IRCloggy #git 2007-12-26

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.

2007-12-26

→ FunkeeMonk joined00:06
→ rhalff joined00:10
← kumbayo left00:15
→ theCarpenter joined00:26
← theCarpenter left00:26
← bdiego left00:36
← ebel left00:43
← janm left00:45
← Yuuhi left00:46
→ janm joined00:52
← rhalff left00:54
← HG` left01:02
→ mattly joined01:07
← cartman left01:11
→ bdiego joined01:11
→ rhalff joined01:17
← bpalmer` left01:17
thiago -r--r--r-- 1 tmacieir 232580636 2007-12-26 01:11 pack-3759a969fcb84bf4fad82d31441eed980b057f3d.idx01:27
thiago has an index file that is larger than the kernel pack file01:27
→ JohnnyL joined01:28
Ilari thiago: Wow...01:29
Debolaz pats thiago01:30
Ilari thiago: Got better compressed packfile yet?01:32
thiago no01:35
not yet01:35
it's at 28% now01:35
I wonder if the performance drops with the number of objects to pack01:35
the pack file is 42 GB in size01:36
I'm running a repack -f now, let's see how far it goes01:37
JohnnyL hey guys how are you doing?01:37
Ilari thiago: It would be probably possible to obtain object list (unless rev-list OOMs too) and feed chunks of it to git-pack-objects with delta research and then finally do repack without -f...01:38
thiago rev-list works01:38
it counts 623354 commits01:39
let's see how fast an annotate goes01:39
← kanru left01:40
Ilari Aww... Except that just splitting it won't do the trick... One would probably need to resort it first and then split it...01:45
← rhalff left01:45
thiago it's interesting that branching & tagging are O(n) operations in CVS01:45
and my importer makes then O(n) too in the commits generated by cvs2svn01:45
Ilari How many lines does 'git rev-list --all --objects' dump?01:47
thiago presses enter01:49
DrNick Ilari: why that instead of git count-objects -v01:49
thiago in-pack: 1001476601:55
packs: 201:55
Ilari Why there are two packs...01:56
thiago because the only way I could make it not OOM was to create a .keep on the first pack01:57
the already-well-compressed initial pack of 2.5 GB01:57
if I remove the .keep file, it OOMs01:57
Ilari The sort apparently could be done using contents of 'rev-list --all --objects' lines after first 40 characters, read backwards(!).01:59
Oh, and the sort should be stable...02:00
← janm left02:01
← mattly left02:02
← bdiego left02:06
thiago gives up on the rev-list command02:08
→ krawek joined02:09
JohnnyL git init is not your friend!02:12
context johnnyl: it is here02:13
← krawek left02:14
JohnnyL context, afaict it delinks objects when run in an existing git repository.02:14
context johnnyl: had zero effect here02:16
meh02:19
johnnyl: i dont see a reason for re-init'ing a repo :/02:19
Running git-init in an existing repository is safe. It will not overwrite things02:20
that are already there.02:20
thiago confirms that the import is faulty02:25
→ dsaxena joined02:33
← kukks left02:37
albertito thiago: just out of curiosity, why don't you experiment on a shorter repository, and when that succeeds, try the big one?02:55
thiago experiment what?02:57
my experiment *is* the KDE's trunk root02:58
→ kanru joined03:05
albertito thiago: I guess every thing you try takes a huge lot of time, so I was wondering if you tried with, for instance, half of the svn repository, and see if it succeeds03:10
thiago that's true03:11
but this is not what I want03:11
albertito thiago: and only when it worked good enough, you would try to use the whole repo03:11
thiago I did a test import of trunk using svn-fast-export. It generated a Git repo of 53 GB I'm still trying to pack and I know the import is faulty.03:11
albertito thiago: I must ask, then, what do you want? =)03:11
thiago and it didn't import branches03:11
I have written now a tool to do the proper import, but this tool of mine requires a file with import rules03:12
this is what I'm doing now.03:12
albertito thiago: I see03:12
thiago: well, your experiment is quite interesting (at least to me, of course), so please keep on posting about how it goes =)03:13
thiago I will03:14
I was surprised to find out that a 32-bit machine with 4 GB of RAM cannot pack 9.8 million objects03:14
→ theCarpenter joined03:25
→ Ori_B joined03:37
Ori_B hi03:37
I'm having trouble with git-daemon03:38
whenever I try to run it, I get the error:03:38
"fatal: unable to allocate any listen sockets on host (null) port 941803:38
→ bobesponja joined03:41
→ DoctorPepper joined03:42
→ wayneeseguin joined03:48
← thiago left04:08
← FunkeeMonk left04:10
← fhobia left04:31
→ srid joined04:32
srid I have all my repo at repo.or.cz and now I can't 'push' to them because I am on a different machine.04:32
I wish it supported password login.04:32
:|04:32
← Ori_B left04:33
theCarpenter ssh?04:36
→ mattly joined04:44
srid theCarpenter: repo.or.cz requires host key and it only accepts one host key per user.04:45
.. which is that of my Linux box. now I'm on Windows.04:45
→ janm joined04:52
← bobesponja left04:56
→ mozammel joined04:57
mozammel Hi, can I change comment of a specific commit?04:57
loops mozammel, git history is immutable.. in order to change the comment for a particular commit.. you must rewrite every commit that comes after it..04:58
mozammel, there is a tool git-filter-branch which is designed to help you do this.04:58
paulproteus git-lie-to-the-cops04:58
loops mozammel, if the commit you're interested in, is actually the most recent commit on a branch.. you can use git commit --amend to change it instead.04:59
mozammel loops, thanks... checking out git-filter-branch04:59
← agib left05:06
← tjafk left05:27
→ tjafk joined05:28
← yashi left05:30
→ yashi joined05:32
→ cartman joined05:40
→ fhobia joined05:43
← JohnnyL left05:54
srid git config --global option is not supported in git v1.5.4?05:58
← theCarpenter left06:01
gitster works fine here.06:05
← kanru left06:12
srid not in windows (vista)06:18
``key does not contain a section: global``06:18
gitster have no idea wnat (vista) is, sorry.06:18
← srid left06:24
→ FunkeeMonk joined06:28
→ kanru joined06:31
← yashi left06:38
→ yashi joined06:38
← FunkeeMonk left07:12
← yashi left07:14
→ yashi joined07:16
→ Pengwn joined07:30
→ bentob0x joined07:38
← madewokherd left07:48
← yashi left07:55
→ yashi joined07:56
→ stni joined07:59
zrajm what is command is the git equivallent of "svn info"?08:04
aeruder isn't really one, what piece of info do you want out of svn info?08:05
zrajm the url to the repository, specifically. found it in .git/config, but i thought there must be a command to obtain it too..08:05
mugwump git remote08:05
zrajm mugwump, that simply states "origin"?08:06
mugwump then git remote show origin08:06
zrajm ah. thanx.08:06
← fhobia left08:19
→ cartman_ joined08:28
← cartman left08:39
← dwmw2_gone_ left08:44
→ mathieu__ joined08:46
← kanru left08:51
← dsaxena left08:54
← yashi left09:00
← mozammel left09:01
← mathieu_ left09:03
cartman_ → cartman09:13
→ desegnis joined09:20
← asdx left09:37
→ yashi joined09:44
yashi how can I disable the compression while fetching from local repo?09:46
mugwump why does it matter, is it taking too long?09:47
I would have thought that git clone -l wouldn't end up repacking anything09:48
or do you mean when using fetch ?09:48
yashi yeah, I'm testing with huge blob and it takes while09:48
yes, with fetch09:48
hmm.. I guess I just use ln by myself09:52
→ ddaa joined09:54
ddaa hello09:57
what's the preferred way of binding git from Python?09:57
← Sho_ left09:57
Randal what's it called, os.process or something? :)09:58
mugwump ddaa, take a look at the python scripts in contrib/09:58
ddaa ok, so process spawning09:58
is the git design amenable to more library-like binding? Process spawning tends to be less than ideal.09:59
mugwump process spawning is great09:59
you get tidy little networks of nodes each performing a small task09:59
it's just a matter of how you plumb them09:59
ddaa mugwump: let's agree to disagree on this point. Can you answer my question?09:59
mugwump right now there's no libgit10:00
at least, not one with a stable API10:00
ddaa Okay, I guess that means the code has some stdio and exit calls scattered all over.10:00
Or is it just that no formal API was defined?10:01
Randal it just hasn't been a high priority10:02
a lot of other cool things have gotten done instead. :)10:02
ddaa wrote python bindings to GNU Arch back in time, so he knows how CLI binding can be a pain in the arse.10:02
Randal a lot of thought has been given to making the plumbing work nicely in scripts though10:03
and that's sufficed enough to keep people from writing a lib interface10:03
ddaa Thanks. I think that answers my question for now.10:03
← Pengwn left10:09
→ kumbayo joined10:11
← yashi left10:15
← mugwump left10:16
→ yashi joined10:17
← LotR left10:31
→ LotR joined10:32
ddaa So, I tried to clone HelloWorld for a start10:45
http://rafb.net/p/P5IEjb12.html10:45
it yields a "fatal: Not a valid object name HEAD"10:45
I feel stumped.10:46
siprbaum ddaa: your url is wrong10:46
ddaa: this is the url to the gitweb and not for cloning10:47
ddaa oh, right10:47
thanks10:47
siprbaum but if you open http://repo.or.cz/w/HelloWorld.git in the browser, you'll find the right urls10:47
ddaa found it10:47
← desegnis left10:48
→ reval joined10:56
iXce fun enough, I thought about doing such a repo around the same date (with HelloWorld's in every language)11:00
Pieter not very interesting imho11:01
ddaa it's just a good sample repo to play around11:02
Pieter I'd rather take something with some interesting merges / conflicts11:02
ddaa one thing at a time11:03
→ Eludias joined11:04
Arjen Pieter: The Git repository itself provides those in abundance :-)11:07
→ Yuuhi joined11:35
→ Dodji joined11:37
← reval left11:40
→ thiago joined11:53
← wayneeseguin left12:09
→ wayneeseguin joined12:10
← DoctorPepper left12:12
→ anttih joined12:13
→ anttih_ joined12:16
← anttih left12:16
→ kanru joined12:21
→ HG` joined12:22
← Fullmoon left12:28
→ robfitz joined12:38
← anttih_ left12:44
→ nud joined12:47
← robfitz_ left12:48
← wayneeseguin left12:53
→ wayneeseguin joined12:54
← wayneeseguin left12:54
→ wayneeseguin joined12:57
→ Sho_ joined13:03
→ ferdy joined13:18
→ FunkeeMonk joined13:19
→ devogon joined13:21
→ Pengwn joined13:25
← HG` left13:32
→ Skyblaze joined13:47
Skyblaze Is it wrong to run always "git add ." in the root of the project instead of specifing the added or changed files?13:49
i mean is this way too heavy for the system?13:49
i point this 'cause maybe we can't always rememeber which files we have changed or added13:49
so doing git add . we can't go wrong13:50
→ ebel joined13:50
→ GyrosGeier joined13:51
Pengwn I am guessing git-add is like an intermediate commit . Before you actually commit. so you can git-diff and ignore files you messed up.?13:52
numbah Skyblaze: you could add the files once and then use git add -u to add the tracked files only13:53
otherwise you'll add some temporary files or such13:54
if you keep doing add .13:54
Skyblaze numbah, i think that is not a problem 'cause anyway i'm sure i will always add the same files 'cause i keep a right .gitignore file13:55
numbah, and then with git add -u i don't bring new files with me13:55
Tv sounds like you might want git commit -a, for even less typing14:04
→ nud_ joined14:23
→ doener joined14:26
← nud left14:40
wayneeseguin Also, if you use a shell you can for example (bash) do allias gca="git commit -a" and then when you want to commit all changes simply do $ gca14:41
:D14:41
s/allias/alias14:41
→ G_SabinoMullane joined14:42
← doener_ left14:42
← context left14:46
→ anttih joined14:47
Arjen git config alias.gca 'commit -a'14:49
or, if you want to set it in your ~/.gitconfig:14:49
git config --global alias.gca 'commit -a'14:49
→ bdiego joined14:51
→ kukks joined14:51
numbah shouldn't that be ca then? since the "git" part is there already14:51
ddaa Am I correct that "git fast-import --export-marks=marks" now use write-then-rename to update the marks file atomically?14:56
That breaks some code I have here that assumes that it just writes in place.14:57
Arjen numbah: Yes, you're right :-)14:58
numbah :)14:58
ddaa that also defeats the secure use of tempfile that is made by this code :(14:58
thiago ddaa: you shouldn't read from the marks file until fast-import is finished14:59
ddaa thiago: sure14:59
The problem I have is that the marks file is opened before git is spawned.14:59
thiago then wait for it to finish before you open the file15:00
ddaa And unless I reopen this file name after it completes, it contains no output.15:00
thiago: the thing with temp files, is that it's much more convenient to just open them once at creation time.15:00
thiago I'm not following...15:01
ddaa nevermind15:01
thiago "you shouldn't open the file before it finishes" -> it doesn't matter when it was created15:01
it finishes, you open the file.15:01
ddaa I have a temp file API that creates a temp file securely with an unpredictable name.15:01
and gives an open file with a known name15:02
thiago temp files always have random names.15:02
you cannot reopen them15:02
ddaa unless you don't unlink it right away to let git write to it15:02
I just realized that's insecure anyway15:02
no big deal, it's code in a test suite15:03
→ context joined15:03
← FunkeeMonk left15:13
→ chris2 joined15:16
← kanru left15:18
→ Oejet joined15:27
cbrake_away → cbrake15:33
→ redmatt joined15:35
redmatt anyone seen keith packard, parsecvs author, around?15:37
siprbaum redmatt: don't think he is in here. at least I can't remember him saying anything15:42
redmatt siprbaum, ahh, ok, thanks15:42
siprbaum redmatt: try /whois keithp to see where i lurks around :-)15:43
s/i/he/15:43
redmatt brilliant, thank you15:44
Pengwn ok15:46
← Pengwn left15:46
← redmatt left15:53
← nud_ left15:53
→ nud joined16:00
→ bobesponja joined16:00
→ ropiku joined16:04
→ z3om joined16:06
→ dsaxena joined16:07
→ cannonball joined16:08
ropiku Hello, How do I authenticate using git-clone ? I have a webdav with basic auth. Is there another method besides .netrc ?16:13
thiago sshd authenticates.16:13
ropiku without ssh16:15
I want to build a system to host git repos and thought that using http auth (like in svn) is easier for users then uploading their public keys16:16
Ilari ropiku: SSH transport is 'smart', unlike HTTP, and thus is more bandwidth-efficent... And SSH can do password auth (if you don't care that much about security, but HTTP can't do much better without using TLS-SRP, which is very recent)...16:18
thiago you can use password-based ssh authentication16:19
ropiku and how can I automate that ? Using known-keys and command I know16:20
thiago the /bin/passwd command16:21
how to automate ssh is off-topic for this channel16:21
ropiku ok, sorry. so you recommend using ssh with git16:22
Randal it's not *entirely* off topic... you just need to google a bit :)16:23
but yes, you eventually want ssh repo-host to not ask any questions16:23
← zeom left16:23
z3om → zeom16:23
Randal and then use git+ssh://repo-host/ as your transport16:23
ropiku thanks16:24
git+ssh is better then ssh:// ?16:24
Randal same16:24
I'm old school16:24
the syntax has gone through many iterations :)16:24
the problem with being involved with the git project for 2.5 years, is you've seen a lot go by16:25
ropiku nice, I heard about git recently and i like it a lot16:25
Randal did you see my google talk?16:27
search video.google.com for "randal schwartz git"16:27
← zjason left16:28
vmiklos or you can use just host:/path like with scp:)16:29
ropiku I first saw Linus talk then yours, that's when I switched to git :)16:30
shenson_not_here → shenson16:31
Oejet thiago: Did your huge import finish?16:32
thiago the import, yes16:33
the repacking, no16:33
Oejet That must take a while.16:34
thiago it's been running for 36 hours now16:34
Oejet Can you see what the bottleneck is?16:35
thiago yes: the number of objects16:36
9.8 million16:36
the process has VSZ=13.5 GB and RSS=5.3 GB16:38
Oejet Well, it should be roughly linear in the number of objects as far as I understood.16:38
So it takes a lot of page faults?16:38
→ HG` joined16:40
← davidm123_ left16:41
thiago it's getting page faults, aye16:41
according to /proc/<pid>/stat16:41
← anttih left16:46
vmiklos the final plan is to have kdebase.git, kdelibs.git, etc, right?16:46
→ raxor joined16:47
thiago right16:48
Ilari thiago: Major page faults presumably?16:48
thiago I hate reading /proc/<pid>/stat16:48
raxor Hi #git, I am trying to setup git-daemon on my server where my gitsosis repos are hosted using the command: "sudo -u git git-daemon --base-path=/home/git/repositories/" but it is not finishing... any ideas why16:49
thiago the 10th and 12th fields16:49
raxor there are no errors, it just seems to hang there16:49
vmiklos they why do you bother with such a big repo? or do you want to stress-test git ;)16:49
thiago just because I can16:49
vmiklos ah:)16:49
thiago I know the repo is faulty now16:49
vmiklos raxor maybe you want --detach?16:50
Ilari raxor: Also, one could see 'netstat -plant' output. It should show the server listening...16:52
raxor thanks guys will check16:53
vmiklos: if I run it like that would it survive a reboot?16:53
or do I need that in init.d?16:53
Ilari raxor: To make it survive reboot, you need add script starting it to initscripts.16:53
→ zjason joined16:55
raxor Ilari: in netstat -plant what port am I looking for?16:55
Ilari raxor: IIRC, 9418... And why do you seem have lots of ports?16:56
raxor Ilari: a few yes16:56
But 9418 is definitely there.. seems to be working greaqt16:57
Tv haha netstat -plant16:57
→ technoweenie joined16:57
Tv netstat -Faceplant16:58
technoweenie hey dudes, i have a quick question regarding remote tracking branches16:58
i created a repo like this: "git init ; git remote add origin git@whatever ; git push origin master:refs/heads/master"16:58
raxor )16:58
:)16:58
technoweenie now i want the master branch to track origin/master. is there a command to do that without having to modify .git/config ?16:59
seems git branch/checkout both want to drop the old one and recreate it.16:59
Ilari technoweenie: Well, there's 'git config'...17:00
technoweenie guess i could just remove and reclone it too17:01
Ilari: thanks for the assist: git config branch.master.remote 'origin' ; git config branch.master.merge 'refs/heads/master'17:04
raxor Ilari: does this look good for init script? "git stream tcp nowait nobody /usr/bin/git-daemon git-daemon --inetd --verbose --base-path=/home/git/repositories"17:04
Ilari raxor: Ah, inetd. If your inetd is in initscripts, then that works...17:06
s/works/should work/17:06
raxor :)17:06
cheers17:06
Ilari raxor: Of course, you should test with actual repository...17:07
raxor Trying now...17:07
← kumbayo left17:12
raxor Ilari: do you know if you need to restart git-daemon when you add a repo for export (with git-daemon-export-ok)?17:15
Ilari raxor: No.17:16
raxor: No, it is not needed, that is.17:16
technoweenie raxor: are you using gitosis? it's all pretty slick for managing that stuff17:16
has specific config options for gitweb and gitdaemon17:17
raxor technoweenie: I am using gitosis17:20
what are the config options? the scientist article said to use git-daemon separately for anon read-only access17:20
has that changed?17:20
→ Ingmar_ joined17:21
Ilari raxor: Yes, you should use git-daemon for anon read-only access...17:22
technoweenie you have to add entries like [repo my-repo]\ndaemon = yes17:24
and it adds the git-daemon-export-ok files for you17:24
raxor reads the latest gitosis config Readme and it says you can manage the git-daemon-export-ok with daemon = yes17:24
raxor ah17:24
→ EmilMedve joined17:24
technoweenie for every_single_repo17:25
raxor <- types too slow17:25
raxor It looks like I still need the initd script though17:25
→ nessundorma joined17:26
technoweenie raxor: http://weblog.techno-weenie.net/2007/12/25/moving-to-git17:26
wrote up my experiences setting up the repos for my plugins. i also wrote a post-update hook that generates a public html file listing only public repos17:26
raxor ooh, thats nice17:27
I was actually reading that entry this morning :)17:27
← Ingmar left17:36
→ kirr joined17:43
← niki_ left17:44
← kirr left17:44
Ingmar_ → Ingmar17:48
raxor → njero18:02
← ropiku left18:05
ddaa is there an equivalent of "git rev-parse" that is usable with git:// urls?18:05
mh... wait18:06
Randal how would it know?18:06
it has to fetch them before it has info18:07
ddaa uh18:07
I'd imagine that the git:// server would be able to provide this sort of information.18:07
Randal that's not part of the protocol18:07
thiago ddaa: first you clone, then you parse18:07
ddaa: or, alternatively, run git-rev-parse via ssh on the remote host18:07
ddaa okay... so is there a network-only way of finding the hash of HEAD through git://?18:08
Randal not without fetching18:09
(is this mike on?)18:09
thiago git-ls-remote18:09
Randal ahh, there's that18:09
what's the application? why do you need to know?18:09
ddaa hacking on bzr-git18:09
Randal so how does this knowledge help?18:10
thiago is hacking on svn-git18:11
ddaa thiago: that's probably the other way around18:11
I'm working on the "make bzr embrace git" side not the other way around :)18:11
thiago I'm working on an svn-to-git conversion tool18:12
ddaa Randal: I'm still exploring. Bascally, I'm trying to get something like "bzr log git://repo.or.cz/HelloWorld.git" to work.18:12
there's some existing code, and I'm fixing one problem at a time18:12
the latest was that the existing code was getting the hash of HEAD, and that did not work with git://18:13
see, I'm not being very smart at the moment, just playing around.18:13
← sgrimm left18:13
ddaa one feature of bzr that some people like is the ability to have lightweight checkouts with a remote repository kind of like a svn checkout.18:14
Actually, you can get a bzr checkout from a svn repo, but that's another story.18:14
So I am working under the assumption that git:// would support that level of functionality.18:15
thiago you can get a shallow clone of git18:16
ddaa I got that impression18:16
but unless it's absolutely necessary, it would be better if bzr-git was able operate with not local .git at all.18:16
thiago not a good idea, IMHO18:18
that's almost perforce18:18
ddaa why would that be a problem?18:18
thiago contacting the server to obtain a "diff" ?18:18
unless I'm misunderstanding you18:18
ddaa a more realistic use case is "bzr branch git://foo", et voila, a bzr branch.18:18
thiago you'd want to fetch the necessary objects first18:19
ddaa thiago: I do not care about what is transmitted on the wire, I'd imagine git would be able to fetch remote packs that could be processed by the client-side git process as needed.18:20
thiago right18:20
← priidu__ left18:20
thiago but you should cache those objects first18:21
→ csc` joined18:21
thiago no need for a .git, but you should use whatever bzr uses for caching objects18:21
ddaa I see, you are talking about performance.18:22
well, in the "bzr branch git://" case, the cache would be the bzr repository.18:22
there can be a special bit of conversion code to handle the bridging if necessary18:23
njero Any way to trackdown why git is telling me fatal: The remote end hung up unexpectedly when I try to clone my repo... I am sure I have it setup wrong. netstat -plant says git-daemon is running18:23
dsop_ njero: did you run git-update-server-info in the repository on the server side?18:26
← bentob0x left18:26
njero dsop_: not that I know of... but I am trying to use gitosis to manage the repos... so maybe it does18:26
should I?18:26
→ anttih joined18:27
← csc` left18:29
→ Fullmoon joined18:32
→ Phoop joined18:38
← nessundorma left18:38
← stni left18:42
→ KirinDave joined18:43
→ k_dub joined18:47
← tclineks left18:54
ddaa thiago: apparently git-fetch-pack requires a local git repository to work18:55
→ archfear joined18:56
ddaa unless I'm missing something important, it seems quite hard (have to reimplement the clients) to use git in a network-only way.18:56
← HG` left19:00
→ hendry joined19:00
hendry i want to do something in git that in SVN i think is called changing the "roots"19:01
wayneeseguin svn switch --relocate ?19:01
hendry i have a faq.git on my local machine and i created a shared git repo on a remote machine19:01
the shared git repo is empty and i want to "clone", the local faq.git19:02
wayneeseguin: yeah, something like that19:02
perhaps a clone is what i need, but with "git --bare init --shared" type options19:02
wayneeseguin I use git to clone an svn repo19:03
git svn clone <url for svn repo>19:03
Ilari hendry: 'git push <URL> --all'?19:05
← archfear left19:12
→ fhobia joined19:16
→ captnmark__ joined19:17
← captnmark_ left19:18
← Fullmoon left19:18
→ sgrimm joined19:34
← captnmark__ left19:43
← anttih left19:52
← k_dub left19:52
← chris2 left19:56
hendry thank you Ilari20:00
→ csc` joined20:09
→ anttih joined20:13
→ tclineks joined20:16
→ pedro_mg joined20:27
← thiago left20:34
→ thiago joined20:36
→ desegnis joined20:51
→ nexneo joined20:52
← dsaxena left20:53
← fhobia left21:15
← ddaa left21:30
→ kumbayo joined21:30
→ HG` joined21:37
→ dsaxena joined21:58
← Phoop left22:01
← Yuuhi left22:01
← sverrej left22:03
← desegnis left22:04
← wagle left22:05
← anttih left22:11
→ wagle joined22:17
→ zdra joined22:18
← wagle left22:24
← Oejet left22:30
← mattly left22:30
→ wagle joined22:32
bdiego → midgar22:34
→ sr joined22:39
← HG` left22:41
← Dodji left22:43
← nexneo left22:43
→ chris2 joined23:02
dsop_ does somebody know how submodules and bare repositories work?23:13
or better question: is it possible to clone a repos that contains submodules together with submodules?23:13
just a question..hmm i should try that23:13
→ mattly joined23:13
vmiklos yes, it is23:14
cbrake → cbrake_away23:14
vmiklos have you read the submodules chapter of the user manual?23:15
dsop_ hmm not yet, i just used git-submodules manpages yet23:15
← zdra left23:18
→ orospakr joined23:18
← GyrosGeier left23:28
→ wagle_ joined23:39
← wagle left23:45
→ bdiego joined23:48
← fdr- left23:56
← cannonball left23:57
← ebel left23:58
→ fdr- joined23:58
← ferdy left23:59

Logs Search ←Prev date Next date→ Channels Documentation