IRCloggy #git 2012-08-11

Provider of IRC logs since 2005.

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

	2012-08-11
	→ chinopozo joined	00:00
	tizzo → tizzo-afk	00:01
	→ ISF joined	00:02
	← zackiv31 left	00:02
	← nifty left	00:02
	← skyf left	00:04
	→ agile joined	00:04
jnewt	i'm trying to clean up an svn repo that i've git svn cloned, using filter-branch. what i want is to remove all the generated files that have been committed over the history of the repo.	00:05
	← nightcrawler786 left	00:06
jnewt	but i get a Warning: Ref 'refs/heads/master' is unchanged. and nothing is removed. i'm using git filter-branch --tree-filter 'rm -rf *.exe' HEAD as a trial. not doing anything	00:06
	← DrHouse\|TARDIS left	00:07
jnewt	the example shows : git filter-branch --tree-filter 'rm filename' HEAD as removing a file from all commits. seems my command would do the same for all files ending in .exe.	00:07
	tizzo-afk → tizzo	00:08
	← kpreid left	00:08
	← beneggett left	00:08
	→ kpreid joined	00:08
	← lirt left	00:08
	→ capiira joined	00:09
	← capiira left	00:09
	← dreiss left	00:10
	← Sebo left	00:10
	→ beneggett joined	00:10
	→ DrHouse\|TARDIS joined	00:10
	← endra left	00:11
	← v0n left	00:11
	← DrHouse\|TARDIS left	00:12
mmb	grawity: thanks it looks like that will work	00:13
	← cybersphinx left	00:15
	→ capiira joined	00:15
	← capiira left	00:15
	← SeySayux left	00:17
	← etcetera left	00:18
	→ marhaban1 joined	00:20
	← SirCmpwn left	00:20
	← marhaban left	00:20
	→ nyx joined	00:22
	← gtrak left	00:24
	← eletuchy left	00:25
	→ mcluskydodallas joined	00:26
	→ SeySayux joined	00:26
	→ eletuchy joined	00:27
	← guampa left	00:27
	→ gnoel joined	00:29
	← imachuchu left	00:29
	← maletor left	00:29
	← SamB_MacG5 left	00:29
	→ guampa joined	00:29
	← mmb left	00:30
	→ SamB_MacG5 joined	00:32
	← beneggett left	00:32
	← sawdisk left	00:32
	← erynofwales left	00:33
	← mwrosen left	00:33
	← nyx left	00:33
	→ sawdisk joined	00:33
AAA_awright	Anyone know how git-add detects if it detects if it should add a subtree or a commit to a tree?	00:34
	i.e. for submodules	00:34
	← msmithng left	00:34
	← maestrojed left	00:36
	→ etcetera joined	00:36
	← etcetera left	00:38
	← hyperair left	00:38
	← loops left	00:39
	→ reds joined	00:39
	→ loops joined	00:39
	reds → Guest22555	00:40
	→ hyperair joined	00:40
	← apok left	00:41
	→ nyx joined	00:43
	← Guest22555 left	00:44
	← mcluskydodallas left	00:45
	→ reactormonk joined	00:45
	← todd_dsm left	00:46
	← dougsland left	00:47
	→ mwrosen joined	00:47
	← shruggar left	00:50
	→ etcetera joined	00:51
	← hyperair left	00:51
	← marhaban1 left	00:51
	← nowhere_man left	00:51
	← oakdog8 left	00:51
	← cakehero left	00:51
	← dorkmafia left	00:51
	← wago left	00:51
	← jstemmer left	00:51
	← marc_kdab_ left	00:51
	← Adys left	00:51
	← orafu left	00:51
	← intripoon left	00:51
	← flippo left	00:51
	← Arokh left	00:51
	← nyxx left	00:51
	← Swathe left	00:51
	← anildigital left	00:51
	← LittleFool left	00:51
	→ oakdog8 joined	00:52
	→ Adys joined	00:57
	→ Arokh joined	00:57
	→ intripoon joined	00:58
	→ hyperair joined	00:58
	→ marhaban1 joined	00:58
	→ nowhere_man joined	00:58
	→ cakehero joined	00:58
	→ wago joined	00:58
	→ dorkmafia joined	00:58
	→ jstemmer joined	00:58
	→ marc_kdab_ joined	00:58
	→ nyxx joined	00:58
	→ AABBOV9 joined	00:58
	→ orafu joined	00:58
	→ Swathe joined	00:58
	→ anildigital joined	00:58
	→ LittleFool joined	00:58
	→ BinGOs joined	00:58
	→ rook2pawn joined	00:58
	→ xrogaan joined	00:58
	→ JPT joined	00:58
	→ bronsen joined	00:58
	morgan.freenode.net set mode: +v	00:58
	← cakehero left	00:58
	← LittleFool left	00:58
	← Eridius left	00:58
	→ raz joined	00:58
	→ cakehero joined	00:58
	→ flippo joined	00:58
	→ intripoon_ joined	00:58
	← hwrd\|work left	00:59
	← anildigital left	00:59
	← JPT left	00:59
	← diegoviola left	00:59
	→ jjbohn joined	01:00
	→ LittleFool joined	01:00
	→ anildigital joined	01:00
	← intripoon left	01:01
	← cakehero left	01:01
	← marhaban1 left	01:02
	→ JPT_ joined	01:02
	→ marhaban joined	01:02
	→ dougsland joined	01:03
	← orafu left	01:03
	→ w09x joined	01:03
	tizzo → tizzo-afk	01:04
	→ orafu joined	01:04
	→ diegoviola joined	01:04
	← Dimitr left	01:04
	→ cakehero joined	01:04
	→ erynofwales joined	01:05
	← mwrosen left	01:07
	→ kenperkins joined	01:08
	→ axl_ joined	01:08
	← opus left	01:08
	← chuckharmston left	01:08
	→ chuckharmston joined	01:08
	→ rehat joined	01:10
	rohan\|building → rohan\|afk	01:14
	← segher left	01:16
	← tweedledumber left	01:16
	→ segher joined	01:17
	→ skyf joined	01:18
	← wago left	01:19
	← bizhanMona left	01:20
	← Lgb left	01:20
	← palm_vue left	01:20
	→ Lgb joined	01:21
	→ bizhanMona joined	01:21
	→ palm_vue joined	01:21
	→ msmithng joined	01:22
	rohan\|afk → rohan32	01:24
	← adityamenon left	01:25
	← ISF left	01:25
	→ intripoon joined	01:27
	→ capiira joined	01:29
	← capiira left	01:29
	→ igotnolegs joined	01:30
	← intripoon_ left	01:30
	← marhaban left	01:33
	→ Lord_DeathMatch joined	01:33
	← codeFiend left	01:34
	→ dreiss joined	01:34
	→ maletor joined	01:34
	→ Vampire0 joined	01:36
	→ jepler joined	01:39
	← Vampire0_ left	01:40
	→ storrgie joined	01:40
	→ duckxx joined	01:41
	← maletor left	01:43
	← flippo left	01:44
	→ flippo joined	01:44
	→ converge joined	01:44
	← nyx left	01:44
	→ maletor joined	01:48
	→ russt_ joined	01:49
	← jjbohn left	01:51
	← russt left	01:52
	russt_ → russt	01:52
	← Sym3tri left	01:52
	← artnez left	01:53
	← axl_ left	01:55
	→ simesy joined	01:57
	→ jjbohn joined	01:58
	← russt left	01:58
	→ Araxia joined	01:58
	→ russt joined	01:59
	← russt left	02:00
	→ beneggett joined	02:00
	← pdtpatrick left	02:00
	← edude03 left	02:01
	→ Anxi80 joined	02:02
	rohan32 → rohan\|afk	02:05
	→ jey joined	02:08
	→ apok joined	02:11
	← Lord_DeathMatch left	02:11
jey	hi, I performed a subtree merge of git://somewhere/foo.git into my repository's foo/ subdirectory, but now when I do "git log foo/something.py" I only see the merge commit. how can I see the full history for that file? I've read through the git-log manual and I feel like I'm missing something trivial	02:13
	← tnzr left	02:15
	gitinfo set mode: +v	02:17
	→ capiira joined	02:17
	← yoklov left	02:18
	← magcius left	02:19
	→ edude03 joined	02:20
	← chuckharmston left	02:20
	← edude03 left	02:20
	→ edude03 joined	02:21
	→ magcius joined	02:21
	← dreiss left	02:22
	→ Super3 joined	02:23
	→ Anders_J joined	02:23
	→ joey77 joined	02:23
	← storrgie left	02:24
	← Anxi80 left	02:25
	→ latrommi joined	02:25
	← Super3 left	02:26
	← apok left	02:27
	→ andrewhl joined	02:28
	← bubbles10101 left	02:28
	← joey77 left	02:28
	← goshawk left	02:29
	→ johntron joined	02:29
	← johntron left	02:29
	→ BinGOs_ joined	02:31
	→ bubbles10101 joined	02:31
	→ xrogaan_ joined	02:31
	← xrogaan_ left	02:31
	→ xrogaan_ joined	02:31
	← hyperair left	02:31
	→ hyperair joined	02:31
	← raz left	02:31
	← marc_kdab_ left	02:31
	← BinGOs left	02:31
	← xrogaan left	02:31
	→ raz joined	02:31
	← raz left	02:31
	→ raz joined	02:31
	→ hwrd\|work joined	02:32
	→ marc_kdab_ joined	02:32
	→ Chib joined	02:33
	→ jobin joined	02:36
	← EricKamsky left	02:37
	← jobin left	02:38
	→ jobin joined	02:39
	← jobin left	02:40
	→ jobin joined	02:40
	← jobin left	02:41
	→ red82 joined	02:41
	← beneggett left	02:41
	→ jobin joined	02:42
	← bubbles10101 left	02:43
	← jobin left	02:43
	→ pierre__ joined	02:43
	→ jobin joined	02:43
	→ bubbles10101 joined	02:44
	← jobin left	02:44
	← simesy left	02:45
	→ jobin joined	02:45
	← jobin left	02:45
	→ jobin joined	02:45
	jobin → andy55	02:45
	← orafu left	02:46
	→ OuLouFu joined	02:46
	OuLouFu → orafu	02:47
	← nowhere_man left	02:47
	← converge left	02:50
	← arooni-mobile left	02:50
	→ angelsl joined	02:50
	← hwrd\|work left	02:54
	guampa → vtolcow	02:58
	vtolcow → guampa	02:58
	→ hwrd\|work joined	03:03
	→ dfr\|mac joined	03:03
	← blupp left	03:03
	→ reds joined	03:04
	reds → Guest98891	03:04
	→ blupp joined	03:06
	→ simesy joined	03:07
	→ chuckharmston joined	03:09
	← Guest98891 left	03:09
	→ Anxi80 joined	03:11
	→ gavin_huang joined	03:11
	→ chuckhar_ joined	03:12
	← chuckhar_ left	03:12
	→ jackh joined	03:12
	← chuckharmston left	03:13
	← edude03 left	03:14
	← localhost left	03:14
	→ edude03 joined	03:15
	→ localhost joined	03:17
	← maletor left	03:17
	← alberto56 left	03:18
	← bubbles10101 left	03:21
	← duckxx left	03:21
	← towski left	03:24
	← latrommi left	03:25
	← erynofwales left	03:26
	→ nyx joined	03:27
	→ AaronMT joined	03:27
	← Anxi80 left	03:28
	← metasansana left	03:29
	→ arkiver joined	03:30
	← AaronMT left	03:30
	← rohan\|afk left	03:30
	← Chillance left	03:31
	← diegoviola left	03:32
	→ diegoviola joined	03:36
	← hwrd\|work left	03:37
	→ subbyyy joined	03:37
	→ rchavik joined	03:38
	← rchavik left	03:38
	→ rchavik joined	03:38
	← kukks left	03:40
	← TomyLobo left	03:40
loops	jey, try --follow	03:41
jey	no luck, then I don't even see the merge commit anymore	03:41
	← jjbohn left	03:41
	← gavin_huang left	03:42
loops	hmmm.. what about git log -- /something.py	03:42
	→ jjbohn joined	03:43
	← jjbohn left	03:44
	→ irqq joined	03:45
loops	git log -- foo/something.py /something.py	03:45
	should work	03:45
	← arkiver left	03:46
jey	$ git log -- pokerd/poker-network/pokernetwork/pokercashier.py /poker-network/pokernetwork/pokercashier.py	03:46
	fatal: Could not switch to '/poker-network/pokernetwork': No such file or directory	03:46
loops	ummm...	03:47
	→ api joined	03:47
	← thelazydeveloper left	03:47
loops	jey, no idea.. what version of git? you may have a bad entry in your .git/config	03:52
	← rehat left	03:52
	← Goplat left	03:53
jey	1.7.9.5	03:53
	← aruiz left	03:53
loops	jey.. there is a bug which generated similar message fixed in 1.7.10	03:53
	is worth upgrading before straining brain too much	03:54
jey	oh hm, ok, I'll give that a shot :)	03:54
	→ zackiv31 joined	03:54
	→ jdav_gon` joined	03:54
	→ kortti joined	03:55
	← mritz_ left	03:55
	← gnoel left	03:56
	← jdav_gone left	03:57
	→ noyb joined	03:57
	→ gtuckerkellogg joined	03:59
	→ zamn joined	04:01
	← zamn left	04:01
	→ zamn joined	04:01
	→ landypro joined	04:01
zamn	i accidently broke my staging area :o. I ran git reset --hard HEAD^ to undo my last commit not realizing that it throws away all the changes i made..can i undo this or am i sol?	04:01
	→ hwrd\|work joined	04:03
	← noyb left	04:03
	→ api_ joined	04:04
	← nutron left	04:04
	← api_ left	04:04
	← api left	04:04
	→ russt joined	04:06
	← chinopozo left	04:06
	← savid left	04:09
avpx	zamn: Were the changes ever stored in git objects? (That is, did you run git add on them?)	04:09
zamn	no :(	04:09
	← nyx left	04:10
avpx	Hmm. That's no good;	04:10
	you're probably not going to be able to recover the changes.	04:10
loops	no luck then zamn, --hard is like running rm -rf on em	04:11
zamn	yeah this sucks	04:11
	← dfr\|mac left	04:11
loops	need to commit early n commit often... git can't help ya til you commit stuff	04:11
avpx	Yeah, sorry. Unsafe commands should probably prompt you for confirmation.	04:11
	→ gnoel joined	04:12
	→ noyb joined	04:12
loops	i'm surprised git hasn't added some protection though, it could stash changes before doing --hard	04:12
avpx	I personally think it would be reasonable to prompt before doing anything destructive	04:12
loops	this must have been raised on the list before, i think the response was something like, we're not going to penalize power users to protect noobs	04:14
	→ v0n joined	04:14
avpx	Still, there could be a default configuration	04:14
loops	avpx, agreed. sorry zamn, sucks	04:14
avpx	Power users could just git config that away	04:14
	→ tnzr joined	04:15
avpx	Meh, porcelain could probably use some work	04:15
loops	even power users can have a thinko, auto stashing would be a nice safety net	04:15
zamn	yeah im an idiot and didnt read the comments of the SO question	04:15
	i commited something else first	04:16
	then i was going to commit this work heh	04:16
	← russt left	04:16
avpx	Yeah, automatically stashing would be nice.	04:16
	← andrewhl left	04:16
	→ russt joined	04:16
	→ gavin_huang joined	04:18
loops	zamn, do you use vim ?	04:18
zamn	yes	04:18
loops	zamn, it may be possible to recover from undo history	04:18
zamn	ah its okay	04:18
	i got the code back	04:18
	luckily i pasted it to pastie for help on something 20 minutes ago	04:18
	so its not too bad	04:18
	← andy55 left	04:19
	→ Textmode joined	04:20
	← russt left	04:22
	→ russt joined	04:23
	→ dfr\|mac joined	04:24
	→ herdingcat joined	04:24
	← rootfield left	04:26
	→ velikasha joined	04:30
	← velikasha left	04:31
	→ towski joined	04:31
	← herdingcat left	04:31
	← gavin_huang left	04:31
	→ herdingcat joined	04:34
	→ kleist joined	04:35
	← juvenal left	04:35
	→ thirdknife joined	04:38
	← guampa left	04:48
	← herdingcat left	04:50
	← hwrd\|work left	04:51
	→ Guest74974 joined	04:52
	→ PovAddict joined	04:52
PovAddict	what is the default order of revisions returned by git rev-list?	04:52
	I always thought it was topo-order, but that turned out to be my own confusion because gitk uses topo-order by default (it has to, since it has to render graphs)	04:53
	→ varg joined	04:53
PovAddict	manpage says "reverse chronological order"; isn't that date-order?	04:53
	→ herdingcat joined	04:53
	→ gavin_huang joined	04:54
zamn	yes	04:54
	→ lyddon joined	04:55
	← lyddon left	04:55
	→ lyddon joined	04:55
PovAddict	zamn: but "diff -u <(git rev-list --all) <(git rev-list --all --date-order)" is non-empty	04:56
	← lyddon left	04:56
zamn	i was confirming chronological order is date order	04:56
	lol	04:56
	im a git newb	04:57
	sorry	04:57
	→ andy55 joined	04:57
PovAddict	it's interesting how I never come here with simple questions	04:57
	last time someone ended up looking at the git source code and scratching his head himself	04:58
	→ lyddon joined	04:58
	→ hwrd\|work joined	04:58
	→ Turicas joined	04:59
	← reactormonk left	05:00
	← gavin_huang left	05:00
	→ nyx joined	05:02
	← SamB_MacG5 left	05:03
	→ reds joined	05:04
	reds → Guest70923	05:05
	→ gavin_huang joined	05:05
	← andy55 left	05:06
	← nyx left	05:08
	← cakehero left	05:08
	← simesy left	05:09
	← Guest70923 left	05:09
	← gavin_huang left	05:10
	→ simesy joined	05:12
	← varg left	05:13
	← simesy left	05:14
	→ gavin_huang joined	05:15
	→ nyx joined	05:17
	← Anders_J left	05:17
	→ dinesh6654 joined	05:18
	← gavin_huang left	05:19
	← nyx left	05:20
	→ varg joined	05:20
	← hyperair left	05:20
	→ dang joined	05:21
	← red82 left	05:21
	→ hyperair joined	05:21
	→ red82 joined	05:21
	→ gavin_huang joined	05:22
	← etcetera left	05:22
	→ theCole joined	05:25
	← zamn left	05:27
	→ arvind_khadri joined	05:28
	← gavin_huang left	05:28
	→ ipoval joined	05:29
	← shawngo left	05:29
	← noyb left	05:31
	→ noyb joined	05:32
	← arvind_khadri left	05:32
	→ arvind_khadri joined	05:33
	← dfr\|mac left	05:36
	← dvide left	05:36
	← arvind_khadri left	05:37
	→ arvind_khadri joined	05:37
	← landypro left	05:39
	← marc_kdab_ left	05:40
	→ marc_kdab_ joined	05:40
	→ landypro joined	05:41
	← arvind_khadri left	05:41
	→ rmr joined	05:42
	→ pfFredd joined	05:42
	← gnoel left	05:44
	← noyb left	05:44
	← russt left	05:45
	→ noyb joined	05:46
	← w09x left	05:46
	← fireh left	05:47
	← rmr left	05:47
	← Davey left	05:48
	→ etcetera joined	05:50
	← hwrd\|work left	05:51
	→ tewecske joined	05:52
	→ rmr joined	05:53
	← dang left	05:53
	← gtuckerkellogg left	05:54
	← tjadc left	05:54
	→ felipe_Brz joined	05:54
	→ tjadc joined	05:54
	→ dang joined	05:54
	→ s1n4 joined	05:55
felipe_Brz	what is the benefit of one using submodules rather than just initializing another git repo in a child directory of your main repo?	05:55
	→ api joined	05:55
	← dinesh6654 left	05:55
s1n4	I'm looking for an awesome code review system, does anyone know one?	05:55
	← aspotashev left	05:56
PovAddict	felipe_Brz: are you going to make a repository's .git metadata directory be just data committed into another repository?	05:56
	→ gavin_huang joined	05:56
	→ guns joined	05:56
felipe_Brz	is that too high an overhead?	05:56
PovAddict	no, it's just stupid	05:57
	you change a file, git-add it, change your mind and revert	05:57
	inner repo has no changes, but outer repo will see a new file in innerrepo/.git/objects/	05:58
felipe_Brz	i could add all children metadata to the parent repo's .gitignore?	05:58
PovAddict	define 'metadata'	05:59
	objects are certainly not metadata, that's where the repository data is :P	05:59
felipe_Brz	well all info git keeps to itself inside the .git directory	05:59
PovAddict	if you leave out the .git repo, you don't have a git repository	05:59
	er	05:59
	the .git dir	05:59
felipe_Brz	no, the .git directory is still there, just hidden from the parent repo	06:00
PovAddict	so the .git directory of the inner repo won't be added to the parent repo?	06:00
felipe_Brz	in other words, make the parent repo not know about each .git directory on each of its children	06:00
	yeah I guess	06:01
	← gavin_huang left	06:01
PovAddict	so now you just have two repositories with the same data that you commit at different times...	06:01
	→ arvind_khadri joined	06:02
felipe_Brz	yes but what if each of the inner repos are separate entities like classes	06:02
	→ lcc joined	06:02
	← lcc left	06:02
felipe_Brz	i could imagine myself updating one class	06:02
	only that one class	06:02
PovAddict	and you commit to the inner repo for that one class	06:02
	← stepnem left	06:02
felipe_Brz	yeah	06:03
	and revert/do whatever i want on that one class	06:03
PovAddict	and the outer repository still sees modified files in myclass/	06:03
felipe_Brz	yes	06:03
	← red82 left	06:03
felipe_Brz	is this too dumb? sorry i'm new to git	06:03
PovAddict	so you have two parallel and unrelated histories of the same data?	06:03
	so far it sounds like you don't need multiple repos at all :P	06:04
felipe_Brz	you are a developer? oo?	06:04
PovAddict	let's say you have just one repo	06:04
	you modify a class and you can make a commit with just the modification to that file (and not changes to other files that you may have)	06:05
	or before committing you think the change is bad, so you 'git checkout myclass.cpp' and revert that single file	06:05
	← v0n left	06:05
PovAddict	why do you think you need repositories inside others?	06:05
	→ gavin_huang joined	06:06
	→ lcc joined	06:06
	← blast_hardcheese left	06:07
	→ shreya joined	06:07
felipe_Brz	it's just that my web site has 4 clearly defined components	06:08
	← s1n4 left	06:08
felipe_Brz	and I thought that each one having their own repo would make it easier to track where each one is	06:08
	and each of them can vary independently of the other	06:09
	→ dinesh6654 joined	06:09
felipe_Brz	and they are at different places too	06:09
	each on a different server	06:09
	→ Davey joined	06:09
PovAddict	perhaps having them on separate repos is good then	06:09
	← ahayweh left	06:09
PovAddict	but in that case, do you really need a "parent repo" containing them?	06:10
felipe_Brz	i could go back and forth on each component's history withour affecting the others	06:10
	yeah, good point	06:10
	→ AmirBehzad joined	06:10
felipe_Brz	perhaps I don't	06:10
	but isn't this the exact kind of thing submodules are meant to address?	06:11
	varying components when inside a single project	06:11
	→ bubbles10101 joined	06:12
	← pfFredd left	06:14
	→ ahayweh joined	06:17
	← bubbles10101 left	06:17
	← etcetera left	06:18
	← varg left	06:19
	→ evanz joined	06:20
	← esa left	06:21
	← arvind_khadri left	06:23
	→ esa joined	06:24
	→ simesy joined	06:24
	← PigeonFriend left	06:25
	→ dc5ala joined	06:25
	→ Hasbro joined	06:27
	← rmr left	06:27
	→ varg joined	06:27
	→ kcj joined	06:28
	← simesy left	06:29
	← unreal left	06:29
	→ unreal joined	06:30
	← neurodrone left	06:35
	← varg left	06:35
	← esa left	06:35
	→ Dave^\| joined	06:40
	→ dstorrs joined	06:41
	→ varg joined	06:42
	← osmosis left	06:43
	→ esa joined	06:44
	← noyb left	06:44
	→ marthinal joined	06:45
	→ msmithng_ joined	06:46
	→ varg_ joined	06:47
	← varg left	06:48
	← angelsl left	06:48
	← msmithng left	06:48
	msmithng_ → msmithng	06:48
	← PovAddict left	06:49
	← madewokherd left	06:49
	← avpx left	06:51
	→ timthelion joined	06:52
	← imrobert left	06:52
	← towski left	06:53
	→ gtrak joined	06:53
	→ CpuID3 joined	06:54
	← Textmode left	06:55
	→ arvind_khadri joined	06:56
	← CpuID2 left	06:58
	→ rurufufuss joined	06:58
	→ simesy joined	06:58
	← tjadc left	06:58
	← arvind_khadri left	07:01
	← dinesh6654 left	07:02
	→ w09x joined	07:04
	← felipe_Brz left	07:05
	→ arvind_khadri joined	07:05
	→ reds joined	07:05
	reds → Guest59418	07:05
	← arvind_khadri left	07:06
	→ arvind_khadri joined	07:06
	→ bubbles10101 joined	07:07
	← arvind_khadri left	07:07
	→ arvind_khadri joined	07:07
	← eletuchy left	07:07
	→ tystr joined	07:08
	→ Textmode joined	07:09
	← dorkmafia left	07:09
	← Guest59418 left	07:10
	← subbyyy left	07:10
	→ arturaz joined	07:13
	← Textmode left	07:13
	→ Textmode joined	07:15
	→ juvenal joined	07:16
	← bubbles10101 left	07:16
	→ beneggett joined	07:17
	← gtrak left	07:17
	← stockholm left	07:17
	← maxwellcorbin left	07:19
	→ bubbles10101 joined	07:20
	← palm_vue left	07:20
	← bizhanMona left	07:20
	← Lgb left	07:20
	→ palm_vue joined	07:21
	→ Lgb joined	07:21
	→ bizhanMona joined	07:21
	← herdingcat left	07:21
	→ tvtue joined	07:23
	← Textmode left	07:25
tvtue	hello all	07:25
gitinfo	tvtue: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	07:25
tvtue	are there any known problems using different git versions for accessing the same central bare repo?	07:27
	!backup	07:28
gitinfo	Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	07:28
	→ skorgon joined	07:28
	← berndj left	07:29
	→ aspotashev joined	07:29
	→ carloshh joined	07:31
	← Chappo left	07:33
	→ herdingcat joined	07:33
	→ Chappo joined	07:35
	← varg_ left	07:35
	← w09x left	07:38
	→ Textmode joined	07:38
	← esa left	07:41
	→ esa joined	07:41
	← igotnolegs left	07:43
	→ varg joined	07:43
	← beneggett left	07:44
	← f0i left	07:44
	← BinGOs_ left	07:45
	→ BinGOs joined	07:45
	← Davey left	07:46
	← tvtue left	07:46
	← api left	07:46
	← jey left	07:49
	→ airborn joined	07:49
	← Transcended left	07:50
	← AmirBehzad left	07:51
	← Dave^\| left	07:51
	← carloshh left	07:51
	→ cybersphinx joined	07:54
	→ BiggFREE joined	07:55
	← Turicas left	07:57
	→ dorkmafia joined	07:59
	← simesy left	08:01
	→ pantsman joined	08:02
	→ w09x joined	08:03
	Droggl__ → Droggl	08:03
	← skorgon left	08:03
	→ private joined	08:03
	private → Guest34398	08:04
	← Textmode left	08:08
	← _huoxito left	08:10
	→ compojoom joined	08:10
	→ d3vic3 joined	08:11
	← esa left	08:11
	→ stepnem joined	08:12
	→ eletuchy joined	08:14
	→ simesy joined	08:14
	→ esa joined	08:15
	← juvenal left	08:15
	→ w41 joined	08:15
	← eletuchy left	08:18
	← simesy left	08:21
	→ Textmode joined	08:22
	→ skorgon joined	08:23
	→ Lord_DeathMatch joined	08:23
	→ _huoxito joined	08:23
	← arturaz left	08:24
	→ simesy joined	08:24
	← arvind_khadri left	08:24
	← tewecske left	08:26
	→ blacktulip joined	08:26
	← darren left	08:26
	→ dsirijus joined	08:26
	→ rtjure joined	08:28
	← simesy left	08:30
	→ leeN joined	08:31
	→ varg_ joined	08:31
	→ djh joined	08:31
	← _sundar_ left	08:31
	← jonn left	08:32
	← varg left	08:33
	→ f0i joined	08:34
	← dorkmafia left	08:34
	→ arvind_khadri joined	08:36
	← varg_ left	08:36
	← thirdknife left	08:39
	→ varg joined	08:39
	→ tktiddle joined	08:39
tktiddle	Does anyone know what might cause this error? "fatal: Unable to create temporary file: Permission denied"	08:40
	→ thirdknife joined	08:40
tktiddle	the bare repo is owned by git:git and i just tried "chmod g+w myrepo.git" but that didn't work	08:41
	shouls I add execute permission for the git group?	08:41
	→ simesy joined	08:41
	→ cesc joined	08:43
	← varg left	08:43
	← airborn left	08:48
	→ mastro joined	08:48
	← scientes left	08:50
tktiddle	OK got it	08:50
	→ angelsl joined	08:51
	← arvind_khadri left	08:51
	← herdingcat left	08:51
	→ txomon\|home joined	08:51
	→ varg joined	08:52
	← thirdknife left	08:52
	← JStoker left	08:53
	← Textmode left	08:56
	→ fayimora joined	08:57
	→ gtuckerkellogg joined	08:57
	→ nutron joined	08:57
	← Samos123 left	08:59
	→ Samos123_ joined	08:59
	← skyf left	08:59
	← fayimora left	09:01
	→ bloony joined	09:01
	→ shruggar joined	09:04
	← shreya left	09:05
	← tktiddle left	09:05
	→ reds joined	09:05
	reds → Guest13307	09:06
	← Guest34398 left	09:06
	→ jiangenj joined	09:07
	← teotwaki_ left	09:07
	gitinfo set mode: +v	09:07
jiangenj	Help needed: when I do git clone, remote seems broken:	09:08
	Cloning into adreno200...	09:08
	fatal: The remote end hung up unexpectedly	09:08
	fatal: early EOF	09:08
	fatal: index-pack failed	09:08
	→ cmn joined	09:08
jiangenj	what can do to fix it?	09:09
	← Guest13307 left	09:10
	→ Textmode joined	09:11
	→ EricKamsky joined	09:12
	← EricKamsky left	09:13
	→ dzonder joined	09:13
	→ nisseni joined	09:14
	→ SliMM joined	09:14
	← Textmode left	09:14
	← gtuckerkellogg left	09:16
	→ JStoker joined	09:16
	← Internet13 left	09:16
	← ahayweh left	09:16
	→ thiago joined	09:17
	← Droolio left	09:18
	← MissionCritical left	09:18
	← bubbles10101 left	09:18
	→ nifty joined	09:20
	← kcj left	09:20
jiangenj	Cloning into adreno200...	09:21
	fatal: The remote end hung up unexpectedly	09:21
	fatal: early EOF	09:21
	fatal: index-pack failed	09:21
	← BiggFREE left	09:22
	→ _iron joined	09:23
	→ arvind_khadri joined	09:25
	← aspotashev left	09:26
	← dsirijus left	09:27
	→ aspotashev joined	09:30
	→ ahayweh joined	09:31
	→ skyf joined	09:31
	→ MissionCritical joined	09:32
thiago	jiangenj: server problem. Check the URL and try again.	09:35
jiangenj	i can clone other projects	09:35
thiago	from the same server?	09:36
jiangenj	when i do git fsck --full --verbose on git server, it shows fatal: index file open failed: Too many open files	09:36
	← skyf left	09:36
jiangenj	no, from local	09:36
thiago	ulimit -n	09:36
	what's the number?	09:36
jiangenj	and on the server, I can clone by /home/git/project/path format	09:36
	1024	09:37
thiago	it's not especially low	09:37
jiangenj	possible too many objects in loose stored?	09:37
thiago	how about /proc/sys/fs/nr_open ?	09:37
jiangenj	1048576	09:38
thiago	the regular number	09:38
	finally: git count-objects -v	09:38
	what's the count?	09:38
	or just without -v	09:39
jiangenj	count: 1404	09:39
	size: 6856	09:39
	in-pack: 151819	09:39
	packs: 1097	09:39
	size-pack: 1238286	09:39
	prune-packable: 0	09:39
	garbage: 0	09:39
thiago	it's not especially high	09:39
	→ arturaz joined	09:39
thiago	there's no special reason why it would fail	09:39
	back up that repository first, then run: git repack -a -d -f	09:39
	you can leave -f out to speed it up	09:39
jiangenj	I git clone --mirror on server, git fsck --full --verbose return fatal: object aa19e00a1fdf78d89e4573c491cedb1caaf455c2 is corrupted	09:40
	→ sq-one joined	09:40
	← arvind_khadri left	09:40
	← aspotashev left	09:40
jiangenj	repacked with -f but returns fatal: object 20369b8eb8a79cdb0554f49758b4136249da1e1a is corrupted	09:41
	← Onkeltem left	09:42
	→ airborn joined	09:44
	→ aspotashev joined	09:44
	→ caseymcg joined	09:45
jiangenj	still fail, other possibility?	09:46
	→ Dimitr joined	09:47
	← aspotashev left	09:47
thiago	it's corrupted	09:48
	fsck can't recover the object	09:48
	← f0i left	09:50
	→ Textmode joined	09:50
	→ aspotashev joined	09:51
jiangenj	so ....	09:51
	→ nioncode joined	09:52
	← varg left	09:53
	← gavin_huang left	09:56
	→ thirdknife joined	09:56
	→ gavin_huang joined	09:58
	→ arietis joined	09:59
	← Textmode left	09:59
	→ compojoom_ joined	10:00
	→ rtjure_ joined	10:00
	→ irqq_ joined	10:00
	← Araxia left	10:00
	→ petteyg_ joined	10:02
	→ xiangfu joined	10:02
	→ nor5trom joined	10:02
	← nifty left	10:02
	→ rudi_s_ joined	10:03
	→ koivula_ joined	10:03
	→ jepler_ joined	10:04
	→ unreal_ joined	10:04
	→ dirker_ joined	10:04
	→ w0m_ joined	10:04
	← Chappo left	10:04
	→ kcj joined	10:04
	← sq-one left	10:05
	→ FernandoBasso joined	10:05
	← gavin_huang left	10:05
	← mklappstuhl left	10:05
	→ nifty_ joined	10:06
	→ schlumpi_ joined	10:07
	→ ivan`_ joined	10:07
	→ Glitch\| joined	10:08
	→ quadz_ joined	10:08
	← diegoviola left	10:08
	→ flexd1 joined	10:08
	← rtjure left	10:09
	← compojoom left	10:09
	← unreal left	10:09
	← irqq left	10:09
	← jepler left	10:09
	← ninive left	10:09
	← SunilThaha left	10:09
	← aep left	10:09
	← ivan` left	10:09
	← w0m left	10:09
	← petteyg left	10:09
	← Mpenzi left	10:09
	← quadz left	10:09
	← flexd left	10:09
	← koivula left	10:09
	← rudi_s left	10:09
	→ SunilThaha joined	10:09
	← boombatower left	10:09
	→ arvind_khadri joined	10:09
	→ urlwolf joined	10:09
jiangenj	hi still there? very strange, if using git clone git:// will fail, but if git clone http:// will succeed	10:10
	→ aep_ joined	10:10
	gitinfo set mode: +v	10:11
	→ diegoviola joined	10:11
	gitinfo set mode: +v	10:11
urlwolf	I can see that /docs in not tracked, but I cannot find it git-ignored anywhere. What could I do to force-add to the stage?	10:11
	→ codeFiend joined	10:11
	← codeFiend left	10:11
FauxFaux	urlwolf: git add --force, perhaps?	10:11
	jiangenj: Dumb http clone has probably cloned the corruption without noticing ('cos he's a cool guy); git fsck the clone?	10:12
urlwolf	FauxFaux: nope, doesn't work	10:12
	← airborn left	10:13
	→ Mpenzi joined	10:14
urlwolf	it doesn't even appear in untracked files	10:14
jiangenj	fauxFaux: on remote server, git fsck shows corrupted objects	10:15
	no idea how to fix	10:15
	and why the corruption occurs?	10:16
	← shruggar left	10:17
	→ TomyLobo joined	10:17
cbreak	broken hardware	10:17
	crashes	10:17
	malicious manipulation	10:17
	→ mklappstuhl joined	10:18
jiangenj	oops	10:18
	have to check with other admin if they got good backup	10:19
	→ airborn joined	10:19
cbreak	you have an other clone, right?	10:19
	of course you do	10:19
jiangenj	but not updated	10:20
cbreak	a server repository is bare after all	10:20
	you still have the one that you committed in	10:20
	and pushed to the server	10:20
jiangenj	no, it's script generated, and already been deleted	10:21
	← rtjure_ left	10:21
	← diegoviola left	10:21
jiangenj	but if no other solution, for sure need this way to recovery as the commits after that old repo is on gerrit	10:21
cbreak	sounds dumb.	10:21
	why the hell would you do that?	10:22
jiangenj	you meant?	10:22
	→ varg joined	10:22
cbreak	why would you delete a git repository after pushing?	10:22
jiangenj	we cherry-pick commits from other site automatically, and then pushing to our gerrit server	10:23
cbreak	sounds dumb. automated cherry-picking ...	10:23
	don't automate things that are designed to fail.	10:23
jiangenj	well, we have to do this way, one person cannot do merge/cherry-pick of hundres one day	10:23
	so what you normally do?	10:24
	→ sedavand_ joined	10:24
	→ thelazydeveloper joined	10:25
	← arvind_khadri left	10:27
	← urlwolf left	10:27
	→ shreya joined	10:29
cbreak	jiangenj: more people	10:29
	→ macIvy joined	10:30
	← jackh left	10:30
cbreak	also, merging hundreds of commits is easy	10:30
	you can merge hundreds of commits in one operation just by merging a branch with them	10:30
jkitchen	why not just have those sites submit directly to gerrit?	10:30
	→ urlwolf joined	10:33
	→ Element9 joined	10:34
	← flavius left	10:34
	← aspotashev left	10:35
	→ r0bby joined	10:37
	← flexd1 left	10:38
	→ flexd joined	10:39
	→ aspotashev joined	10:39
	→ AL\|EN joined	10:40
	← caseymcg left	10:40
	← skorgon left	10:42
jiangenj	well, it's inside group but different	10:42
	contributors	10:42
	→ Chappo joined	10:42
	→ gavin_huang joined	10:42
	→ Tevfxds32 joined	10:43
	gitinfo set mode: +v	10:43
	gitinfo set mode: +v	10:43
	← varg left	10:43
	← Tevfxds32 left	10:43
AL\|EN	hi	10:43
gitinfo	AL\|EN: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	10:43
AL\|EN	!backup	10:44
gitinfo	Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	10:44
AL\|EN	well i have simple question	10:44
	whare i can send a feature request	10:44
	a simple one	10:44
	for the git svn info command	10:44
	?	10:44
	→ f0i joined	10:45
grawity	My guess would be the mailing list?	10:45
AL\|EN	http://vger.kernel.org/vger-lists.html#git	10:45
	this ?	10:45
	← gavin_huang left	10:47
	← ipoval left	10:47
	→ oriba joined	10:47
	→ flavius joined	10:49
	← guns left	10:50
	← sedavand_ left	10:53
	→ sedavand joined	10:55
	← arturaz left	11:02
	→ gavin_huang joined	11:02
	→ shruggar joined	11:02
	← txomon\|home left	11:02
	→ diegoviola joined	11:04
	→ reds joined	11:06
	reds → Guest69013	11:06
	← gavin_huang left	11:07
	→ varg joined	11:07
	← Guest69013 left	11:10
	← shruggar left	11:10
	→ gavin_huang joined	11:10
	→ tqmz joined	11:12
	→ Internet13 joined	11:13
	→ babilen joined	11:14
	← gavin_huang left	11:15
	← AL\|EN left	11:16
	→ txomon\|home joined	11:17
	→ juvenal joined	11:17
	→ gavin_huang joined	11:19
	→ oxidative joined	11:19
	→ unterstrom joined	11:19
	→ elias joined	11:23
	→ shruggar joined	11:23
	→ Droolio joined	11:24
	← gavin_huang left	11:24
	← nioncode left	11:24
	→ gavin_huang joined	11:26
	→ skorgon joined	11:27
	← elias left	11:27
	→ elias joined	11:27
elias	where could i find a post-receive hook which sends one email for each commit? (the default email hook seems to send one email with all commits)	11:28
cbreak	you can just modify it	11:28
	post-receive gets its data on stdin	11:28
elias	true, just wondering if someone already did	11:28
	→ herdingcat joined	11:29
	→ w09x_ joined	11:29
	→ dr_lepper joined	11:29
cbreak	it is a bit tricky	11:29
	if you want to send an email for each commit, do you only want to do that for each new commit?	11:30
	→ noscript joined	11:30
cbreak	for example, if someone merges a branch and pushes that, would you want to send an email for each commit he merged?	11:30
	if someone pushes two branches with the same commits, do you want to send two emails for each commit?	11:30
	← d3vic3 left	11:30
	→ sedavand_ joined	11:31
cbreak	I ended up doing something like git rev-list <commitsha1> <commitsha2> --not --all in a pre-receive hook when I wanted to post messages, one for each commit	11:31
elias	yeah. it's how we had it with svn... it was a bit annoying getting 100 mails on each merge - but still nice to have one mail for each commit in the normal case	11:31
cbreak	that way I only get commits that are completely new	11:31
	← sedavand left	11:32
	→ dinesh6654 joined	11:32
cbreak	merges give no messages if they are ff, and one if they are not	11:32
	← w09x left	11:32
cbreak	<commitsha1> are the new branch tips	11:32
	→ d3vic3 joined	11:32
cbreak	I use pre-receive and not post-receive since in pre-receive, the local branch heads are not yet changed	11:32
	in post-receive, they are	11:33
	post-receive means post-update, unfortunately	11:33
	← jiangenj left	11:34
	← shruggar left	11:35
	→ SkAzZ joined	11:35
elias	cbreak: thanks, I see... just have to figure out how to run multiple pre-receive then in a way so the mail only gets sent if the others succeed	11:37
cbreak	why multiple?	11:38
	it gets run once per push	11:38
elias	yeah, i meant another pre-receive already there	11:39
	← d3vic3 left	11:39
	← varg left	11:39
cbreak	wonders of bash scripting	11:39
	← macIvy left	11:39
	→ d3vic3 joined	11:39
cbreak	bash firstscript && bash secondscript && bash thirdscript	11:39
	maybe even a return of that	11:39
elias	ok, so that's the easy part :)	11:39
	→ skorgon_ joined	11:40
	→ marhaban joined	11:40
	← diegoviola left	11:41
	→ jjbohn joined	11:42
	→ storrgie joined	11:43
	→ simplechat joined	11:43
	← simplechat left	11:43
	→ simplechat joined	11:43
	← skorgon left	11:43
	→ varg joined	11:45
	← blacktulip left	11:48
	→ blacktulip joined	11:48
	← dinesh6654 left	11:53
	← aspotashev left	11:53
	← thirdknife left	11:53
	→ codeM0nK3Y joined	11:56
	← urlwolf left	11:57
	→ thirdknife joined	11:57
	→ urlwolf joined	11:57
multi_io	I'm trying to track a directory with git, including uid/gid/mode/symlink information. Does anyone have some advice on how to do that? Is etckeeper an option?	11:59
	→ sbfds1 joined	12:00
	← quazimodo left	12:00
cbreak	multi_io: git on its own is not an option	12:01
	it does not track any file meta data apart from executability (and wether it is a symlink or not)	12:01
	also, tracking uid/gid sounds stupid	12:01
	how would it restore that?	12:01
	can you even create files with a different UID other than your own without being root?	12:02
	→ yshah joined	12:03
	skorgon_ → skorgon	12:03
	→ Speed joined	12:04
	← leeN left	12:04
	→ leeN joined	12:05
	gitinfo set mode: +v	12:05
	← irqq_ left	12:06
	← varg left	12:06
yshah	quick question to mac lion users. Does the current mac git installer work for the Lion OS?	12:06
	← cesc left	12:06
cbreak	you can just install xcode	12:07
	it comes with git	12:07
	or you compile git from source which also works	12:07
yshah	will Xcode have the latest git version?	12:08
	← storrgie left	12:08
cbreak	no	12:08
	but one sufficiently late	12:08
	git version 1.7.9.6 (Apple Git-31.1)	12:09
yshah	fair enough, thank you cbreak.	12:09
cbreak	of course, xcode is huge :)	12:09
	but it's nice to have anyway	12:09
	(don't forget to install the cli tools)	12:10
	← zackiv31 left	12:10
	→ psoo joined	12:12
	→ zackiv31 joined	12:13
	← Chappo left	12:14
	→ Chappo joined	12:15
	← yshah left	12:19
	→ longbeach joined	12:19
	← marthinal left	12:22
	→ storrgie joined	12:22
	← simplechat left	12:23
	← Internet13 left	12:23
multi_io	cbreak: I would run it as root (this is for tracking config data outside /etc).	12:25
	→ Internet13 joined	12:26
multi_io	cbreak: and I guess the uids/modes etc. would have to be tracked in a separate versioned file and be captured/restored from there before commits / after checkouts	12:26
	(by some script that's invoked via a hook or so)	12:27
cbreak	yeah. You could do hackery like that	12:27
	but be aware that UIDs are rather pointless to track	12:27
	they have different meanings on each system	12:28
	← urlwolf left	12:29
	← theCole left	12:32
	← psoo left	12:34
	← zackiv31 left	12:35
	→ psoo joined	12:36
grawity	Then track usernames?	12:36
cbreak	they also have different meanings on each system.	12:36
	my root is someone else than your root	12:36
	despite them having the same uid and name	12:37
	← rue left	12:38
	← Mpenzi left	12:40
	→ Mpenzi joined	12:42
	← pilgrum left	12:43
	← gavin_huang left	12:45
multi_io	yeah, I'd track usernames, that's good enough for me I guess. This is just for having a complete record of all my config changes.	12:47
	← thirdknife left	12:47
multi_io	etckeeper already does that sort of thing for the /etc directory, but I was having problems using it elsewhere.	12:47
	→ gavin_huang joined	12:50
	→ Fandekasp joined	12:51
	→ archdori joined	12:51
	→ alberto56 joined	12:52
	← moshe742 left	12:54
	← d3vic3 left	12:54
	→ andrewhl joined	12:56
	→ jasmina87SA__ joined	12:56
	← jasmina87SA__ left	12:56
	← gavin_huang left	12:57
	← sbfds1 left	12:58
	← oriba left	13:00
	→ gavin_huang joined	13:00
	→ gnoel joined	13:02
	← lcc left	13:02
	→ metasansana joined	13:05
	→ stkrzysiak joined	13:05
	← gavin_huang left	13:05
	← agile left	13:05
	← metasansana left	13:06
	→ reds joined	13:06
	→ pw1 joined	13:07
	reds → Guest91642	13:07
	→ gavin_huang joined	13:08
	← longbeach left	13:08
	→ rue joined	13:09
	← Guest91642 left	13:11
	← gavin_huang left	13:12
	← rudi_s_ left	13:15
	→ rudi_s joined	13:15
	→ gavin_huang joined	13:16
	← palm_vue left	13:20
	← bizhanMona left	13:20
	← Lgb left	13:20
	→ palm_vue joined	13:20
	→ Lgb joined	13:21
	→ bizhanMona joined	13:21
	→ shruggar joined	13:21
	→ ciscon joined	13:23
	→ guampa joined	13:24
	← xiangfu left	13:25
	→ xiangfu joined	13:27
	→ cwang joined	13:27
	→ Yuuhi joined	13:27
	← psoo left	13:31
	→ metasansana joined	13:31
	← noscript left	13:33
	← SliMM left	13:33
	→ gtrak joined	13:35
	jepler_ → jepler	13:35
	← kcj left	13:36
	→ stkrzysiak_ joined	13:38
	← mastro left	13:38
	← stkrzysiak left	13:39
	stkrzysiak_ → stkrzysiak	13:39
	→ lcc joined	13:39
	→ Onkeltem joined	13:39
	gitinfo set mode: +v	13:39
Onkeltem	Hi	13:40
	I'm reparing a patch with git format-patch. How to minimize patch to remove all intermediate edits?	13:40
	I don't want history in it but it is there now	13:41
Bigcheese	You mean format patch is outputting multiple patches?	13:41
canton7	Onkeltem, you probably want to squash first. or use git diff to produce the path	13:42
Onkeltem	Bigcheese: yes. In my branch I edited file twice. Now when I do: git format-patch origin/master I get my two commits to that branch	13:43
Bigcheese	Yeah, you want what canton7 said.	13:43
Onkeltem	Ok, squash... what is that? :)	13:43
rue	You could rebase them into a single commit	13:43
Bigcheese	Although I forget the exact syntax for dif...	13:43
	diff*	13:43
canton7	it's where you take several commits and make a single commit out of them. there are several ways to do this, but rebase -i is probably the most famous	13:44
Bigcheese	git diff HEAD..firstcommit	13:44
	Or something.	13:44
canton7	that's a backwards diff I think	13:44
	→ stkrzysiak_ joined	13:44
	← stkrzysiak_ left	13:44
grawity	should be master.. (or master..HEAD), yes	13:45
Onkeltem	rue: I read about it, ok! Just need some hints...	13:45
	→ gtuckerkellogg joined	13:45
Onkeltem	1) I did: git reset --hard origin/master then 2) git merge mybranch	13:45
	Should I just change merge to rebase?	13:45
	← stkrzysiak left	13:46
Onkeltem	worth to try it now	13:46
grawity	checkout yourbranch && rebase -i master	13:46
canton7	I have no idea what you're trying to do. why did you reset --hard?	13:46
	← jcromartie left	13:47
Onkeltem	canton7: to get origin without my modifications	13:47
	→ jcromartie joined	13:47
	← f0i left	13:48
	← storrgie left	13:48
Onkeltem	grawity: heh, I'm very new to git, so can ask stupid things but... what is the difference between 1) "standing" on my branch and doing like you said: rebase -i master and 2) "standing" on master and doing: rebase mybranch?	13:49
	(my local master is tracking origin/master)	13:49
	(and I have no write access - just making patches)	13:49
	→ agileadam joined	13:49
canton7	git checkout branch && git rebase master <-- takes branch and rebases it onto master. git checkout master && git rebase branch <-- takes master and rebases it onto branch	13:50
	← txomon\|home left	13:50
	→ AdrienBrault joined	13:51
	→ gry joined	13:52
	← _huoxito left	13:53
	← esa left	13:54
Onkeltem	I did git checkout mybranch, git rebase -i master. Now I get some kind of action file in Nano editor. Having three picks. How would I make minimal patch now? Should change pick to squash or what?	13:54
canton7	kepe the first pick, but change the second 2 to squash/fixup	13:54
	*keep	13:55
	→ esa joined	13:55
	← flavius left	13:55
	→ iamjarvo joined	13:55
	← rook2pawn left	13:55
	← gavin_huang left	13:56
	→ psoo joined	13:57
Onkeltem	Thank you all! Now everything is like I wished	13:59
	← gtrak left	14:00
Onkeltem	canton7, grawity, rue, Bigcheese: t.y.	14:00
	Onkeltem starts loving git more and more	14:00
	Onkeltem and thinks about migrating all his projects to it from Mercurial	14:01
	→ longbeach joined	14:01
	← Droolio left	14:01
	← landypro left	14:02
canton7	do it! do it!	14:02
	heh, just out of curiosity, what do you prefer about git over hg?	14:02
grawity	tig	14:03
	how the hell do hg/bzr users live without something like tig.	14:03
	← juvenal left	14:03
cbreak	tig?	14:03
canton7	curses gitk iirc	14:03
Onkeltem	canton7: Git Book :) It makes you feel git is the only option	14:04
	← gtuckerkellogg left	14:04
	→ Chillance joined	14:04
	→ gavin_huang joined	14:04
canton7	haha! nic	14:04
	*nice	14:04
Onkeltem	canton7: like how would i live now knowing all all these features!	14:04
	heh	14:04
	← cwang left	14:05
Onkeltem	What I can't decide for now - is how to properly organize a big .git project (like a Drupal website) with many .git subprojects (modules).	14:06
	→ arturaz_ joined	14:06
	→ vpetro joined	14:07
	← john3 left	14:07
	→ flavius joined	14:08
	→ dleonardi joined	14:08
AAA_awright	What happened to "Why Git is better than X"?	14:08
	← rchavik left	14:09
	→ john3 joined	14:09
Onkeltem	lol, how VCS could be better then source?	14:09
	← skorgon left	14:10
	← herdingcat left	14:10
Onkeltem	its like a bus stop better then busses?	14:10
	→ reactormonk joined	14:10
	→ cakehero joined	14:11
	→ AaronMT joined	14:12
	→ acalbaza joined	14:13
	← jjbohn left	14:14
	← simesy left	14:14
acalbaza	hi, quick question about git-svn... im trying to clone a remote branch given a revision and that works as expected, but i also want to track remote branches in the same repo. when i do a git svn fetch, shouldnt i see all the remote branches?! how do i teach git about the remote branches after a clone using a revision? i cant pull down the history as its way too big.	14:15
	← Chappo left	14:16
	← lcc left	14:17
acalbaza	i feel like git branch -r should show me all remote branches created after the revision given to clone	14:17
	← gnoel left	14:17
	→ dfr\|mac joined	14:17
acalbaza	but i just get "git-svn" after git branch -r	14:17
cbreak	acalbaza: git svn clone clones all branches	14:17
	acalbaza: run git svn info and pastebin the result. Also pastebin the content of .git/config	14:18
	← flavius left	14:18
	← Pupeno left	14:20
	→ jjbohn joined	14:21
	← psoo left	14:21
	← vpetro left	14:22
	→ simesy joined	14:22
acalbaza	cbreak: http://pastebin.com/LUXVphK3	14:23
	→ psoo joined	14:23
cbreak	well, the clone is already wrong	14:23
	with git you don't clone branches	14:23
	you clone repositories	14:23
	the repository you cloned will always ever only track that svn repository path	14:23
	→ juvenal joined	14:24
	→ SamB_MacG5 joined	14:24
acalbaza	cbreak: got it	14:24
cbreak	you-ll have to do something like	14:25
	git svn clone url://server/name/ -s	14:25
	if name/trunk, name/branches and name/tags exist	14:25
	for trunk, other branches and tags respective	14:25
	→ Dimitr\|2 joined	14:26
acalbaza	cbreak: and i should still be able to add the -r	14:26
cbreak	I wouldn't	14:27
	it'll cut off history	14:27
	← arturaz_ left	14:28
	→ arturaz__ joined	14:28
	→ RaptorX joined	14:28
	← RaptorX left	14:28
	← Dimitr left	14:28
	← madsy left	14:29
	← Internet13 left	14:29
acalbaza	cbreak: ok, thanks. let me play with this a little...	14:29
	→ tjadc joined	14:29
	→ madewokherd joined	14:31
	→ flavius joined	14:32
	→ Chappo joined	14:33
	→ leotr joined	14:34
	← longbeach left	14:34
acalbaza	ok, so i'm probably running into "Couldn't find revmap for trunk" since i tried clone the repo using a revision...	14:35
	← tjadc left	14:36
	→ vitor-br joined	14:36
	→ Internet13 joined	14:36
	← psoo left	14:37
	→ tjadc joined	14:38
acalbaza	seeing if i can get away with just "enough" history	14:38
	→ kukks joined	14:38
	→ psoo joined	14:38
	→ circlcious joined	14:40
	→ Anders_J joined	14:40
circlcious	my team made messed up codebase and pushed. i pulled it. now i wanted to go back in commits so i tried git rebase -i commitsha1, bu i think i have messed up things, what should i do ?	14:40
	← mrwalker left	14:41
	← dleonardi left	14:42
leotr	hi! i accidentally pushed to remote repo master branch. I tried git reset --hard old_rev && git push -f, couldn't do it. What will happen with my last commits if i do git reset --soft on remote repo? Will they disappear?	14:42
	→ toabctl joined	14:42
	→ raminoid joined	14:44
	← alberto56 left	14:44
cbreak	circlcious: git rebase is MOST DEFINITELY not the right command to go back	14:45
	do you want to go back permanently? Then git reset --hard to nuke away any changes	14:45
	← gavin_huang left	14:45
cbreak	reset to some commit in the history, or the reflog	14:45
	→ Goplat joined	14:46
circlcious	cbreak: ok i reset	14:47
	but how do i do the same on remote repo :( i can tpush	14:47
	→ gavin_huang joined	14:47
circlcious	it says me to pull, if i do that its going to pull the commits again	14:47
	← raminoid left	14:48
	→ arturaz_ joined	14:48
	→ varg joined	14:48
circlcious	cbreak: ???	14:49
cbreak	what?	14:49
	→ skyf joined	14:49
circlcious	i reset on my local repo to the commit which was good	14:49
	how to do the same on remote repo ?	14:49
cbreak	the same way.	14:49
circlcious	but remote repo is bare	14:50
cbreak	you can reset it locally with reset anyway	14:50
	or you can force push if you have the permissions	14:50
	← arturaz__ left	14:50
cbreak	you'll also have to do it on every single repository that fetched from there	14:50
circlcious	:S	14:50
	no other way?	14:50
cbreak	nope.	14:50
	every single one of them	14:50
	you're rewriting history after all	14:51
circlcious	so anyone who does something weird and pushes, to revert that its a hard process like this ? :/	14:51
cbreak	no	14:51
	you're not reverting	14:51
	you're rewriting history	14:51
	if you wanted to revert, you would just run git revert	14:52
	→ abletony84 joined	14:52
circlcious	actually i did a git revert after git reset	14:52
	how would i revert on serber too?	14:52
cbreak	that would naturally not get you back to the old commit	14:52
circlcious	i mean remote repo	14:52
abletony84	Hey! How do I "git add ." but at the same time automatically delete all the files/folders that have been deleted?	14:52
	← xiangfu left	14:52
cbreak	after the reset? then revert is absolutely pointless	14:52
circlcious	how would i revert to the proper commit, i.e., the commit before the messy one ?	14:53
cbreak	if you reset, then you've rewritten history.	14:53
circlcious	ok	14:53
cbreak	circlcious: first: get back to the messy one	14:53
	← marhaban left	14:53
cbreak	then, revert the messy one.	14:53
	then you are one commit after the messy one	14:53
circlcious	$ git revert messy_commit_sha1 ?	14:53
cbreak	revert does NOT get you back to old commits	14:53
	it just gets you into the state without a commit	14:53
	← nisseni left	14:54
cbreak	you said you wanted to go back to an old commit so revert won't help you	14:54
circlcious	i just wanted to undo my team mate's commit	14:54
	i pulled his changes and its a mess	14:54
cbreak	well, then revert it and push :)	14:54
	if your pull was fast forward that is	14:54
circlcious	ok but since i did a reset already	14:54
	m,aybe i will pull again and doa revert now?	14:55
cbreak	undo the reset	14:55
	you can git pull --ff-only	14:55
	if that works, then you'll have the original I guess	14:55
abletony84	Sorry guys but how do I "git add ." but at the same time automatically delete all the files/folders that have been deleted?	14:55
cbreak	abletony84: not at all	14:55
	abletony84: just do git add -A	14:55
abletony84	How come?	14:56
	oh cool	14:56
cbreak	because . has different semantics	14:56
jnewt	how do i get a git log that shows the complete changesets?	14:56
cbreak	jnewt: git log -p ?	14:56
circlcious	not possible to fast forward, aborting	14:56
cbreak	what do you mean with changeset?	14:56
circlcious	i will do a force push for now	14:56
cbreak	circlcious: that must be because you reverted	14:56
	reset the revert away and try again	14:56
	(I thinjk you've messed the repo up more than your team by now :)	14:57
circlcious	lol	14:57
	i force pushed anywya	14:57
jnewt	circlcious: like --summary, but summary only seems to show create / deletes. i want to see modified file list too	14:57
circlcious	maybe that rewrote the history but ok	14:57
	how do you revert reverts anyway ?	14:57
cbreak	not maybe	14:57
circlcious	i mean undo revert	14:57
cbreak	most definitely	14:57
circlcious	cbreak: :P	14:57
cbreak	every non-fast-forward push rewrites history	14:57
jnewt	cbreak, not a diff, only a modified / create / delete list	14:58
	← simesy left	14:58
cbreak	jnewt: git log --status or git log --num-stat	14:58
	→ neurodrone joined	14:58
	← neurodrone left	14:58
	→ neurodrone joined	14:58
cbreak	jnewt: also git log --name-only if you want just names	14:58
	jnewt: read man git-log for more	14:58
gitinfo	jnewt: the git-log manpage is available at http://git-scm.com/docs/git-log.html	14:58
circlcious	cbreak: git reset --hard HEAD^ undo's the revert?	14:59
	→ f0i joined	14:59
circlcious	i guess git revert creates a new commit which has reverted everything from the current/last commit	14:59
cbreak	circlcious: if the revert is HEAD, yes	14:59
	no	14:59
	git revert creates a new commit that reverts what you tell it to	14:59
	git revert 123456 reverts commit 123456's changes	14:59
circlcious	yah right	14:59
	should have done that then	14:59
	→ xiangfu joined	15:01
jnewt	cbreak: i'm aware of the man page. i've been trying to figure out the proper command from that, but i can't seem to get it. summary and num-stat aren't switches for git log	15:01
cbreak	they are	15:01
	← varg left	15:01
cbreak	git log --stat, git log --num-stat	15:01
	git log --name-only	15:01
	all work	15:01
	--name-status I mean :)	15:02
	just read the man page, it has all of them	15:02
	← kpreid left	15:02
jnewt	neither are what i want.	15:03
	→ kpreid joined	15:03
	→ Mogget joined	15:04
	→ ISF joined	15:04
cbreak	what do you want?	15:04
jnewt	I want to output a condensed summary of extended header information such as creations, renames, mode changes AND file modifications.	15:05
cbreak	what's an extended header?	15:05
	← xiangfu left	15:05
jnewt	read the man page under --summary. that's it's terminology	15:05
cbreak	--stat and --name-status both show renames, changes, deletes, ...	15:06
	→ reds joined	15:07
	reds → Guest10793	15:07
jnewt	yeah, but instead of create mode file.c, you get shit like file.c Bin 0 -> 1684354bytes	15:07
	← juvenal left	15:08
jnewt	oh, you're right, name-status does it	15:08
	thank you!	15:08
cbreak	np	15:09
jnewt	i have no idea what they mean by extended header though, can't help there.	15:10
	but i dont' really care. so back to work. later, thanks again	15:10
	← Guest10793 left	15:12
acalbaza	so here is a question, can i skip fetching tags during an svn clone?	15:13
cbreak	don't specify a --tags? :)	15:15
acalbaza	ahh, i used -s	15:16
	so if i specify branches and trunk, tags will ignore?	15:16
cbreak	maybe.	15:18
acalbaza	let me try... either that or ignore paths.	15:19
	→ Araxia joined	15:20
	→ gnoel joined	15:21
	← letharion left	15:22
	← Dimitr\|2 left	15:25
	→ chuckharmston joined	15:25
	→ letharion joined	15:25
	← jjbohn left	15:25
	→ Dimitr joined	15:25
	← Learath2 left	15:28
	→ xiangfu joined	15:29
	→ chinopozo joined	15:30
	← kiffer84 left	15:33
	→ AmirBehzad joined	15:33
	→ Davey joined	15:33
	→ Droolio joined	15:33
	← xiangfu left	15:35
	← FernandoBasso left	15:35
	← AdrienBrault left	15:36
	← AmirBehzad left	15:36
	→ tewecske joined	15:36
	→ subbyyy joined	15:36
	→ vibhav joined	15:37
vibhav	How do I contribute to the git project?	15:37
	→ red82 joined	15:37
	→ madsy joined	15:38
	← madsy left	15:38
	→ madsy joined	15:38
	→ velikasha joined	15:40
	← gnoel left	15:40
	→ mrwalker joined	15:41
	→ jjbohn joined	15:43
	← RaceCondition left	15:44
	← AaronMT left	15:46
	→ AaronMT joined	15:46
	→ xiangfu joined	15:47
	← Droolio left	15:47
	← dang left	15:48
	← skyf left	15:49
	→ Droolio joined	15:49
	→ urlwolf joined	15:50
	← dc5ala left	15:51
	← spyro left	15:52
	→ spyro joined	15:52
	← xiangfu left	15:54
loops	vibhav, start off by joining the git mailing list	15:54
	vibhav, you can submit patches there for review	15:54
	← iamjarvo left	15:55
vibhav	loops: I know C, which parts of git can I start with?	15:55
	→ branduren joined	15:56
	→ ipoval joined	15:56
loops	vibhav, do something you care about. scratch your own itch	15:56
	vibhav, if you just want to get your feet wet, check out https://git.wiki.kernel.org/index.php/Janitor	15:57
	toothr → toothrot	15:58
	→ skorgon joined	15:58
	→ yoklov joined	15:59
	← circlcious left	16:01
	→ lausi joined	16:03
	→ tete_ joined	16:04
	→ sh0ne joined	16:05
	← urlwolf left	16:05
	→ nyx joined	16:05
	→ beneggett joined	16:06
	← nyx left	16:07
	← Lord_DeathMatch left	16:07
	→ nyx joined	16:07
	→ FatDarrel joined	16:08
	← Droolio left	16:11
	→ Drool joined	16:11
	→ fission6 joined	16:11
fission6	i did git pull, and got a new remote branch, how do i merge to master	16:11
	← mrwalker left	16:12
cbreak	merge it.	16:12
tete_	i use a gui (tortoisegit) for such stuff fission6	16:12
	just switch to master and click on merge branch, then choose the branch and thats it	16:12
	→ savid joined	16:12
fission6	cbreak: i tried that	16:13
	← jargon- left	16:13
cbreak	fission6: so?	16:13
	→ orrc joined	16:14
cbreak	while on master, git merge origin/otherbranch	16:14
fission6	cbreak: i am in master, i see this branchi want to merge in remotes/origin/feature/issue	16:14
cbreak	done.	16:14
fission6	fatal: 'origin/feature' does not point to a commit	16:14
cbreak	use your brain...	16:14
	you'll have to use the name of an existing branch of course	16:14
	you can't just invent one	16:15
fission6	ok got it going and got CONFLICT (content): Merge conflict in geospatial/api/GeoIP-1.4.8/config.log	16:15
	how do i just take "mine" and complete the merge	16:16
cbreak	git checkout --ours on the file	16:16
	then check git status, and git commit	16:16
	also, what kind of idiot committed a config .lg in the first place?	16:16
fission6	agreed	16:16
cbreak	you're better off removing it from the tracking	16:16
fission6	on the lstnote	16:16
	i agree	16:16
	it will be going into .gitignore	16:16
	← metasansana left	16:16
cbreak	git rm --cached the file	16:17
	← beneggett left	16:17
	→ unterstrom_ joined	16:17
	← bartman left	16:18
	← Davey left	16:18
	← unterstrom left	16:19
fission6	will do.	16:19
	→ Droolio joined	16:21
flavius	is there anything similar to git config --local --add remotes.<remote-group> <remote> but for grouping branches?	16:22
	→ cesc joined	16:22
	← tjadc left	16:22
fission6	cbreak: if i do git rm --cached and commit that, will it delete the file from soneone who pulls it from disk	16:22
cbreak	yes	16:22
	unless he has local modifications	16:22
	in which case there will be a conflict	16:22
fission6	ok	16:22
	→ andrewhl_ joined	16:23
flavius	fission6: they should merge in a sepparate branch "designed" for merges, not directly into feature/branch or master or similar	16:23
	← Drool left	16:23
flavius	merge, run tests, then merge again in the parent branch	16:24
	→ dinesh6654 joined	16:24
	← orrc left	16:24
flavius	(bisect helps here, btw, as a little extra)	16:24
	← Bearcode left	16:25
	→ pppaul joined	16:25
flavius	does anyone have a clue to my question?	16:25
	→ bartman joined	16:25
	→ darren joined	16:26
	← andrewhl left	16:27
cbreak	what would the semantics of grouping branches be?	16:27
	← yoklov left	16:27
tango_	name them as commonstring/branchname	16:28
fission6	weird, i am adding a file to gitignore and it keeps showing up in my changes for commit	16:28
	→ palyboy joined	16:28
tango_	fission6: if the file is already in the index, it will not be ignored	16:29
	→ yoklov joined	16:29
	← lonewulf` left	16:29
fission6	right, but how do i remove from being tracked and dont have it delete when someone pulls?	16:29
	→ varg joined	16:30
cbreak	finch: as I told you	16:30
	you have to untrack the file	16:30
	→ lonewulf` joined	16:30
	→ jargon- joined	16:30
cbreak	fission6: git tracks state. if a file vanishes, it was deleted.	16:30
	if someone merges a delete, the result should be obvious	16:31
flavius	tango_: yeah I do that already, but I was wondering	16:31
	cbreak: for instance feature branches, bugfix branches, etc :)	16:31
	pull-requests! how could I forget that :D	16:31
cbreak	flavius: what?	16:31
	← nyx left	16:32
	← nifty_ left	16:32
	→ juvenal joined	16:32
flavius	cbreak: what what?	16:32
cbreak	flavius: why did you write that?	16:32
	was it an answer to something I said?	16:33
flavius	because you asked "cbreak \| what would the semantics of grouping branches be?"	16:33
cbreak	semantics would be:	16:33
	if I check out a branch, it checks out all branches	16:33
	or if I delete a branch, it deletes all branches	16:33
	that's semantics	16:33
	how they act	16:33
flavius	yeah	16:33
cbreak	I want to know how you want the group to act	16:33
flavius	if you want to fetch all pull requests for instance	16:33
cbreak	because git has nothing that allows that	16:33
	try a custom fetch spec	16:34
flavius	a hook?	16:34
	← setmeaway left	16:34
cbreak	no	16:34
	in your config file, look for lines like fetch = +refs/heads/:refs/remotes/origin/	16:34
	they tell git what to fetch if you fetch a whole remote	16:35
	you can change that	16:35
flavius	ok thanks	16:35
cbreak	you can add new remotes too	16:35
flavius	I'll look it up, thanks :)	16:35
	finch waves to cbreak	16:35
finch	I was very confused why I was highlighted for a moment :)	16:35
	← subbyyy left	16:36
cbreak	first letter confusion :/	16:36
	→ Davey joined	16:36
	→ stkrzysiak joined	16:36
	bryanmv`away → bryanmv	16:36
finch	no worries	16:36
	→ pfFredd joined	16:36
	→ maletor joined	16:37
	← codeM0nK3Y left	16:38
	→ BadDesign joined	16:40
	→ noyb joined	16:40
	→ setmeaway joined	16:41
	→ stkrzysiak_ joined	16:41
	← stkrzysiak left	16:42
	stkrzysiak_ → stkrzysiak	16:42
	← dissidentat left	16:43
	→ Error404NotFound joined	16:43
	← palyboy left	16:44
	→ ESphynx joined	16:44
	← guampa left	16:44
ESphynx	hey guys... I'm looking for advice to how to have 'unrelated' branches?	16:44
	this is for the debian packaging stuff...	16:45
grawity	git checkout --orphan foo	16:45
ESphynx	so I have these debian branches... where there's only debian/	16:45
	and I have the normal branches (master, dev) where I have everything else but the debian folder	16:45
	I already have both now though... what should I do?	16:46
	git checkout --orphan debian ?	16:46
	even if debian exists already ?	16:46
	→ palyboy joined	16:46
	→ fireh joined	16:46
ESphynx	or should I just delete all the branches and start over?	16:47
kpreid	ESphynx: no, you'll have to delete the debian branch first	16:47
	then git checkout --orphan debian, then delete all the stuff you don't want on that branch and commit	16:47
ESphynx	ok thanks guys	16:47
	hmm	16:47
	ok	16:47
	great :)	16:47
	← letharion left	16:48
	→ rchavik joined	16:49
	← varg left	16:49
	← lausi left	16:50
	→ Ad0 joined	16:50
Ad0	hi :)	16:50
gitinfo	Ad0: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	16:50
Ad0	if I have one main branch, and 2 branches I merge together in main branch. I am not doing anything else in that main branch, it's sole purpose is to merge the 2 other branches together	16:51
	is there a way to kill all commits and start over?	16:51
	start fresh with a merge from branch 1 and branch 2	16:51
	is it git rebase or something	16:52
ESphynx	What would be a good prefix for those packaging only branches ? deb? ppa? :)	16:52
	→ lausi joined	16:52
	→ letharion joined	16:53
Ad0	I've seen name-PPA before	16:53
	← Chib left	16:53
ESphynx	I'm thinking of debNatty, debOneiric, debPrecise, debDebian	16:54
grawity	deb/natty works too	16:54
ESphynx	Or ppaNatty, ppaOneiric	16:54
	you can put / ?	16:54
Ad0	I would have it with Natty-ppa	16:54
	but that's me	16:54
ESphynx	hehe yeah i'd rather have a prefix, so they're all grouped together	16:55
m1sc	Ad0: I don't really get what you're looking for, maybe merge --squash?	16:55
grawity	ESphynx: yeah, I think refs are actually hierarchical like that	16:55
Ad0	m1sc: I have been merging branch A and B together in main C	16:55
	← rchavik left	16:55
hyperair	i use ppa/natty	16:55
Ad0	and I did some custom changes which is biting me in the ass	16:55
	so I want to just start fresh with a basic merge from latest from branch A	16:56
	maybe not kill all commits, but at least get to that stage	16:56
	where nothing from branch B is there	16:56
hyperair	ESphynx: you can take a look at the banshee packaging repository see Vcs-Git in apt-cache showsrc banshee	16:56
	s/see/$see/; s/$/$/	16:57
m1sc	Ad0: so you have some committed changes and want to put them on a new base?	16:57
	← psoo left	16:57
Ad0	yes I have a ton of commited merges with custom changes to make them work	16:58
	but I want to wipe them and start fresh	16:58
	→ specialGuest joined	16:58
ESphynx	OK i'll do ppa/natty then	16:58
	thanks guys	16:59
	← FatDarrel left	16:59
m1sc	Ad0: you could reset C to some commit and then merge A again. !backup first if you're not sure about what you're doing..	16:59
gitinfo	Ad0: Taking a backup of a git repository is always a good idea, especially when taking advice over IRC. Usually, the best way to TACTICALLY back up a git repo is `git clone --mirror`. However, some unusual maintenance might require `tar cf repo-backup.tar repodir`. Testing in a clone is also an excellent idea. See also http://sethrobertson.github.com/GitBestPractices/#backups	16:59
	← specialGuest left	17:00
	→ specialGuest joined	17:00
Ad0	hehe thanks	17:01
	→ palyer joined	17:02
ESphynx	so so I did my checkout --orphan	17:02
	← palyboy left	17:02
ESphynx	I have a bunch of 'staged' changes, but reset is being dumb	17:02
	fatal: Failed to resolve 'HEAD' as a valid ref.	17:02
	Can't I clear all that staged stuff?	17:02
grawity	maybe `rm .git/index`?	17:03
m1sc	ESphynx: that's because there's no root commit yet	17:03
	→ surma joined	17:03
ESphynx	m1sc: I still say git reset should delete everything staged :P	17:03
	→ jey joined	17:03
	→ guampa joined	17:04
ESphynx	grawity: that did work. thanks a lot :P	17:04
m1sc	ESphynx: reset can't, because there's no HEAD	17:04
ESphynx	m1sc: I say in this special case it can rm .git/index	17:04
	→ theCole joined	17:05
m1sc	ESphynx: when you pass git reset a ref or commit, it works	17:06
	← Davey left	17:06
	← elias left	17:07
	→ FatDarrel joined	17:07
ESphynx	m1sc: but here I want to reset to 'nothing' :P	17:07
m1sc	ESphynx: well, your index.?	17:07
	→ aspotashev joined	17:08
m1sc	ESphynx: when you create your orphan branch from master, you would do "git reset master", which brings back your index to what master looked but keeps your worktree	17:09
	*like	17:09
	← Zenopus left	17:09
ESphynx	m1sc yes but I want an orphan branch that's empty :P	17:11
	I want to reset to an empty repository	17:11
	← leotr left	17:13
m1sc	ESphynx: check the last paragraph for --orphan in man git-checkout	17:13
gitinfo	ESphynx: the git-checkout manpage is available at http://git-scm.com/docs/git-checkout.html	17:13
	→ pfFred joined	17:13
	← pfFredd left	17:14
	→ duckxx joined	17:14
	→ tshauck joined	17:15
ESphynx	m1sc: that would delete my actual working tree files though, which I did not want	17:16
	← theCole left	17:16
ESphynx	that's the whole point for me of having a disconnected history, so I can switch branch witohut losing my different contents :P	17:16
	(running git rm -rf .)	17:17
	← esa left	17:17
	← FatDarrel left	17:17
	→ esa joined	17:17
	← pfFred left	17:18
m1sc	ESphynx: then go ahead, prepare everything you want it and commit. also, there's --cached for git tm	17:18
	*rm	17:18
kpreid	no, you do that in the orphaned branch after you create it, before commit	17:18
ESphynx	ah right... --cached :P forgot about that	17:18
	I could have done git rm -rf --cached . :P	17:18
	OK I still can't switch branch :\|	17:24
	going back to the same branch as before I get error: The following untracked working tree files would be overwritten by checkout:	17:24
	bryanmv → bryanmv`away	17:24
kpreid	yes, because you didn't delete them	17:25
	← abletony84 left	17:25
kpreid	so you try to check out your old branch and it sees these files	17:25
	ordinarily, when you switch from that branch to debian git automatically removes the files from working tree, so there is no conflict when you switch back	17:25
	← airborn left	17:26
kpreid	just rm the files — they will be restored when you checkout	17:26
ESphynx	I wish it left the files alone	17:26
	→ mrwalker joined	17:26
ESphynx	when I switch branch	17:26
	← maletor left	17:26
kpreid	switching branches is switching trees. you don't get a mix	17:26
ESphynx	in fact I hate the whole idea of branches for debian packaging :\|	17:26
	← rurufufuss left	17:26
	→ airborn joined	17:28
	← darren left	17:29
	← Goplat left	17:29
	→ SegFaultAX\|work2 joined	17:30
ESphynx	this seriously messed up my working directory :\|	17:31
	→ gtrak joined	17:32
cbreak	what did you do?	17:32
	well, you can always git reset --hard HEAD to nuke all changes out of the wd	17:32
	→ erynofwales joined	17:32
ESphynx	cbreak: that might have been a better idea	17:32
	→ marhaban joined	17:33
	→ longbeach joined	17:33
	→ ro0 joined	17:35
	← ro0 left	17:35
	→ ro0 joined	17:35
	← ro0 left	17:36
	→ Sonderblade joined	17:37
	← airborn left	17:37
	← skorgon left	17:37
	← fireh left	17:37
	→ tjadc joined	17:38
	→ airborn joined	17:40
	→ ro0 joined	17:40
	→ dorkmafia joined	17:42
	→ Drool joined	17:42
	→ r0bby_ joined	17:42
	r0bby_ → robbyoconnor	17:42
	← Chappo left	17:42
	← marhaban left	17:42
	← Droolio left	17:42
	→ Chappo joined	17:42
	→ marhaban joined	17:42
	← r0bby left	17:44
	← Sonderblade left	17:44
	→ iori joined	17:45
	→ dc5ala joined	17:46
	← chinopozo left	17:46
	← candybar left	17:46
	← dinesh6654 left	17:46
	← marhaban left	17:47
	→ apok joined	17:47
	→ codertux joined	17:47
	← apok left	17:48
	→ Sonderblade joined	17:49
	→ osmosis joined	17:50
	→ towski joined	17:51
	→ candybar joined	17:53
	← mrwalker left	17:53
	→ macIvy joined	17:53
	→ sunny256 joined	17:54
	→ fireh joined	17:55
	→ Davey joined	17:56
	→ reds joined	17:57
	reds → Guest33206	17:58
	← towski left	17:58
	← acalbaza left	17:59
	← reactormonk left	17:59
	→ wolftankk joined	18:02
	← Guest33206 left	18:02
	→ lupo7 joined	18:04
	→ oxidative_ joined	18:04
	→ towski joined	18:06
	→ RaceCondition joined	18:06
	→ darren joined	18:06
	← oxidative left	18:07
	→ Psycho_pr joined	18:08
	← shreya left	18:09
	← dfr\|mac left	18:09
Psycho_pr	If I have some changes in my work tree, and I pull something from my origin, and the changes that I have are exactly the same as the changes that will be pulled, will there be a conflict?	18:09
	→ shreya joined	18:09
	← towski left	18:09
cbreak	yes	18:10
Psycho_pr	shit	18:10
cbreak	uncommitted changes will always conflict	18:10
Psycho_pr	why?	18:10
cbreak	because that's the only sane thing	18:10
Psycho_pr	so I guess I better first revert the changes on these files before pulling.	18:11
cbreak	you could do it like subversion and risk data loss	18:11
	but obviously that's dumb	18:11
	so only subversion does it	18:11
AAA_awright	Does there exist any plumbing to get meta-data from a remote repository? I just want a list of refs from a remote repository, without cloning any files	18:12
	← longbeach left	18:13
	→ sq-one joined	18:13
	→ dsirijus joined	18:13
cbreak	you can get the current branch tips	18:13
	← chuckf left	18:13
cbreak	git ls-remote	18:13
AAA_awright	charon: Awesome	18:15
	← jrdnn left	18:17
dropdrive	Hi all. Is there a recommended/widely-accepted way to refer to a specific branch on a specific repo? E.g. $url#$branch_name or $url:$branch_name	18:17
grawity	#foo would make the most sense in an URL	18:17
	→ tjholowaychuk joined	18:18
	← angelsl left	18:18
cbreak	dropdrive: "branch Foo in repo ssh://server/repo.git"	18:18
grawity	cbreak: Doesn't work well with software that has to parse the thing	18:19
	→ subbyyy joined	18:20
	→ jrdnn joined	18:21
	← juvenal left	18:21
	← dsirijus left	18:21
	← dr_lepper left	18:22
EugeneKay	I would use SSH syntax, with #refs/heads/foo at the end	18:22
	You won't find a # in any (sane) repo URI or branch name, so it should be parse-safe	18:22
	← palyer left	18:22
	→ palyboy joined	18:22
	← jcromartie left	18:23
grawity	It goes well with other URIs too, remotepart#localpart	18:23
	← pw1 left	18:25
	→ scientes joined	18:25
	← magcius left	18:25
EugeneKay	YOu're referring to HTML "anchors"? Yeah.	18:26
	← kleist left	18:26
	→ magcius joined	18:27
	→ r0bby joined	18:27
	→ cwang joined	18:28
dropdrive	Thanks guys.	18:29
	← compojoom_ left	18:29
	← robbyoconnor left	18:29
	→ dr_lepper joined	18:29
	← Hasbro left	18:31
	← nor5trom left	18:31
	← cwang left	18:33
	→ dom2012 joined	18:34
ESphynx	git://github.com/ecere/sdk.git,branch=dev	18:34
	why is this not working?	18:34
	My branch is called 'dev' ...	18:34
CareBear\	ESphynx : what do you want to do ?	18:35
	Esphynx : clone?	18:35
	→ diegoviola joined	18:35
ESphynx	CareBear\ well I was trying to import on launchpad	18:35
	but when I try a clone it gives me that error too	18:35
CareBear\	ESphynx : oh ok - have to ask #launchpad then	18:35
ESphynx	it's not a launchpad issue, doing a clone does that	18:35
CareBear\	ESphynx : the git URL is git://github.com/ecere/sdk.git and nothing else	18:35
	branches are never part of the url	18:36
ESphynx	hmm right	18:36
	→ lirt joined	18:36
	← ketas left	18:36
ESphynx	it must be a launchpad thing then	18:36
	thanks	18:36
CareBear\	you're welcome	18:36
	← meLon left	18:37
	← diegoviola left	18:38
EugeneKay	I wonder if a patch to clone to take a #refspec as part of the clone URL(instead of -b)	18:41
	→ FatDarrel joined	18:41
EugeneKay	would be accepted	18:42
	← unterstrom_ left	18:42
	← dom2012 left	18:42
	← Davey left	18:42
	→ diegoviola joined	18:42
	→ ketas joined	18:43
milki	EugeneKay: !tias	18:43
	→ dvide joined	18:43
gitinfo	EugeneKay: 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://sitaramc.github.com/1-basic-usage/tias.html may help with git-specific TIAS.	18:43
milki	:P	18:43
	→ Davey joined	18:44
	→ mrwalker joined	18:44
EugeneKay	I'd have to write the patch first	18:44
	← RaceCondition left	18:45
	→ jrdnn_ joined	18:45
	← jrdnn left	18:45
nyxx	Hi, I'm using git on Windows. When ever I do a `git push`, I have to enter my username and password. How can I make it log in automatically?	18:46
	← FatDarrel left	18:46
	→ cannonball joined	18:47
	→ nor5trom joined	18:47
	→ palyer joined	18:49
	→ ph^ joined	18:49
	← palyboy left	18:49
	→ arcane joined	18:50
	→ avpx joined	18:51
	→ dfr\|mac joined	18:52
	← macIvy left	18:53
	← arcanin left	18:53
CareBear\	nyxx : where do you push to?	18:56
	nyxx : what protocol - ssh or http?	18:56
	← tshauck left	18:56
CareBear\	nyxx : if ssh you have to configure it in the ssh client (use putty, even if git is calling plink)	18:57
	nyxx : if http you can add it into the url http://user:pass@host/path/repo.git	18:58
	nyxx : the latter works only for http. with ssh you must configure in your ssh client.	18:58
EugeneKay	nyxx - !ssh_github	18:58
gitinfo	nyxx: If you are having problems accessing github repos over HTTPS you should switch to SSH (the default used by git, for many very good technical reasons): https://help.github.com/articles/generating-ssh-keys	18:58
	← Davey left	18:59
	→ nwest joined	18:59
	← wolftankk left	19:00
	← toabctl left	19:01
	← jey left	19:01
	← jrdnn_ left	19:02
	→ Davey joined	19:03
	→ hwrd\|work joined	19:03
Psycho_pr	Why is there no post-pull or post-pop hook?	19:04
	← hwrd\|work left	19:05
EugeneKay	!cuz	19:05
gitinfo	[!why] Why? Because screw you, that's why.	19:05
grawity	Psycho_pr: post-merge	19:05
Psycho_pr	:-(	19:05
EugeneKay	I can think of a few uses for a post-fetch hook	19:06
_ikke_	This is a mild version, the previous was much more rude	19:06
Psycho_pr	Did the previous version punch you in the face?	19:06
EugeneKay	"Because fuck you, that's why."	19:06
	→ Ether_Man joined	19:07
Psycho_pr	:\|	19:07
	→ jrdnn joined	19:07
Psycho_pr	Man why so harsh?	19:07
EugeneKay	you asked. :-p	19:07
_ikke_	that's EugeneKay	19:07
EugeneKay	Psycho_pr - have a !botsnack, they're good.	19:08
gitinfo	Psycho_pr: Om nom nom	19:08
Psycho_pr	:\|	19:08
	→ asahi joined	19:09
	← asahi left	19:09
Ether_Man	Im having a small issue with gitweb which I hope someone could help me with. I've set up a vhost and I reach gitweb itself fine. I have it pointed to the base dir of where all my git projects are stored. When browsing to the gitweb page though, I get No such projects found. (havnt searched or anything).. List all projects just gets the same. If I press the TXT button, it correctly shows all the project/.git entries, but nothing in the html o	19:11
	utput. Anyone with any idea what I've missed?	19:11
	← palyer left	19:12
	→ palyboy joined	19:14
	← velikasha left	19:15
	→ arkas joined	19:15
	→ maletor joined	19:16
	→ redrocket joined	19:16
	← ku left	19:16
	← arkas left	19:16
	← palm_vue left	19:20
	← bizhanMona left	19:20
	← Lgb left	19:20
	→ palm_vue joined	19:20
	→ Lgb joined	19:21
	→ bizhanMona joined	19:21
	→ todd_dsm joined	19:21
	← Mogget left	19:26
	→ tripolar joined	19:26
tripolar	hi	19:26
gitinfo	tripolar: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	19:26
tripolar	is it possible to move changes made in the master branch not commited to a new branch? or create a branch that looks like master right now with the changes?	19:27
	→ Mogget joined	19:27
raek	tripolar: yes, just do "git checkout -b new-branch" and commit after that	19:28
	!float	19:28
gitinfo	If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there. If the files you changed differ between branches, the checkout will fail. In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.	19:28
tripolar	raek thanks	19:30
	→ goshawk joined	19:30
	→ FernandoBasso joined	19:32
	→ e-dard joined	19:32
	→ HyperNerdV2 joined	19:32
e-dard	Excuse the cruddy question but… I use virtualenwrapper and my usual workflow is to automagically create a virtual environment and a project folder. Once I execute this I'm sat in my project folder.	19:33
	My next step would be to make an empty git repo. But how do I do it such that I don't create a new folder? I just want the .git to sit in the current folder if that makes sense?	19:33
	← segher left	19:34
e-dard	Gah. And as I wrote the question the answer hit me in the face. `git init .` Sorry for noise.	19:34
	→ chinopozo joined	19:34
_ikke_	:)	19:35
raek	this is also what happens when you run git-init without an argument	19:35
e-dard	There should just be a /scratch channel where you write a question to just get the answer pop in your head.	19:35
	→ segher joined	19:35
	← pppaul left	19:36
_ikke_	/join ##scratch	19:36
	→ Kabaka joined	19:38
	← palyboy left	19:39
	← ska-fan left	19:44
	← mrwalker left	19:44
	→ ska-fan joined	19:45
	← dropdrive left	19:45
	← shreya left	19:48
EugeneKay	e-dard - "." is the default for the two-word "git init"	19:49
	→ felgru joined	19:49
	← gtrak left	19:50
	← dc5ala left	19:50
	← Error404NotFound left	19:52
	→ klj613 joined	19:53
	← cagedwisdom left	19:54
	→ pppaul joined	19:55
	← Guest74974 left	19:58
	→ reds joined	19:58
	reds → Guest62697	19:58
	← iori left	19:58
	→ khaije\|minder joined	19:59
	← SegFaultAX\|work2 left	20:00
	← Guest62697 left	20:02
	← subbyyy left	20:04
	← Araxia left	20:05
	→ subbyyy joined	20:05
	← specialGuest left	20:06
	→ fvg joined	20:08
	← osmosis left	20:08
	← Dimitr left	20:08
	→ gtrak joined	20:09
	← jjbohn left	20:10
	→ noyb_ joined	20:15
	← noyb left	20:15
	→ zackiv31 joined	20:16
jnewt	how do you rebase all branches that are based on a "original base SHA" to a "new base SHA". git rebase --onto "new base SHA" "old base SHA" only takes the branch you're on.	20:16
	→ jonn joined	20:17
jnewt	do i have to do them one at a time?	20:17
	→ LeMike joined	20:18
	→ Shusshu joined	20:19
	→ SegFaultAX\|work2 joined	20:19
	→ specialGuest joined	20:19
	→ tellone joined	20:20
	← zackiv31 left	20:21
	← Shusshu left	20:21
	← Anders_J left	20:22
	← LeMike left	20:22
	→ zackiv31 joined	20:23
	→ Dimitr joined	20:23
	← Fullmoon left	20:30
	→ palyboy joined	20:31
	→ specialGuest1 joined	20:32
	→ irqq joined	20:33
	← specialGuest left	20:35
	Guest69382 → wolog	20:35
	wolog → Guest30223	20:36
	→ LeMike joined	20:37
	← chinopozo left	20:38
	← duckxx left	20:40
	← tjholowaychuk left	20:40
	← irqq left	20:40
	→ jey_ joined	20:44
	← cannonball left	20:44
	→ Goplat joined	20:46
	← arietis left	20:47
	→ FatDarrel joined	20:48
	← gtrak left	20:48
	← e-dard left	20:50
	Guest30223 → wolog	20:54
	→ teotwaki_ joined	20:55
SethRobertson	jnewt: Yes, but you can get the list via `git branch --contains` and then put a script around it	20:55
	→ bubbles10101 joined	20:55
	→ mike007 joined	20:55
	← klj613 left	20:55
	← specialGuest1 left	20:58
	→ LeMike1 joined	20:58
	→ specialGuest joined	20:59
	← FatDarrel left	20:59
	← specialGuest left	20:59
	→ specialGuest joined	20:59
	→ aesthetikargonau joined	20:59
	→ theCole joined	21:00
	← aesthetikargonau left	21:00
	← LeMike left	21:00
	← agileadam left	21:01
	→ guns joined	21:03
	→ leila_ joined	21:06
	→ s0ber_ joined	21:06
	← bubbles10101 left	21:08
	← s0ber left	21:08
	s0ber_ → s0ber	21:08
	← arturaz_ left	21:09
Ad0	m1sc: thanks	21:11
	git reset did it	21:11
	← tellone left	21:12
	← esa left	21:12
	← _iron left	21:12
	← bartman left	21:13
	→ esa joined	21:13
	→ blast_hardcheese joined	21:14
	→ e-dard joined	21:15
	→ Mariiii joined	21:16
	← Mariiii left	21:17
	→ bartman joined	21:19
	← ph^ left	21:19
	← maletor left	21:22
	r0bby → robbyoconnor	21:22
	→ artnez joined	21:22
	← cakehero left	21:22
	← leila_ left	21:23
	← aspotashev left	21:24
	← sq-one left	21:26
shruggar	is there any fast (as in simple to stick into a script) way to turn an arbitrary directory into a tree object?	21:26
	← chuckharmston left	21:26
	← ipoval left	21:29
Ad0	shruggar: like ps aux --forest?	21:29
	→ leila_ joined	21:30
jnewt	whats the syntax to stage all unstaged deletes in a folder? git rm dir/. gives me a not removing the folder recursively error, and git rm dir/* gives me errors about changes that are already staged needing --cached. but when i do rm --cached, it adds the deletes, and removes the modifications from the index.	21:30
Ad0	shruggar: http://www.computerhope.com/unix/tree.htm	21:30
	great little tool	21:30
	→ etcetera joined	21:31
jnewt	i can rm each file independently, but I assume there's a catch-all way	21:31
	← stkrzysiak left	21:33
	← leila_ left	21:34
	→ jragon joined	21:35
jragon	Hello	21:35
gitinfo	jragon: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	21:35
shruggar	Ad0: I think this is #git, so I mean a git tree object :)	21:36
	← capiira left	21:36
jragon	I'm trying to push a commit to github. It worked earlier, but now it seems to fail. I get fatal: could not read Username for 'https://github.com': No such device or address	21:36
cbreak	shruggar: go over the index	21:37
	then you can use git write-tree	21:37
	shruggar: see man git-write-tree	21:38
gitinfo	shruggar: the git-write-tree manpage is available at http://git-scm.com/docs/git-write-tree.html	21:38
shruggar	cbreak: yeah, that's the plan at the moment. I just tend to want to say "hey, arbitrary directory, be a tree!" I guess I should just write a dash command for it	21:38
	← tete_ left	21:39
	← subbyyy left	21:40
Ad0	shruggar: heh didn't know what it was :)	21:40
	← savid left	21:40
	← Milossh left	21:44
	→ kcj joined	21:44
	→ Milossh joined	21:45
	← e-dard left	21:45
	→ juvenal joined	21:47
	→ maletor joined	21:48
	→ resmo__ joined	21:48
	← codertux left	21:50
	→ codertux joined	21:50
	← resmo_ left	21:50
	→ linoj joined	21:51
	← specialGuest left	21:51
	→ nadavoid joined	21:53
	← pppaul left	21:55
	← nadavoid left	21:55
	← mike007 left	21:56
	→ specialGuest joined	21:56
	← specialGuest left	21:56
	→ specialGuest joined	21:56
	← etcetera left	21:58
	→ reds joined	21:58
	reds → Guest69510	21:58
	← kart0ffelsack left	21:59
	→ macIvy joined	22:00
	→ eletuchy joined	22:00
	← microcolonel left	22:00
	→ microcolonel joined	22:00
	→ unterstrom joined	22:01
	← LeMike1 left	22:01
	→ alanp joined	22:01
	→ kart0ffelsack joined	22:02
	← kart0ffelsack left	22:02
	→ kart0ffelsack joined	22:02
	← Guest69510 left	22:03
	← naee left	22:04
	← Internet13 left	22:04
	← segher left	22:04
	← Twey left	22:04
	← ISF left	22:04
	← lindenle left	22:04
	← ology left	22:05
	→ ology joined	22:05
	→ segher joined	22:05
	→ nadavoid joined	22:05
	← alanp_ left	22:05
	← juvenal left	22:06
	← Davey left	22:06
	→ Twey joined	22:07
	→ Internet13 joined	22:08
	← diegoviola left	22:08
	→ juvenal joined	22:09
	← specialGuest left	22:09
	→ capiira joined	22:09
	← capiira left	22:09
	→ etcetera joined	22:09
	→ eean joined	22:09
	→ Davey joined	22:10
	→ Error404NotFound joined	22:11
	→ urlwolf joined	22:11
	← jragon left	22:12
	→ Fernando1asso joined	22:12
	← etcetera left	22:12
	→ marhaban joined	22:12
	→ specialGuest joined	22:13
	← specialGuest left	22:13
	→ specialGuest joined	22:13
	← macIvy left	22:14
	→ osmosis joined	22:14
	← maletor left	22:15
	→ EhevuTov joined	22:15
	← FernandoBasso left	22:16
	→ quesada joined	22:17
	Fernando1asso → FernandoBasso	22:17
	→ maletor joined	22:18
	← maletor left	22:18
	← fvg left	22:20
	← urlwolf left	22:21
	← blacktulip left	22:21
	← zackiv31 left	22:22
	← Davey left	22:22
	← pantsman left	22:23
	→ legumbre joined	22:23
	← quesada left	22:25
	→ brotatos joined	22:26
brotatos	is there a commit command that allows me to edit a message specific to each file committed?	22:26
tango_	brotatos: commits are tree-wise, not file-wise	22:26
	if you need file-wise commits, commit a single file	22:27
	change	22:27
jnewt	you'd have to split your commits	22:27
brotatos	tango_: thanks	22:27
	← brotatos left	22:27
	→ pcapriot` joined	22:28
	← EhevuTov left	22:28
	← babilen left	22:32
	← nor5trom left	22:35
	← wolog left	22:35
	← SkAzZ left	22:36
	← Targen left	22:36
	→ nor5trom joined	22:37
	→ david1010 joined	22:37
	← SegFaultAX\|work2 left	22:38
david1010	I am trying to test with rspec and getting this error "/Users/davidmac/.rvm/gems/ruby-1.9.3-p125/gems/rspec-core-2.11.1/lib/rspec/core/option_parser.rb:21:in `parse!': missing argument: --format (OptionParser::MissingArgument)"	22:39
	Any idea what is wrong?	22:39
	sorry wrong window	22:39
canton7	the channel? :P	22:39
	aha	22:39
	← bartman left	22:41
	→ Raydiation joined	22:44
Raydiation	how can i undo a pull?	22:45
	which is some commits behind and didnt touch my files	22:45
	← sh0ne left	22:45
	← esa left	22:45
	← HyperNerdV2 left	22:46
	→ diegoviola joined	22:46
	← Goplat left	22:46
	← guampa left	22:47
offby1	Raydiation: if the pull didn't create a merge commit, then you can reset your branch back to where it was before	22:47
	heck, you can still do that even if it _did_ create a merge commit, now that I think about it	22:48
	"git log -g" will help you figure out where it was before, if you've forgotten	22:48
	→ bartman joined	22:49
	← BadDesign left	22:49
	← flavius left	22:49
Raydiation	offby1: i want to keep my changes: https://pastee.org/teffr	22:49
	and undo 9b54713	22:49
	← oxidative_ left	22:50
	← gavin_huang left	22:51
	→ beneggett joined	22:52
Raydiation	offby1: git rebase -i 9b54713~1 ?	22:53
offby1	hole on	22:55
	← timthelion left	22:55
offby1	yes, that rebase sounds right.	22:55
	I'd stare at "gitk --all" for a while to make sure I understood what I was about to do	22:55
	→ subbyyy joined	22:56
Raydiation	ray@arch /srv/http/apps/news $ git rebase -i 9b54713~1	22:58
	error: could not apply 0cdd144... changes namespace to OCA\News	22:58
	m	22:58
	i aborted the rebase	22:58
	gitk --all	22:59
	/usr/bin/gitk: line 3: exec: wish: not found	22:59
	→ capiira joined	23:00
	← capiira left	23:00
	→ capiira joined	23:00
	← capiira left	23:00
	→ dreiss joined	23:01
	← ology left	23:01
offby1	boo	23:03
	sounds like an incomplete installation of git :-(	23:03
	→ wolog joined	23:03
Raydiation	im using gitg atm	23:03
offby1	it's a shame, since gitk is super-handy	23:03
	if you've got some other graphical viewer, that should be OK	23:03
	the point is to see how the commits relate to one another	23:03
	→ DrHouse\|TARDIS joined	23:04
	→ maletor joined	23:04
Raydiation	the problem is that someone changed something at the same time	23:04
	← candybar left	23:04
Raydiation	in the same branch	23:04
	then later theres a merge	23:05
	well i actually dont know wth happened	23:05
offby1	If gitg isn't making that clear, then it's no good.	23:06
	what _is_ gitg, anyway?	23:06
Raydiation	offby1: can i store my patches somehow?	23:06
	→ ladder4 joined	23:07
Raydiation	then subsequently apply patch afert patch after reset	23:07
	→ japankurt joined	23:07
	← Will\| left	23:07
	→ etcetera joined	23:07
canton7	that's a description of what rebase does, surely?	23:08
	← giallu left	23:08
Raydiation	offby1: problem is the marked entry http://s14.directupload.net/images/120812/v3y5ylnz.png	23:08
offby1	Raydiation: at worst, you can put a branch or tag at your current head; reset to before the merge; then run "git format-patch" to retrieve your work, and apply those patches	23:09
	← zivester left	23:09
	← asakura left	23:09
	← cmn left	23:09
	← ro0 left	23:10
	← koki left	23:10
	← wereHamster left	23:10
offby1	it looks like there are six commits on the "newapp" branch that happened after the merge.	23:10
	Those are the ones you want to preserve?	23:10
	→ steevel joined	23:10
offby1	canton7: I dimly recall there's something about rebasing past a merge that is more complex than usua.	23:11
	I don't want to recommend it since I'm not clear on how to do it.	23:11
canton7	ah, there's a merge involved. fair enough	23:11
	→ Leonidas_ joined	23:11
	→ koki joined	23:11
	→ avar joined	23:11
	← avar left	23:11
	→ avar joined	23:11
	→ gtrak joined	23:11
Raydiation	offby1: i want to preserve everything after the marked entry	23:11
offby1	Raydiation: in the newsapp branch, yes?	23:12
Raydiation	yes	23:12
	→ wereHamster joined	23:12
	→ asakura joined	23:12
	→ blueyed joined	23:12
	← specialGuest left	23:12
	→ cmn joined	23:13
offby1	I'd reset your newsapp branch back before the faulty merge, merge from owncloud/newsapp, then git-cherry-pick "setting items as important", "mark items also", "changed marking of titles", and "fixed the wording".	23:13
	I'm guessing nobody will mind if you forcibly reset newsappfork/newsapp ?	23:14
	→ stfn joined	23:14
	← dorkmafia left	23:14
	→ Will\| joined	23:14
Raydiation	offby1: i forked the branch and want to do a merge request	23:14
	→ ro0 joined	23:15
	→ Deck` joined	23:15
offby1	I'm not sure if that answers my question, but I'll be optimistic	23:15
Raydiation	so git reset --hard commit_before_pull	23:15
	← linoj left	23:16
Raydiation	git pull origin newsapp	23:16
	→ dorkmafia joined	23:16
Raydiation	and then i can just cherry pick with the ids of the future commits?	23:16
	← etcetera left	23:16
Raydiation	shas*	23:16
	→ knittl joined	23:16
	← knittl left	23:16
	→ knittl joined	23:16
	→ liori joined	23:16
Raydiation	or ommit the --hard	23:17
	→ HyperNerdV2 joined	23:17
dorkmafia	hey i'm in the middle of a merge and I use vimdiff as my default mergetool … it's a really large merge and there are about 20 large files that I are resulting in merge conflicts… these 20 files I would like to take their changes and discard mine what's the best way to do this?	23:17
	I checked out their branch	23:18
	they started merging my branch into theirs	23:18
	← vitor-br left	23:18
	→ capiira joined	23:19
	← beneggett left	23:19
	← AaronMT left	23:20
	← nadavoid left	23:21
	← felgru left	23:22
	→ tshauck joined	23:22
dorkmafia	ok i figured it out just did git checkout origin/theirs — file	23:23
	how do I find out all of the files with merge conflicts?	23:23
	→ candybar joined	23:25
offby1	Raydiation: sounds right	23:25
	I'd use --hard though	23:25
	dorkmafia: "git status"	23:26
	← juvenal left	23:26
	→ FatDarrel joined	23:27
	→ nadavoid joined	23:28
	← nadavoid left	23:29
	→ Araxia joined	23:29
	← else- left	23:31
	← unterstrom left	23:31
dorkmafia	offby1: I'm trying to do git checkout origin/theirs — assets/*	23:32
	but it won't let me do it on some files so it's throwing errors	23:32
	error: path 'assets/file'	23:32
	→ skyf joined	23:33
	→ else- joined	23:34
	→ AaronMT joined	23:36
dorkmafia	git checkout origin/branchname --force assets	23:37
	worked	23:37
	← FatDarrel left	23:38
	← skyf left	23:38
	→ trtzrdb joined	23:39
Raydiation	offby1: ty, that cherrypick thing rocks	23:39
	i had to copy some newer things over though	23:39
	now the only problem is To [email@hidden.address]	23:41
	! [rejected] newsapp -> newsapp (non-fast-forward)	23:41
	error: failed to push some refs to 'git@gitorious.org:~raydiation/owncloud/raydiations-apps.git'	23:41
	← dreiss left	23:42
	→ Psycho_pr1 joined	23:43
	← Psycho_pr left	23:43
	← airborn left	23:44
Raydiation	tells me that untracked files would be overwritten by merge	23:44
	→ metasansana joined	23:44
Raydiation	working tree	23:44
	← lirt left	23:44
	→ ehsan joined	23:45
Raydiation	well, fk this, ill just create a new repo and copy my files over	23:47
	← w09x_ left	23:47
	← ehsan left	23:49
	→ skyf joined	23:50
jnewt	i'm trying to rebase branch dev onto branch testing both of which have master as the common ancestor. i want the changes from dev to overwrite any conflicts in testing. so i do $git checkout dev $git rebase -s ours --onto testing master	23:50
	and as it clicks through the revisions on testing, it says "already applied"	23:50
	it moves the branch pointer dev to the testing commit and nothing else changes.	23:51
	← david1010 left	23:51
jnewt	what am i doing wrong there?	23:52
	→ david1010 joined	23:52
	← Speed left	23:53
	→ danmackay joined	23:56
	← JStoker left	23:56
Freeaqingme	I have a machine on which I clone all branches within a certain repo using a bash script. However, when a branch is deleted my script never picks those up. is there any way to retrieve what branches were deleted using git?	23:57
	→ chuckharmston joined	23:58
	→ reds joined	23:58
	reds → Guest97802	23:59

Logs Search ←Prev date Next date→

Channels Documentation