IRCloggy #git 2012-11-29

Provider of IRC logs since 2005.

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

	2012-11-29
steveoh	cbreak: you got a minute for some git branch -m	00:00
cbreak	that's branch renaming, not what you want	00:00
steveoh	it is what i want	00:01
cbreak	then do it	00:01
	→ eijk joined	00:02
	steveoh sigh	00:02
	→ Mag-en00b joined	00:02
	← samuel02 left	00:02
cbreak	what's your problem?	00:03
	← shruggar left	00:03
cbreak	is it not knowing how to use git branch -m?	00:03
	← sirdancealot8 left	00:04
	← kevwil left	00:04
Mag-en00b	http-backend sets GIT_COMMITTER_NAME, yet when i push or commit, everything in the repository is made with the original committers name..?	00:04
	→ Parhs joined	00:04
wereHamster	Mag-en00b: pushing doesn't change the committer identity	00:05
	in fact, pushing does not modify the transmitted data at all	00:05
	← rednaks left	00:05
	← hyperair left	00:05
cbreak	Mag-en00b: committer-name only affects committing	00:06
	http has nothing to do with committing	00:06
	← infynyxx left	00:06
	→ slamice_ joined	00:07
	← fly9 left	00:07
	← rgr left	00:07
	← bubbles10101 left	00:07
	← nitrix left	00:07
	← slamice left	00:07
	slamice_ → slamice	00:07
	→ kerrick joined	00:08
	→ bubbles10101 joined	00:08
	← bubbles10101 left	00:08
	→ bubbles10101 joined	00:09
	← btree left	00:09
Mag-en00b	cgi git http-backend in question, so what is this line supposed to do? https://github.com/git/git/blob/master/http-backend.c#L335	00:09
	→ Back2Basics joined	00:09
	→ cbreak\|evil joined	00:10
wereHamster	Mag-en00b: https://github.com/git/git/commit/e32a4581bcbf1cf43cd5069a0d19df07542d612a	00:10
	→ chaiz joined	00:11
	← eijk left	00:11
	← kerrick left	00:11
	← BD- left	00:11
	→ kerrick joined	00:12
	→ eijk joined	00:12
Mag-en00b	that i know, i posted report to the mailing list about the documentation bug relating to that commit message	00:12
	→ TeamColtra joined	00:12
	← v0n left	00:12
Mag-en00b	but as i just tried to set the variable GIT_COMMITTER_NAME for the cgi-program and push some data, nothing changed. committer's names were same as before. how is that variable supposed to help anything when all that cgi http-backend does is receive packs?	00:13
	← cbreak left	00:13
	cbreak\|evil → cbreak	00:13
wereHamster	Mag-en00b: ask peff .. ?	00:13
cbreak	Mag-en00b: as I said	00:13
	it's not supposed to do anything unless you commit	00:13
	http has nothing to do with committing	00:13
	pushing has nothing to do with committing	00:13
	← kermit left	00:14
steveoh	phew	00:14
	i made it work!	00:14
	wow	00:14
Mag-en00b	yeah. thus it seems seems that functionality is there for nothing	00:14
cbreak	easy, wasn't it?	00:14
steveoh	i'm a git phanboi now	00:14
	no i hated it	00:14
cbreak	Mag-en00b: no, for committing	00:14
steveoh	it scared me	00:14
FauxFaux	Mag-en00b: The tests talk about the reflog.	00:14
steveoh	:)	00:14
wereHamster	cbreak: can http-backend commit?	00:14
steveoh	thanks	00:14
cbreak	no	00:14
FauxFaux	It updates the reflog, though.	00:14
cbreak	as I said	00:14
	it's for committing	00:14
	not for pushing	00:14
wereHamster	cbreak: so.. if http-backend can not commit, why does it export GIT_COMMITTER_NAME	00:15
	?	00:15
FauxFaux	Including committing updated reflog entries.	00:15
	For the reflog.	00:15
cbreak	wereHamster: don't ask me. Maybe so hooks can check it or something?	00:15
FauxFaux	wereHamster's link is very explicit.	00:15
	→ emankcin joined	00:16
Mag-en00b	i'm quite confused. though might have something to do with the fact that i'm not so familiar with git.. reflog committing.. doesnt open up a bit to me :x	00:16
	→ ninegrid joined	00:16
wereHamster	Mag-en00b: reflog committing is the wrong word. It's simply a file which contains how refs changed. And it includes who updated the refs. See the test of that commit.	00:17
	Mag-en00b: https://github.com/git/git/commit/e32a4581bcbf1cf43cd5069a0d19df07542d612a#L2R236	00:17
	← eijk left	00:17
Mag-en00b	yeah, thanks. i'm reading it	00:17
	→ eijk joined	00:18
	← hotovson left	00:19
	← jdunck left	00:20
	→ juvenal joined	00:20
	→ dlam` joined	00:20
	← Spaceghostc2c left	00:21
	← infogulch left	00:21
	→ BadProphet joined	00:22
alnr	After a push, git-diff is confusing me. Eg, I do: 1) local:(master)$ git merge myfeature. 2) local:(master)$ git checkout myfeature 3) local:(myfeature)$ git push prod myfeature. 4) prod:(master)$ git merge myfeature. 5) local:(master)$ git diff prod/master. Here, prod/master looks pre-merge, but on prod, the working tree correctly reflects step 4). What am I missing, shouldnt the diff match?	00:22
	← picoguy left	00:22
	→ infogulch joined	00:22
	→ infynyxx joined	00:22
	← Anders_J left	00:22
	→ Guest11213 joined	00:23
	← MrNibbles left	00:23
wereHamster	alnr: before 5: git fetch prod;	00:23
	→ hellopat joined	00:24
alnr	wereHamster: ok, i think i fundamentally dont understand why thats needed	00:24
	is the diff just looking at the local copy of prod/master? i thought it was going across the wire	00:25
wereHamster	alnr: because local has no clue at all that you have done something on prod.	00:25
	→ adymitruk joined	00:25
Mag-en00b	thanks. had problems because bare repo didn't record reflog by default, but now i can see the point of those env-vars	00:25
wereHamster	alnr: the only commands that go across the wire are fetch and push. Everything else is local.	00:26
adymitruk	anyone ever get the "fatal: You are on a branch yet to be born"? however, I am on a branch... this is not allowing me to reference a submodule..	00:26
alnr	wereHamster: thanks, i had a faulty understanding there which was really causing problems	00:26
	← thiago left	00:27
	← eijk left	00:28
	← infogulch left	00:29
	→ eijk joined	00:29
	← steveoh left	00:30
	→ gerard0 joined	00:31
	← cakehero left	00:31
	→ infogulch joined	00:31
	→ eijk__ joined	00:31
	→ rideh joined	00:33
	→ simesy joined	00:33
	← simesy left	00:33
alnr	followup, this workflow cant be right. what would be the more efficient way to deploy myfeature, developed on local, to prod?	00:33
	← gtrak left	00:34
	→ simesy joined	00:34
	→ yakko joined	00:34
	← eijk left	00:35
	→ slamice_ joined	00:35
	→ eijk joined	00:35
	← eijk__ left	00:36
	← nkuttler left	00:36
	← benbergman left	00:37
	← slamice left	00:37
	slamice_ → slamice	00:37
wereHamster	alnr: a deployment tool.	00:38
	→ MrNibbles joined	00:39
	→ nkuttler joined	00:39
	← nkuttler left	00:39
	→ nkuttler joined	00:39
	← gusnan left	00:39
	← eijk left	00:40
	→ btree joined	00:40
	← slamice left	00:41
	← bronson left	00:41
	← aruiz left	00:42
	→ slamice joined	00:42
	→ jcao219 joined	00:43
	→ aynik_ joined	00:44
	→ benbergman joined	00:44
	← TomWij left	00:45
	← aynik left	00:45
	→ kvothe joined	00:46
	← kvothe left	00:46
	→ kvothe joined	00:46
	→ sirdancealot8 joined	00:46
	→ TomWij joined	00:46
	← yshh left	00:46
	← aynik_ left	00:47
	← juril left	00:48
	→ dragon3_ joined	00:51
	→ marhaban joined	00:55
	→ kermit joined	00:55
	← lresende left	00:56
	→ Imaginativeone joined	00:56
	→ dermoth\|work joined	00:57
	→ Aikar joined	00:57
	← tinybluepixel left	00:57
	← sirdancealot2 left	00:59
	→ yshh joined	00:59
	← glennpratt left	00:59
	← ejd left	01:00
	← rideh left	01:00
	← Guest7061 left	01:00
	← yakko left	01:00
	→ Febul joined	01:00
	← psoo left	01:00
	← jasonmog left	01:01
	Febul → Guest95221	01:01
	← rchavik left	01:01
	→ yakko joined	01:02
	→ xiangfu joined	01:02
	← quazimodo left	01:02
	→ cbreak\|evil joined	01:02
	→ alunduil joined	01:03
	→ nurey joined	01:03
	← adamben left	01:04
	← cbreak left	01:04
	cbreak\|evil → cbreak	01:04
	→ SBoudrias joined	01:05
	← Imaginativeone left	01:05
	← tewecske left	01:05
	→ angelsl joined	01:08
	→ dragon3__ joined	01:09
	← eft left	01:11
	→ petteyg joined	01:12
	→ hyperair joined	01:12
	← dragon3_ left	01:12
	→ JoeAngel_ joined	01:13
	← apok left	01:13
	→ apok joined	01:13
	← darkclown left	01:14
	→ beavel joined	01:15
	← etcetera left	01:16
	← imrobert left	01:16
	→ DarkClown joined	01:17
	→ lenswipee joined	01:17
	← Mag-en00b left	01:17
lenswipee	hi, can someone tell me an editor that window users use for git?	01:17
	i don't like the default vi editor	01:17
wereHamster	notepad!	01:18
lenswipee	something more advanced than notepad.	01:18
	← JStoker left	01:18
Chani	lenswipee: I hear notepad++ is popular. never used it myself	01:18
wereHamster	visual studio 2013?	01:18
Chani	haha	01:18
	yeah probably	01:18
	but wait.. for git? like, for commit messages?	01:19
	what about tortoiseGit?	01:19
	← OOPMan left	01:19
lenswipee	tortoiseGit has built in editor?	01:19
	→ quazimodo joined	01:19
	→ fisted joined	01:20
Chani	well, tortoiseSvn does. it's not exactly hard to make a text input box	01:20
lenswipee	just need a editor that does two things: show line numbers and highlight code	01:21
	← martinjlowm-lapt left	01:21
	→ GeertB joined	01:21
lenswipee	like gedit, but that doesn't work on git	01:21
wereHamster	lenswipee: have you tried asking in #windows?	01:21
Chani	code?	01:21
	lenswipee: what has code got to do with git?	01:21
	→ LuxuryMode joined	01:21
lenswipee	Chani: thats silly	01:22
Chani	the internet is silly.	01:22
	→ etcetera joined	01:22
lenswipee	for some	01:22
Chani	lenswipee: I'm really not clear on what you want this text editor for.	01:22
lenswipee	Chani: to edit text. what else do you think?	01:23
	→ imrobert joined	01:23
Chani	lenswipee: what kind of text?	01:23
lenswipee	what do you noobs use?	01:23
wereHamster	vim	01:23
lenswipee	lol	01:23
	Chani too	01:23
lenswipee	too slow	01:23
Chani	actually no, I don't give a fuck :) I have video games to play	01:24
wereHamster	Chani: WoW?	01:24
	← TeamColtra left	01:24
Chani	wereHamster: ocarina of time	01:24
	→ OOPMan joined	01:24
wereHamster	Chani: is that like tetris?	01:24
Chani	lol	01:24
wereHamster	pacman?	01:24
	← infynyxx left	01:24
	← ninegrid left	01:25
	Chani just found out that most of the old zelda games are available for wii	01:25
lenswipee	wereHamster: by video games he means porn	01:25
	→ abhatnag joined	01:25
	→ tewecske joined	01:25
	← devth left	01:26
	← Mode-M left	01:26
abhatnag	quick question about git stash: when I go git stash show, the files listed include changes in ALL my stashes, ie from stash@{0} to stash@{10}. The same happens when I go git stash show stash@{0} or any other reference. Is that how it is expected to work?	01:26
	→ TeamColtra joined	01:26
	→ unreal joined	01:26
lenswipee	notepadd++ is the one for me :), thanks	01:27
	→ ninegrid joined	01:27
	→ jacobt joined	01:27
	→ oojacobo_ joined	01:28
	← zivester left	01:29
	← juvenal left	01:29
skorgon	abhatnag, sounds wrong. and behaves differently for me	01:29
abhatnag	skorgon: hmm, how do you create your stashes? I usually go, stash save "comment"	01:30
	← oojacoboo left	01:30
	← sideone left	01:30
	oojacobo_ → oojacoboo	01:30
skorgon	i usually just use 'git stash'	01:30
abhatnag	interesting. As far as my knowledge goes, they both should be the same.	01:30
	← jacobt left	01:31
	→ jfs joined	01:31
EugeneKay	git stash is a hackjob command. I wouldn't be surprised if that was a bug.	01:32
	I recommend creating a temp branch and making a quick commit to that instead	01:32
	This leaves you with something you can push to a central repo and then fetch to any other boxen you have a copy of the repo on	01:32
	← fisted left	01:33
	→ tinybluepixel joined	01:34
	→ fisted joined	01:35
	← lenswipee left	01:35
	→ zivester joined	01:35
	→ mikepack joined	01:35
	fisted → Guest43294	01:35
	← Guest95221 left	01:36
	→ mneorr_ joined	01:36
	← benbergman left	01:37
	← m0ltar left	01:37
	← zivester left	01:37
	→ paolooo joined	01:37
	← BadProphet left	01:38
	← tinybluepixel left	01:39
	→ juvenal joined	01:40
	← kenperkins left	01:41
	← paolooo left	01:42
	→ BadProphet joined	01:42
	← Guest43294 left	01:42
	← dankest left	01:44
	← slamice left	01:44
	← angelsl left	01:45
	→ JStoker joined	01:45
	← LuxuryMode left	01:46
	→ benbergman joined	01:49
	→ fisted_ joined	01:50
	← nurey left	01:50
	← abhatnag left	01:50
	→ slamice joined	01:51
	← ninegrid left	01:52
	→ carbohydrate joined	01:53
	→ rideh joined	01:53
offby1	gasp	01:53
	git stash is the way and the life!	01:54
	→ lresende joined	01:54
psilo_	offby1: yep	01:54
	→ ninegrid joined	01:55
	← kerrick left	01:56
	← AaronMT left	01:57
carbohydrate	hello everyone. does anyone here have experience using git from the command line using eclipse? if so, how well does it work? i'm not really interested in egit, but i'd like to start tracking my development better	01:57
EugeneKay	I HATE SUBMODULES DIE DIE DIE	01:58
Vinnie_win	agreed	01:58
	← hwrd\|work left	01:58
	← vpetro left	01:58
	→ jackh joined	02:00
	← russt left	02:01
	→ luc_ joined	02:01
	→ jdyin joined	02:02
	← alexim left	02:02
	← dlam` left	02:04
	← jfs left	02:07
	← cyphase left	02:08
	← sluther left	02:09
	→ delirium joined	02:09
	→ dankest joined	02:09
Back2Basics	quick terminology question: Pull is upload? push is what?	02:10
EugeneKay	Other way around	02:10
	push copies from your local repo to a remote one	02:11
	fetch grabs from a remote repo to your local one	02:11
Back2Basics	a pull request is ?	02:11
imMute	and pull = fetch + merge	02:11
EugeneKay	pull does a fetch, then merges the current branch with its "upstream" equivalent	02:11
	→ v0n joined	02:11
imMute	Back2Basics: a github thing	02:11
EugeneKay	A Pull Request is a request that somebody else Pulls from you	02:11
Back2Basics	imMute: it's a bad design thing.	02:13
imMute	huh?	02:14
	← ehsan_ left	02:14
Back2Basics	I don't have "fetch" (using the github app) would "sync branch" not issue a pull request?	02:16
	← the_cybersphinx left	02:16
imMute	Back2Basics: github... app??	02:16
Back2Basics	http://mac.github.com/	02:17
imMute	yuck.	02:17
offby1	carbohydrate: I use git with eclips.	02:17
	carbohydrate: I don't use egit; I use the command line.	02:17
	carbohydrate: the only tricky bit (apart from Eclipse being a worthless piece of shite) is that the projects that I use Eclipse with tend to be made by people who don't know anything about revision control, and therefore they check in stuff that shouldn't be checked in ...	02:18
	→ aspotashev joined	02:18
	← Internet13 left	02:18
	← oojacoboo left	02:19
	← tvw left	02:19
	← fisted_ left	02:19
	← Gentlecat left	02:20
	→ m0ltar joined	02:20
	→ nurey joined	02:21
	→ Gentlecat joined	02:22
	← mneorr_ left	02:22
	← ninegrid left	02:22
	→ mneorr joined	02:22
	→ hwrd\|work joined	02:24
	→ ninegrid joined	02:25
	→ cyphase joined	02:26
	← dankest left	02:26
	→ spiderweb joined	02:26
someprimetime	so i just downloaded an entire website from my server so i can work on it locally and then push to github. so now i have a bunch of folders that i want to add and push to my repo on github… well I can add some of the folders but there is one folder that keeps giving me this on `git status` modified: app (modified content, untracked content)	02:27
	← benbergman left	02:29
	← kpreid left	02:29
	→ kpreid joined	02:30
	← Squarism left	02:30
	→ bronson joined	02:30
	→ fisted joined	02:33
	fisted → Guest21723	02:33
	→ kerrick joined	02:34
	← mneorr left	02:34
	← Guest21723 left	02:34
	← undecim_ left	02:35
	→ benbergman joined	02:35
	→ undecim_ joined	02:35
	← todd_dsm left	02:35
	→ dankest joined	02:36
	← kerrick left	02:40
	→ Internet13 joined	02:41
	← nurey left	02:41
	← benbergman left	02:43
	→ simesy_ joined	02:43
	→ fisted_ joined	02:43
	→ benbergman joined	02:43
EugeneKay	!repro	02:43
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session. This will help immensely with troubleshooting.	02:43
	→ kerrick joined	02:44
	← kerrick left	02:45
	→ kerrick joined	02:46
	← simesy left	02:47
	→ rchavik joined	02:47
	→ todd_dsm joined	02:47
	→ hafeez3000 joined	02:48
	← yakko left	02:49
	→ Vampire0_ joined	02:51
	← kerrick left	02:54
	← Vampire0 left	02:54
	→ silves joined	02:56
	← boichev left	02:56
	← JSharpe_ left	02:58
	← Internet13 left	02:58
	→ nurey joined	02:58
	→ tinybluepixel joined	02:58
	← gurps left	02:59
	← marhaban left	03:00
	→ pantsman joined	03:00
	← pantsman left	03:00
	→ pantsman joined	03:00
	← dankest left	03:01
	← simesy_ left	03:01
	→ oojacoboo joined	03:01
	← Gentlecat left	03:01
	→ Internet13 joined	03:02
	→ simesy joined	03:02
	Guest11213 → Spaceghostc2c	03:02
	← silves left	03:03
	→ jacobt joined	03:03
	→ dragon3_ joined	03:04
	← lresende left	03:05
	← oojacoboo left	03:06
	jacobt → oojacoboo	03:06
	→ echo-area joined	03:07
	→ silves joined	03:07
	← dragon3__ left	03:08
	← pantsman left	03:10
	→ russt joined	03:10
	→ ehsan joined	03:11
	→ gtrak joined	03:13
	→ boichev joined	03:13
	← silves left	03:14
	→ cgar joined	03:14
	→ infynyxx joined	03:15
	→ tmcmahon joined	03:17
	← Chibby_ left	03:18
	← cgar left	03:19
	→ jetole joined	03:19
jetole	Hey guys. I'm diving into learning git. I've used it here and there over the years but never on my own project or one I contributed on. Anyways, how do I list all of the branches that exist in a repo?	03:20
	ah never mind. `git branch` does the trick	03:21
	→ gurps joined	03:22
jetole	would I use git branches for project revision numbers?	03:22
	← sjhuang left	03:22
	→ svetlyak40wt joined	03:24
	→ sjhuang joined	03:24
	→ cgar joined	03:25
	← fletch` left	03:25
imMute	jetole: there are plenty of branching "strategys", but tags should probably be used to designate release versions. although, if you have to support / develop more than one version simultaneously, then having a branch-per version would be a good way to go	03:25
	← LekeFly left	03:26
imMute	I kinda like the model described by http://nvie.com/posts/a-successful-git-branching-model/ (except for a few things)	03:26
someprimetime	yo	03:26
	i'm trying to add a folder that had an instance of .git inside of	03:26
	→ Chibby_ joined	03:26
jetole	imMute: I understand about the different means a project can be "versioned" but I mean with my own versioning system, how would I use this with git say if I want to save revision 1 and then sometime later while working on 3.5 I want to go back to revision 1 for review?	03:26
	jetole checks the link	03:27
psilo_	jetole: you checkout the old version, that's all	03:27
imMute	jetole: oh, you mean checking out previous revisions of the project? yeah, you just checkout the old version	03:27
someprimetime	i deleted that .git instance and tried to add it again and i'm getting this: fatal: Path 'app/Config' is in submodule 'app'	03:28
	← svetlyak40wt left	03:28
jetole	psilo_: ok. I am very new to understanding git and I'm not a programmer by profession. from what I understand git saves revisions as hex keys? if I have my revision 1 ready to commit now, how would I commit it so that I could reference it later via the revision I'm using	03:28
imMute	someprimetime: sounds like the first add was adding a submodule. not sure how you'd get git to undo that though. maybe 'git rm --cached app' ??	03:29
	← OOPMan left	03:29
JPeterson	how do i list all unindexed commits?	03:29
imMute	jetole: you can use tags to attach your own "version numbers" to commits.	03:29
	JPeterson: unindexed?	03:29
someprimetime	imMute: you fuckin genius	03:29
	+1!!!	03:29
JPeterson	imMute: commit not in log	03:30
jetole	imMute: and I just pulled up the git books tag page which I'm about to read but I could reference commits later via the tags I assign?	03:30
imMute	JPeterson: probably via the reflog	03:30
	← mikepack left	03:30
psilo_	jetole: what imMute said. alternatively if you don't want to do that yet, just commit it, then `git log` to find the commit hash. you can then checkout that hash	03:30
someprimetime	can i tell git to ignore image files?	03:30
	like when adding files	03:30
psilo_	someprimetime: .gitignore -- this is definitely getting into "read a tutorial" territory	03:31
imMute	jetole: yes, whenever a command wants a commit hash (I forget the wording git uses... commitish?) you can provide a hash, a partial hash, a tag name or a branch name	03:31
	← echo-area left	03:31
jetole	psilo_: I don't know if I want to do that yet. about to read about it and I'm more on the path of learning git instead of how it applies to an actual project at the moment. playing with it now on some demo directories and trying it out	03:31
psilo_	someprimetime: sounds like you are using `git add *` or something, not really a good practice	03:31
jetole	imMute: thanks. I'm going to read this page (http://git-scm.com/book/en/Git-Basics-Tagging)	03:32
	← myx left	03:32
someprimetime	psilo_: i'm adding an entire folder which is my entire website	03:32
jetole	psilo_: thank you too	03:32
someprimetime	problem?	03:32
psilo_	someprimetime: .gitignore	03:32
	→ echo-area joined	03:32
someprimetime	ah so you suggest going in and adding that to each folder i want ignored?	03:32
imMute	someprimetime: no, we suggest you go read the docs on how to use .gitignore	03:33
psilo_	jetole: http://cbx33.github.com/gitt/intro.html This book may be interesting to you, it's rather slow-paced if you already know some versioning concepts but it's good	03:33
	← hafeez3000 left	03:33
jetole	oh that reminds me, if I put a .gitignore in my project and I commit the .gitignore then that means all other developers who clone/checkout the project would be using the same .gitignore on that project?	03:33
psilo_	imMute++	03:33
imMute	jetole: correct.	03:33
jetole	imMute: ah cool. thanks again	03:33
	→ JaredMiller joined	03:33
jetole	psilo_: I'll book mark it but I'm diving in already and I used to use svn a few years ago so I don't want to take it too slow	03:34
psilo_	jetole: yeah probably some other book is better in that case	03:34
imMute	git for svn users?	03:34
JPeterson	why is it so convoluted http://stackoverflow.com/questions/359424/detach-subdirectory-into-separate-git-repository/1591174#1591174 to clear unlogged commits?	03:34
imMute	http://git.or.cz/course/svn.html # that. good for diving right in, but you'll have to "forget" svn to use git properly	03:34
	→ OOPMan joined	03:35
	→ robbyoconnor joined	03:35
imMute	JPeterson: that message is rewriting history - not clearing "unlogged commits"	03:35
JPeterson	my convern is that an unindexed commit might have sensitive data that should be removed	03:35
	← benbergman left	03:35
someprimetime	imMute and psilo_ thanks gitignore looks like what i need to be using	03:36
	do you recommend doing a global gitignore file?	03:36
imMute	JPeterson: if a commit is really not part of any history (not reachable through any branch) then the commit will eventually be deleted via git-gc	03:36
	someprimetime: I use them all the time.	03:36
someprimetime	word! thx	03:36
jetole	imMute: well it has been a few years so I don't think I have too much to forget	03:36
	:-)	03:36
imMute	jetole: righto. also, you may or may not be interested, but reading some about how the git object store works (even if you don't understand it fully) would help you understand how to use git as well.	03:37
	← Back2Basics left	03:37
JPeterson	imMute: so "git commit --amend; git gc" removes the sensitive data that was overwritten with amend?	03:37
	→ jdunck joined	03:38
imMute	JPeterson: not immediately	03:38
	← OOPMan left	03:38
jetole	I'm a systems engineer and lately, mostly, I have been using bzr but I haven't been using for code maintenance as much as on the spot repos for system files I am editing or other aspects where I need to be able to immediately restore but point being I don't use bzr in depth enough there to have to worry about fogetting too much either	03:38
JPeterson	what does?	03:38
	← tmcmahon left	03:38
imMute	JPeterson: the old commit will be held on to until it expires. there are ways to make it expire immediately, but you'd have to read teh git-gc manpages to figure out how	03:38
JPeterson	imMute: what about "git reflog expire --all --expire=now"?	03:38
jetole	imMute: two steps ahead of you. I read a good article about git in Linux Journal a month or two ago which described the details plus the book I referenced above which I think is the stock reference also had some good details about how it worked	03:39
imMute	that'd probably be part of it, sure.	03:39
	← orafu left	03:40
jetole	oh. I hope this isn't off topic but, I read that any copy of the most recent git checkout can be used to rebuild the master server from that same revision as the checkout and I guess this applies to why it's considered distributed development but what I'm wondering is if anyone is familiar with MS Team Foundations VCS (I'm not but my co-workers are) and if it has this same distributed nature where a recent checkout/clone can be ...	03:41
	... used to rebuild a failed/lost/gone master	03:41
imMute	TFS? that's about as far into centralization as you can get..	03:42
	← berserkr1 left	03:42
	→ benbergman joined	03:42
	→ lenswipee joined	03:42
imMute	also, s/checkout/clone/ of your last statement.	03:42
jetole	imMute: meaning a client checkout/clone can't be used to rebuild the master server?	03:43
imMute	with TFS? nope. with git? yes	03:43
jetole	good to know	03:43
imMute	in git (as with every distributed VCS), every repository (or clone) contains the entire history of the project	03:44
jetole	glad you used s/checkout/clone/ and not s/checkout/clone/g since I also used "checkout/clone" later on the same line. lol	03:44
	→ orafu joined	03:44
	← nsadmin left	03:44
jetole	imMute: yeah that's what I read but I was wondering if TFS had that capability. I haven't used TFS and seldom use windows but uh, yeah, really good to know that it doesn't have that feature / capability	03:45
imMute	jetole: if you want my honest opinion about TFS - I'd rather use .zip files than that pile of shit. >.>	03:45
lenswipee	what is the command to show all git commands?	03:45
imMute	lenswipee: all of them? probably isn't one.	03:45
Spaceghostc2c	man git	03:45
gitinfo	the git manpage is available at http://jk.gs/git.html	03:45
imMute	listing all of them wouldn't be of much help anyway	03:46
Spaceghostc2c	Or for me, git<tab><tab>	03:46
	→ ryez joined	03:46
jetole	imMute: also good to know. trying to keep my personal opinion quiet since you never when and where IRC could be logged	03:46
	→ qmanjr5 joined	03:46
jetole	imMute: but I'm not arguing with you at all	03:46
Spaceghostc2c	I do the opposite	03:46
qmanjr5	Would this be the channel to ask a question regarding the Git Bash prompt?	03:46
imMute	qmanjr5: possibly. we wont know until you ask :)	03:47
Spaceghostc2c	IRC is my distributed and federated blog platform. Your hard disks are my mirrors!	03:47
jetole	qmanjr5: you never know until you ask	03:47
lenswipee	imMute: say i'm looking for a command to do something, but i don't know how to use git to show that command.	03:47
imMute	lenswipee: then google "how do I do X"	03:47
jetole	Spaceghostc2c: nicely put	03:47
Spaceghostc2c	jetole: Thanks. :)	03:47
imMute	lenswipee: since the names of the git commands are hardly indicitive of what they do	03:47
Spaceghostc2c	indicative* (Sorry)	03:48
lenswipee	imMute: are you being sarcastic?	03:48
	→ sangi joined	03:48
jetole	qmanjr5: do you know what the action of the command you want is? Try describing it and see if anyone can tell you what the command is	03:48
qmanjr5	Well I'm trying to output the current Git branch and status to my Bash prompt. I have http://pastebin.com/H7ujr1cJ as my .bashrc right now	03:48
imMute	lenswipee: not at all. "reset" does a bunch of different things, depending on waht you feed it.	03:48
	→ glennpratt joined	03:48
	← jcao219 left	03:48
qmanjr5	It's plastered all over the internet as how to do it, but it doesn't work.	03:48
imMute	lenswipee: and in some cases, you might actually want "checkout" to do something that you'd think "reset" would cover.	03:48
qmanjr5	Or at least for me. :P	03:49
jetole	I think I'm going to git clone git for my demo dir of learning git :-D	03:50
	after reading "The Git source repo, for instance, contains more than 240 tags"	03:50
lenswipee	imMute: i think it would be useful to learn how to use git in CLI to do what you want, and not have to rely on the internet for the manual or google.	03:50
	→ h0bbit joined	03:50
	→ jcao219 joined	03:51
imMute	jetole: looking at the git or linux repos would probably be a really good indication of how git is used :)	03:51
lenswipee	imMute: to get git to list the commands so that you can then select one of them for further inspection.	03:51
imMute	qmanjr5: what doesn't work on it?	03:51
	→ mikepack joined	03:51
jetole	I agree but I'm trying to find the git git repo. lol. I'm looking at http://git-scm.com/download/linux now	03:51
imMute	lenswipee: the manual is on your hard drive. `man git`	03:51
gitinfo	lenswipee: the git manpage is available at http://jk.gs/git.html	03:51
	→ R_Macy joined	03:51
	→ cakehero joined	03:52
imMute	jetole: https://github.com/git/git	03:52
	→ jackh_ joined	03:52
qmanjr5	imMute: When I go into a repo, nothing is displayed in my prompt.	03:52
jetole	I was just about to say found it @ https://github.com/git/git.git	03:52
	thanks imMute	03:52
qmanjr5	imMute: Or rather, nothing relating to Git.	03:52
	← jackh left	03:52
jetole	wait why does mine have a .git at the end and your doesn't	03:52
	?	03:52
	nevermind	03:53
	jetole slaps forehead	03:53
psilo_	I like how more tokens in that url are git than not	03:53
lenswipee	imMute: how to launch the manual from git CLI?	03:53
	→ jnewt joined	03:53
psilo_	git git git://git.com/git/git.git	03:53
	← infynyxx left	03:53
	← kukks left	03:53
jetole	lol	03:53
imMute	qmanjr5: it's probably because $PS1 isn't reevaluated every time it needs to print the prompt. so it ran __git_ps1 when you sourced the .bashrc but never runs it again	03:53
qmanjr5	Ohhh	03:54
imMute	qmanjr5: you need to use bash's prompt_command to re-create your PS1 every time it does a new prompt	03:54
jetole	imMute and qmanjr5: There is a way to do that with PS1. I don't remember how but it was a ticky var to write	03:54
qmanjr5	jetole: It's already written.	03:54
	imMute: I see. Thank you.	03:54
jetole	I'm trying to remember the bash wiki I read it on. woo something but #bash would be a good place to ask	03:54
imMute	jetole: really? every dynamic prompt I've seen has always used prompt_command	03:55
jetole	qmanjr5: I'm refering to having PS1 run a command on each instance it's displayed and not just when .bashrc is run	03:55
	→ chuckharmston joined	03:55
jnewt	repo is up to about 2gb, and git checkout . is not working anymore (well maybe it is, but for the past 10 mins it's just been hanging there). have i reached a limit of some sort? seemes to me i remember someone doing 250g or even 500g somewhere.	03:55
qmanjr5	jetole: This is the PS1 variable.	03:55
jetole	imMute: I don't know what you just meant but I ran into this issue before with a PS1 I wanted and I solved it and it used an out of the usual method for specifying the command within the variable	03:55
imMute	jnewt: uh, what are you trying to do with 'git checkout .' ?	03:56
	← JEEB left	03:57
jetole	hmmm. according to something I just read it's as simple as using $(command) in the variable but I thought I recalled that what you said was true where the typical way you specify it only executed it once	03:57
	→ echo-are` joined	03:57
jetole	oh I remember I was trying to have bash history pushed to the file on every command run	03:57
	this is off topic so I'm going to shut up	03:58
imMute	jetole: ah, well thats likely controlled by a different variable..	03:58
	← echo-area left	03:58
	→ kerrick joined	03:59
jnewt	imMute: I'm trying to take over the world...thanks for asking.	03:59
jetole	no it was PS1 as it was said (I don't know if it's true) but it's the only means to have a command run with every other command that's run on the system i.e. after you hit enter after each command	03:59
psilo_	jnewt: what is the transport protocol?	03:59
imMute	jetole: that sentence kinda makes sense... are you sure you aren't talking about prompt_command?	04:00
	← echo-are` left	04:00
jetole	imMute: I don't believe so but it was quite a while ago I did it and I don't recall	04:00
jnewt	psilo_: ssh	04:00
	→ echo-area joined	04:00
	jetole just pulled up PROMPT_COMMAND in `man bash`.	04:01
	← m0ltar left	04:01
psilo_	jnewt: 10 mins isn't too long really	04:01
jetole	imMute: I don't think it was but that looks interesting and I didn't realize it was there	04:01
psilo_	jnewt: maybe confirm that something is still happening, and be patient	04:01
lenswipee	does the '--' indicate it is an argument passed in?	04:01
jnewt	psilo_: lock file is still there, so i think something is happening...	04:02
	→ ergZay joined	04:02
	→ michaelmartinez joined	04:02
imMute	lenswipee: the what on what?	04:03
jnewt	2gb just doesn't seem that big. i can transfer 2gb on our internal network in no time, def not 10 min	04:03
lenswipee	imMute: I think it is a switch to indicate what options are turned on.	04:03
imMute	lenswipee: example?	04:04
lenswipee	imMute: 'git config --help'	04:04
jnewt	ignore my poor use of units	04:04
psilo_	jnewt: oh I didn't see that it was internal.	04:04
	← juvenal left	04:04
imMute	lenswipee: '--foo' is a fairly common way of passing "switches" to CLI programs...	04:04
lenswipee	imMute: applies to UNIX commands?	04:05
jnewt	psilo_: server is across the room from me, so, not internal to my pc, but internal to my lan	04:05
imMute	lenswipee: yes.	04:05
	→ tinyblue1ixel joined	04:05
	← tinybluepixel left	04:07
	← hwrd\|work left	04:08
jnewt	ok, it finished and i started a push, i'm getting 1.51MiB/s	04:09
	second thought, checkout is on the index, so it was local to the pc	04:09
	← JaredMiller left	04:09
	← xiangfu left	04:10
	← nurey left	04:10
lenswipee	how to output the contents of global config in git bash?	04:11
imMute	man git-config	04:11
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	04:11
	← undecim_ left	04:11
	→ undecim joined	04:12
	→ edgar-rft joined	04:12
jnewt	i'm down under 300 KiB/s any way to see which file it's working on? wonder if i have a binary in there somewhere that's killing it	04:12
	→ apok_ joined	04:13
	← apok_ left	04:13
lenswipee	'git config --edit global' <-- failed	04:13
	→ hwrd\|work joined	04:14
imMute	lenswipee: did you even read the manpage?	04:14
	or did you even open it?	04:15
jetole	does anyone know where I can read what the git version tags mean on the git source? i.e. is it major.minor.bugfix.build (probably not) ?	04:15
	← eto left	04:15
lenswipee	'git config --edit --global'	04:15
imMute	lenswipee: you said you wanted a list	04:15
	← apok left	04:15
	→ eto joined	04:16
	← mikepack left	04:16
lenswipee	lenswipee: i wanted to echo contents of global config into gitbash	04:16
	that's the closest command that i can see	04:16
	→ revyd joined	04:16
imMute	lenswipee: look one line up from --edit and you'll see --list	04:17
	← kvothe left	04:17
lenswipee	C:\Program Files\Git\doc\git\html\git-config.html	04:17
	i see no list	04:17
	nevermind, i'll try it	04:18
	← simesy left	04:18
imMute	I sincerely doubt that "--list" doesn't appear in that file at all. It might not be exactly one line up from --edit, but it's definitely there.	04:19
lenswipee	actually, it says -l	04:20
	thats works	04:20
	cheers	04:20
	imMute facepalms	04:20
	← tavist0_ left	04:21
	← cakehero left	04:21
	→ irqq joined	04:21
	→ rurufufuss joined	04:22
	← FabTG left	04:22
	← tjbiddle left	04:22
	← eto left	04:22
	→ eto joined	04:24
	← Cassiel left	04:25
	→ Back2Basics joined	04:25
lenswipee	how to use git to get a specific file in the last commit in text editor?	04:26
imMute	pretty sure git doesn't care about the text editor part..	04:26
lenswipee	hmm	04:27
	→ aruiz joined	04:27
	← etcetera left	04:27
lenswipee	text editor is usually where I look at code.	04:27
	← benbergman left	04:27
lenswipee	any ideas imMute?	04:28
	← R_Macy left	04:28
	→ FabTG joined	04:29
	→ Quix0te joined	04:31
	→ benbergman joined	04:32
	→ yarco joined	04:32
milki	cat-file?	04:33
imMute	lenswipee: git show HEAD~:<filename>	04:33
	to see what <filename> looked like in the previous commit	04:33
yarco	how to list only annotated tag in a good format? like	04:34
	→ OOPMan joined	04:34
yarco	v0.0.3 \n messages \n	04:34
	v0.0.2 \n messages \n	04:34
	?	04:34
offby1	well, you can _identify_ annotated tags with "git for-each-ref"	04:35
	the second column will be "tag" instead of "commit"	04:35
lenswipee	imMute: what does '~:' denote?	04:35
	→ Back2Basics_ joined	04:35
offby1	then you can grab the sha1 from the first column, and do whatever you want with it -- presumably send it to "git log" or "git show" with a suitable --format option	04:35
	→ infynyxx joined	04:36
yarco	so that's the way they creating release notes and change log?	04:36
imMute	lenswipee: "HEAD~" means "one commit before HEAD	04:36
offby1	yarco: I Doubt it :-) There's probably a simpler way	04:36
imMute	yarco: my CHANGELOG is "see 'git log'"	04:36
lenswipee	imMute: I should have said what 'HEAD' denote?	04:36
yarco	but i think release notes is necessary to tell user what new features included.	04:37
imMute	lenswipee: simply? whatever the current branch is	04:37
	→ Back2Basics__ joined	04:37
imMute	yarco: and you think useful release notes can be automatically generated?	04:38
yarco	yes	04:38
	i think it can be written into annotated tag	04:38
lenswipee	imMute: I want the current commit not "one commit before HEAD".	04:38
imMute	lenswipee: then do just HEAD:<filename> duh	04:38
yarco	if git can show only annotated tag with message	04:38
	← undecim left	04:39
	← Back2Basics left	04:39
	Back2Basics__ → Back2Basics	04:39
lenswipee	imMute: to confirm thats the latest commit in the repos?	04:39
yarco	git tag -a v0.0.1 -m '…release…notes...'	04:39
	→ undecim joined	04:39
imMute	lenswipee: "latest"? no.	04:39
yarco	git <some command> > RELEASE	04:40
	→ nurey joined	04:40
	← spiderweb left	04:40
	← infynyxx left	04:41
	→ private joined	04:41
	← Back2Basics_ left	04:41
	private → Guest42778	04:41
	→ spiderweb joined	04:42
	← jnewt left	04:42
	← someprimetime left	04:43
	→ tavist0 joined	04:43
	→ someprimetime joined	04:43
	← Back2Basics left	04:45
	← jso-` left	04:46
lenswipee	imMute: from what i read HEAD means current branch.	04:46
	imMute: so HEAD~ can't mean current.	04:47
	← rchavik left	04:47
	→ andrew__ joined	04:48
	andrew__ → andrewhl	04:48
	→ Back2Basics joined	04:48
	→ simesy joined	04:48
imMute	jesus	04:49
	I spent 10 minutes trying to use for-each-ref and git show to only get the annotated messages	04:49
	turns out ' git tag -l -n' is enough	04:49
	yarco: so there you go :)	04:49
	← Guest42778 left	04:49
imMute	lenswipee: yes, HEAD is current.	04:50
lenswipee	imMute: which means it's the lastest commit	04:50
imMute	not necessarily	04:50
	← vsayer left	04:50
imMute	someone could have committed after you did, but in a different branch	04:50
	← joeytwiddle left	04:51
lenswipee	imMute: then HEAD would point to that.	04:51
imMute	no it wouldn't	04:51
	← codertux left	04:51
lenswipee	imMute: it should update	04:52
imMute	no it shouldn't. HEAD points to the currently checked out branch. if someone commits to a different branch, then HEAD wouldn't change	04:52
lenswipee	imMute: how is that 'cuurent'?	04:52
	→ rchavik joined	04:52
	← rchavik left	04:52
	→ rchavik joined	04:52
	← simesy left	04:53
	→ lresende joined	04:53
	→ simesy joined	04:53
lenswipee	each time you commit you create a new branch correct?	04:54
imMute	no	04:54
lenswipee	oh gawd	04:55
imMute	do you do any C programming? a branch is merely a pointer to a commit. when you make a new commit, you just update that pointer to point at a new commit object	04:55
	← glennpratt left	04:55
lenswipee	you update pointer or create a new one?	04:56
	→ Praise- joined	04:56
imMute	update.	04:56
	creating a new would be creating a new branch	04:56
	← nurey left	04:56
lenswipee	if you update pointer then you lose old data	04:56
imMute	no you don't	04:56
lenswipee	the pointer points to a new object	04:57
imMute	because the new commit object has a pointer to the parent commit	04:57
	← benbergman left	04:57
	← Praise left	04:57
	→ codertux joined	04:57
lenswipee	parent commit?	04:58
	← aetcore left	04:58
lenswipee	previous commit?	04:58
imMute	yes (in the case of merge commits - there will be 2 or more parent commits)	04:58
lenswipee	and i assume the parent commit has pointers to its parents as well?	04:59
	and so on and so forth	05:00
imMute	obviously.	05:00
	→ zheng joined	05:00
lenswipee	each user has their own branch?	05:01
imMute	yes, since it's just a pointer	05:01
lenswipee	so HEAD~ <-- gets the most extended branch?	05:02
imMute	no. HEAD~ is the commit before HEAD	05:02
	→ longbeach joined	05:02
	← zheng left	05:03
	← aspotashev left	05:03
	→ r0bby joined	05:04
lenswipee	the word 'before' doesn't make it sound like the most recent commit.	05:05
imMute	here we go again.	05:05
	→ benbergman joined	05:05
lenswipee	this is very confusing	05:05
imMute	maybe it's just you...	05:06
	← robbyoconnor left	05:06
lenswipee	when you commit you create a blob correct?	05:06
imMute	no.	05:06
	← simesy left	05:07
	← irqq left	05:07
lenswipee	when is it created?	05:08
imMute	blob objects store files. commit objects store commits	05:08
	http://git-scm.com/book/en/Git-Internals-Git-Objects	05:08
hyperair	when you add a file to the index, a blob is created for the file, and its hash is referenced in the index.	05:09
lenswipee	hyperair: what's in the blob?	05:09
imMute	the file	05:09
hyperair	lenswipee: the contents of the file.	05:10
lenswipee	that's it?	05:10
imMute	thats it	05:10
	→ nurey joined	05:10
imMute	read that link I just posted and you'd know that	05:10
	oh wait, you like "learning" by listening to other people, not reading.	05:10
	← ergZay left	05:11
lenswipee	not exactly by listening.	05:11
hyperair	simple solution: copypaste the contents of the article, and prefix every line with imMute:	05:11
	← redrocket left	05:11
hyperair	or <imMute> to be exact	05:11
	\o/	05:11
imMute	hyperair: you mean lenswipee ?	05:12
	hyperair: and actually, that's not a bad idea :/	05:12
hyperair	imMute: nah (s)he likes listening to you.	05:12
	heheheh	05:12
lenswipee	i think imMute likes to confuse me by giving the most simplistic vague responses to my questions.	05:13
imMute	where's that face-stabbing-over-ip RFC when you need it	05:13
	→ pretty_function joined	05:13
	→ zheng joined	05:13
hyperair	imMute: there's an RFC like that?	05:13
imMute	hyperair: I wish.	05:13
	→ redrocket joined	05:14
hyperair	http://www.urbandictionary.com/define.php?term=RFC the first definition has a mention of it.	05:14
lenswipee	hyperair was actually being serious sheesh	05:14
hyperair	"Someone needs to draft an RFC for a stab-idiot-IRC-users-in-the-face protocol.	05:14
	"	05:14
	→ ergZay joined	05:14
imMute	lenswipee: seriously. go read some articles. you'll learn much more, and faster, than going back and forth with people on IRC	05:15
hyperair	read the articles and come back with questions.	05:15
	← JoeAngel_ left	05:15
lenswipee	imMute: particularly with you. You like to drag people through mud first.	05:16
	→ JoeAngel_ joined	05:16
pdtpatrick	lenswipee: https://peepcode.com/products/git	05:16
	← benbergman left	05:16
imMute	lenswipee: no, I (as with many other people) don't like teaching the basics that are easily self taught	05:16
lenswipee	imMute: I think you should just speak for yourself ;)	05:17
	going to read	05:17
	brb	05:17
imMute	lenswipee: for some anecdotal evidence, notice how, 45 minutes ago, offby1 helped yarco with his question, but has been ignoring you all night. that should tell you something.	05:18
	← unnamedwill left	05:19
	→ joshsmith joined	05:19
	← SBoudrias left	05:20
	← hwrd\|work left	05:20
lenswipee	imMute: why? because offby1 represents everyone here.	05:20
imMute	nope. never said he did.	05:20
	← v0n left	05:21
	→ simesy joined	05:21
	→ macmartine joined	05:21
lenswipee	imMute: I think you should have taken what I said eariler and ignore me if you can't be bothered to help a beginner at git.	05:21
imMute	ignoring you for me, is as hard as it is for you to RTFM.	05:22
	← h0bbit left	05:22
	→ koltroll joined	05:22
lenswipee	bloody germans	05:23
	laterz	05:23
	→ drewery joined	05:23
	→ v0n joined	05:24
	→ benbergman joined	05:24
	← simesy left	05:26
	← jcao219 left	05:27
	→ simesy joined	05:27
	← someprimetime left	05:28
	→ someprimetime joined	05:28
	→ a_L_p joined	05:32
lenswipee	not many spoon feeders around these days ... what a shame.	05:33
pdtpatrick	you'll be alright	05:34
	← iamjarvo left	05:34
	→ infynyxx joined	05:35
	→ mikepack joined	05:35
	lenswipee is sinking in quicksand, and suddenly hears a whisper of a sound from above that sounded like "you'll be alright".	05:37
	← infynyxx left	05:37
	← andrewhl left	05:38
	→ hwrd\|work joined	05:39
	← hwrd\|work left	05:40
pdtpatrick	lenswipee: there's plenty of videos out there on Git or documents. For instance, a quick search on youtube would give you plenty of things to get started on: http://www.youtube.com/watch?v=74yphulj3uo&list=PL5B3829DDD57E9C23&feature=plcp	05:41
	→ fly9 joined	05:41
	← v0n left	05:42
	← Davey left	05:42
pdtpatrick	there's also stackoverflow which answers a lot of random questions people have asked that you'll have now	05:42
	so if ur drowning, then it's probably because you want to be there not because you don't have the ability to get out.	05:42
	← BadProphet left	05:43
lenswipee	pdtpatrick: when i commit a file does the stage ref gets replaced with a commit ref, and the whole index gets backed up?	05:44
	← koltroll left	05:44
	← fly9 left	05:46
pdtpatrick	lenswipee: read this before u ask any other questions .. http://git-scm.com/documentation	05:46
	← yarco left	05:47
	← benbergman left	05:47
	← simesy left	05:48
lenswipee	pdtpatrick: read it all yesterday. still confused.	05:48
pdtpatrick	my guess is as u read, u didn't try it out huh? ur gonna need to try out what is being said so you can see it work and you'll understand what's happening	05:49
	so if ur expecting to just readi t all and then become an expert, then you're up for disappointment and frustration	05:50
	← madewokherd left	05:50
pdtpatrick	so take ur time, go through each chapter, test it out -- create projects along the way. Force yourself to incorporate it into your workflow, eventually you'll see and understand why some things happen the way they do.	05:51
	then you can start asking better questions and get better answers	05:51
	just my $.02	05:51
	← maroach left	05:52
lenswipee	why can't you answer my complex question?	05:53
	pwhen i commit a file does the stage ref gets replaced with a commit ref, and the whole index gets backed up?	05:53
pdtpatrick	okay let's flip this around	05:53
	what does staging area mean to you ? what exists there ?	05:53
lenswipee	files added to the index	05:54
	next question	05:54
	→ benbergman joined	05:54
	→ the_cybersphinx joined	05:54
pdtpatrick	so what happens to the staging area when u commit ?	05:54
	← axl_ left	05:55
lenswipee	a new commit ref points to same object	05:55
kpreid	commits and refs are two different things	05:56
	a new commit is _created_ and an existing ref is changed to point to the new commit	05:56
lenswipee	thats what i said	05:56
	← macmartine left	05:57
	← linguini left	05:57
	→ laurav joined	05:57
	→ x82_nicole joined	05:58
pdtpatrick	maybe that's what u wanted to say but didn't sound like what he said. Albeit, play nice - he's trying to help you :)	05:58
	read this and see if it makes sense: http://gitready.com/beginner/2009/01/18/the-staging-area.html	05:59
	And here's the doc on references that is pretty clear	06:00
	http://git-scm.com/book/en/Git-Internals-Git-References	06:00
lenswipee	"a new commit" as in a another blob	06:00
	→ pucks joined	06:00
	→ v0n joined	06:00
pdtpatrick	What is a "blob" to you ?	06:00
lenswipee	contents of file	06:00
	← gerard0 left	06:01
lenswipee	doesn't make sense to create another blob	06:01
	← laurav left	06:01
hyperair	lenswipee: contents of files go into blob objects. directories are represented by tree objects, which reference blobs and other trees (directories contain files and subdirectories, see). and a commit references the top-level tree.	06:02
lenswipee	start again	06:02
kpreid	any given commit refers to many blobs, not just one	06:02
hyperair	lenswipee: and a ref (e.g. a branch name -- refs/heads/foo) is just a pointer to a commit.	06:03
	→ o] joined	06:03
o]	how can I stash a single file?	06:03
	"Please, commit your changes or stash them before you can merge."	06:03
pdtpatrick	o]: http://stackoverflow.com/questions/3040833/how-to-stash-only-one-file-out-of-multiple-files-that-have-changed	06:04
o]	I am doing a pull and there is a file that was changed locally, so I want to stash it before pulling to not lose that local changes	06:04
hyperair	o]: git stash --patch	06:04
	i'm not sure if you can stash a single file	06:04
	→ jcao219 joined	06:05
lenswipee	does the existing ref to deleted?	06:05
pdtpatrick	well i think if u add everything u want, then u can stash whatever is left	06:05
lenswipee	does the existing ref gets deleted?	06:05
hyperair	o]: add the files you don't want to stash into the index, then git stash --keep-index.	06:05
pdtpatrick	lenswipee: how does git log work ?	06:06
	← michaelmartinez left	06:06
lenswipee	kpreid> a new commit is _created_ and an existing ref is changed to point to the new commit <-- sounds wrong	06:07
o]	nice! keep-index worked	06:07
	thanks	06:07
pdtpatrick	lenswipee: why is it wrong?	06:07
hyperair	lenswipee: a ref is like a pointer or symbolic link.	06:07
	and commits are like linked lists (dag, really)	06:08
lenswipee	for one file in the staging area does " a new commit is _created" mean a new blob is created/	06:08
hyperair	the blob is created when you add the file into the staging area.	06:09
	when the commit is created, the hash of the tree object in the staging area goes into the commit.	06:09
	→ xiangfu joined	06:09
	← xiangfu left	06:09
	→ xiangfu joined	06:10
lenswipee	hyperair: as i thought the existing ref simply gets renamed.	06:10
	← mikepack left	06:11
pdtpatrick	becareful with ur wording "renamed"	06:11
hyperair	lenswipee: no, a ref is like a symlink.	06:11
lenswipee	hyperair: why create another pointer to the same object?	06:11
hyperair	lenswipee: a ref isn't an object.	06:12
	it's a reference to an object	06:12
	lenswipee: are you familiar with symbolic links?	06:12
	← tinyblue1ixel left	06:12
lenswipee	i get that it's a reference	06:12
hyperair	it's just a pretty name, e.g. "master"	06:12
	isn't "master" easier to remember than a hash, e.g. "1238719789afefaefeaf"?	06:12
	→ sluther joined	06:13
	→ Gentlecat joined	06:13
lenswipee	ah	06:13
hyperair	"master" is the name of the default branch you get when you create a git repository.	06:13
	it's a ref.	06:13
	branches are refs.	06:13
	the name "master" points to the latest commit in the branch.	06:13
	that's all there is to it.	06:13
lenswipee	hyperair: so the existing ref stays as in	06:13
	← benbergman left	06:14
hyperair	when you make a new commit, say "deadbeef000000000000" (i didn't count the number of characters)	06:14
lenswipee	hyperair: so a commit just takes a "snapshot" of the index, and a ref to that snapshot is created.	06:14
hyperair	in the "master" branch, then "deadbeef000000000000" has a parent "1238719789afefaefeaf"	06:14
	and "master" now points to "deadbeef000000000000"	06:14
	in fact, .git/refs/heads/master is just a normal text file	06:15
	open it and check what's inside	06:15
lenswipee	hyperair: read my last comment.	06:15
hyperair	a commit is a snapshot of the index, with other metadata like dates, author information, parent(s) (previous commit(s))	06:15
	and the ref is already there	06:16
	it's just changed to point at the new commit.	06:16
	for example, when you create a new repository...	06:16
	$ git init	06:16
	→ gerard0 joined	06:16
	→ skul joined	06:17
hyperair	then your repository HEAD (which is also a ref) points to the "master" ref	06:17
lenswipee	hyperair: my warped understanding was that the existing ref to a stage object was changed when you commit also.	06:17
hyperair	$ cat .git/HEAD --> ref: refs/heads/master	06:17
	$ cat .git/refs/heads/master --> no such file or directory	06:17
	lenswipee: there's no ref to a stage object	06:18
	i think you're being very confused.	06:18
	← TeamColtra left	06:18
hyperair	did you read any of the articles linked earlier?	06:18
lenswipee	hyperair: when you add a new file it's added to the index.	06:18
	hyperair: i'm refering to that ref	06:19
hyperair	okay, don't call that a "ref"	06:19
	"ref" refers to something else.	06:19
lenswipee	hyperair: what do i call it	06:19
hyperair	erm	06:19
lenswipee	hyperair: staging pointer	06:19
hyperair	there isn't a specific term for it	06:19
	specifically, the stage object contains the hash of the blob object.	06:19
	that's all there is to it	06:20
	→ ampz joined	06:20
	← ag4ve left	06:20
	← cenuij left	06:20
	→ jiunec joined	06:20
	→ _dzonder joined	06:20
lenswipee	hyperair: alright. Now HEAD refers to current users current branch correct?	06:21
	→ benbergman joined	06:21
hyperair	yes.	06:21
	→ gmarseliss joined	06:21
lenswipee	hyperair: and current branch is what I have most recently committed?	06:21
hyperair	no, the current branch is what you have most recently checked out.	06:22
	"git checkout $branch"	06:22
	← gerard0 left	06:22
	→ ag4ve joined	06:22
	→ geonetix` joined	06:22
lenswipee	hyperair: depends on the type of check out. what check out you mean?	06:22
hyperair	14:22:22 <hyperair> "git checkout $branch"	06:22
	→ adhawkinz joined	06:23
lenswipee	hyperair: right, and that means?	06:23
	→ Villadelfia_ joined	06:23
lenswipee	hyperair: stuff in the staging area	06:23
hyperair	...do you know how to use git?	06:23
lenswipee	hyperair: not fully	06:23
hyperair	do you know what git checkout does?	06:24
	→ laurav joined	06:24
	→ no_maam_ joined	06:24
	← daniel-s left	06:24
	→ daniel-s joined	06:24
	← heinrich5991 left	06:24
	← adhawkins-away left	06:24
lenswipee	hyperair: i would be teaching git and not asking help questions if i knew it properly	06:24
	← Villadelfia left	06:24
	← dzonder left	06:24
	← project2501b left	06:24
	Villadelfia_ → Villadelfia	06:24
	geonetix` → geonetix	06:24
hyperair	okay, stop.	06:24
	this information that you're asking is advanced information	06:24
	learn how to use git.	06:24
	and THEN learn this stuff.	06:24
lenswipee	hyperair: here is my understand.	06:25
hyperair	git checkout is one of the most basic things.	06:25
	if you don't know how to use git checkout, you don't know how to use git.	06:25
	← chuckharmston left	06:25
lenswipee	hyperair: it puts the files you last committed in your working tree	06:25
pdtpatrick	lenswipee: question - what do u do? I'm trying to gauge ur experience level.	06:25
	→ chuckharmston joined	06:25
lenswipee	pdtpatrick: i deliver pizza.	06:26
pdtpatrick	then my suggestion is start small	06:26
	seems like ur trying to learn everything overnight	06:26
	← quazimodo left	06:26
	→ daniel-s_ joined	06:26
lenswipee	hyperair: it puts the files you last committed in your working tree. correct?	06:26
hyperair	lenswipee: no, it switches branches	06:26
lenswipee	pdtpatrick: im a sponge. i can take it.	06:27
	← pucks left	06:27
pdtpatrick	read the docs, watch the videos. Work on a few projects with git and when u run into something u want to do but don't know the command. Then ask how to do it and why (after you've searched the internet that is)	06:27
	→ r0bby_ joined	06:27
hyperair	rather, git checkout does multiple things. it takes the tree from the commit or branch specified on the command line and puts it in the index. then it expands all of that out onto your working tree.	06:27
lenswipee	hyperair: does a commit create a new branch?	06:28
	r0bby_ → robbyoconnor	06:28
pdtpatrick	lenswipee: if u haven't noticed by now ur not getting it then i dunno what to tell u. Notice how everyone is repeating the same thing to you. It's not going to make sense right now. Just pace yourself and learn the basics	06:28
hyperair	lenswipee: no.	06:28
	lenswipee: there's an order to the things you have to learn to understand git.	06:28
lenswipee	hyperair: when is a new branch created?	06:28
hyperair	lenswipee: before one understands how a car works, one needs to understand what a car does.	06:28
	lenswipee: you're trying to undersatnd how a car works without understanding what it does.	06:29
	← daniel-s left	06:29
hyperair	my suggestion is to read the git tutorial	06:29
	← chuckharmston left	06:29
hyperair	git help tutorial	06:29
lenswipee	hyperair: i deliver pizza. i think i know how a car works.	06:29
hyperair	hmm wasn't there a bot that spat out links when you did that?	06:29
	← xiangfu left	06:29
hyperair	lenswipee: i'm sure you do. but you don't fully understand what git does yet.	06:30
lenswipee	hyperair: the branch concept is the hardest part.	06:30
hyperair	lenswipee: and yet you're trying to understand how git works.	06:30
	→ heinrich5991 joined	06:30
pdtpatrick	lenswipee: question - in your understanding - what's a symbolic link ?	06:30
	← r0bby left	06:31
lenswipee	pdtpatrick: a pointer	06:32
	hyperair: it puts the files you last committed in your working tree, and updates to index to that branch. correct?	06:32
pdtpatrick	let's say i have file A. I create a sym link to A called file B. If i delete file B - what happens to file A ?	06:33
	→ TeamColtra joined	06:33
lenswipee	nothing	06:33
	← qmanjr5 left	06:33
pdtpatrick	what happens if move file A.. what happens to B ?	06:33
	← boichev left	06:33
lenswipee	nothing	06:34
pdtpatrick	see that's what hyperair was talking about - u need to learn the basics	06:34
lenswipee	pdtpatrick: maybe u need to learn pointers, not me.	06:35
pdtpatrick	maybe i do :) - if u care to educate me, i'm all ears.	06:36
lenswipee	lol	06:36
	actually fileB points to new location of fileA	06:36
pdtpatrick	keep trying	06:37
	→ aruiz_ joined	06:37
lenswipee	you tell me then.	06:37
pdtpatrick	how about u try it on ur unix box if u have one	06:37
	← Arokh left	06:37
	→ infynyxx joined	06:37
lenswipee	typical	06:38
	give me back hyperair	06:39
	← infynyxx left	06:39
pdtpatrick	touch fileA && echo "blah" > fileA; ln -s fileA fileB ; rm fileA ; cat fileB	06:39
	what happens?	06:39
hyperair	lenswipee: unfortunately i'm supposed to be working.	06:39
lenswipee	what is touch?	06:39
	hyperair: it puts the files you last committed in your working tree, and updates to index to that branch. correct?	06:40
	→ maleknet joined	06:40
hyperair	errrr no.	06:40
	→ ebouchut_ joined	06:41
pdtpatrick	u've got a lot of practice/reading to do. I'm gonna get back to my scripting - goodluck soaking everything into your porous sponge :)	06:41
lenswipee	hyperair: for a single user on a project, there is usually only one branch in the tree?	06:41
	→ hotovson_ joined	06:41
	← rideh left	06:41
lenswipee	pdtpatrick: why ask questions when you're not prepared to give answers?	06:42
	→ fly9 joined	06:42
hyperair	lenswipee: nope, a typical use case is to have topic branches -- one branch for every pending feature.	06:42
	→ Plouj- joined	06:43
lenswipee	pdtpatrick: ok, how is a new branch usually created?	06:43
pdtpatrick	lenswipee: its a simple question and if u cannot answer that. How do u expect to understand references and symbolink in git?	06:43
	← benbergman left	06:43
pdtpatrick	the simple answer is if u delete A and try to cat the contents .. u'll get an error that the file does not exist	06:44
lenswipee	pdtpatrick: references/pointers/symbolinks are the same thing.	06:44
pdtpatrick	as B is a sym link to A which no longer exists	06:44
lenswipee	the answer to your second question.	06:44
	→ BadProphet joined	06:44
	← ebouchut_ left	06:44
lenswipee	pdtpatrick: you will sleep better tonight if you help me than to work on your script.	06:45
	→ JohnBat26 joined	06:45
	→ soee joined	06:45
	→ shruggar joined	06:46
	← Plouj left	06:46
	hyperair mumbles i'd probably not get any sleep tonight if i don't work now.	06:46
	← fly9 left	06:46
	→ ebouchut_ joined	06:47
	→ boichev joined	06:47
lenswipee	lol	06:47
	→ liori_ joined	06:48
lenswipee	pdtpatrick: still waiting on the answer to your second question.	06:48
	→ axiom joined	06:48
pdtpatrick	so the fact that u didn't understand that shows u've not spent time learning the basics and u'd rather everyone spend their time in here catering to your needs when u've not spent the time yourself.	06:48
	remember - a closed mouth can't be fed.	06:48
	so learn the basics and then u can ask better questions that would enable u to ask even better questions. Progression	06:48
	← majoh left	06:48
	← chee left	06:48
	→ dc5ala joined	06:48
	→ simesy joined	06:48
	← daniel-s_ left	06:48
	→ chee joined	06:49
lenswipee	<pdtpatrick> what happens if move file A.. what happens to B ?	06:49
pdtpatrick	maybe you should stick to delivering pizza (And I mean that in the nicest way possible).	06:49
lenswipee	no you don't	06:49
	← nurey left	06:50
	→ benbergman joined	06:50
lenswipee	you just like to answer questions with questions.	06:50
	and not give any answers	06:50
	→ daniel-s_ joined	06:50
lenswipee	then end the conversation with "...stick to delivering pizza..."	06:51
	← Goplat left	06:51
lenswipee	thats not good teaching.	06:51
	thats wasting my time.	06:52
	and yours.	06:52
pdtpatrick	or maybe it is difficult to teach someone something for which he has already made up his mind?	06:52
lenswipee	again i'm still waiting for your answers.	06:52
	← BadProphet left	06:52
lenswipee	i'm sure i won't get them	06:52
o]	whine whine whine	06:53
	I agree with the pizza delivery thing	06:53
pdtpatrick	if u scroll up - and read what I wrote u'll see that i said if u delete A then trying to read contents out of B, u'll get an error because B is a pointer to A	06:53
	← simesy left	06:53
lenswipee	<lenswipee> <pdtpatrick> what happens if move file A.. what happens to B ?	06:53
	answer to that question	06:53
	for the one billionth time	06:54
pdtpatrick	maybe you should stick to delivering pizza (And I mean that in the nicest way possible).	06:54
	→ zivester joined	06:54
pdtpatrick	u clearly don't plan on putting forth the effort to further your growth so stick with what u already know.	06:54
o]	I want anchovies	06:55
	→ cirwim joined	06:57
	← ampz left	06:57
lenswipee	you just like to answer questions with questions.	06:57
	← m0viefreak left	06:57
lenswipee	and waste peoples time	06:57
	bye	06:57
pdtpatrick	take care	06:57
	← shruggar left	06:58
	→ ThomasLocke joined	06:58
	← jdunck left	06:58
lenswipee	thats enough soaking for today.	06:58
	← ag4ve left	06:59
	→ AdrienBrault joined	06:59
drewery	are you still soaking?	06:59
	:)	06:59
	→ samuel02 joined	07:00
pdtpatrick	print this and put it in ur delivery car. paraphrase (Give a man a fish, he's good for a day. Teach a man to fish, he's set for life). When that makes sense to you - come back and try to soak up your sponge(maybe this time it'll retain some of the liquid).	07:00
	← v0n left	07:00
	→ kennethreitz joined	07:00
	← longbeach left	07:00
lenswipee	pdtpatrick: thats enough BS	07:01
	→ ag4ve joined	07:01
	→ mishok13 joined	07:02
	→ xpen joined	07:02
someprimetime	i want to delete a file from github but not actually remove it from my clone repo… i added the file to my gitignore global settings, but when i do `git rm filename` it removes it also from my local cloned repo	07:03
	→ simesy joined	07:04
	← jcao219 left	07:05
milki	someprimetime: man git rm --cached	07:05
gitinfo	someprimetime: the git-rm manpage is available at http://jk.gs/git-rm.html	07:05
milki	someprimetime: and ignore doesnt work for tracked files	07:05
	← boichev left	07:05
someprimetime	milki: i did that and got it back, but i still want to remove it from github	07:05
	ah	07:05
	yeah i do know that	07:05
	→ jcao219 joined	07:06
someprimetime	i reverted everything, but i still don't know how to remove that file from github (it has some stuff in it i don't want to be public)	07:06
	e.g. passwords	07:06
	→ adamben joined	07:06
drewery	lenswipee: I think you need to hire a guy to give you a private tutorial and you should pay for it since 3 days on a row and many questions is a good amount of time out of peoples lives 3 days on a row	07:06
milki	someprimetime: !sensitive and !untrack might help	07:06
gitinfo	someprimetime: [!filter_sensitive] You can use filter-branch to remove sensitive data from a repository's history. http://help.github.com/remove-sensitive-data/	07:06
	someprimetime: To remove a file from git's tracking, without deleting it from your working tree, `git rm --cached <file>`. Note that any repo which pulls this change will delete their local copy of that file. You can "bring it back" using `git checkout HEAD^ file`	07:06
	← AdrienBrault left	07:06
someprimetime	awesome thanks	07:06
	← adamben left	07:06
	← simesy left	07:07
	→ a-l-e joined	07:07
lenswipee	drewery: I'll be okay. Thanks.	07:07
	drewery: How many days is it still taking you to learn git?	07:08
drewery	half a day and a bit practice to remember	07:08
	→ infynyxx joined	07:09
lenswipee	drewery: rubbish. Takes alot longer than that to read the online documentation.	07:09
	→ dankest joined	07:09
drewery	rubbish to you then :)	07:09
cirwim	no-one ever read all of gits documentation before they were pretty expert at git	07:10
	→ marthinal joined	07:10
cirwim	it's not exactly light reading :)	07:10
lenswipee	exactly	07:10
drewery	I don't know anyone and I could care less	07:11
rking	Chiming in for my periodic reminder, "Forward-port local commits to the updated upstream head"	07:11
drewery	but I like reading and	07:11
milki	lol online documentation?	07:11
drewery	if you read the first two chapters	07:11
cirwim	rking: forward-port?	07:11
milki	i didnt spend time reading documentationm	07:11
drewery	the rest is just adds up to the show	07:11
	in the first chapter git is explained in full portrait	07:11
	did you read the first chapter?	07:11
lenswipee	drewery: that comment made me throw up on my pc.	07:12
rking	cirwim: Don't ask me! That's Junio C. Hamano's doing: whatis git-rebase	07:12
milki	whatis?	07:12
rking	Marginal improvement: https://github.com/git/git/commit/c3f0baacadbd7b5710052213a2ec3cdd5b77bb6e	07:12
cirwim	man git-rebase	07:12
gitinfo	the git-rebase manpage is available at http://jk.gs/git-rebase.html	07:12
cirwim	starts with that	07:12
drewery	lenswipee: you sound like a young kid killing peoples time for attention	07:13
milki	oo i heard of those	07:13
	vampires right?	07:13
	→ xpen_ joined	07:13
	← benbergman left	07:13
lenswipee	drewery: of course i read the first chapter.	07:13
	→ mneorr joined	07:13
lenswipee	drewery: that why i know git doesn't record changes but record all files like you think it does.	07:14
	→ eXpired joined	07:14
	← aruiz_ left	07:14
drewery	is that your argument, I didn't make any comments regarding recording files? what are you playing son?	07:14
	← xpen left	07:15
lenswipee	go back to your logs.	07:15
	brb	07:15
drewery	you do that	07:15
	I don't care really	07:15
	what a douche	07:15
cirwim	you guys are rubbish at irc drama; my popcorn's not even ready yet	07:16
pdtpatrick	haha	07:16
drewery	pop me some too cirwim	07:16
pdtpatrick	this is lenswipee: http://www.youtube.com/watch?v=fqs9DYisSsg	07:16
cirwim	will do :)	07:16
milki	popcorn doesnt take that long to make	07:16
	rking swaps cirwim's rubbish and popcorn bins.	07:17
milki	corn kernels are pretty amazing	07:17
	→ swiety joined	07:17
	← ebouchut_ left	07:17
	← Back2Basics left	07:17
milki	o, what? regular corn cant be popcorn?	07:17
	→ jdyin1 joined	07:17
	← infynyxx left	07:18
	← huoxito left	07:18
lenswipee	pdtpatrick: ha ha. but thats not me.	07:18
	← imrobert left	07:18
drewery	oh thats so you!	07:18
	lol	07:18
	→ quazimodo joined	07:18
	→ giallu joined	07:18
	gitinfo set mode: +v	07:19
	→ Alvo joined	07:19
	← jcao219 left	07:19
	→ papegaaij joined	07:19
	→ benbergman joined	07:20
	← jdyin left	07:20
	→ nafg_ joined	07:20
	→ boichev joined	07:20
eXpired	What's your problem drewery? I joined too late.	07:21
drewery	I don't have a problem, what is the problem?	07:21
	← zeks left	07:21
	← JoeAngel_ left	07:22
eXpired	<drewery> is that your argument, I didn't make any comments regarding recording files? what are you playing son?	07:22
	→ zeks joined	07:22
drewery	what are you quoting about? can you state what you are trying to say?	07:22
	→ Tegi joined	07:22
	→ poolshark joined	07:23
eXpired	Did you fuck a git commit up or something?	07:23
	→ cbreak-work joined	07:23
drewery	?	07:23
	→ xiangfu joined	07:23
milki	there is no problem	07:23
	milki waves hands	07:23
drewery	what is the problem?	07:23
	eXpired: what is the problem? what are you talking about?	07:24
eXpired	Like I said, that was the first thing I saw. I thought you fucked something up and was going to offer help.	07:24
drewery	oh :)	07:24
	→ jceb joined	07:24
	← swiety left	07:24
drewery	no no no, I was on the other site, I was trying to help	07:24
	pdtpatrick grabs a glass of wine. It's getting fiesty in here :)	07:24
eXpired	Gotcha.	07:24
drewery	lol	07:24
	thanks though	07:25
eXpired	The top thing I've learned about git since starting to use it: it will start a fucking argument.	07:26
	milki gives eXpired an A for effort and a dead horse	07:26
	← catsup left	07:26
	→ tjadc joined	07:26
	→ hoschi joined	07:27
pdtpatrick	well u can always go into #vim and say something along the lines like -- emacs > vim :)	07:28
	← TeamColtra left	07:29
eXpired	Sure. But that would be a filthy damned lie.	07:29
	← xpen_ left	07:29
drewery	depends on the type of the argument, oh well there is always one anyways	07:29
	← PaulCapestany left	07:30
	→ rath_ joined	07:30
eXpired	Convincing my team to move from CVS to Git resulted in more arguments than should be humanly possible.	07:30
	→ xpen joined	07:30
	→ PaulCapestany joined	07:30
pdtpatrick	yeah that's usually a good way to make friends at work	07:30
eXpired	Seriously.	07:31
	→ Hounddog joined	07:31
	→ dvide joined	07:31
	→ flijten joined	07:32
	→ herdingcat joined	07:32
	→ martinjlowm-lapt joined	07:32
pdtpatrick	at an old - i just ended putting together Gitlab on a VM and demonstrated in a team meeting. Long story short, I had many converts.	07:33
eXpired	I have no idea what Gitlab is.	07:33
	← jceb left	07:34
	→ h0bbit joined	07:34
drewery	I think git is good enough	07:34
pdtpatrick	http://gitlabhq.com/	07:34
drewery	but I see many people having problems because they are trying to adapt git to what they traditionally think of a revision control than	07:34
	adapting their project to the system	07:34
pdtpatrick	it's Github-like.	07:34
eXpired	Is this another Bitbucket/Github, or does this have a server?	07:34
	Ah. Neat.	07:35
	drewery: Exactly. DVCS != RCS.	07:35
	Explaining how Distributed works and why they should make lots of commits and squash before a push, and how branching is cheap, is basically impossible at times.	07:35
drewery	yes	07:35
eXpired	"What do you mean you didn't see my changes? I did a commit!"	07:36
	→ tjbiddle joined	07:36
eXpired	That alone takes several e-mails of explaining.	07:36
VeggieMeat	gitlab is great :)	07:36
	← Alvo left	07:37
eXpired	Is gitlab something that's confined to Rails or something?	07:37
	→ hafeez3000 joined	07:37
pdtpatrick	it rails	07:37
eXpired	"Fast, secure and stable solution based on Rails & Gitolite."	07:37
	Ah	07:37
	I was never sold on RoR	07:37
pdtpatrick	every thing has their place right?	07:38
eXpired	Oh, for sure.	07:38
	Seemed really cool in the beginning times of MVC	07:38
	→ fRAiLtY- joined	07:39
eXpired	I've read some things that kind of crap on its scalability. But I have no idea.	07:39
	I'm a filthy Java server dev.	07:39
pdtpatrick	i'm no rails expert but it's come a long way	07:39
	cirwim has scaled rails to hundreds of servers	07:39
eXpired	Seriously?	07:39
cirwim	yes	07:39
	← benbergman left	07:39
eXpired	That's impressive.	07:39
pdtpatrick	and if u want JVM .. there's JRUBY :)	07:40
drewery	cirwim also burned all the popcorn in the microwave :P	07:40
pdtpatrick	ha	07:40
eXpired	Which could be neat, but are there any reasons to use something like that if you're familiar with Java already?	07:40
	I mean, you get the Ruby syntax which I'll admit does some neat things.	07:40
pdtpatrick	dude metaprogramming in ruby is the cat's PJ!	07:41
	← rhl left	07:41
	← cbreak-work left	07:41
eXpired	But, it'll all compile to JRE code at some point.	07:41
	I thought Scala was the new big thing.	07:41
pdtpatrick	if java does what u already need then use it.	07:41
cirwim	eXpired: jruby has much of the dynamicness of ruby	07:41
	it's really nice not to have to edit xml to get things done	07:41
	→ cbreak-work joined	07:41
eXpired	Yeah. Not a huge fam of the XML crap.	07:41
	← samir left	07:41
pdtpatrick	cirwim: however some ruby libs don't play nice with threads - so i think some still have issues with jruby no?	07:42
eXpired	Java's going more towards annotations over XML files at this point, or Spring.	07:42
pdtpatrick	if i recall correctly	07:42
	→ fly9 joined	07:42
cirwim	pdtpatrick: perhaps... I've not had any problems with that myself	07:42
	← Serghar left	07:42
	→ samir joined	07:42
cirwim	but it's true that ruby programmers don't think so hard about concurrency as java programmers	07:42
eXpired	Which would be nice	07:42
	Same thing applies to Erland	07:42
	*Erlang	07:42
	← txomon\|home left	07:43
pdtpatrick	concurrency in ruby? splain yourself.	07:43
	AFAIK its still the same thread. GIL locks that shit down	07:43
eXpired	But, once you understand "synchronized" code blocks and such, and use concurrent compatible Apache collections, you're good.	07:43
	Wait, JRUBY can't multi-thread? It has to be able to. Or else nobody would use it. No?	07:44
pdtpatrick	Jruby sure but not ruby itself	07:44
cirwim	jruby can multithread	07:44
	mri is somewhat limited	07:44
	though most of the IO is non-blocking	07:44
	so your main sticking points are going to be external C libraries that have a lot of CPU intensive stuff	07:44
eXpired	But ruby is like PHP, like a cgi-bin process... no? Spawned on request, anatomically?	07:44
cirwim	no	07:44
	ruby processes are long-lived	07:44
eXpired	Oh, OK.	07:44
	→ rhl joined	07:44
pdtpatrick	i've tried to get myself to learn more Java since almost everything within my team at work is written in Java	07:44
eXpired	I'm not going to say Java is Jesus or anything. But once you learn it, you can do some incredible things with servers.	07:45
	→ vsayer joined	07:45
	← ninegrid left	07:45
eXpired	So, ruby typically has daemon threads?	07:45
	Does it run independently as a web server?	07:45
cirwim	depends what you're doing	07:46
eXpired	Or, require something like "mod_ruby"?	07:46
cirwim	yeah, passenger is like mod_ruby	07:46
	it manages a process pool	07:46
	← rchavik left	07:46
	← tjadc left	07:46
pdtpatrick	are u talking about running it via apache or nginx?	07:46
	there's passenger	07:46
rking	eXpired: Usually there's a backend process serving the actual requests, then a frontend web server sending images, css, and shuffling requests to the backend Ruby	07:46
cirwim	otherwise you just boot up as many processes as you need and put a load-balancer in front	07:46
eXpired	Huh. That's interesting.	07:46
	← fly9 left	07:46
	← Hounddog left	07:47
eXpired	rking: Which does make since, then, since the frontend could handle caching and all.	07:47
	→ tjadc joined	07:47
cirwim	yeah	07:47
eXpired	I should play more with Ruby :\|	07:47
cirwim	for sure :)	07:47
	← someprimetime left	07:47
	→ pantsman joined	07:47
	→ abec5 joined	07:47
	← pantsman left	07:47
	→ pantsman joined	07:47
eXpired	People are trying to convince me that I should spend time on Node.js instead.	07:47
	Not sure how I feel about that.	07:47
cirwim	it's fun	07:47
	a bit extreme though	07:47
wereHamster	eXpired: spend time with both and decide for yourself.	07:47
drewery	JavaScript is great !	07:47
eXpired	I love Javascript. For sure. But a server written in JS?	07:48
	That frightens me.	07:48
drewery	I think it will be a lot more dominant than what it is today	07:48
pdtpatrick	Github has an interesting setup with rails	07:48
	i cannot find the page but they combine tons of things	07:48
drewery	especially if you are working front end development	07:48
pdtpatrick	https://github.com/blog/517-unicorn	07:48
cirwim	I'm waiting until the yeild statement comes out	07:48
	but node is still fun without that	07:48
eXpired	That Unicorn thing looks neat.	07:49
pdtpatrick	coffeeScript is fun	07:49
eXpired	Coffeescript and Angular are next on my "crap to learn" list, after Node.js.	07:49
pdtpatrick	this is neat: http://socket.io/	07:49
drewery	learning is easy but mastering is the most fun	07:49
	← mishok13 left	07:50
pdtpatrick	yet to play around with it	07:50
eXpired	Mostly just like to play with that stuff. For some reason I got stuck on YUI2 for client side, and still have gripes with YUI3. Only spent a few hours with jquery, though I know it wins. I love JS.	07:50
	→ jargon joined	07:50
	→ tjadc_ joined	07:50
lenswipee	should i learn gitlab or git?	07:51
rking	cirwim: What's your basic Node.js sales pitch?	07:51
eXpired	Wait, so that socket.io... what kind of comm does it support?	07:51
	→ simesy joined	07:51
cirwim	rking: event loops are awesome	07:51
	← tjadc left	07:51
cirwim	do one thing and do it well => node does event loops	07:51
	← simesy left	07:52
	→ undecim_ joined	07:52
eXpired	Basic AJAX, I'm sure... but XDR/Cross domain? WebSockets? Looks neat.	07:52
	← samir left	07:52
pdtpatrick	eXpired: yeah so many possibilities	07:52
	i get a headache just thinking of the things out there	07:52
	→ azgard joined	07:52
	← The-Compiler left	07:53
Nevik	lenswipee: i dont have any experience with gitlab, but from glancing at it, i'd say you need to know git to use gitlab	07:53
	→ samir joined	07:53
rking	cirwim: What's an exemplary Node.js app that demonstrates this?	07:53
lenswipee	Nevik: fair call.	07:53
	→ The-Compiler joined	07:53
eXpired	Wait, so node.js runs on Jetty, right?	07:53
cirwim	rking: socket.io is pretty neat	07:54
	eXpired: uh, probably not	07:54
	it's javascript, not java	07:54
	← martinjlowm-lapt left	07:54
lenswipee	serverside js	07:54
	← zivester left	07:55
	← Slush-_ left	07:55
	→ AdrienBrault joined	07:55
pdtpatrick	node is also pretty cool to use via console to test ur js. Sorta like u'd use python's console or ruby's IRB	07:55
eXpired	I know, but it still sits on a basic server. For some reason I was thinking it was Jetty.	07:55
cirwim	eXpired: nah, it's all hand-written C code	07:55
	← lenswipee left	07:55
cirwim	or C++	07:55
	← undecim left	07:55
cirwim	one of those old-fashioned "actually use the real machine" languages	07:55
pdtpatrick	sublime text 2 and vim -- have good builds for this. https://github.com/tanepiper/SublimeText-Nodejs	07:55
	→ Slush- joined	07:55
eXpired	Yeah, Google tells me I'm wrong too.	07:56
	← abec5 left	07:56
eXpired	I think I'm thinking of "Play", which is some Java MVC/REST thing I was playing with recently.	07:56
	→ abec5 joined	07:56
eXpired	Is anybody else here a Java persion?	07:57
	I thought that was the lionshare of devs, actually.	07:57
drewery	I want to learn actually, I looked at it before	07:57
	I do JavaScript which has nothing to do but a common Java name	07:58
eXpired	Right, but JavaScript is rad.	07:58
drewery	I am a big fan and still learning	07:58
eXpired	I love writing JS, actually.	07:58
drewery	deep ocean	07:58
pdtpatrick	i'm slowly learning Java but i'm too comfortable with some of the things I can do with Ruby and Python that sometimes having to write all those lines in Java when when i can do it in 2 lines in ruby turns me away.	07:59
eXpired	You're totally right	07:59
cirwim	pdtpatrick: jump to scala	07:59
drewery	ruby as in framework or	07:59
	ruby as in language?	07:59
eXpired	I know Python better than Ruby, and the way it deals with collections would take some Java code a shitload of lines to match.	07:59
cirwim	ruby is a language	07:59
	→ CrazyHorse18 joined	08:00
wereHamster	eXpired: if you want to try Play, use the scala version, not java ;)	08:00
cirwim	though I guess scripting languages are really just very complete frameworks	08:00
drewery	yes but I just wanted to make sure not referring to rails	08:00
CrazyHorse18	does git pull always pull the entire remote repository including all branches?	08:00
wereHamster	CrazyHorse18: pull = fetch + merge.	08:00
eXpired	wereHamster: Thought about it. I attended an OpenWorld session on Scala, and that's about all I know about it. As far as Java goes, it and Gradle are on my list of things to check out.	08:00
CrazyHorse18	yeah	08:00
pdtpatrick	cirwim: why scala? a friend told me about it earlier as well. Might take a look at it. Though i don't think the community is as large as that of Ruby and Python. I like being able to find resources easily sometimes.	08:01
	→ afuentes joined	08:01
eXpired	Scala has some need syntax and can be used interactively	08:01
	Which is huge for learning	08:01
	*neat	08:01
CrazyHorse18	but lets say the remote repo has had a new branch added to it, will it automatically appear.. if i do a pull	08:01
	or do i have to do something manually	08:01
	basically i want someway of getting a one to one copy of the remote repo	08:02
eXpired	@CrazyHorse18: It should. If you want to check it out, do a "git -t branch <name>"	08:02
canton7	CrazyHorse18, it'll be fetched, and will appear as origin/whateverbranch. It won't create a new local branch though	08:02
CrazyHorse18	hmm ok	08:02
canton7	!local_branch_from_remote	08:02
gitinfo	The following commands are all equivalent, assuming <branch> doesn't yet exist: 'git checkout -b <branch> <remote>/<branch>', 'git checkout -t <remote>/<branch>', 'git checkout <branch>'. The latter invokes some magic.	08:02
eXpired	Er, maybe that's checkout	08:02
	→ timj1 joined	08:02
drewery	just out of curiosity... when you are planning to learn a new language, what really attracts you to learn that language?	08:02
eXpired	Career usefulness.	08:03
	Which is lame.	08:03
drewery	as in pay or gets the job done for a specific task?	08:03
eXpired	At this point, I can learn a new language in a few days. It's just syntax, if it's a functional Turing language.	08:03
CrazyHorse18	maybe there's some more advanced command.. is there any command that will copy the entire repository.. i don't even need a working version. This git repo will only be used for deploying.. (i.e. doing an export from)	08:03
eXpired	Get a specific task done, since I have the power to do that in my job.	08:03
CrazyHorse18	eXpired: usefulness...	08:03
wereHamster	eXpired: yeah.. you go learn haskell 'in a few days' and then come back and tell me if you were successful!	08:04
eXpired	But I need to fight do battle with licenses and whether it's allowed on occasion	08:04
drewery	since you know a language, learning any new language wouldn't be much time for you to tackle	08:04
	← cirwim left	08:04
wereHamster	CrazyHorse18: git clone	08:04
eXpired	wereHamster: Hah... that's one I avoided. Though I spent 2 semesters in LISP hell at school.	08:04
CrazyHorse18	wereHamster: right.. and all the remote branches will exist after i clone right?	08:05
wereHamster	CrazyHorse18: yes	08:05
canton7	eXpired, a few days yes.. Then you have the learn the standard library, which is as important but quite a bit slower :P	08:05
CrazyHorse18	ok.. and if i do a git pull later.. will new branches also be copied across?	08:05
	← timj left	08:05
wereHamster	CrazyHorse18: yes.	08:05
CrazyHorse18	awesome :D	08:05
eXpired	Is there any practical reason to know Haskell at this point?	08:05
	I consider it to be like FORTRAN or COBOL.	08:05
wereHamster	eXpired: it's a much better language than anything else :)	08:06
CrazyHorse18	COBOL's very very useful	08:06
	there's soooo much stuff writtain in cobol	08:06
canton7	CrazyHorse18, you might also like the --mirror option to git clone	08:06
eXpired	COBOL is useful if the application you're working on uses it, which pretty much always means legacy at this point.	08:06
pdtpatrick	:) i'm sold	08:06
	→ adamben joined	08:06
	← adamben left	08:06
	→ adamben joined	08:06
eXpired	wereHamster: But what platforms actually use it in production, and for what?	08:06
	→ MaximLacrima joined	08:06
eXpired	I'm ignorant about Haskell :(	08:07
pdtpatrick	s/legacy/enterprise	08:07
CrazyHorse18	canton7: what does —mirror do?	08:07
canton7	man git-clone	08:07
gitinfo	the git-clone manpage is available at http://jk.gs/git-clone.html	08:07
eXpired	pdtpatrick: Maybe with financial institutions?	08:07
wereHamster	eXpired: lots of companies. They are just not so public about it.	08:07
pdtpatrick	even more scary	08:07
canton7	CrazyHorse18, basically a fetch will overwrite your local branches, rather than updating your remote-tracking ones	08:07
wereHamster	eXpired: you know.. finanacial companies don't usually go around and brag about their code.	08:08
CrazyHorse18	actually.. hmm —bare might be a better option	08:08
eXpired	wereHamster: Yeah, obviously. And a LOT of it is legacy. If it works, it works.	08:08
wereHamster	but there are also startups who us it.	08:08
CrazyHorse18	canton7: because their should no reason to modify anything on the server.. it's just a deployment cache	08:08
eXpired	wereHamster: Cool. Every startup I hear about is about the newest craze, seems like.	08:09
	← bauruine left	08:09
	← spiderweb left	08:09
	← jstemmer left	08:09
	→ dpino joined	08:09
	→ jstemmer joined	08:09
	→ Element9 joined	08:10
	← joshsmith left	08:11
	→ suy joined	08:11
eXpired	Any of you guys ever toy with that checkio.org site?	08:11
	It's kinda fun.	08:11
	Basically thinly veiled college labs that you solve in Python	08:11
	→ rob_w joined	08:12
	← pantsman left	08:12
	→ pantsman joined	08:13
	← pantsman left	08:13
	→ pantsman joined	08:13
	→ resmo__ joined	08:13
	← Bootstrapper left	08:14
	← tjbiddle left	08:14
	← resmo left	08:14
	← resmo__ left	08:14
pdtpatrick	eXpired: i've used codeacademy	08:14
	that's a fun site as well and free :)	08:14
	→ jlg13 joined	08:15
pdtpatrick	http://www.codecademy.com/#!/exercises/0	08:15
eXpired	Agreed. Codeacademy is quite a bit more basic (handholding).	08:15
pdtpatrick	yup	08:15
eXpired	I've actually sent a lot of friends to CodeAcademy.	08:15
	→ psoo joined	08:15
	← maleknet left	08:15
eXpired	I get the "Am I too old to learn programming" question a lot	08:15
pdtpatrick	likewise .. although some just bookmark it and that's the end of it :)	08:15
eXpired	I send them there, and tell them to figure that out :)	08:16
drewery	there are also couple nice links on	08:16
	developer.mozilla.org	08:16
eXpired	For what?	08:16
drewery	for nice sites like codeacademy	08:16
eXpired	You guys tell me. I was asked last week of it's too old to have a realistic career in programming if they started at age 32.	08:17
pdtpatrick	i think it dependso n their expectation	08:17
eXpired	Architect asked this. Clearly had legitimate interest, but can you really land a career if you start then?	08:17
drewery	yes	08:18
	you can	08:18
jast	well, you will have to learn, of course	08:18
eXpired	I mean, you have to have at least 4 years of REAL experience before you land a job, right?	08:18
jast	and most courses are worthless because they teach the programming languages rather than the mindset	08:18
drewery	there is nothing really can stop you but yourself	08:18
	it all depends	08:18
eXpired	jast: That's the difference between a Univ CS degree and a vocational, I think.	08:18
jast	or something along the lines of, hey, here's what object orientation is	08:18
drewery	if you are comfortable and naturally capable	08:18
	it would be easy to produce some nice work demonstrates how good you are	08:19
eXpired	drewery: If you only plan to contract, for sure. But would your company hire an older programmer who's a beginner? I don't think mine would, to be honest.	08:19
jast	eXpired: I dunno. I've yet to see any formal training that turns people without the mindset into people with the mindset	08:19
drewery	it depends on the credentials	08:19
	not persons age	08:19
	but functionality and then the company policy (like expectations)	08:19
CrazyHorse18	uni developers are mostly useless	08:20
	first year out	08:20
eXpired	jast: I can see that. My freshman year was at WSU in 2,000. Everybody there thought "Computers! Get rich quick!". Then the instructer started talking about pointers.	08:20
CrazyHorse18	and it takes a couple of years to get them to be able to do stuff properly	08:20
eXpired	35% of the students were booted for cheating	08:20
	Not exaggerating.	08:20
drewery	on the other hand	08:20
pdtpatrick	eXpired: Washington State? are u in seattle ?	08:20
eXpired	pdtpatrick: I'd planned on it initially, but nope. I love Seattle.	08:20
drewery	older person would have a more mature manner to work ethic and project handling	08:20
	→ _ikke_ joined	08:20
CrazyHorse18	drewery: it really depends on the person	08:21
eXpired	CrazyHorse18: They are. Internships are HUGE!	08:21
	But some college students just think they're assured a job on graduation.	08:21
CrazyHorse18	i've met 16yr old's who are excellent developers.. and 50yr olds who can't even write a document in english to describe how something simplistic works	08:21
drewery	CrazyHorse18: it always does, but a person CAN do it if he wills it.	08:21
pdtpatrick	yeah Seattle is pretty cool. I'm still trying to like the weather	08:21
CrazyHorse18	drewery: the best developer i ever had was a 21yr old chinese guy	08:21
drewery	I think people are more successfull at things that they can intrinsicly do versus extrinsicly do	08:22
CrazyHorse18	who came from an engineering background	08:22
	i.e. he went from doing hardware coding to web-development	08:22
drewery	CrazyHorse18: I think everybody has a different ... like gift... can't find the word	08:22
eXpired	pdtpatrick: Wanted to mvoe there for the music scene when I was a kid, so did WSU. And Seattle has a great tech community.	08:22
	← psilo_ left	08:22
CrazyHorse18	i hired him as an intern.. and he was the best developer in 6 months from 0 knowledge	08:23
pdtpatrick	eXpired: it still. Amazon, Microsoft :)	08:23
	← sajbar left	08:23
	→ catsup joined	08:23
	→ sajbar joined	08:23
eXpired	CrazyHorse18: Exactly. At work, I deal with a lot of dyed in the wool systems devs who are almost 60. They just can't wrap their minds around server development.	08:23
CrazyHorse18	eXpired: some college students just think they're assured a job on graduation <<< :O	08:23
	yeah.. i have this issue as well	08:23
eXpired	Some of them are damned geniuses. But there might be an age, I guess.	08:23
CrazyHorse18	i've been thinking of taking a different approach	08:24
	→ jceb joined	08:24
eXpired	pdtpatrick: True. I have a few friends that work at F5. But MS... I kinda feel like that's a sinking ship. Amazon is amazing, however.	08:24
CrazyHorse18	and writing up a detailed training course.. going to all the universities	08:24
	→ psilo joined	08:24
CrazyHorse18	and saying if you learn this.. we will give you an internship	08:24
	→ mishok13 joined	08:24
CrazyHorse18	and then teaming up with a few of the other software companies in the area.. so we can start having developers who can actually code on day 1	08:25
	← sangi left	08:25
eXpired	CrazyHorse18: Exactly! But not necessarily "if you learn this"... you need to separate passion from the book geeks who just want the money,.	08:25
	I always ask about personal projects when there's a potential hire.	08:25
CrazyHorse18	oh yeah.. 95% of people that apply	08:25
drewery	exactly, thats what sets these people apart from the competition	08:25
CrazyHorse18	don't have any code they can show	08:25
	wtf is with that	08:25
eXpired	Right	08:26
CrazyHorse18	it doesn't make sense though.. software developers applying for jobs	08:26
	→ zeppo joined	08:26
CrazyHorse18	but haven't done any software	08:26
	→ aisbaa joined	08:26
eXpired	Something like, "I wrote an Android add that shows my personal photo album from Flickr" means a lot more than, "I got a 3.7 GPA and surf on the weekends"	08:26
	*app	08:26
drewery	it happens, and happened with me since I traveled way a lot :)	08:26
	→ ninegrid joined	08:26
drewery	and lost my older stuff in an external drive...	08:27
	← jeremy_c left	08:27
CrazyHorse18	here's some peice of code i wrote to steal the photos from the city to bay half marathon, because i couldn't be bother paying $25 per photo :)	08:27
	e.g.	08:27
eXpired	Haha	08:27
	Honestly, that kind of crap too.	08:27
CrazyHorse18	actually	08:27
	→ svetlyak40wt joined	08:27
CrazyHorse18	i got 3 jobs	08:27
	because of a web app	08:27
eXpired	I grew up doing stupid "script kiddy" things.	08:27
CrazyHorse18	i wrote that was for recording music videos	08:28
eXpired	That's rad!	08:28
	What did it do?	08:28
	→ alfadir joined	08:28
	→ sangi joined	08:29
CrazyHorse18	ok.. connected to the abc's website.. downloaded the list of music videos to be played on friday night and saturday morning	08:29
eXpired	RSS scrape?	08:29
	→ chimay joined	08:29
	→ maleknet joined	08:29
CrazyHorse18	parsed out the hour times.. counted number of music videos per hour.. worked out average length of videos and calculated expected time would be on	08:29
eXpired	I had no idea that ABC ever played music videos	08:30
CrazyHorse18	did an ls of a local hard drive with the existing ones	08:30
	→ simesy joined	08:30
CrazyHorse18	australian broadcasting comission	08:30
	parsed the artist name and song name	08:30
eXpired	So, basically, you timeshifted recordings for music videos and compared it against a cache	08:30
CrazyHorse18	and then built an ajax clickable interface	08:30
	with colours	08:30
eXpired	That's awesome!	08:30
CrazyHorse18	green for already got, white for not, blue for record tonight.. and if the video was recorded twice	08:30
	it then produced a recording file.. which it would sent to the application	08:31
	that sets up the recording	08:31
eXpired	So... I'm guessing you've seen Sickbeard?	08:31
CrazyHorse18	and had all this helping shit for copying and pasting file names when we were chopping up the videos	08:31
	we ended up getting 12000 music videos	08:31
eXpired	That's effing amazing	08:31
	← thyseus left	08:31
CrazyHorse18	then .. on one fateful day	08:31
eXpired	None of that database is still online, is it?	08:32
CrazyHorse18	youtube came up.. and we lost interest	08:32
eXpired	Damn. Damn damn damn damn. I miss music videos :(	08:32
	→ xxc_michael joined	08:32
CrazyHorse18	we've still got the entire collection	08:32
	was never online	08:32
wereHamster	CrazyHorse18: you should have sold your system as youtube!	08:32
CrazyHorse18	it was like 100GB	08:32
	or more..	08:32
eXpired	Which, at the time, I'm sure was huge	08:32
	→ skyf joined	08:32
CrazyHorse18	oh yeah.. double-pass xvid encoding.. ahh the good old days	08:32
	eXpired: oh we had the biggest collection out of anyone in 2004.. by 100 times	08:33
eXpired	I used to burn CD-R's full of xvid music videos.	08:33
CrazyHorse18	oh yeah then the police raided our server.. that was funny	08:33
eXpired	Oh, crap... really?	08:33
CrazyHorse18	took them 4yrs to give the fucker back	08:33
	nah for something else though	08:33
	→ thyseus joined	08:33
CrazyHorse18	the house it was in	08:33
eXpired	So you have those videos still?	08:33
CrazyHorse18	was owned by a laywer	08:34
	who was reprsenting some other clients.. and for some reason they decided that my friend was involved	08:34
	so they turned up took all the computers	08:34
	kept them for 4 years	08:34
	then gave them back all scratched up	08:34
eXpired	Bastards!	08:34
CrazyHorse18	we knew all the forensics guides	08:34
	and we asked how they went	08:35
	because it was a freebsd, using a custom raid5 software driver	08:35
eXpired	I've helped a felon friend of his decrypt his federally returned hard drives.	08:35
CrazyHorse18	which was partially hand-coded	08:35
	and they said.. after they'd imaged the drives, they had no idea how to stich them together again :)	08:35
eXpired	Oh, so they didn't mount them correctly for reading?	08:35
	→ senny joined	08:35
CrazyHorse18	nah they couldn't mount	08:35
eXpired	That's beautiful	08:35
CrazyHorse18	you'd have to use the original hardware	08:36
	or buy identical hardware	08:36
	because it was software raid.. and compiled on that machine	08:36
	← ninegrid left	08:36
	liori_ → liori	08:36
eXpired	I urge you to rebuild that array and archive those videos.	08:36
CrazyHorse18	only had music videos and tv on it	08:36
drewery	http://speedtest.net/android/303206319.png	08:36
CrazyHorse18	i've got them all	08:36
drewery	wrong window	08:36
	→ JoeAngel_ joined	08:36
eXpired	That many videos is probably nearly an entire collection	08:36
CrazyHorse18	yeah	08:37
eXpired	I mean, videos have been made since... but not really. It's a lost art form.	08:37
CrazyHorse18	means nothing now	08:37
	you can just download them in hd from youtube	08:37
eXpired	I suppose.	08:37
	Still, neat app idea.	08:38
CrazyHorse18	lol.. took 35 hours	08:38
	got me 3 jobs	08:38
	haha	08:38
	axiom → majoh	08:38
	← sangi left	08:38
	→ lenswipee joined	08:38
	→ marhaban joined	08:38
eXpired	When I was a teen I developed a custom FTP server that operated internally by routing FXP commands. In other words, a custom client registered people's FTP servers with mine, and it was a distributed FTP server.	08:39
CrazyHorse18	first lot i showed it too were wetting themselves.. and were talking about how they were going to integrate some of the functionality into their webapp. because they'd never seen any javascript stuff before	08:39
eXpired	That was neat.	08:39
drewery	night everyone, time to take off	08:39
CrazyHorse18	eXpired: what year was this?	08:39
eXpired	But I was 16, and wrote it in VB and soon discovered women.	08:39
	1997	08:39
	← drewery left	08:39
	← marhaban left	08:39
CrazyHorse18	lol.. vb	08:39
eXpired	Yeah	08:39
CrazyHorse18	yeah… i used to code in vb	08:39
eXpired	I was big in the whole "warez" thing as a teen	08:39
lenswipee	take care drewery	08:40
eXpired	It was my counter culture, I guess.	08:40
CrazyHorse18	yeah.. we used to download tv shows over dial-up	08:40
eXpired	Sharing files was something that was neat to me.	08:40
CrazyHorse18	2 days for one episode!	08:40
	:D	08:40
eXpired	Oh crap	08:40
	I know.	08:40
CrazyHorse18	4.6kbs :D	08:40
	haha	08:40
	→ flyz_ joined	08:40
CrazyHorse18	modem set to auto-redial.. scripting ftp servers to reconnect	08:40
	irc commands :D	08:40
	→ Cromulent joined	08:40
CrazyHorse18	ah … "the shit old days"	08:41
eXpired	First modem was a 2400bps. I'll never forget my dedicated phone line downloading Diablo 1, which was 600 Meg, for over two days.	08:41
CrazyHorse18	none of this.. click one button magnetic torrent shit	08:41
eXpired	I mean, by then it was a 33.6k, but crap.	08:41
lenswipee	is the object store of index located differently from the object store of commits?	08:41
	← Cromulent left	08:41
	→ rednaks joined	08:41
wereHamster	lenswipee: what is an 'object store of index'?	08:41
CrazyHorse18	eXpired; you downloaded diablo 1 :O back then it took an hour to download an mp3	08:42
eXpired	It's all within the .git directory in binary blobs, if that's what you mean by objects?	08:42
	← mneorr left	08:42
lenswipee	wereHamster: the collection of blobs	08:42
eXpired	CrazyHorse18: Hahaha. Exactly! And MP3's took up a sizable portion of the hard drive. Which is weird.	08:42
	→ fly9 joined	08:42
	→ Pugnax joined	08:42
wereHamster	lenswipee: there is only .git/objects.	08:42
	→ mneorr joined	08:42
	→ Cromulent joined	08:43
lenswipee	wereHamster: listen. the blobs that are created when you do 'git add [filename]' are stored as objects	08:43
	← adamben left	08:43
wereHamster	correct.	08:43
	← whitman left	08:44
	← simesy left	08:44
lenswipee	wereHamster: in a 'object store'.	08:44
eXpired	Essentially, yes.	08:44
wereHamster	in .git/	08:44
eXpired	Git's object store doesn't do entire file copies. It operates on diffs, and stores binary blobs.	08:44
lenswipee	right now listen to next bit,	08:45
	→ sangi joined	08:45
lenswipee	when you do a commit 'git commit [filename]' you also create an object that represents the index.	08:46
	→ whitman joined	08:46
alfadir	hi, have a git of articles, would like to tag them with a doc revision. they are independent and many. any pointers how to deal with such tagging ? they are to small to make a git for each, and pretty many so before going wild on tagging i thought i'd check if there are any best practices. pointers welcome, thank you.	08:46
	← skyf left	08:46
wereHamster	lenswipee: yes (I'll play along	08:47
eXpired	lenswipee: The object you're referring to is a SHA1 hash, but sure.	08:47
	← fly9 left	08:47
	← Pugnax left	08:48
eXpired	alfadir: Create a root git repo and add the documents to it. No reason for multiple repo's/branches.	08:48
lenswipee	eXpired: i think of it as a reference (SHA1 hash) to the object (index), but anyway.	08:48
eXpired	Yup, you're not wrong.	08:48
lenswipee	oh good. now	08:49
alfadir	eXpired: i have a root git repo, but want to tag sub trees so to say..	08:49
lenswipee	are the object database for the 'add objects' stored separately from the object database for the 'commit objects'.	08:50
wereHamster	lenswipee: no	08:50
eXpired	alfadir: Are you talking about sub-modules?	08:50
	That's something that more recent versions of git support.	08:50
	→ Pugnax joined	08:51
lenswipee	wereHamster: how are they stored?	08:51
wereHamster	lenswipee: http://git-scm.com/book/ch9-2.html	08:51
alfadir	no, i have a git of documents, they have or will have revision history. As they are many git tag -a v1.4 will not do. then i would have to do git tag -a fullpathtodoc-v1.0 ?	08:52
	← Gentlecat left	08:52
alfadir	a bit like having one git for many small software projects that are not big enough to put in separate gits.	08:53
wereHamster	alfadir: why use tags? Just change, and commit, and in the commit message describe your changes and include the version (vx.y)	08:53
	alfadir: and when you want to list all versions of a document, use git log	08:53
alfadir	sure.. ok.. well git log . would work for that was looking for git tag . :)	08:54
	→ jeremy_c joined	08:54
	← dankest left	08:54
lenswipee	wereHamster: according to that link its all stored in one big object database.	08:54
	→ resmo joined	08:54
alfadir	it is an unusual usecase i know	08:54
eXpired	lenswipee: More importantly, what are you trying to do?	08:54
wereHamster	lenswipee: if that's what it says.. then it must be true.	08:54
lenswipee	eXpired: learn git.	08:55
eXpired	alfadir: Definitely an unusual usecase. You can tag sets of file, but you'll need to split the repos up (and you could create them as sub-modules of something)	08:55
	← AdrienBrault left	08:55
eXpired	lenswipee: Ah. OK :) Using git doesn't need you to learn all of the internals, though. I totally see where you're coming from, however.	08:56
lenswipee	eXpired: it seems the object database is structured like a tree with branches.	08:56
alfadir	i know, already using subtree merging to follow vendor branches in some cases.	08:56
wereHamster	lenswipee: no.	08:56
lenswipee	wereHamster: no? pls elaborate.	08:56
wereHamster	lenswipee: it's a tree in the sense that the objects are stored in a filesystem which uses a tree-like structure. But that's about it.	08:56
	← mneorr left	08:57
	→ mneorr joined	08:57
alfadir	but for this it is not worth the work, git log works but is not so elegant as local tags..	08:57
lenswipee	wereHamster: sounds like what i said but in more detail :)	08:58
alfadir	svn comparison would be to tag just a subtree with svn cp xx/doc1 tags/doc1/1.1 or so.. well been a while since i did svn	08:58
	→ timj joined	08:59
	→ narfnarfnarf joined	08:59
	→ LimeBlast joined	08:59
	→ jlopez joined	08:59
wereHamster	lenswipee: by that argument, everything is stored in a tree. Even mysql databases. Because they store data in a filesystem.	08:59
Bombe	Yay, I’m stored in a tree! \o/	09:00
wereHamster	Bombe: shut up, trees can't talk!	09:00
	:P	09:00
	→ dlam joined	09:00
	← maleknet left	09:01
	← zheng left	09:01
lenswipee	wereHamster: I hear you. How do you explain the concept of tree and branches in git, since the tree concept has nothing to do with the filesystem?	09:01
	← timj1 left	09:02
	← Cromulent left	09:02
	→ jn_ joined	09:02
wereHamster	lenswipee: I'm not sure I understand you.	09:03
lenswipee	wereHamster: It's the way the objects are stored.	09:03
wereHamster	lenswipee: doesn't that article from the git book explain it?	09:03
lenswipee	wereHamster: lol	09:03
	→ adamben joined	09:04
	→ montrotyr__ joined	09:04
	← mneorr left	09:04
	→ _Qman joined	09:05
lenswipee	yeah it says the index is a branch	09:05
wereHamster	where?	09:05
	→ mneorr joined	09:05
Bombe	wereHamster, I am not a tree, I’m only stored in a tree!	09:05
	→ irqq joined	09:05
lenswipee	Bombe: weren't you told to "shutup"? :D	09:06
	→ Anders_J joined	09:06
lenswipee	wereHamster: I don't remember the exact line and page number.	09:06
	← montrotyr_ left	09:07
wereHamster	lenswipee: no book or tutorial would wrie that, because it's simply not true.	09:07
	→ alkino joined	09:08
	← xpen left	09:08
	→ Desproges joined	09:08
lenswipee	wereHamster: the index is somehow part of the repos	09:08
	← Vile left	09:09
wereHamster	yes it is. But how does that make it a branch?	09:09
lenswipee	wereHamster: talk later.	09:10
	wereHamster: i'll share notes with you later i mean.	09:11
	→ robotmay joined	09:11
lenswipee	wereHamster: take care.	09:11
eXpired	lenswipee: It's stored as a tree in essentially what is metadata in binary form.	09:11
	It requires the git binary to know how to deal with the blobs	09:11
lenswipee	eXpired: what are you referring to by "It's"	09:12
eXpired	You're not going to be able to manually dissect the storage of this	09:12
	Anything. The changesets.	09:12
lenswipee	eXpired: the intial index is a complete file copy.	09:14
eXpired	Right.	09:14
	← ping-pong left	09:14
eXpired	And then changesets are composed of diffs, stored in binary blob form.	09:14
lenswipee	eXpired: the rest are diffs,	09:14
	← zeppo left	09:15
Bombe	wat.	09:15
	Git doesn’t store diffs.	09:15
eXpired	That was my impression.	09:15
	It stores deltas between versions, no?	09:15
Bombe	No.	09:16
	Because that would be a diff.	09:16
	And Git doesn’t store diffs.	09:16
	→ simesy joined	09:16
lenswipee	bingo	09:16
eXpired	I thought that was what made checkouts and branching dirt cheap in git?	09:16
Bombe	Yes, not storing diffs makes all that dirt-cheap.	09:16
eXpired	I also assumed the diff storage was what made merging so quick and simple.	09:17
lenswipee	Bombe: then the blobs are complete file contents?	09:17
eXpired	How does it internally store and index things?	09:17
Bombe	lenswipee, yes.	09:17
	And directories, and commits, and tags.	09:17
	That’s about it.	09:17
	← JStoker left	09:17
Bombe	Use the fucking internet.	09:17
	← abec5 left	09:17
wereHamster	eXpired: maybe you should also have a look at http://git-scm.com/book/ch9-2.html	09:18
lenswipee	We have.	09:18
	→ xpen joined	09:18
lenswipee	wereHamster: :D	09:18
	← samuel02 left	09:18
lenswipee	eXpired: at least you learnt something new today.	09:18
eXpired	Yeah, definitely	09:22
	→ samuel02_ joined	09:22
	← a_L_p left	09:22
	← kennethreitz left	09:22
eXpired	Never claimed to be a git expert, especially on how it works under the covers.	09:22
	I will now use the fucker internet to figure that out.	09:23
lenswipee	eXpired: thats OK. you know more than I do.	09:23
CrazyHorse18	is there a way to do an "export" of a git repository..	09:24
	i.e. so it doesn't create the .git directory	09:24
	→ TeamColtra joined	09:24
	← hafeez3000 left	09:24
wereHamster	CrazyHorse18: man git-archive	09:25
gitinfo	CrazyHorse18: the git-archive manpage is available at http://jk.gs/git-archive.html	09:25
CrazyHorse18	cheers	09:25
	git archive master \| tar -x -C /somewhere/else	09:26
	→ maleknet joined	09:26
wereHamster	CrazyHorse18: note that this won't delete files	09:27
	← samuel02_ left	09:27
CrazyHorse18	hmm	09:27
	← eXpired left	09:27
	→ ping-pong joined	09:28
CrazyHorse18	ok.. so basically i've got my repo in /repos/application_name and copies in my app in both /apps/application_name-int and /apps/application_name-uat	09:28
	→ thiago joined	09:28
CrazyHorse18	int needs to match master and uat will always be a particular tag	09:28
lenswipee	wereHamster: Thr problem with," http://git-scm.com/book/ch9-2.html " is that is doesn't explain the index and how the commit tree uses it.	09:28
CrazyHorse18	what's the easiest way of syncing those changes up?	09:28
wereHamster	CrazyHorse18: have /repos/application_name a normal repo that you update with fetch+reset --hard, and rsync it over to the two othre directories.	09:29
	→ Hounddog joined	09:29
CrazyHorse18	right	09:29
	so checkout the branch / tag	09:29
	i want?	09:29
wereHamster	lenswipee: you mean the relation between the index and the commit object?	09:29
	→ ampz joined	09:29
CrazyHorse18	rsync excluding the .git repo	09:30
lenswipee	wereHamster: i know that relationship.	09:30
CrazyHorse18	git direcotry	09:30
	← ampz left	09:30
wereHamster	lenswipee: so what exactly is unclear?	09:30
CrazyHorse18	wereHamster: hmm. you run into problems then if one developer deploys to uat and another one deploys to int at the same time	09:30
	the other option would be to clone it.. and then just delete the .git directory i suppose	09:31
	→ bux joined	09:31
wereHamster	CrazyHorse18: then use /repos/application_name-int-deployment-checkout and /repos/application_name-uat-deployment-checkout	09:31
lenswipee	wereHamster: earlier discussions you made it sound like the index and commit objects are all part of the "tree".	09:31
CrazyHorse18	some of these repos are large and the internet connections are shit, so i wanted to avoid having duplicates	09:32
wereHamster	lenswipee: well, yes, of the filesystem tree: ls .git/index, ls -la .git/objects/*/	09:32
	→ zenx joined	09:35
	← thiago left	09:35
	→ EugenA joined	09:36
lenswipee	wereHamster: the index is simply an extraction of the commit tree with changes from the working tree.	09:36
	← mtk left	09:37
	→ alireza joined	09:37
wereHamster	lenswipee: no. index is list of files. Literally. No relation at all with any commit object or a tree object.	09:37
EugenA	git remote shows me one remote repo: origin. But i don't see remote branches	09:38
wereHamster	EugenA: git branch -r	09:38
	← Milossh left	09:39
lenswipee	wereHamster: then the index is stored in a different dir than the tree objects.	09:39
	← maleknet left	09:39
wereHamster	yes.	09:39
EugenA	what is the proper way to push my commits to special remote branch?	09:40
wereHamster	lenswipee: well, the index is a single file: .git/index	09:40
	← jeremy_c left	09:40
wereHamster	it's stored separately of the git objects, which are stored in .git/objects/*	09:40
lenswipee	bingo	09:41
	→ mtk joined	09:41
	← lresende left	09:42
lenswipee	how do i list the index in git?	09:42
	← sluther left	09:42
	→ berserkr joined	09:42
MacroMan	EugenA, git push origin mybranchname	09:43
	← MaximLacrima left	09:43
	→ marhaban joined	09:43
wereHamster	lenswipee: git ls-files	09:44
	← slamice left	09:44
	← OOPMan left	09:45
EugenA	MacroMan: refusing to update checked out branch: refs/heads/master	09:45
	MacroMan: that is why I want to push to different remote branch	09:45
	→ slamice joined	09:45
MacroMan	EugenA, I can't say I've come across that error, a push to my branch name usually works fine for me.	09:46
	→ JStoker joined	09:46
EugenA	it looks like my workflow is not corrent then	09:46
lenswipee	wereHamster: the index is all the files i initially added when I first created the repos, which then gets updated after every commit?	09:46
MacroMan	EugenA, Are you currently in master or your new branch?	09:46
	→ AdrienBrault joined	09:47
	→ zeppo joined	09:47
MacroMan	Not sure if it'll make a difference or not, but I push while in the branch I'm pushing	09:47
	→ Error404NotFound joined	09:47
EugenA	on remote server is master checked out. On my local system i've cloned it like this: git clone [email@hidden.address] .	09:47
	← marhaban left	09:48
	→ jgomez joined	09:49
MacroMan	Do you have persmission on the remote server to create a new branch?	09:49
	→ hafeez3000 joined	09:49
	← JoeAngel_ left	09:49
EugenA	yes	09:49
	→ BadProphet joined	09:49
	← slamice left	09:50
MacroMan	EugenA, I'm not really sure I can help any further, I'm no expert. Maybe someone else will know what's wrong.	09:50
wereHamster	lenswipee: index is the list of all files which you have git-add'ed	09:50
	← jlg13 left	09:50
wereHamster	lenswipee: minus all files which you have git-rm'ed	09:51
	→ arvind_khadri joined	09:51
	← laurav left	09:51
	→ chepre joined	09:52
lenswipee	wereHamster: whats in the index. I just see filenames and alot of unreadable stuff	09:53
wereHamster	lenswipee: mostly filename + stat info	09:53
	← jackh_ left	09:53
lenswipee	wereHamster: stat <-- means status?	09:54
EugenA	so, I cannot push to checked out branch. I need bare repository for that, right?	09:54
	← BadProphet left	09:54
wereHamster	lenswipee: man 2 stat	09:54
	→ Bootstrapper joined	09:54
wereHamster	lenswipee: everything git needs to be able to : 1) tell you that the fies have changed in your working tree, and 2) to construct a tree object for when you git commit.	09:55
	← e66 left	09:55
	← afuentes left	09:57
lenswipee	wereHamster: why is it not human readable?	09:57
	← babilen left	09:57
	← Bootstrapper left	09:58
	← x82_nicole left	09:58
wereHamster	lenswipee: because it does not have to be.	09:58
	→ marhaban joined	09:58
wereHamster	lenswipee: it's slower to parse a human readable format.	09:58
MacroMan	lenswipee, Plus there is git gui if you want more human readable stuff.	09:59
	→ Milossh joined	09:59
wereHamster	or git ls-files	09:59
	← edgar-rft left	09:59
lenswipee	wereHamster: the index contains sha1 hashes/references to the objects in .git/objects?	09:59
	wereHamster: that gives you just the filenames.	10:00
wereHamster	lenswipee: yes to your question and no idea what you mean by your second statement.	10:00
lenswipee	git ls-files	10:01
	← Rix left	10:01
lenswipee	wereHamster: that gives you just the filenames.	10:01
	→ dsirijus joined	10:01
	← rath_ left	10:01
wereHamster	yes, so?	10:01
	→ chuckharmston joined	10:02
lenswipee	wereHamster: these index objects are different to the commit objects that form the tree.	10:02
wereHamster	commit objects don't form a tree. Blobs and trees form a tree.	10:02
	lenswipee: since the index only contains files, it only references blob objects.	10:03
lenswipee	wereHamster: the blob objects of files are different to those blob objects created from a commit.	10:04
wereHamster	Nevik: no 'blob object' is created from a commit	10:05
	lenswipee: ^^	10:05
	→ rath_ joined	10:06
lenswipee	wereHamster: previous discussion spoke of 'git add [filename]' creating a blob object of file	10:06
EugenA	how do I need to organize my remote repository (repositories) to enable different developert uploading their commits and checking them in browser?	10:06
wereHamster	lenswipee: that is true.	10:06
lenswipee	wereHamster: that is why i said:	10:07
	wereHamster: these index objects are different to the commit objects...	10:07
	→ tvw joined	10:07
lenswipee	wereHamster: replace objects with blobs if u want.	10:07
wereHamster	this is like saying the number 1 is different from the numb r2	10:07
	→ jeremy_c joined	10:07
	→ lausi joined	10:07
	→ thierryp joined	10:08
	→ myonebuddy joined	10:08
	→ Rix joined	10:08
wereHamster	the term 'commit object' has a very special meaning, and is very different from a 'blob object'.	10:08
lenswipee	wereHamster: right. let me rephrase. the index blob are different from the commit blob?	10:09
wereHamster	lenswipee: the term 'commit blob' does not make sense (in the context of git)	10:10
	same with 'index blob', what is that?	10:11
	→ hinst joined	10:11
lenswipee	wereHamster: As above, 'index blob' is the blob created for file when you do git add	10:12
	← j0ran left	10:12
jast	files are always stored as the same kind of blob, no matter whether they're referenced from the index or from a commit	10:12
	→ Vile joined	10:12
wereHamster	lenswipee: just call it a blob	10:12
	→ j0ran joined	10:12
lenswipee	jast: right.	10:13
	→ Alvo joined	10:13
jast	they can even be referenced from both at the same time... happens quite often in practice	10:14
	so it wouldn't make sense to have two different formats	10:14
wereHamster	lenswipee: so your question was: When I create a commit, does the commit reference the same blobs as are referenced by the index?	10:14
lenswipee	wereHamster: yes, and i'm guessing the answer is yes.	10:15
	→ polymar joined	10:16
	→ scroogey joined	10:16
	gitinfo set mode: +v	10:16
	← EugenA left	10:16
scroogey	hello	10:16
	i am having trouble with getting git stash and git svn working :/	10:17
	→ tsdh joined	10:17
	→ Xizor joined	10:17
	← alireza left	10:18
	→ shruggar joined	10:20
	→ fger joined	10:20
fger	hi guys	10:20
gitinfo	fger: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	10:20
fger	one question: can i have a specific tag of a repository as submodule of another GIT repo ?	10:21
lenswipee	jast: are the blobs of the index stored in a different file to those blobs created after commit?	10:21
wereHamster	lenswipee: no.	10:21
lenswipee	wereHamster: do I get a confirmation?	10:21
fger	or do i always have to the master for a submodule ?	10:21
wereHamster	confirmation of what?	10:21
ethanol	I have trouble understand git pull when working in multiple branches. By default when you clone, you get a master branch that tracks the master branch from origin, correct?	10:22
	→ samuel02 joined	10:22
	← dlam left	10:22
	→ ianfleeton joined	10:22
ethanol	then if I create a new branch and set it up to track origin/newbranch, and do a pull --all, why do changes from origin's master not propogate into my local master branch?	10:23
	I have to actually switch to it (checkout) and do the pull again	10:23
	how come?	10:23
fger	anybody ?	10:23
	tag of a repo as submodule in another ?	10:23
wereHamster	fger: sure	10:23
fger	ok thx :)	10:23
lenswipee	wereHamster: so the initial blobs created from index are stored in .git/objects/00	10:23
wereHamster	lenswipee: .git/objects/*, the subdirectory depends on the file contents	10:24
lenswipee	wereHamster: what about the file contents?	10:26
wereHamster	lenswipee: what about them?	10:26
lenswipee	wereHamster: u said " depends on the file contents"	10:27
	wereHamster: whats the dependance?	10:27
wereHamster	lenswipee: a file with contents "foo" is stored in .git/objects/19/.. while a file with contents "bar" is stored in .git/objects/ba/...	10:28
	but that is all explained in the git book chapter I linked to earlier	10:28
lenswipee	wereHamster: each blob must then contains previous versions.	10:29
wereHamster	lenswipee: what 'previous versions'?	10:29
lenswipee	wereHamster: each blob of a file must contain previous versions of itself.	10:30
	← flyz_ left	10:30
scroogey	my git repo keeps complaining about the LF / CRLF issue	10:30
wereHamster	lenswipee: no, how.. why.. what.. I don't get how your brain works..	10:31
	→ hafeez3000_ joined	10:31
Nevik	wereHamster: i assume you didnt mean to hilight me there ?	10:31
wereHamster	Nevik: no	10:31
Nevik	kay	10:31
	← marthinal left	10:32
	→ nsadmin joined	10:32
lenswipee	wereHamster: i see your not going to make it easy for me.	10:33
	wereHamster: thanks and bye.	10:33
	← hafeez3000 left	10:33
	hafeez3000_ → hafeez3000	10:33
wereHamster	lenswipee: you haven a 1GB file in your working tree. You git add that file; git reads that file into memory, creates a blob object from it and writes it somewhere ti .git/objects.	10:33
	→ marthinal joined	10:33
wereHamster	lenswipee: your .git directory has justgrown by 1GB	10:34
	lenswipee: you change one bit in that file. you git add that file again. git reads that fie into memory, creates a blob object from it and writes it somewhere to .git/objects	10:34
	lenswipee: your .git directory has justgrown by another 1GB	10:34
	lenswipee: your .git directory is now ~2GB big.	10:34
	lenswipee: git ad also updates the index, after it writes the blob to .git/objects. It updates the index with: "Hey, there is a file /foo/bar which has contents <blob sha1>"	10:36
	→ beautifulmind joined	10:36
	← hyperair left	10:36
lenswipee	wereHamster: your a master of over simplifying things and being vague about the complex workings of git.	10:36
jast	git isn't all that complex	10:37
lenswipee	hmm	10:37
Nevik	indeed	10:37
_ikke_	!complex	10:37
Nevik	dahw	10:37
_ikke_	!simple	10:37
gitinfo	At its heart git is made up of many concepts that are individually simple. Getting the whole picture right is often tricky, and it is usually about breaking up the complex concept into its simple, individual parts and grokking those. Both !bottomup and !cs will help with that.	10:37
Nevik	ah	10:37
scroogey	:))	10:37
_ikke_	.trigger_edit complex @!simple	10:37
gitinfo	_ikke_: Okay.	10:37
Nevik	lawl :D	10:37
	thats a good bot	10:37
nsadmin	y	10:38
jast	fger: in fact submodules always point to specific commits, so it doesn't matter whether you use a branch or a tag	10:38
	→ soc42 joined	10:39
jast	Nevik: and nobody else has it! mwahaha!	10:39
Nevik	what does the bot run "on" (or, what kind of bot is it?) -- or is custom-made?	10:39
jast	!bot	10:39
gitinfo	[!gitinfo] I am an IRC bot which responds to certain keywords to provide helpful(?) information to humans. Please see http://jk.gs/git/bot for more information about how to use me.	10:39
_ikke_	Nevik: custom made	10:39
Nevik	cool	10:39
	← sangi left	10:39
Nevik	(requires Perl, common::sense <--- LOL	10:39
	→ mikef joined	10:40
Nevik	gotta love coders' naming schemes	10:40
jast	hmm, docs on that page are not quite complete	10:40
	I'm sure we have a few additional features	10:40
fger	jast: and they are "state-of-the-art" for connecting git repos with another ? what i look for is sth like svn externals	10:40
	jast: or is there some other git construction that is suited better ?	10:40
Nevik	there is no other	10:40
	→ bodom joined	10:41
lenswipee	wereHamster: explain in same manner now when you do a commit.	10:41
jast	fger: pick your poison: !subpro	10:41
gitinfo	fger: [!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.	10:41
	← marthinal left	10:41
fger	thx jast :)	10:41
Nevik	or use them in a private message	10:41
	if you wanna read them all	10:41
	← Error404NotFound left	10:41
wereHamster	lenswipee: git reads the index, creates a tree object from it, writes the tree object to .git/objects. It then creates a commit object (which references the just created tree object), writes that object to .git/objects. It then updates HEAD to point to the new commit you created.	10:41
mikef	i messed up my code, and i'm currently commiting to something git has called "no branch" how can i rescue this?, how do i merge this to master or turn it into a branch?	10:41
	← svetlyak40wt left	10:42
Nevik	jast, _ikke_: the trigger should be augmented with an "or in a pm"	10:42
jast	mikef: write down the commit ID(s) (git log), checkout master, git cherry-pick <commit ID>	10:43
	or do this: git checkout -b temporarybranch	10:43
	then you can merge as normal	10:43
	← eli left	10:43
mikef	jast: what does -b do?	10:43
	→ marthinal joined	10:43
jast	mikef: create a branch from your current status	10:43
_ikke_	fixed	10:44
mikef	jast: perfect thanks	10:44
jast	_ikke_: what? :)	10:44
_ikke_	!subprojects	10:44
gitinfo	So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.	10:44
_ikke_	ugh	10:44
mikef	!subtree	10:45
_ikke_	Now for real	10:45
gitinfo	The subtree merge method is great for incorporating a subsidiary git repo into your current one with "unified" history. Read http://git-scm.com/book/en/Git-Tools-Subtree-Merging for more info, or try one of the !subtree_alternatives	10:45
	→ babilen joined	10:45
Nevik	_ikke_: cool :)	10:45
	← hoschi left	10:45
	← simesy left	10:45
	← luc_ left	10:45
fger	guess i will use submodules then...	10:46
	but if i have to switch to another commit in a submodule, i have to re-wire it again ?	10:47
lenswipee	wereHamster: which HEAD? in /git or /git/ref?	10:47
wereHamster	lenswipee: there is only one: .git/hEAD	10:47
fger	so i use tag 4.5.1 as submodule and then i would have to switch to tag 4.5.2 for that submodule to be connected to another repo...	10:48
Nevik	fger: !submodule	10:48
gitinfo	fger: 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	10:48
	→ sangi joined	10:48
fger	yes i read that	10:48
	th	10:48
	x	10:48
	← samuel02 left	10:48
Nevik	fger: the submodule will behave like a normal git repo	10:48
	within your super-repo	10:48
	you can just cd into it, and check out another commit	10:48
fger	ah ok thx	10:48
	← mikef left	10:48
fger	good to hear :)	10:49
Nevik	then cd into the super repo, and commit the change there, so it will remember (and publish) that it now depends on that other commit	10:49
	→ simesy joined	10:49
fger	i want to have an app core as submodule	10:49
	thats why im asking	10:49
Nevik	as was pointed out, the super-repo will always save which commit in the submodule is currently used.	10:49
fger	the sub core comes from a remote repo on the internet	10:49
	ok	10:49
	← yshh left	10:50
Nevik	fger: in your super-repo, use "git submodule add <remote url> <local target folder>"	10:50
	that will create the submodule, clone it, and check out the default branch	10:50
fger	ok	10:50
Nevik	fger: more details can be found in !book	10:50
gitinfo	fger: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	10:50
Nevik	and on man git-submodule	10:50
gitinfo	the git-submodule manpage is available at http://jk.gs/git-submodule.html	10:50
	→ robustus joined	10:51
	← simesy left	10:51
	→ Suprano joined	10:51
fger	ok i will check that out and implement a sample	10:51
	thx guys	10:51
	← tvw left	10:51
	← fger left	10:51
	→ palyer joined	10:52
	← palyboy left	10:52
	← jdyin1 left	10:52
	→ alireza joined	10:52
lenswipee	wereHamster: the tree objects and commit objects are not blobs?	10:52
Nevik	lenswipee: theyre short structs with mainly pointers to other things	10:53
	like parent commit(s) and relevant blob(s)	10:53
	→ benbergman joined	10:53
jast	lenswipee: correct. blobs, trees, commits, tags. those are the types of git objects.	10:53
	→ achadwick joined	10:54
	← Marenz left	10:54
lenswipee	different file types so to speak	10:54
	→ bauruine joined	10:55
	→ JoeAngel_ joined	10:55
	→ TheHaven joined	10:55
lenswipee	i suppose the numeric folders and alpanumeric folders in .git/objects hold differnt git objects	10:55
wereHamster	lenswipee: the directory name is the first two characters of the hex sha1 representation of the object	10:56
	so an object with sha1 ba0e162e1c47469e3fe4b393a8bf8c569f302116 will be stored in .git/objects/ba/0e162e1c47469e3fe4b393a8bf8c569f302116	10:56
	that, also, is explained in the git book chapter	10:56
	← xiangfu left	10:57
	← Pugnax left	10:57
	→ kennethreitz joined	10:57
	→ hoschi joined	10:59
	← Rix left	11:01
	← chuckharmston left	11:01
	→ simesy joined	11:02
	← herdingcat left	11:02
	→ eijk joined	11:02
	→ lenswipee_ joined	11:03
lenswipee_	wereHamster: how does git know the difference between various objects types stored in .git/objects? when its hex sha1 referenced?	11:03
_ikke_	lenswipee_: It's stored in the object	11:03
lenswipee_	_ikke_: that would make it slow	11:04
wereHamster	lenswipee_: read http://git-scm.com/book/ch9-2.html#Object-Storage	11:04
	lenswipee_: what makes you htink that?	11:04
	← ianfleeton left	11:05
lenswipee_	it would have to parse all objects to find say tree object	11:05
_ikke_	lenswipee_: No it doesn't	11:05
wereHamster	lenswipee_: the question is, why do you want to find that tree object?	11:05
	→ caddoo joined	11:05
wereHamster	in which situation do you want that?	11:05
_ikke_	lenswipee_: All objects have to be referenced	11:05
lenswipee_	the commit object would reference the tree object	11:06
_ikke_	it does	11:06
	← simesy left	11:06
wereHamster	lenswipee_: yes, so you already know its sha. So you go and open that file in .git/objects/XX/YYY...	11:06
	← lenswipee left	11:06
lenswipee_	which it has to look for in .git/objects among all the other types of objects	11:06
wereHamster	no. it goes and opens a single file.	11:06
	→ eijk_ joined	11:07
_ikke_	there is a 1-on-1 relation between the sha and the file	11:07
lenswipee_	i don't know the commit objects sha1	11:07
wereHamster	lenswipee_: the commit knows the tree it references.	11:07
	lenswipee_: http://git-scm.com/book/ch9-2.html#Commit-Objects	11:07
	← eijk left	11:07
	← pansdr left	11:08
lenswipee_	lets say i want the 3rd commit.	11:08
wereHamster	third commit from where?	11:08
	→ pansdr joined	11:08
lenswipee_	i know the sha1 is there in .git/objects	11:08
_ikke_	!xy	11:08
gitinfo	This sounds like an "XY Problem" http://mywiki.wooledge.org/XyProblem So let's step back for a minute. What are you actually trying to achieve? Why are you doing it this way?	11:08
wereHamster	lenswipee_: you mean you want the HAED^^^ commit?	11:08
	→ eijk__ joined	11:09
	← _Qman left	11:09
	→ _Qman joined	11:10
lenswipee_	i want the third commit that i did.	11:10
	← eijk__ left	11:10
	→ mikecmpbll joined	11:10
	→ eijk__ joined	11:10
lenswipee_	git commit	11:10
wereHamster	time wise? History wise?	11:10
lenswipee_	x3	11:10
	← eijk__ left	11:10
wereHamster	git doesn't have an efficient way to do that.	11:11
Nevik	lenswipee_: get HEAD, go back to "end" of history, back up three commits	11:11
	lenswipee_: git history is a singly-linked-list of commits	11:11
	← eijk_ left	11:11
_ikke_	!gcs	11:11
gitinfo	[!concepts] "Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://sitaramc.github.com/gcs/	11:11
	← kimico left	11:11
	→ eijk joined	11:11
	→ LekeFly joined	11:12
wereHamster	there are numerous ways how to approximate that. But none of it is reliable. The commit might not even exist. Or be the second commit (history-wise) because you rebased etc..	11:12
	← LimeBlast left	11:12
	→ kimico joined	11:12
lenswipee_	lets try again	11:12
	i do:	11:12
	git init	11:13
	→ tvw joined	11:13
lenswipee_	git add file	11:13
	git commit	11:13
	change file	11:13
	git add file	11:13
	git commit	11:13
	change file	11:13
	git add file	11:13
	git commit	11:13
	thats 3 times	11:14
wereHamster	HEAD is the third commit, the one you just made.	11:14
lenswipee_	now i want to get the first commit	11:14
	→ scroogey_ joined	11:14
wereHamster	HEAD^^	11:14
	that's the grandparent of HEAD.	11:14
Nevik	lenswipee_: PLEASE. READ what we tell you. git does NOT know what the "first" commit is	11:15
	→ Rix joined	11:15
Nevik	git looks back on history	11:15
	→ kirin` joined	11:15
lenswipee_	when do you HEAD^^	11:15
	← cworth left	11:16
lenswipee_	it looks for HEAD in /git	11:16
	→ cworth joined	11:16
lenswipee_	reads the history	11:16
	grabs the sha1	11:16
	→ Cromulent joined	11:16
lenswipee_	uses that is .git/objects	11:16
	*in	11:16
wereHamster	it looks for the HEAD ref, parses the commit it points to, takes its parent, parses that commit, takes the parent again and that's what you want.	11:16
lenswipee_	and grabs the tree object	11:16
	→ LimeBlast joined	11:17
	→ simesy joined	11:17
	→ ircmouser joined	11:17
jast	you haven't told it to look at the tree object	11:17
wereHamster	no trees involved at all.	11:17
	it's just commit walking.	11:17
	← senny left	11:17
	← bux left	11:17
	← scroogey left	11:17
	→ JEEB joined	11:17
	← rednaks left	11:19
	← kimico left	11:19
	← jaalto left	11:19
	← hafeez3000 left	11:19
lenswipee_	what does HEAD^^ return?	11:19
_ikke_	The parent of the parent of the head commit	11:20
wereHamster	lenswipee_: when given to which git command? git show HEAD^^? git log HEAD^^, git rev-parse HAED^^?	11:20
jast	try it yourself on the command line: git rev-parse HEAD^^	11:20
	← h0bbit left	11:20
lenswipee_	The parent is a tree object no?	11:20
jast	no	11:20
_ikke_	no	11:20
	It's a commit	11:20
	→ hafeez3000 joined	11:21
jast	when you make a new commit, the previous commit becomes its parent	11:21
wereHamster	lenswipee_: your parent is a human. you are a human and yoru parent is also one. The same type.	11:21
lenswipee_	which references a tree object	11:21
_ikke_	You're confusing what a tree object is	11:21
jast	when you merge, the merge commit gets two parents	11:21
	← kirin` left	11:21
_ikke_	A tree object is a tree of files that belong to a commit	11:21
	it has nothing to do with the history	11:21
	← asaf left	11:22
_ikke_	(it's part of history)	11:22
	→ kirin` joined	11:22
lenswipee_	so it returns a sha1?	11:22
	← lausi left	11:23
wereHamster	lenswipee_: the tree object describes how your project looked (on the filesystem) at the time you did the commit.	11:23
	→ h0bbit joined	11:23
	→ Kyzz_ joined	11:23
	→ lausi joined	11:23
lenswipee_	bingo a sha1	11:23
wereHamster	lenswipee_: typing HEAD^^ in the terminal produces an error. However, 'git rev-parse HEAD^^' will print the SHA1 of the grandparent of HEAD.	11:23
	but, 'git rev-parse HEAD^^' produces something different from 'git show HEAD^^', which is different from what 'git log HEAD^^' produces.	11:24
lenswipee_	and then i use this sha1 to get the corresponding tree object	11:24
wereHamster	if that's what you want, yes.	11:24
	→ scroogey joined	11:25
	← scroogey left	11:25
	← zenx left	11:25
	→ zenx joined	11:25
lenswipee_	git show HEAD^^ just gives the comment message	11:25
jast	it should also give the diff	11:25
	→ Error404NotFound joined	11:25
jast	though maybe it doesn't do that for the root commit	11:26
wereHamster	and author identity. and date.	11:26
jast	never tried that	11:26
	← Kyzz left	11:26
wereHamster	it does.	11:26
	→ ampz joined	11:26
	← nafg_ left	11:26
	← lausi left	11:26
Nevik	yeah	11:26
	← scroogey_ left	11:26
	← chimay left	11:27
	← gmarseliss left	11:27
lenswipee_	git log HEAD^^ is just a subset of git log	11:27
Nevik	yes	11:27
	← ampz left	11:27
wereHamster	lenswipee_: 'git log' == 'git log HEAD' == show me the log starting at HEAD.	11:27
	← echo-area left	11:27
wereHamster	lenswipee_: git log X == show me the log starting at X.	11:27
Nevik	lenswipee_: since you still have not done !xy as weve told you twice, we still dont know what youre trying to do	11:27
gitinfo	lenswipee_: This sounds like an "XY Problem" http://mywiki.wooledge.org/XyProblem So let's step back for a minute. What are you actually trying to achieve? Why are you doing it this way?	11:27
	← sangi left	11:27
wereHamster	lenswipee_: hence: git log HEAD^^ == show me the log starting at the grandparent of HEAD	11:27
	→ sangi joined	11:27
	→ project2501b joined	11:28
	← kirin` left	11:28
	→ kirin` joined	11:29
lenswipee_	sweet	11:29
	← benbergman left	11:29
	← Cromulent left	11:29
	gitinfo set mode: +v	11:30
lenswipee_	Just realised I was mislead by someone who said HEAD^ was the most recent commit.	11:32
_ikke_	Even HEAD is not per definition the most recent commit	11:33
	!HEAD	11:33
gitinfo	HEAD is a 'pointer' to the currently checked out branch (or commit, if HEAD is !detached). In bare repositories it serves a different function: it tells clients which branch to checkout initially after cloning. HEAD is not something that exists separately for every branch; that's a common misunderstanding. It also is not the newest commit in the repo (which is hard to define in a DVCS anyway)	11:33
	→ Pugnax joined	11:34
lenswipee_	_ikke_: how does one get the newest commit?	11:34
_ikke_	Newest in what way?	11:34
	← kirin` left	11:35
lenswipee_	well certainly HEAD^ doesn't qualify as the newest commit.	11:35
_ikke_	HEAD^ is the parent of HEAD, which points to the currently checked out commit	11:35
lenswipee_	It also is not the newest commit in the repo (which is hard to define in a DVCS anyway)	11:35
	what that meant	11:35
	→ kirin` joined	11:36
	→ svetlyak40wt joined	11:36
	← tsdh left	11:36
	→ benbergman joined	11:36
	→ devsharpen joined	11:37
	← nsadmin left	11:38
	← polymar left	11:38
	→ polymar joined	11:38
	← hafeez3000 left	11:39
	← zenx left	11:40
	→ Donno joined	11:40
Nevik	lenswipee_: just like it doesnt know the first, git doesnt know the last commit; it does not care about such things	11:42
	← kirin` left	11:42
	← lenswipee_ left	11:42
Nevik	it only cares about which commit is related to which others	11:42
	← polymar left	11:43
	→ OOPMan joined	11:43
	→ kirin` joined	11:44
	← benbergman left	11:44
	← simesy left	11:46
	→ samuel02 joined	11:47
	← tvw left	11:48
	← kirin` left	11:49
	← xpen left	11:49
	→ kirin` joined	11:50
	← marhaban left	11:50
	→ polymar joined	11:50
	→ benbergman joined	11:51
	→ drizzd joined	11:51
	← CompMike left	11:52
	← drizzd_ left	11:54
	→ zenx joined	11:55
	← kirin` left	11:55
	← benbergman left	11:57
	→ kirin` joined	11:57
	→ rednaks joined	11:57
	→ angelsl joined	11:58
	→ blami joined	12:01
	← soc42 left	12:01
	→ blami_ joined	12:02
	→ pmdarrow joined	12:02
	← kirin` left	12:03
	→ kirin` joined	12:04
	← svetlyak40wt left	12:04
	← rednaks left	12:06
	→ bux joined	12:06
	← dsirijus left	12:07
	→ hafeez3000 joined	12:08
	→ senny joined	12:08
	→ benbergman joined	12:08
	← dragon3_ left	12:09
	← kirin` left	12:09
	← hafeez3000 left	12:09
	→ Ryback_ joined	12:09
	← Ryback_ left	12:09
	→ Ryback_ joined	12:09
	→ hafeez3000 joined	12:10
	← sangi left	12:10
	→ kirin` joined	12:10
	→ DarkDonno joined	12:11
	→ kimico joined	12:12
	→ Raging_Hog joined	12:13
Raging_Hog	I want to know all that's happened in a branch foo after it was branched from master. Trouble is, it has been merged to master so I can't use master..foo	12:13
	← benbergman left	12:13
Raging_Hog	right now I'm using the number 3. in the first answer here http://stackoverflow.com/questions/1527234/finding-a-branch-point-with-git	12:14
	← Donno left	12:14
	Raging_Hog is happy that he managed to ask this today without accidentally pasting five lines and getting klined from the network	12:14
	→ fisted joined	12:15
	← robbyoconnor left	12:15
	← kirin` left	12:15
	→ rchavik joined	12:16
	← rchavik left	12:16
	→ rchavik joined	12:16
	→ kirin` joined	12:17
	→ e-dard joined	12:17
	→ glennpratt joined	12:17
	→ hyperair joined	12:18
	← fisted_ left	12:18
	DarkDonno → Donno	12:20
	← eijk left	12:20
	→ zheng joined	12:20
	← hotovson_ left	12:21
	→ nioncode joined	12:21
	← kirin` left	12:22
	→ svetlyak40wt joined	12:22
	← kennethreitz left	12:22
	← Bombe left	12:23
	→ kirin` joined	12:23
	→ kennethreitz joined	12:24
	→ Bombe joined	12:25
	← GeertB left	12:25
	→ benbergman joined	12:25
	→ quazimod1 joined	12:27
	→ R1KRDO joined	12:27
	← kirin` left	12:29
	→ kirin` joined	12:30
	→ sylr joined	12:31
	← arvind_khadri left	12:31
	← kerrick left	12:33
	→ marhaban joined	12:33
	→ juvenal joined	12:35
	← kirin` left	12:35
	→ jfs joined	12:35
	← jfs left	12:35
	→ jfs joined	12:35
	→ picoguy joined	12:36
	→ rgr joined	12:36
	→ kirin` joined	12:36
	← kirin` left	12:41
	← hafeez3000 left	12:42
	→ kirin` joined	12:43
	→ bitkiller joined	12:43
	→ hafeez3000 joined	12:43
	← jhutson456_ left	12:44
	← bitkiller left	12:44
	← senny left	12:44
	← h0bbit left	12:44
	→ TomF joined	12:46
	← jargon left	12:46
	← IridiumScaffold left	12:46
osse	Raging_Hog: has 'foo' since been deleted?	12:47
	→ btree_ joined	12:47
Raging_Hog	osse, no, it's active. The point is that I'm developing on a branch that was once merged but has been reopened, and I want to get the big picture of everything that has happened in it during its whole lifetime	12:48
	← kirin` left	12:48
	→ kirin` joined	12:49
	→ tvw joined	12:50
	→ bitkiller joined	12:51
	← btree left	12:51
Nevik	Raging_Hog: do you know when it was first created (i.e. the commit it was based on) ?	12:51
	← ehsan left	12:52
	← TomF left	12:52
	→ cakehero joined	12:52
	← hafeez3000 left	12:53
	→ hafeez3000_ joined	12:53
	← hafeez3000_ left	12:53
Raging_Hog	Nevik, that's what I'm after. I used to find it in the commit log, but that's a bit tedious. The SO answer allows me to do git diff $(git oldest-ancestor master foo) and git log, in a similar way, or gitk or whatever I feel like	12:54
	→ hafeez3000 joined	12:54
Nevik	ah, hm	12:54
	← kirin` left	12:55
Nevik	we've had the same question a few days ago, but there's probably no one way of going at it	12:55
	→ thomasf1021 joined	12:55
Nevik	Raging_Hog: afaik, git cannot discern which branch a commit has belonged to once it was merged somewhere else	12:55
	→ kirin` joined	12:55
Nevik	that has to do with the way git saves history	12:55
	Raging_Hog: so you probably have no reliable way of actually finding out where the branch started	12:56
	unless you can find someone with a really old clone ;P	12:56
wereHamster	or still have the reflog	12:56
Nevik	yeah	12:56
	which is unlikely, from the way it sounds	12:56
wereHamster	(and the parse the reflog somehow to get that information)	12:56
Nevik	how long is the reflog usually, until it gets truncated ?	12:57
	→ cannonball joined	12:57
wereHamster	Nevik: man git-reflog, man git-gc, man git-config	12:58
gitinfo	Nevik: the git-reflog manpage is available at http://jk.gs/git-reflog.html	12:58
	Nevik: the git-gc manpage is available at http://jk.gs/git-gc.html	12:58
	Nevik: the git-config manpage is available at http://jk.gs/git-config.html	12:58
	← tvw left	12:58
	→ MrDHat joined	12:59
	← quazimod1 left	12:59
	→ eijk joined	12:59
Nevik	ah, good to know	12:59
	→ iamjarvo joined	13:00
Nevik	so basically the reflog does not auto-expire if i dont manually gc or so ?	13:00
	← kirin` left	13:01
Raging_Hog	will the command in that SO answer give a reliable result? If I understood it correctly, it looks at all the commits in both branches and selects the oldest (or first?) common one	13:01
	→ hotovson joined	13:01
wereHamster	some git comands do run git-gc	13:01
Nevik	ah	13:01
	→ kirin` joined	13:02
Nevik	Raging_Hog: it's not 100% reliable, but probably the best you can get	13:02
	← hellopat left	13:02
jast	I believe autogc doesn't expire anything	13:02
	→ mp_ joined	13:02
jast	but I haven't verified this	13:02
	→ BadProphet joined	13:02
Raging_Hog	Nevik, in what situations will it fail?	13:02
	← undecim_ left	13:02
Nevik	Raging_Hog: i cant think of a concrete scenario, but as i pointed out, youre trying to determine something which git doesnt usually remember	13:03
	→ undecim_ joined	13:03
Nevik	so it might well be that it cant be reconstructed properly	13:03
	oh i have an example	13:03
	Raging_Hog: the only way to tell the first "appearance" of a branch is when it diverges from another (before that, they share all history)	13:04
	← letharion left	13:04
Nevik	if you create a branch and dont diverge it immediately (but ff it on the "parent" branch for a while), it wont be reflected	13:04
	← picoguy left	13:04
Nevik	Raging_Hog: so actually the method in the SO answer will probably find the oldest non-common commit	13:05
	← skul left	13:05
	← BadProphet left	13:06
	← kirin` left	13:07
Raging_Hog	that sounds fitting for my needs, because I'm interested in what's been happening in the branch that wasn't in master at the time. If I were to just diff origin/master or something like that, I'll see the new changes in master, too. Then I will create a foo branch and rebase it on origin/master and then check the diff, but that's a bit more work	13:07
	→ kirin` joined	13:08
	← narfnarfnarf left	13:08
Nevik	alright, if that's all you need, the SO answer should serve you well enough :)	13:08
	→ yshh joined	13:09
	→ chimay joined	13:09
	← mikecmpbll left	13:10
	← sylr left	13:10
	→ sylr joined	13:10
	→ thomasf10211 joined	13:11
	← thomasf1021 left	13:11
	→ thiago_ joined	13:12
	← fRAiLtY- left	13:12
	← JoeAngel_ left	13:13
	← pretty_function left	13:13
	← kirin` left	13:13
	← thomasf10211 left	13:13
	→ thomasf1021 joined	13:14
	→ rgsteele joined	13:14
	→ kirin` joined	13:14
	→ martinjlowm-lapt joined	13:14
	→ sebrock joined	13:16
	→ sgambino joined	13:16
	← samuel02 left	13:16
	← alireza left	13:17
thomasf1021	Good Day. I have a question about how to get my SVN to GIT conversion working. I currently run an Unstable trunk in SVN. I've used SubGit to start running the two in parallel. Trunk became my master, so it is inherently unstable as well. I've got a Staging branch that I tag off of for production releases. The problem I'm running into now is that it seems I have to cherry pick every commit when going from master -> staging ? Any thoughts would be apprecia	13:17
	← beautifulmind left	13:19
	← irqq left	13:20
	→ MiKom joined	13:22
Nevik	thomasf1021: i dont know how subgit works, but you should be able to just merge staging into master	13:23
	or actually, master into merging	13:23
	if you want all the changes to be taken over	13:23
	← alunduil left	13:23
Nevik	if you only want single commits, then cherry-picking is the way to go	13:23
	← gtrak left	13:24
thomasf1021	SubGit works by using commit hooks to pass the commits to the other VCS.	13:24
Nevik	oh, then, i have no idea how to not break it. the guys who make subgit will be able to help you with that	13:25
thomasf1021	So assuming I've created a ticket branch, completed my work and merged it back into master, I'd have to then cherry pick each revision to get it into Staging?	13:25
	→ samuel02 joined	13:25
Nevik	thomasf1021: is there any reason youre not completely migrating to git (yet) ?	13:26
	← Donno left	13:26
Nevik	thomasf1021: if there is nothing in master you DONT want in staging, you can just merge master into staging	13:26
thomasf1021	It's working perfectly, just that I'm struggling on how to use an unstable master with GIT. I have lots of changes in master that are not yet ready to be migrated, so a standard merge won't work	13:26
	gotcha	13:26
Nevik	thomasf1021: git doesnt care that it's called master	13:26
	← MrDHat left	13:26
thomasf1021	I haven't migrated completely to GIT since my use case isn't yet satisfied. It seems like GIT is meant to only ever use a stable master	13:27
Nevik	it's just a bit of a convention to call the main branch master, but it's just like any other branch	13:27
	no	13:27
	thomasf1021: you can use however many stable or unstable branches you want, and name them whatever you want	13:27
	(e.g. trunk if that feels comfy to you)	13:27
	if youre interested, you might also look at !workflows and !book in general	13:28
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	13:28
	→ rednaks joined	13:28
Nevik	er whats that other trigger	13:28
	!workflow	13:28
gitinfo	Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	13:28
	→ letharion joined	13:28
Nevik	_ikke_: can you alias !workflows to @!workflow (or can i do that too?)	13:28
	thomasf1021: the book and the articles tell you a bit about what workflows are possible with git (or rather, they give you examples; git is generally not limiting you in terms of workflow)	13:29
jast	everyone can do it	13:30
thomasf1021	Thank you i've read them all but they don't answer the problem of mainline development being done and migrated later on. They all base their assumptions that you're branching from production. Then at some point merging to an integration branch to try and ensure conflicts are handled. Let me try to describe it another way	13:30
jast	you just need to be logged into nickserv	13:30
	→ mustafavelioglu joined	13:31
	→ wdouglass joined	13:31
Nevik	.trigger_edit workflows @!workflow	13:31
gitinfo	Nevik: Okay.	13:31
Nevik	!workflows	13:31
gitinfo	[!workflow] Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	13:31
Nevik	yay for working technology ^_^	13:32
jast	you're welcome ;)	13:32
Nevik	thomasf1021: they might all just not mention that it doesnt matter which way you do it	13:32
mustafavelioglu	Hi, i want to add files to gitignore without deleting in my repos and other developers. How can i do this ?	13:32
	→ m0viefreak joined	13:32
jast	mustafavelioglu: if the files are kept in the repo, the ignore patterns don't apply to them, period	13:33
wdouglass	hey all! i've got a question. with 'git log' i can get a list of commits that deal with a specific subdirectory in my git repo. with 'git format-patch' i can get patchsets for a subset of commits. how do i get a set of patches to build just a subdirectory of my repo? is this possible?	13:33
bremner	certainly deleting other developers is tempting sometimes, but illegal in most jurisdictions.	13:33
Nevik	lol bremner	13:33
jast	wdouglass: not a format-patch style patch	13:33
Nevik	thomasf1021: but yes, if you can describe your workflow a bit, maybe i can help you figure it out	13:33
jast	best you can do is, say, git diff commit1 commit2 -- path > file.patch	13:33
	→ alireza joined	13:34
mustafavelioglu	jast : I have to kept files in my repo. Just, i dont want to track them changes	13:34
wdouglass	jast: i was afraid of that. i guess i could write a script to do 'git log' and then iterate from commit to commit, generating patches. thanks!	13:34
bremner	mustafavelioglu: have you read !config	13:34
gitinfo	mustafavelioglu: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas	13:34
thomasf1021	Let's take a CMS example. If I'm adding say a video object to the system (call Feature A) and also the ability to comment on the videos (Feature B). If I branch off of the production release for Feature A, I'll be ok. But for Feature B I have to wait until Feature A is released or I have to branch from Feature A directly, which assumes that I've pushed/exposed Feature A's branch	13:34
	→ tvw joined	13:35
	← pansdr left	13:35
jast	wdouglass: you can use 'git rev-list' instead of 'git log' for easier scriptability	13:35
	same arguments but you get only the commit IDs	13:36
Nevik	thomasf1021: that depends on who is doing feature b. if youre doing it yourself, you can just branch off feature a any time	13:36
	← Pugnax left	13:36
	→ shamrock joined	13:36
Nevik	but if someone else is doing it, they need to share your feature a branch, yes	13:36
wdouglass	jast: cool! i always forget about that :)	13:36
	← heroux left	13:36
	shamrock → Guest28501	13:36
	rgr → Guest59987	13:37
	← boichev left	13:37
	← Guest28501 left	13:37
	→ xiangfu joined	13:37
	→ EugenA joined	13:37
thomasf1021	Nevik: Thank you for your help. I need to change my workflow entirely then but such is life if I want to use GIT. :)	13:37
	→ rgr joined	13:37
Nevik	thomasf1021: many people here have more experience with git than i; feel free to hang around and talk some more, maybe someone can help you figure out a way that you dont have to change it all	13:38
	→ heroux joined	13:38
EugenA	i have two repos: normal - site and bare - hub. Git should go to site and pull the updates from hub after push to hub	13:38
Nevik	thomasf1021: however, to use all features of git, it is good to use a workflow that's aware of the git features and concepts	13:38
	→ luc_ joined	13:38
	thiago_ → thiago	13:38
Nevik	EugenA: what do you mean "go to site" ?	13:39
EugenA	go to repo "site"	13:39
	i've created hook "pust-commit" in bare repo, right?	13:39
Nevik	you should mention such things :P	13:40
	→ jackh joined	13:40
Nevik	im not a hook-expert so i'll step out here	13:40
EugenA	no, i mean "post-update"	13:40
	→ jcromartie joined	13:40
	→ MrDHat joined	13:41
	→ yoh joined	13:42
	→ tjadc joined	13:42
	← Tectu left	13:42
	← tjadc_ left	13:42
	→ erichynds joined	13:43
yoh	[Hi to all gitters] in gitk, highlighting of commits and files "touching path" works fine when files are not specified via globs -- with globs (e.g. a*.c) commits highlighted but not the files ... looks like a bug -- or am I missing smth?	13:43
	← SegFaultAX left	13:43
	← alireza left	13:44
	← tr4656 left	13:45
	← Alvo left	13:45
	← emankcin left	13:45
	← davidh left	13:46
	→ emankcin joined	13:46
	← marhaban left	13:46
	← peper left	13:46
	← hachque left	13:46
	← tm1000 left	13:46
	→ h0bbit joined	13:46
	← d1b left	13:46
	← sharpobject left	13:47
	← Aaeriele left	13:47
	→ tmcmahon joined	13:47
	→ Eridius joined	13:47
	← AdrienBrault left	13:47
	→ hola123 joined	13:47
	← shihan left	13:47
	← roto left	13:47
	→ freeed joined	13:47
	→ cmyers joined	13:48
	← Hamled\|Erp left	13:48
	→ jlg13 joined	13:48
	→ tm1000 joined	13:48
	→ tr4656 joined	13:48
	→ chrisgeorge joined	13:48
	→ dgtlmoon joined	13:48
	→ davidh joined	13:48
dgtlmoon	greetlings	13:48
	→ shihan_ joined	13:48
	→ Aaeriele joined	13:48
	→ Hamled joined	13:48
	→ peper joined	13:49
dgtlmoon	question, when i make a new branch, does that branch move along with 'master' ? or do i have to merge in any changes from master back into that branch?	13:49
	→ hachque joined	13:49
	→ SegFaultAX joined	13:50
	→ maleknet joined	13:50
thiago	you need to move it manually	13:50
rking	So, if I 'git checkout upstream/master' or if I 'git branch upstream/master', both leave me in a detatched HEAD. How do I get the 'upstream' remote to act like my normal 'origin'?	13:50
dgtlmoon	thiago, ahh ok	13:51
thiago	rking: the second command does not change your current branch	13:51
rking	Aha	13:51
thiago	rking: therefore, it is incorrect to say that it left you in a detached-HEAD state.	13:51
dgtlmoon	thiago, you are replying to me?	13:51
thiago	rking: it did not change the state	13:51
	← maleknet left	13:51
rking	Gotcha	13:51
thiago	dgtlmoon: yes, the first answer was to you	13:51
dgtlmoon	ok :)	13:51
thiago	rking: now, the first command does move you to a detached HEAD, since upstream/master is not a local branch	13:52
	→ boichev joined	13:52
	← bitkiller left	13:52
	hola123 → arkiver	13:52
	→ narfnarfnarf joined	13:52
	← MrDHat left	13:52
rking	thiago: OK, so how do I get a not-detached HEAD from upstream/master	13:53
Nevik	rking: you will need to create a tracking branch first	13:53
	man git-branch and man git-checkout for that	13:53
gitinfo	the git-branch manpage is available at http://jk.gs/git-branch.html	13:53
	the git-checkout manpage is available at http://jk.gs/git-checkout.html	13:53
	→ kenyon joined	13:53
	→ Arokh joined	13:53
thiago	rking: non-detached-HEAD means "I am in a local branch"	13:53
	→ roto joined	13:53
	→ sharpobject joined	13:53
rking	Oh, hrm.	13:53
	→ d1b joined	13:53
thiago	rking: therefore, you need a local branch that is at the commit you want it to be	13:53
Nevik	thiago: it also means "im on the tip of a local branch"	13:53
	→ v0n joined	13:54
thiago	rking: you want to use the git branch command to create such a branch, then git checkout to check it out	13:54
rking	OK	13:54
	← jlg13 left	13:55
	← zheng left	13:55
	← CrazyHorse18 left	13:55
	← jn_ left	13:56
	→ sideone joined	13:56
	→ mikecmpbll joined	13:56
rking	Hrm. The git-branch manpage should mention that --set-upstream is deprecated. (The CLI does)	13:57
	← adamben left	13:57
	← EugenA left	13:58
	← marc_kdab_ left	13:58
	→ marc_kdab_ joined	13:59
	← rednaks left	13:59
	← lacrymol1gy left	13:59
	→ bitkiller joined	14:00
Nevik	that man page you got linked to is not the very latest version	14:01
	it has been deprecated since 1.8.0 afaik	14:01
	← tmcmahon left	14:01
Nevik	as long as the option is still working, that's fine	14:01
	→ GeertB joined	14:01
thiago	--set-upstream is a design flaw	14:02
	← samuel02 left	14:02
thiago	when given one argument, it does exactly the opposite of what any sane person would want	14:02
rking	I'm looking at the same manpage from my git install	14:03
	Git 1.8.0.1.250.gb79	14:03
	But this is my first day on HEAD git, so maybe I should be glad I found a small spot to contribute.	14:04
	It'd be nice if someone could confirm if I'm correct about the manpage, though.	14:04
wdouglass	thiago: agreed. that functionality should just be in --track	14:04
thiago	rking: by the way, saying "I'm on HEAD" is redundant with Git and does not mean what you wanted to say	14:05
	rking: you want to say you're at the tip of a given branch, usually the development one	14:05
rking	thiago: Oh? How do I say that right?	14:05
	OK, cool.	14:05
thiago	rking: you're always on HEAD because HEAD is what you have checked out	14:05
wdouglass	lol	14:06
rking	A lot of people get that wrong, thank you for correcting me.	14:06
	→ marhaban joined	14:06
rking	So what's the shorthand for it?	14:06
	"on tip"?	14:06
	← nioncode left	14:06
rking	"I'm using tip git"?	14:06
Nevik	on the dev branch	14:06
rking	Nevik: That doesn't say where on that branch one is	14:06
thiago	just say the name of the branch	14:06
Nevik	rking: naturally you would be on the tip of it	14:06
	unless you specify something else	14:06
rking	OK	14:06
Nevik	rking: technically, a branch only identifies a tip, heh	14:06
rking	So for other projects, would it be like, "I'm on tig master"?	14:06
Nevik	so if you wanna be nit-picky, saying youre on the branch, means your on the tip	14:07
rking	Cool	14:07
Nevik	rking: no, just "i'm on master"	14:07
	← iamjarvo left	14:07
	← GeertB left	14:07
Nevik	rking: a case in which you could be on an out-dated branch is that you havent pulled latest changes yet (i.e. remote branch has moved, but you dont know yet)	14:08
	→ nioncode joined	14:08
rking	Let me put it this way, can this be golfed down?: /j foosoft "Hi. I'm on foosoft's master branch…"	14:08
Nevik	yes	14:08
	that would assume that before asking the question, youve made sure you actually have the latest version of it (i.e. doing a fetch and possibly a merge)	14:08
	→ jlg13 joined	14:09
Nevik	or specified a commit SHA if you have a good reason not to update	14:09
	specifying*	14:09
rking	Wait, you're saying it is correct? (Not that it can be golfed down?)	14:09
	→ myx joined	14:09
Nevik	yeah thats correct	14:09
	→ _Vi joined	14:09
Nevik	it's what ive been telling you ò.ó	14:09
rking	But I'm asking if it can be said more briefly	14:09
	→ GeertB joined	14:09
Nevik	what could possibly be any shorter	14:10
rking	"I'm on foosoft HEAD"	14:10
Nevik	"i'm on master" is about as short as it gets	14:10
rking	Ok, so it's not necessary to say the "branch"	14:10
Nevik	rking: if you join a project's channel, it would be assumed you mean THEIR master	14:10
	Suprano → Marenz	14:10
Nevik	rking: with master, no; it is general convention to name the main branch master, but not required	14:11
	also, due to the state of the english language, "master" is not often used in general conversation	14:11
	→ trungqtran joined	14:11
Nevik	so it should generate enough context to infer your meaning :P	14:11
	→ rideh joined	14:11
rking	OK. Thank you both for fixing my terminology on this.	14:11
	→ m0ltar joined	14:12
Nevik	sure, no problem	14:12
	→ axrfnu joined	14:12
	← sylr left	14:14
	→ sylr joined	14:14
	→ zivester joined	14:14
	← marhaban left	14:18
	→ Pugnax joined	14:18
	→ marhaban joined	14:19
	→ ehsan joined	14:21
	← cagedwisdom left	14:21
	→ hellopat joined	14:22
	← yshh left	14:22
	→ AaronMT joined	14:22
	← kermit left	14:24
mastro	just curios.. If I had to create a commit message with the # symbol as first character of a line, would that be possible ? usually # are ignored	14:24
	→ adamben joined	14:25
	← adamben left	14:25
	→ adamben joined	14:25
	→ TeckniX joined	14:25
rking	mastro: Hehe, under what circumstances would you 'have to' do that? But have you tried the commit -m '# …' form?	14:25
	→ freshpixel joined	14:26
	← robustus left	14:26
	→ kermit joined	14:26
	← rob_w left	14:27
	→ rob_w joined	14:27
	→ JaredMiller joined	14:27
	← polymar left	14:27
	→ alunduil joined	14:28
	← tjadc left	14:28
	→ robustus joined	14:29
	← arkiver left	14:30
	→ geekbri joined	14:30
	← alunduil left	14:30
	← rideh left	14:31
	← marhaban left	14:31
thiago	mastro: git commit --cleanup=verbatim	14:31
	← JaredMiller left	14:31
	→ alunduil joined	14:32
	← fisted left	14:32
	→ senny joined	14:33
	→ fisted joined	14:33
	← juvenal left	14:33
	→ minusthetiger joined	14:34
	← rgr left	14:34
	← Guest59987 left	14:34
	→ gusnan joined	14:34
	→ kukks joined	14:35
	← kukks left	14:35
	→ kukks joined	14:35
	→ BadProphet joined	14:36
	← jackh left	14:36
	→ jackh joined	14:36
	→ bartek_ joined	14:37
	→ JaredMiller joined	14:37
	← jlg13 left	14:39
	→ spuz joined	14:40
	→ i42n joined	14:40
	← fisted left	14:42
	→ fisted joined	14:42
	← RandalSchwartz left	14:43
	→ jn_ joined	14:44
	→ nitrix joined	14:45
	← xiangfu left	14:45
	→ Zenopus joined	14:46
mastro	thiago, tnx	14:47
	→ dvsharp joined	14:47
mastro	rking, thiago replied me.. anyway I would never want to do that... but since many developers just use Eclipse or graphical tool to write commits (which do not strip out # lines) it may happen to find a narrow-minded team that set up as a must-follow comment structure with the bug reference written as #<number> as first word of the commit message :) I just like to be prepared if something like that ever happen to me	14:48
	→ iamjarvo joined	14:49
	→ tsdh joined	14:50
	← o] left	14:50
	← devsharpen left	14:50
	→ o] joined	14:51
	← GeertB left	14:51
	← v0n left	14:52
	← jackh left	14:52
	→ macabre joined	14:52
	← thiago left	14:54
	← chaiz left	14:54
	→ juvenal joined	14:56
	← suy left	14:59
	← BadProphet left	15:00
	← soee left	15:00
	→ L-Chymera joined	15:02
L-Chymera	hi guys, what'S wrong with git if I get this error? laptophost sci # git pull official master error: cannot open .git/FETCH_HEAD: Read-only file system	15:03
	→ suy joined	15:03
	← ehsan left	15:03
	→ sbodo joined	15:04
bremner	are you perhaps on a Read-only file system?	15:04
	→ fly9 joined	15:04
	→ BadProphet joined	15:05
	→ samuel02 joined	15:05
	→ RaceCondition joined	15:06
L-Chymera	bremner: not that I know of, I'm logged in as root on my laprop	15:07
	→ NiteRain joined	15:08
bremner	network file system?	15:08
	→ normanrichards joined	15:09
	→ guampa joined	15:09
	→ acl_ joined	15:09
	← normanrichards left	15:09
	← OOPMan left	15:09
L-Chymera	bremner: evverything except the repo from where I want to pull is local	15:10
	← ThomasLocke left	15:12
	← Hounddog left	15:12
	→ zoned joined	15:12
	← benbergman left	15:13
	→ vpetro joined	15:13
	→ omani joined	15:14
omani	I want to do backups of all my configuration files on all productive systems. I thought about git as a backup/recory tool	15:14
	← i42n left	15:14
	→ infynyxx joined	15:14
	→ jfsthesecond joined	15:15
omani	like: git on / and change files. add them, and only them, in stage. and commit.	15:15
	← NiteRain left	15:15
omani	then copy .git to NAS/SAN server for backup.	15:15
	now everytime I want to recover a system I just have to clone it into /	15:15
	on the system	15:15
	→ d2dchat joined	15:15
	→ normanrichards joined	15:16
omani	what do y'all say 'bout this idea?	15:16
	→ polymar joined	15:16
	→ minusthetiger_ joined	15:16
	→ Hounddog joined	15:17
	← kermit left	15:17
	← L-Chymera left	15:17
	→ dfaure joined	15:17
dfaure	why does "git diff ." work, but not "git diff .." ?	15:18
	← jfs left	15:18
	← minusthetiger left	15:18
	minusthetiger_ → minusthetiger	15:18
	← normanrichards left	15:18
	→ benbergman joined	15:18
	← jfsthesecond left	15:18
	→ JoeAngel_ joined	15:18
wereHamster	dfaure: define 'not work'	15:19
	→ jfs joined	15:19
dfaure	fatal: ambiguous argument '..': both revision and filename	15:19
	oh, I see	15:19
wereHamster	dfaure: are you in a subdirectory of the repository?	15:19
dfaure	yes	15:19
	git di -- .. works	15:20
wereHamster	dfaure: what's your git version?	15:20
FauxFaux	And git diff .. --	15:20
dfaure	1.7.7	15:20
	FauxFaux gets the same on 1.7.10.4.	15:20
dfaure	I forgot that .. also means something as a revision range	15:20
	→ skyf joined	15:20
dfaure	not sure what it actually means though, with nothing on either end	15:20
	← m0viefreak left	15:20
FauxFaux	Quite possibly means HEAD..HEAD, which is bizarre.	15:21
wereHamster	maybe it got changed, because I don'tg get any error when git diff ..	15:21
	← hafeez3000 left	15:21
wereHamster	with 1.8	15:21
FauxFaux	Indeed it does.	15:21
dfaure	that's never going to lead to a very interesting diff :-)	15:21
	ok if it's fixed in 1.8 then all is well :-)	15:21
	← RaceCondition left	15:21
bremner	omani: maybe look at !etckeeper	15:21
gitinfo	omani: etckeeper is a collection of tools to let /etc be stored in a git, mercurial, darcs, or bzr repository. It hooks into various package managers. http://kitenet.net/~joey/code/etckeeper/	15:21
	← zivester left	15:21
	→ RaceCondition joined	15:22
	← mikecmpbll left	15:22
	← Pugnax left	15:22
	← nkuttler left	15:22
	→ Fingolfin joined	15:23
	→ laurav joined	15:23
omani	bremner: thank you	15:23
	→ Pugnax joined	15:23
omani	this is the right tool I need	15:23
	→ barq joined	15:24
	→ strk joined	15:24
FauxFaux	wereHamster: dfaure: rev-parse was basically rewritten in bc5fd6d3 (v1.7.11-rc0~10^2~13), and has lots of code to deal with ".." now.	15:24
	→ fommil joined	15:24
	gitinfo set mode: +v	15:24
fommil	hi all, my git has magically started spitting this out: "fatal: The upstream branch of your current branch does not match the name of your current branch."	15:24
	→ i42n joined	15:25
strk	I'm in the middle of a conflict, is there a command to just pick modifications of "head" for a given file ?	15:25
FauxFaux	strk: git checkout HEAD path/to/file	15:25
fommil	I have origin, upstream and my local. Only one "master" on each.	15:25
	← benbergman left	15:25
wereHamster	strk: git checkout --ours <file>	15:25
	→ kermit joined	15:25
	← eijk left	15:26
	→ subhojit777 joined	15:26
	← jfs left	15:27
	← Pugnax left	15:27
dfaure	FauxFaux: excellent, thanks	15:28
	→ steve6 joined	15:28
	→ huoxito joined	15:28
	← dfaure left	15:28
	→ EugenA joined	15:28
omani	bremner: what if I want to this etckeeper thing on a remote NAS/SAN?	15:29
EugenA	how to merge two branches in bare repository?	15:29
bremner	omani: it's just git. So if you know how to answer that question for git...	15:29
	← BadProphet left	15:29
	← revyd left	15:29
FauxFaux	EugenA: Clone it, merge them, then push the result back.	15:29
	→ revyd joined	15:30
EugenA	ok	15:30
	→ pansdr joined	15:32
	→ benbergman joined	15:32
	→ axl_ joined	15:33
	← fommil left	15:33
	← revyd left	15:35
	← zeppo left	15:36
	→ mikecmpbll joined	15:36
	→ zeppo joined	15:36
	→ xmas88 joined	15:36
	← xmas88 left	15:37
	→ BooK joined	15:37
	← pansdr left	15:37
	gitinfo set mode: +v	15:37
	← poolshark left	15:37
	← tsdh left	15:39
	← _Qman left	15:40
	← alkino left	15:40
	← luc_ left	15:40
	← benbergman left	15:41
	→ unnamedwill joined	15:41
	→ maleknet joined	15:42
	→ kevwil joined	15:42
	→ tsdh joined	15:43
	← papegaaij left	15:43
	→ GeertB joined	15:44
	→ alkino joined	15:44
	→ barq1 joined	15:44
	← kennethreitz left	15:44
	← maleknet left	15:45
	→ maleknet joined	15:45
	← delirium left	15:45
	← barq left	15:46
	→ kennethreitz joined	15:47
	← iamjarvo left	15:48
	← adhawkinz left	15:49
	→ gtrak joined	15:49
	← tsdh left	15:50
	→ benizi joined	15:50
	← maleknet left	15:50
	→ iamjarvo joined	15:51
	→ x82_nicole joined	15:52
	← minusthetiger left	15:53
	→ pogo joined	15:53
	← x82_nicole left	15:54
	→ ejd joined	15:54
	→ ehsan joined	15:54
	→ v0n joined	15:54
	← polymar left	15:55
	→ fluxdude joined	15:55
	→ minusthetiger joined	15:56
	← rchavik left	15:56
	← hyperair left	15:57
	→ FernandoBasso joined	15:59
	→ andrewhl joined	15:59
	← azgard left	16:00
omani	bremner: yes	16:00
	I don't know the answer yet, but I will figure it out	16:00
	I think I can copy the .git directory to remote machine and clone from it	16:01
	I just wasn't sure about whether to copy the wordking directory or just basre git	16:01
bremner	omani: typically remote repos like that are !bare	16:03
gitinfo	omani: an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/	16:03
	← cbreak-work left	16:03
	← Marenz left	16:04
	→ Suprano joined	16:04
	← zeppo left	16:04
	→ benbergman joined	16:05
	→ nurey_ joined	16:05
	→ joeytwiddle joined	16:06
	← JohnBat26 left	16:06
	← R1KRDO left	16:07
	→ marhaban joined	16:07
	→ normanrichards joined	16:07
	← Gx4 left	16:08
	← jn_ left	16:09
	→ ShooterMG joined	16:10
	→ tsdh joined	16:10
	← freshpixel left	16:10
	← Raging_Hog left	16:11
	→ tomchop joined	16:11
	← blami_ left	16:12
	→ __20h__ joined	16:12
__20h__	Greetings. I have the problem that over a cgit exported repository no tags can be pulled from. The git update-server-info is run as post-update hook.	16:13
	Anyone knows what I am missing?	16:13
wereHamster	__20h__: are the tags shown in git ls-remote?	16:13
	← mp_ left	16:14
__20h__	wereHamster, no.	16:14
	Only refs/heads/master and HEAD is shown.	16:14
	← kevwil left	16:15
	← strk left	16:15
	← FernandoBasso left	16:15
	← m0ltar left	16:16
	→ FernandoBasso joined	16:16
	← flijten left	16:16
	→ kevwil joined	16:17
	→ danillonunes joined	16:17
BooK	I was considering using something modeled around the git object database to store "extended objects" (e.g., tree with uid/gid info for the file)	16:17
	→ m0ltar joined	16:17
BooK	shouldn't be a problem with loose format, since the type is a string in the header (format being "<type> <size>\0<content>")	16:17
	but for packs, if I understood correctly, the type is stored on 3 bits, and all	16:17
	values are already used	16:17
	← benbergman left	16:17
wereHamster	BooK: man git-notes	16:18
gitinfo	BooK: the git-notes manpage is available at http://jk.gs/git-notes.html	16:18
	← i42n left	16:18
	← rath_ left	16:18
	← e-dard left	16:19
grundt	oh, cool - you could use that to add TODOs and FIXMEs?	16:19
	← krosenvold left	16:19
BooK	wereHamster: should I understand that as "store that extra info as a note attached to the tree object" ?	16:21
wereHamster	to any object	16:21
BooK	uh uh	16:22
	← samuel02 left	16:22
__20h__	wereHamster, hm, the refs/tags directory in the bare repository is empty.	16:22
	They are only mentioned in the info/refs file.	16:22
wereHamster	__20h__: don't rely on that. Use git commands, such as git show-ref or git for-each-ref	16:22
	← kennethreitz left	16:22
	← tomchop left	16:23
BooK	wereHamster: thanks, I'll do some research on notes	16:23
wereHamster	__20h__: you should not poke around the .git directory.	16:23
	→ tomchop joined	16:23
	← jceb left	16:23
__20h__	wereHamster, yeah, but I thought that the http pull needs some more files available.	16:23
	→ bLUEE joined	16:23
	gitinfo set mode: +v	16:24
	→ chuckharmston joined	16:24
bLUEE	PLEASE HELP!	16:25
wereHamster	NO!	16:25
	→ asaf joined	16:25
bLUEE	i had a project, and added the remote, did a pull/merge	16:25
	← chepre left	16:25
bLUEE	and then all my new code disappeared	16:25
	then i tried to undo merge	16:25
	→ kennethreitz joined	16:25
bLUEE	and now i dont know what's going on :/	16:26
	→ i42n joined	16:26
wereHamster	I don't know either.	16:26
bLUEE	i commitd my work - so how do i get back to that one?	16:26
	→ thl0 joined	16:26
wereHamster	git log	16:26
mgedmin	bLUEE, do you still have that terminal open? can you pastebin the entire session (starting with pull)?	16:26
bLUEE	mgedmin: oh man, i cant, since the terminal window stalled and i crashed it out and opened a new one !	16:27
	← Hounddog left	16:29
	→ joshsmith joined	16:29
	← cakehero left	16:30
mgedmin	well, if you had your code committed, it's still there	16:30
	you just need to find it	16:30
	← tomchop left	16:30
bLUEE	wereHamster: how do i page in git log ?	16:30
mgedmin	git reflog might help	16:30
	→ OOPMan joined	16:30
bLUEE	mgedmin: okay, i see my commit there..	16:31
	how can i get it back?	16:31
	mgedmin waits for experts to answer	16:31
	→ tomchop joined	16:32
	→ tmcmahon joined	16:32
	← h0bbit left	16:32
	→ Davey joined	16:32
mgedmin	it'll be probably something like git reset --hard commithash, after which you should repeat your pull + merge, except more carefully	16:32
	← xxc_michael left	16:33
spuz	bLUEE: try typing git reflog	16:33
	→ argentino joined	16:34
spuz	oh you said you found your lost commit	16:34
	← acl_ left	16:34
spuz	bLUEE: are you on a personal or a shared branch?	16:34
	→ kenperkins joined	16:35
bLUEE	when i do git reflog i see some entries	16:35
	← OOPMan left	16:35
bLUEE	http://pastebin.com/v0ub1D8W	16:36
	→ BadProphet joined	16:36
bLUEE	spuz: shared	16:36
	← argentino left	16:36
	← Bombe left	16:36
	← FernandoBasso left	16:36
bLUEE	HEAD@2 is my commit, if i can get that - then thats all good :/	16:36
	→ argentino joined	16:36
spuz	bLUEE: ok, just type 'git reset --hard 5c27441'	16:37
	→ polymar joined	16:38
	← tomchop left	16:38
	← argentino left	16:38
	→ jlg13 joined	16:38
bLUEE	spuz: !!!!!!!!!! thank you	16:39
	→ revagomes joined	16:39
	→ eijk joined	16:39
	← Davey left	16:40
	→ ajeffco joined	16:41
spuz	bLUEE: no problem. Not sure how you lost your changes though. if you merge with a branch, you should receive a conflict warning before losing any of your changes	16:41
bLUEE	i originally creted the project and we had been working on git repo	16:41
	→ swiety joined	16:42
bLUEE	then i recreated the project then added the remote and tried to push, but i did a pull/merge	16:42
	and then all my crap went	16:42
	← kennethreitz left	16:42
	← bauruine left	16:42
	← mneorr left	16:44
	← EugenA left	16:45
	← svetlyak40wt left	16:45
	→ kennethreitz joined	16:45
	→ benbergman joined	16:46
	→ cakehero joined	16:47
	← tsdh left	16:47
	→ OOPMan joined	16:48
	← TheHaven left	16:50
	← kennethreitz left	16:50
	← LucasTizma left	16:50
	→ ampz joined	16:52
	→ a_L_p joined	16:53
	← ampz left	16:54
	← a_L_p left	16:54
	→ kennethreitz joined	16:55
	← sylr left	16:57
	→ tomek joined	16:57
	→ Pugnax joined	16:57
	→ hafeez3000 joined	16:58
	← kennethreitz left	16:59
	→ dankest joined	17:00
	→ Bombe joined	17:00
	→ Davey joined	17:01
	← angelsl left	17:02
	← psoo left	17:03
	→ kennethreitz joined	17:04
	→ torjus_ joined	17:04
	→ aselder joined	17:05
	← aselder left	17:05
	→ aselder joined	17:06
	← aselder left	17:06
	← wdouglass left	17:08
	→ arietis joined	17:09
	← jlopez left	17:09
	← polymar left	17:09
	→ witul joined	17:09
	← witul left	17:10
	← tdowg12 left	17:10
	← jlg13 left	17:10
	← d1b left	17:10
	← MiKom left	17:10
	← zenx left	17:10
	← achadwick left	17:10
	← zeks left	17:10
	← quazimodo left	17:10
	← txomon\|nqas left	17:10
	← smw left	17:10
	← karstensrage left	17:10
	← mastro left	17:10
	← Pieter left	17:10
	← notfunk left	17:10
	← jds_ left	17:10
	← sjk left	17:10
	← yhager left	17:10
	← pielgrzym left	17:10
	← eightyeight left	17:10
	← bapa left	17:10
	← Gizmokid2005 left	17:10
	← c98_ left	17:10
	← mgedmin left	17:10
	← poseidon left	17:10
	← heytrav_ left	17:10
	← setmeaway2 left	17:10
	← reikalus1kka left	17:10
	← mattp_ left	17:10
	→ d1b joined	17:10
	→ reikalus1kka joined	17:10
	→ heytrav joined	17:10
	→ Pieter joined	17:10
	→ MiKom joined	17:10
	→ sjk joined	17:10
	→ jds joined	17:10
	→ pielgrzym joined	17:10
	→ quazimodo joined	17:10
	→ notfunk joined	17:10
	→ jlg13 joined	17:10
	→ yhager joined	17:10
	→ c98_ joined	17:10
	→ poseidon joined	17:10
	← notfunk left	17:11
	→ achadwick joined	17:11
	→ zeks joined	17:11
	→ mattp_ joined	17:11
	→ txomon\|nqas joined	17:11
	→ setmeaway2 joined	17:11
	→ notfunk joined	17:11
	→ eightyeight joined	17:11
	→ Guest43947 joined	17:11
	→ mastro joined	17:11
	← marhaban left	17:12
	→ geekbri_ joined	17:12
	→ smw joined	17:12
	→ andrewhl_ joined	17:12
	→ bapa joined	17:12
	→ Hounddog joined	17:12
	→ mgedmin joined	17:12
	→ martinjl` joined	17:13
	→ odin_ joined	17:13
	→ mikef joined	17:14
	→ Gizmokid2005 joined	17:14
	→ x82_nicole joined	17:14
	→ Shusshu joined	17:15
	← Shusshu left	17:15
	→ sharpobj1ct joined	17:16
	← sharpobject left	17:16
	→ juril joined	17:16
	← andrewhl left	17:16
	← unreal left	17:17
	→ tjbiddle joined	17:17
thomasf1021	Good afternoon, what's the best way to just the changes made to a branch since it was created? In SVN I was able to essentially say merge the changes made since branching and apply it to another branch.	17:17
	← geekbri left	17:18
	← martinjlowm-lapt left	17:18
	← Quix0te left	17:18
	→ Quix0te joined	17:18
	← kennethreitz left	17:18
FauxFaux	thomasf1021: "to just the changes"?	17:19
wereHamster	thomasf1021: git doesn't have a concept of 'a branch was created'. Or the ability to merge only selected commits from a range of commits.	17:19
	→ kennethreitz joined	17:20
	→ psoo joined	17:21
	→ unreal joined	17:21
	→ Goplat joined	17:21
	→ zenx joined	17:22
	← infynyxx left	17:22
thomasf1021	wereHamster: thanks	17:22
	FauxFaux: I meant "to just apply the changes"	17:22
	→ _iron joined	17:22
	→ jceb joined	17:23
	→ itatitat joined	17:23
	→ Ontology joined	17:24
	→ tdowg12 joined	17:25
	→ nitrix_ joined	17:26
EugeneKay	Look into cherry-pick	17:26
	← mikef left	17:27
deryni	If you want all the changes on an unmerged branch the rev..rev syntax can get you that easily, or did I misunderstand something.	17:29
	→ devth joined	17:29
	← nitrix left	17:29
	← surfdue left	17:29
	← tjbiddle left	17:29
	← alkino left	17:29
	← Desproges left	17:30
	← asaf left	17:30
	← mikecmpbll left	17:31
	→ kerrick joined	17:31
	→ lindenle joined	17:32
	→ surfdue joined	17:32
	→ jfs joined	17:33
	← jfs left	17:33
	→ jfs joined	17:33
	← aisbaa left	17:33
	← _iron left	17:34
	→ _iron joined	17:34
	→ leeN joined	17:35
	← LimeBlast left	17:35
	← i42n left	17:35
	← emankcin left	17:35
	← psoo left	17:36
	→ grizwako joined	17:36
	→ R1KRDO joined	17:36
	← jgomez left	17:36
	→ Serghar joined	17:37
	← kennethreitz left	17:37
	← nigelb left	17:38
	→ nyxynyx joined	17:38
	→ tenmilestereo joined	17:38
	→ tomchop joined	17:38
	← tmcmahon left	17:38
	→ SBoudrias joined	17:38
	← SBoudrias left	17:39
	→ wermel joined	17:39
nyxynyx	Hi I have a question: I created a branch 'v1' on my local system, and want to push the files to github, how should it be done? I tried `git branch v1` `git checkout v1` `git push --set-upstream origin/v1 v1` but i get the error mmsg saying that its not a git repo and it can't read from remote repo	17:40
	→ nigelb joined	17:41
	← jlg13 left	17:42
	← torjus_ left	17:42
	→ kennethreitz joined	17:43
	← rurufufuss left	17:43
	→ Sonderblade joined	17:43
	← Sonderblade left	17:43
	→ Sonderblade joined	17:44
	← mishok13 left	17:44
	← kevwil left	17:45
	→ johnanderson joined	17:45
	→ mneorr joined	17:45
	← john_anderson left	17:45
	johnanderson → john_anderson	17:45
EugeneKay	nyxynyx - 'git push -u origin v1`	17:46
nyxynyx	works thanks!	17:46
EugeneKay	You just had the wrong arguments is all	17:46
	← giallu left	17:46
	→ cirwin joined	17:46
	→ jlg13 joined	17:47
	→ Jippi joined	17:47
	← itatitat left	17:48
	← john_anderson left	17:49
	→ mikef joined	17:50
	← dpino left	17:51
	→ etcetera joined	17:52
	← bux left	17:52
	→ m0viefreak joined	17:52
	← kerrick left	17:54
	→ alexim joined	17:55
	← jlg13 left	17:55
	← alexim left	17:55
	← axl_ left	17:56
	← spuz left	17:56
	← kennethreitz left	17:56
	→ jlg13 joined	17:56
	→ metcalfc joined	17:56
	→ nickoe_ joined	17:56
	← letharion left	17:57
	→ axl_ joined	17:57
	→ JohnBat26 joined	17:58
	← JohnBat26 left	17:58
	→ JohnBat26 joined	17:59
	→ kennethreitz joined	17:59
	← nioncode left	18:00
	← dgtlmoon left	18:00
	→ dermoth\|home joined	18:01
	← thl0 left	18:02
	← rue_XIV left	18:02
	← alunduil left	18:03
	→ dlam joined	18:03
	→ tjbiddle joined	18:03
	← kennethreitz left	18:04
	← bnovc left	18:04
	→ lresende joined	18:06
	← tomchop left	18:07
	→ kennethreitz joined	18:07
	← caddoo left	18:08
	→ bibinou joined	18:08
	→ Sigma joined	18:09
	← metcalfc left	18:11
	← dlam left	18:11
	← adamben left	18:11
	← surfdue left	18:12
	← ergZay left	18:13
	← axl_ left	18:13
	→ surfdue joined	18:13
	← senny left	18:14
	→ kerrick joined	18:15
	→ adamben joined	18:15
	← adamben left	18:15
	→ adamben joined	18:15
	← suy left	18:15
	→ rgr joined	18:16
	← dvide left	18:16
	→ d3c joined	18:16
	← bronson left	18:16
	← bitkiller left	18:17
	→ sluther joined	18:17
	→ asaf joined	18:18
	→ bitkiller joined	18:20
	← m0ltar left	18:20
	→ Squarism joined	18:21
	→ m0ltar joined	18:22
	→ letharion joined	18:24
	← m0viefreak left	18:25
	→ hwrd\|work joined	18:25
	← hwrd\|work left	18:27
	→ i42n joined	18:27
	← narfnarfnarf left	18:27
	← timj left	18:27
	→ timj1 joined	18:27
	← subhojit777 left	18:30
	← achadwick left	18:30
	→ rednaks joined	18:31
	→ brayn joined	18:34
	→ las joined	18:34
las	good day. i just built 1.8.0.1 on a PowerPC Mac, and running "make test" reveals 1 failed test. should i care?	18:34
EugeneKay	No.	18:35
PerlJam	las: if it's the test for "git clone" or "git checkout", yes ;)	18:35
	← bodom left	18:35
las	* t1402-check-ref-format.sh * => not ok - 31 ref name 'heads/foo' is invalid	18:35
	dankest → dankest\|away	18:35
EugeneKay	Isn't there a !packaged version available? :-p	18:36
gitinfo	Unless you have a specific technical need for the "latest and greatest" you should go with a packaged version. Compiling and installing your own is not something that we will walk you through or troubleshoot (though it is fairly easy for developers with GNU/Linux experience).	18:36
	→ krosenvold joined	18:37
las	EugeneKay: not for PPC Macs running Tiger	18:38
	← Aikar left	18:39
EugeneKay	You poor git. Get a job; buy computer manufactured in the right decade.	18:39
	← robotmay left	18:39
las	EugeneKay: err, excuse me	18:39
EugeneKay	:-p	18:39
	→ ChrisBorgia joined	18:39
EugeneKay	In seriousness though, that doesn't sound like a terribly fatal issue. It's probably a borked test with 1.8.x	18:40
	→ bauruine joined	18:40
milki	well, if things break. you could always post to the mailing list	18:40
	im sure there are more powerpc users there	18:40
	EugeneKay: i have a powerpc powerbook too!	18:40
	← tomek left	18:40
milki	compiling anything for it is a pain	18:40
EugeneKay	milki - I have a shotgun you can borrow	18:40
	← poisond left	18:40
milki	haha	18:40
	it looks nicer than my lenovo laptop though	18:41
	→ poisond joined	18:41
milki	and the battery lasts longer!	18:41
	← cakehero left	18:41
	← jlg13 left	18:41
las	wow, no make uninstall ? seriously?	18:41
milki	a lot of projects dont come with uninstall	18:41
	→ buibex joined	18:42
EugeneKay	--prefix=/usr/local helps	18:42
	← Zenopus left	18:42
las	milki: but i thought this was a new decade :)	18:43
milki	laziness will never be solved with time	18:43
las	EugeneKay: well, that's the thing. i used makefile prefix=/usr/local as mentioned in INSTALL. it appeared to install into ~	18:43
milki	otherwise, we'd be out of jobs	18:43
	← wermel left	18:44
EugeneKay	I've honestly never touched git's build scripts.	18:44
las	EugeneKay: me neither, until today. i had a package version of 1.5 but that started to have issues today	18:44
	← cirwin left	18:44
	→ Bootstrapper joined	18:44
	→ les_sylvains joined	18:45
	← nyxynyx left	18:45
	→ tomek joined	18:47
	← tomek left	18:47
	→ tomek joined	18:47
	← tomek left	18:47
	→ tomek joined	18:47
	→ hwrd\|work joined	18:49
	← etcetera left	18:50
	← blami left	18:51
	→ otherj joined	18:51
otherj	when i use git rm, does that just remove the file from what git tracks going forward, or does that remove the file from previous commits as well?	18:51
FauxFaux	Just from going forwards.	18:52
otherj	ok, thanks. I was just doing that for a bunch of files i merged into one and realized it would be a bummer if i lost the history of them before merging :P	18:53
	you're aces FauxFaux	18:53
	→ absynce joined	18:53
	→ Adget joined	18:53
	← rednaks left	18:54
	→ bronson joined	18:54
	← TeamColtra left	18:54
	→ etcetera joined	18:55
	← fly9 left	18:55
	← benbergman left	18:55
	← etcetera left	18:56
	← otherj left	18:56
	→ blami joined	18:57
EugeneKay	otherj - remember that git tracks trees, not files. All you're doing with git-add/rm/etc is creating new trees based upon your filesystem.	18:57
	→ bux joined	18:57
EugeneKay	Unless you rewrite your history it isn't going anywhere	18:57
	← Serghar left	18:57
	← las left	18:59
cmn	it's also the whole reason you use version control	18:59
	← babilen left	18:59
	← zenx left	18:59
	→ etcetera joined	18:59
	→ jayallen joined	19:01
	→ msch joined	19:02
	← Leeol2 left	19:02
	→ Leeol joined	19:02
	→ benbergman joined	19:03
Leeol	Question.. if you use `git update-index --assume-unchanged` on a file, does that get stored and shared to other repos properly? Or is that a local repo only setting?	19:03
	brycec → ^_^	19:03
	← rgsteele left	19:03
	^_^ → brycec	19:04
	← TeckniX left	19:04
	← bibinou left	19:04
	→ cakehero joined	19:04
	→ metcalfc joined	19:04
	→ TeckniX joined	19:05
	Praise- → Praise	19:05
	→ dgtlmoon joined	19:06
	← cakehero left	19:07
drizzd	Leeol: local only, index only	19:07
Leeol	drizzd: Hmm, i was afraid of that. Do you recommend anything similar that is tracked? For skeleton files / config files / etc	19:08
	← Quix0te left	19:08
	→ Quix0te joined	19:08
drizzd	!config	19:08
gitinfo	[!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas	19:08
	← i42n left	19:08
	← thierryp left	19:09
Leeol	So yea, nothing really :/ (aside from not tracking it at all)	19:09
msch	hi, got a question about merging two repositories. So I have to repos A and B and I want to merge them into a new repo C where the contents of A is in subdir A and the contents of B is in subdir B. so far so good, i can do that. now i also want to rewrite the history of these repositories so that i can do git log / blame without --follow. i can do that too.	19:09
drizzd	see also https://git.wiki.kernel.org/index.php/GitFaq#How_do_I_tell_git_to_ignore_tracked_files.3F	19:09
msch	but i have a lot of branches, how can i preserve the different branches so that the tree of branches still looks the same?	19:10
drizzd	Leeol: no, there is no way to configure git to automatically globally ignore changes to tracked files	19:10
	← nickoe_ left	19:10
msch	e.g. branch master and branch feature-x in repo A should later be master and feature-x in repo C but the commits in feature-x should still be on top of master	19:10
drizzd	msch: git filter-branch --all	19:11
	→ IridiumScaffold joined	19:11
msch	drizzd: ah so that doesn't create two completely different commit trees for each branch?	19:12
drizzd	no	19:12
msch	cool	19:12
	thanks	19:12
drizzd	the histories of the two repos are still separate, of course. No way to fix that.	19:12
	dankest\|away → dankest	19:12
	← adamben left	19:12
	← bronson left	19:13
msch	drizzd: yeah, but then i add them both to one repo as remotes, fetch from them and git merge. right?	19:13
drizzd	yes	19:13
	→ infynyxx joined	19:13
	← normanrichards left	19:14
	← shruggar left	19:14
	→ travisgriggs joined	19:15
	← d3c left	19:15
	→ aetcore joined	19:16
	→ i42n joined	19:17
	→ x74beers joined	19:17
	→ maleknet joined	19:18
	→ wilsonj joined	19:18
travisgriggs	a colleague and i are trying to learn (somewhat unsuccessfully) to work together using git. we have a smallish project, a makefile and about 20 .c/.h files. i'm making some local changes in a .c file. he skypes me and tells me he made a change to the makefile. i go to do a pull, but get told that "Your local changes to the following files would be overwritten by merge:". so i assume he changed more than just the makefile. does this mean i shouldn't be usin	19:18
	gitinfo set mode: +v	19:18
wilsonj	anyone good with gitolite conf files?	19:18
	← nurey_ left	19:19
	← dvsharp left	19:19
	→ john_anderson joined	19:19
wilsonj	http://pastie.org/5454284 - i am part of the buildmasters group and I want to be skipping the VREF check, which is why I have RW+ =@buildmasters above the VREF, but the VREF is still running	19:20
	← kennethreitz left	19:20
cmn	travisgriggs: it means that someone at some point pushed some history which modifies a file you have also modified locally but not committed	19:21
	← maleknet left	19:23
	← Vile left	19:23
	→ kennethreitz joined	19:24
travisgriggs	cmn: yes i get that, he changed the same.c file that i did. the question is what is the standard work flow for dealing with that. should i commit first then pull?	19:25
cmn	travisgriggs: some history meaning history you don't have locally	19:25
	→ jpieper joined	19:25
	→ flijten joined	19:25
cmn	you don't merge until you need to	19:25
	is there some big reason why you need that makefile change?	19:25
	← jceb left	19:25
	→ cirwin joined	19:25
deryni	wilsonj: I think you probably need an extra VREF pass line for that not a regular rule for that (no I'm not at all sure).	19:26
travisgriggs	yes, i want it. he added a feature i asked him to add to it, so i could see code space listings, and when i get it, i can do what i need to	19:26
cmn	but how is that related to the change you were making without it?	19:26
wilsonj	ok thx, i'll try that	19:26
	→ thl0 joined	19:28
cmn	or rather, is his change a necessary prerequisite for you to commit your changes?	19:28
	← Xethron left	19:28
travisgriggs	in the .c file you mean? i'm changing main.c, i needed his makefile change to emit code space values, so that i could get a feel for how my changes are going to pan out from a code space pov. is the real problem that he just committed all of his changes, instead of being more structured/controlled about it?	19:29
	← Tegi left	19:29
cmn	there is no way to know the anwer to that question	19:29
	look at what he did	19:29
travisgriggs	i know what he did, i can ask him. i think i'm not communicating my question clearly, sorry :(	19:30
cmn	you don't need to ask anything	19:30
	look at the history he pushed up	19:30
	then you can have a definitive answer of what's there	19:30
	→ d3c joined	19:31
	→ dvsharp joined	19:32
	← dankest left	19:32
travisgriggs	in an svn world, i would have said "hey Dan, can you commit your makefile changes", he would have svn ci'd and thrown some of his in progress tweaks to main.c along for the ride. i would have then done an svn up and it would have given me a new makefile and merged his main.c changes in with mine (provided it could be done so without a conflict). i'm trying really hard not to bring that kind of mental baggage with me, but i'm trying to figure out what the c	19:32
	→ dankest joined	19:33
cmn	your messages are getting cut	19:33
	→ emankcin joined	19:33
travisgriggs	i'm trying really hard not to bring that kind of mental baggage with me, but i'm trying to figure out what the common commands one uses to get a similar fluid workflow between the two of us	19:33
cmn	if he committed to main.c and you've local changes, then that'll be why git is telling that's what happened	19:33
milki	travisgriggs: !vcbe might be good	19:34
gitinfo	travisgriggs: 'Version Control By Example' gives a good overview of the different VCSes available. The author will even mail you a dead-tree copy for free. http://ericsink.com/vcbe/	19:34
	← bux left	19:34
	→ polymar joined	19:34
travisgriggs	i just want it to merge the two of them like svn up would have :( i don't want it to tell me there are changes. i knew that already	19:34
cmn	then either commit your changes or stash	19:35
	→ maleknet joined	19:35
cmn	and merge his changes when you're ready to proceed	19:36
	→ ^Mike joined	19:38
ChrisBorgia	ChrisBorgia: <- Digging the new iTunes	19:38
cmn	you can rebase the latest commit if you really feel that his change should happen before yours	19:38
^Mike	when I do git blame in a git-svn repo, every line is listed as "not committed yet". I gather this is because of something finnicky happening with line endings - how can I see the real blame instead?	19:38
	→ thiago_ joined	19:39
^Mike	in .gitconfig, I have core.autocrlf=input - should I be using something else (on a linux machine)?	19:39
cmn	ask it to start at history, say HEAD	19:40
	← ryez left	19:40
	← benbergman left	19:40
cmn	rather than from the workdir	19:40
	← nitrix_ left	19:40
	→ cakehero joined	19:40
^Mike	excellent, thanks!	19:41
	← cakehero left	19:41
	← kennethreitz left	19:42
	→ jceb joined	19:42
	← chimay left	19:42
	→ kennethreitz joined	19:43
	→ torjus_ joined	19:43
	→ nitrix_ joined	19:43
	← geekbri_ left	19:44
	← kerrick left	19:45
	→ diminoten joined	19:47
	← torjus_ left	19:47
	← maleknet left	19:48
	← jayallen left	19:48
	← minusthetiger left	19:48
	→ mikeit joined	19:48
	→ kerrick joined	19:48
	← Adget left	19:49
	→ daniel_b joined	19:49
	→ Adget joined	19:49
	← diminoten left	19:49
	→ diminoten joined	19:49
	← d3c left	19:49
daniel_b	Is there a way to have git clear the terminal screen between hunks in git add --patch?	19:49
	→ minusthetiger joined	19:50
	← hpa left	19:50
	→ hpa joined	19:50
	dankest → dankest\|away	19:50
	← hafeez3000 left	19:51
	← mikeit left	19:51
	← zoned left	19:52
	→ Adget1 joined	19:52
	→ zoned joined	19:52
	thiago_ → thiago	19:52
	← phlipper left	19:53
	→ Superlokkus joined	19:54
	← Superlokkus left	19:55
	← polymar left	19:55
	← Adget left	19:55
	→ axl_ joined	19:57
^Mike	daniel_b: I've been asking for that for years, and I think it is still not possible	19:58
	→ Chillance joined	19:59
	→ jso- joined	19:59
daniel_b	Bummer.	19:59
^Mike	Truly.	19:59
FauxFaux	It's a shell or perl script, just patch it on live	20:00
	→ geekbri joined	20:01
cmn	it's perl, print "<whatever ^L is>" should do it	20:01
	→ j416 joined	20:01
	← dc5ala left	20:01
	← dvsharp left	20:02
	→ towski joined	20:02
	→ tplaner_ joined	20:02
	← pmdarrow left	20:02
	← absynce left	20:02
	← mneorr left	20:04
	→ benbergman joined	20:04
	← Fingolfin left	20:04
	← infynyxx left	20:04
	→ Samual joined	20:04
	← buibex left	20:06
	← laurav left	20:06
	← les_sylvains left	20:07
	← revagomes left	20:07
	← hwrd\|work left	20:08
	← kennethreitz left	20:09
	dankest\|away → dankest	20:11
	→ shruggar joined	20:12
	← Chaser_ left	20:12
	→ kennethreitz joined	20:12
	← tvw left	20:13
	← rgr left	20:13
	→ nioncode joined	20:13
	→ simesy joined	20:13
	→ infynyxx joined	20:14
	← JohnBat26 left	20:14
	→ dvsharp joined	20:14
	→ pandeiro joined	20:14
	→ alansaul joined	20:14
	→ Vile joined	20:15
	← shruggar left	20:16
	→ Iuz joined	20:17
	→ rgr joined	20:17
	→ mikeit joined	20:17
	← geekbri left	20:18
	→ shruggar joined	20:19
	→ BD- joined	20:20
	← BD- left	20:20
Iuz	Where is the SSH keys my git is using to try and connect with github on my system and where can I reset them?	20:20
	← hinst left	20:20
cmn	wherever your ssh client puts and/or expects them; on unixy systems, under ~/.ssh/	20:21
vpetro	Iuz: ~/.ssh?	20:21
Iuz	on my user ~ right ?	20:22
FauxFaux	Obviously not, as you just checked and they're not there.	20:22
vpetro	/home/your_user_name/.ssh/	20:22
	← Squarism left	20:22
	→ mneorr joined	20:22
cmn	on some systems, sure	20:22
	→ cagedwisdom joined	20:22
Iuz	yeah, there was nothing there	20:22
cmn	which is why we use ~/ or $HOME	20:22
	→ BD- joined	20:22
vpetro	what system are you on?	20:22
	→ Squarism joined	20:22
Iuz	now I mv'd id_rsa there	20:23
	and Its still not working	20:23
FauxFaux	What?!	20:23
cmn	!doesnt	20:23
gitinfo	[!doesntwork] Sorry to hear it doesn't work. What happened? Did it fall asleep on your couch, with your sister? Eat your homework? What did you want it to do? What happened instead? Please be specific!	20:23
FauxFaux	Please ask your real question.	20:23
cmn	this is also a ssh issue, not a git one	20:23
FauxFaux	Hint: It's "I'm trying to connect to github, but it's not using my keys at all."	20:23
	Oh god why does this make me so angry. Off I go again.	20:24
	← FauxFaux left	20:24
	← shruggar left	20:24
cmn	also notice that reset is not something you do with ssh keys	20:24
Iuz	Why bother taking the time to do that? Im only asking a question here	20:24
	did I call his nickname ?	20:25
	← batrick left	20:25
cmn	you're not doing it, though	20:25
	you're dancing around the real issue	20:25
	you haven't explained the actual problem you're trying to solve	20:25
	← m0ltar left	20:25
vpetro	you could also try to generate a new key-pair and see where it written	20:25
Iuz	well, I came around having tried to put my keys on /.ssh/ and they did not work	20:25
	so I asked where was the dir that git used to connect with github	20:26
	→ Serghar joined	20:26
	← __20h__ left	20:26
haarg	git just runs ssh. it uses whatever ssh uses.	20:26
Iuz	and thats it, you guys answered that	20:26
cmn	that is not a real question, either	20:26
	← chris2 left	20:26
	→ giallu joined	20:26
cmn	git doesn't use any dir to connect to github	20:26
	if you're having ssh issues, try to troubleshoot that, but you're mixing a lot of things and not providing information about what you're doing	20:27
	→ m0ltar joined	20:27
	← Squarism left	20:27
	→ AdrienBrault joined	20:27
Iuz	http://pastebin.com/GhSpJHrg	20:28
	can you give me a direction to where the problem might be?	20:28
	Samual → Twat	20:28
	← JoeAngel_ left	20:28
	Twat → Twatual	20:28
cmn	your ssh client isn't providing the right key	20:29
	→ shruggar joined	20:29
vpetro	Iuz: try this: ssh -T [email@hidden.address]	20:30
cmn	ask in an openssh forum or look at the github help pages which show some help	20:30
	← etcetera left	20:30
Nevik	you probably just have the wrong key entered on github (wild guess)	20:30
	→ batrick joined	20:30
Iuz	vpetro, Permission denied (publickey).	20:31
vpetro	Iuz: well, here is the process for setting up ssh keys with github: https://help.github.com/articles/generating-ssh-keys	20:31
	← Twatual left	20:31
	→ Samual joined	20:31
cmn	repeating the error doesn't help; go seek ssh support	20:31
	→ etcetera joined	20:32
	← mneorr left	20:32
	→ cstephens4 joined	20:32
	→ h0bbit joined	20:33
	→ chris2 joined	20:33
	← _Vi left	20:34
	dankest → dankest\|away	20:35
	← tdowg12 left	20:36
	← AdrienBrault left	20:36
	→ mneorr joined	20:36
	← kerrick left	20:36
	← Adget1 left	20:38
	← Davey left	20:38
	→ Davey joined	20:40
	← cstephens4 left	20:41
	→ cstephens4 joined	20:41
	← tplaner_ left	20:41
	← arietis left	20:41
	← BD- left	20:41
	→ tdowg12 joined	20:41
	← lresende left	20:42
	← tdowg12 left	20:42
	→ tdowg12 joined	20:43
	→ lresende joined	20:43
	← Sigma left	20:43
	→ marhaban joined	20:45
cwillu_at_work	how do I update the current branch and index to a different branch, without touching the working tree at all?	20:46
	← tdowg12 left	20:46
cwillu_at_work	(i.e., pull, without touching the working tree)	20:46
	(first person to say git stash gets a wet fish across the face :p)	20:46
cirwin	git reset --mixed <other-commit>	20:46
	← infynyxx left	20:47
	cwillu_at_work investigates	20:47
cirwin	man git-reset	20:47
gitinfo	the git-reset manpage is available at http://jk.gs/git-reset.html	20:47
	→ TheJH joined	20:47
cirwin	there's quite a few modes	20:47
	I usually try to use --keep	20:47
	→ kerrick joined	20:47
cirwin	which maintains the current diff, rather than the actual current working tree	20:47
	→ cakehero joined	20:47
	→ tdowg12 joined	20:47
	→ fly9 joined	20:48
	cwillu_at_work huggles cirwin, and puts the fish where it belongs	20:48
	← tdowg12 left	20:48
	→ psoo joined	20:49
	← TeckniX left	20:49
	→ TeckniX joined	20:49
cirwin	wow, huggling and fish-slapping on irc; are you a wikimedian?	20:49
	← metcalfc left	20:49
cmn	how is that like a pull?	20:50
cwillu_at_work	no, but I did commit from a holiday-inn express last night	20:50
	cmn, like how a git pull is like a git fetch + git merge	20:51
	this was a git fetch + git reset	20:51
cmn	the point of pull is the merge	20:51
	← Milossh left	20:51
cwillu_at_work	which will be happening, very carefully, thank you very much	20:51
	→ rideh joined	20:52
	→ les_sylvains joined	20:52
	← Gizmokid2005 left	20:52
	dankest\|away → dankest	20:54
	→ metcalfc joined	20:54
	← les_sylvains left	20:54
	→ jayallen joined	20:55
	← mikeit left	20:56
	← Ryback_ left	20:56
	← Dweezahr left	20:56
	→ Dweezahr joined	20:57
	← etcetera left	20:57
	← danillonunes left	20:59
	→ lacrymology joined	20:59
	→ danillonunes joined	20:59
	← jceb left	21:00
	→ thierryp joined	21:00
	→ etcetera joined	21:00
	→ bedouin joined	21:01
	→ infynyxx joined	21:01
	← Davey left	21:02
	← Leeol left	21:02
	← cstephens4 left	21:02
	← thierryp left	21:03
	← v0n left	21:03
	→ Gentlecat joined	21:03
	→ yshh joined	21:04
	← benbergman left	21:04
	← TheJH left	21:04
	→ hwrd\|work joined	21:04
	← shruggar left	21:05
	→ nyxynyx joined	21:06
	→ Gizmokid2 joined	21:07
	← x74beers left	21:07
	→ imrobert joined	21:07
	← erichynds left	21:09
	dankest → dankest\|away	21:09
	← GeertB left	21:09
	→ super8 joined	21:09
	→ nurey joined	21:10
	→ TeckniX_ joined	21:11
	← psoo left	21:11
	← TeckniX left	21:11
	TeckniX_ → TeckniX	21:11
	→ benbergman joined	21:12
	← axl_ left	21:12
	← Gizmokid2 left	21:12
	→ s0ber_ joined	21:13
	→ Leeol joined	21:13
	→ tvw joined	21:13
	← skyf left	21:14
	← s0ber left	21:15
	s0ber_ → s0ber	21:15
	→ axl_ joined	21:16
	→ v0n joined	21:18
	← andrewhl_ left	21:18
	→ opus joined	21:19
	← Leeol left	21:19
	→ Synthead joined	21:20
	← simesy left	21:20
	← barneybook left	21:21
	← marthinal left	21:22
	← marhaban left	21:22
	→ barneybook joined	21:22
	← juril left	21:23
	← R1KRDO left	21:23
	→ marhaban joined	21:23
	← etcetera left	21:24
	← jayallen left	21:26
	→ _Vi joined	21:26
	← nurey left	21:27
	← _iron left	21:27
	← martinjl` left	21:28
	→ otherj joined	21:29
	← nyxynyx left	21:29
	← Goplat left	21:30
	→ Goplat` joined	21:30
	← pandeiro left	21:30
	← otherj left	21:31
	→ wuddigel joined	21:32
	← brayn left	21:32
	→ dfaure joined	21:32
	← alansaul left	21:33
dfaure	how to git blame a file that moved, to see things as they were before the move?	21:33
	git blame f929f67b~ kiconloader.cpp doesn't work, with f929f67b the commit that moved the file	21:34
	→ decasm joined	21:34
	← opus left	21:35
	→ ergZay joined	21:35
cirwin	git blame -C	21:35
	← wuddigel left	21:35
dfaure	works somewhat, with -M -C -C, but can't pass an older revision as param to go back in time....	21:38
	← lresende left	21:39
cirwin	dfaure: git blame <revision>:<file>	21:39
	→ juril joined	21:39
cirwin	or <revision>^:<original-file>	21:39
	→ BiggFREE joined	21:40
	← h0bbit left	21:41
	→ Leeol joined	21:41
cmn	yes, you can pass an older version	21:42
	just pass it git blame <flags> <commit> -- <file>	21:42
	← i42n left	21:43
	← Sonderblade left	21:43
	← BiggFREE left	21:45
	← juril left	21:46
	elliottcable → RangeError	21:47
	→ jimmyeatworld joined	21:47
	← lindenle left	21:49
	→ jayallen joined	21:49
	→ Guest75728 joined	21:49
	→ lindenle joined	21:50
	← bubbles10101 left	21:50
	→ bubbles10101 joined	21:50
dfaure	cmn: with the current or old path to the file?	21:50
cmn	whichever path you want to look at that commit	21:51
	← Putr left	21:51
dfaure	ok, old path, works.	21:51
	amazing that the "--" is necessary	21:51
	← wilsonj left	21:51
cmn	it's not	21:52
	← super8 left	21:53
	→ wuddigel joined	21:53
	← wuddigel left	21:53
	← glennpratt left	21:54
cirwin	it is if you're accessing the file before the latest rename	21:55
	→ glennpratt joined	21:55
	← lindenle left	21:56
	→ TeckniX_ joined	21:56
dfaure	cmn: I get errors without it	21:56
	← TeckniX left	21:56
	TeckniX_ → TeckniX	21:56
cmn	namely?	21:56
	→ Putr joined	21:57
	→ juril joined	21:57
	← brabo left	21:57
	RangeError → elliottcable	21:58
	← Guest75728 left	21:58
	→ brabo joined	21:59
	← ajeffco left	21:59
	← alfadir left	21:59
	→ Blue0ctober joined	21:59
	← WinstonSmith left	21:59
	← Blue0ctober left	22:00
	← freeed left	22:00
	← benbergman left	22:01
	→ boombatower joined	22:01
	→ shruggar joined	22:01
	← _Vi left	22:01
	→ WinstonSmith joined	22:02
	← metcalfc left	22:02
dfaure	cmn: git blame 4f57a80b~ kiconloader.cpp	22:02
	fatal: cannot stat path 'kdecore/4f57a80b~': No such file or directory	22:02
	with -- it works.	22:02
cmn	that's very weird	22:02
dfaure	but it's all sorted out now, I traced the orig code finally, through 4 or 5 moves :)	22:02
	→ AdrienBrault joined	22:03
	← eijk left	22:03
	→ eijk joined	22:03
	→ _Vi joined	22:04
	← iamjarvo left	22:04
	→ jcao219 joined	22:05
	← jfs left	22:06
	→ R_Macy joined	22:06
	→ FernandoBasso joined	22:06
	← cannonball left	22:07
	→ psoo joined	22:07
	← mikef left	22:07
	dankest\|away → dankest	22:07
	← bartek_ left	22:08
	→ benbergman joined	22:08
	← rob_w left	22:17
	← mel- left	22:17
	← shruggar left	22:17
	→ etcetera joined	22:18
	→ metcalfc joined	22:18
	← sebrock left	22:19
	← sbodo left	22:22
deryni	If git can't figure out that you gave it a filename (because it can find the file) it tries to look it up as a ref I believe. You need -- to force it to think of it as a file. Or something like that.	22:22
thiago	correct explanation	22:23
	← flijten left	22:24
	→ normanrichards joined	22:24
wereHamster	dfaure: this kind of history archaeology is easier with gitk	22:25
	→ opus joined	22:25
	← swiety left	22:25
cmn	files always go after revisions, the problem is that it's treating a revision as a path, not the other way around	22:25
	← steve6 left	22:26
wereHamster	you can click an line and it takes you to the commit which last modifies it, you don't have to copy sha's around on the commandline and such..	22:26
	← hoschi left	22:26
	← hellopat left	22:26
wereHamster	and it has a 'show origin of this line' option	22:26
	← kerrick left	22:27
	← leeN left	22:28
	→ kerrick joined	22:28
	→ pmdarrow joined	22:29
	← pmdarrow left	22:29
deryni	Ah, true. The failure is the other direction. You need to force it to interpret the rev as a rev there apparently.	22:30
	← sideone left	22:30
	← m0ltar left	22:31
	→ Gizmokid2005 joined	22:31
	→ simesy joined	22:31
	← BadProphet left	22:31
	← chuckharmston left	22:31
	← Pugnax left	22:32
	→ m0ltar joined	22:32
	→ BadProphet joined	22:32
	← minusthetiger left	22:33
	→ tbrock joined	22:33
	→ tylerwillingham joined	22:34
	← Hounddog left	22:34
	← OOPMan left	22:34
tylerwillingham	Is there any sort of script that I can install that can post a message in an IRC channel any time someone pushes a new set of commits?	22:34
	lavaman → lavaman_afk	22:34
cirwin	tylerwillingham: are you using github?	22:34
	there's a builtin service hook for that	22:34
tylerwillingham	cirwin: no, there are all hosted on our work servers	22:35
cirwin	bad luck, there are probably scripts you can steal off the internet	22:35
	← simesy left	22:35
tylerwillingham	cirwin: i've been trying to look but perhaps i'm not entering worthwhile keywords… or no one has anything	22:35
	we have project-specific IRC channels at work and it would be really nice if we were able to be notified in the IRC channel whenever someone pushes a changeset	22:36
	→ guns joined	22:36
	← bitkiller left	22:36
wereHamster	tylerwillingham: http://www.catb.org/esr/irker/	22:37
	← marhaban left	22:37
tylerwillingham	wereHamster: hero	22:38
	← etcetera left	22:38
	← jcao219 left	22:38
	→ yakko joined	22:39
	→ RenJuan joined	22:41
	← SirCmpwn left	22:41
	← Serghar left	22:41
	→ SirCmpwn joined	22:41
RenJuan	you can pull a single file can't you?	22:41
EugeneKay	man git-grep	22:42
gitinfo	the git-grep manpage is available at http://jk.gs/git-grep.html	22:42
	← sgambino left	22:42
	← resmo_ left	22:43
	→ bitkiller joined	22:43
RenJuan	y or n would have worked better	22:43
wereHamster	RenJuan: no	22:44
dfaure	deryni: but why would git blame take two revs?	22:44
RenJuan	wereHamster, ty	22:46
	← project2501a left	22:47
	← JaredMiller left	22:47
	→ lresende joined	22:48
	← berserkr left	22:48
	← TeckniX left	22:49
	← dfaure left	22:49
	→ TeckniX joined	22:50
	← FernandoBasso left	22:50
	→ jarred_ joined	22:51
EugeneKay	I HATE SUBMODULES	22:51
deryni	Yeah. I don't think I understand that blame behaviour.	22:52
	← robustus left	22:53
	← TeckniX left	22:54
	← jokajak left	22:55
	← v0n left	22:55
	→ rcu joined	22:56
	→ madewokherd joined	22:56
	→ BD- joined	22:56
	← axl_ left	22:56
	← BD- left	22:56
	← metcalfc left	22:57
	← thiago left	22:57
	→ marhaban joined	22:57
	→ jokajak joined	22:57
	← SJr left	22:58
	← ChrisBorgia left	22:58
	← infynyxx left	22:59
	→ SJr joined	22:59
	← macabre left	23:01
	← _Vi left	23:02
	← smaddock left	23:03
	→ infynyxx joined	23:03
	→ metcalfc joined	23:04
	→ Blue0ctober joined	23:05
	← Blue0ctober left	23:05
	→ Blue0ctober joined	23:05
	← Blue0ctober left	23:05
	→ Blue0ctober joined	23:06
	← nitrix_ left	23:08
	→ zivester joined	23:09
	→ Raydiation joined	23:10
	← j416 left	23:10
Raydiation	how can i undo a cherry-pick	23:10
	im getting tons of warning: LF will be replaced by CRLF in tests/enable_all.php.	23:10
	The file will have its original line endings in your working directory.	23:10
	→ brokenladder joined	23:10
	→ babilen joined	23:10
	← ircmouser left	23:10
	→ simesy joined	23:11
Raydiation	simply git reset --hard HEAD?	23:11
	← MrNibbles left	23:11
	← AdrienBrault left	23:11
brokenladder	Is there a convention I can follow with ssh identity files so that ssh will try to use an identity matching the current github.user or something to that effect?	23:11
	← lresende left	23:12
	→ xrq joined	23:13
	← trungqtran left	23:13
	← myx left	23:13
	→ daniel_- joined	23:13
	→ MrNibbles joined	23:14
cirwin	brokenladder: add to your .ssh/config	23:14
	→ myx joined	23:14
	→ Davey joined	23:15
	← tylerwillingham left	23:15
	← garrys_game_ left	23:16
brokenladder	that doesn't work.	23:16
	→ garrys_game_ joined	23:16
	← ketas left	23:16
	← actel left	23:16
brokenladder	they are all password protected so it needs to know which one to use.	23:16
	→ pl0sh__ joined	23:16
	→ ketas joined	23:16
brokenladder	are you saying it will pick the right one based on user.password?	23:16
cirwin	brokenladder: you can set up aliases for hosts in your ssh config	23:16
	→ actel joined	23:16
cmn	you tell it which one to use	23:17
cirwin	and then use those aliases in your git remotes	23:17
cmn	they're all for the same user, git@	23:17
brokenladder	no, we use a "git pair" utility.	23:17
	if i'm pairing with jason, i'd type "git pair clay jason"	23:17
	the user.name would then be "Clay Smith and Jason Doe"	23:17
	← jayallen left	23:18
EugeneKay	git.... pair?	23:18
brokenladder	yes. for pair programming.	23:18
	you want to commit as the pair.	23:18
	this is something Pivotal Labs does.	23:18
EugeneKay	Ahh. Split brain syndrome	23:18
brokenladder	?	23:18
cirwin	brokenladder: add two remotes	23:18
	← asakura left	23:19
EugeneKay	brokenladder - https://bitbucket.org/spooning/ ;-)	23:19
brokenladder	this is easy if you just have a single ssh key per machine, and you add it to a single github account called something like "workstation"	23:19
haarg	you can tell it to use an ssh wrapper and have that wrapper decide what key to use based on some criteria	23:19
	← jarred_ left	23:19
brokenladder	but at Change.org, we want to let each user add a Change-specific key to his github account, that is on all machines and requires a password.	23:19
	i want to automate all this..	23:19
	using Chef and Soloist.	23:19
	→ jarred_ joined	23:19
	→ _Vi joined	23:20
	→ PacketCollision joined	23:20
	← Raydiation left	23:20
brokenladder	haarg, yeah, i had thought of that. i think it's $GIT_SSH you set right?	23:21
cirwin	brokenladder: it doesn't really matter who pushes the commits — providing they have the committer and author on the commits	23:21
brokenladder	it certainly matters who has access to our repo.	23:21
	← rideh left	23:21
cirwin	if you're going to put the password for all users on each machine, then per-user access control is useless anyway?	23:23
brokenladder	the password won't be on each machine.	23:23
PacketCollision	I'm trying to set up a deploy script as a post-receive hook, it works when someone pushes a single branch but when two branches are pushed, only the first one is handled by my script. I'm using "while read oldrev newrev refname; do; ... ;done" which I thought would work. Any thoughts?	23:23
	← pantsman left	23:23
	← BadProphet left	23:23
brokenladder	each user will have his private key on each machine he wants to use, but it will be password protected.	23:23
	← psoo left	23:23
cirwin	brokenladder: ok	23:24
	← kermit left	23:24
	← tenmilestereo left	23:25
brokenladder	i'm open to other ways of implementing this. it's tricky. pivotal labs just tends to use one private key per machine, but an engineer could copy those private keys to his laptop, get fired, and....	23:25
	← vpetro left	23:25
	← jarred_ left	23:26
cirwin	brokenladder: you'll never solve that problem	23:26
	just trust people	23:26
	→ kermit joined	23:26
brokenladder	well, you can certainly solve that problem.	23:26
	you make sure that each dev has his own private key.	23:27
	← opus left	23:27
cirwin	the default private key encryption for ssh is not very secure	23:27
	← senior75151 left	23:27
brokenladder	you can set an arbitrarily high key size.	23:27
	← Chillance left	23:28
	← glennpratt left	23:28
	cirwin just runs an offline password guesser on your private key files	23:28
cirwin	unless you're using pbkdf2 already?	23:28
brokenladder	surely not.	23:28
	← tinti left	23:29
	→ glennpratt joined	23:29
	← danillonunes left	23:29
	→ jarred_ joined	23:29
	← melder left	23:30
	← jokajak left	23:30
	→ jokajak joined	23:31
	← hotovson left	23:31
	→ senj joined	23:32
	← jokajak left	23:32
	← Element9 left	23:32
	← nowhere_man_ left	23:33
	→ jokajak joined	23:33
	→ project2501a joined	23:33
	← project2501a left	23:34
	← Goplat` left	23:34
	→ Goplat joined	23:34
	→ senior75151 joined	23:34
	→ dragon3_ joined	23:35
	← Blue0ctober left	23:35
	→ nowhere_man joined	23:35
	← Jippi left	23:36
	← marhaban left	23:37
	→ hyperair joined	23:37
	← guns left	23:37
	← metcalfc left	23:38
	→ deiu joined	23:38
deiu	Hello	23:38
gitinfo	deiu: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	23:38
deiu	Does anyone know how to merge changes from the main branch into a different one?	23:39
	I've tried to do 'git checkout branchname' and then 'git merge main', but it doesn't work	23:39
	→ hotovson_ joined	23:40
	→ rideh joined	23:40
	← jokajak left	23:40
deryni	master not main?	23:40
	← simesy left	23:41
	→ txomon\|home joined	23:41
	→ frogonwheels joined	23:41
	→ jokajak joined	23:41
	← ergZay left	23:42
deiu	ah yes, silly me	23:42
	thanks! :)	23:42
	← jokajak left	23:43
	← dragon3_ left	23:43
	→ jokajak joined	23:44
	→ CrazyHorse18 joined	23:44
	→ martinjlowm-lapt joined	23:49
	← nioncode left	23:49
wereHamster	"But it doesn't work" - The Greatest Thing (tm) ever said on IRC.	23:49
	→ cirenyc joined	23:49
	← cakehero left	23:51
	→ rchavik joined	23:51
	← rchavik left	23:51
	→ rchavik joined	23:51
	→ simesy joined	23:54
	← hyperair left	23:54
	→ RaceCondishun joined	23:57
	→ jdunck joined	23:58
	→ m0viefreak joined	23:59
	→ jayallen joined	23:59
	← deiu left	23:59

Logs Search ←Prev date Next date→

Channels Documentation