IRCloggy #git 2013-10-02

Provider of IRC logs since 2005.

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

	2013-10-02
	← breck7 left	00:00
Dodde	just add -i somewhere in that line?	00:00
	→ dsdeiz joined	00:00
	ojacobson_ → ojacobson	00:00
	← aldodelgado left	00:00
	→ minim joined	00:00
samd	@Dodde what are you asking? are you asking how to set config options?	00:01
	← King_Hual left	00:01
Dodde	git grep --all-match -i '^.?inserted\s+more.?$'	00:01
	→ petey joined	00:02
	← jkremser left	00:02
	→ threesome joined	00:02
Dodde	sorry I just know perl, not git :/	00:02
	← Zwei_ left	00:02
samd	git has a grep.patternType config	00:02
	which can be set to perl	00:02
	to use perl regex	00:03
Dodde	yes, but you have to tell it that it should ignore case, with -i flag	00:03
samd	yes	00:03
	that is the problem	00:03
	It doesn't work	00:03
	or at leat it seems that way	00:03
	→ natechan joined	00:03
Dodde	but in your example example: git log --all-match --grep='^.?inserted\s+more.?$' --grep='^.?Review\s+URL.?$' I can't see an -i anywhere	00:03
	but in your example example: git log --all-match --grep='^.?inserted\s+more.?$' --grep='^.?Review\s+URL.?$' I can't see an -i anywhere	00:04
	oh	00:04
samd	no so that is the first case which works and gives me the match	00:04
	but then if I use -i and cuange the URL -> uRL	00:04
	it gives nothing	00:04
	← petey_ left	00:05
	← hwrd\|work left	00:05
Dodde	if you added -i at the right place, it should..	00:05
	→ Squarism joined	00:06
Dodde	I can't verify the git syntax though	00:06
samd	yeah I tried before and after, tried the --regexp-ignore-case	00:06
	well the search does work	00:06
	where it is case sensitive	00:06
	→ Pullphinger joined	00:07
samd	I found the single entry I expected	00:07
	← msheiny left	00:07
samd	but as a test I did as I mentioned and flipped the U -> u	00:07
Dodde	yeah it's strange	00:07
samd	and found nothing	00:07
Dodde	hope someone else can answer	00:07
	← cmpera left	00:08
samd	Linux version 1.8.4	00:08
	← kukks left	00:09
	← fluxdude left	00:09
samd	using Ubuntu PPA: deb http://ppa.launchpad.net/git-core/ppa/ubuntu raring main	00:09
	→ gitpicker joined	00:10
	→ Swayz joined	00:11
	→ LuxuryMode joined	00:12
	← mfulz left	00:14
	← salamanderrake left	00:14
	← erichynds left	00:14
	← LuxuryMode left	00:16
	→ LuxuryMode joined	00:16
	← maestrojed left	00:16
	← Davey_ left	00:17
	→ Davey joined	00:18
	→ msheiny joined	00:18
	← carbohydrate left	00:18
	← maroloccio left	00:19
	→ Davey_ joined	00:20
	← Hello71 left	00:21
	← ErKa left	00:21
	→ bbloom joined	00:23
	→ Hello71 joined	00:23
	→ marhaban joined	00:24
	← samd left	00:24
	→ salamanderrake joined	00:25
	← CheckDavid left	00:25
	← msheiny left	00:26
	← dsdeiz left	00:26
	→ vigs joined	00:28
	→ emergion joined	00:29
	→ rchavik joined	00:29
	← Dodde left	00:30
	→ dyng joined	00:30
	← Ivo left	00:32
	→ jefflyne joined	00:32
	← boombatower left	00:33
	← stamina left	00:33
	← Roots47 left	00:34
	← malinus left	00:34
	← salamanderrake left	00:36
	→ BlueProtoman joined	00:36
BlueProtoman	test	00:36
	→ Roots47 joined	00:37
	→ spaceone joined	00:38
	← jason237 left	00:38
spaceone	hi	00:38
	→ john_doe_jr1 joined	00:38
spaceone	how can i combine 2 commits to one (so when i push that only 1 commit is displayed)	00:38
BlueProtoman	My friend and I are having some merge conflict issues with Git. My friend has some code she wants to commit to our GitHub repo, but it says the commit is not possible due to unmerged files. We're not quite sure how to fix merge conflicts, we're new to Git.	00:39
	→ cmpera joined	00:39
BlueProtoman	Any tips?	00:39
	← mitsuhiko left	00:39
ojacobson	!conflicts	00:41
	ojacobson squints at gitinfo.	00:41
	← btanaka left	00:41
	← goshawk left	00:41
ojacobson	BlueProtoman: broadly, go through the files listed as conflicted in 'git status' one at a time. For the ones that have been edited on both sides of the merge, open them in an editor, find the conflict markers (<<< === >>> lines, by default; sometimes also \|\|\| lines)	00:41
	figure out the appropriate resolution for each conflict	00:41
	git-add the file	00:42
	repeat until done, then git-commit	00:42
	if you have other kinds of conflict (edit-move, move-move, edit-delete), use 'git add' or 'git rm' as appropriate	00:42
bremner	see also man git-mergetool	00:42
gitinfo	the git-mergetool manpage is available at http://jk.gs/git-mergetool.html	00:42
ojacobson	there are more detailed writeups out there; it wouldn't surprise me if pro-git has some	00:42
	← john_doe_jr1 left	00:42
ojacobson	if you're wondering how you got there in the first place, I bet one of you ran 'git pull'	00:42
	which is, pretty explicitly, a merge.	00:42
	→ john_doe_jr joined	00:42
	→ jaequery joined	00:43
BlueProtoman	ojacobson: Yes, it looks like us working on the same file turned out to be a bad idea.	00:43
ojacobson	well, SCM tools are not a replacement for communication and planning :)	00:44
	← dreiss left	00:44
	← shinnya left	00:44
ojacobson	you can edit the same file at the same time all you want, if you work out how your respective changes will affect one another	00:44
BlueProtoman	ojacobson: OK, so we fix the line conflicts. What then?	00:44
ojacobson	If the result looks sane, stage it	00:44
	→ ewnd9 joined	00:44
ojacobson	if you've staged all the conflicts, and the overall result looks sane, commit	00:45
	→ hwrd\|work joined	00:45
	← hwrd\|work left	00:49
	→ GODDOG joined	00:49
BlueProtoman	OK, now there's a vim window with an automatically-generated commit message. How do I pull out of it?	00:50
	→ carbohydrate joined	00:50
BlueProtoman	(On a Mac)	00:51
ojacobson	It's a commit message.	00:51
	You fill it out, just like any other commit.	00:51
	Then you save and exit the editor.	00:51
	→ ErKa joined	00:51
ojacobson	(If you don't like vim, configure git to use a different editor. :)	00:51
	← osmosis left	00:51
BlueProtoman	OK, but I don't see which key combo it is. On my own computer it's Control-X, but that doesn't seem to be working on my friend's computer.	00:51
ojacobson	Same as usual in vim: escape to get back to command mode, then :wq	00:52
	(I find vim obnoxiously modal, but knowing how to exit it is useful :)	00:52
	→ mmitchell joined	00:52
	← mmitchell left	00:52
	→ warthog9 joined	00:52
	← ivan__ left	00:52
	→ ivan joined	00:53
	→ m0ltar joined	00:54
	ivan`_ → ivan`	00:54
	→ breck7 joined	00:55
	→ g0bl1n joined	00:55
	→ tobscore joined	00:57
BlueProtoman	ojacobson: That returns an error requiring me to commit with -m or -F. When I try that, I can't push to the repo	00:57
	← GODDOG left	00:57
	→ amoe joined	00:57
ojacobson	!books	00:57
gitinfo	[!book] 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	00:57
ojacobson	pro-git has a section on merging.	00:57
	When you are collaborating with others, there will be points where you have to integrate the other guy's changes. Git has a few tools for that; merge (and pull, which calls it) is one such tool.	00:58
	Take some time and work through a few contrived examples, then come back to this.	00:58
BlueProtoman	Our homework is due in three hours.	00:58
bremner	speaking as a teacher, I'm pretty sure you had more time to work on it ;)	00:58
BlueProtoman	We're almost done.	00:59
	← mizu_no_oto left	00:59
BlueProtoman	It's just this one merge issue. I finished my part of the homework, my friend is just about done with hers, and I'm helping her finish her part of it.	00:59
	← harish left	01:02
	← normanrichards left	01:04
	← hahuang61 left	01:04
	← donmichelangelo left	01:04
	← revyd left	01:04
	→ donmichelangelo joined	01:05
	→ causasui joined	01:06
bremner	BlueProtoman: so, you don't know how to commit, is that the problem?	01:06
	← m0ltar left	01:06
BlueProtoman	bremner: No, it was some merge issues, but it looks like we may have fixed it.	01:07
	→ chrisshattuck joined	01:07
	→ bonhoeffer joined	01:07
	← TeckniX left	01:07
bonhoeffer	i have a repo locally and one on github, I _think_ they might be the same -- how can i compare them to find out	01:07
	?	01:08
	← SHyx0rmZ left	01:08
	→ hyperair joined	01:08
	← mam[idl] left	01:08
bremner	you could use git ls-remote	01:08
	→ GODDOG joined	01:08
	→ salamanderrake joined	01:08
bonhoeffer	bremner, thanks!	01:09
	→ m0ltar joined	01:09
	← carbohydrate left	01:09
bonhoeffer	bremner, i should read up on this, but I get a MD5 and HEAD, and the same MD5 refs/heads/master -- does this mean they are equal?	01:10
ojacobson	SHA1, actually	01:10
bremner	also, up to possibly having other branches	01:10
bonhoeffer	ha! good point, sorry	01:10
ojacobson	and, yes, it means both names refer to the same git object (generally, commit)	01:10
	you'll want to check if those objects are present in your local repo; try 'git show (that sha1)'	01:11
bonhoeffer	fatal: bad object . . . .	01:11
	→ mam[idl] joined	01:11
ojacobson	that's indicative that they're not the _same_ repo, but they could still be related	01:12
	how much do you care? :)	01:12
bonhoeffer	well, i would love to compare their code -- i could just clone the old one	01:12
BlueProtoman	Ah, I think we've got everything sorted out.	01:12
	→ tobscore_ joined	01:12
ojacobson	or add it as a remote to your existing repo	01:12
	git's perfectly happy to have multiple unrelated histories in the same repo	01:12
	(it's a bit dumb in most cases, but answering "do these histories overlap" is one of the better uses for it)	01:13
	→ rudak1 joined	01:13
	→ tinti joined	01:13
	← jaequery left	01:14
	← petey left	01:14
	← diegoviola left	01:14
	→ quazimodo joined	01:14
	← g0bl1n left	01:15
	← marhaban left	01:15
	← tobscore left	01:15
	← warthog9 left	01:16
	→ petey joined	01:16
bonhoeffer	i hope i didn't mess it up: https://github.com/tbbooher/ptp_nova_theme/	01:17
	→ tobscore__ joined	01:18
	← jhulten left	01:18
bonhoeffer	in my last commit -- all the files look new	01:19
	→ fridim_ joined	01:19
	→ dslr joined	01:19
	← tobscore_ left	01:19
	→ tobscore_ joined	01:20
	← valgrind left	01:20
	← tobscore__ left	01:21
	→ tmcmahon joined	01:21
	← dslr left	01:21
	← ErKa left	01:21
	→ vishal joined	01:23
	→ tobscore__ joined	01:24
	← quazimodo left	01:25
	← fridim_ left	01:26
	← tobscore_ left	01:27
	→ tobscore_ joined	01:28
	← nadirvardar left	01:30
	→ revyd joined	01:30
	← tobscore__ left	01:31
	← tmcmahon left	01:31
	← Octalot left	01:32
	→ FatDarrel_ joined	01:32
	← FatDarrel left	01:32
	FatDarrel_ → FatDarrel	01:32
	→ Vampire0_ joined	01:33
	→ quazimodo joined	01:33
	← rchavik left	01:34
	← Remram_ left	01:34
	← chrisshattuck left	01:35
	→ rchavik joined	01:35
	→ Remram joined	01:35
	← mam[idl] left	01:35
	→ mam[idl] joined	01:35
	← whunt left	01:36
	← Vampire0 left	01:36
	→ bitkiller joined	01:38
	→ tobscore__ joined	01:38
	→ intripoon joined	01:38
	→ jon_r joined	01:39
	→ bwreilly joined	01:39
	← tobscore_ left	01:39
	→ tobscore_ joined	01:40
	← LucasTizma left	01:40
	→ normanrichards joined	01:41
	← breck7 left	01:41
	→ mizu_no_oto joined	01:41
	← intripoon_ left	01:42
	→ LucasTizma joined	01:42
	← ewnd9 left	01:43
	→ breck7 joined	01:43
	← bwreilly left	01:43
	← rasto left	01:43
	← tobscore__ left	01:43
	→ TN_Jonesboro joined	01:45
	← revyd left	01:45
	← maqr left	01:45
	← TN_Jonesboro left	01:46
	→ TN_Jonesboro joined	01:46
	← ColKurtz left	01:46
	← v0n left	01:49
	← quazimodo left	01:49
	← salamanderrake left	01:51
	← Squarism left	01:52
	→ echo-area joined	01:52
	→ chrisshattuck joined	01:53
	→ tjbiddle_ joined	01:53
	← tjbiddle left	01:54
	tjbiddle_ → tjbiddle	01:54
	→ harish joined	01:54
	→ boombatower joined	01:55
	→ daito joined	01:56
	← ojacobson left	01:56
	← Mars`_ left	01:56
	← FatDarrel left	01:56
	→ cozby joined	01:57
	→ warthog9 joined	01:57
	→ Milossh joined	01:58
cozby	I'm running into some frustrating issues. When I try and checkout -f a branch I get error:unable to unlink old '.gitignore' (permission denied). This file has all the right perms... its doing this for a few files now. What gives?	01:58
	its also throwing errors thats its unable to rmdir because directories are not empty?	02:00
	I just want to blow away anything local and replace it with the remote branch	02:00
	→ w0ng joined	02:01
	← warthog9 left	02:02
	← thiago left	02:03
	→ salamanderrake joined	02:04
	→ tobscore__ joined	02:06
	→ cred joined	02:07
	→ luxgeek joined	02:07
	← tobscore_ left	02:07
	→ tobscore_ joined	02:08
	← mam[idl] left	02:08
	→ ewnd9 joined	02:08
	→ rainbyte joined	02:08
	← sriracha left	02:09
	→ nardev joined	02:10
	→ iamjarvo joined	02:11
	← tobscore__ left	02:12
	→ goshawk joined	02:15
	← breck7 left	02:15
	← pbujok left	02:17
	← BlueProtoman left	02:18
	← cred left	02:19
	← dyng left	02:20
	← thomas_b left	02:21
	← iamjarvo left	02:22
	→ dyng joined	02:23
	→ iamjarvo joined	02:23
	→ thomas_b joined	02:24
	← chrisshattuck left	02:24
	→ iamjarvo_ joined	02:26
	← m0ltar left	02:26
	← rudak1 left	02:27
	← thomas_b left	02:28
	← LuxuryMode left	02:28
	→ thomas_b joined	02:28
	→ mikepack joined	02:28
	← iamjarvo left	02:29
	← mikepack left	02:30
	→ maqr joined	02:30
	← jxpx777 left	02:31
	→ LuxuryMode joined	02:31
	→ jxpx777 joined	02:31
	→ tobscore__ joined	02:32
	← mizu_no_oto left	02:33
	← thomas_b left	02:33
	← tobscore_ left	02:33
	→ tobscore_ joined	02:34
	← Targen left	02:34
	→ eli joined	02:34
	← tobscore__ left	02:35
	← joelteon left	02:35
	→ rurufufuss joined	02:36
	← tinti left	02:36
	→ jaequery joined	02:36
	← GODDOG left	02:38
	→ thomas_b joined	02:39
	← LucasTizma left	02:41
	→ hwrd\|work joined	02:42
	← hwrd\|work left	02:42
	← cmpera left	02:43
	→ AWinterman joined	02:45
	→ kemist joined	02:46
	→ aldodelgado joined	02:47
	← maqr left	02:47
	← Devastator_ left	02:47
	→ Devastator_ joined	02:47
	Devastator_ → Devastator	02:47
	→ dustyfields joined	02:48
	← logix812 left	02:48
	← ewnd9 left	02:48
	→ GODDOG joined	02:50
	→ quazimodo joined	02:50
	→ tobscore__ joined	02:50
	→ thiago joined	02:50
	→ mouse__ joined	02:51
	→ chrisshattuck joined	02:51
	← mouse__ left	02:52
	← joki left	02:52
	→ D-Chymera1 joined	02:52
	gitinfo set mode: +v	02:53
D-Chymera1	hi guys - how is it possible to pull/merge a single file from one project into another (the projects are quite different but I would like to use a file from one in another - and record that provenience of the file)	02:53
	← tobscore_ left	02:53
	← mouse_ left	02:54
milki	D-Chymera1: you cant. however, you can certainly copy the file and ignore history	02:55
	D-Chymera1: alternatively, you can mangle with man git filter-branch and extract the history of the single file	02:55
gitinfo	D-Chymera1: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	02:55
	→ bwreilly joined	02:55
milki	D-Chymera1: it is most likely not worth it	02:55
	← iamjarvo_ left	02:55
	← jefflyne left	02:55
milki	D-Chymera1: i suppose you can merge and then delete everything not the file too	02:56
	→ iamjarvo joined	02:56
	→ tobscore_ joined	02:56
	→ chairabanta joined	02:57
	→ joki joined	02:57
	← dharrigan left	02:57
D-Chymera1	milki: so no specific functionality for this? :(	02:57
milki	right	02:58
	git doesnt deal with files like that	02:59
	← tobscore__ left	02:59
	← Earnestly left	03:00
	← iamjarvo left	03:00
	← quazimodo left	03:01
	← dustyfields left	03:01
	← bunbury left	03:01
	→ bunbury joined	03:02
	→ Targen joined	03:02
	← donmichelangelo left	03:04
	← aldodelgado left	03:04
	→ tobscore__ joined	03:04
	→ preflex_ joined	03:05
	→ donmichelangelo joined	03:05
	← salamanderrake left	03:05
	← thiago left	03:05
	← preflex left	03:06
	→ dharrigan joined	03:06
	preflex_ → preflex	03:06
	→ Mars` joined	03:07
	← chairabanta left	03:07
	← tobscore_ left	03:07
	→ dreiss joined	03:07
	→ tobscore_ joined	03:08
	→ salamanderrake joined	03:09
	→ ingof joined	03:09
	→ dsdeiz joined	03:10
	← petey left	03:10
	← tobscore__ left	03:11
	← yhager left	03:11
	← gitpicker left	03:12
	← Mars` left	03:13
	→ koda joined	03:13
	← salamanderrake left	03:14
	→ salamanderrake joined	03:14
	→ gsvolt joined	03:14
	→ devoldmx joined	03:14
	← GODDOG left	03:15
	← Risto- left	03:15
	→ thiago joined	03:16
	→ Risto- joined	03:16
	→ mizu_no_oto joined	03:16
	← AWinterman left	03:17
	→ Guest27333 joined	03:18
	← Swayz left	03:18
	Guest27333 → diegoviola	03:19
	← Dougie187 left	03:19
	← CKyle left	03:19
	→ AWinterman joined	03:20
	← dyng left	03:22
	→ warthog9 joined	03:22
	← warthog9 left	03:22
	→ tobscore__ joined	03:24
	→ hahuang61 joined	03:24
	→ Swayz joined	03:25
	→ gitpicker joined	03:25
	→ cads joined	03:25
	← DrNick left	03:25
	← tobscore_ left	03:25
	→ breck7 joined	03:26
	→ DrNick joined	03:26
	→ tobscore_ joined	03:26
	← tobscore__ left	03:27
	← pen left	03:28
	→ pen joined	03:29
	← goshawk left	03:29
	→ pencilcheck joined	03:29
	← salamanderrake left	03:30
	← breck7 left	03:30
	← jaimef left	03:31
	← natechan left	03:32
	→ jaimef joined	03:32
	→ tobscore__ joined	03:32
	→ yhager joined	03:32
	← tobscore_ left	03:33
	← pen left	03:33
	→ mikepack joined	03:33
	→ tobscore_ joined	03:34
	← alvaro_o left	03:34
	← dreiss left	03:35
	→ cmpera joined	03:35
	← AWinterman left	03:35
	← the_cybersphinx left	03:36
	← nardev left	03:36
	← tobscore__ left	03:37
	→ tobscore__ joined	03:38
	← tobscore__ left	03:38
	→ tobscore__ joined	03:39
	← tobscore_ left	03:39
	→ relipse joined	03:39
	→ tobscore_ joined	03:40
	← ISF left	03:40
relipse	is there anything wrong with storing the mysql structure AND the data as a dump file in a git repository? it seems smart because if something goes wrong, we can just look at the data diff to see what happened	03:40
	← tobscore__ left	03:41
	← LuxuryMode left	03:42
	← trq left	03:44
	→ R1KRDO joined	03:46
	← tureba left	03:48
	← cmpera left	03:48
	← mizu_no_oto left	03:48
	→ tureba joined	03:48
	→ dreiss joined	03:48
	← arvind_khadri left	03:50
	→ tmcmahon joined	03:52
	→ arvind_khadri joined	03:52
	→ dyng joined	03:52
	← tureba left	03:53
	→ tureba joined	03:53
	← tmcmahon left	03:54
	← danmackay left	03:55
rohieb	relipse: nothing wrong with that. but your repo will blow up in size very soon	03:56
	→ tobscore__ joined	03:56
	← tobscore_ left	03:57
	→ tobscore_ joined	03:58
	← tobscore__ left	03:59
	← Gamecubic left	04:00
	← donmichelangelo left	04:00
	← dyng left	04:00
	→ tobscore__ joined	04:00
	→ donmichelangelo joined	04:00
	→ pbujok joined	04:01
	← bitkiller left	04:01
	← tobscore_ left	04:02
	→ tobscore_ joined	04:02
	← surfdue left	04:02
	← Malukenho left	04:04
	← tobscore__ left	04:06
relipse	rohieb, hmm	04:06
	→ Tommy[D] joined	04:06
	→ beautifulmind joined	04:07
	← beautifulmind left	04:07
	→ CKyle joined	04:07
	→ Mars` joined	04:08
	← Tommy[D]_ left	04:09
	← huoxito left	04:10
	← pbujok left	04:12
	← infogulch left	04:12
	→ tobscore__ joined	04:12
	→ russt joined	04:12
	← tobscore_ left	04:13
	→ tobscore_ joined	04:14
	→ surfdue joined	04:14
	← diegoviola left	04:16
	← daito left	04:16
	← fuzzybunny69y left	04:16
	← pencilcheck left	04:16
	→ fuzzybunny69y joined	04:17
	← tobscore__ left	04:17
	← lavaman left	04:19
	→ Ivo joined	04:19
relipse	How can i find big files in my repo?	04:19
	← yhager left	04:19
	→ petey joined	04:20
milki	uh	04:20
	← ingof left	04:20
milki	ls -l?	04:20
	→ ingof joined	04:21
	→ Katafalkas joined	04:23
	← ingof left	04:25
	← Ivo left	04:26
	→ beautifulmind joined	04:27
	← bunbury left	04:29
	← Risto- left	04:30
	→ girishr_ joined	04:33
	← m0viefreak left	04:34
	→ m0viefreak joined	04:34
	→ salamanderrake joined	04:34
	→ dyng joined	04:34
	← m0viefreak left	04:34
	← Pullphinger left	04:35
	← CKyle left	04:35
	← akurilin left	04:36
	→ akurilin joined	04:36
	← tjbiddle left	04:42
	→ wanders joined	04:42
	→ karstensrage joined	04:42
karstensrage	:( where might my gitolite remote be on freebsd?	04:43
	im trying to find a repo and its not where im sure it should be	04:43
	so i want to check the remote but i cant find it	04:44
	→ CKyle joined	04:44
	← russt left	04:44
	→ subhojit777 joined	04:44
	← subhojit777 left	04:45
	← malloc left	04:45
karstensrage	found it	04:45
	and the repo is not there	04:45
	→ AWinterman joined	04:45
karstensrage	i cant be imagining this	04:45
	→ malloc joined	04:47
	→ jehaby joined	04:49
	→ nadirvardar joined	04:50
	← AWinterman left	04:50
	← normanrichards left	04:50
	← mFacenet left	04:52
	→ pen joined	04:53
	← malloc left	04:53
	→ malloc joined	04:53
	→ ryxai joined	04:54
	→ rasto joined	04:54
	→ leeN joined	04:56
	→ fairuz joined	04:56
	→ jefflyne joined	04:57
	← CKyle left	04:58
	← R1KRDO left	04:59
	→ mdiers joined	05:00
	← chrisshattuck left	05:01
	→ subhojit777 joined	05:02
	← mdiers left	05:02
	← subhojit777 left	05:02
	→ subhojit777 joined	05:03
	← geomyidae left	05:03
	← cozby left	05:04
	→ jas\|work joined	05:06
	← matanya left	05:06
	← raijin left	05:06
	← wanders left	05:08
	← jon_r left	05:09
	← Swayz left	05:11
	← jas\|work left	05:11
	→ Ivo joined	05:13
	→ irqq joined	05:13
	→ jas\|work joined	05:13
	← madewokherd left	05:14
	→ Xiti joined	05:14
	→ jceb joined	05:14
	← emergion left	05:15
	→ ewnd9 joined	05:16
	← threesome left	05:18
	→ freestyl3r joined	05:18
	→ raijin joined	05:19
	← fuzzybunny69y left	05:19
	← dyng left	05:20
	← Davey left	05:21
	→ dyng joined	05:21
	→ daito joined	05:21
	→ geomyidae joined	05:23
milki	o.O	05:24
	→ simesy joined	05:24
milki	phantom repo	05:24
	← simesy left	05:25
	← geomyidae left	05:25
	→ arjun_ joined	05:25
	→ fuzzybunny69y joined	05:26
	→ whunt joined	05:26
	← arjun_ left	05:27
	→ arjun_ joined	05:28
	← arjun_ left	05:28
	→ Swayz joined	05:28
	→ Roy_SP joined	05:29
	→ arjun_ joined	05:30
	← arjun_ left	05:31
	→ arjun_ joined	05:33
	→ _iron joined	05:33
	← jperry left	05:33
	→ Caritade joined	05:34
	← arjun_ left	05:34
	→ jperry joined	05:35
	→ arjun_ joined	05:35
	← Roy_SP left	05:36
	← jxpx777 left	05:36
	← arjun_ left	05:36
	→ samd joined	05:38
	→ pantsman joined	05:38
	← pantsman left	05:38
	→ pantsman joined	05:38
	→ jxpx777 joined	05:38
	← samd left	05:38
	← girishr_ left	05:40
	→ ThomasLocke joined	05:40
	← relipse left	05:40
	→ arjun_ joined	05:41
	→ tobscore__ joined	05:42
	→ hoschi joined	05:42
	← tobscore_ left	05:42
	→ tobscore_ joined	05:44
	→ wanders joined	05:44
	→ simesy joined	05:44
	→ ciampix joined	05:44
	← tobscore__ left	05:48
	← arvind_khadri left	05:48
	→ TomyLobo joined	05:49
	→ cff joined	05:49
	→ jargon joined	05:51
	→ ianmcorvidae\|alt joined	05:51
	← ianmcorvidae\|alt left	05:51
	→ ianmcorvidae joined	05:52
	→ geomyidae joined	05:54
	← mj12albe_ left	05:56
	→ tobscore__ joined	05:56
	← jefflyne left	05:57
	← tobscore_ left	05:57
	← pantsman left	05:58
	→ ncd\|leeN joined	05:58
	→ tobscore joined	05:58
	→ mj12albert joined	05:58
	→ zacts joined	05:59
	← ics left	05:59
	← dhruvasagar left	05:59
	← tobscore__ left	05:59
	→ tobscore_ joined	06:00
	← FrodoTheHobbit left	06:00
	→ FrodoTheHobbit joined	06:00
	← leeN left	06:01
	→ dhruvasagar joined	06:01
	← nadirvardar left	06:01
	→ Xethron joined	06:02
	← tobscore left	06:03
	← vigs left	06:03
	→ tobscore__ joined	06:04
	→ papegaaij joined	06:06
	→ vigs joined	06:07
	← tobscore_ left	06:07
	→ arvind_khadri joined	06:08
	→ fragje joined	06:12
	← petey left	06:13
	← fragje left	06:14
	→ fragje joined	06:14
	→ zarubin joined	06:15
	← PaulCapestany left	06:15
	← dreiss left	06:15
	→ soee joined	06:15
	→ petey joined	06:16
	→ cbreak-work joined	06:16
	→ _nova joined	06:17
	→ tobscore_ joined	06:18
	← hahuang61 left	06:19
	→ cmpera joined	06:19
	← akurilin left	06:19
	← falconfox left	06:19
	← dyng left	06:19
	← tobscore__ left	06:20
	← koda left	06:20
	→ dyng joined	06:20
	→ tobscore joined	06:21
	← ncd\|leeN left	06:21
	→ D9_ joined	06:21
	→ falconfox joined	06:21
	← ryez left	06:23
	← tobscore_ left	06:24
	→ russt joined	06:25
	→ hahuang61 joined	06:27
	← jaequery left	06:27
	→ KillYourTV joined	06:28
	← bwreilly left	06:28
	← tpollard left	06:29
sitaram	.away	06:30
	oops	06:30
	karstensrage: gitolite puts repos inside ~/repositories, but the path component "repositories" does not appear in remote URLS	06:30
	URLs	06:30
	← Ziber_ left	06:32
	→ clone1018_ joined	06:32
	→ ayaz joined	06:32
	→ jcp joined	06:33
	← georgl left	06:33
	→ georgl_ joined	06:33
	← javawizard left	06:33
	← D-Chymera1 left	06:33
	→ Ziber joined	06:34
	← pen left	06:34
	← TomyLobo left	06:34
	← Swayz left	06:35
	→ Ozch joined	06:35
	← pepper_chico left	06:35
	→ FatDarrel joined	06:38
	→ druonysuse joined	06:39
	← druonysuse left	06:39
	→ druonysuse joined	06:39
	← georgl_ left	06:41
	→ netiul joined	06:43
	← mishok13 left	06:44
	← jas\|work left	06:44
	← minim left	06:45
	→ rudak joined	06:45
	← Caritade left	06:45
	→ georgl joined	06:45
	→ jas\|work joined	06:45
	→ ingof joined	06:46
	→ matanya joined	06:48
	→ tobscore_ joined	06:50
	→ Misan joined	06:50
	→ D-Chymera1 joined	06:50
	→ marthinal joined	06:51
	← bket left	06:51
	← matanya left	06:51
	← tobscore left	06:52
	→ tobscore__ joined	06:52
	→ matanya joined	06:53
	← echo-area left	06:55
	→ terravires joined	06:55
	← tobscore_ left	06:55
	← yacks left	06:55
	→ dsdeiz_ joined	06:56
	← dsdeiz left	06:56
	→ docwhat joined	06:56
	gitinfo set mode: +v	06:57
terravires	hi all, I'm using git 1.8.1 on windows, with git-gtk. I had a file that I renamed foo.cs -> bar.cs and git showed the file was deleted. But now it's ignoring anything to do with bar.cs as if it didn't exist? Any advice?	06:57
_ikke_	terravires: Is bar.cs ignored?	06:57
terravires	_ikke_ no	06:57
	→ zendeavr joined	06:57
_ikke_	what does git status show?	06:57
	→ harobed joined	06:58
terravires	noting to commit, working directory clean	06:58
	→ tatsuya_o joined	06:59
_ikke_	And if you edit bar.cs?	06:59
	← raijin left	06:59
terravires	same	06:59
	it totally ignores the file and any chances	06:59
	It used to be called foo.cs, and was fine, I renamed it outside of git and when I went to commit it said the file was deleted	06:59
	← boombatower left	07:00
terravires	I expected it to either catch the rename or at the least add a new bar.cs	07:00
	→ RaceCondition joined	07:00
	→ ciaranm joined	07:00
_ikke_	git ls-files \| grep bar.cs	07:00
terravires	no output	07:00
_ikke_	Then it's not tracked	07:01
terravires	I know, like I said git appears to be ignoring this file for some reason	07:01
	it's tracking other files in the same directory	07:01
_ikke_	git ls-files -i --exclude-standard \| grep bar.cs	07:02
	→ kbart joined	07:02
osse	does git add -f bar.cs work ?	07:02
jast	did you do 'git add bar.cs' at some point?	07:02
	→ tobscore_ joined	07:02
terravires	not listed ikke	07:02
	I was using git-gtk, so most of that was done for me. :-/	07:03
	← cmpera left	07:03
	← tobscore__ left	07:03
	→ canton7 joined	07:03
	→ echo-area joined	07:03
	→ shuaib joined	07:04
	← Goplat left	07:04
	→ tobscore__ joined	07:04
_ikke_	git ls-files -io --exclude-standard \| grep bar.cs	07:04
	← RaceCondition left	07:04
	← jehaby left	07:04
	← s0undt3ch left	07:04
terravires	ok, so git bash + git add tells me the file is in fact being ignored	07:04
	→ ianclarksmith joined	07:04
	→ emergion joined	07:04
_ikke_	:-)	07:04
terravires	but I'm kind of shocked as why... it used to work.	07:05
	→ s0undt3ch joined	07:05
	← kfunk_ left	07:05
_ikke_	What used to work?	07:05
terravires	is it possible to debug .gitignore?	07:05
	→ jon_r joined	07:05
	ianclarksmith → ics	07:06
terravires	well the file was added and being tracked, and then I renamed the file (foo.cs to bar.cs) and then git now ignores it	07:06
osse	Actually, Git 1.8.4 as a check-ignore command	07:06
	→ simesy_ joined	07:06
_ikke_	terravires: git ignores files based on name	07:07
	← tobscore_ left	07:07
terravires	http://pastebin.com/hxsAc0Vk	07:07
_ikke_	so if a file gets a name that is being ignored, git just happely ignoes it	07:07
jast	what's the path of bar.cs, relative to repo root?	07:08
terravires	I'm trying to ignore everything but Assets/_* (and any sub dirs/files)	07:08
	Assets/_Scripts/UI/HUD	07:08
jast	then try this:	07:08
	→ mishok13 joined	07:09
jast	http://pastebin.com/GJAGzVN6	07:09
	→ lb1a joined	07:09
	→ jehaby joined	07:09
	← simesy left	07:09
jast	oh, hang on, they changed the rules, didn't they	07:09
	should work anyway	07:10
	← zarubin left	07:10
	→ robbyoconnor joined	07:11
terravires	do I need to commit the .gitignore before it works?	07:11
moritz	no	07:12
	→ rath joined	07:12
	→ h_kon joined	07:12
	→ dreiss joined	07:12
terravires	still says it is ignored	07:13
	→ RaceCondition joined	07:13
terravires	I just need to get all files/directory that start with _ in Assets/, but ignore everything else.	07:14
iveqy	terravires: do you've anything in .git/info/excludes?	07:14
	← _Norm_ left	07:14
terravires	I didn't change any of my ignores so I don't know why this stopped working... it was tracked	07:14
jast	oh, you want files in Assets, too	07:15
iveqy	terravires: how did you rename the file? And if you track foo.cs and then move it to bar.cs (not with git mv) bar.cs isn't tracked	07:15
terravires	iveqy, nope, all empty	07:15
jast	in that case it has to be !/Assets/_* instead of !/Assets/_*/	07:15
	→ dsdeiz joined	07:15
terravires	I did rename it outside of git, but had to because of my other tools. I figured git would best case pick up the new filename, worst case just add the new file	07:16
	→ tobscore_ joined	07:16
osse	renaming outside of git is no problem. git will detect the delition and if you git add the new name it'll be fine	07:17
_ikke_	terravires: Yes, that's normally not a problem, but git has been told to ignore that file, and it does	07:17
osse	well, that's how it's supposed to be at least	07:17
	← dsdeiz_ left	07:17
	→ pen joined	07:18
osse	but if git ls-files -i --exclude-standard \| grep bar.cs is empty...	07:18
_ikke_	osse: it might be that -io is needed	07:18
	← irqq left	07:19
_ikke_	if -i doesn't cancel -c (cached) it'll only list files that would be ignored but are tracked	07:19
	← tobscore__ left	07:19
	← simesy_ left	07:19
	→ kewah joined	07:19
	→ tobscore__ joined	07:20
	→ tagava joined	07:20
	→ simesy joined	07:20
	→ yezariaely joined	07:20
	→ mitsuhiko joined	07:20
	← tobscore_ left	07:20
	→ alorence joined	07:21
	← matanya left	07:21
osse	you're right	07:21
	→ yacks joined	07:22
terravires	the problem seems to be recursive directories. So I have to do an inverse ignore... ignore , then !_/ and !_/ but then it doesn't got deeper	07:22
	→ koda joined	07:22
terravires	how should I tell it !_* (AND all subdirectories/files) ?	07:23
	← ingof left	07:23
	→ ingof joined	07:23
	← Mars` left	07:23
_ikke_	perhaps !ignore_whitedir helps	07:23
gitinfo	[!gitignore_whitedir] A global/directory .gitignore blacklist with a (potentially) subdirectory whitelist is not easy to specify in git. However, something like `echo -e '/\n!/a/\n/a/\n!/a/b/\n/a/b/\n!/a/b/c/' > .gitignore` (ignore everything but a/b/c directory) or `echo -e '\n!/\n!.c' > .gitignore` (ignore everything but *.txt files) may do what you want.	07:24
terravires	so I can't just whitelist _* files/directories?	07:26
	→ Cerise joined	07:26
terravires	does git support regexp in .gitignore?	07:27
_ikke_	no	07:27
iveqy	terravires: man !gitignore , it support shell	07:27
gitinfo	terravires: [!gitignore_whitedir] A global/directory .gitignore blacklist with a (potentially) subdirectory whitelist is not easy to specify in git. However, something like `echo -e '/\n!/a/\n/a/\n!/a/b/\n/a/b/\n!/a/b/c/' > .gitignore` (ignore everything but a/b/c directory) or `echo -e '\n!/\n!.c' > .gitignore` (ignore everything but *.txt files) may do what you want.	07:27
	← thiago left	07:27
_ikke_	It uses fnmatch	07:27
	← fuzzybunny69y left	07:28
	← jas\|work left	07:28
iveqy	bad gitinfo =(	07:28
	or bad me maybe...	07:28
	← ingof left	07:28
_ikke_	yeah, you prefixed a !	07:28
	→ ngladitz joined	07:29
terravires	sorry guys, but I'm just not seeing how I can do this without whitlisting all my _* directories and files individually	07:30
	→ gua_ould joined	07:30
terravires	I've tried adding all of these: !_, !_/, !_/, !_//, but it always seems to stop at the last depth I manually add	07:31
osse	!_/*	07:32
	As of Git 1.8 I think	07:32
	← tatsuya_o left	07:32
	← akiva left	07:33
	→ psoo joined	07:33
	→ zarubin joined	07:33
terravires	osse, didn't work, stopped at one deep	07:33
	→ fuzzybunny69y joined	07:33
osse	terravires: with two asterisk, right ?	07:35
terravires	osse, yes	07:35
	← dyng left	07:35
terravires	\n!_\n!_/*	07:35
	← beautifulmind left	07:35
	← russt left	07:36
	→ tobscore_ joined	07:36
	→ dyng joined	07:36
	← rchavik left	07:37
	← tobscore__ left	07:37
	← woollyams left	07:37
	→ tobscore__ joined	07:38
	→ GeertV joined	07:38
	→ Asenar joined	07:39
	← jon_r left	07:40
	→ Squarism joined	07:41
	→ nova_ joined	07:41
	← _nova left	07:41
	→ OOPMan joined	07:41
	← tobscore_ left	07:41
	→ rchavik joined	07:42
	← rchavik left	07:42
	→ rchavik joined	07:42
	→ maleknet joined	07:42
	→ GeertV_ joined	07:42
	← GeertV left	07:42
	→ Anticom joined	07:42
	→ tvw joined	07:43
	← cads left	07:44
	→ tobscore_ joined	07:44
terravires	oh well, I'll keep trying, thanks for the suggestions guys.	07:45
	→ quazimodo joined	07:45
	← terravires left	07:45
	→ Martz joined	07:45
	← Martz left	07:45
	→ Martz joined	07:45
	← Ozch left	07:45
	→ threesome joined	07:46
	← rue left	07:47
	← tobscore__ left	07:47
	→ jnavila joined	07:48
	← nfroidure left	07:48
	← fuzzybunny69y left	07:48
	→ SonicX joined	07:49
	→ nfroidure joined	07:49
	← nfroidure left	07:49
	→ nfroidure joined	07:49
	→ tobscore__ joined	07:50
	→ buibex joined	07:50
	← tobscore_ left	07:51
	→ g41n joined	07:51
	→ Kabaka joined	07:51
	→ rue joined	07:51
	→ konadr joined	07:51
	→ Amandil joined	07:52
	→ tobscore_ joined	07:52
	→ matanya joined	07:52
	← zendeavr left	07:53
	→ zendeavor joined	07:53
	← tobscore__ left	07:53
	→ tobscore__ joined	07:54
PaulePanter	In coreboot we need to store patch/diff files, which have whitespace error by design.	07:54
	Is there a way to exclude files from `git diff --check`?	07:54
	http://review.coreboot.org/#/c/3340/1/util/lint/lint-stable-003-whitespace	07:54
	→ jefflyne joined	07:55
	← tobscore_ left	07:55
	→ nikolai_ joined	07:55
	→ tobscore_ joined	07:56
	← druonysuse left	07:56
	→ druonysuse joined	07:57
	← Martz left	07:57
	← druonysuse left	07:57
	→ druonysuse joined	07:57
nikolai_	hi there! When using gitk, creating a new view, and setting the Author field to my name, I see also other commits that do not even mention my name...it seems to be different to "git log --author=Foo" ...how come?	07:57
	→ atomC0de joined	07:57
	→ dfanjul joined	07:58
	← vigs left	07:58
	← tobscore__ left	07:59
	→ dkpi joined	08:00
	→ hachque_ joined	08:01
	→ mikecmpbll joined	08:01
	→ tobscore__ joined	08:02
	→ fuzzybunny69y joined	08:04
	→ dsdeiz_ joined	08:05
	← tobscore_ left	08:05
	← dsdeiz left	08:05
	→ renumi joined	08:06
	← dreiss left	08:08
	← ncopa left	08:08
	→ frobware joined	08:10
	→ Octalot joined	08:10
	← whunt left	08:12
	→ tobscore_ joined	08:12
	← FatDarrel left	08:12
	→ yuppie joined	08:14
	→ whunt joined	08:14
	← tobscore__ left	08:15
	← tobscore_ left	08:17
	→ tobscore_ joined	08:17
	→ intellix joined	08:19
	→ sq-one joined	08:20
	→ kwmiebach joined	08:22
	← Amandil left	08:22
	→ Shiva joined	08:23
	→ jon_r joined	08:23
	→ rafeca joined	08:24
	→ dsdeiz joined	08:24
	→ tobscore__ joined	08:24
	→ marhaban joined	08:24
	→ aChatir joined	08:24
	← dsdeiz_ left	08:25
	← whunt left	08:25
	← staafl left	08:26
	← maleknet left	08:26
	→ whunt joined	08:26
	← tobscore_ left	08:27
	← jon_r left	08:28
	→ tobscore_ joined	08:28
	→ Amandil joined	08:28
	metze_ → metze	08:29
	→ menace joined	08:29
	→ maleknet joined	08:29
	→ jon_r joined	08:29
	← tobscore__ left	08:30
	→ tobscore__ joined	08:30
	→ the_codebot joined	08:30
	← simesy left	08:31
	← tobscore_ left	08:31
	← aChatir left	08:31
	→ tobscore_ joined	08:32
theForce	lets say there are multiple commits which have all a same token in the commit message (e.g. a ticketnumber). what is the best way to review those commits all at once?	08:33
_ikke_	theForce: git log --all --grep "token"	08:33
iveqy	theForce: man git log see --grep	08:34
gitinfo	theForce: the git-log manpage is available at http://jk.gs/git-log.html	08:34
	→ Mars` joined	08:34
	→ simesy joined	08:35
	← tobscore__ left	08:35
theForce	_ikke_: yeah i found that too but that way i only see the commit message, author etc. i actually would like to see all the code changes in those commits merged together, for instance as a patch	08:35
	← CodeShark left	08:35
	← the_codebot left	08:36
	→ zbcm joined	08:37
	→ matteo joined	08:37
_ikke_	Hmmm, if only git log had something like --patch	08:37
	;-)	08:37
	→ noname001__ joined	08:37
	← simesy left	08:37
	← Mars` left	08:38
	← zbcm left	08:39
iveqy	theForce: git log --all --grep "token" \| grep "^commit " \| cut -c7-47	08:39
	→ jan1337z joined	08:39
iveqy	you could even to:	08:39
	theForce: git log --all --grep "token" \| grep "^commit " \| cut -c7-47 \| git show	08:39
_ikke_	iveqy: Or just git log --all --grep "token" -p	08:40
iveqy	nah =(, it just takes the first... you need to loop it	08:40
	_ikke_: haha well, that would be too easy ;)	08:40
	→ howlymowly joined	08:40
	← kcj left	08:42
iveqy	so, that's a list of patches, they're still not merged together	08:42
_ikke_	If you want something more complex: git rev-list --all --grep "token" \| xargs -n1 git show	08:42
iveqy	I would solve this by just applying them to a branch	08:42
_ikke_	yes	08:42
	→ tobscore__ joined	08:42
	← tobscore_ left	08:43
	→ jas\|work joined	08:43
	→ tobscore_ joined	08:44
	← petey left	08:44
	→ GodEater joined	08:44
	← GodEater left	08:44
	→ GodEater joined	08:44
	→ tatsuya_o joined	08:44
	← jas\|work left	08:45
	← fragje left	08:45
	→ jas\|work joined	08:46
	← Amandil left	08:46
osse	iveqy: what do you mean by "just takes the first" ?	08:47
	← tobscore__ left	08:47
iveqy	osse: I just get the the first sha1 showed	08:47
	→ Streward joined	08:48
	→ tobscore__ joined	08:48
theForce	thx guys	08:49
	→ Norm_ joined	08:49
	← menace left	08:49
iveqy	theForce: however IMHO it seems like you're doing something wrong. Shouldn't all commits that have something incommon be on a branch?	08:49
	← tobscore_ left	08:51
theForce	iveqy: well this branching workflow is not really established here currently. hopefully in the future	08:52
	← deryni left	08:52
	→ tobscore_ joined	08:54
	← OOPMan left	08:54
	→ achadwick joined	08:54
	→ valgrind joined	08:56
	→ juril joined	08:57
	← tobscore__ left	08:57
	yuppie → realeyes	08:58
	→ tobscore__ joined	08:58
	→ GODDOG joined	08:58
	← Rh0nda left	08:58
	← jefflyne left	08:59
	← netiul left	08:59
	← tobscore_ left	09:00
	→ aChatir joined	09:00
	→ tobscore_ joined	09:00
	→ pfFredd joined	09:00
	→ Martz joined	09:01
	→ netiul joined	09:01
	← quazimodo left	09:01
	← buibex left	09:02
	← maleknet left	09:02
	← tobscore__ left	09:03
	→ buibex joined	09:04
	← ph88 left	09:04
	← ketas left	09:05
	← Martz left	09:05
	→ vince_prignano joined	09:07
	→ Martz joined	09:07
	→ ketas joined	09:07
	→ tobscore__ joined	09:08
	→ the_codebot joined	09:08
	← hoschi left	09:10
	← tobscore_ left	09:11
	→ tobscore_ joined	09:12
	→ OOPMan joined	09:12
	→ sharpshooter joined	09:12
	← ewnd9 left	09:13
	← tobscore__ left	09:15
	← jophish left	09:16
	→ hoschi joined	09:16
	→ dbuch joined	09:16
	→ tobscore__ joined	09:16
	→ ewnd9 joined	09:16
	← tobscore_ left	09:17
	← Martz left	09:18
	← ayaz left	09:21
	← marhaban left	09:21
	→ tobscore_ joined	09:22
	→ deryni joined	09:22
	→ Martz joined	09:23
	← Martz left	09:23
	→ EvilDMP joined	09:23
	→ Martz joined	09:23
EvilDMP	if I rewrite history of a branch when pushing to a repository, what happens to other branches on the repository that may have shared some of that history?	09:24
iveqy	EvilDMP: nothig	09:25
EvilDMP	iveqy: thanks	09:25
	← tobscore__ left	09:25
	→ mindriot101 joined	09:26
	→ kcj joined	09:26
	→ darnir joined	09:27
	← Streward left	09:27
	→ cads joined	09:28
	← dkpi left	09:29
	→ tobscore__ joined	09:30
jds	"assume-unchanged" sets things up to completely ignore the referenced file, even if it's later changed. Is there a way of setting things up so that the file in its current state is ignored, but any future changes to it get shown?	09:30
	← Martz left	09:31
	← the_codebot left	09:31
_ikke_	nope	09:31
	except for something like inotify that does that	09:32
	→ marhaban joined	09:32
	→ Streward joined	09:33
	← tobscore_ left	09:33
	→ tobscore_ joined	09:34
	→ asQuirreL joined	09:35
	← darnir left	09:35
	→ darnir joined	09:35
	→ ayaz joined	09:35
	← Georg3 left	09:36
	→ VictorCL joined	09:36
	← tobscore__ left	09:37
	→ tobscore__ joined	09:38
	← tobscore_ left	09:39
jast	jds: how would git be able to tell that it was changed except by checking whether it was changed? :}	09:39
	→ tobscore_ joined	09:40
	← harish left	09:40
_ikke_	It could check if it was changed but not report on it	09:40
	← Juri_Agarin left	09:40
	→ Juri_Agarin joined	09:40
jds	jast: I wasn't sure how assume-unchanged was implemented. Thought it maybe had a private staging area akin to the regular staging area where the file changes were recorded	09:41
iveqy	the solution here is to have some sort of api for other programs to tell git that a file is changed (like an inotify server or an IDE)	09:41
	→ introom joined	09:42
	→ LimeBlast joined	09:42
	← subhojit777 left	09:42
	← asQuirreL left	09:43
	← hyperair left	09:43
jast	jds: it does not. it simply makes a note not to check whether the file was changed, and to pretend that it wasn't	09:43
	(by the way, there are several operations that silently unset the assume-unchanged flag)	09:44
	← tobscore__ left	09:44
	→ stamina joined	09:45
	→ tobscore__ joined	09:46
	← rndbit` left	09:48
	→ rndbit joined	09:48
	← Juri_Agarin left	09:48
	→ Stygia joined	09:49
	→ saimaz joined	09:49
	← tobscore_ left	09:49
	→ tobscore_ joined	09:50
bonhoeffer	sorry, just learning git -- i made a bunch of changes that I hate -- i want to go back to a specific commit -- but might want to go back to "some" of my changes --	09:50
	should i revert to the commit where things were good (if so, what happens to the old changes?)	09:51
	← pen left	09:52
	← zero7 left	09:52
	← akitada left	09:52
iveqy	bonhoeffer: !reset you want e), your changes will be lost, but if you save the sha1 of the commits you need you can cherry-pick them for another 2 weeks	09:52
gitinfo	bonhoeffer: A good resource explaining git-reset is http://git-scm.com/2011/07/11/reset.html	09:52
	→ DoniRS joined	09:52
	← xerxas left	09:52
	← ELLIOTTCABLE left	09:53
	← jmcantrell left	09:53
	← tobscore__ left	09:53
	← robstanley left	09:53
	← heidar left	09:54
	← jab416171 left	09:54
	→ dsdeiz_ joined	09:57
	← dsdeiz left	09:57
	→ tobscore__ joined	09:58
	→ Cymew joined	09:59
	← alexw left	09:59
	→ PeterBooker joined	09:59
	← tobscore_ left	09:59
	→ tobscore_ joined	10:00
	← Shiva left	10:00
	→ subhojit777 joined	10:01
	← mikecmpbll left	10:01
Cymew	Hi. I'm trying to clone over http, and get the famous "/info/refs not found: did you run git update-server-info on the server?" I have indeed done that, and ls -l proves that the info/refs file has been accessed. My http log say "/var/lib/git/nagios-se-plugins.git/info/refs?service=git-upload-pack HTTP/1.1" 404 332 "-" "git/1.7.1"" which does indicate that it failed. Any suggestions?	10:01
	→ mikecmpbll joined	10:01
cbreak-work	Cymew: sounds likeyou use dumb http	10:02
	Cymew: run update-server-info	10:02
	on the server	10:03
	← tobscore__ left	10:03
	→ hseg joined	10:03
Cymew	Again? Did it and still no cookie. :(	10:03
cbreak-work	hmm...	10:04
	wonder if they finally removed dumb http support from git	10:04
Cymew	Really?	10:05
hseg	Is there a way to move an edit to an old commit where it belongs? i.e. I divided a patch into several commits and I didn't notice that I forgot about one line until a couple of commits later.	10:05
cbreak-work	hseg: not without history rewriting	10:05
_ikke_	cbreak-work: version 1.7.1, I don't think it's removed from that version	10:05
hseg	History is unpublished.	10:05
cbreak-work	hseg: which is problematic if you already pushed	10:05
hseg	Unpushed.	10:05
cbreak-work	hseg: then just git rebase -i someparent and edit the first commit	10:06
hseg	BTW, I guess this implies that anything pushed is set in stone, right?	10:06
Cymew	It would totally ruin all our setup if they did remove it	10:06
cbreak-work	no. It implies that it's a pain to change published history	10:06
	it's possible though	10:06
Cymew	Sure would be nice to at least have the option...	10:06
cbreak-work	Cymew: dumb http is really worthless	10:06
	there are lots of alternatives	10:06
	← valgrind left	10:06
hseg	So git rebase -i commit_foo^ and edit the diff?	10:07
cbreak-work	like ssh://, git:// or even smart http	10:07
	hseg: not edit the diff	10:07
	just git commit --amend	10:07
	then rebase --continue	10:07
hseg	OK.	10:07
cbreak-work	you know commit --amend?	10:07
	← johnlcox left	10:07
hseg	I think so. Doesn't it reopen the commit for changes?	10:07
Cymew	cbreak-work: ssh doesn work for me. I need unauthenticated checkout. Any good pointer for what smart http means?	10:08
cbreak-work	Cymew: you need to install some scripts on the server.	10:08
	It's much more efficient (and faster) than dumb http.	10:08
	I don't know any details	10:08
	hseg: it just commits a changed version of the current commit in its place	10:08
Cymew	I have no speed problems, but it sounds like I need to look it up. Thanks for the suggestions.	10:09
cbreak-work	hseg: basically, change something, git add it, then commit --amend, and that change will be part of the new version of the current commit	10:09
	→ wrouesnel1 joined	10:09
jast	man git-http-backend	10:09
gitinfo	the git-http-backend manpage is available at http://jk.gs/git-http-backend.html	10:09
Cymew	thx	10:09
hseg	OK. So commit --amend merges staging and the previous commit?	10:09
jast	sort of	10:10
	→ tobscore__ joined	10:10
	→ aandy joined	10:10
jast	it recreates the commit but uses the current index instead of what was in the original commit	10:10
	← SirCmpwn left	10:10
	→ mtk0 joined	10:11
	→ SirCmpwn joined	10:11
	← tobscore_ left	10:12
hseg	So it replaces the current commit by the current code in staging with the old commit message.	10:12
jast	yeah (and you can change the commit message)	10:12
hseg	Nice.	10:13
jast	so, some people use --amend just for changing the commit message	10:13
	→ tobscore joined	10:13
hseg	One thing is annoying - once pushed, your history, along with your stupid commits, is a pain to fix.	10:14
	← tobscore__ left	10:15
	→ tobscore_ joined	10:16
	→ dsdeiz joined	10:16
	← tobscore left	10:17
	→ beautifulmind joined	10:18
hseg	How can I see at which commit origin/master is?	10:18
	← gua_ould left	10:18
	→ tobscore__ joined	10:18
Asenar	git log --all	10:18
	← zarubin left	10:18
	← tobscore_ left	10:19
	← dsdeiz_ left	10:19
hseg	nm, git log repo/branch works perfectly	10:19
	madduck_ → madduck	10:19
	← ghoti left	10:20
	→ ghoti joined	10:20
cbreak-work	hseg: !lol	10:20
gitinfo	hseg: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	10:20
	← emergion left	10:21
	→ jeffreyio2 joined	10:22
	→ tobscore_ joined	10:22
	→ jkremser joined	10:23
	← jkremser left	10:23
	→ jkremser joined	10:23
	→ jfoy joined	10:23
hseg	Wow. That one's going into my .bashrc	10:23
	← VictorCL left	10:24
	→ khushildep joined	10:24
	→ VictorCL joined	10:24
Cymew	Hmm. This smart http thing sure became complicated. Now I need to find an apache channel to ask questions on as well. Oh well.	10:25
	→ harish joined	10:26
	← tobscore__ left	10:26
cbreak-work	hseg: how about a git alias? :)	10:26
hseg	What are those?	10:27
	→ tobscore joined	10:27
	← tobscore left	10:27
	→ tobscore joined	10:27
	← jfoy left	10:27
	← jeffreyio2 left	10:28
	→ Ozch joined	10:29
	← tobscore_ left	10:29
_ikke_	git config alias.lol 'log --oneline --graph --decorate --all'	10:30
	← marhaban left	10:30
	→ gua_ould joined	10:30
_ikke_	then git lol would get you that	10:30
	→ the_cybersphinx joined	10:30
	→ jeffreyio2 joined	10:31
hseg	OK. Why the name lol?	10:31
_ikke_	I guess Log OneLine, but I'm not sure	10:32
hseg	I'd call it bird or bev	10:33
_ikke_	You can give it the name you want, but it can't be the name of built-in command	10:33
	→ malukenho joined	10:33
	→ gratz joined	10:34
gratz	hi	10:34
gitinfo	gratz: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	10:34
	→ simesy joined	10:34
	← frgtn left	10:35
	← vince_prignano left	10:35
	→ jason237 joined	10:36
	→ Stalkr^ joined	10:36
gratz	i have this branch history: http://postimg.org/image/dh6cev3gj/ but where dev stops it should have been merged in (master continued from the dev branch at that point in time), how would i do this? I'd like to then get dev to the same point as master to continue working on it..	10:36
	← Norm_ left	10:37
_ikke_	gratz: is this public history (ie, has it been pushed)?	10:37
	→ xerxas joined	10:37
_ikke_	There are tags on master, so I assume it is	10:37
gratz	it's been pushed but the repo isn't public, only i am working on it at the moment	10:38
	→ Norm_ joined	10:38
	← dpwrussell left	10:38
	→ dpwrussell joined	10:38
	← suy_ left	10:39
	→ suy joined	10:39
	← fuzzybunny69y left	10:40
	← jason237 left	10:41
_ikke_	You have to rewrite history if you want it to be merged at the point where the branch stops	10:41
	→ ChrisAnn joined	10:41
_ikke_	It also means you have to recreate the tags	10:41
	← Milossh left	10:43
	← cads left	10:43
	→ shoshin joined	10:43
	→ emergion joined	10:45
hseg	How do I use a tag instead of the hash corresponding to that tag? Specifically, how do I git log since a specific tag?	10:45
_ikke_	hseg: You can just use the tag	10:46
	→ tobscore_ joined	10:46
	← Spyro left	10:47
	← Misan left	10:47
	→ Spyro joined	10:48
	← arjun_ left	10:48
	→ Milossh joined	10:48
hseg	Resolved. Didn't know revision range syntax, which caused problems.	10:48
gratz	_ikke_, if at the point where dev stops, the commit on master immediately after it is exactly the same as dev at that point do the tags still have to be recreated?	10:48
	also, when you say rewrite history, would this be a merge rebase?	10:49
	→ simesy_ joined	10:49
	← aChatir left	10:49
	← tobscore left	10:49
	→ tobscore__ joined	10:50
	→ fridim_ joined	10:50
_ikke_	a merge rebase is not really a term in git	10:50
	← gua_ould left	10:51
	← dbuch left	10:51
_ikke_	But yes, because every commit that is following it still has to change	10:51
	← simesy left	10:51
	→ dbuch joined	10:52
	← netsroht left	10:52
gratz	sounds painful	10:52
	if i didn't do that, can i just ff dev to the same state as the head of master somehow?	10:53
	← tobscore_ left	10:53
	← PeterBooker left	10:54
_ikke_	You need to merge both branches	10:54
	→ PeterBooker joined	10:54
	→ tobscore_ joined	10:54
hseg	Can someone please explain the warning I got when I tried to git add a file I was previously tracking which I'd just deleted?	10:55
	i.e. http://pastie.org/8371842	10:55
	→ Martz joined	10:55
	← saimaz left	10:56
j416	hseg: 'git add' will add the deletion by default starting in Git 2.0, but currently the default is to not do so	10:56
hseg	Why the change and why the old behaviour?	10:57
j416	mailing list should give you the exact reasons	10:57
	I speculate that it is more intuitive	10:57
	→ netsroht joined	10:57
	← tobscore__ left	10:57
j416	old behaviour is not so intuitive, perhaps	10:57
	→ Chiko joined	10:58
	→ tobscore__ joined	10:58
hseg	Clash of intuitions. Does git add mean "add this change" or "track this new file".	10:58
	→ jason237 joined	10:58
Hello71	both.	10:58
	← tobscore_ left	10:59
	← bonhoeffer left	10:59
	← paulweb515 left	11:00
	← fairuz left	11:00
	← ciampix left	11:00
	→ jefflyne joined	11:01
	→ cmpera joined	11:01
	→ maleknet joined	11:02
	lb1a → helloworld_	11:02
	← kcj left	11:02
	helloworld_ → lb1a	11:02
	→ Misan joined	11:02
	← Xethron left	11:03
Squarism	what dist of git should you run on windows?	11:04
	→ gua_ould joined	11:05
Squarism	...for git beginners	11:05
j416	Squarism: I used to run the version available from http://git-scm.com/ called git-bash, it worked well under the circumstances	11:05
Hello71	msysgit	11:05
j416	(not sure if that is actuall the name)	11:05
	Hello71: yes, maybe that is the correct name. But it comes bundled with bash.	11:05
Hello71	that's like calling coreutils "sed"	11:06
j416	heh	11:06
	→ Balliad joined	11:07
j416	it comes with core utils and it has something called "git bash", which opens a bash prompt	11:07
bremner	well, sed _is_ really all you need	11:07
j416	anyway, that's the version I've been most successful running, Squarism	11:07
Squarism	j416, thanks.. looks most OFFICIAL to me	11:08
	ill try that	11:08
	← fridim_ left	11:08
	→ ryez joined	11:08
	→ koptein joined	11:08
	→ Sonderblade joined	11:09
	← Sonderblade left	11:09
	→ Sonderblade joined	11:10
	→ Xethron joined	11:10
	gitinfo set mode: +v	11:11
	→ _nova joined	11:14
	← darnir left	11:15
	← conan left	11:15
	← nova_ left	11:17
	→ OOPMan1 joined	11:17
	← OOPMan left	11:19
	← Squarism left	11:20
	→ ciampix joined	11:20
	→ arjun_ joined	11:21
	→ OnO joined	11:21
	→ alexandrerussel joined	11:22
	← simesy_ left	11:24
	→ Earnestly joined	11:27
	← jon_r left	11:28
	→ Caritade joined	11:28
	→ Guest17847 joined	11:28
	→ kcj joined	11:28
	→ frgtn joined	11:29
	← GeertV_ left	11:29
	← frgtn left	11:30
	→ frgtn joined	11:30
	→ GeertV joined	11:30
	← SonicX left	11:30
	← thierryp left	11:32
	← introom left	11:33
	→ normanrichards joined	11:35
	→ starfox21 joined	11:35
	← daito left	11:36
	← cmpera left	11:37
	← OnO left	11:37
	→ ghoti_ joined	11:39
	→ leandrosansilva joined	11:45
	→ erichynds joined	11:45
	→ jazz264 joined	11:46
	← Misan left	11:46
	→ vurma joined	11:48
	← jas\|work left	11:48
	→ SonicX joined	11:49
	→ cmpera joined	11:49
	← stamina left	11:51
	← PeterBooker left	11:54
	→ PeterBooker joined	11:54
	← boblee left	11:55
	← SonicX left	12:00
	→ pbujok joined	12:00
	← jefflyne left	12:01
	← mikecmpbll left	12:01
	→ Squarism joined	12:02
	← RaceCondition left	12:02
	← buibex left	12:02
	→ angelsl joined	12:02
	← Boohbah left	12:04
	→ buibex joined	12:04
	→ cannonball joined	12:04
	→ mribeirodantas joined	12:05
	← pbujok left	12:05
	→ Boohbah joined	12:07
	→ johnlcox joined	12:08
	lb1a → is	12:09
	is → rt	12:09
	rt → fo	12:09
	fo → lb1a	12:09
	→ geekbri joined	12:10
	← PeterBooker left	12:12
	→ mattp_ joined	12:13
	→ heidar joined	12:13
	← emergion left	12:13
	→ jab416171 joined	12:14
	→ zero7 joined	12:14
	→ ELLIOTTCABLE joined	12:14
	→ akitada joined	12:15
	← erichynds left	12:15
	→ tdm00 joined	12:16
	→ PeterBooker joined	12:17
	← EvilDMP left	12:17
	→ robstanley joined	12:18
	→ jmcantrell joined	12:19
	← tdm00 left	12:19
	→ CheckDavid joined	12:20
	→ sriracha joined	12:21
	← normanrichards left	12:21
	← Novacha_ left	12:23
	→ Novacha joined	12:24
	→ wakd joined	12:24
	→ RaceCondition joined	12:24
	← RaceCondition left	12:25
	→ Juri_Agarin joined	12:26
	→ paulweb515 joined	12:27
	← Chiko left	12:28
	→ mw44118 joined	12:28
	→ Misan joined	12:28
	← tureba left	12:30
	← cmpera left	12:30
	→ tureba joined	12:30
	→ cmpera joined	12:31
	← paulweb515 left	12:33
	→ decotii joined	12:35
	→ Droolio joined	12:35
	← salamanderrake left	12:38
	→ TheBrainWork joined	12:38
	← Nickeeh_ left	12:39
	→ Nickeeh joined	12:40
	← nfroidure left	12:40
	→ simesy joined	12:40
	→ spaceonline joined	12:40
	← tureba left	12:41
	→ nfroidure joined	12:41
	→ tureba joined	12:41
	← devoldmx left	12:42
j416	is there an opposite of 'git add -p' that allows me to patch by patch unstage changes?	12:42
	I wish to add everything but a small change	12:42
_ikke_	git reset -p	12:43
j416	cool, thanks	12:43
	quite obvious. heh.	12:43
	← hoschi left	12:43
	→ devoldmx joined	12:44
j416	bah I can't edit the hunks	12:44
	oh well	12:44
	← Roots47 left	12:44
_ikke_	cna't	12:45
	can't?	12:45
	→ mikecmpbll joined	12:45
	← dyng left	12:45
	→ m8 joined	12:46
j416	only for files that are already tracked	12:46
	but not for new files (which is why I couldn't git add -p them in the first place)	12:46
	← sq-one left	12:46
	→ Guest33959 joined	12:46
j416	solved it by temporarily removing the parts that I didn't want to add	12:46
	→ stamina joined	12:46
j416	physically from the files	12:46
	→ dsdeiz_ joined	12:47
	→ mizu_no_oto joined	12:47
	← dsdeiz left	12:47
	→ breakingthings joined	12:48
	← chronossc left	12:48
	→ paulweb515 joined	12:48
	← devoldmx left	12:48
_ikke_	befor add -p, you can do git add -N <file>	12:48
	For untracked files	12:48
	→ devoldmx joined	12:49
	← alexandrerussel left	12:51
j416	oh, handy	12:52
	thanks.	12:52
	already solved it now	12:52
	:)	12:52
	← j0ran1 left	12:52
	→ schemanic joined	12:52
	→ tmcmahon joined	12:53
	→ normanrichards joined	12:53
	→ arussel joined	12:53
arussel	I'm trying to clone a repo without user intervention, I've tried 'yes \| git clone xxx' but I still get a question about host and I don't find a -y option	12:54
	← tureba left	12:54
arussel	how can I do it ?	12:54
	→ tureba joined	12:54
	← rchavik left	12:55
jast	sounds like a prompt from ssh	12:56
iveqy	arussel: what's a -y option? And what question about the host are you getting?	12:56
	→ hoschi joined	12:57
	→ levine joined	12:57
iveqy	arussel: and which protocoll are you using for cloning? http, git, ssh?	12:57
arussel	traditionally on unix system, when a prog ask question, you can pass a -y option to answer yes to all	12:57
	← ewnd9 left	12:57
	→ str joined	12:57
iveqy	arussel: oh, well you always learn something new =). So what question is it you're getting?	12:58
arussel	The authenticity of host 'bitbucket.org ()' can't be established.	12:58
	RSA key fingerprint is	12:58
	Are you sure you want to continue connecting (yes/no)?	12:58
	this is a ssh question	12:58
iveqy	arussel: yeah it is, con't you fix that in your ssh-config?	12:58
jast	ssh doesn't allow bypassing that question except if you change your configuration	12:58
	← tmcmahon left	12:58
iveqy	that seems like a dangerous thing to bypass btw.	12:59
	→ serapath_ joined	12:59
jast	either by adding a known_hosts entry, or by having an ssh pubkey configured as a CA that the hostkey has been signed with	12:59
	← levine left	12:59
	lov_ → lov	12:59
	← gua_ould left	12:59
jast	I don't think ssh has an option for skipping the prompt if it doesn't know the host	12:59
	→ salamanderrake joined	12:59
jast	oh, you can set StrictHostKeyChecking to 'no'	13:00
	no terribly advisable, of course	13:00
arussel	if I answer yes, everything goes well, I don't want to bypass, I just want to answer yes	13:00
jast	you can't, ssh doesn't let you	13:00
	→ hyperair joined	13:00
iveqy	arussel: to answer yes automatic is the same thing as bypass that question	13:00
jast	setting that option has the same effect as answering 'yes', though: it automatically adds the host key	13:00
	← buibex left	13:01
arussel	jast: ssh let me do it, otherwise it wouldn't ask the question.	13:01
	if I answer yes, it let me do it	13:01
	StrictHostKeyChecking seems the answer, thans	13:02
	→ chronos joined	13:03
	← OOPMan1 left	13:04
	→ gua_ould joined	13:04
	← serapath_ left	13:05
	→ serapath_ joined	13:05
	→ m0ltar joined	13:05
	← serapath_ left	13:06
	← kcj left	13:06
	→ devoldmx3 joined	13:06
	→ serapath joined	13:07
serapath	hi, if it happens that i "git checkout label", where label can be a tag or a commit id or head or a selector (e.g. HEAD~) so that my HEAD ends up on a commit object that is not the front of a branch. If i now do a "git commit", could i say that i am still on a branch? (probably "git branch") will say so... but what technical effect has that for any move i might do now?	13:07
	→ dsdeiz joined	13:07
	→ Internet13 joined	13:07
	← dsdeiz_ left	13:08
_ikke_	serapath: If you checkout everything except for a branch, your HEAD gets !detached, and any commit you then do will not be referenced and can get lost	13:08
gitinfo	serapath: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	13:08
	← pdurbin left	13:09
	← devoldmx left	13:09
serapath	_ikke_: i know, but that wasnt the question. in that state, what will "git branch" tell me?	13:09
_ikke_	* (detached)	13:10
	or something	13:10
	like that	13:10
serapath	ah ok, so i left any branch	13:10
_ikke_	yes	13:10
serapath	i am not on any branch	13:10
_ikke_	nope	13:10
serapath	ok	13:10
	awesome... because the tutorial i'm doing says otherwise. it says: "note that the current branch is not changed" ...it might have meant something else, but i was assuming that i cannot leave a branch, thus "git branch" will always tell me that i am on a branch, even if that might not make any sense at all	13:11
	→ abstrusenick joined	13:11
iveqy	serapath: a branch is just a pointer to a commit, when you add a new commit that branch tag will move to point to the new commit, but only if you add that new commit to the commit pointed to by the old branch (well you can always manually move git branch pointers with man git reset)	13:11
gitinfo	serapath: the git-reset manpage is available at http://jk.gs/git-reset.html	13:11
	→ Chiko joined	13:12
serapath	yes, i know	13:12
	i just thought, that maybe, some actions will have an effect based on the last branch i was on	13:12
	if that would be true, one could probably say i am still on a branch in some sense	13:13
	but i guess thats not the case	13:13
	→ darnir joined	13:13
	→ huoxito joined	13:13
	← shoshin left	13:13
	→ v0n joined	13:13
	→ mklappstuhl joined	13:14
	clone1018_ → clone1018	13:14
	→ CKyle joined	13:15
	← CKyle left	13:15
serapath	the "git checkout" command is strange. it is a tool to get files from the staging area ... or get files from a commit id or some other kind of label and copy it to the staging area AND the working copy	13:16
	how is that technically related so that its one command?	13:16
	← PeterBooker left	13:17
_ikke_	git checkout has two main modes	13:17
	→ brisbin joined	13:17
	← mizu_no_oto left	13:17
	→ PeterBooker joined	13:17
Nevik	serapath: the git CLI ui is not very well-thought-out in some parts; it just grew over time	13:17
	→ AaronMT joined	13:18
Nevik	there's always certain justifications for why things are the way they are; but you're right, checkout has some very distinct tasks	13:18
	→ scampbell joined	13:19
	← normanrichards left	13:19
jast	serapath: that part isn't so complicated, really... it fetches stuff from the index into the working tree, and can optionally fetch a different version into the index first. the fun part is that it has a completely different mode that is used for switching branches. :)	13:19
	← Xethron left	13:19
Nevik	^	13:19
	→ buibex joined	13:20
Nevik	where the latter mode will not lose uncommitted changes (unless --force), while the former one may do that	13:20
	→ albel727 joined	13:21
	← jazz264 left	13:23
	← devoldmx3 left	13:24
	← sriracha left	13:24
	→ devoldmx joined	13:24
	→ OOPMan joined	13:24
	← stamina left	13:24
	→ ewnd9 joined	13:25
	→ jfoy joined	13:25
	← papegaaij left	13:25
	→ Pullphinger joined	13:26
	← Pullphinger left	13:26
	→ OnO joined	13:26
	→ jerius joined	13:26
	← devoldmx left	13:27
	→ Risto- joined	13:27
	→ devoldmx joined	13:27
	→ pbujok joined	13:27
	→ fission6 joined	13:27
	→ Pullphinger joined	13:27
	→ echo-are` joined	13:28
	← echo-area left	13:28
	← jfoy left	13:29
	→ devoldmx3 joined	13:30
	→ RaceCondition joined	13:30
	→ raonyguimaraes joined	13:31
	← devoldmx left	13:32
	→ TeckniX joined	13:33
	→ LuxuryMode joined	13:35
	← GODDOG left	13:35
	→ worstadmin joined	13:36
	← Chiko left	13:37
	→ petey joined	13:37
	→ Dougie187 joined	13:38
	→ sriracha joined	13:38
	→ rfoust joined	13:39
	→ phile joined	13:39
	→ frankstallone joined	13:39
	← cmpera left	13:40
	← abstrusenick left	13:41
	← echo-are` left	13:41
	← Katafalkas left	13:41
	→ thierryp joined	13:42
	← jnavila left	13:42
	→ echo-area joined	13:43
	→ Destos joined	13:43
	← thierryp left	13:43
	→ thierryp joined	13:43
	← fission6 left	13:44
	← OnO left	13:45
	→ anli joined	13:45
anli	Can I stash push being in branch B, then switch to branch A and stack pop the changes?	13:45
	So I dont have to commit changes to branch B	13:45
masak	yes.	13:46
	!tias	13:46
gitinfo	Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://gitolite.com/1-basic-usage/tias.html may help with git-specific TIAS.	13:46
serapath	can i tag 2 commit objects with the same tag?	13:46
masak	no.	13:46
_ikke_	no	13:46
serapath	why is it called tags then?	13:46
_ikke_	serapath: Because it's an established term in vcs	13:47
serapath	confusing	13:47
moritz	and older than the terminology used on blogs, I believe	13:47
	→ PaulCapestany joined	13:47
dr_lepper	anli: if the difference between branches allows it, you can just switch to branch A	13:47
moritz	which should really say "labels" instead of "tags"	13:47
	← sharpshooter left	13:47
serapath	so a tag is a label instead	13:47
dr_lepper	anli: e.g. if you've changed files that are the same in both branches	13:47
anli	thanks :)	13:48
	← Stalkr^ left	13:48
serapath	s/instead/then	13:48
dr_lepper	serapath: no, tag is a tag; there's a 1-1 relationship between tags and commits	13:48
	→ cmpera joined	13:48
serapath	semantically, thats a label	13:49
	← tqrst left	13:50
	→ CKyle joined	13:50
	← h_kon left	13:50
	→ hwrd\|work joined	13:51
Remram	is there an easy way to remove untracker branches?	13:52
	← RaceCondition left	13:52
	→ kesha joined	13:52
Remram	untracked*	13:52
anli	the stash trick worked fine	13:53
_ikke_	What do you mean exactly with untracked branches	13:53
	?	13:53
	← matanya left	13:53
	→ shinnya joined	13:54
	→ jefflyne joined	13:54
	← dbuch left	13:54
Remram	_ikke_: remote-tracking branches in refs/remotes/*/	13:55
	which do not have a fetchspec	13:55
	→ EvilDMP joined	13:55
	→ conan_ joined	13:55
	← simesy left	13:56
Remram	remote prune doesn't do it	13:57
	branch -d doesn't even accept it	13:57
	oh, branch -rd does... but still, something automatic would be better	13:57
_ikke_	afaik there is not something automatic	13:58
	→ vince_prignano joined	13:58
	→ bitkiller joined	13:58
	← kesha left	13:59
	← kermit left	14:00
	← soee left	14:00
	← donmichelangelo left	14:00
	← yezariaely left	14:00
	→ sivler joined	14:00
	← Squarism left	14:00
	→ donmichelangelo joined	14:01
	← mw44118 left	14:01
	→ dyng joined	14:01
	→ EdPoole joined	14:01
	← EdPoole left	14:02
	→ Guest60019 joined	14:02
	← petey left	14:02
	→ iamjarvo joined	14:03
	→ dwmw2_gone joined	14:03
	→ petey joined	14:03
	→ tang^ joined	14:03
serapath	git checkout master	14:03
	git checkout HEAD~1	14:04
	am i still on master or am i detached?	14:04
moritz	detached	14:04
serapath	thx	14:04
	← Guest60019 left	14:04
	→ Seb joined	14:04
Seb	hi fellows	14:04
	→ ox_ joined	14:04
_ikke_	serapath: both git status and git branch will tell you that	14:05
	← tvw left	14:05
Seb	i'm simulating a merge of branch bar over branch foo, and finding some conflict; what is the cleanest way of changing foo (adding a commit to it, ideally), to ensure that next time the merge will have no conflicts ?	14:05
jast	serapath: have you ever tried using the same dog tag for two dogs at the same time? they don't like that ;)	14:06
	→ shoshin joined	14:06
serapath	_ikke_: im asking, because "git log --graph" visuals make me believe that a branch is something like a path	14:06
	→ devoldmx joined	14:06
	→ kesha joined	14:06
Seb	hrm, of changing bar I mean	14:06
	← arjun_ left	14:06
moritz	Seb: you could merge foo into bar	14:06
	→ dbuch joined	14:07
moritz	Seb: then the next time you merge bar into foo, all the conflicts will already be resolved	14:07
	← Fandekasp left	14:07
	← devoldmx3 left	14:07
serapath	jast: actually i only know snoopy and rantanplan and i tag them both with dog	14:07
	← subhojit777 left	14:07
jast	serapath: I mean these things: http://en.wikipedia.org/wiki/Pet_tag	14:07
Seb	moritz: unfortunately I can't do that yet, that's too many changes	14:08
jast	try using one collar for two dogs	14:08
	→ mmitchell joined	14:08
jast	I guarantee you it'll be rather difficult	14:08
Seb	moritz: i'd just like to "incorporate" the conflict resolution at this point into bar	14:08
moritz	Seb: merging foo into bar is not more difficult than merging bar into foo	14:08
	→ tvw joined	14:08
jast	this is the meaning of "tag" that presumably coined the usage in revision control	14:09
	→ mw44118 joined	14:09
	→ kcj joined	14:09
moritz	anyway, I have no other idea; maybe somebody else here does.	14:09
	← sivler left	14:09
serapath	jast: i see, so if a tag is some kind of unique identifier, then how should those thinge be called which are "tagged", as in web stuff	14:09
	← kbart left	14:10
serapath	i still have a question regarding branches.... what do those paths shown by "git log --graph" tell me if its not branches?	14:10
jast	if you want to distinguish them, I guess you could call those "keyword tags"	14:10
	← D9_ left	14:10
	→ mizu_no_oto joined	14:11
	→ sriracha1 joined	14:11
	← malukenho left	14:11
	← Dougie187 left	14:11
serapath	jast i will just call them "cats" instead :P	14:11
	← sriracha left	14:11
	→ GT4066 joined	14:11
jast	feel free to :)	14:11
	→ sivler joined	14:11
	← sivler left	14:11
jast	to answer your question of branches, let's distinguish two concepts	14:12
	→ sivler joined	14:12
Seb	moritz: i can't; foo is just not meant to be merge into bar (think "master" branch vs. "bugfix" branch)	14:12
jast	one is the "branch name", implemented as a ref	14:12
Seb	merged*	14:12
	→ Cromulent joined	14:12
	← tvw left	14:12
	← sivler left	14:12
jast	the other is a property of history: two lines of history that have the same parent in the history DAG, possibly merging back together at some point	14:12
serapath	jast: not only that, but as a moving ref, if u commit something new	14:12
jast	you can make the latter happen without ever defining any branch name	14:12
	← gua_ould left	14:12
	← vince_prignano left	14:13
jast	yeah... that, in turn, is implemented for symrefs (symbolic refs), most notably HEAD	14:13
	when you commit to HEAD and it's symbolic (i.e. points to a ref rather than a commit ID), the ref is updated	14:13
serapath	but how can one distinguish which is the "real path" of a given branch if there has been a merge in the past	14:13
moritz	Seb: so maybe you want to rebase the bar branch on top of foo	14:13
jast	you can't fully distinguish that after the fact	14:14
	← nemesit\|znc left	14:14
moritz	serapath: all of them are real	14:14
	← cff left	14:14
jast	what you can tell is which line of history was the target of the merge, because the list of parents of a merge commit is ordered	14:14
serapath	moritz: but "git log --graph" visualization does not let one assume that	14:14
jast	if you're on branch foo and do 'git merge bar', the merge commit will have the commit at foo as its first parent, and the commit at bar as its second parent	14:14
ezakimak_work	i popped a stash, and it had merge conflicts. I resolved them and added the changes to the index, what do I do next to clear that stash? git stash clear? git stash drop?	14:15
jast	--graph represents that by drawing the first parent line to the left and the second to the right	14:15
ezakimak_work	i don't see a --continue like with cherry-pick	14:15
jast	ezakimak_work: drop	14:15
	→ acharlieh joined	14:15
serapath	jast but that is kind of random and artificial... i mean like people who marry and then u flip a coin which name continues	14:16
	← thierryp left	14:16
	← shoshin left	14:16
	→ axl_ joined	14:16
	← devoldmx left	14:16
jast	the names are pure convenience	14:16
ezakimak_work	can a merge commit have any number of parents?	14:17
jast	all that matters to git, in the end, is history	14:17
j416	ezakimak_work: no, only 2 or more	14:17
	→ Dougie187 joined	14:17
	← hwrd\|work left	14:17
jast	ezakimak_work: two or more, but merges with more than two parents are rare and not particularly useful	14:17
ezakimak_work	that's what i meant, no upper bound	14:17
	← PeterBooker left	14:17
	→ devoldmx joined	14:17
	→ PeterBooker joined	14:17
	← robbyoconnor left	14:17
jast	by specifying more than one ref when calling 'git merge', you can create an arbitrarily octopus-y merge	14:18
	→ hwrd\|work joined	14:18
jast	don't do that, though :}	14:18
ezakimak_work	wasn't planning on it	14:18
	just curious	14:18
jast	sure	14:18
	← Cromulent left	14:18
jast	it's a property of the data structures that turned out to be interesting but not useful	14:18
	and it never got removed	14:18
serapath	so HISTORY is a tree than, but "git log --graph" shows me a STORY LINE, right?	14:18
	→ crankharder joined	14:18
	← dsdeiz left	14:19
serapath	s/than/then	14:19
jast	history isn't a tree, it's a DAG (directed acyclic graph)	14:19
	→ Swayz joined	14:19
jast	with the addition of an ordering on the outgoing edges	14:19
	and --graph shows you a visual representation of that	14:19
serapath	jast i think i was refering to something different. ...i tried to refer to the history of a commit object or a branch if u want	14:19
	jast u probably were refering to the whole repository i guess	14:20
	→ kermit joined	14:20
jast	all right, maybe I don't understand what it is that you're looking for. let's say you have a commit X. what seems to be missing?	14:20
	← Davey_ left	14:20
	← beautifulmind left	14:20
serapath	jast: following all possible pointers from that commit X creates a tree	14:21
	→ Davey joined	14:21
	← paulweb515 left	14:21
jast	it doesn't	14:21
serapath	jast: commit X doesnt know its children	14:21
	→ jasonp joined	14:21
jast	it creates a DAG	14:21
serapath	ah	14:21
	ok i understand	14:21
jast	look at a merge commit	14:21
serapath	yes, u are correct	14:21
jast	the history splits in two, and joins back together eventually, as you travel through the past	14:21
serapath	there might be fork and merge	14:21
jast	(in most cases, anyway)	14:21
	exactly	14:22
	← devoldmx left	14:22
serapath	would u call it a fork? because one might assume that i clone a repo of someone	14:22
jast	'fork' is a term used for many different things	14:22
serapath	i would like to choose those terms that do not conflict with each other were possible	14:22
jast	its most general meaning is just what you said: a new line of development branching off of an existing one	14:23
serapath	so is there another word for when i create a new branch from an existing one and commit on that too?	14:23
jast	most people specifically mean "make a new clone and do stuff in it" when they say "fork"	14:23
	→ devoldmx joined	14:23
jast	github and some other hosting platforms encourage that terminology	14:23
serapath	jast: so its "branching" and "merging" instead?	14:23
jast	well, I'd call it "branching" :)	14:23
serapath	ok awesome	14:23
	→ andrewSC joined	14:23
andrewSC	hi all	14:23
	so I'm writing a bash alias where I want git fetch && git rebase current-branch master	14:24
	→ macabre joined	14:25
jast	andrewSC: why? git has 'git pull' for that	14:25
	just configure it to do that and you're done	14:25
	← gratz left	14:25
andrewSC	jast: i don't want to merge my changes on my current/feature branch?	14:25
jast	not to mention that in your example' git fetch' does nothing	14:25
	'git pull' uses merge by default, but can be configured to use rebase instead	14:25
andrewSC	tl;dr i have commits on my current/feature branch that i want on top of the latest master	14:25
	hmmm	14:26
jast	well, that's not too hard	14:26
andrewSC	jast: that makes sense right?	14:26
jast	but the command you had in mind won't do that :)	14:26
andrewSC	oh haha	14:26
jast	you need at least one more command to make that happen... but OTOH you don't need to fetch, so that still makes two :}	14:27
andrewSC	hmmmm	14:27
	→ thierryp joined	14:27
serapath	can i delete a branch name? or rename it from "master" to "(detached)"?	14:27
jast	well, thinking about it, your command will work, but only if current-branch is based on the current version of master already	14:27
	serapath: git branch -d	14:28
	or -m to rename, but '(detached)' is a weird name :)	14:28
serapath	if i do that while i am on master, my head still points to the current active commit, right?	14:28
andrewSC	jast: right, I only branch from master when I work on features	14:28
jast	you can't delete the branch you're on	14:28
	but you could detach HEAD first	14:28
serapath	ah ok	14:29
	thx	14:29
jast	andrewSC: okay. so long as master hasn't changed in the meantime, your rebase command should do it... the fetch is unnecessary	14:29
	→ Malukenh_ joined	14:29
	→ fission6 joined	14:30
	← Cymew left	14:30
	← johnkpaul_ left	14:31
andrewSC	jast: what do you classify as change? Just to clarify, all I'm trying to do is get the latest commits from master and rebase them into my feature branch so that any commits i have in the feature branch stay on top	14:31
	← fission6 left	14:31
jast	andrewSC: okay. so you want to change the feature branch, rather than master?	14:32
	→ lavaman joined	14:32
	→ dsdeiz joined	14:32
	→ pauli1 joined	14:32
jast	(and maybe update master from a remote repository before that)	14:32
	→ ErKa joined	14:33
	← jeffreyio2 left	14:33
	← OOPMan left	14:34
	→ danicampa901 joined	14:34
	→ fission6 joined	14:35
	← petey left	14:35
	→ Burer joined	14:35
	← Burer left	14:35
	→ Burer joined	14:35
	→ nemesit\|znc joined	14:36
andrewSC	jast: right, I want to update the feature branch	14:36
	and you can assume my master branch locally has had git pull run on it	14:36
	so it is latest	14:36
	→ hholtmann joined	14:38
	← arvind_khadri left	14:38
	← nemesit\|znc left	14:38
	→ nemesit\|znc joined	14:38
	← ErKa left	14:38
	← harobed left	14:38
	→ petey joined	14:39
	← iamjarvo left	14:41
	→ iamjarvo joined	14:42
	→ harobed_ joined	14:42
	← alorence left	14:42
	← kcj left	14:43
onethfour	you can check out a commit?	14:43
	i thought you could only check out branches	14:43
	→ maroloccio joined	14:43
	→ quizzi joined	14:43
onethfour	what is HEAD?	14:43
	gitinfo set mode: +v	14:44
j416	onethfour: HEAD is the commit you are currently on	14:44
onethfour	oh	14:44
j416	onethfour: if you are on a branch, HEAD points to the branch head	14:44
onethfour	so what is the point of doing git checkout HEAD^2	14:44
j416	onethfour: if you are not on a branch (detached HEAD), your HEAD points to a commit directly	14:44
	← mikecmpbll left	14:45
cbreak-work	"Rip HEAD off and put it onto the second parent of the current merge commit"	14:45
	→ mikecmpbll joined	14:45
j416	onethfour: maybe you want to look at the state of your project at a certain commit without having to create a brnach for it?	14:45
pauli1	Hi everybody. In the git book on the git web site, at chapter 2.2, the Figure 2.1 show that adding a "untracked" file make it become "unmodified", but if I add a untracked file in my working dir it becomes "Staged". Is that figure wrong?	14:45
onethfour	i've been using git for probably 9 or 10 months now and i've never had to checkout a specific commit	14:45
	only branches	14:46
j416	onethfour: so don't	14:46
	→ chrisshattuck joined	14:46
onethfour	why would you want to? what power does it give you?	14:46
	← maleknet left	14:47
j416	onethfour: there are a lot of use cases	14:47
	→ fss7546 joined	14:47
onethfour	to modify that specific commit?	14:47
	← jefflyne left	14:47
fss7546	is there a way to sign all previous commits in a git repository with a gpg key if they weren't signed before?	14:47
j416	onethfour: debugging to name one. Perhaps there is an error and you want to run your test suite on an earlier commit to see if the error is there too.	14:47
	fss7546: filter-branch perhaps. Not sure.	14:48
	fss7546: yes, you should be able to do it with filter-branch.	14:48
	→ Jackneill joined	14:48
	→ mastro joined	14:48
j416	fss7546: although, history will of course be altered.	14:48
fss7546	thanks, I will try it	14:48
jast	andrewSC: then all you should need to do is, while on the feature branch, 'git rebase master'	14:49
	← Norm_ left	14:49
	→ kevwil joined	14:49
	← Sakako left	14:49
andrewSC	jast: hmmm	14:49
	→ paulweb515 joined	14:49
	← m0ltar left	14:50
	→ moben joined	14:50
moben	Is there a setting I can set to make commit -a add untracked files?	14:51
	→ jnavila joined	14:51
	→ ojacobson joined	14:51
	→ marhaban joined	14:51
	← j0ran2 left	14:51
	→ j0ran2 joined	14:52
PerlJam	moben: sure ... just "git add" them first ;)	14:52
Remram	hmm my company is using a automatic script to extract the list of bugfixes and new features from the log	14:52
	into a changelog	14:52
	← fss7546 left	14:53
	← bitkiller left	14:53
Remram	currently they are using <feature>added more colors</feature> and <ticket>123</ticket>	14:53
	what do you think is a good format?	14:53
	→ Cromulent joined	14:54
Remram	it's better if it doesn't interfere with github's parsing of commit messages	14:54
	← AaronMT left	14:54
_ikke_	Why not: Feature: The feature\nTicket: #123	14:54
	← jceb left	14:54
moben	PerlJam: the point is that I mostly edit files and mostly use 'commit -a'	14:54
Remram	that's what I was thinking	14:55
	← quizzi left	14:55
moben	PerlJam: and then sometimes I add a new file and forget to add	14:55
Remram	it's close to the Signed-off-by: format	14:55
	→ rabbite joined	14:55
	→ OnO joined	14:55
	← xxpor left	14:55
pauli1	moben: why not use add -A?	14:55
	← cmpera left	14:56
rabbite	In order to see the difference between develop and master branches is	14:56
	← jnavila left	14:56
rabbite	(master)$ git diff develop	14:56
	→ jnavila_ joined	14:56
	→ mbease joined	14:56
rabbite	that the correct syntax?	14:56
Remram	git diff develop..master	14:56
	→ bwreilly joined	14:56
Remram	git diff develop will diff develop with your working tree	14:56
_ikke_	git diff takes two commits, not a range	14:56
jast	it supports ranges, too	14:57
rabbite	hm	14:57
	→ cmpera joined	14:57
Remram	git diff --cached develop will diff develop with your index	14:57
ojacobson	It "supports" ranges by parsing them as if they were two commits	14:57
_ikke_	yes, but it's just the same as two commits	14:57
jast	git diff a..b is the same as git diff a b	14:57
_ikke_	yes	14:57
jast	git diff a...b is more interesting	14:57
rabbite	I want to see if my master and develop branches are identical after a merge, is diff the right tool?	14:57
_ikke_	yes	14:57
	→ m0ltar joined	14:57
jast	personally I prefer looking at the history	14:58
Remram	probably not	14:58
jast	but if you prefer diffs, why not	14:58
	ojacobson: three-dot ranges actually have a different meaning	14:58
rabbite	Here's a related question. Assume master has 1 hotfix that is not represented in develop. Come time to merge develop to master, is there a faster way to get develop up to date than develop > master, master > develop?	14:58
jast	git diff a...b == git diff `git merge-base a b` b	14:58
Remram	yiy should do git log master...develop	14:59
ojacobson	jast: sometimes they're the symmetric set difference of the two histories! Sometimes they're the merge base of the two commits!	14:59
Remram	if you want the refs to be the same, and not point to versions with the same content	14:59
	→ heinrich5991 joined	14:59
	← harobed_ left	14:59
jast	ojacobson: the only time the merge base comes in is while diffing. in all other cases it's symmetric difference.	14:59
	→ ManDay joined	14:59
ojacobson	jast: it's also the merge base for at least one rebase option, and I think also at least one checkout option	14:59
	← jerius left	14:59
ManDay	I've committed and pushed. What's the correct way to return to the direct parent and push that change to the remote?	14:59
	→ introom joined	15:00
jast	rabbite: ... just merge develop to master? :)	15:00
ojacobson	jast: man git-rebase, check the --onto option :)	15:00
gitinfo	jast: the git-rebase manpage is available at http://jk.gs/git-rebase.html	15:00
ManDay	I tried git set --hard <theparent> ; git push ; but that wont work	15:00
	← mbease left	15:00
	→ taptapdan joined	15:00
_ikke_	ManDay: try git reset instead of set	15:00
ManDay	_ikke_: That's what I did	15:00
	typo	15:00
rabbite	jast: That won't put the hotfix on develop though, it only brings develops changes into master, not the reverse	15:00
ManDay	_ikke_: Still it says, when I try to push, that the remote is ahead of me	15:01
	→ g0bl1n joined	15:01
rabbite	ManDay: git reset —hard <parent>	15:01
_ikke_	ManDay: You have to force pusbh	15:01
ManDay	_ikke_: ah	15:01
jast	rabbite: right... so you have to do two merges because you want two merges. makes sense to me. :}	15:01
_ikke_	ManDay: !rewrite	15:01
gitinfo	ManDay: [!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), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	15:01
jast	ojacobson: I wasn't aware of that synta	15:01
	x	15:01
rabbite	also, git reset —HEAD <parent> works sometimes	15:01
ManDay	_ikke_: what's the alternative?	15:02
	→ thiago_ joined	15:02
jast	--HEAD doesn't exist :}	15:02
	→ jerius joined	15:02
ManDay	How can I just add the reverted commit on top of itsself?	15:02
ojacobson	jast: I think git-checkout accepts it in one mode too, one sec while I remember	15:02
jast	ManDay: 'git revert'	15:02
ojacobson	jast: ah, here it is:	15:02
jast	ojacobson: yeah, it does... interesting	15:02
ManDay	jast: ty	15:02
ojacobson	:)	15:02
rabbite	_ikke_: instead of a reset, you'd recommend reversing the code change in a second commit?	15:03
	→ valgrind joined	15:03
jast	that's pretty crazy	15:03
ojacobson	IMO it's useful enough that it deserves its own syntax, rather than overloading ...	15:03
rabbite	I get the reasoning there, just want to be clear that's whats suggested.	15:03
ojacobson	or that symmetric diff should be a separate syntax, whichever	15:03
_ikke_	If I know no one has pulled from the repo, I would just force push	15:03
ojacobson	Then it'd work in other places	15:03
jast	rabbite: it's vastly preferable if other people already have a copy of the undesired history	15:03
	in other cases rewriting history makes sense	15:03
	← konadr left	15:04
jast	it might make sense even if history has been shared already... but fixing all the copies is a major headache	15:04
rabbite	jast: you're correct, i was mistaking -head for -hard	15:04
	← netiul left	15:04
jast	so really, rewriting public history is a last resort	15:04
rabbite	jast: assuming that you've pushed, but it was to a feature/ branch, you should be fine doing a reset before a merge to develop	15:05
	→ jefflyne joined	15:05
jast	rabbite: unless someone else is working on the same feature branch and got your changes	15:05
rabbite	I suppose it depends on your shop's workflow	15:05
	← mw44118 left	15:05
rabbite	in my shop, it's rare to have more than one dev working on a feature branch	15:05
	← EvilDMP left	15:06
rabbite	And we don't care about crazy dirty git logs	15:06
	← jnavila_ left	15:06
	→ apipkin joined	15:06
	→ fss7546 joined	15:06
	→ tjbiddle joined	15:07
	→ simesy joined	15:07
	← tjbiddle left	15:07
	← dougsland left	15:07
rabbite	ugh, one of my team members keeps putting # signs in his branch names >.<	15:07
	← thierryp left	15:08
	→ boblee joined	15:08
	→ gua_ould joined	15:08
fss7546	I've got one more question on signed git commits	15:09
	if git objects are named by an SHA1 hash over the contents of the commit, the date, and the SHA1 hashes of the commit parents	15:10
	if I understand correctly how git works	15:10
	would it suffice to just sign the latest commit with a gpg key	15:10
	in order to trust the whole repository?	15:10
	← cyphase left	15:10
	→ Stalkr joined	15:10
	→ Xethron joined	15:11
	← ics left	15:11
	← simesy left	15:11
jast	fss7546: the history of that particular commit, yes.	15:11
	← ox_ left	15:12
jast	in fact that's why signed tags exist (and they've been around for much longer than signed commits; that's a comparably recent addition): they allow you to easily sign releases	15:12
rabbite	Hmm, my GitLab repo is showing branches that command line git can't find.	15:12
_ikke_	rabbite: does git branch -r show them?	15:12
	rabbite: It's also possible that they live in namespaces that don't get automatically fetched	15:13
	→ maestrojed joined	15:13
fss7546	thanks, I think it's not necessary to sign all the previous commits if I just started signing the new ones	15:13
rabbite	ahh yes, there are tons in -r	15:13
	is there any way to delete them?	15:13
_ikke_	fss7546: Nope	15:13
	← moben left	15:14
_ikke_	rabbite: If you only delete them locally, they'll reappear the next time you fetch	15:14
	→ cyphase joined	15:14
jast	rabbite: git push --delete	15:14
	← cbreak-work left	15:15
	← fss7546 left	15:15
	← Stalkr left	15:15
	→ qknight joined	15:15
	→ dbolser joined	15:15
	→ shoshin joined	15:16
rabbite	I see. Can I specify a branch with git push —delete?	15:16
dbolser	sorry for noobs... how do I list the commits different between two branches (assuming one can be ff onto t'other)	15:16
_ikke_	rabbite: yes	15:16
	→ FernandoBasso joined	15:16
	← schemanic left	15:16
	← Risto- left	15:16
	→ pepper_chico joined	15:16
_ikke_	dbolser: git log <branch1>..<branch2>	15:16
	→ Risto- joined	15:16
_ikke_	That shows commits that are in branch2, but excludes the commits that are in branch1	15:17
	→ schemanic joined	15:17
dbolser	_ikke_: tys	15:17
	← PeterBooker left	15:17
dbolser	and '..<branch2>' assumes current branch on the lhs?	15:17
	→ PeterBooker joined	15:17
_ikke_	yes	15:18
dbolser	thanks again	15:18
	← LuxuryMode left	15:19
	← pauli1 left	15:19
	← ayaz left	15:19
	← rudak left	15:19
qknight	hi. i have a strange git problem: when i add 18 branches to a remote repo and do a fresh 'git fetch --all' it works. if i add another branch it does not anymore. i have made some traces using wireshark and i can see that the failing git fetch --all seems to send tcp data over and over and the client 'git fetch --all' will never terminate	15:21
	→ Stalkr joined	15:21
	→ yhager joined	15:21
	→ cigarshark joined	15:21
	← ThomasLocke left	15:22
	← schemanic left	15:22
qknight	i have generated a log of the request but i don't understand that either, seems to be some internal git protocol: https://gist.github.com/qknight/6795474	15:22
	any idea welcome!	15:22
	← threesome left	15:23
	→ schemanic joined	15:23
	→ adac joined	15:23
adac	in a hook, is it possible to execute a file from within the git repository without checking out the repo into another folder?	15:24
	→ stamina joined	15:24
qknight	maybe this is a git bug, on the server i use: git 1.7.9.5 while i use: git 1.8.3.2 on the client; in the /var/log/apache/errors.log i see: fatal: protocol error: bad line length character: �	15:24
	→ Zarthus joined	15:24
	→ jaiballistic joined	15:24
	← Dougie187 left	15:24
	gitinfo set mode: +v	15:25
	→ jaequery joined	15:25
jaiballistic	hi new to git, just did a git svn clone of my svn repository which cloned the whole thing, I want to just checkout one of the tags, do I just go into newgitrepo/tags/mytagver1 and start hacking away?	15:25
	← rath left	15:25
	← Droolio left	15:25
adac	gtg. I'll ask you later again :)	15:25
	← adac left	15:25
	→ jfoy joined	15:25
	← Misan left	15:25
	← Stalkr left	15:25
	→ aartist joined	15:26
	→ pandeiro joined	15:27
	→ tuxx joined	15:27
	← Seb left	15:27
	← tuxx left	15:27
	→ Ziber_ joined	15:28
	← Ziber left	15:28
	← lb1a left	15:29
	→ LuxuryMode joined	15:29
	← gua_ould left	15:29
	→ tuxx joined	15:29
	← VictorCL left	15:30
	← schemanic left	15:30
	← tuxx left	15:30
	← Ziber_ left	15:30
aep	if i have a merge between two large branches, how do i make git log only display changes from one branch or the other?	15:30
	→ Ziber joined	15:30
	← aartist left	15:30
	← jfoy left	15:31
	← mikepack left	15:32
	→ Squarism joined	15:33
	→ Dougie187 joined	15:33
	→ cilly joined	15:33
	← cagedwisdom left	15:33
	→ leeN joined	15:34
	← DoniRS left	15:36
	→ aartist joined	15:36
	← hwrd\|work left	15:36
milki	aep: /me reads man git-log	15:36
gitinfo	aep: the git-log manpage is available at http://jk.gs/git-log.html	15:36
	← dyng left	15:37
	→ harobed joined	15:37
	← mindriot101 left	15:37
	→ bamdad joined	15:38
	← dfanjul left	15:38
milki	aep: maybe --first-parent maybe?	15:39
bamdad	Hi	15:39
	← Cromulent left	15:39
aep	milki: yeah thought so too, but i get the same result	15:39
	← yhager left	15:39
milki	--left/right-only?	15:39
aep	same	15:39
bamdad	Github API doesn't have support for revert. I wonder if there is anyway to replicate the same effect of revert	15:39
	milki shrugs	15:39
	→ simesy joined	15:39
	→ mam[idl] joined	15:39
milki	aep: try !lol	15:39
gitinfo	aep: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	15:39
	← Squarism left	15:39
milki	and derive something useful from that	15:40
	← consumerism left	15:40
	← mklappstuhl left	15:40
aep	yeah, but i have a coworker who is very confused by how branches work. he doesnt want to see the changes that come in via a merge	15:40
	← simesy left	15:40
	→ mikepack joined	15:40
	→ simesy joined	15:40
aep	i was hoping --left-only does that :/	15:41
osse	--first parent does that.	15:41
	← dhruvasagar left	15:41
	← mikepack left	15:41
	→ nmpro joined	15:41
	← renumi left	15:42
osse	at least it should, because it does that here	15:42
milki	well, if the first parent is actually correct	15:42
	→ Neon1024 joined	15:43
aep	hm it works fine for me too	15:43
	→ primitive joined	15:43
milki	occiasionally i see the parents mixed	15:43
	you just said it didnt	15:43
	milki shakes aep	15:43
nmpro	I have folders 1, 2, 3, etc in my repo. can I add a .gitignore inside the folder I want ignored during git pulls?	15:43
	← ManDay left	15:43
	gitinfo set mode: +v	15:43
Neon1024	Hey all	15:43
	← cmpera left	15:43
milki	nmpro: gitignore has nothing to do with git pull	15:43
	→ samtuke joined	15:43
nmpro	milki: ahh ok.. new to git .. just trying to find a easy way to ignore folders in my repo ..	15:44
Neon1024	Can anyone please give me some pointers on moving submodules around inside my repo? I have moved them and updated my .gitmodules and .git/config accordingly, but now git submodule doesn't output anything	15:44
milki	nmpro: !explain	15:44
	hmm doesnt work here	15:44
	theres something..	15:44
nmpro	lol	15:44
milki	uh	15:44
aep	milki: doesnt work on his machine. works fine for me	15:44
milki	step back and explain what you are trying to achieve	15:44
aep	maybe he screwed up the merge	15:44
	← vurma left	15:44
milki	aep: use git lol to visualize it	15:45
aep	yeah	15:45
	thanks	15:45
	← bamdad left	15:45
	→ mikepack joined	15:45
	← cilly left	15:45
	→ akiva joined	15:46
	← ciampix left	15:47
	← Anticom left	15:48
	→ vkz joined	15:48
	→ AWinterman joined	15:48
	← rafeca left	15:49
	← Balliad left	15:50
	← mikepack left	15:50
	← iamjarvo left	15:51
	→ iamjarvo joined	15:51
	← juril left	15:52
	→ vince_prignano joined	15:52
	← geekbri left	15:53
	→ BSaboia joined	15:53
	→ OOPMan joined	15:53
	→ buibex_ joined	15:55
	→ dhruvasagar joined	15:55
	→ natechan joined	15:55
	← psoo left	15:56
	← buibex left	15:56
	← iamjarvo left	15:56
	← bwreilly left	15:57
	← eletuchy left	15:57
	→ eletuchy joined	15:57
	← boblee left	15:59
	→ boblee joined	15:59
	→ mindriot101 joined	15:59
	→ Alagar joined	15:59
	→ cred joined	16:00
	→ dyng joined	16:01
	← eletuchy left	16:02
	→ johnkpaul joined	16:02
	→ rath joined	16:03
	← simesy left	16:03
	← dyng left	16:03
	← mam[idl] left	16:04
	← valgrind left	16:05
	← Skrypter left	16:05
	→ dyng joined	16:05
	← serapath left	16:06
	← dhruvasagar left	16:06
	← mindriot101 left	16:06
	→ yhager joined	16:06
	→ EvilDMP joined	16:07
	← harobed left	16:08
	→ aChatir joined	16:09
	→ matanya joined	16:09
	→ breck7 joined	16:09
	← g0bl1n left	16:10
	→ normanrichards joined	16:10
	→ Squarism joined	16:11
	← mizu_no_oto left	16:11
	← w0ng left	16:11
	→ Stalkr joined	16:11
	← dyng left	16:12
	← user258467 left	16:12
	→ mindriot101 joined	16:13
	→ D9_ joined	16:13
	← LimeBlast left	16:13
	← matanya left	16:13
	→ Skrypter joined	16:14
	← mindriot101 left	16:15
	← stamina left	16:16
	← jaiballistic left	16:16
	← Stalkr left	16:16
	→ mindriot101 joined	16:17
	← PeterBooker left	16:17
	← fission6 left	16:18
	← EvilDMP left	16:18
	← mindriot101 left	16:19
	→ bwreilly joined	16:19
	← fRAiLtY- left	16:19
	← Octalot left	16:20
	← Stygia left	16:20
	← mishok13 left	16:20
	→ jfoy joined	16:22
	← dsdeiz left	16:22
skorgon	is there something like a merge strategy 'theirs' available?	16:22
	→ matanya joined	16:23
jgornick	Hey guys, branching strategy question here. I'm kind of following git-flow and my scenario is I merge a hotfix back into master, tag, then run regression tests. If my regression tests fail for the new release, does this mean I go in and make another hotfix and bump the version again for the fix?	16:23
	→ Octalot joined	16:24
	← jargon left	16:25
	← frankstallone left	16:25
	← crass left	16:26
	→ Mars` joined	16:26
	→ raijin joined	16:26
	← kewah left	16:26
	→ Norm_ joined	16:26
	→ CreekRock joined	16:26
	← nemesit\|znc left	16:27
	→ Error404NotFound joined	16:28
	← CreekRock left	16:28
moritz	jgornick: why not test in the hotfix branch before merging?	16:28
	→ mizu_no_oto joined	16:28
	→ violinappren joined	16:28
	→ TomyLobo joined	16:29
jgornick	moritz: Good question, but in our case regression tests are run on our build server as we tests multiple browsers, environments, etc...	16:29
violinappren	Hello all, how may I reject all pushes to a branch from all users? Effectively making it a read-only branch.	16:29
jgornick	So, there's only one location to do regression tests.	16:29
	← dbuch left	16:29
	← donmichelangelo left	16:30
	← hahuang61 left	16:30
	→ RaceCondishun joined	16:30
	← mikecmpbll left	16:30
	rohieb → rehieb	16:30
	→ DoniRS joined	16:30
	← jxpx777 left	16:30
	→ donmichelangelo joined	16:31
	← john_doe_jr left	16:31
	← ewnd9 left	16:31
	→ fission6 joined	16:31
	← petey left	16:32
	→ GODDOG joined	16:32
	→ nemesit\|znc joined	16:32
moritz	jgornick: maybe tag commits you want tested, and have the build server test them (instead of master, or in addition to master) automatically?	16:33
	→ King_Hual joined	16:34
	→ simesy joined	16:34
canton7	violinappren, if you're using gitolite, use that gitolite config. If you're hosting the repo yourself another way, see the pre-receive hook in man githooks. If you're using something like github, no	16:34
gitinfo	violinappren: the githooks manpage is available at http://jk.gs/githooks.html	16:34
	→ johnnyfuchs joined	16:34
	→ boombatower joined	16:35
	← noname001__ left	16:35
	→ dougsland joined	16:36
	→ dhruvasagar joined	16:36
	← FernandoBasso left	16:36
	← achadwick left	16:36
	← breck7 left	16:37
	← hyperair left	16:38
jgornick	moritz: Hmmm	16:39
	→ nateshr joined	16:39
	→ w0ng joined	16:40
	← nateshr left	16:41
	← introom left	16:41
	← RaceCondishun left	16:41
jgornick	moritz: I think I can look at it this way. I tag the release (1.1), regression tests run and fail, I create another hotfix to fix that issue, tag a new release (1.2), regression tests pass, a user finds a bug, I create a new hotfix for that bug, tag a new release (1.3), etc....	16:41
	→ eletuchy joined	16:41
	→ Katafalkas joined	16:42
	← jasonroyle left	16:42
jgornick	moritz: With that type of a flow, I'm essentially treating regression tests the same as if a user finds an issue and reports it.	16:42
	← simesy left	16:42
	→ threesome joined	16:42
jgornick	moritz: So, I'm always creating a new tag for each merge into master (which is what git-flow recommends) and then when there are any issues with that release, you create another hotfix and bump the version.	16:43
	← Katafalkas left	16:46
	← samtuke left	16:46
	→ marcio_ joined	16:46
	→ RaceCondition joined	16:47
	→ FatDarrel joined	16:47
	← darnir left	16:47
	→ stamina joined	16:47
	→ kanzure_ joined	16:47
moritz	jgornick: well, if it were my workflow, I'd like to keep master free of regressions, so that you can really release from it any time	16:49
	← RaceCondition left	16:49
	← koda left	16:50
	← devoldmx left	16:50
	← spaceonline left	16:50
PaulePanter	In coreboot we need to store patch/diff files, which have whitespace error by design.	16:51
	Is there a way to exclude files from `git diff --check`?	16:51
	http://review.coreboot.org/#/c/3340/1/util/lint/lint-stable-003-whitespace	16:51
	→ Cromulent joined	16:52
	← kanzure_ left	16:52
	← thiago_ left	16:52
	→ cmpera joined	16:53
	→ geekbri joined	16:53
moritz	jgornick: another model (that the perl 5 core developers use) is to have special branch names if you want automatic regression tests; but I guess that doesn't mix too well with gitflow	16:54
	→ devoldmx joined	16:54
	→ mishok13 joined	16:55
	← howlymowly left	16:55
	← acharlieh left	16:56
	← rurufufuss left	16:56
	← CKyle left	16:57
	← Mars` left	16:59
	→ jaiballistic joined	16:59
	← geekbri left	16:59
	← aChatir left	17:00
	← pepper_chico left	17:00
	← ChrisAnn left	17:00
	← D9_ left	17:01
	→ jceb joined	17:02
	jceb → Guest14421	17:02
	→ revyd joined	17:03
	← Xethron left	17:03
	→ Amandil joined	17:03
	← johnnyfuchs left	17:03
	← Devastator left	17:03
	← dhruvasagar left	17:04
	← Cromulent left	17:04
	← Risto- left	17:04
	→ johnnyfuchs joined	17:04
	← shinnya left	17:04
	→ Risto- joined	17:04
	← Risto- left	17:05
	← raijin left	17:05
	→ Risto- joined	17:05
	← whunt left	17:05
	← matanya left	17:05
	→ Stalkr^ joined	17:05
	→ cwoebker joined	17:06
	→ dsdeiz joined	17:06
	← kesha left	17:07
	→ iamstalkr joined	17:07
	← cwoebker left	17:07
	← Stalkr^ left	17:07
	→ cwoebker joined	17:07
	← iamstalkr left	17:08
hakril	Hi ! just a question about remotes/origin/HEAD : why some repos show this ref with "git branch -a" and some others don't ?	17:08
	← Risto- left	17:08
	← Sonderblade left	17:08
	← marhaban left	17:08
	→ Risto- joined	17:08
	→ Stalkr^ joined	17:08
	→ iamjarvo joined	17:08
	→ coldboot\|laptop joined	17:08
	← stamina left	17:08
	← johnnyfuchs left	17:08
	→ staafl joined	17:09
coldboot\|laptop	How do you get Git's pager.log settings to apply to aliases of the log command?	17:10
	← violinappren left	17:10
	← khushildep left	17:10
	→ kesha joined	17:10
coldboot\|laptop	I have this config file here: http://pastie.org/8372632 but the settings in pager.log don't apply when I'm using one of my log aliases.	17:10
	← tang^ left	17:11
	→ hahuang61 joined	17:11
	← mmitchell left	17:11
	→ violinappren joined	17:11
	→ Roots47 joined	17:11
	← tatsuya_o left	17:11
	← taptapdan left	17:12
	→ RaceCondition joined	17:12
	← intellix left	17:12
	← Risto- left	17:12
	→ Holomorphism joined	17:12
	← druonysuse left	17:12
violinappren	canton7: thanks	17:12
	← violinappren left	17:12
	→ Risto- joined	17:12
	→ ErKa joined	17:12
	→ hyperair joined	17:13
	← Stalkr^ left	17:13
	← g41n left	17:13
	→ Lady_Day joined	17:14
	← Risto- left	17:14
	→ PeterBooker joined	17:15
	← iamjarvo left	17:15
	→ johnnyfuchs joined	17:15
	→ Risto- joined	17:15
	→ iamjarvo joined	17:16
	→ whunt joined	17:17
Holomorphism	I'm getting this error when I try to `git fetch`: http://pastebin.com/qzUqJqzV. I think the issue is related to Mac OS X… There are two branches, 'Hotfix' and 'hotfix/SearchPageProductGate', but I'm on an HFS filesystem (case insensitive), so I think the issue is with git creating .git/refs/origin/Hotfix as a file, and .git/refs/origin/hotfix therefore cannot be a directory. The question is… what do I do to stop seeing this error?	17:17
	→ juniorsilva7 joined	17:17
	← Lady_Day left	17:17
Holomorphism	Er, that is, .git/refs/remotes/origin/[hH]otfix	17:17
	← PeterBooker left	17:17
	→ PeterBooker joined	17:18
	← juniorsilva7 left	17:18
	← m0ltar left	17:18
	→ juniorsilva7 joined	17:18
	→ Exio4 joined	17:18
	← angelsl left	17:19
	→ dhruvasagar joined	17:19
	← buibex_ left	17:20
	← iamjarvo left	17:20
	← arussel left	17:20
	→ vigs joined	17:20
	→ arussel joined	17:20
	→ Mars` joined	17:21
	← dsdeiz left	17:21
	← ryxai left	17:21
	← machouinard left	17:22
	← Guest33959 left	17:24
	→ raijin joined	17:24
	→ iamjarvo joined	17:24
	→ matanya joined	17:24
	← Risto- left	17:25
	→ Risto- joined	17:25
	← Risto- left	17:26
	← ErKa left	17:26
	← normanrichards left	17:26
	← revyd left	17:27
	← geomyidae left	17:27
	← matanya left	17:28
	← gitpicker left	17:29
	→ geomyidae joined	17:29
	← catsup left	17:30
	→ matanya joined	17:30
	→ catsup joined	17:31
	→ mmitchell joined	17:32
	← hyperair left	17:33
	→ strategicpause joined	17:33
	← eletuchy left	17:34
	→ nadirvardar joined	17:34
	→ eletuchy joined	17:34
	→ jhulten joined	17:34
jgornick	moritz: Thanks for your insight.	17:35
	← jfoy left	17:35
	← matanya left	17:35
	← Spyro left	17:35
	→ geekbri joined	17:36
	→ Devastator joined	17:36
	← cwoebker left	17:37
	→ aard_ joined	17:37
	→ matanya joined	17:37
	→ thiago_ joined	17:37
	→ badass joined	17:38
	← eletuchy left	17:38
	→ refried_ joined	17:38
badass	Hello, I am having troubles getting a hook to execute. Specifically, I need post-receive to execute a simple 1-liner bash script after a push finishes to the server	17:39
	The problem is that the hook does not seem to be executing at all	17:39
osse	badass: chmod +x ?	17:39
milki	is it executable?	17:39
	→ jfoy joined	17:39
milki	is it named post-receive?	17:39
badass	yes, the file is 755 and it is named appropriately	17:39
milki	make it echo to stdout as well	17:40
osse	not post-receive.sh or anything like that?	17:40
badass	I have tried this	17:40
	→ Misan2 joined	17:40
badass	osse, no	17:40
	thiago_ → thiago	17:40
badass	I've even tried touching a file into /tmp	17:40
	with no success	17:40
	I do know that the push is going through successfully	17:40
milki	!repro with some ls -l please	17:40
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.	17:40
badass	I will need to reproduce an environment which I can view the terminal session more appropriately	17:41
	as right now I'm using a stash plugin to mirror (push) the repo	17:42
osse	badass: the hook is on the server, right?	17:42
badass	yes	17:42
	the hook is on the server that is receiving the push	17:42
	→ mikepack joined	17:42
	→ ingof joined	17:42
	← vince_prignano left	17:43
	→ cilly joined	17:43
	→ Risto- joined	17:43
	← foist left	17:43
	← aartist left	17:43
	← vishal left	17:44
osse	and it's placed inside /path/to/bare-repo/hooks ?	17:44
badass	yes	17:44
osse	and you're using git, not jgit?	17:44
badass	correct	17:44
	← matanya left	17:44
osse	hmm	17:44
	← KaibutsuX left	17:44
osse	The only explanation I can think of is: some of what you've said must be a lie :P	17:45
	does the hook have a shebang?	17:45
	I think it'll work regardless, though	17:45
	← juniorsilva7 left	17:46
	← AWinterman left	17:46
	→ matanya joined	17:46
	→ juanmabc joined	17:47
	← danicampa901 left	17:47
badass	it does	17:47
	and i have not lied... lol	17:47
	I can pastebin all of my stuff if I really must	17:47
osse	badass: what protocol do you push over?	17:47
badass	ssh	17:47
	juanmabc → imbakbabedoll	17:47
badass	its just git push root@host:/path/to/repo develop	17:48
osse	you can make it says just e.g. echo "It works!" and it'll be displayed in your terminal	17:48
	← cilly left	17:48
	→ stamina joined	17:49
	← nmpro left	17:49
canton7	badass, seen !hooks ?	17:49
gitinfo	badass: [!hook_pitfalls] Guidelines for writing hooks: 1. Consume all input (cat >/dev/null if you don't want it). 2. If you use any 'cd', also 'unset GIT_DIR'. 3. Don't git-pull in a hook (or any other script).	17:49
	← Risto- left	17:50
	→ cirenyc joined	17:50
badass	lol	17:50
	i know all that	17:50
	thx though	17:50
	→ mam[idl] joined	17:50
	← mam[idl] left	17:50
	→ mam[idl] joined	17:50
cirenyc	I mistakenly force pushed after a rebase before doing a pull. Is there any way to see if I removed any commits from origin?	17:50
canton7	badass, I hadn't seen anyone mention it, and the "consume all input" thing catches lots of people out	17:50
milki	canton7: !fixup	17:50
gitinfo	canton7: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	17:50
	← ojacobson left	17:51
canton7	milki, hmm?	17:51
	→ ojacobson joined	17:51
badass	thanks canton7	17:51
	→ breck7 joined	17:51
	← breck7 left	17:51
	→ vurma joined	17:51
milki	er	17:51
	→ breck7 joined	17:51
milki	thats not you :P	17:51
	cirenyc: !fixup	17:51
gitinfo	cirenyc: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	17:51
badass	lul	17:51
	← przemoc left	17:52
	← cred left	17:52
	→ tonyj joined	17:52
canton7	badass, ok, let's see some pastebins - I'm out of ideas too	17:52
	← Burer left	17:52
	← tsmith left	17:53
badass	Well, I just tried it from another repo (pushing from a local repo instead of stash) and I get my echos and the commands are being issued	17:53
	so I'm going to rule out the hook being the issue now	17:53
	→ tsmith joined	17:53
canton7	aha cool	17:53
	→ maqr joined	17:53
	→ Xethron joined	17:54
	→ przemoc joined	17:55
	← przemoc left	17:55
	→ przemoc joined	17:55
cirenyc	milki: Not finding yet what I'm looking for. I intentionally did a git push -f, but I don't see whether I removed commits on origin/remote by doing so. I stupidly forgot to pull and don't want to lose those commits.	17:55
milki	cirenyc: hm, so we need to go over what you want. what do you mean by "origin/remote" and "origin"	17:56
	do you mean the remote repo or your remote branches?	17:57
	→ svm_invictvs joined	17:57
cirenyc	milki: The remote repo, named origin.	17:57
svm_invictvs	If I am resolving conflicts and I want to resolve with "theirs"	17:57
	git checkout --theirs /path should be all I need right?	17:57
milki	cirenyc: ok, if you git push -f, then the commits on the remote side become unreachable	17:57
	← antoks left	17:57
milki	cirenyc: however, like any other git repo, the commits are available locally until git gc	17:57
	cirenyc: also, they should be avialable in your local repo as well	17:58
cirenyc	milki: how do I find them locally?	17:58
milki	cirenyc: if you git reflog origin/branchname, you should see the history of the remote branch	17:58
	→ AWinterman joined	17:58
	← mribeirodantas left	17:58
milki	or on the remote repo itself, git reflog branchname	17:58
	→ acharlieh joined	17:58
	← vurma left	17:59
cirenyc	milki: I see "updated by push" and "fetch: sorting head"	17:59
	← mikepack left	17:59
milki	i cant help you figure out what the commits are. you have to examine them yourself	17:59
cirenyc	Is this fetch: sorting head the HEAD commit prior to the git push -f?	17:59
milki	git log the sha and find out yourself	18:00
	← aard_ left	18:00
	→ jnavila__ joined	18:00
	← jehaby left	18:00
	→ dsferreira joined	18:00
cirenyc	got it. looks like i'm good. phew. thank you.	18:00
	^ milki	18:01
	→ Stalkr joined	18:02
	← maqr left	18:03
	→ maqr joined	18:04
	→ ErKa joined	18:04
	→ hyperair joined	18:05
	← melee left	18:06
	← Stalkr left	18:07
	→ minim joined	18:07
	→ melee joined	18:07
staafl	why is "rebase" called that way?	18:07
	I get what it means, but isn't "replay" more intuitive?	18:07
	→ HG` joined	18:08
	→ dreiss joined	18:08
	← OnO left	18:08
canton7	it rewrites a set of commits such that they're based on a new commit - such that they have a new base	18:08
	← DoniRS left	18:10
	← imbakbabedoll left	18:10
milki	staafl: git config alias.replay rebaes	18:11
	done	18:11
	← ErKa left	18:11
staafl	canton7, yes, but that's not the way i think about the process	18:11
milki	uh	18:11
	whats the acronym	18:11
	ftfy	18:11
	there	18:11
	→ dorkmafia joined	18:11
cmn	replay is called cherry-pick or sequencer depending on what you mean	18:11
	← matanya left	18:12
staafl	at least my intuition is about replaying diffs rather than rewriting snapshots	18:12
	← iamjarvo left	18:12
staafl	is that naive?	18:12
canton7	staafl, well that's how the process worked, before the likes of rebase -i. It was purely to do with moving branches around - none of this editing business	18:12
dorkmafia	I have a bunch of files that are showing up as merge conflicts is there a way I can just take mine? i did git co --ours -- $(files) but still show up as AU	18:13
	← melee left	18:13
	← catsup left	18:13
	→ cff joined	18:13
	→ ErKa joined	18:14
	← luxgeek left	18:15
	→ hwrd\|work joined	18:15
	← w0ng left	18:15
	← GODDOG left	18:16
	← refried_ left	18:17
	→ pantsman joined	18:17
	← pantsman left	18:17
	→ pantsman joined	18:17
	← PeterBooker left	18:18
	→ PeterBooker joined	18:18
	→ aldodelgado joined	18:20
	→ TGiFallen_ joined	18:21
	→ matanya joined	18:21
	← TGiFallen left	18:21
	← apipkin left	18:21
	→ gitpicker joined	18:21
	→ iamjarvo joined	18:22
	← raonyguimaraes left	18:23
	→ jazz264 joined	18:24
	→ \|Freeman\| joined	18:25
	← tobscore__ left	18:25
	→ tatsuya_o joined	18:26
	← wakd left	18:27
	← chrisshattuck left	18:27
	← kesha left	18:28
	→ Fervicus joined	18:28
	→ oonska joined	18:28
	← mizu_no_oto left	18:28
Fervicus	Hey. I am getting Permission denied (publickey) when pushing my repo (to both github and heroku).	18:28
	Even though I have uploaded my keys	18:28
	Any idea what's wrong?	18:29
	→ nicknisi joined	18:29
	← nicknisi left	18:29
	← Xethron left	18:30
	→ kongthap joined	18:30
	→ nicknisi joined	18:31
	← acharlieh left	18:31
Holomorphism	I'm getting the error "unable to create directory for .git/logs/refs/remotes/origin/hotfix/SearchPageProductGate" when I try to `git fetch` (http://pastebin.com/qzUqJqzV). I think the issue is related to HFS being case insensitive… There are branches 'Hotfix' and 'hotfix/SearchPageProductGate', so I think the issue is with git creating .git/.../Hotfix as a file, and .git/.../hotfix therefore cannot be a directory. The question is… what do I do to stop	18:31
	→ Xethron joined	18:31
	→ JSharpe joined	18:31
	→ Stalkr joined	18:32
	← Caritade left	18:32
	← kongthap left	18:32
	← TGiFallen_ left	18:32
	→ kongthap joined	18:32
	← minim left	18:32
	← dhruvasagar left	18:32
	← Error404NotFound left	18:33
	→ eletuchy joined	18:35
	→ tobscore__ joined	18:35
	← cmpera left	18:35
	← HG` left	18:36
	→ simesy joined	18:37
	← Holomorphism left	18:37
	← nemesit\|znc left	18:37
	→ Caritade joined	18:38
	→ TGiFallen joined	18:38
	→ shabob joined	18:38
	→ cmpera joined	18:38
	→ osmosis joined	18:38
	← dsferreira left	18:39
	→ dhruvasagar joined	18:39
	← huoxito left	18:40
	← jefflyne left	18:40
	← tobscore__ left	18:40
	→ aeontech joined	18:41
	← strategicpause left	18:41
	→ doev joined	18:41
	→ nemesit\|znc joined	18:41
	← aeontech left	18:42
	→ codeFiend joined	18:42
	← shuaib left	18:42
	← stfn left	18:43
	← eletuchy left	18:43
	← staafl left	18:44
	← GeertV left	18:44
	← Guest14421 left	18:44
	→ kukks joined	18:44
	← kukks left	18:44
	→ kukks joined	18:44
	→ staafl joined	18:45
	← tsmith left	18:45
	→ diegoviola joined	18:45
	→ zackiv31 joined	18:46
	→ antoks joined	18:46
	← thiago left	18:46
	← codeFiend left	18:47
	← jerius left	18:47
	→ russt_ joined	18:48
	← russt_ left	18:48
	← kongthap left	18:49
	← antoks left	18:49
	← Caritade left	18:49
	← iamjarvo left	18:49
	← nwertman left	18:49
	← flaguy48 left	18:50
	→ Caritade joined	18:51
	→ flaguy48 joined	18:52
	← flaguy48 left	18:52
	→ flaguy48 joined	18:53
	→ codeFiend joined	18:53
	← kwmiebach left	18:53
	→ kwmiebach joined	18:54
	→ druonysus joined	18:54
	→ catsup joined	18:54
	→ thiago joined	18:54
	→ a____ joined	18:54
	→ g0bl1n joined	18:55
	← frgtn left	18:55
	→ frgtn joined	18:55
	→ tinjaw joined	18:56
	← vigs left	18:57
tinjaw	for my question, I shall refer to this image from the online version of the Pro Git book	18:58
	http://git-scm.com/figures/18333fig0332-tn.png	18:58
jds	"git fetch" is hanging for me, not quite sure why. "git fetch -vvv" shows a bunch of "got ack"s, then "Receiving objects: 60% (181/301)", at which point it doesn't seem to be doing anything	18:58
	← aldodelgado left	18:58
	→ jerius joined	18:58
	← TGiFallen left	18:58
EugeneKay	Fervicus - many ideas; `ssh -v [email@hidden.address] will tell you most of tem.	18:58
	But spelled right.	18:58
tinjaw	What I want to do is take C10 (and NOT C4 or C3) and merge it onto client	18:58
	→ jonkri joined	18:59
EugeneKay	tinjaw - man git-cherry-pick	18:59
	← maqr left	18:59
gitinfo	tinjaw: the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html	18:59
tinjaw	I have never used patches before and am wondering if that is what I need to do.	18:59
canton7	tinjaw, that's not a merge - merges merfe different branches, not individual commits	18:59
	a cherry-pick is the closest you'll get - that duplicates C10 after C9', so you'll have both C10 and C10'	18:59
	→ maqr joined	18:59
	← matteo left	19:00
	→ tobscore joined	19:00
	→ TGiFallen joined	19:00
	← frgtn left	19:00
	→ cmpera_ joined	19:00
	→ mFacenet joined	19:01
	← dhruvasagar left	19:01
fission6	i just did git add .; git commit -m 'whatever' on the wrong machine. how do i remove this commit ./ undo it and have the files become unstated as if nothing had happened	19:02
tinjaw	git-cherry-pick -- thx -- off to RTFM	19:02
	→ mizu_no_oto joined	19:02
	→ iamjarvo joined	19:02
canton7	fission6, if you didn't push, git reset HEAD^ removes the commit, git reset --soft HEAD^ does the same but keeps the commit's contents staged	19:02
	(both keep all of the changes from the commit being removed in the working copy)	19:03
	← cmpera left	19:03
	cmpera_ → cmpera	19:03
	→ ryxai joined	19:03
fission6	i want the files committed to be brought back into the working copy (not staged though)	19:03
canton7	git reset HEAD^	19:03
fission6	but does that actually remove the commit	19:04
staafl	currently, is there a significant distinction between cherry picking and rebasing?	19:04
fission6	that just resets	19:04
	← jkremser left	19:04
canton7	yes, resets to HEAD^	19:04
	removing HEAD	19:04
fission6	right but that commit is still in the git status right	19:04
	its still a commit	19:04
canton7	hmm?	19:04
staafl	other than the direction in which the commits move between the branches?	19:05
canton7	fission6, just try it :P	19:05
	trust me	19:05
	← TGiFallen left	19:05
	← MacWinner left	19:05
canton7	fission6, !reset	19:05
gitinfo	fission6: A good resource explaining git-reset is http://git-scm.com/2011/07/11/reset.html	19:05
	→ thelamest joined	19:06
	← Caritade left	19:06
	→ tang^ joined	19:06
	← iamjarvo left	19:07
	→ nwertman joined	19:07
	→ huoxito joined	19:07
fission6	thanks canton7	19:07
canton7	basically, reset is to do with manipulating the current branch. reset --soft moves the current branch around without touching the index or working copy. --mixed (the default) moves the branch and index, and --hard moves all 3	19:07
	now if you 'git reset [--soft\|--mixed\|--hard] HEAD', you're moving the current branch to itself, so that bit's a no-op - you're only manipulating the index and working copy	19:08
	→ Caritade joined	19:08
	← freestyl3r left	19:08
	← tinjaw left	19:09
	→ bauruine joined	19:09
	→ aldodelgado joined	19:10
	→ boki joined	19:10
Fervicus	EugeneKay: I just needed to ssh-add	19:11
	thanks	19:11
	← zequence-work left	19:12
	→ youlysse` joined	19:12
	→ dhruvasagar joined	19:13
	→ gtmanfred joined	19:13
	→ infynyxx joined	19:14
	← LuxuryMode left	19:14
	→ druonysuse joined	19:16
	← druonysus left	19:16
	→ jehaby joined	19:16
	→ vince_prignano joined	19:16
	← vince_prignano left	19:16
	→ nix joined	19:16
	← boki left	19:17
	nix → Guest6975	19:17
	→ vince_prignano joined	19:17
	← Guest6975 left	19:17
maxorator	is there any way to include the password in command line when doing git pull/push/etc?	19:18
	← PeterBooker left	19:18
	→ PeterBooker joined	19:18
maxorator	I do not want to permanently store any credentials on the machine	19:18
	← vince_prignano left	19:19
	→ vince_prignano joined	19:19
canton7	use ssh auth, a passphrase, and ssh-agent?	19:19
	← vince_prignano left	19:19
	→ yaroslav joined	19:20
	→ vince_prignano joined	19:20
	← nemesit\|znc left	19:20
	← Jackneill left	19:20
	← vince_prignano left	19:20
	← doev left	19:20
maxorator	any tutorial anywhere?	19:20
	← Davey left	19:20
	← tobscore left	19:20
	← ryxai left	19:21
	→ vince_prignano joined	19:22
	← vince_prignano left	19:22
	← osmosis left	19:23
	← jazz264 left	19:23
	← breck7 left	19:23
cbreak	a tutorial for something as simple as that?	19:24
	→ foucist joined	19:24
	← jonkri left	19:24
foucist	i'm curious if it's possible to have some kind of git branch that is exactly like the master branch, but has configuration differences.. stays up to date with the master branch except for a few commits that are on that branch and automagically merge no matter what is done on the master branch?	19:24
bremner	maxorator: https://help.github.com/articles/working-with-ssh-key-passphrases	19:24
foucist	could be useful for having different server deployment setups	19:24
cbreak	foucist: just make a branch, then commit changes.	19:25
	← tatsuya_o left	19:25
foucist	cbreak: hmm i suppose i could maintain a branch indefinitly and merge changes from master into that branch	19:25
cbreak	yes	19:25
	I hope you like subway maps.	19:26
	← marcio_ left	19:26
	→ vince_prignano joined	19:26
	→ mribeirodantas joined	19:26
	→ TGiFallen joined	19:26
foucist	cbreak: i don't :(	19:27
	→ hwrd\|wor_ joined	19:27
foucist	heh	19:27
	← vince_prignano left	19:27
	← hwrd\|work left	19:27
	→ vince_prignano joined	19:27
	→ minim joined	19:27
	→ ryxai joined	19:28
	→ Davey joined	19:29
	→ TGiFallen_ joined	19:29
	→ danman joined	19:29
	→ SHyx0rmZ joined	19:30
	← vince_prignano left	19:30
	→ vince_prignano joined	19:30
	← TGiFallen left	19:31
	→ LucasTizma joined	19:32
	← vince_prignano left	19:32
	← mFacenet left	19:33
	→ tobscore joined	19:34
	← yacks left	19:35
	← diegoviola left	19:35
	← Mars` left	19:36
	→ Ted_ joined	19:37
	→ alvaro_o joined	19:37
	→ jwmann_ joined	19:38
	→ diegoviola joined	19:38
	→ asQuirreL joined	19:39
	→ CKyle joined	19:39
	← ingof left	19:40
	→ JPeterson joined	19:40
	→ [twisti]_i joined	19:40
	→ mfulz joined	19:41
	→ m0ltar joined	19:41
	← codeFiend left	19:41
	→ marjorip joined	19:42
	← JSharpe left	19:42
Squarism	i merged one branch into another (wo conflicts) but now i regret that - can it be undone easilty?	19:42
	→ Gamecubic joined	19:42
[twisti]_i	hello, i have a question, i use egit, and dont really understand, so i went with the default settings. i commited an existing project to a local repository, and it moved all my files there, and now works directly on the file it moved to that folder. is that right ? it folder also contains a subfolder named .git. are copies of my files in there ? if i do a local 'commit', will my data exist twice	19:43
	? if i empty that folder, but keep the .git subfolder, will i still have my data ?	19:43
	→ chrob joined	19:43
	← shoshin left	19:44
	→ nemesit\|znc joined	19:44
	← GT4066 left	19:44
milki	[twisti]_i: that is called a non-bare repo. see !bare	19:45
gitinfo	[twisti]_i: 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/	19:45
[twisti]_i	thanks	19:45
milki	[twisti]_i: it is fine to work on those files	19:45
	→ LuxuryMode joined	19:45
cbreak	Squarism: did you push?	19:46
	← Pullphinger left	19:46
	→ ingof joined	19:47
	→ codeFiend joined	19:48
	→ lazers joined	19:48
Squarism	cbreak, nope	19:48
	found : git reset --hard HEAD^	19:48
	is that correct?	19:49
andrewSC	alright, so here's the situation: I have a remote branch (let's call it feature) I created with commits that are different/need to be merged into master at some point. However, since my first push of this feature branch I've since pulled master and rebased master into my feature branch locally so those couple of commits are now at HEAD of my feature branch. When I try to push my local feature branch to my remote feature branch I get a	19:49
cbreak	Squarism: if you want to nuke it completely, and it is the msot recent commit, then yes	19:49
	→ serapath joined	19:49
	← ryxai left	19:49
andrewSC	is behind my remote	19:49
cbreak	Squarism: it will also remove uncommitted changes	19:49
andrewSC	even though it's ahead?	19:49
Squarism	cbreak, then im ok	19:49
	→ fotis joined	19:49
andrewSC	tl;dr how do i resolve this? I don't like the idea of forcing the push	19:49
Squarism	cbreak, thanks for holding my hand there =D	19:49
cbreak	andrewSC: rebase rewrites history	19:50
andrewSC	dudeee.....	19:50
cbreak	andrewSC: rebasing master into anything is braindead. don't do it	19:50
	← g0bl1n left	19:50
andrewSC	------_________________________________________------	19:50
	← nadirvardar left	19:50
cbreak	andrewSC: if you have rewritten public (pushed) history, you have to throw away either your history or the remote one	19:50
	so either force push, or reset away the local history.	19:51
	→ frgtn joined	19:51
cbreak	(you can merge it, but that'd be quite bad)	19:51
andrewSC	cbreak: I'll just force pull the remote then	19:51
cbreak	force pull?!	19:51
andrewSC	rather than reset?	19:51
	I'm not sure where the branches differ at this point?	19:52
	← realeyes left	19:52
cbreak	andrewSC: git fetch and then run !lol	19:52
gitinfo	andrewSC: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	19:52
	→ yuppie joined	19:52
	→ jasdel joined	19:53
	← jasdel left	19:53
andrewSC	cbreak: getting broken pipe on the git fetch...... i think github is flaking out	19:53
	pain	19:54
	everywhere	19:54
	← badass left	19:55
onethfour	Has anyone wrote a script when pushing to github, it afterwards displays a link to that commit on github?	19:55
	→ Balliad joined	19:55
	← yaroslav left	19:55
	→ shoshin joined	19:55
	→ Balliad_ joined	19:57
andrewSC	cbreak: called it: https://twitter.com/githubstatus/status/385493053718675456	19:57
	→ breck7 joined	19:58
	← Balliad left	19:58
	→ s2013 joined	19:59
s2013	i keep getting a 403 error while trying to push to github	19:59
	← jnavila__ left	19:59
andrewSC	s2013: https://twitter.com/githubstatus/status/385493053718675456	19:59
rabbite	So if I run "git push —delete" will it remove all listed refs or just the ones with no tracked branches?	19:59
	→ Pyrogerg joined	20:00
cbreak	rabbite: all you tell it to delete	20:00
	rabbite: and you have to use --	20:00
s2013	k thanks andrewSC	20:00
	← s2013 left	20:00
	→ apipkin joined	20:00
	← jwmann_ left	20:00
	→ yacks joined	20:00
	← Kabaka left	20:01
rabbite	For example: "git branch -r" reveals "origin/pre_jquery" which is no longer a branch	20:01
	→ jnavila joined	20:01
rabbite	git push —delete will remove it from the list?	20:01
	← Squarism left	20:01
	← simesy left	20:01
rabbite	I don't really understand how branch namespace works.	20:02
	→ Kabaka joined	20:02
	→ codyrobbins joined	20:02
moritz	git fetch --prune	20:03
rabbite	that will remove any old (and dead) branches?	20:03
	← codyrobbins left	20:03
	→ normanrichards joined	20:03
	← Devastator left	20:03
rabbite	I'm having an issue where branches that were deleted are still showing up in the gitlab dropdowns	20:03
	moritz: that helped, thanks.	20:04
	← crankharder left	20:04
	← codeFiend left	20:04
cbreak	rabbite: if you want to delete a branch from the remote, you have to tell it which	20:04
	rabbite: so if you want to delete origin's foo, git push --delete origin foo	20:05
rabbite	Hm, so git branch -rd origin/pre_jquery	20:05
	hm	20:05
	→ acharlieh joined	20:06
rabbite	fatal: --delete doesn't make sense without any refs	20:06
	:/	20:06
	→ Mars` joined	20:06
	← chamunks left	20:06
	← acharlieh left	20:06
	→ chamunks joined	20:07
	→ druonysus joined	20:07
moritz	git push origin :pre_jquery # iirc	20:07
cbreak	rabbite: branch has nothing to do with it	20:07
	← druonysus left	20:07
	→ druonysus joined	20:07
cbreak	rabbite: it only works on local branches	20:07
	← arussel left	20:07
	← druonysuse left	20:07
cbreak	rabbite: look at my example	20:08
rabbite	git push --delete origin pre_jquery	20:09
	error: unable to delete 'pre_jquery': remote ref does not exist	20:09
	It should be noted that we migrated from Github to GitLab	20:09
	← Balliad_ left	20:09
	← shoshin left	20:09
	→ nmeum joined	20:09
rabbite	so it probably carried these entries with it, though they are not available on gitlab	20:09
	← Mars` left	20:10
	← nicknisi left	20:11
	→ Balliad joined	20:11
	→ Mars` joined	20:12
cbreak	rabbite: so it's already deleted	20:12
rabbite	Why does it still show up on git branch -r?	20:13
	→ acharlieh joined	20:13
cbreak	tracking branches are not pruned automatically	20:14
	once a branch is fetched, no one from the outside can take it away.	20:14
	(they can rewrite history though, remote tracking branches are usually forced)	20:14
	← mribeirodantas left	20:14
rabbite	Ahh, so someone else is still tracking that branch?	20:14
cbreak	rabbite: you are.	20:15
	you have a tracking branch for it	20:15
	← Xiti left	20:16
	← rfoust left	20:16
rabbite	How do I stop tracking it then?	20:16
cbreak	rabbite: as said above, git fetch --prune will do that	20:16
	see man page for details	20:16
rabbite	It didn't though.	20:17
	But I'll read up anyway.	20:17
	→ sq-one joined	20:17
	→ el3ctron joined	20:18
	← PeterBooker left	20:18
	→ cads joined	20:20
onethfour	How would I write a post-push hook which spits out a link to github.com/foo/bar/compare/<SHA1>...<SHA2>	20:21
	→ chrisshattuck joined	20:21
	← tobscore left	20:21
andrewSC	cbreak: so how do you recommend I go about making this better?	20:21
	← minim left	20:21
cbreak	onethfour: which ever language you want.	20:21
onethfour	ok where do i put it though?	20:22
	how do i get the last 2 hashes?	20:22
cbreak	onethfour: you probably want a post-receive though	20:22
	onethfour: read man githooks for your options	20:22
gitinfo	onethfour: the githooks manpage is available at http://jk.gs/githooks.html	20:22
	→ minim joined	20:22
cbreak	andrewSC: making what better?	20:22
	→ PeterBooker joined	20:23
	← cads left	20:23
cbreak	andrewSC: first, look at your history, then think what you would like instead.	20:23
andrewSC	I mean, I know what I want--my commits on top of the master i rebased in	20:24
	but i think that screws up the pull request	20:24
	if i force pushed	20:24
	i think the diff would just get messed up	20:24
onethfour	How do i get the last 2 commit shas ?	20:24
	→ cads joined	20:24
onethfour	individually for usage in a script	20:25
andrewSC	realistically I think the best/simplest way to get this fixed is just to start over from the last push to my remote feature branch	20:25
cbreak	andrewSC: then go to your branch, git rebase master	20:25
andrewSC	i only changed one file	20:25
	i did	20:25
cbreak	so you didn't rebase master on your branch after all? good	20:25
	andrewSC: then, force push that branch	20:25
	that will kill the remote version and replace it with your local one	20:26
	onethfour: git rev-list -2	20:26
	← normanrichards left	20:26
cbreak	onethfour: of which ever ref you want	20:26
	← cads left	20:27
onethfour	HEAD?	20:27
andrewSC	cbreak: right, but I don't know what happens to the pull request in github with that. because the diff from the previous version is gonna be a whole lot different since i did run git rebase master locally in my feature branch	20:27
cbreak	andrewSC: no	20:27
andrewSC	haha	20:27
cbreak	andrewSC: I doubt it'll change	20:27
andrewSC	really?	20:27
cbreak	just make a new pull request and throw your old away	20:27
	andrewSC: did you request a pull with a branch or a hash?	20:27
	← BSaboia left	20:28
andrewSC	but comments from other users and stuff...	20:28
	cbreak: branch	20:28
	→ jceb joined	20:28
	← jceb left	20:28
	→ cads joined	20:28
cbreak	then it'll either be the new branch or the old	20:28
	just try it out...	20:28
	→ Guest13625 joined	20:28
cbreak	if you tag the old hash before force pushing, then you can force push it later if you want to go back.	20:29
onethfour	how many chars is usually used for the short SHA-1 version?	20:29
cbreak	six or more	20:29
	→ consumerism joined	20:30
consumerism	is there a way to have git show me what would change in my working tree if i were to run git stash pop/apply? git stash show compares with HEAD but i'm interested in comparing against the working tree	20:31
	← Xethron left	20:32
	← starfox21 left	20:32
	→ hholtmann_ joined	20:32
onethfour	How do I get the repository name?	20:33
rabbite	You could git stash pop, git diff >	20:33
	→ zarubin joined	20:33
rabbite	*git diff	20:33
	onethfour: git remote -v ??	20:34
	no ??	20:34
consumerism	rabbite: i don't want to apply the stash, i only want to see what would change if i did	20:34
rabbite	just saying, try it	20:34
	Hmm, dunno if you can diff a stash	20:34
	git diff [options] <blob> <blob>	20:35
	This form is to view the differences between the raw contents of two blob objects.	20:35
	→ antoks joined	20:35
rabbite	Perhaps?	20:35
	← hholtmann left	20:35
	hholtmann_ → hholtmann	20:35
ojacobson	rabbite, consumerism: the stash is implemented as a ref (refs/stash) and a bunch of reflog entries, both of which are valid arguments for anything accepting a revision	20:36
	including git-diff	20:36
consumerism	what's the ref for the working tree? i've gotten so far as "git diff HEAD stash@{0}" but this is the same as "git stash show -p"	20:37
	what do i use instead of HEAD to refer to the working tree	20:37
andrewSC	cbreak: alright, resolved the issue and chickened out with a reset --hard	20:38
cbreak	andrewSC: that's the other way.	20:38
andrewSC	moving forward, how do i keep my feature branch up to date with master and also maintain/be able to push new commits i make into the pull request?	20:39
	do i always just do a git pull and deal with the merges?	20:39
	git pull in my feature branch?	20:39
ojacobson	There's no ref for the work tree, because it's not an object	20:39
	→ rednaks joined	20:40
ojacobson	however, 'git diff' with one argument compares the work tree and the named object	20:40
consumerism	ojacobson: ah, got it. thanks	20:41
	git diff stash@{0} or git diff refs/stash are what i was looking for	20:41
	← aldodelgado left	20:42
	→ stfn joined	20:43
	→ refried_ joined	20:44
ojacobson	'git diff stash' should be sufficient	20:45
	man gitrevisions to see why	20:45
gitinfo	the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	20:45
	→ danicampa901 joined	20:45
	← geekbri left	20:45
	← youlysse` left	20:45
	← diegoviola left	20:45
	→ mikepack joined	20:46
	← scampbell left	20:46
	→ tatsuya_o joined	20:46
	← infynyxx left	20:47
	← markovh left	20:47
	← fission6 left	20:48
	→ markovh joined	20:49
	← Guest17847 left	20:49
	← dhruvasagar left	20:51
	← jhulten left	20:51
	→ rcu joined	20:52
	← threesome left	20:52
iveqy	onethfour: git doesn't have repository names	20:52
mastro	What happen if I rebase an history that contain a merge at some point?	20:53
	Can I re-do the merge at that point?	20:53
_ikke_	rebase tends to flatten out merges	20:53
cbreak	andrewSC: you can either merge master into your feature branch	20:54
	andrewSC: or rebase your feature branch onto master (which you did)	20:54
SamB	mastro: rebase is documented not to do particularly nice things to merges	20:54
cbreak	the latter will rewrite history, so you can only force push	20:54
SamB	!tias	20:54
cbreak	the former will create merge commits (unless it is fast forward)	20:54
gitinfo	Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://gitolite.com/1-basic-usage/tias.html may help with git-specific TIAS.	20:54
onethfour	iveqy, ok its a github thing thanks	20:54
_ikke_	.version	20:55
gitinfo	_ikke_: .version: still at 1.8.4, not updating topic.	20:55
andrewSC	cbreak: so realistically speaking i want to fast forward only with the former to preserve commit history	20:55
_ikke_	meh	20:55
	→ fission6 joined	20:55
	→ Fervicus_ joined	20:55
cbreak	andrewSC: you can't fast-forward, you have commits on the feature branch	20:55
andrewSC	:(	20:55
	← TheBrainWork left	20:56
	→ normanrichards joined	20:56
	→ infynyxx joined	20:56
	← Fervicus left	20:57
	← FatDarrel left	20:59
	→ alexw joined	20:59
	→ LuxuryMo_ joined	20:59
	→ ryxai joined	21:00
mastro	which means my best bet is to manually rewrite my commit history and next time be more careful :D	21:00
	thanks SamB cbreak	21:00
	→ kcj joined	21:01
	← LuxuryMode left	21:01
	→ samtuke joined	21:01
	→ skorgon joined	21:01
	← phile left	21:02
	← mizu_no_oto left	21:02
	← Fervicus_ left	21:02
	→ koda joined	21:04
	← sq-one left	21:04
milki	huh, --mirror is very destructive	21:05
	TIL	21:05
	itll delete and force push all day long without asking	21:05
canton7	indeedy	21:06
bremner	yep. this led one person into such a rage the ended up being banned from #git	21:06
Nugget	I'm disappointed I missed that.	21:07
iveqy	milki: and dangerous, one big project (kde?) had a corrupt repo which lead to all their "backup repo" also destroyed	21:07
milki	its in the logs	21:07
iveqy	milki: do we have logs for this channel?	21:07
milki	!logs	21:07
gitinfo	[!irclog] Public logs of #git are kept at: http://colabti.org/irclogger/irclogger_log/git or look at stats for the last 30 days: http://itvends.com/irc/git.html or since Jan 1, 2012: http://itvends.com/irc/git-all.html	21:07
_ikke_	!bot	21:07
gitinfo	[!gitinfo] I am an IRC bot which responds to certain keywords to provide helpful(?) information to humans. Please see http://jk.gs/git/bot for more information about how to use me.	21:07
_ikke_	when was this?	21:07
milki	idunno when it was though	21:07
bremner	long time ago.	21:08
	well, 1 year?	21:08
iveqy	milki: thanks	21:08
	→ sbrath joined	21:08
	← BombaUcigasa left	21:08
	← fotis left	21:08
	← [twisti]_i left	21:09
sbrath	I some how have detached my local git repo from my origin and am lost on how to re-attach, anyone willing to help me?	21:09
	→ MacWinner joined	21:09
	→ BombaUcigasa joined	21:10
	← BombaUcigasa left	21:10
canton7	sbrath, man git remote add	21:10
	→ BombaUcigasa joined	21:10
gitinfo	sbrath: the git-remote manpage is available at http://jk.gs/git-remote.html	21:10
	← BombaUcigasa left	21:10
ojacobson	sbrath: "detached my local git repo from my origin" isn't entirely clear	21:11
	→ BombaUcigasa joined	21:11
	← BombaUcigasa left	21:11
ojacobson	can you show us (gist/pastie a transcript showing the problem)?	21:11
	→ BombaUcigasa joined	21:11
	← BombaUcigasa left	21:11
sbrath	sbrath@SBRATH-ENVY /c/users/sbrath/Documents/NetBeansProjects/Guild_Legacy ((201	21:11
	309292330-release))	21:11
	$ git branch -a	21:11
	* (detached from origin/sbrath-20131001-really-change-dynamic-tags)	21:11
	sbrath-20130926-shipping	21:11
	temp	21:11
	remotes/origin/master	21:11
	remotes/origin/sbrath-20130926-shipping	21:12
	remotes/origin/sbrath-20131001-really-change-dynamic-tags	21:12
	→ fotis joined	21:12
	→ BombaUcigasa joined	21:12
	← BombaUcigasa left	21:12
tang^	ow	21:12
sbrath	This is from Git Bash on my windows box, under the covers of NetBeans.	21:12
	→ FatDarrel joined	21:12
	→ BombaUcigasa joined	21:12
	← BombaUcigasa left	21:12
	← cmpera left	21:12
	← worstadmin left	21:13
ojacobson	Ah, you ran 'git checkout origin/sbrath-20131001-really-change-dynamic-tags' didn't you	21:13
	!detached	21:13
gitinfo	A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	21:13
	← apipkin left	21:13
ojacobson	That's case (a)	21:13
	!reattach	21:13
gitinfo	Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue	21:13
	→ BombaUcigasa joined	21:13
	← BombaUcigasa left	21:13
	← acharlieh left	21:14
	→ BombaUcigasa joined	21:14
	← BombaUcigasa left	21:14
	→ cmpera joined	21:14
sbrath	So I did a git checkout -b temp and that pulled in my local changes. Now I guess I just need to push that up	21:14
	→ BombaUcigasa_ joined	21:14
	→ FernandoBasso joined	21:14
canton7	aah, that form of detached	21:15
sbrath	ojacobson: And I did do what you said.	21:16
	← samtuke left	21:17
	← rcu left	21:17
	→ tpollard joined	21:17
	→ rcu joined	21:17
	← ngladitz left	21:17
	← hoschi left	21:18
	→ joshh__ joined	21:19
	→ apipkin joined	21:20
	← rcu left	21:20
	← marthinal left	21:20
	→ acharlieh joined	21:20
	← zarubin left	21:21
	← joshh__ left	21:22
	→ jantman joined	21:22
	← rabbite left	21:22
	← BombaUcigasa_ left	21:23
	→ worstadmin joined	21:23
jantman	I'm trying to write a server-side update hook that runs syntax checks on changed files (only changed, not the whole repo). works fine for existing branches, but for a new branch, I can't seem to figure out how to get the changed files...	21:23
	anyone happen to know?	21:23
	→ BombaUcigasa joined	21:23
	← BombaUcigasa left	21:23
	→ BombaUcigasa joined	21:24
	← BombaUcigasa left	21:24
	← jnavila left	21:24
iveqy	jantman: how is your script now?	21:24
	→ BombaUcigasa joined	21:24
	← BombaUcigasa left	21:24
iveqy	jantman: and btw. shouldn't this be a part of your code-review process? I mean a good code review will catch things a syntax check would not	21:24
	→ BombaUcigasa joined	21:25
jantman	so for non-new branches, "git diff-tree -r $2 $3" does what I want	21:25
	iveqy: this is way before code review....	21:25
	this is "no, you can't push to origin at all if automated syntax checks fail"	21:25
	← BombaUcigasa left	21:25
jantman	we have a bunch more CI and peer review beyond this	21:25
	→ BombaUcigasa joined	21:26
	← BombaUcigasa left	21:26
	→ BombaUcigasa joined	21:26
iveqy	jantman: hmm, but that should work for a new branch too afaik. The new branch has a parent, yes?	21:26
	← BombaUcigasa left	21:26
	← cmpera left	21:27
	→ BombaUcigasa joined	21:27
	← BombaUcigasa left	21:27
	→ fridim_ joined	21:27
jantman	well for a new branch, $2 (the "old hash") is 0{40}	21:27
	which diff-tree barfs on as an invalid ref, rightly so	21:27
	I don't see anything coming into the update hook arguments about a parent branch...	21:28
	← Pyrogerg left	21:28
	← breakingthings left	21:28
iveqy	jantman: but if you look at your local log, does the branch have a parent?	21:29
	(or is it orphan (git checkout --orphan))	21:29
	← minim left	21:29
	← No_One left	21:29
jantman	um...	21:29
	← axl_ left	21:29
jantman	standby one	21:29
	← donmichelangelo left	21:30
	← acharlieh left	21:30
	← kermit left	21:30
jrnieder	jantman: for new branches, you want diff-tree style format for comparing the empty tree to the new value?	21:30
	→ donmichelangelo joined	21:30
jantman	jrnieder: well all I want is a list of files that are changed in that branch	21:30
jrnieder	I think 'git diff-tree -r $(git mktree </dev/null) "$3"' should do the trick	21:30
	→ BombaUcigasa joined	21:30
	← BombaUcigasa left	21:30
jantman	jrnieder: wouldn't that give me all files in the branch? period?	21:31
jrnieder	I mean	21:31
	← mastro left	21:31
	→ BombaUcigasa_ joined	21:31
	← BombaUcigasa_ left	21:31
	→ kermit joined	21:31
jrnieder	if test "$2" = 0000000000000000000000000000000000000000; then (that thing); else git diff-tree -r "$2" "$3"; fi	21:31
_ikke_	Can't find anything in my logs about someone being banned after git clone --mirror	21:31
	→ BombaUcigasa_ joined	21:32
jantman	iveqy: what's the easy way to tell?	21:32
	← BombaUcigasa_ left	21:32
	← el3ctron left	21:32
	→ JokerxHarley joined	21:32
	← Zarthus left	21:32
iveqy	jantman: !lol	21:32
	← macabre left	21:32
gitinfo	jantman: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all	21:32
jantman	jrnieder: so if I run that in my local git checkout on the branch in question, I get a list of EVERY file in the repo...	21:32
jrnieder	jantman: sorry, let me back up a bit	21:33
	← ojacobson left	21:33
jrnieder	you are writing a server-side "update" hook	21:33
	→ ShooterMG joined	21:33
jrnieder	in English, what should this "update" hook do?	21:33
jantman	jrnieder: syntax check every file that was added or changed	21:33
	← LuxuryMo_ left	21:33
jantman	ignore any file that was already in the repo	21:33
jrnieder	added or changed = since before the fetch?	21:34
	or since when?	21:34
jantman	since the last "thing" that origin knows about...	21:34
jrnieder	you can have lots of branches, after all --- if I were doing the work of this hook by hand, which branch would I compare to?	21:34
	→ akurilin joined	21:34
jrnieder	hm, last "thing" = old value of that branch?	21:34
jantman	for an existing branch, the answer is easy: since the state of that branch that's OLD_HASH	21:34
jrnieder	or last "thing" = 'If I push first "master", then "next", then compare master..next'?	21:34
	← cirenyc left	21:35
	→ BombaUcigasa_ joined	21:35
jantman	old value of that branch	21:35
jrnieder	ok, great	21:35
	← BombaUcigasa_ left	21:35
jantman	however... if there is no old value of that branch on origin...	21:35
	→ BombaUcigasa_ joined	21:35
	← BombaUcigasa_ left	21:35
jrnieder	so, I think 'git diff-tree -r --name-only "$2" "$3"' would do what you want when the branch already existed	21:35
	← rednaks left	21:35
jantman	it should be (in english) "the latest version on origin, of whatever branch this one was cut from"	21:36
jrnieder	that leaves us two more cases: when the branch didn't already exist, and when the push is to delete the branch	21:36
	→ BombaUcigasa_ joined	21:36
dorkmafia	I have a bunch of files that are showing up as merge conflicts is there a way I can just take mine? i did git co --ours -- $(files) but still show up as AU	21:36
jrnieder	jantman: I see. git has no notion of what branch this one was cut from	21:36
jantman	jrnieder: yeah, that's what I'm using now, works perfectly for existing branches	21:36
	← BombaUcigasa_ left	21:36
	← devoldmx left	21:36
	→ BombaUcigasa_ joined	21:36
jrnieder	refs are just pointers into history. There can be more than one ref that points to an ancestor of the commit you're pushing.	21:36
	← BombaUcigasa_ left	21:36
	→ yetet joined	21:36
	← apipkin left	21:36
jantman	jrnieder: yeah I know, hence why I specified english. git speak would have to be... some sort of awful "since the last common commit with..... ????"	21:36
	→ devoldmx joined	21:37
jrnieder	well, even in English, if you hand me the existing history and the stuff you are pushing, you haven't handed me enough information to tell you the baseline to compare to	21:37
jantman	iveqy: so the graph looks like a grenade blew up in a pot of spaghetti...	21:37
	→ acharlieh joined	21:37
jrnieder	one possible answer is "for new branches, always compare to refs/heads/master"	21:37
	i.e., hard-code it	21:37
jantman	jrnieder: exatly...	21:38
	yeah... i suppose I could do that	21:38
	yeah, I guess it's a reasonably safe expectation. "if it's not in master, it's suspect, syntax check it"	21:38
jrnieder	another answer might be more appropriate. it depends on your workflow	21:38
jantman	if you cut the branch off something other than master, that's your problem...	21:38
	cool. thanks	21:39
jrnieder	sure thing	21:39
jantman	yeah that simplified it a lot :)	21:39
	have to run, but thanks again	21:39
jrnieder	np. cheers	21:39
	← JokerxHarley left	21:39
	→ BombaUcigasa_ joined	21:40
	← BombaUcigasa_ left	21:40
jrnieder	jantman: don't forget to handle the $3 = 0{40} case, too :)	21:40
	→ minim joined	21:40
	→ BombaUcigasa_ joined	21:40
	← BombaUcigasa_ left	21:40
	← jantman left	21:41
	← Gamecubic left	21:42
	→ eletuchy joined	21:42
	← bdavenport left	21:42
	← cads left	21:43
	→ JokerxHarley joined	21:43
	← PeterBooker left	21:43
	→ platzhirsch joined	21:44
platzhirsch	Can I git subtree a single file? I just want to link repo1/data/file.txt and repo2/file.txt	21:44
jrnieder	platzhirsch: I don't quite follow. what are you trying to do?	21:47
platzhirsch	I have a list in repo2, I don't want to copy the content all the time to repo1 which makes use of the same list	21:48
	← JokerxHarley left	21:48
platzhirsch	so I thought git subtree would be useful	21:48
iveqy	platzhirsch: no you can't	21:48
	→ Gamecubic joined	21:48
jrnieder	platzhirsch: repo2 has other files, too?	21:48
platzhirsch	yes	21:49
jrnieder	I see. yeah, iveqy's right --- git doesn't have any good way to record that you're bringing in a subset of another repository	21:49
	→ threesome joined	21:49
platzhirsch	I thought that's what subtree is for, as opposed to git submodule, include a part of another repo	21:49
jrnieder	nope :)	21:49
platzhirsch	:\|	21:49
jrnieder	you can make a single-file repo out of repo2	21:50
	with git filter-branch	21:50
	and then include it as a submodule or a subtree in repo1	21:50
	but that would be making a new history, with new commit ids, for the single-file history	21:51
platzhirsch	Seems a bit to much effort, but thanks for the idea :)	21:51
jrnieder	depending on how it gets used, just copying over the file now and then is probably more convenient	21:51
platzhirsch	yep	21:51
iveqy	platzhirsch: let's say that subtree really could pick just one file for you, you would still have to fetch the whole repo and track the whole repo so you wouldn't benefit anything	21:51
	→ diegoviola joined	21:52
platzhirsch	iveqy: yeah, I see that now	21:52
	← cff left	21:52
	→ freeed joined	21:54
	→ psoo joined	21:54
	← psoo left	21:55
	→ Droolio joined	21:55
	→ psoo joined	21:55
	→ BombaUcigasa_ joined	21:56
	← BombaUcigasa_ left	21:56
	→ BombaUcigasa_ joined	21:56
	← BombaUcigasa_ left	21:56
	→ BombaUcigasa_ joined	21:57
	← BombaUcigasa_ left	21:57
	← pantsman left	21:57
	→ tinjaw joined	21:57
	← coldboot\|laptop left	21:58
	← m0ltar left	21:59
	→ BombaUcigasa_ joined	21:59
	← BombaUcigasa_ left	21:59
	← tinjaw left	21:59
	→ coldboot\|laptop joined	21:59
	← hholtmann left	22:01
	← Ted_ left	22:01
	← Dougie187 left	22:01
	→ Dreadnaught joined	22:01
	→ Ted_ joined	22:01
	← Internet13 left	22:02
	→ Internet13 joined	22:03
	← jehaby left	22:04
	→ Spyro joined	22:04
	← staafl left	22:04
	← mikepack left	22:07
	← jerius left	22:08
	→ [twisti] joined	22:08
	← AAA_awright left	22:08
	← freeed left	22:08
[twisti]	how do i delete something from a remote repo	22:08
	← devoldmx left	22:08
[twisti]	i did git rm file; git push	22:08
	but it says everything is up to date	22:09
	and the file is still there	22:09
	did i completely misunderstand how that works ?	22:09
tang^	is that remote bare?	22:10
	→ doev joined	22:10
osse	[twisti]: you have to commit after git rm	22:10
[twisti]	oops	22:11
	← bwreilly left	22:11
[twisti]	thanks	22:11
	← Swayz left	22:11
	→ AAA_awright joined	22:11
	→ harobed_ joined	22:12
	← infynyxx left	22:14
	→ cmpera joined	22:16
	rehieb → lehieb	22:17
	lehieb → rohiob	22:17
	← koda left	22:19
	← Balliad left	22:21
	← fission6 left	22:21
	← fotis left	22:23
	← chrob left	22:24
	← minim left	22:24
	→ m0ltar joined	22:25
	→ minim joined	22:25
	← v0n left	22:25
	→ infynyxx joined	22:26
	← m8 left	22:26
	→ el3ctron joined	22:27
	← ryxai left	22:28
	← threesome left	22:28
	← Ted_ left	22:30
	← raijin left	22:30
	→ Ted_ joined	22:30
	← [twisti] left	22:31
	→ Fandekasp joined	22:32
	← cmpera left	22:32
	← TeckniX left	22:32
	→ jhulten joined	22:33
	← Guest13625 left	22:36
	← eletuchy left	22:37
	→ eletuchy joined	22:37
	yuppie → realeyes	22:38
	→ cjwelborn joined	22:39
	← doev left	22:39
	← harobed_ left	22:39
	← andrewSC left	22:39
	← decotii left	22:39
cjwelborn	omg, i suck at git. I'm having to manually fix a merge with lines like '<<<<<< HEAD (the good stuff) ======== (the bad stuff) >>>>>>> myrepo/master' ....is there a command that will just say keep everything in HEAD, delete everything in 'myrepo/master' ?	22:40
	← danicampa901 left	22:40
osse	cjwelborn: git checkout --ours -- path/to/file	22:41
cjwelborn	osse: awesome, thank you. i knew there was something, I just didnt know what.	22:41
osse	cjwelborn: or maybe you just want to push --force to nuke away whatever you had in myrepo	22:42
	← rath left	22:42
	← johnnyfuchs left	22:42
	← _iron left	22:42
cjwelborn	osse: also good. i may just do that.	22:43
	→ cmpera joined	22:43
	← rigel left	22:43
	← acharlieh left	22:44
	→ Swayz joined	22:45
	← infynyxx left	22:48
	← mmitchell left	22:50
	← DrNick left	22:51
	→ DrNick joined	22:52
	→ tjbiddle joined	22:52
	← jaequery left	22:52
	← coldboot\|laptop left	22:52
	← el3ctron left	22:53
	→ tinti joined	22:54
onethfour	Does git version 1.7.10.4 have any serious flaws?	22:54
	I just staged only 1 file. Did a commit, and it says it committed like 15 files, the files had nothing to do with what I staged.	22:55
	→ el3ctron_ joined	22:55
	→ jaequery joined	22:55
	← primitive left	22:56
	→ cagedwisdom joined	22:56
osse	did you use git commit -a ?	22:56
	had you added a bunch of files before and forgotten about i`t?	22:56
	← jaequery left	22:57
	← babilen left	22:57
	← Destos left	22:57
	→ JosephFerano joined	22:59
onethfour	no i can show you the log, it is absolutely bizar	22:59
	→ babilen joined	22:59
	→ adaptr joined	22:59
onethfour	bizzar	22:59
	← SJr left	23:00
onethfour	crap its "bizzare"	23:00
	→ SJr joined	23:00
	← druonysus left	23:00
	→ druonysus joined	23:01
	← druonysus left	23:01
	→ druonysus joined	23:01
	← _nova left	23:01
	→ boboc joined	23:02
	← jayrobot left	23:04
	→ Lunatrius` joined	23:04
onethfour	osse, can you look at this: https://gist.github.com/anonymous/6801846#file-2013-10-02_18-04-08-txt-L37	23:05
	← Lunatrius left	23:05
	← Kabaka left	23:05
	Lunatrius` → Lunatrius	23:05
onethfour	line 37 is where it says 15 files changed???/ and not one of them is the file I committed	23:05
	← Internet13 left	23:05
	→ jayrobot joined	23:05
	→ woollyams joined	23:06
onethfour	i think my git repo is screwed	23:06
	← jhulten left	23:07
onethfour	either 1, my repo is screwed, my hard drive is hosed, my vm hard drive is hosed, or my git version 1.7.10.4 is hosed	23:07
	→ Kabaka joined	23:07
ScottE	or you did something wrong :-)	23:07
	→ Internet13 joined	23:08
	← boboc left	23:09
	← a-l-e left	23:09
onethfour	i just posted my log ScottE	23:10
	← ingof left	23:10
onethfour	look at the times, all within 1 minute	23:11
	i did nothing	23:11
	← diegoviola left	23:11
onethfour	and if I did, please tell me :(	23:11
	← kevwil left	23:13
	→ fuzzybunny69y joined	23:13
	← Misan2 left	23:14
	→ Destos joined	23:14
ScottE	onethfour: I guess I would look at what the script 'gitbcommit' does, it's when that was executed that the files showed up...	23:16
onethfour	i made that script and it output the command it sent	23:17
	but i'll double check it now	23:17
	← lavaman left	23:18
	→ infynyxx joined	23:18
	← cjwelborn left	23:18
	← infynyxx left	23:19
	→ infynyxx joined	23:19
	← Ozch left	23:19
	← tagava left	23:20
	← pfFredd left	23:20
onethfour	https://gist.github.com/anonymous/6802001 <-- this script just commits a message starting with the current branch name	23:21
	it is benign	23:21
	← Mars` left	23:21
onethfour	see, its not my fault, something serious is wrong	23:21
	← jasonp left	23:22
	→ Mars` joined	23:22
	← boombatower left	23:22
onethfour	time for dinner!	23:22
	later	23:22
	← platzhirsch left	23:23
	← Stalkr left	23:23
	→ wrp6 joined	23:24
	→ babilen_ joined	23:25
wrp6	I'm getting the message: "No submodule mapping found in .gitmodules for path 'php/Zfc/ZfcAdmin" upon executing git submodule update --recurisve	23:25
	Basically, I just cloned a repo with a lot of submodles, and I was hoping to pull all of them	23:25
	← TomyLobo left	23:25
iveqy	wrp6: have you done a git submodule init?	23:25
	← babilen left	23:26
wrp6	Oh, right, sorry	23:26
	Actually, that does give me the same error	23:26
	(But it initializes the others as expected)	23:26
iveqy	wrp6: then I guess you've cloned a corrupted repo	23:27
	→ shinnya joined	23:27
	← leeN left	23:27
iveqy	wrp6: if there's no mapping in .gitmodules then git won't know where to clone the submodule from	23:27
wrp6	How does it know it's a submodule if there's no entry in .gitmodules ?	23:27
	(I thought .gitmodules was the onlyrecord of them)	23:28
iveqy	wrp6: no, the submodule is stored as a special file (which contains data on which commit to point the submodule to)	23:28
	the .gitmodules tells you where to find a submodule	23:28
wrp6	Ah okay	23:30
	← Amandil left	23:31
wrp6	A number of the submodules just aren't being pulled -- but they have entries in .submodules	23:32
	What kind of problem could this be?	23:32
	→ Dougie187 joined	23:32
wrp6	In fact only half of them were pulled, trying to figure out what could have been different between them	23:33
	← tang^ left	23:34
	→ D9_ joined	23:34
iveqy	wrp6: are they present in your commit?	23:35
	and do you get any error message?	23:36
	← normanrichards left	23:37
wrp6	Well, the directories are all present, but in the clone they are empty. In fact, when it lists teh registered submodules at the start of submodule update --recursive, they seeem to be missing	23:38
	(But there are entries in .gitmodules in the local clone I have)	23:38
	← pandeiro left	23:38
	→ LuxuryMode joined	23:38
	← LuxuryMode left	23:39
	→ boombatower joined	23:39
iveqy	wrp6: is this a public project?	23:39
wrp6	Nope, I just committed and then cloned remotely	23:39
iveqy	wrp6: what does git submodule tell you?	23:40
wrp6	Outputs the same list as returned by submodule update --recursive	23:40
	(Of course with the hashes etc)	23:40
	← jaiballistic left	23:42
	← psoo left	23:42
	← vkz left	23:42
wrp6	Thanks for the help by the way. I will probably just scp the data over for now but it would be good to solve it in the long run.	23:43
	← boombatower left	23:44
iveqy	wrp6: well, that's the list of submodules that commit contains, if you've more submodules listed in .gitmodules they are probably in an other commit or faulty removed	23:45
	← danman left	23:45
	→ carpii joined	23:46
	← gitpicker left	23:46
carpii	i just created a branch, made a commit, and then merged into master. I then started pushing to remote, but noticed id branched from the wrong branch. Now my master is diverged from origin/master and even git reflog cant seem to get me back to a clean setup. Is it safe to checkout local master and 'git reset origin/master --hard' ?	23:47
	i forgot to say, i CTRL-C'd the push before it completed, and having checked out remote repo, it looks unaffected	23:48
	→ SilverKey joined	23:48
	← SilverKey left	23:49
iveqy	carpii: define safe?	23:51
	it sounds like it what you want to do. But no, it's not safe, it will destroy data	23:51
carpii	id like to just get local master to match remote exactly, so its not diverged anymore	23:51
	and the sort out the damage from there	23:51
	destroy which data ?	23:51
	→ R1KRDO joined	23:52
iveqy	carpii: it will destroy your worktree and move your master branch possible to leave some commits unrefered and then they will be deleted by git gc in 14 days	23:53
carpii	hm	23:53
	→ Stalkr joined	23:54
carpii	is there a way to know which commits will be left dangling?	23:54
	my local master was up to date before i made this bad merge	23:54
	→ normanrichards joined	23:54
	← natechan left	23:54
	→ D-Chymera joined	23:54
carpii	admittedly i dont understand why its diverged really	23:54
	← jason237 left	23:55
	← D-Chymera1 left	23:55
	← thiago left	23:56
	→ relipse joined	23:56
iveqy	carpii: the commits that are in your master but not in origin/master (nor any other branch) will be dangling	23:56
	if you did a merge into master of course it has diverged, it contains a merge	23:56
carpii	ok, thats fine. they will still be in local branch and i can remerge when ready	23:56
	but merging a feature branch into master is my usual workflow. it very rarely diverges	23:57
	unless something has gone wrong (or i need to pull origin)	23:57
iveqy	carpii: did someone else maybe pushed to origin/master?	23:58
	← Stalkr left	23:58
	← oonska left	23:58
carpii	no, ive only one other developer	23:59
	and he's not working atm	23:59

Logs Search ←Prev date Next date→

Channels Documentation