Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-05-19

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

All times shown according to UTC.

Time Nick Message
00:02 rominronin joined #git
00:03 HenriettaSalt joined #git
00:04 unbalancedparen joined #git
00:07 mwhooker joined #git
00:07 mda1 joined #git
00:08 jaguarmagenta joined #git
00:12 chachasmooth_ joined #git
00:13 MattMaker joined #git
00:14 livingstn joined #git
00:15 eka joined #git
00:17 rkazak joined #git
00:17 eletuchy joined #git
00:26 diegoviola joined #git
00:28 fahadash joined #git
00:29 nobso joined #git
00:31 tomog999 joined #git
00:32 zhiwliu joined #git
00:32 zhiwliu left #git
00:32 _xor joined #git
00:32 inflames joined #git
00:33 fuchstronaut joined #git
00:39 mda1 joined #git
00:42 Sasazuka joined #git
00:44 d0nn1e joined #git
00:46 prsn joined #git
00:47 nettoweb joined #git
00:47 mwhooker joined #git
00:49 maestrojed joined #git
00:49 mda1 joined #git
00:54 aidalgol Why might the changes listed by `tig abbadab..branch -- path/to/file.cpp` be missing some that are listed by `gitk abbadab..branch -- path/to/file.cpp` ?
00:57 aidalgol oh, never mind, they're just laying out the graph slightly differently...
00:59 clemf joined #git
01:00 clemf joined #git
01:01 Ali1 joined #git
01:01 MattMaker joined #git
01:03 fracting joined #git
01:03 rominronin joined #git
01:04 MattMaker joined #git
01:05 pks joined #git
01:05 purrkode joined #git
01:05 dermoth__ joined #git
01:05 clemf_ joined #git
01:11 raijin joined #git
01:14 Gsham joined #git
01:14 jaguarmagenta joined #git
01:14 s00pcan_ joined #git
01:16 baba_ joined #git
01:18 aeontech joined #git
01:18 milki huh
01:18 milki new bare repos have a branches folder
01:19 milki whats that for
01:19 milki man gitrepository-layout
01:19 gitinfo the gitrepository-layout manpage is available at http://jk.gs/gitrepository-layout.html
01:19 milki huh i never noticed itbefore
01:23 Voyaging joined #git
01:24 Asiajey joined #git
01:25 baba_ joined #git
01:27 ojacobson I don't think there are any use cases for that sysem at this point
01:27 slappymcfry joined #git
01:32 steve joined #git
01:32 rgrinberg joined #git
01:33 sd512 joined #git
01:33 flaguy48 left #git
01:35 dermoth__ joined #git
01:35 haway joined #git
01:36 fuchstronaut joined #git
01:37 fscala joined #git
01:37 Dougie187 joined #git
01:37 ochorocho__ joined #git
01:46 rkazak joined #git
01:46 Mikaela 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 Gsham joined #git
01:51 mablae joined #git
01:51 livingstn joined #git
01:52 sangy joined #git
01:54 mwhooker joined #git
01:54 harish joined #git
01:55 Gsham joined #git
01:57 gnoel joined #git
02:02 Dougie187 left #git
02:02 mmlb joined #git
02:03 joki joined #git
02:04 rominronin joined #git
02:06 jaguarmagenta joined #git
02:09 mizu_no_oto joined #git
02:10 jaguarmagenta joined #git
02:10 rainbyte joined #git
02:10 aghalarp joined #git
02:13 slappymcfry joined #git
02:15 thiago joined #git
02:16 nerium Is there a reason ”git rev-list --before=xx” doesn’t include commits from xx while ”git rev-list —after=’xx’” does it?
02:16 nerium Feels very inconsitent
02:17 dalias having half-open intervals is actually more useful/consistent
02:17 dalias because they add up to the right thing
02:18 dalias otherwise you get missing commits or duplicate commits when adding up ranges of commits
02:19 ojacobson nerium: to make that a little more concrete, if you're paging through commits with --after and --before, the set "after A, before B" directly abuts the set of commits "after B, before C" - there are no intervening commits, and there are no commits in both sets
02:19 ojacobson no commit is in both sets, rather
02:20 ojacobson --not --after and --not --before can be combined with --after and --before to give closed intervals (or fully open intervals) if you prefer, I believe,
02:20 ojacobson oop, no, I'm lying
02:21 hexagoxel joined #git
02:21 kadoban joined #git
02:21 stantonk joined #git
02:21 Lunatrius` joined #git
02:24 nerium ojacobson: Okay, interesting
02:25 mda1 joined #git
02:27 nerium Can’t wrap my head around it. I’m trying to use the commits for git blame, but the results are not correct
02:29 Limes_ joined #git
02:30 nerium So; git rev-list --after='2008-09-01' master | tail -1 => yyy, git rev-list --before='2008-09-03' --first-parent master | head -1 => xxx, then git blame yyy..xxx should be correct, right?
02:30 dfeuer joined #git
02:30 dfeuer joined #git
02:31 ojacobson git rev-list --after DATE --reverse -1 master
02:32 nerium ojacobson: That returns HEAD
02:34 nerium ojacobson: Yeah, samething as git log | head -1
02:35 nerium I don’t think —reverse works. Tried it before without results
02:37 fuchstronaut joined #git
02:38 johnwhitton joined #git
02:40 Bowel joined #git
02:43 nerium Also, why is git rev-list --first-parent commit^1 -1, where ”commit” is a merge, jumping 2 commits back?
02:45 NeverDie joined #git
02:46 jonmorehouse joined #git
02:49 dalias nerium, presumably because you're asing for the first parent of the previous commit; i'm not sure about that tho
02:52 nerium dalias: Aha, thanks
02:57 kutsuya joined #git
03:01 chachasmooth joined #git
03:05 rominronin joined #git
03:09 faux___ joined #git
03:17 nettoweb joined #git
03:17 f3r70rr35f joined #git
03:18 purrkode joined #git
03:20 CJKinni joined #git
03:22 slappymcfry joined #git
03:24 peepsalot joined #git
03:25 Goplat joined #git
03:26 dalias my setting of color.ui = false is being ignored for git format-patch --stdout
03:26 dalias why?
03:28 aghalarp joined #git
03:30 dalias it looks like this is intentional for some awful reason
03:30 prsn joined #git
03:30 dalias maybe idiots have color.ui=always in their config and were sending corrupt patches? :(
03:32 dalias *sigh* at least i found a workaround
03:32 dalias a [color "diff"] section with all colors set to "normal
03:33 * Arahael _likes_ colour.
03:33 dalias even if you like it, git format-patch should respect the option to suppress it
03:34 dalias and having it set to always is certainly broen
03:34 dalias broken
03:34 PryMar56 joined #git
03:35 zivester joined #git
03:37 mischat joined #git
03:38 fuchstronaut joined #git
03:38 ochorocho__ joined #git
03:41 nobso joined #git
03:45 mehola joined #git
03:47 Lipp joined #git
03:49 bronson joined #git
03:50 elastix joined #git
03:51 boombatower joined #git
03:53 stantonk joined #git
03:56 SimonNa joined #git
03:59 fracting joined #git
04:01 josuebrunel joined #git
04:03 d4rklit3 joined #git
04:04 nettoweb joined #git
04:05 shauna joined #git
04:06 rominronin joined #git
04:12 blackwind_123 joined #git
04:13 br0kenman joined #git
04:16 Mobutils joined #git
04:17 rkazak joined #git
04:18 M-mistake joined #git
04:21 aeontech joined #git
04:22 hendry joined #git
04:22 slappymcfry joined #git
04:23 irqq joined #git
04:24 BSaboia joined #git
04:25 baba_ joined #git
04:25 dharriga1 joined #git
04:33 aghalarp joined #git
04:34 Macaveli joined #git
04:39 fuchstronaut joined #git
04:39 ochorocho__ joined #git
04:40 aeontech joined #git
04:44 aavrug joined #git
04:46 Cabanossi joined #git
04:47 tinajohnson___ joined #git
04:55 jaguarmagenta joined #git
04:56 stantonk joined #git
04:56 fracting joined #git
05:00 WayToDoor joined #git
05:02 aavrug joined #git
05:02 Vinnie_win joined #git
05:04 cagedwisdom joined #git
05:04 hahuang65 joined #git
05:05 sauvin joined #git
05:07 rominronin joined #git
05:13 fmcgeough joined #git
05:13 fracting joined #git
05:17 mariuscc joined #git
05:18 subhojit777 joined #git
05:18 aeontech joined #git
05:18 Psychiatrist joined #git
05:22 baba_ joined #git
05:25 mischat joined #git
05:27 mofai joined #git
05:27 qt-x joined #git
05:27 stantonk joined #git
05:39 fuchstronaut joined #git
05:39 d0nn1e joined #git
05:39 rscata joined #git
05:41 xall joined #git
05:43 muthu joined #git
05:45 clemf joined #git
05:48 ThomasLocke joined #git
05:48 JeroenT joined #git
05:49 xall_ joined #git
05:49 snh joined #git
05:51 programmr joined #git
05:51 bronson joined #git
05:55 sd512 the difference between fetch and pull is that pull autocommits, correct?
06:02 aeontech joined #git
06:05 joako joined #git
06:06 hahuang65 joined #git
06:06 aghalarp joined #git
06:08 blackwind_123 joined #git
06:08 rominronin joined #git
06:10 Lipp joined #git
06:10 zeroed joined #git
06:12 LucaTM joined #git
06:14 fees joined #git
06:15 Seveas no
06:15 Seveas the difference between fetch and pull is that pull = fetch + (merge or rebase)
06:16 TomyLobo joined #git
06:20 maxis joined #git
06:22 acolson joined #git
06:27 aeontech joined #git
06:27 Oog joined #git
06:27 dharrigan joined #git
06:30 joako joined #git
06:32 baba_ joined #git
06:32 inflames joined #git
06:32 Panacea joined #git
06:33 stantonk joined #git
06:33 lamppid joined #git
06:35 SteffanW joined #git
06:37 khmarbaise joined #git
06:38 acetakwas joined #git
06:39 aavrug joined #git
06:39 ochorocho__ joined #git
06:40 fuchstronaut joined #git
06:41 pantsman- joined #git
06:42 tvw joined #git
06:42 zarel joined #git
06:42 [Brain] joined #git
06:42 aeontech joined #git
06:44 josuebrunel joined #git
06:44 eletuchy joined #git
06:46 mohabaks joined #git
06:47 moriko joined #git
06:52 mischat joined #git
06:53 weyer joined #git
06:54 Waskeend joined #git
06:58 aghalarp joined #git
07:01 rominronin joined #git
07:03 xbob joined #git
07:06 neanderslob joined #git
07:08 emPi joined #git
07:09 cyan__ joined #git
07:10 Junior` joined #git
07:11 Junior` left #git
07:11 Junior joined #git
07:12 mischat joined #git
07:15 JeroenT_ joined #git
07:16 jceb joined #git
07:16 TomyWork joined #git
07:17 BinGOs joined #git
07:17 bojangles joined #git
07:18 bojangles hey i've got a commit i want to do, however i havent pulled the most recent version. is it possible to save my version and then pull, then implement my version onto it?
07:18 osse yes
07:18 bojangles what commands would i need to do osse ?
07:19 osse commit as usual, then pull. by how you phrased the question you want pull --rebase
07:20 Sceorem joined #git
07:22 aavrug joined #git
07:24 lindenle joined #git
07:24 j416 bojangles: normally you would work on a separate branch to make life easier, then rebase that on top of the new things that happen in the other branch.
07:24 j416 bojangles: well, that's a common way to work, at least
07:27 pkug joined #git
07:28 djb-irc joined #git
07:28 rinwa joined #git
07:29 ElGrandeAmigo joined #git
07:37 txdv joined #git
07:38 txdv if i do a checkout of a specific file of a specific commit, it stages the file
07:38 ceaucari joined #git
07:38 txdv is there a way to just get an older version of a file without staging it?
07:39 chris_wot joined #git
07:40 ahmedelgabri joined #git
07:40 fwdit joined #git
07:41 fuchstronaut joined #git
07:41 osse txdv: do a git reset after
07:41 osse txdv: or do  git show commit:file > file
07:42 Garen joined #git
07:42 fracting joined #git
07:46 prsn joined #git
07:47 DevAntoine joined #git
07:47 Balliad joined #git
07:49 Guest36303 joined #git
07:50 JeroenT joined #git
07:53 bronson joined #git
07:53 clmsy joined #git
07:55 mischat joined #git
07:56 stantonk joined #git
07:57 attila_ joined #git
07:58 blahdodo joined #git
07:59 kurkale6ka joined #git
08:01 Levitator joined #git
08:03 xall joined #git
08:08 dars joined #git
08:08 dars hey, i want to know how to deploy my website on github ?
08:09 dars I just made a static page and it ran well
08:09 dars by using .github.io extension
08:09 dars but what if i want to add database also
08:09 hahuang65 joined #git
08:09 dars I mean using PHP
08:09 dars ?
08:10 osse I doubt they offer that, but ask #github
08:11 txdv github is static pages only
08:12 dars txdv, i saw some peoples using it for dynamic pages also
08:12 aavrug joined #git
08:12 Anupkumar joined #git
08:12 bongjovi joined #git
08:12 bgamari- joined #git
08:12 txdv dars: what do you mean with 'dynamic' pages?
08:13 dars txdv using databases
08:13 dars also
08:13 Xethron joined #git
08:13 eka joined #git
08:13 txdv i doubt that
08:13 txdv anyway, they have the code on github then
08:13 txdv and you can look into it
08:13 Anupkumar how can I clone a branch from master to remote?
08:14 dars Anupkumar, git clone <branchaddress>
08:14 Ch4rAss joined #git
08:14 Gonzo89 joined #git
08:14 txdv that question doesn't make sense, master is a branch, remote is a ... remote?
08:14 dan2k3k4 joined #git
08:15 settermjd joined #git
08:15 Anupkumar I already have a fork of the master cloned, now I want to clone a branch from master to local
08:16 dars txdv , i am confused here , If code is available on git then someone can hack the website very easily
08:16 txdv not if it is just static content
08:16 dars No but there are some with the databases also
08:17 txdv there are none
08:17 txdv either you show me one or there are none
08:17 dars ashishkedia.me
08:17 osse russell's teapot
08:17 Anupkumar osse: hi
08:17 dars https://www.ashishkedia.me
08:18 Rish joined #git
08:18 dars Anupkumar, do you type the command that i mentioned
08:18 dars txdv, go to his github repository and watch the database configuration in settings .py file
08:19 osse dars: that the source code for a webpage is on github doesn't mean that github serves the webpage as well
08:20 dars osse, yes I know
08:20 txdv the site is hosted on digital oceans
08:20 txdv not on github
08:20 txdv this is just the source code of the site
08:20 dars txdv, can u explain please
08:21 txdv i dont know how to express myself simpler
08:21 jon-mac joined #git
08:21 Trieste joined #git
08:21 txdv this site is using another service (digital oceans) for hosting
08:21 txdv he just put the code on github
08:22 osse dars: you can have C++ source code on github. That doesn't mean Github can compile it for you.  You can have database configuration on Github. That doesn't mean Github can run the database for you.   The stuff on Github is simply a copy of the code.
08:22 sssilver joined #git
08:22 fracting joined #git
08:23 dars txdv, how you cam to know the site is hosted by digital ocean
08:23 dars osse , can't i see the password of database then
08:24 txdv resolve the hostname, get the ip, use whois
08:24 osse dars: if  he put it there, sure. this seems like a guy who knows what's he's doing, so probably not.
08:25 txdv is the password in the configuration file?
08:25 osse not that I can see
08:25 Lipp joined #git
08:25 SimonNa joined #git
08:26 acetakwas joined #git
08:26 matiaslanzi joined #git
08:27 kam270 joined #git
08:27 dars osse, i am not understanding if the website is hosted by something else then why he has put code on github
08:27 dars '?
08:28 osse to show it
08:28 sanketdg joined #git
08:29 dars osse, that means the code their is just for show off and all the actual work is done by hosting provider
08:30 kam270 joined #git
08:31 osse dars: the hosting provider didn't write the code
08:31 osse the hosting provider just runs it
08:31 dars yes i mean to say that
08:32 aidalgol joined #git
08:32 osse then yes
08:32 osse it's just for show off
08:32 eletuchy joined #git
08:33 marcogmonteiro joined #git
08:34 baba_ joined #git
08:34 dars osse, isn't it possible that someone might be by looking at his coding style
08:35 dars can hack it
08:35 Ch4rAss joined #git
08:36 Fijit joined #git
08:38 LekeFly joined #git
08:38 osse maybe
08:39 mcemil joined #git
08:40 mohit7 joined #git
08:40 ochorocho__ joined #git
08:40 prsn joined #git
08:41 razieliyo joined #git
08:44 digidog joined #git
08:49 sans_s3r1f joined #git
08:52 osse Is there a significant difference between Git For Windows and the git package in MSYS2 ?
08:55 txdv in what aspect?
08:55 txdv im using msys2 and its slow
08:55 txdv well, i probably dont have enough ram
08:56 notebox joined #git
08:56 slappymcfry joined #git
08:56 osse txdv: any aspect at all.
08:57 osse and yes I think it's slow too. But I'm not convinced that it's actually slower than Git for Windows
08:58 sanketdg joined #git
08:58 JeroenT joined #git
08:59 nnyk joined #git
08:59 gaen joined #git
09:02 ahmed_elgabri joined #git
09:08 Panacea joined #git
09:09 GavinMagnus joined #git
09:10 achadwick joined #git
09:10 raven-au_ joined #git
09:10 bgamari joined #git
09:10 eletuchy joined #git
09:11 txdv to me windows seems to be a slugish system compared to linux
09:11 txdv its just worse for absolutely no reason
09:11 EY___ joined #git
09:11 txdv although, it might be an msys2 effect, as all the os calls are mapped somehow
09:11 davisonio joined #git
09:12 mischat joined #git
09:12 jast mappings aren't that expensive
09:12 dsfreak93 joined #git
09:13 mdw joined #git
09:13 jast I think a big reason you notice that much of a difference with git is that it's optimized for the way linux does filesystem caching (I could claim linux does a much better job of caching in general but I don't actually know what windows does)
09:13 dars_ joined #git
09:13 acetakwas joined #git
09:15 acetakwas joined #git
09:15 marcogmonteiro joined #git
09:18 ronny joined #git
09:19 Sceorem joined #git
09:21 ahmed_elgabri joined #git
09:26 lindenle joined #git
09:29 Lipp_ joined #git
09:31 SteffanW joined #git
09:31 LeMike joined #git
09:33 Trieste joined #git
09:34 Macaveli joined #git
09:35 prsn joined #git
09:36 timini joined #git
09:36 timini Hey how can i get a list of commits which are different between 2 branches?
09:38 Trieste joined #git
09:39 neanderslob joined #git
09:39 ferr joined #git
09:42 fuchstronaut joined #git
09:42 Jellyg00se joined #git
09:43 _ikke_ timini: man git cherry
09:43 gitinfo timini: the git-cherry manpage is available at http://jk.gs/git-cherry.html
09:44 Arahael timini: git log --oneline --boundary BRANCH1 BRANCH2
09:44 Arahael Or git cherry, indeed. Depends on what you mean.
09:47 aghalarp joined #git
09:49 Heisenberg1276 joined #git
09:50 subhojit777 joined #git
09:50 mdw joined #git
09:54 roelmonnens joined #git
09:55 bronson joined #git
09:55 thomas joined #git
09:57 Lipp joined #git
09:57 Lipp joined #git
09:59 ahmed_elgabri joined #git
09:59 Macaveli joined #git
10:00 Trieste joined #git
10:01 sanketdg joined #git
10:01 dars_ joined #git
10:01 aghalarp joined #git
10:01 gaen joined #git
10:03 fscala joined #git
10:03 mase-tech joined #git
10:04 eijk joined #git
10:06 harish joined #git
10:06 Trieste joined #git
10:06 Amnesia joined #git
10:06 Amnesia question, I lost a commit because I pulled, after somehow issued a forced update
10:07 Amnesia is there some way to get a list of all pushed commits?
10:10 Trieste joined #git
10:11 hahuang65 joined #git
10:12 dars__ joined #git
10:13 Panacea joined #git
10:13 eijk joined #git
10:13 nnyk joined #git
10:15 Trieste joined #git
10:17 aavrug joined #git
10:17 roelmonnens joined #git
10:19 EY__ joined #git
10:20 Trieste joined #git
10:20 csd_ joined #git
10:22 EduardoMartins joined #git
10:22 irqq joined #git
10:22 doener joined #git
10:23 fracting joined #git
10:24 Trieste joined #git
10:27 wgrant joined #git
10:28 Amnesia left #git
10:29 prsn joined #git
10:30 toogley joined #git
10:30 rahtgaz joined #git
10:30 dan2k3k4 joined #git
10:30 maxis joined #git
10:34 Arahael Amnesia: did you go?
10:38 nnyk joined #git
10:38 cyan__ joined #git
10:39 ahmed_elgabri joined #git
10:39 ahmedelgabri joined #git
10:41 DomeMaster joined #git
10:41 ochorocho__ joined #git
10:41 DomeMaster Hi
10:41 DomeMaster I'm really struggling here, trying to use "git email"
10:41 DomeMaster I have gmail account but with custom domain
10:41 DomeMaster whatever I do it says "Username and password not accepted. Learn more at....."
10:42 DomeMaster there are tutorials all over the Internet for gmail
10:42 DomeMaster so I assume it must be the custom domain that is the problem
10:42 DomeMaster anyone had similar issue?
10:43 nettoweb joined #git
10:43 fuchstronaut joined #git
10:44 Pieter joined #git
10:45 Seveas DomeMaster: make sure you use your full email address as username
10:45 modin joined #git
10:45 zincrokx joined #git
10:45 Seveas and that you enable 'unsafe applications' in gmail (unsafe just means 'doesn't do oauth but requires a password')
10:47 Panacea joined #git
10:47 aguimaraes1986 joined #git
10:47 DomeMaster ok ty
10:48 DomeMaster doh it was the username after all
10:48 DomeMaster I had @gmail.com thanks for making me double check
10:48 DomeMaster feel embarrassed now :-$
10:48 CuteHoor joined #git
10:52 mehola joined #git
10:53 mozzarella joined #git
10:53 timini hey I want to merge a set of 5 commits from one mybranch to master, but not all the commits on mybranch
10:53 timini what should i use
10:56 LekeFly joined #git
10:56 ToBeCloud joined #git
10:58 kus joined #git
10:59 JeroenT joined #git
11:00 dsdeiz joined #git
11:01 CuteHoor joined #git
11:02 M-mistake joined #git
11:03 heftig timini: git cherry-pick
11:04 roelmonnens joined #git
11:04 Lipp joined #git
11:06 johnmilton joined #git
11:07 rasto joined #git
11:07 fuchstronaut joined #git
11:07 fuchstronaut joined #git
11:08 JeroenT_ joined #git
11:09 Macaveli joined #git
11:10 aguimaraes1986 joined #git
11:11 hahuang65 joined #git
11:12 unbalanced joined #git
11:14 codeliger joined #git
11:18 aavrug joined #git
11:20 Anupkumar joined #git
11:22 TheDcoder joined #git
11:22 TheDcoder Hello, how do I sync a local repo with a remote repo?
11:22 TheDcoder There are no changes in the local repo
11:23 osse most likely pull
11:23 TheDcoder oh...
11:23 TheDcoder ok
11:23 TheDcoder I am a newbie, sorry
11:23 prsn joined #git
11:23 bremner !tutorial
11:23 gitinfo A list of interactive git tutorials: http://try.github.com
11:23 TheDcoder I am spoiled by GitHub's GUI Client
11:23 TheDcoder :P
11:23 bremner hmm. I was thinking of man gittutorial
11:23 gitinfo the gittutorial manpage is available at http://jk.gs/gittutorial.html
11:25 mizu_no_oto joined #git
11:25 bibble joined #git
11:25 flaguy48 joined #git
11:26 JanC_ joined #git
11:27 lindenle joined #git
11:29 TheDcoder Thanks guys :)
11:29 TheDcoder Bye!
11:30 rnsanchez joined #git
11:32 dsfreak93 joined #git
11:35 modin joined #git
11:38 maucar joined #git
11:41 ahmed_elgabri joined #git
11:41 chris_wot joined #git
11:42 Dropbox joined #git
11:42 heftig timini: git cherry-pick
11:42 heftig whoops
11:43 Dropbox left #git
11:43 Dropbox joined #git
11:44 Dropbox test
11:44 Dropbox left #git
11:45 espindola joined #git
11:45 bibble Hello.
11:45 gitinfo bibble: 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.
11:46 fwdit joined #git
11:46 bibble I'm attempting to host a private shared repo using git-shell.
11:46 bibble But git-shell allows sftp access; meaning other users can sftp remote repo, edit hooks, and reupload
11:47 bibble Meaning they can do normal shell things like wget bad file
11:47 bibble What is a secure way to share a repo with another user ?
11:48 Pulp joined #git
11:48 moritz bibble: !gitolite
11:48 gitinfo bibble: 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/
11:49 deadnull joined #git
11:49 bibble Thank you moritz. I will check it out now. I had come across it, but was attempting the most basic way to do it.
11:51 glauxosdever joined #git
11:51 fuchstronaut joined #git
11:51 |2701 joined #git
11:51 moritz bibble: in my experience, installing gitolite is far less effort than trying to come up with your own solution
11:51 glauxosdever Hello. Does anyone here have experience with gitpython?
11:52 Sceorem joined #git
11:52 gnoel joined #git
11:52 chris_wot joined #git
11:53 bibble moritz: Great. Will gitolite prevent non root users editing the remote hooks ?
11:54 Shashikant86 joined #git
11:55 ojdo joined #git
11:55 MadHatter42 joined #git
11:56 baba_ joined #git
11:57 moritz bibble: yes
11:57 bronson joined #git
11:58 bibble Awesome moritz. Installing it now to try it out. Thanks.
11:58 stantonk joined #git
11:59 flaguy48 left #git
11:59 fstd joined #git
12:00 prsn joined #git
12:02 nnyk joined #git
12:02 hetii joined #git
12:02 hetii Hi
12:03 hetii Q: If I clone my repository that is a for of some other one, I need to on each time add: git remote add upstream ... to set upstream repository
12:03 hetii is it possible to have such information inside my repository
12:04 Guest7219 left #git
12:04 hetii so when someone clone it  it will be able to have my changes and be in sync with upstream code ?
12:05 ronny joined #git
12:05 sdothum joined #git
12:05 vaxim joined #git
12:05 vaxim git is awesome
12:05 moritz if you sync your repo with upstream yourself, they don't need to
12:05 vaxim there, i said it
12:05 moritz hetii: I don't think there's a direct mechanism
12:05 ronny can any admin do something about the guest bann, its kind of missing the point in case of authentication failure
12:07 hetii moritz, hmm so what is a best way to be in sync with originate project and also commits own code, of course in case that don`t produce some conflicts
12:07 hetii till now I usually prepare my Makefile + use git submodule functionality
12:07 timini Hey Im running a cherry pick on a bunch of commits, its asking me to resolve the same conflict every time it it applies a new commit
12:07 hetii but its some sort of workaround
12:07 vaxim left #git
12:07 moritz hetii: well, merge upstream featurs into your own branches, and submit your own features to upstream as pull requests
12:07 rolha joined #git
12:08 settermjd joined #git
12:10 hetii moritz, is it possible somehow to trigger this merge automaticly when I clone repository ?
12:10 JeroenT joined #git
12:10 moritz hetii: I don't think so
12:11 ahmedelgabri joined #git
12:11 rwb joined #git
12:12 rafalcpp joined #git
12:13 caog joined #git
12:15 synthroid joined #git
12:18 jost_ joined #git
12:19 GavinMagnus joined #git
12:20 hetii moritz, is it ok if I will just do inside my repository: git pull git://remote_origin/xx.git ?
12:21 rahtgaz joined #git
12:21 hetii to be in sync ?
12:21 nopf joined #git
12:21 glauxosdever Does anyone know of good python modules that interact with git repositories?
12:21 PCatinean joined #git
12:25 tobiasvl glauxosdever: gitpython?
12:25 glauxosdever Yes, it's hard to use and not really simple...
12:25 tobiasvl oh.
12:26 tobiasvl glauxosdever: gittle seems more user friendly but I haven't used it myself
12:26 glauxosdever It does, but there is no compiler on that server where I intend to use the module.
12:27 glauxosdever I tried it, it wouldn't install.
12:27 tobiasvl compiler?
12:27 glauxosdever Yes.
12:27 glauxosdever It needs a compiler.
12:27 glauxosdever To install, not to work.
12:27 bremner isn't dulwich pure python?
12:28 bremner dunno if it's "good"
12:28 tobiasvl look at that, C extensions are optional
12:28 jimmybuffet joined #git
12:29 glauxosdever I tried to install it with pip. Do you by chance know how to disable C extensions?
12:29 dopesong joined #git
12:30 tobiasvl did you read the readme?
12:30 tobiasvl pip install dulwich --global-option="--pure"
12:30 glauxosdever I meant for gittle.
12:30 tobiasvl oh ok. I don't think you can.
12:31 glauxosdever Ok, I'll be back a bit later. Sorry.
12:31 Mateon1 joined #git
12:31 jimmybuffet joined #git
12:33 mdw joined #git
12:34 King_Hual joined #git
12:34 Macaveli joined #git
12:35 King_Hual joined #git
12:35 King_Hual joined #git
12:40 acetakwas joined #git
12:40 liquidjamm joined #git
12:42 ochorocho__ joined #git
12:42 ferr1 joined #git
12:42 irqq joined #git
12:43 baba_ joined #git
12:43 liquidjamm joined #git
12:44 mizu_no_oto joined #git
12:44 davisonio joined #git
12:45 puff joined #git
12:45 unbalancedparen joined #git
12:48 netj joined #git
12:48 |ifei5g00d joined #git
12:50 Shashikant86 joined #git
12:50 blackwind_123 joined #git
12:54 italoacasas joined #git
12:54 mingwandroid joined #git
12:54 prsn joined #git
13:00 Gsham joined #git
13:05 bitkiller joined #git
13:06 zincrokx joined #git
13:06 zincrokx joined #git
13:07 zincrokx joined #git
13:07 zincrokx joined #git
13:09 livingstn joined #git
13:09 prsn joined #git
13:10 Psychiatrist joined #git
13:12 _pix_ joined #git
13:12 MarkyC joined #git
13:12 hahuang65 joined #git
13:13 _pix joined #git
13:13 c0ded joined #git
13:14 rnsanchez_ joined #git
13:14 glauxosdever Ok, I installed dulwich, seems to work.
13:15 _pix joined #git
13:15 glauxosdever I'm wondering, though, how can I get the list of branches in a repo.
13:17 bibble moritz: Do you ever enable the "create" option in gitolite.rc ?
13:17 bibble I'm unable to get it to work
13:17 bibble ssh gito@localhost create something
13:17 zivester joined #git
13:17 bibble FATAL: repo already exists or you are not authorised to create it
13:17 Psychiatrist joined #git
13:18 cebor joined #git
13:18 jimi_ joined #git
13:19 puff Good morning.
13:19 bibble Morning puff
13:19 JeroenT joined #git
13:19 bollullera joined #git
13:19 Psych joined #git
13:19 ahmedelgabri joined #git
13:20 puff glauxosdever: http://stackoverflow.com/questions/14517​343/check-which-branches-contain-a-speci​fic-git-commit-sha-using-python-dulwich
13:21 glauxosdever Yes, obviously I have been through this question.
13:21 glauxosdever Also, it doesn't ask exactly what I want.
13:21 puff glauxosdever: Yeah, it does more, have you tried just dumbing down his example, i.e. just using this bit:  for ref in repo.refs.keys("refs/heads/")
13:22 flaguy48 joined #git
13:23 ocbtec joined #git
13:23 moritz bibble: I haven't; I create repos by commiting them to the gitolite-admin repo
13:23 moritz erm, adding them to the config file in there
13:24 bibble Cool. I have been doing the same. Just thought "create" would allow users to.
13:24 bibble But unable to get it working anyway.
13:24 FrontPsych joined #git
13:25 bibble In .rc file it's commented out initially.
13:25 dillz joined #git
13:26 Xethron joined #git
13:26 wawrek joined #git
13:28 dillz hello
13:28 gitinfo dillz: 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.
13:28 Gsham joined #git
13:28 fuchstronaut joined #git
13:28 wawrek I want to merge a branch from one repository to the other. it is not about subtree merging, because I want to merge the entire branch. what is the best approach? subtree and submodule docs assume that I want to merge subtrees from one repo to the other.
13:28 moritz wawrek: you can add the other repo as a remote to your own
13:29 moritz wawrek: or even just git merge <url-of-remote> branch
13:29 moritz no wait, pull, not merge is the right command, I think
13:29 Asiajey joined #git
13:29 * moritz usually works with adding the remotes
13:29 wawrek moritz: that sounds nice.
13:29 dillz who officially "owns" Git? who runs the project? i need to answer to an authority that is checking our project; they asked about the "vendor" of Git
13:30 lindenle joined #git
13:30 WayToDoor joined #git
13:30 wawrek moritz: I will try adding the remote , then I merge the new branch to my test
13:30 Edward_Stryfe joined #git
13:30 osse dillz: it has no vendor
13:30 Macaveli joined #git
13:31 moritz dillz: the contributors own the project. The developers can be reached through the !mailing_list
13:31 gitinfo dillz: The mailing list can be reached via git@vger.kernel.org. You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist
13:31 dillz but is there any organization i can refer to?
13:31 dillz such as, for SVN its Apache SF
13:32 osse The closest I can think of is this the Software Freedom Conservancy
13:32 dillz yes, they own the trademark AFAIK
13:32 Edward_Stryfe left #git
13:32 mmlb joined #git
13:34 osse It provides "financial and administrative assistance" to Git
13:35 osse Good enough
13:35 fracting joined #git
13:35 tlaxkit joined #git
13:36 ash_workz joined #git
13:37 BigWillie joined #git
13:37 dillz hopefully :)  thanks!
13:38 jaguarmagenta joined #git
13:39 wawrek moritz: worked perfectly - it is simple at the end. thanks
13:39 Dougie187 joined #git
13:39 leonarth joined #git
13:39 sitaram bibble: http://gitolite.com/gitolite/wild.html (for that 'create' thing)
13:40 bibble Thanks sitaram, reading now...
13:42 Raging_Hog joined #git
13:43 diegoaguilar joined #git
13:43 Lipp joined #git
13:44 Arahael joined #git
13:45 prsn joined #git
13:45 chris_wot joined #git
13:46 duderonomy joined #git
13:46 bibble Thanks moritz & sitaram. Seems great and got "create" working!
13:47 durham joined #git
13:48 bibble Used "repo CREATOR/..*" in .rc file and then "ssh git@host create user/reponame"
13:48 bibble I mean admin conf, not .rc file
13:49 Arahael joined #git
13:49 AaronMT joined #git
13:49 mmlb joined #git
13:49 irqq_ joined #git
13:51 fahadash joined #git
13:51 sitaram bibble: umm where did you get that syntax from?
13:51 clemf joined #git
13:52 sitaram bibble: scratch what I said; /me is tired and is talking nonsense.  Syntax is perfectly fine
13:52 zotherstupidguy joined #git
13:52 bibble sitaram: great :)
13:53 jast joined #git
13:53 bibble I prefer that to editing the admin conf.
13:53 glauxosdever How can I get the list of commits in a branch using dulwich?
13:55 sangy joined #git
13:56 tach___ joined #git
13:59 glauxosdever If anyone has experience with this module, it would be great if he answered.
14:00 cerrar joined #git
14:02 lindenle joined #git
14:03 davisonio joined #git
14:03 Ch4rAss joined #git
14:04 leonarth joined #git
14:05 govg joined #git
14:06 Shashikant86 joined #git
14:07 mohabaks joined #git
14:09 eletuchy joined #git
14:09 King_Hual joined #git
14:09 emPi joined #git
14:11 andyhoang1 joined #git
14:12 xcesariox joined #git
14:13 hahuang65 joined #git
14:13 yqt joined #git
14:14 fracting joined #git
14:14 csd_ joined #git
14:16 baba_ joined #git
14:16 Asiajey joined #git
14:16 Ardethian\work joined #git
14:17 marksei joined #git
14:17 knod_away joined #git
14:17 scarabx joined #git
14:17 Arahael joined #git
14:18 gmurop joined #git
14:19 chachasmooth joined #git
14:19 Kicer86 joined #git
14:20 marksei joined #git
14:20 Gitzilla joined #git
14:23 arahael1 joined #git
14:25 mostlybadfly joined #git
14:26 leonarth joined #git
14:26 ceaucari joined #git
14:26 chachasmooth joined #git
14:26 mohabaks joined #git
14:26 scarabx joined #git
14:28 gajus joined #git
14:28 gajus Git is showing file as modified
14:28 gajus but when I do `git diff ./src/date/constants.js`
14:28 gajus there is no output
14:28 gajus how else to check whats changed?
14:29 _rgn you might have staged the file
14:29 _rgn in which case you need git diff --cached ...
14:30 gajus in git status it appears as non-staged
14:31 puff gajus: Might be a whitespace difference.
14:31 puff glauxosdever: Did you try what I suggested?
14:31 TomyWork joined #git
14:32 wviana_ joined #git
14:32 eijk joined #git
14:32 glauxosdever puff: Yes.
14:33 glauxosdever Except now I want to list the commits.
14:33 knod left #git
14:34 bibble sitaram: Just wanted to check "CREATOR/..*" same as "CREATOR/.*", single dot. Normal regex ?
14:34 axl_ joined #git
14:36 King_Hual` joined #git
14:36 chachasmooth joined #git
14:37 sentriz joined #git
14:37 mmlb joined #git
14:37 knutstrut joined #git
14:37 knutstrut joined #git
14:37 WayToDoor joined #git
14:38 knutstrut joined #git
14:38 synthroid joined #git
14:39 glauxosdever Note that I don't have deep knowledge of git internals.
14:40 phaleth joined #git
14:40 emPi joined #git
14:40 wviana joined #git
14:40 Tortoise1 joined #git
14:41 jantje Hi. I have  tens of windows users that suddenly get out of memory errors when running a pull (rebase) , the error is https://paste.fedoraproject.org/368512/36687641/ ,  it looks like allocating 2MB fail, but all users have 8 to 16G memory and it definitely isnt running out. How can I debug this??
14:41 gajus left #git
14:41 cdg joined #git
14:41 tach___ joined #git
14:41 monoprotic joined #git
14:42 ochorocho__ joined #git
14:45 sitaram bibble: .* would end up matching a string ending in a slash
14:46 bibble How ?
14:46 Tortoise1 Hi guys, We push to deploy from a local machine to a remote server. The actual directory pulls from the /var/git/ folder I believe. Basically the middle repo is missing that the local pushes to and the actual directory pulls from (via hooks). The remote server was migrated some time ago and the /var/git folder has no entry for a website anymore, although others do. How to generate this? The actual files used to serve the site
14:46 bibble Confused
14:46 Gsham joined #git
14:46 bibble sitaram, how does the extra dot work ?
14:46 jceb joined #git
14:47 TomyWork joined #git
14:47 bremner are you mixing globs and regexes?
14:47 bibble I'm thinking regex...
14:47 sitaram bibble: the extra dot forces at least one character after the slash
14:47 sitaram bibble: the correct syntax would thus be "CREATOR/.+" but see http://gitolite.com/gitolite/wild.ht​ml#regex-pattern-versus-normal-repo
14:47 sitaram (for why we use "..*" instead of ".+")
14:47 bibble Ah, understand!
14:48 bibble Thanks sitaram. Will continue using CREATOR/..*
14:48 mosh joined #git
14:50 arahael1 joined #git
14:52 clemf joined #git
14:54 stantonk joined #git
14:57 stantonk joined #git
14:57 mroth_ left #git
14:57 SimonNa joined #git
14:57 mjuszczak joined #git
14:59 rkazak joined #git
15:00 bronson joined #git
15:01 chachasmooth joined #git
15:02 stantonk_ joined #git
15:02 glauxosdever Could anybody, please, tell me how I can list commits in a branch using dulwich from python?
15:03 stantonk_ joined #git
15:04 unbalancedparen joined #git
15:05 Ch4rAss joined #git
15:06 zivester joined #git
15:06 NateAGeek joined #git
15:06 medeirosthiiago joined #git
15:07 baxx joined #git
15:07 nilsi joined #git
15:07 roelmonnens joined #git
15:09 puff glauxosdever: Did you try what I suggested?
15:09 ejnahc joined #git
15:09 glauxosdever puff: Yes.
15:09 puff glauxosdever: Ah, I see your current question is different.  It should be in the dulwich docs.
15:10 puff glauxosdever: Last time I looked at dulwich was a year ago and it seemed somewhat skimpy.
15:11 synthroid joined #git
15:13 nettoweb joined #git
15:13 christophler joined #git
15:14 nnyk joined #git
15:15 hahuang65 joined #git
15:15 subhojit777 joined #git
15:15 christophler im utterly stumped. Im pushgin a github repo [master], and then pulling [master] adn the two filesets are completely different. I have checked the urls in the config, plus Ive been using the setup for 6 months? Any knowldges about this?
15:15 hobodave joined #git
15:15 christophler the repo on github is uptodate, but what Im getting on the command line is totaly different
15:16 christophler there is only one repo on that github account
15:16 han-solo joined #git
15:17 han-solo do i need to go to the github website to create an account?
15:17 unbalancedparen joined #git
15:17 leonarth joined #git
15:19 osse no
15:19 mdw joined #git
15:19 knutstrut joined #git
15:20 roelmonnens joined #git
15:22 nnyk_ joined #git
15:22 glauxosdever puff: Do you by chance remember anything?
15:22 fuchstronaut joined #git
15:25 sentriz joined #git
15:26 CJKinni joined #git
15:27 kadoban joined #git
15:28 puff glauxosdever: Sorry, now.  Check the docs?  This would seem like a pretty basic thing.
15:29 artemisart joined #git
15:29 glauxosdever It's not a basic thing there...
15:29 christophler Hmm. I deleted the local repo and cloned it again from github, got the expected current file.... hmmm indeed
15:29 moei joined #git
15:30 okdamn joined #git
15:30 okdamn Hi, do i move all the files from master branch to new-branch ?
15:30 okdamn i have no changes i just need to move all to new-branch and delete master
15:30 hashpuppy joined #git
15:30 NeverDie joined #git
15:31 narendraj9 joined #git
15:31 madewokherd joined #git
15:31 osse okdamn: rename master to new-branch
15:32 okdamn osse:  how ? :D
15:32 jfive hello
15:32 mjuszczak joined #git
15:32 jfive is the pre-commit hook run before 'git add' ?
15:32 osse okdamn: git branch -m new-branch
15:32 osse jfive: before git commit
15:33 jfive osse: okay, so let's say i wanted to compile my sass files to style.css prior to doing 'git add' is there a way to do that? or should i just do 'git commit -am <message>' instead?
15:34 okdamn jfive:  i already pushed everything to master
15:34 osse jfive: the only sensible thing to do is to not compile them at all.
15:34 okdamn i would like to rename master to gh-pages only
15:34 CJKinni joined #git
15:35 osse jfive: git commit -a simulates git add -u && git commit. my guess is the hook is run after the add part
15:35 jfive osse: yeah, i suppose thinking it through more i would need to compile as i develop to make sure everything looks correct on the site
15:35 osse that said, you can unstage the sass and stage the css there I suppose
15:35 osse jfive: compile as part of deployment
15:35 bango2 joined #git
15:35 blueyed joined #git
15:35 osse jfive: ie. like any other source code
15:36 osse you commit hello.c, not a.out, u feel me bruh
15:36 thiago joined #git
15:36 jfive osse: makes perfect sense, thanks :D
15:37 durham joined #git
15:37 leonarth joined #git
15:39 jaguarmagenta joined #git
15:39 baba_ joined #git
15:39 Gsham joined #git
15:39 LeBlaaanc joined #git
15:41 iatkotep joined #git
15:41 bibble All the best moritz & sitaram !! Thanks. I'll probably be back tomorrow to stay :)
15:42 bibble Hope everyone else has a great day.
15:42 circ-user-bUyXf joined #git
15:43 zumba_addict joined #git
15:44 jceb joined #git
15:44 zumba_addict good morning. Is it possible to customize .gitignore per branch?
15:44 zumba_addict it's because there is a file that only needs to exists in master branch
15:44 canton7 sure. it's a file that's tracked just like any other file, so can be different between branches
15:44 jbitdrop joined #git
15:45 zumba_addict do you mean, everything a developer does a git checkout -b new-feature, we will have to make sure to edit .gitignore?
15:45 frem_ joined #git
15:46 rewt what do you want to do exactly?  branch off of master but not include that file in the new branch?
15:46 nnyk_ joined #git
15:48 crose joined #git
15:48 zumba_addict i'm back rewt
15:50 zumba_addict there is a configuration (app.config) file that's being watched by a script our devops wrote. When this file gets added in a branch and gets pushed remotely, the script will see it and does things like creating Jenkins job
15:51 zumba_addict i just confirmed from the devops team that it doesn't matter which branch the file gets created
15:51 elastix joined #git
15:51 zumba_addict that means, if let's say we have a good app.config in master and then someone does a git checkout -b new-feature, that app.config will get pulled as well
15:52 _ikke_ In general, it doesn't turn out great if you add files to git which should be different between environments
15:52 rewt `git checkout -b new-feature` just creates a new branch pointer and checks it out -- no file movements involved
15:52 _ikke_ using branches to differentiate between files makes it hard to merge the branches back
15:53 zumba_addict now, new-feature contains codes that aren't working. Since it's just a branch, dev will always make a commit/push. Problem is, the devops script will see the file
15:53 rewt if they don't touch the file, it won't be marked as a newly pushed file
15:53 AaronMT_ joined #git
15:53 zumba_addict rew however, it will use the branch that got push. That's what the script will be pulling and not master
15:54 rewt that script pulls the new branch, and checks for the existence of the ifle?
15:54 _ikke_ why does that branch need a different version of the file?
15:54 rewt file*
15:54 zumba_addict i can do another test to make sure this is the behaviour i'm seeing
15:54 zumba_addict _ikke_: it doesn't
15:54 _ikke_ but?
15:54 Bodger joined #git
15:55 Bodger left #git
15:55 _ikke_ Make your build tools smarter to only look in master if that's the goal..
15:55 zumba_addict but devops just told me that when this file also exists in the new branch(though it wasn't touch), the existing job will get overwritten
15:55 nettoweb joined #git
15:55 zumba_addict yeah, that's what we want
15:55 _ikke_ Not a git problem..
15:55 zumba_addict it is not but i was hoping if we can tailor .gitignore
15:55 _ikke_ no
15:55 zumba_addict got it
15:55 zumba_addict thanks
15:56 notebox joined #git
15:56 _ikke_ Don't mean to sound harsh
15:56 Eugene !ignore_tracked
15:56 gitinfo Git only applies ignore patterns to untracked files. You can't use ignore patterns to ignore changes to files that are already tracked by git. To remove files only from git, but keeping them on disk, use git rm --cached <file>. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.
15:57 grawity joined #git
15:57 aard joined #git
15:57 moon23 joined #git
15:57 adamretter joined #git
15:57 rewt you'd have to make the new branch and delete app.config... i'm not sure how .gitignore would help here
15:58 _ikke_ merging that branch would remove app.config from master
15:58 rewt and then on merge make sure it doesn't get deleted in master
15:58 _ikke_ If you only want to build master, the build tool should only look inside master
15:58 _ikke_ simple, nothing to do with git :-)
15:59 adamretter If I have a branch B (which was forked from branch A), and then use filter-branch to remove a file from a branch B (which was introduced after the fork from branch A), will it cause any problems to anyone who isn't using branch B, i.e. will it effect branch A (or any other branches of Branch A) in any way?
15:59 rewt agreed ... set up a git hook to trigger the script on push to master
15:59 _ikke_ Depending on content being different between branches never works
16:00 rewt adamretter, as long as filter-branch only rewrites commits after the branch of B, other branches shouldn't be affected
16:01 mda1 joined #git
16:01 _ikke_ Agreed
16:01 todd_dsm joined #git
16:01 rewt adamretter, after using filter-branch, it should tell you what was rewritten -- if only branch B is listed, you're ok
16:02 adamretter rewt: cool that was what I figured
16:03 rewt and the good thing about git is that you can clone the repo and try things without affecting anything at all
16:03 _ikke_ And even filter-branch backs up all refs
16:03 rewt something like:  cd /tmp ; git clone /path/to/repo ; cd repo
16:03 rewt yeah, but if you don't like how it turns out, you have to undo it
16:04 rewt with a clone you can just rm -rf it
16:04 jantje Hi. I have  tens of windows users that suddenly get out of memory errors when running a pull (rebase) , the error is https://paste.fedoraproject.org/368512/36687641/ ,  it looks like allocating 2MB fail, but all users have 8 to 16G memory and it definitely isnt running out. How can I debug this??
16:04 cyberbit joined #git
16:05 _ikke_ jantje: First step might be, what changed?
16:05 rewt it allocates in chunks as needed, and it tried to go over the available memory
16:05 ForgottenPlayer joined #git
16:05 mwhooker joined #git
16:05 Eugene jantje - that error is pretty straightforward; malloc() fails.
16:05 _ikke_ jantje: new version of git, large files pushed?
16:05 Eugene Is your repo unreasonably large?
16:05 adamretter rewt: so I should pass the common ancestor commit of Branch A and Branch B to git filter-branch to start after the fork then?
16:06 _ikke_ adamretter: iirc, git only rewrites commits if necessary
16:06 _ikke_ So it should not even matter
16:06 Eugene Also, to eliminate noise, use `git fetch` separately from merge or rebase
16:08 jantje _ikke_: git for windows, 2.5.2 (64bit), yes, we recently had some merge commits pushed to the master branch. It's very random, not all users have the issue, but we've seen it frequently enough to know there is an issue somewhere.
16:08 adamretter _ikke_: okay thanks
16:08 Lipp joined #git
16:08 jantje Eugene: yes, it is very large!
16:08 Eugene !binary
16:08 gitinfo Storing binary files in git causes repo balloon, because they do not compress/diff well. In other words, each time you change a file the repo will grow by the size of the file. See !annex for some solutions
16:09 Eugene One of the weak points in git is handling repos that are "large", which usually means a significant fraction of RAM
16:09 jantje Eugene: but not with binary files. It has over 12 years of history, total 750k commits
16:09 Eugene Ah.
16:09 Eugene Is it kernel.git by chance?
16:09 jantje No.
16:09 _ikke_ jantje: you can try experimenting with the packing config settings
16:10 _ikke_ !large_repo
16:10 gitinfo [!large_repos] Git can be slow in the face of large repositories.  There are git-config options which can help.  pack.threads=1; pack.deltaCacheSize=1; pack.windowMemory=512m; core.packedGitWindowSize=16m; core.packedGitLimit=128m. Other likely ones exist.
16:11 _ikke_ packing and deltafying can take up a lot of memory
16:12 mekhami joined #git
16:12 jantje the end result is that the repository gets in a bad state. It lost its branch info
16:13 mekhami so the dev before my committed a ton of source libraries to our main repository. now the repository is unfathomably large. This goes back for so long I can't just revert the commit where he put all this stuff in. What are my options? Do I just have to remake the repository?
16:14 blueyed joined #git
16:14 _ikke_ branch info?
16:14 jantje _ikke_: I suppose there are different optimal settings for windows and linux, I have not seen the issue on Linux.  The problem did occur when rebasing, is it repacking then?
16:14 _ikke_ more unpacking
16:15 moritz mekhami: you could remove the libraries from the history, but that would entail rewriting the history of the complete repo
16:15 hahuang65 joined #git
16:15 jantje _ikke_: yes, for example, it doesn't know its on master. 'git status'says  Not currently on any branch.
16:17 leonarth joined #git
16:17 _ikke_ Right, but that's just because rebase aborted
16:17 _ikke_ rebase detaches first
16:17 _ikke_ so checkout out a branch fixes it
16:18 _ikke_ g2g
16:18 BadApe joined #git
16:18 BadApe hello, where on windows does git install itself to?
16:19 synthroid joined #git
16:20 mekhami moritz ▸ is it smarter to just recreate the repository from a stable master?
16:21 moritz mekhami: you mean from a snapshot? Then you won't have history at all
16:21 mekhami moritz ▸ i understand but it's not sustainable in its current format
16:22 moritz mekhami: that's why i'd propose to rewrite the history and remove the libraries from it
16:22 moritz something like git filter-branch or !bfg can help you with that
16:22 gitinfo A tool designed to remove large files, or passwords from history: https://rtyley.github.io/bfg-repo-cleaner/ (!rewrite applies)
16:22 mekhami okay thank you, i'll look into that
16:22 mekhami !rewrite
16:22 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
16:23 mekhami hm.
16:23 moritz better to have a rewritten history than no history.
16:23 mekhami i'm currently the only dev but we are bringing one person on
16:23 mekhami so it might be less painful to rewrite in that case
16:25 imachuchu joined #git
16:25 MistahKurtz joined #git
16:31 stantonk joined #git
16:31 thiago joined #git
16:31 isxek joined #git
16:32 medeirosthiiago joined #git
16:32 bollullera left #git
16:33 kpease joined #git
16:37 swahili joined #git
16:37 swahili hi!
16:37 swahili Q: A commit was merged to a Branch and broken some stuff; Is there a way to remove it from history ?
16:38 swahili Who committed didn't work on a separate branch unfortunately
16:38 nate_c joined #git
16:38 osse that commit sounds public. I suggest you use git-revert to get the job done
16:39 osse if that's not to your taste, and the history after that commit is simple, git rebase -i commit~1. delete it from the editor, save and close
16:39 osse if it is complex, git filter-branch
16:40 fracting joined #git
16:40 osse But to clarify: use git revert
16:42 johnwhitton joined #git
16:42 CharlieWONG joined #git
16:42 netj joined #git
16:42 InfoTest joined #git
16:43 ochorocho__ joined #git
16:43 Mobutils joined #git
16:44 WayToDoor joined #git
16:45 khmarbaise joined #git
16:53 ngladitz joined #git
16:53 ngladitz joined #git
16:54 mmlb joined #git
16:54 alex1a joined #git
16:55 akushner joined #git
16:56 sbasso joined #git
16:59 jstimm joined #git
17:02 akushner joined #git
17:03 bronson joined #git
17:03 hahuang65 joined #git
17:04 stantonk joined #git
17:04 Noldorin joined #git
17:04 stantonk joined #git
17:04 Kehet_ left #git
17:05 vacho joined #git
17:05 vacho hello everyone..
17:06 vacho I am new to git..and trying to learn. Is there a "recommended" way of pushing new features to git. I was thinking of a workflow that for each JIRA task that is completed, one commit is done to the git branch. But how should I go about this?
17:07 ojacobson !workflow
17:07 gitinfo Finding the right workflow for you is critical for the success of any SCM project.  Git is very flexible with respect to workflow.  See http://sethrobertson.github.c​om/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.
17:08 stanlord joined #git
17:09 brandonw joined #git
17:10 nwkj86 joined #git
17:13 srathel joined #git
17:14 srathel left #git
17:16 Ahmed90 joined #git
17:16 wiky joined #git
17:17 baba_ joined #git
17:18 fuchstronaut joined #git
17:18 crose joined #git
17:18 TomyLobo joined #git
17:18 GavinMagnus joined #git
17:24 victorhck joined #git
17:25 mdw joined #git
17:27 pantsman- joined #git
17:27 xbob joined #git
17:28 settermjd joined #git
17:30 bluezone joined #git
17:31 shinnya joined #git
17:31 Gsham joined #git
17:31 tematibr joined #git
17:32 dunpeal joined #git
17:33 dunpeal Hi. I have a feature branch that was branched off master. I need to review all code changes it introduces. What's the best way to do that?
17:34 Sasazuka joined #git
17:34 preaction dunpeal: git log -u --reverse master..mybranch
17:36 dunpeal preaction: can I see the changes consolidated?  for example, if the committer added a line `x = 1`, then later changed it to `x = 2`, I'd rather only see the single `x = 2` line.
17:36 SporkWitch joined #git
17:36 preaction then you just want git diff
17:36 ojacobson git diff master...my-branch (count the dots)
17:37 preaction but in a review, i'd review each commit atomically. that's what your history is going to look like
17:37 ojacobson git log master...my-branch is also useful
17:37 Dyonisos joined #git
17:37 rominronin joined #git
17:37 csd_ joined #git
17:37 jimmybuffet joined #git
17:37 eletuchy joined #git
17:38 PryMar56 joined #git
17:39 ferr1 joined #git
17:39 jaguarmagenta joined #git
17:42 Gianormagantrous joined #git
17:42 dunpeal Thanks. Is there a better renderer for the diff than the unified-diff output on the console?
17:43 preaction git difftool?
17:43 dunpeal I'm on OS X (and kind of a Mac newb). What would be the best, most readable way to present this diff?
17:43 preaction as a series of commits
17:44 preaction everyone's going to find their own workflow. i use what i said first, the log of commits with the diff inside. there are a bunch of tools for viewing diff output, which can be used in git difftool
17:44 preaction perhaps one of these? https://www.git-tower.com/blog/diff-tools-mac/
17:44 dunpeal what does 'git diff master...my-branch' actually show?
17:45 ojacobson kaleidoscope is nice :)
17:45 thiago_ joined #git
17:45 preaction or these? http://stackoverflow.com/questions/214​86481/is-it-possible-to-view-git-diffs​-using-a-gui-side-by-side-tool-on-mac
17:45 ojacobson the diff(1) between the last common ancestor of master and my-branch, and my-branch
17:46 eijk joined #git
17:47 preaction in case master has moved on since my-branch was branched
17:47 wcpan joined #git
17:47 dunpeal OK, so effectively all the changes introduced by my-branch since the point it diverged off master.
17:47 ahmed_elgabri joined #git
17:48 stantonk_ joined #git
17:48 stantonk_ joined #git
17:48 preaction yes. the ".." two dots requires that you rebase, which is part of my workflow
17:49 dunpeal what would the two dots mean as opposed to 3?
17:50 preaction what master is right now
17:51 Gloomy joined #git
17:52 InfoTest1 joined #git
17:52 dunpeal ah, OK.
17:52 dunpeal I think I'll just use good-old vimdiff. thanks!
17:52 ojacobson !dots
17:52 gitinfo A..B = stuff that happened between A and B (if A and B are related; otherwise refer to "man gitrevisions"), A...B = (a) in history viewers: stuff that isn't in both A and B yet; (b) in "git diff": stuff that happened in B since the two diverged; (c) in "git checkout": the merge base of A and B. "master.." is the same as "master..HEAD" and "..master" is the same as "HEAD..master", and so forth.
17:54 GavinMagnus joined #git
17:55 davisonio joined #git
17:56 elastix joined #git
17:56 raijin joined #git
17:58 dunpeal ojacobson: so ... has two separate meanings?
17:58 dunpeal b and c are exactly the same, but a looks different.
17:59 dunpeal b and c both mean "the merge-base of A and B"
17:59 Domhack joined #git
17:59 ojacobson rev-list parses x...y as the symmetric difference of commit sets x and y
17:59 ojacobson diff and checkout parse it as something to do with the merge-base (and, often, the right-hand commit)
17:59 rahtgaz joined #git
17:59 ojacobson diff, log, and checkout
18:00 ojacobson the two senses are related, but not the same, yes
18:03 dunpeal ojacobson: well, if my HEAD is B, and I do `git diff A...`, the (A...) part is interpreted exactly to mean the merge-base of master and B
18:04 dunpeal it's equivalent to `git diff B merge-base(A, B)`
18:04 dunpeal and since B is head: git diff merge-base(A, B)
18:05 jsolano joined #git
18:07 CubiK joined #git
18:07 bjv joined #git
18:07 Gsham joined #git
18:08 bjv 64bit git v2.1.4, here - is there an expected memory envelope when doing 'git push' to an SSH remote?
18:09 digidog joined #git
18:09 bjv "Writing objects: " starts working, until git uses all my available memory+swap (~8gig) and OOM killed by kernel
18:10 josuebrunel joined #git
18:10 bjv seems... unusual? surely git is streaming the local copy to remote?
18:10 romankl joined #git
18:11 cbreak bjv: no
18:11 cbreak bjv: git creates a deltafied transfer pack
18:12 arahael1 joined #git
18:12 cbreak bjv: but it should never use this much memory unless you did something horribly stupid, like commit big binary files
18:12 bjv cbreak: yes, this repo has several large binary files and i've added some more
18:13 iota_pi joined #git
18:13 cbreak bjv: well then, get more ram
18:15 jsolano Hi, I have a complicated setup, can anybody help? I have a project mixing 2 repos, from repo A I pull snapshots regularly  (I have only read permissions), while in repo B I keep my changes, but I am not allowed to push code from repo A into repo B. Whould would be the best way of using these 2 repos and work efficiently?
18:15 christophler hi. Im trying to commit, after deleting some directories and git says "cannot switch to [directory]: No such file or directory"
18:15 christophler (searching for this on google just gives loads of stuff about branches)
18:15 Eugene christophler - !transcript including `git status`
18:15 gitinfo christophler: 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
18:16 cbreak bjv: I think git can be expected to use two or three times as much memory as the biggest file
18:16 cbreak maybe more
18:16 bjv cbreak: okay so "Next, the client sends a packfile of all the objects the server doesn’t have yet." https://git-scm.com/book/en/v2/G​it-Internals-Transfer-Protocols
18:16 cbreak you are before that.
18:16 faux___ joined #git
18:17 christophler thanks : https://gist.github.com/christopherr​eay/15c15f7f763457129957ba5eecacfba4
18:17 kappy joined #git
18:17 christophler Eugene: ^-^ thanks
18:17 troulouliou_div2 joined #git
18:18 Eugene christophler - so you want to remove the deleted directories from git? Try `git add -u`
18:19 bjv cbreak: really, it said "Writing objects:
18:19 maxis joined #git
18:19 fuchstronaut joined #git
18:19 cbreak bjv: yeah... into the pack file
18:19 christophler same exact output
18:19 christophler two errors
18:20 Eugene How about `git rm -r javscript/` ?
18:20 attente joined #git
18:20 bjv cbreak: oh wow, the whole push packfile is constructed in memory?
18:20 Eugene Is this a !submodule_rm situation?
18:20 gitinfo [!submodules_rm] To delete a submodule, do this in the superproject: Edit/delete .gitmodules to remove the submodule, and `git add .gitmodules`. Then `rm -rf submodulepath; git rm -f --cached submodulepath; git commit -am "Removed submodules!"`  Inspect .git/config for "submodule" entries to remove.  Inspect .git/modules for caches to remove.  Possible alternatives to submodules: "!gitslave" or "!subtree"
18:21 cbreak bjv: I don't think so
18:21 sbasso joined #git
18:21 cbreak bjv: but computing the delta requires the files to be in memory
18:21 christophler Eugene: ok, lots of git rming taking place, i will fiddle
18:21 a3Dman_ joined #git
18:21 cbreak I haven't written the protocol, just read about it :)
18:21 christophler oh wow, hang on ill update the gist
18:22 kutsuya joined #git
18:23 bjv cbreak: well, largest version controlled file is 3.2G but largest packfile is 9.3G (which i'm guessing contains my most recent local commit: the 3.2G file plus some additional model archives)
18:23 khmarbaise joined #git
18:23 khmarbaise joined #git
18:23 cbreak bjv: you should really consider wether git is the right tool for your problem
18:24 bjv guess i'll get myself 19G of virtual memory & see how it goes
18:24 marc_v92 joined #git
18:24 cbreak your problems will only get bigger
18:24 cbreak (because you will not be able to delete any of the files without rewriting history)
18:24 raijin joined #git
18:24 christophler Eugene: I deleted did what you asked, seems to be something to do witha submodule? Ive updated the gist
18:25 fscala joined #git
18:25 DWestCoast joined #git
18:25 Eugene Try the procedure to delete the submodule, see bot above
18:26 avar joined #git
18:26 avar joined #git
18:26 suy joined #git
18:26 zrts joined #git
18:26 APic joined #git
18:26 txdv joined #git
18:26 tazle_ joined #git
18:26 IndigoTiger joined #git
18:26 ffledgling joined #git
18:26 MaddHatter joined #git
18:26 Orbitrix joined #git
18:26 Kaisyu joined #git
18:26 marlun joined #git
18:26 Puffball joined #git
18:26 SirFunk joined #git
18:26 jalnt joined #git
18:26 go1111111 joined #git
18:26 kontoo joined #git
18:26 leprechau joined #git
18:26 MikeChernev joined #git
18:26 kontoo joined #git
18:26 timthowtdi joined #git
18:26 HarryHallman joined #git
18:26 preaction joined #git
18:26 sarbs joined #git
18:26 BombStrike joined #git
18:26 Cata|Work joined #git
18:27 Vaelatern joined #git
18:27 sulky joined #git
18:27 DolphinDream joined #git
18:27 nickthetait joined #git
18:27 r00twrh joined #git
18:27 christophler Eugene: Ive movbed the submodule path to the correct path, and I cannot delete anyting from /javascript as there is nothing left to delete, but the error is the same ;)
18:27 myme joined #git
18:27 PaulCapestany joined #git
18:27 bgamari joined #git
18:27 Peuc joined #git
18:27 jas4711 joined #git
18:27 toogley can i search through the git commit history, which file versions don't have a specific line of text?
18:27 anth0ny joined #git
18:27 MeltedLux joined #git
18:27 snowinferno joined #git
18:27 christophler Eugene: i have updated hte gist
18:27 rom1504 joined #git
18:28 tomboy65 joined #git
18:28 bodie_ joined #git
18:28 suffice joined #git
18:28 bcc joined #git
18:28 cbreak bjv: have you looked at git annex?
18:29 bjv cbreak: I have
18:29 polyzen joined #git
18:29 Sleepy63 joined #git
18:29 christophler Eugene: and I updated it again. .gitsubmodules is empty, the submodules are in the config as to be shared
18:29 Eugene Ya got me, sorry
18:29 toogley bjv: i haven't read your question, but maybe git lfs is sth for you?
18:30 synthroid joined #git
18:30 christophler Eugene: thanks
18:30 blackwell joined #git
18:30 toogley bjv: that deals well with big files, i manage my videos with that
18:30 christophler Eugene: can I just blow a new head over the top of the last one?
18:30 bjv i believe at some point the operator of the GitLab system that hosts these repos will be upgrading to something that bundles a git lfs/git annex capability
18:30 warthog9 joined #git
18:31 christophler some kind of "force" Eugene
18:31 Dro joined #git
18:32 christophler Eugene: I found it. Its in .git/modules/javascript... there is something
18:32 Lipp joined #git
18:33 WizJin joined #git
18:35 jstimm joined #git
18:35 christophler Eugene: .. hmm changed the error message... btu still referencing the module somehow.. cant find it writtne anywhere in any config files.. wll try greping for the name
18:35 parliament joined #git
18:38 rominronin joined #git
18:39 christophler Eugene: so the word "color" does not appear anywhere in the .git  hierarchy but git returns with ".git/modules/javascript/colorPicker" is not a module. Fatal
18:39 christophler that diretory does not exist
18:40 cbreak christophler: git ls-tree -R
18:40 cbreak | grep commit
18:42 christophler cbreak: https://gist.github.com/christopherr​eay/2474163ec707496a49c9a1cb27b35954
18:42 diogenese joined #git
18:42 cbreak christophler: it is case sensitive
18:43 cbreak christophler: show HEAD of course
18:43 christophler both -R or -r return the same output. the --help says -r
18:43 cbreak christophler: you should see a commit entry for every submodule in the tree
18:44 ochorocho__ joined #git
18:44 christophler the repo is now broken I would say. Have a look I updated the git
18:44 christophler *gist
18:44 christophler cbreak: ^-^
18:45 cbreak ls-tree of HEAD, not checkout
18:45 leeN joined #git
18:45 netcarver joined #git
18:45 christophler that works  :)
18:45 UTAN_dev joined #git
18:45 cbreak got some commit entries?
18:45 cbreak each of them should have a corresponding entry in .gitmodules
18:46 LeBlaaanc joined #git
18:46 christophler cbreak: ok, yes, i updated the gist
18:46 christophler loko prmising.
18:46 christophler cbreak: .gitmodules is empty
18:46 cbreak then fix that
18:47 nilsi joined #git
18:47 cbreak [submodule "path"], then on individual lines, path and url entries
18:47 christophler cbreak: I added my .git/config to the gist
18:48 christophler they are there
18:48 cbreak the .gitmodules is what is needed for everyone
18:48 cbreak .git/config is just for you
18:48 cbreak add entries for all four submodules to .gitmodules
18:49 christophler cbreak: ok I added the content to gitmodules, and updated the gist
18:49 christophler ther should oly be two modules,
18:49 cbreak there are four.
18:49 christophler well..
18:49 unbalancedparen joined #git
18:49 cbreak git rm --cached path/to/other
18:50 cbreak that should delete it
18:50 cbreak christophler: also, you forgot the path entries for the submodule
18:54 ril joined #git
18:54 christophler ok. cbreak Ive done my beset. Ive editted the .gitmodules, and tried to somehow delete whatever reference exists to the old path for the old position of the two modules
18:54 christophler ive updated the gist
18:55 satbyy joined #git
18:57 christophler im going to blow the repo away and rebuild it
18:57 cbreak christophler: so, run git submodule --init
18:57 cbreak sorry, git submodule update --init
18:57 cbreak you might have to delete the modules dir
18:58 Assimilater joined #git
18:59 christophler ah yes, thats all works. thank yuo :)
18:59 christophler very very much
19:03 Lipp joined #git
19:04 Shakeel_ joined #git
19:05 bronson joined #git
19:05 dimi1947 joined #git
19:07 Mateon1 joined #git
19:08 Assimilater So I'm locally forking the stable branch of a public repository on github to local branch "github/stable" and then branched to "github/dev" where I modified some visual studio project files to compile on our enrionment and have another branching procedure into our main code development branches which significantly modify the functionality of the github branch. I'm wondering if there's a way I can merge back from "github/dev"
19:09 zotherstupidguy joined #git
19:09 jsolano Hi, I am trying to work mixing locally 2 repositories A and B, but any file updated being pushed goes only to repository B, so that both repos contain codebases as split as possible. Any ideas how to do that? Do you recommend google repo?
19:11 tristanp joined #git
19:12 bronson joined #git
19:13 arahael1 joined #git
19:15 diegoaguilar joined #git
19:15 bjv jsolano: do Repository A and B contain the same files, and you are forking A (and maintaining a changed version in B) or are the contents of A and B non-overlapping?
19:15 multi_io joined #git
19:16 khmarbaise joined #git
19:16 Assimilater maybe it would just be better to branch off github/stable when doing bugfixes and modify local environment to match environment used by remote repository (this should not happen often) rather than going through github/dev back to github/stable
19:18 arooni joined #git
19:18 Assimilater github doesn't seem to handle merging renamed files as well as I seem to recall (might be pulling memories from my SVN days)
19:18 Assimilater git* not github
19:18 circ-user-bUyXf joined #git
19:18 Assimilater or maybe hg....
19:19 arahael2 joined #git
19:20 mwhooker joined #git
19:20 lenswipe joined #git
19:20 lenswipe hey guys
19:20 lenswipe can someone tell me how I completely remove all trace of a file from a git repo?
19:20 pantsman- left #git
19:20 lenswipe This repo has a framework user manual committed into it that serves absolutely no purpose whatsoever
19:21 fuchstronaut joined #git
19:22 zalatovo joined #git
19:22 bremner ,rewrite
19:22 bremner oops.
19:22 bremner !rewrite
19:22 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
19:22 lenswipe yes im aware
19:22 lenswipe I'm aware of the risks etc.
19:23 bremner so then man git-filter-branch
19:23 gitinfo the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html
19:23 lenswipe putting that aside
19:23 lenswipe okay
19:23 bremner see the examples
19:23 lenswipe thanks
19:23 asgasgasgsdhdfh joined #git
19:23 Gsham joined #git
19:23 asgasgasgsdhdfh hello
19:23 gitinfo asgasgasgsdhdfh: 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.
19:24 asgasgasgsdhdfh I m trying to make my first commit but I am getting this message: fatal: unable to auto-detect email address (got 'ariel@EscritorioPC.(none)')   :(
19:24 fracting joined #git
19:24 mwhooker joined #git
19:24 asgasgasgsdhdfh anyone can help me with this ?
19:25 avar git config --global user.email your@email.address
19:28 ehynds joined #git
19:28 asgasgasgsdhdfh avar, i tried that already :(
19:30 texasmynsted joined #git
19:31 texasmynsted why do I see ?? for files when I do git status on a file that has been ignored in my .git/info/excludes file?
19:31 ZGD joined #git
19:32 texasmynsted It should be ignored right?
19:32 texasmynsted even if other files in that directory are stored in git?
19:33 ojacobson show us the excludes file and the status output
19:33 nowhere_man joined #git
19:34 bjv cbreak: with additional memory added to the local client, git push now "Writing objects:" until 2.95GiB then stops & hangs there - cpu usage drops to 0%
19:35 Xethron joined #git
19:35 cbreak hmm... weird
19:35 bjv setting .ssh/config LogLevel=DEBUG2, shows that pack-receive is called.. then client starts its work compressing & writing
19:36 * texasmynsted facepalm
19:36 texasmynsted Was a type-o
19:36 bjv all the while SSH channel is 'rcvd adjusting' up and down, generally staying at one value (131072).. however just as git client gets to the magic 2.95GiB total (about 38% packed) the SSH channel falls totally silent
19:36 texasmynsted Thank you
19:37 _ikke_ jantje: Did you already find something?
19:38 sigma__ joined #git
19:38 ojacobson texasmynsted: no prob
19:38 bjv in each case (run twice, so far with debug on), the final 'debug2' statement printed is an "rcvd adjust" down, to something under 131072 (e.g.: 8192 ) not certain what that means
19:38 Siegfried joined #git
19:38 Siegfried Hi all
19:38 gitinfo Siegfried: 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.
19:38 sigma__ I seemed to have ended up working in a "branch" called (detached from 0119137) and I've committed work that I really don't want to lose.
19:39 sigma__ I'm working on understanding how this happened, but can someone advise me on making sure I don't lose my work?
19:39 Siegfried Can someone explain git push -u origin [branch] vs git push origin [branch] ?
19:39 sigma__ Can I just checkout the branch I thought I was on...?
19:39 akushner joined #git
19:39 rominronin joined #git
19:39 bjv cbreak: wait, wait - i was patient this time (an distracted typing into this channel)!    The client progress has resumed
19:39 GavinMagnus joined #git
19:39 Siegfried sigma__: You can just navigate to where you want the commit
19:40 sigma__ Siegfried: I've already committed the changes, just to be clear.
19:40 Siegfried use git reflog to find the commit hash of where you commited on the detatched head
19:40 rewt sigma__, if you've committed already, you may be able to rebase onto the branch you want to be on
19:40 Siegfried and then use git cherry-pick to copy the commit
19:40 sigma__ Oh.
19:40 Balliad joined #git
19:40 Siegfried ez pz
19:40 sigma__ I've never done this before. I want to get it right.
19:40 sigma__ Heh.
19:40 Siegfried Now help with my question pls :D
19:40 sigma__ It is when you understand it.
19:40 Siegfried Fair
19:41 Siegfried Okay I'll walk you through it
19:41 sigma__ Please.
19:41 Siegfried first you can go ahead and git checkout to where you want this commit to actually be
19:41 texasmynsted left #git
19:41 satifant joined #git
19:41 sigma__ Warning: you are leaving 4 commits behind
19:41 Siegfried hmmm
19:42 rewt if there's no ref pointing to it, git 'forgets' them
19:42 Siegfried try a direct rebase then?
19:42 Siegfried Oh I see
19:42 Siegfried Okay
19:42 ams__ joined #git
19:42 sigma__ rewt: That's what I was afraid of.
19:42 Siegfried Then you can just do git rebase to directly pop the 4 commits onto another branch
19:42 rewt i'm guessing rebase should work
19:42 sigma__ So while detached, git rebase [branch-I-thought-I-was-on] ?
19:42 rewt yes
19:42 sigma__ One sec.
19:43 sigma__ Wait!
19:43 rewt gitk can be useful to see exactly where everything is
19:43 sigma__ I did the checkout as previously suggested and it did switch branches.
19:43 sigma__ But I don't have my changes.
19:43 Siegfried from what I recall
19:43 nettoweb joined #git
19:43 rewt yeah, they're in the other branch
19:43 Siegfried git won't delete everything right away
19:43 Siegfried it does so in waves every so often
19:43 rewt do you have the hash of the last commit?
19:43 Siegfried I forget when/how
19:43 sigma__ But.. where is the other branch? The detached one?
19:44 sigma__ rewt: Oh boy. Probably in my shell history. :)
19:44 Siegfried Nope
19:44 Siegfried type git reflog
19:44 Siegfried It shows you a history of the commits where your HEAD has been pointing
19:44 sigma__ I see it.
19:44 sigma__ 8dbf536 HEAD@{0}: checkout: moving from f1980ad8fcc84a9366b72a32081d2130883230d8 to v1.1.0
19:44 sigma__ Or rather, the one just before that is the commit.
19:45 rewt so you can `git checkout f1980` to go back there
19:45 sigma__ f1980ad HEAD@{1}: commit: [my message here]
19:45 rewt that's the same commit
19:45 sigma__ Either way, there's my hash.
19:45 sigma__ Yep.
19:45 sigma__ So check that out and rebase v1.1.0?
19:45 rewt yes
19:45 sigma__ One sec.
19:46 sigma__ Should that have been done the other way?
19:46 sigma__ checkout v1.1.0 and rebase f1980ad?
19:46 rewt you want f1980 to be at the very end, right?
19:46 Kallis joined #git
19:46 sigma__ Yeah.
19:46 sigma__ I suppose I do.
19:46 rewt so checkout f1980 and rebase v1.1.0
19:46 sigma__ Conflicts. I can work through them, though.
19:47 sigma__ This has been a learning experience. Yay!
19:47 sigma__ :)
19:47 rewt resolve conflicts; git add changed files; git rebase --continue
19:47 * sigma__ increases git cred by +1
19:47 sigma__ Heh, I'm very good at resolving conflicts, actually. xD
19:47 EY joined #git
19:47 sigma__ *That's* a skill I use often.
19:47 Siegfried lol
19:47 sigma__ Thanks for the help!
19:48 Siegfried Glad we could help :)
19:48 LeMike joined #git
19:48 Siegfried ANyone know the difference between git push origin and git push -u origin?
19:48 Siegfried docs are not clear
19:48 Siegfried cause im dumb
19:49 sigma__ Siegfried: Since you helped me, I'll offer my interpretation from the man page. It sounds like, after pushing, if -u was used, a remote tracking branch is automatically created locally for that remote branch, so ease the use of git pull in the future, and probably for other reasons.
19:50 Siegfried sigma__: Hmm okay, that makes sense.  Thank you :)
19:50 sigma__ e.g. git push -u origin dev creates local tracking branch origin/dev, which will be used if you do a git pull or git checkout dev without having a local branch 'dev' already created.
19:50 sigma__ Sure!
19:50 Siegfried Yeah that makes sense
19:50 sigma__ Try it out to confirm. :)
19:50 Siegfried sigma__: Way ahead of you :D
19:51 sigma__ :D
19:51 Siegfried rewt: Can I ask you a quick question?
19:51 rewt you just did
19:51 Siegfried :(
19:51 rewt :)
19:51 Siegfried Let's say I check out a master branch that's tracking an upstream master branch
19:51 Siegfried I branch off of master, work on it, and then rebase onto master
19:52 Siegfried since master is tracking the remote, it won't update after the rebase right?
19:52 jsolano bjv: the contents of A and B are mostly non-overlapping, but if there is some minor overlap, them modified files would go into repo B, so that repo A is untouched
19:52 rewt yeah, sounds about right
19:52 adamretter left #git
19:52 EduardoMartins joined #git
19:53 Siegfried okay cool
19:53 Siegfried thanks
19:53 rewt but maybe a ff merge would be better in that case to bring master up to date
19:54 Siegfried gotcha
19:54 newbsie joined #git
19:55 newbsie Is it normal that you create a new branch, then commit, then merge to master, when you push to remote, the new branch doesn't appear?
19:55 _ikke_ depends on what you push..
19:55 _ikke_ pushing just master does not push the branch
19:55 _ikke_ ie, the reference
19:56 _ikke_ The commits are obviously pushed
19:56 sigma__ Siegfried: My rebase was successful, but I can't seem to get this detached commit ref onto my real branch...
19:57 LeBlaaanc are tags unique to a branch or no?
19:57 Siegfried newbsie: Yeah it's normal depending on what you're pushing to
19:57 Siegfried sigma__: What do you mean by get the ref onto your branch?
19:58 sigma__ I've rebased using the branch I thought I was working on.
19:58 newbsie Siegfried: So what happened to the branch? It got deleted along with it's history?
19:58 Siegfried the detatched branch will get deleted eventually
19:58 sigma__ I've got commit hash 176ca01 checked out with my changes committed, rebase complete, conflicts resolved.
19:58 Siegfried git has some kind of cleanup mechanism
19:58 sigma__ Now I want to get my work from detached commit 176ca01 over to branch v1.1.0
19:59 sigma__ I checked out v1.1.0 and tried to git merge 176ca01 but that didn't work.
19:59 newbsie Siegfried: I thought the whole point of branches is to keep it in history, but I see people deleting it. Is it more a tool to work concurrently as opposed to a history keeping?
19:59 DolphinDream joined #git
19:59 rewt sigma__, is v1.1.0 a tag or a branch?
19:59 sigma__ rewt: It's a branch.
20:00 raijin joined #git
20:00 mmlb joined #git
20:00 rewt the new commit is 176ca01, and it's a descendant of v1.1.0?
20:00 irqq joined #git
20:00 sigma__ rewt: It is now, as I've rebased, but also, I branched *from* it originally, or so I thought.
20:00 sigma__ 99% sure, yes. :)
20:01 sigma__ Confirmed.
20:01 sigma__ I see it in git log --oneline --decorate.
20:01 sigma__ I'm two commits ahead of it.
20:01 bjv joined #git
20:01 m0viefreak joined #git
20:01 sigma__ Can I create a branch out of my current HEAD position (terminology.. ugh) and then merge that branch into v1.1.0?
20:02 pckz joined #git
20:02 sigma__ As in, git checkout -b newbranch right from where I am?
20:02 sigma__ From a detached state?
20:02 rewt try this:  git checkout v1.1.0 ; git merge --ff 176ca01
20:02 sigma__ One sec.
20:02 sigma__ Wicked.
20:02 sigma__ Perfect.
20:02 mdw joined #git
20:03 rewt and yes, you should be able to create a branch on a detached HEAD
20:03 rewt but the hash is just as good, if the branch won't be used for anything else
20:03 sigma__ If you want to keep them by creating a new branch, this may be a good time to do so with: git branch new_branch_name 176ca01
20:03 sigma__ I could've done that, I guess.
20:03 sigma__ Like you say.
20:04 rewt yeah
20:04 sigma__ Awesome. This all made sense, but wasn't intuitive at all. :P
20:04 sigma__ Thanks again. I'm done for the day, finally.
20:04 UTAN_dev joined #git
20:05 sigma__ left #git
20:06 Lipp joined #git
20:06 Lipp_ joined #git
20:07 pckz left #git
20:07 jimi_ joined #git
20:08 arahael2 joined #git
20:09 zeroed joined #git
20:10 WaReZ joined #git
20:12 Gsham joined #git
20:13 raijin joined #git
20:15 bronson joined #git
20:22 fuchstronaut joined #git
20:27 aidalgol joined #git
20:28 Siegfried left #git
20:29 Rayq joined #git
20:29 cdg joined #git
20:30 bjpenn joined #git
20:30 bjpenn i'm using atlassian stash/bitbucket, any way to find where a commit hash is from out of all the repos?
20:32 Macaveli joined #git
20:32 Oog joined #git
20:34 ochorocho__ joined #git
20:35 nnyk_ joined #git
20:36 Gsham joined #git
20:38 stangeland joined #git
20:38 stangeland Hi, i try to pull from my github repo, but when i do git pull it just hangs for a long time and then does: error: RPC failed; result=28, HTTP code = 0. What am I doing wrong?
20:39 leonarth joined #git
20:39 radsy joined #git
20:39 _ikke_ Seems like a github issue
20:39 _ikke_ code 28 is a timeout
20:39 _ikke_ either that, or a network issue
20:41 johnmilton joined #git
20:42 stangeland _ikke_, hmm ok...how could it be a network issue if i just do clone https://... and I can browse https adresses in my browser witout problems?
20:43 toogley joined #git
20:44 faux___ joined #git
20:44 tobiasvl different stuff that serves git repos via http and web pages? who knows
20:44 tobiasvl !sgc
20:44 gitinfo For bug reports, feature requests, suggestions for improvement, or specific problems with an individual repo (or gist, etc.), please contact support via support@github.com or https://github.com/support - support tries to answer all questions within 24 hours, including weekends and holidays.
20:46 liquidjamm joined #git
20:47 diegoviola joined #git
20:50 mjuszczak joined #git
20:52 guardian hello, is there an equivalent of git checkout --orphan for bare repositories?
20:53 jimi_ joined #git
20:53 csd_ joined #git
20:53 diego2 joined #git
20:53 diego2 joined #git
20:53 rewt bare repos have nowhere to checkout
20:54 radsy joined #git
20:54 avar The "equivalent" would be: git init /tmp/foo && cd /tmp/foo && touch a && git add a && git commit -m"first" && git remote add origin /path/to/bare/repo && git push origin master:your-new-orphan-branch
20:54 guardian so I have a bare repo, and for each branch I would like to truncate history to latest commit
20:55 ojacobson You can do it with the plumbing commands (git commit-tree et al) but you'll have to build it yourself
20:55 ojacobson and the usual !rewrite cautions apply
20:55 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
20:56 LeBlaaanc joined #git
20:56 smithbone joined #git
20:57 guardian oh well it's a submodule which history contains large binaries
20:57 guardian I just want to create a new repository and switch submodule
20:57 guardian but since it contains many branches I would like to automate that
20:58 jagob joined #git
20:58 diegoaguilar joined #git
20:59 guardian and I would like to keep the commit message and author etc
20:59 guardian for the latest commit that becomes the new initial commit
20:59 dmto joined #git
21:00 johnmilton joined #git
21:01 diego2 joined #git
21:02 Lipp joined #git
21:04 diego2 joined #git
21:04 c0ded joined #git
21:04 efco joined #git
21:06 jimmybuffet joined #git
21:06 jason237 joined #git
21:10 jagob joined #git
21:11 diego2 joined #git
21:13 IWishIKnew joined #git
21:15 astrofog joined #git
21:16 cyan__ joined #git
21:19 Lipp joined #git
21:20 mmlb joined #git
21:20 deanj joined #git
21:21 Dougie187 left #git
21:23 deanj Hey all, I was curious if anyone else has made the transition from svn -> git and found a way (tool, system, or otherwise) to easily maintain the visible order of commits/revisions. There's a project I'd like to switch from svn -> git but the biggest argument against the move is that using hashes for commmits makes it infinitely more difficult to figure out how it lines up with surrounding commits.
21:25 c0ded joined #git
21:26 Gsham joined #git
21:27 apotry joined #git
21:28 _ikke_ deanj: Not sure what your reasons are, but I hardly have a reason to have this revision numbering
21:28 _ikke_ Can you perhaps explain what your usecase is?
21:28 _ikke_ Note that things like branches make this number hard to define
21:29 fuchstronaut joined #git
21:30 fracting joined #git
21:31 sebboh deanj, from memory, there is a tool called repo-surgeon that might be related to what you want to do.  But I'm not sure.  Good luck!
21:31 acetakwas joined #git
21:32 sebboh deanj, but _ikke_ is right.  In git, you use the log to see the order of commits.  In svn, they are numbered sequentially, yes.. but there is a sequence in git, too.  You just have to look at the log.  (Which you have to do in svn anyway.)
21:33 mwhooker joined #git
21:33 tobiasvl how does seeing the number "1762" on a commit in svn help you figure out how it lines up with surrounding commits
21:33 sebboh tobiasvl: it's right after 1761.
21:33 zincrokx joined #git
21:33 EvilPenguin joined #git
21:33 tobiasvl :)
21:34 troyt joined #git
21:34 tobiasvl and a67b23 is right after... a67b23^
21:34 zincrokx joined #git
21:35 sebboh tobiasvl: oh, sure, I forgot about those suffix chars.
21:35 chrisshattuck joined #git
21:35 zincrokx joined #git
21:35 unbalancedparen joined #git
21:36 zincrokx joined #git
21:36 j416 deanj: this is never a problem. I'd challenge them to give a real use-case where it's a problem.
21:37 j416 deanj: sure, you can get a feel for how old a commit it by its revision number. but, why do you need this information outside of its context?
21:38 dopesong joined #git
21:39 MaddHatter sebboh: right after in time, but not necessarily right after in commit history
21:39 dopesong joined #git
21:40 rominronin joined #git
21:40 sebboh svn has branches?  Okay.  ...I misread the original question anyway, I thought the user wanted to convert an existing repo.  ...Back to idle for me. :)
21:41 jaguarmagenta joined #git
21:41 annoymouse joined #git
21:42 MaddHatter CVS and RCS have branches, too
21:44 a3Dman joined #git
21:48 Dougie187 joined #git
21:48 _ikke_ sebboh: Not if I created a commit on another branch
21:49 _ikke_ sorry deanj
21:51 kus joined #git
21:51 minot joined #git
21:54 gaen joined #git
21:54 briankc left #git
21:57 fuchstronaut joined #git
21:57 fuchstronaut joined #git
22:00 thebloggu joined #git
22:01 mizu_no_oto joined #git
22:02 jstimm joined #git
22:03 arahael2 joined #git
22:05 Assimilater left #git
22:06 ochorocho__ joined #git
22:06 Shentino joined #git
22:07 akushner joined #git
22:08 stantonk joined #git
22:08 arahael3 joined #git
22:09 crayon joined #git
22:10 leonarth_ joined #git
22:12 Guest31 joined #git
22:13 eletuchy joined #git
22:14 davisonio joined #git
22:14 deanj sebboh: _ikke_ j416 Tracking regressions across a repository can be difficult when you have to identify all regressions by a commit hash rather than an easily identifiable (and memorable) number that also shows where in the tree a given regresssion happened.
22:15 arahael4 joined #git
22:15 madewokherd joined #git
22:18 MaddHatter I don't understand what's being "tracked" in that case. If you really want to attach a human-friendly name to a hash, you can give it a tag
22:18 a3Dman joined #git
22:19 deanj MaddHatter: Commits. That introduce regressions (build failures, performance, etc.)
22:19 j416 deanj: if you find the commit that introduced the issue, just: git branch --contains thatcommit
22:19 j416 deanj: done
22:19 MaddHatter deanj: what is there to track? There's a commit, and unless it got svn diff | svn patched into some other place, there's only one commit, in only one place?
22:20 deanj But sometimes we need to see if commit <Y> is between commit <X> and <Z> or if it came before <X> or after <Z>. Because we keep track of dozens of regressions at once, using numbers makes things much easier to coordinate and remember, whereas using hashes would make it much more difficult.
22:20 tobiasvl it's just a different workflow is all
22:21 khmarbaise joined #git
22:21 j416 deanj: if you're working on multiple regressions at once, perhaps create branches for them
22:21 arahael joined #git
22:21 MaddHatter deanj: fair enough. There are commands in git to give you relative positioning and contains and such. In SVN, you're probably relying on the assumption that 1679 is between 1678 and 1680 (which isn't always true, but is probably "true enough" for the use cases you have)
22:22 j416 deanj: if you're not working on them, I'm not sure why you need to keep track of them all at the same time. maybe you have some use-case.
22:23 dan2k3k4 joined #git
22:23 deanj tobiasvl: thanks :) All I'm trying to do is see if anyone else has moved from svn -> git and figured out a goodeasy way to perserve the idea of a human readable commit/hash to make it easier to track how commits have landed.
22:23 j416 deanj: I've moved from svn to git.
22:23 FilipNortic joined #git
22:23 deanj MaddHatter: Yeah... I've brought that up, but some people don't like the much higher percieved effort to track commits in that way.
22:23 tobiasvl I've also moved from svn to git, didn't miss the sequential commit numbers
22:24 tobiasvl didn't use them much before
22:24 MaddHatter "numbers are smaller (and easier to remember) than abbreviated hashes" is just a thing one has to deal with. And in hindsight (having been through the transition) it's not as bad as it looks in front of you, but that's a hard thing to sell to someone without already having it in the rearview mirror
22:25 deanj MaddHatter: Yeah, I'm just trying to mediate a solution that allows us to move to a better tool (git), in my opinion, while being able to let others maintain their workflows as easily as possible.
22:25 MaddHatter the largest project I work on is > 1M commits now, so even revision numbers aren't very memorable
22:25 sssilver_ joined #git
22:25 tobiasvl that's a lot of commits
22:25 j416 deanj: also, in case you are not aware, you only need so much of a hash to make it unique in the repo. "git show 23cc2" may be enough.
22:25 MaddHatter tobiasvl: the .git is ~4GB
22:25 ojacobson deanj: git-describe gives you a weak approximation of a commit sequence number, relative to some reference point; it also includes the hash
22:25 tobiasvl MaddHatter: the largest project I work on has commit history back to 2002, but still only 18k commits I think
22:26 MaddHatter this history dates to 2005, it looks like
22:26 mwhooker joined #git
22:28 arahael1 joined #git
22:29 deanj j416: Yeah, usually 5 characters or so has been fine for me in the past, but the main argument is still that it's less representative of order than r87277 for example.
22:29 CubiK joined #git
22:31 j416 deanj: why is order important out of context?
22:31 hahuang61 joined #git
22:31 rahtgaz joined #git
22:32 deanj j416: Tracking regressions is what they say, but I imagine it's probably more than looking at a 6 character commit hash tells you nothing where as a 5-6 character revision you can at least glean some info from. For example, we could know that a r50000 commit was from a couple years ago where was r100000 was from yesterday
22:33 j416 deanj: I grasped about 70% of what you wrote, but ok
22:34 j416 deanj: why is that important?
22:35 stantonk joined #git
22:35 j416 deanj: if it's a bug you need to fix, does it matter if it's old or new?
22:35 j416 deanj: and if you are trying to find the commit, you already have the context right in front of you, and it's trivial to see how old it is
22:35 stantonk joined #git
22:37 LeBlaaanc joined #git
22:37 hexagoxel joined #git
22:37 j416 deanj: I see your point. I'm just trying to explain that I don't think there is a use-case.
22:38 j416 deanj: in 7+ years of using git almost daily, I've never had the need of the information you describe, out of context.
22:38 j416 deanj: in context, I can easily find all info I need.
22:38 arahael1 left #git
22:38 Arahael joined #git
22:39 deanj j416: Sure, it's not that the info "isn't available" whatsoeever. It's that the info isn't "instantly available"
22:39 axl__ joined #git
22:39 Gsham joined #git
22:39 j416 deanj: and, if revision numbers is the biggest dealbreaker for moving from svn to git, then that's a pretty small obstacle to get over for what git gains you in other aspects.
22:39 AmR|EiSa joined #git
22:39 j416 deanj: so, that sounds quite positive to me. :)
22:40 Arahael deanj: the date has always been "instantly" available to me... it's available jn every context you would want it.
22:40 Arahael deanj: git blame, gitk, git show...
22:40 ojacobson deanj: git-describe gives you a weak approximation of a commit sequence number, relative to some reference point; it also includes the hash, which means it works properly in a distributed context in ways a bare rev number can'
22:40 * j416 has an odd feeling of déjà-vu
22:40 ojacobson indeed
22:41 AmR|EiSa How can that be achieved if the files in the project are the changes and if yes how I can get files list of these change ?
22:41 deanj j416: I'm a huge fan of git and I hate svn (for variable reasons), you don't have to convince me :)
22:41 deanj AmR|EiSa: git diff HEAD
22:41 j416 deanj: I'm trying to give you ammo to convince your peers :)
22:42 deanj j416: Haha, unfortunately that argument falls on deaf ears when it's a team needing to learn a new complex tool/ecosystem, and they don't immediately see most of the benefits because they primarily use it only for checkouts/checkins.
22:42 deanj But I do largely appreciate all your points, so thanks!
22:42 j416 deanj: assuming you are the main driver of moving from svn, I'd probably start with setting up a workflow that fits your needs
22:43 j416 deanj: yeah, I was in what seems from your description as the exact same situation a couple of years ago.
22:43 Arahael deanj: also, consider the team workflow and infrastructure.  not having git may well be a comparatively minor issue if you, evidently, don't have a decent issue tracking system.
22:45 sssilver joined #git
22:45 johnnyfive joined #git
22:47 a3Dman joined #git
22:47 Omen joined #git
22:50 italoacasas joined #git
22:55 stantonk_ joined #git
22:56 Lipp joined #git
22:57 arahael1 joined #git
22:59 Linkandzelda joined #git
22:59 Linkandzelda joined #git
23:00 axl_ joined #git
23:01 deanj j416: Yeah, that's why my original question asked about if anyone had a workflow they used that preserved the revison numbers but allowed them to switch to git. :) But yeah, change is hard.
23:02 j416 the answer is, you don't need them
23:02 j416 if you need to keep track of certain versions, use tags
23:03 j416 if you need to start work from some certain state, create a branch
23:03 arahael2 joined #git
23:08 Shakeel__ joined #git
23:10 arahael3 joined #git
23:11 faux___ joined #git
23:12 dmto_ joined #git
23:13 LeBlaaanc joined #git
23:14 fuchstronaut joined #git
23:14 AmR|EiSa deanj, It's give me " warning: LF will be replaced by CRLF in Code/Libs/CrashHandler/BreakPad/sr​c/client/ios/BreakpadController.h.
23:14 AmR|EiSa The file will have its original line endings in your working directory.
23:14 AmR|EiSa " I don't need that.
23:15 dmto_ joined #git
23:15 ahmedelgabri joined #git
23:16 arahael3 joined #git
23:22 fahadash joined #git
23:22 arahael4 joined #git
23:22 deanj AmR|EiSa: Running which command gives you there?
23:22 deanj that*
23:24 AmR|EiSa deanj, git diff HEAD
23:25 dmto joined #git
23:33 Lipp joined #git
23:34 ochorocho__ joined #git
23:36 nl joined #git
23:38 dsantiago joined #git
23:41 mablae_ joined #git
23:41 rominronin joined #git
23:42 jaguarmagenta joined #git
23:46 aghalarp joined #git
23:46 jimi_ joined #git
23:48 |ifei5g00d joined #git
23:48 chrisshattuck joined #git
23:49 CJKinni joined #git
23:50 BSaboia joined #git
23:52 Guest83419 joined #git
23:58 mattfoo joined #git
23:59 aghalarp joined #git
23:59 fstd joined #git

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