IRCloggy #git 2013-02-20

Provider of IRC logs since 2005.

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

	2013-02-20
	→ tinti_ joined	00:00
	← jalama left	00:00
joshsmith	hmm I've done a git pull and it says already up-to-date	00:00
	ferdev → ferdev\|away	00:00
joshsmith	but git diff origin/master master shows a difference	00:01
	← maxmanders left	00:01
thiago	joshsmith: maybe the change is something new in your branch	00:01
	← mmitchell left	00:02
	← Remram left	00:02
	→ bmac_ joined	00:02
joshsmith	hmm it is…but it shouldn't have been	00:02
	that's very odd	00:02
	← AnimeCYC left	00:03
	← tevfik left	00:03
	→ Remram joined	00:03
	ferdev\|away → ferdev	00:03
	→ tevfik joined	00:04
	← tevfik left	00:04
	← babilen left	00:04
	← sgambino left	00:05
	← p3rror left	00:06
	→ whitemage joined	00:06
	← gusnan left	00:08
	→ daddio joined	00:09
	→ comps joined	00:09
	← v0n left	00:09
	→ Calio joined	00:10
	→ bryanstedman joined	00:10
	→ rchavik joined	00:11
daddio	I found a bug in git-svn 1.7.10.4 (the one packaged by ubuntu). It is in Git.pm. I'm having a hard time figuring out where the master repository is ... to know whether it has since been fixed upstream. Pointers plz?	00:12
	← lavaman left	00:13
	← rgr left	00:14
	→ alunduil joined	00:14
	← Aktau left	00:14
	→ sajan joined	00:14
	→ normanrichards joined	00:15
puppet_help	must gitolite administration be done locally?	00:15
	as in, must I clone the admin git repo in the localhost or can I clone it to a remote host as with any normal git repo	00:15
	← m0viefreak left	00:16
	→ v0n joined	00:17
	← Dooom left	00:21
	← eijk_ left	00:21
	← PaulCapestany left	00:21
	← kermit left	00:22
	← Anders_J left	00:22
	→ Korkenknaller joined	00:23
	← kevwil left	00:23
	→ kermit joined	00:24
	← Korkenknaller left	00:24
	← jdunck left	00:24
	→ Korkenknaller joined	00:25
	→ Aymeric joined	00:25
	← tinti_ left	00:26
	→ SicoWork joined	00:26
	gitinfo set mode: +v	00:27
Aymeric	hello	00:27
gitinfo	Aymeric: 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.	00:27
	→ SethRobertson joined	00:27
Aymeric	I need some help with git. I have a repo I cloned from github. I have pushed a few commits in origin/master, but I now want to move them in a separate branch, and be able to get the latest version of my upstream repo. How could I achieve that? (note that there are some merge commits in there)	00:27
	→ jdunck joined	00:27
	→ TDJACR joined	00:27
	← SicoAnimal left	00:27
	← Galaris__ left	00:28
	→ mikeputnam joined	00:28
	→ Galaris__ joined	00:29
	← Immutable left	00:29
	→ up_the_irons joined	00:29
	→ Mez joined	00:29
	← Mez left	00:29
	→ Mez joined	00:29
Aymeric	Is there anyone here?	00:29
	→ relipse joined	00:29
thiago	yes	00:29
	→ Immutable joined	00:30
	← yshh left	00:30
	← Immutable left	00:31
thiago	Aymeric: push your commits to the new branch	00:31
Aymeric	Hi thiago, are you comfortable with rebase / reset cherry-pick commands?	00:31
thiago	yes	00:31
Aymeric	I have already pushed my commits to origin/master	00:32
	← natechan left	00:32
Aymeric	so my origin/master has a mix of my commits and upstream commits	00:32
	I want to move all my commits to a separate branch	00:32
	so that I can easily rebase from upstream with minimum conflicts	00:33
	not sure if I am clear sorry :(	00:33
thiago	Aymeric: just rebase	00:33
	→ Immutable joined	00:33
	← TDJACR left	00:34
	← chisler left	00:35
	→ novaweb joined	00:36
Aymeric	Currently I have origin/master myCommit1 mergeCommit1 theirCommit2 myCommit2 my Commit3 theirCommit3	00:37
	← Galaris__ left	00:37
Aymeric	I want: origin/master theirCommit2 theirCommit3 origin/mybranch myCommit1 myCommit 2	00:37
	← Immutable left	00:37
	→ Galaris__ joined	00:37
	→ nitrix- joined	00:38
	→ Immutable joined	00:38
	← giallu left	00:39
thiago	Aymeric: indeed. Just rebase.	00:39
	← itatitat left	00:40
Aymeric	what would be the command?	00:40
	→ glaksmono joined	00:40
glaksmono	anyone is around?	00:40
Aymeric	first I need to move the HEAD of myBranch back before my first commit	00:40
	← pantsman left	00:40
Aymeric	and then do a rebase?	00:40
	← maestrojed left	00:41
glaksmono	Changes not staged for commit: <-- i didn't even add any of the files and when i do commit, it seems like it's picking up those files that's listed there?	00:41
thiago	Aymeric: if you've set everything properly, just "git rebase"	00:41
	Aymeric: to download and rebase: git pull --rebase	00:41
nfk	cmn, using git commit -a has made my life better	00:41
thiago	glaksmono: !repro	00:42
gitinfo	glaksmono: 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.	00:42
glaksmono	ah that's because i'm using git commit -am	00:42
	that's probably why..	00:42
	→ dankest joined	00:42
thiago	glaksmono: right	00:42
	→ BiggFREE joined	00:42
glaksmono	makes sense	00:42
	thanks thiago	00:42
	:)	00:42
	← glaksmono left	00:42
	→ babilen joined	00:43
	→ jaequery joined	00:43
Aymeric	ok thanks	00:43
	← puppet_help left	00:44
	→ maestrojed joined	00:44
	→ alp82 joined	00:45
	← Aymeric left	00:45
	→ codekoala joined	00:47
	← lethjakman left	00:47
	← maestrojed left	00:48
	← jdunck left	00:49
	← Galaris__ left	00:49
	→ Galaris__ joined	00:49
	← bmac_ left	00:50
	← adamben left	00:50
	← TheJH left	00:51
	← infynyxx left	00:52
	← BradLilley left	00:52
	← codekoala left	00:53
	→ pity joined	00:53
	← chuckharmston left	00:54
	← babilen left	00:55
	← rlubke left	00:55
	→ rurufufuss joined	00:57
	← alp82 left	00:58
	← novaweb left	00:58
	→ alp82 joined	00:58
	← Calio left	00:59
	← Korkenknaller left	00:59
	← alp82 left	00:59
	→ alp82 joined	01:00
	→ Jake232 joined	01:00
	→ yshh joined	01:00
	← alp82 left	01:00
	← ejd left	01:00
	→ alp_ joined	01:00
	← bootstrapper left	01:00
	← bryanstedman left	01:00
	← jonshea left	01:01
	← alp_ left	01:01
	→ Galaris_ joined	01:01
	← Galaris__ left	01:02
	← normanrichards left	01:03
	← fructose left	01:05
	← Galaris_ left	01:05
	→ Galaris_ joined	01:06
	→ mehwork joined	01:06
mehwork	i did 'git reset --hard' but 'git status' still shows a bunch of untracked files... what should i do?	01:06
	git version 1.7.4.1	01:07
grawity	either tell git to start tracking them (`git add` and commit), or tell git to start ignoring them (.gitignore)	01:07
	resetting does not affect this	01:07
	← jerikl left	01:07
	← TheFuzzball left	01:08
mehwork	so i should just manually rm them	01:08
	gitignore would be brushing it under the rug in this case	01:08
grawity	depends on what those files are	01:08
mehwork	i've had reset --hard work before like this, idk why it's not now	01:08
	these files were generated by 'rails generate controller' but i changed my mind and wanted to undo	01:09
	→ codekoala joined	01:09
grawity	in that case, just `rm` them	01:09
	or, see man git-clean	01:10
gitinfo	the git-clean manpage is available at http://jk.gs/git-clean.html	01:10
	← d2dchat left	01:10
	→ Galaris__ joined	01:11
	← Galaris_ left	01:11
	← snkcld left	01:11
	← Jake232 left	01:11
	→ Jake232 joined	01:13
	← v0n left	01:13
	← dreiss left	01:13
	← marhaban left	01:14
	→ v0n joined	01:14
	← dankest left	01:15
	→ jdunck joined	01:16
	← jarr0dsz left	01:17
	→ Galaris_ joined	01:19
	← Galaris__ left	01:19
mehwork	git clean -fd worked, thanks	01:19
	→ bootstrapper joined	01:20
	→ __BigO__ joined	01:20
	← warez left	01:21
	← Imaginativeone left	01:21
	← lolwhut left	01:22
	→ mmitchell joined	01:22
	→ PaulCapestany joined	01:23
	← __BigO__ left	01:24
	← macrover left	01:25
	← relipse left	01:25
	← codekoala left	01:25
	→ Fandekasp joined	01:25
	→ senorflor joined	01:25
	← intripoon left	01:26
	→ Etheco joined	01:26
	← aartist left	01:26
	→ Progster joined	01:27
	→ intripoon joined	01:27
	← jjore left	01:30
	→ natechan joined	01:31
	← NiteRain left	01:31
	→ d2dchat joined	01:34
	← Jake232 left	01:35
	→ znode joined	01:36
	← Galaris_ left	01:36
	← AartBluestoke left	01:37
	→ Galaris joined	01:37
	→ isomorph_ joined	01:38
	← Noldorin left	01:39
	← znode left	01:39
	← thiago left	01:39
	→ znode joined	01:40
	← senorflor left	01:40
	← isomorph left	01:41
	→ senorflor joined	01:41
	→ Jake232 joined	01:42
	← d2dchat left	01:42
	← sluther left	01:43
	→ Troyan0 joined	01:43
	→ babilen joined	01:44
	→ sluther joined	01:44
	← ErKa left	01:44
	← agile left	01:46
	← etcetera left	01:47
	← senorflor left	01:47
	→ alan__ joined	01:47
	→ znode_ joined	01:47
	→ dlan joined	01:48
	← hyperair left	01:49
	← bootstrapper left	01:50
dlan	is there any command that I can use to find which commit completely delete one directory?	01:50
	← isomorph_ left	01:50
	→ senorflor joined	01:51
	← znode left	01:51
	← bigkm left	01:51
	→ stliu joined	01:51
	← Progster left	01:52
	→ Progster joined	01:52
	← tjbiddle left	01:53
	→ madewokherd joined	01:53
	← plumpkin left	01:53
	→ jdyin joined	01:54
	→ plumpkin joined	01:54
	← nerd left	01:54
	ferdev → ferdev\|away	01:54
	→ willob joined	01:55
	← Davey left	01:55
	→ Noldorin joined	01:56
	← brianpWins left	01:57
	→ QaDeS__ joined	01:57
	→ megha joined	01:57
	→ amigojapan joined	01:58
	← plumpkin left	01:58
	← senorflor left	01:59
	← imachuchu left	01:59
	→ goshawk joined	01:59
	→ nickn joined	02:00
	← goshawk left	02:00
	→ goshawk joined	02:00
	← jaequery left	02:01
	← QaDeS_ left	02:01
	→ techlife joined	02:01
	← AaronMT left	02:02
	→ d2dchat joined	02:02
	← maletor left	02:03
	→ agile joined	02:03
	← Sigma left	02:03
	← goshawk left	02:05
	← znode_ left	02:05
sitaram	dlan: try 'git log --diff-filter=D -- dirname', and perhaps the top commit is what you want (it actually finds any commit that deleted any file or subtree of 'dirname')	02:05
	→ chuckharmston joined	02:05
	← tonymet left	02:06
	← Guest34610 left	02:08
	← yshh left	02:09
	← nickn left	02:10
	→ yshh joined	02:10
	← BiggFREE left	02:10
	← chuckharmston left	02:10
	← joshsmith left	02:11
	← yhager left	02:12
	→ znode joined	02:12
	← simesy left	02:13
	→ joshsmith joined	02:14
	→ wrouesnel joined	02:15
	→ erkules_ joined	02:15
	← AdrienBrault left	02:15
	→ Vampire0_ joined	02:16
	← QaDeS__ left	02:16
	→ tjbiddle joined	02:16
	← intripoon left	02:17
	← opus left	02:17
	→ intripoon joined	02:18
	← tjbiddle left	02:18
	← boze left	02:19
	ldlework → cadgu`a	02:19
	→ isomorph_ joined	02:19
	← Vampire0 left	02:19
	→ maletor joined	02:19
dlan	sitaram: thanks, but I may not know the exact 'dirname', I'd like to traverse all commits, and list all directories which been deleted	02:20
	→ goshawk joined	02:21
	→ santosh joined	02:22
	← alan__ left	02:23
	← daniel_- left	02:23
santosh	My master branch is more developed (has more commits) compared to dev branch. Can I update a single file of dev branch to master branch?	02:23
	→ simesy joined	02:24
	→ przemoc joined	02:26
santosh	Anybody on the channel?	02:26
dlan	santosh: can cherry-pick do? or you can manually update using "git show"	02:28
	→ yhager joined	02:28
	← rchavik left	02:29
	→ fuzzybunny69y joined	02:30
	→ francisfish joined	02:30
	→ rchavik joined	02:31
	← eletuchy left	02:33
	→ eletuchy joined	02:33
	→ JaredMiller joined	02:33
	→ Error404NotFound joined	02:34
	→ alan__ joined	02:35
	→ jaequery joined	02:35
	← maletor left	02:35
lyang0	what does it mean by PR	02:36
	→ danmackay joined	02:37
	← soypirate left	02:37
	← eletuchy left	02:37
	← isomorph_ left	02:38
	← goshawk left	02:38
	← JaredMiller left	02:38
	→ intripoon_ joined	02:38
	← santosh left	02:39
	→ johnny_rose joined	02:39
	→ NiteRain joined	02:40
	→ isomorph_ joined	02:40
	→ relipse joined	02:40
	← d2dchat left	02:40
	← nfk left	02:41
	← znode left	02:41
	← intripoon left	02:42
	← willob left	02:42
	← alan__ left	02:43
	→ znode joined	02:44
	→ redrocket joined	02:44
	→ Goplat joined	02:45
	← ehsan_ left	02:47
	→ jalama joined	02:47
	→ ehsan joined	02:47
	→ hyperair joined	02:48
	← yhager left	02:49
	→ tjbiddle joined	02:52
	← ehsan left	02:52
	→ bababooey joined	02:52
	→ jjore joined	02:53
	→ cornfeedhobo joined	02:53
	→ goshawk joined	02:53
	← chichou left	02:55
	→ breck joined	02:55
	← jjore left	02:56
	gitinfo set mode: +v	02:56
	→ etcetera joined	02:56
bababooey	Anyone here to answer a question about branching?	02:56
	→ PrinceAMD joined	02:56
	← sluther left	02:57
	megha → bsd_freak	02:57
	← tjbiddle left	02:58
	bsd_freak → BSD_freak	02:58
	→ cornfeedhobo_ joined	02:58
engla	bababooey: branching questions are on thursdays	02:58
	just kidding. don't ask to ask	02:58
	!ask	02:59
gitinfo	Yes, it's okay to ask questions here.... in fact, you just asked one! ;-) Pretty much any question is fine. We're not terribly picky, but we might be asleep. Please be patient and you should get an answer soon.	02:59
	→ eletuchy joined	03:00
	← goshawk left	03:00
	← cornfeedhobo left	03:01
	cornfeedhobo_ → cornfeedhobo	03:01
	→ infynyxx joined	03:01
	← eletuchy left	03:01
	← etcetera left	03:01
	← fuzzybunny69y left	03:01
	→ maestrojed joined	03:01
	→ eletuchy joined	03:02
	← RaceCondition left	03:03
	← znode left	03:03
	→ normanrichards joined	03:04
	← Progster left	03:05
	→ thiago joined	03:06
	← Noldorin left	03:06
	→ eletuchy_ joined	03:06
	→ wwwmmmzzz joined	03:06
	← wwwmmmzzz left	03:07
	← eletuchy left	03:07
	→ m0ltar joined	03:08
	→ znode joined	03:08
	← eletuchy_ left	03:08
	→ eletuchy joined	03:09
	← breck left	03:09
	→ breck joined	03:10
	← pr0ton left	03:11
	← m0ltar left	03:11
	← eletuchy left	03:13
	→ etcetera joined	03:14
	← breck left	03:14
	→ alan__ joined	03:15
	→ Gentlecat joined	03:16
	→ juril joined	03:17
	← cornfeedhobo left	03:18
	→ R1KRDO joined	03:18
	← isomorph_ left	03:18
	← juril left	03:19
	→ yhager joined	03:19
	← PaulCapestany left	03:20
	← joe9 left	03:22
	→ eletuchy joined	03:23
	← znode left	03:24
	← natechan left	03:24
	← Jake232 left	03:25
	→ luc_ joined	03:26
	→ weiyang joined	03:27
	→ orafu joined	03:28
	← alan__ left	03:30
	→ saki007ster joined	03:32
	← rurufufuss left	03:32
	→ herdingcat joined	03:34
	→ jonshea joined	03:35
	← jalama left	03:39
	→ alan__ joined	03:42
	← johnny_rose left	03:43
	→ rurufufuss joined	03:45
	→ gh403 joined	03:46
gh403	Question: what is the minimum needed on a server to host a git repo for a few developers?	03:46
	→ Carlitos__ joined	03:47
Carlitos__	hi all	03:47
	please	03:48
gh403	hi Carlitos__	03:48
Carlitos__	I have a repo in a server with ssh	03:48
	→ juvenal joined	03:49
	← jcromartie left	03:49
Carlitos__	in my local machine git clone server.hosting	03:49
	→ dreiss joined	03:49
Carlitos__	I add some file and commited	03:49
gh403	ok	03:49
	← Error404NotFound left	03:49
	→ ehsan joined	03:49
Carlitos__	master is in the server hosting right ?	03:49
	how can I call another name than origin ? example git remote casaservergit ?	03:50
	← Gentlecat left	03:50
gh403	Well, sort of. origin/master is on the server. master is on your computer.	03:50
engla	gh403: you don't need more than installed git and ssh access. You might want to use gitolite though	03:51
SamB	origin/master is on your computer, too; it's just your record of what the remote repository had for "master" last you checked	03:51
	→ bootstrapper joined	03:51
gh403	Read http://git-scm.com/book/en/Git-Branching-Remote-Branches for how branching works with remotes.	03:51
	or http://git-scm.com/book/es/Ramificaciones-en-Git-Ramas-Remotas for the Espanol version ;)	03:52
	← Sky[x] left	03:52
Carlitos__	I only use it it is ok for just one user ?	03:52
thiago	yes	03:52
	→ fuzzybunny69y joined	03:53
	Troyan0 → Troyan0\|comiendo	03:53
gh403	and read http://git-scm.com/book/en/Git-Basics-Working-with-Remotes to learn how to add new remotes	03:53
	(or http://git-scm.com/book/es/Fundamentos-de-Git-Trabajando-con-repositorios-remotos)	03:53
Carlitos__	I try to read in spanish	03:54
	:D	03:54
gh403	engla, it's not my server :/ and afaik they won't run anything else on it. I do have samba (or maybe ssh) access to it.	03:55
	Can you do interesting things like permissions with just an ssh account?	03:55
	← fisted left	03:58
	← SBoudrias left	03:58
	→ fisted joined	03:58
	← Bankzilla left	03:58
Carlitos__	gh403: git fetch origin , I have to put this comman in the server or in my computer?	03:59
engla	I don't know gh403	03:59
gh403	on your computer, Carlitos__	03:59
	ok engla	03:59
Carlitos__	gh403: after that git push ?	03:59
	← eletuchy left	04:00
	← maestrojed left	04:00
thiago	Carlitos__: "fetch" is the same as download	04:00
	Carlitos__: so you run that in the one that you want to receive the changes	04:00
	→ eletuchy joined	04:00
engla	I'm writing a script to store encrypted remotes on any dumb server, but it doesn't support more than rsync protocol and sftp so far. Any 3rd-party remote protocol that is "dumb" (does not depend on smart server) can work	04:00
Carlitos__	I made changes in my computer and now I wanna upload the new files	04:00
gh403	Carlitos__, if the remote branch has changed, you have to merge that branch (for example origin/master) with the branch you are going to push.	04:01
thiago	engla: in other words, encrypt and dumb-transfer?	04:01
	→ dankest joined	04:02
engla	yes	04:02
	← yhager left	04:02
gh403	Carlitos__, if nothing has changed after your git fetch, then you are free to push. (The server will complain if there is an issue. It is not hard to figure out how to fix it.)	04:02
	→ semarjt joined	04:02
Carlitos__	is it correct ? git push -u origin master ?	04:03
	from my local computer	04:03
	→ ak5_ joined	04:03
	← LekeFly- left	04:03
gh403	Carlitos__, yes. If master tracks origin/master then you can omit -u origin	04:04
	I think	04:04
ak5_	hi, is there any way I can use gitolite or something to mirror github two ways? I am in China and concerned with latency	04:04
	and speed in general	04:04
	→ eephillip joined	04:04
Carlitos__	remote: error: 'receive.denyCurrentBranch' configuration variable to 'refuse'.	04:04
	← eletuchy left	04:05
	← kukks left	04:05
	← R1KRDO left	04:05
gh403	Carlitos__, it seems you are trying to push into a repository that has a currently checked out branch (i.e., it is not a bare repository).	04:06
	→ johnny_rose joined	04:06
gh403	http://stackoverflow.com/questions/804545/what-is-this-git-warning-message-when-pushing-changes-to-a-remote-repository has an excellent explanation.	04:06
	← fuzzybunny69y left	04:06
gh403	It is generally good practice to only push to a bare repository.	04:06
Carlitos__	gh403: I can use bare give and error the hosting server	04:07
	git config receive.denyCurrentBranch ignore	04:07
	→ pr0ton joined	04:07
gh403	Huh? I don't understand what you mean.	04:08
Carlitos__	sorry I can not use bare	04:08
	the server hosting give an error eberytime I use git init --bare	04:08
gh403	Oh, ok. Then you must either change the remote's currently checked out branch or set the option like you said.	04:08
	What is the error it gives?	04:09
	→ Progster joined	04:09
Carlitos__	fatal: Out of memory? mmap failed: No such device	04:10
	Can I fix that ?	04:10
	→ rastar joined	04:10
gh403	O.o that is a very strange error	04:11
Carlitos__	I find this http://stackoverflow.com/questions/5670132/git-init-bare-error-fatal-out-of-memory-mmap-failed-no-such-device	04:11
	but I am lazy right now, maybe try for tomorrow	04:12
	or now?	04:12
	:)	04:12
	Troyan0\|comiendo → Troyan0	04:12
gh403	:) one sec	04:12
	let me read...	04:13
	← joshsmith left	04:13
	← rastar left	04:14
gh403	It would seem the best thing to do is to make a local, bare copy of the repository, then move that up to the server.	04:14
	Like the page you linked to explains.	04:14
	Once the repository is on the remote server you will be okay; you would have to add origin manually but you would have to do that anyway	04:15
Carlitos__	gh403: ok	04:16
gh403	Feel free to ask me here or in a /query if you have any more questions	04:17
Carlitos__	gh403: thanks man	04:17
gh403	Sure no problem	04:17
	→ snql joined	04:18
	→ breck joined	04:20
	→ dallen joined	04:21
	← mmitchell left	04:21
dallen	Hi all. I'm using Git Tower for Mac. I have hooks that aren't being run because, as Git Tower's FAQ says: "You need to make sure you're creating the required shell environment in the hook script itself (like modifying "PATH"), not in your shell profile, as the hook script is called from the Tower process environment which is not running in a shell environment (hence your shell profile is not loaded)."	04:22
	← Fandekasp left	04:22
dallen	I honestly have no idea what that means. My post-commit is a simple #!/bin/sh git ftp push	04:22
	← pr0ton left	04:24
	← breck left	04:25
	← ak5_ left	04:25
	→ subhojit777 joined	04:25
	→ OOPMan joined	04:28
	← Chillance left	04:28
	Troyan0 → Troyan0\|working	04:29
	← dallen left	04:32
	← ehsan left	04:34
	← eephillip left	04:34
	→ maestrojed joined	04:35
	→ bigkm joined	04:35
	→ eletuchy joined	04:35
	← eletuchy left	04:35
	→ eletuchy joined	04:36
	← pppaul left	04:38
	← scorphus left	04:40
	→ boze joined	04:40
	→ beautifulmind joined	04:43
	← Milossh left	04:43
	→ preflex_ joined	04:45
	preflex_ → preflex	04:47
	→ Bombe joined	04:47
	← the_cybersphinx left	04:48
	← maestrojed left	04:49
	← arvind_khadri left	04:49
	→ dhruvasagar joined	04:51
	→ h0bbit joined	04:51
	← snql left	04:52
	→ macmartine joined	04:52
	← madewokherd left	04:53
	→ isomorph_ joined	04:55
	→ JDuke128 joined	04:55
	← jwmann left	04:57
	← alan__ left	04:58
	→ jwmann joined	04:58
	← jargon left	05:00
	→ maestrojed joined	05:01
	→ lun_ joined	05:07
	← zequence left	05:07
	→ robbyoconnor joined	05:08
	→ kevwil joined	05:09
	← eletuchy left	05:09
	← Progster left	05:09
	← jwmann left	05:10
	→ eletuchy joined	05:10
	→ LimeBlast joined	05:10
	→ TheJH joined	05:11
	→ pretty_function joined	05:11
	← kevwil left	05:14
	← eletuchy left	05:14
	← etcetera left	05:14
	← dreiss left	05:16
	← TheJH left	05:17
	→ megha joined	05:18
	→ senorflor joined	05:18
	← BSD_freak left	05:19
	→ cornfeedhobo joined	05:19
	← bigkm left	05:22
	→ pppaul joined	05:22
	→ mike_pizza joined	05:25
	← mehwork left	05:28
	→ dreiss joined	05:30
	→ etcetera joined	05:31
	← CreativeEmbassy left	05:32
	→ alan__ joined	05:32
	→ mmitchell joined	05:32
	→ CreativeEmbassy joined	05:33
	← lun_ left	05:34
	← isomorph_ left	05:35
	→ lun_ joined	05:35
	← mmitchell left	05:37
	→ lun__ joined	05:38
	→ zequence joined	05:38
	← lun_ left	05:38
	← dreiss left	05:38
	→ dreiss joined	05:39
	← lun__ left	05:39
	→ Tommy[D]_ joined	05:40
	← saki007ster left	05:42
	→ sluther joined	05:42
	← Tommy[D] left	05:43
	← CreativeEmbassy left	05:45
	← infynyxx left	05:46
	→ Fandekasp joined	05:46
	→ infynyxx joined	05:46
	→ CreativeEmbassy joined	05:46
	← Damgaard left	05:47
	→ Skaag joined	05:48
	← CreativeEmbassy left	05:48
	← maestrojed left	05:48
Skaag	does git add * make sense when adding a bunch of files and folders into a newly initialized repository?	05:48
	→ i0s joined	05:48
Skaag	or is there a more correct way?	05:48
	← dreiss left	05:49
	← gh403 left	05:50
	← OOPMan left	05:50
	→ gh403 joined	05:50
	← gh403 left	05:50
	→ gh403 joined	05:50
	← PrinceAMD left	05:51
	← amigojapan left	05:52
	→ dreiss joined	05:52
	→ lethjakman joined	05:52
	→ CreativeEmbassy joined	05:53
	→ maestrojed joined	05:54
	← fisted left	05:54
	→ fisted joined	05:54
	← senorflor left	05:55
	→ senorflor joined	05:55
mike_pizza	I think it makes sense if you're using bash. The wildcard should subsitute every item in the current directory and add it to the index. I don't know if it will work with subdirectories.	05:55
	→ Jackneill joined	05:56
	← senorflor left	05:56
	← Jackneill left	05:57
	← linpc left	05:57
	→ linpc joined	05:57
	→ snql joined	05:59
	→ blackmat` joined	06:02
	gitinfo set mode: +v	06:02
	→ fuzzybunny69y joined	06:03
	← wrouesnel left	06:03
	← v0n left	06:04
	→ Sadeukko joined	06:07
	→ maleknet joined	06:07
	→ AartBluestoke joined	06:08
	← fuzzybunny69y left	06:10
	→ eletuchy joined	06:10
	→ j416 joined	06:10
	erkules_ → erkules	06:12
	← Ziggy` left	06:14
	→ fuzzybunny69y joined	06:14
	← Vampire0_ left	06:15
lethjakman	join #ruby	06:15
gh403	ha	06:15
	← Goplat left	06:18
	→ fsvehla joined	06:18
	→ isomorph_ joined	06:20
	→ Vampire0 joined	06:20
	→ rastar joined	06:22
	← lethjakman left	06:22
	← alan__ left	06:23
	← kerrick left	06:23
	→ kerrick joined	06:23
	→ philips_ joined	06:24
	← rastar left	06:24
	→ lethjakman joined	06:25
	← etcetera left	06:26
	← macmartine left	06:27
	← bootstrapper left	06:30
	← CreativeEmbassy left	06:32
	← whitemage left	06:33
	→ whitemage joined	06:34
	← fuzzybunny69y left	06:35
	→ fuzzybunny69y joined	06:37
	→ Cromulent joined	06:40
	← lresende left	06:42
	← johnny_rose left	06:43
	← tewecske left	06:44
Skaag	mike_pizza: seems to have worked, recursively too	06:44
	→ WorkeD joined	06:44
	← dreiss left	06:46
	← boze left	06:48
	← pppaul left	06:49
	→ saki007ster joined	06:50
	→ longbeach joined	06:51
	← dankest left	06:52
	← normanrichards left	06:52
	→ tevfik joined	06:54
	→ Anders_J joined	06:54
	→ pppaul joined	06:54
	← JDuke128 left	06:55
	← whitemage left	06:55
	← Remram left	06:56
	→ toothr joined	06:59
	→ TonyM joined	07:00
	← lethjakman left	07:00
	→ alan__ joined	07:01
	← fuzzybunny69y left	07:03
	→ ThomasLocke joined	07:04
	← isomorph_ left	07:05
	→ etcetera joined	07:06
	← kcj left	07:07
	← etcetera left	07:07
	→ isomorph_ joined	07:07
	→ etcetera joined	07:08
	← b1rkh0ff left	07:09
	→ lifelike joined	07:09
	← maleknet left	07:10
	← alan__ left	07:10
	→ maleknet joined	07:10
	→ rakl joined	07:11
	→ nafg_ joined	07:11
	→ Error404NotFound joined	07:12
	→ fsvend joined	07:14
	← sluther left	07:14
	→ rtjure joined	07:14
	→ weekes joined	07:15
	→ perlstein joined	07:15
	gitinfo set mode: +v	07:15
	← simesy left	07:15
	← nafg_ left	07:16
	→ simesy joined	07:16
	← toothr left	07:16
	→ dreiss joined	07:17
	→ cbreak-work joined	07:18
	→ nafg_ joined	07:18
	← rakl left	07:19
	← weekes left	07:19
	← simesy left	07:20
	→ xiangfu joined	07:20
	← Cromulent left	07:21
	→ simesy joined	07:21
	→ b1rkh0ff joined	07:22
	→ tjbiddle joined	07:22
	→ Sigma joined	07:22
	← gh403 left	07:22
	→ ns5 joined	07:22
	→ jamescarr joined	07:22
jamescarr	yo yo yo	07:22
	→ marthinal joined	07:23
ns5	how to find out the original url I used to clone a repo?	07:23
jamescarr	I added a submodule to my project that I then found out was a mistake so I removed it	07:23
	I cannot get git to stop bitching about it	07:23
	No submodule mapping found in .gitmodules for path 'modules/puppetdbquery'	07:23
	every single time	07:23
	I removed it from .git/config, .gitmodule, and did git rm modules/pupetdbquery	07:23
	it still complains	07:23
	fatal: git rm: 'modules/puppetdbquery': Operation not permitted	07:24
	← nafg_ left	07:24
	→ rath joined	07:24
cbreak-work	ns5: man git-remote -v	07:24
gitinfo	ns5: the git-remote manpage is available at http://jk.gs/git-remote.html	07:24
CoJaBo	so, figured out how the git repo got wedged. heh.	07:25
	→ jargon joined	07:25
cbreak-work	jamescarr: git rm --cached submodule/path	07:25
	← maestrojed left	07:25
	→ Sky[x] joined	07:25
cbreak-work	jamescarr: be careful to not use a trailing /	07:25
	← tjbiddle left	07:25
	→ hoschi joined	07:26
	← infynyxx left	07:26
	→ toothr joined	07:26
	← francisfish left	07:26
	→ Turicas joined	07:27
	→ soee joined	07:28
	→ jMCg joined	07:28
jMCg	Hey folks o/~	07:28
	← netiul left	07:29
	← hoschi left	07:29
jMCg	We've got our git repositories mirrored onto the redmine server every 5 minutes by cron - is there any way or another to stop cron mailing us when everything went actually fine?	07:29
_ikke_	Not really a git question, is it?	07:30
	→ Guest14353 joined	07:31
	← drodger left	07:31
	← etcetera left	07:31
	→ hoschi joined	07:32
	→ drodger joined	07:32
jMCg	_ikke_: What I'm saying is: How do I stop git from reporting what is actually a success on stderr, and only report errors on stderr. Getting mails like this: http://apaste.info/ajvT is not useful.	07:33
jamescarr	cbreak-work: thanks	07:33
	→ papegaaij joined	07:34
	→ alan__ joined	07:34
mgedmin	jMCg, redirect stdout to /dev/null? I don't think that message is printed to stderr	07:34
	← mishok13 left	07:34
	→ jarr0dsz joined	07:35
	→ netiul joined	07:35
	→ giallu joined	07:36
	← isomorph_ left	07:37
	→ GeertV joined	07:37
	→ lng joined	07:37
thiago	jMCg: git push --quiet	07:37
	→ ErKa joined	07:38
	← alan__ left	07:38
lng	Hi! I have rebsed old commit message but could not push now - error: failed to push some refs	07:38
	← Error404NotFound left	07:38
cbreak-work	lng: you rewrote public history	07:38
lng	cbreak-work: it is okay	07:39
cbreak-work	you can either give up and use the old one, or kill the old history and force push	07:39
	or merge old and new history by pulling, and then push	07:39
lng	cbreak-work: need to force it	07:39
	→ isomorph joined	07:39
	→ arvind_khadri joined	07:39
lng	cbreak-work: i tried merging but message appeared to be as old version	07:39
	→ etcetera joined	07:40
lng	git push -f worked	07:40
	← lausi left	07:40
cbreak-work	yes. that'll kill remote history and replace it with yours	07:40
	← Guest14353 left	07:40
	→ lausi joined	07:40
lng	great	07:41
	← rtjure left	07:41
lng	thanks	07:41
	← etcetera left	07:41
	→ diegoviola joined	07:42
	← relipse left	07:42
	→ _goTAN joined	07:42
	→ fRAiLtY- joined	07:43
	→ Cromulent joined	07:44
	← GeertV left	07:44
	→ ewokchewy joined	07:45
	→ GeertV joined	07:45
	← isomorph left	07:45
	→ Error404NotFound joined	07:46
	→ isomorph joined	07:46
	→ mastro joined	07:47
	→ Xethron joined	07:48
	← dreiss left	07:48
	→ Remram joined	07:48
	← giallu left	07:49
	→ giallu joined	07:49
	← tapout left	07:49
	← simesy left	07:52
	← Cromulent left	07:52
	→ asaf joined	07:52
	← nigelb left	07:52
	→ Gonzo89 joined	07:52
	→ nigelb joined	07:52
	→ dankest joined	07:52
	← isomorph left	07:55
jMCg	thiago: apparently we have it implemented with git remote update. And that doesn't seem to take any --quiet flags	07:55
	← megha left	07:56
	← Immutable left	07:56
	→ jceb joined	07:56
	← pppaul left	07:56
	→ infynyxx joined	07:56
	→ Immutable joined	07:56
	→ pantsman joined	07:57
	← pantsman left	07:57
	→ pantsman joined	07:57
_ikke_	jMCg: git remote update > /dev/null	07:58
	← letharion left	07:58
	→ tapout joined	07:59
	→ flijten joined	07:59
jMCg	_ikke_: again: It's coming out on stderr.	08:00
	→ tim joined	08:01
	→ pppaul joined	08:02
	← beautifulmind left	08:02
	→ JohnBat26 joined	08:02
	→ TheFuzzball joined	08:03
jMCg	But looking at the code, all that git remote update does, the way we're calling it is: git fetch --all - so I can replace our cron job with git fetch --quiet --all	08:03
	→ xxc_michael joined	08:05
	→ xxc_michael_ joined	08:05
	JEEBcz → JEEB	08:05
	← JEEB left	08:05
	→ JEEB joined	08:05
	← xxc_michael left	08:05
	→ mishok13 joined	08:05
	← infynyxx left	08:06
	← xxc_michael_ left	08:06
	→ xxc_michael joined	08:06
	→ xxc_michael_ joined	08:07
	→ SaCu joined	08:07
_ikke_	jMCg: Odd that it ouputs that to stderr	08:07
	← toothr left	08:08
cbreak-work	stderr makes sense, doesn't it?	08:08
	it's output for the user, not an other program	08:08
	→ relipse joined	08:09
	→ relipse2 joined	08:10
	← relipse2 left	08:10
	→ maestrojed joined	08:10
	→ adamben joined	08:10
_ikke_	Well, it's a bit odd to report: "Everything went fine" to stderr	08:11
	← bababooey left	08:11
	→ unix joined	08:11
	→ joshsmith joined	08:11
	← dankest left	08:11
	← arvind_khadri left	08:13
	→ arvind_khadri joined	08:13
cbreak-work	why?	08:13
	stdout and stderr are just ways to separate user facing and program facing output	08:13
	→ tPl0ch joined	08:13
cbreak-work	that way you can chain programs with \| and still give output to the user	08:13
_ikke_	The name of stderr implies errors	08:13
cbreak-work	so, it's a bad name.	08:14
_ikke_	yes	08:14
	stdstat	08:14
cbreak-work	not like git and posix isn't littered with those.	08:14
_ikke_	:P	08:14
	→ luc___ joined	08:15
	← bauruine_ left	08:16
jMCg	There are two hard challenges in computer science: Naming Things. Cache Invalidation. Off-by-One Errors.	08:16
	→ suy joined	08:16
_ikke_	Yup	08:16
	→ toothr joined	08:17
	→ OOPMan joined	08:17
	← arvind_khadri left	08:17
	← albel727 left	08:17
	→ arvind_k joined	08:18
	→ Pugnax joined	08:18
	← suy left	08:19
	← snql left	08:19
	← luc_ left	08:19
	→ JDuke128 joined	08:20
	→ albel727 joined	08:20
	← Sigma left	08:21
	→ JeremyWei joined	08:21
	→ rob_w joined	08:21
	→ simesy joined	08:22
	→ adam12 joined	08:22
	→ AdrienBrault joined	08:23
	← AdrienBrault left	08:24
	← luc___ left	08:24
	→ avjpl joined	08:24
	→ anderse joined	08:24
	← avjpl left	08:24
	→ luc_ joined	08:25
	→ isomorph joined	08:25
	← Remram left	08:25
	← dvide left	08:25
	→ teask joined	08:26
	→ Cromulent joined	08:26
	ferdev\|away → ferdev	08:27
	→ alorence joined	08:29
	→ mgedmin_ joined	08:29
	← thiago left	08:30
	→ tjbiddle joined	08:30
	← tjbiddle left	08:32
	← mtk left	08:32
	← ErKa left	08:32
	← marthinal left	08:32
Infiltrator	Hah.	08:33
	→ marthinal joined	08:35
	→ mtk joined	08:36
	→ suy joined	08:37
	← pppaul left	08:38
	→ boichev joined	08:38
	← boichev left	08:38
	→ maxmanders joined	08:41
	→ sjefen6 joined	08:41
	← pantsman left	08:41
	← isomorph left	08:42
	← mgedmin_ left	08:42
	← betz2 left	08:43
	← anderse left	08:44
	→ goshawk joined	08:45
	← thyseus left	08:45
	← Cromulent left	08:46
	→ psoo joined	08:46
	→ thyseus joined	08:47
	→ boichev joined	08:47
	→ valgrind joined	08:48
	← sjefen6 left	08:48
	→ dreiss joined	08:48
	→ bootstrapper joined	08:49
	→ alexwhitman joined	08:52
	← fission6 left	08:53
	← flijten left	08:53
	→ flijten joined	08:53
	→ Sigma joined	08:53
	→ LeMike joined	08:54
	← joshsmith left	08:54
	← jargon left	08:55
	→ jargon joined	08:55
	← dreiss left	08:56
	→ howlymowly joined	08:56
	→ laurav joined	08:57
	← flijten left	08:57
	→ Element9 joined	08:57
	← howlymowly left	08:58
	→ sjefen6 joined	08:58
	→ timj1 joined	08:58
	→ asaf1 joined	09:00
	← bootstrapper left	09:00
	← timj left	09:00
	→ flijten joined	09:01
	→ youf joined	09:01
	→ kevinfagan joined	09:02
	← ns5 left	09:02
	← asaf left	09:03
	→ QaDeS joined	09:06
	← WorkeD left	09:07
	→ letharion joined	09:07
	→ alkisg joined	09:09
	→ betz joined	09:09
	gitinfo set mode: +v	09:11
	→ francisfish joined	09:11
alkisg	Hi, I'm completely new to git. I'm doing: git rm data/themes/greek/words/easy.txt; git add .; git status, and it tells me that:# renamed: data/themes/greek/words/easy.txt -> data/themes/greek/words/words1.txt	09:11
	But that didn't happen, I deleted easy.txt and I added words1.txt, why would git think I renamed the file?	09:11
	→ ozette joined	09:12
	→ BiggFREE joined	09:13
alkisg	Ah, it probably happened because I deleted the file from nautilus before issuing "git rm file". By allowing git to do the actual file deletion, it seems OK now.	09:14
	← Pugnax left	09:14
	→ m0viefreak joined	09:14
	← m0viefreak left	09:14
	→ m0viefreak joined	09:14
alkisg	Err no sorry it's still happening	09:15
	→ bartman joined	09:16
	→ bauruine_ joined	09:17
	← betz left	09:17
	← ferdev left	09:18
	→ Damgaard joined	09:19
alkisg	Here are my git commands and their output in a pastebin: http://paste.debian.net/235853/	09:20
	← tevfik left	09:20
	→ tevfik_ joined	09:20
alkisg	Do I really have to separate that into 2 commits?	09:20
Bombe	Separate what?	09:21
	Ah.	09:21
_ikke_	alkisg: Do those files have similar content?	09:21
	→ Kicer86 joined	09:21
alkisg	Add file "a", remove file "b", it says "a was renamed to b" instead of what I want	09:21
	→ ivan\ joined	09:21
_ikke_	alkisg: And for what it matters, git does not record renames	09:21
alkisg	_ikke_: yes, but only a bit	09:21
Bombe	alkisg, git will do what you want. It only shows it differently.	09:22
alkisg	Ah, so I can just go ahead and commit it? Thanks guys	09:22
	→ ThePawnBreak joined	09:22
_ikke_	alkisg: yes	09:23
	alkisg: It only matters when you tell git to follow renames	09:23
	with commands like git diff and git log	09:23
alkisg	Which I'm not, so all is fine (except for the really confusing output), thanks again	09:23
	→ chichou joined	09:24
ThePawnBreak	hello ; take a look here: https://github.com/ThePawnBreak/unknown-horizons/commits/master ; I'd like to merge the last 4 commint into a single one; I've been trying to do it using rebase, but I've no idea what I;m doing; can anybody help, please?	09:25
	→ acrocity joined	09:25
m0viefreak	git rebase -i 75a3a	09:26
	change pick / pick / pick / pick to pick / squash / suqash / squash	09:26
	but:	09:26
	!rewrite	09:26
gitinfo	[!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to). If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstream_rebase	09:26
	← laurav left	09:27
ThePawnBreak	history is not a problem, nobody pulled this yet	09:27
m0viefreak	ok	09:27
	→ laurav joined	09:27
	→ bootstrapper joined	09:27
ThePawnBreak	Could not apply 9b9867b... Add necessary resources to tooltip	09:29
	m0viefreak: what do I do now?	09:31
	← maestrojed left	09:31
m0viefreak	uhm	09:31
	that shouldnt happen, it should apply cleanly	09:31
cbreak-work	ThePawnBreak: git rebase --abort and do it again, this time correctly	09:31
	(what did you select?)	09:31
ThePawnBreak	http://pastebin.com/5bKr4nXm	09:32
	I selected pick for the first one and squash for the rest	09:32
	→ vic2011 joined	09:32
	→ maestrojed joined	09:33
	← bauruine_ left	09:34
	← vic2011 left	09:34
cbreak-work	that should work normally. Does that range contain merges?	09:34
	→ jkremser joined	09:34
	→ snql joined	09:34
	← ewokchewy left	09:34
	→ zoresvit joined	09:35
ThePawnBreak	I think so; there will be a merge conflict when I push; I tried to do a merge earlier but couldn't manage to get it done properly	09:35
m0viefreak	well it looks like you have local changes then	09:35
	and your local repo isn't the same as the github link	09:35
ThePawnBreak	so what should I do?	09:35
	→ phluks joined	09:36
	← GeertV left	09:36
m0viefreak	if the merge isn't what you wanted and you want to discard any of the local changes you can reset your branch to the state of the github link	09:36
	← bootstrapper left	09:36
	← soee left	09:37
	→ GeertV joined	09:37
m0viefreak	git reset --hard remotes/<yourgithubremote>/master	09:37
ThePawnBreak	the local version is the correct version; I want the local version on github with only one commit in the history instead of 4	09:37
	→ ferdev joined	09:37
m0viefreak	then you need to show us your local history	09:37
ThePawnBreak	how?	09:37
m0viefreak	!lol	09:37
gitinfo	A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	09:37
	→ spuz joined	09:37
ThePawnBreak	can I just print a log on pastebin?	09:38
m0viefreak	copy/paste ist	09:38
	← Tom234 left	09:38
	← laurav left	09:39
fsvehla	Is there a plumbing command that returns the current branch name?	09:39
	→ nikolaidante joined	09:39
	← tevfik_ left	09:39
nikolaidante	Maxdamantus: what's wrong with this code? http://pastie.org/6234498#8; For fb-like button	09:39
	Maxdamantus: It doesn't actually show anything	09:40
	← JeremyWei left	09:40
m0viefreak	fsvehla: git symbolic-ref HEAD	09:40
ThePawnBreak	m0viefreak: is there any way to just force the github version to be exactly like the local one, including the history, while avoiding any merges?	09:41
sajbar	nikolaidante: ahem, how does this relate to git?	09:41
fsvehla	m0viefreak: That’s it. thanks	09:41
m0viefreak	ThePawnBreak: yes, push -f	09:41
	← herdingcat left	09:42
Maxdamantus	nikolaidante: you can't use document.getElementsByTagName until the document's loaded.	09:43
nikolaidante	sajbar: oh, sorry	09:43
	← Damgaard left	09:43
	→ howlymowly joined	09:43
nikolaidante	Maxdamantus: Oh, thanks	09:43
ThePawnBreak	m0viefreak: the local history seems to be different from the one on github, even after push 0f	09:44
m0viefreak	that can't be	09:44
	→ Damgaard joined	09:44
cbreak-work	maybe you pushed the wrong branch.	09:44
m0viefreak	just copy/paste the first bunch of lines of	09:44
	git log --oneline --graph --decorate --all	09:44
	← weiyang left	09:44
ThePawnBreak	http://pastebin.com/KP4JN7yC	09:45
	→ bauruine joined	09:46
m0viefreak	ah i see, so they are merges	09:46
	← zoresvit left	09:46
ThePawnBreak	I want only one commit saying Add necessary resources tooltip to build menu	09:46
	→ beautifulmind joined	09:47
	← sajan left	09:47
ThePawnBreak	can I do that?	09:47
cbreak-work	don't mess with merges.	09:47
	← Turicas left	09:48
	← maestrojed left	09:48
cbreak-work	you can reset them away I guess. Or you can linearize history	09:48
ThePawnBreak	cbreak: this is a trivial 10 line change, nobody pulled the change, I want to make only one commit	09:48
cbreak-work	but everything seems pushed	09:48
ThePawnBreak	cbreak-work: please be more explicit; I have no experience using git	09:49
	→ AdrienBrault joined	09:49
cbreak-work	well, go ahead then. reset --soft back to the past, then make a single commit	09:49
	→ zenx joined	09:49
cbreak-work	for example, git reset --soft 75a3a62	09:49
ThePawnBreak	what does that do?	09:49
cbreak-work	that will make you be on 75a3a62, but with all the staged changes from HEAD	09:49
	→ myx joined	09:50
	← AdrienBrault left	09:50
ThePawnBreak	then I commit and push?	09:50
cbreak-work	then you can just git commit -c ef63fe2	09:50
	and look at history if it is as you want	09:50
	→ eijk joined	09:50
cbreak-work	(be sure git show shows what you want)	09:50
	→ soee joined	09:50
	→ alkino joined	09:51
ThePawnBreak	commit -c ef63fe2 -- what is -c. what is ef63... ?	09:51
m0viefreak	-c == take commit message from	09:51
	but before you do anything: you are in detached head state	09:51
ThePawnBreak	oh , I'll just write it again	09:51
m0viefreak	checkout master first	09:51
	→ AdrienBrault joined	09:51
	→ aero1 joined	09:52
	→ lionel_ joined	09:53
aero1	is there any presentation that was aimed at "Git for Management" aka "Git for Pointy Haired Bosses"	09:53
ThePawnBreak	m0viefreak: afte checkout master I lost all my changes, now I have 4 commits again :((	09:54
	→ FernandoBasso joined	09:54
m0viefreak	yes, now do what cbreak said	09:54
	git reset --soft ef63fe2	09:54
	git commit	09:54
	enter your new message	09:54
	done	09:54
	uhm wait	09:55
	git reset --soft 75a3a62	09:55
	ofcourse	09:55
ThePawnBreak	already used ef6..	09:55
	gitinfo set mode: +v	09:55
ThePawnBreak	can i go back?	09:55
lionel_	hi all, i have a very very strange pb with git, a read error:bad file descriptor on a fetch, but git branch -r works ???	09:55
m0viefreak	just enter git reset --soft 75a3a62	09:56
	← teask left	09:56
lionel_	any idea ?	09:56
m0viefreak	then use git diff to verify what you see is correct	09:56
	then git commit	09:56
ThePawnBreak	git diff outputs nothing	09:56
Maxdamantus	lionel_: `branch -r` just lists branches under remotes.	09:57
m0viefreak	give us another log	09:57
ThePawnBreak	ok, my file is correct now	09:57
	and the log shows none of my commits	09:57
Maxdamantus	lionel_: it doesn't ask for anything outside of the repository.	09:57
	→ arietis joined	09:57
m0viefreak	oh	09:57
ThePawnBreak	now I just commit and push -f, right?	09:57
m0viefreak	use git diff --cached	09:57
	if thats right, commit	09:57
	then push -f	09:57
lionel_	ah, it use the local copy of the remote ?	09:57
Maxdamantus	lionel_: yes.	09:57
lionel_	while fetch goes really looking for the remote data ? that's it ?	09:58
Maxdamantus	Yes. fetch could try to open an SSH session or connect to a git or http server.	09:58
ThePawnBreak	m0viefreak: git status says : # Your branch and 'origin/master' have diverged,	09:58
	# and have 1 and 4 different commits each, respectively.	09:58
	; the log looks ok; do I still push?	09:58
m0viefreak	yeah thats what you get when you rewrite pushed history	09:59
lionel_	in our case, the repository is another folder on the same server	09:59
m0viefreak	just push -f now and overwrite github's history	09:59
	→ nkr joined	09:59
ThePawnBreak	yay, it worked; thanks, m0viefreak	09:59
Maxdamantus	lionel_: unless you've set up a file:// remote (or just one with no explicit protocol), it won't look like the "same" server to git.	10:00
	→ zoresvit joined	10:00
	→ gyre007 joined	10:00
Maxdamantus	lionel_: what does the remote look like?	10:00
cbreak-work	lionel_: git remote -v	10:00
	→ private joined	10:01
lionel_	it has no explicit protocol	10:01
	→ ryez joined	10:01
	private → Guest79562	10:01
Maxdamantus	Try using file://	10:02
	→ kcj joined	10:03
	→ anderse joined	10:04
	← maleknet left	10:05
	→ saki007s_ joined	10:05
	→ Guest55175 joined	10:06
	← goshawk left	10:07
	→ gustavnils joined	10:08
	← xiangfu left	10:08
	← saki007ster left	10:08
	→ maleknet joined	10:09
lionel_	doing that	10:10
	← khmarbaise left	10:10
	→ khmarbaise joined	10:10
	→ jerikl joined	10:12
	→ cristi_ joined	10:12
	cristi_ → Guest4238	10:13
	← Guest4238 left	10:13
	→ kandinski joined	10:14
lionel_	ah, learn a new thing : the git commands are launch via a daemon. Is there gotchas with that ?	10:15
	← ThePawnBreak left	10:15
cbreak-work	environment.	10:16
	← _goTAN left	10:16
	← jdyin left	10:17
lionel_	@cbreak could you elaborate ? what could I check ?	10:18
cbreak-work	check the environment variables.	10:18
_ikke_	via a daemon?	10:18
cbreak-work	check the file system permissions	10:18
	check if it works if you do it yourself	10:19
	← Guest55175 left	10:19
	→ ThePawnBreak joined	10:19
	→ xiangfu joined	10:19
	← kcj left	10:20
	← JDuke128 left	10:20
	→ sdeboey joined	10:22
	← BiggFREE left	10:22
	→ DachesBella joined	10:25
lionel_	env variable are ok, I added GIT_TRACE=1	10:25
	← Guest79562 left	10:26
	← lifelike left	10:26
	← LimeBlast left	10:27
	→ lifelike joined	10:27
	→ wasanzy joined	10:28
	→ mnx\| joined	10:28
	→ jamescarr_ joined	10:28
	→ marhaban joined	10:28
	← jamescarr left	10:29
	jamescarr_ → jamescarr	10:29
	→ phiwo joined	10:29
	→ tevfik joined	10:29
lionel_	ohoh, i got fatal: read error: bad file descriptor on exec: '/bin/sh' '-c' 'git-upload-pack '\''/servers/ataoservcom.git/'\''' 'git-upload-pack '\''/servers/ataoservcom.git/'\'''	10:30
	@cbreak, I am remote and sadly don't have a direct access or ssh...	10:31
	← jamescarr left	10:32
	→ LimeBlast joined	10:33
	→ RaceCondition joined	10:33
	← DachesBella left	10:33
	→ chrisward joined	10:34
	← myx left	10:34
	chrisward → Guest54377	10:34
	→ jamescarr joined	10:35
	→ myx joined	10:36
	← xxc_michael left	10:37
	← xxc_michael_ left	10:37
	→ xxc_michael joined	10:37
	→ xxc_michael_ joined	10:37
	← Desproges left	10:37
	← RaceCondition left	10:38
	→ cebor joined	10:40
	→ Desproges joined	10:40
	→ dherzog joined	10:41
	← eletuchy left	10:42
	→ eletuchy joined	10:42
	→ gusnan joined	10:43
	→ bootstrapper joined	10:45
	← OOPMan left	10:47
lionel_	this is something wrong with the daemon, I suspect some pipes redirections but I cannot do more for now, I will see if the on site operator can see what it is.	10:47
	→ luc___ joined	10:48
	→ Imaginativeone joined	10:49
	→ heidar joined	10:49
	→ bzyx joined	10:50
	→ bzyx_ joined	10:50
	← simesy left	10:51
	← khmarbaise left	10:51
	← cagedwisdom left	10:51
	← luc_ left	10:51
	→ rgr joined	10:52
	← xxc_michael_ left	10:54
	→ Pugnax joined	10:54
Cork	i'm trying to make an alias that goes to the root of the git tree, but i can't figure out what i'm doing wrong	10:56
	cdroot = !bash -c 'echo cd ./$(git rev-parse --show-cdup)'	10:56
	← Sky[x] left	10:57
	← bootstrapper left	10:57
	← bzyx left	10:57
jast	Cork: what happens?	10:58
Cork	it goes to the current map :/	10:58
	cd ./	10:58
	$(git rev-parse --show-cdup) isn't executed...	10:58
	or something	10:58
	but theres where i fail to understand why	10:59
	← fisted left	11:00
Cork	aha	11:01
	→ fisted joined	11:01
Cork	you can't do cd in a git alias...	11:01
	→ aspiers joined	11:01
Cork	it will revert back to your current location when its done	11:01
	← stliu left	11:02
	→ khmarbaise joined	11:02
	← davveo left	11:02
	→ RaceCondition joined	11:04
jast	yeah, it's executed in a subshell, after all	11:05
	also, apparently shell aliases run from the top level of the git repo...	11:05
charon	yeah... hysterical raisins	11:05
	← luc___ left	11:05
	← marthinal left	11:06
charon	you have to use a git-foo executable if you want to work around this	11:06
jast	fun	11:06
	← dhruvasagar left	11:06
charon	or maybe someone invented a silly syntax like !$@#%&! to denote that you want it to run in the current dir	11:06
	→ Merwin joined	11:07
	→ aqabiz joined	11:10
	→ zerba joined	11:11
charon	oh, a different solution was found: http://article.gmane.org/gmane.comp.version-control.git/172198	11:11
	it's magic!	11:11
	Cork: ^	11:12
	→ Yuuhi` joined	11:12
	→ marthinal joined	11:12
Cork	charon: thx	11:12
	← dherzog left	11:13
	← Pugnax left	11:13
	→ the_cybersphinx joined	11:13
	← marthinal left	11:14
	← Yuuhi left	11:14
	← rgr left	11:15
Cork	charon: not sure how that differs from the problem i have though	11:15
	only difference i can see is that is uses sh	11:16
charon	erm, ok, i only read enough of the backlog to understand your actual issue just now	11:16
	and the answer is: you can't do that from within git because the working directory is a property of the shell and only the shell itself can change it	11:16
	← snql left	11:16
charon	so you need a shell alias of some sort	11:16
Cork	yup	11:17
	that is what i found out before :)	11:17
	when i tested echo $PWD and it showed that it already was in the root directory i managed to put two and two together	11:17
	→ rosco_y joined	11:18
charon	alias reporoot='test -n "$(git rev-parse --show-cdup 2>/dev/null)" && cd "$(git rev-parse --show-cdup)"'	11:18
	or some such	11:18
	← Damgaard left	11:18
charon	bonus points for refactoring the two calls to git into a variable	11:18
	→ snql joined	11:18
	→ plumpkin joined	11:18
rosco_y	I used "git rm myFile" and now myFile shows in the status as "deleted". How do I add it back to my commit list as a tracked object?	11:19
	→ Gentlecat joined	11:19
	← hyperair left	11:19
	→ daniel_- joined	11:19
	→ _goTAN joined	11:19
charon	rosco_y: read the commentary in 'git status' output	11:20
rosco_y	charon: ty, I did try "git add", but that had no effect....	11:20
Cork	charon: a simple export gitroot='cd ./$(git rev-parse --show-cdup)' is all thats needed	11:20
charon	that's because git-rm deleted the file, and you can't add a nonexistent file	11:20
	Cork: umm... so you call that as $gitroot?	11:21
aero1	is there any presentation that was aimed at "Git for Management" aka "Git for Pointy Haired Bosses"	11:21
rosco_y	it physically deletes the file?	11:21
Cork	no just gitroot	11:21
charon	s/export/alias/ then?	11:21
Cork	ya	11:21
	sorry	11:21
	blackout :D	11:21
	← daniel_- left	11:21
charon	rosco_y: of course, saying 'git rm file' means 'please delete <file> and consider yourself notified of the deletion'	11:21
arand	rosco_y: yes, and stages the deletion for next commit.	11:21
rosco_y	well, it's not the end of the world. If that's all it took to derail the train, I'd definately have to quit.	11:22
	← ferdev left	11:22
charon	rosco_y: git-status should point you at 'git checkout HEAD -- <file>' to restore it (assuming you haven't committed the deletion yet)	11:22
rosco_y	ty charon, arand, Cork et. all., "I think I can, I think I can"....	11:23
	← youf left	11:23
	→ ferdev joined	11:24
charon	rosco_y: do you understand on a conceptual level what the staging area does? otherwise you should probably read "the thing about git" or some such	11:24
	after that it gets easier: git rm F = rm F && remove F from the staging area completely	11:25
	→ bootstrapper joined	11:25
charon	which will appear to stage a deletion, but really the effect is just that the next commit doesn't contain F.	11:25
	← jaequery left	11:25
rosco_y	charon: ty. I'm actually getting fairly comfortable with git, I was just surprised that it actually deleted the file prior to the commit	11:25
	I love git, it's really working well for me	11:25
spuz	aero1: I don't see what git has to do with point-haired managers - they shouldn't care about it	11:26
rosco_y	I'm branching with impunity and everything!	11:26
	rosco_y takes a bow	11:26
rosco_y	just throw money	11:26
	← RaceCondition left	11:26
	← jerikl left	11:26
rosco_y	so, even beginners can benefit with git	11:26
aero1	spuz: I have a manager. They dictate what infrastructure gets paid for. My shop doesn't already have VC set up. I need to convince them that it's worth it to invest in training people to use it etc.	11:26
rosco_y	ty :)	11:26
spuz	aero1: git is not an infrastructure	11:27
	maybe you're thinking of github or some other service	11:27
aero1	...	11:27
charon	aero1: if anyone knows about this, surely it's sitaram	11:27
aero1	neckbeards	11:27
	You know exactly what I am talking about	11:27
rosco_y	aero1: I once had a manager who really tried to keep us developers happy, but we didn't know it--he was really underappreciated....	11:27
	a rarity for sure	11:27
aero1	Indeed	11:27
	I already have github+gitolite etc etc etc already set up	11:28
	→ dhruvasagar joined	11:28
	← marhaban left	11:28
rosco_y	He had his faults, but we all do....	11:28
aero1	I just need to explain to a PHB what exactly git is so I can get the A-OK to provide paid-for or time-away from work to provide training to people.	11:29
rosco_y	crap, I gotta get to it	11:29
aero1	I'm sure I've seen something like this	11:29
rosco_y	c-ya :)	11:29
aero1	cya	11:29
	← bzyx_ left	11:30
spuz	aero1: wait you said your shop doesn't already use VC, maybe you want an article explaining version control	11:31
aero1	Exactly how dense are you?	11:31
	← yshh left	11:31
	→ yshh joined	11:32
spuz	aero1: sorry - I'm just trying to help	11:32
aero1	I use version control, I use all the niceties that surround git. I love it. My place of work doesn't. I explained that already.	11:32
	Nevermind, you don't have an answer.	11:32
spuz	aero1: I understand that - that's why I suggested you might like to explain to your bosses the benefits of version control	11:33
	I'm only saying that because getting into the details of git over other tools might not be what you want	11:33
	→ Sky[x] joined	11:33
aero1	My god - that's exactly what I was asking for! A presentation/material of this kind of thing to a PHB.	11:33
	nevermind	11:33
	← jamescarr left	11:34
aero1	I found something	11:34
	http://tosbourn.com/2011/03/git/explaining-version-control-git-in-simple-terms/	11:34
	← jargon left	11:35
spuz	looks good	11:35
	→ murosai joined	11:35
murosai	i accidently commited to master but did not push yet, how can i undo this commit and commit it instead to "develop" branch?	11:36
	→ OOPMan joined	11:36
	← yshh left	11:36
nicoo	murosai: git reset HEAD~1	11:37
	git stash	11:37
	git co develop	11:37
	git stash pop	11:37
	→ marthinal joined	11:37
nicoo	I would say	11:37
	← bootstrapper left	11:37
	→ zoonfafer joined	11:38
nicoo	Sorry, probably HEAD^1 instead of HEAD~1	11:39
	→ johnny_rose joined	11:39
arand	Umm, HEAD^ and HEAD~1 are equivalent, right? Does HEAD^! even work?	11:39
	*HEAD^1	11:40
jast	HEAD^1 == HEAD~1 == HEAD^ == HEAD~	11:40
arand	Oh, well then.	11:40
jast	and HEAD^^ == HEAD~~ but HEAD^2 != HEAD~2 :}	11:40
aero1	so much head	11:41
nicoo	jast: That's the « HEAD^2 != HEAD~2 » that confused me :P	11:41
_ikke_	nicoo: ^n means nth parent. ~n means nth ancestor	11:42
	← pity left	11:42
	alkisg → work_alkisg	11:42
nicoo	_ikke_: That you :)	11:43
	Thank*	11:43
_ikke_	^1 and ~1 are the same	11:43
	→ jargon joined	11:44
	← jargon left	11:44
	→ jargon joined	11:44
	→ marhaban joined	11:47
	→ hatake joined	11:49
	→ rakl joined	11:49
	→ chisler joined	11:50
	← rakl left	11:50
nicoo	Yeah, read that	11:51
	← FernandoBasso left	11:52
	→ JPeterson joined	11:52
	→ avjpl joined	11:53
	← avjpl left	11:53
	→ avjpl joined	11:53
	← jargon left	11:55
murosai	seems like the git stash fails because it can't create some files?	11:55
	→ albel727_ joined	11:56
sajbar	!repro	11:56
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.	11:56
	← alkino left	11:56
	← albel727 left	11:58
	→ RaceCondition joined	11:59
	← avjpl left	11:59
	→ bootstrapper joined	12:04
	← ferdev left	12:04
	← rosco_y left	12:05
	← Sky[x] left	12:05
	→ ferdev joined	12:05
	→ Sky[x] joined	12:06
	← blackmat` left	12:07
	→ blackmat` joined	12:07
	← marhaban left	12:08
	→ marlun joined	12:08
	→ alkino joined	12:10
marlun	If I got a file which has changed in 4 commits and in the first change it was converted by mistake to utf8 and all the swedish letters (å,ä,ö) failed to be converted so now we've got ? in their place. Any tips on how we could fix the encoding errors but still have all the other changes? :)	12:11
	← giallu left	12:11
_ikke_	marlun: Maybe git checkout -p, but that only works if the corrupt characters are in their own hunks	12:12
	← phluks left	12:13
	→ phluks joined	12:14
	← juvenal left	12:14
	← bootstrapper left	12:15
	← Kicer86 left	12:18
	→ z8z joined	12:19
	← gustavnils left	12:20
	→ cdeszaq joined	12:21
sitaram	aero1: you should be looking for horror stories not good stuff	12:22
	→ juvenal joined	12:24
sitaram	horror stories from not using a VCS, I mean. (In case you turn around and ask me "how dense are you" like you did that poor guy who was trying to help)	12:24
	charon: most of my efforts have been svn versus git, not "nothing" versus git. (Also clearcase in some projects)	12:25
	TCS has long had policies that no project shall run without a VCS. It's a different matter that the policy doesn't say anything about using brain dead ones :)	12:26
	→ iflista joined	12:26
	→ Damgaard joined	12:26
	← arietis left	12:28
	→ sgambino joined	12:29
	→ arietis joined	12:30
	← papegaaij left	12:30
	→ papegaaij joined	12:30
jast	horror stories first, then good stuff	12:30
fasta	What does am in (git)-[master\|am]- mean?	12:31
	← aqabiz left	12:31
jast	fasta: is that some kind of shell prompt enhancement? probably that you're on master and there's an unfinshed rebase/am operation	12:32
	(am as in man git-am)	12:32
gitinfo	the git-am manpage is available at http://jk.gs/git-am.html	12:32
	← phiwo left	12:32
fasta	jast: nothing to commit (working directory clean)	12:32
	jast: it is what zsh displays via vcs_info.	12:33
jast	then you (or whoever is faced with this) probably forgot to finish the operation and just continued as normal	12:33
	← RaceCondition left	12:33
	→ RaceCondishun joined	12:33
jast	for example, suppose you do a rebase and it comes up with conflicts	12:33
	now, suppose you miss that fact and just keep working normally, stage things, commit stuff, etc.	12:34
	now you have a clean working dir, but the rebase is still in progress	12:34
fasta	jast: the last commit is by Linus. I didn't do anything.	12:34
jast	if you were to type 'git rebase --continue' in that situation, unspeakable things would happen	12:34
	← fsvehla left	12:34
jast	well, maybe the 'am' operation applied commits by linus	12:34
	← xiangfu left	12:34
jast	or someone did a 'git pull' with the 'am' still unfinished	12:35
fasta	How can unfinished operations be shared via the network?	12:35
jast	they can't	12:35
fasta	That seems like a very bad idea.	12:35
	← arietis left	12:35
jast	rebase/am progress is always completely local	12:35
	← Imaginativeone left	12:35
	→ Infiltrator joined	12:35
jast	it's completely impossible for it to be transmitted from somewhere else via git-native methods	12:35
fasta	So, how is it then possible that it shows it?	12:36
jast	probably because the person who works in this local repository did a corresponding operation at some point in the past	12:36
	→ R1KRDO joined	12:36
	→ danillonunes joined	12:37
jast	might have been months ago if this is the first time you're actually displaying this information	12:37
fasta	jast: I now did a git reset --hard and it still shows it.	12:37
	→ corimarisol joined	12:37
	← corimarisol left	12:37
	→ corimarisol joined	12:37
jast	rebase and am progress has nothing to do with the index and the branches	12:37
	it's independent information, stored somewhere in the bowels of .git	12:37
	← marthinal left	12:38
	← corimarisol left	12:38
jast	you could do 'git am --abort', but it would restore your situation back to what it was before the operation began, possibly a long time ago	12:38
	you can also hunt down the data in .git and remove it	12:38
	→ corimarisol joined	12:38
	← corimarisol left	12:38
	→ corimarisol joined	12:38
jast	I'm not completely sure on the location, possibly something like .git/am or .git/applypatch or .git/dotest	12:38
	← AdrienBrault left	12:39
	→ bzyx joined	12:39
fasta	jast: so, you are saying it is impossible that I got the repository via a pull into this state?	12:40
jast	yes	12:40
	→ picoguy joined	12:40
jast	well	12:41
fasta	jast: well, then I must have gremlins on my system.	12:41
	← kevinfagan left	12:41
jast	if the detection of this vcs_info thing is poor, it might actually refer to a merge that's still in progress	12:41
	← chisler left	12:41
fasta	Your suggestion worked.	12:41
	For other repositories it returned more expected results.	12:41
	→ phiwo joined	12:41
fasta	Thank you	12:41
jast	but then again, a 'git reset --hard' does abort unfinished merges	12:42
	← sgambino left	12:42
	→ kevinfagan joined	12:42
	→ bioman_ joined	12:42
	← nikolaidante left	12:43
	→ marthinal joined	12:43
	← WinstonSmith left	12:43
	→ bootstrapper joined	12:45
	← bioman_ left	12:45
	→ bioman joined	12:47
	→ normanrichards joined	12:47
bioman	Hello :)	12:47
gitinfo	bioman: 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.	12:47
bioman	I'd like to use git revert <bad commit>	12:47
	I git kernel.org (3.8), then I add nouveau (open source Nvidia)	12:48
	← RaceCondishun left	12:50
	→ CreativeEmbassy joined	12:50
	→ _main_ joined	12:51
	→ etcetera joined	12:51
	← etcetera left	12:51
	→ WinstonSmith joined	12:51
	← robbyoconnor left	12:52
	→ sebuba joined	12:53
	← snql left	12:53
	← eletuchy left	12:54
	→ MrDHat joined	12:54
	→ pity joined	12:54
	← CreativeEmbassy left	12:54
	_main_ → __main__	12:55
	← bootstrapper left	12:55
bioman	Here is how I get kernel.org 3.8 : http://ix.io/4tn	12:55
	→ sq-one joined	12:56
	→ alexbligh joined	12:56
bioman	Here is how I add nouveau : http://ix.io/4to	12:57
alexbligh	What's the easiest way to do a git pull on all local branches and not change which branch is checked out?	12:57
	(against a single remote)	12:58
bioman	After that, I type git revert <older bad commit> and then it says "fatal: bad object <the commit I want to git revert>	12:58
	← sebuba left	12:58
	→ daniel_- joined	12:59
	→ saki007ster joined	12:59
spuz	alexbligh: you can't because a git pull can modify the state of a branch git requires you to have it checked out	12:59
bioman	I have to say I'm dump as regards git, I just can't understand how it works Q_Q	12:59
	← saki007s_ left	12:59
bioman	please help me	13:00
alexbligh	spuz, I am prepared for it to check each out in turn and restore	13:00
spuz	alexbligh: how would you resolve a conflict in that situation?	13:00
	→ Doomground joined	13:00
alexbligh	spuz, I suppose I'm interested on what your 'I'm going travelling, without connectivity" workflow. I'm guessing this involves "git fetch" then manually doing the merges.	13:01
	spuz, I'd hardly ever see merge conflicts on a pull as all the dev work is done on feature branches which are local.	13:01
spuz	alexbligh: you can do a git fetch --all to get all the commits that are on your remote	13:01
alexbligh	So I'd always see ff pulls	13:01
	and then how do you merge them in? IE do the pulls on each branch?	13:01
m0viefreak	check them out and do a git merge	13:02
	← bzyx left	13:02
spuz	alexbligh: either do a merge or a rebase or whatever you want to do with your remote/branch	13:02
m0viefreak	which will essentially fast-forward to your upstream branch	13:02
	→ warez joined	13:02
	→ cannonball joined	13:02
alexbligh	Oh right, so it's as simple as git checkout foobranch && git merge remotes/origin/foobranch	13:02
spuz	yep	13:02
m0viefreak	you dont even need the ref	13:02
	"git merge" is just fine	13:03
	→ hellopat_ joined	13:03
	← khmarbaise left	13:03
alexbligh	OK, so my workflow before losing connectivity is git submodule foreach git fetch --all (essentially).	13:03
	→ khmarbaise joined	13:04
	← johnny_rose left	13:05
	→ bwidmer joined	13:06
	← hellopat left	13:06
	← babilen left	13:06
jast	bioman: what does this say: git cat-file -t <commit you want to revert>	13:06
	← hellopat_ left	13:07
	→ bzyx joined	13:08
	→ RaceCondition joined	13:08
	← robustus left	13:09
	← saki007ster left	13:09
	→ robustus joined	13:10
	← MrDHat left	13:11
	← RaceCondition left	13:11
	→ MrDHat joined	13:11
	← MrDHat left	13:12
	→ nesti joined	13:12
	← z8z left	13:13
	→ hyperair joined	13:13
	→ freeed joined	13:14
bioman	jast: hello :) "error: unable to find 2c6a05f8d481fef46ea89248e4c8a4c9453cb493" "fatal: git cat-file 2c6a05f8d481fef46ea89248e4c8a4c9453cb493: bad file"	13:14
jast	that looks like your commit does not exist or is broken (e.g. due to filesystem corruption)	13:14
	→ itatitat joined	13:14
jast	you'll probably find that 'git log <commit>' won't work either	13:15
	→ CreativeEmbassy joined	13:15
jast	actually you'd get another error if the object was simply missing, so my guess is it's corrupted	13:16
	→ MrDHat joined	13:16
	→ sebuba joined	13:16
nicoo	.wg #code	13:16
	Oops, sorry	13:16
	→ johnny_rose joined	13:17
	→ chimay joined	13:17
	→ baba joined	13:18
	← plumpkin left	13:19
	← unix left	13:19
	→ sgambino joined	13:20
	→ AdrienBrault joined	13:21
bioman	jast: this commit exists : http://cgit.freedesktop.org/nouveau/linux-2.6/commit/?id=2c6a05f8d481fef46ea89248e4c8a4c9453cb493	13:23
	→ Guest6475 joined	13:25
	← alunduil left	13:27
	→ geekbri joined	13:28
bioman	jast: First of all, is my way to get kernel.org + nouveau with git ?	13:28
	→ whitemage joined	13:28
	→ IntricateRuin joined	13:29
	← normanrichards left	13:29
	ferdev → ferdev\|away	13:29
	← Fandekasp left	13:30
bioman	correct	13:31
	← __main__ left	13:31
	← whitemage left	13:32
	→ _main_ joined	13:32
	→ giallu joined	13:32
	← pretty_function left	13:33
	→ babilen joined	13:33
	← skyfaller left	13:34
	→ skyfaller_ joined	13:34
	← Guest6475 left	13:35
	→ skyfaller joined	13:36
	← skyfaller_ left	13:36
bioman	jast: ?	13:36
	→ MrDHat_ joined	13:36
	_main_ → __main__	13:36
	← NiteRain left	13:37
	← MrDHat_ left	13:37
	← MrDHat left	13:37
	← skyfaller left	13:37
	→ skyfaller joined	13:38
	→ jargon joined	13:38
	← jargon left	13:38
	→ jargon joined	13:38
jast	bioman: I dunno, I've never had any problems like that. perhaps freedesktop.org's server is broken? try fetching nouveau again in a separate repository and see if the same error shows up there	13:38
	→ anonymuse joined	13:38
	→ m0ltar joined	13:39
	→ W8uiE5 joined	13:41
	→ chisler joined	13:41
	→ asaf joined	13:42
	← asaf1 left	13:42
W8uiE5	git apply patch returned some whitespace errors. i manually removed the trailing whitespace errors, but still get error:patch failed filename:33 what can i do?	13:42
	→ achadwick joined	13:42
	← Gonzo89 left	13:43
	← bzyx left	13:44
	→ ojacobson joined	13:44
	← mnx\| left	13:45
	→ rmichnik joined	13:45
	← h0bbit left	13:45
bioman	jast: ok. I thank you. I'll let you know	13:46
	← marlun left	13:46
	← soee left	13:46
	→ soee joined	13:48
	← jarr0dsz left	13:49
	← CodeShark left	13:49
	← arand left	13:49
	→ JaredMiller joined	13:49
	→ kimico joined	13:50
	→ saki007ster joined	13:50
	← JaredMiller left	13:50
	→ etcetera joined	13:50
	← jonshea left	13:51
	← jkremser left	13:52
	→ jcromartie joined	13:52
	→ jonshea joined	13:52
	← Damgaard left	13:54
	→ maverick joined	13:54
	maverick → Guest90509	13:54
	→ jamescarr joined	13:56
	← Guest90509 left	13:56
	← adamben left	13:57
	→ KevinSjoberg joined	13:57
	← lng left	14:00
	← iflista left	14:01
	← Immutable left	14:02
	→ wting joined	14:03
	→ rolfvandekrol joined	14:03
	→ jfdi joined	14:03
	→ jamesez joined	14:03
	→ Nebros joined	14:04
	← Nebros left	14:04
	→ amigojapan joined	14:04
	← nilg left	14:04
	→ bootstrapper joined	14:04
	→ JDuke128 joined	14:05
	→ jkremser joined	14:05
	← beautifulmind left	14:05
	← rolfvandekrol left	14:06
	← Galaris left	14:06
	← jkremser left	14:07
	← Skaag left	14:08
	→ Skaag joined	14:08
	← itatitat left	14:09
	← kevinfagan left	14:09
	→ breakingthings joined	14:11
	← subhojit777 left	14:11
	→ nickn joined	14:13
W8uiE5	how can i check if a hook was called?	14:14
	← bootstrapper left	14:14
	← maleknet left	14:14
	← jamescarr left	14:15
	→ znode joined	14:15
ojacobson	W8uiE5: does it generate any output? Most hooks have their output sent to the git command's output.	14:16
	Alternately, check to see if the thing the hook does has been done.	14:16
	→ scampbell joined	14:16
	← breakingthings left	14:17
W8uiE5	ojacobson: just testing a simple pre-commit hook with 1 line: echo 'test'. then i make a commit, but theirs no test	14:17
	→ sspeaker joined	14:17
ojacobson	W8uiE5: is the hook executable? 'ls -l .git/hooks/' will tell you.	14:17
	← sspeaker left	14:17
	→ kukks joined	14:17
	← kukks left	14:17
	→ kukks joined	14:17
	→ arand joined	14:18
W8uiE5	ojacobson: i use win 7	14:18
	toothr → toothrot	14:18
YoungFrog	try using "touch ~/sometempfile", and see if that file was touched.	14:18
	→ jatt joined	14:19
W8uiE5	YoungFrog: touch a creates a if it does not exist. tocu a if it exist does not show a different time stamp	14:19
ojacobson	W8uiE5: it also updates the mtime of files that do exist.	14:20
	→ willob joined	14:20
	→ jamescarr joined	14:20
	→ subhojit777 joined	14:21
	→ nilg joined	14:21
YoungFrog	W8uiE5: sorry I don't know win 7 ; but you can use it on a non-existent file then, and see if it gets created	14:21
	I mean, it's easier than trying to figure out how stdout and stderr are redirected.	14:21
W8uiE5	YoungFrog: ok, one moment	14:22
	→ RaceCondition joined	14:22
	← jfdi left	14:22
	→ breakingthings joined	14:22
	→ v0n joined	14:23
	→ jkremser joined	14:23
W8uiE5	YoungFrog: hmm. the file hasn't be created	14:23
	→ Footur joined	14:23
Footur	hello there	14:24
gitinfo	Footur: 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.	14:24
	→ thl0 joined	14:24
	→ kogent joined	14:24
Footur	is there a recommended way to handle large files in git?	14:24
	← arvind_k left	14:24
YoungFrog	W8uiE5: As ojacobson suggested, check if the file is executable and has the right name.	14:25
	→ macabre joined	14:25
arand	Footur: There are some third-party extensions like git-media and git-annex, otherwise the recommendation would be "don't" probably.	14:26
	← znode left	14:26
	→ chuckharmston joined	14:26
Footur	arand: thanks	14:26
	← cmmagid left	14:26
	→ jxpx777 joined	14:26
	→ trungqtran joined	14:27
	→ AaronMT joined	14:27
	← xxc_michael left	14:29
	→ eletuchy joined	14:29
W8uiE5	ok, thanks	14:29
	git apply patch returned some whitespace errors. i manually removed the trailing whitespace errors, but still get error:patch failed filename:33 what can i do?	14:29
	→ Pugnax joined	14:30
	← przemoc left	14:30
	→ xxc_michael joined	14:30
	→ btree joined	14:31
	← OOPMan left	14:32
	← btree left	14:32
	→ Milossh joined	14:32
	← xxc_michael left	14:32
	→ znode joined	14:32
	← znode left	14:33
	← rurufufuss left	14:33
	→ mmitchell joined	14:34
	→ fsvehla joined	14:35
	← marthinal left	14:35
	→ jfdi_maverick joined	14:35
	← jamescarr left	14:36
	→ marthinal joined	14:36
	→ hellopat joined	14:37
	← pity left	14:39
	→ theRoUS joined	14:39
	← theRoUS left	14:39
	→ theRoUS joined	14:39
	→ albel727 joined	14:40
	→ nielsbusch joined	14:40
	ferdev\|away → ferdev	14:40
	→ mikeym_ joined	14:41
	← saki007ster left	14:41
	→ gustavnils joined	14:41
	→ betz joined	14:41
	← nielsbusch left	14:41
	→ jaequery joined	14:42
_ikke_	manually editing patches is very tricky	14:42
	→ jamescarr joined	14:42
	← albel727_ left	14:42
	← betz left	14:44
arand	W8uiE5: Have you tried using the --whitespace option to git apply instead?	14:44
	→ Vert joined	14:44
	→ bootstrapper joined	14:44
	→ cristi_ joined	14:45
	cristi_ → Guest86534	14:45
	→ breck joined	14:46
	→ przemoc joined	14:46
W8uiE5	arand: there is no such option. tried --ignore-whitespace	14:46
	→ nielsbusch joined	14:47
	← valgrind left	14:47
	← Pugnax left	14:48
	→ shadok joined	14:48
	← ThePawnBreak left	14:48
arand	man git apply	14:50
gitinfo	the git-apply manpage is available at http://jk.gs/git-apply.html	14:50
arand	W8uiE5: --whitespace=(nowarn\|warn\|fix\|error\|error-all)	14:51
	← nielsbusch left	14:52
	→ lov joined	14:52
	← hatake left	14:53
lov	When I see ssh://foo.bar.com/some/path/Project.git , does Project.git actually exist? or is it something that the git client interprets?	14:53
bremner	lov: that's up to the server, but probably there is a Project.git somewhere on the server	14:53
W8uiE5	arand: ok, thanks. the error is still the same	14:54
	← eletuchy left	14:54
	← Sky[x] left	14:54
lov	bremner: Is there any sort of documention on this? I'm having trouble getting google to understand ".git file"	14:54
	→ Ziggy` joined	14:54
	→ sweb joined	14:54
	→ eletuchy joined	14:54
bremner	lov: perhaps you could tell us what the problem you are trying to solve is?	14:54
	← sweb left	14:55
lov	I'm really just trying to understand better how git works. I've seen foo.git as the end part of repo urls, but the foo.git itself isn't part of the clone. I'm trying to figure out what's actually happening.	14:55
ojacobson	lov: the magic phrase is "git repository" and, yes, generally it exists on disk. See !bare for some details. Answering bremner's question will help, too.	14:55
gitinfo	lov: 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/	14:55
jast	usually it's something that actually exists	14:55
lov	ojacobson: thanks	14:55
jast	git just parses the name to determine a nice name for your local target path	14:55
	→ kevinfagan joined	14:55
ojacobson	When you 'git clone foo.git' you end up with a non-bare clone in a directory named foo/.git with a working tree in foo/	14:56
	so, in effect, foo.git is copied to the clone, just under a different name.	14:56
jast	on the remote server, chances are that you have a physical directory like /some/path/Project.git and it contains similar things to your local .git subdir in a repo	14:56
	→ hatake joined	14:56
jast	(when you have only that metadata, we call that a bare repository)	14:56
	← bootstrapper left	14:56
ojacobson	(the local repository in .git isn't a byte-for-byte copy, but it contains much of the same data)	14:56
	→ P4C0 joined	14:56
lov	ok, I think I'm starting to understand.	14:57
	→ nielsbusch joined	14:57
P4C0	hello, is it possible to rebase n commit in the middle of a log? (I mean, rebase n, n-1 and n-2, but n is not head, head will be n+1)	14:58
lov	so Project.git isn't an actual file on the remote server at all. It's just a reference to a bare repository.	14:58
bremner	well, it's also a directory	14:58
	probably	14:58
	← Guest86534 left	14:58
	← hatake left	14:58
	→ ThePawnBreak joined	14:58
lov	meaning it would be /some/path/Project/.git/ ? or /some/path/Project.git/ ?	14:59
bremner	the latter	14:59
	← eletuchy left	14:59
ojacobson	Probably some/path/project.git	14:59
lov	hmm.	14:59
ojacobson	'git init --bare project.git' will create such a directory	14:59
	(or 'git clone --bare', or 'git clone --mirror')	14:59
	← W8uiE5 left	14:59
	← aliljet left	15:00
lov	right. ok. So is there a reason behind using this kind of nomenclature? Why not just provide ssh://...../some/path/Project ?	15:00
ojacobson	You can do that, too	15:00
	→ aliljet joined	15:00
	→ mklappstuhl joined	15:00
lov	is it just to help humans understand more easily that the url is a git repo?	15:00
jast	that, and it helps on the server	15:00
	← thl0 left	15:00
ojacobson	git-upload-pack (the part that handles the server side of git clone and git fetch) will append .git if necessary	15:00
	→ ehsan joined	15:00
jast	if you look at a server filesystem and see foo.git, you know immediately that it's a bare git repo	15:00
lov	hmm, interesting.	15:00
ojacobson	it'll try without .git first, though, if the passed path doesn't end with .git	15:00
	→ senorflor joined	15:01
jast	this is just a convention, though; technically you can call it whatever you want	15:01
lov	ok.	15:01
	→ hatake joined	15:01
ojacobson	yeah, the convention is that bare repos have names ending in .git and working repos do not, but it is just a convention	15:01
	you can violate it if you think you have a good reason	15:01
	git mostly doesn't care about names, for bare repos	15:01
	→ Bass10 joined	15:01
	→ thl0 joined	15:01
	← nielsbusch left	15:01
	← soee left	15:02
	→ garbagecollectio joined	15:02
	→ NiteRain joined	15:02
	← ShooterMG left	15:02
lov	ok, thanks. I was looking at someone's post where he was pushing from one desktop to another desktop which had a post-update hook to perform tests/etc. I was confused since his remote was Project.git, and I had never really understood that part.	15:02
	← phluks left	15:03
	→ znode joined	15:03
	→ oposomme joined	15:04
	→ aisbaa joined	15:04
	→ rudivs joined	15:05
	→ Pugnax joined	15:05
	← fsvend left	15:06
	→ normanrichards joined	15:06
	← zerba left	15:06
rudivs	following some bad advice, I did "git reset --hard HEAD^" when I think I was actually looking for "git reset --hard". In the process I lost my last commit. Is there a safe way to get it back?	15:07
	→ nielsbusch joined	15:07
ojacobson	rudivs: it should be listed in 'git reflog'	15:07
	you can reset back to it	15:07
	→ yshh joined	15:08
	→ pfefferz joined	15:08
	← daddio left	15:08
	→ AnimeCYC joined	15:09
	← hatake left	15:09
rudivs	ojacobson, I can see it as HEAD@{2} - would the command then be git reset HEAD@{2}?	15:09
	← senorflor left	15:09
ojacobson	Yes, or you can reset to the bare hash	15:10
	→ __BigO__ joined	15:10
	→ axl_ joined	15:10
ojacobson	The HEAD@{number} syntax (explained in man gitrevisions) is notation for "the n'th item of the reflog of HEAD"	15:10
gitinfo	the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	15:10
	→ senorflor joined	15:10
	→ Kadma joined	15:10
	← rmichnik left	15:10
	← jamescarr left	15:11
	← alkino left	15:12
	← nielsbusch left	15:12
rudivs	thanks ojacobson	15:12
	→ johnpcamara joined	15:13
	← znode left	15:13
	← breck left	15:13
	→ beaky joined	15:13
beaky	hello	15:13
	how do I combine two git repos into one	15:14
	→ thatguycraig joined	15:14
	→ SeoFood joined	15:15
	← daniel_- left	15:15
	→ datadevil joined	15:15
_ikke_	beaky: !subtree	15:15
gitinfo	beaky: 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	15:15
YoungFrog	beaky: wait, did you just ask "how do I combine two universes into one" ?	15:15
beaky	ah thanks	15:15
YoungFrog	_ikke_: and did you just answer that straightforwardly ?	15:15
beaky	yep; git lets you become master of the universe	15:15
Skaag	i just git rm -f the wrong directory	15:15
	can i get it back from the master?	15:15
	i didn't commit yet	15:16
datadevil	If I it checkout master && git rebase someotherbranche	15:16
	does that rebase master onto someotherbranche or the other way round?	15:16
beaky	also, what is the best way to drop all uncommitted changes? 'git revert' or 'git reset' or 'git reset --hard' or something else?	15:16
	→ hatake joined	15:16
_ikke_	Skaag: git checkout path/to/dir	15:17
	datadevil: It rebases master on someotherbranch	15:17
YoungFrog	datadevil: master is then rebased onto someotherbranche.	15:17
_ikke_	beaky: the latter	15:17
beaky	ah thanks	15:17
datadevil	ok. Would that explain master saying it is ahead of the origin if I run a status on master?	15:18
Skaag	_ikke_: thanks, still learning the git :)	15:18
	→ nielsbusch joined	15:18
	← jaequery left	15:18
	→ natechan joined	15:18
Skaag	_ikke_: it says doesn't match any files known to git	15:19
	→ TheJH joined	15:19
_ikke_	Skaag: Does git status say the directory is removed?	15:19
	→ OOPMan joined	15:19
YoungFrog	datadevil: if you rebase master onto someother branch, you usually diverge from origin. So you're in a nice situation, where the only commits you changed were actually commits you had made yourself.	15:19
	datadevil: but yes, it can explain.	15:20
	→ isomorph joined	15:20
	← flijten left	15:20
	← johnny_rose left	15:20
	→ kpreid joined	15:21
	→ madewokherd joined	15:21
datadevil	YoungFrog: so how would I resolve that, rebase master on origin? The weirder thing seems to be that this master == origin, if I read it correctly	15:21
arand	datadevil: Well, master would now be diverged from origin/master, since you've rewritten it's history.	15:21
	← ostree left	15:21
	→ tbmiller joined	15:21
datadevil	arand: makes sense	15:21
	← nielsbusch left	15:22
	→ adamben joined	15:22
	← zoonfafer left	15:23
YoungFrog	datadevil: you have your own master branch, which was intially a copy of origin/master (which itself is a copy of the true master branch at origin when you fetched it)	15:23
	→ Sulculac joined	15:23
	← rudivs left	15:23
arand	datadevil: If origin has already been published, you would need to now merge your new master and origin/master and then push that, or as you said rebase master on top of origin/master. I'm not sure you want to do the rebase you did in the first place though if you intend to push stuff...	15:24
datadevil	arand: not quite sure yet why they thought of this workflow	15:25
arand	where did this "someotherbranch" come from, did it start off from the original master somewhere?	15:25
	→ nir joined	15:25
datadevil	arand: think so, but not quite sure tbh	15:26
	→ ErKa joined	15:26
	← Sulculac left	15:26
arand	In that case it would make more sense to rebase someotherbranch on top of master, and then merge that into master (fast-forward) and push that...	15:27
	→ bootstrapper joined	15:27
	→ nielsbusch joined	15:27
	← nir left	15:27
	→ nir joined	15:27
	→ alkino joined	15:27
datadevil	arand: yeah..or even get rid of the notion of remote master as the master I'm talking about has no real remote..think that is a configuration issue	15:28
	remote origin master...my head spins, sorry	15:28
	← jatt left	15:30
	← KevinSjoberg left	15:32
	← nielsbusch left	15:32
datadevil	I do understand it all a bit better now, thanks for that	15:33
	→ fission6 joined	15:34
	← thl0 left	15:35
	← babilen left	15:35
	→ babilen joined	15:35
	← hatake left	15:35
	→ rmichnik joined	15:36
	← babilen left	15:36
	→ flijten joined	15:36
	→ saki007ster joined	15:36
	→ babilen joined	15:37
	→ hatake joined	15:37
	→ nielsbusch joined	15:37
	← babilen left	15:37
	← Chryyz left	15:37
	← datadevil left	15:37
	→ SqREL joined	15:38
	← matthewt left	15:38
	→ MrDHat joined	15:38
	← letharion left	15:39
	→ babilen joined	15:39
	→ thl0 joined	15:39
	→ grn_ joined	15:39
	← nir left	15:39
	→ eletuchy joined	15:39
	← eletuchy left	15:40
	← babilen left	15:40
	→ eletuchy joined	15:40
	→ kjetilho joined	15:41
	← OOPMan left	15:41
	← arand left	15:41
	← bauruine left	15:41
MetaCosm	I am curious about best practices for maintain multiple changesets on top of an open-source project (hosted on github in this case). We have general improvements that might be merged into the upstream proejct, then we have system specific changes that never will be merged. Is there a well-worn pattern for this?	15:41
	← nielsbusch left	15:42
	→ moore_ joined	15:42
MetaCosm	We are just using topic branches at the moment, but I sense there is a better way.	15:42
	← hatake left	15:42
	← kjetilho left	15:42
	→ hatake joined	15:43
	→ OOPMan joined	15:43
	→ babilen joined	15:44
_ikke_	1Well. there is very little better than topic branches, to keep a set of changes seperate	15:44
	← shezbaig_wk left	15:44
	← pfefferz left	15:44
	→ jphillips joined	15:44
	← Pugnax left	15:45
	→ jerikl joined	15:45
	← JPeterson left	15:45
	→ [1]JPeterson joined	15:45
	→ sirdancealot joined	15:46
	→ tomswartz07 joined	15:46
tomswartz07	hey all, i have a question, but i'm uncertain of the terms for what I'm looking to do.	15:46
_ikke_	tomswartz07: Well, try to explain it	15:47
	← agile left	15:47
tomswartz07	i would like to automatically configure my git commit messages to include a 'tested-on:" field, that could be filled out at the time of commit. As in, this commit was tested on this device	15:47
	is there a way to do this?	15:47
	→ nielsbusch joined	15:47
	← hatake left	15:47
_ikke_	tomswartz07: Look at the commit-msg hook in man githooks	15:48
gitinfo	tomswartz07: the githooks manpage is available at http://jk.gs/githooks.html	15:48
_ikke_	though, hooks need to be set up by each developer themselfs	15:48
	→ hatake joined	15:48
	← hatake left	15:49
	← thl0 left	15:49
	→ hatake joined	15:49
tomswartz07	_ikke_: i see- theres a similar thing for gerrit as well. I've never worked with hooks before; what should i know before i begin?	15:49
	→ thl0_ joined	15:49
	→ CodeShark joined	15:50
	→ chrisgeorge joined	15:51
	← mikeym_ left	15:51
_ikke_	tomswartz07: Very little, they're just executables that get run on certain events	15:51
	→ agile joined	15:51
tomswartz07	i see- after digging in that page, it seems that there's a git commit.template option	15:51
_ikke_	Ah. I meant prepare-commit-msg	15:51
	ah ok	15:52
	← nielsbusch left	15:52
	← papegaaij left	15:52
tomswartz07	would the template do what i need?	15:52
_ikke_	tomswartz07: Most likely, yes	15:52
	→ amigojapan_ joined	15:52
	← saki007ster left	15:52
	← amigojapan left	15:53
	→ mikeym_ joined	15:53
	← MrDHat left	15:54
tomswartz07	whats the difference between prepare-commit-msg and a commit template?	15:54
_ikke_	tomswartz07: The one is static, the other dynamic	15:54
tomswartz07	_ikke_: template is static, and prepare is dynamic?	15:55
_ikke_	aye	15:55
	← CodeShark left	15:55
	→ kristofers joined	15:55
_ikke_	The template is just a file that's the base for the commit message, the hook can then do the things it want to the message	15:56
	← hatake left	15:56
tomswartz07	excellent. I think the template is exactly what i need! thanks _ikke_	15:56
	oposomme → oposomme\|away	15:57
	→ nielsbusch joined	15:57
	← wasanzy left	15:58
tomswartz07	one last question: when using git templates in conjunction with gerrit configs, will the gerrit.createchangeid hook still operate as expected?	15:59
	→ L0cky joined	15:59
	→ Remram joined	15:59
	← thl0_ left	16:02
	→ thl0 joined	16:02
_ikke_	tomswartz07: I don't know	16:02
tomswartz07	ok- i guess i'll have to play with it and see.	16:04
	thanks again.	16:04
	← tomswartz07 left	16:04
	← JDuke128 left	16:05
	→ JDuke128 joined	16:05
	← yshh left	16:05
	→ infynyxx joined	16:05
	→ yshh joined	16:06
	← GeertV left	16:06
	← aero1 left	16:07
	→ GeertV joined	16:07
	→ johnpcamara_ joined	16:08
	→ maletor joined	16:08
	→ znode joined	16:09
	→ alunduil joined	16:10
	← yshh left	16:11
	→ aartist joined	16:11
	← GeertV left	16:11
	← nielsbusch left	16:12
	← johnpcamara left	16:12
	← tPl0ch left	16:12
	← bioman left	16:13
	← rchavik left	16:13
	← jonshea left	16:14
	← thl0 left	16:14
	→ thl0 joined	16:14
	→ Kicer86 joined	16:14
	→ Noldorin joined	16:14
	← netiul left	16:15
	← znode left	16:15
	→ itatitat joined	16:17
	← cbreak-work left	16:17
	→ nielsbusch joined	16:17
	→ arand joined	16:20
	→ maestrojed joined	16:20
	← alunduil left	16:20
	→ alunduil joined	16:21
	← mishok13 left	16:21
	← jerikl left	16:21
	← nielsbusch left	16:22
	← LeMike left	16:22
	→ marhaban joined	16:23
	← psoo left	16:24
	→ boze joined	16:24
	oposomme\|away → oposomme	16:24
	← oposomme left	16:25
	← flijten left	16:25
	→ matthewt joined	16:26
	← jfdi_maverick left	16:26
	← ThomasLocke left	16:27
	← thl0 left	16:28
	→ nielsbusch joined	16:28
	→ thl0 joined	16:29
	← thl0 left	16:29
	→ thl0_ joined	16:29
	← bootstrapper left	16:30
	← Kicer86 left	16:31
	→ hays_ joined	16:32
	→ LeMike joined	16:32
	→ Sky[x] joined	16:33
	← RaceCondition left	16:35
	→ Jackneill joined	16:35
	← Jackneill left	16:35
	→ Jackneill joined	16:35
	← rath left	16:35
	← tim left	16:37
	→ Cromulent joined	16:38
	→ h0bbit joined	16:39
	→ fsvend joined	16:40
	← jxpx777 left	16:40
	← willob left	16:41
	← OOPMan left	16:41
	← johnpcamara_ left	16:41
	→ johnpcamara joined	16:42
	← thl0_ left	16:42
	→ Gentlecat_ joined	16:42
	→ thl0 joined	16:42
	← lifelike left	16:43
	← thl0 left	16:43
	← johnpcamara left	16:43
	→ thl0 joined	16:43
	→ lifelike joined	16:43
	→ johnpcamara joined	16:43
	← Gentlecat left	16:43
	← madewokherd left	16:43
	← johnpcamara left	16:43
	→ OOPMan joined	16:44
	→ jpcamara joined	16:44
	→ madewokherd joined	16:44
	→ Kicer86 joined	16:45
	← lifelike left	16:46
	→ lifelike joined	16:46
	→ bootstrapper joined	16:46
	→ jensenc joined	16:47
	← lifelike left	16:47
	← LeMike left	16:48
	→ lifelike joined	16:48
	→ jonshea joined	16:48
	→ oriba joined	16:49
	→ RaceCondition joined	16:49
	← JohnBat26 left	16:50
	→ bauruine joined	16:50
	→ local joined	16:50
	→ bioman joined	16:50
bioman	Hep	16:50
	→ agileadam joined	16:50
bioman	trying git revert : "error: 'revert' is not possible because you have unmerged files". What does this mean and how can I "fix" this please ?	16:51
agileadam	I have the pleasure of trying to determine which files a colleague has modified. The code has a git repo with about 15 commits. The commits contain a LOT of garbage… he didn't know what he was doing. Is there a way I can export a diff of the changes made to all files EXCLUDING certain directories?	16:52
charon	bioman: !eekaconflict	16:52
gitinfo	bioman: Merge conflicts are a natural part of collaboration. When facing one, don't panic. Read "How to resolve conflicts" in man git-merge. Carefully go through the conflicts. Picking one side verbatim is not always the right choice!	16:52
jast	bioman: your pull/merge still isn't complete. you'd know this if you'd read the message it gave to you. ;)	16:52
	I gotta go... good luck	16:53
TonyM	I've just started getting this message when using git on github: /data/github/shell/bin/gh-environment: line 84: /etc/ghrc-secrets: Permission denied	16:54
	← thl0 left	16:54
	→ fobban joined	16:54
TonyM	anyone know whether it's my fault or just something awry on github?	16:54
	→ thl0 joined	16:55
	→ sfalcon joined	16:55
	← thl0 left	16:55
	→ thl0 joined	16:55
	→ _iron joined	16:56
milki	lol probably not	16:56
	to 1	16:57
	check statuspage?	16:57
	→ jaequery joined	16:57
	→ rideh joined	16:57
	→ texasmynsted joined	16:58
	→ tonymet joined	17:00
	→ sluther joined	17:00
	→ lavaman joined	17:00
texasmynsted	So when I do a push (git 1.8.1.3) over https I get a gui prompt for credentials. How do I prevent a gui prompt? I checked here https://www.git-scm.com/docs/gitcredentials and found nothing about gui/non-gui. Any ideas?	17:01
	← revyd left	17:01
	← moore_ left	17:02
	→ robbyoconnor joined	17:02
	→ enr joined	17:02
	→ rtjure joined	17:03
	← chisler left	17:03
bioman	Ok, read... Just don't understand anything. Infact, I've never understood git, I use basics	17:03
	→ ejd joined	17:03
bioman	Please help me... Q_Q	17:03
	→ wasanzy joined	17:03
lov	bioman: you're still in the middle of a merge.	17:04
agileadam	Nevermind. I've worked something out to determine modified files excluding certain directories	17:04
lov	are you trying to back out of the merge? "revert" is not the same as what you might be used to in, say, svn.	17:04
	← thl0 left	17:05
bioman	lov: I have a commit that causes regression. On #nouveau, I've been said to use 'git revert <commi_that_causes_regression>	17:05
	→ thl0 joined	17:06
	← thl0 left	17:06
	← AdrienBrault left	17:06
	← senorflor left	17:06
	→ thl0 joined	17:06
lov	right, that's fair. But you're in the middle of merging something, because there's a merge conflict.	17:06
	You need to resolve that before you can do anything else.	17:06
	So, your problem here is with a merge conflict, not reverting.	17:06
	→ boombatower joined	17:07
	← howlymowly left	17:07
	← AartBluestoke left	17:07
	bioman wonders why God has made him so dumb as regards git ;'(	17:07
	← OOPMan left	17:07
milki	o.O	17:07
	← skul left	17:08
	← rmichnik left	17:08
_ikke_	bioman: Have you read anything from !book?	17:08
gitinfo	bioman: 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	17:08
	← jargon left	17:09
bioman	lov: How can I "resolve" things that do not understand ? :(	17:09
	→ goshawk joined	17:10
	→ OOPMan joined	17:10
	→ rmichnik joined	17:10
lov	http://stackoverflow.com/questions/161813/how-do-i-fix-merge-conflicts-in-git	17:10
	also, !book	17:10
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	17:10
	→ lresende joined	17:11
	→ znode joined	17:11
	→ psoo joined	17:12
	→ tavist0 joined	17:14
	→ looper joined	17:14
	← alorence left	17:14
Skaag	_ikke_: git status shows them as deleted	17:15
	_ikke_: how do I undo this removal and bring it back from git?	17:16
	← thl0 left	17:16
	→ yshh joined	17:16
	→ thl0 joined	17:17
_ikke_	Skaag: What command did you execute?	17:17
	← znode left	17:17
	← thl0 left	17:17
	→ willob joined	17:17
Skaag	ok i think i managed with git reset HEAD media	17:17
	then git checkout -- media	17:17
	→ thl0 joined	17:17
	← natechan left	17:17
	gitinfo set mode: +v	17:17
jensenc	hi can git-p4 understand p4 branches ?	17:18
texasmynsted	anybody know how to prevent a gui prompt on push?	17:18
spyroboy	is it for an ssh password?	17:18
	← fsvehla left	17:18
	← lionel_ left	17:19
spyroboy	I think I've seen that of some distros. they spawn a gui for some password inputs on the command line.	17:19
_ikke_	Skaag: Ooh, did you git add it?	17:19
	→ scorphus joined	17:19
texasmynsted	hmm. yes, I guess the remote.origin.url is ssh at a third party	17:21
	← yshh left	17:21
bioman	lov: I've launched git mergetool. Now I'm in the twilight zone... I'm not a developer, I know nothing at all about this. I'm just a linux user that try to maintain his distro... How can I quit this tool please ?	17:22
	→ natechan joined	17:23
	← rob_w left	17:23
	← itatitat left	17:23
_ikke_	you're probably in vimdiff?	17:24
	→ Melrian joined	17:24
	→ swiety joined	17:24
grawity	try Esc Esc :qa! Enter	17:24
	← dhruvasagar left	17:24
	← blackmat` left	17:25
	→ tewecske joined	17:25
	→ blackmat` joined	17:25
	→ kevwil joined	17:25
bioman	grawity: thank you.	17:26
	← jaequery left	17:26
	← fRAiLtY- left	17:27
texasmynsted	is there no way to prevent the gui prompt?	17:27
bioman	If I understand correctly, a basic linux user will never manage to merge conflicts in git isn't it ?	17:27
	← thl0 left	17:27
grawity	bioman: a basic linux user can merge conflicts using their favourite text editor	17:27
	the mergetool is just a convenience – and there are many better ones than vimdiff	17:28
	→ thl0 joined	17:28
grawity	e.g. meld, kdiff	17:28
	← thl0 left	17:28
	→ thl0 joined	17:28
	→ Shusshu joined	17:29
	→ Chillance joined	17:29
	← breakingthings left	17:30
	→ edude03 joined	17:30
	← Shusshu left	17:31
	→ jaequery joined	17:31
	→ tureba joined	17:31
	← Cromulent left	17:32
	← francisfish left	17:32
bioman	grawity: I see lines of code, how do you want me to solve a conflict ???	17:32
	← spuz left	17:32
	← adamben left	17:33
	→ jarr0dsz joined	17:33
	→ Chryyz joined	17:35
	← hyperair left	17:35
grawity	ask #nouveau guys, they told you to use `git revert` and it's their code in the first place	17:35
	→ oriba_ joined	17:35
	← kogent left	17:35
texasmynsted	hmm. looks like it could be an ssh problem and not specifically a git problem	17:35
	← oriba left	17:35
grawity	I can't imagine that many situations in which a basic linux user will encounter merge conflicts in code they don't understand a thing about	17:36
Skaag	_ikke_: initially yes	17:36
	→ dhruvasagar joined	17:36
Skaag	_ikke_: it was added, then I realized the mistake and added it to .gitignore	17:36
	_ikke_: I didn't know git rm -r would also delete the physical files :)	17:36
grawity	either they have to merge their own changes with upstream's changes (which means they know enough about the code to have changed & committed it...)	17:36
	← subhojit777 left	17:36
	← amigojapan_ left	17:36
grawity	or they run a command that was given by the developers on IRC (which means the same developers should be helping with the failed command, too)	17:37
	← sq-one left	17:37
	→ sq-one joined	17:37
bioman	grawity: I've asked them, there I'm know as "bothering bioman". They won't help me anymore	17:38
esc	does git already have a stage command?	17:38
	never versions	17:38
	← thl0 left	17:38
	← fsvend left	17:38
esc	just correcting an exam, a student reckons he can use 'git stage' to stage a file	17:38
	→ thl0 joined	17:38
grawity	esc: No, afaik it has always been called `git add`.	17:38
	→ protux joined	17:39
grawity	oh, wait a moment	17:39
	← thl0 left	17:39
esc	maybe he had a newer version on his machine, so i better double check	17:39
grawity	apparently `git stage` exists and is a synonym for that	17:39
	→ thl0 joined	17:39
esc	!man stage	17:39
	man git stage	17:39
gitinfo	the git-stage manpage is available at http://jk.gs/git-stage.html	17:39
	← protux left	17:40
esc	wow	17:40
YoungFrog	same here, git stage works. git v1.8.1.2	17:40
bremner	get off my lawn you kids	17:40
grawity	`git stage` was added in v1.6.1, commit 11920d28da1ac1	17:40
	← oriba_ left	17:40
	→ lolwhut joined	17:41
bremner	really?	17:41
	I don't have it in 1.7.10.4	17:41
YoungFrog	I have it in 1.7.2.5 on a Debian	17:41
	→ hatake joined	17:41
	← ozette left	17:41
YoungFrog	(Debian squeeze)	17:42
bremner	oh, I have it, just not zsh completion	17:42
	→ senorflor joined	17:42
	→ Corren joined	17:42
	→ protux joined	17:42
	← protux left	17:42
YoungFrog	What, zsh does not scan source code to make its completion list ? That is disappointing.	17:43
	→ protux joined	17:44
	← suy left	17:44
	← senorflor left	17:45
	→ pppaul joined	17:45
	→ shezbaig_wk joined	17:46
	→ lavaman1 joined	17:46
	← lavaman left	17:46
	← mmitchell left	17:46
texasmynsted	fwiw, the problem was an ssh related issue.	17:49
	← jceb left	17:49
	← texasmynsted left	17:49
	← thl0 left	17:51
	→ thl0 joined	17:52
	← thl0 left	17:52
	→ thl0_ joined	17:52
	→ opus joined	17:53
	→ techhelp joined	17:54
	→ lavaman joined	17:54
	→ OliverJAsh joined	17:55
	→ tsdh joined	17:56
	→ betz joined	17:57
	→ jjore joined	17:57
	← dhruvasagar left	17:57
	→ codertux joined	17:57
	← lavaman1 left	17:58
	← maxmanders left	17:58
	→ erichynds joined	17:59
	← bbloom left	18:00
	← jcromartie left	18:00
	← Desproges left	18:00
	→ kkwak joined	18:00
	← kkwak left	18:01
	→ thiago joined	18:01
	← thl0_ left	18:02
	→ daniel_- joined	18:02
	← daniel_- left	18:02
	→ daniel_- joined	18:02
P4C0	Hello, when doing git diff between commits, is there an easy way to apply that?	18:02
	→ thl0 joined	18:03
	← thl0 left	18:03
	← RaceCondition left	18:03
	→ thl0 joined	18:03
P4C0	(my making a patch mabye, or a command to overwrite current file with the one on commit x?	18:03
grawity	what do you actually want to achieve?	18:03
	← hatake left	18:04
grawity	the first line sounds like you're asking for `git apply` or `git am` or `git cherry-pick`	18:04
	but the second line sounds like you want `git checkout`	18:04
P4C0	grawity: yes I just checked the man page for git checkout, that's what I need, with -p option	18:05
	→ senorflor joined	18:05
	← JDuke128 left	18:05
	→ fsvehla joined	18:05
esc	YoungFrog: have fun implementing this :)	18:06
YoungFrog	:)	18:06
P4C0	grawity: oh, but it failed, since my file is not merged :( (ignoring unmerged:)	18:06
	← Melrian left	18:06
	← Kadma left	18:06
	→ fly9 joined	18:07
	← opus left	18:07
	→ alp82 joined	18:07
	→ pr0ton joined	18:07
	→ NaN joined	18:08
	→ asker joined	18:08
NaN	what is a best practice after doing a merge? git diff <current branch> <new branch>?	18:08
	← fobban left	18:08
	← wasanzy left	18:09
	← tureba left	18:09
	→ joshsmith joined	18:10
	← pr0ton left	18:10
	→ dreiss joined	18:10
P4C0	can format-patch be used only in one file? it doesn't seem to be the case :(	18:10
	→ pr0ton joined	18:10
grawity	so are you going to explain what you're trying to achieve by running random patch-related commands?	18:11
	← infynyxx left	18:11
	→ undecim joined	18:11
	← diegoviola left	18:11
	→ khmarbaise_ joined	18:12
	→ znode joined	18:12
	→ opus joined	18:12
	→ mihamina joined	18:13
	← thl0 left	18:13
	→ thl0 joined	18:13
	← thl0 left	18:14
	← khmarbaise left	18:14
	khmarbaise_ → khmarbaise	18:14
	→ thl0 joined	18:14
	← relipse left	18:16
	→ ozette joined	18:16
	→ fobban joined	18:16
	→ kumbayo joined	18:17
	← NaN left	18:17
	Corren → kenperkins	18:17
	← znode left	18:17
	← mikeym_ left	18:19
	← marthinal left	18:19
P4C0	Is it possible to squash 2 commits together, but using the commit log of the first one?	18:19
	→ glaksmono joined	18:19
	← sq-one left	18:20
	← techhelp left	18:20
	→ mikeym_ joined	18:20
	→ gyre008 joined	18:20
	→ techhelp joined	18:20
	→ jerikl joined	18:22
	→ pfefferz joined	18:22
	← etcetera left	18:22
	→ kit joined	18:22
	→ hyperair joined	18:22
	← kevinfagan left	18:22
	← gyre007 left	18:24
	→ sh42 joined	18:24
	← thl0 left	18:24
	← sh42 left	18:24
	→ sh42 joined	18:24
	→ thl0 joined	18:24
	→ marthinal joined	18:24
	← thl0 left	18:24
	→ thl0 joined	18:25
	← gyre008 left	18:25
	→ etcetera joined	18:25
	→ howlymowly joined	18:25
	← robbyoconnor left	18:26
	→ wasanzy joined	18:26
	← ferdev left	18:26
tango_	P4C0: use the fixup keyword instead of squash during the interactive rebase	18:26
	→ HardDisk_WP joined	18:27
HardDisk_WP	Hi	18:27
	→ robbyoconnor joined	18:27
HardDisk_WP	In a repository, I have created a file with >100 commits	18:27
P4C0	tango_: what's the difference? I'm not 100% sure	18:27
HardDisk_WP	Now, the time has come to move this and another file into its own repository	18:27
	How do i do this without losing history?	18:27
	I'm not concerned about the other files affected by the commits	18:28
tango_	P4C0: squash asks you for the commit message. fixup uses the first commit message	18:28
P4C0	tango_: thanks :)	18:28
	← betz left	18:28
	← garbagecollectio left	18:28
tango_	HardDisk_WP: you probably want something like filter-branch or something with the subtrees, but I can't give you specific incantations	18:28
HardDisk_WP	k	18:29
	← h0bbit left	18:30
	← senorflor left	18:31
	← asker left	18:32
	← jaequery left	18:32
esc	so, the exam question was: Describe a common work-flow (series of a few commands) when using git	18:33
	→ jaequery joined	18:33
	← mike_pizza left	18:33
esc	someone wrote: I use the github APP, it has a nice GUI	18:33
	← kumbayo left	18:34
esc	man git synchronize	18:34
gitinfo	the git manpage is available at http://jk.gs/git.html	18:34
	→ kumbayo joined	18:34
	→ tallest_red joined	18:34
P4C0	what's the difference between edit and reword? is amending different than editing the commit message?	18:34
	→ EmpiresBane joined	18:34
	← Sigma left	18:35
	→ breakingthings joined	18:35
	→ Cromulent joined	18:35
	← thl0 left	18:36
	← zenx left	18:36
	→ punknology joined	18:36
	→ PaulCapestany joined	18:36
	→ thl0 joined	18:36
	← thl0 left	18:36
	→ thl0 joined	18:37
	→ hatake joined	18:37
	→ chisler joined	18:37
	← mikeym_ left	18:37
	→ whyrusleeping joined	18:37
	→ mikeym_ joined	18:38
whyrusleeping	is there a way to make a repo automatically checkout a new branch that i push to it? (for publishing a website)	18:38
	← kumbayo left	18:38
	← tjadc left	18:39
	→ kumbayo joined	18:39
	← alexbligh left	18:39
	→ system64 joined	18:40
Rhonda	whyrusleeping: ikiwiki does that	18:40
	→ jceb joined	18:41
	← robbyoconnor left	18:41
	← alkino left	18:42
	→ bzyx joined	18:42
	→ kumbayo_ joined	18:43
	← kumbayo left	18:44
	← Cromulent left	18:45
	← chimay left	18:46
	→ senorflor joined	18:46
	← jaequery left	18:46
	→ maxmanders joined	18:46
	← marhaban left	18:46
whyrusleeping	isnt it just using hooks?	18:47
	← maestrojed left	18:48
	← protux left	18:48
	← achadwick left	18:48
	→ snql joined	18:48
	→ Sicp joined	18:48
bremner	whyrusleeping: see also !deploy	18:49
gitinfo	whyrusleeping: Git is not a deployment tool. You can build one around it for simple environments. http://sitaramc.github.com/the-list-and-irc/deploy.html	18:49
	← snql left	18:49
Rhonda	whyrusleeping: In theory, yes.	18:49
	→ snql joined	18:49
	→ maestrojed joined	18:49
	← hatake left	18:49
	→ AdrienBrault joined	18:49
	← _goTAN left	18:50
	→ rakl joined	18:51
	← baba left	18:51
	← zoresvit left	18:51
	← snql left	18:51
	→ snql joined	18:52
	← chuckharmston left	18:52
	← cebor left	18:52
	← shezbaig_wk left	18:52
	← pr0ton left	18:53
	← ryez left	18:53
whyrusleeping	alright, thanks!	18:53
	→ cebor joined	18:54
	→ pr0ton joined	18:54
	← thl0 left	18:55
	→ thl0 joined	18:55
	← thl0 left	18:55
	→ thl0 joined	18:56
	← pfefferz left	18:56
	→ pfefferz joined	18:56
	→ ferdev joined	18:57
	→ LeMike joined	18:57
	← anderse left	18:57
	← jonshea left	18:58
	→ jonshea joined	18:58
	← Vampire0 left	18:58
	→ jaequery joined	18:58
	← senorflor left	19:00
	→ diegoviola joined	19:00
	← tonymet left	19:00
	→ tureba joined	19:01
	sh42 → zeroSignal	19:01
	→ tonymet joined	19:01
	→ ARNHOE joined	19:01
	→ Vampire0 joined	19:03
	→ Parhs joined	19:03
	← babilen left	19:06
	→ alorence joined	19:06
	← jensenc left	19:07
	← opus left	19:07
	← diegoviola left	19:07
	← etcetera left	19:08
	→ infynyxx joined	19:08
	→ garbagecollectio joined	19:09
	← chisler left	19:09
	→ rjc joined	19:10
	→ etcetera joined	19:11
	← EmpiresBane left	19:12
	→ pantsman joined	19:12
	← pantsman left	19:12
	→ pantsman joined	19:12
	← kumbayo_ left	19:13
	→ leeN joined	19:13
	→ znode joined	19:13
	→ hobodave joined	19:13
	← thl0 left	19:13
	→ thl0 joined	19:14
	← maestrojed left	19:14
	← thl0 left	19:14
	→ thl0 joined	19:14
	← blackmat` left	19:14
	→ kevinfagan joined	19:16
	← kevinfagan left	19:16
	→ kevinfagan joined	19:16
	→ maestrojed joined	19:16
garbagecollectio	hey	19:16
	→ lacrymology joined	19:17
garbagecollectio	git push says everything is up to date	19:17
	after i did git add	19:17
	git commit -m	19:17
	and git push	19:17
	within a branch	19:17
	??	19:18
	→ kogent joined	19:18
	← znode left	19:18
	→ rlubke joined	19:18
bremner	garbagecollectio: probably that branch is not present in the remote repo.	19:19
	← cebor left	19:19
	← jaequery left	19:19
garbagecollectio	how do i make it present	19:19
	i did git checkout -b 'name'	19:19
bremner	git push remotename branchname	19:19
	→ jaequery joined	19:20
garbagecollectio	git push name-of-remote-repo-that-isnt-present local-branch	19:20
	??	19:20
bremner	ummn. What is the name of your remote, origin?	19:21
garbagecollectio	yeah	19:21
	well how do i find out to be sure	19:21
bremner	git remote -v	19:21
garbagecollectio	says origin and then a url	19:21
bremner	ok, so git push origin branchname	19:22
	→ Saby joined	19:22
	← Saby left	19:22
	→ Saby joined	19:22
garbagecollectio	so anytime i cut a branch locally i need to do a git push origin branchname?	19:22
	→ nelsn joined	19:24
	← thl0 left	19:24
	→ thl0 joined	19:25
	← thl0 left	19:25
	← mklappstuhl left	19:25
	→ thl0_ joined	19:25
	← jpcamara left	19:26
canton7	garbagecollectio, that's a good thing, trust me	19:26
garbagecollectio	i didn't have to do that at my old shop	19:26
	→ Ziggy_ joined	19:26
canton7	if you want, you can 'git config --global alias.pushit "push -u origin HEAD"'	19:26
	→ shezbaig_wk joined	19:26
canton7	then 'git pushit' will push the currently-checked-out branch to origin	19:27
	← bioman left	19:27
	→ Cassiel joined	19:27
	← __BigO__ left	19:27
	← Ziggy` left	19:27
canton7	also, if you can, grab a git 1.8, which has push.default set to 'simple' (or if you've got a recent git, set it to 'simple' yourself)	19:28
	→ __BigO__ joined	19:28
garbagecollectio	2h3n eo3w i5 eo 5hq5?	19:28
	when does it do that?	19:28
canton7	what?	19:28
	← __BigO__ left	19:28
garbagecollectio	like when does it push	19:28
	at the time of branch creation?	19:28
	→ Ziggy` joined	19:29
canton7	my alias? when you run 'git pushit'	19:29
	like I said	19:29
garbagecollectio	yeah but why at my old shop did it not work	19:29
	← sdeboey left	19:29
garbagecollectio	did it push	19:29
	Tommy[D]_ → Tommy[D]	19:29
garbagecollectio	automatically	19:29
canton7	I don't know what your old shop is, what you were trying to do, and what didn't work	19:29
garbagecollectio	so i can use git push from now on after running git pushit once	19:30
	← Ziggy_ left	19:31
	← kevinfagan left	19:31
	→ chuckharmston joined	19:32
	→ antonishen joined	19:33
	→ kogent_ joined	19:33
cbreak	garbagecollectio: new branches need to be pushed explicitly.	19:33
canton7	yeah	19:34
	← Vampire0 left	19:34
canton7	once a branch has been pushed once, it's pushed automatically from then on	19:35
	read push.default in man git-config	19:35
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	19:35
	→ apok joined	19:35
	← kogent left	19:37
	kogent_ → kogent	19:37
	← thl0_ left	19:37
	→ thl0 joined	19:38
	→ mklappstuhl joined	19:38
	← thl0 left	19:38
	← enr left	19:38
	→ opus joined	19:38
	← leeN left	19:40
garbagecollectio	right so i can do git push after pushing it once	19:40
	← PaulCapestany left	19:40
garbagecollectio	after doing the pushit command	19:41
	→ leeN joined	19:42
	← IntricateRuin left	19:42
	→ alexbligh joined	19:42
	→ ncd\|leeN joined	19:42
	→ thegattaca joined	19:43
	← R1KRDO left	19:43
	← sebuba left	19:44
	→ satiricon joined	19:44
	→ dvide joined	19:46
	→ bnason joined	19:46
	← leeN left	19:47
	← przemoc left	19:47
	← ferdev left	19:48
	→ jpcamara joined	19:49
	← axl_ left	19:49
	→ CodeShark joined	19:49
bnason	Is there a way to get an existing dir setup to use an existing repo without effecting any of the files? IE our vendor setup a new tier and I need to get some of the code under git, so I need to get it to the state where it thinks its checkout out the master branch without modifying any of the existing files. Should I just copy the .git folder from an existing tier?	19:49
	← jpcamara left	19:49
	→ jpcamara joined	19:50
bnason	hope that makes sense lol	19:50
cbreak	you can git init	19:50
	then git remote add and git fetch	19:51
	... you can just copy the .git dir... but that might not work...	19:51
	you'll notice if it does not work.	19:51
	→ axl_ joined	19:51
bnason	then how do I get it to think it has already checked out master, or conversely, how do I check out master actually without effecting any files heh	19:52
	→ mishok13 joined	19:52
	← system64 left	19:52
cbreak	hmm...	19:52
	just move the files away	19:52
	and move them back after the checkout	19:53
bnason	yea i was just thinking about doing all this in another dir and moving over the .git folder then	19:53
	← the_cybersphinx left	19:53
	→ the_cybersphinx joined	19:55
	→ Cromulent joined	19:55
	← kevwil left	19:55
	← Noldorin left	19:55
	→ przemoc joined	19:56
	← przemoc left	19:56
	→ przemoc joined	19:56
	→ ostree joined	19:56
	→ bbloom joined	19:56
	→ kevwil joined	19:56
bnason	bah now to figure out the best way to get the repo over there... stupid firewalls and multiple nets heh	19:57
cbreak	bnason: zip it up	19:58
	scp it over	19:58
	→ SirCmpwn joined	19:58
	cbreak got a hard disk fedexed from the states a few days ago	19:58
	→ Vampire0 joined	19:58
bnason	well I need to be able to fetch as well	19:59
	ill have to get them to open up some ports or nat it	19:59
cbreak	bnason: ssh reverse proxy	20:00
	an easy way to get a backdoor into a network that contains a server you control :)	20:00
	bnason: or take a look at git bundles	20:00
bnason	thanks	20:00
	← tonymet left	20:02
	← kermit left	20:03
	← opus left	20:03
	← bbloom left	20:03
	→ kermit joined	20:03
	← Gentlecat_ left	20:04
	→ mmitchell joined	20:05
	← jarr0dsz left	20:05
	← normanrichards left	20:06
	→ sStewie joined	20:06
	← tsdh left	20:08
	→ nicoulaj joined	20:08
	← jkremser left	20:08
	← pfefferz left	20:08
	← pr0ton left	20:10
	← tavist0 left	20:11
	← tallest_red left	20:11
	→ chrisgeorge_ joined	20:11
	← Cromulent left	20:11
	→ Sonderblade joined	20:12
	← Sonderblade left	20:12
	→ Sonderblade joined	20:12
	→ dankest joined	20:13
	→ skoop joined	20:13
	→ i42n joined	20:13
	→ tavist0 joined	20:13
	→ fsvend joined	20:13
	← chrisgeorge left	20:14
	→ WorkeD joined	20:14
	→ Bankzilla joined	20:15
	← Bankzilla left	20:15
	→ Bankzilla joined	20:15
	← infynyxx left	20:15
	→ juril joined	20:16
	← CreativeEmbassy left	20:17
	→ PaulCapestany joined	20:17
	→ jarr0dsz joined	20:18
	→ infynyxx joined	20:19
	→ cilly joined	20:20
	→ kevinfagan joined	20:20
	→ relipse joined	20:21
	← beaky left	20:21
	← rmichnik left	20:23
	← ErKa left	20:24
	→ jamescarr joined	20:24
	← fsvend left	20:26
	→ gianmarco joined	20:27
	← phiwo left	20:27
	→ zackiv31 joined	20:27
	→ sdeboey joined	20:27
	→ phiwo joined	20:27
	← joshsmith left	20:27
	← mtk left	20:27
	→ heytrav joined	20:28
	→ n3m joined	20:28
	chrisgeorge_ → chrisgeorge	20:29
	← gianmarco left	20:29
	→ ferdev joined	20:30
	→ Kartagis joined	20:30
	→ gianmarco joined	20:31
	→ dijonyummy joined	20:32
	← dijonyummy left	20:32
	← gianmarco left	20:33
	← relipse left	20:34
	← TheJH left	20:34
	← glaksmono left	20:34
	→ Sigma joined	20:35
	← thegattaca left	20:35
	→ dijonyummy joined	20:35
	← ARNHOE left	20:36
	→ Squarism2 joined	20:37
	→ john_anderson joined	20:37
	→ tibs01 joined	20:37
	← john_anderson left	20:37
	→ relipse joined	20:38
	← kogent left	20:38
	← WorkeD left	20:38
	← chuckharmston left	20:39
	← n3m left	20:39
	← shezbaig_wk left	20:42
	→ fsvend joined	20:43
	→ archonix joined	20:44
	← i42n left	20:44
	→ i42n joined	20:44
	← i42n left	20:45
	→ ghoti joined	20:45
	→ sq-one joined	20:46
	→ MrKMG joined	20:46
	← erichynds left	20:48
	→ carandraug joined	20:48
	← thatguycraig left	20:49
	← _iron left	20:49
	→ glaksmono joined	20:50
	← nelsn left	20:50
	→ snkcld joined	20:50
	← Squarism2 left	20:50
snkcld	why do i have to constantly re add all my files when i do a commit?	20:50
	← scampbell left	20:51
carandraug	hi! I have been playing with filter branch and trying to also remove merges that are actually empty commits. --prune-empty does not do it (as the documentation says), but I couldn't find a link for a clear explanation on how to do it. Could someone help me out, please?	20:51
lov	snkcld: because you may not want to commit all of the files that you changed.	20:51
	Actually, I just performed the exception that proved the rule; I added via --patch to only add part of one file to a commit	20:52
	→ bbloom joined	20:52
	← relipse left	20:52
snkcld	lov, that makes perfect sense. thanks. so its part of a typical git workflow that you "git add" prior to all "git commits"?	20:52
lov	in other news, git add --patch makes my heart sing	20:52
	snkcld: effectively, though there's a -a (I think?) arg for git commit which will automatically add everything	20:52
	→ diegoviola joined	20:52
lov	the idea that you tell git what you actually want to commit (including parts of files instead of the whole file) is a feature, not a bug.	20:53
	→ pr0ton joined	20:53
snkcld	lov, agreed	20:53
	→ scampbell joined	20:53
snkcld	so 'git add' is a command which is directly related to commits	20:53
cbreak	snkcld: you do not have to add files	20:53
	you have to add changes	20:53
snkcld	it has nothing to do with branches	20:53
	?	20:53
lov	it has nothing to do with branches.	20:54
cbreak	it is completely unrelated to branching.	20:54
	it stages things for commit	20:54
lov	although files are what you're changing 99% of the time.	20:54
	← bbloom left	20:55
	→ relipse joined	20:55
	← archonix left	20:55
	→ jwmann joined	20:55
	← maletor left	20:56
	→ gamja9e joined	20:58
	← nicoulaj left	20:58
Kartagis	hi	20:59
gitinfo	Kartagis: 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.	20:59
	← sfalcon left	21:00
	→ Noldorin joined	21:00
Kartagis	I've committed, and pushed; but they don't show	21:00
	← juril left	21:00
	← AdrienBrault left	21:00
Kartagis	how can I trace?	21:00
	→ martyl joined	21:00
cbreak	show?	21:01
	→ beastd joined	21:01
	← dreiss left	21:01
lov	You'll have to be more specific. "they don't show" where?	21:02
	← relipse left	21:02
	← martyl left	21:02
Kartagis	lov: http://drupal.org/project/tcmb	21:02
	it says my last commit was 1 day ago	21:02
	I did one 10 minutes ago, and it doesn't show	21:03
cbreak	git log	21:03
	note the sha of the commit	21:03
	git ls-remote remotename	21:03
	is the sha in the branch you want it to be?	21:03
	→ relipse joined	21:03
	← antonishen left	21:03
	→ protux joined	21:04
	← protux left	21:05
	← huoxito left	21:05
	← HardDisk_WP left	21:05
	← jpcamara left	21:06
	→ chuckharmston joined	21:06
	→ opus joined	21:06
	← Jackneill left	21:06
	← lacrymology left	21:07
	← jwmann left	21:07
Kartagis	yes, but my commit from 10 minutes ago doesn't show	21:08
	→ peteski joined	21:09
cbreak	then the thing you use to show it is broken	21:10
	→ mtk joined	21:10
cbreak	use git to show it	21:10
	then it'll work	21:10
	→ hacktose joined	21:10
	→ Nneon joined	21:11
Kartagis	I used git to show it	21:11
	← Cassiel left	21:12
	→ protux joined	21:12
cbreak	how?	21:12
	→ __BigO__ joined	21:13
Kartagis	git show	21:13
cbreak	where?	21:13
Kartagis	in the directory where .git is	21:14
	← Sicp left	21:14
cbreak	that's the wrong place	21:14
	you have to do it on the server	21:14
	where you pushed to	21:15
	← zeroSignal left	21:15
	← apok left	21:15
Kartagis	I have no SSH access there	21:15
	to it*	21:15
cbreak	then you can't push	21:15
	of course you can't see the commits	21:16
	← bzyx left	21:16
	→ HardDisk_WP joined	21:16
	← HardDisk_WP left	21:16
	→ HardDisk_WP joined	21:16
	← kevinfagan left	21:16
	→ bzyx joined	21:16
	← nkr left	21:16
	← punknology left	21:17
	→ jpcamara joined	21:18
Kartagis	cbreak: I could push to this day	21:18
	← Saby left	21:18
	→ maletor joined	21:18
	← jpcamara left	21:18
	→ punknology joined	21:18
	← punknology left	21:18
	→ _FrnchFrgg_ joined	21:19
	→ jpcamara joined	21:19
	→ punknology joined	21:19
	← punknology left	21:19
	→ lresende_ joined	21:19
	→ punknology joined	21:20
	← punknology left	21:20
	→ bigkm joined	21:20
	→ punknology joined	21:20
	← punknology left	21:20
	← hacktose left	21:20
	← sdeboey left	21:21
	← tinti left	21:21
	→ punknology joined	21:21
	← punknology left	21:21
	→ punknology joined	21:22
	← punknology left	21:22
	→ punknology joined	21:23
	← punknology left	21:23
	→ RaceCondition joined	21:23
	← lresende left	21:23
	lresende_ → lresende	21:23
	→ punknology joined	21:23
	← punknology left	21:23
	→ s0ber_ joined	21:24
	← jphillips left	21:24
	→ punknology joined	21:24
	← punknology left	21:24
	→ youf joined	21:24
	← protux left	21:24
	→ punknology joined	21:25
lov	I want to create a new remote branch based on a certain commit, rather than my HEAD. How would I go about doing this? I'm just confused about what to use for the refspec	21:25
	← punknology left	21:25
	→ kevinfagan joined	21:25
	→ punknology joined	21:25
lov	nevermind, got it	21:25
	← punknology left	21:26
	→ protux joined	21:26
	s0ber_ → s0ber	21:26
cbreak	lov: git branch blah blabhabhla	21:26
	← picoguy left	21:26
	→ punknology joined	21:26
	← punknology left	21:26
cbreak	lov: git push -u origin blabhabhla	21:26
	← dankest left	21:26
lov	cbreak: thanks.	21:27
	← jceb left	21:27
lov	actually, it was the blahblahblah I was asking about	21:27
	→ punknology joined	21:27
	← punknology left	21:27
	→ jwmann joined	21:27
	→ RonScagz joined	21:27
cbreak	that happens a lot	21:27
	← swiety left	21:28
	→ punknology joined	21:28
	← bzyx left	21:28
	← jarr0dsz left	21:29
	→ jarr0dsz joined	21:29
	← jarr0dsz left	21:29
	→ Davey joined	21:29
	← b1rkh0ff left	21:30
	→ tinti joined	21:30
	← RaceCondition left	21:31
	→ cagedwisdom joined	21:31
	→ ErKa joined	21:31
	← ferdev left	21:32
	← rakl left	21:32
	→ ferdev joined	21:33
	→ sdeboey joined	21:33
	← peteski left	21:34
	← alorence left	21:35
	→ tallest_red joined	21:35
	← gustavnils left	21:36
	← jamesez left	21:36
	← kit left	21:37
	→ normanrichards joined	21:37
	← rideh left	21:37
	→ itatitat joined	21:38
	→ Hounddog joined	21:39
	→ skipm joined	21:40
	→ b1rkh0ff joined	21:41
	← P4C0 left	21:41
	→ diamonds joined	21:41
diamonds	http://pastie.org/pastes/6258003/text	21:41
	I committed, had one extra file I didn't mean to commit	21:41
	← jaequery left	21:42
diamonds	reset --soft HEAD~ && git add ... && git commit; now I have the proper commit & the messed up one as a little "branch"?	21:42
	← skipm left	21:42
diamonds	so I want to go back, discard all those commits...	21:43
	← wasanzy left	21:43
diamonds	basically how do I destroy an index	21:43
	← hoschi left	21:44
	← RonScagz left	21:44
	← fsvend left	21:44
	← alexbligh left	21:44
	← tbmiller left	21:46
	← asaf left	21:47
	→ GothAlice joined	21:47
	→ asaf1 joined	21:47
GothAlice	How do I rebase a branch based off of my current branch, without switching to the branch I'm rebasing?	21:47
	← geekbri left	21:47
	← fission6 left	21:47
GothAlice	(I'm in an environment where switching branches is a Serious Thing™.)	21:47
bnason	Does anyone have recommendations for project deployment? My predecessor set us up where we are actually deploying with git.	21:47
ojacobson	GothAlice: nil.	21:47
diamonds	here's an image from gitk of my issue http://screencast.com/t/9IY5X78EQ	21:47
ojacobson	You could do the rebase in a scratch repo, though	21:48
diamonds	I want to discard those last two tips	21:48
bnason	GothAlice, make a copy of the repo?	21:48
GothAlice	bnason: I also deploy using Git. https://github.com/marrow/WebCore-Template is an example of the base for apps developed using my web framework.	21:48
bnason	bah beat me hehe	21:48
	→ Skipp_OSX joined	21:48
GothAlice	If a rebase is not an option (other than making a new clone), is merging the current branch into another local branch an option?	21:49
	→ fission6 joined	21:49
_ikke_	GothAlice: merge and rebase result in the same code, only the history looks different	21:49
diamonds	I'd reset --hard but I'm concerned about losing my changes	21:49
Skipp_OSX	hello, is there a way to get git to give you a warning if you acidentally commit nothing, that is if you commit without adding any files or git screws up?	21:49
	← freeed left	21:49
diamonds	obviously reset is not the command I should be using	21:50
_ikke_	Skipp_OSX: That's default behaviour	21:50
ojacobson	Skipp_OSX: that would be 'git commit''s default behaviour.	21:50
Skipp_OSX	I just lost a commit and I'm so angry at myself for it	21:50
	← infynyxx left	21:50
_ikke_	Skipp_OSX: What do you mean, lost a commti?	21:50
	← lresende left	21:50
GothAlice	bnason: We then have GitHub push notify the app server of updates. https://gist.github.com/amcgregor/11d78f5ed471376a0ee7#file-updater-py is a naive implementation of this.	21:50
_ikke_	Skipp_OSX: It's very hard in git to loose commits	21:50
ojacobson	loose/lose	21:50
	→ lresende joined	21:50
_ikke_	:)	21:50
Skipp_OSX	I added files, did a git commit, wrote a commit message, added some more files, another commit message, but when I pushed, the first commit was not there, only the second	21:51
Nevik	s/loo/s\//	21:51
	er	21:51
Skipp_OSX	and all the changes were in the second commit	21:51
Nevik	s/loo/s\/loo/	21:51
	→ eimajenthat joined	21:51
	← protux left	21:51
	→ tonymet joined	21:51
_ikke_	Skipp_OSX: Then you didn't commit the first time	21:51
Skipp_OSX	which was really the first commit as there was only one commit...	21:51
	_ikke_, yeah, apparently. I'm just not sure how that happened	21:52
	← sdeboey left	21:52
_ikke_	!repro?	21:52
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.	21:52
eimajenthat	Okay, I'm trying to setup a pre-commit hook, but it doesn't appear to be running. Looking at the sample pre commit hook, it says it is run when you run "git commit" with no arguments. Does this mean that if I use -m for my commit message, the pre-commit hook won't run? Is there a workaround?	21:53
Skipp_OSX	unfortunately I closed the term window so that's gone now	21:53
_ikke_	eimajenthat: is the pre-commit hook executable?	21:53
bnason	GothAlice, I'll probably have to setup something custom to get what I'm looking for. Essentially I want a 'guided deploy' (at least if I stick with git deployment). Where each step is reported on and the user can Stop or Continue with the deployment based on the results. I'm not confortable with automated deployment yet at least.	21:53
ojacobson	'git-commit' won't even prompt for a message if it's not going to allow the commit	21:53
	→ kogent joined	21:53
ojacobson	pre-commit and emptiness checking both happen first	21:53
	→ bbloom joined	21:53
	→ dankest joined	21:54
	← kogent left	21:54
eimajenthat	_ikke_: Derp, probably not. I'll check	21:54
_ikke_	Skipp_OSX: Did you remove the commits message and saved + closed the editor?	21:54
Skipp_OSX	I'm trying to reproduce what I did	21:54
	→ kogent joined	21:54
	← Nneon left	21:54
Skipp_OSX	_ikke_, I believe that I did	21:54
Nevik	Skipp_OSX: your shell probably keeps a history file somewhere	21:54
	you can look at that	21:54
_ikke_	Skipp_OSX: when the commit message is empty, git aborts the commit	21:54
Skipp_OSX	_ikke_, I know, maybe I did a q! instead of a wq ???	21:56
	(vim)	21:56
	← satiricon left	21:56
	→ jaequery joined	21:56
GothAlice	bnason: We are pretty strict about our pre-commit hooks and use of git flow to ensure only QA-approved commits get into master.	21:56
Skipp_OSX	It's possible I suppose...	21:56
	→ todd_dsm joined	21:56
GothAlice	bnason: e.g.: https://github.com/amcgregor/snippits/tree/hooks	21:57
Nevik	Skipp_OSX: if no one has based work off your published commit, you can split it in two by rebasing interactively	21:57
eimajenthat	_ikke_: that was it, thanks	21:57
Skipp_OSX	Nevik, unfortunately I've already pushed to a public repo so it's too late to rewrite history non-destructively	21:57
Nevik	Skipp_OSX: as i said, if no one's based anything off it, it wont hurt anyone if you rewrite	21:58
	← lifelike left	21:58
Skipp_OSX	I could perhaps do a revert and then recommit	21:58
Nevik	since you say that was the first commit at all, it's probably unlikely someone has done anything with it	21:58
	→ lifelike joined	21:58
	→ satiricon joined	21:58
Skipp_OSX	Nevik, yeah but the history would be rewritten	21:58
	→ pandeiro joined	21:58
_ikke_	Skipp_OSX: Doing a revert only makes things more complicated	21:58
Nevik	Skipp_OSX: so?	21:59
	← SirCmpwn left	21:59
_ikke_	Skipp_OSX: It's only a problem if someone has committed his changes on top of that history	21:59
Skipp_OSX	Nevik, so if somebody has pulled from that branch and I rewrite history their local tree will diverge from origin...	21:59
	← opus left	22:00
Nevik	Skipp_OSX: is it likely that someone has pulled in the meantime?	22:00
Skipp_OSX	Nevik, yes, it is likely, big project	22:00
	I can't say for certain though	22:00
Nevik	well, then just leave it as it is	22:00
Skipp_OSX	I'm less concerned with fixing this though and more concerned with preventing future occurances	22:01
	→ protux joined	22:01
Nevik	well, git wont lose commits	22:01
	so it was probably your own fault :)	22:01
_ikke_	Skipp_OSX: Better read the output from git to see what actually happened	22:01
Nevik	exactly	22:01
	git will always report when it does something	22:01
	→ jxpx777 joined	22:02
Skipp_OSX	_ikke_, well, the commit log shows 1 commit with all files added	22:02
	I'm guessing what happened is there was an error in my first commit, perhaps git warned me and I didn't see it.	22:02
_ikke_	Skipp_OSX: Yes, you probably aborted the first commit	22:02
	yes, indeed	22:02
	→ opus joined	22:02
	→ honschu_ joined	22:03
	→ Octalot joined	22:04
	→ adamben joined	22:04
	→ SirCmpwn joined	22:04
	← asaf1 left	22:05
	← rtjure left	22:05
	← GothAlice left	22:05
Skipp_OSX	it looks like it does warn you if you abort a commit	22:06
_ikke_	It does	22:06
Skipp_OSX	darn, I wish I didn't close my session...	22:06
	aha!	22:07
	→ simesy joined	22:07
ojacobson	Skipp_OSX: this is why my terminal prompt has an exit status indicator in it :)	22:07
	I don't close windows that end with :-(	22:07
_ikke_	lol	22:07
Nevik	ha	22:07
Skipp_OSX	I just got the following error while testing: error: There was a problem with the editor 'vim'. Please supply the message using either -m or -F option.	22:07
	← honschu left	22:07
	← kevinfagan left	22:07
ojacobson	https://gist.github.com/ojacobson/5000113	22:07
_ikke_	ojacobson: bash?	22:08
ojacobson	_ikke_: I'm sure oh-my-zsh has something similar you can use.	22:08
_ikke_	right	22:08
	← kevwil left	22:08
_ikke_	Does bash call ps1_status itself?	22:09
Skipp_OSX	okay, this might be a vim problem	22:09
ojacobson	it does	22:09
	The single-quoted form there stops bash from calling it while defining the prompt	22:09
	the result is that a literal $(...) sequence appears in PS1, which then gets evaluated when PS1 is rendered for a prompt	22:10
	← AaronMT left	22:11
_ikke_	That's not really hard to build	22:12
	→ sfalcon joined	22:12
	→ infynyxx joined	22:13
Skipp_OSX	okay, this is interesting, if I do a git commit, edit the commit message in vim, do a :q and vim will yell at me that there are unsaved changes, then I do a :wq and vim writes and exits, but, git does not perform the commit and I get no warning	22:14
_ikke_	hmm	22:14
ojacobson	Skipp_OSX: What's the command and the full output from git? pastebin please	22:14
	← howlymowly left	22:14
	→ Michael__ joined	22:14
Skipp_OSX	if I then do an echo $? I get 1 indicating that there was a non-zero exit code	22:15
	→ sh4rm4 joined	22:15
	→ znode joined	22:15
	← jpcamara left	22:16
_ikke_	Skipp_OSX: Your vim is bust	22:17
	somehow	22:17
Skipp_OSX	http://pastebin.com/VGPD1Uxx	22:18
	I got a warning this time though	22:18
osse	Skipp_OSX: are you sure you have staged files?	22:18
	ahh, never mind	22:19
	→ grzlwopf joined	22:19
Skipp_OSX	lemme try zsh to see if it warns me, might not have used bash	22:19
	← youf left	22:19
	→ FernandoBasso joined	22:19
	← znode left	22:20
Skipp_OSX	the first commit I did :q then :wq, and it failed, the second commit I did :wq and it succeeded	22:20
osse	Skipp_OSX: do you get the output of 'git status' in the editor?	22:20
jhford	anyone have any ideas why a commit id can be rev-parsed but not used in git log? http://pastebin.mozilla.org/2162234	22:20
Skipp_OSX	osse, yes	22:20
	→ dsantiago joined	22:20
	← fobban left	22:20
ojacobson	jhford: rev-parse doesn't actually try to read the object	22:20
	afaik	22:20
	git-log does.	22:20
	I can run git rev-parse 4c760023217ec7ae1a55e20993e28efe2cff6f01 in a repo I have here that has no such commit, for example. :)	22:21
	(it behaves identically)	22:21
jhford	hmm, the weird thing is that the object definatey exists on github :S https://github.com/mozilla-b2g/gaia/commit/4c760023217ec7ae1a55e20993e28efe2cff6f01	22:21
	← bgerber left	22:21
ojacobson	But is it reachable from any refs on github?	22:21
	(github runs git gc somewhat lazily)	22:22
Skipp_OSX	okay, I'm not able to reproduce not getting the warning so I assume that I must have just ignored it... but I've still got a vim problem here	22:22
	← j416 left	22:22
	→ dreiss joined	22:22
ojacobson	(un-gc'd but unreachable objects can remain available on github for quite a while even though 'git fetch' will no longer retrieve them.)	22:22
	← pandeiro left	22:22
ojacobson	git fetch; git branch --contains <hash> --all	22:22
	← ncd\|leeN left	22:22
Skipp_OSX	I think I've figured it out: http://tooky.co.uk/2010/04/08/there-was-a-problem-with-the-editor-vi-git-on-mac-os-x.html	22:22
engla	ojacobson: so github also exposes unreachable objects?	22:23
jhford	ojacobson: sorry for the hassle, i needed to git fetch	22:23
	jhford feels like a dummy	22:23
jhford	i have a script that's supposed to do that	22:23
	engla: they cache the html for the objects	22:23
ojacobson	engla: if you happen to have the URL of, say, a commit that has since become unreachable, yes, that URL will remain valid for "a while"	22:23
engla	ok	22:23
ojacobson	the unreachable object will eventually be GCed and become unavailable from github	22:23
	there's also HTML caching	22:24
jhford	i had lots of fun learning about that :)	22:24
	→ Fissure joined	22:24
engla	I've done some experiments recently that should have created a lot of garbage on github so I wondered about gc too. good to know that they at least to it at some point	22:25
	← Carlitos__ left	22:25
engla	do the gc/prune	22:25
osse	Skipp_OSX: does your vimrc contain errors?	22:25
	← Anders_J left	22:25
	→ bgerber joined	22:25
ojacobson	engla: I know that in github enterprise all repos are created with the reflog enabled	22:25
Skipp_OSX	osse, not that vim complains about...	22:25
ojacobson	so there's at least a 2-week window where the unreachable object will remain available	22:25
	← dsantiago left	22:25
ojacobson	that's probably also true of real github	22:25
_ikke_	ojacobson: afaik, the 2 week has nothing to do with the reflog	22:25
ojacobson	_ikke_: I thought the default reflog expiry was 2 wk	22:25
engla	reflog would be 90 days	22:26
ojacobson	oop	22:26
	poor memory on my part, then	22:26
	← hellopat left	22:26
_ikke_	git prune doesn't remove objects younger than 14 days	22:26
engla	2 weeks is for loose objects	22:26
_ikke_	afaik	22:26
	← skorgon left	22:26
	← _FrnchFrgg_ left	22:26
Skipp_OSX	I put the following in my .bash_profile and it fixed it GIT_EDITOR='/usr/bin/vim -f'	22:27
	← agile left	22:27
Skipp_OSX	well, with export in front of that...	22:27
	→ agile joined	22:27
	← dankest left	22:27
	→ duosrx joined	22:27
	→ Dooom joined	22:28
osse	Skipp_OSX: weird you should need -f, is it gvim/macvim you're using?	22:28
engla	does anyone here use expiring stash? I can't get it to expire after a fixed time (What I see, either it lives forever, or git gc kills it directly)	22:29
	→ asaf joined	22:29
jhford	ojacobson: do you know how lazy the github gc'ing is?	22:29
Skipp_OSX	osse, it's vim on a Mac, but just regular vim	22:29
	← scampbell left	22:29
	← agileadam left	22:29
	← protux left	22:30
	← ferdev left	22:30
_ikke_	Skipp_OSX: Does it open in graphic window?	22:30
Skipp_OSX	the one that comes with the system (10.6)	22:30
	_ikke_, no, terminal	22:30
	← tevfik left	22:30
bnason	So I've inherited a not well maintained repo. We just did a fresh install of the base software without any of our customizations. Is there a way I can make a new fresh repo, and apply the changes I want from the old repo on it?	22:30
Kartagis	idk what I did, but my commits are showing now	22:31
_ikke_	bnason: Where does this base version come from?	22:31
osse	Skipp_OSX: weird. According to the manpage -f shouldn't have any effect on terminal vim.	22:31
	oh well	22:31
	→ protux joined	22:31
bnason	_ikke_, vendor installation	22:31
_ikke_	bnason: And these changes? How do they relate to that vendor installation?	22:31
	← protux left	22:32
Skipp_OSX	it looks like the vim that comes with 10.6 is slightly broken in that way from what I read	22:32
	→ smw joined	22:32
bnason	_ikke_, they modify the files created by the vendor install.	22:32
_ikke_	bnason: Create a new repo, commit the base version	22:32
osse	Skipp_OSX: if you have macvim you can use that as a replacement for the regular vim by using mvim -v, or making a symlink from vim to mvim	22:33
_ikke_	bnason: create patches for the changes, and apply them to the new repo	22:33
osse	Skipp_OSX: it's newer and more fully featured (I've heard)	22:33
	→ ferdev joined	22:33
Skipp_OSX	osse, good advice	22:33
osse	Skipp_OSX: a lot of people aren't aware that macvim also works in the terminal :)	22:34
jhford	ojacobson: it seems that my issue is that fetching from an on disk repository wasn't fetching the commit	22:34
Skipp_OSX	I certainly wasn't	22:34
	jhford curious why fetching from a repo that has that commit wouldn't give me the commit	22:34
bnason	_ikke_, thanks. Just gonna take some time. I definitely like my history to not be full of crap commits ;)	22:35
osse	Skipp_OSX: another soul saved!	22:35
Skipp_OSX	well, thank you all for your help	22:35
HardDisk_WP	In a repository, I have created a file with >100 commits	22:35
_ikke_	bnason: another way would be to fetch from the old repo, and to cherry-pick	22:35
HardDisk_WP	Now, the time has come to move this and another file into its own repository	22:35
	How do i do this without losing history?	22:35
_ikke_	HardDisk_WP: with filter-branch afaik	22:36
	← dreiss left	22:36
_ikke_	man git filter-branch	22:36
gitinfo	the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	22:36
_ikke_	depending on whether those files are in their own directory	22:36
bnason	_ikke_, yea the old commits are all 'commit missing file' type commits heh	22:37
	but a lot of random files	22:37
	→ dankest joined	22:37
HardDisk_WP	_ikke_, file #1 is examples/ftpfs.php, but in the examples/ directory there are other files which should not be exported	22:37
	← Kartagis left	22:39
	→ skorgon joined	22:39
_ikke_	HardDisk_WP: Not sure if it would work, but maybe you can experiment by moving those files in a seperate directory, comitting that, and then use git filter-branch with the --subdirectory-filter	22:40
arand	HardDisk_WP: Create a clone and use something like "git filter-branch --index-filter rm {all except wanted files}"	22:41
HardDisk_WP	ok	22:41
	← adamben left	22:41
HardDisk_WP	first the subdir filter to remove the non-examples stuff	22:41
arand	_ikke_: I think you can't make it forget directory history like that.	22:41
HardDisk_WP	then the index-filter rm to remove the unneeded exaples	22:41
arand	HardDisk_WP: That's probably the quickes, yeah.	22:41
	HardDisk_WP: Make sure to do this in a copy of the mmain repo ;)	22:42
HardDisk_WP	sure :D	22:42
_ikke_	arand: It shouldn't forget history	22:42
HardDisk_WP	the project is on github anyway^^	22:42
	→ j^2 joined	22:42
	← danillonunes left	22:42
	← ferdev left	22:42
	← trungqtran left	22:43
arand	_ikke_: if ever a file was in that dir, I think ---subdir would still keep it.	22:43
_ikke_	arand: In what dir?	22:43
	→ ferdev joined	22:43
arand	_ikke_: the one you suggested moving stuff out from.	22:44
_ikke_	arand: I mean he should create a new dir	22:44
	But not sure if the subdir filter follows renames	22:44
arand	I don't think it does, given the wording in the manpage	22:44
	"history which touches the given subdirectory"	22:45
	→ Gitzilla joined	22:45
	→ sdeboey joined	22:45
	→ Gentlecat_ joined	22:46
HardDisk_WP	hmm	22:46
	the git subdir filter worked	22:46
	← fission6 left	22:46
HardDisk_WP	git filter-branch -f --index-filter "rm example_1.php"	22:46
	Rewrite 12d4130db6181fb2fb5117a1e72d7da06fd4d18f (1/59)rm: cannot remove `example_1.php': No such file or directory	22:46
	→ CausaMortis joined	22:47
_ikke_	--ignore-unmatch	22:47
arand	1. Use "git rm --cached" 2. use --ignore-unmatch	22:47
	← Skipp_OSX left	22:48
	→ rwestphal joined	22:48
	← juvenal left	22:48
	→ dreiss joined	22:48
	→ avjpl joined	22:49
boze	I moved my head back locally ... I want the commits I made to the server to disapear, but when I try to push it says non-fast-forward updates were rejected. How can I force it?	22:49
_ikke_	git push -f	22:49
	!rewrite applies	22:49
gitinfo	[!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to). If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://git-scm.com/docs/git-rebase.html#_recovering_from_upstream_rebase	22:49
HardDisk_WP	ok, now the files were removed	22:50
	very good	22:50
	but git log still shows "bad" commits	22:50
boze	_ikke_: thank you	22:50
	← infynyxx left	22:50
HardDisk_WP	"bad" as in "these commits have not touched the file"	22:51
_ikke_	HardDisk_WP: In what way are they bad?	22:51
	HardDisk_WP: Well, they probably have touched those other files	22:51
HardDisk_WP	there are a number of commits which are before ftpfs.php was created!	22:51
_ikke_	You need to prune them	22:51
HardDisk_WP	how?	22:51
arand	HardDisk_WP: --prune-empty	22:51
	← psoo left	22:51
	← sdeboey left	22:51
	→ dsantiago joined	22:52
	→ sdeboey joined	22:53
HardDisk_WP	Ran git filter-branch -f --prune-empty -- --all	22:53
	, didn't help	22:53
	also, how do I remove all the refs and branches?	22:53
	← relipse left	22:53
	← dreiss left	22:54
	→ RaceCondition joined	22:54
_ikke_	HardDisk_WP: Did git give a message?	22:54
HardDisk_WP	Yeah, something about warning: ignoring duplicate parent, and some warning: ref xxx is unchanged	22:55
	→ zoonfafer joined	22:56
	→ eimajenthat_ joined	22:56
_ikke_	I'm not sure	22:56
alp82	wow, finally found a nice gui for git in linux	22:56
_ikke_	alp82: What?	22:56
	HardDisk_WP: Can you check if those commits are actually empty?	22:56
	git show <commit>	22:56
alp82	_ikke_: smartgithg	22:56
_ikke_	hmmm...	22:57
	Sounds interesting	22:57
	→ soypirate joined	22:57
	← bbloom left	22:58
HardDisk_WP	_ikke_, yeah the commit shows up there	22:58
	but a git checkout 6c4bacd555194bffb42e811d0d4c4c6904990d72 followed by ls -lah shows no files except .git/	22:58
	→ damienbrz joined	22:58
	← fsvehla left	22:58
	← alunduil left	22:59
	← breakingthings left	22:59
	← eimajenthat left	22:59
	eimajenthat_ → eimajenthat	22:59
_ikke_	HardDisk_WP: Well, git considers a commit "empty", when the diff with the previous commits is empty, not when the tree is empty	22:59
	← jdunck left	22:59
HardDisk_WP	ok	22:59
	well 6c4bacd555194bffb42e811d0d4c4c6904990d72 is the first commit	22:59
	← fisted left	23:00
HardDisk_WP	aah, and the next commit is the one where I created the file.	23:00
_ikke_	you probably need to rebase --root	23:00
	← simesy left	23:00
_ikke_	and leave out that first commit	23:00
HardDisk_WP	Oh, crap	23:00
_ikke_	?	23:00
HardDisk_WP	I can't get back to HEAD	23:00
	→ simesy joined	23:01
_ikke_	HardDisk_WP: You are always on HEAD ;)	23:01
arand	--prune-empty ignores merge points, I think.	23:01
_ikke_	arand: it does	23:01
	← simesy left	23:01
HardDisk_WP	after git checkout $firstcommit I got You are in 'detached HEAD' state. You can look around, make experimental	23:01
	← opus left	23:01
_ikke_	correct	23:01
	→ simesy joined	23:01
	→ fisted joined	23:01
HardDisk_WP	how do I get back to the "old" HEAD?	23:01
_ikke_	if you want to go back, git checkout -	23:01
	→ jdunck joined	23:01
HardDisk_WP	ah :)	23:01
arand	HardDisk_WP: or "git checkout master", if you were on master	23:02
	← avjpl left	23:02
_ikke_	Well, I don't think all the refs got rewritten	23:02
	→ AartBluestoke joined	23:02
_ikke_	but the reflog has you covered ;)	23:02
	→ hwrd\|work joined	23:02
	→ dreiss joined	23:03
HardDisk_WP	hmm strange	23:03
_ikke_	HardDisk_WP: What is?	23:03
HardDisk_WP	I did git rebase --root --onto=4eb48fe2ab0aeb4dcce96d91ab2f7adbac0599a8	23:03
	but the first commit in git log still is 6c4bacd555194bffb42e811d0d4c4c6904990d72	23:03
_ikke_	you probably want to use git rebase -i --root	23:04
	← TonyM left	23:04
	← nickn left	23:05
_ikke_	because 6c4bacd555194bffb42e811d0d4c4c6904990d72 is probably a child of the root commit, I guess it's a noop	23:05
	← lausi left	23:05
HardDisk_WP	_ikke_, now this is bad, neither the old root, nor 4eb48fe2ab0aeb4dcce96d91ab2f7adbac0599a8 show up in the editor	23:05
	→ lausi joined	23:05
	← anonymuse left	23:05
HardDisk_WP	only that what was the 3rd commit is now the 1st commit to show in the editor	23:05
_ikke_	remove all	23:06
	(ie, abort the rebase)	23:06
	← ojacobson left	23:06
HardDisk_WP	But the comment line says # Rebase 5d2032d onto 4eb48fe	23:06
	→ texasmynsted joined	23:06
	→ pierre1_ joined	23:07
	→ opus joined	23:07
	← AnimeCYC left	23:07
HardDisk_WP	_ikke_, can I safely proceed?	23:07
_ikke_	HardDisk_WP: Not sure	23:08
	→ AnimeCYC joined	23:08
HardDisk_WP	ah well i'll just try it and redo if I mess up :D	23:08
	← jamescarr left	23:08
texasmynsted	what is the simplest way to uncommit some changes to a submodule? In error some changes were made in a submodule and they were committed. I think it should be possible to remove the submodules and re-add them but it seems like it should also be possible to rebase or something	23:08
	← j^2 left	23:08
	← pr0ton left	23:09
_ikke_	i g2g	23:09
	nite	23:09
nevyn	texasmynsted: reset?	23:09
	→ pr0ton joined	23:09
HardDisk_WP	wthf, the edited commit messages show up just fine	23:09
	← jonshea left	23:10
	← dreiss left	23:10
HardDisk_WP	but there's still the wrong root	23:10
	← giallu left	23:10
texasmynsted	oops yes I mean reset. I guess this should work for for the changes to a submodule also	23:11
	thank you nevyn	23:11
	← bnason left	23:12
	← theRoUS left	23:12
	→ bmac_ joined	23:13
	← bmac_ left	23:13
	← Hounddog left	23:13
	→ NiallIain joined	23:13
	→ oren1 joined	23:13
	→ apok joined	23:15
	← arand left	23:16
	→ nanun joined	23:16
	← ThePawnBreak left	23:17
	→ j^2 joined	23:17
	← v0n left	23:17
	← honschu_ left	23:18
	← oren1 left	23:18
	→ Caraes_Naur joined	23:18
	→ owen1 joined	23:18
HardDisk_WP	ah craaap, now I totally messed up...	23:18
	← Caraes_Naur left	23:18
owen1	how to delete master branch and replace it with foo ?	23:18
	← mikeym_ left	23:21
	← maqr left	23:21
	→ maqr joined	23:22
	→ SamB_ joined	23:22
	← diamonds left	23:23
	← fly9 left	23:24
	← dankest left	23:24
	← bauruine left	23:24
	← damienbrz left	23:25
	→ jpcamara joined	23:25
	← __BigO__ left	23:26
	← tewecske left	23:26
	→ marhaban joined	23:29
HardDisk_WP	hmm, ok now I got it to a state where git log -p shows a ton of empty commits as beginning	23:30
	how to delete or squash them? git rebase fails, it says "cant apply..."	23:30
	→ macrover joined	23:31
	→ jarr0dsz joined	23:31
	← kristofers left	23:31
	← diegoviola left	23:32
	→ lalaland joined	23:32
	← lalaland left	23:32
	← natechan left	23:33
	← aartist left	23:33
carandraug	HardDisk_WP, filter-branch with --prune-empty ?	23:34
	← pantsman left	23:34
	← soypirate left	23:34
	← lolwhut left	23:35
HardDisk_WP	doesn't do all the work	23:35
	https://github.com/msmuenchen/php-ftpfs/commits/master?page=2	23:35
	← simesy left	23:35
HardDisk_WP	I want https://github.com/msmuenchen/php-ftpfs/commit/4eb48fe2ab0aeb4dcce96d91ab2f7adbac0599a8 as root	23:36
	instead of https://github.com/msmuenchen/php-ftpfs/commit/6c4bacd555194bffb42e811d0d4c4c6904990d72	23:36
	← beastd left	23:36
	← sStewie left	23:36
	→ AaronMT joined	23:36
	→ simesy joined	23:37
	← macabre left	23:37
	← chrisgeorge left	23:37
	→ dankest joined	23:37
	→ chrisgeorge joined	23:37
	← AaronMT left	23:38
	← nanun left	23:38
	→ prsn joined	23:38
	← willob left	23:39
	→ willob joined	23:39
	→ relipse joined	23:39
	→ AdrienBrault joined	23:40
	→ juvenal joined	23:41
	← PaulCapestany left	23:41
	← etcetera left	23:42
	← Element9 left	23:43
	→ daniel_-_ joined	23:43
	← chrisgeorge left	23:44
	← RaceCondition left	23:45
carandraug	HardDisk_WP, is that a merge point?	23:45
	← mastro left	23:45
HardDisk_WP	nope	23:45
	→ RaceCondition joined	23:45
HardDisk_WP	6c4 is the commit where the directory got created	23:45
carandraug	HardDisk_WP, if you only have 4 commits, why don't you export them, and use them to start a new repo?	23:46
HardDisk_WP	carandraug, this is page 2 :D	23:46
	should be 39 commits	23:46
	← NiallIain left	23:46
HardDisk_WP	or such	23:46
	← lavaman left	23:46
carandraug	HardDisk_WP, ah! ok	23:46
	← daniel_- left	23:46
carandraug	HardDisk_WP, I don't know. I'm having problems myself rewriting the history because I can't get rid of empty commits because they are merges	23:47
	← sq-one left	23:47
prsn	hi all. For some reason, starting about a week ago, pulling from our repo results in all local files being -rw-r--r-- and owned by the person pulling. These permissions are then maintained when pushed, with annoying results when files are checked out directly from origin/master into a shared directory	23:47
carandraug	It seems that I have to play with parent-filter but the documentation is chinese to me	23:48
	← zoonfafer left	23:48
	→ etcetera joined	23:49
	← opus left	23:49
carandraug	prsn, what filesystem?	23:49
prsn	carandraug: um, the shared dir is on a debian box, some of us are on osx, some windows	23:50
carandraug	prsn, tell windows devs to use "git config core.filemode false"	23:50
	← hobodave left	23:51
carandraug	prsn, or anyone that keeps their files in funny behaving filesystems such as FAT. Not sure what happens in NTFS	23:51
	← dankest left	23:51
	← etcetera left	23:52
carandraug	prsn, can't you check on the log --stat which commit changed it the first time?	23:53
prsn	carandraug: that sounds useful, how will it be noted? there have been a LOT of commits since then	23:53
	→ eddd joined	23:54
	→ nanun_ joined	23:55
	→ _jmg joined	23:55
	← kermit left	23:56
	← AnimeCYC left	23:56
	→ bmac_ joined	23:56
	→ tvw joined	23:56
	→ kermit joined	23:57
	→ breakingthings joined	23:58
carandraug	prsn, "git log --summary \| grep -C 10 -iP "mode change" " or something like that	23:58
	← normanrichards left	23:59
	← duosrx left	23:59

Logs Search ←Prev date Next date→

Channels Documentation