IRCloggy #git 2022-04-14

Provider of IRC logs since 2005.

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

	2022-04-14
	← Codaraxis left	00:02
	← causa_sui left	00:02
	→ causa_sui joined	00:04
	← bindu left	00:05
	← ghost43 left	00:05
	→ ghost43 joined	00:05
	→ bindu joined	00:05
	→ lantech19446 joined	00:05
lantech19446	I'm having an issue getting ssh to work with my account when I run ssh -vT git@github everything authenticates properly but when I run git pull I get permission denied (publickey) and so far the documents on github and the stuff i've found on stack overflow haven't helped	00:06
	← causa_sui left	00:09
	→ jjt joined	00:11
	← Jong left	00:11
	← jclsn left	00:15
BtbN	Do you have access to the repo you're trying to clone?	00:15
lantech19446	yes	00:18
	it's my repo and I'm trying to pull right now I set the remote already but even when I specify it I get this error	00:18
	I don't want to delete the existing personal access tokens until I'm sure ssh is functional	00:19
	it says cannot read from remote repository	00:19
	→ causa_sui joined	00:20
	→ jclsn joined	00:20
	→ rgrinberg joined	00:21
	← duxsco left	00:24
lantech19446	do I need to clone it before it'll work?	00:24
	the site's all up and running so I didn't think I would need too	00:24
	← causa_sui left	00:24
BtbN	not sure what you mean, clone it? What else are you trying to do?	00:25
	You need to add your public key to your github account settings, and then make sure ssh uses the corresponding private key	00:26
lantech19446	the repo was already up and running on my computer under https I want it to run under ssh I didn't realize I had to start over	00:27
bookworm	you don't	00:27
BtbN	you'll need to update the remote to use ssh	00:27
bookworm	git remote add if you want to keep the old one, git remote --set-url if you don't	00:27
lantech19446	yea neither fixed the issue I tried them	00:28
	I just destroyed the repo and cloned it again that seems to have fixed whatever it was	00:28
BtbN	Changing the remote url to ssh will sure make it use ssh	00:28
lantech19446	oh it was trying the problem was it couldn't read the remote repository for some reason even though the ssh key was working perfectly	00:29
dzho	getting the form of the ssh URL correct can be a tricky business though, depending on the nature of the remote	00:29
	← Sven_vB left	00:30
	→ mven joined	00:30
BtbN	it	00:31
	it's just github	00:31
lantech19446	so here's another kink in the armor, 2wks ago I got hit by ransomware, this site runs on hyper-v I didn't know it at the time but hyper-v doesn't set an archive bit so you need a special version of the backup software I use to backup vm's anyway ended up with a 3yr old setup and had to do a git pull over it all it worked but clearly it screwed something up in the meantime	00:31
BtbN	Did you forget to turn the initial / to : when converting from https to ssh?	00:31
lantech19446	nope I copied it straight off the site	00:31
dzho	maybe it's that I run old versions of git I don't know but I haven't gotten the URLs copied from github to work, I always have to change them	00:32
lantech19446	I think I'll never know what happened but deleting the local repo and re-cloning has cleared it up	00:32
dzho	the ssh ones at least	00:33
lantech19446	dzho: I mean for all I know that was my problem and since cloning does it for me it did whatever it needed done automatically	00:33
dzho	lantech19446: you might want to take a look then at the remote now as cloned to see what the form of the URL is for future reference	00:33
lantech19446	good idea	00:33
dzho	git remote -v	00:33
	← leftyfb left	00:33
dzho	something like that	00:33
lantech19446	there's not a single thing different about it	00:34
	dzho shrugs	00:34
dzho	oh well	00:34
	← _9pfs left	00:34
lantech19446	yea hey I appreciate your and BtbN's help thanks for being here when I needed you	00:34
	→ leftyfb joined	00:34
	→ twb joined	00:38
twb	So today I wondered "has git switched to an cryptographically-secure hash yet?"	00:39
bookworm	no	00:40
twb	Based on https://git-scm.com/docs/hash-function-transition with "core.objectFormat = sha256" in a NEW repo, I get 64-byte hashes instead of 40-byte hashes.	00:40
	Can I viably switch real repos to SHA-2 yet?	00:40
	I guess an initial easy question is: does github understand SHA-2	00:41
BtbN	tbf, git does not use the hashes for anything that requires cryptographic security	00:42
	so there is really no rush to break half the world and use something newer	00:43
	→ chenqisu1 joined	00:43
twb	maybe, but it's a lot easier for me to prove "SHA-1 is gone" than to prove "SHA-1 is safe, in THIS case"	00:43
	I do not care if my repo cannot be checked out by people running Fedora 17	00:43
BtbN	No idea what you're afraid of. It's not like someone can somehow "hack you" because your repo uses sha1	00:44
	→ causa_sui joined	00:44
twb	I am not asking about that. I'm asking whether SHA-2 works right now.	00:45
	If it doesn't work in current versions of git, then I'll put this on hold for another 2 years.	00:45
	← Strom left	00:45
twb	If it DOES work in current versions THEN we can argue about how much old shit will be unable to use a SHA-2 repo	00:46
	→ Strom joined	00:47
	→ Guest3765 joined	00:48
	← causa_sui left	00:49
twb	It looks like if you have any existing commits, extensions.objectFormat=sha256 "breaks" them, and extensions.compatObjectFormat is not understood as at 2.30.2, so at a minimum you'd have to fast-export and fast-import back into a new repo that has always been sha2	00:49
	← gildasio left	00:49
BtbN	There is zero interoperability, and pretty sure it's still considered experimental	00:50
	so I wouldn't use it for production	00:50
twb	Fair enough	00:50
	← sudoforge left	00:51
	→ sudoforge joined	00:52
twb	FWIW it looks like github doesn't support it yet.	00:54
	When pushing a SHA-2 repo into an empty repo on github, it says "fatal: protocol error: unexpected capabilities^{}"	00:54
mackerman	Most hosting is not at the stage of supporting it, no.	00:55
twb	Okey dokey. /me forgets about this for another couple years	00:56
mackerman	git init --object-format=sha256 # will create a repo with the new object storage	00:58
	Note the man page has experimental warnings.	00:58
BtbN	For git, there is really little to no benefit for a more crypographically secure hash. So any transition with regressiona won't be acceptable. Hence it takes a while.	00:58
	← nyah left	01:00
	→ gildasio joined	01:01
mackerman	https://stackoverflow.com/questions/42433126/how-does-the-newly-found-sha-1-collision-affect-git	01:02
	→ causa_sui joined	01:02
	→ bambanxx joined	01:03
	→ Sulky_ joined	01:05
mackerman	I have yet to see sha1 in git intentionally broken. The fast attacks are probaby in the tens of GPU years scale of effort.	01:05
	→ mud joined	01:07
	← Sulky left	01:07
	→ Dotz0cat- joined	01:08
	← Dotz0cat left	01:08
	Dotz0cat- → Dotz0cat	01:08
	← causa_sui left	01:09
	→ robo- joined	01:13
	← sudoforge left	01:14
	← bertvvvs left	01:14
	← madewokherd left	01:15
	→ sudoforge joined	01:15
	← ThorMojito left	01:17
	→ _9pfs55 joined	01:18
	← srinidhi left	01:20
	→ softirq joined	01:21
softirq	Is there a way to clone a branch such that the description added with --edit-description will be copied to the new branch?	01:23
	→ snake joined	01:24
	→ Jong joined	01:24
	→ causa_sui joined	01:24
snake	hey that was my cover letter "i like graphs"	01:24
	didnt get any calls from that one	01:25
	anyways, for git on windows, is there a way to make the portable exe i have on my thumb drive use the global configuration on my thumb drive or would it just need to be reconfigured on the new PC for my user there?	01:25
	→ R2robot joined	01:28
	← causa_sui left	01:29
	→ vysn joined	01:33
	→ Shentino joined	01:35
	→ mns joined	01:35
	→ causa_sui joined	01:37
	← jjt left	01:41
	← causa_sui left	01:42
	→ causa_sui joined	01:43
	→ kaiseisei joined	01:50
	→ sa0 joined	01:52
	← causa_sui left	01:54
	→ reset joined	01:54
	← junktext_ left	01:56
	→ junktext_ joined	01:58
	← littlebobeep left	02:00
	← makara left	02:04
	← makara1 left	02:05
	→ makara joined	02:05
	→ makara1 joined	02:06
	→ madewokherd joined	02:08
	→ causa_sui joined	02:12
	← jclsn left	02:15
	→ jclsn joined	02:16
	← causa_sui left	02:17
	← justache left	02:18
	→ causa_sui joined	02:20
	→ justache joined	02:20
	← jclsn left	02:22
	← causa_sui left	02:24
	← sa0 left	02:26
	→ jclsn joined	02:27
	→ Jacques joined	02:30
	→ maryo87 joined	02:32
	← Codaraxis_ left	02:34
	← coaxing left	02:34
	→ codedchaos joined	02:35
	← Jacques left	02:37
	← FinnElija left	02:38
	→ finn_elija joined	02:38
	finn_elija → FinnElija	02:38
	← FinnElija left	02:43
	→ FinnElija joined	02:44
	← bambanxx left	02:44
	← terrorjack left	02:47
	→ causa_sui joined	02:47
	→ terrorjack joined	02:48
	→ bluepenquin_ joined	02:52
	→ bambanxx joined	02:53
	← jclsn left	02:53
	← causa_sui left	02:54
	← maryo87 left	02:55
	→ mhugh joined	02:56
	→ furrymcg1e joined	02:58
	→ infogulch5 joined	02:58
	← infogulch left	02:59
	infogulch5 → infogulch	02:59
	→ jclsn joined	03:00
	← furrymcgee left	03:01
	← bambanxx left	03:02
	→ bambanxx joined	03:03
	← softirq left	03:03
	→ w_lehinsun joined	03:07
	→ maryo joined	03:07
	→ Jacques joined	03:08
	→ elastic_1 joined	03:09
	← elastic_dog left	03:10
	elastic_1 → elastic_dog	03:12
	→ Codaraxis joined	03:15
	← cdown left	03:16
	→ bytehackr joined	03:21
	→ littlebobeep joined	03:26
	→ Guest4 joined	03:29
	→ causa_sui joined	03:29
	← Shentino left	03:32
	→ Shentino joined	03:33
	← causa_sui left	03:34
	← Sulky_ left	03:35
	← wonko-the-sane left	03:36
	← _9pfs55 left	03:36
Guest4	HELLO there , i am unable to change my default branch name by this command "git config --global init.defaultBranch <nameOfnewbranch> can anybody help me with that	03:41
	→ Wulf joined	03:44
	← bloody left	03:44
Wulf	Hello. I've got commits A and B. A changes things, B undoes this and adds something else. How can I squash them and keep the meta data of B?	03:45
	→ causa_sui joined	03:46
	← causasui left	03:48
	← Guest6932 left	03:48
	→ k8yun joined	03:50
	→ trollanon joined	03:51
	← Guest4 left	03:53
rewt	Wulf, `git rebase -i A^`, then in the editor remove the first one (A), and change `pick` to `edit` for the 2nd one (B), and you'll probably have to resolve some conflicts (the things B undoes because now A never happened)	03:56
Wulf	rewt: thanks. But it's even better: for some reason there's nothing to resolve. I removed A and magially those changes disappeared from B too.	03:59
	← Jacques left	04:01
	→ Willtech joined	04:08
	← trollanon left	04:10
	← bambanxx left	04:14
	← zebrag left	04:17
	← ChmEarl left	04:19
	→ pridence joined	04:19
	← causa_sui left	04:20
	← pridence left	04:20
	← rgrinberg left	04:24
	→ Jacques joined	04:30
	← bindu left	04:31
	→ causa_sui joined	04:31
	→ bindu joined	04:31
	← dob1 left	04:31
	→ dob1 joined	04:32
	← causa_sui left	04:39
	→ u0_a147 joined	04:39
	← junktext_ left	04:41
	← u0_a147 left	04:41
	→ u0_a147 joined	04:42
	← mhugh left	04:42
	← carl-- left	04:45
	← u0_a147 left	04:46
	→ rektide joined	04:46
	→ u0_a147 joined	04:46
	← u0_a147 left	04:46
	← gast0n left	04:52
	← howdoi left	04:57
hpfr	I have a repo I just aggressively gc'd and `git rev-list --objects --all \| wc -l` is 810. `fd \| wc -l` (which ignores `.git`) outputs 201. I have one branch and two commits. what forms the remaining ~600 objects?	04:57
	I tried git show on some of the outputs of the rev list command and I get some trees that have no output from `git ls-tree`?	04:59
	Guest3765 → blahboybaz	05:00
	→ dlwomw^ joined	05:01
	← bluepenquin_ left	05:02
	← geosmile left	05:03
	→ u0_a147 joined	05:06
	← u0_a147 left	05:07
	→ madewokherd` joined	05:07
	→ u0_a147 joined	05:08
	← u0_a147 left	05:08
	→ u0_a147 joined	05:08
	← madewokherd left	05:09
	← u0_a147 left	05:10
	→ u0_a147 joined	05:11
	→ rgrinberg joined	05:12
	← u0_a147 left	05:12
	→ u0_a147 joined	05:12
	← thiago left	05:13
	→ _9pfs9 joined	05:14
	← u0_a147 left	05:15
	→ u0_a147 joined	05:15
	← u0_a147 left	05:17
	← infogulch left	05:23
	← tyson2 left	05:30
	→ causa_sui joined	05:34
	bkircher_afk → bkircher	05:38
	← mud left	05:38
	← causa_sui left	05:39
	→ causa_sui joined	05:40
hpfr	well in order to debug that, I tried removing a worktree and got `cannot remove working tree: <worktree-path>/.git is not a .git file, error code 2`. the worktree's .git is a symlink created by `git worktree add`, so it should work, no?	05:42
	← maryo left	05:43
	← great_taste left	05:43
	← causa_sui left	05:44
	← Misotauros left	05:45
	→ causa_sui joined	05:49
	→ srinidhi joined	05:51
	← Willtech left	05:52
	← k8yun left	05:52
	→ jianwu joined	05:52
	← rgrinberg left	05:52
	← sudoforge left	05:52
	← causa_sui left	05:53
	→ mexen joined	05:53
	→ causa_sui joined	05:55
	← causa_sui left	06:04
	→ gurkenglas joined	06:05
	→ carl-- joined	06:05
	← RiFo left	06:06
	→ RiFo joined	06:07
	← piwu left	06:07
	→ jbg joined	06:07
	→ piwu joined	06:07
	→ xkoncek joined	06:08
	→ samuelbernardo joined	06:09
	→ feriman joined	06:09
	→ bluepenquin_ joined	06:12
	→ roadie joined	06:17
	→ dionysus69 joined	06:20
	← Vonter left	06:21
	→ Guest77 joined	06:25
	← m5zs7k left	06:30
	→ m5zs7k joined	06:30
	← _9pfs9 left	06:31
	← m1dnight left	06:33
	← w0ng left	06:37
	→ carl--- joined	06:39
	→ m1dnight joined	06:39
	← carl-- left	06:41
	→ EvilDMP joined	06:41
	→ kenanmarasli joined	06:43
	← dionysus69 left	06:47
	← Shentino left	06:50
	→ Shentino joined	06:51
	← feriman left	06:51
	← jianwu left	06:52
	→ rfuentess joined	06:53
	← carl--- left	06:55
	→ carl- joined	06:55
	← carl- left	06:57
	→ feriman joined	06:58
	← Gustavo6046 left	06:59
	→ Gustavo6046 joined	07:00
	→ iomari891 joined	07:00
	→ Misotauros joined	07:02
	→ arch_angel joined	07:10
	→ \\ joined	07:11
	→ palasso joined	07:12
	← iomari891 left	07:12
	→ pretty_dumm_guy joined	07:14
	← pretty_dumm_guy left	07:14
	→ iomari891 joined	07:15
	← \\ left	07:16
	→ \\ joined	07:16
	→ pretty_dumm_guy joined	07:17
	← tmz left	07:19
	← Millicent left	07:19
	← aindilis_ left	07:19
	→ ___nick___ joined	07:19
	→ Millicent joined	07:19
	← fjmorazan left	07:19
	→ tmz joined	07:20
	← blahboybaz left	07:20
	→ fjmorazan joined	07:21
	→ aindilis joined	07:23
	→ tinhear joined	07:29
	→ coot joined	07:30
	← dlwomw^ left	07:33
	tinhear FREENODE AND LIBERA MOVE TO irc.worldhacker.org is open for public #0day , [ 0day (xc) Our ] WhiteHat Hacker Team And Dev Official Public Channel : https://worldhacker.org , REQUEST PROGRAM , SOURCE CODE , OR ANYTHING HERE ... . NEED STORAGE OR VPS ??? Free	07:34
	ChanServ set mode: +o	07:35
	ikke kicked tinhear (tinhear)	07:35
	→ TheCatBehemoth joined	07:38
	← Jacques left	07:39
	← Thanatermesis left	07:40
	→ igemnace joined	07:42
	→ duxsco joined	07:45
	→ sa0 joined	07:46
	→ Sulky joined	07:46
	← user24037 left	07:49
	→ user24037 joined	07:53
	← user24037 left	07:53
	→ user24037 joined	07:53
	→ vimal joined	07:54
	← jimklimov1 left	07:55
	← hololeap left	07:56
	→ hololeap joined	07:56
	→ dionysus69 joined	07:57
	→ causa_sui joined	07:59
	← ajfriesen1 left	08:00
	← igemnace left	08:00
	→ vitali64 joined	08:03
	→ ajfriesen1 joined	08:03
	← causa_sui left	08:04
	→ Vonter joined	08:05
	→ Haohmaru joined	08:06
	← Jong left	08:09
	← ajfriesen1 left	08:11
	→ ajfriesen1 joined	08:14
	→ [sr]\|2 joined	08:14
	→ austin_ joined	08:14
	← austin987 left	08:14
	← leshaste left	08:15
	← Techcable left	08:15
	→ leshaste joined	08:15
	→ Techcable joined	08:15
	→ awmv joined	08:15
	← euidzero left	08:16
	→ euidzero joined	08:16
	← sa0 left	08:16
	→ analpaper_ joined	08:17
	→ Roy_Mustang joined	08:18
	← A_Dragon left	08:18
	← roadie left	08:19
	Roy_Mustang → A_Dragon	08:19
	→ toulene8 joined	08:19
	→ dskullz joined	08:19
	→ haskl[error] joined	08:20
	→ constxo joined	08:22
	← analpaper__ left	08:22
	← [sr] left	08:22
	← toulene left	08:22
	← dskull left	08:22
	← haskl left	08:22
	← constxd left	08:22
	← realtime-neil left	08:22
	dskullz → dskull	08:22
	toulene8 → toulene	08:22
	→ realtime-neil joined	08:23
	→ causa_sui joined	08:24
	→ Jong joined	08:24
	← Gustavo6046 left	08:25
	← causa_sui left	08:28
	→ causa_sui joined	08:29
	→ rsx joined	08:32
	← michele left	08:32
	← emanuele6 left	08:33
	← causa_sui left	08:33
	→ cousteau_ joined	08:35
	cousteau_ → cousteau`	08:36
	→ coldmountain joined	08:36
	← malteger left	08:37
	→ cousteau joined	08:39
cousteau	Hi! (I'm on a mobile and very unstable connection, so my apologies in advance if I suddenly disconnect)	08:39
	← fflam left	08:40
cousteau	Question: To rebase, or not to rebase?	08:40
	← cousteau` left	08:40
cousteau	(there goes my ghost!)	08:41
selckin	if you can rebasing branches before merging makes life easy, so you need a reason to not do it, instead of one to do it	08:41
	→ malteger joined	08:41
	→ igemnace joined	08:42
	← ajfriesen1 left	08:42
	→ cousteau_ joined	08:44
	cousteau_ has seen 1 message from selckin before disconnecting	08:44
	→ roadie joined	08:45
	← cousteau left	08:45
cousteau_	Personally I think it looks neat in the git history to have the actual snapshots of code you had while you were working, and could be useful as a map to see how development worked, and for debugging / finding regressions	08:45
selckin	no all worthless info	08:46
	→ ajfriesen1 joined	08:46
cousteau_	Whereas rebasing means "making up fake snapshots of code that never actually existed", so maybe you're creating commits that don't actually work	08:46
bookworm	commits can help the reviewer, which is kinda good	08:47
	and chances are you weren't on the best track from the very beginning	08:47
cousteau_	(not sure what the policy usually is, but I imagine that "all commits must work, at least partially" is a common policy)	08:48
bookworm	throw CI at it and test that if you want	08:48
	might be expensive to run though	08:48
	and if you have fixup commits in between that got reverted/rewriten down the line your code doesn't work either	08:49
cousteau_	selckin, bookworm: wait, are we talking about collapsing all commits into one, or cherry-picking them individually on top of origin/master?	08:49
selckin	you want small logical commits that change single logical thing and cmopile and all tests pass, a branch is a bunch of these, you keep it rebased on top of master, till its merged	08:49
	and you should be free to edit the commits on the branch to those	08:49
cousteau_	I was thinking more on the latter, but I now realize you might have been thinking about collapsing the feature branch into a single commit	08:50
bookworm	no	08:50
	not at all	08:50
	that's refered to a s a squash merge	08:50
cousteau_	That was the word, "squash"	08:50
	← bluepenquin_ left	08:50
	← roadie left	08:51
	→ fflam joined	08:52
	← vdamewood left	08:52
	→ txtsd joined	08:53
	→ [Tux] joined	08:53
[Tux]	Good-day from NL	08:53
	→ nyah joined	08:54
[Tux]	Testing 2.35.1 hangs in t7528-signed-commit-ssh.sh and it killed my ssh-agent	08:54
	reproducable on openSUSE Tumbleweed and Leap 15.3	08:54
cousteau_	selckin: do you mean "on top of your local heads/master", or "on top of remotes/origin/master, which you should be periodically merging into your heads/master and then rebasing your feature branch"?	08:54
selckin	those are the same thing	08:55
cousteau_	...not if you don't periodically pull master	08:55
[Tux]	when I restart ssh-agent, the test finishes (and PASSes)	08:56
selckin	well if you don't eat you die	08:56
	← austin_ left	08:58
cousteau_	The flow I had in mind was: (1) pull master, (2) create feature branch and append commits to it, (3) pull master again so that it is up to date, and (4) merge feature branch into master	08:58
	furrymcg1e → furrymcgee	08:59
	→ austin987 joined	08:59
	→ causa_sui joined	08:59
selckin	4 rebase on master, 5 resolve any conflicts 6 test everything again 7 merge	08:59
cousteau_	"Everything" being "the last commit", or "every single commit"?	09:00
	Also, 4: with squash or as individual commits?	09:00
selckin	i'd do last commit	09:00
	→ vdamewood joined	09:01
bookworm	cousteau_: generally if people just say merge, they do mean a merge A proper one, the thing that generates a merge commit	09:01
selckin	you need a clean commit history, every commit should be a logical change, so if you have that don't squash, some ppl are stupid and commit every 5min and every little experimetan and wrong code they tried and just before lunch, and before thye left the office, so if htose commits are in there squash em into the void	09:01
cousteau_	OK that makes sense. But it leaves the risk of having non working commits, which could be problematic for git bipart	09:02
bookworm	bisect*	09:02
cousteau_	(but probably ONLY for git bipart, so I don't know how much of a problem that would be)	09:02
bookworm	bisect*	09:02
cousteau_	Er that, bisect	09:02
	Er that, bisect	09:02
	(sorry, I stop reading other people's messages when typing)	09:03
	→ enkeyz joined	09:03
	← causa_sui left	09:04
cousteau_	selckin: I've seen people do that in Wikipedia and it gets on my nerves because Wikipedia doesn't allow you to "squash commits"	09:04
	bookworm: so "merge the changes", not "merge the commits" or "merge the branches", right?	09:06
bookworm	it's all the same	09:06
	← vimal left	09:07
	bkircher → bkircher_afk	09:07
bookworm	a branch is just a pointer to a commit, merge will merge whatever commits are reachable from the thing you merge that are not yet in the branch you merge into (or rather, take the two trees end result and merge that, the commits get pulled in by virtue of being a graph)	09:08
	→ Willtech joined	09:10
cousteau_	or "merging the... graphs?" Maybe I'm confusing "merge" with "merge two trees, and then create a commit with two parents pointing to the commits that you merged, containing the resulting tree"	09:12
	ie, maybe conceptually "merge" actually means just "combine codes / worktrees", and not "do what the `git merge` command does", is that what you mean?	09:14
bookworm	eh? You seem confused. The two statements don't contradict one another?	09:15
cousteau_	ie, "merge the content of commits", not "merge two paths as seen in the commit graph". I'm too used to seeing git as a graph, and only a graph.	09:15
	→ roadie joined	09:18
cousteau_	One refers to the content of the commits (the code), the other refers to the graph (the commit history and parent relationships). They're not mutually exclusive, but they're not the same either.	09:18
	(...to clarify, I didn't want to imply that I'm not confused. I probably am.)	09:20
bookworm	a commit points to a tree	09:21
	← Guest77 left	09:21
	← roadie left	09:23
	→ causa_sui joined	09:24
cousteau_	...also I think I misread "they do mean a merge, a proper one" as "they don't mean a merge", which contributed to the confusion	09:25
bookworm	have you read https://git-scm.com/book/en/v2/Git-Internals-Git-Objects ? Maybe that helps?	09:25
	← igemnace left	09:26
	→ lgc joined	09:26
	← skapata left	09:26
tsdh	Is there a reason "git diff" shows symlinks as "+target-file" (which makes sense) but also with "\ new newline at end of file"?	09:27
	← cousteau_ left	09:29
	← lgc left	09:30
	→ han-solo joined	09:31
	→ drogas joined	09:33
	← causa_sui left	09:33
	→ roadie joined	09:37
	→ jimklimov joined	09:39
	← kaiseisei left	09:42
	← roadie left	09:42
	→ roadie joined	09:42
	← AbleBacon left	09:54
	→ cousteau joined	09:56
cousteau	This connection is really poor... ok, thanks for the help/info, bookworm and selckin, bye!	09:57
	← cousteau left	09:57
	← OnkelTem left	09:59
	← rfuentess left	10:00
	→ OnkelTem joined	10:02
	→ sa0 joined	10:03
	← vishal left	10:08
	→ sandipan joined	10:08
	← bytehackr left	10:09
	→ vishal joined	10:09
	← awmv left	10:10
	→ Tobbi_ joined	10:15
	→ jazzy2 joined	10:15
	← jazzy left	10:18
	→ nwoob joined	10:22
nwoob	How do i commit from forked repo to original repo directly	10:22
	commit and push	10:22
	→ causa_sui joined	10:24
selckin	!remote	10:24
gitinfo	[!remote_commands] The only git porcelain commands which might interact with remote repos are: fetch, pull, push, remote, ls-remote, archive, submodule (plus plumbing, remote helpers, etc)	10:24
	← iomari891 left	10:25
selckin	you can add multiple remotes in the same local repo, and then push/fetch to the one you specify	10:25
	→ awmv joined	10:29
	← causa_sui left	10:33
	→ causa_sui joined	10:34
	← drogas left	10:35
nwoob	selckin: so I have to add original repo as another remote and then push by specifying on which remote's branch I want to push	10:35
selckin	git remote add original https://somewhere; git push original master	10:36
	etc	10:36
	→ Guest82 joined	10:39
	← causa_sui left	10:39
	← Guest82 left	10:39
	→ causa_sui joined	10:39
nwoob	and this will push to the repo from which I have forked my repo	10:39
bookworm	well, if you set the new remote to be the origin	10:42
	if you cloned, your remote "origin" is set to the fork generally, not "upstream"	10:42
	→ coaxing joined	10:43
	← [twisti] left	10:43
bookworm	chances are that if you had to "fork" you aren't in control of the upstream repo, else you wouldn't have had to fork in the first place	10:43
	so you can't directly write to it either, you don't have write permissions (generally)	10:44
nwoob	but if I want to push to upstream from my fork	10:44
bookworm	then you'd request the upstream to pull your changes	10:44
nwoob	then I can do git remote add upstream url and then git push upstream master	10:44
	← zulutango left	10:44
bookworm	in GH speak that's a pull request, Gitlab it's merge request, gerrit it's change list or whatever, git send-email, you get the idea	10:44
nwoob	I don;t want to create PR. I want to write script	10:45
bookworm	do you have write access to upstream?	10:45
nwoob	that will push to upstream from forked repo	10:45
	yes I have write access	10:45
bookworm	what do you consider a "forked repo" ?	10:45
	generally, forking means you take code and push it to some other place, which diverges from upstreams code	10:46
nwoob	yes	10:46
	I understand that	10:46
	I forked a repo and now it;s in my userspace	10:46
bookworm	but sure, just `git remote add upstream $url` && `git push upstream $branch`	10:47
	the remote only needs to be added once, naturally	10:47
nwoob	now url is https://me_user/project	10:47
	and original repo url will be https://project	10:47
	understood	10:48
	→ bluepenquin_ joined	10:49
	← causa_sui left	10:49
	← sa0 left	10:49
	→ sa0 joined	10:49
	← vitali64 left	10:49
	→ zulutango joined	10:50
	← nwoob left	10:53
	→ vitali64 joined	10:53
	← BlessJah left	10:56
	← konrados left	10:58
	→ konrados joined	10:58
	→ BlessJah joined	10:59
	← zer0bitz left	11:03
	← DeFreeZe left	11:06
	← jazzy2 left	11:07
	← awmv left	11:12
	→ cdown joined	11:13
	→ structoha1 joined	11:15
	← gildasio left	11:15
	→ gildasio joined	11:16
	→ kaiseisei joined	11:20
	→ bertvvvs joined	11:24
	→ a1eaiactaest joined	11:25
	← XV8 left	11:26
	→ XV8 joined	11:26
	← bindu left	11:27
	→ bindu joined	11:29
	← structoha1 left	11:30
	→ geosmile joined	11:31
	← Misotauros left	11:34
	← littlebobeep left	11:35
	→ littlebobeep joined	11:35
	← kaiseisei left	11:36
	→ olle joined	11:42
olle	Hmmmmm, for me, --word-diff=color is working for some changes but not others; for some changes, it still shows the entire line, or at least more than what was changed. Normal behaviour?	11:43
	→ Yorick_ joined	11:43
	→ ws2k3 joined	11:43
	→ _rgn joined	11:45
_rgn	is there a better way to alter the latest commit by picking hunks manually?	11:45
	git reset HEAD~ && git add --patch && git commit --reuse-message='@{1}' && git reset --hard	11:45
	← pulse left	11:46
	← moldorcoder7 left	11:48
	← tealtabbycat left	11:49
	← chenqisu1 left	11:50
	→ _sa0sin_ joined	11:53
	→ causa_sui joined	11:54
	← hololeap left	11:54
	← elastic_dog left	11:56
	→ hololeap joined	11:56
	→ epolanski joined	11:56
	← sa0 left	11:56
	← causa_sui left	11:59
	← Murr left	11:59
	→ Murr joined	11:59
	→ causa_sui joined	11:59
	← FinnElija left	12:00
	← _sa0sin_ left	12:00
	→ FinnElija joined	12:00
	→ pulse joined	12:01
	→ elastic_dog joined	12:01
	← causa_sui left	12:04
	→ sa0 joined	12:07
	→ chenqisu1 joined	12:08
	→ rafasc joined	12:09
	← rafasc left	12:10
	→ rafasc joined	12:10
	← konrados left	12:12
	→ konrados joined	12:12
	→ SteveR joined	12:15
	→ saroy\|afk joined	12:15
SteveR	Hi there, i get this error, https://dpaste.org/YEkiP	12:15
	any idea on how to proceed further?	12:15
bremner	the message tells you?	12:16
ikke	It's not an error, just a warning	12:16
	'You are in 'detached HEAD' state. '	12:16
	← sandipan left	12:18
	→ sandipan joined	12:19
	← a1eaiactaest left	12:20
	→ causa_sui joined	12:20
	→ gh34 joined	12:21
	← sandipan left	12:21
	→ sandipan joined	12:22
	← saroy\|afk left	12:22
	← sandipan left	12:24
	→ cdown_ joined	12:24
	→ sandipan joined	12:26
	← cdown left	12:27
	→ meator joined	12:27
	cdown_ → cdown	12:27
meator	Hello. How can I ignore uncommited changes in git blame?	12:28
	← SteveR left	12:29
	→ faceface joined	12:30
faceface	hello	12:30
	when I say `git branch` I see, "* (no branch, rebasing master)", but I'm not sure where I'm stuck...	12:30
	→ wolfdale8 joined	12:30
faceface	git am --show shows, public/stylesheets/report.css changed, but I don't see a 'both modified' warning anywhere	12:31
	← gurkenglas left	12:33
	→ bloody joined	12:36
	→ gurkenglas joined	12:36
	→ tyson2 joined	12:40
	→ Milos joined	12:42
	← roadie left	12:48
	→ roadie joined	12:48
	← Tobbi_ left	12:49
osse	faceface: what about git status?	12:51
	→ mjh4386 joined	12:52
	→ Tobbi_ joined	12:53
	→ wakeup joined	12:56
	→ iomari891 joined	12:57
	← wakeup left	12:57
	→ saroy\|afk joined	12:58
	← chenqisu1 left	13:00
	← sandipan left	13:00
	→ saroy joined	13:00
	→ sudoforge joined	13:02
	← coot left	13:02
	← saroy\|afk left	13:03
	→ wakeup joined	13:05
wakeup	if I does "git checkout master" and then "git merge other-branch"	13:05
	which branch is merged into what?	13:05
	what are the source and destination branches?	13:05
jast	the currently checked out branch is always the destination	13:06
wakeup	so that would merge (i.e. add) the commits of other-branch into master?	13:06
jast	adding to a branch, be it a normal merge or a commit, always requires it being checked out, unless you use black magic	13:06
	yeah	13:06
	*normal commit or a merge	13:06
_rgn	git has very few commands, if any, that change a branch you're not on	13:07
wakeup	git on github:P	13:09
	https://github.com/git/git	13:09
	git versioned by git	13:09
	← olle left	13:09
jast	well... yeah :)	13:10
wakeup	https://github.com/git/git/blob/master/mergesort.c :D	13:11
	→ wonko-the-sane joined	13:12
	→ igemnace joined	13:13
	← analoq left	13:13
	→ analoq joined	13:15
	→ cdown_ joined	13:19
	← wonko-the-sane left	13:20
	← cdown left	13:22
	cdown_ → cdown	13:22
	→ wonko-the-sane joined	13:25
	→ Rashad joined	13:31
	→ coot joined	13:35
	→ DeFreeZe joined	13:43
	← han-solo left	13:44
	← vdamewood left	13:46
	→ mgd joined	13:46
	← bluepenquin_ left	13:48
	→ lejocelyn joined	13:53
	← mgd left	13:56
	→ iffraff joined	13:58
wakeup	It seems that rebase can be used to solve similar problems of those solved by merge, but merge doesn't destroy the history and creates an additional commit automatically. It seems that you would want to use rebase only locally, because, locally, you can destroy, in a way, the history, because only you knows it. Now, I am still not grasping the idea	14:06
	of rebasing. I understand that when we merge we basically pick the changes from one branch and insert them in the current branch. But when we rebase, what exactly happens? I read that commits are somehow rewritten (i.e. the history is rewritten), but I don't understand how this is related to 2 branches. Maybe I have this doubt because I used rebase	14:06
	(with the interactive option) to squash multiple commits into one, and that's the only time I remember having used rebase	14:06
	← konrados left	14:10
	→ konrados joined	14:10
wakeup	so, in that occasion, I used rebase and only one branch was involved	14:10
mackerman	Yes, rebase rewrites history. Sometimes rewinding to a point in time, and applying the same patch on top of some other tree.	14:11
	Rebase is not required, you don't need to use it if you do not want to.	14:11
	→ thiago joined	14:12
	← pranshukhandal left	14:12
	→ pranshukhandal joined	14:13
igemnace	wakeup: rebase and rebase -i do quite different things, but both have to do with rewriting commits (immutably, so it's more of an "edit-then-write-as-new-commit" operation)	14:17
	← snake left	14:18
igemnace	wakeup: rebase, imagine two branches. A has C -> A1 -> A2 -> A3, B has C -> B1 -> B2 -> B3 where C is a common base commit for the two. checked out on A, doing git merge B would get you C -> A1 -> A2 -> A3 -> M where merge has two parents: A3 and B3	14:19
jast	!rebase	14:20
gitinfo	'git rebase' takes away your local commits, updates your branch with new stuff from <upstream> (argument), then re-applies your local commits on top. This makes it look like your commits were created "after" the new stuff, and it can look cleaner than doing a !merge. Beware of !rewriting_public_history, though. Not to be confused with !interactive_rebase.	14:20
igemnace	wakeup: doing git rebase B instead of git merge B would get you C -> B1 -> B2 -> B3 -> A1' -> A2' -> A3'. A1, A2, and A3 are all rewritten into new commits (the primed versions, A1' and so on)	14:20
	← wakeup left	14:22
	→ wakeup joined	14:23
	← gurkenglas left	14:24
	← GNUmoon left	14:24
	→ GNUmoon joined	14:24
	← feriman left	14:29
wakeup	igemnace wait, in the example of rebase that you've just given, you do git rebase B after you've checkout A, i.e. you're in branch A, right?	14:30
igemnace	yes	14:30
	→ zebrag joined	14:32
faceface	hi osse	14:33
wakeup	but why are the commits of B placed before those of A, if we're in A?	14:33
faceface	Here is the error and git status: https://gist.github.com/CholoTook/fb881b043049556f1969bbc2fe4cfb73	14:34
wakeup	anyway, it seems that rebase just inserts the commits that you MANUALLY made in the branch in the other branch, while merge just AUTOMATICALLY creates this merge commit that represents really a merge of all the changes from the other branch	14:34
	← Shentino left	14:34
faceface	So I don't know what to do ...	14:35
	→ Shentino joined	14:37
	robo- → roboirc	14:37
faceface	OK, hitting git rebase --continue tells me what to do	14:37
igemnace	wakeup: B's pointing at B3, and so git rebase B says "make B3 the new base of A1, A2, and A3"	14:38
	← XV8 left	14:38
	← DeFreeZe left	14:39
	→ XV8 joined	14:39
	→ DeFreeZe joined	14:39
igemnace	wakeup: well, yes, in a loosey-goosey way (manual and automatic won't be the first words i use to describe the difference). merge turns two branches into one by creating one commit with two parents, rebase turns two branches into one by rewriting some commit's parents such that they never branched in the first place	14:41
	→ vodkra joined	14:41
	← BlessJah left	14:51
	→ BlessJah joined	14:52
wakeup	alright, I guess that is clearer now	14:52
	I hope not to forget	14:52
	anyway, git rebase and git rebase interactive are different	14:53
	why?	14:53
	why did u say that	14:53
	I mean, of course, one is "interactive"	14:53
	but it seemed that you implied that one is fundamentallly different from the other	14:53
mackerman	rebase is scripted and does a defined thing. rebase --interactive lets you do arbitrary things.	14:57
	To understand exactly what you did with -i we would want the entire todo list	14:58
	→ rgrinberg joined	14:58
	→ _9pfs joined	14:59
wakeup	I wanted to change the last commit to include a new file	15:02
	so I did	15:02
	git add my_file	15:02
	git commit --amend	15:02
	didn't change the file that was opened to change the commit message	15:02
	then git push	15:02
	→ madewokherd joined	15:03
wakeup	but it says : https://pastebin.com/D8JibhAb	15:03
	but actually before that I did git pull and it was up-todayte	15:03
	→ cdown_ joined	15:05
igemnace	wakeup: git rebase -i is more for rewriting your own local commits in	15:05
mackerman	Did you push the unmodified commit before it was rebased?	15:06
igemnace	whoops. pressed ctrl-j accidentally. that was "...in different ways"	15:06
wakeup	now I did git pull and it says "Pulling without specifying how to reconcile divergent branches is discouraged."	15:06
	← madewokherd` left	15:06
wakeup	as I said, I did 1. git pull (up-to-date), 2. git add my_file, 3. git commit --amend, 4. git push (gives the error in the pastebin)	15:06
igemnace	wakeup: that error message usually indicates that you amended a commit that you already pushed previously	15:07
wakeup	5. git pull (gives the hint "Pulling without specifying how to reconcile divergent branches is discouraged" and opens a file)	15:07
	← cdown left	15:07
	cdown_ → cdown	15:07
wakeup	yes, I already pushed it indeed	15:07
	so what I should do it	15:07
	I want to amend the already pushed commit	15:07
igemnace	!rewrite	15:08
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/waqum	15:08
igemnace	how many people work on that repo, and on that branch?	15:08
wakeup	I don't think anyone pulled anything	15:08
	this is a personal project	15:08
	→ great_taste joined	15:08
wakeup	there are some forks, but they are old	15:08
igemnace	alright. just push -f	15:08
wakeup	but what do I do with the file that opened after that last git pull? just close the file?	15:09
	the file says "Merge branch 'master' of https://github.com/..."	15:09
	so it seems that it's going to merge anything	15:09
igemnace	yeah but you'd want to cancel the pull (that'd likely merge both amends). what editor is it open in, vim?	15:09
wakeup	I don't want to merge anything	15:09
	ThiagoCMartins[m → TMartins	15:10
wakeup	visual studio code... I changed the default editor	15:10
	cuz I am more used to code	15:10
igemnace	i don't know if there's a way to exit vscode with a non-zero exit code. just exit the editor, and we'll reset the merge commit	15:10
wakeup	ok	15:11
	→ nwoob joined	15:11
wakeup	so we should undo the merge?	15:11
igemnace	once the pull and merge resolves, i'd git reset --hard @^ (to reset back to before the merge) then git push -f	15:11
	verify before and after every command with git log, to make sure	15:12
	→ ChmEarl joined	15:13
	← jinsun left	15:14
wakeup	yeah, that seems to have worked	15:15
	what does @^ mean?	15:15
	exactly	15:15
	in fact, in the past, I think I used git pull -f	15:16
igemnace	a very terse way of saying "the (first) parent of the current commit". see git help revisions	15:16
wakeup	so I forgot -f there	15:16
	and that cuased this	15:16
	→ beuys joined	15:24
	→ Church joined	15:24
beuys	The recent git update borked my workflow :/ Investigating why ...	15:24
Church	Heh same here.	15:24
	→ jinsun joined	15:25
	← sgen left	15:25
	← coldmountain left	15:26
	→ a1eaiactaest joined	15:26
beuys	I see why. root cannot do "git status /path/repo" anymore if /path/repo does not have the owner root.	15:26
	That seems pretty insane. What to do about it?	15:27
bremner	uh. you'll need to tell us more about your setup.	15:27
beuys	bremner: Setup is that I create a list of repos with changes.	15:27
	→ Jacques joined	15:28
beuys	The script checks a bunch of repos and outputs a list of those that have changes.	15:28
bremner	git status /path/repo sounds wrong, for starters.	15:28
	← bindu left	15:28
beuys	Wow, it does not even work if /path/repo is owned by root.	15:28
igemnace	beuys: the drop-in configuration is https://git-scm.com/docs/git-config/2.35.2#Documentation/git-config.txt-safedirectory	15:28
bremner	most git commands operate on a "current repo"	15:29
igemnace	beuys: it was likely this change that borked your setup: https://github.blog/2022-04-12-git-security-vulnerability-announced/	15:29
	→ bindu joined	15:29
beuys	igemnace: YEs	15:29
Church	Is there a wildcard setting for safe.directory? Trying to fix a half broken phabricator install now.	15:29
	And I'd rather not manually do that for ever repo...	15:29
beuys	I don't want to define repos as "safe" at all. I just want to be able to "git status" them.	15:30
	Oh, the problem seems to be different than I thought.	15:31
	← meator left	15:31
	← wakeup left	15:31
beuys	Let me dive in deeper ...	15:31
	Yeah, it is even worse.	15:31
	← _9pfs left	15:31
beuys	root can not do "git status" if the owner of the dir that holds the repo is not root.	15:32
	Insane.	15:32
	→ Gustavo6046 joined	15:33
	← Gustavo6046 left	15:34
	→ Gustavo6046 joined	15:34
	→ Thanatermesis joined	15:36
	pranshukhandal → ghost1999	15:37
	→ softirq joined	15:37
	← Jacques left	15:38
	← bertvvvs left	15:38
	← Tobbi_ left	15:39
beuys	They should have just disabled the "execute code from repos" by default and have a setting for it.	15:41
	→ gurkenglas joined	15:43
beuys	By the logic that is imposed now, many tools would have to disallow input from dirs not owned by root.	15:43
	"source /some/script" -> Sorry, you have to mark "/some/" as a safe dir globally to do that :/	15:43
	→ bertvvvs joined	15:44
	→ mat001__ joined	15:45
	← mat001_ left	15:46
	← mat001__ left	15:46
	→ mat001_ joined	15:47
	← great_taste left	15:48
	← softirq left	15:52
	← beuys left	15:52
	← arch_angel left	15:54
	→ softirq joined	15:54
	← softirq left	15:55
Church	Yeah I've just taken to downgrading my git binary for now till I find a happy path.	15:56
	→ great_taste joined	15:56
	← Murr left	15:58
jast	Church: good news: https://github.com/git/git/blob/1ac7422e39b0043250b026f9988d0da24cb2cb58/Documentation/RelNotes/2.30.4.txt#L11	15:58
	→ Murr joined	15:59
	→ FFY00_ joined	15:59
jast	too bad beuys left, might have been useful for them too	15:59
Church	Rad	16:00
	I'll wait for that to make it out and then upgrade	16:00
jast	I think it's been out for a while	16:00
	← nwoob left	16:00
jast	hm, no, it was added yesterday on master	16:00
	unfortunately this will likely not make it into many distro packages because distros like to push only security fixes into updates, excluding quality-of-life improvements like this	16:02
	← FFY00 left	16:02
Church	Yeah probably, here's hoping Ubuntu pushes this.	16:03
jast	fingers crossed :)	16:03
	← pretty_dumm_guy left	16:03
jast	this whole change completely blindsided us at work, our testing machines update nightly and suddenly our rebuilds didn't work anymore	16:03
	→ bambanxx joined	16:04
	← bambanxx left	16:04
	→ bambanxx joined	16:05
	→ pretty_dumm_guy joined	16:05
	← rewt left	16:07
	→ kix joined	16:14
	← ___nick___ left	16:16
	→ ___nick___ joined	16:17
	← ___nick___ left	16:18
	← gildasio left	16:19
	→ ___nick___ joined	16:20
	→ oncall-pokemon joined	16:21
	→ BeatButton joined	16:22
	→ pretty_d1 joined	16:22
	← lejocelyn left	16:22
	← pretty_d1 left	16:22
	← ghost1999 left	16:22
	← redoc left	16:23
	← pretty_dumm_guy left	16:24
	← Haohmaru left	16:24
nedbat	am I misreading this release note? It seems self-referential: https://github.com/git/git/blob/1ac7422e39b0043250b026f9988d0da24cb2cb58/Documentation/RelNotes/2.35.3.txt	16:25
	← rafasc left	16:25
	→ rewt joined	16:26
	→ rafasc joined	16:26
igemnace	nedbat: haha. well, it's not false	16:26
	← BeatButton left	16:26
	← bertvvvs left	16:27
	← mat001_ left	16:27
	← rafasc left	16:27
	→ rafasc joined	16:27
nedbat	igemnace: i'm not sure if there's anything I can do to draw attention to it. I can't comment on it.	16:27
	← rgrinberg left	16:28
nedbat	igemnace: and all the release notes in that commit are like that: https://github.com/git/git/commit/d516b2db0af2221bd6b13e7347abdcb5830b2829	16:28
	→ nwoob joined	16:30
igemnace	hmm, good point. perhaps the mailing list?	16:30
nedbat	I don't read the mailing list. Maybe someone here who is more embedded in git development could do it?	16:31
	← tyson2 left	16:32
	→ mat001_ joined	16:32
	→ rgrinberg joined	16:36
	← mat001_ left	16:36
	→ mat001_ joined	16:37
	← JanC left	16:38
	→ JanC joined	16:38
	→ redoc joined	16:41
	→ orbyt joined	16:42
	← kix left	16:44
	→ lionkor joined	16:45
	→ Guest8629 joined	16:46
phryk	how do i add just a single line (NOT a hunk)?	16:49
	← Lcawte left	16:51
	→ Lcawte joined	16:51
igemnace	phryk: i don't remember if you can split it all the way down to single-line hunks, but try that. you could also edit the index by hand, that's just error-prone. i tend to rely on editor integration for such a task	16:52
	← mexen left	16:52
	→ tyson2 joined	16:52
	→ feriman joined	16:52
	→ infogulch joined	16:54
	→ mexen joined	16:55
	← DeFreeZe left	16:56
	→ DeFreeZe joined	16:57
	→ bertvvvs joined	16:57
	→ James0r joined	16:57
	← vicfred left	16:57
phryk	mhh, git add -p also offers "e" to edit the patch, but i don't really understand patch files, so that's just as error prone… :/	16:58
	i just want to be able to do something like `git add foo.code:23` or whatever… :F	16:58
	← redoc left	16:58
James0r	Never quite ran into this before, i'm on a mac and when i rename a file with different casing i see it treats it as a new file and retains the old file, causing failed builds. is there a git setting to tell git to treat that as the same file and replace/rename it?	16:59
	i tried git config core.ignorecase false but that didn't seem to do much if anything	16:59
phryk	James0r: did you use git mv to move the file?	17:00
James0r	phryk: i did not. i just simply renamed with my editor and expected magical things to happen lol	17:00
phryk	use git mv for the rename and git'll take care of tracking it right.	17:00
	→ redoc joined	17:01
James0r	so i use git mv instead of renaming with editor or CLI?	17:01
	→ HumanG33k joined	17:02
mackerman	If it has already been renamed outside of git, git rm --cached the old one	17:03
	→ tulpa_ joined	17:03
	→ Yruama joined	17:03
BtbN	"git mv" is just a shortcut for deleting and creating a new file. It doesn't set a magic mark ala "this file was moved"	17:03
James0r	mackerman: ya that's what i did to fix it this time. how the hell have i never run into this before. one of those moments haha	17:03
BtbN	git determines that itself when building the commit, if a file vanished, and a strikingly similar one appeared elsewhere	17:04
James0r	i guess cause it's being built on netlify which is linux, still surprised	17:04
	→ squirrel joined	17:04
	← saroy left	17:04
James0r	so should i leave git config core.ignorecase false ?	17:05
	← JanC left	17:06
	→ JanC joined	17:06
	→ subopt joined	17:07
	← a1eaiactaest left	17:07
mackerman	core.ignoreCase true may have reasonable workarounds on case-insenstive file systems.	17:07
	Such as on your Mac.	17:07
	← subopt left	17:08
	→ subopt joined	17:08
	Timvde still doesn't understand how two of the biggest tech companies with some of the smartest engineers in the world came to the conclusion that case insensitive file systems are a good idea	17:09
	← igemnace left	17:10
	→ igemnace joined	17:11
mackerman	Because humans have unreasonable demands to be flexible with their inputs.	17:11
	← Gustavo6046 left	17:11
mackerman	Non-technical people may have difficulty that File.txt and file.txt are different.	17:11
	← Codaraxis left	17:12
	→ bambams joined	17:16
	← hqdruxn08 left	17:22
	→ hqdruxn08 joined	17:23
	→ analpaper__ joined	17:24
	→ pulse_ joined	17:24
	→ lllh joined	17:24
	← great_taste left	17:24
	→ Shentino_ joined	17:25
	→ jclsn6 joined	17:25
	→ rhe7 joined	17:25
	→ realies7 joined	17:25
	→ dansan_ joined	17:26
	→ HobGoblin joined	17:26
	→ jgn` joined	17:26
	→ dskullz joined	17:26
	→ lemonsni- joined	17:26
	→ MIsAn joined	17:26
	→ krushia_ joined	17:27
	→ zachel_ joined	17:27
	→ Albright_ joined	17:27
	→ wacko_ joined	17:27
	← wacko_ left	17:27
	→ wacko_ joined	17:27
	→ jinsun__ joined	17:27
	→ Roy_Mustang joined	17:28
	→ tomek_ joined	17:28
	→ Anarchic` joined	17:28
	← coot left	17:28
	→ h4x0riz3d joined	17:28
	→ Novi_ joined	17:28
	← A_Dragon left	17:28
	Roy_Mustang → A_Dragon	17:28
	→ ferr_ joined	17:28
	→ furrymcg1e joined	17:28
	→ Rh0nda joined	17:28
	→ multi_io_ joined	17:28
	→ osse1 joined	17:29
	→ mingdao_ joined	17:29
	→ constxd joined	17:29
	→ jlc_ joined	17:29
	→ lamein2 joined	17:29
	→ TeXMaster joined	17:29
	→ nuala2 joined	17:29
	→ coaxing_ joined	17:29
	→ mooz_ joined	17:29
	→ schoty joined	17:29
	→ Henry151_ joined	17:30
	← Yruama left	17:31
	bkircher_afk → bkircher	17:31
	← LuKaRo left	17:31
	→ tsdh` joined	17:33
	→ great_taste joined	17:33
	← infogulch left	17:33
	← orbyt left	17:33
	← jinsun left	17:33
	← ChmEarl left	17:33
	← Shentino left	17:33
	← zebrag left	17:33
	← wolfdale8 left	17:33
	← pulse left	17:33
	← coaxing left	17:33
	← constxo left	17:33
	← haskl[error] left	17:33
	← dskull left	17:33
	← toulene left	17:33
	← analpaper_ left	17:33
	← Vonter left	17:33
	← jclsn left	17:33
	← furrymcgee left	17:33
	← mns left	17:33
	← jlc left	17:33
	← realies left	17:33
	← mingdao left	17:33
	← llh left	17:33
	← mw_ left	17:33
	← Oblomov left	17:33
	← wolfshappen left	17:33
	← mooz left	17:33
	← multi_io left	17:33
	← wacko left	17:33
	← pcarphin left	17:33
	← antto left	17:33
	← Henry151 left	17:33
	← petaflot left	17:33
	← lemonsnicks left	17:33
	← dslegends left	17:33
	← Eustachy_Kapusta left	17:33
	← fercell left	17:33
	← dansan left	17:33
	← cweiss0765321 left	17:33
	← gumnos left	17:33
	← lamein left	17:33
	← tsdh left	17:33
	← Rhonda left	17:33
	← Church left	17:33
	Rhvs → Rhys	17:33
	rhe7 → rhe	17:33
	dskullz → dskull	17:33
	realies7 → realies	17:33
	← iffraff left	17:34
	→ LuKaRo_ joined	17:34
	→ w0ng joined	17:35
	→ Misotauros joined	17:35
	← Rashad left	17:35
	pulse_ → pulse	17:36
	→ gumnos joined	17:37
	→ cweiss07653212 joined	17:37
	→ orbyt joined	17:37
	→ Church joined	17:37
	→ zebrag joined	17:37
	→ wolfdale8 joined	17:37
	→ haskl[error] joined	17:37
	→ Vonter joined	17:37
	→ mns joined	17:37
	→ mw_ joined	17:37
	→ wolfshappen joined	17:37
	→ pcarphin joined	17:37
	→ petaflot joined	17:37
	→ dslegends joined	17:37
	→ jb1277976 joined	17:37
	→ led_belly joined	17:37
	→ deepy joined	17:37
	→ avalon joined	17:37
	→ majors joined	17:37
	→ lechner joined	17:37
	→ teear joined	17:37
	→ vikonen joined	17:37
	dansan_ → dansan	17:37
	← led_belly left	17:37
	← wolfshappen left	17:37
	→ wolfshappen joined	17:38
	→ dlwomw^ joined	17:39
	→ ChmEarl joined	17:39
	→ vicfred joined	17:40
	→ APic joined	17:41
	→ coot joined	17:41
	→ led_belly joined	17:42
	→ Sulky_ joined	17:42
	→ MajorBiscuit joined	17:43
	→ moldorcoder7 joined	17:43
	← Sulky left	17:45
	→ pyeverything joined	17:48
pyeverything	hey how can I fix this problem with git pull/push? https://bpa.st/KP2A	17:48
	← vysn left	17:48
	← duxsco left	17:49
	→ duxsco joined	17:49
JumpinBanana	pyeverything: you really need to read about merge conflicts and how to solve them. Also a git status in between commands can help to tell what is going on.	17:51
	← Guest8629 left	17:55
	→ Jacques joined	18:03
	← FinnElija left	18:03
	← tyson2 left	18:04
	→ FinnElija joined	18:06
	← dlwomw^ left	18:07
	← rsx left	18:08
rewt	pyeverything, your initial `git merge` command failed... you should resolve that before trying to `git push`	18:10
	h4x0riz3d → antto	18:13
	← orbyt left	18:17
	→ xkuru joined	18:19
	→ softirq joined	18:20
	→ orbyt joined	18:22
	← MajorBiscuit left	18:23
	→ Codaraxis joined	18:23
	← great_taste left	18:23
	← softirq left	18:23
	→ impermanence joined	18:25
	→ great_taste joined	18:26
	← t0mm13b left	18:32
	← bambanxx left	18:35
	← konrados left	18:35
	← vicfred left	18:36
	→ bambanxx joined	18:38
	← gh34 left	18:38
	TeXMaster → Oblomov	18:38
	← orbyt left	18:39
	← iomari891 left	18:39
	← tulpa_ left	18:39
	→ infogulch joined	18:42
	← lllh left	18:44
	← koo5 left	18:46
	→ koo5 joined	18:47
	← great_taste left	18:51
	→ great_taste joined	18:52
	furrymcg1e → furrymcgee	18:56
	→ MajorBiscuit joined	18:57
	← DeFreeZe left	18:59
	→ DeFreeZe joined	18:59
	← dionysus69 left	19:01
	← vitali64 left	19:01
	← rgrinberg left	19:04
	→ vicfred joined	19:04
	← roadie left	19:04
	← Jacques left	19:05
	← impermanence left	19:06
	← epolanski left	19:06
	← bambanxx left	19:09
	→ dsrt^ joined	19:12
	→ sobkas joined	19:16
	→ t0mm13b joined	19:16
	→ roadie joined	19:16
	→ arturaz joined	19:23
	→ skapata joined	19:23
	← skapata left	19:23
	→ skapata joined	19:23
	← squirrel left	19:23
	→ Jacques32 joined	19:25
	→ bambanxx joined	19:26
arturaz	Hey. I have a super weird problem. I am using Git on Windows and Git does not seem to see changes to my submodules. I change the commit in the submodule, then run `git status` or `git gui` in the parent git repository and there's... nothing. The thing this is happening to only one submodule. If I run `git submodule` I see a weird thing next to the faulty submodule in the parenthesis (quantum_unity/vendor/fp_csharp_unity (heads/master-14-gfcbcd7b7	19:27
	)), which looks different from other submodules (for example, quantum_unity/Assets/Plugins/modules/code-events (heads/pzd)). Anybody could point me to any direction? Like, what the things in parenthesis are when I run `git submodule`?	19:27
	→ dionysus69 joined	19:27
	→ Rashad joined	19:31
	→ hololeap_ joined	19:32
	← hololeap left	19:33
	← ColdKeyboard left	19:36
	→ Jacques78 joined	19:37
	hololeap_ → hololeap	19:38
	← Jacques32 left	19:39
	← adhawkins left	19:40
	→ impermanence joined	19:43
	← nwoob left	19:51
	→ AbleBacon joined	19:54
	→ ColdKeyboard joined	19:56
	→ tyson2 joined	19:56
	→ moldorcoder7_ joined	19:58
	← Murr left	19:58
	→ Murr joined	19:58
	← moldorcoder7 left	19:59
	← bambanxx left	19:59
	→ zeenk joined	20:00
	→ shan joined	20:01
	← ___nick___ left	20:03
	Shentino_ → Shentino	20:05
	→ justsomeguy joined	20:07
	→ Guest76 joined	20:07
	← Guest76 left	20:07
	→ bambanxx joined	20:08
	→ softirq joined	20:08
	→ svn2gitDev joined	20:08
svn2gitDev	Hey What's the best way to convert a Single SVN repo into multiple github projects, is it best to split them into their own repositories while using SVN still, then clone them to GIT?	20:09
	→ gh34 joined	20:10
	← gh34 left	20:11
	← duxsco left	20:16
	→ hololeap_ joined	20:16
	← hololeap left	20:17
	hololeap_ → hololeap	20:19
	→ awmv joined	20:23
	→ Lord_of_Life_ joined	20:25
	← Lord_of_Life left	20:25
	← arturaz left	20:26
	Lord_of_Life_ → Lord_of_Life	20:28
	→ Gustavo6046 joined	20:28
	→ liefer9 joined	20:29
	← hololeap left	20:30
	← liefer left	20:31
	liefer9 → liefer	20:31
	→ cdown_ joined	20:32
	← cdown left	20:34
	→ cdown joined	20:35
	→ hololeap joined	20:35
	→ junktext_ joined	20:37
	← cdown_ left	20:38
	← Jacques78 left	20:38
	← softirq left	20:39
	← hololeap left	20:39
	← cdown left	20:40
	← bertvvvs left	20:41
	→ hololeap joined	20:42
realtime-neil	How do I discover if a tag is signed _without_ attempting to verify it?	20:42
	← txtsd left	20:44
	→ softirq joined	20:45
realtime-neil	How do I print the keyid with which a given tag was signed?	20:45
	→ txtsd joined	20:46
	← Codaraxis left	20:46
	→ Codaraxis joined	20:46
	← jimklimov left	20:46
	→ forgotmynick joined	20:48
	← Codaraxis left	20:50
	→ jazzy joined	20:52
	← feriman left	20:54
	← softirq left	20:56
	← justsomeguy left	20:58
	← James0r left	20:58
	→ arturaz joined	20:59
	→ pretty_dumm_guy joined	21:00
	← oncall-pokemon left	21:01
Gustavo6046	How do I do a commit amend that at the same time behaves as a separate commit (can be pushed in front of the previous commit)?	21:03
	← svn2gitDev left	21:03
Gustavo6046	Instead of doing a dirty git push -f.	21:03
	← coot left	21:03
	← awmv left	21:04
Gustavo6046	Or alternatively, is it in bad form to make a commit with the exact same message as the previous one? (git commit -C HEAD~1)	21:05
mackerman	Gustavo6046: A regular "git commit" will not cause problems with rewriting history	21:05
	Styles vary, but I would not expect the exact same commit message. Seems confusing.	21:05
Gustavo6046	I can rewrite history with commit amend and push -f and it wouldn't really matter, but dunno	21:06
mackerman	You don't have to amend or rewrite if you do not want to	21:06
Gustavo6046	yeah	21:08
	→ jimklimov joined	21:08
Gustavo6046	I can just make another commit	21:08
	→ awmv joined	21:11
	→ Guest6935 joined	21:12
	← arturaz left	21:13
Gustavo6046	thanks!	21:13
	mackerman, it's also because it's just a routinely cleanup – like running a formatter tool on code	21:14
	← TheCatBehemoth left	21:18
	← DeFreeZe left	21:20
	← w_lehinsun left	21:20
	→ DeFreeZe joined	21:21
	← MajorBiscuit left	21:22
	→ MajorBiscuit joined	21:23
	← roadie left	21:24
	→ roadie joined	21:24
	← Techcable left	21:25
	→ Techcable joined	21:26
	← awmv left	21:27
	← rafasc left	21:27
	→ structoha1 joined	21:28
	← dionysus69 left	21:29
	← roadie left	21:29
	← dansan left	21:30
	→ dansan joined	21:31
	krushia_ → krushia	21:34
	→ dionysus69 joined	21:36
	→ awmv joined	21:38
	← sudoforge left	21:40
	→ a1eaiactaest joined	21:42
	← lemonsni- left	21:43
	← MIsAn left	21:43
	→ JoeyeS joined	21:43
	← bambanxx left	21:43
	JoeyeS → Guest3746	21:44
	← Guest3746 left	21:44
	→ funhouse joined	21:47
	← MajorBiscuit left	21:47
	→ lemonsnicks joined	21:47
	← kenanmarasli left	21:52
	→ yuljk joined	21:54
	→ roadie joined	21:55
	← structoha1 left	21:55
	← EvilDMP left	21:56
	← Murr left	21:58
	→ Murr joined	21:58
	← glik22_ left	21:59
	→ glik22_ joined	22:01
	← roadie left	22:04
	→ Jacques joined	22:06
	← jimklimov left	22:11
	→ jimklimov joined	22:13
	← palasso left	22:13
	Novi_ → Novi	22:14
	← a1eaiactaest left	22:15
	→ iomonad joined	22:15
	← thiago left	22:19
	→ bambanxx joined	22:21
	← travisghansen left	22:23
	→ travisghansen joined	22:23
	← XV8 left	22:24
	→ XV8 joined	22:25
	→ roadie joined	22:30
	→ sudoforge joined	22:31
	← dionysus69 left	22:35
	← XV8 left	22:39
	← Jacques left	22:40
	← roadie left	22:40
	→ roadie joined	22:41
	← shokohsc left	22:43
	→ shokohsc joined	22:45
	← roadie left	22:46
	→ gast0n joined	22:57
	→ softirq joined	23:06
	→ roadie joined	23:07
	→ rgrinberg joined	23:09
	→ Foxboron joined	23:09
	← sa0 left	23:09
	← rgrinberg left	23:10
Foxboron	Yo, I managed to mudge two sets of changes into a commit. I want to keep the commit message so git rebase -i, select the commit to edit	23:10
	names	23:10
	but you never really get the staged/unstaged files. So things like git restore --staged --patch doesn't work.	23:10
	→ sa0 joined	23:10
Foxboron	Is there any easy way to use rebase and still be able to deal with the staged changes?	23:10
	→ rgrinberg joined	23:11
	← nyah left	23:11
	← jazzy left	23:11
	← gurkenglas left	23:11
	← bambanxx left	23:12
	← Rashad left	23:13
	→ jazzy joined	23:15
	← sa0 left	23:15
	← jimklimov left	23:16
	← Jong left	23:17
	← forgotmynick left	23:18
	→ earthToAdam joined	23:19
	← roadie left	23:19
earthToAdam	Hello git!	23:19
	← very_sneaky left	23:21
	← DeFreeZe left	23:23
	→ michele_ joined	23:23
	→ DeFreeZe joined	23:23
	← awmv left	23:26
	→ thiago joined	23:29
bn_work	hi, git 1.8.3.1 connecting to repos hosted on github. I am running this in a shared account and don't want to save my credentials permanently so I added something like `helper = cache --timeout 28800` (8 hours) and then `2592000` (~4.3 weeks) in the hope that it would cache it for a short time but no matter what I annoying keep getting prompted for credentials in a much shorter duration, why?	23:35
	(added to my gitconfig)	23:36
	under `[credential]`	23:36
	user gitconfig in ~/.gitconfig	23:36
	getting prompted like every 20-30 mins?	23:37
	← rgrinberg left	23:39
	→ rgrinberg joined	23:42
	→ arch_angel joined	23:43
bn_work	FWIW, I am using my token to authenticate	23:44
	← agrosant left	23:44
	→ roadie joined	23:45
	← FinnElija left	23:45
	→ DoubleJ joined	23:47
	→ FinnElija joined	23:48
	← Wulf left	23:50
	→ Wulf joined	23:52
	← lionkor left	23:52
BtbN	git 1.8 is _really_ damn old, are you sure you want to be doing anything with your credentials on that machine?	23:55
	← roadie left	23:56
	→ mitrokov joined	23:57
	→ Jong joined	23:58
	← Jong left	23:59

Logs Search ←Prev date Next date→

Channels Documentation