IRCloggy #git 2024-06-14

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.

2024-06-14

→ Quack joined00:00
← jenia2 left00:01
→ hritik joined00:06
→ chandash joined00:10
← henry40408 left00:11
← hritik left00:11
→ henry40408 joined00:11
→ cpackham joined00:25
← iomari891 left00:33
→ iomari891 joined00:35
→ hritik joined00:39
← pedahzur left00:48
← thiago left00:56
← foul_owl left01:02
← jenia2` left01:03
← mns left01:04
← hritik left01:05
→ mns joined01:05
→ gsi_ joined01:15
→ foul_owl joined01:16
← chandash left01:18
← gsi left01:18
← Arokh left01:19
→ hritik joined01:25
→ tirnanog joined01:26
← misanthropos left01:27
→ Arokh joined01:27
← mgedmin left01:30
← hritik left01:30
← TomyLobo left01:32
→ mgedmin joined01:32
→ cdown_ joined01:34
→ oopzzozzo joined01:37
← cdown left01:37
cdown_ → cdown01:37
sam113101 if my branch name has multiple words in it, should I separate them with hypens (-) or underscores (_)?01:40
→ hritik joined01:43
→ thiago joined01:46
← hritik left01:48
← fflam left01:49
→ fflam joined01:52
← jab416171 left01:52
→ jab416171 joined01:58
→ jenia2` joined01:59
→ hritik joined02:01
← jab416171 left02:03
← henry40408 left02:04
← rigid left02:04
→ henry40408 joined02:04
→ rigid joined02:05
← hritik left02:06
→ jab416171 joined02:09
← jenia2` left02:11
← jab416171 left02:14
← navi left02:16
→ jab416171 joined02:19
← krushia left02:20
→ misanthropos joined02:21
← leftyfb left02:22
→ leftyfb joined02:22
→ printfdebugging joined02:22
← jab416171 left02:22
→ jab416171 joined02:23
← iomari891 left02:26
→ iomari891 joined02:28
← oopzzozzo left02:28
← PantelicGR left02:29
→ wys joined02:30
← iomari891 left02:34
← jab416171 left02:35
→ iomari891 joined02:35
→ Vonter joined02:36
→ jab416171 joined02:40
→ haxalot joined02:41
→ hritik joined02:44
← jab416171 left02:45
← hritik left02:50
→ jab416171 joined02:52
→ br__ joined02:56
← jab416171 left02:58
← iomari891 left03:02
→ hritik joined03:03
→ jab416171 joined03:05
→ Gary_Sinise joined03:08
→ iomari891 joined03:08
← hritik left03:09
→ xdminsy joined03:10
→ printfdebugging1 joined03:10
→ toric_ joined03:10
← toric left03:11
← xdminsy left03:11
← printfdebugging left03:12
printfdebugging1 → printfdebugging03:12
→ xdminsy joined03:22
→ toric joined03:23
← toric_ left03:24
← gast0n left03:26
→ hritik joined03:27
← printfdebugging left03:27
← firewyre left03:28
← ChmEarl left03:29
← iomari891 left03:30
→ iomari891 joined03:31
← hritik left03:32
→ firewyre joined03:32
→ hbautista_ joined03:33
→ hritik joined03:35
← Betal left03:37
← iomari891 left03:37
→ iomari891 joined03:39
← hritik left03:40
← jab416171 left03:45
← dlh404 left03:50
← bloody_ left03:52
→ jab416171 joined03:52
→ hritik joined03:52
→ bloody joined03:55
← bloody left03:55
← hritik left03:57
→ hritik joined04:09
← hritik left04:15
← grogu left04:17
→ grogu joined04:17
← iomari891 left04:18
← jab416171 left04:20
→ jab416171 joined04:21
→ hritik joined04:22
← Gary_Sinise left04:25
← jab416171 left04:26
→ jab416171 joined04:27
← SuperLag left04:30
→ SuperLag joined04:30
← SuperLag left04:30
→ SuperLag joined04:30
← jab416171 left04:31
← remedan left04:33
← hritik left04:33
→ fdan joined04:34
→ jab416171 joined04:35
← grogu left04:37
← br__ left04:41
→ hritik joined04:47
→ remedan joined04:49
← hritik left04:51
← SuperLag left04:55
→ SuperLag joined04:55
← kurahaupo left04:57
← jab416171 left05:01
← cpackham left05:01
→ jmd joined05:03
← jmd left05:03
→ hritik joined05:03
→ jab416171 joined05:07
← hritik left05:08
→ jmd joined05:12
→ qqq joined05:13
← lemuria left05:18
→ hritik joined05:22
→ becm joined05:24
← hritik left05:27
← jmd left05:35
← FH_thecat left05:41
← misanthropos left05:42
→ hritik joined05:47
→ jmd joined05:49
← jmd left05:52
← hritik left05:53
← thiago left05:56
← wys left06:03
→ hritik joined06:05
→ FH_thecat joined06:09
→ duxsco joined06:10
← hritik left06:11
← a6im left06:14
← jab416171 left06:14
← hbautista_ left06:15
→ jab416171 joined06:21
← Crispy left06:23
→ hritik joined06:24
← fdan left06:24
→ odinsbane joined06:29
→ misanthropos joined06:30
← haxalot left06:34
← hritik left06:36
→ Crispy joined06:43
← dsrt^ left06:44
→ hritik joined06:49
← jab416171 left06:49
→ peirik__ joined06:49
← krumelmonster left06:53
← hritik left06:54
→ jab416171 joined06:54
→ rfuentess joined07:00
j416 sam113101: I prefer hyphens, but I don't think there is a convention per se.07:02
→ rsx joined07:03
→ danse-nr3 joined07:03
→ iamtheranger joined07:03
← danse-nr3 left07:04
→ danse-nr3 joined07:04
← jab416171 left07:04
→ krumelmonster joined07:05
→ rosco joined07:07
→ hritik joined07:08
→ textyash joined07:09
← iamtheranger left07:09
textyash git switch doesn't prompt me to stash my changes for both tracked and untracked files while switching branches. Why is that? When does switching prompt me to stash my changes?07:10
→ jab416171 joined07:10
← qqq left07:12
← hritik left07:13
osse textyash: if any of the files you've changed are different in the branch you switch to07:15
→ gandhibobandhi_ joined07:16
← danse-nr3 left07:26
textyash osse Yes I tried that. The same files exist on the both the branches but it's still not prompting me.07:27
osse textyash: but are they different?07:27
→ hritik joined07:30
textyash osse the changes are different in both the branches to the same file.07:32
osse textyash: !repro07:33
gitinfo textyash: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see07:33
→ gloomy_desktop joined07:35
→ danse-nr3 joined07:36
osse textyash: When you switch branches if Git doesn't have to ovewrite a file when doing so then it won't complain. That's the gist of it.07:36
← hritik left07:39
← textyash left07:43
← CovertDuck left07:44
→ textyash joined07:48
textyash osse For example, I created a test branch with and commited a test.txt file with one line. Then created another branch from test and overwrote that line, didn't stage the changes & switched to test branch. It just switched.07:49
osse textyash: try committig the change in the other branch, make another change and try to switch this time07:50
← troyt left07:52
→ troyt joined07:56
← skapate left07:57
→ nivag joined07:57
→ nivag_2nd joined07:57
osse textyash: remember that when you make a new branch it will initially be equal to the branch you created it from. and untracked changes don't "belong" to a branch07:59
→ Left_Turn joined08:00
← rosco left08:01
→ rosco joined08:01
← defec7ed left08:02
← nivag left08:03
← nivag_2nd left08:03
→ defec7ed joined08:03
→ nivag joined08:03
← nivag left08:03
→ nivag joined08:05
→ jmd joined08:06
→ Ocupado joined08:07
osse textyash: remember that when you make a new branch it will initially be equal to the branch you created it from. and untracked changes don't "belong" to a branch08:09
oopd08:09
← jmd left08:14
← stagas left08:19
→ someone235 joined08:19
← splud left08:21
→ wys joined08:23
→ bluesmonk joined08:23
← wys left08:23
← danse-nr3 left08:28
→ ferdna joined08:29
← textyash left08:31
← CrunchyFlakes left08:32
→ CrunchyFlakes joined08:36
→ danse-nr3 joined08:37
→ stagas joined08:40
← CrunchyFlakes left08:45
→ CrunchyFlakes joined08:50
→ splud joined08:50
→ hritik joined09:02
← hritik left09:07
→ hritik joined09:07
→ konsolebox joined09:17
← hritik left09:19
→ hritik joined09:20
← memset left09:22
→ memset joined09:22
→ MGF_Fabio joined09:23
← duxsco left09:23
← theobjectivedad left09:24
→ theobjectivedad joined09:25
→ duxsco joined09:25
← hritik left09:25
→ hritik joined09:26
← hritik left09:27
→ hritik joined09:28
→ KnoP joined09:30
← Crispy left09:35
← hritik left09:39
→ hritik joined09:40
← danse-nr3 left09:52
→ danse-nr3 joined09:53
→ printfdebugging joined09:57
← printfdebugging left09:57
← hritik left10:01
→ hritik joined10:06
← hyperair left10:09
→ davispuh joined10:11
→ textyash joined10:18
textyash osse now i can recreate it sometimes. So basically you're saying that it only happens for conflicting changes between the branches? How can I read on this? Like what makes a conflicting change to be aware of it every time?10:19
← hritik left10:20
→ hritik joined10:21
← SJrX left10:22
→ SJrX joined10:25
→ Turn_Left joined10:25
← jab416171 left10:27
← Left_Turn left10:30
← hritik left10:30
→ jab416171 joined10:33
← memset left10:34
→ memset joined10:35
← misanthropos left10:36
→ YuGiOhJCJ joined10:37
← Lights left10:42
→ hritik joined10:44
→ misanthropos joined10:49
← hritik left10:54
→ Betal joined10:58
← peirik__ left10:59
← textyash left10:59
→ AmR_ joined11:10
← AmR left11:13
→ peirik__ joined11:17
→ jenia2` joined11:17
← peirik__ left11:17
→ peirik__ joined11:18
→ peirik___ joined11:18
→ hritik joined11:20
← peirik_ left11:20
← peirik__ left11:22
← bluesmonk left11:26
← hritik left11:26
→ bluesmonk joined11:28
← bluesmonk left11:28
→ bluesmonk joined11:28
← danse-nr3 left11:31
← nightstrike left11:33
→ delay joined11:35
← someone235 left11:38
← dsimic left11:45
→ dsimic joined11:47
← rustyshackleford left11:47
← ferdna left11:48
← memset left11:50
→ memset joined11:50
← rosco left11:51
→ dlh404 joined11:53
← jab416171 left11:55
← dermoth left11:56
→ jab416171 joined12:00
→ dermoth joined12:01
← delay left12:03
→ jenia2`` joined12:03
→ delay joined12:04
← jenia2` left12:05
← memset left12:10
← delay left12:11
← jab416171 left12:12
→ delay joined12:17
→ jab416171 joined12:18
← duxsco left12:19
← bluesmonk left12:19
← delay left12:21
→ halloy2089 joined12:21
→ duxsco joined12:21
→ bluesmonk joined12:21
→ Ram-Z joined12:21
halloy2089 → Alexey-Pelykh12:21
→ memset joined12:22
→ navi joined12:23
→ Lights joined12:23
← jab416171 left12:23
→ tlatelolco3 joined12:25
tlatelolco3 Hello. I wanted to make my branch point one commit earlier. I issued the command `git update-ref dev 5212c` which didn’t did the trick. Instead, when I checkout around, I get `warning: refname 'dev' is ambiguous.`. How to revert my mess?12:26
← becm left12:27
→ becm1 joined12:27
osse tlatelolco3: git update-rev -d dev first off. then: Is your current branch named dev and you want to hop back and pretend the currnt commit didn't exist?12:28
→ jab416171 joined12:28
← odinsbane left12:29
← mrmoneyc left12:29
tlatelolco3 `git update-ref -d dev` : `error: refusing to update ref with bad name 'dev'`.12:29
becm1 → becm12:29
→ mrmoneyc joined12:29
tlatelolco3 The last commit is part of an other branch. But I want dev to point one commit earlier.12:30
→ jmd joined12:31
osse tlatelolco3: can you run git for-each-ref | grep -w dev12:32
tlatelolco3 `9031065fe0679794cc610ec489fda26045a2af56 commit refs/heads/dev` is the only answer12:33
← Novi left12:33
osse good12:33
tlatelolco3 it’s the actual "too far" commit12:33
osse then perhasp the first update-ref failed also, so there is nothing for my command to delete12:33
the answer is git reset (--hard) 5212c12:33
--hard if you also want to revert the files on disk12:34
→ krushia joined12:35
→ Novi joined12:35
tlatelolco3 Ok. It’s reset to an early commit. But when I switch branches I still have this ambiguous warning12:39
And then I changed a tracked file on disk but `git status` doesn’t see it. Even if I `git add` it manually again. That’s really weird12:40
Alexey-Pelykh Hi! Why having a folder named ".gitattributes" causes and error and just not ignored for the purpose of processing the .gitattributes? It's possible to have a folder named .gitignore at the same time12:41
← din0 left12:43
→ din0 joined12:43
osse tlatelolco3: you also have a file or directory named dev ?12:43
tlatelolco3 No I don’t12:44
← jmjl left12:44
← kanzure left12:45
→ kanzure joined12:45
← mrmoneyc left12:46
selckin /2112:46
osse tlatelolco3: does .git/dev or .git/refs/dev exist ?12:47
tlatelolco3 .git/dev yes. But it’s not the good commit number dev is pointing to now12:48
→ trace1918 joined12:48
osse remove that file12:49
that causes the warning12:50
tlatelolco3 Thank you. Indeed.12:50
But I still can’t commit from the right-positionned dev branch12:51
osse !repro12:51
gitinfo [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see12:51
tlatelolco3 My bad. I had got confused. It’s alright :)12:55
osse it'12:55
it's weird that update-ref lets you create a ref that you cannot remove12:55
→ mrmoneyc joined12:58
← becm left13:00
← skered left13:03
→ duxsco_ joined13:05
→ otisolsen70 joined13:05
→ becm joined13:06
← duxsco left13:08
← duxsco_ left13:13
← tlatelolco3 left13:15
← dlh404 left13:21
← peirik___ left13:21
→ peirik_ joined13:22
→ dlh404 joined13:23
→ peirik__ joined13:24
→ hyperair joined13:25
← jmd left13:27
← trace1918 left13:27
→ dmj` joined13:31
← peirik__ left13:33
→ delay joined13:46
→ skered joined13:49
← celphi left13:50
← becm left13:53
→ celphi joined13:53
→ becm1 joined13:53
← henry40408 left13:54
→ henry40408 joined13:54
← delay left13:55
becm1 → becm13:55
→ thiago joined13:57
← AmR_ left14:04
← thiago left14:15
← fflam left14:19
→ fflam joined14:21
→ stef204 joined14:25
→ delay joined14:26
← fflam left14:26
→ iomari891 joined14:27
→ fflam joined14:28
→ jensen joined14:30
→ peirik__ joined14:36
← peirik__ left14:37
→ lucasta joined14:38
→ peirik__ joined14:38
← jab416171 left14:39
← peirik_ left14:40
→ roli joined14:45
roli hi14:45
← delay left14:45
→ jab416171 joined14:45
← jab416171 left14:45
→ jab416171 joined14:45
← roli left14:46
→ john_johnk joined14:52
→ roli joined14:52
→ thiago joined14:53
← roli left14:53
← john_johnk left14:54
→ roli joined14:55
← roli left14:55
jensen → aYS114:57
← thiago left14:59
→ ChmEarl joined15:01
← Vonter left15:04
→ Vonter joined15:04
→ roli joined15:05
← roli left15:05
→ roli joined15:06
← skered left15:06
→ thiago joined15:08
← roli left15:08
→ B-| joined15:09
B-| how can a github commit patch hash change15:10
https://github.com/llvm/llvm-project/commit/c2aabcfc839515:10
https://github.com/llvm/llvm-project/commit/c2aabcfc8395.patch15:10
https://aur.archlinux.org/cgit/aur.git/commit/?h=clang15&id=c36e10c91559289b3264ab806a5d3ae6e5035b5915:10
and it's still wrong somehow15:10
make this make sense15:10
→ OverlordDuck joined15:12
← OverlordDuck left15:12
selckin first you have to make sense15:12
→ skered joined15:12
B-| https://github.com/llvm/llvm-project/commit/c2aabcfc8395.patch15:13
this changed the checksum somehow15:13
selckin why and compared to what15:13
B-| https://aur.archlinux.org/cgit/aur.git/commit/?h=clang15&id=c36e10c91559289b3264ab806a5d3ae6e5035b5915:13
link didn't change, checksum changed15:13
isn't c2aabcfc8395 the commit hash15:14
how can the patch file change but the commit hash part not15:14
selckin nothing you're saying makes any sense15:14
B-| what do you mean...................15:14
selckin how are those links proving anything15:14
→ delay joined15:14
B-| aur package maintainer had to update the checksum for the file at that link15:15
i'm asking how it could have possibly changed15:15
selckin ah so the arch link is a checksum from the other link?15:15
B-| yes15:15
checksum being updated15:15
← xdminsy left15:18
← delay left15:19
B-| selckin: i'll assume you were killed on the spot for noticing this15:22
selckin the current url doesn't even hash to either of those, ask the maintainer15:23
B-| 🙄15:23
← fflam left15:27
← bluesmonk left15:29
→ fflam joined15:29
← mkosmo267 left15:30
→ bluesmonk joined15:30
→ BurningHuman joined15:34
→ stuv joined15:36
BurningHuman Hello everyone! Has anybody set up himself a plain git server at some time?15:37
selckin !gitolite15:37
gitinfo Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/ Quick example: http://gitolite.com/gitolite/overview/#basic-use-case15:37
BurningHuman Thanks for the link, but sadly it's dead15:38
→ epicout joined15:38
← epicout left15:38
→ epicout joined15:38
BurningHuman I actually don't want a server like that, I just literally want to have my laptop, the git user, and manually put in all the ssh key pairs. That's it.15:39
→ jmd joined15:39
BurningHuman It's for demonstration purposes and people should be able to just connect over ssh to the computer, copy in the public key and then be able to clone, pull push, from a bare repository lying around under the git user.15:39
← rfuentess left15:40
selckin that just works (tm), don't need anything special15:41
BurningHuman What I get wrong, or have problems with is the file structure in comparison to GitLab. There is the "git" user, from which we can issue push,pull, clone requests and unter it, as far as I understand it, there are the account names and their repositories lying in it. Or am I wrong?15:41
osse BurningHuman: If you want to do it manually you can just do it. There's very little magic to it. A dedicated user with a bunch of bare repos in its home dir15:41
Everyone uses the same OS user on the server. The actual user is determined by which key is used15:42
BurningHuman Yes, but OS is not the concern here :). Currently my setup is like that, that the public keys are under /home/git/.ssh/authorized_keys. Anyone with the correct keypair has then ssh access to the computer over the "git" user.15:44
osse Perfect15:44
→ Mooncairn joined15:44
BurningHuman But since, there are other multiple computers which should get "accounts", just like in GitLab, I guess, their home directory is then under /home/git/USER/?15:44
Because repositories in GitLab always start with: [email@hidden.address]15:45
Or do I get something wrong?15:45
selckin thats just a convention gitlab enforces, you can name it anything you want15:45
BurningHuman Wait.. so I actually don't need all those extra accounts? I can just work with the one "git" user?15:46
selckin from what you said probably15:46
osse BurningHuman: Not sure it literally works like this, but you can at least model it as just a directiory hierarchy under the git user's HOME15:46
selckin unless you want to restrict access between users15:46
BurningHuman @selckin, I mean would be great to have more control over that, yes. But if I get multiple computers working with the very simple configuration, this should already be some kind of success :)15:48
selckin the multiple computers is very confusing what that means15:49
← robobub left15:49
BurningHuman Okey, so I have one Laptop. Where the sshd is running with the git user to be used.15:49
Other Computers, so also laptops, should be able to connect to this one laptop over ssh and then use the git repositories15:50
selckin thats how ssh works yes15:50
B-| BurningHuman: https://wiki.archlinux.org/title/Git15:50
everything's ez on arch15:51
BurningHuman I am on Arch ;)15:51
← MGF_Fabio left15:51
B-| epico15:52
BurningHuman I just was very confused with the pathing in comparison to GitLab, I tried to recreate the same behaviour15:52
But, it seems like it is not really necessary15:52
B-| BurningHuman: can you even reach your puter from the internet15:52
BurningHuman Since I have a dynamic IP, not really15:52
B-| then you need to fix that somehow15:53
also i think you literally just need an http server to make your git repo available15:53
BurningHuman 1. I don't want to, 2. I would have to pay my ISP more for that15:53
you literally have to buy that thing15:53
B-| BurningHuman: you could proxy15:53
BurningHuman hm15:53
B-| ssh would make that twork15:54
→ AmR joined15:54
B-| > also i think you literally just need an http server to make your git repo available15:54
no ssh or anything15:54
unless you want people to write to it15:54
BurningHuman I actually want ssh in this case15:54
This too15:54
B-| i don't see why this wouldn't be well documented on the arch wiki15:55
BurningHuman No no, it's fine there is no problem in that15:55
The general goal here is, to have at least one repository on the server laptop, from where people can clonse/push/pull to/from. But this seems like an easy thing now since I don't need those path shenanigans :D15:57
B-| git@your-domain-or-maybe-ip-idk:repository.git15:59
i imagine a git user on the system15:59
and a repository.git bare git repo in ~/15:59
you would need port 22 accessible from the network16:00
i assume you want it accessible from the internet16:00
BurningHuman Yes everything right, except the internet16:01
local network16:02
let me just test it first with this very simple approach, so I kind of get it all right :)16:02
B-| port forward, buy static ip, get a proxy server16:02
etc etc etc16:02
pick one solution16:02
ipv6 is not an option i'll tell you rn16:02
BurningHuman Yes, I know, but I can just do it in my local network by typing in the 192.168.1.2 IP, that's enough16:03
B-| ok so you don't need access from the internet16:03
BurningHuman Yes, that's what I meant :)16:03
B-| oh my zoodness16:03
your device still needs the ip to not change16:04
BurningHuman Yes16:04
B-| you may get lucky or you may need to change router config16:04
BurningHuman This laptop wont reboot for 3h, that's enough16:04
← iomari891 left16:04
B-| also i don't see why [email@hidden.address] wouldn't work16:04
BurningHuman I just tell the people my current local IP Address and that's fine16:04
yea yea16:04
← epicout left16:05
B-| also i think git comes with a server already16:05
BurningHuman But there is one last thing, I cannot create repository on my computer and the kind of push it as a bare repository to the server?16:05
yes it does16:05
B-| pgit /usr/lib/systemd/system/git-daemon.socket16:05
git /usr/lib/systemd/system/git-daemon@.service16:05
BurningHuman I am actually not running it. Hm16:06
B-| BurningHuman: why not first try doing it locally, you can push to a local bare repo16:06
BurningHuman But the git-daemon uses the git protocol16:06
not ssh16:06
B-| and then scp -p it over to the server16:06
BurningHuman Yes, but I would have to create the bare repository first on the server, right?16:07
B-| i said you can do it locally16:07
and then copy each file verbatim over to the server with scp -p16:08
scp -rp16:08
then update the url on your normal repo to point to the server one16:08
and delete local bare repo16:08
i'm telling you this so you can test16:08
BurningHuman You mean, create the bare repository on the one laptop, scp it to the server, and then clone again basically?16:08
B-| i don't care nor do i know how you can do it another way16:09
~/repos/x-server-bare.git ~/repos/x16:09
on your normal laptop16:09
BurningHuman The thing is, the user @git uses the git-shell, and not bash, so it's capabilites in accessing programs is limitted.. currently I have changed it to bash16:10
B-| i'm telling you that you can first test locally16:10
BurningHuman ooh okey16:10
B-| BurningHuman: just scp -pr with root16:10
and then update with chown git:git16:10
← gloomy_desktop left16:10
B-| chown -R git:git ~git16:10
or whatever16:11
← jab416171 left16:11
B-| i'd first look at what perms ~git has with ls -lhd ~git16:11
just in case it's something quirky16:11
→ epicout joined16:13
← epicout left16:13
→ epicout joined16:13
BurningHuman That's what I did before, it just requires to be active on the server.16:13
but wait.. I have an Idea16:13
→ jmjl joined16:13
← OnkelTem left16:14
→ haxalot joined16:16
← tirnanog left16:17
→ mud joined16:19
→ jab416171 joined16:20
← jab416171 left16:20
→ jab416171 joined16:20
→ pedahzur joined16:20
osse BurningHuman: https://freedns.afraid.org/ been using this for a long time works pretty well16:23
→ hdh joined16:24
osse Free subdomains16:24
← jab416171 left16:25
← pebble left16:28
→ pebble joined16:29
→ tirnanog joined16:29
BurningHuman Oh this is very cool! Thx!16:29
j416 or, you know, cloudflare.16:30
→ PocketLagger joined16:30
→ jab416171 joined16:31
← PocketKiller left16:31
→ cdown_ joined16:33
→ AmR_ joined16:35
← cdown left16:36
cdown_ → cdown16:36
← Alexey-Pelykh left16:38
← AmR left16:39
← tirnanog left16:42
osse how so?16:42
← specing left16:42
→ donofrio joined16:44
← gandhibobandhi_ left16:47
BurningHuman Okey! Good news! So, I just came to the conclusion I don't need the user git@, I just create as many users as I want, let the users ssh into the server and let them there create a bare git repository. When they are back on their laptops they can just clone from it directly16:47
← YuGiOhJCJ left16:49
← pebble left16:50
← peirik__ left16:51
→ peirik_ joined16:51
← DragonMaus left16:54
→ DragonMaus joined16:55
→ ferdna joined16:55
← otisolsen70 left17:00
→ skapata joined17:01
← thiago left17:01
← rsx left17:02
osse BurningHuman: That will work, but it can become burdensome to deal with FS permissions if the users want to push to another user's repo17:03
→ delay joined17:04
← shored left17:05
BurningHuman Yes, this is the next thing I will try17:06
I just want to see how far I can go with this approach17:06
← ocra8 left17:09
→ hbautista_ joined17:09
→ shored joined17:09
→ ocra8 joined17:11
← Turn_Left left17:12
← jab416171 left17:15
→ jab416171 joined17:17
← jab416171 left17:22
→ peirik__ joined17:22
← donofrio left17:24
→ thiago joined17:25
← peirik_ left17:25
→ jab416171 joined17:26
← stuv left17:26
→ stuv joined17:27
← jenia2`` left17:32
→ pebble joined17:34
← jab416171 left17:36
← delay left17:36
→ jab416171 joined17:41
AmR_ → AmR17:41
← jab416171 left17:46
→ jab416171 joined17:46
→ delay joined17:48
→ hbautista__ joined17:49
→ newpy joined17:50
→ donofrio joined17:50
← hbautista_ left17:51
← delay left17:53
← becm left17:58
→ ocra8_ joined18:00
← AmR left18:01
← ocra8 left18:03
← pebble left18:08
→ AmR joined18:10
→ wootehfoot joined18:15
← epicout left18:27
→ iomari891 joined18:30
← B-| left18:30
← jab416171 left18:31
→ epicout joined18:32
← epicout left18:32
→ epicout joined18:32
← tomboy64 left18:35
→ jab416171 joined18:37
← peirik__ left18:41
→ delay joined18:41
← delay left18:41
→ peirik_ joined18:41
→ tomboy64 joined18:44
← tomboy64 left18:45
→ tomboy64 joined18:46
← hdh left18:47
→ peirik__ joined18:50
← AmR left18:51
→ peirik___ joined18:51
← peirik_ left18:52
← peirik___ left18:53
→ peirik_ joined18:54
← peirik_ left18:55
→ hdh joined18:55
→ peirik_ joined18:55
← peirik__ left18:56
→ jamiejackson joined18:56
← iomari891 left19:00
→ AmR joined19:07
← hbautista__ left19:09
→ TomyLobo joined19:17
→ hbautista__ joined19:21
→ jinsun joined19:22
← hbautista__ left19:26
BurningHuman So individual repositories for each user works perfectly now, now I try to have a shared repository where for example two users can clone,pull and push to/from.19:31
I have now created a directory under /shared/sharedtest, where sharedtest is a basre repository created by "user1". Access rights to all files look like this: "drwxrwx--- 3 root sharedgit", where user1 and user2 are withing the sharedgit group. So far so good19:32
cloning with user1 is without any problems19:32
but with user2 it get this error: fatal: detected dubious ownership in repository at '/shared/sharedtest'19:33
Do you guys know how to solve this properly?19:33
Oh yea, and the repository is of course owned by "user1" and not root.19:34
← jmd left19:37
→ specing joined19:37
← henry40408 left19:44
← CrunchyFlakes left19:44
→ henry40408 joined19:44
→ ferdna_ joined19:46
→ CrunchyFlakes joined19:46
← ferdna left19:47
→ jenia2`` joined19:47
→ currenthandle joined19:53
fossdd_ → fossdd19:57
← davispuh left20:08
→ peirik__ joined20:09
← donofrio left20:09
← ferdna_ left20:11
← lucasta left20:11
← jenia2`` left20:15
← benaldo left20:17
→ benaldo joined20:18
→ jenia2`` joined20:19
← dmj` left20:21
BurningHuman Someone here?20:24
mefistofeles nope20:24
BurningHuman A damn it20:24
← BurningHuman left20:27
→ BurningHuman joined20:30
BurningHuman Heyo has somebody ever done a plain git repository in a network where mutliple users can have access to it? Even though other users have access to the directory and files, they cannot clone or push/pull from it. Is there anything I am missing?20:31
← TomyLobo left20:33
→ hbautista__ joined20:35
BtbN you mean like a repo on a samba/nfs share, and people want to commit?20:35
for git, those are all one user, and it'll be messy20:35
much better to not do that20:35
BurningHuman No No not like this20:36
Simply, one simple linux server20:37
where under /shared/repository.git is a repository that users of a specific group have access to20:37
← hbautista__ left20:37
BurningHuman No webbrowser, just plain git command line stuff20:37
and ssh20:37
→ hbautista__ joined20:38
weaksauce BurningHuman https://git-scm.com/book/en/v2/Git-on-the-Server-Getting-Git-on-a-Server20:40
BtbN That's the normal way of doing that, and how basically everyone does it20:40
weaksauce you read through that?20:40
→ MrCryo joined20:43
BurningHuman Yes I have20:54
→ Crispy joined20:57
← peirik__ left20:59
→ donofrio joined21:00
← MrCryo left21:01
← tec left21:06
→ tec joined21:06
← wootehfoot left21:08
← donofrio left21:18
→ m0viefreak joined21:19
← jamiejackson left21:24
→ Turn_Left joined21:27
← AmR left21:28
← pedahzur left21:28
→ gast0n joined21:29
← jenia2`` left21:29
→ ash_worksi joined21:30
ash_worksi suppose I have commits with 1 letter hashes from a..h21:31
BtbN h?21:32
ash_worksi if I `git diff d h` I would think that would be the differences between file systems on d and file systems on h21:32
← konsolebox left21:32
ash_worksi BtbN: it's jsut a random letter in the middlish of the alphabet21:32
BtbN It's not a valid hash though21:33
and yes, that command will generate a diff/patch to get from d to "h"21:33
ash_worksi BtbN: oh right, sorry, yes I am just thinking of the little git tutor which I thought were one letter commits21:34
→ pebble joined21:35
ash_worksi anyway, so, the reason I'm doing this is because 2 servers are not in sync21:36
which I will simply call node1 and node221:37
→ antranigv joined21:37
ash_worksi anyway so when I rev-parse node1, it reports "h" (:P)21:37
when I rev-parse node2 it reports "d"21:37
← hbautista__ left21:38
ash_worksi and when I look at the history for node1, it shows that it does indeed contain a "d" commit21:38
BtbN So, push the missing stuff to it?21:38
ash_worksi BtbN: well, I am trying to see if there's anything that could potentially conflict with how node2 is currently operating before I push, but the point is I seem to be missing something about the diff21:39
BtbN Unless something got really messed up, if it's just missing the latest commits, just push them21:40
ash_worksi because when I `git diff d h` it shows changes to a particular file, when I use `tig` to look at the history of the file on node2, the latest commit is the same changes that were reflected in my diff21:40
BtbN If the two nodes diverged in history, that won't work though, unless you just discard one half of it21:40
ash_worksi my point is, it's weird that that would show up21:41
so I seem to be missing something21:41
BtbN I don't understand what's weird21:41
It's just two instances of the same git repo, and one is older than the other?21:41
← thuna` left21:43
← currenthandle left21:51
→ Left_Turn joined21:53
← Turn_Left left21:56
→ Turn_Left joined21:56
← haxalot left21:58
→ currenthandle joined21:59
← Left_Turn left22:00
→ Left_Turn joined22:01
→ epicout_ joined22:01
← Turn_Left left22:02
BurningHuman Is there a way to tell in the .git files which branches it should protect, so that only one user can merge and push branches to the main? Or is that the part where the actual "gitserver" takes over?22:05
← epicout left22:05
→ Turn_Left joined22:05
BtbN That's done via hook scripts or even higher level stuff22:07
git itself has no concept of branch protection22:07
BurningHuman aaah okey thx!22:08
I just had a misconception how I should handle multiple users, but since it is so "prmitive", it's okey for now. I'll what I can do next22:09
← ajfriesen1 left22:09
← Left_Turn left22:09
→ ajfriesen1 joined22:09
ash_worksi can I set tig variables from within tig?22:10
→ systemdlete joined22:19
→ cdown_ joined22:20
← cdown left22:23
cdown_ → cdown22:23
← capn left22:25
→ AmR joined22:26
← epicout_ left22:26
← stef204 left22:28
← aYS1 left22:29
→ haxalot joined22:32
intelfx I'm not entirely sure but perhaps this should be better asked in tig channel or issue tracker?22:32
← thiago left22:34
intelfx but to answer your question, if you mean the built-in configuration variables, they can be changed à la Vim with ":set", like so: ":set tab-size = 8" (all spaces are significant)22:37
if you mean the state variables (e.g. "%(commit)"), they reflect the UI state and cannot be set22:37
in either case, there are no user-defined variables in tig22:38
ash_worksi: ^22:38
← firewyre left22:40
ash_worksi intelfx: it's incredibly frustrating, and I don't see anything on github about it, nor is there really a #tig channel...22:43
intelfx there's gitter channel I think? it is pretty dormant though, but fwiw I'm on it :-)22:43
ash_worksi k, I'll jump into that22:44
← Turn_Left left22:45
ash_worksi intelfx: I haven't used gitter in awhile... how do I add tig?22:45
oh wait, I see it22:45
it just took a long time to show up, its jonas/tig right?22:45
you use @ on gitter right?22:47
intelfx yes it's jonas/tig, it's always equal to the name of the repo on github (which was the point of gitter :-))22:50
→ jensen joined22:57
→ so-offish joined23:00
← jensen left23:03
← BurningHuman left23:05
→ user_____ joined23:06
→ BurningHuman joined23:06
← BurningHuman left23:07
← weaksauce left23:08
→ ferdna joined23:09
← user_____ left23:10
→ user_____ joined23:10
→ jensen joined23:11
→ guardianxxx joined23:13
← Dotz0cat left23:16
→ Dotz0cat joined23:17
→ weaksauce joined23:23
→ BurningHuman joined23:23
→ pedahzur joined23:26
← travisghansen left23:29
→ travisghansen joined23:30
← user_____ left23:30
← rigid left23:34
→ rigid joined23:35
→ euouae joined23:36
euouae Hello I have a website template in a git project23:36
I'm wondering if it is a good idea to script git to have the 'web' tag or branch automatically build the website and serve it?23:36
and if that is something that sounds reasonable, how do I do it?23:37
the idea would be that I can update my website by updating the tag/branch on the server23:37
→ rsx joined23:38
← rsx left23:38
→ lucasta joined23:40
→ thiago joined23:40
← jab416171 left23:41
→ jab416171 joined23:42
← Lord_of_Life left23:42
→ Lord_of_Life joined23:43
← BurningHuman left23:43
← CrunchyFlakes left23:44
→ CrunchyFlakes joined23:46
aninternettroll That's exactly how github pages works (depending on the configuration)23:48
euouae how do they do it then?23:49
aninternettroll https://docs.github.com/en/pages/getting-started-with-github-pages/configuring-a-publishing-source-for-your-github-pages-site#publishing-from-a-branch23:51
Overly simplified you push the compiled static pages to a branch and github puts them on their servers23:52
euouae I'm not asking for how to do it on GH, I'm asking how GH does it23:52
This is about git, not github23:52
how do you trigger an action on git when branch/tag changes?23:52
aninternettroll Github doesn't use git the same way you do. They have custom servers that hook into the push process and trigger what they have to trigger with those. For all intents and purposes it's magic23:53
though I guess git has the update hook23:54

Logs Search ←Prev date Next date→ Channels Documentation