Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-09-15

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

All times shown according to UTC.

Time Nick Message
00:00 idmistir nah, I thought of that too. phpstorm does not allow you to set --force from its gui unless you explicitly remove the restriction
00:00 preaction then you could not have pushed that to bitbucket
00:00 idmistir from an entirely different panel
00:00 durham joined #git
00:00 idmistir Then how did the staging server lose the changes -_-
00:00 preaction either they were never in bitbucket, or you didn't push to bitbucket
00:00 Noodlewitt joined #git
00:00 idmistir Either I'm wrong about the --force, somehow.. or this is an impossible situation
00:01 idmistir Because the staging server only reads from bitbucket
00:01 preaction or, yes, you force pushed and deleted the code
00:01 idmistir Alright, let's consider that for a second... if I force pushed.. buh bye?
00:02 idmistir Furthermore, can I somehow prove or disprove that I force pushed using a git log command or so?
00:02 preaction yes, if you force-pushed, it's gone. it exists on Smith's computer though, perhaps
00:02 idmistir preaction; inside the reflog?
00:03 preaction maybe bitbucket keeps a history?
00:03 preaction no, it exists on smith's computer because they did the work. unless they pulled and then yes, the reflog
00:03 idmistir they definitely pulled
00:03 idmistir multiple times
00:03 idmistir is it worth checking the reflog?
00:03 preaction does it cost anything to check the reflog?
00:04 idmistir Alright, wrong phrasing
00:04 Sasazuka__ joined #git
00:04 idmistir Just asking if there's any chance at all, to your knowledge, that because he wasn't the one to force push, that he would have a track of this in his reflog
00:04 dsdeiz_ joined #git
00:05 preaction yes
00:05 idmistir Alright.. thank you very much
00:05 idmistir preaction; may I pm?
00:05 preaction no, why?
00:05 idmistir I promise I'm not going to bug you
00:05 idmistir Just one sentence :-)
00:07 preaction that's not an answer to "why?"
00:08 bruce_lee joined #git
00:08 bruce_lee joined #git
00:09 om_henners joined #git
00:09 LeBlaaanc joined #git
00:10 madewokherd joined #git
00:11 courrier________ joined #git
00:11 noordinaryspider left #git
00:13 YuGiOhJCJ joined #git
00:13 bitmod joined #git
00:14 Kaisyu joined #git
00:14 shdw joined #git
00:18 Oatmeal joined #git
00:19 safe joined #git
00:19 nydel07 joined #git
00:21 ashpokemonZ joined #git
00:22 brent__ joined #git
00:23 debakarroy2 joined #git
00:24 debakarroy2 joined #git
00:24 Cabanossi joined #git
00:25 PtxDK joined #git
00:31 dsdeiz_ joined #git
00:32 tunekey joined #git
00:37 Gurkenglas joined #git
00:41 shdw joined #git
00:44 kculpis joined #git
00:47 xaa joined #git
00:48 kerrick joined #git
00:48 cagedwisdom joined #git
00:48 bronson joined #git
00:52 raijin joined #git
00:53 Sootherer left #git
00:56 multi_io_ joined #git
00:58 ahrs joined #git
00:59 roelmonnens joined #git
01:01 sunri5e joined #git
01:02 Feral_Cat joined #git
01:03 dsdeiz_ joined #git
01:03 Argylelabcoat joined #git
01:06 raijin joined #git
01:08 rnsanchez joined #git
01:10 Cabanossi joined #git
01:14 cmosguy There is this one file in my repo that I need to keep as an LF instead of CRLF for Windows. I am on Windows. How do I force that one file to always be LF in my work area?
01:14 jnewt joined #git
01:15 dreiss joined #git
01:16 Bennux_dev joined #git
01:18 LeBlaaanc joined #git
01:18 stutter joined #git
01:19 kadoban cmosguy: Should be something in .gitattributes , though you'd have to look up what the exact flag is
01:20 Khisanth joined #git
01:24 finalbeta joined #git
01:26 piremi joined #git
01:27 hherman1 joined #git
01:28 Murr joined #git
01:29 brent__ joined #git
01:29 jokester joined #git
01:31 sunri5e joined #git
01:31 ntz joined #git
01:31 ntz hello
01:32 kadoban Hello
01:32 robertparkerx joined #git
01:32 ntz sorry for stupid Q: but how do I ask some git cloned repository to tell me what version it reports ? it's not in gut log
01:32 ntz **git log
01:33 rewt what do you mean?
01:34 ntz latest stable version of ModemManager is 1.6.8 and I want to ask what version reports back current dir git://anongit.freedesktop.org/ModemManager/ModemManager
01:34 kadoban ntz: It should be the first commit in git log. Or you can do like uhm, 'git rev-parse --verify HEAD' or something with git describe
01:34 nd- joined #git
01:34 kadoban Oh, should be git describe, that should help you out. Assuming they use tags correctly. Otherwise it's not really possible to tell.
01:35 ntz 0.7.990-1998-gbfff4f53 holy sh*t :D
01:36 ntz kadoban: obviously they dnb't use tags correctly since it should be >= 1.6.8
01:36 kadoban Hmmmm
01:36 Blackpajamas joined #git
01:37 ozcanesen joined #git
01:38 rewt no, they do have tags
01:38 ntz it's so annoying ... so if I download latest stable from here https://www.freedesktop.org/software/ModemManager/ and check this https://cgit.freedesktop.org/ModemManager/ModemManager
01:39 ntz I still have no idea what all has changed :(
01:39 kadoban You can try 'git tag' and see what tags exist
01:39 rewt so what do you really want?  the latest tag, or a changelog?
01:39 kadoban If one exists you want, you can just check it out
01:40 ntz rewt: I'd prefer changlog since latest stable 1.6.8
01:40 Evalix joined #git
01:40 ntz because git tag ends with 1.6.8
01:40 blaeks joined #git
01:41 wxg joined #git
01:41 d^sh joined #git
01:55 ilbot3 joined #git
01:55 Topic for #git is now Welcome to #git, the place for git help and ><(((('> | Public logs at https://gitirc.eu/log | First visit? Read: https://gitirc.eu | Current stable version: 2.14.1 | Getting "cannot send to channel"? /msg gitinfo .voice | Have you ever seen a git eat its own HEAD?
01:56 piremi joined #git
01:58 cqi joined #git
02:00 aFeijo left #git
02:05 jameser joined #git
02:06 Noodlewitt joined #git
02:10 Cabanossi joined #git
02:11 roygbiv joined #git
02:12 bongozig joined #git
02:13 Bhootrk_ joined #git
02:13 ntz rewt: stupid Q:
02:14 ntz there's always missing a configure ... am I supposed to use instead of configure an autogen.sh ?
02:16 rewt that sounds like a question for that project's dev
02:17 ntz ok, fair, thanks
02:17 bongozig_ joined #git
02:19 jimmyrcom joined #git
02:20 bongozig_ joined #git
02:20 orbyt_ joined #git
02:22 bongozig_ joined #git
02:23 bongozig joined #git
02:23 justan0theruser joined #git
02:24 shdw- joined #git
02:27 ntz okay .. I still have a same problem
02:28 ntz I need to build something and I need to determine for it a version string in some sensible way
02:28 brent__ joined #git
02:29 ntz for example this is version string for kernel-vanilla package in suse: kernel-vanilla-4.13.rc5.67.g510c8a8-1.1.g9766d60.x86_64
02:29 ntz which is 4.13-rc5 + 67 commits
02:30 ntz lets's say I need to somehow set version (string) for this: git://anongit.freedesktop.org/libmbim/libmbim
02:31 ntz git tag says that last version is 1.14.2 but how should I build some release substring that wil increment in sensible way with commits until the tag will report back higher main version string ?
02:32 piremi joined #git
02:33 Puffball joined #git
02:34 enkrypt joined #git
02:36 hash-z3r0 joined #git
02:39 chachasmooth_ joined #git
02:39 courrier________ joined #git
02:40 lagothri1 joined #git
02:42 Vortex35 joined #git
02:47 dviola joined #git
02:47 roelmonnens joined #git
02:49 Gurkenglas joined #git
02:49 bronson joined #git
02:53 shdw- joined #git
02:57 courrier________ joined #git
02:57 thiago joined #git
02:57 kfoonamalik joined #git
03:03 govg joined #git
03:04 pks joined #git
03:05 ahr3n joined #git
03:06 prg318 joined #git
03:06 foxkit joined #git
03:06 foxkit hi there, I was wondering if it was possible to maybe have a bit more 'live' help than ML for a thread I've recently posted: https://public-inbox.org/git/59BB3E40.7020804@adelielinux.org/T/
03:09 Plouescat joined #git
03:09 kartheek joined #git
03:09 Cabanossi joined #git
03:14 dka_ joined #git
03:17 fahadash joined #git
03:19 diqidoq_ joined #git
03:19 igemnace joined #git
03:21 Nizumzen joined #git
03:21 prg318 joined #git
03:22 xaxxon joined #git
03:22 shdw- joined #git
03:23 bayman joined #git
03:24 shdw- joined #git
03:25 igemnace joined #git
03:26 kfoonamalik joined #git
03:26 Noodlewitt joined #git
03:27 ozcanesen joined #git
03:28 Murr joined #git
03:29 dviola joined #git
03:29 justan0theruser joined #git
03:31 bayman left #git
03:32 michelangelo joined #git
03:33 v4ult joined #git
03:37 pijiu joined #git
03:38 Cabanossi joined #git
03:39 dsdeiz joined #git
03:40 LuksLux joined #git
03:42 courrier________ joined #git
03:43 danieli joined #git
03:44 robotroll joined #git
03:47 danieli joined #git
03:50 dka_ joined #git
03:51 sarbs joined #git
03:51 upperdeck joined #git
03:52 MattMaker joined #git
03:54 YuGiOhJCJ joined #git
03:56 dka__ joined #git
03:59 Gurkenglas joined #git
03:59 upperdeck_ joined #git
04:04 pijiu2 joined #git
04:08 Bl4ckpajamas joined #git
04:08 beerdrop joined #git
04:09 noyb joined #git
04:12 cdown joined #git
04:16 cdown_ joined #git
04:19 steeze joined #git
04:20 mkali joined #git
04:23 tarkus joined #git
04:27 Kavothe joined #git
04:31 _ikke_ foxkit: hi
04:32 foxkit hi _ikke_
04:32 dka_ joined #git
04:35 _ikke_ foxkit: So do I understand that gc does not seem to respect the lock and does run pack-ref (so the loose ref is gone)?
04:35 roelmonnens joined #git
04:35 courrier________ joined #git
04:35 foxkit _ikke_: yes
04:36 _ikke_ interesting
04:36 _ikke_ and it only happens in 2.14.1 (not 2.14.0)?
04:38 _ikke_ You could try to bisect it to find out what commit introduced the breakage
04:38 upperdeck joined #git
04:39 Cabanossi joined #git
04:41 noyb joined #git
04:41 MattMaker joined #git
04:42 MattMaker joined #git
04:45 MattMaker joined #git
04:45 MattMaker joined #git
04:46 rkta_ joined #git
04:46 Noodlewitt joined #git
04:47 Gurkenglas joined #git
04:49 XVar joined #git
04:49 XVar joined #git
04:50 LuksLux joined #git
04:51 bronson joined #git
04:53 _ikke_ foxkit: I can at least reproduce in on 2.14.1
04:53 MattMaker joined #git
04:56 wxg joined #git
04:58 _ikke_ 2.14.0 fails too
05:01 _ikke_ foxkit: does building git for you also give soo much warnings? (about redirecting sys/pol.h to pol.h)
05:03 rscata joined #git
05:04 AnujSharma joined #git
05:07 joeyy joined #git
05:07 MattMaker joined #git
05:07 _ikke_ foxkit: this is the commit that introduced the test that is failing (introduced in 2.14.0)
05:08 imack joined #git
05:09 parsnip can we git fetch a second "initial commit" from github?
05:09 parsnip `git fetch` says "warning: no common commits"
05:10 osse In that case you just did
05:10 kadoban warning doesn't sound like it's going to stop you?
05:10 dec0n joined #git
05:10 parsnip okay, will think on it
05:11 foxkit _ikke_: what is the commit?  I don't see any.  and yes, poll.h gives lots of warnings, but it is harmless and part of musl
05:13 foxkit _ikke_: oh, I see.  the commit that introduced the test.  yeah
05:14 _ikke_ foxkit: c45af94dbc960cbb46ac482ec9cb7f73f55ea270
05:14 freimatz joined #git
05:14 _ikke_ https://github.com/git/git/commit/c45af94dbc
05:20 Anthaas joined #git
05:21 xaxxon joined #git
05:21 Bock joined #git
05:22 fatihgenc joined #git
05:26 kerrick joined #git
05:27 upperdeck joined #git
05:28 Murr joined #git
05:28 courrier________ joined #git
05:30 Plouescat joined #git
05:33 steven joined #git
05:33 mkali joined #git
05:35 MrMojit0 joined #git
05:36 upperdeck joined #git
05:38 lss8 joined #git
05:38 Cabanossi joined #git
05:40 Noodlewitt joined #git
05:46 dreiss joined #git
05:48 ppang joined #git
05:49 dsdeiz_ joined #git
05:52 ThomasLocke joined #git
05:53 Raging_Hog joined #git
05:55 hash-z3r0 joined #git
05:57 Plouescat joined #git
05:59 zulutango joined #git
05:59 mkali joined #git
05:59 sekmo joined #git
06:05 hash-z3r0 joined #git
06:06 overlord_tm joined #git
06:06 dre_ joined #git
06:10 upperdeck joined #git
06:14 abba joined #git
06:14 hash-z3r0 joined #git
06:15 ams__ joined #git
06:15 dka_ joined #git
06:19 Noodlewitt joined #git
06:19 roelmonnens joined #git
06:19 roelmonnens joined #git
06:22 steven joined #git
06:24 ZyX-I joined #git
06:28 wandering_vagran joined #git
06:28 jstimm joined #git
06:29 theoceaniscool joined #git
06:34 cagedwisdom joined #git
06:35 fstd joined #git
06:39 Cabanossi joined #git
06:39 Tobbi joined #git
06:42 causasui joined #git
06:43 ltaylor joined #git
06:44 roelmonn_ joined #git
06:46 roelmonn_ joined #git
06:47 dreiss joined #git
06:50 shdw joined #git
06:51 bronson joined #git
06:54 MattMaker joined #git
06:54 dka_ joined #git
06:58 luc4 joined #git
06:58 roelmonnens joined #git
06:59 dsdeiz joined #git
06:59 dsdeiz joined #git
06:59 Zoxc joined #git
06:59 roelmonnens joined #git
07:03 JeroenT joined #git
07:04 wandering_vagran joined #git
07:07 high_fiver joined #git
07:09 Cabanossi joined #git
07:09 jameser_ joined #git
07:10 luc4 joined #git
07:11 jmccarthy joined #git
07:12 TomyWork joined #git
07:13 lkolstad joined #git
07:20 schleppel joined #git
07:21 MACscr joined #git
07:22 shdw- joined #git
07:24 dege joined #git
07:26 tvw joined #git
07:27 Mattiaslndstrm joined #git
07:27 venmx joined #git
07:28 Murr joined #git
07:29 imrg90 joined #git
07:30 sergio joined #git
07:31 sergio does git have any built-in command to untar a file ? I need do write a hook that that needs to untar something, but this will also be used on Windows, which usually doesn't have tar installed
07:32 _ikke_ No, git itself has nothing
07:32 igemnace sergio: don't think so, git implements its own zlib compression. maybe git bash on windows has tar?
07:32 venmx_ joined #git
07:32 osse worth checking if Git for Windows has a tar tagging along
07:32 czart_ joined #git
07:32 osse wouldn't surprise me
07:33 sergio ok, thanks
07:33 osse another solution would be to change the tar archive you have in your repo into a zip :P
07:33 osse Or simply unpackit
07:34 sergio the project has a 40 GB 3rdparty folder, I tried using git-lfs, but there's many small files so takes time, tar would be faster
07:34 piremi joined #git
07:34 imrg90 joined #git
07:35 osse ah
07:36 luc4 joined #git
07:36 King_Hual joined #git
07:36 King_Hual joined #git
07:36 osse do you happen to use cmake in your project?
07:36 osse that can tar and untar files
07:37 JeroenT_ joined #git
07:37 cyberz joined #git
07:38 sergio unfortunately not, but I might just document that the user must untar it himself :)
07:39 czart joined #git
07:40 DaGoaty joined #git
07:43 Bhootrk_ joined #git
07:44 dsdeiz joined #git
07:44 Jak_o_Shadows joined #git
07:47 energizer joined #git
07:48 imrg90 joined #git
07:49 dsdeiz_ joined #git
07:51 JeroenT joined #git
07:51 Lunatrius joined #git
07:54 mescalinum joined #git
07:54 MattMaker joined #git
07:54 JeroenT_ joined #git
07:55 aboya joined #git
07:56 acidjnk joined #git
07:58 pbandark joined #git
07:58 JeroenT joined #git
08:00 ShalokShalom joined #git
08:02 jameser joined #git
08:03 mostlybadfly joined #git
08:05 aard_ joined #git
08:06 robinb joined #git
08:06 imrg90 left #git
08:09 Cabanossi joined #git
08:10 aboya Is there any explanation for this behavior in the webkit repositories? Files with valid CRLF line endings are being modified... yet are still the same byte-per-byte, and there is no way to get those "changes" out of git status. https://dpaste.de/6HaT
08:12 cstratak joined #git
08:12 CheckDavid joined #git
08:12 osse 'What doesn't git diff say?
08:12 osse *does
08:13 aboya osse: https://dpaste.de/NG3v
08:13 aboya minus all the lines, plus all the lines
08:13 aboya it's quite wierd too
08:14 sphalerite joined #git
08:14 aboya If you clone the webkit repo in a linux computer, run git status, copy the repository to another folder or transfer it via rsync and run git status on the copy, you get those nasty "changes" in your git status.
08:15 courrier________ joined #git
08:15 aboya There are many other .sln and .vcproj files unaffected. It's always these four.
08:15 cdg joined #git
08:17 cdg_ joined #git
08:17 sphalerite Is it safe/sensible to run a tool that finds identical files and hardlinks them across the .git dir for severla checkouts of closely related repos?
08:17 osse aboya: while does  `git show HEAD:thatfile | file -` and `file thatfile` say?
08:18 JeroenT joined #git
08:19 aboya osse: https://dpaste.de/PPSu
08:21 ertes-w joined #git
08:21 osse aboya: git check-attr --all Source/ThirdParty/libwebrtc/Source/third_party/jsoncpp/source/makefiles/vs71/jsoncpp.sln  ?
08:21 aboya Source/ThirdParty/libwebrtc/Source/third_party/jsoncpp/source/makefiles/vs71/jsoncpp.sln: eol: crlf
08:22 osse ahh
08:22 osse ok, I get it. The file as stored in the repo has CRLF
08:23 osse it shouldn't. the attributes just say that when the file is checked out it should have CRLF on disk. it should still have LF *inside the repo*
08:24 cdg joined #git
08:24 cbreak joined #git
08:26 osse I think 'dos2unix those files' && git commit those files will take care of it
08:27 aboya I think I understand now...  Thank you very much.
08:27 aboya It should be fine if I just committed those changes, right?
08:28 jlebrech joined #git
08:28 aboya Even if I see them as CRLF, they should already be LF in the git database, if they are normalized.
08:28 aboya Is that correct?
08:28 osse yes exactly
08:28 osse but they are CRLF in the databse
08:28 aboya They are now, but if I commit they will turn into LF.
08:29 dka_ left #git
08:29 bongjovi joined #git
08:29 wandering_vagran joined #git
08:29 jlebrech we're using a static site generator that creates a build directory, and we have a build branch in git. how to pull the contents of the master/build into the build branch?
08:31 osse aboya: alternatively you can keep them as they are and make git *not* attempt any form of conversion
08:31 osse aboya: instead of 'eol=crlf' you can mark them '-text'
08:31 aboya But then I would have modified .gitattributes and I would have the same problem, right?
08:31 clmsy joined #git
08:31 aboya Unless there is an override file in .git/
08:32 acidjnk2 joined #git
08:32 cdg joined #git
08:33 osse yes you can override gitattributes
08:33 osse but both my suggestions are imho proper fixes that you might as well commit and share with the world
08:34 aboya That's what I'm going to try. It may not be as easy as I wanted because webkit  uses git-svn and I'm not yet a committer.
08:35 bongjovi joined #git
08:36 j08nY joined #git
08:36 Kavothe joined #git
08:37 osse aboya: printf '*.sln -eol -text\n*.vsproj -eol -text' >> .git/info/attributes
08:37 osse see if that works
08:37 osse hmm, add an \n at the end there
08:38 osse jlebrech: can you rephrase? I don't get it
08:38 aboya Yes it does, thank you
08:38 osse What is "master/build" ? The generated build directory when you have master checkoed out+
08:38 osse aboya: wonderful. praise gaben
08:38 tvw joined #git
08:39 jlebrech i have a directory in master called build want to copy that to another branch but at the root level
08:40 kent\n you can't have branch directories and branches share names
08:40 kurkale6ka joined #git
08:40 kent\n eg:  master + master/builds # bang, because you can't create a node inside a node
08:40 kent\n master + builds/master # is fine though, because you can create nodes in directories
08:41 osse jlebrech: git add build; treehash=$(git write-tree --prefix=build/); git checkout build; git commit-tree -p HEAD -m 'New build' $treehash
08:41 osse oops, needs a reset
08:43 cagedwisdom joined #git
08:44 jlebrech wow, I don't know whey middleman (site generator) did it this way. I think it used to build into another branch but they might have removed that.
08:44 jlebrech it might be better to just point vhost to that build folder. and have less git magic.
08:44 jsears joined #git
08:47 cdg joined #git
08:48 elsevero joined #git
08:49 kharloss joined #git
08:52 bronson joined #git
08:53 Codaraxis_ joined #git
08:55 MattMaker joined #git
08:57 sekmo joined #git
08:59 Tobbi joined #git
09:00 Floflobel joined #git
09:00 bergice_ joined #git
09:04 cdg joined #git
09:04 abrcdbr joined #git
09:06 achadwick joined #git
09:06 chll_ joined #git
09:06 se1_ joined #git
09:07 Ladus joined #git
09:08 gxt joined #git
09:11 dsdeiz joined #git
09:11 dsdeiz joined #git
09:13 jmccarthy left #git
09:17 cdg joined #git
09:19 aboya osse: why gaben?
09:20 aboya Is there a way to force a normalization rescan in git? I cleared .git/info/attributes but now `git status` does not show the files I needed to change.
09:22 manuelschneid3r joined #git
09:23 cdg joined #git
09:25 Richard_Cavell Hello everyone.  I'm a newb and I have a github project for which I am the only contributor.
09:25 Richard_Cavell I want to see all of the changes in my project between revision A and revision B, for all files but not file C.  How do I do that?
09:25 _ikke_ aboya: delete the index and run git reset
09:26 dminuoso Through what mechanism can a pre-commit hook block a commit from happening?
09:26 _ikke_ dminuoso: exiting with a non-zero value
09:26 dminuoso Ahh.
09:27 dminuoso _ikke_: So it's also the hooks job to put an appropriate error message to stdout?
09:28 _ikke_ yes
09:28 Murr joined #git
09:29 i7c Richard_Cavell: g diff A B -- ':(exclude)path/to/file/C'
09:29 i7c g=git
09:30 cdg joined #git
09:30 i7c diff --name-status or something respectively
09:30 aboya _ikke_: thanks, that worked :)
09:31 foxpup joined #git
09:31 Solvaring joined #git
09:33 TikityTik joined #git
09:34 roelmonnens joined #git
09:34 cdg_ joined #git
09:34 steffen left #git
09:35 cdg joined #git
09:36 murii joined #git
09:36 Richard_Cavell i7c: https://ideone.com/dpBCKX
09:44 johnnyfive joined #git
09:46 brent__ joined #git
09:46 Mixologic joined #git
09:48 Saliedro joined #git
09:50 dsdeiz_ joined #git
09:50 basiclaser joined #git
09:51 Richard_Cavell i7c: adding a period to the command line fixed it.
09:51 Richard_Cavell thanks
09:53 jhe_ joined #git
09:56 MattMaker joined #git
10:05 cdg joined #git
10:06 Saliedro joined #git
10:08 renarc joined #git
10:08 Cabanossi joined #git
10:16 Saliedro joined #git
10:18 Acerific joined #git
10:18 Acerific joined #git
10:18 Anticom joined #git
10:21 mescalinum joined #git
10:24 sarri joined #git
10:24 sarri joined #git
10:25 cagedwisdom joined #git
10:29 belmoussaoui joined #git
10:29 rokups joined #git
10:30 Saliedro_ joined #git
10:31 Borg- joined #git
10:31 musca` joined #git
10:33 belmoussaoui joined #git
10:34 osse aboya: gaben is the light of the world
10:37 aboya The only gaben I know is the one from Valve. I don't see what he could have to do with line terminators, unless he was the one to decide that Windows should use CRLF while he worked there.
10:39 mescalinum joined #git
10:40 daguiam joined #git
10:43 osse if he was the one who decided that then condemn gaben!!
10:43 venmx joined #git
10:43 venmx_ joined #git
10:44 ShalokShalom joined #git
10:45 mkali joined #git
10:46 PuzzledFolk joined #git
10:47 PuzzledFolk Hello everybody
10:47 gitinfo PuzzledFolk: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
10:50 PuzzledFolk I hate to bother other but I ran into a problem with merging I can't solve. I have a master branch and a feature branch and a problematic file. When I branched off of master, the problematic file had a bunch of code in it that at this time has been moved (copy-pasted) in a new file)
10:50 PuzzledFolk My master branch is fine and everything, all those modifications are in. My feature branch though, still has that problematic file with the bunch of code that was removed.
10:51 luc4 joined #git
10:51 PuzzledFolk I merged master into my feature branch several times, last a few seconds ago, but nothing helped. Master keeps having the correct file and feature keeps having the older, wrong version of that file.
10:52 PuzzledFolk Even github tells me that there's no merge conflict (I of course had checked and solved them before opening a PR)
10:52 PuzzledFolk Any idea what this might be due too and how to solve it?
10:53 Cabanossi joined #git
10:54 bronson joined #git
10:54 osse PuzzledFolk: git thinks the feature branch version is "newer"
10:54 jimsio joined #git
10:54 rawste joined #git
10:56 PuzzledFolk osse: Yeah I figured, but how do I convince it that it's not? Would I just be better off creating a new commit in my feature branch with the changes removed?
10:56 osse PuzzledFolk: can you pastebin the output of `git log --decorate --oneline --graph master feature` ?
10:56 osse as long as is needed to show where the feature branch originally started
10:57 AndreasLutro usually it's easier to rebase your feature branch onto master, not merge master into the feature branch
10:57 govg joined #git
10:57 PuzzledFolk osse: Might have to go back quite a lot, this feature branch is several weeks old
10:57 osse PuzzledFolk: for all I care you can pastebin everything :p
11:00 PuzzledFolk osse: How do I even tell from the log where that branch started? o.o
11:00 osse I don't know because I can't see it. Presumably if you scroll down far enough there is only one line?
11:01 osse ah
11:01 osse try `git log --decorate --oneline --graph --first-parent master feature`
11:02 _28_ria joined #git
11:02 roelmonn_ joined #git
11:02 JeroenT joined #git
11:03 PuzzledFolk I think I found it
11:04 oxymoron93 joined #git
11:04 PuzzledFolk Though, now that I think of it, I branched from master with an older branch (which got abandoned) and branched my current feature branch from that one
11:04 Nizumzen joined #git
11:07 marinin joined #git
11:12 gloomy_ joined #git
11:13 upperdeck joined #git
11:15 raijin joined #git
11:18 upperdeck_ joined #git
11:18 imack joined #git
11:18 aboya Fixed the issue with CRLF terminators in webkit. https://bugs.webkit.org/show_bug.cgi?id=176991
11:19 pbandark joined #git
11:20 plexigras joined #git
11:21 wandering_vagran joined #git
11:21 justache joined #git
11:21 hsiktas joined #git
11:21 JeroenT joined #git
11:22 osse <3
11:25 jesk joined #git
11:25 jesk stupid question, but 'git commit -a' when do I need that?
11:25 jesk the '-a'
11:26 osse jesk: when you are too lazy to run git add first
11:27 jesk so I always do 'git add -A .' first
11:27 _ikke_ When you don't care about how your history looks
11:27 _ikke_ jesk: That doesn't matter
11:28 _ikke_ jesk: The difference is whether you commit all uncomitted changes, or only selected changes that logically belong together
11:28 jesk can't follow
11:28 jesk hm ok
11:28 _ikke_ jesk: git allows you to specify exactly what you want to commit
11:28 jesk but dont I want to commit always all changes? what are 'logically belong together'-changes?
11:28 jmccarthy joined #git
11:29 oxymoron93 not all changes go into commit
11:29 _ikke_ jesk: The reason is that you can create logical commits, people can follow, rather than a single bulk commit which contains many unrelated changes
11:29 marinin joined #git
11:29 Murr joined #git
11:29 jesk and which ones would be excluded?
11:30 jesk i have some auto checkin script
11:30 jesk which fetchs things, which are some configuration backups
11:30 jesk then deletes the old content of that folder
11:30 jesk and put in complete new content
11:30 jesk then checks that in
11:30 osse jesk: say you fix a bug. during that you added som printf()s or printouts or whatever.
11:30 jesk so each commit is complete different content, without the old content
11:30 osse git add the bug fix. don't add the other crap
11:31 jesk but that would be related to the git-add and less to git-commit,. not?
11:31 netj joined #git
11:32 osse jesk: it was an example of a case where you don't want to just commit everything unconditionally
11:33 oxymoron93 and git commit -a does that ^  jesk
11:33 oxymoron93 except new files iirc
11:34 belmoussaoui_ joined #git
11:34 ertes-w joined #git
11:36 theoceaniscool joined #git
11:38 belmoussaoui joined #git
11:38 Cabanossi joined #git
11:39 venmx joined #git
11:39 venmx_ joined #git
11:40 Cavallari joined #git
11:40 SkyPatrol joined #git
11:42 kurkale61 joined #git
11:46 piremi joined #git
11:50 dsdeiz_ joined #git
11:51 ShalokShalom_ joined #git
11:52 lupine joined #git
11:55 Acerific joined #git
11:55 Acerific joined #git
11:56 ntz I have question about this: git://anongit.freedesktop.org/ModemManager/ModemManager .... if I ask git tag it tells me the last used tag is 1.6.8 however there's still N commits since then (eg ``git log master ^1.6.8'')
11:56 ntz ^^ how you'd reliably and sanely determine the release substring for that, eg like for exampe is substring derived in here: kernel-vanilla-4.13.rc5.67.g510c8a8-1.1.g9766d60.x86_64
11:56 ntz ^^ eg 4.13-rc5 + 67 commits since the + ?1.1 (not sure why)
11:57 osse ntz: 'git describe'
11:57 MattMaker joined #git
11:57 ntz osse: 0.7.990-1998-gbfff4f53
11:57 ntz what ?
11:57 ntz osse: you mean 1.6.8-$(git describe)
11:58 osse maybe git describe --tags
11:58 ntz 1.6.0-668-gbfff4f53 what ????? git tag say 1.6.8 is latest, and it is, we have in suse ofc 1.6.8 main version
11:59 ntz osse: ^^ perhaps they have broken tags ?
11:59 Gurkenglas joined #git
11:59 osse git describe will describe the current commit.
11:59 osse are you on a different branch? have you fetched recently?
11:59 ntz yeah, sure
12:00 ntz osse: for example this git://anongit.freedesktop.org/libmbim/libmbim reports 1.14.2 with git tag but describe --tags says 1.14.0-60-g0cd2280
12:01 ntz I don't catch how it works :(
12:01 JeroenT_ joined #git
12:01 southernirc joined #git
12:01 osse git describe has several options
12:02 osse and in that particular repo it seems they maintain many releases simultaneously
12:02 osse I just cloned it and I get 0.0.1-392-g0cd2280
12:02 ntz what o.O
12:02 osse anyway, git describe is *usually* how strings like that are produced
12:02 AndreasLutro $ git describe --tags origin/mbim-1-14
12:02 AndreasLutro 1.14.2-3-g9e6d5e6
12:02 AndreasLutro 1.14.2 just isn't on the master branch
12:03 ntz osse: okay, thanks ... so describe --tags should be preffered
12:03 ntz AndreasLutro: o.O
12:04 AndreasLutro same for the modemmanager you linked earlier: $ git describe --tags origin/mm-1-6
12:04 AndreasLutro 1.6.8-46-gfd47dd53
12:04 osse ntz: git log ^master 1.6.8 has many commits as well.... So master doesn't contain the 1.6.8 tags
12:04 osse *tag
12:05 roelmonnens joined #git
12:06 ntz osse: then I don't indeed catch how it works ... so what branch I'll be building when I do in root_dir `NOCONFIGURE=1 sh autogen.sh' ?
12:06 noidea joined #git
12:06 noidea joined #git
12:06 osse ntz: the branch you have checked out
12:06 ntz master
12:07 ntz but you've confused me with 14:02 < AndreasLutro> $ git describe --tags origin/mbim-1-14
12:07 osse this is the list of tags that are reachable from master: git tag --merged master
12:08 osse 1.6.8 is not on that list
12:08 masuberu joined #git
12:08 Cabanossi joined #git
12:08 osse seems only the mm-1.6 branch contains that tag
12:09 ntz osse: can you please use this ``git clone git://anongit.freedesktop.org/libmbim/libmbim libmbim-git''
12:10 osse sure
12:10 ntz AndreasLutro: is using for his example git://anongit.freedesktop.org/libmbim/libmbim and you use git://anongit.freedesktop.org/ModemManager/ModemManager
12:10 ntz it's for me confusing, I'm now in libmbim, it's same problem / same principle, thanks much
12:11 osse 'git describe' describes the CURRENT commit.
12:11 osse you are probably on the wrong branch
12:11 AndreasLutro it's the same answer - the 1.14.2 tag is not on the master branch, it's on the mbim-1-14 branch, so you'd need to check out that branch before `git describe --tags` shows you the "correct" version number
12:11 oxymoron93 yep checkout to origin/mbim-1-14
12:12 ntz okay
12:12 ntz thanks
12:12 str joined #git
12:13 synthroid joined #git
12:15 kurkale6ka joined #git
12:17 mentazoom joined #git
12:21 Yakusaji joined #git
12:26 jameser joined #git
12:27 robinb joined #git
12:27 marinin joined #git
12:28 Yakusaji joined #git
12:29 jeffreylevesque joined #git
12:30 JeroenT joined #git
12:31 str joined #git
12:32 jameser joined #git
12:34 kurkale61 joined #git
12:34 Tobbi joined #git
12:35 jameser_ joined #git
12:37 kurkale6ka joined #git
12:37 shdw joined #git
12:38 Guest38326 joined #git
12:39 Cabanossi joined #git
12:39 shdw- joined #git
12:41 gloomy joined #git
12:41 glad joined #git
12:43 foxpup left #git
12:45 chriys joined #git
12:45 bongjovi joined #git
12:46 overlord_tm joined #git
12:46 donnib osse: u around ?
12:46 navidr joined #git
12:48 Munt joined #git
12:48 dsdeiz_ joined #git
12:50 osse donnib: maybe
12:50 donnib ;)
12:50 donnib i got an issue with git subtree split which i wanted to ask about, maybe you had a hint for me
12:50 oskarkv joined #git
12:50 donnib I did git subtree split -P Folder/MyFolder -b onlyMyFolder to split a folder into a branch so i can move it to a separete repository
12:51 donnib i did that while beeing on master
12:51 donnib now when the operation is done i can see that split did not take last commit into the new branch although the changes in that commit are solely done to that path that i split. Any ideas what might cause that ?
12:51 donnib i should point that this lastcommit was a merge commit
12:52 osse Sorry, I've never used subtree.
12:52 donnib ok, np
12:52 bsanford joined #git
12:55 bronson joined #git
12:56 str joined #git
12:58 dviola joined #git
13:01 cdg joined #git
13:03 Yakusaji joined #git
13:06 stutter joined #git
13:07 cbd_ joined #git
13:07 whitby joined #git
13:07 jimsio joined #git
13:07 dre_ joined #git
13:08 robinb joined #git
13:09 Cabanossi joined #git
13:09 MadLamb joined #git
13:10 jeffreylevesque_ joined #git
13:11 YuGiOhJCJ joined #git
13:11 MadLamb One branch was accidentably merged and instead of rebasing and dropping the commit, some developer did a revert. The branch was merged to mainline, but now all the commits are in the mainline, so its not possible to merge/rebase. Is there a way to keep the same commits (even if with another identifier) and do another merge?
13:12 tobiasvl MadLamb: !faulty_merge
13:12 gitinfo MadLamb: 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
13:12 FrostCandy joined #git
13:12 tobiasvl hm maybe not so interesting now in hindsight
13:12 tobiasvl MadLamb: but you can revert the revert
13:13 Droolio joined #git
13:13 diqidoq joined #git
13:14 MadLamb I would like to avoid reverting the revert if there is another alternative.
13:17 dermoth joined #git
13:17 tobiasvl sure, you can probably do what you said then. give the old commits new SHAs and merge again
13:18 tobiasvl sounds find to me
13:18 tobiasvl look at the bottom of the website I linked you to to see how to do it with rebase --no-ff
13:25 igemnace joined #git
13:30 Zoxc joined #git
13:32 boombatower joined #git
13:35 dimi1947 joined #git
13:35 VladGh joined #git
13:39 jrm joined #git
13:39 jason237 joined #git
13:41 daguiam joined #git
13:41 rawste joined #git
13:41 freimatz joined #git
13:42 jeffreylevesque_ joined #git
13:44 deltab joined #git
13:45 Hello71 joined #git
13:46 Cavallari joined #git
13:46 brent__ joined #git
13:47 robinb joined #git
13:48 a3Dman joined #git
13:50 theoceaniscool joined #git
13:50 selckin joined #git
13:50 igemnace joined #git
13:51 aw1 joined #git
13:53 aw1 "theirs" and "ours" for git rebase is the opposite in git merge ... is that right?
13:53 stutter joined #git
13:54 osse yes
13:56 aw1 thanks osse :)
13:57 MattMaker joined #git
13:58 TheWild joined #git
13:58 TheWild hello
13:59 TheWild when I do "checkout -b newbranch", nothing will be changed in my files except .git directory, right?
14:01 aw1 [non-expert warning] yes TheWild
14:01 TheWild ok, thank you.
14:02 _ikke_ correct
14:04 em3 joined #git
14:04 ResidentBiscuit joined #git
14:08 Gurkenglas joined #git
14:09 njbair joined #git
14:11 kurkale61 joined #git
14:13 acidjnk2 left #git
14:13 acidjnk joined #git
14:15 chriys joined #git
14:16 Argylelabcoat joined #git
14:17 fission6 joined #git
14:17 tunekey joined #git
14:17 upperdeck joined #git
14:17 SkyPatrol_ joined #git
14:18 YuGiOhJCJ joined #git
14:19 kpease joined #git
14:19 Munt joined #git
14:22 kurkale6ka joined #git
14:24 ppang joined #git
14:24 upperdeck joined #git
14:25 Darcidride joined #git
14:25 robinb joined #git
14:25 phluks joined #git
14:26 maxb joined #git
14:28 marinin joined #git
14:28 thiago joined #git
14:29 phinxy joined #git
14:29 shdw joined #git
14:30 relipse__ joined #git
14:31 tvw joined #git
14:31 jstimm joined #git
14:31 mustmodify left #git
14:31 renarc joined #git
14:32 dreiss joined #git
14:34 phinxy left #git
14:34 upperdeck_ joined #git
14:34 em3 joined #git
14:34 relipse___ joined #git
14:35 em3 joined #git
14:36 netj joined #git
14:37 upperdec- joined #git
14:37 Lynxium joined #git
14:42 str joined #git
14:43 upperdeck joined #git
14:44 Yakusaji joined #git
14:45 robotroll joined #git
14:46 Phylock joined #git
14:46 montyboy joined #git
14:47 madewokherd joined #git
14:48 roelmonn_ joined #git
14:49 raijin joined #git
14:50 kurkale61 joined #git
14:51 h12o joined #git
14:52 kurkale6ka joined #git
14:53 snowkidind joined #git
14:53 Hello71 joined #git
14:56 bronson joined #git
14:57 chrisshattuck joined #git
15:00 multi_io_ left #git
15:01 gloomy joined #git
15:03 nadley joined #git
15:03 kurkale61 joined #git
15:04 ToBeFree joined #git
15:04 ntz left #git
15:04 ToBeFree joined #git
15:07 kurkale6ka joined #git
15:08 roelmonnens joined #git
15:09 Cabanossi joined #git
15:09 jimsio joined #git
15:15 kculpis joined #git
15:17 h12o joined #git
15:22 al-damiri joined #git
15:22 dimi1947 joined #git
15:24 TomyLobo joined #git
15:26 kpease joined #git
15:27 marinin joined #git
15:27 Murr joined #git
15:30 upperdeck joined #git
15:30 rkta joined #git
15:33 upperdeck_ joined #git
15:35 CheckDavid joined #git
15:36 netj joined #git
15:36 iNCoNFuN joined #git
15:37 Plouescat joined #git
15:38 blurp joined #git
15:38 blurp hi guys! how can I ignore upstream changes in only one dir of a specific clone instance?
15:39 blurp without replicating .gitignore through all of my instances
15:39 Gustavo6046 joined #git
15:39 thiago blurp: .gitignore wouldn't help you
15:40 brent__ joined #git
15:40 thiago what do you mean by "ignore"? If you meant just not pay attention, don't pay attention.
15:40 thiago do you mean you want to keep a certain directory at the state it was a previous point in time?
15:40 thiago git checkout thatpointintime -- path/to/dir
15:40 upperdeck joined #git
15:40 thiago those count as new changes for you
15:40 blurp thiago: I mean I do not want any more changes to that directory from now on
15:41 blurp but only for that specific cloned repo
15:41 thiago do that git checkout every time you update
15:42 PaulCapestany joined #git
15:42 thiago in fact, git reset --hard to throw away your change before you update; update; then do the checkout
15:42 thiago every time
15:43 blurp I see. thanks thiago
15:43 _ikke_ There is git update-index --skip-worktree, but use at your own risk
15:49 Garen joined #git
15:50 chriys joined #git
15:51 blurp _ikke_: that's it!
15:51 blurp thanks
15:53 diogenese joined #git
15:53 sammyo joined #git
15:53 rkta_ joined #git
15:53 sergio left #git
15:53 JanC_ joined #git
15:53 Lothe joined #git
15:54 durham joined #git
15:54 _sathed joined #git
15:54 stoopkid__ joined #git
15:55 holdsworth_ joined #git
15:55 SoniEx2 joined #git
15:55 ghormoon_ joined #git
15:55 olbat[m]1 joined #git
15:56 MikeChernev joined #git
15:56 em3 joined #git
15:56 Dyonisos joined #git
15:56 ertes joined #git
15:56 cverna joined #git
15:56 binary106 joined #git
15:56 sven^ joined #git
15:56 sven^ joined #git
15:56 SirFunk joined #git
15:57 Kazuto joined #git
15:57 Kazuto joined #git
15:57 Nothing4You joined #git
15:59 MattMaker joined #git
16:00 sergio joined #git
16:00 cyberz joined #git
16:00 roelmonnens joined #git
16:00 cbreak joined #git
16:01 upperdeck joined #git
16:01 opr joined #git
16:05 shdw- joined #git
16:06 manuelschneid3r joined #git
16:08 upperdeck joined #git
16:09 venmx joined #git
16:09 venmx_ joined #git
16:09 kurkale61 joined #git
16:10 dsdeiz_ joined #git
16:11 em3 joined #git
16:11 MattMaker joined #git
16:11 peepsalot joined #git
16:12 svm_invictvs joined #git
16:13 lukky513 joined #git
16:13 eshlox joined #git
16:19 gloomy joined #git
16:20 Tobbi joined #git
16:20 Achylles joined #git
16:21 synthroid joined #git
16:22 Nizumzen joined #git
16:24 Mattiaslndstrm joined #git
16:24 kurkale6ka joined #git
16:25 muep joined #git
16:26 dreiss joined #git
16:26 guardian joined #git
16:26 chriys joined #git
16:30 ResidentBiscuit joined #git
16:30 dimi1947 I have created a new repo and when I open GUI all the folders are listed as unstaged changes...why that?
16:31 ciscobandit joined #git
16:31 ahr3n joined #git
16:31 bremner !gui
16:31 gitinfo Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.
16:32 cdown joined #git
16:33 Murii|osx joined #git
16:34 upperdeck joined #git
16:36 jlewis joined #git
16:36 dimi1947 ok for git CLI
16:36 masuberu joined #git
16:37 dimi1947 after typing git status I see that all folders are untracked...what next?
16:37 crose joined #git
16:37 dviola joined #git
16:38 bremner if you want to track them, you can use git add, and git commit.
16:38 orbyt_ joined #git
16:38 bremner of course if you commit 4G VM images or equivalent, you will probably regret it
16:38 edvb joined #git
16:39 mkali joined #git
16:39 Cabanossi joined #git
16:39 dimi1947 bremmer: can git add work for so many folders at once
16:39 dimi1947 ???
16:40 bremner sure, it sounds like you need to follow a tutorial though
16:40 bremner !tutorial
16:40 gitinfo A list of interactive git tutorials: http://try.github.com
16:40 bremner see also man gittutorial
16:40 gitinfo the gittutorial manpage is available at https://gitirc.eu/gittutorial.html
16:40 durham joined #git
16:40 bremner "Importing a new project"
16:43 MattMaker joined #git
16:47 cagomez joined #git
16:51 thiago joined #git
16:51 royal_screwup21 joined #git
16:54 royal_screwup21 I have a heroku repo with like a 100 commits. Now I want to make a new github repo and sync it with the heroku repo. But I want it so that the github repo starts out fresh, with just 1 commit (the "initial commit"), and not 100 preceding commits that I used on my heroku repo. How do I accomplish this?
16:54 dsdeiz joined #git
16:54 dsdeiz joined #git
16:54 roelmonnens joined #git
16:55 LuksLux joined #git
16:56 _ikke_ Not without creating a new repo
16:56 svm_invictvs joined #git
16:57 bronson joined #git
16:57 TSS_ joined #git
16:58 royal_screwup21 I think a better question is: how do you sync a github repo with an existing heroku repo? Supposed it took me 100 commits to create 3 files on my heroku repo. Okay. Now I want to create a github that is synced with the heroku repo and has all of those 3 files BUT it shows that I've just made one commit
16:59 upperdeck joined #git
16:59 wintrytbs joined #git
16:59 dsdeiz_ joined #git
16:59 royal_screwup21 so the next commit I make on heroku will be the 101st commit on heroku and the 2nd commit on github
17:02 gargola joined #git
17:02 cjohnson No such thing, really
17:02 cjohnson You can take the files, discard the history, and commit the current state of things as a new commti in a new repo
17:02 cjohnson and then go from there
17:02 cjohnson but if you want two repos to actively be synced up you need them to share remotes which means when you push the 100 commits the other remote will see that
17:03 royal_screwup21 aw man
17:03 royal_screwup21 I made a lot of stupid commits on heroku, just to test things out
17:03 _ikke_ You can create a new root commit
17:03 cjohnson Just blow away that heroku repo with a new one
17:04 cjohnson whose first commit is the current state of hte old repo
17:04 royal_screwup21 ah good idea
17:04 royal_screwup21 thanks!
17:04 h12o joined #git
17:05 iNCoNFuN joined #git
17:08 rk joined #git
17:09 guardian joined #git
17:09 chriys joined #git
17:11 energizer joined #git
17:15 bitmod joined #git
17:15 thinkdoge joined #git
17:16 TSS_ joined #git
17:22 cverna joined #git
17:22 raijin joined #git
17:23 stutter joined #git
17:24 upperdeck joined #git
17:24 h12o joined #git
17:26 shdw joined #git
17:27 jesst joined #git
17:27 cdown joined #git
17:28 Murr joined #git
17:30 Borg- joined #git
17:30 abrcdbr joined #git
17:30 Noldorin joined #git
17:31 BSaboia joined #git
17:34 Gustavo6046 joined #git
17:34 robinb joined #git
17:36 ArcaTech joined #git
17:39 em3 joined #git
17:41 nixjdm joined #git
17:41 royal_screwup21 joined #git
17:41 em3 joined #git
17:42 upperdeck joined #git
17:43 Bobdude joined #git
17:44 cagomez joined #git
17:49 dsdeiz joined #git
17:49 dsdeiz joined #git
17:49 raynold joined #git
17:50 matoro joined #git
17:54 fission6 joined #git
17:54 King_Hual joined #git
17:55 orbyt_ joined #git
17:56 aw1 joined #git
17:58 pioto joined #git
17:59 pioto Hi, what's the best way to find the point where a branch diverged from a branch I'm merging in? I need to do some more complicated merge conflict resolution, and wanna view the changes made to some specific files on this branch
17:59 grawity git merge-base, with or without --fork-point
18:01 pioto grawity: perfect, thanks!
18:01 grawity also `git log <other>..HEAD` and `HEAD..<other>` for commits exclusive to one side
18:01 causasui joined #git
18:01 grawity e.g. `git log ..feature` or `git diff ..feature` to see what you're bringing in
18:01 jmccarthy left #git
18:02 Sasazuka joined #git
18:04 salamanderrake joined #git
18:07 h12o joined #git
18:11 _28_ria joined #git
18:12 maryo joined #git
18:13 gargola joined #git
18:16 seds joined #git
18:18 cdown joined #git
18:19 stutter joined #git
18:24 Ldip joined #git
18:24 holodoc joined #git
18:29 InfoTest joined #git
18:33 njbair joined #git
18:33 upperdeck joined #git
18:35 maestroj_ left #git
18:35 Achylles joined #git
18:36 AaronMT joined #git
18:39 Brain joined #git
18:39 joeyy joined #git
18:44 phelix joined #git
18:44 prg318 joined #git
18:45 phelix I have these files that are caught in my branch tha are to large to push. I have deleted them from the files system but it keeps telling me they are to big when I try to commit.. How in the world can I get them out?
18:45 fission6 joined #git
18:46 _ikke_ phelix: commit != push
18:46 _ikke_ phelix: what does git status say?
18:46 phelix Well i deleted them and then tried to commit and push. The push keeps getting rejected about these file sizes but they don't even exist anymore.. how can I get them out?
18:47 oldnsalty joined #git
18:47 _ikke_ phelix: deleting the files and comitting that does not remove the files from older commits
18:47 _ikke_ phelix: You need to make sure they are removed from the previous commits too
18:47 phelix How can I do that?
18:47 phelix I am kind of a git newbie..
18:47 phelix not sure how to get into more complicated stuff.
18:48 h12o joined #git
18:48 _ikke_ phelix: https://help.github.com/articles/removing-files-from-a-repository-s-history/
18:49 _ikke_ phelix: (the steps described at the bottom is similar to what you already tried)
18:49 Borg- file size hooks :) nice.. smart.. they should be on local repo too
18:50 phelix _ikke_,  when I try to use git rm --cached file it tells me that it doesn't match any files
18:50 svm_invi_ joined #git
18:51 Borg- phelix: when did you added that large files?
18:51 Borg- how many comits away?
18:51 phelix I have no idea. I think it was a while ago
18:51 Borg- thats bad...
18:51 phelix yea.
18:51 Borg- but.. not that bad :)
18:51 Borg- what I would do.. is
18:51 Borg- tag your current head.. with something like temp
18:52 Borg- git reset --hard to commit just before add of those files
18:53 Borg- and then.. cherry pick every single commit again.. one by one.. eventually.. removing that big files... and commiting
18:53 Borg- other option is... if you dont care much about history
18:53 Mattiaslndstrm joined #git
18:53 Borg- git reset --hard <commit_before_big_files_add>
18:53 Borg- git merge --squash temp
18:53 Borg- remove big files.. commit...
18:53 phelix Yea. I like where my current system is. I might just delete the repository and re create it.
18:53 Borg- woot?
18:53 Cabanossi joined #git
18:53 Borg- its not necessary
18:54 Borg- just do backup...
18:54 _ikke_ !bfg
18:54 Borg- just in case
18:54 gitinfo A tool designed to remove large files, or passwords from history: https://rtyley.github.io/bfg-repo-cleaner/ (!rewrite applies)
18:54 durham_ joined #git
18:54 phelix k let me check that out. Thanks
18:54 Borg- damn.. I play way too much DOOM ;)
18:55 _ikke_ hehe
18:55 omnigoat joined #git
18:57 ELFrederich joined #git
18:57 nixjdm joined #git
18:57 robotroll joined #git
18:57 ELFrederich is there an equivalent of worktree that will just get me a working copy of a particular tree and put it somewhere, but without the destination needing to be a git repo and without requiring a branch?
18:58 bronson joined #git
18:58 Borg- worktree. no. but you can use git archive
18:58 osse ELFrederich: git archive | tar -C somewhere -xf -
18:58 cagomez I'm trying to setup CODEOWNERS. how do I match all all files within a directory? is it simply `foo/` or is it `foo/*`? https://github.com/blog/2392-introducing-code-owners
18:59 ELFrederich osse, cool.  I'll try it.  So something like "git archive HEAD^^^ | tar -C /tmp/yesterday -xf -"
18:59 osse sure
19:01 fission6 joined #git
19:01 em3a joined #git
19:02 manuelschneid3r joined #git
19:04 em3a joined #git
19:06 Cavallari joined #git
19:06 Gustavo6046_ joined #git
19:06 vals_ joined #git
19:10 Calinou joined #git
19:11 mescalinum joined #git
19:11 MadLamb joined #git
19:11 gargola joined #git
19:12 ahr3n joined #git
19:13 Gustavo6046 joined #git
19:18 cagomez joined #git
19:18 preyalone joined #git
19:19 Mattiaslndstrm joined #git
19:20 bronson joined #git
19:21 joeyy joined #git
19:24 h12o joined #git
19:26 mostlybadfly joined #git
19:27 misosoup joined #git
19:28 Murr joined #git
19:30 synthroid joined #git
19:31 hagridaaron joined #git
19:32 belmoussaoui_ joined #git
19:35 hfp_work joined #git
19:36 blaeks joined #git
19:38 Cabanossi joined #git
19:39 Fallen0223 joined #git
19:40 Satsuki joined #git
19:40 energizer joined #git
19:41 wintrytbs joined #git
19:44 kadoban joined #git
19:44 Calinou hi, I'm doing a Git rebase of many commits, how do I ignore all conflicting files coming from the branch I'm rebasing from?
19:44 Calinou without solving the conflict in each commit manually
19:45 osse git checkout --ours {the files here}
19:47 griffindy joined #git
19:49 Calinou hmm, when should I use this?
19:49 cagomez joined #git
19:49 Calinou after starting the rebase, before…?
19:49 osse whenever git rebase stops because of a conflict
19:49 Calinou I did that, then added the file, then did "git rebase --continue", it's telling me there are no changes
19:50 osse git rebase --skip then
19:50 Solvaring joined #git
19:51 Calinou that doesn't work, ie. it's asking me for every commit :|
19:51 Calinou I have over 30 of them
19:51 Smokes_ joined #git
19:51 phluks joined #git
19:52 osse Calinou: git rebase -X ours ...  etc
19:52 osse try that
19:52 Plouescat joined #git
19:53 Calinou this seems to work! it takes quite a bit of time to apply, though
19:54 gopar joined #git
19:55 Calinou well crap, it didn't
19:55 Calinou and since I force-pushed my changes, they're gone :DDDD
19:55 osse !reflog
19:55 gitinfo The git reflog (`git log -g`) records the SHAs of your HEADs for 2+ weeks. `git checkout -b myrestore OLDSHA` and `git reset --hard OLDSHA` will relink to that state via a new and current branch respectively, see http://sethrobertson.github.com/GitFixUm/ for full details.  WARNING: reset --hard will trash any uncommitted changes!  Visualize with: gitk --all --date-order `git log -g --pretty=%H`
19:55 Calinou phew, I forgot about the reflog
19:55 Smokes_ joined #git
19:55 Calinou despite me using it daily
19:56 nixjdm joined #git
20:00 Gurkenglas joined #git
20:07 chrisshattuck joined #git
20:07 tsia joined #git
20:09 Calinou this will do, thanks :)
20:09 chrisshattuck joined #git
20:10 whitby joined #git
20:12 Sasazuka__ joined #git
20:12 chriys joined #git
20:16 h12o joined #git
20:16 parsnip i guess `git lol` doesn't indicate if you have two initial commits?
20:17 osse you can spot it if you look carefully
20:17 osse the easiest is to make git tell you:   git rev-list --count --max-parents=0 --all
20:18 parsnip right, that shows two, but i don't see `git lol` indicating
20:19 orbyt_ joined #git
20:20 diqidoq joined #git
20:21 hagridaaron joined #git
20:22 osse then you have to look more carefully
20:23 osse does rev-list print to also if you replace --all with HEAD?
20:23 osse print two
20:27 jesst joined #git
20:28 cdg joined #git
20:30 cdg joined #git
20:30 cagomez joined #git
20:31 ozcanesen joined #git
20:31 cdg joined #git
20:32 parsnip oh dang, just rebased. one initial now. what was idea with HEAD?
20:32 BSAlb joined #git
20:32 osse count roots commit reachable from the current commit, not from wherever
20:33 theoceaniscool joined #git
20:33 ToBeCloud joined #git
20:35 mattcen joined #git
20:36 hagridaaron joined #git
20:36 parsnip oh interesting, i wonder if git lol would show those
20:36 parsnip what i mean is, maybe it would "connect" them
20:37 tsia joined #git
20:38 Oatmeal joined #git
20:41 h12o joined #git
20:43 Kiruwa Is there any way to easily get a "short relative date" in a log format?
20:43 Kiruwa IE, "1w" instead of "1 week ago", "3d" instead of "3 days ago"
20:44 Kiruwa I was playing around with %<(3,trunc), but it insists on replacing with dots instead of JUST truncating
20:45 disque joined #git
20:46 diogenese joined #git
20:50 cdg_ joined #git
20:52 seds joined #git
20:53 Cabanossi joined #git
20:57 nixjdm joined #git
21:00 jeffreylevesque joined #git
21:01 ahr3n joined #git
21:05 ZyX-I joined #git
21:05 doener joined #git
21:07 seds joined #git
21:07 svm_invictvs joined #git
21:14 thiago joined #git
21:15 zzz_ joined #git
21:16 Tobbi joined #git
21:17 gopar joined #git
21:17 petersaints joined #git
21:17 Nizumzen joined #git
21:20 h12o joined #git
21:23 tsia joined #git
21:24 bayman joined #git
21:24 gopar joined #git
21:24 manuelschneid3r joined #git
21:25 bayman left #git
21:25 cagedwisdom joined #git
21:27 re1 joined #git
21:28 Murr joined #git
21:28 gopar joined #git
21:29 SkyPatrol joined #git
21:30 petersaints joined #git
21:33 hagridaaron joined #git
21:34 sandmangil joined #git
21:34 hash-z3r0 joined #git
21:35 elsevero joined #git
21:35 robinb joined #git
21:40 h12o joined #git
21:41 gopar joined #git
21:42 wilbert joined #git
21:44 Guest76 joined #git
21:45 marinin joined #git
21:45 hash-z3r0 joined #git
21:49 energizer joined #git
21:52 seds joined #git
21:53 Guest76 left #git
21:53 Guest76 joined #git
21:53 cdown joined #git
21:55 m0viefreak joined #git
21:56 str joined #git
21:57 nixjdm joined #git
22:01 h12o joined #git
22:01 dsdeiz_ joined #git
22:05 thiago joined #git
22:06 seds joined #git
22:11 grombles joined #git
22:11 h12o joined #git
22:15 bitmod joined #git
22:15 nahtnam joined #git
22:15 bitmod joined #git
22:17 roelmonnens joined #git
22:18 xaxxon joined #git
22:19 seds joined #git
22:19 baxx joined #git
22:19 Raed joined #git
22:19 baxx is there an issue with using https and git or something?
22:20 baxx for the remote url
22:20 _ikke_ not in general
22:20 njbair joined #git
22:20 _ikke_ It should just work
22:20 _ikke_ git uses libcurl to communicate over http(s)
22:20 baxx so i should be able to just use the repos url from the browser right?
22:21 _ikke_ not necesserily
22:21 osse that depends on the host. probably
22:21 _ikke_ The repo url might be different from the webinterface
22:21 baxx what do you mean ? I have this locally on my system... hrm, ok
22:22 baxx where's the repo url ?
22:22 osse what we mean is that the URL to the nice looking repo page might not be clonable. but it usually is
22:22 baxx well idk how to check whether that's the case or not though. Unless i just create a new project or something
22:22 morgoth88 joined #git
22:23 osse you can check it by cloning?
22:23 cagomez joined #git
22:23 baxx the link to clone looks the same as the one in the browser url bar with the exception of .git at the end
22:24 baxx and it doesn't work - just tested, hrm... ok
22:25 cagomez joined #git
22:25 thiago joined #git
22:25 baxx i've not had this before - what does this ( likely ) mean?
22:27 baxx i'm getting  " fatal: Unable to find remote helper for https "
22:29 foxkit joined #git
22:29 baxx https://stackoverflow.com/a/13018777/3130747
22:30 baxx uninstall git, install curl-devel , reinstall git and try again?
22:30 baxx i installed git from a PPA, and it's worked with bitbucket, so I'm not sure why it's not working with github here
22:31 tunekey joined #git
22:31 drodger joined #git
22:32 bronson joined #git
22:35 MattMaker joined #git
22:39 baxx ok i think it's working with ssh fine - so i'm guessing i have done something stupid with the http settings or something ¯\_(ツ)_/¯
22:39 Cabanossi joined #git
22:45 kerrick joined #git
22:47 parsnip does `git fetch --all` grab not only all distant branches, but other repos too? or is `group` needed for that?
22:47 parsnip i don't think man page "--all\n\tFetch all remotes." makes it obvious.
22:48 parsnip oh wait, each repo is _a_ remote, so that would suggest plural, so yes, all repos, right?
22:48 dsdeiz joined #git
22:48 dsdeiz joined #git
22:49 theoceaniscool joined #git
22:50 fury joined #git
22:50 orbyt_ joined #git
22:51 foxkit joined #git
22:52 thiago joined #git
22:56 Elon_Satoshi joined #git
22:57 nixjdm joined #git
22:58 Paraxial joined #git
22:59 h12o joined #git
22:59 seds joined #git
23:02 ahr3n joined #git
23:02 Bishop_Bautista joined #git
23:05 robinb joined #git
23:05 safe joined #git
23:08 feodoran_ joined #git
23:20 wilbert joined #git
23:22 hash-z3r0 joined #git
23:22 peepsalot joined #git
23:24 Cabanossi joined #git
23:28 Murr joined #git
23:28 dsdeiz joined #git
23:32 salamanderrake joined #git
23:34 abrcdbr joined #git
23:34 cdg joined #git
23:41 musca joined #git
23:44 BoucyKnightMishu joined #git
23:51 h12o joined #git
23:51 ertes joined #git
23:56 zulutango joined #git
23:59 justan0theruser joined #git

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