShooterM_ left	16:18
	← ShooterMG left	16:18
	→ ShooterMG joined	16:18
	→ ShooterM_ joined	16:19
	→ camdenx joined	16:19
	→ diegoviola joined	16:19
	← pro_metedor left	16:19
	→ mritz joined	16:19
	← reels left	16:22
	→ starsinmypockets joined	16:22
	← BarryCrayon left	16:22
cmn	true, OSX does have some non-base Apple provided packages, and that includes git	16:24
	but most other packages, you'll have to build yourself using a third-party tool	16:25
	→ yoklov joined	16:25
	→ qt-x joined	16:25
	→ shirish joined	16:25
starsinmypockets	This is very basic but if pull from my upstream repo and add a branch (say, 'dev') - when I push this back up, it will ad the branch to the upstream repo?	16:26
	→ sayd joined	16:27
	tizzo-afk → tizzo	16:27
	→ maletor joined	16:27
cmn	add a branch as in create it locally?	16:28
	← ph^ left	16:28
cmn	the push semantics depends on what you've configured	16:28
	by default, a 'git push' won't try to create branches	16:29
	→ jceb joined	16:30
	← flijten left	16:30
	→ n2diy_ joined	16:30
	→ agile joined	16:30
	→ hydrozen joined	16:30
	← adhawkins left	16:31
	← LimeBlast left	16:31
cmn	starsinmypockets: ^^	16:33
	→ SegFaultAX\|work joined	16:33
	← n2diy left	16:34
	← asedeno_work left	16:34
	← Iota_ left	16:34
	→ Iota joined	16:34
Iota	I made a modification on a file without making a branch. I commited it so I was 1 commit ahead of the server. I then git checkout origin/master, how do I switch back to my other version? Using git branch it just shows "* (no branch)" and "master"	16:34
	→ asedeno_work joined	16:34
	← williamcotton left	16:35
cmn	and what did you want to do when you ran 'git checkout origin/master'?	16:35
Iota	I wanted to see if the files there had changed (stupidly remembering after that I didn't push them). I'm still learning. :(	16:36
cmn	in the remote-tracking branch? that gets updated by fetch	16:36
	you modified whatever branch you were on	16:36
Iota	Ah yes. So checking out a remote branch is never a good idea?	16:37
cmn	rarly	16:37
	neersighted\|AFK → neersighted	16:37
	← asedeno_work left	16:37
Iota	Ah, thanks. I've gotten back to where I was meant to be.	16:37
	← theCole left	16:38
	→ arvind_khadri joined	16:39
	→ dcarper joined	16:39
	← master-lincoln left	16:40
	→ vama89 joined	16:40
	← JohnBat26 left	16:40
	← dcarper left	16:41
	← felgru_ left	16:41
	→ dcarper joined	16:41
	→ mishok13 joined	16:41
	→ leavittx joined	16:42
	← TheHaven left	16:42
starsinmypockets	cmn - right, if I create a branch locally, will that be pushed up, or just the master (by default)	16:42
cmn	neither of them	16:42
	the default is 'matching' so it'll try to push any branches that have the same name	16:43
	take a look at man git config push.default	16:43
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	16:43
starsinmypockets	cmn I do have this in my config git config remote.unfuddle.push refs/heads/master:refs/heads/master	16:43
	→ thebignoise joined	16:44
	→ marc_kdab joined	16:44
cmn	I've never heard of that config option	16:44
	← knopkodav left	16:44
	← Dragooon left	16:44
cmn	and it doesn't exist on my version of the documentation	16:45
	→ ph^ joined	16:45
cmn	if it's an option for some third-party tool, git won't honour it	16:45
	← pmatthews left	16:45
	← poohtarou left	16:46
	← ed8 left	16:47
	← pandeiro left	16:48
	← thearthur left	16:48
	← gavin_huang left	16:48
	→ usmanity joined	16:49
	→ avinashhm joined	16:49
	← fir31ight left	16:49
	→ gusnan joined	16:49
	← Davey left	16:49
starsinmypockets	cmn interesting - it's unfuddle.com which is a project management tool... not sure what git's doing with it... I'll read those docs.. thanks	16:50
cmn	git's won't do anything with it	16:50
	→ sq-one joined	16:50
cmn	that tool might use git repos	16:50
	→ duckxx joined	16:50
	← shirt left	16:51
	← Poapfel left	16:51
	→ Poapfel joined	16:51
	neersighted → neersighted\|AFK	16:52
	← berserkr left	16:52
	→ Shusshu joined	16:52
	→ berserkr joined	16:54
	→ asedeno_work joined	16:54
avinashhm	Hi friends , how do we create a http weblink to the current git repo .. we are having a working git repo in a group of 4 to 5 ppl .. we want a web interface ... something similar to kernel.org .. what things i need to install so create http links ..	16:55
	← Mowsh left	16:55
	← vama89 left	16:56
	→ JohnyBaby joined	16:56
	← amigojapan left	16:57
	→ spaceonline joined	16:58
	→ Davey joined	16:58
	→ luxurymode joined	16:58
	→ elasticdog joined	16:59
cmn	kernel.org uses gitewb	16:59
	gitweb*	16:59
avinashhm	cmn, great .. thanks ... i will check gitweb and install ..	17:00
	← iocor left	17:00
	← eph3meral left	17:00
	← cedeon left	17:01
	← ipalaus_ left	17:01
	→ iocor joined	17:01
starsinmypockets	Is there a way to use git to show diffs between files and/or directories (not commits)?	17:01
cmn	just tell it what you want to diff	17:01
	→ nioncode joined	17:02
elasticdog	is there a plumbing command to see if a commit is necessary? http://hastebin.com/qeronupoga.sh	17:02
	I'm wanting to auto-commit things, but want to test and see if it's necessary after I've updated the index	17:03
wereHamster	elasticdog: check if working tree is dirty?	17:03
	test -z "$(git status -s)" && exit	17:03
	quick and dirty	17:03
elasticdog	that may work	17:04
wereHamster	of course it does. But it's still dirty	17:04
	← iocor left	17:04
	→ BadCodSmell joined	17:04
BadCodSmell	How do I ignore a folder's contents but not the folder its self?	17:05
elasticdog	wereHamster: yeah, I'm wondering if there's a command like status that will give exit codes based on whether or not there is anything staged	17:05
	← bolt left	17:05
canton7	there is something... let me find it	17:06
elasticdog	note I do have a typo in that paste /ls-file/ls-files/	17:06
	← PigeonFriend left	17:06
	← shruggar left	17:06
canton7	elasticdog, https://github.com/nvie/gitflow/blob/ab7fda21f1009d8cc073fc9ae6152731f6bfee8a/gitflow-common#L82 . It's not my code, and I don't know how well-advised those techniques are, but it's a start. a return value of 1 = unstaged changes, 2 = uncommitted changes	17:07
	→ iocor joined	17:07
elasticdog	BadCodSmell: is the trouble that it's not including the "empty" folder because of your ignores?	17:07
	canton7: thanks!	17:07
	← Dave^\| left	17:08
	← upasna left	17:09
	→ thelazydeveloper joined	17:09
cmn	BadCodSmell: git doesn't track directories	17:10
	← jatt left	17:10
	→ Targen joined	17:10
	→ JohnyBaby_ joined	17:10
	← yoklov left	17:10
	← Targen_ left	17:11
canton7	BadCodSmell, stick a .gitignore with the contents "*" in that folder, and track that .gitignore	17:11
elasticdog	BadCodSmell: in order to include the directory, you'd have to commit a hidden file or something under it	17:11
	← achadwick left	17:11
elasticdog	that's a nice solution canton7	17:11
	→ ipalaus_ joined	17:11
	→ isao joined	17:11
	→ codeshah joined	17:12
	← iocor left	17:12
	→ upasna joined	17:13
	→ wbednarski joined	17:13
	→ dreiss joined	17:13
	← kadoban left	17:14
	→ isao1 joined	17:14
	→ cakehero joined	17:15
	← isao left	17:16
	← brngardner left	17:16
	→ williamcotton joined	17:16
	→ bolt joined	17:17
	→ brngardner joined	17:18
	← brngardner left	17:18
	← rockets left	17:19
	→ akiress joined	17:20
	← Spockz left	17:20
	→ master-lincoln joined	17:21
	← gigamonkey left	17:21
	→ thierryp joined	17:21
	→ yoklov joined	17:22
	→ poohtarou joined	17:22
	← dstorrs left	17:23
	→ gigamonkey joined	17:23
	→ jeshwanth joined	17:24
	→ kenperkins joined	17:25
	← fisted_ left	17:25
	→ NukeTanuki joined	17:25
	→ MyTallest joined	17:28
	← marcogmonteiro left	17:30
	→ jstout24 joined	17:30
	← netiru left	17:30
	← blorbx left	17:30
	← kevinfagan left	17:31
elasticdog	canton7: `git diff-index --cached --quiet HEAD` was exactly what I needed...thanks!	17:31
	→ vama89 joined	17:31
	→ Dave^\| joined	17:32
	→ OOPMan joined	17:32
	← jeshwanth left	17:33
	→ tchan joined	17:33
	→ ColKurtz joined	17:34
	← quazimodo left	17:34
	→ quazimodo joined	17:34
	← sq-one left	17:35
	← spaceonline left	17:35
	→ brngardner joined	17:36
	← h0bbit left	17:36
	← Targen left	17:36
	→ osmosis joined	17:38
	→ fir31ight joined	17:40
	→ RaceCondition joined	17:41
	← RaceCondition left	17:41
	→ krz joined	17:41
	← mikepack left	17:41
	→ aristidesfl1 joined	17:41
	← aristidesfl1 left	17:41
	→ irqq joined	17:41
	→ ben225 joined	17:42
	→ RaceCondition joined	17:44
	← berserkr left	17:44
canton7	elasticdog, cool :)	17:44
	← yoklov left	17:44
	← whitman left	17:44
	← babilen left	17:44
	→ ntelford joined	17:45
	gitinfo set mode: +v	17:45
	← tvw left	17:45
ntelford	can you commit a single file irrespective of what's staged for commit?	17:45
	i.e. git commit -m <msg> <file>	17:45
	← Vile left	17:47
	← _sundar_ left	17:48
cbreak	ntelford: yes	17:48
	but I recommend against it.	17:48
ntelford	excellent	17:48
cbreak	and I recommend against using -m too	17:48
ntelford	cbreak, why?	17:48
	to both questions	17:48
cbreak	-m because it makes people write shitty oneliners as commit messages	17:49
	→ j416 joined	17:49
cbreak	singlefilecommit because of granularity and workflow considerations	17:49
	→ alberto56 joined	17:49
	→ Spockz joined	17:50
	← lausi left	17:50
	→ lausi joined	17:50
ntelford	cbreak, the -m one I understand	17:50
	← chrisshattuck left	17:50
ntelford	the only reason I wanted to commit a single file was because I had a small change that was orthogonal to my current set of changes already staged	17:51
	→ alvaro_o joined	17:51
cbreak	well, it's your choice	17:51
	technically there's no problem	17:51
	but if you get used to it, you might do it once and then later notice that you committed two orthogonal things because they happened to be in the same file	17:51
	→ btanaka joined	17:52
cbreak	or that you committed something that does not compile but you never tested that because of side effects of your other changes	17:52
	infogulch\|afk → infogulch	17:52
	← fission6 left	17:52
cbreak	as with everything in git: You have the power, you just have to know how to use it :)	17:52
	→ talexb joined	17:54
	← fir31ight left	17:55
	← oso96_2000 left	17:55
	← squentin left	17:55
	→ squentin joined	17:56
	→ fir31ight joined	17:57
	← isao1 left	17:57
	→ isao joined	17:57
	← jameswarren left	17:57
	← DerekPioneer left	17:59
	← asedeno_work left	17:59
	→ asedeno_work joined	18:01
	← Gentlecat left	18:01
	→ Gentlecat joined	18:02
	→ Chillance joined	18:02
	← candybar left	18:02
	→ shruggar joined	18:02
	→ gigamonkey_ joined	18:03
	→ sneakyness_wk joined	18:03
	→ pandeiro joined	18:03
	→ berserkr joined	18:03
	← bartman left	18:03
	← dreiss left	18:03
	→ oso96_2000 joined	18:04
	← gigamonkey left	18:04
	gigamonkey_ → gigamonkey	18:04
	→ sigmonsays joined	18:04
sigmonsays	Is there an ascii way to print all branches as a tree of where they stem from?	18:04
	← ceej left	18:04
	→ scientes joined	18:04
	→ ISF joined	18:04
	← syphar left	18:05
cmn	--graph shows the commits as a graph	18:05
	with --oneline for readability	18:05
	← drvanon left	18:06
sigmonsays	thank you	18:06
cbreak	throw in --decorate	18:07
	← gigamonkey left	18:08
milki	--extra-icing	18:08
	--with-a-cherry-on-top	18:08
	← Araxia_ left	18:09
elasticdog	log --all --date-order --oneline --graph --decorate -M -C	18:09
	→ Araxia_ joined	18:09
	→ bartman joined	18:10
	→ soc42 joined	18:10
elasticdog	I alias `git lo` to that, and sing to myself every time I type it	18:11
	← alvaro_o left	18:12
	← quazimodo left	18:13
	← Nibato left	18:13
	→ Textmode joined	18:14
	→ ulm joined	18:15
NukeTanuki	git lo, git lo, it's off to code I go...	18:16
	← kermit left	18:16
	← scientes left	18:17
	→ scientes joined	18:17
	→ dreiss joined	18:17
	→ adhawkins joined	18:17
	→ idl0r joined	18:18
	→ Nibato joined	18:18
	e-ndy\|afk → e-ndy	18:19
ulm	hi, I've a repo that was converted with svn2git, but unfortunately some of the commits contain a bad author line	18:19
	git fsck reports:	18:19
	error in commit 09c857cb2313767cc108e8933bcc5f1f58f53324: invalid author/committer line - missing space before email	18:19
	so my question is, is there any way to fix these commits?	18:20
cmn	rebase -i and edit them; use commit --amend --author=...	18:21
	← jlebar left	18:21
	← BadCodSmell left	18:21
kevlarman	filter-branch might be more practical depending on the number of affected commits	18:21
ulm	cmn: I've tried that already, but it chokes at the bad commits	18:21
	and generates merge conflicts as well	18:22
cmn	of course it chokes, that's when you fix it	18:22
	but the flattening of the history will be a problem, so try with filter-branch, like kevlarman said	18:22
	← senny left	18:23
ulm	kevlarman: 15 commits out of a total of about 1000	18:23
	→ Vile joined	18:23
ulm	ok, I	18:24
	'll look into filter-branch then	18:24
	thanks	18:24
	← arvind_khadri left	18:25
	→ janhouse joined	18:26
janhouse	error: The following untracked working tree files would be overwritten by merge:	18:27
	→ nowhereman joined	18:27
janhouse	how can I ignore those?	18:27
	as in, leave my own and just... I dont'know. remove them from the repo? no clue.	18:27
	← jim0203 left	18:27
idl0r	git stash	18:28
	← ipalaus_ left	18:28
bremner	I don't think that will help here	18:28
	→ yoklov joined	18:28
janhouse	so, what would?	18:29
bremner	janhouse: is this a one-time thing, or you want to permanently untrack the file?	18:29
	← sneakyness_wk left	18:29
janhouse	well it is a different configuration file that I want to ignore on this box.	18:29
	like forever	18:29
	the whole app/Config/ folder.	18:30
bremner	ah, so I guess you want !config	18:30
gitinfo	[!configfiles] the recommended way is to change your application so that you can (perhaps optionally) store site-specific configuration/data in separate files, then keep those files out of version control. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.	18:30
	← nowhere_man left	18:30
bremner	in particular the link	18:30
janhouse	and then there is some stupid log folder that was added. can I remove that one?	18:30
	→ chrisshattuck joined	18:31
	← Nibato left	18:31
	→ iocor joined	18:33
	← iocor left	18:34
	→ felgru joined	18:34
	→ jameswarren joined	18:34
	→ TravisCarden joined	18:34
janhouse	how can I untrack stuff in git?	18:34
	← TravisCarden left	18:34
kevlarman	janhouse: git rm	18:34
	(--cached if you want your working copy to stay around)	18:35
	→ shruggar1 joined	18:35
	→ Nathandim joined	18:35
	← shruggar left	18:36
	→ spaceonline joined	18:36
talexb	I'm interested in using a post-commit hook, but I have Linux and Windows users .. will the same hook work on all platforms?	18:36
janhouse	git rm --cached?	18:36
	hmm	18:36
	← raatiniemi left	18:37
	→ _iron joined	18:37
	→ Russell^^ joined	18:37
talexb	To complicate matters, I have command line users as well as Eclipse users going through eGit .. which I've heard doesn't just map the UI to git commands, but rather talks to the git repo directly.	18:37
	← jacky left	18:39
	← pdtpatrick left	18:39
	→ Nibato joined	18:40
	← shruggar1 left	18:41
	→ jsoa_ joined	18:43
	← juanfer left	18:44
	→ juanfer joined	18:44
	← jsoa left	18:45
	jsoa_ → jsoa	18:45
	→ StubbornTurtle joined	18:45
StubbornTurtle	For some reason, after rebasing a feature branch ("git rebase master"), it won't let me push that branch anymore. It gets rejected. However, if I pull, it starts taking me through all of the same merge conflicts I just dealt with in the rebase…. why?	18:47
	← hydrozen left	18:47
milki	explain wont let you push	18:48
	← Dave^\| left	18:48
FauxFaux	You can't push because you just !rewrit your history. Pull also makes no sense anymore.	18:48
gitinfo	[!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	18:48
StubbornTurtle	So, should I not have rebased in the first place? Done a merge instead?	18:49
	← vama89 left	18:49
FauxFaux	As is normally the case with git, it depends what you want. Sounds like you either want to: a) Delete your feature branch and push what you have to master, or b) force the push to your feature branch and ignore the consequences.	18:50
	← eli left	18:50
	→ vama89 joined	18:50
	← ISF left	18:51
StubbornTurtle	FauxFaux: I'm trying to use it as a working branch. Making changes, and them being merged into master every once in a while.	18:52
	FauxFaux: I'm rebasing to bring in the latest changes from master into my feature branch so I can be working on the "latest greatest"	18:52
	← brngardner left	18:54
	→ mikepack joined	18:54
	← stodan left	18:54
	← jsoa left	18:54
	→ EuroTrash joined	18:55
	→ wedgeV joined	18:55
	→ jsoa joined	18:55
EuroTrash	How do I tell git not to use my ssh key but prompt for a password instead?	18:55
	→ ipalaus_ joined	18:55
wedgeV	hi, how would i get a list of commits to a branch that happened after a certain commits for which i have the SHA?	18:56
FauxFaux	EuroTrash: That's an ssh setting. Although, for it to happen, the server must be accepting your ssh key. Do you have a weird authorised_keys setup?	18:56
wedgeV	i tried "git log --since <sha> <branch>", but not sure wether that works	18:56
FauxFaux	wedgeV: git log [sha1] # does that, doesn't it?	18:56
	Oh, after.	18:56
EuroTrash	FauxFaux: bitbucket might have...	18:56
FauxFaux	EuroTrash: Do bitbucket allow passwords at all?	18:57
EuroTrash	FauxFaux: yeah, I've done it on other machines that don't have ssh keys.	18:57
	← codeshah left	18:57
EuroTrash	I suspect this key is added to a different account than the one I'm trying to clone from.	18:57
FauxFaux	[email@hidden.address] doesn't support passwords. At all.	18:57
	← dreiss left	18:58
wedgeV	FauxFaux: git log <sha>..<branch> maybe?	18:58
	FauxFaux: it seems to work, but i don't really understand what that command exactly does	18:58
EuroTrash	FauxFaux: I'm pretty sure I've used passwords before, not sure if it was git@ though.	18:58
FauxFaux	wedgeV: ".." means between. So from that commit until the commit that's at the top of the branch.	18:58
	EuroTrash: I'd wildly guess that you were using an https url instead. man git remote if you want to change url.	18:59
gitinfo	EuroTrash: the git-remote manpage is available at http://jk.gs/git-remote.html	18:59
wedgeV	FauxFaux: ok, thats what i want, thanks!	18:59
EuroTrash	Odd, this key is on my account.	19:01
	EuroTrash is baffled	19:01
FauxFaux	EuroTrash: !ssh	19:01
gitinfo	EuroTrash: [!gitolite_ssh] See http://sitaramc.github.com/gitolite/sts.html for steps to troubleshoot ssh/gitolite	19:01
EuroTrash	Woa it works now.	19:01
	Must have been a hickup	19:02
	← isao left	19:02
EuroTrash	Thanks for the pointers anyhow	19:02
	→ eli joined	19:02
	→ brngardner joined	19:04
	← jargon- left	19:04
	→ redstone joined	19:04
	← n1x left	19:05
	← redstone left	19:06
	→ nax joined	19:06
	nax → Guest32975	19:06
	→ iocor joined	19:06
	→ isao joined	19:07
	← JohnyBaby left	19:09
	→ francisfish joined	19:09
	← Peste_Bubonica left	19:10
	→ thirdknife joined	19:11
	← JohnyBaby_ left	19:11
	→ dreiss joined	19:11
	← marthinal left	19:13
	→ mister_m joined	19:14
	→ Peste_Bubonica joined	19:14
	← osmosis left	19:15
	← codewrangler left	19:15
	← brngardner left	19:16
	→ osmosis joined	19:16
	← Shusshu left	19:16
	← ToBeFree left	19:16
	← quuxman_ left	19:16
	→ gtrak joined	19:16
gtrak	hey, is there any way to to make a git svn clone go faster? the svn server is local to my box and it's not using https	19:17
	← j416 left	19:17
cmn	use the filesystem instead of going through the svn protocol	19:17
	other than that... it's a slow process	19:17
	← luxurymode left	19:18
gtrak	oh? how do I do that?	19:18
	← soc42 left	19:19
	→ sq-one joined	19:19
cmn	you use file:///path/to/repo	19:19
	r0bby_ → robbyoconnor	19:19
gtrak	ah, excellent	19:19
	→ drizzd joined	19:20
	← osmosis left	19:21
gtrak	that's brilliant	19:22
	← ntelford left	19:22
gtrak	i imagine it also avoids any interuption errors from the server	19:23
	← ipalaus_ left	19:23
cmn	no idea what that is	19:23
	→ ISF joined	19:23
cmn	basically, all the reading from svn is done by svn itself, so whatever makes svn fast will work here as well	19:24
gtrak	ah, I did not know that	19:24
	← spaceonline left	19:24
	→ shruggar joined	19:24
selckin	linux?	19:24
	← subho left	19:25
gtrak	windows actually	19:25
	gitinfo set mode: +v	19:25
gtrak	everything's cygwin	19:25
	← tatsuya_o left	19:25
selckin	it'll be about 100x faster if you conver it on linux	19:25
	no joke	19:25
asoltys	hi, I want to remove multiple files at once with one call to git rm. some have spaces in the filename, so I've quoted them in double quotes. I've put the list of files in a bash variable but i think i'm getting bash variable or quote interpolation wrong somewhere. can someone help me? https://gist.github.com/2502223	19:26
	→ spaceonline joined	19:26
gtrak	how about a linux vm?	19:26
	→ CannedCorn joined	19:26
gtrak	selckin, why would that be the case, though?	19:27
selckin	possiblty, its mostly cygwins fault & perl, there is always some wua thing you can turn off that gives a little improvement	19:27
gtrak	it doesn't seem like an I/O problem, and cpu usage is right weird, 46% usage, yet system idle process is at 75%	19:28
	doing the 'show processes from all users'	19:28
selckin	trust me, you won't believe your eyes if you run the same command on linux from the text flying by, while its now like 1 line a second	19:28
kevlarman	asoltys: you have too many quotes	19:29
	asoltys: just use tab completion to escape them properly	19:29
gtrak	if I do it in my dual-boot install, can I move the git repo over to a windows fs and still work with there?	19:29
selckin	it wil	19:29
gtrak	it'll be alright?	19:29
selckin	yes	19:29
gtrak	awesome	19:30
	will do, then	19:30
	→ h0bbit joined	19:30
	← gtrak left	19:30
asoltys	kevlarman: i actually have a list of over a hundred that i want to put into a bash variable, so what would be the right way to quote them assuming i wanted to pass them as a variable?	19:34
	→ gtrak joined	19:34
kevlarman	asoltys: um, how are you generating this list?	19:35
	asoltys: sounds like you should be using xargs or something instead of trying to stuff it in a variable	19:35
asoltys	it's the result of a find command. i'm going to be doing it in a filter-branch operation on my whole history so i'm looking for performance hence trynig to avoid xargs or looping	19:36
	→ marcogmonteiro joined	19:36
	← FACEFOX left	19:36
kevlarman	asoltys: xargs is fast	19:36
	asoltys: though you can just find -exec git rm --cached --ignore-unmatch {} + to skip the extra pipe	19:37
	(+ does xargs style expansion rather than calling it for each file)	19:37
	← gusnan left	19:38
asoltys	i'm trying to avoid doing even the find every time though by doing it once up front and caching the results in a variable	19:38
	← gtrak left	19:39
selckin	is it actually slow? the fs stuff will be cached by the OS, find should be instant	19:39
	after the first time	19:39
	→ FACEFOX joined	19:39
	← FACEFOX left	19:39
	← spaceonline left	19:39
kevlarman	asoltys: git ls-files is fast	19:39
	if you want to avoid the fs hit	19:39
	→ FACEFOX joined	19:39
	← FACEFOX left	19:39
asoltys	selckin: it is slow, yeah. and i'm not sure why so experimenting with different things. i also tried doing the whole filter operation in my /run (tempfs) mounted folder so it would all be done in memory	19:39
cmn	if tempfs is slow, the problem is somewhere else	19:40
	→ Aaeriele joined	19:40
	→ kylef joined	19:40
bartek	I dont seem to understand git pull --rebase. Will git basically put anything on my branch up and push any merged changes below those?	19:41
asoltys	i guess it's just a lot of (large, binary) files and a lot of revisions.	19:41
kevlarman	recursing over a tmpfs is still slower than reading the list from the index	19:41
selckin	unless he has 100s of thousands of files shouldn't notice much	19:42
	← freeed left	19:42
	← iocor left	19:42
kevlarman	selckin: actually it makes a huge difference when you switch from a tree filter to an index filter	19:42
asoltys	give me a sec and i'll paste the full command i'm using	19:42
	→ freeed joined	19:42
	e-ndy → e-ndy\|afk	19:43
kevlarman	even writing to tmpfs is slow when you have to do it for every commit	19:43
	→ FACEFOX joined	19:43
selckin	find over my 30gb of source code is like 0.5s	19:43
	← FACEFOX left	19:43
asoltys	kevlarman: i am using index-filter btw	19:43
	→ FACEFOX joined	19:43
	← FACEFOX left	19:43
kevlarman	selckin: now multiply that by 5k commits	19:43
	→ drizzd_ joined	19:44
selckin	and divive by 300x less files	19:44
	← RaceCondition left	19:44
	danmackay_ → danmackay	19:44
	→ FACEFOX joined	19:45
	← FACEFOX left	19:45
	← freeed left	19:45
	← d0k left	19:45
	→ astm joined	19:45
	← _Vi left	19:45
	→ d0k joined	19:45
	→ FACEFOX joined	19:45
	← FACEFOX left	19:45
	← astm left	19:46
	→ jargon- joined	19:46
asoltys	shiver me timbers.. seems to be working reasonably fast now using the find -exec command you posted earlier kevlarman.	19:46
	well, thanks :D	19:46
	→ FACEFOX joined	19:46
	← FACEFOX left	19:46
	→ spaceonline joined	19:47
	← drizzd left	19:47
	← idl0r left	19:48
	← jceb left	19:48
	→ FACEFOX joined	19:48
	→ vpopov joined	19:49
	→ cjb` joined	19:50
	← cjb` left	19:51
	→ cjb joined	19:51
cjb	Oops, I just did "rm .git/index" instead of "rm .git/index.lock". Any tips?	19:51
	→ gtrak joined	19:51
cjb	(I don't need to recover any changes, just want to get an index back.)	19:52
selckin	clone again?	19:52
	← kylef left	19:52
	→ Tasser joined	19:52
cjb	ah, found something:	19:53
	selckin: http://robots.thoughtbot.com/post/8484599870/how-to-fix-rm-f-git-index	19:53
	→ singingwolfboy joined	19:53
singingwolfboy	When I run `git commit`, my editor opens with a listing of the files that were changed in the commit. Is there a way to make it instead display the full diff for that commit?	19:54
selckin	if you don't need to recover, seem like the obvious way to fix anything :)	19:54
cjb	selckin: obvious but slow; it's a kernel repo.	19:54
	← Vert left	19:54
selckin	true	19:54
cmn	cjb: read-tree	19:55
cjb	cmn: yep, that's what the URL says, thanks	19:55
	← spaceonline left	19:55
cjb	it worked	19:55
kevlarman	singingwolfboy: -v	19:55
cmn	or read-tree HEAD if it's being stubborn	19:55
cjb	("git read-tree --reset HEAD")	19:55
	→ kermit joined	19:55
	← sp3ctum left	19:55
kevlarman	singingwolfboy: although that isn't commented out, so it's up to you to deal with the diff somehow	19:55
cmn	doesn't it get created under the three dashes?	19:56
	nevermind, that's git-am, nothing to do with commit	19:56
	→ Fleep joined	19:56
	→ airborn joined	19:57
	← Jackneill left	19:59
	← Poapfel left	20:00
	← ronniy left	20:01
	→ spaceonline joined	20:03
	→ kadoban joined	20:03
	→ Poapfel joined	20:03
	→ beatak_m joined	20:03
	← Davey left	20:04
	→ ogzy joined	20:04
	← thirdknife left	20:04
ogzy	when i was commiting, i excidentally wrote git commit -m"message" so all my changes are sent, but i was planning to do it only for one file, how can i fix it?	20:04
selckin	did you push already?	20:05
	← StubbornTurtle left	20:05
	→ ipalaus_ joined	20:05
selckin	git reset HEAD^, the undo the last commit	20:05
ogzy	selckin: no	20:06
	← beatak left	20:06
	← ben225 left	20:07
	← FernandoBasso left	20:07
	← enmand left	20:09
	← skylamer` left	20:09
	→ Davey joined	20:11
	→ nutron joined	20:12
	→ tatsuya_o joined	20:12
	← francisfish left	20:12
	← i10neorg left	20:14
	→ i10neorg joined	20:14
	→ oriba joined	20:14
	→ whitman joined	20:14
cmn	if you didn't push, your changes weren't sent anywhere	20:14
	use commit --amend	20:14
	→ francisfish joined	20:14
	→ Dave^\| joined	20:14
	← beatak_m left	20:14
	← cjb left	20:15
	← keller left	20:15
	→ brngardner joined	20:16
	← Milossh left	20:16
	← janhouse left	20:16
ogzy	cmn: git reset HEAD^ was fine	20:17
	→ janhouse joined	20:17
	→ Milossh joined	20:17
	← FACEFOX left	20:17
	← isao left	20:19
	← francisfish left	20:19
	→ FACEFOX joined	20:19
	← imarcusthis left	20:20
selckin	yw	20:22
	→ blorbx joined	20:22
	← brngardner left	20:23
	← Sigma left	20:23
	← marcogmonteiro left	20:24
	→ imarcusthis joined	20:24
	← speeder left	20:25
	← na3r left	20:25
	→ user64 joined	20:25
	← janhouse left	20:26
	→ dabd joined	20:26
	← ehsan left	20:26
	→ ehsan joined	20:26
	→ isao joined	20:27
	← singingwolfboy left	20:29
	← hagabaka left	20:29
	→ teweWORK joined	20:29
	← drizzd_ left	20:31
	→ nevyn joined	20:32
	→ marcogmonteiro joined	20:32
	→ claaud joined	20:32
	← claaud left	20:33
	← ipalaus_ left	20:33
	→ FreeArtMan joined	20:34
	← sq-one left	20:34
	← FreeArtMan left	20:35
	← user64 left	20:35
	← Icedream left	20:36
	← whitman left	20:36
	← spyro left	20:36
	→ ShadeTornado joined	20:36
	→ RaceCondition joined	20:38
sayd	yay or nay on submodules for dependencies?	20:38
selckin	nay	20:38
cmn	nay for the most part	20:39
sayd	better alternative?	20:39
cmn	that depends on what you want to do	20:39
selckin	for a problem you haven't explained	20:39
sayd	so	20:39
cmn	but, you know, 42	20:39
	→ ipalaus_ joined	20:39
	← RaceCondition left	20:39
sayd	:)	20:39
	← elasticdog left	20:39
sayd	i have a dependency on twitter bootstrap, which has its own git repo.	20:40
	right now its added as a submodule to vendor/bootstrap	20:40
	← usmanity left	20:40
sayd	im not sure how to keep things in sync as different commits of my super repo may be tied to different commits of the submodule repo	20:41
	→ janhouse joined	20:42
	← Davey left	20:43
	← dabd left	20:43
	← vpopov left	20:44
	→ Vert joined	20:44
	→ ben225 joined	20:45
	← ulm left	20:45
	→ nemish1 joined	20:45
	← nemish1 left	20:46
	→ brngardner joined	20:47
	← jsoa left	20:49
	← spaceonline left	20:51
	→ pdtpatrick joined	20:51
	→ gigamonkey joined	20:52
	← janhouse left	20:52
	→ janhouse joined	20:52
	→ spaceonline joined	20:53
	neersighted\|AFK → neersighted	20:56
	→ marhaban joined	20:56
	← NukeTanuki left	20:58
	← irqq left	20:58
	→ babilen joined	21:00
	→ thirdknife joined	21:01
	← ben225 left	21:02
	← janhouse left	21:02
	→ janhouse joined	21:03
	← chrisshattuck left	21:03
	→ fisted joined	21:03
	→ jatt joined	21:03
	→ ben225 joined	21:04
	→ ceej joined	21:04
	← pantsman left	21:04
	← rurufufuss left	21:04
	← mister_m left	21:05
	→ whitman joined	21:07
	← mrwalker left	21:09
	← martinjlowm-lptp left	21:09
	→ martinjlowm-lptp joined	21:09
	← cannonball left	21:11
	← mishok13 left	21:11
	→ spyro joined	21:12
	← Quadrescence left	21:12
	← janhouse left	21:13
	→ janhouse joined	21:13
	← MyTallest left	21:14
	→ Quadrescence joined	21:16
isao	Hiya.. so I have an svn repo S and a git repo R of the same files. Unfortunately, R was not created with git-svn clone, and patches to one have been manually applied to the other.	21:17
	Now.. I must make the git repo R the source of truth (or die trying).	21:18
	→ aspotashev joined	21:18
isao	what's the best way to sync git to svn at this point?	21:18
	automatically..	21:18
	← dcarper left	21:18
kevlarman	isao: can you just get rid of the svn repo?	21:19
isao	yes	21:19
kevlarman	do that	21:19
	→ beatak joined	21:19
isao	ok..	21:20
FauxFaux	You /could/ git-svn import the svn repo, git-replace together two old commits (causing it to look like history has forked), then merge history back together (and pray).	21:20
kevlarman	FauxFaux: ew	21:21
	FauxFaux: there are better ways to repair history	21:21
	← spaceonline left	21:21
	→ vork joined	21:21
vork	What's the opposite of stg init? How can I un-stgit a branch?	21:21
	→ RaceCondition joined	21:21
	← jatt left	21:22
FauxFaux	kevlarman: I think that maps quite well to what's gone on, and is quite likely to give the expected result, especially if you just commit-tree dirty-merge the tip; what would you do?	21:22
kevlarman	FauxFaux: git svn clone, then some careful rebasing	21:23
	→ willmars_ joined	21:24
FauxFaux	kevlarman: Rebasing is going to ruin the order of history. (Normally I'd never, ever suggest merging.)	21:25
kevlarman	FauxFaux: huh? i mean for commits that exist only in git	21:25
	FauxFaux: a merge will ruin any attempts to write back to svn	21:25
	← martinjlowm-lptp left	21:25
	→ martinjlowm-lptp joined	21:26
kevlarman	(and if you can just get rid of svn, then there's nothing to be done)	21:26
adhawkins	Any way in git to get the last modified date of a file? Thinking for a commit hook to modify a date in the header	21:26
kevlarman	adhawkins: terrible idea	21:27
milki	that info isnt tracked	21:27
adhawkins	kevlarman: I'm getting fed up with myself forgetting to update copyright notices at the top of files :)	21:27
	→ spaced0g joined	21:27
	← martinjlowm left	21:28
adhawkins	Any suggestions?	21:28
kevlarman	adhawkins: then write a hook that nags you	21:28
	→ martinjlowm joined	21:28
isao	adhawkins: git filters?	21:28
kevlarman	automatically modifying it is going to cause you problems	21:28
adhawkins	isao: Explain? kevlarman: What problems?	21:28
	← agile left	21:29
	← martinjlowm left	21:30
	→ martinjlowm joined	21:30
isao	adhawkins: I've never used, but… here http://progit.org/book/ch7-2.html re "Keyword Expansion"	21:30
adhawkins	isao: Yes, that's what I was envisaging	21:31
	← starsinmypockets left	21:31
	← marhaban left	21:31
adhawkins	kevlarman: What problems would it cause?	21:31
	→ rockfruit joined	21:34
isao	FauxFaux, kevlarman: so if I just forget about the svn-side for now, here's what I think you were suggesting: 1) git svn clone newsvn 2) cd newsvn; git pull git-truth master, 3) git svn dcommit… 4) can I repeat and automate this safely, assuming no users write to "newsvn" ?	21:36
kevlarman	adhawkins: let's say you do something like rebase -i	21:36
	and it changes old commits	21:36
	← airborn left	21:36
	→ dcarper joined	21:37
kevlarman	isao: no, you just created a merge	21:37
isao	:(	21:37
adhawkins	So probably the best idea is a pre-commit hook that looks for a Copyright line, and rejects it if it doesn't contain the current year?	21:37
	Or would that have the same issue with rebasing etc.?	21:38
	→ HyperNerdV2 joined	21:38
	← EuroTrash left	21:38
	← tr-808 left	21:38
	← t4nkd left	21:38
kadoban	why do you even have a line in every file with the year?	21:38
sayd	to cherry pick a range of commits: cherry-pick start-hash:end-hash?	21:39
	→ tr-808 joined	21:39
milki	or make it a variable	21:39
	that you replace when you package	21:39
	!	21:39
	← Element9 left	21:39
	→ martinjl` joined	21:39
	→ mister_m joined	21:40
	← _iron left	21:40
	← martinjlowm left	21:41
	→ ixti joined	21:42
	← mister_m left	21:43
	← revagomes left	21:43
	← pandeiro left	21:43
	→ w09x joined	21:43
	← thirdknife left	21:44
	← saml left	21:46
	← nrichards left	21:46
	← ixti left	21:48
	← Russell^^ left	21:48
	→ ixti joined	21:48
	→ juvenal joined	21:49
	→ jalama joined	21:49
	→ mister_m joined	21:50
	→ Diemuzi joined	21:52
wedgeV	what's the best way to delete a branch in a remote repo?	21:52
kevlarman	wedgeV: git push remote :branchname	21:52
	← ShooterM_ left	21:52
	← jalama left	21:52
wedgeV	thanks	21:53
vork	How can I find out which remote branch the current branch is tracking?	21:53
	← berserkr left	21:53
	← ISF left	21:53
kevlarman	vork: git config branch.<yourbranch>.remote and .merge	21:53
	← ixti left	21:54
	→ berserkr joined	21:54
	← dcarper left	21:55
	← macabre left	21:55
	→ ShooterM_ joined	21:56
	← arietis left	21:56
	→ kcj joined	21:57
	→ Leftmost joined	21:57
	→ davidbanham joined	21:57
	← mritz left	21:58
	← sebrock left	21:58
canton7	vork, git branch -vv also iirc	21:58
Leftmost	I have a branch which split from master at commit A. The branch is now at commit B and master at commit C. I want to revert all changes between A and C in master and replace it with the commits between A and B. What is the correct way to do this?	21:59
canton7	Leftmost, has master been pushed anywhere?	21:59
Leftmost	Yes.	21:59
	Reversing the patch is fine.	21:59
canton7	do you care about anyone who might be tracking master? :P	21:59
	→ enmand joined	22:00
Leftmost	It's checked out elsewhere, but no one should have made any changes at all.	22:00
	Recloning is okay.	22:00
	← Dave^\| left	22:01
canton7	hrm, people pulling rewritten history is a bad idea... it tends to merge together the two versions of history	22:01
	your options are 1. pretend that the commits on master never existed, and replace them with those from the branch	22:01
	→ marhaban joined	22:01
canton7	2. keep the commits on master, make a commit which resets the working copy back to commit A, then stick the branch's commits on top	22:01
	← glcrazy left	22:02
	← nioncode left	22:02
	→ Slix` joined	22:02
	← btx94 left	22:02
canton7	the former is !rewriting_public_history, the latter is safe	22:02
gitinfo	Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	22:02
	← ehsan left	22:02
	→ kaen joined	22:03
	← kaen left	22:03
	→ kaen joined	22:03
	← ben225 left	22:03
	← Destos left	22:03
vork	Thanks canton7, kevlarman.	22:03
Leftmost	Thank you, canton7.	22:04
	→ LittleFoot joined	22:04
LittleFoot	hi	22:04
gitinfo	LittleFoot: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	22:04
Leftmost	Is there a speedy way to create a commit to revert the changes? git revert foo..bar will work in a pinch, but a single commit seems cleaner.	22:04
LittleFoot	anyone want to test out my diff alg?	22:04
	http://noteys.com/notes/details/181	22:05
	→ btx94 joined	22:05
Leftmost	Oh, nevermind.	22:05
	Found -n. :P	22:05
canton7	Leftmost, there's also a trick: git rm -rf *; git checkout <commit> -- .	22:06
	← brngardner left	22:06
	← ogzy left	22:06
	← madsy left	22:06
	→ ixti joined	22:07
	← jargon- left	22:07
	→ madsy joined	22:07
	← madsy left	22:08
	→ madsy joined	22:08
	← ixti left	22:09
	← eijk left	22:09
	→ seank-efi joined	22:09
	→ bartek_ joined	22:10
sayd	it i want to cherry-pick a range of revisions that happend in the same branch but in the past, whats the best way?	22:10
vork	sayd: rebase	22:10
kevlarman	sayd: um, cherry pick them?	22:10
	← juanfer left	22:10
cmn	what's wrong with using cherry-pick to cherry-pick?	22:11
	→ thearthur joined	22:11
sayd	kevlarman: the range is the tricky thing. i just read through man gitrevisions and im still unclear	22:11
gitinfo	kevlarman: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	22:11
vork	Don't know. Rebase might be easier.	22:11
cmn	vork: not if what he wants is to cherry-pick stuff from the past	22:11
sayd	git cherry-pick 131231:3923492 kinda thing?	22:11
vork	cmn: Uh..?	22:11
cmn	sayd: the range is negative..positive	22:11
sayd	or is it ...	22:11
	← martinjl` left	22:11
vork	cmn: Since when did rebase care about when the commits were made?	22:12
cmn	sayd: where did you get the colon for ranges from?	22:12
sayd	ok	22:12
kevlarman	vork: rebasing published history is a terrible idea	22:12
cmn	vork: in the past means that they're not at the tip	22:12
sayd	so oldhash..newerhash	22:12
cmn	yeah	22:12
vork	kevlarman: cherry-picking published history is just as terrible.	22:12
kevlarman	vork: no it isn't	22:12
sayd	thats inclusive, yeah?	22:12
cmn	vork: completely different hings	22:12
kevlarman	vork: cherry-picking doesn't rewrite	22:12
vork	Why isn't it? It's pretty much the same thing, isn't it?	22:12
	→ martinjlowm joined	22:12
	← tqmz left	22:12
kevlarman	vork: not even close to it	22:13
cmn	sayd: yeah, it'll start at newerhash and stop when it finds oldhash	22:13
	← bartek left	22:13
vork	kevlarman: What do you mean, it doesn't rewrite?	22:13
	cherry-pick creates a new commit object, with a new SHA-1.	22:13
kevlarman	vork: cherry-pick doesn't rewrite history, rebase does	22:13
cmn	vork: they're not, one puts your commits on top of other ones and the other puts other commits on top the current tip	22:13
	they're pretty much opposites	22:13
	gitinfo set mode: +v	22:14
	← martinjlowm left	22:14
vork	cmn: Well, it's the same operation, just different ways of specifying the commits.	22:14
	→ martinjlowm joined	22:14
sayd	hmm, Automatic cherry-pick failed	22:14
vork	kevlarman: I don't understand, "cherry-pick doesn't rewrite history".	22:14
canton7	vork, the main difference is that rebase changes commits. cherry-pick copies commits. When you cherry-pick, the original commit remains	22:14
vork	It rewrites it just as much as rebase does. The main difference is that git-rebase tends to want to move your source ref, whereas cherry-pick doesn't.	22:15
canton7	..which is a critical difference	22:15
vork	canton7: Rebase does not change commits, it creates new ones, and (by default) deletes the old ref.	22:15
	→ adamm joined	22:15
canton7	vork, I know. I know where you're coming from. The difference, as I say, is that rebase removed the ref to the old commits. It re-writes a branch, if you want to define such a thing	22:15
	← juvenal left	22:15
	← AaronMT left	22:16
	→ HyperNerdV2_ joined	22:16
	← dr_lepper left	22:16
	← ShooterM_ left	22:16
vork	canton7: That's about the only difference.	22:16
sayd	why would cherry-pick conflict?	22:16
vork	If you create a new ref in the middle of an old branch, and then rebase it, no refs are lost.	22:16
cmn	sayd: becase you've made changes	22:17
	→ ShooterM_ joined	22:17
	← martinjlowm left	22:17
	→ martinjlowm joined	22:17
cmn	fix the conflict and continue	22:17
	vork: the other critical difference is that rebase takes the whole branch; cherry-pick lets you specify which commits you're interested in	22:18
	← HyperNerdV2 left	22:18
sayd	yeah, i see the conflict is within a block that i changed removed	22:18
vork	cmn: Yes.	22:18
	← dv310p3r left	22:18
sayd	so i can just blast out the whole <<>> ?	22:18
vork	What sayd said sounded like an old topic branch, that's why I thought of rebasing.	22:18
	← mikepack left	22:19
cmn	sayd: you're the only one who knows	22:19
	← bartek_ left	22:19
	← Briareos1 left	22:19
	← CellKill_ left	22:19
	← HyperNerdV2_ left	22:20
sayd	ok.. im just a machine that does git's bidding	22:20
	← teweWORK left	22:20
	→ ben225 joined	22:21
	← tewecske left	22:21
	← duckxx left	22:21
vork	Git exists to serve man, and not to rule over him.	22:21
	→ CellKill joined	22:21
milki	no it does	22:21
	nt	22:21
seank-efi	Hi guys, I think i did something dumb. My initial commit contains sensitive files and I cant seem to do a rebase -i and pick my first commit. Is there any hope?	22:22
milki	you cant rebase the first commit	22:22
sayd	i didnt get any of the commits i thought that range would give me :(	22:22
	→ linguini joined	22:22
milki	but you can probably filter-branch it still	22:22
seank-efi	hmmm	22:22
	→ menardII joined	22:22
milki	or just create an orphaned branch	22:22
	redo the commit	22:22
	and name the new branch master	22:23
	and delete the old master	22:23
	← fmPirate left	22:23
milki	not in that order	22:23
	its just the first commit	22:23
	make a new repo	22:23
	>.>	22:23
	← thiago left	22:23
milki	unless theres more commits...	22:23
seank-efi	no there are several commits, but i suppose I dont really need the history	22:23
milki	if you want the history, i suggest filter-branch	22:23
	← TeckniX left	22:24
seank-efi	Ill have a read about filter-branching	22:24
	thx!	22:24
milki	github has a nice page on sensitive data removing	22:24
	→ redbrain joined	22:24
	← martinjlowm-lptp left	22:25
sayd	do I just continue to go through each conflict, fix, commit and then retry the cherry-pick?	22:25
milki	context?	22:25
vork	sayd: Yep.	22:25
milki	o	22:25
	milki reads up	22:25
sayd	a conflicting cherry-pick	22:25
	there doesnt seem to be a --continue flag like rebase	22:26
vork	I think cherry-pick should automatically skip already-applied commits.	22:26
	sayd: Actually there is.	22:26
sayd	seems like i could diff rev2..rev10 and create a pactch	22:27
	would that still conflict?	22:27
	← martinjlowm left	22:27
vork	It's likely to.	22:28
cmn	there is --continue	22:28
	← willmars_ left	22:28
vork	I prefer resolving conflicting commits one at a time.	22:28
	→ Adys joined	22:29
	→ thiago joined	22:29
	← shruggar left	22:29
	← Leftmost left	22:30
sayd	it just keeps conflicting	22:30
vork	What, the same commit?	22:31
	→ williamcotton_ joined	22:31
	gitinfo set mode: +v	22:31
menardII	i'm trying to figure out how to "unsquash" a big commit -- bisect has told me that it's the cause of a problem, but when i eyeball its changes i can't figure out what they are	22:31
sayd	cherry-pick --continue \nusage: git cherry-pick [options] <commit-ish>	22:31
	← williamcotton left	22:32
	williamcotton_ → williamcotton	22:32
	→ paphio joined	22:32
vork	menardII: git checkout -b problem <problem-commit> && git reset HEAD~1 && git add -p	22:32
	← felgru left	22:32
cmn	then you should stop using an ancient version of git	22:32
	← v0n left	22:33
	← LucasTizma left	22:33
cmn	you'll have to pick one-by-one I guess	22:33
menardII	vork: aha! so obvious now that you mention it. thanks	22:33
	→ frogonwheels joined	22:34
	← whitman left	22:34
	← marhaban left	22:34
	← gtrak left	22:34
	→ marhaban joined	22:34
	← beatak left	22:34
	← Chillance left	22:36
	← Fullmoon left	22:38
	← QaDeS left	22:38
linguini	I'm trying to "rewrite history"; git merge --no-commit --no-ff is almost what I want; but I'm not sure if I want a "merge" or a "rebase".	22:39
	← d0k left	22:40
	← paphio left	22:43
	→ paphio joined	22:43
	← vork left	22:44
	→ iocor joined	22:44
	← oso96_2000 left	22:46
	← marcogmonteiro left	22:46
	← thelazydeveloper left	22:46
linguini	http://paste.lisp.org/+2ROD	22:48
	← Araxia_ left	22:48
	→ Briareos1 joined	22:48
	← CannedCorn left	22:49
	→ mikegerwitz joined	22:49
	← paphio left	22:49
	← maestrojed left	22:49
	← rgr left	22:51
	→ brngardner joined	22:51
	← Briareos1 left	22:52
	→ oso96_2000 joined	22:52
	← berserkr left	22:52
	← Diemuzi left	22:52
	← ology left	22:53
	← isao left	22:54
	→ berserkr joined	22:54
	← vama89 left	22:54
	→ pastime joined	22:54
	← pastime left	22:55
	→ john_anderson joined	22:55
	→ fling joined	22:58
seank-efi	hmmm	22:58
	git filter-branch -f --index-filter 'git rm --cached --ignore-unmatch InterractingWithFreeEMS.png' --prune-empty -- --all	22:58
	Rewrite 13e888c8e6c4faf95816d5a3061a61bd6c68d3c5 (10/10)	22:58
	WARNING: Ref 'refs/heads/master' is unchanged	22:58
	WARNING: Ref 'refs/remotes/origin/master' is unchanged	22:58
	WARNING: Ref 'refs/stash' is unchanged	22:58
	→ texasmynsted joined	22:59
	← dreiss left	22:59
Vinnie_win	can git-svn import from the local file system or must it have a server?	23:00
	← master-lincoln left	23:00
seank-efi	if I do git branch temp f6a5494f61bacf703bc8a527d681396dfa1e813e the InterractingWithFreeEMS.png is still there	23:00
	→ pnorman joined	23:01
	← karsten left	23:01
	→ karsten joined	23:01
	→ yogo joined	23:02
	→ chronos joined	23:02
	← chrisgeorge left	23:03
cmn	Vinnie_win: it uses svn underneath and can use whatever svn can	23:04
pnorman	I have a local repo and a remote one that I want to merge into it. I don't want the remote one to end up in a subfolder, I want it in the same folder. There is one file with a duplicate name that I do not want to get from the remote repo. I would like to preserve history. Any pointers?	23:04
cmn	seank-efi: if it was in that commit, it will always be in that commit	23:04
	seank-efi: what you have to look at is the rewritten history, not the one you already had	23:04
Vinnie_win	cmn: hmm... It seems I have "svnadmin" and when I run it on the svn repo everything is verified properly (its a local drive). How can I import that to git ?	23:04
	← btanaka left	23:04
Vinnie_win	pwd	23:04
	^^ forget about that	23:04
cmn	git svn clone <path>	23:04
	svn likes its fs-paths to have file:// in front	23:05
pnorman	If all else fails I suppose I could just copy the files.	23:05
	\toothrot → toothr	23:06
	← yogo left	23:07
	→ krzkrzkrz joined	23:07
Vinnie_win	cmn: http://codepad.org/nHv6qbq7	23:07
	cmn: When I run svnadmin from /e/OldDrive/svn/Dev it seems fine.	23:07
	← menardII left	23:07
cmn	that's not a valid fs URI	23:08
Vinnie_win	cmn: I'm on Windows	23:08
cmn	you're missing the root slash for the path	23:08
Vinnie_win	so three / ?	23:08
	← Guest32975 left	23:08
cmn	two for the protocol specifier and one because that's how you root paths	23:08
Vinnie_win	http://codepad.org/PRNxpDHS	23:08
cmn	your svn version is too old	23:09
Vinnie_win	that is something I can probably remedy	23:09
	Nice work thanks...you sure know your svn!	23:09
	← krz left	23:09
	krzkrzkrz → krz	23:09
gigamonkey	Is there any way to git stash just certain files?	23:09
Vinnie_win	cmn: my svn is sliksvn from mar 8 2012	23:10
cmn	no idea wth that version is meant to be	23:10
Vinnie_win	cmn: is it my git, or my svn, that needs updating?	23:10
cmn	the important version is the one that libsvn has	23:10
	that's what git-svn uses	23:11
	I don't think git-svn forces a version, but I prefer not to look at that monstrosity	23:11
Vinnie_win	subversion.tigris.org	23:11
	→ rpk_ joined	23:11
Vinnie_win	Oh...well, don't worry I'm on my way out. I just got my foot caught in the door.	23:11
	→ LucasTizma joined	23:11
	→ WillMarshall joined	23:12
	→ dreiss joined	23:13
	← rpk_ left	23:13
	→ rpk_ joined	23:13
rpk_	I just ran merge --squash --no-commit branch and committed. Some files that were "Auto-merged" with no conflict now have duplicate sections of code in them. Is there anything I can do to fix / prevent this?	23:14
	← TomyLobo left	23:15
seank-efi	cmn: is there anyway to eradicate it ? I messed up and it's part of my initial commit.	23:15
cmn	it will get deleted once the safe-time is passed	23:16
	just make sure it's not part of the history you're going to pusblish	23:16
	→ rgr joined	23:16
	← LucasTizma left	23:16
	→ project2501a joined	23:16
seank-efi	cool and thx !	23:16
cmn	you'll have to delete refs/original/*	23:16
	once you're sure the rewrite gave you something you want	23:16
	→ LucasTizma joined	23:16
	← scientes left	23:17
	← adhawkins left	23:18
	→ scientes joined	23:18
Vinnie_win	cmn: Even after using the latest binaries from Apache I still get the format error when doing git svn clone	23:18
	→ cbx33 joined	23:18
	← blorbx left	23:18
cmn	the problem isn't in the binaries, it's in the library that git-svn uses	23:18
	which may or may not be the ones that come with the distribution you just downloaded	23:19
Vinnie_win	cmn: ah so the command line version of svn has nothing to do with what git-svn is using	23:19
cmn	the command-line svn uses libsvn, same as git-svn	23:19
Vinnie_win	I'm on windows	23:19
cmn	yes, so you keep saying	23:19
	which is why this is likely	23:20
Vinnie_win	the svn I installed came with its own libsvn_client.dll etc separate from what git has	23:20
cmn	the version you're using when you type 'svn' may not be the same one that git-svn is using	23:20
	right	23:20
	← williamcotton left	23:20
	← brngardner left	23:20
Vinnie_win	damn	23:20
	← giallu left	23:20
Vinnie_win	the libsvn files are from 2009	23:21
rpk_	Also, I consistently end up having to remerge changes (that is, both local and the merge branch have the same changes, and the destination file shows old changes)	23:21
Vinnie_win	cmn: I'm going to try splicing in a newer client dll and see what happens	23:21
	← Nathandim left	23:21
cmn	rpk_: if you're using --squash, you're not merging	23:22
	putting in a new DLL might work, or it's just as likely that the bindings will blow up	23:23
Vinnie_win	cmn: It didnt blow up it just generated a different error	23:23
cmn	you might want to try using msysgit to build a newer version	23:23
Vinnie_win	cmn: Hmm...you mean, build a new version of the svn dlls ?	23:23
cmn	yes	23:23
rpk_	cmn: I need to squash my changes before pushing them back	23:24
Vinnie_win	cmn: It would probably be easier to just run a local svn server temporarily	23:24
cmn	Vinnie_win: or use svn-dump-fast-export	23:24
	rpk_: if that's what you need to do fine, but you're not merging	23:24
rpk_	cmnL and the output message is literally "Auto-merging rails-root/app/models/file.rb"	23:24
Vinnie_win	cmn: Does that include history? I'm trying to splice the old svn history at the back of my existing git repo	23:25
cmn	and whatever do you mean by "destination file"?	23:25
	rpk_: that's a different thing	23:25
	← ajpiano left	23:25
cmn	Vinnie_win: yes	23:25
	it takes the dump and replays it	23:25
	→ brngardner joined	23:25
rpk_	cmn: I meant the file that I merge into. I end up with local, server, and then merge into a file which currently has changes from a revision back, with local and server both different (and sharing some identical changes) into that file	23:26
Vinnie_win	cmn: I've never used svn command lines before. So I would issue svnadmin dump .... and then run this other tool on that to produce a git repo? I should try that before installing apache and svn daemon and all that jazz	23:26
cmn	the svn-dump-fast-export produces a fast-import stream, which you can use to create a git repo	23:27
Vinnie_win	so I dont use svnadmin at all	23:27
cmn	you need to create a dump	23:27
rpk_	cmn: (I'm aware that the file just delineates differences using some tags, but I'm trying to explain that git is flagging some things as changed that haven't)	23:27
	← batmanian left	23:27
	→ ajpiano joined	23:27
	→ yogo joined	23:27
Vinnie_win	cmn: svnadmin dump then svn-dump-fast-export then a git command	23:27
cmn	roughly, the webpage has the exact commands	23:28
	→ noahnoah joined	23:28
Vinnie_win	I cant find a .exe	23:28
	← rgr left	23:28
cmn	I doubt he has them	23:29
noahnoah	Hello,	23:29
gitinfo	noahnoah: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	23:29
cmn	rpk_: git tries to do a three-way merge whenever possible, so you might be seeing the base version which git is showing you for comparison	23:29
	← maqr left	23:30
	→ mritz joined	23:30
cmn	rpk_: and if you keep bringing in changes without telling git about it, then it's no wonder it gets confused about what version is which	23:30
noahnoah	I have made a bunch of updates on a local machine that I want to push to the remote repository. The problem is that there may have been some other updates made from another machine, so things might be out of sync. I don't care about any of the other updates and want to just force the current version on my machine as the most recent in the remote. However, when I push, Git gives me an error, "non fast-forward updates were	23:30
	rejected. How can I resolve this?	23:30
rpk_	cmn: yes, but its doing it incorrectly in two different ways. 1) Auto merging and creating duplicates ( I added =begin =end around a block of code and the entire section now appears twice in the auto-merged file, the original and commented version). 2) Detecting changes where there are none.	23:31
thiago	noahnoah: push -f. That will erase from history any updates made from any other machines.	23:31
noahnoah	Awesome. Thank You!!	23:31
thiago	noahnoah: it's as if they've never existed. Do not pull from those machines, you'll get into trouble.	23:31
	← marhaban left	23:31
noahnoah	So I lose all history??	23:31
thiago	noahnoah: you lose the history created from those machines	23:32
et	!rewriting	23:32
gitinfo	[!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	23:32
	← mritz left	23:32
thiago	noahnoah: in general, if you created history in other nodes, it's because such history is useful	23:32
	→ marhaban joined	23:32
Vinnie_win	cmn: it doesnt build for me in msvc2010	23:32
thiago	noahnoah: therefore, you don't want to waste it. You want to merge it, and that's what git pull does	23:33
	← john_anderson left	23:33
	→ maqr joined	23:33
noahnoah	I'm the only developer, but juggle between home machine and work machine. I was screwing around on home machine and made a small update. Later, at work, I forgot to pull, and did a TON of work. The work machine is the current version. I don't care about what is done on the home machine. So, I just want to force the version on the work machine as current, and then pull that down to the home machone.	23:33
thiago	noahnoah: git pull and solve the conflicts	23:34
cmn	rpk_: if the diff isn't descriptive enough, it's possible for it to be applied wrong	23:34
	→ sorahn joined	23:34
Vinnie_win	noahnoah: you could branch your home changes then roll back the master to the place where work split off, and resolve the conflicts later	23:34
cmn	and as I said, if you keep lying to git about the origin of changes, it's going to be confused	23:34
gigamonkey	Does doing a git push origin --delete somebranch have any effect on the local version of somebranch, either immediately or the next time you pull?	23:35
sorahn	hey guys, i have a question, I'm using a hook to add the current branch name to my commit message, but if you do that, there's no way to abort the commit, as that the message is already different from what git is comparing against. Anyone know a way around this?	23:35
noahnoah	pulling and merging seems difficult. I mad a ton of changes on the work machine. Really have not need for the changes that I pushed last from the other machine.	23:35
rpk_	how am I lying about the origin of the changes? I have a tracking branch, and a release branch. all of my changes get merged (no squash) into release, and then squashed into the tracking branch	23:35
	cmn ^	23:35
	→ Lord_DeathMatch joined	23:36
noahnoah	I think git push -f should do what I want.	23:36
rpk_	There have not been any other changes on the remote tracking branch. It should appear that I just made a series of edits to a set of files all in one commit	23:36
thiago	noahnoah: if you are 100% sure, then do this: git fetch && git merge -s ours ${upstream}	23:36
rpk_	and then merged that commit back from the branch where they were made	23:36
noahnoah	OK. Thanks!!	23:37
thiago	noahnoah: sorry, that's @{upstream}	23:37
noahnoah	OK	23:37
	→ john_anderson joined	23:37
	← ben225 left	23:37
	← cakehero left	23:38
cmn	rpk_: the squashing is where you lie to git about the source of the changes	23:38
	→ user64 joined	23:38
	← user64 left	23:38
	→ limoges joined	23:38
	← noahnoah left	23:38
cmn	and what meaning are you giving to "tracking branch" there?	23:38
rpk_	cmn: then how do I produce a single aggregation of my changes that will merge correctly? The changes themselves are simple	23:39
cmn	you create a merge	23:39
	that's what it's there for	23:39
rpk_	cmn: my git branch is proxied to another version control system	23:39
cmn	then that's probably where the problems come from	23:39
rpk_	I have a remote branch tracking the other VCS. Merging multiple changes breaks the proxy	23:39
	← brngardner left	23:39
rpk_	cmn: but locally, they're just two git branches	23:39
cmn	there are two conflicting histories	23:40
	→ davidban_ joined	23:40
	→ DevilGirl joined	23:40
	← wilmoore left	23:40
rpk_	cmn: so given that, what should I do in order to get it to start merging and picking up changes correctly?	23:41
	← ph^ left	23:41
cmn	rebasing	23:42
Vinnie_win	cmn: I got svnserve running how do I figure out the URL (its on my LAN)	23:42
	← tatsuya_o left	23:42
cmn	svn://localhost/something I guess	23:42
	I've never run a svn server	23:42
	← davidbanham left	23:42
Vinnie_win	cmn: its a miracle...the import is happening	23:43
	→ brngardner joined	23:44
	← thearthur left	23:44
	← cbx33 left	23:45
	← Nibato left	23:45
	→ ph^ joined	23:45
	→ rjwill joined	23:46
	gitinfo set mode: +v	23:46
rjwill	hello, i have a quick git question. i have a master branch with 2 checkins, the first is a vanilla codebase and the second contains a few local changes to the codebase	23:48
Vinnie_win	That's a statement.	23:48
rjwill	i created a new branch called vendor to hold the pristine code since i need to upgrade	23:48
	→ rootfield joined	23:48
rjwill	so now i need to merge the new vendor branch and then apply only the one checkin from the current master to it	23:49
cmn	that's called rebasing	23:49
rjwill	sorry if my git terminology is incorrect	23:49
cmn	then again, since you still haven't asked that quick question	23:49
rjwill	how do i do that	23:49
	← tharkun left	23:50
rjwill	i had to explain the current situation to ask the question	23:50
cmn	that depends on whether you actually want to merge or rebase	23:50
	there is still no question mark in sight, but whatever	23:50
	what you're describing is a rebase	23:50
	← DevilGirl left	23:50
cmn	you take one branch and apply another branch's commits on top	23:50
	→ thearthur joined	23:50
	→ Nibato joined	23:50
rjwill	yes exactly, except I dont want the first commit since it is a previous version of what is currenlty in vendor	23:51
	→ Nathandim joined	23:51
rjwill	it should have been separated out initially i suppose	23:51
	← seank-efi left	23:51
	→ tharkun joined	23:52
	→ elijah joined	23:52
cmn	if you don't want it, what help would it be to have separated out?	23:52
sorahn	st	23:52
	← tharkun left	23:52
	→ tharkun joined	23:52
	← RaceCondition left	23:53
	← ajpiano left	23:53
cmn	and it should already be, from what you explained	23:53
rjwill	i should have had a vendor branch in the beginning	23:53
	tahts what i meant	23:53
cmn	and why didn't you base the vendor branch on the vendor commit you had?	23:54
	→ Jordan_ joined	23:54
elijah	I did a git checkout <hash> to about 10 versions back to match a database that should sync with it. It doesn't appear to be working because doing a git status leaves all sorts of untracked files. Is this normal, am I doing something wrong?	23:54
cmn	a branch can be started at any point	23:54
	← JEntrep left	23:54
cmn	so if it should be based on the root commt, then that's where it should start	23:54
	and fixing it involves rebases as well	23:55
	man git rebase has graphs and explanations of different ways of doing it	23:55
gitinfo	the git-rebase manpage is available at http://jk.gs/git-rebase.html	23:55
	→ dcarper joined	23:55
	→ ajpiano joined	23:55
rjwill	thanks i'll go read that	23:55
	appreciate it	23:55
	→ alamaiso` joined	23:55
rjwill	have a good night	23:56
	← rjwill left	23:56
	← Lord_DeathMatch left	23:56
	← thearthur left	23:56
	← SkAzZ left	23:56
	← heinrich5991 left	23:56
	← limoges left	23:56
	→ RaceCondition joined	23:57
	← alamaiso` left	23:58
	→ alamaiso` joined	23:58
	→ somedude joined	23:58
	← tharkun left	23:59
smw	hi all. I am currently at an old commit in git. Is there any way to see the commits that came right after the current commit on the master branch?	23:59

Logs Search ←Prev date Next date→

Channels Documentation