IRCloggy #git 2018-11-05

Provider of IRC logs since 2005.

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

	2018-11-05
	→ Bombe joined	00:00
	← Essadon left	00:04
	→ pity_ joined	00:04
	← finalbeta left	00:04
	← dege left	00:05
	← pity left	00:05
	→ muelleme joined	00:06
	← ixian left	00:07
	→ fstd_ joined	00:09
	← lig left	00:09
	← Bombe left	00:10
	→ mollusk joined	00:11
	→ CodingRacoon joined	00:12
	← fstd left	00:12
	fstd_ → fstd	00:12
	← olalonde left	00:12
	→ lig joined	00:13
	← muelleme left	00:13
	→ ixian joined	00:14
	← kerframil left	00:14
	→ Bombe joined	00:14
	→ kapil____ joined	00:14
	← jlamur left	00:15
	→ kerframil joined	00:17
	→ afuentes joined	00:18
	← thebope left	00:18
	→ weltall joined	00:19
	→ lig1 joined	00:20
	→ boombatower joined	00:21
	← pity_ left	00:23
	← lig left	00:23
	← JoshuaD left	00:24
	← weltall left	00:26
	← Bombe left	00:26
	← RoriconKnight left	00:27
	← dilfridge left	00:27
	← profmac left	00:28
	→ Garbanzo joined	00:29
	← venmx left	00:29
	← t0dd1v left	00:29
	→ sanches joined	00:30
	→ weltall joined	00:31
	→ pionen joined	00:31
	novumsensum → \|	00:34
	← n3wborn left	00:34
	\| → n^	00:34
	← boombatower left	00:35
	→ dilfridge joined	00:36
	← weltall left	00:36
	← monr0e left	00:36
	← sanches left	00:38
	→ Bombe joined	00:40
	n^ → ^\s\zs\w\{-}	00:41
	← pionen left	00:42
	→ weltall joined	00:42
	→ pionen joined	00:42
	^\s\zs\w\{-} → n^	00:43
	← dilfridge left	00:46
	→ profmac joined	00:47
	← CodingRacoon left	00:47
	← Bombe left	00:47
	→ thebope joined	00:48
	← profmac left	00:48
	→ andyhuzhill joined	00:49
	→ profmac joined	00:49
	← Tykling left	00:50
	← pionen left	00:50
	→ Tykling joined	00:50
	← tled left	00:51
	← liefer left	00:54
	← weltall left	01:01
	→ sanches joined	01:05
	→ weltall joined	01:05
	← mlopezviedma left	01:06
	→ JoshuaD joined	01:07
	← justanotheruser left	01:08
	← JoshuaD left	01:09
	→ maheshpec joined	01:09
	← yohnnyjoe left	01:11
	← Tykling left	01:12
	← orbyt_ left	01:12
	→ Bombe joined	01:12
	→ nurdie joined	01:16
	→ cagomez joined	01:17
	← cagomez left	01:18
	← weltall left	01:18
	← Acelogic left	01:19
	→ cagomez joined	01:19
	← nurdie left	01:20
	← thebope left	01:21
	→ justanotheruser joined	01:22
	← jasmith left	01:23
	→ weltall joined	01:23
	→ t0dd1v joined	01:24
	→ Tykling joined	01:24
	→ peacememories joined	01:25
	→ Acelogic joined	01:25
	→ liefer joined	01:26
	← weltall left	01:29
	← Bombe left	01:30
	→ realig joined	01:30
	← lig1 left	01:33
	← lolmac__ left	01:33
	→ lolmac joined	01:34
	← jstein left	01:34
	← LeMike__ left	01:34
	→ weltall joined	01:35
	← LeMike_ left	01:35
	← weltall left	01:36
	→ weltall joined	01:37
	→ libertyprime joined	01:38
	→ orbyt_ joined	01:39
	← weltall left	01:39
	→ weltall joined	01:40
	→ Bombe joined	01:41
	← weltall left	01:42
	← maheshpec left	01:43
	← cagomez left	01:43
	→ cagomez joined	01:44
	← cagomez left	01:44
	→ weltall joined	01:45
	← SpeakerToMeat_ left	01:48
	→ SpeakerToMeat_ joined	01:50
	← ilmostro left	01:50
	← jtrzebiatowski left	01:51
	← Bombe left	01:51
	→ ilmostro joined	01:52
	← mollusk left	01:52
	← weltall left	01:54
	← fassl left	01:55
	→ thebope joined	01:56
	→ mollusk joined	01:58
	← jakogut_ left	01:58
	→ esrse joined	02:00
	→ Bombe joined	02:00
	→ weltall joined	02:01
	→ prakashdanish joined	02:02
	← jetchisel left	02:02
	← peacememories left	02:03
	→ finalbeta joined	02:03
	→ igemnace joined	02:03
	→ nurdie joined	02:04
	→ jetchisel joined	02:07
	← weltall left	02:07
	→ cagomez joined	02:09
	← Bombe left	02:12
	← nurdie left	02:15
	→ parallel21 joined	02:16
	→ Bombe joined	02:17
	← cagomez left	02:19
	→ cagomez joined	02:20
	← mollusk left	02:21
	→ weltall joined	02:23
	← Bombe left	02:23
	← prakashdanish left	02:24
	→ xco joined	02:24
xco	hi. how do i checkout to a branch that’s a PR on my upstream remote?	02:25
	i have two remotes; origin and upstream	02:25
	on upstream someone submitted a PR i want to pull down to check	02:25
	if it were origin (the one i cloned) i’d just do something like ‘git checkout the-remote-branch-for-the-pr	02:26
	but now that it’s on a different remote i’m not sure how to checkout into someone’s pr	02:26
	any ideas?	02:27
	← thebope left	02:29
	→ Bombe joined	02:30
orbyt_	git fetch remote_name && git checkout -b some_branch remote_name/some_branch	02:32
	or...something like that	02:32
	← m0viefreak left	02:33
	← mofai left	02:35
	→ nurdie joined	02:36
	McKraken_zzzzz → McKraken	02:39
	← foul_owl left	02:41
	→ nurdie_ joined	02:41
	← nurdie_ left	02:42
	← F0rTex left	02:43
	→ F0rTex joined	02:45
	← nurdie left	02:45
	← archergodson left	02:46
	← Mattiaslndstrm left	02:49
	← Bombe left	02:49
	→ nurdie joined	02:50
	→ nurdie_ joined	02:51
	→ prakashdanish joined	02:54
	← nurdie left	02:54
	→ foul_owl joined	02:56
	→ thebope joined	02:57
	→ mollusk joined	02:59
	→ Bombe joined	03:03
	← mollusk left	03:04
	← nurdie_ left	03:07
	← Bombe left	03:08
	→ nurdie joined	03:08
	← xco left	03:09
	← sanches left	03:10
	→ Bombe joined	03:11
	← nurdie left	03:12
	← wreed left	03:15
	← jetchisel left	03:15
	→ cfoch joined	03:16
	→ jetchisel joined	03:17
	→ wreed joined	03:17
	← nephyrin left	03:19
	← Bombe left	03:20
	→ Bombe joined	03:20
	← CheckDavid left	03:24
	→ mollusk joined	03:24
	→ Toadisattva1 joined	03:25
	← Toadisattva left	03:26
	← thebope left	03:30
	→ nurdie joined	03:33
	→ g00s joined	03:37
	← nurdie left	03:37
	← Bombe left	03:43
	→ Bombe joined	03:45
	→ sanches joined	03:46
	← Toadisattva1 left	03:46
	← sanches left	03:50
	→ monr0e joined	03:51
	→ nurdie joined	03:53
	→ stuser joined	03:53
	← liefer left	03:54
	→ liefer joined	03:55
	lagothrix → Guest83591	03:56
	← Guest83591 left	03:56
	→ lagothrix joined	03:56
stuser	I'm looking to read the documentation of a project when I'm offline, but I don't need all of the code from the actual project. Is there a way to clone a repo, remove files I don't need, then track commits to the remaining files?	03:58
thiago	no	03:59
	← dka left	03:59
	← nurdie left	04:00
	← Bombe left	04:00
	→ howdoi joined	04:01
	→ dka joined	04:01
stuser	darn. I was hoping there was some low-level mechanism for doing such a thing	04:03
	← mrCyborg\|zzz left	04:05
	← monr0e left	04:05
	→ mrCyborg\|zzz joined	04:05
	→ thebope joined	04:06
	→ Bombe joined	04:07
	← n^ left	04:08
	← cagomez left	04:12
	← timj left	04:18
hendry	how do i checkout a file i deleted in the last commit? `git checkout .env.sample HEAD~1` doesn't work	04:19
	← Bombe left	04:19
	← SJr left	04:20
hendry	ah; git checkout HEAD~1 -- .env.sample # works	04:21
	→ SJr joined	04:21
	← andyhuzhill left	04:23
	→ andyhuzhill joined	04:23
	→ [[]] joined	04:24
	→ timj joined	04:32
	→ matsaman joined	04:32
	← matsaman left	04:33
	← netj left	04:36
	→ netj joined	04:36
	→ nurdie joined	04:36
	← thebope left	04:39
	→ Bombe joined	04:39
	→ sanches joined	04:39
ayjay_t	stuser: who is the git remote server?	04:40
	stuser: and are you an admin on it?	04:40
	← nurdie left	04:41
	← mooasaurus left	04:42
	→ mooasaurus joined	04:42
	← sanches left	04:43
	→ Goplat joined	04:47
	← Bombe left	04:48
	← t0dd1v left	04:49
stuser	ayjay_t: various projects (some github hosted, some not), none of which I have admin to. It was just an idea to read project docs on my tablet during long trips without wifi, then git pull when I was able to connect. I was hoping to save space by not pulling the source code- just the documentation	04:50
	← prakashdanish left	04:52
	→ pks_ joined	05:01
	← pks left	05:02
	pks_ → pks	05:02
	→ prakashdanish joined	05:07
	→ VaultTec joined	05:07
	← libertyprime left	05:09
	← Flaim left	05:10
	→ thebope joined	05:11
	→ jasmith joined	05:15
	← TheRuckus left	05:25
	→ libertyprime joined	05:26
	→ crimastergogo joined	05:29
	→ Bombe joined	05:30
stuser	If I have a nested submodule (i.e. superproject, submodule of superproject, and submodule of that submodule) will "git submodule update" update the sub-sub-module?	05:37
	→ miczac joined	05:37
	← libertyprime left	05:38
	→ Kaisyu joined	05:39
	← andyhuzhill left	05:39
	→ andyhuzhill_ joined	05:39
	andyhuzhill_ → andyhuzhill	05:40
	← Bombe left	05:42
	→ jiji606 joined	05:43
	← thebope left	05:45
	← jim left	05:46
	→ jim joined	05:47
	← _xor left	05:48
	→ _xor joined	05:48
bn_work	what was the git command to split apart multiple changes from staged files (index?) into separate commits?	05:52
	git 2.13.6	05:52
	→ Bombe joined	05:53
	← xcm left	06:00
	→ xcm joined	06:00
	← realig left	06:00
	→ Mattiaslndstrm joined	06:01
	← stuser left	06:07
	→ miczac_ joined	06:08
	← miczac left	06:08
bn_work	was it `git add --patch` or `git add -i`?	06:09
igemnace	bn_work: --patch splits apart changes from UNstaged files, then stages the hunks you want	06:10
	you can get to patch from add -i as well	06:10
	← mooasaurus left	06:13
	→ mooasaurus joined	06:13
	← ferdna left	06:14
	→ neunon_ joined	06:15
	← neunon left	06:15
	→ sanches joined	06:15
	neunon_ → neunon	06:15
	→ thebope joined	06:18
	← prakashdanish left	06:18
	← sanches left	06:20
	→ prakashdanish joined	06:21
	← Arokh left	06:25
	→ libertyprime joined	06:26
	← igemnace left	06:28
	← vjacob_ left	06:28
	→ vjacob_ joined	06:28
	→ prakashd1 joined	06:30
	← prakashdanish left	06:32
	← mollusk left	06:32
	← GreenJello left	06:32
	→ igemnace joined	06:32
	→ GreenJello joined	06:33
	← miklcct left	06:33
	← thiago left	06:34
	← miczac_ left	06:34
	→ miklcct joined	06:35
	← kerframil left	06:39
	← Bobdude left	06:41
	← orbyt_ left	06:41
	← kanzure left	06:42
	→ gooooootime joined	06:44
	→ pity joined	06:44
	→ sQVe joined	06:45
	← dg_vrani left	06:46
	← thebope left	06:52
	← Acelogic left	06:59
	→ oxymoron93 joined	07:00
	← djapo left	07:02
	→ nydel09 joined	07:06
	← Goplat left	07:09
	← donofrio left	07:11
	→ shabius joined	07:13
	→ dsx_ joined	07:14
	→ dreiss joined	07:14
	→ yyy joined	07:14
	← kapil____ left	07:14
	← dsx_ left	07:16
	→ st-gourichon-fid joined	07:18
	→ pawle joined	07:18
	← sQVe left	07:21
	← VaultTec left	07:21
	→ chele joined	07:22
	→ thebope joined	07:23
	← st-gourichon-fid left	07:23
	→ st-gourichon-fid joined	07:24
	← dreiss left	07:29
	→ m_ joined	07:29
	← Mat001 left	07:32
	→ jlamur joined	07:36
	→ VaultTec joined	07:39
	← Garbanzo left	07:40
	← ExoUNX left	07:40
	← kent\n left	07:40
	← Riviera left	07:41
	← g00s left	07:41
	← akushner left	07:45
	→ dilfridge joined	07:46
	→ PEG joined	07:52
	→ muelleme joined	07:54
	← crimastergogo left	07:55
	← yyy left	07:55
	← thebope left	07:55
	→ crimastergogo joined	07:56
	→ sanches joined	07:58
	→ mollusk joined	07:58
	→ Jackneill joined	07:58
	← muelleme left	07:59
	← sanches left	08:03
	→ thebope joined	08:06
	→ n3wborn joined	08:07
	← thebope left	08:10
	→ venmx joined	08:10
	→ g00s joined	08:11
	→ mona_x joined	08:15
	← mollusk left	08:17
	→ rsrx joined	08:18
	← crimastergogo left	08:20
	→ crimastergogo joined	08:21
	→ dege joined	08:22
	→ sQVe joined	08:22
	← m_ left	08:27
	→ Noti joined	08:29
	← jlamur left	08:29
	← dege left	08:29
	→ dege joined	08:30
	→ plexigras joined	08:30
	← Tobbi_ left	08:31
	← veegee left	08:32
	→ Tobbi_ joined	08:33
	← pawle left	08:37
	← prakashd1 left	08:37
	→ mofai joined	08:41
	→ sylario joined	08:41
	← oxymoron93 left	08:41
	_wrksx → wrksx	08:41
	← g00s left	08:43
	← staafl_ left	08:43
	← austin987 left	08:43
	→ staafl_ joined	08:43
	→ thebope joined	08:46
	→ doubtful joined	08:49
	→ oxymoron93 joined	08:50
	← afuentes left	08:56
	→ tomfarr joined	09:01
	→ ZyX-I joined	09:03
	← Impaloo left	09:04
	→ Impaloo joined	09:05
	→ eduardas_m joined	09:07
	→ floppydh joined	09:08
	← Impaloo left	09:09
	→ Impaloo joined	09:10
	→ prakashd1 joined	09:12
jim	bn_work, if you have committing to do, you can add just what you want per commit, using git add -e and in this variant, you're given all the changes you'd see in git diff, and you're literally editing the diff itself	09:12
	→ fairuz joined	09:12
dcpc007	hi, i add another remote on my local git. i fetch updates from this one (same base history, but newer branches/commits)	09:12
	now i do a git checkout remote2/master and it warns me about detached head !	09:13
	think it's not what should happens	09:13
oxymoron93	dcpc007: that happens when you checkout remote tracking branch	09:13
	!detached	09:14
gitinfo	A detached HEAD (aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	09:14
dcpc007	ha ... don't remember this :-(	09:14
	janx\|away → janx	09:15
dcpc007	yes but from memory when i git co remote2/branch, he automatically add the track locally too	09:15
	and in the prompt i got the granch name	09:15
	then i "only" have to add the tracking manually this time	09:15
	← fjasle left	09:16
dcpc007	(boring to have to maintain 2 different central repo !)	09:16
oxymoron93	you do not remember seei .. oh wait wrong line.. you don't remember well, you used `git checkout <branch>` and git automatically created local branch with tracking info	09:16
dcpc007	oxymoron93, yes for me it should, but not the case	09:17
	i check the .git/config, i have the remote2 configured but no branch tracking	09:17
	i have 3 blocks : remote for origin, branch master for origin, remote for repo2	09:18
	← thebope left	09:19
oxymoron93	hm... git checkout <branch> does not create local <branch> when you already have <remote>/<branch>, is that what you are saying? then there is something more I don't know , and probably involves having more than one remote repository	09:20
	← crimastergogo left	09:21
	← Noti left	09:21
dcpc007	ha ok it works for a different branch, but not for remote2/master because already a local master tracking origin/master	09:21
	if i do git co remote2/branch1 it works !	09:21
	branch "branch1" is set to track "remote2/branch1"	09:22
	if i try to force like 'git co --track remote2/master' i got	09:22
	→ crimastergogo joined	09:23
dcpc007	fatal : a branch 'master' already exists	09:23
oxymoron93	well as expected	09:23
dcpc007	i don't remember how i do that before	09:23
	lol ! i remember	09:23
	→ sanches joined	09:23
oxymoron93	don't do it, you should only have one `master` which works with both repositories, but default is origin	09:23
	if you wan't to have explicit new branch which tracks repo2/master	09:24
dcpc007	my second remote is fir gitlab, and asi use protected branches, i was never working in the master :)	09:24
	then think it should be like that that since start, but never try to git co remote2/master maybe	09:24
	→ elsheepo joined	09:25
dcpc007	but strange i never do that check before ... i work like this sonce more than 1 year	09:25
	← cfoch left	09:25
dcpc007	(as project of migrate gitolite -> gitlab is in stand-by ...)	09:25
	→ archergodson joined	09:25
	→ mollusk joined	09:25
dcpc007	my workflow is : working solo in gitlab with issues and MR, when all is finished and branch merged, i push history to origin	09:26
	to maintain both sources uptodate	09:26
	to keep	09:26
	maybe i can name my remote2/master in a local "remote2_master" ?	09:27
oxymoron93	yeah, was about to tell you that, if that suits you	09:28
	I really have no idea what your workflow is, but if it works for you then yeah	09:28
	← sanches left	09:28
	← crimastergogo left	09:28
dcpc007	boring workflow, ... need to update all works 2 times	09:29
	only me on gitlab	09:29
	all users are still on gitolite	09:29
	but need to works with gitlab to learn it and try implement it for all users ...	09:30
	→ cd joined	09:31
	← andyhuzhill left	09:32
	← mollusk left	09:32
	→ psprint_ joined	09:33
	← [[]] left	09:34
	→ crimastergogo joined	09:35
	→ nakkht joined	09:36
	← staafl_ left	09:36
	← sQVe left	09:37
	→ sQVe joined	09:37
	← nakkht left	09:37
	→ nakkht joined	09:37
	→ staafl_ joined	09:39
	← psprint_ left	09:41
	→ zulutango joined	09:43
	← elsheepo left	09:45
	← nakkht left	09:45
	← dpyro left	09:48
	→ thebope joined	09:49
	← justanotheruser left	09:49
	→ PsyStemUpdate joined	09:50
PsyStemUpdate	does anyone else here manager a git server using one git user through ssh but with different public private keypairs all representing different users? if so, how do you guys tell who is working on what if everyone uses the same user?	09:51
	→ elsheepo joined	09:51
	→ Kulrak joined	09:53
PsyStemUpdate	or do you guys create a user on the linux server for each person with access to only one folder which has one git repo mounted to that mountpoint?	09:53
	many users all with the same target mounted to a folder(the repo folder) they are stuck in?	09:54
grawity	PsyStemUpdate: for more than 1 user, let Gitolite deal with it	09:55
	it works out who's connecting to git@ based on extra parameters in the authorized_keys file	09:55
PsyStemUpdate	havent heard of it, ill look into it. before gitolite came along though, was this the solution(all users having a mountpoint to the same target?or maybe a hard link)?	09:58
	→ rkta joined	09:59
	→ realig joined	10:00
	→ Raimondii joined	10:00
grawity	before gitolite there was gitosis	10:01
	before gitosis, I dunno, just regular file permissions	10:02
	→ bsaboia joined	10:02
	realig → lig	10:02
grawity	not sure why you'd need each user to have a mountpoint – they can all access the same path directly	10:02
	as long as they're not chrooted, that is	10:02
	← Raimondi left	10:03
	Raimondii → Raimondi	10:03
	→ justanotheruser joined	10:03
	← tomfarr left	10:09
	→ steven_ joined	10:11
	MrSpooky → MrC	10:11
	← raven-au left	10:11
	← steven left	10:12
dcpc007	psYchotic, same this is how gitolite works. but i'm switching to gitlab soon	10:12
	more easy to manage and a lot more complete :)	10:12
	→ nurdie joined	10:12
	← gooooootime left	10:13
psYchotic	PsyStemUpdate: I think that ^ was meant for you =)	10:13
dcpc007	and i start have problems with some cases like a user using 2 ssh keys, i can't affect both to the same user ... then i need to create fake users like "user_bis" and duplicate the access rights	10:13
	with the common next problem, forget to update both account when modifying rights	10:13
	but if you have a very few people, and repo, and basics access rules, and no need of nothing unless a basic share git (without the many functionalities possible), gitolite is simple and ok	10:15
	← sunri5e left	10:15
	→ sunri5e joined	10:15
dcpc007	psYchotic, oups yes :)))	10:15
	← nurdie left	10:17
	← catsup left	10:17
	→ catsup joined	10:18
	← catsup left	10:18
	→ catsup joined	10:18
	← genius_monkey left	10:18
	← lig left	10:20
	← AWindowsKrill left	10:22
	← thebope left	10:23
	→ nowhere_man joined	10:23
PsyStemUpdate	grawity ya my chroot was the problem.	10:24
	→ AWindowsKrill joined	10:24
	← mona_x left	10:24
PsyStemUpdate	it made me have to have a folder put in each of their jail folders but copying the repo wasnt an option	10:24
	so i thought of hard linking or mounting which i havent done before, im not even sure u can have a target mounted to several mountpoints at once.	10:25
grawity	`mount --bind` if you mean Linux	10:32
	→ MrJones joined	10:37
MrJones	hi	10:37
	← crimastergogo left	10:37
MrJones	I'm having trouble checking out the master branch of a remote to a local branch of another name	10:37
	$ git checkout -b tlsfix tlsfix/master < complains that 'tlsfix/master' is not a commit. isn't the syntax <remote>/<branch>? I'm not sure what I'm doing wrong	10:37
	(the remote is named 'tlsfix', and I want to have its master as local branch 'tlsfix')	10:37
	← Luyin left	10:38
	← cdown left	10:38
	← n3wborn left	10:39
selckin	does it show it in' git branch -a' ?	10:40
	steven_ → steven	10:40
MrJones	selckin: interesting, it doesn't :O I did git fetch tlsfix though	10:40
	no branches of that origin are listed there, I don't know why	10:41
	← flippedouttt left	10:41
MrJones	(other origins are listed fine)	10:41
selckin	and fetch doesn't error?	10:41
	→ kerframil joined	10:41
	← kerframil left	10:42
MrJones	selckin: no, gives zero output and exit code is 0	10:42
	← doubtful left	10:42
selckin	how did you add the remote? you can configure to limit the branches you want etc	10:42
	→ crimastergogo joined	10:43
	→ jtrzebiatowski joined	10:44
MrJones	selckin: ah nice, bingo. had the branches set up to be added as remote/origin/ which of course doesn't make sense, now it works	10:44
	← Raimondi left	10:44
	← esrse left	10:46
dcpc007	i forget how to import with history a small repo inside a main repo (with no duplicated path/data)	10:48
	i have both repo locally, and the small to import = one folder + 2 files (easy case this time)	10:49
	i remember an option to not frget like --allow-external-xxx or similar (boring, not written the process)	10:50
	(i need to find a centralized way to store all those mini procedures)	10:50
	← hussam left	10:55
	→ LondonAppDev joined	10:55
	→ thebope joined	10:57
	→ pawle joined	10:58
	→ gooooootime joined	10:59
	← PsyStemUpdate left	10:59
	← lankanmon left	11:00
	← vvor left	11:00
	← LondonAppDev left	11:06
	→ sanches joined	11:10
dcpc007	found, seems ok, i try this : https://thoughts.t37.net/merging-2-different-git-repositories-without-losing-your-history-de7a06bba804	11:10
	esran_ → esran	11:15
	← sanches left	11:15
	→ LondonAppDev joined	11:17
	→ cdown joined	11:17
	← fairuz left	11:19
	→ veegee joined	11:21
	→ mona_x joined	11:25
	→ halide joined	11:31
	← thebope left	11:31
	← staafl_ left	11:31
	← elsheepo left	11:34
	← rkta left	11:37
	→ jlamur joined	11:37
	→ hussam joined	11:40
	← gooooootime left	11:41
	→ psprint_ joined	11:43
	← PEG left	11:46
	→ rockyh joined	11:51
	→ staafl_ joined	11:52
rockyh	Hi!	11:53
	I tried to make a git clone from a remote repository and I noticed that some directories are (in the web repository page) named `name @ <7_characters_hex_value>'. These directories are created in my local clone, but they are empty, while in the web repository they are not. What could be the reason? I would like to clone them as well as the other ones	11:55
cbreak	rockyh: git ls-tree -R	11:58
	→ thebope joined	11:59
cbreak	rockyh: if you see "commit" entries, you'll have to clone --recursive.	11:59
	or later git submodule update --init	11:59
	→ AfroThundr\|main joined	12:02
rockyh	cbreak: `-R' gives `error: unknown switch'	12:03
_ikke_	it's -r	12:03
	small	12:03
cbreak	hmm...	12:03
	so inconsistent	12:03
	← AfroThundr\|main left	12:03
_ikke_	with what?	12:03
	← jlamur left	12:04
cbreak	ls, cp, and so on	12:05
	← staafl_ left	12:05
rockyh	thanks for the clarification. Should I run `git ls-tree -r' inside the already cloned repository?	12:07
_ikke_	git ls-tree -r HEAD	12:07
	→ staafl_ joined	12:08
cbreak	yes, inside the repository	12:08
	it'll list your chosen tree, recursively	12:08
	if there are commit entries, that means you have submodules	12:08
	and you'll have to init and update them (if you want them)	12:08
	→ Essadon joined	12:09
	→ lankanmon joined	12:10
	← halide left	12:11
	← xcm left	12:11
	→ xcm joined	12:12
rockyh	thanks for `HEAD' and for all the suggestions. There are several `commit' entries. Following your first messages, I ran `git submodule update --init' and they have been cloned	12:12
	(Yes, I wanted those submodules)	12:12
	→ revoltingPeasant joined	12:13
	← eset left	12:14
	→ rafasc joined	12:15
	← Kulrak left	12:17
	← LondonAppDev left	12:18
	→ LondonAppDev joined	12:19
	→ bolovanos joined	12:19
	← crimastergogo left	12:20
	→ eset joined	12:20
	← igemnace left	12:21
	→ yorick_ joined	12:25
yorick_	Hi. I want to merge multiple modules / subprojects with one core project (both with versioning via tags); but what is the best way to do this? I've read something about sub-trees and submodules. In case the modules / subprojects which need to be merged with the core also need to be merged with some files from the core (so not in a sub dir); I'd read that submodules isn't the right options. Is sub-tree the right option to do this?	12:28
cbreak	submodules track hashes, not tags	12:29
yorick_	And what if merging the core and subprojects doesn't need to be a separated repository; is it possible to merge these repositories with composer?	12:30
cbreak	you can try submodules, it will work, but it won't be without work	12:30
	what's composer?	12:30
yorick_	the composer file which sometime will be sed for versioning of projects (what i'd see a lot @ github). I'm not very fimiliar with it tbh.	12:31
	← thebope left	12:32
cbreak	never heard of that.	12:32
	if you want submodules, git supports them	12:32
	as said, the binding to them works via commit hash	12:32
	but I think there are ways to make submodules pull from branches via merging, rebasing or similar	12:32
	if you want to completely absorb a repository, git supports that too	12:33
	but then, pushing back is probably super annoying	12:33
	→ Noti joined	12:33
yorick_	but you can't use tags to pull specific versions of a module into the project?	12:34
	← xcm left	12:34
	→ xcm joined	12:35
_ikke_	You can checkout a tag and then record that hash in the parent repo	12:37
	cbreak: composer is a package manager for PHP	12:37
cbreak	so not git related.	12:38
_ikke_	no	12:38
cbreak	good... :)	12:38
_ikke_	Though if you use composer, there is little need for submodules (at least for things managed with composer0	12:39
	← prakashd1 left	12:40
	← triad left	12:41
	→ thebope joined	12:42
yorick_	oky, then i will read more about submodules. To be sure I'd understand it correctly; it's possible to merge two repositories into one repository with submodules?	12:42
_ikke_	no	12:42
	They remain seperate repositories	12:43
yorick_	so they both need their own directory within the new repo?	12:43
	← oxymoron93 left	12:44
	→ shabius_ joined	12:44
	→ oxymoron93 joined	12:44
	← shabius left	12:46
_ikke_	What do you mean?	12:46
	← thebope left	12:47
	→ CodeSlingerPaul joined	12:48
	→ PEG joined	12:51
	← xcm left	12:51
	→ xcm joined	12:52
	← mona_x left	12:53
	→ mona_x joined	12:54
	← mona_x left	12:55
	← PioneerAxon left	12:55
	← cd left	12:56
cbreak	yorick_: a submodule is exactly one repository in exactly one place in the parent repository	12:57
	if you want two submodules, they'll better have different paths.	12:57
	→ PioneerAxon joined	13:00
	← LondonAppDev left	13:03
	→ cd joined	13:04
	← random_yanek left	13:05
	→ monr0e joined	13:10
	→ wildermind joined	13:11
	← netj left	13:14
	→ netj joined	13:14
	→ thebope joined	13:14
	← xcm left	13:17
	→ xcm joined	13:18
	→ random_yanek joined	13:18
	← tonyj left	13:22
	← rasb left	13:25
	→ TheRuckus joined	13:26
	→ yyy joined	13:26
	← rockyh left	13:26
	→ Raimondi joined	13:40
	→ fairuz joined	13:41
	← dege left	13:41
	→ nurdie joined	13:45
	→ mona_x joined	13:47
	← netj left	13:47
	→ netj joined	13:48
	← veegee left	13:49
	← thebope left	13:49
	← nurdie left	13:49
	← GreenJello left	13:50
	← cdown left	13:51
	→ cdown joined	13:52
	→ freeman42x joined	13:58
	← jtrzebiatowski left	13:59
	→ gooooootime joined	14:01
	→ LondonAppDev joined	14:03
	← yyy left	14:03
	→ yyy joined	14:04
	← Mattiaslndstrm left	14:06
	← yyy left	14:13
	→ LondonAp_ joined	14:13
	← plasticle left	14:15
	← xcm left	14:15
	→ Zexaron joined	14:15
	→ plasticle joined	14:16
	→ Mattiaslndstrm joined	14:16
	← ilmostro left	14:16
	← LondonAppDev left	14:17
	← Noti left	14:18
	→ ilmostro joined	14:19
	← newnick left	14:20
	→ sanches joined	14:21
	→ xcm joined	14:21
	→ rnat joined	14:23
	→ Mat001 joined	14:23
	→ thebope joined	14:24
	← fairuz left	14:26
	← sanches left	14:26
	→ kapil____ joined	14:28
	← Mattiaslndstrm left	14:29
	→ apeiros_ joined	14:30
apeiros_	hi there. I used `git tag --sort=committerdate` to get tags sorted by date. but now we have some tags which are annotated, and they do not have a commiterdate but instead use taggerdate. how can I get a list of all tags sorted correctly?	14:32
	i.e. --sort=taggerdate gets unannotated tags out of order, --sort=committerdate gets annotated tags out of order - I look for a way to get all of them in order	14:33
	→ yyy joined	14:34
	→ kanzure joined	14:38
	← PEG left	14:38
	← VaultTec left	14:42
	→ thebetrayer joined	14:42
rafasc	apeiros_: try creatordate	14:42
	→ donofrio joined	14:42
	← dilfridge left	14:43
	← monr0e left	14:44
rafasc	or: git for-each-ref --format="%(refname:short)" refs/tags --sort=creatordate	14:45
	→ jlamur joined	14:45
apeiros_	rafasc: hm, I guess in our case that should be indeed the same as taggerdate \|\| committerdate	14:46
	or close enough at least	14:46
	apeiros_ → apeiros	14:46
apeiros	thanks	14:46
rafasc	apeiros: creator date is an abstraction to deal with your exact problem.	14:46
apeiros	my other approach would have been to emit both dates as strict iso and run it through a script which sorts/formats it	14:46
	oh? interesting. are all the different types of dates, how they come to be and their purposes somewhere concisely listed in the git docs?	14:47
	← TinkerT left	14:48
rafasc	apeiros: the creatordate is mentioned in git for-each-ref;	14:48
	← revoltingPeasant left	14:52
	← thebetrayer left	14:54
apeiros	ok, I see it's exactly as you say. thanks a lot.	14:56
rafasc	apeiros: annotated tags and lightweight tags are stored using different object types. Lightweight tags are just simple pointers to commits, there's no extra information stored. So the 'author/date' information you get is the information from the commit being tagged.	14:56
	→ jtrzebiatowski joined	14:56
apeiros	thanks for the explanation. I actually understood the cause for the issue, I just didn't know the solution :)	14:57
rafasc	annotated tags, have their own 'object' that is stored.	14:57
apeiros	sob, apparently `git tag --format` does not accept %C(color) like git log --format does :(	14:57
	← plasticle left	14:58
	apeiros throws hands in the air -- well, properly sorted must suffice for the day :D	14:58
	→ YuGiOhJCJ joined	14:58
osse	apeiros: have you tried --color	14:58
	← thebope left	14:58
	← LondonAp_ left	14:59
apeiros	osse: na, it's apparently %(color:<colorname>)	15:00
rafasc	apeiros: log --format and for-each-ref format use different languages.	15:01
	← yyy left	15:01
	→ LondonAppDev joined	15:02
	→ yyy joined	15:02
rafasc	I've read on the ML that ideally these two should converge to a unified thing, but when that will happen is TBD.	15:02
	→ plasticle joined	15:02
apeiros	I can imagine that it's not on the top of the list. also because it probably raises the question of "how much old code relying on the 'old' format are we willing to break?"	15:04
	→ relipse joined	15:04
rafasc	apeiros: --format='%(color:yellow bold)tag: %(refname:short)'	15:04
	apeiros: definitely not on top.. But I am not sure that's the main issue. The main issue is having someone willing to get his hand dirty.	15:05
	← mofai left	15:06
rafasc	the regular devs are busy with more complex things.	15:07
	→ Raed\|Laptop joined	15:08
apeiros	:)	15:08
	→ temhawk joined	15:08
rafasc	hash transition and changes to the protocol to allow partial clones, on top of the usual fix the existing bugs. :)	15:09
	← oxymoron93 left	15:10
	→ muelleme joined	15:10
rafasc	there's also the commit-graph series that promises some nice performance improvements for things that deal with walking the commit history.	15:11
temhawk	How do I reset a file to the last commit? git-reset refuses me, git-checkout seems to be for whole branches/commits, git-clean is for removing untracked files, git-revert is for remote/public stuff, git-rm is for un-tracking files. I'm confused...	15:11
rafasc	temhawk: you can use checkout with files: git checkout HEAD -- filename; <- This will discard any uncommited changes!!	15:12
temhawk	rafasc: ok, thanks mate!	15:12
rafasc	just be aware that mode is one of the few ways you can lose things without being able to recover.	15:13
	that will discard any changes in 'filename' without asking for confirmation.	15:13
temhawk	yup, that's fine	15:14
rafasc	usually not an issue for things tracked, because you have ways to recover, but untracked changes get nuked.	15:14
	← muelleme left	15:15
	← Mir4g3 left	15:16
	→ bsanford joined	15:18
	← sQVe left	15:22
	→ D630 joined	15:24
	→ Mir4g3 joined	15:25
	← bsanford left	15:26
	→ thebope joined	15:27
	← mona_x left	15:28
	→ mona_x joined	15:29
	→ igemnace joined	15:36
	← jwest left	15:41
	← ilmostro left	15:41
	→ jwest joined	15:43
	→ ilmostro joined	15:43
	← Raed\|Laptop left	15:44
	← yorick_ left	15:45
	← pawle left	15:45
	← ilmostro left	15:47
	→ Dougie187 joined	15:48
	→ ilmostro joined	15:49
	→ tomfarr joined	15:49
	→ GreenJello joined	15:51
	← GreenJello left	15:51
	→ czart joined	15:54
	← apeiros left	15:55
	McKraken → McKraken_zzzzz	15:55
	→ AbleBacon joined	15:57
	← ilmostro left	15:58
	→ thiago joined	15:58
	← thebope left	15:59
	→ ilmostro joined	16:00
	← temhawk left	16:01
	← SpeakerToMeat_ left	16:05
	→ dpyro joined	16:06
	→ pawle joined	16:07
	→ SpeakerToMeat_ joined	16:07
	← SpeakerToMeat_ left	16:07
	→ spl33n joined	16:08
	→ SpeakerToMeat_ joined	16:08
spl33n	hello all can i give access on a private gitlab repo some people for only see commit ?	16:08
dzho	yeah probably	16:09
spl33n	dzho: do you knows how can i do this ? please	16:09
dzho	well, are you on a private gitlab instance, or gitlab.com's instance?	16:11
spl33n	dzho: i have a private instance	16:11
dzho	so, for starters, if I knew exactly, I probably wouldn't have said "probably" I'd have just said yes.	16:12
teward	spl33n: I answered you in #gitlab	16:12
dzho	and though I've worked with ...	16:12
	aha	16:12
	teward: thanks	16:13
teward	crossposting is a pain in the butt so please don't do that :P	16:13
	dzho: yep.	16:13
spl33n	teward: i see, but i must let repo private :(	16:14
teward	dzho: TL;DR: No you can't give access in GitLab private on-prem hosted to see just a single commit, you'd have to give readonly to the entire repository to give them that level of access	16:14
	spl33n: then you have a mutually exclusive set of circumstances	16:14
	a discussion which should be taken up with the GitLab developers	16:14
	← SpeakerToMeat_ left	16:14
teward	about permissions/features changes	16:14
	(but that's less about git and more about GitLab's UI and permissions implementations, so #gitlab is probably where the convo should continue)	16:14
	← YuGiOhJCJ left	16:14
	→ thebope joined	16:15
teward	spl33n: if the repo must remain private, then you can't give users access to see the single commit. If you want to share the single commit directly from your GitLab, then the repo cannot remain private, or you must give access to the specific users in readonly mode to see commits.	16:15
	→ prakashd1 joined	16:15
	→ SpeakerToMeat_ joined	16:15
teward	spl33n: alternatively, take the unified diff of your specific single commit and copy it out to somewhere non-private	16:15
	that way they can only see the exact commit/change you pointed at	16:16
	but you won't eb able to do what you're looking to with just GitLab and its existing permissions schema	16:16
	even if youw ere an Enterprise grade customer.	16:16
	but as I said earlier, that's less 'git' related and more GitLab related	16:16
	and you should really be asking GitLab Permisisons related questions in #gitlab and not #git	16:16
	dzho nods	16:17
	← SpeakerToMeat_ left	16:17
dzho	I should probably put a little more effort into staying in #gitlab, speaking of	16:17
teward	drags dzho into #gitlab, then anchors them there	16:17
	:P	16:17
	gitinfo set mode: +v	16:18
	→ SpeakerToMeat_ joined	16:18
	← yyy left	16:20
	→ yyy joined	16:20
	← SpeakerToMeat_ left	16:20
	→ SpeakerToMeat_ joined	16:21
dzho	hah	16:23
	→ rkta joined	16:25
	← chele left	16:26
teward	dzho: I think I did a good job summarizing my minor monologue here in #gitlab :P	16:28
dzho	cool	16:29
	← Mir4g3 left	16:29
	→ voolik joined	16:35
	← prakashd1 left	16:35
	← eduardas_m left	16:38
	→ Envil joined	16:39
	→ SmashingX joined	16:39
SmashingX	I would like to pull from the repo with "git pull" but I get this error: error: cannot open .git/FETCH_HEAD: Permission denied	16:40
	I noticed that the file is owned by jenkins, -rw-r--r--. 1 jenkins jenkins 0 Nov 5 09:35 FETCH_HEAD	16:41
	← bsaboia left	16:41
	→ lig joined	16:42
	← howdoi left	16:42
	← SpeakerToMeat_ left	16:43
	← LondonAppDev left	16:44
	→ SpeakerToMeat_ joined	16:45
	→ durham joined	16:48
	→ LondonAppDev joined	16:48
	← SpeakerToMeat_ left	16:50
	→ jorvis joined	16:51
	→ SpeakerToMeat_ joined	16:52
jorvis	I've inherited a project where a git directory on a server runs production code, and development continued for months with commits to GitHub outside of it. Is there a way to see from within the server's code directory where it is in the timeline of the trunk so I can branch it there?	16:53
	← Tobbi_ left	16:53
jorvis	That is, I want to ask of the directory "Where are you in relation to the commit history of the master?"	16:53
	← SpeakerToMeat_ left	16:53
	→ SpeakerToMeat_ joined	16:56
	→ dege joined	16:56
	← lig left	16:57
	→ comptroller joined	16:58
	→ wildlander joined	16:58
	← Kaisyu left	16:58
	← SpeakerToMeat_ left	17:00
	→ SpeakerToMeat_ joined	17:01
	→ peacememories joined	17:03
	← peacememories left	17:03
	← SpeakerToMeat_ left	17:04
	→ SpeakerToMeat_ joined	17:06
	← rsrx left	17:09
	← D630 left	17:09
	← liefer left	17:09
	→ cbrt64 joined	17:10
rafasc	jorvis: take the last commit that touched that directory: git merge-base master <commit>	17:10
	← thiago left	17:10
	→ miczac joined	17:11
	← SpeakerToMeat_ left	17:12
jorvis	rafasc: Does that change the local directory though? I'm not wanting to change anything, just know where it is. It looks like I found the last pulled commit # by inspecting .git/That is, I want to ask of the directory "Where are you in relation to the commit history of the master?"	17:13
	Bad paste. By inspecting .git/FETCH_HEAD	17:13
	← gooooootime left	17:13
	← yyy left	17:14
rafasc	jorvis: git merge-base is just a querying tool. It doesn't change anything.	17:14
dawik	jorvis: you can do git diff master	17:14
	→ SpeakerToMeat_ joined	17:14
	← SpeakerToMeat_ left	17:14
dawik	asusming you are on the dev branch	17:14
rafasc	and it does exactly what are you asking: given two branches, find me the common commit in their history.	17:14
	→ gooooootime joined	17:14
	→ pkx1 joined	17:15
jorvis	That's the problem, this isn't a branch. They just had a long development path on the trunk but just stopped updating the code on a server. Now they need to know where that one was left off to make a branch and fix something. It's super ugly, but inherited and I'm trying to fix it.	17:15
	→ SpeakerToMeat_ joined	17:15
rafasc	jorvis: use git log -- dir; to figure out what was the last commit that touched that directory. And use that commit.	17:16
	→ doubtful joined	17:16
jorvis	There it is, thank you. The latest commit there matches what I found in FETCH_HEAD, but that command is a cleaner way to do it	17:18
	→ GreenJello joined	17:18
	← SpeakerToMeat_ left	17:19
	→ liefer joined	17:20
	→ SpeakerToMeat_ joined	17:20
	→ rsrx joined	17:23
	← floppydh left	17:26
	← mniip left	17:26
	← heroux left	17:27
	→ mniip joined	17:29
	← pawle left	17:30
	→ D5N_ch3ck joined	17:32
	← jstimm left	17:33
	← pkx1 left	17:33
	← durham left	17:33
	→ sanches joined	17:33
	→ leeN joined	17:33
	→ blackandblue joined	17:34
	→ dilfridge joined	17:36
	→ Mattiaslndstrm joined	17:36
	→ djapo joined	17:37
	← sanches left	17:37
	← SpeakerToMeat_ left	17:40
	→ nurdie joined	17:41
	→ SpeakerToMeat_ joined	17:42
	← doubtful left	17:42
	← jorvis left	17:43
	← rnat left	17:43
	← jlamur left	17:51
	← miczac left	17:53
	← ilmostro left	17:54
	→ miczac joined	17:55
	→ ilmostro joined	17:56
	→ orbyt_ joined	17:56
	→ jlamur joined	17:57
	← LondonAppDev left	18:00
	← rkta left	18:01
	→ sanches joined	18:01
	→ muelleme joined	18:02
	→ thiago joined	18:02
	← diogenese left	18:03
	→ diogenese joined	18:04
	← djapo left	18:06
	→ m0viefreak joined	18:06
	← Revan- left	18:07
	← muelleme left	18:07
	→ Revan- joined	18:08
	← SpeakerToMeat_ left	18:11
	→ SpeakerToMeat_ joined	18:13
	← mona_x left	18:16
	← SpeakerToMeat_ left	18:16
	← sauvin left	18:17
	→ SpeakerToMeat_ joined	18:17
	→ revoltingPeasant joined	18:19
	← D5N_ch3ck left	18:21
	→ Mike11 joined	18:22
	→ D5N_ch3ck joined	18:22
	← profmac left	18:23
	← opalepatrick left	18:25
	← blackandblue left	18:30
	→ mostlybadfly joined	18:30
	← SpeakerToMeat_ left	18:32
	→ hbautista joined	18:32
	→ durham joined	18:32
	← cdown left	18:32
	→ SpeakerToMeat_ joined	18:33
	← gooooootime left	18:35
	← SpeakerToMeat_ left	18:35
	→ beefjoe joined	18:35
	→ SpeakerToMeat_ joined	18:36
	← igemnace left	18:37
	→ peacememories joined	18:38
	← D5N_ch3ck left	18:39
	→ opalepatrick joined	18:39
	← SpeakerToMeat_ left	18:40
	→ D5N_ch3ck joined	18:40
	← peacememories left	18:40
	→ SpeakerToMeat_ joined	18:41
	← venmx left	18:41
	← D5N_ch3ck left	18:45
	→ D5N_ch3ck joined	18:46
	→ camsbury joined	18:46
	← st-gourichon-fid left	18:53
	→ spacesuitdiver joined	18:53
	→ yyy joined	18:55
	→ Matt\|home joined	18:59
Matt\|home	hi .. what's the quickest way to do an overwrite of all files from one branch to another except the .git history thing? in other words: "copy from master to mybranch all files overwriting everything"	19:00
	← SpeakerToMeat_ left	19:01
	→ SpeakerToMeat_ joined	19:02
	→ noob joined	19:04
	→ Sasazuka joined	19:05
noob	Hi ! I am in one repo and what to find the root level path of another repo . is there a git command for it	19:05
	want*	19:05
	← SpeakerToMeat_ left	19:05
	← virmin left	19:06
	→ profmac joined	19:07
	→ SpeakerToMeat_ joined	19:07
	→ treehug88 joined	19:08
	→ virmin joined	19:08
	→ st-gourichon-fid joined	19:08
	← SpeakerToMeat_ left	19:09
rafasc	noob: your question is not clear.	19:10
	→ SpeakerToMeat_ joined	19:11
	← SpeakerToMeat_ left	19:11
	→ SpeakerToMeat_ joined	19:12
RandomGuyOnIrc	If I checkout a tag, is the head in a detached state?	19:12
rafasc	RandomGuyOnIrc: yes. !detached	19:12
gitinfo	RandomGuyOnIrc: A detached HEAD (aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach	19:12
	← profmac left	19:14
	← revoltingPeasant left	19:14
RandomGuyOnIrc	thanks	19:14
rafasc	Matt\|home: probably git checkout mybranch; git checkout master -- .;	19:14
	← SpeakerToMeat_ left	19:15
	→ profmac joined	19:15
rafasc	Matt\|home: but more often than not, that's tells that there is something wrong with your workflow.	19:15
	← beefjoe left	19:16
	→ SpeakerToMeat_ joined	19:16
	← libertyprime left	19:17
	← virmin left	19:17
	← SpeakerToMeat_ left	19:18
	→ SpeakerToMeat_ joined	19:19
	← angular_mike_2 left	19:19
	← leeN left	19:20
	→ anna-sara joined	19:20
	→ angular_mike_2 joined	19:21
	← profmac left	19:21
	→ profmac joined	19:21
	← SpeakerToMeat_ left	19:23
	← ilmostro left	19:23
	→ wootehfoot joined	19:23
	→ SpeakerToMeat_ joined	19:24
	→ jrnieder joined	19:26
	← spl33n left	19:27
	← profmac left	19:27
	→ goTAN joined	19:27
	← SpeakerToMeat_ left	19:27
j416	Matt\|home: are you at work?	19:29
	j416 runs	19:29
	→ SpeakerToMeat_ joined	19:29
	← nurdie left	19:29
	→ lig joined	19:29
	→ AtumT joined	19:29
Matt\|home	j416 : i work remote from home, but yeah at the moment. and rafalcpp : i found out how to do it easy. and yeah the issue with my workflow was that too many people are working on my things at the same time :p	19:30
	yeah at the moment im working*	19:30
j416	Matt\|home: ah, it wasn't clear that you were at home.	19:30
	→ ilmostro joined	19:30
Matt\|home	:p	19:32
	so i like git, and it's really powerful.. there are a few idiosyncracies i not happy with and i have a few minor complaints, but i really wish after having used this stupid program for over a year i'd at least memorize all the commands needed -_-	19:32
	should work on that..	19:33
	→ pawle joined	19:33
	← czart left	19:34
	→ n3wborn joined	19:35
	→ TikityTik joined	19:37
	→ triad joined	19:37
	← utrack left	19:38
	→ nowhereman joined	19:38
	nowhereman → Guest70792	19:39
	← nowhere_man left	19:39
	→ utrack joined	19:40
	→ stuser joined	19:41
	→ yohnnyjoe joined	19:42
	→ Sonderblade joined	19:42
rafalcpp	Matt\|home: do I look like rafasc	19:42
_ikke_	yes	19:42
Matt\|home	ah	19:43
	rafasc , sorry *	19:43
rafasc	rafalcpp: this isn't the first and will not be the last!	19:43
	← sanches left	19:49
	← anna-sara left	19:50
	→ sanscoeur joined	19:52
stuser	trouble with submodules and remotes: I have a repo, "my-projects", which exists on my laptop (normal repo) and on my server (bare repo). On my laptop, I then create a submodule, "dotfiles", and then commit some stuff to the submodules. I then set the remote and try to push (with --set-upstream) but it fails with this error: https://upaste.de/Rk6 . from this I gather that I have to create the submodule on the server, inside of the existi	19:55
	maybe I'm getting some concepts mixed up, but how do I ensure that other people on my team can pull the newly create submodule?	19:57
	← TikityTik left	19:58
	← orbyt_ left	19:59
	→ dsx_ joined	20:00
	→ TikityTik joined	20:01
	← sanscoeur left	20:01
	← tafa2 left	20:02
	→ tafa2 joined	20:02
	→ profmac joined	20:02
	← dsx_ left	20:03
	← Envil left	20:07
	→ virmin joined	20:07
j416	the submodule needs its own repo remote	20:07
	it needs to be accessible via some URL.	20:08
	← plasticle left	20:08
j416	a submodule is in essence just a repo contained within a parent repo, locked to a specific revision	20:08
	→ nurdie joined	20:09
j416	what's your use-case?	20:09
	← BooK_ left	20:09
	→ plasticle joined	20:11
j416	stuser: ^	20:13
stuser	j416: my use case: I'm working on a project with my team and want to share my config files (dotfiles) with my team. We all use git, and I'm looking to VC my dotfiles, so I thought sharing them as a submodule would be good	20:16
	→ kerframil joined	20:16
j416	why do you need a submodule?	20:17
	(if you don't know why, I'd suggest you don't use a submodule until you figure out why)	20:19
	→ revoltingPeasant joined	20:23
	← jiji606 left	20:26
stuser	j416: I don't know why, but I have concerns about having many repos. The main concern being that I'll forget to push a repo that I did work in, then end up with conflicts. I thought converting my existing repos into submodules of a superproject would help with the management	20:26
_ikke_	You still need to keep the submodules up-to-date	20:27
stuser	but if I run git status in the superproject, wouldn't it detect a change in the submodule?	20:28
	← Mike11 left	20:29
rafasc	stuser: I think submodules are not ideal for your usecase. What you want is something like https://myrepos.branchable.com/	20:30
	→ sanscoeur joined	20:30
rafasc	disclaimer: I don't use this.	20:30
	but it is something that got talked some days ago around here.	20:31
stuser	ah I'll look into it. thank you	20:32
rafasc	stuser: yes, it will detect a change in the submodule if you do one. But it won't tell you if that change is pushed or not.	20:32
	← tomfarr left	20:33
rafasc	it will complain the submodule is not at the recorded hash, until you commit that change in the superproject.	20:33
	← Raging_Hog left	20:35
	→ tomfarr joined	20:37
	← mostlybadfly left	20:40
	← pawle left	20:41
j416	stuser: submodules are a way to lock versions of things; it's not a way to conveniently manage multiple unrelated repos	20:42
	← D5N_ch3ck left	20:42
j416	stuser: as for keeping things up-to-date, does it really matter if it's outdated if you're not using it?	20:43
	stuser: if you're working on something, it's easy to keep it up-to-date.	20:43
	stuser: most of my local repos where there are more people than me involved are out of date until I pick up work on them.	20:44
	works fine.	20:44
	s/most/all/	20:44
	even.	20:44
	← yohnnyjoe left	20:44
rafasc	> The main concern being that I'll forget to push a repo that I did work in, then end up with conflicts.	20:45
	← inkbottle left	20:46
	→ jstimm joined	20:46
rafasc	on top of all j416 said, this bit makes me think if you're not falling in that 'race for master' trap.	20:47
j416	while true; do sleep 300; script-that-pops-up-modal-window -message "Don't forget to push your changes"; done	20:47
	it's a human issue not a git issue	20:47
rafasc	j416: I find the '>' symbol in my prompt enough of a warning.	20:48
j416	what's that	20:48
	did I forget a quote	20:48
rafasc	symbol in the git prompt that tells you're ahead of upstream.	20:49
j416	oh.	20:49
	fancy things you have. I don't have that.	20:49
rafasc	how? o.O	20:49
	or, better, why? o.O	20:49
j416	never knew it existed	20:50
	is that part of the default contrib prompt?	20:50
rafasc	yeah..	20:50
j416	this is what I have currently https://github.com/jsageryd/.config.d/blob/2e4455283c67/profile/common/git.inc#L7-L19	20:51
	maybe I forgot something	20:51
stuser	thank you all, I think I'm understanding submodules better now. but 1 question: suppose I have to wipe my computer or get a new device and I want to get all of my projects quickly. If I have all of my repos setup as submodules of a general "projects" repo, then could "$git clone --recurse-submodules /url/to/projects" get them all in one command?	20:51
	← TikityTik left	20:52
j416	submodules still aren't the right approach	20:52
	get a list of your repos, clone them	20:52
	or, well, the one(s) you need.	20:52
rafasc	j416: add this one: GIT_PS1_SHOWUPSTREAM="auto"	20:54
j416	ooh.	20:54
	a > appears	20:54
rafasc	> < = depending if you're behind ahead or synced	20:55
j416	yeah, I read up on it	20:56
	I'm annoyed by the = though	20:56
stuser	:/ ah alright. guess I'll make a script. I don't suppose git maintains a list of local repos?	20:56
rafasc	in relation to the configured upstream	20:56
	→ yohnnyjoe joined	20:56
j416	stuser: that'd be your filesystem	20:56
	stuser: "ls"	20:56
	→ Acelogic joined	20:57
rafasc	stuser: why make a script when someone else already did for you? myrepos being just one example.	20:57
	→ fukakai joined	20:57
j416	or that ^	20:57
	← opalepatrick left	20:57
rafasc	j416: the = is helpful though, to distinguish when you have an upstream set.	21:00
	← nurdie left	21:00
j416	I suppose	21:00
	I don't like that it's white	21:00
	→ nurdie joined	21:00
	← durham left	21:01
	← yyy left	21:01
rafasc	search for # calculate the result and patch it.	21:01
	→ inkbottle joined	21:01
j416	nah; hassle	21:01
	:P	21:01
rafasc	it isn't configurable...	21:02
j416	dunno	21:02
	← armin left	21:02
	← tombusby left	21:02
	→ ZetFury joined	21:02
	→ armin joined	21:02
	← lig left	21:03
	→ tombusby joined	21:03
	→ orbyt_ joined	21:03
	← armin left	21:04
	→ yyy joined	21:04
rafasc	that wasn't a question. It was a statement.	21:04
j416	fitured	21:04
	figured*	21:04
	"auto" doesn't seem to be a thing btw	21:04
	it just needs to be set afaict	21:04
	I set it to "foo", works fine	21:05
	← triad left	21:07
	← wootehfoot left	21:08
rafasc	If you would like to see the difference between HEAD and its upstream, set GIT_PS1_SHOWUPSTREAM="auto".	21:08
j416	yeah I saw that	21:08
	try searching for "auto"	21:08
	← SpeakerToMeat_ left	21:09
j416	:P	21:09
	auto is probably just the "current default"	21:09
	→ armin joined	21:09
j416	i.e. no logic needed	21:09
	← voolik left	21:10
rafasc	I think any word that doesn't match the ones that change the behaviour are considered "auto".	21:10
j416	yeah	21:10
	→ opalepatrick joined	21:12
rafasc	j416: GIT_PS1_SHOWUPSTREAM=name; doesn't seem to be working though.	21:12
	← nurdie left	21:13
	← avar left	21:14
	→ polydo_s joined	21:15
	← armin left	21:15
	→ armin joined	21:16
	← armin left	21:17
j416	seems to work here	21:18
	← Sonderblade left	21:19
rafasc	can you show me the result?	21:20
j416	2221 q:~/.config.d [master *=] j$	21:21
rafasc	name was supposed to show the upstream branch too.	21:22
	→ Tobbi joined	21:22
j416	oh.	21:23
Tobbi	Is there an easy (one line) code to squash all the last 86 commits on a branch into one?	21:23
	→ venmx joined	21:23
j416	I thought you meant it didn't show anything. You need verbose though	21:23
	← jtrzebiatowski left	21:23
	← yohnnyjoe left	21:24
rafasc	j416: yes, but name seems to be undoing verbose in my machine.	21:25
j416	just "verbose" gives me 2225 q:~/.config.d [master * u=] j$	21:25
	"name verbose" gives me 2225 q:~/.config.d [master * u= origin/master] j$	21:25
	Tobbi: git reset --soft <commit>; then commit again	21:26
rafasc	j416: derp, I read space-separated as comma-separated.	21:26
j416	rafasc: I did too. Took me a minute to figure that out. :D	21:26
rafasc	<commit> = HEAD~86	21:27
	→ yohnnyjoe joined	21:27
j416	nod	21:27
rafasc	depending if their 86 is counting from 0 or not, off-by-one errors might apply.	21:28
	j416: GIT_PS1_DESCRIBE_STYLE also seems cool. Didn't knew it was configurable.	21:30
j416	nod	21:31
	pretty neat	21:32
	"branch" seems good	21:33
rafasc	I was split between tag and branch.	21:35
	← plexigras left	21:35
j416	tag seems useless for repos that don't have tags	21:35
	most of the repos I work on daily don't have tags	21:36
	also, tags aren't that interesting when doing actual work	21:36
	at least not to me	21:36
rafasc	yea, this only applies to labeling the 'detached head' state, so yeah. It will depends if you're interested in comparing with the future or the past.	21:37
	← kapil____ left	21:37
	← goTAN left	21:37
j416	also, I think I prefer to see what newer branch or tag _contains_ HEAD, rather than an old one that isn't really realted	21:37
	it could be based on whatever past	21:38
	multiple pasts	21:38
	come to think of it that's true for the opposite as well	21:39
rafasc	I'll stick with branch for now, I guess using it is the way to find out if I like it.	21:39
j416	yeah	21:40
rafasc	but for people used to describe output, it's also handy: this commits happened 15 commits after some tag.	21:40
j416	I think I'll try out "verbose" for the other one	21:40
	rafasc: yeah. Maybe.	21:40
	→ SpeakerToMeat_ joined	21:40
	← jlamur left	21:40
rafasc	yea, me too. I actually changed for verbose name.	21:41
j416	my branches are typically origin/samename anyway	21:42
rafasc	I changed my prompt to take up two lines anyways. [rafasc@something somepath] \n $	21:42
j416	ah.	21:42
	never liked that	21:42
rafasc	so the long prompts are not an issue for me.	21:42
j416	two lines per newline	21:43
	but seems handy	21:43
	→ cdown joined	21:43
j416	might change my mind one day.	21:43
rafasc	It annoyed me at first, but having to write line wrapped commands annoyed me more I guess.	21:44
j416	yeah	21:44
	that's annoying	21:44
	→ jnewt joined	21:44
	→ peacememories joined	21:44
j416	toyed with the thought of putting the prompt info in the tmux status bar	21:44
	might work	21:44
rafasc	give it a try for some days, only way to know.	21:45
j416	but perhaps it's useful to be able to see in the history in what directory you were previously	21:45
	nod; will see.	21:45
rafasc	Don't use tmux... i3+termite does the job.	21:45
	→ d10n-work joined	21:46
j416	arch?	21:46
rafasc	yes.	21:46
j416	(it was the first page I got when I googled termite i3)	21:46
	I'm on a mac so no proper window manager for me	21:46
	→ cdown_ joined	21:46
rafasc	virtual env prompt + git prompt + long paths ... at the end of that I was typing two or the command words then getting line wrapped.	21:46
j416	haha	21:47
rafasc	this way, all the information junk gets its own line, I get one line for myself.	21:47
	← peacememories left	21:49
	← cdown left	21:49
	cdown_ → cdown	21:49
rafasc	triple clicking the line, you get only the '$ some commands'	21:49
	which is nice to paste, without telling the name of your user/machine.	21:50
j416	should be rather easy to copypaste a whole chunk of scrollback too; just remove lines starting with [user	21:50
rafasc	yeah... sed /[rafasc/d or something like that	21:51
	→ durham joined	21:53
	← durham left	21:53
	→ MrRothstein joined	21:54
	← TheRuckus left	21:56
MrRothstein	hello, i have a large directory down in the commit history that is shared between two branches i'd like to remove before pushing to remote. Is there a way to do it without losing the ability to later merge the branches?	21:56
	← cdown left	21:57
MrRothstein	to clarify... the directory was committed by accident and removed in a later commit. I'm fine squashing the commits that contained the directory, but i'm not sure how to do it in a way that doesnt break merging	21:57
	→ Toadisattva joined	21:59
	→ majuscule joined	22:02
majuscule	Why might I be getting this error when cherry-picking from a branch? fatal: empty ident name (for <>) not allowed	22:02
	→ jlamur joined	22:03
majuscule	All I can find about it is that it might have something to do with a broken diff but... this isn't a diff, it's a branch that git has already	22:03
	← netj left	22:05
	→ netj joined	22:05
	→ leeN joined	22:06
	→ mitchellsalad__ joined	22:12
rafasc	you either don't have your indentity set, or the commit you're trying to cherry-pick has corrupt identity information...	22:12
	← miczac left	22:13
	← xcm left	22:14
	← rafasc left	22:15
	→ xcm joined	22:16
MrRothstein	so... is there a way to squash erroneous commits in a history shared between branches?	22:20
	← kerframil left	22:29
	← wildermind left	22:31
	→ TikityTik joined	22:31
	← leeN left	22:32
	← hbautista left	22:33
	← fukakai left	22:38
	→ nurdie joined	22:39
	← ZyX-I left	22:40
	← Toadisattva left	22:42
	← nurdie left	22:44
	← yyy left	22:46
	← Dougie187 left	22:47
	→ g00s joined	22:48
	← orbyt_ left	22:51
	← MrRothstein left	22:54
	→ fassl joined	22:55
	→ miczac joined	22:56
	← Fissure left	22:57
	← Revan- left	23:05
	← miczac left	23:05
	→ Revan- joined	23:06
	→ novumsensum joined	23:07
	→ mofai joined	23:07
	→ miczac joined	23:09
	← relipse left	23:10
	← miczac left	23:10
	← staafl__ left	23:14
	→ igemnace joined	23:14
	← esotericnonsense left	23:14
	← freeman42x left	23:14
	→ mklaang joined	23:14
	← spacesuitdiver left	23:16
	→ esotericnonsense joined	23:16
	← mklaang left	23:18
	→ djapo joined	23:20
	← TikityTik left	23:21
	← revoltingPeasant left	23:22
	← stuser left	23:27
	← Acelogic left	23:29
	← Kaisyu72 left	23:32
	← thiago left	23:33
	← M6HZ left	23:33
	→ aLeSD_ joined	23:35
	← aLeSD left	23:37
	← AtumT left	23:38
	← ZetFury left	23:42
	→ CodingRacoon joined	23:45
	→ fradanovic_ joined	23:46
	→ aLeSD joined	23:47
	← aLeSD_ left	23:50
	← venmx left	23:51
	← bolovanos left	23:51
	← opalepatrick left	23:52
	← CodingRacoon left	23:55
	← Peetz0r left	23:55
	→ Praxi joined	23:56
	→ piggah joined	23:57
	← dege left	23:58
	→ Peetz0r joined	23:59

Logs Search ←Prev date Next date→

Channels Documentation