Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-12-15

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

All times shown according to UTC.

Time Nick Message
00:00 joshszep joined #git
00:00 arescorpio joined #git
00:00 texasmynsted would it be safer to always push to a named branch and delete and re-create the branch prior to pushing?
00:01 kadoban Gotta run, good luck. Koognod: only way to find out is to ask (specify details, etc.)
00:01 texasmynsted Or is there another way to purge all history?
00:02 matsaman purge all history?
00:02 Koognod trying to build cyanogenmod for my device from source however when syncinc with repo I keep getting fatal error "not a git repository" I'm guessing this has something to do with how I init the repo.
00:02 diogenese joined #git
00:03 Koognod it says I got at least 11 gigs in the local folder so I must have been doing something right.
00:03 safe joined #git
00:03 dersand joined #git
00:04 texasmynsted all history
00:05 dersand Maybe a question more targeted for linux/terminal environment. But if i'm writing the commit message in e.g. vim and want to check the diff/the changed files, is there any command i could perform?
00:05 texasmynsted how are you syncing the repo?
00:05 tmsmith joined #git
00:05 matsaman dersand: the diff of what, the commit?
00:05 dersand matsaman, the staged diff yes
00:06 matsaman dersand: can you clarify how vim is related to this question?
00:06 Koognod repo sync -f -j8 --force-sync
00:07 qqx dersand: For that I always run commit in verbose mode, so that git includes the diff in the file it passes to the editor.
00:07 ekneuss joined #git
00:07 githelp @gitinfo that git log command was actually enough to send me on my way! Turned out it was not missing and indeed in the graph just wasn't displayed in git log!
00:07 mizu_no_oto joined #git
00:07 dersand qqx it was something like that I was looking for, many thanks
00:08 qqx dersand: For recent versions of git theres the commit.verbose config to enable that. But I generally use an alias for commit that includes the -v option.
00:08 dersand qqx, [alias]ci=commit -v?
00:08 qqx dersand: Exactly what I have for it.
00:09 nettoweb joined #git
00:10 mojave all right, looks like i got a basic handle on git.. but that was a lot of fenangling
00:10 Koognod do I have to init repo everytime I'm trying to sync or just once
00:10 mojave or *finangling rather
00:11 Koognod and why is my sync ending at 97% so close to being finished.
00:12 Neelixx joined #git
00:13 diogenese joined #git
00:14 texasmynsted repo sync?
00:15 unbalanced joined #git
00:15 texasmynsted I did a "man git-sync" and get nothing
00:15 Anja joined #git
00:15 mizu_no_oto joined #git
00:16 prg3 joined #git
00:19 mizu_no_oto joined #git
00:20 matsaman joined #git
00:21 texasmynsted if I set an alias with git config with the same name as an existing alias the new version will replace the old, correct?
00:21 raspado joined #git
00:23 rewt theoretically
00:31 bluepixel joined #git
00:31 nebg joined #git
00:32 Koognod trying to build cyanogenmod for my device from source however when syncinc with repo I keep getting fatal error "not a git repository" at 97%, can anyone help me figure out why this is happening?
00:32 zzz_ joined #git
00:32 Neelixx joined #git
00:33 Koognod herrrow pwreeze
00:34 matsaman Koognod: what's the command you're using?
00:34 dermoth_ joined #git
00:35 Koognod repo sync - j8 --force-sync
00:35 matsaman Koognod: 'repo' is not 'git'
00:35 Koognod oh
00:36 matsaman using mr or radmind?
00:36 matsaman anyway, if you're having a problem (& you are), try using a plain old 'git' command first
00:36 SexgirLL joined #git
00:39 Koognod ty
00:43 Gsham joined #git
00:43 sweatsuit joined #git
00:45 cqi joined #git
00:47 Darren_ joined #git
00:51 tmk Unable to initialize SMTP properly. Check config and use --smtp-debug. VALUES: server=smtp.gmail.com encryption=tls hello=fedora.localdomain port=587 at /usr/libexec/git-core/git-send-email line 1383.
00:52 tmk can anybody figure out what mistake i am doing ?? got error while sending email
00:52 rewt did you do as requested?
00:52 Neelixx joined #git
00:54 tmk you mean debug?
00:54 nicksloan joined #git
00:55 rewt and the other thing
00:55 tmk rewt: i also tried --smtp-debug but still getting the same error
00:55 rewt and nothing else?
00:56 tmk tcpdump
00:56 rewt i mean no other output than that single line?
00:56 tmk yes
00:57 qqx tmk: did tcpdump indicate that it was able to connect?
00:57 rewt did you put just "--smpt-debug" or "--smtp-debug=1" ?
00:57 tmk no it was not able to capture packets
00:57 askb joined #git
00:58 tmk yes i made --smtp-debug=1 for enable
00:58 qqx Are you able to connect to port 587 of smtp.gmail.com?
00:59 DolpheenDream joined #git
00:59 tmk qqx: i am not getting what you are asking,how to test that
01:00 qqx tmk: My usual way to check that would be `telnet smtp.gmail.com 587`
01:01 tmk qqx: not tried
01:01 tmk how to do that?
01:01 qqx tmk: With the command I quoted
01:02 tmk qqx: ok, let me try this
01:02 FilipNortic joined #git
01:03 dzho openssl s_client -starttls smtp -crlf -connect smtp.gmail.com:587
01:04 dzho then you type literally anything in there, hit return, and it will either reply having recognized it as a valid command, or will tell you it doesn't recognize it as a valid command, but at least you'll know you are able to reach it.
01:05 mjago joined #git
01:05 qqx At least with telnet it should indicate if the connection succeeded or not even without typing anything more.
01:05 qqx Easy command that will be recognized and useful is `QUIT`.
01:05 madewokherd joined #git
01:06 tmk qqx:this telnet command just gives the output   Trying 74.125.200.108
01:06 qqx tmk: Sounds like something is blocking those connections then.
01:06 tmk qqx:connection time out
01:06 tmk network is unreachable
01:07 qqx tmk: Could be your ISP trying to force you to use their own SMTP relay.
01:07 qqx tmk: The issue then has nothing to do with git.
01:08 tmk qqx: but when i use my mobile internet still it shows the same thing.
01:10 unbalanced joined #git
01:10 zzz_ joined #git
01:10 kenoob joined #git
01:10 kenoob joined #git
01:10 bouncy joined #git
01:12 tmk dzho; i typed openssl s_client -starttls smtp -crlf -connect smtp.gmail.com:587  and the www.google.com
01:13 dzho and the what
01:13 tmk it gives socket:Bad file descriptor
01:13 dzho ah
01:13 tmk connect:errno=9
01:13 tmk what does it mean?
01:14 dzho it probably means your network connection is being filtered or blocked
01:15 tmk how do i make it work?
01:15 dzho move someplace with a full network connection?
01:15 tmk dzho: should i try with my mobile data?
01:15 dzho contact your network provider and report the bug?
01:15 dzho maybe?
01:15 dzho tmk: where are you?
01:16 tmk are you asking me location>
01:16 dzho yes
01:16 tmk Bangalore
01:16 dzho hmm.
01:16 dzho not an obvious problematic locale
01:17 dzho what qqx said, though, your ISP may wish you to use their mail relays
01:18 dzho hmm, 587? not 465?
01:18 dzho https://www.fastmail.com/help/technical/ssltlsstarttls.html
01:18 qqx It's a fairly common practice for the traditional SMTP port (25), it really shouldn't be done for port 587 though. The whole point of that is to allow that.
01:18 qqx dzho: 587 is the standard port for message submission.
01:18 dzho oh, right.
01:18 * dzho nods
01:19 dzho sorry, I would be even more at sea about all this than I already am had I not been reading that fastmail explainer recently.
01:19 dzho skimmed again just now too quickly.
01:20 nettoweb joined #git
01:21 anonymuse joined #git
01:23 MissionCritical joined #git
01:24 Sasazuka joined #git
01:26 tmk joined #git
01:31 tmk dzho:but it worked few hours before only for one time and i was able to send my patch
01:34 tmk dzho,qqx: it worked :)
01:35 buffal0 joined #git
01:36 Sasazuka joined #git
01:37 serialoverflow joined #git
01:42 mjago joined #git
01:46 heroux joined #git
01:47 tmk joined #git
01:51 str joined #git
01:53 matsaman joined #git
01:55 netj joined #git
01:57 mojave i heard git can destroy your data remotely if you aren't careful.. is that true?
01:58 matsaman mmm, SexgirLL is spamming all new joiners
01:58 matsaman mojave: git can barely destroy data at all
01:58 Nilesh_ joined #git
01:58 matsaman although it can give the appearance of it being destroyed
01:59 mojave and i mean destroying the data locally
02:00 matsaman mojave: git can barely destroy data at all
02:00 matsaman I don't particularly doubt it's true that you _heard_ it can, though =P
02:00 mojave i got it from this http://benno.id.au/blog/2011/10/01/git-recursive-merge-broken  .. but it appears as if he just did a bad thing
02:00 mojave that's what i got out of it
02:00 matsaman aha, well that's a blog
02:00 matsaman you see, blogs are where misinformation is born & lives
02:00 mojave well sure, it's his story
02:01 * matsaman looks
02:01 mojave but i did want to see what you guys thought about it
02:01 mojave or, the idea
02:01 matsaman personally I don't need git to destroy data, I already have rm for that
02:01 matsaman every tool you use you need to learn how ot use
02:01 mojave lol
02:01 mojave that is true
02:02 * matsaman continues to scroll, looking for the point in this blog post
02:02 Olipro joined #git
02:02 mojave i think his major point is in the fourth paragraph
02:02 mojave like, how it happened for him
02:04 matsaman okay, well
02:04 matsaman for starters, he's talking about losing data that he never committed
02:05 garethdaine joined #git
02:05 matsaman that's like... the whole reason you use a VCS: so you can commit and not lose data
02:05 mojave right.
02:05 mojave i see some of the commenters telling him that it's because he did a bad thing
02:05 mojave i.e. pulled without comitting
02:05 matsaman yeah the first comment on a blog post is usually where you see the explanation that the essay typed up is wrong
02:06 matsaman anyway, I don't like git, but when you learn how to use it halfway the way it was intended to be used, you're fine
02:06 tens0r joined #git
02:06 matsaman you also don't usually want to pull into your working feature branch, really, IMO
02:06 matsaman and you do want to commit regularly, just like saving regularly
02:07 matsaman or, at the very least... before you tell git to do something else unrelated? =P
02:07 matsaman anyway, this is the kind of thing you would figure out in the first half hour of using git
02:07 mojave i'm starting to get it. somebody suggested it to me so i have better project organization.. but i've also been looking at this thing called SVN
02:07 matsaman and if it isn't, you're probably never going to figure out how to keep your data from disappearing
02:08 mojave it sounds like him and his team had been using it for awhile
02:08 mojave i don't know
02:08 austin987 joined #git
02:08 matsaman well, svn
02:08 matsaman is very old
02:08 matsaman and has a different approach
02:09 matsaman and honestly, I _have_ seen svn literally break
02:09 mojave oh.
02:09 matsaman while you could argue this persons misuse of git has cost him some data
02:09 inovas joined #git
02:09 matsaman I could argue that using svn _normally_ can cost you a lot of time
02:09 mojave sure
02:09 mojave oh, good to know.
02:09 matsaman I would imagine at this point the development, compared to things like git, has essentially halted
02:10 matsaman but the main reason to choose svn or git is down to the question of whether you want a "distributed" VCS, which almost everyone always does
02:10 howitdo joined #git
02:10 mojave i'm going to ask a dumb question, but it's because i don't know anything.. is something like FTP a viable alternative for this kind of thing?
02:10 matsaman handily, you can actually, if you want, use svn upstream and git on any or all downstreams
02:10 matsaman because git can interface with svn
02:10 finalbeta joined #git
02:10 matsaman no, ftp isn't a viable solution for anything, really
02:10 matsaman sane people don't even use it for file transfers anymore
02:10 mojave lol
02:10 mojave all right
02:10 matsaman they use ssh (or "sftp")
02:11 mojave i've been really out of the loop
02:11 matsaman the biggest things git has going for it is 1) Linux uses it and Linux is a beast that isn't going anywhere anytime soon, and so it has a certain code quality and community size & support, and 2) it's distributed
02:12 matsaman so take those into account, test a few other clients, do some trivial comparisons, and then bite a bullet and try one longer
02:12 mojave but you can still use svn on a linux machine right?
02:12 matsaman yup
02:12 mojave that's good.
02:12 matsaman they're both cross platform, AFAIk
02:12 matsaman although you'd naturally be silly to use them on something other than GNU/Linux =)
02:12 mojave i don't hate git as much as i did a couple hours ago when i was feeling some frustration.
02:13 mojave lol
02:13 mojave i have a windows machine running git.. but i switch back and forth
02:13 mangoleaf left #git
02:14 mangoleaf joined #git
02:14 d^sh joined #git
02:14 matsaman yeah I really hate git, but
02:15 matsaman I hate all its competitors I've used even more
02:15 matsaman the thing about git is there's a specific way you're meant to use it
02:15 matsaman and unfortunately you have to read about that, you can't just assume the git commands will steer you perfectly
02:15 mojave that's becoming apparent
02:15 mojave sure
02:15 matsaman and the commands are kind of imperfectly named
02:16 mojave so far if i'm just sticking with the main three commands and i don't try to get fancy about it.. it has been behaving
02:16 matsaman and some of the commands really would be better used with some extra opts, rather than as they do things by default, blah blah
02:16 matsaman yeah it's
02:16 matsaman I mean it definitely works
02:16 mojave that is true.. like "pull" does not intuitively do what i might think it does
02:16 matsaman but you have to learn those few fundamental commands, and their names aren't useful, you just have to learn them
02:17 matsaman http://rogerdudler.github.io/git-guide/ is fine for just getting going, IMO
02:17 matsaman as you compare what various people's published workflows are, you'll start to see a general pattern where almost everyone is doing almost the same thing
02:17 matsaman and that's how you use git =P
02:18 mojave oh yeah, that's the first guide i read through
02:18 mojave even though i resorted to more dangerous things like git push -u
02:18 matsaman clone, pull, branch, add, commit, merge, push, not so bad
02:19 matsaman yeah there are some sticky situations, but
02:19 matsaman since it is a VCS, it's job is to keep data (at least data you've committed)
02:19 kadoban Once you're to the point where you're wondering what workflow you should use, since git doesn't particularly enforce one, I usually recommend git-flow (you can google for), it's a decent place to start. It's a bit restrictive, but IMO that's actually kind of helpful at first. But first you have to kind of figure out what the hell any of the commands do and what the basic model of git is (which is ... better than the interface)
02:19 mojave why couldn't they just call "clone" "copy to" or something...?
02:19 matsaman so the worst thing you risk, usually, is embarrassment, from colleagues seeing you screwup
02:19 matsaman clone doesn't bother me, but
02:19 mojave kadoban, i will look into it
02:20 matsaman for example if you compare 'checkout' between git and svn, they don't compare at all
02:20 mojave matsaman, oh yeah, because your stuff is al lright there
02:20 mojave *all right
02:20 matsaman yeah, http://nvie.com/posts/a-successful-git-branching-model/ is definitely nice if you need something that complex
02:21 mojave matsaman, will read
02:22 matsaman mostly just more "interesting" than useful: https://www.google.com/trends/explore?date=all&q=%22svn%22,%22git%22
02:23 mojave whoa
02:23 kadoban I don't really see svn anymore; I definitely wouldn't recommend it. Though if it's solving your problems already ...
02:23 matsaman mojave: and that's part of what kadoban was mentioning, FYI
02:23 mojave for sure
02:24 matsaman I was fairly indifferent to svn vs git about a year ago, but I've actually seen the application screw up a few times now, it's annoying to fix
02:24 kadoban But it's pretty quick to learn enough git that it'll just be a slightly better version of the same model svn uses. And then you can add stuff later that's super useful without a big bunch of work.
02:24 matsaman and makes me wonder how else it's failing, and if it does so in destructive ways =P
02:24 matsaman yeah, and like I said, you can even use svn as your repo and git as your interface to an svn repo (rather than svn)
02:24 matsaman a lot of people stuck with a legacy svn repo do that
02:25 matsaman git has a much better branches model for development
02:25 matsaman and it can save you from having to learn both svn and git if you already use git for other things <shrug>
02:26 inovas joined #git
02:26 zzz joined #git
02:27 matsaman yeah, I sure would like to have the time and inclination to make something better than git =P
02:27 matsaman pity it's such a complex problem (in the implementation details, if nothing else)
02:28 kadoban There's probably quite a bit of room to write a better interface on top of git. The underlying model is great. The interface leaves something to be desired. There's projects out there that try to do that, but I haven't particularly tried any of them.
02:29 mojave and i hear people complaining about being stuck with git
02:29 matsaman yeah it's definitely the bits one directly interfaces with that I find the most imperfect
02:29 matsaman wellll
02:30 matsaman the thing is, if you're going to be a developer and not live on an island, you've got to learn well enough one or two of the most popular VCSes
02:30 kadoban mojave: There's two types of projects: 1) The ones nobody uses 2) The ones people complain about
02:30 matsaman regardless of all other things =)
02:30 matsaman and yeah, useful complaining utilizes comparisons
02:30 matsaman and/or replacement code/patches/etc.
02:30 mojave kadoban, a fair point
02:31 matsaman oh svn is apache licensed, no wonder it's dying
02:31 matsaman teehee
02:31 mojave lol
02:31 mojave i've heard quite a few folks just refer to SVN as "apache"
02:31 mojave like kleenex to tissue
02:31 matsaman oh really? weird
02:32 mojave yeah
02:32 matsaman everyone knows that "apache" is only the correct wrong name for apache's httpd =P
02:32 converge joined #git
02:32 mojave lol
02:32 matsaman mostly it's just old, and not incredibly maintained
02:33 matsaman otherwise I'm sure they would've just implemented a branching model similar to git's, and a distribution model similar
02:33 matsaman but nope
02:33 qqx I definitely wouldn't be considering svn in response to a horror story about problems with merging in git.
02:33 matsaman there's hardly ever a really great reason to not implement what people clearly like about your competitors' offerings
02:33 mojave qqx, yeah, that is what i am gathering, lol
02:33 matsaman macOS can get away with that to a certain extent, because they're a cult and the whole idea is to be one with the cult
02:33 tens0r joined #git
02:33 qqx At least with git if you're using it at all close to right, you'll have your changes committed before a merge is started. That's basically never done with svn.
02:33 matsaman svn is not macOS =)
02:34 kadoban SVN is explicitly the best implementation of CVS (a really old model) that could be done, that was the goal way back when. So I don't think changing the basic model would have made sense for svn really.
02:34 matsaman qqx: particularly if it's to do with having lost data you never told git to not lose
02:34 Jonno_FTW I have a detached state, how do I make this current state the new head?
02:34 mojave qqn, right, so it "just happens", right?
02:35 matsaman well I mean back when it was a business thing, collabnet?
02:35 * matsaman shrugs
02:36 matsaman wouldn't it suck to make something that lived forever, anyway? That would mean there wasn't any room for progress
02:36 matsaman that's the kind of thing that doesn't merge well with the human condition =P
02:36 svm_invictvs joined #git
02:36 kadoban Jonno_FTW: Not really clear what you want to happen. What's "the new head"?
02:37 mojave matsaman, well, nobody should "update" the game of chess
02:37 mojave it's fine the way it is
02:37 Jonno_FTW kadoban: nvm I fixed it by making a new branch
02:37 matsaman sure, but chess has a different goal
02:37 matsaman to be chess
02:37 mojave that is true
02:37 matsaman if its goal were to be the most challenging game there were, then it'd suck if it'd never been updated
02:37 matsaman but things aren't software, either
02:37 mojave but, like.. Dungeons and Dragons just wants to be Dungeons and Dragons, and it has some real design problems.
02:37 mojave I still love it, though
02:37 matsaman software is for solving problems, not being singular works of art
02:38 matsaman when you find a better way to solve a problem, you stop using antiquated tools; when you find another piece of art, you just have another piece of art
02:38 matsaman heh
02:38 watabou joined #git
02:38 matsaman people play dnd at my new job
02:39 mojave matsaman, that's cool
02:39 shgysk8zer0_ joined #git
02:39 WhiskerBiscuit joined #git
02:40 matsaman it's a pretty hilarious activity
02:41 mojave i have a deep love of role playing
02:41 mojave like, the tabletop kind, i mean
02:41 matsaman heh
02:46 matsaman good night all =)
02:48 ilbot3 joined #git
02:48 Topic for #git is now Welcome to #git, the place for git help and serious business | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | The git-jokes project has been suspended for lack of contribution. Help revive it! https://madeitwor.se/git-jokes
02:48 inflames joined #git
02:48 pks joined #git
02:52 howdoi joined #git
02:52 Sasazuka joined #git
02:52 heroux joined #git
02:52 darkbit joined #git
02:59 Sasazuka joined #git
02:59 aavrug joined #git
02:59 bruce_lee joined #git
03:09 derk0pf joined #git
03:17 str joined #git
03:18 Noodlewitt joined #git
03:20 Noodlewitt joined #git
03:23 Koognod joined #git
03:24 Goplat joined #git
03:24 ki0 joined #git
03:26 jimi_ joined #git
03:27 chachasmooth joined #git
03:27 bronson joined #git
03:28 tmk joined #git
03:29 jimi_ joined #git
03:31 jimi_ joined #git
03:32 sunri5e joined #git
03:33 strugee joined #git
03:36 Andrew_K joined #git
03:37 boombatower joined #git
03:38 breezy_ joined #git
03:39 tang^ joined #git
03:39 MrWoohoo joined #git
03:43 heroux joined #git
03:47 mizu_no_oto joined #git
03:48 watabou joined #git
03:50 duderonomy joined #git
03:51 Limes_ joined #git
03:51 saml joined #git
03:51 venmx joined #git
03:52 hexagoxel joined #git
03:53 sbulage joined #git
03:54 tmsmith joined #git
03:58 _rht joined #git
03:59 a_thakur joined #git
04:03 mischat joined #git
04:07 venmx joined #git
04:07 saml joined #git
04:12 fstd_ joined #git
04:14 Vortex35 joined #git
04:19 brewmaster joined #git
04:23 spacelord_ joined #git
04:29 aruns joined #git
04:30 breezy_ joined #git
04:34 alyptik joined #git
04:36 alyptik left #git
04:37 gfixler joined #git
04:40 durham joined #git
04:41 subhojit777 joined #git
04:41 bronson_ joined #git
04:42 moei joined #git
04:42 durham joined #git
04:43 durham joined #git
04:44 strugee joined #git
04:45 tterrag joined #git
04:45 Sasazuka joined #git
04:48 saml joined #git
04:56 svm_invictvs joined #git
05:00 aruns joined #git
05:02 mojave er, i have two branches in github, master and test. and for some reason i can't push into test, only master?
05:02 mojave it's saying 'error: src refspec test does not match any.'
05:04 mischat joined #git
05:05 JanC_ joined #git
05:05 Dragonbird joined #git
05:06 habbasi joined #git
05:10 m1m3-50 joined #git
05:10 mojave well, it seems i have a handle on it.. and faster than expected
05:11 mojave last problem took me over an hour to solve
05:11 mojave lol
05:11 Kamilion joined #git
05:12 mojave i know how to make and push to branches now. (yay)
05:12 mojave lol
05:13 yaewa joined #git
05:23 dan2k3k4 joined #git
05:25 PioneerAxon joined #git
05:25 ki0 joined #git
05:32 [0xAA] joined #git
05:37 howitdo joined #git
05:40 WhiskerBiscuit joined #git
05:43 bocaneri joined #git
05:45 pks_ joined #git
05:46 beatdown joined #git
05:47 Cabanossi joined #git
05:56 bilb_ono joined #git
05:56 bilb_ono how do I change the user I am pushing to a repo with?
05:56 _jan_ joined #git
05:57 bilb_ono I am getting this error: remote: Permission to AlexMarshall12/AlexMarshall12.github.io.git denied to alex-conduce. when I try and push a commit
05:57 bilb_ono I do git config user.name and it shows AlexMarshall12
05:57 bilb_ono but clearly it still thinks Im alex-conduce
06:00 breezy_ joined #git
06:01 esokia joined #git
06:05 grawity user.name has nothing to do with logins
06:05 d0nn1e joined #git
06:05 grawity it's the actual name you sign commits with
06:05 mischat joined #git
06:07 thiago bilb_ono: change the URL you're pushing to
06:07 thiago bilb_ono: run git remote -v to see it
06:07 thiago bilb_ono: git remote set-url <remotename> <new URL>
06:07 venmx joined #git
06:07 thiago if you're in the case of identity given by SSH key, edit ~/.ssh/config
06:08 bilb_ono thiago: git remote -v shows originhttps://github.com/AlexMarshall12/AlexMarshall12.github.io.git (push)
06:08 m1m3-50 joined #git
06:08 bilb_ono this is correct
06:09 bilb_ono I think the problem is that its saying its denied to alex-conduce. It think’s Im alex-conduce. I want to be pushing as AlexMarshall12
06:11 thiago change the username in the same place that the password is saved
06:11 tarkus joined #git
06:11 Derderderd joined #git
06:12 rscata joined #git
06:12 markus3 joined #git
06:12 bilb_ono thiago: Im not sure what you mean by the place the password is saved?
06:12 bilb_ono how would I find this?
06:12 bilb_ono is there a git config file somewhere?
06:13 bket joined #git
06:14 mrmadcat joined #git
06:15 _ikke_ bilb_ono: git config -l will list all config settings it knows about (global and for the current repo)
06:15 breezy_ joined #git
06:15 _ikke_ bilb_ono: But git also uses credential helpers
06:15 _ikke_ bilb_ono: If you set on up, you have check that specific credential helper
06:16 bilb_ono _ikke_: strange git config -l doesn’t mention anything about alex-conduce. I presume I didn’t set that specifically
06:17 bilb_ono is one of those variables the thing its referring to when It says “remote: Permission to AlexMarshall12/AlexMarshall12.github.io.git denied to alex-conduce”
06:17 ogny joined #git
06:18 _ikke_ bilb_ono: Does it mention anything about credential helpers?
06:18 _ikke_ the config
06:18 Hink joined #git
06:18 bilb_ono _ikke_: ah yes it says credential.helper=osxkeychain
06:18 _ikke_ there you go
06:18 _ikke_ Look at your osx keychain
06:20 bilb_ono _ikke_: thanks, worked perfect!
06:23 texasmynsted I am looking at git subtree split.  I see this in the man page " <branch> must not already exist" for the -b switch on git subtree split.  In practice it _seems_ like this is fine if the branch was used for a prior split of the same directory.  Is this true?
06:23 Derderderd texasmynsted: is twice i think
06:23 texasmynsted is twice?
06:24 Andrew_K|2 joined #git
06:25 Sample joined #git
06:26 Oatmeal joined #git
06:27 ShapeShifter499 joined #git
06:28 Dragonbird joined #git
06:29 govg joined #git
06:32 tmk joined #git
06:35 Derderderd what do you mean texasmynsted
06:35 tobiasBora joined #git
06:39 texasmynsted I mean the branch exists from a prior subtree split of the same directory.
06:39 texasmynsted It seems to be working fine
06:40 [Saint] joined #git
06:41 aidalgol joined #git
06:45 Derderderd texasmynsted: simple question - is twice
06:46 _August_ joined #git
06:47 texasmynsted I do not understand your "is twice" question.
06:49 Nitesh joined #git
06:50 zapb_ joined #git
06:51 _jan_ joined #git
06:51 qt-x joined #git
06:54 Takiito joined #git
06:58 dreiss joined #git
07:00 Raging_Hog joined #git
07:00 dermoth joined #git
07:04 Anja joined #git
07:06 mischat joined #git
07:10 ki0 joined #git
07:11 PioneerAxon joined #git
07:12 Alienpruts joined #git
07:15 svm_invictvs joined #git
07:15 subhojit777 joined #git
07:15 lb1d joined #git
07:16 JeroenT joined #git
07:20 mahendratech joined #git
07:22 inovas joined #git
07:23 romerocesar_ joined #git
07:23 mischat joined #git
07:23 freimatz joined #git
07:25 tango_ joined #git
07:25 FilipNortic joined #git
07:26 Raging_Hog joined #git
07:28 freimatz joined #git
07:30 sea-gull_ joined #git
07:30 lordjancso joined #git
07:31 sbulage joined #git
07:32 mindfart joined #git
07:35 zeroed joined #git
07:36 a_thakur joined #git
07:37 fullstack_ joined #git
07:38 gingitsune joined #git
07:38 _August_ joined #git
07:39 inovas joined #git
07:40 eduardas_m joined #git
07:40 fullstack_ joined #git
07:41 GodGinrai joined #git
07:43 fullstack_ joined #git
07:44 zeroed joined #git
07:44 zeroed joined #git
07:45 Orbitrix joined #git
07:45 warface joined #git
07:45 fullstack_ left #git
07:45 _August_ joined #git
07:46 ahmedelgabri joined #git
07:47 venmx joined #git
07:49 Sceorem joined #git
07:50 jnavila joined #git
07:50 Sample joined #git
07:50 jfr joined #git
07:52 Anja joined #git
07:54 theoceaniscool joined #git
07:55 romerocesar joined #git
07:57 JeroenT_ joined #git
08:02 ochorocho__ joined #git
08:03 noqnio1 joined #git
08:03 noqnio1 hey, i accidentally used git add on my virtual environment folder, if i git rm -r venv it will not delete the actual files... right?
08:04 a_thakur joined #git
08:04 selckin git status output tells you how to unstage
08:04 Sceorem joined #git
08:05 venmx joined #git
08:05 noqnio1 oh cool thanks
08:06 JeroenT joined #git
08:07 acetakwas joined #git
08:08 a_thakur_ joined #git
08:08 diogenese joined #git
08:08 gingitsune What excatly is neccesary to serve git over http
08:09 gingitsune I cloned a --bare repo to /var/www/git/repo_name
08:09 gingitsune I set post-update.sample to post-update
08:10 cyan__ joined #git
08:10 ws2k3 joined #git
08:11 ochorocho__ joined #git
08:12 govg joined #git
08:13 govg joined #git
08:13 gingitsune nginx conf, permisions of the served directory
08:13 gingitsune and chrome output
08:13 gingitsune http://hastebin.com/apokocehip.rb
08:14 _ikke_ you have to run git update-server-info (which the hook does after push)
08:14 selckin and +x the hook
08:14 JeroenT joined #git
08:14 _ikke_ selckin: I think the samples are +x already
08:14 danslo joined #git
08:14 gingitsune they are
08:15 JeroenT joined #git
08:15 selckin _ikke_: ok, it was in the docs
08:16 _ikke_ If you create new hooks, you need to do that
08:16 selckin https://git-scm.com/book/en/v2/Git-on-the-Server-The-Protocols
08:16 selckin just to make sure i guess
08:16 chele joined #git
08:16 rasto joined #git
08:16 derk0pf joined #git
08:17 howitdo joined #git
08:17 gingitsune sudo chmod -R ugo+x hooks/
08:17 gingitsune no dice
08:17 FreakySneaky joined #git
08:17 selckin did you do update-server-info ? no reply on that
08:18 gingitsune selckin: i don't see a message form you on that
08:18 selckin from _ikke_
08:20 gingitsune That did the trick
08:20 tnecniv joined #git
08:20 gingitsune Thank you both!
08:21 ayogi joined #git
08:23 elect joined #git
08:23 mablae joined #git
08:23 mischat joined #git
08:27 theoceaniscool joined #git
08:27 tmk joined #git
08:29 QuestionTime joined #git
08:32 ochorocho__ joined #git
08:37 netj joined #git
08:38 roxlu joined #git
08:39 metachr0n joined #git
08:40 PCatinean joined #git
08:41 andres joined #git
08:42 elastix joined #git
08:42 Balliad joined #git
08:43 diget joined #git
08:48 bernardio joined #git
08:49 encod3 joined #git
08:49 j416 joined #git
08:51 shams joined #git
08:51 GavinMagnus joined #git
08:51 PioneerAxon joined #git
08:52 ahmedelgabri joined #git
08:52 ki0 joined #git
08:53 bernardio_ joined #git
08:55 geoid_ joined #git
08:55 bernardio_ joined #git
08:55 TomyWork I'm trying to write an index-filter that removes everything except 4 specific directories from every commit of a repo. Now "echo !(dir1|dir2|dir3|dir4)" shows me all the directories except those 4, but "git rm --cached '!(dir1|dir2|dir3|dir4)'" tells me "fatal: pathspec '!(dir1|dir2|dir3|dir4)' did not match any files".
08:55 bernardio_ joined #git
08:56 selckin your shell is expanding that
08:56 TomyWork i know
08:56 selckin add thee '' in the echo
08:56 watabou joined #git
08:56 TomyWork i had assumed that git is at least as powerful as the shell when it comes to globbing :)
08:56 selckin bash zsh sh tsh dash?
08:56 TomyWork bash
08:57 bernardio joined #git
08:57 bernardio joined #git
08:57 TomyWork i tried "git rm --cached . && git reset dir1 dir2 dir3 dir4" as the index-filter, but that appears to have picked up the files from the latest commit every time...
08:58 TomyWork in other words, all my commits but the root commit have no file changes
08:58 TomyWork just commit messages
08:58 jfr joined #git
08:58 bernardio joined #git
08:59 TomyWork More precisely, i tried "git rm --cached -qrf . && git reset -q -- dir1 dir2 dir3 dir4" as the index-filter
08:59 bernardio joined #git
09:01 mkoskar joined #git
09:01 selckin does index filter not start with the changes stages, can't remember trying to lookup
09:01 willlondon joined #git
09:01 j416 joined #git
09:01 TomyWork i dont know. i assumed it would
09:02 selckin then you would need to reset everything and git add dir dir2 no?
09:02 TomyWork git add in an index filter?
09:02 leosw joined #git
09:02 leosw Hello world !
09:03 bongjovi joined #git
09:03 leosw I'm using gogs.io to manage my git server.
09:03 metachr0n i have been tasked with migration of document resources from sharepoint to gitwiki ... i have found pandocs very helpful ... however does anyone have any suggestions for quickly taking existing directory structure ( containing docs ) and to just rip it to markdown
09:03 ToBeFree joined #git
09:03 leosw Is there a way to recover git repositories since I only have raw data like in this example : http://pastebin.com/peePDsxA
09:03 aw1 joined #git
09:03 TomyWork leosw good for you, but the specifics of that server are best answered by people actually using that server :)
09:04 cdown joined #git
09:04 tobiasvl leosw: that looks like the git repository…
09:04 TomyWork leosw that looks like temporaries for a download link
09:04 aruns joined #git
09:04 TomyWork the archives stuff i mean
09:04 tobiasvl looks like the actual git repo too, plus lots of archives
09:04 TomyWork you can probably remove the archives directory and the rest is your git repo
09:05 TomyWork also pastebin is terrible for ascii art
09:05 selckin metachr0n: doesn't sound like a git question
09:05 TomyWork metachr0n try #sharepoint or #gitwiki
09:05 TomyWork or #pandocs
09:05 venmx joined #git
09:06 leosw tobiasvl: You're right. Is there a way to make a git server with only that data
09:06 TomyWork (i have no idea what either of these are... i think i heard sharepoint in a microsoft context before though)
09:06 leosw I do not need an interface in the first step
09:06 selckin leosw: it is the git repo, all the data you need, you can run git status in it or clone it
09:06 GavinMagnus left #git
09:06 tobiasvl leosw: a git server? sure, you don't need anything else than what you have there
09:06 jknetl joined #git
09:06 bernardio joined #git
09:06 tobiasvl do anything you want with it
09:06 TomyWork leosw what do you want to do with it?
09:07 TomyWork push it to your gogs.io server?
09:07 leosw i just want to make back my server to get my data back to my computers.
09:07 TomyWork leosw what do you mean by that, exactly?
09:08 TomyWork you exported that dump from gogs.io?
09:08 leosw The problem with gogs.io (I know it's not git related) is that Gogs doesn't seems to be able to parse the git repo to add the repo to the web interface
09:08 overlord_tm joined #git
09:08 TomyWork leosw ok so you ARE moving to gogs.io
09:08 TomyWork right?
09:08 TomyWork be specific, then highlight me
09:08 leosw TomyWork: No, gogs.io failed one day, and I only stores this data, hoping it contains all my stuff.
09:09 leosw So my first question is to be sure I stores everything important.
09:09 netj joined #git
09:09 leosw Than, I want to put that back to a fresh install of Gog.io (I asked theyre team since 2 weeks ;) )
09:09 TomyWork leosw ok let's assume you are moving to github. (it works the same way with a different url for other sites)
09:09 selckin leosw: go in one of those dirs and run git log
09:09 tobiasvl leosw: it contains all the git stuff, it looks like (although only you can verify that), but no idea what gogs needs in addition
09:09 TomyWork cd binindexexporter.git && git push git@github.com:leosw/binindexexporter.git --mirror
09:10 TomyWork leosw adjust the push url, then enter that and it'll push all the branches and tags to your repo
09:10 leosw Ok TomyWork, I'll try to do that, but that will be long to do that for all repos.
09:11 TomyWork leosw no problem, you can loop through it:
09:11 bernardio joined #git
09:11 nickthetait joined #git
09:11 TomyWork for repo in *.git; do cd "$repo" && git push git@github.com:leosw/"$repo" --mirror; done
09:11 leosw (I have a structure of organization and repositories.
09:11 TomyWork just make sure you create all the repos on github first :)
09:12 leosw Ok, That's a way, I'll try that ;)
09:12 TomyWork try it with one repo first
09:13 bernardio_ joined #git
09:13 TomyWork selckin so do you have an idea how to solve that index filter problem?
09:14 Flowinho joined #git
09:15 djb-irc joined #git
09:16 systemovich joined #git
09:16 selckin from quick test i believe my last thing was correct
09:17 marcogmonteiro joined #git
09:17 selckin hmm or only with modifications
09:17 selckin should still work
09:18 clmsy joined #git
09:18 TomyWork well i can't run "git add" in an index-filter since it will pick up the files in my work tree
09:18 selckin get a clean work tree
09:19 somebody joined #git
09:19 TomyWork index-filter doesnt check out the worktree
09:19 TomyWork that's kind of the point
09:19 osse TomyWork: you forgot to cd out again. I suggest git -C "$repo"
09:19 TomyWork osse oops, i actually forgot the ()
09:19 kurkale6ka joined #git
09:20 TomyWork leosw for repo in *.git; do ( cd "$repo" && git push git@github.com:leosw/"$repo" --mirror ); done
09:20 TomyWork osse easier to see what's going on this way :)
09:21 derk0pf joined #git
09:21 osse i guess
09:21 jaguarmagenta joined #git
09:21 selckin TomyWork: i don't really know how it works in detail, why i didn't want to awnser, but if i git status in filter, the changes are already staged, so if you rm everything, then reset what you want to keep, the things you want to keep aren't staged anymore right
09:22 TomyWork ah i see
09:24 nettoweb joined #git
09:25 selckin you seem good with shell scripting, so maybe you want to filter on bash -c 'git rm yourglboal' and get the quotes right
09:26 TomyWork hmm i could figure out a way to list the files in the index, then subtract the 4 dirs from that and put that into rm --cached
09:26 manuels joined #git
09:27 selckin git status | grep ': '  | cut -d ':' -f1- | grep -v dir (not tested)
09:27 reuf_ joined #git
09:27 TomyWork yipes
09:27 TomyWork i was more thinking along the lines of git ls-files or what it's called
09:27 selckin your original glob should also work if you let bash run it with bsah -c
09:27 selckin but then you need to figure out the magic qutoes you need too
09:28 selckin ah no, it won't work for same reason my add won't work i'm stupid
09:28 selckin =)
09:28 TomyWork the problem is there were directories added and removed throughout history
09:29 selckin parsing status should work then no?
09:29 selckin put it in a bash script and call that, easier with the quotes
09:29 TomyWork too fragile for my taste
09:30 TomyWork i'd rather go to plumbing
09:30 osse git rm --cached -- . :!dir1 :!dir2
09:30 selckin one time runs you will manually verify anyway, but sure
09:31 TomyWork did i mention it's 40000 commits? :)
09:31 TomyWork osse thank you!
09:31 selckin the hero arrives
09:32 bernardio_ joined #git
09:32 osse test it before you sing my praises :p
09:32 TomyWork "fatal: pathspec ':dir2' did not match any files"
09:32 TomyWork oh nm
09:33 TomyWork --ignore-unmatch
09:33 osse also the ! mysteriously disappeared
09:33 TomyWork oh
09:34 TomyWork yep, with the ! it works better :)
09:34 djb-irc joined #git
09:35 * TomyWork sings praises
09:36 paul424 joined #git
09:36 Tucky joined #git
09:36 synthmeat joined #git
09:36 TomyWork also twice as fast as my previous index-filter :)
09:37 * osse takes a bow
09:38 * selckin throws roses
09:38 sunri5e_ joined #git
09:39 cdown joined #git
09:40 bernardio joined #git
09:46 chll_ joined #git
09:47 ochorocho__ joined #git
09:49 dwts joined #git
09:49 cmalmber joined #git
09:50 metachr0n TomyWork: i've already written a script to do it myself ... its quicker than asking for help
09:51 frobnic joined #git
09:51 leosw left #git
09:51 metachr0n TomyWork: but thanks for not dissapointing
09:52 * selckin slaps metachr0n to another dimenision
09:54 manuel_ joined #git
09:55 * metachr0n weeps ... sitting solitarily in the darkness of the void with no hope of escape
09:56 PaulCapestany joined #git
09:56 jast that sounds a little dreary
09:57 watabou joined #git
09:57 * metachr0n suddenly hears bells ringing ... and to his surprise its santa claus with a bottle of coka cola
09:59 sunri5e joined #git
09:59 TomyWork run away!
09:59 * metachr0n realizes it was just a sick prank by selckin ( the evil general of TommyWork ) both archons of the highest class with magic far exceeding his own
09:59 jast run away, turn away, run away, turn away, run away...
10:00 TomyWork http://vignette4.wikia.nocookie.net/en.futurama/images/e/ed/Santa_bot.JPG/revision/latest?cb=20071101205312
10:00 metachr0n :)
10:00 rahtgaz joined #git
10:01 metachr0n XD
10:01 jast btw migrating from sharepoint to gitwiki? that's an... interesting switch...
10:01 govg joined #git
10:02 jozwior__ joined #git
10:02 Sample joined #git
10:02 metachr0n well ... its a long story ... but lets say that the company norm is sharepoint
10:03 metachr0n but our team can do pretty much what we want ... and we want everything in git
10:03 TomyWork so you're versioning the docs alongside the program?
10:03 Serpent7776 joined #git
10:04 metachr0n well its more everything from docs, user guides, general informaion, and of course programs ... often not interrelated
10:04 metachr0n network diagrams ... etc
10:05 metachr0n i've used pandoc to get rid of all the .docx crap and now we hve a more searchable goodness
10:05 metachr0n i don't have microsoft ... they have some good products
10:05 metachr0n but i just don't think many of them are great
10:05 metachr0n brb
10:08 overlord_tm joined #git
10:09 geoid_ joined #git
10:11 howitdo joined #git
10:11 TomyWork osse do you happen to know a way of filtering out commits that are empty but werent?
10:12 osse TomyWork: --prune-empty
10:15 TomyWork osse doesnt that remove commits that were empty before the rewrite as well?
10:17 osse TomyWork: dunno
10:18 DaveTaboola joined #git
10:19 pbrewczynski joined #git
10:20 achadwick joined #git
10:24 _August_ joined #git
10:24 nukeu666 joined #git
10:25 nukeu666 If i want to redirect my url to github pages, what sort of a dns entry would i need?
10:25 harwood joined #git
10:25 selckin !github
10:25 gitinfo Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
10:26 * nukeu666 whooshes away
10:26 selckin https://help.github.com/articles/using-a-custom-domain-with-github-pages/
10:33 QuestionTime joined #git
10:33 _jan__ joined #git
10:33 venmx joined #git
10:33 ChiLL_SurF joined #git
10:34 _jan__ joined #git
10:36 _August_ joined #git
10:36 GT4066 joined #git
10:37 MaryH joined #git
10:40 QuestionTime joined #git
10:40 QuestionTime joined #git
10:41 QuestionTime joined #git
10:47 encod3_ joined #git
10:52 encod3 joined #git
10:52 govg joined #git
10:53 alex1a joined #git
10:53 davimore_ joined #git
10:53 Sircle_ left #git
10:54 encod3_ joined #git
10:54 mar77i joined #git
10:54 derk0pf joined #git
10:55 paul424 joined #git
10:56 Nicmavr joined #git
10:57 yazdmich joined #git
10:57 tcsc joined #git
10:57 watabou joined #git
11:03 GT4066 joined #git
11:04 marscher joined #git
11:04 dut joined #git
11:05 ThomasTang joined #git
11:06 dut Hello. I'm sorry, can someone help with git? I can't run git add --patch or --interactive, getting error http://pastebin.com/9qBW8Pyd
11:07 selckin you shell is fucking something up
11:07 selckin try : 'commoand git add -p'
11:07 dut Oh. I see. Thank you
11:07 selckin try : 'command git add -p'
11:09 dut Same. I'll try to find something about shell
11:09 Silox| joined #git
11:10 selckin the error dropped the space, so your shell is doing that before starting git
11:10 cebor_ joined #git
11:10 dut selckin: Thank you! I'll try to investigate it.
11:11 ams__ joined #git
11:12 selckin what does "which git" output
11:12 finalbeta joined #git
11:12 osse if bash, that won't reveal much.  try "type git"
11:12 aw1 joined #git
11:14 dut git is hashed (/usr/bin/git) and git: /usr/bin/git /usr/share/git /usr/share/man/man1/git.1
11:14 King_Hual joined #git
11:14 dut if it's distro issue, i'll ask maintainers then
11:14 dut thank you :3
11:15 selckin wierd
11:15 buzzLightBeer joined #git
11:16 ThomasLocke joined #git
11:18 mda1 joined #git
11:19 selckin bash --noprofile --norc -c "git add -p"
11:19 selckin dut ^ does that work?
11:20 dut no, but i'm sorry. It seems i've find it out. it was perl issue (oh my)
11:21 osse selckin: FYI bash -c doesn't read .profile or .bashrc anyway
11:21 osse but I agree with the investigation nonetheless
11:21 osse ah yes, the interactive part is a perl script
11:21 dut i'm very sorry about bothering you
11:21 * osse shakes fist at git-send-email
11:22 selckin osse: thanks
11:22 dersand joined #git
11:25 JuliusN a developer did a bad merge to our dev branch. we want to revert the dev state to what it was before. what are our options?
11:26 mischat joined #git
11:26 osse JuliusN: as in any such sitation: overwrite history, or make a new commit that undoes the effect of the merge.
11:26 osse JuliusN: how many other developers have pulled ?
11:27 JuliusN everybody i think
11:27 JuliusN but i have an untouched feature branch
11:31 selckin reset --hard to the prev commit and force push, and deal with everyone having a hard time updating
11:32 selckin !undo_merge
11:32 selckin !_merge
11:32 gitinfo [!evil_merge] What you describe _can_ be done, but results in an "evil merge": a commit that purports to be a merge, but does not actually take all changes from all sides. This violates the assumptions that tools and humans make about merges, and usually results in confusion and pain down the road.
11:32 selckin pff
11:32 selckin !revert_merge
11:32 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
11:33 byte512 joined #git
11:34 venmx joined #git
11:37 JuliusN so to make it less painful for other developers an optimal way of undoing an evil merge is to just fix the code?
11:37 wyt joined #git
11:37 osse yes
11:37 JuliusN makes sens
11:37 JuliusN e
11:38 JuliusN git branch -r
11:38 JuliusN whoops
11:41 feodaron joined #git
11:43 WayToDoor joined #git
11:43 _August_ joined #git
11:43 freimatz joined #git
11:47 converge joined #git
11:49 tango_ 33
11:50 OOPMan joined #git
11:54 govg joined #git
11:55 rt joined #git
11:55 diogenese joined #git
11:57 romerocesar joined #git
11:58 watabou joined #git
11:59 jaguarmagenta joined #git
12:01 Sample joined #git
12:02 tvw joined #git
12:02 spacelord_ joined #git
12:04 feodaron joined #git
12:05 ahmed_elgabri joined #git
12:05 Dragonbird joined #git
12:07 Outsider___ joined #git
12:07 hussam joined #git
12:07 noahsussman_ joined #git
12:07 howdoi_ joined #git
12:07 pbrewczynski_ joined #git
12:07 ams___ joined #git
12:07 evanc joined #git
12:07 demize- joined #git
12:07 sunri5e joined #git
12:08 Anthaas joined #git
12:08 navidr_ joined #git
12:08 MLM_ joined #git
12:08 wcpan joined #git
12:09 misha joined #git
12:09 Gonzo89 joined #git
12:09 shauna joined #git
12:10 kenoob joined #git
12:10 kenoob joined #git
12:12 ghostlight joined #git
12:13 Singmyr joined #git
12:13 WayToDoor joined #git
12:13 ski7777 joined #git
12:13 matthiaskrgr joined #git
12:13 krillr joined #git
12:13 hexagoxel joined #git
12:14 jeffreylevesque joined #git
12:17 Klumben joined #git
12:17 sdothum joined #git
12:21 timlyo[m] joined #git
12:21 Dhanya[m] joined #git
12:21 saintaquinas[m] joined #git
12:22 norotartagen joined #git
12:22 patrickr[m] joined #git
12:22 _flow__ joined #git
12:22 buffal0 joined #git
12:23 _Cyclone_ joined #git
12:24 dsdeiz joined #git
12:24 daurnimator joined #git
12:24 cyan__ joined #git
12:24 flaviodesousa joined #git
12:25 jaguarmagenta joined #git
12:25 _August_ joined #git
12:27 dwts hello guys, quick question: on git version 1.7.12.4, when running git -rev parse HEAD returns just a commit hash. on git version 2.11.0 it retuns the branch name too. Is there a way to get the same output with 1.7.12.4?
12:27 dwts commit hash+branch name that is with git version 1.7.12.4
12:29 shams joined #git
12:30 selckin can't reproduce
12:30 _ikke_ git rev-parse just returns a hash by default
12:31 Masber joined #git
12:31 dwts _ikke_: yes, but with passing HEAD will also give the branch name
12:32 dwts hash + branch name
12:32 shams joined #git
12:32 YuGiOhJCJ joined #git
12:35 venmx joined #git
12:38 jknetl joined #git
12:43 hramrach joined #git
12:44 hramrach hello
12:44 gitinfo hramrach: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
12:45 gingitsune I'm still trying to setpu a dumb http setup
12:45 hramrach I cannot pull: http://paste.debian.net/902304/
12:45 hramrach WTF?
12:45 gingitsune It worked on my laptop but its not working out on our git server
12:46 hramrach btw I don't care about either of the branches mentioned in the error
12:47 hramrach I don;t have them checked out, don't have a local branch tracking them, nothing
12:47 osse hramrach: try git update-ref -d refes/remotes/blabla and then pull again
12:47 osse hramrach: i've had this happen on case insensitive file systems
12:47 Ozymandy joined #git
12:47 jast git can't have 'foo' and 'foo/bar' branches at the same time. in this case, presumably someone created 'foo', later deleted it and even later created 'foo/bar'
12:48 jast and you've got remnants of 'foo' still because git doesn't automatically prune its mirrors
12:48 vio4 joined #git
12:48 jast the straightforward way to fix this is to run 'git remote prune origin'
12:48 f3r70rr35f joined #git
12:49 gingitsune Figured it out
12:49 grawity I think `git fetch --prune` just does both automatically
12:49 osse git config --global fetch.prune true \o/
12:53 LordRyan joined #git
12:55 ojdo joined #git
12:57 hramrach yes, pruning origin fixed the problem. Thanks
12:58 Jackneill joined #git
12:58 DolpheenDream joined #git
12:58 hramrach It is a repo with especially hairy branch tree
12:59 rnsanchez joined #git
12:59 johnmilton joined #git
12:59 watabou joined #git
13:00 zzz joined #git
13:01 rafalcpp When you have multiple branches and multiple repositiores, how to see/visualize/list any "lost" commits compared to given git version?
13:02 rafalcpp e.g. I have mainrepo/branch version that I hope includes in it all work that existed on any other repo/branch. How to confirm
13:02 osse rafalcpp: git merge-base --is-ancestor otherrepo/branch mainrepo/branch
13:05 cqi joined #git
13:05 m1m3-50 joined #git
13:06 cqi joined #git
13:06 converge_ joined #git
13:07 cdown joined #git
13:08 synthroid joined #git
13:08 mattcen joined #git
13:09 sunri5e_ joined #git
13:09 manuel_ joined #git
13:09 frobnic_ joined #git
13:11 systemovich joined #git
13:12 qqx rafalcpp: git branch -a --no-merged
13:12 rafalcpp hanks
13:12 rafalcpp thanks
13:13 systemovich joined #git
13:14 systemovich joined #git
13:16 cqi joined #git
13:16 GodGinrai joined #git
13:17 shinnya joined #git
13:18 mischat joined #git
13:21 mischat joined #git
13:21 govg joined #git
13:28 _ikke_ dwts: Not for me
13:29 m1m3-50 joined #git
13:29 _ikke_ git rev-parse HEAD -> d2da68a14ada34b7842a51c88751ffe4aaf8a1af
13:30 mischat_ joined #git
13:31 elastix joined #git
13:32 eduardas_m joined #git
13:32 [0xAA] joined #git
13:34 howdoi joined #git
13:36 mischat joined #git
13:36 venmx joined #git
13:37 synthroi_ joined #git
13:39 tchan joined #git
13:42 b3kN joined #git
13:50 perlpilot joined #git
13:51 tmsmith joined #git
13:52 eduardas_m joined #git
13:53 ahmedelgabri joined #git
13:57 Henkie--drop-tab joined #git
13:59 joshsyn joined #git
14:00 watabou joined #git
14:01 jeffreylevesque joined #git
14:02 rnsanchez_ joined #git
14:04 skelterjohn joined #git
14:04 Iacobus__ joined #git
14:04 jaguarmagenta joined #git
14:04 mehdi___ joined #git
14:05 twisted` joined #git
14:08 cdg joined #git
14:10 cdg_ joined #git
14:11 Gloomy joined #git
14:14 valize joined #git
14:17 sbulage joined #git
14:19 CEnnis91 joined #git
14:21 ExoUNX joined #git
14:21 theoceaniscool joined #git
14:22 mda joined #git
14:23 jimi_ joined #git
14:23 eycsigfy joined #git
14:24 giraffe joined #git
14:25 lordjancso joined #git
14:26 buzzLightBeer joined #git
14:27 ahmedelgabri joined #git
14:28 GavinMagnus joined #git
14:28 nicksloan joined #git
14:33 ShekharReddy joined #git
14:34 noqnio1 joined #git
14:34 GodGinrai joined #git
14:35 shgysk8zer0_ joined #git
14:37 aw1 joined #git
14:37 venmx joined #git
14:41 AfroThundr joined #git
14:44 nanga joined #git
14:48 shgysk8zer0_ joined #git
14:49 GodGinrai joined #git
14:49 Gsham joined #git
14:52 denisMone joined #git
14:54 frobnic joined #git
14:56 nickabbey joined #git
14:56 cdown joined #git
14:57 minikN joined #git
15:00 anonymuse joined #git
15:01 BrianJ joined #git
15:02 Es0teric joined #git
15:07 Es0teric joined #git
15:07 GavinMagnus joined #git
15:08 Dragonbird joined #git
15:08 e14 joined #git
15:11 JuliusN how do i get the total amount of branches on a repository?
15:12 finalbeta joined #git
15:12 rwb joined #git
15:12 Darren_ joined #git
15:14 fahadash joined #git
15:16 guampa joined #git
15:16 boombatower joined #git
15:18 jaafar joined #git
15:19 davidosomething joined #git
15:19 GodGinrai JuliusN: `git branch -a | wc -l`
15:19 jaguarmagenta joined #git
15:23 mda1 joined #git
15:24 bejeb joined #git
15:25 jedahan joined #git
15:25 Guest30569 joined #git
15:26 idmistir joined #git
15:26 wjp1738 joined #git
15:27 Es0teric joined #git
15:28 jedahan How do I set '--color-words' or '--word-diff=color' in ~/.gitconfig to be a default option for git diff and git show without creating a new alias for those commands?
15:28 shgysk8zer0_ joined #git
15:29 nettoweb joined #git
15:29 nickabbey joined #git
15:30 unbalanced joined #git
15:30 Noodlewitt joined #git
15:31 AaronMT joined #git
15:31 cjpls_ joined #git
15:31 setuid jedahan: https://gist.github.com/vdemeester/bd17bdd2e87b87efb106
15:31 setuid That should help
15:32 mindfart joined #git
15:33 _Vi joined #git
15:34 jedahan setuid thanks but still doesn't work for me :(
15:34 encod3 joined #git
15:35 subhojit777 joined #git
15:36 zivester joined #git
15:36 setuid Are you sure you're properly sourcing it?
15:36 setuid IOW, is anything at all in ~/.gitconfig being enabled?
15:37 GavinMagnus left #git
15:38 venmx joined #git
15:38 Sample joined #git
15:42 qqx setuid: I don't see anything in that gist that would answer jedahan's question. Only thing I see about word diff is an alias, and the question excluded that.
15:42 qqx I think the real answer is that you can't.
15:42 chardan joined #git
15:42 jedahan this smells like my first pull request to git :)
15:43 jedahan seems like it would be simple and uncontentious
15:43 setuid http://git.661346.n2.nabble.com/Configuring-diff-color-words-using-git-config-td6426306.html
15:44 setuid What if you used: git config --global color.diff auto
15:44 setuid or git config --global color.diff always
15:44 nickabbey joined #git
15:44 jedahan yeah i set them to always, no dice
15:44 jedahan i want git diff to do what your git wdiff does
15:45 dimi1947 joined #git
15:45 texasmynsted If I used git submodule add with the -b <branch> option and I now want to change to a different branch, do I re-add or is there a way to "git submodule update" it to a different branch?
15:45 selckin can maybe hack something with using external diff program
15:45 xall joined #git
15:47 serialoverflow joined #git
15:48 netcarver joined #git
15:48 robotroll joined #git
15:49 grawity texasmynsted: edit .gitmodules
15:51 jnavila joined #git
15:51 texasmynsted So just edit .gitmodules then update?  Doesn't init write to .git/config then changes to .gitmodules are ignored? (is this true)
15:54 Darcidride joined #git
15:55 netj joined #git
15:55 BrianJ joined #git
15:57 Guest50076 joined #git
15:57 Gsham joined #git
15:58 romerocesar joined #git
16:01 leeN joined #git
16:01 JamesBaxter joined #git
16:08 watabou joined #git
16:08 blackwind_123 joined #git
16:08 al-damiri joined #git
16:08 nfk joined #git
16:11 eduardas_ joined #git
16:13 jnavila joined #git
16:14 kumarsuraj joined #git
16:18 LionsMane joined #git
16:20 netcarver left #git
16:20 anonymuse joined #git
16:21 g5pw joined #git
16:24 Dragonbird joined #git
16:27 King_Hual joined #git
16:27 kier joined #git
16:27 romerocesar joined #git
16:29 King_Hual joined #git
16:31 hardenedapple joined #git
16:32 overlord_tm joined #git
16:33 King_Hual` joined #git
16:34 mahendratech joined #git
16:36 tang^ joined #git
16:39 Shashikant86 joined #git
16:39 venmx joined #git
16:40 converge joined #git
16:40 converge joined #git
16:44 Dougie187 joined #git
16:45 kadoban joined #git
16:45 jccn joined #git
16:45 Jackneill_ joined #git
16:48 travelion joined #git
16:50 elastix joined #git
16:50 dreiss joined #git
16:51 Dragonbird joined #git
16:51 dviola joined #git
16:51 troyt joined #git
16:52 nickabbey joined #git
16:53 Dragonbird joined #git
16:54 Domhack joined #git
16:57 Dragonbird joined #git
17:00 Orbitrix joined #git
17:00 romerocesar_ joined #git
17:00 madewokherd joined #git
17:00 breezy_ joined #git
17:00 netj joined #git
17:01 Dragonbird joined #git
17:01 shams joined #git
17:02 dreampwnzor joined #git
17:03 travelion left #git
17:04 synthmeat joined #git
17:04 multi_io joined #git
17:05 multi_io can git log --pretty:format output path names? There's no % for it apparently?
17:05 Dragonbird joined #git
17:05 multi_io this would only make sense for log that track a single file, but that's what I have
17:05 multi_io this would only make sense for logs that track a single file, but that's what I have
17:07 jor joined #git
17:08 ahmedelgabri joined #git
17:09 austin987 joined #git
17:09 jdiez joined #git
17:09 CtrlC joined #git
17:09 tabakhase joined #git
17:10 Dragonbird joined #git
17:11 howitdo joined #git
17:11 nickabbey joined #git
17:12 Smiv joined #git
17:14 penanze joined #git
17:16 afuentes joined #git
17:17 howitdo joined #git
17:17 Dragonbird joined #git
17:18 raspado joined #git
17:18 dimi1979 joined #git
17:18 ExeciN joined #git
17:18 menip joined #git
17:19 anonymuse joined #git
17:19 Derderderd joined #git
17:20 thiago joined #git
17:20 jaguarmagenta joined #git
17:20 mischat joined #git
17:20 Dragonbird joined #git
17:21 e14 joined #git
17:21 mischat_ joined #git
17:23 nickabbey joined #git
17:24 irqq joined #git
17:24 danslo joined #git
17:26 Dragonbird joined #git
17:30 durham joined #git
17:33 metachr0n joined #git
17:33 thiago joined #git
17:35 denisMone joined #git
17:39 cdown_ joined #git
17:40 venmx joined #git
17:40 systemov` joined #git
17:40 orb258 joined #git
17:40 tonyj joined #git
17:40 dwts _ikke_: I found a solution, but I'm currently not at working laptop
17:41 johnmilton joined #git
17:42 blunaxela joined #git
17:44 svm_invictvs joined #git
17:44 orb258 is there any way to run git-http-backend as fastcgi instead of cgi on windows?
17:45 orb258 nginx doesn't support cgi, only fastcgi, and i haven't found any wrappers for fastcgi to cgi
17:45 synthroid joined #git
17:46 SwingShock joined #git
17:46 alex1a joined #git
17:46 robotroll joined #git
17:46 canton7 orb258, there are loads of sample configurations online for running git-http-backend behind nginx
17:46 synthroid joined #git
17:47 orb258 canton7: they use fcgiwrap which is linux only, nginx doesn't natively support cgi as it is old and outdated
17:47 sea-gull_ joined #git
17:48 acetakwas joined #git
17:49 rominronin joined #git
17:51 kazantip joined #git
17:51 CheckDavid joined #git
17:53 dreiss joined #git
17:54 cdown joined #git
17:54 Electrometro joined #git
17:55 tang^ joined #git
17:56 Balliad joined #git
17:57 ResidentBiscuit joined #git
17:57 synthroi_ joined #git
17:57 ResidentBiscuit joined #git
17:58 ResidentBiscuit joined #git
17:58 nickabbey joined #git
17:59 nickabbey joined #git
17:59 orb258 alright so i'm guessing nobody has an answer, whoever is maintaining the code for git-http-backend, i suggest they convert it to fastcgi as nobody uses cgi anymore
17:59 mischat joined #git
18:00 nidr0x joined #git
18:00 matsaman joined #git
18:01 ReptilianBrain joined #git
18:01 dsdeiz joined #git
18:02 nukeu666 joined #git
18:02 vdl joined #git
18:04 Dougie187 joined #git
18:04 e14 joined #git
18:07 gingitsune joined #git
18:09 xall joined #git
18:09 kier joined #git
18:10 romerocesar joined #git
18:10 leeN joined #git
18:11 synthroid joined #git
18:11 Sample joined #git
18:12 neilthereildeil joined #git
18:12 neilthereildeil hey guys
18:12 neilthereildeil i just git pulled from a remote
18:13 neilthereildeil now git status says "HEAD detached from 01a55e4"
18:13 neilthereildeil doesnt pull automatically make HEAD point to the tip of the branch?
18:14 neilthereildeil "git log --all --tags --decorate=full" doesnt show any special tags or refs at 01a55e4
18:14 neilthereildeil why is it using 01a55e4 as a reference based on which to give me errors?
18:15 neilthereildeil why is it expecting the head to be attached there?
18:15 kadoban neilthereildeil: What happened in between when you ran 'git pull' and the 'git status' ?
18:16 neilthereildeil nothing. it was the very next command
18:16 kadoban neilthereildeil: No output, no nothing? What did 'git pull' say?
18:16 matsaman I've gotten that error before, after a git reset
18:17 ahmedelgabri joined #git
18:17 neilthereildeil however, git status BEFORE running pull gave the same message "HEAD detached from 01a55e4"
18:17 neilthereildeil yea i did git reset on this repo yesterday
18:17 mischat joined #git
18:17 neilthereildeil --hard
18:17 kadoban neilthereildeil: So then the pull is pretty much unrelated?
18:18 neilthereildeil yesterday's pull had a commit that i didnt want
18:18 kadoban Detached HEAD just means that you're not on a branch
18:18 kadoban There's a decent number of ways to get there.
18:18 watabou joined #git
18:18 neilthereildeil so i reset --hard to the commit before what i pulled. but after talking to the other engineers, i understood why they did what they did, and i just re-pulled
18:19 thiago reset --hard does not detach HEAD
18:19 thiago it was something else
18:19 thiago did you try checkout or rebase?
18:19 neilthereildeil never done rebase
18:19 neilthereildeil i can try to checkout
18:19 kadoban So most likely you should checkout a branch? It's unclear how you've gotten where you are though. You could check 'gitk --all' and see what your current history looks like.
18:19 neilthereildeil but we dont know how i got into this state/
18:19 kadoban Probably you did a checkout when you meant to do a reset or something
18:20 neilthereildeil ok ill checkout and try to check status before pulling next time. thanks
18:22 synthroi_ joined #git
18:23 matsaman hum, is it ordinary that gitk would call itself 'Wish' in its menu? I'm stuck on macOS ATM, if that matters
18:24 kadoban I think 'wish' is some gui library, likely it's what is used by gitk. I don't recall it being evident in the UI, but I don't know that I really looked either.
18:25 matsaman too curious, will have to check on a proper OS
18:27 synthroid joined #git
18:27 manuel_ joined #git
18:27 converge joined #git
18:27 TomyLobo joined #git
18:28 qqx gitk is implemented as a TCL/TK script. `wish` is the name of the interpreter for that.
18:30 sawk joined #git
18:31 QuestionTime joined #git
18:31 ojacobson NAME
18:31 ojacobson wish — Simple windowing shell
18:34 peeps[lappy] joined #git
18:34 matsaman bah, sensibly, I hadn't even built in GTK+ support over here over here
18:34 geoid joined #git
18:40 matsaman kadoban: says Wish for you on GNU/Linux?
18:40 dreiss joined #git
18:40 venmx joined #git
18:41 kadoban Nowhere I've seen
18:41 WayToDoor joined #git
18:41 e14 joined #git
18:42 cagedwisdom joined #git
18:42 [0xAA] joined #git
18:43 ojacobson The backstory is that for a program to get its own name and dock icon, it needs to be in its own .app bundle with appropriate metadata. If it isn't - and wish isn't - you get a generic icon and the binary name, or the baked-in metadata for the binary, as the app name.
18:43 ojacobson I assume there's an API for fiddling that, somewhere, but I'd be surprised if the tcl/tk folks were interested in supporting it just for gitk
18:43 ojacobson the "right" fix (from a mac dev point of view) would be to create a gitk app bundle, and invoke _that_ (through launchservices, probably) to start gitk
18:44 ojacobson this is, for example, what tools like subl and stree (for Sublime Text and SourceTree, respectively) do
18:44 ojacobson Linux desktops don't rely so heavily on metadata for labelling; most desktop managers make it very easy for a window to set its own display name or set it based on the window title
18:45 dave0x6d joined #git
18:45 moop joined #git
18:47 aw1 joined #git
18:49 robics joined #git
18:49 jnavila joined #git
18:57 tag joined #git
18:59 xall joined #git
19:02 sea-gull_ joined #git
19:03 manuel_ joined #git
19:03 vdv joined #git
19:04 NathanI_ joined #git
19:06 katherine joined #git
19:06 Balliad joined #git
19:06 NathanI_ i'm fairly new to git. I clone a repository and then renamed the origin to upstream, then pointed the origin at another gihub repository that I made. Recently I pushed a change, and the changes were pushed to upstream.. I'm not sure why
19:08 neilthereildeil left #git
19:08 NathanI_ i suppose I should mention I recently did a pull from the upstream.
19:12 nikivi joined #git
19:12 jccn joined #git
19:12 aw1 joined #git
19:12 Salander27 joined #git
19:12 Sasazuka joined #git
19:14 _nwkj86 joined #git
19:15 aw1 joined #git
19:15 btree joined #git
19:16 geoid joined #git
19:17 dsantiago joined #git
19:17 jedahan joined #git
19:18 sea-gull joined #git
19:18 breezy_ joined #git
19:19 NathanI_ anyone able to help?
19:20 btree left #git
19:20 yaudu joined #git
19:21 jaguarmagenta joined #git
19:22 qqx NathanI_: That's probably because the branch which you pushed was set to go to upstream, when you renamed the remote that would generally be altered as well.
19:22 QuestionTime joined #git
19:22 grawity 'git remote rename' also adjusts all the config settings
19:23 nedbat NathanI_: can you pastebin your .git/config file?
19:23 qqx You can use `git branch --set-upstream-to=origin/somebranch` to change that setting.
19:23 qqx That would change it for the currently checked out branch.
19:24 johnmilton joined #git
19:25 danslo joined #git
19:25 NathanI_ so what did I do wrong. How should i go about creating a new repository in the future, using an existing repository as a starting point?
19:26 grawity you didn't tell git where to push, so I assume it pushed to the default upstream configured for that branch
19:27 tang^ joined #git
19:27 soc42 joined #git
19:29 NathanI_ it pushed to both my origin and upstream, as far as i can tell.
19:30 SteffanW joined #git
19:31 astrofog joined #git
19:31 svara1 joined #git
19:31 mloy joined #git
19:32 romerocesar joined #git
19:33 tnecniv joined #git
19:34 mloy Hi! What versioning system do y'all recommend for projects that don't have public APIs? I am working on an ES6 + react + redux frontend that talks to a couple of different APIs (which are their own projects) and semver seems great, but according to the RFC having a public API is required, since changes in the major version denote backwards-compatibility breaking releases
19:35 NathanI_ qqx: thanks for the help
19:36 e14 joined #git
19:37 nunchuck joined #git
19:37 cdown_ joined #git
19:38 cjohnson mloy: public could mean many things
19:38 HedgeMage mloy: semver only sort-of requires a public API... semver is based on an older, de facto standard that applied to both projects with and without external APIs.
19:39 phroa joined #git
19:39 HedgeMage mloy: for "public API change" that would cause a major version change, you can substitute e.g. a compatibility-breaking change in your config file format or how you take command line options, just as a couple of examples.
19:39 mloy Or how it talks to its backends
19:39 mloy That makes sense, thank you both
19:40 * HedgeMage nods
19:40 mloy I need to add tests to this project. I'm implementing Driessen's branching model to this project I just got onboarded to
19:40 HedgeMage You are very welcome :D
19:40 mloy So I think I'll start with 0.1.0 and call it pre-alpha
19:40 mloy v0.1.0*
19:41 nettoweb joined #git
19:41 mloy Hmm would v0.0.0 be bad?
19:41 venmx joined #git
19:41 mloy I was thinking v0.1.0 since there already exists a significant chunk of progress
19:41 mloy that I could retroactively refer to as v0.0.0 haha
19:42 matsaman anything <1 you don't have to _call_ pre/alpha/beta
19:42 mloy Yeah
19:42 Dougie187 joined #git
19:42 inflames joined #git
19:44 kulelu88 joined #git
19:46 askb joined #git
19:47 LionsMane joined #git
19:48 irqq joined #git
19:48 frostythesnowman joined #git
19:48 frostythesnowman hi, i'm trying to pull a remote git branch and i'm getting a bunch of whitespace conflicts.  how can i ignore these?
19:49 frostythesnowman i've tried -Xignore-all-space but i'm still getting all the same conflicts
19:49 msonntag joined #git
19:49 pypi_ joined #git
19:50 nicksloan joined #git
19:50 frostythesnowman gist.e.g. https://gist.github.com/boxbeatsy/5615af752336e4c34ce74ecee19ff313
19:51 synthroi_ joined #git
19:51 harwood joined #git
19:52 multi_io joined #git
19:52 pbrewczynski joined #git
19:53 frostythesnowman anyone? :\
19:55 j416 I just spontaneously found the weirdest workflow. I ended up with a nice end result and committed that as. Then I realised that the diff was kind of hard to read since there were many unrelated changes, so I went ahead and undid part of the changes, committed that as "intermediate". Then went ahead and undid some more, and committed that as "first". So I had three commits whose state was correct but in reverse
19:55 j416 order, so I git-replace'd + git-filter-branch'd them incrementally, and now, they are in the right order. Except the author timestamp is all backwards. :D
19:56 j416 s/committed that as./committed that./
19:56 matsaman :D
19:56 matsaman hide that sausage!
19:57 matsaman looking at gitk --all output really makes me wonder if it's even worth trying to
19:57 harwood joined #git
19:57 kadoban I think "hide the sausage" is a different game
19:57 j416 I try hard to make commits that are good examples to follow, and pray that my fellow comrades will at least draw some inspiration from it.
19:57 Guest72933 joined #git
19:58 j416 (actually I don't try very hard, it's quite natural)
19:58 breezy_ joined #git
19:58 matsaman kadoban: that's salami, isn't it? =P
19:59 synthroid joined #git
19:59 matsaman perfect commits are nice, but productivity is even better
20:00 preaction i'm more productive when the other members of my team write good commits
20:04 j416 preaction: exactly.
20:04 j416 and I assume vice-versa, hence it's worth spending a little time extra on making good commits.
20:04 j416 I think git-replace is my new favourite.
20:05 j416 it feels like a nicer interactive rebase
20:05 _Vi joined #git
20:08 nd joined #git
20:09 frostythesnowman i'm trying to pull from a remote repo with "git pull origin -Xignore-all-space" but i'm getting a ton of whitespace conflicts. e.g. https://gist.github.com/boxbeatsy/5615af752336e4c34ce74ecee19ff313
20:09 frostythesnowman could anyone help me figure out how to ignore whitespace conflicts? i've been running up against walls all morning
20:10 romerocesar_ joined #git
20:10 hahuang61 joined #git
20:11 nickabbey joined #git
20:11 ki0 joined #git
20:11 jccn joined #git
20:11 bannakaffalatta joined #git
20:11 mischat joined #git
20:12 johnmilton joined #git
20:12 pancakes9 joined #git
20:13 pancakes9 I created a new folder in a git tracked directory and there's nothing in my .gitignore or .git/info/exclude but it won't pick up this new folder so I can't commit and push it -- any idea what I should be doing to make sure it gets tracked and pushed?
20:13 vuoto joined #git
20:14 bxkq joined #git
20:14 valize joined #git
20:14 ahmed_elgabri joined #git
20:14 Spec-Chum joined #git
20:15 matsaman pancakes9: "pick up"?
20:15 matsaman and what command/s have you used to tell it to "pick up" this directory?
20:15 pancakes9 the new folder doesn't show up under git status
20:15 khagler joined #git
20:16 denisMone joined #git
20:16 pothepanda joined #git
20:16 matsaman pancakes9: you know, git doesn't track directories, it tracks files
20:16 khagler joined #git
20:17 matsaman if you actually want an empty dir in a git repo, you probably want to touch dir/.gitkeep or something like that, I don't keep track of the most popular conventions
20:17 venmx joined #git
20:17 jeffreylevesque joined #git
20:18 ilmari or dir/.gitignore, since the empty drectory presumably exists to hold files you don't want to check in
20:18 ilmari I've also seen dir/.exists
20:19 synthroi_ joined #git
20:19 dodobrain joined #git
20:20 watabou joined #git
20:21 AaronMT joined #git
20:21 Gsham joined #git
20:23 KV joined #git
20:23 joshszep_ joined #git
20:23 nickabbey joined #git
20:23 khagler joined #git
20:23 PityDaFool joined #git
20:24 hahuang65 joined #git
20:24 joshszep_ joined #git
20:24 matsaman that does seem an annoying limitation of git
20:24 hahuang61 joined #git
20:24 Sasazuka joined #git
20:25 jgrasser joined #git
20:25 ixjlyons joined #git
20:25 JamesBaxter joined #git
20:26 joshszep joined #git
20:26 Phrohdoh joined #git
20:26 Phrohdoh o/
20:27 joshszep joined #git
20:28 nicksloan joined #git
20:29 cbegin_ joined #git
20:30 cbegin_ left #git
20:31 Sasazuka_ joined #git
20:33 roygbiv joined #git
20:35 Phrohdoh left #git
20:36 anonymuse joined #git
20:37 Anja joined #git
20:37 Serus joined #git
20:37 duderonomy joined #git
20:37 Serus hello
20:38 Serus I'm trying to enforce lf line endings in my repo, I have core.eol=lf, core.autocrlf=false, and * -text eol=lf in my .gitattributes
20:38 APic Heya
20:38 Serus but I still get windows line endings in my source files
20:38 Serus what am I doing wrong?
20:39 Sasazuka joined #git
20:39 Serus and should I have done this at the beginning or something?
20:39 markus3_ joined #git
20:39 anonymus_ joined #git
20:40 Guest69356 joined #git
20:40 Guest69356 joined #git
20:42 nikivi joined #git
20:43 Orbitrix joined #git
20:43 Orbitrix joined #git
20:44 sdothum joined #git
20:45 anonymuse joined #git
20:46 aidalgol joined #git
20:46 geoid joined #git
20:47 [0xAA] joined #git
20:47 jikz joined #git
20:48 jnavila joined #git
20:48 Singmyr joined #git
20:48 romerocesar joined #git
20:52 johnmilton joined #git
20:53 LionsMane joined #git
20:54 venmx joined #git
20:55 e14 joined #git
20:55 _ikke_ Serus: you don't want -text
20:55 _ikke_ that removes the text attribute
20:55 blackwind_123 joined #git
20:55 _ikke_ which you disable normalization
20:57 Serus I see
20:57 Serus so * text eol=lf should enforce unix line endings, right?
20:57 _ikke_ yes, on checkout
20:57 Serus I want them enforced in the working copy
20:58 Serus or well, too
20:58 _ikke_ "This setting forces Git to normalize line endings to LF on checkin and prevents conversion to CRLF when the file is checked out."
20:58 _ikke_ So it does not convert to lf on checkout
20:58 _ikke_ Serus: Basically, you want to make sure what gets into the repo is already lf
20:59 Sasazuka joined #git
21:00 matsaman really you want to make sure that nobody even uses an OS that uses anything but LF in the first place, heheh
21:00 Serus so, * text eol=lf should work, is what you're saying?
21:00 axisys joined #git
21:00 Serus the problem is
21:00 Serus Qt Creator creates files with crlf endings on windows, with no way to change it
21:00 matsaman oh I'm sure there's a way to change it
21:01 _ikke_ Serus: Right, so just convert it once, and with the attributes file, it should prevent new CRLF to come in
21:01 matsaman but yeah, if you're on Windows, you'd be fighting the norm
21:01 matsaman for Windows
21:01 Serus nope, somebody made a bug report and they declared it "out of scope"
21:01 Serus meanwhile visual studio supports this :|
21:01 matsaman eh, at the very least qt creator is probably open source
21:01 matsaman but it is out of scope
21:02 matsaman too bad VS is an abomination
21:02 Serus what's out scope by configuring an editor to save files as LF?
21:02 d0nn1e joined #git
21:02 matsaman it's just more than required, particularly when the entire OS you're using desires CRLF
21:02 MRChao joined #git
21:02 matsaman it's not like something that is actually broken
21:02 matsaman it's just an imperfection
21:03 matsaman it's not Qt people's job to make your OS use line endings it doesn't want to use: out of scope =)
21:03 matsaman definitely nice to find projects that prefer destroying bugs rather than closing them, though
21:04 nickabbey joined #git
21:05 ThunderChicken joined #git
21:05 ThunderChicken joined #git
21:06 sea-gull joined #git
21:06 cdg joined #git
21:08 Balliad joined #git
21:10 kenoob joined #git
21:10 Orbitrix joined #git
21:11 dsdeiz joined #git
21:11 dreaddymck joined #git
21:12 howitdo joined #git
21:15 Orbitrix joined #git
21:15 Serus okay so
21:16 Serus I changed all files to unix line endings with dos2unix
21:16 peepsalot joined #git
21:17 howitdo joined #git
21:18 MRChao left #git
21:20 WhiskerBiscuit joined #git
21:21 watabou joined #git
21:21 matsaman Serus: nice
21:21 matsaman you could make a nice hacky alias if you wanted
21:21 matsaman to find crlf files & dos2unix them, before committing, etc.
21:22 jaguarmagenta joined #git
21:22 Serus according to an SO article I should * text=auto in .gitattributes, then core.eol lf and core.autocrlf input
21:22 Orbitrix joined #git
21:26 danslo joined #git
21:26 markus3_ left #git
21:27 Serus but I want this option to be enforced on the whole repo
21:27 diogenese joined #git
21:29 durham joined #git
21:29 Michael-Dot joined #git
21:29 Michael-Dot left #git
21:29 Orbitrix joined #git
21:29 romerocesar_ joined #git
21:29 matsaman wouldn't .gitattributes do that, if you commit it?
21:30 Serus and apparently you need to manually run a command that picks up your local .gitconfig
21:30 Serus honestly
21:30 Serus I have no idea how .gitattributes works
21:30 matsaman yeah it's harder to work backwards
21:30 matsaman but it can be done
21:31 Serus it's not often I say this with git
21:31 Serus but I genuinely have no clue how this works
21:31 matsaman https://help.github.com/articles/dealing-with-line-endings/ seems to have a rundown
21:31 Puffball joined #git
21:31 converge joined #git
21:33 Atm0spher1c joined #git
21:33 zzz_ joined #git
21:35 Serus so according to this * text eol=lf should say: every file is a text file with unix line endings
21:36 canton7 ... lf line endings in the repo. the line endings in your working copy depend on core.eol and core.autocrlf
21:36 canton7 ah, no, it uses lf on checkout
21:36 Orbitrix joined #git
21:37 Milos joined #git
21:37 snowkidind joined #git
21:37 durham joined #git
21:37 Oatmeal joined #git
21:38 Serus but what does it do on commit?
21:38 _ikke_ It does not change anything on commit
21:38 _ikke_ All conversion happens between the working tree and the index
21:39 zzz_ joined #git
21:39 canton7 when you 'git add' a file which has crlf, it'll change that to lf in the version that's added to git's index
21:40 breezy_ joined #git
21:40 moei joined #git
21:40 anonymuse joined #git
21:40 Serus warning: CRLF will be replaced by LF in random crlf.txt.
21:40 Serus The file will have its original line endings in your working directory.
21:40 Serus that's what happens if I make a simple file crlf line endings
21:41 Serus so that's half the battle I guess
21:41 canton7 yeah, crlf is being converted to lf on checking
21:41 canton7 *checkin
21:42 re1 joined #git
21:42 canton7 see https://help.github.com/articles/dealing-with-line-endings/#refreshing-a-repository-after-changing-line-endings
21:42 jimi_ joined #git
21:43 Orbitrix joined #git
21:44 Sasazuka_ joined #git
21:44 Serus I think I did it correctly after all
21:45 Serus but notepad++ doesn't recheck the line endings, so it appears as if nothing changed
21:45 Serus reopening the file does show LF
21:46 matsaman notepad++ probably has the following pref options, on Windows: always use crlf, use whatever it already is, always use lf, etc....
21:46 matsaman oh well done
21:46 fuchsi joined #git
21:46 Serus it defaults to crlf, but you can go to edit->EOL conversions->Unix line endings
21:47 lolmac joined #git
21:47 qwerty2 joined #git
21:47 Serus http://puu.sh/sQOyg.png wow, amazing, if only Qt Creator had this option
21:48 geoid joined #git
21:49 Serus so I know you can set hooks
21:49 foul_owl joined #git
21:50 foul_owl How do I see a diff between the last commit of a file and the unstaged changes to the same file?
21:50 _ikke_ git diff HEAD <file>
21:50 frostythesnowman joined #git
21:50 foul_owl Thank you!!
21:50 Serus he said unstaged, so git diff <file> should just work
21:50 Serus unless he means staged
21:51 kyan joined #git
21:52 qqx Serus: He said between the last commit and current. Without specifying HEAD it wouldn't show any staged changes.
21:52 Noldorin joined #git
21:52 _ikke_ by default it shows the diff between the working tree and the index
21:52 mikecmpbll joined #git
21:52 Orbitrix joined #git
21:52 foul_owl Ok I suppose I need know both :)
21:52 Serus I assumed that's what he meant
21:53 foul_owl staged and unstaged
21:53 foul_owl Thank you!
21:53 qqx foul_owl: Specifying HEAD will show both staged and unstaged changes.
21:53 foul_owl Sweet
21:53 xall joined #git
21:55 theoceaniscool joined #git
21:55 theoceaniscool joined #git
21:55 nettoweb joined #git
21:56 msonntag joined #git
21:57 Orbitrix joined #git
21:57 jnavila joined #git
21:59 romerocesar joined #git
22:00 rwb joined #git
22:02 Orbitrix joined #git
22:03 BSaboia joined #git
22:06 cagedwisdom joined #git
22:06 m1dnight_ joined #git
22:07 Orbitrix joined #git
22:10 steeze joined #git
22:10 steeze im trying to use --skip-worktree but it wont let me switch branches since upstream has a differing version of the file
22:10 Lunatrius joined #git
22:10 steeze force checkout of branch doesnt work either. what's the best way to remedy this?
22:11 steeze it's for a config file that is different for each developer and for production so we --skip-worktree locally
22:11 moritz not keeping the file in that repo at all
22:12 moritz use a lightweight configuration management system with support for multiple environments, like Ansible
22:12 moritz (and typically have the configuration in a different repo, but that's optional)
22:13 Bexa joined #git
22:13 steeze moritz for a quick stop gap (need to switch branches for a hotfix) is there something else i can do in the meatime
22:13 steeze ?
22:13 steeze i hear you on ignorinng the config files. im pushing for that but eed to fix this ow
22:14 moritz git stash, maybe?
22:15 Orbitrix joined #git
22:17 raspado joined #git
22:18 DANtheBEASTman joined #git
22:18 devster31 joined #git
22:18 chrysaor joined #git
22:19 MeltedLux joined #git
22:19 saidelike joined #git
22:19 CustosLimen joined #git
22:19 Virox joined #git
22:19 Bilge joined #git
22:21 Orbitrix joined #git
22:21 nikivi joined #git
22:22 lu_zero joined #git
22:24 steeze it says no local changes to save
22:24 freeroute joined #git
22:25 nettoweb joined #git
22:26 nickthetait left #git
22:26 Spydar007 joined #git
22:26 freeroute hi. I want to gitignore everything but home/.cinnamon/configs* and home/.cinnamon/cinnamon*
22:26 freeroute right now I have http://apaste.info/RCtVn but it doesn't seem to work
22:28 hahuang61 joined #git
22:29 Orbitrix joined #git
22:30 jimi_ joined #git
22:30 Dougie187 joined #git
22:30 romerocesar joined #git
22:31 WaReZ joined #git
22:31 freeroute what am I doing wrong?
22:33 Asmodius joined #git
22:35 asmodius_ joined #git
22:35 matsaman freeroute: I thought exceptions (!) were meant to come afterward/last
22:36 venmx joined #git
22:37 matsaman more provisos: http://stackoverflow.com/questions/3203228/git-ignore-exception
22:38 hahuang61 joined #git
22:40 steeze i thought one of the main points of --skip-worktree was to ignore local changes and not have local changes get overwritten on checkout and stash. but adding skip-worktree to a file, modifying that file, then trying to switch branches throws an error saying my local changes would get overwritten, but like i said i thought that was the point?
22:41 Asmodius joined #git
22:41 dreaddymck joined #git
22:41 Asmodius test
22:44 Asmodius Hello, Assistance required for a git implementation on RHEL that pulls from TFS. I was tasked to setup git to pull on a schedule with a service account. I setup a script that decrypts the system account password and passes it into git pull https://user:pass@tfsurl/git. Unfortunantly the password is being logged in plain text in the logs/HEAD. any ways to prevent this?
22:44 ki0 joined #git
22:44 tmsmith joined #git
22:44 oaao joined #git
22:45 matsaman Asmodius: on what, a Unix system?
22:45 Asmodius yeah RHEL red had enterprise linux
22:45 Asmodius hat^
22:46 matsaman oh sorry you did say that
22:46 matsaman you can't use keys?
22:49 Asmodius the script uses a key to decrypt the password into a variable and then pass that into the url so the git pull occurs with the services accounts credentials. that works well but its loggin the command with the URL including the password into the /logs/HEAD file
22:49 geoid joined #git
22:49 matsaman_ joined #git
22:49 matsaman_ joined #git
22:50 matsaman joined #git
22:50 sonOfRa Asmodius: TFS should support cloning over SSH, I'd recommend you set up a key for the user, and pull via git
22:51 Orbitrix joined #git
22:51 sonOfRa You can encrypt the private key, and decrypt it the same way you currently pass a passphrase into the git url
22:51 Asmodius thanks do you perchance have a url to a decent guide for that?
22:52 sonOfRa https://www.visualstudio.com/en-us/docs/git/use-ssh-keys-to-authenticate Not sure if that helps. I'm not really a windows person and haven't worked with TFS in ages, but I remember that it was possible
22:52 Atrumx joined #git
22:52 sonOfRa Step 2 is probably the most interesting, how to add the public key to an account
22:52 Asmodius your a legend tyvm
22:54 Asmodius ill give this a go and jump back in if it didnt work. *virtual high five*
22:55 elastix joined #git
22:56 Orbitrix joined #git
22:56 Vauff joined #git
22:57 Salander27 joined #git
22:59 axisys joined #git
23:00 durham joined #git
23:00 holgersson joined #git
23:01 Vauff joined #git
23:01 fmeerkoetter joined #git
23:01 freeroute joined #git
23:01 freeroute matsaman: thanks I'll take a look
23:01 lacour joined #git
23:02 durham_ joined #git
23:03 [Saint] joined #git
23:04 Orbitrix joined #git
23:08 Gsham joined #git
23:11 przemoc joined #git
23:14 penanze joined #git
23:17 jcadduono left #git
23:18 Anja joined #git
23:18 tag joined #git
23:19 tag joined #git
23:22 watabou joined #git
23:22 jaguarmagenta joined #git
23:23 denisMone joined #git
23:23 mloy Is there a way to configure git to only GPGsign for commits to master, and not other branches?
23:23 johnny56 joined #git
23:24 mda joined #git
23:24 danslo joined #git
23:24 mloy Right now in .git/config I have `[commit] gpgsign = true` set, but I only want that to apply for commits where I merge release branches into master.
23:26 kadoban mloy: Probably just sign those manually, or do signed tags or something?
23:29 dsdeiz joined #git
23:30 mloy kadoban: Thanks. I was being dumb and had it set to always sign commits in .git/config, but I removed that
23:31 kadoban Well, it's not an unreasonable request, I just don't think there's a config that does exactly that.
23:34 BrianJ joined #git
23:34 dsdeiz joined #git
23:34 dsdeiz joined #git
23:35 troyt joined #git
23:36 AfroThundr joined #git
23:38 svm_invictvs If I modify a commit comment
23:38 svm_invictvs And then re-push is it going to make a mess?
23:39 kadoban svm_invictvs: If anyone is using the history you already pushed, possibly.
23:39 kadoban !rewrite
23:39 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
23:39 mojave maybe better to just leave the mistake in, then?
23:41 AfroThundr joined #git
23:41 kadoban It depends, but can be.
23:41 cyan__ joined #git
23:42 AfroThundr joined #git
23:43 Vampire0 joined #git
23:44 mojave I know i've accidentally had repeat comments in commits.. i've just left them. not sure if that's bad or not
23:44 _28_ria joined #git
23:46 sea-gull joined #git
23:49 arescorpio joined #git
23:49 geoid joined #git
23:50 bannakaffalatta joined #git
23:55 romerocesar joined #git
23:57 durham joined #git

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