IRCloggy #git 2012-05-30

Logs Search ←Prev date Next date→ Channels Documentation

Provider of IRC logs since 2005.
WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

2012-05-30

← mmc left00:00
frogonwheels comps: but cherry-pick doesn't handle failure, continuing from failure .. or (I believe) the -i00:00
→ Anxi80 joined00:01
TheShagg for this situation, i would probably use a three aguement rebase00:01
← dreiss left00:01
← TeckniX left00:01
frogonwheels TheShagg: still need to create the temp branch so the original branch remains intact.00:02
→ rchavik joined00:02
← rchavik left00:02
→ rchavik joined00:02
TheShagg frogonwheels: yes00:02
although the help is not telling me how to do three arguement rebase00:02
frogonwheels TheShagg: .. and I actually know the --onto form :)00:02
comps frogonwheels: no, it doesn't00:02
frogonwheels: git rebase --onto HEAD~2 HEAD~7 HEAD~5 works fine for me, while being on (no branch)00:02
→ cybersphinx joined00:02
← b1rkh0ff left00:03
← chrisgeorge left00:03
frogonwheels comps: so it just leaves it as a detached head? interesting. Still, for my money, it's easier to grok with a temp branch name00:03
TheShagg i remember doing git rebase --onto=new_base oldest^..newest00:03
comps I often do things like git rebase --onto v3.3.7 v3.3.4 master since I have several commits on top of the vanilla linux kernel00:04
frogonwheels TheShagg: does it update new_base ?00:04
TheShagg in a pinch you could open up two sessions of git rebase -i, and copy commits from one into the other00:04
hehe00:04
drags frogonwheels: using this strategy when I finished editing the picklist for the interactive rebase and it begings to actually replay the commits: it stops at every commit saying "Could not apply SoMeShA... commit message"00:04
except there's no merge conflict to cleanup00:05
I can then do rebase --continue to get by it, but I have to imagine the commit is then not being applied00:05
TheShagg new_base..... no I think it it actually updates the current branch, not the newbase00:05
← the_cybersphinx left00:05
← fly9 left00:05
frogonwheels TheShagg: exactly, and he wanted to keep the current branch as it was.00:05
→ fly9_ joined00:05
TheShagg right00:05
← Myhro left00:06
TheShagg so yes, checkout the current branch, type "branch newbranch" then "checkout newbranch", and then do the rebase00:06
→ grantbdev joined00:06
TheShagg as you probably said already00:06
← lnostdal left00:06
frogonwheels :)00:06
TheShagg I've actually been doing a lot of this00:06
cleaning up commit hell from my coworker00:07
→ Myhro joined00:07
← andrew__ left00:07
TheShagg eventually I started doing git rebase -i00:07
much better way of working00:07
and git add -p00:07
→ rvsjoen joined00:07
← rvsjoen_ left00:08
→ lnostdal joined00:08
← btx94 left00:09
frogonwheels TheShagg: i've been using rebase to copy between 2 conversion of the same code (converted from dev and release of a file-based VSS style RCS)00:09
← libertyprime left00:10
→ libertyprime joined00:10
→ edude03 joined00:11
TheShagg frogonwheels: keeping two parallel branches?00:12
→ JEntrep joined00:13
→ mritz_ joined00:13
← JEntrep left00:13
← mritz left00:13
← Chillance left00:14
← felgru left00:16
← Zenith77 left00:16
frogonwheels TheShagg: I'm trying to gradually wean off one of them. nearly there. only taken a year00:16
← libertyprime left00:17
← peppyheppy left00:17
← panthrodeb left00:18
→ libertyprime joined00:18
→ Slim joined00:18
← berserkr left00:18
← p3rror left00:21
← ahayweh left00:21
milki commit hell~00:21
← libertyprime left00:22
→ libertyprime joined00:22
← Kabaka left00:23
→ davidbanham joined00:23
← dinduks left00:24
→ ahayweh joined00:25
← fuzzybunny69y left00:25
← libertyprime left00:25
frogonwheels milki: you talking about my situation? yeah. but it will be worth it in the end.00:25
→ libertyprime joined00:26
milki TheShagg's actually00:26
← markus_ left00:26
← notmypudding left00:28
→ jgoss joined00:29
drags frogonwheels: thanks the temp branch strategy seems to have worked, unfortunately this is puppet configuration and not buildable (testable) code, so I'll find out in a bit if I stepped on my own toes anywhere, but at least the merge went through, woo00:30
→ sgronblo joined00:30
→ fp1 joined00:31
← DoniRS left00:31
→ vitor-br joined00:33
← nowhereman left00:33
← vitor-br left00:33
→ vitor-br joined00:34
→ nowhere_man joined00:34
→ p3rror joined00:34
jexmex if you delete branches how do you push those deletes through?00:35
bob2 it's not conceptually really pushing at all00:36
it's just that 'push' is about the only command that mutates a remote repo at all00:36
← kenperkins left00:36
jexmex so do I just commit my master and push?00:36
bob2 ?00:37
jexmex just wondering how to tell the remote to remove those branches I just deleted00:37
bob2 if you want to delete remote branch, 'git push :zomg'00:37
comps git push remotename :refs/heads/branchtodelete00:37
→ notmypudding joined00:39
jexmex ok thanks00:39
→ Davey_ joined00:39
← boombatower left00:40
→ whompwhomp joined00:40
whompwhomp how can i replace the contents of my git repo folder with the folder on the server?00:41
jexmex git push remote :branchname is all that is needed00:41
→ bruno_coelho joined00:41
→ kaos01 joined00:41
comps whompwhomp: clone --mirror .. ?00:41
kaos01 hi how do i "clone" a git repository into a non empty directory ?00:42
the directory has .gitignore and data files in it00:42
comps kaos01: why would you want to do that?00:42
kaos01 true actually but unfortunatly the directory already contans the data files related to the project00:43
if it was in a parent directory it would be easier :)00:43
comps just clone the project to some other dir and copy those files over .. ?00:44
← makubi left00:44
whompwhomp cloning the whole repo takes a long time00:44
i want to clone just one folder00:44
comps that was for kaos0100:44
→ Per4u3e joined00:44
comps whompwhomp: and you can't clone just one dir like svn can00:45
whompwhomp when i write git clone --mirror, it can't recognize the command00:45
comps whompwhomp: you can fetch one ref (like one specific branch) though00:45
← marcogmonteiro left00:46
← wilmoore left00:46
→ g0bl1n joined00:47
← g0bl1n left00:47
→ g0bl1n joined00:47
← fly9_ left00:47
whompwhomp sigh00:48
← davidbanham left00:48
← ZanQdo left00:49
← aboudreault left00:51
→ makubi joined00:51
← ryez_ left00:52
← boo5 left00:53
TheShagg whompwhomp: why, do you have all of your companies projects in one git repo?00:53
→ thursday joined00:55
→ Jojo1234 joined00:55
drags TheShagg: like facebook? :)00:56
kaos01 comps, well copying it over is an option but not so great as everytime a change is made one nedst to copy the changes to be able to do the testing00:57
← jonshea left00:57
bruno_coelho → brunocoelho00:57
← abetaha_ left00:58
← grantbdev left00:58
→ fuzzybunny69y joined01:00
→ Kabaka joined01:00
← teweWORK left01:00
← Davey_ left01:00
→ tomkralidis joined01:01
→ xiangfu joined01:03
tomkralidis hi, my pull request has 5 submodules which (somehow) have been diffed (https://github.com/GeoNode/geonode/pull/248/files#diff-31), but I want to revert to what they are in master. Any advice on how to revert submodules? tia01:03
bob2 submodules :|01:03
← ColKurtz left01:03
→ erkules|away joined01:05
← erkules left01:07
← aantix left01:08
→ `Assassin joined01:08
← fisted left01:09
tomkralidis do I have to rebase the submodule's or something?01:09
→ dv310p3r joined01:09
← TTSDA left01:11
→ ivan12348 joined01:11
→ ivan12349 joined01:11
→ ivan12350 joined01:11
← ivan12347 left01:11
← ivan12345 left01:12
→ cenuij joined01:12
← cenuij left01:12
→ cenuij joined01:12
← TheShagg left01:12
→ Davey_ joined01:13
→ TTSDA joined01:14
← ivan12346 left01:14
← libertyprime left01:15
→ HyperNerdV2 joined01:15
→ libertyprime joined01:15
← asedeno_work left01:17
→ tchan joined01:18
→ tmcmahon joined01:19
← hakunin left01:19
← tchan1 left01:19
← Lgb left01:20
← libertyprime left01:20
→ intripoon_ joined01:20
→ libertyprime joined01:21
ProLoser|Work tomkralidis: stilll there?01:21
tomkralidis hi ProLoser|Work01:21
ProLoser|Work did you figure it out?01:21
tomkralidis not yet01:21
ProLoser|Work so what do you want? the submodules to be on the latest version of their master? or you want them to be at the state the parent recognizes them at?01:22
→ Lgb joined01:22
→ asedeno_work joined01:22
ProLoser|Work s/parent/primary repo/01:22
← Vampire0 left01:22
→ hakunin joined01:23
tomkralidis my pull request somehow committed changes to submodules (https://github.com/GeoNode/geonode/pull/248/files#diff-31). I need to take the submodule changes out of the pull request/commit.01:23
so basically go back to that commit for each submodule01:23
ProLoser|Work that's a huge commit01:24
← intripoon left01:24
ProLoser|Work you can ammend the commit01:24
tomkralidis so I want it such that those commits on the submodules never happened.01:24
ProLoser|Work or just make a new commit with the submodules where they should be01:24
← alexim left01:25
ProLoser|Work change the submodules to what they should be and then commit the parent project01:25
you can just do git checkout 4c6d2ce285cfa000ed0ae108f7ac6ce5d7a804b501:25
or what-have-you for each submodule01:25
tomkralidis I get fatal: reference is not a tree01:26
→ flippo joined01:26
← fuzzybunny69y left01:28
← thearthur left01:29
→ lunchub_ joined01:29
← g0bl1n left01:30
→ tonesfrommars joined01:31
← tewecske left01:32
← pheaver left01:32
← klj613 left01:32
→ klj613 joined01:32
→ lfaraone_ joined01:33
← klj613 left01:33
→ klj613 joined01:33
← eletuchy left01:33
← mocramis left01:34
→ mocramis joined01:34
SethRobertson tomkralidis: You can look at !fixup instructions on reverting commits, even commits of submodule SHA1 changes.01:34
gitinfo tomkralidis: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!01:34
← domino14 left01:36
→ g0bl1n joined01:36
← g0bl1n left01:36
← twisted` left01:36
→ Vampire0 joined01:36
ProLoser|Work tomkralidis: you have to show your folder structure and where you're running it from01:37
← libertyprime left01:38
→ libertyprime joined01:38
→ amigojapan joined01:38
→ zdk joined01:42
→ KiiK joined01:43
← Bartzy left01:43
← Per4u3e left01:44
→ Spudster joined01:44
KiiK I do a "git clone -l A B" on same dir. but it doesn't use hardlinks ?01:44
→ wilmoore joined01:45
offby1 KiiK: I bet it is doing so, but you're not looking in the right place01:45
→ abetaha joined01:45
offby1 !repro01:45
gitinfo Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session, or at least explain exactly what you did that led up to the problem. This will help immensely with troubleshooting.01:45
← Davey_ left01:45
→ gavin_huang joined01:45
frogonwheels KiiK: ie. what are you looking at that suggests it isn't using hardlinks?01:46
← aristidesfl left01:46
offby1 I think we both know ... :)01:47
but he's playing coy01:47
frogonwheels offby1: :)01:47
→ aristidesfl joined01:47
KiiK im cloning a local repo and want to saving the space.....01:48
frogonwheels KiiK: yes.. and .. what makes you think it's not?01:48
offby1 stretches and settles in ... it's gonna be a long night01:49
→ alberto56 joined01:49
KiiK frogonwheels: im looking at the space & ls -l, no symlinks01:49
frogonwheels KiiK: I suspect ls -l .git/objects/pack/ may be enlightening01:49
offby1 whistles innocently01:50
offby1 also !man git-clone01:51
gitinfo the git-clone manpage is available at http://git-scm.com/docs/git-clone.html01:51
frogonwheels and offby1 await the dropping of the penny.01:51
→ Mark___ joined01:51
→ chaowang joined01:51
frogonwheels KiiK: you on windoze?01:52
KiiK https://gist.github.com/283255901:52
tomkralidis ProLoserWork|Work: structure is https://github.com/GeoNode/geonode/tree/master/src/geonode-client/externals01:52
so cd ringo and git checkout <SHA>, and get reference is not a tree01:53
offby1 KiiK: if you know how to read the output from "ls", the answer is right in front of you: the two big "pack-*" files have more than one link01:53
SethRobertson KiiK: Yeah, that isn't how hard links work. du will report the same size, but df will show that no more space is added01:53
← w09x left01:54
ProLoser|Work anyne use sublime2?01:54
frogonwheels KiiK: once there's a hard link, there's no real distinction over the 'original'.. they're just links in the directory to the same inode afaik01:55
tomkralidis ProLoser|Work: structure is https://github.com/GeoNode/geonode/tree/master/src/geonode-client/externals01:55
ProLoser|Work where are you calling the method from?01:56
*command01:56
tomkralidis cd ringo/ && git checkout <SHA>01:56
← Araxia_ left01:56
SethRobertson KiiK: Example: df .; ln bigfile otherfile; df .; cp bigfile otherotherfile; df .01:57
ProLoser|Work try git status01:57
git remote -v01:57
etc01:58
tomkralidis ah, on (no branch). Check out master and then git checkout <SHA> ?01:58
ProLoser|Work well i figured you weren't on a branch01:58
but that shouldn't have mattered01:58
go ahead and try checkout master01:58
and try pulling01:58
then try checking out the sha01:58
← alberto56 left01:58
ProLoser|Work although fetch + checkout sha should work in theory too01:58
tomkralidis same result (fatal: ref is not a tree)01:58
ProLoser|Work how did you put the submodule into the project to begin with?01:59
→ newbie999|3 joined01:59
ProLoser|Work just copy paste or git submodule add?01:59
← dwmw2 left01:59
→ erkules joined01:59
← goshawk left01:59
← erkules|away left02:01
← jhford left02:01
← whompwhomp left02:01
→ dwmw2_gone joined02:01
→ peppyheppy joined02:01
→ jhford joined02:01
KiiK thanks @sethRobertson, forgonwheels, offby1 :)02:01
← newbie999 left02:01
← mlt- left02:02
← klj613 left02:03
← peppyheppy left02:03
→ boombatower joined02:03
← toothr left02:04
tomkralidis oh wait, I had to git pull, then git checkout SHA, that worked, thx02:06
← dv310p3r left02:06
← tomkralidis left02:06
→ Zenith77 joined02:07
→ peppyheppy joined02:08
← oriba_ left02:09
→ marhaban1 joined02:12
→ juvenal joined02:12
← marhaban left02:12
← jgoss left02:14
← rrn_ left02:15
→ AlbireoX`Laptop joined02:15
→ rrn joined02:16
→ w09x joined02:17
← mikepack left02:19
← ill1cit left02:19
→ ill1cit joined02:20
→ Chib joined02:20
← ProLoser|Work left02:20
→ dv310p3r joined02:21
← Peste_Bubonica left02:21
→ fisted joined02:24
← aristidesfl left02:25
← HyperNerdV2 left02:26
← apok left02:26
→ kenperkins joined02:26
Karmaon → doesUbuntuHateAw02:26
doesUbuntuHateAw → AwayNickAwayNick02:27
AwayNickAwayNick → Karmaon02:27
← kenperkins left02:28
→ zanneth joined02:29
→ thearthur joined02:29
← p3rror left02:30
← guampa left02:30
← Sky[x] left02:30
→ Per4u3e joined02:31
← Pookz left02:31
← dv310p3r left02:32
← ill1cit left02:32
→ ill1cit joined02:32
← w09x left02:33
← libertyprime left02:34
← zanneth left02:34
→ zanneth joined02:34
→ libertyprime joined02:34
→ glennpratt joined02:35
← k-man left02:36
← Caelum left02:37
→ herdingcat joined02:37
→ davidbanham joined02:38
← Zyrkon left02:38
← davidbanham left02:39
→ Zyrkon joined02:39
→ mikepack joined02:42
← mikepack left02:43
→ p3rror joined02:43
← Eridius left02:43
← peppyheppy left02:43
→ Davey_ joined02:44
→ mdhopkins joined02:45
→ ipoval joined02:47
→ ISF joined02:48
→ mikepack joined02:48
→ Eridius joined02:51
← palyboy left02:52
← KiiK left02:53
→ bubbles10101 joined02:54
← amigojapan left02:55
← Mark___ left02:56
← osmosis left02:59
← williamcotton left03:00
→ weiyang joined03:01
← koo3 left03:01
← libertyprime left03:01
→ libertyprime joined03:02
scientes does posix git push require mmap()?03:03
fatal: sha1 file './objects/pack/tmp_pack_XGJziB' write error: Operation not supported03:03
bremner "posix git push"??03:05
scientes to a file url03:05
rather than ssh://03:05
hence, POSIX03:06
← duckxx left03:06
→ koo4 joined03:08
← cccaldas left03:08
← Spudster left03:09
→ andrew_ joined03:09
← smw left03:11
→ kenperkins joined03:12
→ lonewulfcloud joined03:12
→ Hodo joined03:13
→ theCole joined03:13
→ davidbanham joined03:15
← drl left03:17
→ aristidesfl joined03:19
aristidesfl hi03:19
→ smw joined03:19
aristidesfl lets suppose I've committed some changes in branch B then merged them into branch A.03:19
← kukks left03:19
aristidesfl if I amend the commit in branch B, will the amend affect branch A as well?03:20
offby1 scientes: I don't know but strace will tell you03:20
→ dreiss joined03:22
Eridius aristidesfl: no03:24
← Internet13 left03:24
Eridius aristidesfl: commits are immutable objects. "Ammending" a commit really creates a brand new commit and changes the branch pointer to point at the new one instead of the old one. But anything else that referenced the old one (merges, other branch pointers, etc) will continue to point at the old one03:24
→ w09x joined03:25
aristidesfl Eridius thank you03:26
→ mischat_ joined03:26
← mischat left03:27
mischat_ → mischat03:27
→ drl joined03:31
scientes mmap(NULL, 613712, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f078fb2500003:32
close(3) = 003:32
bing03:32
it needs to test03:32
and then fall back03:32
or rather it should03:32
;)03:32
← davidbanham left03:32
← gerard0 left03:33
wereHamster it needs to test what03:34
← plasticboy left03:34
← cakehero left03:34
← Per4u3e left03:35
← kerrick left03:38
→ d4rklit3 joined03:38
← Davey_ left03:38
← kenperkins left03:39
scientes wit im not sure that was he issue03:41
← aristidesfl left03:41
scientes i only suspected it cause fuse doens't support mmap03:41
→ toothr joined03:42
→ aristidesfl joined03:43
wereHamster the mmap succeeded.03:43
→ kerrick joined03:43
wereHamster it returned a valid address.03:43
→ pheaver joined03:43
scientes oh03:44
← Eridius left03:46
→ barrbrain joined03:49
← Hodo left03:52
→ kenperkins joined03:53
← glennpratt left03:54
← vitor-br left03:56
← tmcmahon left03:56
→ Eridius joined03:57
← resmo left03:59
→ duckxx joined03:59
→ tshauck joined03:59
← marhaban1 left04:00
← scientes left04:00
→ singh_abhinav joined04:01
→ igotnolegs joined04:01
← mischat left04:01
← OOPMan left04:03
→ davidbanham joined04:04
← rchavik left04:04
→ dgs joined04:06
→ resmo joined04:06
dgs can you have two identically named git tags in two seperate branches? eg a v1.0 tag in both the development and production branches?04:06
or do tag names need to be globally unique?04:07
just curious to how if it would handle a checkout of the duplicate tag name - eg how would it know what branch to check it out from...04:08
← mdhopkins left04:08
offby1 nope04:08
tag names gotta be unique; and they're not in the least affected by what branch they're "on" (that doesn't even make sese)04:08
sense04:08
→ WAA3I0Z joined04:08
offby1 dgs: you're very confused about git's innards.04:08
I recommend you read "Git For Computer Scientists" (http://eagain.net/articles/git-for-computer-scientists/)04:09
dgs yup, that was pretty much what I figured04:09
just wanted to confirm04:09
bob2 tag = name for a commitid04:09
dgs k. thanks. I will have a read =)04:09
bob2 branch = moving label on a commitid04:09
offby1 s/name/label that typically doesn't move/04:09
← brunocoelho left04:10
→ renzhi joined04:10
bob2 heh04:11
→ beautifulmind joined04:11
→ Per4u3e joined04:12
renzhi hide, I have the following situation. I clone a remote repo (let's call it remote A) on my local machine, do a branch --track04:12
I meant "hi", sorry for that04:12
← madewokherd left04:13
renzhi nowhere_man, I'd like to push that branch to another remote repo (called remote B), and I'd like that remote B tracks remote A.04:13
how could I do that?04:13
→ scientes joined04:14
offby1 well, "git remote add B url://b/whatever"04:14
then "git checkout that-branch; git push -u B:that-branch"04:15
I think that'll do it04:15
← duckxx left04:15
← koo4 left04:15
renzhi :O04:15
let me try this04:15
→ koo4 joined04:16
offby1 I hope you can delete branches on B, in case I've given you bad advice :-|04:16
← singh_abhinav left04:16
← davidbanham left04:16
→ sangi joined04:18
← resmo left04:18
→ Lord_DeathMatch joined04:18
← optixx left04:18
renzhi I have full control of B, I just want to keep a copy on B (it's faster), but we need to share between more than one computer04:19
← d4rklit3 left04:19
renzhi B is supposed to be the hostname?04:20
→ optixx joined04:20
gitinfo set mode: +v04:21
← batrick left04:21
← theCole left04:22
bob2 no04:22
← koo4 left04:22
bob2 git remote add someonename someurl04:22
← Goplat left04:22
← weiyang left04:23
kaos01 is there a way to see what changes are in a remore repo without doing a fetch first ?04:25
← scientes left04:26
← beautifulmind left04:26
← mtytel left04:26
sitaram not without logging on to the remote machine04:28
← Bass10 left04:28
← aristidesfl left04:29
→ maleknet joined04:29
← d_roge left04:30
← mikepack left04:31
→ mikepack joined04:31
← Textmode left04:32
← Slix` left04:33
← p3rror left04:34
→ batrick joined04:34
← mikepack left04:36
← iwoj left04:37
→ fisted_ joined04:38
→ raatiniemi joined04:38
→ ColKurtz joined04:39
→ scientes joined04:39
← fisted left04:39
← codeM0nK3Y left04:42
← raatiniemi left04:42
← edude03 left04:44
→ weiyang joined04:44
→ codeM0nK3Y joined04:44
← Vert left04:44
→ edude03 joined04:45
← codeM0nK3Y left04:45
← _huoxito left04:46
→ theCole joined04:46
→ p3rror joined04:48
→ Textmode joined04:48
← thearthur left04:48
→ _huoxito joined04:49
→ SegFaultAX joined04:50
→ mangdood joined04:51
→ glennpratt joined04:51
← meisth0th left04:54
← Robbo_ left04:55
→ WillMarshall joined04:55
→ rsewill joined04:55
← xiangfu left04:55
← w09x left04:56
← makeawish left04:56
← boombatower left04:56
← EyesIsMine left04:56
→ fuzzybunny69y joined04:57
→ Hamled joined04:58
Hamled What does it mean when I use the git branch --contains command and the response is "* (no branch)"04:58
← Textmode left04:59
→ shruggar joined05:00
→ peppyheppy joined05:00
renzhi when I do "git push -u B:that-branch", it's always trying to consider B as a hostname and connect to that host05:01
← glennpratt left05:02
← fuzzybunny69y left05:03
wereHamster Hamled: probably because HEAD contains that commit05:03
renzhi: git push <remote> <refspec>05:03
→ fuzzybunny69y joined05:03
wereHamster renzhi: you probably want git push -u origin B:that-branch05:03
renzhi wereHamster: I just do "git push --all url://repo-B"05:05
infogulch → infogulch|afk05:09
mangdood What's a noob-friendly diff/merg tool? I'm having some difficulty understanding what exactly is happening when I'm merging05:10
→ FreeArtMan joined05:10
bob2 your editor probably05:10
→ williamcotton joined05:11
→ Textmode joined05:11
EugeneKay SmartGit's 3-pane merge view works pretty good.05:11
bob2 most merges are conflictless, when they do conflict, open up the files in your editor and look at them05:12
merge by hand05:12
then check the diff05:12
→ Robbo_ joined05:12
← SegFaultAX left05:13
→ cbreak-work joined05:13
→ SegFaultAX joined05:13
→ j416 joined05:13
mangdood bob2: I'm just working with a local repos of a single file, and I realized what was confusing me; I made a change in one of my commits that I wasn't supposed to..05:13
← lunchub_ left05:13
→ mmc joined05:15
→ meisth0th joined05:15
← dhruvasagar left05:15
← mmc left05:19
← gtrak left05:22
→ fuzzybunny69y1 joined05:23
Hamled Is there something I can add to a refspec in a git pull command to cause the merge to only include the individual commit specified in the refspec05:24
→ mishok13 joined05:25
← rsewill left05:25
j416 Hamled: to apply just a single commit, you can use git cherry-pick05:25
Hamled thanks05:25
→ Dave^| joined05:25
← fuzzybunny69y left05:25
fuzzybunny69y1 → fuzzybunny69y05:25
bob2 merge merges branches, it has no concept of merging a single commit05:25
j416 it will replicate that commit to a new one.05:25
bob2 unless you've only diverged by one commit05:25
mangdood I made an error in one of my commits; I was working with an old copy I believe (because of my editor). Is there a way I can "merge" this commit with the previous to reverse my mistake?05:25
→ resmo joined05:25
Hamled bob2, ah, thanks05:25
bob2 did you push it anywhere05:26
j416 mangdood: have a look at 'git revert', it can record a commit with the opposite effect of a selected commit or series of commits05:26
← lindenle left05:26
← robbyoconnor left05:27
← stkrzysiak left05:28
← Slim left05:28
→ robbyoconnor joined05:28
→ Slim joined05:28
→ w09x joined05:29
mangdood j416: does revert only revert whole files, or does it allow partial changes?05:29
kevlarman mangdood: revert operates on commits not files05:30
← lapistano left05:30
→ lapistano joined05:30
→ Dragooon joined05:30
j416 mangdood: you can use 'git revert --no-commit' and edit what parts are being reverted, then commit the result yourself.05:31
mangdood kevlarman: It feels terrible to not know the correct terminology, even after doing a tutorial.05:31
j416 but it can of course be confusing with partial reverts. :)05:31
don't worry, we've all been newbies.05:31
:)05:31
kevlarman mangdood: revert basically takes the diff of the commit you tell it to, applies it with -R, and commits the result if there were no conflicts05:32
→ davidbanham joined05:33
→ benjo2 joined05:33
→ airborn joined05:33
← FreeArtMan left05:33
→ singh_abhinav joined05:35
← project2501a left05:35
← zendeavor left05:35
→ qawsedrf joined05:36
→ Guest31877 joined05:36
qawsedrf in my project, one of my folder has subfolders and files. it's weird that they do not exist on remote repo, howcome ? :O05:37
Guest31877 → automage05:38
← Textmode left05:38
mangdood Okay, so git revert doesn't make any changes for me except the conflicts. This is rather confusing05:38
automage → zendeavor05:38
← singh_abhinav left05:40
mangdood ...let me rephrase that in a less confusing way: git revert isn't making any changes to my file in my working directory (only one file :P in the repos) except their conflicts.05:41
→ Textmode joined05:41
wereHamster mangdood: can you pastebin the output from git status?05:42
qawsedrf someone care to help me with this - http://pastie.org/private/edfzwxmoefop9xewcw1q ?05:42
← khmarbaise left05:42
← benjo2 left05:43
mangdood wereHamster: http://pastebin.com/E1H9pNgB05:43
← gavin_huang left05:43
→ tomoj joined05:43
→ warthog9 joined05:44
qawsedrf anyone ? :(05:44
mangdood qawsedrf: I looked it at, but I'm no help05:45
wereHamster mangdood: that's a conflict. Fix it, then commit.05:45
qawsedrf i am having problem on live deployment05:45
need to solve it quick05:45
wereHamster qawsedrf: cd libs/vendor/fb_php_sdk; git add *; ...05:45
though I'm not sure if that's really what you want...05:46
→ martinjlowm joined05:46
qawsedrf it should had been part of the entire repo05:46
well, i just got cloned from github05:46
wereHamster but it's not.05:46
it's a submodule05:46
mangdood wereHamster: I understand, but I expected that when I reverted, all the changes that don't conflict should be reverted, yes?05:46
qawsedrf now when i noticed its not on server, i delete libs/vendor/fb_php_sdk/.git05:46
wereHamster mangdood: correct.05:47
→ eletuchy joined05:47
wereHamster mangdood: so it looks like all the changes from the commit conflicted05:47
qawsedrf git add does nothing05:47
it omplains bla blka is in submodule05:47
mangdood wereHamster: In one of the files, it only lists the the conflicts and didn't revert the other parts of the file to the file of the previous commit (forgive me for my crap use of terms)05:48
qawsedrf wat te hell05:48
← mocramis left05:49
qawsedrf git thinks it a submodule or something05:50
doesnt makes it a part of entire repo05:50
help me :|05:50
← bolt left05:50
mangdood qawsedrf: ...from what I understand, you shouldn't be removing your .git directory. What do you mean by "it's not on the server"05:50
qawsedrf on the live deployment05:50
and remote repo05:50
but i have the files on local repo05:50
mangdood qawsedrf: What do you mean by "it" when you said "it's not on the the server"?05:51
qawsedrf libs/vendor/fb_php_sdk/*05:51
→ pantsman joined05:51
← pantsman left05:51
→ pantsman joined05:51
mangdood qawsedrf: THat's not confusing at all. Why did you remove the .git directory?05:52
qawsedrf i removed libs/vendor/fb_php_sdk/.git05:53
cuz i thought maybe thats causing the issue05:53
mangdood qawsedrf: did you try adding before you added the .git directory?05:53
qawsedrf i did'nt remove the .git in the root of my project05:53
mangdood Oh, I see05:53
← fuzzybunny69y left05:53
qawsedrf i didnt add the .git, i cloned from github05:53
← ColKurtz left05:53
qawsedrf now on my local repo, wqhen i remove all contents of fb_php_sdk, and just add a hello.txt , it doenst shows in git status05:54
marked untracked05:54
:|05:54
← Textmode left05:54
mangdood qawsedrf: have you tried adding the individual files in sdk?05:55
qawsedrf: I'm thinking it might be a problem with *?05:55
because shouldn't your shell expand *?05:56
→ rtjure_ joined05:56
→ jceb joined05:56
← shruggar left05:57
→ singh_abhinav joined05:57
qawsedrf tried that yes05:58
no luck05:58
← j416 left05:59
→ arti_t joined05:59
mangdood qawsedrf: wait, what does git status say if you haven't added them?05:59
tomoj I have A<-B<-C and want to squash into C - that is, squash forward in time instead of backward - possible?05:59
mangdood qawsedrf: sigh, I'm sure I can't help much, but I'll do what I can06:00
← spaceonline left06:00
qawsedrf nothing to commit - on master rbanch06:00
← _huoxito left06:00
← rtjure left06:00
mangdood qawsedrf: wait, so why are you trying to add libs/blah/* if you haven't changed anything06:00
← edude03 left06:00
→ Longbeach joined06:02
qawsedrf leave it06:02
i renamed the folder and it worked fine06:02
← scientes left06:02
mangdood qawsedrf: alright, best of luck. Back to my own git crisis06:02
qawsedrf maybe git has somewhere in config that fb_php_sdk is still a submod or w/e and so won't let me use that :|06:03
hehe ok06:03
← qawsedrf left06:03
← weiyang left06:03
← Slim left06:04
→ thiago joined06:06
← tchan left06:06
← gchristensen left06:07
→ mocramis joined06:07
← thursday left06:09
mangdood How can I delete my last commit? (this has probably been asked for ages) I accidently reverted to a parent commit without any changes.06:10
← mishok13 left06:12
→ _huoxito joined06:13
→ Sitethief joined06:13
→ wah_ joined06:13
wereHamster mangdood: git reset HEAD^06:14
← Sitethief left06:14
→ Sitethief joined06:15
→ marthinal joined06:15
mangdood sigh, I hate asking so many questions: How can I get the diff between a file between the working directory and a commit (or any two places) and save it to a file so I can resolve the changes manually?06:17
→ benjo2 joined06:17
← wah_ left06:17
bob2 resolve?06:17
← benjo2 left06:18
mangdood bob2: Er...06:18
(whenever I'm on IRC, I'm usually having a problem that drives me nuts)06:19
→ senny joined06:19
→ JohnBat26 joined06:19
mangdood bob2: I want to make some of the changes indicated by the diff, but not the rest06:19
wereHamster mangdood: man git-diff06:20
gitinfo mangdood: the git-diff manpage is available at http://git-scm.com/docs/git-diff.html06:20
→ ThomasLocke joined06:20
bob2 maybe you just want git add -i06:20
→ papegaaij joined06:20
→ tsdh joined06:20
mangdood wereHamster: I have the manpage open, I just can't seem to find it06:20
Seveas mangdood, follow bob2's advice:if you want to add only certain parts of your changes to a commit, use git add -i06:21
→ DuzyZiemniak joined06:21
→ Sigma joined06:22
wereHamster mangdood: find what06:22
"This form is to view the changes you have in your working tree relative to the named <commit>. [...]". Is that not what you want?06:22
mangdood ...git add -i is going to save my brains to live another day. bob2: thanks06:23
← ISF left06:23
bob2 ps06:23
read about xyproblem06:23
mangdood wereHamster: sorry I wasn't specific. I was looking to have the diff outputted to a file06:24
← MissionCritical left06:24
wereHamster git diff ... > teh-file.diff06:24
mangdood Seveas: thanks as well :D06:24
← Per4u3e left06:24
→ beautifulmind joined06:24
← nevyn left06:25
_ikke_ shortcut, git add -p06:25
mangdood wereHamster: It looks like it's already being piped to less(more? never know the difference)though, which was why I was hesitant to do that06:25
→ Per4u3e joined06:25
wereHamster yes, git pipes to less if the output is a tty. If it's a file then obviously not06:25
→ harshpb joined06:26
← adamm left06:26
mangdood wereHamster: ...being the noob that I am, I didn't realize that programs were aware of where the output was being sent06:26
wereHamster mangdood: why didn't you just !try it ?06:26
gitinfo mangdood: [!tias] Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://sitaramc.github.com/1-basic-usage/tias.html may help with git-specific TIAS.06:26
→ shruggar joined06:27
→ EyesIsMine joined06:28
mangdood wereHamster: One reason was that I hoped to get the whole file with the indications of changes so I could just make the changes and add that whole file; that seemed like it would give the changes only06:29
← _huoxito left06:29
wereHamster eh?06:29
like a blame?06:29
mangdood wereHamster: ...why is my wording so bad06:29
Uh, I wanted the entire parts that were unchanged as well as the diff output, so that after I changed the parts indicated by diff, I could just commit the file06:31
→ drollwit joined06:32
mangdood add the file* FFFF06:32
_ikke_ mangdood: It's still unclear06:33
!xy06:33
gitinfo [!doinitrong] It sounds like you're approaching this problem in the wrong manner. Let's step back for a minute - What are you actually trying to achieve? Why are you trying to do it this way?06:33
→ Psycho_pr joined06:33
→ mneorr joined06:33
gitinfo set mode: +v06:33
→ mishok13 joined06:33
Psycho_pr Hi. How do I checkout from master all files that have somewhere in their path a certain word?06:34
_ikke_ Psycho_pr: That's not something git supports afaik06:34
← akiress left06:34
Psycho_pr Hrmmm06:35
Ok, thanks.06:35
mangdood _ikke_: I'm trying to take the difference between a file in HEAD (I hope to whatever that I'm using the right word) and the file in the parent commit, since there are changes that I didn't want to make. I want to reverse some of these changes.06:35
← bubbles10101 left06:35
→ dpino joined06:36
_ikke_ git checkout -p06:36
mangdood ...I'm going promptly back to that tutorial after I'm done explaining.06:36
_ikke_ Is that what you want?06:36
git checkout -p HEAD^06:36
→ tchan joined06:37
→ dc5ala joined06:37
← SegFaultAX|work2 left06:37
→ SegFaultAX|work2 joined06:38
→ weiyang joined06:38
← drev1 left06:38
mangdood _ikke_: I want to reverse only some of the changes from the difference between the files of these two commits, so I don't think that's the solution06:38
← drollwit left06:39
→ mneorr_ joined06:39
→ w09x_ joined06:39
→ giallu joined06:39
→ akiress joined06:39
← mneorr left06:40
← p3rror left06:40
_ikke_ mangdood: You can select what part you want to 'reverse'06:42
mangdood: By using e, you get the editor where you can select per line if you want it reversed or not06:42
← w09x left06:42
wereHamster Psycho_pr: probably git ls-files piped to git checkout06:43
_ikke_ wereHamster: with a grep in between I guess06:43
Or can git ls-files do the filtering?06:43
mangdood _ikke_: Hm. I'm actually doing it another way at the moment, as my experimenting has led me to do. I reverted to the file of the parent commit, and now I'm using git add -i file to make the changes06:44
→ w09x joined06:46
→ ph^ joined06:47
→ strk joined06:47
Psycho_pr Thanks!06:47
strk do you know of a git2cl playing nicely with git-svn metadata ? (using revisions of SVN)06:48
← WillMarshall left06:48
← yatiohi left06:48
← arti_t left06:48
→ arti_t joined06:49
Psycho_pr Does git checkout take stdin, or do I need to use xargs type of thing?06:49
thiago Psycho_pr: you need xargs06:49
← w09x_ left06:50
cbreak-work checkout takes a ref and/or file as command line arg06:50
← tchan left06:50
← airborn left06:51
→ dhruvasagar joined06:51
← strk left06:52
→ p3rror joined06:53
mangdood I used add -i to add hunks of text, and now my working directory is different from the index. Should I make the file in my working directory to be the same as the index, or is it preferred to just commit it?06:54
_ikke_ It depends on what you want06:54
← papegaaij left06:55
← hyperair left06:55
→ mischat joined06:55
mangdood I would like the file that is in the index and discarding the differences from the one in my working directory06:55
_ikke_ git checkout -- file06:55
← theCole left06:56
→ hyperair joined06:56
← shruggar left06:56
→ lopin joined06:56
← thiago left06:56
→ afuentes joined06:58
← beautifulmind left06:59
→ beautifulmind joined06:59
mangdood Thanks a lot to the whole channel XD. I almost died06:59
→ bosspetta joined07:00
← mangdood left07:02
← Per4u3e left07:02
← hyperair left07:09
→ boob00 joined07:10
← AlbireoX`Laptop left07:10
← ajpiano left07:10
→ shruggar joined07:11
← beautifulmind left07:11
← b00b00 left07:12
← singh_abhinav left07:13
→ ajpiano joined07:14
← shruggar left07:15
← Korakoff left07:16
→ Korakoff joined07:16
← kerrick left07:17
→ takus joined07:18
→ kerrick joined07:18
← Lgb left07:20
← mischat left07:20
→ rurufufuss joined07:20
← igotnolegs left07:21
← andrew_ left07:21
Psycho_pr I ran this but it doesn't seem to do anything.... git ls-files | grep PDFMaker | xargs git checkout master07:21
← zanneth left07:21
→ gavin_huang joined07:21
→ theCole joined07:21
→ pro_metedor joined07:21
_ikke_ Psycho_pr: try it with xargs echo07:21
→ Lgb joined07:22
Psycho_pr it shows everything but without new lines07:24
→ boombatower joined07:25
_ikke_ That shouldn't be necessary07:26
Psycho_pr: try to add --07:26
git ls-files | grep PDFMaker | xargs git checkout master --07:27
→ hyperair joined07:27
← gavin_huang left07:28
Psycho_pr still nothing07:29
_ikke_ if you do one manually, does it then work?07:29
→ xiangfu joined07:31
Psycho_pr It works but somehow still appears on git status as modified.07:31
frogonwheels Psycho_pr: what did you expect to happen?07:31
→ thiago joined07:31
frogonwheels oh. status as modified. you on master currently?07:32
← ipoval left07:32
→ gavin_huang joined07:33
← thiago left07:33
Psycho_pr Yes. I expected that it would be unmodified.07:33
→ thiago joined07:33
← davidbanham left07:36
→ ipoval joined07:37
→ jonnu joined07:37
→ MissionCritical joined07:38
→ stephanmg joined07:39
→ jargon- joined07:39
← ipoval left07:39
← zorzar_ left07:40
→ jackh joined07:40
→ OOPMan joined07:42
→ jn_ joined07:43
← cenuij left07:44
→ SilverWlf joined07:44
← takus left07:46
→ takus joined07:46
← eagles0513875 left07:47
→ eagles0513875 joined07:47
← eagles0513875 left07:47
→ eagles0513875 joined07:47
← Desproges left07:48
→ nevyn joined07:48
← kadoban left07:49
boob00 hello07:49
while running CCnet with git, getting timeout, any HOME var need to be set? (windows)07:50
cbreak-work what's ccnet?07:50
what does it have to do with git?07:50
→ whitman joined07:51
← mneorr_ left07:52
→ Gentlecat joined07:53
→ singh_abhinav joined07:53
boob00 CruiseControl.Net07:54
CI07:54
← babilen left07:56
← fp1 left07:56
→ francisfish joined07:57
← pheaver left07:58
→ AlbireoX`Laptop joined07:59
← keltvek left08:00
← SilverWlf left08:01
→ keltvek joined08:02
← weiyang left08:02
← acrocity left08:03
→ und3f joined08:03
← pantsman left08:05
→ zorzar joined08:07
Psycho_pr Why are there many changes in git status even though I've just run git reset --hard HEAD?08:07
← d3vic3 left08:08
→ LimeBlast joined08:08
→ BarryCrayon joined08:08
wereHamster Psycho_pr: windows?08:09
cbreak-work untracked?08:09
→ Sitethief{afk} joined08:09
bob2 bpaste.net 'git status' output08:10
→ yatiohi joined08:10
← Sitethief{afk} left08:11
→ d3vic3 joined08:11
→ Sitethief{afk} joined08:11
← Sitethief left08:12
→ shruggar joined08:12
→ marc_kdab joined08:13
→ bubbles10101 joined08:14
← cyphase left08:14
→ Sitethief joined08:14
→ Desproges joined08:15
→ dharrigan joined08:15
← dharrigan left08:15
→ dharrigan joined08:15
← peppyheppy left08:16
→ papegaaij joined08:17
← Sitethief{afk} left08:17
← jpic_ left08:18
← spb left08:18
← Layke left08:18
← Rafusek_ left08:19
← teukka left08:19
→ spb joined08:20
→ teukka joined08:20
← teukka left08:20
→ teukka joined08:20
→ Rafusek joined08:21
→ scratcha joined08:21
← scratcha left08:21
→ scratcha joined08:21
← hakunin left08:21
→ marcogmonteiro joined08:22
← dreiss left08:22
→ tchan joined08:22
← Sitethief left08:22
→ Sitethief joined08:23
← optixx left08:24
→ Sitethief{afk} joined08:24
→ optixx joined08:25
← Robbo_ left08:27
← Sitethief left08:27
→ Robbo_ joined08:28
→ fp1 joined08:28
→ babilen joined08:28
→ tvw joined08:29
← j0ran1 left08:30
→ j0ran joined08:30
→ cyphase joined08:30
← beneggett left08:30
→ ron joined08:31
← Eridius left08:31
← kenperkins left08:31
ron okay, say my developers and I have quite a few branches... some opened from our main branch, some opened from the sub branches. is there any way to visualize all the open branches and see which is branched off of which (and hopefully which is synchronized with which)?08:32
← williamcotton left08:32
et gitk --all?08:33
ron you're asking me? :)08:33
← xxtjaxx left08:33
et well, try and see if you get what you want08:33
← lonewulftp left08:33
ron ;)08:33
→ beneggett joined08:34
ron okay, looks great, thanks.08:35
→ mischat joined08:35
← ron left08:36
← pro_metedor left08:36
→ achadwick joined08:37
→ pro_metedor joined08:37
→ mneorr joined08:38
→ dark_element joined08:38
→ tatsuya_o joined08:40
← rtjure_ left08:41
→ Sitethief joined08:43
← ping-pong left08:44
← p3rror left08:44
← pro_metedor left08:44
→ pro_metedor joined08:45
← Sitethief{afk} left08:46
← ferdna left08:48
→ berserkr joined08:50
← d3vic3 left08:50
→ stodan joined08:50
← pro_metedor left08:51
→ d0k joined08:52
→ d3vic3 joined08:52
← dharrigan left08:52
← singh_abhinav left08:56
→ p3rror joined08:58
← boob00 left08:59
→ pro_metedor joined08:59
→ sh0ne joined09:01
→ b00b00 joined09:02
← Sitethief left09:04
→ Sitethief joined09:05
Psycho_pr http://bpaste.net/show/BYoZ7yt2LU1J1CA95Yys/09:09
This shows I first do a reset and then status and there are many files modified...09:10
← fp1 left09:10
→ TheHaven joined09:10
→ _iron joined09:10
← papegaaij left09:11
wereHamster Psycho_pr: what does git diff show?09:14
← stodan left09:14
→ psakrii joined09:14
→ tewecske joined09:16
→ stodan joined09:17
Psycho_pr All these files have new modes.09:18
→ diegoviola joined09:19
FauxFaux Psycho_pr: Running on a non-linux filesystem? man git config core.filemode09:19
gitinfo Psycho_pr: the git-config manpage is available at http://git-scm.com/docs/git-config.html09:19
→ caddoo joined09:19
Psycho_pr I'm on ubuntu.09:19
and I'm tracking file-modes.09:20
→ papegaaij joined09:20
← j0ran left09:20
→ j0ran joined09:20
→ _Qman joined09:21
→ acrocity joined09:21
→ diegovio1a joined09:22
→ simesy joined09:22
← diegoviola left09:24
← renzhi left09:24
→ xxtjaxx joined09:27
← Cros79 left09:32
→ twisted` joined09:32
← txomon|home left09:33
diegovio1a → diegoviola09:33
← vandemar left09:34
→ yizumi joined09:36
→ txomon|home joined09:36
← OOPMan left09:38
→ spaceonline joined09:38
→ cortana2 joined09:38
Psycho_pr How do I remove a remote branch that I'm never going to use again?09:39
FauxFaux man git push --delete09:39
gitinfo the git-push manpage is available at http://git-scm.com/docs/git-push.html09:39
← dhruvasagar left09:39
→ dhruvasagar joined09:40
Psycho_pr That takes references as a parameter, but I would like to give a name of a branch as a parameter.09:41
jast with --delete it will accept a branch name and magically turn it into a refspec09:41
Psycho_pr $ git push --delete remotes/OldSandboxServer/master09:42
fatal: --delete doesn't make sense without any refs09:42
wereHamster Psycho_pr: git push [options] <remote> <refspec>09:43
→ dr_lepper joined09:43
← dr_lepper left09:43
→ dr_lepper joined09:43
Psycho_pr I don't understand, I take the name I get from branch -avv09:46
→ ethanol joined09:46
Psycho_pr and that name, I want to delete.09:46
→ Sitethief{afk} joined09:46
← ethanol left09:47
→ ethanol joined09:47
← chayin left09:47
ethanol oops. anyway, I wanted to ask. if I have svn repo locally. how can I git svn clone that into a git repo? do I need to specify the directory? or the .svn directory inside it as source?09:47
wereHamster Psycho_pr: git branch -r -d OldSandboxServer/master09:47
→ vandemar joined09:47
wereHamster Psycho_pr: that will delete the branch in your local repository09:48
ethanol I get E: 'trunk' is not a complete URL and a separate URL is not specified09:48
:<09:48
→ eurek joined09:48
← Sitethief left09:49
← yizumi left09:49
→ yizumi joined09:50
← davv3 left09:50
← SunilThaha left09:50
← KyleXY left09:51
→ KyleXY joined09:51
← nutron left09:51
← spyro left09:51
→ cvanes joined09:51
→ b1rkh0ff joined09:51
→ davv3 joined09:51
← ethanol left09:51
← theCole left09:52
→ theCole_ joined09:52
Psycho_pr wereHamster, Thanks!\09:52
→ G0re2 joined09:52
cvanes hi, does anyone know if it's possible to use env vars in git config?09:52
← kryl99 left09:52
→ nutron joined09:53
gitinfo set mode: +v09:53
G0re2 hi, my coworker commited ~6000 unecessary files, i deleted them, but not a commit lasts like forever09:53
is there something i can do?09:53
not = now09:53
→ kryl99 joined09:54
selckin define deleted09:54
← yizumi left09:54
→ spyro joined09:54
← jackh left09:54
→ caseymcg joined09:54
← Desproges left09:54
selckin rewrote history to remove the commit?09:54
shruggar G0re2: git filter-branch can rewrite old history (and git rebase -i can rewrite simple history, more-easily). buyut see !rewrite09:54
gitinfo G0re2: [!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)09:54
→ SunilThaha joined09:55
G0re2 hmm09:56
i deleted them and commited that change09:56
→ linusoleander joined09:58
selckin then it has to track that you did that in the history like anything else09:58
if you want the repoto be smaller again and not have those files, you need to rewrite history as if it never happened09:59
G0re2 oh boy09:59
thx :)09:59
← tyatpi left10:00
→ TrahDivad joined10:01
→ harish joined10:01
selckin it's easy with rebase -i10:01
← AlbireoX`Laptop left10:01
selckin depending on how many other things changed since then10:01
← txomon|home left10:02
→ txomon|home_ joined10:03
→ dbdoclet joined10:03
← dbdoclet left10:04
→ pfFredd joined10:05
← linusoleander left10:06
← Sitethief{afk} left10:07
→ Desproges joined10:10
ohcibi hi, i have a repository which is just biiiiig (in file size) because at some point, someone committed a really large file... when i now want to shrink my repo, i'll delete that commit by rebasing and then force-push??10:11
← dhruvasagar left10:11
ohcibi or is there a better way?10:11
← chaowang left10:11
→ dhruvasagar joined10:12
→ Internet13 joined10:12
et ohcibi: filter-branch maybe? see the first example of http://www.kernel.org/pub/software/scm/git/docs/v1.7.3/git-filter-branch.html10:12
(with the below note)10:13
cehteh better is subjective .. but to shrink the repository really you need to be sure all traces of the big object are purged10:13
ohcibi cehteh: yes, thats what i want to do..... i just want to get rid of that object in objects database10:14
cehteh and any change in history has some headache capability as it invalidates any pre-change clones. people will curse or accidentally merge the crap in again10:14
→ beautifulmind joined10:15
cehteh ohcibi: locally you need to wipe the reflog, on the server you need to check that there are no stale branches or tags left behind, or just create a new repository with only the new branch initially10:15
think about a non destructive workflow :P10:16
ohcibi cehteh: i know that... i have a bunch of students repos and some of the students made a big mess in their repo.. i now want them to clean their repository. after that, they'll of course have to clone a fresh new copy to not merge any crap10:16
cehteh that will happen again :P ..10:17
ohcibi i know10:17
cehteh you could reject pushes from a hook checking for some policies10:17
ohcibi my second name is sisyphus10:17
cehteh: i could have done that in advance 8-)10:18
cehteh: but now i gotta clean the mess10:18
*they have to10:18
but i need to tell them how 8-)10:18
→ tyatpi joined10:18
← Robbo_ left10:20
← cortana2 left10:20
→ Element9 joined10:21
cehteh bondage and dicipline :) ... pre-receive hook checking for proper formatted commit messages, file and diff sizes and so on :)10:21
← herdingcat left10:21
→ Robbo_ joined10:21
cehteh once i thought about running astyle on the code to reformat it sanitize/relax it a bit and compare with how it looked initially and only accept pushes which fulfil coding styles :)10:22
ohcibi cehteh: i'll keep that in mind for next year10:22
cehteh }:>10:23
ohcibi hm the filter-branch method does not quite work.. i just have a new commit without the file, but the old commit with the file still exist, so the repo has the same size10:24
cehteh iirc filter-branch leaves a backup branch behind10:25
(i rarely used that and thats long ago)10:26
← bubbles10101 left10:26
cehteh also reflog, remotes, tags, whatsnot will have references to the old commit10:26
ohcibi ah i see, the commit with the file is now on no branch10:27
cehteh you have to purge that all carefully or what i would recommend, fix your repo in this nondestructuve way ... init a new repo and push the new sanitized branch there10:27
then check this new repo very carefully ..10:27
if that worked then it becomes the new working repository and you can push that to the server (also delete all old traces there)10:28
→ cortana2 joined10:28
cehteh and sooner or later delete the old repo (when you are sure everything worked in a few days/weeks)10:28
→ hotfloppy joined10:28
→ gerard0 joined10:28
ohcibi hm sounds reasonable10:29
how to clean the reflog?10:29
cehteh otherwise you can git-gc or git-repack your repo but by default these preserve some cruft for some configured time to be sure nothing gets lost10:29
man git-reflog10:29
gitinfo the git-reflog manpage is available at http://git-scm.com/docs/git-reflog.html10:29
← ipsion left10:30
ohcibi ah, so the person who committet the crap has to clean his reflog?10:30
cehteh not necessary10:31
reflogs are local10:31
ohcibi yes10:31
cehteh anyone has to force update his repo from the server10:31
reflogs tags and branches may have old big stuff around in unused leaves, but that will stay local10:32
ohcibi k i understand...10:32
cehteh unless they merge it in again and push it up :P10:32
ohcibi so i will do it with one of the students, he'll force-push, the others will clone10:32
hotfloppy Hi guys. I got a 1 bare repo on my server which my developer colleagues is using. Before this, everything is working but now we got a "No submodule mapping found in .gitmodules for path" error when trying to clone the repo. Actually, we did not create any submodules and there's no .gitmodules on top-level directories. How can I fix this? Is this a bug?10:32
→ arietis joined10:32
cehteh tags are good candidates for such fuckups :P10:32
ohcibi cehteh: for this purpose i always clone a new copy for myself, so after sanitizing, also the students who sanitized the remote has to clone a new repo10:33
← rurufufuss left10:33
→ Superlokkus joined10:34
← takus left10:34
← sangi left10:34
cehteh hotfloppy: what did you change? looks to me like something done wrong somehow10:34
← txomon|home_ left10:34
cehteh someone committed a submodule?10:35
← dhruvasagar left10:35
← tvw left10:35
→ tcurdt joined10:36
← tcurdt left10:36
ohcibi hotfloppy: some of my students copied a folder which contains a .git-directory to their repo and then commited and pushed them.. now git thinks there is a submodule,10:36
hotfloppy cehteh, that's also what I thinks. Is there's any ways to check if someone create a submodules? Log or something? And I didn't change anything. I just create and let them use.10:36
→ dhruvasagar joined10:36
ohcibi hotfloppy: i fixed it by cloning the repo.. deleting the folder with rmdir, then git rm <foldername> and then commit and push10:36
→ txomon|home joined10:36
hotfloppy ohcibi, lemme try your method first.. :)10:37
→ koo4 joined10:38
← babilen left10:39
→ w09x_ joined10:39
ohcibi hotfloppy: git rm -rf <foldername> wont work because git complains that <foldername> is a directory. however after deleting the folder from hard-disk, git notices that theres something missing anyway10:40
hotfloppy ohcibi, those folder name is the same name with my repo. For example, my repo is SherlockCase and those folder is on SherlockCase/Case1/Case2/SherlockCase. Is it okay to delete, commit and push? Is there any chances that this will delete the repo also?10:40
ohcibi hotfloppy: this reallly looks like my students work 8-)) i assume someone cloned the repo in a subdir of his working-copy10:41
hotfloppy f*ck..10:41
ohcibi hotfloppy: and yes, if the folder is empty after cloning, just purge that b**sterd10:41
← w09x left10:41
hotfloppy ohcibi, purge from local repo or need to push?10:41
I'm a bit paranoid.10:42
ohcibi, oh the folder is empty.10:42
→ S1oX joined10:42
ohcibi hotfloppy: it is because git wont checkout your submodules automatically... you have to do it manually after cloning10:43
hotfloppy: but since there is no submodule -> delete it10:43
→ veyres joined10:44
hotfloppy ohcibi: so, no need to push? But, the problem is, my friend use XCode and he cannot clone the repo at all because of this error.10:44
→ w09x joined10:44
ohcibi hotfloppy: of course you have to push... after all this is just a normal delete-file-from-repository operation... except that its not actually a file what youre deleting10:45
→ w09x__ joined10:45
ohcibi hotfloppy: and if you can clone it without any problems then xcode seems to do anything further (like automatically checking out submodules) which fails because of the wrong submodule-assumption10:46
so maybe he can clone it normally via console and then import it into xcode, but i dont know this ide10:46
hotfloppy ohcibi: Okay. So, the steps is like this: 1. rmdir SherlockHolmes 2. git commit 3. git push10:46
is it correct ?10:46
ohcibi almost10:46
at step 2 you wont have anything to commit 8-)10:47
hotfloppy eh ?10:47
no need to commit ?10:47
ohcibi so 1. rmdir SherlockHolmes 2. git rm SherlockHolmes 3. git commit 4. git push10:47
← w09x_ left10:47
hotfloppy I see.. Okay, let me try.10:47
→ khmarbaise joined10:48
← Jojo1234 left10:49
→ sangi joined10:49
← w09x left10:49
← neurodrone left10:51
← arti_t left10:52
← stephanmg left10:53
→ arti_t joined10:53
← simesy left10:56
→ tvw joined10:57
← `Assassin left10:57
teotwaki or just git rm -r SherlockHolmes10:58
→ barrbrain_ joined10:58
← jspiros left10:58
← xerxas left11:00
hotfloppy ohcibi & teotwaki: I've try deleting some folder from testing repo and it's seems that every folder on the level below SherlockHolmes got deleted also. For example, Sherlock/Case1/Case2/SherlockHolmesToDelete, Case1 & Case2 & SherlockHolmesToDelete will be remove. Is it really like this?11:01
ohcibi teotwaki: it wont work because SherlockHolmes is a directory11:01
→ |bnz joined11:02
ohcibi hotfloppy: if case1/case2 is empty otherwise yes11:02
gitinfo set mode: +v11:02
ohcibi hotfloppy: git doesnt tracks directories, it tracks files... empty directories wont tracked unless you do anything fancy like your colleague did11:02
*doesnt track11:02
← beneggett left11:03
teotwaki $ git rm -r foo/11:03
rm 'foo/bar'11:03
# deleted: foo/bar11:03
it works.11:03
|bnz Hi was wondering if someone could help me understand the .git directory. The client I am using is asking me if I want to stage changes to the .git directory itself. Should I?11:03
ohcibi teotwaki: SherlockHolmes is empty11:03
→ babilen joined11:03
_ikke_ |bnz: You can't and you shouldn't11:03
teotwaki ohcibi: meh, weird scenario.11:04
|bnz so should i have .git in .gitignore?11:04
_ikke_ |bnz: No11:04
hotfloppy ohcibi, I see. So, its safe to push eh? My fingers is already excited to hit the enter key.11:04
_ikke_ |bnz: git doesn't look at the .git directory11:04
|bnz: You can'11:04
ohcibi hotfloppy: push it to somewere else, clone it, and check if everything is right if you want to get sure11:04
_ikke_ |bnz: You can't add .git11:04
|bnz so if i do something like local branch11:05
then commit - it talks about files in .git11:05
hotfloppy teotwaki, Yes, this is a weird scenario. I'm lucky that ohcibi is here and was experiencing this before :)11:05
_ikke_ |bnz: Yes11:05
|bnz should i be just making sure i do not stage those files?11:05
_ikke_ |bnz: .git contains all the history11:05
|bnz or is there a better way?11:05
_ikke_ |bnz: You don't have to care about that11:05
|bnz will it be actually breaking anything then?11:05
_ikke_ |bnz: try git add .git; git status11:05
ohcibi teotwaki: init a repository, push it somewere, clone it into your working-copy, add and push this, then clone it again at some other place, and then try to git rm -r delete the clone in the clone 8-))11:06
_ikke_ |bnz: (nothing is added)11:06
|bnz strange11:06
very confused why this is happening11:06
_ikke_ |bnz: Of course not, why should git add it'11:06
|bnz if e.g. i add a new git remote repository11:06
_ikke_ |bnz: Of course not, why should git add it's own repository to itself11:06
|bnz then commit, i get asked if i want to stage changes to certain files within .git11:06
→ angelsl joined11:07
_ikke_ |bnz: Have you added a git repository within another git repository?11:07
ohcibi lol11:07
if he did he gotta pay the beer11:07
he is the third now 8-)11:07
|bnz ikke - thanks for all this, i'm not sure!11:07
bit new to all this11:08
_ikke_ |bnz: do you use the command line?11:08
|bnz no using aptana studio11:08
← Dragooon left11:09
_ikke_ |bnz: It might have messed something up. Not sure11:09
← gavin_huang left11:09
_ikke_ |bnz: I have no experience with it, so it's hard to help you with it11:09
|bnz so my .git directory should live within my project?11:09
_ikke_ |bnz: It should be at the root of your project11:09
|bnz yep11:09
good11:09
← khmarbaise left11:10
|bnz so that .git directory contains info about say, remote repositories right?11:10
← beautifulmind left11:12
→ gavin_huang joined11:12
→ Lummoxx joined11:14
→ delx joined11:14
← gavin_huang left11:14
delx Is it possible to give a git-user permissions to only read one certain subdirectory of the git-repository?11:15
← zdk left11:15
→ jspiros joined11:15
→ alberto56 joined11:15
vandemar |bnz: git should not be mentioning anything inside .git/ when you do git status or git add, if it does something is broken11:15
→ hakunin joined11:15
_ikke_ |bnz: delx no11:16
← koo4 left11:16
|bnz vandemar - it only happens occasionally11:16
i think its when i do things with remote repositories11:16
← tsdh left11:16
delx _ikke_, any alternatve?11:16
_ikke_ delx: What do you want to achieve?11:17
delx: And why do you need that11:17
eregon_ → eregon11:17
delx _ikke_, my designer wants to have access to the templates of my webproject, but I don't want to give him the full repository access - then he would be able to see the code, too.11:17
vandemar |bnz: it still should not be doing that. you must be doing something crazy (and probably dangerous) to get that to happen. It might mention heads or refs informationally, but if it talks about anything inside .git/ as if they're files to be tracked, something is very wrong11:17
→ marhaban joined11:18
vandemar |bnz: can you pastebin actual messages where you see this?11:18
|bnz im trying to reproduce it at the moment11:18
SethRobertson delx: You probably want to use git-subtree to extract the subtree into a new repo which you can provide. git-filter-branch can do the same thing.11:19
← alberto56 left11:19
→ Peste_Bubonica joined11:19
|bnz i could but there'd be a lot of obfuscation needed11:19
example though is this:11:19
in aptana, add new remote repository11:19
then commit11:19
aptana/git asks if i want to stage .git/config11:19
SethRobertson delx: You could use !subprojects if you wanted to permanently split the templates from the code11:19
gitinfo delx: 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 seperate. Otherwise your best options are "!submodule" "!gitslave" and "!subtree" Try typing those commands into this IRC channel.11:19
delx SethRobertson, thanks for your answer. Can I merge the subtree back into my original repostory later on?11:19
!subtree11:20
gitinfo The git subtree merge method is ideal to incorporate a subsidiary git repositories directly in to single git repository with "unified" git history, where you only need to pull changes in from external sources not contribute your own changes back (which if technically possible is at least difficult). See http://progit.org/book/ch6-7.html Type "!subtree_alt" for more options11:20
← jspiros left11:20
|bnz the change to .git/config talks about the new [remote "gitserver"] and it's url etc11:21
← S1oX left11:21
→ jspiros joined11:22
_ikke_ |bnz: Could you try to do git status on that repository on the command line?11:22
delx SethRobertson, thanks so far.11:22
← delx left11:22
→ Yuuhi joined11:23
|bnz yes, i get:11:24
$ git status11:24
# On branch master11:24
# Changes to be committed:11:24
# (use "git reset HEAD <file>..." to unstage)11:24
#11:24
# modified: application/controllers/StuffController.php11:24
#11:24
_ikke_ |bnz: Please use a pastebin or something next time11:25
|bnz (from within aptana's terminal window)11:25
_ikke_ |bnz: That seems all normal11:25
|bnz yes11:25
→ DoniRS joined11:25
_ikke_ |bnz: So it's something with aptana11:25
|bnz is it something to do with the fact git should never be doing anything with . folders11:25
i.e. hidden11:25
← sh0ne left11:25
|bnz but from within aptana it is11:25
?11:25
_ikke_ |bnz: No11:25
|bnz should one be able to push hidden folders?11:26
_ikke_ |bnz: with push, you mean add? Yes, you can add hidden folders11:26
and files11:26
For git they are just normal files11:26
|bnz but if .git folder contains info about say repositories11:26
(which it clearly does)11:26
_ikke_ |bnz: info about repositories is stored in .git/config11:26
|bnz then does it matter if those changes are under version control?11:27
_ikke_ |bnz: It doesn't make sense to put them under version control11:27
← marc_kdab left11:27
|bnz would one usually not git add the .git directory?11:27
← preflex left11:27
|bnz is that what's gone wrong perhaps?11:27
(perhaps aptana's gone and done it?)11:28
_ikke_ |bnz: One would not, and with git itself, it's not even possible11:28
jast git doesn't add .git directories11:28
→ preflex_ joined11:28
_ikke_ try git ls-files .git11:28
or git ls-tree .git11:28
|bnz strange i get nothing11:28
→ koo4 joined11:28
vandemar |bnz: that's not strange, that's normal11:28
jast yeah, that's what is supposed to happen11:28
preflex_ → preflex11:29
vandemar |bnz: whatever aptana is doing sounds dangerous and bizarre11:29
|bnz git ls-tree .git11:29
returns fatal: Not a valid object name .git11:29
vandemar - to be clear - it may just be the way i am using it11:29
jast yeah, that doesn't work11:29
|bnz (i may have done something wrong)11:29
jast git ls-tree HEAD:.git11:29
|bnz again not a valid object11:30
← jspiros left11:30
|bnz so now when i try to commit11:30
i get the chances to stage .git/index and .git/objects/9aabcbdde...etc11:31
vandemar maybe aptana is trying to put .git under version control in some other repo, but trying to put dir/.git under version control in the dir/.git repo will not work11:31
|bnz could it be the way i set up my remote repository/git daemon?11:31
git init -bare foo ... was what i used to set up the remote repository11:32
→ klj613 joined11:32
_ikke_ nope11:32
vandemar it's bare locally or bare on some remote?11:32
← Element9 left11:32
|bnz on the remote11:32
before i pushed to it11:32
incidentally i am only playing with this at the moment and have backups etc, this is a learning experience at this point but i want to know what i am doing before i start using this properly11:33
vandemar can you dig the commands you used to set up your local and remote repos, out of your shell history, and pastebin it?11:34
_ikke_ vandemar: He uses aptana11:34
|bnz local i used aptana's "initialise", and remote i used the command above11:35
i.e. git init --bare foo11:35
← tshauck left11:36
jast the remote command is fine11:37
you're not having any problems on the remote, anyway, are you :)11:37
|bnz not afaict!11:37
so to be clear is git hard coded to ignore .git directories?11:37
_ikke_ |bnz: yes11:37
|bnz and seemingly aptana is not11:37
is it possible that its just that aptana is listing them and they are not actually being committed11:38
→ fleetfox joined11:38
_ikke_ |bnz: that might be possible, depending on what aptana is actually doing11:38
fleetfox how do you force clone in non empty directory?11:38
_ikke_ fleetfox: That question doesn't make sense11:38
fleetfox: Oh, now it does :P11:39
fleetfox fatal: destination path '.' already exists and is not an empty directory.11:39
i have some files there11:39
← JEEB left11:39
_ikke_ fleetfox: It's not possible afiak11:39
afaik*11:39
← w09x__ left11:39
fleetfox really?11:40
→ w09x joined11:40
barrbrain_ sometimes I wonder if we need a global lock for responding to n00b questions on the mailing list11:40
ohcibi okay, now i have succesfully deleted some files from all commits and force-pushed that to the remote, so that the .git-directory of a clone consumes less space. objects seem to be gone. but on the remote they still exist. how can i delete the objects on the remote? remote is github11:40
_ikke_ ohcibi: You have to wait for them to be garbage collected11:41
ohcibi so github will take care of it in this case?11:41
i triggered garbage collection on my server but the directory still consumes the same space11:42
_ikke_ ohcibi: Because of the reflog11:43
→ JEEB joined11:43
← psakrii left11:44
← kcj left11:44
ohcibi _ikke_: the reflog on the remote??11:45
← harish left11:45
jast fleetfox: git init; git remote add origin <URL>; git fetch; git checkout master11:46
inside that directory11:46
→ pyhead joined11:48
→ jspiros joined11:48
← eurek left11:48
← Mowsh left11:51
← dpino left11:53
ohcibi is it possible to trigger garbage collection on remote from local?11:53
_ikke_ ohcibi: nope11:53
ohcibi k... if i do gc --prune=NOW, the objects will be deleted, so i guess they'll disappaer at sometime on github11:54
← JEEB left11:54
→ sh0ne joined11:55
→ simesy joined11:55
jameslord hi all11:55
_ikke_ ohcibi: by doing gc --prune=NOW will not make it expire any quicker on a remote11:55
jameslord anybody here who know how to use repo script?11:55
why google develop a tool such as repo?11:56
_ikke_ ohcibi: You can only make sure that there is nothing referencing the object and just wait11:56
ohcibi _ikke_: ?? it actually does... http://stackoverflow.com/questions/1904860/how-to-remove-unreferenced-blobs-from-my-git-repo11:56
← madsy left11:56
→ madsy_ joined11:56
→ tshauck joined11:56
ohcibi _ikke_: see the quote in the first answer11:56
_ikke_ ohcibi: I'm talking about remote repositories, not local repositories11:57
ohcibi _ikke_: i just did that on the remote repository11:57
_ikke_: i copied the stuff to my server to try it out11:57
_ikke_: the local is easy done by re-cloning (which is okay in my case)11:58
_ikke_ ohcibi: What I meant is that doing gc on a local repository does not make it expire quicker on a remote repository11:59
→ replore joined11:59
← Robbo_ left11:59
ohcibi _ikke_: i know 8-))11:59
i was about to write the same 8-)12:00
→ j0k3r joined12:00
→ ph^_ joined12:00
→ erichynds joined12:00
→ cccaldas joined12:02
← ph^ left12:03
→ Robbo_ joined12:03
← twisted` left12:03
← Lisimba left12:05
→ Lisimba2 joined12:06
→ JEEB joined12:08
fleetfox thanks jast :\12:10
← wilmoore left12:11
→ the_cybersphinx joined12:11
← Robbo_ left12:11
← hotfloppy left12:11
→ xerxas joined12:11
→ bubbles10101 joined12:14
← cybersphinx left12:14
→ Robbo_ joined12:14
→ SidGBF joined12:15
SidGBF Hi all :)12:15
Guys, Im taking some time to finally learn how to use git.12:16
_ikke_ SidGBF: !hi12:16
gitinfo SidGBF: [!welcome] Welcome to #git, a place full of helpful gits. If you have a question, feel free to just go ahead and ask—somebody should answer shortly. For more info on this channel, see http://jk.gs/git/ Take backups (type !backup to learn how) before taking advice.12:16
← harshpb left12:18
SidGBF I was checking GitImmersion, but at this moment I have an specific doubt about how the remote hosting works. I mean, Im confused about that bare stuff... I mean, I can start working on my project, init a repository here in my machine and stuff. But, I want everything backed up in my server. In SVN I would create there a repository and check out it in my machine. Im confused about how this works on git...12:18
jast SidGBF: create a bare repository on the server and push to it12:19
bare means all the history is stored, but git doesn't bother checking the files out on the server12:19
which is similar to how SVN server-side repositories work, only faster and saner ;)12:19
→ ryez joined12:20
→ glennpratt joined12:20
SidGBF jast: ok, so, a BARE repository is like a 'control center', where everyone can clone from (but seem that there's also the .git file that looks like allowing you to check out from another user)12:20
(checkout -> clone... Im trying hard to fix the language)12:21
→ alem0lars joined12:21
jast yeah, you can clone from non-bare repositories, too12:21
but there are less complications with pushing to _bare_ repositories12:22
here's an article about it: !bare12:22
gitinfo an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/12:22
_ikke_ The problem with pushing to non-bare repositories is updating the working directory12:22
→ bnz joined12:22
gitinfo set mode: +v12:22
jast btw, one of my secret plans is to supplement gitrecipes with a series of fairly silly yet instructional videos :}12:23
→ Bass10 joined12:23
→ intransit joined12:23
SidGBF the hard part is the language transition...12:23
jast it's actually not hard at all, it just takes a bit of time. don't sweat it.12:23
SidGBF 'commit' now is to 'backup work locally' and to really "commit" is the push.12:24
and I must say: it is a great idea12:24
to destroy a repo, is it just rm the folder?12:24
bnz got dropped out, who was kindly helping me earlier?12:24
← sh0ne left12:25
bnz may be interested to know that i fired up a copy of git-scm (it's gui) and its doing the expected now12:25
(i.e. not listing the stuff in .git to be committed)12:25
so i have a feeling this could be something funny aptana is doing12:25
_ikke_ bnz: Yes, obviously.12:26
bnz or the git gui it provides12:26
ah ikke - thanks v. much for help earlier12:26
do you think what has happened is actually dangerous though?12:26
im not so sure12:26
→ jackh joined12:26
Psycho_pr $ git clone --bare /var/www/vtigercrm/.git/ ~jacob/source/vTiger_RedCad.git12:27
Initialized empty Git repository in /home/jacob/source/vTiger_RedCad.git/12:27
Why do I get an empty Git repo instead of cloning the xisting one?12:27
← sangi left12:27
→ shadowe989 joined12:27
_ikke_ bnz: Well, git is not built for tracking itself12:27
bnz yeah12:27
i guess if you did a pull12:27
it could break things actually cos i may merge changes from an old .git!12:28
← und3f left12:28
← pfFredd left12:29
← Kulrak left12:29
← jspiros left12:29
kkiero how to deal with fragile data files in git? for example I want to version file that contains password but I don't want to show this passwords in repository12:32
← arti_t left12:32
← macabre left12:32
cbreak-work kkiero: don't track it12:32
track some template file instead12:32
_ikke_ !config12:33
gitinfo [!configfiles] the recommended way is to change your application so that you can (perhaps optionally) store site-specific configuration/data in separate files, then keep those files out of version control. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.12:33
bnz ikke - would you agree as long as i never stage/commit the .git stuff in aptana i can carry on as normal?12:33
← alem0lars left12:34
→ stephanmg joined12:34
_ikke_ bnz: I would first try to find out what went wrong12:34
→ alem0lars joined12:34
bnz its a bug that aptana is offering them to be staged perhaps?12:34
ok12:34
its possible they are not being tracked either12:34
i.e. selecting them for staging/committing results in nothing happening12:34
→ jspiros joined12:35
kkiero alright so how to track one file template with clear data only once, then replace this data with passwords and stop tracking it but have first version in repo all the time?12:35
← ronny left12:35
← Desproges left12:36
→ gchristensen joined12:36
← tyatpi left12:36
bremner kkiero: that is what the link about is about.12:36
← ivan12350 left12:38
→ Kulrak joined12:38
← Kulrak left12:38
→ Kulrak joined12:38
SidGBF jast: Dude, so, the first thing to do is a) create a local repo, with file, and then the bare in my server b) first the bare ?12:39
→ Octalot joined12:39
← ivan12349 left12:39
← ivan12348 left12:39
→ g0bl1n joined12:39
_ikke_ kkiero: Best solution is to never store the config file itself into git12:40
→ Codif joined12:42
vandemar or include filedata.example but have the real filedata file in .gitignore12:43
→ bullicon joined12:43
← Codif left12:43
→ j0k3r_ joined12:44
→ Caelum joined12:44
→ brentratliff joined12:45
→ Hounddog joined12:46
Hounddog what was the parameter to ignore hooks?12:46
abackstrom_ → abackstrom12:47
_ikke_ Hounddog: which hooks?12:47
→ neurodrone joined12:47
SidGBF Can anyone here help me with the bare repository? how do I create it?12:47
cbreak-work Hounddog: man git-commit12:47
gitinfo Hounddog: the git-commit manpage is available at http://git-scm.com/docs/git-commit.html12:47
cbreak-work SidGBF: man git-clone/git-init12:47
gitinfo SidGBF: the git-clone manpage is available at http://git-scm.com/docs/git-clone.html12:47
Hounddog _ikke_: pre-commit hook... i have a phpmd hook with me but a co worker has not... now after pulling changes i cannot commit cause his code is shit12:48
_ikke_ --no-verify12:48
→ Milossh_ joined12:48
Hounddog thx :)12:48
i was looking for ignore hook :)12:48
← Milossh left12:49
← ph^_ left12:49
← tabakhase left12:49
→ tabakhase joined12:49
→ bartek joined12:49
ironfroggy is it possible to find all of my commits across all of my local branches?12:49
→ dzonder joined12:49
ironfroggy essentially i'm trying to answer the question "what did you do this month?"12:50
→ tyatpi joined12:50
cmn git rev-list --all12:50
FauxFaux git log --after=.. --author= .. --all12:50
← kaen left12:50
FauxFaux -p12:50
→ h0bbit joined12:51
SidGBF mmmm12:51
got the idea12:51
→ ivan12345 joined12:51
ironfroggy FauxFaux: thank you12:51
← jspiros left12:52
→ ivan12346 joined12:52
→ ivan12347 joined12:52
→ pandeiro joined12:52
→ jspiros joined12:52
← j0k3r_ left12:54
← j0k3r left12:54
→ bitkiller joined12:55
SidGBF that's weird :) the bare is created AFTER12:55
← ben_alman left12:55
→ hellopat joined12:56
← axl_ left12:56
← ranamalo left12:57
→ revagomes joined12:58
_ikke_ SidGBF: For git, it doesn't matter in what sequence12:58
But you can't clone from empty bare repositories12:59
SidGBF _ikke_: yea, so, you must create first a local repo, then clone a bare, and then the things go12:59
→ alex88 joined12:59
SidGBF this is what I wasn't getting13:00
cbreak-work _ikke_: you can clone empty bare repositories13:00
since git 1.6 or so?13:00
maybe longer13:00
et you just get a warning13:00
→ Jackneill joined13:01
SidGBF but I can't 'git init --bare .' for instance13:01
can I?13:01
cmn if it's empty, you should be able to13:01
not that it makes any sense, try it13:01
→ taylanub joined13:02
← hide left13:02
→ ph^ joined13:02
alex88 hello guys, look at this pic, http://i.imgur.com/VW29Y.png i had development and staging on the purple dot, production on the actual stage, i merged development on production and it merged correctly, done the same with staging and it created a new commit, why it didn't merged on same?13:02
→ brotspinne joined13:02
jast SidGBF: there are actually many ways to do it. you can create the bare repository before or after creating the local one. cloning from it (so creating it first) is slightly easier but by no means the only way.13:02
_ikke_ cmn: I recall it giving an error about there being no branch or something13:02
→ macabre joined13:02
_ikke_ cbreak-work*13:03
cbreak-work yeah13:03
no big deal13:03
_ikke_ ah ok13:03
that used to be13:03
taylanub how do i absolutely go back to a previous commit ? is reset --hard enough ?13:03
cbreak-work of course it doesn't have a branch :)13:03
_ikke_ taylanub: yes13:03
cbreak-work taylanub: you have to specify where you want to go to13:03
jast alex88: you did something like "git merge branch1 branch2", right?13:03
taylanub cbreak-work: that i know. _ikke_: ok, thanks both13:03
→ logik-bomb joined13:03
brotspinne hello! is there a way to do a git branch --contains which only compares the "content" of the commits without comparing sha sums? because I have a branch with a bunch of commits which seems to be also in the master branch but I'm not sure13:03
→ Ryback_ joined13:03
← Ryback_ left13:04
→ Ryback_ joined13:04
alex88 jast: i've switched to staging, clickend on merge (it's sourcetree app) and selected the production stage commit13:04
cmn brotspinne: take a look at man git cherry13:04
gitinfo brotspinne: the git-cherry manpage is available at http://git-scm.com/docs/git-cherry.html13:04
SidGBF jast: Ok, let's say I have no repos. How to create the first one already being the bare one?13:04
jast alex88: the commit message suggests that you merged two branches simultaneously13:04
brotspinne thanks13:04
logik-bomb If I have uncommited/untracked files on my branch and I switch to another branch, what happens? Do those changes get carried around between branches?13:04
→ bglusman joined13:04
_ikke_ logik-bomb: yes13:04
logik-bomb: unless the other branch would have to overwrite them13:05
→ AaronMT joined13:05
_ikke_ logik-bomb: then git aborts13:05
jast alex88: that never works as a fast-forward, because git tries to represent that you merged them simultaneously, and for that it needs to create a merge commit13:05
logik-bomb can I just stash my stuff between swapping branches?13:05
*before13:05
FauxFaux logik-bomb: !float13:05
gitinfo logik-bomb: If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there. If the files you changed differ between branches, the checkout will fail. In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.13:05
taylanub now i have a modified version of a specific commit. how do i merge with the latest upstream commit ?13:05
_ikke_ logik-bomb: yup13:05
alex88 jast: ok thank you… so merging to a commit with more then 1 branches on them creates this, thank you13:05
FauxFaux taylanub: git merge @{upstream}13:06
logik-bomb great13:06
Thanks!13:06
taylanub FauxFaux: after commiting my changes, right ?13:06
jast alex88: merging one branch will create a mereg commit, too, if there are new commits in both branches. I'm still not entirely sure whether that was the case in your scenario... I can't think straight enough right now. ;)13:06
FauxFaux taylanub: Everything fun deals only with committed changes, yes.13:06
→ ben_alman joined13:06
alex88 jast: not a problem, it was just for having a cleaner view on the log13:06
jast SidGBF: on server, create directory, go in there, git init --bare.13:06
taylanub FauxFaux: ok :)13:06
brotspinne cmn: sorry I can't find any option with cherry-pick which only tests for presence of a given commit or branch in another branch according to its content13:06
← maleknet left13:07
cmn brotspinne: that's not the command I told you about13:07
SidGBF jast: that's what I was looking for :)13:07
cbreak-work brotspinne: man git-cherry13:07
gitinfo brotspinne: the git-cherry manpage is available at http://git-scm.com/docs/git-cherry.html13:07
SidGBF now the stuff started making sense13:07
cnm: Yo dude ;)13:07
jast SidGBF: sounds good!13:07
brotspinne ah just "cherry". I will have a look13:08
SidGBF this BARE was making no sense :p13:08
jast it's all explained at !bare13:08
gitinfo an explanation of bare and non-bare repositories (and why pushing to a non-bare one causes problems) can be found here: http://bare-vs-nonbare.gitrecipes.de/13:08
SidGBF jast: I mean cause most of docs will tell you to clone into a bare, and not init the bare13:09
jast though it doesn't include commands for creating a repository of any kind13:09
cloning *into* a bare repository?13:09
that's a pretty rare use case13:09
SidGBF yeah, git clone --bare ....13:09
drue not really. everytime you fork on github it clones into a bare repo.13:10
→ marc_kdab joined13:10
drue everytime you mirror a repo.. etc13:10
jast yeah, *it* does. you don't. :}13:10
you don't create all that many mirrors, typically13:10
also in that case I'd recommend git clone --mirror13:10
and chances are github does something else anyway. I know repo.or.cz uses a different approach to forking, internally...13:11
drue the funny thing about git is it allows for a wide variety of use cases. it's hard to assume what people do and don't do with hit.13:11
jast indeed, but we were pretty much speaking about the first steps here13:11
cmn but if someone tells you to clone into a bare repo, soemthing's fishy13:12
jast mirroring repositories is not a 'first steps' topic13:12
← acrocity left13:12
← caseymcg left13:13
← fusion27 left13:13
→ juanfer joined13:14
→ j4mie joined13:16
brotspinne cmn: nice one. cherry did the job. thanks13:16
Psycho_pr Why is it that git status shows me something is staged for commit but git diff shows me the same file has no changes in it?13:17
← j4mie left13:17
cmn Psycho_pr: because those are two different things13:17
drue Psycho_pr, because git diff shows files that aren't staged. try git diff --staged13:17
_ikke_ It shows the diff between the working dir and the index13:18
if you want both, you'd do git diff HEAD13:18
Psycho_pr I see.13:19
← tatsuya_o left13:19
→ _huoxito joined13:20
← Lgb left13:20
→ tatsuya_o joined13:20
→ ehsan joined13:20
Psycho_pr Thank you.13:21
→ chrisdev joined13:21
→ Lgb joined13:22
← Robbo_ left13:22
← tomoj left13:23
→ kent\n joined13:23
← ph^ left13:23
→ w09x_ joined13:24
← PapaSierra left13:24
→ Robbo_ joined13:26
← w09x left13:26
→ ipalaus_ joined13:27
→ jgoss joined13:27
→ freeed joined13:27
→ ph^_ joined13:28
← kpreid left13:28
← ipalaus_ left13:28
→ ipalaus_ joined13:28
← tatsuya_o left13:28
ipalaus_ → ipalaus13:28
← v0n left13:29
→ kpreid joined13:29
→ gavin_huang joined13:29
taylanub what's the quickest way to get events for when internet connectivity is lost/gained ?13:30
_ikke_ taylanub: has this to do with git?13:30
taylanub whoops, wrong channel :P13:30
→ axl_ joined13:31
← marhaban left13:33
← pyhead left13:34
→ t4nkd joined13:34
← _huoxito left13:35
← dc5ala left13:35
→ _huoxito joined13:35
→ pyhead joined13:36
→ rjgonza joined13:36
→ NiteRain joined13:37
→ cannonball joined13:38
→ dv310p3r joined13:38
← brotspinne left13:38
→ ronniy joined13:40
imMute taylanub: the answer is ifplugd13:41
but yeah, wrong channel13:41
→ TeckniX joined13:42
← bitkiller left13:42
← gitinfo left13:43
jast oh, I think I broke something13:43
→ hide joined13:43
→ gitinfo joined13:43
ChanServ set mode: +o13:43
← w09x_ left13:44
→ bitkiller joined13:46
← SidGBF left13:47
← gavin_huang left13:49
→ eacameron joined13:50
← eacameron left13:50
← Samual left13:51
← NiteRain left13:51
→ kukks joined13:51
← t4nkd left13:51
taylanub imMute: the question was targeted to #iphonedev but thanks :P13:52
→ ToBeFree joined13:53
jast an easy mistake to make... 'iphonedev' and 'git' type pretty much the same :P13:54
→ eijk joined13:56
Psycho_pr How do I make sure all branches (including remote ones) are merged FROM my current branch? I'll explain: say I was in some branch randomly, and all my branches were synchronized to the same commit. Suddenly I found a bug and fixed it, committed it, but now I have to do tens of checkouts and merges to propagate the change..13:57
← Psycho_pr left14:00
→ dcarper joined14:01
← jceb left14:02
→ stamina joined14:03
← xiangfu left14:04
→ chrisgeorge joined14:04
cmn you do the same thing you do for any other bug/feature, you identify the earliest branch point you want it to affect, create a branch off of that, fix the bug and merge wherever you need to14:04
→ Samual joined14:05
cehteh but he left :P14:05
cmn meh14:05
then he doesn't get the answer14:05
→ SidGBF joined14:05
→ mdhopkins joined14:05
SidGBF guys14:06
http://pastebin.com/rYxJnPHC14:06
IIm failing to clone from github14:06
jast your local configuration is screwed up14:07
SidGBF May it be some git setting?14:07
(wrote it before reading)14:07
→ psoo joined14:07
jast do you happen to have a http_proxy environment variable? ("set" on the command line to see all of 'em)14:07
← f0i left14:08
SidGBF jast: doesn't seem so... I'll check out repo configs14:08
jast shouldn't matter14:09
cmn env is probably more readable14:09
→ gusnan joined14:09
SidGBF global config has ' proxy = ' inside '[http]'14:09
← simesy left14:09
← _Qman left14:09
SidGBF may be this?14:09
← barrbrain_ left14:09
← blacktulip left14:10
cmn that looks extremely likely14:10
→ davidjdagino joined14:10
_ikke_ the git rebase -i --autosquash, how does it know where it should fix it up?14:10
SidGBF yeah14:10
that was it :)14:10
_ikke_ (where it should squash with)14:10
→ t4nkd joined14:10
jast _ikke_: it matches the commit summaries, I think14:11
_ikke_ jast: Ok14:11
← ph^_ left14:11
jast e.g. if you have a commit that has "add quantum entanglement" and a commit that has "fixup! add quantum entanglement"14:12
SidGBF btw, if I use git clone inside a repo, does it will work as an external (submodule) or will screw my repo?14:12
jast SidGBF: it won't automatically turn into a submodule... it will just happen to live in that particular location on your computer. but no, nothing will implode.14:12
if your version of git isn't ancient, that is14:13
SidGBF I just update it14:13
jast good enough!14:13
cmn the commit message14:13
SidGBF I was getting a lot of problems cause it was a bit old14:13
CalimeroTeknik how does git manage symlinks?14:13
→ oscar_ joined14:13
← ThomasLocke left14:13
jast CalimeroTeknik: it records them as symlinks14:14
→ scientes joined14:15
SidGBF jast: and how should I use the command in order to be a submodule? I mean, in my idea, when I order to sync, all the submodules will be synced14:15
← __root__0 left14:15
CalimeroTeknik it doesn't look like it :|14:15
I have new: new: etc14:15
not on the symlinks but as if I had done cp -r14:15
this kind of freaks me out14:15
cmn are you sure it's a symlink and not a hardlink?14:15
jast SidGBF: submodules are a topic unto its own. I keep meaning to write a decent article about 'em. in the meantime, let's try !submodule14:15
gitinfo SidGBF: git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://book.git-scm.com/5_submodules.html14:15
CalimeroTeknik it's symlinks, yes14:16
→ gtrak joined14:16
← frogonwheels left14:16
cmn and where do you have "new: new:"? what happens when you commit? is it still recorded as a new file?14:16
→ frogonwheels joined14:17
jast yeah, I'd like some !repro here14:17
gitinfo Please paste (using https://gist.github.com/ or similar) a transcript (https://gist.github.com/2415442) of your terminal session, or at least explain exactly what you did that led up to the problem. This will help immensely with troubleshooting.14:17
CalimeroTeknik cmn: I'm not trying to commit, it's irreversible so…14:17
cmn huh?14:17
→ herdingcat joined14:17
CalimeroTeknik ah I get it, absolute symlinks14:17
← kerrick left14:17
cmn that makes no snese14:18
CalimeroTeknik that's why git doesn't record them as symlinks but as hard ones14:18
cmn what's an absolute symlink?14:18
CalimeroTeknik that has an URI starting with /14:18
cmn that could be it14:18
jast shouldn't matter, though14:18
cmn if it looks like it might be pointing outside the repo14:18
CalimeroTeknik yes. can I unstage everything for commit?14:18
jast normally git doesn't care where the symlink points14:19
cmn do what status tells you14:19
→ megalomania joined14:19
CalimeroTeknik I'm unsure if that's what I want14:19
I don't want git to touch my sources14:19
← TeckniX left14:19
CalimeroTeknik checkout does that.14:19
→ TeckniX joined14:20
← diegoviola left14:20
CalimeroTeknik so how do I un-add?14:20
reset?14:21
I'd like to be sure nothing will be touched14:21
→ kerrick joined14:22
jast here's a ttyplay recording of git treating an absolute symlink as a symlink: http://jk.gs/absolute-symlink.tty14:22
→ kenperkins joined14:22
→ tatsuya_o joined14:22
cmn for some reason my system thinks wireshark should be opening that...14:23
← kenperkins left14:23
→ itzhack joined14:24
← psoo left14:24
→ masonkramer joined14:24
→ ishi joined14:25
ishi hi14:25
gitinfo ishi: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.14:25
← replore left14:26
gitinfo set mode: +v14:26
gitinfo set mode: +v14:27
← juvenal left14:27
cbreak-work git needs a patience option for git diff14:27
gitinfo set mode: +v14:27
ishi I'm using git+gerrit. I commited, pushed... in gerrit I noticed that I forgot to do soemthing, so I did this fetch/checkout FETCH_HEAD... I amended my commit, pushed it14:27
so far so good.14:27
itzhack hi, i just git rm a file but did not git commit, how do i un-rm this file and put it back in status "modified" ?14:27
cbreak-work itzhack: git checkout HEAD -- filename14:28
or14:28
if you did not actually use git rm but rm14:28
git checkout -- filename14:28
itzhack yes, i used git rm14:28
ishi unfortunetly I continued working on my 'fix' branch... but I thought it was OK, so I commited my changes, pushed them14:28
cmn cbreak-work: git diff does have a --patience option14:28
itzhack but i had it modified14:28
cbreak-work cmn: yeah, but git add -p does not14:29
ishi but the problem is that now when I try to do git checkout master, I get the following error:14:29
itzhack checking it out wont erase my changes?14:29
ishi error: Your local changes to the following files would be overwritten by checkout:14:29
cbreak-work cmn: and git diff doesn't have a git config flag to set patience always14:29
ishi can I force this checkout?14:29
← juanfer left14:29
cbreak-work ishi: just undo the changes with git checkout or git reset14:29
or stash them with git stash14:29
or commit them with git commit14:29
ishi I don't have anything uncommited14:29
cmn cbreak-work: IIRC that was proposed on the list, though I can't remember if it got far14:29
→ jceb joined14:29
cbreak-work ishi: git disagrees14:30
← _huoxito left14:30
ishi $ git status14:30
# On branch fix14:30
nothing to commit (working directory clean)14:30
jast itzhack: you deleted the file. so, obviously, your uncommitted changes are gone.14:30
itzhack no i didn't14:31
cbreak-work git rm will delete the file14:31
AND stage the file for removal14:31
→ fly9 joined14:31
cbreak-work so both the filesystem and the index will not contain the file14:31
itzhack git rm didn't remove my file14:31
jast if you did what you said you did, then yes, you deleted the file14:31
otherwise you didn't do what you said you did14:31
cbreak-work git rm WILL remove the file14:31
if it didn't remove the file, then you didn't use git rm but git rm --cached14:32
ishi cbreak-work: my git status tells me that it does agree14:32
itzhack i did git rm --cached <file>14:32
yes14:32
cbreak-work then just git add the file again14:32
jast itzhack: well, in that case 'git status' tells you how to undo that14:32
itzhack but i don't want to add it14:32
← DuzyZiemniak left14:32
itzhack i want it to come back to mode "modified"14:32
jast right. do what 'git status' suggests.14:32
ishi cbreak-work: what I want to get is *exactly* the same sate I have at the moment, but on branch master (instead of branch 'fix')14:33
and the origin/master has this state, afaik14:33
itzhack jast: thanks a lot14:33
yes14:33
it did, and did not revert the file to its un-modified state14:33
great14:33
thanks14:33
ishi so I could just remove the bloody thing and clone it again... but I'd rather not do it every time I bump into a problem like this :(14:33
→ NiteRain joined14:34
jast ishi: if you really don't care about things getting overwritten, checkout has a --force flag14:35
if it still complains, delete the offending files14:35
← hide left14:35
ishi jast: but I don't have any local changes :(14:36
→ kevwil joined14:36
ishi at least not in the 'fix' branch14:36
or at least that's what the git status tells me :(14:36
→ hide joined14:36
infogulch|afk → infogulch14:36
jast ishi: the message refers to *untracked* files that you currently have that are *tracked* in the target branch14:36
git tries to avoid clobbering untracked files unless you tell it otherwise14:37
← warthog9 left14:37
→ airborn joined14:38
→ stkrzysiak joined14:38
→ fjfish joined14:39
← francisfish left14:40
et who manages git-scm.com?14:40
→ guampa joined14:40
FauxFaux et: See the committers on the github link at the bottom.14:40
→ StuckMojo joined14:41
FauxFaux Which is a fucking ruby project that doesn't build, as expected, because that's what ruby projects do. Just for reference.14:41
StuckMojo morning. does git have issues with utf8 filenames?14:41
i've got: カードデザイナAIテンプレート（表裏）.zip14:41
FauxFaux It was poor on Windows before very recently.14:42
StuckMojo and it shows up as a bunch of escapes in git status14:42
← itzhack left14:42
StuckMojo this is on debian squeeze14:42
with git from backports i think14:42
jast StuckMojo: they are shown escaped in some places but are stored literally (IIRC)14:42
→ juanfer joined14:42
jast not sure who thought it was a good idea to escape 'em in 'git status'14:42
StuckMojo hmmm...ok, how do i git add it ;)14:42
jast git add <filename> ;)14:42
et FauxFaux: well, that's basically true for any language14:42
StuckMojo oh wow, that worked14:43
heh14:43
thanks :)14:43
→ marhaban joined14:43
← dark_element left14:43
cmn status and others show it escaped by default just in case; you can tell it not to with some falg14:43
kpreid config core.quotepath sounds relevant14:44
bullicon StuckMojo: what does カードデザイナAIテンプレート（表裏）mean by the way14:44
StuckMojo uhhhhh14:45
honestly i don't know14:45
bullicon AI Designer card templates inside and outside !14:45
StuckMojo it's an adobe illustrator template file inside it14:45
← giallu left14:45
bullicon cool StuckMojo14:46
StuckMojo inside and outside, eh? it should really say "front and back"14:46
→ diegoviola joined14:46
← cagedwisdom left14:46
→ _huoxito joined14:46
StuckMojo i'm helping do Japanese translation of a django music site14:46
jast dang, still far from #1 on the karma top list14:46
_ikke_ jast: Is gitinfo counting karma?14:47
← jspiros left14:47
jast _ikke_: yes, since a few days ago. but not using the ++ syntax.14:47
_ikke_ jast: yeah, that's better14:47
jast: probably looking for things like thanks14:47
jast yes. thanks for your thanks just now. ;)14:47
StuckMojo thanks jast14:48
thanks cmn14:48
thanks kpreid14:48
as you all helped ;)14:48
_ikke_ jast: :D14:48
→ giallu joined14:48
← giallu left14:48
→ giallu joined14:48
→ gavin_huang joined14:48
→ jspiros joined14:49
jast !karma14:49
gitinfo This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them.14:49
→ martinjl` joined14:49
_ikke_ jast: Is there an overview somewhere?14:49
jast StuckMojo: stop thanking people who are above me in the ranking, dammit! :P14:49
StuckMojo that's better than ++ because people don't have to know any magic to make it work14:49
jast yeah, there's a bot command to get a list of the top 5 or an arbitrary list of folks14:50
right now it's prone to highlighting lots of people, though. I'll have to fix that before announcing it.14:50
← tatsuya_o left14:50
← martinjlowm left14:50
→ tatsuya_o joined14:51
→ edude03 joined14:51
_ikke_ jast: Maybe only allow it in pm's14:51
← dhruvasagar left14:51
← jonnu left14:56
← marc_kdab left14:57
← edude03 left14:57
→ duckxx joined14:57
→ edude03 joined14:57
→ Element9 joined14:58
→ dark_element joined14:59
← edude03 left15:00
← quazimodo left15:00
→ quazimodo joined15:00
→ markgifford joined15:01
ishi jast: ok, so... what can I do with it? :)15:02
duckxx how do i display files changed in a git log ?15:02
ishi jast: as I said, I want my master to look like my 'fix' branch15:03
:)15:03
← wolog left15:03
duckxx git log -p --online15:03
_ikke_ duckxx: --stat or something15:03
→ jim0203 joined15:04
jast gotta go15:04
→ Pookz joined15:04
← cbreak-work left15:04
duckxx how do u add color to it ??15:04
git log is just all white15:04
← d3vic3 left15:04
duckxx i mean single color15:04
_ikke_ duckxx: man git config15:05
gitinfo duckxx: the git-config manpage is available at http://git-scm.com/docs/git-config.html15:05
_ikke_ see the color options15:05
bremner duckxx: change your terminal configuration?15:05
→ ph^ joined15:05
← jackh left15:06
← bitkiller left15:06
→ wolog joined15:06
StuckMojo ok thanks folks15:06
← StuckMojo left15:06
→ d3vic3 joined15:06
→ cakehero joined15:06
→ bitkiller joined15:08
→ tcurdt joined15:08
← tcurdt left15:08
← stodan left15:09
→ cbreak-work joined15:09
← gusnan left15:10
← cbreak-work left15:10
← d3vic3 left15:11
SidGBF how to revert all non-commited changes?15:12
→ fp1 joined15:12
FauxFaux git stash15:12
SidGBF mmm15:12
→ d3vic3 joined15:12
← abetaha left15:13
→ abetaha joined15:13
→ Lisimba3 joined15:13
← Lisimba2 left15:15
← dcarper left15:16
→ harshpb joined15:16
→ rath joined15:18
→ rgenito joined15:18
→ noel__ joined15:18
← noel__ left15:18
→ dcarper joined15:20
rgenito i run "git rebase --onto trunk master~2 master" and i get the error, "needed a single revision"15:20
...doesn't "master~2" grab the last 2 commits for the branch 'master'?15:20
← pro_metedor left15:21
→ andrew_ joined15:22
→ lonewulftp joined15:23
→ pdtpatrick joined15:24
← DoniRS left15:24
rgenito *sigh*15:24
rgenito i wish #git was actually a community =\15:24
← JohnBat26 left15:24
← jim0203 left15:25
→ aristidesfl joined15:25
← harshpb left15:26
→ jim0203 joined15:26
→ harshpb joined15:26
→ wilmoore joined15:27
← dvide left15:28
bremner a community being defined as a consultant with a <= 5 minute response time?15:28
gitinfo set mode: +v15:30
bremner rgenito: what does "git rev-parse" return for each of those refs you pass to rebase?15:31
i.e., does it error out for any of them?15:31
→ CannedCorn joined15:32
→ TrahDivad_ joined15:33
← TrahDivad left15:33
TrahDivad_ → TrahDivad15:33
rgenito bremner: exactly :)15:33
lemme see...15:34
← t4nkd left15:34
rgenito "git rev-parse" doesn't show anything15:34
btw i figured out the issue where i was getting "needed a single revision"15:34
now i'm trying to do, "git svn dcommit" :D15:34
← jceb left15:34
→ mikepack joined15:35
rgenito git tells me, "unable to determine upstream SVN information .... Perhaps the repository is empty. ...."15:35
→ kenperkins joined15:35
rgenito which is true...the svn repository is completely empty. i basically wanna push my entire GIT repository to SVN repository15:35
→ Zenopus_ joined15:36
rgenito i have a linear history on my local GIT repository; however, there's no home for this repository. and the home will be (unfortunately) SVN. so it would be nice to push the history and changes to that SVN repository15:36
talexb rgenito, Can #svn help you with that?15:36
rgenito if not, i'll just upload it.15:36
← hakunin left15:36
rgenito talexb: nope ;(15:36
→ Goplat joined15:37
kpreid rgenito: I would use svn to create the trunk directory, then git svn clone that and rebase your work onto it15:37
talexb rgenito, My only experience in changing VCS' is Perforce to git .. can't help you.15:37
rgenito kpreid: ahhh good idea...thanks!15:37
thanks tho talexb15:37
kpreid note that git-svn is *not* an arbitrary bidir channel - what git-svn produces on the git side is *explicitly* "mirror of svn"15:37
it will insert extra metadata into your git revisions15:38
← Zenopus left15:38
Zenopus_ → Zenopus15:38
rgenito kpreid: wish i knew how to create the trunk directory for the svn repository lol15:38
the repository is completely empty....15:38
← papegaaij left15:38
rgenito google time :D15:38
kpreid rgenito: svn mkdir <repository url>/trunk/15:38
→ rgsteele joined15:38
kpreid that will perform a commit which creates only that dir15:38
← jargon- left15:39
← mishok13 left15:39
→ blacktulip joined15:40
→ sh0ne joined15:40
← Robbo_ left15:40
← SidGBF left15:41
rgenito thanks kpreid!15:42
→ TrahDivad_ joined15:42
← brokenshakles left15:43
← bglusman left15:44
← nb left15:44
→ arvind_khadri joined15:44
← arvind_khadri left15:44
→ arvind_khadri joined15:44
→ brokenshakles joined15:44
rgenito kpreid: do you know of a way for me to "svn co" the empty svn/trunk into a new branch?15:44
← TrahDivad left15:44
TrahDivad_ → TrahDivad15:44
rgenito or should i use "git svn clone" to bring it into a new branch?15:45
← com4 left15:45
kpreid don't bother with svn, stick to git svn clone15:45
rgenito (i have the rebasing experience... just not the svn part)15:45
← kenperkins left15:45
rgenito any special parameters i'll need to pass through git svn clone? (i'm looking at the help now)15:46
← rphillips left15:46
← thorie left15:46
→ jim0203_ joined15:46
← sontek left15:47
kpreid --stdlayout15:47
other stuff is mostly if you want to make a nicer git mirror, but since you are keeping git locally and publishing svn, that doesn't matter15:47
← thiago left15:48
→ marc_kdab joined15:48
rgenito is that normal for it to clone into a directory?15:48
...i was thinking cloning into a branch where i could rebase to =|15:49
→ Davey_ joined15:49
← neurodrone left15:49
kpreid just take the new repository and copy your existing commits into it15:49
← jim0203 left15:50
jim0203_ → jim020315:50
rgenito ooooh i got ya... pull the changes into the new git-svn repository?15:50
cool...thanks :)15:50
this feels like a puzzle, hehe15:50
→ kenperkins joined15:50
← andrew_ left15:51
→ DuzyZiemniak joined15:51
→ andrew_ joined15:51
rgenito oh wait, now how do i commit? =\15:52
← Dave^| left15:52
→ iwoj joined15:52
rgenito i'm getting farther tho... thanks again15:52
kpreid You mean, how do you create SVN commits for your Git work? that is git svn dcommit's job. But first you will need to rebase onto that initial revision so that git svn knows what you want to do with them.15:54
→ maletor joined15:54
→ bglusman joined15:54
← cccaldas left15:54
rgenito ahh ok15:54
← bglusman left15:55
→ v0n joined15:56
← bnz left15:56
← |bnz left15:56
rgenito hmm, i still get "Unable to determine upstream SVN information from HEAD history."15:57
let me pastie.org15:57
← mdhopkins left15:57
→ Robbo_ joined15:57
← harshpb left15:57
← shadowe989 left15:57
→ kadoban joined15:57
← abetaha left15:58
→ peppyheppy joined15:58
← gchristensen left15:58
← freeed left15:59
← lonewulftp left15:59
→ linusoleander joined16:00
rgenito kpreid: does this make sense?16:01
http://pastie.org/private/7cco4utv9peheeu0tpioqw16:01
← hyperair left16:01
→ mdhopkins joined16:01
→ ShadesofGraylin| joined16:01
kpreid ok, you have that backwards16:02
← daethorian left16:02
← taylanub left16:02
kpreid you need to take your existing work and rebase *onto* the svn branch, but you did the opposite, or something — can't quite tell16:03
← whitman left16:03
→ comptekki joined16:03
→ lonewulftp joined16:03
rgenito when you say "svn branch", do you mean http://example.com/repository/branches/thebranch ... or do you mean a branch listed in "$ git branch -a" ?16:04
kpreid Sorry, I mean the git branch which is git-svn created.16:04
→ CannedCo_ joined16:04
← masonkramer left16:04
rgenito how do i create the "git-svn created" branch?16:04
i think i'm lost on just that one part16:05
kpreid So, starting from after you've git svn cloned, do NOT git pull, but rather git fetch from your other repository. Then git checkout -t the other repository's branch of interest, and then on *that* branch do git rebase --onto master16:05
→ williamcotton joined16:05
← Davey left16:05
Davey_ → Davey16:05
kpreid git svn clone creates it! probably named master16:05
← senny left16:05
rgenito ahh16:05
kpreid but when you did 'git pull' that broke things because git pull does a merge, i.e. creatign nonlinear history, by default16:05
→ Davey_ joined16:05
← Davey_ left16:05
→ Davey_ joined16:05
kpreid (either that or it did nothing)16:06
← LimeBlast left16:06
← lonewulf` left16:06
diminoten wait, when I check files into git, do the permissions I give those files stick with them?16:06
← CannedCorn left16:06
rgenito ah i see16:06
thanks kpreid i'll give this a shot now :)16:06
kpreid rgenito: when mucking around like this I like to keep a `gitk --all` open to visualize the state of all the branches16:06
rgenito "git checkout -t the other repositories branch" ...16:07
kpreid diminoten: execute bit only16:07
diminoten bleh16:07
← axl_ left16:07
diminoten I didn't set that before checking in, how depressing16:08
rgenito kpreid: "git fetch ../cbt" warns me: "warning: no common commits"16:08
→ axl_ joined16:08
← jn_ left16:08
kpreid Yes, that is correct. Our rebase is to fix that :)16:08
rgenito i mean, of course there aren't going to be any common commits...16:08
kpreid "no common commits" = "no common history"16:08
rgenito ok cool >)16:08
hmm, "get checkout -t svnrebase" tells me, "fatal: Missing branch name; try -b"16:08
svnrebase isn't an actual branch yet lol16:09
i only have master16:09
oh wait! i also have the branch remotes/git-svn hehe16:09
kpreid hm16:09
→ alvaro_o joined16:10
rgenito "Then git checkout -t the other repository's branch of interest, and then on *that* branch do git rebase --onto master"16:10
← comptekki left16:10
kpreid You want to checkout -t what you fetched, I'm not sure what name it goes under. git branch -a will list all your branches including tracking branches16:10
rgenito so "git checkout -t master"? the branch master is the only i'm interested in16:10
← Chib left16:10
rgenito ahh ok16:10
kpreid No, master is your "empty" branch. You want to checkout -t the fetched copy of your work, creating ANOTHER branch off it which it is ok to mess around with it.16:10
rgenito hmm16:11
→ pheaver joined16:11
← fjfish left16:11
← candybar left16:11
rgenito if my branches are *master and remotes/git-svn ... how do i do that ?16:11
"git fetch ../cbt" worked fine btw :)16:11
kpreid Sounds like the fetch didn't do anything useful. OK, we need to set up a named remote.16:11
→ aantix joined16:11
kpreid git remote add cbt ../cbt && git fetch cbt16:11
Then after that git branch -a should list something like "cbt/master"16:12
← Hounddog left16:12
← aantix left16:12
← alvaro_o left16:12
kpreid (there's probably a nice and pretty way to get all this done simpler, but I don't know it so I'm using the basic stuff)16:12
← lfaraone_ left16:12
→ lfaraone_ joined16:13
→ aantix joined16:13
rgenito ok, but before i do that...16:13
http://pastie.org/private/c2wjykvokgla0wzgih9q16:13
offby1 If I do "git stash drop", and then later wish I hadn't ... is there a way (perhaps with "git reflog") to get that stash back?16:13
rgenito that's my actual history so you can get an idea. it's short, but shows what i did16:13
← Peste_Bubonica left16:13
← brokenshakles left16:13
rgenito offby1: from my understanding, there's no way to get it back :(16:13
← aantix left16:14
kpreid I said to use git remote add.16:14
← logik-bomb left16:14
rgenito ok will do that now16:14
kpreid That way there is a name to store what you fetched under.16:14
rgenito ok sweeet16:14
$ git remote add cbt ../cbt && git fetch cbt16:14
From ../cbt16:14
* [new branch] master -> cbt/master16:14
* [new branch] master.bak -> cbt/master.bak16:15
← stephanmg left16:15
kpreid Right, now you can git checkout -t svnrebase cbt/master16:15
This will create your new branch 'svnrebase' which is derived from 'cbt/master'16:15
← kerrick left16:16
rgenito when i do "git checkout -t svnrebase cbt/master" it still says fatal: Missing branch name; try -b16:16
← ShadesofGraylin| left16:16
→ f0i joined16:16
rgenito ah wait, remotes/cbt/master is the branch... let me try specifying that hehe :)16:16
kpreid NO16:17
Don't do that.16:17
rgenito ok that didn't work anyways so we're good16:17
hehe16:17
kpreid I was just confusing git branch and git checkout16:17
← pheaver left16:17
rgenito ah ok16:17
→ ipoval joined16:17
rgenito sweeet i got it16:17
"Branch svnrebase set up to track remote branch master from cbt."16:17
kpreid git checkout generally assumes you're asking for a branch exists; in this case -b is right. git checkout -b -t svnrebase cbt/master16:18
→ Peste_Bubonica joined16:18
rgenito cool :) from here, i think it is just rebase and then dcommit16:19
kpreid Right, but, one last point:16:19
← mikepack left16:19
kpreid this is a special case because we're rebasing all the way to the root of 'svnrebase', not just a portion. So instead of the usual syntax, it will be git rebase --onto master --root16:20
→ mmc joined16:20
rgenito listening16:20
kpreid (ordinarily we would need to specify the "upstream" point which rebase is supposed to snip off history at)16:20
That's it.16:20
← caddoo left16:20
→ brokenshakles joined16:20
← pyhead left16:20
rgenito so originally it would be master~2 or something, but here... git rebase --onto master --root ?16:20
kpreid Um, what do you think master~2 is?16:21
← herdingcat left16:21
rgenito ^^ that will rebase from the "root" of master into the current branch?16:21
i have no idea what master~2 is haha. just got that from a bunch of google results.16:21
kpreid No, we are rebasing the current branch onto master.16:21
The current branch will have more history than it used to.16:21
rgenito ahh16:21
kpreid --root means that, of the *current* branch, we are taking *all* of it back to the root rather than a more recent fragment.16:22
master~2 means "master branch two commits ago" and is completely irrelevant here16:22
offby1 rgenito: bum-mair16:22
rgenito holy god that's a lot of changes!16:22
← v0n left16:22
kpreid Yep, it's going to list every commit you have, because it is rewriting every one of them.16:23
rgenito kpreid: ty for that clarification16:23
kpreid That's what rebase *does*, and why it can be dangerous.16:23
rgenito http://pastie.org/private/c2wjykvokgla0wzgih9q16:23
that probably looks about right :) hehe16:23
← yitz_ left16:23
→ drev1 joined16:23
rgenito yes, i can see why now16:24
kpreid Yeah, that's a normal rebase.16:24
Now, here is something you can do to spot check it is good:16:24
git diff svnrebase cbt/master16:24
That should produce no output.16:24
→ ISF joined16:24
rgenito correct! no output :)16:24
kpreid Congratyerlations.16:24
rgenito so quick question before finalizing this...16:25
kpreid Now a git svn dcommit should succeed.16:25
rgenito when i was on branch svnrebase...16:25
and then i did "git rebase --onto master --root",16:25
what is that command translated to english?16:25
"rebase what is on this branch onto the root of master"?16:26
kpreid You understand about the git model of a graph of commits, yes?16:26
rgenito no, i dont. i do know what gitk looks like tho. (that comment was just to make you laugh :) )16:26
← Sweetshark left16:26
rgenito at this point, i would love to research/understand that model16:26
dcommit is doing all sorts of awesome stuff @_@16:27
kpreid Each "commit" object contains two relevant things: a snapshot of the current tree, and one (or more) parent commits. When you commit, the snapshot is saved and what was the head of the branch becomes the parent of the new commit. Got it?16:27
Furthermore, all of the commits we are interested in here have exactly one parent (more are for merges).16:28
→ Sweetshark joined16:28
← Sweetshark left16:28
→ Sweetshark joined16:28
← alem0lars left16:28
rgenito so then...16:29
kpreid What a rebase does is it takes a series of commits and rewrites them so that the *oldest* commit has a different parent than it used to. (It also treats each one as a diff so that changes made in the parent aren't lost.)16:29
rgenito the "snapshot" of the current tree... what does that mean?16:29
→ hotfloppy joined16:29
kpreid All of your files.16:29
rgenito oooh, so the latest commit has a copy of all of the raw files16:29
cmn each commit has that16:30
rgenito that snapshot has the diff information? (so that when you revert or something... it knows how to reverse)16:30
kpreid *EVERY* commit contains a copy (well, a reference) to all of the files at the time.16:30
Diffs are generated by comparing two commits' trees.16:30
cmn rgenito: snapshts and diffs are opposites16:30
rgenito ahh, ty for the clarification16:30
cmn yes i understand16:30
→ dreiss joined16:30
kpreid OK, so. Rebase rewrites the oldest commit of interest's parent. What "--onto master" means is explicitly specifying that we want that parent to be the commit which is currently the master branch's head.16:31
← giallu left16:31
← noisebleed left16:31
kpreid What "--root" means is that the oldest commit of interest is the *very first* commit on the branch we are rebasing.16:31
← arvind_khadri left16:32
→ candybar joined16:33
→ und3f joined16:33
rgenito i understand --root, but a little hazy on "--onto master". to re-iterate: "--onto master" means "we want this branch's (svnrebase) parent to be the parent commit on master" ?16:33
kpreid Now, rebase doesn't *just* modify parents; it actually takes the diff from each commit and its parent and sequentially *applies* them onto the new base. (that's what all the "applying" lines are.) This way, the result of rebase combines the new parent with the work you are rebasing. In this particular case, master contained no files, so the result is just the same as the original.16:33
rgenito: not "the parent commit on master" but "the commit which is currently master"16:33
← tonesfrommars left16:33
rgenito ahh16:33
← andrew_ left16:33
← meisth0th left16:33
← CodeVision left16:34
← ajpiano left16:34
→ gchristensen joined16:34
→ abetaha joined16:34
→ hyperair joined16:35
→ abetaha_ joined16:35
rgenito kpreid: you kick ass :)16:35
where can i find more information on this?16:35
← jgoss left16:35
rgenito any git books come to mind? or online resources that explain this?16:35
...probably git's official site :)16:35
← afuentes left16:36
← tr-808 left16:36
kpreid I didn't learn git that way, so I can't say, but the channel has its official recommendations16:36
!book16:36
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 !designers !gitt !vcbe and !parable16:36
kpreid There is no official web site for git.16:36
← dreiss left16:36
→ Sky[x] joined16:36
← ill1cit left16:37
kpreid It's all decentralized and stuff, don'tchaknow.16:37
→ kerrick joined16:37
← ipoval left16:37
rgenito hehe16:37
→ cccaldas joined16:37
→ ill1cit joined16:37
→ ajpiano joined16:37
→ lonewulf` joined16:38
→ jgoss joined16:38
← jgoss left16:38
→ jgoss joined16:38
→ bglusman joined16:38
← rapha left16:39
← lapistano left16:39
→ booginga joined16:39
→ lapistano joined16:39
← abetaha left16:39
← cccaldas left16:39
rgenito thanks again! i greatly appreciate all of the help! :D16:39
← BarryCrayon left16:40
← lopin left16:42
← g0bl1n left16:44
→ CodeVision joined16:44
→ superdmp joined16:44
← kerrick left16:44
superdmp what does: previous rebase directory /home/daniele/images-2012-apr-17/django-cms/.git/rebase-apply still exists but mbox given. mean?16:44
→ apok joined16:44
→ neurodrone joined16:45
← fleetfox left16:46
→ mrkuchbhi joined16:46
← mrkuchbhi left16:46
← bglusman left16:47
→ beneggett joined16:47
→ raatiniemi joined16:48
← achadwick left16:48
bremner superdmp: at a guess, you're in the middle of a rebase already?16:48
superdmp bremner: I simply don't know!16:49
I have a patch file, that want to apply with: git am -3 1117.patch16:49
I did this a few days ago, but the patch has been updatd since then16:49
→ Dragooon joined16:50
→ mrkuchbhi joined16:50
hotfloppy ohcibi, thanks about the SherlockHolmes.. fixed! :D16:50
bremner superdmp: what does "git status report"16:50
err, "git status" report16:51
→ thiago joined16:51
mrkuchbhi hi .. i just did a git rm -rf dirname16:51
is there someway to revert that ?!16:51
cmn mrkuchbhi: do what git status says16:51
ohcibi hotfloppy: good! 8-)16:51
superdmp bremner: dpaste16:51
bremner: I'll try again http://dpaste.com/753412/16:52
→ mishok13 joined16:52
← robbyoconnor left16:52
bremner you are in the middle of something, because you have unmerged paths16:52
superdmp mm16:52
→ robbyoconnor joined16:53
superdmp bremner: I will try to unravel myself and get back somewhere sensible16:53
→ Shusshu joined16:53
mrkuchbhi cmn: the prob is... i added a new directory using `git add dirname` ... then realized that i only wanted a few files from it ...so without committing i did `git rm -rf dirname` ... and now i cant find the directory. ...`git status` does not show the directory or the files from that directory in untracked files or in changed files16:54
→ sirpistache joined16:56
→ lacrymology joined16:56
→ docgnome joined16:56
lacrymology if I do git push --all and I have a number of remotes, each tracking different branches, will them be respected?16:57
PerlJam mrkuchbhi: well ... you removed them.16:57
← lonewulftp left16:57
← Poapfel left16:58
docgnome i have a submodule in a submodule and it used to work fine but with 1.7.10 i can no longer git submodule update...16:58
→ Poapfel joined16:58
→ gtrak` joined16:59
← gtrak left16:59
docgnome i get... fatal: Not a git repository: ../../../../../../../../../lxc/cetusa/rootfs/home/docgnome/crm/.git/modules/vendor/plugins/phrails-paperclip/modules/lib/cloud/rsc16:59
→ bglusman joined17:00
mrkuchbhi PerlJam: so git rm removes it from disk as well?17:00
bremner unless you say --cached17:00
PerlJam mrkuchbhi: "Remove files from the index, or from the working tree and the index." You did the latter.17:01
mrkuchbhi :(17:01
okay..thanks17:01
→ phishy joined17:01
→ Araxia_ joined17:01
→ tr-808 joined17:01
← bullicon left17:02
← phishy left17:02
← hyperair left17:03
→ ripero joined17:03
← shruggar left17:04
→ BBBThunda joined17:04
← koo4 left17:04
→ daethorian joined17:06
← tr-808 left17:06
→ kerrick joined17:06
← dtang_ left17:07
← klj613 left17:07
→ dtang joined17:07
← kerrick left17:08
→ kerrick joined17:09
← sgronblo left17:09
→ ProLoser|Work joined17:09
← cvanes left17:09
→ Praise- joined17:09
← ProLoser|Work left17:09
← kerrick left17:10
← Praise left17:10
→ kerrick joined17:10
ripero Hi; I forgot to add a file a few commits ago; is it possible to add it, keeping all commit timestamps, using some git filter-branch magic?17:12
← Lord_DeathMatch left17:12
→ pheaver joined17:13
← TrahDivad left17:13
← brentratliff left17:13
→ luxurymode joined17:13
← angelsl left17:13
← dark_element left17:14
← fisted_ left17:14
→ osmosis joined17:14
→ TheJH joined17:15
cmn mrkuchbhi: if they're cached, looking though git-fsck's dangling blobs can help you get the contents back17:16
but it's a lot of work17:16
→ goshawk joined17:16
cmn s/cached/staged/17:16
→ juvenal joined17:16
m1sc ripero: you can do this by interactive rebasing -- commit time won't be changed until you choose to do so17:17
→ koo4 joined17:17
← juanfer left17:17
m1sc ripero: !rewrite17:17
gitinfo ripero: [!rewriting_public_history] Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)17:17
← Shusshu left17:17
← hadees left17:18
neersighted → neersighted|AFK17:18
→ fisted joined17:18
→ dreiss joined17:19
→ fuzzybunny69y joined17:19
← kerrick left17:20
→ acrocity joined17:20
neersighted|AFK → neersighted17:20
→ markus_ joined17:20
← h0bbit left17:21
← zinjacoder left17:23
→ gusnan joined17:23
→ kerrick joined17:24
← Superlokkus left17:24
← cortana2 left17:24
← kerrick left17:25
← juvenal left17:25
→ khmarbaise joined17:26
→ kerrick joined17:27
← kerrick left17:27
← ipalaus left17:27
→ juvenal joined17:27
→ kerrick joined17:28
← markus_ left17:28
→ Superlokkus joined17:28
→ v0n joined17:29
neersighted → neersighted|AFK17:29
← kerrick left17:29
→ hadees joined17:29
neersighted|AFK → neersighted17:30
→ alberto56 joined17:30
→ AeroNotix joined17:30
→ tworkin joined17:30
tworkin !submodule_rm17:30
gitinfo [!submodules_rm] You want to delete submodules? Excellent choice! All commands are in the superproject. Edit/delete .gitmodules to remove the submodule. Then `rm -rf submodulepath; git rm -f --cached submodulepath; git commit -am "Removed submodules!"` Inspect .git/config for "submodule" entries to remove. Inspect .git/modules for caches to remove "!gitslave" or "!subtree" (type them!) might be alternatives17:30
tworkin hmmm, doesnt address the case where `git submodule add` failed in a way that left a .gitmodules entry but no folder17:31
← marcogmonteiro left17:32
← mischat left17:32
cmn skip that part17:32
← alberto56 left17:32
cmn do the steps that don't involve removing that directory from disc17:33
ripero m1sc: ok, thanks... but then how do I keep the old CommitDate when committing?17:33
→ kerrick joined17:34
← mdhopkins left17:36
← Spockz left17:36
→ nbari joined17:38
← nbari left17:38
→ nbari joined17:38
← daethorian left17:38
→ edude03 joined17:39
m1sc ripero: you don't have to keep it. just get familiar with interactive rebasing and you'll see (backup your repo, start playing..)17:41
ripero: !book will help17:41
gitinfo ripero: 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 !designers !gitt !vcbe and !parable17:41
nbari hi all, inside a branch I did a merge branchX but that deleted a directory created from branchA, now when I do a merge branchA for trying to restore the delated files, I just get an Already up-to-date. message, but the directory is not pulled17:41
any idea of how to refresh or re-merge so that I can get the directories ?17:41
← gandhijee_ left17:42
bremner nbari: you could checkout the directory from the other branch and make a new commit.17:43
→ FreeArtMan joined17:43
→ Dave^| joined17:43
← peppyheppy left17:44
← Robbo_ left17:44
ripero m1sc : ok, so I did a first interactive rebase, and then a second one with --committer-date-is-author-date ;17:46
→ ConstantineXVI joined17:49
nbari bremner: when doing the git checkout branchA I don't get resoted the dir17:49
← kerrick left17:50
bremner nbari: see man git checkout, with a path17:50
gitinfo nbari: the git-checkout manpage is available at http://git-scm.com/docs/git-checkout.html17:50
→ ryez_ joined17:51
← ryez_ left17:51
→ ryez_ joined17:51
nbari i created another file on the branchX and later tryied to merge it17:51
→ Tinti joined17:51
nbari that worked but the other files still not been restored17:52
like if the merge that deleted the files had a priority17:52
→ tr-808 joined17:52
→ kerrick joined17:52
→ daethorian joined17:55
→ edlothiol joined17:55
→ AlbireoX`Laptop joined17:56
← v0n left17:56
← tr-808 left17:57
→ bolt joined17:57
← kerrick left17:57
→ freeed joined17:57
→ Spockz joined17:57
→ plasticboy joined17:58
→ bullicon joined17:58
→ Eridius joined17:59
← awx left18:00
← hellopat left18:01
→ harshpb joined18:01
→ peppyheppy joined18:02
alex88 i've found a webapp similar to github, someone knows the name?18:04
← bubbles10101 left18:06
← _huoxito left18:06
← JEEB left18:06
← tworkin left18:06
→ bubbles10101 joined18:07
← BBBThunda left18:08
EugeneKay !github_cheap18:08
gitinfo [!host_gui] If you want to mange your hosted git repos through a web GUI, you have a few options: !gitlab !gitorious and !gerrit are the big ones. Keep in mind that all of these are less flexible and more resource-intensive than !gitolite18:08
← jgoss left18:09
→ kerrick joined18:10
← juvenal left18:11
offby1 rgenito: FYI -- http://stackoverflow.com/questions/89332/recover-dropped-stash-in-git18:11
→ hunterp joined18:12
→ jcromartie joined18:12
jcromartie Is this documentation wrong? http://git-scm.com/book/en/Git-Tools-Submodules18:12
hunterp I checked out the clone at a revision. However, none of the files changed. why ?18:12
jcromartie I added a submodule to my repo, but it's empty18:13
← ripero left18:13
→ alvaro_o joined18:14
→ HyperNerdV2 joined18:14
← alvaro_o left18:14
jcromartie there are so many things wrong with this tut18:14
it's a shame that it's the first Google result!18:14
← pheaver left18:14
← raatiniemi left18:14
hunterp jcromartie: yes18:15
I checked out the clone at a revision. However, none of the files changed. why ?18:15
→ pheaver joined18:15
→ hgb joined18:15
→ juvenal joined18:15
→ adhawkins joined18:16
jcromartie hunterp: what clone? none of which files?18:16
hgb Hi. I have this setup where we have a subversion server, a common git-svn repo which I've cloned, and where I use git svn dcommit to commit (so asynchronously, in a way), but 'git fetch/pull --rebase' to update from upstream.18:16
hunterp jcromartie: when I git checkout, my expectation is that my clone will "rewind" in time to 4 checkouts ago. Why has it not ?18:17
hgb Now, somehow there was a fuckup during my dcommit, and it failed. I did a git svn rebase (stupid), which seemed to fix things, but now my remotes/mirror/trunk (upstream trunk) is wrong. It's four commits (the four I dcommited) too old.18:17
Any ideas how to fix?18:17
→ tr-808 joined18:18
hunterp hgb: cp -a your repo, and fuck around with one18:18
jcromartie hunterp: what is your checkout command, exactly18:18
← daethorian left18:18
hunterp git checkout <rev>18:18
→ macmartine joined18:18
hgb hunterp: I suppose that's an alternative.. However, now that I look at it, it appears that my problem might not be that my local repo is screwed, but that the common git-svn ("read-upstream") is...18:19
Which is worse, I suppose...18:19
hunterp hgb: minimize your work to your goal.18:19
hgb Sounds ideal.18:20
jcromartie hunterp: that will give you a detached head at that commit18:20
→ tworkin joined18:20
jcromartie hunterp: what are you expecting, exactly?18:21
you probably want to reset18:21
unless you want to create a new branch at that revision18:21
in which case you should use: git checkout -b my-temp-branch <rev>18:21
→ v0n joined18:21
hunterp jcromartie: I see. thank you18:21
jcromartie but otherwise, if you really just want to reset a branch to the revision, use: git reset <revision>18:22
→ com4 joined18:22
hunterp jcromartie: I reset, but nothing happened. do i need to commit after reset?18:22
← tr-808 left18:23
hgb hunterp: It seems I have a git-svn repo to unfuck, a repo I don't really think I have any write access to... Oh well, someone will be in in the morning to assist..18:23
jcromartie I think you've got a mess on your hands… what is your starting point? Are you sure you didn't already checkout that commit as a detached head?18:23
hunterp jcromartie: i don';t care about the repo. the moment I get my files, I'm no longer git'ing this clone18:24
← oscar_ left18:24
→ tr-808 joined18:24
jcromartie OK so do you have a fresh clone?18:24
is it a public repo I can look at, BTW?18:24
← Poapfel left18:24
← tatsuya_o left18:24
→ tatsuya_o joined18:25
← v0n left18:25
→ rphillips joined18:25
← vmiklos left18:26
→ v0n joined18:26
→ thorie_ joined18:26
→ hydrozen joined18:27
→ Poapfel joined18:27
← hotfloppy left18:27
← tr-808 left18:29
hunterp jcromartie: http://code.google.com/p/marketbilling/source/list18:29
→ Rotwang joined18:29
← optixx left18:29
← tatsuya_o left18:29
jcromartie OK now which commit do you want18:29
hunterp jcromartie: 6a18:29
← Rotwang left18:29
hunterp jcromartie: aka the first18:30
jcromartie lolwut?18:30
THAT'S WHAT YOU HAVE!18:30
hunterp no18:30
→ vmiklos joined18:30
→ Rotwang joined18:30
← Rotwang left18:31
← harshpb left18:31
jcromartie yes18:31
there are 4 commits in the entire repo, on a single branch18:31
→ Rotwang joined18:31
← pandeiro left18:31
← marc_kdab left18:31
hunterp jcromartie: no, i confirmed that the file AndroidManifest.xml has "singleTask" in the last (most current) revision, and lacks it in the first (6a) revision18:31
jcromartie when you clone the repo, you automatically get the default branch (master) at the latest commit (6a669ef39c)18:31
→ optixx joined18:31
← Rotwang left18:32
hunterp jcromartie: the latest commit is not 6a18:32
its ef18:32
→ nb joined18:32
jcromartie right sorry18:32
← theCole_ left18:32
jcromartie OK so you want the file at that state18:32
→ Rotwang joined18:32
→ sdeboey joined18:33
→ whitman joined18:33
jcromartie that's the tip of the branch though18:33
hunterp jcromartie: i want the entire repo at 6a18:33
→ hyperair joined18:34
← Rotwang left18:34
→ oojacoboo joined18:34
jcromartie I'm sorry no, 6a is the latest18:34
just run "git log" and see for yourself18:34
hunterp jcromartie: no, 6a is march, ef is ,may18:34
→ laen0k joined18:35
jcromartie what's the full SHA18:35
6a669ef39cc1388164f12cfb53b54699af388a5f is Mar 22 201218:35
hunterp correct18:35
jcromartie ef5e96db315dfbf8f9557565cb4ad572d95181ea is Apr 5 201118:35
hunterp that is what I need18:35
I need to go back to march18:35
← Poapfel left18:36
jcromartie OK then: git checkout -b march 6a669ef39cc1388164f1218:36
then you'll be in a new branch at that commit18:36
→ beastd joined18:36
jcromartie but if we're looking at literally the *same* repo, then that is the latest commit anyway… but I have a feeling you are working on a private copy already with additional commits right?18:36
hunterp jcromartie: no, singleTask is still in the manifest, no files have changes18:36
jcromartie: no CLEAN 100% clean18:36
jcromartie so you just cloned this repo18:37
→ ProLoser|Work joined18:37
jcromartie from Google code18:37
hunterp yes18:37
ProLoser|Work i can't figure out how to get sublime to ignore the git folder18:37
→ JEEB joined18:37
hunterp I did git checkout -b march 6a669ef39cc1388164f12 and no files have changed. WHY ?18:37
← gavin_huang left18:38
← kerrick left18:38
jcromartie because it's ALREADY THE NEWEST commit!18:38
→ sontek_ joined18:38
jcromartie OK tell me18:38
in your clone18:38
→ monk12 joined18:38
hunterp :)18:38
jcromartie run this: git log --oneline18:38
and paste somewhere please18:38
hunterp jcromartie: http://pastebin.com/3hXgjZcf18:39
monk12 hello all, I just cloned a git repository and i would like to 'merge' in a pull request. not sure how to use git pull on the command line. all i know is it is pull request #85 in the project and a long commit number.18:39
jcromartie OK yes, so you see 6a669ef is the latest18:39
monk12 (I am very new to git.)18:40
jcromartie and "singleTask" was added in 7638aa818:40
which is a *parent* of 6a669ef18:40
hunterp jcromartie: right and 76 is in april and 6a is march. I am still not in march, i am in april18:40
jcromartie so *of course* it's "still there" in 6a669ef, and *of course* nothing has changed because 6a669ef IS THE CURRENT HEAD OF master ANYWAY!18:40
hunterp i need to get to march18:40
jcromartie you are not looking at the year18:40
as I said18:41
2011 vs 201218:41
hunterp 6a == march == "no singleTask"18:41
jcromartie: appreciate it. I see now. I am the dyslexic one :)18:41
→ oscar_ joined18:41
jcromartie HOLY SHIT18:41
OK great18:41
← v0n left18:41
jcromartie sorry18:41
← barrbrain left18:41
PerlJam jcromartie++ (patience :)18:42
oojacoboo hey, if I had files that were under version control then I add the master folder to .gitignore, how do I remove that folder from version control so it's not part of the repo anymore?18:42
hunterp jcromartie++18:42
Vinnie_win can I cherry-pick a range of commits?18:42
hunterp PerlJam++18:42
jcromartie oojacoboo: git rm --cached18:42
oojacoboo jcromartie, that sounds so simple, is that it?18:42
jcromartie yup18:43
→ zarac joined18:43
oojacoboo where would I run that as well if I want to preserve my local files?18:43
PerlJam monk12: You're talking about github pull requests? They a help page that explains how to do it.18:43
jcromartie it will be ignored, and then removed from the index18:43
→ v0n joined18:43
jcromartie oojacoboo: that will preserve your local files18:43
oojacoboo ah ok, so it just removes the index18:43
thanks jcromartie18:43
→ tr-808 joined18:43
← linusoleander left18:44
→ kerrick joined18:44
→ mcluskydodallas joined18:44
← Lummoxx left18:44
zarac Hey folks! I just read the guide for mercurial. They make use of clone quite a bit for features etc. Is this normal in git too or do branches cover the good-you-get-out-of cloning?18:45
← spaceonline left18:45
→ g0bl1n joined18:45
→ h0bbit joined18:45
bremner branches18:45
→ raatiniemi joined18:45
zarac Thanks! ; )18:45
→ EtgarDizz joined18:45
gitinfo set mode: +v18:46
EtgarDizz hi guys18:46
gitinfo EtgarDizz: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.18:46
zarac Why would anyone use mercurial over git?18:46
← peppyheppy left18:46
PerlJam zarac: dunno.18:46
EtgarDizz I have a local repository, but I'd like to push to it as another user on this computer18:46
bremner why don't you ask them?18:46
zarac Because they'll lie ; )18:46
← blacktulip left18:46
EtgarDizz so I'm logged in as user a, but want to push as user b18:47
bremner zarac: mercurial users love nothing better than to talk about how much better hg is than git.18:47
PerlJam EtgarDizz: you probably want to use the ssh protocol18:47
zarac EtgarDizz: Perhaps you can just add user@ before your host18:47
EtgarDizz I've tried, it just hangs!18:47
zarac bremner: ; )18:47
EtgarDizz I've tried to change the push remote def into ssh://b@domain.com:port/home/a/repo18:48
this doesn't work18:48
→ Lummoxx joined18:48
zarac I'll just stick to git. I've had few problems with it (and those were of course due to ignorance).18:48
← Sigma left18:48
← kerrick left18:48
← tr-808 left18:48
EtgarDizz sorry - ssh://b@domain.com:port/home/a/repo18:48
zarac EtgarDizz: I'm not sure that's how you specify port. The path usually comes directly after the colon.18:49
PerlJam zarac: when faced with choices of hg, bzr, or git, I chose git because it fit my brain best. I'm sure for other people hg or bzr fit their brain.18:49
EtgarDizz this is how I've cloned it to another machine...18:49
→ tr-808 joined18:49
EtgarDizz git clone ssh://user@domain: port/path/to/repo18:49
zarac PerlJam: Thanks. It fits my brain well too. : )18:49
bremner EtgarDizz: you could make an ssh host alias with the user and port18:49
← hunterp left18:49
EtgarDizz bremner, how's that done?18:49
monk12 PerlJam, k think i found some good blog on github site explaining.18:50
zarac EtgarDizz: Is it on another port other than 21? Otherwise just do ssh://a@domain.com:/home/a/repo (b cannot push to a)18:50
← adhawkins left18:50
EtgarDizz bremner, I should also note that when trying to use the EXACT same command from another machine, there's no problem. just when trying to use it on the machine where the repo resides, it hangs...18:50
yes it's on another port18:50
← ISF left18:50
zarac hmm18:51
bremner well, probably a question of ssh setup on that machine18:51
EtgarDizz what you mean "setup"? the ssh works fine when I try to use is remotely18:52
madsy_ → Madsy18:52
EtgarDizz is there a chance it's "blocked" for local use?18:52
bremner do you have the right key locally?18:52
→ replore joined18:53
EtgarDizz bremner, I'm not even being asked for a key... it just hangs after I hit enter18:53
→ kerrick joined18:53
yrlnry Suppose I have made a bunch of commits with subjects like 'fixup! ...'. I want to squash these into the commts that they fix up. I can do git-rebase -i --autosquash, which opens the editor to edit the rebase plan, and then quit the editor without making any changes. Is there a way to skip the pointless editor step?18:53
bremner can you ssh directly without git?18:53
yrlnry git-rebase does not seem to have a --no-edit option.18:53
EtgarDizz you mean locally?18:53
bremner yes18:53
→ mdhopkins joined18:54
EtgarDizz damn!18:54
no, I can't18:54
bremner good news, it's not a git problem ;)18:54
EtgarDizz ok :) where should I direct this question now?...18:54
bremner !ssh18:55
gitinfo Please use SSH to talk to remote repos. http:// and git:// are completely unsecured. You can try to configure "smart" https:// access, but it is usually much easier to just stick with ssh://18:55
bremner hmm.18:55
← freeed left18:55
EtgarDizz !ssh18:55
gitinfo Please use SSH to talk to remote repos. http:// and git:// are completely unsecured. You can try to configure "smart" https:// access, but it is usually much easier to just stick with ssh://18:55
EtgarDizz a ha18:55
bremner it has the sound of a firewall problem18:55
→ juanfer joined18:56
→ shruggar joined18:58
← marthinal left18:58
← Dulak left18:58
→ peppyheppy joined18:59
← FreeArtMan left19:00
→ jgoss joined19:01
monk12 okay to use 'git pull' to pull in someone's changes i see on many blogs/tutorial i need to know the branch_name (to do 'git pull http:..blah.git branch_name'). i dunno how to figure that out by looking at this page: https://github.com/digitalBush/jquery.maskedinput/pull/8519:02
bremner monk12: use git fetch19:02
← dv310p3r left19:02
osmosis i am trying to git push origin to put my locally created branch on the server, I get the error ! [rejected] master -> master (non-fast-forward)19:03
monk12 k i shall try that. just assumed needed 'git pull'19:03
→ mneorr_ joined19:03
→ dv310p3r joined19:04
← docgnome left19:04
EtgarDizz osmosis, what command are you typing?19:05
→ freeed joined19:05
osmosis EtgarDizz, git push origin19:06
← mneorr left19:06
EtgarDizz osmosis, unless you've configured your new branch to be the default branch to act upon, you need to add the branch name in the command. so first "git status" to make sure you're on the new [branchname], and the "git push origin [branchname]"19:07
rgenito after doing, "git svn fetch" on a local branch that is tracking a remote svn repository....19:09
...how do i merge those changes into my local branch?19:09
← chrisdev left19:10
osmosis EtgarDizz, thx19:11
EtgarDizz osmosis, np19:11
← rgsteele left19:11
← raatiniemi left19:12
→ dhruvasagar joined19:13
← bitkiller left19:15
← dr_lepper left19:16
→ chrisdev joined19:17
← replore left19:18
→ tqmz joined19:19
← Lgb left19:20
← beneggett left19:21
← p3rror left19:21
← aristidesfl left19:21
→ Lgb joined19:22
→ Codif joined19:22
→ texasmynsted joined19:22
← macmartine left19:23
→ fayimora joined19:24
← Peste_Bubonica left19:24
← dhruvasagar left19:26
texasmynsted what do I do with a corrupt lose object? I have one that is preventing me from gc and from repack, etc. If I git cat-file -p , I just see author, committer etc. I do not know what the file is.19:27
← maletor left19:27
→ jcapinc joined19:27
→ j416 joined19:28
texasmynsted if its corrupt seems like I should be able to re-fetch it from the origin/master somehome right?19:28
← gusnan left19:28
→ Peste_Bubonica joined19:28
→ ColKurtz joined19:28
cbreak move it out of the way19:28
texasmynsted how?19:28
cbreak mv19:28
jcapinc ok, so how am I supposed to revert back to a commit temporarily in such a way that I can go back to the tip19:28
cbreak or mouse click & drag if you like19:28
texasmynsted lol19:28
← Codif left19:29
texasmynsted so I just mv that file and it should get replaced?19:29
cbreak jcapinc: git checkout -b <tmpbranchname> <sha>19:29
texasmynsted: I don't think it'd get replaced19:29
→ JohnBat26 joined19:29
← peppyheppy left19:29
cbreak but you would get a different error, maybe telling you what that blob was for19:29
← EugeneKay left19:29
cbreak which then allows you to easier get a working version19:30
texasmynsted can I see the content?19:30
← khmarbaise left19:30
texasmynsted When view it with vi, it looks binary19:30
cbreak it's compressed19:30
jcapinc thankyou, cbreak19:30
cbreak git cat-file -p shahere would show it19:30
but since it's corrupted, maybe not19:31
texasmynsted I did that but it is not something that makes sense19:31
← mishok13 left19:31
texasmynsted ok moved it and now git gc19:31
→ EugeneKay joined19:32
cbreak I would not gc19:32
you should do a git fsck19:32
texasmynsted lol, now and error that it can not be read19:32
I keep having stuff like this happen. How can so many files get corrupted all the time?19:32
← ToBeFree left19:33
→ todd_dsm joined19:33
← alex88 left19:33
→ khmarbaise joined19:33
texasmynsted hmm. Looks like the file was a "missing commit"19:33
← koo4 left19:34
texasmynsted this does not sound good, as I have two dangling commits, a dangling blob, a dangling tree and that missing commit.19:34
→ p3rror joined19:35
cbreak texasmynsted: dangling things don't matter much19:35
→ bitkiller joined19:35
texasmynsted ok good19:35
cbreak just mean some history rewriting happened or so19:35
← EtgarDizz left19:35
cbreak if you have a lot of corruption, maybe your HDD or SSD is deffective19:35
or a cabling/memory issue19:35
→ shreya joined19:35
texasmynsted ok19:35
cbreak my mac book pro is in repair since two weeks for a defective ssd :(19:35
texasmynsted yikes19:36
cbreak anyway, once you found the missing thing, think: do you have something locally that you want to keep? Unpushed commits or so?19:36
texasmynsted doing git status19:36
texasmynsted I should not have unpublished commits, I assume that means adds that are not committed.19:37
← ajpiano left19:37
texasmynsted I have some submodule adds, but I they were failing so need to re-do them anyway19:37
→ dinduks joined19:37
← fayimora left19:37
→ ajpiano_ joined19:37
ajpiano_ → ajpiano19:37
← fuzzybunny69y left19:38
texasmynsted This is a remotely mounted disk mounted of sshfs. I think that doing it this way is not ideal. I will switch to shelling in over ssh and then executing the git commands, but I would need to create a new key and can not do that at the moment19:39
I think the sshfs was causing the corruption perhaps.19:39
only changes to be committed are "modified: .gitmodules"19:39
should I commit that then do something?19:39
I think the change is from when I attempted to git submodule add19:40
cbreak texasmynsted: unpublished changes means things you have not PUSHED19:40
what you can do is try to git fetch from a remote to get the missing commit object19:41
→ mbutubuntu joined19:41
texasmynsted oh, sorry. I have been wondering if there is a way to see what has/has not been pushed but I do not know how to ask (or how to look it up)19:41
→ btanaka joined19:41
mbutubuntu hello folks, is there any way to automatically run some tasks when a user commits something to a remote server?19:41
cbreak just use log/diff with the remote tracking19:42
texasmynsted: what happens if you run git remote update?19:42
→ yaymukund joined19:42
→ koo4 joined19:42
texasmynsted 'git remote update' ?19:42
mbutubuntu *when a user commits and pushes19:42
raek mbutubuntu: yes, hooks in the remote repo19:43
→ littledwarf joined19:44
texasmynsted going to git remote -v update19:44
← littledwarf left19:44
← j0ran left19:45
mbutubuntu raek, sorry but I'm relatively new with git... any link/resource/doc about hooks?19:45
texasmynsted = [up to date] master -> origin/master19:45
← axl_ left19:46
← Jackneill left19:46
cbreak texasmynsted: I assume the commit is still missing19:46
texasmynsted <nod>19:46
→ axl_ joined19:46
cbreak you can try to git remote add it again under a new name19:46
raek mbutubuntu: this is one place to read about them: http://git-scm.com/book/en/Customizing-Git-Git-Hooks19:46
cbreak if you remote update after that, it might download the whole thing again19:46
mbutubuntu raek, great! thanks19:47
cbreak but to be honest, if I was in your situation, I'd just either clone fresh (if I wanted to solve the problem)19:47
or I'd go to the server repo, get the file, and scp it down manually19:47
texasmynsted I would like to go with plan b19:48
would that exact file name exist on the server repo?19:48
← mcluskydodallas left19:48
texasmynsted or perhaps I could clone fresh to a different location and copy the file from the fresh copy to the original?19:49
← iwoj left19:49
→ maletor joined19:50
← Octalot left19:50
→ Chinorro joined19:51
→ grantbdev joined19:51
← p3rror left19:52
cbreak texasmynsted: only if it has not been packed19:52
texasmynsted: but if it's packed, then there are ways to get to it anyway, like git cat-file or git unpack-objects19:53
rgenito after doing, "git svn fetch" on a local branch that is tracking a remote svn repository...how do i merge those changes into my local branch?19:53
cbreak I've not used the latter, but it should probably be able to do it19:53
texasmynsted It is packed19:54
→ beneggett joined19:54
← CannedCo_ left19:54
→ anjen joined19:55
texasmynsted so I would simply cd into git/objects/pack then git unpack-objects ?19:55
or rather git unpack-objects foo.pack?19:56
→ alester joined19:56
cbreak I've not used it19:57
→ fuzzybunny69y joined19:57
cbreak ... make backups before using it :)19:57
texasmynsted hmm. I did "git unpack-objects < pack-aabc708407e70a20bfefcd8ffd7718c170fdbf62.pack" but I do not know where it put the objects19:57
→ mishok13 joined19:57
→ blacktulip joined19:57
cbreak texasmynsted: it'll make them loose19:57
according to docs19:57
← __null left19:58
nbari how to force merge branch on another branch, I am doing merge branch but I just getting Already up-to-date. but I dont see the dirs19:58
texasmynsted heh19:58
cbreak so something called 14205faev19:58
→ Shusshu joined19:58
texasmynsted I do not know what that means really. I just want the file so I can copy it19:58
cbreak would be in 14/205fae and so on19:58
milki nbari: did you previously do a git revert?19:58
texasmynsted perhaps I need to git cat-file or something19:58
→ __null joined19:58
← tshauck left19:58
cbreak git cat-file would give you the contents, but I don't think that's as easy to use19:58
nbari yes19:59
→ ISF joined19:59
milki nbari: you cannot remerge something. this is in the warnings in the man git revert docs19:59
gitinfo nbari: the git-revert manpage is available at http://git-scm.com/docs/git-revert.html19:59
texasmynsted hmm.19:59
→ drizzd joined19:59
cbreak nbari: if you merged a branch, then it will stay merged for ever19:59
you can never merge it again19:59
because all the commits are already part of history19:59
← beneggett left20:00
nbari I have created another branch and tried to merge this branch and still dont get the files20:00
texasmynsted I am not sure what to try next20:00
milki did you rebase the branch or just make a new branch with the same history?20:00
← maletor left20:01
cbreak nbari: branches are identified by the commits they have20:01
→ mcluskydodallas joined20:01
cbreak just making a new branch has no purpose20:01
← dv310p3r left20:01
→ larpede joined20:01
cbreak in this context20:01
→ awx joined20:01
cbreak texasmynsted: if it worked then you're done20:01
nbari I created a new branch, lets say branchA an inside branhA I did a merge brachX, gut I can't et the files of branchX on bracchA20:01
cbreak if not, find out why not20:02
and if in doubt, just clone again :)20:02
texasmynsted no, I did it on a fresh clone so I could copy it over20:02
← hagna left20:02
→ tshauck joined20:02
milki no, you want to create a new version of branchX, rebase branchX on branchA, and then merge branchX into branchA20:02
nbari I was trying to found a better way to handle themes on a code, so I decided to create a branch per each theme leaving the core intact but I think is becomiing a mess20:03
milki or simply undo the initial merge with a reset20:03
and merge again20:03
texasmynsted cloning again would be a major problem in the current locaion20:03
nbari before I had evertying in a huge repositories20:03
cbreak nbari: you might think that branches are about files20:03
they are not20:03
they are about history20:03
if history has been merged, then that's it.20:03
revert can't unmerge it20:03
texasmynsted I just need a good copy of that corrupt commit20:03
nbari ok20:03
cbreak reset could nuke it away via history rewriting20:03
texasmynsted: yeah, what I wrote above should give you that20:03
(git unpack-objects)20:04
← g0bl1n left20:04
← anjen left20:04
texasmynsted ok will try in the original clone20:04
→ mangdood_ joined20:04
→ trend joined20:05
← magcius left20:05
→ iwoj joined20:05
→ magcius joined20:05
texasmynsted ok, so this is silly question but why would there be more than one pack file?20:05
trend I keep trying to add a folder, but every time I commit it thinks it is a binary file w/ 0 bytes. these folders are submodules fyi20:06
← nbari left20:06
← mcluskydodallas left20:06
→ cristian joined20:06
← stamina left20:06
texasmynsted ok the unpack, unpacked (used the latest pack file)20:06
so then try git fsck?20:07
to see if all good?20:07
mbutubuntu I'm reading this http://git-scm.com/book/en/Git-on-the-Server-Setting-Up-the-Server to set up a git server. I'd like to know why that page says to create a repo on the server in /opt/git/project.git. I thought that repositories were stored in the 'git' $HOME.20:07
← Ryback_ left20:09
→ p3rror joined20:09
texasmynsted hmm. still getting "bad sha1 file: .git/objects/a4/ff96a4868ceefc38a32d995bce2f9379b28361.old" which is the "renamed" version of the file20:10
and "missing commit a4ff96a4868ceefc38a32d995bce2f9379b28361" which is the orignal.20:10
← oscar_ left20:10
→ hakunin joined20:11
texasmynsted So I guess my only option is to blow away my clone and fetch a new clone?20:12
← erichynds left20:12
texasmynsted (the nuclear option)20:12
← eletuchy left20:12
→ jefftimesten joined20:12
yaymukund how would I show the diff for all the commits in `git log master..HEAD`20:13
→ Jippi joined20:15
thiago yaymukund: add -p20:16
→ aartist joined20:16
yaymukund git thiago it's like magic!20:17
thanks20:17
← shreya left20:17
→ lonewulftb joined20:17
← tshauck left20:18
mbutubuntu hey guys, I've added an origin called foo that is mapped to a ssh://git@host:port/path/to/repo ... I'd like to use SSH key instead of sending the user password... how to do this?20:19
← larpede left20:19
← JohnBat26 left20:20
→ francisfish joined20:21
← karsten left20:21
→ karsten joined20:21
→ mcluskydodallas joined20:21
→ theCole joined20:23
→ fayimora joined20:23
← mcluskydodallas left20:23
← marhaban left20:24
wereHamster mbutubuntu: set up ssh keys20:24
mbutubuntu wereHamster, I've just ssh keys because I use bitbucket20:25
wereHamster I can't parse that...20:25
mbutubuntu what I'd like to know is how to use the same ssh key for both git servers (bitbucket and the other one)20:25
← ProLoser|Work left20:25
← jefftimesten left20:26
wereHamster upload the public key to whichever server you want to access20:26
mbutubuntu wereHamster, you can't parse... ahahahah!!!! :-) sorry for my enlgish20:26
*english20:26
← francisfish left20:26
mbutubuntu wereHamster, uploaded and added to authorized_keys in the git $HOME/.ssh/20:26
I've done it yet20:27
← ronniy left20:27
wereHamster does 'ssh git@host' work?20:27
← bitkiller left20:27
mbutubuntu yes but I don't want to be able to log in with the git user20:28
→ oscar_ joined20:28
wereHamster I don't get your problem then.20:28
→ floppydisk joined20:29
mbutubuntu wereHamster, ssh git@host works but asks password... it doesn't use an ssh key20:29
floppydisk i dont have to checkout a branch to push to it right?¨20:29
wereHamster mbutubuntu: ssh -vv git@host, and see which key it's trying to use, and why it fails20:29
← cristian left20:29
← markgifford left20:31
mbutubuntu wereHamster, it tries to use the right key but it continues ask me the git@host password and not the key password...20:32
→ mcluskydodallas joined20:32
→ alberto56 joined20:32
wereHamster looks like you did not set up the key on the server correctly20:33
← ill1cit left20:33
← megalomania left20:33
← mcluskydodallas left20:34
mbutubuntu wereHamster, I've appended my public key to the git $HOME/.ssh/authorized_keys on the git server20:34
isn't it right?!20:34
← oojacoboo left20:35
← Pookz left20:35
mbutubuntu wereHamster, I've done what suggested on http://git-scm.com/book/en/Git-on-the-Server-Setting-Up-the-Server20:35
wereHamster are the permissions correct? Is anything useful in the server logs?20:35
→ markgifford joined20:35
→ daethorian joined20:35
→ oojacoboo joined20:36
← alberto56 left20:37
← mneorr_ left20:37
→ mneorr joined20:38
mbutubuntu wereHamster, sorry, maybe ssh -vv git@host doesn't use a key, watch this https://gist.github.com/a7557ed248b856060c2e20:39
→ dv310p3r joined20:39
→ mischat joined20:39
→ min|dvir|us joined20:39
min|dvir|us Hi.20:39
gitinfo min|dvir|us: welcome to #git, a place full of helpful gits. If you have a question, just ask it–somebody should answer shortly. For more information about git and this channel, see the links in the topic. It can't hurt to do a backup (type !backup for help) before trying advice you receive here.20:39
→ edude03_ joined20:39
wereHamster mbutubuntu: what makes you think it doesn't use the key?20:40
← lonewulftb left20:40
min|dvir|us I'm fairly new to git, so I apologize for my inability to understand/explain simple concepts.20:40
I did `git fetch bobby dan`.20:40
← edude03 left20:40
← kadoban left20:40
mbutubuntu wereHamster, I'm guessing... :-)20:40
min|dvir|us bobby is a remote, and dan is a branch on bobby.20:40
← toothr left20:40
→ dr_lepper joined20:40
min|dvir|us On bobby itself, I can see changes that I can't see on my local repo.20:40
→ krosenbl joined20:41
min|dvir|us Why?20:41
mbutubuntu wereHamster, so why it asks the git@host password and not the key password?20:41
← tewecske left20:41
wereHamster min|dvir|us: take a look at the output from that fetch command. What do you see and what does it tell you?20:41
mbutubuntu: probably because the server rejects the key20:41
→ tatsuya_o joined20:42
→ toothr joined20:42
← krosenbl_ left20:42
mbutubuntu wereHamster, mmhhh.... why would my server reject a key correctly added to .authorized_keys ?20:42
← mneorr left20:43
wereHamster do any other keys in that authorized_keys file work?20:44
min|dvir|us wereHamster: so, I'm using github.20:44
mbutubuntu wereHamster, there is only one key :-(20:44
→ barrbrain joined20:44
min|dvir|us wereHamster: I can see on github that the dan branch on the bobby remote has the data I want.20:44
wereHamster mbutubuntu: does that file have the correct permissions?20:44
min|dvir|us wereHamster: but my local repo does not.20:45
wereHamster min|dvir|us: that's not what I asked. take a look at the output from that fetch command. What do you see and what does it tell you?20:45
mbutubuntu wereHamster, owner git.git that has all read permissions20:45
← the_cybersphinx left20:45
wereHamster mbutubuntu: try chmod 060020:45
→ cybersphinx joined20:45
wereHamster mbutubuntu: and chmod 0700 $HOME/.ssh20:45
→ arooni-mobile joined20:45
← mischat left20:45
arooni-mobile is there a fast way to add a bunch of changes without doing one at a time?20:45
i know i could do git add *20:46
but i have some stuff i dont want to add too20:46
wereHamster arooni-mobile: git add -p ?20:46
mbutubuntu wereHamster, ooooh.... now it works... :-) it was a simple permissions problem!!! thank you very much20:46
:-)20:46
min|dvir|us wereHamster: I see a bunch of organized colors. It tells me words I am having difficulty understanding.20:46
gitinfo set mode: +v20:46
arooni-mobile wereHamster, so i would basically be asked for each file whether to add or not?20:46
wereHamster mbutubuntu: you would have probably seen that if you looked at the srever logs..20:47
mangdood_ ...I didn't even realize my messages weren't sending until now20:47
wereHamster min|dvir|us: maybe you could pastebin the output.. ?20:47
mbutubuntu wereHamster, sorry I'm late for an appointment... :-)20:47
→ Sigma joined20:47
← TheHaven left20:48
← hakunin left20:48
mangdood_ I want to patch a renamed file with changes from another commit, how can I do this?20:48
min|dvir|us wereHamster: http://dpaste.org/6fNIo/20:48
→ beneggett joined20:48
wereHamster min|dvir|us: like 6: From github.com:user/repo. Git fetched stuff from that repository.20:48
→ hakunin joined20:49
arooni-mobile how can i add all untracked files20:49
mbutubuntu wereHamster, this was why I didn't see the server log... sorry and thanks again20:49
→ trend_ joined20:49
← trend left20:49
arooni-mobile but still follow my .gitignore rules20:49
trend_ → trend20:49
wereHamster min|dvir|us: line 7: * branch dan -> FETCH_HEAD. It fetched a branch dan from that github repository, and stored it in the local ref FETCH_HEAD.20:49
← oscar_ left20:49
← hydrozen left20:49
wereHamster min|dvir|us: now you can look at FETCH_HEAD and there will be what you are looking for20:49
min|dvir|us wereHamster: how can I confirm that?20:49
What is FETCH_HEAD?20:49
wereHamster arooni-mobile: git add .20:50
min|dvir|us: a ref name20:50
min|dvir|us wereHamster: how do I look at refs? What is a ref?20:50
wereHamster min|dvir|us: !book20:50
gitinfo min|dvir|us: 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 !designers !gitt !vcbe and !parable20:50
→ the_cybersphinx joined20:51
→ g0bl1n joined20:52
← ben_alman left20:53
← cybersphinx left20:54
← Goplat left20:54
← freeed left20:54
← jcapinc left20:54
← booginga left20:55
→ Turicas joined20:55
→ RaptorX joined20:56
→ mcluskydodallas joined20:56
→ ilogger2 joined20:57
→ eletuchy joined20:57
→ booginga joined20:58
→ lindenle_ joined21:01
gitinfo set mode: +v21:01
lindenle_ HI can I reset a directory to match a different branch?21:01
→ ben_alman joined21:01
thiago git checkout thatbranch -- directory21:01
→ peppyheppy joined21:02
lindenle_ thiago: what about files that i have removed? How can I see which ones to delete?21:02
thiago lindenle_: if you have removed the files, they have been removed already21:03
lindenle_: you can't delete files that don't exist21:03
lindenle_: you probably meant something else, so please rephrase21:03
lindenle_ thiago: they exist in this branch, but not in the branch I want to reset this directory to match21:03
apok is there a way to undo a reset?21:04
thiago by the way, please don't say "reset" because the verb "to reset" has a special meaning in Git21:04
apok i can checkout the previous commit again, but them i'm in detached head state21:04
thiago lindenle_: when you do that checkout, it will remove the files that don't exist in the other branch21:04
lindenle_: but git status will tell you what it removed. You can undo the change for specific files.21:04
lindenle_ oh ok, cool that is what I want21:04
→ ipalaus_ joined21:04
→ ill1cit joined21:04
thiago apok: reset back to where you were before21:05
apok oh so just git reset --hard <commit-hash>21:05
thiago apok: yes21:05
→ linusoleander joined21:06
lindenle_ thiago: it did not delet the files when I did the checkout21:07
→ mritz joined21:07
→ mneorr joined21:08
→ Araxia_ joined21:09
→ Mowsh joined21:10
lindenle_ thiago: I can just delete them again if I can figure out which ones to remove21:10
→ stamina joined21:14
lindenle_ So... can i compare which files exist on two branches in an easy way?21:16
thiago git diff --name-status branch1 branch221:16
apok so i'm confused about cherry-pick, i just tried to cherry pick one commit from one branch to another but it included changes in previous commits as well…. is that how it's supposed to work?21:17
what if I only want the changes from one single commit?21:17
thiago apok: no, it's supposed to be one single commit21:17
apok hmmm guess i'm doing something wrong then21:17
my test is so simple i've done it twice now21:18
lindenle_ thiago: awesom thx21:18
thiago: awesom ethx21:18
→ littledwarf1 joined21:19
← littledwarf1 left21:19
apok crazy so when i do a diff of the cherry picked commit and it's previous commit i get one change, but when i cherry pick that same commit to another branch it includes that change plus the one from the previous commit, how is that possible?21:19
→ bitkiller joined21:20
→ dv310p3r joined21:22
← peppyheppy left21:22
→ harish joined21:22
mangdood_ How do I know what diff tool git is currently using?21:22
→ libertyprime joined21:24
← libertyprime left21:26
→ libertyprime joined21:27
← mneorr left21:28
→ peppyheppy joined21:29
→ CrazyTux[w] joined21:30
apok so does cherry-pick include all changes included in the diff between the cherry-picked commit and HEAD ?21:30
or is it supposed to be only changes from the commit itself (not other changes in the diff)21:30
CrazyTux[w] I've got a local branch and a remote branch and they both have slight changes, how can I merge them back into a single branch?21:30
apok because I thought it was the latter, but it's behaving like the former....21:30
→ kadoban joined21:32
→ johnanderson joined21:32
apok man that's kind of a bummer that it does a whole diff, i don't want all that other stuff, just the changes in that one commit >:/21:34
→ Schweby joined21:34
→ bremner_ joined21:35
→ n2diy joined21:35
→ simesy joined21:36
→ macmartine joined21:36
→ bglusman joined21:36
→ pro_metedor joined21:38
→ codeM0nK3Y joined21:38
→ pantsman joined21:40
← pantsman left21:40
→ pantsman joined21:40
← bglusman left21:42
mangdood_ :X Sorry for asking so many questions but is it a good idea to take some changes from another branch into the branch I'm on? What's the best way to do this?21:42
FauxFaux mangdood_: man git cherry-pick21:42
gitinfo mangdood_: the git-cherry-pick manpage is available at http://git-scm.com/docs/git-cherry-pick.html21:42
FauxFaux Took me far too long to type "man[tab]man".21:43
→ dougsland joined21:43
← dougsland left21:43
→ dougsland joined21:43
mangdood_ FauxFaux: I just looked up cherry-pick, though I'm not sure that's what I want. Is it for all of the changes from its entire history to that commit or...21:43
apok yeah i'm a little confused on that point too21:44
when i test it, it seems to be all of the changes, i was hoping it was limited to the changes in one commit21:44
FauxFaux mangdood_: No, it just takes the actual changes done in that commit. Like generating a patch for the commit, then applying it (but a little bit smarter (or stupider, depending on your luck)).21:44
apok though maybe i'm just doint it wrong somehow21:45
FauxFaux: that's not the behavior i'm seeing21:45
→ beneggett joined21:45
apok for me it includes all changes as a result of a diff between HEAD and the commit i'm trying to cherry pick21:45
mangdood_ FauxFaux: is there a way I can just get the patch from cherry pick without applying it to see what it looks like?21:45
→ vervic joined21:45
apok cherry-pick -n <commit>21:46
FauxFaux mangdood_: Why bother? Just do the cherry-pick and undo it if you don't like it.21:46
→ EugeneKay joined21:47
ChanServ set mode: +v21:48
→ w09x joined21:48
← w09x left21:48
→ w09x joined21:48
mangdood_ FauxFaux: This renaming is driving me mad; I'm trying to take a renamed file from another commit and take those changes, but it won't let me21:49
Should I just rename the file from the branch I'm on to match the one in the other commit? or is there a better way to do this21:49
→ arooni-mobile joined21:50
texasmynsted what is the best way to maintain a directory that looks like the git archive extraction from head?21:50
→ kinetic joined21:50
FauxFaux texasmynsted: !website21:50
gitinfo texasmynsted: [!deploy] Git is not a deployment tool, but you can build one around it for simple environments. Here is an example hook to get you started: http://goo.gl/l68hq21:50
→ HelloComputer15 joined21:51
→ adamm joined21:52
→ _W_ joined21:52
← HelloComputer15 left21:55
→ williamcotton joined21:55
→ phin joined21:56
→ lonewulftp joined21:56
→ warrick_ joined21:56
← phin left21:57
← warrick_ left21:57
→ uncleshebly joined21:57
texasmynsted That is close but really the idea would be to avoid replicating the .git directory on the destination server. It is not an asset but rather a liability.21:58
I would rather like to find a way to see the head without the .git director then replicate that to the server.21:59
→ pheaver joined21:59
texasmynsted I searched but did not see a fuse filesystem for git. That might be ideal.21:59
EugeneKay That's not really what git is for.22:00
I think ! deploy, as above, is close to what you want22:00
→ b1rkh0ff joined22:00
← Schweby left22:01
→ madewokherd joined22:01
texasmynsted that is creating two repositories on the server. One is the production public web directory with .git and the other is the working git, if I read it correctly22:02
← dougsland left22:03
apok FauxFaux, lindenle_: if behavior of cherry pick is like you said then how is this possible? https://gist.github.com/283921122:03
← linusoleander left22:03
texasmynsted I suppose that I should just write a script to git archive to a directory and manage a symlink to the latest expanded archive …22:04
apok you can see i am trying to cherry pick the latest commit from branch-1 onto master, but it tries to apply changes from all commits in branch-1 since I branched from master22:04
texasmynsted :-)22:04
→ HelloComputer15 joined22:04
texasmynsted hmm. this could work22:04
→ prudnikov joined22:04
→ cagedwisdom joined22:05
→ yoklov joined22:05
→ TeckniX joined22:06
← madewokherd left22:06
→ madewokherd joined22:06
← TeckniX left22:06
← Araxia_ left22:10
→ Milossh joined22:10
→ Araxia_ joined22:11
← libertyprime left22:11
→ libertyprime joined22:12
← dv310p3r left22:12
→ kcj joined22:12
→ marhaban joined22:13
→ littledwarf joined22:13
← pantsman left22:14
→ DofLeFou joined22:15
→ Maccer joined22:17
→ rsumi joined22:17
← littledwarf left22:17
→ dougsland joined22:19
rsumi hello, i'm having trouble with this help of converting svn to git... http://help.github.com/import-from-subversion/... when i do "git push origin master" i get this error " ! [rejected] master -> master (non-fast-forward) error: failed to push some refs to 'git@github.com:rsumiim/mytest.git'". how do i fix this?22:19
→ _root_ joined22:19
← peppyheppy left22:19
rsumi i'm doing to git-svn instructions22:20
cbreak you already have remote history22:20
← libertyprime left22:20
rsumi cbreak: i'm new to git. what is that trying to tell me?22:20
← arooni-mobile left22:20
cbreak there are commits that someone else pushed into that repository22:20
→ tyfighter joined22:20
cbreak you can not push divergent history22:20
either merge or rebase before pushing22:21
but if you convert a fresh svn repository22:21
you should push to a fresh git repository too22:21
→ libertyprime joined22:21
rsumi cbreak: it's an old svn repo to a brand new git repo. there are no changes on the git end22:21
← prudnikov left22:21
rsumi cbreak: well, it created a README.md file on github when creating the repo. is that the problem?22:22
cbreak there is git history22:22
that's why the push failed22:23
it is22:23
→ g0bl1n joined22:23
← g0bl1n left22:23
→ g0bl1n joined22:23
rsumi cbreak: if i do a "git merge master" right before the command it says its already up to date22:24
then i retry and the same error22:24
cbreak of course...22:24
you are probably on master22:24
merging with the current branch doesn't do anything22:24
if you want to kill remote history, just git push -f22:24
texasmynsted Where can I find more information about chains of remotes. example Repository A is a clone of B which is a clone of C, where A has everything that B has but also manages a few extra things that B does not and B has everything that C has but manages a few extra things than C.22:26
does that make sense?22:26
→ srtim joined22:28
← HelloComputer15 left22:29
rsumi cbreak: ok great that worked but my authors are still what they were in svn and not what i did with "git svn --authors-file=path/to/authors_file clone SVN_REPO_URL LOCAL_DIR". am i missing something here also?22:30
cbreak texasmynsted: it doesn't matter to git22:30
rsumi: maybe. I never used that22:31
→ hedgomatic joined22:36
→ yizumi joined22:37
→ arooni-mobile joined22:38
→ fayimora_ joined22:38
← libertyprime left22:40
→ libertyprime joined22:40
← ipalaus_ left22:42
fayimora_ → fayimora22:42
→ alberto56 joined22:43
← g0bl1n left22:43
→ leavittx joined22:44
→ SegFaultAX|work2 joined22:45
← srtim left22:47
→ pl0sh joined22:49
← booginga left22:50
← alberto56 left22:52
← johnanderson left22:57
→ w09x_ joined22:58
→ hyperair joined23:00
← w09x left23:01
← pheaver left23:03
→ fuzzybunny69y1 joined23:05
← DofLeFou left23:05
fuzzybunny69y1 → fuzzybunny69y23:05
→ ISF joined23:06
→ john_anderson joined23:07
→ alberto56 joined23:08
← harish left23:08
→ btree_ joined23:08
← btree_ left23:08
→ btree_ joined23:08
→ w09x joined23:09
→ Praise joined23:09
texasmynsted cbreak, ok23:11
← alberto56 left23:12
→ palyboy joined23:12
→ dv310p3r joined23:12
← w09x_ left23:13
→ hadees joined23:15
→ masonkramer joined23:17
→ solussd joined23:19
← ISF left23:20
→ nhooey|mini joined23:20
nhooey|mini You can use `git log --grep` to search log messages, and `git log -S<string>` to search code, but is there a way to search the diffstat for a file name pattern with `git log`?23:21
← hedgomatic left23:23
→ Chillance joined23:24
← macmartine left23:24
← beneggett left23:25
← john_anderson left23:25
→ optixx joined23:25
→ Vert joined23:25
← dv310p3r left23:26
→ peppyheppy joined23:26
→ john_anderson joined23:27
← solussd left23:28
→ w09x_ joined23:28
→ dv310p3r joined23:28
→ Goplat joined23:30
← w09x left23:31
→ alberto56 joined23:34
→ mdhopkins joined23:35
→ davidbanham joined23:36
→ cenuij joined23:36
← cenuij left23:36
→ cenuij joined23:36
← simesy left23:38
← dv310p3r left23:39
→ dv310p3r joined23:43
→ solussd joined23:43
→ w09x joined23:44
→ acrocity joined23:44
← w09x_ left23:47
→ chayin joined23:47
→ sdeboey joined23:47
← sdeboey left23:48
→ stkrzysiak joined23:51
→ chrisgeorge joined23:54
← solussd left23:55
→ mikepack joined23:56
← tyfighter left23:59
→ mtytel joined23:59
← peppyheppy left23:59

Logs Search ←Prev date Next date→ Channels Documentation