Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-10

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 ocelotsloth joined #git
00:02 Gsham joined #git
00:02 pfrench joined #git
00:03 PCatinean joined #git
00:10 MindAndGo joined #git
00:10 pfrench joined #git
00:11 Flotho66 joined #git
00:11 mda1 joined #git
00:11 Flotho66 good evening everyone
00:12 Flotho66 I would like to know if it's possible to make the difference between 2 commit for a specific folder and apply it as a patch
00:12 Eugene
00:13 ResidentBiscuit joined #git
00:13 Cabanossi joined #git
00:14 bremner git diff a b folder > foo.patch
00:16 xissburg joined #git
00:16 Bennux joined #git
00:17 nopf joined #git
00:20 pfrench joined #git
00:20 DolphinDream joined #git
00:21 Flotho66 thanks bremmer
00:22 gopar joined #git
00:22 TikityTik joined #git
00:22 madewokherd joined #git
00:22 Flotho66 thanks bremner
00:23 xissburg joined #git
00:23 mischat joined #git
00:23 rkazak_ joined #git
00:26 automaton joined #git
00:27 vtols joined #git
00:27 Abbott joined #git
00:27 King_Hual` joined #git
00:28 Abbott if I want to remove all extra files in a directory to make it match a repository exactly, I could do rm -rf * then git checkout -- .
00:28 Abbott but is there a command that does that all in one? git reset with a flag or something
00:29 Nicmavr joined #git
00:29 Eugene !resets
00:29 gitinfo tl;dr of man git-reset: --soft moves HEAD, --mixed moves HEAD+index, --hard moves HEAD+index+work-tree, -- foo.txt will !unstage
00:29 marsmellow joined #git
00:30 Eugene This only affects tracked files; if you want to delete untracked ones do something like `git clean -fdx`. Please read man git-clean first because this WILL delete things.
00:30 gitinfo the git-clean manpage is available at http://jk.gs/git-clean.html
00:31 xissburg_ joined #git
00:32 marianina8 joined #git
00:32 holodoc joined #git
00:32 Abbott okay I think git-clean is what I'm looking for
00:32 Abbott thanks Eugene I'll read that
00:33 King_Hual_ joined #git
00:39 euphxenos joined #git
00:39 Sasazuka joined #git
00:40 UTAN_dev joined #git
00:44 urda joined #git
00:44 menip joined #git
00:44 anuxivm left #git
00:49 mrslaughter joined #git
00:50 pfrench joined #git
00:51 dvaske joined #git
00:51 DarkPsydeLord joined #git
00:52 xissburg joined #git
00:54 TbobbyZ_ joined #git
00:56 nettoweb joined #git
00:57 mmattice joined #git
00:58 Cabanossi joined #git
01:03 jimklo joined #git
01:06 rchavik joined #git
01:07 a_thakur joined #git
01:07 cdg joined #git
01:10 luke_nukem joined #git
01:10 cdg joined #git
01:11 variable joined #git
01:13 UTAN_dev joined #git
01:15 pfrench joined #git
01:15 jimklo joined #git
01:18 WayToDoor joined #git
01:18 luke_nukem joined #git
01:19 rominronin joined #git
01:19 keepguessing joined #git
01:21 skered joined #git
01:21 livingstn joined #git
01:22 jimklo joined #git
01:25 mischat joined #git
01:25 jimklo joined #git
01:28 xissburg joined #git
01:29 Vampire0_ joined #git
01:31 aidalgol joined #git
01:33 cqi joined #git
01:33 UTAN_dev joined #git
01:35 Sasazuka joined #git
01:37 durham_ joined #git
01:38 raspado joined #git
01:40 Sasazuka_ joined #git
01:43 sarbs joined #git
01:45 pfrench joined #git
01:46 livingstn joined #git
01:48 dreiss joined #git
01:50 dvaske joined #git
01:51 skered joined #git
01:51 Sasazuka joined #git
01:55 altinu joined #git
01:56 dvaske joined #git
01:56 uklot joined #git
01:58 tmg joined #git
01:58 Cabanossi joined #git
02:00 UTAN_dev joined #git
02:03 Gsham joined #git
02:05 Nilesh_ joined #git
02:09 Sasazuka joined #git
02:10 xall joined #git
02:16 nowhere_man joined #git
02:16 howitdo joined #git
02:16 howitdo joined #git
02:18 d0nn1e joined #git
02:18 raspado joined #git
02:19 raspado joined #git
02:19 A124 joined #git
02:19 YottaByte joined #git
02:19 pfrench joined #git
02:19 mrslaughter1775 joined #git
02:20 YottaByte hi, if I want to get the changes from another branch but I don't want to check that other branch out locally and then do pull then switch back to my branch to merge it, can I simply "git pull origin the-brach"?
02:20 luke_nukem joined #git
02:21 SwiftMatt joined #git
02:21 nedbat YottaByte: if you don't mind me asking, why don't you want to have the branch locally?
02:22 YottaByte because that's so many extra steps
02:22 Sasazuka joined #git
02:22 YottaByte the idea is this: I work in my own branch, then make a merge request in the branch I want to pull in. I pull in so I can resolve any merge conflicts that may occur, so when I finally make the merge request, everything has already been resolved (by me)
02:22 nedbat YottaByte: you don't switch to the other branch to merge it anyway. You should be on the branch you want to bring the changes into
02:23 YottaByte I am
02:23 streetwitch joined #git
02:23 YottaByte wait I'm confused now
02:24 nedbat YottaByte: git fetch --all;  git merge other-branch
02:24 streetwitch Is there a way to clear a directory of all git information and then upload the entire directory to github?
02:24 YottaByte there's a branch A, I'm in branch B, I want to bring my changes from B into A (via a merge request), so first I want to pull A into B to see if any merge conflicts arise
02:24 nedbat streetwitch: what do you mean, clear a directory of git information?
02:24 streetwitch like .gitignore
02:25 streetwitch and and everything else hiding
02:25 nedbat streetwitch: why would you want to get rid of .gitignore?
02:25 streetwitch So I can upload everything
02:25 nedbat streetwitch: you should upload the .gitignore file.
02:25 streetwitch But the .gitignore file excludes important data
02:25 streetwitch I just want a fresh start!
02:26 nedbat streetwitch: that's odd, why did you exclude those files?
02:26 streetwitch nedbat, I don't know
02:26 nedbat streetwitch: you can make any changes you like and then push to github
02:26 mischat joined #git
02:27 streetwitch nedbat: I don't want to push changes
02:27 streetwitch I want to start a new repo with everything in my source directory
02:27 nedbat streetwitch: rm -rf .git;  git init
02:27 burzum joined #git
02:28 streetwitch Thanks
02:28 pks joined #git
02:28 YottaByte yeah, I think I can just git pull origin branch
02:30 d^sh joined #git
02:31 ISmithers joined #git
02:31 finalbeta joined #git
02:38 gopar joined #git
02:38 izacht13 joined #git
02:38 gugah joined #git
02:40 UTAN_dev joined #git
02:42 hexagoxel joined #git
02:46 clemf joined #git
02:46 hiroki joined #git
02:46 hiroki joined #git
02:46 MattMaker joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help. Just git help, apparently. | We are aware of the SHA-1 collision. See: http://jk.gs/t/a4 | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.12.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This is serious business. https://madeitwor.se/git-jokes
02:51 rominronin joined #git
02:52 peterbecich joined #git
02:53 MattMaker joined #git
02:54 UTAN_dev joined #git
02:54 NeverDie joined #git
02:54 Rodya_ joined #git
02:54 aavrug joined #git
02:55 sunri5e joined #git
02:58 Aeterne joined #git
03:04 ojdo joined #git
03:05 chipotle joined #git
03:08 a_thakur joined #git
03:08 ejb joined #git
03:13 duderonomy joined #git
03:13 rkazak_ joined #git
03:14 mizu_no_oto joined #git
03:15 impliednude joined #git
03:18 dsantiago joined #git
03:19 izacht13 left #git
03:20 UTAN_dev joined #git
03:21 pfrench joined #git
03:21 streetwitch Is anyone here?
03:22 dtrott joined #git
03:25 zrts_ yes
03:25 Vortex35 joined #git
03:26 lagothri1 joined #git
03:28 Cabanossi joined #git
03:28 mischat joined #git
03:29 pfrench joined #git
03:32 chachasmooth joined #git
03:38 mrslaughter1775 joined #git
03:39 brent__ joined #git
03:40 keepguessing joined #git
03:40 UTAN_dev joined #git
03:41 MTecknology no
03:43 streetwitch Hi
03:43 streetwitch When I try to upload my git to a github repository, it doesn't include one very important directory
03:43 streetwitch and I can't figure out how to add it!
03:43 streetwitch Can anyone help guide me?
03:47 xall joined #git
03:48 MTecknology git add
03:49 arret joined #git
03:49 dvaske joined #git
03:50 streetwitch MTecknology, funny
03:50 blunaxela joined #git
03:51 MTecknology That's how you add things. If you need more, then maybe include some details...
03:53 zivester joined #git
03:54 cjohnson I have an unpushed commit that I want to rewind
03:54 cjohnson none of my other changes were impacted by this commit
03:54 cjohnson but I'd rather not have a revert commit
03:55 cjohnson Can I use rebase to basically just delete the commit and undo the changes it had on my code?
03:55 cjohnson I've had other commits since then so I can't just reset to it
03:56 jameser joined #git
03:56 cjohnson If I git rebase -i HEAD~9 and put "drop" on that line, will it basically remove the commit and undo the changes the commit introduced?
03:56 cjohnson the commit is 7 commits back
03:57 thib joined #git
03:58 jerrcs joined #git
03:58 MTecknology git reset HEAD~
03:58 hfp_work joined #git
03:59 cjohnson I've had other commits since then so I can't just reset to it
03:59 hfp joined #git
03:59 hexagoxel joined #git
03:59 Isla_de_Muerte joined #git
03:59 cjohnson Anyway, got it sorted. git rebase -i HEAD~9 and using drop on that commit was correct
03:59 xall joined #git
04:00 c0ded joined #git
04:00 a_thakur joined #git
04:00 cwre joined #git
04:02 svm_invictvs joined #git
04:05 pfrench joined #git
04:05 SkyFire joined #git
04:06 SkyFire Can someoen help me with git? im using it on Visual Studio and im having trouble publishing my branch to remotee
04:06 impliednude joined #git
04:10 pfrench joined #git
04:12 Cabanossi joined #git
04:14 pfrench joined #git
04:20 UTAN_dev joined #git
04:20 lucido-cl joined #git
04:21 keepguessing joined #git
04:21 rominronin joined #git
04:21 gtristan joined #git
04:23 hashtagstrashtag joined #git
04:27 UTAN_dev_ joined #git
04:27 pfrench joined #git
04:29 mischat joined #git
04:31 pfrench joined #git
04:35 hahuang65 joined #git
04:35 akkad joined #git
04:37 xall joined #git
04:38 pfrench joined #git
04:41 dcy joined #git
04:45 muhannad_______ joined #git
04:45 MarioBranco joined #git
04:45 peterbecich joined #git
04:51 mehola joined #git
04:52 keepguessing joined #git
04:56 dvaske joined #git
04:57 Cabanossi joined #git
04:59 akkad joined #git
05:03 TheHackax joined #git
05:03 pfrench joined #git
05:05 xall joined #git
05:06 rkazak_ joined #git
05:09 sveinse joined #git
05:09 MineCoins joined #git
05:10 ayogi joined #git
05:10 pfrench joined #git
05:11 sveinse I have somewhat a tricky challenge. We used to have a git (native) server that have been replaced by another http based git server. Problem is that a lot of old script still refer to the old server. Is there a way, a proxy or anything, that is capable of presenting the old git service to the new URL somehow? Any ideas on how that might work?
05:13 dandaman joined #git
05:13 ResidentBiscuit joined #git
05:17 NwS joined #git
05:17 Andrew_K joined #git
05:19 dreiss joined #git
05:20 UTAN_dev joined #git
05:20 dcy joined #git
05:21 preaction sveinse: it's just an http proxy, no?
05:21 Goplat joined #git
05:25 mbrumbelow joined #git
05:28 pur3eval joined #git
05:31 mischat joined #git
05:31 sveinse preaction: no, because I don't think the git daemon protocol supports redirects. If it only would be that easy.
05:32 pfrench joined #git
05:36 waveclaw joined #git
05:36 imack joined #git
05:37 irqq joined #git
05:40 UTAN_dev joined #git
05:42 WayToDoor joined #git
05:46 pfrench joined #git
05:46 cqi joined #git
05:48 rkazak_ joined #git
05:51 sword joined #git
05:51 matt_m joined #git
05:52 marianina8 joined #git
05:52 thiago joined #git
05:53 UTAN_dev joined #git
05:53 ankit01ojha joined #git
05:54 rominronin joined #git
05:55 dreiss joined #git
05:56 dcy joined #git
05:56 peterbecich joined #git
05:58 mda1 joined #git
06:00 Andrew_K joined #git
06:00 grawity regular git-daemon + regular TCP relay (like socat/stunnel/rinetd)
06:00 pfrench joined #git
06:01 a_thakur joined #git
06:01 gopar joined #git
06:01 xcesariox joined #git
06:01 xall joined #git
06:02 dec0n joined #git
06:03 sveinse grawity: that requires the relay destination to be git-daemon, right? Problem is that the new repo is http.
06:03 cesario_ joined #git
06:04 sveinse I probably need to setup a git-server and clone regularely from the master http repo. There is never going to be made any pushes that way
06:06 grawity "http" isn't an immutable property of the repo
06:06 grawity do you mean that you can't *possibly* set up git-daemon on the new repo's server?
06:06 grawity like, if it's github/bitbucket hosted type
06:07 xcesariox joined #git
06:07 sveinse grawity: yes, it's offsite, like github
06:08 pfrench joined #git
06:08 rscata joined #git
06:11 dreiss joined #git
06:11 comradepashka joined #git
06:12 pfrench joined #git
06:14 comradepashka hello guys! I have a noob question, sorry: if I clone repo from github to my local PC and there is one pull request, not confirmed yet, is it possible to apply it to my local repository?
06:15 mmattice comradepashka: sure.  git remote add [otherreponame] [otherrepourl] ; git remote update
06:15 tmg joined #git
06:17 manuelschneid3r joined #git
06:19 dandaman joined #git
06:22 comradepashka hm.. probably I was incorrect. I mean, there is pull request (and it is not my pull request) already at this repository, but it is not merged yet, and I need it, so I can rebuild project with this fix before community include it officially (they will, indeed)
06:22 qt-x joined #git
06:23 pfrench joined #git
06:23 comradepashka it is proposed pull request and I need it locally
06:24 WayToDoor joined #git
06:24 masuberu joined #git
06:25 sbulage joined #git
06:28 navidr joined #git
06:29 pfrench joined #git
06:30 Vampire0_ joined #git
06:32 ldleworker left #git
06:32 mischat joined #git
06:38 robattila256 joined #git
06:41 MrC joined #git
06:41 g0d355__ joined #git
06:42 sbulage joined #git
06:42 Cabanossi joined #git
06:43 Andrew_K joined #git
06:43 dvaske joined #git
06:44 jay-m joined #git
06:45 akkad joined #git
06:45 TheHackax joined #git
06:46 pfrench joined #git
06:47 bocaneri joined #git
06:53 pur3eval joined #git
06:53 pfrench joined #git
06:55 joeco joined #git
06:58 pfrench joined #git
06:59 Silmarilion joined #git
07:02 pur3eval joined #git
07:04 pfrench joined #git
07:05 keepguessing joined #git
07:06 UTAN_dev joined #git
07:07 cdown joined #git
07:08 a_thakur joined #git
07:09 pfrench joined #git
07:11 kurkale6ka joined #git
07:11 senaps joined #git
07:12 Cabanossi joined #git
07:12 flobin joined #git
07:14 ResidentBiscuit joined #git
07:15 pfrench joined #git
07:15 lordjancso joined #git
07:15 thiago joined #git
07:18 w9qbj joined #git
07:20 UTAN_dev joined #git
07:20 pfrench joined #git
07:24 manuelschneid3r joined #git
07:24 imack joined #git
07:24 srandon1213 joined #git
07:24 srandon1213 hello all, is there a way on git to emulate a merge, in order to see if there would be some conflicts ?
07:25 rominronin joined #git
07:26 chele joined #git
07:28 Seveas srandon1213: yes.
07:28 Seveas see https://git.seveas.net/previewing-a-merge-result.html#previewing-a-merge-result
07:28 srandon1213 Seveas: ok i mean... how  ?
07:28 srandon1213 okok thanks
07:28 rominronin joined #git
07:28 srandon1213 thanks a lot Seveas
07:33 pfrench joined #git
07:34 mischat joined #git
07:34 ali1234 is there a way to get an atomic "time of last change" for an entire repository and working dir?
07:34 ali1234 including dirty files
07:36 ali1234 also including things like changing the currently checked out branch
07:37 Flotho66 joined #git
07:37 overlord_tm joined #git
07:40 nevodka joined #git
07:42 jnavila joined #git
07:47 Chrismeister joined #git
07:47 theoceaniscool joined #git
07:52 aidalgol joined #git
07:53 pfrench joined #git
07:53 dersand joined #git
07:53 luke_nukem joined #git
07:56 pfrench joined #git
07:58 masuberu joined #git
07:59 elect joined #git
08:00 Seveas including dirty files: no.
08:00 Seveas excluding dirty files: look at the last reflog entry for HEAD
08:04 Tobbi joined #git
08:06 nowhere_man joined #git
08:06 sarbs joined #git
08:07 luke_nukem left #git
08:07 fk joined #git
08:08 freedrul1 left #git
08:08 j08nY joined #git
08:10 ali1234 Seveas: looks like mostly what i need, thanks
08:11 rkazak_ joined #git
08:12 ak77 how to do "git submodule add" in a filter-branc --tree-filter script ?
08:13 flobin joined #git
08:14 dmj` joined #git
08:15 ak77 I do "git submodule add" the submodule repo is downloaded, but in the next iteration (next commit), the submodule directory is missing.
08:15 absinthe joined #git
08:16 wxg joined #git
08:16 dvaske joined #git
08:17 imack joined #git
08:17 dvaske_ joined #git
08:20 Tobbi joined #git
08:20 absinthe joined #git
08:22 PHPanos joined #git
08:23 pur3eval joined #git
08:23 n1ce joined #git
08:26 pfrench joined #git
08:27 oskarkv joined #git
08:27 flobin left #git
08:29 elect joined #git
08:29 luke_nukem joined #git
08:29 ak77 anyone has an idea? submodule file gets created, the submodule directory is present with content, but on the next iteration the submodule is missing
08:31 flaviodesousa joined #git
08:32 markzen joined #git
08:32 markzen hi
08:33 aLeSD joined #git
08:33 aLeSD hi all
08:34 aLeSD how could I create a copy of my master brach without version control ?
08:34 variable joined #git
08:34 markzen what are the best practices when there are a few bugs in a pull request?
08:34 variable Is there a porcelin version of git branch --merged=origin/master ?
08:35 sotoz joined #git
08:35 variable markzen if this is github specific you may want to ask in #github
08:35 peterbecich joined #git
08:35 markzen variable: this is about dvcs workflow I guess
08:35 variable does anyone know if there is a version of "git merge --no-commit --no-ff" but that doesn't actually change the working tree? I
08:36 moritz aLeSD: you mean a copy of all files? git archive might help you
08:36 variable markzen you can either fix the bugs for the person, or otherwise tell them and have the OP fix them
08:36 variable markzen the former is faster, the latter is better mentorship
08:36 pfrench joined #git
08:36 moritz variable: what should it do? just check if the other side is ahead of the current branch?
08:36 aLeSD moritz, yes I want a copy of my project in another folder without .git files
08:37 variable moritz: I want to see if a merge will conflict without actually affecting the working tree
08:37 Silenced joined #git
08:38 moritz variable: you could do the merge with --no-commit and then git merge --abort
08:38 variable moritz but then I'd need to be on the right branch
08:38 moritz variable: yes
08:38 variable moritz  what I'm actually trying to do is iterate over 1000+ branches
08:38 variable and see if master merges into them without conflicts
08:38 JeroenT joined #git
08:38 variable and I don't want to spend the i/o cost of checking out each branch
08:39 Silenced joined #git
08:39 variable ideally I could see "will this merge" using the index / stored files, and just bail if there would be conflicts
08:40 a-l-e joined #git
08:41 pfrench joined #git
08:42 a_thakur joined #git
08:45 Balliad joined #git
08:45 Vortex34 joined #git
08:46 pfrench joined #git
08:50 pfrench joined #git
08:52 Dumblez0r joined #git
08:53 adac joined #git
08:53 adac git push --all origin shouldn't that also push all branches?
08:53 byte512 joined #git
08:55 adac for some reasons it does not do that for me actually
08:56 osse what happens instead
08:56 pfrench joined #git
08:57 adac osse, only the master is pushed
08:57 osse adac: some sanity check: do you actually have more than one branch? and are any of those different from those at origin ?
08:57 osse and what is the output of git push
08:59 _ADN_ joined #git
08:59 adac osse, actually that is the ouptut of my local repository I cloned from my old gilab installation:
08:59 adac git branch -a https://gist.github.com/anonymous/03c996a64b5da288d3bf71492daa55b9
09:00 adac and I then tried to push it onto my new gitlab installation via
09:00 adac git push -u origin --all
09:00 adac git push -u origin --tags
09:00 adac but there is then only the master branch pushed
09:00 osse there is only master branch
09:01 osse the others are remote tracking branches. the don't count as far as --all is concerned
09:01 pluszak joined #git
09:01 osse *they
09:01 adac osse I see. So a git clone does not pull them at all
09:01 osse yes it does
09:01 osse but it gives them a different name
09:02 osse namely 'origin/master' instead of 'master' etc.
09:02 osse if it didn't then they wouldn't even be in the list
09:02 adac osse, yes I see
09:02 osse i think maybe you want --mirror
09:03 osse but then the branches become remote tracking branches in the destination too
09:03 adac how can i download these branches propery so they get uploaded?
09:03 osse I would write individual push commands
09:03 cdown joined #git
09:03 g0d355__ joined #git
09:04 osse or you could just   clone --mirror in the first place, then push --mirror
09:04 adac osse, hmm ok I try the clone --mirror then
09:04 adac there are a lot of branches
09:04 adac thanks!
09:06 wrouesnel1 joined #git
09:08 armyriad joined #git
09:08 bongjovi joined #git
09:09 pfrench joined #git
09:11 adac osse, awesome that mirror thing works!
09:11 adac thank you again!
09:11 mikecmpbll joined #git
09:11 marcogmonteiro joined #git
09:12 Cabanossi joined #git
09:15 ResidentBiscuit joined #git
09:16 pfrench joined #git
09:18 PHPanos joined #git
09:18 TurBoss joined #git
09:19 TurBoss Hello
09:19 TurBoss I have 5 unstaged files but when i commit them they start a new tree from bottom don't continue from the lattest commit
09:19 TurBoss what i'm doing wront?
09:19 TurBoss *wrong
09:20 sathed_ joined #git
09:20 canton7 "start a new tree from bottom don't continue from the lattest commit" -- I'm afraid I can't work out what you're trying to say
09:20 pfrench joined #git
09:20 sathed_ joined #git
09:20 TurBoss canton7: hmm the commit doesn't follow history
09:21 TurBoss starts from the begining
09:21 canton7 oh, so not a "tree" in the git sense? you mean a "tree" of commits?
09:21 canton7 (we call that a graph in git)
09:21 TurBoss oh thanks
09:21 TurBoss yeys
09:21 TurBoss *yes
09:22 canton7 a tree in git is used to store a directory, full of files and other directories
09:22 osse sounds like you're in an orphan branch or soem such
09:22 canton7 right, so when you make a commit, it doesn't have a parent commit?
09:22 TurBoss yes thats it
09:22 canton7 yeah, did you 'git checkout --orphan' at any point?
09:22 TurBoss no
09:22 chipotle joined #git
09:22 TurBoss i should?
09:22 canton7 what makes you think that there isn't a parent commit?
09:22 canton7 no, I'm trying to work out how you got into the state you're in
09:23 osse let's do a big standard 'git status' and see what it says
09:23 osse *bog
09:23 TurBoss the graph line has no start commit
09:24 TurBoss http://dpaste.com/1F8Q7CC
09:24 TurBoss here is the output of git status
09:24 canton7 what "graph line" do you mean?
09:24 TurBoss in the gui
09:24 canton7 what gui?
09:24 canton7 help us please
09:25 TurBoss sek i'll took a screenshot
09:25 Darcidride joined #git
09:25 a_thakur joined #git
09:25 SwiftMatt joined #git
09:26 TurBoss http://i.imgur.com/58EKato.png
09:26 TurBoss the NukedLibs branch should be parent of master
09:26 canton7 where do you see a graph line with no start commit?
09:27 TurBoss sorry again I mean a commit without parent
09:27 canton7 where do you seee a commit without a parent?
09:28 TurBoss the in the NukedLibs branch
09:28 canton7 where?
09:28 canton7 all of the commits in the NukedLibs branch have parents
09:28 canton7 (well, the single commit in the NukedLibs branch)
09:28 TurBoss it should be parent of master
09:28 canton7 that's not what you said. You said it makes a commit with *no parent*
09:29 TurBoss sorry child
09:29 canton7 I'm trying to find out where there is a commit with *no parent*
09:29 TurBoss sorry i'm confused
09:29 TurBoss this never happended to me before
09:29 pfrench joined #git
09:29 canton7 you said (and I'm rephrasing), "I stage some files and make a new commit, and that new commit doesn't have a parent"
09:29 canton7 I'm asking what makes you think that you make a new commit which doesn't have a parent
09:30 canton7 so far I've seen nothing to suggest that you make a commit without a parent
09:30 TurBoss so my question is wrong
09:30 canton7 what is your question, then?
09:30 TurBoss http://i.imgur.com/bimz6Em.png
09:30 canton7 what are you trying to show in that screenshot?
09:30 TurBoss here is a screenshot of the initial commit
09:31 TurBoss the blue line is the NukedLibs branch
09:31 canton7 hmm, looks like GitKraken's confused
09:31 canton7 pastebin the output of `git log --graph --oneline --decorate --all`
09:31 GT4066 joined #git
09:31 TurBoss tryed source treer
09:31 TurBoss ok
09:31 navidr joined #git
09:32 leeN joined #git
09:33 rchavik joined #git
09:33 TurBoss canto7: http://dpaste.com/1WB6M6B
09:33 markzen left #git
09:33 TurBoss ops
09:33 TurBoss sorry canton7 ^
09:34 canton7 TurBoss, looks like GitKragen isn't showing remote branches
09:35 canton7 the " Removed some libs by accident" commit is actually based on TurBoss/master, which is "Merge pull request #5"
09:35 canton7 I've never used GitKraken, so i don't know how to make it display remote branches, or update local branches
09:35 TurBoss so history is fine?
09:36 TurBoss source tree also shows this
09:36 canton7 ah, it's also bsed on 'master', which GitKraken shows correctly
09:36 jay-m joined #git
09:36 osse why won't it show the rest of the commits? :O
09:36 osse but I guess if it truly was a commit without a parent there wouldn't be a line in the first place
09:36 canton7 hmmm actually, I don't know whether that git log output shows orphaned commits correctly
09:37 pfrench joined #git
09:37 TurBoss so i should shearch for orphaned commits?
09:38 inflames joined #git
09:38 canton7 uh, not orphaned commits, sorry. Commits without a parent
09:38 inflames joined #git
09:38 canton7 hmm, no it doesn't
09:38 _ADN_ joined #git
09:38 canton7 so GitKraken might be right
09:38 canton7 how did you create the NukedLibs branch?
09:39 TurBoss from master then create branch
09:39 chll_ joined #git
09:39 TurBoss i have tryed from pycharm and also from sourcetree
09:39 TurBoss to create the branch
09:39 mischat joined #git
09:40 seni13 joined #git
09:40 canton7 just to check
09:40 PHPanos joined #git
09:41 TurBoss this happend even if I commit to master
09:41 canton7 what's the output of 'git cat-file -p HEAD'?
09:41 TurBoss ok checking
09:41 canton7 (when you have NukedLibs checked out)
09:41 JeroenT joined #git
09:41 TurBoss http://dpaste.com/272AC8N
09:41 TurBoss done
09:42 TurBoss nukedLibs checked and command issued
09:42 canton7 right, so it *does* have a parent
09:42 canton7 the parent is 38e06d7001f
09:42 Cabanossi joined #git
09:42 canton7 and that's "Merge pull request #5 "
09:42 canton7 so goodness only knows what's confusing GitKraken
09:42 canton7 try 'gitk --all &' from the command line - that's reliable
09:43 TurBoss gitk shows fine
09:43 TurBoss sek
09:43 canton7 so, definitely a GitKraken issue /surug
09:43 canton7 *shrug
09:44 canton7 maybe the date on the commit "Removed some libs by accident" is older than the date on the commit "Merge pull request #5" (due to someone's clock being wrong), and that's confusing it?
09:45 TurBoss posible
09:45 canton7 just 'git show 38e06d7' and 'git show fb3733b'
09:45 TurBoss just to http://i.imgur.com/QSIaQ4g.png
09:45 canton7 ... I've got a meeting now, be bac, in 20 mins or so
09:45 TurBoss thank you
09:45 canton7 ah, gitk has the dates, don't worry about the git show
09:45 canton7 yeah, there we go: 9:25 and 9:54
09:45 canton7 so GitKraken's sorting by date, rather than by which commit is based on which other commit
09:46 canton7 and that's confusing it
09:46 TurBoss so if I wait 20 mins
09:46 TurBoss and make the commit again
09:46 TurBoss :D
09:46 canton7 so it's drawing "Removed some libs by accident" underneath "Merge pull request #5", and so failing to draw the line between them
09:46 canton7 heh, maybe :P
09:46 canton7 or fix your clock
09:46 TurBoss now its fixed
09:46 TurBoss duno what happened
09:46 canton7 right, I'm off!
09:46 TurBoss Thabks you so much
09:47 TurBoss *thank
09:47 thebird joined #git
09:54 marcogmonteiro joined #git
09:59 TurBoss bye
09:59 TurBoss left #git
10:02 pfrench joined #git
10:04 aard_ joined #git
10:05 dsdeiz joined #git
10:08 pfrench joined #git
10:08 JeroenT_ joined #git
10:09 JeroenT joined #git
10:09 Zio joined #git
10:09 Zio left #git
10:10 Silenced joined #git
10:11 kristofferR joined #git
10:12 Karazhan joined #git
10:13 pfrench joined #git
10:17 zeroed joined #git
10:17 zeroed joined #git
10:18 sarbs joined #git
10:24 pfrench joined #git
10:25 murii joined #git
10:25 sbasso joined #git
10:26 kadoban joined #git
10:27 Snugglebash joined #git
10:29 Insurget joined #git
10:29 Insurget left #git
10:30 pfrench joined #git
10:30 rchavik joined #git
10:34 ak77 answering to myself, ... to add git submodule during filter-branch --tree-filter, you have to add it at each step, set commit revision with  git update-index --cacheinfo 160000 $submodulegithash $submodulepath
10:34 orvckb joined #git
10:35 leeN joined #git
10:36 pfrench joined #git
10:37 SilverBe joined #git
10:37 irqq joined #git
10:40 pfrench joined #git
10:44 King_Hual joined #git
10:45 pfrench joined #git
10:47 jantje_ left #git
10:48 jantje joined #git
10:50 NwS joined #git
10:53 ferr1 joined #git
10:56 blitz206 joined #git
10:57 blitz206 Hi guys, quick question
10:57 Cabanossi joined #git
10:58 blitz206 I started working on master accidentally, then switched over to a feature branch. Right now, I need to pull down a fresh master and merge my features into this for a code review.
10:58 dsdeiz joined #git
10:59 osse blitz206: reset master back to where it should be
10:59 osse then you can pretend you worked on the feature branch all along
10:59 blitz206 when I 'git fetch origin master', I get ! [rejected]   master  -> master  (non-fast-forward)
10:59 mischat_ joined #git
10:59 canton7 are you *sure*. You're sure that's exactly what you type, and that's the exact message?
10:59 canton7 'git fetch' shouldn't want to update 'master', unless you've got some very weird configuration going on
11:00 blitz206 git fetch origin master:master
11:00 canton7 right, please make sure you give the commands exactly as you type them
11:00 PCatinean joined #git
11:00 blitz206 My apologies :)
11:01 canton7 the more usual approach is to get 'git fetch' to update origin/master, then reset master to origin/master
11:01 blitz206 Will I at all run the risk of doing anything to my feature branch given this odd workflow I've followed?
11:02 canton7 no, 'git fetch' will never update any of your local branches (unless you explicitly tell it to, like you did)
11:04 blitz206 Okay, so from here - what is recommended to do?
11:04 canton7 if you want to reset master to origin/master, 'git checkout master && git reset --hard origin/master'. I'm not entirely clear on what commits you've got where however
11:04 canton7 can you pastebin the top bit of 'git log --graph --oneline --decorate --all', which shows master and your feature branch?
11:04 * osse feels ignored
11:05 gtristan joined #git
11:05 canton7 also, listen to osse :D
11:05 chipotle joined #git
11:05 osse <3
11:05 blitz206 osse - sorry mate! I am not ignoring you at all
11:05 blitz206 I am afraid to lose anything :)
11:05 canton7 !backup
11:05 gitinfo Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.com/GitBestPractices/#backups
11:06 blitz206 You guys on on top of it
11:06 pfrench joined #git
11:07 blitz206 Ooph this includes a lot of commits
11:07 blitz206 from my entire company
11:07 canton7 I just want to see the bit containing your feature branch, and how it relates to master
11:07 osse blitz206: https://github.com/Osse/git-stuff/blob/master/git-reset-explanation.txt
11:07 canton7 so if you've only done a couple of commits on the feature, the top 5-10 lines should be fine
11:08 osse blitz206: that addresses your first point. move master back, but feature will still hold on to your recently created commits
11:08 blitz206 Even if said commits were done on master originally?
11:09 osse yews
11:09 osse assumnig you started the feature branch from where amster was then
11:09 osse as shown in my beautiful asciiart
11:09 blitz206 Some* that is. Messy work flow was feature, master, feature
11:09 osse ehh, ok. then i've misunderstood
11:09 osse back to showing us git log then
11:10 osse git log --graph --decorate --oneline master feature | head -n 20 is probably plenty.
11:10 borkr joined #git
11:10 Vampire0 joined #git
11:10 osse it just needs to show the "structure" of it. from right befire feature existed and up to now.
11:11 blitz206 Okay, so this is all I have
11:11 blitz206 *   361c9d9 (HEAD -> feature/LP-3118) fixed merge conflicts
11:11 osse both canton7 and myself are spoiled trust fund babies anyway so we have very little incentive to screw you over
11:12 canton7 I'm not exactly sure what "done on master originally" means. I want to make sure that your feature branch points to those commits, and that they haven't been pushed somewhere
11:12 JeroenT joined #git
11:13 JeroenT_ joined #git
11:13 osse blitz206: is that the complete output? :O
11:13 blitz206 My last two commands were: 'git fetch origin master:master' and 'git merge origin feature/LP-3118'
11:14 blitz206 That is the pertanant output - presumably because I fetched master and potentially overwrote anything I had going on there
11:15 ResidentBiscuit joined #git
11:16 blitz206 Here is another after sifting through... * | | | | e180b31 (LP-3118-explore, LP-3118) Fix bug in trimDefaultDateRange_
11:17 achadwick joined #git
11:19 osse let's try something else
11:19 osse what does 'git rev-list --count master..feature' say?
11:19 osse and 'git rev-list --count feature..master'
11:20 blitz206 What I can tell you is that my feature branch is working perfectly fine, and my master branch does not reflect the changes on my feature branch right now
11:20 blitz206 0, 109
11:21 osse ehhh
11:21 burzum left #git
11:22 osse ok, so if feature is fine, and you wish you didn't work on master in the first place, then i suggest you forcibly rewrite master so it's equal to origin/master and then be on your way
11:22 marenz_ joined #git
11:23 osse then you can merge feature into mster if you wish
11:23 blitz206 Okay, let me make sure feature is 100% :)
11:25 thebird joined #git
11:28 govg joined #git
11:30 mknod joined #git
11:31 JeroenT joined #git
11:32 j`ey joined #git
11:33 j`ey joined #git
11:33 j`ey possibly strange question. I want something like --no-prefix.. but for a directory. our source code is under 'a/b'. so 99% of the time I'm in the b directory
11:33 j`ey is there a way to get git diff etc to start from b?
11:34 osse j`ey: git diff --relative
11:34 moritz also when you apply the diff, you can strip prefixes with -p<number of prefixes>
11:35 j`ey osse: cool thanks
11:35 j`ey maybe I should make an alias
11:35 osse j`ey: what is outside the b directory?
11:36 j`ey some scripts and stuff
11:36 osse we're in a similar situation. we're planning of moving stuff outside b to some network share to be forgotten about and move the worthwhile bits up :)
11:38 j`ey theyre in lock-step with the rest of the code]
11:38 osse then we're not in a similar situation
11:39 osse we have a bunch of old crap from back when this used to be an SVN repo
11:39 j`ey it's just that I don't touch it 99% of the time, so pretending my git repo starts at b is nice
11:39 multi_io joined #git
11:42 jay-m joined #git
11:44 sbulage joined #git
11:47 jay-m joined #git
11:50 reverse_light joined #git
11:50 zeroed joined #git
11:53 imack joined #git
11:53 rnsanchez joined #git
11:54 osse ok, so we are in the same situation, but in my case we're more free to do something about it :p
11:54 j`ey :D
11:54 osse because in my case all of us are in this situation. so to speak
11:55 pfrench joined #git
11:55 pks joined #git
11:57 Cabanossi joined #git
11:58 Vampire0_ joined #git
11:58 davimore joined #git
12:02 pfrench joined #git
12:03 theoceaniscool joined #git
12:06 mizu_no_oto joined #git
12:06 jameser joined #git
12:06 Bennux joined #git
12:07 JeroenT joined #git
12:09 sarri joined #git
12:09 sarri joined #git
12:11 Stutters joined #git
12:13 sunri5e joined #git
12:14 sQVe joined #git
12:14 ResidentBiscuit joined #git
12:14 PHPanos joined #git
12:17 sbasso joined #git
12:19 pfrench joined #git
12:20 leitao joined #git
12:21 Snugglebash joined #git
12:22 muhannad_______ joined #git
12:23 thiago joined #git
12:23 xaviergmail joined #git
12:24 thecomedian_ left #git
12:26 thecomedian joined #git
12:28 pfrench joined #git
12:30 thecomedian joined #git
12:30 Bizkit joined #git
12:32 jimi_ joined #git
12:32 pfrench joined #git
12:33 Silenced_v2 joined #git
12:38 sbulage joined #git
12:38 absinthe joined #git
12:39 pfrench joined #git
12:40 chele joined #git
12:41 ToBeCloud joined #git
12:41 d0nn1e joined #git
12:42 jay-m joined #git
12:42 dviola joined #git
12:45 leeN joined #git
12:46 a3Dman joined #git
12:49 mischat joined #git
12:51 wxg joined #git
12:51 sbasso joined #git
12:51 jay-m joined #git
12:51 ankit01ojha joined #git
12:52 MattMaker joined #git
12:52 JeroenT joined #git
12:52 absinthe joined #git
12:53 MattMaker joined #git
12:53 geek2nerd joined #git
12:53 JeroenT_ joined #git
12:53 MattMaker joined #git
12:54 Tobbi joined #git
12:54 JeroenT joined #git
12:54 MattMaker joined #git
12:55 MattMaker joined #git
12:56 MattMaker joined #git
12:57 Cabanossi joined #git
12:57 ebcz joined #git
12:58 Vampire0 joined #git
12:59 thiago joined #git
12:59 pfrench joined #git
13:03 dviola joined #git
13:05 geek2nerd joined #git
13:05 masuberu joined #git
13:05 GreySunshine joined #git
13:06 ChadTaljaardt joined #git
13:09 xaviergmail joined #git
13:11 Darcidride joined #git
13:12 pfrench joined #git
13:13 kexmex joined #git
13:14 eh3- joined #git
13:15 kpease joined #git
13:15 kpease_ joined #git
13:17 jay-m joined #git
13:20 str joined #git
13:21 sQVe joined #git
13:25 xkr47 joined #git
13:27 ShapeShifter499 joined #git
13:27 geek2ne17 joined #git
13:28 geek2ne54 joined #git
13:32 Asenar joined #git
13:32 sgfgdf hello, guys! is it possible to create a custom git command like `git mycommand` without creating separate git-mycommand script which is in my PATH? can i do this with simple bash function?
13:32 dviola joined #git
13:32 osse sgfgdf: you can create an alias that runs shell code
13:32 JeroenT joined #git
13:33 bongjovi joined #git
13:33 echo2 joined #git
13:33 sgfgdf osse: but will it work like `git mycommand` or will be something like `g-mycommand`?
13:34 osse it will work like git mycommand
13:34 deepakbndeepu joined #git
13:34 geek2nerd joined #git
13:34 sgfgdf osse: how should i setup it? can you give a simple example, please?
13:34 osse git config --global alias.mycommand '!f() { shell code here; }; f'
13:35 a-l-e joined #git
13:36 sgfgdf osse: what is a better way to extend with such custom commands? create a function in some bash file and then this alias or just create a script with 'git-' name prefix?
13:37 osse sgfgdf: I'd go with script
13:37 osse because in both cases you need an extra file anyway
13:37 qqx I agree. Especially since I don't use bash.
13:38 mischat joined #git
13:38 King_Hual joined #git
13:38 deepakbndeepu left #git
13:39 osse sgfgdf: besides in the script case you don't have to touch .gitconfig.
13:39 pfrench joined #git
13:40 ij joined #git
13:40 kexmex joined #git
13:41 a_thakur joined #git
13:41 dsdeiz joined #git
13:42 dsdeiz joined #git
13:44 ij I've gotten "refname '…' is ambiguous", each time I was scratching my head as to why. How do I list all the refs that matched the name?
13:44 sgfgdf osse: will go with a script. thank you very much!
13:45 jameser joined #git
13:45 osse ij: git for-each-ref
13:45 osse ij: a classic is to have both origin/master and a local branch whose name is 'origin/master'
13:45 JeroenT joined #git
13:46 thebird joined #git
13:46 j`ey left #git
13:47 sz0 joined #git
13:47 pfrench joined #git
13:48 Bennux joined #git
13:50 ujjain joined #git
13:50 ujjain joined #git
13:50 j08nY joined #git
13:52 GodGinrai joined #git
13:56 jameser joined #git
13:56 manojbagari joined #git
13:57 Cabanossi joined #git
13:57 MattMaker joined #git
14:01 DBag joined #git
14:01 troulouliou_div2 joined #git
14:01 DarkPsydeLord joined #git
14:01 livingstn joined #git
14:02 JeroenT joined #git
14:04 jameser joined #git
14:04 JeroenT joined #git
14:06 jeffreylevesque joined #git
14:09 awal joined #git
14:09 thiago joined #git
14:10 elsevero joined #git
14:10 ExoUNX joined #git
14:10 awal hi. how exactly am I supposed to do this:
14:10 awal > If you’ve allowed everyone to connect with a single user (like “git”) via public-key authentication, you may have to give that user a shell wrapper that determines which user is connecting based on the public key
14:10 awal (from https://git-scm.com/book/en/v2/Customizing-Git-An-Example-Git-Enforced-Policy)
14:11 moritz awal: there are ready-made solutions for that, for example gitolite
14:12 nomadsoul left #git
14:12 moritz awal: I highly recommend using gitolite, it's awesome if you want a lightweight solution
14:12 MacGyver And I'd recommend not reimplementing that yourself.
14:12 MacGyver Because it's quite tricky to do in such a way that guarantees a user can't use it to break out.
14:12 danslo joined #git
14:12 wxg joined #git
14:13 mizu_no_oto joined #git
14:14 awal MacGyver, moritz: I am not sure I want to use gitolite - we have only two users and a handful of repos. Would you still recommend it?
14:14 perlpilot awal: I would.
14:15 awal Does gitolite start a daemon or something? Or does it just provide the shell wrapper?
14:15 mischat_ joined #git
14:15 perlpilot awal: it just leverages features of ssh (you need a gitolite user though)
14:16 Masber_080 joined #git
14:16 King_Hual joined #git
14:16 King_Hual joined #git
14:17 awal perlpilot: I really would like to know what exactly it does...
14:17 Gsham joined #git
14:18 perlpilot awal: read gitolite.com.  sitaram has lots of documentation
14:19 publio joined #git
14:19 uebera|| joined #git
14:19 thiago joined #git
14:19 porsche944 joined #git
14:21 awal perlpilot: Ah it is plenty descriptive and doesn't seem intrusive, I'll use this then.
14:21 awal thanks everyone :)
14:24 ResidentBiscuit joined #git
14:24 s2013 joined #git
14:26 bernardio joined #git
14:27 _ADN_ joined #git
14:27 iamthad joined #git
14:29 Bhootrk joined #git
14:29 JeroenT joined #git
14:30 danslo joined #git
14:30 dendazen joined #git
14:31 matt_m joined #git
14:34 JeroenT joined #git
14:34 Gsham joined #git
14:36 Gustavo6046 joined #git
14:38 ccolorado left #git
14:38 danslo_ joined #git
14:39 puff joined #git
14:39 puff Good morning.
14:40 cdg joined #git
14:41 cdg joined #git
14:42 boombatower joined #git
14:42 Gsham joined #git
14:43 mizu_no_oto joined #git
14:45 porsche944 a question on git branch --contains
14:45 porsche944 if you cherry-pick a commit from one branch to another, doesn't it get a new sha
14:45 avar yes, --contains won't catch that
14:45 porsche944 so how does contains know the commit is in both branches
14:46 zeth joined #git
14:46 avar it doesn't
14:46 Rhonda joined #git
14:46 porsche944 then what's the point of --contains
14:46 avar to find cases where multiple branches contain the same commit
14:46 porsche944 so are there cases where cherry-picking doesn't create a new SHA
14:47 qqx Plus it's more encouragement for people to use proper workflows (which don't use cherry picking).
14:47 porsche944 I've cherry-picked a few commits from our dev branch to our release
14:47 avar no, but there's plenty of cases where you want to find what commit a branch is on without the cherry-picking
14:47 porsche944 and --contains is finding them
14:47 porsche944 on both
14:48 porsche944 git branch -r --contains 39a2b7f
14:48 porsche944 origin/HEAD -> origin/develop  origin/develop  origin/release/9.0
14:48 perlpilot porsche944: the branch probably contains that commit *and* your cherry pick
14:48 oaao joined #git
14:48 porsche944 from a merge?
14:49 perlpilot yes, that's a possibility.
14:49 avar Yes, if you cherry pick e.g. from develop to release, then later merge develop to release, it'll --contain the non-cherry-picked version
14:49 porsche944 I'm pretty sure we didn't merge all of develop into release
14:49 mischat joined #git
14:50 PCatinean Hey guys
14:50 JeroenT joined #git
14:50 PCatinean I find git submoduled very confusing, how can I simply add a submodule targeting branch X inside another repo?
14:50 PCatinean I did it once and now I cannot redefine it
14:50 porsche944 so how would you ever be able to tell if you cherry-picked a commit from develop into release
14:51 osse porsche944: by using git cherry
14:51 porsche944 nothing is launching
14:51 porsche944 do I need to install something
14:51 avar You can also run git-patch-id, but more generally improve your workflow in some way to make this more obvious, because the connection in git's own model is always going to be loose
14:51 avar e.g. sometimes the patch-id isn't the same if the hunks moved around
14:52 avar One technique is hack some tool to tag commits you cherry-pick into release
14:52 osse porsche944: no it's built in
14:52 qqx My preferred way is to never cherry pick, so that the answer is always no.
14:52 porsche944 git-cherry - Find commits yet to be applied to upstream
14:52 porsche944 that doesn't sound like the same thing
14:52 avar and that tag can contain the sha1 of both commits involved..
14:52 porsche944 ok
14:53 osse porsche944: it is. for each commit in a range it will look for corresponding commits somewhere else and mark them accordingly
14:53 porsche944 it sounds like next time I should just work on the release branch and then merge to develop?
14:53 porsche944 osse: ok
14:53 qqx porsche944: Yes, that's the way I'd do that.
14:53 porsche944 all right makes sense
14:54 PCatinean is it because I used ammend?
14:54 qqx It also ensures that the change gets more testing in the release branch.
14:54 avar A technique we use @work where I am is whenever we cherry-pick to "release" we merge that cherry-picked commit back into the dev branch (where the merge is a noop 99% of the time)
14:55 avar So you leave a merge marker in the permanent history indicating what got cherry-picked
14:55 TomyWork joined #git
14:55 avar And that workflow also ensures that you can make *original* commits on the release branch, e.g. to fix some issue there, but then you'll always merge that to develop
14:56 ispn joined #git
14:56 avar So you don't have cases where you hotfix release and then rollout a new release and your "fix" gets undone
14:56 Stutters joined #git
14:58 PHPanos joined #git
14:58 MattMaker joined #git
14:58 Bennux joined #git
14:58 crose joined #git
15:02 Philippe joined #git
15:02 Timvde joined #git
15:02 mmlb joined #git
15:03 GreySunshine joined #git
15:05 Philippe left #git
15:06 Starky joined #git
15:06 ayogi joined #git
15:06 ij osse, It prints nothing. :(
15:07 ij git for-each-ref master, that is. rev-parse still says `master' is ambiguous.
15:07 awal I installed gitolite but when I try to clone the testing repository, it says Permission denied (public key). Do I have to restart sshd on the server?
15:08 canton7 awal, no, no need to restart sshd
15:08 danslo joined #git
15:08 awal I do the have private key for the public key on my machine and ssh-agent running and the key added
15:08 awal canton7: what could be wrong?
15:08 canton7 awal, you've definitely edited the config correctly, and pushed it to gitolite?
15:08 awal canton7: I haven't edited any config, I can't clone the gitolite-admin repo
15:09 Emperor_Earth joined #git
15:09 awal unless you are asking me to edit it on the server itself?
15:09 canton7 ah, you said the "testing repo", not the admin repo
15:09 canton7 no, don't do that
15:09 awal I can't clone either
15:09 IrishPrime joined #git
15:09 awal simple `ssh git@myserver` doesn't work either
15:09 canton7 you gave your private key to 'gitolite setup'?
15:09 awal public
15:09 plos joined #git
15:09 canton7 oops, yes
15:10 awal yes, I did
15:10 awal it is present in /home/git/.ssh/authorized_keys
15:10 awal with gitolite's preamble commands and options
15:10 PHPanos joined #git
15:10 canton7 and the correct public key is being given to ssh on the client? 'ssh -vvv git@yourserver' will tell you
15:10 canton7 what command exactly are you using to try and clone the gitolite-admin repo?
15:11 awal yes, it offers many keys, one of them is the corresponding key
15:11 awal git clone <url to gitolite on github>
15:11 canton7 the gitolite-admin repo, not gitolite itself
15:11 awal followed by gitolite install -ln ~/bin and gitolite setup -pk awal.pub
15:11 scoobert1on joined #git
15:11 awal ahh
15:12 Cabanossi joined #git
15:12 awal git clone git@myserver:gitolite-admin
15:12 awal I am not sure why though even ssh'ing to the server doesn't work and gives me a public key error...
15:13 mizu_no_oto joined #git
15:13 sitaram awal: http://gitolite.com/gitolite/sts/  <-- all your ssh troubles solved
15:13 canton7 yeah, that suggests it's a key issue
15:14 canton7 nice new theme btw :)
15:14 gugah joined #git
15:15 gugah joined #git
15:15 osse ij: git for-each-ref | grep master
15:15 DDBag joined #git
15:16 cagmz joined #git
15:17 gareppa joined #git
15:17 ij Ah!
15:17 AaronMT joined #git
15:17 ij But doesn't the man-page say it can do patterns?
15:18 osse ij: try '*master*'. They're not regexes
15:18 awal canton7: hmm using a different key seems to work. Sorry, my bad!
15:18 canton7 no worries! sometimes it's good to rubber-duck
15:18 awal sitaram: hi! thanks for the great software
15:18 Rodya_ joined #git
15:19 TomyWork i want to use git as a binary patch tool. i want to create repeatable commits, down to the SHA1 sum, on both sides, then add one commit on the source side and transfer it over as a bundle. Is that possible like that?
15:19 SupaHam joined #git
15:20 awal TomyWork: the diff'ing operates on line-level, which would suggest this is not possible, no? Just a guess
15:20 impliednude joined #git
15:21 TomyWork awal no i dont want a (textual) diff/patch file as a result
15:21 TomyWork i want to upgrade a file on the target machine without transferring a ton of data
15:21 TomyWork (the reason being the line is extremely thin)
15:21 TomyWork and unreliable
15:22 TomyWork so i want to condense it down to the minimum by only sending the different parts
15:23 bremner TomyWork: it would be simpler to use xdelta
15:24 fees joined #git
15:24 TomyWork bremner oh, i can get that on both sides, neat
15:26 absinthe joined #git
15:26 Darren_ joined #git
15:26 Silenced_v2 joined #git
15:27 TomyWork shit, it's barely smaller :D
15:27 TomyWork i think the files are internally compressed :/
15:28 aca20031 joined #git
15:29 dandaman joined #git
15:29 robotroll joined #git
15:31 rkazak_ joined #git
15:33 stduolc joined #git
15:33 jagob joined #git
15:33 sitaram canton7: :)
15:34 sitaram awal: you're welcome!
15:34 sitaram awal: if you're using ssh-agent, and you *don't* have the line 'IdentitiesOnly yes' in ~/.ssh/config, ssh picks unexpected keys
15:35 raijin joined #git
15:36 sitaram canton7: I did it mainly because of the built-in search in this one
15:36 awal sitaram: Oh? I never knew this. I am using ssh-agent indeed, and I don't have that line in my config. Do you mean it picks random keys?
15:37 Grigora joined #git
15:37 canton7 how come my 'ssh -vvv' trick didn't show that, then?
15:37 awal ssh -v did show the key being offered to the server
15:37 awal so I guess in this case my key pairs are malformed or something
15:38 awal I will check that later on
15:39 fstd_ joined #git
15:39 EV3RYDAYPR0GRESS joined #git
15:40 Grigora Hi, I have probably a rather basic question - I made a commit which was supposed to be broken into three commits (interactive/patch) so I did git reset HEAD~, then committed the same changeset in three commits. My question is - is this the correct approach and why am I seeing the original commit which I was trying to undo (presumably doesn't hurt anything but still, not clean).  This is a local repository, no pushing/pulling. Thanks!
15:41 WeiJunLi joined #git
15:43 mizu_no_oto joined #git
15:43 Learath2 Grigora: where do you see the original commit?
15:43 Guest4561 joined #git
15:43 Guest4561 left #git
15:44 Grigora Learath2: git log --oneline file
15:45 Learath2 idk what you did exactly but what you were looking for is "git reset --mixed HEAD~ && git add -i && git commit..."
15:46 pfrench joined #git
15:46 Learath2 shouldn't see the original tho :/ hmm
15:47 Grigora Learath2: I did almost that, except the --mixed part. My git is v1.7.1, don't ask me why
15:48 Learath2 --mixed is the default behaviour for git reset tho
15:48 Grigora Learath2: one more question - what would happen if I did git reset HEAD~ filename ?
15:49 shinnya joined #git
15:49 Grigora Learath2: I may have done that, don't remember.
15:50 astrofog joined #git
15:51 orbyt_ joined #git
15:51 fk joined #git
15:51 Learath2 Grigora: it would just reset that specific file, leaving the commit in place i believe
15:52 thebird joined #git
15:52 Grigora Learath2: then maybe that explains the behavior I'm seeing, no? Thanks for your help!
15:52 marianina8 joined #git
15:53 ansimation joined #git
15:54 sitaram awal, canton7: specifically, if you use "-i" or "IdentityFile", and that file is *not* in "ssh-add -l"'s output, then it won't be used.  Instead, it will use whatever is held by the agent, ignoring the "-i" or "IdentityFile"
15:54 LeBlaaanc joined #git
15:54 ansimation joined #git
15:56 Vampire0_ joined #git
15:56 bruce_lee joined #git
15:56 ij Will "git fetch remote" download a ref that's only pointed to by remote's HEAD? It doesn't seem so, but could I make it do that?
15:56 elect_ joined #git
15:57 _Alex-N_ joined #git
15:57 selckin create a branch instead
15:58 elect_ joined #git
15:58 ij :(
15:58 al-damiri joined #git
15:58 mizu_no_oto joined #git
15:58 raspado joined #git
15:59 MattMaker joined #git
15:59 nettoweb joined #git
16:01 elect__ joined #git
16:03 NotThizTime joined #git
16:04 zivester joined #git
16:04 tang^ joined #git
16:05 ResidentBiscuit joined #git
16:05 SupaHam joined #git
16:08 DDDBag joined #git
16:11 cluelessperson joined #git
16:12 dgonzo joined #git
16:12 Cabanossi joined #git
16:12 SupaHam joined #git
16:13 a3Dman joined #git
16:14 TbobbyZ joined #git
16:15 TbobbyZ joined #git
16:15 overlord_tm joined #git
16:17 danslo joined #git
16:17 clemf joined #git
16:19 SupaHam joined #git
16:20 ResidentBiscuit joined #git
16:21 thiago joined #git
16:22 miczac joined #git
16:24 SupaHam joined #git
16:24 ych joined #git
16:24 dviola joined #git
16:25 jozwior joined #git
16:26 dgonzo joined #git
16:27 dreiss joined #git
16:28 Aeterne joined #git
16:28 mizu_no_oto joined #git
16:31 dgonzo joined #git
16:33 pfrench joined #git
16:35 brent__ joined #git
16:37 brent__ joined #git
16:41 murii_ joined #git
16:41 xissburg joined #git
16:42 bernardio joined #git
16:46 dsteplight__ joined #git
16:47 s1scha joined #git
16:47 eroux joined #git
16:47 sathed joined #git
16:48 YuGiOhJCJ joined #git
16:49 GreySunshine left #git
16:53 geek2nerd joined #git
16:55 metachr0n joined #git
16:55 thiago_ joined #git
16:56 ALIASE joined #git
16:57 ALIASE left #git
16:57 jstimm joined #git
16:58 thebird joined #git
16:59 InfoTest joined #git
16:59 awal sitaram: I see, thanks for the info
17:00 MattMaker joined #git
17:00 raspado joined #git
17:04 Dougie187 joined #git
17:07 alexandre9099 joined #git
17:09 Murii__ joined #git
17:10 madewokherd joined #git
17:12 Cabanossi joined #git
17:12 jost joined #git
17:12 inflames joined #git
17:13 BackEndCoder joined #git
17:13 mizu_no_oto joined #git
17:13 multi_io joined #git
17:14 LiquidatorBrunt joined #git
17:15 ayogi joined #git
17:16 LiquidatorBrunt sup git
17:16 a-l-e left #git
17:16 j08nY joined #git
17:18 nsh joined #git
17:21 enkrypt joined #git
17:21 eroux joined #git
17:24 ensign_ joined #git
17:24 elsevero joined #git
17:25 muhannad_______ joined #git
17:26 cdown joined #git
17:27 jimklo joined #git
17:30 LiquidatorBrunt question with visual studio and git, if i checkout a branch and then compile it
17:30 LiquidatorBrunt the output in bin is that branch?
17:33 Random832 probably - you might want to clean just to make sure
17:33 gugah joined #git
17:36 svm_invictvs joined #git
17:38 Bennux joined #git
17:39 jungsubk joined #git
17:39 dave0x6d joined #git
17:43 Starkythefox joined #git
17:44 raijin joined #git
17:44 chardan joined #git
17:45 Ryanar joined #git
17:45 Starky joined #git
17:45 Starky joined #git
17:46 rwp LiquidatorBrunt, No. The output in bin is just built files. Those initially will be new "untracked" files. You should probably add those patterns to the .gitignore file so that they are not checked in.
17:46 eh3 joined #git
17:46 thiago joined #git
17:48 clandestino joined #git
17:48 gopar joined #git
17:52 theoceaniscool joined #git
17:54 ertesx joined #git
17:56 chrisgeorge joined #git
17:58 bernardio joined #git
18:01 vtols joined #git
18:01 durham joined #git
18:01 Random832 i think i'd been assuming they were already ignored, but may not have all been rebuilt when he compiles it
18:02 chardan joined #git
18:02 thiago LiquidatorBrunt: build products are not in any branch
18:02 thiago they don't belong in the repository
18:03 Tobbi joined #git
18:04 brent__ joined #git
18:06 radoslavcap joined #git
18:07 Taikibansei joined #git
18:10 kexmex joined #git
18:10 Hozy joined #git
18:10 max3 joined #git
18:11 rivarun joined #git
18:12 radoslavcap joined #git
18:13 mizu_no_oto joined #git
18:15 cgdub joined #git
18:18 dviola joined #git
18:19 LiquidatorBrunt i know, they are not in the repo
18:19 Hozy joined #git
18:20 LiquidatorBrunt but i am asking, how do i compile the branch i am on?
18:20 grawity click "Build", I assume
18:20 grawity or "Clean and build" if that's a thing?
18:20 DBag joined #git
18:20 DBag left #git
18:20 osse LiquidatorBrunt: that has nothing to do with git. git just changes the files on disk when you switch branches
18:21 osse presumably the build procedure is the same amongst them
18:21 jnavila joined #git
18:21 grawity either way, it just compiles whatever actual files it finds
18:21 Nacho_Libre and doesn't change branches
18:21 grawity if git has checked out branch1, then VS will compile branch1
18:22 durham joined #git
18:24 pifon joined #git
18:24 drs joined #git
18:24 LeBlaaanc joined #git
18:26 hahuang65 joined #git
18:28 mizu_no_oto joined #git
18:28 brokensyntax joined #git
18:33 eroux joined #git
18:36 dreiss joined #git
18:37 TikityTik joined #git
18:39 einnjo joined #git
18:40 thebird joined #git
18:40 cdg joined #git
18:42 Cabanossi joined #git
18:43 adac joined #git
18:43 howitdo joined #git
18:43 howitdo joined #git
18:43 mizu_no_oto joined #git
18:44 hahuang61 joined #git
18:45 grawity hmm, didn't know `git checkout` requires that much RAM
18:45 grawity over 500 MB when trying to checkout a linux.git commit
18:45 SwiftMatt joined #git
18:46 peeps[lappy] joined #git
18:46 enleeten joined #git
18:46 cdg joined #git
18:47 duderonomy joined #git
18:48 LiquidatorBrunt what's a good way to handle something a in config files such as a database connection that are needed for an application to run?
18:49 fstd joined #git
18:51 brent__ joined #git
18:51 brent__ joined #git
18:52 Jack[z] joined #git
18:53 qqx LiquidatorBrunt: !configs
18:53 gitinfo LiquidatorBrunt: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas
18:56 Flotho66 joined #git
18:56 miczac joined #git
18:59 MattMaker joined #git
19:02 war joined #git
19:02 SlimmyJimmy joined #git
19:02 einnjo joined #git
19:03 ratdeptrai joined #git
19:03 Bob8989|2 joined #git
19:05 SlimmyJimmy Im trying to install a package with NPM and that package apparently requiers that I install GIT. Is GIT installed on a per-project basis or should I install that permanently for use with NodeJS?
19:06 Gsham joined #git
19:07 grawity it's a standalone program and is usually installed system-wide
19:12 raynold joined #git
19:13 jozwior joined #git
19:13 TomyLobo joined #git
19:14 diogenese joined #git
19:15 kasual joined #git
19:16 davidfetter_ge joined #git
19:16 davidfetter_ge hi
19:17 davidfetter_ge i have accrued the work of a couple of co-workers into a patch. is there some standard way to name multiple authors? git 2.12.0 if it matters.
19:17 Naan joined #git
19:17 einnjo joined #git
19:18 durham joined #git
19:18 grawity not really
19:19 nixjdm joined #git
19:19 grawity adding some footers like "Coauthored-By: Fred Foobar" is common though
19:19 davidfetter_ge welp. i guess i'll just mention it in the commit message, just like that
19:19 robinsmidsrod joined #git
19:19 qqx Fairly recent patch and discussion about that: http://public-inbox.org/git/1485713194-11782-1-git-send-email-schumacher@kde.org/
19:20 Sasazuka joined #git
19:21 hahuang65 joined #git
19:21 apotry joined #git
19:23 cdown joined #git
19:24 _joes__ joined #git
19:25 * davidfetter_ge notes that this is a Wicked Problem™ and moves on
19:27 SlimmyJimmy I think the "Error: not found: git" actually means the package folder now that I read it again.
19:27 eh3 How can i see if a repo has any submodules without cloning it and running the submodule command? I need to check through a couple dozen or so, and find all the submodule usage.
19:28 grawity all "how do I look into a remote repo's contents" questions have the same answer: clone it first
19:28 grawity --depth=1 if you're low on bandwidth
19:28 eh3 grawity : thanks!
19:29 xissburg joined #git
19:29 cdg joined #git
19:30 qqx If there's a web interface (e.g. github) you could also look for a .gitmodules file at the top level.
19:30 durham joined #git
19:32 davidfetter_ge left #git
19:32 max3 joined #git
19:34 MatthewChatham joined #git
19:36 MatthewChatham hello
19:36 gitinfo MatthewChatham: 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.
19:36 MatthewChatham hello
19:39 Eugene grawity - FWIW, Co-Authored-By is something of a standard. https://git.wiki.kernel.org/index.php/CommitMessageConventions
19:40 _joes__ joined #git
19:41 cdg_ joined #git
19:42 jnavila joined #git
19:46 fstd joined #git
19:47 faux___ joined #git
19:49 gugah_ joined #git
19:52 cwillu_at_work joined #git
19:54 pur3eval joined #git
19:55 math-alpha joined #git
19:55 thiago joined #git
19:56 Vampire0_ joined #git
19:56 s1scha joined #git
19:56 tango_ joined #git
19:57 Noldorin joined #git
19:58 t0by joined #git
20:01 MattMaker joined #git
20:05 barteks2x after a few years of using git I still don't remember which way git merge works... current branch into specified branch or specified branch into current? Anyone knows some "trick" to remember it? It's a bit annoying to always look it up
20:06 bremner barteks2x: git generally modifies the current branch
20:07 nickabbey joined #git
20:07 bremner reset, merge, cherry-pick
20:07 blitz206 joined #git
20:07 barteks2x I guess that would make sense then
20:08 blitz206 Okay - how do I deal with a non-fast-forward?
20:09 ToxicFrog blitz206: as in a non-fast-forward-merge? "deal with" in what sense?
20:09 blitz206 I have a feature branch and a master
20:09 ToxicFrog Like, did you want it to be an FF merge and want to know how to roll back and make that happen? Are you having a conflict and not sure how to resolve it?
20:09 blitz206 the feature branch contains everything I want to submit for code review, and the master is recently pulled down
20:10 qqx In that situation I think the best thing to do is nothing. Just submit the branch for review.
20:10 Vampire0 joined #git
20:10 Vampire0 joined #git
20:11 blitz206 The instructions to submit via Arcanist are as follows:
20:11 sq-one joined #git
20:11 blitz206 1) Rebase your branch on top of master. Do not merge master into your branch again until the code review has been accepted.
20:11 blitz206 git fetch origin master:master git rebase master
20:11 qqx Well then it told you what needs to be done.
20:12 blitz206 Right, but I am receiving a rejected error message.
20:12 irqq joined #git
20:12 blitz206 [rejected]        master     -> master  (non-fast-forward)
20:13 qqx You can use `git fetch origin +master:master`, but that will cause any local commits on master to be potentially lost.
20:13 tmg joined #git
20:13 ToxicFrog blitz206: that looks like the output of 'git push'
20:14 qqx Another option would be `git fetch origin; git rebase origin/master`.
20:15 blitz206 Will that option affect my feature branch in any way?
20:16 qqx The first command I stated wouldn't. It would need to be followed by the rebase command from the Aranist instructions. My second command included the rebase. Combined either of those would rewrite history in your feature branch.
20:17 a_thakur joined #git
20:17 Silox| joined #git
20:19 blitz206 Well I am not looking to overwrite my feature branch
20:19 blitz206 I am definitely a bit confused on the fundamentals
20:20 qqx blitz206: Apparently if you're going to submit to that project you're required to rewrite history.
20:20 qqx That *doesn't* mean that the contents of your commits are lost.
20:20 mizu_no_oto joined #git
20:20 qqx It's mainly that the commit IDs change. You may also need to resolve conflicts during the rebase.
20:21 blitz206 Okay, so I do not run the risk of losing any content, only commit history?
20:22 pfrench joined #git
20:22 qqx Right, even if something goes horribly wrong your original commits will still be in the reflog for some time.
20:22 qqx At least as long as you don't completely wipe out the repo.
20:22 blitz206 Okay, let me give this a shot :) Thanks for the help and I will let you know if something does go horribly wrong.
20:23 blitz206 I made a repo backup as well.
20:23 johnny56 joined #git
20:23 qqx The most likely problem would be an error while resolving a conflict. That can always be undone with `git rebase --abort` and restart the rebase.
20:25 geoid joined #git
20:26 Starky joined #git
20:28 Darren_ joined #git
20:29 jimklo joined #git
20:29 jimi_ joined #git
20:30 johnny56 joined #git
20:30 Vampire0_ joined #git
20:32 radoslavcap joined #git
20:35 gopar joined #git
20:37 gopar joined #git
20:39 xaviergmail joined #git
20:39 dmin7b5 joined #git
20:39 nickabbey joined #git
20:42 Cabanossi joined #git
20:42 Dazzyp joined #git
20:44 geoid_ joined #git
20:46 radoslavcap joined #git
20:47 irqq_ joined #git
20:48 Flotho66 joined #git
20:48 m0viefreak joined #git
20:52 kbs joined #git
20:53 Jackneill joined #git
20:55 Sasazuka_ joined #git
20:58 Vampire0 joined #git
20:59 pfrench joined #git
21:01 MattMaker joined #git
21:02 dhollinger joined #git
21:04 Tobbi joined #git
21:05 blitz206 I'm still getting [rejected]        master     -> master  (non-fast-forward)
21:05 blitz206 When git fetch origin master:master in my feature branch
21:07 osse blitz206: you need to fetch then merge
21:07 osse blitz206: just git fetch, then git merge origin/master
21:07 johnny56 joined #git
21:12 blitz206 When I did my rebase, my changes disappeared
21:14 thiago did you submit them to the upstream repository?
21:14 SilverBe joined #git
21:14 aielima joined #git
21:14 blitz206 No, nothing is getting pulled
21:15 blitz206 pushed rather*
21:15 pur3eval joined #git
21:16 thiago are you sure that the rebase finished?
21:17 evanc joined #git
21:18 ankit01ojha joined #git
21:18 Led_Zeppelin joined #git
21:19 Led_Zeppelin need some help reverting to a certain hash. Basically, I want to checkout a certain version and check in to make it the head.
21:19 thiago Led_Zeppelin: are you sure you want to throw away everything that has happened since?
21:19 Led_Zeppelin no. def don't want to two away anything.
21:19 WeiJunLi joined #git
21:19 thiago ok, so you want to undo the changes that that commit made?
21:20 Led_Zeppelin realized that the previous commit was good. the new changes I made are wrong. so just want to re-commit the previous hash
21:20 osse is there only one commit made after the last good one?
21:21 Led_Zeppelin no. more than 1 commit. I think 4 commits
21:21 thiago is the commit you don'tw ant the last one?
21:21 thiago do you just want to throw away that commit?
21:22 Led_Zeppelin this is a good commit: 83de6a. I made 4 more commits (these all failed my regression test). So, I want to checkout 83de6a and make that my head. I still wait to keep those old commits.
21:22 valize joined #git
21:22 thiago you can't have both
21:23 nedbat Led_Zeppelin: you can keep those 4 bad commits on a branch
21:23 muhannad_______ joined #git
21:23 thiago if you make 83de6a your master (not HEAD), then everything that has happened since gets thrown away
21:23 Led_Zeppelin hmm. was thinking of checkout 83des seperately, then do a rsync and then git add --all then commit :-p
21:24 Led_Zeppelin already working on a branch, BTW. don't have access to commit on master.
21:24 thiago it's your repository. You have access to all branches.
21:24 thiago your repository on your machine
21:25 Led_Zeppelin true. so what you suggest?
21:25 kharloss joined #git
21:25 masuberu joined #git
21:26 thiago do you know if the mistake is in a single commit?
21:26 mizu_no_oto joined #git
21:26 Led_Zeppelin made multiple mistakes. I tried to fix it by commit->push->run CI. Then relized i was going the wrong path. so I just want to revert back to my old state
21:27 thiago so you do want to throw away all that has happened since the original mistake in that branch
21:27 thiago correct?
21:28 Led_Zeppelin no. don't want to trow away. I want to commit saying. "reverting back to 83de6a" while still keeping the bad 4 states.
21:28 Led_Zeppelin sorry for being so stubborn. don't want to throw away my mistakes :p
21:28 thiago git checkout 83de6a -- .
21:28 thiago git commit
21:29 thiago don't forget the . at the end
21:29 Led_Zeppelin let me try that
21:33 Led_Zeppelin You rock! worked like a champ!
21:33 Led_Zeppelin only spent 2 hrs trying to do this in many variations....
21:35 luke_nukem joined #git
21:38 osse keep in mind that won't remove new files that where added
21:38 osse *were
21:41 averell joined #git
21:42 Cabanossi joined #git
21:44 Silenced joined #git
21:44 WayToDoor joined #git
21:48 vuoto joined #git
21:51 lucido-cl joined #git
21:53 pfrench joined #git
21:55 absinthe joined #git
21:56 Led_Zeppelin ah I see.
21:59 causasui joined #git
21:59 causasui joined #git
21:59 elsevero joined #git
22:00 eclecticjohny joined #git
22:02 MattMaker joined #git
22:05 luke_nukem joined #git
22:11 durham joined #git
22:16 Lunatrius` joined #git
22:23 anuxivm joined #git
22:23 durham joined #git
22:31 pur3eval joined #git
22:32 FlavioZ joined #git
22:33 borkr joined #git
22:35 durham joined #git
22:36 ocelotsloth joined #git
22:40 pfrench joined #git
22:42 Noldorin joined #git
22:45 Kaisyu joined #git
22:45 durham joined #git
22:46 avar Why is it that when someone does:
22:47 chachasmooth joined #git
22:47 avar git push origin some/branch-name:some/other-branch-name
22:47 avar they might get "error: unable to push to unqualified destination "
22:47 avar unless they do:
22:47 avar git push origin some/branch-name:refs/heads/some/other-branch-name
22:47 avar But other clients can push a local branch like that just fine?
22:47 preaction my guess is because tags are a thing
22:47 dviola joined #git
22:48 avar I thought it was some local ref, but had them run "git for-each-ref | grep some" and they just had that local branch
22:48 avar So this was me helping someone today, I found the relevant code, which has to do with ref disambiguation: https://github.com/git/git/blob/master/remote.c#L1197c
22:48 avar I thought it might have to do with push.default somehow, but git -c push.default=upstream ... didn't help
22:48 avar we had the same git version (2.12.0)
22:49 avar but for some reason they needed refs/heads/ on the RHS of the :, and I didn't
22:49 pfrench joined #git
22:51 avar So given that there's no conflicting branches/tags locally, I can't find what might cause that disambiguation to be needed..
22:51 avar But I didn't get access to their repo/machine, so I couldn't test it myself
22:52 SwiftMatt joined #git
22:55 durham joined #git
23:03 MattMaker joined #git
23:04 tvw joined #git
23:07 Sound joined #git
23:07 smurfendrek123 joined #git
23:07 durham joined #git
23:08 luke_nukem joined #git
23:08 smurfendrek123 Hello, is there a git provider that allows you to generate a link that provides read only access only to people that have the link?
23:08 smurfendrek123 kinda like you can in google docs
23:09 Learath2 joined #git
23:09 avar github?
23:09 avar any github-like provider that allows you to create repos and configure collaborators seperately
23:09 smurfendrek123 no without them having an account
23:10 smurfendrek123 i want to give people read only access with the link
23:10 smurfendrek123 without them even having an account
23:10 osse still github.
23:10 smurfendrek123 but i don't want to make my repo public
23:10 chachasmooth joined #git
23:10 osse hmm
23:10 osse maybe bitbucket then
23:10 osse they offer free private repos
23:11 gugah_ joined #git
23:11 osse not sure if you can generate such a link. but it wouldn't surprise me if you could. makes sense
23:12 raspado joined #git
23:12 fletom joined #git
23:17 gugah__ joined #git
23:19 belak smurfendrek123: github's gists let you do that, but it's only really recommended for smaller repos
23:19 chardan joined #git
23:21 smurfendrek123 belak, that won't do
23:21 smurfendrek123 it has a lot of folders
23:21 xissburg joined #git
23:21 smurfendrek123 i can also just put it on drive and link to that
23:21 smurfendrek123 won't be a repo but all the files will be there
23:21 belak I'm not aware of a service which lets you share a link to a private repo
23:21 smurfendrek123 thanks for the suggestion anyways
23:21 YuGiOhJCJ joined #git
23:21 raijin joined #git
23:22 gopar joined #git
23:23 pfrench joined #git
23:23 gopar joined #git
23:23 wxg joined #git
23:27 solution_ joined #git
23:27 anry joined #git
23:32 pur3eval joined #git
23:34 dendazen joined #git
23:34 Darren_ joined #git
23:39 xaviergmail joined #git
23:39 xissburg joined #git
23:40 durham joined #git
23:40 pfrench joined #git
23:43 livingstn joined #git
23:44 pfrench joined #git
23:48 sgfgdf left #git
23:49 cust joined #git
23:51 Semtex-h joined #git
23:56 peterbecich joined #git
23:56 dandaman joined #git
23:57 Cabanossi joined #git

| Channels | #git index | Today | | Search | Google Search | Plain-Text | summary