Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-20

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

All times shown according to UTC.

Time Nick Message
00:00 mrslaughter1775 joined #git
00:02 Cxcf Does anyone know what magit means when it lists Head, Merge, and Pull in the magit:git-status buffer? Is it telling me to merge or telling me to pull?
00:09 irc3k im a fascist now, survival of the fittest
00:10 irc3k no more free education, who gives a fuck about humanity and shit.
00:10 irc3k who knows what'll happen
00:10 irc3k genes mix and shit, so it gets complicated
00:12 irc3k i have to get out of here
00:13 sbasso joined #git
00:14 Cxcf irc3k: u ok?
00:14 irc3k lol no, haven't eaten in 2 days
00:14 irc3k isnt painful xD
00:15 irc3k anyway, Im' more worried abuot society
00:15 irc3k i dont understand it
00:15 irc3k I ate a few nuts today
00:15 Cxcf irc3k: https://www.soylent.com/
00:15 irc3k my eggs were gone? can't afford rent if I eat
00:15 irc3k i must steal food from mom, I can't take the screaming so I rather not eat
00:17 irc3k i wanna pee in a glass
00:17 irc3k must go up
00:17 irc3k they just ban me when I talk about savages and parasites
00:17 irc3k im not getting a cent
00:17 irc3k im just existing
00:17 irc3k and we're all going to die and my genes are everywhere in the genepool anyway so
00:18 irc3k i wont pee
00:18 WantFood wtf is this.
00:18 irc3k WantFood: your nick lol
00:18 irc3k WantFood: i must get up to pee
00:19 irc3k no bremner
00:19 irc3k stop.
00:19 bremner irc3k: stop.
00:19 irc3k bremner: what do you want me to do instead
00:19 TbobbyZ joined #git
00:19 bremner irc3k: if you have a git support question, ask it.
00:19 amdi_ joined #git
00:19 irc3k im just getting banned lol. admit it, people like me should get banned and killed
00:19 bremner otherwise, find something else to do, you're annoying people.
00:20 irc3k bremner: ive tried to discuss the value of human life.
00:21 _28_ria joined #git
00:23 Cxcf How does git determine which branches are available to contributors and which are available to collaborators? I think I may be misunderstanding something. I guess my question is, how do contributors collaborate on specific branch?
00:24 bremner Cxcf: git per se doesn't do that. Do you mean github?
00:24 Cxcf bremner: yes, you're right, github
00:24 Cxcf Is there a different channel for github related issues?
00:24 bremner you may try asking on #github
00:25 neunon joined #git
00:25 bremner or someone may answer here
00:25 Cxcf gotcha, thanks, I'll keep eyes on both:)
00:27 amdi_ joined #git
00:29 andrew710 joined #git
00:35 Cxcf to clarify, git fetch just updates the tracking branches. will this tell you if you have merge conflicts? Or does only happen with a pull
00:36 mizu_no_oto joined #git
00:40 minn joined #git
00:40 roelmonnens joined #git
00:40 cqi joined #git
00:41 bremner you only find merge conflicts when you merge
00:41 gugah joined #git
00:42 livingstn joined #git
00:43 Guest2_ joined #git
00:45 nwmcsween_ joined #git
00:53 Cxcf bremner: thanks
01:09 rewt pull is a shortcut for fetch+merge
01:09 DarkPsydeLord joined #git
01:09 anth0ny joined #git
01:11 shgysk8zer0_ joined #git
01:13 arescorpio joined #git
01:16 dreiss joined #git
01:17 nwmcsween__ joined #git
01:17 jimmyrcom_ joined #git
01:23 TbobbyZ joined #git
01:26 a_thakur joined #git
01:28 Emperor_Earth joined #git
01:29 meLon joined #git
01:35 roelmonnens joined #git
01:38 qpdb joined #git
01:40 drodger joined #git
01:41 gugah joined #git
01:47 dsdeiz joined #git
01:47 dsdeiz joined #git
01:50 karstensrage is there a way to undo changes like add --patch but checkout --patch -
01:51 dvaske joined #git
01:54 Cabanossi joined #git
01:55 nowhereman joined #git
01:55 skered joined #git
01:57 tang^ joined #git
01:59 karstensrage wow git checkout has --patch too
01:59 karstensrage thats awesome
01:59 preaction karstensrage: reset --patch. reset undoes add
02:00 gugah_ joined #git
02:02 Naan left #git
02:04 nevodka joined #git
02:08 pks joined #git
02:10 madewokherd joined #git
02:14 Es0teric joined #git
02:14 karstensrage no this wasnt committed yet
02:15 karstensrage checkout --patch was exactly what i wanted
02:17 Nilesh_ joined #git
02:18 pickfire j416: https://public-inbox.org/git/CACsJy8D+VKGEtq6yMe+ZhfxKX2peCVXGZfZFckkZF_L=bWBqTg@mail.gmail.com/
02:18 pickfire Does that mean that I can't join anymore? Looks like the application deadline has passed.
02:19 d^sh joined #git
02:19 nowhereman joined #git
02:21 mda1 joined #git
02:21 andrew710 joined #git
02:24 finalbeta joined #git
02:25 jon-mac joined #git
02:27 mizu_no_oto joined #git
02:27 menip joined #git
02:29 Cxcf joined #git
02:29 Cxcf joined #git
02:30 roelmonnens joined #git
02:31 meLon joined #git
02:31 LeBlaaanc joined #git
02:34 jameser joined #git
02:35 TbobbyZ joined #git
02:39 Noldorin joined #git
02:46 hahuang65 joined #git
02:46 sunri5e 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
02:50 fstd_ joined #git
02:51 nebodka joined #git
02:55 moL-Ou joined #git
02:56 Derperperd joined #git
02:58 learning_ joined #git
02:59 nevodka_ joined #git
03:01 Cxcf joined #git
03:01 Cxcf joined #git
03:01 sbulage joined #git
03:03 xaviergmail joined #git
03:05 ojdo joined #git
03:07 nevodka joined #git
03:08 v0yAgEr joined #git
03:13 luke_nukem joined #git
03:13 v0yAgEr left #git
03:14 Geo joined #git
03:15 rewt joined #git
03:15 mizu_no_oto joined #git
03:16 luke_nukem joined #git
03:17 lagothri1 joined #git
03:17 Geo Hi, I have a develop branch, a release/X.Y branch that we package up release candidates and stable on (all the specific release stuff we dont want on develop), and then a stable/X.Y branch where the final version goes as a squash commit, which we then tag as vX.Y.Z  . As such, the stable/ branch only holds single commits that encompass all the work done since the last version.
03:18 Geo Question is- is there some way I can use git rev-list vX.Y..vX.Z and get all the individual commits that took place between those two tags on the develop branch?
03:18 Vampire0_ joined #git
03:19 chachasmooth joined #git
03:20 nebodka joined #git
03:22 andrew710 joined #git
03:24 nebodka joined #git
03:24 rewt Geo, git doesn't track where commits originate, only that they're committed after some other commit... you could look at the timestamps of vX.Y and vX.Z, and get all commits on dev branch that fall between those timestamps
03:24 rewt with --after and --before
03:25 rewt of git log
03:27 Derperperd joined #git
03:27 a_thakur joined #git
03:27 Geo ok
03:28 Geo and just in the name of thoroughness, if not all commits from develop were used (say I had 10 commits, and cherrypicked 8 of them as an example), all 10 of those would show up
03:28 Geo ...in the rev-list outout
03:30 roelmonnens joined #git
03:31 port0 joined #git
03:31 rewt yes
03:32 Geo cool, thank you
03:32 a_thakur joined #git
03:33 a_thakur joined #git
03:37 howdoi joined #git
03:37 menip joined #git
03:40 peeps[lappy] joined #git
03:42 venmx joined #git
03:43 prg318 joined #git
03:45 svm_invictvs joined #git
03:47 hexagoxel joined #git
03:53 Vortex35 joined #git
03:59 port0 joined #git
04:01 andrew710 joined #git
04:01 hahuang65 joined #git
04:09 tristanp joined #git
04:14 nevodka_ joined #git
04:18 hashtagstrashtag joined #git
04:18 Ystah joined #git
04:19 satya4ever_ joined #git
04:19 Ystah guys i'm new to git, I want to roll back the last 3 commits in a remote repo, like completely remove them from the history
04:19 Ystah how should I do that?
04:20 Geo Ystah, take a look at git revert?
04:21 Geo I think thats the 'right' way if you're doing this on a remote repo- it will undo commits as a 'new' commit
04:21 Ystah "as a 'new' commit"
04:21 Ystah can't I just remove the last 3 without any additional commits?
04:21 Geo commit D removes the changes done by A, B, and C
04:22 Geo if its on a remote repo, you generally don't want to do that
04:22 Ystah its only me and a friend that uses it
04:22 Geo if your friend has pulled from it since then, it will hose his repo
04:22 Ystah said friend fucked up and did those 3 useless commits
04:23 miczac joined #git
04:23 Ystah I just want to undo this to unpollute the commit history
04:23 Geo have you pulled it yet?
04:23 Ystah its triggering me
04:23 Ystah no i didn't
04:23 Geo so this is your one and only warning, bad things could happen if you do this
04:23 Geo ie, make it more of a pain to try and fix
04:24 Cabanossi joined #git
04:24 Geo your friend in this case, needs to take a look at git log, find the last 'good' commit, and do a git reset --hard <good SHA>
04:24 Geo then, push to the remote
04:24 Geo it'll tell you this is a bad idea and fail
04:25 Ystah can't I just revert the last 3 commits and ask him to pull?
04:25 Geo I'm only saying that so you have no misconception that its a bad idea
04:25 Geo then you'll follow the directions the error gives you, and do a push -f
04:25 Geo doing that messed up the commit history
04:25 Geo *messes
04:25 Geo so no
04:26 Geo you could, i guess, but when you push it back, your friends repo will be all out of whack
04:26 Geo and if you are asking these types of questions here, it will likely only make a bad situation worse
04:26 Ystah I see, thats sad
04:27 Geo no, its configuration management
04:27 Ystah i thought this was an easy and simple task
04:27 Geo it is
04:27 Geo i just told you how to do it
04:27 Ystah deleting commits from a remote repo you own
04:28 Geo once its on a remote repo, you dont want to mess with it. Because when you do, it breaks things for other people who have already pulled data from it, and it gets out of synch
04:28 Geo With great push permissions comes great responsability
04:32 Ystah I see, ok then
04:32 Ystah guess i'll just get over it
04:33 hexagoxel joined #git
04:35 durham joined #git
04:36 cqi joined #git
04:38 dvaske joined #git
04:39 TbobbyZ joined #git
04:40 _xor joined #git
04:41 ahrs joined #git
04:41 andrew710 joined #git
04:42 Derperperd joined #git
04:45 kasual joined #git
04:47 go1111111 joined #git
04:48 Cabanossi joined #git
04:50 ChiLL_SurF joined #git
04:51 khfeng joined #git
04:51 dreiss joined #git
04:53 MacWinner joined #git
04:54 Geo left #git
04:55 joki joined #git
04:56 raijin joined #git
04:56 MineCoins joined #git
04:59 robotroll joined #git
05:02 eghdk joined #git
05:02 eghdk When I send a file up to my remote git, does it save the mode (chmod) settings of files as well?
05:04 rwp eghdk, Short answer, yes. Long answer, you don't actually send files individually. You push the entire repo. And the repo records file modes along with contents.
05:06 eghdk rwp: Hm. Reading the answers from http://stackoverflow.com/questions/3207728/retaining-file-permissions-with-git make it seem like only executable permissions are saved.
05:07 variable joined #git
05:09 rwp eghdk, I had thought it saved the regular permissions. I will try to test and test it out.
05:10 eghdk okay. I'm kinda a noob, so would appreciate your results! hahah Thanks
05:10 ayogi joined #git
05:15 dweelie joined #git
05:15 blomp joined #git
05:15 roelmonnens joined #git
05:16 rwp eghdk, It does seem that git does not record actual permissions but just uses a "reasaonable default". Which I had not noticed since the default it picks is the same as I would normally use.
05:16 eghdk rwp: Thanks for checking. I appreciate it.
05:18 Soni joined #git
05:21 matoro joined #git
05:24 TbobbyZ joined #git
05:28 variable rwp by "reasonable default" do you mean the umask ?
05:28 variable git also saves +x -x
05:29 variable * only saves
05:29 xaviergmail joined #git
05:31 MineCoins joined #git
05:33 Cabanossi joined #git
05:34 a_thakur joined #git
05:37 sbulage joined #git
05:39 andrew710 joined #git
05:41 Anja joined #git
05:41 MineCoins joined #git
05:50 lordjancso joined #git
05:54 cqi joined #git
06:01 rscata joined #git
06:02 freimatz joined #git
06:03 Cabanossi joined #git
06:07 dec0n joined #git
06:12 Anja joined #git
06:12 drodger joined #git
06:16 [[thufir]] joined #git
06:18 _28_ria joined #git
06:18 MineCoins joined #git
06:18 [[thufir]] I normally use git from the command line with github.  how do I run my own server so that I "push origin master" locally?  something like "git remote add origin <localhost>"?
06:19 hexagoxel joined #git
06:19 Derperperd joined #git
06:20 kadoban [[thufir]]: You can use local directories as "remote" repositories if that's what you mean
06:20 kadoban If you want an actual server, gitolite is nice
06:20 kadoban If you don't need like web features and etc.
06:20 kadoban If you do, there's others ... none of which I have really any experience with.
06:20 nevodka joined #git
06:21 [[thufir]] kadoban: thanks, I'm looking at gitolite.
06:21 Cxcf joined #git
06:21 Cxcf joined #git
06:28 Rotaerk joined #git
06:29 overlord_tm joined #git
06:32 cqi joined #git
06:34 ToxicFrog joined #git
06:37 masuberu joined #git
06:39 math-alpha joined #git
06:39 PtxDK joined #git
06:40 nowhereman joined #git
06:45 dersand joined #git
06:51 justan0theruser joined #git
06:54 hexagoxel joined #git
06:54 nowhereman joined #git
06:55 Raging_Hog joined #git
06:57 ThomasLocke joined #git
06:59 imack joined #git
07:00 PCatinean joined #git
07:03 qt-x joined #git
07:04 dreiss joined #git
07:06 freimatz joined #git
07:11 chele joined #git
07:12 Cxcf joined #git
07:12 Cxcf joined #git
07:15 JeroenT joined #git
07:17 vrach joined #git
07:17 Silenced joined #git
07:18 Silenced joined #git
07:18 masuberu joined #git
07:18 drodger joined #git
07:24 roelmonnens joined #git
07:26 martinhez joined #git
07:26 zeroed joined #git
07:26 Brski joined #git
07:29 hexagoxel joined #git
07:30 seishun joined #git
07:32 rokups joined #git
07:33 JJtG joined #git
07:33 Cabanossi joined #git
07:34 JJtG Hey guys I have a question. Git Status shows me a bunch of modified untracked files. But git clean -n doesnt catch them, and git pull doesnt work without getting rid of the files first. So how would I get rid of the files?
07:34 lucido-cl joined #git
07:34 a_thakur joined #git
07:35 jnavila joined #git
07:36 kadoban JJtG: Are you sure they're untracked? Also, git clean doesn't usually do anything with more flags. I wouldn't be surprised if -n just does nothing without the same extra flags.
07:38 JJtG kadoban, if they show red in git status did that mean they were untracked?
07:38 kadoban JJtG: I don't remember the color coding, if there is one. There are headings that will say what each list means.
07:39 _ikke_ JJtG: A file can not be both modified and untracked
07:39 JJtG kadoban, then I guess its just modified
07:39 t0by joined #git
07:41 JJtG well I guess git reset --hard might do it
07:42 JJtG yeah, that did it
07:43 venmx joined #git
07:44 JJtG kadoban, _ikke_ alright its fixed, ty guys
07:47 qt-x joined #git
07:48 Skyfinn joined #git
07:49 Skyfinn joined #git
07:50 hexagoxel joined #git
07:51 dsdeiz joined #git
07:51 yaewa joined #git
07:54 schleppel joined #git
07:54 elect joined #git
07:56 acetakwas joined #git
07:56 dsdeiz joined #git
07:56 dsdeiz joined #git
07:56 elect joined #git
07:56 Raging_Hog joined #git
07:58 Raging_Hog joined #git
08:00 acetakwas joined #git
08:03 pwnz0r joined #git
08:03 acetakwas joined #git
08:03 raijin joined #git
08:05 Cxcf joined #git
08:06 King_Hual joined #git
08:06 zxd joined #git
08:06 zxd hi
08:06 Cxcf joined #git
08:07 i7c Ohai
08:07 qt-x joined #git
08:07 zxd how does git ls-remote  , knows what is a remote referene
08:07 zxd oops
08:07 zxd it's misleading no?  it can list local repositories as well
08:09 dvaske joined #git
08:09 no_gravity joined #git
08:10 no_gravity Hello! Are "Pull Requests" something GitHub invented or are they part of Git?
08:10 _ikke_ no_gravity: git already had the 'request-pull' command
08:10 _ikke_ which basically formatted an e-mail with the details how and where to pull it from
08:11 _ikke_ no_gravity: github integrated that process in their web interface
08:11 no_gravity Oh.. emails... hmm... we don't use those.
08:11 _ikke_ The e-mail is just a notification to someone
08:11 grawity you can turn anything into a "pull request"
08:11 grawity be it email or Slack or IRC
08:11 jast request-pull isn't used by terribly many people, I think
08:12 _ikke_ Mostly decentralized projects
08:12 grawity but GitHub integrated that with a web UI and some nice review tools
08:12 grawity jast: probably just linux.git and git.git? :D
08:12 no_gravity If you have 2 repos on one server and the owner of one repo wants to send a pull-request to the owner of the other repo, how would you do that?
08:12 jast git.git uses it rarely, if at all
08:12 grawity 1) publish your own repo somewhere other people can clone/pull/etc
08:13 jast no_gravity: if the server doesn't have a UI for notifying the other owner, send an e-mail or something
08:13 grawity 2) push your PR branch there
08:13 no_gravity grawity: It's on the same server.
08:13 grawity 3) tell the other person to pull branch X from url Y
08:13 _ikke_ jast: other side projects use it in git.git
08:13 _ikke_ jast: i18n, gitk
08:13 grawity no_gravity: well then it's fine, step 1 is done
08:13 no_gravity grawity: I see. So the "tell" part has to be handled via some channel outside of git.
08:13 jast yeah, I had a vague memory of seeing that, just wasn't sure
08:13 _ikke_ jast: They tell junio to pull in their commits
08:14 no_gravity The more I look into Git, the more I think it's a low level tool that can be used to build tools on top of it.
08:14 jozwior joined #git
08:14 no_gravity So if you want to set up a workflow, you will inevitable build some tools on top of Git.
08:14 grawity that's exactly what Git has always been
08:14 no_gravity And eventually end up with you own implementation of GitHub.
08:14 grawity see e.g. cogito
08:15 _ikke_ it started out as some very low level tools
08:15 jast nobody uses cogito anymore, though :)
08:15 TheWild joined #git
08:15 grawity in fact I think it started as 'dircache'
08:15 TheWild hello
08:15 TheWild how to make "git stash" default to "git stash list"?
08:15 aidalgol joined #git
08:15 no_gravity Anybody here using git without additional tools to manage a workflow with multiple coders?
08:15 jast yes, the developers of Git
08:15 _ikke_ no_gravity: I do
08:15 i7c Me too.
08:16 _ikke_ Mostly use cli + tig
08:16 _ikke_ (once in a while I also use gitlab for code review though)
08:16 jestjb joined #git
08:16 jast though I guess that's not "anybody here" for the most part
08:16 i7c I use just cli + mergetools.
08:16 no_gravity _ikke_, i7c: How does one coder notify the other coders that he has something ready that they can use?
08:16 jast e-mail, IRC
08:16 no_gravity jast: Well, I asked about git only.
08:16 jast in the development of Git, e-mail pretty much exclusively
08:17 _ikke_ We do use gitlab for that
08:17 no_gravity _ikke_: Oh, so you *do* use tools on top of git.
08:17 i7c I thought you meant like git-wrapping tools like UIs or whatever.
08:17 _ikke_ no_gravity: well, I still do most of the work without it
08:17 jast most of it goes via the Git mailing list, there. the process is people sending patches to the list, getting feedback there, and the maintainer applies/merges them if the feedback is positive enough
08:17 no_gravity Ok, so nobody uses git without tools on top of it.
08:18 jast the Git tool for sending patches as e-mails is git send-email
08:18 i7c Well without even Email it’s getting difficult to communicat at all. :P
08:18 i7c communicate*
08:18 i7c So you need *something* to pass your patches around.
08:18 no_gravity Yeah. Looks like I will have to write some tools then.
08:18 i7c No.
08:18 i7c There are tons out there already. :P
08:18 _ikke_ no_gravity: Like what kind of tools?
08:18 i7c You can write one, but you really don’t have to. :)
08:18 no_gravity _ikke_: Good question. Not sure yet.
08:19 jast I'm not sure what you were expecting... a subcommand in Git that sends neural waves to the target maintainer to have a vision appear to them saying, "hey, no_gravity has new code for you"?
08:19 i7c :D
08:19 i7c That’d be cool.
08:19 jast yeah, but I'm not sure how you'd implement that :)
08:19 no_gravity jast: Well, git could have a function that shows you pull requests.
08:19 _ikke_ hmm, I have this sudden urge to pull from person X
08:19 jast where would those pull requests be stored?
08:19 no_gravity jast: Git would have to handle that.
08:19 i7c In ... a git repo!
08:19 jast yeah, but how
08:20 _ikke_ if only git were decentralized :P
08:20 jast you're basically saying people can only send you pull requests if you are already allowing them to upload that information to them
08:20 i7c _ikke_, so we can try hard to centralize it again? ;)
08:20 jast that's a bit circular in logic if you ask me
08:20 no_gravity jast: Well, look at GitHub. It has that functionality.
08:20 jast yeah, but that's a web interface with account management and all
08:20 no_gravity That's what I mean by "tools oon top".
08:20 _ikke_ jast: You can give people read-only access to a repository where they can pull from
08:21 jast Git itself is decentralized, unlike GitHub
08:21 jast you're basically asking for Git to limit its scope so your use case happens to be easier
08:21 no_gravity jast: Well it would be possible in a decentralized way. Git could run a deamon that accepts pull requests.
08:21 no_gravity For example.
08:21 jast _ikke_: no_gravity is asking for Git to let you know that those commits exist, so you'd have to be able to send the target maintainer information via Git
08:21 _ikke_ right
08:22 jast which does require some amount of write access to that destination
08:22 no_gravity jast: Yup.
08:22 jast no_gravity: so you'd want to accept arbitrary data from arbitrary sources?
08:22 jast I mean, there's a lot of details you have to take care of
08:22 lb1c joined #git
08:22 no_gravity jast: I don't. I just gave GitHub as an example of how a workflow is implemented on top of Git.
08:22 jast suppose someone uploads 4 TB of data to you. now you have to think about limits.
08:22 no_gravity I will implement my own workflow that fits my company.
08:23 grawity well no, pull *requests* probably wouldn't upload the whole thing
08:23 grawity just the request itself
08:23 jast grawity: but GitHub's pull requests include a diff you can view
08:23 grawity but there's still a bit of opportunity for spam
08:23 no_gravity grawity: In a company you would limit pull requests to employees.
08:23 jast no_gravity: sure, but GitHub does the hard work for you of fighting spam and stuff like that. if you have your own public place where you accept messages from everyone, you have to do it all yourself
08:24 grawity no_gravity: in a company you wouldn't need Git's decentralized approach in the first place
08:24 jast anyway, if you want something "like" GitHub, you can just install GitLab or gogs and you're done
08:24 no_gravity jast: I just gave GitHub as an example of tooling on top of Git. I will not implement public pull requests.
08:24 lorddoskias joined #git
08:24 venmx joined #git
08:24 grawity # this error message should only be triggered by cogito usage
08:24 grawity I guess that part of git-bisect.sh should go
08:24 zefferno joined #git
08:25 no_gravity grawity: We don't *need* it but it is useful because most coders already know git.
08:25 RichiH moar graVVity
08:25 lorddoskias joined #git
08:25 RichiH jast: clearly, you have the wrong nick
08:25 RichiH tsk
08:25 lorddoskias hello if i'm cherry-picking some patches which modify a file in dirA/file.c but in my current branc the file structure is dirA/dirB/file.c is there a way to apply the patch without manually resolving the changes?
08:26 jast RichiH: sorry :(
08:27 RichiH lorddoskias: as git tracks diffs and not files, i am not sure there's a nice way
08:27 RichiH you could branch off of whereever you want to cherry-pick from, git mv the files, and then pick from that
08:27 RichiH but that's not exactly nice
08:27 lorddoskias RichiH: and if i get the patch and manually insert the dirA/dirB/ file.c prefix would that work ?
08:28 Masber_080 joined #git
08:29 no_gravity I wonder if the reason for the rise of Git is Git. Or if it's GitHub.
08:30 no_gravity It probably goes hand in hand.
08:30 jast fwiw I started using Git before I first heard about GitHub
08:31 jast it's probably safe to say that GitHub contributed to making it popular, though the most prominent competitors have/had more or less comparable hosting services, too
08:32 no_gravity Well, hosting. But GitHub is pretty much a social IDE.
08:33 selckin what are you smoking can i get some of that
08:33 cbreak joined #git
08:33 no_gravity You can fork a repo, click the editor button, edit a file, send a pull request... all without leaving GitHub.
08:33 jast I can only assume that the fork/pull request stuff didn't map as easily to the other systems as it did to Git, so that may be why the others didn't offer the code sharing part in a similar way. all the other features are pretty similar on the competing platforms, though.
08:33 selckin that far away from anything ide
08:34 jast true
08:34 gunnaro joined #git
08:34 no_gravity You can even publish the result on GitHub so people can use it.
08:34 jast and other open source hosting platforms don't allow you to publish your code
08:34 selckin wordpress is closer to ide then github
08:35 no_gravity jast: Not the code. The working app for people to use.
08:35 jast don't be silly now, wordpress is nothing
08:35 jast no_gravity: we must be speaking about different GitHubs
08:35 no_gravity jast: yourthing.github.io
08:35 jast that works for HTML and its associates, nothing else
08:35 zxd joined #git
08:35 jast which is a pretty narrow use case
08:35 no_gravity jast: the browser plattform is pretty common these days.
08:35 no_gravity most people have one.
08:36 no_gravity And there can use yourthing.github.io
08:36 borkr joined #git
08:36 jast yes, and how will you use GitHub pages to implement, say, an e-mail server?
08:36 no_gravity therefore
08:36 jast or a web-based chat system, to stay in your "platform"
08:36 jast you can't, because anything more complex than a bit of fanciness needs a server-side component
08:36 no_gravity jast: A web-based chat might be possible via webrtc
08:36 jast without any way to do discovery?
08:37 ismithers joined #git
08:37 no_gravity jast: You can use services from around the net for storage/discovery etc.
08:37 jast oh, so suddenly I can't run my code on GitHub alone
08:37 jast if only someone had told me before
08:38 no_gravity jast: The services you use would not be "your code".
08:38 jast you can't offload everything to ready-made services
08:38 zeroed_ joined #git
08:38 jast and if you can it's not necessarily the most cost-effective choice
08:39 no_gravity You always rely on the rest of the world.
08:39 no_gravity Nobody builds their own universe to make a chat app.
08:39 selckin move those goal posts
08:39 jast so basically you're not making your own chat system, you're using someone else's chat system and providing a different UI
08:39 no_gravity Well I gotta go. Cu!
08:40 jast incidentally I mentioned chat systems because at work we do develop web chat stuff, serving many millions of HTTP requests daily. GitHub would suspend us pretty much instantly if we ran it on GitHub pages. :)
08:40 RichiH lorddoskias: that should work
08:40 jast ... a classic "oh hey look at the time"
08:42 lorddoskias RichiH: or another thing which i started doing: 1. git mv the files to match the path in the patch being cherry-picked. 2. Do the cherry-pick 3. revert 1. And finally squash everything into a single commit - the cherry-pick
08:43 RichiH jast: to be fair, what gh did was lower the bar for non-tech people
08:43 RichiH (even though their windows client is a desaster)
08:43 jast sure
08:43 gunnaro_ joined #git
08:44 jast I'm not saying GitHub didn't do anything valuable, that would be silly... but that bordered on worshipping
08:44 _ng joined #git
08:45 selckin git put branching/merging and performance first, and therefor made it nicer to use then all the alternatives, even with all the cli UX complaints
08:46 selckin and so it won, and all related services helped it rise fruther
08:46 selckin it had a big visiable project (kernel) to prove it works and bootstrap it usage
08:46 jast actually I'm convinced now. henceforth I shall use GitHub exclusively, and stop using Git
08:47 TheWild how comes "git status" points out there are files modified but "git diff" shows nothing?
08:47 selckin --cached
08:47 jast 1) 'git diff' without extra args shows unstaged changes only
08:47 anli_ joined #git
08:48 anli_ I would like to gitignore lines of a file :)
08:48 selckin !config
08:48 gitinfo [!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
08:48 anli_ Not all lines however
08:48 jast !ignore_tracked
08:48 gitinfo Git only applies ignore patterns to untracked files. You can't use ignore patterns to ignore changes to files that are already tracked by git. To remove files only from git, but keeping them on disk, use git rm --cached <file>. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.
08:48 Cabanossi joined #git
08:48 jast in short, there's no such thing as ignoring parts of a file
08:48 anli_ I know
08:48 TheWild "git diff --cached" worked. What happened then? These files appear as modified. Are they staged or not by default?
08:49 TheWild "git diff" alone always worked before
08:49 _ikke_ TheWild: You staged them
08:49 jast TheWild: 'git diff' without args shows unstaged changes only. 'git diff --cached' shows staged changes only. 'git diff HEAD' shows both.
08:49 anli_ I am using a diff tool to deselect the lines I will not commit and am using "restore after commit"
08:49 jast anli_: you can do partial commits, I do this pretty much all the time: git commit -p (instead of 'git add'ing the files)
08:50 jast there's "git add -p", too
08:50 anli_ I know, I do partial commits all the time in the way I just described
08:50 jast sure, but your way seems like a lot more work :)
08:50 SCHAAP137 joined #git
08:50 anli_ hm, I use a gui actually
08:50 Balliad joined #git
08:51 anli_ What is a partial commit if not deselecting the lines that I do not want in the commit?
08:51 anli_ I always start with no checked files, then I add those I want
08:51 TheWild ok, thanks selckin, _ikke_, jast. Though I haven't staged them explicitly, I think I've messed up something in the middle.
08:51 anli_ Sorting by modified date
08:51 anli_ So I have the most recently modified in the top of the list
08:52 _ikke_ git has no built-in way to do what you want
08:52 _ikke_ anli_:
08:52 osse TheWild: git status tells you whether changes are staged or not
08:52 alcohol joined #git
08:52 osse it will show up to two lists of modified files
08:53 xaa joined #git
08:53 anli_ Actually, what I really need is maybe to extend the diff tool
08:53 selckin anli_: maybe you just need add -p
08:53 anli_ So it will be remembering which lines that are not yet candidates for committing, I use a hotkey to walk down to the next lines that have been changed, that hotkey could be improved maybe
08:54 TheWild omg. Got it. I added space to some random file in the project and it also appears as modified, but in 'Changes not staged for commit' and in red.
08:54 qt-x joined #git
08:54 TheWild thanks osse.
08:54 nilg joined #git
08:54 TheWild Seems like I staged them some time ago but I just did not remember.
08:56 _ikke_ TheWild: git status is your friend :-)
08:57 Tantamounter joined #git
08:58 nilg joined #git
08:59 Rotaerk joined #git
09:00 tojo joined #git
09:01 qt-x joined #git
09:01 wrouesnel1 joined #git
09:01 Rotaerk joined #git
09:03 elect_ joined #git
09:04 bariscant joined #git
09:05 NeverDie joined #git
09:06 elect__ joined #git
09:07 marcogmonteiro joined #git
09:09 nikitasius joined #git
09:11 Rotaerk_ joined #git
09:11 xaviergmail joined #git
09:11 dvaske joined #git
09:15 polyrob joined #git
09:17 eahmedshendy joined #git
09:18 dvaske_ joined #git
09:19 Snugglebash joined #git
09:23 drodger joined #git
09:24 ayogi joined #git
09:25 bariscant joined #git
09:26 ayogi joined #git
09:26 polyrob joined #git
09:28 masuberu joined #git
09:31 seni joined #git
09:31 sbasso joined #git
09:32 Murii joined #git
09:34 Dumblez0r joined #git
09:35 zxd joined #git
09:35 zxd hi
09:36 choice joined #git
09:36 zxd what happens if I do 'git add'  and there is a .git dir on the dir I am running the command but there are also other .git  and .gitconfig files in subdirs
09:36 Snugglebash joined #git
09:36 choice Hi! When I do changes do to a merge conflict, these changes do not show up in "git log --patch". Why?
09:37 _ikke_ choice: try git log --patch --cc
09:37 acetakwas joined #git
09:38 choice _ikke_: That shows the merge patch. Now all the changes are in the log twice. Hmm...
09:40 choice Now I have 3 commits "+line1", "+line2" and "+line1, +line2". Isn't that a bit strange?
09:40 hexagoxel joined #git
09:40 tens0r joined #git
09:42 czeq joined #git
09:42 math-alpha joined #git
09:43 i7c zxd, stuff happens in the .git where you currently are, and the .git in subdirs are unchanged.
09:44 i7c git goes up the dir tree to find a .git directory, but it does not descend.
09:45 choice Would svn show the same behaviour? That the history shows all merged changes twice?
09:46 cbreak git only shows all merged changes once in history
09:47 Avaq joined #git
09:47 aielima joined #git
09:47 choice cbreak: I see them twice. Let me make a paste...
09:48 choice cbreak: http://pastebin.com/raw/shyGTr5X
09:49 cbreak choice: the merge commit doesn't show original changes
09:49 choice Line1 was added in one repo a, Line2 was added in repo b. Then repo a pulled from repo b.
09:49 cbreak it only shows a summary of the two branches that were merged, IF there were conflicts
09:49 cbreak (you can influence how merge commits are visualized)
09:50 choice cbreak: Yes, the conflict was that in both repos a line was added to f.
09:50 cbreak but this doesn't show a change, it shows a conflict resolution
09:51 choice Is that indicated by the spacing before/after the +?
09:51 cbreak no, by the fact that it's a merge commit
09:51 cbreak (it has more than one parent)
09:51 cbreak choice: try !lol
09:51 gitinfo choice: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
09:52 choice But the merge commit could have also created Line1 out of thin air.
09:52 lorddoskias left #git
09:52 choice Can I roll back the merge commit?
09:53 cbreak yes it could have. That'd be an "evil merge"
09:53 choice Aha! Evil merge. Ok.
09:53 cbreak you can revert merge commits, but that's somewhat problematic
09:54 tojo left #git
09:54 choice How would I do it?
09:55 choice "git revert 069319b553af8fce271bbb73e325d8a6c785c93"?
09:55 cbreak choice: man git-merge
09:55 gitinfo choice: the git-merge manpage is available at http://jk.gs/git-merge.html
09:55 cbreak it should show you the flag for reverting merges
09:55 cbreak you have to specify which parent line you want to keep
09:55 cbreak I think it's -m
09:56 choice Jesus, the revert also ends up as a commit in the history.
09:57 kadoban That's the point of revert, yeah.
09:58 choice Strangely, when I do "git pull ../b/" now, it tells me "Already up-to-date"
09:58 Cxcf joined #git
09:58 choice How can that be? I just reverted the merge.
09:59 choice And the repos are different!
09:59 choice What the hell is going on?
10:01 osse did you do pull in the repo you didn't make the revert in ?
10:01 yqt joined #git
10:01 choice I pulled in a. Then I reverted in a.
10:02 flaviodesousa joined #git
10:02 choice Step 1: git pull ../b/
10:03 choice Step 2: git revert 069319b553af8fce271bbb73e325d8a6c785c93a
10:03 choice Step 3: git pull ../b/
10:03 choice "Already up to date".
10:03 choice Even though the repos are different.
10:03 Cabanossi joined #git
10:03 osse choice: at is there to pull ?
10:04 osse *what
10:04 cbreak choice: there's no new history to pull
10:04 cbreak choice: so of course it says there's nothing to pull
10:04 choice In a, f contains "Line1"
10:04 kexmex joined #git
10:04 choice In b, f contains "Line2"
10:04 cbreak choice: files don't matter
10:04 choice Both are commited.
10:04 osse choice: but the repo in ../b/ hasn't changed
10:04 cbreak git pull only cares about history
10:04 jast you have already merged. Git won't merge the same thing twice.
10:04 choice Well, it *did* pull in Step 1.
10:05 cbreak choice: is there something new to pull?
10:05 choice Then I reverted the pull.
10:05 choice cbreak: Sure, "Line2" in f.
10:05 cbreak choice: no
10:05 cbreak choice: you already pulled that
10:05 jast yeah, but what you think revert does isn't what revert does
10:05 cbreak it's part of history
10:05 jast revert applied the reverse of the changes from the merge, but the merge commit is still there
10:05 cbreak choice: git isn't stupid
10:05 cbreak it knows you already merged it
10:05 cbreak and it won't waste time merging it again, ever
10:05 jast and so the second merge/pull won't see any need to do it again
10:05 jast see !revert_merge
10:05 gitinfo [!faulty_merge] If you have published a merge commit that turned out to be faulty and you'd like to get rid of it, you're probably not going to have a good time. Here's the full story: https://raw.github.com/git/git/master/Documentation/howto/revert-a-faulty-merge.txt
10:06 choice How long will it remember that? Can it bite you in the ass 5 years later? That you pull from a repo and end up with a repo *different* from the one you pulled from?
10:06 jast forever
10:06 cbreak choice: it is part of history
10:06 choice Wow, that's dangerous.
10:06 jast it's not dangerous at all
10:06 cbreak choice: it'd be a quite worthless SCM if it would forget about merges
10:06 choice So after a pull you might end up with a *different* state then the repo you pulled from.
10:06 cbreak choice: yeah, DUH
10:06 cbreak that's the point of pulling
10:06 jast merge looks at the history of the source and the destination, and merges only those commits that exist in the source but not in the destination. that's all.
10:07 cbreak choice: if you wanted the same state you wouldn't pull
10:07 cbreak you'd fetch and reset --hard
10:07 jast you wouldn't want pull to always get rid of your own local changes, would you?
10:07 choice Is there a real "revert"? One the undoes the pull?
10:07 cbreak sure
10:07 cbreak reset --hard
10:07 jast merges can't be undone except by removing history from Git
10:08 cbreak with reset you can kill history as much as you want
10:08 cbreak careful, it will kill history :D
10:08 cbreak (and local changes)
10:08 jast which is straightforward if you're the only one using the repo, and can be rather tricky in other cases
10:08 choice How would you undo the last pull with reset --hard?
10:08 cbreak reset to where you want to be in history
10:08 jast you figure out the commit before the merge, and pass that commit ID to reset --hard
10:08 cbreak choice: as I said above, look at !lol
10:08 gitinfo choice: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
10:09 _ikke_ choice: If the last thing you did is the pull, then you can use: git reset --hard HEAD@{1}
10:09 choice cbreak: I want to be before the pull.
10:09 cbreak choice: so?
10:09 jast boom, merge gone, and everything after it, too
10:09 aielima joined #git
10:09 jast _ikke_: we're two steps further, minimum :)
10:09 cbreak it's probably HEAD^1
10:09 jast remember the revert? it's not HEAD^1
10:09 cbreak (directly after the merge)
10:09 cbreak but with !lol you can see what ever is needed
10:09 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
10:10 manuelschneid3r joined #git
10:10 jast also, given that reverting evidently worked without -m, it was probably a fast-forward merge
10:10 zxd joined #git
10:11 choice "git reset --hard HEAD^1" seems to revert the last commit.
10:11 zxd how does git go about staging files to current .git GIT_DIR if it encounters .git dirs in subdirectories does it ignore these ?
10:11 choice The {1} thing did ... i don't know. Some crazy stuff.
10:11 zxd or if it encounters .gitignore files
10:11 rominronin joined #git
10:12 cbreak choice: it won't revert anything
10:12 cbreak choice: it'll kill history
10:12 choice cbreak: I used revert in human language.
10:12 irqq joined #git
10:12 jast revert can mean many things, !revert
10:12 gitinfo That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?
10:12 Ystah joined #git
10:13 jast zxd: AFAIK git prevents you from adding things in .git subdirs
10:13 cbreak zxd: it'll consider those containing directories "submodules" and stages their hash as commit entry in the tree
10:13 cbreak they are opaque as far as the outer git repo is concerned.
10:13 acetakwas joined #git
10:14 tvw joined #git
10:14 jast only if you add the root path of the subrepo
10:14 choice Thanks for the info guys! I learned something.
10:14 jast if you add individual files directly instead, it'll take them
10:17 ferr1 joined #git
10:17 acetakwas joined #git
10:21 Glooomy joined #git
10:24 fees joined #git
10:25 lordjancso joined #git
10:27 Odaym joined #git
10:27 Odaym hey guys
10:27 Odaym anyone use BFG before? https://rtyley.github.io/bfg-repo-cleaner/
10:28 acetakwas joined #git
10:28 jast not me personally but I've heard of it :)
10:28 Odaym I'm not sure how to use it to remove a certain commit in history
10:29 Odaym where I changed sensitive keys to "------"
10:29 Odaym it still shows in history, it's embarrassing
10:29 jast so you want to get rid of the changes made in that commit?
10:29 Odaym no I want to keep them, I just don't want the log of them to show
10:29 Odaym cause the commit shows the diff, lol
10:29 jast well, if you simply remove the commit, it will show up in the changes for the next commit
10:29 cbreak Odaym: the diff is generated by git
10:29 cbreak git does not store changes
10:29 cbreak it stores snapshots
10:30 Odaym if I remove the Snapshot_18238u1.hprof file?
10:30 jast after all, the difference is still there, only now it can't be pinpointed to that one commit anymore
10:30 jast no idea what file that is. it's not something created by git.
10:30 cbreak Odaym: what?
10:31 Odaym well as I was pushing the repo for the first time to Github, it told me that that file was 79 MB
10:31 Odaym and that I should use git-large-storage something to store it
10:31 Odaym is that where the history is?
10:31 jast anyway, BFG doesn't do what you want. it's for getting rid of a file completely, not for making a commit vanish
10:31 jast no, it's some random file created by something totally unrelated to git
10:32 Odaym you know what? I should just delete .git completely
10:32 Odaym and re-init
10:32 jast okay, if you don't think you're going to need that history...
10:32 Odaym ok what if I do, how do I solve the current issue
10:33 Odaym here's the issue man https://github.com/Odaym/Snipit/commit/4bfc751318bf9af32118cf55b1ea23fe98831649
10:33 Cabanossi joined #git
10:34 jast well, basically what you really want is for the keys to disappear from all the commits before that, right? (*and* vanish that commit)
10:34 Odaym they only appear in the commit which adds the file which they exist in
10:34 Odaym well no, they will exist anywhere in history where I changed
10:35 Odaym yea that's what I want, yes
10:35 jast don't let the list of changes in a commit view deceive you. the content exists in all commits from where you first introduced it to where you got rid of it.
10:35 navidr joined #git
10:35 jast the changes just show you the differences, but each commit contains everything
10:36 Odaym yes
10:36 Odaym that I know
10:36 Odaym so, given that
10:36 jast okay, just making sure :)
10:36 acetakwas joined #git
10:37 jast now, to get rid of the keys you're going to have to start by finding the commit in which you first added the keys
10:37 Odaym I can do that, hold my beer 1 sec
10:38 jast and then you're going to use deep magic (interactive rebase) to retroactively undo that change in that commit. this will alter history. fair warning: if anyone else has the current version of history, they'll have to throw away and re-clone or manually fix their repo, too.
10:39 marcogmonteiro joined #git
10:40 Odaym so listen
10:40 Odaym I went to the very first commit of that file
10:40 Odaym and it still has other keys :P
10:40 Odaym stuff I also don't want to show, this is not going to work like this
10:40 Odaym I am going to remove .git and re-init
10:41 Odaym screw the history, let this be a lesson
10:41 PioneerAxon joined #git
10:42 Ardethian\work joined #git
10:43 dendazen joined #git
10:45 jast that certainly makes things easier :)
10:46 Odaym listen what about deleting the file
10:46 Odaym but also, deleting its history
10:46 Odaym and then re-pushing
10:46 Odaym re-adding it
10:48 jast that would work, too, I guess
10:48 Odaym that's what BFG can do
10:48 jast yep
10:48 masuberu joined #git
10:48 mehola joined #git
10:49 cymen left #git
10:49 sbasso joined #git
10:50 jast in fact BFG will automatically keep the current version of the file in the latest commit, but that's easy to change manually
10:53 kpease joined #git
10:54 acetakwas joined #git
10:55 zxd joined #git
11:00 Decorater joined #git
11:02 Chinggis6 joined #git
11:03 Cabanossi joined #git
11:05 Decorater Guys
11:05 Decorater I am so annoyed when there is no setting to disable to default behavior of git push where it forces an git merge with merge commits that are annoying when you are not aware someone pushed things right before you
11:06 Odaym alright so I ran --delete-files with BFG and it succeeded, but the file is still there and nothing has changed :o
11:06 Decorater and then you did not pull them as it happens so fast that it beats it
11:06 tobiasvl Decorater: forces a git merge? push won't merge anything
11:07 Decorater but it creates merge commits for me that is annoying because I have rules to only have merge commits on pull requests only as it creates useless and clutter commits
11:07 tobiasvl pushing doesn't create merge commits for you
11:08 osse Decorater: it's the pull that git will suggest that creates the merge
11:08 tobiasvl Decorater: if someone pushed before you, you won't be able to push. then you need to either pull, which by default does create a merge commit, or you can rebase
11:08 Decorater actually it does when you push right after someone pushed other changes it goes into vim and does ``Merge branch master in ``
11:09 tobiasvl no. git does not do that.
11:09 tobiasvl are you using any other tools with git?
11:09 Decorater Well I am using git from git bash
11:10 atomicb0mb joined #git
11:10 osse then you must have set something up that makes it pull also
11:10 osse is the EXACT command you run 'git push' ?
11:10 ghoti joined #git
11:10 Decorater yes
11:10 Decorater and then it forces an merge instead of pull
11:11 Ice_Strike joined #git
11:11 tobiasvl pull is a merge
11:11 tobiasvl but it shouldn't force anything
11:11 mizu_no_oto joined #git
11:11 Decorater yeah pull is a merge but it does not create commits
11:11 tobiasvl sure it does, as you can see
11:11 sbasso joined #git
11:12 tobiasvl "git pull" will create a merge commit if needed. "git pull --ff" will refuse to merge if that means making a merge commit. "git pull --rebase" will not merge at all, but rebase.
11:12 tobiasvl sounds like you want the latter
11:12 tobiasvl but git shouldn't make that choice for you
11:13 math-alpha joined #git
11:15 ssarah joined #git
11:18 mozzarella joined #git
11:19 afuentes joined #git
11:20 achadwick joined #git
11:20 Darcidride joined #git
11:22 tvw joined #git
11:23 jameser joined #git
11:23 seni joined #git
11:23 aard_ joined #git
11:24 graff joined #git
11:25 chigang joined #git
11:25 graff I have a patch submitted by a developer. how do i apply it?
11:25 ephimetheus joined #git
11:25 osse graff: what format is it in?
11:25 graff just: git apply ../some.patch
11:25 graff osse: they used git --diff
11:25 osse graff: git apply should do the trick
11:25 cbreak git diff isn't a good tool to make patches at all :(
11:26 osse or ye old regular patch program
11:26 graff cbreak: really? i thought that was the latest hip new thing
11:26 cbreak if you want proper patches, use git format-patch
11:26 grawity well, the diff itself is okay
11:26 cbreak or be sure to tell diff to make proper patches
11:26 grawity it's just that it lacks all the commit metadata
11:26 cbreak (normally it makes human facing patches, with colors and so on)
11:26 osse in my experience the patches produced by a standard 'git diff > bleh' work just fine
11:26 masuberu joined #git
11:26 graff hmm. i see
11:26 grawity cbreak: surely it doesn't write colors when redirected to a file
11:26 graff thanks you guys
11:27 cbreak depends on settings
11:27 graff ok so then i just push after i apply?
11:27 osse no, commit first
11:27 cbreak graff: you have to commit it if you use apply
11:27 cbreak graff: git am will auto-commit
11:27 cbreak it's made for git format-patch output
11:27 ssarah joined #git
11:27 graff i don't want to mess this guy's work up. ok commit then push
11:28 graff cbreak: ok, well i will use that in the future then. when i learn this other way you are talking about
11:28 * cbreak would just push / fetch
11:28 graff for now a ;   git apply ../some.patch ; git commit -m "Other dev's message" ; git push origin master
11:28 graff will work?
11:29 cbreak maybe
11:29 cbreak as I said, normal git diff is not good for making patches
11:29 cbreak if the changes are limited to text-only files it has good chances to work
11:29 dsdeiz joined #git
11:29 dsdeiz joined #git
11:29 osse graff: you have to git add first, or use git apply --index
11:30 drodger joined #git
11:32 Decorater oh it was git apply I thought it was git patch.
11:36 graff that recorded me as the author though :/
11:36 graff heh oh shite
11:36 grawity that's the "lacks all the commit metadata" part, right
11:36 grawity you can amend with --author="Fred Foobar <foo@bar.com>"
11:38 cdown_ joined #git
11:38 graff alright, and then just push again?
11:38 grawity oh you already pushed it?
11:39 graff yeah
11:39 graff it's messed up
11:39 grawity hmm
11:39 MineCoins joined #git
11:39 grawity depending on the number of developers, either try to ninja `git push --force-with-lease` or undo the amend and go on with your life
11:39 atomicb0mb joined #git
11:40 rnsanchez joined #git
11:46 jameser joined #git
11:46 pickfire Is it okay if I work on "Add more builtin patterns for userdiff" for gsoc?
11:46 pickfire I mean as a microproject.
11:48 magentar joined #git
11:48 Cabanossi joined #git
11:48 tmg joined #git
11:50 kpease joined #git
11:51 bremner pickfire: pretty sure you need to ask on the mailing list
11:51 Tobbi joined #git
11:51 bremner not many git devs here
11:51 JeroenT_ joined #git
11:52 netj joined #git
11:52 magentar any idea why 'git pull' may only pick up changes 5 minutes late (it says already up to date even though the changes are visible in cgit for example), even after restarting httpd from which it is served?
11:53 _ikke_ pickfire: to add to that: you don't have to ask for permission to work on a micro project, but it helps to find something someone else hasn't done yet
11:54 jast magentar: using git-http-backend, or plain webdav?
11:55 jameser joined #git
11:55 magentar jast: http backend
11:55 jast that's weird, then
11:56 magentar yeah, i have no idea where could be the problem
11:56 jast with webdav it's usually because someone is running git update-server-info in a cronjob (and it does need to be run so clients can see the changes)
11:56 magentar perhaps i should update the git on the machine where i pull it's only 1.8
11:56 jast but if it uses the http-backend CGI the changes should show up instantly
11:57 magentar git pull even ask me for username password so it connects to the server at least
11:59 jast can you share the httpd access logfiles from an unsuccessful ("up-to-date") pull?
12:00 osse should probably do git update-server-info in a post-receieve hook, no ?
12:00 magentar one sec
12:00 synthroid joined #git
12:00 jast if you have git-http-backend, you don't need update-server-info. if you don't, nothing executes the hooks.
12:01 synthroi_ joined #git
12:02 magentar http://pastebin.com/f1GLsyjA i hope that's it
12:02 magentar that should be the pull
12:03 Snugglebash joined #git
12:04 magentar ah okay, i use cgit as remote, i'm stupid ;)
12:04 Sonderblade joined #git
12:05 magentar (the url should be /git instead of /cgit)
12:06 parted joined #git
12:07 ronator joined #git
12:07 jast so that means the problem is solved? :)
12:08 magentar absolutely ;)
12:08 Silenced joined #git
12:08 Silenced joined #git
12:09 parted Hello. What am I doing wrong, if my git-user has a git-shell-commands directory, including the no-interactive-login file and upon SSH login, I am presented with a git-shell? git-user owns the home dir, git-shell-commands dir and no-login file. The shell of git-user is /usr/bin/git-shell. Thanks in advance :)
12:10 parted The no-login file has execution permissions and contains the default sh printf’s
12:10 jameser joined #git
12:13 gingitsune joined #git
12:13 dviola joined #git
12:13 gingitsune I have pushed some sensetive info to github repo. Is there a way to get it out of the repo history?
12:13 Raging_Hog joined #git
12:14 grawity they happen to have a guide for that https://help.github.com/articles/removing-sensitive-data-from-a-repository/
12:16 gingitsune grawity: thanks
12:17 ankit01ojha joined #git
12:17 scottjl joined #git
12:21 muhannad__ joined #git
12:24 parted Is there any way to troubleshoot, why my git ssh user has interactive access, even if the no-interactive-login file exists?
12:24 jameser joined #git
12:25 _Cyclone_ joined #git
12:26 venmx joined #git
12:29 Vampire0 parted, this is more a question for #openssh or similar
12:29 Derperperd joined #git
12:29 parted Vampire0: okay, thanks.. will try there
12:31 perlpilot joined #git
12:32 clmsy joined #git
12:35 jonathan_x joined #git
12:35 muhannad___ joined #git
12:37 jonathan_x left #git
12:37 zomaar joined #git
12:38 zomaar I've done a remote commit and a local commit of the same file. How do I ensure a fetch where I rebase the local changes on top of the remote changes?
12:38 xaviergmail joined #git
12:39 zomaar I don't want any merging, I want the remote changes first and then just rebase on top of that.
12:39 martinhez joined #git
12:39 Vampire0 zomaar, well, just do it, where is the problem?
12:40 YuGiOhJCJ joined #git
12:40 zomaar how do I do it?
12:40 Vampire0 zomaar, either do `git fetch` and then `git rebase ...`, or you can also do `git pull --rebase ...` or set configuration options to make pull use rebase instead of the default merge
12:41 zomaar hmm git pull --rebase (without parameters) resorted to a 3-way-merge.
12:42 zomaar do I need to specify origin/master as something to rebase --onto?
12:43 Vampire0 zomaar, it should do a 3-way-merge of the changes, but not a merge in Git terms
12:43 zomaar it gave me a patch file in some .git/... directory
12:43 zomaar and asked me to resolve the conflict, but there shouldn't be any.
12:43 Vampire0 zomaar, !repro
12:43 gitinfo zomaar: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
12:44 Vampire0 zomaar, what makes you think there shouldn't be a conflict? You edited the file in both branches, so there is a conflict potentially
12:45 zomaar because replacing the local file first, and then applying the commit on top of that, (rebasing) should never produce a conflict.
12:45 _ikke_ it would, otherwise it would throw away changes
12:46 zomaar in other words, if I clone the remote repo elsewhere, and then transplant my local commit on top of that, it should work right.
12:46 Vampire0 zomaar, of course it does, the changes you did were made against a different base file, so there is a conflict that you need to resolve
12:46 masuberu joined #git
12:46 _ikke_ zomaar: git doesn't 'zomaar' throw away changes ;-)
12:46 zomaar okay, but if I now copy my file to a backup location, then pull, I will have the remote version.
12:47 zomaar if I now copy back my file on top of that and commit, I will have what I want
12:47 zomaar I was just trying to do it with the file in place
12:47 _ikke_ zomaar: right, but that's not default behaviour of git rebase
12:47 Vampire0 zomaar, but you will loose the changes from the remote, wont you?
12:48 Vampire0 zomaar, if that is what you want, you can easily tell Git to use your version and discard what is coming in from remote
12:48 _ikke_ zomaar: look at man git merge merge strategy 'ours' (it's also valid for rebase)
12:48 gitinfo zomaar: the git-merge manpage is available at http://jk.gs/git-merge.html
12:48 Cabanossi joined #git
12:48 zomaar no because I would pull them. I would throw away local "changes" first (but keep the commit)
12:48 zomaar at least that is what I was trying to do...
12:48 zomaar I could try to rebase on top of origin/master :p
12:48 ShapeShifter499 joined #git
12:49 Vampire0 zomaar, sure, but if you copy away the file as it is before the rebase, and later copy the file over again, you will force your local version without the changes from remote
12:49 dfas joined #git
12:49 inflames joined #git
12:50 Vampire0 zomaar, and if origin/master is the upstream branch of your local branch, then `git rebase` or `git pull --rebase` will exactly do that, rebase on top of origin/master
12:51 zomaar euhm that's true. But the remote would still be in the commit log :p
12:51 zomaar I guess I'm trying to do something unusual :p
12:51 _ikke_ yes
12:51 zomaar there is actually no local commit of that file yet prior to the remote commit :p.
12:52 _ikke_ what is it that you want to achieve (separate from how you want to achieve it)?
12:52 Vampire0 zomaar, it is not uncommon to say "take my version of the file regardless of the changes coming in". That is way Git supports this if you tell it to
12:52 zomaar I committed the file first on GitHub instead of through the local repo
12:53 zomaar then I made local changes
12:53 zomaar so It's okay to throw away the remote
12:53 zomaar I just want to have the correct commit log
12:54 zomaar I guess I should just have stashed the changes prior to the pull and then checkout -p stash
12:54 zomaar instead of committing first
12:54 jeffreylevesque joined #git
12:55 _ikke_ zomaar: try git pull --rebase -Xtheirs
12:55 zomaar ok
12:56 Vampire0 zomaar, like _ikke_ suggested, but it will take all files from your branch iirc
12:56 Vampire0 zomaar, you can also after the conflict is recognized do
12:57 Vampire0 zomaar, `git checkout --ours -- path/to/your/file` or `git checkout --theirs -- path/to/your/file`
12:57 Vampire0 zomaar, and the add the file and continue the rebase with `git rebase --continue` to only solve that conflict by using either version of the file completely
12:57 fahadash joined #git
12:58 eroux joined #git
12:58 pickfire bremner, _ikke_: Thanks a lot for the suggestion, I will ask on the mailing lists.
12:59 JeroenT joined #git
12:59 zomaar ok so in the mids of the rebase, I can checkout my own commit using --ours, then add the file and --continue, and then it will use my local copy as the final verison.
13:00 zomaar that is perfect
13:00 Vampire0 zomaar, exactly. You can and should validate the result by looking at it of course. :-)
13:00 netj joined #git
13:01 Vampire0 zomaar, you can also use a mergetool like p4merge (the best available free one in my opinion) with `git mergetool` to look at the local and remote changes and decide what to do
13:01 59NAAN4TO joined #git
13:01 _ikke_ Vampire0: is p4merge cross-platform?
13:02 zomaar right, well that is exactly what I needed, thank you. I already used the first option but I could redo it to learn etc.
13:02 zomaar but that's great, that's exactly what I needed :). Thanks.
13:02 Derperperd joined #git
13:03 bleb joined #git
13:04 Vampire0 _ikke_, yes, it is
13:04 pbrewczynski joined #git
13:05 zefferno joined #git
13:05 zomaar actually I needed to do --theirs to get the local commit :-/.
13:05 _ikke_ yeah, correct
13:06 _ikke_ zomaar: your commit is applied on top of what has been fetched
13:06 Vampire0 zomaar, this is because you actually first (yours) have the one you rebase onto and then you apply the other version (theirs) you rebase onto it
13:06 _ikke_ so ours is the 'remote' commit, and theirs is the local commit
13:06 Vampire0 zomaar, if you would do a merge it would be the other way around
13:06 bleb I made a git repo within another git repo which I didn't know existed, and now my git repo is messing things up for the outer one. is there an easy way to (a) make the outer git repo behave as if there is no inner git repo or (b) merge my commits to the inner repo with the outer repo?
13:07 ExoUNX joined #git
13:07 zomaar right
13:09 Vampire0 _ikke_, I just verified, there are versions for Linux, Windows and OS X. I use the first two ones since years. I don't use OS X, thus I was not fully sure, but it is supported
13:09 zomaar yeah I tried to avoid merges at all costs since it is so hard to do from the command line or I just don't know how to do it yet :p.
13:10 Vampire0 zomaar, then use p4merge, it is easy as pie to do it with a proper GUI tool for merging and as I said, P4Merge is the best free on I know
13:10 Vampire0 s/on/one/
13:10 JeroenT joined #git
13:12 joeco joined #git
13:12 zomaar yeah I was just working on a remote ssh session instead of a local filesystem :p.
13:13 JeroenT_ joined #git
13:13 muhannad___ joined #git
13:15 Vampire0 zomaar, so what? For this X-forwarding exists. I regularly run GUI tools on servers via SSH connection that open up on my local desktop, from both, Linux and Windows. You just have to have a running X Server on Windows like XMing or the one included in Cygwin
13:15 zomaar ooh okay
13:16 imMute Vampire0: there's an X server included with Cygwin?  how does it compare to Xming?
13:16 ij joined #git
13:17 ij Is it condemnable to repush the latest of the version tags while you're developing?
13:17 Vampire0 imMute, I don't know, I use the Cygwin one since years and have no problem running GUI tools from Cygwin or running remote GUI commands via SSH
13:17 Vampire0 imMute, you have to install the package that contains it of course, I think it is Cygwin-X or something like that
13:18 Vampire0 imMute, and I added it to my autostart configuration
13:18 Vampire0 ij, dunno if someone else understood what you mean, I didn't though :-(
13:19 _Vi joined #git
13:19 ij Would anyone be as kind as to either confirm or deny the supposed incomprehensibility of my question?
13:20 grawity ij: once a version is released, it should be immutable
13:20 BlackMaria joined #git
13:20 grawity ij: tags are not suitable for things like 'latest'
13:20 grawity use regular branches or something else
13:20 ij Ah, so while I'm developing, I should point to master in other places — gotcha.
13:21 Vampire0 ij, actually you can do what you want. If you want to maintain a tag "deploy-this" that for anyone is clear is mutable and will change over time, this is probably fine also
13:21 leeN joined #git
13:21 Vampire0 ij, but typically for changing refs it is better to use branches
13:22 roelmonn_ joined #git
13:23 aielima joined #git
13:24 muhannad__ joined #git
13:25 al-damiri joined #git
13:25 mazert_ joined #git
13:26 JanC_ joined #git
13:29 Olipro joined #git
13:30 HardlySeen joined #git
13:31 seni joined #git
13:33 sarri joined #git
13:33 sarri joined #git
13:35 bleb git status says modified: src/mydir (modified content)
13:35 _ikke_ bleb: that's a submodule
13:35 bleb then if I do git add ., git status says the same thing
13:36 _ikke_ bleb: cd to that directory, and run git status again
13:36 bleb are submodules create automatically if I have a git repo within another repo?
13:36 lmatteis joined #git
13:36 _ikke_ Not complete submodules
13:36 bleb running git status in mydir says nothing to commit, working directory clean
13:37 sunri5e joined #git
13:37 bleb I'm happy to have the inner git repo be separate from the outer one for now, until I learn about submodules
13:38 sunri5e joined #git
13:38 bleb would adding the other .git to .gitignore help this?
13:38 roelmonnens joined #git
13:38 bgerber joined #git
13:40 GodGinrai joined #git
13:41 pinkkis joined #git
13:42 jozwior joined #git
13:43 Dougie187 joined #git
13:46 Vampire0 bleb, maybe !supre will help you
13:46 Vampire0 bleb, mistyped, I meant !subrep
13:46 gitinfo bleb: [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
13:48 Cabanossi joined #git
13:48 cdown joined #git
13:52 bleb thanks
13:53 bleb seems strange that git seeks out other .git directories and interprets them as submodules without explicitly setting them up as such
13:53 bleb but submodules looks like what I need so I'll look into it
13:53 GodGinrai joined #git
13:55 davidfetter_ge joined #git
13:55 manuelschneid3r joined #git
13:55 GodGinrai Hey guys, is there a way to get interactive rebase to specially mark commits that are specific to that branch? (not in the branch you are rebasing from *or* to)
13:56 Odaym why is it that every command with BFG leaves the repo in a state where it's behind origin?
13:56 Odaym and I have to force
13:56 manuelschneid3r joined #git
13:56 Gsham joined #git
13:56 Odaym maybe because they were all followed by git reflog expire
13:57 davidfetter_ge is that the bfg9000?
13:57 Odaym lol, from Doom?
13:57 regedit joined #git
13:57 davidfetter_ge i see people still follow the classics.
13:57 Odaym it's Doom
13:58 Vampire0 GodGinrai, I guess not, commits do not belong to a branch. A branch is just a pointer to some commit and this commit has some history. But that branch can at any point in time made pointing at a different commit
13:59 Vampire0 Odaym, well BFG modifies the history, so you always will suffer from !rewrite if you use it
13:59 gitinfo Odaym: Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
13:59 davidfetter_ge anyhow, since i didn't actually play Doom back when, I should probably get to my actual issue, which is that recently, things, like git checkout on a local branch have slowed to a crawl
13:59 davidfetter_ge how do i diagnose?
14:00 Odaym the issue was with sensitive data
14:00 Es0teric joined #git
14:00 Odaym I had to rewrite as I need the history
14:00 sespiros joined #git
14:00 Odaym it went swimmingly though
14:00 Vampire0 Odaym, then you probably need it, yes
14:00 GodGinrai Vampire0: but git can determine what are specific to the branch when rebasing to a newer version of the same branch, so why couldn't it figure it out before attempting to rebase to a different branch?
14:00 Vampire0 Odaym, but still it is a rewrite of history, so you need to force push and anyone has to rebase their work
14:01 Odaym because their branch is now out of sync with mine? it's not just out of sync actually, it's corrupted
14:01 Odaym the history has been changed
14:02 Vampire0 Odaym, read !rewrite and / or the section of "how to recover from an upstream rebase" in i think man git rebase
14:02 gitinfo Odaym: Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
14:02 gitinfo Odaym: the git-rebase manpage is available at http://jk.gs/git-rebase.html
14:02 davidfetter_ge for example, https://paste.fedoraproject.org/paste/a90J~-k-4mZVTyBKGADhtF5M1UNdIGYhyRLivL9gydE=
14:02 davidfetter_ge over 3 seconds for "git stash list"
14:03 AaronMT joined #git
14:03 Derperperd joined #git
14:04 davidfetter_ge is there some routine maintenance operation I haven't done?
14:04 moei joined #git
14:04 Vampire0 davidfetter_ge, routine maintenance is done automatically every so often if needed
14:05 davidfetter_ge Vampire0, ok. what would help diagnose and (ideally) fix this?
14:05 Vampire0 davidfetter_ge, I don't know, maybe strace will give you additional information?
14:06 davidfetter_ge no strace on osx atm.
14:06 zomaar joined #git
14:07 learning joined #git
14:07 zomaar Vampire0: I have managed to get p4merge running using XMing and PuTTY, thanks.
14:07 kaushal joined #git
14:08 Vampire0 zomaar, you're welcome :-)
14:08 zomaar Vampire0: I have yet to test it but it starts fine and I have it selected with --global config.mergetool or whatever :p.
14:08 govg joined #git
14:08 zomaar thanks :).
14:08 Vampire0 davidfetter_ge, isn't there some os x supplement
14:08 dglambert joined #git
14:08 davidfetter_ge dunno. i get mine from homebrew
14:08 dglambert yo
14:09 Vampire0 davidfetter_ge, dtruss maybe? https://opensourcehacker.com/2011/12/02/osx-strace-equivalent-dtruss-seeing-inside-applications-what-they-do-and-why-they-hang/
14:09 * davidfetter_ge bookmarks this for future reference
14:09 Vampire0 davidfetter_ge, or dtrace: https://8thlight.com/blog/colin-jones/2015/11/06/dtrace-even-better-than-strace-for-osx.html
14:10 dglambert Hey, what is the preferred method of diffing the mainline branch against your feature branch before merging?
14:10 Vampire0 davidfetter_ge, just Google results, I didn't use either
14:10 * davidfetter_ge thinking this is less a git binary problem and more a PEBKAC issue
14:11 Vampire0 dglambert, depends on what you want to see. difference between mainline and feature? differences on mainline that are not yet in feature? differences on feature that are not yet in mainline?
14:11 sunrunner20 joined #git
14:11 Vampire0 davidfetter_ge, possible :-)
14:11 Sound joined #git
14:12 hvxgr joined #git
14:12 dglambert Vampire0: here is the scenario, I have already merged mainline into feature, so I want to see the differences on feature that are not yet in mainline
14:12 dglambert However, maybe knowing how to do the inverse would be helpful in the future
14:14 Vampire0 dglambert, `git diff mainline..feature` should show you the diff of commits that are in feature but not in mainline iirc
14:14 oaao joined #git
14:15 dglambert that will use my default diffing tool then correct
14:16 dglambert Can I diff against a branch in a remote repo?
14:16 dglambert origin:mainline
14:16 dglambert ?
14:16 dsteplight__ joined #git
14:17 Ryanar joined #git
14:17 osse dglambert: if you have a remote branch corresponding to it, sure
14:18 osse dglambert: git diff origin/mainline HEAD
14:18 osse or something like that
14:18 theoceaniscool joined #git
14:18 dglambert not :
14:19 marcogmonteiro joined #git
14:19 zomaar no it's a /
14:21 dglambert it openned in bash
14:21 dglambert is that expected, or is it have to do with having a default diff tool setup
14:21 zomaar it just outputs the diff in "unified diff" format.
14:21 guest in post-checkout git hook, if I use "find ." command, will it run from the root of the repo, or from whichever directory I was in at the time?
14:22 King_Hual joined #git
14:22 zomaar dglambert: you get the same output you can feed to the "patch -u" command
14:22 eroux joined #git
14:22 guest nvm answered my own question by putting "pwd" in there - it always runs from the root of the repo.
14:23 zomaar dglambert: actually patch automatically determines the type... but it's the same as -u ...
14:23 davidfetter_ge so i'm on a branch, and it took about 5-6 seconds for "git rebase -i HEAD~2" to give me an editor window. how do i diagnose this?
14:23 Vampire0 dglambert, `git diff`will show you a textual diff. `git difftool` will use your configured diff tool.
14:23 dglambert what man page should I look at for patch and -u
14:23 osse davidfetter_ge: are you on windows ?
14:24 ghoti joined #git
14:24 davidfetter_ge osx git 2.12.0
14:24 dglambert Vampire0: :)
14:24 drodger joined #git
14:24 zomaar dglambert: it's just the tool that people use for applying patches.
14:24 zomaar dglambert: but it would be "man patch"
14:25 zomaar dglambert: unified format is just the standard format.
14:25 dglambert oh -u aka unified
14:25 zomaar yeah
14:25 dglambert what the bot command to pull up the man page
14:25 Vampire0 zomaar, not all people use it. You can also use `git apply` ;-)
14:25 Vampire0 dglambert, man git foo
14:25 gitinfo dglambert: the git manpage is available at http://jk.gs/git.html
14:25 davidfetter_ge osse, basically, doing just about anything takes ~5 seconds
14:25 zomaar okay ;-).
14:25 dglambert TY
14:26 dglambert man git patch
14:26 gitinfo the git manpage is available at http://jk.gs/git.html
14:26 Vampire0 dglambert, patch is not a git subcommand, but a totally different binary.
14:26 Vampire0 dglambert, what you mean is man git apply
14:26 gitinfo dglambert: the git-apply manpage is available at http://jk.gs/git-apply.html
14:26 dglambert oh
14:26 CEnnis91 joined #git
14:26 dglambert is patch part of bash?
14:26 davimore joined #git
14:26 Vampire0 dglambert, no, it is a totally different tool
14:26 nurey joined #git
14:26 dglambert ohh
14:27 sword joined #git
14:27 nd_ joined #git
14:27 zomaar dglambert: part of the "patch" package on Linux ;-).
14:27 davidfetter_ge osse, any ideas how i go about diagnosing this?
14:28 dglambert thanks zomaar
14:28 dglambert thanks Vampire0
14:28 dglambert thanks osse
14:28 osse davidfetter_ge: add GIT_TRACE=1 in front of the command and see if you can identify a single step that takes much longer than all the rest
14:29 Vampire0 osse, he knows about GIT_TRACE already, he posted a transcript with it, he searches on how to dig deeper
14:29 davidfetter_ge https://paste.fedoraproject.org/paste/a90J~-k-4mZVTyBKGADhtF5M1UNdIGYhyRLivL9gydE=
14:29 davidfetter_ge i don't see any particular thing that's slowest
14:29 notebox joined #git
14:30 Ryanar joined #git
14:30 osse Woops
14:31 osse how long does status take?
14:31 zefferno joined #git
14:32 davidfetter_ge ~0.03 seconds
14:33 aielima joined #git
14:34 LeBlaaanc joined #git
14:35 tvw joined #git
14:36 sunrunner20 joined #git
14:36 dglambert Hey Vampire0, you asked if I wanted to see the changes going into mainline or the changes going into feature, how does that make a difference
14:36 masuberu joined #git
14:37 Vampire0 dglambert, I don't understand the question
14:37 dglambert Well, I guess I am only seeing changes from my feature branch right now
14:37 dglambert im just curious if inverting the to commit args in difftool would make a difference
14:37 cdown joined #git
14:37 yehowyada joined #git
14:38 zomaar dglambert: inverting the args would just change the + to - and the - to +
14:38 Vampire0 dglambert, yes, it should if you do feature..mainline it should show the changes that are in mainline but not yet in feature
14:38 zomaar dglambert: which they would call a "reverse" patch.
14:38 davidfetter_ge osse, any other ideas about stuff i should poke at?  one thing that's changed recently is that i now have another upstream, which i use for reviewing that person's PRs
14:38 osse davidfetter_ge: I was basically out of ideas once windows was not the issue
14:39 osse how big is the repo?
14:39 osse does git gc help?
14:39 dglambert I did feature mainline, and I am seeing my changes
14:39 * davidfetter_ge tries git gc
14:39 dglambert I haven't seen any changes yet, that I don't recognize, so I am thinking maybe it is just inverted
14:39 gugah joined #git
14:40 ThomasLocke joined #git
14:40 davidfetter_ge du -hs .
14:40 davidfetter_ge 7.8M.
14:40 davidfetter_ge is there some othere size measure i should look at?
14:40 gugah joined #git
14:40 dglambert should, git difftool feature mainline, even show file that only I changed in feature?
14:40 jast at that size I don't think anything else is required
14:41 zomaar dglambert:    you're getting a reverse patch this way.
14:41 dglambert ok, I'll try inverting it and see the difference
14:41 Vampire0 dglambert, ah, I was false. It is not `git diff mainline..feature`, that would diff the actual heads against each other
14:41 Vampire0 dglambert, it is `git diff mainline...feature` instead
14:41 dglambert side question, is ctrl+c the correct way to quit the current command?
14:41 dglambert in bash that is
14:42 Vampire0 dglambert, this should only show the changes in feature against the merge base
14:42 dglambert I do it but then I cant type afterwards
14:42 Vampire0 dglambert, yes, ctrl+c is correct for aborting and I can type fine after that. Maybe some local issue on your box?
14:42 jast depends on what you're interrupting
14:42 zomaar Vampire0: there is a difference between .. and ... ?
14:42 jast zomaar: yes
14:42 tribly joined #git
14:42 Vampire0 zomaar, or course, otherwise it would be non-sense to have both
14:43 zomaar Vampire0: sure, I just didn't know about it yet :p.
14:43 jast for diff specifically, a...b shows differences between {merge base of a and b} and {b}
14:43 jast for everything else (rev lists) it has a different meaning :}
14:43 zomaar so the merge base is the commit where both paths diverged.
14:44 jast man gitrevisions for all the details
14:44 gitinfo the gitrevisions manpage is available at http://jk.gs/gitrevisions.html
14:44 jast yes, or the most recent merge of the two
14:44 dglambert Yeah, I am still seeing my changes
14:44 zomaar that seems ordinarily more useful than a diff against a diverged branch...
14:45 zomaar thanks
14:45 dglambert Maybe I will wait till I find a file that has changed on the mainline
14:45 Vampire0 jast, the gitrevisions man page does not have the info about the meaning in git diff, has it? I think this is only mentioned on the git-diff manpage
14:45 osse davidfetter_ge: try LANG=C git stash list
14:45 LeBlaaanc joined #git
14:46 osse davidfetter_ge: some of the things that take an unreasonable amount of time compared to on my computer is git-sh-i18n--envsubst
14:46 jast Vampire0: yeah, but since I already mentioned that part, I thought that page should be sufficient for the rest at least
14:46 davidfetter_ge real0m4.821s
14:46 davidfetter_ge user0m1.493s
14:46 davidfetter_ge sys0m3.260s
14:46 Vampire0 jast, sure, just wanted to make sure. Could have changed since I last searched :-)
14:46 jast I know just as much as you do :)
14:46 sunri5e joined #git
14:47 davidfetter_ge osse, https://paste.fedoraproject.org/paste/CDLKgORTeiQH9hl7PTZLNF5M1UNdIGYhyRLivL9gydE=
14:48 svm_invictvs joined #git
14:48 Cabanossi joined #git
14:48 Vampire0 jast, I don't think so :-D
14:48 raspado joined #git
14:49 osse davidfetter_ge: time git 'config' '--get-colorbool' 'color.interactive'
14:49 jast Vampire0: correction: I now as least as much as you do :P
14:49 jast *at least, dangit
14:49 davidfetter_ge osse, real0m0.005s
14:49 jast *know
14:49 jast curses
14:49 osse davidfetter_ge: ok, so that's short and sweet. yet when you run git stash it takes 0.3
14:50 osse i think i am out of ideas
14:50 davidfetter_ge 3.3 :P
14:50 ngqvhx joined #git
14:50 synthroid joined #git
14:50 osse what?
14:51 davidfetter_ge git stash list takes ~3.3 seconds
14:51 LyndsySimon joined #git
14:51 notebox joined #git
14:51 davidfetter_ge now 4.7
14:52 mehola joined #git
14:54 Derperperd joined #git
14:55 Snugglebash joined #git
14:56 skered I want to view a file from a tree from an old commit that has been deleted.  Normally I would just 'git checkout commit path'  However, this stages the file, is there a different command to just get the file?  Also is it possible to get that file to a different filename?
14:56 sbasso joined #git
14:57 jast skered: git show commit:/path/to/file > mynewname
14:57 jast actually remove that first '/', that was a typo
14:57 masuberu joined #git
14:57 jast it works with the /, too, I think, but it's not necessary
14:58 jast you can use ./ if you want something relative to your current dir
15:00 skered Thanks.
15:00 gargola joined #git
15:03 AstralStorm joined #git
15:04 AstralStorm hello, is there something easier than git-subtree that does the same thing? esp. shares remotes and remembers tracking branches
15:06 dcpc007 joined #git
15:06 graff well i finally got it osse. thanks for the help
15:07 graff i just used a really hacky method. renamed myself and pushed
15:07 graff seems ok
15:07 graff left #git
15:10 oleo joined #git
15:11 King_Hual joined #git
15:11 BSaboia joined #git
15:12 noobie_ joined #git
15:14 JeroenT joined #git
15:15 crose joined #git
15:16 JanC_ joined #git
15:16 bleb left #git
15:17 Guest55 joined #git
15:18 Derperperd joined #git
15:18 cdg joined #git
15:19 JeroenT joined #git
15:20 sbasso joined #git
15:21 warthog9 joined #git
15:21 kegster OK so I have a branch called "testing" that when I checkout, it says detached head state. When I'm in said state, how can i remove whatever is causing this and just revert back to the branch state?
15:22 Swiftb0y joined #git
15:23 Groovin joined #git
15:25 math-alpha joined #git
15:27 Vampire0 kegster, !repro and !zomb
15:27 gitinfo kegster: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
15:27 gitinfo kegster: [!zombie] Brainzz! More brainzzzz!  For information about detached heads, see !detached
15:27 venmx joined #git
15:27 BSAlb joined #git
15:28 Snugglebash joined #git
15:28 dreiss joined #git
15:29 brent__ joined #git
15:29 brent__ joined #git
15:30 Gustavo6046 joined #git
15:30 noobie_ left #git
15:30 noobie_ joined #git
15:33 noobie_ left #git
15:33 Cabanossi joined #git
15:33 madewokherd joined #git
15:36 WeiJunLi joined #git
15:36 WeiJunLi simplest way to git a branch?
15:37 Vampire0 WeiJunLi, what is "to git"?
15:37 star_prone joined #git
15:37 WeiJunLi i did git branch branch-name
15:38 WeiJunLi but no files were downloaded
15:40 leeN joined #git
15:40 Vampire0 WeiJunLi, with `git branch branch-name` you create a new branch called branch-name pointing at the commit you currently have checked out
15:40 tvw joined #git
15:41 WeiJunLi Vampire0: https://android.googlesource.com/kernel/tegra/
15:41 WeiJunLi i did the clone
15:41 Vampire0 WeiJunLi, with `git checkout branch-name` you switch to a branch. With `git checkout -b branch-name` you do both together, create a new branch called branch-name and checking it out.
15:41 WeiJunLi and then git branch android-tegra-3.10
15:41 Vampire0 WeiJunLi, just as I said, you created a new branch
15:41 WeiJunLi alright
15:42 Vampire0 WeiJunLi, you probably are after !where
15:42 gitinfo WeiJunLi: [!wherearemybranches] Git clones all branches, but initially makes only the main branch (usually 'master') available as a normal local branch. To work on another branch, simply 'git checkout <branch>' (or, in versions older than 1.6.6, git checkout -b <branch> origin/<branch>). This works even if the branch doesn't show up in the normal list of branches. For technical details, see !rtb
15:42 WeiJunLi understood
15:42 Naan joined #git
15:42 LeBlaaanc joined #git
15:42 col joined #git
15:43 WeiJunLi Vampire0: i did git checkout -b branch-name
15:43 WeiJunLi Already on 'android-tegra-dragon-3.18-marshmallow-dr-dragon'
15:43 WeiJunLi but no files there
15:44 WeiJunLi probably i need to delete the branch i created?
15:44 Guest55 joined #git
15:45 dviola joined #git
15:45 Vampire0 WeiJunLi, what means "no files there" mean? !repro
15:45 gitinfo WeiJunLi: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
15:45 WeiJunLi Vampire0: there are no folders inside the dir
15:46 pickfire WeiJunLi: Are you from china?
15:47 WeiJunLi no
15:47 Vampire0 WeiJunLi, still !repro, guessing is hard without enough information
15:47 gitinfo WeiJunLi: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
15:48 pickfire WeiJunLi: Did you do ls <branch>?
15:48 xaviergmail joined #git
15:49 sbulage joined #git
15:54 d0nn1e joined #git
15:55 nurey joined #git
15:55 WeiJunLi Vampire0: it shows a bunch of files when i do git checkout branch-name
15:55 WeiJunLi and in the end "Already on 'branch-name" '
15:55 WeiJunLi the list of files is quite big and upload it take some time and my connection is slow
15:56 WeiJunLi i just did git clone and then now git checkout branch-name
15:56 WeiJunLi and the directory is still empty...
15:56 Es0teric joined #git
15:57 Es0teric_ joined #git
15:57 porsche944 joined #git
15:57 porsche944_ joined #git
15:57 cyberz joined #git
15:58 star_prone joined #git
15:58 star_prone joined #git
15:58 WeiJunLi you can try yourself
15:59 shgysk8zer0 joined #git
16:00 synthroid joined #git
16:01 durham joined #git
16:03 roelmonn_ joined #git
16:04 GodGinrai hey guys, when I use rebase --onto, it seems that it just replaces my branch with the branch I wanted to rebase onto...
16:07 Vampire0 WeiJunLi, of course I can try myself, but it will work for me
16:07 WeiJunLi already figured it out
16:07 WeiJunLi tks
16:07 Vampire0 WeiJunLi, ok, what was it?
16:07 chardan joined #git
16:07 gitinfo WeiJunLi: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx", "ty" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
16:08 madewokherd GodGinrai: I don't know what you're trying to do, but I think of it this way: "upstream" and "branch" determine the set of commits to be applied, and "onto" is the base they are applied to
16:08 Vampire0 GodGinrai, !repro
16:08 gitinfo GodGinrai: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
16:08 theoceaniscool joined #git
16:13 GodGinrai madewokherd: it goes --onto <newbase> <oldbase> <branch>, right?
16:13 GodGinrai Vampire0: https://pastebin.mozilla.org/8982649 <-- not gonna help much
16:14 madewokherd GodGinrai: seems like all commits in feature are already in either stage or develop
16:15 GodGinrai madewokherd: ah, good point.  They are likely already merged into develop
16:15 TbobbyZ joined #git
16:17 GodGinrai madewokherd: I think that was the problem, thanks :)
16:18 raspado joined #git
16:19 matoro joined #git
16:20 orbyt_ joined #git
16:21 Skyfinn joined #git
16:22 Gustavo6046_ joined #git
16:25 noecc joined #git
16:25 matoro joined #git
16:27 marcogmonteiro joined #git
16:27 col joined #git
16:30 Gustavo6046 joined #git
16:30 irc2000 joined #git
16:32 aielima joined #git
16:38 zeroed joined #git
16:38 zeroed joined #git
16:38 bariscant joined #git
16:39 sz0 joined #git
16:40 tlaxkit joined #git
16:40 sgen git wont let me apply a set of stashed changes, it gives me a list of files with the error: 'already exists, no checkout' after them. How can I apply this stash?
16:44 Takumo Hi all, is there a way to squash/prune commit history from >2 years ago?
16:44 Takumo I've got a repo where the number of commits is making any kind of log/diff operation very slow, and I never need such old data
16:46 acetakwas joined #git
16:47 gugah_ joined #git
16:50 HardlySeen joined #git
16:50 Vampire0 Takumo, with Git you can do almost anything you want, but I don't think the number of commits should significantly influence the log / diff speed
16:51 Vampire0 Takumo, be aware that you !rewrite history if you do what you asked for. If you don't care about the implications, describe in more detail how exactly you want the output to look like, to be able to give a proper advice
16:51 gitinfo Takumo: Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
16:53 Takumo fair enough
16:53 Silenced joined #git
16:53 Takumo I guess 11k commits over 5 years isn't that many raelly
16:53 Silenced joined #git
16:55 d0nkeyBOB joined #git
16:55 hashtagstrashtag joined #git
16:56 Vampire0 Takumo, I think so too
16:56 gingitsune joined #git
16:56 Vampire0 Takumo, I'm currently mainly working with a repo with 512.758 commits and I'm fine
16:56 Takumo I think its probably more to do with the rather poor structure of some of the code in this repo
16:56 Takumo at least that'd explain the slow diffs
16:57 Vampire0 Ah, bullshit, 62.773 commits
16:57 Vampire0 forgot the --oneline
16:57 Takumo Vampire0: I just used `git rev-parse HEAD --count` to get commit count
16:57 Takumo which gave me 11.130
16:57 Vampire0 with which version of Git?
16:58 Takumo 2.12.0
16:58 Takumo but should work on much older versions
16:58 j08nY joined #git
16:58 Vampire0 git rev-parse has no --count option
16:58 Takumo sorry its `git rev-list HEAD --count`
16:58 Vampire0 ah, ok
16:59 Vampire0 50.510, but there the branches are missing of course
16:59 Vampire0 git rev-list --all --count
16:59 Vampire0 That gives 62.078
17:00 Takumo this repo used to be in svn so not much in the branch department
17:02 Mobutils joined #git
17:02 matoro joined #git
17:02 liberpater left #git
17:03 Cabanossi joined #git
17:04 diogenese joined #git
17:05 Vampire0 Takumo, the one has nothing to do with the other. You can have many things in branches in SVN too. The repo I actually talk about is  a git-svn clone, so 12.000 commits are in branches other than trunk
17:05 Takumo ok, they didn't use branches back in svn days
17:08 GodGinrai Takumo: they had them, people just didn't like to use them because they were expensive
17:08 Takumo I know, but for that same reason, we didn't use them
17:09 cdown_ joined #git
17:10 jagob joined #git
17:10 Vampire0 GodGinrai, it was less that they were expensive, that was the problem of the IT guys, it is more that merging is a PITA with SVN.
17:10 s1scha joined #git
17:11 svm_invictvs joined #git
17:11 tang^ joined #git
17:11 d1z joined #git
17:12 cdown_ joined #git
17:13 GodGinrai Vampire0: I won't deny the point about merging.  Although you could also consider that expensive :P
17:13 Vampire0 yes, of course :-P
17:13 boombatower joined #git
17:15 multi_io joined #git
17:16 gunnaro joined #git
17:18 hahuang65 joined #git
17:19 Ryanar joined #git
17:23 kaushal joined #git
17:25 puff Is there an easy way to do "git log -n i" where "i" defaults to "back to where this branch started"?
17:26 hahuang61 joined #git
17:26 ash_workz I think I accidentally pulled to master when I shouldn't have... how do I roll back?
17:26 Vampire0 puff, depends on how you define "where the branch started". A branch is just a post-it note on a commit. And it actually started at the root commit.
17:27 Vampire0 puff, but you are probably after "git log master..feature" which gives you the commits that are reachable from feature but not from master.
17:28 Vampire0 ash_workz, is this only locally or already pushed?
17:28 ash_workz Vampire0: already pushed
17:28 Vampire0 ash_workz, then !rewrite applies
17:28 gitinfo ash_workz: Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
17:29 rt joined #git
17:29 ash_workz Vampire0: I mean, right now I just want to rewind 1 merge
17:29 ams__ joined #git
17:30 ash_workz remotely
17:30 Chinggis6 joined #git
17:30 Vampire0 ash_workz, if that merge is already pushed, !rewrite still applies, no matter how much you pushed, every commit counts
17:30 gitinfo ash_workz: Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
17:30 Vampire0 ash_workz, if you are aware of the problems and don't care, you can simply reset to before the merge and do a force-push if the remote server allows it
17:30 puff Vampire0: Thanks.
17:31 joshszep joined #git
17:31 sns joined #git
17:31 ash_workz Vampire0: I don't get it; but let me explain the situation more clearly; I had committed something (or so I thought) and pushed it and I pulled the whole branch remotely and I think it messed up some scripts somehow, so I just want to undo that specific merge on the remote server
17:32 joshszep joined #git
17:33 Snugglebash joined #git
17:33 exarch joined #git
17:33 Cabanossi joined #git
17:33 Vampire0 ash_workz, I'm still not sure about your situation. Do you use Git for deployment and just want to checkout another commit on the production server, or do you want to undo a merge in the Git repository?
17:33 Mobutils joined #git
17:34 robattila256 joined #git
17:34 Duder9999 joined #git
17:34 Snugglebash joined #git
17:35 crayon joined #git
17:37 j416 is there a difference between "git pull --rebase" and "git fetch && git rebase"? I seem to recall that pull --rebase has some peculiar difference as compared to regular rebase, but I may be mistaken.
17:38 publio joined #git
17:41 matoro joined #git
17:41 Vampire0 j416, yes, there is, git fetch will fetch all branches, git pull --rebase will only fetch the branch to be rebased
17:41 thiago joined #git
17:42 wrkrcoop joined #git
17:42 wrkrcoop hi when i run “git show-branch mybranch” it shows only thte last commit, how can i get the last maybe 20?
17:42 nedbat wrkrcoop: git log mybranch
17:42 wrkrcoop nedbat: thank you
17:44 grayjoc joined #git
17:45 TomyLobo2 joined #git
17:45 j416 Vampire0: false
17:47 orbyt_ joined #git
17:47 GodGinrai #github is about as useful as a treehugger at a logging festival
17:48 GodGinrai no wonder everyone asks their github questions here
17:48 bremner I guess they have some web forums or something?
17:48 GodGinrai bremner: there are 510 people in there.  It's just that no one talks
17:49 j416 Vampire0: http://pastebin.com/pgtThaQv
17:49 * bremner adds the silence of #github to the list of SEPs
17:51 thiago joined #git
17:51 puff SEP?
17:51 bremner Someone Else's Problem (TM) ;)
17:51 alexandre9099 joined #git
17:52 j416 :D
17:52 hahuang62 joined #git
17:54 muhannad__ joined #git
17:56 GodGinrai bremner: while it is someone else's problem, it does have bearing here, since it is one of the reasons we have to answer so many github questions :P
17:56 dreiss joined #git
17:57 theoceaniscool joined #git
17:58 Noldorin joined #git
17:59 drizbit joined #git
17:59 ajpiano joined #git
17:59 miczac joined #git
18:01 roelmonnens joined #git
18:01 Es0teric joined #git
18:02 abh joined #git
18:02 livingstn joined #git
18:04 cdg joined #git
18:04 dglambert joined #git
18:04 Starky joined #git
18:05 Skyfinn joined #git
18:05 rominron_ joined #git
18:06 parasite_ joined #git
18:07 freimatz joined #git
18:07 cdown joined #git
18:08 Sasazuka joined #git
18:09 Es0teric_ joined #git
18:11 ertesx joined #git
18:12 lorenzomaieru joined #git
18:15 kexmex joined #git
18:16 tango_ joined #git
18:17 eroux joined #git
18:18 kristofferR joined #git
18:22 raijin joined #git
18:22 xaviergmail joined #git
18:24 roentgen joined #git
18:24 Anja joined #git
18:24 shgysk8zer0 joined #git
18:25 miczac joined #git
18:25 raynold joined #git
18:26 skalpin a lot of people thing github = git
18:26 skalpin at least windows devs :/
18:28 warthog9 joined #git
18:28 roelmonnens joined #git
18:29 venmx joined #git
18:30 GodGinrai skalpin: well it is when you assign it like that :P
18:32 marcogmonteiro joined #git
18:32 skalpin lol
18:33 Cabanossi joined #git
18:34 star_prone joined #git
18:34 roelmonnens joined #git
18:35 j416 omg what did you do
18:36 j416 at least you didn't git = github
18:36 j416 ...wait
18:36 cbreak joined #git
18:36 dviola joined #git
18:37 WeiJunLi joined #git
18:37 ij GodGinrai, That's not assignment, that's a truth statement whose truth value was asserted only in the case of some observers, not generally.
18:37 timetrex joined #git
18:39 cbreak joined #git
18:39 gopar joined #git
18:40 roelmonnens joined #git
18:41 nod joined #git
18:41 ankit01ojha joined #git
18:42 skalpin some would say alternative facts
18:44 Darcidride joined #git
18:44 TbobbyZ_ joined #git
18:46 GodGinrai ij: you're under arrest for murder of an innocent joke
18:46 GodGinrai you have the right to remain silent
18:46 GodGinrai anything you say can and will be logged by the many bots sitting in this channel
18:46 _ikke_ lol
18:46 finalbeta1 joined #git
18:48 UrsoBranco joined #git
18:51 notebox joined #git
18:52 mns joined #git
18:53 mehola joined #git
18:55 dsteplight__ joined #git
18:57 minibar left #git
18:59 pinkkis joined #git
18:59 AndroIRCScreensh joined #git
19:00 Vampire0 j416, hm, ok, then this either changed or I remembered wrongly
19:00 Vampire0 j416, but you can set GIT_TRACE=1 and then run the commands in question, then you will see whether there are differences
19:01 j416 good point
19:01 j416 I'll try that
19:01 j416 thanks
19:01 j416 rm -rf'd my test repos so I'll do it some other day I guess ;)
19:05 satifant joined #git
19:06 aidalgol joined #git
19:07 dvaske_ joined #git
19:08 cyberz joined #git
19:08 Discovery joined #git
19:09 raijin joined #git
19:09 fedepad joined #git
19:09 CaptainKnots joined #git
19:12 aielima joined #git
19:12 TbobbyZ joined #git
19:12 d1z joined #git
19:14 Derperperd joined #git
19:15 overlord_tm joined #git
19:16 Sample joined #git
19:16 inflames joined #git
19:17 dave0x6d joined #git
19:17 Sample joined #git
19:17 MineCoins joined #git
19:18 star_prone joined #git
19:18 jstimm joined #git
19:18 star_prone joined #git
19:18 _Vi joined #git
19:19 tctara joined #git
19:21 Sanqui joined #git
19:21 Guest55 joined #git
19:21 livingstn joined #git
19:22 Balliad joined #git
19:23 cyberz joined #git
19:24 Sanqui hi.  after a `git clone` && `git submodule update --init --recursive`, why would `git status` say the submodule folder is modified?
19:24 tobiasvl Sanqui: what does it say is the modification?
19:25 ugjka joined #git
19:25 Sanqui tobiasvl: the difference is
19:25 Sanqui -Subproject commit 0e1798937a4bf723813574281d0dc12c471c9199
19:25 Sanqui +Subproject commit 0e1798937a4bf723813574281d0dc12c471c9199-dirty
19:25 Vampire0 line endings?
19:25 Sanqui dunno, i'm on linux
19:25 Vampire0 hm, then probably not
19:25 Sanqui hm
19:25 Sanqui actually I can cd into the submodule
19:26 Sanqui and it does look like there are newline differences
19:26 Sanqui because entire files are diffed
19:26 fahadash joined #git
19:26 Sanqui but it's kind of stressing me out that the submodule is 'dirty' immediately after a init/update
19:28 Es0teric joined #git
19:28 ejb joined #git
19:28 kyan joined #git
19:29 Es0teric_ joined #git
19:29 cdown joined #git
19:29 peterbecich joined #git
19:30 Sanqui thanks for the hints, I'm gonna resolve this by --ignore-submodules=all for now
19:30 courrier joined #git
19:34 Maxxed_ joined #git
19:34 buffal0 joined #git
19:35 muhannad__ joined #git
19:35 Maxxed_ left #git
19:36 Starky joined #git
19:36 socomm joined #git
19:36 socomm left #git
19:38 livingstn joined #git
19:38 Maxxed joined #git
19:39 Maxxed i have a noob question..
19:39 _ikke_ !just_ask
19:39 gitinfo You can just ask your question. If anybody knows the answer, they will answer soon (most of the time)
19:39 Maxxed $ git push
19:39 Maxxed remote: Permission to miramax99/General.git denied to xyman-it.
19:39 Maxxed fatal: unable to access 'https://github.com/miramax99/General/': The requested URL returned error: 403
19:40 gareppa joined #git
19:40 Maxxed where is that user name coming from?
19:40 Maxxed is there some setting or something im missing?
19:40 Maxxed xyman-it, idk wtf that is :p
19:40 _ikke_ Maxxed: git remote -v show
19:41 e14 joined #git
19:41 Maxxed $ git remote -v show
19:41 Maxxed originhttps://github.com/miramax99/General (fetch)
19:41 Maxxed originhttps://github.com/miramax99/General (push)
19:41 _ikke_ Ok, not from there
19:41 Emperor_Earth joined #git
19:42 _ikke_ git config credential.helper
19:42 Maxxed ah hah!
19:42 Maxxed $ git config credential.helper
19:42 Maxxed osxkeychain
19:42 Maxxed think i might have it from here..
19:42 crose joined #git
19:43 ftsp joined #git
19:43 Maxxed yep, got it, thanks man :D
19:45 _ikke_ yw
19:46 Guest55 joined #git
19:46 raijin joined #git
19:46 Sanqui left #git
19:47 cyberz joined #git
19:49 Darcidride joined #git
19:50 cr34ton joined #git
19:51 crose joined #git
19:52 learning joined #git
19:53 Sasazuka joined #git
19:53 cyberz joined #git
19:53 MasterNayru joined #git
19:55 ajpiano joined #git
19:55 ejb left #git
19:55 yorkshre joined #git
19:55 irqq joined #git
19:56 cdg joined #git
19:56 mda1 joined #git
19:57 GodGinrai joined #git
19:57 yorkshre left #git
19:57 cdg joined #git
20:00 luke_nukem joined #git
20:03 tx32 joined #git
20:04 tx32 left #git
20:04 msonntag joined #git
20:06 cr34ton joined #git
20:08 kexmex joined #git
20:13 Emperor_Earth_ joined #git
20:13 BSAlb joined #git
20:14 Sample_ joined #git
20:14 a3Dman joined #git
20:14 Hez_ joined #git
20:15 Nevik_ joined #git
20:16 HandheldPenguin` joined #git
20:16 dfas` joined #git
20:16 Neic0 joined #git
20:16 gargola_ joined #git
20:16 frankie64 joined #git
20:17 King_Hual` joined #git
20:17 wget_ joined #git
20:17 lucasem_ joined #git
20:18 ntnn_ joined #git
20:18 heroux_ joined #git
20:18 l2y_ joined #git
20:18 maqr_ joined #git
20:18 thunderbirdtr joined #git
20:18 nephyrin` joined #git
20:18 BlessJah_ joined #git
20:18 lmat_ joined #git
20:19 draget_ joined #git
20:19 Kulrak|2 joined #git
20:19 runlevl4_ joined #git
20:19 Vector_Gorgoth joined #git
20:19 someperson joined #git
20:19 tilpner_ joined #git
20:19 styler2go_ joined #git
20:19 manuels joined #git
20:19 alphor_ joined #git
20:19 Kronuz_ joined #git
20:19 worstadmin_ joined #git
20:19 kaosine_ joined #git
20:19 obiiwahn joined #git
20:19 Tyklol joined #git
20:19 nowhereman joined #git
20:19 d0nn1e_ joined #git
20:20 NightMonkey_ joined #git
20:20 Ice_Strike2 joined #git
20:20 bket1 joined #git
20:20 xkr47_ joined #git
20:20 hvxgr___ joined #git
20:20 Loomaanaatii_ joined #git
20:20 prions_ joined #git
20:20 PaulePan1er joined #git
20:20 KG4PEQ__ joined #git
20:20 kaosine_ left #git
20:20 rwp_ joined #git
20:20 bl0m1_ joined #git
20:21 Xarodon_ joined #git
20:21 kaosine joined #git
20:21 byyte512 joined #git
20:21 shennyg_ joined #git
20:21 sonOfRa_ joined #git
20:22 bauruine_ joined #git
20:22 DuncanT_ joined #git
20:22 cyphase_eviltwin joined #git
20:22 fuzzmz_ joined #git
20:22 dwmw2_gone joined #git
20:22 utrack_ joined #git
20:22 da5id_ joined #git
20:23 HedgeMage_ joined #git
20:23 ahrs_ joined #git
20:23 mloy joined #git
20:23 ngladitz_ joined #git
20:23 oleo_ joined #git
20:23 qassim_ joined #git
20:23 sitaram_ joined #git
20:23 sitaram_ joined #git
20:23 thejoecarroll_ joined #git
20:23 meZee- joined #git
20:23 leah2_ joined #git
20:23 HelloAtMega joined #git
20:23 pwndave_ joined #git
20:23 toydestroyer_ joined #git
20:24 BackEndCoder_ joined #git
20:24 Rodya_ joined #git
20:24 ome_ joined #git
20:24 tcsc_ joined #git
20:24 BrianBlaze420 joined #git
20:24 Sasazuka joined #git
20:24 strugee_ joined #git
20:24 regedit_ joined #git
20:24 atrus- joined #git
20:25 Kobazz joined #git
20:25 madprops_ joined #git
20:25 navidr_ joined #git
20:25 Balliad joined #git
20:25 dminuoso_ joined #git
20:26 scarcry_ joined #git
20:26 Elysia joined #git
20:26 lolmac_ joined #git
20:26 monomon[m]1 joined #git
20:27 askb joined #git
20:28 ulkesh_ joined #git
20:28 UncleCJ1 joined #git
20:28 MindfulMonk_ joined #git
20:29 trelane joined #git
20:29 trelane joined #git
20:29 trelane joined #git
20:29 mns joined #git
20:29 Remramm joined #git
20:29 ponyofdeath joined #git
20:30 skelterjohn joined #git
20:30 cmalmber joined #git
20:30 SpeakerToMeat joined #git
20:30 pickfire joined #git
20:31 varesa_ joined #git
20:31 quite joined #git
20:31 xaviergmail joined #git
20:32 y_sekino joined #git
20:32 Sleepy63 joined #git
20:32 PCatinean joined #git
20:32 joshszep joined #git
20:32 Soni joined #git
20:32 boxrick1 joined #git
20:33 Awesomecase joined #git
20:33 gajus joined #git
20:33 mehdi___ joined #git
20:33 Cabanossi joined #git
20:33 elijah joined #git
20:33 fahadash joined #git
20:34 g0th joined #git
20:34 elmcrest joined #git
20:34 Iacobus joined #git
20:34 eshlox joined #git
20:34 mda1 joined #git
20:34 anry joined #git
20:34 swalladge joined #git
20:34 cmosguy joined #git
20:34 wizonesolutions joined #git
20:34 themill joined #git
20:34 nachoman joined #git
20:34 timwis joined #git
20:35 ahrs joined #git
20:35 ome joined #git
20:35 raynold joined #git
20:35 Gunni joined #git
20:36 apotry joined #git
20:36 learning joined #git
20:37 ome joined #git
20:37 Pseudocrat joined #git
20:38 joshszep joined #git
20:38 Srep joined #git
20:38 MrWoohoo joined #git
20:38 linuxmodder joined #git
20:39 unholycrab joined #git
20:40 ribasushi joined #git
20:41 jottr joined #git
20:44 krabat joined #git
20:44 cust joined #git
20:45 forgot joined #git
20:45 raijin joined #git
20:45 frekoya_ joined #git
20:45 roelmonnens joined #git
20:46 tom67 joined #git
20:47 dserodio joined #git
20:47 aard_ joined #git
20:49 musicmatze joined #git
20:49 Celelibi joined #git
20:50 simongoransson joined #git
20:50 spm_draget joined #git
20:51 SCHAAP137 joined #git
20:51 roelmonnens joined #git
20:52 wrkrcoop joined #git
20:52 d1z joined #git
20:53 c0ded joined #git
20:53 seni joined #git
20:54 gunnaro joined #git
20:54 mehola joined #git
20:54 AaronMT joined #git
20:55 crose joined #git
20:55 roentgen joined #git
20:56 CuriousMan joined #git
20:56 flopsec joined #git
20:58 CuriousMan Hello, I imported a gitlab project in github with lfs. The github project can't find the lfs files on the server now though. Does anyone know how to properly import a lfs project?
20:58 tmg joined #git
20:59 star_prone joined #git
21:04 finalbeta joined #git
21:04 Rodya_ joined #git
21:07 Eugene Nope! Welcome to proprietary(non-core) git extensions. THanks, Github!
21:07 Eugene </tinfoilhat>
21:07 steeze joined #git
21:08 bremner I thought there was a free reference implimentation of git-lfs.
21:08 steeze i get this error when trying to
21:08 steeze `git branch stash fooooo`
21:08 steeze Not a valid object name: 'app/fixes/saving-user-carts'.
21:08 Eugene That doesn't make the development and specification any less controlled by a single entity
21:08 bremner ack.
21:08 steeze but it works when not doing the stash, is there a way to do git-flow like folders with git branch stash?
21:09 bremner git-flow has folders?
21:09 _ikke_ git branch stash would create a branch called stash
21:09 steeze ah did i just do them out of order?
21:09 cdg joined #git
21:10 steeze silly me
21:10 _ikke_ bremner: I think he meant the foo/bar/bax structure
21:10 cdg joined #git
21:11 steeze yeah just talking about the naming convention (that a lot of guis put into "folders"), sorry i was being ambiguous
21:11 bariscant joined #git
21:12 Atm0spher1c joined #git
21:13 jost_ joined #git
21:16 aidalgol joined #git
21:16 sword joined #git
21:17 sq-one joined #git
21:19 xkr47_ left #git
21:19 xkr47 joined #git
21:20 dvaske joined #git
21:23 m0viefreak joined #git
21:24 dsdeiz joined #git
21:24 theoceaniscool joined #git
21:24 frekoya_ joined #git
21:24 someperson left #git
21:24 nobso joined #git
21:26 brent__ joined #git
21:26 brent__ joined #git
21:26 t-mart joined #git
21:27 mda1 joined #git
21:32 nixjdm joined #git
21:33 Cabanossi joined #git
21:35 _W_ joined #git
21:40 mikeric joined #git
21:41 plantedseed joined #git
21:41 adas0496_ joined #git
21:42 plantedseed left #git
21:45 wrkrcoop joined #git
21:46 livingstn joined #git
21:46 masuberu joined #git
21:46 _Alex-N_ joined #git
21:47 oncall-pokemon joined #git
21:49 nowhereman joined #git
21:49 jstimm joined #git
21:51 raijin joined #git
21:52 cdown_ joined #git
21:54 IntruderSRB joined #git
21:55 atomi joined #git
21:58 Atm0spher1c joined #git
22:01 mrkake joined #git
22:03 venmx joined #git
22:03 bket joined #git
22:03 Cabanossi joined #git
22:03 lucido-cl joined #git
22:05 kyan joined #git
22:06 Sasazuka joined #git
22:07 oleo joined #git
22:08 joshszep joined #git
22:09 Mathiasdm joined #git
22:09 CheckDavid joined #git
22:10 a3Dman joined #git
22:10 bket joined #git
22:10 Dougie187 left #git
22:11 Werring joined #git
22:14 kristofferR joined #git
22:14 username1 joined #git
22:14 livingst_ joined #git
22:14 username1 is there a way to stash commit objects?
22:14 Atm0spher1c joined #git
22:14 username1 i have ~10 commits i want to rebase into master eventually
22:14 username1 but for now, i just want to rebase one
22:14 _ikke_ branches
22:15 username1 _ikke_: ive already made the commit objects
22:15 username1 will that work?
22:15 username1 your saying put those 9 commits into a branch, except for the one i want to rebase into master
22:15 username1 then update origin master
22:15 username1 then bring back those commits into master?
22:16 _ikke_ I think cherry-pick is better in this case, instead of rebase
22:16 polyrob joined #git
22:16 polyrob joined #git
22:17 username1 _ikke_: thank you kindly
22:17 username1 will look into it
22:18 matoro joined #git
22:19 ToBeCloud joined #git
22:20 snowalpaca joined #git
22:24 irc2000 joined #git
22:26 Atm0spher1c joined #git
22:27 gopar joined #git
22:29 aielima joined #git
22:29 courrier joined #git
22:30 nurey joined #git
22:37 Groovin joined #git
22:38 dirtyroshi joined #git
22:39 dviola joined #git
22:41 raijin joined #git
22:46 buffal0 joined #git
22:48 thiago joined #git
22:49 TikityTik joined #git
22:52 Sasazuka joined #git
22:53 CaptainKnots joined #git
22:54 mikecmpbll joined #git
22:55 mehola joined #git
22:57 vamiry_ joined #git
23:00 timwis joined #git
23:00 raynold joined #git
23:00 ome joined #git
23:00 unholycrab joined #git
23:00 flopsec joined #git
23:00 oncall-pokemon joined #git
23:00 CheckDavid joined #git
23:00 aielima joined #git
23:01 joshszep_ joined #git
23:01 livingstn joined #git
23:03 gugah joined #git
23:03 Cabanossi joined #git
23:07 venmx joined #git
23:08 ggg__ joined #git
23:08 raijin joined #git
23:08 dvaske joined #git
23:08 Derperperd joined #git
23:09 username1 is it possible to stash commits away including their messages?
23:13 Sasazuka joined #git
23:13 Random832 jut make a branch from your current head
23:13 __rgn you can drop them with rebase and cherry-pick back by sha
23:14 Random832 a rebase is essentially a sequence of cherry-picks anyway
23:19 menip joined #git
23:19 gopar joined #git
23:21 username1 Random832: git rebase onto a new branch?
23:21 username1 alternatively, I could just do "git push master hash"?
23:21 username1 Or is that bad practice
23:23 username1 "git push origin hash:master" **
23:23 bitmod joined #git
23:23 username1 as long as its the the oldest commit locally
23:24 username1 __rgn: do you mean git rebase onto the new branch?
23:26 gopar joined #git
23:26 leeN joined #git
23:28 __rgn interactive rebase
23:33 kristofferR joined #git
23:35 livingstn joined #git
23:35 matoro joined #git
23:42 min_ joined #git
23:42 Anja_ joined #git
23:43 NeverDie joined #git
23:44 RalphJr45 joined #git
23:46 irqq joined #git
23:47 janx joined #git
23:49 miczac joined #git
23:51 _Cyclone_ joined #git
23:52 pasky joined #git
23:55 anuxivm joined #git
23:57 duderonomy joined #git

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