Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-01

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

All times shown according to UTC.

Time Nick Message
00:05 whaley joined #git
00:08 rivarun joined #git
00:10 User458764 joined #git
00:11 thiago joined #git
00:12 rivarun joined #git
00:13 ws2k3 joined #git
00:14 anuxivm left #git
00:14 ChadTaljaardt joined #git
00:18 matoro joined #git
00:22 dendazen joined #git
00:28 cagomez joined #git
00:28 SwiftMatt joined #git
00:29 bvcosta joined #git
00:29 dsdeiz joined #git
00:29 dsdeiz joined #git
00:29 smurfendrek123 joined #git
00:29 zumba_addict joined #git
00:30 Aboba joined #git
00:36 pks joined #git
00:39 Sasazuka__ joined #git
00:42 raijin joined #git
00:42 Darren_ joined #git
00:43 Fallen0223 joined #git
00:43 humboldt joined #git
00:43 nowhereman joined #git
00:52 bgerber joined #git
00:53 manuels joined #git
00:54 b1tchcakes joined #git
00:54 aidalgol joined #git
00:55 matoro joined #git
00:56 DolpheenDream joined #git
00:56 peacememories joined #git
00:56 bgerber joined #git
00:57 rivarun joined #git
00:58 alexandre9099_ joined #git
01:01 Sasazuka joined #git
01:01 raspado_ joined #git
01:05 kristofferR joined #git
01:05 sgen joined #git
01:06 marianina8 joined #git
01:09 gugah joined #git
01:11 a_thakur joined #git
01:12 OnkelTem phy1729: sorry was afk. Now. Currently they are listed in the untracked section
01:12 madewokherd joined #git
01:14 aruns joined #git
01:16 indistylo joined #git
01:16 OnkelTem fixed
01:16 OnkelTem that's odd
01:16 OnkelTem I had to start with current directory!
01:16 OnkelTem i.e. with www/sites it works, and without www - doesn't
01:16 Levex joined #git
01:23 jameser joined #git
01:29 moei joined #git
01:32 dsdeiz joined #git
01:33 aruns__ joined #git
01:34 dingbat joined #git
01:35 jason237 joined #git
01:35 cqi joined #git
01:36 dingbat Hello kind people. I am trying to understand gitignore syntax. I want to ignore all files of a certain suffix, except when they are in a specific directory. I have tried "*.s" + "!foo/asm/*.s" and it does not seem to work
01:36 dingbat Any suggestions?
01:36 sbasso joined #git
01:36 alzagros joined #git
01:37 ResidentBiscuit joined #git
01:39 Sasazuka joined #git
01:40 offtopik joined #git
01:41 pur3eval joined #git
01:41 Sasazuka what's the difference between: "git rebase master" on my foo branch versus branching off of foo then "git rebase master" ?
01:42 Sasazuka I usually run into less issue when I do the latter so I'm trying to understand what's happening
01:43 thiago joined #git
01:44 menip joined #git
01:45 aolusr joined #git
01:47 Darren_ joined #git
01:48 matoro joined #git
01:49 a_thakur joined #git
01:49 mattcen dingbat: Have you read the gitignore manpage?
01:49 dingbat mattcen: I read this: https://git-scm.com/docs/gitignore
01:50 mattcen OK
01:50 gclaugus left #git
01:50 dingbat And what I had above is what I thought would work, based on said reading. But perhaps there is something subtle I am not understanding?
01:51 thebope joined #git
01:53 mattcen dingbat: seems to work for me: http://sprunge.us/WEfF. what version of git?
01:54 mattcen Oh. Almost forgot why I'm here.
01:54 mattcen Hi folks! How can I reject any attempt to generate a merge commit, to enforce linear history (by whatever means necessary; rebasing, cherry-picks, etc.)?
01:56 Eugene You really can't/shouldn't. Non-linear history is not the devil. But anyway, man githooks or you should be able to do this through your hosting provider(eg, Github has options for it per-repo)
01:56 gitinfo the githooks manpage is available at http://jk.gs/githooks.html
01:57 Eugene You will need to find(or write) a hook that inspects & sanity-checks history however you like; the pre-commit hook runs locally before a commit is created. pre-receive is a server-side hook that is executed when a client runs"git push" , before it accepts the new commit(s).
01:59 cdg_ joined #git
01:59 mattcen Eugene: Thanks; I assumed I'd need a hook; just wasn't sure quite how to craft it. And I know merges aren't the devil; the company I've started working for enforces this on their end (for reasons they will explain to me tomorrow, but for now I'm just trusting are sane), and I wanted to enforce locally so it didn't get to their end before I noticed
02:00 dingbat mattcen: "git version 2.10.1 (Apple Git-78)". Same result in both Github desktop and command line
02:00 mattcen dingbat: Odd. "git version 2.6.4 (Apple Git-63)" is me
02:00 dingbat Hmm ok I'll keep trying, maybe the nested folder is confusing things
02:01 Eugene The story usually goes something like "We've always used SVN, but we've been hearing about git. Let's move to that!". This is quickly followed by "what the fuck is a merge and why did it break all of our code this is garbage"
02:01 mattcen dingbat: I recommend trying the sample I used above for sanity-checking
02:01 Eugene And then the inevitable dumbshit CTO Decision.
02:01 dingbat Eugene: hahahaha I've heard that story before
02:01 dingbat mattcen: will do, thanks
02:01 Eugene A few days or months later, the poor coders end up here in #git asking why the hell and how do they get around it
02:02 thebope joined #git
02:03 Sasazuka__ joined #git
02:04 mattcen dingbat: http://sprunge.us/gDUh is what I ran, basically
02:05 mattcen Eugene: Heh. Thanks for the commentary :)
02:05 ankit01ojha joined #git
02:05 jameser joined #git
02:06 Eugene Any time. Good luck
02:07 Eugene FWIW, you can make a simple pre-commit hook that checks for the existence of .git/MERGE_HEAD
02:07 Eugene And abort the commit(exit 1) if so
02:10 Levex joined #git
02:10 dingbat mattcen: Oooh I think I got it. I didn't notice that command line git was actually picking it up
02:11 dingbat mattcen: so it looks like the problem is that Github Desktop doesn't match the behaviour of git for gitignores
02:12 dingbat basically, when I run "git status", I see untracked files "sandbox/asm/", but Github Desktop doesn't show that
02:12 dingbat Welp, this is awkward. Looks like Github Desktop has a bug
02:13 marianina8 joined #git
02:14 pur3eval joined #git
02:15 raijin joined #git
02:15 Sasazuka joined #git
02:16 olbat joined #git
02:16 olbat joined #git
02:18 MattMaker joined #git
02:18 dreiss joined #git
02:18 nettoweb joined #git
02:19 Gsham joined #git
02:20 tgp_ joined #git
02:24 finalbeta joined #git
02:24 nettoweb joined #git
02:25 clemf joined #git
02:31 cagedwisdom joined #git
02:34 dingbat Aha. Looks like a bug in libgit2: https://github.com/libgit2/libgit2/issues/3224
02:36 ngladitz joined #git
02:36 ngladitz joined #git
02:37 dingbat However, manually adding the files in question with 'git add' makes Github Desktop "see" the files
02:38 d^sh joined #git
02:38 prg3 joined #git
02:40 d0nn1e joined #git
02:42 Emperor_Earth joined #git
02:46 Vampire0_ joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help. Just git help, apparently. | We are aware of the SHA-1 collision. See: http://jk.gs/t/a4 | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.12.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This is serious business. https://madeitwor.se/git-jokes
02:49 Richard_Cavell_ joined #git
02:51 Electrometro joined #git
02:51 ResidentBiscuit joined #git
02:51 durham joined #git
02:52 Electrometro joined #git
02:53 Tomi joined #git
02:53 cdown joined #git
02:56 Tomi I'm reading the git book at git-scm and everything is going smoothly, but I can't explain this. When I do "git clone https://github.com/Tomi/private-repo" on a clean directory it is successfull. How does git have access to my private repo?
02:57 Tomi What chapter in the git book explains how this works?
02:57 aavrug joined #git
02:58 dsantiago joined #git
02:59 clemf joined #git
03:03 cqi joined #git
03:03 ojdo joined #git
03:06 fstd_ joined #git
03:07 hexagoxel joined #git
03:11 duderonomy joined #git
03:12 clemf joined #git
03:13 dave0x6d joined #git
03:13 chachasmooth joined #git
03:17 Electrometro joined #git
03:17 DaYwalKer joined #git
03:17 Vortex34 joined #git
03:18 raijin joined #git
03:19 jottr joined #git
03:20 trip joined #git
03:22 Gsham joined #git
03:22 ayogi joined #git
03:22 jon-mac joined #git
03:23 MattMaker joined #git
03:24 MattMaker joined #git
03:26 Goplat joined #git
03:33 arescorpio joined #git
03:34 salamanderrake joined #git
03:35 dirtyroshi joined #git
03:35 ayogi joined #git
03:37 sbasso joined #git
03:40 thebope_ joined #git
03:42 matoro joined #git
03:42 dzho !github
03:42 gitinfo Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
03:43 dzho so, the git book won't talk about this.
03:43 dzho because this is a github-specific thing.
03:43 dzho Tomi: have you set up ssh keys with github?
03:43 dzho if so, that's one way it can authenticate you without seeming to need to.
03:45 sunri5e joined #git
03:47 avatar_ joined #git
03:48 watabou joined #git
03:49 raijin joined #git
03:51 red82 joined #git
03:53 pigeon joined #git
03:53 a_thakur joined #git
03:55 snowalpaca joined #git
03:56 govg joined #git
03:56 mizu_no_oto joined #git
03:57 jason237 joined #git
04:02 krarkrrrc left #git
04:02 jimsio joined #git
04:03 Limes_ joined #git
04:06 gugah joined #git
04:07 dsdeiz joined #git
04:07 dsdeiz joined #git
04:11 Electrometro joined #git
04:11 Darren_ joined #git
04:12 gtristan joined #git
04:13 lowercaseman joined #git
04:14 mattcen dingbat: Glad you've narrowed it down!
04:15 a_thakur_ joined #git
04:15 dingbat mattcen: aye, thanks for the help. Unfortunately, it seems like a bug that nobody cares enough to fix, but also explains why my editor also greys out those files :), must be using libgit2 as well
04:16 sbulage joined #git
04:17 pur3eval joined #git
04:18 mattcen Unfortunate indeed. At least now it's a *know* bug to you!
04:20 sparr Is there a way I can force git to recognize a file that has moved and changed, short of reverting the changes and committing the move then reapplying the changes?
04:20 rubyonrailed joined #git
04:21 ISmithers joined #git
04:24 mattcen That sort of detection is done at run-time, not at commit time. As in, when you do "git log" or similar, only then does it compare the files and realise that they match or almost match. I think the answer to your question is "no"
04:25 raspado joined #git
04:26 kadoban With a just to make sure: redoing the commit would be useless, 'git mv' doesn't record anything special.
04:26 jfelchner joined #git
04:27 cahoots joined #git
04:32 Darren_ joined #git
04:34 Eugene joined #git
04:35 blunaxela joined #git
04:37 anohigisavay joined #git
04:37 dtrott joined #git
04:43 nellicus_ joined #git
04:47 DarkPsydeLord joined #git
04:48 Cabanoss- joined #git
04:48 thebope joined #git
04:49 durham joined #git
04:55 ayogi joined #git
04:56 cahoots joined #git
04:56 muhannad_______ joined #git
04:59 diogenese joined #git
05:01 durham joined #git
05:02 mattcen kadoban: but when git log sees 2 separate operations, a deletion+addition of two files whose hashes match, and then a change to the latter file in a separate commit, it's likely to have an easier time distinguishing the shared history, right?
05:03 cahoots joined #git
05:03 mattcen I mean, there's always git log --follow, which probably adds more smarts
05:03 mattcen also -M
05:05 venmx joined #git
05:08 svm_invictvs joined #git
05:12 Levex joined #git
05:15 gtristan joined #git
05:15 dreiss joined #git
05:17 rivarun joined #git
05:17 dsdeiz joined #git
05:18 pur3eval joined #git
05:21 MrWoohoo joined #git
05:23 acetakwas joined #git
05:27 dtrott joined #git
05:29 lucido-cl joined #git
05:33 cahoots joined #git
05:34 rawburt I'm ignoring some files in my repo, but I don't know how exactly. I think I told git not to track them, and forgotten how. Don't know where to list them. git ls-files --others -i --exclude-standard didn't work. Any ideas?
05:35 MattMaker joined #git
05:36 dtrott rawburt: Typcially either a .gitignore file in your repo or you may have added an excludesfiles option in ~/.gitconfig
05:36 pks joined #git
05:38 sbasso joined #git
05:39 SwiftMatt joined #git
05:40 causasui joined #git
05:40 rawburt So I don't have a .gitconfig file anywhere :/
05:40 causasui joined #git
05:40 kadoban rawburt: man git check-ignore
05:40 gitinfo rawburt: the git-check-ignore manpage is available at http://jk.gs/git-check-ignore.html
05:40 matoro joined #git
05:41 kadoban Especially notice the -v flag, which makes it very useful for this type of thing
05:42 rchavik joined #git
05:42 rawburt git check-ignore -v <file>
05:42 rawburt Nada.
05:42 rawburt When I do git diff <file> there are no differences shown, but there very clearly are differences.
05:43 rawburt There are other files in that directory that are showing changes, and git ls-files lists the file that has been changed but not showing up in git status.
05:44 rawburt Hmm.. maybe I'm getting that path wrong or something, it's 1 layer deep. I'm in the root. git check-ignore -v dir/file.csv
05:45 pur3eval joined #git
05:46 rawburt Ok, found something.
05:46 watabou joined #git
05:47 rawburt They have lowercase h's whereas everything else has uppercase H's in git ls-files -v *
05:48 dersand joined #git
05:48 _ikke_ rawburt: assume-unchanges
05:48 _ikke_ rawburt: assume-unchanged
05:48 rawburt _ikke_: Thank you!
05:48 rawburt That is what I forgot.
05:48 Sample joined #git
05:49 chachasmooth joined #git
05:49 rawburt Thank you kadoban and dtrott
05:50 kadoban Eww, assume-unchanged
05:54 hoodrat joined #git
05:55 cahoots joined #git
05:55 diogenese joined #git
05:55 rawburt I'm only using it on these CSV files because when the application runs it populates with sensitive data.
05:56 rawburt The original commit has the fields at the top of the CSV with no data, serves as a reference/starting point.
05:56 pks_ joined #git
05:56 bocaneri joined #git
05:57 _ikke_ Why not use a separately named template?
05:57 rawburt Hmmm... well
05:57 rawburt I think the main reason would because I don't know what I'm doing
05:57 Guest9 joined #git
05:58 _ikke_ !assume
05:58 gitinfo [!assume_unchanged] git update-index --assume-unchanged is sometimes suggested for keeping yourself from committing changes to a file, but it's actually a promise to Git that the file is identical to the committed version, to reduce unnecessary disk reads. Not only will Git feel free to discard the flag, it will also happily overwrite your changes whenever a checkout/merge/rebase/... updates it.
05:58 Guest9 joined #git
05:58 potatoe left #git
05:58 rawburt Interesting
05:58 kadoban much of !config could be interesting, that's not actually config files, but it sounds similar regardless
05:58 gitinfo [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas
05:59 windy-black joined #git
05:59 kadoban tl;dr version: probably just ignore them and have the program create the headers when the file doesn't exist.
06:00 thebope joined #git
06:00 rawburt This changes some things for me. However this is much better with gitignore.
06:01 rawburt Thanks again _ikke_ and kadoban
06:01 rawburt You guys are awesome.
06:01 kadoban :)
06:01 fracting joined #git
06:05 MattMaker joined #git
06:08 rscata joined #git
06:08 dec0n joined #git
06:10 ozmage joined #git
06:11 hexagoxel joined #git
06:15 noteness joined #git
06:15 clemf joined #git
06:17 marianina8 joined #git
06:17 overlord_tm joined #git
06:24 MineCoins joined #git
06:29 sz0 joined #git
06:33 sarri joined #git
06:33 sarri joined #git
06:36 wilbert joined #git
06:37 Darren_ joined #git
06:37 nivag joined #git
06:38 thebope_ joined #git
06:38 King_Hual joined #git
06:38 King_Hual joined #git
06:39 darkbit joined #git
06:39 wilbert_ joined #git
06:39 akkad joined #git
06:43 tgp__ joined #git
06:44 zumba_addict joined #git
06:45 zumba_addict evening folks. When is it ideal to execute git tag?
06:45 hexagoxel joined #git
06:47 brent__ joined #git
06:49 sbulage joined #git
06:50 thebope joined #git
06:52 modin joined #git
06:53 ThomasLocke joined #git
06:53 nivag joined #git
06:53 zefferno joined #git
06:56 chachasmooth joined #git
06:58 lowercaseman joined #git
06:59 anohigisavay zumba_addict, when you want to create a release and add a version number
06:59 zumba_addict got it
07:00 Pavel-K joined #git
07:00 zumba_addict our app gets built in Jenkins. The other tool we need if we want to be able to revert requires the repo to be tagged. Is it ok to tag everytime our Jenkins make a build?
07:01 zumba_addict i'll write a script to tag based on jenkins build number
07:02 MattMaker joined #git
07:02 Raging_Hog joined #git
07:04 thebope joined #git
07:04 pur3eval joined #git
07:06 venmx joined #git
07:07 eroux joined #git
07:08 MattMaker joined #git
07:11 eden_ joined #git
07:13 muhannad_______ joined #git
07:13 mangueLinux_ joined #git
07:16 qt-x joined #git
07:18 chele joined #git
07:20 clemf_ joined #git
07:22 gelei007 joined #git
07:23 JeroenT joined #git
07:23 bschindler joined #git
07:24 bschindler Is there a way to get the location of git root within a clean filter?
07:26 eroux joined #git
07:26 thiago git rev-parse --show-prefix
07:26 thiago or --show-toplevel
07:27 bschindler so if I need the output of that within the clean filter, I'd have to do `git rev-parse --show-prefix`/something - is this supported on non-unix platforms?
07:28 tgp_ joined #git
07:28 zumba_addict hey folks, if I upgrade to private account in github, will I still be able to create public repo?
07:30 jagob joined #git
07:30 hexagoxel joined #git
07:30 _ikke_ !github
07:30 gitinfo Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
07:30 Mobutils joined #git
07:31 _ikke_ zumba_addict: but yes, when you create a repo, you can decide whether it's public or private
07:32 zumba_addict oh cool. Thanks _ikke_
07:33 ronskisika joined #git
07:35 ShapeShifter499 joined #git
07:39 lordjancso joined #git
07:39 sbasso joined #git
07:40 zumba_addict i now have a private account :)
07:40 GT4066 joined #git
07:40 wrongdevice joined #git
07:42 jnavila joined #git
07:43 b100s joined #git
07:43 Balliad joined #git
07:44 thebope_ joined #git
07:47 dedicated joined #git
07:52 ResidentBiscuit joined #git
07:52 ski7777 joined #git
07:55 jungsubk joined #git
07:56 finalbeta joined #git
07:56 thebope joined #git
07:56 chachasmooth joined #git
07:56 tgp_ joined #git
07:57 koneko joined #git
07:59 a_thakur joined #git
07:59 ToBeCloud joined #git
07:59 koneko_ joined #git
08:00 ski7777 joined #git
08:01 Phylock joined #git
08:02 Joost how would you add `git submodule update --init` to a Makefile of a project?
08:03 ngladitz joined #git
08:03 ngladitz joined #git
08:04 _ikke_ How would you add any shell command to a Makefile?
08:04 _ikke_ (I'm not a Makefile expert)
08:05 bashcoder joined #git
08:05 pur3eval joined #git
08:05 Joost sure, I could just add it as a command to execute whenever my target builds
08:05 Joost I hoped to do something like only execute if it's necessary
08:05 Joost but I guess `git submodule update --init` includes that check by itself
08:05 Joost so there's no point
08:06 Joost i.e. it would be a practically free no-op if the module exists and is up to date
08:07 _ikke_ correct
08:08 _ikke_ Joost: you would have to do the same amount of work yourself as that command does
08:08 Joost yeah
08:08 Joost that's pretty convincing :)
08:09 User458764 joined #git
08:11 michael_mbp joined #git
08:11 mcortinas joined #git
08:11 humboldt joined #git
08:11 TheWild joined #git
08:12 TheWild hello
08:12 TheWild on Windows: is this safe to run multiple git instances at once (git pull, different repositories)?
08:13 mehola joined #git
08:13 _ikke_ sure
08:14 gelei007 joined #git
08:14 TheWild oh good. Thank you _ikke_
08:15 elect_ joined #git
08:15 GodGinrai joined #git
08:15 TomyWork joined #git
08:16 mindfart joined #git
08:18 elect joined #git
08:21 elect_ joined #git
08:21 zeroed joined #git
08:21 zeroed joined #git
08:23 ckruczek joined #git
08:23 elect__ joined #git
08:24 jajvpy joined #git
08:25 dvaske joined #git
08:25 dvaske_ joined #git
08:25 armyriad joined #git
08:26 elect joined #git
08:26 MattMaker joined #git
08:26 b100s joined #git
08:26 King_Hual joined #git
08:28 elect_ joined #git
08:30 elect__ joined #git
08:30 trip joined #git
08:32 a_thakur joined #git
08:32 modin joined #git
08:33 jave joined #git
08:34 humboldt joined #git
08:35 arunpyasi joined #git
08:35 Snugglebash joined #git
08:37 armyriad joined #git
08:37 elect joined #git
08:39 YuGiOhJCJ joined #git
08:40 mattcen joined #git
08:42 Xatenev joined #git
08:42 Xatenev Hi
08:42 Xatenev Quick question
08:42 Xatenev I did git reflog --date=local origin/myBranch
08:42 Xatenev that gives me one line: fe78d4e refs/remotes/origin/myBranch@{Mon Feb 27 13:57:01 2017}: update by push
08:42 Xatenev That means there was one push to that branch on mon feb 27 13:57:01 correct?
08:45 Xatenev Would be cool if anybody could tell me if im correct here or not :]
08:46 ^7heo I can tell you that you're correct.
08:46 ashokrajar joined #git
08:46 ^7heo But you shouldn't trust me.
08:46 Xatenev Are you sure? I ran that on my local branch
08:47 Xatenev Cause my colleague said I didnt push my stuff - but im 100% sure I did. lol
08:47 Xatenev Just wanna check whether I pushed my stuff( on what date and time )
08:47 ^7heo ok now I'm curious, so I'll read you question
08:47 ^7heo yor*
08:47 ^7heo ah fucking phone...
08:48 modin joined #git
08:48 osse Xatenev: do a git fetch then check if your commits are on that branch
08:48 Xatenev osse: They are now >definetly< on that branch because my colleague said he went to my pc and pushed it from there.
08:48 Xatenev So I can just check the time when it has been pushed.
08:49 ^7heo lock your workstation...
08:49 ^7heo what kind of workplace is that?
08:49 jast push time is difficult to figure out
08:49 Xatenev damn
08:49 Xatenev jast: so git reflog --date=local origin/myBranch isnt doing what I think?
08:49 osse i agree with ^7heo that the reflog indicates you pushed
08:50 jast well, reflog doesn't know what happened on the server. it will record that your 'git push' invocation changed the ref locally
08:50 ^7heo Xatenev: all that log says is that your LOCAL copy of that remote branch was updated then
08:50 jast which *is* a pretty good indication that you pushed. assuming fe78d4e is the revision you wanted to push, that's actually fairly clear
08:50 jast but your reflog doesn't know what pushes other people did
08:50 Xatenev I check that again, sec
08:51 Xatenev Yep it is fe78d4e
08:51 jast for all we know, someone else did another push a short while later and rewound the branch, undoing your push
08:51 overlord_tm joined #git
08:51 Xatenev So but I can say I pushed it at 13:57:01
08:51 ^7heo yeah
08:51 Xatenev okay thank you people :)
08:51 ^7heo not that it worked
08:52 ^7heo or that it stayed that was
08:52 jast if it hadn't worked at the time, push wouldn't have updated the ref
08:52 ^7heo or that you even pushed on the right remote branch
08:53 ^7heo jast: that's assuming the server is git
08:53 ^7heo jast: you'd be surprised what people work with
08:54 jast if it's not git, the meaning of 'worked' becomes much less clear anyway
08:54 elect joined #git
08:55 jast over a dumb transport, for example, updating the refs is all nice and arguably the push can be successful but people still won't actually fetch the new version of the ref (if the server has not run update-server-info)
08:56 ^7heo jast: yeah, I am no that familiar with the details
08:57 ^7heo jast: but my point was to warn him of what he shouln't claim :p
08:57 MattMaker joined #git
08:58 MattMaker joined #git
09:00 jwmann joined #git
09:00 flaviodesousa joined #git
09:01 User458764 joined #git
09:01 apotry joined #git
09:02 kurkale6ka joined #git
09:02 a_thakur joined #git
09:03 brent__ joined #git
09:04 matoro joined #git
09:04 hexagoxel joined #git
09:04 mikecmpbll joined #git
09:06 bongjovi joined #git
09:06 delboy1978uk joined #git
09:06 venmx joined #git
09:06 acetakwas joined #git
09:07 delboy1978uk is there a way i can do a git pull, one commit at a time? trying to find an offending commit that breaks something, i've reset back to a point where it works and want to go forward until i find it
09:07 delboy1978uk or must i iterate either side of it? hopefully not
09:07 ckruczek just do git fetch and check the commits
09:08 marcogmonteiro joined #git
09:08 ckruczek if you already pulled it and merged it you can do git bisect to find the breaking commit
09:08 ckruczek man bisect
09:08 delboy1978uk i dont know whjat code i'm looking for though, just want to get the next commit, check it works, get the next one, etc, until it breaks
09:08 Xatenev definetly bisect
09:08 Xatenev :P
09:09 ckruczek jop
09:09 delboy1978uk bisect? never even heard of that one
09:09 delboy1978uk Xatenev: o/
09:09 ckruczek go ahead and google it ;) it will be from tremendous help for you
09:09 kadoban man git bisect
09:09 gitinfo the git-bisect manpage is available at http://jk.gs/git-bisect.html
09:09 lowercaseman joined #git
09:09 ckruczek man git bisect
09:09 gitinfo the git-bisect manpage is available at http://jk.gs/git-bisect.html
09:09 ckruczek why is it not working
09:09 ckruczek !bisect
09:09 ckruczek DAMN IT
09:10 delboy1978uk wow bisect looks awesome!
09:10 Alienpruts joined #git
09:11 delboy1978uk !man git bisect
09:11 gitinfo The git man pages are available online at http://jk.gs/git.html. Or were you looking for the "man git-foo" syntax(without the !) ?
09:11 gitinfo the git-bisect manpage is available at http://jk.gs/git-bisect.html
09:11 ckruczek why didnt it worked for me kadoban
09:11 delboy1978uk ha, first try :-P
09:11 kadoban It did
09:11 metalraiden34 joined #git
09:11 ckruczek ok i didn't get any visual feedback, maybe gthe client is lacking I am using
09:12 finalbeta joined #git
09:16 cqi joined #git
09:18 cqi joined #git
09:18 koneko joined #git
09:19 dcpc007 joined #git
09:19 bruce_lee joined #git
09:21 humboldt joined #git
09:23 imack joined #git
09:23 JeroenT_ joined #git
09:25 smurfendrek123 joined #git
09:25 troyt joined #git
09:26 aard_ joined #git
09:26 jab416171 joined #git
09:30 notebox joined #git
09:33 AciD` joined #git
09:33 AciD` joined #git
09:37 chll_ joined #git
09:37 MattMaker joined #git
09:39 seni joined #git
09:40 acetakwas joined #git
09:40 Anja joined #git
09:41 manuelschneid3r joined #git
09:42 cr34ton joined #git
09:45 netj joined #git
09:45 howdoi joined #git
09:47 moei joined #git
09:50 bashcoder joined #git
09:51 muhannad_______ joined #git
09:52 elect_ joined #git
09:52 Murr joined #git
09:53 ResidentBiscuit joined #git
09:54 pur3eval joined #git
09:57 Cabanossi joined #git
10:03 cjbrambo joined #git
10:03 delboy1978uk git bisect is freaking awesome!!!
10:04 delboy1978uk 02e9a8e4b7787fbe4ce72b9225c214b77c19c632 is the first bad commit
10:04 delboy1978uk bloody brilliant
10:06 aswen joined #git
10:10 Tobbi joined #git
10:11 humboldt joined #git
10:15 jungsubk joined #git
10:15 ahr3n joined #git
10:16 GodGinrai joined #git
10:17 auzty joined #git
10:17 tvw joined #git
10:18 xall joined #git
10:20 JanC_ joined #git
10:20 aw1 joined #git
10:20 marenz_ joined #git
10:23 humboldt joined #git
10:24 hexagoxel joined #git
10:27 ismithers joined #git
10:28 finalbeta joined #git
10:31 ifra201 joined #git
10:31 noodle joined #git
10:33 timini joined #git
10:34 MattMaker joined #git
10:34 timini hey is there anyway to get all the patents of a commit?
10:35 timini imagine this senario: commit1: create readme, commit2: create howto, commit3: edit redme and howto
10:36 WayToDoor joined #git
10:36 timini is there some way in git to show that in order to apply commit 3 you have to apply commit1 and commit2 ?
10:36 timini actually
10:37 timini imagine this senario: commit1: create readme.txt, commit2: create howto.txt, commit3: create authors.txt, commit4: edit redme and howto
10:38 timini maybe that is a better illustration
10:38 JeroenT joined #git
10:40 delboy1978uk how do i diff one file with a specific commit?
10:42 cdown joined #git
10:45 sunri5e joined #git
10:46 auzty joined #git
10:47 JeroenT joined #git
10:47 jungsubk joined #git
10:48 afuentes joined #git
10:48 scarcry joined #git
10:50 ckruczek you do not diff files, you only diff commits delboy1978uk
10:53 mehola joined #git
10:53 bgerber joined #git
10:54 pur3eval joined #git
10:57 tobiasvl delboy1978uk: do you mean you want to diff one file from HEAD (aka the commit you have currently checked out) with the same file from another commit?
10:57 AlHafoudh joined #git
10:57 thebope joined #git
10:59 git821 joined #git
10:59 git821 joined #git
10:59 git821 joined #git
10:59 git821 hi guys... one quick question.
11:00 git821 let's say i checkout a branch from current develop and so does my collegue, i add something to file A, create merge request to develop... my collegue also adds something to file A, then rebases to develop...
11:01 User458764 joined #git
11:01 git821 problem is that my changes somehow get lost, and only his are visible.
11:02 git821 is this normal? how would i go around that?
11:02 humboldt_ joined #git
11:04 brent__ joined #git
11:04 ntonjeta joined #git
11:04 scarcry joined #git
11:05 delboy1978uk git821: your code hasnt been merged, so rebasing wouldnt have your code
11:07 venmx joined #git
11:09 dsdeiz joined #git
11:09 mda1 joined #git
11:11 hussam joined #git
11:12 ashokraj_ joined #git
11:17 rnsanchez joined #git
11:18 djb-irc joined #git
11:20 DaYwalKer joined #git
11:23 ferr1 joined #git
11:28 jameser joined #git
11:28 cdleonard joined #git
11:34 davimore_ joined #git
11:38 hexagoxel joined #git
11:38 User458764 joined #git
11:39 User458764 joined #git
11:40 dendazen joined #git
11:43 jmd joined #git
11:44 delboy1978uk left #git
11:45 smurfendrek123 joined #git
11:46 sbasso joined #git
11:46 stamina joined #git
11:48 chipotle joined #git
11:49 p4trix joined #git
11:50 nettoweb joined #git
11:51 LucaS05 joined #git
11:52 thebope joined #git
11:52 bvcosta joined #git
11:53 LucaS05 Hi. My team is experiencing this problem: https://hastebin.com/epequsufog.vbs I've just jumped on the project that has this "tag problem" and is the first time for me, approaching it. Do you have any suggestions on what to do? Also because those lines appear as an "error" (in red) in SourceTree. Thanks in advance
11:53 selckin whats the problem?
11:54 mehola joined #git
11:54 ^7heo SourceTree obviously.
11:54 tobiasvl LucaS05: can you explain what you mean by "tag problem"? there's no apparent "problem" in your hastebin
11:55 LucaS05 I've missed this message: Updates were rejected because the tag already exists in the remote sorry
11:56 koneko joined #git
11:56 xanadu_ joined #git
11:56 n1ce joined #git
11:57 LucaS05 tobiasvl: ^
11:57 grawity if you check `git ls-remote origin --tags` and `git ls-remote . --tags`, does the same tag point to different commits?
11:57 grawity if it points to the same object in both, then I'm not sure why sourcetree would even complain about that
11:58 LucaS05 I will check that. Anyway I've found this: https://www.raymonschouwenaar.nl/git-error-updates-rejected-tag-already-exists-remote-sourcetree/
11:58 LucaS05 Maybe is not a git problem, but a SourceTree one
11:58 LucaS05 as ^7heo has said ;)
11:58 rvgate joined #git
12:00 bvcosta joined #git
12:02 jost joined #git
12:05 davimore_ joined #git
12:06 venmx joined #git
12:07 smurfendrek123 joined #git
12:08 pks Since when can `git rebase` without --merge strategy detect and correctly merge renamed files?
12:08 elect joined #git
12:12 explody joined #git
12:13 dviola joined #git
12:14 JrCs joined #git
12:14 marenz_ joined #git
12:14 ntonjeta joined #git
12:14 aw1 joined #git
12:16 GodGinrai joined #git
12:17 ntonjeta_ joined #git
12:21 nd joined #git
12:21 zefferno joined #git
12:25 timini I have written up my question on stack overflow - help would be greatly appreciated (i might put a bounty on it if anyone cares)
12:25 timini http://stackoverflow.com/questions/42531657/find-the-direct-ancestors-parents-of-a-git-commit
12:29 notebox joined #git
12:30 cerkauskas joined #git
12:31 cerkauskas hello. I have a file, that disappeared. I want to track down in which commit it was made. How can I do that?
12:32 jmd How do I pull updating all submodules?
12:32 SwiftMatt joined #git
12:33 ckruczek cerkauskas: http://bfy.tw/AN5r
12:36 jimi_ joined #git
12:42 clmsy joined #git
12:43 pur3eval joined #git
12:43 troyt joined #git
12:45 jameser joined #git
12:46 thebope joined #git
12:47 netj joined #git
12:52 DolpheenDream joined #git
12:56 mkalfon_ joined #git
12:57 tcorneli joined #git
12:57 TheWild timini: try git rev-list --all | git cat-file --batch
12:57 koneko joined #git
12:58 OnkelTem joined #git
12:58 masuberu joined #git
12:59 windy-black joined #git
13:00 TheWild or just git cat-file -p (hash)
13:00 flavius joined #git
13:00 gclaugus joined #git
13:01 gclaugus left #git
13:02 synthroid joined #git
13:03 cahoots joined #git
13:03 synthroi_ joined #git
13:05 ankit01ojha joined #git
13:05 smurfendrek123 joined #git
13:06 gugah joined #git
13:06 Raging_Hog joined #git
13:07 MineCoins joined #git
13:07 jottr joined #git
13:09 jameser joined #git
13:10 govg joined #git
13:10 m4sk1n joined #git
13:10 boombatower joined #git
13:12 jnavila joined #git
13:13 fahadash joined #git
13:13 hiwk joined #git
13:13 hiwk can I get git merge to add conflict files to my log message?
13:13 hiwk some configure option?
13:14 bremner they are usually added by git commit
13:15 hiwk bremner: in my case they seem to be prepended by #?
13:15 bremner sure. So remove the #?
13:16 hiwk can I make it do that automatically? simlar to what the merge.log config option does
13:16 bremner I don't know about that.
13:17 hiwk ok
13:17 hiwk thanks anyways
13:18 justanotheruser joined #git
13:18 NimeshNeema joined #git
13:19 brent__ joined #git
13:21 bashcoder joined #git
13:23 PCatinean joined #git
13:24 explody joined #git
13:27 jottr_ joined #git
13:27 hurricanehrndz joined #git
13:28 windy-black joined #git
13:29 serialoverflow joined #git
13:31 leeN joined #git
13:32 dermoth_ joined #git
13:33 muhannad_______ joined #git
13:33 hiwk hm, it seems like recording the conflicts in the log message varies with git client version :/
13:33 gituser123 joined #git
13:34 bremner uh. what is "git client"?
13:35 shinnya joined #git
13:36 qqx hiwk: Yes, it used to be that conflicts were put into the commit template without the comment markers. That was changed at some point, since most of the developers view that as useless information to include in the history.
13:36 hiwk bremner: sorry, git isn't the client/server model, right. I meant the "git" program
13:36 bremner hiwk: ah, ok. I thought you might refer to some third-party front end.
13:37 hiwk qqx: ok, thanks!
13:37 Gustavo6046 joined #git
13:37 qqx I found the `git client` reference to be completely understandable. And can make sense to differentiate between what's running on the local system vs. what's on a remote.
13:39 King_Hual joined #git
13:39 dendazen joined #git
13:40 peacememories joined #git
13:40 thebope joined #git
13:41 scottjl joined #git
13:42 scottjl joined #git
13:42 scottjl joined #git
13:43 BlackMaria joined #git
13:45 MattMaker joined #git
13:46 publio joined #git
13:47 brent__ joined #git
13:47 scottjl joined #git
13:48 Gsham joined #git
13:49 livingstn joined #git
13:50 jeffreylevesque joined #git
13:53 tinanoo joined #git
13:55 mehola joined #git
13:56 JuliusN joined #git
13:56 utrack joined #git
13:56 timini thanks theWild
13:56 Zamyatin joined #git
13:57 JuliusN i'm trying to enter 'git merge --abort' but i get 'error: Entry 'Podfile.lock' not uptodate. Cannot merge.'
13:58 ronskisika joined #git
14:00 masuberu joined #git
14:00 thebope joined #git
14:00 oskarkv joined #git
14:02 xissburg joined #git
14:02 Oatmeal joined #git
14:04 JuliusN how do i abort merge?
14:04 aw1 joined #git
14:05 selckin think its already aborted?
14:05 selckin ie merge was not done because work dir is not clean
14:05 JuliusN got it to work, not exactly sure why
14:05 davimore joined #git
14:06 rwb joined #git
14:06 JuliusN if fast-forward fails, isn't the repository left in a merging state?
14:07 grawity no
14:07 _ikke_ ff is not a real merge
14:07 grawity if fast-forward fails (and wasn't explicitly requested), an actual merge is performed
14:07 grawity if the *actual* merge fails, you're left in a merging state
14:09 paul424 joined #git
14:10 Tobbi joined #git
14:10 timini hey how can i get the history of a file?
14:10 _ikke_ timini: By rewriting every commit that touches the file
14:10 _ikke_ timini: git doesn't store history per file
14:11 ^7heo _ikke_: rewriting?!
14:11 ^7heo _ikke_: to GET a history?
14:11 _ikke_ ^7heo: d'oh, need to read better
14:11 ^7heo :P
14:11 _ikke_ I read 'get rid of'
14:11 Knirch :D
14:11 ^7heo you wish
14:11 ^7heo :D
14:12 _ikke_ timini: git log --all <path/to/file>
14:12 Murii_ joined #git
14:12 Levex joined #git
14:13 jottr_ joined #git
14:13 Knirch --all is quite peculiar, is that used to see across branches etc?
14:13 _ikke_ Knirch: Yes, to see it across all branches and tags (and remote tracking branches)
14:14 _ikke_ otherwise, it would only show commits that touch the file reachable from HEAD
14:14 Knirch oh, and reflog, picked a file that's changed in a branch I've rebased a couple of times :D
14:15 Knirch ^7heo: --follow can be nice if you've moved the file about
14:15 ^7heo I guess.
14:15 Knirch oh, wrong person
14:15 Knirch timini: --follow can be nice if you've moved the file about
14:16 ^7heo Knirch: you should drink less Knirch.
14:16 * ^7heo hides
14:17 jottr joined #git
14:18 Knirch you wish
14:18 Knirch :D
14:18 timini ok what about the history of a file for one branch
14:18 visual joined #git
14:18 visual hey
14:18 Knirch timini: git log --follow file
14:19 timini i think that shows all branches non?
14:19 al-damiri joined #git
14:20 visual http://prntscr.com/eeofko
14:20 visual Why is git so retarded?
14:20 gitiser12 joined #git
14:21 bvcosta joined #git
14:21 Learath2 joined #git
14:21 Knirch visual: ... origin is the remote name
14:21 visual Knirch: oh, thank you
14:22 Knirch visual: like it says in the top example, git remote rm destination, as you can see on the green lines destination is to the far left, the name of the remote.
14:23 vmffkn joined #git
14:23 visual dang it, missed that one
14:23 visual thought it was just a repository
14:23 visual thanks Knirch
14:24 suttonwilliamd joined #git
14:25 rubyonrailed joined #git
14:27 ntonjeta joined #git
14:27 newbie1 joined #git
14:28 finalbeta joined #git
14:29 cr34ton joined #git
14:29 Nilesh_ joined #git
14:29 GodGinrai joined #git
14:30 zxd joined #git
14:32 pur3eval joined #git
14:33 ismithers joined #git
14:36 ExoUNX joined #git
14:36 thebope joined #git
14:37 eblip joined #git
14:37 borkr joined #git
14:38 marcogmonteiro joined #git
14:38 chitopunk joined #git
14:39 bauruine joined #git
14:39 staplezebra joined #git
14:40 red82 joined #git
14:40 jstimm joined #git
14:42 peacememories joined #git
14:42 Levex joined #git
14:42 geoid joined #git
14:43 finalbeta joined #git
14:44 doppo joined #git
14:44 amrits joined #git
14:45 matumbo joined #git
14:46 cyphase joined #git
14:47 AaronMT joined #git
14:47 sgen joined #git
14:47 msrkmo joined #git
14:49 flavius joined #git
14:51 Rodya_ joined #git
14:53 le_melomane joined #git
14:54 cahoots joined #git
14:55 Es0teric joined #git
14:58 Vinnie_win joined #git
15:00 doppo joined #git
15:00 kpease joined #git
15:00 aw1 joined #git
15:00 e14 joined #git
15:01 aw1 joined #git
15:02 sbulage joined #git
15:02 Kulrak joined #git
15:02 Kulrak joined #git
15:02 synthroid joined #git
15:02 aw1 joined #git
15:04 GodGinrai joined #git
15:04 Alienpruts joined #git
15:04 chigang joined #git
15:05 aw1 joined #git
15:06 nickabbey joined #git
15:07 LeBlaaanc joined #git
15:15 Levex joined #git
15:15 uranellus joined #git
15:15 uranellus joined #git
15:17 le_melomane joined #git
15:19 flopsec joined #git
15:21 le_melomane joined #git
15:22 DaYwalKer joined #git
15:23 aw1 joined #git
15:24 bambams I appear to be in need of some kind of case-insensitive branch names fix. Colleagues push various case paths to TFS, and when I fetch it screws up my remote tracking branches.
15:24 bambams E.g., I can push and that works fine, but if I then fetch --all it shows the remote branch one commit behind again.
15:25 bambams My best guess is that it's the varying case in ref names confusing it... Since I'm also running Windows and file-system is case-insensitive.
15:26 Levex joined #git
15:27 sweatsuit joined #git
15:27 suttonwilliamd joined #git
15:28 aielima joined #git
15:29 cdg joined #git
15:30 einnjo joined #git
15:31 cdg joined #git
15:32 synthroid joined #git
15:34 emin joined #git
15:34 bashcoder joined #git
15:34 Kulrak joined #git
15:34 Kulrak joined #git
15:36 bambams E.g., foo/bar/baz versus foo/BAR/baz.
15:36 Levex joined #git
15:36 roent joined #git
15:37 n1ce joined #git
15:37 bambams I wonder if it would be possible for the local Git to force remote branch names to lower case automatically... Would cause problems if there were actually two different branches with case-insensitive equality, but that's probably unlikely. Maybe Git could recognize it too and skip one of the conflicts...
15:37 ChadTaljaardt joined #git
15:39 jottr joined #git
15:43 gugah joined #git
15:44 finalbeta joined #git
15:45 Darren_ joined #git
15:45 ayogi joined #git
15:45 scottjl joined #git
15:47 Raging_Hog joined #git
15:47 thiago joined #git
15:49 a_thakur joined #git
15:51 notebox joined #git
15:51 lohfu joined #git
15:52 Rodya_ joined #git
15:52 kritzikratzi joined #git
15:52 Gsham joined #git
15:53 bambams Seems `git config core.ignoreCase true' helps, but doesn't completely eliminate the issues.
15:53 malt3 joined #git
15:54 User458764 joined #git
15:55 DarkPsydeLord joined #git
15:57 dviola joined #git
15:58 bashcoder joined #git
15:59 jstimm joined #git
15:59 Vivekananda joined #git
16:01 xall joined #git
16:02 cdg joined #git
16:02 jost_ joined #git
16:05 Noldorin joined #git
16:05 Bizkit joined #git
16:05 chachasmooth joined #git
16:06 dorkmafia joined #git
16:06 buro joined #git
16:07 _Vi joined #git
16:10 qqx joined #git
16:12 tang^ joined #git
16:12 crose joined #git
16:13 nckpz joined #git
16:14 mda1 joined #git
16:18 Es0teric joined #git
16:18 gtristan joined #git
16:19 grift joined #git
16:19 hasc joined #git
16:19 nickabbey joined #git
16:20 g5pw joined #git
16:21 pur3eval joined #git
16:21 Muzer joined #git
16:21 Jackneill joined #git
16:23 jeffreylevesque_ joined #git
16:23 muhannad_______ joined #git
16:26 ntonjeta joined #git
16:27 s1scha joined #git
16:28 marianina8 joined #git
16:30 einnjo joined #git
16:30 brokensyntax joined #git
16:30 synthroid joined #git
16:34 dipapa joined #git
16:39 tchan joined #git
16:39 tchan joined #git
16:39 raspado joined #git
16:39 Guma joined #git
16:39 nettoweb joined #git
16:40 xissburg joined #git
16:44 ejb joined #git
16:45 Zamyatin joined #git
16:47 raynold joined #git
16:48 tekniq joined #git
16:48 e14 joined #git
16:49 username1 joined #git
16:49 NeXTSUN joined #git
16:50 pawelbx joined #git
16:54 thiago joined #git
16:54 eroux joined #git
16:55 texinwien_ joined #git
16:58 Tobbi joined #git
17:00 akkad joined #git
17:02 raynold ahh it's a wonderful day
17:02 pawelbx hi, i have two branches m and p. When i merged m into p, git would fast-forward, but then I needed to cherry pick a commit from m into p and since then git is no longer doing a fast-forward. Is there a way I can get back into a state that merging m into p does a fast-forward?
17:03 muhannad_______ joined #git
17:05 * bambams conceives of a script to pack all refs, lower-case them, and push the fixes out to remote...
17:10 nickabbey joined #git
17:11 mikecmpbll joined #git
17:12 xissburg_ joined #git
17:12 multi_io joined #git
17:12 tchan joined #git
17:12 bambams pawelbx: For a fast-forward to be possible, p needs to be an ancestor of m. That is, the same commit that p points to already exists on m. That way, no history is lost by fast-forwarding. Since you cherry-picked, p has an additional commit that m does not have (since the cherry-pick made a *copy* of the commit).
17:13 guardian hello, I just looked at the documentation for .gitattributes in hope to find a way to control filemode per extension. Does anybody know why Git doesn't consider it makes sense to do so?
17:13 bambams pawelbx: If you merged p into m that would allow you to fast-forward again, but that may be semantically wrong or undesirable (though likely it'll make no changes).
17:15 Noldorin joined #git
17:15 nickabbey joined #git
17:15 King_Hual` joined #git
17:18 Rodya_ joined #git
17:19 orbyt_ joined #git
17:19 thiago joined #git
17:19 ayogi joined #git
17:20 madewokherd joined #git
17:24 mmattice joined #git
17:25 nowhereman joined #git
17:25 durham joined #git
17:26 jstimm joined #git
17:26 bashcoder joined #git
17:27 Electrometro joined #git
17:28 kpease joined #git
17:29 muhannad_______ joined #git
17:29 ivyb joined #git
17:30 xall joined #git
17:30 orbyt_ joined #git
17:31 jimi_ joined #git
17:33 svm_invictvs joined #git
17:37 tyreld joined #git
17:38 kpease_ joined #git
17:41 thebope joined #git
17:42 benbrown_ joined #git
17:42 theparadoxer02 joined #git
17:43 theparadoxer02 hey everyone , i am not able to push changes, it shows everythine up to date, even i have made so many changes
17:43 _ikke_ theparadoxer02: Then you are not pushing the correct branch
17:44 PioneerAxon joined #git
17:46 theparadoxer02 _ikke_, i am pushing to the same branch !
17:46 hahuang65 joined #git
17:47 _ikke_ theparadoxer02: How are you pushing?
17:48 theparadoxer02 _ikke_, "git push origin master"
17:48 benbrown_ Hi, would converting a bare repository to non-bare by moving the directory to non-bare/.git, then setting core.bare for false and running a `git reset --hard` be robust or sane in any sense?
17:49 _ikke_ benbrown_: There are some more settings you want to change, but that approach should work
17:50 Aboba joined #git
17:50 kpease joined #git
17:51 xissburg joined #git
17:51 zeroed joined #git
17:51 akkad joined #git
17:52 benbrown_ _ikke_: cool
17:53 chardan joined #git
17:53 pdgwien joined #git
17:53 ertesx joined #git
17:54 bket joined #git
17:54 benlieb joined #git
17:54 benlieb I’m confused about remote branchses
17:55 benlieb I meant to make a new branch, but somehow I ended up on a branch that was “remote”, of the same name, one I made a long time ago
17:55 StuckMojo joined #git
17:55 benlieb when I did git branch, I only saw master
17:55 ash_workz joined #git
17:55 stevenspasbo joined #git
17:55 benlieb of course my app went into a crazy state since the branch was old, and took me a while to figure out. What is the recommended way to avoid this confusion. How can I clean up remote branches?
17:56 nickabbey joined #git
17:56 ribasushi joined #git
17:56 mehola joined #git
17:56 Sound joined #git
17:56 StuckMojo hi. so on my work laptop, when i "git checkout -b blah" i can subsequently do "git push origin" without issue, but on my desktop at home it always complains that the current branch has no upstream branch. i prefer the way my laptop is configured. any hints on what setting controls this?
17:58 thebope joined #git
17:58 _ikke_ theparadoxer02: Just to verify, why do you think whatever is in master is not yet pushed?
17:58 corentin StuckMojo: probably push.default
17:59 corentin can be set to simple or matching
17:59 corentin StuckMojo: guess you have simple on one computer and matching on the other
17:59 [Brain] joined #git
18:00 corentin to show the value: git config push.default
18:01 StuckMojo ah yeah, that sounds right
18:01 StuckMojo i.e. rings a bell. thanks!
18:01 msonntag joined #git
18:01 Electrometro joined #git
18:02 jason237 joined #git
18:02 StuckMojo desktop is "simple", laptop is "current"
18:02 bipolar joined #git
18:03 jstimm joined #git
18:03 StuckMojo good call. thanks corentin !
18:03 StuckMojo ciao
18:03 StuckMojo left #git
18:03 corentin you're welcome
18:03 akkad joined #git
18:03 StuckMojo joined #git
18:03 StuckMojo ah crap. one more question:
18:04 Dazzyp joined #git
18:04 StuckMojo so i've got two directories on my desktop, on for my work repos and one for personal. just realized as i edited ~/.gitconfig that since i've been working from home this week, my work commits have had the wrong email associated with them :(
18:05 zeroed joined #git
18:05 zeroed joined #git
18:05 StuckMojo i realize i could override the user.email in each repo, but is there some easier way?
18:05 StuckMojo s/each/each individual/
18:05 masuberu joined #git
18:06 corentin StuckMojo: you can set the user.email system wide
18:06 einnjo joined #git
18:06 pdgwien joined #git
18:06 StuckMojo yeah but then my personal repo commits would have the wrong email...although i've not made many in a while so...
18:06 corentin but on two different computer it would not work, and you probably cannot do it
18:07 corentin StuckMojo: so you just want to fix the commits to have the proper email?
18:07 StuckMojo nah i'm not worried about retroactively fixing the incorrect ones
18:07 StuckMojo i just want it to be correct going forward
18:07 StuckMojo so i think that i have two choices:
18:08 StuckMojo either set it globally, and have the wrong email sometimes, or override it per-repo
18:08 StuckMojo which...would be easier on the personal ones because i've got a lot less of those
18:09 bgamari joined #git
18:09 pur3eval joined #git
18:09 HardlySeen joined #git
18:10 corentin StuckMojo: git config --global user.email "blah@mail.com"
18:10 corentin this will configure your email for all repos on the computer where this command is executed
18:10 bgamari Is rev-parse, rev-list, or something similar able to give me a list of SHAs if a range is specified (e.g. A..C) or just a single SHA if a single ref is given (e.g. master)?
18:10 corentin but you'll have to do the same on every computer you use to commit
18:10 ozmage joined #git
18:11 StuckMojo right but that will effect everything. i have my repos divided into to directories, think ~/work_git ~/personal_git and i want different emails to be used for commits in repos in each of those dirs
18:11 dreiss joined #git
18:11 StuckMojo corentin: sorry, i think i was unclear
18:11 corentin StuckMojo: hah, I don't know a solution for that, but I would be interested
18:11 StuckMojo corentin: i'm ass-u-me-ing that i can put [user] \n email = foo@foo.com inside a given repo/.git/config file to override it
18:12 corentin of course you can
18:12 StuckMojo so i'll just do that in the personal repos, and set the global to the work email since there's a shit-ton of work repos
18:12 corentin sounds good
18:12 StuckMojo kk. thanks again
18:12 corentin StuckMojo: although you might be ending commiting on your personal projects with your work email
18:13 corentin ending up
18:13 staplezebra joined #git
18:13 StuckMojo yeah. i only have like 3 personal repos where i need to override.
18:14 StuckMojo plus it's probably better to accidentally use my work email in my personal repo than to  use my personal in my work commits
18:15 StuckMojo oh so... i can correct the email on my most recent un-pushed work commit with --ammend right?
18:15 StuckMojo or i suppose since i've not pushed i could just reset back one and re-commit
18:17 Darren_ joined #git
18:17 StuckMojo meh. had to go the reset route. amend doesn't change the author
18:18 kasual joined #git
18:19 phillco joined #git
18:20 i7c git commit --amend --reset-author
18:21 muhannad_______ joined #git
18:21 StuckMojo ah
18:22 StuckMojo well anyway, it's fixed. good to have that in my toolbox though, thanks i7c
18:22 StuckMojo l8rz
18:22 StuckMojo left #git
18:23 marc_v92 joined #git
18:23 m4sk1n joined #git
18:23 e14 joined #git
18:24 matoro joined #git
18:25 theparadoxer02 i have made ssh connection to my git hub
18:25 theparadoxer02 how do i push commit so that it does not ask for password everytime ?
18:26 GodGinrai theparadoxer02: use ssh-agent
18:26 GodGinrai theparadoxer02: `ssh-add ~/.ssh/keyfile` *enter password*
18:27 theparadoxer02 GodGinrai, i have done that
18:27 MattMaker joined #git
18:27 cyphase joined #git
18:27 User458764 joined #git
18:27 theparadoxer02 still when i hit "git push origin master", its asking me for password !
18:28 _ikke_ theparadoxer02: if it asks for a password, it means it's not using your key
18:28 theparadoxer02 _ikke_, GodGinrai shold i have to again clone my repo ?
18:28 GodGinrai theparadoxer02: https://wiki.archlinux.org/index.php/SSH_keys#ssh-agent
18:29 _ikke_ theparadoxer02: No
18:29 _ikke_ theparadoxer02: what does git remote -v show
18:29 _ikke_ ?
18:29 GodGinrai ssh-agent probably is not set up correctly
18:29 GodGinrai read the wiki page I linked, it has a couple of setup steps you can try
18:30 theparadoxer02 _ikke_, its showing my remote links!
18:30 theparadoxer02 GodGinrai, i have cloned my repo earlier
18:30 GodGinrai theparadoxer02: this has *nothing* to do with when you cloned
18:31 _ikke_ theparadoxer02: What is it showing, is it aan http url, or an ssh connection
18:31 jagob joined #git
18:32 goTAN joined #git
18:33 theparadoxer02 _ikke_, its showing the http url
18:33 DaYwalKer joined #git
18:33 nextloop joined #git
18:34 nextloop hello! is there some status on resumable clones?
18:34 _ikke_ nextloop: Nothing new
18:34 Es0teric joined #git
18:34 nextloop is there something blocking this feature _ikke_?
18:35 nextloop or just lack of interest and manpower?
18:35 _ikke_ nextloop: Yes, a whole lot of development
18:35 bremner most people find it easier to fix their network ;)
18:35 Eugene Or not have a gigabyte-scale repo
18:37 d0nn1e joined #git
18:38 _ikke_ nextloop: this is one attempt (no idea if it was accepted): https://public-inbox.org/git/1473984742-12516-1-git-send-email-kevin.m.wern@gmail.com/
18:38 nextloop don't really know git internals, but incremental shallow clones could essentially allow that?
18:38 theparadoxer02 GodGinrai, _ikke_ should i use ssh link in remote ?
18:39 _ikke_ theparadoxer02: Probably, yes
18:39 User458764 joined #git
18:40 nextloop Eugene: repo is several hundred megabytes, bremner unfortunately not possible. it's not a very direct connection
18:41 allenr joined #git
18:41 bremner sure. But it's also not very practical for you to fix git, presumably.
18:42 nextloop yeah. we still use monotone for that reason
18:43 _ikke_ nextloop: Another thread: https://public-inbox.org/git/CANtyZjSJf5_xbsBC5tUaazFT3uiEgJnx2_kHLwYwKcn50Jy_qg@mail.gmail.com/
18:44 GodGinrai nextloop: "several hundred megabytes" is a problem?  Our company has repos that are multiple gigs... <.<
18:44 rubyonrailed joined #git
18:44 _ikke_ GodGinrai: Depending on you network connection, that may be a problem
18:44 Eugene I'm amused at Microsof's approach for the "several hundred gigabyte" Windows repo: just write a filesystem driver, because THAT makes sense.
18:46 GodGinrai _ikke_: sure, but there are plenty of ways around it. For example, you could ssh to a server that does not have said net problems, clone there, and then rsync, scp, or use any other utility that works well for your shitty net connection from that box.
18:46 Mobutils joined #git
18:46 Rodya_ joined #git
18:47 Sample joined #git
18:48 ahr3n joined #git
18:48 Mobutils left #git
18:48 nextloop GodGinrai: the only solution around it for us would be a monthly torrent archive of the repo
18:48 _ikke_ you can create a git bundle
18:50 thebope joined #git
18:52 Spacew00t joined #git
18:53 nextloop but obviously not that convenient
18:53 _ikke_ It can bootstrap you
18:53 nextloop that's the idea behind it, yes
18:54 Groovin joined #git
18:54 nextloop ...or work with shallow-clones
18:54 nextloop is there any drawback?
18:54 nextloop for "normal" development work?
18:54 Eugene !shallow
18:54 gitinfo A shallow clone will clone a repository to a certain "depth", which sounds like a good idea for only obtaining the latest revision of a given repo. However, they often don't work as-advertised(breaking on fetch and push operations). Instead you should consider git-archive's --remote feature, or perhaps you're really encountering a !binary problem
18:54 Eugene Shallow clones have improved since i wrote that factoid, but I still dont like them
18:55 matoro joined #git
18:56 bashcoder joined #git
18:57 thiago joined #git
18:58 nohop_ joined #git
18:59 phy1729 Why?
19:00 theparadoxer02 _ikke_, my branch was detached that why i was not able to push changes!
19:00 theparadoxer02 but when i checked the master branch i lost all the commits ?
19:00 theparadoxer02 where did they gone ?
19:01 cdg_ joined #git
19:02 Darcidride joined #git
19:03 Eugene !reattach
19:03 gitinfo Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue
19:03 Eugene !detached
19:03 gitinfo A detached HEAD (aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach
19:03 Munt joined #git
19:03 Eugene I think !fixup covers this
19:03 gitinfo So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!
19:03 WayToDoor joined #git
19:03 Groovin joined #git
19:03 Munt joined #git
19:04 mikecmpbll joined #git
19:04 pigeon joined #git
19:07 cany0n joined #git
19:07 stoopkid joined #git
19:07 marc_v92 joined #git
19:12 bashcode1 joined #git
19:13 Electrom_ joined #git
19:15 Sasazuka joined #git
19:15 snowalpaca joined #git
19:17 b1tchcakes joined #git
19:19 Synthead joined #git
19:20 joeco joined #git
19:21 vuoto joined #git
19:21 dreiss joined #git
19:25 manuelschneid3r joined #git
19:27 Noldorin joined #git
19:28 MattMaker joined #git
19:30 nohop_ left #git
19:30 nohop joined #git
19:30 nohop if a GET from "https://api.github.com/repos/AA/BB/tags" gives me:  "zipball_url": "https://api.github.com/repos/AA/BB/zipball/v0.42"
19:30 nohop how would i download that url it returns? When i'm using the same method as getting the tags, i will receive no data
19:31 GodGinrai nohop: this is the git channel, not github channel
19:31 daemon left #git
19:32 Sample joined #git
19:32 nohop fair enough. I didn't even know there was a specific github channel :) thanks
19:33 GodGinrai np
19:34 Guest19319 joined #git
19:38 bashcode1 joined #git
19:40 pur3eval joined #git
19:42 dreiss joined #git
19:42 einnjo joined #git
19:43 aax joined #git
19:45 robotroll joined #git
19:46 e14 joined #git
19:46 b100s joined #git
19:49 kiltzman joined #git
19:50 kiltzman joined #git
19:51 tekniq joined #git
19:51 kiltzman joined #git
19:51 kiltzman joined #git
19:52 kiltzman joined #git
19:52 kiltzman joined #git
19:54 aielima joined #git
19:55 kubuntu_ joined #git
19:55 kubuntu_ joined #git
19:56 kubuntu_ joined #git
19:56 kubuntu_ joined #git
19:56 kubuntu_ joined #git
19:56 sbasso joined #git
19:57 kubuntu_ joined #git
19:58 mehola joined #git
19:58 fstd_ joined #git
19:58 crose joined #git
19:58 le_melomane joined #git
20:00 noteness_ joined #git
20:02 bomb-on joined #git
20:02 ok91 joined #git
20:02 julius left #git
20:04 Sleepy63 joined #git
20:05 ahr3n joined #git
20:06 finalbeta joined #git
20:09 Sample_ joined #git
20:09 troulouliou_dev joined #git
20:10 flavius joined #git
20:13 Anja joined #git
20:13 Starky joined #git
20:13 cahoots joined #git
20:14 cahoots hi, i want to set a branch to point to a commit without checking that commit out. how can i do this?
20:14 m0viefreak joined #git
20:15 ejb joined #git
20:15 thiago joined #git
20:17 alezandro joined #git
20:17 Sample joined #git
20:17 aielima joined #git
20:17 Sample joined #git
20:20 ojacobson cahoots: git branch <name> <commit>
20:20 scottjl1 joined #git
20:20 ojacobson with --force if you want to blindly overwrite an existing branch
20:21 scottjl joined #git
20:23 cahoots ojacobson, thanks
20:24 tvw joined #git
20:27 bashcoder joined #git
20:27 cahoots ojacobson, now how would i push it without checking it out? when i try, i get fatal: 'me-string-backup' does not appear to be a git repository
20:27 cahoots fatal: Could not read from remote repository.
20:28 Electrometro joined #git
20:28 ejb joined #git
20:28 MattMaker joined #git
20:29 brent__ joined #git
20:31 BashCode1 joined #git
20:32 osse cahoots: git push remotename branchname
20:33 bashcode1 joined #git
20:35 e14 joined #git
20:36 Sasazuka__ joined #git
20:36 bashcode1 joined #git
20:38 tubaman joined #git
20:38 scottjl joined #git
20:38 MattMaker joined #git
20:39 GodGinrai joined #git
20:41 le_melomane joined #git
20:44 mozzarella joined #git
20:45 Vgr_ joined #git
20:46 aielima joined #git
20:47 lucido-cl joined #git
20:47 _Vi joined #git
20:48 g0th_ joined #git
20:48 reto_ joined #git
20:48 freedrul1 joined #git
20:48 joey_ joined #git
20:48 dzho_ joined #git
20:48 exussum_ joined #git
20:49 joey_ left #git
20:49 nikivi- joined #git
20:51 joey00 joined #git
20:51 MillerBoss joined #git
20:51 wilbert_ joined #git
20:52 Deamon_ joined #git
20:52 thunderb- joined #git
20:52 username_ joined #git
20:53 nickabbey joined #git
20:54 fletom joined #git
20:54 joey_ joined #git
20:54 orbyt_ !lol
20:54 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
20:55 ccolorado joined #git
20:55 ablackack joined #git
20:55 Munt joined #git
20:55 gopar joined #git
20:56 Tatou I need off svn
20:56 Tatou How well does this bridge work?
20:56 joey_ joined #git
20:58 joey_ Hi i need some help with merging can someone help me?
20:58 moritz joey_: depends on your problem :-)
20:58 madprops joined #git
20:58 joey_ I want to try to merge a branch to master. The problem is that a lot of files are in the changed file list and lots of files wil be changed wrong.
20:59 cmalmber joined #git
20:59 moritz why will they be changed wrongly?
20:59 joey_ Some new files wil be deleted in master (files are created in master after the branch was made) Some new files have conflicts on delete (files are created in master after branch was made) Some new files wil be changed (some lines wil be deleted so i guess some commits will be reverted) (files are created in master after the brach was made) Changed files in master wil be undone or partly undone after merge (I did not touch these files in t
20:59 musicmatze joined #git
20:59 umbSublime joined #git
21:00 arooni if i dont plan on commiting changes to a github repo (just use tagged releases); is there any problem in checking it out in detached head state?  should i use a branch instead?
21:00 cdg joined #git
21:01 jeffreylevesque joined #git
21:02 avar Branches are literally just a "when the head changes, let's echo the sha1 to a file" facility
21:02 avar so if you don't want to refer to a commit by a nice name, it doesn't matter
21:02 benlieb joined #git
21:03 arooni does anything change between those two different approaches when i switch to a different tagged release?
21:03 arooni i.e. when the project gets updated
21:04 joey_ I have no idee why it wants to do this but there are over 1000 changes and i am not planning in looking over all of them
21:04 pkiller joined #git
21:05 avar arooni: if you're not making commits, no
21:05 klusark joined #git
21:05 Blendify_ joined #git
21:05 avar branches and tags are just pointers to a sha1, you can just cehckout the sha1, or tagged release
21:05 Blendify_ when I run git svn fetch https://svn.blender.org/svnroot/bf-manual/trunk/blender_docs/
21:05 Blendify_ i get: [svn-remote "https://svn.blender.org/svnroot/bf-manual/trunk/blender_docs/"] unknown
21:05 sathed If I do a git push --mirror, that will overwrite commit history that doesn't exist in the repository that is being copied. Right?
21:05 Blendify_ do I have to add the svn remote somehow?
21:06 arooni avar: got it
21:06 avar Blendify_: you start with git svn clone
21:07 askb joined #git
21:07 joey_ Does someone have an idee?
21:07 invisbl joined #git
21:07 neilthereildeil joined #git
21:08 neilthereildeil hey guys
21:08 neilthereildeil i am trying to pull a remote change to my local repo
21:08 neilthereildeil both local and remote have 2 branches
21:09 neilthereildeil HEAD is at the tip of my local branch, and i do git pull REMOTE_URL
21:10 neilthereildeil but its giving me conflicts even though HEAD is currently at the 2nd latest commit of remote and the latest of local.
21:10 neilthereildeil how can i fix this?
21:10 neilthereildeil remote has f5870537152ad1668c97b6319dbcc8b8bacc5c0d followed by 15e215a4413ef9b5cc5e59ba5113a452dfd93e2a
21:10 neilthereildeil local HEAD is at f5870537152ad1668c97b6319dbcc8b8bacc5c0d
21:10 neilthereildeil and pull gives issues
21:11 chachasmooth_ joined #git
21:11 whaley joined #git
21:14 Random832 what are the conflicts? are you sure you don't have uncommitted changes?
21:15 smurfendrek123 joined #git
21:15 sfix joined #git
21:15 jnewt joined #git
21:17 neilthereildeil Random832: git status shows nothing modified although there are untracked files which i dont care about
21:17 scottjl joined #git
21:18 neilthereildeil and it saysy im ahead of remote by 3 commits
21:18 joey_ if someone wants to help me please look at http://stackoverflow.com/questions/42541231/merging-to-master-deletes-new-files There i have listed all
21:19 alezandro joined #git
21:19 Random832 how are you ahead of remote if head is f58 and remote has f58
21:19 neilthereildeil not sure...
21:19 neilthereildeil could that be the issue?
21:19 Random832 can you post output of "git log --decorate --oneline --graph HEAD remote/branchname"
21:19 crayon joined #git
21:20 bashcode1 joined #git
21:21 Starky joined #git
21:21 neilthereildeil Random832: ohh i see origin/version_local_disk 3 commits behind. is that a problem?
21:21 llamapixel push then pull
21:21 neilthereildeil i havent made any local commits ever. why would i push?
21:22 neilthereildeil i just use this tree for reading code
21:22 Random832 ...what does your log output look like, now i'm confused as to what's behind what
21:22 Starky joined #git
21:22 llamapixel aye
21:22 llamapixel hastbin your log
21:23 Random832 also what is the remote? maybe someone else has been doing some !rewriting
21:23 gitinfo [!rewrite] 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
21:24 neilthereildeil Random832: even if they rewrote, the commits still match
21:24 neilthereildeil http://pastebin.com/RkKQXuAd
21:24 bashcode1 joined #git
21:24 dsdeiz joined #git
21:25 Random832 ok that says your local is ahead of origin
21:25 neilthereildeil yea exactly
21:25 neilthereildeil so my local is ahead of remote
21:25 neilthereildeil how does that happen and how can i fix it?
21:25 Random832 ok i'm a bit confused since i don't understand why you're not expecting to have conflicts
21:26 Random832 or what you mean by not having any local commits ever
21:26 neilthereildeil why would i expect to have conflicts?
21:26 Random832 ...because you have local commits
21:26 neilthereildeil my local history is a subset of remote history
21:26 llamapixel Sometimes when you open a project manager it will change files locally even though you have not added new media or code.
21:26 Random832 huh?
21:26 llamapixel IS there a project manager in the mix?
21:26 Random832 your local history has three commits that are not part of the remote history, how is that a subset
21:27 * llamapixel slows down and waits Random832 to ask.
21:27 neilthereildeil locally i have A->B->C, remote has A->B->C->D. im just trynna pull D
21:27 Random832 ...huh?
21:27 Random832 the log establishes that remote has A only.
21:28 Random832 unless you've deleted stuff out of the log before pasting it (why?)
21:29 Random832 ok just paste the first few dozen lines or so of git log --all --decorate --oneline --graph
21:29 pur3eval joined #git
21:29 neilthereildeil locally, the latest commit i see is f5870537152ad1668c97b6319dbcc8b8bacc5c0d. sshed into remote, the latest 2 that i see are f5870537152ad1668c97b6319dbcc8b8bacc5c0d followed by 15e215a4413ef9b5cc5e59ba5113a452dfd93e2a
21:30 neilthereildeil so remote has 1 newer commit than local, and im trynna pukll it
21:30 neilthereildeil pull*
21:30 NeXTSUN joined #git
21:32 joey_ Is there anyone that can help me with my merging problem?
21:33 bket joined #git
21:34 exo_ joined #git
21:34 llamapixel what does git status say on your local side neilthereildeil ?
21:34 kbs neilthereildeil: the available evidence indicates otherwise :) but to confirm maybe (a) start with "git fetch origin oldBranch" followed by (b) Random832 's advice of pastbin'ing log output - without this it's unclear what's going on
21:36 kbs joey_: It sounds like fundamentally master and your desired branch have diverged, where git needs your input to figure out what files fall into your four outlined buckets.
21:36 Random832 neilthereildeil, how exactly do you see f58 and 15e remotely?
21:36 neilthereildeil http://pastebin.com/skeJ4wPs
21:37 neilthereildeil thats the output of local repo
21:37 Random832 and that's after a fetch?
21:37 neilthereildeil no, i have notfetched. i just tried pull, which failed with conflicts
21:37 Random832 pull would have fetched first
21:38 joey_ so you want to say that i have to look over all the 1000 files and change them all according to solving this?
21:38 neilthereildeil on remote, i am viewing f58/15e using "git log --all --tags --decorate=full --graph
21:38 neilthereildeil "
21:38 Random832 is there some other branch involved other than "oldBranch"? is that log output really with --all ?
21:38 neilthereildeil yea ias i mentioned earlier, there is another branch
21:38 llamapixel aha
21:38 neilthereildeil im just interested in getting the new commit from this branch
21:38 harmlessdrudge joined #git
21:38 neilthereildeil the single latest commit
21:39 peterbecich joined #git
21:39 neilthereildeil my local is on same branch that i amd hoping to get newest commit of
21:39 neilthereildeil local HEAD
21:41 neilthereildeil should i be running git fetch?
21:42 synthroid joined #git
21:42 tunjin joined #git
21:43 Random832 ...if there's another branch why didn't it show up in the log output
21:43 Random832 i specifically asked for the log with --all so we could get the *whole* picture, it feels like you're playing games
21:43 neilthereildeil you asked for git log --decorate --oneline --graph HEAD
21:44 Random832 at no point did i ask for that. first i asked for git log --decorate --oneline --graph HEAD remote/branch, then I asked for git log --all --decorate --oneline --graph [no refnames]
21:44 neilthereildeil ohh shoot. sry i think i misread the second one thinking it was the same
21:45 Random832 what branch are the commits actually on in the remote, and what branch is checked out?
21:45 neilthereildeil both on oldBranch
21:46 ablackack joined #git
21:47 Atm0spher1c joined #git
21:47 Random832 then this makes no sense
21:48 Random832 are you sure origin is actually pointed to the repository that you're looking at?
21:48 kbs I suppose it's always possible the origin url... :-)
21:48 hahuang65 joined #git
21:49 joey_ what do you mean?
21:49 neilthereildeil thats --all on my local repo
21:49 neilthereildeil http://pastebin.com/mrd7ymD4
21:49 neilthereildeil i removed descriptions
21:50 kbs joey_: sorry, that was meant for neilthereildeil
21:50 alex__ joined #git
21:50 neilthereildeil does that --all output help?
21:51 joey_ no prob but is there a way i can merge them without having over 2000 changes? or is this going to be a long day looking at all the changes and seeing what needs to be done and what not?
21:51 e14 joined #git
21:51 Exorades_ joined #git
21:51 Limes joined #git
21:52 Whiskey Can git detect "git reset --soft @^" in post-receive hook?
21:52 Random832 well i need to go sorry i couldn't help you
21:52 neilthereildeil ah thanks for your help
21:53 neilthereildeil kbs: joey_: any ideas?
21:54 Gustavo6046 joined #git
21:54 kbs joey_: hard to say - if the knowledge about what changes should be in the final result can be expressed as one of the merge rules (e.g. if you see a conflict, just use "my" branch, or "master") you might look at git merge strategies (-s option)
21:54 satifant joined #git
21:54 kbs joey_: otherwise - if each change is case-by-case, hard to escape only you can decide what to do in each case
21:55 joey_ okay then i think that is gonne be a long hard day the thing that i find strange is that it will delete new files from master that never existed on the new branch
21:55 kbs neilthereildeil: just to be on the safe side, can you run git fetch origin oldBranch, and see if there's any change in the local git log output as above?
21:55 benbrown_ left #git
21:56 joey_ so i dont know why git thinks they need to be deleted
21:57 molloy joey_: have you tried merging master into branch 'c'?
21:57 molloy rather than vice versa
21:57 joey_ yes then the files that where deleted when merging from c -> master wont even be in the changelog
21:58 joey_ so those files wont be added to branch c
22:02 joey_ so if there is any other way someone can help i am verry happy. I am even willing to share my screen so someone can look together with me to find a solution
22:03 wilbert_ joined #git
22:06 Munt joined #git
22:06 molloy joey_: have you tried soft resetting so that all of the changes you want to keep are staged, then stashing your staged changes, checking out master, then popping the stashed changes?
22:07 blackwind_123 joined #git
22:07 Whiskey no one?
22:08 bvcosta joined #git
22:09 Vampire0 Whiskey, depens on what you do later. If you try to push the rewritten history, this should easily be detectable in a hook by checking whether the old sha is a parent of the new sha
22:10 Vampire0 Whiskey, except if the resetting only happens within non-pushed commits where this is daily usage business
22:10 Whiskey Vampire0: i try to detect when i remove stuff from the git, to remove the files it created from the commit
22:10 Whiskey thats it outside git ofc
22:10 diegoll joined #git
22:11 Vampire0 Whiskey, sorry, I didn't get that
22:11 joey_ no i have not tried that but how do i soft rest my changes?
22:12 Whiskey git commit "something", git push will push it, and also create some files in a commit in "post-receive" hook, if i reset the last changes i also need to delete the files that was created as it not valid anymore
22:12 Darren_ joined #git
22:12 molloy joey_: `git reset --soft HEAD~1`, if your commit consists of the changes you want
22:12 Whiskey Don´t ask what "files in a commit in" is it sounds strange
22:13 joey_ but how wil resetting help with merging? I don't have any open changes at this moment
22:13 dsteplight__ joined #git
22:14 Whiskey Vampire0: so basicly i just need to detect the reset, to make it possible to delete the files from the other hook thats not valid anymore
22:14 Vampire0 Whiskey, well, I don't know what files you create, but how is that case different from a revert that is pushed, meaning simply undoing the effects of the last commit with a new commit and pushing?
22:14 Whiskey Vampire0: files that is created in the post-receive, it has nothing to do whit git its just "stuff" outside git
22:15 Vampire0 Whiskey, without saying how this "stuff" is related to the pushed changes, you cannot get a helpful answer
22:15 Whiskey Vampire0: the files are not importent ether, what is if its possible to detect the reset command
22:16 g105b joined #git
22:16 neilthereildeil kbs: oh wow that worked!
22:16 smurfendrek123 joined #git
22:16 neilthereildeil kbs: what was i doing wrong?
22:16 Whiskey Vampire0: did i not tell that 100 times allready i make some zips in the hook thats all
22:17 kbs neilthereildeil: fetch doesn't actually change anything, so celebrations may be premature
22:17 neilthereildeil what does it do then?
22:17 neilthereildeil i see the latest commit in my local history
22:18 neilthereildeil however, oldBranch is not updated to point to new commit. oldbranch and HEAD both point to the latest local commit before i ran fetch
22:18 Vampire0 Whiskey, zips containing what? how are those zips related to the commit?
22:18 Vampire0 Whiskey, and btw. a) be polite b) realise that I already told you that you can and how
22:19 kbs neilthereildeil: it might be productive to rerun the git log as before in a pastebin, etc
22:19 Whiskey Vampire0: i allredy told you that two, its contain files in the last push, it make now files for evrey push, but if i do a reset and go back in the code i also need to delete the files as it not valid anymore
22:20 jenia joined #git
22:20 jenia hello
22:20 fracting joined #git
22:21 Vampire0 Whiskey, and what is wrong with simply making a new ZIP from the new push?
22:21 jenia is there a way to merge develop to a feature branch but keep all the files in dir X unchanged?
22:21 Whiskey Vampire0: why the hell do you ask questions you don´t need any answer for?
22:21 jenia in other words, I need to resolve dir X by keeping all the changes made in feature-branch
22:22 Whiskey Vampire0: every file is well not every but many are uniq, so its need to be deleted if its not valid
22:22 Whiskey hmm "uniqe"
22:22 Whiskey Vampire0: for an example do i save the files only that was changed in the current commit, whit the sha1 hash
22:23 Vampire0 jenia, make the merge with --no-commit, then revert the files in that directory before you finally do the commit manually maybe?
22:23 Whiskey Vampire0: git reset --soft @^ and then git push --force and last file is invalid
22:24 jenia Vampire0, thanks!!
22:24 Vampire0 jenia, yw
22:24 phos1 joined #git
22:24 Whiskey anyways, so its need to detect the reset so it can delete "last created file" but thats another problem and not handle by git at all i think
22:25 Vampire0 Whiskey, please stop highlighting me, I wont read your messages anymore. I'm not keen of helping ppl that are unpolite as you. For that my spare time is too precious, otherwise I would just upset.
22:25 phos1 I did a git diff and it’s a lot of changes. How can.I scroll to end?
22:25 Whiskey Vampire0: you ask 100 questions that don´t make any sense to me, i just asked if if it was possible to detect the reset and i get 1000 questions that don´t matter
22:26 Vampire0 phos1, if your pager is the default less, just press "G" to go to the end or "g" to go to the beginning
22:27 thebope joined #git
22:27 phos1 omg thank you, I’ve been scrolling forever!
22:27 Vampire0 phos1, yw
22:27 gitinfo phos1: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
22:27 Whiskey Vampire0: i don´t think that was fair, i did one easy question it was a stright yes or no and get 1000 questions back
22:27 homerovich joined #git
22:28 Vampire0 Whiskey, if you could evaluate which questions make sense, you wouldn't need help here. You cannot always see the sense behind the questions. Besides that, I already told you after your first question how to detect it, but you kept on questioning, so there must be more
22:28 bvcosta joined #git
22:29 Vampire0 Whiskey, you got a clear yes and even how to do it. So why did you keep on writing questions if you already had your simple straight answer?
22:29 Whiskey Not again
22:29 Vampire0 Whiskey, Not again what?
22:30 Whiskey Well i missed that part, i was more hope that it could show me like $command = "git reset..............."
22:30 Whiskey And no you right, but i just did know what wanted and know what i need to make it done
22:32 Tatou Hows is git/svn bridge?
22:33 rwb joined #git
22:34 Es0teric joined #git
22:34 Whiskey Vampire0: i still don´t get you, parent will always be the same, is just the new SHA thats created
22:34 Noodlewitt joined #git
22:35 Whiskey aaa yeh, is its a new commit its changed its place
22:36 Maxdamantus When fetching from gerrit with something like `git fetch gerrit refs/changes/3/2/1`, it only updates FETCH_HEAD and doesn't create a ref at remotes/gerrit/refs/changes/3/2/1 .. what is it that dictates whether a ref is stored under the the remote on the local repository?
22:37 Whiskey Only thing i don´t get it was sha to check whit, as the old one is removed and don´t exist anymore
22:37 Vampire0 Whiskey, for each pushed ref you get on stdin the old SHA, the new SHA and the ref name. If the old SHA is not an ancestor of the new SHA, there was some reset / rebase / filter-branch / whatever-history-rewriting done
22:38 borkr joined #git
22:38 Vampire0 Whiskey, that is not in the history of the ref anymore does not mean that the commit is not in the repository anymore instantly
22:38 Whiskey whatever that means, i just know git reset :D
22:38 jstimm joined #git
22:38 Whiskey damit i hate this, control over nothing, on really thin ice :D
22:38 Vampire0 sorry?
22:39 talkJS joined #git
22:39 Whiskey Vampire0: i know to littel about git to control it myself
22:39 Whiskey and i don´t like the feeling
22:39 YuGiOhJCJ joined #git
22:40 talkJS when i run git blame, i see the guy that does the merge there.  I need to find out where the original source entry came from/which branch and stuff.  Any idea?
22:40 Vampire0 Maxdamantus, what Git version are you using?
22:40 Whiskey Vampire0: anyways how do i get out the old SHA, i have the current allredy thats made i think
22:40 Vampire0 Whiskey, well, then learn more about git ;-)
22:41 Whiskey Vampire0: well thanks for that :D
22:41 Vampire0 Whiskey, get out of what?
22:41 Munt joined #git
22:41 Whiskey the SHA that was deleted
22:41 Vampire0 Whiskey, as I and the documentation told you, you don't need to figure it out, it is told the hook on stdin
22:42 Whiskey förälder
22:42 Whiskey 1b241f635c
22:42 Vampire0 Whiskey, if you don't know what stdin is, you should maybe start with learning some basic Bash scripting before you try to write some Git hooks
22:42 Whiskey ops
22:42 Whiskey <old-sha1> SP <new-sha1> <- that part
22:42 Vampire0 exactly
22:42 Whiskey I don´t know eactly what sdin
22:42 Whiskey but i would guss its something that software sends in
22:42 Whiskey variables that allredy exist :D
22:43 Whiskey or commands it sends in to a script to tell it what to do or work whit :D
22:43 phos1 How can i revert one file to what it is in a differnt branch?
22:43 Vampire0 Whiskey, well, if you run a program on the command line, everything you type is reaching the program on stdin
22:44 Vampire0 Whiskey, in the case of the hook git calls the hook and sends the specified information on stdin
22:44 Whiskey yeh exactly
22:44 Vampire0 Whiskey, to learn how to use that information, as I said, please learn bash scripting. This is not the right place for that. ;-)
22:45 Whiskey I don´t do my hooks in bash but :)
22:45 Vampire0 phos1, `git checkout different-branch -- path/to/file && git add path/to/file && git commit -m "revert file to different branch"`
22:45 Vampire0 Whiskey, then learn how to do it in whatever language you do
22:46 Whiskey Ummm my hook is on 99 lines i manage just fine :)
22:46 Maxdamantus Vampire0: 2.11.0
22:46 Whiskey Vampire0: how can i what do you say "test a commit" =
22:46 Whiskey whit the hook script
22:46 Maxdamantus Vampire0: I'm guessing the behaviour is independent of git version. I'm just wondering what it is.
22:47 Whiskey well "pretend" maybe a batter word for it, ooo emulate :D
22:47 Vampire0 Maxdamantus, hm, ok, I thought this is maybe !fetch4, but that is only pre-1.8.4
22:47 gitinfo Maxdamantus: [!fetchfour] [pre 1.8.4 only] We recommend against using 'git fetch/pull <remote> <refspec>' (i.e. with branch argument), because it doesn't update the <remote>/<branch> ref. The easy way to fetch things properly is to get everything: 'git fetch' or 'git pull' are sufficient if you have one remote; otherwise we recommend 'git fetch <remote>' (plus 'git merge <remote>/<branch>' if you wanted to pull/merge).
22:48 Maxdamantus If I fetch from a repository on my local filesystem, it will create a ref under remotes with that name, so it's obviosly not determined simply by the name.
22:48 venmx joined #git
22:48 dviola joined #git
22:48 Maxdamantus Okaf, yeah, not using that old version.
22:49 Maxdamantus s/f/y/
22:51 bashcoder joined #git
22:51 pur3eval joined #git
22:52 Maxdamantus It should be noted that simply doing `git fetch` doesn't get that ref at all.
22:52 phos1 left #git
22:52 mda1 joined #git
22:52 Maxdamantus All it gets are the main branches, not all changesets.
22:53 Whiskey Vampire0: i need to simulate it :D
22:54 Vampire0 Whiskey, you need to simulate what?
22:54 manuelschneid3r joined #git
22:54 Whiskey i need to simulate a push to test the script untill it works or can´t that be done whitout actully do it?
22:55 Whiskey well reset would be batter :D
22:55 Vampire0 Maxdamantus, well, that is configurable. You can configure the refspecs that should be fetched, so that all changesets are fetched if that is what you want
22:55 Vampire0 Whiskey, just call your script with the desired test input on stdin
22:56 Whiskey but i don´t know what the defualt is
22:56 Vampire0 echo "345634634 345634563456 refs/heads/master" | ./post-receive
22:56 Vampire0 Whiskey, what do you mean by "default"?
22:56 Whiskey i mean, what its actully inside sdin when i try to run it for real, to try it i need to know how it looks in "real cases"
22:57 Vampire0 Whiskey, of course you know, you are kidding me aren't you?
22:57 Whiskey but it seems like you have show that above
22:57 Vampire0 like I, the docs and yourself said already
22:57 Whiskey no i have never look, and i can´t see it ether how can i? its running in the background
22:57 Maxdamantus Ah, I see. It fetches things under refs/heads by default and creates refs for those locally.
22:57 Vampire0 it is "<old sha> <new sha> <ref name>"
22:58 talkJS how can i search git history for a piece of code and have all commits shown with it?
22:58 Vampire0 Whiskey, so just take some sha as old and its parent as new and give it to your script
22:58 Vampire0 Maxdamantus, exactly
22:58 Whiskey yeh right
22:59 Whiskey i don´t get this
23:00 tvw joined #git
23:00 Vampire0 Whiskey, you don't get what?
23:01 Whiskey its complicated, is sha always full?
23:01 fmeerkoetter joined #git
23:01 kyan joined #git
23:04 Vampire0 Whiskey, what is complicated?
23:04 Vampire0 Whiskey, just compare `git merge-base <old> <new>` with <old> and <new>
23:04 matoro joined #git
23:05 Vampire0 Whiskey, if it is equal to <old>, it was a fast-forward push, if it is equal to <new>, it was some history rewriting like a reset
23:06 Whiskey lets see if i can get something
23:07 Gsham joined #git
23:08 xissburg joined #git
23:08 cr34ton joined #git
23:08 Whiskey is the parent sha the current when the hook is run?
23:11 Vampire0 Whiskey, is the parent of what the current of what?
23:11 cliluw joined #git
23:11 Whiskey Vampire0: what does git merge base do?
23:11 Whiskey Im not sure its what im after
23:13 jimi_ joined #git
23:13 rubyonrailed joined #git
23:13 Whiskey okay lets see i think i come somewhere
23:14 xissburg_ joined #git
23:17 jeffreylevesque joined #git
23:18 venmx joined #git
23:19 ejb joined #git
23:20 Vampire0 Whiskey, it gives you the nearest common ancestor of the two commits
23:21 Vampire0 Whiskey, if the nearest common ancestor is the old one, the new one was a fast-forward push, if not (equal to the new one, or even none), then it was a reset or something else
23:21 thebope joined #git
23:22 Whiskey i miss something here
23:22 peeps[lappy] joined #git
23:23 dirtyroshi joined #git
23:23 Whiskey if [lindex [exec git log --format=%H -n 2] 1] != [lindex $argv 0] i think thats correct
23:24 Whiskey it checks if parent commit is same as the old
23:24 xissburg joined #git
23:24 Whiskey it should be true if the new commit is made on that one, if its removed it won´t exist and fail
23:26 anuxivm joined #git
23:27 duderonomy joined #git
23:29 ntonjeta joined #git
23:29 Vampire0 Whiskey, I don't fully understand your language, but I don't think you are right. 1. are you sure it is exactly the parent that is pushed, not reset to parent, some other commit done and then pushed, or reset to grand-parent and then pushed?
23:30 Vampire0 Whiskey, besides that, "$argv" sounds like command parameters. Those will always be empty for the hook.
23:30 Whiskey well thats another problem
23:30 Vampire0 Whiskey, and I also told you to use merge-base for a reliable recognition
23:31 Whiskey argv shows me what i do in ./hook this is the options
23:31 Vampire0 Whiskey, if you don't take ANY of my advice, why are you asking at all?
23:31 Vampire0 Whiskey, now I'm really not reading your messages any longer, as you ignore my answers anyway
23:32 Whiskey Vampire0: no why would i ?
23:35 dsdeiz joined #git
23:35 dsdeiz joined #git
23:36 peterbecich joined #git
23:36 Whiskey I actully things that right code
23:36 jmckind joined #git
23:37 Arokh joined #git
23:37 Whiskey argv is not a file?
23:38 DonVitoCorleone joined #git
23:39 neilthereildeil left #git
23:46 nydel07 joined #git
23:49 jstimm joined #git
23:50 ejb joined #git
23:50 dreiss joined #git
23:52 pur3eval joined #git
23:53 kiltzman joined #git
23:54 kiltzman joined #git
23:54 kiltzman joined #git
23:55 ahr3n joined #git
23:55 nickabbey joined #git
23:56 kiltzman joined #git
23:56 elisa87 joined #git
23:57 elisa87 can you please help me solve this issue? http://pastebin.com/0wYaUT26
23:57 kiltzman joined #git
23:59 mehola joined #git

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