IRCloggy #git 2014-01-23

Provider of IRC logs since 2005.

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

	2014-01-23
grawity	it's still useful to find out where someone else screwed it up	00:00
	→ thiago joined	00:00
	← skorgon left	00:00
	→ Exancillatus joined	00:00
	← tbmiller_ left	00:01
xiong	It's your fault.	00:01
	→ halu joined	00:01
	← lavaman left	00:01
	→ cagedwisdom joined	00:02
grawity	it's not my fault and `git blame` can back me up on that	00:02
	→ Dreadnaught joined	00:02
xiong	:)	00:02
	← jnoon left	00:02
xiong	It's a hardware problem.	00:02
	→ johnkevinmbasco joined	00:02
	← gelardi_ left	00:03
	← LeShadow left	00:03
	← WaReZ left	00:03
	elliottcable → ELLIOTTCABLE	00:03
	← infynyxx left	00:04
	→ WaReZ joined	00:04
	→ skorgon joined	00:04
	→ CKyle joined	00:04
	← tang^ left	00:05
	← halu left	00:05
	→ jperry joined	00:07
	← xSmurf left	00:07
	← nZac left	00:10
	← skorgon left	00:10
	→ ben_alman joined	00:10
	→ ferreum joined	00:11
	→ scarabx joined	00:11
	← Kayra left	00:12
	→ MetalMatze joined	00:13
	→ halu joined	00:14
	→ john2x joined	00:15
	→ Nizumzen joined	00:15
	← ben_alman left	00:15
	→ mribeirodantas joined	00:16
	← blinkdesign left	00:17
	← boombatower left	00:17
	→ el3ctron_ joined	00:17
	← ferreum left	00:18
	← Exancillatus left	00:18
	← Norm_ left	00:21
	← primitive left	00:22
	← aubreypw_ left	00:22
	→ guampa_ joined	00:22
	→ keyrage joined	00:22
	→ aubreypwd joined	00:22
	guampa → Guest62410	00:23
	guampa_ → guampa	00:23
	← makubi left	00:23
	dermoth_ → dermoth\|home	00:24
	← Guest62410 left	00:24
	→ skorgon joined	00:25
	→ makubi joined	00:25
	→ alexlindenlevy joined	00:27
	← hobodave left	00:27
	→ hobodave joined	00:27
	← hobodave left	00:28
	→ hobodave joined	00:28
	→ githogori joined	00:28
	→ cads joined	00:29
	← lindenle left	00:29
	← SocalNick_ left	00:29
	← skorgon left	00:30
	→ SKYRAN joined	00:30
	← Fandekasp left	00:30
	→ EngierkO joined	00:30
	zz_karupanerura → karupanerura	00:31
	→ ambodi joined	00:33
	← standyro1 left	00:34
	→ dyng joined	00:34
	→ skorgon joined	00:34
	→ bhldev joined	00:34
	← SKYRAN left	00:34
	→ sohail joined	00:34
	← sohail left	00:34
	→ sohail joined	00:34
	→ PhearTheCeal joined	00:35
	← PhearTheCeal left	00:35
	→ PhearTheCeal joined	00:35
	← amcsi left	00:36
	← thiago left	00:36
	→ ojacobson joined	00:37
	← decotii left	00:37
	← madewokherd left	00:37
	[emmajane] → emmajane	00:37
	← ojacobson left	00:37
	→ acl_ joined	00:37
	← ambodi left	00:39
	→ mikecmpbll joined	00:39
	← skorgon left	00:39
	← hyperair left	00:41
	→ skorgon joined	00:44
	→ nathancahill_ joined	00:46
	← nathancahill left	00:47
	nathancahill_ → nathancahill	00:47
	← PhearTheCeal left	00:47
	→ thiago joined	00:47
	→ tmcmahon joined	00:47
	← Silmarilion left	00:47
	← mizu_no_oto left	00:48
	← skorgon left	00:49
	→ bwjt joined	00:51
	→ rpg joined	00:52
	→ rpg_ joined	00:52
	← MetalMatze left	00:53
bwjt	I did agit I'm confused by rebasing. git checkout mychange; git add .; git commit -m "blah"; git rebase master; fix the small conflict; git add .; git rebase --continue;	00:53
	→ skorgon joined	00:53
bwjt	Though when I do git rebase --continue, it just tells me to do git rebase --continue again	00:53
	# HEAD detached from 607e5fb # You are currently rebasing branch 'angulartimertesting' on '607e5fb'. # (all conflicts fixed: run "git rebase --continue")	00:54
	I thought 'git rebase --continue' would append these changes to my angulartimertesting ( 'mychange' )	00:55
	→ JerryT joined	00:55
	→ danecando joined	00:56
	← rpg left	00:56
	→ johnkevi_ joined	00:58
	← tmcmahon left	00:59
	← conan left	00:59
	← eli left	00:59
	bwjt notes it's oddly quiet in here for 1k+ people, this must be LT's botnet	00:59
	→ Sabbath joined	00:59
	← johnkevinmbasco left	01:00
	→ conan joined	01:00
	→ lindenle joined	01:00
	→ aberrant joined	01:00
	← skorgon left	01:01
	→ alif joined	01:02
	← thekkid left	01:03
	← mellernoia left	01:03
	← bwreilly left	01:03
	← asQuirre_ left	01:04
	→ kbouwser joined	01:04
	← becks___ left	01:04
	→ becks__ joined	01:05
	→ weiyang joined	01:05
	← rq601 left	01:07
	← Not_Whatever left	01:07
	← hellopat left	01:07
	← rafaqueque left	01:07
	← becks__ left	01:09
	← nnyk_ left	01:09
	← alexlindenlevy left	01:11
	→ eli joined	01:11
	← alif left	01:12
	emmajane → [emmajane]	01:13
	← starkhalo left	01:16
	← maestrojed left	01:17
	→ sykopomp_ joined	01:18
	→ alif joined	01:18
	← imachuchu left	01:19
	→ freimatz joined	01:19
sykopomp_	hi all. I'm trying to figure out what a good workflow could be for an open source project where there's a "stable" version that only receives backwards-compatible patches, and a "next major version" branch, which might be maintained for a longish time before actually being tagged, which would be used to preview new, backwards-incompatible changes, but would still need a backporting of all the maintenance patches	01:20
	→ alexlindenlevy joined	01:20
	→ nerdfiles joined	01:20
	← joelm left	01:20
sykopomp_	or if there's some workflow that achieves something similar that doesn't exactly involve those two branches. What have folks found to be manageable/successful? I'm used to the "gitflow" style of doing things, but it's not very friendly towards long-lived branches that only have some patches in common.	01:21
	← renatow left	01:21
	← redrocket left	01:21
	← Nizumzen left	01:21
	← aberrant left	01:22
	→ Kayra joined	01:23
	← cjz left	01:23
	← Diemuzi left	01:24
	→ livingstn joined	01:24
	← hahuang65 left	01:24
	← FatDarrel left	01:25
	← mikecmpbll left	01:25
	← alexlindenlevy left	01:25
	→ hyperair joined	01:26
	→ jade_ joined	01:27
	→ woollyams joined	01:27
	← Kayra left	01:27
	→ jday_ joined	01:27
	← bgerber left	01:28
	→ groglogic joined	01:29
	← jayallen left	01:30
	← cmpera left	01:31
	← SeptDePique left	01:32
	→ thekkid joined	01:33
	→ solker joined	01:34
	→ bwreilly joined	01:34
	← sohail left	01:35
	← panpainter left	01:35
	← bartman` left	01:36
	← mlimaloureiro left	01:36
	← githogori left	01:37
	← jperry left	01:37
	← danecando left	01:38
	← rpg_ left	01:38
	← solker left	01:38
	← dmarkham left	01:39
	→ bgerber joined	01:39
	← ColKurtz left	01:39
	← wolfman left	01:39
	→ aslaen joined	01:40
	← aslaen left	01:40
	→ Watabou joined	01:41
	→ wongon joined	01:42
	→ nnyk_ joined	01:43
	→ gelardi_ joined	01:46
	← Dreadnaught left	01:48
	← halu left	01:49
	→ kpreid joined	01:49
	→ buzzybron joined	01:51
buzzybron	how can i list the versions of a file?	01:51
	← gelardi_ left	01:51
	→ h0bbit_ joined	01:52
cjh`	`git log file` will show you every commit from the current branch that touches the file	01:52
	→ xjiujiu joined	01:52
	→ p4denis joined	01:52
	← CheckDavid left	01:52
	← zxd left	01:53
	→ ehmatthes joined	01:53
	→ nZac joined	01:53
	→ mitin001 joined	01:55
	→ jatoben joined	01:57
	← p4denis left	01:57
	← EngierkO left	01:57
	← mitin001 left	01:57
	← nZac left	01:58
	← psyl0n left	01:58
	← MacWinner left	02:00
dmarr	how are prs that are dependent on other prs usually handled? for example if merge an upstream pr and resolve conflicts, then the upstream pr changes	02:01
	← cads left	02:01
	→ Vampire0_ joined	02:02
	← gerard0 left	02:02
	← Vampire0 left	02:02
	← aubreypwd left	02:03
milki	the downstream pr should be updated either with merges or rebases	02:04
	→ githogori joined	02:04
	→ psyl0n joined	02:04
	← shruggar left	02:05
	← awakecoding left	02:05
	← v1n left	02:06
	← ErKa left	02:06
	→ gerard0 joined	02:06
	→ daito joined	02:07
	← teran left	02:07
	← Voting left	02:09
dmarr	how would a rebase work in that case?	02:09
	→ huahax joined	02:10
	→ derptacos joined	02:10
	← kbouwser left	02:10
	→ arizuka joined	02:11
	← xjiujiu left	02:12
	← derptacos left	02:12
	→ xjiujiu joined	02:12
	→ z8z joined	02:13
	← z8z left	02:14
	→ Exancillatus joined	02:14
	← vkrest left	02:14
	← h0bbit_ left	02:15
	→ vkrest joined	02:15
	← nlogn left	02:15
	→ cads joined	02:16
	← psyl0n left	02:18
	← Exancillatus left	02:19
	← kim_plausible left	02:20
	← LucasKA left	02:22
	← pheaver left	02:23
	→ shotgunblues joined	02:23
	→ acharlieh joined	02:24
	← xjiujiu left	02:25
	→ xjiujiu joined	02:25
	← shotgunblues left	02:26
	← alif left	02:26
	→ alexlindenlevy joined	02:26
	→ danecando joined	02:27
	→ solker joined	02:28
	← polyzen left	02:29
	← lazers left	02:29
	← xjiujiu left	02:31
	← alexlindenlevy left	02:31
	→ xjiujiu joined	02:31
	→ yacks joined	02:32
	← liamkeily left	02:33
	→ echo-area joined	02:33
	← solker left	02:33
	→ solker joined	02:33
	→ rouf_ joined	02:33
	← teasherm left	02:33
	→ theRealGentleman joined	02:35
	→ jgoss joined	02:35
	→ james_d_h joined	02:35
	← rouf left	02:35
	← bwjt left	02:35
	← pingupingu left	02:36
	← james_d_h left	02:37
	← insejn left	02:37
	← livingstn left	02:37
	← huahax left	02:37
	→ dobby156_ joined	02:37
	← dobby156 left	02:37
	← rouf_ left	02:38
	→ Goplat joined	02:38
	← solker left	02:38
	→ hellopat joined	02:39
	→ livingstn joined	02:39
	← xjiujiu left	02:39
	← bitkid left	02:39
	← dreiss left	02:39
	→ bitkid joined	02:40
	→ xjiujiu joined	02:40
	→ jbarker joined	02:40
	← jbarker left	02:40
	→ intripoon_ joined	02:40
	→ alif joined	02:41
	← acharlieh left	02:41
	← hellopat left	02:43
	← jade_ left	02:43
	← WaReZ left	02:43
	→ droid_ joined	02:43
droid_	hello, new to git, i got a problem	02:44
	← intripoon left	02:44
	→ fairuz joined	02:44
	← fairuz left	02:44
	→ fairuz joined	02:44
droid_	remote repository in github already existing and also my local codes but not in sync	02:45
	→ munderwo joined	02:45
droid_	how can i push my local changes into the remote project ?	02:45
	← droid_ left	02:46
	→ luc_ joined	02:47
Hello71	git push	02:47
	→ [1]JPeterson joined	02:48
	← xjiujiu left	02:48
	→ xjiujiu joined	02:48
	← munderwo left	02:49
	← JPeterson left	02:49
	→ axl_ joined	02:49
	→ alexlindenlevy joined	02:50
	← setmeaway left	02:50
	← githogori left	02:53
	← sykopomp_ left	02:53
	← mribeirodantas left	02:54
	→ wbdecker joined	02:54
	← xjiujiu left	02:55
	← yacks left	02:55
	← Walex2 left	02:56
	→ xjiujiu joined	02:56
	← xjiujiu left	02:56
	← ariscop left	02:56
	→ madewokherd joined	02:56
	→ yacks joined	02:57
	→ xjiujiu joined	02:57
	→ Walex joined	02:57
	→ WaReZ joined	02:58
	digifiv5e → guyz	02:58
	→ ariscop joined	02:58
	← axl_ left	02:59
	→ axl_ joined	02:59
	← alexlindenlevy left	03:00
	→ Vampire0 joined	03:00
	→ alexlindenlevy joined	03:00
	← lindenle left	03:02
	← Vampire0_ left	03:03
	→ aberrant joined	03:05
	← ehmatthes left	03:05
	← alif left	03:06
	→ phipes joined	03:07
	→ v1n joined	03:08
	→ alif joined	03:08
	→ pikaren joined	03:08
	← pikaren left	03:08
	→ pikaren joined	03:09
	← pikaren left	03:09
	→ pikaren joined	03:10
	← pikaren left	03:10
	← xjiujiu left	03:10
	→ pikaren joined	03:10
	→ Rex357 joined	03:10
	→ xjiujiu joined	03:11
Rex357	Do you guys lay into noobs?	03:11
	← ncthom91 left	03:11
Rex357	I'm having trouble figuring out what the address is for a repository	03:11
	← pikaren left	03:11
	→ LucasKA joined	03:11
Rex357	Googling not really breaking it down for me	03:11
	→ pikaren joined	03:12
	← elle left	03:12
	← pikaren left	03:12
	→ pikaren joined	03:12
	← pikaren left	03:13
	→ pikaren joined	03:13
	→ githogori joined	03:13
	→ eletuchy_ joined	03:14
	← eletuchy left	03:14
	← pikaren left	03:14
Rex357	I'm trying to figure out the address to git this: https://github.com/archlinuxarm/PKGBUILDs/tree/master/aur/cower	03:14
	→ m0ltar joined	03:14
	wbdecker → wbdecker\|away	03:14
	→ pikaren joined	03:14
	→ Exancillatus joined	03:14
	wbdecker\|away → wbdecker	03:14
	← pikaren left	03:15
	← dglambert left	03:15
	→ pikaren joined	03:15
	← pikaren left	03:15
	→ pikaren joined	03:16
	← xjiujiu left	03:16
	→ SocalNick joined	03:16
	→ xjiujiu joined	03:16
	← pikaren left	03:17
	← WaReZ left	03:17
	→ lindenle joined	03:17
	→ pikaren joined	03:17
	← pikaren left	03:18
	→ pikaren joined	03:18
	← JerryT left	03:18
	← pikaren left	03:18
	→ cmpera joined	03:19
	→ pikaren joined	03:19
	← Exancillatus left	03:19
	← githogori left	03:19
	← AaronMT left	03:19
	← pikaren left	03:19
	→ pikaren joined	03:20
	← pikaren left	03:20
	→ pikaren joined	03:21
	← pikaren left	03:21
	→ pikaren joined	03:22
	← pikaren left	03:22
fairuz	Rex357: [email@hidden.address]	03:22
	→ Starcraftmazter joined	03:23
Starcraftmazter	hi	03:23
	→ pikaren joined	03:23
Starcraftmazter	i wanted to ignore only a top level directory called media so i put "media/" in my gitignore, but it ignores every media directory anywhere	03:23
	← pikaren left	03:23
Starcraftmazter	how can i only make it ignore it at the very top levle?	03:23
	like ./media/ ?	03:23
	→ pikaren joined	03:23
	← jday_ left	03:24
fairuz	Starcraftmazter: I believe it's /media/*	03:24
	Starts with a slash	03:24
	← pikaren left	03:24
Starcraftmazter	but not a dot?	03:24
	→ mizu_no_oto joined	03:24
	→ pikaren joined	03:25
Rex357	fairuz: I only wanted one directory in there. Was there any way I could of done that?	03:25
	← pikaren left	03:25
	← xjiujiu left	03:25
	→ pikaren joined	03:25
fairuz	Starcraftmazter: A leading slash matches the beginning of the pathname. For example, "/*.c" matches "cat-file.c" but not "mozilla-sha1/sha1.c". (From the documentation)	03:25
	→ xjiujiu joined	03:26
	← pikaren left	03:26
fairuz	Rex357: YOu need to clone the project	03:26
	*you	03:26
	→ pikaren joined	03:26
	← pikaren left	03:27
Rex357	So I had to take the whole thing then?	03:27
	→ pikaren joined	03:27
	← pikaren left	03:28
fairuz	Rex357: Yes	03:28
	→ pikaren joined	03:28
	← pikaren left	03:28
	→ pikaren joined	03:29
	← pikaren left	03:29
	← el3ctron_ left	03:30
	→ pikaren joined	03:30
Rex357	fairuz: OK thanks. What is the -b thing for then? (Sorry was just looking at the help and I was trying to do that the whole time thinking I could just get the branch I was after)	03:30
	← pikaren left	03:30
	→ pikaren joined	03:31
xiong	git ignores nested repos. Is it possible to place the same dir under two repos?	03:31
	→ WaReZ joined	03:31
	← xjiujiu left	03:31
	← pikaren left	03:31
	← william1 left	03:31
	→ xjiujiu joined	03:32
	→ pikaren joined	03:32
offby1	I am not able rightly to apprehend the kind of confusion of ideas that could provoke such a question.	03:32
	← pikaren left	03:32
xiong	Ha ha, it's nasty, eh?	03:32
	← dyng left	03:32
	→ githogori joined	03:33
	→ dyng joined	03:33
	→ starkhalo joined	03:33
Starcraftmazter	i see, thanks fairuz	03:33
	← m0ltar left	03:34
	← dougsland left	03:34
xiong	Let's say I have two repos, one nested inside the other; however I have done this, I'm completely happy with that... to a point. Now I want to push the outside repo to a remote, disregarding the fact of the nested repo. Perhaps I'll have a third repo that disregards the first two.	03:34
	→ munderwo joined	03:34
xiong	In any case I want to make the project monolithic.	03:34
fairuz	Rex357: It's for pointing the HEAD to specified branch name	03:35
xiong	If you say 'no' then that's quite okay. Obviously I can copy both outside and inside project and trash their repos, leaving me with a monolith.	03:36
	← mr_red left	03:36
	← Devels left	03:37
	← dyng left	03:37
	← alvaro_o left	03:38
	→ Destos joined	03:38
	← GT4066 left	03:38
	← alexlindenlevy left	03:40
	← livingstn left	03:42
	→ nZac joined	03:42
Rex357	fairuz: OK. Thank you for your help. Have a good night.	03:42
	← Rex357 left	03:43
	→ JulianGindi joined	03:43
	← johnkevi_ left	03:44
	→ aubreypwd joined	03:45
	← nZac left	03:47
xiong	Ah. $ git add sub/ ..... will add sub/ as a subdir of the outer project, ignoring that it's in its own repo.	03:47
	← nedbat left	03:47
	← xjiujiu left	03:48
	→ xjiujiu joined	03:48
	→ mr_red joined	03:49
	← thiago left	03:49
	→ Devels joined	03:50
	← Devels left	03:51
	← groglogic left	03:52
	→ Nikoli joined	03:54
	→ zumba_addict joined	03:56
	← theRealGentleman left	03:56
	← danecando left	03:58
dmarr	git has put me through some tough times	03:58
	← aisbaa left	03:58
	→ aisbaa joined	03:58
xiong	Hope it's helped out out of more rough spots than it's put you in.	03:58
	→ jKlaus joined	03:59
jKlaus	hey guys	03:59
	→ Devels_ joined	03:59
	Devels_ → Devels	04:00
jKlaus	Is there a linux git client out there that will allow you to create repositories?	04:00
	← Devels left	04:00
	→ Devels joined	04:00
	→ pheaver joined	04:00
	← scarabx left	04:00
	← JulianGindi left	04:01
	← xjiujiu left	04:01
	→ xjiujiu joined	04:01
dmarr	i have this state where all these rebased commits are in the history of a branch. i have a dev branch that i'd like to just apply the changes to in one commit. there have been conflicts along the way but i cant seem to get things to work :(	04:02
starkhalo	jKlaus: all clients can create local repositories	04:02
	← kermit left	04:03
	→ craigp joined	04:03
jKlaus	starkhalo, I want to be able to create the remote repository	04:03
	→ kermit joined	04:03
jKlaus	.. so I don't have to manually remote into the server and create the repo	04:03
	← craigp left	04:03
	→ mdub joined	04:03
starkhalo	jKlaus: something along http://hub.github.com/ for github?	04:03
	→ dyng joined	04:04
starkhalo	I'm sure there are others for other services	04:04
	← woollyams left	04:04
jKlaus	I'm looking for a graphical UI though	04:04
	trying to find something so I might be able to talk my company into migrating to git rather than M$'s TFS	04:05
	→ saurabh joined	04:05
	→ ewnd9 joined	04:05
offby1	they're gonna base their decision on the presence or absence of a GUI ?	04:05
	tsk tsk tsk	04:05
	saurabh → Guest60494	04:05
jKlaus	offby1.. did I mention the bulk of them are sharepoint developers?	04:06
starkhalo	jKlaus: then check out https://github.com/gitlabhq/gitlabhq, perhaps that'll help you	04:06
jKlaus	if you want to call them that	04:06
	→ dreiss joined	04:07
	← xjiujiu left	04:07
	→ xjiujiu joined	04:08
jKlaus	starkhalo, that is actually interesting, thanks	04:08
	← dyng left	04:09
	→ scarabx joined	04:09
starkhalo	jKlaus: np	04:09
	→ fire joined	04:09
	← mdub left	04:11
	← chrisshattuck left	04:12
	→ woollyams joined	04:12
	→ sohail joined	04:13
	← sohail left	04:13
	→ sohail joined	04:13
	→ chrisshattuck joined	04:13
	→ dik_dak joined	04:15
	→ Exancillatus joined	04:15
	← xjiujiu left	04:17
	→ xjiujiu joined	04:17
	→ ethlor joined	04:18
	→ gittish joined	04:18
	← Exancillatus left	04:20
	← daito left	04:21
	→ MattStratton joined	04:21
	→ hellopat joined	04:21
	← friesk left	04:22
	→ craigp joined	04:22
	→ subhojit777 joined	04:24
relipse	why is git stuck on this: Unpacking objects: 100% (9/9), done	04:24
SamB	is that really the last thing it shows?	04:25
relipse	yes then i ctrl+c'd it and did git pull again and now it is stuck on Updating 3255052..7080a02	04:25
	really odd	04:26
	oh now it finished, oh well	04:26
	← hellopat left	04:26
	← Dougie187 left	04:26
	← Khisanth left	04:26
	← craigp left	04:27
	→ s4nchez joined	04:27
	→ arvind_khadri joined	04:27
	← arvind_khadri left	04:28
	← woollyams left	04:28
jKlaus	anyone have a clue as to why my laptop resolves to my server given the domain I setup via DDNS but my desktop does not?	04:29
	wbdecker → wbdecker\|away	04:29
milki	spam?	04:29
relipse	add to hosts file jKlaus l	04:30
	on your desktop	04:30
	← SocalNick left	04:30
relipse	wait nevermind i have no idea what i'm talking about	04:30
	→ Kayra joined	04:30
jKlaus	I was going to say..? lol	04:30
	wbdecker\|away → wbdecker	04:30
jKlaus	hmm	04:30
	strange.. chrome couldn't resolve it.. I accessed it via fire fox and now chrome can	04:31
	lol	04:31
	← s4nchez left	04:31
jKlaus	man I really like gitlab..	04:31
	seems like extreme overkill to set it up if they don't opt to use it but.. I like it	04:32
	→ thiago joined	04:32
	→ SocalNick joined	04:32
	wbdecker → wbdecker\|away	04:32
	→ kcj joined	04:33
	← xjiujiu left	04:33
	→ xjiujiu joined	04:34
	→ solker joined	04:34
	← axl_ left	04:35
	→ jayrobot joined	04:35
	wbdecker\|away → wbdecker	04:35
	→ pawprint joined	04:37
	← thrakcattack left	04:37
pawprint	how can i ask "git diff" to show me more lines of context?	04:37
sunny256	pawprint: The -U option	04:37
pawprint	thank you	04:37
sunny256	git diff -U 20	04:37
	wbdecker → wbdecker\|away	04:37
sunny256	np :)	04:38
	← yhager left	04:38
	← kukks left	04:38
	← aubreypwd left	04:38
	← Noldorin left	04:38
	→ aubreypwd joined	04:39
	→ woollyams joined	04:39
	→ thrakcattack joined	04:39
	← solker left	04:39
	← pawprint left	04:39
	→ Khisanth joined	04:39
	← WaReZ left	04:39
	→ yhager joined	04:40
	→ wolfman joined	04:42
	← xjiujiu left	04:42
	Guest41679 → rastus_vernon	04:43
	← rastus_vernon left	04:43
	→ rastus_vernon joined	04:43
	→ xjiujiu joined	04:43
	wbdecker\|away → wbdecker	04:43
	wbdecker → wbdecker\|away	04:45
	→ cenuij joined	04:46
	← cenuij left	04:46
	→ cenuij joined	04:46
	wbdecker\|away → wbdecker	04:46
	→ alester joined	04:46
	← yhager left	04:46
	→ DolphinDream joined	04:47
	← aubreypwd left	04:47
	→ bl33 joined	04:47
	← LucasKA left	04:48
	→ aubreypwd joined	04:49
	← fire left	04:49
	→ fire joined	04:50
	← xjiujiu left	04:50
	→ xjiujiu joined	04:51
	← danielcompton left	04:52
	→ karstensrage joined	04:53
	← Destos left	04:54
	→ h0bbit_ joined	04:55
	→ WaReZ joined	04:55
	← woollyams left	04:55
	→ beautifulmind joined	04:56
	← bwreilly left	04:56
	← takamichi left	04:57
	→ koda joined	04:57
	← Kayra left	04:57
	← ariscop left	04:58
	→ dyng joined	04:58
	→ takamichi joined	04:58
	← fire left	04:59
	← xjiujiu left	04:59
	← aberrant left	05:00
	← donmichelangelo left	05:00
	← urn left	05:00
	→ donmichelangelo joined	05:00
	← koda left	05:01
	→ woollyams joined	05:02
	← ethlor left	05:03
DolphinDream	what does SCM stand for ?	05:05
	source control managmenet?	05:05
SamB	!SCM	05:05
	!scm	05:05
Starcraftmazter	it stands for me name	05:05
DolphinDream	:)	05:05
Starcraftmazter	obviously	05:05
DolphinDream	nice try SamB	05:05
	ok. kool	05:05
SamB	how about "scheme"	05:05
DolphinDream	scheme? hm.. u sure?	05:06
SamB	probably not what you were looking for, but .scm is a common extension for scheme programs ...	05:06
	← woollyams left	05:08
	← alif left	05:08
	← alester left	05:09
	→ nafg_ joined	05:09
	← nafg__ left	05:10
	→ daito joined	05:11
	← daito left	05:12
	← ianmcorvidae left	05:13
	→ ianmcorvidae joined	05:13
	→ daito joined	05:13
	→ ahmedelgabri joined	05:14
	← scarabx left	05:14
	→ urn joined	05:14
	← CKyle left	05:15
	→ Exancillatus joined	05:16
	→ johnkevinmbasco joined	05:17
	→ koda joined	05:19
	→ ariscop joined	05:19
	← Exancillatus left	05:20
	→ Tommy[D] joined	05:21
	← vkrest left	05:23
	→ cbot joined	05:23
	→ marthinal joined	05:23
	← Tommy[D]_ left	05:24
	← mizu_no_oto left	05:24
	→ alezost joined	05:24
	→ Fandekasp joined	05:25
	→ LucasKA joined	05:26
	← nicedice left	05:27
	← bgerber left	05:27
	→ bgerber joined	05:28
	→ solker joined	05:28
	← wongon left	05:30
	→ craigp joined	05:31
	→ FatDarrel joined	05:32
	← solker left	05:33
	← hobodave left	05:34
	→ solker joined	05:34
	← SocalNick left	05:34
	← thekkid left	05:35
	← luxgeek left	05:35
	→ ClientAlive joined	05:35
	← sohail left	05:35
ClientAlive	stupid question... is it possible to branch a branch?	05:35
SamB	yes	05:36
Eugene	ClientAlive - you can branch anything!	05:36
	It's just a pointer	05:36
ClientAlive	cool	05:36
	thx man	05:36
	→ pepper_chico joined	05:37
	← solker left	05:38
	← chrisshattuck left	05:42
	→ teran joined	05:44
	→ Nizumzen joined	05:46
	← cbot left	05:47
	→ shingshang joined	05:47
	← madewokherd left	05:50
	→ mmcdaris joined	05:50
	← mmcdaris left	05:50
	← pepper_chico left	05:50
	← bl33 left	05:53
	← cads left	05:55
	→ bl33 joined	05:55
	→ yhager joined	05:58
	→ woollyams joined	05:58
	→ irqq joined	05:59
	← Nizumzen left	05:59
	→ vkrest joined	06:01
	← Freeman42 left	06:02
	→ Freeman42 joined	06:02
	← tpollard left	06:03
	← v1n left	06:04
	→ Raging_Hog joined	06:05
	→ ehmatthes joined	06:05
	← Walex left	06:07
	→ Walex joined	06:07
	← ewnd9 left	06:08
	← D9 left	06:13
	← teran left	06:14
	→ Nizumzen joined	06:15
	→ Exancillatus joined	06:17
DolphinDream	ClientAlive: branch a branch ? :) that's why it's called a branch ;p	06:18
ClientAlive	har har	06:18
DolphinDream	master is a branch.. from which you branch other branches ;0	06:18
	← jKlaus left	06:18
	← cehteh left	06:18
	← freestyl3r left	06:20
	← Exancillatus left	06:22
	← Sabbath left	06:24
xiong	What is this mode 160000	06:24
	?	06:24
	→ Sabbath joined	06:25
	→ cehteh joined	06:27
	← hike left	06:27
	→ bwreilly joined	06:28
	← midnightmagic left	06:28
ClientAlive	I've been reading this book and now I'm starting to form a thought lol	06:28
j416	xiong: submodule, likely	06:28
ClientAlive	I have this graaaand idea how to do a repository workflow...	06:28
xiong	Not only that, ClientAlive; it's highly recommended. I don't like the word 'master' itself; it means too many things. I call the branch suitable for ordinary user download 'trunk'. Then an integration branch is 'devel'; and from that, perhaps, 'feature-foo' branches.	06:28
	← Nizumzen left	06:29
j416	xiong: it happens when git finds a .git directory inside a directory you are committing	06:29
xiong	j416, it is indeed a submodule. I'd like to know what kind of a mode that is -- no read, write, or execute for user, group, or all??	06:29
	← lindenle left	06:29
	← cehteh left	06:29
j416	xiong: it doesn't have to be a submodule, but if it isn't, you probably are doing something weird	06:29
ClientAlive	xiong: Yes, I think I get it.	06:29
	→ midnightmagic joined	06:29
j416	xiong: not sure, special git-mode afaik	06:29
	← tabeaux left	06:30
xiong	Is it a fake mode, perhaps? Because the thing pointed to is a directory, which generally has plenty of permissions.	06:30
ClientAlive	It's late here. Catch ya guys another time	06:31
	j416 is off to work o/	06:31
	← ClientAlive left	06:31
j416	xiong: it didn't point to a directory, it pointed to a commit	06:31
	j416 off to work fr	06:31
	← the_cybersphinx left	06:31
	← donmichelangelo left	06:32
	→ donmichelangelo joined	06:32
xiong	There's some confusion here, if only in my mind.	06:32
	← takamichi left	06:33
	→ hike joined	06:34
xiong	I need to understand submodules better; I've been reading on them for two weeks. Some authors hate them, other love them. Some say they are broken, others just say, Don't forget to XYZ.	06:35
	← FatDarrel left	06:35
	→ ThomasLocke joined	06:36
	← ahmedelgabri left	06:36
	← wbdecker left	06:37
xiong	I don't think I understand the complaints -- I understand what they're saying but not why the complaints are any more substantial than "You have to do more work and if you forget a step it's bad."	06:38
	I agree there's more work involved, more decisions to be made; which is why I'm working on a script and some hooks to smooth that along. But the same could be said for nearly any complex task -- once you figure out how you want to do it, write a script.	06:39
	→ abrkn joined	06:41
	→ eslwork joined	06:41
	← vkrest left	06:41
	← daito left	06:42
	→ vkrest joined	06:42
	← glebihan left	06:43
	→ ewet joined	06:44
	← ewet left	06:44
	→ ewet joined	06:44
	→ ambodi joined	06:45
	→ StarFire joined	06:45
	← ewet left	06:46
moritz	xiong: one thing that's constantly annoying is that so few git commands are aware of submodules	06:47
	want to create a tar ball? sorry, git-archive doesn't understand submodules	06:47
xiong	Agree, moritz. Hence the wrapper script and hooks.	06:47
	→ hoschi joined	06:47
moritz	wnat to work around that by creating a file list with git ls-files? sorry, git ls-files doesn't know about submodules	06:47
	and the list goes on	06:48
xiong	Well, I'd actually take that list pretty seriously if you want to link it.	06:48
	I'm determined to beat these issues down.	06:48
moritz	by patching the git commandsß	06:48
	s/ß/?/	06:48
	→ pikaren joined	06:49
	→ glebihan joined	06:49
	Starcraftmazter → SCM\|NotHere	06:49
xiong	How should it be done?	06:49
	← bl33 left	06:50
xiong	I'd rather not tamper with git itself; hence the wrapper -- a partial wrapper; I'd like to let the user use git normally whenever possible.	06:50
	→ intellix joined	06:51
xiong	I particularly plan to put in a pre-commit hook so that submodules are scanned. All subs must be clean and pushed before committing the super.	06:51
	→ jargon joined	06:52
	→ mishok13 joined	06:52
xiong	Um, I award you bonus points for using a letter (I think) even the Germans have dropped. ;)	06:53
	← Gamecubic left	06:53
	← Mamoru left	06:54
	← dreiss left	06:54
xiong	If you have suggestions on the fixes, naturally I'll want to hear them. But what I want most is rational and specific complaints -- stuff needing fixing, such as the git-archive and git-ls-files you mentioned, moritz.	06:55
	→ ewet joined	06:56
	← woollyams left	06:56
	← EisNerd left	06:57
	→ a-l-e joined	06:57
	→ kbart joined	06:58
	← misterpink left	06:58
	→ woollyams joined	06:59
	→ nilg joined	07:00
	→ jayallen joined	07:00
	→ jaseemabid joined	07:00
	← jaseemabid left	07:01
	← yacks left	07:01
	← koda left	07:01
	← Tatsh left	07:01
	→ jaseemabid joined	07:02
	→ ewnd9 joined	07:02
	← ThomasLocke left	07:03
	→ ThomasLocke joined	07:03
moritz	xiong: but why not patch git itself?	07:04
	xiong: it would be the most direct solution with the most benefit for every user	07:04
	← woollyams left	07:04
xiong	Well, I already worry about my users. Will they be willing to download and install wrapper and hooks?	07:04
	You are probably thinking in bigger terms; my goals are more humble. For me, this is a special-purpose tool. You are thinking about the general-purpose fix to submodules -- submodules done right. I don't think I have the experience or resources to do that.	07:05
moritz	which is why patching it seems a much more logical choice	07:05
	fwiw I'm not proposing a general re-think of submodules, just wider support in core	07:06
xiong	Well, let me turn it around: People have been complaining about submodules for years. Why has nobody else heeded the complaints and reworked submodules closer to the heart's desire?	07:06
	→ mikar joined	07:07
hyperair	because it's controversial?	07:07
xiong	Nobody, that is, with much more ability and cred than I have....	07:07
	?	07:07
moritz	there are other approaches to solve the same problem that submodules tries to solve	07:07
	or some of the problems	07:07
xiong	Controversial in that some don't like?	07:07
moritz	!sub	07:07
gitinfo	[!subtree] The subtree merge method is great for incorporating a subsidiary git repo into your current one with "unified" history. Read http://git-scm.com/book/en/Git-Tools-Subtree-Merging for more info, or try one of the !subtree_alternatives	07:07
	← yhager left	07:08
moritz	for example	07:08
	!subproj	07:08
gitinfo	[!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.	07:08
xiong	I looked hard at alternatives and did not like at all, thanks.	07:08
moritz	xiong: but it answers your "why did noone tried to fix them?" question with "several people did"	07:08
xiong	I took about a week to seriously consider either subtrees or gitslave.	07:08
	← gittish left	07:09
	← starkhalo left	07:09
xiong	Oh, sorry; I see this through other lenses. Both of those handle the issue of subprojects in different basic ways from submodules. I thought we were talking about improving the submodules feature directly.	07:09
moritz	well, there have been improvements	07:10
	for example git clone --recursive	07:10
	which hasn't been there from the start	07:10
	it just looks to me that submodule development is resource starved	07:10
	though I'm not on the !list, so I could be wrong	07:11
gitinfo	[!mailing_list] The mailing list can be reached via [email@hidden.address] You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist	07:11
xiong	Exactly. Update git to a recent version and some objections are already answered. One difficulty with my kind of catch-up research is that I wallow in objections ten years old.	07:11
	← acl_ left	07:11
	→ daito joined	07:11
xiong	I'm probably the wrong guy, moritz; although it's flattering even to hear the suggestion. My C is so rusty it's peeling off in orange flakes.	07:12
	← jaseemabid left	07:15
	→ cbreak-work joined	07:15
	→ simesy joined	07:16
	← chepre left	07:17
	→ roelmonnens joined	07:18
	→ dreiss joined	07:18
	→ h_kon joined	07:20
	→ longbeach joined	07:20
	→ ahmedelgabri joined	07:20
	→ xlinkz0 joined	07:21
	→ yhager joined	07:21
	→ Nizumzen joined	07:22
	→ bozek joined	07:22
	← m0viefreak left	07:22
	← ewet left	07:24
	→ corgankidd joined	07:24
	→ JohnBat26 joined	07:24
	← ambodi left	07:26
	gitinfo set mode: +v	07:26
	→ fire joined	07:26
	→ user258467 joined	07:26
	→ jaseemabid joined	07:27
	→ danielcompton joined	07:28
	→ arvind_khadri joined	07:28
	← bozek left	07:28
	→ FatDarrel joined	07:28
	← mishok13 left	07:30
	→ papegaaij joined	07:30
	→ rouf joined	07:30
	← craigp left	07:31
	→ ewet joined	07:31
	← ewet left	07:31
	→ skorgon joined	07:33
	← fire left	07:33
	→ ambodi joined	07:33
	→ thevishy joined	07:33
	→ macben joined	07:34
	← nilg left	07:34
	→ solker joined	07:34
	→ donnib joined	07:36
donnib	Hi	07:36
	i am doing a git svn fetch every day to have my git repo in sync with the svn but today i got an error and i don't know how to get back on track	07:37
	i get : Incomplete data: Delta source ended unexpectedly at /usr/lib/perl5/site_perl/Git/SVN/Ra.pm line 290	07:37
	can somebody help me ?	07:37
	← corgankidd left	07:38
_ikke_	donnib: not a usual error	07:38
	← danielcompton left	07:38
_ikke_	It's an svn error aparently	07:38
donnib	i tried searching online but have not been able to figure out what is causing this	07:38
_ikke_	https://stackoverflow.com/questions/13137473/tortoise-svn-errordelta-source-ended-unexpectedly	07:38
	https://stackoverflow.com/questions/211025/git-svn-rebase-incomplete-data-delta-source-ended-unexpectedly	07:39
	← evilbug left	07:39
	→ dobalina joined	07:39
	← solker left	07:39
donnib	_ikke_: i just don't know where to start, the article says top level delete of my local copy but where is that ? What is safe to delete ?	07:40
_ikke_	My knowledge about this is fairly limited, so I can't really help	07:40
	→ ewet joined	07:40
donnib	I shold add that that i am in middle of a svn -> git migration and i do this git svn fetch until we are ready so i don't have a copy of the repo (git)	07:40
	← phipes left	07:41
donnib	oh ok, np maybe somebody else can help	07:41
	← aubreypwd left	07:41
	← simesy left	07:41
_ikke_	It's a bit quiet currently	07:41
	→ fRAiLtY- joined	07:42
donnib	:(	07:42
	vocifer-away → vocifer	07:42
	→ nilg joined	07:43
	← Goplat left	07:44
	→ hahuang65 joined	07:44
	← thierryp left	07:45
	→ rob_w joined	07:45
vocifer	sorry, don't have any experience with git svn, I assume svn is still working properly with the repository?	07:46
	→ thierryp_ joined	07:46
donnib	vocifer: yup i just did an update on my repo with svn up and it worked fine	07:46
ewet	so, please help me read this:	07:46
	→ koda joined	07:46
ewet	When the .gitattributes file is missing from the work tree, the path in the index is used as a fall-back. During checkout process, .gitattributes in the index is used and then the file in the working tree is used as a fall-back.	07:47
	at what time would there be a .gitattributes in the index but not in the work tree?	07:47
vocifer	ewet, `git add .gitattributes` `rm .gitattributes`	07:48
	→ takamichi joined	07:48
	← macben left	07:49
	← MissionCritical left	07:50
	→ HelperW joined	07:50
	→ zxd joined	07:50
ewet	vocifer: of course ...	07:50
	← thierryp_ left	07:50
	→ chepre joined	07:51
	→ jnavila joined	07:51
	← beautifulmind left	07:53
	→ eijk_ joined	07:53
vocifer	donnib, is a fresh `git svn clone` successful? (just curious)	07:53
	→ mishok13 joined	07:54
	→ olle joined	07:54
	← olle left	07:54
	→ olle joined	07:55
	→ jnavila_ joined	07:56
	← jnavila left	07:56
	→ introom joined	07:56
introom	hi. I want to only checkout a portion files	07:56
	sorry, my plan is to not checkout several files, while checking the remaining	07:57
	← Bass10 left	07:58
	→ matanya joined	07:58
	→ Mission-Critical joined	07:58
moritz	xiong: much of git is written in shell, not C; also some perl	07:58
	→ hellopat joined	07:58
	→ rendar joined	07:59
	→ s4nchez joined	07:59
	← arizuka left	07:59
xiong	Well, Perl for me.	08:00
	I'd better stick to my special project. Perhaps once I'm done with that I'll have some ideas how to generalize it.	08:00
	← ahmedelgabri left	08:00
	→ arizuka_ joined	08:01
	→ ahmedelg_ joined	08:01
	← zxd left	08:01
	→ zxd_ joined	08:01
_ikke_	introom: The only way it to checkout each file you want seperately	08:01
	introom: git checkout <commit> <file>	08:01
	→ teran joined	08:01
vocifer	moritz, 45.2% is c 34.7% is shell	08:01
introom	_ikke_: there is only one file I don't wanna checkout	08:01
vocifer	moritz, but i think they do write a lot of test code in sh scripts as well ... so	08:02
_ikke_	introom: If it has uncomitted changes, the easiest way is to temporarily move it away, check out the commit, and then put the file back	08:02
	← hellopat left	08:03
	← ahmedelg_ left	08:04
	→ simesy joined	08:05
	← thiago left	08:05
	→ shruggar joined	08:06
	← tadasZ left	08:07
	→ lb1a joined	08:09
	→ tvw joined	08:10
	→ TomyWork joined	08:10
	← arvind_khadri left	08:10
	← s4nchez left	08:11
	→ puppeh joined	08:11
puppeh	I found the following in a .gitattribute file: "* text=auto !eol"	08:12
gitinfo	The option/attribute to use when you need to care about different line-endings in your file: http://timclem.wordpress.com/2012/03/01/mind-the-end-of-your-line/	08:12
puppeh	what does it mean?	08:12
	→ Anticom joined	08:12
	← dreiss left	08:12
	← dik_dak left	08:12
	Mission-Critical → MissionCritical	08:13
_ikke_	puppeh: see man gitattributes	08:13
gitinfo	puppeh: the gitattributes manpage is available at http://jk.gs/gitattributes.html	08:13
_ikke_	puppeh: basicaly, it has to do with how git treats the line-endings of files	08:13
	→ daedeloth joined	08:13
	← hahuang65 left	08:13
daedeloth	so, long time subversion user, moving to git.	08:13
	← nnyk_ left	08:13
daedeloth	I have a project that consists of 2 parts, a website and an api.	08:14
	in git terminology, I think I should use 2 repositories for that	08:14
_ikke_	dean\|away: correct	08:14
	→ vurma joined	08:14
	→ gittish joined	08:14
daedeloth	but the website also has a "build" folder, which is basically the minified version of the website	08:15
	and a "builder" folder, containing the script to do the actual minifying	08:15
	← freimatz left	08:15
daedeloth	however, I think it would be cleanest if the "website" repository would simply contain the website files, not the buildscripts	08:15
moritz	why?	08:15
	← przemkovv left	08:15
moritz	is the build script specific to the website?	08:16
daedeloth	yes	08:16
	→ przemkovv joined	08:16
moritz	then it's probably simplest to leave it in the website repo	08:16
daedeloth	but in git, you can't just clone part of the project, no?	08:16
	← simesy left	08:16
moritz	no	08:16
	but why would you want to?	08:16
daedeloth	like, I'm a developer, I have set up /var/www/website-part, I want to put the website there	08:16
moritz	if the builder is specific to the website, you never want the builder without the website	08:16
	← h0bbit_ left	08:16
daedeloth	indeed	08:17
puppeh	_ikke_: yes but what does the "!" symbol in front of eol means	08:17
	→ CharlesFinley joined	08:17
daedeloth	but as a developer, in subversion I just just checkout repostiry-url.com/website and I'd end up with /var/www/website-part/index.html	08:17
	→ solker joined	08:17
daedeloth	in git, I can't do that, so I'd end up with /var/www/website-aprt/website/index.html	08:18
	right?	08:18
moritz	yes	08:18
	so move your document root	08:18
	and all is fine	08:18
	→ Exancillatus joined	08:18
daedeloth	and put the buildscript and build folder in the website root, yea, could work	08:18
moritz	or have the repository in your home, and use the builder to create /var/www/website-part/ from it	08:18
	→ h0bbit_ joined	08:19
daedeloth	well the builder is not for development, in development the regular source is used	08:19
moritz	ok	08:19
arand__	Or write an appropriate deploy hook that only checks out a subdirectory.	08:19
	← cenuij left	08:19
	← ambodi left	08:20
_ikke_	puppeh: According the documentation, it means leave it unspecified	08:20
daedeloth	I'll just put everything in the root then	08:20
_ikke_	puppeh: "Sometimes you would need to override an setting of an attribute for a path to Unspecified state. This can be done by listing the name of the attribute prefixed with an exclamation point !."	08:20
	→ ambodi joined	08:20
	← Freeman42 left	08:20
daedeloth	ok, so next question :) right now I have one subversion repository containing both parts. I'm now using git-svn and this tutorial to convert it to one git repository	08:21
	http://john.albin.net/git/convert-subversion-to-git	08:21
	→ harobed joined	08:21
	→ Xiti` joined	08:21
	→ Freeman42 joined	08:21
_ikke_	puppeh: It basically disables eol	08:21
daedeloth	then I'm going to push it to 2 remote locations and in both repository remove half of the files	08:21
	is this extremely stupid?	08:21
	or just regular stupid	08:21
moritz	just regular stupid	08:22
_ikke_	daedeloth: are the different parts self contained in a single directory or something?	08:22
	← neal__ left	08:22
daedeloth	yes	08:22
	← joki left	08:22
	← ribasushi left	08:22
	← bipolar left	08:22
moritz	there once was an svn repo from which I extracted about a half dozen git repositories	08:22
arand__	daedeloth: if you have directory boundries you could split it from the beginning.	08:22
daedeloth	yes but each git repository will contain the history of the whole project	08:23
	→ LionKing joined	08:23
_ikke_	daedeloth: You can use filter-branch --subdirectory-filter to split them out	08:23
	or what arand says indeed	08:23
	← Exancillatus left	08:23
_ikke_	Just clone the svn project from the subdirectory	08:23
daedeloth	arand__, yes, but then I lose my tags and branches	08:23
_ikke_	yeah, true	08:23
	← andres left	08:24
	→ andres joined	08:24
	← potatoe left	08:24
	← gunmuc left	08:24
	→ gunmuc joined	08:24
	← Xiti left	08:24
	→ potatoe joined	08:24
	→ lavong joined	08:24
	← ambodi left	08:25
	← Cork left	08:25
	→ ribasushi joined	08:25
	→ aard_ joined	08:25
	→ neal__ joined	08:26
	← ariscop left	08:26
	← StarFire left	08:26
	→ bipolar joined	08:26
	← TomyWork left	08:26
	← skorgon left	08:27
	→ TomyWork joined	08:27
	zz_frobware → frobware	08:27
	→ noname001__ joined	08:28
	→ Cork joined	08:28
	→ joki joined	08:28
	← puppeh left	08:28
	zxd_ → zxd	08:29
	→ vince_prignano joined	08:29
	← kadoban left	08:29
	← ehmatthes left	08:29
	← nathancahill left	08:30
	← kossy left	08:31
	→ kossy joined	08:31
	→ poulson joined	08:32
	→ aChatir joined	08:33
	→ alorence joined	08:34
	→ juril joined	08:35
	← thevishy left	08:35
	→ danicampa901 joined	08:35
	← bernalex left	08:36
	→ bernalex joined	08:36
	→ Exancillatus joined	08:36
	← johnkevinmbasco left	08:37
	→ crocket joined	08:38
	→ rethus joined	08:38
	→ blinkdesign joined	08:38
crocket	My coworker doesn't know how to resolve git conflicts and refuses to learn it.	08:38
	Mediocrity rules my workplace.	08:38
	← blinkdesign left	08:38
grawity	And still has commit rights?	08:39
lb1a	crocket, slap him with a spoon	08:39
crocket	To them, anything that they need to take time to learn feels too academic.	08:39
lb1a	crocket, http://www.youtube.com/watch?v=9VDvgL58h_Y	08:39
	→ rst0 joined	08:40
crocket	He deals with git repositories everyday, but he refuses to learn how to resolve git conflicts.	08:40
_ikke_	What does he do when encountering them?	08:41
	→ s4nchez joined	08:41
	← aChatir left	08:41
	← Khisanth left	08:41
	→ aChatir joined	08:42
crocket	_ikke_, He asks help from others who know git well.	08:42
	→ Khisanth joined	08:42
crocket	_ikke_, He asks others to resolve git conflicts.	08:42
	→ vince_pr_ joined	08:42
_ikke_	s/git conflicts/merge conflicts	08:43
	And as long as you keep helping him, he won't ever bother	08:43
	← FatDarrel left	08:43
crocket	_ikke_, He doensn't ask my help.	08:43
moritz	<@Zefram> there's a distinct moral hazard in pandering to people who don't cooperate	08:44
_ikke_	you, plural	08:44
crocket	ok	08:44
moritz	seen yesterday on #p5p	08:44
crocket	You didn't tell me "you" were plural.	08:44
_ikke_	moritz: What does pandering mean?	08:44
	← akiva left	08:44
_ikke_	google translate isn't helping	08:44
	or does it mean something like connect	08:45
	← vince_prignano left	08:45
moritz	_ikke_: to favor someone, to adapot to their way of doing things	08:45
daedeloth	ok, so, now I have: local git with both parts and a working copy of that local git with a commit to remove half of the code	08:45
	now I want to push this commit to a new remote repository, right?	08:45
	and hten remove the working copy and do the same for the other part?	08:46
	← s4nchez left	08:46
_ikke_	daedeloth: Have you looked into git filter-branch --subdirectory filter?	08:46
	← Novacha left	08:46
	← rst0 left	08:46
	→ simesy joined	08:46
_ikke_	--subdirectory-filter	08:46
daedeloth	ah, no, I figured I'd just move & remove...	08:47
lb1a	if it does the job ....	08:47
daedeloth	hm but that filter branch is actually going to remove the history of the second part?	08:47
	well, not remove... filter out?	08:47
	→ Novacha joined	08:48
lb1a	yes	08:48
	→ pretty_function joined	08:48
_ikke_	daedeloth: it recreates history with the subdirectory as root	08:48
daedeloth	that's way better.	08:48
	do I do that in a working copy?	08:48
_ikke_	yes	08:48
	→ akiva joined	08:49
	→ arvind_khadri joined	08:49
	← rue left	08:50
	← arand left	08:50
	← aChatir left	08:52
	→ ahmedelgabri joined	08:52
	← shruggar left	08:52
	← vince_pr_ left	08:53
	→ _iron joined	08:53
ewet	so, technically git applies the gitattributes clean filter on git-add, right? it wouldn't make sense to do this on commit ...	08:53
	→ ArSn joined	08:53
grawity	right	08:53
	→ vince_prignano joined	08:54
grawity	commit just makes a tree out of the exact same things that are in the index/cache	08:54
	h_kon → hkon_	08:54
ewet	yeah	08:54
	← simesy left	08:55
	→ ghostlines joined	08:55
	← ghostlines left	08:55
_ikke_	Those filters work between the index and the working tree	08:55
	→ aChatir joined	08:55
	→ ghostlines joined	08:55
	← ghostlines left	08:56
	← yuwang^^ left	08:56
	→ konadr joined	08:56
daedeloth	ok, I did that subtree thing and I'm pushing now, but I'm not sure what is happening to the tags and branches now	08:56
	→ rue joined	08:56
	← BBonifield left	08:56
daedeloth	and since my collegues are breathing down my neck, wanting to commit all kinds of stuff... :)	08:57
	→ ghostlines joined	08:57
	→ armin_w joined	08:57
	→ BBonifield joined	08:57
	→ stodan joined	08:58
daedeloth	or, apparently all tags and branches are gone?	08:58
	→ p4denis joined	08:59
	← Fandekasp left	08:59
	→ Fandekasp joined	08:59
	→ s4nchez joined	08:59
	→ macben joined	09:00
	← ahmedelgabri left	09:01
ewet	_ikke_: I know. Here the point in time was of importance to me.	09:02
	→ ahmedelgabri joined	09:03
_ikke_	ewet: Yes, but knowing where the filters apply say something about when they are applied	09:03
	← s4nchez left	09:04
ewet	_ikke_: that's why I assumed it wouldn't make sense on commit	09:04
_ikke_	And my comment was confirming that ;-)	09:05
	→ s4nchez joined	09:05
	← macben left	09:05
ewet	but I see, I skipped that in the man page, it's right there ;)	09:05
	← Khisanth left	09:06
	← joki left	09:06
	← teran left	09:06
	← midnightmagic left	09:06
	← munderwo left	09:06
	← Kabaka left	09:06
	← veebers left	09:06
	← geomyidae left	09:06
	← cwillu_at_work left	09:06
	→ Silmarilion joined	09:06
	→ carbon1 joined	09:07
	← salamanderrake left	09:07
	→ dsdeiz_ joined	09:07
	← rue left	09:07
	→ rue joined	09:08
	→ arand joined	09:08
	← dsdeiz left	09:08
	→ yuwang joined	09:09
	→ Kabaka joined	09:09
	← carbon1 left	09:09
	→ munderwo joined	09:10
	→ s4nchez1 joined	09:10
	← s4nchez1 left	09:10
	→ s4nchez1 joined	09:10
	← s4nchez left	09:11
	→ g0bl1n joined	09:11
	→ Sonderblade joined	09:11
	→ Khisanth joined	09:11
	→ teran joined	09:11
	→ midnightmagic joined	09:11
	→ veebers joined	09:11
	→ cwillu_at_work joined	09:11
	→ arand_ joined	09:11
	→ francisfish joined	09:12
	← rue left	09:12
	← g0bl1n left	09:12
	← dsdeiz_ left	09:13
	→ joki joined	09:13
	→ cllamas joined	09:13
	← Sonderblade left	09:13
	← Khisanth left	09:13
	← teran left	09:13
	← midnightmagic left	09:13
	← veebers left	09:13
	← cwillu_at_work left	09:13
	→ HelperW_ joined	09:13
	→ Khisanth joined	09:14
	→ Sonderblade joined	09:14
	← HelperW_ left	09:14
	→ mlechner joined	09:14
	→ Acidnerd joined	09:14
	→ HelperW_ joined	09:14
	→ teran joined	09:14
	→ LeMike joined	09:14
	→ veebers joined	09:15
	→ midnightmagic joined	09:15
LeMike	Hello. Some here on work still got it wrong. Is there a global config have `git clone` always checkout the develop instead of master?	09:15
	← HelperW left	09:16
grawity	`git clone` picks the branch that the remote repo's HEAD points to	09:16
	← DolphinDream left	09:17
	→ salamanderrake joined	09:18
LeMike	so i make a symbolic-ref in the bare repo, grawity ?	09:19
grawity	yes	09:19
LeMike	thanks! :)	09:19
	← michael_mbp left	09:20
	← midnightmagic left	09:20
	← arvind_khadri left	09:20
	→ ambodi joined	09:21
	← ahmedelgabri left	09:21
	→ Not_Whatever joined	09:21
	→ michael_mbp joined	09:21
	→ kent\n_ joined	09:21
	→ rue joined	09:22
	→ midnightmagic joined	09:22
	← kent\n left	09:23
	→ lazers joined	09:23
	→ rath joined	09:23
	← s4nchez1 left	09:23
	→ s4nchez joined	09:24
	→ shruggar joined	09:27
	→ ferreum joined	09:27
	→ _Qman joined	09:27
	→ kadoban joined	09:28
	→ thevishy joined	09:29
	→ woollyams joined	09:29
	→ Norm_ joined	09:29
	→ thierryp joined	09:29
	← bwreilly left	09:30
	→ blinkdesign joined	09:30
TomyWork	the git ls-tree manpage says "When paths are given, show them (note that this isnt really raw pathnames, but rather a list of patterns to match)." - what kind of pattern?	09:31
	i tried * and .*	09:31
	→ Asenar joined	09:32
	← jaseemabid left	09:34
daedeloth	does anyone know how I encode an "@" character in a http auth username?	09:35
_ikke_	daedeloth: I don't think you need to encode it	09:36
daedeloth	[email@hidden.address] ?	09:36
	← woollyams left	09:36
_ikke_	yes, that should work	09:36
	If not, try to percent encodfe it	09:37
TomyWork	i'm sure there are a few url validators that fail that :)	09:37
	even if it's valid	09:37
	→ arvind_khadri joined	09:37
_ikke_	I use mc to connect to ftp, and it has no problems with that	09:37
grawity	daedeloth: standard URL-escaping, '@' → %40	09:37
	→ SHyx0rmZ joined	09:39
	→ Art44 joined	09:39
	→ halu joined	09:40
	← s4nchez left	09:40
	→ s4nchez joined	09:41
	→ jackneill joined	09:41
	→ skorgon joined	09:41
	← cmpera left	09:42
	← pretty_function left	09:42
	karupanerura → zz_karupanerura	09:43
	← sitaram left	09:43
	← thevishy left	09:44
	→ ariscop joined	09:44
	← rue left	09:44
	→ rue joined	09:45
	← rue left	09:45
	← Guest29396 left	09:45
	→ thevishy joined	09:46
	→ hellopat joined	09:47
	→ jooools joined	09:47
	← Nizumzen left	09:48
	← skorgon left	09:48
	→ Spyro joined	09:48
	Spyro → Guest53173	09:49
	→ Advocation joined	09:49
	← CharlesFinley left	09:50
	→ freimatz joined	09:51
	← hellopat left	09:51
	← jatoben left	09:52
	→ sitaram joined	09:54
osse	TomyWork: the pattern should match directories	09:54
	→ ben_alman joined	09:54
osse	actually, it doesn't seem to like any form of wildcard	09:54
	→ jnavila__ joined	09:55
	← SHyx0rmZ left	09:56
	→ davividal joined	09:56
	← Guest53173 left	09:56
jast	ls-tree does use the PATH_GLOB option when parsing the pathspec	09:57
	I recommend quoting globbing pathspecs so that the shell doesn't sink its teeth into them	09:58
	← jnavila_ left	09:58
	→ pretty_function joined	09:58
	← ben_alman left	09:58
jast	that said, I can't get it to work either ;)	09:59
	← francisfish left	10:00
	← eijk_ left	10:00
	← nafg_ left	10:00
daedeloth	what's the proper way to maintain a stable branch?	10:02
	→ eijk_ joined	10:02
ghostlines	morning, someone pushed some changed and we now have merge conflict because we edited the same file. After 'fixing' the conflict when I run the a git status I see a bunch of modified files that the other person pushed. Why is this?	10:02
	← HelperW_ left	10:02
daedeloth	just create a branch" stable"?	10:02
	→ ChrisAnn joined	10:02
	→ HelperW_ joined	10:03
ghostlines	I would expect only the the file that has a conflict to be modified	10:03
	← HelperW_ left	10:04
	kent\n_ → kent\n	10:04
	← kent\n left	10:04
	→ kent\n joined	10:04
_ikke_	ghostlines: when a merge conflict happens, the files aren't comitted	10:04
	→ HelperW_ joined	10:04
grawity	your branch didn't have those files at all, and now you're making a merge commit that does have them, so they show up as added	10:04
ghostlines	grawity, ahh that makes sense	10:05
_ikke_	daedeloth: sure	10:05
	daedeloth: look at something like !gitflow for inspiration	10:06
gitinfo	daedeloth: The description of the gitflow branch workflow model is at http://nvie.com/posts/a-successful-git-branching-model/ while a tool to help implement this workflow is at https://github.com/petervanderdoes/gitflow See http://sethrobertson.github.com/GitBestPractices/#workflow for other workflow suggestions/references	10:06
ghostlines	grawity what I also noticed is that the I can't see the changes that the other person added in the file that had the merge conflict	10:06
	I just moved around some lines of code, how do I know if the conflict is 'really' solved?	10:06
	← teran left	10:07
daedeloth	_ikke_, hm, that's the opposite of what I had in mind, I thought the master branch was for development	10:07
ghostlines	and I deleted the HEAD and commit string that showed up in the merge conflict file	10:07
	← Art44 left	10:07
_ikke_	daedeloth: It doesn't matter	10:07
	← Silmarilion left	10:07
grawity	ghostlines: did you delete all the markers though?	10:07
ghostlines	yeah	10:08
_ikke_	daedeloth: The master branch is in no way special (except, that it's created by default, and by default the branch that gets checked out when cloning)	10:08
grawity	ghostlines: usually it shows up as "<<< HEAD" + your changes + "======" + other guy's changes + ">>> origin/blah"	10:08
ghostlines	but I could undo that	10:08
daedeloth	ah, alright.	10:08
	→ Guest53173 joined	10:08
	← konadr left	10:09
ghostlines	grawity, ahh thx, now I understand the format!	10:10
	should I then just delete the markers, check if the code still runs good then add and commit?	10:10
	← yuwang left	10:10
	→ alexandernst joined	10:10
	→ dort joined	10:11
	→ ahmedelgabri joined	10:12
	← obert left	10:12
	← thevishy left	10:12
	← alexandernst left	10:13
lb1a	ghostlines, no you should pick or meld the two different versions	10:13
	← Guest53173 left	10:13
	← s4nchez left	10:14
	← bgerber left	10:14
	→ msodrew joined	10:14
	→ m0ltar joined	10:14
lb1a	if you just remove the markers you end up with code that have both parts ( your code and the conflicting code from the other person) and most likely will be broken	10:14
_ikke_	ghostlines: A mergetool helps greatly with this	10:14
	→ s4nchez joined	10:14
	← vince_prignano left	10:14
ghostlines	ok thanks alot	10:14
	→ vince_pr_ joined	10:15
	← lazers left	10:15
_ikke_	ghostlines: Use something like meld or kdiff3	10:15
ghostlines	cool	10:15
lb1a	(git mergetool has some proposals)	10:15
	→ cbot joined	10:16
	← s4nchez left	10:16
	→ s4nchez joined	10:17
	→ Spyro joined	10:17
	→ bgerber joined	10:17
	Spyro → Guest58859	10:17
	← vince_pr_ left	10:18
machine-x	WAIT 2000	10:18
_ikke_	soo long?	10:18
	← Guest58859 left	10:18
	→ Guest58859 joined	10:18
	Guest58859 → Spyro	10:18
	→ beautifulmind joined	10:19
	← jayallen left	10:19
	→ arif-ali joined	10:20
introom	l;/join ##morse	10:20
	← arizuka_ left	10:20
	→ arizuka joined	10:21
	→ nilg` joined	10:21
	→ Davey joined	10:21
	← s4nchez left	10:21
	← Exancillatus left	10:22
	← xlinkz0 left	10:22
	← hyperair left	10:23
	← dyng left	10:23
	→ dyng joined	10:24
	→ simesy joined	10:24
	← h0bbit_ left	10:25
lb1a	:wq	10:25
	← fairuz left	10:25
_ikke_	E_NO_VIM	10:25
	← arif-ali left	10:25
	← luc_ left	10:26
	dwmw2_gone → dwmw2	10:26
	→ acharlieh joined	10:28
	← dyng left	10:28
	→ arif-ali joined	10:29
	← echo-area left	10:29
	→ jizanthapus-work joined	10:29
	→ jayallen joined	10:30
	→ francisfish joined	10:31
	← francisfish left	10:32
	← endiruna left	10:32
	→ francisfish joined	10:32
	→ Silmarilion joined	10:32
	← acharlieh left	10:32
	→ vince_prignano joined	10:33
	← arif-ali left	10:33
	← vkrest left	10:35
	→ eaj joined	10:36
	→ arif-ali joined	10:36
	← hkon_ left	10:38
	→ lotsofcode joined	10:38
	→ Voyage joined	10:39
Voyage	I am using intelij idea IDE and github. I made a fork out of a main repo on github. Now I should checkout from the main and commit/push to the fork?	10:39
	→ hkon_ joined	10:39
	← jayallen left	10:39
	← arif-ali left	10:41
jizanthapus-work	Voyage, you should clone from your fork	10:41
_ikke_	It doesn't really matter where you clone from	10:41
	← Davey left	10:43
jizanthapus-work	_ikke_, it does wrt the tracking he'll have set up	10:43
_ikke_	jizanthapus-work: You can always adjust tracking if you want	10:43
	← ewet left	10:44
	← eletuchy_ left	10:44
_ikke_	You can pull / push from / to many repositories from one repo	10:44
	→ eletuchy joined	10:44
jizanthapus-work	_ikke_, in this case he will only pull and push from his fork	10:44
	→ dyng joined	10:44
jizanthapus-work	hence no need for extra step	10:45
_ikke_	Nope, he should also pull from the main repo	10:45
	If you intend to make pull requests that is	10:45
Voyage	jizanthapus-work so where to checkout and were to commit? where /how to make pull request	10:45
	→ teran joined	10:46
jizanthapus-work	_ikke_, well ok, but still pushing to the main doesn't make sense	10:46
	Voyage, clone from your fork, push to it, use github's GUI for pull requests	10:46
	→ Exancillatus joined	10:46
_ikke_	jizanthapus-work: If you already cloned to repo, you don't need to reclone. If you still need to clone, it's indeed easier to just clone from your fork	10:47
lb1a	jizanthapus-work, that will make the upstream maintainer angry, if he doesn't pull/merge the upstream prior to starting his PR	10:47
Voyage	jizanthapus-work I should checkout from fork also? or from main?	10:47
	← Advocation left	10:47
lb1a	Voyage, doesn't matter, should be the same anyway	10:47
_ikke_	Voyage: you use checkout the way svn uses it	10:47
	Voyage: checkout in git means something different	10:47
	← weiyang left	10:48
	← jnavila__ left	10:48
	→ jnavila joined	10:48
Voyage	_ikke_ I didnt used svn	10:48
jizanthapus-work	Voyage, checkout is when you already have a clone and want to retrieve a file at a specific revision	10:48
	what you want here is "clone"	10:48
	← eletuchy left	10:49
jizanthapus-work	lb1a, noted; Voyage if you want to contribute to the main repository, add it as a remote and pull from it before making a pull request	10:49
_ikke_	jizanthapus-work: That was what I was trying to tell ;-)	10:49
	→ ciampix joined	10:49
	→ bvi joined	10:49
jizanthapus-work	_ikke_, ok :-)	10:50
lb1a	all credits belong to _ikke_	10:50
	← dort left	10:50
	← bvi left	10:50
	→ eijk joined	10:50
	← abrkn left	10:51
	← eijk_ left	10:51
	→ nedbat joined	10:52
	→ pantsman joined	10:52
	→ jayallen joined	10:53
	← eaj left	10:53
	→ huahax joined	10:53
	→ eaj joined	10:53
	→ dstuxo joined	10:55
	← ambodi left	10:55
	← yhager left	10:55
	→ bePolite joined	10:56
bePolite	How large mus a repo be for gith to hand	10:56
	handle?	10:57
grawity	are you asking for the maximum, or the minimum?	10:57
Voyage	jizanthapus-work _ikke_ lb1a when I will make a pull request to main (from my fork) and the project manager entertaines my pull requests, merges. Will my fork be also updated auto matically and the main will merge to my fork? (the changes of other collaborators / forks all get merged in main. so My fork will get those too?)	10:58
bePolite	grawity: For a maximum please	10:58
	→ s4nchez joined	10:59
RichiH	bePolite: you mean the maximum git can handle?	10:59
jizanthapus-work	Voyage, if you've made a pull request with some changes, you will already have the changes in your own repo	11:00
	→ ambodi joined	11:00
bePolite	Yes RichiH	11:00
	I mean the maximum size of a repository which git can handle	11:01
	→ h0bbit_ joined	11:01
RichiH	there is no clear-cut maximum	11:01
	bePolite: but unless your use case is _extreme_ you will be fine	11:01
bePolite	Ok so 500MB is fine	11:01
RichiH	bePolite: heh	11:02
	yes	11:02
lb1a	Voyage, no changes in another remote (main) will not make any changes to your remote (fork) unless you fetch/pull from main and push to fork	11:02
	→ diegoviola joined	11:02
RichiH	bePolite: below a few dozen GB, i would not even think about scalability	11:02
	→ s4nchez1 joined	11:03
RichiH	(and my largest git _annex_ repository is almost 1 TB in size)	11:03
	← s4nchez left	11:03
	← buzzybron left	11:04
	← solker left	11:04
murosai	how can i check if a file has been added to git repository?	11:05
	→ craigp joined	11:05
	← daito left	11:05
	→ ben_alman joined	11:05
moritz	murosai: do you mean 'added' as in 'added to the index'? or committed?	11:06
	if the latter, git log -- filename	11:06
	← jayallen left	11:06
	→ vkrest joined	11:06
murosai	yeah the latter	11:06
Voyage	lb1a how can i pull from the main?	11:07
	← vkrest left	11:07
	← dstuxo left	11:07
	→ vkrest joined	11:08
Voyage	lb1a how can i pull from the main in my fork?	11:08
	← s4nchez1 left	11:08
	→ correcto joined	11:09
	← robban666 left	11:10
	← ben_alman left	11:10
bePolite	Voyage: "git pull upstream master"	11:11
	but you first have to add the upstream	11:12
	← Exancillatus left	11:13
lb1a	Voyage, not possible to "pull" from "main" into "fork"	11:13
	Voyage, you always have to clone either of these two, add the other one as remote (which you should have done by now) fetch both of them, merge them together as you like and push to remote if you want to publish your changes/merges/whatever.	11:14
	← sluther left	11:14
lb1a	i sense a lack of knowledge about git !workflow	11:15
gitinfo	Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	11:15
	← bePolite left	11:15
	→ eletuchy joined	11:15
Voyage	thanks	11:15
	← ambodi left	11:16
	→ robarelli joined	11:16
	→ ambodi joined	11:17
	← veebers left	11:17
diegoviola	some developer is asking me to rewrite my history and use --force, that's most of the time a bad idea, right?	11:18
	he said it's not dangerous if it's in my branch	11:18
shruggar	you mean with git-filter-branch?	11:18
_rgn	afaik if it's your own branch and other people don't work on it it's probably safe	11:19
diegoviola	i think he mentioned git commit --amend	11:19
	ok thanks	11:19
lb1a	diegoviola, !rewrite	11:19
gitinfo	diegoviola: [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	11:19
shruggar	rewriting history for unpublished work is part of regular workflow	11:19
	← Sabbath left	11:20
lb1a	diegoviola, the git push -f part is the dangerous one	11:20
shruggar	understand the git reflog. History rewriting is (usually) safe	11:20
diegoviola	thanks	11:20
	i mentioned that i'll just keep the history as-is	11:20
	but they insisted	11:20
shruggar	rewriting published work is annoying at worst. "safe" to me means "has potential data loss"	11:21
	err, "does not have potential data loss" ;)	11:21
_rgn	rewriting published work others haven't touched == ?? (safe afaik)	11:21
Voyage	lb1a jizanthapus-work If I am the project manager, the main repo is on my account. If i make code changes, i should also make another fork or just commit to the main?	11:21
lb1a	diegoviola, you can rewrite your unpublished history as much as you like, but rewriting published history involves "potential data loss" and "screw other peoples work" and "GTFO" and more ;)	11:21
	← jargon left	11:22
lb1a	Voyage, your choice.	11:22
jizanthapus-work	Voyage, up to you	11:22
diegoviola	thanks	11:22
	how do i rewrite unpublished history?	11:22
	like merging few commits into one and such	11:22
_rgn	git commit --amend, git rebase -i	11:23
Voyage	lb1a jizanthapus-work whats a better appraoch? 2. if I make another fork, do I need another account?	11:23
diegoviola	thanks	11:23
	git commit --amend is only for changing the message of a commit no?	11:23
	→ jargon joined	11:23
_rgn	it will apply what you have staged to the commit also	11:23
	so not just the message	11:23
diegoviola	ok thanks	11:23
lb1a	diegoviola, everything that involves 'modifiying' existing commits is a "rewrite" of history, because in git objects are immutable. so every operation that changes existing commits is a rewrite. e.g. git commit --amend or git rebase (-i)	11:23
shruggar	git rebase -i @{u} <--- allows you to edit all history not yet published to the upstream	11:23
jizanthapus-work	Voyage, I'm not sure github allows you to fork your own repos	11:24
	Voyage, but you can always branch	11:24
Voyage	jizanthapus-work ya. your a right	11:24
	oh so, I can do a branch? that will work exactly like the fork?	11:24
	jizanthapus-work ok, the workers who have to fork the main, must be contributors of main repo too. so they can commit directly too? right?	11:25
lb1a	diegoviola, a git reset to some older point and contiuing another path of development (and pushing that) is a rewrite of existing history	11:25
diegoviola	right thanks	11:25
	→ Exancillatus joined	11:25
shruggar	"fork" is just a github word which can be translated into "branch in a way which doesn't show up in the main repository's list of branches"	11:25
	← introom left	11:25
lb1a	githubs terminology is not the best :/	11:25
ohcibi	jizanthapus-work: of course it allows you to fork, it just means copy it to another user/organization you have rights for	11:26
Voyage	jizanthapus-work <shruggar> "fork" is just a github word which can be translated into "branch in a way which doesn't show up in the main repository's list of branches"	11:26
	true i guess	11:26
shruggar	I don't know how the internals of github work, but I'd be surprised if it actually did a "git clone" behind the scenes whenever "fork" was clicked	11:27
	→ macben joined	11:27
Voyage	lb1a jizanthapus-work ok, the workers who have to fork the main, must be contributors of main repo too. so they can commit directly too? right?	11:27
	→ hellopat joined	11:27
lb1a	Voyage, no how said that ?	11:28
Voyage	sorry?	11:29
	→ Sly1986 joined	11:29
	→ s4nchez joined	11:29
TomyWork	shruggar it might be on some dedup-capable fs	11:29
Voyage	if one can fork the main, he can commit to main too.	11:29
	← macben left	11:29
	← jargon left	11:29
Voyage	how to make a new branch on github site?	11:29
lb1a	Voyage, the whole point of "forking" in github terms is, that you have no commit rights on the "MAIN" repo. so everyone plays in his little sandbox (fork) and if he done he'll message the projectmanager and creates a Pull request, so the manager can integrate the work of the sandbox into the main repo	11:30
TomyWork	why else would they take actual time when fork happens?	11:30
	→ DoniRS joined	11:30
lb1a	Voyage, no, forking requires read access, commiting requires write access. 2 seperate things	11:30
TomyWork	branching, too	11:31
	→ skorgon joined	11:31
	← hellopat left	11:32
ohcibi	TomyWork: no branching only requires write access if you want to push the branches	11:32
	8-)	11:32
	actually committing too	11:32
TomyWork	local branching is not a github operation	11:33
	i thought we were talking about github	11:33
ohcibi	oh, i forgot.. you can even branch and commit on github now....	11:34
TomyWork	yep and then accidentally make a pull request	11:34
ohcibi	which essentially makes students use github as a gui for git	11:34
	even confusing their remote repository with their local ones	11:34
	just because github	11:35
TomyWork	i once or twice used github as a student	11:35
	because it's free git hosting and who cares about the source of their homework?	11:35
	→ rchavik joined	11:35
ohcibi	we use it for a course where the students have to work on a project for one semester... proper usage of git is covered by our topic, but I'm mostly fighting against git-guis	11:36
	← donnib left	11:36
	← roxlu left	11:37
shruggar	I have set up git for several companies, and invariably someone comes in who only has any interest in using the "github client", which doesn't seem to support any basic operation, it feels like dropbox with optional commit messages	11:37
	← skorgon left	11:38
ohcibi	github client	11:38
	shiver	11:38
shruggar	I have yet to successfully walk someone through its use in a way which doesn't involve them, two days later, pushing up 20 merge commits and 7 revert commits with no explanation as to why they were done, because the person using it didn't actually intend to do any of that	11:39
TomyWork	shruggar what's your take on gitlab?	11:39
	← machine-x left	11:39
	← subhojit777 left	11:39
	→ mikecmpbll joined	11:39
	→ machine-x joined	11:40
	→ solker joined	11:40
shruggar	never used it, but I expect to have an opinion by the end of the weekend. I'm at a client now who is the first to want to use git but not github (most clients just want to switch to using "github", don't know that git is a separate thing)	11:40
	→ vector_ joined	11:40
TomyWork	that's what i set up for the company. setup was quite painless for it being ruby	11:40
	← arvind_khadri left	11:41
	← vkrest left	11:41
ohcibi	TomyWork: setting up rails apps is quite painless in general nowadays... gitlab is nice	11:42
shruggar	ah, ruby will likely be a problem. They only have one aging server which they're afraid to install any new packages on (trying to convince them to let me set up some puppet scripts or similar for re-building a server image for disaster recovery and testing...)	11:42
	→ hejia joined	11:42
ohcibi	but I miss proper project management	11:42
	← m0ltar left	11:42
ohcibi	its great for repositories but nothing beats redmine on projec tmanagement	11:42
	issues/calendar/wiki/documents and stuff	11:42
	← vector_ left	11:43
shruggar	I seem to recall trying redmine once, but either I never got it set up, or the company never used it after it was set up, so it was forgotten	11:43
ohcibi	shruggar: if the server has sufficient packages to build ruby, it is not necessary to install additional packages... rvm ftw.. the deps are not that much	11:43
	← Silmarilion left	11:44
shruggar	I use rvm locally, never tried it on a server environment (I use rvm to match server environments)	11:44
ohcibi	shruggar: it was intentionally build for managing server environments...	11:44
	→ dougsland joined	11:45
TomyWork	repo management and a bug tracker are two different things	11:45
	even if you call it project management for some reason :)	11:45
ohcibi	TomyWork: i know, still a project consists of repos, so its nice to combine them	11:45
	→ mvelioglu joined	11:45
ohcibi	TomyWork: its just that I wouldnt need the issue managment of gitlab just because of that reason	11:46
	← kcj left	11:46
TomyWork	oh	11:46
	i dont even use that either	11:46
shruggar	I hate using multiple package managers, really. I wish they'd all talk to each-other better	11:46
ohcibi	TomyWork: we use it at work as a full project management suite... but integrating redmine is planned	11:46
TomyWork	shruggar like rpm and dpkg?	11:46
telmich	good day	11:47
moritz	telmich: lies!	11:47
	:-)	11:47
ohcibi	shruggar: rvm is not a package manager, it is a version manager that even uses the local package manager to install builddeps	11:47
telmich	moritz: always	11:47
shruggar	like rpm and dpkg and ruby gems and CPAN and rvm and composer and puppet modules	11:47
	→ setmeaway joined	11:47
shruggar	if you don't consider rvm to be a package manager, you either have a poor version of the word "package" or a poor version of the word "manager"	11:47
ohcibi	shruggar: you create a user for your web-app instal rvm for him and then you can bundle install all along, not affecting any other user/packages on the system	11:47
	← Watabou left	11:48
	← ahmedelgabri left	11:48
telmich	how can I find out, which remote branch I am tracking? i.e. I want git to return only remotes/origin/name/to/	11:48
TomyWork	for python there is stdeb which makes debian packaged from pip packages	11:48
ohcibi	shruggar: rvm is not a package manager..	11:48
TomyWork	maybe there's something like that for the others as well?	11:48
	→ vector_ joined	11:48
TomyWork	packages*	11:48
ohcibi	shruggar: it does way more than a package manager does	11:48
	← robarelli left	11:49
	→ ahmedelgabri joined	11:49
ohcibi	shruggar: the package manager is rubygems	11:49
TomyWork	ohcibi it should do what i mean	11:49
	like install the correct ruby versions for my apps	11:50
	→ wbdecker joined	11:50
	← wbdecker left	11:50
ohcibi	and rvm puts the packages of rubygems into a sandbox, that way you dont have to clutter up the systems packages with the rubygems	11:50
	← arand_ left	11:50
	← pretty_function left	11:50
	→ skorgon joined	11:50
moritz	telmich: git rev-parse --symbolic-full-name '@{u}'	11:51
	→ arand_ joined	11:51
	← aard_ left	11:51
	← arand left	11:51
	→ bartman` joined	11:51
	arand_ → arand	11:51
	← Spyro left	11:52
	← s4nchez left	11:53
ohcibi	TomyWork: it does exactly that.. you tell it to rvm install ruby-version and it installs the ruby version for the current user (create one for the app, e.g.), all gems will now be installed into the specific environment for that ruby version, bundler will then handle the needed versions of the gems for each app (or you can create specific envs for each app, depends on your needs)	11:53
	← Guest60494 left	11:53
	→ shinnya joined	11:54
	→ Spyro joined	11:54
	Spyro → Guest69162	11:55
	→ s4nchez joined	11:56
	← skorgon left	11:57
	→ kriks joined	11:58
	← thierryp left	11:58
	→ thierryp joined	11:58
	→ arvind_khadri joined	11:58
TomyWork	ohcibi hmm, the init script of that app seems to ignore it though	11:59
	→ Devastator joined	11:59
ohcibi	TomyWork: you mean there is a shell script that does not respect the current rvm environment?	11:59
	TomyWork: lets continue this in #rvm and have a look at the manuals on rvm.io, lots of common problems are explained there	12:00
TomyWork	it's an init.d script	12:00
	not gitlab's	12:00
	← paulweb515 left	12:00
TomyWork	i'm working on something else right now though, but thanks for the offer	12:00
ohcibi	TomyWork: /j #rvm	12:01
	TomyWork: https://rvm.io/integration/init-d	12:01
	← Devastatr left	12:01
	← intellix left	12:01
	← Devastator left	12:02
	→ Devastator joined	12:02
	← zxd left	12:03
	dean\|away → dean	12:04
	← MattMaker left	12:04
	← nedbat left	12:05
	← danicampa901 left	12:05
	→ ben_alman joined	12:06
	→ intellix joined	12:07
	← arizuka left	12:07
	← diegoviola left	12:08
	← ahmedelgabri left	12:08
	← ben_alman left	12:10
	← Xiti` left	12:11
	→ Nizumzen joined	12:11
	→ angelsl joined	12:12
	← jooools left	12:12
	→ Xiti joined	12:12
	→ cmpera joined	12:13
	→ zxd joined	12:14
	← vector_ left	12:15
	→ jargon joined	12:16
	← angelsl left	12:16
	← pantsman left	12:16
	→ Stygia joined	12:17
	→ EngierkO joined	12:18
	← EngierkO left	12:18
	← vurma left	12:18
	← ambodi left	12:20
	→ irqq_ joined	12:20
	→ ambodi joined	12:20
	→ konadr joined	12:20
	← konadr left	12:21
	→ konadr joined	12:21
	← arvind_khadri left	12:22
	→ dfanjul joined	12:22
	← juril left	12:22
	→ EngierkO joined	12:22
	← EngierkO left	12:23
	← irqq left	12:23
	← h0bbit_ left	12:23
	→ EngierkO joined	12:23
	irqq_ → irqq	12:23
	→ h0bbit_ joined	12:24
	→ lazers joined	12:24
	→ CheckDavid joined	12:24
	← ambodi left	12:25
	← h0bbit_ left	12:25
	← dyng left	12:25
	→ dyng joined	12:26
	← rudisimo left	12:26
	← machine-x left	12:26
	← vince_prignano left	12:27
	→ rudisimo joined	12:27
	→ machine-x joined	12:27
	→ Advocation joined	12:30
	← dyng left	12:31
	→ the_cybersphinx joined	12:31
	← rudisimo left	12:31
	← Nizumzen left	12:31
	← Chais left	12:32
	→ diegoviola joined	12:33
Voyage	lb1a but github normal account has both read / write acces	12:33
	lb1a if one is a collaborator in a private repo. he gets full acces	12:34
	→ vtunka joined	12:34
	← nfroidure left	12:34
Voyage	lb1a what you said is correct for public repos	12:34
	→ ambodi joined	12:35
	← ambodi left	12:35
	← daedeloth left	12:35
	→ ambodi joined	12:35
	→ netcarver joined	12:36
lb1a	Voyage, i dont knot about private repos access levels, but i work with a public organisation.	12:37
	← ciampix left	12:37
	→ freeed joined	12:37
lb1a	Voyage, there you can add/edit teams with their members in read/write/admin access levels	12:38
	→ vkrest joined	12:38
lb1a	Voyage, i'd think this is similar to private repos.	12:38
	(if not, just create an organisation	12:38
	organization*	12:38
	← poulson left	12:38
	← jxpx777 left	12:40
	→ macben joined	12:40
	→ poulson joined	12:41
	← vkrest left	12:42
	← _Vi left	12:42
	→ ahmedelgabri joined	12:43
	← Exancillatus left	12:44
	← francisfish left	12:44
	← macben left	12:46
	→ ZenGeist joined	12:46
	→ cannonball joined	12:47
ZenGeist	Hi all! I attempted to create new remote branch in repo with no changes, just new branch and after pushing it I saw, that git is sending all files to remote	12:48
	so I get doublesized repo	12:48
	← jnavila left	12:49
	← eslwork left	12:52
	← dobalina left	12:52
	→ Exancillatus joined	12:52
ohcibi	ZenGeist: git actually only pushs the new branch-ref... what did you actually see that makes you think that this repo is double sized now?	12:53
ZenGeist	ohcibi: too long process of pushing. cca 30min with a lot of files	12:54
	→ jnavila joined	12:54
ohcibi	ZenGeist: maybe you did some commits after you pushed the other branches the last time but before you branched off?	12:55
ZenGeist	ohcibi: no	12:55
	→ tmcmahon joined	12:55
ohcibi	ZenGeist: then what exactly have you done?	12:55
ZenGeist	git checkout master	12:56
	git status	12:56
	git branch -b live	12:56
	git push -u origin live	12:56
ohcibi	if that really pushed new commits then either master wasnt pushed before or you did what I wrote before	12:56
	or you misinterpreted something and it hasn't actually pushed new commits	12:57
	← Baluse left	12:58
ZenGeist	ohcibi: there are no new commits just the same repo one more	12:58
	← ahmedelgabri left	12:58
ZenGeist	it was 2gb and now it's 4gb	12:58
	← Advocation left	12:58
ohcibi	ZenGeist: have you rebased?	12:58
ZenGeist	no	12:58
	← nilg left	12:58
	→ liuzhoou_ joined	12:59
ohcibi	again: if master was pushed before and you typed the commands in exactly this order with no commands at all in between, no commits can possibly be pushed	12:59
	or you pushed master to another remote...	12:59
	→ Macaveli joined	12:59
ohcibi	ZenGeist: can you show the repo, or reproduce an example?	12:59
	← tmcmahon left	13:00
	← lotsofcode left	13:00
	→ fire joined	13:00
	→ Baluse joined	13:01
ZenGeist	ohcibi: unf. no =( it was repo only with master branch	13:01
	→ Malice joined	13:01
	← intellix left	13:03
	← mlechner left	13:03
ohcibi	ZenGeist: when git pushes a branch, it checks what commit that branch points to, if that commit is already pushed (like when master was pushed before), then git only creates a new branch object remotely and let it point to the appropriate remote-commit, there is not even something uploaded here	13:03
	← ewnd9 left	13:03
ohcibi	ZenGeist: this means there must have been something else you did, or you misintepreted.. as you cant reproduce it, remember this for the next time 8-))	13:03
	→ mlechner joined	13:04
ZenGeist	ohcibi: ok, thx	13:04
	← simesy left	13:06
	→ m0ltar joined	13:06
	← mdpatrick left	13:06
	← boblee left	13:07
	→ boblee_ joined	13:08
	← huahax left	13:08
	← zumba_addict left	13:08
	→ nnyk_ joined	13:09
	→ smashwilson joined	13:10
	→ canton7-mac joined	13:10
	← halu left	13:10
	→ Kayra joined	13:11
	→ Gamecubic joined	13:12
	→ vince_prignano joined	13:14
	← makubi left	13:14
	→ jxpx777 joined	13:15
	→ jxpx777_ joined	13:15
	← m0ltar left	13:15
	← jxpx777_ left	13:15
	← ZenGeist left	13:15
	← Exdaix left	13:15
	→ _Vi joined	13:15
	→ aberrant joined	13:15
	→ CharlesFinley joined	13:16
CharlesFinley	Hi guys	13:16
	→ makubi joined	13:16
CharlesFinley	1. I cloned a repo and switched to a new branch	13:16
	← correcto left	13:16
CharlesFinley	git clone xxx	13:16
	git checkout -b feature-something-something	13:17
bremner	3. profit!	13:17
CharlesFinley	did some changes	13:17
	did a git add .	13:17
	and got a middle finger in return	13:17
canton7-mac	what exact type of middle finger?	13:17
bremner	please be more specific.	13:17
	bah.	13:17
	→ dyng joined	13:17
CharlesFinley	http://hastebin.com/degerucige.vbs	13:17
	← diegoviola left	13:18
	→ m0ltar joined	13:18
canton7-mac	it's not looking like you actually made any changes?	13:18
	[emmajane] → emmajane	13:18
CharlesFinley	http://hastebin.com/foqemipici.xml	13:19
	here's what it shows when i do git log	13:19
	→ nfroidure joined	13:19
CharlesFinley	it actually logged my commits	13:19
bremner	right, so you already committed your changes	13:19
	→ mellernoia joined	13:19
_ikke_	yes	13:19
	You have already comitted them	13:19
	so git says everything is clean	13:19
	← jxpx777 left	13:19
CharlesFinley	ok then - why can't i push?	13:19
_ikke_	CharlesFinley: Don't know	13:19
canton7-mac	I don't know. why can't you push	13:19
	?	13:19
	→ correcto joined	13:20
bremner	did you try "git push" ?	13:20
CharlesFinley	http://hastebin.com/korexugole.vhdl	13:20
_ikke_	CharlesFinley: Have you tried reading?	13:20
canton7-mac	looks like it pushed to me	13:20
	→ diegoviola joined	13:20
	← Norm_ left	13:20
CharlesFinley	and then i go to github and can't see neither my new branch	13:20
	nor any of my commits	13:20
	→ cads joined	13:21
canton7-mac	git log --graph --oneline --decorate --all	13:21
	pastie the result of that	13:21
_ikke_	canton7-mac: You haven't set push.default	13:21
	CharlesFinley: *	13:21
	→ Norm_ joined	13:21
canton7-mac	that's just a warning though - the push succeeded with 'everything up to date' at the bottom	13:21
_ikke_	CharlesFinley: git push origin master	13:21
canton7-mac	good point, matching as some odd behaviour, I always forget	13:21
_ikke_	canton7-mac: yes, but git is trying to push master	13:21
	CharlesFinley: git push origin <branch_name> rather	13:22
	git push by default pushes every branch that the remote also has	13:22
canton7-mac	yeah, CharlesFinley, by (the current) default, 'git push' pushes all branches where a branch or the same name exists on the remote. since your branch doesn't yet exist on the remote, vanilla 'git push' didn't push it	13:22
	→ lindenle joined	13:22
	→ skorgon joined	13:22
canton7-mac	(as that warning says, this behaviour is changing in git 2.0, and you can adopt the new behaviour now)	13:22
CharlesFinley	_ikke_: thanks, it worked	13:22
	git push origin branchname worked ;)	13:23
_ikke_	CharlesFinley: if you do git push -u origin branchname, it also sets up tracking information (you only have to do that once)	13:23
canton7-mac	(so that 'git pull' will know what to fetch and merge)	13:23
CharlesFinley	thanks :)	13:24
_ikke_	And also for git status	13:24
	← takamichi left	13:24
canton7-mac	and some others	13:24
	← Macaveli left	13:24
	→ Exdaix joined	13:24
	→ Macaveli joined	13:24
_ikke_	git branch -vv	13:25
canton7-mac	yup	13:25
	← Macaveli left	13:26
	→ ben_alman joined	13:26
	→ Macaveli joined	13:26
	← msodrew left	13:27
	→ vurma joined	13:27
	← beautifulmind left	13:29
	→ angelsl joined	13:30
	← dyng left	13:31
	→ takamichi joined	13:31
	→ CKyle joined	13:31
	← tbrock left	13:31
	← skorgon left	13:32
	→ tbrock joined	13:32
	→ stephCoue joined	13:33
	← diegoviola left	13:33
	← Exancillatus left	13:33
	→ alexlindenlevy joined	13:35
	→ rudisimo joined	13:35
	→ PaddyPatPat joined	13:35
	→ Advocation joined	13:36
	→ ewnd9 joined	13:36
	→ simesy joined	13:37
	→ baba joined	13:37
	← takamichi left	13:38
	← fire left	13:38
	→ vkrest joined	13:38
	baba → fire	13:39
	← mikecmpbll left	13:39
	→ takamichi joined	13:40
	← vkrest left	13:43
	→ subhojit777 joined	13:44
	← simesy left	13:45
	→ arvind_khadri joined	13:45
	→ diegoviola joined	13:45
	← nilg` left	13:45
	← solker left	13:46
	← ItSANgo left	13:46
	→ paulweb515 joined	13:47
	← paulweb515 left	13:47
	→ paulweb515 joined	13:48
	→ berzerka joined	13:48
	→ thevishy joined	13:48
	← thevishy left	13:48
	← olle left	13:48
	→ thevishy joined	13:48
berzerka	after i cloned a repo, do i need to fetch remote branches (from origin) to see them, or should i already see them with git branch -a?	13:48
_ikke_	berzerka: git branch -r (although, git branch -a shows them also)	13:49
LionKing	berzerka: You have to ¨checkout¨ them	13:49
berzerka	_ikke_, well then that project seems not to use any branches, was a bit confused. thanks!	13:49
LionKing	if you want to work on them	13:49
_ikke_	Just checking them out is not enough	13:49
	→ ehmatthes joined	13:50
_ikke_	You have to create 'local' branches for them	13:50
	→ v0n joined	13:50
	→ dort joined	13:50
LionKing	_ikke_: But does´t checkout exactly do that?	13:50
_ikke_	LionKing: There is some automagic that does that, yeah	13:51
berzerka	when i do checkout origin/foo, will a local branch foo be created implicityly?	13:51
	→ intellix joined	13:51
berzerka	-y	13:51
_ikke_	berzerka: git checkout foo will do that	13:51
canton7-mac	!local_branch_from_remote	13:51
gitinfo	The following commands are all equivalent, assuming <branch> doesn't yet exist: 'git checkout -b <branch> <remote>/<branch>', 'git checkout -t <remote>/<branch>', 'git checkout <branch>'. The latter invokes some magic.	13:51
berzerka	alright then.	13:51
_ikke_	And it only works if there is no ambiguity	13:51
canton7-mac	('git checkout origin/foo' will detach the HEAD - so you'll be able to see the files as they are on origin/foo, but any commits you make won't be attached to the branch. make sure you checkout a proper branch again before committing	13:51
	→ decotii joined	13:52
	→ ItSANgo joined	13:53
	→ JerryT joined	13:53
	→ mizu_no_oto joined	13:54
	→ yacks joined	13:54
	→ polyzen joined	13:54
	→ solker joined	13:54
	→ IrishPrime joined	13:55
	→ skorgon joined	13:56
	→ lotsofcode joined	13:57
bernalex	how do I get the behaviour of git rebase -i when I only have my master branch and no upstream?	13:58
	→ Diemuzi joined	13:58
	gitinfo set mode: +v	13:58
bernalex	what I want to achieve is reordering two commits and then squeezing them together.	13:58
_ikke_	bernalex: By manually providing an upstream commit	13:58
moritz	bernalex: you can do git rebase -i <hash>	13:58
bernalex	_ikke_: there is no upstream	13:58
	moritz: oh yes of course	13:59
moritz	or HEAD~10 or so # for the last 10 commits	13:59
	→ sohail joined	13:59
	→ liuzhoo__ joined	13:59
	← liuzhoou_ left	13:59
bernalex	it shows that I have been busy with meetings and planning and crap keeping me away from programming for far too long, considering this is something I do daily (almost hourly) usually :-P	13:59
	← liuzhoo__ left	13:59
bernalex	doh. how do I sqaush a commit to the initial commit though?	14:00
	← alexlindenlevy left	14:00
	→ luxgeek joined	14:00
bernalex	I know I can checkout and reset and ammend, but is there no way to do it more conveniently á la rebase -i?	14:00
_ikke_	bernalex: newer version of git have git rebase --root	14:00
	→ alexlindenlevy joined	14:00
	→ phile joined	14:01
bernalex	_ikke_: of course. I have used this before plenty of times as well. :-P thanks.	14:01
	← craigp left	14:01
	→ cmpera_ joined	14:01
_ikke_	bernalex: Note that rebase calls to commit to base on upstream	14:02
	← cmpera left	14:02
	cmpera_ → cmpera	14:02
_ikke_	It's not tied to a tracking branch	14:02
bernalex	_ikke_: yes, that's fine. there's only master in this pet project right now.	14:02
	hopefully I can write something interesting today and put it on github or whatever.	14:02
	→ jerius joined	14:03
	→ el3ctron_ joined	14:03
	← skorgon left	14:03
	→ nedbat joined	14:04
	← nedbat left	14:04
	← alexlindenlevy left	14:05
	← Exdaix left	14:05
	← takamichi left	14:08
	← fire left	14:09
	← Kayra left	14:09
	→ fire joined	14:09
	← m0ltar left	14:09
	fire → Guest31108	14:10
	← hakril left	14:10
	← Guest31108 left	14:10
	→ Davey joined	14:10
	→ baba joined	14:10
	→ takamichi joined	14:11
	← dougsland left	14:12
	→ dsdeiz joined	14:12
	→ dougsland joined	14:12
	dobby156_ → dobby156	14:12
	→ g105b joined	14:13
	← teran left	14:13
	→ Exdaix joined	14:13
	← githogori left	14:13
	→ teran joined	14:13
	→ matheus23 joined	14:13
	dean → dean\|away	14:14
g105b	When a branch is deleted from my remote (in this case, Github), how do I reflect these deletions in the `git branch -a` list?	14:14
_ikke_	git remote prune	14:14
	← aberrant left	14:14
	→ liamkeily joined	14:15
	→ ojacobson joined	14:16
	← liamkeily left	14:17
	← hejia left	14:17
	→ mtk joined	14:18
	← gua_ould left	14:18
	→ FortySix2 joined	14:19
	→ m0ltar joined	14:19
	← FortySix2 left	14:20
	← dobby156 left	14:20
	→ ahmedelgabri joined	14:21
	→ BenL joined	14:21
	← ahmedelgabri left	14:21
	→ ezekielnoob joined	14:21
	← takamichi left	14:22
BenL	if I have two git repos, allegedly the same. and I think one might've been tampered with, how would I check?	14:22
	← mizu_no_oto left	14:22
	← Lunatrius left	14:22
	← CKyle left	14:22
armin_w	BenL: run a diff tool against them	14:23
TomyWork	what do i do with https://raw.github.com/git/git/v1.7.2.5/contrib/completion/git-completion.bash on squeeze?	14:23
	(which comes with 1.7.2.5)	14:23
	→ FortySix2 joined	14:24
BenL	armin_w: you mean, check out from each repo and manually diff each branch?	14:24
armin_w	clone both repos and do a file comparision	14:24
	← lindenle left	14:24
	← calcifea left	14:24
	→ liamkeily joined	14:24
	→ nszceta joined	14:24
BenL	oh, on the repo itself?	14:25
	→ takamichi joined	14:25
armin_w	yeah its just files too	14:25
	← cff left	14:25
_ikke_	BenL: You can compare refs	14:25
	BenL: git show-ref	14:25
	And compare those	14:25
BenL	armin_w: right, so would work if they were at the same point, but if one is ahead of the other, that's trickier, right?	14:26
	_ikke_: how do I know the refs have not been doctored?	14:26
_ikke_	for each ref, check if the ref in the other repo contains that commit	14:27
	→ falconfox joined	14:27
	→ francisfish joined	14:28
	← correcto left	14:28
	→ chromano joined	14:28
	→ Exancillatus joined	14:29
	→ Dougie187 joined	14:29
	← liamkeily left	14:29
nszceta	Alright I have a practical problem... Me and my buddies are working on an application but we are all relatively new to Git and we are using Bitbucket. So far they clone the core repo and submit pull requests back to the core repo from their own repo. However, this is causing me (the maintainer of core) and the pull requests to trash the view of the git log and it is hard to understand WTF is going on anymore. How would you handle a branching	14:29
	workflow without millions of merges?	14:29
	→ wongon joined	14:29
	→ AaronMT joined	14:30
_ikke_	nszceta: !workflow	14:30
gitinfo	nszceta: Finding the right workflow for you is critical for the success of any SCM project. Git is very flexible with respect to workflow. See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.	14:30
	← MattStratton left	14:30
chromano	I'm trying to generate commit hashes and I'm following this format: commit %d\0%s % (len, body)... but the hash I get never matches the one in git log, any ideas?	14:31
	→ skorgon joined	14:31
	→ livingstn joined	14:33
	← salamanderrake left	14:34
	← Dougie187 left	14:35
_ikke_	http://stackoverflow/com/questions/1869885	14:35
ojacobson	_ikke_: now how did you manage that	14:35
_ikke_	manage what?	14:35
	← Advocation left	14:36
ojacobson	Mangling a URL like that	14:36
_ikke_	manually typing it	14:36
CharlesFinley	best tool for macs to resolve conflicts?	14:36
_ikke_	http://stackoverflow.com/questions/1869885	14:36
chromano	_ikke_: yeap, but like I said I didn't forget to wrap inside the header stuff	14:37
_ikke_	Are you sure you have the exact same data?	14:37
nszceta	I'm still trying to wrap my head around ff-only vs rebase	14:37
chromano	yes, checked twice against git show -s --pretty=raw	14:38
	← skorgon left	14:38
	→ strk joined	14:38
relipse	I have an issue, for some reason my coworker is making a commit and then a push, and it is addign 2 commits "Merging ...." and then the commit he specified, but it is not properly updating the site, is something wrong with his repo or is he not on the latest branch	14:38
_ikke_	nszceta: --ff-only is a merge option that aborts the merge if it would result in an actual merge commit	14:38
canton7-mac	nszceta, if an ff-only merge will work, trying to do a rebase will do an ff-only merge	14:38
strk	I've changed a remote's name manually and now "git branch -r" contains a lot of references to the old name	14:38
	→ marthinal_ joined	14:38
	→ hellopat joined	14:38
nszceta	hmmm	14:38
strk	prune with the old name doesn't work, how to "clean" that up ?	14:39
nszceta	thank you - this clears things up	14:39
	→ vkrest joined	14:39
	← chromano left	14:40
	← marthinal left	14:40
	← qsuscs left	14:40
	strk tried: rm -rf .git/refs/remotes/upstream/	14:40
strk	safe ?	14:41
	→ nZac joined	14:41
strk	(fixed branch -r)	14:41
_ikke_	yes, but not comprehensive	14:41
	→ Dougie187 joined	14:41
nszceta	So what I'm seeing is that the best way to minimize the overload of merges in the git log is to rebase	14:41
	marthinal_ → marthinal	14:41
strk	comprehensive ?	14:41
	← clarkfischer left	14:41
strk	_ikke_: you mean it's not enough to clean all refs to the old name ?	14:42
	→ King_Hual joined	14:42
	← cotton left	14:42
	→ leakybuckit joined	14:43
_ikke_	strk: the refs could also be stored in a pack file	14:44
	← vkrest left	14:44
	→ cotton joined	14:44
	← nerdfiles left	14:45
	→ alexlindenlevy joined	14:45
strk	would that affect use of the repository ?	14:45
	← lotsofcode left	14:45
strk	in any common use ?	14:46
	← dsdeiz left	14:47
	→ mikecmpbll joined	14:48
	→ lotsofcode joined	14:48
	→ pbujok joined	14:48
	← arvind_khadri left	14:48
	→ welly joined	14:49
	← welly left	14:49
	→ welly joined	14:49
	→ skorgon joined	14:52
	→ Karpuragauram joined	14:52
	← heroux left	14:52
	→ pduin_ joined	14:52
	← pduin_ left	14:53
	→ pduin joined	14:53
	→ sctt joined	14:54
	→ lindenle joined	14:55
	→ ahmedelgabri joined	14:56
	→ malkomalko joined	14:56
	← Karpuragauram left	14:56
	← ahmedelgabri left	14:56
	→ pantsman joined	14:56
	→ clarkfischer joined	14:57
	← ruskie left	14:57
	→ ahmedelgabri joined	14:57
	→ jrobichaud joined	14:57
	← welly left	14:58
	← sctt left	14:58
	→ holodoc joined	14:58
relipse	Does anyone know why this commit is getting packed in with my other commits: Merge branch 'master' of bitbucket.org:...	14:59
	← lindenle left	14:59
	→ liamkeily joined	14:59
	← skorgon left	14:59
relipse	not only is that 2nd commit getting packed in there but the 2 commits are not getting updated	14:59
	How do i grab the latest and greatest from origin master and discard all local changes	14:59
	→ faBimon joined	15:00
	→ Usama_ joined	15:00
	← faBimon left	15:00
moritz	git reset --hard origin/master	15:00
	→ faBimon joined	15:00
nszceta	^ what he said	15:00
	→ ahmedelg_ joined	15:01
	→ CKyle joined	15:02
	→ prometh joined	15:02
	← ahmedelgabri left	15:02
prometh	i'm looking to clean up my repo history... i have a bunch of simple readme commits that i'd like to merge	15:02
	they're way back in history.. not recent	15:03
_ikke_	Is this repo shared?	15:03
moritz	prometh: have you published that history?	15:03
prometh	it's published, but it's private	15:03
	i tried doing a rebase -i, but it tried to merge everything	15:03
	merge/squash	15:03
	i'm looking to squash the oldest 20 or so commits only	15:04
	← cjsarette left	15:04
moritz	rebase -i doesn't merge or squash anything, unless you tell it to	15:04
	oh, and modifying the root commit doesn't work out of the box	15:04
prometh	yeah, the oldest 20 after root	15:04
moritz	so leave the first commit alone, and only squash the 3rd into into the 2nd etc.	15:05
	← diegoviola left	15:05
	← hkon_ left	15:05
prometh	i tried that, but it somehow tried to squash everything	15:05
moritz	!repro please	15:05
gitinfo	Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session -- or, even better for complex issues, design a minimal case in which your problem can be reproduced, and share it with us. This will help immensely with troubleshooting.	15:05
	→ nedbat joined	15:06
prometh	well, i'm using sourcetree	15:06
	(gui)	15:06
	i guess i'll have to do this with a text editor, then	15:06
moritz	!gui applies :-)	15:06
gitinfo	Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.	15:06
	→ tbmiller joined	15:07
prometh	yeah, i understand.. i just thought that maybe i was doing something obviously wrong and could translate it to the gui myself	15:07
	← cmpera left	15:09
	→ m0viefreak joined	15:09
	→ welly joined	15:09
	→ Silmarilion joined	15:10
	← matanya left	15:10
	→ robarelli joined	15:11
	→ mizu_no_oto joined	15:11
	← kbart left	15:11
	← Exancillatus left	15:11
	← Macaveli left	15:11
	→ Exancillatus joined	15:11
	→ chrisshattuck joined	15:11
	← chrisshattuck left	15:11
	← leakybuckit left	15:12
	→ Macaveli joined	15:12
	← rchavik left	15:12
	→ jasonroyle joined	15:12
	→ shawndumas joined	15:12
	← Exancillatus left	15:12
	← cotton left	15:13
	→ Rylee joined	15:13
Rylee	Hey! Question: How would one list all commits on the master branch since a given commit?	15:14
	← wolfman left	15:14
	→ godgodgodgo joined	15:14
	← welly left	15:14
	→ asQuirre_ joined	15:14
mage_	git log givencommit..HEAD ?	15:15
offby1	git log GIVEN..master	15:15
	→ zumba_addict joined	15:16
Rylee	That excludes some commits seemingly at random	15:16
	[VPS] rylai@wisp:~/code/atheme $ git log --oneline cbdb528...HEAD \| tail -1	15:16
	85fe039 Atheme 7.0-beta3.	15:16
	→ Bass10 joined	15:16
Rylee	Shouldn't the last commit be cbdb528 and not 85fe039?	15:16
	← stodan left	15:17
_ikke_	No, the first one is excluded	15:17
	A..B is ^A B is everything reachable by B, but everything reachable by A excluded	15:17
	← armin_w left	15:17
jwmann	Hey _ikke_, did you get my messages before? :3	15:18
	→ danecando joined	15:18
_ikke_	jwmann: which?	15:19
	← Malice left	15:19
jwmann	Where I figured out the reason why it couldn't my 'secret key' ?	15:19
	couldn't find*	15:19
_ikke_	I haven't seen them	15:19
	→ tang^ joined	15:19
jwmann	I'll just re send ti	15:19
	_ikke_: I figured out why it wouldn't work.	15:19
	_ikke_: It's because I put something in the 'Comment' field.	15:20
_ikke_	ah ok	15:20
	And it wants to match that too, right?	15:20
jwmann	I suppose! :S	15:20
_ikke_	I guess so	15:20
jwmann	But after removing the comment field 'just for fun to see if it might work' and it worked	15:20
	I'm wondering, is there a way to check my tag to see if it was 'signed' properly?	15:21
	→ skorgon joined	15:21
jwmann	Because when it signed it said "You need a passphrase to unlock the secret key for user: "	15:21
prometh	moritz: so, to squash old commits, i would not use: git rebase -i HEAD~4 correct?	15:21
jwmann	Although the tag was still made	15:21
	→ becks__ joined	15:22
	← strk left	15:22
Rylee	Hmm, alright	15:22
	thanks, mage_, offby1, _ikke_	15:22
	→ Exancillatus joined	15:22
	→ welly joined	15:23
	← welly left	15:23
	→ BenL1 joined	15:23
	← jasonroyle left	15:23
	→ aslaen joined	15:24
prometh	i'm looking to clean up my repo history... i have a bunch of simple readme commits way back in history that i'd like to squash	15:24
	← BenL left	15:24
	→ rq60 joined	15:24
	← faBimon left	15:24
	← papegaaij left	15:25
_ikke_	jwmann: git tag -v	15:25
	← Bass10 left	15:25
	→ ErKa joined	15:25
	← Davey left	15:26
	→ aberrant joined	15:26
	← nnyk_ left	15:26
jwmann	Hmmm nothing showed up :S	15:26
_ikke_	You have to pass the tagname iirc	15:27
jwmann	git describe shows v0.1.0	15:27
moritz	prometh: if HEAD~4 takes you bag far enough, yes	15:27
	← pikaren left	15:27
prometh	moritz: it only takes me back 4 from current	15:27
moritz	prometh: yes, that's the 4 in there	15:27
_ikke_	yes	15:27
prometh	i want to go back like 100 to 1005	15:27
	i don't want those 4	15:27
	← jargon left	15:27
moritz	then don't type 4 :-)	15:27
_ikke_	prometh: You can also pass in hashes	15:27
prometh	i don't want HEAD-4... i want ROOT-20	15:27
jwmann	_ikke_: Oh, now I see more info on the tag. gpg says "Good signarture" so I guess it worked xD	15:27
_ikke_	prometh: It doesn't work that way	15:27
prometh	i have to list all the hashes?	15:28
	or a hash range ?	15:28
moritz	no	15:28
_ikke_	prometh: The parent of the last hash	15:28
	you want to change	15:28
moritz	just the hash from which the rebase is to start	15:28
prometh	ah, okay thanks	15:28
	both	15:28
	← skorgon left	15:28
_ikke_	prometh: Do you have merges in between?	15:28
	Because merges can cause problems	15:28
prometh	nope	15:28
_ikke_	ok	15:28
prometh	thanks for teh heads up though	15:28
	← shinnya left	15:28
	→ macben joined	15:29
	→ Lunatrius joined	15:29
	← alexlindenlevy left	15:30
	→ welly joined	15:31
	← welly left	15:31
	← macben left	15:31
	← facundor left	15:35
	→ daivyk joined	15:36
	← mvelioglu left	15:36
	→ Guest6850 joined	15:37
	← lotsofcode left	15:38
Guest6850	Hi, my group hosts its stuff on GitHub. I had a question about the number of repos. Is it better to have a repo for each project, or one repo with lots of subfolders?	15:38
	→ Kayra joined	15:38
prometh	_ikke_ moritz: sorry for the delay.. i used the has of my root commit... now all commits are listed.. i'd like to pick the 20th oldest commit as the place to commit all older ones into	15:39
moritz	prometh: so do that	15:39
	→ Usama joined	15:40
	→ vkrest joined	15:40
prometh	having a little trouble figuring it out though	15:40
	← Usama left	15:40
	→ Usama joined	15:41
	← Usama left	15:41
	→ lotsofcode joined	15:41
prometh	i'll post a screenshot	15:41
	→ friesk joined	15:42
	→ Usama joined	15:42
	← Usama_ left	15:42
	← Usama left	15:43
	← robarelli left	15:43
	→ Usama joined	15:43
	→ ruskie joined	15:44
	← Usama left	15:44
moritz	no need to screen-shot; just copy&paste the text into a nopaste service	15:44
	→ Usama joined	15:44
	→ ebetancourt joined	15:44
	← vkrest left	15:44
	→ axl_ joined	15:45
	← Usama left	15:45
	→ Usama joined	15:45
	← Usama left	15:45
prometh	oh yeah... well, already got a screenshot	15:46
	http://snag.gy/eBwBl.jpg	15:46
	← Exancillatus left	15:46
prometh	i'd like to squash all from the top into v0.1.0	15:46
	→ Usama joined	15:47
prometh	do i "pick" v0.1.0 ?	15:47
	← _Qman left	15:47
	→ obert joined	15:47
prometh	or do i pick the top "no message" ?	15:47
	→ el3ctron2 joined	15:48
moritz	you pick those on that top that you want to leave unmodified	15:48
	and then 'squash' or 'fixup' those that you want to combine	15:48
prometh	oh i see	15:49
	→ hakril joined	15:49
	← Norm_ left	15:50
	← liamkeily left	15:50
prometh	so, fixup will let me set a new commit message?	15:50
	→ welly joined	15:50
ashooby	no, fixup discards the commit message	15:50
prometh	actaully.. i get the chance to change the commit message afterwards, anyway, right ?	15:50
	← welly left	15:50
ashooby	if you use squash or edit	15:51
	← karstensrage left	15:51
prometh	what if i use fixup on all of them ?	15:51
	it will squash and remove commit messages, making the commit message editing easier, right ?	15:51
	← vurma left	15:52
	→ tadasZ joined	15:52
	← thevishy left	15:52
YoungFrog	you could "edit" the first one, and "fixup" all others ?	15:53
moritz	you only get a chance to change the commit message if you use 'squash', or 'reword' instead of 'pick' in the baser commit	15:53
	→ bwreilly joined	15:53
	→ welly joined	15:53
prometh	"Cannot 'fixup' without a previous commit"	15:53
	← wongon left	15:53
prometh	"edit" will still do a squash, right?	15:54
	→ githogori joined	15:54
	→ robarelli joined	15:54
prometh	so... squash first one, fixup rest ?	15:54
ashooby	if I were you I would do pick for the topmost, squash for the v0.1.0 commit and fixup for the rest	15:54
prometh	or edit first one, fixup rest?	15:54
YoungFrog	I meant "reword", if you only want to reword the commit msg.	15:54
	prometh: no, "edit" or "reword" the first, fixup rest	15:54
prometh	ok	15:55
	→ heroux joined	15:55
	← chepre left	15:55
	← jrobichaud left	15:55
	← githogori left	15:56
prometh	taking a long time to --abort	15:56
	← zxd left	15:56
	← FortySix2 left	15:56
	→ infynyxx joined	15:57
prometh	ok, now where will this be placed in history ?	15:58
	→ chrisshattuck joined	15:58
	→ ambodi_ joined	15:58
prometh	if i give it a commit message of "v0.1.0", will it appear higher than v0.4.5 ?	15:58
	or, will it date it where my "edit" was ?	15:59
YoungFrog	err the order is not dependent on the commit message	15:59
prometh	i know, but the date	15:59
YoungFrog	I think the original date is preserved	15:59
	not sure thoug	15:59
prometh	i've done this before, but only from head down	15:59
	not root up	15:59
zumba_addict	should I run git stash pop first or git merge master into my feature branch?	16:00
prometh	hmm, actauly i don't think this worked	16:00
	it says only 1 file was changed	16:00
	← M00sL0gB0t left	16:01
	→ M00sL0gB0t joined	16:01
ashooby	if you used "edit" or "reword" it will only have the files changed in that specific commit	16:01
	← ehmatthes left	16:01
	→ skorgon joined	16:02
	→ FortySix2 joined	16:02
	← ambodi left	16:02
ashooby	while you're editing the commit anyway	16:02
	→ el3ctron4 joined	16:02
	← FortySix2 left	16:02
	← roelmonnens left	16:03
	← rath left	16:03
prometh	ashooby: http://pastebin.com/Nevm64YZ	16:03
	oh, it won't list all changes from all squashed commits ?	16:03
	← Suprano left	16:03
	← chrisshattuck left	16:04
	→ zizzl joined	16:04
zizzl	Hey guys working on the same app on heroku from two different computers.. cloned it on my second one and worked on it, now I'm back at the original one, what's the best method to update it with the new stuff?	16:04
ashooby	if you do edit, it will stop at that commit and allow you to edit that commit's commit message	16:05
prometh	yes, that's what i'd like	16:05
	→ cmpera joined	16:05
ashooby	if you do fixup, it will just get rid of the commit message	16:05
prometh	but it will still squash everything and preserve the date ?	16:05
ashooby	if you do squash, it will show all the commit messages and allow you to edit them all into one message as you see fit	16:05
PerlJam	zizzl: you can commit the changes to a common "central" repo like github. You can just add the other repo as a remote and fetch the changes directly.	16:05
	← Silmarilion left	16:05
prometh	yeah, i'm replacing all messages	16:05
	← bartman` left	16:05
prometh	the main thing i'm worried about is if it will still squash everything and preserve the date	16:06
	→ chrisshattuck joined	16:06
PerlJam	zizzl: I find it hard to answer "best method" questions because there's always some hidden constraints that no one mentions.	16:06
zizzl	PerlJam well I'm using heroku as that repo	16:06
	← HelperW_ left	16:06
zizzl	so should I just clone it back over here?	16:06
	→ kukks joined	16:06
ashooby	it will preserve the date	16:07
PerlJam	zizzl: do the simplest thing that could possibly work :)	16:07
	→ HelperW joined	16:07
zizzl	PerlJam well I'm new to this so I'm asking	16:07
	basically easiest way	16:07
	in your opinion	16:07
	← aChatir left	16:07
prometh	ashooby: and squash everyting ?	16:07
ashooby	yes, it should	16:07
prometh	ok, thank you	16:07
	← Guest6850 left	16:07
	← polyzen left	16:08
	← mishok13 left	16:08
	← cbreak-work left	16:08
ashooby	anyways, you should just do it, and if it doesn't work then that's what the reflog is for :P	16:08
	rather, if it doesn't do quite what you want	16:08
	← skorgon left	16:09
prometh	never heard of reflog :)	16:09
	i may end up learning.. but i kinda hope not	16:09
	my tags are interfering with hiding history.. gotta remove them	16:09
	→ cbreak-work joined	16:09
	→ senj joined	16:10
	→ liamkeily joined	16:10
	← LionKing left	16:10
	← dougsland left	16:11
	← john2x left	16:11
prometh	fuck	16:12
	→ Strogg joined	16:12
prometh	it created a separate branch of some kind that i cannot get back to	16:12
Strogg	'lo 'lo	16:12
	← liamkeily left	16:12
	← WaReZ left	16:12
prometh	because of my tags, it did not rewrite history completely	16:12
	→ liamkeily joined	16:12
prometh	it looked like a branch in sourcetree (used to to picture what was going on)	16:12
	→ _tpavel joined	16:12
Strogg	When you use git log with --pretty:format, %d shows you all of the refnames for each commit. Is there a way to make it not show tags?	16:12
	← konadr left	16:13
prometh	i switched back to master and removed the tags... now i can't get back to my new squashed shit	16:13
Strogg	I tried --no-decorate, but that doesn't work	16:13
	→ lavaman joined	16:13
	← cbreak-work left	16:13
	← tadasZ left	16:13
	← liamkeily left	16:14
prometh	ashooby	16:14
	→ t0r joined	16:14
	→ liamkeily joined	16:14
ashooby	git reflog	16:14
	→ cbreak-work joined	16:14
prometh	k	16:14
	← Parnic left	16:15
	→ tadasZ joined	16:15
	← bdavenport left	16:15
	→ dougsland joined	16:15
	← dougsland left	16:15
	→ dougsland joined	16:15
ashooby	you can use the HEAD@{#} annotations to do stuff like git reset HEAD@{3}	16:15
	← liamkeily left	16:15
ashooby	and that will put you back in the state at that point	16:15
	though, you might want git reset --hard	16:15
	→ liamkeily joined	16:15
prometh	ugh	16:15
	← malkomalko left	16:15
prometh	i just did some checkouts n stuff	16:15
ashooby	i'm actually not sure how reflog deals with tags	16:15
prometh	trying to find my commit	16:15
	← cbreak-work left	16:16
ashooby	or if it does	16:16
	→ bdavenport joined	16:17
	← ThomasLocke left	16:17
	← liamkeily left	16:18
ashooby	also, when i rebase it has no problems with blowing away tags	16:18
	→ liamkeily joined	16:18
prometh	so, hwy did mine not work?	16:18
	→ AciD` joined	16:18
AciD`	hi	16:18
gitinfo	AciD`: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	16:18
prometh	how do i get out of this reflog?	16:18
	← Anticom left	16:18
ashooby	q	16:18
prometh	contrl+c won't let me exit	16:18
	thank you	16:18
AciD`	!backup	16:18
gitinfo	Worried about your data while trying out stuff in your repo? To back up commit history on all branches/tags: `git clone --mirror`. To backup everything, including work tree and staging area: `tar cf repo-backup.tar repodir`. Or do your experiment in a throwaway clone instead. See also http://sethrobertson.github.com/GitBestPractices/#backups	16:18
	→ Exancillatus joined	16:19
	← rob_w left	16:19
	→ joelm joined	16:19
prometh	ashooby: reflog only lists 15 heads	16:19
ashooby	git reflog --all	16:20
	oh wait	16:20
	no	16:20
prometh	http://pastebin.com/J4vj9sTL	16:20
	→ Parnic joined	16:21
	← nZac left	16:21
ashooby	so, if you do something like `git reset --hard HEAD@{15} then it will put you back at the original state of the repo right after you cloned it	16:21
	this must be a new repo	16:21
prometh	nope	16:22
	→ ngladitz joined	16:22
	← ngladitz left	16:22
	→ ngladitz joined	16:22
prometh	but i'm squashing my initial commits	16:22
	← liamkeily left	16:22
prometh	from back when the repo was new	16:22
	oh... i'm working wiht a fresh copy	16:22
	as my other has changes in it and didn't wanna risk losing a stash	16:23
	→ aubreypwd joined	16:23
ashooby	not a bad idea	16:23
prometh	why would epxlain why there's only 15 heads i guess... i didn't really know what a head was	16:23
	→ h0bbit_ joined	16:24
ashooby	the HEAD is where you currently are in the tree	16:24
prometh	whcih*	16:24
	which*	16:24
	← gittish left	16:24
ashooby	so HEAD{1} is where the head was 1 step ago	16:24
prometh	hmm, that's what i originally thought... then why aren't there heads available from before my clone?	16:24
	what is a "tree" then ?	16:24
	the whole repo tree, or a local clone	16:25
Eugene	reflog is a log of the LOCAL state of your refs; it does not have historical information about other repos	16:25
prometh	local state, yeah, ok	16:25
Eugene	If you want history-based refs you're looking for ^ ~ and friends - !treeish	16:26
gitinfo	A tree-ish is something that looks like a tree. Read 'man gitrevisions' and http://git-scm.com/book/en/Git-Tools-Revision-Selection	16:26
prometh	ok, so i need to reflog to the start of my local state	16:26
	→ FleaFlicker joined	16:26
	← FleaFlicker left	16:26
prometh	hmm	16:27
	ok, i did: git reset --hard HEAD@{15}	16:27
	→ Bass10 joined	16:27
prometh	then i try to start a new rebase.. but it says that a rebase dir already exists	16:27
	my tags have not returned	16:27
Eugene	What are you trying to do/	16:28
AciD`	noob question : I started working on a project few months ago (located in /old/path/to/proj), using git to manages changes. I now want to totally reorganize my files in a new directory structure like /new/path/to/proj (cherry picking each file to mv it to a specific new directory, then later renaming it). How should I proceed if I want to keep the history for each file ?	16:28
	→ ezakimak_work joined	16:28
	← HelperW left	16:28
AciD`	I only have 5 files in git, so I can do that manually	16:28
prometh	Eugene: undo a rebase -i that didn't appear to do antyhing at all	16:28
Eugene	Ahhhh I see.	16:29
	→ DolphinDream joined	16:29
DolphinDream	howdy	16:29
prometh	it did, but when my tags interferred, i switched to master to remove them... then i couldn't get back to my squash	16:29
ashooby	if the rebase directory already exists then do git rebase --abort to get rid of it	16:29
	← pheaver left	16:29
prometh	git rebase -i b08821e5817155be50527e15909b3f8585f1ba7f	16:29
	warning: refname 'HEAD' is ambiguous.	16:29
	warning: refname 'HEAD' is ambiguous.	16:29
	warning: refname 'HEAD' is ambiguous.	16:29
	wtf is this shit	16:30
DolphinDream	what is the recommended comment (first line) length ? (following the rule first line:summary, second line: empty, third line and below: details)	16:30
osse	DolphinDream: 50 chars typically	16:30
PerlJam	DolphinDream: 60 chars I think.	16:30
osse	oh	16:30
DolphinDream	i'll take 55	16:30
	:)	16:30
PerlJam	heh	16:31
osse	54.9 then	16:31
	→ calcifea joined	16:31
DolphinDream	*sqrt(pi)/phi	16:31
prometh	ashooby: ok, i aborted it.. but now i get the above weirdness	16:31
ashooby	if you use vim, it will complain at you when you go past the recommended length	16:31
	← DoniRS left	16:31
ashooby	though, I don't actually know what it is	16:31
PerlJam	If you have an appropriate .vimrc or .emacsrc for your git commit messages, it'll color the text different when you hit the "limit"	16:31
prometh	is my repo fucked ?	16:31
	→ nicedice joined	16:32
Eugene	It sounds like there's something odd going on, yes.	16:32
DolphinDream	do you have to explicitely break comment lines to 72 or something ? (if you're not using vim;)	16:32
	→ Kicer86 joined	16:32
osse	DolphinDream: you "have to" do that no matter what editor you use	16:32
Eugene	Have you done any actual work in this repo? Or would a fresh `git clone` be viable?	16:32
	← h0bbit_ left	16:32
prometh	yes, i have work in it	16:32
	→ juliohm joined	16:32
prometh	err, wait... no, this is a fresh clone	16:33
	→ nZac joined	16:33
prometh	but i've force pushed my rebase	16:33
Eugene	My advice: rm -rf	16:33
	Oh. I see.	16:33
	→ alan_ joined	16:33
Eugene	Well hilarity ensues!	16:33
DolphinDream	osse, i assumed vim does it for u. i sometimes use smartgit	16:33
Eugene	Do you have any /other/ clones running around?	16:33
prometh	and now ashooby isn't responding	16:33
	← LucasKA left	16:33
	alan_ → Guest31292	16:33
	← ezakimak_work left	16:33
ashooby	well I was going to say what Eugene did :P	16:33
prometh	then why did you tell me all would be fine?	16:34
ashooby	I've never actually seen the head is ambiguous comment so I don't know what to make of it	16:34
prometh	when it's busted now	16:34
	Eugene didn't tell you anything :v	16:34
prometh	seemingly... i mean, my files are intact	16:34
	ashooby did, though	16:34
osse	DolphinDream: vim's default setup for commit messages is to automatically insert a linebreak whenver you type and you pass 72 characters. But if you paste something or do other edits then you have to fix it yourself	16:34
ashooby	well, if you are only doing it in a fresh clone then your original repo will be find	16:34
prometh	i'm not blaming, i just wanna figur eit out	16:34
Guest31292	i'm on ubuntu 12.04 and i have git version 1.7.9.5. according to git's website the latest version is 1.8.5.3	16:34
	is there a reason apt-get doesn't have that version yet	16:34
prometh	ashooby: i did a force push	16:34
ashooby	but if you pushed it before you were sure it was right...	16:34
DolphinDream	osse, gotcha. why this 72 rule exactly ?	16:35
prometh	i was sure	16:35
	← jnavila left	16:35
osse	prometh: my guess is that somehow a ref (branch/tag/...) named HEAD was created, and that screws things up	16:35
ashooby	if the problem is the tags, then I don't really know anything about tags	16:35
prometh	i asked questions which reassured it	16:35
	← chrisshattuck left	16:35
Nevik	DolphinDream: 80cars plus tab	16:35
	chars* even	16:35
osse	DolphinDream: 80 - a little bit to allow for indentation in the output of git/log and for when you quote stuff in emails	16:35
	I think	16:35
prometh	there're no branches	16:35
	→ chrisshattuck joined	16:35
Nevik	DolphinDream: so as to make it displayable in an 80col terminal	16:35
prometh	just checked on github	16:35
ashooby	oh wait	16:35
Eugene	DolphinDream - because there are a lot of people still stuck in the 1970s and don't understand how to have reflowed text. Make your commit message lines as long as you goddamn well please IMO. Try to keep the first line short(treat it as a "Subject" a la email"), but to hell with it on the rest.	16:35
ashooby	prometh: do you have a branch named HEAD?	16:35
DolphinDream	Eugene, :) oh boy	16:36
prometh	locally, yes	16:36
	← ambodi_ left	16:36
ashooby	ok, get rid of that	16:36
	or rename it	16:36
	git branch -m HEAD newname	16:36
	to rename	16:36
	→ thiago joined	16:36
ashooby	that's why you're getting that warning	16:36
	→ ambodi joined	16:36
DolphinDream	Eugene, i;d vote for you as the presient :)	16:36
	president	16:36
	→ blinkdes_ joined	16:36
Eugene	\o/	16:37
prometh	delete it remotely or locally ?	16:37
	ashooby	16:37
ashooby	both	16:37
	← swaechter left	16:37
osse	prometh: you should be able to specify refs/HEAD or refs/heads/HEAD to resolve the ambiguity. But if the error comes from the internals of a git command then there's nothing you can do	16:37
DolphinDream	Nevik, how many are stuck ? :) can we stash them ?	16:37
ashooby	HEAD is used by git to refer to your current branch, so having a branch called HEAD is a little screwy	16:37
	← pantsman left	16:37
	← pduin left	16:38
Nevik	DolphinDream: it's just a convention. many are of Eugene's opinion by now and dont stick to it too much. I personally do because my editor is already set up that way.	16:38
DolphinDream	Eugene, .. but for the first line what's your vote for the length ?	16:38
	← Raging_Hog left	16:38
Eugene	DolphinDream - "about a sentence"	16:38
ashooby	^	16:38
Nevik	DolphinDream: i think it's a good idea not to let the lines get too long, so they're easily readable in editors/viewers that dont have line wrapping on. but other than that, it's up to your own judgement	16:38
prometh	ok, done	16:38
DolphinDream	Nevik, I do still stick to it but only because i typically split my editor vertically in two.. so each one has about 80 or so	16:38
	← ghostlines left	16:39
Eugene	Almost all viewers have line wrapping; it's reflowing at word breaks that's the issue.	16:39
DolphinDream	Eugene, so 3 words :)	16:39
	→ ezakimak_work joined	16:39
ashooby	at my job we use the first line to say somethign like "fix issue xx"	16:39
DolphinDream	I think short first line is important also for the graph ..	16:39
tang^	assume that somebody will read that first line on a phone at some point... short and sweet	16:39
ashooby	so most of ours actually are only 3 words :D	16:40
	← blinkdesign left	16:40
Eugene	:Frobnaz the foobar by bazbaring" is fairly typical of me.	16:40
	→ rchansen joined	16:40
	← lb1a left	16:40
prometh	yeah, it's not working correctly ashooby	16:40
	← Guest31292 left	16:40
	→ vkrest joined	16:41
ashooby	what exactly did you push to the remote?	16:41
	→ Nizumzen joined	16:41
	← ambodi left	16:41
DolphinDream	ok.. tweeter style messages :) except only 55 characters	16:41
osse	I'm a bit skeptical about referencing issue numbers etc in commit messages without any other info. What if you change bug trackers? Stuff like that	16:41
ashooby	or rather, what is the actual problem, I'm not sure I completely understand	16:41
	→ mani2399 joined	16:41
	← LeMike left	16:41
ashooby	osse: then we'll have to look at the rest of the commit to see what it actually was I guess	16:41
	and then...	16:42
	→ pheaver joined	16:42
	← takamichi left	16:42
ashooby	refer to them as "<new bugtracker> issue xx" probably	16:42
	→ macben joined	16:42
prometh	ashooby: here's what sourcetree says has happend: http://snag.gy/hFQsD.jpg	16:42
tang^	osse: I just ran into that with some bitbucket/jira testing I was doing	16:42
	← noname001__ left	16:42
	← chrisshattuck left	16:42
tang^	imported bitbucket issues into a jira project and the numbers changed	16:43
	→ RandalSchwartz joined	16:43
RandalSchwartz	why does git-pull omit the git-fetch --prune switch?	16:43
ashooby	so, HEAD is a tag?	16:43
prometh	i guess so... i never added it	16:43
RandalSchwartz	it has nearly everything else fetch does	16:43
prometh	but why is it up there... why did it not squash antying ?	16:43
RandalSchwartz	am I missing something?	16:43
Eugene	Booze	16:44
	→ kesha joined	16:44
	→ diegoviola joined	16:44
ashooby	I'm not entirely sure what is happening in that picture prometh	16:44
	maybe someone more knowledgable can chime in	16:44
	it looks like you have a merge commit	16:45
	that is tagged HEAD and also has no message?	16:45
Eugene	RandalSchwartz - I'm guessing some historical reason, but I can't be arsed to hunt through the mailing lists for it	16:45
	← vkrest left	16:45
	← user258467 left	16:45
	→ chrisshattuck joined	16:45
prometh	this is what i did: http://pastebin.com/Nevm64YZ	16:45
	and this is what happend: http://snag.gy/hFQsD.jpg	16:45
	anyone?	16:46
	oh yeah, i then amended the commit message to be "v0.1.0" and did a force push	16:46
	and that screenshot shows me what happend.... looks like nothing.	16:46
	→ dreiss joined	16:47
	← subhojit777 left	16:47
jrnieder	RandalSchwartz: looks like it was documented in an ifndef::git-pull block since day one	16:47
	← mikecmpbll left	16:48
	→ thevishy joined	16:48
	→ skorgon joined	16:48
jrnieder	RandalSchwartz: for no obvious reason. probably worth trying a patch to take it out of that block?	16:48
	← heroux left	16:48
osse	perhaps in case the current branch's tracking branch is removed from the remote?	16:48
jrnieder	I don't see how that would be harmful	16:48
osse	oooh, it actually works but isn't present in the docs?	16:48
	← macben left	16:48
osse	jrnieder: it would be harmful if pull then tries to merge	16:48
	I guess	16:48
prometh	Eugene: can you help me ?	16:49
jrnieder	if I'm on master and origin/master goes missing, 'git pull' should fail whether it wants to remove the remote-tracking branch or not	16:49
Eugene	prometh - do you have shell access to your remote repo?	16:49
prometh	maybe, i dunno	16:49
	it's on github	16:49
ashooby	prometh it might help if you had a log of what you actually typed into the terminal	16:49
Eugene	Then no.	16:49
	Any other clones of it handy at all?	16:49
ashooby	to get your repo into this state	16:49
	→ heroux joined	16:49
prometh	it rebase -i b08821e5817155be50527e15909b3f8585f1ba7f	16:50
	git commit --amend	16:50
jrnieder	s\|origin/master\|origin's master\| in my example	16:50
	← heroux left	16:50
prometh	then i did a force push to master.. using sourcetree, though	16:50
	← paulweb515 left	16:50
	→ blizzow joined	16:50
	→ starkhalo joined	16:50
	→ misterpink joined	16:50
prometh	looking at sourcetree's log.. it did this: git -c diff.mnemonicprefix=false -c core.quotepath=false -c credential.helper=sourcetree push -v --force origin master:master	16:51
	→ ehmatthes joined	16:51
	← g105b left	16:51
	→ pheanex joined	16:51
	← blinkdes_ left	16:51
	→ huahax joined	16:51
ashooby	link to the github?	16:52
	← axl_ left	16:52
prometh	https://github.com/stevenvachon/can-boilerplate	16:52
	← dreiss left	16:52
	→ blinkdesign joined	16:52
	→ pen joined	16:53
	→ wrouesnel joined	16:53
	← cagedwisdom left	16:54
	← bwreilly left	16:54
	→ m8 joined	16:55
	← skorgon left	16:55
prometh	i guess github won't let us rewrite ancient history	16:55
	→ axl_ joined	16:55
ashooby	the github looks fine	16:55
prometh	yes, the rebase appears to have done nothing at all	16:56
	how can i rewrite ancient history?	16:56
ashooby	just a second, let me look at it	16:56
	← axl_ left	16:56
	→ lindenle joined	16:56
	← cmpera left	16:57
	← tbrock left	16:58
	→ blindingdawn joined	16:58
	← Nizumzen left	16:58
pheanex	How would you use git log --format to show only the title(first line) of the commit message. "%s" gives me more than that. => further parsing needed. Is there a way to do this in git only?	16:58
	← zumba_addict left	16:58
	→ chepre joined	16:58
	→ cmpera joined	16:59
	← rchansen left	16:59
osse	pheanex: strange. %s is supposed to only be the first line. Does it give the same output if you use %B ?	16:59
	← francisfish left	17:00
	← intellix left	17:00
	→ calamari joined	17:00
	→ LeMike joined	17:00
	← baba left	17:00
prometh	ashooby: anything?	17:01
	← lindenle left	17:01
	← kukks left	17:01
	← m0ltar left	17:01
	→ nathancahill joined	17:01
	→ maestrojed joined	17:01
pheanex	@osse %s seems to be the same like %B withouth the newlines	17:02
prometh	k, well, i guess that means you don't know	17:02
	prometh needs lunch	17:02
osse	prometh: hmm, do you have an empty line between the first line and the rest?	17:02
	→ mikecmpbll joined	17:03
	→ qsuscs joined	17:04
pheanex	@osse %s hast no empty lines at all and %B has one at the end but not in between. btw git version 1.8.5.3	17:04
calamari	had a guy that claims he committed his code to his branch. then he pushed the branch to the remote. he then did a pull, and said his commit was gone. I checked and it was found as a dangling commit, and we recovered it. how could this have happened?	17:04
	→ zumba_addict joined	17:04
	← pheaver left	17:05
calamari	(the pul was the next day)	17:05
osse	pheanex: no I mean the actual log messages themselves (as with a plain git log). Do they have an empty line between the first and the last?	17:05
	→ afhammad joined	17:05
shruggar	calamari, how many people on your team?	17:05
	→ gnu-man joined	17:05
	← juliohm left	17:06
	← ErKa left	17:06
	← senj left	17:06
	← dougsland left	17:06
	→ rjen joined	17:06
pheanex	osse: plain "git log" without the pretty/format shows it the right way, meaning commit <id> newline author: <author> newline date: <date> newline newline <title>newline<body>newline newline	17:07
	→ imachuchu joined	17:07
Nevik	calamari: check his reflog to find out? ;)	17:07
osse	pheanex: usually there's two newlines between the title and the body. Maybe Git thinks everything before the first empty line is the title	17:08
jast	pheanex: %s and friends rely on the assumption (and common convention in git) that the first line is followed by a blank line	17:08
	i.e. the first 'paragraph' is the subject	17:08
	→ m0ltar joined	17:08
calamari	shruggar: about 10	17:08
	Nevik: will do	17:09
pheanex	osse: i guess that solves the problem, when committing we dont use a blank line to separate title with body	17:09
jast	and that's all from me. on my way into the internet-free zone... :)	17:09
pheanex	osse: thanks for your help	17:09
	→ SocalNick joined	17:09
shruggar	calamari, I had the same suggestion as Nevik, but 10 people might be a small enough number to say "have everyone check their reflog to find out"	17:09
	→ psyl0n joined	17:10
shruggar	because it could be he pushed, then someone else forced a push to clobber it	17:10
	← alorence left	17:10
shruggar	if you're not enforcing forward-only	17:10
	← cmpera left	17:10
Nevik	however	17:10
	a pull would then have told him	17:10
	← nZac left	17:10
calamari	what will a force push look like?	17:10
Nevik	that it's non-ff	17:10
	calamari: what do you mean	17:10
	← mizu_no_oto left	17:10
	→ FatDarrel joined	17:11
calamari	how will I know from looking at reflog output that someone did a force push	17:11
	→ devbug joined	17:11
	→ RogerWS_ joined	17:11
Nevik	you wont. you would be able to tell if history-rewriting happened before. you'd most especially probably find the action on $claimant's reflog that shows him skipping away from his commit	17:12
shruggar	calamari, check not for the presence of a forced push, but for the presence of a command which would move towards one (such as a hard reset)	17:12
Nevik	probably resetting his branch	17:12
	calamari: basically, a pull cannot lose his commit	17:12
	if someone lost his commit by force-pushing, pull would have merged it back in for him	17:12
calamari	so it will say hard reset?	17:12
	→ cmpera joined	17:12
shruggar	iirc, some git commands dangerously suggest resetting in the event of an unknown problem	17:12
Nevik	so he did some other operation of his own (probably a hard reset or plumbing commands) that lost his commit	17:13
	← RogerWS left	17:13
	← pheanex left	17:13
	← el3ctron2 left	17:13
	→ halu joined	17:14
shruggar	and if the reflog is empty or short, it may mean he deleted and re-created the branch (I've seen people around here do that when getting changes for branches they hadn't intended to modify)	17:14
	← el3ctron_ left	17:14
	← el3ctron4 left	17:14
Nevik	calamari: to summarize, there's no 100% sure way to find out what happened	17:14
	but you might find indications. and it's most probably his own fault	17:14
	← RandalSchwartz left	17:15
	← solker left	17:15
Nevik	calamari: another thought: if he uses the command-line, check his command history	17:15
	→ kwstas joined	17:15
	→ bwreilly joined	17:15
	→ wongon joined	17:15
	← bwreilly left	17:16
	← mlechner left	17:16
	→ bwreilly joined	17:16
	← pen left	17:16
calamari	"reset to origin" is that anything to worry about?	17:16
	→ ncthom91 joined	17:17
Nevik	calamari: what's the exact line?	17:17
kwstas	hello there! I'm have a remote branch named 'mybranch' and a local one with the same name and configured to track the remote. I know that the remote has 3 additional commits. Why when i call "git fetch remote mybranch" the origin/mybranch does not update?	17:18
	← Sly1986 left	17:18
	→ nZac joined	17:18
calamari	HEAD@{11}: branch: Reset to origin/ded_v2_addInspector	17:18
kwstas	instead i need to fetch all remotes 'git fetch origin' to get the new commits	17:18
Nevik	kwstas: !fetch4	17:18
gitinfo	kwstas: [!fetchfour] We recommend against using 'git fetch/pull <remote> <refspec>' (i.e. with branch argument), because it doesn't update the <remote>/<branch> ref. The easy way to fetch things properly is to get everything: 'git fetch' or 'git pull' are sufficient if you have one remote; otherwise we recommend 'git fetch <remote>' (plus 'git merge <remote>/<branch>' if you wanted to pull/merge).	17:18
calamari	that was the line right before his commit in the reflog	17:18
	← Voyage left	17:18
	← freimatz left	17:18
Vampire0	calamari, origin is just the name of your configured remote in your repository	17:19
Nevik	calamari: before his commit? so is his commit @12 or @10?	17:19
	→ kukks joined	17:19
	← kukks left	17:19
	→ kukks joined	17:19
kwstas	gitinfo: but what if I want to update only a specific branch and not everything?	17:19
Nevik	kwstas: why do you not want to update everything?	17:20
	they're just remote branches	17:20
	← zumba_addict left	17:20
calamari	his commit is 12	17:20
	← mani2399 left	17:21
	→ beyondcreed joined	17:21
	← beyondcreed left	17:21
	→ beyondcreed joined	17:21
Nevik	calamari: then the reset came after (the reflog counts from now backward in time), so yes, that reset is probably what lost his commit	17:21
kwstas	Nevik: For no specific reason :)	17:21
	← m0ltar left	17:21
Nevik	kwstas: !fetchfourwhy	17:21
	→ fire joined	17:21
Nevik	argh what was that trigger again	17:21
	kwstas: !fetch4why	17:21
gitinfo	kwstas: The four-word version of git-fetch doesn't update remote-tracking branches and tags. It fetches into FETCH_HEAD only; you probably don't want to have to deal with that. Pull in four-word form automatically uses FETCH_HEAD, but your remote-tracking will still be outdated. Even updating them manually with 'git fetch origin master:remotes/origin/master' (clunky) still doesn't update tags.	17:21
Nevik	yay consistency	17:21
	→ paulweb515 joined	17:22
	← blinkdesign left	17:22
	← LeMike left	17:22
Nevik	kwstas: tl;dr, if you want to update only that one branch, you have to use pull or enter the full refspec	17:22
	← TomyWork left	17:22
Nevik	(of course, pull will merge your local into the remote as well)	17:22
kwstas	Nevik: :) ok, thanks. I thought that I was doing something wrong, because many online tutorial advertise the command "git fetch <remote> <branch>" as the appropriate to update a branch	17:23
Nevik	oh wait 4-word pull doesnt update remote tracking either	17:24
	kwstas: that's why we have a trigger for it :P	17:24
	you're not the first person to have trouble with this	17:24
calamari	Nevik: http://pastebin.com/SWeA282P I marked the commit that was later lost with asterisks	17:24
	← vince_prignano left	17:25
calamari	Nevik: ah so the top line in relog is most recent?	17:25
kwstas	Nevik: nice! thanks again. I can now finally sleep in peace :)	17:25
Nevik	calamari: yes	17:25
	→ svm_invictvs joined	17:25
Nevik	kwstas: no problem ;)	17:25
svm_invictvs	how do you specify the default branch?	17:25
calamari	Nevik: thanks!!!	17:25
	→ MistuhKurtz joined	17:25
	→ BenL joined	17:25
svm_invictvs	is there a configuration option that makes the "master" branch the actual master?	17:26
Nevik	svm_invictvs: no	17:26
	git init will create the master branch by default	17:26
	iirc that's hard-coded into git init	17:26
	but after initing the repo, you can just treat any branch as the main (you can delete or rename master too, if you want)	17:26
	nothing is special about the "main" branch	17:26
	gitinfo set mode: +v	17:26
	← BenL1 left	17:27
svm_invictvs	Nevik: Ah, I see.	17:27
Nevik	calamari: what i find odd about that reflog is those lines where he changed from <branch> to <samebranch>	17:27
	← cmpera left	17:27
calamari	Nevik: we are using a gui tool called GitExtensions, it is probably doing silly things behind the scenes	17:28
Nevik	hm, it might	17:28
svm_invictvs	Nevik: I renamed my remote master branch and created a new master branch. Evidently when somebody clonest he repository they get the old master checked out by dfault.	17:28
	→ skorgon joined	17:28
	← shruggar left	17:29
Nevik	calamari: i just found http://git.kaarsemaker.net/git/commit/c512b035556eff4d8f869afeda5fd78bc7a4966d/	17:29
	← mikecmpbll left	17:29
	← ncthom91 left	17:29
Nevik	calamari: so my guess is that he used git branch -f (intentionally or by suggestion of git extensions), which seems to include a reset operation. that might well have lost his commit	17:29
	← becks__ left	17:29
Nevik	calamari: the option is probably a rather inconspicuous checkbox in git extensions	17:30
	→ blinkdesign joined	17:30
	← donmichelangelo left	17:30
	→ becks__ joined	17:30
Nevik	svm_invictvs: ah, yes. the remote probably has a HEAD symref pointing to that	17:30
	← ecky--ptang-zoob left	17:30
calamari	Nevik: I suspect you are right, thanks a lot	17:30
DolphinDream	is there a reason for this difference between log and diff : git diff origin/master...master gives me the changes between the master and the common ancestor of origin/master and master, WHILE git log origin/master...master (gives me the log between the two branches). I have to user the .. notation instead of ... to get the log between master and the common anceestor of origin/master and master	17:30
Nevik	calamari: sure :)	17:30
	→ donmichelangelo joined	17:30
	→ ecky--ptang-zoob joined	17:30
DolphinDream	it's almost like the git log got the .. and ... switched	17:31
	→ arietis joined	17:32
Nevik	DolphinDream: from man git diff:	17:32
gitinfo	DolphinDream: the git-diff manpage is available at http://jk.gs/git-diff.html	17:32
	← fRAiLtY- left	17:32
Nevik	git diff [--options] <commit>...<commit> [--] [<path>...]	17:32
	This form is to view the changes on the branch containing and up to the second <commit>, starting at a common ancestor of both <commit>. "git diff A...B" is equivalent to "git diff $(git-merge-base A B) B".	17:32
	You can omit any one of <commit>, which has the same effect as using HEAD instead.	17:32
	(tl;dr: dont use commit ranges with git diff)	17:32
	→ takamichi joined	17:32
	← arand left	17:32
Nevik	DolphinDream: a diff is always between two points (commits), not across a commit range	17:32
	← nZac left	17:33
DolphinDream	Nevik, what's a commit range?	17:33
Nevik	A...B	17:33
	and A..B	17:33
	are both ranges	17:33
	← huahax left	17:33
Nevik	DolphinDream: see man gitrevisions for what the notations mean exactly	17:33
gitinfo	DolphinDream: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	17:33
	→ huahax joined	17:34
DolphinDream	Nevik, if i want to see what changes were introduced by both my local master and the remote/master i do git diff master..origin/master	17:34
	is that not the way to go ?	17:34
atrus	i've written up a little script that compares a range to a range. it's.... interesting.	17:34
Nevik	DolphinDream: that description doesnt make sense. you seem to want a diff of two diffs	17:35
	"changes introduced by master" is a diff (assuming some arbitrary point from which you diff master's current state)	17:35
	← JohnBat26 left	17:35
DolphinDream	Nevik, show me the changes between master and origin/master	17:35
Nevik	DolphinDream: git diff master origin/master	17:36
	← skorgon left	17:36
Nevik	is the difference between the current state of master and origin/master	17:36
	→ ncthom91 joined	17:36
DolphinDream	and the chagnes on master that are NOT on origin/master ?	17:36
	← p4denis left	17:36
Nevik	DolphinDream: git diff (that mergebase expression from above) master	17:36
DolphinDream	Nevik, why not git diff origin/master...master ?	17:36
Nevik	that's equivalent to what i just said	17:37
	→ p4denis joined	17:37
Nevik	but it's confusing	17:37
	because it's not interpreted as a commit range	17:37
	→ dmarkham joined	17:37
	← lotsofcode left	17:37
Nevik	because commit ranges don't make sense with a diff	17:37
	→ jceb joined	17:37
	→ cmpera joined	17:37
DolphinDream	so.. my question is then.. why doesn't git log origin/master...master do the same thing (except that for logs instead of diffs)	17:37
	→ codeforfun joined	17:37
	← nbouscal left	17:37
Nevik	well, technically, A...B is a commit set, not a pure range	17:37
	DolphinDream: if you had looked at the man page as i told you, you'll notice that A...B means "everything reachable from A and everything reachable from B, except not everything reachable from both"	17:38
	which is not what you say you want	17:38
	← cllamas left	17:38
DolphinDream	ok. i'll have to revisit that page. thanks	17:39
	← blinkdesign left	17:39
	← yacks left	17:39
	← mr_red left	17:39
	→ dreiss joined	17:39
	→ alexlindenlevy joined	17:39
Nevik	sure	17:40
	→ yhager joined	17:40
	→ mgolawala joined	17:40
	← SocalNick left	17:41
	→ xsbeats joined	17:41
	← p4denis left	17:41
	← nszceta left	17:41
	→ vkrest joined	17:41
	→ mr_red joined	17:42
	← welly left	17:42
	→ nszceta joined	17:42
	← tang^ left	17:42
	← jceb left	17:42
	→ nbouscal joined	17:43
	← Sonderblade left	17:43
	→ rue joined	17:44
	← PaddyPatPat left	17:44
	← _tpavel left	17:44
	← huahax left	17:45
	→ zumba_addict joined	17:45
	→ huahax joined	17:45
	← Exancillatus left	17:45
	← edwardly left	17:45
	→ craigp joined	17:45
	← vkrest left	17:46
	ashooby → ashooby\|lunch	17:46
	→ edwardly joined	17:46
	← canton7-mac left	17:46
	→ m0ltar joined	17:47
	← koda left	17:47
	← becks__ left	17:49
	nsfw → xnrand	17:49
	← thierryp left	17:49
	→ becks__ joined	17:49
	← edwardly left	17:50
	→ SocalNick_ joined	17:50
	← m0ltar left	17:51
	← richwestcoast left	17:52
	→ richwestcoast joined	17:52
	→ becks___ joined	17:52
	← infynyxx left	17:53
	← becks__ left	17:53
	→ pepper_chico joined	17:53
	→ mmcdaris joined	17:54
	→ ulkesh joined	17:54
calamari	Nevik: I recreated it. then he suddenly remembered what he did. he checked out a remote branch, and he selected the "reset" option	17:55
Nevik	lol	17:55
	whack him on the head from me please	17:55
	← huahax left	17:55
calamari	Nevik: that is why it looked like he was switching from the same branch to itself	17:55
Nevik	next time he does that, that day's pay goes to me	17:55
	→ sluther joined	17:55
	→ huahax joined	17:55
	← eaj left	17:56
	→ ezrafree joined	17:56
	← becks___ left	17:56
ezrafree	hello all	17:56
	→ becks__ joined	17:57
ezrafree	so me and a group of other developers all share a user on our dev server, is there any way to specify what my name/email should be set to when i commit? without setting it globally, for everyone, that is...	17:57
calamari	Nevik: thanks a lot for your help, I understand reflog a tiny bit better!	17:57
	→ edwardly joined	17:57
	← Kayra left	17:58
	→ DolpheenDream joined	17:58
	← thiago left	17:58
	← mgolawala left	17:59
	→ skorgon joined	17:59
	← yhager left	17:59
Nevik	no problem :)	17:59
	→ dougsland joined	18:00
	← a-l-e left	18:00
	→ infynyxx_ joined	18:00
	← SocalNick_ left	18:00
	← teran left	18:00
	← lavaman left	18:01
	→ lavaman joined	18:02
	→ SocalNick joined	18:02
prometh	is it possible to ammend a very old commit message?	18:02
	→ acharlieh joined	18:02
prometh	or must i rebase first?	18:02
	→ m0ltar joined	18:02
	→ Malice joined	18:02
ojacobson	It's possible, but it's rarely a good idea.	18:02
	!xy	18:02
gitinfo	Woah, slow down for a bit. Are you sure that you need to jump to that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...	18:02
prometh	ojacobson: it's a private repo	18:02
	→ FortySix2 joined	18:03
	→ waako joined	18:03
prometh	hello?	18:04
	→ cjsarette joined	18:04
	frobware → zz_frobware	18:05
lov	prometh: !filter-branch	18:05
	← robarelli left	18:05
milki	prometh: if !rewriting doenst matter to do, then you would use interactive rebase to reword the commit message	18:05
gitinfo	prometh: [!rewriting_public_history] Rewriting public history is usually bad. Everyone who has pulled the old history have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push -f` to force (and the remote may reject that, anyway). See http://goo.gl/waqum	18:05
lov	oh ffs :\|	18:05
	→ SocalNick_ joined	18:05
	→ solker joined	18:05
ezrafree	is it possible to set the "name" and "email" with each commit command?	18:05
	→ ahmedelgabri joined	18:05
prometh	oh, i have to do a rebase- i ?	18:05
lov	ezrafree: I believe so, check the options for git commit	18:05
prometh	i can't just rebase then ammend ?	18:05
	← ArSn left	18:05
	← skorgon left	18:06
	→ SeptDePique joined	18:06
	← SocalNick left	18:06
	SocalNick_ → SocalNick	18:06
prometh	oh yeah, ammend just does the absolute latest.. not latest relative to head	18:06
	→ nZac joined	18:06
ezrafree	lov: ahh there's an --author option, thanks!	18:06
ojacobson	if you're only looking for confirmation, why not try it on a copy of your repo and see what happens?	18:06
	← edwardly left	18:06
	← waako left	18:06
	← huahax left	18:07
	← f8l left	18:07
	→ robarelli joined	18:07
	→ thierryp joined	18:07
	← FortySix2 left	18:07
	→ huahax joined	18:07
	→ pheaver joined	18:08
	→ edwardly joined	18:08
	← kesha left	18:08
	← eletuchy left	18:08
	← yano left	18:09
	← ahmedelg_ left	18:09
	→ promethazin_ joined	18:09
	→ tuocuggino joined	18:09
	← devbug left	18:10
	← solker left	18:10
	→ Kayra joined	18:10
	← m0ltar left	18:10
	→ m0ltar joined	18:11
	→ rouf_ joined	18:11
	← patrickkettner left	18:11
	← thierryp left	18:11
	← rouf left	18:11
	← tadasZ left	18:12
ezrafree	one thing "git commit -h" doesn't really explain is the format of the value provided to --author, it should be: --author="Author Name [email@hidden.address]	18:12
	→ yano joined	18:12
	← ewnd9 left	18:12
	← mmcdaris left	18:12
	← edwardly left	18:13
	← ChrisAnn left	18:14
	→ mgolawala joined	18:14
	→ ErKa joined	18:14
	← kermit left	18:14
	→ mmcdaris joined	18:15
	← huahax left	18:15
	← m0ltar left	18:15
	→ huahax joined	18:15
	→ nlogn joined	18:15
	← kukks left	18:15
	→ InstaGrams5 joined	18:16
prometh	fucking godamnit! i just did a rebase -i to squash ancient commits... all teh dates were reset to today	18:16
	so much for a fucking history	18:16
	← wongon left	18:16
ojacobson	it's spelled out in the actual man page (git help commit, man git-commit)	18:16
gitinfo	the git-commit manpage is available at http://jk.gs/git-commit.html	18:16
prometh	why does git suck so fucking much?	18:16
ojacobson	prometh: what part of "it's usually not a good idea" wasn't clear to you	18:17
	← kriks left	18:17
ojacobson	it's not like I say things just to screw with people. Not here, anyways.	18:17
prometh	the makers of git should have thought about it.	18:17
ojacobson	(Also, git keeps separate author date and commit date)	18:17
	Rebasing commits resets the commit date, but usually not the author date	18:17
prometh	i asked if it reset the date... everyone said that it keeps it	18:17
	it didn't keep it.	18:17
	experts do not exist.	18:17
	← cads left	18:17
nlogn	heh	18:17
	→ Shusshu joined	18:17
InstaGrams5	Anybody like to play poker online? TRying to get players to sign up.100% match Msg me	18:18
	← Shusshu left	18:18
ojacobson	prometh: fortunately, you can get back to the pre-rebase state using the reflog	18:18
	and using git-reset	18:18
prometh	and if i push it? all dates gone?	18:18
ojacobson	try not to blow your feet off, I'd rather not tell you more in case you blame me	18:18
	→ LBV_User joined	18:18
prometh	i wish people just knew wtf they were doing, that's all	18:18
	i like advice... if it's correct	18:18
	← promethazin_ left	18:19
ojacobson	The part where you ignored my advice determined that that statement was a lie	18:19
prometh	everyone says it's not a good idea.. i know that part	18:19
ojacobson	And then you did it.	18:19
prometh	everyone's reason is always "could mess with ohter devs" ... no ohter devs!	18:19
ojacobson	Now "experts don't exist".	18:19
	→ promethazin_ joined	18:19
	→ mmcgrath joined	18:19
	← SocalNick left	18:20
prometh	i'm not a git expert.... clearly	18:20
	i wish i was, though	18:20
mmcgrath	is there any way to force git to not keep history? (Have a user that wants to use git but company policy prevents keeping history)	18:20
prometh	then i could sleep at night	18:20
ojacobson	mmcgrath: well, if you never run 'git commit'...	18:20
prometh	i want to merge history to clean it up	18:20
	it should be possible	18:20
	i did and dates fucked.	18:20
	← nZac left	18:20
jtrucks	I use Xcode with git repos. If I checkout and pull a repo, add it to a workspace, and then edit, commit, and push something, things work great. Then I made another edit, but can't commit and push. I think this is because I may not have done a pull in between, even though there shouldn't be anything different in the repo (I confirmed that nobody else had pushed commits to it).	18:20
ojacobson	I suspect and hope the policy isn't intended to mean what your user thinks it means...	18:20
prometh	to me that's an incomplete product	18:20
	← stephCoue left	18:21
	→ promethazin__ joined	18:21
atrus	prometh: are you sure it didn't preserve the author date? i've never seen rebase -i reset the author date. the commit date, sure. did you look at git log --pretty=fuller ?	18:21
	← calamari left	18:21
	→ SocalNick joined	18:22
	← InstaGrams5 left	18:22
	→ g0bl1n joined	18:22
	← g0bl1n left	18:22
	→ g0bl1n joined	18:22
jtrucks	so, anyone know how to fix this problem? If I now have xocde discard changes and do a pull, my locally committed change to a file is still in place, which means I can't pull a clean copy, manually redo my edits, and commit+push. The only solution I've seen before was to delete the entire local repo tree from disk and start over.	18:22
	anyone know a better way to do this?	18:22
	→ hahuang65 joined	18:22
prometh	atrus: thank you.. yes author date is preserved	18:22
	github shows commit date, unfortunately :(	18:22
atrus	prometh: also, maybe try a modicum of politeness when asking people for free support help.	18:22
prometh	so the history lookes weird	18:22
jtrucks	atrus: :)	18:23
prometh	atrus: i'm polite when i'm not mad and confused	18:23
	→ Norm_ joined	18:23
prometh	which is, probably surprisingly, most often	18:23
	→ m0ltar joined	18:23
jtrucks	most often mad and confused? ;)	18:23
	← promethazin_ left	18:23
	→ tadasZ joined	18:23
prometh	:P	18:23
atrus	prometh: maybe don't go asking for help when you're in a mental state where you can't control yourself.	18:23
prometh	maybe	18:23
	← ehmatthes left	18:24
	→ braj joined	18:24
Strogg	When you use git log with --pretty:format, %d shows you all of the refnames for each commit. Is there a way to make it not show tags?	18:24
	→ gittish joined	18:24
atrus	prometh: because seriously, a moment after i posted my suggestion, i regretted it, because (seeing how you've ranted) i assumed you were just trolling.	18:24
	← gittish left	18:24
	→ edwardly joined	18:24
	ashooby\|lunch → ashooby	18:25
	← vtunka left	18:25
prometh	ok	18:25
	→ GeertV joined	18:25
	← mgolawala left	18:25
prometh	is there any way to change commit date? heh	18:25
atrus	prometh: to be fair, i've been guilty of that in the past. eventually you learn that it's not in your best interest, or you move on to a career that doesn't involve interacting with people :)	18:26
	→ leeN joined	18:26
	→ DrShoggoth joined	18:26
prometh	and... if i do a reflog, and recommit, will it change commit date to today ?	18:26
atrus	prometh: you mean picking a commit out of the reflog, and doing a reset --hard to it?	18:26
prometh	yes	18:27
	will i get my dates back?	18:27
atrus	i think so, only one way to find out :)	18:27
prometh	ughh	18:27
	you don't know? :P	18:27
	git bothers me with stuff like this... it should be simple	18:27
	emmajane → ejfood	18:27
atrus	commit-date changes ~ if the commit hash changes. it's the date/time that the commit hash became meaningful. so i believe the reset applies.	18:28
	← l0x3py left	18:28
	← m0ltar left	18:28
prometh	reset will return original hashes ?	18:28
	→ francisfish joined	18:28
prometh	if i do a reset locally, i have to push it to the server... won't that create a new commit date ?	18:28
	→ standyro1 joined	18:29
	← Kayra left	18:29
	← huahax left	18:29
	← guampa left	18:29
atrus	keeping in mind i don't use github or know much about it, if you're doing a push --force to blast away the history that's already there, you're literally replacing the commits that are there with the commits you have, including their commit-dates.	18:30
	(all the usual caveats about using push --force apply, and if you don't know them, you may want to read about that )	18:30
	← edwardly left	18:30
	→ kriks joined	18:32
	→ mgolawala joined	18:32
osse	prometh: reset doesn't change commits, it only changes which branches point to them	18:33
prometh	hmm, trying to visualize that	18:33
	i only have a master branch	18:33
	→ axl_ joined	18:34
	← mgolawala left	18:34
	← SocalNick left	18:34
atrus	you may find http://marklodato.github.io/visual-git-guide/index-en.html and/or http://eagain.net/articles/git-for-computer-scientists/ helpful.	18:35
	→ promethazin___ joined	18:35
	→ eletuchy joined	18:35
	→ SocalNick joined	18:35
	→ guampa joined	18:35
	→ infynyxx joined	18:36
osse	prometh: think of it as if you drew the commits as connected dots no a whiteboard and you have a post-it not with 'master' written on it. git-reset moves the post-it from wherever it is now to where you specify it to go	18:36
	arrgh, on a whiteboard and a post-it note with	18:36
atrus	prometh: git (the implemented tools) really are quite simple, but it does mean you sometimes want to know a bit more about what's going on under the hood than you would with VCS tools.	18:37
	→ rath joined	18:37
atrus	er, "simpler" VCS tools i mean :) VCS tools where you can't do the sorts of things you can do with git	18:37
	← blizzow left	18:37
prometh	the only reason i have sourcetree is for visualizing branches	18:37
	→ vkrest joined	18:37
	← promethazin__ left	18:37
prometh	all of its commands are named the same as their command line counterparts	18:37
	so i know a bit about hwat's going on under the hood in git	18:38
atrus	prometh: i little further "under" :) that is to say, what the actually git commands are doing.	18:38
	← infynyxx_ left	18:38
	→ pinage404 joined	18:38
jtrucks	so, turns out, the only fix I can figure out is to wipe my fork and checkout again and add to xcode again :(	18:38
	→ Pullphinger joined	18:38
	← Macaveli left	18:39
prometh	atrus: yeah, i dunno if i have the time for that, hehe	18:39
	git is pretty complex	18:40
	→ blinkdesign joined	18:40
	→ jsilver_ joined	18:40
	← blinkdesign left	18:40
jsilver_	woohoo!!! successfully mirrored a svn path with subgit !!!	18:40
	don't be fooled, no git/svn integration is better than subgit	18:40
	if you need mirroring, there is no other option	18:40
	→ knuth joined	18:41
jsilver_	if all you need is import, it does that too but you may also live git2svn	18:41
j416	prometh, atrus: git's basic concept is not hard to understand	18:41
jsilver_	err, svn2git..	18:41
	→ m0ltar joined	18:41
	← pinage404 left	18:41
	← chrisshattuck left	18:41
j416	prometh, atrus: but indeed it does some magic to make things work well	18:41
	→ blizzow joined	18:41
	→ pinage404 joined	18:41
atrus	prometh: yes/no. there's a few things that, if you spend a little time getting to grok (and i mean a little, it's really not that scary :) ), you'll kind of unlock the secret mysteries of what's going on. super-rewarding, in a "i feel like i just leveled-up as a software developer" kind of way	18:41
	→ chrisshattuck joined	18:41
prometh	cool	18:42
	← ezrafree left	18:42
	→ vince_prignano joined	18:42
prometh	i'll have to try and squeeze it in, then... i just have other stuff i gotta learn too :(	18:42
atrus	reality will be what it is	18:42
prometh	while building public tools and doing client work	18:42
	← pinage404 left	18:42
	← promethazin___ left	18:42
j416	prometh: spending a few hours or even a few _days_ (yes) really learning git, will save you _a lot_ of time later. I promise.	18:42
	→ ezrafree joined	18:43
	← ezrafree left	18:43
	→ ezrafree joined	18:43
	→ promethazin___ joined	18:43
	→ jizanthapus joined	18:43
	→ simesy joined	18:43
	→ edwardly joined	18:43
prometh	ok, i'll put it near the top of the list	18:43
	thank you both	18:43
	← m0ltar left	18:43
	← mmcdaris left	18:44
j416	prometh: read all of http://git-scm.com/documentation and if you should get tired of that, watch this http://vimeo.com/46010208 (I think it's a pretty good intro), then continue reading	18:44
	→ arif-ali joined	18:44
	→ thiago joined	18:44
osse	\o/ That's the video I got the whiteboard and post-it not analogy from!	18:44
	arrrgh, note	18:45
j416	heh	18:45
	not sure it's a very good analogy	18:45
	lol	18:45
ashooby	and make sure you read this too http://stevelosh.com/blog/2013/04/git-koans/	18:45
	→ SHyx0rmZ joined	18:45
	← halu left	18:45
j416	nice way to do presentations of git, though	18:45
	→ halu joined	18:45
osse	j416: heh yeah. I guess it's not a good analogy but it's effective for visualization	18:45
	← takamichi left	18:45
	← vince_prignano left	18:46
j416	:)	18:46
	← vkrest left	18:47
prometh	5 long links to study, lol	18:47
	← simesy left	18:47
	→ kermit joined	18:47
prometh	that'll be a weekend	18:47
	best to get it in before summer	18:48
	→ tarkus joined	18:48
arand__	One thing about jessitron's talk though is that the presented "changeset" description is mixed up between Git and SVN.	18:48
j416	s/summer/monday/	18:48
	→ salamanderrake joined	18:48
prometh	haha	18:48
	← tbmiller left	18:48
	← chepre left	18:48
j416	arand__: indeed	18:48
osse	j416: although I was able to throw in the automatic gc by thinking of it as the cleaning lady coming by after hours and wiping away stuff that doesn't have a post-it note at the tip :P	18:48
tarkus	When sombody made a pull request, but you as a repo owner, want to make some final changes before merghing it into your repo.. how do you do it?	18:48
prometh	i say before summer, though, because in the summer i'd be missing sunshine	18:48
j416	osse: haha.	18:49
	osse: would that be the whiteboard eraser?	18:49
	lol	18:49
	→ pinage404 joined	18:49
j416	cleaning lady :D	18:49
osse	Setting: office	18:49
	She doesn't touch the coffee maker though. That's your problem	18:50
	← ezrafree left	18:50
j416	prometh: I say before monday, though, because the longer you go not knowing git, the angrier you will get because 1) git drives you insane 2) you are not as productive	18:50
	→ dik_dak joined	18:50
	j416 speculates	18:50
	→ ezrafree joined	18:50
	← ezrafree left	18:50
	→ ezrafree joined	18:50
prometh	as long as i don't do any squashing, i'm fine with git	18:50
j416	no you're not	18:50
prometh	i have been, is what i mean	18:51
	j416 squashes things _every single day_	18:51
prometh	well, ancient squashes	18:51
j416	ok not so much on weekends.	18:51
Nevik	ashooby: that link is awesome :D	18:51
ashooby	isn't it?	18:51
j416	lol I'm reading it @ link	18:51
	← aslaen left	18:51
ashooby	it should be on the bot	18:51
prometh	very recent squashing works fien for me	18:51
ashooby	!koans	18:51
	→ Kayra joined	18:51
	← pepper_chico left	18:52
emacer	does anyone know if there's a way to store git repositories within git repositories? I want them to appear hierarchically in the web frontend for my git server.	18:53
prometh	guys, if i do a hard reset and force push when i'm 43 ahead and 19 behind, will it change commit dates ?	18:53
	← GeertV left	18:53
vocifer	ashooby, I've never seen that blog with the git-koans but they are amasing	18:53
	← arif-ali left	18:53
	→ shruggar joined	18:53
	→ aslaen joined	18:53
	→ GeertV joined	18:53
j416	fun read ashooby. :)	18:54
	→ tbmiller joined	18:54
	← godgodgodgo left	18:54
j416	prometh: commits are immutable, so their dates never change.	18:54
prometh	hmm	18:54
j416	prometh: you'd understand what I just said if you had read the docs ;)	18:54
prometh	commit dates, not author dates	18:54
	that'll take me all day.. at this point, ij ust gotta move forward :(	18:55
j416	not meaning to be rude, but, really, read it.	18:55
	← aubreypwd left	18:55
j416	prometh: you're mistaken	18:55
	prometh: the sum of all time you will spend here asking will be much greater than the time it takes for you to read and learn it :)	18:55
	← thiago left	18:55
	→ mgolawala joined	18:55
	← Guest69162 left	18:55
	→ Guest69162 joined	18:55
Nevik	indeed	18:55
	Guest69162 → Spyro	18:55
prometh	j416: i understand.. but this is just one task tha ti need to get done for today	18:56
	if i read docs, i'll be trying to understand for days	18:56
j416	prometh: then tell us what result you want and what you have	18:56
prometh	i'm a visual learner, so i'm a lot slower from words	18:56
	and even the pics in some of those docs are confusing	18:56
j416	prometh: hence, I gave you a video	18:56
prometh	yes, 40min.. i will watch it later today	18:56
	→ looper_ joined	18:56
Nevik	.trigger_edit koans Looking for some enlightenment? Five koans about Master Git: http://stevelosh.com/blog/2013/04/git-koans/	18:57
gitinfo	Nevik: Okay.	18:57
Nevik	ashooby: ^	18:57
	j416 is off o/	18:57
	← alexlindenlevy left	18:57
prometh	j416: i did an ancient history squash and force pushed... all my commit dates wre changed to today... so i did a hard reset, dates are restored locally.... if i force push now, what will happen to those commit dates ?	18:57
Nevik	prometh: the pics are made to be accompanied by text. there are probably pics around the web that go with less text	18:57
ashooby	yay!	18:57
	!koans	18:57
gitinfo	Looking for some enlightenment? Five koans about Master Git: http://stevelosh.com/blog/2013/04/git-koans/	18:57
ashooby	<3	18:57
Nevik	;)	18:57
	→ alexlindenlevy joined	18:57
Nevik	prometh: the will be exactly what they are locally	18:57
prometh	ok, i will try	18:58
	← ezrafree left	18:58
	→ ezrafree joined	18:58
	→ grampajoe joined	18:59
	→ fragamus joined	18:59
prometh	whew	18:59
	ok, everything is back to normal now	18:59
	→ solker joined	18:59
prometh	i will never again try to do an ancient history rewrite	18:59
Nevik	lulz	18:59
prometh	recent history, sure... ancient, no	18:59
vocifer	prometh, why not? you got it fixed didn't you?	19:00
prometh	wow...3.5 hours wasted	19:00
	yes, but.. ^	19:00
Nevik	vocifer: he just restored the same state as before	19:00
prometh	yeah	19:00
Nevik	as in: chicken	19:00
prometh	back to the start	19:00
vocifer	yes the point is even though you feel like you screw up nothing bad happend	19:00
	← diegoviola left	19:00
vocifer	remember that instead	19:00
prometh	i like chicken... is that bad?	19:00
	← tbmiller left	19:00
	← salamanderrake left	19:00
	← s0beit left	19:00
Nevik	prometh: you chickened out instead of pulling it through like a git ;)	19:00
	→ tbmiller joined	19:01
prometh	oh, no, i finished my task.. but what i ultimately wanted was not possible	19:01
	i wanted to rewrite history and preserve commit dates	19:01
Nevik	which is what happens when you rewrite	19:01
prometh	so i reset everything to get my dates back	19:01
	→ Annak joined	19:01
	gitinfo set mode: +v	19:01
prometh	yeah :(	19:01
	→ teran joined	19:01
	→ thiago joined	19:01
Nevik	oh wait, i never remember if it's commit dates or author dates that are updated in the rewritten commits	19:01
prometh	is it possible, with shell access, to change git commit dates ?	19:01
Nevik	it doesnt matter because one remains the same	19:02
	→ mlimaloureiro joined	19:02
prometh	commit dates are changed, author dates are preserved	19:02
	it matters to me because of the github history page	19:02
	→ msodrew joined	19:02
Nevik	prometh: change them? no. set new ones while rewriting commits? yes	19:02
	← kwstas left	19:02
prometh	it said that all my work was doen in one day, heh	19:02
Nevik	so?	19:02
	stats are meaningless	19:02
prometh	so... public image, man!	19:02
	i started my project a while ago	19:02
Nevik	prometh: if your github stats are that important, you have way bigger issues than your commit dates	19:02
prometh	yeah.. my penis is small.	19:03
	→ teran_ joined	19:03
Nevik	at least you're honest about it	19:03
prometh	lol	19:03
	→ FortySix2 joined	19:03
	← eletuchy left	19:04
Nevik	prometh: you can manually set the commit date when (re)creating a commit	19:04
	it just requires some scripting	19:04
	← xsbeats left	19:04
Nevik	for a certain value of "some"	19:04
	→ s0beit\|ZNC joined	19:04
	← solker left	19:04
	s0beit\|ZNC → s0beit	19:04
prometh	scripting?	19:04
	← Diemuzi left	19:04
prometh	like, lots of commands?	19:04
Nevik	like, commands that are run in a certain automated fashion	19:04
prometh	hmm	19:05
Nevik	i would have to look it up myself.	19:05
prometh	sounds way over my head, then	19:05
Nevik	i dont know if rebase has appropriate support for custom commands or options	19:05
prometh	maybe when i'm rtfm'ed	19:05
Nevik	if it doesnt, filter-branch could probably do it	19:05
prometh	s/i'm/i've/	19:05
Nevik	if THAT cant, you can always use plumbing commands and do it the bare metal way	19:05
	prometh: well, yes :P	19:05
prometh	i wish there was just a simple GUI program to do all this	19:06
Nevik	when you finally know why commits are immutable and what rewriting history actually means, this'll be a lot clearer to you	19:06
prometh	select, merge.. done	19:06
	preserve commit dates? yes	19:06
Nevik	prometh: you're not doing easy things	19:06
	← teran left	19:06
prometh	they could be made easy, for sure	19:06
Nevik	you're rewriting history. something that no other VCS can actually do	19:06
prometh	can mercurial ?	19:07
Nevik	i dont know :D	19:07
prometh	hehe	19:07
	i've used it briefly... i kinda liked it more than git, btu was quite similar	19:07
Nevik	but i know that git can, and it's versatile enough to do just about anything	19:07
prometh	i've used svn as well... kind of a pin	19:07
	s/pin/pain	19:07
	s/pin/pain/	19:07
Nevik	i dont see why so many coders have trouble with git	19:07
	git is basically a programmer's dream	19:07
	→ thierryp joined	19:07
prometh	not in our nature, i guess	19:07
Nevik	if the UI doesnt do it, just script it	19:07
	← edwardly left	19:08
Nevik	i guess it's more of a linux mentality than a general programmer one	19:08
	→ helterscelter joined	19:08
prometh	it's like those retards in high school that have trouble grasping math	19:08
	that's me	19:08
	→ waako joined	19:08
ojacobson	Nevik: some people want an SCM system, not all the parts they need to build an SCM system	19:08
Nevik	prometh: good that you're honest. here, wear this idiot badge whenever you're here :D	19:08
prometh	i failed gr11 math	19:08
ojacobson	That's my biggest complaint about Git, really	19:09
	that and the terribad UI	19:09
prometh	lol, yeah.. i'm naturally a designer... but learned and learned to love programming	19:09
	so, git is very unnatural to me	19:09
Nevik	yes, the UI is terribly inconsistent. but the idea of the git core is ingenius	19:09
	it's not for everyone	19:09
	→ thierryp_ joined	19:09
Nevik	but it doesnt want to be	19:09
	← thierryp left	19:09
	← Kayra left	19:09
prometh	it should want to be... they'd have more happy campers	19:09
Nevik	prometh: not everyone needs others to like them to further their own self-esteem ;D	19:10
prometh	complicated things are, from my perspective, often unsophisticated things	19:10
Nevik	git is very self-confident	19:10
	→ kukks joined	19:10
	← kukks left	19:10
	→ kukks joined	19:10
prometh	Nevik: it's not about self esteem... it's about being able to walk outside without a bag over your head :P	19:10
	← cmpera left	19:10
	prometh must have his prada	19:10
	← pinage404 left	19:10
Nevik	in the case of the git UI, it's more "unplanned and naturally grown" than truly unsophisticated	19:10
	prometh: git is so self-confident, it walks around naked :P	19:11
prometh	is it fat?	19:11
Nevik	no, it's quite slim actually	19:11
prometh	well... git is stupid!	19:11
Nevik	yes	19:11
	by definition	19:11
vocifer	prometh, and a content tracker	19:11
	← SocalNick left	19:12
prometh	it's a naked content tracker?	19:12
	→ pen joined	19:13
vocifer	it's a stuped content tracker	19:13
prometh	oh, i wouldn't know.. i'm pretty basic with git	19:13
Nevik	so stoopid, it can't spell stoopit correkt	19:13
prometh	i'm a stupid git user	19:13
Nevik	prometh: read the top of man git :D	19:13
gitinfo	prometh: the git manpage is available at http://jk.gs/git.html	19:13
prometh	oh no.. anohter link to read	19:13
	there goes TWO weekends!	19:13
Nevik	prometh: just the first sentence after the headings, lol	19:14
	← thierryp_ left	19:14
prometh	oh, ok, i'll read that!	19:14
	→ SocalNick joined	19:14
prometh	really... that's their subtitle?	19:14
	← infynyxx left	19:14
	→ william1 joined	19:14
prometh	that's..... angering... i want to call git stupid	19:14
	and now i cannot	19:15
vocifer	XD	19:15
prometh	they have disarmed me	19:15
	← CKyle left	19:15
	→ edwardly joined	19:15
	→ infynyxx joined	19:15
Nevik	Linus predicted you	19:15
	Annak → Diemuzi	19:15
prometh	skynet exists!	19:15
Nevik	he put out an email two years ago saying that you would show up here on this day	19:15
	we were waiting all day for you to show up	19:15
prometh	i'm john connor's cousin	19:15
	→ salamanderrake joined	19:15
prometh	skynet knew i would come... i'm important n stuff	19:16
Nevik	lol	19:16
	← promethazin___ left	19:16
	← infynyxx left	19:16
prometh	git is dumb	19:17
Nevik	prometh: so yeah. if you're gonna use git a lot/regularly, it's probably worth learning it. and learning git usually works best from the inside out	19:17
prometh	there... humans win	19:17
Nevik	because, yes, git is dumb	19:17
	its core is incredibly simple	19:17
	→ promethazin___ joined	19:17
	← pen left	19:17
vocifer	some might even say elegant	19:17
prometh	yeah, i gotta spend some time on it	19:17
Nevik	that, compounded by the UI is terribly inconsistent and can be utterly confusing, makes it much easier to use git when you know what happens on the inside	19:17
prometh	will they ever make it consistent ?	19:18
	v2 kinda thing ?	19:18
	→ infynyxx joined	19:18
Nevik	prometh: if you can make yourself read it, the pro-git book (you probably already have the link) is good to give you an overview (chapters 1-3 and 5 are relevant)	19:18
	prometh: v2 is shrouded in mystery!	19:18
prometh	mysteries of the sith	19:18
Nevik	we (=non-git-developers) all hope for improvement	19:18
	→ diegoviola joined	19:19
luto	git is perfect	19:19
prometh	on the inside... so, it's like an unattractive girl ?	19:19
luto	lol	19:19
prometh	git... or should we call her betty?	19:19
Nevik	prometh: if you feel comfy with the basic everyday commands (possibly after having watched a video or two), i can recommend tackling !gcs and/or !cs	19:19
gitinfo	prometh: [!concepts] "Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs.html	19:19
	prometh: "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup	19:19
	← mlimaloureiro left	19:19
prometh	git v2... it's not betty anymore!	19:20
	← edwardly left	19:20
	→ mfulz joined	19:20
luto	but I like betty :(	19:20
Nevik	you'll notice those are relatively slim documents (with lots of pictures)	19:20
prometh	hmm... git v2: doesn't have to be betty if you don't want it to be betty anymore!	19:20
Nevik	once you get the graph idea anchored in your head, git is easy peasy	19:20
luto	DAG!	19:21
prometh	yeah, i have 5 links to read so far.. one of them a video	19:21
	Nevik watches prometh read a video	19:21
	← mgolawala left	19:21
Nevik	prometh: if you find you dont like that one, we have more !talks	19:21
gitinfo	prometh: Some good video talks about Git: [yt] http://goo.gl/z72s (Linus Torvalds: History&Concepts); [yt] http://goo.gl/R9H2q (Scott Chacon: Git basics, live examples); http://vimeo.com/35778382 (Randal Schwartz: Git basics, descriptional); http://vimeo.com/46010208 (Jesica Kerr: Git basics, descriptional)	19:21
prometh	it's the new thing.. haven't you heard ?	19:21
	video books	19:21
	← promethazin___ left	19:21
Nevik	lulz	19:21
	luto gives prometh a DAG to chew on	19:21
	→ valgrind joined	19:21
	← jason237 left	19:21
Nevik	luto: are you crazy, he's a designer! he'll put a cycle into that graph!	19:22
luto	no cycles allowed!	19:22
Nevik	prometh: i believe the video you were given before is Jessica's talk	19:22
luto	prometh: don't put cycles in your DAGs.	19:22
Nevik	that would make it a DCG	19:22
	and that's just wrong	19:22
prometh	how do you chew on a dag?	19:23
	is a dag something you eat?	19:23
	→ hobodave joined	19:23
Nevik	something just occurred to me	19:23
vocifer	start from the roots?	19:23
prometh	you didn't eat lunch?	19:23
Nevik	it'd be nice if it were a Directed Acyclic Non-reflexive Graph	19:23
vocifer	It would be?	19:23
prometh	dang ?	19:24
	or danrg ?	19:24
Nevik	DANG	19:24
	→ dp joined	19:24
Nevik	we would be doing operations on the DANG	19:24
	← shawndumas left	19:24
Nevik	bwahahaha	19:24
	im nearly rofling here	19:24
prometh	lol	19:24
	← asedeno left	19:25
vocifer	what would a it mean for a graph to be reflexive?	19:25
dp	is there a command that will show me the first time a line of text existed in a commit (from a specific file)?	19:25
	→ shawndumas joined	19:25
vocifer	dp, `git log -G <line>` i think	19:25
luto	-G = --grep ?	19:25
osse	luto: no	19:25
luto	:(	19:25
prometh	everyone in Nevik's office is wondering why he's laughing so hard	19:25
	he responds with "DANG!"	19:25
luto	man git log	19:26
gitinfo	the git-log manpage is available at http://jk.gs/git-log.html	19:26
osse	dp: git blame may fit the bill too	19:26
prometh	they nod in approval	19:26
	→ edwardly joined	19:26
luto	prometh: I don't think that Nevik is in a office right now.	19:26
prometh	in the assylum ?	19:27
Nevik	vocifer: as a graph represents a relation, a reflexive graph is one in which every node is related to itself	19:27
dp	awesome. git log -G worked perfectly	19:27
	thanks!	19:27
	← dp left	19:27
Nevik	vocifer: as the simplified git DAG usually represents history, that would mean every commit is its own ancestor	19:27
prometh	at the funny farm	19:27
	← mellernoia left	19:27
	← lavong left	19:27
luto	prometh: at home	19:27
prometh	aye, that's where i am	19:27
Nevik	luto: stop watching me through my webcam please	19:27
luto	if I have the right timezone in mind	19:27
prometh	people look at my funny here too, though.... are they really there?	19:28
luto	you're UTC+1 or something, right, Nevik?	19:28
	→ pen joined	19:28
luto	wasn't there a command?	19:28
	.time Nevik	19:28
	.timezone Nevik	19:28
	!time Nevik	19:28
gitinfo	[!reply_time] I only respond to !ECHO-REQUEST	19:28
luto	wat?	19:28
	!timezone Nevik	19:28
prometh	!echo-request nevik	19:28
	← tarkus left	19:28
prometh	gitinfo has failed us.	19:28
Nevik	luto: im figuratively breathing down your neck	19:28
	so yes	19:28
luto	!ECHO-REQUEST	19:29
gitinfo	I only respond to !ping	19:29
luto	lol..	19:29
prometh	lol	19:29
luto	!ping	19:29
gitinfo	I only respond to !reply_time	19:29
luto	!reply_time	19:29
gitinfo	I only respond to !ECHO-REQUEST	19:29
prometh	!fuck-you	19:29
luto	it's a loop!	19:29
prometh	yup	19:29
Nevik	i was gonna check my whois, but my bouncer is not in my own timezone	19:29
	i was gonna say, check my whois*	19:29
	→ tarkus joined	19:29
	→ tieTYT joined	19:30
Eugene	About time somebod figured that out	19:30
prometh	skynet put it there	19:30
	→ eletuchy joined	19:30
	← LBV_User left	19:30
	← Not_Whatever left	19:30
tieTYT	so @{-1} means the previous branch I was on, right? How do I get the name of that branch?	19:30
	ChanServ set mode: +o	19:30
	→ icarus_ joined	19:31
	→ LBV_User joined	19:31
	← GeertV left	19:31
	← prometh left	19:31
	Eugene changed the topic to: Welcome to #git, the place for git help and stale leftovers \| Current stable version: 1.8.5.3 \| Start here: http://jk.gs/git \| Seeing "Cannot send to channel" or unable to change nick? /msg gitinfo .voice \| I bet you thought this was a joke. Nope, git-testa.	19:31
	Eugene set mode: -o	19:31
luto	Eugene: lol..	19:31
vocifer	`git rev-parse --abbrev-ref ${-1}`	19:31
Eugene	s/$/@	19:31
	But yes	19:31
	← arietis left	19:31
vocifer	thank you	19:31
	Eugene, you are a live saver	19:32
tieTYT	is that for me?	19:32
vocifer	XD	19:32
	tieTYT, yes	19:32
tieTYT	ok thanks	19:32
	nice	19:32
	oh that's confusing... it shows the branch I'm already on	19:32
	does that mean I checked out a branch I was already on?	19:32
	→ devbug joined	19:32
vocifer	tieTYT, I don't know why do you think it would be a different branch?	19:33
tieTYT	doesn't -1 mean "the previous branch you were on"?	19:33
vocifer	tieTYT, no the previious commit	19:33
	fuck	19:33
	nop	19:33
	nvn I'm screwing up	19:33
	i'm tired	19:34
	← kukks left	19:34
	← tadasZ left	19:34
	→ jnavila joined	19:34
Nevik	yeap	19:34
	→ wongon joined	19:35
tieTYT	is there a way to see a history? So I can see what -2, -3, -4...	19:35
	→ tadasZ joined	19:35
icarus_	If i git clone a repo, and i later want to update my local clone of changes from the master repo, what would i run?	19:35
Nevik	tieTYT: git reflog \| grep "moving from"	19:35
ojacobson	tieTYT: man gitrevisions tells you how each is interpreted	19:35
gitinfo	tieTYT: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html	19:35
tieTYT	hm, this is an XY problem. My real annoyance is I'm ready to merge and I forget what my branch is called. I have to clear my console and do a 'git branch' which takes me out of my workflow	19:35
ojacobson	the branch you're on or the branch you're merging from?	19:36
	→ CKyle joined	19:36
ojacobson	('git branch --contains HEAD' and 'git branch --contains MERGE_HEAD' respectively)	19:36
Eugene	If you forget your branch name after 30 seconds try writing it down. On paper.	19:36
	Or just scroll up in your terminal	19:36
	← dfanjul left	19:37
Nevik	or use a prompt that shows it	19:37
vocifer	\nick vocifer-away	19:37
	fuck me	19:37
ojacobson	Please don't	19:37
	vocifer → vocifer-away	19:37
ojacobson	nobody cares, just set /away :)	19:37
Nevik	vocifer-away: ojacobson is right	19:38
vocifer-away	was writing latex	19:38
ojacobson	Don't tell /counts/ several hundred people you're going away	19:38
Nevik	dont use away-nicks	19:38
	→ vkrest joined	19:38
	→ Kayra joined	19:38
	vocifer-away → vocifer	19:38
osse	oops. This is my away nick. I'm stuck forever.	19:38
Nevik	vocifer: good boy	19:38
	ChanServ set mode: +o	19:38
	Eugene set mode: +b	19:38
ojacobson	hah	19:38
	← ahmedelgabri left	19:38
Eugene	</problem>	19:38
vocifer	see I real really should get some sleep	19:38
Nevik	Eugene: that's rather pointless :D	19:38
	← daivyk left	19:38
	← angelsl left	19:39
ojacobson	Nevik: bans on freenode sometimes also mute matching participants	19:39
Nevik	not only sometimes	19:39
ojacobson	I don't know the specifics but we use it on ##java to restore order	19:39
	← edwardly left	19:39
Nevik	but the annoying thing is the nick change	19:39
Eugene	And keeps them locked to that nick, forcing them to realize their mistake	19:39
ojacobson	Oh, sure	19:39
Nevik	not them saying something	19:39
ojacobson	Also, the meta-discussion about nicks	19:39
Nevik	lol	19:39
	→ edwardly_ joined	19:39
Eugene	The real problem here is you gits don't know how to Hide Join/Parts/Nickchanges in your IRC client. In a room of 1000 people it screws the SNR	19:40
	→ opc0de joined	19:40
Eugene	Sure I miss a few ban opportunities for ##fix_your_connection, but it's worth the peace+quiet that it affords my window	19:40
rawtaz	whats SNR in this context?	19:41
opc0de	hi, anyone can help me with gitlist, i'm on fedora for test and i get this error message: "cache" folder must be writable for GitList to run	19:41
Eugene	Signal-Noise Ratio	19:41
rawtaz	ah	19:41
	→ nnyk_ joined	19:41
grawity	relevant http://wiki.xkcd.com/irc/Hide_join_part_messages	19:41
	← tarkus left	19:41
	→ tarkus joined	19:42
	← vkrest left	19:42
	→ aubreypwd joined	19:42
	→ Not_Whatever joined	19:42
Eugene	That's the one!	19:42
	→ Apocalis_ joined	19:43
	→ asedeno joined	19:43
	→ jatt joined	19:43
	← grampajoe left	19:44
	← Apocalis_ left	19:44
	← devbug left	19:44
Su7	Hello #git	19:44
	I have a question for you	19:44
rawtaz	so do we, for you	19:44
	→ el3ctron1 joined	19:45
Su7	I'd like to set up a cron to git pull from my repo to my testing environment	19:45
	→ XndrK joined	19:45
	→ el3ctron3 joined	19:45
Su7	but I don't want my testing env's changes to be erased at each pull	19:45
XndrK	Hello	19:46
gitinfo	XndrK: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	19:46
	→ thierryp joined	19:46
Su7	(let's say I have a config file that would be erased at each pull. I want to keep this file as-is and apply al the other changes from origin)	19:46
	→ cmpera joined	19:47
XndrK	I am using the Git GUI and trying to clone a repository from GitHub.	19:47
	→ ehynds joined	19:47
	← mfulz left	19:47
ojacobson	Su7: !config and !deploy are probably good starting points	19:48
gitinfo	Su7: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas	19:48
	Su7: Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html	19:48
XndrK	When I click Browse for folder next to the target folder window and select one, it always says it already exists.	19:48
	Two questions:	19:48
	1: Why is it important that it already exists?	19:49
	← _Vi left	19:49
Eugene	git-gui or the Github application?	19:49
XndrK	git-gui	19:49
	2: How can I make it shut up?	19:49
Eugene	(I'm useless with both, but wanted to make sure you knew they were separate)	19:49
XndrK	I know they are different.	19:49
ojacobson	'git clone' tries not to replace existing directories, ever, in case you have files in them you care about	19:49
	'mkdir foo && git clone ... foo' will do the same thing, and fail the same way	19:50
XndrK	so what's the browse for folder button for?	19:50
ojacobson	(it probably could detect the empty-dir case, but to my knowledge it doesn't. Git.)	19:50
XndrK	I finally solved the problem by going to the folder in the browse for folder thing and then adding the new folder thing	19:51
	→ motto joined	19:51
	← opc0de left	19:51
XndrK	Is that the only option, and if so, what's the New Folder button for?	19:51
Su7	ojacobson: awesome, thanks !	19:52
	→ devbug joined	19:52
	→ el3ctron4 joined	19:52
	→ cozby joined	19:53
	← m8 left	19:53
	→ narrenschiff joined	19:53
	← jatt left	19:53
	← narrenschiff left	19:54
	← poulson left	19:55
	XndrK will be doing something else for a minute, so please put his username before your message to alert him.	19:55
	→ jatt joined	19:55
rawtaz	i find it rather silly when ppl write about themselves in third person :P	19:57
	→ mfulz joined	19:57
	← thierryp left	19:57
Eugene	You're puppets.	19:57
	← eletuchy left	19:57
	← jatt left	19:58
	← girishr left	19:58
	→ cads joined	19:58
ojacobson	rawtaz: clearly you've never used IRC for collaborative fiction :)	19:59
	← fire left	19:59
rawtaz	haha	19:59
	→ mgolawala joined	20:00
	→ jatt joined	20:00
	→ veebers joined	20:01
	← mgolawala left	20:01
	← tvw left	20:01
	ehynds → erichynds	20:01
	← mfulz left	20:01
	← oojacoboo left	20:02
milki	rawtaz: what about when they speak about themselves in third person?	20:02
	← fragamus left	20:02
	← jerius left	20:02
	← msodrew left	20:02
	← erichynds left	20:02
rawtaz	milki: what about it? i find it silly :)	20:03
	→ jerius joined	20:03
	→ g3orge joined	20:03
	→ msodrew joined	20:03
	→ mgolawala joined	20:03
milki	i think some languages/cultures have it built in	20:03
rawtaz	that might be true	20:04
	i read it one way, other cultures probably read it differently	20:04
	→ nZac joined	20:05
	→ TeraJL joined	20:05
TeraJL	hi there, how can i cancel a GIT RM, without reverting the file to my hardrive?(i've already added the file to gitignore	20:05
	← shawndumas left	20:06
	← unreal left	20:06
	→ Freeman1337 joined	20:06
	← mgolawala left	20:07
	← afhammad left	20:07
	← XndrK left	20:07
	→ bartman` joined	20:07
	→ jason237 joined	20:08
	→ kevwil joined	20:08
	→ unreal joined	20:08
rawtaz	TeraJL: git add ?	20:08
	i dont know :D	20:08
	i'd expect git status to tell me	20:08
	→ danielcompton joined	20:08
TeraJL	i have removed the files , and added them on gitignore, but they still show up on git status	20:08
	as removed	20:08
	← Freeman42 left	20:08
TeraJL	i don't want them removed from the repository, i just don't want them locally	20:09
rawtaz	TeraJL: yeah files that are tracked will still be tracked even if you add them to gitignore. you need to untrack them	20:09
	← nZac left	20:09
	→ ZaeDrake joined	20:09
rawtaz	this took me literally four seconds to find: https://stackoverflow.com/questions/1274057/making-git-forget-about-a-file-that-was-tracked-but-is-now-gitignored	20:09
	perhaps it is relevant	20:09
	→ Freeman42 joined	20:09
	→ ciampix joined	20:09
rawtaz	unless i misunderstood you	20:10
	← LBV_User left	20:11
	→ danman joined	20:11
	← edwardly_ left	20:12
	← ZaeDrake left	20:12
	← Freeman1337 left	20:12
	→ alvaro_o joined	20:13
	→ yhager joined	20:13
	→ aChatir joined	20:14
	← jatt left	20:14
diegoviola	i did a rebase/squash of a few commits that i have in a branch that nobody else has, but i already pushed that branch to my remote	20:15
	→ danillonunes joined	20:15
	← Vivekananda left	20:15
diegoviola	if i push that branch to the remote the commits i squashed will disappear and will show only the single commit i have as a result from the squash, right?	20:15
	in the remote i mean	20:15
Eugene	Yes, if you push -f it	20:15
	→ dglambert joined	20:16
	← ulkesh left	20:17
	→ DynamiteXXL joined	20:18
	→ vkrest joined	20:18
TeraJL	i've cloned again.. how can i remove certain folders from pulling and pushing and delete them locally?	20:20
	← crazedpsyc left	20:20
	← jason237 left	20:20
	→ Ardit joined	20:21
rawtaz	huh?	20:21
TeraJL	almost all the solutions remove the file from the origin repository	20:22
	→ edwardly joined	20:22
rawtaz	solutions to what problem?	20:22
	← sohail left	20:23
TeraJL	like git rm -r --cached .	20:23
	→ _Vi joined	20:23
rawtaz	solutions to what problem?	20:23
	→ sohail joined	20:23
TeraJL	i don't want to change anything from the server... i want to remove a folder locally, and work (pull and push) and just forget about that folder from/to this computer	20:24
	→ crazedpsyc joined	20:24
	→ eletuchy joined	20:24
	← freeed left	20:24
diegoviola	Eugene: ty	20:24
	← SocalNick left	20:25
rawtaz	TeraJL: wouldnt you want that folder to be in a .gitignore file in the repo then?	20:25
TeraJL	yes	20:25
rawtaz	so do that?	20:25
Eugene	TeraJL - man git-update-index, see "assume unchanged". Hilarity will likely ensue.	20:26
gitinfo	TeraJL: the git-update-index manpage is available at http://jk.gs/git-update-index.html	20:26
	→ Jurko joined	20:26
	← ezakimak_work left	20:26
rawtaz	hehe looks like it can get messy in the end :)	20:27
	→ MattStratton joined	20:27
	← salamanderrake left	20:28
	← mikar left	20:28
	→ jatt joined	20:30
	← veebers left	20:30
	← jackneill left	20:30
TeraJL	i will try to read it... i'll just try to explain one last time, and thanks.... i want this: on computer A, i work with everything, so i "pull" and "push" as normal to the origin.... but on computer B i don't work with images and static files.. so i wanted to "pull" and "push" evrything but the "static/" folder... but keep it on the server	20:30
	→ veebers joined	20:31
	← zizzl left	20:31
	→ kukks joined	20:31
	← kukks left	20:31
	→ kukks joined	20:31
rawtaz	interesting.	20:32
	TeraJL: what is the reason for this. again, what problem are you trying to solve by wanting to do this?	20:32
TeraJL	space	20:32
rawtaz	that is way more relevant than how to do the thing you ask about	20:32
	ok. not bandwidth etc?	20:32
TeraJL	yes	20:32
rawtaz	just disk space	20:32
	ok	20:32
	→ mikar joined	20:33
rawtaz	interesting question anyway	20:33
	→ jason237 joined	20:33
	← dougsland left	20:34
rawtaz	i have no idea if thats doable. telling git to not fetch certain parts of the repo	20:34
	← ben_alman left	20:34
	→ bl33_ joined	20:34
	→ FortySix_ joined	20:34
	→ SocalNick joined	20:34
	← TeraJL left	20:35
	← kermit left	20:36
	ejfood → emmajane	20:36
	→ arif-ali joined	20:36
	→ matanya joined	20:36
	← mikar left	20:36
	← wongon left	20:37
	→ mikar joined	20:37
	→ dougsland joined	20:37
	← dougsland left	20:37
	→ dougsland joined	20:37
	← FortySix2 left	20:38
	← mikar left	20:38
	→ thierryp joined	20:38
	→ Foldager joined	20:38
	→ mellernoia joined	20:40
	→ joelsmith joined	20:40
	→ rafaqueque joined	20:41
	→ tiglionabbit joined	20:42
	← bl33_ left	20:42
tiglionabbit	How can I get a diff of two versions of a binary file in git?	20:42
	can I get it to use HexFiend as a difftool or something?	20:42
	← thierryp left	20:43
arand__	tiglionabbit: you can configure difftools per filetype.	20:43
tiglionabbit	I don't know how to use difftool	20:43
moritz	man git-config, look for diff.external	20:43
gitinfo	the git-config manpage is available at http://jk.gs/git-config.html	20:43
	← imachuchu left	20:43
	→ ben_alman joined	20:44
	→ tvw joined	20:44
moritz	... and the related config options	20:44
tiglionabbit	moritz: ok but none of the builtin options are going to be meant for binary	20:45
	how do I set up a binary diff tool?	20:45
	→ correcto joined	20:45
	→ ngladitz_ joined	20:45
	← ngladitz_ left	20:45
	→ ngladitz_ joined	20:45
	→ mizu_no_oto joined	20:46
	→ daivyk joined	20:46
	← ngladitz left	20:47
	→ thierryp joined	20:48
	← g0bl1n left	20:48
	← tvw left	20:48
	→ simesy joined	20:49
	→ tvw joined	20:49
	← kukks left	20:49
	→ prudhvi joined	20:49
	← holodoc left	20:49
prudhvi	Hi, how can git detect changes to the working directory so quickly?	20:49
_ikke_	prudhvi: By using stat information	20:50
	→ imachuchu joined	20:50
	→ kermit joined	20:50
	← simesy left	20:50
	→ simesy_ joined	20:50
_ikke_	it caches the stat information in the index, and compares it next time you do a git status	20:50
	← eletuchy left	20:50
	← sohail left	20:52
	← bartman` left	20:53
	← marthinal left	20:53
	← MistuhKurtz left	20:54
	← francisfish left	20:54
	← yhager left	20:55
	← simesy_ left	20:55
	← jatt left	20:56
	← ebetancourt left	20:56
	→ mvelioglu joined	20:56
	→ promethazin___ joined	20:57
	→ robban666 joined	20:58
	→ gelardi_ joined	20:58
	← FortySix_ left	20:59
	→ revele joined	20:59
	→ FortySix2 joined	20:59
	→ Not_Whatever2 joined	20:59
	← hoschi left	21:00
	← halu left	21:01
	← robbyoconnor left	21:01
	← Pullphinger left	21:02
	→ mgolawala joined	21:02
	→ xSmurf joined	21:02
	mgolawala → KT	21:03
	← tiglionabbit left	21:03
	← promethazin___ left	21:04
	← jizanthapus left	21:04
	→ tpollard joined	21:04
	← nedbat left	21:04
	→ promethazin___ joined	21:05
	→ oojacoboo joined	21:05
	← Kabaka left	21:05
	← thierryp left	21:05
	→ solker joined	21:05
	→ tiglionabbit joined	21:05
	← davividal left	21:06
	→ afhammad joined	21:06
	← robban666 left	21:06
	→ Kabaka joined	21:08
	← correcto left	21:08
	← codeforfun left	21:08
	← veebers left	21:08
	→ robban666 joined	21:08
	→ lsm5 joined	21:09
	← promethazin___ left	21:09
	→ eletuchy joined	21:09
	← solker left	21:10
	→ shawndumas joined	21:10
	← friesk left	21:10
	← cmpera left	21:11
	→ nedbat joined	21:11
	← ngladitz_ left	21:12
	← darius93 left	21:12
	→ ngladitz joined	21:12
	← ngladitz left	21:12
	→ ngladitz joined	21:12
	← Foldager left	21:13
rawtaz	is there a way to make git status show renames in the "rename {private/protected/backend => backend}/components/BackendController" format (the {} part) instead of like it does it by default?	21:13
relipse	how do i get back to the latest and greatest master branch discarding all local changes?	21:13
	→ bodom joined	21:13
rawtaz	relipse: afaik you can do git checkout master && git reset --hard	21:14
	← Rix left	21:14
rawtaz	im not sure if you should follow up with a git clean -ndf and then git clean -df too though, but its only for getting rid of untracked stuff	21:15
	← nedbat left	21:15
	→ irashadul joined	21:15
	← irashadul left	21:16
	← infynyxx left	21:16
	→ cmpera joined	21:16
	→ GeertV joined	21:17
	← ezekielnoob left	21:17
	← GeertV left	21:18
	→ cogitos joined	21:18
	← jason237 left	21:18
	→ GeertV joined	21:18
_ikke_	rawtaz: I don't think git supports that	21:20
	→ sohail joined	21:20
	← cogitos left	21:20
	→ infynyxx joined	21:21
	→ cogitos joined	21:21
	→ bwreilly_ joined	21:21
	→ veebers joined	21:21
	→ Sabbath joined	21:21
	→ ezekielnoob joined	21:21
	← bwreilly_ left	21:22
	← diegoviola left	21:22
	← bwreilly left	21:22
	→ bwreilly joined	21:22
	→ joke2k joined	21:22
joke2k	hi	21:22
	← sohail left	21:23
joke2k	there is some italians?	21:23
	→ kcj joined	21:23
	→ Noldorin joined	21:23
	→ mlimaloureiro joined	21:24
	→ Exancillatus joined	21:24
	→ holodoc joined	21:24
	← joke2k left	21:25
	← Sabbath left	21:26
	← standyro1 left	21:27
	← veebers left	21:29
	→ nDuff joined	21:29
	← SHyx0rmZ left	21:29
nDuff	Is there a way to retrieve the origin URL for a remote intended for programmatic consumption? Trying to parse the human-readable output of "git remote show [remote-name]" seems pretty unfortunate.	21:30
	→ blinkdesign joined	21:31
	→ ham1 joined	21:31
relipse	how do i undo a merge	21:32
	gitinfo set mode: +v	21:32
	→ MistuhKurtz joined	21:32
nDuff	ahh, git config --get remote.<name>.url	21:32
	relipse: just reset the branch pointer to where it was beforehand	21:32
	← tbmiller left	21:33
	→ tbmiller joined	21:34
	→ darius93 joined	21:34
	→ Silmarilion joined	21:34
	→ SHyx0rmZ joined	21:35
	→ iderik joined	21:35
	→ teran joined	21:35
	→ Rix joined	21:36
	← danielcompton left	21:36
	← Usama left	21:36
	→ Macaveli joined	21:36
rath	relipse: man revert , it's the "-m" option you're looking for	21:37
	!man revert	21:37
	← tbmiller left	21:38
	→ thierryp joined	21:38
	← teran_ left	21:38
	→ salamanderrake_ joined	21:38
osse	nDuff: depending on your setup url might be different from pushurl.	21:38
	rath needs to learn how to activate the bot	21:38
	← MattStratton left	21:39
	← Freeman42 left	21:40
	← salamanderrake_ left	21:41
	→ salamanderrake joined	21:41
rawtaz	_ikke_: ok. well if it did you'd probably know about it	21:41
	← KT left	21:41
	← IrishPrime left	21:41
rawtaz	too bad, it's a bit odd that it uses two different formats	21:41
	← Macaveli left	21:41
	← jnavila left	21:42
	← CharlesFinley left	21:43
	← thierryp left	21:43
	→ mribeirodantas joined	21:43
	→ joke2k joined	21:44
	→ SocalNick_ joined	21:45
	→ githogori_ joined	21:46
	← githogori_ left	21:46
	← machine-x left	21:47
	← tarkus left	21:47
	← SocalNick left	21:47
	SocalNick_ → SocalNick	21:47
	← ezekielnoob left	21:47
	← arif-ali left	21:48
	← mlimaloureiro left	21:49
	→ arif-ali joined	21:50
	→ rj11 joined	21:50
rj11	Is there any easy way to make diff --check include the hash of the commit that introduced a whitespace error?	21:51
AciD`	noob question : I started working on a project few months ago (located in /old/path/to/proj), using git to manages changes. I now want to totally reorganize my files in a new directory structure like /new/path/to/proj (cherry picking each file to mv it to a specific new directory, then later renaming it). How should I proceed if I want to keep the history for each file ?	21:53
	→ veebers joined	21:53
AciD`	I only have 5 files in git, so I can do that manually	21:53
	→ standyro1 joined	21:53
osse	rj11: I doubt it since git compares two commits directly without looking at the history between them. I think it outputs a line number though. You can do 'git blame <commit2> <file>' and go to that line number	21:53
	→ nZac joined	21:54
	← FatDarrel left	21:54
	→ primitive joined	21:54
	→ cenuij joined	21:55
rj11	osse: Yeah, that's what I've been doing. A bit of a pain when more than a few files/commits contain whitespace errors. :)	21:56
	← arif-ali left	21:56
osse	rj11: they're all probably from the same guy :P	21:57
	→ arif-ali joined	21:57
	→ nafg joined	21:57
	← oojacoboo left	21:57
osse	maybe there's a way to script it	21:58
	← nZac left	21:59
	← iderik left	21:59
rj11	osse: Most likely. :P Scripting it would be doable.	22:00
	← ham1 left	22:00
	→ iderik joined	22:00
	→ xMYTHICx joined	22:01
	→ oojacoboo joined	22:01
	← standyro1 left	22:01
	→ nedbat joined	22:02
	→ endiruna joined	22:02
	← nathancahill left	22:03
	→ Fandingo joined	22:03
	← danecando left	22:03
	← asQuirre_ left	22:04
	← cannonball left	22:04
	→ FatDarrel joined	22:04
	← aChatir left	22:04
	← shruggar left	22:04
	→ solker_ joined	22:05
	← Fandingo left	22:05
	← _iron left	22:07
	xnrand → a	22:07
	→ thierryp joined	22:08
	← arif-ali left	22:08
rj11	If anyone knows about an indescribably basic introduction to git I'd appreciate a link.	22:09
	→ dsdeiz joined	22:09
rawtaz	rj11: have you tried the book?	22:09
	!book	22:09
gitinfo	There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable	22:09
	→ arif-ali joined	22:09
	← kossy left	22:09
	← solker_ left	22:09
rj11	Yeah, I've tried the book (Pro Git). And git flow. And Google. But getting my coworkers to grasp git is impossible.	22:10
	→ kossy joined	22:10
	→ simesy joined	22:10
rj11	I see merges from develop into master, from master into develop. Hotfixes branched off of feature branches later merged directly into master.	22:10
	→ teran_ joined	22:10
	← cmpera left	22:10
rawtaz	gee	22:11
	whats wwrong with them	22:11
	← dort left	22:11
	← xMYTHICx left	22:11
osse	meh, they're all just branches anyway	22:11
rawtaz	i mean those things are not typing mistakes, it's fundamental lack of understanding	22:11
	← calcifea left	22:11
rawtaz	or they dont know which branches theyre on	22:11
	→ calcifea joined	22:12
rawtaz	its not like it's a matter of git cli being hard either, theyd probably have the same problems with another branchy VCS	22:12
	rj11: try making them watch a video about git flow?	22:12
	← thierryp left	22:12
rawtaz	do you need that workflow model anyway or can you settle on something simpler until they become smarter?	22:13
	→ cagedwisdom joined	22:13
rj11	No, not really on that project. But I figured learning different workflows for different projects would be even harder.	22:13
	← teran left	22:14
	→ simesy_ joined	22:14
	← rudisimo left	22:14
	← bodom left	22:14
rj11	But trying the video approach might we a good idea.	22:14
rawtaz	mebe. i was just thinking perhaps they would find it easier if there was just a master branch and all they do is feature branches (and merging to it from master)	22:14
	← DolphinDream left	22:15
	DolpheenDream → DolphinDream	22:15
rawtaz	and you could be the main integrator so they make a PR for their feature branches and you are the only one allowed to merge that into master	22:15
osse	what's the difference between a hotfix branch and a feature branch? (Except its lifetime)	22:15
rawtaz	yeah i think so. it might make it clearer how simple it is to use git flow	22:15
	← DolphinDream left	22:15
	→ SocalNick_ joined	22:15
	SCM\|NotHere → Starcraftmazter	22:16
rj11	osse: If the bug is from current production it is a very bad idea to branch off of a feature branch. Merging that bug fix back into master to release a new version would include the partially finished feature.	22:16
	→ friesk joined	22:16
	← hellopat left	22:16
	→ skorgon joined	22:16
	← Kicer86 left	22:17
	← simesy left	22:17
	← SocalNick left	22:17
	SocalNick_ → SocalNick	22:17
osse	rj11: yeah I get that.	22:17
	rj11: but by the same reasoning it's basd to branch a feature branch off another feature branch	22:17
	← ben_alman left	22:18
rawtaz	osse: features go into the next release, while hotfixes are for small patches/whatever that cannot wait for the next release. so you make a hotfix and merge it back to master (and develop). while feature branches is not something that go into the mainline master branch, they go into the next release (and with that into master)	22:18
rj11	In the end it comes down to common sense. But there must be some logic behind it.	22:18
rawtaz	its just a workflow, which is just a set of rules basically	22:19
	policy	22:19
osse	then rj11 must come in a position to enforce that policy	22:19
	→ shruggar joined	22:19
rawtaz	yeah	22:19
	if they dont listen thre's nothing he can do really	22:19
	zoo.	22:20
	← GeertV left	22:20
rj11	Keep in mind I've been working there for a few months. Before that they use Dropbox for "version control". :)	22:20
	used*	22:20
rawtaz	hah	22:20
	are you regretting your choice of working place et? :P	22:20
	yet*	22:20
rj11	Haha, not yet. They're quite interested in learning.	22:21
	← cbreak left	22:21
	→ standyro1 joined	22:21
	← kossy left	22:21
	← bernalex left	22:21
rawtaz	thats good :)	22:21
	← harobed left	22:22
	← eletuchy left	22:22
	→ bernalex joined	22:22
	← arif-ali left	22:22
	→ kossy joined	22:22
	→ eletuchy joined	22:23
	← gnu-man left	22:23
	← hobodave left	22:23
	→ cbreak joined	22:23
	→ ambodi joined	22:23
	← bernalex left	22:23
	→ bernalex joined	22:23
	← rendar left	22:24
	← vkrest left	22:24
AciD`	nobody ?	22:24
rawtaz	everyone	22:24
AciD`	idle	22:25
rawtaz	no way	22:25
	never!	22:25
	→ sudoman joined	22:25
	← tiglionabbit left	22:25
rj11	we're all stale	22:25
	← skorgon left	22:25
	← matheus23 left	22:25
AciD`	maybe my question wasn't that nooish ?	22:25
	*noobish	22:25
	← endiruna left	22:26
	→ Roots47 joined	22:26
	← craigp left	22:26
rawtaz	AciD`: im sorry i dont get it. if your repo is in foo/bar/blah and you want to move your project to /star/snar then you can just move it. the .git folder in the repo/project folder will move along with it	22:26
	thats one of the nice things, that everything is in a .git folder at the root of the repo	22:27
Roots47	Hey all, I'm having trouble with rebase -i ... I want to rename a bunch of commits, but there is a merge from master a few commits down which is causing trouble	22:27
rawtaz	Roots47: what happens/whats the trouble?	22:27
	← eletuchy left	22:27
Roots47	I'm running 'git rebase -i HEAD~12' ... selecting reword on the commits i want to rename, but something weird seems to happen with the rebase.. it displays the merge commits as no longer being a merge	22:27
	→ ciampix_ joined	22:28
	← ciampix left	22:28
AciD`	rawtaz > let's say I have my files in /old/proj/disorganized. then if I follow you I should copy paste all those files and folder structure into the new (and cleaner) folder structure, then use `git mv` to move the files where I actually want them to be, then commit -a, then rename my files, then commit all again ?	22:29
	→ darklajid joined	22:29
	→ Vivekananda joined	22:29
AciD`	(including the .git folder ofc)	22:29
Vivekananda	on git commit and it says vi is not installed	22:29
AciD`	(then delete the empty folders	22:29
	→ endiruna joined	22:29
rawtaz	AciD`: holy shit i have no idea what youre saying. first of all, what is the root folder of your project and what is the root of your repo (i.e. where is the .git folder located)?	22:29
Vivekananda	how do I choose an editor	22:29
osse	Vivekananda: git config --global core.editor foo	22:30
	← donmichelangelo left	22:30
	← SocalNick left	22:30
	→ Mamoru joined	22:30
	→ donmichelangelo joined	22:30
	← sudoman left	22:31
darklajid	Hi there. Looking at the git datamodel right now. I understand the format for blobs and trees, but commits are still.. unclear. I assummed that commits are just texts in the mandatory tree/parent/author/committer/message form (as in cat-file's output) and stored just like blobs (header: "commit $length\0"), but I fail to verify that in my tests. Any idea what I might be missing?	22:31
	← mvelioglu left	22:31
	→ SocalNick joined	22:31
	a → o	22:31
	→ BSaboia joined	22:32
AciD`	rawtaz > my project root folder is /var/www/foobar, I want to move files like /var/www/foobar/index.php into a newly created ~/dev/web/src, and /var/www/foobar/dir1/dir2/foo.php into a newly created ~/dev/web/src/blah/, while retaining the history for those 2 files	22:32
	→ aard_ joined	22:33
	← aard_ left	22:33
	o → xnrand	22:33
Vivekananda	oose. git commit says : cannot start vi no such file or editor but on trying just $: vi the vi starts up. What is wrong here	22:33
	← pheaver left	22:33
bernalex	can I add a sign-off commit without reseting author date?	22:34
AciD`	the repo is currently in /var/www/foobar/.git	22:34
Vivekananda	osse: sorry	22:34
rawtaz	AciD`: that is two separate repositories, how do you expect them to keep history?	22:34
bernalex	oh right, amend doesn't reset date by default, I forgot, sorry.	22:34
	← pbujok left	22:34
rawtaz	it might be doable somehow though. but i dont know how. but there are hits on google	22:35
	→ amcsi joined	22:35
rawtaz	AciD`: you can always just copy the entire repository though, but im not sure if thats what you want	22:35
AciD`	rawtaz > that's the point of my question, as it's basicly just some move/rename files, I'd like to keep all the changes I made to those files	22:35
osse	Vivekananda: sounds like an environment issue (weird PATH or something). I don't know sorry	22:35
	Vivekananda: is vi actually an alias?	22:35
	darklajid: interestingly I cannot find much documentation on it.	22:36
AciD`	rawtaz > there are indeed numerous hits on google, but you need to master git to fully understand those (that's why I'm reading git docs for hours now..)	22:36
darklajid	osse: Neighet can I so far. And reading the C source isn't quite as easy if you're not familiar with the codebase AND the language..	22:37
	← thevishy left	22:37
	→ yhager joined	22:37
	→ p3lim joined	22:37
rawtaz	AciD`: well i dont think i understand the problem. do you basically want to copy your entire project, or do you want to just copy a few of the files from one repo/project into another existing one?	22:38
osse	darklajid: I would make a very simple commit and then open the file in a hex editor	22:38
	→ vkrest joined	22:38
	← zumba_addict left	22:39
frogonwheels	AciD`: Git does nottrack individual file history; it tracks snapshots of all files within the directory structure.. So the only way of doing this would be to start off with a copy of the repo filter-branched down to just those files..	22:39
	← valgrind left	22:39
darklajid	osse: Well, I cannot even find that file unfortunately :)	22:39
osse	darklajid: no wait. They're compressed.	22:39
	darklajid: unless you recently made a commit then it's likely packed	22:39
	ie. not directly available in .git/object/abcdef1234	22:39
frogonwheels	Vivekananda: what does 'which vi' give you?	22:40
AciD`	rawtaz > let me explain a bit more ; when I started the project, I never heard of the yeoman tools, and build my directory structure by hand. Now I used `yo` to scaffold a brand new directory/file structure. I now need to 'copy-paste' some of my source code in the right folders (and hoped to be able to keep the history for each file)	22:40
darklajid	osse: I just created a commit. I don't find a file that corresponds to the commit though	22:40
Vivekananda	frogonwheels: which is not found but whereis : whereis vi	22:41
	vital: /usr/share/man/man1/vi.1.gz	22:41
AciD`	frogonwheels > ok, I'll research filterbranching	22:41
	→ craigp joined	22:41
osse	darklajid: you know about the "two first characters of the SHA is the directory" thing right	22:41
	Vivekananda: ''type vi''	22:41
	→ nZac joined	22:41
frogonwheels	AciD`: the other option is just to take a copy of what you've got, then make a single commit that switches over to your new structure	22:41
	← Kayra left	22:42
AciD`	frogonwheels > what do you mean ?	22:42
	→ l0x3py joined	22:42
	← chadi left	22:42
frogonwheels	AciD`:Question - is this new copy like a second go at the same project - with the new tools?	22:42
	← alezost left	22:43
	← CKyle left	22:43
darklajid	osse: I grepped the whole tree for the commit message. Now I see that the commit file is totally unreadable. Hrm..	22:43
	← afhammad left	22:44
	← aslaen left	22:44
AciD`	frogonwheels > sorry I'm not sure I understood your question (I'm not a native english speaker) are you asking me if I plan to still modify my file after copying them in the new project ? if yes, then the answer is yes	22:44
rawtaz	AciD`: i agree with frogonwheels. looka t it this way; your project is still that same project. you want to change the files and contents so it matches yeoman style instad of your manual one. so make those changes, and commit it	22:44
	→ nathancahill joined	22:45
AciD`	ah ok, I guess you can indeed first make the changes, THEN just move the folder elsewhere to the new location	22:45
	→ arif-ali joined	22:45
frogonwheels	AciD`: How about; make a clone of your old repo, then just clear all the working files and copy in the new structure with the old index.php and commit that whole change.	22:45
	← mizu_no_oto left	22:46
	← rath left	22:46
AciD`	sounds great, I'll try that !	22:46
	→ pheaver joined	22:46
rawtaz	AciD`: move+rename the folder /var/www/foobar so that it becomes ~/dev/web/src (and move all the contents with it of course). then make the edits you want to make in it, e.g. moving dir1/dir2/foo.php into blah/	22:47
frogonwheels	Vivekananda: ok, then type 'alias vi'	22:47
rawtaz	or what frogonwheels	22:47
	said	22:47
	← FortySix2 left	22:47
	→ whitby joined	22:47
	← matanya left	22:47
	← leeN left	22:48
AciD`	rawtaz, frogonwheels > should I use basic `mv` cmds, or the one bundled with git `git mv` ?	22:48
	← bernalex left	22:49
	→ Mars__ joined	22:49
	← longbeach left	22:49
	→ phracker joined	22:49
rawtaz	i give up, sorry.	22:49
frogonwheels	AciD`: Personally, I'd just use standard mv commands and fire up git gui to commit all the changes.	22:50
rawtaz	to move the repo you just move the base folder as usual	22:50
	→ bernalex joined	22:50
rawtaz	to move files within, you can for simplicity's sake use `git mv`	22:50
AciD`	ok, thanks	22:50
	← jerius left	22:50
rawtaz	that will stage the move for you	22:50
	Mars__ → Mars`	22:51
osse	darklajid: they're zcompressed	22:51
	darklajid: zlib	22:51
	→ Finkelmann joined	22:52
	← bernalex left	22:52
	→ bernalex joined	22:52
darklajid	osse: 'file' reports VAX COFF executable not stripped - version 5060 ;-)	22:52
	← kossy left	22:52
darklajid	Let me see if I can extract that thing	22:52
	→ kossy joined	22:52
osse	darklajid: http://stackoverflow.com/questions/3178566/deflate-command-line-tool	22:53
	← ariscop left	22:53
	← SocalNick left	22:53
	→ skorgon joined	22:53
osse	bed	22:54
	← acharlieh left	22:55
	← ngladitz left	22:55
	→ SocalNick joined	22:55
	← kossy left	22:56
	→ kossy joined	22:56
darklajid	osse: Thanks	22:57
	→ wongon joined	22:59
	→ mizu_no_oto joined	22:59
	← Silmarilion left	23:00
	← Dougie187 left	23:00
	← skorgon left	23:01
	← mizu_no_oto left	23:02
	← nedbat left	23:02
	→ jabowering joined	23:02
xiong	I've heard it said it's bad practice to use tags on anything except released versions. Anybody care to discuss this?	23:02
	← Malice left	23:03
	← danman left	23:04
	← pheaver left	23:05
	← el3ctron3 left	23:05
	← el3ctron1 left	23:05
	← el3ctron4 left	23:06
	← becks__ left	23:06
	← smashwilson left	23:06
	← robban666 left	23:06
	→ becks__ joined	23:06
	→ BullSherd joined	23:07
BullSherd	Wow, what a things of google. http://goo.gl/pF3kWN	23:07
	Google is funny haha	23:08
	← BullSherd left	23:08
	← netcarver left	23:08
	→ thierryp joined	23:08
	← rjen left	23:09
rawtaz	funny guy	23:09
	← cozby left	23:09
	← devbug left	23:10
	← becks__ left	23:11
	→ danecando joined	23:12
	→ fragamus joined	23:12
	→ Norm__ joined	23:12
	← thierryp left	23:13
	← dsdeiz left	23:14
	← Norm_ left	23:15
xiong	???	23:17
	← fragamus left	23:17
xiong	Tags, no tags? No opinion??	23:17
helterscelter	personally I use tags for a couple different reasons.. but most of those tags never leave my repo	23:19
	→ skorgon joined	23:19
helterscelter	my public tags only refer to commits for which I have delivered "something"	23:19
xiong	Well... I don't have a dog in the fight yet. What you are saying is what I've heard from others.	23:19
helterscelter	generally a TAG is supposed to be immutable.. though you can change them -- many many times if you choose --	23:20
xiong	Agreed.	23:20
helterscelter	some projects choose to have a "latest" tag that moves along with their dev workflow	23:20
	so the latest gets updated "every release"	23:21
	the interesting thing to be about this is "latest" still refers to a delivery	23:21
	← v0n left	23:21
xiong	Agreed, that is, that one should not alter tags. I think of a tag as a fixed pointer, some sort of promise that the thing is something.	23:21
helterscelter	that is how I treat them as well -- annotated tags anyway	23:21
	→ tiglionabbit joined	23:21
	← calcifea left	23:21
xiong	Well, what about lightweights?	23:21
helterscelter	lightweights never make it out of my repo	23:22
	I use them for internal keeping track of things	23:22
	mostly they are short lived	23:22
	especially useful with iterative rebase sessions	23:22
	with lots of possible reording and merge conflicts	23:22
	→ pheaver joined	23:23
xiong	I don't think I've ever rebased except to fix a mistake in my use of git. Not sure the idea appeals to me.	23:23
helterscelter	I may choose to throw away a couple iterations of squashed/reordering depending on how the history is turning out.. and having a light tag sitting there for some "I know I'm going to go with this state" is handy	23:23
	reflogs also give you the same thing.. but not as easy to remember where you were	23:24
	← eqyiel left	23:24
xiong	If two of us have checked out 'devel' and worked on different features, one of us will have to re-pull devel and rebase; I understand that. Otherwise I don't see it.	23:24
helterscelter	rebase is a very powerful thing, once you wrap your head around it	23:24
	not necessarily	23:24
	← Finkelmann left	23:24
helterscelter	you don't HAVE to rebase.. you could merge	23:24
xiong	I really do not like the idea of squashing. Anything that alters history worries me.	23:25
helterscelter	ideally you'd not do your development in your local repo against the actual "devel" branch.. you'd create another branch off of it to keep your "devel" tracking clean	23:25
	← revele left	23:25
xiong	Exactly. Always.	23:25
helterscelter	rebase is handy for more than squashing	23:25
	some times it's handy to pull-apart a commit into several or re-order the history a little	23:25
xiong	http://imgur.com/7cAou	23:25
helterscelter	yep	23:26
	that branching workflow is very handy	23:26
	hard to get folks used to legacy (svn) workflows to grok it tho	23:26
	→ madewokherd joined	23:26
xiong	I've had some arguments about it. I say, I didn't swallow it whole; I modified it until it worked for me. You can do the same, okay.	23:27
helterscelter	yep	23:27
	thats what ended up happening	23:27
	← skorgon left	23:27
xiong	My primary motive was to minimize the SVG file size; the original looks much better but the fuzzy circles cost more.	23:27
helterscelter	for my team we ended up being forced to have multiple develop lines concurrently, so that we could align with the business's planned release schedule..	23:28
	← craigp left	23:28
helterscelter	so we "changed it up" some to make the flow work for us and the "business"	23:28
	→ polyzen joined	23:28
	→ eqyiel joined	23:29
	← Exancillatus left	23:29
	← dglambert left	23:29
	→ bhldev_ joined	23:30
	← bhldev left	23:30
xiong	Sorry, that was the wrong link: http://imgur.com/EGOzs	23:30
	← bernalex left	23:31
	→ bernalex joined	23:32
	← tuocuggino left	23:33
	← aberrant left	23:33
	← motto left	23:33
	→ afhammad joined	23:33
	← dougsland left	23:34
	← nZac left	23:34
	→ dougsland joined	23:34
	→ eletuchy joined	23:34
	← cogitos left	23:35
helterscelter	the only diff between that and what my team does, is we don't use develop -- our "release" branches are our "develop" branches and we merge-forward periodically between releases to keep future target dates up-to-date with "next release" work	23:35
	← jabowering left	23:35
	→ amcsi_ joined	23:35
	→ dsdeiz joined	23:36
helterscelter	we also find it handy to tag on the release branch(es) to designate QC and UAT code drops.. so we can easily tell when a drop was made and what changed between drops	23:36
	→ whatever___ joined	23:36
xiong	Okay, well, so one shortcoming of that image is that all the little balloons don't represent the same things. The balloons along support- and trunk represent actual tags; the others are just annotations from me to you.	23:36
helterscelter	ya	23:36
	→ vince_prignano joined	23:36
	← ojacobson left	23:37
	→ sanmarcos joined	23:37
sanmarcos	is there a quick command to "cat" a file before all my changes, cat whatever is at head?	23:37
	→ p4denis joined	23:38
xiong	My current project is a super/sub tool; I've been studying up and discussing various aspects here. I see a use for tags to mark commits in subs, to identify commits made at the same moment in time.	23:38
	← max_weller left	23:38
	← phile left	23:39
	→ max_weller joined	23:39
	← amcsi left	23:39
	← Not_Whatever left	23:39
xiong	Um, therefore $ rhea tag 'foo' and each sub is so tagged.	23:39
	← pheaver left	23:39
xiong	I fear to go crosswise with the school that advises against tags -- to one degree or another.	23:40
helterscelter	for "sub" you mean sub-projects?	23:40
	or git submodules?	23:41
	→ CKyle joined	23:41
xiong	Same thing, in my case -- I think. The more I look at them, the thinner submodules appear.	23:41
	← berzerka left	23:41
	→ ariscop joined	23:41
	← p4denis left	23:42
	← eijk left	23:42
	← amcsi_ left	23:43
xiong	If I always push --follow-tags then lightweight tags will not be pushed, correct? Then perhaps rhea-tag should limit itself to lightweight and we avoid half the objections at least.	23:43
helterscelter	for git submodules -- which I'll readily admit I don't use currently -- you don't need to tag submodules if you tag the parent	23:43
	yes --follow-tags only pushes annotated tags	23:44
xiong	Well... "need"?	23:44
	If I'm working in a sub then I'm ignorant of the parent context.	23:44
helterscelter	the current commit that the submodule is on is recorded as part of the commit in the parent	23:44
	← danecando left	23:44
helterscelter	so a tag on the parent is effectively a tag on the child	23:44
	← vince_prignano left	23:45
helterscelter	unless you need to interrogate the state (ie tag) in the child individually, there is no reason to tag both	23:45
xiong	Um, through a layer of indirection. If I'm in subA and want to know which commit was made in subB at the same time, it's up and down.	23:45
	→ vince_prignano joined	23:45
	→ vurma joined	23:45
helterscelter	true.. tho I wonder why you'd need to know that	23:45
	;]	23:45
xiong	Well, they wouldn't be in the same super if they weren't related.	23:46
helterscelter	not necessarily	23:46
	suppose the super is a project "A" and subs "B" and "C" are libraries that "A" uses.. B and C aren't related in that case	23:46
xiong	I'm going to have to take a break from actually working on the issue and write it up.	23:46
	← nnyk_ left	23:46
xiong	I usually do that, if not first then soon; but it's taken awhile just to clarify the matter in my head.	23:47
	→ Fizzbot joined	23:47
xiong	In short, in my particular case, there's at least one change I can picture very sharply that needs to be made in every sub and for the same reason.	23:48
	← plasticle left	23:49
Fizzbot	Hi all, trying to get git installed onto OSX Mavericks.. but the installer is coming up as damaged and unable to install. what's the way to deal with this?	23:49
	← aubreypwd left	23:49
	→ nnyk_ joined	23:49
	← Acidnerd left	23:49
	← tvw left	23:49
	→ MacWinner joined	23:52
	→ Kayra joined	23:52
	← edwardly left	23:53
	← Fizzbot left	23:54
	← whitby left	23:55
	← Kayra left	23:57
	← SHyx0rmZ left	23:57
helterscelter	ttyl	23:58
	← helterscelter left	23:58
	← bernalex left	23:58
	← nszceta left	23:59
	→ pheaver joined	23:59
	← shruggar left	23:59
	→ bernalex joined	23:59
	← g3orge left	23:59

Logs Search ←Prev date Next date→

Channels Documentation