IRCloggy #git 2021-01-07

Provider of IRC logs since 2005.

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

	2021-01-07
	← dqx left	00:00
impermanence	https://bpa.st/IOVA	00:01
	nedbat	00:01
nedbat	impermanence: yes, the new directory is still there when you switch branches.	00:01
	impermanence: git doesn't track empty directories, so it's independent of git	00:01
impermanence	okay I'll switch channels?	00:01
nedbat	impermanence: i don't understand.	00:02
impermanence	nor do I	00:02
	I don't follow you	00:02
nedbat	impermanence: what is the problem? Why switch channels?	00:02
impermanence	I've no idea; I thought you were saying "it isn't git"	00:02
nedbat	impermanence: you were asking about how git works.	00:02
	impermanence: the behavior you are seeing is expected. It's the way git works.	00:02
	theree's no problem heree	00:03
impermanence	it happens with everything	00:03
nedbat	impermanence: what do you mean by everything?	00:03
impermanence	literally anything I do in one branch gets spammed to another	00:03
	I'll show you	00:03
	one sec	00:03
nedbat	impermanence: if you make other files, and switch branches, those files still exist also.	00:04
impermanence	yes	00:04
nedbat	it's because you haven't added those files to git yet, so git doesn't manage them.	00:04
impermanence	oh	00:04
	hm	00:04
	sorry	00:04
	I feel adequately foolish	00:05
nedbat	impermanence: no need, these things have to be learned	00:05
impermanence	I don't know why that took me by surprise today	00:05
	wait no...	00:06
	something is definitely wrong though I think	00:06
	yeah I wish it was that simple;	00:06
	this happens for directories that are being tracked by git	00:07
	so say I'm on branchA and it has tracked files in a dir called dirA; and on branchB there is no dirA; when I switch to branchB now all of sudden dirA is on branchB...	00:08
	so yeah I'm not crazy that's not how git works	00:08
	I'll provide you a screenshot	00:08
	er: text	00:08
nedbat	impermanence: that would be good	00:08
	→ gjergj_kastrioti joined	00:11
	→ spacesuitdiver joined	00:12
	← tiin57 left	00:14
impermanence	nedbat: surreal. https://bpa.st/IDQQ	00:14
	← spacesuitdiver left	00:15
nedbat	impermanence: you made a new file in layer-0, but you didn't add it to git.	00:16
	← milkt_ left	00:16
nedbat	so it's a file outside of git, so git doesn't manage it.	00:16
impermanence	so?	00:16
nedbat	impermanence: so switch branches won't remove it.	00:16
impermanence	oh	00:16
nedbat	impermanence: so it's still there after you switch the bbranch	00:16
impermanence	I had no idea that was a git thing	00:16
	that makes absolutely no sense	00:17
	→ tiin57 joined	00:17
impermanence	so if what you're saying is correct then if I add it layer-0 won't show up on that other branch...	00:17
nedbat	impermanence: right	00:18
impermanence	k	00:18
	good	00:18
	← Sleepy63 left	00:19
	→ thiago joined	00:19
	→ Sleepy63 joined	00:22
	→ Newami joined	00:23
impermanence	man	00:24
	what a relief	00:24
	that was it nedbat	00:24
	I never noticed before *shrug	00:24
nedbat	great :)	00:25
	← Newami left	00:28
	→ milkt joined	00:30
	→ adlternative joined	00:32
	→ oprypin joined	00:35
	→ spacesuitdiver joined	00:39
	← stats4647 left	00:42
	← tiin57 left	00:44
	→ tiin57 joined	00:47
	← bittyx left	00:49
	→ stats4647 joined	00:50
	→ filoppi joined	00:51
	→ ppppppp joined	00:53
	← ppppppp left	00:54
	→ sudoforge joined	01:01
	← Sleepy63 left	01:04
	← moldorcoder7 left	01:06
	→ moldorcoder7 joined	01:06
	→ Sleepy63 joined	01:06
	← Jitta left	01:07
	→ nomadicmathmo joined	01:11
	← moldorcoder7 left	01:12
	→ nomadicmathmo1 joined	01:12
	← tiin57 left	01:14
	← nomadicmathmo left	01:16
	nomadicmathmo1 → nomadicmathmo	01:16
	→ moldorcoder7 joined	01:16
	→ tiin57 joined	01:17
	→ kalbasit joined	01:20
	→ bittyx joined	01:20
	→ fstd_ joined	01:22
	fstd_ → fstd	01:27
	← CodeSlingerPaul left	01:34
	← AnAverageHuman left	01:35
	← PJBoy left	01:35
	← bolovanos left	01:36
	→ kevr joined	01:37
	→ jaziz1 joined	01:38
	← mat001 left	01:43
	← tiin57 left	01:44
	→ Wetmelon joined	01:44
	← al3x27 left	01:47
	→ tiin57 joined	01:47
	← wootehfoot left	01:48
	← cloud69 left	01:48
	→ Xenguy joined	01:49
	→ adl_ joined	01:49
	← adlternative left	01:50
	→ cim209 joined	01:53
	→ AnAverageHuman joined	01:57
	→ maestrojed joined	01:57
	← adl_ left	01:59
	→ adl_ joined	01:59
	← cbreak left	02:00
	← Kurozen left	02:05
snappy	is there a way to do like a git reset --soft except without moving HEAD. I kind of want to revert back to a git commit in history, but just have all those changes staged.	02:07
	→ INeedAHandle joined	02:07
	→ nsc joined	02:11
	nsc → Guest58778	02:11
	← Guest49216 left	02:13
	← tiin57 left	02:14
	← moldorcoder7 left	02:15
	→ tiin57 joined	02:17
	→ moldorcoder7 joined	02:17
	← maestrojed left	02:17
	→ maestrojed joined	02:18
	← F0rTex left	02:18
BtbN	snappy, git diff + patch -R?	02:20
	→ F0rTex joined	02:20
	← magic_ninja left	02:20
BtbN	Or hard reset to that commit, and then reset to your previous HEAD	02:20
	← zebrag left	02:22
	→ zebrag joined	02:23
	← filoppi left	02:26
	def_jam → eb0t	02:27
snappy	BtbN: ah yeah that might do the trick, thanks	02:27
	eb0t_ → eblip	02:27
	← justan0theruser left	02:34
	← heyitsrama left	02:35
	← mowcat left	02:42
	← tiin57 left	02:44
	← Jackneilll left	02:44
	← shush left	02:45
	→ dqx joined	02:47
	→ tiin57 joined	02:47
	← gjergj_kastrioti left	02:48
Hello71	git checkout HEAD^ .	02:50
	or the modern slightly less confusing git restore -S -s HEAD^	02:50
	→ Technodrome joined	02:51
	← gigasu_shida left	02:54
	→ gigasu_shida joined	02:55
	→ drbean joined	02:56
	→ Jackneilll joined	02:56
	→ perrier-jouet joined	02:58
	← amerigo left	03:03
	← nomadicmathmo left	03:04
	← milkt left	03:04
	lagothrix → Guest44665	03:05
	→ xirhtogal joined	03:05
	← Guest44665 left	03:05
	xirhtogal → lagothrix	03:05
	→ milkt joined	03:05
	← AnAverageHuman left	03:07
	→ AnAverageHuman joined	03:08
	→ Muimi joined	03:10
	→ heyitsrama joined	03:12
	← heyitsrama left	03:12
	← tiin57 left	03:14
	← p-core left	03:14
	→ p-core joined	03:14
	→ tiin57 joined	03:17
	← maestrojed left	03:21
	→ Muimi_ joined	03:21
	← retrospectacus left	03:22
	← zebrag left	03:22
	← Muimi left	03:23
	→ zebrag joined	03:23
	Muimi_ → Muimi	03:23
	→ retrospectacus joined	03:23
	→ maestrojed joined	03:25
	→ n000g joined	03:30
	→ mgedmin joined	03:35
	← Sleepy63 left	03:38
	← benfelin left	03:42
	→ Sleepy63 joined	03:42
	← n000g left	03:42
	→ FH_thecat joined	03:43
	← tiin57 left	03:44
	→ tiin57 joined	03:47
	→ R2robot joined	03:47
kevr	when i use `git add -p`, it seems to select stages for me. is there a way i can split up one of those stages?	03:50
	→ n000g joined	03:51
	→ pyeverything joined	03:52
pyeverything	you know how i can do git diff between a file that has not yet been "git add"ed and a file on master branch already pushed? same file name	03:53
kevr	say what?	03:54
	→ geosmile joined	03:54
kevr	you can run a diff between commits, and using a branch name will use HEAD of that branch	03:54
	like `git diff HEAD..master`	03:54
	if it's not in a local copy though, you need the commits from master fetches, then you can diff between your current state and origin/master	03:55
	(or upstream/master, whatever.)	03:55
	← n000g left	04:03
	← Iarfen left	04:03
	← rosco_y left	04:04
	← spacesuitdiver left	04:06
	→ codebam__ joined	04:06
	→ n000g joined	04:09
	← codebam_ left	04:09
	← pyeverything left	04:09
	← cliluw left	04:09
	← n000g left	04:09
	← Guest58778 left	04:12
	← tiin57 left	04:14
	→ cliluw joined	04:14
	→ magic_ninja joined	04:15
	→ Anthaas joined	04:15
	→ mat001 joined	04:15
	→ nsc joined	04:16
	nsc → Guest55008	04:16
	→ tiin57 joined	04:17
	← gast0n left	04:18
	← drbean left	04:18
	← cliluw left	04:18
	← iNs left	04:21
	→ iNs joined	04:22
	← zebrag left	04:22
	→ zebrag joined	04:23
	← maestrojed left	04:23
	→ pyeverything joined	04:25
	← pyeverything left	04:25
	← FH_thecat left	04:25
	→ FH_thecat joined	04:29
	→ Guest20 joined	04:30
	→ mindCrime joined	04:34
	← thiago left	04:34
	← acercle left	04:36
	← Wetmelon left	04:37
	← stats4647 left	04:38
	← tiin57 left	04:44
	→ Goplat joined	04:46
	← Xenguy left	04:47
	→ tiin57 joined	04:47
	← zsoc left	04:52
	→ Hounddog joined	05:07
	← dking left	05:09
	← AnAverageHuman left	05:09
	← Guest20 left	05:13
	← tiin57 left	05:14
	→ tiin57 joined	05:17
	→ cliluw joined	05:17
	← zebrag left	05:22
	→ zebrag joined	05:23
	← cliluw left	05:23
	← FH_thecat left	05:23
	→ Guest20 joined	05:26
	→ cliluw joined	05:26
	← cliluw left	05:31
	→ al3x27 joined	05:34
	← Betal left	05:43
	← tiin57 left	05:44
	← geosmile left	05:45
	→ dermoth_ joined	05:46
	dermoth_ → dermoth	05:47
	→ tiin57 joined	05:47
	← arcatech left	05:48
	← TheConjurer left	05:53
	← Maxattax left	05:54
	→ shush joined	05:56
	← alyawn left	06:00
	← shush left	06:01
	← Hounddog left	06:02
	→ fuzzmz_ joined	06:02
	fuzzmz_ → fuzzmz	06:03
	→ sord937 joined	06:05
	← tiin57 left	06:14
	→ jrm2 joined	06:15
	jrm2 → jrm	06:16
	→ tiin57 joined	06:17
	→ crose joined	06:18
	→ ferdna joined	06:21
	← zebrag left	06:22
	→ zebrag joined	06:23
	← yrzr left	06:25
	← mindCrime left	06:27
	→ cliluw joined	06:27
	→ spyroboy joined	06:29
	← spyroboy left	06:29
	→ spyroboy joined	06:29
	← cliluw left	06:32
	← kenwoodfox left	06:34
	→ mindCrime joined	06:34
	→ qassim joined	06:43
	← tiin57 left	06:44
	→ tiin57 joined	06:47
	← qassim left	06:52
	→ TheConjurer joined	07:01
	→ Xiti joined	07:05
	← ferdna left	07:05
	→ toolbox joined	07:06
	→ qassim joined	07:07
	← madewokherd left	07:07
	→ cliluw joined	07:08
	← kalbasit left	07:08
	← opal left	07:11
	→ mobidrop joined	07:13
	← cliluw left	07:13
	← tiin57 left	07:14
	→ rfuentess joined	07:15
	→ opal joined	07:15
	→ thrill_seeker joined	07:15
	→ tiin57 joined	07:17
	→ Wetmelon joined	07:23
	→ nomadicmathmo joined	07:23
	← andy25225 left	07:24
	→ ravndal joined	07:26
	← nomadicmathmo left	07:27
	← mindCrime left	07:27
	← thrill_seeker left	07:28
	→ thrill_seeker joined	07:28
	← sudoforge left	07:36
	→ andy25225 joined	07:38
	→ TomyWork joined	07:38
	→ riksteri joined	07:38
	→ Fokst joined	07:41
	→ Haohmaru joined	07:42
	→ skyroveRR joined	07:43
	← tiin57 left	07:44
	→ erred joined	07:45
	→ codebam_ joined	07:46
	→ tiin57 joined	07:47
	← erred left	07:48
	← codebam__ left	07:48
	← qassim left	07:50
skyroveRR	Hey guys, I've got a linux distribution repo from which I'd like to revert/restore only a couple of files which were removed from a particular commit, which is this: http://git.pktsurf.in/smlinux/commit/?id=4257f6f8d51f95a5b81d939c1af95ebde7e888bd . I'd like to restore files in that commit that have the path "source/lxde" ... can someone kindly provide pointers on how this can be done?	07:52
	→ otisolsen70 joined	07:53
ikke	git checkout 4257f6f8d51f95a5b81d939c1af95ebde7e888bd -- source/lxde	07:55
	this would chekout any file in source/lxde from that commit	07:55
skyroveRR	ikke: I ran your command and I get this: error: pathspec 'source/lxde' did not match any file(s) known to git	07:56
osse	you have to use that commit's parent	07:57
	Boo, ikke, boo!	07:57
ikke	:(	07:57
osse	Also, git diff --diff-filter=D --name-only 4257f6f8d51f95a5b8~ 4257f6f8d51f95a5b8 is the actual list of files you want	07:57
skyroveRR	Uh.. I had files in staging... they've disappeared from there.	08:00
	→ cbreak joined	08:01
skyroveRR	... maybe not. Sorry.	08:01
	Yup. Got the files back.	08:01
	Thanks osse	08:02
	← unluckyshrubbery left	08:02
	← andreas303 left	08:05
	→ qassim joined	08:05
	→ andreas303 joined	08:06
	→ theoceaniscool joined	08:07
	← bn_work left	08:10
skyroveRR	osse: a question: why use the parent commit hash? That commit doesn't have anything relating to LXDE...	08:11
osse	skyroveRR: commits are not diffs or patches. they are snapshots	08:11
	skyroveRR: and the parent commit is the last commit those files existed in	08:11
skyroveRR	Ohh. Crystal clear.	08:11
osse	skyroveRR: so in principle you could use any commit you like to grab them from, but you might get different versions of them then	08:13
	well... any commit they exist in, of course. Like ikke FAILED TO DO	08:13
	<3	08:13
	← dqx left	08:13
	← tiin57 left	08:14
	→ vdamewood joined	08:14
ikke	😢	08:14
	→ rcoote joined	08:14
	← theoceaniscool left	08:15
	→ tiin57 joined	08:17
	→ theoceaniscool joined	08:17
	← zebrag left	08:22
	→ zebrag joined	08:23
	← Goplat left	08:23
jaziz1	when 'cloning' a remote tracking-branch, is the better practice to create a branch and merge the remote into it, or something like `git checkout -b localbranch origin/remotebranch`	08:27
	what is the term for bringing a remote-tracking branch into use, actually?	08:27
osse	you clone a repo, not a branch. Usually after cloning you already have a bunch of remote tracking branches. By "use" I assume do development, make commits and such. For that you should create a local branch	08:28
ikke	(so your last suggestion)	08:29
jaziz1	I see, sorry for poor jargon	08:29
	→ Borkr joined	08:30
	← gigasu_shida left	08:30
vdamewood	jaziz1: The term for making a branch from another branch is 'branching'. This applies to making a local branch to track a remote as well.	08:30
	or 'branching off'	08:31
jaziz1	oh I see, thanks	08:31
	I think I'll take a break from this book after I finish this chapter on branching, prob have enough now to get to work	08:32
vdamewood	jaziz1: And the best practice is to directly branch off rather than do any merging, first.	08:32
jaziz1	I'll need to figure out a workflow, not used to structured coding	08:32
	vdamewood, oh great thanks, that felt like the right thing to do	08:32
	the book mentioned merging as well, though, so was curious	08:32
	← mat001 left	08:34
	→ codebam__ joined	08:34
	← codebam_ left	08:36
	→ shush joined	08:38
	→ codebam_ joined	08:39
	→ sgn1 joined	08:41
	← codebam__ left	08:42
	→ nabil_ joined	08:43
	← shush left	08:43
	← tiin57 left	08:44
	← sgn left	08:44
skyroveRR	osse: thanks a lot. Laters.	08:44
	← skyroveRR left	08:44
	← nabil__ left	08:46
	← g00s left	08:47
	→ tiin57 joined	08:47
	→ g00s joined	08:48
	← iNs left	08:54
	→ iNs joined	08:54
	→ mozgalom joined	08:58
	→ __Aqua_Vitae__ joined	09:00
	→ chele joined	09:02
	→ Fissure1 joined	09:05
	← tiin57 left	09:14
	→ tiin57 joined	09:17
	← hnOsmium0001 left	09:18
	← zebrag left	09:22
	→ zebrag joined	09:23
	← cd left	09:23
	→ shivam_ joined	09:24
	← impermanence left	09:24
	→ PJBoy joined	09:24
	→ another joined	09:25
	→ mindCrime joined	09:28
	→ clime joined	09:29
	→ calcul0n_ joined	09:40
	→ unluckyshrubbery joined	09:40
	← tiin57 left	09:44
	→ tiin57 joined	09:47
	← Wetmelon left	09:47
	→ lofo_ joined	09:49
	← lofo left	09:53
	→ dudek joined	09:59
	→ nabil__ joined	10:01
	← nabil_ left	10:04
	→ benfelin joined	10:12
	← tiin57 left	10:14
	→ tiin57 joined	10:17
	← zebrag left	10:22
	→ zebrag joined	10:23
	→ justanotheruser joined	10:29
	→ biryukov joined	10:29
jaziz1	Would it be okay to specify certain sections of a file for certain code additions?	10:30
	to prevent merge conflicts?	10:30
	maybe in code comments or something	10:30
osse	Git has no way of enforcing that	10:31
	→ shabius joined	10:32
osse	But you can have any agreements you like human to human	10:32
	← shabius left	10:32
	→ Sp4rKy joined	10:32
	→ shabius_ joined	10:32
	← shabius_ left	10:32
jaziz1	Yeah, was just wondering if people did that I guess	10:32
	given I haven't really been part of a team	10:32
osse	There's no "I" in "team", but there is one in "merge conflict"	10:33
	→ shabius joined	10:33
jaziz1	lol	10:34
	← austin987 left	10:37
vdamewood	There's a 'me' in team, scrambled in there.	10:38
	Come to think of it, team is an anagram of "at me".	10:39
	← Sp4rKy left	10:39
	← tiin57 left	10:44
	← __Aqua_Vitae__ left	10:44
	→ tiin57 joined	10:47
	→ austin987 joined	10:49
	→ humanface22 joined	10:51
humanface22	hi all	10:51
gitinfo	humanface22: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.	10:51
	← mindCrime left	10:52
	← thrill_seeker left	10:53
	→ blahdodo joined	10:53
	→ thrill_seeker joined	10:53
humanface22	is the git able to do the following?: checkout a previous commit on branch A (lets call this 1. commit), create a branch B on that, miss branch A's 2. and 3. commit, and add only the 4. 5. 6. 7. commits into the branch B	10:54
	should I cherry pick 4. 5. 6. 7. commits in order?	10:55
	→ __Aqua_Vitae__ joined	10:57
	← maxwilliamson left	11:00
	→ maxwilliamson joined	11:01
	→ cheater joined	11:01
cheater	hi! what's the best way to: create a new branch; which tracks a remote on origin of the same name; and switch to that branch?	11:02
	git checkout -b newbranch doesn't track remote	11:02
ikke	git checkout -t origin/newbranch, or git checkout newbranch	11:04
	the latter is automatically creating the branch when it does not exist yet, and there is only one remote branch with that name	11:04
osse	humanface22: that is not a problem	11:05
	← Trieste left	11:05
humanface22	osse what do you mean?	11:05
osse	humanface22: the thing you described is straight forward to do	11:06
humanface22	osse ok, so what is the answer?	11:06
	osse should I cherry pick 4. 5. 6. 7. commits in order?	11:06
	so should I do those steps in order? checking out that commit, creating branch, cherry picking those commits?	11:07
osse	humanface22: create a new branch named B at the point where you want B to start from. check it out, then cherry-pick in date order	11:07
cheater	ikke: both don't work. the first one says "fatal: 'origin/newbranch' is not a commit and a branch 'newbranch' cannot be created from it" and the second one says "error: pathspec 'newbranch' did not match any file(s) known to git.".	11:08
	→ Trieste joined	11:08
humanface22	osse ok is it differ from what I said already?	11:08
ikke	cheater: If it's a completely new branch, then there is no remote tracking branch yet to track	11:08
osse	humanface22: maybe not. depends a bit on what you meant by the numbers	11:08
cheater	well yes, i said "new branch", that's what i meant	11:09
ikke	cheater: then you first create it locally, and when you push, you specify --set-upstream	11:09
cheater	what can i do so that i don't need to --set-upstream explicitly?	11:09
humanface22	osse the date order :) firstly you create the 1. commit, then the 2. commit, and so on	11:09
cheater	i just want it to know it should be set to remote/newbranch	11:09
	is there no command that does both?	11:09
ikke	cheater: not when remote/newbranch does not exist yet	11:10
	which gets created after you push	11:10
cheater	so what would i do, just git checkout -b to create a local branch, and then --set-upstream later?	11:10
ikke	yes	11:10
cheater	ok. thanks	11:10
ikke	git push -u <remote> <branch>	11:10
osse	humanface22: then what you describe is spot on	11:10
humanface22	osse ok, thanks	11:11
	→ aw1 joined	11:11
	→ PacKetSlayer joined	11:12
	← tiin57 left	11:14
cheater	thank you ikke	11:17
	→ tiin57 joined	11:17
	← humanface22 left	11:19
	← zebrag left	11:22
	→ nabil_ joined	11:22
	→ zebrag joined	11:23
	← nabil__ left	11:25
	→ BH23_ joined	11:25
	→ thiago joined	11:28
	→ mellernoia joined	11:30
	→ nabil__ joined	11:34
	← sgn1 left	11:36
	← nabil_ left	11:37
	→ aliss joined	11:37
	← Jookia left	11:37
	← caveman left	11:38
	→ sgn joined	11:41
	→ Guest20_ joined	11:41
	← Guest20 left	11:42
	→ lacrymology joined	11:42
	→ DevAntoine joined	11:43
	← tiin57 left	11:44
	← milkt left	11:44
	← adl_ left	11:45
	→ Masklin joined	11:45
	→ adl_ joined	11:45
	→ Atlenohen joined	11:46
	← adl_ left	11:47
	→ adl_ joined	11:47
	→ tiin57 joined	11:47
	→ calcul0n__ joined	11:48
	← DevAntoine left	11:48
	← calcul0n_ left	11:50
	← j7k6 left	11:51
	→ milkt joined	11:52
	← al3x27 left	12:00
	→ rosco_y joined	12:01
	← adl_ left	12:05
	→ adl_ joined	12:05
	→ NitrousPG joined	12:07
	→ dwmw2_gone joined	12:09
	← adl_ left	12:10
	→ adlternative joined	12:10
	← tiin57 left	12:14
	→ gjergj_kastrioti joined	12:17
	← __Aqua_Vitae__ left	12:17
	→ tiin57 joined	12:18
	← andreas303 left	12:18
	→ tgunr joined	12:21
	← shivam_ left	12:22
	← zebrag left	12:22
	→ zebrag joined	12:23
	→ cliluw joined	12:24
	→ davve joined	12:25
davve	hey, is there an interactive or convenient way of creaning up ones branches (local and remote?)	12:26
	i tend to make a lot of them ^^	12:26
osse	local: not directly. git branch --merged master produces a list of branches that's at least safe to remove	12:27
davve	like interactive rebase where they all default to keep or something	12:27
osse	remote: try git fetch --prune	12:27
davve	osse: do you mean me?	12:27
	ah right	12:27
	ofc you do lol. cool i'll try that, thought local and remote were two people in here :)	12:28
	the --merged master arguments to branch is great	12:29
	← cliluw left	12:29
davve	prune cleaned up a few too. neat	12:29
	→ __Aqua_Vitae__ joined	12:31
	← luke-jr left	12:31
davve	but i still have a bunch, will have to write up a script i think. I'll try googling too	12:31
osse	davve: local ones?	12:32
	you can remove those too in one swoop if you really wish	12:32
davve	yeah	12:32
	no i'd rather have a prompt or somthing :)	12:32
	some are keepers	12:32
	→ luke-jr joined	12:32
davve	but for science, how would you do that?	12:33
	→ andreas303 joined	12:35
osse	something like:	12:35
	→ wootehfoot joined	12:35
osse	git for-each-ref --merged master --format='delete %(refname)' -- refs/heads \| sed '/\/master$/d' \| git update-ref --stdin	12:35
	for-each-ref is the script-friendly version of git branch/tag	12:36
	→ Peetz0r joined	12:39
	→ shivam_ joined	12:40
	→ shush joined	12:40
	→ Tazmanya joined	12:41
	← Tazmanya left	12:43
	← tiin57 left	12:44
	← shush left	12:45
	→ tiin57 joined	12:47
	← Guest20_ left	12:47
	→ bolovanos joined	12:48
jjakob	I run git rebase commithash^ , select the commithash to edit, but git rebase shows the commit I selected to edit as already done? Last command done (1 command done): edit abcdefg mycommit	12:50
	→ worc3131 joined	12:50
jjakob	running git reset resets no files...	12:50
osse	jjakob: edit means git rebase stops after that commit, and you can change it with commit --amend	12:51
	then run rebase --continue	12:51
jjakob	ah right, but I want to split it into 2	12:51
	reset HEAD~ ?	12:52
osse	yes	12:52
	but since it is only the latest commit you want to change, you can do it all without the rabase stuff	12:52
jjakob	it's not the latest, it's quite a few back	12:52
osse	oh	12:53
	i misunderstood the initial question	12:53
	in that case, carry on	12:53
jjakob	I selected it with commithash~	12:53
	→ DevAntoine joined	12:55
	← cheater left	12:57
	← vdamewood left	12:59
	← sgn left	13:00
	→ Guest20 joined	13:01
	← worc3131 left	13:01
	→ worc3131 joined	13:02
	→ sgn joined	13:06
relipse	I have a fork of the primary repo, how do I make my fork's master the exact same as the primary repo? i tried git rebase remote/master but it had some 3 way merge stuff	13:07
	→ Jitta joined	13:09
	← bolovanos left	13:11
relipse	i tried git pull remote master but that din't work eather said fix conflicts	13:12
j416	relipse: reset	13:13
	← tiin57 left	13:14
relipse	git reset --hard upstream/master	13:14
	that did it. thank you j416	13:14
	→ jl4 joined	13:17
jaziz1	what are some good git workflows I should look into?	13:17
	→ tiin57 joined	13:18
	→ j7k6 joined	13:18
	← sweatsuit left	13:20
	← zebrag left	13:22
	→ zebrag joined	13:23
jaziz1	never done any real work in git, just kind of haphazardly wrote things, copy/pasted for backup, etc	13:23
	→ Jookia joined	13:23
	→ caveman joined	13:24
	→ rsx joined	13:24
	← luke-jr left	13:25
	→ sweatsuit joined	13:25
	→ luke-jr joined	13:25
osse	if you work alone I would not focus much on workflows	13:26
	that's for your boss to decide, unless you are a boss	13:26
	← justanotheruser left	13:26
osse	make commits that make sense, and use different branches for different things if you switch between areas you work on	13:27
jaziz1	I see	13:28
	Maybe one 'main' branch for stable things and then branch off for individual features?	13:28
osse	yes	13:31
	→ dionysus69 joined	13:31
	→ mat001 joined	13:32
	← spyroboy left	13:32
	← Guest20 left	13:32
	→ zsoc joined	13:35
	← zsoc left	13:35
	→ zsoc joined	13:35
	← biryukov left	13:35
	→ justanotheruser joined	13:40
	→ dd7 joined	13:41
	← tilpner left	13:42
jaziz1	reading about this 'rebase' and 'fast forward' stuff	13:42
	→ tilpner joined	13:42
jaziz1	is it not bad to lose history that you don't have to lose?	13:43
	thought one of the points of git was to easily identify what has happened in the past?	13:43
	← tiin57 left	13:43
dd7	I notice that if the `core.pager` process reads git config by invoking `git config --list`, that it picks up modifications to the config made on the command line of the parent git process by e.g. `git -c foo.bar=baz`. How does this happen? I would have thought that `git -c` causes the config object to be mutated in the memory of the parent process	13:44
	whereas `git config --list` is simply reading from disk again.	13:44
	→ tiin57 joined	13:47
	← dionysus69 left	13:48
	→ Voltum joined	13:48
	← Voltum left	13:48
osse	jaziz1: some history is not worth keeping	13:50
	jaziz1: let's say you make a commit. then you find you forgot to add a file to it. so you do that, and modify the commit you just created instead of making a second commit	13:50
	technically, the "bad" commit is still lying around	13:50
ikke	Some use git append-only, sadly	13:51
	← __Aqua_Vitae__ left	13:51
osse	jaziz1: but it will be lost, and no one sheds a tear for it	13:51
jaziz1	osse, mm right	13:52
	that seems less consequential than something like rebase though	13:52
osse	depends on how you use it	13:52
	the usual form of a rebase is to just put your commits on top	13:53
	→ n0mn0m joined	13:53
dd7	jaziz1: a big factor is whether you're talking about work that only exists on your personal machine, or work that you've pushed to a remote and others may be sharing.	13:53
	← xelxebar left	13:53
jaziz1	dd7, that makes sense, local history might not be relevant to the repo as a whole	13:53
dd7	If it's your local personal work, it's common to make a bunch of messy commits and then tidy them up afterwards	13:54
	"tidy up" the history I mean	13:54
jaziz1	mm	13:54
	← rcoote left	13:54
dd7	Personally my rules are (a) I do that a lot and (b) but I don't if it involves any kind of merge conflict resolution	13:54
jaziz1	I'm just trying to get a good sense of best practices since I won't be able to do it on the job	13:54
	since no job lmao	13:54
osse	as for fast-forwarding, that has nothing to do with losing history	13:55
jaziz1	dd7, I see	13:55
	osse, you could be losing some history, though, right?	13:56
	as shown here? https://i.snipboard.io/9goF0P.jpg	13:56
osse	this is showing two end results	13:57
	the new yellow dot on the left hand side doesn't exist before you start	13:57
ikke	In that case you're note losing history, at most, you cannot distinguish in the 2nd example that those commits belong together	13:57
jaziz1	osse, hmm	13:58
osse	you could argue there is some history that will never be created	13:58
	and in many cases you don't want that history, for example when you're not merging in a feature, but just merging in origin/master (ie. gettnig your local master up to date)	13:58
jaziz1	right	13:59
	ikke, well, knowing where you' branched off for a feature is still history right?	14:00
osse	what if the feature branch has been rebased multiple times before it is merged in? :P	14:00
jaziz1	no idea haha	14:01
	visualizing rebasing is still kinda tough for me honestly haha	14:01
osse	my point is that the branched-off point might not be the actual point at which the feature development started	14:01
	https://github.com/Osse/git-stuff/blob/master/rebase.txt	14:02
ikke	(in the 2nd example, the old commits are still there where they were, just not referenced anymore by the current branch	14:02
	0	14:02
	n0mn0m → n0mn0m[away]	14:03
	→ __Aqua_Vitae__ joined	14:03
	→ haslo joined	14:03
	← n0mn0m[away] left	14:04
jaziz1	osse, hmm I guess I've misunderstood rebase, too haha	14:05
	it was my understanding that it'd just overlap commits starting from where it branched off	14:06
osse	jaziz1: rebase = re-base = changing the "base" of a branch (ie. where it appears to have branched off)	14:06
	← dd7 left	14:06
jaziz1	I see	14:07
osse	jaziz1: using my fancy ascii art for reference: Before the "base" of the feature branch was D, but after the rebase it is H	14:07
	the base has changed => the branch has been re-based	14:08
jaziz1	mhmm	14:08
	← thiago left	14:13
	→ mowcat joined	14:13
	← tiin57 left	14:14
osse	that means rewriting some commits and stuff. as ikke said the old commits are still around, technically	14:15
	but they have been LOST :O :O	14:15
	→ sudoforge joined	14:15
osse	Kate, we have to go back!	14:15
jaziz1	oooh I see	14:15
	so you take the base of the branch	14:15
	and the just kind of append the commits in the same progression to the end of the other one?	14:15
	and if the work is totally independent you come out with no conflicts?	14:16
selckin	you have a tree, saw off a branch, and glue it on somewhere else	14:17
jaziz1	lmao	14:17
ikke	(while you are standing on it)	14:17
	→ tiin57 joined	14:18
jaziz1	how are you going to commit if you're splattered all over the ground?	14:18
osse	jaziz1: yep, that's essentially what happens	14:19
jaziz1	coool	14:19
osse	d'accord	14:21
	← zebrag left	14:22
	→ zebrag joined	14:23
	← gjergj_kastrioti left	14:23
	→ xelxebar joined	14:26
	← mowcat left	14:27
	→ herren joined	14:28
	← ghost43 left	14:28
	→ ghost43 joined	14:29
	← retrospectacus left	14:31
	→ retrospectacus joined	14:31
	→ bolovanos joined	14:37
	← __Aqua_Vitae__ left	14:39
	← Jck_true left	14:41
	→ Jck_true joined	14:42
	← perrier-jouet left	14:43
	← milkt left	14:43
	← tiin57 left	14:44
	→ tiin57 joined	14:47
	→ milkt joined	14:49
	→ impermanence joined	14:49
	→ __Aqua_Vitae__ joined	14:52
	→ k3yp joined	14:55
	→ perrier-jouet joined	14:56
	← perrier-jouet left	14:58
	→ AnAverageHuman joined	14:58
	→ k3yp- joined	14:59
	← k3yp left	15:00
	k3yp- → k3yp	15:01
	← adlternative left	15:02
	→ Maxattax joined	15:02
	→ adlternative joined	15:02
	→ CodeSlingerPaul joined	15:07
	← Borkr left	15:08
	← jl4 left	15:08
	← impermanence left	15:09
	→ maxwilli- joined	15:09
	← maxwilliamson left	15:10
	→ perrier-jouet joined	15:11
	→ acercle joined	15:11
	← tiin57 left	15:14
	→ Newami joined	15:15
	← Newami left	15:16
	→ spacesuitdiver joined	15:16
	→ gavlee joined	15:17
	→ tiin57 joined	15:17
	→ wrobinso1 joined	15:18
	← aw1 left	15:20
	→ aw1 joined	15:21
	← wrobinson left	15:22
	← zebrag left	15:22
	→ zebrag joined	15:23
	← adlternative left	15:26
	→ redcaptrickster joined	15:34
	→ lofo joined	15:36
	← lofo_ left	15:39
	→ Xenguy joined	15:40
	← worc3131 left	15:41
	← justanotheruser left	15:42
	← tiin57 left	15:44
	← austin987 left	15:44
	→ trinityblade joined	15:44
	← acercle left	15:47
	→ tiin57 joined	15:47
	→ acercle joined	15:48
	← Xiti left	15:57
	→ austin987 joined	15:57
	→ hnOsmium0001 joined	15:59
	← DevAntoine left	15:59
	→ DevAntoine joined	15:59
	→ al3x27 joined	16:02
	← sudoforge left	16:02
	→ sudoforge joined	16:04
	→ Borkr joined	16:04
	← DevAntoine left	16:04
southerntofu	hey folks i don't know if this is okay according to your etiquette, but i'd like to bring attention to a bug report i submitted two days ago on the ML (with no answer so far): https://public-inbox.org/git/X%2FSq38YKmLjY4KmD@thunix.net/T/#u	16:04
	thanks and have a great day :)	16:05
	← fling left	16:05
	← thrill_seeker left	16:06
	→ nkz joined	16:06
	→ thrill_seeker joined	16:07
nkz	Hey, noob here. When I master into feature branch, git creates a merge-commit. When I do that other way around git doesn't create a merge-commit, just adds on commits. Why is that?	16:07
	When I merge master into feature branch	16:08
	← sudoforge left	16:10
	→ fling joined	16:10
	← shivam_ left	16:12
osse	nkz: git doesn't make a merge commit unless it has to	16:12
	so if one branch can be "fast-forwarded", that's what happens	16:12
	when you do it the other way I would suspect it just says "already up-to-date" and does nothing at all	16:13
	→ mindCrime joined	16:13
	← haslo left	16:13
	← tiin57 left	16:14
nkz	osse: makes sense, thanks!	16:14
	→ Xiti joined	16:16
	← thrill_seeker left	16:16
	→ tiin57 joined	16:17
	→ thrill_seeker joined	16:17
	← thrill_seeker left	16:17
	→ thrill_seeker joined	16:17
	← thrill_seeker left	16:18
	→ mowcat joined	16:18
	← TomyWork left	16:18
	→ DevAntoine joined	16:19
	→ thrill_seeker joined	16:19
	← thrill_seeker left	16:19
	→ thrill_seeker joined	16:20
	← thrill_seeker left	16:20
	→ Iarfen joined	16:20
	→ thrill_seeker joined	16:20
	← thrill_seeker left	16:20
	→ thrill_seeker joined	16:21
	← caterfxo left	16:21
	← thrill_seeker left	16:21
	← fling left	16:21
	→ lss joined	16:21
	→ thrill_seeker joined	16:21
	← thrill_seeker left	16:21
	→ caterfxo joined	16:22
j416	nkz: someone posted an image of exactly this scenario just a while ago; https://i.snipboard.io/9goF0P.jpg	16:22
	well, close enough.	16:22
	→ Guest20 joined	16:22
j416	if git can fast-forward, it'll do that by default.	16:23
	→ thrill_seeker joined	16:23
	← aliss left	16:23
	← DevAntoine left	16:23
	← thrill_seeker left	16:23
	→ thrill_seeker joined	16:24
	← thrill_seeker left	16:24
	→ thrill_seeker joined	16:24
	← thrill_seeker left	16:24
	→ sgn1 joined	16:24
nkz	j416: thank you. I understand now	16:24
	→ thrill_seeker joined	16:25
	→ DevAntoine joined	16:25
	← sgn left	16:28
	→ moviuro joined	16:28
	→ madewokherd joined	16:28
	→ Raging_Hog joined	16:29
	← DevAntoine left	16:29
	→ fling joined	16:29
	→ wnbeser joined	16:30
	→ heyitsrama joined	16:30
j416	o/	16:32
	← spacesuitdiver left	16:36
	← nkz left	16:38
	→ gast0n joined	16:38
	→ stats4647 joined	16:39
	→ shush joined	16:42
	← tiin57 left	16:44
	→ mirrorbird joined	16:44
	→ rcoote joined	16:44
	← bolovanos left	16:45
	← zebrag left	16:46
	→ zebrag joined	16:46
	← shush left	16:47
	→ tiin57 joined	16:47
	→ daniele76 joined	16:48
	← daniele76 left	16:49
	← lss left	16:55
	→ aliss joined	16:55
	← mozgalom left	16:55
	→ shivam_ joined	17:01
	→ shush joined	17:01
	→ aquery joined	17:01
	← Guest20 left	17:01
	← shivam_ left	17:05
aquery	Hello all, I'm a bit confused, hope you can help me. I wanted do create a pull-request for a project using git (codeberg.org). I forked it, cloned the forked repo, pushed my changes and submitted a pull request.	17:05
	← Haohmaru left	17:07
aquery	It got accepted and all is good. But the forked repo didn't go anywhere, it's still sitting there. Now I have a new feature (in the same project) that I need to develope. But I can't fork the master branch - because that forked repo is still sitting there!	17:07
	← Hello71 left	17:07
ikke	You need to fork only once	17:07
	You update that fork through your local repo	17:07
	Generally, you fetch the main branch from upstream, and create new feature branches, and push those to your fork	17:08
	← rfuentess left	17:08
	→ Hello71 joined	17:08
aquery	ikke: ok, but the master branch has changed since I forked it, and my forked repo doesn't show those changes	17:08
ikke	aquery: generally that does not matter	17:08
	not for making pull requests at least	17:08
	You add the upstream (the repo you forked from) repo as a remote	17:09
	to your local repo	17:09
aquery	Hmmmm that might be the key step I missed. One moment.	17:10
	← zebrag left	17:11
aquery	ikke: ok so on my local repo, running "git remote get-url --all origin" gives me only one url - to the forked repo	17:12
ikke	you could do git remote add upstream <upstream-url>	17:12
aquery	ikke: is that the classic way to do things? I'm trying to realize what step in the workflow I missed.	17:13
ikke	yes, it is	17:13
	← Technodrome left	17:13
	← tiin57 left	17:13
	← caterfxo left	17:14
aquery	ikke: ok cool. I'll do that. After that, git pull origin master should give me the up-to-date repo?	17:14
ikke	no	17:14
	you need to fetch from upstream	17:14
	origin is your fork, which does not have any changes	17:14
	→ OnkelBlade joined	17:14
	→ caterfxo joined	17:14
	→ zebrag joined	17:15
ikke	git pull --ff-only upstream master (while on master)	17:15
aquery	ikke: oh, ok, right. Let me try it real quick :)	17:15
	→ Guest20 joined	17:16
	→ spacesuitdiver joined	17:17
	→ tiin57 joined	17:17
	← Muimi left	17:18
	→ jl4 joined	17:18
aquery	ikke: awesome.	17:21
	ikke: many thanks for your help :))	17:22
	← raven-au left	17:23
ikke	np	17:25
	→ aesthe joined	17:27
	← mindCrime left	17:27
	← aquery left	17:28
	→ Technodrome joined	17:29
	← Sleepy63 left	17:29
	→ arcatech joined	17:31
	→ Betal joined	17:33
	← HumanG33k left	17:36
	← trinityblade left	17:36
	→ IvanZhuro joined	17:36
	← shush left	17:36
	→ shush joined	17:36
	← IvanZhuro left	17:37
	→ HumanG33k joined	17:37
	← shush left	17:38
	→ nvmd joined	17:38
	→ shush joined	17:38
	→ mindCrime joined	17:38
	← caterfxo left	17:42
	→ mozgalom joined	17:42
	→ caterfxo joined	17:43
	← tiin57 left	17:43
	← HumanG33k left	17:45
	→ orbyt_v7 joined	17:45
	← jaziz1 left	17:45
	← zebrag left	17:45
	→ Guest20_ joined	17:46
	→ zebrag joined	17:46
	→ Sleepy63 joined	17:46
	→ tiin57 joined	17:47
	← Guest20 left	17:48
	← zebrag left	17:49
	→ HumanG33k joined	17:51
	→ kalbasit joined	17:52
	→ bn_work joined	17:53
	← HumanG33k left	17:54
	← moviuro left	17:54
	← mindCrime left	17:54
	→ HumanG33k joined	17:55
	← fachinformatiker left	17:58
	→ fachinformatiker joined	18:02
	← fachinformatiker left	18:02
	→ fachinformatiker joined	18:02
	→ moviuro joined	18:02
	→ zebrag joined	18:03
	← moviuro left	18:03
	→ moviuro joined	18:06
	← chele left	18:07
	→ n0mn0m joined	18:10
	→ soifou joined	18:11
	← tiin57 left	18:13
	← arcatech left	18:14
	← wnbeser left	18:16
	→ tiin57 joined	18:17
	→ Wetmelon joined	18:17
	← nabil__ left	18:17
	← mirrorbird left	18:19
	n0mn0m → n0mn0m[away]	18:20
	← wootehfoot left	18:20
	← n0mn0m[away] left	18:20
	→ dqx joined	18:21
	← zebrag left	18:22
	→ zebrag joined	18:23
	→ cliluw joined	18:24
	→ DevAntoine joined	18:28
	← cliluw left	18:29
	→ alyawn joined	18:33
	← DevAntoine left	18:34
	← aesthe left	18:34
	← thrill_seeker left	18:36
	→ trinityblade joined	18:37
	→ napping joined	18:37
	→ plexigras joined	18:40
	→ cfoch joined	18:40
	← DarthRadar left	18:41
	← tiin57 left	18:43
	← mellernoia left	18:46
	→ aesthe joined	18:46
	→ reactormonk joined	18:47
	→ mirrorbird joined	18:47
	→ tiin57 joined	18:48
	→ dionysus69 joined	18:49
	→ physikoi joined	18:51
	← TheConjurer left	18:52
physikoi	#git, does `git reset --hard HEAD~1` permanently delete last commit? I accidentaly did hard instead of soft. I had just commited some changes, but accidentally includes some changes that I didn't want in the commit. So, I tried to undo the commit. Now I'm afraid I lost a lot of work. Help?	18:53
	→ arcatech joined	18:53
physikoi	I tried going through reflog, but none of the recent commits in that log bring me back to the state i had been in	18:54
	← aesthe left	18:54
	→ Lord_of_Life_ joined	18:56
	← Iarfen left	18:58
	→ cliluw joined	18:58
	← Lord_of_Life left	18:59
	Lord_of_Life_ → Lord_of_Life	18:59
	← luke-jr left	18:59
	→ luke-jr joined	18:59
imMute	physikoi: if it's been less than 14 days, that commit should be somewhere in the reflog	18:59
physikoi	it's been less than 2hours	19:01
	To be perfectly clear, I did these two things: add all, commit, then `git reset --hard HEAD~1`	19:02
	→ gonzo joined	19:02
imMute	definitely should be in the reflog then	19:03
	→ amahl_ joined	19:03
	← amahl_ left	19:03
	→ amahl_ joined	19:04
	← arcatech left	19:06
physikoi	imMute: you're right. Not sure if I took the right next step to recovery (git reset HEAD@{15} in this case). This did unstage those changes. I think i'm good from here. Thank you!	19:07
Jitta	Hi, I would like to use a credential helper for a specific remote (i.e. github) but not for others. Any idea/hints how to do that?	19:13
	← tiin57 left	19:13
shush	Anyone have some documentation on `:/` syntax? It was mentioned in a article about `git commit --fixup`	19:14
physikoi	Jitta: Like SSH?	19:14
shush	Apparently you can reference a commit by it's message	19:14
	I'm wondering if you can reference a commit by it's message and use a specific downstream commit when search upstream	19:15
	← mirrorbird left	19:16
imMute	shush: can you link to the article?	19:16
Jitta	physikoi: like "git config --global credential.helper xyz" but only for a specific remote	19:16
	→ mellernoia joined	19:16
	→ arcatech joined	19:16
	→ tiin57 joined	19:17
	→ wnbeser joined	19:17
	← Wetmelon left	19:18
	← Sleepy63 left	19:18
	← aw1 left	19:20
	← alyawn left	19:21
	→ alyawn joined	19:21
shush	https://thoughtbot.com/blog/autosquashing-git-commits	19:23
	→ k3yp- joined	19:24
	← k3yp left	19:25
imMute	shush: I've never seen that syntax before and it's not mentioned on the git-commit or git-rebase manpages	19:26
	k3yp- → k3yp	19:26
shush	That's intersting	19:27
	s/intersting/interesting	19:28
j416	you'll find the syntax in man gitrevisions	19:28
gitinfo	the gitrevisions manpage is available at https://gitirc.eu/gitrevisions.html	19:28
osse	autosquashing is the best kind of squashing	19:30
shush	https://stackoverflow.com/questions/52039149/refer-to-a-git-commit-by-commit-message-headline	19:30
j416	the man page says youngest commit; note that this is by date, not by topology.	19:31
	→ Sleepy63 joined	19:31
shush	j416: Is it using the commit date or the author date I wonder	19:32
j416	shush: didn't know about <rev>^/text; thanks.	19:32
imMute	is that commit date or author date?	19:32
j416	I forget. but it should be easy to check.	19:32
shush	j416: Almost: HEAD^{/text}	19:32
j416	shush: HEAD is a revision.	19:32
shush	You forgot the curly brace	19:32
j416	oh, it has to be {}	19:33
shush	It appears so	19:33
j416	makes sense	19:33
	indeed; doesn't work without {}	19:33
	thanks	19:33
shush	This is pretty cool. Now I have all I need to mimic the way fixup! commits work for rebase	19:33
	← OnkelBlade left	19:34
j416	the pitfall with the youngest by timestamp rather than topology is that it might pick a commit from say a different branch -- not always what you'd intend	19:35
	→ gjergj_kastrioti joined	19:35
shush	Isn't it only looking upstream?	19:36
j416	no	19:36
	but for --fixup, as long as you verify that the commit message subject is the correct one, it'll work just fine. All that matters for --fixup is that the subject matches.	19:37
shush	:/<text> will find the one that is reachable from any ref	19:37
j416	yes	19:37
shush	But HEAD^{/<text>} will only search what is reachable from HEAD	19:38
j416	yes.	19:38
	but it's quite a bit to type	19:38
	← gonzo left	19:38
shush	j416: It is	19:38
	← rsx left	19:38
j416	at that point HEAD~5 or what have you seems faster	19:38
shush	I'm not going to be typing it	19:38
j416	(which can be further shortened to @~5)	19:38
shush	`git portal some-branch` will by my alias	19:39
	→ cd joined	19:39
napping	I've been testing different orders of making commits and running gc. How is it that 'git gc --aggressive' helps a later 'git gc' find a better result? The description of pack heuristics at https://git-scm.com/docs/pack-heuristics doesn't seem to mention looking at the old pack	19:39
shush	git rebase --onto ':/future! some-branch' 'HEAD^{/future! some-branch}'	19:40
j416	napping: I don't have your answer, but you may find it interesting to experiment with git-repack.	19:40
shush	That'll move the current branch to the latest "future commit"	19:40
j416	shush: heh. seems like quite the roundabout.	19:41
	← lacrymology left	19:41
shush	j416: How else can you move a branch that depends another feature branch that isn't yet merged into master?	19:42
napping	j416: its man page mentions git-pack-objects, and a --no-reuse-delta option that sounds relevant	19:42
shush	j416: All while having a commit that stands in place for that merge for the peer-review stage?	19:43
	← tiin57 left	19:44
j416	shush: I typically just use aforementioned HEAD~<n> syntax, or hashes directly	19:44
shush	How do I test a specific revision?	19:44
	Like how can I see what the hash is for a specific revision parameter?	19:44
	→ lacrymology joined	19:44
j416	shush: what do you mean by revision parameter?	19:44
shush	j416: Yes, but it'll take longer to create the merge commit, and then deal with rebasing manually	19:45
ikke	git rev-parse <revision>	19:45
shush	j416: A rev param would be HEAD~3 for example	19:45
j416	if you have that, you don't need the hash	19:45
	just use HEAD~3, then	19:45
shush	ikke: Thanks	19:45
j416	but, what ikke said.	19:45
shush	j416: It's to confirm things are working	19:45
	That my rev works before I use it	19:45
	Sure I can just do `git undo`	19:46
j416	unfamiliar with that	19:46
shush	which is an alias for undoing the last change to a branch	19:46
	:D	19:46
j416	what about the worktree	19:46
shush	undo = "!f() { git reset --hard $(git here)@{${1:-1}}; }; f"	19:46
j416	scary	19:47
shush	scary how?	19:47
j416	if you hadn't committed and did git undo	19:47
	but, of course that goes for reset --hard itself, too.	19:47
	→ tiin57 joined	19:47
shush	Then you looks your worktree	19:47
	← mozgalom left	19:47
j416	just thought undo seems more.. harmless	19:47
shush	I'm open to suggestions	19:48
	--soft then?	19:48
	or --mixed	19:48
j416	do whatever you prefer :)	19:48
shush	Well, it would be good to not lose changes that haven't been checked in	19:48
	But I don't want to keeps all the shit that was undo	19:48
j416	commit or stash them?	19:48
shush	undone*	19:48
	do a git stash and then pop between the reset	19:49
	← Technodrome left	19:49
shush	Maybe warn the user of undo that the worktree or index must be empty?	19:49
	I wouldn't know how to do that check...	19:49
	→ thiago joined	19:50
j416	shush: read -p "rly do this [y/n]? " -n 1 -r	19:50
	maybe	19:50
	(and then you check $REPLY)	19:51
	seems like a lot of effort though	19:51
	→ Narrat joined	19:52
	← gast0n left	19:52
	← lacrymology left	19:53
	→ vdamewood joined	19:54
shush	How do I turn a branch name into a symbolic ref?	19:54
napping	Ah, looks like git gc runs git repack with -f only for --aggressive	19:54
	→ wootehfoot joined	19:55
	← epoll left	19:55
j416	shush: do you mean you want the full ref?	19:56
	→ epoll joined	19:56
j416	shush: git rev-parse --symbolic-full-name <branch>	19:56
shush	j416: Well `git parse-rev branch^{:/commit text}` wont work	19:56
j416	shush: try rev-parse instead	19:57
	→ manj-gnome joined	19:57
j416	(also skip the :)	19:58
shush	I miss typed	19:58
	I'm using rev-parse	19:58
j416	then it'll work fine	19:58
	what's the issue?	19:58
	← jl4 left	20:00
	→ nabil joined	20:00
shush	`git rev-parse branch` returns fatal: ambiguous argument 'branch': unknown revision or path not in the working tree.	20:00
	→ TheConjurer joined	20:00
	← arcatech left	20:00
	← manj-gnome left	20:01
thiago	does it exist as abranch?	20:01
j416	shush: does it actually say 'branch', or 'branch^{:/commit text}' ?	20:02
shush	Ah,	20:02
	I was messing up	20:02
	looks like it works	20:02
	← p-core left	20:02
shush	Needed ^{/ after my branch	20:02
	← riksteri left	20:02
j416	..yes.	20:02
shush	I think I had {/ only	20:03
	→ p-core joined	20:03
shush	And also there was no commit matching my text on the branch I was testing	20:03
	→ nabil_ joined	20:03
shush	Btw	20:03
	← napping left	20:04
shush	HEAD^{/text} can be shorter: @^{/text} which is only 3 more characters than :/text	20:04
j416	shush: note that all of this relies on your commits having different subjects. If you have two commits that are "Update docs" or such, it'll only ever find one of them.	20:04
shush	j416: Yes	20:04
	The commit message will always be formatted as "future! branch-name"	20:04
	← nabil left	20:05
	← cliluw left	20:06
shush	This way I can do `git portal upstream downstream` which translates to `git rebase --onto @^{/future! upstream} downstream^{/future! upstream} downstream`	20:06
	Of course to create the "future!" commit, I'd have `git future branch` which will do `git merge -m"future! branch" branch`	20:08
	That's the workflow I'm thinking of using	20:08
	← plexigras left	20:08
	→ nomadicmathmo joined	20:11
	← R2robot left	20:11
	→ vs^ joined	20:11
	← Borkr left	20:13
	← tiin57 left	20:13
	→ tiin57 joined	20:17
	← redcaptrickster left	20:19
	← p-core left	20:20
	← toolbox left	20:20
	→ kupi joined	20:22
	← GankMove left	20:23
	→ jnavila joined	20:23
	← HumanG33k left	20:24
	→ cliluw joined	20:27
	← vdamewood left	20:27
	← cliluw left	20:30
	→ cliluw joined	20:31
	→ rossome joined	20:31
	← cliluw left	20:35
	→ otisolsen70_ joined	20:36
	← Gustavo6046 left	20:37
	carbolymer_ → carbolymer	20:38
	→ p-core joined	20:39
	→ laria joined	20:39
	← otisolsen70 left	20:40
	→ Iarfen joined	20:43
	← tiin57 left	20:43
	→ HumanG33k joined	20:45
	← Narrat left	20:45
	→ erin1983684 joined	20:46
	→ tiin57 joined	20:47
	← erin1983684 left	20:49
	→ erin1983684 joined	20:49
	← erin1983684 left	20:50
	→ Narrat joined	20:51
	← otisolsen70_ left	20:51
	← mellernoia left	20:51
	→ BrunoSpr joined	20:54
	→ codebam__ joined	20:55
	← trinityblade left	20:55
	← codebam_ left	20:57
	→ Gustavo6046 joined	20:58
	← spacesuitdiver left	20:59
	→ justanotheruser joined	20:59
	→ otisolsen70 joined	21:00
	← wnbeser left	21:04
	→ bolovanos joined	21:05
	→ cliluw joined	21:08
	← nomadicmathmo left	21:09
	→ spacesuitdiver joined	21:10
	← Maxattax left	21:11
	← dionysus69 left	21:13
	→ nomadicmathmo joined	21:13
	← tiin57 left	21:14
	→ dionysus69 joined	21:14
	→ jstein joined	21:14
	→ Wetmelon joined	21:15
	← dwmw2_gone left	21:15
	← ghost43 left	21:15
	→ ghost43 joined	21:15
	→ arcatech joined	21:16
	→ tiin57 joined	21:17
	← kalbasit left	21:20
	← aliss left	21:21
	← BrunoSpr left	21:22
	← vs^ left	21:22
	← Gustavo6046 left	21:26
	← Xiti left	21:30
	← Atlenohen left	21:31
	→ Gustavo6046 joined	21:33
	← laria left	21:33
	← Raging_Hog left	21:34
	→ dwmw2_gone joined	21:35
	→ osse joined	21:35
	→ gonzo joined	21:36
	← sord937 left	21:38
	← dwmw2_gone left	21:40
	← rcoote left	21:43
	← tiin57 left	21:44
	→ BrunoSpr joined	21:46
	→ tiin57 joined	21:47
	→ shored1 joined	21:48
	← gonzo left	21:48
	← Sleepy63 left	21:48
	← shush left	21:51
	→ shush joined	21:52
	← Guest20_ left	21:52
	← Wetmelon left	21:52
	→ FFY00 joined	21:53
	← nomadicmathmo left	21:54
	→ trinityblade joined	21:56
	← cfoch left	22:00
	← dionysus69 left	22:01
	→ lofo_ joined	22:01
	← __Aqua_Vitae__ left	22:02
	← TheConjurer left	22:05
	← lofo left	22:05
	→ TheConjurer joined	22:06
	→ tsrt^ joined	22:07
	→ Sleepy63 joined	22:07
	← caveman left	22:07
	← tiin57 left	22:13
	← cliluw left	22:15
	→ dwmw2_gone joined	22:16
	→ tiin57 joined	22:17
	← crose left	22:18
	← jnavila left	22:20
	→ mirrorbird joined	22:21
	← endiruna1 left	22:22
	← mirrorbird left	22:22
	→ gigasu_shida joined	22:22
	→ durham joined	22:23
	← BrunoSpr left	22:25
	← theoceaniscool left	22:29
	← orbyt_v7 left	22:30
	→ gast0n joined	22:35
	← durham left	22:40
	→ durham joined	22:40
	→ cliluw joined	22:41
	← tiin57 left	22:43
	← dwmw2_gone left	22:44
	→ dwmw2_gone joined	22:44
	→ Wetmelon joined	22:45
	← dwmw2_gone left	22:45
	→ dwmw2_gone joined	22:45
	→ tiin57 joined	22:47
	← otisolsen70 left	22:48
	← calcul0n__ left	22:51
	→ devslash joined	22:54
devslash	I did git add . on my repo but meant to add only 1 file	22:55
	how do i undo add .	22:55
zsoc	git reset	22:55
devslash	ok thanks	22:55
	← arcatech left	23:01
ackyshake	git reset -p if you want to unstage selectively, while keeping the file you want, staged.	23:03
	or git reset <file>	23:03
	→ royal_screwup21 joined	23:03
	→ Xiti joined	23:09
	← gigasu_shida left	23:11
	→ n0mn0m[away] joined	23:12
	← luke-jr left	23:12
	→ gigasu_shida joined	23:12
	→ luke-jr joined	23:13
	← n0mn0m[away] left	23:13
	← tiin57 left	23:13
	→ tiin57 joined	23:18
	→ arcatech joined	23:19
	← dwmw2_gone left	23:20
	→ Benzi-Junior joined	23:20
	← jstein left	23:21
	→ dwmw2_gone joined	23:21
	→ sgn joined	23:29
	← sgn1 left	23:30
	← durham left	23:32
	← m0viefreak left	23:37
	→ karvad joined	23:41
	← dudek left	23:41
	← tiin57 left	23:43
	← DiffieHellman left	23:43
	→ DiffieHellman joined	23:45
	→ codebam_ joined	23:46
	→ tiin57 joined	23:47
	→ orbyt_ joined	23:48
	← codebam__ left	23:48
	← royal_screwup21 left	23:51
	→ royal_screwup21 joined	23:51
	→ codebam__ joined	23:51
	← Betal left	23:52
	← herren left	23:53
	← codebam_ left	23:54
	→ Thorn joined	23:54
	→ powerhouse joined	23:54
	→ sudoforge joined	23:57
	→ durham_ joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation