IRCloggy #git 2019-04-21

Provider of IRC logs since 2005.

WARNING: As Freenode became unjoinable and lost all warnings in topics, we cannot log channels on Freenode anymore.

	2019-04-21
rafasc	git diff --staged; git diff; covers the first two.	00:00
	← de-facto left	00:01
	→ de-facto joined	00:01
rafasc	-1 --base, -2 --ours, -3 --theirs cover the last ones.	00:01
enigmus	rafasc: ah! Thanks. I can't write "git diff -1 refs/heads/xxx" though, right?	00:05
rafasc	So the tools are there, and you can place your own wrapper with that logic on your path.	00:06
	→ fphilipe joined	00:12
	→ jottr joined	00:12
	← kadoban_ left	00:12
	← fphilipe left	00:17
	→ z\|bandito joined	00:18
	← jottr left	00:18
rafasc	enigmus: well, 2 will be HEAD, 3 will be MERGE_HEAD, and 1 will be merge-base HEAD MERGE_HEAD;	00:19
	also, the nature of merges and how they can mismerge things is something that is hard to automate.	00:20
	https://youtu.be/8ET_gl1qAZ0?t=2495	00:22
	→ plongshot joined	00:24
rafasc	enigmus: I don't think you can, but you can specify the stage in diff commands for specific files.	00:27
	git diff :1:README HEAD~5000:README	00:28
plongshot	I'm not quite sure how to phrase my question.. I'm assuming a bitbucket wiki repository is it's own separate repo from the repositry the code is in (even thought it's the same project)? If that is true then my question here is - ways to manage that? I heard about something called "grafting". Could that help in my case?	00:29
	I mean. I guess my ultimate, first decision is whether to manage two separate repos under the same project or to try to combine them?	00:30
rafasc	plongshot: grafting is something else.	00:32
plongshot	oh	00:32
	→ learningc joined	00:32
osse	plongshot: One git repository can contain two branches that don't share a common root.	00:32
rafasc	it's something that allows you to "lie" about history	00:32
osse	In that sense you can manage two different repositories in one	00:33
plongshot	osse: So you are also saying 'more than one root in a single repository' ?	00:33
	Like two repos in one?	00:33
osse	Essentially yes	00:34
rafasc	seems like bitbucket wiki are actually their own repos, in contrast to things like gh-pages which are just a branch.	00:34
osse	Branches with no common root can be merged, but they don't have to	00:34
plongshot	I still have not verified for certain but it seems as thought they are separate repos	00:35
	→ mven joined	00:35
rafasc	plongshot: https://confluence.atlassian.com/bitbucket/clone-and-edit-pages-317195798.html	00:35
plongshot	osse: Oh fuck! I just thought of something glorious!!	00:36
	ohmygod!	00:36
rafasc	you can clone them directly, so they are their own standalone repo.	00:36
plongshot	I think.. I think I can just add the additional remot to the wiki repo on my code repo (locally) and that will...	00:36
	well be the start of something I don't know how to finish	00:36
rafasc	plongshot: that's the approach osse was mentioning.	00:37
osse	plongshot: Not sure about the details but it sounds like you're on the right track.	00:37
plongshot	that popped into my head but not certain on the details how to handle that	00:37
	→ R2robot joined	00:37
	← savolla left	00:37
osse	At least you've "grokked" something fundamental about how git works	00:37
rafasc	but the advantages of doing that are limited... Especially since it can be a pain to switch branches.	00:37
	because the files are completely unrelated.	00:38
plongshot	rafasc: i think thats one of the questions I didn't know to ask (nor the answer to)	00:38
osse	plongshot: The git repo of git itself has seven roots	00:38
plongshot	oh wow	00:39
rafasc	the simpler solution is to use worktrees in that case, but then you just better off having as separate repo directories.	00:39
nedbat	osse: why does it have seven roots?	00:40
plongshot	Idk what good it does me to add a remote to the repo going to the wiki. Unless its to pull or clone or sync or fetch right into the existing code repo locally. But idk what the result would be after doing it?	00:40
rafasc	nedbat: some projects were started separated and were merged in, like git gui.	00:40
plongshot	If I'm on the right track?	00:40
	a worktree	00:41
rafasc	osse: I have 13 roots! 0.0	00:41
osse	rafasc: Replace --all with origin/master :P	00:42
	← mven left	00:43
osse	plongshot: Fetching just updates stuff. You can add the Linux repo as a remote to your dotfiles repo and fetch.	00:44
rafasc	the kernel also has a couple.. At least one of them was introduced accidentally, and Linus was a bit upset about not catching it sooner.	00:44
osse	plongshot: !fetch	00:44
gitinfo	plongshot: When you work with remote repositories, Git stores copies of the remote's branches in !tracking_branches (basically mirrors). You can use 'git fetch' to update those. If you want to actually apply changes from the remote to your checked out branch, a second step is needed, usually 'git merge' or 'git rebase'. There's also 'git pull' which combines both steps.	00:44
osse	!remote_tr	00:44
gitinfo	[!remote_tracking_branch] Remote-tracking branches (branches which start with e.g. 'origin/', listed by 'git branch -r') are read-only mirrors of the branches in another repository. They're updated by 'git fetch'. You can't edit them directly (trying to check them out results in a !detached HEAD), but you can create a local branch based on a remote-tracking branch using e.g. 'git checkout -b <branch> <remote>/<branch>'	00:44
enigmus	rafasc: by the way, thanks a lot for all the details	00:46
	→ kadoban_ joined	00:47
	→ mven joined	00:48
rafasc	accidental root: http://lkml.iu.edu/hypermail/linux/kernel/1603.2/01926.html	00:49
	→ donofrio_ joined	00:49
	← donofrio left	00:50
rafasc	plongshot: it will be a lot easier on you to manage it as separate repos. Especially because when you're writting documentation, it is usually useful to reference the code. But if you keep it on the same repo as different branches, you won't be able to do that.	00:52
	→ fphilipe joined	00:52
	→ causa_sui joined	00:53
	← Phylock left	00:53
	← mooasaurus left	00:54
	→ mooasaurus joined	00:55
	← causa_sui left	00:58
	← Trieste left	00:58
	← kadoban_ left	00:59
	← freeman42y left	01:00
	→ Trieste joined	01:08
	← hussam left	01:13
	→ pupsikov joined	01:15
	← Ignacy left	01:15
	← rafasc left	01:17
	← F0rTex left	01:19
	→ F0rTex joined	01:20
	→ finalbeta joined	01:21
	← mven left	01:25
	← fphilipe left	01:25
	← zonixo left	01:27
	← m0viefreak left	01:28
	→ Ignacy joined	01:29
	→ kadoban_ joined	01:29
	← kadoban_ left	01:32
	→ YuGiOhJCJ joined	01:34
	← donofrio_ left	01:35
	← royal_screwup21 left	01:37
	→ jottr joined	01:38
	→ donofrio joined	01:40
	← MACscr left	01:41
	→ boombatower joined	01:42
	← jottr left	01:43
	← Ignacy left	01:45
	→ Beam_Inn__ joined	01:48
	← plongshot left	01:48
	← Beam_Inn_ left	01:50
	Beam_Inn__ → Beam_Inn_	01:50
	← ferdna left	01:56
	← igemnace left	01:57
	→ mowcat joined	02:07
	→ causa_sui joined	02:08
	← causa_sui left	02:13
	lagothrix → Guest14695	02:17
	← Guest14695 left	02:17
	→ lagothrix joined	02:17
	← pupsikov left	02:18
	→ ferdna joined	02:20
	← boombatower left	02:29
	→ mven joined	02:32
	→ kadoban_ joined	02:33
	← Anthaas left	02:37
	→ Anthaas joined	02:37
	← mven left	02:37
	→ mven joined	02:40
	← kadoban_ left	02:43
	→ duderonomy joined	02:49
	← RustyShackleford left	02:52
	← rsrx left	02:55
	← HotBeefDip left	02:57
	→ axsuul joined	03:00
	← duderonomy left	03:01
	→ HotBeefDip joined	03:01
	← HotBeefDip left	03:01
	→ jottr joined	03:04
	→ kerframil joined	03:08
	← jottr left	03:09
	← omnireq left	03:16
	← rkta left	03:17
	→ rkta joined	03:17
	→ omnireq joined	03:18
	← omnireq left	03:19
	→ omnireq joined	03:20
	→ fphilipe joined	03:22
	→ causa_sui joined	03:23
	← rosco_y left	03:25
	← causa_sui left	03:27
	← jstimm left	03:28
	← moldybits left	03:30
	← jnewt_ left	03:30
	← enigmus left	03:37
	← mven left	03:40
	← th_ left	03:41
	→ nurdie joined	03:41
	→ jottr joined	03:42
	mudmat → mud	03:42
	← mud left	03:43
	→ mud joined	03:43
	← mud left	03:43
	→ mud joined	03:43
	← arecacea1 left	03:44
	→ arecacea1 joined	03:45
	→ moldybits joined	03:45
	← zulutango left	03:46
	← molusc left	03:47
	→ molusc joined	03:47
	← jottr left	03:48
	← Beam_Inn_ left	03:52
	→ n3wborn joined	03:52
	← mowcat left	03:53
	← fphilipe left	03:56
	→ mven joined	04:00
	→ duderonomy joined	04:02
	← Shentino left	04:04
	→ Shentino_ joined	04:04
	→ netj joined	04:08
	← duderonomy left	04:12
	→ kadobanana joined	04:13
	← craftyguy left	04:15
	→ nabil joined	04:17
	→ craftyguy joined	04:17
	← mven left	04:21
	← kadobanana left	04:21
	→ bambanx joined	04:22
	→ kadobanana joined	04:22
	→ fphilipe joined	04:29
	← APic left	04:30
	← kadobanana left	04:31
	→ mven joined	04:33
	→ veegee joined	04:36
	→ causa_sui joined	04:38
	← mven left	04:39
	→ jottr joined	04:39
	→ APic joined	04:40
	← causa_sui left	04:43
	← jottr left	04:44
	→ zulutango joined	04:45
	← Xeyame left	04:47
	→ dpyro joined	04:57
	→ pks_ joined	05:03
	← fphilipe left	05:04
	← pks left	05:05
	pks_ → pks	05:05
	← g00s left	05:13
	→ jottr joined	05:20
	← jottr left	05:25
	→ tigos2 joined	05:28
	→ RustyShackleford joined	05:29
	← tya99 left	05:29
	← gmh left	05:30
	← kadoban left	05:37
	→ g00s joined	05:47
	← xcm left	05:52
	→ xcm joined	05:52
	→ causa_sui joined	05:53
	← veegee left	05:55
	← causa_sui left	05:58
	→ rohit[m] joined	06:00
	→ jottr joined	06:00
	← victorqueiroz left	06:00
	← jottr left	06:05
	← thiago left	06:05
	← rohit[m] left	06:06
	→ rohit[m] joined	06:06
	← Goplat left	06:06
	← xcm left	06:08
	← rohit[m] left	06:09
	→ xcm joined	06:09
	→ rohit[m] joined	06:09
	→ fstd_ joined	06:12
	← rohit[m] left	06:13
	→ rohit[m] joined	06:13
	← fstd left	06:16
	fstd_ → fstd	06:16
	→ Envil joined	06:20
	← yonder left	06:20
	→ cgi joined	06:21
cgi	I have a few people in a project who checked in with different email addresses. I want to rewrite those email addresses to a common email address. Is there an easy way of doing that? For example [email@hidden.address] and [email@hidden.address] -> rewrite them to the same email	06:21
	→ flavius joined	06:24
	← cdown left	06:30
	→ cd joined	06:31
	→ Tawonga joined	06:31
	→ cdown joined	06:33
	→ plexigras joined	06:37
_ikke_	man git mailmap	06:41
gitinfo	the git manpage is available at https://gitirc.eu/git.html	06:41
_ikke_	hmm	06:41
	man git check-mailmap	06:42
gitinfo	the git-check-mailmap manpage is available at https://gitirc.eu/git-check-mailmap.html	06:42
	→ noobJS joined	06:48
	← noobJS left	06:52
	→ czart joined	06:54
	→ thefatma joined	06:59
	→ fphilipe joined	06:59
	← bambanx left	07:04
	→ kapil____ joined	07:06
	→ causa_sui joined	07:08
	→ achen_ joined	07:12
	← causa_sui left	07:13
	→ Mattiaslndstrm joined	07:15
	← thefatma left	07:16
	→ duderonomy joined	07:16
	← Mattiaslndstrm left	07:19
	→ zonixo joined	07:22
	→ lomex joined	07:25
	← kjartan left	07:29
	← tigos2 left	07:30
	→ kjartan joined	07:34
	→ ggSo joined	07:38
	← zonixo left	07:40
	→ f00lest joined	07:41
f00lest	will the owner of a private gitlab repository know if I have cloned it using ssh, will my ssh information be recorded?	07:41
	← ferdna left	07:43
	→ tm007 joined	07:48
	→ Noti joined	07:51
	← ggSo left	07:55
	← lomex left	08:03
	← Noti left	08:03
j416	they probably have logs	08:03
	← nic-hartley left	08:03
	→ Regon joined	08:03
	→ Noti joined	08:07
	→ royal_screwup21 joined	08:10
	→ leeN joined	08:12
	← Noti left	08:16
	→ Mattiaslndstrm joined	08:19
f00lest	j416: If I clone using http instead of ssh will it be the same?	08:24
	→ causa_sui joined	08:24
xcm	f00lest: if they are in control of the server, they can log anything the server does, including http requests	08:26
f00lest	xcm: repository is on gitlab as a private repo	08:27
xcm	gitlab is a piece of software, but now i'm starting to think you're talking about gitlab.com	08:28
f00lest	xcm: pardon me for not being clear, aye gitlab.com	08:28
	→ veegee joined	08:29
xcm	afraid i'll have to redirect you to gilab.com's support :(	08:29
	← causa_sui left	08:29
xcm	since it's a private repo, i imagine you still need to auth to fetch things. but whether they make these logs available to the repo owner has even less to do with #git	08:30
f00lest	xcm: thanks anyways, :p	08:30
	← f00lest left	08:30
	→ goron joined	08:32
	← plexigras left	08:33
	← dpyro left	08:33
	→ lomex joined	08:33
	← fphilipe left	08:36
	← R2robot left	08:37
	← goron left	08:39
	→ R2robot joined	08:40
	veremitz → {{veremitz}}	08:58
	→ fphilipe joined	09:00
	← dhollinger left	09:00
	{{veremitz}} → veremitz	09:00
	→ dhollinger joined	09:01
	→ cybrNaut joined	09:01
	← g00s left	09:01
	← azwieg103 left	09:05
	→ HotBeefDip joined	09:06
	→ yyy joined	09:16
	← nabil left	09:18
	← catsup left	09:19
	→ catsup joined	09:19
	→ nowhere_man joined	09:20
	← catsup left	09:20
	→ catsup joined	09:20
	← rohit[m] left	09:26
	→ rohit[m] joined	09:26
	← jcbitter left	09:30
	→ mowcat joined	09:34
	→ chinhoye joined	09:38
	← chinhoye left	09:39
	→ causa_sui joined	09:39
	→ KnoP joined	09:40
	← causa_sui left	09:43
	→ jcbitter joined	09:46
	← achen_ left	09:46
	→ Noti joined	09:46
	→ savolla joined	09:47
	← jcbitter left	09:54
	→ Puffball joined	09:54
	→ achen_ joined	09:56
	← lankanmon left	10:01
	→ basiclaser joined	10:03
	→ tmm88 joined	10:05
	← HotBeefDip left	10:06
	← finalbeta left	10:07
	→ finalbeta joined	10:07
	→ tmm8876 joined	10:08
	← tmm8876 left	10:08
	← tmm88 left	10:09
	→ tmm88 joined	10:09
	→ jas4711 joined	10:13
	← Noti left	10:14
	→ jottr joined	10:15
	→ Noti joined	10:19
	→ Phylock joined	10:20
	← Anthaas left	10:20
	→ Anthaas joined	10:21
	→ m0viefreak joined	10:25
	← lomex left	10:26
	LogicAside → ThatOtherGuy	10:30
	← nuc left	10:33
	← fakhraee left	10:34
	← fphilipe left	10:34
	← yyy left	10:34
	→ fphilipe joined	10:34
	→ nuc joined	10:35
	→ fakhraee joined	10:37
	→ rsrx joined	10:41
	← zulutango left	10:48
	→ lankanmon joined	10:51
	← m0viefreak left	10:53
	→ causa_sui joined	10:54
	← llinguini left	10:56
	← causa_sui left	10:58
	→ yyy joined	11:07
	← jottr left	11:07
	← savolla left	11:08
	→ Lucas_Gray joined	11:09
	← kerframil left	11:11
	← anubani left	11:12
	→ SkarmoutsosV joined	11:14
	→ zulutango joined	11:18
	← bashfulshell left	11:18
	← royal_screwup21 left	11:18
	← zulutango left	11:19
	→ zulutango joined	11:22
	← SkarmoutsosV left	11:26
	← kjartan left	11:41
	→ ricekrispie2 joined	11:44
	← tmm88 left	11:46
	→ kjartan joined	11:46
	← ricekrispie left	11:46
	→ Wryhder joined	11:47
	← Lucas_Gray left	11:48
	Wryhder → Lucas_Gray	11:48
	→ plantroon joined	11:50
	← mooasaurus left	11:50
	→ mooasaurus joined	11:51
	← DropItLikeItsHot left	11:51
	← downer06 left	11:52
	→ ave_milia joined	11:52
	→ downer06 joined	11:53
plantroon	I have the so called bare repositories like project3.git created on my server and I am using them over ssh. Now what's the best practice on adding such remote repository via `git remote add ssh://git.example.com/srv/git/project3.git` - do I need to include the .git when adding the remote? Because it works both ways, I don't know why. Is this .git extension something special?	11:53
	→ DropItLikeItsHot joined	11:54
bremner	yes	11:54
_ikke_	plantroon: the best is to use the name as it's on disk	11:59
	but if you leave out .git, and git doesn't find it, it will try with .git appended	11:59
	← Have-Quick left	11:59
plantroon	thanks :) I'll use .git everywhere to make it perfectly clear for me :)	12:01
	one more thing, can I change the "root" of the git repositories? So that I can do `git remote add ssh://git.example.com/project3.git` without having project3.git in my homedir?	12:01
	I can imagine how to do this with http but not ssh	12:02
_ikke_	plantroon: gitolite does it, not sure how though	12:04
	it uses gitolite-shell, which probably has to do with it	12:05
plantroon	_ikke_: I can imagine many ways, like separate ssh chroot or symlinks to homedir but I was wondering if there's some server-side git config that does just this. I want to make this resillient to migrations in the future so that I don't have to change the remotes one by one when the time comes	12:05
_ikke_	plantroon: there is no git config that can do this	12:06
	→ causa_sui joined	12:09
plantroon	well, I'll just pick better paths then :D thanks for the answers/advice	12:11
_ikke_	it's common to have a dedicated user for it	12:11
plantroon	I noticed that, but I see no need for that. Especially in this case where all the hosts are only used by a single person and I already have elaborate ssh mechanisms set up revolving around just a single user running only very few well known verified apps	12:13
	← causa_sui left	12:13
	← Regon left	12:14
	← Puffball left	12:15
	← AKPWD left	12:15
	→ howdoi joined	12:16
	→ jstimm joined	12:22
	→ AKPWD joined	12:22
	← n000g left	12:27
	→ fphilipe_ joined	12:28
	→ n000g joined	12:30
	← fphilipe left	12:30
	→ royal_screwup21 joined	12:32
	← fphilipe_ left	12:33
	→ victorqueiroz joined	12:36
	← xcm left	12:40
	→ xcm joined	12:41
	→ agowa338 joined	12:41
	← Noti left	12:42
	← arecacea1 left	12:45
	→ arecacea1 joined	12:46
	← de-vri-es left	12:49
	→ de-vri-es joined	12:51
	← n3wborn left	12:56
	→ shawn_ joined	12:56
	shawn_ → scientes	12:58
	→ Emperor_Earth joined	12:58
	← DropItLikeItsHot left	12:58
	→ lomex joined	12:59
	→ nabil joined	13:00
	→ fphilipe_ joined	13:07
	→ bashfulshell joined	13:11
	← RoriconKnight left	13:13
	← rohit[m] left	13:13
	→ savolla joined	13:16
	← omnireq left	13:19
	→ hussam joined	13:21
	→ causa_sui joined	13:24
	→ AfroThundr\|main joined	13:25
	← YuGiOhJCJ left	13:25
	← causa_sui left	13:29
	→ triad joined	13:30
	← savolla left	13:40
	→ mven joined	13:40
	← mven left	13:40
	→ HenryCH joined	13:41
	→ savolla joined	13:41
HenryCH	is there any way to retroactively tell git that a file has been renamed, such that a pull request shows the diff rather than -old +new file?	13:42
	← fphilipe_ left	13:42
HenryCH	changes to history would be fine	13:43
_ikke_	Git does not explicitly track renames	13:45
HenryCH	the commit where i renamed and made changes actually looks fine, the diff recognizes it's been renamed	13:47
	← savolla left	13:47
HenryCH	but then i merged master into my branch and now the diff shows -old +new	13:48
	→ savolla joined	13:49
	← Lucas_Gray left	13:50
	← nurdie left	13:52
	→ beastd joined	14:08
	← beastd left	14:08
	← finalbeta left	14:19
	→ finalbeta joined	14:19
	← lomex left	14:19
	→ Goplat joined	14:20
	← HenryCH left	14:25
	← nabil left	14:38
	→ causa_sui joined	14:39
	← causa_sui left	14:44
	← mowcat left	14:48
	→ nurdie joined	14:52
	← nurdie left	14:57
	← jstein left	15:05
	← howdoi left	15:05
	→ kneeki joined	15:05
	← tm007 left	15:09
	→ ferdna joined	15:14
	→ lomex joined	15:18
	← lomex left	15:19
	← achen_ left	15:19
	→ omnireq joined	15:26
	Shentino_ → Shentino	15:27
	→ jottr joined	15:29
	← kneeki left	15:31
	← cgi left	15:34
	→ cgi joined	15:36
	→ fphilipe_ joined	15:38
	← agowa338 left	15:39
	→ dpyro joined	15:42
	→ JoeBaker joined	15:42
	→ thiago joined	15:44
	← jottr left	15:44
	← JoeBaker left	15:45
	→ JoeBaker joined	15:45
	→ thefatma joined	15:48
	← JoeBaker left	15:50
	→ JoeBaker joined	15:51
	→ m0viefreak joined	15:51
	→ nurdie joined	15:53
	→ achen_ joined	15:53
	← kjartan left	15:54
	→ causa_sui joined	15:54
	← thefatma left	15:55
	← xcm left	15:57
	→ xcm joined	15:57
	← nurdie left	15:58
	← causa_sui left	15:59
	→ kjartan joined	15:59
	← JoeBaker left	15:59
	→ lomex joined	16:02
	→ dngray joined	16:03
dngray	with a rebase, can you squash it into the last commit?	16:03
	i guess i could change the date of the commit after squashing	16:03
	← cdown left	16:06
	→ yyy- joined	16:07
dngray	https://stackoverflow.com/questions/454734/how-can-one-change-the-timestamp-of-an-old-commit-in-git	16:08
	ah	16:08
	← yyy left	16:10
	← fphilipe_ left	16:11
	← achen_ left	16:13
	→ rohit[m] joined	16:18
	← learningc left	16:27
	→ achen_ joined	16:31
	← dngray left	16:39
	→ GGMethos joined	16:39
	→ fphilipe_ joined	16:43
	→ noobJS joined	16:44
	← xcm left	16:46
	→ xcm joined	16:47
	← noobJS left	16:49
	→ greatgatsby joined	16:51
	→ dviola joined	16:54
	← BlessJah left	16:54
	→ SkarmoutsosV joined	16:55
amosbird	what are the upper and lower mean? https://la.wentropy.com/kb3Y.png	16:59
	→ czart_ joined	16:59
	→ Mr__Anderson joined	17:00
	← czart left	17:01
	← R2robot left	17:03
	→ analogical joined	17:05
analogical	hello is it possible to use rsync with git bash for windows?	17:07
	→ venmx joined	17:07
_ikke_	I think so, yes	17:07
	← Mattiaslndstrm left	17:08
analogical	how do I add rsync since it doesn't seem to be included by default?	17:08
	→ causa_sui joined	17:10
	← ave_milia left	17:10
	← causa_sui left	17:15
	→ fatalhalt joined	17:15
	← fphilipe_ left	17:17
	→ mra90 joined	17:19
mra90	so I have a repository which was forked from linux kernel mainline, how can I check the first commit they did after a fork took place?	17:19
	← greatgatsby left	17:24
	→ neiluj joined	17:25
	→ nightshift joined	17:28
	→ agowa338 joined	17:29
nightshift	uggggg, I may have to scrape the best feature of my gitviewer, at least until I switch it to perl	17:30
	→ jottr joined	17:30
	→ shomon joined	17:31
shomon	hi, happy easter :D	17:31
nightshift	happy easter shomon	17:31
shomon	I'm trying to figured out of all thebranches in a repo, which ones have not been committed to master.. is there a command or quick way to get that?	17:31
	← neiluj left	17:31
shomon	I thought it was git log --first-parent master	17:32
osse	mra90: Do you have the mainline Linux master branch in the same repo?	17:32
	shomon: git branch --no-merged master	17:32
nightshift	shomon: you mean the branches that haven't been merged?	17:33
shomon	yes	17:33
nightshift	ah, yeah, osse's answer.	17:33
	I think	17:33
shomon	thanks both!!	17:33
	← theoceaniscool left	17:33
shomon	hmmm but I think I tried that too osse.. it gives me only one result here locally but I don't see all repos in master history	17:34
	→ neiluj joined	17:34
nightshift	shomon, is this a publicly available repo? I think there's a terminology issue going on.....	17:35
shomon	no, a private one on bitbucket	17:36
nightshift	would be easier to solve the issue if we can know what you're seeing versus what you're expecting to see	17:36
	← cgi left	17:39
	→ noobJS joined	17:41
	← achen_ left	17:42
	← neiluj left	17:42
	← czart_ left	17:43
	→ ravndal joined	17:45
	← perrier-jouet left	17:46
	→ plongshot joined	17:46
plongshot	I thought git branch -r was supposed to list remote branches?	17:47
dmc	it does	17:47
shomon	ah so maybe git branch -r --no-merged master	17:47
	I'll try	17:47
	gotcha	17:47
plongshot	I have a repository using ssh connection to bitbucket but when I run git branch -r there is no output. The upstream connection is there and works fine.	17:47
shomon	ok that worked for me too thanks plongshot	17:48
dmc	try fetching first	17:48
rewt	that will list remote branches that have not been merged to the local master, which may not be at the same point as the remote master	17:48
shomon	yeah I wanted remote master	17:49
rewt	git branch -r --no-merged origin/master	17:49
	or if your remote is not named origin, replace that part with the remote you're insterested in	17:49
	→ thefatma joined	17:50
plongshot	dmc: np but help me understand.. The way this repos was created was git init done local then a empty repo created in bb, then git push --mirror <git@bitbucket/repo.git last was done locally. There are no changes or commits ahead on either. What does fetch do for this then?	17:50
	oh, i created some branches before pushing up to bb	17:50
	← nowhere_man left	17:51
nightshift	are all the branches on bb?	17:51
	← noobJS left	17:52
dmc	did you add the repo with `git remote`?	17:52
plongshot	The two repos are identical, new repo with a single (initial commit) file "LICENSE" and 4 branches created locally but nothing moire done. then mirror push rom local to upstream. I can see in my ide that there are no upstream or downstream changes. I will only have downstream since only me working.	17:52
nightshift	plongshot, so you never ran git remote add?	17:53
	← dviola left	17:53
plongshot	oh I'm sorry also one part after creating the repo local is a had to run git remote add git@bitbucket/repo.git run locally	17:53
	So I really dunno why I wouldn see the remote branch when running git branch -r	17:54
	yes	17:54
	or I would not have been able to push	17:54
	I better take a closer look	17:55
	→ HotBeefDip joined	17:55
plongshot	ty	17:55
nightshift	well, you CAN push without the remote add, you just have to specify the url each time	17:55
	I just ran git branch -r on one of mine and saw the remote branches, but, It's not on bb, just a local network server	17:56
	might be a bb issue	17:56
plongshot	nightshift: Well I think that is how I formed the push command when I ran it (with the ssh address specified). So not sure now. I was having problems with bb about ssh working.	17:56
mra90	so I have a repository which was forked from linux kernel mainline, how can I check the first commit they did after a fork took place?	17:56
plongshot	I had filed a ticket then thought it was resolved and closed it	17:56
	thx man peace	17:56
nightshift	hmmmmmm, good luck figuring it out plongshot	17:56
	← thefatma left	17:56
plongshot	I know how to find out	17:57
	nightshift: thx	17:57
	← jottr left	17:57
	→ dviola joined	17:59
	→ plexigras joined	18:02
osse	mra90: Do you have the mainline Linux master branch in the same repo?	18:08
mra90	osse: no	18:09
	separate one	18:10
	how can I have it in the same repo? The second on is forked from mainline	18:10
	→ ggSo joined	18:13
	→ czart joined	18:18
	← venmx left	18:18
	← lomex left	18:20
plongshot	I want to do something specific, I have a realllly good reason why and I'm seeing the same stale information online that I don't think will lead me there. How can I mirror a local repo to the upstream repo? I am the only person (who may ever) work on this project and I am the only one now. I would like to create a situation where the operatoins done with the remote (push, pull, sync, fetch..) are --mirror by default.	18:21
	← KnoP left	18:22
plongshot	I know I prolly said some things that don't fit. I have read the man page several times. I see a flag "-mirror[=(push\|fetch)]" but not sure if it applies or how to use it.	18:22
	The changes made locally are always --mirror pushed and then go disable the ability to pull or sync on the remote end. (So changes can flow in only one direction).	18:24
	→ jottr joined	18:24
	→ nic-hartley joined	18:24
	→ causa_sui joined	18:25
	→ KnoP joined	18:25
	→ JoeBaker joined	18:26
	← rohit[m] left	18:26
	→ rohit[m] joined	18:27
	→ Regon joined	18:27
plongshot	There has to be tools (sugar) to handle all this distributed connections and remotes and stuff	18:28
	→ Mattiaslndstrm joined	18:29
	← causa_sui left	18:29
	← jottr left	18:30
	→ R2robot joined	18:33
osse	mra90: You can add the original repo as a remote, fetch from it, then use git log	18:35
	Something like git log mainline/master..mybranch	18:35
	Assuming you name the remote mainline	18:35
	→ n3wborn joined	18:36
mra90	osse: sounds good but how to add that mainline repo as remote?	18:37
	→ nowhere_man joined	18:37
	← rsrx left	18:38
	← n3wborn left	18:40
	→ greatgatsby joined	18:40
	→ n3wborn joined	18:42
nightshift	plongshot: it's my understanding that most mirrors are handled via hooks	18:42
	But, I haven't looked in to actually maintaining one, so, could be wrong	18:43
	→ manuelschneid3r joined	18:45
osse	mra90: git remote add mainline http://kernel.org/or/whatever.git; git fetch mainline	18:46
plongshot	nightshift: ok. I see information on other use cases which are similar but not my own. And, while I suspect a certain flag for the git remote command may help me, there doesn't seem to be any more friendly information on "-mirror[=(push\|fetch)]" than what the man page says about it and I'm not seeing it.	18:46
	→ antimatroid joined	18:46
plongshot	information\information online	18:46
antimatroid	anyone on here make websites at all? I made my own site manager (https://nifty-site-manager.com) that works a lot like latex and git, I really like it and think others may like it as well..	18:46
nightshift	I understand plongshot, sorry :(	18:46
plongshot	just aking anyone. I appreciate ya man	18:47
antimatroid	I would love to know what people think, whether they just look at the site or have a play..	18:47
nightshift	of course, trying to help best I can!	18:47
	antimatroid, I do websites, but, I HATE page generators. Just personal opinion there....	18:48
	(I do use a couple drag and drop tools to help me play with layout, sometimes, but the idea of using a tool where I don't have full control of the code annoys me)	18:49
mra90	osse: thanks, I will give it a try	18:49
	→ zonixo joined	18:51
	→ azwieg103 joined	18:52
	← ggSo left	18:52
antimatroid	nightshift: what about page generators don't you like? this might not even do that..	18:53
	it has like @input commands similar to latex, so you can break things up in to multiple files, which can also be used for template files etc.	18:53
nightshift	antimatroid, I want full, absolute control of my code. Also, this is actually off topic for this channel	18:54
	← hvxgr left	18:54
plongshot	Does it seem right the way I'm reading this in the man page? ($ man git remote \| grep mirror). It looks like I can run something like git remote add --miirror=push run locally and run it one time to create a special kind of branch? Then every time I push from local to remote is will be done as a --mirror push even if I don't use the --mirror flag ever time I push?	18:54
gitinfo	the git-remote manpage is available at https://gitirc.eu/git-remote.html	18:54
plongshot	It sounds like that's what they are saying but it's a bit ambiguous to me.	18:55
osse	mra90: Do you know that the repo you have is a fork of Linux? Or have they started out with a tar ball of Linux and only have their own history?	18:55
	← azwieg103 left	18:55
osse	Because if it's a separate repo started from some copy of git then my approach won't work	18:55
antimatroid	nightshift: you do have absolute full control of your code with this I think, I'd be curious to know if I'm right there if you do happen to take a look. My apologies for being slightly off-topic, however it's something that I think website makes who use git will really like, I'm not really sure what the best way to try and find those people is? Any suggestions would be awesome..	18:56
nightshift	plongshot: that's what it seems to be saying as well	18:56
	antimatroid, that fact that you (appear) to be using latex to generate the pages says not as much control as I'd like, because I'd still be depending on what you, and latex, think good html looks like	18:58
antimatroid	nah I have my own code for that, just use similar syntax	18:58
	it's a c++ program that 'injects' the text from one text file to another	18:58
	you can have a whole dag of inclusions and it'll throw errors for you if there's an input loop (tell you which file and line number too)	18:59
	I just like the latex syntax, less for me to learn/remember	18:59
	I may end up using something like this for programs that I'd like to be able to 'build' in to one single source file before compiling too, that could be an interesting use of this kind of thing	19:00
	→ Jubei joined	19:00
plongshot	This is what I was talking about earlier. https://imgur.com/a/cd6M8nR git remote add says the remote exists already but git branch -r yeilds no output and git branc -a shows no remotes	19:00
Jubei	hello :) I checked out an earlier commit, and that put me in detached head mode. I have since made a few commits but now I want to make my current state master (i.e. drop the old master)	19:01
	how do I do that?:)	19:01
nightshift	standby plongshot, I just read something that MIGHT explain your problem	19:01
plongshot	cool	19:02
	i b here	19:02
nightshift	plongshot "When a fetch mirror is created with --mirror=fetch, the refs will not be stored in the refs/remotes/ namespace, but rather everything in refs/ on the remote will be directly mirrored into refs/ in the local repository. This option only makes sense in bare repositories, because a fetch would overwrite any local commits."	19:03
	→ jottr joined	19:03
nightshift	notice the stuff about there being no refs/remotes	19:03
plongshot	nightshift: right. But for me it would be the other direction and I think that siad it does retain / push the refs going up	19:04
nightshift	In other words, depending on what you did, the mirror may not know it's a mirror	19:04
Jubei	sorry figured it out plx ignore me.	19:05
plongshot	There is a couple things going on. I'm not sure where the problem lies. In addition to wanting to set up this mirror thing, there seems to be a problem with the existing repo setup.	19:05
	← MylesBorins left	19:05
plongshot	That's the second / other question that popped up first and then again a min ago	19:05
	→ MylesBorins joined	19:05
plongshot	nightshift: I see what you mean	19:05
	ok	19:05
	It is a almost empty repo no consequence to just do over. I'mna try learn something here. thx man	19:06
nightshift	of course. I'm looking at a number of pages here, trying to understand how this is done.	19:07
	Because, honestly, at some point I might want to have a full mirror of at least some of my repos too	19:07
	old article, but, might help http://blog.plataformatec.com.br/2013/05/how-to-properly-mirror-a-git-repository/	19:08
	← MylesBorins left	19:09
	← jottr left	19:09
	→ gloomy joined	19:11
	← Jubei left	19:12
	→ MylesBorins joined	19:13
plongshot	nightshift: Yes. Not only is the use case as a backup in a private repo but also if you chose to use bitbucket or github solely to publish software (so the wiki maybe and maybe some other tools) but not a place of coordingation for a team - only to dissemiinate to the public only.	19:13
	→ fphilipe_ joined	19:13
plongshot	:>	19:13
	could be use case I mean	19:13
	→ probs joined	19:15
nightshift	yeah, that's what I'm thinking, the wiki, and issue tracker, but the actual dev is done locally, the "official" bare repo is on my local network, and, when I think I have an issue fixed, it gets pushed to my local server, then mirrored from there	19:15
probs	Hey I'm using gitbash to try and install make on windows 10	19:15
	I've created a makefile and tried running it with 'make' and I was getting the error "bash: make: command not found" so I googled it, downloaded minGW, installed a few packages (mostly C++ compiling ones), but mingw-get still doesn't work as a command in bash	19:15
nightshift	probs, that kind of stuff is why I've moved away from windows.....	19:17
	maybe try cygwin if you want bash on windows....	19:17
	(it's my understand that gitbash isn't a fully functional bash environment, but I could be wrong)	19:18
	→ steven joined	19:19
probs	ugh I just spent a few hours watching tutorials on how to use gitbash	19:20
	← kapil____ left	19:21
	→ null1337 joined	19:22
nightshift	plongshot, further reading (this time on gitlab) shows that cron/hooks ARE involved it getting the auto-push/pull, at least in their implementation, sounds like it's the same for github	19:24
plongshot	Is git remote add supposed to give output if it succeeds?	19:24
	nightshift: thx. I will want to automate it too	19:25
nightshift	um, I don't think so plongshot	19:25
	standby....	19:25
plongshot	So I run git remote add --mirror=push git@bitbucket/my-repo.git and get a new command line. Then, when I do git branch -r there is no putput (new command line) and git branch -a does not show anything remote	19:26
	← null1337 left	19:26
nightshift	plongshot, you haven't pushed it yet	19:26
	unless the = is a typo	19:26
	All you've done is added the remote	19:27
	now you need to git push	19:27
plongshot	nightshift: I did the git remote add locally before pushing anything up. The remote is sitting there empty at the time I issue the remote add command	19:28
moritz	remote add != push	19:28
plongshot	moritz: man git remote	19:28
gitinfo	moritz: the git-remote manpage is available at https://gitirc.eu/git-remote.html	19:28
moritz	remote add just adds the meta data, it doesn't actually do anything over the network	19:28
	→ gareppa joined	19:28
plongshot	there is an option there see it in the first or second line (for git remote add) that shows "--mirror=..."	19:29
moritz	When a push mirror is created with --mirror=push, then git push will always behave as if --mirror was	19:29
	passed.	19:29
nightshift	git remote add just tells you local .git that you have a remote. the remote doesn't get data until you git push	19:29
plongshot	--mirror=[fetch\|push]	19:29
moritz	you still have to git push	19:29
plongshot	moritz: that's what I'm after	19:29
	you got it!	19:29
nightshift	plongshot, so, you have your git remote add --mirror=push....	19:29
plongshot	When I do git branch -r it is reading from upstream not local?	19:29
nightshift	you get a new prompt. NOW do git push......	19:30
moritz	plongshot: git branch -r is a local operation	19:30
plongshot	nightshift: I try to	19:30
	← gareppa left	19:30
plongshot	moritz: Then if I did a git remote add .... blah... it shousl be listed	19:30
	no?	19:30
	→ gareppa joined	19:30
moritz	plongshot: no	19:30
	plongshot: becaues you haven't done anything yet that talks to the remote	19:30
nightshift	plongshot, not until you actually push, which you haven't done	19:31
plongshot	I understand. ty. I would (at some time) love love LOVE to find out about the internals (git internals) of all that :>	19:31
moritz	to first approximation, fetch/pull and push are the only operations that actually talk to the network	19:31
plongshot	this app blows me away!	19:31
	moritz: glad you addressed that - solid gold	19:32
	ty	19:32
	whn i come here you guys load me down w/ knowledgs. I leave w/ my head swole :p	19:32
moritz	(oh, and clone, and some low-level commands like ls-remote)	19:32
plongshot	right on	19:33
	I have to run for a min but I'll drop a note what happens (you already know).	19:34
	peace	19:34
	← zonixo left	19:35
	← HotBeefDip left	19:39
	→ jnewt joined	19:40
	→ causa_sui joined	19:40
	→ dege joined	19:41
	← causa_sui left	19:44
plongshot	I should be able to just do git push origin then right? (origin was the name given and not anything unique).	19:45
	I guess it won't destroy my repo to try it	19:45
nightshift	won't destroy anything to try it	19:46
	← arecacea1 left	19:46
moritz	having backups never hurts :D	19:46
	→ arecacea1 joined	19:46
	← fphilipe_ left	19:46
nightshift	as for whether you use origin or not, it depends on what you named your remote, which, given the command you gave us for you remote add, may not actually do anything	19:47
	you can check the name by typing git remote while in your repo	19:48
plongshot	I added another screenshot showing the results of git push origin (success!). However, git branch -r and git branch -a still do not show any remote branch. https://imgur.com/7OxqSFV	19:50
	This seems odd to me. I have git init repo, added a file on master, added, and committed it, created 4 branches (all done locally), done git remote add --mirror=push <ssh address>, done git push origin (success) --> but git branch -r \| -a list no remote branches	19:51
nightshift	plongshot, you never set your upstream.....	19:53
	you need the -u on a remote to get the branches there	19:53
	(I think)	19:53
	right now your local git doesn't know it needs to know about your remote branches	19:55
	← thebope left	19:55
plongshot	nightshift: so I needed to run gtit push -u origin? I thought that might limit it to only pushing master though (and defeate the purpose of mirror option)	19:55
nightshift	plongshot, yes	19:56
plongshot	If I run with the -u flag after running the first time, will it reflect in my upstream history as 2 events?	19:56
	I'm nonestly feeling pretty frustrated. Somthing seems retarted and it's probably me. I don't like it	19:57
	why in the hell wouldn't git mirror option on a remote branch not work the way you would expect - Just do it - all of it.	19:57
	:p	19:57
nightshift	I don't think it would show as two events.	19:57
plongshot	lets see :>	19:58
nightshift	And, normally you'd mirror a bare repository, not your working one	19:58
	← probs left	19:58
	← gareppa left	19:58
nightshift	basically, you've missed a step (I get what you're trying to do, but, really, you need a setup closer to what I have, where you'd mirror your "local" bare)	19:59
plongshot	I added a third screenshot https://imgur.com/a/cd6M8nR showing results after running git push -u origin (after fist push). Still no remote branches listed	20:04
	← rohit[m] left	20:08
	→ rohit[m] joined	20:08
nightshift	plongshot, try `git remote show origin`	20:09
	← rohit[m] left	20:10
	→ rohit[m] joined	20:11
	← czart left	20:11
nightshift	I still think this is related to the fact that you're trying to mirror a working repo and not a bare repo	20:12
plongshot	I don't know why but I did a git remot --verbose update. It changed nothign about the output of git branch commands. Yes.. There is pleny of relevant output with show origin. But nothgin explains why the git branch commands wojld not yeild output.	20:13
	→ cdown joined	20:13
nightshift	plongshot, really pretty sure it's because you're using mirror in a way that it's not supposed to be used	20:14
plongshot	nightshift: Earlier (I never said so here) I was looking into making a bare repo upstream to mirror to but had a question about that (assuming its even possible). Then I thought the way man git remote read indicated you didn't need to.	20:15
gitinfo	nightshift: the git-remote manpage is available at https://gitirc.eu/git-remote.html	20:15
plongshot	it on my computer too	20:15
	im reading it over an over	20:15
nightshift	plongshot, do me a favor and try `git push -u origin <branchname>` for every branch you have	20:16
	the manpage is a bit misleading for what you are trying to make it do.......	20:16
plongshot	when it speaks of the --mirror= ... optoin with git remote add, it says it only makes sense when talking about "=fetch" but when talking about "=push" they describe e different behavior.	20:16
nightshift	remember, when you mirror, the mirror things everything is local.	20:16
plongshot	oh wow	20:17
	what!	20:17
	I can picture something like that but what the hell?	20:17
nightshift	?	20:17
plongshot	yeah - I have to find out the interals of that to really get it	20:17
	this is awesome!	20:18
	its working that what counts	20:18
nightshift	:)	20:18
	→ m4rsh joined	20:19
nightshift	you have to set each branch to track upstream, not the repo as a whole.	20:19
plongshot	yeah - upsteram matches but no tags. I guess I gotta do a git push --tags too :)	20:19
nightshift	I would really suggest though, that you give up on trying to mirror your working repo, that's not what mirror is for.	20:19
	yup	20:20
	→ fphilipe_ joined	20:20
nightshift	plongshot, have you read the git book?	20:20
plongshot	nightshift: I thought that's what was indicated in the output of git push -u origin (seen in the screeenshot). It maps each branch to the remote branch with the same name in that output.	20:20
	← moldybits left	20:21
plongshot	nightshift: I can't convince anyone ('ve treid). ... I have to learn how it works for me - jump in deep, way deepp and flail like hell till I get to the top	20:21
	That's what works for me - actually doing it then the conceptual that ties it all together follows	20:21
	but I know of the book - ty	20:21
nightshift	plongshot, git forces you to set the upstream for each branch, because sometimes, in really large projects, a developer will work alone on a branch (usually a small issue or feature), and NEVER push that branch anywhere. To make sure that branch doesn't accidently get pushed, you have to explicitly tell git to push, AND track.	20:22
	I understand, I have to jump deep too, but, there are a few basics that would have saved the last few hours :)	20:23
	as an example, I barely know bash and git, but I'm trying to write a static replacement for gitweb in bash :)	20:24
plongshot	nightshift: Are you speaking in the context of only additional branches from this point? There were 4 branches local before the inital push upstream. That I thought was taken care of.	20:24
	Are you saying that branches created after this will have to explicitly be published?	20:25
nightshift	(and, apparently, one of the things I'm trying to do with that can't be written in pure bash, so I'm doing a massively deep dive into awk because apparently THAT can do it, and can be called from bash)	20:25
plongshot	ok	20:25
	I follow	20:25
	I see	20:26
nightshift	plongshot, the 4 branches you previously created had to have the explicit git push -u, AND all future branches will have to have a git push -u	20:26
	← rohit[m] left	20:26
plongshot	I get so frustrated and bogged down like that I can't go on and I abondon my love, Then all kinds of negative feelings fester. It's not good.	20:26
	→ rohit[m] joined	20:27
cybrNaut	i'm looking for a freedom-respecting repo service that supports web page projects. That means github.com and gitlab.com are ruled out. notabug.org and codeberg.org seem philosophically perfect, but I'm not seeing how to deploy a website on those platforms (apparently no "github pages" equivalent)	20:27
nightshift	If you want them to make it to the server, and have your local repo also know that the server has them, that is	20:27
plongshot	cybrNaut: I use bitbucket and like it	20:27
	They set new repos to private by default and seem to offer a lot for free	20:28
nightshift	plongshot, bitbucket doesn't have his website support though	20:28
cybrNaut	plongshot: i heard bad things about bitbucket using nonfree software and lots of non-free j/s pushed on users.	20:28
nightshift	There's a reason even github removed their github pages	20:28
cybrNaut	ah, well no website support kills it	20:28
plongshot	I didn't know that Sorry to hear it :(	20:28
	→ yyy joined	20:29
plongshot	Is the server version any different? I don't know about github but bb has an installable version that may offer more funcitonality. I have no idea.	20:29
	← rohit[m] left	20:29
plongshot	But that's a whole new ballgame - installing, config, maintianing it youself	20:30
cybrNaut	nightshift: really? i'm using github pages and it seems fine	20:30
nightshift	Yeah, but they're trying to shut it down, I heard	20:30
	→ moldybits joined	20:30
plongshot	there are github pages. I never used them but I know they exist	20:30
	← Envil left	20:31
	← justanotheruser left	20:31
	← basiclaser left	20:32
	← yyy- left	20:32
nightshift	cybrNaut, this is a fairly common question we get (minus the free part), there really aren't any other alternatives.	20:33
	if you don't mind self hosting, gitolite, MIGHT be able to do what you want	20:34
cybrNaut	thanks for the tip but i'm trying to avoid self-hosting ATM.	20:35
	maybe if i get desperate i'll try to put a web server on a openwrt box	20:36
nightshift	there are a lot of people trying to self-roll a solution, haven't heard of anyone with a successful solution	20:36
	cybrNaut, I've got one running on a raspberry pi (along with git and dns), but, my codebase is small. At some point I'll need more cpu and ran	20:37
	er, ram	20:37
	but, I don't use git to populate the sites I have up	20:37
	→ yonder joined	20:38
	← plongshot left	20:39
	→ perrier-jouet joined	20:40
	← ferdna left	20:43
	→ ferdna joined	20:46
	→ Narrat joined	20:47
nightshift	Think I'm gonna shut down for a while and go for a walk.	20:48
	← nightshift left	20:49
	→ HotBeefDip joined	20:51
	→ aruns joined	20:52
aruns	Hi, I am building a Git submodule for common hook scripts we can use at work.	20:52
	What is the best way to handle Windows support?	20:53
moritz	ignore it	20:53
HotBeefDip	I have a group project in one of my university courses that uses a self-hosted GitLab instance...they decided to do maintenance on campus during Easter weekend, so my group doesn't have access to the server...if I push my local repository (that has been linked to my school's GitLab instance up until now) to a brand new repo on GitHub, so my group will have a temporary place to work, will this cause major issues for us when we	20:54
	switch back to the GitLab server?	20:54
moritz	HotBeefDip: no git-level issues	20:54
osse	aruns: Maybe write the scripts in python?	20:54
moritz	just maybe privacy issues, dunno what's in the repo	20:54
Timvde	Huh. Can a git merge overwrite uncommitted changes?	20:55
moritz	Timvde: no, it aborts rather than doing that	20:55
	→ causa_sui joined	20:55
Timvde	moritz: That's what I thought, but it just happened	20:55
HotBeefDip	moritz : nothing of importance...a very poorly designed web-app built by 3rd year undergrads ;-)	20:55
Timvde	File foo and bar. foo is untracked, bar is tracked. Merge in a commit that renames bar to foo. Foo is gone.	20:56
	Is this a bug?	20:56
moritz	Timvde: !repro please	20:58
gitinfo	Timvde: [!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 see	20:58
Timvde	moritz: sure	20:58
moritz	Timvde: I just tried a minimal example, it said: error: The following untracked working tree files would be overwritten by merge:	20:58
	foo	20:58
	is this on a case insensitive (or case-preserving) file system, with different casings for foo involved?	20:59
Timvde	Nope, ext4	20:59
	→ seds joined	20:59
	→ wwwi joined	20:59
Timvde	It's actually a friend of mine who found it, so here's his repo: https://git.principis.be/Principis/zucht	20:59
wwwi	hello	21:00
	← causa_sui left	21:00
Timvde	clone it, reset master to master~, create a file named test.txt and merge master back in	21:00
wwwi	if you use git to upload files in a git server, that means you no longer need to use total commander?	21:00
aruns	osse: That might work, thanks.	21:02
moritz	Timvde: holy shit, you are right. That looks like a bug to me	21:02
	← mooasaurus left	21:02
moritz	I could reproduce it with that repo	21:02
	→ mooasaurus joined	21:02
Timvde	moritz: it's strange that you couldn't reproduce it with your repo	21:02
	I wonder if it has something to do with the "double extension"?	21:02
	but let's start with filing a bug	21:03
	And creating a cleaner version of that repository, because I just noticed that it has bad language (in Dutch) :P	21:04
	→ theoceaniscool joined	21:06
Timvde	moritz: OH	21:07
	I just noticed something that he failed to mention when explaining it to me	21:07
	I'm not sure what behaviour I expect then...	21:07
	moritz: In the first commit (master~), test.txt is in .gitignore; it is removed in the second commit	21:07
	I suppose I can understand the behaviour then... Not sure if I'd consider it to be a bug	21:08
	because as far as git is concerned, the file didn't exist when applying the second commit	21:09
	→ mven joined	21:14
	← JoeBaker left	21:14
	→ sschw joined	21:14
	← sschw left	21:15
aruns	osse: If I create both a shell script and say a batch script	21:19
	Is there a way to get the OS to determine which one to read?	21:20
	Or I suppose I can just call CMD from the shell script.	21:21
nic-hartley	This looks promising: https://stackoverflow.com/a/17623721/1863564	21:25
	← seds left	21:25
	← Emperor_Earth left	21:25
	← SkarmoutsosV left	21:25
nic-hartley	It takes advantage of : being the label marker and only allowing certain characters after it, so if you put anything ELSE it's effectively a comment. Whereas in POSIXish shells it's a valid command with no effect.	21:26
	→ BlessJah joined	21:28
	→ justanotheruser joined	21:36
	→ jcbitter joined	21:37
	→ hvxgr joined	21:38
	← midnightmagic left	21:38
	→ AtumT joined	21:39
	← Phylock left	21:40
	→ nightshift joined	21:47
	← greatgatsby left	21:50
	← theoceaniscool left	21:51
	→ DrunkZZZ joined	22:07
	← dege left	22:10
	→ causa_sui joined	22:10
	← kjartan left	22:11
osse	aruns: Git just runs the hooks directly by name like executables. You'd need to write a wrapper. Hooks don't need to be scripts either.	22:12
	← fphilipe_ left	22:12
	→ g00s joined	22:14
	← causa_sui left	22:15
	→ Phylock joined	22:16
	→ kjartan joined	22:17
	→ midnightmagic joined	22:24
	← nowhere_man left	22:25
	← gloomy left	22:28
	→ seds joined	22:29
	← plexigras left	22:33
	← Globe_Bloater left	22:34
	← aruns left	22:36
	→ fphilipe_ joined	22:37
	← DrunkZZZ left	22:38
	← shomon left	22:39
nightshift	osse: is there an additional reference for hooks, the book, and man pages seem a bit light, for my ability to understand writing custom actions anyway	22:42
	← seds left	22:43
	→ nurdie joined	22:59
	← random_yanek left	23:00
	← nurdie left	23:04
	← finalbeta left	23:04
	→ igemnace joined	23:05
	← g00s left	23:06
osse	nightshift: don't know of any	23:07
	what do you want to know?	23:07
nightshift	thanks, I was hoping I was just missing something	23:07
	→ acercle joined	23:08
	→ random_yanek joined	23:08
	← random_yanek left	23:08
nightshift	at this point, I don't know exactly what I want to know, just that I didn't see much "help" in what I'd found....	23:08
osse	nightshift: What you want to use them for is entirely up to you. Maybe that's why. There's a limit to what the docs actually can say other than how to run them	23:11
	← fphilipe_ left	23:12
nightshift	Primarily what I'm looking at are 2 possible options (the first is the one I care about the most). I'm writing a static repo viewer that I would, ultimately, like to have a hook execute anytime a repo is pushed to	23:13
	(maybe with some optimizations so that it only runs once every x minutes, in case someone goes on a push frenzy, pushing all their repos one after the other)	23:13
	→ Beam_Inn_ joined	23:14
nightshift	the other one is less likely to actually be put to use, but would involve pushing to a mirror	23:14
	← justanotheruser left	23:15
	← n3wborn left	23:15
	→ random_yanek joined	23:16
	← random_yanek left	23:16
	← royal_screwup21 left	23:17
	→ random_yanek joined	23:18
	← random_yanek left	23:18
	→ Olipro joined	23:24
	→ causa_sui joined	23:25
	← Mr__Anderson left	23:26
	← causa_sui left	23:30
	← duderonomy left	23:31
	→ random_yanek joined	23:35
	← lamont left	23:40
	→ meLon joined	23:40
	← alyptik left	23:41
	← Narrat left	23:43
	← leeN left	23:46
	← Phylock left	23:48
	→ justanotheruser joined	23:54
	← manuelschneid3r left	23:59

Logs Search ←Prev date Next date→

Channels Documentation