IRCloggy #git 2018-08-30

Logs Search ←Prev date Next date→ Channels Documentation

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

2018-08-30

wreed joined00:00
LiftLeft left00:03
dsdeiz joined00:03
dsdeiz left00:03
dsdeiz joined00:03
LiftLeft joined00:06
diogenese joined00:06
m00sehead joined00:08
dsdeiz left00:08
tristanp left00:10
cdown_ joined00:10
cdown left00:12
cdown_cdown00:12
xaa left00:15
BallChang-CN left00:17
RoriconKnight left00:19
str left00:19
m0viefreak left00:22
mattcen left00:23
_ADN_ joined00:25
greatgatsby left00:26
CheckDavid joined00:26
mattcen joined00:27
CheckDavid left00:27
Kaisyu joined00:31
lagothrix left00:33
lagothrix joined00:35
BallChang-CN joined00:36
dsdeiz joined00:37
dsdeiz left00:37
dsdeiz joined00:37
ilmostro left00:39
ilmostro joined00:41
jasmith joined00:41
dsdeiz left00:42
zedek joined00:42
zedek ?00:42
yyy joined00:43
al-damiri left00:46
tristanp joined00:46
prakashdanish joined00:47
CheckDavid joined00:48
zedek left00:48
CheckDavid left00:48
tristanp left00:49
tristanp joined00:49
mud joined00:50
_ADN_ left00:53
Peetz0r left00:54
Peetz0r joined00:55
egc left01:03
tristanp left01:03
rosco_y left01:05
arecaceae left01:06
brettgilio left01:08
Narrat left01:12
KeyJoo left01:13
dsdeiz joined01:14
dsdeiz left01:18
dsdeiz joined01:18
dsdeiz left01:18
thiago joined01:20
Lunatrius joined01:20
XenophonF joined01:22
finalbeta joined01:23
yyy left01:27
F0rTex left01:28
wreed left01:28
F0rTex joined01:28
arecaceae joined01:28
James0r left01:31
jakogut_ joined01:37
watabou_ left01:39
watabou joined01:40
RuckusBringer joined01:41
TheRuckus left01:44
MACscr left01:46
str joined01:47
minionofgozer left01:51
minionofgozer joined01:53
dsdeiz joined01:55
dsdeiz left01:55
dsdeiz joined01:55
checkItOut joined01:57
str left01:57
CheckDavid joined01:58
CheckDavid left01:59
libregeekingkid left02:02
Goplat joined02:06
nowhere_man left02:09
nowhere_man joined02:10
igemnace joined02:12
PLLunlock joined02:12
raven-au left02:14
xirhtogal joined02:14
nowhere_man left02:15
nowhere_man joined02:16
lagothrix left02:17
raven-au joined02:24
kyan left02:27
James0r joined02:27
cagomez joined02:29
Praxi left02:30
Lyqyd joined02:32
Lyqyd left02:36
ryouma left02:37
Lyqyd joined02:37
sylario left02:38
prakashdanish left02:43
gendl Hi, how do I add a deploy key to a git repository? (i.e. hosting my own plain git repo, not on github or gitlab)02:45
hendry joined02:47
hendry say i want to back out of complex merge, how do i do that? reset?02:47
i think `git merge --abort`02:49
works02:49
CheckDavid joined02:50
CheckDavid left02:50
yyy joined02:52
inovas left02:57
yyy left03:00
tristanp joined03:01
inovas joined03:04
ilmostro left03:04
brettgilio joined03:05
tristanp left03:05
ilmostro joined03:06
kallisti5 left03:09
m00sehead left03:10
thiago gendl: the typical answer is "don't"03:11
nosferatuman joined03:11
texinwien joined03:12
applecrumble left03:13
brent__ left03:14
applecrumble joined03:14
Onionnion joined03:15
ixian left03:17
figurehe4d joined03:20
jasmith left03:20
yeshurun joined03:22
yeshurun left03:23
yeshurun joined03:23
texinwien left03:24
ixian joined03:26
zrc joined03:34
ekleog gendl: you may want to look at git-crypt, if you really want to do that03:38
Mattiaslndstrm left03:38
zrc left03:38
brettgilio left03:38
applecrumble left03:40
nosferatuman left03:40
jakogut_ left03:47
czart_ joined03:47
graphene left03:49
graphene joined03:50
houami joined03:51
gendl ekleog: thanks - but why can't I set up a read-only deploy key on my own git server?03:51
thiago you can. You shouldn't03:52
gendl Is that a service only offered by services such as github and gitlab?03:52
ekleog I don't really get what you're trying to do03:52
gendl should I also not do it with github03:52
ekleog if it's deploy-on-push, then you don't need to have the deploy key in the repo03:52
it's just a post-receive hook03:52
Wetmelon left03:53
ekleog ( https://git-scm.com/book/en/v2/Customizing-Git-Git-Hooks )03:53
(or maybe update hook would be easier, idk03:53
)03:53
Lunatrius` joined03:53
houami left03:54
gendl It's an automated build process on a client machine which should do a pull/clone from the remote before building03:54
so it needs read-only access to the repo.03:54
thiago the key should be out-of-band. Make the deploying system know it, without the repository.03:55
gendl just like is done with a deploy key on github or gitlab.03:55
thiago that's how you should do it03:55
Lunatrius left03:55
Lunatrius`Lunatrius03:55
thiago you can put the key in the repository. A file is a file, after all, and any file can be added and tracked. That's just a bad idea.03:55
gendl on the repository machine it would be just a public key03:55
the private key is on the client build machine03:55
mud left03:56
gendl I just got it working by appending the public key to the ~git/.ssh/authorized_keys file on the remote repository machine03:56
libregeekingkid joined03:56
gendl but I just can't believe that's how a huge service like github would do it03:56
they have millions of users, they can't have millions of public keys appended to their ~git/.ssh/authorized_keys file...03:57
thiago but they do03:57
gendl srsly?03:57
I thought it can also go somewhere in each repository configuration...03:58
thiago gitolite uses a single user SSH login, with users identified by their keys03:58
gitlab does too03:58
ekleog gogs has all public keys to ~git/.ssh/authorized_keys too03:58
thiago I push to [email@hidden.address] It's the key that identifies me, not something else03:58
github has usernames in SSH, though whether that is important or not and how they organise is entirely unknown03:59
ekleog hmm? I push to [email@hidden.address] personally03:59
gendl Ok. Go figure. Thanks guys.03:59
kbZ joined03:59
thiago oh, right03:59
sorry, I'm confusing with some other service then04:00
so, yes, the SSH username is irrelevant. The key identifies you.04:00
ilmostro left04:00
ekleog hmm, I think you're confusing with github-over-http, which uses username+password for auth :)04:00
thiago yeah04:00
CheckDavid joined04:00
thiago note that GitHub does not use openssh, so there's no ~git/.ssh/authorized_keys04:00
CheckDavid left04:01
thiago gitlab does04:01
ilmostro joined04:02
rhinux left04:02
Jonno_FTW hello04:02
I have 2 branches, master and dev, how can I only merge in specific changes into master?04:03
can I make all diffs a conflict when merging?04:03
Darren joined04:03
DarrenGuest6253104:03
tristanp joined04:04
ismithers joined04:04
ismithers Hey all, is there a way for me to git add -i or -p and have them ignore whitespace changes?04:04
mrCyborg|zzz left04:05
mrCyborg|zzz joined04:05
ismithers Oh I may have found it: --ignore-space-change04:06
tristanp left04:07
tristanp joined04:07
tango_ left04:07
tango_ joined04:09
boombatower left04:16
serialoverflow left04:17
cagomez left04:20
ShapeShifter499 joined04:22
oojacoboo joined04:22
serialoverflow joined04:23
dry left04:26
igemnace left04:27
wpcarro left04:36
libregeekingkid left04:38
thiago Jonno_FTW: Git won'th elp you. You can only merge full branches.04:40
more specifically, whenever you merge a commit, you merge all commits leading to it04:40
Lyqyd left04:42
Bucciarati left04:44
zsoc left04:47
avar left04:48
CheckDavid joined04:51
CheckDavid left04:52
Jonno_FTW thiago: so I have to merge them into a new branch and pick out all the changes myself?04:53
tristanp left04:56
Envil joined04:57
ferdna left04:59
zumba_ad_ joined04:59
wpcarro joined05:00
Oatmeal joined05:00
tsdh joined05:01
pks_ joined05:01
zumba_ad_ left05:04
pks left05:04
pks_pks05:04
Lunatrius left05:05
kotepillar_ joined05:07
kotepillar_ when I with my friends work on one private project, “Forking with feature branches” or “create new feature branch directly”, which one is better05:08
g00s joined05:08
teodorg joined05:09
Lunatrius joined05:11
oojacobo_ joined05:11
avar joined05:12
avar left05:12
avar joined05:12
Bucciarati joined05:12
kotepillar_ left05:12
freimatz joined05:14
prakashdanish joined05:14
oojacoboo left05:14
kotepillar_ joined05:15
kotepillar_ when I with my friends work on one private project, “Forking with feature branches” or “create new feature branch directly”, which one is bette05:15
zumba_ad_ joined05:16
orbyt_ joined05:17
zumba_ad_ left05:20
jenwhite joined05:20
Renter_ joined05:20
schleppel joined05:22
Hounddog joined05:23
orbyt_ left05:24
kotepillar_ left05:24
dry joined05:25
kotepillar_ joined05:25
dreiss joined05:26
sauvin joined05:27
Guest62531 left05:28
hendry kotepillar_: forking is more the usual way05:29
kotepillar_ hendry: forking how to rebase05:30
Darren joined05:30
Darren left05:30
hendry kotepillar_: you don't need to rebase, you merge to keep upto date with master05:30
zlsyx left05:31
Darren joined05:31
Darren left05:31
zlsyx joined05:31
genken joined05:32
Darren joined05:33
Darren left05:33
dreiss left05:33
dc13ff joined05:33
oojacobo_ left05:33
Darren joined05:34
Darren left05:34
kotepillar_ but the branch tree will be cross lots of commits05:34
not very clean05:35
LiftLeft left05:35
KeyJoo joined05:35
brettgilio joined05:35
cur8or joined05:36
graphene left05:36
graphene joined05:37
Darren joined05:37
Darren left05:37
Darren joined05:40
Darren left05:40
Darren joined05:42
Darren left05:42
zsoc joined05:42
zsoc left05:42
zsoc joined05:42
zlsyx left05:44
freimatz left05:44
zlsyx joined05:44
Darren joined05:45
Darren left05:45
checkItOut left05:45
Bobguy left05:46
Darren joined05:46
Darren left05:46
hendry kotepillar_: big projects operate this well. merging is usually pretty painless if you are not working on the exact same file05:47
Darren joined05:48
Darren left05:48
brettgilio left05:48
Darren joined05:49
Darren left05:49
jgornick joined05:50
Darren joined05:51
Darren left05:51
graphene left05:51
johnjay yeah does git give an error if commits conflict?05:51
muelleme joined05:52
Darren joined05:52
Darren left05:52
jacksparling joined05:53
Darren joined05:54
dsdeiz left05:54
Darren left05:54
jacksparling left05:55
Darren joined05:55
Darren left05:55
hendry johnjay: yes05:55
LiftLeft joined05:56
checkItOut joined05:56
Darren joined05:57
Darren left05:57
johnjay ok. i assume it would just go with whichever one was made first and then give an error then05:57
of course when you have commits like "changed *p = 7 to p = 7" it's hard to have a conflict05:58
soee joined05:58
Darren joined05:58
Darren left05:58
Darren joined06:00
Darren left06:00
sz0 joined06:00
segnior left06:01
Darren joined06:01
Darren left06:01
CheckDavid joined06:02
kurkale6ka joined06:03
Darren joined06:03
CheckDavid left06:03
Darren left06:03
LiftLeft left06:04
Darren joined06:04
Darren left06:04
Darren joined06:06
Darren left06:06
PLLunlock left06:06
LiftLeft joined06:07
Darren joined06:07
Darren left06:07
wandering_vagran joined06:08
figurehe4d left06:09
Darren joined06:09
Darren left06:09
Darren joined06:10
soee_ joined06:10
Darren left06:10
soee left06:10
Darren joined06:12
chele joined06:12
Darren left06:12
lowbro joined06:14
lowbro left06:14
lowbro joined06:14
kurkale6ka left06:14
Darren joined06:15
jenwhite left06:15
Darren left06:15
cur8or left06:16
Darren joined06:18
Darren left06:18
Darren joined06:19
Darren left06:19
Darren joined06:21
Darren left06:21
teodorg left06:22
GreenJello left06:23
teodorg joined06:23
Sasazuka joined06:24
Darren joined06:24
Darren left06:24
Sasazuka left06:24
Darren joined06:25
Darren left06:25
Darren joined06:27
Darren left06:27
kbZ left06:28
Hi-Angel joined06:28
Darren joined06:28
Darren left06:28
ocds joined06:29
Darren joined06:33
Darren left06:33
segnior joined06:34
Masplus joined06:34
Darren joined06:34
Darren left06:35
floatleft joined06:35
floatleft left06:35
Darren joined06:36
Darren left06:36
ShapeShifter499 left06:36
ShapeShifter499 joined06:36
Darren joined06:37
Darren left06:38
Darren joined06:39
Darren left06:39
jakogut_ joined06:39
Darren joined06:42
TinkerT left06:42
dakar left06:42
Darren left06:42
Darren joined06:43
MACscr joined06:43
Darren left06:44
TinkerT joined06:44
dakar joined06:44
minionofgozer left06:44
dreiss joined06:45
Darren joined06:45
Darren left06:45
muelleme left06:45
Darren joined06:48
Darren left06:48
Darren joined06:49
Darren left06:50
ocds left06:50
minionofgozer joined06:51
dmone joined06:51
Darren joined06:51
pfleck joined06:51
Darren left06:51
abba joined06:51
genken left06:52
agris joined06:52
Darren joined06:52
agris test06:52
Darren left06:53
_ikke_ !hi06:53
gitinfo [!welcome] Welcome to #git, a place full of helpful gits. If you have a question, feel free to just go ahead and ask—somebody should answer shortly. For more info on this channel, see https://gitirc.eu/ - Take backups (type !backup to learn how) before taking advice.06:53
thiago left06:53
CheckDavid joined06:53
agris Can someone please direct me to a good guide on how to properly setup a GIT server? I've setup cgit but apparently that only views repos. I found out i actually have to have a daemon of some kind.06:54
CheckDavid left06:54
Darren joined06:54
Darren left06:54
agris Also, is it bad to have an open-access-commit server?06:54
_ikke_ agris: I would say yes06:55
Darren joined06:55
Darren left06:56
_ikke_ agris: So you want everyone to be able to push to your repo?06:56
tristanp joined06:56
zlsyx left06:57
Darren joined06:57
Darren left06:57
agris I've just never done this before. What's the proper way to go about easily accepting patches should someone want to contribute06:57
Darren joined06:58
Darren left06:59
Poffer joined06:59
_ikke_ agris: That person could publish their copy of the repository and ask you to pull from them06:59
That's the original decentralized workflow of git07:00
agris wow that is better than sending patches07:00
Darren joined07:00
Darren left07:00
agris I'd love to be able to do that for projects on github.07:00
Instead of owning an account there07:00
gxt left07:01
_ikke_ agris: nothing prevents you from doing that07:01
zefferno joined07:01
tristanp left07:01
_ikke_ agris: in your local repository, you just add their repository as a remote07:01
and fetch from it07:01
Darren joined07:02
Darren left07:02
agris so how do i go about setting up my own git server?07:03
Puffball joined07:03
Darren joined07:03
Darren left07:03
agris Please excuse my greenness. We were all green at some point07:04
Darren joined07:05
Darren left07:05
jakogut_ left07:05
roncone joined07:05
_ikke_ there are 3 main ways to host repositories: 1) via ssh (read-write, authenticated only). 2) http(s) (read-only, unless you use the git http back-end). 3) git deamon (read-only, public)07:05
no worry07:05
Darren joined07:06
Darren left07:06
_ikke_ https://git-scm.com/book/en/v2/Git-on-the-Server-The-Protocols07:07
Darren joined07:08
Darren left07:08
agris wait, so what your saying is all i have to do is set origin to ssh://my.server.tld:~/Public/git/reponame and then tell cgit to view that directory for public read-only access?07:08
It's that simple?07:08
pfleck left07:09
_ikke_ for 'dumb' http to work you have to do a bit of additional work07:09
Darren joined07:09
agris oh?07:09
Darren left07:09
_ikke_ Read that page I linked to, it explains the details07:09
dsdeiz joined07:09
dsdeiz left07:09
dsdeiz joined07:09
agris thank you _ikke_07:10
I can't believe I was using git in such a centralized manner before when it can be used like this07:10
Darren joined07:11
Darren left07:11
cur8or joined07:11
_ikke_ git was meant to be very decentralized, but people found it easier to use a centralized workflow07:11
g00s left07:12
dreiss left07:12
Darren joined07:12
Darren left07:12
fr33domlover left07:13
texinwien joined07:13
Poffer left07:13
Darren joined07:13
Darren left07:14
Darren joined07:15
Darren left07:15
m1m3-50 left07:16
Darren joined07:16
Darren left07:17
kakashiAL joined07:17
Darren joined07:18
Darren left07:18
Darren joined07:19
Darren left07:20
rtjure left07:20
Darren joined07:21
Darren left07:21
sysvalve joined07:21
Tobbi joined07:22
agris if i am pushing to a empty directory over SSH what is the proper way to do that? it complains because a project.git file does not exist07:22
does not exist yet on the remote repository07:22
cbreak agris: git init --bare on the remote07:22
charolastra joined07:22
Darren joined07:22
cbreak that will create a repository you can push to07:23
Darren left07:23
cbreak agris: git obviously only allows pushing to git repositories07:23
Darren joined07:24
Darren left07:24
Onionnion left07:25
Darren joined07:25
Darren left07:26
Darren joined07:27
Darren left07:27
Codaraxis_ left07:28
Darren joined07:29
blackmesa joined07:29
Darren left07:29
Darren joined07:30
Darren left07:30
gxt joined07:30
Codaraxis joined07:31
Darren joined07:31
Darren left07:32
skisela joined07:32
dsx_ left07:33
Darren joined07:33
Darren left07:33
Darren joined07:35
Darren left07:35
Tobbi left07:35
skisela_afk joined07:36
Darren joined07:36
Puffball left07:36
Darren left07:36
Darren joined07:38
Darren left07:38
skisela left07:39
Darren joined07:41
Darren left07:41
m1m3-50 joined07:41
blackmesa left07:42
agris left07:42
plexigras joined07:42
Darren joined07:42
m1m3-50 left07:42
agris joined07:42
DarrenGuest975007:42
Guest9750 left07:43
m1m3-50 joined07:43
bsaboia joined07:44
Darren_ joined07:44
Darren_ left07:44
Darren_ joined07:46
Darren_ left07:46
mmmnootka joined07:46
Darren_ joined07:48
Darren_ left07:48
agris cbreak, the files are not showing up07:49
Darren_ joined07:49
Quetzal2 joined07:50
absence is it possible to visualise the relationship between two specific branches with git log --graph? it's a convoluted mess when all the branches are shown07:50
Darren_ left07:50
cbreak agris: which files?07:50
agris and git init --bare keeps trying to put the management files in . instead of .git . i tried git init --bare .git but the files still don't show07:50
cbreak agris: you should see HEAD at least07:50
agris: and also directories refs, objects07:50
agris: you don't see those files?07:51
mmmnootka left07:51
cbreak agris: git log will only show the branches you tell it07:51
sorry, absence07:51
agris i see the commit messages but i don't see any of the files in that actual dir07:51
Darren_ joined07:51
Darren_ left07:51
cbreak agris: you should see the files HEAD07:51
agris: and directories objects, refs and so on07:51
ismithers left07:52
cbreak agris: if you see those, and git log works, then everything seems fine07:52
don't worry07:52
agris i don't though07:52
cbreak you don't see HEAD?07:52
agris i see the folder head but my code isn't in there07:53
Darren_ joined07:53
cbreak it should be directly in the repo directory you did the git init --bare07:53
Darren_ left07:53
cbreak yeah, that's normal07:53
don't worry about it07:53
you don't need it in a server repo07:53
it's efficiently stored in gits object database07:53
probably in a pack file07:53
agris does it have to be that way?07:53
cbreak yes07:53
that's normal07:53
server repositories are bare07:54
they don't have a working directory with loose files07:54
that's why you can push to them safely07:54
and they're obviously also more efficient :)07:54
Darren_ joined07:54
Darren_ left07:54
agris so how do i get my files out?07:55
Darren_ joined07:55
_ikke_ agris: when you clone from this repository, you get a normal repository with all the files there07:55
Darren_ left07:56
cbreak agris: the usual way is to clone07:57
agris I don't like this at all07:57
cbreak but you can also use git archive07:57
agris: why?07:57
Darren_ joined07:57
Darren_ left07:57
cbreak this is safe, efficient, fast07:57
_ikke_ Because I guess they want to serve their project from this reposiory07:57
repository07:57
Goplat left07:57
cbreak that sounds idiotic07:58
someone didn't read !deploy :(07:58
gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html07:58
agris cgit can't read this07:58
Darren_ joined07:58
cbreak if cgit can't handle git repositories07:59
then it's broken07:59
Darren_ left07:59
_ikke_ It should be able to handle bare repos07:59
cbreak I remember it being able to handle it half a decade ago07:59
absence cbreak: by default there are all kinds of branches visible. if i scroll down, the whole screen is basically full of | characters from all the branches07:59
cbreak but I haven't tried it in ages07:59
absence: you should only see what you tell git07:59
absence: git log --graph master should only show master branch07:59
so only what is reachable from master08:00
nothing else08:00
serialoverflow left08:00
Darren_ joined08:00
Darren_ left08:00
sh3iny left08:03
andres left08:03
absence cbreak: it starts off at master, but there are lots of lines if i scroll down08:10
cbreak that's to be expected08:10
your master likely has lots of history merged in08:10
_ikke_ --simplyfy-by-decoration08:10
cbreak but no worries, this is also part of master's history08:10
Darren_ joined08:10
Darren_ left08:10
Darren_ joined08:10
Darren_ left08:10
andres_f joined08:10
sylario joined08:10
mikecmpbll joined08:10
CheckDavid joined08:10
manuelschneid3r joined08:10
MACscr left08:10
Darren_ joined08:10
Darren_ left08:10
roro247 joined08:10
czart_ left08:10
Renter joined08:10
sea-gull_ joined08:10
hendry left08:10
Darren_ joined08:10
Darren_ left08:10
serialoverflow joined08:10
Darren_ joined08:10
Darren_ left08:10
dex1983 left08:10
Renter_ left08:10
MACscr joined08:10
sylario left08:10
sylario joined08:10
CheckDavid left08:10
CheckDavid joined08:10
dex1983 joined08:10
xcm left08:10
sh3iny joined08:11
czart joined08:11
xcm joined08:11
jgornick left08:12
Darren_ joined08:13
Darren_ left08:14
jgornick joined08:15
Darren_ joined08:15
kurkale6ka joined08:15
Darren_ left08:15
masterasia joined08:15
masterasia Hmm.. why does git rebase --abor08:16
work08:16
quxgyver joined08:16
jast what do you mean? because someone implemented it ;)08:16
masterasia I mistyped but it looks like git is smart enough to determine that --abor or even --abo represents --abort08:16
jast oh08:16
that's common with option parsers... any prefix that is unambiguous works08:16
Darren_ joined08:17
jast GNU tools usually do this, too08:17
Darren_ left08:17
masterasia Heh08:17
I think it shouldn't do that08:17
because it isn't documented anywhere08:18
Thanks jast08:18
jast at this point it's pretty much convention in the linux world08:18
something that everyone Just Knows(tm) - or doesn't ;)08:18
Darren_ joined08:18
Darren_ left08:18
masterasia well I guess I'm joining the realm of Just Knows(TM)08:18
jast welcome to the realm!08:19
man, this Darren guy needs to fix his IRC client08:19
let's see if a short ban will help08:19
xaa joined08:19
osse $ git fetch --prun08:19
error: Ambiguous option: prun (could be --prune or --prune-tags)08:20
Darren_ joined08:20
osse oh08:20
nvm08:20
Darren_ left08:20
ChanServ set mode: +o08:20
jast set mode: +b08:20
osse I thought only git's shell scripts did that08:20
jast it's a standard feature of getopt, for instance08:20
osse git isn't consistent about it08:21
jast I think some of the shell scripts have their own "parsers"08:21
artemis joined08:21
amirpro left08:21
Tobbi_ joined08:22
dre_ joined08:22
supernovah left08:23
artemis left08:24
crane left08:29
stoopkid left08:29
Garen left08:30
jast set mode: -b08:31
Darren_ joined08:32
Darren_ left08:32
Darren joined08:33
Darren left08:33
kerframil left08:34
Darren joined08:35
Darren left08:35
agris ok nevermind cgit can handle bare repos08:36
I've just never seen a bare repo before this08:36
mooasaurus left08:36
Darren joined08:36
Darren left08:36
mooasaurus joined08:37
ilmostro left08:37
Darren joined08:37
Darren left08:38
Darren joined08:39
Darren left08:39
ilmostro joined08:39
Darren joined08:41
Darren left08:41
lanlink joined08:41
Darren joined08:42
Darren left08:42
manatee_ joined08:44
Darren joined08:44
Darren left08:44
manatee_ left08:44
Darren joined08:45
Darren left08:45
Darren joined08:46
Darren left08:47
_ADN_ joined08:48
Darren joined08:48
Darren left08:48
Darren joined08:49
Darren left08:50
Darren joined08:51
Darren left08:51
Snakepit joined08:51
Puffball joined08:52
Darren joined08:53
Darren left08:53
Garen joined08:53
Snakepit left08:54
Darren joined08:54
Darren left08:54
kes joined08:55
Darren joined08:56
Darren left08:56
texinwien_ joined08:57
Darren joined08:57
Darren left08:57
tristanp joined08:58
Darren joined08:59
Darren left08:59
texinwien left09:00
Raimondii joined09:01
Darren joined09:02
Darren left09:02
kurkale6ka left09:02
kurkale6ka joined09:03
Darren joined09:03
Darren left09:03
tristanp left09:03
Raimondi left09:04
RaimondiiRaimondi09:04
_flow_ left09:04
sQVe joined09:04
Darren joined09:05
logithack joined09:05
Darren left09:05
Hi-Angel How do I git-push only the current branch, but without explicitly writing its name every time? Something like "git push origin --current" would be nice. Can't find it in man09:06
Darren joined09:06
cur8or left09:06
Darren left09:06
osse Hi-Angel: you can use HEAD09:07
maybe even @, not sure09:07
Hi-Angel osse, thanks!09:07
osse Hi-Angel: add -u to the command line, so that git will configure it. after that you can do 'git push'09:07
Darren joined09:08
logithack i'd like to show all files in a repo that are currently being tracked. i've found the command "git ls-tree -r master --name-only", but it gives me "fatal: Not a valid object name master". any ideas?09:08
Darren left09:08
Live joined09:08
Live left09:08
Live joined09:08
cur8or joined09:08
osse logithack: there is no master branch, it seems09:09
Darren joined09:09
Darren left09:09
osse logithack: provide a valid branch name. see git branch -a09:09
logithack osse: well, ive cd'ed into the directoy with git bash and the prompt shows "<dir> (master)"09:10
git branch -a gives me no output09:10
osse logithack: then perhaps the repo is completely empty09:11
logithack: what does git status say ?09:11
wildermind joined09:12
Darren joined09:12
Darren left09:12
logithack it says "On branch master", "No commits yet", "Changes to be committed:" shows a number of "new file"s in green colour09:12
osse that explains it09:13
in an empty repo there are no branches. it just says on master because that is the default branch name. that branch is created as soon as you make a commit09:13
dc13ff left09:13
osse if the ls-tree command had worked, it wouldn't have listed anything anyway because there is no content09:13
logithack osse: i'm completely new to this, i'd like to create a local repository from the directory i'm in and then push it to gitlab, creating a new remote repo. my steps so far: cd'ed into dir, "git init", "git add"ed everything inside it.09:14
osse So far, so good. Now do git commit09:14
Darren joined09:14
logithack so the next step is to do a commit, right?09:14
good09:14
Darren left09:14
lanlink left09:14
logithack oh, one question first though. theres a subdir, which i added with "git add <subdir>". there are some files in there though which i dont want to be tracked. how can i remove them before committing?09:15
DirkjanH joined09:15
osse logithack: git reset file(s)09:15
zizo joined09:15
Darren joined09:15
Darren left09:15
osse git status prints usage for the common commands above each group of files09:15
soee_ hi, if i have branch B created from master and it has several chnages i i would now like to add new chnages from master to my branch B, how would i do that properly?09:15
logithack osse: is it possible to add the subdir and only the subdir, ignoring all the files currently in it?09:16
because now its removed my subdir entirely09:17
osse logithack: no. git doesn't let you add empty directories09:17
you can put a dummy file in it09:17
logithack osse: ah, you mean a .gitkeep file09:18
catsup left09:18
Darren joined09:18
osse for example09:18
catsup joined09:18
Darren left09:18
osse but in my opinion the better choice is to just not add it09:18
catsup left09:19
catsup joined09:19
Darren joined09:20
Darren left09:20
osse if you need it to build/run then make the build system or the program itself create it09:20
but that's just my opinion. and I'm stypid09:20
Darren joined09:21
Darren left09:21
Darren joined09:23
Darren left09:23
logithack osse: haha, youre not. you helped me a great deal, thanks!09:23
doev joined09:23
goweol left09:25
zizo Hi all, complete noob here :) Probably I'm using the wrong tool, but please let me know if what I'm trying to do is completely wrong... I've many linux machines on my network, with many common scripts and programs that helps the normal routine. Not all files are on all machines, and it is what I want. So I thought to set up a git repo where all can push but a script fetches and checkouts only existing files. The09:25
problem is that when pushing from machine x, with x that has only a small amount of files, it deletes all the files missing locally but found in remote. Can I avoid it? Is there a way only for "push-add" instead of standard push? Thanks for your help09:25
Darren joined09:26
Darren left09:26
psprint Hey, an alternate Git progress bar: https://asciinema.org/a/pmMhiH3jSULsuWg7rmj4uAQQE09:27
Darren joined09:27
Darren left09:27
fatalhalt left09:28
psprint I would want to write to [email@hidden.address] (or what was the address), with feature request: core.progress_pipe , to be able to set it to above script09:29
_ikke_ [email@hidden.address]09:29
psprint a config option09:29
thanks09:29
_ikke_ If you're able to, it's even better to provide a patch09:29
Darren joined09:30
Darren left09:30
psprint Hmm, maybe, it seems easy to popen() the script, write what would have been written to stderr, read from popen, write to stderr. But adding a config option, I could have problems with this09:31
i7c zizo, you cannot have "partials" of the tracked files with git. If you track a file, it will be there when you checkout the repository, if you delete it and commit+push the deletion it will be gone in that next revision, for everyone that checks it out.09:31
Darren joined09:32
Darren left09:32
_ikke_ psprint: https://github.com/git/git/blob/master/Documentation/technical/api-config.txt09:32
Darcidride joined09:32
psprint thanks09:32
i7c zizo, you could checkout the whole repository in one location and only symlink the parts in-place that you actually want. But afaik there’s now way to checkout partials like that.09:33
libregeekingkid joined09:33
Darren joined09:33
Darren left09:33
dsdeiz left09:34
Darren joined09:35
Darren left09:35
zizo i7c: thanks for your help, the last one is a good suggestion09:35
Darren joined09:36
Darren left09:36
Darren joined09:38
Darren left09:38
i7c Glad if it helps you. Ofc you can also use branches, but it might be extremly expensive in terms of effort to maintain one branch per machine ...09:38
zizo i7c: i found something about the sparse checkout, it seems to fit my requests, let me read about it...09:38
Darren joined09:39
Darren left09:39
texinwien_ left09:40
Darren joined09:41
Darren left09:41
planetcall|work joined09:41
texinwien joined09:41
Darren joined09:42
Darren left09:42
kotepillar_ left09:43
Darren joined09:45
Darren left09:45
Darren joined09:47
Darren left09:47
Darren joined09:48
Darren left09:48
Darren joined09:50
Darren left09:50
Darren joined09:51
psprint if one would spare a $10 donation I would code the progress-pipe patch quickly09:51
Darren left09:51
Darren joined09:53
Silenced joined09:53
Darren left09:53
Darren joined09:54
Darren left09:54
tty left09:55
Darren joined09:56
Darren left09:56
kakashiAL in git terms, what does09:56
slacker_nl joined09:56
kakashiAL working tree or working directory means?09:56
does it mean the .git folder?09:56
_ikke_ The files you actually work on09:56
no09:57
Darren joined09:57
_ikke_ "The tree of actual checked out files. The working tree normally contains the contents of the HEAD commit’s tree, plus any local changes that you have made but not yet committed." according to man gitglossary09:57
gitinfo the gitglossary manpage is available at https://gitirc.eu/gitglossary.html09:57
Darren left09:57
kakashiAL _ikke_: lets say I cd my_git_test, and do git initi --bare test_repo.git09:58
_ikke_ A bare repository doesn't have a working tree09:58
kakashiAL what does it mean for the my_get_test folder?09:58
_ikke_ That's just the repository directory09:59
kakashiAL it still have a .git folder, right?09:59
_ikke_ no09:59
it *is* the .git folder basically09:59
kakashiAL does git creates some hidden folder in this folder (in my case my_git_test ?09:59
_ikke_ no10:00
kakashiAL hmm...so git saves that internaly?10:00
Darren joined10:00
Darren left10:00
_ikke_ no, everything is in there10:01
lankanmon left10:01
_ikke_ try to execute git init --bare my_git_test.git and check the contents of that folder10:01
hussam left10:01
Darren joined10:02
Darren left10:02
kakashiAL _ikke_: okay, as you mentioned it contains all the files that .git has10:02
zulutango joined10:03
kakashiAL now, as I know you cant execute git commit in that folder10:03
kini left10:03
_ikke_ no, correct10:03
a bare repository is not meant to directly work with10:03
Darren joined10:03
Darren left10:03
kakashiAL _ikke_: yes, it is only used for a shared repository10:04
you only can push stuff in it10:04
supernovah joined10:04
kakashiAL not commit10:04
_ikke_: for example if you want to create a shared-server-repository10:04
Darren joined10:05
Darren left10:05
kakashiAL and again, it is "commits protected"10:05
DirkjanH left10:05
kakashiAL _ikke_: is that correct so far?10:05
DirkjanH joined10:05
Darren joined10:06
Darren left10:06
Darren joined10:08
Darren left10:08
hussam joined10:09
Darren joined10:09
i7c You are inventing new terms ("commit protected") which are not well-defined and ask if they are correct.10:09
Darren left10:09
DirkjanH left10:10
kini joined10:10
kakashiAL i7c: ahh okay10:10
i7c That’s the response: 12:03 <_ikke_> a bare repository is not meant to directly work with10:10
kakashiAL with "commit protected" I mean "you cant do commits inside"10:10
you only can push10:10
Darren joined10:11
Darren left10:11
i7c You can probably fiddle-in a commit with plumbing ("low-level") commands, though I’m not sure of it. But yes, it’s not meant for that, as you said. It’s meant to be pushed to and fetched from.10:12
Bucciarati left10:12
i7c I’m just answering this way, because "protected" sounds like a security measure. And there is none.10:12
Darren joined10:12
CodeSlingerPaul joined10:12
Darren left10:12
i7c If somebody has access to the bare repo, they can tinker stuff.10:13
There is no secret protection.10:13
Darren joined10:14
Darren left10:14
Kaisyu left10:14
Darren joined10:15
Darren left10:15
Darren joined10:17
Darren left10:17
spY|da joined10:17
Darren joined10:18
Darren left10:18
Darren joined10:20
Darren left10:20
Darren joined10:21
Darren left10:21
kurkale61 joined10:22
kurkale6ka left10:22
Darren joined10:23
Darren left10:23
Bucciarati joined10:23
Darren joined10:24
Darren left10:24
oxymoron93 joined10:25
kakashiAL i7c: so with other words, I as "Bruce Wayne" create a bare repo, I as "Bruce Wayne" push a new repo to it to master and I as "Bruce Wayne" can decide what to merge10:26
Darren joined10:27
Darren left10:27
kurkale61 left10:29
jast git itself doesn't decide who is allowed to push to a repo10:30
jast set mode: -o10:30
Darren joined10:30
yyy joined10:30
jast in a simple setup, it's decided by filesystem permissions (e.g. you push via SSH and your SSH user's permissions determine what repos you can update)10:30
Darren left10:30
jast more fancy setups for git servers usually check your permissions in their own way in a pre-receive/update hook or in whatever ways they can think of :)10:31
Darren joined10:32
dpyro left10:32
Darren left10:32
Darren joined10:33
Darren left10:33
kakashiAL jast: okay, but if I create a bare repo as "Bruce Wayne", amd I not the owner of that repo?10:35
Darren joined10:35
Darren left10:35
kurkale6ka joined10:35
kakashiAL and am I not the one who can decide what to merge and what not?10:36
dre_ left10:36
tmm88 joined10:36
Darren joined10:36
Darren left10:36
Darren joined10:38
Darren left10:38
agowa338 joined10:38
Darren joined10:39
Darren left10:39
lankanmon joined10:41
tomfarr joined10:41
Darren joined10:42
Darren left10:42
Darren joined10:44
Darren left10:44
Darren joined10:45
Darren left10:46
karstenk joined10:46
karstenk I try to pull a project and ignoring a file I have changed. But I write the file with full relative path to .gitignore and .git/info/exclude , but it still want to merge or not to continue befor stashed10:48
Darren joined10:48
Darren left10:48
Darren joined10:50
Darren left10:50
bremner karstenk: gitignore does not work for tracked files10:51
Darren joined10:51
bremner karstenk: sounds like you need to read !config10:51
gitinfo karstenk: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas10:51
Darren left10:52
wandering_vagran left10:52
Darren joined10:53
Darren left10:53
Wetmelon joined10:53
jast kakashiAL: depends. Git itself doesn't manage access rights at all. if you create a bare repo on your own local filesystem, then filesystem permissions decide who owns the repo and thus who can change it (e.g. push to it). on github, for example, the SSH/HTTPS service they wrap around git decides whether you're allowed, based on your authentication.10:55
livcd left10:56
Darren joined10:56
Darren left10:56
Mimicize joined10:57
libregeekingkid left10:57
fission6 joined10:58
hoek_ left10:58
tmm88 left10:58
hoek_ joined10:58
hoek_ left10:58
hoek joined10:58
hoek left10:58
elichai2 joined10:58
hoek joined10:59
Darren joined10:59
Darren left10:59
m1m3-50 left11:00
Darren joined11:00
Darren left11:01
throstur_tp joined11:03
Darren joined11:03
Darren left11:04
paco joined11:04
Darren joined11:05
Darren left11:05
Darren joined11:06
Darren left11:07
checkItOut left11:07
DirkjanH joined11:08
Darren joined11:08
Darren left11:08
Darren joined11:09
Darren left11:10
Darren joined11:11
Darren left11:11
Darren joined11:12
Darren left11:13
tmm88 joined11:13
Darren joined11:14
Darren left11:14
amirpro joined11:14
leeN joined11:15
Darren joined11:15
Darren left11:16
ilmostro left11:16
c0c0 joined11:16
Darren joined11:17
Darren left11:17
Darren joined11:18
Darren left11:19
Darren joined11:20
Darren left11:20
kotepillar_ joined11:20
Darren joined11:21
Darren left11:22
Silenced left11:22
Darren joined11:23
kfoonamalik left11:23
Darren left11:23
inovas left11:24
miczac joined11:24
Darren joined11:24
Darren left11:25
Silenced joined11:26
Darren joined11:26
Darren left11:26
Darren joined11:27
Darren left11:28
KBme joined11:28
KBme hello11:28
Darren joined11:29
inovas joined11:29
Darren left11:29
KBme I just got an ambiguous object name error from git, so I've started converting my code to using explicit object names like refs/heads/mybranch or refs/tags/mytag. My issue is that now I try to git branch -D refs/heads/mybranch and git says it can't find that branch.11:29
git show-ref does show refs/heads/mybranch11:30
and it does show up in git branch11:30
Darren joined11:30
Darren left11:31
osse KBme: i think git branch unconditionally prepends "refs/heads" to stuff11:31
KBme so it's an inconsistent naming of branches11:31
sQVe left11:32
Darren joined11:32
Darren left11:32
cdown left11:33
KBme but yes that does seem to be the case11:33
Darren joined11:33
Darren left11:34
Darren joined11:35
Darren left11:35
Darren joined11:36
Darren left11:37
_ikke_ git branch is porcelain11:37
KBme: use git update-ref instead11:37
(porcelain, being meant for end users)11:38
Darren joined11:38
kakashiAL left11:38
Darren left11:38
Darren joined11:39
Darren left11:40
Darren joined11:41
James0r left11:41
Darren left11:41
KBme ah uh11:41
_ikke_, okay thanks i'll check11:41
Darren joined11:42
Darren left11:43
Darren joined11:44
Darren left11:44
timj joined11:44
agowa338 left11:45
Darren joined11:45
Darren left11:46
Darren joined11:47
Darren left11:47
Darren joined11:48
Darren left11:49
KBme thanks, bye.11:49
KBme left11:49
ubuntuisloved joined11:49
Darren joined11:50
Darren left11:50
kakashiAL joined11:51
Darren joined11:51
str joined11:52
Darren left11:52
lkurusa left11:52
paco left11:52
Darren joined11:53
Darren left11:53
Masplus left11:54
Darren joined11:54
Darren left11:55
e_l1997 joined11:55
Darren joined11:56
Darren left11:56
tvw joined11:57
cdown joined11:57
Darren joined11:57
Darren left11:58
amirpro_ joined11:58
_flow_ joined11:58
Darren joined11:59
Darren left11:59
cr1mson_king joined12:00
Darren joined12:00
amirpro left12:01
amirpro_amirpro12:01
Darren left12:01
cr1mson_king left12:01
jokajak left12:01
amirpro left12:01
cr1mson_king joined12:01
cr1mson_king left12:01
amirpro joined12:02
Darren joined12:02
Darren left12:02
Darren joined12:03
Darren left12:04
Darren joined12:05
prakashdanish left12:05
Darren left12:05
fmcgeough joined12:06
fmcgeough left12:06
Darren joined12:06
Darren left12:07
kurkale6ka left12:07
Darren joined12:08
Darren left12:08
roban joined12:09
Darren joined12:09
Darren left12:10
jokajak joined12:10
lkurusa joined12:10
roban left12:10
lkurusaGuest6927212:10
Darren joined12:11
Darren left12:11
Darren joined12:12
Darren left12:13
Darren joined12:14
Darren left12:14
Darren joined12:16
Darren left12:16
kurkale6ka joined12:16
Darren joined12:17
Darren left12:17
Darren joined12:18
Darren left12:19
Darren joined12:20
kurkale6ka left12:20
Darren left12:20
Darren joined12:21
Darren left12:22
Regon joined12:23
Darren joined12:23
parsnip Darren:12:23
Darren left12:23
timj1 joined12:24
parsnip Darren fix your connection12:24
Darren joined12:24
Darren left12:25
brettgilio joined12:25
Darren joined12:26
cdown left12:26
Darren left12:26
boombatower joined12:26
kurkale6ka joined12:27
timj left12:27
Darren joined12:27
Darren left12:28
DirkjanH left12:28
Darren joined12:29
Darren left12:29
Darren joined12:31
Darren left12:31
Darren joined12:32
Darren left12:32
throstur_tp left12:33
Darren joined12:33
Darren left12:34
wandering_vagran joined12:34
amirpro left12:35
amirpro joined12:35
Darren joined12:35
Darren left12:35
Darren joined12:37
Darren left12:37
Darren joined12:38
Darren left12:38
c0c0 left12:39
Darren joined12:39
Darren left12:40
Darren joined12:41
Darren left12:41
Darren joined12:42
BallChang-CN left12:43
Darren left12:43
ilmostro joined12:43
BallChang-CN joined12:43
Darren joined12:44
Darren left12:44
Darren joined12:45
Darren left12:46
Darren joined12:47
Darren left12:47
rchavik joined12:48
Darren joined12:48
Darren left12:49
jast okay, I'm gonna make this permanent12:49
ChanServ set mode: +o12:49
jast set mode: +b12:50
jast set mode: -o12:50
skisela_afk left12:51
kakashiAL I dont understand the command:12:53
git add -u12:53
jast it automatically stages all modifications/deletions on files that are already tracked (in the current directory and subdirs if you don't give a path argument)12:54
one use of -u is to avoid accidentally adding new files12:54
DelTree sometimes better than -A, indeed...12:55
jast another is to just stage all of your changes in tracked files (and also unstage deleted files, unlike 'git commit -a')12:55
-A goes even further by also staging new files12:55
planetcalls|work joined12:56
applecrumble joined12:56
mkalfon_ joined12:56
planetcalls|work left12:57
kakashiAL jast: hmm, so lets say I have a folder with the files foo.txt and bar.txt12:58
and they allready staged (with git add -A)12:58
zefferno left12:58
kakashiAL if I delete bar.txt and run git -u, what will happen?12:59
jast bar.txt will be unstaged12:59
ableto joined12:59
jast i.e. the next commit won't have it12:59
kakashiAL jast: and if I add baz.txt and run git add -u it will be staged?12:59
osse you mean staged for deletion, right?13:00
planetcalls|work joined13:00
jast kakashiAL: no, because now it's technically a new file, which -u doesn't do anything with13:01
kakashiAL jast: so "git add -u" only means "check if the files are still there, and if some files are missing, remove them from the repo, so on next commit they are not part of the commit" ?13:02
stoopkid joined13:02
Hounddog left13:03
m00sehead joined13:03
jast it also stages changes in files that are already tracked that you modified in the meantime13:03
kakashiAL jast: what does that means? could you give me a simple example?13:03
mkalfon_ left13:03
al-damiri joined13:03
Hounddog joined13:04
planetcalls|work left13:04
jast step 1: foo.txt is tracked, no changes in "git status". step 2: you edit foo.txt, changes now show up as "not staged for commit". step 3: you run "git add -u", changes now show up as "to be committed"13:05
agowa338 joined13:05
avar left13:06
kakashiAL jast: wait!13:07
geekodour08 joined13:08
kakashiAL jast: does this means that you have to run "git add -u" bbefore every "git commit" so "git commit knows what files it has to commit" ?13:08
fmcgeough joined13:09
eset joined13:12
gunnarx joined13:12
gunnarx left13:12
gunnarx joined13:12
geekodour08 left13:12
amirpro left13:14
fmcgeough_ joined13:14
cdown joined13:14
fmcgeough left13:14
KeyJoo left13:14
fmcgeough_fmcgeough13:14
kakashiAL jast: it makes sense for me to run "git add -u" if I delete a file13:14
jast: but I dont understand why you need to other "refresh" behavior of "git add -u"13:15
osse if you remove a tracked file and run git add -u then that file will be staged for removal13:15
parsnip esc:13:15
dendazen joined13:15
parsnip oops13:15
kakashiAL osse: sure, a deleted file + "git add -u" means that this deleted file is not part of the next commit13:16
osse: but why do I have to run "git add -u" if I change a file?13:16
parsnip eset has passwords in his repo, sounds like they can't just amend an earlier commit.13:17
BallChang-CN left13:17
osse kakashiAL: you don't have to13:17
kes left13:17
osse kakashiAL: the whole intermediate step is there so that you can select what to commit or not13:17
kakashiAL osse: okay, but whats about the other behavior of "git add -u" that is described as "updating file changes"13:18
kes joined13:18
osse what about it?13:18
kakashiAL I dont understand it13:18
osse it git adds all changed files13:19
"changed" includes "removed"13:19
kakashiAL osse: what files does it adds that are not removed? and how can I imagine that?13:19
zsocw joined13:20
zsoc left13:20
osse kakashiAL: files that you have edited13:20
zsocwzsoc13:20
zsoc left13:20
zsoc joined13:20
kerframil joined13:20
kakashiAL osse: okay, so lets say I have a file called foo.text with the content Apple, and I add Banana to it13:20
now I would only run "git commit"13:21
kurvivor joined13:21
kakashiAL so why do I need "git add -u", even if I edited the file?13:21
parsnip eset: you could potentially do an interactive rebase.13:21
osse i explained that already13:21
kakashiAL osse: I dont get that :(13:21
osse you need to tell git commit what to commit13:21
in that case you can use git commit -a as a shortcut13:22
kakashiAL osse: you mean "git commit" alone would ignore that file changes, unless I run "git add -u" before or "git commit -a"13:22
correct?13:22
osse yes13:22
git commit alone only cares about what is staged13:23
eigenvalue left13:23
kakashiAL osse: thanks!!!13:23
eigenvalue joined13:23
osse kakashiAL: the reason is: suppose you add Banana to foo.text like you said, but also add Orange to bar.text, and you only want to commit one of the tings13:25
things13:25
without the git add step, you would have to undo one change before making a commit13:25
kakashiAL osse: I see, for example if you want to experiment or if its still in process13:25
brettgilio left13:26
osse yes, or you add some extra printouts for debugging13:26
tobiasvl or you added Banana to foo.text and also Orange to foo.txt, for that matter. I use `git add -p` a lot13:26
osse that too13:26
zefferno joined13:28
kakashiAL osse: thanks, makes sense!13:30
nowhere_man left13:31
kurkale6ka left13:32
throstur_tp joined13:32
kurkale6ka joined13:34
mostlybadfly left13:35
dry left13:35
TheRuckus joined13:36
zefferno left13:38
RuckusBringer left13:39
Silenced left13:40
fission6 left13:41
eset Hi some developer had added a commit and pushed to github enterprise password with plaintext13:42
satoriprints change the password13:42
eset he realise that he done it after few commits later when he changed the password in code and pushed that then he saw that in commit history he sees those changed (he didn't realise that this password was there previously added)13:43
satoriprints: I said that but this is a generic something13:43
plexigras left13:43
eset he asked me to urease those commits13:43
erase13:43
zumba_ad_ joined13:43
zefferno joined13:44
eset satoriprints: but you are right13:45
I will blaim him that security was breached need to change this password13:45
sauvin left13:46
AndresInSpace joined13:46
TomyWork joined13:47
mostlybadfly joined13:48
Hedgework eset: So what's the question? Change the password and move on... or if you're especially thorough... change the password, write a git hook to reject future commits with passwords in likely places for mistakenly committed passwords, and move on.13:51
cdown left13:53
Puffball left13:53
teodorg left13:53
plexigras joined13:53
oxymoron93 left13:54
timj1 left13:54
eset Hedgework: basically no further questions :)13:54
osse yer 'onah13:55
tristanp joined13:55
rchavik left13:56
mud joined13:56
ilmostro left13:58
Masplus joined13:59
ilmostro joined13:59
fluxAeon_ joined14:00
Envil left14:00
kerframil left14:01
Mimicize left14:01
fission6 joined14:02
m1m3-50 joined14:02
kurvivor hello! sourry for stupid question, but what perl modules need to be in system for git svn to work properly?14:03
assuming one adds them one by one by hand14:03
or with cpan14:03
cdown joined14:05
mud left14:05
mrtnt joined14:06
crose joined14:06
m1m3-50 left14:06
RoriconKnight joined14:10
miczac left14:10
libregee1ingkid joined14:10
crose left14:11
fission6 left14:13
tsdh left14:13
SpeakerToMeat_ joined14:16
RzR left14:16
RzR joined14:16
avar joined14:17
avar left14:17
avar joined14:17
Guest69272 left14:18
lkurusa joined14:19
thiago joined14:20
kakashiAL left14:21
f2k joined14:24
greggerz joined14:24
fission6 joined14:26
karstenk left14:28
dsdeiz joined14:30
f2k left14:33
notnull left14:33
TheRuckus left14:36
TheRuckus joined14:37
zumba_addict joined14:37
rsrx joined14:37
zumba_ad_ left14:38
irwiss left14:44
rsrx left14:44
irwiss joined14:44
doev left14:44
houami joined14:45
netj left14:45
netj joined14:45
Envil joined14:47
houami left14:49
eigenvalue left14:50
XenophonF kurvivor: maybe just Term::Readkey?14:50
plus Subversion's Perl bindings14:50
which I don't think are in CPAN but I could be wrong14:53
thiago left14:55
XenophonF (this is all from my reading of the FreeBSD port definitions for devel/git-subversion and devel/p5-subversion)14:56
TheRuckus left14:56
Dougie187 joined14:57
lowbro left14:59
introom left14:59
introom joined14:59
bsaboia left15:01
abba left15:02
kurkale6ka left15:02
T_UNIX joined15:03
kurkale6ka joined15:03
chele left15:05
fluxAeon_ left15:06
fstd left15:07
fstd joined15:07
kneeki joined15:07
applecrumble left15:07
SpeakerToMeat_ left15:09
xirhtogallagothrix15:10
Pugabyte25 joined15:13
Pugabyte25 left15:15
ajbura joined15:18
tjbp left15:21
amirpro joined15:21
gareppa joined15:22
TheRuckus joined15:25
tjbp joined15:25
cur8or left15:26
spacesuitdiver joined15:27
c0c0 joined15:28
Hi-Angel left15:28
fluxAeon_ joined15:28
TheRuckus left15:29
jhaenchen joined15:29
jhaenchen left15:29
jhaenchen joined15:29
Hi-Angel joined15:30
bsanford joined15:31
manuelschneid3r left15:32
teodorg joined15:34
Hi-Angel left15:35
lomex joined15:36
houami joined15:36
jab416171 left15:37
rsrx joined15:38
zefferno left15:39
olux joined15:40
plexigras left15:40
olux left15:40
rsrx left15:40
olux joined15:40
houami left15:40
rafasc joined15:42
plexigras joined15:42
ajbura left15:42
ajbura` joined15:43
ajbura` left15:43
mattcen left15:43
ajbura` joined15:43
ilmostro left15:45
ajbura` left15:46
thiago joined15:46
ajbura` joined15:47
thiago left15:47
TinkerT left15:48
ajbura` left15:48
ajbura` joined15:48
kurkale61 joined15:48
TinkerT joined15:49
kurkale62 joined15:50
kurkale6ka left15:50
ilmostro joined15:50
mattcen joined15:50
TripFandango joined15:52
TripFandango left15:52
kurkale61 left15:53
jab416171 joined15:53
ajbura` left15:54
dfaught joined15:54
manuelschneid3r joined15:55
Sasazuka joined15:56
TheRuckus joined15:56
xfil joined15:57
thiago joined15:57
thebope joined15:57
xfil left15:58
BrianBlaze8 joined16:01
TheRuckus left16:01
tombusby left16:02
tombusby joined16:03
miczac joined16:04
manuelschneid3r left16:05
fryst1 joined16:06
fryst1 left16:06
str left16:07
BrianBlaze8 left16:07
libregee1ingkid left16:07
charolastra left16:08
applecrumble joined16:08
wpcarro left16:09
wpcarro joined16:10
wpcarro left16:10
wpcarro joined16:11
Tobbi_ left16:11
ZetFury joined16:13
veegee joined16:13
MLock joined16:14
MLock left16:16
agowa338 left16:17
opalepatrick joined16:18
jhaenchen left16:18
MLock joined16:18
ubuntuisloved left16:18
jhaenchen joined16:20
jhaenchen left16:20
jhaenchen joined16:20
qpdb joined16:20
fission6 left16:20
geekodour08 joined16:26
throstur_tp left16:27
TheRuckus joined16:29
throstur_tp joined16:30
oojacoboo joined16:30
mikecmpbll left16:31
Poffer joined16:32
m1m3-50 joined16:34
TheRuckus left16:34
kerframil joined16:35
kotepillar_ left16:37
Tobbi joined16:37
profall24 joined16:37
kotepillar_ joined16:37
kotepillar_ left16:37
kurkale62 left16:37
Sasazuka left16:37
Masplus left16:39
wpcarro left16:39
zumba_addict left16:40
wpcarro joined16:41
profall24 left16:42
eggy16 joined16:45
jhaenchen left16:45
m1m3-50 left16:45
quxgyver left16:46
Guest7814 left16:46
gunnarx left16:46
zaratustra7 joined16:47
MLock left16:48
MLock joined16:48
eggy16 left16:49
eNbass27 joined16:49
eNbass27 left16:50
TheRuckus joined16:52
orbyt_ joined16:53
zaratustra7 left16:54
darkbit joined16:54
MLock left16:54
orbyt_ left16:54
fatalhalt joined16:54
gxt left16:54
AtumT joined16:55
orbyt_ joined16:56
HarryS21 joined16:58
HarryS21 left17:00
maryo_ joined17:00
Tobbi left17:00
kegster so i'm trying to go back to a specific commit, and then push it up. i've tried git revert hash and it doesnt seem to do the trick17:02
jast !revert17:03
gitinfo That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?17:03
thiago left17:03
jast or f) make a new commit on top of everything that contains exactly the same file versions as an older commit :)17:03
cdown left17:04
ski7777 left17:05
thiago joined17:05
kegster maybe i'm confused on all the types of reerts haha17:05
maryo_ left17:05
kegster reverts17:05
i don't want to lose my commits, so git checkout -p seems to be the tool17:06
maryo_ joined17:06
amirpro left17:07
ski7777 joined17:08
chaosruler joined17:10
sysvalve left17:10
bsanford left17:10
sauvin joined17:10
amirpro joined17:12
Theking^17 joined17:15
tombusby left17:15
Theking^17 left17:16
tombusby joined17:18
lomex left17:18
str joined17:18
ski7777 left17:22
nuxdie_ left17:22
nuxdie joined17:22
d^sh joined17:23
cylon512_ joined17:23
houami joined17:24
rafasc left17:25
ubuntuisloved joined17:27
muelleme joined17:27
houami left17:29
maryo_ left17:29
ski7777 joined17:29
_ADN_ left17:30
akushner joined17:30
dmone left17:32
cylon512_ left17:32
netj left17:33
netj joined17:33
soee_ left17:34
soee joined17:35
muelleme left17:36
throstur_tp left17:37
mr1holmes joined17:39
karstensrage joined17:40
wandering_vagran left17:40
olbat[m] left17:41
olbat[m] joined17:41
olbat[m] left17:41
olbat[m] joined17:41
ice9 joined17:43
roygbiv joined17:43
ice9 how to completely reset things i've done in the current branch?17:43
kes left17:44
Muzer left17:44
rond_ joined17:46
heftig ice9: git reset --hard will throw away all uncommitted changes17:47
rond_ hi! I want to delete some commits from my branch (they ONLY contain commit message, no files are changed). However, after that there were some meaningful commits that I want to keep. How to delete the unwanted ones?17:47
heftig ice9: git clean -xfd will purge all untracked files and directories17:47
ice9: needless to say both commands are dangerous17:47
mr1holmes left17:48
heftig rond_: git rebase -i <commit-before-first-commit-you-want-to-throw-away>17:48
rond_: then replace the relevant "pick"s with "drop"17:49
dpyro joined17:50
rond_ heftig: so, basically, it means "Show all commits from <commit-before-first-...> to the recent ones. there will be "pick" for all of them - so it would change nothing. having "drop" in some cases will result in deleting them17:50
heftig: am I getting it right?17:51
heftig rond_: are you familiar with how cherry-pick works?17:52
soee_ joined17:52
rond_ heftig: not really17:53
so my understanding was wrong?17:53
heftig rond_: no, it was pretty much right17:53
soee left17:53
fstd left17:54
fstd_ joined17:54
Praxi joined17:54
fstd_fstd17:54
heftig rond_: rebase basically resets the branch to the commit you specified and then tries to reapply ("cherry-pick") the commits that were removed17:54
tvw left17:54
heftig rond_: -i lets you change the behavior interactively17:54
rond_ nice. thank yo17:55
you17:55
heftig in actuality it's a lot more complex (there are at least three backend mechanisms to rebase that get selected by different options, but that's the basics)17:55
T_UNIX left17:56
Tobbi joined17:59
floppydh left17:59
manuelschneid3r joined18:00
rond_ heftig: can I pass commit id instead of counting "Head-N" ?18:01
_ikke_ yes18:02
heftig rond_: you can pass anything understood by rev-parse (see man git-rev-parse, /SPECIFYING REVISIONS)18:02
gitinfo rond_: the git-rev-parse manpage is available at https://gitirc.eu/git-rev-parse.html18:02
darkbit left18:05
manuelschneid3r left18:06
libregee1ingkid joined18:08
Muzer joined18:09
m1m3-50 joined18:09
texinwien_ joined18:12
yyy left18:12
yyy joined18:12
shicks2525 joined18:14
Sasazuka joined18:14
shicks2525 left18:14
soee joined18:15
texinwien left18:15
shicks2525 joined18:15
soee_ left18:15
Mattiaslndstrm joined18:15
thomasross left18:16
GreenJello joined18:16
jhaenchen joined18:16
jhaenchen left18:16
jhaenchen joined18:16
thomasross joined18:16
rond_ heftig: If I use "squash" instead of "pick", then there will be no commit, but the code will remain?18:16
heftig: or, maybe the better solution would be: create new branch, pull from remote master. copy relevant commits from the other branch18:18
heftig rond_: yes, the changes will be added into the current head at that step of the rebase18:18
so pick 1 squash 2 squash 3 will create a commit combining 1, 2 and 318:19
yyy left18:19
rond_ heftig: I think that the thing I would want is to to the latter one: create new branch, pull from remote master. copy relevant commits from the other branch18:20
m1m3-50 left18:20
kurkale62 joined18:20
heftig git provides many ways to get the same result18:20
tmz left18:21
GreenJello left18:21
GreenJello joined18:21
thebope left18:22
haivron joined18:23
vavkamil left18:24
rond_ heftig: could you help me with "create a new branch X, sync it with remotes/origin/HEAD, copy commit C from local branch Y to X. push" ?18:25
kurkale62 left18:25
jnavila joined18:26
thebope joined18:26
haivron left18:27
heftig rond_: git fetch origin; git checkout -b X origin; git cherry-pick C; git push origin --set-upstream X:X18:30
cur8or joined18:31
rond_ thank you!18:32
Sna4x8 joined18:32
houami joined18:35
Guest6180 left18:36
fury joined18:37
netlar left18:37
virmaha joined18:37
nosferatuman joined18:38
virmaha Hello. Need some help. Earlier, I was on say branch x. I then did git checkout tags/mytag -b branch_y. But since it had similar files, I got unable to rmdir, file already exists error. So I manually deleted all the folders (but made sure not to commit them since I dont want branch x to be affected)18:39
after that, i was able to checkout. I added 3 git submodules on branch_y. But when I do git status, I see all the deleted files18:39
how do I fix this mess18:39
houami left18:39
rond_ heftig: "git remote show origin" shows "Fetch URL: MY REPO", not the original one that I forked. I think that after your commands I've reproduced my local master, not the master from original rpeo18:41
repo18:41
heftig rond_: did you clone your own repo again?18:41
none of the commands I gave you alters the remote config18:42
nosferatuman left18:42
rond_ i think so18:42
heftig (--set-upstream only affects the branch "X" config)18:42
rond_: if you clone a repo, you get a new repo that's downstream of the cloned repo18:43
rond_: no matter if that repository is local or remote18:43
zumba_addict joined18:44
roygbiv left18:45
jhutchins joined18:45
jhutchins left18:46
Sna4x8 I have a project that uses a submodule. The submodule is versioned using semver, and each version is tagged (so theres a 1.0.0 tag, a 1.1.0, a 2.0.1, etc.). I want my project to use a specific tag, but I can't seem to get that working right.18:47
I tried putting branch = tags/1.0.3 in .gitmodules, but that doesn't seem to do the trick. Any help is appreciated.18:48
TomyWork left18:49
virmaha Sna4x8: git clone parentRepo ; git submodule update --init --recursive; cd submodule; git checkout tags/1.1.0 ; cd .. '; git add submodule;18:49
cdown joined18:49
virmaha git commit and push18:49
brettgilio joined18:51
rond_ heftig: I don't get one thing. I performed "git fetch upstream; git checkout -b NEW_BRANCH; git merge upstream/master; git push origin NEW_BRANCH" and the github says it's still "17 commits ahead of projectname:master"18:52
git remote -v shows that upstream is set correctly to the original repo, not my forked one18:53
tmz joined18:53
ice9 how to commit with multi line message?18:54
virmaha ice9: git commit. That'll open a text editor and you can put multi line message18:54
rond_ ice9: maybe commit -m "First Line\nSecond line", but I've never tried18:55
tty joined18:55
Sna4x8 virmaha: Thank you. When I do want to bump the version of my submodule, I should just checkout the new tag that I want in the submodule folder?18:55
plexigras left18:55
virmaha yes Sna4x8. That'll work.18:56
Sna4x8: also, what I'd recommend is create a sample branch in your parent - say test. Try out the submodule bump after checking out test18:56
RoriconKnight left18:56
Sna4x8 virmaha: If I do a git submodule update --remote it pulls in the latest version from master. Is there a way to use the .gitmodule file to specify the tag I want to use?18:57
virmaha in another terminal, do git clone parentRepo; git checkout test; git submodule update --init --recursive; cd submodule; git log -n 118:57
RoriconKnight joined18:57
Sna4x8 ^^ I'm using a test branch and a test submodule.18:57
virmaha verify that it has latest tag18:57
and then do the same on your main/master branch18:57
czart left18:57
ice9 how to set vim as editor for commit messages?18:57
dskotnik joined18:57
ilmostro left18:57
pawle joined18:57
virmaha Sna4x8: i am just a beginner..so I am unaware18:57
plexigras joined18:58
virmaha ice9: export EDITOR=/usr/bin/vim or whatever18:58
Sna4x8 ice9: editor = vim under [core] in ~/.gitconfig18:58
virmaha I've my own doubt. How do I make sure that if I am in branch_x, I am not having any stale files of branch_y18:58
basically, I want to make sure my local repo has only those files as branch_x on remote18:59
ilmostro joined18:59
planetcall|work left19:00
sdwrage joined19:00
miczac Hi, I can't push my local branch to origin "because the tip of your current branch is behind its remote counterpart." - I'm looking at gitk and git log 3.10..remotes/origin/3.10 but I can't see why. I just don't want simple push --force ...19:00
nowhere_man joined19:01
dskotnik left19:01
netj left19:01
virmaha miczac: git pull. Looks like your remote had more commits which your local doesn't have19:01
netj joined19:02
mikecmpbll joined19:02
tvw joined19:02
cliluw joined19:02
pawle left19:04
miczac virmaha: yeah, that's what logic suggests. This is why I'm confused about that: I forked upstream on github (getting origin). I recursively cloned origin to local. I'm working locally and fetch/merge updates from _upstream_. Actually I don't touch origin at all, except pushing from local and issue PRs to upstream.19:06
virmaha miczac: sorry no idea then. Some expert might be able to help you out19:07
miczac np, thx virmaha19:08
dsdeiz left19:09
timj joined19:10
kerframil left19:10
plexigras left19:11
plexigras joined19:14
MegafRoaryTiger19:14
rond_ If I want to change the branch in pull_request, can I do something else then closing the old pull_request and opening a new one ?19:17
dex1983 hi19:18
j416 rond_: that's a github question better answered at ##github; but I suspect the answer is no.19:18
rond_ j416: right19:19
pawle joined19:19
demize rond_: Which branch do you want to change?19:19
aFeijo joined19:20
diogenese left19:20
trineroks hey all, I posted this question on reddit but I am wondering if such a monorepo is possible to build https://www.reddit.com/r/git/comments/9beldj/is_it_possible_to_create_a_monorepo_with_this/19:20
aFeijo hi folks. Is it possible to make git track a folder inside an ignored folder?19:20
rond_ demize: Let' say there is branch X and Y. X is under pull_request. Instead, I want to now make pull_request with Y.19:20
aFeijo /web/sites/files is ignored, but I want to track /web/sites/files/images19:21
rond_ I'd be nice to have the discussion intact19:21
GreenJello rond_, you can replace a branch with another in git19:21
rond_ GreenJello: what do you mean?19:21
git reset --hard origin/Y19:21
?19:21
GreenJello rond_, git reset --hard other-branch; git push -f19:21
yeah19:21
j416 rond_: what is "pull_request"? is it the github feature or something else?19:21
rond_ j416: well, it is...19:21
GreenJello and before that you'd git checkout -b branch-backup; and push that19:22
libregee1ingkid left19:22
rond_ push what?19:22
GreenJello rond_, probably best to open a new PR and avoid screwing up history19:22
rond_, you'll be erasing any changes on the branch, so copy the current branch to another branch to preserve it19:22
then you can switch back and hard reset the PR branch19:23
rond_ I have all I need on branch Y19:23
olux left19:23
j416 rond_: it is what?19:23
dopplergange left19:23
libregee1ingkid joined19:24
rond_ there were some old unnecessary commits on branch X which is currently under pull_request19:24
GreenJello PRs are weird, since you often need to force push things19:24
ktanner joined19:25
rond_ i think I'll go with reset --hard and push --force19:25
thank you19:25
GreenJello I haven't had this specific case before, but like if you rebase on master you have to force push that19:25
rond_ (I've already cherry-picked what's necessary :) )19:25
GreenJello I don't understand how it works, but if you do a normal push, all of the changes you rebased show up in the PR19:26
trineroks left19:26
dopplergange joined19:27
plexigras left19:27
trineroks joined19:27
rond_ GreenJello: Worked like a charm!! :)19:27
Darcidride left19:27
_ADN_ joined19:29
Mattiaslndstrm left19:29
dpyro left19:29
dopplergange left19:31
dopplergange joined19:31
miczac Hi, what is git log 3.10..remotes/origin/3.10 different from git log remotes/origin/3.10..3.10 ?19:32
Celelibi left19:34
bubo left19:34
planigan_ left19:35
RoaryTigerMegaf19:35
bubo joined19:35
nowhere_man left19:36
lkurusa left19:37
dege joined19:37
planigan joined19:37
kneeki left19:37
j416 miczac: man gitrevisions19:39
gitinfo miczac: the gitrevisions manpage is available at https://gitirc.eu/gitrevisions.html19:39
lkurusa joined19:39
lkurusaGuest5487019:39
Celelibi joined19:40
crawler joined19:41
Dougie187 left19:42
Envil left19:42
planigan left19:42
Guest54870 left19:43
soee left19:44
lkurusa_ joined19:44
houami joined19:46
aFeijo left19:48
houami left19:50
wpcarro left19:50
ableto left19:51
gareppa left19:52
pfleck joined19:53
heftig rond_: yes, because you have more commits than master19:54
ice9 what is 'upstream'?19:54
rond_ heftig: hmm? (I've already solved the issue with fetching from upstream, cherry-picking commits of interest and pushing them)19:54
ice9: https://help.github.com/articles/configuring-a-remote-for-a-fork/19:55
heftig [20:52:30] <rond_> heftig: [...] github says it's still "17 commits ahead of projectname:master"19:55
brettgilio left19:55
rond_ heftig: I think I understand clearly what happened and what I was doing wrong and why my last solution worked :)19:56
Thank you very much for you help!!!19:56
heftig okay, yw19:56
planigan joined19:58
ice9 i forked a repo, how can i pull to my local branch with the main repo that i forked from?19:58
miczac virmaha: I couldn't find it. I deleted origin, did a new fork and pushed --all from local. Now all is fine.19:58
codenix joined19:58
tristanp_ joined19:59
virmaha miczac: cool19:59
anubani joined19:59
tristanp left19:59
grumble left20:00
foul_owl_ left20:01
Hobby joined20:01
HobbyGuest540920:02
kurvivor left20:04
todevil joined20:05
todevil left20:06
Guest5409 left20:06
pfleck left20:06
Sample joined20:08
Sample left20:08
Sample joined20:08
elichai2 left20:08
Sample left20:09
nolfie joined20:10
rond_ ice9: rephrase your question20:12
m0viefreak joined20:12
fmcgeough left20:12
libregee1ingkid left20:13
ice9 rond_, i forked repo B from repo A, I create a branch X in repo B; Now I want to sync branch X with repo A master20:14
grumble joined20:14
geekodour08 left20:14
yyy joined20:14
lf94 left20:15
rond_ ice9: https://help.github.com/articles/syncing-a-fork/20:15
notnull joined20:17
notnull left20:17
notnull joined20:17
foul_owl_ joined20:18
schleppel left20:18
taotaomona joined20:19
taotaomonadiogenese20:19
wpcarro joined20:19
Poffer left20:25
grumble left20:27
grumble joined20:28
aOssed28 joined20:28
cr1mson_king joined20:28
cr1mson_king left20:30
aOssed28 left20:30
planigan left20:32
planigan joined20:33
dendazen left20:33
eset_ joined20:35
dfaught left20:35
m00sehead left20:35
applecrumble left20:36
ilmostro left20:37
Dougie187 joined20:38
pawle left20:38
applecrumble joined20:39
mlopezviedma left20:41
sourcream joined20:41
mlopezviedma joined20:42
hypercore joined20:42
hypercore do you guys use git even for projects where it's just you working on it?20:42
nedbat hypercore: yes20:43
j416 hypercore: yes20:43
nedbat hypercore: it's useful to be able to go back in time.20:43
hypercore: and you never know when a project will grow beyond just you20:44
Dougie187 The uses are endless.20:44
notnull left20:44
j416 I tend to put most things in a repo. Just in case.20:44
hypercore maybe i don't understand it's utility, at the moment all i do is add the new/modified file, and then commit it. I never do anything else, so i guess that makes me question if i really need it or not20:44
Dougie187 It also makes it easy to share, and test out changes without having to lose older versions.20:45
cr1mson_king joined20:45
Dougie187 Sometimes it's great if you want to add the new modified file, test it, and if it doesn't work, remove the changes.20:45
hypercore (and i create branches too, but often they conflict with one another and then it becomes a hassle because i can't see the full current state of the application)20:45
cr1mson_king left20:45
cr1mson_king joined20:45
Dougie187 hypercore: I don't know why you couldn't see the full current state of the application.20:46
hypercore Dougie187: i definitely see the advantages for teams, just not so much when it's just me20:46
j416 hypercore: spend a couple of days reading up on it. You'll see. It saves a ton of time and allows you to do things that would be very painful otherwise. !book20:46
gitinfo hypercore: There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable20:46
hypercore ok i'll give it another shot20:46
Dougie187 hypercore: It's saved my butt many times just working by myself.20:46
j416 hypercore: if you understand the basic concepts, the commands (save the syntax of them) will be mostly self-explanatory20:47
at least, that's what I'd think20:47
hypercore i think i understand the "basics", maybe it's just a habit i have to form20:47
Dougie187 hypercore: What were your complaints from the last time you used it?20:47
hypercore (it feels a lot quicker just to prototype without using it, rather than having to commit every single change i make. Or maybe i'm just using it wrong)20:48
j416 hypercore: you don't need to make proper commits to start with20:48
make temporary commits as you go, just to keep track of state20:49
mattcen left20:49
Dougie187 If it's literally just you, and you don't want to make your commits useful, you can even use it to just auto-commit.20:49
hypercore: https://stackoverflow.com/questions/420143/making-git-auto-commit20:49
j416 when you have something isolated that would make a good commit, squash into a proper commit20:49
eset_ left20:49
hypercore Dougie187: ok so the last time i used it for a personal app (being worked on only by me), i would change so styles in my header template, e.g. linking feature x or y, then i would checkout feature z and start working on that, but then there was a mismatch between all the features (because none were merged), so i didn't really get a good feel for the current progress of the application20:50
chaosruler left20:50
anubani hypercore: git can save you the nightmare of creating a hundread backup folders of your work20:50
dpyro joined20:50
andres_fandres20:50
mattcen joined20:50
Dougie187 hypercore: I'd suggest in that case that your branches should either be merged, or you shouldn't use branches for those separate projects.20:50
hypercore: Fundamentally, the branching structure is up to you as the project creator.20:51
hypercore: You could only have a single master branch, until you have a *need* for branches.20:51
hypercore anubani: i must admit i do create a few backup folders (for saving state) as i develop applications, but i'm not sure how to reference the different states using git20:51
anubani it also makes it easier for others to track your changes if you ever publish your repo20:51
Dougie187 hypercore: Referencing the different states is easy. We could help you with that. This would just take away the manual upkeep of maintaining the different versions, and keeping track of what's what.20:52
anubani it helps others understand why you certain changes too20:52
scrptktty joined20:52
hypercore anubani: yeah i definitely see the benefits when more than one person is working on the project, no doubt20:52
Dougie187: how would you recommend saving the project at a particular state?20:52
plexigras joined20:52
anubani from your git commits and their body (which had an extended explanation of the commit msg)20:53
Dougie187 hypercore: I'd recommend commiting when you have something that's testable. Either that, or commit as frequent as you can, and squash when it's testable like was suggested earlier.20:53
hypercore: it largely depends on your workstyle.20:53
hypercore anubani: so "git checkout some-hash-here"?20:53
Dougie187 hypercore: If you work on a project until it's complete, or if you context switch a lot.20:53
hypercore Dougie187: what is squashing?20:53
Dougie187 hypercore: It's the process of taking multiple linear commits, and combining them to make a single commit.20:54
hypercore oh ok20:54
Guest44294ec20:54
Rutice joined20:55
anubani hypercore: if others do 'git log' in your project20:55
they can every commit you made20:55
Dougie187 So, you could make checkpoint commits as frequent as you want, and then at some interval (once you're done with the project) you can squash all the checkpoints into a single working commit.20:55
anubani the msg of the commits says what it did20:55
and thus helps peple reading your code20:55
people*20:56
jp left20:56
Nick`24 joined20:56
anubani even if you are working alone, I mean if you ever publish it for others to read and use20:56
Oatmeal left20:57
opalepatrick left20:57
Dougie187 It sometimes helps you by yourself understand what your changes were as well.20:57
hypercore guess i need to force myself to use it until it becomes a habit, objectively speaking i know using git is beneficial, im just lazy20:57
Dougie187 If you find a bug for example.20:57
hypercore: Doesn't sound like your that lazy if you manage backups manually. :P20:57
cr1mson_king left20:58
Nick`24 left20:59
Praxi left20:59
anubani hypercore: 'git show <hash>' is also useful21:00
jnavila left21:00
mattcen left21:03
surfist left21:03
surfist joined21:04
m1m3-50 joined21:04
dsdeiz joined21:05
dsdeiz left21:05
dsdeiz joined21:05
kurkale62 joined21:05
kerframil joined21:07
dsdeiz left21:09
amirpro left21:11
amirpro joined21:12
Quetzal2 left21:12
kiltzman left21:13
libregee1ingkid joined21:17
Oatmeal joined21:18
m1m3-50 left21:18
KindOne27 joined21:20
mattcen joined21:20
rafasc joined21:20
kurkale62 left21:22
KindOne27 left21:22
Mattiaslndstrm joined21:23
greatgatsby joined21:24
MACscr|lappy joined21:25
tureba left21:25
Oatmeal left21:25
wreed joined21:26
sdwrage left21:26
tureba joined21:27
angular_mike_2 left21:28
angular_mike_2 joined21:28
mattcen left21:29
akushner_ joined21:29
nowhere_man joined21:31
kurkale62 joined21:32
akushner left21:32
jp joined21:33
ubuntuisloved left21:34
MACscr|l_ joined21:34
leeN left21:34
alcohol left21:36
alcohol joined21:37
MACscr|lappy left21:37
eset left21:38
tvw left21:38
Tobbi left21:38
m00sehead joined21:38
rond_ left21:38
ubuntuisloved joined21:39
nowhere_man left21:40
AndresInSpace left21:41
spacesuitdiver left21:42
Tobbi joined21:43
str left21:44
MACscr|l_ left21:45
MACscr|lappy joined21:46
dc13ff joined21:47
houami joined21:47
MACscr|l_ joined21:48
fluxAeon_ left21:50
marlinc left21:50
mattcen joined21:51
MACscr|lappy left21:51
houami left21:51
kurkale62 left21:52
kurkale62 joined21:52
Sophira6 joined21:53
Puffball joined21:54
TheRuckus left21:56
Sophira6 left21:57
eset joined21:58
eset left21:58
eset joined21:58
tmm88 left21:58
marlinc joined21:59
jnavila joined22:04
jnavila left22:04
dsdeiz joined22:05
dsdeiz left22:05
dsdeiz joined22:05
CodeSlingerPaul left22:05
Tobbi left22:05
m00sehead left22:07
brettgilio joined22:08
dsdeiz left22:09
JJonah_ left22:09
thebope left22:09
JJonah joined22:09
m00sehead joined22:09
eset left22:10
MACscr|lappy joined22:10
fstd_ joined22:10
MACscr|l_ left22:11
eset joined22:12
eset left22:12
eset joined22:12
yaymuffins24 joined22:14
zulutango left22:14
Tobbi joined22:14
fstd left22:15
fstd_fstd22:15
Smithe joined22:15
pmn joined22:16
Smithe If I already have raw data of a project but not the git data (commits and such) how can I retrieve it?22:16
_ikke_ Smithe: You have to still clone the project22:18
str joined22:18
philipp64|laptop joined22:18
Dougie187 left22:18
pmn left22:18
philipp64|laptop Hi… question about filter-branch and this example… http://gbayer.com/development/moving-files-from-one-git-repository-to-another-preserving-history/22:18
yaymuffins24 left22:19
kfoonamalik joined22:19
philipp64|laptop I tried this, but after doing the filter-branch and running "git branch -a" I'm only seeing the branch I was on...22:19
how do I do this for all branches?22:20
mattcen left22:20
MACscr|lappy left22:20
_ikke_ what command did you run?22:20
philipp64|laptop I thought the -- --all handled that but it doesn't seem to.22:21
applecrumble left22:21
_ikke_ it should22:22
philipp64|laptop ah… the "git remote rm origin" blows away all the other branches but the current one…22:24
Rostfrei joined22:24
ferdna joined22:24
roro247 left22:25
philipp64|laptop is it safe to omit that?22:25
mattcen joined22:27
agris how would one go about hosting git repos and collaborating with them over a distibuted peer-to-peer transport protocol like IPFS?22:28
amirpro left22:28
rafasc left22:28
rafasc joined22:29
dzho https://github.com/clehner/git-ssb22:29
agris hmm. not exactly what i had in mind. Is ScuttleButt better suited for this than IPFS?22:31
kus joined22:31
ice9 left22:33
profmac left22:33
profmac joined22:34
z|bandito joined22:34
dviola joined22:36
tmsmith joined22:37
jasmith joined22:39
amirpro_ joined22:39
crawler left22:39
nivag left22:42
kurkale62 left22:43
cur8or left22:45
z|bandito left22:46
justanotheruser left22:46
z|bandito joined22:46
z|bandito left22:47
justanotheruser joined22:47
nivag joined22:47
ubuntuisloved left22:48
brettgilio left22:49
z|bandito joined22:49
z|bandito left22:49
z|bandito joined22:49
str left22:51
z1haze joined22:52
z1haze is it possible to delete a remote commit that is not the last commit?22:53
_ikke_ fix your local repo, then force push22:53
!rewrite22:53
gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum22:53
olik joined22:54
mikecmpbll left22:54
PioneerAxon left22:56
PioneerAxon joined22:56
dendazen joined23:04
finalbeta left23:04
goweol joined23:08
PLLunlock joined23:09
PioneerAxon left23:11
dege left23:11
goweol left23:12
goweol joined23:13
TheRuckus joined23:14
mud joined23:14
TheRuckus left23:18
bucketm0use left23:19
mattcen left23:24
vTox left23:25
mattcen joined23:26
thiago left23:32
greggerz left23:33
AtumT left23:34
UTAN_dev_ joined23:35
akushner_ left23:36
UTAN_dev left23:36
darkbit joined23:36
avar left23:36
akushner joined23:37
m0viefreak left23:37
Bucciarati left23:37
houami joined23:38
avar joined23:38
avar left23:38
avar joined23:38
Bucciarati joined23:38
teej joined23:38
plexigras left23:39
oojacoboo left23:41
houami left23:42
plexigras joined23:42
Bobguy joined23:42
figurehe4d joined23:45
figurehe4d left23:45
Vortex-Sec joined23:45
dtux joined23:46
dtux I have a clone on GitHub that was made with `git push --mirror`, and I want to convert that to a plain old repo. Do I need to do anything special?23:48
Forkk left23:48
Forkk joined23:49
mud dtux: Only thing --mirror does is push extra refs (more than it usually would). So if you don't care that those exist, no.23:50
dtux mud: ok, cool. so it would be a reasonable thing to create a cross-server "fork" by doing the initial upload with --mirror and then never doing that again?23:51
akushner left23:51
SpeakerToMeat left23:52
mud dtux: I suppose it wouldn't be that bad, the only thing is I think it also pushes refs/remotes/*, which could be a tad odd to have at the destination. Or I might be misremembering that23:53
SpeakerToMeat joined23:53
dtux mud: i thought so too, but if i browse the branches on the mirror, it doesn't link to public github (which i suspect it would if the remote refs weren't adjusted somehow?)23:54
e_l1997 left23:55
mud Hm, well if it doesn't list them then it's probably fine. Even if it did it probably wouldn't really hurt anything23:55
plexigras left23:55
bucketm0use joined23:55
z8z joined23:56
Rutice left23:57
dtux ok, sounds kosher... we're not trying to sever from the upstream necessarily, it's just abandoned basically23:58
z1haze left23:59

Logs Search ←Prev date Next date→ Channels Documentation