IRCloggy #git 2012-05-09

Provider of IRC logs since 2005.

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

	2012-05-09
	← Destos left	00:00
	← Lord_DeathMatch left	00:01
	← Element9 left	00:01
	← Xizor left	00:02
	→ RobertLaptop joined	00:02
	← ferdna left	00:03
	← ISF left	00:03
	→ yoklov joined	00:03
	← dreiss left	00:04
	→ iwoj joined	00:04
	→ rgwu joined	00:04
	→ theCole joined	00:05
	← kenperkins left	00:07
	→ n0tk joined	00:07
	← theCole left	00:07
	→ RoboTedd_ joined	00:08
	← TeckniX left	00:08
	← fission6 left	00:09
	← smw left	00:12
	→ KWKdesign joined	00:14
	← vervic left	00:15
	← lonewulf` left	00:15
	→ Slix` joined	00:16
	→ vervic joined	00:16
	← codewrangler left	00:16
	→ TeckniX joined	00:17
	→ lonewulf` joined	00:17
	← adhawkins left	00:17
	→ adhawkins joined	00:18
	→ smw joined	00:19
	← AaronMT left	00:19
	← JohnyBaby left	00:19
	→ JohnyBaby joined	00:19
	→ nrichards joined	00:21
	→ MichaelRpdx joined	00:21
	← daaku left	00:21
	→ tmcmahon joined	00:21
	← neurodrone left	00:22
	→ daaku joined	00:22
	← tmcmahon left	00:23
	← JohnyBaby left	00:24
	← adhawkins left	00:24
	← TeckniX left	00:25
	← daaku left	00:26
	→ envi_ joined	00:26
	→ rurufufuss joined	00:29
	→ tfittsy joined	00:29
	← lonewulf` left	00:29
ScottNYC	is iterm2 pretty much the standard for terminal use? among heavy terminal users anyway	00:31
	→ gavin_huang joined	00:31
bremner	umm. Were you looking for a mac user channel?	00:32
	→ thsd joined	00:33
	→ pjn_oz joined	00:33
	→ lonewulf` joined	00:33
	← juvenal left	00:34
ScottNYC	nah	00:37
	sry for the random question	00:37
	→ cccaldas joined	00:38
	← SegFaultAX\|work left	00:40
	→ fission6 joined	00:41
	← nb left	00:42
	← relurk left	00:42
	← MacGyverNL left	00:42
	→ vervic_ joined	00:43
	→ JohnyBaby joined	00:43
	← vervic_ left	00:43
	← Adys left	00:44
	← squentin left	00:45
	← vervic left	00:45
	→ squentin joined	00:46
	→ relurk joined	00:46
	→ MacGyverNL joined	00:46
	← MacGyverNL left	00:46
	→ MacGyverNL joined	00:46
	← tharkun left	00:48
	→ tharkun joined	00:48
	← MacGyverNL left	00:49
	← cccaldas left	00:50
	→ nb joined	00:50
	→ friskd joined	00:51
	→ aristidesfl joined	00:52
	→ kenperkins joined	00:53
	← friskd left	00:54
	→ Adys joined	00:54
	→ WillMarshall joined	00:54
	→ MacGyverNL joined	00:54
	← MacGyverNL left	00:54
	→ MacGyverNL joined	00:54
	← LucasTizma left	00:55
sitaram	good lord! is mac becoming so common people will say "standard" instead of "standard on mac"? I think I'll retire and become a blacksmith or something	00:55
	→ acl_ joined	00:57
	← sdeboey left	00:58
	← joaoh82 left	00:58
kevlarman	sitaram: wait what?	00:59
	→ k-man joined	01:00
	→ theCole joined	01:00
	← maletor left	01:01
	gitinfo set mode: +v	01:02
	← iwoj left	01:04
	→ heptadecagram joined	01:04
	→ Macca- joined	01:05
	← wilmoore left	01:05
	→ neurodrone joined	01:06
	→ cakehero joined	01:06
	gitinfo set mode: +v	01:07
Macca-	so, under windows i'm having trouble with ssh-agent.. at work it works fine but at home it won't work. `ssh-add` gives me "Could not connect to authentication agent"	01:07
	← heptadecagram left	01:07
Macca-	i disabled windows firewall, uninstalled mcafee, but still no dice.	01:07
ivan	sometimes you can use Process Monitor and hope there's a clue in the flood of output	01:08
cmn	have you tried starting the agent?	01:10
Macca-	cmn many times	01:12
cmn	how?	01:12
Macca-	i've killed the process and restarted it.. it seems to start fine but `ssh-add` just won't connect	01:12
	kill PID	01:12
cmn	what output does it give?	01:12
	← williamcotton left	01:12
cmn	how do you start it?	01:12
Macca-	`ssh-agent`	01:13
	→ weiyang joined	01:13
Macca-	https://gist.github.com/2640920	01:13
cmn	and have you tried 'eval $(ssh-agent)'	01:13
Macca-	is the output from starting ssh-agent	01:13
	← MistuhKurtz left	01:13
cmn	which is how you should start it?	01:13
	those variables are what the shell needs to consume so ssh can find the agent	01:13
	so either type it, or run it with eval	01:14
Macca-	that doesn't work either.	01:14
	← Eridius left	01:14
	← Slix` left	01:14
	← envi_ left	01:14
Macca-	https://gist.github.com/2640926	01:15
	the variables are set	01:16
	← Vampire0 left	01:16
Macca-	echo $SSH_AGENT_PID returns the correct pid	01:16
	← jstout24 left	01:16
cmn	then that's really something you should talk about with the ssh people	01:16
	← thsd left	01:18
	← TomyLobo left	01:18
	← Lgb left	01:20
	→ kaos01_ joined	01:21
kaos01_	hi, i like to start a branch called testing in a bare repo, do i need to create amaster branch first ?	01:22
	→ Lgb joined	01:22
	← bglusman left	01:23
	← codeM0nK3Y left	01:23
frogonwheels	kaos01_: I guess you could, but it's not really the normal way of doing it.	01:23
	→ iamjarvo joined	01:23
frogonwheels	kaos01_: presumably you have a non-bare clone somewhere?	01:23
	kaos01_: !master	01:23
gitinfo	kaos01_: [!master_branch] Master is the default name for the first branch created by git and is in no way special. You could just delete it if you want (though in a bare repository you'll probably want to change HEAD first).	01:23
	← tharkun left	01:25
	→ tharkun joined	01:25
kaos01_	well i just want to have a repo sith dev/test/prod brnaches	01:26
	← Bombe left	01:26
kaos01_	i guess i could create amster branch woth just a README or something in it ?	01:27
	so whats the standard way to do this ?	01:27
frogonwheels	kaos01_: I'm not really sure of the point of creating the branches until there's something there.	01:27
	→ QF-MichaelK joined	01:27
frogonwheels	kaos01_: when you need the branches, you push them up would be more standard...	01:27
	← QF-MichaelK left	01:27
frogonwheels	kaos01_: but sure, if you want to start somewhere, a README or .gitignore or something could be your base commit	01:28
	kaos01_: and you could just make your dev/test/prod branches and push them to the remote	01:28
	kaos01_: but I'd do it from a non-bare clone	01:28
	kaos01_: if you don't want a master branch (which is fine) just remember to set a new default branch for your bare repo	01:29
	→ Vampire0 joined	01:29
	← yorick left	01:29
kaos01_	aah so i could just push the master up as a branch with the name deve ?	01:29
	dev	01:29
frogonwheels	kaos01_: You could, but I wouldn't. git checkout -b dev && git push origin dev && git branch -d master	01:30
	kaos01_: master is just a name	01:30
	← acl_ left	01:30
	→ zdk joined	01:30
frogonwheels	kaos01_: or you could rename the branch	01:31
	kaos01_: actually that would be better, git branch -m master dev	01:31
	→ yorick joined	01:32
kaos01_	cool, i give that a go thanks	01:32
	← thearthur left	01:32
	→ thearthur joined	01:32
frogonwheels	kaos01_: while you CAN push a local branch to a specified remote branch, I would generally say it's to be avoided unless you really, really, really (really) know what you're doing	01:32
	← Lulzon left	01:33
	→ xiangfu joined	01:33
kaos01_	mm, im confused now	01:33
	→ xian9fu joined	01:33
kaos01_	so how would i handle different environmnets like dev/test/prod ?	01:34
	not branches ?	01:34
frogonwheels	kaos01_: different environments?	01:34
	→ Bombe joined	01:34
kaos01_	well versions of code, dev /test/prod	01:35
frogonwheels	kaos01_: you're best off handling them as branches.	01:35
	← thearthur left	01:35
kaos01_	isnt that what i was trying to do ? sorry confused	01:36
	→ kadoban_ joined	01:36
frogonwheels	kaos01_: ok, so start will all the branches pointing to the single base commit.	01:37
	→ cccaldas joined	01:37
	← xiangfu left	01:38
	← cespare left	01:38
	← iljakuklic left	01:38
	← glennpratt left	01:38
	→ sophacles joined	01:39
	← n0tk left	01:39
	→ Chib joined	01:44
	→ intripoon joined	01:44
	← sophacles left	01:45
	← kcj left	01:46
	→ kcj joined	01:46
	→ LucasTizma joined	01:47
	← intripoon_ left	01:47
	← osmosis left	01:48
	← chrisgeorge left	01:50
	→ sophacles joined	01:52
	→ cyb3r3li0g joined	01:53
	← cyphase left	01:54
	→ AaronMT joined	01:54
	→ bodom joined	01:56
	← brokenshakles_ left	01:56
	← kadoban_ left	01:58
	← TTSDA left	01:58
	← Chaser left	01:59
	← eletuchy left	01:59
	→ wilmoore joined	01:59
	→ ymasory joined	01:59
	→ _huoxito joined	02:01
	← `Assassin left	02:01
	→ Slix` joined	02:01
	→ iwoj joined	02:02
	← theCole left	02:02
	→ Chaser joined	02:02
	→ Sahua joined	02:02
	← sayd left	02:03
	→ TTSDA joined	02:04
	→ acl_ joined	02:07
	→ replore_ joined	02:08
	→ localhost joined	02:09
	← weiyang left	02:10
	← cakehero left	02:10
	→ daaku joined	02:10
	→ jimubao joined	02:11
	→ cyphase joined	02:13
	← juanpabloaj left	02:18
	← archdori left	02:20
	→ solofight joined	02:21
	← RoboTedd_ left	02:22
	← solofight left	02:22
TheShagg	if I have 2 branches with a common parent, what is the simplest way to rebase one on top of another	02:22
SethRobertson	Just do it?	02:23
	git rebase -p theotherbranch	02:23
	→ thearthur joined	02:23
	← tharkun left	02:25
	→ tharkun joined	02:25
	→ alexim joined	02:25
	← m0ve left	02:25
	→ weebs__ joined	02:25
	→ mritz joined	02:26
	← bodom left	02:26
	→ bodom_ joined	02:26
	← weebs__ left	02:26
	← bitkiller left	02:26
	→ wassa joined	02:26
	gitinfo set mode: +v	02:27
wassa	anyone have experience getting cruisecontrol to pull from a private repo on github? resources?	02:27
SethRobertson	There is of course nothing magic about github. Any usage of git would be equivalant	02:28
wassa	?	02:28
	← bodom_ left	02:29
	→ herdingcat joined	02:29
	← AaronMT left	02:29
	← sporous left	02:30
	← yoklov left	02:30
	→ sporous joined	02:30
	← Araxia left	02:30
	→ juvenal joined	02:31
	→ yoklov joined	02:31
	→ dv310p3r joined	02:32
	→ weiyang joined	02:32
SethRobertson	When you are doing google research, don't restrict your queries to github. The problem barely has anything to do with git, let alone github.	02:32
	→ harish joined	02:32
	→ JoeyJoeJo joined	02:33
JoeyJoeJo	How can I clone a repo at a specific commit, not the most recent commit/	02:34
	?	02:34
SethRobertson	Typically you clone the repo and then `git checkout SHA-1`	02:34
frogonwheels	JoeyJoeJo: when you clone a repo, you get the whole repo.	02:34
	← dabd left	02:34
	→ jackh joined	02:34
	← kcj left	02:34
	← orafu left	02:36
	→ orafu joined	02:36
	→ maletor joined	02:36
	→ cybersphinx joined	02:38
	→ igotnolegs joined	02:39
	→ sirdancealot joined	02:39
	→ dfr\|mac joined	02:40
	← the_cybersphinx left	02:41
	← alexim left	02:41
	← hwrd\|work left	02:41
	→ hwrd\|work joined	02:42
	← hwrd\|work left	02:49
	→ brokenshakles joined	02:50
	→ robbyoconnor joined	02:50
	← tharkun left	02:50
	→ tharkun joined	02:51
	→ alexim joined	02:52
	→ kerrick joined	02:52
	→ kcj joined	02:53
	← kcj left	02:53
	→ kcj joined	02:53
	← rgwu left	02:54
	→ mwrosen joined	02:55
	→ hwrd\|work joined	02:57
	→ kadoban_ joined	02:57
	→ Yuuhi` joined	02:58
	→ rchavik joined	02:59
	← Yuuhi left	03:00
	← brokenshakles left	03:01
	→ envi_ joined	03:03
	→ dreiss joined	03:05
	← ScottNYC left	03:06
	← tharkun left	03:08
	← jackh left	03:08
	→ jackh_ joined	03:08
	← TheShagg left	03:12
	← txomon\|nqas left	03:13
	→ drup4l_n00b joined	03:13
	→ tharkun joined	03:13
	→ txomon\|nqas joined	03:15
	← cyb3r3li0g left	03:15
	→ chrisgeorge joined	03:16
	← jameswarren left	03:16
	→ SirFunk joined	03:16
	→ jameswarren joined	03:17
	← jameswarren left	03:17
	← tharkun left	03:18
	→ irqq joined	03:20
	→ neunon joined	03:23
	← Orbitrix left	03:24
	→ tharkun joined	03:24
	→ edude03 joined	03:24
	← elijah left	03:24
	→ Orbitrix joined	03:25
	← replore_ left	03:25
	→ cyb3r3li0g joined	03:26
	← subbyyy left	03:27
	← chrisgeorge left	03:27
	← kcj left	03:28
	← tharkun left	03:29
	→ davidbanham joined	03:29
	→ joshsmith joined	03:31
	→ subbyyy joined	03:33
	→ brokenshakles joined	03:34
	→ tharkun joined	03:34
	← apok left	03:35
	← Sahua left	03:36
	← JohnyBaby left	03:36
	← acl_ left	03:37
	← tharkun left	03:38
	← OOPMan left	03:38
	← sophacles left	03:38
	← Chib left	03:42
	→ tharkun joined	03:44
	→ alanp_ joined	03:44
	→ kcj joined	03:46
	→ ToBeFree joined	03:46
	→ sophacles joined	03:47
	→ acl_ joined	03:47
	← alanp left	03:47
	→ theDrumm_ joined	03:47
	→ Grunewald joined	03:49
	← theDrumm left	03:50
	← tharkun left	03:51
	→ tharkun joined	03:54
	← irqq left	03:55
	← weiyang left	03:57
	← drev1 left	03:57
	← az left	03:57
	← maletor left	03:57
	→ apok joined	03:59
	← davidbanham left	04:00
	→ az joined	04:01
	← dreiss left	04:01
	→ brokenshakles__ joined	04:02
	→ dreiss joined	04:02
	← brokenshakles left	04:02
	→ ivanoats joined	04:03
	→ __class__ joined	04:03
	← ivanoats left	04:03
	→ ivanoats joined	04:03
	← tharkun left	04:04
	← leonardotartari left	04:04
cccaldas	hi guys, how can i show the tag on git log --pretty ?	04:06
	?	04:07
	← _huoxito left	04:09
	→ apok_ joined	04:09
	← dreiss left	04:09
	→ tharkun joined	04:09
	← d3vic3 left	04:10
	← apok left	04:11
	apok_ → apok	04:11
	→ h0bbit joined	04:11
	→ d3vic3 joined	04:11
	← mcluskydodallas left	04:15
offby1	cccaldas: try --decorate	04:15
	← lonewulf` left	04:15
	← flippo left	04:16
	← bgerber left	04:16
	→ JohnyBaby joined	04:17
	→ Per4u3e joined	04:17
cccaldas	offby1: thkx	04:17
	→ lonewulf` joined	04:18
	← mritz left	04:19
	← tharkun left	04:22
	→ bgerber joined	04:22
	→ enigmabomb joined	04:22
enigmabomb	Hello	04:22
gitinfo	enigmabomb: 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.	04:22
	→ iamntz joined	04:22
	← JohnyBaby left	04:23
enigmabomb	I just switched my repo from being owned by a person to being owned by an organization. Now I can't commit to it. how do I change my information to commit from my personal account?	04:23
	→ Athanasius joined	04:23
	← Athanasius left	04:23
	→ mcluskydodallas joined	04:24
	→ yappy joined	04:25
EugeneKay	enigmabomb - !github	04:25
gitinfo	enigmabomb: For support with Github-specific features (Forks, Pull Requests, Wikis, etc) please /join #github and ask them. Feel free to ask us, but there are no guarantees.	04:25
Grunewald	enigmabomb where did you "switch" that? Git doesn't really have any permission system built in	04:25
	← iwoj left	04:25
	→ kershal joined	04:25
enigmabomb	It's github, many apologies.	04:25
Grunewald	no worries	04:25
	← kukks left	04:25
	← alexim left	04:26
	→ JohnyBaby joined	04:26
	← kershal left	04:26
	← madewokherd left	04:27
	← joeytwiddle left	04:27
	← cccaldas left	04:27
	→ iwoj joined	04:28
	→ Fandekasp joined	04:28
	→ skylamer` joined	04:30
	→ apok_ joined	04:30
	→ d00d3 joined	04:31
	→ jameswarren joined	04:31
	← JohnyBaby left	04:32
	← apok left	04:33
	apok_ → apok	04:33
	→ tharkun joined	04:37
	← iamjarvo left	04:37
	← acl_ left	04:38
	← yappy left	04:39
	← d00d3 left	04:40
	→ maleknet joined	04:42
	← fission6 left	04:43
	← mcluskydodallas left	04:44
	→ sangi joined	04:44
	→ drev1 joined	04:45
	→ subbyyy1 joined	04:45
	← subbyyy left	04:45
	← subbyyy1 left	04:45
	← const left	04:45
	→ beautifulmind joined	04:46
	→ apok_ joined	04:46
	← Slix` left	04:47
	← thearthur left	04:48
	← EyesIsMine left	04:48
	← neurodrone left	04:48
	← apok left	04:49
	apok_ → apok	04:49
	→ variable joined	04:49
	← yacc left	04:49
	→ yacc joined	04:50
SerajewelKS	is there an argument to git-log similar to --source but that will show every ref (not just one) and only on the commits that the ref points to?	04:50
	--source will show exactly one ref that the commit is reachable from. i just want to see where each branch is. (basically, trying to get a visual graph similar to gitk's, but on the terminal. "git log --all --source --graph" is close, but not quite there.)	04:51
	derp, --decorate	04:52
	← ToBeFree left	04:52
SerajewelKS	once again, i solve my problem only after asking	04:52
	← harish left	04:52
	→ harshpb joined	04:52
	← daniel-s__ left	04:53
	← ivanoats left	04:54
	→ ivanoats joined	04:57
	→ isao joined	04:58
	← pppaul left	04:58
	→ pppaul joined	04:59
	→ Ecesis joined	05:00
	→ superdmp joined	05:01
	← ivanoats left	05:01
	→ michaelmartinez joined	05:01
	→ shruggar joined	05:02
	→ khmarbaise joined	05:02
	← shruggar left	05:02
	← Maccer left	05:03
	← twisted` left	05:03
	← malex left	05:03
	← JonasF left	05:03
	→ JonasF joined	05:03
	→ malex joined	05:03
	← antix left	05:04
	← Milossh left	05:07
	← edude03 left	05:07
	← rawtaz left	05:08
	← et left	05:09
	→ Milossh joined	05:09
	← iwoj left	05:09
	→ neurodrone_ joined	05:11
	→ Sigma joined	05:11
	→ antix joined	05:11
	→ et joined	05:12
	← jimubao left	05:12
	→ rawtaz joined	05:12
SerajewelKS	is there a difference between "git fetch" and "git remote update"?	05:14
	← rchavik left	05:14
SerajewelKS	rather, "git fetch --all" and "git remote update"	05:14
	← tharkun left	05:15
	← brokenshakles__ left	05:15
	← wassa left	05:16
	→ weiyang joined	05:17
	← michaelmartinez left	05:17
	→ ivanoats joined	05:18
	→ michaelmartinez joined	05:18
	← ivanoats left	05:18
	→ ivanoats joined	05:18
	→ davidbanham joined	05:18
	← ivanoats left	05:19
	→ tharkun joined	05:21
	← kerrick left	05:21
	← khmarbaise left	05:22
	← adamm left	05:22
	← cehteh left	05:24
	← drev1 left	05:24
	→ subbyyy joined	05:24
	← subbyyy left	05:25
	→ subbyyy joined	05:25
	→ iwoj joined	05:26
	← tharkun left	05:27
	→ oogatta joined	05:29
	→ tharkun joined	05:31
	→ jceb joined	05:32
	→ brokenshakles joined	05:32
	→ Dave^\| joined	05:33
	← iwoj left	05:33
	← neurodrone_ left	05:35
	→ arietis joined	05:36
	→ fission6 joined	05:36
	→ airborn joined	05:38
	→ mineiro joined	05:38
	→ twisted` joined	05:39
	→ Maccer joined	05:41
	← tharkun left	05:41
	→ rtjure joined	05:41
	← jceb left	05:41
	← davidbanham left	05:42
	→ martinjlowm joined	05:43
	→ _sundar_ joined	05:44
	← macmartine left	05:45
	← lausi left	05:46
	→ lausi joined	05:46
	→ tharkun joined	05:47
	→ davidbanham joined	05:47
	→ Unirgy joined	05:47
	→ resmo_ joined	05:48
	→ dr0id joined	05:48
	← dr0id left	05:48
	→ dr0id joined	05:48
Unirgy	hi, does anyone use git subtree? how would I push the changes back to the original lib repo?	05:48
	→ emptyflask joined	05:48
Unirgy	is it just as simple as git subtree push?	05:48
_ikke_	!subtree	05:49
gitinfo	The git subtree merge method is ideal to incorporate a subsidiary git repositories directly in to single git repository with "unified" git history, where you only need to pull changes in from external sources not contribute your own changes back (which if technically possible is at least difficult). See http://progit.org/book/ch6-7.html Type "!subtree_alt" for more options	05:49
_ikke_	subtree merges aren't really meant to contribute back	05:50
	← twisted` left	05:50
	← hwrd\|work left	05:50
	← dfr\|mac left	05:50
Unirgy	_ikke_: thanks but this page looks outdated, doesn't use git subtree * syntax.. so I've seen `git subtree push` which suppose to do what I'm after, but I didn't find any examples in txt	05:51
	the library is evolving together with the project, that's why i need to push back	05:51
	→ LongBeach joined	05:51
	← tharkun left	05:52
Unirgy	is there anything else that would suite my needs? submodule doesn't do anything beyond just having repo inside copy.	05:52
_ikke_	Unirgy: There is also !gitslave	05:53
gitinfo	Unirgy: gitslave (http://gitslave.sf.net) is useful to add subsidiary git repositories to a git superproject when you control and develop on the subprojects at more or less the same time as the superproject, and furthermore when you typically want to tag, branch, push, pull, etc. all repositories at the same time.	05:53
_ikke_	but I wouldn't have a clue if it's suitable for you	05:53
	← Grunewald left	05:53
Unirgy	thanks!	05:53
	looks like exactly what i need	05:54
_ikke_	Never used it, so can't help you with it	05:55
	← michaelmartinez left	05:55
	→ KWKdesign_ joined	05:57
	→ tharkun joined	05:57
	← thursday left	06:03
	← davidbanham left	06:03
frogonwheels	Unirgy: that's not quite true about submodule.. but up to y ou	06:04
	← skylamer` left	06:04
	→ ferdna joined	06:04
frogonwheels	Unirgy: when you commit a submodule change to the parent module, you commit the sha1 of the submodule into the parent module.	06:04
Unirgy	frogonwheels: basically i found learning the intricacies/fighting with it not worth the trouble. having (no branch) commits, not updating always, etc	06:05
frogonwheels	Unirgy: Which means if you checkout an older commit in your main repo, do git submodule update - then the submodule will be in the same state as when you comitted that main repo commit	06:05
	→ davidbanham joined	06:05
Unirgy	frogonwheels: what's the point of (no branch) state?	06:06
_ikke_	Unirgy: That you always have the same version in all repositories	06:07
	← alanp_ left	06:07
frogonwheels	Unirgy: it checks out a commit, not a branch	06:07
Unirgy	well i guess i just don't have enough brain real estate left to understand this	06:08
	→ marthinal joined	06:08
	← martinjlowm left	06:08
_ikke_	Unirgy: When you add a submodule to git, you say: "I want this specific version of the submodule"\	06:08
frogonwheels	Unirgy: when you have a branch checked out, HEAD points to ref:/refs/heads/branch .. if you check out a commit, then HEAD points to SHA1	06:08
_ikke_	Unirgy: Because you don't want your app to break when upstream adds commits	06:09
	← LucasTizma left	06:09
frogonwheels	Unirgy: !detached	06:09
gitinfo	Unirgy: A detached HEAD occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	06:09
	→ ThomasLocke joined	06:10
	→ martinjlowm joined	06:10
	→ alanp joined	06:10
	→ jceb joined	06:10
Unirgy	thanks for the info. given all the needs i have for managing modules, i think for now i'll work with subtree, and then if i'll have time i'll revisit submodules and/or gitslave	06:11
	← davidbanham left	06:11
	→ LucasTizma joined	06:12
	→ mishok13 joined	06:13
	← h0bbit left	06:13
	← KWKdesign left	06:13
	← LucasTizma left	06:16
	← ferdna left	06:17
	→ cbreak-work joined	06:17
	→ papegaaij joined	06:19
	→ archdori joined	06:19
	← ThomasLocke left	06:20
	← mishok13 left	06:22
	← Fandekasp left	06:23
	← mwrosen left	06:24
	← subbyyy left	06:24
	← welly left	06:25
	← josefig left	06:25
	→ senny joined	06:25
	→ ph^ joined	06:25
	→ dreiss joined	06:27
	← archdori left	06:27
	→ harish joined	06:28
	→ EyesIsMine joined	06:29
	→ mastro joined	06:30
	→ phluks joined	06:30
	← brokenshakles left	06:30
	← Rafusek left	06:31
	← igotnolegs left	06:31
	→ anderse_ joined	06:32
	→ dc5ala joined	06:32
	→ Rafusek joined	06:33
	← Milossh left	06:33
	→ flijten joined	06:33
	← Chaser left	06:33
	← kaos01_ left	06:36
	← tharkun left	06:36
	→ hwrd\|work joined	06:36
	→ Chaser joined	06:36
	→ rchavik joined	06:37
	→ XMLSDK joined	06:40
	→ mikepack joined	06:40
	→ Milossh joined	06:41
	→ giallu joined	06:42
	→ tharkun joined	06:42
	← bowarv left	06:42
	→ iwoj joined	06:44
	← pppaul left	06:45
	→ pppaul joined	06:45
	← airborn left	06:45
	← mikepack left	06:45
	→ mishok13 joined	06:46
	← cafaro left	06:46
	→ mikepack joined	06:46
	← yoklov left	06:46
	→ cafaro joined	06:46
	← cafaro left	06:46
	→ cafaro joined	06:46
	→ Fandekasp joined	06:48
	→ ehsan joined	06:50
	← mikepack left	06:51
	← emptyflask left	06:51
	← tharkun left	06:52
	→ pantsman joined	06:55
	→ amcsi_work joined	06:55
	→ afuentes joined	06:55
diegoviola	hmm, i can't upload files to github bug tracker?	06:56
	logs, etc	06:56
	i could just use markdown i guess?	06:56
Silex	yes, or create a gist and link to it	06:56
diegoviola	ok, ty	06:57
	← amcsi_work left	06:57
	→ amcsi_work joined	06:57
	→ tharkun joined	06:57
	← hwrd\|work left	06:57
	→ brokenshakles joined	06:58
	→ marhaban joined	06:59
	← Nibato left	06:59
	→ txomon\|n1as joined	06:59
	← txomon\|nqas left	07:00
	← Bigcheese left	07:01
	→ Bigcheese joined	07:01
	→ DofLeFou joined	07:04
	→ valtih joined	07:06
	→ Nibato joined	07:06
	← txomon\|n1as left	07:06
	→ txomon\|nqas joined	07:08
	← txomon\|nqas left	07:08
	→ txomon\|nqas joined	07:08
	← marhaban left	07:08
	→ h0bbit joined	07:08
boob00	hello	07:09
gitinfo	boob00: 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.	07:09
boob00	is there good document over the net for migrate working SVN to working GIT?	07:09
	← oogatta left	07:10
	← iamheavymetal left	07:10
	→ thiago joined	07:11
	← daaku left	07:12
_ikke_	boob00: one time conversion?	07:13
	boob00: Maybe this helps: http://help.github.com/import-from-subversion/	07:14
	→ alecm_ joined	07:15
	→ thiago_ joined	07:15
	← robbyoconnor left	07:15
boob00	_ikke_: Thanks, But I will mention that I want keep the SVN A live :)	07:15
	← thiago left	07:15
	← alecm_ left	07:16
	← thiago_ left	07:16
	→ OOPMan1 joined	07:16
	← kcj left	07:17
_ikke_	Then just use git svn	07:17
	→ sgronblom joined	07:17
	← DofLeFou left	07:17
	→ thiago joined	07:17
serban	is it just me or the git plugin for eclipse is just horrible?	07:18
sgronblom	How should I do my prepare-commit-msg hook so that I can close the editor without editing the message and have the commit be aborted.	07:18
	← bigkm left	07:19
	← pantsman left	07:19
_ikke_	sgronblom: Not	07:19
	sgronblom: The hook is applied before the editor is invoked	07:19
	← Lgb left	07:20
	→ xhr joined	07:20
sgronblom	_ikke_: yeah but if I overwrite the message file then git will make the commit even if I didn't make any manual changes to the commit file... in some cases you might want that though i guess	07:20
_ikke_	sgronblom: So it can only be used to modify the message, not change the behaviour after the editor is closed	07:21
sgronblom	yeah so if you have a prepare-commit-msg hook that changes the file you will always end up with a commit...	07:22
	← xhr left	07:22
	→ irqq joined	07:22
	→ Lgb joined	07:22
sgronblom	well i can always reset	07:22
_ikke_	sgronblom: The only way to abort the commit in the editor is to result in an empty commit message (delete everything)	07:23
sgronblom	ah that works fine too	07:24
	← thiago left	07:25
	→ skylamer` joined	07:26
	← brokenshakles left	07:26
	→ veyres joined	07:26
	← dvide left	07:28
	→ rtuin joined	07:29
	← jameswarren left	07:29
boob00	_ikke_: the svn2git utility is ruby script?	07:30
	_ikke_: the svn2git utility is ruby script?	07:30
	sorry for twice here :)	07:30
	← tharkun left	07:32
	→ tharkun joined	07:32
_ikke_	boob00: It's a perl script afaik	07:33
	← Per4u3e left	07:34
_ikke_	But it's integrated into git	07:34
	← yacc left	07:36
	→ yacc joined	07:37
	← pjn_oz left	07:38
	← drup4l_n00b left	07:38
	→ TheHaven joined	07:39
	← dr_lepper left	07:40
	→ oogatta joined	07:41
	← WillMarshall left	07:42
	← dreiss left	07:46
	← enigmabomb left	07:46
	← babilen left	07:47
	→ dr_lepper joined	07:47
	→ snickersnack joined	07:48
	← XMLSDK left	07:48
	← mastro left	07:49
	→ BarryCrayon joined	07:49
	← tharkun left	07:49
	→ enigmabomb joined	07:50
	→ kcj joined	07:50
	← thomas_b left	07:52
	→ tharkun joined	07:55
	→ adhawkins joined	07:56
	→ jds_ joined	07:56
	← enigmabomb left	07:57
	← beautifulmind left	07:57
	→ shruggar joined	07:58
	← anderse_ left	07:58
	← jds left	07:58
	→ seoaqua joined	07:59
	← Kicer86 left	07:59
	→ Rafusek_ joined	07:59
	← seoaqua left	07:59
	← snickersnack left	07:59
	→ seoaqua joined	07:59
	← Rafusek left	08:00
	→ Czesiu joined	08:00
	← virtuose left	08:00
Silex	I have a big history that I'd like to cleanup with git rebase -i. However I have merge commits on top of this history that I don't want to lose. What are the suggestions?	08:01
	git rebase -i -p fails	08:01
	I was thinking of creating a branch taht points to where before the merge occur, then rebase -i that one, then rebase -p --onto or smth the stuffs with merges	08:02
_ikke_	Silex: You basically have to reacreate the merges manually	08:02
	s/reacreate/recreate/	08:02
Silex	(what I want to cleanup is like 50+ commits named "Work in progress")	08:03
_ikke_	Are there many merges?	08:03
Silex	_ikke_: it's merge commits that would otherwise be fine with --ff	08:03
_ikke_	yeah, but rebasing doesn't work well with merges	08:03
Silex	they were created with --no-ff following the git successful branching model	08:03
	I thought rebase -p worked "fine"	08:03
	except with -i	08:03
	→ aspotashev\| joined	08:03
_ikke_	yeah, that's what I meant	08:04
	← aspotashev left	08:04
Silex	I think my idea would work	08:04
_ikke_	Might, no experience	08:04
	→ mastro joined	08:04
Silex	given A-B-C-D-MERGESHITSTARTSHERE	08:04
	I create a branch on D	08:05
	rebase the shit out of it	08:05
	then rebase -p on D'	08:05
	← xian9fu left	08:05
Silex	with the range MERGESHIFTSTARTSHERE	08:05
	I'll try. Thanks anyway :)	08:05
_ikke_	Just make a backup and !tias :P	08:06
gitinfo	Try it and see™. You learn much more by experimentation than by asking without having even tried. http://sitaramc.github.com/1-basic-usage/tias.html may help with git-specific TIAS.	08:06
	→ Kicer86 joined	08:06
Silex	now I have to go deep in the abyss of --onto	08:06
	→ flype joined	08:07
	← changedNicks left	08:08
	→ changedNicks joined	08:08
	changedNicks → kindHuman	08:08
	← mastro left	08:09
	← sig^ left	08:10
	← isao left	08:10
	→ mastro joined	08:12
	← dr_lepper left	08:13
	→ LemonBE\|work joined	08:13
	→ dr_lepper joined	08:15
	← dr_lepper left	08:15
	→ dr_lepper joined	08:15
	→ ThomasLocke joined	08:15
	→ LimeBlast joined	08:15
	→ Vert joined	08:16
	← optixx left	08:18
	← Vile left	08:18
	→ arvind_khadri joined	08:20
	→ tsdh joined	08:20
	→ babilen joined	08:22
	← ThunderChicken left	08:23
	→ ThunderChicken joined	08:27
	← mineiro left	08:27
_ikke_	onto is nothing special	08:28
	→ mineiro joined	08:28
	← mineiro left	08:28
	→ mineiro joined	08:28
_ikke_	you are just explicitly telling where to base onto	08:28
	← flype left	08:31
veyres	_ikke_: Thank you for all the interesting url you give with gitinfo. I'm a noob with git and it helps a lot.	08:33
	→ flype joined	08:33
	← Rafusek_ left	08:33
	← iwoj left	08:34
	← Kicer86 left	08:37
	→ marcogmonteiro joined	08:37
	← flype left	08:38
	← zdk left	08:39
	→ thomas_b joined	08:39
	→ zdk joined	08:39
	→ stodan joined	08:39
	→ iwoj joined	08:39
	← iwoj left	08:40
	→ tatsuya_o joined	08:40
_ikke_	That's why it's there. It helps us to give information very easily	08:40
Silex	damnit, git rebase --onto expects a common parent	08:40
	← apok left	08:41
_ikke_	does it?	08:41
	Might be	08:41
	→ Pupeno_W joined	08:41
	→ dark_element joined	08:41
	→ iwoj joined	08:42
	← stodan left	08:42
afuentes	how to use git rm -rn path? whats the -r recursive for if git does not track directories?	08:43
	← kadoban_ left	08:43
afuentes	it gives me an error fatal: pathspec 'test' did not match any files :/	08:43
	→ Kicer86 joined	08:43
	← thomas_b left	08:44
	← ThunderChicken left	08:44
Silex	how do I merge two completely unrelated branches while keeping merges?	08:44
	→ thomas_b joined	08:45
	→ optixx joined	08:45
selckin	define merge in that context	08:45
	→ iwoj_ joined	08:45
	← lekefly_ left	08:46
FauxFaux	With "git merge otherbranch"? (Or commit-tree _¬_¬_¬)	08:46
	← iwoj left	08:46
	→ tcurdt joined	08:46
	← tcurdt left	08:46
	← keltvek left	08:46
	← harshpb left	08:47
	→ anderse_ joined	08:47
	→ noisebleed joined	08:48
	← noisebleed left	08:48
	→ noisebleed joined	08:48
	→ ThunderChicken joined	08:48
	← babilen left	08:48
	→ stodan joined	08:49
	← tharkun left	08:49
	→ chrischambers joined	08:50
	→ yappy joined	08:51
zendeavor	yurgh, screwed the pooch on a merge	08:51
	what is the command i'm looking for to revert back to a prior commit/	08:51
	git reset?	08:52
_ikke_	zendeavor: yes	08:52
	zendeavor: as long as you haven't published (pushed) it yet	08:52
Silex	FauxFaux: ah right, thanks	08:53
zendeavor	shit i think i did by reflex	08:53
	← preflex left	08:53
Silex	btw I found that "git rebase -i -p" works actually fine as long as you only do "fixup" and not reorder commits	08:53
zendeavor	_ikke_: can i use something else to step back	08:53
	← sporous left	08:53
Silex	now of course the problem is taht I want to flatten some merge commits but not others	08:53
_ikke_	Well, there is git revert, but you have to look out with merges	08:53
	Silex: I guess that's what's described in the bugs part of the man page	08:54
zendeavor	yeah i thought i checked my files but i definitely missed some crucial ones	08:54
Silex	_ikke_: yep	08:55
	→ tharkun joined	08:55
	→ pro_metedor joined	08:55
Silex	oh, of course I lost all my tags too	08:56
	→ keltvek joined	08:56
pro_metedor	how can I move a commit from top of my branch to new branch ?	08:56
	bolt → bolt_	08:56
	bolt_ → bolt	08:56
FauxFaux	pro_metedor: !fixup	08:56
gitinfo	pro_metedor: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	08:56
_ikke_	pro_metedor: git checkout -b new_branch	08:56
pro_metedor	I want to get back actual branch by one commit and push my changes into new branch	08:56
zendeavor	okay let me get this right. if i want to merge a branch into master with intent to essentially make the branch the new master, what is the proper step?	08:57
	← Textmode left	08:57
FauxFaux	git branch -f master master~ #!	08:57
zendeavor	checkout master, merge branch?	08:57
FauxFaux	zendeavor: Yep.	08:57
	→ Desproges joined	08:58
pro_metedor	how can I move last commit into new branch and delete it form existing branch ?	08:58
zendeavor	okay looks good now	08:58
_ikke_	pro_metedor: You have to think in terms of where the branch points to	08:59
zendeavor	FauxFaux: _ikke_ thanks guys. git still frightens me with its power ha	08:59
	→ dpino joined	08:59
_ikke_	pro_metedor: A commit doesn't really 'belong' to a branch	08:59
	→ Vile joined	08:59
pro_metedor	yeah, commit is only a change	08:59
_ikke_	not really	08:59
	Well, not in the meaning that it's a diff	09:00
	→ Rafusek joined	09:00
_ikke_	It's rather a snapshot	09:00
afuentes	i untracked some files from master... but i still track them in branches... how to checkout branches now? it says its going to overwrite files if i do checkout :S	09:00
pro_metedor	ok	09:00
	→ Sitethief joined	09:00
_ikke_	pro_metedor: git branch <new_branch>; git reset --keep HEAD~; git checkout <new_branch>	09:00
	afuentes: That's a tough problem	09:01
afuentes	_ikke_, thanks	09:03
	← Unirgy left	09:03
afuentes	:) i manage to break stuff often	09:03
	no matter how rock-solid is the design	09:03
_ikke_	afuentes: git can't overwrite those files (because it doesn't track them anymore)	09:03
	So it doesn't know what to do	09:03
afuentes	well, neither do i	09:04
	is there a way to remove those files form the first commit on from the branches without checking out?	09:05
	that would solve it :)	09:06
_ikke_	afuentes: man git filter-branch	09:06
gitinfo	afuentes: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html	09:06
	→ berserkr joined	09:08
	← harish left	09:09
	→ NetJunky joined	09:09
	→ d0k joined	09:10
	→ achadwick joined	09:10
	← Rafusek left	09:11
	→ Rafusek joined	09:11
	→ linusoleander joined	09:12
afuentes	_ikke_, that sounds complicated :S what about git rebase branch? wouldnt that put the rm commit to the head without checking out?	09:12
	→ LekeFly joined	09:14
	→ JohnyBaby joined	09:14
	→ francisfish joined	09:14
	gitinfo set mode: +v	09:14
NetJunky	Hello!	09:14
gitinfo	NetJunky: 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.	09:14
afuentes	nah, it tries to rewind, so it does not allow to do each checkout... damn	09:15
	ill read more about git filter-branch... to try to figure it out what it does	09:15
NetJunky	@afuentes can I ask you about git-scm?	09:15
afuentes	NetJunky, wel, you can certanly ask the channgel about git-scm	09:16
	NetJunky, just what gitinfo told you if you read it :)	09:17
NetJunky	:)	09:17
	Okey.	09:17
	← jds_ left	09:17
Silex	ah, it worked! \o/ git rebase --onto rebased_branch_before_A A~ master	09:19
	→ isao joined	09:19
	→ archdori joined	09:19
	← arvind_khadri left	09:20
	← TheHaven left	09:20
_ikke_	Silex: Nice	09:20
	← iwoj_ left	09:20
	→ nenjordi joined	09:21
Silex	_ikke_: yeah, basically rebase the chunks you want individually each in its own branch then "rebase --onto" the hell out of it	09:22
zendeavor	man merging branches in git is easy but tedious huh	09:22
_ikke_	zendeavor: In what way tedious?	09:22
	→ eIf joined	09:22
esc	http://sprunge.us/aJUU	09:23
	help?	09:23
	← Fandekasp left	09:23
cbreak-work	it should not be tedius	09:23
	← linusoleander left	09:23
cbreak-work	esc: what's up?	09:23
zendeavor	well i'd setup all these branches something like a month ago, and then redid my sandbag branch from the bottom up and now want to essentially make it the master branch from which to tweak the rest of the other branches.	09:23
	so there's a month worth of commits, maybe longer, that i'm having to merge around and stuff before i've got clean working branches of everything	09:24
cbreak-work	just make new branches when you need them	09:24
	→ rking joined	09:24
esc	i have it falling over bad	09:24
_ikke_	esc: do you have another repository that might have that object?	09:24
cbreak-work	making branches long ahead is usually dumb.	09:24
esc	yeah, but its a different object every time	09:24
	← Vile left	09:25
zendeavor	cbreak-work: i am still learning git, using my dotfiles as a guinea pig for the process	09:25
	and i have branches of 3 different machines	09:25
eIf	hello, can you please help me with this issue? http://pastebin.com/ED4QfR2L thank you a lot	09:25
esc	now i am getting this one: http://sprunge.us/DdYL	09:25
	i think the filesystem is goen haywire	09:26
	→ Vile joined	09:26
FauxFaux	Yeah, have fun with that. :)	09:26
zendeavor	esc: go fsck it right now	09:27
	→ angelsl joined	09:27
esc	lol, its coming in via nfs4	09:27
	→ TheHaven joined	09:27
zendeavor	oh no.	09:27
	→ eijk joined	09:27
esc	my sysadmin will have fun	09:27
	← JohnyBaby left	09:27
	→ sdeboey joined	09:27
esc	zendeavor: http://sprunge.us/fiEU	09:28
	check the end of it	09:28
cbreak-work	zendeavor: well, merging is easy	09:28
zendeavor	esc: yeah doesn't look fun	09:28
cbreak-work	just merge all the things you want into the branch you want them	09:28
	then you're done.	09:28
	you can rename branches too if you want	09:29
	but I recommend you to only make branches for topics, and maybe some very specific other things	09:29
	← diegoviola left	09:29
zendeavor	cbreak-work: yup it's just that i'm doing the activity in a silly way by keeping all these branches at all	09:29
	← troyt left	09:30
	→ TomyLobo joined	09:30
zendeavor	but i don't have root on one system, so my configs need some tinkering	09:30
	← archdori left	09:30
zendeavor	i could manage it as a separate repo, but that's too easy. no learning!	09:30
	→ iljakuklic joined	09:30
	→ alexukf joined	09:30
	← eIf left	09:30
	→ JohnyBaby joined	09:30
	← lapistano left	09:31
	→ lapistano joined	09:31
Silex	hum, one last thing: git rebase -p didn't reuse the commiter date	09:33
	→ cortana2 joined	09:33
	→ andrew293 joined	09:33
Silex	how do I tell it to reuse it too? (for merge commits)	09:33
	→ Element9 joined	09:34
	→ mastroWork joined	09:37
jast	you can't	09:37
	← asakura0 left	09:37
esc	grrr	09:37
jast	newly created commits always use the current time for committer date	09:37
	← mastro left	09:39
	→ asakura0 joined	09:40
	← AlexanderS left	09:40
	→ AlexanderS joined	09:41
	← rtjure left	09:42
	← tharkun left	09:43
	→ linusoleander joined	09:44
	← mineiro left	09:45
	← seoaqua left	09:46
	→ beautifulmind joined	09:46
	← w09x left	09:47
	← jackh_ left	09:47
	→ tharkun joined	09:48
	→ iwoj joined	09:49
esc	i managed to copy my file, just in time	09:49
	lost the history though	09:49
	→ cslcm joined	09:50
cslcm	Hi folks. I have a problem with git returning a "index-pack abnormal exit" error, on a push. I've established that the git-index-pack process does run - but I can't see what the return code is, or what the error is. Is there any way to force some verbosity in index-pack?	09:51
	← weiyang left	09:51
	→ w09x joined	09:51
	→ flype joined	09:52
_ikke_	cslcm: not sure, but maybe EXPORT GIT_TRACE=1	09:53
	s/EXPORT/export/	09:53
	export GIT_TRACE=true	09:53
	→ theCole joined	09:54
	→ Fandekasp joined	09:54
_ikke_	If this variable is set to "1", "2" or "true" (comparison is case insensitive), git will print trace: messages on stderr telling about alias expansion,	09:54
	built-in command execution and external command execution.	09:54
FauxFaux	It might honour GIT_TRACE=1	09:54
_ikke_	Yeah, but numeric values refer to an open file descriptor	09:54
	1 would be STDOUT and 2 would be STDERR then I guess, right?	09:55
FauxFaux	Oh dear.	09:56
_ikke_	FauxFaux: ?	09:56
	→ babilen joined	09:56
	→ Lord_DeathMatch joined	09:56
cslcm	unfortunately that appears to have had no effect	09:56
FauxFaux	cslcm: Is the repo okay? git fsck && git repack # on the remote?	09:57
cslcm	it's a brand new empty repo	09:57
	so yes :)	09:57
FauxFaux	Uh oh.	09:57
_ikke_	FauxFaux: Doesn't sounds good	09:58
FauxFaux	git push file:///a/repo	09:58
cslcm	pushing locally works fine	09:58
	pushing to ssh://user@127.0.0.1 works fine	09:58
	but pushing remotely doesn't	09:58
FauxFaux	I find that statement hard to believe. What else is different?	09:59
cslcm	the data being pushed is different	09:59
	the local tests were just small files	09:59
FauxFaux	How big is the other stuff? i.e. is index pack running out of memory?	10:00
	← JohnyBaby left	10:00
cslcm	ah, that is possible, the server doesn't have a lot of memory	10:00
	the data i'm trying to push is ~200mb	10:01
FauxFaux	Mmm, things are normally fine up to ram+swap size.	10:01
cslcm	there's plenty of swap	10:01
	I strongly suspect a permissions problem (We recently had a sysadmin run a chmod -R 0600 / - needless to say he's gone now)	10:02
	but I have replaced all the git binaries and checked their perms	10:03
	and the repo perms	10:03
	i really need to get some feedback from index-pack	10:03
cbreak-work	wonder what that guy was thinking	10:05
cslcm	I suspect it was a typo / great power great responsibility fail etc etc	10:05
	→ harish joined	10:07
	→ pydonny joined	10:07
	← ph^ left	10:07
cslcm	alright, so i'm gonna move the data to the remote server and try pushing it locally	10:07
et	like the thing on github that had rm /usr /share/mylib in the uninstall script	10:08
	with "-rf" of course	10:08
cslcm	heh	10:09
	← Fandekasp left	10:11
	← linusoleander left	10:11
	← tatsuya_o left	10:13
	→ tatsuya_o joined	10:13
	← Desproges left	10:14
	→ Desproges joined	10:14
	→ tatsuya__ joined	10:14
	← tatsuya_o left	10:14
	← tatsuya__ left	10:14
	→ pjn_oz joined	10:15
	← theCole left	10:16
	← herdingcat left	10:17
	← iwoj left	10:17
	← berserkr left	10:17
	→ avar joined	10:18
	← avar left	10:18
	→ avar joined	10:18
	→ berserkr joined	10:19
	→ Yewbacca joined	10:19
_ikke_	bumblebee or something like that	10:20
	gitinfo set mode: +v	10:21
	← fisted left	10:22
	→ fisted_ joined	10:22
_ikke_	yes, bumblebee, but the repository has been abandoned	10:23
Yewbacca	Hey. Git keeps track of files by hash, but what if a file is both renamed/moved AND slightly modified; how does Git know it is a modified version of an older tracked file? Does it look at a list of all files that it can no longer find that were there before and does a delta compare? Or maybe it doesn't know, and stores the whole file?	10:23
	→ dvide joined	10:23
	← babilen left	10:23
Yewbacca	I ask because I am implementing something similar to how Git works but for an entirely different purpose	10:24
cslcm	Yewbacca: in that case you would normally see a delete and an add in the log	10:24
Yewbacca	Ah, it stores the whole file as if it was new?	10:24
canton7	I think git's rename detection will detect a rename if <50% of the files in question are the same?	10:24
	oops, >50%	10:25
cslcm	you may be right, but it isn't my experience	10:25
Yewbacca	Mm, sounds like a delta compare of a list of missing files to all ne, untracked files then, canton. I think that is how I will do it too.	10:25
	to all new*	10:25
canton7	also, -M and -C to git-log	10:25
cslcm	when i've moved AND changed a file, it's always decided it's a new object. However, this very rarely happens of course.. commit often! :)	10:26
	→ xiangfu joined	10:26
canton7	also --follow	10:26
	← andrew293 left	10:28
Yewbacca	Hehe yeah, commit often. I'm actually implementing a space-saving, incremental installer format that delivers application updates as deltas, and am tracking files via hash (and also variable length hashed chunks), and was working on the problem of renamed and modified files, hence why Git came to mind. Thanks a lot for the info, I will do something similar and if the files are mostly the same I'll decide it's a rename+modify.	10:28
	← yappy left	10:28
cslcm	an sha1 won't cut it for that, a slightly modified file will generate a completely different hash	10:29
	← intripoon left	10:30
jast	Yewbacca: you may want to look at how rsync works	10:30
cslcm	will have to use tigertree or something similar	10:30
	→ babilen joined	10:30
Yewbacca	Yep, I use a main hash for the file itself (fast way to verify updates after deltas are applied, as well as tracking moved files) AND variable-length hashed chunks. The latter moves a sliding window over the file and identifies patterns and only stores them once.	10:30
jast	there are delta libraries you can reuse, in fact	10:31
	such as xdelta	10:31
Yewbacca	Thanks for the suggestion jast but rsync uses this principle I just described, even simpler in fact.	10:31
	Aye	10:31
jast	I know	10:31
Yewbacca	Might use xfelta	10:31
jast	then again xdelta is only good for generating deltas when you have both versions of the file available	10:32
cslcm	I feel that these packages have failed in not adopting the name "delta skelter"	10:32
jast	if you deliver update bundles you don't really need the extra abilities of rsync	10:32
	→ snearch joined	10:32
	→ linusoleander joined	10:33
Yewbacca	Yeah, my system basically keeps both the latest + newly released versions, and derives the deltas from that info, spits out an updater package, and throws away the previous full installer, so the server keeps only the latest full installer, plus incremental breadcrumb updates for all older versions.	10:33
	Rofl cslcm	10:33
	← rtuin left	10:34
	→ simenbrekken_ joined	10:34
	← snearch left	10:34
	← nrichards left	10:35
	→ nrichards joined	10:35
	→ pfFredd joined	10:35
	→ cagedwisdom_ joined	10:35
	← ihptru left	10:36
	→ rtuin joined	10:36
	→ intripoon joined	10:41
	→ mastro__ joined	10:43
	→ pfFred joined	10:45
	← pfFredd left	10:45
	→ ronniy joined	10:46
	← rtuin left	10:46
	← mastroWork left	10:46
	← MissionCritical left	10:47
	← tharkun left	10:49
	→ tharkun joined	10:49
	→ rtuin joined	10:50
	← hexa2k9 left	10:55
	← xiangfu left	10:56
	→ gjoseph joined	10:56
	→ hexa2k9 joined	10:57
	→ blacktulip joined	10:57
	→ marhaban joined	10:58
	→ Mission-Critical joined	10:58
gjoseph	hi. i'm running this git svn clone command with 2 —branches and 2 —tags flags. it sits there for a few minutes, no output, stops, and the result git repo is empty or broken (git log says "fatal: bad default revision 'HEAD'", for example). Now, if I only pass the first of each —branch and —tags, it works fine (except it obviously doesnt fetch the other tags/branches). if i only pass the second ones, it "doesn't work". What's strange is that i have anot	10:59
	splitting init and fetch has the same results btw	11:00
	→ nonuby joined	11:00
	← juvenal left	11:00
	← linusoleander left	11:04
	← j0ran left	11:04
	→ linusoleander joined	11:05
	← linusoleander left	11:06
	→ linusoleander_ joined	11:06
	Mission-Critical → MissionCritical	11:09
gjoseph	ho well in fact, that doesn't work for that other project either. it used to, with git 1.7.9.4, we've upgraded to 1.7.10 since. does that ring a bell ?	11:09
	→ jvasak joined	11:10
	← ben_alman left	11:10
	→ ben_alman joined	11:12
gjoseph	(it says nothing after "Initialized empty Git repository in" — while the working ones output a lot, going through each and every revision etc)	11:13
	→ ph^ joined	11:15
	→ troyt joined	11:17
	← oogatta left	11:19
	→ j0ran joined	11:20
	← amitz left	11:21
	→ amitz joined	11:21
	← pydonny left	11:22
	← squentin left	11:23
	← marhaban left	11:24
	→ squentin joined	11:24
	← mikaelj left	11:24
Silex	"git filter-branch" also rewrite tags... is there such an option for git rebase?	11:25
	← tharkun left	11:25
	→ mikaelj joined	11:26
Silex	(filter-branch does it with the option --tag-name-filter cat)	11:26
	← sangi left	11:27
	→ teotwaki joined	11:27
	← Yewbacca left	11:27
	→ tvw joined	11:31
	→ tharkun joined	11:31
	→ Peste_Bubonica joined	11:31
	→ Per4u3e joined	11:33
	→ syamata joined	11:35
	→ b00b00 joined	11:36
	→ buttsies joined	11:36
	→ edude03 joined	11:36
	→ chayin_ joined	11:36
	← tomex left	11:36
	→ w09x_ joined	11:37
	→ Droggl_ joined	11:37
	→ Doobie_ joined	11:37
	← pjn_oz left	11:37
	→ roland_ joined	11:37
	→ Yuuhi`` joined	11:37
	← dv310p3r left	11:39
	→ Shanacha1 joined	11:39
	→ shabble_ joined	11:39
	→ PerlPilot joined	11:39
	→ eno_ joined	11:39
	← shabble_ left	11:39
	→ shabble_ joined	11:39
Silex	ahah! I'd use the post-rewrite hook apparently	11:39
	→ rook2paw1 joined	11:39
	→ doki_pen_ joined	11:39
	→ Serajewe1KS joined	11:39
	→ Nom-_ joined	11:39
	→ heptadecagram joined	11:39
	→ Smirnov joined	11:39
	→ Milossh_ joined	11:39
	← heptadecagram left	11:39
	→ nick_h_ joined	11:39
cmn	if you need automated tag rewrite, you shouldn't be using tags for that	11:39
	→ hanneso_ joined	11:39
	→ Weiss joined	11:39
	→ MattRich1rdson joined	11:39
	→ FabTG joined	11:39
	→ noqqe joined	11:39
	→ chihiro_ joined	11:39
	→ asoltys_ joined	11:39
	→ anildigital_ joined	11:40
	→ ppr joined	11:40
	→ sophacle1 joined	11:40
	→ xxtjaxx_ joined	11:40
	→ tommyvyo_ joined	11:40
Silex	cmn: context is I removed binary files from the repo, and now I want to rebase -i to clean the mess	11:40
	→ tabakhase_ joined	11:40
	→ anoob joined	11:40
Silex	existing tags don't get automatically transfered to the rebased commits	11:41
	→ stodan_ joined	11:41
cmn	and why not filter-branch?	11:41
	which is the tool made for that?	11:41
Silex	I already used filter branch to remove teh big binaries	11:41
	→ iOPK joined	11:41
Silex	now I want to reword commits and squash a lot of them together etc	11:41
	the history is really messy	11:41
	← Element9 left	11:42
	← butts left	11:42
	← hwrdprkns left	11:42
	← beautifulmind left	11:42
	← cslcm left	11:42
	← boob00 left	11:42
	← chayin left	11:42
	← w09x left	11:42
	← Milossh left	11:42
	← OPK left	11:42
	← tabakhase left	11:42
	← roland left	11:42
	← Doobie left	11:42
	← Vert left	11:42
	← tommyvyo left	11:42
	← SeySayux left	11:42
	← maxorator left	11:42
	← anildigital left	11:42
	← teotwaki left	11:42
	← SerajewelKS left	11:42
	← BrainFreak__ left	11:42
	← eno left	11:42
	← thorbjornDX left	11:42
	← Shanachan left	11:42
	← PerlJam left	11:42
	← stodan left	11:42
	← TTSDA left	11:42
	← daethorian left	11:42
	← syamata0 left	11:42
	← Nom- left	11:42
	← et left	11:42
	← sophacles left	11:42
	← FabTG_ left	11:42
	← unreal left	11:42
	← hanneso left	11:42
	← noqqe__ left	11:42
	← MattRichardson left	11:42
	← OOPMan1 left	11:42
	← Yuuhi` left	11:42
	→ Vert joined	11:42
	→ thorbjornDX joined	11:42
	→ unreal joined	11:42
	→ maxorator joined	11:42
	← unreal left	11:42
	→ unreal joined	11:42
	tommyvyo_ → tommyvyo	11:42
	→ Element9 joined	11:42
	→ Zyrkon_ joined	11:42
	→ et joined	11:42
	→ hwrdprkns joined	11:42
	→ cslcm joined	11:42
	→ teotwaki joined	11:42
	→ TheDudeKnows joined	11:43
	→ arvind_khadri joined	11:43
	→ OOPMan joined	11:43
	← arvind_khadri left	11:43
	→ arvind_khadri joined	11:43
	→ markatto joined	11:43
	← chihiro_ left	11:44
	→ maxb joined	11:44
	← aristidesfl left	11:45
	→ chihiro joined	11:45
	→ didi joined	11:45
	→ TTSDA joined	11:46
	← tharkun left	11:46
	→ tatsuya_o joined	11:47
	← chihiro left	11:47
	← nonuby left	11:47
	← LimeBlast left	11:47
	→ LimeBlast joined	11:47
	← linusoleander_ left	11:48
didi	I have a branch called `foo' that has a file called `bar'. I'm at `master' and I want to patch master's `bar' with the changes from the foo's `bar', but only for testing. I'm not ready to create a new commit. Is it possible?	11:48
	→ chihiro joined	11:48
	→ SeySayux_ joined	11:48
_ikke_	What do you mean with patch it with foo's bar	11:49
	← Orbitrix left	11:50
	→ linusoleander joined	11:50
	← lukus left	11:51
didi	_ikke_: I mean, the `bar' file from the `foo' branch has some differences from the `bar' file from the `master' branch. I want those differences merged at the `master' branch.	11:51
	→ lukus joined	11:51
Silex	btw for my problem before I found this: http://blog.nilsjonsson.com/post/4421450571/rebasing-tags-in-git-repositories	11:51
	→ tharkun joined	11:52
	→ NimeshNeema joined	11:52
FauxFaux	Which has a man page reference and an admission that he's silly at the top?	11:52
	← flype left	11:52
	→ PrinceAMD joined	11:52
	← zdk left	11:52
_ikke_	FauxFaux: wat?	11:53
	Droggl_ → Droggl	11:53
	← linusoleander left	11:55
didi	`git checkout foo bar'	11:55
	That will do.	11:56
	← Sigma left	11:56
nevyn	so I'm trying to track multiple patch sets on the kernel	11:56
	→ linusoleander joined	11:56
nevyn	so from some kernel base there's then the rt quilt patches (using quiltimport)	11:56
	← didi left	11:57
nevyn	then a particular topic branch from sound/usb-audio which I'm cherrypicking	11:57
	then a manual patch to fix some hardware stuff.	11:57
	→ Chibby_ joined	11:57
	→ squeakytoy joined	11:57
nevyn	is there a sane way to move that last set of stuff to a new -rt without re-doing all the above.	11:58
	← Desproges left	11:58
	← OOPMan left	11:59
	← tharkun left	12:00
	→ _class_ joined	12:01
	← __class__ left	12:01
Silex	--committer-date-is-author-date doesn't work for git rebase -p	12:01
	any idea around this?	12:01
	← edude03 left	12:01
	→ Desproges joined	12:02
	→ sgncfl joined	12:03
	→ freeed joined	12:04
	← MacGyverNL left	12:05
	← dvide left	12:05
	→ tharkun joined	12:05
Silex	ah, filter-branch should be able to (http://stackoverflow.com/questions/5520425/how-can-i-rebase-a-commit-made-by-another-author-without-adding-myself-as-the-co)	12:05
	_class_ → __class__	12:06
anoob	I've added "website/config.php" to my .gitignore file and then commit the changes and pushed. My changes in this files are not ignored at all, they are still marked. My friend tried in his repo but the same happens...	12:07
	← wilmoore left	12:07
_ikke_	anoob: Git doesn't ignore tracked files	12:08
	!config	12:08
gitinfo	[!configfiles] the recommended way is to change your application so that you can (perhaps optionally) store site-specific configuration/data in separate files, then keep those files out of version control. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.	12:08
	→ dv310p3r joined	12:10
	← anoob left	12:11
	→ d3c joined	12:11
	← krz left	12:12
	shabble_ → shabble	12:13
Vinnie_win	how do I discard all my untracked files and just reset index and working to HEAD ?	12:14
	→ krz joined	12:14
	→ FernandoBasso joined	12:15
	→ iamjarvo joined	12:16
cmn	clean && reset --hard HEAD	12:16
	→ Orbitrix joined	12:17
	← arvind_khadri left	12:18
	→ edude03 joined	12:18
	→ jackh joined	12:19
Vinnie_win	clean	12:19
	→ cccaldas joined	12:19
Vinnie_win	that's the ticket. thanks	12:19
	→ t4nkd`workin joined	12:20
	← asakura0 left	12:20
	Serajewe1KS → SerajewelKS	12:21
	→ jsoa joined	12:24
	← ph^ left	12:25
	← camdenx left	12:26
	→ cehteh joined	12:27
	→ camdenx joined	12:27
	← MichaelRpdx left	12:28
	← nenjordi left	12:29
	→ daethorian joined	12:29
	← alexukf left	12:30
	→ asakura0 joined	12:30
	→ eli joined	12:30
	← ehsan left	12:30
	← camdenx left	12:31
	← sgncfl left	12:32
	→ plasticboy joined	12:33
	→ Sigma joined	12:34
	t4nkd`workin → t4nkd`	12:35
	← linusoleander left	12:36
	← dv310p3r left	12:37
	→ bartek joined	12:37
	→ davv3 joined	12:38
	← sirdancealot left	12:38
	← rurufufuss left	12:39
	← kaen left	12:39
	→ ph^ joined	12:39
	→ davidbanham joined	12:39
	← ph^ left	12:41
	← Chibby_ left	12:42
	→ AaronMT joined	12:42
	← nrichards left	12:44
	→ sgncfl joined	12:47
	→ amigojapan joined	12:47
	→ martinjlowm-lptp joined	12:48
	← pfFred left	12:49
	→ revagomes joined	12:50
	→ glennpratt joined	12:51
	→ subbyyy joined	12:51
	← sgncfl left	12:51
	→ bodom joined	12:52
	← cccaldas left	12:55
	← dr0id left	12:56
	→ glosoli joined	12:56
glosoli	how to check status of pull ? I mean to check if there is any updates available ?	12:56
	→ RaceCondition joined	12:56
	→ anoob joined	12:57
	→ vitor-br joined	12:57
cmn	git remote show tells you about that	12:57
	or you can just fetch and get the new data directl	12:57
glosoli	git remote show sounds fairly	12:58
	thanks cmn wish you a good day	12:58
	← PerlPilot left	12:58
	← kcj left	12:58
	→ TeckniX joined	12:58
	→ bglusman joined	13:00
	→ PerlJam joined	13:00
	← glosoli left	13:00
	← Per4u3e left	13:01
	→ macabre joined	13:01
cslcm	so i've established that the "unpack error" seems to be caused by a certain file being checked in / pushed. I haven't been able to work out which one yet, or why.	13:02
	← davidbanham left	13:04
b00b00	If I ran git svn clone on SVN repo, and it build new git repo, during this time new commits appended to the SVN, is there a command to update the cloned git repo?	13:05
	→ oriba joined	13:05
	→ sq-one joined	13:05
bremner	b00b00: man git-svn fetch	13:05
gitinfo	b00b00: the git-svn manpage is available at http://jk.gs/git-svn.html	13:05
bremner	also git-svn rebase iirc	13:06
b00b00	Thanks	13:06
	← kermit left	13:06
	→ juvenal joined	13:07
	← krz left	13:08
	→ kermit joined	13:08
	← edude03 left	13:08
	→ juanfer joined	13:09
	← ocharles left	13:09
	← maleknet left	13:10
	← glennpratt left	13:10
	→ neurodrone joined	13:10
	→ glennpratt joined	13:11
	→ envi_\|2 joined	13:11
	→ mastroWork joined	13:12
	← subbyyy left	13:12
	→ leonardotartari joined	13:12
	← envi_ left	13:13
	→ edude03 joined	13:13
	→ sgncfl joined	13:15
	← mastro__ left	13:15
malex	k	13:16
	← harish left	13:16
	→ Rickardo1 joined	13:17
	→ ehsan joined	13:18
	→ barrbrain joined	13:18
Rickardo1	I am new to git and now I am fighting with a merge conflict.. when I open the file I see <<<<<<<<< HEAD and some hash ... How am I supposed to handle this? Is there any tool?	13:18
	← Lgb left	13:20
FauxFaux	git mergetool # will launch vimdiff if you know how to operate that. Or, you can just make the file look like you want, save, git add, and declare victory.	13:20
	→ kodie joined	13:20
	→ cannonball joined	13:20
	→ barrbrain_ joined	13:21
	← localhost left	13:21
	← kodie left	13:22
	← mastroWork left	13:22
	→ Lgb joined	13:22
	→ linusoleander joined	13:24
	← barrbrain left	13:24
	barrbrain_ → barrbrain	13:24
	← dpino left	13:24
	← Rickardo1 left	13:26
	→ ocharles joined	13:27
	→ nrichards joined	13:27
	→ tyatpi joined	13:29
	→ pmatthews joined	13:30
	→ nadavoid joined	13:30
	→ elijah joined	13:30
	→ mischat joined	13:31
	← ocharles left	13:31
_ikke_	I can recommend kdiff3	13:31
	I find it easy to use, and quite clear	13:31
	→ mritz joined	13:33
	→ ocharles joined	13:33
	← lcpfnvcy left	13:33
et	p4merge is also quite nice	13:33
	→ erichynds joined	13:34
	← mritz left	13:36
	→ Mowsh joined	13:37
	→ joeytwiddle joined	13:37
	→ Chib joined	13:38
	→ mastroWork joined	13:40
	→ petershaw joined	13:41
	→ w0m joined	13:41
	← oriba left	13:41
	← tharkun left	13:41
	→ tharkun joined	13:41
w0m	.	13:42
	← isao left	13:42
	← gjoseph left	13:42
	← LekeFly left	13:43
	→ ToBeFree joined	13:43
	→ isao joined	13:44
	← luka12345 left	13:44
	← __class__ left	13:45
	→ __class__ joined	13:46
	→ luka12345 joined	13:46
	→ twisted` joined	13:47
	→ yoklov joined	13:49
	Nom-_ → Nom-	13:49
	→ ph^ joined	13:49
	→ hendry joined	13:50
hendry	is there a way to print the location of origin without doing a `cat .git/config`?	13:50
isao	git remote -v	13:51
	→ jkal joined	13:51
	← phluks left	13:51
	→ mritz joined	13:53
hendry	isao: thanks!	13:53
	→ rstory-work joined	13:53
	← isao left	13:54
rstory-work	hi.. quick bisect question... what do you do when a bisect point won't build due to somone else's bug somewhere else? e.g. bisect chose a commit in a series of N commits, and the project won't build at that point..	13:55
_ikke_	skip	13:55
	man git bisecr	13:55
	man git bisect	13:56
gitinfo	the git-bisect manpage is available at http://jk.gs/git-bisect.html	13:56
cbreak-work	so... we had quite a weird problem on windows with git just now	13:56
	git clone of a new repository failed	13:56
	it worked fine on my mac	13:56
	← Dave^\| left	13:56
rstory-work	perfect, thanks! sorry, i should have checked the man page instead of relying on a blog post. :-(	13:56
cbreak-work	but cloning on windows always aborted during the checkout phase	13:56
cmn	are you using non-ASCII chars?	13:57
cbreak-work	no	13:57
cehteh	always? .. i had some transitent bugs in win2000 running under virtualbox	13:57
cbreak-work	on windows 7 64 bit	13:57
	every single one	13:57
	← Shanacha1 left	13:57
cbreak-work	turns out, the folder had an illegal name.	13:57
cehteh	as usual with windows .. trying the same thing again worked	13:57
	→ camdenx joined	13:58
cehteh	proof: windows is nondeterministic :)	13:58
cbreak-work	the name was "aux"	13:58
	(without the quotes)	13:58
cmn	ah, yes	13:58
	← ugi left	13:58
cbreak-work	I mean, what the hell	13:58
et	it's more likely that you don't remember what you did the first time and did something differently now	13:58
cbreak-work	how's that name illegal?	13:58
cehteh	and irregular :)	13:58
cmn	I believe you'll have to thank Mr. DOS for that	13:58
	← bartman` left	13:58
cbreak-work	this is windows 7	13:58
_ikke_	is aux forbidden in windows?	13:58
cbreak-work	based on NT	13:58
	_ikke_: yes.	13:58
	it seems that way	13:58
cehteh	backwards compatibility cows :)	13:59
	bug for bug	13:59
	→ harish joined	13:59
	→ OOPMan joined	14:00
_ikke_	cbreak-work: How did you find out?	14:00
	→ tbird72 joined	14:00
cbreak-work	_ikke_: I tried to make the folder in explorer	14:00
	← rstory-work left	14:01
_ikke_	http://www.blindedbytech.com/2006/11/16/forbidden-file-and-folder-names-on-windows/	14:01
cmn	yeah, Wikpedia says so too: AUX COM1 … COM9 CON LPT1 … LPT9 NUL PRN	14:01
FauxFaux	http://msdn.microsoft.com/en-us/library/windows/desktop/aa365247(v=vs.85).aspx is the official spec.	14:02
	→ drev1 joined	14:02
cehteh	on unix only 'core' sux :)	14:02
	← t4nkd` left	14:02
FauxFaux	Used to be called "Naming a file", which is a much cuter title. ("John.").	14:02
	→ newbie999\|3 joined	14:02
	← gavin_huang left	14:03
	→ t4nkd`workin joined	14:03
	← newbie999 left	14:03
	→ oriba joined	14:04
	← joeytwiddle left	14:04
_ikke_	cehteh: I can create a dir called core on linux	14:04
cehteh	yes	14:04
	← blacktulip left	14:05
	→ bartman` joined	14:05
	← oriba left	14:05
cehteh	but some (few/many/whatever) tools do ugly things :)	14:06
	← mischat left	14:06
cmn	you can create a file as well	14:06
	but if your program dumps core, it might get overwritten	14:06
	← pppaul left	14:08
	← OOPMan left	14:08
	→ joeytwiddle joined	14:09
	→ pppaul joined	14:09
_ikke_	ah ok	14:09
	← Ecesis left	14:10
	→ popcorn joined	14:10
	→ Sahua joined	14:10
	← kpreid left	14:10
	← FernandoBasso left	14:11
	→ valtih_ joined	14:11
	→ intransit[work] joined	14:11
	← intransit[work] left	14:12
	← pro_metedor left	14:12
	← linusoleander left	14:12
	→ Sonja joined	14:12
Sonja	is it possible to compare git commits to instances of "saving", and if so, what is the equivalent of "undo", i.e. return to a specific "save"	14:13
_ikke_	Sonja: Instances of saving?	14:14
bob2	ado you mean stash?	14:14
	← valtih left	14:14
	← mastroWork left	14:14
	→ pro_metedor joined	14:14
	valtih_ → valtih	14:14
cbreak-work	Sonja: no. Why?	14:14
Sonja	i.e. "ok my app works, and i just added a feature. i'll commit this."	14:14
PerlJam	Sonja: you can checkout any commit you want. That's not quite "undo" though.	14:14
	→ FernandoBasso joined	14:14
Sonja	then i dabble and try to add a new feature, but i mess up	14:14
	and i want to roll back to the last functioning version, i.e. the most recent commit	14:14
	→ oriba joined	14:14
cbreak-work	git checkout -p	14:14
PerlJam	Sonja: so checkout that version	14:14
Sonja	thanks	14:15
_ikke_	Sonja: Also: commit often, commit early	14:15
cbreak-work	and always commit running code	14:15
	(or rewrite history to make it appear that way)	14:16
	otherwise bisecting will suck.	14:16
	→ Sahua_ joined	14:16
	← Sahua left	14:16
	← w09x_ left	14:17
	← sdeboey left	14:17
	← j0ran left	14:17
	→ kpreid joined	14:18
Sonja	committing feels like clicking Control-S in LIbreOffice Writer	14:18
	i.e. I'm saving my work	14:18
cmn	but you're not overwriting it	14:18
cbreak-work	also, it's not saving it	14:19
Sonja	maybe a better analogy is edit history of an article in wikipedia	14:19
	you can go and rollback to a particular version of the article	14:19
cmn	for some values of "rollback"	14:19
	← Lord_DeathMatch left	14:19
Sonja	:-S	14:20
	→ whitman joined	14:20
cmn	rollback isn't something that's well-defined in the context of igt	14:20
	→ bitkiller joined	14:20
cmn	and it means something different for different people	14:20
cbreak-work	it's more like copying all files that you track into a new directory	14:20
	somewhere away from your normal files	14:21
	→ shondm joined	14:21
cmn	what's the trigger for the parable? !parable	14:21
gitinfo	'The git parable' provides some good reasoning behind git. http://tom.preston-werner.com/2009/05/19/the-git-parable.html	14:21
Sonja	what's the word for "hmm let's make everything look like git commit "version 3" again, let's restore to back then and undo what i currently have"	14:21
cmn	it's the undo part	14:21
	→ intransit joined	14:21
_ikke_	git checkout -f <version 3>	14:21
cmn	you don't undo it, you split off a new line	14:21
cbreak-work	git reset --hard kills history	14:21
	git checkout visits history	14:21
cmn	eventually, yeah	14:21
cbreak-work	I recommend doing the latter	14:21
Sonja	thank you for the parable	14:21
	and <version 3> here would be that long string that uniquely identifies the commit?	14:22
cmn	the problem with speaking about rolling back when you don't like a particular way the code is progressing, is that it sounds like braches don't exist	14:22
	→ eicto joined	14:22
	→ sirdancealot2 joined	14:22
cmn	and they're a central part of git	14:22
Sonja	i guess i'm not using the branching feature, or somehow i don't intend to	14:23
_ikke_	Sonja: Yet	14:23
Sonja	i'm using git as a way for me to go forward and backward in time when i mess up	14:23
	does branching become relevant when i work in a team?	14:23
_ikke_	Sonja: Even if you don't	14:23
cmn	no, it becomes relevant when you have more than a simple script	14:24
cbreak-work	you have at least one branch	14:24
Wooga	xsl-трансформеры ( http://git.neverb.net/mine/aloe.git/tree/sufd-files/src/main/xslt?id=4ed59d8783a09bb912b75c00b14242cd4602f687 ) берут такие вот файлы: http://git.neverb.net/mine/aloe.git/tree/sufd-files/src/main/generators?id=4ed59d8783a09bb912b75c00b14242cd4602f687 и генерят вот этого монстра :	14:24
	http://git.neverb.net/mine/aloe.git/tree/sufd-files/src/main/resources/aloe/sufdfiles/layer.xml?id=4ed59d8783a09bb912b75c00b14242cd4602f687	14:24
	oops	14:24
	sorry, wrong channel	14:24
	→ chrisgeorge joined	14:25
Wooga	and network...	14:25
	← shondm left	14:25
	→ shondm joined	14:25
shondm	:)	14:25
	→ harshpb joined	14:25
Sonja	if i'm developing a web app by myself, what is an example of why i would want to use branches in my project?	14:25
_ikke_	Sonja: Developing different features in paralel for example	14:26
	building a feature and in the meantime fix something	14:26
cmn	or experimentation	14:26
Wooga	or refactoring	14:26
Sonja	thanks for your wisdom :)	14:26
	← yoklov left	14:26
cmn	and also for clarity	14:26
	→ mastroWork joined	14:27
_ikke_	Sonja: You master branch for example, could be defined as always deployable	14:27
	→ consumerism joined	14:27
_ikke_	So any development happens on a different branch	14:27
Sonja	my current programming methodology is. add one tiny feature. check if it works. save. if not, then undo.	14:27
_ikke_	(topic branches as they are called)	14:27
cmn	if you always develop all features linearly, you'll have a harder time figuring out where particular commits come from	14:27
shondm	Anyone know how I could use Tower (or any UI) with a git repo on another computer over lan? The working copy needs to be on the remote (it's a web server), but I'm getting massive lag on NFS, SMB, SSH	14:27
	← shondm left	14:27
	→ shondm joined	14:27
	← shondm left	14:27
	→ shondm joined	14:28
Sonja	i'm always living on the edge of things breaking	14:28
cmn	shondm: the working directory is local, see !deploy	14:28
gitinfo	shondm: Git is not a deployment tool, but you can build one around it for simple environments. Here is an example hook to get you started: https://github.com/EugeneKay/scripts/blob/master/bash/git-deploy-hook.sh	14:28
k-man	when i do a git pull to my branch, it askes me for a commit message, as i have some local changes commited - is there a way to prevent it prompting me for the message and just use the default message?	14:28
cmn	Sanja: that's why VCS exists	14:28
	Sonja, I mean	14:28
shondm	I don't want to deploy, I want to use a git UI tool locally, and dev on the remote, push from the remote, etc	14:28
cmn	so you don't have to worry about breaking things	14:28
shondm	Using a UI through the network	14:28
_ikke_	k-man: No, it's designed this way	14:28
cmn	shondm: then mount	14:28
_ikke_	k-man: merges should be meaningful	14:29
Sonja	thanks for the guidance, everyone	14:29
cmn	and use that repo as though it were local	14:29
shondm	Tried that, everything is horribly slow	14:29
	→ thiago joined	14:29
cmn	there's no other way	14:29
	← glennpratt left	14:29
_ikke_	k-man: You might want to rebase instead	14:29
consumerism	i have a master branch and a testing branch. a lot of people have been making commits on testing for a while, some have bothered to merge master from time to time but most have not. master also contains cherry-picked commits from testing. i just merged master into testing with -s recusive -X theirs. (i wanted to simply make sure any code that is on testing is working with everything that's on master).	14:29
	now git status says i am 84 commits ahead of origin/testing. this seems wrong to me. any idea what i did wrong here? i expected 1 commit ahead: my merge i just did.	14:29
cmn	either you run the software on the other computer or you put a repo on yours	14:29
_ikke_	consumerism: !repro	14:29
gitinfo	consumerism: Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session, or at least explain exactly what you did that led up to the problem. This will help immensely with troubleshooting.	14:29
k-man	_ikke_, --no-edit?	14:29
	→ glennpratt joined	14:29
k-man	_ikke_, whats the advantage of rebasing instead?	14:30
_ikke_	k-man: cleaner history	14:30
	k-man: No unnecessary merges	14:30
k-man	_ikke_, sounds good	14:30
shondm	@cmn I even tried setting --git-dir locally and --work-tree as the remote, but it's slow. Still no other way?	14:30
k-man	so i just do a git-rebase rather than a git-pull?	14:30
consumerism	_ikke_: literally, while on testing with a clean tree, 'git merge master -s recursive -X theirs'	14:30
	← KWKdesign_ left	14:30
cmn	shondm: your problem is with the speed of our file access	14:30
	nothing git can do about that	14:31
consumerism	not sure what else you're looking for	14:31
cmn	s/our//	14:31
mikegerwitz	consumerism: Every commit that was in master but not in testing has been merged, in addition to a merge commit. So if you had 84 commits in master that were not in testing, then that is correct.	14:31
	83*	14:31
shondm	@cmn Is there a recommended lower limit? Something like 100kb/s?	14:31
	← kenperkins left	14:32
consumerism	mikegerwitz: ok, in that case i guess everything is ok. i wish my fellow coders would merge more frequently. :)	14:32
cmn	shondm: that's something that your FS does	14:32
	shondm: you might get some advantage by repacking	14:32
	but the main problem probably isn't speed, but latency	14:32
	any network FS is going to have horrible latency	14:32
shondm	Alright so the short answer is find out how fast git needs to be, for me, and then find a network protocol that can do at least that, plus very low latency	14:32
cmn	you can't change that	14:32
	there is no need for it to be any speed	14:33
	it's whatever you feel isn't too slow	14:33
Sonja	'git add --all' is the same as 'git add .' ?	14:33
	→ code joined	14:33
cbreak-work	shondm: latency is not an issue for git	14:34
shondm	Well, status through ssh on the remote takes ~1 second, from local through sshfs takes ~1 minute	14:34
	After a repack	14:34
cbreak-work	it works fine with extremely high latency protocols	14:34
	like hours or days	14:34
	if you use the appropriate transport	14:34
cmn	right, that's all just your FS	14:34
	git usually does small reads	14:34
shondm	Interesting	14:34
	← ShooterMG left	14:35
	← drev1 left	14:35
shondm	The remote is ext4, local is Mac OS Extended (journaled)	14:35
cmn	just clone the damn repo	14:35
	← oriba left	14:35
	← NetJunky left	14:35
cmn	it's going to be faster than trying to figure out how to do magic with physics	14:36
shondm	You suggest a post-commit push on the local and a post-receive push on the remote then?	14:36
cmn	no	14:36
	→ linusoleander joined	14:36
	← code left	14:36
cmn	I suggest you work on your computer and push out changes whenever you need to	14:37
shondm	Well there's 3 machines. My local dev (mac), remote web (ubuntu), and then a cloud (production)	14:37
	→ code joined	14:37
	← code left	14:37
	→ code joined	14:37
cmn	and?	14:37
shondm	I'd like to write code locally and use a git UI, where the git repo/work tree it on the remote web, and I push, using the UI, to the cloud	14:37
PerlJam	shondm: are you concerned that you can't test your changes locally?	14:38
shondm	Basically yes	14:38
cmn	then push to a temp branch	14:38
	and run the tests there	14:38
shondm	Specifically, PHPUnit takes a very, very long time and I'm not allowed to push to production without running it	14:38
cmn	and why is five seconds that takes you to push so important then?	14:38
shondm	I'm trying to find a way not to add steps to my workflow	14:38
cmn	then what you need is to figure out how to test locally	14:39
shondm	I run tests / push maybe ~20 times a day (urgent client requests, fixes, etc)	14:39
PerlJam	shondm: not quite the right optimization I would think.	14:39
	← LemonBE\|work left	14:39
PerlJam	shondm: you want to only add simple steps to your workflow	14:39
cmn	use a VM locally or whatever	14:39
PerlJam	shondm: and then hopefully you can figure out how to automate them :)	14:39
cbreak-work	work via x11	14:39
	→ LemonBE\|work joined	14:39
cmn	trying to use a very slow link for the FS is going to cost you more time	14:39
	cbreak-work: tower is OSX, remote is Linux	14:40
cbreak-work	who cares	14:40
shondm	I guess I'll change the short answer to "no-go"	14:40
cbreak-work	if he wants to work remote	14:40
	he can use a linux git client	14:40
	→ beatak joined	14:40
cmn	what he wants is to use tower, from what I see	14:40
cbreak-work	then he wants to work locally	14:40
cmn	right	14:40
cbreak-work	because that is a local program	14:40
	that works on local repositories	14:41
cmn	indeed	14:41
shondm	That's why mounting was the first idea	14:41
cbreak-work	you could try to somehow mount the remote filesystem	14:41
	but I'd doubt you'd be happy with the performance :)	14:41
cmn	except that's way too laggy	14:41
PerlJam	shondm: you could always run ubuntu in a VM on your mac for testing.	14:41
cmn	and the real problem is that testing takes long, so figure out how to test faster	14:41
shondm	It's not the ubuntu OS I'm looking for, it's the dedicated hardware	14:41
cmn	huh?	14:42
PerlJam	shondm: either way you want to separate development from deployment (even if it's just to a testing box)	14:42
cmn	you have your own computer, no?	14:42
	→ barrbrain_ joined	14:42
shondm	Believe me, I've optimized the tests (40% reduction), and I'm on a 2011 Pro with all the upgrades including RAID 0 SSD's, and still it drags	14:42
cbreak-work	git does not need dedicated hardware	14:42
	it works fine on the side	14:43
cmn	shondm: then make the tests on push	14:43
shondm	So back to adding steps. Dang.	14:43
	← linusoleander left	14:43
cmn	then go sit on that computer	14:43
	you're adding computers to the mix, so you need more steps	14:43
shondm	I'll do a commit hook locally that pushes to the test PC, a post-receive hook that runs tests, and a post-test-pass hook that pushes to production	14:44
cbreak-work	shondm: you should remove steps	14:44
	remove the step of communicating to the remote host uselessly	14:44
	and work locally	14:44
	it'll be much faster	14:44
shondm	Non-option	14:44
	← ThomasLocke left	14:44
cbreak-work	believe me:	14:44
	a mac is capable to edit text without trouble	14:44
PerlJam	shondm: why is that a non-option exactly?	14:44
shondm	You both missed the first parts of the conversation I think	14:45
	→ _huoxito joined	14:45
cmn	shondm: the step you should remove is the NFS mount	14:45
	that's adding a gazillion steps to every edit you make	14:45
shondm	Yes, I'm about to	14:45
cbreak-work	yeah, as I said: work locally	14:45
	FauxFaux giggles.	14:45
cbreak-work	all those useless steps just cost you time	14:45
sitaram	shondm: git config core.preloadindex true #... might help	14:45
bob2	you want a CI server to manage this	14:45
shondm	@sitaram Will try	14:45
	← barrbrain left	14:46
	barrbrain_ → barrbrain	14:46
	→ airborn joined	14:46
shondm	@bob2 have one, but it's on the cloud, twice as slow, and runs tests from 10 other devs	14:46
bob2	lolz	14:46
	also, not twitter	14:46
	← frogonwheels left	14:47
shondm	Impossibly request is impossible I guess	14:47
	→ drev1 joined	14:47
	→ linusoleander joined	14:48
cmn	yes, quite	14:48
	→ barrbrain_ joined	14:48
	→ frogonwheels joined	14:48
	← drev1 left	14:48
cmn	why is this computer somewhere on the Internet instead of the office?	14:49
	← barrbrain_ left	14:49
	← linusoleander left	14:49
	→ Per4u3e joined	14:50
cbreak-work	cmn: the cloud is the new paradigm	14:50
sitaram	and it costs a pair of dimes now too	14:51
cmn	and that's all fine for background jobs	14:51
	← barrbrain left	14:51
	→ klj613 joined	14:54
shondm	@cmn CI server is in the cloud because the team is all remote (9 countries, 10 devs)	14:55
	→ Tinti joined	14:55
	→ nj3ma joined	14:55
cmn	then there's no way anything other than pushing is going to work	14:56
shondm	Well, again I mis-explained myself	14:56
	→ xiangfu joined	14:57
	← jackh left	14:57
	← shruggar left	14:57
	← FauxFaux left	14:57
	→ dv310p3r joined	14:58
	← h0bbit left	14:59
	← Sigma left	14:59
PerlJam	shondm: un-mis-explain yourself :)	15:00
	→ Sigma joined	15:00
PerlJam	shondm: I still don't understand your constraints. They seem arbitrary to me. Why can't you develop and test locally? Why do you need the 3rd machine?	15:00
shondm	There's only 2 machines in my problem, and 4 total	15:01
cmn	it sounds like it's clear. You want to edit text on your computer and have another computer do the tests, because they apparently need high-end software to even run	15:01
shondm	I would like to write code and use Tower (git ui) locally, but have the working tree on a remote (which is in my house) because it's a web app, where I also run tests	15:01
PerlJam	shondm: yeah, all that is arbitrary.	15:01
	← xiangfu left	15:01
cmn	then where on earth are you?	15:01
	why aren't you at home?	15:02
shondm	I'm at home	15:02
	Me, my laptop, and my desktop are all here	15:02
cmn	but you said the test server is somewhere in the cloud	15:02
shondm	That's the CI server, can ignore	15:02
	core. preloadindex didn't help	15:02
SethRobertson	I think you need to step back and re-explain your problem.	15:03
sitaram	shondm: let me get this straight:	15:03
cmn	why is that repo in that other computer that isn't even running the tests?	15:03
	why is that location magic?	15:03
shondm	Sublime Text 2 to write the code, Tower to manage the repo, and it's all hosted on another machine in my home	15:03
sitaram	shondm: everything is on LAN, no WAN involved. You want to use a UI on one machine while keeping the work tree on another	15:03
shondm	^	15:03
sitaram	oh good lord; what the heck is sublime text -- poetry writing software?	15:04
cmn	hihi	15:04
sitaram	the names people come up with	15:04
shondm	Just a text editor I like	15:04
PerlJam	sitaram: it's a textmat-like editor	15:04
	→ Jackneill joined	15:04
	← Jackneill left	15:04
	→ Jackneill joined	15:04
PerlJam	er, textmate-like even	15:04
cmn	but why must the code be on this other computer?	15:04
sitaram	PerlJam: I know nothing about textmate either, honestly :) But have at least heard of it :)	15:04
cmn	if it's not running the tests, it just adds complexity	15:04
shondm	It's dedicated to this application, my macbook isn't	15:04
PerlJam	shondm: and why do you keep talking about "the repo" as if there's only one sacred repo?	15:04
cmn	dedicated how?	15:05
	→ glcrazy joined	15:05
	← glcrazy left	15:05
cmn	if it needs Linux, why not a VM?	15:05
shondm	It's running ubuntu server 11.10, and configured to be a web server, with just one web app, so minimal resources used for anything but this lone app	15:05
cmn	and why can't you run this on your mac?	15:05
	→ glcrazy joined	15:05
cmn	it can run web servers as well	15:06
PerlJam	yeah, I do stuff like that all the time.	15:06
shondm	Yes, that was my original incantation	15:06
PerlJam	but ... ?	15:06
cmn	and you decided to add this other computer why?	15:06
shondm	After the app got larger and larger, my macbook is no longer able to run the tests within 10 minutes	15:06
	10 minutes per push x 20 pushes per day = 200 minutes of tests	15:06
cmn	so this other computer /is/ running tests?	15:06
	← t4nkd`workin left	15:07
shondm	On the desktop it is only 30 minutes total	15:07
	→ sdaone joined	15:07
PerlJam	shondm: so you still develop locally (have a clone of the repo on your mac) and deploy to the testing machine whenever you want to run tests (it has it's own clone too)	15:07
shondm	I don't have a clone on my mac, I'd like to keep only one copy of the app on one computer (the desktop)	15:08
cmn	you keep contradicting yourself	15:08
	→ alexukf joined	15:08
PerlJam	shondm: why only one copy? that's the crazy part for me.	15:08
SethRobertson	And why do you care how many repos you have?	15:08
	→ FauxFaux joined	15:08
shondm	Reducing the total steps to production	15:08
cmn	but you're adding time	15:09
sitaram	shondm: why dont you have a clone on your mac?	15:09
SethRobertson	"Doctor doctor, it hurts when I do this!" "Don't do that."	15:09
	→ yoklov joined	15:09
PerlJam	shondm: so ... back to the beginning. Having another clone reduces the steps. NOT having another clone increases the number of steps	15:09
	← glennpratt left	15:09
	→ sporous joined	15:09
shondm	I basically want to dev, test, and sync as if my macbook is my desktop	15:10
PerlJam	shondm: not having another clone increases the steps because of all the hoops you have to jump through to maintain that arbitrary constraint.	15:10
	→ glennpratt joined	15:10
	→ preflex joined	15:10
cbreak-work	shondm: you can clone the repository	15:10
shondm	It was simple at first, when I was using the terminal for repo management	15:10
cbreak-work	then you can work locally as if your macbook is your desktop	15:10
	because it has the repository local	15:10
shondm	It's a web app, that's the catch. I have to refresh the page / do whatever between changes, then commit those changes	15:11
	I guess it's hard to understand when you're not actively doing it	15:11
PerlJam	shondm: what it sounds like is this: you're creating a whole bunch of other work because you're trying to avoid having to type "git push"	15:11
	← flijten left	15:12
shondm	That's not the case	15:12
cmn	just work on your desktop	15:12
shondm	There's additional factors	15:12
cmn	it reduces even more steps	15:12
PerlJam	shondm: is there some part of the app that you can't run on your mac?	15:12
shondm	Tests	15:12
PerlJam	tests aren't part of the app	15:12
cmn	but you don't need to run tests after every keystroke	15:12
shondm	No but I do before committing	15:12
	← Sitethief left	15:12
shondm	Well, pushing	15:12
cmn	no you don't	15:12
	but you don't	15:13
PerlJam	shondm: so stop doing that	15:13
shondm	I must test before I push	15:13
	→ xiangfu joined	15:13
cbreak-work	no	15:13
PerlJam	shondm: you said you have the ubuntu box for testing. Let the tests run there.	15:13
cmn	no you don't	15:13
cbreak-work	don't be stupid	15:13
shondm	I push directly to production with important fixes	15:13
SethRobertson	You must test before you share your work with others	15:13
cmn	you have to test before you deploy to production	15:13
cbreak-work	don't push to production!	15:13
cmn	shondm: then don't push directly to production	15:13
shondm	It's my job to do so	15:13
cbreak-work	push to the test server that does the testing	15:13
	then push from there to where ever you want	15:14
	after testing	15:14
	got it?	15:14
	simple.	15:14
	← sporous left	15:14
sitaram	he won't	15:14
cbreak-work	even a web dev should get that :)	15:14
cmn	huh? it's your job to do what?	15:14
sitaram	(get it I mean)	15:14
cbreak-work	cmn: probably "push directly to production"	15:14
	ah well...	15:14
	work end! :D	15:14
	← cbreak-work left	15:14
PerlJam	sitaram: don't be disparaging; some people need more help than others and that's okay :)	15:14
	← tvw left	15:15
	← xiangfu left	15:15
cmn	I'm pretty sure the job isn't "the push to production must happen directly to production without pushing to a third computer to run tets"	15:15
	that just doesn't make sense	15:15
	→ xiangfu joined	15:15
shondm	I don't see how this is a complex workflow; I dev on my laptop, the app runs on the desktop, where the git repo is, and through ssh I tell the desktop to test, then push to production	15:15
sitaram	PerlJam: I agree, but the amount of traffic, and the number of people who have said essentially the same thing for the past few minutes... shrug	15:15
PerlJam	shondm: it's not complex except that you make it so by requiring only one repo	15:16
cmn	shondm: it's not complex	15:16
	you're making it complicated	15:16
	instead of pushing directly to production, push to a computer that runs tets fast	15:16
shondm	That's the goal	15:16
cmn	then why aren't you doing it?	15:16
shondm	I feel like I need a diagram at this point	15:17
	→ Porpado joined	15:17
PerlJam	shondm: for us or for you? ;)	15:17
sitaram	PerlJam: join the club!	15:17
shondm	It's just the UI tripping me, so I'll just switch back to using the terminal	15:17
SethRobertson	Let us try again. You make changes on mac-repo. You commit. You push to test-repo. Test run. If all is good, you push to production-repo. Done!	15:17
shondm	Problem solved	15:17
PerlJam	shondm: what SethRobertson said	15:18
	→ sporous joined	15:18
	← sporous left	15:18
	→ sporous joined	15:18
shondm	I don't do anything with mac-repo because I don't have one, I only have test-repo	15:18
SethRobertson	GAAAAAAAAA!	15:18
	← Porpado left	15:18
PerlJam	shondm: and that's the problem.	15:18
sitaram	shondm: (laptop, local clone, edit, save, commit, push) -> (desktop, ubuntu, post-receive hook runs test suite and reports errors to STDERR)	15:18
	← dark_element left	15:18
PerlJam	shondm: clone the repo on your mac	15:18
shondm	Sillyness	15:19
	← FernandoBasso left	15:19
cmn	there is no reason why you can't have your repo on the mac	15:19
	← sporous left	15:19
cmn	it will solve your problem with latency	15:19
shondm	Between edits (before commit) I'd like to load the app and see what it's doing	15:19
	→ sporous joined	15:19
PerlJam	shondm: and you can do that on your mac	15:19
SethRobertson	Commit early and often. !best	15:19
gitinfo	[!best_practices] There is no one right answer for git best practices, but a consensus from #git is available at http://sethrobertson.github.com/GitBestPractices/	15:19
cmn	then do that	15:20
	← popcorn left	15:20
cmn	you can run a simple web server on your laptop	15:20
shondm	The desktop is loads faster	15:20
	350ms laptop, 120ms desktop on average	15:20
PerlJam	shondm: you aren't a recent svn user are you?	15:20
shondm	No, git was my first cvs	15:20
PerlJam	okay, just checking	15:21
	← Vert left	15:21
	→ phishy joined	15:21
cmn	but you're suffering more latency with the file system	15:21
PerlJam	shondm: wait ... how do you know the desktop loads it faster?	15:21
cmn	and you're suffering it more often	15:21
shondm	Because I switched from laptop to desktop this morning	15:21
	→ Fullmoon joined	15:22
PerlJam	shondm: okay, so you actually had a local copy on your laptop at one point	15:22
	← Fullmoon left	15:22
shondm	Still have one, yes	15:22
PerlJam	shondm: oh, just not a clone of the repo	15:22
	→ Dave^\| joined	15:22
sitaram	shondm: /ll before	15:22
	oops	15:22
	← sporous left	15:23
shondm	Right, not a clone, a copy-paste	15:23
sitaram	shondm: http://dpaste.org/qqQwX/	15:23
	→ sporous joined	15:23
shondm	The edit/save/commit part needs to happen on the desktop to maintain productivity	15:23
	→ emptyflask joined	15:23
	← thiago left	15:24
cmn	then why are you doing it on your laptop?	15:24
shondm	There are n edits before save and n saves before commit, between save and commit there is loading in a browser	15:24
	My desktop is a server	15:24
PerlJam	shondm: how often will you incur the extra 230 ms latency if you were to use your laptop?	15:24
shondm	About 45 minutes / day	15:24
	→ Fullmoon joined	15:24
shondm	The whole reason I'm switching is because it's more than twice as fast over an entire day to get fixes live	15:25
	← shondm left	15:25
	→ stkrzysiak joined	15:25
	→ rtjure joined	15:25
PerlJam	sigh	15:25
	→ linusoleander joined	15:25
PerlJam	oh well	15:25
SethRobertson	You just can't help some people	15:25
	→ shondm joined	15:25
shondm	And network drop-offs are annoying in the countryside	15:26
	← neurodrone left	15:26
	← dc5ala left	15:26
PerlJam	shondm: so, the real bottle neck was your edit/commit cycles, not running your tests?	15:26
	← sq-one left	15:27
	← stfn left	15:27
shondm	That's the next thing to be addressed, but unfortunately relies on more factors than software/network	15:27
	← envi_\|2 left	15:27
	→ kevwil joined	15:27
	PerlJam isn't sure if that's a "yes" or not	15:27
	→ envi_ joined	15:27
shondm	It's "yes, but not now"	15:28
	← amcsi_work left	15:28
shondm	Well, this has opened my eyes to certain limitations in the workflow I desire	15:28
	← kevwil left	15:29
shondm	Thanks everyone (PerlJam, cmn, sitaram)	15:29
	→ kenperkins joined	15:29
shondm	Have a swell day all of you	15:29
	← shondm left	15:29
	→ wilmoore joined	15:29
	→ kevwil joined	15:30
	← simenbrekken_ left	15:30
	← linusoleander left	15:30
superdmp	is it possible to checkout a file from a different branch (upstream)?	15:31
PerlJam	superdmp: sure.	15:31
	← amigojapan left	15:31
superdmp	git checkout remote_branch file	15:32
	← tsdh left	15:32
superdmp	that's correct, isn't it?	15:32
	← harshpb left	15:33
	← _sundar_ left	15:33
	→ snearch joined	15:34
	← xiangfu left	15:36
	→ shruggar joined	15:37
	→ linusoleander joined	15:37
FauxFaux	Wish git add -p could split a single hunk containing the replace of a line.	15:37
	← linusoleander left	15:37
	← sporous left	15:37
	← senny left	15:37
cmn	how would it get split?	15:38
	→ sporous joined	15:38
	← sporous left	15:38
	→ sporous joined	15:38
	→ flype joined	15:38
	→ OOPMan joined	15:39
	← preflex left	15:39
	→ FernandoBasso joined	15:39
	← TheHaven left	15:39
	→ iflista joined	15:40
	← jceb left	15:40
	→ t4nkd`workin joined	15:40
	→ localhost joined	15:40
	← anderse_ left	15:40
	← fukushima left	15:41
	→ preflex joined	15:41
	← papegaaij left	15:41
	← berserkr left	15:42
	← rtuin left	15:43
	→ enigmabomb joined	15:43
	t4nkd`workin → t4nkd`	15:44
	← lapistano left	15:45
	→ berserkr joined	15:45
	→ shwaiil joined	15:45
shwaiil	hi	15:45
gitinfo	shwaiil: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	15:45
	→ lapistano joined	15:45
	→ tvw joined	15:45
	← mastroWork left	15:46
	← shruggar left	15:46
	← Sonja left	15:46
	→ click joined	15:47
	→ hwrd\|work joined	15:48
	← stkrzysiak left	15:48
	← babilen left	15:48
	→ stkrzysiak joined	15:49
	← nj3ma left	15:49
	→ raatiniemi joined	15:50
	← Fullmoon left	15:51
	← t4nkd` left	15:51
	← Nibato left	15:51
	→ pl0sh joined	15:52
	← sgncfl left	15:52
	→ isao joined	15:52
	→ stfn joined	15:52
	→ newbie999 joined	15:53
	← ehsan left	15:53
sitaram	FauxFaux: git gui manages to do line-by-line adds	15:53
	→ tonebender joined	15:54
	→ ceez joined	15:54
	← squeakytoy left	15:54
	→ sgncfl joined	15:55
	← jkal left	15:55
	← angelsl left	15:55
	← newbie999\|3 left	15:55
	→ shruggar joined	15:55
	→ _iron joined	15:57
	→ arvind_khadri joined	15:58
	→ newbie999\|3 joined	15:59
	→ just0x41man joined	15:59
	← newbie999 left	15:59
	→ thelazydeveloper joined	16:00
	← mishok13 left	16:00
	← Per4u3e left	16:00
FauxFaux	cmn: I wanted to commit the delete, then the add. (i.e. remove feature one and add feature two, method ended up in the same place).	16:01
	→ Nibato joined	16:01
	→ drev1 joined	16:01
cmn	oh, I see	16:01
	editing that diff shouldn't be hard, just delete the + line	16:01
	→ umc joined	16:01
cmn	but it's certainly annoying	16:02
	dwmw2_gone → dwmw2_HIO	16:02
	← theDrumm_ left	16:02
	← whitman left	16:02
	→ ddelrio1986 joined	16:02
	→ neurodrone joined	16:04
	← ruskie left	16:06
spyro	sometimes I even diff the diffs	16:06
	and once it even came in handy	16:06
	← phishy left	16:06
	← shruggar left	16:06
	cslcm smells a "yo, dawg" coming up	16:06
	→ theDrumm joined	16:06
cslcm	oh, it's already been done http://cdn.memegenerator.net/instances/400x/11462942.jpg	16:07
	→ jameswarren joined	16:07
	← weebs_ left	16:07
	← snearch left	16:08
	← hwrd\|work left	16:08
	← giallu left	16:10
	← calimocho left	16:11
	→ maletor joined	16:13
	→ mikepack joined	16:13
	← drev1 left	16:14
	→ SegFaultAX\|work joined	16:15
	→ ruskie joined	16:15
	→ shruggar joined	16:16
	→ kadoban_ joined	16:16
	← shruggar left	16:17
	← fission6 left	16:18
	← rtjure left	16:19
	variable → trout	16:20
	→ michaelmartinez joined	16:20
	← maletor left	16:20
	← rrn left	16:20
	← francisfish left	16:20
	→ ehsan joined	16:21
	→ maletor joined	16:21
	→ Vert joined	16:21
	← cagedwisdom_ left	16:21
	← cagedwisdom left	16:21
	→ LekeFly joined	16:21
	→ fly9 joined	16:23
	→ williamcotton joined	16:24
	← juvenal left	16:24
	→ alexim joined	16:24
	→ dfr\|mac joined	16:25
	→ juvenal joined	16:26
	→ alvaro_o joined	16:26
	→ rath joined	16:26
	← alvaro_o left	16:27
	→ anderse_ joined	16:28
	→ h0bbit joined	16:30
	← LimeBlast left	16:30
	→ jceb joined	16:31
	← enigmabomb left	16:31
	← adhawkins left	16:32
	→ Err404NotFound joined	16:32
	→ intransition joined	16:32
	← afuentes left	16:32
	← intransit left	16:32
	← stkrzysiak left	16:33
	← chrischambers left	16:34
	← tatsuya_o left	16:34
	→ ShooterMG joined	16:34
	→ hwrd\|work joined	16:34
	→ tatsuya_o joined	16:34
	← noisebleed left	16:35
	← cslcm left	16:35
	← Desproges left	16:35
	→ enigmabomb joined	16:37
	← tyatpi left	16:38
	→ notVert joined	16:38
	← Vert left	16:38
	→ stkrzysiak joined	16:40
	← Chib left	16:41
	← sporous left	16:42
	→ Marchitos joined	16:42
	← enigmabomb left	16:42
	← irqq left	16:43
	← ronniy left	16:43
	← valtih left	16:45
	← kadoban_ left	16:45
	→ pfFredd joined	16:46
	← jason237 left	16:46
	→ Fullmoon joined	16:46
	→ Synthead joined	16:46
	← Fullmoon left	16:46
	→ diegoviola joined	16:46
	← tvw left	16:47
	← jave left	16:47
	← LekeFly left	16:49
	← LemonBE\|work left	16:49
	→ kerrick joined	16:49
	← Err404NotFound left	16:49
	← dfr\|mac left	16:49
	→ t4nkd`workin joined	16:52
	← Marchitos left	16:53
	→ John_Preston joined	16:53
	→ thiago joined	16:54
	→ MistuhKurtz joined	16:57
	→ tystr joined	16:57
	← achadwick left	16:58
	← code left	16:58
	← thiago left	16:58
	→ thiago joined	16:58
	← John_Preston left	17:00
	→ enigmabomb joined	17:00
	← kevwil left	17:00
	← preflex left	17:00
	→ sdeboey joined	17:00
	← michaelmartinez left	17:01
	→ smoge joined	17:01
	← rchavik left	17:01
	← veyres left	17:03
	→ calimocho joined	17:03
	→ khmarbaise joined	17:03
	← EgS left	17:04
	→ fission6 joined	17:04
	← jameswarren left	17:04
	→ dsantiago joined	17:05
	→ kukks joined	17:06
	→ jstout24 joined	17:06
	dwmw2_HIO → dwmw2_gone	17:07
	← Twey left	17:07
	← fisted_ left	17:08
	← notVert left	17:09
	← dr_lepper left	17:09
	→ QF-MichaelK joined	17:09
	→ darkmethod joined	17:10
	→ Twey joined	17:10
	→ shirish joined	17:11
	→ btree_ joined	17:12
	→ preflex joined	17:13
	← lonewulf` left	17:13
	→ dreiss joined	17:14
	← QF-MichaelK left	17:15
	→ alberto56 joined	17:15
	→ jsoa_ joined	17:15
	← BiggFREE left	17:16
	← jsoa_ left	17:16
	→ lonewulf` joined	17:16
	→ jsoa_ joined	17:16
	← ThunderChicken left	17:17
	← az left	17:17
	→ tonis_ joined	17:17
	→ ThunderChicken joined	17:17
	→ fisted joined	17:17
	→ rtuin joined	17:18
	← tonis_ left	17:19
	← jsoa left	17:19
	jsoa_ → jsoa	17:19
	→ theCole joined	17:19
	→ jason237 joined	17:20
	→ sureaint joined	17:21
	→ rrn joined	17:22
	→ IchGuckLive joined	17:23
	← shoerain left	17:23
	gitinfo set mode: +v	17:24
	→ codeshah joined	17:24
EugeneKay	!refund	17:25
gitinfo	If you are not satisfied with git for whatever reason, you are entitled to a full refund of the purchase price, and are invited to use another VCS. Elsewhere.	17:25
IchGuckLive	hi all i did a fetsch and checkout to someone elses branch it says im in the branch how i do now get the updatet files there are 5 updatet inside	17:25
SJS	But I paid in sweat and blood! That'll be a biohazard by now!	17:25
EugeneKay	Shush, it was so I could steal the factoid for another channel.	17:26
	→ drev1 joined	17:27
EugeneKay	IchGuckLive - what are youtrying to do, exactly? Merge somebody else's branch into yours?	17:27
	← cortana2 left	17:27
IchGuckLive	yes i guess i put a patch to the head and maild it to the devels and they put it in there personel branch for testing	17:28
	→ Ryback_ joined	17:29
	← Ryback_ left	17:29
	→ Ryback_ joined	17:29
IchGuckLive	so im now trying to get this 5 files back from there edit	17:29
	← calimocho left	17:29
	→ ISF joined	17:29
EugeneKay	IchGuckLive - OK. 'checkout' the branch you want the merge to be on, and then 'git merge PersonalBranch' to bring it to the currently-checked-out-branch.	17:29
	Have you read the !book yet, by chance?	17:30
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://progit.org/book/ but also look at !bottomup !cs !designers !gitt !vcbe and !parable	17:30
IchGuckLive	i will idid this on some like fetsch	17:30
	← jason237 left	17:30
lonewulf`	testing.	17:31
	→ jcrotty joined	17:31
IchGuckLive	i think i made it as git pull gives uptodate	17:31
	i will compile and see the results	17:31
EugeneKay	lonewulf` - !hi	17:32
gitinfo	lonewulf`: [!welcome] Welcome to #git, a place full of helpful gits. If you have a question, feel free to just go ahead and ask—somebody should answer shortly. For more info on this channel, see http://jk.gs/git/ Take backups (type !backup to learn how) before taking advice.	17:32
	← pppaul left	17:32
lonewulf`	EugeneKay: hi, and tnx.	17:32
	gitinfo set mode: +v	17:32
	← marcogmonteiro left	17:33
jcrotty	how to see which commit sha1 the super project points to for a particular submodule?	17:33
	→ pppaul joined	17:33
	← IchGuckLive left	17:33
	→ iwoj joined	17:34
	→ calimocho joined	17:35
MistuhKurtz	hm	17:37
cmn	status submodule should give you that	17:37
	submodule status, rather	17:37
MistuhKurtz	If I have an application that's built, and I want to re-build it mostly from scratch, on a totally new/different frameworks & everything, should I build it as a branch of my current repo?	17:37
	← just0x41man left	17:37
MistuhKurtz	or make a new repo?	17:38
	→ az joined	17:38
	← flype left	17:38
	← Goplat left	17:38
	→ ronniy joined	17:39
raek	or something in between perhaps? git checkout --orphan	17:39
	→ diago joined	17:39
	→ lacrymology joined	17:39
diago	if I did a git reset <rev> --hard, can I get those changes back?	17:40
raek	you could then merge over any shared files	17:40
EugeneKay	MistuhKurtz - I would say that it depends upon how you're gonna present it, version wise. New Foo, or Foo 2.0?	17:40
raek	diago: committed or uncommitted?	17:40
diago	I have not commit	17:40
	I just messed up	17:40
	← FernandoBasso left	17:40
EugeneKay	MistuhKurtz - and how long will you be keeping 1.0 around / supporting it?	17:41
MistuhKurtz	the first version is sort of a proof of concept kind of	17:41
	it's really roughly done.	17:41
	→ kevwil joined	17:41
	← tatsuya_o left	17:42
EugeneKay	If it's a TOTAL rewrite I would just init a new repo and name it -new, -future, or something like that.	17:42
	→ blacktulip joined	17:43
cbreak	diago: you can only get committed changes back, sorry	17:43
	→ jave joined	17:43
diago	oh sorry, they were commited locally	17:43
cbreak	(if you staged but not committed them, you might be able to get them back if you're prepared to invest a few hours)	17:43
	then type git reflog and find the commit just before you reset	17:44
	and then reset --hard to that commit	17:44
diago	ok cool. Thanks	17:44
	← consumerism left	17:44
	← jceb left	17:45
superdmp	i'm on a new branch called pageflags-testing; it's only a local branch, but now I want to put my work into my pageflags branch locally, and then push it to the pageflags branch at github	17:46
	what's the best way to proceed?	17:46
cbreak	check out the branch, commit	17:46
	then git push remotename branchname	17:47
	→ drev1_ joined	17:47
	dwmw2_gone → dwmw2_HIO	17:47
	→ jceb joined	17:47
superdmp	cbreak: checkout which branch?	17:47
cbreak	the one you want the changes to be in	17:47
	← drev1 left	17:47
	drev1_ → drev1	17:47
cbreak	just make one branch look like you want it, by merging, committing, rebasing, what ever	17:47
	(maybe merge the testing branch, or rename it?)	17:47
	and then push	17:47
	→ jameswarren joined	17:47
	← jcrotty left	17:47
	→ rtjure joined	17:48
superdmp	cbreak: so why will that bring my changes from pageflags-testing with it, rather than replacing them with what was in pageflags before?	17:48
	→ eletuchy joined	17:48
cbreak	because you do that.	17:48
	you have a branch. you can change it how ever you like.	17:49
	do you like merges? Then git merge	17:49
	→ Spedge joined	17:49
cbreak	Or do you like linear history? Then you could cherry-pick all the stuff. Or do something with rebases.	17:49
	or did you not even commit your changes? Then commit them in that branch.	17:49
	← jceb left	17:49
EugeneKay	superdmp - !sausage	17:49
gitinfo	superdmp: [!sausage_making] Some developers like to hide the sausage making (pretend to the outside world that their commits sprung full-formed in utter perfection into their git repository). `git rebase -i`, `git add -p`, and `git reset -p` can fix commits up in post-production by splitting different concepts, merging fixes to older commits, etc. See also http://sethrobertson.github.com/GitBestPractices/#sausage	17:49
	→ n1x joined	17:49
	→ jceb joined	17:50
superdmp	cbreak: ok, so now I am back in pageflags, so I can to merge pageflags-testing into it	17:50
	i have no need to hide my sausages	17:50
	← jceb left	17:50
	→ psoo joined	17:50
	→ jceb joined	17:50
cbreak	then git merge the other branch	17:50
	← isao left	17:50
	← enigmabomb left	17:50
	→ notjohn joined	17:51
superdmp	thanks	17:52
	← alexukf left	17:53
	← pfFredd left	17:55
	→ pfFredd joined	17:55
	← pfFredd left	17:56
	→ EgS joined	17:56
	← thiago left	17:56
	raek plans to suggest people to paste the output (possibly censored) of "git log --all --graph --decorate --oneline --abbrev-commit" to aid communication	17:58
	→ eucalyptus joined	17:59
	← d3c left	18:00
	← smoge left	18:00
	→ osmosis joined	18:01
	← iwoj left	18:01
	→ adhawkins joined	18:01
	← Vile left	18:02
	← Milossh_ left	18:04
	→ drev1_ joined	18:04
	→ srcerer joined	18:04
	→ JohnyBaby joined	18:04
	← drev1 left	18:05
	drev1_ → drev1	18:05
	← theCole left	18:07
	→ robbyoconnor joined	18:08
	→ Spedge- joined	18:09
	→ cakehero joined	18:09
	← Spedge left	18:09
	→ shruggar joined	18:09
	→ iwoj joined	18:10
	← arietis left	18:10
	← iwoj left	18:11
	→ iwoj joined	18:12
	→ tsdh joined	18:12
	→ metalfan_ joined	18:13
	→ j416 joined	18:13
shwaiil	Q: Not sure where to ask this, but when we push, is it possible to run minifications of JS and CSS files ? How to people do that stuff ? I'm using Ubuntu at home and Mac Osx at work. I guess there's command line programs that do the minifcation and maybe we have a script we run instead of git push ?! Tks for looking	18:14
	← pppaul left	18:14
tango_	shwaiil: you can do that when DEPLOYING, not when pushing	18:14
	!deploy	18:15
gitinfo	Git is not a deployment tool, but you can build one around it for simple environments. Here is an example hook to get you started: https://github.com/EugeneKay/scripts/blob/master/bash/git-deploy-hook.sh	18:15
tystr	I'm using this to create a repository from a subdirectory within another repository…. git filter-branch --subdirectory-filter ./src/path/to/directory -- --all	18:15
	but I don't end up with all the contents of that folder….	18:15
	← _huoxito left	18:16
shwaiil	tango_: oh right, tks for looking. I'll read that tks. Tango, but by the way, we could use hooks to run other programs or something ?!	18:16
tango_	yes	18:16
	→ dr_lepper joined	18:17
	← dr_lepper left	18:17
	→ dr_lepper joined	18:17
cmn	hooks are executable files	18:17
	you can do whatever you want, just remember that git isn't a deployment tool and that the hooks block the push ending	18:17
	← Spedge- left	18:17
	→ HyperNerdV2 joined	18:17
shwaiil	cmn: I see. I think for now, I'll do something basic. Because I usually do push from master on production server, not sure if its a good practice, but it works fine. So I could, let's say, run a script to do this and after he finishes minify the original files or something	18:18
	← bolt left	18:19
cmn	the problem is that the minification could take a long time, and it's really the deploy system's jbo	18:19
	→ bolt joined	18:19
	← diegoviola left	18:19
cmn	you should probably make the post-receive hook signal or enqueue something that tells the deployment system to run	18:19
	→ Spedge joined	18:20
tystr	nvm im tripping	18:20
	→ BartlomiejB joined	18:21
	→ linguini joined	18:21
	→ cespare joined	18:21
	← khmarbaise left	18:21
	→ ben225 joined	18:21
	→ thiago joined	18:22
	→ sporous joined	18:23
	→ kadoban_ joined	18:23
	dwmw2_HIO → dwmw2_gone	18:23
shwaiil	cmn: ok thanks for your sugestions, I'll try to learn a deployment tools. I got that one from tango_ but I dont think I can use that now, don't have enough knowledge	18:23
	← rtjure left	18:23
	→ tewecske joined	18:23
	← tewecske left	18:23
	← blacktulip left	18:24
	→ flijten joined	18:24
	→ blacktulip joined	18:25
cmn	it's quite simple reall, it just needs to make sure there are not odd edge cases; the basic idea is to use git-archive to generate a tarball of the files we want and then use rsync to send them to the right place	18:25
	→ pppaul joined	18:25
cbreak	you don't even need rsync	18:26
	you can just untar them to a directory and then atomically swap it	18:26
	← pmatthews left	18:26
cbreak	well, "atomically"	18:26
thiago	atomic objects are neither active nor radioactive	18:26
shwaiil	cmn: can you give a example of its usage ? Like the commands you'd use on production server ?	18:27
	I mean..if you have the time	18:27
cbreak	atomically means undividable. Ironically, one of the most potent energy sources comes from dividing atoms...	18:27
shwaiil	anyway, i'm searching google, many of the articles they're basicly doing what i'm doing right now.. git pull origin master :P	18:28
cmn	it's mostly just git archive \| tar xf - -C /var/www/some.dir.$RANDOM; ln -s /var/www/mysite.com /var/www/some.dir.$RANDOM	18:28
	where $RANDOM is the same in both uses	18:28
	that's what cbreak is talking about	18:28
	← kerrick left	18:28
cbreak	shwaiil: there are a lot of morons on the internet.	18:29
	morons that do stuff like merging in a script	18:29
	→ cvprog joined	18:29
cmn	with rsync, you'd unpack to /tmp/something.$RANDOM and then rsync -r --delete /tmp/something.$RANDOM/ /var/www/mysite.com/	18:29
cvprog	one question, i just pulled from origin master how do i merge does changes into my branch while keeping my changes ?	18:29
cbreak	if you pull, that means you merged	18:30
	pull is fetch + merge	18:30
	← Spedge left	18:30
	→ Spedge joined	18:30
	→ jcrotty joined	18:30
shwaiil	cmn: to start I think I should use rsync	18:30
	this doesn't look that easy, I'll plan some reading on this	18:31
cmn	that works mostly for static sites; if you need DB config or somesuch, it gets a bit more complicated	18:31
	this is why places like heroku exist; they do the annoying deploying steps for you	18:32
shwaiil	cmn: oh nice :)	18:32
	right now, for DB and stuff I do it manually, but configuration params are set based on server name	18:32
	cmn: to be clear to me. the commands you showed to me, is what https://github.com/EugeneKay/scripts/blob/master/bash/git-deploy-hook.sh does for me, or is this the commands I should use with it ? I didn't read git-deploy-hook.sh yet, I only got it a few moments ago	18:33
cmn	it's the basic steps	18:34
	that script does more	18:34
	and lets you configure what to give to rsync	18:34
	and makes sure you don't accidentally try to deploy a deleted brach, that kind of thing	18:34
shwaiil	I guess I'm starting to understand	18:34
	Thank you cmn	18:34
	← raatiniemi left	18:35
	→ duckxx joined	18:35
	← sdaone left	18:35
	→ tyatpi joined	18:35
	→ eephillip joined	18:36
duckxx	im trying to do a pull.. but getting "please commit your changes before you can merge... but i dont want to commit my changes, cuz its my config file for my local environment.... screenshot: http://img513.imageshack.us/img513/1361/croppercapture166.jpg	18:36
	← Spedge left	18:37
cmn	!config	18:37
gitinfo	[!configfiles] the recommended way is to change your application so that you can (perhaps optionally) store site-specific configuration/data in separate files, then keep those files out of version control. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.	18:37
cmn	you can stash those changes	18:37
	but you're going to run into this issue a lot	18:38
	← eephillip left	18:38
	→ whitman joined	18:39
	→ eephillip joined	18:39
	→ mcluskydodallas joined	18:39
	← eno_ left	18:39
	gitinfo set mode: +v	18:39
	→ eno joined	18:40
	← eno left	18:40
	→ eno joined	18:40
duckxx	omg.. how do i just do a full reset to the last sucessful pull i had	18:40
	→ MattLo_SWX joined	18:40
MattLo_SWX	hey	18:40
	→ Eridius joined	18:40
	→ Spedge joined	18:40
cmn	reset --hard will get you back to whatever commit you wish	18:40
MattLo_SWX	question to anyone: I'm pulling from Github and its showing a different Author name, I changed all my configs but its still showing a different author name when I pull down	18:40
cmn	but be aware that it will destroy all changes	18:40
MattLo_SWX	Has anyone had this issue before?	18:40
	→ tatsuya_o joined	18:41
cmn	MattLo_SWX: the author information is in the commit	18:41
rath	duckxx: we use .gitignore in that case	18:41
	← thiago left	18:41
MattLo_SWX	its possible to change that right	18:41
	?	18:41
cmn	rath: the file is tracked, otherwise git wouldn't have complained	18:41
	MattLo_SWX: only by !rewriting the history	18:41
gitinfo	MattLo_SWX: [!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	18:41
MattLo_SWX	gotcha	18:42
	yeah I dont want to do anything problematic, the pull merge just happened to be configured to the wrong account, I just realized thats what is in the logs.	18:42
	← marthinal left	18:42
rath	cmn: i mean that "settings.php" .. there is an example in the repo ..	18:42
duckxx	git reset hard ?	18:43
	is that how u reset it fully to original ?	18:43
rath	in our case at work it's enough to track that sample, not a specific version	18:43
cmn	but ignore rules won't affect that file	18:43
	← jcrotty left	18:44
	← drev1 left	18:44
	→ drev1_ joined	18:44
	← j416 left	18:44
	← cafaro left	18:45
rath	at work we don't track such a file...that's what i want to say	18:45
	→ irqq joined	18:45
	← iamntz left	18:45
	→ cafaro joined	18:45
	← cafaro left	18:45
	→ cafaro joined	18:45
	→ mishok13 joined	18:45
cmn	defaults or explanations are a good thing to track, but not the dev environment	18:45
rath	+1	18:46
	← JohnyBaby left	18:46
	→ giallu joined	18:46
	← cvprog left	18:47
cbreak	I've heard that in other companies, they'd track the whole toolchain including compilers and all libraries	18:49
	not with git though :)	18:49
	→ JohnyBaby joined	18:49
SerajewelKS	cbreak: whenever i have to track a compiled, third-party blob i always die a bit inside... but what else can you do	18:50
raek	I can see the value of doing that (in a repo distinct from the source repo)	18:50
	← lapistano left	18:51
	← tabakhase_ left	18:51
	→ tabakhase_ joined	18:51
	tabakhase_ → tabakhase	18:52
	→ thiago joined	18:52
	→ lapistano joined	18:52
	← thiago left	18:52
cmn	sometimes you have to because all they give you is some binary blob	18:52
	→ thiago joined	18:52
cmn	but IIRC android does track some build results, because they take hours to build	18:52
	← MattLo_SWX left	18:53
	← Spedge left	18:53
	← rtuin left	18:53
	← tsdh left	18:54
	→ DofLeFou joined	18:55
	← shwaiil left	18:55
	← DofLeFou left	18:56
	→ Shusshu joined	18:56
	→ DofLeFou joined	18:56
	← pppaul left	18:56
	→ rtuin joined	18:56
	→ pppaul joined	18:57
	← anoob left	18:57
	← ISF left	18:57
	← ben225 left	18:58
	← beatak left	18:59
	→ diegoviola joined	19:00
	← stodan_ left	19:00
	→ arietis joined	19:00
	→ beatak joined	19:01
	← pppaul left	19:01
	← bnjmn left	19:02
	→ beatak_m joined	19:02
	→ pppaul joined	19:02
	← rath left	19:04
	→ tvw joined	19:04
	← drev1_ left	19:05
	← beatak left	19:05
EugeneKay	A dual hex-core buildbox with 8GB of RAM and a 128GB SSD will put Android together in about 15-20 minutes	19:06
	My desktop with a 8GB ramdisk for out/ will do it in about 30	19:06
FauxFaux	Distcc.	19:07
	→ pandeiro joined	19:08
infogulch	how does git know the change was in "class DataBase"? @@ -143,6 +143,10 @@ class DataBase	19:08
	indentation?	19:08
FauxFaux	infogulch: ctags	19:10
infogulch	ah	19:11
	→ glphvgacs joined	19:11
	← n1x left	19:11
infogulch	but it's an obscure little scripting language o_O	19:11
glphvgacs	so there is a patch i wanna apply to a git repo and then i need to "rebase" apparently. what's rebasing?	19:11
	→ intransit joined	19:12
cmn	man git rebase has diagrams	19:12
gitinfo	the git-rebase manpage is available at http://jk.gs/git-rebase.html	19:12
FauxFaux	glphvgacs: man git rebase	19:12
gitinfo	glphvgacs: the git-rebase manpage is available at http://jk.gs/git-rebase.html	19:12
FauxFaux	infogulch: It generally assumes it's like C if it looks like C, i.e. the first thing before the top-level { is a method name	19:13
	← t4nkd`workin left	19:13
infogulch	oh ok. cool	19:13
	← intransition left	19:13
cmn	there are regexes for a few popular languages	19:13
	← neurodrone left	19:14
	→ neurodrone joined	19:15
	← glphvgacs left	19:15
ohcibi	i have extracted almost any content of a file A to a file B. after git add i have renamed A->B plus modified A oO. how to force git to put it as 'new file B' and 'modfied A'?	19:16
cmn	ohcibi: that's not how git works	19:17
	git doesn't store that information, it's calculating it at runtime	19:17
ohcibi	hm..	19:17
	← CaptainKnots left	19:17
ohcibi	it clashs with svn dcommit	19:17
cmn	there's -M and friends to limit the similarity at which git will consider something renamed	19:17
	→ t4nkd`workin joined	19:17
cmn	well, svn is a different beast	19:17
SethRobertson	You can always create the file B in a different commit if git-svn is causing problems.	19:17
	→ Radium joined	19:18
	← jvasak left	19:18
ohcibi	hm, not nice but possible.... well its svn, so who asks for nice	19:18
	→ codeM0nK3Y joined	19:19
ohcibi	cmn: which command has -M?	19:19
cmn	the log family	19:19
	→ tizzo joined	19:19
	← Lgb left	19:20
cmn	though it's probably handled by the diff code	19:20
	← osX64 left	19:20
	← ceez left	19:20
	← cakehero left	19:21
tizzo	anyone have any idea why git-filter-branch would refuse to remove submodules?	19:21
	→ bibi23 joined	19:21
	→ ben225 joined	19:22
	gitinfo set mode: +v	19:22
bibi23	hi, what is the proper to move a bare repository from a server to another? thanks	19:22
	→ jason237 joined	19:22
cbreak	tizzo: it's not filter-branch that refuses it. It's probably your script	19:22
	bibi23: you can just clone them over	19:22
	or rsync I guess	19:22
cmn	bibi23: either clone --mirror; or rsync	19:22
	→ Lgb joined	19:23
	← envi_ left	19:23
	← pppaul left	19:23
cmn	or copying it to an external medium and copying it over	19:23
	or anything that will transfer files from one place to another	19:23
	← Sigma left	19:23
milki	aka, nothing special	19:23
bibi23	ah ok great, so basically just copying it, then it will be ok there is no reference to the current server in git internal file for example right?	19:23
milki	right	19:24
	→ pppaul joined	19:24
bibi23	ok thanks guys ;)	19:24
milki	only existing repos pointing to it as a remote	19:24
tizzo	cbreak: git filter-branch -f --tree-filter 'rm -rf folder_containing_submods' HE AD	19:24
	cbreak: that kills everything but the submodule folders	19:24
	← bibi23 left	19:25
cbreak	tizzo: that's not the way to remove submodules	19:25
	you should use git rm --cached on them	19:25
tizzo	cbreak: hrm, tried that too but nuked the .gitmodules folder first	19:25
	→ syphar joined	19:25
tizzo	so maybe I need to start over and go in a different order?	19:26
cbreak	.gitmodules is a file not a folder	19:26
SerajewelKS	the directory where the submodules get cloned into shouldn't even be tracked in the parent repo	19:26
cbreak	you can delete it if you want to remove all submodules I guess	19:26
SerajewelKS	the only thing that gets tracked when you add a submodule is .gitmodules	19:26
	the submodule working directory should be ignored	19:26
cbreak	no	19:26
	the main thing is the commit entry in the tree	19:27
	.gitmodules is just some meta data to get the remotes up and running	19:27
tizzo	cbreak: yeah, I know, misspoke	19:27
SerajewelKS	cbreak: .gitmodules contains the remote repo url as well as the commit to check out	19:27
cbreak	no	19:27
	the tree contains the commit to check out	19:27
	→ drev1 joined	19:27
	→ LekeFly joined	19:28
tizzo	right, the tree contains a blob where the submodule goes (I think with a special permission) that contains the hash	19:28
SerajewelKS	ah, right. forgot that detail.	19:28
	tizzo: not a blob, if i'm reading cbreak right	19:28
tizzo	thought it was a blob..	19:28
	but that entry isn't getting deleted by a filter-branch	19:29
cbreak	git cat-file -p HEAD:path/to/dir	19:29
tizzo	with git rm -r --cached	19:29
SerajewelKS	tizzo: the object referenced in the tree will be a commit object in the submodule repo. right, cbreak/	19:29
cbreak	where dir is the folder that contains the submodule folder	19:29
tizzo	ah	19:29
	dig it, I was wrong	19:29
cbreak	so HEAD: if the submodules are in root	19:29
	that'll show you how they are stored	19:29
	tizzo: git rm --cached will delete it	19:29
	tizzo: filter-branch could not care less	19:29
	you should see entries like 160000 commit a8eb8b16c34bc6b240e471c8efc7f7d14b0a6563 tinyxml	19:30
tizzo	cbreakso are you saying this should work? - `git filter-branch -f --tree-filter 'git rm --cached folderContainingSubmods' HEAD`	19:31
	→ preflex_ joined	19:31
cbreak	I'd use --index-filter	19:32
	→ moltar_net1 joined	19:32
	← preflex left	19:32
cbreak	and it depends on what you mean with your path	19:32
	is it a folder that contains submodules?	19:32
	or is it the actual submodule directory?	19:32
	preflex_ → preflex	19:32
SerajewelKS	tizzo: --tree-filter requires the tree to be checked out before you can mess with it... --index-filter will let you tweak the index tree directly	19:32
tizzo	cbreak: contains submods	19:33
	everything else got blown away	19:33
	→ osX64 joined	19:33
cbreak	hmm?	19:33
	throw in an -r for recursive then	19:33
SerajewelKS	tizzo: therefore --tree-filter will take longer, since it has to check out the tree, while --index-filter does not.	19:33
tizzo	I keep getting `did not match any files`	19:33
	it's there, if I clone it I have them	19:34
cbreak	your path might be wrong	19:34
cmn	have you checked every single revision and checked?	19:34
	→ teweWORK joined	19:34
cbreak	but you can also throw in an --ignore-unmatched to not get that	19:34
tizzo	ah	19:35
	that'll help	19:35
	it might not be in the early commits	19:35
	← anderse_ left	19:36
tizzo	ah ha!	19:37
	cmn cbreak SerajewelKS: you are my saviors	19:37
	that did it	19:37
	you guys rock, thanks	19:37
	→ cakehero joined	19:37
	→ vervic joined	19:37
SerajewelKS	glad my half-truth was helpful :)	19:37
tizzo	so is there any automated way to squash commits that don't have changes after a filter-branch?	19:38
SerajewelKS	tizzo: --prune-empty	19:39
	tizzo: (add that to the filter-branch invocation)	19:39
tizzo	SerajewelKS: AWESOME! thanks :)	19:39
SerajewelKS	\o/	19:39
	← thiago left	19:40
	→ sq-one joined	19:40
	← JohnyBaby left	19:40
	→ GreekFreak joined	19:40
	← codeM0nK3Y left	19:42
GreekFreak	Hey all. I would like to back up my code, with the git repos, but not on github. Do I simply copy my code's folder, paste it and it's fine, or just the .git folder?	19:42
	→ codeM0nK3Y joined	19:42
SerajewelKS	GreekFreak: either will work	19:42
	GreekFreak: if you copy just the .git folder you technically won't have a "proper" repo unless you change the repo config to indicate that it is bare	19:43
tizzo	GreekFreak: if you have ssh access to a server it's easy to host your own repos too	19:43
SerajewelKS	GreekFreak: but if your goal is just to have your crap in case the worst happens, a plain copy is more than sufficient	19:43
	→ thearthur joined	19:43
	← sdeboey left	19:43
SerajewelKS	GreekFreak: and what tizzo said	19:43
	GreekFreak: taking just the .git will take less space and will still contain the entire repo (obviously, it won't back up changes to your working directory though, if that is important to you)	19:44
GreekFreak	SerajewelKS: thanks. All I want is to basically backup for the worst. So I'll just do a straight copy	19:45
	tizzo: I don't have one yet, but soon	19:45
	Would I be able to have that whole folder in a flash drive (as an example) load it on another machine and carry on working, and back to the original? Like basically carrying the repo with me?	19:46
	← codeM0nK3Y left	19:46
	← yoklov left	19:46
cmn	yes	19:46
	you can copy and carry it however you feel	19:47
GreekFreak	cmn: that's pretty nifty. Allows you to switch machines on the fly	19:47
cmn	but git isn't for backups, it's for developing	19:47
GreekFreak	thanks guys	19:47
	cmn: I am devloping	19:47
	and I do want it for the version cntrol and branching	19:47
	← pppaul left	19:47
	gitinfo set mode: +v	19:47
moltar_net1	Hi! Several commits ago, I deleted several files, that I though were no longer needed. Now I realized it was a bad idea, and need to bring them back. What is the appropriate way of brining them back from dead? Thanks!	19:48
	→ pppaul joined	19:48
cmn	git checkout <commit> -- <files>	19:48
SerajewelKS	GreekFreak: basically, you would just clone the repo on the flash drive	19:48
cmn	where <commit> is the commit where they exist	19:48
moltar_net1	cmn: ah saaweet. always so easy!	19:48
SerajewelKS	GreekFreak: what OS are you using?	19:48
GreekFreak	cmn : I just want a quicker way to move the code around without uploads, and as a backup	19:48
	SerajewelKS: Ubuntu. Why would I clone and not just copy/paste?	19:49
moltar_net1	GreekFreak: I've heard people using Dropbox to sync Git repos. Or does that count as "upload"?	19:49
	← ronniy left	19:49
cmn	by that you mean copy?	19:49
	moltar_net1: that counts as a very bad idea	19:49
EugeneKay	!dropbox	19:49
SerajewelKS	GreekFreak: because what you will have is not a proper repo, if you only copy the .git	19:49
gitinfo	Storing a bare repository on DropBox is a Very Bad Idea (tm). Everybody who has tried it has reported repo corruption, slowness in synchronizing, and general pain. Use a free code-hosting service such as Bitbucket or GitHub instead.	19:49
moltar_net1	got it! I'd never do this myself under these same assumptions :)	19:49
	I run gitolite on rackspace	19:50
SerajewelKS	GreekFreak: if you copy the .git to /media/flashdrive/repo-backup then you can "clone /media/flashdrive/repo-backup myproject" later	19:50
	→ twittard joined	19:50
GreekFreak	moltar_net1: I've read that that causes sync errors	19:50
SerajewelKS	GreekFreak: you might want to, instead, create a directory on the flash drive, "git init --bare" inside it, then add it as a remote to your dev repo and push	19:50
	GreekFreak: this is The Right Way(tm) to do this	19:50
	← twittard left	19:51
SerajewelKS	GreekFreak: the downside is that if you forget to push a local branch, it won't exist in the backup	19:51
GreekFreak	SerajewelKS: agreed, it's the right way, but why not just copy paste every night. The Right Way sounds like a hell of a lot more work for the same thing	19:51
SerajewelKS	GreekFreak: actually, once you get it set up it's much easier	19:52
GreekFreak	and this way I can carry my code around (so to speak)	19:52
fission6	if i made changes on master, then checked out a branch, i do git rebase master, to rebase the branch with my master changes, correct?	19:52
	→ limoges joined	19:52
SerajewelKS	GreekFreak: $ cd /media/flashdrive && mkdir myproject.git && cd myproject.git && git init --bare	19:52
GreekFreak	SerajewelKS: oh ok. I'm assuming it will be the same concept, but instead of to Github its to your "folder"	19:52
SerajewelKS	GreekFreak: go to your repo and: $ git remote add flashdrive /media/flashdrive && git push flashdrive --all	19:53
	GreekFreak: and you are done	19:53
	GreekFreak: then just "git push flashdrive --all" whenever you want to back up	19:53
yrlnry	This morning while my tests were running on branch A, I wanted to make some changes on branch B. I cloned the local repo, committed the changes in the clone, and pushed them back to the local repo before the tests finished running.	19:53
	→ codeM0nK3Y joined	19:53
yrlnry	Is that peculiar or surprising?	19:54
	→ JohnyBaby joined	19:54
	← emptyflask left	19:54
EugeneKay	Sounds like a valid thingto do, the usual caveats about pushing to non-bare still apply.	19:54
	→ borior joined	19:54
SerajewelKS	GreekFreak: if you have the flash drive and you want to edit code, you will have to clone the repo first though, that's one downside. however, this is still simpler since a non-bare repo introduces pushing issues (you can't push a branch to a remote when that branch is checked out at the remote. bare repos can't check out since they don't have a working directory, so this sidesteps that issue.)	19:54
GreekFreak	SerajewelKS: thank you. that is simpler. And to work on another PC would i have to clone then? As opposed to just opening a file and working and committing, (like I would've done with the cope/paste way)	19:54
SerajewelKS	GreekFreak: yes, exactly	19:54
borior	hi all. is there any way I can suppress particular files/patterns from the output of git diff?	19:55
SerajewelKS	GreekFreak: it sounds like more work but it's actually a bit better, since now you have three copies of the repo :) two computers and one flash drive	19:55
	GreekFreak: and that way if you commit on one box and forget to push, you can do the proper merge or rebase later	19:55
	→ dvide joined	19:55
SerajewelKS	GreekFreak: you are treating the flash drive as a central github-like repo, basically	19:55
GreekFreak	SerajewelKS: thank you. that makes sense. I'll sort something out like that	19:56
	← mcluskydodallas left	19:56
SerajewelKS	GreekFreak: the key to this whole thing is specifying --bare when you init the repo on the flash drive	19:56
yrlnry	there's nothing confusing about pushing to a non-bare repository as long as you don't push to the current HEAD ref.	19:56
cbreak	I did use a flash drive as kind of "portable server"	19:56
GreekFreak	When I cloned from my one pc from github, it marked it as clone. never master. does that make a difference?	19:57
	→ P4C0 joined	19:57
cbreak	but then I switched to a gitolite server on the wide network which is more convenient :)	19:57
SerajewelKS	GreekFreak: what does "marked it as clone" mean?	19:57
GreekFreak	yrlnry: what happens if you do?	19:57
	← skylamer` left	19:57
cbreak	GreekFreak: doesn't matter	19:57
	all you have are tracking infos	19:57
SerajewelKS	GreekFreak: the checked-out ref is updated but the working directory is not	19:57
cbreak	you can change them as much as you want, and add more remotes	19:57
P4C0	Hello, I made some changes to a branch and commit them, but now I realized I want those changes on a different branch (and the current back to original), I haven't pushed anything yet, how can I fix it?	19:58
SerajewelKS	GreekFreak: which results is confusion when you go to work on the remote repo, since your files are all dirty and you don't know why	19:58
cbreak	P4C0: easy: switch to the branch you want it	19:58
yrlnry	GreekFreak: what happens if you do what?	19:58
cbreak	git cherry-pick the commit	19:58
	then switch back and git reset --hard HEAD~1 it away	19:58
GreekFreak	yrlnry: commit the HEAD, but the guys just told me	19:59
	→ frikinz_ joined	19:59
	→ enigmabomb joined	19:59
GreekFreak	thanks guys. I wanted a git repo but thought I needed a server (which is the impression I got from the book). This solves my problem :)	19:59
P4C0	cbreak: thanks!	19:59
	→ bryanmv`away joined	19:59
	← diegoviola left	19:59
yrlnry	Yeah. Even I find that confusing and try to avoid it.	19:59
SerajewelKS	GreekFreak: yeah, git can use an sshd/httpd/gitd as a remote, but it can also use a local directory, which is really awesome	20:00
yrlnry	But pushing to a not-HEAD ref is not confusing at all.	20:00
	← iwoj left	20:00
cbreak	GreekFreak: git works just as well stand-alone	20:00
SerajewelKS	GreekFreak: incidentally, so can subversion (you can svnadmin create and then checkout that directory somewhere else)	20:00
GreekFreak	At least I know that the copy/pastes I've been doing haven't been for naught.	20:00
cbreak	unlike ... other tools :)	20:00
	← Ryback_ left	20:00
GreekFreak	cbreak: I've noticed	20:00
SerajewelKS	GreekFreak: indeed -- if the purpose is backup then copying .git works. however, pushing changes this way means there is less data to be copied, since git will copy only the new stuff	20:01
GreekFreak	I've grown to like it quite a bit :P	20:01
	→ CaptainKnots joined	20:01
	→ glennpratt_ joined	20:01
	← ToBeFree left	20:01
	← bryanmv`away left	20:01
	→ bryanmv`away joined	20:01
	← bryanmv`away left	20:01
	→ bryanmv`away joined	20:02
GreekFreak	SerajewelKS: I wanted to be able to just work with it on another machine (the purpose of the original question). Now I'll just do it The Right Way and kill 2 birds with one stone ;)	20:02
	← bryanmv`away left	20:02
SerajewelKS	GreekFreak: yup, it's a great solution	20:02
	→ bryanmv`away joined	20:02
SerajewelKS	fyi, i do use git for distributed backup and sync, but don't tell cmn ;)	20:03
GreekFreak	thanks for the help guys	20:03
SerajewelKS	git is at its core a content-addressable filesystem with versioning, and that's perfect for backups if you want to retain history forever	20:03
	dwmw2_gone → dwmw2_HIO	20:03
cmn	as long as you take backups of that	20:04
	→ HyperNerdV2_ joined	20:04
SerajewelKS	i use git to maintain a repo for my oblivion save game files, so i can sync them	20:04
cmn	the problem isn't making git part of it	20:04
SerajewelKS	well yeah, you have to have your backup repo somewhere else too	20:04
cmn	the problem is thinking that git is the only part you need to worry about	20:04
	← glennpratt left	20:04
GreekFreak	SerajewelKS: I was just about to say... I can use it for ANY kind of file then, not just text files for code :P	20:04
	→ Sigma joined	20:05
SerajewelKS	the cool thing is that i decided i wanted to experiment with vampire gameplay in oblivion. so what did i do? created a "vampire" branch.	20:05
	← psoo left	20:05
SerajewelKS	so i can checkout master or vampire depending on which fork of my character i want to play	20:05
GreekFreak	hahahahahahahaha	20:05
	that's taking it to a whole different level bru :P	20:05
SerajewelKS	it was the most obvious thing to do :)	20:05
	← drev1 left	20:05
SerajewelKS	i also use git to create daily snapshots of my minecraft server's world	20:06
	→ drev1 joined	20:06
	← jceb left	20:07
	→ alexukf joined	20:07
	← t4nkd`workin left	20:07
	→ t4nkd`workin joined	20:07
	← HyperNerdV2 left	20:07
	→ gusnan joined	20:08
	← Jackneill left	20:08
	← HyperNerdV2_ left	20:08
	← pppaul left	20:09
	→ pppaul joined	20:09
	→ alvaro_o joined	20:09
SerajewelKS	git can fill so many roles	20:10
	→ sayd joined	20:11
sayd	my coworker says using rebase on branches that are shared is a bad thing, is this true? ie should it never be used on shared branches	20:12
SethRobertson	Yes	20:12
	!rewriting_public_history	20:12
gitinfo	Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)	20:12
PerlJam	sayd: you can rebase all you want the stuff you haven't shared yet	20:12
SethRobertson	It can be done in carefully controlled circumstances, like in git's pu branch	20:12
GreekFreak	would be interesting to see if we could make it work with Google Drive haha	20:13
	← adhawkins left	20:13
sayd	PerlJam: SethRobertson thanks, it seems the 'stuff you haven't shared yet' is where i was confused	20:13
	does that mean a branch that has never reached a remote that would others use	20:13
	/s/would//	20:13
	→ Fullmoon joined	20:14
PerlJam	sayd: not just a branch but any commits you haven't pushed to a remote yet.	20:14
	→ iwoj joined	20:15
	→ khmarbaise joined	20:16
	← syphar left	20:17
	← arietis left	20:17
	← stkrzysiak left	20:17
	← GreekFreak left	20:17
	→ syphar joined	20:17
	→ booginga joined	20:18
	→ brentdanley joined	20:18
	← alfplayer left	20:18
	→ diegoviola joined	20:19
	← Sigma left	20:19
	← dr_lepper left	20:19
	→ teotwaki_ joined	20:21
	→ lonewulftp joined	20:21
	→ anderse_ joined	20:21
	← syphar left	20:22
	← shirish left	20:23
	← ben225 left	20:24
	← airborn left	20:24
	→ chrisgeorge_ joined	20:24
	← pppaul left	20:24
	→ camdenx_ joined	20:24
	→ wilmoore_ joined	20:24
	→ drev1_ joined	20:25
	→ tonebender_ joined	20:25
	→ vervic_ joined	20:25
	→ pppaul joined	20:26
	← drev1 left	20:26
	drev1_ → drev1	20:26
	→ MindDrive joined	20:27
	→ shirish joined	20:27
	← Dreamer3 left	20:27
	→ Dreamer3 joined	20:28
MindDrive	Hopefully quick question: is there a way to get a list of tags in (reverse) 'chronological' order? 'git tag' sorts alphanumerically, unfortunately, and all the other 'solutions' I'm finding online don't actually seem to work...	20:28
	→ thorbjor1DX joined	20:28
	← Shusshu left	20:29
	← rtuin left	20:29
	← pppaul left	20:30
EugeneKay	MindDrive - man git-log; look at --decorate=	20:30
gitinfo	MindDrive: the git-log manpage is available at http://jk.gs/git-log.html	20:30
	→ pppaul joined	20:30
EugeneKay	--tags too	20:31
	← unreal left	20:31
	→ unreal_ joined	20:32
	→ dr_lepper joined	20:32
	← dr_lepper left	20:32
	→ dr_lepper joined	20:32
	unreal_ → unreal	20:32
	→ jrnieder joined	20:32
	→ ben225 joined	20:33
	← ben225 left	20:33
	← vervic left	20:34
	← bolt left	20:34
	← tonebender left	20:34
	← wilmoore left	20:34
	← chrisgeorge left	20:34
	← camdenx left	20:34
	← Mowsh left	20:34
	← Element9 left	20:34
	← thorbjornDX left	20:34
	← maxorator left	20:34
	← Womkes left	20:34
	tonebender_ → tonebender	20:34
	vervic_ → vervic	20:34
	→ Element9 joined	20:34
	← flijten left	20:34
	→ ben225 joined	20:35
	→ bolt joined	20:36
	→ s0ber_ joined	20:37
	← limoges left	20:37
	→ ISF joined	20:39
	← s0ber left	20:39
	→ maxorator joined	20:39
	→ Mowsh joined	20:39
	→ Womkes joined	20:39
	s0ber_ → s0ber	20:39
	← Mowsh left	20:40
MindDrive	Thanks for the pointers, folks, though I think I found something that will do what I need (in this case, time should be relatively safe): git for-each-ref --format="%(taggerdate): %(refname)" --sort=-taggerdate --count=10 refs/tags \| awk '{print $NF}'	20:40
	→ Mowsh joined	20:40
	← anderse_ left	20:40
	→ aspotashev\|\| joined	20:40
	→ siprbaum_ joined	20:41
	← enigmabomb left	20:41
	→ [Czesiu] joined	20:41
	→ genehack_ joined	20:41
	→ jrdavid_ joined	20:41
	→ chilts_ joined	20:41
	→ Favorit_ joined	20:42
	→ harv_ joined	20:42
	→ SegFault1X\|work joined	20:42
	→ Aaeriele_ joined	20:42
	→ tharkun_ joined	20:42
	→ w0m_ joined	20:42
	→ MarcWebe1 joined	20:42
	→ heoa_ joined	20:42
	→ HobGoblin joined	20:42
	→ Textmode joined	20:42
	→ djh_ joined	20:42
	→ rasmusth_ joined	20:42
	→ mjessome_ joined	20:42
	→ Sil4nc4_ joined	20:42
	→ milki_ joined	20:42
	→ speakman_ joined	20:42
	→ kloeri_ joined	20:42
	→ thomas_b_ joined	20:42
	→ Spockz_ joined	20:42
	HobGoblin → Guest57075	20:43
	→ mikegerw1tz joined	20:43
	→ limoges joined	20:43
	→ selckin1 joined	20:43
	→ drev1_ joined	20:44
	← pppaul left	20:45
	→ enigmabomb joined	20:45
	→ pppaul joined	20:45
	← mikegerwitz left	20:45
	← drev1 left	20:45
	drev1_ → drev1	20:45
	← mikegerw1tz left	20:46
	← mishok13 left	20:46
	→ Vile joined	20:46
	→ mikegerwitz joined	20:46
	mikegerwitz → 15SABPL2S	20:46
	← neunon left	20:46
	← milki left	20:46
	← kloeri left	20:46
	← Spockz left	20:46
	← knittl left	20:46
	← selckin left	20:46
	← tystr left	20:46
	← Aaeriele left	20:47
	← thomas_b left	20:47
	← heoa left	20:47
	← Sil4nc4 left	20:47
	← mjessome left	20:47
	← speakman left	20:47
	← djh left	20:47
	← ChoiZ left	20:47
	← UukGoblin left	20:47
	← rasmusth left	20:47
	← MarcWeber left	20:47
	← w0m left	20:47
	← harv left	20:47
	← SegFaultAX\|work left	20:47
	← Favorit left	20:47
	← Czesiu left	20:47
	← siprbaum left	20:47
	← jrdavid left	20:47
	← genehack left	20:47
	← chilts left	20:47
	harv_ → harv	20:47
	← aspotashev\| left	20:47
	← cantoma left	20:47
	← [E]sc left	20:47
	← qwertz_ left	20:47
	← arvind_khadri left	20:47
	← segher left	20:47
	← newbie999\|3 left	20:47
	← jesse left	20:47
	→ knittl joined	20:47
	→ neunon joined	20:47
	← pandeiro left	20:47
	→ [E]sc joined	20:47
	→ arvind_khadri joined	20:47
	→ segher joined	20:47
	→ newbie999\|3 joined	20:47
	→ cantoma joined	20:47
	→ qwertz_ joined	20:47
	→ jesse joined	20:47
	← kloeri_ left	20:48
	→ kloeri joined	20:48
	→ alvaro_o_ joined	20:49
	← alvaro_o_ left	20:49
	→ pantsman joined	20:49
	← pantsman left	20:49
	→ pantsman joined	20:49
	← iwoj left	20:50
	← pppaul left	20:50
	← khmarbaise left	20:50
	→ pppaul joined	20:50
	→ tystr joined	20:51
	→ ChoiZ joined	20:52
	← alvaro_o left	20:53
	← infogulch left	20:53
	← tharkun left	20:53
	→ khmarbaise joined	20:53
	→ mcluskydodallas joined	20:53
	← chrisgeorge_ left	20:53
	← jsoa left	20:53
	← duckxx left	20:53
	→ duckx joined	20:53
	← vitor-br left	20:53
	← Anubisss_ left	20:54
	→ chrisgeorge joined	20:54
	← joeytwiddle left	20:55
	← JEEB left	20:56
	← aried3r left	20:56
	→ Anubisss_ joined	20:56
	← kloeri left	20:56
	→ kloeri joined	20:56
	← revagomes left	20:56
	→ revagomes joined	20:57
	← shirish left	20:57
	← theDrumm left	20:57
	← ben225 left	20:57
	← bartek left	20:57
	← freeed left	20:57
	→ theDrumm joined	20:57
	← ddelrio1986 left	20:58
	← [Czesiu] left	20:58
	← MistuhKurtz left	20:59
	← gtrak` left	20:59
	← jrnieder left	20:59
	← heoa_ left	21:00
	← diegoviola left	21:00
	← TTSDA left	21:00
	← asoltys_ left	21:00
	← apetrescu left	21:00
	← ocharles left	21:01
	← dv310p3r left	21:03
	← tchan left	21:03
	← atomicity left	21:03
	→ apetrescu joined	21:03
	→ bremner joined	21:03
	← revagomes left	21:03
	← bitkiller left	21:03
	← Internet13 left	21:03
	→ P-l joined	21:04
	→ cmn joined	21:04
	→ ben225 joined	21:04
	→ gtrak` joined	21:04
	← codeshah left	21:04
	← ThunderChicken left	21:05
	← meLon left	21:05
	→ Czesiu joined	21:06
	← pppaul left	21:06
	→ FreeArtMan joined	21:06
	→ isao joined	21:06
	→ pppaul joined	21:06
	← glennpratt_ left	21:07
	← coopsh left	21:07
	← libertyprime left	21:07
	← jmcantrell left	21:07
	← P4C0 left	21:07
	→ Rafusek_ joined	21:07
	← lukus left	21:07
	← Rafusek left	21:07
	← Adys left	21:07
	→ JEEB joined	21:08
	→ heoa joined	21:08
	→ pureluck_ joined	21:08
	→ robins joined	21:08
	→ isao1 joined	21:08
	← darkmethod left	21:08
	← klj613 left	21:08
	← cyphase left	21:08
	← gtrak` left	21:10
	← theDrumm left	21:10
	→ diegoviola joined	21:10
	→ lukus joined	21:10
	← mitsuhiko left	21:10
	← vmiklos left	21:10
	→ asoltys joined	21:11
	P-l → Guest72236	21:12
	→ iwoj joined	21:12
	← kloeri left	21:12
	→ aried3r joined	21:12
	← FreeArtMan left	21:12
	← robins left	21:12
	→ kloeri joined	21:12
	→ robinsmidsrod joined	21:12
	← Peste_Bubonica left	21:12
	→ vmiklos joined	21:12
	← tvw left	21:12
	→ infogulch joined	21:12
	→ tchan1 joined	21:12
	← AaronMT left	21:12
	← ymasory left	21:12
	→ dv310p3r joined	21:12
	→ tvw joined	21:12
	→ ocharles joined	21:13
	← Guest72236 left	21:13
	← isao1 left	21:13
	← erichynds left	21:13
	← defaultro left	21:16
	→ joeytwiddle joined	21:16
	← teotwaki_ left	21:16
	← Tohuw left	21:16
	← isao left	21:16
	→ TTSDA joined	21:16
	→ jstemmer joined	21:16
	← dv310p3r left	21:16
	← siprbaum_ left	21:16
	← __main__ left	21:16
	→ Bass10 joined	21:16
	← lukus left	21:18
	← Bass10 left	21:19
	← Orbitrix left	21:19
	← selckin1 left	21:19
	← BartlomiejB left	21:19
	→ dv310p3r joined	21:19
	→ __main__ joined	21:20
	← Vinnie_win left	21:20
	← dmilith left	21:20
	→ joeytwiddle_ joined	21:20
moltar_net1	I have a project structure like this: ./dir1/lib ./dir2/lib. Both "lib" are dirs as well. I decided that I don't need top level dir1 & dir2 anymore. So I did "git mv dir1/lib .", then "git mv dir2/lib". On the second move I get "fatal: cannot move directory over file, source=dir2/lib, destination=lib". Why is that?	21:21
	← jstemmer left	21:22
	← Chaser left	21:23
	→ lukus joined	21:24
	← bremner left	21:24
	← dv310p3r left	21:24
	← eicto left	21:24
	→ P-l- joined	21:25
	→ brokenshakles joined	21:26
	→ revagomes joined	21:26
	→ dv310p3r joined	21:26
	← joeytwiddle_ left	21:26
	← joeytwiddle left	21:26
	← Twey left	21:26
	← pppaul left	21:26
	→ bitkiller joined	21:26
	→ gtrak``` joined	21:26
	→ meLon__ joined	21:26
	→ ThC joined	21:26
	← dv310p3r left	21:26
	← Synthead left	21:26
	→ Adys_ joined	21:26
	← limoges left	21:26
	← fission6 left	21:26
	→ pppaul joined	21:26
	← juanfer left	21:26
	→ glennpratt joined	21:27
	→ libertyprime joined	21:27
	→ geeKISSexy joined	21:27
	→ jmcantrell_ joined	21:27
	→ joeytwiddle joined	21:27
	→ Internet13 joined	21:27
	→ theDrumm joined	21:28
	← ipalaus left	21:28
	→ Bass10 joined	21:28
	← iamjarvo left	21:28
	← trout left	21:28
	← Bass10 left	21:28
	← joeytwiddle left	21:28
	15SABPL2S → mikegerwitz	21:28
	← vmiklos left	21:28
	→ vmiklos joined	21:28
	→ atomicity joined	21:28
	← pppaul left	21:28
	→ pppaul joined	21:28
	→ ymasory_ joined	21:28
	→ AaronMT joined	21:28
	→ variable joined	21:28
	← pantsman left	21:28
	← whitman left	21:28
	→ joeytwiddle joined	21:28
	← MindDrive left	21:28
	→ defaultro joined	21:28
	← bolt left	21:28
	→ ipalaus joined	21:28
	← joshsmith left	21:28
	→ bolt joined	21:28
	→ d1b joined	21:28
	→ xaka joined	21:28
	→ JPohlman1 joined	21:28
	→ siprbaum joined	21:28
xaka	if i have commited, but not pushed changes yet, how can i see the diff? "git diff <commit>" or "git diff --cached <commit>" doesn't show anything	21:28
	← joeytwiddle left	21:28
SerajewelKS	xaka: "git diff <commit>" compares the commit to HEAD	21:28
xaka	but i can see my commit via "git log"	21:28
	→ CellKill_ joined	21:28
SerajewelKS	xaka: if the commit you named is the commit you have checked out, then of course you are not going to see anything	21:28
	→ tonyj joined	21:28
	→ huevos joined	21:28
SerajewelKS	xaka: "git show <commit>" will display a diff of the commit to its parent(s)	21:28
	← berserkr left	21:28
	→ batrick_ joined	21:28
	→ ghost_olli joined	21:28
cmn	xaka: HEAD is the tip of the branch you're currentlyh on	21:28
SerajewelKS	xaka: if the commit is the commit you have checked out, "git show HEAD" will do the trick	21:28
xaka	SerajewelKS: git show <commit> works, thank you!	21:28
	→ Tohuw joined	21:28
cmn	so trying to diff that with the worktree if you've just committed will be empty	21:28
	← Tohuw left	21:28
	→ Tohuw joined	21:28
	→ Vinnie_win_n joined	21:28
	→ Bass2 joined	21:28
	→ BartZilla joined	21:28
	→ kcj joined	21:28
	→ jimubao joined	21:28
	← BartZilla left	21:28
	← Dave^\| left	21:28
	← variable left	21:28
	→ dmil joined	21:28
	← ehsan left	21:28
	→ actel joined	21:28
	← Maccer left	21:28
	gitinfo set mode: +v	21:28
	← pppaul left	21:28
	→ morgoth_ joined	21:28
	→ Chaser__ joined	21:28
	→ variable joined	21:28
	→ pppaul joined	21:28
	→ xou_ joined	21:28
	← lukus left	21:28
	→ lukus joined	21:28
FauxFaux	moltar_net1: Ignore git mv, it's only a helper. git add afterwards is identical.	21:28
	→ mage_ joined	21:28
	→ Twey joined	21:28
	→ bremner joined	21:28
	← sq-one left	21:28
	→ dv310p3r joined	21:28
	→ shad joined	21:28
	← glcrazy left	21:28
	← chrisgeorge left	21:28
	→ glcrazy joined	21:28
	→ chrisgeorge joined	21:28
	Vinnie_win_n → Vinnie_win	21:28
moltar_net1	FauxFaux: so I can just manually re-arrange everything, and git would pick it up as intended?	21:28
	→ juanfer joined	21:28
cmn	you'd need to tell git	21:28
	← tbird72 left	21:28
moltar_net1	cmn: but of course .. :)	21:28
	I meant "add" would tell git to pick it up as intended?	21:29
FauxFaux	moltar_net1: Yeah. git add or git add -A afterwards.	21:29
moltar_net1	it'll know that files have moved?	21:29
	← LittleFool left	21:29
FauxFaux	Yes.	21:29
moltar_net1	I guess it can tell by some sort of checksum, right?	21:29
	← fly9 left	21:29
moltar_net1	too good :)	21:29
	→ LittleFool joined	21:29
	→ fly9 joined	21:29
	← chrisgeorge left	21:29
	→ chrisgeorge joined	21:30
	← ghost_olli left	21:30
	→ joeytwiddle joined	21:31
cmn	git calculates movements on the fl	21:31
	→ ghost_olli joined	21:31
	← fly9 left	21:31
cmn	it will simply record the new files and not record the old ones	21:31
	→ fly9 joined	21:31
	→ Lord_DeathMatch joined	21:31
	→ lcpfnvcy joined	21:31
moltar_net1	cmn: but will it maintain history?	21:31
	i.e. when I check log for the moved file, will I see commits for when it was under a different name?	21:32
	→ jstemmer joined	21:32
cmn	with --follow	21:32
	blame will tell you if the code originally came from the old one	21:32
	git keeps snapshots	21:32
	any changes are all calculated at runtime	21:33
	→ w09x joined	21:33
	← Bass2 left	21:33
	← joeytwiddle left	21:34
Dreamer3	can someone help me with rebasing?	21:34
	← chrisgeorge left	21:35
	← umc left	21:35
Dreamer3	i have a conflict… is there any way to see the two files from both sides of the merge… or tell it which one to use outright instead of merging manually?	21:35
	→ Oliver_ joined	21:36
	← LongBeach left	21:36
	← d0k left	21:37
	← dzonder left	21:37
	→ joeytwiddle joined	21:38
	→ Superlokkus joined	21:38
	→ pandeiro joined	21:38
	→ dzonder joined	21:38
cmn	checkout --theirs or --ours will checkout one of the versions	21:38
	git mergetool will start up a merge tool	21:38
	→ chrisgeorge joined	21:38
cmn	by default vimdiff, you can use many others	21:38
Dreamer3	hmmm	21:38
	← ph^ left	21:38
Dreamer3	so if i'm on a branch… and rebasing on master...	21:38
	--theirs would be?	21:39
	i.e. i'm running "git rebase master"	21:39
cmn	your version	21:39
	a rebase is doing things the "wrong" way around	21:39
	← JohnyBaby left	21:39
Dreamer3	so theirs is from my local branch?	21:39
cmn	you can also see which side has which version from the contents is the file	21:39
Dreamer3	which am i checking out if i don't specify?	21:39
	← ghost_olli left	21:39
Dreamer3	sounds like the kind of thing i need to play with in a test project	21:40
cmn	you have to specify the file(s)	21:40
	well, you can always use !fixup to recover	21:40
gitinfo	So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	21:40
Dreamer3	right but i can just run "checkout" without --theirs or --ours	21:40
cmn	you can	21:40
Dreamer3	i would assume i'm getting my branches copy?	21:40
cmn	but you'd have to figure out exactly which version you want	21:41
	and checkout already knows which exact file is each	21:41
	← ben225 left	21:41
	← glcrazy left	21:41
	← t4nkd`workin left	21:41
	← Sahua_ left	21:42
	← joeytwiddle left	21:42
	→ ben225 joined	21:42
Dreamer3	ok i have a good test	21:42
	so if i clobble it with checkout --theirs or --ours… is there a quick way to get the merged file with conflicts back?	21:43
	it's trying to find p4merge	21:44
	hmmm	21:44
	any really great guy merge tools for OS X?	21:44
	← petershaw left	21:44
	← blacktulip left	21:46
	→ joeytwiddle joined	21:46
	→ yoklov joined	21:46
	Spockz_ → Spockz	21:47
rking	Oops. I'm drawing a blank. I have a repo I forked quite a while back, but now I'm about to work on a new Pull Request -- how do I update it so that I'm in sync with the current upstream repo?	21:47
	← Superlokkus left	21:48
	dwmw2_HIO → dwmw2_gone	21:49
	← dv310p3r left	21:49
	← Oliver_ left	21:49
	← booginga left	21:49
	→ ghost_olli joined	21:49
raek	rking: if you have a clean working tree and no local commits, then just do "git pull origin"	21:50
	→ theCole joined	21:50
rking	But my origin is me?	21:51
	Oh no it isn't.	21:51
	← brokenshakles left	21:52
raek	rking: you can add the fork that's not present using "git remote add <name> <url>"	21:52
	→ pfFredd joined	21:52
	← pandeiro left	21:52
rking	OK	21:52
	→ smoge joined	21:52
	→ isao joined	21:52
rking	Then basically I do a git pull origin; but a git push rkingstuff ?	21:52
	→ saurik joined	21:53
	← ben225 left	21:54
raek	yeah	21:54
	it's probably good to specify the branch when you do the push	21:54
rking	OK, thanks. That does sound like what I had been informed of, before.	21:54
	Oh yeah.	21:54
	← ghost_olli left	21:54
raek	git push rkingstuff issue-123-patch	21:54
rking	That will automatically make a branch for it?	21:54
	Or I have to git branch first?	21:54
	→ ghost_olli joined	21:55
raek	that will push your local issue-123-patch branch to the remote issue-123-patch branch	21:55
	← saurik left	21:55
raek	no	21:55
	→ saurik joined	21:55
raek	well, the remote branch will be created	21:55
	→ MistuhKurtz joined	21:55
	← ghost_olli left	21:56
rking	Here's the whole process, as I understand it:	21:56
raek	you might want to do "git fetch origin" and then "git checkout -b issue-123-patch origin/master" to set up your branch, or something similar	21:56
	→ ghost_olli joined	21:56
	← cakehero left	21:56
	← ghost_olli left	21:57
	→ berserkr joined	21:57
	← pppaul left	21:57
raek	(git pull fetches from the remote and then merges the remote branch with your local one)	21:57
rking	1) "Fork" through the web UI (is there a CLI tool for this?) 2) clone that resultant URL, 3) git branch issue-123-patch, 4) git push 5) ...time passes.. 6) git pull origin... crap I'm getting confused now.	21:58
	← tyatpi left	21:58
	→ cagedwisdom joined	21:58
	← ISF left	21:58
raek	that should work	21:58
	← edude03 left	21:58
rking	Then that's it? Simply git branch issue-124-patch; git push ?	21:59
EugeneKay	rking - you probably want to add a remote named "upstream" to your local clone which points to the original repo	21:59
	→ pppaul joined	21:59
	← dvide left	21:59
EugeneKay	Then you would fetch upstream master; checkout master; merge upstream/master	21:59
rking	Hrm.. merge?	21:59
	Because I did branch I assume.. but when does that happen?	22:00
raek	rking: I would recommend using the non-abbreviated form of git push (where you specify remote and branch) when learning	22:00
EugeneKay	Branch creates a divergence from the upstream master to your working branch	22:00
rking	Aha, I can get into that.	22:00
	← shruggar left	22:01
raek	the defaults you get when you ommit the arguments to pull and push tend to be confusing	22:01
EugeneKay	The reason you need to merge upstream/master into your local master(and probably pus hthat back to your clone's master) is that you don't automagically get new commits that were made to the upstream.	22:01
raek	and depends on how you created the current branch	22:01
rking	OK. I think I follow.	22:02
	I'll give it a whirl and probably be back in a few weeks when I need to do it again, unless I happen to retain this.	22:03
	At the moment it looks like I'm sorted.	22:03
	← leonardotartari left	22:03
rking	Now I have an entirely different question - I did a `git blame` and determined that commit ce52b38e is dubious, but I want to know more about it. I did a `git log ce52b38e` and found out that the commit message doesn't justify the change, so now I want to see the whole diff of ce52b38e	22:04
	→ java joined	22:04
	← defaultro left	22:04
	→ defaultro joined	22:04
rking	But git diff ce52b38e seems to do something different.	22:04
cmn	rking: show	22:04
EugeneKay	rking - !lol helps to keep it straight	22:04
gitinfo	rking: git config --global alias.lol "log --oneline --graph --decorate"	22:04
cmn	git diff <commit> compares HEAD with that commit	22:04
	or git log -p	22:04
	→ BlueProtoman joined	22:04
	← java left	22:05
rking	WHOA.	22:05
	--decorate goes to town.	22:05
BlueProtoman	I'm getting an error, something about fast forwards. Someone mind explaining it to me? http://pastebin.com/Dt9t5ASG	22:05
cmn	someone else pushed	22:05
	or you rewrote published history	22:05
raek	--all is nice too	22:05
	← nrichards left	22:05
	← lacrymology left	22:05
rking	Strangely `git lol` doesn't show ce52b38e	22:05
BlueProtoman	cmn: Well, I DID check "Amend Last Commit" on git-gui.	22:06
cmn	so your push would overwrite history and possibly lose infromation	22:06
BlueProtoman	So how can I fix this?	22:06
cmn	BlueProtoman: that would be it then	22:06
	← albel727 left	22:06
cmn	first make sure	22:06
	fetch from that remote and compare	22:06
	→ kinetic joined	22:06
	→ dvide joined	22:06
	→ Maccer joined	22:06
	← Lord_DeathMatch left	22:06
	← juanfer left	22:07
cmn	once you're sure that the only change is that ammend, you can force-push with -f	22:07
	← fly9 left	22:07
	← pppaul left	22:07
cmn	and then apologise to the people who can see that branch	22:07
rking	Hrm, I am definitely appreciating these tips, but I'm still not quite able to accomplish the task - I basically need to learn everything about a certain chunk. I've gotten as far as what `git blame` shows me, but I need to go further back in time to find the first time that chunk was introduced.	22:08
	→ pppaul joined	22:08
cmn	rking: blame -C -C	22:08
	← sgncfl left	22:08
	← w09x left	22:08
rking	Oh, sounds cool.	22:08
	→ limoges joined	22:08
cmn	that looks for the original place the code was introduced	22:08
	it will only work if the change came in as part of a chunk of code	22:08
BlueProtoman	cmn: This is just my own project, I'm the only one working on it.	22:08
cmn	following one single line won't trigger the detection	22:08
rking	Hrm, does it expect more args? `git blame -C -C foo.asdf` is giving me the same commit hash for that line.	22:09
cmn	then taht's what git thinks that line was introduced	22:10
	where	22:10
rking	Hrm.	22:10
	cmn: OK, thanks.	22:10
	I'm nuking it. =)	22:10
	→ ferdna joined	22:10
	← txomon\|nqas left	22:10
	→ txomon\|nqas joined	22:12
	← txomon\|nqas left	22:12
	→ txomon\|nqas joined	22:12
	← pppaul left	22:12
	→ mineiro joined	22:12
	→ pppaul joined	22:12
	→ Chillance joined	22:12
BlueProtoman	cmn: Thanks for the tip!	22:13
	← BlueProtoman left	22:14
cmn	make sure you don't abuse force-pushes	22:14
	← tizzo left	22:16
	→ TheShagg joined	22:18
	gitinfo set mode: +v	22:18
TheShagg	I am trying to fetch commit/tag shown here:	22:18
	http://arago-project.org/git/projects/?p=linux-davinci.git;a=commit;h=2acf935c01b9adb50164d421c40cdc5862b9e143	22:18
	→ JohnyBaby joined	22:18
FauxFaux	Need to write me a hook that allows trivial force-pushes, but not scary ones. For some definition of the above. Actually, I bet someone's written a hook that allows force-pushes of unpulled history in gitolite or similar.	22:18
	← pppaul left	22:18
TheShagg	but when I try and fetch it it says Couldn't find remote ref	22:18
FauxFaux	TheShagg: git fetch origin --tags	22:19
	→ pppaul joined	22:19
	← RaceCondition left	22:20
	← pppaul left	22:20
	→ brokenshakles joined	22:20
	→ drev1_ joined	22:20
	→ albel727 joined	22:20
	← juvenal left	22:21
TheShagg	FauxFaux: wow that takes a long time on a linux kernel	22:21
	← twisted` left	22:21
	→ pppaul joined	22:21
	← obert- left	22:22
	← yoklov left	22:22
FauxFaux	It might be fetching a bit more data, too. Or a lot, depending on how crazy the author is.	22:22
	← drev1 left	22:22
	drev1_ → drev1	22:22
	→ twisted` joined	22:24
TheShagg	FauxFaux: so if I grep my git show-ref I can see it in refs/tags	22:24
	but when I try and: "git fetch arago v2.6.33-rc4_DAVINCIPSP_03.20.00.13" it says Couldn't find remote ref v2.6.33..."	22:25
FauxFaux	But, if it's in refs/tags, you have it? So why do you need to fetch it?	22:25
cmn	TheShagg: use the real name	22:25
	if you don't specify a namespace, git assumes you mean a branch	22:25
TheShagg	elaborate?	22:25
	real name?	22:25
	← isao left	22:25
TheShagg	refs/tags/.... ?	22:26
cmn	yes	22:26
TheShagg	k	22:26
cmn	I think fetch also lets you say fetch tag v1... and does it for you	22:26
	but you just fetched it	22:26
	← duckx left	22:26
cmn	so there isn't really anything for git to do	22:26
TheShagg	ah, that worked	22:26
	thank you	22:26
	was not aware that its not smart enough to search for it as a tag	22:26
	→ alesan joined	22:26
TheShagg	I figured it used all namespaces	22:26
alesan	hi what is a guide for a friend that wants to learn git and has never used any svn or group development instrument?	22:27
	he's a programmer for 15 years but a lot in php or individual stuff	22:27
	← dzonder left	22:27
TheShagg	and he hasn't used version control?	22:27
	does he use windows?	22:27
	or she	22:28
alesan	no ubuntu now	22:28
	HE	22:28
	→ wizbit joined	22:28
alesan	I know only one female that knows how to use git well	22:28
	but that is not relevant	22:28
TheShagg	I learned version control on windows using tortoise SVN and i think that helped me maintain sanity\	22:28
kevlarman	alesan: what's the question?	22:28
wizbit	does git come with any global configuration examples?	22:28
TheShagg	the git tools are really clunky for newbs	22:28
alesan	kevlarman, <alesan> hi what is a guide for a friend that wants to learn git and has never used any svn or group development instrument?	22:28
	← radii left	22:28
kevlarman	alesan: !book is probably a good place to start	22:29
gitinfo	alesan: There are several good books available about git; 'Pro Git' is probably the best: http://progit.org/book/ but also look at !bottomup !cs !designers !gitt !vcbe and !parable	22:29
	← jorts left	22:29
cmn	TheShagg: it's not about being smart; the rule is that an unqualified name for fetch is a branch	22:29
alesan	so "pro git"?	22:29
kevlarman	alesan: learning git is actually much easier if you've never learned svn	22:29
	(or cvs, etc.)	22:29
alesan	well if you know svn there is the git svn crash guide	22:30
TheShagg	cmn: if it were smart, it would look for tags too :)	22:30
offby1	I have a whole range of commits that have the wrong email address (because I forgot to update ~/.gitconfig). Can I update them all in one go? As opposed to doing an interactive rebase and editing each, then doing "git commit --amend --reset-author" on each?	22:30
cmn	TheShagg: see my earlier comment	22:30
	← macabre left	22:30
	← diago left	22:30
kevlarman	alesan: problem is when you know svn you expect git to act like it, which git doesn't do because svn's way of doing things is terrible	22:30
TheShagg	cmn: see my earlier comment	22:30
	kevlarman: are you actualy linus, trolling the git irc	22:31
kevlarman	TheShagg: hehe	22:31
	i wish	22:31
alesan	kevlarman, are you saying it would be much easier to leran Java if you have absolutely no C experience?	22:31
kevlarman	alesan: no	22:31
alesan	learn	22:31
	what is the difference then	22:31
	← giallu left	22:31
alesan	git is like jave if svn is C, not?	22:31
	Java	22:31
bremner	no	22:31
kevlarman	alesan: no	22:31
TheShagg	kevlarman: my specific statement also mentioned using tortoise SVN, which is probably highly different from the command line	22:31
ohcibi	no	22:31
bremner	first of all, that's a crazy analogy	22:31
alesan	I do not understand then	22:31
	→ avinashhm joined	22:32
alesan	anyway	22:32
kevlarman	alesan: make that visual basic 6 and lisp, and maybe the analogy kinda works	22:32
alesan	I love LISP	22:32
	Scheme actually	22:32
bremner	alesan: don't worry about it then. But many svn users expect git to be like svn	22:32
	and then they are sad	22:32
	← glennpratt left	22:32
TheShagg	git is like svn	22:32
	just like lisp is like visual basic	22:33
	← revagomes left	22:33
kevlarman	alesan: i spent a ton of time banging my head on the desk before i managed to unlearn svn	22:33
	offby1 suddenly has an idea: run 'git-format-patch' and then "git am"	22:33
kevlarman	once i got past that learning git was easy	22:33
TheShagg	you used the command line, didn't you?	22:33
	← Tinti left	22:33
alesan	!bottomup	22:33
gitinfo	'Git from the bottom up' starts with explaining the building blocks of git and proceeds to tell you how they fit together. http://ftp.newartisans.com/pub/git.from.bottom.up.pdf	22:33
kevlarman	TheShagg: yes, for both	22:33
alesan	!cs	22:33
gitinfo	'Git for Computer Scientists' explains the basic structures used by git, which is very helpful for understanding its concepts. http://sitaramc.github.com/gcs/	22:33
TheShagg	kevlarman: thats the problem	22:33
kevlarman	alesan: you can /msg the bot as well	22:33
alesan	ok	22:34
TheShagg	use the gui to learn the concept with a rich interface before you start trying to figure out some bass ackwards gui command structure	22:35
	sorry, command line structure	22:35
ohcibi	o-O	22:35
	→ Tinti joined	22:35
	→ drizzd joined	22:35
bremner	TheShagg: ok, that's your opinion. I think you'll find not many here agree.	22:35
TheShagg	that's fine	22:36
ohcibi	"use the gui to learn concepts".... _wrong_ 8-))	22:36
Dreamer3	is there something better than gitk to show branch history?	22:36
	i can't even figure out how to resize the top pane	22:37
ohcibi	Dreamer3: i like giggle	22:37
	← pppaul left	22:37
TheShagg	if you use windows, tortoise GIT works nicely	22:37
Dreamer3	i need mac software	22:37
TheShagg	gitk is the only one I have seen for linux that does proper multithreading	22:37
bremner	that is the first nice thing I have ever heard about tortoise git	22:38
ohcibi	8-)	22:38
	→ pppaul joined	22:38
Dreamer3	i have two branches, how can i tell the point from which they diverged?	22:38
	← hwrd\|work left	22:39
TheShagg	I like tortoise git a lot, gets better all the time, except the problem with it is that it doesn't exist for linux, and using it with a repo that is also used in linux presents all the typical windows/linux incompatibility issues (such as file name case and newlines)	22:39
ivan	I'm using git-svn and it's pulling ~7000 revisions per tag repeatedly per tag, and there are 200 tags. Is there a script somewhere that imports tags without grabbing history from r1 for each?	22:39
	I don't mind being unable to obtain updates on those tags	22:40
ohcibi	TheShagg: there is no need for a git-gui for linux, we know how to deal with git 8-)	22:40
	← chrisgeorge left	22:40
ohcibi	in fact i tell my students to NOT use any gui because they'll learn the gui and not git	22:40
	*they'd	22:41
Dreamer3	hmmm	22:41
	how can i rewrite my commit history and remove a commit completely?	22:41
bremner	rebase -i	22:41
Riviera	Sounds silly, make them learn concepts first, applying these concepts then on the command line should be much easier.	22:41
	(And a GUI might be rather helpful in learning concepts.)	22:41
Dreamer3	bremner: that tells me it doesn't know what to rebase against	22:41
	← milki_ left	22:42
	→ shuvo joined	22:42
bremner	Dreamer3: I guess I should have said main git rebase -i	22:42
	→ milki joined	22:42
bremner	err. you know, check the man page	22:42
ohcibi	Riviera: for the history ok, but the rest is better done with the command line	22:42
	→ macmartine joined	22:42
	→ RaceCondition joined	22:43
	gitinfo set mode: +v	22:43
ohcibi	plus experience with windows users shows: gui-users dont learn any concepts 8-))	22:43
Dreamer3	man	22:43
shuvo	hello everyone, I had a question	22:43
	can I have branches of my source code that excludes certain directories from the main branch? think "views" in a SQL DB	22:43
	← JohnyBaby left	22:43
shuvo	something like .gitignore, but for branches	22:43
Dreamer3	i seem to remember something giving me a list of commits and i could just commit one out and then be done	22:43
ohcibi	but enough trolling... good night everyone 8-)	22:43
	batrick_ → batrick	22:44
Riviera	nite ohcibi 8)	22:44
TheShagg	ohcibi: you seem as equally snobbish as I, perhaps we could arrange some mortal enemyship?	22:44
kevlarman	shuvo: no	22:44
bremner	Dreamer3: yes, that is rebase -i, but you need the right parameter. see !fixup	22:44
gitinfo	Dreamer3: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!	22:44
	← keltvek left	22:45
	← teweWORK left	22:45
shuvo	thanks kevlarman. there is no way to have additional directories "removed" from the "main branch"?	22:46
Dreamer3	qhmmm	22:46
	← RaceCondition left	22:46
	→ w09x joined	22:47
Dreamer3	http://sethrobertson.github.com/GitFixUm/fixup.html#remove_deep	22:47
	ok	22:47
	← dvide left	22:47
Dreamer3	git rebase -p --onto SHA^ SHA	22:47
	→ royale1223 joined	22:47
wizbit	how can i make sure a hook gets executed?	22:48
	← royale1223 left	22:48
wizbit	post-update	22:48
	does it just need chmod +x ?	22:48
	← limoges left	22:48
Dreamer3	so i get that i'm rebating onto the revision BEFORE my "bad" commit… but what does adding the SHA again do	22:48
	← fisted left	22:48
	→ fisted joined	22:49
	→ thebignoise joined	22:50
	→ limoges joined	22:51
	→ dvide joined	22:51
TheShagg	is there a shorter way to write: "gitk HEAD~30..HEAD" ?	22:51
	← pppaul left	22:52
dr_lepper	gitk HEAD~30..	22:52
	← aspotashev\|\| left	22:53
	→ aspotashev joined	22:53
TheShagg	dr_lepper: cool, but nothing shorter than that to show the last 30 commits?	22:53
	→ pppaul joined	22:53
dr_lepper	no idea, never used gitk	22:54
TheShagg	huh	22:54
	→ hwrd\|work joined	22:55
dr_lepper	you could try, for example gitk -30	22:55
	← txomon\|home left	22:55
dr_lepper	it works with qgit because it uses git log internally	22:55
Dreamer3	ok that was neat	22:55
dr_lepper	so i strongly suspect gitk does	22:55
TheShagg	dr_lepper: Brilliant! yes so does gitk (i see the errors when I feed it bad args)	22:56
	thats perfect	22:56
	thank you	22:56
dr_lepper	anytime	22:56
	→ krz joined	22:56
	← aspotashev left	22:57
	→ coopsh joined	22:57
	→ adammokan joined	22:59
	← alexukf left	22:59
	← neurodrone left	23:00
	← TeckniX left	23:00
	→ mineiro_ joined	23:01
	← mritz left	23:02
	→ ConstantineXVI joined	23:03
	← mineiro left	23:04
	→ mxweas_ joined	23:04
	→ tizzo joined	23:04
	← mxweas_ left	23:07
	← martinjlowm-lptp left	23:07
	← martinjlowm left	23:08
	← chilts_ left	23:09
	→ mritz joined	23:10
	← _iron left	23:10
	→ nrichards joined	23:10
	→ keltvek joined	23:12
	← kermit left	23:12
osmosis	"Your branch is behind 'origin/master' by 10 commits, and can be fast-forwarded." How?	23:12
	← AaronMT left	23:13
bremner	osmosis: man git merge	23:13
gitinfo	osmosis: the git-merge manpage is available at http://jk.gs/git-merge.html	23:13
bremner	osmosis: in particular, git merge origin/master	23:13
	→ JohnyBaby joined	23:13
	← kevwil left	23:13
	← tonebender left	23:13
	→ RaceCondition joined	23:14
	← nrichards left	23:15
	→ selckin1 joined	23:15
	→ spaceonline joined	23:16
	← xaka left	23:17
	← pppaul left	23:18
	← tatsuya_o left	23:18
	→ pppaul joined	23:19
shuvo	kevlarman, @all, in that case, I could do the opposite: maintain the minimum common files/directories as the "main branch" and have the files/directories that can change (get added) across branches in each branch (removing the need to "remove" files/directories from the "main branch" instead)?	23:19
	→ cyphase joined	23:19
kevlarman	shuvo: yes	23:20
	→ lindenle_ joined	23:20
	→ Drakevr joined	23:20
	← Drakevr left	23:20
	→ Drakevr joined	23:20
	← drev1 left	23:20
	→ zorzar_ joined	23:21
shuvo	and later, if I WANTED TO, I could merge in some branches (and their disparate directories) using git merge or similar?	23:21
	← macmartine left	23:21
	← superdmp left	23:23
	← Element9 left	23:23
	→ AaronMT joined	23:24
	← eijk left	23:24
	→ ScottE_ joined	23:24
	→ mitsuhiko_ joined	23:25
sitaram	FauxFaux: "force pushes of unpulled history"? Could you explain?	23:25
	→ tizzo_ joined	23:25
	← ScottE left	23:25
	← CellKill_ left	23:26
FauxFaux	sitaram: Allow you to force a push iff nobody has observed the history you're destroying.	23:26
	→ CellKill joined	23:26
	← Tinti left	23:26
	← tizzo left	23:26
	tizzo_ → tizzo	23:26
FauxFaux	i.e. if, for the simplest case, if someone's done a pull since push-time of merge-base master@{1} master, disallow the forciness of the push?	23:27
	→ aZnmAn joined	23:27
	→ sjefen6_ joined	23:27
	← zendeavor left	23:27
	→ serj joined	23:27
	← joeytwiddle left	23:28
	← aried3r left	23:28
	← arvind_khadri left	23:28
	← JPT left	23:28
	← ymasory_ left	23:28
	← alberto56 left	23:28
	← piti left	23:28
	← serj_ left	23:28
	← sjefen6 left	23:28
	← sente` left	23:28
	→ tyatpi joined	23:28
aZnmAn	Hi, trying to remove remove unchanged files from unstaged changes. I found a post mentioning crlf/lf line-ends, and to disable core.whitespace = fix... I checked git config and don't see it enabled. How do I fix this?	23:28
	← asakura0 left	23:29
	← catsup left	23:29
FauxFaux	aZnmAn: It also picks up on encoding and case changes, are you on a non-linux-like filesystem?	23:29
	→ JPT joined	23:29
aZnmAn	yeah, my remote repo is on CentOS, but my local deev machine is running win 7 pro	23:30
	*dev	23:30
	→ CellKill_ joined	23:30
	← CellKill left	23:31
	← bglusman left	23:31
sitaram	FauxFaux: even in theory, only that simplest case is possible, I think	23:31
	FauxFaux: because a fetch is a fetch; gitolite doesn't/can't know what was fetched, far less what was done with the fetched branch(es)	23:32
osmosis	how do I pull the new changes from master into my current branch?	23:33
	→ alberto56 joined	23:33
FauxFaux	Even assuming everything was fetched would probably work for me. I'm thinking of forces where you realise you made a mistake shortly after pushing, because I'm a retard. Or after the CI server / deploy script notices, or whatever. Maybe you'd argue that forces in this situation are bad anyway.	23:33
	→ Turicas joined	23:33
	← morgoth_ left	23:34
	→ zendeavor joined	23:34
	← tvw left	23:34
	→ marhaban joined	23:35
	→ ymasory_ joined	23:35
kevlarman	FauxFaux: tbh i'd use gerrit for something like that	23:35
	→ aried3r joined	23:36
	→ sente` joined	23:36
	→ joeytwiddle joined	23:36
	→ arvind_khadri joined	23:36
aZnmAn	FauxFaux: ? Is there a fix?	23:37
	→ madewokherd joined	23:37
	→ asakura0 joined	23:37
	→ piti joined	23:37
	← mritz left	23:37
	→ sgncfl joined	23:37
	← sgncfl left	23:38
	← eephillip left	23:38
	← drizzd left	23:38
	→ kerrick joined	23:39
sitaram	kevlarman: how would gerrit do this? (other than adding a temp "for-master" branch in between)	23:40
kevlarman	sitaram: the latter	23:40
	sitaram: since everything is so heavily patch based, there isn't anything to rebase except the things that you know are going to be rebased	23:41
	← thebignoise left	23:41
	← iOPK left	23:41
	← RaceCondition left	23:41
sitaram	kevlarman: in theory you could do the same with gitolite also; push to a branch that people are told not to fetch, and post-receive triggers CI off of that branch only and if it works moves it to the main one	23:41
	← hwrd\|work left	23:41
kevlarman	sitaram: yup	23:41
	sitaram: i just love the idea of forcing code review like gerrit does	23:42
	← nadavoid left	23:42
	→ OPK joined	23:42
sitaram	kevlarman: yeah gerrit is apparently pretty good that way; I always tell people who want code review to swicth to gerrit from gitolite	23:42
	switch*	23:42
	→ catsup joined	23:42
	← Chillance left	23:42
milki	gerrits custom git stack is weiiiiierd	23:43
kevlarman	ew it uses a custom git stack?	23:43
FauxFaux	But it's written in Java. It muts be good.	23:43
milki	haha	23:43
sitaram	and custom ssh, and a database, and....	23:43
FauxFaux	sitaram: But they're apache projects! They must be good.	23:43
sitaram	even the sshd is in Java	23:43
milki	cant use normal git-hooks	23:43
	which means if you are mirroring and stuff	23:43
	you need to do wierd stuff	23:44
FauxFaux	kevlarman: Same thing eclipse-git uses (jgit).	23:44
sitaram	FauxFaux: why "must be"?	23:44
FauxFaux	Sarcasm. ¬_¬	23:44
sitaram	FauxFaux: aah ok!	23:44
milki	lol	23:44
	you have those characters in your smiley easily accessible?	23:44
kevlarman	FauxFaux: double-ew	23:44
sitaram	FauxFaux: my serious opinion is that they would have some credibility as being from a non-small player, but Apache is too spread out, and anyone who is still putting all their projects on one SVN repo (last I heard) needs a lobotomy	23:45
milki	uh	23:45
	milki averts eyes	23:45
milki	its not my fault. its legacy!	23:45
	→ shwaiil joined	23:45
shwaiil	hi	23:46
gitinfo	shwaiil: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.	23:46
sitaram	and since I have paid my dues with COBOL from 1986 to about 1993, I refuse to use the "COBOL of the internet" for anything serious. (Freemind is the only exception)	23:46
milki	sitaram confirms his ancientness	23:46
sitaram	as if you didn't know	23:46
	already	23:46
milki	:P	23:46
	→ morgoth_ joined	23:46
sitaram	:)	23:47
FauxFaux	sitaram: We have one big 870,000 revision svn repo at work! It's a brilliant idea!	23:47
	→ rchavik joined	23:47
	← rchavik left	23:47
	→ rchavik joined	23:47
sitaram	how many subprojects?	23:47
FauxFaux	All of them. (Hundreds.)	23:47
sitaram	oh nice; you're a corporate version of Apache!	23:47
	← morgoth_ left	23:48
	→ thebignoise joined	23:48
	→ eicto joined	23:48
shwaiil	Q: I'm trying to learn how to deploy using Git. I came across (https://github.com/EugeneKay/scripts/blob/master/bash/git-deploy-hook.sh), that was sent to me via this room. Since I've done some reading (but not enough yet, still in learning process), I realised that to deploy to my production server, I need to add it as a Repo. When push to this repo, that's when the "hook" (like in the link), runs. Is this correct ? Thanks! OBS: My main	23:48
osmosis	how do I pull the new changes from origin/master into my current branch?	23:48
	← kenperkins left	23:48
FauxFaux	Yeah. We heard it was the least efficient way to run anything, so we were right in there. (no sarcasm this time)	23:48
	→ Pupeno_W_ joined	23:49
	← adammokan left	23:49
	→ dv310p3r joined	23:49
	← kadoban_ left	23:49
	→ Vert joined	23:50
	← keltvek left	23:50
	← rchavik left	23:50
	→ rchavik joined	23:51
	← rchavik left	23:51
	→ rchavik joined	23:51
	← Pupeno_W left	23:51
	Pupeno_W_ → Pupeno_W	23:51
	← pfFredd left	23:52
	→ Jippi joined	23:52
	→ neurodrone joined	23:52
osmosis	git pull origin master	23:52
	→ txomon\|home joined	23:53
	← txomon\|home left	23:53
	→ txomon\|home joined	23:53
	← notjohn left	23:53
	→ mxweas_ joined	23:56
	← DofLeFou left	23:58
	→ kermit joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation