IRCloggy #git 2012-03-15

Provider of IRC logs since 2005.

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

	2012-03-15
	→ wroathe joined	00:00
	← mister_m left	00:00
cespare	Xano: well, i assumed your central server was a remote called "origin", as that is typically the case.	00:00
	→ JulieHernandez joined	00:00
cespare	I don't know how you cloned the repo in the first place.	00:00
Xano	cespare: that might be the case, but git reset doesn't accept that as an argument	00:00
	cespare: It only accepts a commit hash	00:00
cespare	Xano: false	00:00
	← mritz left	00:00
wxr	can thee help me?	00:00
cespare	Xano: git remote	00:01
	Xano: git branch -r	00:01
	→ sattu94 joined	00:01
cespare	Xano: you can give git reset any ref, including say a sha, or HEAD~12, or origin/master, whatever	00:01
wxr	Can thee tell if thee can help me?	00:02
	I'm really confused	00:02
cespare	wxr: i don't know what you're typing. Clearly there's a big disconnect here.	00:02
	if 'git' gives you command not found.	00:02
Xano	cespare: So that's what the docs mean with "path"?	00:02
	cespare: Because git gave me an error saying origin was not in the path (or something like that)	00:02
cespare	Xano: what does 'git remote' say?	00:03
Xano	cespare: origin	00:03
	→ ben225 joined	00:03
cespare	and what does git branch -r say?	00:03
Xano	cespare: origin/$branchname	00:04
wxr	http://i.imgur.com/R7iiQ.png	00:04
	^ cespare	00:04
	← satyanash left	00:04
cespare	Xano: ok, so you'll notice in my original message I said "assuming your remote is origin, you're on master..."	00:04
	you're not on master	00:04
	so you would do "git reset --hard origin/$branchname"	00:04
Xano	ah right	00:04
cespare	whatever that is	00:04
Xano	geti it	00:04
wxr	cespare: ideeas?	00:05
	→ satyanash joined	00:05
cespare	wxr: ok so maybe some command-line familiarity would be good	00:05
	you typed E:\	00:05
Xano	cespare: That works indeed. I guess I tried origin/master and just origin, both of which failed (the reason for the first fail is obvious)	00:05
cespare	i meant for you do go to that location	00:05
Xano	cespare: Thanks for the clarification :)	00:05
wxr	erm	00:05
cespare	wxr: so	00:06
	cd E:\	00:06
	i think	00:06
	or no \	00:06
	so cd E:	00:06
	just guessing here, I don't use windows	00:06
	then run the git command.	00:06
	Xano: np	00:06
wxr	not a git repository	00:06
	do i have to git init?	00:06
	← Makoryu left	00:06
	→ Fullmoon joined	00:06
cespare	wxr: no.	00:07
	← Mowsh left	00:07
cespare	wxr: one sec, let me track down this repo you're trying to clone	00:07
	→ EvanR joined	00:07
wxr	I'm not trying to clone that one	00:07
EvanR	can you use git comments to manipulate a repo in a script without cding into that repo dir	00:07
	git commands	00:07
	← sattu94 left	00:07
EvanR	not comments	00:07
	→ boombatower joined	00:08
wxr	I'm gonna get from #git to #suicide-watch	00:09
	← mfridh left	00:09
	→ williamcotton_ joined	00:09
cespare	wxr: settle down	00:10
	← williamcotton left	00:10
	williamcotton_ → williamcotton	00:10
cespare	EvanR: Sure. One way is by using GIT_DIR	00:10
wxr	It says it's receiving the objects BUT THERE ARE NO OBJECTS	00:10
cespare	EvanR: so for instance GIT_DIR=path_to_your_thing/.git git ls-files	00:10
EvanR	ah --git-dir	00:10
cespare	EvanR: oh yeah that too :D	00:10
EvanR	alrigh	00:10
wxr	I'm dumb as fuck	00:10
	fuck me	00:10
EvanR	are you female?	00:11
cespare	wxr: what is the exact command you're entering at this point?	00:11
	EvanR: 100% no	00:11
EvanR	dammit	00:11
wxr	It actually worked, thanks	00:11
cespare	wat	00:11
wxr	I'm still gonna join suicide-watch anyways..	00:11
	thanks	00:11
cespare	ಠ_ಠ	00:11
wxr	cespare ur the man	00:11
EvanR	was talking to wxr	00:11
wxr	have a nice day	00:11
EvanR	lol	00:11
	← Russell^^ left	00:11
	← wxr left	00:11
cespare	EvanR: yeah i know. I was saying 100% no	00:12
	→ wxr joined	00:12
	← wxr left	00:12
	← bartek left	00:12
	→ mister_m joined	00:12
	← dotemacs left	00:12
	← fujin left	00:13
	← tfnico left	00:13
	← denysonique__ left	00:13
	← pl0sh left	00:13
	← necromancer left	00:13
	← akraut left	00:13
	← arnihermann left	00:13
	← NimeshNeema left	00:13
	← shennyg left	00:13
	← claudiu__ left	00:13
	← lyddonb__ left	00:13
	← Dr4g_ left	00:13
	← etcetera left	00:13
	← MetaCosm left	00:13
	← CEnnis91 left	00:13
	← jab416171\|Cloud left	00:13
	← rjgonza left	00:13
	← jmcantrell left	00:13
	← GameGamer43 left	00:13
	→ mneorr joined	00:14
	→ neurodrone joined	00:15
	→ Dr4g_ joined	00:15
	→ MetaCosm joined	00:15
	← JulieHernandez left	00:16
	→ akraut joined	00:16
	→ necromancer joined	00:16
	→ mikepack joined	00:16
	← RaceCondition left	00:16
	→ nik0 joined	00:16
	← wroathe left	00:16
	→ warthog9 joined	00:17
	→ gchristensen joined	00:17
	← gmarseliss left	00:17
	→ RaceCondition joined	00:17
	→ wuj joined	00:18
	→ project2501b joined	00:18
	← mneorr left	00:18
	→ wroathe joined	00:18
	gitinfo set mode: +v	00:18
	← ehsan left	00:18
	→ friskd_ joined	00:18
	→ pdtpatr1ck_ joined	00:19
	← EvanR left	00:19
	← pdtpatr1ck_ left	00:20
	→ stkrzysiak joined	00:20
	→ CEnnis91 joined	00:20
	← Anubisss_ left	00:20
	← glcrazy left	00:21
	← williamcotton left	00:21
	← friskd left	00:21
	← mikepack left	00:21
	friskd_ → friskd	00:21
	→ arnihermann joined	00:21
	← pdtpatr1ck left	00:21
	→ sorahn joined	00:22
	→ basepi_ joined	00:22
	← basepi_ left	00:23
	→ SamB_ joined	00:23
sorahn	hey guys, when i run git diff $FILE in my terminal, it spits out something like this. Any idea why?	00:23
	https://gist.github.com/e933a1b0f05fd087a84c	00:23
thiago	you have colouring turned on	00:23
	but I guess your terminal is not colour-capable	00:24
sorahn	it's definitely color capable...	00:24
	→ shennyg joined	00:24
	← SamB left	00:24
	← gusnan left	00:24
sorahn	example: http://sorahn.com/s/9jq4g5mzgyibs4w.png	00:24
	← arvind_khadri left	00:24
	→ fujin joined	00:25
	→ rjgonza joined	00:25
	← basepi left	00:25
	→ denysonique__ joined	00:25
	→ jmcantrell joined	00:26
	← ojimac left	00:26
	← nik0 left	00:28
	→ basepi joined	00:28
cthuluh	sorahn: do you explicitely use a pager (as in git diff ... \| pager)?	00:28
	→ glennpratt joined	00:28
sorahn	cthuluh: no. I just enabled the 'color.ui true' and then some of the colors work (pull requests) and some don't (diff)	00:29
cthuluh	if so, the problem could be your use of "always" instead of "auto" for the color.* config buttons	00:29
	← shibayu36 left	00:29
cthuluh	"always" is the same as "true", iirc	00:29
sorahn	right.	00:29
	→ dotemacs joined	00:29
sorahn	but i don't explicitly push it into a pager	00:29
	it pages on it's own	00:29
	also if i pipe the diff output into a different diff (colordiff) it works fine.	00:30
	← lasers left	00:30
sorahn	'git diff foo.html \| colordiff'	00:30
	→ etcetera joined	00:30
	← warthog9 left	00:31
	← agathecur left	00:31
cthuluh	see if "auto" doesn't solve your problem, in the first place	00:31
	← yoklov left	00:31
cthuluh	(what's a "pull request"? is it related to git or just to some github-like stuff?)	00:31
sorahn	'$ git config --global ui.color auto'	00:31
	er, sorry, i mean, when I did get pull it showed all the +'s in colors.	00:32
	git* pull	00:32
	→ GameGamer43 joined	00:32
cthuluh	ok	00:32
	← thiago left	00:32
	← Noxville left	00:32
	→ pl0sh joined	00:32
sorahn	so, setting ui.color to auto stopped it from showing all the escapes, but it's also not in color.	00:32
	← Xano left	00:33
	← stkrzysiak left	00:34
project2501b	sorahn: echo $TERM	00:34
sorahn	xterm-256color	00:34
	→ tortvanc joined	00:34
project2501b	hmm. that should be just fine. i got color in the same type of $TERM	00:34
cthuluh	sorahn: then your problem is with the pager you use	00:34
sorahn	i wonder if it's a mac thing.	00:34
	← soulcake left	00:35
	→ ojimac joined	00:35
	→ lyddonb__ joined	00:35
sorahn	because on our remote server, (same terminal) the colors show up fine	00:35
cthuluh	sorahn: what's the output of ''git config core.pager''?	00:35
	→ ZGirl joined	00:35
sorahn	nothing	00:35
ZGirl	Hello. I'm curious what might be the way to switch branches with dirty work if I'm expecting merge conflicts?	00:35
cthuluh	then use 'less -R'	00:35
	← ojimac left	00:36
ZGirl	I mean, not merge conflicts but conflicts.	00:36
kevlarman	ZGirl: no	00:36
	← pfFredd left	00:36
sorahn	cthuluh: Yay!	00:36
	← tortvanc left	00:36
kevlarman	ZGirl: you'll have to stash your code, switch branches, and then apply the stash	00:36
	→ mritz joined	00:36
sorahn	thanks!	00:36
	← arietis left	00:37
kevlarman	(and that will generate merge conflicts)	00:37
	→ yshh joined	00:37
cthuluh	sorahn: you're welcome	00:37
ZGirl	Ah sweet, thanks	00:37
	→ arvind_khadri joined	00:38
	→ NimeshNeema joined	00:38
	→ claudiu__ joined	00:39
	→ tfnico joined	00:39
	→ bartek joined	00:39
	→ JulieHernandez joined	00:40
	→ lasers joined	00:41
	→ yoklov joined	00:41
	← ScottE left	00:42
	← y3di left	00:42
	← sudoquai left	00:43
	→ shibayu36 joined	00:44
	→ becomingGuru joined	00:45
	→ nik0 joined	00:46
	→ rbrooks joined	00:46
	→ juvenal joined	00:47
	→ macrover joined	00:48
	→ cooloney joined	00:49
	← dreiss left	00:49
	→ seanh joined	00:49
	← LongBeach left	00:49
	← maletor left	00:50
	← aspotashev left	00:52
	← seanh left	00:52
	← macrover left	00:54
	→ ehsan joined	00:55
	→ seanh joined	00:56
	← seanh left	00:56
	← ericmuyser left	00:56
	← yizumi_ left	00:57
	← ryez_ left	00:57
	→ heppy_ joined	00:57
	← nik0 left	00:57
	← heppy_ left	00:58
	→ Mowsh joined	00:58
	→ kaen joined	00:59
	← kaen left	00:59
	→ kaen joined	00:59
	← friskd left	00:59
	← arvind_khadri left	01:00
	→ friskd joined	01:00
	← heppy left	01:00
	← friskd left	01:00
	→ PsiliPharm joined	01:00
	→ felgru_ joined	01:00
	→ CannedCorn joined	01:01
CannedCorn	how would you git clone all branches	01:01
	?	01:01
	← felgru left	01:01
	→ valtih joined	01:01
	→ friskd joined	01:01
	← friskd left	01:02
	→ warthog9 joined	01:02
	→ yizumi joined	01:02
	→ sattu94 joined	01:03
	← beatak left	01:04
bob2	clone --help	01:04
	note that by default it does what you ask	01:04
	← envi_ left	01:04
bob2	but it does not make local matching branches	01:05
	← spaceonline left	01:05
	→ xpot-mobile joined	01:06
	← satyanash left	01:07
	← warthog9 left	01:07
	→ mo0nykit joined	01:08
	→ friskd joined	01:09
	→ Turicas joined	01:09
mo0nykit	How do I fetch one file from another branch and bring it into the current branch?	01:09
	→ Slix` joined	01:10
	← SegFaultAX\|work left	01:10
	← yshh left	01:11
	→ arvind_khadri joined	01:12
	← arvind_khadri left	01:12
	→ arvind_khadri joined	01:12
	← marsilainen left	01:12
bob2	I think you mean 'git checkout SOMEREF -- SOMEFILE'	01:12
	→ pdtpatr1ck joined	01:12
bob2	but do run 'git status' after that to understand what it did	01:12
	→ ChaoWang joined	01:13
mo0nykit	bob2: SOMEREF can be the name of that other branch?	01:15
bob2	it's a reference to a commit yes	01:16
	← diogogmt left	01:16
mo0nykit	bob2: thank you, got it now	01:16
	→ ericmuyser joined	01:16
	← ZGirl left	01:16
	→ nik0 joined	01:16
	← muddo left	01:16
	→ muddo joined	01:17
	→ diogogmt joined	01:17
	→ xiangfu joined	01:19
	← Destos left	01:19
	→ heppy joined	01:20
	→ forsto joined	01:20
	← andres left	01:20
	← valtih left	01:21
	→ andres_f joined	01:22
	← Riviera left	01:22
	→ Goplat joined	01:23
	← adulion22 left	01:23
	→ Riviera joined	01:25
	← nik0 left	01:27
	← btanaka left	01:28
	← yizumi left	01:28
	← eletuchy left	01:28
	→ yizumi joined	01:31
	→ Skipp_OSX joined	01:32
Skipp_OSX	hello, is there a way to see a diff for a specific commit given the commit number?	01:32
bremner	sure, man git-diff	01:33
gitinfo	the git-diff manpage is available at http://jk.gs/git-diff.html	01:33
project2501b	there is no "commit number"	01:33
	but yes, sure	01:33
Skipp_OSX	by commit number I mean signature	01:33
	→ Samual joined	01:34
project2501b	you mean the sha1 hash	01:34
	terminology is important.	01:34
Skipp_OSX	yeah, that is what I mean	01:34
	if I do `git diff <hash>` it gives me all the changes from that sha to HEAD, I just want to see the changes committed in that hash	01:35
	← zoran119 left	01:36
	← XayOn_ left	01:36
	→ XayOn joined	01:36
	← candybar left	01:36
SethRobertson	`git show SHA` will work	01:37
	as will `git diff SHA^...SHA`	01:37
	→ mfridh joined	01:37
Skipp_OSX	SethRobertson: hey, that worked great! thanks	01:39
SethRobertson	!thanks	01:39
gitinfo	Feeling thankful? Type "ExampleUser++", and ExampleUser will score karma points at http://carmivore.com (our preferred way to objectify self-worth). There's really no point to thanking me – instead, why not thank the person who made me help you?	01:39
Skipp_OSX	SethRobertson++	01:40
	→ candybar joined	01:40
	← _inc left	01:40
SethRobertson	cespare: Did you get an answer to your question? !submodule_rm	01:40
gitinfo	cespare: [!submodules_rm] You want to delete submodules? Excellent choice! All commands are in the superproject. Edit/delete .gitmodules to remove the submodule. Then `rm -rf submodulepath; git rm -f --cached submodulepath; git commit -am "Removed submodules!"` Inspect .git/config for "submodule" entries to remove. Inspect .git/modules for caches to remove "!gitslave" or "!subtree" (type them!) might be alternatives	01:40
	← candybar left	01:40
	→ _inc joined	01:41
	→ candybar joined	01:41
	→ warthog9 joined	01:41
	← CannedCorn left	01:42
	← zivester left	01:43
	→ herdingcat joined	01:44
	→ CannedCorn joined	01:45
	→ zdk joined	01:46
	→ nik0 joined	01:46
	→ resmo__ joined	01:46
	← _inc left	01:47
	→ hadronzo` joined	01:47
	→ zivester joined	01:47
	→ subbyyy joined	01:47
	← felgru_ left	01:47
	→ oriba_ joined	01:47
	← CannedCorn left	01:48
	← gllera left	01:48
	← cooloney left	01:48
	← herdingcat left	01:49
	← mister_m left	01:49
	→ cooloney joined	01:49
	← resmo_ left	01:50
	← hadronzoo left	01:50
	← oriba left	01:50
	← tatsuya_o left	01:52
	→ tatsuya_o joined	01:52
	→ _inc joined	01:53
	→ BiggFREE joined	01:55
	→ QshelTier joined	01:55
	← Bombe left	01:56
	← berserkr left	01:56
	← Riviera left	01:57
	← nik0 left	01:57
	← zdk left	01:58
	→ zdk joined	01:58
	← xpot-mobile left	02:00
	→ pw_ joined	02:03
	← mcluskydodallas left	02:05
	← matthiaskrgr left	02:05
	← apok left	02:05
	← gnychis left	02:05
	← ben225 left	02:06
	← Turicas left	02:06
	→ matthiaskrgr joined	02:07
	← jameswarren left	02:07
	← yoklov left	02:08
	→ iamjarvo joined	02:10
	→ herdingcat joined	02:11
	← PsiliPharm left	02:11
	← wroathe left	02:11
	→ guns joined	02:11
	→ zoran119 joined	02:12
	← Skipp_OSX left	02:12
mo0nykit	bob2++	02:13
	← oriba_ left	02:14
	→ kai_ joined	02:14
	← BlindOracle left	02:15
	← diogogmt left	02:16
	→ nik0 joined	02:16
EugeneKay	SethRobertson - it's worth noting that the karmabot is no longer idling here	02:16
	Which makes me a sad karmawhore	02:16
SethRobertson	EugeneKay++	02:17
	← pw_ left	02:17
SethRobertson	That is odd since they have data as of yesterday in their system	02:18
	← jigal left	02:18
	→ yoklov joined	02:18
	← aeontech left	02:19
	← JulieHernandez left	02:19
	→ diogogmt joined	02:19
	← mfridh left	02:19
	→ Targen joined	02:20
	EugeneKay shrugs	02:21
	→ wroathe joined	02:21
	→ pandeiro joined	02:21
	← Vampire0_ left	02:21
	→ williamcotton joined	02:21
	→ friskd_ joined	02:23
	→ mister_m joined	02:24
	← friskd left	02:24
	friskd_ → friskd	02:24
	→ jangell joined	02:24
	← RaceCondition left	02:25
	← friskd left	02:25
	→ mfridh joined	02:25
	← enmand left	02:27
	→ friskd joined	02:27
	← pdtpatr1ck left	02:28
	← nik0 left	02:28
	← bitkiller left	02:30
	kennethreitz → _kennethlove	02:30
	_kennethlove → daniellindsley	02:31
	← pro_metedor left	02:31
	← cannonball left	02:31
	← warthog9 left	02:32
	← becomingGuru left	02:33
	→ becomingGuru joined	02:33
	→ maletor joined	02:33
	← daniellindsley left	02:33
	→ BlindOracle joined	02:34
	→ LimitedAtonement joined	02:35
	← heppy left	02:35
	→ Vampire0_ joined	02:35
	→ kennethreitz joined	02:35
	← maletor left	02:36
	← ura left	02:37
	← LmtdAt left	02:38
	← unreal left	02:39
	→ Per4u3e joined	02:40
	→ unreal joined	02:41
	← edheldil left	02:41
	→ SegFaultAX joined	02:42
	← adamm left	02:43
	→ adamm joined	02:45
	→ laxatives_ joined	02:46
	→ nik0 joined	02:46
	← Araxia left	02:46
laxatives_	how do I commit to a url? can I commit to a url? I've just remote added, but when I try to remote push, theres nothing to push	02:47
SethRobertson	!repo	02:47
gitinfo	[!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.	02:47
SethRobertson	Sorry, !repro	02:47
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript of your terminal session. This will help immensely with troubleshooting.	02:47
laxatives_	https://gist.github.com/2041438	02:47
	nevermind, i have no idea what changed, but it went through this time	02:49
SethRobertson	You were trying to use `git remote add -A filename` when you wanted to use `git add filename`	02:49
	→ Yuuhi` joined	02:50
SethRobertson	`git remote` manages your connection to other repositories, and typical+l-y w	02:50
	→ mo0nykit1 joined	02:50
SethRobertson	those repositories would not have the name of a .c source code file	02:50
	← unreal left	02:51
laxatives_	eh nevermind it didn't stick. I got the adding part, I thought I cuold use -A like I could for my main repo, but I guess not. The file added, but now that it exists in the remote repo, I can't figure out how to push a newer version since it tells me everything is up to date	02:51
	but I can clearly see on github its the old version	02:52
	← Yuuhi left	02:52
SethRobertson	You can use `git add -A` if you must	02:52
	← mo0nykit left	02:52
SethRobertson	Continue with your pastebin of what you did	02:52
	← arvind_khadri left	02:53
laxatives_	http://pastebin.com/raZK2Nzy	02:54
SethRobertson	We already had the discussion that `git remote add` is not the droid you are looking for. I strongly suggest reading !book, but as a shortcut:	02:55
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://progit.org/book/ but also look at !bottomup !cs !designers !gitt !vcbe and !parable	02:55
SethRobertson	git add sgemm-scall.c; git commit -m "I changed this file"	02:55
	git push	02:55
	← friskd left	02:56
	→ JessicaHernandez joined	02:56
SethRobertson	The git-push may or may not work depending on how you created/cloned this repo	02:56
	← xiangfu left	02:56
	→ xiangfu joined	02:57
	← nik0 left	02:57
laxatives_	thanks Seth, seems to have solved the issue	02:58
	→ sgm[-1] joined	02:59
	→ pdtpatr1ck joined	02:59
	→ dv310p3r joined	03:04
	← wroathe left	03:04
	→ CannedCorn joined	03:05
	→ arvind_khadri joined	03:06
	← nazgul101 left	03:08
	→ i-make-robots joined	03:09
	→ unreal joined	03:10
i-make-robots	hi, git. how do i change the config so git stops forcing me to upload with windows line endingS?	03:10
SethRobertson	i-make-robots: !crlf	03:10
gitinfo	i-make-robots: to fix problems with line endings on different platforms, check out http://line-endings.gitrecipes.de/	03:10
	→ duckxx joined	03:11
	← starlon left	03:11
	← wuj left	03:13
	← el2ro left	03:14
i-make-robots	sorry, that article doesn't seem to help. git is expecting me to convert to CRLF before I commit and I want to make it stop that.	03:14
SethRobertson	Either you have a pre-commit hook which is doing it, or you have eol/crlf settings in your .git/config	03:15
	← duckxx left	03:15
SethRobertson	The article talks about the latter	03:15
	→ nik0 joined	03:16
	← marcuy_ left	03:19
	→ aeontech joined	03:20
	← Fullmoon left	03:21
	→ Fullmoon joined	03:22
	→ nazgul101 joined	03:22
	← CannedCorn left	03:23
i-make-robots	so... all I do is create a file called .gitattributes, tell it that my problem file is text, and then add & commit that file?	03:23
	→ gtar joined	03:24
bob2	to be clear, by default git doesn't fiddle your files	03:24
SethRobertson	Seems unlikely	03:24
	→ CannedCorn joined	03:24
	→ el2ro joined	03:25
SethRobertson	i-make-robots: Check to see if you have a file .git/hooks/pre-commit which is executable	03:26
cmn	that should work, if you tell git that you want crlf conversion	03:26
	← gtar left	03:26
SethRobertson	he claims git isn't letting him commit with wrong EOL settings	03:26
	→ Fullmoon_ joined	03:27
	→ AlbireoX joined	03:27
	→ Fullmoon__ joined	03:27
i-make-robots	fatal: lf would be replaced by crlf in [problem file]	03:27
bob2	when doing what? commit? push?	03:27
i-make-robots	commit	03:27
cmn	why would git transform to crlf?	03:28
	that has to be a setting you've put	03:28
	← Fullmoon left	03:28
	Fullmoon__ → Fullmoon	03:28
	← CannedCorn left	03:28
i-make-robots	shrug this is my first project, I'm lucky i got it running at all.	03:28
	i mean, "yes, i agree. but what to do about it?"	03:28
	← nik0 left	03:28
bob2	to clarify, everyone is kinda implying that someone did something odd to your config or repo	03:29
	did you make the repo yourself?	03:29
cmn	check your config	03:29
	git config --list	03:29
SethRobertson	core.safecrlf is the man git-config setting you want to look at	03:29
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	03:29
	← killfill left	03:29
i-make-robots	cmn - core.autocrlf=true, core.safecrlf=true	03:30
SethRobertson	That or core.autocrlf or core.eol	03:30
i-make-robots	i don't see a core.eol in my config.	03:30
	→ killfill joined	03:30
SethRobertson	which means native, and I will wildly guess windows for you	03:30
i-make-robots	yep.	03:30
	git config --global core.safecrlf="false" ?	03:31
	← Fullmoon_ left	03:31
SethRobertson	well, changing autocrlf to warn will allow the commits, but also see http://stackoverflow.com/questions/1598260/make-crlf-warnings-go-away	03:31
	← becomingGuru left	03:32
	← Mowsh left	03:33
	← johnkpaul left	03:34
i-make-robots	thank you! :)	03:38
	← JessicaHernandez left	03:38
	→ gabeh joined	03:39
	← Slix` left	03:42
	← kadoban_ left	03:43
	sattu94 → satyanash	03:44
	← laxatives_ left	03:44
	→ stkrzysiak joined	03:46
	→ nik0 joined	03:46
	← i-make-robots left	03:47
	← jangell left	03:50
	→ dr0id joined	03:50
	← xsbeats left	03:51
	→ alx- joined	03:51
	→ igotnolegs joined	03:52
	← ericmuyser left	03:53
	← Bass10 left	03:56
	→ lyddonb joined	03:56
iamjarvo	do you guys prefer a gui like source tree or gitx or command line	03:57
	← nik0 left	03:57
nevyn	depends what it's for.	03:57
	mostly I tend to use commandline	03:57
	sometimes if I think the history mgiht be a bit of a tangle it's useful to visualise it in gitk	03:58
	→ ericmuyser joined	03:58
	→ jony joined	03:58
	→ ThePawnBreak joined	03:58
	→ malandro95 joined	03:59
iamjarvo	nevyn: true. i like command line but im starting to like the guis. seems to help visualizations	04:00
	← orafu left	04:00
	→ orafu joined	04:00
	→ jangell joined	04:02
bob2	gitk is handy for history stuff	04:04
	magit is awesome for almost everything else	04:04
milki	emacs mode for git?	04:05
	o, no, thats probably something else	04:05
bob2	more like a little application inside emacs then a mere mode	04:05
milki	or...not	04:05
	← f0i left	04:06
iamjarvo	wonder if fugitive for vim is good	04:06
	→ dfr\|mac joined	04:07
	→ gnufied joined	04:08
	← gabeh left	04:08
	← griotspeak left	04:09
	→ zanneth joined	04:12
	← BiggFREE left	04:12
	← kai_ left	04:15
	← twisted` left	04:16
	→ twisted` joined	04:16
	→ nik0 joined	04:16
	→ gavin_huang joined	04:17
	← BryanRuiz1 left	04:19
	← nik0 left	04:19
	← Per4u3e left	04:20
	→ friskd joined	04:20
	← Fullmoon left	04:20
	← twisted` left	04:20
	← JackWinter left	04:20
	← dfr\|mac left	04:21
	← flippo left	04:22
	← infogulch left	04:22
	← lindenle_ left	04:24
	→ Noxville joined	04:24
	→ beautifulmind joined	04:26
	← lyddonb left	04:29
	→ `Assassin joined	04:29
	→ envi_ joined	04:29
	→ lyddonb joined	04:29
	→ harshpb joined	04:30
	→ gabeh joined	04:31
	← lyddonb left	04:32
	← subbyyy left	04:33
	→ bindaasomatic joined	04:33
	← bindaasomatic left	04:33
	← pandeiro left	04:33
	→ jameswarren joined	04:34
	← mgpcoe left	04:34
	← Sadin left	04:35
	← kukks left	04:35
	→ kenperkins joined	04:36
	→ lindenle_ joined	04:37
	← OOPMan left	04:37
	→ heppy joined	04:38
	→ arvind_k joined	04:38
	← arvind_khadri left	04:38
	← seoaqua left	04:40
	→ MaskRay joined	04:40
	← sporous left	04:43
	→ sporous joined	04:43
	→ tyatpi joined	04:43
	← kenperkins left	04:43
	← friskd left	04:45
	→ arvind__ joined	04:46
	→ nik0 joined	04:46
	→ JackWinter joined	04:47
	← arvind_k left	04:47
	→ mo0nykit joined	04:47
	← mo0nykit left	04:47
	→ mo0nykit joined	04:47
	← mo0nykit left	04:47
	→ mo0nykit joined	04:48
	← mo0nykit left	04:48
	→ mo0nykit joined	04:48
	← mo0nykit left	04:48
	→ mo0nykit joined	04:49
	← mo0nykit left	04:49
	→ mo0nykit joined	04:49
	← mo0nykit1 left	04:49
	→ mgpcoe joined	04:51
	→ ura joined	04:52
	← arvind__ left	04:52
	← beautifulmind left	04:52
	→ Per4u3e joined	04:53
	← ThePawnBreak left	04:53
	← ericmuyser left	04:53
	→ Davey_ joined	04:54
	← Davey_ left	04:54
	→ Davey_ joined	04:54
	← mo0nykit left	04:54
	→ pidus joined	04:54
	→ mo0nykit joined	04:54
	← xinming left	04:55
	← eno left	04:56
	← nik0 left	04:58
	→ eno joined	04:58
	← thursday left	04:59
	→ arvind__ joined	05:00
	← SegFaultAX left	05:01
	→ DofLeFou joined	05:06
	← diogogmt left	05:07
	← jony left	05:09
	→ mo0nykit1 joined	05:10
	← mritz left	05:11
	← madewokherd left	05:12
	← mo0nykit left	05:14
	→ mo0nykit joined	05:14
	← mo0nykit1 left	05:15
	← kaen left	05:15
	→ nik0 joined	05:16
	← joshsmith left	05:20
	← neurodrone left	05:21
	← `10 left	05:22
	← mandric left	05:23
	← williamcotton left	05:25
	→ mritz joined	05:26
	← mister_m left	05:26
	konr → nomad	05:27
	← nomad left	05:28
	← nik0 left	05:28
	→ rurufufuss joined	05:29
	→ weiyang joined	05:30
	← zanneth left	05:31
	← pidus left	05:31
	→ stepnem joined	05:33
	→ pidus joined	05:34
	← mo0nykit left	05:35
	→ pidus_ joined	05:35
	← pidus left	05:38
	pidus_ → pidus	05:38
	← stkrzysiak left	05:38
	← envi_ left	05:39
	← arvind__ left	05:40
	→ Davey__ joined	05:40
	← Davey_ left	05:40
	← glennpratt left	05:40
	→ mo0nykit joined	05:41
	→ arti_t joined	05:41
	→ jony joined	05:42
	→ ericmuyser joined	05:42
	→ krz joined	05:43
	← timoxley left	05:43
	→ freimatz joined	05:43
	← mo0nykit left	05:45
	→ nik0 joined	05:46
	→ timoxley joined	05:47
	→ Fullmoon joined	05:48
	← alx- left	05:49
	→ dreiss joined	05:51
	→ wtking joined	05:52
	→ caseymcg joined	05:52
wtking	Does anyone know the status of ETags with gitweb (especially snapshots)?	05:52
	← satyanash left	05:53
	→ satyanash joined	05:54
	→ warthog9 joined	05:55
	← hadronzo` left	05:57
	→ f0i joined	05:57
	→ hw1979 joined	05:57
	← yizumi left	05:58
	← nik0 left	05:58
	→ yizumi joined	05:58
	← malandro95 left	06:01
	← yizumi left	06:01
	→ yizumi joined	06:01
	← harshpb left	06:02
wtking	nevermind, I'll ask on the mailing list	06:03
	← wtking left	06:03
	→ mwrosen joined	06:04
	→ harshpb joined	06:04
	← timoxley left	06:05
	← mgpcoe left	06:05
	→ FACEFOX joined	06:06
	← FACEFOX left	06:06
	← WillMars_ left	06:07
	→ FACEFOX joined	06:08
	← FACEFOX left	06:08
	← Kewlj1313 left	06:08
	← arti_t left	06:09
	→ arti_t joined	06:09
	→ giallu joined	06:15
	→ nik0 joined	06:16
	← dreiss left	06:18
	→ kaen joined	06:18
	← kaen left	06:18
	→ kaen joined	06:18
	→ Kewlj1313 joined	06:19
	← freimatz left	06:21
	→ fairuz_ joined	06:23
	← jony left	06:27
	← nik0 left	06:27
	→ Dave^\| joined	06:28
	→ freimatz joined	06:29
	→ shruggar joined	06:30
	→ OOPMan joined	06:31
	→ friskd joined	06:31
	← giallu left	06:31
	← Davey__ left	06:31
	← arti_t left	06:32
	→ arti_t joined	06:33
	← shruggar left	06:37
	→ LongBeach joined	06:41
	→ dvide joined	06:42
	← yoklov left	06:42
	← toothr left	06:42
	← arti_t left	06:43
	→ arti_t joined	06:43
	← mwrosen left	06:44
	→ flijten joined	06:44
	→ mishok13 joined	06:46
	→ libertyp1ime joined	06:46
	→ nik0 joined	06:46
	← libertyprime left	06:47
	← libertyp1ime left	06:47
	→ MestreLion joined	06:47
	→ libertyprime joined	06:48
	→ toothr joined	06:48
	→ erickr joined	06:48
	→ marthinal joined	06:49
	→ JohnBat26 joined	06:49
MestreLion	guys, I've found weird behavior in git 1.7.1... $GIT_EDITOR, $VISUAL, $EDITOR env vars are NOT set, core.editor in config is NOT set, vi executable is set to vim, and still git opens.. nano! How come?	06:49
	→ dreiss joined	06:50
MestreLion	(the only "especial" thing about nano in my system is that /usr/bin/editor points to it... but Git's documentation does not ever mention using editor command as one of the alternatives)	06:51
	← `Assassin left	06:52
MestreLion	any clues? is this a known issue?	06:52
	← yizumi left	06:52
MestreLion	man git-vars	06:54
	→ martinjlowm joined	06:54
MestreLion	bot offline?	06:55
	← number_number left	06:57
	← Goplat left	06:58
	→ yizumi_ joined	06:58
MestreLion	man git vars	06:59
gitinfo	the git manpage is available at http://jk.gs/git.html	06:59
	← nik0 left	06:59
	→ `Assassin joined	07:00
	← vdv left	07:01
	← iamjarvo left	07:02
MestreLion	ok, git var -l shows me GIT_EDITOR=editor , but this is NOT coming from environment var $GIT_EDITOR, since this is empty. Any hint on where this value came from?	07:03
	→ jokergoo joined	07:03
cmn	core.editor	07:03
	or just the default	07:03
MestreLion	nope, core.editor is also not set	07:03
cmn	then it's the default	07:03
MestreLion	and where this default is set?	07:04
cmn	somewhere in the source code	07:04
	open_editor() or something like that	07:04
MestreLion	so docs should be fixed	07:04
nevyn	MestreLion: what's update-alternatives --list editor say?	07:04
	or default-system-editor maybe	07:04
cmn	maybe, what docs are you talking about?	07:04
nevyn	check /etc/alternatives	07:04
MestreLion	nevyn: as I said earlier, editor IS set to nano... but there's nowhere in git's docs hat say it will try editor command	07:05
	← bigkitty left	07:05
nevyn	editor is what an application should open as a text editor on a system with alternatives	07:05
	unless the user overrides	07:05
	→ cck joined	07:05
cmn	MestreLion: so the default should be to fail?	07:06
	← craigntammy left	07:06
	← pidus left	07:06
	← hw1979 left	07:06
nevyn	the whole point of editor was to stop stupid vi vs emacs vs blah arguments based on individual packager/developer's whim	07:06
MestreLion	nope... it should try vm, as man git says	07:07
gitinfo	the git manpage is available at http://jk.gs/git.html	07:07
nevyn	vm?	07:07
	or vi	07:07
MestreLion	vi, sorry	07:07
nevyn	the docs are buggy file a bug.	07:07
MestreLion	I'm trying to find the URL / man page that described the behavior	07:07
nevyn	what does the man page FOR your os say?	07:07
	not the generic git man page?	07:08
MestreLion	but I remember reading: $GIT_EDITOR, core.editor , $VISUAL, $EDITOR, vi	07:08
	in that order	07:08
nevyn	MestreLion: what os are you running?	07:08
MestreLion	Mint 10	07:09
cmn	it's in var	07:09
MestreLion	it was man git-var	07:09
gitinfo	the git-var manpage is available at http://jk.gs/git-var.html	07:09
cmn	but your distribution can override it	07:09
nevyn	which mint varient?	07:09
	← ianmcorvidae left	07:09
MestreLion	" The order of preference is the $GIT_EDITOR environment variable, then core.editor configuration, then $VISUAL, then $EDITOR, and then finally vi."	07:09
	→ oehman joined	07:09
cmn	unless whoever builds the code overrides it	07:09
nevyn	that's in the mint manpage?	07:09
	file a bug.	07:10
MestreLion	nevyn: main edition, x64, based on Ubuntu 10.10	07:10
nevyn	if it doesn't work that way.	07:10
cmn	git does default to vi	07:10
	it's probably your distro	07:10
MestreLion	but "editor" should be in the list	07:10
	→ number_number joined	07:10
nevyn	cmn: upstream git might	07:10
	→ tsdh joined	07:10
cmn	what might?	07:10
MestreLion	hummm, maybe my git was patched by distro to try editor before vi?	07:10
nevyn	but debian->ubuntu->mint git should not	07:10
cmn	MestreLion: then file a bug with your distro	07:10
nevyn	yes	07:11
MestreLion	I'll check if there are any patches	07:11
cmn	editor.c:6:#define DEFAULT_EDITOR "vi"	07:11
nevyn	that's what I'm saying the docs are buggy. because the package was altered to conform to policy but the docs wern't	07:11
	← friskd left	07:11
MestreLion	i'll download the sources for my distro and check that out cmn	07:11
	→ Sigma joined	07:12
	→ ianmcorvidae joined	07:12
	← ianmcorvidae left	07:12
	→ ianmcorvidae joined	07:12
MestreLion	(btw, Mint uses Ubuntu's repos for git... so it's either Ubuntu or inherited from Debian	07:12
nevyn	http://www.debian.org/doc/debian-policy/ch-customized-programs.html#s11.4	07:12
	→ _AlbireoX joined	07:13
	→ krisfrem1n joined	07:13
	→ cbreak-work joined	07:14
	→ RickXy joined	07:14
	→ Targen_ joined	07:14
	→ tewecske joined	07:15
	→ com4_ joined	07:15
nevyn	MestreLion: the git-var manpage on debian has that it calls vi	07:15
cmn	yes, Debian changes the default to 'editor'	07:16
	so it matches the systerm	07:16
	→ jmd joined	07:16
	→ pidus_ joined	07:16
	→ XayOn_ joined	07:16
	→ Quanteek joined	07:16
	→ nik0 joined	07:16
	→ rogg joined	07:16
	← pidus_ left	07:16
	→ esc_ joined	07:16
	→ cubb` joined	07:17
nevyn	so the docs are wrong	07:17
	on debian	07:17
MestreLion	but they didn't patch that file	07:17
jmd	How can I update my local branch, from a remote, discarding any changes in the local one?	07:18
nevyn	jmd: git reset --hard remote/branch	07:18
	after doing git remote update	07:18
jmd	nevyn: Thanks.	07:18
	← ianmcorvidae left	07:20
	→ papegaaij joined	07:20
	→ ph^ joined	07:20
MestreLion	weird... there are no patches concerning editor in the package	07:20
	→ heidar_ joined	07:20
	→ _main_ joined	07:21
nevyn	and $EDITOR isn't set in your shell?	07:21
	by /etc/profile	07:21
	→ airborn joined	07:21
	← AlbireoX left	07:21
	← Targen left	07:21
	← XayOn left	07:21
	← forsto left	07:21
	← rjgonza left	07:21
	← esc left	07:21
	← sobel left	07:21
	← __main__ left	07:21
	← lonewulf` left	07:21
	← philips left	07:21
	← bolt left	07:21
	← com4 left	07:21
	← charon left	07:21
	← heidar left	07:21
	← krisfremen left	07:21
	← akiress left	07:21
	← rrs left	07:21
	← Strogg left	07:21
	← Akari` left	07:21
	→ ianmcorvidae joined	07:21
	← ianmcorvidae left	07:21
	→ ianmcorvidae joined	07:21
MestreLion	oh...	07:22
cmn	MestreLion: it's not a patch	07:22
MestreLion	its a built-time configure	07:22
nevyn	it's not?	07:22
MestreLion	it's a configure option	07:22
	→ philips joined	07:22
nevyn	ah	07:22
	← gabeh left	07:22
nevyn	hrm	07:22
MestreLion	check configure.ac	07:22
	hrm ?	07:22
	← jokergoo left	07:22
nevyn	so if you can build it using a configure option in a way that makes the manpage wrong...	07:23
	who should fix it?	07:23
MestreLion	git, since this is an upstream configure option	07:23
	← mishok13 left	07:23
	→ shruggar joined	07:23
MestreLion	it should change the man page accordinly	07:23
nevyn	that's sortof what I was thinking	07:23
MestreLion	or edit the manpage to a more generic version	07:23
	a simple "This can be overruled at compile time" in man git-var would be enough	07:24
gitinfo	the git-var manpage is available at http://jk.gs/git-var.html	07:24
	_main_ → __main__	07:24
	→ Sitethief joined	07:25
nevyn	cmn: thoughts?	07:25
	← ianmcorvidae left	07:25
	→ ianmcorvidae joined	07:26
	← ianmcorvidae left	07:26
	→ ianmcorvidae joined	07:26
MestreLion	./configure --help \| grep editor	07:26
cmn	if you can provide a patch that doesn't break things, it might get accepted, but it's not really an issue	07:26
MestreLion	--with-editor=VALUE Use VALUE as the fall-back editor instead of 'vi'.	07:26
	→ bolt joined	07:27
	→ TTSDA_ joined	07:27
MestreLion	theres even an example... which says... /usr/bin/editor :D	07:27
	← TTSDA left	07:28
	← fisted_ left	07:28
	→ akiress joined	07:28
	← nik0 left	07:28
	← TTSDA_ left	07:28
	→ TTSDA_ joined	07:28
	→ rjgonza joined	07:28
	TTSDA_ → TTSDA	07:28
MestreLion	What would be the best approach? ./configure edits the manpage at build time, or manpage text says about the overridable ./configure option?	07:30
	→ fisted joined	07:30
MestreLion	are you guys familiar with git's contribution workflow? I have no idea where to start	07:31
cmn	read the docs on it	07:31
	Documentation/SubmittingPatches	07:32
	→ flavius joined	07:32
MestreLion	where are these docs? /usr/share/doc ?	07:32
nevyn	MestreLion: the latter of those sounds easier.	07:32
	MestreLion: in the source	07:32
	→ ipalaus joined	07:32
	← adamm left	07:32
nevyn	MestreLion: the former sounds "better"	07:33
	→ sq-one joined	07:33
	← arti_t left	07:34
	→ bigkitty joined	07:34
	→ arti_t joined	07:35
	→ giallu joined	07:35
cmn	they're in git.git	07:35
	→ jony joined	07:36
	→ arvind_khadri joined	07:37
	← arvind_khadri left	07:37
	→ arvind_khadri joined	07:37
	neersighted → neersighted\|AFK	07:38
MestreLion	and is there an official git.git URL?	07:39
cmn	there are a few; use whichever one you feel like	07:39
	kernel.org, github.com, git.or.ca	07:40
	git.or.cz, I mean	07:40
	or code.google.com	07:40
	→ jceb joined	07:40
	→ Fullmoon_ joined	07:40
	← Fullmoon_ left	07:40
	← Per4u3e left	07:40
	← jony left	07:40
	← erickr left	07:41
	→ Error404NotFound joined	07:41
	← Error404NotFound left	07:42
	← Sigma left	07:43
	← Fullmoon left	07:43
	→ rtuin joined	07:43
	← siprbaum_ left	07:43
	→ psoo joined	07:45
	→ aspotashev joined	07:45
	→ mishok13 joined	07:46
	← arti_t left	07:46
	→ twb joined	07:46
	→ nik0 joined	07:46
	→ arti_t joined	07:47
twb	I have a repo that contains secrets, which other users on the system should not be able to access. This is straightforward to achieve -- just make sure you run "umask 077" before doing any work in that repo. However, can I tell git to do this "inside itself", just for that repo? I looked at git-config manpage and I couldn't see how to do it.	07:47
	← cooloney left	07:48
cmn	git doesn't track permissions, only the executable bit	07:48
	→ Chib joined	07:48
	← dr0id left	07:49
	→ glcrazy joined	07:50
twb	git has --git-dir and --work-tree args, which avoid having to cd into the repo before running git.	07:50
	I'm hoping for something similar w.r.t. umasks	07:50
	← tyatpi left	07:50
twb	Otherwise if I go into that repo and do something like git bisect or checkout, and forget to set umask first, the files git creates will be more permissive than they should	07:51
shruggar	cmn: I think he means that others shouldn't be able to access the entire repo	07:51
twb	IOW I don't need git to remember the permissions of a single file; I need it to remember that files it creates (including in .git) should not be group/world readable.	07:51
shruggar	twb: I believe umask is the only way to achieve this	07:51
	→ ThomasLocke joined	07:51
twb	shruggar: okey dokey	07:51
	I guess there's still one feature darcs has that git doesn't ;-)	07:52
	← acrocity left	07:52
cmn	git isn't interested in that feature	07:52
	← bigkitty left	07:53
	→ acrocity joined	07:53
twb	Yeah, well, fair enough, it's just a little annoying in this specific edge case	07:53
shruggar	twb: you should set your umask anyway, as it's important that other programs you use to edit your worktree respect the setting too	07:53
	twb: if you only care about the repo, not the worktree, use gitolite	07:53
twb	shruggar: I do set my umask, but my general umask is different from my repo umask	07:53
	← ptrb left	07:54
twb	shruggar: this repo is basically stuff like .ssh/id_rsa, so I do care about the work tree, and it's not really feasible to say "oh just put a web smarthost in the middle"	07:54
cmn	that wasn't a proposed solution	07:55
	→ napoletani1 joined	07:55
cmn	you shouldn't be putting secret keys in a repo anyway	07:55
twb	cmn: why not?	07:55
	How else do you know why they changed?	07:55
cmn	why would private keys change?	07:55
	they're cryptographically secured	07:55
twb	Because the expire, or are compromised, or a new algorithm is released	07:55
	→ martinjlowm-lptp joined	07:56
	→ gldn joined	07:56
	← napoletani1 left	07:56
cmn	but what for?	07:56
twb	Or you have a new email address and need to add its subkey	07:56
	← Karmaon left	07:56
cmn	for ssh keys? that sounds more like gnupg	07:57
twb	Sure. They're also in the repo	07:57
cmn	PGP/GPG does need sync	07:57
	but the private ssh keys don't	07:58
	→ erickr joined	07:58
	← nik0 left	07:58
	← jmd left	07:58
twb	Well, suppose we're only talking about GPG keys instead, then.	07:58
cmn	then you need rsync	07:58
twb	I have no fucking clue what you're talking about	07:58
cmn	man rsync	07:59
shruggar	"I'm trying to use git to manage something other than what git is for, and I'm running into problems because of it. Please help"	07:59
	→ seoaqua joined	07:59
cmn	you don't need versioning	07:59
	→ mabrand joined	07:59
	← airborn left	07:59
twb	How would rsync remember who/what/why a change was made to a keyring?	07:59
cmn	you need syncrhonisation	07:59
	→ sajbar joined	07:59
	← ph^ left	08:00
twb	shruggar: I want to control versions. With a version control system. It would be nice if git had could help me keep it secure, but you said it can't, so that's fine. But now I seem to need to defend the need for VCS at all, so that's what I'm doing.	08:00
cmn	if you do feel your keyring absolutely needs to be versioned, then you need to set the umask when you checkout	08:00
	→ ph^ joined	08:01
twb	cmn: right. And any other operation that might create a file in the work tree or index or object tree, of course, since otherwise an attacker could just pull a historic version out of an insecure object tree.	08:01
	← f0i left	08:01
cmn	right, which becomes a deployment problem	08:02
	→ rendar joined	08:03
	← aspotashev left	08:03
	→ iocor joined	08:03
	← arti_t left	08:03
	→ arti_t joined	08:04
	→ kremiak joined	08:04
	← igotnolegs left	08:05
twb	Incidentally, core.sharedRepository appears to exist so people can go the other way -- have a restrictive default umask, and a more permissive repo	08:05
	← ph^ left	08:05
twb	Hmm, wait, what happens if I just set that to 0600...	08:06
	→ as3llus joined	08:06
cmn	no, sharedrepository exists so several users can use the same repo	08:06
	← ipalaus left	08:07
cmn	without making stuff unreadable/unwritable to people who come afterwards	08:07
twb	I don't understand why you think that's different from "permissive"	08:07
	← Quanteek left	08:07
	← shruggar left	08:07
	← hyperair left	08:08
	← stringoO left	08:08
cmn	it's about the sticky bit	08:09
	→ shruggar joined	08:10
twb	Huh, OK. That's not obvious (to me) from the documentation.	08:10
cmn	it's not about letting other people read you data, but work with them	08:10
	that's because you use that stuff so you don't have to set the bits yourself	08:10
	→ jutaro joined	08:11
	→ Adaptee joined	08:11
twb	Looks more like setgid bit than sticky bit from my simple test	08:11
	← mritz left	08:12
	→ khmarbaise joined	08:13
twb	http://paste.debian.net/159799/	08:13
	→ scientes joined	08:13
scientes	!lol	08:13
gitinfo	git config --global alias.lol "log --oneline --graph --decorate"	08:13
	→ senny joined	08:14
cmn	right, that might be it; sticky is probably the /tmp one	08:15
twb	Unsurprisingly, it looks like it affects permissions within GIT_DIR but not GIT_WORK_TREE	08:15
	cmn: yep	08:15
	→ ipalaus joined	08:15
cmn	those are not meant to have a worktree	08:15
twb	cmn: sticky says you can only delete files in it if you own them, or thereabouts. setgid says any files made in there default to being grouped same as the parent dir	08:15
	← shruggar left	08:16
twb	although setgid doesn't do anything about permissions, which is probably why the sharedRepo variable is needed -- so git will make sure the permissions are OK	08:16
	→ nik0 joined	08:16
	→ tqmz joined	08:17
	← heppy left	08:18
	← ipalaus left	08:20
	→ Akari` joined	08:22
	← boombatower left	08:22
	← lausi left	08:25
	← gavin_huang left	08:25
	→ lausi joined	08:25
	→ gavin_huang joined	08:25
	→ bigkitty joined	08:27
	→ Karmaon joined	08:27
	← Karmaon left	08:27
	→ Karmaon joined	08:27
	← kerrick left	08:27
	_aeris_\|away → _aeris_	08:28
	← nik0 left	08:28
	← oehman left	08:28
	← nowhere_man left	08:28
	→ nowhere_man joined	08:29
	→ kpreid joined	08:30
	← seoaqua left	08:30
	→ dpino joined	08:31
	→ ph^ joined	08:32
	→ ojimac joined	08:33
	→ shachaf_ joined	08:33
	← shachaf_ left	08:33
	→ solofight_ joined	08:34
	← tatsuya_o left	08:35
solofight_	wow - https://bitbucket.org/	08:35
	so till 5 users we can have unlimited private and public repos without any charge and without a time constraing ! ?	08:35
	constraint*	08:35
	their service is reliable ?	08:35
	solofight_ → solofight	08:36
cmn	it's likely a loss leader for Atlassian	08:36
	← ology left	08:36
solofight	cmn: ? loss leader ?	08:36
	→ macrover joined	08:37
	→ whitman joined	08:37
jast	business term	08:37
	→ hyperair joined	08:37
jast	http://en.wikipedia.org/wiki/Loss_leader	08:37
solofight	got it	08:37
	to gain customers	08:37
	→ seoaqua joined	08:37
jast	yeah, that's the idea	08:37
cmn	right, once you're at their servers, you're more likely to keep using them when you outgrow the free offer	08:37
	← ojimac left	08:38
	← ericmuyser left	08:38
solofight	:) learned two new things today till now	08:39
	thanks jast cmn	08:40
	← flavius left	08:41
	← tonesfrommars left	08:42
	← twb left	08:42
	→ drizzd joined	08:42
	← arvind_khadri left	08:43
	→ ipalaus joined	08:43
	→ eijk joined	08:44
	→ nik0 joined	08:46
solofight	cmn: how do you know that they are incurring loss with that service initially to get business ?	08:47
	bitbucket is the only provider who offers unlimited private repositories for free ?	08:48
	till 4 users ?	08:48
	5 users*	08:49
FauxFaux	solofight: Well, they're not making any money from it, are they..	08:49
solofight	FauxFaux: dont know - wildly guessing - ads ? like how bloggers make money .. ?	08:50
jast	well you can use your repository without even visiting the site much, can't you :)	08:50
FauxFaux	solofight: There aren't any ads. :)	08:51
	→ shruggar joined	08:51
	← toothr left	08:52
	← MaskRay left	08:54
	→ anderse_ joined	08:54
	→ toothr joined	08:56
	→ Sigma joined	08:56
solofight	FauxFaux: yeah :)	08:56
	jast: yes :P	08:56
	you need to use the site for creation alone i think	08:57
	→ LmtdAt joined	08:57
solofight	unless they have console access as well	08:57
cmn	solofight: you need to pay money to keep the servers running and to pay for bandwidth and for storage	08:58
	if you're not paying for it, they're losing money on you	08:58
	← nik0 left	08:58
	→ d3vic3 joined	08:59
	← marthinal left	09:00
	← sawgij left	09:00
	← LimitedAtonement left	09:00
solofight	cmn: understood	09:00
	thanks	09:00
	→ mo0nykit joined	09:01
mo0nykit	!remote	09:01
gitinfo	[!remote_branches] Remote branches are not created locally on clone. They show up as remote branches (git branch -r). To create local branches from them use git checkout -b local_name remote/branch_name or more recently git checkout -t remote/branch-name	09:01
mo0nykit	How do I inspect the URL of an existing remote?	09:02
FauxFaux	git remote -v	09:02
wereHamster	mo0nykit: git remote -vv	09:02
FauxFaux	git remote -vvv	09:02
wereHamster	git remote -vvvv	09:02
FauxFaux	JINX	09:02
mo0nykit	thank you	09:02
	→ blacktulip joined	09:04
	→ sami joined	09:04
sami	Hi, i've got a problem which i don't know how to explain. How can i clone a git repo without an absolut path.	09:05
FauxFaux	With a relative path!	09:06
	← bruce__ left	09:06
sami	like the following git@<server>:/dont/want/full/path/to/this.git	09:06
	→ beautifulmind joined	09:06
FauxFaux	sami: You mean, you want it relative to your home directory? Or you want to alias away the /dont/want/full/path bit?	09:06
	→ nik0 joined	09:07
	→ diegoviola joined	09:07
	→ soulcake joined	09:07
sami	FauxFaux: if the repo is called this.git i want to clone it by <server>:this.git that's what i'm having trouble to explain. Sorry	09:07
	nik0 → nicko	09:08
	→ CannedCorn joined	09:08
sami	FauxFaux: But i guess i want it relative to my home dit	09:08
	dir*	09:08
FauxFaux	I don't think you can do that (beyond changing the user's home or symlinks). It might be easier just to install !gitolite	09:09
gitinfo	Want to host as many git repos (and users!) as you like, on your own server, with fine-grained access control? You want gitolite: https://github.com/sitaramc/gitolite	09:09
	→ Vile joined	09:09
FauxFaux	It /is/ relative to your home dir. "git clone you@server:foo.git" looks in /home/you/foo.git.	09:09
	→ libertyp1ime joined	09:09
sami	I'm sorry FauxFaux :). Didn't think it was relative to my homedir.	09:11
	I've got gitosis installed so the repos exist within the ~repositories directory	09:11
	← beautifulmind left	09:12
	→ tcurdt joined	09:13
	← sami left	09:13
	← nicko left	09:14
	← xiangfu left	09:15
	← bigkitty left	09:17
	→ defts joined	09:17
	→ arvind_khadri joined	09:19
	← CannedCorn left	09:20
	→ thiago joined	09:20
	← Orbitrix left	09:21
	→ forsto joined	09:22
	→ beautifulmind joined	09:23
	→ tvw joined	09:23
	← drizzd left	09:25
	→ BiggFREE joined	09:25
	← Sigma left	09:26
	← QiBala left	09:26
	← [mbm] left	09:26
	→ icdark joined	09:26
	← neurodamage left	09:26
	← shuffle2 left	09:26
	← legumbre left	09:27
	← flazz left	09:27
	← artagnon left	09:27
	→ mo0nykit1 joined	09:28
	← hyperair left	09:29
	← mo0nykit left	09:30
	→ marcogmonteiro joined	09:31
	→ zub joined	09:31
	→ envi_ joined	09:32
	→ LimeBlast joined	09:33
	← L42y left	09:34
	→ jameslordhz joined	09:34
jameslordhz	hi	09:34
	→ L42y joined	09:34
	← kremiak left	09:36
	→ Element9_work joined	09:37
	→ overbythere joined	09:37
jameslordhz	hi	09:39
	← OOPMan left	09:40
	← mo0nykit1 left	09:41
cbreak-work	hihihi	09:42
	→ harish joined	09:42
	→ Guest16470 joined	09:42
	→ sneakret joined	09:43
	→ artagnon joined	09:43
	← khmarbaise left	09:43
	→ flazz joined	09:43
	→ Riviera joined	09:43
	→ shuffle2 joined	09:44
	→ neurodamage joined	09:44
	→ teotwaki joined	09:44
	← Guest16470 left	09:44
	→ Guest16470 joined	09:44
	← lasers left	09:44
	← dreiss left	09:45
	→ charon joined	09:45
	← charon left	09:45
	→ charon joined	09:45
	→ _iron joined	09:46
	→ tatsuya_o joined	09:46
	→ mo0nykit joined	09:47
	← BlindOracle left	09:51
	→ soc42 joined	09:52
	→ hyperair joined	09:53
	→ Fullmoon joined	09:53
	← cyphase left	09:53
	→ falena joined	09:54
	QshelTier → Bombe	09:54
	← weiyang left	09:55
	← jameslordhz left	09:55
	→ BlindOracle joined	09:55
	← JackWinter left	09:56
	← sgm[-1] left	09:58
	← shruggar left	09:58
	← iocor left	09:59
	→ JackWinter joined	09:59
	→ f0i joined	09:59
	→ xiangfu joined	10:00
	→ vervic joined	10:03
sigs	hey give me some too	10:03
	→ stodan joined	10:03
	→ iocor joined	10:04
	← Tabmow left	10:05
	← iocor left	10:05
	← herdingcat left	10:06
	← harish left	10:06
	→ Mowsh joined	10:06
	→ shruggar joined	10:07
	→ y3di joined	10:08
	→ Tabmow joined	10:08
	→ Orbitrix joined	10:09
	→ francisfish joined	10:09
	← mo0nykit left	10:10
	→ timoxley joined	10:10
	← forsto left	10:10
	→ felgru joined	10:11
	→ enmand joined	10:15
	→ johnathanb joined	10:15
	← dpino left	10:15
	→ Shusshu joined	10:17
	← babilen_ left	10:18
	← enmand left	10:18
	← Shusshu left	10:19
	→ fayimora joined	10:19
	→ Quanteek joined	10:20
	← seoaqua left	10:22
	→ zoran119_ joined	10:24
	← erickr left	10:24
	→ erickr joined	10:25
	→ cyphase joined	10:25
	← thiago left	10:27
	← zoran119 left	10:27
	→ dpino joined	10:27
MestreLion	is there an official bug tracker or something for git?	10:28
	← hyperair left	10:29
	← felgru left	10:29
charon	MestreLion: just send an email with as much detail as possible to [email@hidden.address] (no subscription required)	10:29
MestreLion	tank you charon	10:29
	thank*	10:30
charon	of course, prepare for "please test with v1.7.9.4" if your report states you're way behind :)	10:30
	MestreLion: you can search the list archives at gmane (tinyurl.com/gitml) if you want to look for old reports and such	10:30
	→ cortana2 joined	10:30
	← caseymcg left	10:31
MestreLion	the docs urls that gitinfo bot provide are from latest git, correct?	10:31
	← tcurdt left	10:31
	← `Assassin left	10:31
charon	right. i suspect they're even built from the master branch, but i'm too lazy to check	10:33
	→ gusnan joined	10:35
	→ iocor joined	10:37
	← kpreid left	10:37
	← selckin left	10:38
	→ cmyk_ joined	10:39
	← cmyk_ left	10:39
	← iocor left	10:40
	← canton7 left	10:42
	→ canton7 joined	10:42
	← solofight left	10:42
	→ bigkitty joined	10:43
bigkitty	hi	10:43
	→ enmand joined	10:45
	← dhruvasagar left	10:45
MestreLion	programming question about building git from source: i've noticed there's no ./configure in tarball/repo. Which files are read when make configure is executed to generate it?	10:45
	← Chib left	10:46
	→ Chibby joined	10:46
MestreLion	because the bug lies in the ./configure executable.. and I just realized git repo has none	10:46
shruggar	MestreLion: have you tried just running "make" without configure?	10:46
	→ dhruvasagar joined	10:47
MestreLion	not really shruggar... but I need to run ./configure to reproduce the bug	10:47
	or at least to properly document it	10:47
	← enmand left	10:48
	← fisted left	10:48
shruggar	MestreLion: there's a configure.ac, I expect that's related, but I've never been able to wrap my head around autoconf/automake	10:48
	there's also a config.mak.in	10:49
m1sc	MestreLion: details in INSTALL	10:49
MestreLion	it's a bug in the manpages... git-var fails to mention that the last fallback for editor, vi, is actually configurable at compile time	10:49
	← cck left	10:49
MestreLion	in my distro source package there is a ./configure , in git repo there is not	10:50
	m1sc: I was reading it before coming here	10:51
	it only says "make configure"	10:51
cmn	that's because ./configure scripts aren't ever in source control	10:51
	→ eijk_ joined	10:51
cmn	it gets created by autoconf	10:51
ndim	Should not ever be in source control	10:52
cmn	right	10:52
	if it is, then it's a bug	10:52
	EugeneKay makes rude noises about autoconf	10:52
	← arti_t left	10:53
cmn	you probably don't want to use it in git, anyway	10:53
	let the Makefile do it for youi	10:53
	→ arti_t joined	10:53
ndim	And the creation usually happens not by directly calling autoconf. Instead you call "autoreconf" (or, in some cases, a custom "autogen" or "bootstrap" script which is in source control).	10:53
	Can't remember what git uses off hand.	10:54
	← eijk left	10:54
	← fayimora left	10:55
	← ChaoWang left	10:55
MestreLion	I'm just trying to determine who made the --with-editor configure option, upstream or distro	10:56
bremner	that doesn't sound like a git question	10:58
	necessarily	10:58
	→ thiago joined	10:59
	→ netiru joined	11:00
	← netiru left	11:00
MestreLion	bremner: it is about giving git an accurate and detailed bug report	11:01
bremner	ok	11:01
MestreLion	instead of simply saying "man git-var does not document the actual behavior regarding editor choice"	11:02
gitinfo	the git-var manpage is available at http://jk.gs/git-var.html	11:02
	→ tingo joined	11:02
bremner	MestreLion: so, according to upstream, the way to build git (documented in INSTALL) is "make"	11:04
	→ FernandoBasso joined	11:04
bremner	or, you can read more about autoreconf in that same file	11:05
	notice that "make configure" is the first step.	11:05
	→ RaceCondition joined	11:05
	← BiggFREE left	11:05
	→ cmyk_ joined	11:06
	→ fisted joined	11:06
	← zdk left	11:07
	← ph^ left	11:08
	← Dave^\| left	11:08
	→ sattu94 joined	11:08
	→ seanh joined	11:09
	→ libertyp2ime joined	11:10
MestreLion	ok, --with-editor is indeed provided by a ./configure created in a defautlt "make configure" run	11:10
	← ehsan left	11:10
	→ ph^ joined	11:10
MestreLion	so yes, this is an upstream issue	11:10
	← libertyp1ime left	11:11
	→ hyperair joined	11:11
	→ BiggFREE joined	11:11
	← icdark left	11:12
	bremner grumbles about reading INSTALL for people, demands a raise.	11:12
	→ harish joined	11:13
	→ noisebleed joined	11:13
	← noisebleed left	11:13
	→ noisebleed joined	11:13
	← f0i left	11:14
	→ icdark joined	11:14
	→ fayimora joined	11:14
	← fayimora left	11:15
	← jceb left	11:15
	← kaen left	11:15
EugeneKay	!refund	11:16
gitinfo	If you are not satisfied with git for whatever reason, you are entitled to a full refund of the purchase price, and are invited to use another VCS. Elsewhere.	11:16
EugeneKay	I move we give bremner a 50% raise, from $0.00 to $0.00.	11:17
cmn	overriding the default editor is provided by the build system	11:17
	and the default is correct in git-var.txt	11:17
	→ spaceonline joined	11:18
	← dhruvasagar left	11:19
	← flaguy48 left	11:19
	→ dhruvasagar joined	11:19
	→ flaguy48 joined	11:19
	← spaceonline left	11:21
	→ jatt joined	11:21
	→ cooloney joined	11:22
	→ Lujeni joined	11:23
	→ iocor joined	11:23
	→ berserkr joined	11:23
	← vervic left	11:24
	← cooloney left	11:24
	← BiggFREE left	11:24
	→ traviscj joined	11:26
	→ spaceonline joined	11:27
	→ armetiz joined	11:27
	← seanh left	11:27
	← Sitethief left	11:29
	→ psakrii joined	11:29
	→ pidus joined	11:29
	← macrover left	11:31
	→ i42n joined	11:31
	→ RaceCondishun joined	11:32
MestreLion	cmn: true, but it should mention that this is configurable at build time	11:33
	vi being just a default	11:33
cmn	maybe, but "the default defaults to 'vi'" is a stupid sentence	11:34
MestreLion	or, as you said, git-var.txt should be changed by make to reflect the selected choice	11:34
	← pidus left	11:34
MestreLion	I'll suggest this workaround sentence:	11:34
	This last fallback editor can be chosen at build-time using configure option --with-EDITOR, which is, by default, 'vi'.	11:34
	right after the "finally, 'vi'." part	11:35
	← RaceCondition left	11:36
	RaceCondishun → RaceCondition	11:36
MestreLion	is that a good phrasing?	11:36
	→ enmand joined	11:38
MestreLion	maybe add "hardcoded" as in "This last hardcoded fallback editor" ?	11:38
cmn	man git-var shouldn't talk about build options	11:38
gitinfo	the git-var manpage is available at http://jk.gs/git-var.html	11:38
EugeneKay	SethRobertson - apparently I started working on that ssh_config doc while drunk.... https://gist.github.com/1917102	11:39
MestreLion	cmn: I could omit defailts like the name of the configure option, but what do you think about mentioning this is configurable at build time?	11:40
	← sq-one left	11:40
MestreLion	is it worth it, at least until someone properly chang the build system to edit git-var.txt at build time/	11:42
	← Quanteek left	11:42
	→ delight joined	11:44
	→ enmand_ joined	11:44
	← enmand left	11:44
cmn	I don't see it being a big deal	11:45
	you can say that the fallback is configurable, possibly vi	11:45
	but you should be setting your own editor anyway	11:45
	and the default value isn't that important	11:45
	← i42n left	11:45
	→ fanaticgeek joined	11:45
	→ i42n joined	11:46
	→ Cromulent joined	11:46
	← arti_t left	11:46
	→ iamjarvo joined	11:47
	→ impr0bable joined	11:48
	→ erichynds joined	11:48
	← erichynds left	11:49
	→ erichynds joined	11:50
MestreLion	i agree this is a very minor documentation issue	11:50
	← scientes left	11:50
MestreLion	but it did puzzle me for a while, and I had to investigate quite a bit to find out what was going on	11:50
	a single clarification would have saved me (and possibly others who stumble on that) an hour or so	11:51
	→ digitalfiz joined	11:53
IcePic	why is getting the fallback "right" so important?	11:54
	I would understand if it was poorly documented how you get your own favourite editor used, but the fallback is .. the fallback.	11:54
MestreLion	not getting the fallback 'right', IcePic .. getting the docsright	11:55
	→ johnkpaul joined	11:55
IcePic	yes, so why would people spend time in docs to change the build-time fallback?	11:55
	builders, that is.	11:55
	← RaceCondition left	11:55
MestreLion	i don't want to change it...	11:55
	→ the_cybersphinx joined	11:55
MestreLion	i was just puzzled where git was choosing my editor, since I don't have any of the documented vars set	11:55
	→ cannonball joined	11:56
	← d3vic3 left	11:56
jast	yeah, I think the question is: who would benefit from having that documented? only someone who wants to change it. how many such people are there? :}	11:56
	← fanaticgeek left	11:56
jast	I'm not saying that I oppose a doc patch	11:56
MestreLion	and it was supposed to open 'vi' in this case, but it was opening nano instead	11:56
EugeneKay	You know, for the amount of time you've <s>wasted</s> spent talking about it, you could submit a patch to the mailing list.	11:56
MestreLion	I am doing that, EugeneKay ..	11:57
	Just discussing what a proper phrasing would be	11:57
	← cybersphinx left	11:58
jast	[...] and then finally _vi_ (or whichever editor was set at build time using DEFAULT_EDITOR or --with-editor).	11:58
	just an idea	11:58
	gitinfo set mode: +v	11:59
sajbar	hey there, is there a way to modify post-recieve hook so it only triggers on certain branches?	11:59
IcePic	I'm just wondering, if I am a builder and set Eclipse (or whatever) as the last resort and dont communicate that to my package users, would that text really help?	11:59
MestreLion	jast: cmn suggested a manpage should not discuss build options... so I guess i'll go with: "... `$VISUAL`, then `$EDITOR`, and then finally, a hardcoded fallback editor set at build time, by default 'vi'." - How about that?	11:59
jast	IcePic: sure, but then again no variant would help in that case	12:00
MestreLion	yes it would IcePic ... because git-var is the only place in docs that document Git's search order for editor	12:00
IcePic	if I get nano, which isnt my EDITOR nor VISUAL, and its not vi, what use is the manpage for me as a user of said buidl-time-changed git	12:00
EugeneKay	sajbar - of course. Loop through the stdin and do an if [] check on the ref-name.	12:00
jast	we can't really patch the manpage at build time because manpages are often generated completely separately	12:00
MestreLion	IcePic: I was puzzled because i don't have ANY of the env vars set, nor core.editor.. so where git was reading nano from? that's a legitmate user question, not just builder's	12:02
	→ cccaldas joined	12:02
	→ Kenjin joined	12:02
Kenjin	hello	12:02
IcePic	MestreLion: and the last manpage "quote" you suggested would not help you with that either.	12:03
	so I guess i'll go with: "... `$VISUAL`, then `$EDITOR`, and then finally, a hardcoded fallback editor set at build time, by default 'vi'.	12:03
	that one.	12:03
	→ arti_t joined	12:03
	→ arietis joined	12:03
MestreLion	of course it would... I would know that nano (or /usr/bin/editor) was proabaly picked as the hardcoded fallback	12:04
	which, for my distro, was not the default 'vi'...	12:04
	← envi_ left	12:04
IcePic	how would that differ from your experience of seeing nano today?	12:04
MestreLion	mistery solved in 5 minutes instead of 50	12:04
	← eno left	12:04
IcePic	something wants nano started <- thats all you really know.	12:04
	in both cases	12:04
MestreLion	it would differ as I would understand what is going on	12:05
	→ CannedCorn joined	12:05
	→ eno joined	12:06
MestreLion	IcePic: my point is simply that it was confusing (and surprising) to see nano poping up	12:07
	← tingo left	12:07
jast	my suggestion is don't spend too much time debating the wording before submitting it. resending patches is cheap. if it turns out the maintainer is violently opposed to accepting any such change at all, you'll end up wasting much more time if you polish it beforehand. ;)	12:07
	← yizumi_ left	12:07
MestreLion	good point jast	12:07
jast	just make it 'good enough'	12:07
bigkitty	hi all	12:08
	← iamjarvo left	12:08
	← cccaldas left	12:08
cmn	!hi	12:09
gitinfo	[!welcome] Welcome to #git, a place full of helpful gits. If you have a question, feel free to just go ahead and ask—somebody should answer shortly. For more info on this channel, see http://jk.gs/git/ Take backups (type !backup to learn how) before taking advice.	12:09
	← mklappstuhl left	12:09
	← hyperair left	12:10
	→ RaceCondition joined	12:11
	→ cccaldas joined	12:12
	→ boo1 joined	12:13
	→ Ryback_ joined	12:13
	→ griotspeak joined	12:14
	→ dr0id joined	12:14
	← soulcake left	12:17
	← boo1 left	12:18
	→ babilen joined	12:18
	← bigkitty left	12:19
	← arti_t left	12:19
	→ bigkitty joined	12:19
	← dv310p3r left	12:20
	→ arti_t joined	12:20
MestreLion	and guys, sorry about being what can be seen as picky. I was trying to help others that may stumble on this too, and may get puzzled like I was. Maybe a doc patch may not help much, but i think it would at least lead users to the right direction, ie, check the build options used in his package. Sometimes just knowing that this fallback is configurable is enough for him.	12:20
	→ mklappstuhl joined	12:22
CannedCorn	hey guys, how would you clone all branches	12:24
	of a repo	12:24
MestreLion	CannedCorn: a remote repo?	12:24
	CannedCorn: when you clone a repo, you already get all its branches	12:25
jast	CannedCorn: by default all branches are cloned, they just don't automatically get created as local branches in the target repo (they do show up in git branch -r)	12:25
	← berserkr left	12:25
	→ kryl99 joined	12:26
soc42	hi #git	12:26
	are there any security issues (flaws or known vulnerabilites) while using git?	12:27
	→ angelsl joined	12:28
CannedCorn	how do i make them show up in git branches	12:28
	i.e. have them locally	12:28
	← arti_t left	12:29
	→ arti_t joined	12:29
	← beautifulmind left	12:29
Kenjin	is it possible to do a partial cherry-pick. Something along the lines off add -p ?	12:31
MestreLion	CannedCorn: if all you want is to list them, use branch -a (shows all branches, local and remote)	12:31
	→ marthinal joined	12:31
	← mfridh left	12:31
SethRobertson	soc42: nop	12:31
MestreLion	CannedCorn: but there is little point in locally creating all remote branches. Create only the ones you will actually work on	12:32
soc42	hi SethRobertson	12:32
	→ mandric joined	12:32
jast	CannedCorn: in general they get created automatically when you do "git checkout <name of branch>"	12:32
SethRobertson	Kenjin: Yes, run with -n and then you can evict from the index and add -p.	12:32
	→ d0k joined	12:33
CannedCorn	yeah, ok	12:33
	← martinjlowm left	12:33
CannedCorn	there are 100s of branches on this project and i want to be able to switch to them all without having internet connectivity so that isn't going to work :-(	12:34
	oh well	12:34
	→ tommyvyo joined	12:34
SethRobertson	Um, yes it is	12:34
MestreLion	CannedCorn: you don't need internet connectivity	12:34
SethRobertson	You already have a copy of all bits. It doesn't go anywhere when it createa a local branch	12:34
jast	CannedCorn: the contents are cloned automatically. "git checkout" just makes them visible as a local branch, too.	12:34
MestreLion	CannedCorn: a "remote branch" is a local name that you have in your cloned repo	12:35
jast	just try it, you'll see	12:35
Kenjin	SethRobertson: cool thanks	12:35
jast	it happens way too fast for any connection thing to be involved :)	12:35
	let's test the newest change to the bot...	12:36
	→ jasttest joined	12:37
	← gitinfo left	12:37
jasttest	hi there!	12:37
jast	well that didn't work. :}	12:37
CannedCorn	MestreLion SethRobertson cool, thanks	12:37
	← jasttest left	12:37
MestreLion	CannedCorn: think of a remote branch as just a "tag" that points to a given commit. A commit you have locally because you cloned the repo. Once you clone a repo, you have all the data you need to work, no more internet required	12:37
jast	interestingly the bot is still running, just not connected anymore	12:37
	→ gitinfo joined	12:38
	ChanServ set mode: +o	12:38
SethRobertson	!branch	12:38
	→ Dave^\| joined	12:38
SethRobertson	?	12:38
jast	it takes a while to sync, remember?	12:38
gitinfo	A branch and a tag are just convenient ways of spelling the name of a particular commit. A commit represents a specific set of files and the history of all commits which came before it, and the SHA-1 hash tag official name provides cryptographic assurance of the lineage of a particular commit (and thus branch or tag). A branch's reference may change. A tag usually doesn't.	12:38
SethRobertson	Not in the slightest	12:38
jast	oh, it wasn't actually running. that was just the keepalive shell script. waits a bit and then restarts it.	12:39
	→ jasttest joined	12:39
jasttest	hi there. testing the bot and stuff.	12:39
gitinfo	jasttest: 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 a>nd 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.	12:39
jasttest	hi there. should not trigger again.	12:39
MestreLion	CannedCorn: so the remote branches are just names that reflect the current status of the server remote repo the last time you fetched from it	12:39
jasttest	excellent.	12:40
	← jasttest left	12:40
SethRobertson	jast: "a>nd this channel"?	12:40
	→ mfridh joined	12:40
jast	whoops	12:40
	that must have crept in while copying the line from one terminal to another	12:40
	→ amigojapan joined	12:41
jast	now, did I have a command for rehashing the config file?	12:41
	yes I did!	12:41
	← cortana2 left	12:41
jast	.rehash	12:41
gitinfo	jast: Okay.	12:41
jast	fixed :}	12:41
	← martinjlowm-lptp left	12:41
	→ seanh joined	12:41
	← seanh left	12:42
	→ seanh joined	12:43
	← arti_t left	12:43
	→ d3vic3 joined	12:43
	→ arti_t joined	12:44
	→ beautifulmind joined	12:45
	← jargon- left	12:45
	← cccaldas left	12:45
	← harshpb left	12:45
	→ cccaldas joined	12:45
	← thiago left	12:46
	→ memoryleak joined	12:47
	→ reels joined	12:47
	→ neurodrone joined	12:49
	← neurodrone left	12:49
	→ neurodrone joined	12:49
	← schlumpi left	12:50
	← tatsuya_o left	12:50
	→ tatsuya_o joined	12:51
	→ vmassuchetto joined	12:52
	→ TheDudeKnows joined	12:52
	→ jargon- joined	12:52
	← Cromulent left	12:52
vmassuchetto	How can I 'force' a push to be made again? I want a 'post-update' script that reads branches to be executed again, but for another branch this time.	12:53
	← arti_t left	12:54
SethRobertson	make a change	12:54
	→ ralf_e joined	12:54
vmassuchetto	SethRobertson, yeah... but I was wondering if there's something else that could be done.	12:55
SethRobertson	Once the bits have been transferred and the branch pointed updated remotely, that is it.	12:55
	→ schlumpi joined	12:56
	→ arti_t joined	12:57
	→ dzonder joined	12:58
	← dpino left	12:58
	← marcogmonteiro left	12:58
	→ marcogmonteiro joined	12:59
	← as3llus left	12:59
	← jatt left	12:59
vmassuchetto	Is there a way to make git to get the remote servers from a commited file? Like it's done with files on .gitignore?	12:59
	← psakrii left	13:00
FauxFaux	vmassuchetto: I don't understand your question.	13:00
SethRobertson	I don't either, but the answer is no	13:00
	→ wroathe joined	13:00
	← marcogmonteiro left	13:00
	→ as3llus joined	13:00
vmassuchetto	FauxFaux, Like a '.gitconfig' file to be at the root of the repository with the remote servers, branches and so...	13:00
	The information that currently goes into .git/config	13:01
	→ AaronMT joined	13:01
FauxFaux	You want the .gitignore not to be part of the repository, such that it's not committed or transfered around with clone? man gitignore "excludes".	13:01
gitinfo	the gitignore manpage is available at http://jk.gs/gitignore.html	13:01
FauxFaux	Uh, "exclude". I always get that wrong. Causes so much confusion.	13:01
vmassuchetto	Faux. No.	13:02
FauxFaux	Explain further!	13:02
	← seanh left	13:03
	→ jsoa joined	13:03
	← kryl99 left	13:03
	→ zoran119 joined	13:03
vmassuchetto	FauxFaux, I want a file to distribute to other developers that tells their local gits where's the information that I'm putting now into '.git/config' file related to '[branch]' and '[remote]' options.	13:03
jast	we don't have any such thing at this time	13:03
FauxFaux	What he said.	13:03
	→ hyperair joined	13:04
	← zoran119_ left	13:04
vmassuchetto	Ok.... But would it be useful?	13:04
jast	it would raise a few hairy questions, too	13:04
	for example: what if that file clashes with .git/config?	13:04
	personally I wouldn't find it terribly useful	13:04
vmassuchetto	jast, Yeah.. Some sort of precedence would be needed.	13:05
FauxFaux	git config was going to be getting #include at some recent point, wasn't it, so I can start versioning my .gitconfig again.	13:05
	→ seanh joined	13:05
jast	yeah, I think I saw someone working on that	13:05
vmassuchetto	jast, Nice. If you have a link or anything would be nice.	13:06
	← soc42 left	13:06
jast	I was talking about includes	13:06
	← shruggar left	13:06
jast	that wouldn't really do anything for your situation	13:06
	← ThomasLocke left	13:06
	→ TeckniX joined	13:06
	→ chilversc_ joined	13:07
FauxFaux	Well, you could have a template of #include "../.vmassuchettoattributes" for your .gitconfig.	13:07
	← arti_t left	13:07
	← Guest58816 left	13:07
	→ Guest58816 joined	13:07
	Guest58816 → UukGoblin	13:08
vmassuchetto	FauxFaux, Ok. I'll try to get more information about it.	13:08
	→ vervic joined	13:08
vmassuchetto	Thanks guys.	13:08
	← vervic left	13:08
	→ dv310p3r joined	13:11
	→ thiago joined	13:11
	← Kuba left	13:11
	→ empity joined	13:11
	→ john2496 joined	13:12
	→ mikewintermute joined	13:13
	→ glennpratt joined	13:14
	→ nenjordi joined	13:14
	← iocor left	13:14
	→ stkrzysiak joined	13:15
	→ Bass10 joined	13:16
	← vmassuchetto left	13:16
	→ ceej joined	13:16
	→ n1x joined	13:16
	→ cortana2 joined	13:17
	→ pidus joined	13:17
	→ williamcotton joined	13:17
	← diegoviola left	13:17
	→ jeffisabelle joined	13:17
	→ Vortex34 joined	13:18
	→ bitkiller joined	13:18
	→ bLUEEE joined	13:19
	gitinfo set mode: +v	13:19
	gitinfo set mode: +v	13:19
bLUEEE	hi there	13:19
gitinfo	bLUEEE: 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.	13:19
	← hyperair left	13:19
	→ hyperair joined	13:19
	← teotwaki left	13:20
jast	it works! \o/	13:20
	← Vortex35 left	13:21
	→ newbie1 joined	13:22
EugeneKay	high time.	13:22
bLUEEE	small question, i pushed everything to the origin. but then i edited .gitignore file and pushed again. i was thinking that the origin would actually ignore the newly added file.. but it didnt.. how can i do that ?	13:23
EugeneKay	bLUEEE - !untrack	13:23
gitinfo	bLUEEE: to stop git from tracking a file, without deleting the file, use "git rm --cached <file>"	13:23
bLUEEE	thanks	13:23
EugeneKay	gitignore ignores only untrackd files. You have to remove it from git in order to untrack it. ;-)	13:23
	s/untrack/ignore	13:23
	← newbie1 left	13:23
bLUEEE	got it !	13:23
EugeneKay	You can also force addition of an ignored file with git add -f	13:24
	→ Sigma joined	13:24
	→ teotwaki joined	13:24
	→ mgpcoe joined	13:25
bLUEEE	oh man, i deleted the file accidently	13:25
EugeneKay	man git-checkout	13:25
gitinfo	the git-checkout manpage is available at http://jk.gs/git-checkout.html	13:25
EugeneKay	git checkout HEAD -- foo.txt will undisappear it	13:26
	→ prgmctan joined	13:26
	→ fanaticgeek joined	13:27
prgmctan	anyone ever use git to version whole database sql files? let's say you have development, staging, and production databases	13:28
	EugeneKay makes a rude noise	13:28
jast	prgmctan: you're probably better off using the snapshot/dump feature of your database	13:28
EugeneKay	Datatbase schema, or database data?	13:28
jast	merging is pointless with database images, anyway	13:28
	→ Zenith77 joined	13:28
MacGyverNL	I have, though I have only used it as a secondary backup solution w/ history.	13:28
jast	all you need is snapshots	13:29
	→ cakehero joined	13:29
	← eno left	13:29
EugeneKay	The Daily WTF did a decent piece on DB schema changes and updates about a week ago	13:29
prgmctan	jast: ok, just a thought that popped into my head lol	13:29
MacGyverNL	Intended for recovery attempts if all else fails.	13:29
jast	prgmctan: that thought has assaulted a few other people before ;)	13:29
EugeneKay	You can use git to store backups(it is fundamentally a content-addressable data store), but it's not really a good idea.	13:29
prgmctan	EugeneKay: I was talking about the schema	13:29
EugeneKay	prgmctan - look up TDWTF's piece on doing it. git is a good place to store your schema changes(alongside your code, as an upgrade binary or such), but there are many pitfalls to look out for.	13:30
prgmctan	I heard about that, basically using git and overwriting your backup every time	13:30
	→ eno joined	13:30
	← johnkpaul left	13:30
	← krz left	13:31
EugeneKay	The general problem is that you need to handle changes gracefully(no dropping tables!) and making the update script work on "desired state", rather than a process.	13:31
	The first is usually easy-to-middling, the second if difficult.	13:31
	→ sobel joined	13:31
EugeneKay	s/if/is	13:31
	← gavin_huang left	13:32
bLUEEE	EugeneKay, all good - thanks !	13:32
prgmctan	right	13:32
	→ gavin_huang joined	13:32
prgmctan	EugeneKay: you have a link to that article?	13:33
EugeneKay	!googl the daily wtf database upgrade	13:33
	Oh wait, gitinfo doesn't even do that	13:33
jast	I could add it, but then I'd have to think about how to do async HTTP requests	13:33
	I've been trying to avoid that :)	13:33
prgmctan	I googled that, but I wasn't sure which one you were refering to	13:34
EugeneKay	It was dated a week or two ago	13:34
	rogg → Strogg	13:35
prgmctan	does 'Database Changes Done Right' sound about right?	13:35
EugeneKay	Yup	13:35
prgmctan	cool, thanks	13:35
	← eno left	13:36
EugeneKay	jast - how are you doing the main loop? Can you add an outgoing queue somewhere that it'll read through?	13:36
	← xiangfu left	13:36
jast	EugeneKay: I use POE. there's an HTTP client component for it. just never bothered looking properly before.	13:37
	← memoryleak left	13:37
	→ eno joined	13:38
EugeneKay	Apathy is a hell of a drug	13:39
	← philips left	13:39
	→ philips joined	13:39
jast	can't get enough of it :}	13:39
	← philips left	13:39
	→ philips joined	13:39
	→ yoklov joined	13:40
	← empity left	13:41
	→ chrisgeorge joined	13:42
	→ shruggar joined	13:42
	← Targen_ left	13:42
	→ sebrock joined	13:43
	→ OOPMan joined	13:44
	← d3vic3 left	13:44
	← ralf_e left	13:44
	← nenjordi left	13:46
	→ Quanteek joined	13:47
	→ boonrumor joined	13:47
	→ bLUEEEE joined	13:47
	gitinfo set mode: +v	13:47
bLUEEEE	say that a colleague updates the origin and before i start working can i see whats different between my local files and the origin ?	13:48
	← stkrzysiak left	13:48
	← Maccer left	13:48
	→ beatak joined	13:49
	→ vervic joined	13:49
	→ Maccer joined	13:49
sajbar	bLUEEEE: yes, git fetch origin and then you can diff your working copy with the origins/master or what ever your branch is called.	13:50
	and now off work	13:50
	→ lyddonb joined	13:50
	→ dddk1 joined	13:51
	← bLUEEE left	13:51
	← canton7 left	13:51
	→ jceb joined	13:51
	← dddk1 left	13:51
	← gavin_huang left	13:52
	← vervic left	13:52
	Vampire0_ → Vampire0	13:52
	← L42y left	13:53
	← OOPMan left	13:54
	← Orbitrix left	13:54
	→ Evanescence joined	13:54
	→ ology joined	13:55
	→ juanfer joined	13:55
	→ canton7 joined	13:55
	← canton7 left	13:56
	→ ralf_e joined	13:56
	← eno left	13:56
	→ snearch joined	13:57
	← DofLeFou left	13:58
	← cccaldas left	13:58
	→ weiyang joined	13:58
	→ eno joined	13:59
	→ gavin_huang joined	14:00
	→ Orbitrix joined	14:02
	→ alansaul joined	14:02
	← i42n left	14:02
	→ i42n joined	14:02
	← kennethreitz left	14:04
	→ EvanR joined	14:05
	← surfdue left	14:05
	EvanR → Guest54644	14:05
	→ pandeiro joined	14:06
	→ surfdue joined	14:06
	→ L42y joined	14:06
	→ krz joined	14:07
	← betovar left	14:07
	→ iamjarvo joined	14:07
	andres_f → andres	14:09
	← coredumb left	14:09
	→ cccaldas joined	14:12
	← frogonwheels left	14:13
	→ iocor joined	14:13
	→ marcogmonteiro joined	14:13
	→ Malar joined	14:14
	→ wildalaskan joined	14:14
	→ dannyanderson joined	14:16
	→ ehsan joined	14:16
	→ craigntammy joined	14:17
	→ canton7 joined	14:17
	← stodan left	14:17
	← canton7 left	14:18
	← gldn left	14:19
	→ mneorr joined	14:19
	→ tommcdo joined	14:19
	← yoklov left	14:20
	→ davpoind joined	14:20
	→ xsbeats joined	14:21
	← eno left	14:21
	→ yoklov joined	14:21
	← sattu94 left	14:23
	→ kevwil joined	14:23
qwertz_	git remote -a still shows remote branches that are gone. how do i tell git that they're gone? git fetch doesn't do the job apparently	14:23
tommcdo	hey all, i have a question that's a little involved, and i'm not well versed in git terminology, so please bear with me...	14:23
jast	qwertz_: git remote prune	14:23
	tommcdo: you're not the first, don't worry :)	14:23
tommcdo	i want to use different git repos on one project in a nested fashion	14:24
	→ adman65 joined	14:24
	→ twisted` joined	14:24
	← adman65 left	14:24
	→ adman65 joined	14:24
	→ eno joined	14:25
tommcdo	here's how i'm actually doing it. i'm working on a PHP site using a framework (Kohana), and there's a 'modules' directory somewhere inside the codebase. i'm using the modules directory to include code that i use for lots of different sites, and i want to be able to put it into its own repository	14:25
jast	tommcdo: !subproj	14:25
gitinfo	tommcdo: [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.	14:25
jast	hope that helps :)	14:25
	→ enmand joined	14:25
	← Guest54644 left	14:26
jast	if not, just complain loudly	14:26
	→ KaZeR_W joined	14:26
qwertz_	jast, thanks a lot.	14:26
KaZeR_W	hi there	14:26
gitinfo	KaZeR_W: 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.	14:26
	← Evanescence left	14:27
	← enmand_ left	14:27
	← cccaldas left	14:27
KaZeR_W	i'm having issue with the submodule of a project. one one host, everything seems pushed, but on the other, i'm getting "fatal: reference is not a tree"	14:27
tommcdo	jast: looks like one of the latter suggestions might do the trick	14:27
	!submodule	14:27
gitinfo	git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html	14:27
shruggar	lots of submodule-related queries recently. Did someone suddenly decide that submodules were a good idea and tell everyone to start using them?	14:28
jast	hey, I do like submodules :)	14:29
	← bLUEEEE left	14:29
	→ oriba joined	14:29
tommcdo	shruggar: i think this is a really common thing in web dev. i work for an agency making lots of sites, and let's face it, all sites need the same stuff. i want one directory for my common stuff to be kept in its own repository (but i still want that stuff to be in each site's repositorty)	14:29
jast	KaZeR_W: what command gave you that?	14:29
	← thiago left	14:29
	← jceb left	14:30
shruggar	I've just always preferred subtree merge	14:30
	it's not perfect, but it's conceptually better than submodules, imo	14:30
	→ cccaldas joined	14:30
	← krisfrem1n left	14:30
	← yoklov left	14:30
	→ krisfremen joined	14:31
	← neurodrone left	14:31
KaZeR_W	jast, git submodule --init on another host (where the use of the submodule is new)	14:31
	← mfridh left	14:32
tommcdo	welp, the more i read about this the more complicated it looks like it's gonna be.	14:34
	← timoxley left	14:34
jast	tommcdo: try one of the other options. different ones appeal to different people and work better for different situations.	14:34
	KaZeR_W: might just need to run 'git submodule update' there	14:34
	→ mfridh joined	14:36
	→ guampa joined	14:36
	→ arti_t joined	14:36
tommcdo	jast: thanks, i'll look into a few different options	14:37
	← Fullmoon left	14:37
	→ fly9 joined	14:37
	← tatsuya_o left	14:37
	← pidus left	14:37
jast	tommcdo: namely !subtree and !gitslave	14:37
gitinfo	tommcdo: The git subtree merge method is ideal to incorporate a subsidiary git repositories directly in to single git repository with "unified" git history, where you only need to pull changes in from external sources not contribute your own changes back (which if technically possible is at least difficult). See http://progit.org/book/ch6-7.html Type "!subtree_alt" for more options	14:37
	tommcdo: gitslave (http://gitslave.sf.net) is useful to add subsidiary git repositories to a git superproject when you control and develop on the subprojects at more or less the same time as the superproject, and furthermore when you typically want to tag, branch, push, pull, etc. all repositories at the same time.	14:37
	→ tatsuya_o joined	14:37
	→ jceb joined	14:38
	→ achadwick joined	14:38
shruggar	look at jast, wielding gitinfo with multiple strikes in the blink of an eye, like some sort of help ninja	14:38
jast	I'm just that magical :}	14:39
KaZeR_W	jast, then i get fatal: reference is not a tree	14:39
	→ Cromulent joined	14:39
	← freimatz left	14:39
	← iocor left	14:39
jast	KaZeR_W: perhaps you need both rolled into one: git submodule update --init	14:39
	if that doesn't work, something is broken in the submodule config	14:39
	← mandric left	14:39
	→ iocor joined	14:40
KaZeR_W	jast, same thing (fatal: reference is not a tree)	14:40
	→ kukks joined	14:40
KaZeR_W	any tip about how i could solve that ?	14:40
jast	KaZeR_W: make sure the submodule URL in .gitmodules is correct and accessible to you	14:41
KaZeR_W	jast, yep the issue is on the push side. i took the url from .gitmodules, cloned it somewhere else on host #2 and the last commit is missing	14:43
	← arti_t left	14:43
	← eno left	14:43
KaZeR_W	but on host#1, if i push from the super project, or the module, it says i'm up to date, and git log in the submodule does show the commit	14:43
jast	I suspect you don't completely understand how submodules are supposed to work	14:43
	→ arti_t joined	14:43
	← fanaticgeek left	14:44
jast	I really need to write an article about that	14:44
	→ Fhsrap joined	14:44
KaZeR_W	jast, i suspect you might be right :)	14:44
	→ diegoviola joined	14:44
KaZeR_W	the thing is : i committed from the submodule, i added the submodule path to the superproject commit, committed, pushed	14:44
	what did i miss?	14:44
	→ eno joined	14:45
jast	you shouldn't commit from the submodule	14:45
	within the submodule all you should do is look	14:45
	because 'submodule update' detaches HEAD inside the submodule, so new commits you make there don't belong to any branch	14:46
	← Sigma left	14:47
jast	this is the slightly annoying thing about submodules :)	14:47
tommcdo	jast (and shruggar): i think subtree is the way to go for me. gitslave looks a little better, but i don't want to install another program. thanks a lot, guys	14:47
KaZeR_W	mmm indeed. so you shouldn't fix issues from the submodules. that differs a bit from svn	14:48
jast	tommcdo: git-subtree hasn't been added to core git yet, but it will be soonish	14:48
	gchristensen → gr[a]hamc	14:48
	← shibayu36 left	14:48
	→ oehman joined	14:49
shruggar	git-subtree? I was just talking about -s subtree	14:49
jast	oh, right	14:49
shruggar	tell me more	14:49
tommcdo	shruggar: is that any different from the description that gitinfo gave?	14:49
jast	KaZeR_W: the right thing to do: have separate clone of submodule, make changes/commit/push there. then, in the actual submodule, checkout the branch and pull. then do a 'git add path/to/submodule' in the superproject and commit/push (that changes the revision of the submodule).	14:49
shruggar	tommcdo: I do not believe so	14:49
jast	I may have been talking about !subtree_alt	14:49
gitinfo	The git subtree merge method is hard to export changes from.. https://github.com/apenwarr/git-subtree provides another method which appears to be easier to export changes from. Also as a no-change-exported method, see https://metacpan.org/module/git-stitch-repo which claims to generate a unified history instead of merged branches.	14:49
	→ yshh joined	14:50
KaZeR_W	thanks jast, i'll do that way from now.	14:50
	how can i fix my current tree, though?	14:50
	← gavin_huang left	14:50
jast	is it just one commit that's missing?	14:50
tommcdo	shruggar: can you give me a "quick start" guide to get a subtree merge going?	14:51
oehman	If I made a local commit and want to go back to uncommited state to add more changes.. how do I do this?	14:51
shruggar	ah.. I tend to avoid making changes directly in a subtree unless the change is not applicable to upstream	14:51
jast	oehman: add your changes, stage them, git commit --amend	14:51
oehman	thx	14:51
jast	shruggar: makes sense, too	14:51
shruggar	tommcdo: personally, I set things up the way I want them, then git merge -s ours with the upstream repository. After that, -s subtree works via magic	14:51
	→ Sigma joined	14:52
shruggar	"set things up the way I want them" meaning "copy files from the subproject over manually and add them"	14:52
	← cakehero left	14:53
	→ hsttlrjeff joined	14:53
	→ Destos joined	14:53
KaZeR_W	jast, yes, only one	14:53
shruggar	-s ours means "record that a merge took place, assume I am telling the truth and don't do anything merge-y yourself, git"	14:53
	→ mandric joined	14:53
tommcdo	shruggar: you're throwing some words at me that i'm not familiar with. i've so far only been using pull, commit, and push lol	14:54
	→ MANCHUCK joined	14:54
shruggar	actually, come to think of it, I git merge -s ours, then git commit --amend after setting things up how I want them	14:54
	→ fanaticgeek joined	14:54
tommcdo	at which point do you actually specify where the remote repository for the subtree lives?	14:55
shruggar	tommcdo: oh, I'd done that already...	14:55
	← Destos left	14:55
	→ Destos joined	14:55
	← dr0id left	14:55
	→ EvanR joined	14:56
EvanR	bah	14:56
shruggar	tommcdo: 1) git remote add mylibrary path/to/mylibrary/.git 2) git merge -s ours mylibrary/master 3) cp /path/to/mylibrary where/I/want/it 4) git add where/I/want/it 5) git commit --amend	14:56
EvanR	--git-dir=/foo/bar/baz always results in	14:56
shruggar	there's the simplified overview	14:56
	← Sigma left	14:56
EvanR	fatal: Not a git repository: '/foo/bar/baz'	14:56
tommcdo	shruggar: all of these commands would be run from the base directory of the main project (the one that wants to include the subtree), right?	14:58
	→ thiago joined	14:59
shruggar	tommcdo: none of them need to	14:59
	→ aldodelgado joined	14:59
tommcdo	but they would at least be done from somewhere within the main project?	14:59
shruggar	yes	14:59
	→ legnaleurc joined	15:00
shruggar	unless you set GIT_DIR and GIT_WORK_TREE :)	15:00
tommcdo	this leads to another thing i don't have a solid grasp on... the identifiers, or whatever they're called. in this case, what you've called "mylibrary", and what for my projects i always call "origin" (and i'm not sure why)	15:01
	→ neurodrone joined	15:01
	← neurodrone left	15:01
	→ neurodrone joined	15:01
	→ mritz joined	15:01
shruggar	that's called a "remote"	15:01
	→ stkrzysiak joined	15:01
shruggar	it's a local alias for a remote repository, so you don't need to type [email@hidden.address] all the time	15:01
	→ yoklov joined	15:02
	← pdtpatr1ck left	15:02
shruggar	once defined, a remote also keeps track of remote branches (via "remote-tracking branches"), which are useful to have local copies of	15:02
	→ mikepack joined	15:02
tommcdo	okay, so after i've done all of that, and i run pull and push, etc, where/when should i have to specify "mylibrary" (as oppoised to "origin", which i've set as my default remote)	15:03
	← jargon- left	15:03
	→ alx- joined	15:03
	← alansaul left	15:04
	→ ojimac joined	15:04
	← eno left	15:04
shruggar	this workflow doesn't involve pushes to mylibrary, and though I don't like "pull" myself: git pull -s subtree mylibrary master	15:05
tommcdo	shruggar: what don't you like about pull? for my case, we have multiple developers working on the same project, so as far as i can tell, it's necessary	15:06
	← cmyk_ left	15:06
	→ eno joined	15:06
	← arvind_khadri left	15:07
	→ Fullmoon joined	15:07
shruggar	"pull" is chaotic, equivalent to saying "just put some changes into my worktree, I don't care what." (for most workflows)	15:07
tommcdo	what's the alternative?	15:07
	← beautifulmind left	15:08
	→ dpino joined	15:08
ndim	fetch first, then manually merge what you want to merge.	15:08
shruggar	only merging in one direction into the master (or "further-upstream") branch, and in general only performing a merge after you've looked at changes and determined that you want them	15:08
	people who use "pull" usually merge for no good reason, and I dislike that	15:09
tommcdo	oh. in my workflow (web dev), it's basically always vital to get everything. we each have our own copy of the site, and without each of us having everything, some things just won't work	15:09
	→ linusoleander joined	15:10
shruggar	what things won't work?	15:10
tommcdo	well, one guy might be working on a database model class, and without the changes he implements, my code will start doing the wrong things, etc	15:12
shruggar	I'm also in web-dev, with each developer on their own copy. Because of that, we only need to merge when we're done working	15:12
tommcdo	well, define "done working"	15:12
	end of day? end of project?	15:12
	→ coredumb joined	15:12
shruggar	end of topic	15:12
cannonball	In our dev environment, each dev has their own db and a script to update table schemas when a merge contains new model definitions.	15:13
tommcdo	i can't tell if you're defining it as that, or telling me that you don't want to talk about this anymore :P	15:13
cannonball	it's not an auto update though, we have to manually run said script.	15:13
shruggar	defining it as that :)	15:13
tommcdo	then that's basically the same flow. we don't use pull all the time, just when we need to pull :D	15:14
shruggar	if one person is working on a model, and another person is working on the use of that model, one of you (or both working together) defines the interface. One of you works on using the interfacing, the other works on making the interface do something other than spit out test data	15:14
	→ jargon- joined	15:14
	← kevwil left	15:14
	← kermit left	15:15
tommcdo	shruggar: yeah, we could probably use some better standards around here to make such things work more fluidly	15:15
	← dcarper left	15:15
	← eno left	15:15
tommcdo	cannonball: do you have a general script that checks for database structure changes, or do you just write little SQL files that do the work? (if the former, i'd be very interested!)	15:15
	← erickr left	15:16
	→ kevwil joined	15:16
shruggar	tommcdo: http://thedailywtf.com/Articles/Database-Changes-Done-Right.aspx	15:16
	← akiress left	15:17
	→ john_anderson joined	15:17
	← linusoleander left	15:17
tommcdo	thanks, shruggar	15:17
	→ eno joined	15:17
shruggar	tommcdo: that's my way of saying "don't diff your databases /instead/ of writing migrations"	15:18
	write migrations, diff databases only when finding out wtf	15:18
qwertz_	oh wow. merging can fail big time if the same files got heavily modify in both branches.	15:18
cannonball	latter. Each ALTER TABLE or CREATE TABLE or whatever is housed a new .sql file. The script uses yaml to keep track of which .sql files have been processed, looks for new ones, and process them in alphabetical order. the dev which committed the change is supposed to gen new model files as part of their commit, so this should bring my schema inline with the original committer's schema and newly checked out models.	15:18
	← Cromulent left	15:19
	← eightyeight left	15:19
	→ jatt joined	15:19
tommcdo	shruggar: that's good advice. i just wanted to be lazy if i could :P	15:19
	← adman65 left	15:19
	→ eightyeight joined	15:19
	→ airborn joined	15:19
tommcdo	cannonball: i've been thinking if implementing a system like that for a while, just haven't decided on the right structure/toolset	15:19
shruggar	tommcdo: programming is the highest form of laziness	15:19
	→ adman65 joined	15:20
adman65	hey	15:20
shruggar	tommcdo: personally, I steal from rails	15:20
adman65	is there a command I can excute to determine if a remote branch is in sync with a local branch?	15:20
FauxFaux	http://code.google.com/p/dbdeploy/ (or http://www.liquibase.org/ ¬_¬)	15:20
	→ linusoleander joined	15:20
cannonball	tommcdo: It's all perl for us.	15:21
shruggar	adman65: git log @{u}…HEAD if that returns nothing, you're good	15:21
	→ RaceCondishun joined	15:21
	→ reels_ joined	15:21
	← RaceCondition left	15:21
	RaceCondishun → RaceCondition	15:21
	← amirite left	15:21
	← thiago left	15:22
	← jsoa left	15:22
	← Fullmoon left	15:23
tommcdo	cannonball: glorious perl.	15:23
cannonball	adman65: If you do just a 'git fetch', then 'git checkout' will show relationship of current branch to the remote branch.	15:23
	→ thiago joined	15:24
	← reels left	15:24
cannonball	I don't know of an all in one way of doing it.	15:24
	→ pdtpatr1ck joined	15:24
adman65	cannonball: more spefically I am looking for a boolean value that says is my local branch in sync with a remote branch	15:24
	← griotspeak left	15:24
adman65	I think git diff head origin/master --exit-code does the trick	15:24
	← williamcotton left	15:25
	← dpino left	15:25
	→ dreiss joined	15:26
	← mandric left	15:26
ndim	My bash prompt tells me that.	15:26
	← i42n left	15:27
cannonball	ndim: what does PS1 look like to do that?	15:27
	← fanaticgeek left	15:27
tommcdo	cannonball: have a $? in there. you can always echo $? to see the last return status	15:27
	← arti_t left	15:27
ndim	cannonball: It contains a part '$(__git_ps1 " \| git %s")'	15:28
FauxFaux	adman65: git diff HEAD @{u} --exit-code # perhaps.	15:28
	← lyddonb left	15:29
adman65	FauxFaux: what is @{u}	15:29
FauxFaux	adman65: man git rev-pares	15:29
gitinfo	adman65: the git manpage is available at http://jk.gs/git.html	15:29
ndim	cannonball: There is also a bunch of GIT_PS1_* env vars to control the __git_ps1 output	15:29
FauxFaux	adman65: man git rev-parse	15:29
gitinfo	adman65: the git-rev-parse manpage is available at http://jk.gs/git-rev-parse.html	15:29
cannonball	Hmmm, I have $(__git_ps1 " (%s)") By doing 'git BRANCHNAME', it shows the relationship to the remote?	15:29
	→ edwardbc joined	15:29
cannonball	it doesn't for me. Maybe I'm using an older version compared to you?	15:30
ndim	cannonball: GIT_PS1_SHOWUPSTREAM=yes	15:30
cannonball	OH NICE!	15:30
haarg	cannonball: https://github.com/gitster/git/blob/master/contrib/completion/git-completion.bash#L46	15:31
	there are other vars you can set as well	15:31
	← ehsan left	15:31
cannonball	yah, I'm reading the git bash_completion file.	15:31
	Lots of things I didn't know about it.	15:31
	→ ehsan joined	15:31
	→ mister_m joined	15:32
	← prgmctan left	15:32
	→ khmarbaise joined	15:32
haarg	be aware that GIT_PS1_SHOWDIRTYSTATE and GIT_PS1_SHOWUNTRACKEDFILES can cause it to be really slow on large repos	15:32
	→ fanaticgeek joined	15:33
cannonball	Our repo is on an nfs homedir, so yeah, anything like that would be painful. But I set GIT_PS1_SHOWUPSTREAM to auto and it works great. Thanks!	15:34
	← surfdue left	15:34
ndim	"really slow" can mean 30+s for a "cd linux" until the cache is warm, and then another 1 or 2 seconds for every prompt.	15:34
	→ dfr\|mac joined	15:35
	→ surfdue joined	15:35
	→ cmyk_ joined	15:36
	→ lyddonb joined	15:36
	→ canton7 joined	15:36
	→ edwardbc_ joined	15:36
	← cmyk_ left	15:37
	← eno left	15:37
	← edwardbc left	15:37
	← mritz left	15:38
	→ dpino joined	15:38
	→ eno joined	15:39
	→ arti_t joined	15:40
	← matthiaskrgr left	15:40
	→ maletor joined	15:40
	Guest16470 → [mbm]	15:40
	← maletor left	15:40
	← anderse_ left	15:40
necromancer	ugh...git is such a blueballer	15:41
	i'm all like OMG I LOST EVERYTHING but then i figure out where it is	15:41
	→ matthiaskrgr joined	15:41
necromancer	i mean i love it because i don't lose shit but it freaks me out sometimes ;)	15:41
	← jangell left	15:41
shruggar	it's hard to lose things in git without saying "git, please lose my things"	15:41
FauxFaux	Blueballer, as in, prevents you having sex? I've actually found the ~10% productivity increase from using git has left me with 10% more of my day to have sex with my waifu.	15:41
	→ Fullmoon joined	15:42
	shruggar chattr -R +euphamism git	15:43
	← aeontech left	15:45
necromancer	the fact that someone who understands git is also capable of snagging a woman blows my mind	15:45
	;-)	15:45
	and it also gives me hope	15:45
shruggar	if it helps, my wife probably wishes I wouldn't talk about directed acyclic graphs in bed	15:45
	← Quanteek left	15:46
	← JohnBat26 left	15:46
	→ P0TaT0 joined	15:46
	→ bLUEE joined	15:47
	gitinfo set mode: +v	15:47
	← EricInBNE left	15:47
	→ soulcake joined	15:47
bLUEE	hi there.. i am getting ... (master\|REBASE -i)	15:47
gitinfo	bLUEE: 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.	15:47
	← weiyang left	15:47
bLUEE	how do i remove rebase ?/	15:47
	← ajt left	15:47
shruggar	bLUEE: in what? your PS1?	15:48
_ikke_	bLUEE: You create a branch called REBASE	15:48
	→ ajt joined	15:48
_ikke_	bLUEE: what does git branch -a return?	15:48
bLUEE	master, remotes/origin/master	15:48
	← Womkes left	15:48
bLUEE	i guess i could delete the remotes/origin/master ?	15:49
_ikke_	bLUEE: You probably don't want that	15:49
bLUEE	ah, how do i remove it ?	15:50
	→ Quanteek joined	15:50
	→ Targen joined	15:50
shruggar	I still don't know what you want to remove :)	15:50
ndim	bLUEE: You want to abort the rebase?	15:50
	→ Womkes joined	15:50
bLUEE	ndim thats correct	15:50
ndim	Ah.	15:50
	→ BryanRuiz1 joined	15:51
BryanRuiz1	hi, when i use: git clone ssh://git@repo	15:51
gitinfo	BryanRuiz1: 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.	15:51
ndim	bLUEE: man git-rebase, then look for "abort".	15:51
gitinfo	bLUEE: the git-rebase manpage is available at http://jk.gs/git-rebase.html	15:51
BryanRuiz1	how can i get it to use a different ssh key?	15:51
shruggar	jast: can you turn off the auto-hi?	15:51
	← armetiz left	15:52
ndim	BryanRuiz1: Do the normal ssh setup such that "ssh git@repo" does what you want.	15:52
	← ph^ left	15:52
bLUEE	ndim ! thanks	15:52
	→ dcarper joined	15:54
tommcdo	BryanRuiz1: that is, creating a file in ~/.ssh/config that contains lines specifying the identity file (key), etc	15:54
BryanRuiz1	thanks ndim, tommcdo	15:55
	← Chibby left	15:57
	← eno left	15:57
	← spaceonline left	15:57
	← rurufufuss left	15:58
	← jeffisabelle left	15:58
	← [mbm] left	15:58
	→ eno joined	15:59
	→ [mbm] joined	15:59
_ikke_	does gitinfo now respond to hi?	15:59
	[mbm] → Guest55788	15:59
jast	shruggar: I can. why should I? ;)	15:59
	← Guest55788 left	15:59
	← falena left	16:00
	← Quanteek left	16:00
surfdue	How do I clone a repo on the server with the gitolite installation	16:00
	ssh: Could not resolve hostname localhost: Name or service not known?	16:00
jast	I suppose I can make it suppress the response if the message appears to be long enough to contain the start of an actual question	16:00
surfdue	nvm diff ports	16:01
	→ Cromulent joined	16:01
shruggar	jast: yeah, was about to say. Perhaps just trigger on "hi" by itself, but not "hi,"	16:01
	← snearch left	16:01
	→ tingo joined	16:02
	→ Quanteek joined	16:02
	← linusoleander left	16:02
jast	committed, will push out once I've got the other stuff done	16:02
	← Fhsrap left	16:03
	→ troyt joined	16:03
	← dreiss left	16:04
	→ x0166636 joined	16:04
	→ williamcotton joined	16:04
	→ Fhsrap joined	16:05
	→ jangell joined	16:05
	→ mritz joined	16:05
	← tqmz left	16:06
	→ alberto56 joined	16:06
	→ kenperkins joined	16:06
	← canton7 left	16:06
	→ canton7 joined	16:07
	→ ScottE joined	16:07
	→ keystr0k joined	16:08
	← krz left	16:08
	← jutaro left	16:11
bLUEE	when you do "git diff origin/master", in bash i am entered into some editor	16:11
	how do i navigate it and exit it ?	16:11
FauxFaux	It's probably `less`. Arrows, q for quit.	16:11
	→ vervic joined	16:11
bLUEE	:D	16:12
SethRobertson	!vcsh	16:12
gitinfo	https://github.com/RichiH/vcsh - a tool to manage dotfiles using git's fake bare repos to put more than one working directory into $HOME. Nothing to do with csh AFAIK.	16:12
	← papegaaij left	16:13
	← eno left	16:13
	← cbreak-work left	16:13
	← ralf_e left	16:13
	← shruggar left	16:14
	→ tomasm- joined	16:15
	→ jay joined	16:15
	jay → Guest24957	16:15
	→ kermit joined	16:15
	← dcarper left	16:15
	← gitinfo left	16:16
	← rbrooks left	16:16
	→ eno joined	16:17
	→ SegFaultAX\|work joined	16:17
tomasm-	hi, something just happened to a repository where in the process of trying to back it up (zipping the code's folder including .git), it's almost like the versioning log got reverted back to approximately 6 months ago... 'master' branch is 6 months old even though I've made many revisions/commits since then.... it seems that all the files are there, just that they are all of the sudden under 'untracked'. I also noticed for a few minutes th	16:17
	at files were showin as 'modified' and the pending changes were a reversion back to an older version... anyone know how this could have happened?	16:17
	→ Fullmoon_ joined	16:17
jast	what the	16:17
	← sorahn left	16:17
jast	for some reason the bot refuses to connect now	16:18
	seems to be a TCP error, though... not my fault! :}	16:18
Malar	is git clone "local git directory" == copying "local git directory" to "current directory"?	16:19
	← Jay_Levitt left	16:19
jast	Malar: no, it uses hardlinks for internal objects where possible	16:19
	tomasm-: sounds like something is missing from the archive	16:19
	← Fullmoon left	16:20
	← jatt left	16:20
	← Quanteek left	16:20
Malar	jast, does that mean if i delete the current directory clone, the original file deletes too?	16:20
jast	Malar: no. that's not how hardlinks work.	16:21
	← brokenshakles_ left	16:21
	→ shruggar joined	16:21
	→ erickr joined	16:21
	← Tabmow left	16:21
Malar	ok, never used hardlinks :P.	16:21
tomasm-	jast, so what to do?	16:21
jast	basically you can safely use clone locally and save disk space too!	16:21
	→ spaceonline joined	16:21
	← Fullmoon_ left	16:21
Malar	oh	16:22
	← adman65 left	16:22
jast	tomasm-: do you still have the original repository? i.e. are we talking disaster recovery here or just trying to figure out what happened?	16:22
Malar	jast, is it like copy on write?	16:22
jast	Malar: sort of, yeah	16:23
	it works particularly well for git because objects/pack files never change	16:23
	a later 'git gc' or similar may cause it to create a completely new pack file at that point, thus losing the benefit of the hardlinks	16:23
	but until then you save space and there's not really any downside. :)	16:24
tomasm-	jast, I'm trying to fix the problem. this IS the original repository.... I don't understand what happened. it was fine yesterday. all I did was tar the folder up.... and the log was there (in gitk) 5 minutes ago, but after ANOTHER backup (just "tar jcvf ../backup.bz2 FOLDER"), the log has reverted back to 6 months ago... =(	16:24
jast	tomasm-: so it seems to you like creating the tarball somehow messed up the original repository?	16:24
shruggar	tomasm-: check your "history" output for anything unexpected?	16:24
tomasm-	history? how?	16:25
shruggar	type "history"	16:25
jast	this blows. apparently the IPv6 link between the bot and freenode is broken	16:25
tomasm-	oh, shell history...	16:25
	← eno left	16:26
RichiH	SethRobertson: ?	16:26
	← L42y left	16:26
SethRobertson	RichiH: Didn't you like my pimp?	16:26
	→ ns5 joined	16:26
RichiH	i do, but i didn't see the context	16:26
tomasm-	jast, I don't know if it's tar that did something or git-cola or what	16:26
	→ L42y joined	16:27
jast	there, magically fixed	16:27
	→ gitinfo joined	16:27
	ChanServ set mode: +o	16:27
ns5	Hi! I need to use gitk, but when I invoke it, I get "Error in startup script: failed to allocate font due to internal system font engine problem", any idea?	16:27
SethRobertson	RichiH: Oh, the context was over on stackoverflow	16:27
	→ alvaro_o joined	16:27
SethRobertson	RichiH: http://stackoverflow.com/questions/436125/two-git-repositories-in-one-directory	16:27
jast	tomasm-: good to know that it might have been something different... :} my best guess is that the ref for your branch got corrupted by something	16:27
tomasm-	I saw it saying it had 'pending' changes that were actually reversions back to something 6 months ago, so I backed it all up, just in case....	16:27
jast	or reset	16:28
tomasm-	nothing deliberate, i'm sure	16:28
jast	tomasm-: is it just one branch that's affected or several?	16:28
tomasm-	I only have one branch	16:28
	← alvaro_o left	16:28
jast	okay	16:28
	check if the most recent commit shows up in "git log -g"	16:28
	the one that's supposed to be the most recent, that is :)	16:28
RichiH	SethRobertson: ah	16:28
	→ friskd joined	16:28
tomasm-	no, the most recent is october 2011... =(	16:29
jast	okay, everyone, watch the bot crash and burn due to the new experiental code!	16:29
	.info git	16:29
	← gitinfo left	16:29
	← friskd left	16:29
	← flijten left	16:29
jast	tomasm-: you used -g, right? and it may not be the topmost one in that list. see if your previous most recent commit shows up anywhere in that list... (but probably somewhat near the top)	16:29
	→ gitinfo joined	16:29
	ChanServ set mode: +o	16:29
	→ friskd joined	16:29
surfdue	I just setup gitolite setup all the keys, and everything. On my local computer I did a clone to test, works great.. External ip. I am on my other server now and trying to do the same clone, same ip, etc I am getting "fatal: 'panda-main' does not appear to be a git repository"	16:29
	→ eno joined	16:30
	→ spaceonline__ joined	16:30
tomasm-	jast, nope, nothing	16:30
	← spaceonline left	16:30
	← spaceonline__ left	16:31
	← k6b left	16:31
jast	tomasm-: that is... highly unusual	16:31
	→ spaceonline joined	16:31
SethRobertson	tomasm-: It really sounds if you extracted over your .git. You can look in lost-and-found to find the appropriate commits to be your HEAD. Specifically, look in !fixup	16:31
gitinfo	tomasm-: 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. For hints type !fixup_hints in IRC. Remember: if you have pushed already, there are only a few things you can do without !rewriting_public_history (type that for more info)	16:31
ns5	Does anyone has any idea how to fix this? http://pastebin.com/t2KvT0zU	16:31
	→ marciodias joined	16:31
	→ k6b joined	16:31
jast	really the only way I can explain it to myself is that something overwrote all of git's metadata with an ancient version	16:31
	for the sake of completeness, see if 'git fsck' finds any useful dangling objects	16:31
	← vervic left	16:31
SethRobertson	ns5: tk font problem when running gitk? Never seen such a thing.	16:31
	← bLUEE left	16:32
	→ brokenshakles_ joined	16:32
	→ arvind_khadri joined	16:33
	← arvind_khadri left	16:33
	→ arvind_khadri joined	16:33
	← tsdh left	16:33
	→ pidus joined	16:33
	← L42y left	16:34
	← gitinfo left	16:34
SethRobertson	ns5: Looks like you might not have the "Times New Roman" font installed.	16:34
	← delight left	16:35
ns5	SethRobertson: How do you know that it's missing "Times New Roman"?	16:35
	→ gitinfo joined	16:35
	ChanServ set mode: +o	16:35
	← oehman left	16:36
jast	more bot crashage about to happen	16:36
	.info git	16:36
	← gitinfo left	16:36
	← _aeris_ left	16:36
	← jceb left	16:36
jast	what? I just fixed that error :C	16:36
	oh. that's a different one.	16:36
tomasm-	ok well I made a backup the other day and wanted to import it back into my local version, so I can at least fix MOST of the stuff.... is that possible?	16:36
	→ gitinfo joined	16:36
	ChanServ set mode: +o	16:36
	→ jceb joined	16:36
	→ dabd joined	16:37
SethRobertson	ns5: Code inspection. However, you can try deleting ~/.gitk or adding fonts you do have to that. I have "set mainfont {Helvetica 9}\nset textfont {Courier 9}\nset uifont {Helvetica 9 bold}\n" in that file	16:37
	← gitinfo left	16:37
	→ gitinfo joined	16:37
	ChanServ set mode: +o	16:37
jast	.info git	16:37
	← gitinfo left	16:37
jast	see, now that's a different error this time	16:37
SethRobertson	ns5: Ah, it might be missing Helvetica/Courier andor Lucida Grande/Monaco as well.	16:38
	→ gitinfo joined	16:38
	ChanServ set mode: +o	16:38
SethRobertson	.info git!	16:38
	No fun	16:38
	→ _blop joined	16:39
jast	.info git	16:39
	more bugs!	16:39
RichiH	SethRobertson: any idea how to delete the csh blurb from the bot?	16:39
SethRobertson	You should add eliza to gitinfo. We can point users with severe problems to it "gitinfo, why does git hate me?"	16:40
	RichiH: Sure, read !gitinfo	16:40
gitinfo	RichiH: I am an IRC bot which responds to certain keywords to provide helpful(?) information to humans. Please see http://jk.gs/git/bot for more information about how to use me.	16:40
	→ sq-one joined	16:40
	→ empity joined	16:40
	← amigojapan left	16:40
	← blop left	16:40
SethRobertson	However, note that it appears to be undergoing maintenance right now	16:41
	← mneorr left	16:41
	← defts left	16:41
jast	all non-new features should work as normal	16:41
	occasional restarts are not to be avoided, though... I'm extending the core part, and it's a rather messy^Witerative process	16:41
	← gitinfo left	16:41
jast	there it goes again!	16:41
tomasm-	if I have an old repo in /local/repo1, and a newer version in /local/repo2, how do I get the new stuff in repo2 to be added to repo1? I tried git fetch /local/repo2 from within repo1, but that didn't do anything	16:42
	→ gitinfo joined	16:42
	ChanServ set mode: +o	16:42
	→ _aeris_ joined	16:42
jast	tomasm-: git pull /path/to/other yourbranch	16:42
	→ Tabmow joined	16:42
ns5	SethRobertson: I installed dejavu font and now gitk works fine	16:42
jast	.info git	16:42
gitinfo	.info result: Git (software), a distributed version control system <http://duckduckgo.com/Git_(software)> \| Georgia Institute of Technology, an American engineering and technological university located in Atlanta, Georgia, USA <http://duckduckgo.com/Georgia_Institute_of_Technology> \| The Gits, a post punk band <http://duckduckgo.com/The_Gits> \| ...	16:42
jast	finally!	16:42
SethRobertson	tomasm-: Or `git remote add repo2 /local/repo2` and then `git pull repo2`	16:42
jast	that's part one working	16:42
	.info the meaning of life, the universe and everything	16:43
SethRobertson	.info gitslave	16:43
gitinfo	.info result: Answer to Life, the Universe, and Everything: The Answer to Life, the Universe, and Everything has a numeric solution in Douglas Adams ' series The Hitchhiker's Guide to the Galaxy. <http://www.thefreedictionary.com/_/search.aspx?pid=aff18&word=Answer%20to%20Life%2C%20the%20Universe%2C%20and%20Everything> [from The Free Dictionary]	16:43
	.info result: gitslave: Software description: superproject mutli-repository management. <http://sourceforge.net/projects/gitslave/> [from SourceForge]	16:43
jast	and it wasn't even that much work :)	16:43
	now the other one that's more tricky to implement because it involves HTML parsing...	16:43
	.search git	16:43
	well it doesn't crash, at any rate :)	16:43
SethRobertson	.info help	16:43
gitinfo	.info result: Help: Help is any form of assistance. <http://en.wikipedia.org/wiki/Help> [from Wikipedia]	16:43
jast	that's a good start	16:43
	→ linusoleander joined	16:43
	← Malar left	16:43
jast	the .info command will attempt to disambiguate/define/solve any query	16:44
	it can do calculations, dictionary lookups, wikipedia articles, many code hosting sites, ...	16:44
	→ d0k_ joined	16:44
SethRobertson	.clearcase: why does clearcase suck?	16:44
	.info why does clearcase suck	16:44
gitinfo	.info result: nothing found.	16:44
	← d0k left	16:44
jast	.info clearcase	16:44
gitinfo	.info result: IBM Rational ClearCase: The Rational ClearCase family consists of several software tools for supporting software configuration management of source code and other software development assets. <http://en.wikipedia.org/wiki/IBM_Rational_ClearCase> [from Wikipedia]	16:44
	d0k_ → d0k	16:44
SethRobertson	.info 1+1	16:44
gitinfo	.info result: [calc] 1 + 1 = 2	16:44
jast	it's pretty magical if I do say so myself	16:45
	it uses duckduckgo's awesome zero click API	16:45
SethRobertson	.info 00:1d:60:cf:32:13	16:45
gitinfo	.info result: nothing found.	16:45
jast	or was that zeroclick info?	16:45
	→ Araxia joined	16:45
IcePic	its pretty magical it did not know why CC sucks a lot. ;)	16:45
jast	I think the syntax for IP stuff is different	16:45
tomasm-	hmmm, ok git pull doesnt work because it says there are untracked working files to be overwritten... can I bypass that?	16:45
	← KaZeR_W left	16:45
SethRobertson	tomasm-: Review the changes: `git diff` If you are QUITE sure you want to nuke them, `git reset --hard HEAD` if you are less sure `git stash`	16:46
jast	well apparently it doesn't do IP lookups or anything	16:46
	→ heppy joined	16:46
jast	tomasm-: you have a backup of the working tree, right? just git reset --hard to obliterate everything, then	16:46
	→ apok joined	16:46
	→ maletor joined	16:47
	→ L42y joined	16:47
	← angelsl left	16:47
	← fanaticgeek left	16:48
	← x0166636 left	16:49
	→ legumbre joined	16:49
	← philips left	16:50
	← johnathanb left	16:50
	→ cakehero joined	16:51
	← jceb left	16:51
	→ philips joined	16:51
	→ johnkpaul joined	16:52
	→ DavidP joined	16:53
	→ jceb joined	16:53
	← whitman left	16:54
	← pidus left	16:54
	→ dreiss joined	16:55
	← yoklov left	16:55
	← giallu left	16:56
	gitinfo set mode: +v	16:56
DavidP	hello, I try to add 2 submodules to a projects. the 1st submodule is added and cloned correctly, but the second one say : "fatal: You are on a branch yet to be born"	16:56
FauxFaux	Did you try committing the first submodule add? </complete guess>	16:57
DavidP	yes, commit and push	16:57
	← johnkpaul left	16:58
	→ pidus joined	16:59
	→ griotspeak joined	16:59
	→ OOPMan joined	16:59
	← mikewintermute left	16:59
SethRobertson	Does the 2nd submodule have anything in it? Are you sure? Try cloning it by itself	16:59
ns5	From gitk I can see that some of the commits belong to branches that I have removed, and looks like I can never go to those commits. How to remove them?	17:00
	Those commits are like orphan commits	17:00
	no pointer is pointing to them	17:00
DavidP	my console : http://pastebin.com/XS9pyxzH (I'll try the clone of the second module)	17:01
SethRobertson	ns5: What arguments did you give to gitk?	17:01
	← eno left	17:02
	← overbythere left	17:02
ns5	SethRobertson: no arguments, just run "gitk"	17:02
	→ flavius joined	17:02
jast	let's see what my blindly coded libxml stuff does	17:02
	.search git	17:02
	← RaceCondition left	17:03
jast	that's not terribly inspiring	17:03
	← jameswarren left	17:03
	→ eno joined	17:03
jast	!ping	17:03
gitinfo	I only respond to !ECHO-REQUEST	17:03
	→ fischli joined	17:03
jast	hmm. wonder how it managed to not send _anything_	17:03
	→ n3m joined	17:04
	← n3m left	17:04
SethRobertson	ns5: Try selecting "View->New" and then accepting the defaults. It feels like your default view is pulling other stuff in	17:04
	→ n3m joined	17:04
jast	.search git	17:04
	← fischli left	17:04
	→ fanaticgeek joined	17:05
	← psoo left	17:05
SethRobertson	ns5: Oh, that doesn't help. The new view is based on the old one. Perhaps sharing a screenshot would help us help you.	17:06
	← libertyp2ime left	17:08
	→ libertyp1ime joined	17:08
	eijk_ → eijk	17:08
	→ yoklov joined	17:09
	← icdark left	17:09
DavidP	FauxFaux: my second module seems dirty... git status say "(working directory clean)" but I've logs and files...	17:09
	← rchavik left	17:09
	← cubb` left	17:10
	← _iron left	17:10
	← reels_ left	17:10
	→ lonewulf` joined	17:10
	← OOPMan left	17:10
	← mister_m left	17:11
ns5	Say that I use "git reset --hard <commit>" to go back to an old commit, is it possible to undo this?	17:12
	→ pidus_ joined	17:12
	← pidus_ left	17:12
	→ pidus_ joined	17:12
FauxFaux	You can find the old commit in the reflog and reset hard back to it, to undo the damage to your branch, but your working copy changes are /gone/. !gka	17:12
gitinfo	For a better way to view the reflog, try: gka() { gitk --all $(git log -g --format="%h" -50) "$@"; }; gka	17:12
	→ alex__c2022 joined	17:13
	→ duckxx joined	17:13
	← pidus left	17:13
	pidus_ → pidus	17:13
	→ adman65 joined	17:14
	→ shwaiil joined	17:15
shwaiil	hi	17:15
gitinfo	shwaiil: 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.	17:15
	→ luxurymode joined	17:15
ns5	FauxFaux: so commits are never removed? What if I remove a branch that has many commits not merged into master?	17:16
	← dpino left	17:16
	← jceb left	17:16
shwaiil	Q: I've done git remote add origin [email@hidden.address] but there was a typo on yyyyy.git and now I can't push because it doesnt exist. What to do ?	17:16
	Thanks for looking	17:16
	← lonewulf` left	17:16
	→ _iron joined	17:16
	→ lonewulf` joined	17:16
	← soulcake left	17:16
FauxFaux	ns5: It'll be eventually garbage collected (after a couple of weeks).	17:16
shwaiil	ah solved it's git rm	17:17
	thanks	17:17
ns5	FauxFaux: git automatically removes them after a few weeks?	17:17
	← P0TaT0 left	17:18
	← pidus left	17:18
	← lyddonb left	17:19
FauxFaux	Yeah. i.e. it's really hard to lose stuff unless you fail pretty damn hard.	17:19
	→ xrated joined	17:19
xrated	Has anyone used SmartGit?	17:19
	→ MistuhKurtz joined	17:20
	→ eletuchy joined	17:21
FauxFaux	xrated: Yes, people have used SmartGit.	17:21
	← pandeiro left	17:21
	← tomasm- left	17:22
xrated	Well it tells me it can't parse the HEAD file but using git through mingw doesn't give me any such error	17:22
	← eletuchy left	17:22
	→ h0bbit joined	17:23
	← Vampire0 left	17:24
	→ Vampire0_ joined	17:24
	← eno left	17:24
	→ scelis joined	17:25
	→ OOPMan joined	17:25
	→ pandeiro joined	17:25
	← mishok13 left	17:26
	→ eno joined	17:27
	← eno left	17:27
	→ eno joined	17:27
	← LimeBlast left	17:29
	→ cg433n joined	17:30
	→ dcarper joined	17:30
	← Element9_work left	17:31
	→ notola joined	17:31
	← notola left	17:31
	← cg433n left	17:31
	→ ibash joined	17:31
	→ momo__ joined	17:32
	→ ericmuyser joined	17:32
	→ Element9_work joined	17:32
	→ pfFredd joined	17:32
	→ sattu94 joined	17:33
	→ aeontech joined	17:33
	← project2501a left	17:33
	→ pidus joined	17:33
	→ project2501a joined	17:33
	← levifig left	17:33
	← francisfish left	17:33
	gitinfo set mode: +v	17:34
	→ levifig joined	17:34
	← gitinfo left	17:35
	→ gitinfo joined	17:35
	ChanServ set mode: +o	17:35
	← satyanash left	17:36
	← gitinfo left	17:36
	→ JJasonClark joined	17:36
	← babilen left	17:36
	← momo__ left	17:37
	→ gitinfo joined	17:37
	ChanServ set mode: +o	17:37
	← Fandekasp left	17:37
	gitinfo set mode: +v	17:37
	→ netiru joined	17:37
	← hsttlrjeff left	17:37
	← proq left	17:38
	→ empity` joined	17:39
	→ mister_m joined	17:39
JJasonClark	hello all. Anyknow know a way to reference more than 1 global .gitconfig? I would like to setup my global config for generic settings and another one for machine specific config.	17:39
	→ marcuy_ joined	17:40
	milki has no clue	17:40
jast	JJasonClark: an include directive will be added to git's config file format soonish	17:40
	← empity left	17:42
jast	also, AFAIK git will fall back to /etc/gitconfig if the option isn't set in .git/config and ~/.gitconfig	17:42
	$(prefix)/etc/gitconfig to be precise :)	17:42
JJasonClark	jast: so I just need to wait a little longer. Not to bad.	17:42
	← achadwick left	17:42
	← canton7 left	17:42
milki	thats cool	17:42
JJasonClark	jast: hey! that could work. I can just put the machine stuff in the /etc/gitconfig	17:42
	→ jameswarren joined	17:43
	→ canton7 joined	17:43
	→ hgb joined	17:43
JJasonClark	jast: thanks. I'm going to give that a try now	17:43
milki	i thought thats what you meant by global...	17:43
jast	global is git's name for the user's .gitconfig file	17:43
	→ pantsman joined	17:43
milki	hmm	17:43
jast	the global-global config file is called system config file	17:44
milki	o	17:44
jast	see options --global/--system to 'git config' :)	17:44
milki	its "global" in the man page	17:44
	and system	17:44
	ok	17:44
hgb	Hi. I have this repo that I've cloned, and I have a patch (imported by git am) which I want to push, but even though git status says my branch is one ahead of origin/trunk, git push says everything is up-to-date. How come?	17:44
	← Karmaon left	17:44
	→ ashisjoshi joined	17:44
milki	your origin/trunk is not up-to-date	17:44
	← h0bbit left	17:45
jast	hgb: use 'git fetch' to bring origin/* up-to-date	17:45
	← zoran119 left	17:45
milki	origin/trunk does not always = remote's trunk	17:45
	← levifig left	17:46
	→ twitchy joined	17:46
	← dannyanderson left	17:46
	→ axl_ joined	17:46
hgb	jast: git fetch doesn't do anything. Same message after doing that.	17:46
	→ moomerman joined	17:46
	→ cybersphinx joined	17:46
	→ levifig joined	17:46
	← canton7 left	17:47
	→ ralf_e joined	17:47
	← ralf_e left	17:47
	→ lyddonb joined	17:47
	→ canton7 joined	17:47
	← rtuin left	17:47
	→ RaceCondition joined	17:47
	→ shibayu36 joined	17:47
	→ Karmaon joined	17:47
	→ j416 joined	17:47
	← pidus left	17:47
	← dhruvasagar left	17:48
	→ pidus joined	17:48
	← bgerber left	17:49
	→ k6b`_ joined	17:49
	← k6b`_ left	17:49
	→ rheddry joined	17:49
	← aldodelgado left	17:49
jast	hgb: please pastebin the output of: git symbolic-ref HEAD; git rev-parse HEAD; git rev-parse origin/trunk; echo =====; git rev-list -n 5 ...@{u}	17:49
	← rheddry left	17:49
	← levifig left	17:50
	→ dhruvasagar joined	17:50
	→ levifig joined	17:50
	→ Raging_Hog joined	17:50
	← the_cybersphinx left	17:50
	→ scientes joined	17:50
hgb	jast: http://fpaste.org/Wgi0/	17:51
	Guess I could'a actually done that with ; instead of five commands... Oh, well.	17:51
	← pantsman left	17:51
jast	I'll live :)	17:51
iamjarvo	SethRobertson: whats the workflow article you have on your site again?	17:51
jast	and what is the exact line in 'git status' that you were talking about? since it's just one line, feel free to paste here	17:52
SethRobertson	iamjarvo: You mean !best_practices?	17:52
gitinfo	iamjarvo: There is no one right answer for git best practices, but a consensus from #git is available at http://sethrobertson.github.com/GitBestPractices/	17:52
iamjarvo	thanks. found you on githubv after i asked :)	17:52
hgb	# Your branch is ahead of 'origin/trunk' by 2 commits.	17:52
	→ vervic joined	17:52
hgb	jast: Then, git push says up-to-date.	17:52
	→ zoran119 joined	17:53
hgb	(There was one commit, now there are two, since I tried adding a small extra one for testing...)	17:53
	→ LmAt joined	17:53
	← Guest24957 left	17:53
	→ kremiak joined	17:53
	→ Jay_Levitt joined	17:53
	→ eletuchy joined	17:53
	Jay_Levitt → jaylevitt	17:53
	← nikuyoshi left	17:53
	← jameswarren left	17:53
	← ashisjoshi left	17:53
jast	hgb: what about the output of: git ls-remote origin refs/heads/trunk	17:53
	→ diogogmt joined	17:53
	→ jameswarren joined	17:54
ns5	why after "git gc --prune=now --aggressive" I can still git show a commit which belongs to a branch already forcefully removed?	17:54
	← Element9_work left	17:54
hgb	jast: No output	17:54
	→ duckx joined	17:55
jast	hgb: that indicates someone removed/renamed the branch in the remote repository, and push doesn't automatically re-create it	17:55
	hgb: perhaps the upstream name is now 'master' like your local branch? please go check if you can	17:55
hgb	jast: Hmm. That may be.	17:55
	→ bgerber joined	17:55
	hgb tries to find out	17:56
	← cyphase left	17:56
	← duckxx left	17:56
	← LmtdAt left	17:56
	→ barneybook joined	17:56
hgb	jast: Seems like it's (still) called trunk	17:56
	→ Sigma joined	17:57
	← twitchy left	17:57
jast	hgb: that's weird... because when ls-remote looked, it didn't find a 'trunk' branch :}	17:57
	← moomerman left	17:57
	← shwaiil left	17:57
jast	you can do a "git ls-remote origin" to get all refs the remote repo knows about	17:57
	might be a longer list	17:58
	← fr\| left	17:58
hgb	jast: HEAD and refs/head/trunk	17:58
	heads, I mean	17:58
	← diogogmt left	17:58
	← netiru left	17:58
	← vervic left	17:58
	→ cyphase_ joined	17:59
jast	hm. what does it say for refs/heads/trunk?	17:59
hgb	46ffabcdb9cb4e4b05a83db21156db451d250838 refs/heads/trunk	17:59
	lonewulf` → doctorwulf	17:59
	doctorwulf → lonewulf`	17:59
	← Destos left	17:59
jast	okay, so that matches your origin/trunk value	17:59
	→ diogogmt joined	17:59
	→ kaen joined	17:59
	→ maxy joined	17:59
	← senny left	17:59
	← eletuchy left	17:59
	← RaceCondition left	18:00
	← mister_m left	18:00
	hgb finds this stuff weird	18:00
jast	oh, hey, it may have something to do with your local branch name being different from the remote name	18:00
	← shruggar left	18:00
	→ cg433n joined	18:00
	← LmAt left	18:00
jast	let's see how your push config is set up. what does this say: git config push.default	18:00
hgb	Nothing.	18:00
	→ LmAt joined	18:00
jast	right	18:00
hgb	Hmm. Come to think about it... Someone suggested I set that to auto	18:00
	← pidus left	18:00
jast	'auto' isn't a valid value	18:01
	← cg433n left	18:01
jast	but yeah, the default is confusing in your situation	18:01
	you may want to set it to 'upstream'	18:01
hgb	No. rebase.	18:01
	I was supposed to use. I think.	18:01
	← duckx left	18:01
	→ williamcotton_ joined	18:01
hgb	Hmm. Or upstream...	18:01
jast	well, we're trying to change the behaviour of push	18:01
	the default is for it to compare the names of the local and remote branches and update branches if there are pairs (e.g. names that exist on both sides)	18:01
	→ mister_m joined	18:01
jast	in your case that breaks because your local branch name is different from the remote one	18:02
hgb	tracking was the suggestion, actually.	18:02
	→ duckxx joined	18:02
jast	'tracking' is deprecated, the new name is 'upstream' :)	18:02
	← shibayu36 left	18:02
hgb	Ah, ha	18:02
jast	don't ask me why... I thought 'tracking' was a perfectly good name :}	18:02
hgb	Ah!	18:02
	Now we're getting somewhere.	18:02
	→ julius_ joined	18:02
julius_	gi	18:02
	hi	18:03
	← LongBeach left	18:03
	→ johnkpaul joined	18:03
	← williamcotton left	18:03
hgb	Then I suppose the next question is if this patch really _should_ be pushed...	18:03
	williamcotton_ → williamcotton	18:03
	← cortana2 left	18:03
hgb	jast: Many thanks.	18:03
jast	that's not my decision to make :} but you're welcome	18:03
hgb	I'll leave that to me :)	18:03
julius_	git clone git@hostname:gitolite-admin - in what directory does this search for the repo gitolite-admin=	18:04
	?	18:04
canton7	julius_, <home of git user/repositories. You really shouldn't be accessing it directly, though	18:04
julius_	canton7, i have to, to configure gitolite	18:05
	← adman65 left	18:05
canton7	julius_, sorry, I meant you shouldn't be accessing it through something other than ssh	18:05
	→ Fandekasp joined	18:05
	→ infogulch joined	18:06
julius_	git uses ssh in the background	18:06
	← johnkpaul left	18:06
	→ harshpb joined	18:06
	→ subh joined	18:07
	← ns5 left	18:07
canton7	ok, there's a lack of understanding somewhere here. I thought you were asking in which path the gitolite-admin repo was stored. I told you, but warned you not to access it over the filesystem (e.g. 'git clone /home/git/repositories/gitolite-admin')	18:07
	← as3llus left	18:07
	→ Russell^^ joined	18:07
	← apok left	18:07
jast	julius_: !xy	18:08
gitinfo	julius_: I suspect that you are trying to achieve your goal in a way that doesn't work very well with git. Let's take a step back. What is it that you ultimately want to achieve by what you're trying to do now?	18:08
	← xrated left	18:08
	← bitkiller left	18:08
	→ apok_ joined	18:08
	→ Element9_work joined	18:09
jast	hmm, not exactly general enough here :) but understandable enough	18:09
	← yshh left	18:09
	← fanaticgeek left	18:09
	← kevwil left	18:09
	← Element9_work left	18:10
julius_	canton7, no i meant in which directory does the above syntax look	18:10
	→ Element9_work joined	18:10
	← Element9_work left	18:10
	← dr_lepper left	18:10
jast	we've answered that question	18:10
	but it shouldn't matter because gitolite's setup should automatically create it and everything	18:11
	← ajt left	18:11
	→ IWorld joined	18:11
	← IWorld left	18:11
jast	and now... will it crash?	18:11
	no.	18:12
julius_	im looking at a gitolite howto, while installing i was asked to run: git clone git@hostname:gitolite-admin to get the admin repo to my client system for configuration....and i was wondering in what directory THAT command would look	18:12
jast	well we covered that already! given gitolite's defaults, it's in /home/git/repositories	18:12
julius_	jast, no he answered the question where the repo is	18:12
	← yoklov left	18:12
jast	well and that's where it looks, courtesy of gitolite's path resolving magic	18:12
julius_	ah	18:13
canton7	julius_, it doesn't "look" anywhere. It connects to hostname through ssh. gitolite answers the connection. Git asks for the gitolite-admin repo. gitolite fetches it from <home dir>repositories/gitolite-admin	18:13
	→ eletuchy joined	18:13
julius_	so gitolite plays with the paths, that got me confused thx	18:13
	← eletuchy left	18:13
	→ paul424 joined	18:13
Kog\|Work	whoa, gitlolite has a picture-guided install doc? I guess they weren't kidding when they claimed to have a lot of documentation	18:14
	→ duckx joined	18:14
jast	yeah, the gitolite maintainer gets a lot of stupid questions and he's been trying to fight it by making the documentation really good :)	18:14
	→ tsdh joined	18:15
Kog\|Work	jast: fair enough, I salute proper docs	18:15
	← duckx left	18:15
jast	well I'm stumped here. for some reason my .info requests work but my .search requests don't	18:15
	it's the same code, supposedly	18:15
	[to the uninitiated: this is not strictly git-related]	18:15
	→ duckx joined	18:16
Kog\|Work	jast: honestly I try doing awesome documentation... then people just ignore the docs and ask stupid questions anyway	18:16
	never found it to help	18:16
	← duckxx left	18:16
julius_	but how does gitolite step in when i do "git clone user@host:repo" that would look into /home/user/repo without any additional magic?	18:16
jast	yeah, I'm not sure it works all that well for sitaram either :)	18:16
	julius_: the magic is in ~git/.ssh/authorized_keys	18:17
	→ oehman joined	18:17
	← Sigma left	18:17
julius_	Kog\|Work, they just dont love the software as you do	18:17
	→ berserkr joined	18:18
julius_	jast ah thx	18:18
	→ Chillance joined	18:18
jast	oh. I'm absolutely stupid.	18:18
canton7	it's got to the point, though, that he's got a doc for pretty much every scenario ever. So dealing with the stupid questions is a case of locating the right doc	18:18
jast	code nested incorrectly. DUH.	18:18
scientes	go i git svn fetched	18:18
	but where is it?	18:18
	and how do i checkout it	18:19
jast	we return to our regularly scheduled program of frequent bot crashes	18:19
	scientes: you may be looking for 'git svn rebase'	18:19
scientes	doesn't work cause there are no commits	18:19
canton7	scientes, git branch -a. it's probably in remotes/trunk	18:19
jast	oh, right	18:19
	→ [mbm] joined	18:20
jast	that's why it's easier to use 'git svn clone' for creating a new git-svn repo	18:20
scientes	i did	18:20
	and then it stalled up	18:20
	→ babilen joined	18:20
jast	ah, I see	18:20
scientes	so i did git svn fetch after that	18:20
	and it continued where it left off	18:20
jast	is that with --stdlayout?	18:20
scientes	with no options	18:20
	← pfFredd left	18:20
jast	so just one branch, right?	18:20
	← j416 left	18:20
	scientes wishes the enlightment git mirror worked	18:21
	→ rtuin joined	18:21
	→ j416 joined	18:21
scientes	i think so, but im not git on svn	18:21
	→ pidus joined	18:21
	→ oposomme joined	18:21
scientes	git svn clone http://svn.enlightenment.org/svn/e/python	18:21
	everything went into .git	18:21
jast	that doesn't even seem to exist	18:21
scientes	and git branch -a shows nothing	18:21
	→ defts joined	18:21
Kog\|Work	julius_: yes, but that's just life: people have varying degrees of awareness and passion	18:22
[mbm]	what's the best way to clean up the mess caused by git submodules? trying to set up a mirror but several nested git submodules each with the old url hardcoded in .gitmodules	18:22
Kog\|Work	jast: it does if you put a hyphen between git and svn	18:22
scientes	http://svn.enlightenment.org/svn/e/	18:22
	ahh	18:22
	i've gotta put trunk in there	18:22
jast	Kog\|Work: I mean the URL	18:22
	_aeris_ → _aeris_\|away	18:22
jast	Kog\|Work: I happen to know how git works ;P	18:22
Kog\|Work	jast: oh, sorry	18:22
	← tommyvyo left	18:22
	→ xrated joined	18:22
	→ Chaser joined	18:22
	← dabd left	18:22
Kog\|Work	jast: you know, everyone says that, but in practice I find that no one seems to know exactly how it works	18:22
jast	I won't claim that I know all the details... but I've contributed a couple of patches and stuff	18:23
scientes	it confused me cause it checked out files	18:23
Kog\|Work	jast: actually, if you've got the deep voodoo I've got a question you might be able to help with	18:23
	← heppy left	18:23
tommcdo	there's only one man who knows how git works :P	18:23
	← rtuin left	18:23
Kog\|Work	tommcdo: jesus?	18:23
jast	Kog\|Work: sure, let's give that a try	18:23
	→ bitkiller joined	18:23
	→ envygeeks joined	18:23
jast	I don't think any one person knows all the details in git	18:23
tommcdo	i'd say it's a bit excessive to call Linus that, but yeah	18:23
Kog\|Work	jast: so one of my co-workers decided to check in a file with an unprintable unicode character in the name	18:24
jast	Linus hasn't been the maintainer of git in _years_	18:24
Kog\|Work	tommcdo: I'm not sure he knows exactly how it works	18:24
tommcdo	oh, fair enough. didn't know that	18:24
	→ tommyvyo joined	18:24
scientes	anything but NULL and / eh? Kog\|Work	18:24
paul424	can you explain int the simplest words what is the diffrence between add -p and commit ? That is between adding a patch and commiting a file ?	18:24
	whats the practical diffrence is ? cause the changes are saved there anyway ?	18:24
jast	paul424: that's not actually what add -p does. more explanation follows. :)	18:25
scientes	i've never used add -p, what does it do?	18:25
cmn	add puts files in the index (staging area)	18:25
scientes	yeah i know that	18:25
cmn	commit creates a new commit with the contents of the index	18:25
	→ Nik05 joined	18:25
	→ kevwil joined	18:25
jast	paul424: 'git add' is to tell 'git commit' what files/changes to include in the commit. if you do 'git commit' without ever adding anything, no commit is created because it wouldn't include any changes	18:25
cmn	scientes: that's not for you	18:25
jast	paul424: ('git commit -a' automatically git adds all changed files)	18:25
scientes	sry, i was wonder what the -p does	18:25
	scientes goes and grep man	18:26
jast	paul424: git add -p, on the other hand, let's you add individual changes, even just parts of files	18:26
cmn	man git add --interactive	18:26
gitinfo	the git-add manpage is available at http://jk.gs/git-add.html	18:26
	← jameswarren left	18:26
	→ pidus_ joined	18:28
	← pidus_ left	18:28
	→ pidus_ joined	18:28
	→ krz joined	18:29
	→ heppy joined	18:29
	← pidus left	18:30
	← cyphase_ left	18:30
paul424	yeah that was the wrong question .... So whats' the practical diffrence between add -p ( patches ) and add <file > (whole files ) ? Both changes are "saved" and the history is written as well.	18:30
	← ipalaus left	18:30
Kog\|Work	jast: sorry, co-worker, brb	18:30
	← oehman left	18:30
	→ williamcotton_ joined	18:31
	← williamcotton left	18:31
	williamcotton_ → williamcotton	18:31
	→ cyphase joined	18:32
paul424	ahh ok I get it ... never mind.	18:32
	← pidus_ left	18:32
	← Nik05 left	18:33
	→ pidus joined	18:34
	← pidus left	18:34
	→ tqmz joined	18:34
	→ Ksugzulut joined	18:35
	→ shruggar joined	18:35
	→ yoklov joined	18:35
	← marcogmonteiro left	18:35
	→ btanaka joined	18:37
scientes	!rm	18:39
gitinfo	[!karma] Karma is nominally tracked by http://carmivore.com/. If you wanted to thank me, for instance, you would say gitinfo++. My karma is on display via http://carmivore.com/gitinfo	18:39
	→ Fullmoon joined	18:39
scientes	nope, how do i remove files?	18:39
	← Fullmoon left	18:39
scientes	with git add	18:39
milki	git add adds files	18:40
	git rm removes fiels	18:40
scientes	-A	18:40
	thar we go	18:40
paul424	Now I have serious question : why the same cmd 'add' is used to add file to the project "tracked area ' and to the 'staged area' from it ?	18:40
kevlarman	paul424: there is no "tracked area"	18:41
SethRobertson	Do you mean working directory by "tracked area"?	18:41
	← _blop left	18:41
paul424	SethRobertson: yeah	18:41
	→ mneorr joined	18:41
SethRobertson	It is added FROM the working directory TO the 'index'	18:41
Vinnie_win	sigh how do I remove stuff from the index that I don't want commited	18:42
	specifically, an entire directory (recursively)	18:42
paul424	.w index	18:42
	← kaen left	18:42
Vinnie_win	say what	18:43
	→ nicxvan joined	18:43
	→ nikuyoshi joined	18:43
	→ alexim joined	18:43
SethRobertson	Vinnie_win: `git status` tells you how	18:43
paul424	SethRobertson: ok so why the same command ?	18:43
Vinnie_win	git reset	18:43
	→ sami joined	18:44
SethRobertson	paul424: Why the same command as what?	18:44
Vinnie_win	ahhhh	18:44
sami	Could i ask gitweb questions here?	18:44
paul424	SethRobertson: why the same cmd 'add' is used to add file to the project "tracked area ' and to the 'staged area' from it ?	18:44
	→ spiette joined	18:44
	← tatsuya_o left	18:44
	← heppy left	18:44
SethRobertson	It isn't. Can you demonstrate the two different usages that you are referring to?	18:45
	→ joshsmith joined	18:45
	← mneorr left	18:45
Vinnie_win	the command to unstage was "git reset HEAD <dir>"	18:46
paul424	SethRobertson: ohh so the "new file" automaticly goes into staged area/	18:46
SethRobertson	paul424: And please start using the preferred git terminology instead of the works you make up. "Working directory" and "index"	18:46
jast	paul424: the area is the same. in both cases, the first time you stage something the file starts being tracked.	18:46
paul424	I see now.	18:46
jast	'tracked' == 'is it in the index'?	18:46
	(basically)	18:46
	← shruggar left	18:47
Vinnie_win	I thought tracked meant "will have history"	18:47
	→ shruggar joined	18:47
Vinnie_win	i.e. "remote tracking branch"	18:47
paul424	nothing added to commit but untracked files present (use "git add" to track)	18:48
	← nikuyoshi left	18:48
SethRobertson	Vinnie_win: No. I recommend reading !book	18:48
gitinfo	Vinnie_win: There are several good books available about git; 'Pro Git' is probably the best: http://progit.org/book/ but also look at !bottomup !cs !designers !gitt !vcbe and !parable	18:48
Vinnie_win	!gitt	18:48
gitinfo	A book designed to teach people about Git in a real world usage model. The book follows a fictional company as they implement and learn about Git. Covers all 21 standard Git commands. Available for free at http://cbx33.github.com/gitt/	18:48
	← libertyp1ime left	18:49
SethRobertson	I'd fire any developers who were like the ones in this fictional company that took months to learn how to use version control	18:49
sami	Well i'll try anyways. I've just pushed new code in to my git repo and after starting the gitweb it's not showing up anymore. (after the push)	18:50
paul424	Vinnie_win: is there hentai version for it :D	18:50
	→ kerrick joined	18:50
Vinnie_win	How is that funny	18:51
	I'm not asian	18:51
paul424	never mind than	18:51
Vinnie_win	funny you should ask...is there a channel for git discussions in Chinese?	18:51
	(serious question)	18:52
	← shruggar left	18:52
	→ ph^ joined	18:52
	→ ajt joined	18:53
paul424	Is there some dictionary for git terms ?	18:53
SethRobertson	paul424: man gitglossary	18:53
gitinfo	paul424: the gitglossary manpage is available at http://jk.gs/gitglossary.html	18:53
	← ajt left	18:53
paul424	ohh greart	18:53
CareBear\	Vinnie_win : don't know of a cn channel, sorry	18:54
	→ shruggar joined	18:54
CareBear\	Vinnie_win : but feel free to start one!	18:54
Vinnie_win	CareBear: How would I go about trying to find out?	18:54
	I've tried search engines but they don't turn anything ujp	18:55
CareBear\	besides asking here, I guess you can also ask on the mailing list	18:55
Vinnie_win	Freenode mailing list?	18:55
CareBear\	or simply join #git-cn	18:55
	← FernandoBasso left	18:55
Vinnie_win	its empty	18:55
CareBear\	I was thinking the git mailing list	18:55
	nod, so you get to start it	18:55
Vinnie_win	That's beyond useless	18:55
	← erickr left	18:55
	→ rcu joined	18:55
tommcdo	alright, i was seeking some help earlier on subtrees and the like, and i ran into an even more complex problem. i wrote out a scenario that i'll paste here. it's kind of long and involved, but i want to be as clear as i can...	18:55
Vinnie_win	First of all I don't speak or write Chinese, and second of all if the channel is empty, how is one to get help?	18:56
	CareBear: but thanks for caring :-)	18:56
	→ adamm joined	18:56
tommcdo	A repository at git://example.com/mod containing a modules directory already exists. It consists of code that is to be shared across all projects. Alice creates a new project and configures a repository for it at git://example.com/proj. She then adds the module to the project by adding a remote to git://example.com/mod and adding it in with read-tree, and putting it under the subdirectory lib. Now there is a directory in lib/modules that	18:56
CareBear\	Vinnie_win : you asked if there was one - I assumed you wanted to use or contribute :)	18:56
tommcdo	(my apologies for the wall of text)	18:56
Vinnie_win	CareBear: I'm mentoring a Chinese programming and he's having the dickens of a time with Git	18:57
CareBear\	Vinnie_win : starting the channel means creating a new forum for chinese speaking users	18:57
Vinnie_win	Yeah I guess I could do that if there was no alternative but if there is no one there now, its doubtful it would get any sort of critical mass. I've started channels before and they are...well, slow to pick up	18:57
SethRobertson	tommcdo: A truncated wall at that. We stopped after "a directory in lib/modules"	18:57
CareBear\	Vinnie_win : I see. if there is a language barrier I guess books are better than live help, because they can be studied at a comfortable pace	18:57
	← tingo left	18:58
	← xrated left	18:58
tommcdo	damn, that's less than half-way through haha. i'll pastebin it	18:58
	→ mrsrikanth joined	18:59
tommcdo	http://pastebin.com/AZwyWj8v	18:59
	i bet it's not every day someone pastebins a question, eh? (much thanks to anyone who actually chooses to read this!)	19:00
	← levifig left	19:00
	← empity` left	19:00
CareBear\	tommcdo : see about !submodules	19:00
gitinfo	[!submodule] git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html	19:00
	← guns left	19:01
	← shruggar left	19:01
	_aeris_\|away → _aeris_	19:01
	→ blop joined	19:02
	→ levifig joined	19:02
	→ xrated joined	19:03
	→ pdtpatr1ck_ joined	19:04
	→ mandric joined	19:04
	← krz left	19:04
	→ jeppy joined	19:05
tommcdo	CareBear\: so, when someone clones the superproject, they'll get all the submodules, and git will remember that they are submodules?	19:05
	← n3m left	19:06
SethRobertson	Yes. Submodules are annoying to use because of that.	19:06
	→ n3m joined	19:06
tommcdo	SethRobertson: care to elaborate?	19:06
SethRobertson	!submodules_change	19:07
	!submodule_change	19:07
gitinfo	In order to change a submodule you must go into the submodule repository, check it out to the appropriate branch, make the needed change (possibly involving git pull), commit the change, cd .. (out of the submodule), git commit -m "Updated submodule" submodulepath	19:07
paul424	huh C is very small language but for modern system programer its just a fraction percent with other requested skills ( make , cmake , git , emacs , compiler options etc :D )	19:07
	→ Destos joined	19:07
	← jeppy left	19:07
	← pdtpatr1ck left	19:07
	pdtpatr1ck_ → pdtpatr1ck	19:07
	→ MANCHUCK_ joined	19:07
	← Fandekasp left	19:07
SethRobertson	tommcdo: There are other options, see !subprojects	19:07
gitinfo	tommcdo: So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.	19:07
CareBear\	and submodules need to be downloaded from their corresponding servers	19:07
	← gitinfo left	19:07
	→ giallu joined	19:07
jast	yay crash!	19:07
	→ gitinfo joined	19:08
	ChanServ set mode: +o	19:08
CareBear\	paul424 : neither cmake nor emacs nor git are really neccessary for programming	19:08
tommcdo	SethRobertson, CareBear\: I was looking into those other options earlier, and subtree seemed to be the most appropriate, but now that i'm trying to implement it, it's not working out as i'd like	19:09
paul424	CareBear\: neither the C language >_< .	19:09
	← kevwil left	19:09
	← MANCHUCK left	19:10
	← teotwaki left	19:11
SethRobertson	tommcdo: Sorry to hear that. I don't have time to look into it right now, perhaps someone else can help you. You can also look into !subtree_alt	19:11
gitinfo	tommcdo: The git subtree merge method is hard to export changes from.. https://github.com/apenwarr/git-subtree provides another method which appears to be easier to export changes from. Also as a no-change-exported method, see https://metacpan.org/module/git-stitch-repo which claims to generate a unified history instead of merged branches.	19:11
	→ phishy joined	19:12
Kog\|Work	jast: sorry about that... speaking of co-workers that can't figure things out heh	19:14
	← gitinfo left	19:14
jast	web search implemented into bot! \o/	19:14
Kog\|Work	jast: so anyway, someone committed+pushed a file with an unprintable unicode char in the name	19:14
	← envygeeks left	19:14
	→ gitinfo joined	19:14
	ChanServ set mode: +o	19:14
	→ libertyp1ime joined	19:14
Kog\|Work	jast: and then people who were behind remote/origin couldn't catch up without doing a hard reset to HEAD	19:14
jast	.search clue	19:15
gitinfo	jast: Got a response here!	19:15
	Hasbro Games - Clue <http://jk.gs/tmp/2> \| Clue \| Board Game \| BoardGameGeek <http://jk.gs/tmp/3> \| Clue (film) - Wikipedia, the free encyclopedia <http://jk.gs/tmp/4> \| ...	19:15
SethRobertson	.search jast porn	19:15
Kog\|Work	jast: the error was totally opaque too	19:15
gitinfo	SethRobertson: Got a response here!	19:15
	<http://jk.gs/tmp/5>	19:15
jast	oh, I forgot a debug message	19:15
Kog\|Work	jast: so the real problem is that people here don't know how to use git, so repairing the history becomes really hard to do	19:15
jast	Kog\|Work: can't say much without knowing the opaque error	19:16
Kog\|Work	jast: I'm not really quite sure what to do about it other than remove both nodes: the original commit, the rename, from remotely tracked history	19:16
	jast: and then telling people to do another pull/fetch	19:16
SethRobertson	We have fairly instructions on recovering from most user errors. Repository corruption, if that is your problem, is hard for anyone to recover from	19:16
Kog\|Work	jast: but the problem being that people keep recommitting the history, so it's hard to reparent things	19:16
jast	corruption has nothing to do with it	19:16
	← oposomme left	19:16
Kog\|Work	yeah, this isn't corruption, this is just stupidity	19:16
	jast: so in the mean time I told them to just reset hard to HEAD, which is fine... it's a single branch	19:17
	it's where we keep some DB scripts	19:17
SethRobertson	Kog\|Work: then after you have fixed the bad history, set the forbid-force-override option	19:17
	→ the_cybersphinx joined	19:17
	→ dannyanderson joined	19:17
Kog\|Work	SethRobertson: hahah, it's funny you should mention that	19:17
	← warthog9 left	19:17
Kog\|Work	SethRobertson: but no, this wasn't the result of pushing via force	19:17
	← Danielpk left	19:18
SethRobertson	Then write a receive hook to forbid that bad history	19:18
	→ whitman joined	19:18
Kog\|Work	the bad history is some idjit slapped a bogus character in a file name	19:18
	I'm not even entirely sure how he did it	19:18
	but yes, we could add a hook to run a regex across the name and look for such things	19:19
	that's rather ex post facto though	19:19
jast	so just renaming the file and comitting and pushing that doesn't work because...?	19:19
Kog\|Work	well, so if you're before that bit of history you can't catch up	19:19
	I guess we could try and amend the commit	19:19
jast	why not?	19:19
SethRobertson	You can either rename the file in history (rewriting history, which is bad) or you can rename the file now. In both cases it should not come back without some user doing something specifically	19:19
Kog\|Work	let me see if I can find the error message, but the short answer is I'm not sure	19:19
SethRobertson	Kog\|Work: See !fixup for instructions	19:20
gitinfo	Kog\|Work: 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. For hints type !fixup_hints in IRC. Remember: if you have pushed already, there are only a few things you can do without !rewriting_public_history (type that for more info)	19:20
Kog\|Work	SethRobertson: we did rename the file, and it works fine on a clone or a hard reset to HEAD	19:20
	but not if you're catching up from a state where you're behind HEAD	19:20
	we could probably go back and amend that commit though	19:20
SethRobertson	Kog\|Work: It should rename that file as part of the pull normally	19:20
jast	.search SethRobertson porn	19:21
	← cybersphinx left	19:21
jast	hmm, it doesn't like you	19:21
	or I broke something	19:21
SethRobertson	I'm just that clean cut	19:21
jast	.search git	19:21
gitinfo	Git - Fast Version Control System <http://jk.gs/tmp/6> \| Git - Fast Version Control System <http://jk.gs/tmp/7> \| Git (software) - Wikipedia, the free encyclopedia <http://jk.gs/tmp/8> \| ...	19:21
	← Nibato left	19:21
jast	I suspect it's related to safe search	19:21
	← qizwiz left	19:22
Kog\|Work	bah, I can't find the error message and I've already reset my repo	19:22
	lame	19:22
	→ dp joined	19:24
	→ notola joined	19:24
	← notola left	19:24
dp	if a git pull gets interrupted, does it start the transfer over from the beginning next time, or does it continue where it left off?	19:24
SethRobertson	restart	19:24
cmn	it needs to start again	19:24
dp	suck	19:24
SethRobertson	!resume	19:24
gitinfo	Git does not yet support resuming, torrenting, or parallel fetching of clones or other network traffic. Some people have successfully created bundles and then used torrents/resuming http download/etc to accomplish what they needed (possibly creating a reliable on-demand VM somewhere to do the initial download, bundle creation, and file hosting)	19:24
	→ alsasounda joined	19:25
dp	is there any potential plan for that stuff?	19:25
	← alsasounda left	19:25
	→ j416_ joined	19:25
SethRobertson	Potential plans abound. Don't hold your breath	19:25
jast	plan, sure... but it's very complicated to implement that	19:25
	VERY.	19:25
	.search SethRobertson porn	19:25
dp	understood	19:25
gitinfo	.search: nothing found.	19:25
	→ jab416171\|Cloud joined	19:26
	← flavius left	19:26
milki	does pull restart only because it pulls in a single packed bundle?	19:26
jast	no idea why it doesn't find the results for that, but oh well	19:26
Vinnie_win	okay I have a working copy changes with a memory leak, and I'm on the "master" branch. Can I make a new branch, check it out, and then commit these working copy changes to it? Or will my checkout revert me to master HEAD?	19:27
	← dp left	19:27
SethRobertson	!float	19:28
gitinfo	If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there. If the files you changed differ between branches, the checkout will fail. In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.	19:28
kevlarman	Vinnie_win: git checkout -b newbranch;git commit	19:28
	→ Nibato joined	19:28
	← j416 left	19:28
jast	.search multiple words	19:28
gitinfo	Multiple Meaning Words <http://jk.gs/tmp/9> \| PATH%20WORDS <http://jk.gs/tmp/a> \| E-Mail Newsletters: Increasing Usability (Jakob Nielsen's Alertbox) <http://jk.gs/tmp/b> \| ...	19:28
Vinnie_win	okay the implication here is that none of these commands will cause me to lose data	19:29
jast	hmm. ah well. I'll just ignore that.	19:29
SethRobertson	How long does your URL shortener results last?	19:29
	← rcu left	19:29
kevlarman	jast: i know the rules say that because SethRobertson exists, there must be porn of him, but that's taking it a bit far :P	19:30
SethRobertson	And "t" is shorter than "tmp"	19:30
	→ Sigma joined	19:30
jast	ah, found it	19:30
	SethRobertson: until the plugin/bot is restarted	19:30
	i.e. usually long enough	19:30
Vinnie_win	SethRobertson: Hey thanks that was really handy! Apparently I stumbled into a common use-case	19:31
	← alx- left	19:31
jast	.search SethRobertson porn	19:32
gitinfo	Seth Robertson \| LinkedIn <http://jk.gs/t/c> \| Seth Roberts Photography - Richmond Wedding Photography, Richmond ... <http://jk.gs/t/d> \| Beth Robertson, Real Estate Broker Sonoma County - Testimonials <http://jk.gs/t/e> \| ...	19:32
jast	there we go!	19:32
	it works after all	19:32
	(but the 'porn' keyword is ignored due to safe search)	19:32
Vinnie_win	I thought I had a black belt in goofing off but apparently I have much to learn	19:33
jast	this is productive, actually	19:33
	I'm testing the new functions of the bot	19:33
	well, semi-productive :}	19:33
tango_	.search giuseppe bilotta	19:33
gitinfo	Guiseppe Bilotta \| Facebook <http://jk.gs/t/f> \| Facebook - Giuseppe Bilotta <http://jk.gs/t/g> \| Giuseppe Bilotta \| Facebook <http://jk.gs/t/h> \| ...	19:33
tango_	not even remotely	19:33
	← felideon left	19:33
jast	well that's useful, isn't it :)	19:34
	→ rcp-ca joined	19:34
jast	thank facebook for indexing everything a billion times	19:34
	it's not like my name is likely to fare any better...	19:34
	.search jan krüger	19:34
gitinfo	Welcome To Hearthside Quilters Nook - Quilt Lectures ... <http://jk.gs/t/k> \| Welcome To Hearthside Quilters Nook - Quilt Lectures Quilting ... <http://jk.gs/t/m> \| RA: Jan Krueger <http://jk.gs/t/n> \| ...	19:34
	← alexim left	19:35
	← denysonique__ left	19:35
jast	I wonder what this does...	19:35
	.info git-scm.com	19:35
gitinfo	.info: nothing found.	19:35
jast	ah, okay...	19:35
	.info weather in aachen, germany	19:35
gitinfo	.info: nothing found.	19:35
Vinnie_win	.search Vinnie Falco	19:36
gitinfo	Facebook - Vinnie Falco <http://jk.gs/t/o> \| Interview: Vinnie Falco, Creator of BearShare <http://jk.gs/t/p> \| Vinnie Falco - Pipl Profile <http://jk.gs/t/q> \| ...	19:36
jast	it doesn't like the 'in', apparently	19:36
	.info whois git-scm.com	19:36
gitinfo	.info: nothing found.	19:36
jast	aww	19:37
	it claims to support that	19:37
	but apparently not via ZCI	19:37
	← davpoind left	19:37
	← arti_t left	19:37
jast	.info U+1AF49	19:38
gitinfo	.info: nothing found.	19:38
	→ nikuyoshi joined	19:39
	→ kaitocracy joined	19:39
jast	the reference doesn't say which ones are available via the API and which ones aren't	19:39
kaitocracy	help I had a repo with two branches: 'master' and 'original'; I pushed it to Github, but I forgot to push my 'original' branch to github; at some point I deleted the repo off my local disk, when I pulled it again from Github my 'original' branch was gone; how can I recover as much of the 'original' branch as possible?	19:40
jast	kaitocracy: well if you deleted your local repo completely, where do you expect traces of the branch to be left?	19:40
	← edwardbc_ left	19:40
jast	I mean, the branch didn't get pushed after all	19:41
Vinnie_win	check the recycle bin	19:41
kaitocracy	jast: but see if I look on here https://github.com/Kiwilight/System-Pkgbuilds/network, the other branch is clearly visible	19:41
	gitinfo set mode: +v	19:41
	→ rbrooks joined	19:42
rcp-ca	I'm building git-1.7.9.4 on a wayback machine - Fedora Core 4 i386 - and have a test failure: "not ok - 31 ref name 'heads/foo' is invalid" in t1402-check-ref-format.sh. Anyone got pointers before I did into the code?	19:42
	← Vile left	19:42
	→ tatsuya_o joined	19:42
	← sebrock left	19:43
jast	kaitocracy: that's not what you're seeing there. you're seeing everything that you merged from original to master, i.e. only stuff that's part of master.	19:43
	← nikuyoshi left	19:43
kaitocracy	jast: yeah that's fine, I mean I don't expect ot recover the whole branch, what's shown in that graph is good enough; but I want it as a separate branch	19:43
jast	kaitocracy: just create original as a copy of master. that's all you can do, really.	19:44
	← tatsuya_o left	19:44
	→ DrNick2 joined	19:44
	→ tatsuya_o joined	19:44
	← DrNick left	19:45
	→ mjt joined	19:45
	← khmarbaise left	19:45
	← linusoleander left	19:46
	→ khmarbaise joined	19:46
	← lindenle_ left	19:47
	→ alexim joined	19:47
	j416_ → j416	19:47
	← Sigma left	19:49
	→ spiette_ joined	19:50
	← spiette_ left	19:50
	← gnufied left	19:50
	← enmand left	19:52
	← boonrumor left	19:52
	→ pantsman joined	19:53
	← pantsman left	19:53
	→ pantsman joined	19:53
mjt	hello. I was in a middle of a rebase -i but forgot about it, and went to a entirely different place for some bisection. Now i - seemingly - returned "back" and did a 'git rebase --continue', but the result was that i went to a very old version of my patcheset.	19:53
	→ ipalaus joined	19:53
jast	mjt: you're now – to use a technical term – screwed. best to use 'git rebase --abort' if you can, or otherwise reset to the previous state of things	19:54
	← harshpb left	19:54
mjt	the thing is -- rebase completed	19:54
jast	right, so that second part :)	19:54
	→ BryanRuiz11 joined	19:55
mjt	i see some git hashes in the terminal still	19:55
jast	look at 'git log -g'	19:55
	actually,	19:55
	look at 'git log -g <your branch>'	19:55
	← BryanRuiz1 left	19:55
jast	that'll show you the last bunch of values of the ref	19:55
	one of those should be the commit that was on the branch before you started the rebase	19:56
	← marcuy_ left	19:56
jast	(is that branch still active, i.e. checked out, by the way?)	19:56
mjt	gosh, that's alot of stuff... :)	19:56
	← subh left	19:56
	→ churp joined	19:56
	→ marcuy joined	19:57
	→ rcu joined	19:57
mjt	um	19:57
	→ patrick99e99 joined	19:57
	← rcu left	19:58
patrick99e99	so I did a auto-merge pull request on github, and see that the history has an additional commit for the merge added to it... I thought I might be able to do git rebase HEAD^2 -i and squash that last commit into the previous, but it didn't even list it... Is there any way to do this?	19:58
mjt	okay, i found my patchet, which is somehow detached now	19:58
	→ rcu joined	19:59
mjt	as one of the hashes still on the screen.	19:59
	branch - as i said - contains some very old version of it	19:59
	so it looks like everything's allright	19:59
wroathe	The company I work for is using a new workflow with pull requests and bug branches and they want everyone to branch off of production and then open a pull request for production	19:59
	If I accidentally branched off of master and pushed it	19:59
mjt	wow	20:00
wroathe	how could I remedy it so that branch is off of production and not master	20:00
	Without deleteing and re-pushing the branch	20:00
	→ Shusshu joined	20:00
	← Shusshu left	20:00
	← jast left	20:01
	→ erickr joined	20:01
cmn	you have to rewrite the branch anyway	20:01
	you can rebase	20:01
	← noisebleed left	20:01
cmn	but that means rewriting the branch	20:01
	→ jast joined	20:02
	→ Fandekasp joined	20:03
	← n1x left	20:04
	→ p3rror joined	20:04
	← sq-one left	20:05
	← Ryback_ left	20:06
	← bitkiller left	20:07
	→ Nik05 joined	20:07
	← griotspeak left	20:08
	← pandeiro left	20:08
	→ subh joined	20:09
	← tvw left	20:11
	← erichynds left	20:12
	→ fisted_ joined	20:12
	→ griotspeak joined	20:12
	→ adman65 joined	20:13
	← fisted left	20:13
	→ Vile joined	20:15
	← j416 left	20:16
	→ axl__ joined	20:19
	→ boombatower joined	20:20
	← axl_ left	20:22
	← ceej left	20:22
	← Raging_Hog left	20:24
	← juvenal left	20:24
	← joshsmith left	20:25
	← ericdc left	20:26
	← Cromulent left	20:26
	ChanServ set mode: +v	20:26
	→ `Assassin joined	20:27
	← defts left	20:28
	→ jameswarren joined	20:29
	← iamjarvo left	20:31
patrick99e99	so I did a auto-merge pull request on github, and see that the history has an additional commit for the merge added to it... I thought I might be able to do git rebase HEAD^2 -i and squash that last commit into the previous, but it didn't even list it... Is there any way to do this?	20:33
	→ dannas joined	20:33
canton7	patrick99e99, do you know what "HEAD^2" means?	20:33
	→ apix_ joined	20:34
patrick99e99	canton7: I thought it means two commits behind head ?	20:34
jast	no, that's what HEAD~2 means	20:34
canton7	patrick99e99, it means the second parent of HEAD	20:34
patrick99e99	aaaaaaaaaaahhh ooops	20:34
	so that was my problem?	20:34
canton7	yup	20:34
patrick99e99	dayum!	20:34
	gitinfo set mode: +v	20:34
apix_	whats the git equivallent of "hg glog" ?	20:35
jast	note that rebase -i destroys merges	20:35
	apix_: what does "hg glog" do?	20:35
apix_	its a glaphical log with all the commits	20:35
	command line graphical log	20:35
jast	!lol	20:35
gitinfo	git config --global alias.lol "log --oneline --graph --decorate"	20:35
apix_	that includes all vranches	20:35
	*brances	20:35
canton7	!lola	20:35
	awww :(	20:35
jast	sounds like the result would be hard to navigate	20:35
canton7	git log --graph --oneline --decorate --all anyway	20:36
	also look at gitk	20:36
jast	or tig if you prefer the console	20:36
apix_	ill check that tig and the git command	20:36
	cheers lads	20:36
	← canton7 left	20:37
	→ drizzd joined	20:37
	→ canton7 joined	20:37
	← CellKill left	20:38
	→ CellKill joined	20:38
	→ alx- joined	20:38
	← tsdh left	20:39
	← Kenjin left	20:39
	→ codeFiend joined	20:39
	← griotspeak left	20:40
	← codeFiend left	20:40
	→ rtuin joined	20:40
	→ mneorr joined	20:42
	→ griotspeak joined	20:42
	→ aspotashev joined	20:43
	← aeontech left	20:43
	← keystr0k left	20:43
	→ varnie joined	20:46
	← tommcdo left	20:47
	→ LiohAu_ joined	20:47
	← mneorr left	20:47
	→ linusoleander joined	20:47
varnie	I commited few commits and now I'd like to see what files exactly will be pushed if I enter 'git push origin/master'. is it possible?	20:47
	← p3rror left	20:48
	← dvide left	20:48
	← ph^ left	20:49
	← rtuin left	20:49
CareBear\	git log -p origin/master.. #the trailing dots are important	20:50
	← Ksugzulut left	20:50
canton7	varnie, you probably meant 'git push origin master'	20:50
jast	varnie: git fetch; git log @{u}..	20:50
	if master is checked out, that is	20:50
varnie	yes, I got it. thank you	20:50
	← MrMagic left	20:53
	← alex__c2022 left	20:54
	← y3di left	20:55
	← craigntammy left	20:55
	← canton7 left	20:56
	→ canton7 joined	20:56
	→ smw_ joined	20:56
	→ cyrou joined	20:58
	← troyt left	20:59
	← cannonball left	21:01
	→ p3rror joined	21:02
n3m	git seems to ignore the PATH set in my shell when calling the editor specified in core.editor.	21:02
	← rcu left	21:03
	← mrsrikanth left	21:03
	→ LmtdAt joined	21:03
	→ siprbaum joined	21:03
	→ y3di joined	21:03
	→ troyt joined	21:04
	← canton7 left	21:04
pom	I don't exactly understand the reasoning behind submodules. Doesn't a deployment break if a sub repo is missing?	21:04
	→ canton7 joined	21:05
	→ adulion22 joined	21:05
m1sc	n3m: !repro	21:06
gitinfo	n3m: Please paste (using https://gist.github.com/ or similar) a transcript of your terminal session. This will help immensely with troubleshooting.	21:06
	← MestreLion left	21:06
	← LmAt left	21:06
varnie	bye	21:07
	→ cjuner joined	21:07
	← varnie left	21:07
	→ hobodave joined	21:07
	← hobodave left	21:07
	→ hobodave joined	21:07
	→ joshsmith joined	21:08
	→ EricInBNE joined	21:09
	← pdtpatr1ck left	21:09
drizzd	Hrm. Just fixed a bug only to realize someone else already fixed it 3 weeks ago, even though the bug has been there for half a year...	21:09
	→ Slix` joined	21:09
	→ pdtpatr1ck joined	21:09
	gitinfo set mode: +v	21:10
	→ PElshoff joined	21:10
	← cyphase left	21:11
	← y3di left	21:11
Vinnie_win	bug in what	21:12
	→ eletuchy joined	21:13
	→ anli_ joined	21:13
anli_	Seems like I fail to push a repo to the server, I can commit locally, but the end server is not setup correctly	21:13
	(using tortoise)	21:14
	← spaceonline left	21:14
canton7	anli_> (using tortoise) <-- that's often a problem in itself. Any more details on the "seems like I fail to push"?	21:15
anli_	I am trying to push to a .git file that is on a smb share	21:15
canton7	push to a file? How does that work?	21:15
anli_	I actually setup that before, at least it seemed like that, but now that "master" is not visible when trying	21:15
	Not file actually, sorry, its a .git directory	21:16
	But I still only have one master, the local one	21:16
	→ cyphase joined	21:17
canton7	anli_, I'm afraid there's not nearly enough information there for someone to try and figure out what's going on	21:17
	→ marsilainen joined	21:17
marsilainen	is it common for 'push' to fail?	21:17
anli_	If the remote repo is not setup, its common	21:17
marsilainen	I'm pushing changes up to a remote location, over webdav, and I often seem to get failures a number of times quite frequently	21:18
	→ kennethreitz joined	21:18
marsilainen	then I'll try again and it works	21:18
	when you say 'not setup' what do you mean?	21:18
	→ alester joined	21:18
anli_	That I can only see the local repository	21:18
marsilainen	oh	21:18
canton7	marsilainen, when you said "fail", what do you mean?	21:18
anli_	That I have nowhere to push the changes	21:18
	Because that location is not setup correctly	21:19
marsilainen	canton7: I get output like:	21:19
anli_	oh, sorry, I must be tired, you talked to marsilainen	21:19
marsilainen	sending 425 objects	21:19
	MOVE 4634ceadd647e8ccd6604de5a9c3b0bd5d29b7e8 failed, aborting (22/404)	21:19
	Updating remote server info	21:19
	fatal: git-http-push failed	21:19
	the actual command that it fails on can vary	21:19
	← marthinal left	21:19
marsilainen	I have other connections to the server via ssh that stay fine during the entire time - so it doesn't just seem like a flaky network connection or something like that	21:20
canton7	marsilainen, looks webdav-related yeah. I don't entirely trust it tbh	21:20
marsilainen	sure	21:20
	perhaps I'd be better off doing it over ssh then?	21:21
	← gusnan left	21:21
canton7	marsilainen, in almost all cases	21:21
marsilainen	ok, I'll give that a try and see if it performs better	21:22
	thanks	21:22
	← Slix` left	21:23
	→ mattsa joined	21:23
	→ Slix` joined	21:23
	← Slix` left	21:23
canton7	anli_, i'm afraid there still isn't enough information. Can you tell us exactly what you're doing, and exactly what happens, and exactly how this differs from what you expect?	21:23
	→ Slix` joined	21:23
anli_	Exactly what happens? I probably cannot tell that. What I want to do is to take my existing repo and push that to a remote repo, which is a bare .git directory	21:24
	← Maccer left	21:24
	→ gusnan joined	21:25
	→ malkia joined	21:25
	→ texasmynsted1 joined	21:25
	→ proq joined	21:26
anli_	Hm, trying to rename the dir with my working copy, is tortoise hindering me to do that?	21:26
	Some process uses that dir	21:26
canton7	anli_, I wouldn't be surprised. It's generally a bit rather shit	21:26
	→ Maccer joined	21:26
	gitinfo set mode: +v	21:27
offby1	anli_: unless you're already thoroughly familiar with git, I suspect that using tortoisegit is making your life harder, not simpler.	21:27
	← tommyvyo left	21:27
anli_	You all think I have choosen it myself, right?	21:27
offby1	which I suppose is just a longer-winded way of saying what canton7 just said :)	21:27
	← griotspeak left	21:27
	← keller left	21:27
offby1	anli_: yes, I probably am assuming that.	21:27
	It's not relevant, though. What's relevant is: if you can use the command line, you ought to give it a try	21:28
anli_	The question whether tortoise git is a good choice or not is not my favorite topic at the moment :)	21:28
offby1	surely you can install msysgit and try it?	21:28
	→ jigal joined	21:28
offby1	is your machine so locked-down that you cannot?	21:28
anli_	yes	21:28
offby1	oh.	21:28
	I have little to offer, then, apart from sympathy.	21:29
	← ibash left	21:29
anli_	Hehe	21:29
canton7	anli_, I was answering "Hm, trying to rename the dir with my working copy, is tortoise hindering me to do that?"	21:29
jigal	Hello i am using a submodule in my git repos. How can i make sure i have the latest version of this submodule?	21:29
anli_	Thanks for not quoting what I am saying	21:29
	→ davpoind joined	21:30
	← dzonder left	21:30
anli_	Seems that the easiest way is to checkout the repo and drag files into the resulting working dir	21:31
offby1	anli_: just out of curiosity: have you told whoever locked down your machine that they're making your job harder?	21:32
	→ madewokherd joined	21:32
anli_	No	21:32
	← adman65 left	21:32
	← paul424 left	21:34
	→ empity joined	21:34
anli_	But with that logic, I would need to speak with Torvalds as well	21:34
	← Vile left	21:35
	← luxurymode left	21:35
	→ ysawej joined	21:35
	← MANCHUCK_ left	21:36
	← mattsa left	21:38
	← adulion22 left	21:38
	← anli_ left	21:39
	← wildalaskan left	21:39
	← dannas left	21:41
	← _iron left	21:44
	← marsilainen left	21:44
cespare	jigal: The latest version of the submodule specified in the parent repo or the latest version of the submodule repository that exists?	21:45
	→ erichynds joined	21:46
	→ martinjlowm joined	21:46
	← eletuchy left	21:46
	← jangell left	21:49
	→ Jackneill joined	21:49
	→ FernandoBasso joined	21:49
	→ de joined	21:50
CareBear\	anli_ : add a remote which refers to your .git directory on the smb server	21:50
	anli_ : I think the URL must be a URL and that a simple z:\folder may not work	21:51
offby1	forward-slashes might also help	21:51
	← aspotashev left	21:52
necromancer	you know how there's the foo bar?	21:53
	i think we should have a git pub	21:53
	← Jackneill left	21:53
	← stkrzysiak left	21:53
	← joshsmith left	21:54
	→ n1x joined	21:54
	← s0ber left	21:54
	→ aspotashev joined	21:54
CareBear\	I'll have a burger and a pint, please	21:54
	← lyddonb left	21:55
	← xrated left	21:55
	← dv310p3r left	21:55
	← dannyanderson left	21:56
	→ s0ber joined	21:56
	→ axl_ joined	21:56
	← mabrand left	21:58
	← spiette left	21:58
offby1	haw	21:58
	← juanfer left	21:59
offby1	necromancer: or even a "git hub" -- a lounge, bar, meeting place	21:59
	← john2496 left	21:59
offby1	crazy	21:59
	← diegoviola left	22:00
	← axl__ left	22:00
necromancer	yeah we should probably have a book of faces of people who come by often	22:00
	maybe we could call it a "face book" of sorts	22:00
	brb im gonna go hang out with some girl coders	22:01
	in the git tub	22:01
	← iocor left	22:01
	← ehsan left	22:02
	← axl_ left	22:02
	→ pfFredd joined	22:02
	← de left	22:03
	← wroathe left	22:03
	← apix_ left	22:04
	← digitalfiz left	22:04
	← Nik05 left	22:05
	← ology left	22:05
texasmynsted1	Could somebody recommend a link or a book for git best practices, especially for a large organization? (I have seen some sites, etc. But git has been around long enough to have some good best practices.)	22:06
SethRobertson	!best_practices	22:06
gitinfo	There is no one right answer for git best practices, but a consensus from #git is available at http://sethrobertson.github.com/GitBestPractices/	22:06
offby1	texasmynsted1: well "gitflow" is one	22:06
Nugget	Don't mess with Texas.	22:06
offby1	!gitflow	22:07
gitinfo	The description of the gitflow branch workflow model is at http://nvie.com/posts/a-successful-git-branching-model/, while a tool to help implement this workflow is at https://github.com/nvie/gitflow See http://sethrobertson.github.com/GitBestPractices/#workflow for other workflow suggestions/references	22:07
texasmynsted1	Nugget, that is good advice.	22:07
offby1	SethRobertson: somehow I'd never come across that before; thanks	22:07
	offby1 reads eagerly	22:07
texasmynsted1	Thank you offby1	22:07
	→ iocor joined	22:08
	→ marcogmonteiro joined	22:08
texasmynsted1	et al	22:08
	→ axl_ joined	22:09
	→ shruggar joined	22:11
	→ phlogisto joined	22:11
	→ _iron joined	22:11
	→ warthog9 joined	22:11
phlogisto	hey guys! I accidentaly pushed something bad to master. How can I revert the changes?	22:12
	← p3rror left	22:13
wereHamster	you push the old commit to master	22:14
offby1	!fixup	22:14
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. For hints type !fixup_hints in IRC. Remember: if you have pushed already, there are only a few things you can do without !rewriting_public_history (type that for more info)	22:14
	← ddv left	22:14
	→ RaceCondition joined	22:14
	← churp left	22:14
	← empity left	22:15
	→ cybersphinx joined	22:15
	← Dave^\| left	22:17
offby1	this Seth Robertson fellow is prolific	22:17
phlogisto	gitinfo: wereHamster: thank you guys!	22:17
gitinfo	phlogisto: you're welcome, but please note that I'm a bot. I'm not programmed to care.	22:17
	← the_cybersphinx left	22:17
	← duckx left	22:19
wereHamster	Y U no care?	22:20
	!botsnack	22:21
gitinfo	Om nom nom	22:21
	← kermit left	22:21
jast	wereHamster: that won't make it care, either! mwahaha!	22:22
	→ localhost joined	22:22
	← boombatower left	22:23
	← beatak left	22:23
	→ Atlantic777 joined	22:23
	← aspotashev left	22:24
	← pantsman left	22:25
	← texasmynsted1 left	22:25
deavid	i need a bit of help with "git svn clone"	22:26
	→ thursday joined	22:26
	← iocor left	22:26
deavid	guess what, i forgot to add the "-s" switch to svn clone, and i got a repo with the folders trunk/ tags/ and branches/	22:27
	→ p3rror joined	22:27
deavid	i want to fix that. But i spent hours cloning 2000+ revisions from svn that i already have in my local machine	22:27
wereHamster	clone again while adding -s	22:27
cbreak	deavid: yeah. You can try to manually fix it	22:27
FauxFaux	It'll be much, much slower.	22:28
cbreak	but you'll spend days on it	22:28
deavid	yeah, but i can do anything to avoid downloading...	22:28
	ouch	22:28
cbreak	well then	22:28
	anything? :)	22:28
	then do the git svn clone on the server	22:28
deavid	sure? so guess is better to clone again :-( :-(	22:28
cbreak	it'll be faster	22:28
deavid	it's not mine :-P	22:28
cbreak	hire someone to steal the hard disk	22:28
wereHamster	maybe you can try svnsync the repo first and then git svn clone that	22:28
deavid	is a server with a slow internt conection	22:28
wereHamster	git svn clone is restartable. so you can stop and continue at any time	22:29
	→ aspotashev joined	22:29
phlogisto	!fixup_hints	22:30
gitinfo	Hints for fixing commits are: (1) NOT PUSHED/PUBLISHED: `git rebase -i $COMMIT^` or perhaps `git commit --amend` (or `git reset HEAD^`). (2) OTHERWISE, `git revert $COMMIT` to make a reverse commit. (3) If you have pushed and MUST remove it, use rebase or filter-branch and type !rewriting_public_history in IRC.	22:30
cyrou	is there an equivalent to git svn clone for CVS :s	22:30
phlogisto	!rewriting_public_history	22:31
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)	22:31
cbreak	maybe there's a cvs fast export	22:31
	← kadoban left	22:31
wereHamster	cyrou: cvsimport	22:31
cyrou	ah, okay	22:31
wereHamster	git-cvsimport	22:31
	← subh left	22:31
cyrou	does that bring histories too?	22:31
	→ kadoban joined	22:32
wereHamster	yeah	22:32
	← fly9 left	22:32
deavid	ok, i give up. I'll clone again with -s. Guess i have all the night to get that	22:33
	→ kadoban_ joined	22:33
	← seanh left	22:33
	← phlogisto left	22:34
	→ yizumi joined	22:34
	← jason237 left	22:34
	← RaceCondition left	22:35
	← cakehero left	22:36
	← scelis left	22:36
	→ iocor joined	22:37
	→ marcuy_ joined	22:38
	→ seanh joined	22:39
	← rendar left	22:40
	→ Frederik joined	22:40
	gitinfo set mode: +v	22:41
Frederik	Hi	22:41
gitinfo	Frederik: 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.	22:41
	→ ph^ joined	22:42
Frederik	I could use some help with git..	22:42
	← marcuy left	22:42
	→ kermit joined	22:42
Frederik	I need to know how I can update my repository from the "master" repository..	22:42
	← AaronMT left	22:43
Frederik	Is that possible?	22:43
cbreak	from the origin repository?	22:43
	git remote update will get remote infos	22:43
	← linusoleander left	22:44
	→ subh joined	22:45
	→ Cromulent joined	22:45
Frederik	Yea.. Well I'll try to explain what I have done so far..	22:45
jigal	why are the DoctrineORMModule files not downloaded from github? http://pastebin.com/uEiyF9gB	22:45
	← alester left	22:46
	→ RaceCondition joined	22:46
jast	jigal: was there already already a repository in the target location?	22:47
Frederik	I forked the project https://github.com/woothemes/woocommerce.git. This created a new repository (https://github.com/my_username/woocommerce.git). In the Git GUI I cloned the newly created repository. Then I added a "remote" called upstream to the source https://github.com/woothemes/woocommerce.git	22:47
	← preflex left	22:47
Frederik	What is the next step?	22:47
jigal	jast, no	22:47
	jast, i deleted it before	22:47
	← patrick99e99 left	22:47
jast	jigal: by the way, 'git submodule add' doesn't accept a --recursive option AFAIK	22:48
FauxFaux	Frederik: git pull upstream master	22:48
Frederik	To synchronize the original repository with my repository..	22:48
	Is there anything else??	22:48
jigal	jast, ok so what should i do to get it as a submodule	22:48
	← alberto56 left	22:48
FauxFaux	Frederik: Do you want to publish your update? If so, you should publish it like any other set of commits.	22:49
	→ preflex joined	22:49
jast	jigal: try git submodule update --init	22:49
	← Fhsrap left	22:49
erickr	I just pushed a local branch to remote origin, as a feature branch. But my local branch doesnt seem to track that remote branch automatically. What do I need to do to track it?	22:49
	→ alberto56 joined	22:49
jast	erickr: -u option to push on next push, or git branch --set-upstream	22:49
jigal	jast, for now i removed the submodule how do i start again ?	22:49
Frederik	i tried to merge the two repositories, and when I pushed the commit button it "updated" my local personal repository on github	22:49
	← hobodave left	22:50
	→ manithree joined	22:50
	→ ptn777 joined	22:50
FauxFaux	Frederik: "local personal repository on github" makes no sense.	22:50
jast	yeah, github isn't exactly local	22:50
Frederik	Sorry not local :( but the repository linked to my account. But when I then tried to push my own updated files from my repository to the original one, it would commit all the commits that I updated from the original one..	22:51
	← Atlantic777 left	22:52
	→ WillMars_ joined	22:52
Frederik	I apologize but I'm totally new to git/guthub	22:52
jast	Frederik: to be able to push directly to the original repository, you need push access to it	22:52
	← cyrou left	22:52
jast	that's the whole reason why you need to fork... because you don't have push access	22:52
erickr	jast: tried that but get, Branch origin/Feature903-new-drop-down-menu set up to track local branch Feature903-new-drop-down-menu.	22:52
	feels like it should be the other way around..	22:52
jigal	jast, how to start again from 0	22:52
jast	a common way to let the original maintainer know about the changes in your fork is to send a pull request via git's interface	22:52
	err, github's interface	22:52
	erickr: the proper syntax is: git branch --set-upstream foo origin/foo	22:53
	→ heppy joined	22:53
jast	jigal: just doing the add again should do it... then again it should have done it before, too	22:53
erickr	jast: ahhh..	22:53
	← martinjlowm left	22:53
Frederik	Yea I know. I have only read permissions, so I should push to my own repository and then make a pull request.. But the thing that confuses me, is the thing that after the update, it would commit alle updated files..	22:54
erickr	jast, I get, warning: refname 'origin/Feature903-new-drop-down-menu' is ambiguous.	22:54
	fatal: Ambiguous object name: 'origin/Feature903-new-drop-down-menu'	22:54
Frederik	Update from original --> my repository --> the original again..	22:54
jast	erickr: yeah, your first try broke it ;) do a 'git branch -d origin/Feature903-new-drop-down-menu' first	22:54
cbreak	what do you mean with update?	22:54
	git remote update as I wrote above?	22:55
	← _iron left	22:55
jigal	jast, but then i won't get the recursive files from this submodule	22:55
jast	jigal: you should. otherwise something very weird is going on.	22:55
erickr	jast, says its not found.. :)	22:55
	→ eletuchy joined	22:55
	→ Malar joined	22:55
jast	erickr: gah... does 'origin/Feature903...' show up in "git branch"?	22:56
Frederik	I fetched the original repository, then merged it with my local files. The local files I comitted/pushed to my personal git repository.	22:56
jigal	jast, how can i find out	22:56
	← TeckniX left	22:56
	← ipalaus left	22:56
cbreak	Frederik: no, git merges with branches, not files	22:56
erickr	jast: no, just the Feature903-new-drop-down-menu without origin/	22:56
cbreak	so if you merged an other branch	22:56
jast	jigal: sorry, based on what you pasted earlier I have no idea what's going on	22:56
cbreak	then you only have to commit yourself if there were conflicts	22:56
	after resolving them	22:56
jigal	jast, how can i give you more info	22:56
	→ luke-jr joined	22:56
cbreak	otherwise it'll commit automatically	22:56
luke-jr	Does git check for integrity while receiving objects from a push?	22:57
	LiohAu_ → LiohAu	22:57
cbreak	also, you can not push files with git	22:57
	you can only push history (commits)	22:57
Frederik	Then I added an update to my personal repository. Then I wanted to make a pull request to the original repository, but it would make a pull request on 69 files instead of the 2 files I edited..	22:57
jast	erickr: does this give you more than one line: git show-ref origin/Feature903-new-drop-down-menu	22:57
	→ tvw joined	22:57
Frederik	hmm..	22:57
jast	luke-jr: yes, by design. corrupted objects are automatically invalid.	22:58
Frederik	I just wanted to know how to do it right..	22:58
	→ _iron joined	22:58
	→ edude03 joined	22:58
luke-jr	I keep getting corrupt objects :/	22:58
	→ TeckniX joined	22:58
jast	luke-jr: RAM fault? or frequent power losses	22:59
cbreak	Frederik: pull requests are not on files	22:59
	they are on history (commits)	22:59
luke-jr	is it possible for a packed object to be corrupt?	23:00
cbreak	maybe you can make cherry-pick requests or so	23:00
	luke-jr: sure it is	23:00
jast	luke-jr: sure, if the pack file is corrupted	23:00
luke-jr	hmm	23:00
	← TeckniX left	23:00
luke-jr	mine are always unpacked objects	23:00
cbreak	just edit it with a hex editor	23:00
	and you can corrupt anything	23:00
luke-jr	cbreak: right, but I mean woudl git notice?	23:00
	→ MestreLion joined	23:00
cbreak	sure it would	23:00
	if you fsck	23:01
jast	luke-jr: it'll notice whenever the object is access	23:01
	ed	23:01
luke-jr	would git gc access it?	23:01
cbreak	as soon as the hashes don't match anymore git will complain	23:01
jast	Frederik: check on github to make sure that your commits contain only the changes you wanted... go to the 'commits' tab on the github interface for your repo and look at the individual changesets	23:01
	git gc would definitely access it	23:01
erickr	jast, gives one return	23:02
	← kerrick left	23:02
Frederik	Is the right way to do it to run the "git pull upstream master" and then push the files I edit?	23:02
jast	erickr: that's weird, because then that command has no business complaining about ambiguities...	23:02
erickr	mm..	23:02
jast	Frederik: to do what?	23:02
erickr	is it that my local branch is named the same perhaps?	23:02
jast	Frederik: push only uploads commits, not uncommitted changes	23:03
MestreLion	is it normal for an email sent to the list take > 10 minutes to show in rss/web interface?	23:03
Frederik	To be sure that the repositories are syncronized..	23:03
luke-jr	any way to get git gc or fsck to report all corrupt objects at once?	23:03
jast	erickr: is your local branch named 'origin/Feature903...'? no it isn't ;P	23:03
	← aspotashev left	23:03
luke-jr	rather than fsck, fix, fsck, fix	23:03
jast	MestreLion: yes, because greylisting	23:03
Frederik	So first I'll make the commit..	23:03
erickr	jast: only Feature903...	23:03
Frederik	and then push?`	23:03
erickr	jast: not the origin-part..	23:03
jast	erickr: yeah, that's a different name. so no ambiguity.	23:03
erickr	jast: I'll try to checkout with a different name,..	23:03
jast	erickr: just to be sure, what's your git version? (git --version)	23:03
MestreLion	jast: Oh... I thought the list wasn't moderated	23:04
jast	MestreLion: has nothing to do with moderation	23:04
	← chrisgeorge left	23:04
jast	it's an automatic process	23:04
erickr	jast: 1.7.9	23:04
jast	erickr: okay. definitely new enough. :)	23:04
erickr	jast: yes, upgraded fairly recently..	23:04
jast	erickr: does "git rev-parse origin/Feature903..." give you the same complaint?	23:05
Frederik	I'll see if I can google something out.. Thank's for the support so far :)	23:05
	→ aspotashev joined	23:05
MestreLion	jast: I've sent the doc patch... when it pops in, any feedback from you guys would be greatly appreciated	23:05
jast	Frederik: if you want to publish changes, then yes, commit them and then push.	23:05
erickr	jast: only a sha1	23:05
	jast: i checked it out with another name with git checkout -b localname origin/remotebranch and that seems to work..	23:06
	→ josx joined	23:06
	← manithree left	23:06
	→ fayimora joined	23:07
Frederik	jast: Thx. Can you easily explain how I should deal with it if the "original" repository update something. So my local files and my repository get's updated?	23:07
	→ explodes joined	23:07
explodes	Production emergency!	23:07
Frederik	jast: and still be able to push commits?	23:08
cbreak	fast! Charge the shield emiters!	23:08
	and depolarize the dilithium core!	23:08
	← ptn777 left	23:08
explodes	I have local code (from another developer) that is broken and was cleared to be pushed onto the master branch, and pushed to production	23:08
	← john_anderson left	23:08
explodes	I need master to be reset to commit 1a5f14fd403faedf22da94937aaf90cfb36e90dc	23:08
FauxFaux	Did everything explodes?	23:08
	← kadoban left	23:08
FauxFaux	explodes: !fixup	23:08
gitinfo	explodes: 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. For hints type !fixup_hints in IRC. Remember: if you have pushed already, there are only a few things you can do without !rewriting_public_history (type that for more info)	23:08
explodes	How do I do this :S	23:08
jast	Frederik: that's precisely the reason(s), yeah :)	23:08
	← kadoban_ left	23:08
explodes	!fixup_hints	23:08
gitinfo	Hints for fixing commits are: (1) NOT PUSHED/PUBLISHED: `git rebase -i $COMMIT^` or perhaps `git commit --amend` (or `git reset HEAD^`). (2) OTHERWISE, `git revert $COMMIT` to make a reverse commit. (3) If you have pushed and MUST remove it, use rebase or filter-branch and type !rewriting_public_history in IRC.	23:08
	→ kadoban joined	23:08
	← kadoban left	23:09
explodes	Guh.	23:09
Frederik	jast: thx.. I'll try it out :)	23:09
	→ kadoban_ joined	23:09
Frederik	exit	23:09
	← Frederik left	23:09
jast	erickr: I'd just try the same --set-upstream command again. if it fails again, I'd like to see the exact command just to make extra sure there isn't anything else wrong with it :)	23:09
cbreak	explodes: you can rewrite any history you want if you're prepared to deal with the fallout	23:09
explodes	i am not	23:10
cbreak	for example with git reset --hard 1a5f14	23:10
SethRobertson	Perhaps you should send someone in here who is prepared to deal with the fallout	23:10
cbreak	explodes: then you can just make new history with revert	23:11
	just git revert the commits that did bad things	23:11
SethRobertson	or just checkout that SHA and do whatever you do to publish that code to production	23:11
	← Destos left	23:12
	← JJasonClark left	23:13
	← apok_ left	23:14
	← whitman left	23:15
	← mister_m left	23:16
erickr	jast: well, what do you know..	23:16
	jast: now it worked..	23:16
	← subh left	23:16
erickr	jast: strange..	23:17
jast	erickr: let's just both whistle and look innocent	23:17
erickr	I typed the exact same command before,	23:17
	according to history	23:17
	→ jangell joined	23:17
erickr	I did switch to master and pulled before this last attempt though, if that might have done any difference..	23:17
jast	not that I can think of	23:18
	← blacktulip left	23:18
erickr	jast: this is the commands, http://pastebin.com/MRwkCr8t	23:19
	oh well, thanks anyway.. :)	23:19
jast	erickr: #1996 seems weird :)	23:20
	ah well, the history is pretty messy at this point. let's just pretend nothing weird ever happened... much easier to understand that way	23:21
	← explodes left	23:21
	← CannedCorn left	23:22
	← dreiss left	23:22
	← Chillance left	23:22
CareBear\	how to find out if a merge will be fast-forward without actually doing the merge?	23:23
cbreak	CareBear\: you can rev-list with the correct arguments	23:23
	if theirs..yours is empty, then it'll be ff	23:24
offby1	CareBear\: interesting question. I just go ahead and do it, and "git reset --hard HEAD^" if I don't like the result.	23:24
cbreak	you can do merge --ff	23:24
	if you want to attempt it	23:24
	offby1 smacks forehead	23:24
offby1	of course.	23:24
	CareBear\: what cbreak said.	23:24
CareBear\	mh, no, merge --ff creates a merge commit if neccessary	23:25
	← PElshoff left	23:25
CareBear\	--no-ff creates a merge commit always	23:25
cbreak	--ff-only of course	23:25
	← EvanR left	23:25
CareBear\	aha! once again my old man pages betray me!	23:25
	man git-merge	23:25
gitinfo	the git-merge manpage is available at http://jk.gs/git-merge.html	23:25
	← yoklov left	23:25
	← dfr\|mac left	23:28
	← alexim left	23:28
	→ Davey_ joined	23:28
	→ Sigma joined	23:29
alberto56	hi, does anyone kwno if it's possible to have a fallback git submodule remote address, so that if someone doesn't have access to the submodule through its principal URL, one can still run git submodule update and have access to the submodules in read-only form; but devs will be able to modify and push submodules. Cheers.	23:29
cbreak	alberto56: just put the read-only ones into .gitmodules	23:30
	the devs can just set a custom url in the .git/config	23:30
	← davpoind left	23:30
cbreak	or just manually fetch/reset inside the submodule	23:30
	you can git-remote-add in submodules and then use them like a normal repo	23:31
kevlarman	alberto56: generally you don't want to try to develop inside a submodule anyway	23:31
	← eletuchy left	23:31
alberto56	cbreak sounds good. So by default, the submodules is readonly for anyone. Thanks!	23:31
	← fayimora left	23:31
cbreak	submodules are designed to be read only	23:32
	← Russell^^ left	23:32
cbreak	you can use them if you commit to them, just be aware of the problems	23:32
	→ cordoval joined	23:32
	← shruggar left	23:32
alberto56	kevlarman cbreak thanks, makes sense. But in my case I'm starting many submodules from scratch on a big project so it's practical de develop them together, if that makes sense.	23:32
cordoval	hi guys i was doing some cherry-picking but now my console has this label	23:33
	~/sites-2/vespolina_sandbox (master *+\|CHERRY-PICKING)	23:33
	and I don't know what to do to go back to normal	23:33
	← n1x left	23:33
cordoval	what can I do?	23:33
cbreak	maybe gitslave is better for your usecase	23:33
	→ CannedCorn joined	23:33
cbreak	cordoval: git status?	23:33
cordoval	what do you mean git status	23:33
	it modified a file	23:33
cbreak	maybe cherry-picking failed and you have to resolve the conflict and then commit. read git status for details	23:33
kevlarman	cbreak: it will keep getting its HEAD detached though, more practical to use a separate clone to develop it	23:33
cordoval	and there seem to be some conflict but git merge --abort git cherry-pick --abort does not work	23:34
	how to abort thouugh	23:34
cbreak	kevlarman: nah, you can switch to a branch if you want	23:34
	you just can't use submodule update :)	23:34
cordoval	cbreak: how to abort a cherry-picking?	23:34
kevlarman	cbreak: exactly	23:34
alberto56	cbreak thanks for your help, I'll look into git slave.	23:34
cbreak	cordoval: git reset --hard the stuff away?	23:34
	and afterwards, maybe delete CHERRY_PICK_HEAD	23:35
	← shuffle2 left	23:35
cordoval	thanks cbreak	23:35
	you rock	23:35
	what is the best way to cherry-pick, I am trying to get some of the latest commits from a repo and go one by one discriminating, even using git cola that shows the commits but I feel like there should be a better bookkeeping tracking safer smarter way	23:36
	→ Prudhvi joined	23:36
	← iivvoo left	23:36
	→ shruggar joined	23:36
cbreak	imho, the best way is to not do it	23:36
	← airborn left	23:37
cordoval	why not cbreak also trying this ~ git log --not upstream/master but it seems something is not working	23:37
	missing	23:37
cbreak	hmm?	23:37
	your understanding is missing	23:37
	git cherry-pick duplicates commits	23:38
	you do NOT get commits from somewhere else	23:38
	you just get their effect and meta copied	23:38
	if you want to get commits from somewhere, use merge, or rebase onto it	23:38
cordoval	right but at least i want to see which commits to cherry pick	23:38
	and that is the difference	23:38
cbreak	if you want to continue using cherry-pick to copy commits	23:38
	then use git cherry to see which you already have	23:38
	→ alexim joined	23:38
cbreak	note that this uses a heuristic	23:38
	git log is not usable for that	23:39
	→ PsiliPharm joined	23:39
cbreak	git log assumes sane histories :)	23:39
cordoval	cbreak: i cannot rebase as i don't want all	23:39
	i did not know git cherry	23:39
	it seems powerful and just what i needed	23:39
	it is not so popular, wonder why	23:39
	maybe then i wonder if i can do single commit merges?	23:40
	git merge sha1	23:40
	is that valid?	23:40
	but I don't want to go one by one	23:40
cbreak	cherry-pick is a mess	23:40
	merging merges history	23:40
cordoval	and cherry?	23:41
	I mean the problem i am trying to resolve is	23:41
	I am parting my ways from a popular repo	23:41
	but i need to eventually copy cheat on some of their commits	23:41
	→ Per4u3e joined	23:41
cordoval	as some things are related, so the question is how or what is the best way for doing this	23:42
	← ph^ left	23:42
	→ Chib joined	23:42
__machine	when i do git svn dcommit, git seems to hard reset to the latest commit, then mixed reset to the svn tip and progressively put changes into the index and commit... but if a 3rd party commits to svn during my dcommit (e.g. 15 commits in our of 30)...	23:42
	git tells me that i need to rebase but i cant rebase because there are working copy changes that need to be committed or stashed... but those changes are already committed... and dont need to be stashed... why does git leave these files in the working copy, and not automatically just cleanup the working copy, rebase again and continue dcommit?	23:42
cordoval	cbreak: I actually want the inverse of git-cherry - Find commits not merged upstream	23:42
	i want git-uncherry - Find commits not merged downstream	23:43
	← Cromulent left	23:43
cbreak	__machine: git svn dcommit includes git svn rebase	23:43
drizzd	__machine: you're asking why git does not irreversibly delete files from your working copy?	23:43
	→ rchavik joined	23:44
	← rchavik left	23:44
	→ rchavik joined	23:44
cbreak	cordoval: and what do you think it does?	23:44
	← shruggar left	23:44
	← williamcotton left	23:45
drizzd	__machine: so you are in the middle of an aborted dcommit? What does git name-rev HEAD say?	23:45
	← rchavik left	23:45
	→ rchavik joined	23:46
cordoval	cbreak: by its description I think it only outputs the commits that are upstream but not downstream	23:46
	→ mishok13 joined	23:46
	← Exposure_ left	23:46
__machine	drizzd: the files are NOT local changes that need to be stashed or committed... i ALREADY did git svn rebase... then... during dcommit... git puts files in the working copy itself (not me)... if a commit to svn happens during my dcommit... git says it needs to rebase again, but that it cant because of working copy files...	23:47
	→ NyB_ joined	23:47
__machine	drizzd: im saying git should be smart enough to know that the files in the working copy that it is currently trying to dcommit are put there by git, not the user, and nothing will be lost by removing them, and git should be able to rebase	23:47
drizzd	__machine: ok, I thought you needed help. Never mind then.	23:48
	→ wroathe joined	23:48
cordoval	maybe someone knwos of a tutorial for git cherry?	23:48
	i just skim git cherry --help but ...	23:49
cbreak	cordoval: I don't get your problem	23:50
	upstream is just a name	23:50
	you can say that your branch is upstream	23:50
	and the other one is the other one.	23:50
	→ mister_m joined	23:50
	← kremiak left	23:50
	← k6b left	23:50
__machine	drizzd: here's the log... http://dpaste.com/717189/	23:50
	→ k6b joined	23:50
	← k6b left	23:50
cordoval	right cbreak, just hard to get my thoughts match the reverse	23:50
	← ScottE left	23:51
cordoval	I am so accustomed to names matching stuff :D	23:51
cbreak	it matches. It's just a matter of perspective	23:51
	← drizzd left	23:51
__machine	drizzd: r10809-108220 and r10822 are svn commits made by git svn dcommit for my local commits... and r10821 is a commit made by a 3rd party while my dcommit was in progress...	23:51
cordoval	right I have a slow changing perspective cbreak	23:52
	that is why a tutorial can be helpful :D	23:52
	→ dv310p3r joined	23:52
	→ Exposure joined	23:52
cordoval	or a blog post	23:52
	→ ScottE joined	23:52
	← tatsuya_o left	23:53
EugeneKay	!blog	23:54
gitinfo	Blog posts, while helpful and informative, are quite often outdated, give bad advice, or are just plain wrong. Please don't rely solely upon them, or treat them as authoritative.	23:54
	← josx left	23:54
	→ fayimora joined	23:55
mythmon	how can I revert all commits between some commit abcdef and HEAD?	23:55
EugeneKay	man git-revert; it takes multiple <commit> parameters.	23:55
gitinfo	the git-revert manpage is available at http://jk.gs/git-revert.html	23:55
	← L42y left	23:55
mythmon	i tried that, and it gave me conflicts.	23:56
	→ enmand joined	23:56
	← Dragooon left	23:56
mythmon	also, is there a way i can give a range of commits, instead of each manually?	23:56
EugeneKay	If abcdef is the commit you want the current state to be at you can checkout abcdef and then commit -a, or reset --hard abcdef then reset --soft to what used to be your HEAD.	23:56
	See also !fixup	23:57
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. For hints type !fixup_hints in IRC. Remember: if you have pushed already, there are only a few things you can do without !rewriting_public_history (type that for more info)	23:57
	→ tatsuya_o joined	23:58
	← warthog9 left	23:58
mythmon	i want to make a commit that undoes everything in the range. will those two do that?	23:58
	ie: no rewriting history.	23:58
EugeneKay	reset will make it exactly as it was at abcdef, effectively an "undo"	23:59
mythmon	thats not what i want.	23:59
	hmm, i was able to revert them all in a bunch of commits, now i just need to squash them	23:59

Logs Search ←Prev date Next date→

Channels Documentation