IRCloggy #git 2012-01-12

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.

2012-01-12

Harzilein cmn: say you use the cmd basic command to set output to rs232. now you use socat to connect that tty and your pty. you can choose all the stty settings. none will give you cr to lf conversion.00:00
s/that tty/that serial port/00:00
cmn I see, so when you connect from a unix machine to a c64, you only get one line that gets overwritten?00:01
Harzilein yes00:02
(of course you can work around that by adding them in your transfer software, but i wanted to avoid that. the idea is that i want to make a c64 rom that starts with i/o redirected to rs232)00:03
→ HardFu joined00:03
← steffo left00:03
← Laibsch left00:03
← HardFu left00:05
← ramsey left00:05
← NetHawk left00:05
← Dunhamzz left00:05
← sureaint left00:06
→ cakehero joined00:06
→ NetHawk joined00:06
→ subbyyy joined00:06
← jjido left00:07
→ KingPin joined00:07
Harzilein it kind of makes sense to have cr in microcontrollers00:08
→ Morphous_ joined00:09
→ TheFuzzball_iOS joined00:09
cmn more than LF?00:10
or just because it's 1 byte instead of 2?00:10
Harzilein a tiny bit more than lf, yes.00:10
→ sattu94__ joined00:11
← DofLeFou left00:11
Harzilein say you want to have a protocol to communicate between two ucs: if you want to send stuff to a buffer then have it execute stuff on it after the cr is detected, then _return_ to the start of the buffer %-)00:12
← johnanderson left00:12
Harzilein -if00:12
cmn so you're sending just one line at a time00:14
← sattu94 left00:14
→ proq`` joined00:15
cmn otherwise I don't see how going to the start of the buffer makes sense00:16
then again, the whole µcontroller thing confuses me00:16
→ michaelh joined00:17
→ Tabmow joined00:18
← fr0sty left00:18
← nrichards left00:19
→ fr0sty joined00:19
→ scelis joined00:19
← proq`` left00:20
← otakutomo left00:22
→ y3di joined00:23
← fr0sty left00:23
y3di how do i remove all the files ive currently added with git add from the commit queue00:23
canton7 y3di, git status will tell you00:24
← baburdick left00:24
offby1 Dear lazyfreenode: is there a tool which will search commits in a range (e.g., origin/master...HEAD) that add the string TODO, and then display them _with context_? "context" means the file name and line number where the TODO is found. Note that the various stuff that http://www.google.com/search?q=git+todo shows me aren't quite it.00:24
if not, I gotta write it myself00:24
→ bigkm joined00:25
offby1 The use case is: I've been hacking for a while, and writing TODO notes to myself; now I want to review them all. And I am not interested in TODOs that were already present when I started hacking.00:25
bremner did you try some combination of options to git-log?00:26
offby1 Well, I stared at the man page for it for a while :)00:26
I suspect I'm gonna have to parse the "patch" output.00:26
y3di ugh, thanks canton7. sorry about that00:26
offby1 gonna need a bigger boat00:26
→ baburdick joined00:26
→ fr0sty joined00:26
offby1 bremner: see, "git log -STODO origin/master..." gets me about 50% of the way there00:26
bremner and then --name-status or so?00:27
offby1 ideally I'd have something whereby I could do M-x git-todo RET in emacs and then hit C-x ` and knock them off one by one :)00:27
→ rchavik joined00:27
← rchavik left00:27
→ rchavik joined00:27
→ pLameN_4o joined00:27
← ChrisAnn left00:27
bremner or --name-only, but that will have false-positives00:28
← neurodrone left00:28
offby1 yeah, --name-status or something like it is probably essential ... but as I say, I suspect I'm still gonna have to parse the patch.00:29
oh well00:29
← pLameN_4o left00:29
→ centipedefarmer_ joined00:29
→ pLameN_4o joined00:30
← pLameN_4o left00:30
← Tabmow left00:30
← bronson left00:30
→ Tabmow joined00:31
scelis canton7: 'git reset HEAD' is a quick way to completely reset the index00:31
sorry, y3di00:32
sitaram offby1: when I need this sort of thing I just go "git log -p A...B" and handle the search for TODO in the pager00:35
→ rtconner_ joined00:35
sitaram offby1: and you surely didn't mean three dots in your original question did you?00:35
offby1 I think I did, but I get that wrong sometimes00:36
ah, it makes no sense for "git log"00:36
sitaram well it would make a difference to my method; you have to use the opposite00:36
two dots00:36
offby1 as usual ... *mumble*00:36
we Emacs weenies like to be able to hit C-x ` without thinking and be presented with the next line to hack on :)00:37
sitaram offby1: although the page was written loooong ago for a differnt purpose, it might serve as a good mnemonic: http://sitaramc.github.com/other-stuff/eg.html#other_stuff_eg_eg_diff__00:37
offby1 unvestigates00:37
← scelis left00:38
← rtconner left00:38
→ csgeek joined00:38
offby1 sitaram: hmm, that just might stick :)00:39
mdel__ → mdel00:39
tizzo → tizzo-afk00:39
offby1 man, coming up with suitable mnemonics is really what it's all about, isn't it?00:39
aides memoires00:39
or "aides-mémoire", as Google Translate tells me00:39
sitaram I have a good memory. When I remember...00:39
offby1 ♩♬I forgot to remember to forget♫♪00:40
← joshteam left00:40
→ kpreid joined00:41
Mikachu 'good memory, but short'00:41
← sudoquai_ left00:42
← sudoquai__ left00:42
← sudoquai left00:42
← sudoq358 left00:42
→ Dragooon joined00:42
Dragooon Hola00:42
← poppiez left00:42
→ jmah joined00:42
→ jryans_ joined00:44
← jryans_ left00:44
→ secoif joined00:44
→ ISF_ joined00:44
← chris- left00:45
← jryans left00:46
← centipedefarmer_ left00:46
← ipalaus_ left00:49
→ ipalaus_ joined00:50
← gfto left00:51
← sattu94__ left00:51
tizzo-afk → tizzo00:51
← yshh left00:53
→ Targen joined00:54
→ bronson joined00:54
→ yshh joined00:54
← baburdick left00:55
→ baburdick joined00:56
← brian_g left00:56
← Karmaon left00:57
tizzo → tizzo-afk00:57
← canton7 left00:57
→ Karmaon joined00:58
tizzo-afk → tizzo00:58
← yshh left00:59
→ pfloyd joined00:59
← baburdick left00:59
pfloyd Where can I find a list of all the git debug environment variables I can set? I'm trying to figure out why a clone goes fast up to 60%, then slows down a lot. git clone -v doesn't add any verbosity at that particular point00:59
→ otakutomo joined01:00
FauxFaux man git , but it probably won't help. The problem is probably that the percentage bar is in objects, not in bytes, so if you suddenly have a 500gb object it'll just hang waiting for the network.01:00
gitinfo the git manpage is available at http://jk.gs/git.html01:00
→ valtih joined01:01
→ neurodrone joined01:01
← brainbox1100 left01:01
→ Kocho joined01:02
← Vampire0 left01:03
tizzo → tizzo-afk01:03
diverdude i wonder if gitlab and gitorious can co-exist01:03
← secoif left01:03
Karmaon → Karmoan01:04
→ sattu94__ joined01:04
Harzilein diverdude: that's not an all or nothing thing i think01:04
diverdude: gitorious _can_ be excluded from certain communities if people do not write vendor agnostic/pluggable api consumers to integrate their stuff with01:05
bremner wut?01:05
pfloyd FauxFaux: ok, that was my suspicion, just weird that it is fast up to 60-ish%, then the last 40 is consistently slow. Given the non-uniformity of the repo in question, it seems strange so I was hoping I could see which objects it was dealing with when it slows down01:05
→ cccaldas joined01:05
→ akosikeno joined01:06
FauxFaux pfloyd: Nah, it sorts them so big things are at the end. I don't know why.01:06
pfloyd ahh! that makes sense, thanks :)01:06
FauxFaux Or, at least, it seems to.01:06
→ HardFu joined01:06
sitaram diverdude: I doubt it. Both will at least want to take over your ssh authkeys file01:06
pfloyd but there is no way to see which objects it is working on while it's cloning?01:07
→ Vampire0 joined01:07
→ alexim joined01:07
diverdude sitaram, ahh right01:07
sitaram, i would a user for each then i guess01:07
sitaram and both want to take over the update hook also, I suspect01:07
diverdude: not enough pain in your life yet?01:07
Karmoan → Karmaon01:07
sitaram aah to be young and masochistic!01:08
Harzilein oh *blush*01:08
→ xiangfu joined01:08
diverdude hahahaaha01:08
→ joshsmith joined01:08
Harzilein i thought this was about some ecosystem-like way of interpreting "coexisting"01:08
diverdude sitaram, believe me...gitorious has caused me SO much pain to install...and aint even done yet01:08
← Textmode left01:08
→ brian_g joined01:09
→ mgpcoe1 joined01:09
tizzo-afk → tizzo01:10
Kocho hey guys, I'm not sure if git is for me (I just found out about it today). I need to make some changes to my website and I'd like to test them to make sure it all works, but I don't want the public to see the changes until I'm ready to publish them. Do I just need a testing server?01:10
→ zm joined01:10
sitaram diverdude: one of the stupidest things I did was the *name* of gitolite. I started with "gitosis-lite" since I really had no intentions of taking over the spot, then shortened it to gitolite. And now with gito(sis|rious|lite|God-alone-knows-what-else) people are getting confused01:10
FauxFaux My current project is called gitopenis.01:11
bremner gitolife.01:11
sitaram git for open Information Systems?01:11
diverdude sitaram, understandable...i was also very very confused01:11
FauxFaux sitaram: Good choice.01:11
sitaram I should have called it something very very obscure. I always wonder how they come up with names like k3b, kudzu, etc01:11
diverdude sitaram, i am starting to learn now what is what...but wont say im like with a full overview01:12
sitaram diverdude: you need to get the gitlab guys to clean up their install instructions. Too many people asking for help... they should fix it or come to the channel01:12
← rjcks left01:12
← rtconner_ left01:12
← mgpcoe left01:12
→ FxMasta joined01:13
FauxFaux got spam from http://bamboowang.com/ earlier, which appears to actually be a legitimate company.01:13
diverdude sitaram, yes...gitlab install instructions are way to "lazy" :(01:13
sitaram, its a showstopper for that frontend i think01:13
sitaram, it needs to be much more accurate and descriptive if people are to use it01:14
← FxMasta left01:14
sitaram FauxFaux: my definition of "legitimate" does not include "sends spam" but that may be just me ;-)01:14
→ lacrymology joined01:14
tizzo → tizzo-afk01:14
← jaequery left01:14
sitaram FauxFaux: at least the owner cannot be a legitimate child ;-)01:14
diverdude sitaram, and its a pitty because it looks like a nice piece of software01:15
← Kocho left01:15
sitaram diverdude: well what are you doing to fix it? did you even report it? join their forums/mailing list/whatever?01:15
→ hyperair joined01:15
→ rtconner_ joined01:15
← rtconner_ left01:15
← cakehero left01:15
← HardFu left01:18
← lacrymology left01:18
← VladGh left01:19
← robotmay left01:19
→ lacrymology joined01:20
← valtih left01:22
→ jameslordhz joined01:24
← lacrymology left01:25
← ScottE left01:25
← ehsan left01:25
→ Tazar joined01:28
→ amigojapan joined01:29
← lonewulftp left01:33
→ secoif joined01:33
← akosikeno left01:33
→ forsto joined01:34
← steveoliver left01:37
← brian_g left01:38
→ dwmw2_gone joined01:39
← TheFuzzball_iOS left01:40
← kenichi left01:42
← mandric left01:42
← BiggFREE left01:42
← v0n left01:45
→ patrick99e99 joined01:46
patrick99e99 I accidentally just typed "git branch -t origin/foobar" and it says "Branch origin/foobar set up to track local branch master01:47
← friskd left01:47
patrick99e99 how can I undo that?01:47
← hyperair left01:47
→ zdk joined01:49
← abetaha left01:49
→ abetaha joined01:50
nevyn git branch -d origin/foobar01:50
will delete the local01:50
← acrocity left01:50
nevyn and you should be able to reference remotes/origin/foobar again01:50
maybe a -D01:50
→ acrocity joined01:51
← nadirvardar left01:52
← Gizmokid2005 left01:52
→ kaushik_ joined01:54
← abetaha left01:54
→ lekefly_ joined01:57
→ rjcks joined01:57
← LekeFly left01:58
← Bombe left01:58
← alexim left01:58
← linusoleander left01:59
→ dzonder joined02:00
→ Bombe joined02:01
← happyface left02:02
→ happyface joined02:02
← chadi left02:03
← msponge left02:03
← kaushik_ left02:03
→ demi` joined02:04
← Bombe left02:06
→ chadi joined02:06
→ ivan joined02:07
→ Gizmokid2005 joined02:07
← quazimodo left02:07
→ ivan` joined02:07
→ ramsey joined02:07
← Sigma left02:08
tizzo-afk → tizzo02:10
→ lonewulftp joined02:11
→ Crewe joined02:11
← iocor left02:11
tizzo → tizzo-afk02:11
← ypcat_ left02:12
→ ypcat joined02:12
→ WAAJ28K joined02:14
→ spearce_ joined02:15
← dzonder left02:16
→ dorisabayon joined02:17
← spearce left02:18
spearce_ → spearce02:18
← tystr left02:18
← Turicas left02:19
← spearce left02:19
← Fandekasp left02:20
← hwrdprkns left02:24
→ IsmAvatar joined02:24
→ Laibsch joined02:26
→ abetaha joined02:27
→ cgd joined02:27
← kenperkins left02:28
← mdel left02:28
IsmAvatar would this be a bad place to ask simple questions about jgit?02:30
→ Bombe joined02:32
→ centipedefarmer_ joined02:32
→ Heimidal joined02:33
→ xarthna joined02:33
← j0ran left02:34
offby1 probably not _too_ bad.02:34
Perhaps the questions will turn out to actually be about git.02:34
← detaer left02:35
← xarthna left02:35
← Laibsch left02:36
← Bombe left02:36
→ j0ran joined02:39
gitinfo set mode: +v02:40
← eletuchy left02:40
→ Kronuz joined02:40
cgd is it possible to get rid of commits in a remote after i've pushed them, and just keep the one i rebased to02:40
→ Bombe joined02:40
← Targen left02:40
offby1 sue02:40
sure02:40
cgd: you can "git push -f". There are all sorts of caveats, though.02:40
!rewrite02:40
offby1 will never learn the bot's secret keywords02:41
cgd yeah i did end up doing that02:41
i have a clone of that repo though that i need to fix02:41
→ mdel joined02:41
← mdel left02:41
cgd git pull --rebase looked promising but i dunno :<02:42
← tsujiban left02:43
offby1 I think that'll work fine.02:43
The caveat is: _everyone_ who has a clone will need to "git pull --rebase" once, lest they suffer cryptic merge conflicts.02:43
That's called "rewriting history", and is useful on occasion, but a pain in the ass.02:43
offby1 stirs dinner02:43
← bitslip left02:45
cgd aye, luckily i'm the only one who has to deal with that headache, no one else has a clone02:45
→ Vampire0_ joined02:45
offby1 then "git push -f" is fine.02:45
IsmAvatar offby1: nevermind then, lol. It's mostly just API questions. I'll just use their forums or whatnot02:45
offby1 s/fine/just what the doctor ordered/02:45
IsmAvatar thanks anyways02:45
← IsmAvatar left02:46
← mando left02:48
→ fisted_ joined02:49
← Vampire0 left02:49
← amigojapan left02:51
← fisted left02:51
→ bitslip joined02:51
← centipedefarmer_ left02:52
→ Behi joined02:54
→ zancas joined02:54
→ SiCKa joined02:55
sitaram offby1: !trigger or !shortcuts02:56
gitinfo offby1: [!triggers] See http://jk.gs/git/bot/trigger.php02:56
← zancas left02:56
EugeneKay .trigger_edit rewrite @!rewriting_public_history02:57
gitinfo EugeneKay: okay.02:57
EugeneKay offby1: !rewrite02:57
gitinfo offby1: [!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)02:57
→ diegoviola joined02:57
EugeneKay ;-)02:57
← Destos left02:57
→ kaushik_ joined02:58
← ramsey left02:59
offby1 sitaram: yeah, but ... I have to then put the web page under my pillow or something. I want to just _know_ them without having to _learn_ them, y'know?02:59
EugeneKay: heh. Thanks for caching that :)02:59
EugeneKay offby1 - I figure it's a decent alias to have around.02:59
sitaram offby1: lazy much? ;-)03:00
offby1 sitaram: I'd tell you but then I'd have to type03:00
← Dragooon left03:01
→ Dragooon joined03:04
→ dys` joined03:04
EugeneKay offby1 - http://www.nuance.com/dragon/03:04
Just babble in the general direction of your computer03:04
→ horieyui joined03:05
← dys left03:05
← icrazyhack left03:06
→ baburdick joined03:08
← Gizmokid2005 left03:08
→ brngardner joined03:09
→ Gizmokid2005 joined03:10
→ dfr|mac joined03:13
← brngardner left03:14
→ dv310p3r joined03:15
→ nadavoid joined03:15
→ eggman2001 joined03:17
← Kronuz left03:18
gitinfo set mode: +v03:18
eggman2001 hello, if I have deleted files under "changes not staged for commit" what should I do with those?03:18
chrisf eggman2001: git rm03:19
→ jryans joined03:19
chrisf eggman2001: or git add -A will pick them up too, in a slightly less fail-prone way03:19
→ krz joined03:20
eggman2001 chrisf: that's what I thought but I'm getting a "fatal: Unable to create ../.git/index.lock: File exists"03:20
← watsonian left03:20
eggman2001 getting the same message with git add -A03:20
chrisf eggman2001: long-running command in progress, or has it been left behind by some previous explosion?03:21
← kukks left03:21
← TomyLobo left03:21
eggman2001 it hasn't been left behind by a previous explosion since I just deleted these files, like a minute ago03:22
← michaelh left03:22
← diogogmt left03:22
offby1 EugeneKay: I was thinking "Siri" when I said that.03:22
chrisf eggman2001: remove it03:22
eggman2001 chrisf: the files are gone though. I already deleted them03:23
chrisf ~ remove .git/index.lock03:23
it shouldnt get left behind by any normal operation though03:24
eggman2001 weird. i can't remove the .lock files either03:26
i simply removed these files using rm03:27
→ remyo joined03:31
← boombatower left03:31
← Araxia_ left03:32
→ boombatower joined03:33
→ nathelm joined03:34
← bitslip left03:34
← flaguy48 left03:35
→ bitslip joined03:35
← boombatower left03:36
→ boombatower joined03:37
offby1 I often have long rebases fail with a complaint about the lock file.03:37
→ heidar joined03:37
offby1 I just "git rebase --continue", and cross my fingers03:37
← orafu left03:37
→ orafu joined03:38
← boombatower left03:38
→ ISF__ joined03:38
← ISF_ left03:38
→ flaguy48 joined03:38
→ boombatower joined03:39
eggman2001 dang, still getting that .lock file and I can't delete it03:39
← boombatower left03:39
eggman2001 because it apparently doesn't exists03:39
← otakutomo left03:40
← jlebar left03:40
→ mdel joined03:40
→ oriba_ joined03:42
offby1 creepy03:44
are you sure there's no git process already running?03:44
← oriba left03:44
eggman2001 i think it might have been because i ran git init twice03:44
i'm a little tipsy03:44
i just remove .git and started over03:44
→ milki joined03:45
← shiba_yu36 left03:45
→ shiba_yu36 joined03:46
← shiba_yu36 left03:46
→ chayin joined03:46
→ bindaasomatic joined03:47
← kaushik_ left03:47
→ mandric joined03:51
→ robde joined03:51
← Nibato left03:51
→ Nibato joined03:52
← secoif left03:53
AlexMax Question. If someone git-svn's their svn repo, how do you push everything to a remote repository?03:55
i.e. if you want to do a one time conversion03:55
→ flip_digits joined03:56
→ digid joined03:57
← jmah left03:57
→ kerrick joined03:58
→ boombatower joined03:58
→ przemoc joined03:59
→ Kiall joined03:59
→ stringoO joined04:00
← Swathe left04:00
← dotpot left04:01
← OOPMan left04:03
→ gavin_huang joined04:05
→ f0i joined04:05
offby1 perhaps "git push --mirror" ... but beware, if the target repository has any branches, this will delete them04:05
→ mindlogger joined04:06
mindlogger how can I go one commit back?04:06
AlexMax Nevermind, I found svn2git04:06
mindlogger I screwed my code :(04:07
→ tellone joined04:08
offby1 mindlogger: "git reset --hard HEAD^" is probably what you want04:08
→ nOStahl joined04:09
mindlogger oh yeah I just remembered, thanks it worked04:09
HEAD^^ is two commits back right?04:09
← kermit left04:09
nOStahl hi all, I am new to git and version control stuff all over.04:11
→ friskd joined04:11
→ kermit joined04:11
→ detaer joined04:18
offby1 mindlogger: yeah I think so.04:18
hi nOStahl! Keep coming back.04:18
eggman2001 hello, how do I find the path to a local git repo?04:20
← Anxi80 left04:20
← tango_ left04:22
→ aspotashev joined04:23
← mandric left04:23
← Behi left04:23
offby1 git rev-parse --show-toplevel04:23
also git rev-parse --cdup04:24
→ pabs3 joined04:24
→ beautifulmind joined04:24
→ OOPMan joined04:24
← koo3 left04:24
eggman2001 offby1: I'm a bit confused. lets say I have the path, home/user/project/.git. would that be the path to the git repo?04:25
← lonewulftp left04:25
pabs3 hmm, would dirname run from git filter-branch --tree-filter return . instead of the correct directory?04:26
s/would/why would/.04:26
→ koo4 joined04:27
→ macRover joined04:28
← Crewe left04:29
← detaer left04:29
→ Swathe joined04:29
← flip_digits left04:30
fr0sty eggman2001: what is the cwd in your hypothetical situation?04:30
← Swathe left04:30
← boombatower left04:31
→ Swathe joined04:32
→ secoif joined04:32
→ brainbox1100 joined04:33
offby1 eggman2001: I'd call that the "git dir"04:33
the repository would be its parent.04:33
I think.04:33
← rpg left04:34
→ Prasad joined04:35
→ sdtr443w joined04:35
← dreiss left04:36
ChanServ set mode: +v04:37
← oriba_ left04:37
sdtr443w My remote repository got corrupted, but I have it cloned locally and I'm the only person developing on the project. Can I just copy my .git in place of the corrupt one and take it over?04:38
← thiago left04:38
sdtr443w I got some grief trying to do it so I figured there was more to it.04:38
offby1 well, your local .git will differ from the remote one, in the names of the refs.04:39
but the objects should be fine.04:39
← josefig left04:39
offby1 I'd move the remote one aside, then create a fresh one with the original name, then "git push" various refs from your local one to it.04:40
sdtr443w I have some commits in the local one. If I copy the remote with it, does that basically make them look like they were pushed?04:40
hmm04:40
what would be the "various" in "various refs?"04:40
← mindlogger left04:42
sdtr443w Well OTOH I could try to repair the remote but I don't know what to really do there.04:43
I just booted into Windows one time and it started checking for inconsistencies. I think it went and nibbled on something. Even the local had some goofs, but they all were fixed with rehashing.04:43
offby1 don't understand your first question.04:44
"various" means "whichever ones you want"04:44
sdtr443w oh04:44
→ igotnolegs joined04:44
sdtr443w If I pushed would that push the whole history into it or just the files as they are?04:45
I mean specifically if I was creating a fresh one and pushing into it04:45
offby1 when you push, you push refs.04:46
Each ref points to a single commit.04:46
Each commit takes with it all its history.04:46
you need to learn some git basics.04:47
→ kaushik_ joined04:49
→ ericmuyser joined04:49
sdtr443w yeah you can say this here is the real crash course. I haven't had to bother too much with the architecture of it while it was working fine for single-user stuff.04:50
Since I've booted up tonight to look at this, I am finding it to be more compliant.04:50
I had shared the repo over the network so I could get it down on my laptop, and I think that share process did something goofy.04:50
At this point it looks like it's just complaining about some merge conflicts.04:51
offby1 I am never comfortable with using network file systems.04:51
Windows' is probably the most reliable one out there, and I don't trust _it_04:51
sdtr443w Yeah I saw fun stuff online about ZFS and friends. Can't be coincidence.04:51
→ DofLeFou joined04:52
→ VladGh joined04:54
← VladGh left04:54
→ VladGh joined04:58
sdtr443w Looks like in all the shuffling the remote claims to have some files newer than mine, but I want it to take my versions wholesale.05:00
If I'm reading this right, I do "git checkout --ours [files I want to plow over]" then "git add -u" then "git commit" Should that acknowledge and take care of the conflicts?05:01
← eggman2001 left05:01
→ caseymcg joined05:02
→ Dave^| joined05:03
← beautifulmind left05:03
← ISF__ left05:04
← tewecske left05:04
diegoviola hey guys how do i make a pull request in command line05:04
sdtr443w git pull05:05
→ ISF joined05:05
← mgpcoe1 left05:05
← berserkr left05:06
→ dfr|mac_ joined05:07
→ nikuyoshi joined05:08
sdtr443w in more newbie goodness, I see I am a bad, bad man for doing all my stuff with master all the time w/o branching05:08
← dfr|mac left05:10
offby1 diegoviola: I don't think "git pull" is a pull request.05:11
← digid left05:11
offby1 diegoviola: I think it's git request-pull05:11
← cilly left05:12
diegoviola offby1: ty05:12
← slonopotamus left05:13
→ josefig joined05:13
→ cilly joined05:14
→ mandric joined05:15
→ friskd_ joined05:16
← friskd left05:16
friskd_ → friskd05:16
→ dreiss joined05:16
→ Fandekasp joined05:17
← josefig left05:17
← gavin_huang left05:17
← macRover left05:19
→ slonopotamus joined05:19
← neurodrone left05:20
← Prasad left05:20
← dorisabayon left05:21
→ Laibsch joined05:22
← nadavoid left05:22
→ sambar joined05:23
sdtr443w So it looks like I got the remote reconciled from a copy, except I had to set denyCurrentBranch to ignore.05:24
→ otakutomo joined05:24
← Laibsch left05:24
→ petisnnake joined05:24
petisnnake Hi. How can I delete a file on a remote repo?05:25
sdtr443w git rm and commit it?05:25
← sambar left05:25
← nathelm left05:26
petisnnake tried that, but doesn't seem to remove file from github05:26
→ lonewulftp joined05:27
← nOStahl left05:27
offby1 petisnnake: my hunch is that you deleted it in branch A, but are looking at branch B in your browser05:28
either that or you forgot to push05:28
← cgd left05:28
offby1 if the latter, then I accuse you of being used to Subversion :)05:28
← OOPMan left05:28
← aspotashev left05:28
→ pidus joined05:29
petisnnake pushed, forgot to commit :D05:29
thanks05:29
EugeneKay git dowhatimeannotwhatisay05:32
← kermit left05:33
→ Prasad joined05:34
→ psakrii joined05:35
→ AndChat| joined05:36
← lonewulftp left05:36
→ dnjaramba joined05:36
AndChat| → lonewulftp05:37
→ LongBeach joined05:37
← nikuyoshi left05:37
→ cloudgroup joined05:38
→ ehsan joined05:38
→ kermit joined05:39
→ nikuyoshi joined05:39
→ v0n joined05:39
→ hotswaphdd joined05:39
→ arvind_khadri joined05:40
← arvind_khadri left05:40
→ arvind_khadri joined05:40
petisnnake How can I add a file that I don't have on my local repo, so that I can pull it from a remote one?05:41
→ Graham joined05:42
Graham → Guest6262905:42
← guampa left05:42
→ mdel__ joined05:43
→ gianlucadv joined05:43
wereHamster petisnnake: you first copy it into your local repo?05:43
← Prasad left05:44
petisnnake wereHamster: using what git option? I'm assuming not simple wget05:44
wereHamster where do you want to download it from?05:45
← mdel left05:46
petisnnake my github repo05:46
I accidentally hit $ git rm foo05:46
wereHamster that would be git fetch05:46
petisnnake: git status05:46
run that, it'll tell you05:46
petisnnake yes, the file is marked as deleted05:46
so... $ git fetch foo?05:47
← dreiss left05:47
wereHamster petisnnake: can you pastebin the output from 'git status'?05:47
→ harshpb joined05:48
petisnnake Changes to be committed: deleted: foo05:48
wereHamster there's more05:48
petisnnake the rest is just untracked files05:48
→ freimatz joined05:48
fr0sty petisnnake: your local repository contains the entire project history. You don't need to get anything from github.05:48
→ cgd joined05:49
← Guest62629 left05:49
petisnnake fr0sty: so I just need to revert?05:49
wereHamster petisnnake: http://sprunge.us/GeNR05:49
how about this 'use "git reset HEAD <file>..." to unstage', don't you have that in your output?05:49
petisnnake wereHamster: I do, and as you can see I'm an idiot05:50
← abetaha left05:51
→ abetaha joined05:51
petisnnake hmmm I did $ git reset HEAD foo, but file is still missing and marked for deletion05:51
→ weiyang joined05:51
wereHamster petisnnake: run git status again05:51
petisnnake I did05:52
→ maddenwg joined05:52
← maddenwg left05:52
petisnnake hmm worked, must have misspelled something05:52
thanks wereHamster05:52
→ maddenwg joined05:53
← slonopotamus left05:54
petisnnake How can I remove all files matching a pattern? tried $ git rm *.py, but isn't working05:57
← jryans left05:57
→ Prasad joined05:57
← sattu94__ left05:58
→ dreiss joined05:59
→ jaequery joined06:00
→ slonopotamus joined06:00
fr0sty Try quoting: git rm "*.py" but note that will also delete the local copies.06:01
petisnnake thanks fr0sty, works06:02
← Bass10 left06:02
← cloudgroup left06:02
← p3rror left06:07
← Vortex34 left06:09
← serj left06:09
→ josefig joined06:11
→ dfr|mac joined06:11
→ aspotashev joined06:12
→ ThomasLocke joined06:12
← gianlucadv left06:12
→ TrineLocke joined06:13
→ yhager joined06:13
→ htoothrot joined06:13
→ serj joined06:13
← v0n left06:13
← dfr|mac_ left06:14
← jameslord1 left06:14
← maddenwg left06:15
← slonopotamus left06:15
← cyphase left06:15
← Dragooon left06:16
→ AmirBehzad joined06:17
← petisnnake left06:18
← ericmuyser left06:18
→ Ratler joined06:20
← jaequery left06:22
← madewokherd left06:24
← ipalaus_ left06:24
→ pro_metedor joined06:24
← kaushik_ left06:25
→ macRover joined06:25
→ anubhaw joined06:26
← aspotashev left06:27
→ v0n joined06:28
← pro_metedor left06:29
← y3di left06:30
← josefig left06:33
← queequeg1 left06:35
← maletor left06:37
→ dvide joined06:38
→ tomislater joined06:39
← fr0sty left06:42
← bigkm left06:42
← dfr|mac left06:43
→ patrakov joined06:43
→ fr0sty joined06:44
patrakov Hi. I tried to revert a commit, got "Automatic revert failed", and, after examining the conflict, decided that I don't actually want to revert it now. How do I abort the revert?06:44
← harshpb left06:45
← welly left06:45
← axl_ left06:46
→ axl_ joined06:46
← Goplat left06:47
→ welly joined06:47
EugeneKay man git-reset06:47
gitinfo the git-reset manpage is available at http://jk.gs/git-reset.html06:47
← jwmann1 left06:48
← sdtr443w left06:49
← Araxia left06:50
← axl_ left06:51
patrakov ok, did a "git reset --hard", it worked06:52
→ flijten joined06:52
→ cyphase joined06:53
cmn there's also git revert --abort06:53
though the effect is probably exactly the same06:54
→ marthinal_ joined06:54
← drev1 left06:54
← txomon|nqas left06:55
→ txomon|nqas joined06:55
→ Araxia joined06:55
← lonewulftp left06:56
marthinal_ → marthinal06:57
← jameslordhz left06:57
→ lonewulftp joined06:58
← mandric left06:59
← Spockz left07:00
→ ilteris joined07:00
→ mandric joined07:01
ilteris hey guys I am trying to ignore .htaccess and this is my gitignore file in my project folder: http://dl.dropbox.com/u/2070808/Screenshots/1u.png07:01
and this is my current git status: http://dl.dropbox.com/u/2070808/Screenshots/1v.png07:02
when I try to change something in the .htaccess it still throws me this :http://dl.dropbox.com/u/2070808/Screenshots/1w.png07:02
why is that?07:02
how can I make sure git status doesn't throw me any error (since I thought that file is not tracked anymore)07:03
→ cloudgroup joined07:05
← macRover left07:05
← thursday left07:05
→ JohnBat26 joined07:07
← Prasad left07:08
fr0sty ilteris: .gitignore does not work on tracked files.07:10
ilteris fr0sty how can I exclude already tracked file without deleting it?07:10
fr0sty You need to do something like 'git rm --cached .htaccess'07:11
That will stop tracking it.07:11
→ jameslordhz joined07:11
→ jmah joined07:11
ilteris fr0sty I tried git rm .htaccess but it removed it07:12
so --cached is the key there?07:12
← btanaka left07:13
fr0sty If you don't want to stop tracking it but still want git to ignore modifications you need to do something with git update-index --assume-unchanged but that is not recommended.07:13
Yes, --cached is important07:13
→ sonja joined07:13
→ cbreak-work joined07:14
← patrick99e99 left07:14
← caseymcg left07:16
→ jceb joined07:16
→ Sitethief joined07:17
← dreiss left07:19
← eighty4 left07:19
→ teweWORK joined07:19
→ eighty4 joined07:20
→ papegaaij joined07:21
→ airborn joined07:22
← Ratler left07:23
← sonja left07:23
→ stoffus joined07:23
→ gebi joined07:24
→ nzee joined07:25
← jceb left07:26
→ frerich joined07:27
← frerich left07:27
→ frerich joined07:27
→ Heisenmink joined07:27
→ OOPMan joined07:28
← ISF left07:28
← mandric left07:29
→ yshh joined07:30
← htoothrot left07:31
← pfloyd left07:32
← lonewulftp left07:34
→ lonewulftp joined07:34
→ Ratler joined07:35
← friskd left07:35
→ jceb joined07:37
→ Spockz joined07:38
→ dreiss joined07:38
← mdel__ left07:39
→ caseymcg joined07:39
→ hyperair joined07:39
→ ngulden joined07:39
← dorkmafia left07:41
← legnaleurc left07:41
→ legnaleurc joined07:42
← dreiss left07:43
→ rendar joined07:44
→ nessundorma joined07:45
→ h0bbit joined07:46
← airborn left07:48
→ batlock666 joined07:48
→ ASIRINeiluj joined07:49
→ mishok13 joined07:49
← DofLeFou left07:53
→ CrazyTux[m] joined07:54
CrazyTux[m] Whats the git log command to show a condensed file name +++++------ log ?07:54
FauxFaux --shortstat07:54
→ DofLeFou joined07:54
FauxFaux Uh, stat.07:54
CrazyTux[m] That is showing me commit messages, dates, etc.07:55
I only want filename and ++ -----07:55
→ dreiss joined07:55
CrazyTux[m] for lets say maybe the last 20 commits or somethnig07:55
cmn if you don't want the messages, you'll have to play around with --pretty07:56
→ phreaknerd joined07:58
← phreaknerd left07:58
← dirkle left07:59
← weiyang left08:00
→ _Qman joined08:01
← CrazyTux[m] left08:02
← Spockz left08:03
→ kaushik_ joined08:04
← Sonderblade left08:05
← arvind_khadri left08:05
→ dc5ala joined08:07
← secoif left08:09
→ whyking joined08:10
← bauruine left08:11
← nessundorma left08:11
← lapistano left08:11
← desowin_ left08:12
→ lapistano joined08:12
→ sykop5am joined08:13
← dreiss left08:13
→ maletor joined08:16
→ dorisabayon joined08:18
→ toabctl joined08:18
→ marthinal_ joined08:18
← danols_work left08:19
→ trez` joined08:19
← trez` left08:20
← kaushik_ left08:20
← igotnolegs left08:20
← marthinal left08:20
→ weiyang joined08:21
← Fandekasp left08:21
→ jutaro joined08:21
→ CrazyTux[m] joined08:24
← teweWORK left08:24
→ giallu joined08:26
← hyperair left08:27
→ strk joined08:27
strk how can I add a partial patch to the index ?08:28
→ dreiss joined08:28
CareBear\ git add -i08:28
?08:28
-ip maybe even08:28
→ whitman joined08:29
strk -p, thanks !08:30
→ padams joined08:31
→ cylence joined08:31
→ beautifulmind joined08:31
→ hyperair joined08:32
← Heisenmink left08:33
→ eijk joined08:33
→ kaushik_ joined08:34
→ ipalaus_ joined08:35
→ glcrazy joined08:36
← f0i left08:36
← toabctl left08:37
← Heimidal left08:37
← nikuyoshi left08:38
← kaushik_ left08:39
← psakrii left08:40
→ ph^ joined08:41
← kermit left08:41
→ stodan joined08:42
→ teweWORK joined08:42
← jmah left08:43
→ johnathanb joined08:47
→ nikuyoshi joined08:48
→ Element9_work joined08:49
← txomon|home left08:49
→ snearch joined08:50
→ txomon|home joined08:50
→ sebrock joined08:51
← fermianyon left08:51
← legnaleurc left08:52
→ xjonc joined08:52
← snearch left08:53
→ snearch joined08:53
→ legnaleurc joined08:54
← lonewulftp left08:54
→ arvind_khadri joined08:54
← arvind_khadri left08:54
→ arvind_khadri joined08:54
→ kaushik_ joined08:54
← diegoviola left08:54
← v0n left08:54
← strk left08:55
→ xjonc666 joined08:55
← beautifulmind left08:57
→ harshpb joined08:57
← xjonc666 left08:57
→ heisenmink joined08:58
← kaushik_ left08:58
→ soulcake joined08:58
→ Spockz joined08:59
→ TheFuzzball_iOS joined08:59
← mabrand left09:00
← JackWinter left09:00
← arvind_khadri left09:01
→ disappea_ joined09:01
→ tango_ joined09:02
← legnaleurc left09:02
→ mabrand joined09:03
→ legnaleurc joined09:04
→ gnufied joined09:06
→ Textmode joined09:07
→ Sigma joined09:07
← maletor left09:07
→ saschagehlich joined09:08
← snearch left09:08
← ehsan left09:09
← cylence left09:09
← gebi left09:09
← legnaleurc left09:09
→ FernandoBasso joined09:09
→ Chib joined09:10
← Textmode left09:10
→ legnaleurc joined09:10
← subbyyy left09:10
→ kaushik_ joined09:11
→ Textmode joined09:11
← disappea_ left09:12
→ eijk_ joined09:12
← eijk left09:12
→ lterje joined09:13
→ disappe__ joined09:14
← AmirBehzad left09:14
→ overbythere joined09:15
← Textmode left09:16
← kaushik_ left09:17
→ Textmode joined09:17
→ tatsuya_o joined09:18
→ arvind_khadri joined09:18
← OOPMan left09:19
← saschagehlich left09:21
← z0n3d left09:22
→ htoothrot joined09:23
→ tapout joined09:24
← whyking left09:24
→ detaer joined09:25
← detaer left09:25
tapout I have 2 servers.. both point to different repositories that are bare. I want to merge them to both point to the same... i did ... git remote add origin... then a git pull will say: You asked me to pull without telling me which branch you09:25
→ canton7 joined09:26
FauxFaux tapout: "git pull origin master" or so.09:26
tapout oh nice, i kept saying.. git pull origin/master :)09:27
← zm left09:27
← weiyang left09:28
→ gavin_huang joined09:29
tapout I have changes on 1 server, and changes on the other server. is there a better way to 'compare' them .. now that they are both pointing at the same repo... than to... move one of the files over... diff file /path/to/other.. and eyeball ?09:29
← xerora left09:29
→ marthinal joined09:30
← jceb left09:30
→ kaushik_ joined09:30
→ LimeBlast joined09:30
← Tazar left09:31
← marthinal_ left09:32
→ crashanddie joined09:32
← crashanddie left09:32
→ crashanddie joined09:32
pabs3 tapout: use your favourite graphical diff tool by running git difftool server1/master server2/master09:33
→ jds joined09:33
← Vile left09:33
→ pro_metedor joined09:34
← krz left09:34
← htoothrot left09:35
→ friskd joined09:35
← kaushik_ left09:35
→ ISF joined09:36
→ kaushik_ joined09:36
→ legnaleurc_ joined09:36
→ tcurdt joined09:37
← TheFuzzball_iOS left09:37
← legnaleurc left09:37
→ dotpot joined09:40
→ marmaladex joined09:40
→ krz joined09:40
→ jceb joined09:40
← GunArm left09:40
ChanServ set mode: +v09:41
← friskd left09:41
→ ryez joined09:41
→ nimrod10 joined09:41
→ friskd joined09:42
→ shruggar joined09:42
→ htoothrot joined09:42
marmaladex hi....I'm wondering if someone can help me with a simple mistake I made....09:42
← htoothrot left09:42
→ htoothrot joined09:42
wereHamster !ask09:42
gitinfo yes, it's okay to ask questions here... in fact, you just asked one! ;) Pretty much any question is fine. We're not terribly picky.09:42
← legnaleurc_ left09:42
wereHamster !asktoask09:43
marmaladex I created a git repo on my documents folder, but I didn't mean to do it there, so I just want to get rid of it09:43
I'm using mac terminal for this09:43
wereHamster rm -rf .git09:43
marmaladex that is depressingly sensible09:43
FauxFaux wereHamster: Arrrrgh! Your question is entirely useless. Fill in the blanks: "I tried [X] in [Y]. I got [A] but I was expecting [B]. The simplest way I can think to reproduce this is [C]." Reply on one line. If it doesn't fit, use a pastebin. Do not even think of speaking before fully filling in the question.09:43
(is my alias ¬_¬)09:43
marmaladex but thank you so much09:43
→ legnaleurc joined09:44
wereHamster alias for what?09:44
FauxFaux ¡asktoask.09:44
← yshh left09:44
→ yshh joined09:45
tapout thanks pabs309:46
→ Dave^|| joined09:47
→ TheFuzzball_iOS joined09:47
→ gusnan joined09:47
→ matleh joined09:48
← yshh left09:49
← Dave^| left09:50
pro_metedor How to get know on which commit I am actually ?09:50
FernandoBasso krz: Hey,09:50
FauxFaux pro_metedor: git branch -avv; git symbolic-ref HEAD; git rev-parse HEAD; git describe --tags; # or something like that.09:52
→ Shusshu joined09:52
pro_metedor a lot of that09:52
ryez regarding shared local clone, given I have 4 repositories, namely remote.git, local.git, b1, b2,09:53
among them, local.git is a normal clone of remote.git; while b1 and b2 are shared-local clone of local.git for branches with the same name respectively09:53
← legnaleurc left09:53
ryez then, when I'm in b1, is it possible to directly pull from remote.git?09:53
tapout so if I do; git tag -a v0.1 -m 'version 0.1' ... Is it in the -m that I should put summaries of what this new version holds? Or how do you guys do it?09:53
→ achadwick joined09:53
→ legnaleurc joined09:53
← matleh left09:54
← vervic left09:54
pro_metedor I want to add Xcode project file to repo, but I don't want git to version it further, just to store the versions that I indicate. How to do it ?09:55
← SiCKa left09:55
nacho charon: excellent, thank you very much.09:56
pro_metedor in other words: I want to add file to repo, but then make it untracked09:56
→ iocor joined09:57
← TheFuzzball_iOS left09:57
pro_metedor am I concluding in a good way ?09:57
tapout git add file.pl, git commit -am "added file" ... git rm file.pl ... git commit -am "removed file" ... now it's tracked and deleted? i dunno09:57
← friskd left09:57
→ rcu joined09:58
← legnaleurc left09:58
→ gianlucadv joined09:58
gitinfo set mode: +v09:58
← htoothrot left09:59
← jds left09:59
wereHamster tapout: if you git rm it, it becomes untracked10:01
→ legnaleurc joined10:01
rendar at every commit i make, git must store the file tree in a tree object, and it does that for every directory, right? i mean, every directory becames a tree object with its SHA-1 based on the SHA-1 on the files in that directory, am i right?10:01
FauxFaux Yep.10:02
→ Vile joined10:02
→ jds joined10:02
FauxFaux rendar: Tree objects aren't recursive, though, they store other trees; any unmodified subtrees won't be re-stored.10:02
← cccaldas left10:03
← weeb1e_ left10:03
rendar FauxFaux: yes, thats exactly what i mean, it works just as for unmodified files.. and, there is a root tree objects which has all tree objects for every directory in the root path of the project, right?10:03
ryez am I clear in the question above? I'm curious about how do you manage multiple working branches10:03
→ htoothrot joined10:04
cgd ryez: you want to pull a branch from a remote into a local clone of that branch?10:05
or am i reading that wrong10:05
← canton7 left10:05
→ weeb1e joined10:06
← legnaleurc left10:06
pro_metedor when I use git fetch, I download remote branch into my repo, but not merge?10:06
rendar FauxFaux: and, if my project is only 1 *huge* directory with thouusands of files, and without any directory, i'll only have a big root tree object which will be entirely rewritten for each commit i make, right?10:06
wereHamster pro_metedor: yes10:06
ryez +cgd, yes10:06
→ legnaleurc joined10:07
pro_metedor thnx10:07
cgd ryez: "git pull origin b1" maybe?10:07
pro_metedor how can I store the initial file and exclude it from tracking then ?10:07
wereHamster pro_metedor: don't add it at all10:08
!config10:08
gitinfo [!configfiles] the recommended way is to change your application so that you can (perhaps optionally) store site-specific configuration/data in separate files, then keep those files out of version control. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.10:08
ryez +cgd, I expect local.git to be a bare clone of remote.git (from a remote server), meanhwile I want to work only in b1, so that I can `git pull` directly from b110:08
→ plopix joined10:08
pro_metedor but I want it to be stored in repo with its first version10:08
cbreak-work don't change it10:08
git can not ignore and track a file at the same time, obviously10:09
pro_metedor next changes are not predictable, cause they are done by IDE10:09
cbreak-work then write-lock the file10:09
wereHamster pro_metedor: read that pastebin. really.10:09
I mean gist10:09
mjt or store a .in file10:09
or .orig, whatever10:09
ryez +cgd, probably not, because for b1 origin is local.git, `git pull origin b1` might not fetch from remote.git10:10
← legnaleurc left10:10
cgd ryez: could make a new remote for remote.git and pull from that10:11
ryez: i.e. upstream10:11
→ legnaleurc joined10:11
ryez +cgd: thanks, I'd like to give it a try10:12
mjt ryez: "+" in front of a nick isn't part of it, it is a flag. JFYI.10:12
ryez mjt: thanks, didn't know that :P10:12
→ JackWinter joined10:12
→ whyking joined10:13
ryez btw, what does that '+' mean? room OP?10:14
mjt heh. and i don't remember what does it mean. +v aka voice?10:14
heh10:14
chanop (operator) is marked with @10:14
but it is all off-topic here in #git :)10:15
cgd wtb a better workflow, paying in productivity boosts10:15
← legnaleurc left10:15
cgd downloading a whole site to develop locally feels like it takes longer in the beginning but i should just start doing it :<10:15
← dnjaramba left10:16
← pabs3 left10:17
→ dnjaramba joined10:17
→ legnaleurc joined10:17
← FernandoBasso left10:19
← AAA_awright left10:19
→ canton7 joined10:19
→ noisebleed joined10:20
← noisebleed left10:20
→ noisebleed joined10:20
← legnaleurc left10:20
→ legnaleurc joined10:20
→ ManDay joined10:21
ManDay Permission denied (publickey).10:21
On git pull - any idea why?10:21
ryez cgd, directly add the remote.git for b1 works quite well, thank you10:22
→ johnkpaul joined10:22
wereHamster ManDay: because you don't have permission? Maybe? Just a wild guess. No idea why I would think that..10:22
cgd ryez: welcome :-)10:22
ManDay wereHamster: I used to. Why would I suddenly stop having it?10:23
ryez afterward in b1, `git du -sh .git` is only 6.0M, that's great10:23
wereHamster ManDay: what did you change on your setup?10:23
ManDay wereHamster: PS: Your answer is stupid.10:23
wereHamster: My hostname and username are still the same.10:23
← xiangfu left10:23
ManDay So is the setup on gitorious10:23
wereHamster ManDay: ps: your question is incomplete, impossible to answer without knowing more about the situation.10:23
ManDay Fair enough.10:24
wereHamster ManDay: ask the server admin what he changed.10:24
ManDay wereHamster: What components are relevant?10:24
cmn ManDay: that comes from ssh, you're failing to give the ssh daemon a key to verify who you are10:24
ManDay s/What/Which10:24
cmn: I do have my usual public key in .ssh.10:25
cmn does ssh know to use it?10:25
← legnaleurc left10:25
mjt ssh does not use public keys10:25
→ Gorkyman joined10:25
mjt ssh client anyway10:25
ManDay I'll create a new key now.10:25
cmn mjt: what?10:25
mjt ssh client uses private key10:26
← forsto left10:26
cmn yes, and the server uses the public one10:26
→ legnaleurc joined10:26
cmn and the auth method which uses these keys is called publickey10:26
ManDay: a new one? gitorious certainly won't know about that key10:27
ManDay After I created a new key it works10:27
cmn: I added it to gitorious of course10:28
← Gorkyman_ left10:28
← rockfruit left10:29
mjt cmn: some time ago it was a common misunderstanding. people placed their public ssh keys on their client machines and wondered why they can't login... but ok.10:30
→ tomislater2 joined10:30
→ angelsl joined10:30
cmn then people need to read the instructions10:30
→ kennethreitz joined10:31
→ legnaleurc_ joined10:31
← legnaleurc left10:32
← tomislater left10:32
→ cccaldas joined10:32
← ryez left10:32
→ shogsbro_ joined10:33
→ hwit joined10:34
→ vervic joined10:34
← DofLeFou left10:35
← caseymcg left10:35
← legnaleurc_ left10:36
← hyperair left10:37
→ karstensrage joined10:37
karstensrage → Guest5451310:38
← hwit left10:38
→ legnaleurc joined10:39
← ManDay left10:39
← Sitethief left10:41
→ rockfruit joined10:41
← cortana` left10:43
← Gorkyman left10:45
→ cortana` joined10:45
dys` → dys10:47
→ Gorkyman joined10:48
← legnaleurc left10:50
← shogsbro_ left10:50
→ gjoseph joined10:51
← Gorkyman left10:52
← gianlucadv left10:54
→ dzonder joined10:56
→ lonewulf` joined10:57
→ Dragooon joined10:58
← robde left10:58
→ legnaleurc joined10:59
→ OOPMan joined11:01
← legnaleurc left11:04
→ fusedreality joined11:04
← OOPMan left11:05
→ soulcake_ joined11:05
← soulcake left11:05
→ umren joined11:07
← segher left11:07
← aro left11:07
→ pur3forlyphe joined11:07
→ psakrii joined11:08
→ legnaleurc joined11:08
→ d0k joined11:10
→ Borbus joined11:11
→ herdingcat joined11:12
← dc5ala left11:12
→ dkannan joined11:12
← legnaleurc left11:12
gitinfo set mode: +v11:13
Borbus I made a mistake the other day: I made a new branch locally, and then ran git push so it made the branch on the remote repo as well, then locally I made another branch and rebased the previous branch to that, forgetting that I had pushed it. Now what shall I do? Will deleting that first remote branch and pushing again work?11:13
→ segher joined11:14
FauxFaux Or push --force, which basically does the same thing, yes.11:14
→ Phylock joined11:16
Borbus Oh yeah, I see, so it loses the history on the remote, which is what I want11:16
→ Fandekasp joined11:18
→ robde joined11:18
← segher left11:18
→ legnaleurc joined11:18
← bartman left11:20
← dorisabayon left11:21
← jonesy left11:21
→ seivan joined11:22
← disappe__ left11:23
← legnaleurc left11:23
→ segher joined11:23
→ uu joined11:25
→ tillMorning joined11:25
→ legnaleurc joined11:27
→ bartman joined11:27
→ OOPMan joined11:28
tillMorning Hello. How I can 'pull request' for branch? I have a Github repository url and here's branch. I have created a branch in my localhost with the same name as it is in a GitHub, but how to add (commit) to github rep?11:28
FauxFaux You have to fork the github repo, then push your branch to your fork, then pull-request. #github.11:29
tillMorning I've done: git branch ; git branch testsoft ; git checkout testsoft ; git commit -m 'test' ; git push git@...11:29
how I can do pull request11:29
1,2 steps i've done11:29
FauxFaux #github11:30
tillMorning ok11:30
← Element9_work left11:31
← dr_ick left11:32
→ Xizor joined11:32
← marthinal left11:32
→ senny joined11:35
← rjcks left11:36
← johnkpaul left11:37
← nzee left11:37
← tellone left11:38
→ faber joined11:39
← shruggar left11:41
→ jatt joined11:44
← Fandekasp left11:44
→ lyth_ joined11:44
→ shruggar joined11:46
→ secoif joined11:47
→ vdv joined11:50
← lyth_ left11:51
→ lyth_ joined11:51
→ tellone joined11:51
← swombat left11:53
← tellone left11:53
ChanServ set mode: +v11:53
lyth_ hi11:54
→ brian_g joined11:54
lyth_ I have a question about using a user-based ACL system11:54
→ yshh joined11:55
← secoif left11:55
_ikke_ !question11:56
gitinfo You know, you have not actually asked a question. Without asking a question, it is hard to get any responses. Please explain what you are trying to do, what you typed in, and exactly what came back. Use http://pastebin.com for exact command sessions.11:56
_ikke_ lyth_: Just ask your question11:56
lyth_ If I have a server-side ACL which specifies which user can push changes to which directory .. How can I assure that the commiting user a is really user a and not b. How does authentication with acl works?11:57
→ gebi joined11:57
→ thiago joined11:58
→ Q| joined11:58
_ikke_ lyth_: systems like !gitolite use ssh keys to identify users11:58
gitinfo lyth_: gitolite allows you to host any number of repositories with extensive access control options for any number of users, using just one system account: http://github.com/sitaramc/gitolite11:58
→ Cromulent joined11:58
FauxFaux lyth_: Most people do security based purely upon the user that is executing the push, not on the committer at all. There exist systems to ensure that you can only push your own commits, but they're stupid.11:58
← shruggar left11:58
→ shruggar1 joined11:58
→ leen^| joined11:58
_ikke_ lyth_: But directory based ACL is very uncommon in git11:59
→ amitprakash joined11:59
jast that's because ACLs aren't used much in general12:01
amitprakash Hi, for a given project, I cloned the remote master branch locally and branched local/master to local/newfeature.. however, while working oin newfeature12:01
← zdk left12:01
← Dave^|| left12:01
amitprakash I also made changes to files existing in master .. meanwhile I've kept my newfeature updated with origin/master using gitl pulls and resolving conflicts12:02
now if I wish to send a clean patch to remote to apply, how can I create it?12:02
← Borbus left12:02
cmn git diff newfeature master ?12:02
wereHamster amitprakash: rebase the commits and clean up the history12:02
amitprakash cmn, with diff.. remote won't be able to sign off will it?12:03
lyth_ Okay ... well .. I´m using EGit and in EGit I change the committer´s and author´s name. Are there any possibilities to disable this?12:03
amitprakash wereHamster, wont rebase cause the conflicts I resolved during git pull all over again?12:03
cmn lyth_: you could use a pre-commit hook, but then a malicious user would just disable it12:03
wereHamster maybe12:03
← [1]Kichael left12:03
→ Element9_work joined12:03
jast lyth_: you'd have to write a hook on the server side that verifies uploaded history12:03
amitprakash wereHamster, whiel I now realize that I should have after every commit updated local/master and done a rebase on local/newfeature12:03
wereHamster amitprakash: you could als git fetch; git rebase origin/master; without having to touch local master at all12:04
← stringoO left12:04
→ secoif joined12:05
_ikke_ lyth_: Maybe something like this: http://mwop.net/blog/236-GPG-signing-Git-Commits12:05
amitprakash wereHamster, that makes sense too.. sadly now I have to re-resolve conflicts I've already resolved12:05
→ johnkpaul joined12:05
← shruggar1 left12:05
→ shruggar joined12:05
← brainbox1100 left12:05
cmn lyth_: if you don't trust people not to lie about their name, why can they push to your repo?12:06
→ mr_mustard joined12:06
← mr_mustard left12:06
wereHamster amitprakash: only once if you use git-rerere12:06
cmn lyth_: such a hook would forbid a new branch based on someone else's work as well12:06
jast cmn: you could do a reachability test using existing branches, too12:06
cmn jast: unless dev A pulled from dev B directly12:07
→ brainbox1100 joined12:07
jast well too bad :)12:07
cmn sure12:08
I find the premise flawed anyway12:08
→ hyperair joined12:08
jast in decentralized development, sure12:08
cmn not just that12:08
why does this person have push permissions if I don't trust them not to lie about the commit's author?12:09
lyth_ I don´t ask these questions because I don´t trust a special person, that can commit to the remote repository. I ask this because I woult like to know, if it is possible in general12:12
cmn it's possible, but it's generally the wrong answer12:14
_ikke_ lyth_: Git is distributed, so it's hard to enforce locally12:16
lyth_: hard / impossible12:16
→ RaceCondition joined12:18
wereHamster hard with current git, probably somewhat possible with signed pushes12:18
s/hard/impossible/12:18
RaceCondition I'm using git-svn and want to do a merge using git and then commit back to SVN... this worked out nicely, except SVN only sees the merge commit.. so I was thinking: is it possible to tell "git merge" to re-commit all commits from the other branch? mmth like cherry-pick'ing one by one but automatically and with proper conflict resolving12:19
wereHamster RaceCondition: use git rebase to make the history linear12:19
← shruggar left12:20
← jceb left12:20
RaceCondition wereHamster: did you read the part where I said "git-svn"?12:20
wereHamster so?12:20
RaceCondition you cannot use git rebase with remote tracked SVN branches12:20
wereHamster ah, you want to merge two svn branches?12:20
RaceCondition yes12:20
← Chib left12:21
wereHamster does git-svn even support svn mergeinfo?12:21
jast not really12:21
wereHamster maybe try to merge with git, resolve the conflicts, then copy the working tree into a svn checkout12:21
RaceCondition I could just loop over `git log mybranch..master` and `git cherry-pick` each commit.. but I'm not sure if I'll then get all the git merge conflict resolution goodness12:21
→ DaveIngram joined12:21
jast well mergeinfo is supported, but you need to do it manually12:22
RaceCondition I want to commit ALL COMMITS from the other branch, not just a single "Merged blabla" commit12:22
jast oh, actually there's a config option now12:22
not sure how well it works12:22
RaceCondition what config option?12:23
wereHamster RaceCondition: in a svn checkout, merge the two branches. You'll get conflicts. go to a git-svn clone, merge the exact same two branches, resolve the conflicts, copy working tree to the svn checkout. in the svn checkout mark all files as resolved and commit12:23
jast for enabling git-svn to record merge information in svn12:23
svn.pushmergeinfo12:23
← harshpb left12:23
cmn git svn --merginfo needs the whole file so it learnt recently how to set the right info itself12:23
RaceCondition wereHamster: that will not keep the history; it will just commit everything in a single commit12:23
→ vervic_ joined12:24
← segher left12:24
→ jceb joined12:24
wereHamster RaceCondition: I assume that svn correctly records the history, right?12:24
if you merge two branches with svn ...12:24
jast sure, but merging in svn is painful12:24
RaceCondition ah, well, in that sense12:24
yeah, OK, I get your idea, but I prefer svn.pushmergeinfo :)12:25
thanks for pointing it out, jast12:25
jast you're welcome12:25
fwiw I didn't know about it until now, either12:25
I found it by reading the manpage12:25
← dv310p3r left12:26
→ shruggar joined12:26
→ the_cybersphinx joined12:26
RaceCondition I didn't even have an idea what to look for in the manpages12:27
also, I didn't know svn had mergeinfo (never used pure SVN... started with darcs and then moved on to git)12:27
← vervic left12:28
vervic_ → vervic12:28
cmn svn:mergeinfo is the only way svn knows that a merge happened12:28
RaceCondition I thought it was simply ignorant about it altogether12:28
cmn that's why svn:merginfo is there12:28
RaceCondition glad to know that it's not the case12:28
cmn svn doesn't handle merges that well12:28
← kennethreitz left12:28
cmn it doesn't recognise a reintegrate-merge, you have to tell it yourself12:28
and in svn:merginfo you have to put each every single merge in the repo12:29
← setmeaway left12:29
cmn and a typo in that file (say you use git-svn --mergeinfo and do it manually) makes the repo unusable12:29
← cybersphinx left12:29
→ segher joined12:29
← tillMorning left12:30
RaceCondition yeah, svn totally sucks, I'm aware of that :)12:30
lyth_ cmn_: could you please give me a tip how it´s possible to disable changing commiters name?12:30
cmn you can't really unless the system has knowledge of the user's name12:32
so in the pre-commit hook, you could ask git for ident and tell it to ignore any configuration12:32
so it'll give you its primary diea12:32
← robde left12:32
cmn which may or may not be right12:32
to check on push, with e.g. gitolite, you can check $GL_USER against some database12:33
jast RaceCondition: svn introduced merge info in version 1.5. before that you were right about it not doing proper merges. and even now it's cumbersome to use.12:33
cmn the $GL_USER check would be in update12:33
← johnkpaul left12:33
→ RaceCondishun joined12:34
tizzo-afk → tizzo12:34
← gebi left12:35
← RaceCondition left12:36
RaceCondishun → RaceCondition12:36
← pidus left12:36
→ Targen joined12:36
← lonewulf` left12:36
→ johnkpaul joined12:37
← seivan left12:39
→ berserkr joined12:41
→ feindbild joined12:41
← cgd left12:41
RaceCondition btw, if I make a git commit under a different author name than myself and then `git svn dcommit`, will SVN simply override the author's name based on my auth credentials?12:45
(this is really not a 100% #git question, but people in #svn would know even less.. :) )12:45
cmn svn will use whatever user you manage to auth against12:45
RaceCondition ok, so unless I can auth as the other user, SVN won't let me commit as him12:46
cmn not as far I know12:46
it'd be silly/dangerous is a centrealised system12:46
← secoif left12:47
← horieyui left12:47
→ plaugg joined12:48
ChanServ set mode: +v12:48
← johnkpaul left12:50
← amitprakash left12:50
← hyperair left12:50
RaceCondition yeah, but it was worth asking :P12:51
→ AAA_awright joined12:51
→ ajaykools joined12:51
← babilen left12:51
ajaykools Hi, I am in detached head state, and I did a commit in that, I am not able to see my other branches. What do I do?12:52
cmn what do you mean you can't see them? git branch doesn't show anything?12:52
→ _pingu joined12:53
← _pingu left12:53
ajaykools cmn- it shows me the present branch I am at.12:53
wereHamster ajaykools: can you pastebin the output from 'git branch' ?12:53
cmn so it only shows (no branch) ?12:53
ajaykools wereHamster- just a moment12:54
cmn- it shows both. I tried to rectify it by fetching branches, seems to be badly gone wrong.12:54
cmn ok, so you only have one local branch (and on detached HEAD you're not on any branch)12:55
what do you want to do?12:55
did you use to have other branches?12:55
local branches, that is12:55
ajaykools cmn- Ya I had. I want to go back to a state where I can get all my branches.12:55
cmn if branch doesn't show them, you've deleted them at some point12:56
wereHamster explain what you mean by a 'state where you can get all your branches'?12:56
you already have all your brnaches, what more do you want?12:56
ajaykools cmn wereHamster http://paste.pocoo.org/show/533749/12:56
but I can see them in github12:56
the remote still has it basically12:56
wereHamster what can you see on github?12:57
ajaykools wereHamster- my branches12:57
wereHamster ajaykools: can you pastebin output from 'git branch -a'?12:57
← alberto56 left12:57
cmn it looks like you've created a branch called HEAD which is going to cause problems12:57
and another one called 'origin/gh-pages'12:57
which is also going to cause problems12:58
ajaykools wereHamster- http://paste.pocoo.org/show/533751/12:58
wereHamster it's interesting to see in what interesting way people can screw up theeir repository12:58
ajaykools cmn- I see, but that it is still at local right, I can clone my remote again.12:59
:)12:59
_ikke_ ajaykools: You can still fix your local repository12:59
cmn they're local but origin/gh-pages looks like a remote branch12:59
gtg12:59
_ikke_ ajaykools: You have some branches with names that confuse git13:00
ajaykools _ikke_- that sounds really motivating. :)13:00
← psakrii left13:00
ajaykools _ikke_- so how do I fix them?13:00
_ikke_ ajaykools: like cmn said, you have a branch called HEAD and a branch called origin/gh-pages13:01
ajaykools _ikke_- so I delete them?13:01
wereHamster if you want that..13:01
_ikke_ ajaykools: If your sure you don't 'loose' some work13:02
ajaykools wereHamster- yes, how do I do that13:02
wereHamster ajaykools: git branch -d13:02
read the man page13:02
ajaykools wereHamster- thanks a lot :)13:02
→ RaceCondishun joined13:02
→ cannonball joined13:03
← uu left13:03
← shruggar left13:03
→ shruggar joined13:03
→ kermit joined13:04
→ bauruine joined13:04
ajaykools wereHamster- _ikke_ So now, how do I get the branches at my remote, when I clone the repo, it does not get fetched.13:04
_ikke_ !remote_branches13:05
gitinfo Remote branches are not created locally on clone. They show up as remote branches (git branch -r). To create local branches form them use git checkout -b local_name remote/branch_name or more recently git checkout -t remote/branch_name13:05
→ Fandekasp joined13:06
← RaceCondition left13:06
RaceCondishun → RaceCondition13:06
← Targen left13:06
→ t4nkd joined13:08
→ robde joined13:08
→ impulze joined13:10
impulze sitaram: ping :P13:10
← robde left13:10
_ikke_ impulze: !anyone13:10
gitinfo impulze: Usually, it does not help to ask for someone specific to answer your question. No one is the sole expert on anything, or even if they may be, there are plenty of others who know enough to help you13:10
ajaykools _ikke_- thanks a lot :)13:11
impulze _ikke_: thanks, don't teach me irc ;)13:11
_ikke_ impulze: I don't know everyone here, so I don't know how much each person knows13:11
impulze: So please see it as a guide, not as some sort of command13:12
impulze yeah, it would be scary if you would know everyone here13:12
sure, i didn't mean it literally, hence the smiley :P13:12
_ikke_ right13:12
tizzo → tizzo-afk13:13
→ TomyLobo joined13:14
→ vervic_ joined13:14
→ swombat joined13:15
← vervic left13:16
vervic_ → vervic13:16
← spvensko left13:16
← fusedreality left13:17
→ lonewulf` joined13:17
← Vampire0_ left13:18
patrakov if I push some commits that change, say, the README file, change other files, change the README back to its original state, would that create conflicts for other people who also want to change the README?13:18
→ akosikeno joined13:19
_ikke_ patrakov: Only if you change the same lines13:19
or they13:19
Mikachu patrakov: git only looks at the endpoints of branches when merging, so no13:20
← gusnan left13:20
Mikachu unless someone is rebasing your branch on their edits.. which they should never do :)13:20
→ nardev joined13:21
← tomislater2 left13:22
→ pfsmorigo joined13:22
← canton7 left13:24
→ johnfarrell joined13:24
← johnfarrell left13:25
→ ahmedkotb joined13:25
→ canton7 joined13:26
→ digitalfiz joined13:27
→ babilen joined13:28
← babilen left13:28
→ babilen joined13:29
tizzo-afk → tizzo13:30
diverdude i have a repo on my server. I also have a git-user, which i created like this: adduser --system --home /var/www/gitorious/ --no-create-home --group --shell /bin/bash git && chown -R git:git /var/www/gitorious. When i try to clone the repo, it asks me for a password. Which password is that? I did not set a password13:30
← brian_g left13:30
→ jsick3 joined13:31
→ nardev1 joined13:31
patrakov you gave to either set a password ("passwd git") or, better, register a public key13:31
→ soc42 joined13:32
← digitalfiz left13:32
→ Vampire0_ joined13:32
diverdude patrakov, but i did register a public key13:32
→ digitalfiz joined13:32
patrakov diverdude: then sshd is for some reason ignoring it13:32
diverdude dammit13:32
patrakov could you please paste the output of the following commands to a pastebin?13:32
wereHamster `sl`13:33
← digitalfiz left13:33
→ digitalfiz joined13:33
→ Ryback_ joined13:33
← Ryback_ left13:33
→ Ryback_ joined13:33
patrakov ls -ld /var/www/gitorious/.ssh/authorized_keys /var/www/gitorious/.ssh /var/www/gitorious /var/www /var /13:33
← nardev left13:33
→ sattu94 joined13:34
→ Error404NotFound joined13:35
← Error404NotFound left13:35
→ Error404NotFound joined13:35
→ brian_g joined13:35
← whyking left13:36
← soc42 left13:37
→ beaufour joined13:38
→ erichynds joined13:38
→ tommyvyo joined13:39
tizzo → tizzo-afk13:39
→ nixmaniack joined13:39
← EricInBNE left13:39
← segher left13:39
→ khmarbaise joined13:40
→ neurodrone joined13:43
← Cromulent left13:44
→ pikpik joined13:44
→ vervic_ joined13:44
pikpik Hi. Is there a way to have Git do its processing on-disk instead of in-memory?13:45
← stodan left13:45
tizzo-afk → tizzo13:45
IcePic add swap files? =)13:45
pikpik What if my platform doesn't have those? :D13:46
← vervic left13:46
vervic_ → vervic13:46
wereHamster you're out of luck13:46
pikpik Great. Thanks.13:47
FauxFaux Build with an allocator that implements swap? I've seen at least one.13:47
wereHamster pikpik: which platform does not have swap if I may ask?13:47
← ahmedkotb left13:47
pikpik wereHamster: MINIX.13:47
FauxFaux Also, when do you run out of memory (what operation)?13:48
diverdude how can i list all available repos`?13:48
pikpik FauxFaux: And rebuild Git with that allocator?13:48
tizzo → tizzo-afk13:48
wereHamster diverdude: available repos where?13:48
in the whole world?13:48
pikpik FauxFaux: When I checkout a branch of pkgsrc (a Git version): http://git.minix3.org/?p=pkgsrc.git13:49
diverdude wereHamster, on my server13:50
pikpik (Pkgsrc is huge: http://pkgsrc.se/statistics.php )13:50
wereHamster diverdude: find / -name .git -type d13:50
→ guampa joined13:51
pikpik FauxFaux: I have 256 MB of memory and ~8 GB of storage.13:51
→ centipedefarmer_ joined13:51
→ jwmann joined13:51
pikpik I think the same thing happens with 512 MB of memory. It just takes a little longer.13:52
→ spvensko joined13:52
← spvensko left13:52
→ spvensko joined13:52
→ pepr joined13:52
ChanServ set mode: +v13:52
diverdude wereHamster, thx....it seems my gitorious is not working after all :(13:52
→ zulax joined13:53
← spvensko left13:53
wereHamster diverdude: that will find only non-bare repos..13:53
diverdude wereHamster, non-bare?13:53
FauxFaux pikpik: I wonder what it's doing that needs memory. Not that I can think what operations checkout uses, except maybe unpacking the pack-files wrong.13:53
→ acl_ joined13:54
wereHamster diverdude: gitorious stores all repositories in a single directory. It is configurable. So just look where it stores them, and look there13:54
→ eggman2001 joined13:54
t4nkd Hey git, I have a single commit in my branch that I want to like, undo/remove -- but I want to keep the other commits above it -- how might I do this?13:55
FauxFaux t4nkd: !fixup13:56
gitinfo t4nkd: So you made a bad commit and want to remove/fix it? (1) NOT PUSHED/PUBLISHED: `git rebase -i $COMMIT^` or perhaps `git commit --amend` (or `git reset HEAD^`). (2) OTHERWISE, `git revert $COMMIT` to make a reverse commit. (3) If you have pushed and MUST remove it, use rebase or filter-branch and type !rewriting_public_history in IRC.13:56
t4nkd sweet, ty13:56
pikpik FauxFaux: Do you know of a verbose/diagnostic/debugging option?13:56
t4nkd What is '$COMMIT^' indicative of?13:56
FauxFaux There's GIT_TRACE=1 git checkout whatever #, or strace.13:56
t4nkd: The name of the commit you want to edit, followed by a "^".13:57
SethRobertson and many git commands support -v13:57
pikpik Ok.13:57
t4nkd oh okay cool, tyty again13:57
SethRobertson t4nkd: aka the (first) parent of $COMMIT13:57
← nikuyoshi left13:57
t4nkd Hrm okie, when I use '^' it says no matches can be found... but ZSH tells me this13:58
FauxFaux Either quote it, or use ~ instead.13:58
(Identical meaning in this case.)13:58
diverdude wereHamster, hmm yeah ok...it does actually create repo13:58
→ cemerick joined13:59
diverdude wereHamster, but it gets some weird name like: f8c/166/6cfce7a632ca258f14ead6d217fc92f862.git13:59
wereHamster, maybe thats why i cannot clone it13:59
cemerick How would I go about getting numstats out of git log via --format?13:59
t4nkd Awesome, thanks for all the quick help guys13:59
→ dv310p3r joined13:59
t4nkd K+ FauxFaux SethRobertson13:59
← helgikrs1 left14:00
→ alisonken1home joined14:01
cemerick Or, let's put it this way: I'd like to get the output of `git shortlog -ns`, but with line change stats added. What is the best way to do that?14:01
→ jonesy joined14:01
← alisonken1home left14:01
→ spvensko joined14:02
← spvensko left14:02
→ spvensko joined14:02
diverdude wereHamster, do you have any idea why i cant clone my repo?14:02
→ segher joined14:03
← Nibato left14:03
→ bartek joined14:03
→ Nibato joined14:04
→ glennpratt joined14:08
← bindaasomatic left14:08
← DaveIngram left14:09
→ AaronMT joined14:09
→ cakehero joined14:10
→ sudoquai joined14:11
→ Cromulent joined14:11
← rchavik left14:12
← centipedefarmer_ left14:12
SethRobertson !repro14:12
gitinfo please paste (e.g. using http://git.pastebin.com/) a transcript of your terminal session in which the problem occurred so that we can troubleshoot more easily.14:12
→ FabTG joined14:13
← eggman2001 left14:13
← anubhaw left14:13
→ anubhaw_ joined14:13
SethRobertson cemerick: You may need to post-process the output of `git log` to get what you want14:13
→ VladimirZ joined14:14
→ vervic_ joined14:14
→ heptadecagram joined14:14
VladimirZ Does anyone know if Fisheye supports smth like github "pull requests" ? making branches or posting diffs at forum just for pre-commit review not good14:15
https://github.com/symfony/symfony/pulls https://github.com/symfony/symfony/pull/310214:15
→ helgikrs1 joined14:15
→ Bass10 joined14:15
VladimirZ now if i want to publish some patch for review i 1) posting diff at forum 2) making a branch at main repo with patch 3) just pushing into master14:15
SethRobertson Why not create a new repo like github does if you don't like the other options14:16
← [M]ax left14:16
→ GunArm joined14:16
← vervic left14:16
vervic_ → vervic14:16
diverdude if i make an empty project, can i then clean it?14:17
sorry, empty repo i mean14:17
FauxFaux What the hell are you even doing?!14:17
diverdude FauxFaux, me?14:17
→ ISF_ joined14:17
← OOPMan left14:18
→ dorisabayon joined14:18
→ alan_w joined14:18
← vervic left14:18
← rcu left14:18
PerlJam diverdude: If it's already empty, why would you want to clean it?14:18
← swombat left14:18
← lyth_ left14:19
diverdude PerlJam, clean it?14:19
ack14:19
i mmean clone it14:19
sorry bout that14:19
→ shruggar1 joined14:19
→ [M]ax joined14:19
PerlJam oh! heh, yes you can clone any repo14:19
← shruggar left14:19
← ISF left14:19
diverdude hmm ok...damn then there is some problem14:19
VladimirZ SethRobertson, its more overhead i guess than simple branch14:20
just need to find a way to cleanup branches from fisheye14:20
← alan_w left14:20
SethRobertson VladimirZ: Why not push a delete of the branch?14:20
→ alan_w joined14:20
→ swombat joined14:21
← Fandekasp left14:22
← spvensko left14:22
sitaram impulze: if it's something simple I can help you right now. Otherwise wait for another 30 min or so14:22
impulze sitaram: i'll wait, i don't know yet if it's simple ;) take your time14:23
sitaram impulze: (and _ikke_ ) I was the one who created the ! anyone trigger on our bot14:23
impulze :)14:24
too many requests? :P14:24
→ ahmedkotb joined14:24
impulze maybe you need #gitolite :D14:24
sitaram have you checked?14:24
impulze not on freenode14:24
sitaram there *is* #gitolite14:24
← ryanakca left14:24
impulze :(14:24
sitaram me "dont teach me IRC" you are wrong14:24
_ikke_: ^ :-)14:25
impulze hihi14:25
diverdude did anybody experience cloning problems with a locally hosted gitorious system?14:25
t4nkd Hrm, I just made a commit, but I actually wanted to checkout to a new branch /then/ commit, what's the easiest way to backtrack and make that happen?14:26
← patrakov left14:26
sitaram s/me/mr/14:26
→ oriba joined14:26
VladimirZ SethRobertson, for some reason it doesn't14:27
← jds left14:27
VladimirZ Looks like Fisheye keeps branch history or something14:28
→ mandric joined14:29
← Cromulent left14:29
← swombat left14:30
← dzonder left14:31
VladimirZ http://gitready.com/beginner/2009/02/02/push-and-delete-branches.html14:31
This is how I'm trying to delete a branch14:31
but for some reason it doesn't do anything14:31
I still see a branch on t he server14:31
the*14:31
any idea ?14:31
→ alansaul joined14:31
→ nikuyoshi joined14:32
← canton7 left14:32
← dnjaramba left14:32
← sh[4]rm4 left14:32
→ gusnan joined14:33
← Spockz left14:35
pepr Hi, I am confused with the following git log results...14:35
→ jpalmer joined14:35
→ Spockz joined14:35
pepr git log --pretty=format:"%ad %s (%an)" --date=short >log.txt14:35
It generates about 80 lines.14:36
FauxFaux Stop pressing return every four words. /o\14:36
VladimirZ Can someone help me please, it's really important =/14:36
→ menace joined14:37
→ psakrii joined14:37
menace if i create a git tag without signing it, will it be pushed too to my origin?14:37
← zulax left14:37
→ zi joined14:37
→ polygonpusher joined14:37
zi → Guest166914:38
polygonpusher whats the cmd to check the status of a remote repo?14:38
pepr The other git log command should add the --tags14:38
git log --pretty=format:"%ad %s (%an)" --date=short --tags >logT.txt14:38
but it generates much more lines and no tag lines. Why?14:38
→ TeckniX joined14:38
diverdude did anybody experience cloning problems with a locally hosted gitorious system?14:39
_ikke_ polygonpusher: What do you want to see?14:39
→ markkalderon joined14:41
← freimatz left14:42
← antix left14:43
← markkalderon left14:43
tizzo-afk → tizzo14:44
→ sh[4]rm4 joined14:45
← cemerick left14:45
→ Borbie joined14:45
→ antix joined14:45
→ tomislater joined14:46
→ canton7 joined14:47
→ OOPMan joined14:47
← cortana` left14:47
→ empity joined14:48
← Borbie left14:48
← kerrick left14:48
→ whyking joined14:48
← alansaul left14:50
← alecat left14:50
← krz left14:51
binjured is there a simple command to ask git if a commit is a merge commit or not?14:51
→ cortana2 joined14:51
binjured i'm trying to cherry-pick a range of commits but if it hits a merge commit without "-m" it dies14:51
← psakrii left14:52
cortana2 is 'git config --global url.foo.bar' supposed to work with 'git submodule update'? when I update, the rewritten URL is not used. a manual clone uses the rewrite, however14:52
→ alansaul joined14:53
→ robde joined14:53
→ dfr|mac joined14:53
→ Gorkyman joined14:54
← Gorkyman left14:54
→ Gorkyman joined14:54
← bandroidx left14:54
← demi` left14:55
cmn binjured: I'm not sure if there's a command, but with 'git cat-file commit <hash> | grep ^parent' yo ucan see how many parents it has14:55
→ shruggar joined14:55
← shruggar1 left14:55
binjured cmn: ah, okay, thanks14:55
← robde left14:56
← neurodrone left14:56
← ProperNoun left14:56
SethRobertson binjured: You can do a `git log -n 1 --pretty="%p" caa9913 | grep ' '`14:56
← polygonpusher left14:56
→ robde joined14:56
→ y3di joined14:56
→ mercutio22 joined14:57
← sudoquai left14:57
→ airborn joined14:58
→ Prasad joined14:58
← robde left14:58
→ squentin joined14:58
→ spvensko joined14:59
→ robde joined14:59
→ FernandoBasso joined14:59
← spvensko left14:59
← sattu94 left15:00
← cbreak-work left15:00
→ cbreak-work joined15:00
menace let's say i have a git repo, pushed it into a bare repo (with a tag) and cloned that repo to another location. if i delete the tag at the other location, and i push --tags and pull from the bare repo (which is origin btw) to the first repo.. is the tag there deleted, too?15:00
→ sudoquai joined15:01
VladimirZ SethRobertson, http://pastebin.com/ndbhAzbf - I'm doing it correctly, right ?15:01
← gavin_huang left15:01
→ bandroidx joined15:01
← tcurdt left15:02
→ jetienne_ joined15:02
← Xizor left15:02
jetienne_ q. on macos, the fs ignore char cases. is there a way to force git to not ignore it ?15:03
← _Qman left15:03
→ kukks joined15:03
SethRobertson VladimirZ: Looks like it. You could log into the server and run `git branch` on the server repo to make sure it was gone15:03
binjured SethRobertson: oh, nice; and if it comes up empty, then it's not a merge commit.15:04
SethRobertson Right, or if it exits true it will be a merge commit15:04
← khmarbaise left15:04
← zivester left15:05
← canton7 left15:05
→ gavin_huang joined15:06
→ marthinal joined15:06
← gavin_huang left15:06
→ canton7 joined15:07
← spiette left15:07
→ spiette joined15:07
→ tcurdt joined15:07
← mercutio22 left15:08
← teweWORK left15:09
← tizzo left15:09
→ teweWORK joined15:09
→ tizzo joined15:10
teweWORK hi15:10
VladimirZ SethRobertson, it's still there :/15:10
SethRobertson VladimirZ: `git branch` shows that it is there? Please pastebin.15:11
← Q| left15:11
← Gorkyman left15:11
teweWORK can i do git submodule foreach on selected submodules with an alias?15:11
← Spockz left15:12
→ Gorkyman joined15:12
SethRobertson teweWORK: It doesn't look like git-submodule-foreach has a selector.15:12
→ dr0id joined15:12
teweWORK I know, that's why I ask :)15:12
SethRobertson teweWORK: Of course you can do the "foreach" work manually15:13
→ mercutio22 joined15:13
← Hapz left15:13
← eeanm left15:14
→ jacobat joined15:15
← JohnBat26 left15:15
→ ufk joined15:15
ufk hello15:15
can i configure .gitignore to ignore *swf file recursively or do i need to add .gitignore to each and every directory in the tree ?15:15
→ Q| joined15:16
jacobat ufk: just add it to the root15:16
VladimirZ SethRobertson, http://img822.imageshack.us/img822/7668/branch.png15:16
ufk just to add "*swf" to the root and it ignores it recursively ?15:16
didn't know that :) cool15:16
jacobat yup15:16
← pepr left15:16
→ snearch joined15:17
→ jryans joined15:17
jacobat Any ideas on how to debug a merge strategy not running based on a gitattributes file?15:17
I've got: "db/schema.rb merge=railsschema" in my ~/.gitattributes15:17
SethRobertson VladimirZ: I'm more interested in what git has to say, not what some GUI has to say. If you don't have server access you can `git ls-remote origin`15:17
jacobat But it seems to have no effect15:17
→ thannoy joined15:17
← iocor left15:17
jacobat I've got [merge "railsschema"] in ~/.gitconfig15:17
→ luxurymode joined15:18
← oriba left15:18
→ amigojapan joined15:18
← dfr|mac left15:18
→ axl_ joined15:18
→ thannoy_ joined15:19
← thannoy left15:19
← brian_g left15:19
luxurymode Simple question here (I think…): git is telling me that in order for me to merge originm/master into master, i need to commit changes to some file first because its gonna be overwritten by the merge. Now, I happen to know that in origin/master, the file was accidentally deleted. How can I merge origin/master but ensure that I can quickly grab back the file I currently have in my working directory?15:19
→ Danielpk joined15:19
thannoy_ → thannoy15:19
→ iocor joined15:20
luxurymode like this? git rev-list -n 1 HEAD -- <file_path>15:21
SethRobertson Do a merge --no-commit and then we can do something like `git checkout HEAD -- filename` to recover the file.15:21
luxurymode SethRobertson, sweet. thanks a lot!15:21
SethRobertson, but i still need to commit first or does --no-commit allow me to merge w/o commit?15:21
→ Chani joined15:22
SethRobertson luxurymode: Strangely enough, man git-merge suggests that using --no-commit asks the system to…wait for it…not commit15:22
gitinfo luxurymode: the git-merge manpage is available at http://jk.gs/git-merge.html15:22
luxurymode lol arite thanks15:22
jetienne_ → jetienne15:23
← h0bbit left15:23
menace pushing/pulling removed/renamed tags aka keeping removed tags in a synchronized state across several repositories seems to be a little issue?15:23
Chani I have a git repo with about a dozen commits. I need to push it into a folder in an existing svn repo. is there an easy way to get the history over to svn, or should I just dump it in one commit?15:23
← airborn left15:24
luxurymode SethRobertson, still says "Please, commit your changes or stash them before you can merge."15:24
← glcrazy left15:24
→ psakrii joined15:24
→ Gorkyman_ joined15:25
← Gorkyman left15:25
SethRobertson Yes, that is true15:25
fr0sty Just move the file aside for now...15:25
mv thefile thefile.save15:25
Then merge...15:25
luxurymode i just committed it15:26
then did the merge15:26
and checked out HEAD for the file name15:26
fine right?15:26
→ hyperair joined15:26
→ Spockz joined15:27
SethRobertson Then commit, talking about hte merge and the file restoral.15:27
Only you can say if the working directory is in the right state15:27
→ centipedefarmer_ joined15:28
luxurymode so i committed my local stuff, then did the merge --no-commit, then right after i checkedout HEAD --filename15:28
now git status says my branch and origin/master have divereged, 4 and 1 different commits each, respectively15:28
and nothing to commit15:29
and i dont see the changes i was expecting from the merge15:29
← kpreid left15:29
← plaugg left15:29
← jutaro left15:29
← koo4 left15:29
→ ProperNoun joined15:29
← gjoseph left15:30
SethRobertson luxurymode: What does `git show HEAD^2` show? A patch hopefully? Does the result of the patch appear in your code?15:30
→ gjoseph joined15:30
gjoseph hey there15:30
← OOPMan left15:31
luxurymode SethRobertson, HEAD^2?15:31
eh?15:31
SethRobertson Quite so15:31
luxurymode fatal: ambiguous argument 'HEAD^2': unknown revision or path not in the working tree.15:31
gjoseph how can i tell with git tag -l (or otherwise) if a tag is annotated or lightweight ? (other than by analyzing and guessing if the message is a tag's or a commit's with -n)15:31
→ eean joined15:31
pikpik FauxFaux: Never mind. I must've been using 128 MB of memory when I had problems.15:32
→ khmarbaise joined15:32
SethRobertson luxurymode: Hmm. It does feel like the merge didn't happen.15:32
→ keystr0k joined15:32
→ LouisJB joined15:32
pikpik FauxFaux: My system's memory did get very low this time, but it scrapped by (which isn't great, but I don't know of a solution).15:32
→ havocologe joined15:32
luxurymode SethRobertson, when i did the merge --no-commit it told me that there was a conflict where one file was deleted in origin and modified in HEAD and it left the HEAD file in tree15:32
→ Aikawa joined15:33
→ beatak joined15:33
luxurymode SethRobertson, and then i didnt actually do anything when it said that. Instead i just did a CHECKOUT of head on the file15:33
whoops15:33
← eean left15:33
luxurymode so try the merge again and just resolve the conflict?15:33
SethRobertson gjoseph: If you `git show` the tag, you will get information about the tag in the annotated case, but only the commit in the non-annotated case15:33
pikpik FauxFaux: This was during the fetch and checkout stages of an initial init, fetch, branch, and checkout series.15:34
luxurymode SethRobertson, also if i try a merge now it says "fatal: You have not concluded your merge (MERGE_HEAD exists)."15:34
cmn I think git cat-file -t refs/tags/whatever should tell you15:34
havocologe if i do a git pull /path/to/my/cloned/repository.git then i get an error: "fatal : not a git repsitory" but it really is a repo15:34
luxurymode SethRobertson, so how do i complete it...15:34
?15:34
cmn gjoseph: ^^15:34
gjoseph SethRobertson: that's if i know the tag15:34
← stoffus left15:34
SethRobertson luxurymode: git merge --abort15:34
cmn git tag tells you what tags there are15:34
SethRobertson gjoseph: Perhaps you can explain your desired end goal15:35
gjoseph i was hoping to be able to list them and get some symbol or so next to their name to tell me which ones are annotated15:35
luxurymode SethRobertson, ok..now what? :)15:35
gjoseph SethRobertson: but really here i wanted to see what kind of tag the tool i'm trying to use just did15:35
← nardev1 left15:35
SethRobertson luxurymode: Now restart the merge15:35
gjoseph so git show should do15:35
luxurymode same way with --no-commit?15:35
→ steveoliver joined15:36
SethRobertson luxurymode: Doesn't matter too much. You can always restore it in a second commit.15:36
→ brngardner joined15:36
← steveoliver left15:36
→ steveoliver joined15:36
diverdude YEEEAHAAA GITORIUS IS WORKING :D15:36
→ brian_g joined15:36
luxurymode ok, so now auto merge fails. so if it just left deleted file in tree, i can simply add it to mark as resolved?15:36
SethRobertson Seems likely15:37
luxurymode git status just shows "unmerged paths and… deleted by them: path/to/file"15:37
← xjonc left15:37
gjoseph is there a switch for git show to *not* show the diff/commit ?15:38
FauxFaux gjoseph: git cat-file commit $COMMIT15:38
→ fisted joined15:38
wereHamster gjoseph: git log -115:38
or git log <commit>^!15:38
luxurymode SethRobertson, so everything that could be auto-merged was merged except this? or do i need somehow complete the merge now after resolving the conflict?15:38
gjoseph wereHamster: i don't think that'll work since i'm (at the moment) looking into using git show with tags15:38
→ kpreid joined15:39
wereHamster gjoseph: so what do you want to get?15:39
just the tag message?15:39
gjoseph everything else ?15:39
and the metadata ? tagger: etc15:39
FauxFaux git show --shortstat. ¬_¬15:39
wereHamster gjoseph: git log with a custom format?15:39
← Spockz left15:39
gjoseph FauxFaux: that was it thx! didn't see that in the man...15:39
wereHamster ah, maybe that won't work15:39
SethRobertson luxurymode: The man git-merge page says: Resolve the conflicts. Git will mark the conflicts in the working tree. Edit the files into shape and git add them to the index. Use git commit to seal the deal15:40
gitinfo luxurymode: the git-merge manpage is available at http://jk.gs/git-merge.html15:40
→ spearce joined15:40
luxurymode SethRobertson, great. thanks very much for your help. and your patience :) really appreciate it15:40
← fisted_ left15:41
← jacobat left15:41
→ scelis joined15:41
← scelis left15:42
→ scelis joined15:42
→ swombat joined15:42
→ joch_n joined15:42
→ airborn joined15:43
→ koo4 joined15:43
← mandric left15:43
→ alex__c2022 joined15:43
→ demi` joined15:44
← flijten left15:45
← Shusshu left15:45
→ Spockz joined15:45
→ ramsey joined15:45
→ neurodrone joined15:46
← robde left15:47
→ kevwil joined15:48
← Guest1669 left15:48
→ robde joined15:51
← papegaaij left15:51
→ papegaaij joined15:52
← tomislater left15:53
← shruggar left15:53
→ gianlucadv joined15:53
→ shruggar joined15:53
→ zivester joined15:53
← y3di left15:53
→ weiyang joined15:54
→ v0n joined15:54
VladimirZ SethRobertson, from 'git ls-remote origin' there is no such branch any more. But I don't know why I still see it in that Gui :/15:54
← JackWinter left15:54
SethRobertson Cannot help you there, sorry15:54
← angelsl left15:54
← zivester left15:55
weiyang could git clone --mirror clone all the branch from remote?15:55
→ zivester joined15:55
diverdude i wonder hwy github isn't opensource15:56
*why15:56
cbreak-work because they want to earn money with it15:56
weiyang they want to earn money...15:56
← zivester left15:56
← kpreid left15:56
→ zivester joined15:57
← robde left15:57
diverdude but they can still earn on hosting15:57
← jetienne left15:57
diverdude if its opensource15:57
cmn the web app is their secret sauce15:58
diverdude hehe yes it is15:58
← teweWORK left15:58
cmn and hosting probably isn't even the main income earner15:58
or maybe it is, but that's not that important anyway15:58
diverdude i wonder what the main income is15:59
cmn you can ask on #github15:59
→ johnanderson joined15:59
cmn but many companies do want to have github locally, so they pay the licence15:59
Chani damnit. I ended up with noting but one merge commit, with a terrible commit message >.<15:59
← herdingcat left15:59
← Chani left16:00
→ robde joined16:00
diverdude its probably also a big work making such a webapp16:01
← papegaaij left16:01
→ EyesIsServer joined16:01
→ JackWinter joined16:01
→ y3di joined16:01
← y3di left16:01
→ papegaaij joined16:02
← robde left16:02
← papegaaij left16:02
gjoseph silly and probably not very useful, but something like this is what i was hoping for initially (albeit with a simpler command :D) git for-each-ref --sort=subject refs/tags | awk '{print $2, $3}' | sed 's,tag refs/tags/, * ,' | sed 's,commit refs/tags/, ,'16:03
→ papegaaij joined16:03
← dorisabayon left16:03
← heisenmink left16:04
→ Fandekasp joined16:04
← kevwil left16:05
→ ScottE joined16:05
FauxFaux gjoseph: git branch; git tag | sed 's/^/ * /' # ?16:05
gjoseph FauxFaux: in my example, the * denotes annotated tags, where as lightweight ones are just prefixed by space16:07
← gianlucadv left16:07
FauxFaux Oh.16:07
← mercutio22 left16:07
→ robotmay joined16:12
← cbreak-work left16:12
← papegaaij left16:12
→ kerrick joined16:13
← Aikawa left16:14
← cyfi left16:14
← brian_g left16:15
← dr0id left16:16
→ n8o-mba joined16:16
→ Codif joined16:17
← jamiejackson left16:17
→ spvensko joined16:18
→ ehsan joined16:18
→ mklappstuhl joined16:19
← scelis left16:19
← robotmay left16:19
← Internet13 left16:19
→ sureaint joined16:20
→ Internet13 joined16:20
→ ddkilzer joined16:20
weiyang how could i clone all the branches from a repo?16:21
FauxFaux weiyang: That's the default behaviour. git branch -avv16:22
→ scelis joined16:22
→ robotmay joined16:22
→ mercutio22 joined16:22
weiyang FauxFaux, really?16:22
FauxFaux, I use git clone, but see there are many other branches16:22
SethRobertson They are remote tracking branches, but you have all of the data16:22
FauxFaux weiyang: Perhaps if you ran what I said you'd see what I mean.16:23
weiyang FauxFaux, let me try again16:23
→ icrazyhack joined16:24
SethRobertson What is there to try again? Just run the command he told you to.16:25
twisted` any Xcode users here?16:25
← pro_metedor left16:25
FauxFaux twisted`: !question16:25
gitinfo twisted`: You know, you have not actually asked a question. Without asking a question, it is hard to get any responses. Please explain what you are trying to do, what you typed in, and exactly what came back. Use http://pastebin.com for exact command sessions.16:25
FauxFaux Is there a good writeup of the fact that remote tracking branches exist, that we can link people to, btw? I can't think of one.16:26
twisted` FauxFaux: technically, it is a question :D16:26
I'm having problems with git and xcode16:26
FauxFaux I'd kick you if I had rights.16:26
twisted` or well, xcode and git, the combo if it :D16:26
← batlock666 left16:26
← sykop5am left16:26
twisted` but, seeing how this channel doesn't neccesarily have anything to do with Xcode it's a normal question imo16:26
impulze still no question16:26
SethRobertson FauxFaux: http://progit.org/book/ch3-5.html is OKish16:27
cmn is it telling you that file doesn't exist?16:27
FauxFaux Seriously, do people like you not actually want help?16:27
→ _iron joined16:27
FauxFaux SethRobertson: The first paragraph does a better job than I probably would have done. :)16:28
twisted` in Xcode I've added a remote repository, it's now (afaik) completely empty cause it has just been made. Now I want to commit/push my whole local git project to the remote server.16:28
← ajaykools left16:28
twisted` I do this using the Source Control -> Push method16:28
however, it's done within 2s and I can't find anything in the remote repo in my Organizer16:28
weiyang FauxFaux, i run the command and shows,16:28
impulze did you commit yet?16:28
twisted` so either I'm doing something wrong (wrong config) or something16:28
→ saml joined16:28
weiyang remote/origin/next16:28
saml how can I undo changes of one file?16:28
weiyang remote/origin/upstream16:28
← senny left16:29
← arvind_khadri left16:29
FauxFaux saml: git checkout -- path/to/file # will destroy your changes for you.16:29
← Spockz left16:29
saml ah. tahnks. iw as doing git reset --hard HEAD16:29
FauxFaux There's basically no reasons to use reset unless git specifically tells you to.16:29
← kaushik_ left16:29
cmn twisted`: it might be using the matching rules, and as nothing matches in an empty repo, it's not pushing anything16:30
weiyang saml, git reset --hard HEAD will reset all the file16:30
← akosikeno left16:30
saml yup16:30
→ DaveIngram joined16:30
twisted` cmn: hmm... I just also read something about creating a new project, specifiying the remote git repo, do a checkout and then move my files into that new project16:31
weiyang FauxFaux, but I can't use git checkout remote/origin/next16:31
→ kaushik_ joined16:31
cmn and did it help or something?16:31
FauxFaux weiyang: Type "git checkout next".16:31
cmn you'd have to look at the xcode docs to see if it'll let you push out one branch16:31
(or --all if you prefer)16:31
twisted` cmn: hmm ok16:32
cmn or just use the command line16:32
→ fermianyon joined16:32
SethRobertson weiyang: Why not? !repro16:32
gitinfo weiyang: please paste (e.g. using http://git.pastebin.com/) a transcript of your terminal session in which the problem occurred so that we can troubleshoot more easily.16:32
weiyang FauxFaux, oh...16:32
→ jamiejackson joined16:32
← sureaint left16:32
impulze oopsi16:32
SethRobertson weiyang: You can of course checkout origin/next (or remotes/origin/next) you just cannot commit to it since it is a detached head16:33
→ brian_g joined16:34
→ mando joined16:35
← spearce left16:35
→ papegaaij joined16:35
→ Yuuhi joined16:36
→ mandric joined16:37
← giallu left16:38
weiyang SethRobertson, http://pastebin.com/chUNRqsh16:38
SethRobertson, maybe i use the wrong command16:39
SethRobertson, so I should use "git checkout next " instead of "git checkout remote/origin/next"?16:39
← brainbox1100 left16:40
SethRobertson weiyang: It depends on what your goal is. What is your goal?16:40
What do you want to do to the branches16:40
→ michaelh joined16:40
twisted` cmn: yeah just tried the commandline: git push reponame master16:40
and I get back: Everything up-to-date16:40
← shruggar left16:40
weiyang SethRobertson, I am asked to compile the code for a specific branch16:40
→ shruggar joined16:41
weiyang SethRobertson, not the master branch16:41
SethRobertson weiyang: Then `git checkout origin/old-master; make`16:41
→ john2496 joined16:41
SethRobertson You do not need a local tracking branch (like next and master are) in order to check out a remote's branch.16:41
← Codif left16:42
← bauruine left16:43
→ akosikeno joined16:43
weiyang SethRobertson, not work, please see http://pastebin.com/BTeVLUa616:43
cmn twisted`: if git says so, then it's probably right16:43
twisted` I know my internet is fast but it cannot have sent 2.3M of files in less then 1s16:44
→ alvaro_o joined16:44
wereHamster twisted`: 'git checkout old-master'16:44
cmn git ls-remote reponame16:44
wereHamster with a recent enough git16:44
← alvaro_o left16:44
twisted` 023065515d5cc8a84d79ec36ae93de04b27608b5HEAD16:45
023065515d5cc8a84d79ec36ae93de04b27608b5refs/heads/master16:45
SethRobertson weiyang: It did work. git is just warning you that you cannot effectively commit.16:45
twisted` is all there is according to it16:45
cmn git rev-parse master should also give you that hash16:45
twisted` yep16:45
cmn then everything's fine16:46
→ gianlucadv joined16:47
→ SeySayux_ joined16:47
weiyang SethRobertson, then what is the difference between "git checkout old-master" and "git checkout origin/old-master" ?16:47
← jceb left16:47
wereHamster weiyang: one detaches head, the other doesn't16:48
SethRobertson `git checkout old-master` will, if old-master does not exist, create a local tracking branch tracking old-master to the remote tracking branch origin/old-master (which in turn tracks the origin's old-master branch)16:48
SeySayux_ → SeySayux16:48
→ cylence joined16:48
SethRobertson weiyang: You can commit to old-master, since that is your local branch. You cannot commit to origin/old-master since it is (your copy of) the remote's branch. So git puts you in a detached head16:48
The working directory will be identical in both cases16:49
weiyang SethRobertson, this makes me more clear16:49
SethRobertson Of course, the process of keeping old-master up to date is a little more complicated than keeping origin/old-master up to date. Not too bad16:49
← LongBeach left16:49
SethRobertson weiyang: Perhaps you should get some opaque makeup. The invisible man showed that being more clear isn't all that you might think it is.16:50
twisted` wtf!!!!16:50
← spvensko left16:50
twisted` it actually commited everything16:50
amazingly quick16:50
weiyang SethRobertson, your last sentence makes me confused.16:51
← brngardner left16:51
SethRobertson weiyang: Don't worry about it. I was poking fun at your language usage by making a reference to http://en.wikipedia.org/wiki/The_Invisible_Man16:51
→ guns joined16:52
weiyang SethRobertson, at least I get the code of a specific branch. Thanks!16:52
oh...16:52
let me seee16:52
← kaushik_ left16:52
→ kaushik_ joined16:53
← johnathanb left16:53
cmn twisted`: you mean pushed? committing should always be very fast16:53
→ robde joined16:54
← robde left16:54
→ robottinosino joined16:54
cmn the 2.3M were probably pushed the first time when you weren't paying attention and you've been doing no-ops since16:54
← n8o-mba left16:54
→ gdmfsob joined16:55
← brian_g left16:55
→ jaequery joined16:55
→ johnathanb joined16:55
← mishok13 left16:56
← Fandekasp left16:56
← weiyang left16:56
twisted` cmn: well, I added the remote branch to Xcode and it was complaining about cannot doing a push cause not everything was comitted locally and once I did I went to Push to remote and it was *poof* done16:56
but no matter...16:56
← robotmay left16:56
→ sureaint joined16:56
twisted` reminds me of compiling a Linux kernel on an insanely fast machine and it was done *too* quick (imo) so something *had* to be wrong ;)16:57
→ drev1 joined16:57
→ joshteam joined16:58
→ pidus joined16:58
← VladimirZ left16:59
→ MiiGo joined17:00
→ brian_g joined17:00
→ DofLeFou joined17:01
← brian_g left17:02
← whitman left17:02
→ AmirBehzad joined17:02
← ipalaus_ left17:02
→ brian_g joined17:03
→ brngardner joined17:03
→ zi joined17:05
← mklappstuhl left17:05
binjured is there a simple one-liner to tell me if my working directory is clean?17:06
SethRobertson git status?17:06
← papegaaij left17:07
← yshh left17:07
→ spearce joined17:08
→ alberto56 joined17:08
← zivester left17:09
← RaceCondition left17:09
→ iamjarvo joined17:10
← centipedefarmer left17:11
← kaushik_ left17:11
← Prasad left17:11
← Element9_work left17:12
→ Know1edge1 joined17:13
gitinfo set mode: +v17:14
← jatt left17:14
Know1edge1 whats the best way to get the contents of a remote repo but not get the .git data...like an svn export17:14
cmn archive --remote17:15
→ Element9_work joined17:15
→ kenichi joined17:15
Know1edge1 ah thanks perfect17:15
thanks cmn17:15
→ dzonder joined17:15
← iamjarvo left17:16
→ tewecske joined17:16
← t4nkd left17:17
→ iamjarvo joined17:17
← iamjarvo left17:17
← Vile left17:18
→ marthinal_ joined17:18
→ vjacob joined17:18
← Element9_work left17:18
→ Vile joined17:18
vjacob hello all17:19
→ Heisenmink joined17:20
← marthinal left17:20
→ jaalto joined17:20
→ Targen joined17:21
→ lugzer joined17:21
← tcurdt left17:21
jaalto Would anyone have idea why simple "git clone" doesn't clone all branches? http://pastebin.com/n44utH21 Or is it something I forgot?17:22
→ Voziv joined17:22
← DaveIngram left17:23
→ zivester joined17:23
vjacob any of you use egit and know how to relocate/or "refacture" the location of a repository in terms of local directory?17:23
zi Why doesn't git work, when I set default shell to /bin/false?17:24
PerlJam jaalto: looks like it cloned all *local* branches to me17:24
← mando left17:24
→ DaveIngram joined17:24
jaalto PerlJam: Why did it not clone remotes/origin/pristine-tar ?17:24
← thannoy left17:24
← joch_n left17:25
jaalto PerlJam: Never mind. I get it now.17:25
PerlJam, obvious, yes. Thanks.17:25
→ joch_n joined17:25
→ macmartine joined17:26
← joch_n left17:26
← dzonder left17:27
→ maletor joined17:27
← gnufied left17:27
→ kennethreitz joined17:27
← soulcake_ left17:28
← marthinal_ left17:28
vjacob so like moving the location of the repositories (multiple) from C:\repositories to X:\git-repos and updatin eclipse/egit config accordingly?17:28
← DaveIngram left17:29
→ bindaasomatic joined17:30
← gianlucadv left17:33
← LimeBlast left17:33
→ marthinal joined17:34
→ friskd joined17:34
← friskd left17:35
→ russ joined17:35
← jajmon left17:36
→ cybersphinx joined17:36
← russ left17:36
→ mgpcoe joined17:36
→ churp joined17:37
→ rustifer joined17:37
ChanServ set mode: +v17:38
→ friskd joined17:38
→ ipalaus_ joined17:38
← the_cybersphinx left17:38
lugzer Soda theme no good anymore, new update is breaking the "modified" flag in a tab17:39
→ ziggles joined17:39
lugzer probably Soda needs an update ;)17:39
→ kenperkins joined17:39
EugeneKay Cool story bro.17:40
lugzer EugeneKay, all - sorry, wrong chan! :/17:40
EugeneKay ;-)17:40
← padams left17:41
← sureaint left17:42
rustifer So, came in to ask a question about setting up a work flow. My normal process is to get some files together, do a git init, set up an unfuddle repository, add a remote master, do a git push.17:43
← ph^ left17:43
→ Gorkyman joined17:43
rustifer I was just handed a zip archive of someone's work dir that already includes .git and configuration17:43
→ sureaint joined17:43
rustifer I'm sure I could just rip it out and set up a new repo but that seems counterproductive.17:43
EugeneKay Unzip & use.17:44
wereHamster NO, THAT'S TOO EASY!17:44
rustifer Don't be silly. I don't have credentials for that other repo.17:44
EugeneKay You'll probably need to fiddle the remotes and such17:44
wereHamster rustifer: of course you have. It's yours17:44
← Gorkyman_ left17:44
wereHamster the zip archive is yours, correct?17:44
rustifer The zip archive was handed to me by the client17:45
wereHamster rustifer: but you now do have access to all the files within it17:45
rustifer It has their .git stuff.17:45
wereHamster so what?17:45
you have read and write access to everythign within the zip if you extract it, right?17:45
← alansaul left17:45
wereHamster so first step: extract the archive somewhere17:45
← lonewulf` left17:45
rustifer Yes, yes, but we keep all our repos at unfuddle.com, less of a distributed model.17:45
→ patrick99e99 joined17:45
wereHamster step 2: cd into that directory, and push the code into your unfuddle repository17:46
→ alester joined17:46
rustifer ... could it be that simple?17:46
← dreiss left17:46
→ disappea_ joined17:46
→ kaushik_ joined17:46
rustifer I'm overthinking the problem, aren't I?17:46
wereHamster you know.. distributed and that stuff..17:46
→ fir31ight joined17:47
rustifer You've been very reassuring. Thanks!17:47
→ tomislater joined17:49
← Vampire0_ left17:49
← VladGh left17:49
← ASIRINeiluj left17:50
→ billy_ran_away joined17:51
← sureaint left17:51
← achadwick left17:51
billy_ran_away Is there a way to revert a single file from a commit?17:51
→ Araxia_ joined17:51
FauxFaux billy_ran_away: git checkout THATCOMMIT~ -- path/to/file will grab the version of the file from the previous commit.17:52
billy_ran_away And how can I get a listing of merge commit's parents? I.e. which parent was 1 and 2?17:52
FauxFaux: Awesome, why didn't I think of that...17:52
← kaushik_ left17:52
billy_ran_away FauxFaux: Thanks!17:52
→ delight joined17:54
wereHamster billy_ran_away: <commit>^1 is first parent, ^2 second parent and so on17:55
FauxFaux billy_ran_away: And "git show" shows the merge parents, in order.17:55
→ tystr joined17:56
billy_ran_away FauxFaux: Okay cool17:56
← marthinal left17:56
billy_ran_away wereHamster: THanks17:56
← csgeek left17:57
← lugzer left17:57
← LouisJB left17:58
← magnus___ left17:59
← shruggar left17:59
← overbythere left17:59
← vjacob left17:59
→ csgeek joined17:59
← disappea_ left18:01
→ marthinal joined18:01
← nikuyoshi left18:02
→ iamjarvo joined18:04
→ lobolife joined18:05
→ davidcoallier joined18:05
→ Spockz` joined18:06
← ipalaus_ left18:06
← scelis left18:06
iamjarvo say there is a branch called new-design and then there is a staging branch, then there is feature branch i am working on. i should merge my branch into new-design and then merge new-design into staging ? question about process18:06
→ ipalaus_ joined18:06
→ Vampire0_ joined18:07
→ kpreid joined18:09
→ apok joined18:09
→ kevwil joined18:09
← rockfruit left18:10
→ codeshah joined18:10
→ t4nkd joined18:12
tizzo → tizzo-afk18:12
→ secoif joined18:14
→ dvide_ joined18:17
→ spvensko joined18:17
← dvide left18:19
← spvensko left18:20
→ aldodelgado joined18:20
← pikpik left18:20
← JEEB left18:22
→ Goplat joined18:22
tizzo-afk → tizzo18:23
→ Chillance joined18:23
→ LouisJB joined18:23
← brian_g left18:24
← brngardner left18:25
→ gianlucadv joined18:26
← abetaha left18:26
→ dreiss joined18:27
→ Topcat joined18:28
Voziv When I do a git push I get the error "! [rejected] HEAD -> refs/for/refs/heads/master (non-fast-forward)" If I do a "git push origin master" it works fine. Any idea why it tries to push HEAD -> refs/for/refs/heads/master by default?18:28
→ fr0sty|mobile joined18:28
steveoliver Should I be able to just pop a .gitmodules file into a repo and git submodule init to get them started up (checked out)?18:28
Topcat is there an equivilent to git add . like git rm . that won't empty my whole project18:28
steveoliver *a populated .gitmodules file, of course18:29
thiago Voziv: did you configure anything in .git/config?18:30
← fr0sty left18:30
Voziv Me personally, no, but I do see the line "push = HEAD:refs/for/refs/heads/master" under my remote18:31
Sorry, in .git/config under the remote section18:32
→ tanob joined18:33
→ nadirvardar joined18:33
→ JEEB joined18:34
← babilen left18:35
tanob hey guys, sometimes I'm getting failures on "unpack-objects" when doing a `git fetch`: https://gist.github.com/1602271 Any idea on why is this happening or how to debug/fix?18:37
matthiaskrgr I'm not a git dev, but did you try checking if /tmp/ is full ?18:38
→ DarkAR joined18:39
→ aspotashev joined18:39
tanob matthiaskrgr: actually no, let me check that18:39
← mugwump left18:39
matthiaskrgr :)18:39
→ Gorkyman_ joined18:40
→ VladGh joined18:40
← psakrii left18:41
← Gorkyman left18:42
← canton7 left18:43
← kevwil left18:43
→ kevwil joined18:43
steveoliver I'm trying to take a populated .gitmodules file from one existing repo and copy it into another, effectively adding all the same submodules to the second repo... init and update don't do anything, and status doesn't show any registration of the submodules.... Ideas?18:43
← cortana2 left18:45
→ abetaha joined18:45
tanob matthiaskrgr: /tmp isn't full18:45
matthiaskrgr mh18:45
meh, it was just an idea :P18:45
tanob but I see that the repo is broken, I can't run `git status`, it says "unable to read tree …"18:46
matthiaskrgr I'd try to run git gc or git fsck and try again no idea if that helps18:46
GNUcifer tanob: maybe permissions?18:46
← ahmedkotb left18:47
SethRobertson steveoliver: You also need to `git add` the submodule at the correct SHA checkout.18:47
GNUcifer or filesystem errors ..18:47
← ziggles left18:48
→ brian_g joined18:48
← menace left18:48
→ Suhana joined18:48
← empity left18:48
SethRobertson steveoliver: submodules consist of three bits of state. .git/config, .gitmodules, and the committed work tree SHA state. The last two are committed state.18:48
→ LouisJB_ joined18:48
← Spockz` left18:49
tanob GNUcifer: permissions are ok, just checked. About FS errors, how could I check that?18:49
← marthinal left18:49
GNUcifer how did you obtained that repo, git clone?18:49
← LouisJB left18:49
LouisJB_ → LouisJB18:49
cmn tanob: can you change the script so it runs on /home/ maybe?18:50
steveoliver SethRobertson: are you saying there's no real way to shortcut like I'm trying to? I've got to add each submodule manually?18:50
→ kahrl joined18:50
→ Prateek joined18:50
Prateek → Guest1595818:50
SethRobertson Yes and no. You can copy the .gitmodules and then you can copy the repos checked out at the right SHA into the correct paths, and then you can `git add` the correct paths. However, you cannot do it by just copying .gitmodules.18:51
tanob cmn: that is the output of a job in the project's CI server, I could copy the repo to the home directory and try to fetch from there, but I believe the repo is a broken state, right?18:51
GNUcifer: yes, the CI server did a git clone in the very first job it ran18:51
cmn maybe, maybe not18:51
the error comes locally18:52
GNUcifer dunno really .. needs some debugging, which CI do you use?18:52
cmn at least the output seems to come from rev-lit18:52
steveoliver SethRobertson: I think that makes sense18:52
← Guest15958 left18:52
← rustifer left18:52
tanob GNUcifer: Go18:52
cmn if the repop were corrupted on the remote, upload-pack should have complained18:52
can you cone it to your machien?18:52
← delight left18:53
← kahrl_ left18:53
tanob cmn: yes, we have the repo cloned in all dev machines and we don't have that issue locally18:53
GNUcifer maybe the ci does some cleanup it isnt supposed to do18:54
← robottinosino left18:54
tanob cmn: and yes, rev-list is crying "error: Could not read df4a53ec7432445ca73bc0ffe4c04c4857a89409"18:54
cmn well if you can clone on your machine, but the clone to /tmp on the CI fails, that's a strong hint that the problem is in whatever the CI does18:55
cloning to /tmp is probably not the best idea18:55
are you sure that you have enough space there? many distros put that in tmpfs, so you don't have that much storage space there18:55
Harzilein can i have git log show me the commits as HEAD~n tree-ishs?18:55
→ jceb joined18:55
tanob I could probably clone the repo again on the CI machine, I believe the repo got corrupted at some point in time18:55
cmn have you run 'git fsck' on the remote repo?18:56
tanob cmn: the remote repo is hosted at codebasehq18:56
← mercutio22 left18:57
cmn I've no idea what that is; some hoster?18:57
tanob cmn: and about the /tmp space, there isn't any mount point for /tmp, just "/dev/simfs on / type simfs (rw,relatime)", there are 6.2G free18:57
cmn: yep, it is like github18:58
→ OOPMan joined18:58
→ blueadept joined18:58
cmn OK, so it's not space, can you do a clone from your machine?18:58
← Vile left18:58
tanob cmn: yes18:58
blueadept how do you get the status of how many lines have been deleted and added before you commit ?18:58
→ Vile joined18:58
cmn then the CI is doing somthing wrong18:58
tanob could it get corrupt because of connection errors?18:58
blueadept like i do git status, but only see modified files18:58
cmn blueadept: you mean the --stat option to diff ?18:59
← lapistano left18:59
→ metcalfc joined18:59
← beatak left18:59
blueadept ah yes, cool18:59
→ lapistano joined18:59
blueadept thanks!18:59
→ detaer joined19:00
→ agliodbs joined19:01
← secoif left19:01
agliodbs what's the name of the base git package on ubuntu? I can't seem to find it19:01
FauxFaux git19:01
→ beatak joined19:01
tanob cmn: i've copied the repo to my machine and it is really corrupted19:02
← steveoliver left19:02
cmn agliodbs: if it's old, it might be git-core19:02
← OOPMan left19:02
agliodbs git-core19:03
cmn tanob: you mean the one that the CI failed to clone properly, right?19:03
agliodbs apparently, thanks19:03
→ mercutio22 joined19:03
tanob cmn: the error is fetching19:03
Harzilein anyone?19:03
cmn tanob: silly me, I've been thinking it was a clone19:04
tanob cmn: yes, I've scp'ed the repo from the CI server to my machine19:04
cmn maybe the tree that's failing was fetched at some point in the past19:04
→ LouisJB_ joined19:05
cmn and has since been corrupted by power failure, bad blocks or some such19:05
does git fsck report heavy damage?19:05
tanob haven't done it yet19:05
← jceb left19:05
← detaer left19:05
← gdmfsob left19:05
tanob lots of broken links and missing blobs19:05
→ detaer joined19:05
← detaer left19:06
→ crisp joined19:06
→ detaer joined19:06
→ marthinal joined19:06
← LouisJB left19:07
LouisJB_ → LouisJB19:07
cmn well, something blew up there19:07
the large amount of missing blobs might just be unpack-objects not finishing properly19:07
← cakehero left19:08
cmn if there were missing or corrupt objects and the server sent a thin pack, it would account for the breakage19:08
→ hwit joined19:09
→ jceb joined19:09
← vdv left19:10
← CrazyTux[m] left19:11
→ dorkmafia joined19:11
← ngulden left19:11
→ robde_ joined19:12
← kermit left19:15
→ Crewe joined19:15
← gjoseph left19:16
tanob cmn: i'm going to delete the existing cache repo and watch if this happens again in the future19:16
thanks guys!19:17
cmn that'll be the easiest option, yeah19:17
you might want to schedule a fsck on the server19:17
GNUcifer does the CI system run client commands (including git) under some restrictions (ulimits)?19:18
cmn git generally doesn't kill its own repo19:18
→ OOPMan joined19:19
→ MANCHUCK joined19:19
← faber left19:20
→ ocharles joined19:22
ocharles hey, does anyone know how to generate a patch from a repository against the very start of it?19:22
ie, dump the entire current tree as a patch file19:22
→ ericdc joined19:22
FauxFaux ocharles: git checkout --orphan OMGPONIES && git diff master19:22
malphaet dude19:23
← BlaDe^ left19:23
ocharles FauxFaux: naw, that doesn't seem to work19:23
→ btx94 joined19:24
→ cccaldas_ joined19:24
tanob GNUcifer: I'm not aware of that19:24
FauxFaux ocharles: Can probably make it work. --cached, rm .git/index, etc.19:25
→ Cromulent joined19:25
GNUcifer well there must be some reason, had a unexpected reboot or power failure?19:25
→ spvensko joined19:26
← spvensko left19:26
→ spvensko joined19:26
→ btanaka joined19:26
← spvensko left19:26
cmn git diff master would compare the worktree with the master branch, which is probably equal19:26
← marthinal left19:26
cmn git diff HEAD master should either work or blow up on your face19:26
ocharles right19:26
yrlnry Is there way to ask git-log to list commits that touch any file whose name has the form *.notyet, regardless of where such files appear in the tree ?19:27
← cccaldas left19:27
→ whyking_ joined19:27
cmn yrlnry: git log -- '*.notyet' should do it19:27
MANCHUCK is there a way i can turn a branch into a submodule?19:28
GNUcifer man git-filter-branch19:28
gitinfo the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html19:28
→ dankest joined19:28
← dankest left19:28
GNUcifer err .. that was for dirs .. branche (heads) can be submodules by just committing them as submodules19:29
← whyking left19:29
yrlnry cmn: Thanks, I see what happened. I tried it without quotes by mistake, and of course the shell expanded the * and so I got only the commits touching files in the current directory.19:29
cmn indeed19:29
GNUcifer uses that to have submodules as branches in the same repo as the main stuff19:29
GNUcifer its a bit tricky to set up19:29
→ spvensko joined19:30
MANCHUCK Well i have the CMS Expression Engine and a branch and Magento in the master I want to make the Expression Engine branch be in its own folder. I figured that converting that branch to a submodule then adding that submodule to my Magento branch will do the trick19:30
← spvensko left19:30
→ spvensko joined19:30
yrlnry Is there a way to get git-log's --stat option to mention zero-length files? It seems to omit them.19:30
→ babilen joined19:30
→ jetienne joined19:30
yrlnry I want to see which *.notyet files were touched in each commit, but each one is zero-length.19:31
Hmm, maybe dirstat.19:31
→ ManDay joined19:31
tanob GNUcifer: definitely there is a reason for this, it happened in 3 of 20 agents that we have in our CI system, but it is difficult to track this kind of issue19:31
cbreak touched?19:31
yrlnry created or deleted.19:31
→ mishok13 joined19:31
cbreak git log -- file?19:32
ManDay Does git create junk in my .git dir? I just stumbled across ".sample" file. What kind of crap is that? Does it also put a full copy of the documentation into every .git dir?19:32
cbreak or if you want to view from the other side, git log --name-only and then grep those you want to have19:32
ManDay: no19:32
yrlnry cbreak: I do git log --stat -- '*.notyet' and I want to see, for each commit, which "noetyet" files were created or deleted. But because they are all 0-length, --stat omits them19:32
→ nwp_ joined19:33
ManDay cbreak: Thank you, here is your award for most obvious answer of the evening.19:33
cbreak not obvious for you to answer it I guess :)19:33
but thanks19:33
GNUcifer --prefix=.git .. make install :)19:33
ManDay I was being sar-cas-tic...19:33
→ drizzd joined19:33
cbreak yrlnry: even --name-only?19:33
← crisp left19:34
yrlnry cbreak: --name-only is perfect, thanks.19:34
→ crisp joined19:35
← ajt left19:36
Harzilein so19:37
← mishok13 left19:37
→ mando joined19:37
→ maheshk_ joined19:38
Harzilein this might be common but i have not found it: i started with a history-less checkout (from last night). now there _is_ another repo whose history i would like to have in my branch (i already added a branch that tracks it). so now i have a history-less branch, a branch with the full history and a commit-id from shortly after my distribution tarball. how can i merge those?19:39
yrlnry Harzilein: you can add the other repo as a remote, and fetch it into the historyless repo.19:39
Then you can rebase the fetched history onto your branch.19:39
Harzilein yrlnry: as i said, i already added the repo for a new branch19:40
yrlnry Then you can rebase the fetched history onto your branch.19:40
Maybe I don't understand the question.19:40
Harzilein yrlnry: so now i have a branch with history and one without. i want all the history (and of course as i modified my version, do merges where necessary)19:40
oops19:41
→ pantsman joined19:41
← pantsman left19:41
→ pantsman joined19:41
cbreak history less?19:41
Harzilein all the history back to a certain commit19:41
→ mloskot joined19:41
cbreak you mean an --orphan?19:41
← mloskot left19:41
→ mloskot joined19:41
Harzilein historyless as in "without the remote history in it"19:41
it has my changes though19:41
← Cromulent left19:41
← tanob left19:41
cbreak git can not make history less clones19:41
one of the few ways to start new history is to git init a fresh repo19:42
Harzilein i did not clone it originally. i used a distribution tarball.19:42
bremner git-archive ?19:42
EugeneKay git clone --depth19:42
cbreak or to use something like git checkout --orphan19:42
EugeneKay It's a hack, but it /can/....19:42
Harzilein sorry?19:42
cbreak --depth doesn't give you that eugen19:42
Harzilein i think we are all confused19:42
cbreak it still has history19:42
you just don't have it local19:42
jetienne q. on macos, the fs ignore char cases. is there a way to force git to not ignore it ?19:42
EugeneKay Terminology, I spose19:42
proq am I correct in noticing that when I have uncommitted changes, then checkout a different branch, I can commit my changes to the branch I just switched to?19:42
cbreak if you git init a fresh repo from an archive19:42
← beatak left19:42
cbreak then you can just merge it19:42
or rebase onto the commit the archive was made form19:43
and then merge it19:43
Harzilein again19:43
cbreak (that's what I'd do)19:43
→ marthinal joined19:43
← pfsmorigo left19:43
EugeneKay jetienne - man git-config for "core.ignorecase"19:44
gitinfo the git-config manpage is available at http://jk.gs/git-config.html19:44
cbreak jetienne: git doesn't ignore case19:44
jetienne EugeneKay: Oh this exists ? too cool, i will read19:44
cbreak not even on mac os19:44
EugeneKay cbreak - I /think/ core.ignorecase might default to true on some OS Xes19:44
cbreak it all depends on the file system you use19:45
jetienne cbreak: macos fs does and this create significant issue with renaming19:45
EugeneKay (rather, git distros)19:45
proq mac os? no one has used that for 11 years19:45
← tgunr left19:45
← LouisJB left19:45
Harzilein so19:45
EugeneKay I'm lucky enough to not own any Apple products myself19:45
cbreak proq: there's a new version that's rather nice19:45
proq cbreak: no, there isn't19:45
9.2.2 isn't new or nice19:45
→ sontek joined19:45
cbreak you should look up version 2019:45
Harzilein i added the local "upstream" branch (which in turn has the upstream as remote) as upstream to my "ugly" branch19:46
jetienne proq is stuck in the 90s :)19:46
→ oso96_2000 joined19:46
sontek I need to pull a commit out of master and push it to a branch (its not as good as I thought when I pushed it)19:46
is this easy?19:46
proq jetienne: no, your lingo is19:46
Harzilein how can i now effect the rebase/merge to start with a certain commit id?19:46
jetienne EugeneKay: well thanks i will look19:46
proq: carefull my lingo is just behind you19:46
cbreak Harzilein: do what I proposed above:19:46
← jetienne left19:46
cbreak git rebase --onto the commit the archive was built from19:46
Harzilein and can i somehow make it incrementally? as in rebase until there is a conflict, merge, then continue adding history until the next conflict?19:46
proq it's called os x. "mac os" hasn't been updated in a decade19:47
→ beatak joined19:47
EugeneKay Apple Mac OS X 10.8 Giant Cat RC2 x64 Intel nocompat Hurd19:47
← bindaasomatic left19:47
cbreak Harzilein: why?19:47
Harzilein cbreak: again, i don't _know_ which commit my files are from. i have identified a commit where it would make sense for my histoy to be merged into theirs19:48
s/my history/my local changes/19:48
cbreak Harzilein: no version number or date/time?19:48
Harzilein no19:48
proq EugeneKay: is that supposed to prove something?19:48
cbreak try to find a commit that looks similar then. or just pick one.19:48
→ tgunr joined19:48
cbreak the closer it is to the one you had, the less likely there will be conflicts19:48
EugeneKay proq - that people are obsessed with names in the face of an obvious reality.19:48
cbreak Harzilein: due to your special circumstances, you probably want to use the --root flag for rebase too19:49
proq EugeneKay: Mac OS is an insecure operating system with no kernel. os x is based on BSD. welcome to 2012.19:49
EugeneKay: it has a different name because they have nothing in common19:49
EugeneKay proq - and it's GNU/Linux, not "linux". Yes, yes, whatever.19:49
cmn my MacBook claims to be running Mac OS version 1019:49
cbreak no kernel?19:49
Harzilein cbreak: that's what i did19:49
cbreak Harzilein: if you get a conflict, resolve it, git add the files, git rebase --continue19:50
ShooterMG → ShooterMG_lunch19:50
cbreak repeat until the branch is rebased completely19:50
after that you can merge it19:50
→ infogulch joined19:50
cbreak I'd advise against doing anything like merging during the rebase19:50
← tatsuya_o left19:50
Harzilein cbreak: basically i wrote a git-tarball.sh: TREEISH="master~$1" ; git archive "$TREEISH" > ../prvm-tarball || exit 1 ; shortcommit=`git get-tar-commit-id < ../prvm-tarball | head -c 12` ; mv ../prvm-tarball ../prvm-tarball-"$1"-"$shortcommit".tar19:50
← t4nkd left19:50
cbreak (because it'd introduce ahuge bit of new history with potential conflicts)19:50
Harzilein cbreak: then ran: for i in `seq 0 100` ; do git-tarball "$i" || break ; done19:51
cbreak: and after for i in `seq 0 7` ; do (cd prvm-check && rm -fr * && tar xf ../prvm-tarball-"$i"-* ; echo "$i" ; git diff --color=always ../prvm/ . ) ; done | less -R19:51
cbreak and what for?19:51
→ jerQ joined19:53
← amigojapan left19:53
← noisebleed left19:53
Harzilein cbreak: first i excluded those commits where i have files they don't have. then i searched for the first that makes sense (does not remove my files, has commits that look like the future, not the past)19:54
→ _macro joined19:54
→ t4nkd joined19:54
cbreak you could just use git log and look for a commit with the same tree hash as your original one19:54
Harzilein i'm not sure such a thing exists... how do i compute the tree hash from some unmanaged directory btw?19:55
← hwit left19:55
cbreak make it a git repository19:55
commit19:56
instant tree hash :)19:56
besides, didn't you say you had a git repository already?19:56
just take the tree hash of the initial commit19:56
← marmaladex left19:56
cbreak if they used git archive, and didn't filter out files, then you should get the same for the first commit...19:56
should.19:56
← jaequery left19:56
cbreak lots of things that can go wrong of course :)19:57
← kennethreitz left19:57
proq EugeneKay: odd notion. I don't refer to my ubuntu and debian systems as "linux"19:58
bremner guys, please, drop it19:59
← ipalaus_ left19:59
← MiiGo left19:59
→ dbdot joined20:00
→ tomislater2 joined20:00
← infogulch left20:00
← pidus left20:00
← swombat left20:00
← tomislater left20:01
← bartek left20:02
→ gjoseph joined20:02
→ infogulch joined20:02
→ vdv joined20:02
→ ASIRINeiluj joined20:03
← FernandoBasso left20:03
← spvensko left20:04
→ tsujiban joined20:04
→ bartek joined20:05
← Targen left20:05
← kevwil left20:06
→ kevwil joined20:06
→ brngardner joined20:07
← kevwil left20:07
→ devth joined20:07
devth there's no short flag for --amend is there? (getting tired of typing amend)20:07
ShooterMG_lunch → ShooterMG20:07
← khmarbaise left20:07
→ ipalaus_ joined20:07
EugeneKay man git-commit20:08
gitinfo the git-commit manpage is available at http://jk.gs/git-commit.html20:08
devth yeah. i read it.20:08
confirming there's no undocumented flag20:08
EugeneKay Doubt it. Feel free to abuse alias, though.20:08
devth k. will do20:08
dbdot i was using drupal features on a blog content type, more files than were expected show as changed20:08
← AmirBehzad left20:09
devth cia = commit -a --amend it is20:09
cmn did those files not change?20:09
EugeneKay calls the black helicopters20:10
dbdot the diff warning indicates crlf used vs lf, how to reset these files?20:10
→ LouisJB joined20:10
dbdot i can run them through dos2unix but they still show as changed20:11
devth EugeneKay: :D20:11
→ tatsuya_o joined20:11
← LouisJB left20:11
← ManDay left20:11
dwmw2_gone → dwmw2_LHR20:12
dbdot and yes i'm the onnly one in my group on wamp20:12
← lobolife left20:13
→ whitman joined20:14
→ boombatower joined20:14
EugeneKay Easy way? Don't use Windows.20:14
→ lobolife joined20:14
EugeneKay Painful way? man git-config; grep for crlf20:14
gitinfo the git-config manpage is available at http://jk.gs/git-config.html20:14
Harzilein cbreak: nopes, no luck. did git log --pretty=raw, copied the tree hash, switched to the branch w/ history, git log --pretty=raw | grep thehash20:15
dbdot lol heard that one a lot, stuck on this until I can afford to replace this one20:15
EugeneKay VMware Player ;-)20:15
cbreak Harzilein: maybe they filtered it first, or somehow an executable bit wasn't transfered20:15
← jpalmer left20:16
→ steffo joined20:16
dbdot could the repository be config'd for the windows crlf when i created it?20:16
← Vile left20:17
dbdot changes were made in drupal not outside at os20:17
→ twisted joined20:17
← tgunr left20:18
← twisted left20:18
EugeneKay dos2unix the files. ;-)20:18
cmn then drupal is messing with stuff, you can configure git to change the line endings20:18
cbreak use a proper text editor20:18
← twisted` left20:18
← gianlucadv left20:18
jast dos2unix ALL THE FILES20:18
→ twisted` joined20:19
← aldodelgado left20:19
→ aldodelgado joined20:19
dbdot thanks all20:19
kevlarman jast: heh20:19
← zivester left20:20
← aldodelgado left20:20
← dbdot left20:20
← kukks left20:20
← axl_ left20:21
→ axl_ joined20:21
→ dep3ndant joined20:21
cbreak dos2unix can destroy files that are not dos text files :/20:21
proq is there an easy way to merge all changes in a branch into master?20:21
cbreak yes.20:22
git checkout master && git merge branchnamehere20:22
→ AAA_awright_ joined20:22
proq thanks20:22
→ tgunr joined20:23
← AAA_awright left20:24
→ canton7 joined20:24
dwmw2_LHR → dwmw2_gone20:25
Harzilein cbreak: i don't understand --root20:25
→ j416 joined20:25
saml hey, git mv good?20:25
← cylence left20:25
→ c0_stawberryy joined20:25
→ CE_BISPAK_SMG joined20:25
→ Emo_Princezz joined20:25
→ woww joined20:25
→ artihidup joined20:25
→ [[Nanda_sweet]] joined20:25
cbreak Harzilein: makes you rebase the whole branch20:25
← Emo_Princezz left20:26
← [[Nanda_sweet]] left20:26
→ sair_ah joined20:26
→ ce_bg02_an joined20:26
→ ce_MuTz joined20:26
→ Cewe_sexI_cantIk joined20:26
→ phil\ joined20:26
→ DimaS40 joined20:26
→ PEJANTAN_TANGGU joined20:26
→ rise_up joined20:26
→ loneguy joined20:26
→ Gordy joined20:26
→ dimana joined20:26
→ COWOK_MANADO joined20:26
→ citronade03 joined20:26
→ Cwo_caRi_tmen joined20:26
→ cew_juaal_haandp joined20:26
→ CIBUBUR_CR_TEMAN joined20:26
→ co_jkt_27_chn_ joined20:26
→ Black_Boss joined20:26
→ exmud_bthceww_4f joined20:26
→ Op-Gaul-Game joined20:26
→ BigEars joined20:26
→ Hiicham joined20:26
→ JustinJ joined20:26
→ kept joined20:26
→ Rhei joined20:26
→ tasikcrew joined20:26
→ eno_ce_loencha joined20:26
→ co`jomblo`esia joined20:26
→ zZzZ_bOys joined20:26
→ wanita_dewasa_bt joined20:26
→ LeD|oC joined20:26
→ stiffmeister joined20:26
→ wolf31r-s joined20:26
→ cari_agen joined20:26
→ PRIA^DEWASA joined20:26
→ ce-debate joined20:26
→ WAPPO joined20:26
→ deep_ joined20:26
→ harry joined20:26
→ c0_fb joined20:26
→ Faisal_36 joined20:26
→ AanG___ joined20:26
→ cwokK-pngen-susu joined20:26
→ CEWE_NDUTZ_BOUKE joined20:26
→ c0_baek joined20:26
→ co_mw_ML joined20:26
→ cow_t4_curhat joined20:26
→ Oki_jie joined20:26
→ Chn_Co_JKt_LIke_ joined20:26
→ man_romantic joined20:26
→ yuke_07 joined20:26
→ StePH[G] joined20:26
→ Aimen joined20:26
→ beaublack joined20:26
→ co_kes3p1an joined20:26
→ RiE_aLOnE joined20:26
→ joetherich joined20:26
→ easly joined20:26
→ KlausFuchs joined20:26
→ Ce_H1N1 joined20:26
→ MarcAndreQc joined20:26
← Cewe_sexI_cantIk left20:26
→ cari_ce-eksibisi joined20:26
→ AnakAyam joined20:26
harry → Guest2696020:26
← loneguy left20:26
← CIBUBUR_CR_TEMAN left20:26
← AanG___ left20:26
← artihidup left20:26
← sudoquai left20:26
→ co_co_co joined20:26
→ CoWw__Cr__tmnCE joined20:26
→ rHya3 joined20:26
→ seSha joined20:26
→ ekki joined20:26
→ Bluey_RoSe joined20:26
→ cO_gOgilzzz joined20:26
→ NoHoper joined20:26
→ buns0til_25 joined20:26
→ Fun_jkp joined20:26
→ IcHanTiQue joined20:26
→ cow_suka_cow joined20:26
→ PijatSensualUtkW joined20:26
→ Adiib joined20:26
→ Aldana_32Cal joined20:26
→ aaderaai joined20:26
→ georges70 joined20:26
→ steven28t joined20:26
← aaderaai left20:26
← BigEars left20:26
← beaublack left20:26
← co_jkt_27_chn_ left20:26
← c0_baek left20:26
← Aimen left20:26
← ce_MuTz left20:26
← kept left20:26
← Chn_Co_JKt_LIke_ left20:26
← Hiicham left20:26
← cwokK-pngen-susu left20:26
← Aldana_32Cal left20:26
← eno_ce_loencha left20:26
← JustinJ left20:26
← Gordy left20:26
← Black_Boss left20:26
← PEJANTAN_TANGGU left20:27
→ Cwok_dws_ajak joined20:27
→ nurse_akbeth joined20:27
← Dragooon left20:27
cmn saml: good in comparison to what?20:27
→ Assuradda joined20:27
→ cwo_ngajak_Nonto joined20:27
→ andreas_mks joined20:27
→ muthia joined20:27
→ sallyna joined20:27
→ co_cakepz_sby joined20:27
→ moccacino joined20:27
→ DinDa_aZa joined20:27
→ yuga joined20:27
→ satria_pingit joined20:27
→ jancok joined20:27
← Fun_jkp left20:27
← exmud_bthceww_4f left20:27
← Adiib left20:27
← jancok left20:27
→ CIBUBUR_CR_TEMAN joined20:27
→ Pryde joined20:27
→ LfcGirl joined20:27
→ gund joined20:27
→ prjonzy87 joined20:27
→ pretty_Nice joined20:27
→ BOLIHUANGGA joined20:27
→ naq joined20:27
→ Cow_kece joined20:27
→ junky[\[a] joined20:27
→ cE_imOoet joined20:27
→ louis_43 joined20:27
→ Co__cr__serius joined20:27
→ Guest89457 joined20:27
→ Hot_Date joined20:27
→ pim joined20:27
→ tara_aja joined20:27
→ shady_777 joined20:27
→ siNgLeLady79 joined20:27
→ maria_dpk joined20:27
→ joni_adams joined20:27
→ sukroo joined20:27
→ ZOE`S joined20:27
→ Bnt_aBohaa joined20:27
→ cr_cewk joined20:27
→ DoGGy_StyLe^ joined20:27
→ Adi_jogja joined20:27
→ GOODFatheR7 joined20:27
→ Leo_Lee joined20:27
→ nee_a joined20:27
→ meLL_InBndung_ju joined20:27
→ Sushmita joined20:27
→ Invierno joined20:27
→ dirka joined20:27
→ cantikz_maniz_im joined20:27
→ agas joined20:27
→ h26 joined20:27
→ co_cr_kepuasan joined20:27
→ caleb32 joined20:27
← Sushmita left20:27
← meLL_InBndung_ju left20:27
← siNgLeLady79 left20:27
← Leo_Lee left20:27
← GOODFatheR7 left20:27
← DoGGy_StyLe^ left20:27
← cr_cewk left20:27
← ZOE`S left20:27
← joni_adams left20:27
← maria_dpk left20:27
→ squ1z5 joined20:27
→ cO_TebInK_pEngEn joined20:27
→ cochn29 joined20:27
→ galnk joined20:27
→ putra_yg_lagi_pu joined20:27
→ and1k4 joined20:27
→ F^i^r^s^t joined20:27
→ co_in_here joined20:27
→ aryo799 joined20:27
→ zibeon joined20:27
→ Zyrcov joined20:27
→ _Reptyl_ joined20:27
→ zie joined20:27
→ MeCindy joined20:27
→ amie joined20:27
→ polarman joined20:27
← aryo799 left20:27
← F^i^r^s^t left20:27
← and1k4 left20:27
← putra_yg_lagi_pu left20:27
← galnk left20:27
← cochn29 left20:27
→ Oddity joined20:27
← deep_ left20:27
→ eric joined20:27
→ Fedotenko joined20:27
→ LNI_communityFOR joined20:27
→ ^pHoLLene^ joined20:27
→ PrAyInG^EyEs joined20:27
→ The_Rev joined20:27
→ CO_LG_NGCOK_KONT joined20:27
→ poelon joined20:27
→ micael_go joined20:27
→ boy_boring joined20:27
→ MarocSlave joined20:27
→ ikki_jie joined20:27
→ c0_aL joined20:27
→ Mr-Lonely joined20:27
→ maya^k}ar8in joined20:27
→ co_pengen_ML_ma_ joined20:27
← Hot_Date left20:27
← citronade03 left20:27
← DimaS40 left20:28
→ and1k4 joined20:28
← ce_bg02_an left20:28
← dimana left20:28
← tasikcrew left20:28
← wanita_dewasa_bt left20:28
→ [iMMoRtaL] joined20:28
→ JoLiEfLeUr20 joined20:28
→ DaveIngram joined20:28
→ A7EB_ALsadaqa joined20:28
← Cwo_caRi_tmen left20:28
← cari_agen left20:28
← micael_go left20:28
← CO_LG_NGCOK_KONT left20:28
← pretty_Nice left20:28
← CoWw__Cr__tmnCE left20:28
← steven28t left20:28
← wolf31r-s left20:28
eric → Guest3458320:28
← vdv left20:28
← boy_boring left20:28
← h26 left20:28
← cantikz_maniz_im left20:28
→ ^co_biasa_aja joined20:28
→ co_MKS joined20:28
← Bnt_aBohaa left20:28
← nee_a left20:28
→ Sapi joined20:28
← Guest34583 left20:28
← Fedotenko left20:28
← seSha left20:28
Harzilein cbreak: i don't get what you mean with "whole". what would be omitted when i don't specify --root? i'm wondering how i can still do the evil thing and suppress the history that clearly is earlier than my branch20:29
→ Cow_santai joined20:29
→ aRaLe joined20:29
→ jpalmer joined20:29
← buns0til_25 left20:29
← co_mw_ML left20:29
← co_kes3p1an left20:29
← aRaLe left20:29
← Mr-Lonely left20:29
← junky[\[a] left20:29
← cE_imOoet left20:29
← nurse_akbeth left20:29
← moccacino left20:29
cbreak just use --root20:29
← cenuij left20:29
← Guest26960 left20:29
cbreak otherwise you have to specify your own merge base, and lose your first commit20:29
→ vdv joined20:30
→ diogogmt joined20:30
← sair_ah left20:30
← rise_up left20:30
→ cenuij joined20:30
← cenuij left20:30
→ cenuij joined20:30
→ sudoquai joined20:30
→ Gorkyman joined20:31
→ mee joined20:31
→ E3FeeNi joined20:31
→ mishok13 joined20:31
← easly left20:31
← co_cr_kepuasan left20:31
← Pryde left20:31
← georges70 left20:31
← ikki_jie left20:31
← satria_pingit left20:31
← tara_aja left20:31
← polarman left20:31
← IcHanTiQue left20:31
← cow_suka_cow left20:31
← cow_t4_curhat left20:31
← CIBUBUR_CR_TEMAN left20:31
← Adi_jogja left20:31
← dirka left20:31
← ce-debate left20:31
← Oddity left20:31
← Cow_santai left20:31
← Assuradda left20:31
← MarcAndreQc left20:31
← _Reptyl_ left20:31
← co`jomblo`esia left20:31
← Zyrcov left20:31
← cO_gOgilzzz left20:31
← ekki left20:31
← DinDa_aZa left20:31
← c0_fb left20:31
← StePH[G] left20:31
← Faisal_36 left20:32
← MarocSlave left20:32
← zibeon left20:32
← Rhei left20:32
← Cwok_dws_ajak left20:32
← E3FeeNi left20:32
← JoLiEfLeUr20 left20:32
← c0_stawberryy left20:32
← cew_juaal_haandp left20:32
← rHya3 left20:32
← pim left20:32
← Oki_jie left20:32
← Cow_kece left20:32
← naq left20:32
← prjonzy87 left20:32
← Guest89457 left20:32
← Sapi left20:32
← CEWE_NDUTZ_BOUKE left20:32
← Co__cr__serius left20:32
← MeCindy left20:32
← caleb32 left20:32
← louis_43 left20:32
← cO_TebInK_pEngEn left20:32
← Ce_H1N1 left20:32
← The_Rev left20:32
← RiE_aLOnE left20:32
← LNI_communityFOR left20:32
← BOLIHUANGGA left20:32
← amie left20:32
← yuke_07 left20:32
← man_romantic left20:32
← WAPPO left20:32
← ^pHoLLene^ left20:32
← co_in_here left20:32
← andreas_mks left20:32
← muthia left20:32
← cwo_ngajak_Nonto left20:32
← phil\ left20:32
← zZzZ_bOys left20:32
← KlausFuchs left20:32
← Op-Gaul-Game left20:32
← co_pengen_ML_ma_ left20:32
← ^co_biasa_aja left20:32
← co_cakepz_sby left20:32
← gund left20:32
← PrAyInG^EyEs left20:32
← and1k4 left20:32
← yuga left20:32
← AnakAyam left20:32
← cari_ce-eksibisi left20:32
← PijatSensualUtkW left20:32
← co_MKS left20:32
← joetherich left20:32
← sallyna left20:32
← NoHoper left20:32
← Bluey_RoSe left20:32
← A7EB_ALsadaqa left20:32
← poelon left20:32
← Invierno left20:32
← zie left20:32
← woww left20:32
← CE_BISPAK_SMG left20:32
← COWOK_MANADO left20:32
← shady_777 left20:32
← LfcGirl left20:32
← PRIA^DEWASA left20:32
← sukroo left20:32
← agas left20:32
← LeD|oC left20:32
← stiffmeister left20:32
← maya^k}ar8in left20:32
← squ1z5 left20:32
Harzilein cbreak: will --onto still work if i don't have commits in common with the upstream?20:32
cbreak yes20:33
I think so20:33
hyperair rebase is like cherry-picking20:33
← Bricktop left20:33
→ zivester joined20:33
→ riskynet joined20:33
← Gorkyman_ left20:33
← dep3ndant left20:33
Harzilein "This has the exact same effect as git reset --hard <upstream>" <- but i would not be able to reset to the upstream's HEAD if i did not have commits in common i think20:33
← co_co_co left20:34
hyperair reset --hard always works20:34
cmn it always does what you tell it to, at any rate20:34
hyperair yeah20:34
cmn whether it makes sense or not...20:34
← [iMMoRtaL] left20:34
hyperair a branch points to a commit20:34
reset --hard $commit makes your current branch point to $commit20:35
← dv310p3r left20:35
cbreak you can reset to anything20:35
hyperair yep20:35
cbreak as long as it's a commit :)20:35
← bburhans left20:35
→ spvensko joined20:36
← spvensko left20:36
→ spvensko joined20:36
→ bburhans joined20:36
Harzilein if i leave out --onto, will it use the HEAD of my upstream?20:36
of my remote i mean20:36
uhm, whatever20:36
cbreak you have one single HEAD20:36
Harzilein the one i'm tracking20:36
cbreak it's the current branch20:36
Harzilein ah20:36
right20:36
← mishok13 left20:38
→ psoo joined20:38
→ cjb joined20:38
← nixmaniack left20:39
→ tazrover joined20:40
cbreak you have to specify an --onto, and --root20:40
→ Gorkyman_ joined20:41
Harzilein yay, it worked :)20:41
it got the old history i would have wanted to avoid, but oh well20:41
AAA_awright_ → AAA_awright20:41
→ rcu joined20:42
← bburhans left20:42
→ bauruine joined20:42
cbreak it got onto what you told it to20:42
← c0_aL left20:43
→ martinjlowm joined20:44
→ bburhans joined20:44
→ DONIEE joined20:44
← Gorkyman left20:44
→ DhimmEr joined20:44
→ Ronin`[CRF] joined20:44
← mercutio22 left20:45
← ipalaus_ left20:45
← GunArm left20:46
→ rtconner joined20:46
→ [David]KS[Lee] joined20:46
Harzilein okay, now i did this on a backup branch...20:46
→ Gorkyman joined20:46
← DONIEE left20:46
Harzilein how can i bring the master branch to the same branch?20:46
← mloskot left20:47
Harzilein same state*20:47
cjb Hi everyone. I'm using signed tags -- I say "git request-pull master <url> tagname", but the generated pull request says "<url> <branch name>", where there happens to be a branch at the same HEAD as the tag, and it doesn't mention the tag name. Should I replace the branch name with the tag name in the pull message?20:47
→ foldedcat joined20:47
→ rpg joined20:48
foldedcat any reason I would be getting On branch master nothing to commit (working directory clean) even after I've done git add first20:48
canton7 foldedcat, git status ?20:48
cjb foldedcat: maybe you did git add on a file that doesn't actually contain any changes20:48
→ empity joined20:48
← [M]ax left20:49
foldedcat it gives me the same error when doing git status after doing git add20:49
albel727 in any case, if git status says "working directory clean" then it's clean. it's not an error.20:49
canton7 it's not an error... It's just telling you that you have nothing to commit20:49
foldedcat hmm the file should contain changes, I've done a push since and the file on github isn't the same20:49
gotcha20:49
← Gorkyman_ left20:49
→ tgunr_ joined20:50
→ whyking joined20:50
← bburhans left20:50
→ Destos joined20:51
→ fairuz_ joined20:51
→ tkellen joined20:51
→ bburhans joined20:51
→ [M]ax joined20:52
tkellen can anyone tell me how would i overwrite files in a branch using files from another branch (in this case, master)20:52
→ Gorkyman_ joined20:52
→ GunArm joined20:52
← rcu left20:53
canton7 tkellen, git checkout branch_to_overwrite; git checkout another_branch -- path/to/file20:54
← whyking_ left20:54
← Gorkyman left20:54
← Ronin`[CRF] left20:55
← kpreid left20:55
← blueadept left20:55
→ mercutio22 joined20:55
→ Gorkyman joined20:56
← Gorkyman_ left20:56
← robde_ left20:56
← bburhans left20:57
→ jryans_ joined20:58
→ bburhans joined20:59
← cccaldas_ left21:00
→ niko joined21:01
← jryans left21:01
jryans_ → jryans21:01
← DaveIngram left21:02
→ kpreid joined21:02
→ dv310p3r joined21:02
→ aliguori joined21:04
← bburhans left21:05
aliguori Hi, i've been thinking about how to better work with cover letters when sending patch series. in many ways, git commit --allow-empty is perfect for this, the only problem is git-rebase doesn't add use --allow-empty so it gets upset about it, i was thinking about trying to fix this but was wondering if I'm barking up the wrong tree21:05
billy_ran_away How can I do a git show with a file that's been deleted but exists in previous commits?21:05
aliguori any advice?21:06
→ kennethreitz joined21:06
→ stefanha joined21:06
→ bburhans joined21:07
← Crewe left21:07
← DhimmEr left21:07
canton7 aliguori, git-format-patch. There's an option to generate a cover letter21:08
aliguori canton7, right, but that doesn't help when you resubmit a patch series21:08
that's the problem i'm trying to solve21:08
i need to keep around the cover letter, and i'd like to store that cover letter in git (in the same branch that I used to send the series)21:08
← cjb left21:09
canton7 git notes maybe? I'm loathe to stick that sort of info in a commit21:09
aliguori git notes tags an object, right? so a note wouldn't survive a rebase21:09
and rebasing a feature branch is a pretty common thing when sending a v221:09
→ bitkiller joined21:09
Mikachu aliguori: notes are a special ref, and rebase can copy them, though it doesn't by default21:10
yrlnry billy_ran_away: git show commitid:path-to-file21:10
← tgunr_ left21:10
→ tgunr_ joined21:10
billy_ran_away yrlnry: Thank you!!!21:10
Mikachu [notes]21:10
rewriteRef = refs/notes/*21:10
that enables it, iirc21:10
yrlnry billy_ran_away: you're welcome21:10
aliguori Mikachu, ah, cool, i'll look into that, thanks!21:10
gitinfo set mode: +v21:11
tgunr_ I just did a `git filter-branch --index-filter 'git rm -r --cached --ignore-unmatch private' --prune-empty -- --all' and watched a slew of rewrites, followed by some warnings like at https://ideone.com/N8xpO, should I run it again with the suggested option? And what is `cat'?21:11
← erichynds left21:12
Mikachu cat passes input to output without changing it21:12
→ Chaser joined21:12
Mikachu which will cause the tag to change to point to the rewritten commit ;)21:12
← anubhaw_ left21:13
← brngardner left21:13
→ kukks joined21:13
← dv310p3r left21:13
tgunr_ ok, I'm not sure where to put that21:13
← bburhans left21:13
Mikachu before or after the --prune-empty would work21:13
tgunr_ cc, trying21:13
Mikachu or after filter-branch21:14
← leen^| left21:14
→ bburhans joined21:14
← sebrock left21:15
tgunr_ hmm, getting `Cannot create a new backup. previous backup already exists'21:15
This is first time I've tried this filter stuff21:15
Mikachu right, did you make a backup of the repo first? :)21:16
tgunr_ not intentional21:16
ly21:16
Mikachu but you have one? :)21:16
tgunr_ h yes21:16
oh yes21:16
← martinjlowm left21:16
← nwp_ left21:16
Mikachu okay, then try rm -rf .git/refs/original and then try the command again ;)21:17
→ Spockz joined21:17
canton7 or you can pass --force to git-filter-branch, irrc?21:17
→ martinjlowm joined21:17
canton7 *iirc21:17
tgunr_ might as well remove the backup, i don't need it21:17
← [David]KS[Lee] left21:18
Mikachu heh21:18
← jdav_gone left21:18
Mikachu at least keep it until this copy is how you want it?21:18
sontek I need to pull a commit out of master and push it to a branch (its not as good as I thought when I pushed it)21:18
→ fisted_ joined21:20
→ GunArm__ joined21:20
← bburhans left21:20
← GunArm left21:20
← heptadecagram left21:20
→ dzonder joined21:21
← keystr0k left21:21
PerlJam sontek: cherry-pick it to the other branch and revert it in master?21:21
→ Gorkyman_ joined21:21
→ bburhans joined21:22
sontek PerlJam: well, its already in master, so if I create a branch it'll already be in there as well.. right?21:22
← fisted left21:22
sontek PerlJam: so if I just git revert it'll do what I want?21:22
← Gorkyman left21:23
PerlJam should.21:23
assuming I understood what you wanted.21:23
← Spockz left21:23
→ jdav_gone joined21:23
tgunr_ ok, i have pruned my local repo and the private folder is gone, can I just `git push origin master' and the folder will be gone on the remote?21:23
sontek thanks21:24
→ jmontross joined21:24
→ i42n joined21:25
← gusnan left21:26
← joshteam left21:27
← fairuz_ left21:27
← thiago left21:27
← marthinal left21:28
→ ipalaus_ joined21:28
→ nardev1 joined21:28
← bburhans left21:28
→ brngardner joined21:28
← nadirvardar left21:28
→ bburhans joined21:29
← signal-9 left21:29
← aliguori left21:30
→ dv310p3r joined21:30
← bartek left21:30
← spvensko left21:31
← psoo left21:31
← proq left21:32
← luxurymode left21:32
← martinjlowm left21:32
← jceb left21:32
← mercutio22 left21:32
→ martinjlowm joined21:33
→ marshall joined21:33
marshall hey git21:33
what's the proper way to update all submodules of a project to their latest remote heads?21:34
I'm trying to update all my submodules in my dotfiles repo, so they're like vim scripts and stuff that have been updated by their authors21:35
← bburhans left21:35
iamjarvo so i merged a branch into another branch and there should have been conflicts but it didnt give me any conflicts and gave me one the version of the file i didnt want21:35
well there was an edit in the one file that i wanted to combine21:35
→ thiago joined21:36
→ cylence joined21:36
SethRobertson marshall: !submodule_change21:37
gitinfo marshall: In order to change a submodule you must go into the submodule repository, check it out to the appropriate branch, make the needed change (possibly involving git pull), commit the change, cd .. (out of the submodule), git commit -m "Updated submodule" submodulepath21:37
← brngardner left21:37
cmn a git submodule foreach git pull might work21:37
SethRobertson iamjarvo: Presumably you can `git merge --no-commit` and then fix up what it did wrong.21:37
cmn: As step one in the process, certainly21:38
Well, step two. You need to be on a branch first21:38
iamjarvo SethRobertson: should i do git merge —abort then? and try what you suggested?21:38
→ EricInBNE joined21:38
cmn oh, right, submodules are detached21:38
SethRobertson ?? If the merge did not complete you can just fix it now. If the merge did complete it is too late to abort. You can `git reset --hard HEAD^` to undo the merge, if you have not pushed21:39
← guns left21:40
→ mercutio22 joined21:40
← Q| left21:41
→ AmirBehzad joined21:41
← AmirBehzad left21:41
← billy_ran_away left21:41
← jonesy left21:41
← churp left21:42
← jryans left21:43
iamjarvo SethRobertson: so there was a change in the branch and i had a change on my feature branch. i need that part of the file on the branch im trying to merge into21:43
→ jryans joined21:44
← noz left21:44
← Know1edge1 left21:44
SethRobertson Copying partial-commit changes always leads to tears21:45
marshall SethRobertson: thanks mate21:45
→ churp joined21:45
iamjarvo not sure what to do21:46
should i just edit it?21:47
→ mloskot joined21:47
← ipalaus_ left21:47
← mloskot left21:47
→ mloskot joined21:47
SethRobertson iamjarvo: If the change you want on the feature branch is a commit by itself, then cherry-pick it. If you have pushed the feature branch then just edit it. If you have not pushed the feature branch, then you can rewrite the commit on the feature branch to split it into the stuff you want to cherry pick and everything else21:48
If after reading that twice your mind is still blown, then just edit it21:48
iamjarvo lol21:49
← Gorkyman_ left21:49
iamjarvo thanks SethRobertson21:49
→ lmi joined21:49
lmi hi21:49
← tazrover left21:49
SethRobertson ~hi21:49
→ Gorkyman joined21:49
lmi can format-patch strip prefix the way patch -p does?21:49
SethRobertson !hi21:50
gitinfo [!welcome] Welcome to this channel. Go ahead and ask your question. Very few people are responding to greetings, but will respond if a question is asked.21:50
SethRobertson git format-patch is designed to feed to git, not to normal patch systems.21:50
← tomislater2 left21:50
SethRobertson However, see --src-prefix and --dest-prefix and --no-prefix21:50
← churp left21:51
lmi well, i'm using git-svn, so want a patch that applies with -p1 in a release tarball21:51
SethRobertson So --no-prefix sounds like it is what you want21:52
lmi aka, want to get rid of trunk/ from it21:52
let me see, thanks21:52
martinjlowm → martinjlowm-lapt21:52
martinjlowm-lapt → martinjlowm-lptp21:53
lmi yeah, that will do, thanks!21:53
→ FernandoBasso joined21:53
← TrineLocke left21:53
malphaet → Malphaet[away]21:53
→ churp joined21:54
→ magnus___ joined21:55
← jsick3 left21:55
← ThomasLocke left21:56
→ noz joined22:00
← detaer left22:01
← tommyvyo left22:01
← saml left22:01
→ detaer joined22:02
→ spvensko joined22:02
← spvensko left22:02
→ spvensko joined22:02
← Heisenmink left22:03
→ linusoleander joined22:03
← mercutio22 left22:03
→ mercutio22 joined22:05
← churp left22:05
← tatsuya_o left22:06
← cannonball left22:06
tgunr_ After doing `git filter-branch --index-filter 'git rm -r --cached --ignore-unmatch private' --prune-empty --tage-name-fileter cat -- --all' and the private folder is now gone from my local repo. But not getting how to update the remotes, if I try to push it complains and wants me to pull but if I pull it wants to overwrite the folder I just deleted.22:06
← bauruine left22:06
SethRobertson You need to do a force push. This may be denied by default by the remote server22:07
tgunr_ ok, so just add a -f eh?22:07
→ the_cybersphinx joined22:08
← i42n left22:08
SethRobertson Assuming you understand the drawbacks of !rewriting_public_history, yes22:08
gitinfo Rewriting public history is a very bad idea. Anyone else who may have pulled the old history will have to `git pull --rebase` and even worse things if they have tagged or branched, so you must publish your humiliation so they know what to do. You will need to `git push -f` to force the push. The server may not allow this. See receive.denyNonFastForwards (git-config)22:08
cbreak filter-branch is meant to rewrite history22:08
← devth left22:08
tgunr_ no one but me is using it :) so i think it is ok22:08
SethRobertson that doesn't necessarily mean they understand the problems22:08
cbreak and as it is tradition, he who writes history controls the future22:08
so use that power responsibly22:09
← phishy left22:09
← cyphase left22:09
tgunr_ thats the whole point of doing this, I want to prune this folder out before someone does use it22:09
← kpreid left22:10
→ tatsuya_o joined22:10
← cybersphinx left22:10
martinjlowm-lptp → martinjlowm22:10
tgunr_ so when the repo is cloned this folder will not be known to exist22:10
→ NiklasFiekas joined22:10
← spvensko left22:10
→ iamjarvo1 joined22:11
← axl_ left22:12
tgunr_ I guess another way would be to create a new bare repo after the folder is pruned no?22:12
← martinjlowm left22:12
→ ipalaus_ joined22:13
← iamjarvo left22:13
SethRobertson Sure, whatever is best for your case22:13
→ cyphase joined22:14
→ rjcks joined22:14
→ gnufied joined22:14
← dotpot left22:15
→ jjasinski joined22:16
→ jdav_gon` joined22:16
← macmartine left22:16
→ stringoO joined22:16
→ kpreid joined22:18
← drizzd left22:18
→ bencahill joined22:18
← mercutio22 left22:18
← gnufied left22:19
← kpreid left22:19
→ axl_ joined22:19
← jdav_gone left22:19
→ kpreid joined22:19
bencahill hey guys, can you recommend a good introduction to git?22:21
SethRobertson Lots. See !book22:21
gitinfo 'Pro Git' is a complete book about git that is available online for free (there's a print version, too): http://progit.org/book/22:21
SethRobertson !cs22:21
gitinfo 'Git for Computer Scientists' explains the basic structures used by git, which is very helpful for understanding its concepts. http://sitaramc.github.com/gcs/22:21
← whitman left22:21
SethRobertson !doc22:21
gitinfo A list of useful documentation of all kinds is here: http://git-scm.com/documentation -- or try the wiki at http://git.wiki.kernel.org/. Also try typing "!book" "!cs" "!bottomup" here in IRC, or reading http://www.webdesignerdepot.com/2009/03/intro-to-git-for-web-designers/22:21
→ cccaldas joined22:21
SethRobertson !bottomup22:21
gitinfo 'Git from the bottom up' starts with explaining the building blocks of git and proceeds to tell you how they fit together. http://ftp.newartisans.com/pub/git.from.bottom.up.pdf22:21
← tgunr_ left22:22
bencahill SethRobertson: online book looks perfect, thanks!22:22
→ tgunr_ joined22:22
← ISF_ left22:22
→ giallu joined22:22
← tgunr_ left22:24
→ ISF joined22:24
marshall \exit22:24
← marshall left22:24
→ tgunr_ joined22:24
← Goplat left22:26
→ soulcake joined22:27
→ Vile joined22:27
→ erichynds joined22:28
→ martinjlowm joined22:28
← erichynds left22:29
clone1018_ → clone101822:30
→ DaveIngram joined22:31
← rleigh left22:32
← claudiu__ left22:32
← martinjlowm left22:33
→ martinjlowm-lptp joined22:33
← maletor left22:35
← d0k left22:36
→ eletuchy joined22:36
→ maletor joined22:37
← whyking left22:37
← jmcantrell left22:37
→ jmcantrell joined22:37
← cccaldas left22:38
← tgunr_ left22:40
← acl_ left22:40
→ tgunr_ joined22:40
→ thursday joined22:40
← vdv left22:43
← tgunr left22:44
← tgunr_ left22:44
→ tgunr joined22:44
← gjoseph left22:45
← rendar left22:45
← giallu left22:47
→ Cromulent joined22:47
← _iron left22:48
→ gebi joined22:48
← Ryback_ left22:49
← digitalfiz left22:50
← jmontross left22:50
← DarkAR left22:51
← davidcoallier left22:51
← mando left22:52
→ vdv joined22:52
← kpreid left22:53
→ kpreid joined22:53
← metcalfc left22:55
→ MarcoSousa joined22:56
→ erichynds joined22:57
→ tellone joined22:58
← dv310p3r left22:58
→ t4nkd`workin joined22:59
← maletor left23:00
bencahill in the progit book, chapter 2,2 it says when you run the git status command on a fresh clone, there are no tracked and modified files.23:00
does it mean that there are no tracked files that have been modified?23:00
→ oriba joined23:01
lmi yes23:02
← thiago left23:02
lmi after clone you shouldn't have untracked files either, anyways23:02
bencahill ok, thanks...if that's maintained, it could use a little more specificity23:02
lmi: correct23:02
(correct = I understand) ;)23:02
→ mdel_ joined23:03
← diogogmt left23:03
← neurodrone left23:03
← soulcake left23:03
← t4nkd left23:04
← alan_w left23:05
← t4nkd`workin left23:05
→ nax__ joined23:05
→ dsheets joined23:06
← alester left23:06
← john2496 left23:06
← pantsman left23:06
← AaronMT left23:06
→ madewokherd joined23:06
← Gitzilla left23:06
→ maletor joined23:06
dsheets When should I use a version-based release branch vs a tag? Aren't tags just like branches?23:06
wereHamster dsheets: tags can't move, branches can23:07
Malphaet[away] → Malphaet23:07
← rpg left23:07
→ eggman2001 joined23:08
← eletuchy left23:08
dsheets wereHamster: by move you mean change? I suspect there may be bugfixes at a later time so I should branch for the minor version and tag the revs?23:08
→ billy_ran_away joined23:08
billy_ran_away Is there a convenient way to look at the commit of the last time a file was changed?23:09
FauxFaux git show $(git log --format=%h -1 path/to/file)23:10
→ eletuchy joined23:10
wereHamster dsheets: yes23:10
billy_ran_away FauxFaux: Thank you23:10
dsheets wereHamster: cool. thanks for your guidance. i really appreciate your help :-)23:11
← DaveIngram left23:11
← aspotashev left23:12
→ s0undt3ch joined23:12
← berserkr left23:12
← akosikeno left23:13
← FernandoBasso left23:13
← Sir_Rai left23:13
→ berserkr joined23:14
← kaimi left23:14
→ basepi joined23:14
← canton7 left23:14
→ cccaldas joined23:16
→ giallu joined23:17
→ kaimi joined23:17
→ Sir_Rai joined23:17
→ FernandoBasso joined23:17
← j416 left23:18
dsheets One more question: should I start my version tags and branches with an alpha like 'v' or can I use %d.%d.%d directly? Will the decimal-first format mess up some system?23:19
cbreak tags and branches are file names23:19
does your system allow file names starting with a number?23:20
bencahill hehe23:20
← MANCHUCK left23:20
→ polygonpusher joined23:20
dsheets it does. I assume there are some systems that may not, though? Windows or something?23:20
← localhost left23:21
dsheets It seems like a common identifier limitation to require the first character to be alpha or _. Is 'v%d.%d.%d' community convention?23:21
← mandric left23:21
bencahill dsheets: Windows may not be that great, but it's not that bad ;)23:22
cbreak I read windows only allows one letter names for volumes23:22
← eggman2001 left23:22
cbreak no numbers23:22
bencahill cbreak: like just 'a' ?23:22
cbreak or C23:22
→ mandric joined23:22
bencahill ok, I'll try...23:22
cbreak C is the most popular for who knows why23:22
dsheets floppy-only machines? A and B are floppy vols23:23
bencahill cbreak: I just touched a file named C on a Windows system (share), and it seemed to work fine...23:23
→ localhost joined23:23
polygonpusher I'm working on a django project where I init the repo inside the source of the project on the remote sever then I pulled the git repo into my local eclipse ide but when I went to push I got error about the master being checked out23:23
is that cause the remote repo considered to be the working folder?23:24
bencahill cbreak: just made one named C with Windows Explorer (myth busted!)23:24
← beaufour left23:24
→ martinjlowm joined23:24
→ beaufour joined23:24
bencahill cbreak: 0 also worked23:25
polygonpusher also is there a guide I could read about best practice when from creating a repo for server hosted project like website23:25
→ thiago joined23:25
← boombatower left23:26
polygonpusher err. Sorry bad typing. I hope you understood that.23:26
← umren left23:26
← mandric left23:27
→ diegoviola joined23:27
← martinjlowm left23:27
← MarcoSousa left23:27
→ martinjlowm joined23:28
← empity left23:28
→ BlaDe^ joined23:28
→ DaveIngram joined23:29
← FernandoBasso left23:29
← beaufour left23:29
→ FernandoBasso joined23:29
← martinjlowm left23:30
→ martinjlowm joined23:30
← eijk_ left23:31
← lobolife left23:31
← alex__c2022 left23:32
→ empity joined23:32
→ iamjarvo joined23:34
← martinjlowm left23:35
← iamjarvo1 left23:35
tizzo → tizzo-afk23:37
→ Gorkyman_ joined23:38
← babilen left23:38
→ nOStahl joined23:38
← Gorkyman left23:39
nOStahl hi all23:39
→ seivan joined23:39
← sudoquai left23:40
← empity left23:40
← ipalaus_ left23:41
→ Gorkyman joined23:42
← Gorkyman_ left23:42
→ macRover joined23:45
← thiago left23:46
← Cromulent left23:47
← snearch left23:47
→ nadavoid joined23:47
→ thiago joined23:48
→ cronin joined23:52
← dsheets left23:52
← Morphous_ left23:53
→ shiba_yu36 joined23:53
← DaveIngram left23:54
→ Anxi80 joined23:56
← Vampire0_ left23:57
→ Vampire0 joined23:58
← zivester left23:58
→ zackiv31 joined23:58

Logs Search ←Prev date Next date→ Channels Documentation