Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-09-22

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

All times shown according to UTC.

Time Nick Message
00:00 snthsnth joined #git
00:02 ochorocho__ joined #git
00:03 faux___ joined #git
00:03 e14 joined #git
00:05 BackEndCoder joined #git
00:07 ochorocho__ joined #git
00:09 meejah_ joined #git
00:09 fuchstronaut joined #git
00:11 arand joined #git
00:12 harish_ joined #git
00:14 mizu_no_oto joined #git
00:16 e14 joined #git
00:17 buglessdr joined #git
00:17 gumnos joined #git
00:19 gumnos Running git on a Debian-derivative, how does git determine certificate authorities for HTTPS? Should it default to using certs provided by the "ca-certificates" package? (I know there's also http.sslCAInfo in the config to modify the CA listing)
00:20 _sjs joined #git
00:21 ochorocho__ joined #git
00:28 e14 joined #git
00:29 BackEndCoder joined #git
00:29 prawnzy joined #git
00:37 mycyber joined #git
00:38 dendazen joined #git
00:41 ochorocho__ joined #git
00:41 BenderRodriguez joined #git
00:48 autoboxer joined #git
00:50 ResidentBiscuit joined #git
00:51 theskillwithin joined #git
00:51 Milijus joined #git
00:53 ochorocho__ joined #git
00:54 sonikspin joined #git
00:56 netj joined #git
00:56 RedNifre_ joined #git
00:59 mizu_no_oto joined #git
00:59 tesuji joined #git
01:01 netj joined #git
01:03 morenoh149 joined #git
01:07 thiago_ joined #git
01:07 RickLee joined #git
01:08 RickLee joined #git
01:08 cqi joined #git
01:08 RickLee joined #git
01:09 kus_ubuntui686 joined #git
01:09 YottaByte joined #git
01:09 RickLee joined #git
01:10 fuchstronaut joined #git
01:10 RickLee joined #git
01:10 kus_ubuntui686 joined #git
01:10 arescorpio joined #git
01:11 RickLee joined #git
01:11 ochorocho__ joined #git
01:12 Shentino joined #git
01:16 snthsnth joined #git
01:16 jaafar joined #git
01:17 actinide joined #git
01:22 Vampire0_ joined #git
01:24 Pyrogerg joined #git
01:25 sonikspin joined #git
01:27 HoierM joined #git
01:29 dreiss joined #git
01:30 d^sh joined #git
01:32 Gsham joined #git
01:33 Milijus joined #git
01:36 _ADN_ joined #git
01:39 sam_yan joined #git
01:41 sam_yan joined #git
01:46 sam_yan how to create a  patch from  the changes in the current working directory
01:46 Atm0spher1c joined #git
01:46 ochorocho__ joined #git
01:46 mingrammer joined #git
01:48 ilbot3 joined #git
01:48 Topic for #git is now Welcome to #git, the place for git help and aquarium keeping | Please take the git survey: https://jk.gs/gsurv | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.10.0 | Getting "cannot send to channel"? /msg gitinfo .voice | Keep good care of your git-shell and your inner snail will thank you
01:50 jamlee joined #git
01:50 Shentino joined #git
01:51 e14 joined #git
01:52 preaction sam_yan: git diff > patchfile
01:52 blunaxela joined #git
01:52 fstd_ joined #git
01:52 fstd_ joined #git
01:53 sam_yan I have give a commit ,so I think "git diff " will not work for me.
01:54 _ADN_ joined #git
01:56 ResidentBiscuit Do you want the changes just for the directory you're in, or for the last commit?
01:56 sam_yan for the directory I am in
01:58 rorro joined #git
01:59 kejoma joined #git
01:59 ResidentBiscuit If you have a commit, you can do `git diff <hash>:directory/`
02:00 duderonomy joined #git
02:00 pks joined #git
02:01 sam_yan What I want to do is generate a patch for all the changes files which is in a directory
02:03 Doginal joined #git
02:03 Goplat joined #git
02:04 raijin joined #git
02:04 danielsh joined #git
02:05 JamesBaxter joined #git
02:05 danielsh How could I create a signed tag when I don't have git and gpg on the same box?
02:05 ojacobson install both on the same box somewhere
02:05 danielsh other than that ;-)
02:05 ojacobson ideally, somewhere you trust with your keys
02:06 ojacobson That's the only practical way. The storage format for signed tags isn't -that- complex, so you could reengineer it, but it's not simple enough that there's a stream you can feed into gpg if you just move it to the right box
02:06 ojacobson (the signature is interleaved in the middle of the tag object)
02:07 ojacobson [object preamble]\0[headers headers headers][signature][optionally more headers][tag message]
02:07 ochorocho__ joined #git
02:07 danielsh I _have_ tried to reengineer it, https://github.com/zsh-users/zsh-syntax-hig​hlighting/issues/360#issuecomment-244558342
02:08 danielsh I also figured I could hack something with a custom gpg.program.
02:08 actinide joined #git
02:09 Rodya_ joined #git
02:09 DolphinDream joined #git
02:11 fuchstronaut joined #git
02:14 danielsh ojacobson, thanks.
02:15 ochorocho__ joined #git
02:15 ojacobson Personally, I'd rent an AWS instance for a few minutes, generate a signing key there, sign the tag there, and then sign the signing key from your personal key later on
02:15 ojacobson For less than a buck you can avoid this whole problem :)
02:16 ojacobson Then go back and fix whatever infrastructure limitation currently stops you from running Git where your keys are
02:18 raijin joined #git
02:19 danielsh If I could've solved my problem by spinning up a VM I'd have done so. :-)
02:20 dendazen joined #git
02:20 evanwang joined #git
02:21 hobodave joined #git
02:22 mikecmpbll joined #git
02:23 jstimm joined #git
02:24 bambams joined #git
02:25 jamlee joined #git
02:27 red82 joined #git
02:29 clemf joined #git
02:34 DolphinDream joined #git
02:35 ojdo joined #git
02:35 ochorocho__ joined #git
02:38 Milijus joined #git
02:39 sargas joined #git
02:39 dreiss joined #git
02:40 mingrammer joined #git
02:40 crayon joined #git
02:40 crayon joined #git
02:49 JanC_ joined #git
02:50 rchavik joined #git
02:50 actinide joined #git
02:51 lagothrix joined #git
02:51 therealGent joined #git
02:52 morenoh149 joined #git
02:53 prawnzy joined #git
02:54 chachasmooth joined #git
02:58 hexagoxel joined #git
03:00 aavrug joined #git
03:02 Milijus joined #git
03:04 clemf joined #git
03:06 shgysk8zer0 joined #git
03:08 theskillwithin joined #git
03:10 Rodya_ joined #git
03:12 fuchstronaut joined #git
03:18 raijin joined #git
03:18 gfixler1 joined #git
03:18 qws-user-1228 joined #git
03:25 beautifulmind joined #git
03:25 krs93 joined #git
03:27 theskillwithin joined #git
03:28 lindenle joined #git
03:29 mehola joined #git
03:32 ejnahc joined #git
03:34 lagothri1 joined #git
03:34 bttf joined #git
03:35 lindenle joined #git
03:35 Cerise joined #git
03:35 Cerise joined #git
03:35 dilfridge joined #git
03:36 dhollinger joined #git
03:37 wget joined #git
03:37 wget joined #git
03:39 duderonomy joined #git
03:41 mizu_no_oto joined #git
03:44 Guest56_ joined #git
03:45 fish3485 joined #git
03:48 fedepad joined #git
03:49 Peuc joined #git
03:51 prawnzy joined #git
03:53 autoboxer joined #git
03:54 AlexxNica joined #git
03:59 romerocesar joined #git
04:02 ochorocho__ joined #git
04:03 Marbug joined #git
04:07 lindenle joined #git
04:09 krs93 joined #git
04:12 fuchstronaut joined #git
04:13 Rodya_ joined #git
04:17 superguest joined #git
04:17 superguest is the term "Head" synonymous with the term "Branch"?
04:18 romerocesar joined #git
04:18 danielsh "HEAD" is the commit that was last checked out.
04:18 danielsh "branch" is a named line of development.
04:19 daey_ joined #git
04:20 thiago_ HEAD is always written in capital letters. It's case sensitive.
04:21 qws-user-1228 joined #git
04:23 enfx joined #git
04:26 a3Dman joined #git
04:27 Norno joined #git
04:27 lindenle joined #git
04:27 Norno For some reason I had to do another merge straight after the first merge commit, is there any decent way to squash the two into one?
04:27 Norno Tried searching stackoverflow and similar, but couldn't find anything suitable
04:28 mingrammer joined #git
04:30 thiago Norno: of the same branch?
04:30 Norno yep
04:30 meZee joined #git
04:31 trelane joined #git
04:31 trelane joined #git
04:31 thiago there's a way, but it's not easy
04:31 thiago Norno: first of all, it's a history !rewrite. Have you pushed this?
04:31 gitinfo Norno: 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
04:32 Peng_ joined #git
04:32 Norno No, and I know
04:32 phanimahesh joined #git
04:32 lindenle joined #git
04:33 thiago one way is to set a graft for the current commit so that its first parent is actually its first grandparent, then do a git filter-branch to make it permanent
04:33 thiago another is to reset to that grandparent (before the merges) and merge again
04:33 thiago I advise doing the second
04:33 Norno I would've, but it was a seriously painful merge >.<
04:34 Norno might be best though
04:34 thiago sure, but you can reuse the entire resolution
04:34 thiago once git merge drops you in the conflicted state, run:
04:34 thiago git checkout <the SHA-1 of the current commit, right now> .
04:34 thiago including the dot at the end
04:34 thiago so get the current commit's SHA-1, before you reset
04:35 Norno hm, current commit, as in the one in the mergee branch right before the merge commit?
04:35 thiago the *current* commit, after all the merges and all the fixes
04:35 thiago before you reset
04:35 thiago the point is to reuse its tree, because you've already resolved all the conflicts
04:36 Norno I see, interesting
04:36 Norno also, how come I can't soft reset the latest merge commit?
04:36 Norno it just kills the changes
04:36 Norno the one before that I can keep them though
04:36 netj joined #git
04:37 thiago what happens when you git reset --soft?
04:37 thiago and why do you think it didn't reset?
04:37 Perspicaciosity joined #git
04:37 Norno it removes the commit, but nothing is staged afterwards
04:37 Norno i.e. it removes the changes as well (like a --hard reset)
04:38 thiago are you sure there were changes in that commit?
04:39 Norno well, if I `git merge master` with the second merge commit nothing happens
04:39 Norno if I undo it and do the same there's further conflicts
04:40 Norno it's quite strange
04:40 Andrew_K joined #git
04:41 morenoh149 joined #git
04:42 bronson joined #git
04:42 lindenle joined #git
04:43 qws-user-1228 joined #git
04:43 mingrammer joined #git
04:44 Norno I'll redo the merge thiago, thanks for the help nonetheless though.
04:46 bronson_ joined #git
04:48 morfin joined #git
04:48 morfin hello
04:49 morfin is there static git - i want to use that on shared hosting where possibly no git
04:49 snthsnth joined #git
04:51 harish_ joined #git
04:51 Cabanossi joined #git
04:51 durham joined #git
04:52 bronson joined #git
04:56 ochorocho__ joined #git
04:59 bronson joined #git
05:00 ochorocho__ joined #git
05:02 dec0n joined #git
05:02 subhojit777 joined #git
05:05 bronson joined #git
05:06 bouncy joined #git
05:08 gusnan joined #git
05:08 gusnan joined #git
05:11 bronson joined #git
05:12 zhangailu joined #git
05:12 zhangailu left #git
05:12 ochorocho__ joined #git
05:13 phanimahesh joined #git
05:13 fuchstronaut joined #git
05:19 fedepad joined #git
05:22 ochorocho__ joined #git
05:23 bronson joined #git
05:23 glebihan_ joined #git
05:26 laumann joined #git
05:27 Peng_ joined #git
05:28 trelane joined #git
05:28 trelane joined #git
05:30 meZee joined #git
05:31 TheMoonMaster joined #git
05:33 bocaneri joined #git
05:35 t0by joined #git
05:35 bronson joined #git
05:35 mohabaks joined #git
05:36 Caelum joined #git
05:37 psprint How to create Object with a file contents, without commit and tree Objects ?
05:38 ochorocho__ joined #git
05:38 satya4ever joined #git
05:41 bronson joined #git
05:42 diogenese joined #git
05:44 Kartagis joined #git
05:44 Kartagis ehlo
05:44 Nex6 joined #git
05:44 Kartagis why would I get "Access denied" when I have access the the repo?
05:45 grawity psprint: git hash-object -w
05:45 grawity Kartagis: you told git to a) use the wrong protocol or b) use the wrong credentials
05:49 Kartagis right, it was wrong protocol
05:49 Kartagis I could've sworn I pushed with the wrong one before
05:51 zeroed joined #git
05:51 Kartagis grawity++
05:51 Kartagis left #git
05:52 glebihan_ joined #git
05:52 stuh84 joined #git
05:54 bronson joined #git
05:56 Raging_Hog joined #git
05:57 Repox joined #git
05:58 bronson_ joined #git
06:00 thiago psprint: git add that file
06:02 Rodya_ joined #git
06:03 Caelum joined #git
06:04 bronson joined #git
06:05 rscata joined #git
06:06 netj joined #git
06:07 niels__ joined #git
06:07 psprint grawity: thanks
06:08 lordjancso joined #git
06:11 bronson joined #git
06:12 glebihan_ joined #git
06:14 fuchstronaut joined #git
06:14 heroux joined #git
06:15 dan2k3k4 joined #git
06:15 lb1c joined #git
06:16 bronson_ joined #git
06:18 evanwang joined #git
06:19 evanwang joined #git
06:20 _ikke_ low-level would be git hash-object -w
06:20 _ikke_ That would only create a blob, but not add it to the index
06:21 bronson joined #git
06:23 geoid joined #git
06:24 psprint _ikke_: yeah, I'm thinking about the wipe-file feature, but will probably do this without clearing changes of file, so action will be [StoreObject], then user might wipe file changes if he like hmm. Anyway, the point of [StoreObject] is: a file stored_objects.lst will be updated with [SHA] FileName\n, so user has this kind of backup, files hanging around until gc gets at them
06:25 _ikke_ then git hash-object -w (write) is what you want
06:25 _ikke_ Otherwise it would be more akin to stage
06:26 xcesariox joined #git
06:27 JeroenT joined #git
06:27 bronson joined #git
06:27 mycyber joined #git
06:28 JeroenT joined #git
06:28 raspado joined #git
06:29 dreiss joined #git
06:30 morenoh149 joined #git
06:30 jnavila joined #git
06:32 dan2k3k4 joined #git
06:34 Sceorem joined #git
06:35 sonikspin joined #git
06:35 ayjay_t joined #git
06:36 danslo joined #git
06:37 star_prone joined #git
06:38 durham joined #git
06:38 star_prone joined #git
06:42 stuh84 joined #git
06:43 AndreSomers joined #git
06:43 cdown joined #git
06:44 rendar joined #git
06:45 Alienpruts joined #git
06:47 t0by joined #git
06:50 ochorocho__ joined #git
06:50 PaulePanter joined #git
06:50 cdown_ joined #git
06:51 ShapeShifter499 joined #git
06:52 lindenle joined #git
06:53 geoid joined #git
06:54 Gloomy joined #git
06:55 eduardas_m joined #git
06:56 _ng joined #git
06:57 danslo joined #git
06:57 ochorocho__ joined #git
06:57 sam_yan joined #git
06:59 prawnzy joined #git
06:59 sam_yan when use "git format-patch -1 <commit value> --stdout  >tmp.patch  " and use "git apply --check tmp.patch",there is some error .
07:00 sam_yan error: patch failed: testcases/kernel_a/fs/fsx-linux/Makefile:29
07:00 thiago you can't apply a patch on top of itself, usually
07:00 thiago it will fail
07:01 laumann joined #git
07:02 thiago sam_yan: did you mean that you applued the patch to another commit?
07:02 sam_yan I apply in its parent directory
07:02 thiago directories are irrelevant
07:02 thiago what commit did you apply it on?
07:03 thiago was it on <commit value>~ ?
07:03 sam_yan yes
07:03 thiago please confirm by doing git log. Is it the same SHA-1 ?
07:04 sam_yan commit f502caf8476baa3b7508d06edd88641412fb29f1
07:04 sam_yan Author: Wenzhong Sun <wenzhong.sun@intel.com>
07:04 sam_yan I use that
07:04 sam_yan f502caf8476baa3b7508d06edd88641412fb29f1
07:05 thiago which one is that?
07:05 thiago the one you're trying to apply? Or the one that is current?
07:05 sam_yan I can not understand what you mean
07:06 thiago is that the current commit in the repository?
07:06 sam_yan NO
07:06 thiago that's the one you're trying to format-patch and apply, then?
07:07 SwiftMatt joined #git
07:07 sam_yan i use git log and get the commit value
07:07 thiago so by "no" you mean "yes"
07:08 thiago what's the SHA-1 of the commit that you passed to git format-patch?
07:08 submitnine joined #git
07:08 AndreSomers joined #git
07:08 Silenced joined #git
07:09 prawnzy joined #git
07:11 sam_yan git format-patch -1 $name -n -o ~/Templates/git_patch/
07:12 thiago do a git log $name~
07:12 thiago what is the SHA-1 of the commit at the top?
07:14 ochorocho__ joined #git
07:14 sam_yan it has multiple number
07:15 fuchstronaut joined #git
07:15 thiago there's exactly one at the top. What is it?
07:15 sam_yan yes
07:15 sam_yan and then?
07:16 m1m3-50 joined #git
07:16 thiago and then you tell me
07:16 sam_yan 72f6f120a3a91346f42cbf5cd6ce2829c146f38a
07:16 thiago ok
07:16 thiago now go to the repository where you're trying to get the patch applied
07:16 thiago do git log. What's the SHA-1 at the top?
07:18 sam_yan 72f6f120a3a91346f42cbf5cd6ce2829c146f38a
07:18 thiago ok, so it is the same
07:18 danslo joined #git
07:18 sam_yan yes
07:18 thiago why are you trying to transfer this commit via format-patch + apply?
07:18 thiago you should use git pull instead
07:19 overlord_tm joined #git
07:19 thiago alternatively, use git am to apply the patch, not git apply
07:20 sam_yan why not git apply?
07:20 thiago because git apply is the reverse of git diff
07:20 thiago the reverse of git format-patch is git am
07:20 danslo joined #git
07:21 TomyWork joined #git
07:21 cagedwisdom joined #git
07:22 sam_yan sorry ,the SHA-1 is not same ,my fault
07:23 thiago then you may be getting a simple conflict
07:23 circ-user-f7FIN joined #git
07:23 thiago are the two repositories on the same computer?
07:23 sam_yan no
07:23 PCatinean joined #git
07:23 thiago how are you transferring the file?
07:23 tango_ joined #git
07:23 thiago the patch
07:23 sam_yan scp
07:23 thiago ok
07:24 thiago so the most likely scenario is a simple conflict
07:24 thiago if you use git am -3 tmp.patch, it will allow you to run git mergetool to help fix the conflicts
07:24 ajhansen joined #git
07:24 sam_yan git am -3 tmp.patch ?
07:24 AndreSomers joined #git
07:24 sam_yan i will try
07:24 thiago yes
07:25 thiago make sure your tree is clean, no uncommitted changes
07:25 Balliad joined #git
07:26 netj joined #git
07:27 lindenle joined #git
07:28 sam_yan thiago:can i have your e-mail?
07:29 thiago why?
07:30 Guest56_ joined #git
07:30 JeroenT_ joined #git
07:30 sam_yan maybe i need to ask you ,because there is so many questions
07:30 fedepad joined #git
07:30 thiago you probably have people closer to you in your office
07:31 dan2k3k4 joined #git
07:31 ldlework oh my god
07:31 ldlework I git rebase --skip'ed a whole rebase
07:31 ldlework and now all my changes are gone
07:31 ldlework oh my god
07:32 ldlework please help me
07:32 AndreSomers joined #git
07:33 ldlework I think I'm gonna throw up
07:33 thiago is the rebase over?
07:33 ldlework yes
07:33 thiago ok
07:33 ldlework all the code is gone...
07:33 thiago have you done anything since it ended?
07:33 ldlework yeah I checkout master, forked a new branch, make a commit, pushed that new branch up
07:33 aidalgol joined #git
07:33 thiago ok
07:34 thiago what branch is the one that is now messed up?
07:34 ldlework noticed that my PR of the branch we're talking about only has 1 commit now..
07:34 theskillwithin joined #git
07:34 ldlework builder_refactor
07:34 thiago ok
07:34 thiago have you done anything to that branch after the rebase ended?
07:34 thiago you can answer "I don't remember"
07:34 ldlework git submodule update I think
07:35 thiago that doesn't change the branch
07:35 thiago anyway, git checkout builder_refactor
07:35 thiago git log -g builder_refactor
07:35 thiago it's going to list all commits that builder_refactor has recently pointed to
07:35 ochorocho__ joined #git
07:35 ldlework git do I just checkout the commit
07:35 thiago from the second one onwards, copy & paste the SHA-1, do a git log and check if it is what you wanted
07:35 ldlework of where it was before the rebase
07:36 thiago it's probably the one after:
07:36 thiago Reflog message: rebase finished: returning to refs/heads/builder_refactor
07:37 thiago once you've found the right one, git reset --hard thatsha1
07:37 ldlework oh glory
07:38 ldlework thiago: thank you
07:38 thiago as long as you commit, git makes it very hard to actually lose data
07:38 ldlework I need to remember that
07:38 ldlework its just horrifying seeing the code actually gone
07:38 ldlework O_O
07:38 thiago the only operation that destroys the reflog is deleting a branch
07:39 m1m3-50 joined #git
07:39 thiago that's why git branch -d says "you haven't merged, are you sure?"
07:42 ochorocho__ joined #git
07:43 ramsub07 joined #git
07:43 Triq joined #git
07:44 Balliad joined #git
07:44 antlarr joined #git
07:47 _ikke_ git reflog is a safety net
07:47 _ikke_ Note that the HEAD reflog still contains the commit you last checked out from that branch
07:48 huggy joined #git
07:49 kurkale6ka joined #git
07:49 thiago yeah, but HEAD's reflog is usually a mess on a busy repository
07:49 jknetl joined #git
07:49 _ikke_ Correct, but still a good last resort
07:50 ahmedelgabri joined #git
07:50 _ikke_ Wel, not a last last resort (git fsck --lost-found would be)
07:50 Rodya_ joined #git
07:51 thiago yeah
07:54 _ADN_ joined #git
07:56 niels__ joined #git
07:57 Triq Hi, does anyone know if it is safe to remove the svn (.git/svn) after migrating and creating the branches/tags in .git/ref? Perhaps it just isn't included when pushed to the server so it doesn't really matter?
07:58 _ikke_ correct, it's not included
07:58 _ikke_ One often used way to un-svn a repo is to just clone it again
07:59 geoid joined #git
08:00 Triq That sound like a good idea. Thanks
08:00 DevAntoine joined #git
08:00 DevAntoine hi
08:02 _ikke_ hai
08:02 DevAntoine I've got multiple files no staged on a given branch. I need to stash them to be able to checkout develop, but I need to pull a single file from this stash. Can I do it?
08:02 bernardio joined #git
08:03 cbreak DevAntoine: stashing is like committing
08:03 crazycoder joined #git
08:03 crazycoder hello
08:03 cbreak DevAntoine: you get a stash commit (pair)
08:03 thiago it's easy to get the exact file that was stashed
08:04 thiago it's not possible to unstash just one file
08:04 cbreak you can then check out from those as you'd normally would
08:04 crazycoder anyone use gitlab? i am using their CI and i would like to merge into master during the deploy stage
08:04 crazycoder how can i do that?
08:04 _ikke_ git checkout stash@{n} <file> where n is the stash number
08:04 DevAntoine if I got a single stash n will be 0?
08:04 _ikke_ crazycoder: automatic merging is not recommended
08:04 _ikke_ DevAntoine: git stash list
08:04 DevAntoine that's nice to know cbreak, _ikke_
08:04 DevAntoine everything is a commit it seems :D
08:05 crazycoder _ikke_, why ? the deploy stage is called after the tests
08:05 _ikke_ crazycoder: whatabout merge conflicts?
08:05 _ikke_ (And they will happen)
08:06 djb-irc joined #git
08:06 crazycoder _ikke_, from dev to master? i think there are no conflict because master is always touched from dev   i always do git merge --no-ff dev (from master)
08:07 navidr joined #git
08:07 _ikke_ Untill someone does a hotfix on master?
08:07 thiago you should use --ff-only actually
08:07 thiago that one is guaranteed to never have a conflict
08:08 compeman joined #git
08:08 re1 joined #git
08:08 _ikke_ But breaks when it cannot ff
08:09 thiago right, which is what you want
08:10 esc joined #git
08:10 crazycoder thiago, --ff-only what is the diff?
08:10 thiago huh?
08:10 crazycoder i mean
08:10 crazycoder --no-ff and --ff-only
08:10 crazycoder what is the difference?
08:10 thiago --no-ff creates a merge commit even if it didn't have to
08:11 thiago --ff-only only merges if it is a fast-forward
08:11 tobiasvl crazycoder: why do you always do --no-ff if you don't know what it does? :)
08:12 crazycoder tobiasvl, i asked for differences between them
08:12 qt-x joined #git
08:13 crazycoder thiago, ok so in this case i will not have problems it will works always (fast.forward) otherwise it fails...that's what i need
08:13 crazycoder ok
08:14 ahmed_elgabri joined #git
08:15 zeroed joined #git
08:15 zeroed joined #git
08:15 ochorocho__ joined #git
08:15 fuchstronaut joined #git
08:17 Mathiasdm joined #git
08:19 morenoh149 joined #git
08:19 nopf joined #git
08:23 djb-irc joined #git
08:25 chll_ joined #git
08:26 ochorocho__ joined #git
08:26 ajhansen joined #git
08:26 ocbtec joined #git
08:29 perrier-jouet joined #git
08:30 perrier-jouet hi all
08:30 therealGent joined #git
08:30 gitinfo perrier-jouet: 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.
08:30 perrier-jouet hi
08:30 mischat joined #git
08:30 perrier-jouet when we do git stash is it possible to recover one ancient file ?
08:31 cbreak perrier-jouet: what?
08:32 cbreak how is the first and second part of your question related?
08:32 elect joined #git
08:34 cqi joined #git
08:34 perrier-jouet cbreak: I was in .zprezto and wanted to update to latest, it told me to do a stash and I did, but I lost my .zshrc I wanted to revert back to the old .zshrc
08:35 niels__ joined #git
08:35 cbreak perrier-jouet: to undo a git stash do a git stash pop
08:35 cbreak git stash stashes away changes, git stash pop gets them back
08:36 balor joined #git
08:36 Serpent7776 joined #git
08:37 perrier-jouet cbreak: I think I lost my ancient .zshrc since git stash pop did not show the previous .zshrc
08:38 cbreak git only knows about things you commit
08:38 achadwick joined #git
08:38 cbreak did you commit that file?
08:38 balor A git clone over dumb HTTP is looking for "GET /AppRepo.git/info/refs?service=git-upload-pack HTTP/1.1", which is a bare repo as described in the docs.  However, my repo has no /info/refs, but does have a /refs.  git 2.1 on the server and 2.7 on one client, 2.1 on the other client.
08:38 balor Why is it looking for 'refs' in the wrong dir?
08:38 cbreak balor: with dumb http, git doesn't need to be on the server
08:39 psprint Git log -p will show subsequent diffs for a file. How to show instead: accumulated subsequent diffs for file, i.e. not A-B, B-C, C-D, etc., but: A-B, A-C, A-D, etc.
08:40 cbreak balor: check out man git-update-server-info
08:40 gitinfo balor: the git-update-server-info manpage is available at http://jk.gs/git-update-server-info.html
08:40 perrier-jouet cbreak: yes I commit that file
08:41 balor thanks all
08:41 cbreak perrier-jouet: then git log -- filename
08:41 cbreak pick a commit
08:41 cbreak and git checkout thatcommit -- filename
08:42 psprint How to make git log --follow -p <file> show only IDs ?
08:42 cbreak psprint: not at all
08:42 cbreak the whole point of -p is to show patches
08:42 cbreak if you don't want patches, don't use -p
08:43 cbreak psprint: check out man git-rev-list instead
08:43 gitinfo psprint: the git-rev-list manpage is available at http://jk.gs/git-rev-list.html
08:43 psprint thanks
08:44 psprint rev-list seems to need commits
08:44 perrier-jouet cbreak: the log tells me that last commit was in 2012 so I guess I did not commit. i use prezto zsh framewokr and it creates .zprezto and I wanted to update it that's when I lost my .zshrc
08:45 psprint axcept there's --all :]
08:45 cbreak psprint: you have to tell it what to list from
08:45 cbreak psprint: tags, branches, what ever
08:45 cbreak psprint: unlike git log it does not default to HEAD
08:46 planet_en joined #git
08:46 sanketdg joined #git
08:46 planet_en how can I delete all the branches merged to a specific branch 'foo' and older than 3 months.
08:47 djb-irc joined #git
08:50 mikecmpbll joined #git
08:53 midnightmagic joined #git
08:56 ToBeCloud joined #git
08:57 leeN joined #git
08:57 nandi_ge___ joined #git
08:58 ochorocho__ joined #git
09:02 Raansu joined #git
09:03 Fissure joined #git
09:04 rorro joined #git
09:05 dreamyspell joined #git
09:06 ochorocho__ joined #git
09:06 djb-irc joined #git
09:08 robogoat joined #git
09:08 cdown joined #git
09:10 JeroenT joined #git
09:13 MattMaker joined #git
09:15 qt-x joined #git
09:16 timewalker joined #git
09:16 stamina joined #git
09:16 fuchstronaut joined #git
09:17 superguest My local working branch is ahead of of 'origin/master' by 1 commit.  That's b/c I've deleted ('git rm file') a file locally.
09:17 ochorocho__ joined #git
09:17 tobiasvl superguest: sounds like you want to push it then
09:18 mycyber joined #git
09:18 superguest tobiasv1, actually I want to experiment by doing a pull from the origin to see if it'll recover the deleted file.
09:19 morfin so - is there static git somewhere?
09:19 cbreak superguest: of course it won't
09:19 tobiasvl superguest: it will not. if you pull from origin, you won't get any new commits, since you are ahead.
09:19 cbreak superguest: you made a change, git'd be braindead to kill it
09:19 tobiasvl morfin: static git?
09:20 morfin i hate shared hosting - it has no Git and we used it for deploy
09:20 cbreak morfin: run it locally and rsync
09:21 ig0r_ joined #git
09:21 morfin well, i was thinking about just static binary )
09:22 mingrammer joined #git
09:22 ShapeShifter499 joined #git
09:22 mingrammer joined #git
09:22 superguest tobiasvl: okay, let me experiment with merge to see if it'll do it, since the Git Cheat Sheet says " *merge*  combines bookmark's branch into current local branch"
09:23 tobiasvl "the Git Cheat Sheet"
09:23 tobiasvl what's "bookmark"
09:23 tobiasvl but anyway, that's correct, it combines the origin's master into your local master – but there's nothing to combine
09:23 tobiasvl you have everything that origin's master has
09:23 tobiasvl already
09:24 superguest lolz.... right.
09:25 superguest tobiasvl: okay, what's the solution then?
09:25 tobiasvl solution?
09:25 tobiasvl I wasn't aware you had a problem
09:26 superguest recover the locally-deleted file from the origin's master
09:26 tobiasvl you don't need to involve origin to do that
09:26 tobiasvl you still have the local history
09:26 tobiasvl you want to throw away your latest commit which deleted the file?
09:28 MattMaker joined #git
09:28 ochorocho__ joined #git
09:28 tobiasvl superguest: !revert
09:28 gitinfo superguest: That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?
09:30 tlaxkit joined #git
09:30 harish_ joined #git
09:31 dreamyspell joined #git
09:31 Lomex joined #git
09:32 DnlXz joined #git
09:33 jophish joined #git
09:34 niels__ joined #git
09:34 jophish I'm rebasing B onto M. I've got a conflict. Is it possible to see this commit's original diff
09:34 jophish the one on the non-rebased B
09:35 jophish it would be very helpful to guide my reosolution
09:35 superguest tobiasvl: is it always the case that trying to merge an older commit (in this case, my origin's master) with a newer commit (in my case, my local) that there will be *nothing* to combine?
09:35 niels__ joined #git
09:35 canton7 jophish, related: tried setting merge.conflictstyle to diff3
09:36 canton7 it adds the original version of the conflicted hunk, so you can see the changes that B originally made
09:36 jophish canton7: ah, that sounds very useful
09:36 jophish perhaps vim can do this
09:36 tobiasvl superguest: yes.
09:36 niels__ joined #git
09:36 canton7 I can't fix conflicts without it any more :P
09:36 qt-x joined #git
09:36 tobiasvl superguest: you merge _history_, remember. there's nothing to combine because you already have that commit in your local history.
09:36 tobiasvl merging that commit into your history would do nothing.
09:38 Rodya_ joined #git
09:38 MattMaker joined #git
09:39 superguest Option c) undo committing [git reset --soft HEAD^];
09:39 superguest what does the '^' in 'HEAD^' mean?
09:40 nikivi joined #git
09:40 tobiasvl HEAD's parent
09:40 tobiasvl in your case, the same commit that origin/master points to
09:40 mikecmpbll joined #git
09:40 mischat joined #git
09:40 tobiasvl (and your HEAD is master)
09:41 superguest tobiasv1, my intuition is often silly, but will HEAD^^ give me the grandpartent?
09:41 a3pq51 joined #git
09:42 tobiasvl yes it will
09:42 tobiasvl your intuition is fine. you can also write HEAD^2
09:43 tobiasvl just note that ^ gives you the _first_ parent. in your case, HEAD only has one parent, but in the case of merges, there are two parents
09:43 tobiasvl man gitrevisions
09:44 gitinfo the gitrevisions manpage is available at http://jk.gs/gitrevisions.html
09:45 tobiasvl I wrote the wrong thing too... HEAD^^ is equivalent to HEAD~2, following first parents
09:46 djb-irc joined #git
09:48 MattMaker joined #git
09:48 prawnzy joined #git
09:55 ochorocho__ joined #git
09:55 courrier joined #git
09:58 nuck joined #git
09:58 theskillwithin joined #git
09:59 Silenced joined #git
10:01 mehola joined #git
10:03 JeroenT joined #git
10:03 MattMaker joined #git
10:06 cdown joined #git
10:08 morenoh149 joined #git
10:10 Pyrogerg_ joined #git
10:11 mischat joined #git
10:11 ochorocho__ joined #git
10:15 Fuco joined #git
10:15 Fuco is there some way to find out which .gitignore rules are no longer required?  We restructured the project quite a bit and many of those are no longer valid, but I'd rather not go through the list manually :p
10:15 Fuco I need rules which don't match any file anymore
10:16 garethdaine joined #git
10:17 osse Fuco: do you have a repo that is sufficiently dirty?
10:17 Fuco I don't know what you mean by that
10:17 fuchstronaut joined #git
10:17 Fuco the .gitignore file has over 100 rules
10:17 ahmed_elgabri joined #git
10:17 Fuco don't even ask :D
10:18 canton7 Fuco, something in man git check-ignore maybe?
10:18 gitinfo Fuco: the git-check-ignore manpage is available at http://jk.gs/git-check-ignore.html
10:18 osse Fuco: if you clone the repo then none of the patterns match anything at all
10:18 MattMaker joined #git
10:18 osse Fuco: so to know which rules are useful and which rules are not you need a repo that has been used (code built etc.)
10:18 canton7 that'll give you a list of all patterns that match, and it'll be easy enough to use that to find the ones that won't
10:19 Fuco osse: right, I have that :)
10:19 Fuco canton7: seems like a plan, will check that
10:19 canton7 Fuco, well, I could write some code that did it in around 5 lines :P
10:19 osse Fuco: git ls-files --others -z | xargs -0 git check-ignore -v
10:20 osse that'll tell you which rule applies to each file
10:20 canton7 doesn't check-ignore tell you that anyway?
10:20 fuchstronaut joined #git
10:20 fuchstronaut joined #git
10:21 osse what do you mean?
10:21 osse you need to give it paths to files
10:22 osse Fuco: pipe all that to  awk -F'\t' '!seen[$1]++ { print $1 }'  to filter it, so each pattern is only mentioned once
10:22 Fuco shouldn't I use find instead of ls-files? Or does that list everything, even non-versioned files?
10:22 canton7 ah righty, I'd assumed you could give it a dir
10:22 osse Fuco: that's what ls-files --others does
10:22 osse lists ONLY non-versioned files
10:22 Fuco great :)
10:23 Fuco this is also rad !seen[$1]++ !
10:23 ahr3n joined #git
10:23 osse i almost soiled my panties writing it
10:23 Fuco amazing, thanks!
10:24 Fuco I see my awk-fu is weak :D
10:27 lomig joined #git
10:28 ochorocho__ joined #git
10:28 osse Fuco: ok, so after you've found the list of patterns not currently in use you have to find out which are actually not useful and which are but just don't apply at the moment.
10:28 osse good luck
10:29 canton7 I'm assuming they're all related to build artefacts or something, so doing a full builds will make them all...
10:30 Fuco yea, this should just be the difference
10:30 Fuco in the worst case we can add back those I purge but should not
10:30 Fuco thank you for the help, you're all awesome :)
10:30 dan2k3k4 joined #git
10:32 Milijus joined #git
10:32 osse i'm starting to think it would actually be easier to remove the obviously useless ones and then remove the supicious once one or a couple at a time and check git status
10:32 osse instead of massaging the output of the pipe i gave
10:33 osse (good thing I put "the output of" into that sentence. or maybe not...)
10:33 osse s/once/ones/
10:36 ochorocho__ joined #git
10:38 MattMaker joined #git
10:40 planet_en joined #git
10:41 ochorocho__ joined #git
10:45 irco joined #git
10:48 rnsanchez joined #git
10:48 MattMaker joined #git
10:49 JeroenT joined #git
10:50 morfin how can i clean up all stuff not related to git?
10:50 tobiasvl morfin: what do you mean
10:50 morfin i mean  lots of generated by build process files/directories
10:51 tobiasvl man git clean
10:51 gitinfo the git-clean manpage is available at http://jk.gs/git-clean.html
10:51 morfin the problem is they're possibly ignored hmm
10:51 vuoto joined #git
10:51 canton7 there's a flag to 'clean' to delete ignored files
10:52 planet_en joined #git
10:53 red82 joined #git
10:53 osse git clean -fdx
10:53 osse it's like you recloned
10:53 ramsub07 joined #git
10:54 nilez joined #git
10:55 morfin also that has submodules
10:55 * osse runs in the opposite direction
10:55 morfin :))
10:55 morfin i will try to clean - seems like git can't even print so many files
10:57 ochorocho__ joined #git
11:00 Milijus joined #git
11:01 ochorocho__ joined #git
11:02 danslo joined #git
11:04 MattMaker joined #git
11:06 BadSnappa joined #git
11:07 danielsh left #git
11:08 JeroenT joined #git
11:08 danslo joined #git
11:10 Rodya_ joined #git
11:10 raijin joined #git
11:11 rchavik joined #git
11:11 planet_en joined #git
11:13 cyan__ joined #git
11:13 clmsy joined #git
11:13 nikivi joined #git
11:18 ochorocho__ joined #git
11:20 fuchstronaut I would like to commit both steps seperately without adding in removing stuff in the editor, is there a neat way with git? http://pastebin.com/ctQ6vgfY
11:20 singalaut joined #git
11:21 Mindfart joined #git
11:22 Jellyg00se joined #git
11:23 ochorocho__ joined #git
11:23 MattMaker joined #git
11:23 mingrammer joined #git
11:25 jnavila fuchstronaut: git add -p then edit the hunk according to instructions in the editor
11:28 MattMaker joined #git
11:34 Vampire0 joined #git
11:35 psprint Can be a bunch of IDs referenced somehow without creating branches or tags, to protect them from garbage collector?
11:36 ochorocho__ joined #git
11:37 nuck joined #git
11:38 tobiasvl why not just create a branch?
11:38 tobiasvl branches are cheap
11:38 netj joined #git
11:38 psprint the IDs are just blobs stored with hash-object
11:41 planet_en joined #git
11:41 Dong__ joined #git
11:42 canton7 create a tree which references them, and point a branch/tag at that tree?
11:42 canton7 at that point you've basically got a branch which has a bunch of files in it - but that's not surprising, as that's kinda what git does
11:43 isysd joined #git
11:43 Dong__ guys, trying to understand smart protocol, is there a plumbing command allows me to specify "wants" and get a thin pack accordingly? Or put another words, what is most easy way to fetch a thin pack to look at?
11:43 tlaxkit joined #git
11:44 jnavila joined #git
11:45 mingrammer joined #git
11:46 ochorocho__ joined #git
11:46 psprint canton7: interesting idea. The branch could be named: "pocket" (sic), would point always to single tree, that would reference blobs. However, I would like to store a file multiple times by using git hash-object, i.e. same path. Maybe I could just append timestamp to file name
11:46 ash_workz joined #git
11:47 canton7 or create a directory named after the filename, and put the file as an index - filename/1, filename/2, etc
11:47 canton7 (or timestamp, etc)
11:48 MattMaker joined #git
11:49 prawnzy joined #git
11:50 Jellyg00se joined #git
11:54 raijin joined #git
11:55 chris2 joined #git
11:56 morenoh149 joined #git
11:58 psprint passing a path would prevent requirement of additional tree for each directory?
11:58 canton7 not sure what you mean by that I'm afraid
11:58 ssitani joined #git
11:58 canton7 passing a path to what? what requirement for additional trees for each directory? what directory@?
11:59 Prisoner-2460_1 joined #git
11:59 ochorocho__ joined #git
12:00 psprint canton7: tree references trees when listing directories. So you traverse tree (main dir) -> tree (sub dir) -> blob.  You did "filename/1" as if you want to put that directly into the "base" tree
12:01 canton7 no, you'd need to create a tree
12:02 chris2 joined #git
12:03 MattMaker joined #git
12:04 mehola joined #git
12:04 KeelOfSteel joined #git
12:04 qt-x joined #git
12:05 ochorocho__ joined #git
12:06 Lomex joined #git
12:08 dsdeiz joined #git
12:09 grawity if you don't want a "branch or tag", create another ref type
12:09 grawity like how git-notes uses a very similar structure under refs/notes/foo
12:09 niels__ joined #git
12:09 psprint grawity: but the objects wouldn't be tied together by parent-relation from commit object, they would be just files/blobs
12:10 grawity that's not the only type of relation
12:10 grawity refs don't have to point to a commit anyway
12:10 ochorocho__ joined #git
12:11 Rodya_ joined #git
12:12 psprint ahm, indeed a directory under refs/ would serve quite well to prevent flooding repo with reference files, they would be isolated
12:12 planet_en joined #git
12:12 mischat joined #git
12:13 psprint grawity: so it's about holding simple single-Object (no parent relation) refs under refs/notes or similar directory, or is there a type of relation different from commit's that can be used?
12:13 grawity git doesn't care
12:13 grawity a ref can point at *any* object
12:13 mkoskar joined #git
12:13 grawity git.git itself has a few blobs tagged; linux.git has a tagged tree
12:13 rgrinberg joined #git
12:14 bruce_lee joined #git
12:14 bruce_lee joined #git
12:14 psprint ok, but it would be nice to limit number of files in refs/notes, and git garbage collection rather cares about parent relation to detect unreferenced objects
12:15 Noldorin joined #git
12:15 grawity git garbage collection doesn't care either
12:16 grawity I mean, it accepts all 'types of relation' that git itself accepts
12:16 mob1le joined #git
12:16 grawity as long as a ref eventually somehow points to an object, it won't be gc'd
12:16 grawity even if it's ref->tree->blob or just ref->blob
12:16 psprint ok that's clear
12:17 grawity so, if you have a bunch of objects
12:17 grawity make a tree with all of them
12:17 grawity point refs/foo/bar to that tree
12:17 grawity that's all
12:18 MattMaker joined #git
12:18 jeffreylevesque joined #git
12:20 trondth joined #git
12:22 dmpeters63 joined #git
12:23 LBV_User joined #git
12:26 balor joined #git
12:26 fmcgeough joined #git
12:27 GodGinrai joined #git
12:28 jimi_ joined #git
12:29 MattMaker joined #git
12:29 Dong__ Anybody knows how to generate a thin pack?
12:30 raspado joined #git
12:30 jast Dong__: git pack-files has a --thin option, if that's what you're going for
12:31 GodGinrai thin packs are?
12:31 jast packs in which some objects are omitted. push uses thin packs to transmit objects without re-transmitting all of the history shared by the sender and the receiver.
12:32 Prisoner-2460_1 joined #git
12:32 GodGinrai ah, I see
12:32 GodGinrai thanks for the explanation :)
12:33 Dong__ jast: pack-files?
12:34 Zilenc3 joined #git
12:34 ToxicFrog joined #git
12:34 jast Dong__: pack-objects, my bad
12:35 Dong__ jast: I saw that --thin there but haven't got clue how to do it
12:36 autoboxer joined #git
12:36 sdothum joined #git
12:38 fuchstronaut how do i remove a file from beeing tracked on the server? I accidentially pushed a commit with a vim temp file (*.swp) and now i did 'git rm --chached filename.swp' and added *.swp to my .git/info/exclude file
12:38 fuchstronaut shall I just commit?
12:39 _ikke_ sure
12:39 _ikke_ Is it the last commit?
12:39 fuchstronaut no sadly its somewhere in the history
12:40 jast Dong__: well, you give it a list of objects, right? normally pack-objects also includes all objects required/referenced by those, but with --thin it will exclude base commits and delta bases and only include related trees/blobs and such
12:40 fuchstronaut but a correction commit isnt too bad i think
12:42 jast fuchstronaut: you can rewrite the history and exclude the file retroactively but it gets painful if other people are involved (i.e. have copies of the current version of history). if in doubt, make a correction commit just like you described.
12:42 netj joined #git
12:42 fuchstronaut jast: thank you+
12:42 jast by the way, I've configured vim to put swapfiles in a different location
12:42 fuchstronaut will do the correction commit
12:42 mizu_no_oto joined #git
12:43 fuchstronaut are you guys pro putting ide-specific files in the global .gitignore file?
12:43 planet_en joined #git
12:43 nikivi joined #git
12:43 jast I don't do it but I'd have no qualms. it's not like it hurts anyone...
12:44 * Bombe thinks those belong in .git/info/exclude.
12:45 Bombe .gitignore is for files created by gradle/maven/your-build-system.
12:46 jast if the project consists of disciplined people, sure ;)
12:46 fuchstronaut i think so too, but on the other hand, files like .swp, .idea, etc. are (99.9%) not needed for a project so excluding these in .gitignore would stop idiots like me adding them to the index
12:46 JeroenT joined #git
12:46 HelloAtMega left #git
12:46 Bombe Just fire everyone who is not disciplined. Problem solved!
12:47 fuchstronaut we would be a 0-developer shop
12:47 fuchstronaut oh well.. not we, I would have to go somewhere else then..
12:47 jast well, great, that way you minimize personnel costs, too
12:47 fuchstronaut so smart...
12:48 fuchstronaut And it would be less workload for each of us
12:48 MattMaker joined #git
12:48 jast win-win
12:48 Bombe I can not see a single downside to that.
12:49 HelloShitty joined #git
12:50 balor joined #git
12:50 johnmilton joined #git
12:51 ochorocho__ joined #git
12:51 whitby joined #git
12:51 mycyber joined #git
12:51 dendazen joined #git
12:52 NeXTSUN joined #git
12:53 synthroid joined #git
12:53 zivester joined #git
12:55 ochorocho__ joined #git
12:56 smithbone joined #git
12:57 p4trix joined #git
12:58 MattMaker joined #git
13:00 ochorocho__ joined #git
13:05 DnlXz joined #git
13:07 ExoUNX joined #git
13:07 algae joined #git
13:08 cdg joined #git
13:09 Dong__ jast: I tried to find a delta object and did this: echo "28c214a32fc93e96c868984bad9d24eee442d105" | git pack-objects --stdout --thin | hd
13:09 Dong__ jast: I thought with or without --thin I should get different output, but they are same.
13:13 yottabyte joined #git
13:13 jast Dong__: how did you determine that this object is a delta, and what type of object is it?
13:13 ShekharReddy joined #git
13:13 planet_en joined #git
13:13 MattMaker joined #git
13:14 WayToDoor joined #git
13:15 Dong__ jast: I tried to find a perfect object by looking at "git verify-pack -v ..........idx"
13:15 ochorocho__ joined #git
13:15 jast okay, sure, that works
13:15 red82 joined #git
13:16 Dong__ but even it is a delta, with or without --thin I seems always get the whole thing in output
13:16 jast so what about the second part of my question? :)
13:17 Dong__ it is a blob
13:17 jast that would explain it, the logic that pulls in more objects works only on the commit level
13:17 Dong__ you mean I should find a commit which is a delta?
13:19 wrouesnel1 joined #git
13:19 cdown joined #git
13:20 jast or any commit that references delta'd objects, I think
13:20 Vampire0_ joined #git
13:21 Dong__ wow, this is funny now, see what happens with a commit http://pastebin.com/HTx3wDMx
13:22 Dong__ giving it --thin packed almost everything, without just packed this single commit
13:22 eshlox joined #git
13:23 Dong__ totally confused now.
13:23 ochorocho__ joined #git
13:23 rendar joined #git
13:23 CEnnis91 joined #git
13:24 Vivekananda joined #git
13:25 psprint By default gc doesn't run automatically?
13:26 canton7 it goes, from time to time
13:26 canton7 *does
13:26 psprint never seen it, maybe Apple's Git is somehow configured
13:26 officer_raccoon joined #git
13:27 NeverDie joined #git
13:27 canton7 you won't see any different output
13:27 jast Dong__: I looked at the logic some more. with no options, pack-objects uses the exact list of objects on the command line. --thin enables calculating objects, but includes less objects than other options that enable this.
13:28 psprint ahsoo, I thought it will be like if I would run git gc. But shouldn't I at least observe very long e.g. commit time?
13:28 MattMaker joined #git
13:28 oxsyn joined #git
13:28 jast psprint: the automatic gc does a lot less than the normal gc, IIRC
13:28 yottabyte what if I think one of my local changes is breaking the build? can I do a sanity check? do I have to stash it or something?
13:29 jast yottabyte: for uncommitted changes, stashing would be a straightforward way to check, yeah
13:29 DnlXz joined #git
13:29 yottabyte how do I do that
13:29 yottabyte for a single file, potentially
13:29 Vampire0 joined #git
13:29 ochorocho__ joined #git
13:30 acr4 left #git
13:30 jast with git stash save -p you can stash individual changes, but with any reasonable workflow where you don't have a ton of different things uncommitted, I'd just stash everything
13:31 _xor joined #git
13:31 Dong__ jast: I still can't generate a thin pack to study into. But let me try another angle, when a server sent over a thin pack over wire, it can't use ofs_delta but must be ref_delta right? Because there is no way to figure out which pack file the offset should refer to, so the only way to find the base object is by ref?
13:32 yottabyte but then how can I see which change is causing the problem
13:32 yottabyte and I did a git pull and now it's saying I have a lot of changes to files I didn't change when I do git status
13:32 yottabyte and they'er all red, what does that mean?
13:32 sangy joined #git
13:33 jast Dong__: as I understand it, ofs_delta can only be used within the same pack... so, in theory, a thin pack could use ofs_delta to refer to another object in the same pack, but not another pack
13:34 jast yottabyte: that would depend on which section of 'git status' they show up in
13:34 jast red *usually* means 'deleted'
13:35 jast but it can also mean "changes that are not staged"
13:35 yottabyte it's modified: file name
13:35 f3r70rr35f joined #git
13:35 jast that's the line, not the section the line shows up in
13:37 yottabyte oh they're in: Changes not staged for commit:
13:37 ochorocho__ joined #git
13:37 jast okay, that indicates the files are changed but the changes are neither staged nor committed
13:37 mischat joined #git
13:37 jast if that happens even though you did not change the file(s), it's usually because of autocrlf, eol/text attributes in .gitattributes, or your filter configuration
13:38 jaafar joined #git
13:38 govg joined #git
13:38 MattMaker joined #git
13:42 troulouliou_div2 joined #git
13:44 planet_en joined #git
13:44 niels__ joined #git
13:45 rnsanchez_ joined #git
13:45 morenoh149 joined #git
13:45 crelix joined #git
13:46 encod3 joined #git
13:50 prawnzy joined #git
13:50 ochorocho__ joined #git
13:50 AaronMT joined #git
13:52 inco6 joined #git
13:55 ouxg joined #git
13:55 ochorocho__ joined #git
13:58 MattMaker joined #git
13:58 patarr joined #git
13:59 Darren__ joined #git
13:59 Rodya_ joined #git
14:00 phaleth joined #git
14:02 raspado joined #git
14:04 zeroed joined #git
14:05 afuentes joined #git
14:05 mehola joined #git
14:08 Mindfart joined #git
14:08 ochorocho__ joined #git
14:08 afuentes joined #git
14:08 planet_en joined #git
14:09 MattMaker joined #git
14:13 dmpeters63 joined #git
14:14 MattMaker joined #git
14:14 iamthad joined #git
14:14 andlabs joined #git
14:15 dmpeters63 joined #git
14:17 re1 joined #git
14:18 ochorocho__ joined #git
14:18 crose joined #git
14:20 mischat joined #git
14:22 andlabs joined #git
14:24 hobodave joined #git
14:24 rgrinberg joined #git
14:24 Triq How are branches normally organized in .git/ref? The reason I'm asking is that when cloning from svn they end up in .....origin/<branch> but tags end up in ....origin/tags/<tag>.
14:25 js-acolyte joined #git
14:26 nedbat Triq: does that affect the git behavior?
14:26 psprint_ joined #git
14:27 ResidentBiscuit joined #git
14:27 Triq nedbat: I don't know. Would assume it doesn't. Just curious if there is a way it's usually done like the structure SVN usually has.
14:27 ResidentBiscuit joined #git
14:27 ResidentBiscuit joined #git
14:28 gdqo joined #git
14:28 ResidentBiscuit joined #git
14:28 MattMaker joined #git
14:28 ResidentBiscuit joined #git
14:29 jast Triq: git-svn does special things to account for the fact that in svn, tags are mutable, whereas in git they're not. the normal layout is explained in man gitrepository-layout
14:29 gitinfo Triq: the gitrepository-layout manpage is available at http://jk.gs/gitrepository-layout.html
14:29 snowkidind joined #git
14:29 inco6 left #git
14:29 inco6 joined #git
14:30 jast to add to that: git doesn't distinguish between local and remote tags the way it does for branches... so you'll have refs/tags/foo but not refs/remotes/origin/tags/foo, but you can have refs/heads/bar and refs/remotes/origin/bar
14:31 fahadash joined #git
14:31 watersoul joined #git
14:32 ochorocho__ joined #git
14:32 troulouliou_div2 joined #git
14:32 actinide joined #git
14:33 rorro joined #git
14:33 actinide joined #git
14:34 fuchstronaut joined #git
14:35 dwsr Hey all, I want to make some commits part of different branches in my GitHub respository. Originally I had all the commits in a single branch, but they all touch separate files and now I want them in two separate branches. Is it possible to base the branches off of the initial time I created the repository and then move the various commits to each branch?
14:38 sarbs joined #git
14:38 WayToDoor joined #git
14:38 Triq jast: Ok, thank you.
14:39 MattMaker joined #git
14:41 wrouesnel1 joined #git
14:41 ochorocho__ joined #git
14:46 Rodya_ joined #git
14:47 inco6 left #git
14:47 inco6 joined #git
14:47 Lomex joined #git
14:47 ochorocho__ joined #git
14:50 inco6 joined #git
14:50 yyzr joined #git
14:50 inco6 joined #git
14:51 inco6 joined #git
14:51 inco6 joined #git
14:54 MattMaker joined #git
14:55 inco6 joined #git
14:55 inco6 joined #git
14:55 encod3 joined #git
14:55 inco6 joined #git
14:56 tlaxkit joined #git
14:56 satbyy joined #git
14:56 stomplee joined #git
14:56 Anupkumar joined #git
14:57 McNozes joined #git
14:57 inco6 joined #git
14:57 robotroll joined #git
14:58 inco6 joined #git
14:58 inco6 joined #git
14:58 inco6 joined #git
14:58 inco6 joined #git
14:59 inco6 joined #git
14:59 inco6 joined #git
14:59 inco6 joined #git
14:59 inco6 joined #git
14:59 inco6 joined #git
15:00 inco6 joined #git
15:01 inco6 joined #git
15:01 inco6 joined #git
15:02 inco6 joined #git
15:02 inco6 joined #git
15:02 ahmedelgabri joined #git
15:02 inco6 joined #git
15:02 inco6 joined #git
15:03 eivarv joined #git
15:06 shgysk8zer0 joined #git
15:06 dreiss joined #git
15:07 ghostlight joined #git
15:08 jncog joined #git
15:09 jncog joined #git
15:09 MattMaker joined #git
15:09 jncog joined #git
15:09 thebope joined #git
15:09 actinide joined #git
15:10 nandi_ge___ joined #git
15:10 huggy joined #git
15:11 GodGinrai joined #git
15:12 InfoTest joined #git
15:12 niax joined #git
15:13 tyreld joined #git
15:14 inkog joined #git
15:15 planet_en joined #git
15:15 GodGinrai joined #git
15:15 actinide joined #git
15:16 subhojit777 joined #git
15:19 dviola joined #git
15:21 govg joined #git
15:21 Sleepy63 joined #git
15:24 MattMaker joined #git
15:24 WayToDoor joined #git
15:24 imMute joined #git
15:24 matsaman joined #git
15:25 clr joined #git
15:27 Milijus joined #git
15:27 IChoseThisOne joined #git
15:29 clr joined #git
15:30 le_melomane joined #git
15:33 dwsr Hey all, I want to make some commits part of different branches in my GitHub respository. Originally I had all the commits in a single branch, but they all touch separate files and now I want them in two separate branches. Is it possible to base the branches off of the initial time I created the repository and then move the various commits to each branch?
15:34 _ikke_ dwsr: I think the easiest way is to work with interactive rebase
15:34 rorro joined #git
15:34 morenoh149 joined #git
15:37 ochorocho__ joined #git
15:37 dwsr _ikke_: I forked an existing repo. Interestingly enough, I can't rebase the branch to the point in time where I forked the repo.
15:37 dwsr Unless I'm missing something.
15:38 MattMaker joined #git
15:38 marlinc joined #git
15:41 cdg joined #git
15:41 antlarr joined #git
15:42 dwsr Can I rebase an entire branch?
15:42 dwsr The branch is currently empty (i.e. no commits)
15:42 inco6 joined #git
15:43 osse dwsr: in that case it's easier to recreate it where you want ti
15:43 osse it
15:43 nikivi joined #git
15:43 quizzi joined #git
15:43 marlinc joined #git
15:45 tabakhase hm, so is gitlabEnterprise still the only thing that can do "rebase before merge"?
15:45 dwsr Can I create a branch from a specific commit?
15:45 tabakhase looks like nothings been happening there in the last 6-12 months :/
15:45 tabakhase dwsr git checkout <commithash>; git checkout -b BRANCH; -- done.
15:46 planet_en joined #git
15:47 _ikke_ tabakhase: How does gitlab handle conflicts when rebasing?
15:47 Gsham joined #git
15:47 ochorocho__ joined #git
15:47 tabakhase _ikke_ it does not, whats fine -- bitching at the dev to update it is all it needs todo
15:47 tabakhase its about the clean cases where its posssible without isssue^^
15:47 quizzi joined #git
15:48 Nex6 joined #git
15:49 MattMaker joined #git
15:50 dwsr How can I move a commit between branches?
15:50 osse dwsr: cherry-pick to make a copy, then use eg. reset to remove the original
15:51 _ikke_ reset if it's the last commit, rebase if not
15:51 rominronin joined #git
15:51 kpease joined #git
15:51 prawnzy joined #git
15:51 EvilPeng1 joined #git
15:51 alex1a joined #git
15:52 NotSure joined #git
15:52 planet_en joined #git
15:54 thebope joined #git
15:54 Electrometro joined #git
15:54 |prawnzy joined #git
15:54 Gsham joined #git
15:56 ochorocho__ joined #git
15:56 le_melomane joined #git
15:56 dwsr Eh, I just blew away the repository and started over. It was all of like 7 commits.
15:56 PioneerAxon joined #git
15:58 eivarv joined #git
15:59 danslo joined #git
16:00 ochorocho__ joined #git
16:00 synthroid joined #git
16:01 morenoh149 joined #git
16:02 guampa joined #git
16:02 heptadecagram joined #git
16:02 madewokherd joined #git
16:03 MattMaker joined #git
16:05 overlord_tm joined #git
16:05 thiago joined #git
16:06 mehola joined #git
16:08 DieguezZ joined #git
16:08 clr anybody have any thoughts on how to merge without a work tree?  if the merge has conflicts that cannot be automatically resolved, fail.  i've written scripts to do this but seems like this could just be a useful mode for git-merge.  is this possibly better supported than i realize?
16:10 clr my use case is integration builds, i.e, merge as many branches together as possible and build the result.  basically linux-next w/o manual intervention
16:11 icedp joined #git
16:11 Rodya_ joined #git
16:11 lagothri1 joined #git
16:14 mgolawala joined #git
16:14 MattMaker joined #git
16:14 bremner clr: when I played with this, I eventually decided life was to short, and local clones very fast, and just make a temporary clone with a working directory
16:14 bremner that wasn't a repo the size of linux though
16:17 durham joined #git
16:18 G__81 joined #git
16:18 clr bremner: yeah, even with a big repo have a work tree isn't a big deal.  iirc, you can point git to an empty directory for the work tree when merging and it will create files with guards for only things that had conflicts, so easy enough to just detect that the merge didn't succeed and continue on.  performance of this was very good.
16:19 psprint In git log is it possible to show (origin/master) instead of (HEAD -> master, origin/master) ?
16:20 shgysk8zer0 joined #git
16:20 G__81 i have a question pertaining to sending patches. I had modified 5 files and sent for review by creating my own branch. Now i feel the changes made in files 4 and 5 are not needed and i have also addressed comments in files 1 to 3. Now I just need to send file 1,2 and 3 changes but i want the original 4 and 5 file unchanged. i.e i dont want 4 and 5 file part of my patch v2. How do i address this
16:21 danslo joined #git
16:21 alex1a_ joined #git
16:24 poohbear82 joined #git
16:24 eivarv joined #git
16:24 G__81 can someone help me out with my question pls
16:25 NeXTSUN joined #git
16:26 clr bremner: i've also scripted an index-only merge.  it works, but it's reinventing a lot of the wheel and surely is not as robust with errors as git-merge.  also, i think was just calling the recursive driver directly.  so, like i said, seems like it could be a nice mode for git-merge, but maybe i'm missing something.
16:26 poohbear82 why is it that everytime i commit it says that the HEAD is detached?!
16:26 poohbear82 I don't need to create a new branch everytime I commit, do i?!
16:26 re1_ joined #git
16:27 kpease joined #git
16:28 konon joined #git
16:28 Balliad joined #git
16:28 clr G__81: you can make the updates and run `git commit --amend'.  That fold the changes into the original commit.
16:29 G__81 I had modified 5 files and sent for review by creating my own branch. Now i feel the changes made in files 4 and 5 are not needed and i have also addressed comments in files 1 to 3. Now I just need to send file 1,2 and 3 changes but i want the original 4 and 5 file unchanged. i.e i dont want 4 and 5 file part of my patch v2. How do i address this
16:29 Lomex joined #git
16:29 MattMaker joined #git
16:29 clr G__81: or if you've already committed the changes on top of the old, you can use interactive rebase to combine the commits into one
16:29 fedepad joined #git
16:30 VladGh joined #git
16:30 nitric joined #git
16:31 clr poohbear82: you need to checkout the branch by name.  you're currently detached, meaning HEAD is pointing directly to a commit, not at a branch name (which would point to a commit)
16:32 clr poohbear82: try something like: git checkout -b my_branch
16:32 G__81 clr: i have committed 5 files in my branch and now i am gonna commit v2 for first 3 files. For 4 and 5 files can i do git reset ?
16:33 clr G__81: yep, if you want your patch to leave those files alone you can do something like: git reset HEAD~ -- file4 file5
16:33 clr G__81: ... assuming the changes you want to back out are in the most recent commit.
16:34 garethdaine joined #git
16:34 G__81 yes its in my latest commit and i want to go back to prev->prev commit for files 4 and 5 alone
16:35 clr poohbear82: that will give the commit you're on the branch name 'my_branch' and point HEAD to it (i.e., make it the current branch)
16:36 clr G__81: prev->prev?  maybe you want HEAD~2 then?
16:37 VladGh joined #git
16:37 jbitdrop joined #git
16:38 JamesBaxter joined #git
16:39 ansimation joined #git
16:40 dan2k3k4 joined #git
16:40 poohbear82 clr, the problem is that i was already on a branch! when i commited, it created a commit and the head was detached. Doesn't make sense to me. So i created another branch and committed that. Now when i do a git push it says everything is up to date, but doesn't push the new branch
16:42 danslo joined #git
16:42 dalias joined #git
16:43 dalias are there any lock files git uses internally other than index.lock?
16:44 dalias i'm trying to synthesize a repo history programmatically with git add/write-tree using a non-default index file so that trees can be synthesized concurrently
16:44 dalias but it seems like some of the adds are failing
16:44 MattMaker joined #git
16:44 armyriad joined #git
16:45 niels__ joined #git
16:45 ochorocho__ joined #git
16:46 clr poohbear82: well, if you were detached after the commit, you were probably detached before it, too, for some reason.  use git-prompt to stay sane.  it gives you important state of repo in your prompt to help you notice when that chnages from what you expect.
16:46 m1m3-50 joined #git
16:47 niels__ joined #git
16:47 t4nkd joined #git
16:48 Apteryx joined #git
16:48 Apteryx joined #git
16:49 LucaTM joined #git
16:50 clr dalias: are you seeing any errors?
16:50 eivarv joined #git
16:51 fuchstronaut joined #git
16:51 dalias clr, it's just a few trees/commits that got botched out of a big 30+ min run and somehow my logging didn't save stderr like it was supposed to
16:51 rominronin joined #git
16:51 dalias (i just re-ran it with logging but the logging didn't work :-p)
16:52 tang^ joined #git
16:52 dalias so i'm not seeing errors but they likely happened
16:53 Chloe joined #git
16:53 Chloe and aquarium keeping?
16:54 Chloe Is there a way to get a list of all the contributors (and their emails) who have code in the current state?
16:55 clr Chloe: git log --format='%an <%ae>' | sort -u
16:55 Chloe and that is for the current HEAD only?
16:56 clr Chloe: yes
16:56 Chloe I assumed it would include all history as well, ok
16:56 navidr joined #git
16:56 clr Chloe: well, all history reachable from HEAD
16:56 dalias chloe might want to exclude contributors whose code was removed
16:56 dalias thats a hard problem tho
16:57 Chloe yes, that's what I'd like
16:57 dalias maybe you could run a git blame across all files and process the results somehow
16:57 clr Chloe: you can run `git blame' on all current files
16:59 dalias this might help catch the bug:
16:59 dalias test "$out" = "4b825dc642cb6eb9a060e54bf8d69288fbee4904" && exit 1
16:59 Balliad joined #git
16:59 AlexxNica joined #git
16:59 rorro joined #git
17:00 lindenle joined #git
17:00 dalias that's a useful sha1 to know :)
17:00 mikecmpbll joined #git
17:02 SteffanW joined #git
17:03 MattMaker joined #git
17:06 lindenle joined #git
17:07 nikivi joined #git
17:07 _ADN_ joined #git
17:08 SwiftMatt joined #git
17:11 snthsnth joined #git
17:11 Salander27 joined #git
17:12 le_melomane joined #git
17:13 Chloe well, I couldn't find anything, so I posted it on SO (if anyone has a solution later) https://stackoverflow.com/questions/39645035/h​ow-to-get-all-authors-for-current-state-of-git
17:15 fuchstronaut joined #git
17:15 tyreld joined #git
17:18 MattMaker joined #git
17:19 sbeller_ joined #git
17:20 incog joined #git
17:23 sylario joined #git
17:23 gfixler1 joined #git
17:26 Repox joined #git
17:29 t4nk642 joined #git
17:29 blackwind_123 joined #git
17:30 t4nk642 ayy
17:31 anonymuse joined #git
17:31 psprint What cool things can be done with a commit, besides checkout?
17:32 hahuang65 joined #git
17:33 MattMaker joined #git
17:36 armin joined #git
17:37 Prisoner-2460_1 joined #git
17:38 mischat joined #git
17:38 shgysk8zer0 joined #git
17:39 shgysk8zer0 joined #git
17:41 barteks2x joined #git
17:41 subhojit_ joined #git
17:41 ahmedelgabri joined #git
17:44 danslo_ joined #git
17:45 nedbat does rerere help with rebase conflicts as well as merge conflicts?
17:46 snthsnth joined #git
17:46 Rodya_ joined #git
17:46 Sasazuka joined #git
17:48 MattMaker joined #git
17:49 nd joined #git
17:49 OnBrokenWings joined #git
17:49 heftig joined #git
17:50 le_melomane joined #git
17:52 chris2 joined #git
17:52 cdown joined #git
17:53 invisbl joined #git
17:53 LionsMane joined #git
17:53 EricPoe joined #git
17:57 js-acolyte joined #git
17:57 Vmmus joined #git
17:58 multi_io joined #git
17:58 inco6 joined #git
17:58 dreiss joined #git
17:58 WayToDoor joined #git
18:00 re1 joined #git
18:01 vassagus joined #git
18:01 sangy joined #git
18:01 inco6 left #git
18:02 raspado joined #git
18:03 MattMaker joined #git
18:04 prawnzy joined #git
18:04 LeBlaaanc joined #git
18:06 danslo_ joined #git
18:07 mehola joined #git
18:08 e14 joined #git
18:11 pokerpoker joined #git
18:13 js-acolyte joined #git
18:14 mikecmpbll joined #git
18:15 ShekharReddy joined #git
18:15 zacts joined #git
18:15 MattMaker joined #git
18:16 anonymuse joined #git
18:16 rnsanchez_ joined #git
18:17 dsantiago joined #git
18:18 Atm0spher1c joined #git
18:19 LionsMane joined #git
18:20 rgrinberg joined #git
18:23 _rnsanchez_ joined #git
18:23 hanthings joined #git
18:24 red82 joined #git
18:25 mlvnd joined #git
18:27 danslo_ joined #git
18:27 MattMaker joined #git
18:28 ResidentBiscuit Yes it does
18:29 tmk joined #git
18:29 tremblerz joined #git
18:29 snthsnth joined #git
18:29 tremblerz Guys, How can I make other users also to appear as contributor in my github repo
18:30 fuchstronaut joined #git
18:30 tremblerz There profile comes in the github commit
18:30 tremblerz But contributor list just consists of one person
18:30 Walakea_ joined #git
18:32 MattMaker joined #git
18:33 aard_ joined #git
18:33 KevinMGranger Did you merge any of their PRs onto master, or whatever your main branch is? I think that's what the list reflects
18:34 dreiss joined #git
18:34 pokerpoker when i do a git show <hash> it pipes the results into an editor.  I want to dump to stdout
18:35 clr pokerpoker: git --no-pager show
18:38 pokerpoker ty!
18:38 Mobutils joined #git
18:39 mrwat joined #git
18:39 tremblerz KevinMGranger: I added her directly as collaborator in settings
18:39 tremblerz So that she can directly make changes into master
18:41 Vampire0_ joined #git
18:41 ahmed_elgabri joined #git
18:41 clr tremblerz: so she probably just needs to contribute to show up as a contributor
18:42 tremblerz clr: Yeah she did, her commits in the master can be seen.
18:42 joki joined #git
18:43 ResidentBiscuit Might ask on #github
18:44 danimal2 joined #git
18:46 Vampire0 tremblerz, which repo is this about?
18:47 tremblerz https://github.com/chrizandr/semantic_web
18:48 snthsnth joined #git
18:48 ResidentBiscuit Oh, I think her user.name and/or user.email are configured wrong
18:48 ResidentBiscuit IIRC they need to match github
18:48 eivarv joined #git
18:49 tremblerz Oh thanks
18:49 m1m3-50 joined #git
18:49 thebope hello all, I'm following this - https://help.github.com/art​icles/changing-author-info/
18:50 thebope except instead of cloning master, I cloned a branch
18:50 thebope would the end result
18:50 danslo_ joined #git
18:50 thebope git push --force --tags origin 'refs/heads/*'
18:50 mingrammer joined #git
18:50 thebope know to push to the proper branch? I'm a little out of my gitfu league at the moemnt
18:51 re1 joined #git
18:51 thebope Or can I do "git push --force --tags origin MYBRANCH 'refs/heads/*'"?
18:52 MattMaker joined #git
18:52 thebope Okay I tried it with MYBRANCH and it worked well
18:53 irco joined #git
18:53 chelly joined #git
18:54 cagedwisdom joined #git
18:56 russoneto joined #git
18:57 nikivi joined #git
18:57 e14_ joined #git
18:57 saml joined #git
18:57 russoneto how can I "go back" a few commits, and from that point move it to another branch?
18:57 albel727 joined #git
18:57 saml i branched off from wrong branch.  how can I change it so that my branch looks like it's branched off from master branch?
18:58 ResidentBiscuit russoneto: You can checkout whatever commit you want
18:58 ResidentBiscuit `git checkout <hash>` or `git checkout HEAD~n` where n is the number of commits you want to go back
18:58 clr russoneto: git checkout -b new_branch_name @~3 (if few == 3)
18:59 ProbabilityMoon joined #git
18:59 fuchstronaut joined #git
18:59 jaafar joined #git
19:00 OSryx joined #git
19:01 saml i'm on detached head. how can I force set abranch to this detached head?
19:01 russoneto ResidentBiscuit clr yes but if I checkout a specific commit, I wont have the commits beyond that
19:02 saml i did  git checkout abranch;  git rebase --onto master sha^ HEAD;    now I want abranch to point to detached
19:04 ResidentBiscuit russoneto: Of course you won't, you've gone back in time. I'm not really sure what you need then
19:04 clr russoneto: maybe i'm not understanding.  you can `git checkout -b new_branch_naame' and then `git rebase -i @~10'.  Then delete the commits you don't want in your new branch.
19:04 monde joined #git
19:04 russoneto I want to get X commits behind and MOVE it to another branch (and remove from master, of course)
19:04 phoenix_p13d joined #git
19:05 ResidentBiscuit Oh
19:05 ResidentBiscuit That's easier
19:05 fahadash joined #git
19:05 ResidentBiscuit Create a new branch from master, reset --hard master back to where you want it to be
19:05 russoneto yeah
19:05 russoneto tks
19:06 thebope_ joined #git
19:06 danslo joined #git
19:06 clr saml: you can run `git checkout -b branch_name' to turn your detached state into a branch
19:06 vassagus joined #git
19:06 saml clr, I want to use existing branch name, abranch
19:07 saml git checkout -B abranch
19:07 MattMaker joined #git
19:08 t0by joined #git
19:08 dviola joined #git
19:08 OSryx left #git
19:09 OSryx joined #git
19:10 Rodya_ joined #git
19:11 dodobrain joined #git
19:13 js-acolyte joined #git
19:13 liquidjamm joined #git
19:14 heroux joined #git
19:15 nidr0x joined #git
19:16 CrazyBonz joined #git
19:19 dodobrain joined #git
19:20 obiwahn joined #git
19:20 Geminizer joined #git
19:21 obiwahn hi how do i reset a single file to the state it was in when i ran into a mergeconflict
19:21 Geminizer Hello everyone.  How can one set up git-http-backend using nginx on a windows box?
19:21 js-acolyte joined #git
19:21 pokerpoker if i have done git reset <sha1> --hard, is there a way to undo it?
19:21 danslo joined #git
19:21 afuentes joined #git
19:21 LionsMane joined #git
19:22 Flaghacker joined #git
19:22 Prisoner-2460_1 no way to undo —-hard that i’m aware of
19:22 obiwahn pokerpoker: yes
19:22 preaction pokerpoker: depending on how long it's been, check the !reflog
19:22 gitinfo pokerpoker: The git reflog (`git log -g`) records the SHAs of your HEADs for 2+ weeks. `git checkout -b myrestore OLDSHA` and `git reset --hard OLDSHA` will relink to that state via a new and current branch respectively, see http://sethrobertson.github.com/GitFixUm/ for full details.  WARNING: reset --hard will trash any uncommitted changes!  Visualize with: gitk --all --date-order `git log -g --pretty=%H`
19:22 rgrinberg joined #git
19:23 obiwahn preaction: do a "git reflog" and find your lost head
19:23 Flaghacker Is there a way to create a temporary "backup" of my current code using git? I'm implementing something new and I would like to try it out in two different ways.
19:23 rahtgaz joined #git
19:24 Prisoner-2460_1 try git stash
19:24 Flaghacker I'll take a look, thanks!
19:25 liquidjamm joined #git
19:25 preaction Flaghacker: or make a branch
19:27 MattMaker joined #git
19:27 Flaghacker preaction, I've never used branches before and they look quite complicated... I have to learn about them later though.
19:28 eivarv joined #git
19:29 Flaghacker Wait git stash reverts my repository to the previous HEAD? I don't want that, so I guess I'll have to use branches, right?
19:30 mlvnd joined #git
19:31 sin3t joined #git
19:31 dodobrain in a given dir, how do i list files that are *not* available in a specified tagged commit?
19:32 clr Flaghacker: either should work, though branches might be a little easier to manage.  just name your branches descriptively, and when you find the right solution you can delete any branches you no longer need.
19:32 dodobrain eg: in dir 'mig' want to list files that are not present in tag '1.0
19:32 chandu20 joined #git
19:32 Flaghacker clr, And merge the "good" branch into master?
19:33 Flaghacker Also, what kind of name should I use? I have no idea what the convention is.
19:33 heftig joined #git
19:33 planet_en joined #git
19:34 red82 joined #git
19:34 clr dodobrain: git diff --diff-filter=A --name-only 1.0 HEAD
19:34 ash_workz joined #git
19:36 dodobrain err.. clr thats showing some other files that aren;t even in the current dir!
19:37 clr Flaghacker: yes, once you have a branch that works you can check out master and merge it in.  i normally use lowercase names with hyphens, e.g., my-useful-feature
19:37 dodobrain and more importantly also not showing files in the current dir that aren;t in the 1.0 tagged commit
19:37 MattMaker joined #git
19:38 danslo joined #git
19:38 clr dodobrain: add " -- ." to the end to get current directory.  are the files that are not showing up untracked?
19:38 dodobrain clr, ah bleh.. i had to add the dir at the end
19:39 dodobrain heh, thanks
19:39 m0viefreak joined #git
19:39 boombatower joined #git
19:39 dodobrain clr, actually yeah.. the untracked ones aren;t showing up
19:39 dodobrain i need those too
19:40 e14_ joined #git
19:41 franl joined #git
19:41 vassagus joined #git
19:42 shgysk8zer0 joined #git
19:43 albel727 joined #git
19:43 inflames joined #git
19:43 clr git ls-files --others --directory
19:44 clr dodobrain: that was for you, that will list untraced files.  again, add " -- ." for current directory
19:44 planet_en joined #git
19:44 dodobrain is it not possible to combine the 2?
19:45 psprint Initial commit's message: "Commit committed"
19:45 mingrammer joined #git
19:45 dodobrain i want everything (apart from ignored files) that is in the current dir that is modified from 1.0 or untracked compared to 1.0
19:46 AlexxNica joined #git
19:46 obiwahn anyone? how do i get a file back to the state at the begin of a merge conflict?
19:46 clr dodobrain: you could add all the files in the directory and commit, then do the diff.  git doesn't generally care much about untracked files, though, so i don't think there's a single command for this.
19:47 dodobrain the seconds command is showing me the untracked files but is including the ignored files
19:47 dodobrain i want to not have the ignored files in this list
19:48 Megaf joined #git
19:48 clr dodobrain: create a temporary commit and then diff, git won't add ignored files and therefore should get you what you want.
19:49 hobodave joined #git
19:50 ResidentBiscuit obiwahn: You can always `git merge --abort` if you need to
19:52 rominron_ joined #git
19:52 snthsnth joined #git
19:53 e14_ joined #git
19:53 mikecmpbll joined #git
19:54 crayon joined #git
19:54 danslo joined #git
19:54 crayon joined #git
19:57 Jellyg00se joined #git
19:57 thiago joined #git
19:57 e14_ joined #git
19:57 chandu20 joined #git
19:58 p4trix joined #git
19:58 drizztbsd joined #git
19:58 stamina joined #git
19:59 WayToDoor joined #git
19:59 yazdmich joined #git
20:00 dalias clr, uhg, so i found the problem
20:00 kulelu88 joined #git
20:01 dalias git_mkstemp_mode is broken
20:01 dalias it generates the temp file name deterministically based on low-res time (gettimeofday)
20:01 dalias and if it has a collision, it simply adds += 7777 and tries again
20:02 MattMaker joined #git
20:02 fuchstronaut joined #git
20:02 fuchstronaut joined #git
20:03 jokester joined #git
20:04 johnmilton joined #git
20:04 stfn joined #git
20:04 ResidentBiscuit Pretty good logic
20:05 drizztbsd joined #git
20:05 dalias i'm not sure _how_ this is colliding, because even tho it's bad it still seems pretty hard
20:05 dalias but the temp file creation is failing with EEXIST
20:05 clr obiwahn: to restore the file with guards without aborting the merge, you should be able to do something like this: git merge-file -p $(git unpack-file :2:"$f") $(git unpack-file :1:"$f") $(git unpack-file :3:"$f")
20:06 user123abc joined #git
20:06 clr obiwahn: ... assuming $f is the file you care about.  the above will dump the content to stdout, so you can try this out and verify.
20:06 dalias it should just use the libc mkstemp
20:07 Vampire0 joined #git
20:07 MattMaker joined #git
20:07 mehola joined #git
20:08 user123abc I want to filter-branch to change the emails associated with a submodule branch. Then I want to somehow update the submodule references of all commits in the parent project to point to the rewritten version of the commit they reference.
20:08 user123abc I see an SO post talking about the process:
20:08 user123abc http://stackoverflow.com/questions/3​7442500/repository-with-submodules-a​fter-rewriting-history-of-submodule
20:09 user123abc does anybody have any recommendations to add to that?
20:09 hobodave joined #git
20:09 user123abc "rewritten version of the submodule commit", I mean
20:10 clr dalias: what is it trying to create a temp file for?  can you tell?  or at least which git command is trying to do it?
20:10 dalias user123abc, it's awful :)
20:10 dalias it's similar to what i'm doing
20:10 dalias clr, for adding objects
20:10 dalias git add
20:11 dalias the way my script works is to git add . with a local index file and local work tree, then git write-tree from that local index file
20:11 dalias so it's parallelizable
20:11 dalias (it's actually run from make -jN :)
20:11 Rodya_ joined #git
20:11 Flaghacker When I create a new branch with "git branch [name]", what will happen to the unstaged files and at what branch will I be?
20:11 ramsub07 joined #git
20:11 _ikke_ Flaghacker: git branch does not touch any files in the working tree
20:12 _ikke_ It only creates a new branch (a ref, which is a pointer to a commit)
20:12 js-acolyte joined #git
20:12 dalias i fixed my script so it errors out rather than generating an empty tree if git add fails
20:12 danslo joined #git
20:13 dalias so now i just have to rerun make if/when it bails with EEXIST
20:13 dalias user123abc, what i'm doing is actually similar to what you're trying to do
20:13 dalias except my input repo and subrepos come from hg (eew) and i'm merging them into a single git repo (no submodules)
20:14 user123abc dalias, I see. Now I suddenly think I should read more about it before I can even ask meaningful questions
20:14 js-acolyte joined #git
20:14 Flaghacker _ikke_, And how would I commit something to the new branch?
20:14 clr dalias: hmm, well you could try using update-index.  that shouldn't create any temporary files and give you total control, though it would require some scripting.
20:14 user123abc for instance, it mentions refs/originals and a way to get a mapping of old->new commits, but I see only a single hash there
20:14 mycelium joined #git
20:14 ToxicFrog Flaghacker: git checkout <branch>; do stuff; git add; git commit
20:14 dalias clr, well then i have to implement my own git add :-p
20:15 ToxicFrog Flaghacker: you can also "git checkout -b <new-branch-name>" to create a new branch and make it active in one step, rather than "git branch foo; git checkout foo"
20:15 dalias clr, i.e. recurse the working tree and add a git object for each file
20:15 _ikke_ Flaghacker: Note that uncomitted changes do not belong to any branch
20:15 user123abc dalias, also, do you mean that post is awful or the problem is awful?
20:15 fstd joined #git
20:15 dalias user123abc, the problem is awful :)
20:15 user123abc yeah
20:15 Flaghacker Okay thanks!
20:15 obiwahn thank you clr!
20:15 user123abc I wish I had just fixed my author settings before crapping all over my net of submodules
20:16 js-acolyte joined #git
20:16 clr dalias: or maybe just try the add again if it fails?
20:16 eight58 joined #git
20:16 Flaghacker What would happen with uncommited/unstaged changes if I switched to another branch?
20:17 _ikke_ nothing
20:17 Pasi joined #git
20:17 js-acolyte joined #git
20:18 _ikke_ if git would need to overwrite those files (because they are tracked in the other branch, git would complain
20:18 dreiss joined #git
20:19 e14 joined #git
20:19 dalias clr, yeah that sounds like a decent idea
20:19 jaafar joined #git
20:19 dalias after rm'ing the index in case it was partially updated..?
20:20 * dalias wonders how many ppl here cringe at "rm'ing the index" :-)
20:20 clr dalias: probably a good idea
20:21 dalias -git add . 1>&2
20:21 clr dalias: well, probably unnecessary.  if it was partially updated, it should only include stuff you want.  but starting from scratch probably doesn't hurt much.
20:21 dalias +until git add . 1>&2 ; do rm "$GIT_INDEX_FILE" ; done
20:22 dalias i could put a retry limit and sleep just in case it fails for some other reason
20:22 dalias so it doesn't hammer the disk for hours until i notice :)
20:24 AlexxNica joined #git
20:24 Flaghacker joined #git
20:25 Seveas dalias: why the 1>&2?
20:26 dalias iirc git add can write stdout; maybe i'm wrong
20:26 dalias just being safe
20:26 dalias because the output of the script has to be the output of git write-tree (the tree sha1)
20:26 Seveas ah
20:26 dalias anything else in the output will make it unusable
20:27 rhizome joined #git
20:27 cdg joined #git
20:28 danslo joined #git
20:28 dalias <3 make -j4 on this
20:28 mrpackethead joined #git
20:29 clr dalias: it's weird you are failing on that.  looking at somewhat recent source it should try to create the tmp file 16384 times before giving up.  do you somehow have that many conflicting files in /tmp?
20:29 dalias 33min to convert a job that used to take hours
20:29 dalias clr, i think the issue is how many files are being added
20:29 dalias unless they're all renamed before the next one is made
20:30 dalias like i said it's hard for me to understand _why_ these rare collisions happen
20:31 dalias but the error message in the log was EEXIST making temp file
20:35 nwkj86 joined #git
20:36 duderonomy joined #git
20:37 nilez joined #git
20:38 Flaghacker left #git
20:38 Flaghacker joined #git
20:39 Flaghacker joined #git
20:39 Flaghacker I dig "git checkout -b <name>" and it showed some lines staring with an M, eg. "M       build.gradle". What do those mean?
20:39 nidr0x joined #git
20:40 planet_e1 joined #git
20:40 Seveas that the file is modified
20:41 thebope joined #git
20:42 john__ joined #git
20:43 lolmac joined #git
20:43 e14 joined #git
20:43 Flaghacker And how so? Shouldn't that command simply create a new branch?
20:45 john__ hi all! i have a question regarding http.sslCAInfo and http.sslCAPath. can they both be configured at the same time or does one override the other?
20:45 danslo joined #git
20:47 thebope joined #git
20:49 e14_ joined #git
20:50 thiago john__: doesn't your system have those settings?
20:50 Keridos joined #git
20:51 hahuang61 joined #git
20:51 nedbat Flaghacker: you must have modified them before you made the branch.  Making the branch didn't change them.
20:52 airdisa joined #git
20:52 Flaghacker nedbat, Okay but did it change anything or is it just saying that I modified them relative to something?
20:53 nedbat Flaghacker: it is saying that those files are modified compared to the local repository, there are changes you haven't committed yet
20:53 Flaghacker And why is it not listing untracked files?
20:54 sparetire joined #git
20:54 nedbat Flaghacker: i don't know
20:54 nedbat Flaghacker: it doesn't.
20:54 elastix joined #git
20:55 thebope when I do git rebase -i -p it squashes all my merges into single lines
20:55 thebope is there a way to do that with git log?
20:56 thebope hmm I can try the author command
20:56 thebope That works~
20:56 thebope !*
20:57 Rodya_ joined #git
20:59 Mobutils joined #git
20:59 hahuang65 joined #git
21:00 Flaghacke joined #git
21:00 Flaghacke left #git
21:00 anuxivm joined #git
21:00 Flaghacke joined #git
21:00 OSryx left #git
21:01 hahuang62 joined #git
21:01 UserOO7 joined #git
21:03 hahuang61 joined #git
21:04 roelj joined #git
21:04 rorro joined #git
21:05 dreiss joined #git
21:05 roelj joined #git
21:07 roelj How can I sign multiple commits (ony-by-one is fine though) before I push them?
21:08 t0by joined #git
21:09 thiago what's the problem with signing each of them?
21:11 roelj thiago: I already committed them without -S (because I didn't have my key at that point).
21:11 roelj I didn't push the commits though, so I can rewrite the history
21:12 thebope joined #git
21:12 hahuang65 joined #git
21:13 raijin joined #git
21:13 strugee joined #git
21:14 thiago roelj: just re-commit, one by one
21:14 ResidentBiscuit Might be able to do it as part of a rebase
21:15 danslo joined #git
21:16 roelj Aha! I just did git rebase -i --root. Then changed 'pick' to 'edit' for the commits I wanted to sign.  Then ran `git commit -S --amend --no-edit' on each of them.
21:16 ayjay_t joined #git
21:17 prawnzy joined #git
21:17 diogomr joined #git
21:18 vassagus joined #git
21:18 franl joined #git
21:23 mingrammer joined #git
21:25 elastix joined #git
21:29 olalonde joined #git
21:31 obiwahn clr: where can i read about the :1:filename syntax?
21:31 danslo joined #git
21:31 obiwahn How is it called?
21:31 sr___ joined #git
21:32 ResidentBiscuit I think you could have done something like `git rebase -root -x "git commit -s --amend --no-edit"`
21:34 SwiftMatt joined #git
21:35 LionsMane joined #git
21:35 AaronMT joined #git
21:37 roelj ResidentBiscuit: That would've automated the recommitting parts, right?
21:38 t0by joined #git
21:38 dviola joined #git
21:38 mischat joined #git
21:38 lindenle joined #git
21:38 fedepad joined #git
21:38 icedp joined #git
21:38 ResidentBiscuit joined #git
21:38 psprint joined #git
21:38 gusnan joined #git
21:38 bouncy joined #git
21:38 Kaisyu joined #git
21:38 blr joined #git
21:38 Bombe joined #git
21:38 _ikke_ joined #git
21:38 Blkt joined #git
21:38 bencevans joined #git
21:38 tristanp joined #git
21:38 souther joined #git
21:38 ffabi joined #git
21:38 sea-gull joined #git
21:38 VoidFox joined #git
21:38 rasto joined #git
21:38 ohcibi joined #git
21:38 minibar joined #git
21:38 Hobbyboy joined #git
21:38 AlexanderS joined #git
21:38 kba joined #git
21:38 Chinorro joined #git
21:38 nemesit|znc joined #git
21:38 ToeSnacks joined #git
21:38 jokajak joined #git
21:38 sweatsuit joined #git
21:38 msonntag joined #git
21:38 sfan786_ joined #git
21:38 optraz joined #git
21:38 rcn joined #git
21:38 shennyg joined #git
21:38 wwalker joined #git
21:38 lucz joined #git
21:38 texasmynsted joined #git
21:38 Vaelatern joined #git
21:38 abh joined #git
21:38 anonnumberanon joined #git
21:38 Neo_ joined #git
21:38 hypera1r joined #git
21:38 drawkula joined #git
21:38 [cust] joined #git
21:38 rossome joined #git
21:38 antispaceman joined #git
21:38 creppe joined #git
21:38 asd5a joined #git
21:38 thecomedian joined #git
21:38 barq joined #git
21:38 rfv joined #git
21:38 Zarthus joined #git
21:38 relgar joined #git
21:38 nilsi joined #git
21:38 jerrcs joined #git
21:38 leprechau joined #git
21:38 Raed joined #git
21:38 batrick joined #git
21:38 discopatrick joined #git
21:38 etcinit joined #git
21:38 BinGOs joined #git
21:38 SupaHam joined #git
21:38 rishi` joined #git
21:38 huhlig joined #git
21:38 parliament joined #git
21:38 trig-ger joined #git
21:38 Celelibi joined #git
21:38 MLM joined #git
21:38 wizonesolutions joined #git
21:38 prosody joined #git
21:38 sebhoss joined #git
21:38 janx joined #git
21:38 adambrenecki joined #git
21:38 paperziggurat joined #git
21:38 pizzaops joined #git
21:38 da5id joined #git
21:38 artgoeshere joined #git
21:38 bathtub_shark joined #git
21:38 ozzloy joined #git
21:38 bittyx joined #git
21:38 iveqy_ joined #git
21:38 michael_mbp joined #git
21:38 jeffl35 joined #git
21:38 Zoxc joined #git
21:38 AKPWD joined #git
21:38 Tsutsukakushi joined #git
21:38 Guest92408 joined #git
21:38 Fiouz joined #git
21:38 cworth joined #git
21:38 gkatsev joined #git
21:38 canton7 joined #git
21:38 CandyAngel joined #git
21:38 astaves joined #git
21:38 DTekk joined #git
21:38 aspiers joined #git
21:38 bipolar joined #git
21:38 Ring0` joined #git
21:38 marienz joined #git
21:38 Guest39784 joined #git
21:38 iveqy joined #git
21:38 SegFaultAX joined #git
21:38 wizardedit joined #git
21:38 synchronous joined #git
21:38 Afforess joined #git
21:38 GingerGeek joined #git
21:38 mafrasi2_ joined #git
21:38 VampiricPadraig_ joined #git
21:38 boshhead joined #git
21:38 marvi joined #git
21:38 e joined #git
21:38 themill joined #git
21:38 hwoarang joined #git
21:38 DarkMukke joined #git
21:38 tfnico joined #git
21:38 brewmast1r joined #git
21:38 harv joined #git
21:38 satifant joined #git
21:38 anonymuse joined #git
21:38 Mo0O joined #git
21:38 SporkWitch joined #git
21:39 Nothing4You joined #git
21:40 theskillwithin joined #git
21:40 jerrcs joined #git
21:40 olalonde joined #git
21:40 raijin joined #git
21:40 anuxivm joined #git
21:40 Mobutils joined #git
21:40 sparetire joined #git
21:40 mrpackethead joined #git
21:40 MattMaker joined #git
21:40 p4trix joined #git
21:40 thiago joined #git
21:40 inflames joined #git
21:40 boombatower joined #git
21:40 liquidjamm joined #git
21:40 ahmed_elgabri joined #git
21:40 chris2 joined #git
21:40 tyreld joined #git
21:40 Guest28265 joined #git
21:40 Caelum joined #git
21:40 dilfridge joined #git
21:40 Cerise joined #git
21:40 mozzarella joined #git
21:40 cspotcode joined #git
21:40 realz joined #git
21:40 jason237 joined #git
21:40 timthowtdi joined #git
21:40 aviraldg joined #git
21:40 bremner joined #git
21:40 grawity joined #git
21:40 troyt joined #git
21:40 iob- joined #git
21:40 DelphicOkami joined #git
21:40 Nitesh joined #git
21:40 TSS_ joined #git
21:40 nedbat joined #git
21:40 tubaman joined #git
21:40 azerus joined #git
21:40 croberts joined #git
21:40 tek81 joined #git
21:40 blackjid joined #git
21:40 miklcct joined #git
21:40 holgersson joined #git
21:40 hide_ joined #git
21:40 pikajude joined #git
21:40 nebg joined #git
21:40 CussBot joined #git
21:40 Ranhir joined #git
21:40 cjbrambo joined #git
21:40 nephyrin joined #git
21:40 cdesai joined #git
21:40 cebor_ joined #git
21:40 GodEater joined #git
21:40 Surfer2010 joined #git
21:40 berndj joined #git
21:40 howitdo joined #git
21:40 asdofindia joined #git
21:40 dhanush joined #git
21:40 NightMonkey joined #git
21:40 armenzg joined #git
21:40 jpleau joined #git
21:40 emilyrose joined #git
21:40 Lowell joined #git
21:40 kanyeezy joined #git
21:40 karstensrage joined #git
21:40 rancyd joined #git
21:40 sparr joined #git
21:40 hagabaka joined #git
21:40 cschneid joined #git
21:40 POQDavid joined #git
21:40 uebera|| joined #git
21:40 Forkk joined #git
21:40 Bigcheese joined #git
21:40 przemoc joined #git
21:40 davidstrauss joined #git
21:40 DistantStar joined #git
21:40 forrestv joined #git
21:40 HedgeMage joined #git
21:40 boxrick1 joined #git
21:40 hexa- joined #git
21:40 habbasi joined #git
21:40 altendky joined #git
21:40 WillMoogle joined #git
21:40 richardlitt joined #git
21:40 daxroc joined #git
21:40 davisonio joined #git
21:40 Tee_Pee joined #git
21:40 contracode joined #git
21:40 ealvarez___ joined #git
21:40 vcoinminer joined #git
21:40 cloudcell_ joined #git
21:40 pfg joined #git
21:40 j416 joined #git
21:40 ygeLN joined #git
21:40 CentaurWarchief joined #git
21:40 Exagone313 joined #git
21:40 WebDawg joined #git
21:40 p1k joined #git
21:40 atrus joined #git
21:40 n-st joined #git
21:40 SleePy joined #git
21:40 bejeb joined #git
21:40 Hello71 joined #git
21:40 Dev0n joined #git
21:40 tjbp joined #git
21:40 monokrome joined #git
21:40 maqr joined #git
21:40 ashnur joined #git
21:40 Maple__ joined #git
21:40 nickthetait joined #git
21:40 bailon joined #git
21:40 rzeka joined #git
21:40 benbergman joined #git
21:40 ariver joined #git
21:40 CIAguy joined #git
21:40 Grum joined #git
21:40 jokke joined #git
21:40 preaction joined #git
21:40 rts-sander joined #git
21:40 lamont joined #git
21:40 Neal_ joined #git
21:40 nkhoit joined #git
21:40 toydestroyer joined #git
21:40 Esya joined #git
21:40 errr joined #git
21:40 Chaser joined #git
21:40 Sigma joined #git
21:40 McKraken_zzzzz joined #git
21:40 madduck joined #git
21:40 fk joined #git
21:40 topi` joined #git
21:40 fhackdroid joined #git
21:40 saidelike joined #git
21:40 zifnab joined #git
21:40 WakiMiko_ joined #git
21:40 tonyj joined #git
21:40 chrysa joined #git
21:40 vassagus joined #git
21:40 dviola joined #git
21:40 courrier joined #git
21:40 elastix joined #git
21:40 mikecmpbll joined #git
21:40 heroux joined #git
21:40 Chloe joined #git
21:40 KeelOfSteel joined #git
21:40 midnightmagic joined #git
21:40 trelane joined #git
21:40 MrWoohoo joined #git
21:40 tokkee joined #git
21:40 roger_raymond joined #git
21:40 tchan joined #git
21:40 polyzen joined #git
21:40 SebastianFlyte joined #git
21:40 LeMike joined #git
21:40 quenode joined #git
21:40 alcohol joined #git
21:40 DolphinDreamm joined #git
21:40 catsup joined #git
21:40 nevyn joined #git
21:40 Arguggi joined #git
21:40 sartan joined #git
21:40 comps joined #git
21:40 Guest22710 joined #git
21:40 IndigoTiger joined #git
21:40 jab416171 joined #git
21:40 same_ol_1eez joined #git
21:40 larrybolt joined #git
21:40 Eryn_1983_FL joined #git
21:40 LukeLR joined #git
21:40 rafalcpp joined #git
21:40 rj1 joined #git
21:40 devster31 joined #git
21:40 Davey joined #git
21:40 markovh joined #git
21:41 Xe joined #git
21:41 tHEMtZ joined #git
21:41 weltall joined #git
21:41 elementalest joined #git
21:41 JCaesar joined #git
21:41 fullstack joined #git
21:41 franl joined #git
21:41 snthsnth joined #git
21:41 sin3t joined #git
21:41 eivarv joined #git
21:41 CrazyBonz joined #git
21:41 cagedwisdom joined #git
21:41 mrwat joined #git
21:41 nd joined #git
21:41 madewokherd joined #git
21:41 antlarr joined #git
21:41 sdothum joined #git
21:41 Zilenc3 joined #git
21:41 harish_ joined #git
21:41 ojdo joined #git
21:41 blunaxela joined #git
21:41 DonRichie joined #git
21:41 Sna4x8 joined #git
21:41 Atlantic777 joined #git
21:41 likewhoa joined #git
21:41 noodle joined #git
21:41 axisys joined #git
21:41 blahdodo joined #git
21:41 biax joined #git
21:41 sitaram joined #git
21:41 chernjie joined #git
21:41 grift joined #git
21:41 MarkusDBX joined #git
21:41 mukmuk joined #git
21:41 dustinm joined #git
21:41 HyP3r joined #git
21:41 Chelsea_Jurgens joined #git
21:41 Moult joined #git
21:41 sleezd joined #git
21:41 corebits joined #git
21:41 arand_ joined #git
21:41 johnnyfive joined #git
21:41 Impaloo joined #git
21:41 KaibutsuX_ joined #git
21:41 scorphus joined #git
21:41 rom1504 joined #git
21:41 Anthaas joined #git
21:41 d3r1ck joined #git
21:41 lohfu joined #git
21:41 sim642 joined #git
21:41 hiroki joined #git
21:41 mischief joined #git
21:41 bathtub_shark joined #git
21:41 ap0c joined #git
21:42 flopsec joined #git
21:42 kulelu88 joined #git
21:42 EvilPeng1 joined #git
21:42 actinide joined #git
21:42 iamthad joined #git
21:42 tango_ joined #git
21:42 xcesariox joined #git
21:42 meLon joined #git
21:42 pigeon joined #git
21:42 rbern joined #git
21:42 chimeracoder joined #git
21:42 timvisher joined #git
21:42 synthmeat joined #git
21:42 fiete joined #git
21:42 sharkz_ joined #git
21:42 jwest joined #git
21:42 enleeten joined #git
21:42 GyrosGeier joined #git
21:42 lmat joined #git
21:42 serialoverflow joined #git
21:42 Nugget joined #git
21:42 grumble joined #git
21:42 brandonw joined #git
21:42 Guest68096 joined #git
21:42 mutex joined #git
21:42 Orion3k joined #git
21:42 spriz joined #git
21:42 quite joined #git
21:42 constantinexvi joined #git
21:42 BlaXpirit joined #git
21:42 mitzip joined #git
21:42 jamick joined #git
21:42 tuor joined #git
21:42 caffeinatedcode joined #git
21:42 rxo joined #git
21:42 thejoecarroll joined #git
21:42 quadz joined #git
21:42 fold4 joined #git
21:42 unbalancedparen joined #git
21:42 Aaeriele joined #git
21:42 babilen joined #git
21:42 MattE joined #git
21:42 parasite_ joined #git
21:42 adante joined #git
21:42 raz joined #git
21:42 HarryHallman joined #git
21:42 BeerSerc joined #git
21:42 rideh joined #git
21:42 dmarr joined #git
21:42 ndim joined #git
21:42 AlgorithmWizard joined #git
21:42 Ekho joined #git
21:42 styx_ joined #git
21:42 ffualo joined #git
21:42 Juri_Agarin joined #git
21:42 azwieg103 joined #git
21:42 Kuukunen joined #git
21:42 taargus41 joined #git
21:42 greymalkin joined #git
21:42 snowinferno_ joined #git
21:42 Yaiyan joined #git
21:42 harleyyyyymayes joined #git
21:42 Karunamon joined #git
21:42 m0ltar joined #git
21:42 colona joined #git
21:42 angseus joined #git
21:42 mattp_ joined #git
21:42 dzho joined #git
21:42 EvanDotPro joined #git
21:42 jasondotstar joined #git
21:42 JoshL joined #git
21:42 Vadtec joined #git
21:42 Ariadeno joined #git
21:43 timvisher joined #git
21:43 rafalcpp joined #git
21:43 fold4 joined #git
21:43 MarkusDBX joined #git
21:43 catsup joined #git
21:44 cdown_ joined #git
21:44 grumble joined #git
21:45 ahr3n joined #git
21:45 neurodrone_ joined #git
21:45 IndigoTiger joined #git
21:46 monokrome joined #git
21:46 nitric joined #git
21:46 danslo joined #git
21:48 d4rklit3 joined #git
21:48 peepsalot is there a way to extend git commands with scripts, so i could have something like "git unadd"
21:49 lolmac joined #git
21:49 d4rklit3 left #git
21:51 cspotcode joined #git
21:51 tfnico joined #git
21:51 aendrew joined #git
21:51 tang^_ joined #git
21:51 daxroc joined #git
21:52 elmcrest joined #git
21:52 nedbat peepsalot: if you put an executable file named git-foob on your path, then "git foob" will run it.
21:52 devhost joined #git
21:52 peepsalot ah, neat
21:52 nedbat peepsalot: you can also put aliases into your ~/.gitconfig file
21:52 ggherdov joined #git
21:53 hahuang61 joined #git
21:54 teadrop_ joined #git
21:54 rominronin joined #git
21:54 MissionCritical joined #git
21:54 richardlitt joined #git
21:55 neurodrone_ joined #git
21:56 sylario joined #git
21:56 jaafar joined #git
21:56 ToeSnacks joined #git
21:57 chelly joined #git
21:58 nikivi joined #git
21:59 al-damiri joined #git
21:59 addshore joined #git
21:59 aidalgol joined #git
21:59 discopatrick joined #git
22:00 angelsl joined #git
22:00 Iacobus joined #git
22:00 ivpvd joined #git
22:00 twisted` joined #git
22:00 harleypig joined #git
22:01 felowe joined #git
22:01 Awesomecase joined #git
22:01 trig-ger joined #git
22:01 martin_g joined #git
22:02 cstrahan joined #git
22:03 duderonomy joined #git
22:03 beckyconning_ joined #git
22:03 gyre007 joined #git
22:03 jottr joined #git
22:04 sr___ joined #git
22:04 DannyFritz joined #git
22:04 zero7 joined #git
22:05 spr0tsuki joined #git
22:05 Apteryx joined #git
22:05 tobie joined #git
22:05 iamdevnul joined #git
22:05 MLM joined #git
22:05 thecomedian joined #git
22:05 GameGamer43 joined #git
22:06 dunk joined #git
22:06 kwmiebach joined #git
22:06 noahsussman joined #git
22:06 zeller joined #git
22:06 olalonde joined #git
22:06 lightstalker joined #git
22:06 hahuang62 joined #git
22:07 adambrenecki joined #git
22:07 yena joined #git
22:07 matsaman joined #git
22:07 vassagus_ joined #git
22:07 mehdi__ joined #git
22:07 henriquev joined #git
22:08 Tee_Pee joined #git
22:08 theresajayne joined #git
22:08 jleclanche joined #git
22:08 ionelmc joined #git
22:08 therealGent joined #git
22:08 mehola joined #git
22:08 imanc joined #git
22:08 DuncanT joined #git
22:09 Bengi joined #git
22:09 cloudcell_ joined #git
22:09 angular_mike_ joined #git
22:09 CEnnis91 joined #git
22:09 unholycrab joined #git
22:09 timwis joined #git
22:09 Lowell joined #git
22:09 flupke joined #git
22:10 worstadmin joined #git
22:10 techntoke joined #git
22:11 kutenai joined #git
22:11 Gsham joined #git
22:11 Lewix joined #git
22:11 janx joined #git
22:11 sfan786_ joined #git
22:11 mroth joined #git
22:12 boxrick1 joined #git
22:12 bcardarella joined #git
22:12 wizonesolutions joined #git
22:12 jmx joined #git
22:12 shennyg joined #git
22:12 epic joined #git
22:13 sebhoss joined #git
22:13 ec joined #git
22:13 SwiftMatt joined #git
22:13 da5id joined #git
22:13 yoyoyo___ joined #git
22:13 DarkMukke joined #git
22:13 huhlig joined #git
22:13 HoierM joined #git
22:14 davisonio joined #git
22:14 paperziggurat joined #git
22:14 [diecast] joined #git
22:14 altendky joined #git
22:14 XDS2010 joined #git
22:14 contracode joined #git
22:14 FritzLakritz joined #git
22:14 barq joined #git
22:15 hahuang61 joined #git
22:15 rfv joined #git
22:15 cory_ joined #git
22:15 JackMc joined #git
22:15 cmosguy joined #git
22:15 elijah joined #git
22:16 ulkesh joined #git
22:16 pizzaops joined #git
22:16 parliament joined #git
22:16 pfg joined #git
22:16 prosody joined #git
22:16 stealthii joined #git
22:17 vcoinminer joined #git
22:17 bbhoss joined #git
22:17 johtso joined #git
22:17 joseph_arnstein joined #git
22:18 mingrammer joined #git
22:18 etcinit joined #git
22:18 navidr joined #git
22:19 jaafar joined #git
22:19 AfroThundr joined #git
22:19 lebster joined #git
22:19 WillMoogle joined #git
22:22 mischat joined #git
22:24 danslo joined #git
22:29 cdg joined #git
22:34 NeverDie joined #git
22:38 nilez joined #git
22:38 airdisa joined #git
22:40 danslo joined #git
22:43 justan0theruser joined #git
22:45 will joined #git
22:45 obiwahn is there a way to find out the names of branches when a merge has started?
22:47 raijin joined #git
22:48 Apteryx Hi! I've converted a SVN repo to Git using "git svn clone --stdlayout". It now has many remotes/origin/tags/* and branches, which is good. Now if I do "git clone" on this local repo, these seem to be lost. Why?
22:49 ash_workz joined #git
22:53 ToBeCloud joined #git
22:54 morenoh149 joined #git
22:56 danslo joined #git
22:56 obiwahn you will have them available as origin/tag#
22:56 shirty joined #git
22:57 obiwahn or you could clone with --mirror and turn the bare repo into a normal one afterwards.
22:57 andlabs joined #git
22:58 xlegoman joined #git
22:58 Vampire0 !svn2git
22:58 gitinfo For a one-time conversion of SVN to git, svn2git should be used instead of git-svn. There are many tools with this name. Probably the best is the KDE one at https://github.com/svn-all-fast-export/svn2git. To analyze the SVN repo history upfront for building up proper rules for svn2git, you might use svneverever from here: http://blog.hartwork.org/?p=763
22:59 nitesh` joined #git
23:00 stomplee joined #git
23:01 Apteryx obiwahn: OK. Not sure how to turn a bare repo into a non bare one, if not by using "git clone"
23:02 clr obiwahn: !rev-parse
23:02 Apteryx OK, thanks!
23:03 nitesh joined #git
23:03 clr obiwahn: ok, that didn't work, look at rev-parse: "A colon, optionally followed by a stage number ..."
23:05 fuchstronaut joined #git
23:05 mablae__ joined #git
23:08 Vampire0 clr you mean man git rev-parse
23:08 gitinfo the git-rev-parse manpage is available at http://jk.gs/git-rev-parse.html
23:09 Vampire0 clr, but how does that help in knowing the branchnames involved?
23:11 ResidentBiscuit joined #git
23:11 danslo joined #git
23:12 Milijus joined #git
23:13 clr Vampire0: he asked me a question before that
23:14 wizonesolutions joined #git
23:14 clr Vampire0: I was trying to get the bot to say something about it, that didn't work
23:14 mischat joined #git
23:15 wizonesolutions joined #git
23:16 cdown joined #git
23:16 mischat joined #git
23:16 mizu_no_oto joined #git
23:17 mischat joined #git
23:17 ResidentBiscuit joined #git
23:17 Gsham joined #git
23:18 Urbany joined #git
23:18 mischat joined #git
23:18 mischat joined #git
23:19 pothepanda joined #git
23:19 mischat joined #git
23:20 mischat joined #git
23:21 NaN joined #git
23:22 dodobrain joined #git
23:22 cdg_ joined #git
23:23 NaN is it possible to rename an already saved stash?
23:23 k6ekb joined #git
23:25 ffabi joined #git
23:25 madewokherd joined #git
23:26 EvilPeng1 joined #git
23:27 danslo joined #git
23:29 dsdeiz joined #git
23:29 dsdeiz joined #git
23:32 UserOO7 joined #git
23:42 clr NaN: what is its current name?
23:43 danslo joined #git
23:45 thiago joined #git
23:45 Apteryx How can I create a git tag and associate it to a remote branch?
23:45 obiwahn reset-merge-of-file = "!_git_fun() { git merge-file -L HEAD -L BASE -L OTHER -p $(git unpack-file :2:\"$1\") $(git unpack-file :1: \"$1\") $(git unpack-file :3:\"$1\"); rm .merge_file_*; }; _git_fun"
23:46 obiwahn is there a way to replace HEAD and other with the actual branch names?
23:47 agileadam joined #git
23:48 NaN clr: stash@{0}
23:48 NaN I want to change the message at least
23:48 obiwahn clr: i found the rev-parse command already but replacing the name is still on my list for tomorrow:)
23:51 mischat joined #git
23:54 rominronin joined #git
23:56 Mobutils joined #git
23:56 Noldorin joined #git
23:57 DolphinDream joined #git
23:58 mizu_no_oto joined #git
23:59 danslo joined #git

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