Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-05-27

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

All times shown according to UTC.

Time Nick Message
00:02 tristanp_ joined #git
00:02 unforgiven512 joined #git
00:03 zivester joined #git
00:03 unforgiven512 joined #git
00:03 jstimm joined #git
00:04 unforgiven512 joined #git
00:04 unforgiven512 joined #git
00:05 unforgiven512 joined #git
00:06 boudicca joined #git
00:09 toothe joined #git
00:09 jstimm_ joined #git
00:09 toothe so, I'm confused about subtrees. Are they basically different git repositories inside of another?
00:10 bronson joined #git
00:10 rkazak joined #git
00:10 mwhooker joined #git
00:11 toothe why/when would you use subtrees?
00:11 MattMaker joined #git
00:14 MattMaker joined #git
00:15 JanC_ joined #git
00:15 jooge joined #git
00:17 nicechap_ joined #git
00:19 kellytk I am also interested in best practices for nesting repositories
00:20 Impaloo joined #git
00:21 mwhooker joined #git
00:21 tristanp joined #git
00:21 Impaloo joined #git
00:23 eletuchy joined #git
00:28 ojacobson joined #git
00:28 bjpenn joined #git
00:31 tomog999 joined #git
00:32 prsn joined #git
00:34 tristanp_ joined #git
00:34 Impaloo joined #git
00:36 Impaloo joined #git
00:38 tristanp joined #git
00:41 interne7y joined #git
00:44 whomp_ joined #git
00:46 fuchstronaut joined #git
00:46 Eugene !subrepos
00:46 gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
00:47 Eugene Submodules are a good fit for "I want to import a specific version of a library"
00:47 tristanp joined #git
00:47 Eugene Ignore-it is a good fit for components that don't need to be tightly coupled, or just happen to live inside of some other git repo
00:48 Eugene Subtree I'm not entirely sure of the use-case on.... things you want to develop in two places at once, I guess?
00:48 Eugene There's a !book chapter on it that goes into more detail
00:48 gitinfo There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
00:49 eletuchy joined #git
00:50 inflames joined #git
00:52 bronson joined #git
00:56 DFrostByte joined #git
00:56 inflames joined #git
00:59 s00pcan joined #git
01:01 primeos left #git
01:02 tristanp_ joined #git
01:06 kellytk I use the first
01:07 kellytk Subtrees could be useful for one of my repositories, thanks for the tip Eugene
01:09 fracting joined #git
01:10 smithbone joined #git
01:14 DARSCODE kellytk, another thing you might want to look in to is git-subrepo. It has been favored lately
01:16 tristanp joined #git
01:16 PaulCapestany joined #git
01:20 MattMaker joined #git
01:21 chachasmooth joined #git
01:24 awsation joined #git
01:26 linduxed joined #git
01:27 jaguarmagenta joined #git
01:31 tristanp_ joined #git
01:31 kellytk Thanks DARSCODE
01:32 astrofog joined #git
01:33 nu11_byte joined #git
01:33 kellytk I'm new to Git and getting a basic system in place.  What do you think of deleting the master branch and instead using production?  In my mind it maps to production deployment, and the develop branch maps to development deployment
01:34 bremner name your branches how you like
01:39 sagerdearia joined #git
01:39 zivester joined #git
01:39 doebi joined #git
01:46 CJKinni`` joined #git
01:48 ilbot3 joined #git
01:48 Topic for #git is now Welcome to #git, the place for git help and zerg rushes | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.8.2 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | YOU MUST SUBMIT ADDITIONAL JOKES https://madeitwor.se/git-jokes
01:49 pks joined #git
01:49 ojacobson "connect"?
01:49 CandyWife yes. i'm taking this online course and the instructor was doing something in the command line then he said "okay let's open sublime text and save this page"
01:49 CandyWife i opened sublime text and had no idea where he got the page from to save it
01:49 CandyWife it was like git was talking to sublime text
01:50 eletuchy joined #git
01:50 CandyWife but he did not explain this. maybe it's because i have git on a shell instead of on my laptop
01:50 hexagoxel joined #git
01:51 Tanger_ joined #git
01:51 CandyWife i tried this that i found on stackoverflow "21:19 -!- mau [~BL4CK@wtfux-CC45DEDD.dlth.qwest.net] has quit [Ping timeout: 241 seconds]
01:51 CandyWife 21:19 -!- mau [~BL4CK@wtfux-CC45DEDD.dlth.qwest.net] has quit [Ping timeout: 241 seconds]
01:51 CandyWife wtf
01:51 CandyWife i'm sorry that's not what i copied
01:52 CandyWife git config --global core.editor "subl"
01:52 CandyWife it didn't give any messages though and when i opened sublime text the page still wasn't there to save or commit or whatever.
01:52 fracting joined #git
01:53 bronson joined #git
01:54 ochorocho__ joined #git
01:57 mizu_no_oto joined #git
01:58 tristanp joined #git
01:59 phroa CandyWife: if you use "git commit" in the shell now, it should bring up a commit message editor in sublime. I assume that's what you're trying to do currently
02:00 dumkunt joined #git
02:00 CandyWife i got this message fatal: Not a git repository (or any of the parent directories): .git
02:00 phroa has the instructor gone over the 'cd' and 'ls' commands?
02:01 CandyWife yes
02:01 phroa you'll probably have to 'cd' to wherever the lesson file repository is
02:02 eletuchy joined #git
02:03 CandyWife okay i did the directory thing and it told me to set my user name and email, which i have done
02:03 CandyWife nothing is happening in sublime text though
02:04 ensyde joined #git
02:04 phroa (this is all assuming you're supposed to be creating a commit at this step in the lesson plan; the plan and any videos are probably way more accurate than my guessing) try 'git commit' again
02:04 dumkunt joined #git
02:04 CandyWife okay it worked in git but nothing happened in sublime text. i guess maybe it is some kind of common knowledge i am just missin
02:04 ojacobson sublime doesn't track your console (in fact, can't)
02:05 scottbot84 joined #git
02:05 ojacobson if you want to open an editor from a given directory, go there first, then open an editor
02:05 ojacobson 'subl .' will open sublime in the current directory, for example
02:05 kellytk left #git
02:05 ojacobson (assuming you have the 'subl' command installed, which... we'll see)
02:05 CandyWife it says it is not found
02:05 CandyWife okay
02:05 CandyWife i guess that thing from stackoverflow didn't work or i did it wrong
02:05 zivester joined #git
02:06 ojacobson Well, since you're paying your prof to teach this stuff, I'd say park this for now and go talk to your prof in the morning
02:06 CandyWife oh it's not someone i'm paying. i'm taking a class at coursera on front end web stuff
02:06 tristanp_ joined #git
02:06 CandyWife the person who did the videos for setting up the development environment did not explain how he got his thing to work but
02:06 CandyWife thank you guys for trying to help
02:07 CandyWife maybe i am not cut out for this kind of thing
02:07 phroa !git_for_designers
02:07 phroa mm, what was it
02:07 phroa !designers
02:07 gitinfo 'Git for Web Designers' is a quick "Who why what when how" on VCS and git: http://www.webdesignerdepot.com/2009/03/intro-to-git-for-web-designers/
02:07 CandyWife thank you
02:08 phroa !books
02:08 gitinfo [!book] There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
02:08 CandyWife thank you phroa
02:08 mwleeds joined #git
02:09 chachasmooth joined #git
02:13 DFrostByte left #git
02:13 Goplat joined #git
02:15 xall_ joined #git
02:16 a_thakur joined #git
02:17 tristanp joined #git
02:20 linuxmint joined #git
02:21 hexagoxel joined #git
02:22 aidalgol joined #git
02:25 johnny56 joined #git
02:26 tristanp_ joined #git
02:26 ash_workz joined #git
02:31 Dougie187 joined #git
02:32 bjpenn joined #git
02:32 tristanp joined #git
02:34 ajf- joined #git
02:37 ajf__ joined #git
02:37 fracting joined #git
02:37 tristanp_ joined #git
02:38 eletuchy joined #git
02:39 nfk joined #git
02:41 PaulCape_ joined #git
02:42 interne7y joined #git
02:43 david007 joined #git
02:44 rchavik joined #git
02:47 tristanp joined #git
02:47 mehola joined #git
02:48 Mobutils joined #git
02:48 Celelibi joined #git
02:48 fuchstronaut joined #git
02:55 Chunk2 joined #git
02:57 dermoth__ joined #git
02:58 dwmw2_gone joined #git
02:59 ryez joined #git
03:01 fscala joined #git
03:03 aavrug joined #git
03:03 arescorpio joined #git
03:03 PaulCapestany joined #git
03:04 aavrug joined #git
03:04 scoobertron joined #git
03:04 tristanp_ joined #git
03:05 a_thakur joined #git
03:06 StuartMI joined #git
03:07 peeps[lappy] joined #git
03:09 boudiccas joined #git
03:11 eletuchy joined #git
03:12 b1tchcakes joined #git
03:12 kadoban joined #git
03:14 peeps[lappy] if core.autocrlf=input and core.safecrlf=true, then git will never alter the line endings, it will only cause it to error out if i try to check in files containing CRLF?  i'm on linux
03:16 peeps[lappy] just trying to confirm that i understand these settings
03:18 ygeLN joined #git
03:20 k3rn31 joined #git
03:22 tristanp joined #git
03:23 Dougie187 left #git
03:24 sangy joined #git
03:26 ImJune joined #git
03:27 steven_a_s joined #git
03:28 jaguarmagenta joined #git
03:30 bjpenn joined #git
03:33 tristanp_ joined #git
03:37 wviana joined #git
03:40 adrian_broher joined #git
03:42 mwhooker joined #git
03:46 baba_ joined #git
03:46 k3rn31 joined #git
03:47 tristanp joined #git
03:48 maestrojed joined #git
03:48 rneco joined #git
03:49 bjpenn joined #git
03:51 adrian_broher joined #git
03:52 adrian_broher joined #git
03:55 ochorocho__ joined #git
03:55 dreiss joined #git
03:57 toogley joined #git
03:59 d4rklit3 joined #git
03:59 arooni joined #git
04:00 PaulCape_ joined #git
04:00 xall joined #git
04:01 boudicca joined #git
04:02 tristanp_ joined #git
04:06 settermjd joined #git
04:07 clemf joined #git
04:09 noark9 joined #git
04:09 sargas joined #git
04:11 bjpenn joined #git
04:11 a_thakur joined #git
04:12 MutantMahesh joined #git
04:13 settermjd joined #git
04:13 tristanp joined #git
04:15 muthu joined #git
04:22 settermjd joined #git
04:22 CrypticGator joined #git
04:28 tristanp_ joined #git
04:36 Cabanossi joined #git
04:39 boudiccas joined #git
04:39 SimonNa joined #git
04:40 tinajohnson___ joined #git
04:40 LeBlaaanc joined #git
04:42 pabs3 joined #git
04:43 pabs3 is there any way to make git commit -v the default?
04:44 jaguarmagenta joined #git
04:44 phroa git config --global alias.com "commit -v"
04:44 phroa $ git com
04:46 tristanp joined #git
04:47 pabs3 doesn't seem to work when I do git commit
04:48 phroa use 'git com'
04:49 fuchstronaut joined #git
04:52 PaulCapestany joined #git
04:53 ilmgb joined #git
04:53 MutantMahesh joined #git
04:55 tristanp_ joined #git
04:57 eletuchy joined #git
05:00 LeBlaaanc joined #git
05:02 gfixler joined #git
05:03 pabs3 I would rather modify the git commit command. I'll make a shell wrapper
05:03 efco joined #git
05:07 pabs3 hmm, that doesn't work either
05:12 MutantMa_ joined #git
05:12 ochorocho__ joined #git
05:13 boudiccas joined #git
05:15 tristanp joined #git
05:16 subhojit777 joined #git
05:16 hahuang65 joined #git
05:17 PaulCape_ joined #git
05:17 ryez joined #git
05:17 rscata joined #git
05:18 harish_ joined #git
05:19 boudicca joined #git
05:22 afuentes joined #git
05:24 boudiccas joined #git
05:25 jstimm_ joined #git
05:27 a_thakur joined #git
05:27 hahuang65 joined #git
05:28 qt-x joined #git
05:29 nuge joined #git
05:31 nilsi joined #git
05:34 boudicca joined #git
05:34 tristanp_ joined #git
05:35 harish_ joined #git
05:36 bjpenn joined #git
05:38 bronson joined #git
05:39 SwiftMatt joined #git
05:39 boudiccas joined #git
05:42 PaulCapestany joined #git
05:43 t2mkn joined #git
05:44 sauvin joined #git
05:48 settermjd joined #git
05:50 ThomasLocke joined #git
05:51 tristanp joined #git
05:51 eletuchy joined #git
05:54 mda1 joined #git
05:56 mariuscc joined #git
05:58 lindenle joined #git
06:00 ImJune joined #git
06:00 jceb joined #git
06:00 Kaisyu joined #git
06:02 josuebrunel joined #git
06:03 rkazak joined #git
06:03 tristanp_ joined #git
06:03 freimatz joined #git
06:03 a_thakur_ joined #git
06:05 boudicca joined #git
06:06 ImJune joined #git
06:07 ImJune joined #git
06:08 ciampix joined #git
06:08 ImJune joined #git
06:09 a_thakur joined #git
06:09 PaulCape_ joined #git
06:10 bjpenn joined #git
06:10 ImJune joined #git
06:11 realz joined #git
06:12 robotroll joined #git
06:13 Macaveli joined #git
06:14 neanderslob joined #git
06:16 mariuscc joined #git
06:17 tristanp joined #git
06:17 Psychiatrist joined #git
06:18 ochorocho__ joined #git
06:19 zeroed joined #git
06:19 zeroed joined #git
06:22 Fijit joined #git
06:24 diogenese joined #git
06:24 radsy joined #git
06:25 PCatinean joined #git
06:28 fwdit joined #git
06:31 boudiccas joined #git
06:31 radicalbiscuit joined #git
06:32 SteffanW joined #git
06:33 radicalbiscuit Hey all. I was in the middle of a merge when I remembered I was in a detached head. So I created a branch and now git has forgotten it was in the middle of a merge. I still have all the changes, of course, but it won't register as a merge commit.
06:33 maestrojed joined #git
06:34 radicalbiscuit Anything I can do to restore the merge-ness before committing?
06:34 Xano joined #git
06:34 boudicca joined #git
06:37 circ-user-plJrZ joined #git
06:37 dreiss joined #git
06:37 tristanp_ joined #git
06:37 PaulCapestany joined #git
06:38 LeMike joined #git
06:40 Sceorem joined #git
06:42 yuhlw joined #git
06:42 hanthings joined #git
06:43 interne7y joined #git
06:43 lucido-cwl joined #git
06:45 eletuchy joined #git
06:47 jstimm_ joined #git
06:48 fees joined #git
06:50 fuchstronaut joined #git
06:52 lamppid joined #git
06:52 MattMaker joined #git
06:52 zeroed joined #git
06:54 akushner joined #git
06:54 MattMaker joined #git
06:57 mda1 joined #git
07:00 JeroenT joined #git
07:00 jagob joined #git
07:00 akushner joined #git
07:02 Psychiatrist joined #git
07:02 PaulCape_ joined #git
07:03 mda1 joined #git
07:03 tristanp joined #git
07:04 artemisart joined #git
07:04 _ikke_ Already left :-(
07:05 mda1 joined #git
07:05 osse _ikke_: what would your advice have been?
07:06 _ikke_ Depends, if the index still had the merge stages, then I think creating .git/MERGE_HEAD would fix it
07:06 Neon1024x joined #git
07:06 tristanp_ joined #git
07:07 Alienpruts joined #git
07:07 osse hmm, that could work
07:08 encod3 joined #git
07:08 osse i interpreted is as ut my interpretation was that the was already done resolving conflicts and everything and the only remaining thing to do was to perform the commit. In that case I would have made the commit to be safe, then potentially rewritten ti with another parent
07:08 osse but I must admit I doubt the premise. will checkout -b ruin everything?
07:08 _ikke_ well, even then creating MERGE_HEAD would be easier
07:09 _ikke_ if that exists, the new commit wil have it as an extra parent
07:09 tristanp joined #git
07:10 osse TIL
07:10 sanketdg joined #git
07:10 TomyWork joined #git
07:11 trista___ joined #git
07:13 nnyk_ joined #git
07:14 tristanp joined #git
07:15 suy joined #git
07:15 themill joined #git
07:16 zapb_ joined #git
07:16 languitar joined #git
07:16 Stummi joined #git
07:17 osse dank. you're right
07:18 mda1 joined #git
07:18 osse I was a bit confused since Git put "it appears your'e committing a merge..." into the editor. But it does that for actual proper merges too
07:19 monokrome joined #git
07:19 MattMaker joined #git
07:21 vjacob joined #git
07:22 prsn joined #git
07:22 osse and yes,checkout -b ruins everything
07:23 SimonNa joined #git
07:23 vjacob joined #git
07:24 kurkale6ka joined #git
07:25 MattMaker joined #git
07:26 pantsman- joined #git
07:29 Gonzo89 joined #git
07:30 psacrifice joined #git
07:32 PaulCapestany joined #git
07:32 GavinMagnus joined #git
07:32 GavinMagnus left #git
07:33 ahmedelgabri joined #git
07:35 JeroenT joined #git
07:36 Balliad joined #git
07:38 Kicer86 joined #git
07:39 willhunt joined #git
07:39 bronson joined #git
07:39 eletuchy joined #git
07:41 tristanp_ joined #git
07:43 GavinMagnus joined #git
07:43 ToBeCloud joined #git
07:43 kami joined #git
07:44 kami Good morning
07:44 tango_ joined #git
07:44 zincrokx joined #git
07:46 jceb joined #git
07:46 tango_ joined #git
07:47 kami I receive a gnutls error when fetching from a https repo: 'GnuTLS recv error (-110): The TLS connection was non-properly terminated.'
07:47 tango_ joined #git
07:47 Hobbyboy joined #git
07:47 osse Speak for yourself. I have a terrible headache, I'm bankrupt and I have pasta sauce stains on my shirt
07:47 tristanp joined #git
07:47 osse Does it happen only with a specific repo? or any repo over https ?
07:48 kami osse: it is one repo out of several on a Phabricator server.
07:48 kami Others seem to have had such an issue in the past
07:49 kami But it couldn't be reproduced: https://secure.phabricator.com/T10296
07:49 kami I have a verbose log of gnutls which I can paste in a second.
07:49 kami (I have to remove some customer-related info)
07:51 lite__ joined #git
07:51 kami The problem is a GET /diffusion/HAAS/haas.git/info/refs?service=git-upload-pack where the response leads to 'Fatal error: The TLS connection was non-properly terminated'
07:51 Xano joined #git
07:54 kami http://pastebin.com/7mJPfSeq
07:55 stratos joined #git
07:55 Rish joined #git
07:56 johnchalekson joined #git
07:57 m0viefreak joined #git
07:57 osse this is above my pay grade
07:58 Mobutils joined #git
07:58 FeyFre joined #git
07:59 bjpenn joined #git
07:59 FeyFre guys. One big question: HOW to add to repo empty directory??? (Answer impossible is not acceptable)
07:59 PaulCape_ joined #git
08:03 goTAN joined #git
08:03 bruce_lee joined #git
08:04 bongjovi joined #git
08:04 tristanp joined #git
08:05 mda1 joined #git
08:08 kami osse: do git developers hang around in this channel?
08:09 josuebrunel joined #git
08:09 Ch4rAss joined #git
08:10 LeMike joined #git
08:10 Alienpruts joined #git
08:10 ErrorHead joined #git
08:10 osse kami: the top brass don't afaik. there is #git-devel but it's very quiet. msot people there are here too I think.
08:11 osse FeyFre: if impossible is not acceptable then you have to write a patch for git
08:11 kami osse: when I think it over, I guess the better channel would be one which is related to gnutls.
08:11 kami osse: thank you
08:12 _pix joined #git
08:12 JeroenT joined #git
08:13 mikecmpbll joined #git
08:14 vjacob joined #git
08:14 FeyFre osse: well, according to google search result, my question is not new, and I suspect such patches was written and herebely rejected ;)
08:14 nnyk_ joined #git
08:14 osse that could be
08:15 osse the least bad answer is to add a "dummy" file there. a common name is .gitkeep, which makes the reason it's there more apparent.
08:15 tristanp_ joined #git
08:15 osse or you could name it .gitignore. It might even come in handy later
08:16 prsn joined #git
08:17 FeyFre well, the problem is it is unknown how build(or other) tools will complain. If vcs causes problem to build process, that is probably bad vcs(no offence)
08:17 testild joined #git
08:17 tristanp joined #git
08:19 osse if a build system is unable to create a directory that is probably a bad build system (no offence)
08:19 osse :D
08:19 Psychiatrist joined #git
08:19 FeyFre that is source directory, not target :)
08:19 ochorocho__ joined #git
08:20 osse if it's a source directory but empty, how is it useful?
08:20 trista___ joined #git
08:21 FeyFre a kind of flag directory. if it present, toolchain will activate more specific build steps...
08:21 osse as for patches. I've found two patch disussions on the mailing list. from what I can tell they're weren't rejected; there just wasn't enough interest
08:21 nuge any gitkraken users here by chance?  wondering if anyone knows how to resolve fetch/pull errors with github
08:21 a_thakur joined #git
08:22 Jellyg00se joined #git
08:22 tristanp_ joined #git
08:22 FeyFre oops. It is appeared I do not need that empty dir anymore.. problem suddenly became irrelevant for now..
08:23 weckl joined #git
08:25 tristanp joined #git
08:25 biertie joined #git
08:26 emPi joined #git
08:26 biertie joined #git
08:26 paul98 joined #git
08:26 osse It's someone else's problems now!
08:27 * osse is reminded of that Futurama episode with the giant ball of garbage
08:27 FeyFre ok. other, irrelevane question
08:27 FeyFre have git repo, it has 4 branches: master, A, B, C
08:28 cyan__ joined #git
08:28 FeyFre master branch contains dirs A B C
08:29 FeyFre I need those dirs reference correspondent branches. master/A reference branch A, etc
08:29 PCatinean joined #git
08:31 tristanp joined #git
08:31 WayToDoor joined #git
08:33 eletuchy joined #git
08:34 skunkz joined #git
08:34 johnchalekson joined #git
08:39 boudiccas joined #git
08:39 tristanp joined #git
08:40 mwhooker joined #git
08:40 bronson joined #git
08:40 jceb joined #git
08:43 makinen joined #git
08:43 nnyk_ joined #git
08:43 patrakov joined #git
08:44 interne7y joined #git
08:45 makinen I made a new branch and added & committed a few files but there were commits in the original remote branch which I didn't fetch before making the new branch. How do I add them to the new branch?
08:46 zmo left #git
08:46 tens0r joined #git
08:46 cdown joined #git
08:47 Psychiatrist joined #git
08:49 osse makinen: merge or rebase. your choice
08:49 boudicca joined #git
08:50 makinen thx
08:50 fuchstronaut joined #git
08:51 stratos @makinen: you can pull the original branch into you new branch.. this will result in a fetch of the original branch and a merge of it into your new branch
08:52 bonii joined #git
08:52 makinen I solved the issue with merge
08:53 cdown_ joined #git
08:53 bonii Hi, I was using git filter-branch to remove a directory from my repository. Using git filter-branch --tree-filter 'rm -rf <dir>' -- --all does remove the directory but keeps the commit messages for commits in that directory, how can I get rid of commit messages as well ?
08:54 bonii Or am I misunderstanding something here ?
08:54 osse bonii: use filter-branch --prune-empty
08:55 EvilDMP Hello, I'd like to rename our "support/x.y" branches on GitHub to "release/x.y". What is the best and safest way to do this? It looks like: git branch -M, push, delete original on github
08:55 troulouliou_div2 joined #git
08:55 patrakov left #git
08:55 EvilDMP Or perhaps I could create it directly on GitHub
08:55 ErrorHead joined #git
08:55 EvilDMP but I would like to ensure that we lose nothing!
08:56 ErrorHead joined #git
08:56 courrier joined #git
08:57 Havvy EvilDMP:  That sounds like it'd work (though I don't know what -M does)
08:57 EvilDMP Havvy: should be -m actually
08:57 EvilDMP I think
08:57 EvilDMP hence my caution...
08:57 Havvy What does `-m` do?
08:58 osse Havvy: --move. it's basically rename
08:58 bonii osse: Thanks :-)
08:59 Havvy Oh. I'd just leave that off personally.
08:59 ErrorHead joined #git
08:59 Havvy And just delete the support one /later/.
09:00 Havvy If it's not private, you could clone it to your account and do a dry run there.
09:00 bronson joined #git
09:01 ahmedelgabri joined #git
09:01 ErrorHead joined #git
09:01 tristanp joined #git
09:01 _rgn joined #git
09:01 durham joined #git
09:01 cold_sau- joined #git
09:04 WayToDoor joined #git
09:05 Guest57 joined #git
09:06 johnny56_ joined #git
09:06 EvilDMP I think that the rename will always involve deleting the old version of the branch on GitHub
09:06 ErrorHead joined #git
09:06 jbitdrop joined #git
09:06 osse not unless you push --delete
09:07 ErrorHead joined #git
09:08 ErrorHead joined #git
09:09 EvilDMP osse: <fear>
09:10 johnchalekson joined #git
09:11 zacsek joined #git
09:11 WayToDoor joined #git
09:13 cmn joined #git
09:18 tristanp joined #git
09:18 CalimeroTeknik joined #git
09:19 khmarbaise joined #git
09:21 ascii-soup joined #git
09:21 a_thakur joined #git
09:26 EY joined #git
09:27 ascii-soup hi, i'm using 'git rev-parse @{-n}' to get branch names but if a branch has been deleted (say, at position @{-15}) rev-parse bails - any way to stop it from having a fatal error for unknown revision?
09:27 MutantMahesh joined #git
09:27 a_thakur joined #git
09:27 WayToDoor joined #git
09:28 ascii-soup for example, if I give it multiple (git rev-parse --abbrev-ref @{-1} @{-2}) and one of them doesn't exist, it stops doing any of the rest
09:29 tristanp_ joined #git
09:29 _ikke_ ascii-soup: --ignore-missing ?
09:30 ascii-soup alas, no :(
09:30 ascii-soup I mean, that's exactly what I want, but it doesn't work
09:31 _ikke_ hmm, i see
09:32 GavinMagnus joined #git
09:32 ascii-soup I think it's not an option for rev-parse maybe
09:32 liamonade joined #git
09:33 tristanp joined #git
09:34 liamonade joined #git
09:35 MattMaker joined #git
09:38 irqq joined #git
09:38 tristanp_ joined #git
09:39 Dirkos joined #git
09:39 Dirkos When running "git clean -dfx" i receive Skipping repository nius-core
09:39 boudiccas joined #git
09:39 Rish joined #git
09:40 chachasmooth joined #git
09:40 jost joined #git
09:44 josuebrunel joined #git
09:44 boudicca joined #git
09:47 xanadu joined #git
09:49 mikecmpbll joined #git
09:51 robotrolll joined #git
09:52 Rish joined #git
09:52 kellytk joined #git
09:53 johnchalekson joined #git
09:53 kellytk Is it correct that `git init` alone is not sufficient to create a new branch?  I receive the error "fatal: Not a valid object name: 'master'." from `git branch production`
09:53 jaguarmagenta joined #git
09:55 _rgn i think you need to commit once
09:55 SteffanW joined #git
09:55 cdown joined #git
09:56 kellytk That's what I suspect.  But that begs the question, why commit something to cause "master" to be generated to then add production and delete master?  I'm genuinely interested in the operation, I don't mean to express a complaint
09:57 troulouliou_div2 joined #git
09:58 sans_s3r1f joined #git
10:00 ascii-soup without a commit, there is nothing for a branch called 'master' to point to
10:00 hekto joined #git
10:00 ascii-soup branches + tags are just 'friendly names' that point to a commit
10:00 dsdeiz joined #git
10:00 dsdeiz joined #git
10:01 mdw joined #git
10:01 j416 kellytk: interesting question
10:01 ascii-soup not sure if there's a way to specify the 'first branch' name, however
10:01 fuchstronaut joined #git
10:01 kellytk Perhaps in .git/config?
10:01 ascii-soup very possibly
10:02 bollullera joined #git
10:02 j416 no
10:02 a_thakur_ joined #git
10:02 j416 it's hard-coded
10:03 bollullera left #git
10:03 j416 (i.e. not configurable)
10:03 kellytk Well, initial state at least, because through a process I can change the `default branch` can I not?
10:04 j416 hang on
10:04 kellytk I've created "production" branch and I'm now looking up the syntax for deleting master and updating any references (HEAD I suspect)
10:04 kellytk I looked for docs on configuring a default branch in .git/config other than "master" but found nothing
10:06 j416 out of curiosity, why do you need to make the initial commit on a branch that is not master?
10:06 j416 what is the use-case?
10:06 kellytk I don't understand your question
10:06 kellytk My intent is to configure my repositories with the default branch being "production" rather than "master", and then to initialize Git flow
10:06 j416 you want to create the very first commit on a branch that is not master, yes?
10:07 kellytk `git init`; ..?
10:07 j416 please just answer the questio
10:07 j416 n
10:07 kellytk It's an irrelevant consequence
10:07 mithenks joined #git
10:07 kellytk Ah but you may be interested for other reasons, ok let me think it over
10:08 ascii-soup not if the problem you're trying to solve can be solved in another way
10:08 j416 kellytk: I want to know the use-case
10:08 kellytk j416: I would prefer to begin commits with a production and develop branch, not master, that's correct
10:08 Junior joined #git
10:08 kellytk Yep, gotcha
10:08 kellytk Let me know if I can help in any other way too
10:08 j416 kellytk: master is a very common convention to use for the name, what is the reason to not use that?
10:09 j416 kellytk: I'm not trying to argue against you, I just want to know why you want to make your project non-idiomatic
10:09 nnyk_ joined #git
10:09 kellytk Thank you for the question.  I've come across a couple of articles describing potential issues with having a master branch so-named when shared with repo hosting services like GitHub
10:09 King_Hual joined #git
10:09 kellytk https://matthew-brett.github.io/pydagogue/gh_delete_master.html for example
10:09 ferr joined #git
10:10 johnchalekson joined #git
10:10 kellytk Second reason, I'm admittedly pedantic with naming and I intend to map repo branches with auto deployment to, wait for it, production and development infrastructure
10:10 tristanp joined #git
10:10 j416 kellytk: that is not tied to the name of the branch, you can accidentally push to any branch
10:10 elastix joined #git
10:11 kellytk I understand there is no concrete linkage, if that's what you mean
10:11 kellytk It's merely developer comfort
10:11 j416 renaming master to something else does not make things safer
10:11 j416 I can understand your second argument though.
10:11 kellytk The issue with hosting services I can't speak to as I've only really pushed code to GitHub, not much collab/social stuff
10:12 mase-tech joined #git
10:12 kellytk Can you help me understand what you concern is the author of the article has? (be it unfounded or otherwise)
10:12 kellytk Sorry, what the concern*
10:12 tlaxkit joined #git
10:14 j416 kellytk: it's a strange workaround
10:15 tristanp_ joined #git
10:15 j416 kellytk: he seems to be scared that he will make a pull request from a branch named "master", which indeed is odd, should be more descriptively named. But, if he doesn't know he is on branch master (should be quite obvious), and even a github pull request shows the name of the branch, then he probably has otehr problems..
10:16 nuge joined #git
10:16 kellytk If a pull request directly to master was filed at a repo hosting site, it could simply be rejected is that right?
10:16 ascii-soup yep
10:16 j416 yes
10:17 tristanp joined #git
10:17 roaman joined #git
10:17 ascii-soup I think that article mentions problems which shouldn't really exist
10:17 j416 kellytk: "if you ever worry about spilling coffee on your clothes? here's how to fix it: buy tea instead"
10:18 j416 s/?//
10:18 kellytk Ok well thank you and moving on; in using "production" for default vs "master", should I remember any tips beyond adding the new branch and deleting master?  I'm still skimming articles on the subject
10:18 ascii-soup I mean, if the author is so slap-dash with how they manage their git branches / workflow, they should probably sort that issue out instead of trying to mess around with git
10:18 kellytk :tea: :party:
10:18 j416 kellytk: you can rename it, there's no issue there
10:18 j416 kellytk: you cannot make it the default name when you init a new repo, unless you wrap git-init to also update .git/HEAD
10:19 j416 kellytk: I would recommend sticking to master.
10:19 roaman left #git
10:19 kellytk Why?
10:20 j416 kellytk: because that's a sensible default, used everywhere
10:20 shaggycat joined #git
10:20 ochorocho__ joined #git
10:20 j416 kellytk: if you have a repo where it does not make sense, just use other names. But for new repos, how does it make sense to start with "production" for instance? it's not even live yet.
10:20 kellytk But then I read plenty of comments and docs that say branches can be anything and even master is arbitrary
10:21 j416 kellytk: I understand your question and I think your argument that it would be nice to have it configurable is valid, though.
10:21 j416 kellytk: yep
10:21 j416 kellytk: but as I said, the name 'master' is hard-coded when you do 'git init'.
10:21 kellytk That's a fair point actually.  And with my workflow that branch wouldn't have any commits to begin with, and until a git flow release start was finished
10:21 j416 kellytk: you can easily rename it after you do the first commit
10:21 User458764 joined #git
10:21 tristanp_ joined #git
10:21 cdown_ joined #git
10:22 Xano joined #git
10:22 kellytk I checked out production and git status returned "fatal: This operation must be run in a work tree" so I'm looking up the cause of that now
10:22 eletuchy joined #git
10:22 jondot joined #git
10:22 j416 kellytk: I tend to start _all_ of my projects on branch master with a commit "Initial commit" containg a readme. Then depending on the project, I might continue work on a separate branch, or not.
10:23 kellytk That's what I've been doing to warm up repos too :-)
10:24 j416 kellytk: I prefer keeping "master" as the stable branch, so that the state of master is always either the latest release, or at least something stable enough to not explode
10:24 ascii-soup For us, master is the current state of production
10:24 j416 kellytk: i.e., safe to start new work from, and good enough to not need history changes
10:24 j416 ascii-soup: yep, same @ dayjob
10:25 kellytk The notion of a master key comes to mind and gives your argument weight
10:25 ascii-soup We then have another long-lived branch called 'deploy' which is the 'working branch' in which we merge/test/etc before a production release happens (which is usually once or twice per day)
10:25 ascii-soup master is _always_ safe to branch from
10:25 kellytk Why don't you use Git flow?
10:25 ascii-soup _always_ safe to deploy
10:25 ascii-soup git flow is more complex than we need
10:26 ascii-soup We have master + deploy, which are 'shared'
10:26 j416 kellytk: you will want to keep your side branches (such as the one you call development) close to prod anyway, and merge often, and naming the main branch "master" kind of indicates that this is the one true branch into which everything that is worth deploying goes into
10:26 ascii-soup we have tooling support around merging/rebasing into/onto those branches so we don't mess it up
10:26 j416 git flow is too complex for us as well, we don't need it
10:26 j416 we don't have tooling, we just use plain git,' works fine.
10:27 ascii-soup developers are free to create other branches as they see fit, but those branches are just 'handy' for the devs, they have no 'meaning' outside of that
10:28 j416 ascii-soup: we prefix branches with our names, devname/somefeature
10:28 j416 ascii-soup: it's very handy.
10:28 ascii-soup when i say tooling, i mean we have a couple of git commands, 'git prep-deploy' and 'git deploy'
10:28 ascii-soup j416: Yeah we do the same thing
10:28 j416 ascii-soup: we don't even have that, works well without :)
10:28 ascii-soup Although for us it's usually a split between  'devname/issueid-description' or 'whatamidoing' :P
10:28 kellytk Prefixing branches with usernames is another smart one
10:28 j416 but I suppose some things could be a little bit sped up with a script.
10:29 makinen I made a new branch and pushed it into the remote repository by git push but why am I not seeing the new branch on gitlab?
10:29 j416 makinen: did you get an error message?
10:29 _ikke_ makinen: how did you push?
10:30 ascii-soup j416: the reason we have those commands is to ensure that we always fetch + reset + rebase on the 'deploy' and 'master' branches; it's more of a 'I can't be bothered to type all that' than for safety if i'm honest
10:30 makinen it gives me a warning "push.default is unset; ..." and after that says everything is up-to-date
10:31 NwS joined #git
10:31 makinen _ikke_: just by a command 'git push'
10:31 Psychiatrist joined #git
10:31 j416 ascii-soup: we don't enforce rebase if it's a clean merge (syntactically and semantically), but we don't do merge conflicts, so anything that conflicts or looks scary, we don't merge unless developer rebases it
10:31 j416 works pretty well
10:32 ascii-soup the beauty of git - supports all the workflows :)
10:32 j416 all teh workflows woo
10:33 King_Hual` joined #git
10:33 fuchstronaut joined #git
10:36 mda1 joined #git
10:39 freimatz joined #git
10:45 t0mex joined #git
10:45 interne7y joined #git
10:45 _ikke_ makinen: That's the problem, first set push.default to a value that works for you (simple is the most basic one)
10:46 JeroenT joined #git
10:47 MattMaker joined #git
10:49 gaen joined #git
10:50 dopesong joined #git
10:53 DropItLikeItsHot joined #git
10:53 ocbtec joined #git
10:53 makinen _ikke_: got it to work thx
10:54 johnchalekson joined #git
10:55 johnchalekson joined #git
10:56 zeroed joined #git
10:59 leeN joined #git
11:00 tristanp joined #git
11:03 CandyWife left #git
11:04 Stath joined #git
11:04 tristanp_ joined #git
11:06 ceaucari joined #git
11:10 benteight joined #git
11:10 ceaucari joined #git
11:11 johnchalekson joined #git
11:12 harish_ joined #git
11:13 Guest57 joined #git
11:13 johnmilton joined #git
11:13 CussBot joined #git
11:14 mdw joined #git
11:15 vjacob joined #git
11:16 eletuchy joined #git
11:16 aspiers hi all. I've just written a new git subcommand called git-splice which wraps around rebase -i and can non-interactively remove/insert commits from within a branch (not just at the tip).  I hope it might be considered for inclusion in git. I'm currently finishing off the unit test suite ...
11:16 bjpenn joined #git
11:17 aspiers It's part of a bigger plan I have
11:17 aspiers next step is to write git-transplant which will wrap around git-splice and allow transplanting of commits from within a branch to elsewhere (e.g. insert into / append to existing or new branch)
11:18 aspiers then final step is to write git-explode which combines git-transplant with git-deps and can automatically explode a linear sequence of commits into independent branches
11:18 kellytk left #git
11:19 aspiers well, actually there's one more step - add support for git-explode to integrate with https://github.com/jonseymour/gitwork or topgit
11:19 aspiers thought I'd mention here in case people are interested
11:19 nettoweb joined #git
11:19 aspiers I should also post on the list about this
11:21 aspiers the alpha code is in this branch, but be warned I will be rewriting history: https://github.com/git/git/compare/master...aspiers:splice
11:21 osse hmm, there's a joke here somewhere
11:21 osse you willb e rewriting the history of a project that helps rewrite project histories
11:21 mohabaks joined #git
11:22 aspiers lol
11:22 aspiers indeed :) but it's not the first time ...
11:23 osse Sorry to burst your bubble, but I'm pretty sure it won't be included in git if it's a bash script. All official git scripts are sh scripts
11:23 aspiers osse: oh, I can easily fix that. I don't think I use any non-POSIX stuff
11:23 osse you do
11:23 aspiers osse: originally I didn't intend to submit it to git
11:23 aspiers osse: OK, well I can happily fix that
11:23 osse args=( stuff ...); cmd "${args[@]}"
11:23 aspiers oh yeah, arrays :-(
11:24 aspiers well that can be fixed
11:24 aspiers hell, I could rewrite in Python
11:24 osse please don't :p
11:24 osse I mean please don't fix it
11:24 aspiers if I'd originally intended to submit to git, I would have used test everywhere
11:24 aspiers why not?
11:24 _ikke_ I think nowadays things like this firsts land in contrib
11:24 osse Rewriting in python could work; but perl might be a better choice since a lot of stuff is perl already
11:24 aspiers _ikke_: right, that was my expectation too
11:25 osse aspiers: don't listen to me. posix scripts are the devil
11:25 aspiers sorry, I refuse to write perl on principle ;-)
11:25 aspiers I wrote perl for ~15 years
11:25 aspiers that's more than enough for me
11:25 aspiers yes, posix is horribly limited
11:26 aspiers _ikke_: how would I handle the unit tests if it's in contrib though?
11:26 aspiers _ikke_: having said that, when I wrote check-ignore, it went straight into core, not via contrib/
11:26 osse if it's in contrib then maybe bash is ok
11:26 tristanp joined #git
11:26 _ikke_ Oh, nice
11:27 aspiers by "straight", I mean after about 6 months and endless reviews :)
11:28 monoxane joined #git
11:28 osse hmm
11:28 leonarth joined #git
11:28 osse maybe i should resurrect my rewrite-stash-in-c branch
11:29 osse but noobs need to be rekt too. choices.
11:29 aspiers I also had another interesting idea along the same lines as git-transplant:
11:29 tvw joined #git
11:29 _ikke_ aspiers: if it fits in the current git style of programming, they're more inclined to include it
11:30 aspiers reuse my d3.js work in git deps --serve to render git history graphs in the browser
11:30 aspiers and then allow operations like selecting a range of commits and then dragging and dropping it to another branch
11:30 aspiers IOW, a modern HTML-based replacement for gitk etc.
11:30 aspiers which could eventually grow to a full-blown web UI
11:31 {AS} joined #git
11:31 tristanp_ joined #git
11:31 strk joined #git
11:31 MutantMahesh joined #git
11:31 strk (how) is it possible to create a branch with no commit in common with any other ?
11:31 aspiers _ikke_: sure - https://github.com/git/git/compare/master...aspiers:splice#diff-30f2671b4625984c97d2979256ef1585 is already moving in that direction
11:32 strk a new top-level branch, that is (or who does github do it for webpages for example?)
11:32 scoobert1on joined #git
11:32 aspiers strk: check out a new branch at the root commit, then amend it
11:32 cdown joined #git
11:33 aspiers strk: yesterday I discovered rebase --root, which is related but overkill for your case
11:33 a_thakur joined #git
11:33 aspiers strk: there are other ways of achieving the same thing, e.g. with plumbing commands
11:33 trista___ joined #git
11:33 MutantMahesh joined #git
11:33 pol123456 joined #git
11:34 johnchalekson joined #git
11:34 strk good point (I did amend the first one)
11:34 ciampix joined #git
11:34 ochorocho__ joined #git
11:36 EY joined #git
11:36 nilsi joined #git
11:36 tristanp joined #git
11:38 johnchalekson joined #git
11:39 TomyWork For a script I'm writing, I'm looking for a solution that non-force-pushes without asking, but asks the user for confirmation if a force push is needed
11:40 TomyWork is there something that git push has out of the box?
11:40 skunkz joined #git
11:40 TomyWork that would certainly be less confusing for a UI than "if ! push; then confirm && push -f; fi"
11:41 j416 TomyWork: just script it? git push, check for error code
11:41 Lunatrius joined #git
11:41 johnchalekson joined #git
11:42 TomyWork well i do want to show the output of "git push", but i dont want to litter the output with irrelevant error messages, if possible
11:42 johnchalekson joined #git
11:43 TomyWork but if git push has no integrated option like that, that's what i'll have to do
11:43 johnchalekson joined #git
11:44 Havvy git push 2> /dev/null     ;)
11:44 j416 TomyWork: >
11:44 j416 what havvy said
11:44 osse this reminds of my patch that implements git push --force-with-tease
11:44 TomyWork that's an odd name
11:44 Lunatrius` joined #git
11:44 j416 there's also -q to git push.
11:45 Havvy By the by, if anybody wants to fix https://github.com/NixOS/nixpkgs/issues/15687 (nix + sh + git issue), I'd be happy.
11:45 osse TomyWork: the idea is that you say to the server "i'm going to push... Just kidding! I'm gonna push --force lolol".
11:45 osse Basically it does git push || git push --force
11:45 osse except built in to git push \o/
11:46 TomyWork with user confirmation?
11:46 interne7y joined #git
11:46 osse nope
11:46 TomyWork then how is it different from --force??
11:46 TomyWork -?
11:47 TomyWork i mean how is the end result different from --force?
11:47 osse it's different in that for a second the server gets to be pleased with itself and then afterwards it gets crushed
11:48 j416 TomyWork: man git-push
11:48 gitinfo TomyWork: the git-push manpage is available at http://jk.gs/git-push.html
11:48 TomyWork osse oww
11:48 TomyWork oh, i see, there's force-with-lease
11:48 osse TomyWork: Spooiler alert: It's not a serious patch.  I thought  "heh, --force-with-tease would be a funny name for an options because it rhymes" and then I write a patch afterwards
11:48 cdesai joined #git
11:49 osse I also have patch that changes the default expire time for reflog from 90 to 89 days
11:49 tristanp joined #git
11:50 mahendra_ joined #git
11:50 TomyWork osse well you could take that force-with-tease patch
11:50 TomyWork rename the option
11:50 TomyWork put user confirmation between the two pushes
11:50 ShapeShifter499 joined #git
11:50 TomyWork and voila, you have that option i'm looking for :)
11:51 osse But on a serious note, it doesn't sound very useful to me. Because in my eyes when you give that option you have already made the confirmation.
11:51 osse g push --force where the --force is in fact not needed isn't different from a regular push anyway
11:52 TomyWork osse well as i said, this is for a script
11:52 osse I am in no way against it, mind you. I just think it's not necessary
11:53 TomyWork git pull isnt necessary either
11:55 ojdo joined #git
11:55 idmistir joined #git
11:55 idmistir joined #git
11:56 NeverDie joined #git
11:57 osse but it saves me time
11:57 osse this wouldn't. but it may save others time, so go for it.
11:57 kellytk joined #git
11:57 osse not sure why I'm objecting in the first place; it's subjective. never mind me
11:58 kellytk What is the point of tag messages and what is good practice?  I plan to name release tags "v1.2.3" but I don't know how the tag message is used
11:58 Sceorem joined #git
11:58 _rgn co-worker new to vim was interested in my vim config, but i was thinking about suggesting vim-sensible as a starting point. thoughts?
11:59 _rgn oops wrong channel
11:59 osse kellytk: even Git's own tag messages a pretty trivial. But it might include stuff like why you created the release
11:59 fstd joined #git
11:59 osse is it for a particular customer? Is it the first version that has Impotant Feature No. 1 ?
12:00 kellytk Simply when new product releases are cut and versioned
12:00 kellytk From "develop" using Git flow
12:00 osse Why release it now? Why not a week ago or tomorrow?
12:01 acetakwas joined #git
12:01 tristanp_ joined #git
12:01 kellytk If I knew how tag messages were used I could better answer that
12:01 osse no
12:01 neuro_sys joined #git
12:02 osse you got it wrong
12:02 ceaucari joined #git
12:02 neuro_sys Hello, trying to checkout a particular file that was deleted from a previous commit, but when I do git checkout REF filepath it fails with the error: did not match any file(s) known to git
12:02 kellytk How?
12:02 osse Disregard the tag message for a second. Just answer the questions plainly
12:02 ntshetty joined #git
12:02 neuro_sys I think it works only files that are not deleted, is there a way to checkout deleted files from previous commits?
12:02 kellytk The release implements the set of features planned for it
12:02 osse BOOM
12:02 kellytk Or rather, it groups those
12:02 osse there you go
12:02 osse write that
12:03 kellytk So tag messages should be a kind of changelog if they refer to versions?
12:03 rnsanchez joined #git
12:03 flaguy48 left #git
12:04 ncd|leeN joined #git
12:04 neuro_sys sorry It was my mistake to checkout the "deleting" commit
12:04 neuro_sys I had to go one before
12:05 MutantMahesh joined #git
12:05 ceaucari joined #git
12:06 tristanp joined #git
12:06 josuebrunel joined #git
12:06 osse kellytk: I woudln't say "should" but sure
12:07 kellytk I notice multiline appears to be supported
12:07 kellytk And this is what my question was about, examples of good uses of tag messages when cutting a release
12:07 fornax joined #git
12:08 osse kellytk: In my opinion it would me useful to write why the tag was created. What this tag signifies. This is why I asked you all the questions above.  But if it's blatantly obvious to everyone involved in the project why (eg. you have som release schedule everyone knows about) then it won't be as useful.
12:08 tristanp_ joined #git
12:09 a_thakur joined #git
12:10 eletuchy joined #git
12:10 ajf- joined #git
12:10 tristanp_ joined #git
12:12 rwb joined #git
12:12 osse You could for example write why it has more or less features than what was originally planned, or why it was released ahead of scheduled or why it was delayed
12:13 tesuji joined #git
12:15 nnyk_ joined #git
12:16 osse You said "If I knew how tag messages were used I could better answer that". That's the wrong way to approach the problem. The questions I asked has nothing directly to do with Git or tags or tag messages.
12:16 jondot left #git
12:17 deadnull_ joined #git
12:19 osse That said: All the projects I have lying around have pretty useless tag messages. So maybe it's simply not a much-used feature.  But I stand by what I said, and I would add that if you don't have anything useful to put there then don't put anything. Better to have a shirt and trivial message than a useless one
12:19 osse *short
12:19 efco joined #git
12:19 tristanp joined #git
12:20 jost_ joined #git
12:20 osse The deafening silence makes me think everyone agrees with me or everyone thinks I'm a lunatic.
12:21 skunkz joined #git
12:23 tomboy64 is there a shorthand for performing the rebase action "edit" on all commits for a set of revisions?
12:23 lss8 joined #git
12:23 tomboy64 basically as if i did rebase and replaced all picks with edits
12:24 LeMike joined #git
12:24 sdothum joined #git
12:24 BtbN :%s/pick/edit/g
12:25 tomboy64 yarp, that's what i'm currently doing
12:25 osse GIT_EDITOR='vim -c "%s/pick/edit | wq"' git rebase -i
12:25 osse i wonder if that actaully works
12:26 BtbN could probably use sed -i instead
12:26 BtbN but that would work i guess
12:26 osse yeah
12:27 troulouliou_div2 joined #git
12:27 tomboy64 thanks :)
12:28 digidog joined #git
12:28 gaurav_ joined #git
12:30 nettoweb joined #git
12:31 sagerdearia joined #git
12:31 jeffreylevesque joined #git
12:32 tomboy64 hmm | doesn't get sourced
12:32 tomboy64 <bar> neither
12:32 tomboy64 they are taken verbatim
12:32 tomboy64 ohwait
12:32 drodger joined #git
12:33 a_thakur joined #git
12:33 nilg joined #git
12:33 * tomboy64 stupid
12:33 tomboy64 GIT_EDITOR='vim -c "%s/pick/edit/g | wq"' git rebase -i master
12:33 tomboy64 thaorks
12:33 tomboy64 that works
12:33 BtbN I'd definitely use sed instead of vim for that task though
12:33 tristanp_ joined #git
12:35 MattMaker joined #git
12:37 M-sbts joined #git
12:39 ChrisWarrick left #git
12:40 fmcgeough joined #git
12:41 tristanp joined #git
12:43 kellytk left #git
12:44 liquidjamm joined #git
12:44 rainbyte joined #git
12:45 EduardoMartins joined #git
12:47 interne7y joined #git
12:47 noidea joined #git
12:49 Meteorhead If I want to reset hard a folder to a particular commit (and not whole tree) is git reset --hard {hash} /folder right ?
12:49 ahmedelgabri joined #git
12:50 _ikke_ No, git reset --hard only operates on whole trees
12:50 M-sbts joined #git
12:50 _ikke_ Meteorhead: You want git checkout <hash> -- <path>
12:51 Meteorhead _ikke_: pheww thank you very much, you saved me haha
12:52 dsdeiz_ joined #git
12:53 tristanp_ joined #git
12:54 a_thakur joined #git
12:54 annoymouse joined #git
12:54 liamonade joined #git
12:55 strk left #git
12:56 jstimm_ joined #git
12:56 EY___ joined #git
12:56 tomboy64 BtbN: sed doesn't seem to work. at least GIT_EDITOR='sed -i "s/bla/blubb/g" -e ' git rebase -i master doesn't
12:56 dopesong joined #git
12:56 fizzgorilla joined #git
12:56 annoymouse When doing `git add -p` what is the difference between `q` and `d`?
12:57 annoymouse joined #git
12:58 tristanp joined #git
12:58 mizu_no_oto joined #git
12:59 _ikke_ annoymouse: d is only per file, q exists the entire process
12:59 annoymouse _ikke_:  Thanks
12:59 _ikke_ had to think about it for a second too
12:59 italoacasas joined #git
13:00 annoymouse _ikke_: The descriptions aren't very clear: "q - quit; do not stage this hunk or any of the remaining ones" and  "d - do not stage this hunk or any of the later hunks in the file"
13:00 _ikke_ "in the file" is the key
13:00 annoymouse I guess they could have emphasized in the file more
13:01 PerlJam joined #git
13:01 cdown joined #git
13:03 skunkz joined #git
13:06 tristanp joined #git
13:07 happy-dude joined #git
13:07 k3rn31 joined #git
13:08 dopesong joined #git
13:09 fuchstronaut joined #git
13:09 c0ded joined #git
13:10 unbalancedparen joined #git
13:11 troulouliou_div2 joined #git
13:12 danchr joined #git
13:12 mdw joined #git
13:16 sagerdearia joined #git
13:16 tristanp_ joined #git
13:16 flaguy48 joined #git
13:19 rnsanchez joined #git
13:21 netj joined #git
13:21 kam270 joined #git
13:22 JosephFerano joined #git
13:23 tristanp joined #git
13:24 JosephFerano hello, we're switching source control to git lfs. We have a large project and most of the cloners will already have the project on their machine. Is there a way to init the repository and have it sync with github master, without having to do a traditional clone?
13:24 JosephFerano I tried googling but the solutions I found haven't worked
13:25 selckin just add it as a remote?
13:25 mbrzuzy joined #git
13:25 tobiasvl I don't think there's a repo yet
13:25 tobiasvl locally
13:26 JosephFerano The directory and the files exist
13:26 dopesong joined #git
13:26 JosephFerano but not the history/repo per se
13:26 EY joined #git
13:26 selckin then no
13:26 JosephFerano it's on github though
13:26 selckin can have one guy clone it in the office and share it tho if network is the bottleneck
13:26 JosephFerano sorry, meant locally, the history is on github, so I could add the remote, but I would need to sync the history between the local repo and the blessed one
13:27 JosephFerano I can give the repo as is in a USB and just have them change the git config?
13:27 JosephFerano like the username and email and stuff?
13:27 heinrich5991 joined #git
13:28 JosephFerano I could probably do that, yeah, that's a good point. Would it work fine between mac and windows? Don't see why not, but I'm asking just in case.
13:28 tristanp_ joined #git
13:28 Rish joined #git
13:29 selckin should work
13:29 selckin also only need the .git  not the checkout
13:32 prsn joined #git
13:32 JosephFerano right
13:32 JosephFerano plus the gitignore and the gitattributes
13:32 JosephFerano sounds good though, should be an easy transition
13:33 _ikke_ Those should be comitted anyway
13:34 stonerfish joined #git
13:35 ochorocho__ joined #git
13:35 mohabaks joined #git
13:36 fiy6 joined #git
13:40 JosephFerano right
13:40 JosephFerano Alright, thanks for the help
13:40 Darcidride joined #git
13:41 User458764 joined #git
13:42 dopesong joined #git
13:43 todd_dsm joined #git
13:45 Darcidride joined #git
13:46 khmarbaise joined #git
13:47 Dougie187 joined #git
13:47 L8D joined #git
13:47 MattMaker joined #git
13:48 interne7y joined #git
13:48 jstimm_ joined #git
13:49 tristanp joined #git
13:49 osse tomboy64: what does that -e do? o_O
13:49 nilg joined #git
13:50 JosephFerano joined #git
13:50 a_thakur joined #git
13:52 cyan__ joined #git
13:53 Gamecubic joined #git
13:55 nilg joined #git
13:57 Waskeend joined #git
13:58 AaronMT joined #git
13:58 eletuchy joined #git
13:58 ash_workz joined #git
13:59 fahadash joined #git
14:00 Rish_ joined #git
14:00 cyan__ joined #git
14:01 acetakwas joined #git
14:03 dopesong joined #git
14:04 cdown joined #git
14:08 tomboy64 osse: that's used to signify an expression; so you can have sed execute multiple expressions with one call.
14:10 osse tomboy64: yes, but that -e has no expression after it
14:10 tomboy64 ...
14:10 osse if GIT_EDITOR works the way I think it works it will try to read the filename of the todo list as an expression
14:10 tomboy64 yeah
14:10 tomboy64 i turned the flags around, silly me
14:11 tomboy64 thanks!
14:11 marcelofr joined #git
14:12 prsn joined #git
14:12 osse \o/
14:12 a_thakur joined #git
14:13 BrianBlaze420 joined #git
14:13 osse but would be nice with git rebase --default-action=edit
14:13 osse or some such
14:13 guardian hello
14:13 osse (and git rebase -d edit of course. Ain't nobody got time to type that!)
14:15 cyan__ joined #git
14:15 guardian I'm using git for-each-ref --shell --format to display information about who committed last on each branch
14:16 guardian is there a direct way with git-for-each-ref to make it consider the penultimate commit in each ref instead of the last one?
14:17 bjpenn joined #git
14:17 dsdeiz joined #git
14:17 dsdeiz joined #git
14:17 tristanp joined #git
14:18 Bombe What is “the penultimate commit in [a] ref?”
14:18 matthiaskrgr joined #git
14:18 osse Bombe: ref~1
14:19 Bombe Ah.
14:19 Bombe So there is a word for that in English.
14:20 osse Bombe: if you're standing in line for the restroom, and there's a guy in front you but no one behind you then that guy is the penultimate guy in the line
14:20 osse u feel me fam
14:21 osse I suspect the answer to his and/or her question is "no"
14:21 zivester joined #git
14:22 osse might need git log. get a list of refs, do some regex magic to append ~1 to each of them
14:23 adron joined #git
14:23 aspiers osse, _ikke_: http://article.gmane.org/gmane.comp.version-control.git/295755
14:24 aspiers this is the point where someone probably tells me I just reinvented a wheel :-o
14:25 tomboy64 with git diff --stat HEAD^ it opens up my pager to view the stuff. with | cat it strips color. anyway to get it into stdout without the colors stripped?
14:25 haasn joined #git
14:26 Bombe tomboy64, set color.ui to “true”. Default is “auto” which does not use color if the output is not a terminal.
14:26 bsanford joined #git
14:26 tomboy64 Bombe: can i do that one-time on the cli?
14:26 frem_ joined #git
14:26 Bombe tomboy64, git-diff does have a --color option, so I’d say yes.
14:27 tomboy64 ahhh, oversaw that
14:27 tomboy64 thanks!
14:27 osse aspiers: I can confirm that you're not the only one diciplined enough
14:27 aspiers osse: haha, good to know ;-)
14:28 aspiers osse: I easily found more evidence of that by googling for "git move commits to another branch"
14:28 osse what I've done usually is to move bugfixes and refactorings to the start of my branch, and then potentionally push them to master to "get them out of the way". That works well for me, but this looks cool
14:28 aspiers osse: yes, that's what I do a lot too
14:28 osse aspiers: you might have come across this: https://github.com/Osse/git-scripts/blob/master/git-move :P
14:29 Panacea joined #git
14:29 User458764 joined #git
14:30 aspiers osse: hehe :) these days I find magit helps enormously with branch maintenance, but it's still not good enough when you have multiple topic branches with dependencies between them
14:30 phaleth joined #git
14:30 aspiers osse: I wrote git-mix / git-mixdown for that https://github.com/aspiers/git-config#upstream-branch-status-reporting--management
14:31 osse :O that's a lot of stuff
14:31 aspiers and also https://github.com/aspiers/git-config/blob/master/bin/ggrbm but it's an ugly hack
14:31 aspiers I also started work on adding rebase support to topgit, but never got very far
14:31 aspiers IMHO gitwork possibly has the most potential
14:32 axisys joined #git
14:32 sangy joined #git
14:34 ocbtec joined #git
14:35 tristanp joined #git
14:37 cyan__ joined #git
14:38 Guest61081 joined #git
14:38 Guest61081 joined #git
14:41 Lunatrius` joined #git
14:42 exarch joined #git
14:44 Darcidride joined #git
14:44 joki joined #git
14:44 nofacade joined #git
14:44 tinajohnson___ joined #git
14:47 cyan__ joined #git
14:48 interne7y joined #git
14:49 Lunatrius joined #git
14:51 Darcidride joined #git
14:52 dreiss joined #git
14:52 tristanp joined #git
14:55 LionsMane joined #git
14:56 diegoviola joined #git
14:57 kulelu88 joined #git
14:57 kulelu88 Hello all
14:57 _rgn joined #git
14:57 kulelu88 is there a service similar to github that is run by a community of volunteers?
14:58 kilo_bravo joined #git
15:00 MattMaker joined #git
15:00 rkazak joined #git
15:00 jimi_ joined #git
15:00 deryni joined #git
15:01 Darcidride joined #git
15:01 grawity repo.or.cz, I think
15:03 afuentes joined #git
15:03 bibble awesome
15:03 Deeroy joined #git
15:04 marcelofr joined #git
15:04 skunkz joined #git
15:05 kpease joined #git
15:07 crose joined #git
15:09 sargas joined #git
15:11 SteffanW joined #git
15:12 kulelu88 grawity: if you had to call your git service something, what name would appeal to you?
15:12 kulelu88 *hosted git service
15:13 grawity Go Away™
15:15 kulelu88 sigh
15:16 jrun joined #git
15:17 _28_ria joined #git
15:18 tristanp_ joined #git
15:18 jrun is branching always based on the current branch?
15:18 liamonade joined #git
15:19 bremner not according to the fine manual
15:19 grawity `git branch` takes a start-point parameter
15:19 subhojit777 joined #git
15:19 durham_ joined #git
15:21 kulelu88 does anybody have any reasonable names to call a public git service? if not, no need to answer, just looking for a few ideas
15:26 tristanp joined #git
15:27 mwleeds joined #git
15:28 amdi_ joined #git
15:29 noark9 joined #git
15:29 rgrinberg joined #git
15:29 f3r70rr35f joined #git
15:30 baraba joined #git
15:31 Jellyg00se joined #git
15:32 m_antis joined #git
15:33 rbr joined #git
15:34 tristanp_ joined #git
15:36 ochorocho__ joined #git
15:37 tristanp joined #git
15:39 bibble kulelu88, a play on gitolite... gitoserve ?
15:40 barteks2x joined #git
15:40 bibble pubgit
15:41 ThiefMaster yet another public git service?
15:41 orcus joined #git
15:42 barteks2x joined #git
15:42 kulelu88 I don't know if you guys realized this, but none of the public available services are run by a community. The biggest ones are all driven by for-profit organizations
15:42 tristanp joined #git
15:42 barteks2x git fetch --unshallow --tags doesn't seem to actually fetch any tags, is there some other way to do that?
15:42 osse Gorcha. Short for The Githarmonic Orchestra
15:43 m_antis joined #git
15:43 bibble gitpit
15:44 osse apogit
15:44 leonarth joined #git
15:44 nettoweb joined #git
15:44 osse Like apogee except it's has git in it
15:45 |meta joined #git
15:45 bibble Had to tfd.com/apogee
15:45 osse Alternative interpretation: Apocalypse Git!
15:45 bibble Ha, nice.
15:45 kulelu88 does the word git included in the name make it more appealing to the user?
15:45 kulelu88 or does the name not matter much?
15:46 bibble Makes it more obvious, or then pick anything.
15:46 bibble banana
15:46 eletuchy joined #git
15:46 osse bibble: Apogee was also the name of the dev studio that made the Commander Keen games
15:46 bibble Ah, didn't know that.
15:47 kulelu88 how does this name seem: kernelhq?
15:47 kulelu88 gitbase?
15:47 bibble Like gitbase
15:47 kulelu88 gitpad?
15:47 tristanp joined #git
15:47 kulelu88 kryo/kryp ?
15:47 thiago joined #git
15:47 bibble gitadd
15:47 osse kernelhq is too similar to kernel.org
15:48 osse IMHO
15:48 rbr joined #git
15:48 kulelu88 with base, the issue I had was that there is a service called "codebase"
15:48 bibble See.
15:49 bibble gitrep; repos listed by reputation.
15:49 jimi_ joined #git
15:49 lindenle joined #git
15:49 interne7y joined #git
15:49 kulelu88 what about: nucleus? bibble
15:50 osse Apogit is the best suggestion ever. I can already envision the logo. Sort of like the Git logo but it's like an orbit
15:50 jfran joined #git
15:51 skunkz joined #git
15:51 osse Dots along the orbit are satellites and the planet represents the server
15:51 bibble nucleus cool, but hard to get domain ?
15:52 _ikke_ http://apogit.pl/
15:52 kulelu88 yeah looks like it's taken
15:52 espindola joined #git
15:52 bibble :D
15:52 kulelu88 any proper english word will be difficult to get as a domain sadly
15:52 osse And instead of "creating a fork" you can launch a satellite.
15:52 diego2 joined #git
15:52 osse It's glorious, Dan!
15:53 osse Aww man :(
15:53 bibble Someone already though of the glory.
15:53 bibble #thought
15:53 acetakwas joined #git
15:53 kulelu88 apogit is a legit company as well
15:53 osse Gorcha it is then. Each person in the orchestra is a contributor
15:53 kulelu88 Polish
15:54 artemisart joined #git
15:54 madewokherd joined #git
15:54 barteks2x how can I fetch tags if git already thinks everything is up to date? I really need that to work because travis creates a clone with --depth 50 and specifies a branch "master", and I need tags for git describe to work.
15:55 akushner joined #git
15:55 osse People who have push access are called conductors. And the main repo in combination with the forks are called the ensemble
15:55 diego3 joined #git
15:55 osse Perfect!
15:56 ajf- joined #git
15:56 _ikke_ and you are sure you don't already have the tags?
15:56 mikecmpbll joined #git
15:56 bibble I like where osse is going, like openshift and digitalocean; having a whole theme.
15:56 barteks2x I'm sure they are not there, git tag gives no output
15:56 Eugene As a Power User, I think "cute" themes like that are annoying, stupid, and confusing
15:56 Eugene YMMV
15:56 deadnull_ joined #git
15:56 diego3 joined #git
15:57 InfoTest joined #git
15:57 nu11_byte joined #git
15:57 bibble True, can be better sticking with actual terminology.
15:58 bibble allgit
15:58 osse bibble: Bur for Halloween it's all about orchs, of course
15:59 osse No, I don't mean a git host for kids. Bur it can still be elegant, ya know.
16:00 bibble githut
16:00 osse Honestly I think the way github uses the word "fork" is kinda misleading. But it's ingrained niw, I guess
16:00 sagerdearia joined #git
16:00 osse Now
16:01 barteks2x Can it be somethign that at some point was a bug and has been fixed in later version? I gave git 2.1.4
16:01 Grimm_ joined #git
16:01 barteks2x *have
16:01 clemf joined #git
16:02 kulelu88 it doesn't bother any of you that the biggest repo providers are for-profit companies?
16:02 kulelu88 *repo providers / git hosting services
16:02 josuebrunel joined #git
16:02 nu11_byte no
16:02 SwiftMatt joined #git
16:03 sangy joined #git
16:03 nu11_byte bitbucket provides free repositories and atlassian yes has JIRA/Stash etc, github charges for private but at the same time you can setup your own remotes
16:03 samson joined #git
16:04 nu11_byte also the biggest providers of most thing turn a profit, thats how they got there
16:04 Eugene kulelu88 - centralized hosting is the sort of thing that naturally attracts business and is diffcult to provide gratis; Free self-hosted alternatives like !gitolite suit people who don't want to use commercial tools
16:04 gitinfo kulelu88: Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/
16:04 osse You mean in a philosophical or political sense? No, it doesn't bother me.
16:04 Eugene So, no, not a problem at all.
16:05 osse Not sure I get why it should
16:05 nu11_byte their servers dont run on the laughter and joy of the open source community
16:05 nu11_byte would be nice though
16:05 nu11_byte one day...
16:05 bibble I though repo.or.cz that grawity mentioned is non-profit ?
16:05 tristanp_ joined #git
16:05 rahtgaz joined #git
16:06 Eugene And thus non-profit-able ;-)
16:06 bibble I'm a gitolite convert; it's awesome.
16:06 Eugene I use gitolite where I need to maintain a private repo. Mostly I just throw up on Github with WTFPL, keeping care not to commit any secret keys
16:07 kulelu88 well whether it is philosophical or political or the fact that you feel profiting directly off open source is kind of a moral grey area, I was just asking to see if anybody else felt slightly disturbed by it (I'm also not the person to start internet arguments, but I think maybe discussing this out in the open with other folks will help me better understand)
16:07 osse bibble: correct, it's community driven
16:07 bibble Kewl.
16:07 Eugene Profiting off of open source is totally fine. If you can convince somebody to give you money for a binary that they could make themselves, that's their problem for being too lazy to run `make`
16:08 Eugene That is Red Hat's entire business model
16:08 Eugene I gladly pay somebody else money to deal with problems that I don't watnt to, like Email(Google Apps)
16:08 osse Also worth noting that the second biggest contributor to Git is a github employee
16:08 kulelu88 the only gripe with github for me, although they haven't strong-armed the community, is that a massive open source eco-system runs on closed-source software
16:09 Eugene !gitlab is another commercial host that sprung up around the open-source software
16:09 gitinfo Gitlab is Git hosting similar to Github, but for your own server or on gitlab.com, giving you unlimited free private repositories: http://gitlab.com/
16:10 osse Also also worth noting is that libgit2 powers much of github and it's open source.
16:10 kulelu88 I apologize if this channel is not meant for this type of discussion, but thank you all for contributing to it so far
16:10 bibble gitlab vs bitbucket ?
16:10 Eugene Totally the right place
16:11 bibble Does gitlab enable page/site hosting like github ?
16:11 osse barteks2x: Can Travis use full clones?
16:11 kulelu88 I know that there are tons of already available hosted git services, but is there room in the world for 1 more that is community-driven but is run as a social enterprise to not depend on donations?
16:11 kulelu88 bibble: I believe they do
16:11 kulelu88 static hosting right?
16:12 Eugene Sure, go for it. I won't use it, but somebody will
16:12 bibble Yeah, handy if they do.
16:12 barteks2x I have no idea, but I couldn't find anything that says it's that it's possible. possible workaround would be to do rm -rf *; cd ..; git clone <url>; cd repositoryname
16:12 Eugene You'll need a revenue stream of some sort if you want it to be anything more than a hobby, which is darned difficult on an ad-free gratis servce
16:12 barteks2x wrong channel
16:12 barteks2x or now
16:12 barteks2x not
16:12 editshare-ericf joined #git
16:12 barteks2x I got confused
16:12 MattMaker joined #git
16:13 kulelu88 thanks for the honest feedback Eugene
16:14 osse kulelu88: for-profit doesn't mean evil. IMHO github's position in the community is mostly well deserved.
16:14 algun joined #git
16:15 osse That said, Gorcha would be cute, and with a better compare function
16:15 imachuchu joined #git
16:16 bibble Really, because of gitolite, it's more realistic to have Eugene's work flow. Private server, github for sharing project.
16:16 bibble Cutest, githu.gs
16:16 ash_workz joined #git
16:17 bjpenn joined #git
16:17 osse nice
16:18 osse refugit
16:18 osse Too soon?
16:18 bibble lol
16:18 dhollinger joined #git
16:19 kulelu88 osse: I know that, but the motives for a for-profit organization are different. If tomorrow github experiences a cash-flow issue, do they serve the community or their shareholders first?
16:21 bremner kulelu88: you may be interested in https://mako.cc/writing/hill-free_tools.html
16:21 bremner also, aparently bitkeeper was opensourced recently
16:23 ORiON-- joined #git
16:25 tgunr joined #git
16:25 Mobutils joined #git
16:26 crose joined #git
16:26 osse kulelu88: It's a trade-off. Drastic steps might improve the situation, but it might hurt cash flow in the long term, and shareholders know this.
16:27 kulelu88 osse: but that is the type of issue a community should not have to be held hostage to.
16:27 kulelu88 thanks for that link bremner . reading through it
16:28 osse Bur if a community driven site starts taking in water it might be taken offline, because it has no obligations anyway :p
16:28 acetakwas joined #git
16:29 mdw joined #git
16:29 osse That reminds me. I feel bad for not having donated to Wikipedia
16:30 nnyk_ joined #git
16:31 liquidjamm joined #git
16:32 maestrojed joined #git
16:32 ciampix joined #git
16:32 kulelu88 osse: the difference (to me) is that with enough vested interest in it, especially for public code hosting, the services can withstand financial strains. I don't know how you feel about it, but would you be more willing to volunteer your time if you know that the time is for the community and not to help someone else profit?
16:33 tristanp joined #git
16:34 osse I definitely would
16:35 kulelu88 thanks for the discussion guys. I'll catch up with you all soon.
16:35 bibble I think syncthing.net/ turns all your servers into a single filesystem ? Be cool to build git hosting upon it; people can donate servers.
16:35 trista___ joined #git
16:36 bibble Decentralised, etc
16:36 blackwind_123 joined #git
16:36 nidr0x joined #git
16:37 bibble ...and open source.
16:37 zrts joined #git
16:38 bibble gitolite installed. Bam! :D
16:38 mbrzuzy joined #git
16:38 osse Sounds awesome, provided you regularly check that wantsToExterminateHumanity() returns false.
16:40 tristanp joined #git
16:40 bibble lol, gitnet4eva
16:40 osse Thinking on it in some more I think Refugit would be a fitting name since it will harbour people who have fled github
16:42 bibble Criminals.
16:42 bibble Github Law.
16:42 isxek joined #git
16:43 cdown joined #git
16:44 tristanp_ joined #git
16:45 osse Refugit will be a place where you can seek your fortune freely, without Big Git shutting on you. There will be less silly rules and regulations, like for example it won't harass you if you don't have a readme
16:45 User458764 joined #git
16:45 osse It'll be a free state, where laissez-faire capita.. oh
16:45 osse Never mind
16:50 osse Ouch, tough crowd.
16:50 interne7y joined #git
16:52 tristanp joined #git
16:53 nopf joined #git
16:54 khmarbaise joined #git
16:54 bibble osse, lol
16:55 nnyk_ joined #git
16:55 hahuang65 joined #git
16:55 kami joined #git
16:55 [cust] joined #git
16:58 mjuszczak joined #git
16:59 mjuszczak joined #git
16:59 SteffanW joined #git
16:59 pwnz0r joined #git
17:01 zincrokx joined #git
17:01 Noldorin joined #git
17:01 bitkiller joined #git
17:01 al-damiri joined #git
17:02 Crash1hd joined #git
17:02 micah joined #git
17:03 micah hi, i have a repo with 20 remote branches, and I'm trying to move that repo and all its branches to a new git remote, how can I push all the remotes to that new one? git push --all will only push the ones I have locally
17:03 zincrokx joined #git
17:03 nofacade joined #git
17:03 Crash1hd I have a git ignore file that ignores all dll files but I have a folder that is called Dlls that I cant ignore the dll files in that folder? is there a way for me to tell git to ignore the dlls elsewhere except that folder?
17:03 marked joined #git
17:03 osse micah: you might want to look at git push --mirror
17:03 osse I *think*
17:04 osse that's what you want. Not entirely sure
17:04 osse no wait, it isn't
17:04 osse you have to specify the refspec yourself:  git push newremote oldremote/foo:foo
17:04 zincrokx joined #git
17:04 osse Crash1hd: the lazy thing to do is to git add -f those dlls
17:05 osse but yeah it's possible. Something like this in gitignore:  *.dll \n !dir/*.dll
17:05 gitinfo [!directory] Git does not track directories, only files in directories.  Create a nonce file as a placeholder (eg .gitignore) to create the directory or create the directory as part of a post-checkout hook
17:07 micah osse: hm that doesn't work
17:07 johnny56 joined #git
17:07 osse how so
17:07 micah osse: git push newremote origin/dovecot_changes:dovecot_changes <-- is what i did
17:07 osse is it actually called newremote? :3
17:07 skyjumper left #git
17:07 micah osse: yes
17:08 tristanp_ joined #git
17:08 micah osse: error: unable to push to unqualified destination: dovecot_changes The destination refspec neither matches an existing ref on the remote nor begins with refs/, and we are unable to guess a prefix based on the source ref.
17:08 osse ah
17:08 osse git push newremote origin/dovecot_changes:refs/heads/dovecot_changes
17:08 shinnya joined #git
17:08 micah osse: ok, that works, now i need to script it :o
17:09 clevodearia joined #git
17:09 Ahmed90 joined #git
17:10 osse micah: ez pz
17:11 osse micah:  git for-each-ref --format='%(refname)' 'refs/remotes/origin' | while read -r x; do y=${x#refs/remotes/origin/}; git push newremote "$x:$y"; done
17:11 * osse crosses fingers
17:12 diegoaguilar joined #git
17:13 micah osse: only problem is it includes HEAD, so I can just cut that out
17:14 osse micah: the remote needs a HEAD too. But maybe it already has one
17:14 micah osse: git for-each-ref --format='%(refname)' 'refs/remotes/origin' | while read -r x; do y=${x#refs/remotes/origin/}; git push newremote "$x:refs/heads/$y"; done
17:14 micah osse: needed to add refs/heads/ in there
17:14 osse goddammit
17:15 osse i knew it was too good to be true
17:15 mjuszczak joined #git
17:15 osse One does not simply write a bunch of shell code directly on irc
17:15 micah heheh
17:15 barteks2x I wanted to solve issue with travis, and now I'm trying to figure out why I can;t compile git...
17:16 osse barteks2x: have you seen that gif from malcolm in the middle?
17:16 osse if you have you know which one I mean
17:16 SwiftMatt joined #git
17:16 rbr joined #git
17:16 tristanp joined #git
17:16 barteks2x no
17:17 barteks2x and apparently now I know that I choulsn't copypaste anything from stackoverflow. ever. there was a hidden forkbomb in there
17:17 adron joined #git
17:18 osse barteks2x: http://i.giphy.com/yoJC2GIU95RkDuPhbG.gif
17:21 barteks2x screw it, I will just add that hacky workaround that will almost definitely break with concurrent builds
17:21 fuchstronaut joined #git
17:26 mwleeds joined #git
17:27 mwleeds joined #git
17:28 InfoTest1 joined #git
17:30 tristanp_ joined #git
17:34 InfoTest joined #git
17:34 fscala joined #git
17:35 moei joined #git
17:36 tristanp joined #git
17:37 ochorocho__ joined #git
17:38 doebi joined #git
17:41 IWishIKnew joined #git
17:43 timewalker joined #git
17:43 bjpenn joined #git
17:46 The_Phoenix joined #git
17:46 The_Phoenix joined #git
17:46 LeMike joined #git
17:46 clemf joined #git
17:47 The_Phoenix joined #git
17:47 clemf joined #git
17:48 User458764 joined #git
17:50 timewalker joined #git
17:51 interne7y joined #git
17:52 tristanp_ joined #git
17:53 barteks2x Now I just want to kill myself... all of my issues were because I didn't actually push my tags when creating the test repository...
17:53 skunkz joined #git
17:54 jamiejackson joined #git
17:55 barteks2x Why the hell "git push" doesn't warn me when it's not pushing tags?
17:55 sbeller joined #git
17:55 mikecmpbll joined #git
17:56 jamiejackson hi folk. i don't know a lot about git, despite my contributing to some projects with it. i'm using eGit in eclipse. i cloned and checked out a branch from the command line in cygwin (i'm on wndows), and when i look at the project in eGit, some files have the "changed" icon, before i've done anything to them. can you help me understand why?
17:56 satifant joined #git
17:57 jamiejackson also i've just set "git config --global core.autocrlf false" on some advice i've seen on SO. that was after the clone/checkout, fwiw.
18:00 The_Phoenix joined #git
18:01 MattMaker joined #git
18:02 SwiftMatt joined #git
18:06 ramsub07 joined #git
18:07 ramsub07 hello, how do i check the list of changes before doing git add ?
18:07 osse jamiejackson: welcome to the world of different line endings
18:07 GavinMagnus left #git
18:07 osse jamiejackson: i suggest you set that thing back to true. it's the recommended setting on windows
18:07 osse or so they say
18:07 osse ramsub07: git diff
18:09 Xano joined #git
18:10 jamiejackson osse: are there any caveats because i also use cygwin?
18:10 mdw joined #git
18:11 osse jamiejackson: afaik no
18:11 cyan__ joined #git
18:11 kadoban joined #git
18:12 osse jamiejackson: but in my experience using the Git For Windows thing is the most comfortable way of using git on windows
18:14 elsevero joined #git
18:14 sebboh joined #git
18:14 jamiejackson now that i've `git config --global core.autocrlf true`, how do i get the files to show as unchanged?
18:14 tristanp joined #git
18:15 osse if that doesn't happen automatically then something else is wrong I think
18:15 relect joined #git
18:15 osse is it only some files?
18:15 _ikke_ You should normalize before enabling that setting
18:15 fuchstronaut joined #git
18:15 osse try simply deleting them and checking them out
18:15 sebboh Hi.  When I try git stash apply, I get a big merge conflict.  I want to tell git to just replace that file with the whole file from the stash.  How might I do that?
18:16 sebboh 1.9.4.msysgit.1
18:16 osse sebboh: try git checkout --theirs file
18:16 rbr joined #git
18:17 sebboh osse, checkout?  Theirs?  This repo has no remotes.
18:17 tristanp_ joined #git
18:17 jamiejackson osse, what were the issues that led you away from, say git on cygwin?
18:18 jamiejackson osse: yeah, it's just a few files
18:18 osse jamiejackson: i have never tried git on cygwin. I have tried git on msys2 and it was slower and had other defaults for exactly stuff like this
18:18 osse sebboh: yes, yes, doesn't matter
18:19 osse jamiejackson: I wound up in a situation where git for windows said everything was fine and msysgit said it wasn't and vice versa
18:19 nitin joined #git
18:19 sebboh osse: ok.  So I should do that right after the git stash apply, yes?  I thought about it a little after my last message.
18:19 osse jamiejackson: as _ikke_ hinted at, it depends a bit on how the files where added to the repo in the first place
18:20 sebboh ...I'll have to get back there first.
18:20 osse sebboh: git checkout --theirs get "their version" of a file when in a merge conflict
18:20 osse the command doesn't make sense in any other setting
18:20 tristanp joined #git
18:20 osse when applying a stash "our version" is whatever was there before, and "their version" is the stash
18:21 sebboh got it.  Thanks!
18:21 Lunatrius joined #git
18:22 sebboh fatal: '--ours/--theirs' cannot be used with switching branches
18:22 relect joined #git
18:22 tristanp_ joined #git
18:24 osse hmm, seems to work here
18:24 sebboh Can I just grab the fileconent out of the stash?  Not the diff, but the raw file?
18:24 osse are you still in the merge conflict state? (ie. git status says unmerged paths)
18:24 osse git checkout stash -- file
18:24 osse hmm, maybe git checkout refs/stash -- file
18:24 ahmedelgabri joined #git
18:25 xall joined #git
18:25 tristanp joined #git
18:25 whomp joined #git
18:26 sebboh osse, I did a reset HEAD to get back to the no-changes state, confirmed that with 'git status', then I did 'git stash apply', got the conflict warning, and immediately did 'git checkout --theirs' and that's when I got the error about no --ours/--theirs with switching branches.
18:26 diego1 joined #git
18:27 osse reset --hard HEAD?
18:27 osse or just reset?
18:27 sebboh I dunno, next time I'll just open a notepad window and copy and paste what I want to stash.
18:28 sargas joined #git
18:28 sebboh This is like the third time I've tried using stash in three years.  Still doesn't work.
18:28 osse you don't know what kind of reset you ran?
18:29 sebboh osse, the one that git told me to.
18:29 sebboh git reset HEAD path/to/file
18:29 sebboh after that status was 'no changes'.
18:30 osse that most likely mean you had uncommitted changes at the time you ran git stash apply. that means git refuses to apply the stash in the first place, so you naturally won't be in a merge conflict state where you can use --theirs
18:30 osse oh, hmm... then I have no idea.
18:30 whomp_ joined #git
18:30 osse if you're asking after the fact then it can be hard to help unfortunately
18:31 sebboh Well I didn't have a problem until I had a problem, sorry. :)
18:31 JFS joined #git
18:31 SwiftMatt joined #git
18:31 osse sure
18:31 iob- joined #git
18:31 osse but as it stand you made the problem go away right?
18:31 sebboh no, I'm chatting.
18:33 sebboh ok, working directory clean.  I had to use checkout -- path/file after the reset HEAD path/file, but here we are.
18:33 multi_io joined #git
18:33 osse ok, so you aren't done yet. Ok, then I suggest that unless you know you had uncommitted changes other than the one in the stash you do git reset --hard HEAD and then git stash apply again
18:34 sebboh I think I'll git show refs/stash > path/newFile; mv path/newFile path/file; git commit #to master.  This will only fail if applying the stash puts me on something other than master, I think.
18:35 sebboh osse, I don't understand.  I don't have uncommitted changes, anywhere, except in the stash.
18:36 diego1 joined #git
18:36 osse then !repro
18:36 gitinfo [!transcript] Please paste (using https://gist.github.com/ or https://vomitb.in/ ) a transcript ( https://gist.github.com/2415442 ) of your terminal session so we can see exactly what you see
18:36 sebboh wait you want me to reset, apply, and then checkout --theirs?  I can't see how it will be different than last time I did that.
18:37 tristanp joined #git
18:37 osse if you're sure you don't have other changes then it won't
18:37 osse i thought you did
18:37 osse i don't know then. it works here :/
18:38 osse sebboh: the error you get from git stash is:
18:38 osse Auto-merging file.txt                                                                                                                   │···
18:38 osse CONFLICT (content): Merge conflict in file.txt                                                                                          │···
18:38 osse right ?
18:38 sebboh yep
18:39 mwhooker joined #git
18:39 osse and you did ''git checkout --theirs file.txt'' right?
18:39 osse with whathever is the correct path to the file, nothing more, nothing less?
18:39 sebboh no, I didn't specify a path, just ''git checkout --theirs''
18:40 sebboh as I believed was the recommendation... ;)
18:40 osse there's your problem
18:40 sebboh ok.  Works for me.
18:40 osse or my problem, as it were
18:40 sebboh ok.  Transcript time.
18:40 osse for some silly git status will say it's still unmerged, but the contents should be what  you expect
18:43 nwkj86 joined #git
18:44 dreiss joined #git
18:45 Fijit2 joined #git
18:45 sebboh https://gist.github.com/daveloyall the one created today, osse.  (Sorry, I can't copy/paste right now!)
18:45 tristanp_ joined #git
18:46 osse "gov", there's your problem
18:46 osse lololol
18:46 osse no, but seriously now you can do a 'git reset' and everything will be fine
18:46 osse then you can further modify or commit
18:47 dopesong joined #git
18:47 sebboh What? I can reset and keep the content? so then it's just a modified file?  That's fine.  How?
18:47 osse had this been an actual merge conflict from when you did git merge that command would have staged and resolved the conflict so that you could commit. but stash seems to be weird
18:48 osse sebboh: yes, exactly. just a plain 'git reset' with no frills should do the trick
18:48 sebboh stash does seem to be weird.
18:48 sebboh vunderbar
18:49 whomp joined #git
18:49 paul424 joined #git
18:51 ErrorHead joined #git
18:52 sebboh perfect!  Thanks osse.  The whole reason I did the stash was, I had a great commit, where I change each call to a logging method from the old way, to the new way.  And when I went to commit it, the diff was ugly because I'd also applied auto-formatting.  So I thought, ok, I'll just stash this, go back to HEAD, perform automatic formatting there, commit that, then apply the stash to the NEW HEAD.  And though the path was a little rocky,
18:52 sebboh that's the result I got.  Now the history will show the auto-formatting as having occurred right before changing the way I call the logger.  As desired.
18:52 UrsoBranco joined #git
18:52 interne7y joined #git
18:53 osse \o/
18:54 LionsMane joined #git
18:54 osse what was the merge conflict then? just whitespace?
18:55 tristanp joined #git
18:56 jamiejackson joined #git
18:56 cdown joined #git
18:57 idmistir_ joined #git
18:57 jamiejackson kiwiirc died on me. anyway, osse, i'm still having trouble, even with git for windows: https://gist.github.com/jamiejackson/625d3f7eded0d53ab3d8034cb5079c98
18:58 jamiejackson those are the files that keep coming up as having changes
18:59 jamiejackson i want to sync from the upstream of my fork, but i'm nervous to do it until i know what's going on locally
19:00 Grimm_ make a local backup of the repository before syncing
19:00 Noldorin joined #git
19:04 mohsenazimi joined #git
19:05 mohsenazimi Hello
19:05 gitinfo mohsenazimi: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
19:06 josuebrunel joined #git
19:06 jamiejackson Grimm_: i didn't change those files. i want them to show as unchanged
19:06 mohsenazimi When I resolve a conflict, I un-comment the list of conflicted files. Is it fair to ask Git to add an option to commit not to comment out those file names?
19:07 mohsenazimi People are trying smart things to get around it: http://stackoverflow.com/questions/34899662/git-default-merge-commit-message-not-including-conflicts
19:07 SonikBoom joined #git
19:08 preaction mohsenazimi: as the bottom answer shows, the merge commit itself has to show what changes were made to resolve the conflicts. they're in the commit already
19:09 ramsub07 joined #git
19:09 mohsenazimi preaction: how you read them if you don't uncomment them?
19:09 jbitdrop joined #git
19:10 preaction mohsenazimi: they're in the commit, not the message, the commit itself. you can use "git show" or "git log --stat" or other commands that show you what a commit has inside
19:11 floatingpoint joined #git
19:11 mohsenazimi Thank you preaction! Learned something useful today! :)
19:13 dopesong joined #git
19:14 tristanp joined #git
19:14 TiagoEDGE joined #git
19:16 cyan__ joined #git
19:17 ramsub07 joined #git
19:18 dopesong joined #git
19:20 dsdeiz joined #git
19:23 ochorocho__ joined #git
19:24 ahmedelgabri joined #git
19:25 fuchstronaut joined #git
19:25 Rooxo joined #git
19:25 mda1 joined #git
19:26 TiagoEDGE joined #git
19:29 mjuszczak joined #git
19:31 inflames joined #git
19:31 espindola joined #git
19:32 drodger joined #git
19:33 ahmed_elgabri joined #git
19:37 paul424 hello , hello I have an push request from the past here https://github.com/OpenDungeons/OpenDungeons/pull/954/files
19:37 paul424 problem is it's about 14 commits ... how do I 1) merge them into one ( which was done in this view ) 2) apply it to existing code
19:38 grawity I don't see how that's a problem tbh
19:38 grawity just `git merge` the whole branch at once
19:39 LeBlaaanc joined #git
19:40 hahuang65 joined #git
19:40 kpease joined #git
19:41 ocbtec joined #git
19:46 tristanp_ joined #git
19:47 SwiftMatt joined #git
19:48 paul424 grawity: problem is my branch does not contain that code anymore
19:48 paul424 development was synced with upstream development
19:48 paul424 origin development was synced with upstream development
19:48 nadim joined #git
19:49 MattMaker joined #git
19:49 paul424 what  I could do is restore the commit from that link
19:50 tristanp joined #git
19:51 grawity git fetch https://github.com/OpenDungeons/OpenDungeons refs/pull/954/head:culling
19:52 tristanp_ joined #git
19:52 paul424 grawity: and what then ?
19:53 Gianormagantrous joined #git
19:53 grawity you'll have a branch named "culling"
19:53 interne7y joined #git
19:53 grawity do whatever you want with it
19:53 paul424 ahhh , now I  do REMEMBER !
19:54 skunkz joined #git
19:54 rahtgaz joined #git
19:57 ril joined #git
19:57 mizu_no_oto joined #git
19:57 tvw joined #git
19:57 cyan__ joined #git
19:58 L8D joined #git
19:59 mwhooker joined #git
20:00 paul424 grawity: how do  I sync to upstream development ?
20:00 tristanp joined #git
20:00 drodger joined #git
20:01 Doginal joined #git
20:03 WayToDoor joined #git
20:03 Doginal hey, im using github and i have setup ssh keys, but for some reason when i try to push its always asking for my password. I have followed the github walkthough and I have had it working on another repo im not sure why my new repo is asking for a password. with ‘ssh-add -l’ i see my keys. It seems git isnt using them at all.
20:04 osse Doginal: what does git remote -v say
20:04 Doginal its the repo plus https://USERNAME@repo
20:04 WayToDoor joined #git
20:05 Eugene Doginal - you'll need to change your repo URL to use ssh://
20:05 Eugene man git-remote; see set-url
20:05 gitinfo the git-remote manpage is available at http://jk.gs/git-remote.html
20:05 Eugene The github URL format is: ssh://git@github.com/Username/Repository.git
20:06 Doginal is it the same for a repo owned by an organizations?
20:06 osse Doginal: http://imgur.com/7qtEqk6
20:06 osse click that
20:06 osse use what it says
20:07 Doginal okay thanks osse
20:07 osse http://imgur.com/waRZfWX
20:07 osse heh, this is what I meant to upload
20:07 Doginal i figured that much haha
20:08 cdown_ joined #git
20:09 Doginal osse, thanks!
20:10 Doginal its working again woot woot! :D
20:10 ToBeCloud joined #git
20:10 CheckDavid joined #git
20:12 NeverDie joined #git
20:13 EricPoe joined #git
20:16 nettoweb joined #git
20:17 bibble Eugene, is that github url for sure ? Normally ssh standard format is ":" after <host>. Shame if github has altered it.
20:17 grawity there is no "standard format" for it
20:17 Eugene man git-clone shows what `git` expects it to be
20:17 gitinfo the git-clone manpage is available at http://jk.gs/git-clone.html
20:17 grawity so git supports both the semi-standard URL format – scheme://user@host/path – and the rcp/scp format – user@host:path
20:17 bibble ? Normally ssh anything is, host:path
20:18 grawity yes, and normally sftp is sftp://host/path
20:18 Eugene ssh:// will be inferred if no scheme is provided
20:18 bibble Ahh, see cool.
20:18 Eugene I prefer to be explicit
20:18 rcpp85 joined #git
20:18 NeverDie joined #git
20:18 paul424 I have solved all the merge conflicts , what now git merge --continue ?
20:18 paul424 that does not work
20:18 grawity git commit
20:20 nwkj86 joined #git
20:21 rcpp85 Hello! I'm having a conflict merge (a failed auto-merge) after a "pull" done in a hook. To solve this conflict I must have access to the server where the "pull" is happening? Are there othe possible solutions?
20:21 mjuszczak joined #git
20:22 osse !deploy > rcpp85
20:22 gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
20:22 osse rcpp85: TL:DR; don't use pull
20:22 italoacasas joined #git
20:25 rcpp85 osse: Thanks for the tips! But: I'm not the one risponsible for the "deployment". I'm not the one who wrote the hook. Actually, until some days ago I was not even  aware of how the "deploy" worked.  We were having a lot of troubles between our branches and some of the prepoduction and production server. Then, they did a "fresh start",which made eveyrthing fine until a push to the prepod - which triggers this hook.
20:25 mjuszcza_ joined #git
20:25 Doginal joined #git
20:25 tristanp_ joined #git
20:26 pandeiro joined #git
20:26 fuchstronaut joined #git
20:26 rcpp85 osse: so, to fix conflicts I must have access to where the "pull" was made, correct?
20:27 mwhooker joined #git
20:27 nu11_byte joined #git
20:28 nivag joined #git
20:28 tristanp joined #git
20:33 eijk joined #git
20:34 Dreamer3 joined #git
20:35 eletuchy joined #git
20:35 ojacobson rcpp85: if you're going to run merges, you're going to have to have a way to resolve conflicts
20:35 ojacobson conflicts are exactly the cases that git doesn't have the information to resolve automatically, so that process is necessarily manual
20:35 ojacobson hence: don't use merge in an automated context. :)
20:35 ojacobson if you must, then design for it; there's no "never conflict" mode you can use
20:36 Vinnie_win joined #git
20:36 SwiftMatt joined #git
20:36 ojacobson you can likely replace the `pull` (which is a merge) with something from sitaram's deploy list, though.
20:36 eletuchy joined #git
20:36 linuxmint joined #git
20:36 xanadu_ joined #git
20:36 rnsanchez_ joined #git
20:37 acetakwas joined #git
20:38 rcpp85 ojacobson: I see! The biggest problem for me, is not having fully access to the repositories. The servers are managed by a third part. Is quite a long story... but in short: I will have to ask them access to solve these - since they probably won't know how to solve that.
20:38 rcpp85 Thank you!
20:38 LionsMane joined #git
20:39 NeverDie_ joined #git
20:39 tristanp joined #git
20:40 anth0ny joined #git
20:42 bruce_lee joined #git
20:42 m0viefreak joined #git
20:45 mdw joined #git
20:47 GeneralVagueness joined #git
20:48 arooni joined #git
20:50 MattMaker joined #git
20:50 tristanp_ joined #git
20:52 eijk_ joined #git
20:52 MattMaker joined #git
20:53 nilsi joined #git
20:54 interne7y joined #git
20:55 L8D joined #git
20:58 MrWoohoo joined #git
20:59 hahuang65 joined #git
21:00 LeMike joined #git
21:00 rcpp85 Time to go! Thanks for the help, fellas.
21:01 liquidjamm joined #git
21:01 blackpajamas joined #git
21:02 ahmedelgabri joined #git
21:02 GavinMagnus joined #git
21:02 m_antis joined #git
21:03 nnyk_ joined #git
21:03 GavinMagnus left #git
21:07 diegoaguilar joined #git
21:09 AaronMT joined #git
21:11 liamonade joined #git
21:11 GavinMagnus joined #git
21:15 tristanp joined #git
21:16 leonarth joined #git
21:16 iob- joined #git
21:17 GavinMagnus left #git
21:19 italoacasas joined #git
21:19 tristanp_ joined #git
21:21 mellernoia joined #git
21:21 tristanp joined #git
21:23 mwhooker joined #git
21:24 ochorocho__ joined #git
21:26 elastix joined #git
21:28 SwiftMatt joined #git
21:28 t0by joined #git
21:28 nnyk_ joined #git
21:28 Celelibi joined #git
21:37 tristanp joined #git
21:42 mjuszczak joined #git
21:42 nwkj86 joined #git
21:42 mjuszczak joined #git
21:44 m0j0 joined #git
21:44 diosmiux joined #git
21:46 dopesong joined #git
21:48 fscala joined #git
21:53 moei joined #git
21:55 interne7y joined #git
21:55 skunkz joined #git
21:55 rbr joined #git
21:56 tristanp_ joined #git
21:56 Dougie187 left #git
21:58 nadim joined #git
21:59 Noldorin joined #git
22:04 iamthad joined #git
22:07 tristanp joined #git
22:12 ploop joined #git
22:13 tristanp_ joined #git
22:14 Topic for #git is now Welcome to #git, the place for git help and musical ditties | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.8.2 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | Dunnn, dudududu Du, dudududu Du, DuuDuu duuduu mmmmm https://madeitwor.se/git-jokes
22:15 gaen joined #git
22:15 mikecmpbll joined #git
22:17 radsy joined #git
22:18 Noldorin joined #git
22:19 UTAN_dev joined #git
22:22 tristanp joined #git
22:23 xlegoman joined #git
22:23 cdown joined #git
22:24 avar Eugene: Since you're changing the topic, the current stable is 2.8.3
22:24 Eugene .version
22:24 Topic for #git is now Welcome to #git, the place for git help and musical ditties | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.8.3 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | Dunnn, dudududu Du, dudududu Du, DuuDuu duuduu mmmmm https://madeitwor.se/git-jokes
22:24 Eugene !botsnack
22:24 gitinfo Om nom nom
22:27 fuchstronaut joined #git
22:29 SimonNa joined #git
22:33 tristanp joined #git
22:35 tristanp_ joined #git
22:36 tristanp_ joined #git
22:38 tristanp joined #git
22:41 johnchalekson joined #git
22:41 johnchalekson joined #git
22:42 johnchalekson joined #git
22:42 madewokherd joined #git
22:47 dopesong joined #git
22:49 mikecmpbll joined #git
22:51 whitby joined #git
22:53 artemisart joined #git
22:56 tristanp_ joined #git
22:57 johnmilton joined #git
22:59 tristanp joined #git
23:01 tristanp_ joined #git
23:02 dsantiago joined #git
23:05 rxc joined #git
23:05 rxc joined #git
23:06 tristanp joined #git
23:07 hahuang65 joined #git
23:07 nwkj86 joined #git
23:10 tristanp_ joined #git
23:10 mikecmpbll joined #git
23:12 annoymouse joined #git
23:12 ecksit joined #git
23:12 IWishIKnew joined #git
23:16 son joined #git
23:16 dsantiago joined #git
23:17 GavinMagnus joined #git
23:17 micah left #git
23:18 GavinMagnus left #git
23:21 tristanp joined #git
23:22 rgrinberg joined #git
23:24 cdown_ joined #git
23:25 ochorocho__ joined #git
23:27 chipmadness joined #git
23:27 chipmadness is there a way to change a previous commit message
23:28 drbean joined #git
23:28 ojacobson Sure, but !rewrite
23:28 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
23:28 ojacobson If it's a published commit, you're probably better off attaching a note to it (man git-notes)
23:28 gitinfo the git-notes manpage is available at http://jk.gs/git-notes.html
23:29 Noldorin joined #git
23:29 GavinMagnus joined #git
23:29 chipmadness ojacobson thank you!
23:29 melbatoast joined #git
23:30 FroMaster joined #git
23:30 tristanp_ joined #git
23:31 GavinMagnus left #git
23:31 Stath joined #git
23:33 iamthad joined #git
23:33 tristanp joined #git
23:34 meLon joined #git
23:35 pamplemousse joined #git
23:36 thunderbirdircte joined #git
23:37 tristanp_ joined #git
23:37 basiclaser joined #git
23:38 Blkt joined #git
23:38 drodger joined #git
23:38 jaequery joined #git
23:42 Olipro_ joined #git
23:42 diogenese joined #git
23:44 _ikke_ joined #git
23:44 diosmiux joined #git
23:47 pamplemousse joined #git
23:47 jaequery joined #git
23:48 tristanp joined #git
23:49 heinrich5991 joined #git
23:50 mattfoo joined #git
23:50 marcosmf joined #git
23:54 theskillwithin joined #git
23:56 interne7y joined #git
23:56 skunkz joined #git
23:58 ril joined #git
23:59 fstd joined #git

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