Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-09-23

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

All times shown according to UTC.

Time Nick Message
00:03 raijin joined #git
00:03 Rodya_ joined #git
00:04 a3Dman joined #git
00:05 mingrammer joined #git
00:06 netj joined #git
00:08 cdg joined #git
00:08 prawnzy joined #git
00:09 sagerdearia joined #git
00:11 cqi joined #git
00:12 inflames joined #git
00:12 AnarchyAo joined #git
00:14 ismaris joined #git
00:15 danslo joined #git
00:15 fission6 joined #git
00:16 cagedwisdom joined #git
00:16 fission6 is it odd or ineffiecient workflow to pull down Pull request branches locally, merge master in to them, run tests, then checkout master and merge the branch into master?
00:20 madewokherd I don't see any reason to merge it in multiple ways like that
00:20 Gsham joined #git
00:20 madewokherd why not fetch, merge into master, then run the tests?
00:20 e14 joined #git
00:21 madewokherd if there's a problem, you can undo the merge locally
00:21 fission6 why undo the merge vs update the branch
00:23 madewokherd just so that you're not doing two different merges, and the thing you tested is what ends up in master
00:23 fission6 that just doesn't feel right to merge into master then go through the pain of unmerging so you can have a clean master
00:23 singalaut joined #git
00:23 fission6 sounds odd, would like to hear other peoples workflows though
00:24 madewokherd well, if you do the merge twice you will also have to resolve any conflicts twice
00:24 fission6 not really
00:24 fission6 i never have to resolve merge conflicts with this workflow
00:25 m1m3-50 joined #git
00:26 nedbat fission6: what do you mean by "unmerge so you can have a clean master"?
00:27 clr fission6: undoing a merge is not much pain: git reset --hard @~
00:27 fission6 well i mean if you merge a branch in locally to master then run tests and it breaks you are left with a bad master
00:27 nedbat fission6: yeah, not hard to undo.
00:27 nedbat fission6: for example:  git reset --hard origin/master
00:28 fission6 what if you had a merge conflict and fixed the merge conflict, ran the tests and something went wrong and you had to get back to a clean master
00:28 ochorocho__ joined #git
00:29 nedbat fission6: git reset --hard origin/master
00:29 fission6 and all this is better than my workflow?
00:29 dzho joined #git
00:29 nedbat fission6: "all this"? That's one command that you only have to run if something goes wrong.
00:29 clr fission6: to answer your question, my workflow would be be to merge into master, build and and test, then push if things look good.  i think you will find this is common.
00:29 nedbat fission6: you have two merges in the best case.
00:30 fission6 clr and what if things do not look good
00:30 danslo joined #git
00:30 nedbat fission6: clr already showed you the command they would use:  git reset --hard @~
00:30 clr fission6: git reset --hard origin/master
00:31 fission6 it just feels more stable to isolate merge issues into the topic branch
00:31 fission6 more sane or inutivie
00:31 fission6 intuitive
00:31 Blackpajamas joined #git
00:31 clr fission6: or git reset --hard @{u}, which is a general way to refer to upstream
00:32 dalias fission6, you can do a local topic branch based on master and merge into it
00:32 dalias then just discard the local topic branch if you don't like it
00:32 clr fission6: the reset basically removes the merge from history as if it didn't exist.
00:32 dalias that's equivalent to just resetting master tho
00:33 dalias the backwards workflow you're using now suggests some sort of misunderstanding of how git works
00:33 e14 joined #git
00:34 HoierM joined #git
00:34 clr fission6: the act of merging master into a feature branch usually would mean you want to depend on something added to master after you created the branch.  what you're doing accomplishes what you want, but it looks weird to someone that is used to common conventions.
00:35 fahadash joined #git
00:35 fission6 ill still with my flow, dont mind the double merge!
00:35 sagerdearia joined #git
00:35 UserOO7 joined #git
00:35 fission6 thanks for everyones approach though
00:35 dalias clr, btw does git make any special distinctions between first/second/... parent anywhere?
00:36 fission6 also i am about to pay for travis-ci / github, is travis-ci worth the money for what it does? just gtting the basic bootstrap service
00:36 clr dalias: git log allows you to pass --first-parent
00:36 clr dalias: which is useful if you are always merging towards upstream
00:38 dalias can i ask you for an idea on what i'm working on?
00:38 clr dalias: e.g., git log --first-parent on the git or linux trees gives a nice birdseye view of the activity.  if you're not following that convention, then --first-parent is probably a random list of stuff.
00:39 dalias *nod*
00:39 dalias that makes sense
00:39 pandeiro joined #git
00:40 dalias in my repo conversion (hg with subrepos to unified git history of whole-tree state) i'm representing "update subrepos" as a multi-way merge of N branches where changes to N different subtrees were made
00:40 dalias this is kinda an abuse of merges but it "works"
00:41 dalias do you have any thoughts on a better way to represent it?
00:42 dalias ideally the top-level repo would only have updated multiple subrepos at once when the intent was to make an atomic change, but that's not what happened :(
00:43 clr dalias: i'm not much of an hg person, but that sounds totally reasonable.  why do you feel like it's abusing merges?
00:43 morenoh149 joined #git
00:45 dalias well most people don't expect N-way merges all over the place
00:45 dalias and git log -p does not show any additional changes that were made with the merge unless you also add -c
00:46 danslo joined #git
00:48 ghostlight joined #git
00:52 clr dalias: yeah, they're somewhat unusual to find in repos but it seems like what you're trying to represent is best done with an N-way merge.  otherwise it seems like you'd need to split up the suprepo updates somehow.
00:52 mischat joined #git
00:53 dalias probably most of the time the changes were not logically intended to be atomic, it was just someone being lazy and doing their commits in subrepos and not doing a new top-level commit to use them until a bunch of subrepo changes had been made
00:54 RedNifre_ joined #git
00:54 snthsnth joined #git
00:57 clr dalias: fixing that is usually so much work, though.  :-(  anyway, seems reasonable to me, and i agree with the caveats you noted.
00:58 tesuji joined #git
01:00 mingrammer joined #git
01:02 sangy joined #git
01:02 danslo joined #git
01:04 diogenese joined #git
01:05 UserOO7 joined #git
01:06 wxg joined #git
01:08 clemf joined #git
01:12 jaafar joined #git
01:13 Apteryx joined #git
01:13 cyphase joined #git
01:13 dalias clr, thanks for the feedback
01:14 Cabanossi joined #git
01:14 anonymuse joined #git
01:16 raijin joined #git
01:16 UserOO7 joined #git
01:17 evanwang joined #git
01:18 bobdobbs joined #git
01:18 danslo joined #git
01:18 ismaris joined #git
01:18 snthsnth joined #git
01:22 somaReverse joined #git
01:23 somaReverse Hello, my team have three correlated git repos in one project. How can I manage them in one directory? Is git submodule the right choice?
01:25 cyphase joined #git
01:26 rchavik joined #git
01:26 ahr3n joined #git
01:27 mrpackethead joined #git
01:28 sturby joined #git
01:28 Vampire0 joined #git
01:28 snthsnth joined #git
01:28 d^sh_ joined #git
01:30 Hello71 !sub
01:30 gitinfo [!subgit] SubGit is a two-way git<->svn bridge. It provides a (self-)hosted alternative to git-svn, with some advantages. See http://subgit.com/
01:30 Hello71 ugh
01:31 Hello71 !submodule
01:31 gitinfo git-submodule is ideal to add subsidiary git repositories to a git superproject when you do not control the subprojects or more specifically wish to fix the subproject at a specific revision even as the subproject changes upstream. See http://www.git-scm.com/boo​k/en/Git-Tools-Submodules
01:31 Hello71 hm...
01:31 Hello71 !subp
01:31 gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
01:32 somaReverse haha, thanks
01:33 somaReverse !subtree
01:33 gitinfo The subtree merge method is great for incorporating a subsidiary git repo into your current one with "unified" history. Read http://git-scm.com/book/en/v2/Git-To​ols-Advanced-Merging#_subtree_merge for more info, see also !git-subtree and !git-stitch-repo.
01:33 somaReverse !git-stitch-repo
01:33 gitinfo The git-stitch-repo tool allows for stitching several git repositories together as a unified history without merges, see https://metacpan.org/module/git-stitch-repo for more info.
01:33 somaReverse !gitslave
01:33 gitinfo gitslave (http://gitslave.sf.net) is useful to add subsidiary git repositories to a git superproject when you control and develop on the subprojects at more or less the same time as the superproject, and furthermore when you typically want to tag, branch, push, pull, etc. all repositories at the same time.
01:33 dermoth_ joined #git
01:34 danslo joined #git
01:34 Milijus joined #git
01:35 somaReverse Hello71: ok, so which one do you suggest?
01:36 Hello71 they do different things
01:36 Hello71 it depends on what you need
01:36 sagerdearia joined #git
01:37 Gsham joined #git
01:38 cyphase joined #git
01:38 Lor_ joined #git
01:40 somaReverse My need is somehow complex. I have people committing to multiple repos. It's a many to many relation between developers and repos.
01:40 Hello71 use the !subp
01:40 gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
01:40 ochorocho__ joined #git
01:41 clr somaReverse: i'd start with submodules.  they're easy to setup and don't require any changes to the other repos.  submodules don't get much love, mostly because you have to learn a new set of commands.  but they are exactly for corrolating a group of independent repos.
01:41 somaReverse clr: the repos aren't independent
01:41 Shentino joined #git
01:42 clr somaReverse: ok, well, i think as far as git is concerned they are
01:46 somaReverse ah, i see
01:46 Shentino2 joined #git
01:46 Shentino joined #git
01:46 nwkj86 joined #git
01:48 Rodya_ joined #git
01:49 clr somaReverse: i should note that i'm not familiar with with gitslave or or stich.  gitslave sounds possibly better from the brief description.  but you're not going to wrong if you start with submodules
01:50 danslo joined #git
01:50 fstd_ joined #git
01:52 fission6 joined #git
01:53 mingrammer joined #git
01:53 mischat joined #git
01:56 rominronin joined #git
01:56 grub joined #git
01:56 CEnnis91 joined #git
01:57 pks joined #git
02:00 whitby joined #git
02:00 grub hi
02:00 gitinfo grub: 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.
02:01 mangolisk joined #git
02:02 grub Quick question. I'm setting up git on my mac using command line. Is there anyway I can establish a password for my account?
02:02 mizu_no_oto joined #git
02:04 harish joined #git
02:05 kireevco joined #git
02:05 kireevco hi all. looking if someone has an idea how to share a file across multiple repos. It is a file, not a directory, submodules won't seem to work
02:06 fuzzmz joined #git
02:06 e14 joined #git
02:06 cyphase_eviltwin joined #git
02:07 salamanderrake joined #git
02:08 mangolisk joined #git
02:09 romerocesar joined #git
02:09 prawnzy joined #git
02:10 mehola joined #git
02:11 e14_ joined #git
02:15 wxg joined #git
02:16 dreiss joined #git
02:19 cyphase_eviltwin joined #git
02:21 danslo joined #git
02:24 durham joined #git
02:27 prawnzy joined #git
02:30 wxg joined #git
02:31 perrier-jouet joined #git
02:32 morenoh149 joined #git
02:33 cyphase_eviltwin joined #git
02:34 ojdo joined #git
02:34 dsantiago joined #git
02:36 cyphase_eviltwin joined #git
02:38 leonarth joined #git
02:39 biertie joined #git
02:40 Gsham_ joined #git
02:41 ochorocho__ joined #git
02:44 therealGent joined #git
02:46 snthsnth joined #git
02:47 e14 joined #git
02:48 Goplat joined #git
02:48 Mogget joined #git
02:53 hexagoxel joined #git
02:54 jmx joined #git
02:54 cyphase joined #git
02:54 mischat joined #git
02:55 LeBlaaanc joined #git
02:58 chachasmooth joined #git
03:03 aavrug joined #git
03:05 aavrug joined #git
03:06 cyphase joined #git
03:07 clemf joined #git
03:11 anonymuse joined #git
03:13 danslo joined #git
03:13 arescorpio joined #git
03:15 cagmz joined #git
03:17 gfixler joined #git
03:17 ndim joined #git
03:18 e14 joined #git
03:19 cyphase joined #git
03:19 Aboba joined #git
03:20 e14 joined #git
03:21 MattMake_ joined #git
03:24 EvilPeng1 joined #git
03:27 LeBlaaanc joined #git
03:27 waveclaw joined #git
03:28 danslo joined #git
03:31 robotroll joined #git
03:35 sargas joined #git
03:35 SwiftMatt joined #git
03:35 duderonomy joined #git
03:35 LeBlaaanc joined #git
03:36 Electrometro joined #git
03:37 Aboba_ joined #git
03:38 durham joined #git
03:38 cyphase joined #git
03:44 danslo joined #git
03:44 aavrug joined #git
03:45 justanotheruser joined #git
03:45 drupald joined #git
03:46 aavrug joined #git
03:48 boombatower joined #git
03:48 olalonde joined #git
03:48 mingrammer joined #git
03:51 mehola joined #git
03:51 _W_ joined #git
03:54 archaic joined #git
03:56 mischat joined #git
03:56 Lyqyd joined #git
03:57 rominronin joined #git
03:57 LeBlaaan_ joined #git
04:00 danslo joined #git
04:01 Electrom_ joined #git
04:01 cyphase joined #git
04:12 mehola joined #git
04:14 red82 joined #git
04:14 cyphase joined #git
04:15 romerocesar joined #git
04:16 danslo joined #git
04:17 daey_ joined #git
04:18 cqi joined #git
04:20 mingrammer joined #git
04:20 morenoh149 joined #git
04:21 nidr0x joined #git
04:22 snthsnth joined #git
04:23 z8z joined #git
04:25 ItSANgo joined #git
04:28 cyphase joined #git
04:29 red82 joined #git
04:32 danslo joined #git
04:33 russoneto joined #git
04:33 Cabanossi joined #git
04:35 snthsnth joined #git
04:35 WizJin joined #git
04:35 russo joined #git
04:36 grawity submodules + well-placed symlinks
04:40 ThunderChicken joined #git
04:42 ochorocho__ joined #git
04:43 dreiss joined #git
04:47 danslo joined #git
04:52 romerocesar joined #git
04:55 cyphase joined #git
04:57 mischat joined #git
04:59 fedepad joined #git
05:00 romerocesar joined #git
05:03 dec0n joined #git
05:03 danslo joined #git
05:05 subhojit777 joined #git
05:06 rscata joined #git
05:09 hahuang61 joined #git
05:14 qws-user-1228 joined #git
05:15 Mixologic joined #git
05:16 lolmac joined #git
05:17 prawnzy joined #git
05:17 ShekharReddy joined #git
05:19 danslo joined #git
05:21 qws-user-1228 joined #git
05:21 cyphase joined #git
05:22 shinnya joined #git
05:22 t0by joined #git
05:23 ewet joined #git
05:23 actinide joined #git
05:24 SwiftMatt joined #git
05:28 Mixologic joined #git
05:31 jorj joined #git
05:31 qws-user-1228 joined #git
05:32 HappyLoaf joined #git
05:35 danslo joined #git
05:35 qws-user-1228 joined #git
05:35 r3c0d3x joined #git
05:41 cyphase joined #git
05:41 jmx joined #git
05:41 Mixologic joined #git
05:42 zeroed joined #git
05:42 zeroed joined #git
05:42 dodobrain joined #git
05:45 stuh84 joined #git
05:45 snthsnth joined #git
05:46 ThomasLocke joined #git
05:46 ThomasLocke joined #git
05:46 rscata joined #git
05:47 overlord_tm joined #git
05:51 danslo joined #git
05:54 anbz joined #git
05:55 linuxmint joined #git
05:56 Raging_Hog joined #git
05:58 rominronin joined #git
05:59 drbean_ joined #git
06:00 arand joined #git
06:01 drbean- joined #git
06:01 mischat joined #git
06:03 drbean joined #git
06:04 drbean_ joined #git
06:06 drbean| joined #git
06:07 danslo joined #git
06:09 hahuang61 joined #git
06:09 singalaut joined #git
06:09 morenoh149 joined #git
06:09 drbean joined #git
06:10 lb1c joined #git
06:10 Andrew_K joined #git
06:10 cyphase joined #git
06:10 qt-x joined #git
06:11 drbean_ joined #git
06:12 drbean- joined #git
06:13 Repox joined #git
06:14 drbean| joined #git
06:14 rorro joined #git
06:18 rendar joined #git
06:18 bocaneri joined #git
06:22 danslo joined #git
06:23 jnavila joined #git
06:23 cbreak kireevco: maybe subtree merging helps
06:23 cbreak kireevco: or use git submodules + symlink
06:24 dan2k3k4 joined #git
06:24 cbreak kireevco: or use plain merging of a separate that-file-only repo
06:28 cyphase joined #git
06:31 lindenle joined #git
06:33 KnownSyntax joined #git
06:34 Sceorem joined #git
06:34 planet_en joined #git
06:35 zeroed_ joined #git
06:35 shgysk8zer0 joined #git
06:36 JeroenT joined #git
06:36 snthsnth joined #git
06:37 jknetl joined #git
06:38 danslo joined #git
06:39 clemf joined #git
06:40 fedepad joined #git
06:40 JeroenT joined #git
06:41 psprint git fsck --lost-found shows many dangling blobs and commits. Looks like git gc doesn't do its job. Running git gc by hand doesn't remove the objects. Why?
06:41 cyphase joined #git
06:42 psprint hm maybe because I've issued 100 commits in like 2 weeks and many amends, so it's just no timeout
06:42 m1m3-50 joined #git
06:42 ap0c joined #git
06:44 nur joined #git
06:44 anbz joined #git
06:44 nur Is it a bad idea to rsync a git repo to another machine? Will all git operations still work as expected on the remote machine?
06:48 Takiito joined #git
06:48 Takiito Hello
06:50 wxg joined #git
06:50 eduardas_m joined #git
06:50 Takiito How can I cause git pull to behave like cvs update in the way that it deletes local files that are no longer in the repository?
06:54 danslo joined #git
06:55 [Brain] joined #git
06:55 fuzzmz but that's how it works
06:55 submitnine joined #git
06:56 fuzzmz if you pull a commit that included deletions then those files will be deleted on pull
06:56 Takiito Hmm, ok. I have to check then if it was propery deleted.
06:56 fuzzmz Takiito, did you actually git rm file1 in a previous commit?
06:57 fuzzmz also, git doesn't track directories, so it won't remove ones that become empty as a result of pulls deleting all of their content
06:58 fuzzmz left #git
06:58 fuzzmz joined #git
06:58 fuzzmz at least that's what I remember regarding dirs
07:00 rominronin joined #git
07:01 Takiito Ah that's the problem. The directory still had files in it which weren't tracked by git
07:01 Takiito The files that were tracked by git are actually gone.
07:02 Takiito So there isn't an equivalent to the -P option for cvs update (which deletes empty directories)?
07:03 Silenced joined #git
07:03 mischat joined #git
07:03 fuzzmz unfortunately not
07:04 fuzzmz the closest you get is git clean -fd , but be warned that that removes untracked files and directories
07:04 fuzzmz to see what would get removed you can do a it clean -nd
07:04 Takiito It's probably easier to remove those directories by hand :)
07:05 fuzzmz yeah
07:06 vuoto joined #git
07:07 Balliad joined #git
07:09 chandu20 joined #git
07:09 encod3 joined #git
07:10 danslo joined #git
07:10 hahuang61 joined #git
07:14 TomyWork joined #git
07:15 danslo joined #git
07:17 jmx joined #git
07:18 prawnzy joined #git
07:21 DevAntoine joined #git
07:21 JeroenT joined #git
07:22 qt-x joined #git
07:24 dan2k3k4 joined #git
07:24 psprint Writing a shell helper tool for git.. Want something cool that can be done with a commit. Nothing's there :( Just standard actions like checkout, log -p :(
07:27 albel727 joined #git
07:27 m1m3-50 joined #git
07:28 Silenced joined #git
07:29 snthsnth joined #git
07:30 cdown joined #git
07:33 canton7 joined #git
07:36 dreiss joined #git
07:38 PCatinean joined #git
07:38 jmx joined #git
07:40 jnavila joined #git
07:41 Lomex joined #git
07:48 chll_ joined #git
07:49 kurkale6ka joined #git
07:50 JeroenT joined #git
07:52 nur left #git
07:52 flaviodesousa joined #git
07:53 hide joined #git
07:54 ahmedelgabri joined #git
07:55 elect joined #git
07:58 russo joined #git
07:58 russo_ joined #git
07:58 mikecmpbll joined #git
07:58 morenoh149 joined #git
07:59 JeroenT_ joined #git
07:59 sonikspin joined #git
08:02 psprint Hm maybe I've got something: find commits that alter the same set of files. Is this doeable directly from git?
08:05 jmx joined #git
08:08 cyan__ joined #git
08:09 BadSnappa joined #git
08:10 Alienpruts joined #git
08:10 garyserj joined #git
08:10 eshlox joined #git
08:11 grawity only the "at least one file from set" kind
08:11 grawity but you can feed such lists to `comm` to get unions or intersections, I think
08:12 hahuang61 joined #git
08:13 quenode joined #git
08:13 garyserj grawity: how many times have you been spotted like a celebrity, from a stackexchange site?
08:14 garyserj like somebody on IRC saying "ah grawity, from there".
08:14 Macaveli joined #git
08:14 quenode joined #git
08:14 psprint grawity: can do complex data processing thanks to Zsh. So I would get list of candidates that then would have to be verified?
08:15 rorro joined #git
08:17 grawity garyserj: about five, I think?
08:17 grawity psprint: I guess
08:17 jmx joined #git
08:17 grawity psprint: try `git rev-list HEAD -- file1`
08:20 [Brain] joined #git
08:23 planet_en joined #git
08:24 biertie joined #git
08:25 clmsy joined #git
08:26 mgedmin joined #git
08:27 mgedmin is it possible to ask git commit --amend to please ignore the index for now and just edit the commit message of the previous commit without updating the commit's tree?
08:27 djb-irc joined #git
08:28 z8z joined #git
08:30 therealGent joined #git
08:30 grawity --only
08:30 p4trix joined #git
08:31 * canton7 has the aliases 'amend = commit --amend --no-edit' and 'reword = commit --amend --only'
08:31 psprint grawity: I see. How to nicely get list of altered files of a commit?
08:32 canton7 git show --name-status <commit>
08:32 canton7 or --name-only
08:32 laumann joined #git
08:32 Serpent7776 joined #git
08:32 krillr joined #git
08:33 lindenle joined #git
08:36 psprint thanks
08:37 devhoag joined #git
08:38 blackjid joined #git
08:40 Guest54448 joined #git
08:43 ssitani joined #git
08:43 Balliad joined #git
08:44 mischat joined #git
08:45 mingrammer joined #git
08:50 harish joined #git
08:51 jguillen joined #git
08:53 anbz joined #git
08:56 jnavila joined #git
08:57 Balliad joined #git
08:57 wxg joined #git
08:58 mgedmin thank you, grawity!
09:00 garethdaine joined #git
09:00 martinsb joined #git
09:01 * mgedmin has amend = git commit --amend -C HEAD and wonders if --no-edit didn't exist back then or if I just missed it in the man page
09:01 * mgedmin now has a git reword alias
09:01 martinsb hi, i have slight problem. I pushed some changes to main repo(gitlab) after merge that everything. Can i do something on my end to delete the changes in main repo or i have to look for a way to do it in gitlab?
09:03 Darcidride joined #git
09:04 jast mgedmin: --no-edit hasn't been around forever, I've been using -C HEAD, too
09:04 jast it's shorter anyway
09:13 xbob joined #git
09:13 hahuang61 joined #git
09:15 bernardio joined #git
09:17 bauruine joined #git
09:18 cyphase joined #git
09:19 prawnzy joined #git
09:22 leeN joined #git
09:24 jnavila joined #git
09:25 navidr joined #git
09:26 Ezzy joined #git
09:26 subby joined #git
09:27 garethdaine joined #git
09:30 BrianBlaze420 joined #git
09:30 niels__ joined #git
09:30 sin3t joined #git
09:33 shinnya joined #git
09:37 tvw joined #git
09:38 mkoskar joined #git
09:39 mingrammer joined #git
09:41 stuh84 joined #git
09:41 SwiftMatt joined #git
09:43 PioneerAxon joined #git
09:44 circ-user-WVSYH joined #git
09:44 fedepad joined #git
09:44 jmx joined #git
09:47 morenoh149 joined #git
09:48 stuh84 joined #git
09:48 weckl joined #git
09:48 fuchstronaut joined #git
09:48 fuchstronaut joined #git
09:49 JeroenT joined #git
09:50 shirty joined #git
09:53 lb1c i'm new to git subtree. is this really still expected behavior that rebase after subtree add is a major pain in the A** ?
09:56 stamina joined #git
10:00 Milijus joined #git
10:07 barteks2x joined #git
10:08 bongjovi joined #git
10:11 planet_en joined #git
10:13 Macaveli joined #git
10:14 hahuang61 joined #git
10:18 GT4066 joined #git
10:18 lb1c no one uses subtree?
10:19 lb1c :/
10:19 lb1c poor me
10:20 lb1c rebasing any history that involves a subtree fails? or just trying to rebase the commits, that subtree generates?
10:20 rorro joined #git
10:21 jmx joined #git
10:21 Balliad joined #git
10:23 Milijus joined #git
10:25 Triq joined #git
10:26 DevAntoine` joined #git
10:26 rorro joined #git
10:27 fling joined #git
10:28 mischat_ joined #git
10:30 DolphinDream joined #git
10:31 dan2k3k4 joined #git
10:34 JeroenT joined #git
10:35 gebbione joined #git
10:35 gebbione a specific file git revert is not working
10:35 gebbione any suggestion
10:35 lindenle joined #git
10:35 bejeb Does anyone on here use gitflow? How do you use git flow hotfix start <blah>? Is <blah> a version number or a descriptive name for the hotifx?
10:37 _ikke_ bejeb: git flow hotfix start [-F] <version> [<base>]
10:37 cdown joined #git
10:37 Darcidride joined #git
10:37 bejeb _ikke_: Thank you. and <version> is a version number the same as your standard versioning, like v1.3.3?
10:39 _ikke_ yes
10:39 _ikke_ I'm looking in the code how it's used
10:39 bejeb _ikke_: thanks
10:39 * bejeb thinks
10:39 bejeb I'm wondering what happens if, while dev A is working on hotfix/v1.3.3, someone else finishes a release for v1.4.0. When you do git flow hotfix finish, the version number now "rolls back", doesn't it?
10:39 codfection joined #git
10:39 Surfer2010_ joined #git
10:40 bejeb I guess I can see a way around it, maybe I'm worrying about nothing.
10:40 _ikke_ Shouldn't matter
10:40 _ikke_ version 1.3.3 can happen later as version 1.4.0
10:41 bejeb _ikke_: ok, now I'm confused.
10:41 _ikke_ Say you released 1.4.0
10:41 _ikke_ Now you find out 1.3.2 has a critical bug that needs to be fixed
10:41 bejeb with you so far
10:42 bejeb oooohh, so the last commit on hotfix/v1.3.3 should really be tagged v1.3.3, and the merge into master should be tagged as v1.4.1.
10:42 _ikke_ No, it's not 1.4.1
10:42 bejeb hence the -T option of git flow hotfix finish
10:42 bejeb oh, sorry, I jumped the gun
10:43 _ikke_ THe code shows no -T parameter
10:43 _ikke_ But anyway
10:43 bejeb odd - https://github.com/petervanderdoes/gitflow​-avh/wiki/Reference:-git-flow-hotfix#git-f​low-hotfix-finish---finish-a-hotfix-branch shows -T
10:43 _ikke_ I don't see it described in the git-flow post, but usually you create a maintenance branch for this
10:44 zoso joined #git
10:44 bejeb ah, of course - it would be on a maintenance branch
10:44 mattcen joined #git
10:44 bejeb _ikke_: as always, thank you, I learn something new every time we talk!
10:44 zoso Hi, is there a flag that I can pass to git-grep so that I get back the nearest function definition that is holding matching pattern?
10:44 _ikke_ No problem, glad to help
10:45 Xethron joined #git
10:45 _ikke_ zoso: -p / --show-function
10:45 _ikke_ man git grep
10:45 gitinfo the git-grep manpage is available at http://jk.gs/git-grep.html
10:45 _ikke_ Didn't know that existed
10:46 zoso _ikke_, When I do that I get the class definition back from the file. E.g "git grep -ie 'fooBar' --show-function" that returns the class definition
10:46 notebox joined #git
10:47 _ikke_ "The name is determined in the same way as git diff works out patch hunk headers (see Defining a custom hunk-header in gitattributes(5))"
10:47 jeffreylevesque joined #git
10:48 red82 joined #git
10:49 zoso _ikke_, thanks a lot man
10:49 _ikke_ bejeb: This is what git flow says about it: "when a release branch currently exists, the hotfix changes need to be merged into that release branch, instead of develop.:
10:50 _ikke_ But it says nothing about a hotfix for a previous version
10:50 _ikke_ But I would go the maintenance branch route
10:52 Darcidride joined #git
10:52 bejeb _ikke_: interesting. There are a few phases then, if you're lucky then simultaneous release and hotfix get published, fetched, and everyone adjusts accordingly. If not then the final fetch before finish on the hotfix will spot that you are out of sync and then you adjust accordingly - rebase and bump version, or put it on a maintenance branch for the older version
10:53 bejeb my poor team is still getting its head around git flow feature start/finish and git flow release start/finish!
10:53 bejeb so this will fry their brains
10:54 _ikke_ bejeb: I don't even use the plugin, just do my own thing
10:54 _ikke_ bejeb: Danger is that people don't see what it actually does
10:54 bejeb fair enough
10:55 bejeb this team I joined is _definitely_ in a better place since we started using it. Before that git was a mess, versioning was a mess, releases were a mess, so it has already helped.
10:55 mingrammer joined #git
10:56 shirty joined #git
10:56 ssitani joined #git
10:57 maret joined #git
10:58 fahadash joined #git
10:58 raijin joined #git
10:58 dan2k3k4_ joined #git
11:00 enckse joined #git
11:00 joki joined #git
11:01 enckse any thoughts on what is the most used/supported method for storing versioned binaries (whether in git or some other tool)?
11:01 enckse clearly something like git-lfs exists but not sure what input anyone here specifically has about it or other options
11:07 JeroenT joined #git
11:08 cdown joined #git
11:09 mizu_no_oto joined #git
11:10 Lomex joined #git
11:11 chandu20 joined #git
11:12 JeroenT joined #git
11:12 rnsanchez joined #git
11:14 red82 joined #git
11:15 hahuang61 joined #git
11:16 cdown joined #git
11:18 Andrew_K joined #git
11:19 prawnzy joined #git
11:19 ayjay_t joined #git
11:25 Pasi joined #git
11:26 djb-irc joined #git
11:30 mingrammer joined #git
11:35 solars joined #git
11:36 morenoh149 joined #git
11:38 nuck joined #git
11:38 tjone270 joined #git
11:39 mehola joined #git
11:42 Pasi joined #git
11:46 afuentes joined #git
11:48 dendazen joined #git
11:50 ssitani joined #git
11:50 anonymuse joined #git
11:52 JamesBaxter joined #git
11:54 a_thakur joined #git
11:55 Meteorhead joined #git
11:56 CussBot joined #git
11:57 hahuang61 joined #git
11:59 planet_en joined #git
11:59 johnmilton joined #git
12:01 star_prone joined #git
12:03 clmsy joined #git
12:05 jbitdrop joined #git
12:05 johnmilton joined #git
12:07 fling joined #git
12:08 WayToDoor joined #git
12:09 midnightmagic joined #git
12:10 Prisoner-2460_1 joined #git
12:11 NeverDie joined #git
12:13 Darcidride joined #git
12:15 johnmilton joined #git
12:16 fedepad joined #git
12:18 GodGinrai joined #git
12:21 r3c0d3x joined #git
12:21 g105b joined #git
12:22 johnmilton joined #git
12:25 JeroenT joined #git
12:25 Walakea_ joined #git
12:30 ShekharReddy joined #git
12:31 planet_en joined #git
12:31 garethdaine joined #git
12:31 zoso joined #git
12:32 ismaris joined #git
12:33 m0r0n joined #git
12:34 ewet joined #git
12:35 liquidjamm joined #git
12:38 _ikke_ bejeb: cool, nice to hear
12:39 _ikke_ enckse: !media
12:39 gitinfo enckse: [!annex] git-annex, git-media, and Git LFS are some solutions to the !binary problem. They work by keeping the blobs outside of the repo, storing a reference to the blob in the repo instead. http://git-annex.branchable.com https://github.com/alebedev/git-media http://git.io/git-lfs
12:39 _ikke_ Note that all solutions are 3rd party solutions, but that should not be a problem
12:40 ash_workz joined #git
12:40 jeffreylevesque joined #git
12:41 ocbtec joined #git
12:41 bejeb enckse: in .net world we publish binaries into nuget packages so they can be pulled in. The packages are built from scripts that are versioned, so if you have the repo then for any given commit you can always produce identical binaries. The binaries are not stored in the repo.
12:42 bejeb */pulled in/pulled in by consumers of the binaries
12:42 sdothum joined #git
12:42 synthroid joined #git
12:42 bejeb and just for extra effect, a jenkins build server is typically used to run that build script so it isn't Works On My Machine developer dependent
12:42 psprint What does "MM" mean in git show --name-status ?
12:45 canton7 psprint, is that on a merge commit?
12:45 troulouliou_div2 joined #git
12:46 sagerdearia joined #git
12:46 psprint canton7: yes
12:46 _ikke_ modified on both sides then
12:47 troulouliou_div2 joined #git
12:48 enckse thanks for the suggestions, I've used nuget and such in the past and we have media (basically we have multiple "different" binary blobs) so I'm very much looking at all solutions
12:48 FilipNortic_ joined #git
12:49 cyphase joined #git
12:50 Lomex joined #git
12:51 Macaveli joined #git
12:52 andlabs joined #git
12:54 oxsyn joined #git
12:56 nikivi joined #git
12:58 DonRichie joined #git
12:59 Hello71 joined #git
13:00 mischat joined #git
13:01 psprint Can there be " M"? Apparently yes?
13:01 Vampire0 psprint, sure
13:02 Vampire0 psprint, left didn't change, right did modify
13:05 esc hey, I was wondering, is there yet a way to force git clone ssh... to use a specific identity?
13:06 bremner ssh host aliases are one way
13:06 cyphase joined #git
13:06 esc I am on a remote server with two ssh keys in my agent, both can connect to github but only one has permissions to fetch a repo
13:06 esc it works locally, but not when doing an ssh-agent forward
13:07 esc mainly because the ssh-agent part of gnome-keyring-daemon is broken
13:07 esc i.e. I am able to clone things locally, but not remotely
13:07 esc ssh-agent -L shows the same keys both locally and remotely
13:07 Vampire0 esc,  use the ssh config file to setup a host alias for github, then assign that host alias the second key and use the host alias instead of github.com
13:07 esc Vampire0: can I alias a key from the ssh-agent?
13:08 esc the key files aren't available locally
13:08 esc on the remote machine
13:08 cdg joined #git
13:08 Rodya_ joined #git
13:08 Vampire0 esc, good question, I'm not sure
13:09 esc I get: ERROR: Repository not found.
13:09 esc fatal: Could not read from remote repository.
13:09 esc Please make sure you have the correct access rights
13:09 esc and the repository exists.
13:09 esc this is a message from github that I get after the ssh auth part is complete
13:10 canuk joined #git
13:10 rorro joined #git
13:10 ExoUNX joined #git
13:10 ilmari are you sure you're using a key belonging to a user that has access to that repo?
13:10 ilmari github doesn't distinguish between not found and access denied, to avoid leaking the existance of private repos
13:12 esc yes, 100% positive
13:12 jimi_ joined #git
13:12 esc the agent has one key that has access and another that doesn't
13:13 des_consolado joined #git
13:14 des_consolado Hey is there a way to ask for all remote branches that have been merged with a particular branch?
13:14 UserOO7 joined #git
13:14 _ikke_ git branch --merged -r
13:14 canuk Is it possible to let git know that a file was renamed and not deleted (after the fact), or do I need to use git mv?
13:14 red82 joined #git
13:14 des_consolado oh cool
13:14 Vampire0 esc, I guess the problem is, that both keys can be used to connect and after the first is tried and was successful of course the second is not tried
13:14 canton7 canuk, git doesn't store that information - it's calculated when needed
13:14 esc Vampire0: that is my guess too
13:14 canton7 canuk, the 'copied' and 'moved' stuff in git status is just for information
13:15 canton7 canuk, !rename
13:15 gitinfo canuk: git does not support explicitly file renaming - the 'git mv' command is shorthand for 'git rm --cached; mv; git add'. All "rename" statistics are generated at runtime when examining history with git-log. See the -M option in the git-log manpage for more info. Also http://thread.gmane.org/gmane.comp​.version-control.git/27/focus=217 and `diff.renames = true`  in man git-config
13:15 esc I'd remove the key from the agent, except that it's not ssh-agent proper but the dreaded gnome-keyring-daemon that comes with ubuntu/unity
13:15 Vampire0 canuk, Git does not record moves or copies explicitly, but it identifies them on-the-fly when using according commands.
13:15 esc and this thing is so buggy, that removing keys from their agent implementation doesn't actually
13:15 esc work
13:15 canton7 bleh, the mail mirror's gone
13:15 lxnch joined #git
13:16 ssitani joined #git
13:16 esc i though that maybe I could bypass that by telling git clone to use a specific identity, but that seems possible only with the ssh config, which in turn only allows key-files, which again are not present on the remote host
13:16 JamesBaxter joined #git
13:17 esc yawn
13:17 hahuang65 joined #git
13:17 esc back to copying across the repo with rsync
13:17 CEnnis91 joined #git
13:17 grawity put the public key on the remote host
13:17 UserOO7_ joined #git
13:17 grawity ssh will use it to pick the correct agent key
13:17 esc grawity: grawity interesting trick
13:18 notebox joined #git
13:18 nikivi joined #git
13:19 mingrammer joined #git
13:19 cyphase joined #git
13:19 Algebr` joined #git
13:19 Algebr` is there a way to say, move/cp everything from one branch to another?
13:20 fmcgeough joined #git
13:20 Algebr` like on one branch I have a directory _site
13:20 CheckDavid joined #git
13:20 Algebr` and I want to mv _site/* to another branch, master
13:20 prawnzy joined #git
13:20 grawity do you want to also get the history, or only the files themselves?
13:21 Algebr` either is satisfactory
13:21 grawity on master, `git checkout otherbranch -- _site`
13:21 Lomex joined #git
13:21 grawity gets you the files
13:21 stuh84 joined #git
13:21 grawity history would involve some subtree mangling
13:21 Algebr` awesome, will try
13:21 Rodya_ joined #git
13:22 esc grawity: lol, I just changed the order in which I load the keys into the agent
13:22 esc if I load the correct key first, the clone works
13:22 grawity :|
13:23 encod3 joined #git
13:23 esc yeah, I can confirm that the order in which one loads keys into the ssh-agent matters!!!
13:24 grawity it matters, sure, but
13:24 grawity *relying* on it sounds like a hack
13:24 esc yeah, definately
13:24 grawity are you going to flush/re-add keys every time you work with another repo?
13:25 codfection joined #git
13:25 esc I presume it doesn't make a difference locally, because one of the keys didn't have a passphrase and was present on disk in .ssh
13:25 morenoh149 joined #git
13:25 esc and I would guess that ssh will try on-disk, passphrase-less keys before asking the agent
13:25 esc but that is a guess
13:26 esc when I move all keys from .ssh to .ssh_backup and add them manually, I can reproduce the behaviour on my local machine
13:26 esc anyway, coming back to your question
13:26 Algebr` grawity: okay insane, someone deleting a directory on one branch makes it disappear on another branch, how is this possible
13:26 Algebr` using jekyll
13:26 blackwind_123 joined #git
13:27 Vampire0 canton7, !rename should be the right one hopefully
13:27 gitinfo canton7: git does not support explicitly file renaming - the 'git mv' command is shorthand for 'git rm --cached; mv; git add'. All "rename" statistics are generated at runtime when examining history with git-log. See the -M option in the git-log manpage for more info. Also public-inbox.org/git/Pine.L​NX.4.58.0504150753440.7211@ppc970.osdl.org and `diff.renames = true`  in man git-config
13:27 Algebr` somehow*
13:27 eizua joined #git
13:27 esc grawity: this is for deployment purposes only
13:27 canton7 Vampire0, yeah I *think* that's the right one - it's been a while since I looked at it
13:27 esc when I work on code I have everything setup correctly locally
13:27 canton7 couldn't find it on google cache or web.archive.org
13:28 sangy joined #git
13:28 esc this issue only arose, because I had to do a hot-needle deployment on a server by sshing in and doing a git-clone on that box
13:28 Vampire0 canton7, http://repo.or.cz/git.git/commitdiff/a37a​17b940f537adbd40c352580ac21cb328b977?hp=8​800e191a80e338a195dd4178fe1496204a3c045
13:29 esc which again hints at much larger issues, since we deploy from github, rather than installing build artifacts
13:29 esc grawity: anyway, thanks for helping me understand, much appreciated!!
13:29 canton7 Vampire0, ah, sweet!
13:29 cdown joined #git
13:31 chelly joined #git
13:31 cyphase joined #git
13:34 Gsham joined #git
13:37 Brain joined #git
13:40 mehola joined #git
13:40 anonymuse joined #git
13:40 arooni joined #git
13:41 deltab joined #git
13:41 Algebr` grawity: checkout used that way brings in the directory, anyway to pull in the contents of that directory directly into the current branch?
13:41 Algebr` besides extra steps by hand
13:41 Algebr` like via git
13:41 grawity ...huh?
13:42 Algebr` _site is a directory, for example it contains index.html. If I just do the checkout, then I get the directory _site. I would rather have everything in _site be checkouted flatly in the current branch
13:42 Vampire0 esc, follow http://superuser.com/questions/273037/u​sing-the-identityfile-directive-in-ssh-​config-when-agentforwarding-is-in-use, which says copy your public key part to the server, use the public key file in the ssh config for IdentityFile and probably also add IdentitiesOnly and you are good to go as the public key is then used to pick the correct private key from the forwarded agent
13:42 Algebr` so that then I have on the branch index.html rather than _site/index.html
13:43 rnsanchez_ joined #git
13:43 Algebr` its just one invokation of mv _site/* .
13:43 Algebr` but neat if could do it directly with git
13:43 ssitani joined #git
13:44 weckl joined #git
13:44 cyphase joined #git
13:44 WayToDoor joined #git
13:44 grawity git read-tree -m -u otherbranch:_site
13:45 grawity actually no,
13:46 cdown joined #git
13:46 sin3t joined #git
13:46 Algebr` grawity: that worked
13:46 fahadash joined #git
13:46 grawity threw away the other files though
13:47 Algebr` meaning the already existing ones in the branch
13:47 Algebr` ?
13:47 Vampire0 probably and probably there are none in your use-case? :-)
13:48 Darcidride joined #git
13:48 Algebr` grawity: this is perfect!
13:48 Algebr` grawity: thank you, need to read the man page now to understand
13:49 danimal2 joined #git
13:50 kulelu88 joined #git
13:50 planet_en joined #git
13:51 encod3_ joined #git
13:51 AaronMT joined #git
13:53 patarr joined #git
13:54 AlexxNica joined #git
13:56 dendazen joined #git
13:58 cyphase joined #git
14:07 ochorocho__ joined #git
14:08 HappyLoaf joined #git
14:10 michael_mbp joined #git
14:12 zeroed joined #git
14:12 Eryn_1983_FL joined #git
14:13 mingrammer joined #git
14:13 JeroenT joined #git
14:13 cdown joined #git
14:14 Darren__ joined #git
14:18 Pasi joined #git
14:19 ThomasLocke joined #git
14:20 wxg joined #git
14:21 PioneerAxon joined #git
14:22 isysd joined #git
14:25 VoidFox joined #git
14:26 robotroll joined #git
14:27 raspado joined #git
14:27 yac left #git
14:27 eclaire joined #git
14:28 trondth joined #git
14:28 Sonderblade i want to reset the second-last commit, can it be done in a easy way?
14:29 Sonderblade easy == doesn't require to many keypresses
14:30 raspado joined #git
14:30 Vampire0 Sonderblade, what does "reset the second-last commit" mean?
14:30 Vampire0 Sonderblade, do you mean *to* the second-last commit?
14:31 michael_mbp joined #git
14:31 * perlpilot wonders what "too many keypresses" means ...
14:31 perlpilot "oops, that's one keystroke too many; it's not easy; I can't do it"
14:31 Vampire0 Sonderblade, and you mean the parent of the current or the grand-parent of the current?
14:31 mangolisk joined #git
14:32 chibby0ne Let's say I want to sparse checkout a repository and I also want to add it as submodule... The steps required to do this, modify files inside the .git directory. Problem is these files are not part of the version control. How would I go about doing this but making it in such a way that anyone that clones the repo gets the submodule and the sparse checkout settings without going ahead and manually recreating the steps?
14:32 hobodave joined #git
14:32 Sonderblade Vampire0: reset as in remove that commit
14:32 absence joined #git
14:33 perlpilot Sonderblade: And "git revert" doesn't work for you?
14:33 absence does anyone know why git for windows includes vim? o_O
14:33 bremner so that git commit works?
14:33 thebope joined #git
14:33 Vampire0 Sonderblade, git rebase -i @~2   then remove the line for the commit and your are done
14:34 thiago joined #git
14:34 absence bremner: does git commit require vim?
14:34 _ikke_ No
14:34 bremner it requires an editor
14:34 bremner dunno if windows ships something that works
14:34 Vampire0 absence, I'd say to have a default editor that works in text mode
14:34 Sonderblade Vampire0: but i want the changes to remain but be uncommitted, like in reset
14:34 Vampire0 absence, if you use another editor you would not need to have vim
14:34 dimi1947 joined #git
14:35 Vampire0 Sonderblade, what do you want to do with the changes? Do you want to squash them to the previous commit, or do you want them applied to the top then, or what?
14:36 Sonderblade applied
14:36 absence bremner, Vampire0: git for windows already ships with a gui editor that it uses by default
14:36 Vampire0 sonderblade still use git rebase -i, then change the ordering of the lines so that your commit in question is bottommost, then after that do the reset
14:36 perlpilot absence: is it called vim?  ;)
14:37 absence either way, if a user wants vim, they'll install vim. it makes about as much sense to include it with git as it would to include an x server
14:37 Vampire0 absence, oh really? which?
14:37 absence perlpilot: i've no idea what it's called. it doesn't look like gvim at least
14:37 absence maybe it's vim concealed, heh
14:37 Sonderblade Vampire0: that's over my keystroke limit, but thanks
14:38 perlpilot Sonderblade: But it's conceptually the easiest thing to do
14:38 absence actually, never mind that. it picks up an editor that doesn't come with git for windows
14:38 Vampire0 Sonderblade, what is your keystroke limit?
14:40 Sonderblade like one or two commands i can remember
14:40 absence how does git know which editor to start?
14:40 perlpilot it starts the one in the EDITOR env var
14:40 perlpilot I dunno what it does wif that's not set
14:40 perlpilot s/wif/if/
14:41 absence yeah, it's not set
14:41 bremner git config core.editor
14:41 perlpilot oh, yeah, and what bremner said :)
14:41 absence ah
14:41 bremner as we all know, because we _all_ did the gittutorial, right?
14:42 Sonderblade Vampire0: your solution: git stash ; git rebase -i @~2 ; <edit edit> ; git reset HEAD~ ; git stash pop
14:42 absence bremner: if i did, it was years ago, and i don't remember :)
14:42 txdv joined #git
14:42 txdv how do I sign my last 5 commits?
14:43 ResidentBiscuit joined #git
14:44 Vampire0 Sonderblade, no problem, you can make this "git r" by defining a git alias
14:44 absence ok, so some git gui tool has stuffed its own editor in global core.editor. i assume it would contain vim on a clean install
14:46 Gsham joined #git
14:46 Vampire0 Sonderblade,     git config alias.x '!git stash && git rebase -i @~2 && git reset @~ && git stash pop'
14:47 Vampire0 Sonderblade, after that you can do "git x", move the line down and you are done
14:47 Vampire0 Sonderblade, if you sometimes want to pull up an even older commit, just increase the 2 in your alias or make it configurable
14:48 sea-gull joined #git
14:49 JanC_ joined #git
14:49 satbyy joined #git
14:49 satbyy joined #git
14:50 subhojit777 joined #git
14:50 Sonderblade Vampire0: i'd like to avoid the interactive part of that. so i can do git evict-commit 7 and have it work automatically
14:52 morenoh149 joined #git
14:52 e14_ joined #git
14:53 tmk joined #git
14:53 leehambley joined #git
14:54 zonality joined #git
14:56 mingrammer joined #git
14:59 dsdeiz joined #git
15:00 hanthings joined #git
15:02 notebox joined #git
15:02 wxg joined #git
15:02 diogenese joined #git
15:03 Vampire0 Sonderblade, sure, no problem:             git config alias.evict-commit '!evict_commit() { git stash && git rebase --onto "@~$1" "@~$(($1-1))" "$(git name-rev --name-only HEAD)" && git cherry-pick -n "@{1}~$(($1-1))" && git stash pop; }; evict_commit'
15:04 Vampire0 Then "git evict-commit 7" will pull up the 7th-last commit
15:04 Balliad joined #git
15:04 Vampire0 If you don't want the changes in the index but only in the worktree, add a git reset before the stash pop
15:05 mizu_no_oto joined #git
15:05 stomplee joined #git
15:06 InfoTest joined #git
15:06 e14_ joined #git
15:08 silverfix joined #git
15:08 Sonderblade Vampire0: seem to work fine. does it work on windows too?
15:09 silverfix hello
15:10 mooncup joined #git
15:10 mooncup hey guys
15:10 mooncup I'm seem to be having submodule issues :<
15:10 Khisanth joined #git
15:10 mooncup !paste
15:10 gitinfo Please use a pastebin for snippets longer than one line. https://gist.github.com and https://vomitb.in both work well.
15:11 silverfix i've changes in my working tree. Is there a way to stash them, change branch, apply them to that branch, commit to that branch, checkout to the previous branch and then unstash changes?
15:11 silverfix that is, i'm writing a script
15:11 mooncup https://vomitb.in/IqkdMkHKhu
15:11 mooncup I'm trying to push up the parent git repo, but it says I have unpushed changes in my submodules
15:12 mooncup however it is also saying there are no unpushed changes in my submodule
15:12 mooncup I'm so confused :<
15:15 wxg joined #git
15:15 wxg joined #git
15:18 hahuang65 joined #git
15:19 Ender13 joined #git
15:19 wxg joined #git
15:20 nidr0x joined #git
15:21 prawnzy joined #git
15:22 harish joined #git
15:22 fuchstronaut joined #git
15:23 kaldoran joined #git
15:27 psprint what does "R100" mean in show --name-status
15:27 thiago 100% content match in a rename
15:27 AlexxNica joined #git
15:27 thiago that is, a pure rename
15:27 psprint ahso, thanks
15:27 wxg joined #git
15:29 al-damiri joined #git
15:30 wxg joined #git
15:30 ShekharReddy joined #git
15:31 wxg joined #git
15:32 hobodave joined #git
15:33 dviola joined #git
15:34 hobodave_ joined #git
15:37 gebbione left #git
15:37 raspado joined #git
15:40 dreiss joined #git
15:40 nedbat I thought there was a git log option that would show the names of the files changed (and ideally, the number of -/+ lines changed), but there doesn't seem to be. What am I thinking of?
15:40 nullie --stat
15:41 dsdeiz joined #git
15:41 nullie nedbat
15:41 nullie ^
15:42 nedbat nullie: YES! thanks
15:43 kpease joined #git
15:45 ahmed_elgabri joined #git
15:46 nedbat ok, hoping someone can double check me here.  I want to see the commits that have touched .underscore files, since branch1 was split off from master, up until the tip of branch2.  This is what I used:   git log --stat $(g merge-base branch1 master)..branch2 -- '**/*.underscore'
15:47 nedbat (oops: g is an alias for git)
15:49 dreiss joined #git
15:51 osse looks good
15:51 synthroid joined #git
15:51 Pasi joined #git
15:51 osse if there are merges between branch1 and master since the original fork point it quickly gets hairy
15:52 endiruna joined #git
15:52 madewokherd joined #git
15:54 dan2k3k4 joined #git
15:54 Rodya_ joined #git
15:55 hanthings joined #git
15:57 le_melomane joined #git
15:58 geoid joined #git
15:59 synthroid joined #git
16:00 EvilPenguin joined #git
16:01 nikivi joined #git
16:02 ahmed_elgabri joined #git
16:04 e14_ joined #git
16:04 nedbat osse: thanks, part of it was from you
16:04 kpease joined #git
16:05 PCatinean_ joined #git
16:06 actinide joined #git
16:08 robotroll joined #git
16:09 Vampire0 Sonderblade, should, yes
16:10 Gsham joined #git
16:10 clemf joined #git
16:11 osse nedbat: i recognized it :p
16:12 durham joined #git
16:13 LeBlaaanc joined #git
16:17 AlexxNica joined #git
16:18 shirty joined #git
16:18 spaceone how to show a specifi file prior to a pecific commit?
16:19 hahuang65 joined #git
16:19 lindenle joined #git
16:21 jnewt joined #git
16:21 spaceone git it
16:21 spaceone git show HASH~1:./file
16:22 mikecmpbll joined #git
16:23 fuchstronaut joined #git
16:23 codfection joined #git
16:23 le_melomane joined #git
16:24 nydel07 joined #git
16:24 cdown joined #git
16:28 tmk joined #git
16:28 Prisoner-2460_1 joined #git
16:28 SwiftMatt joined #git
16:29 thiago joined #git
16:30 hobodave joined #git
16:31 mycelium joined #git
16:32 saml joined #git
16:36 sagerdearia joined #git
16:38 stamina joined #git
16:40 thebope joined #git
16:40 ansimation joined #git
16:40 nedbat osse: what would be a similar command to find commits that *added* .underscore files?
16:40 dreiss joined #git
16:40 LeRieur joined #git
16:40 osse nedbat: --diff-filter=A
16:41 osse ivrhink
16:41 thebope joined #git
16:41 nedbat osse: beautiful! I don't know which is better, git or you :)
16:42 Apteryx joined #git
16:42 durham joined #git
16:42 nitric joined #git
16:42 durham joined #git
16:43 garethdaine joined #git
16:43 mikecmpbll joined #git
16:44 LeRieur joined #git
16:44 MattMaker joined #git
16:44 Pasi joined #git
16:44 osse Mos def me
16:45 osse I am somewhat capable of independent thought
16:45 nedbat that is a plus
16:50 sgrover joined #git
16:50 osse But git doesn't need air or nutrients, except for electricity
16:50 osse soooo
16:53 clemf joined #git
16:53 mingrammer joined #git
16:54 Junior` joined #git
16:54 Junior` left #git
16:54 Junior joined #git
16:54 snowkidind joined #git
16:55 cyan__ joined #git
16:55 Repox joined #git
16:55 freimatz joined #git
16:56 nedbat :) I like your committment to this gag :)
16:56 sgrover Can I indicate a specific merge on develop to merge into master?  We have the usual develop/master branches.  Develop gets a number of merges from the developers that may contain many files.  Do I need to get the list of files and merge the specific files in, for those times when I only need *some* of the changes on the devlop branch moved into master...
16:57 cyan__ joined #git
16:57 nedbat sgrover: are the changes in specific commits? Or files within commits?
16:57 synthroid joined #git
16:57 sgrover usually, yes
16:57 AlexxNica joined #git
16:57 nedbat sgrover: you can cherry-pick commits onto master
16:58 nedbat sgrover: if this happens a lot, then perhaps you want to rethink your flow
16:58 sgrover cherry-pick.. that's the phrase I needed... Off to the Googles I go... Thanks!
16:58 shirty joined #git
16:59 osse sgrover: you can merge a prior commit on a branch as if that was the newest commit
16:59 osse other than that, cherry-pick
16:59 LionsMane joined #git
17:00 sgrover nedbat: We are in the midst of switching from a Feature Branch flow, to a GitFlow workflow.  There is still a lot of mindset insisting on snapshots like you see with the Feature Branches, instead of adopting continuous integration.... So I'm asked to do fun things like... for the next bit only, I hope
17:00 sgrover er... "like this..."
17:02 AnarchyAo joined #git
17:03 mattcen joined #git
17:04 Darcidride joined #git
17:07 Balliad joined #git
17:09 shams left #git
17:10 jimi_ joined #git
17:10 hobodave joined #git
17:11 a3pq51 joined #git
17:11 morenoh149 joined #git
17:12 dreiss joined #git
17:17 txdv I created 5 commits but I forgot to use -S (sign them) on them, is there a singular command I can use to sign those 5 commits retroactively?
17:18 osse txdv: git rebase --exec 'git commit -S --amend' HEAD~5
17:18 osse if that actually works I'm buying myself a new suit
17:21 srcerer joined #git
17:22 anonymuse joined #git
17:22 prawnzy joined #git
17:24 star_prone joined #git
17:25 srcerer joined #git
17:27 Darcidride joined #git
17:28 Darcidride joined #git
17:33 Walakea_ joined #git
17:34 thebope joined #git
17:34 subhojit777 joined #git
17:35 sargas joined #git
17:37 terminal_echo joined #git
17:37 jottr joined #git
17:38 watt_rabbit joined #git
17:38 thebope joined #git
17:40 tang^ joined #git
17:41 stamina joined #git
17:43 Lomex joined #git
17:43 dsdeiz joined #git
17:43 dsdeiz joined #git
17:44 inflames joined #git
17:44 fahadash joined #git
17:46 e14_ joined #git
17:48 mingrammer joined #git
17:51 nd_ joined #git
17:52 Guest18061 joined #git
17:54 greymalkin left #git
17:56 e14_ joined #git
17:58 multi_io joined #git
18:04 chris2 can i run git ls-files in a subdir and see all files in other dirs with ../?
18:05 chris2 ah ls-files :/
18:05 sssilver joined #git
18:05 chandu20 joined #git
18:06 fuchstronaut joined #git
18:07 Repox joined #git
18:08 fstd joined #git
18:08 sargas joined #git
18:12 nd joined #git
18:15 mohabaks joined #git
18:17 nikivi joined #git
18:19 leehambley joined #git
18:19 tyreld joined #git
18:20 fuchstronaut joined #git
18:21 hahuang65 joined #git
18:21 ahmed_elgabri joined #git
18:24 raijin joined #git
18:25 snthsnth joined #git
18:25 Walakea_ left #git
18:25 robotrolll joined #git
18:26 duderonomy joined #git
18:26 user123abc joined #git
18:26 nikivi joined #git
18:27 Pasi joined #git
18:27 eivarv joined #git
18:27 user123abc I'd like to run git filter-branch on a submodule, and then run git-filter branch in a parent project to update each parent commit's submodule reference to point to the newly modified submodule commit
18:28 user123abc I have a way to fetch the desired submodule commit given the original, unmodified submodule commit
18:28 user123abc but I'm not sure how to "update the submodule reference" to this commit in a git filter-branch of the parent project
18:29 user123abc I've tried a tree-filter, wherein I do a submodule foreach which checks out the right commit
18:29 user123abc but that doesn't seem to have an effect: if I then check out the old parent commits, they still point to unmodified submodule commits
18:30 user123abc any advice on how to get this step working? Do you think I might just be doing this tree filter idea wrong, or is there a reason why it shouldn't work?
18:30 robotroll joined #git
18:30 Dazzyp joined #git
18:32 theskillwithin joined #git
18:32 user123abc ah, it looks like git submodule foreach is a no-op in the tree filter
18:32 Pasi joined #git
18:32 Thomsen joined #git
18:33 robotroll joined #git
18:33 Mobutils joined #git
18:35 mrwat joined #git
18:36 leeN joined #git
18:37 eivarv joined #git
18:38 LeRieur joined #git
18:41 user123abc OK, it looks like this is exactly what I want
18:41 user123abc http://stackoverflow.com/questions/1​4785287/how-do-i-use-git-filter-bran​ch-to-update-the-sha-of-a-submodule
18:42 user123abc HOWEVER, I'd like to get submodule paths and commits in an index-filter
18:42 theskillwithin joined #git
18:42 user123abc how can I query the index to tell me this (git submodule doesn't work in index-filter)
18:43 GarethAdams joined #git
18:43 e14_ joined #git
18:43 sargas joined #git
18:43 alex1a joined #git
18:45 pwnawannab[NE] joined #git
18:47 GarethAdams Hi! I've set my Git config to create local branches for Github pull requests, using https://gist.github.com/karlhorky​/88b3c8c258796cd3eb97615da36e07be
18:48 GarethAdams …wait, I just rubberducked my own problem
18:48 star_prone joined #git
18:48 GarethAdams forget I said anything!
18:49 sagerdearia joined #git
18:50 joki joined #git
18:54 planet_en joined #git
18:54 Algebr` joined #git
18:55 SwiftMatt joined #git
18:56 user123abc joined #git
18:57 ochorocho__ joined #git
19:00 cagedwisdom joined #git
19:00 Apteryx joined #git
19:00 e14_ joined #git
19:01 dreiss joined #git
19:03 kulelu88 when you make a commit via SSH to github, how does github restrict the SSH access to that commit only?
19:04 e14_ joined #git
19:04 ochorocho__ joined #git
19:04 preaction kulelu88: you mean restrict to that repository?
19:07 dsdeiz joined #git
19:07 kulelu88 preaction: restrict SSH access in general. what prevents me from doing ssh user@[IP] rm -rf /some/dir
19:07 dsdeiz joined #git
19:07 qsx kulelu88: try it.
19:08 Balliad joined #git
19:08 kulelu88 well not for any user, but for situations where my public key is on their server
19:08 preaction kulelu88: try it
19:08 qsx kulelu88: you can only execute commands like git upload-pack
19:08 leeN joined #git
19:08 preaction if you want to see a free software version of a similar thing, look at gitolite
19:08 snthsnth joined #git
19:08 kulelu88 qsx: how do they restrict access?
19:08 preaction !gitolite
19:08 gitinfo Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/
19:08 ochorocho__ joined #git
19:09 kulelu88 is it some complex setup?
19:09 qsx kulelu88: github?  we can’t know, for it’s closed source
19:09 kulelu88 "with fine-grained access control and many more powerful features."
19:10 qsx but openssh allows one to restrict keys to certain commands and other limitations
19:10 jottr joined #git
19:11 kulelu88 so the restrictions are via openSSH (mostly)?
19:11 m0r0n joined #git
19:11 qsx well it plays an important role
19:11 qsx but so does e.g. gitolite
19:12 sagerdearia joined #git
19:12 kulelu88 aah so that is a problem I probably shouldn't solve right now
19:14 hahuang65 joined #git
19:18 satbyy joined #git
19:18 satbyy joined #git
19:19 navidr joined #git
19:19 tgunr joined #git
19:20 fuchstronaut joined #git
19:23 prawnzy joined #git
19:23 mingrammer joined #git
19:25 cjohnson I believe you could do that with a simple chroot
19:25 cjohnson only copy the allowable commands into your root's bin
19:26 jesk I got a workflow question, therefor lets say this example: I want to share some dotfiles, lets say .zshrc, .zsh/*, .vimrc, and .vim/*. Some guys want to get the newest versions of those, some others just want to get the newest version of .vimrc, nothing else. At the same time they share also files and folders from their home-directories, but not everyone is interested in all of them. My question is now if its
19:26 jesk possible to do this with one repository or does everyone need his own repo and people need to merge between all of them manually?
19:27 cjohnson kulelu88:
19:27 kulelu88 I did find this cjohnson qsx preaction : https://stackoverflow.com/questions/40​2615/how-to-restrict-ssh-users-to-a-pr​edefined-set-of-commands-after-login
19:27 preaction kulelu88: yes. that's how gitolite does it...
19:28 kulelu88 so instead of restricting allowable commands, I will just restrict what can be executed via SSH
19:28 dreiss joined #git
19:28 cjohnson oh that makes sense
19:28 kulelu88 (if that made sense)
19:28 cjohnson right. no interactive login at all
19:28 cjohnson makes sense for a scripted command
19:28 terminal_echo joined #git
19:29 kulelu88 I essentially want to use it for auth without passwords
19:29 ochorocho__ joined #git
19:29 kulelu88 *note: not a webapp
19:29 cjohnson related to git? or for something else
19:31 kulelu88 cjohnson: A simple storage solution, bit I figured I would ask here as someone from the 1200 people here would know more than the *50 (my assumption) odd people (or less) in the #ssh room
19:31 e14_ joined #git
19:31 jesk anyone?
19:32 UserOO7__ joined #git
19:33 kulelu88 jesk: why make life difficult with a complex solution? You can share all the dot files and guys can checkout only the files they want
19:34 jesk what if they want to commit to one of those files?
19:34 preaction jesk: i load my dotfiles into ~/dot-files and then use symlinks to pick out the ones i want into my home directory
19:34 jpic left #git
19:35 kulelu88 they can load the whole repo into a folder, make a commit to that 1 file and push back to merge/review
19:35 e14_ joined #git
19:35 jesk i guess its hard to have a directory where different files correspond to different upstream repos
19:36 Gsham joined #git
19:36 jesk yeah
19:37 cjohnson jesk: they make a branch
19:37 cjohnson in their branch, they modify those files, or maybe replace them entirely
19:37 cjohnson then I guess whenever merging from origin, choose always-mine for those files
19:37 jesk .vimrc is from repo-user-foo while .zshrc is from repo-user-bar
19:38 cjohnson No I mean each person who wants variations does so in that person's own branch
19:38 cjohnson then when they want to update they merge master into their branch
19:38 jesk and when they want to commit their version to the remote repo?
19:38 e14_ joined #git
19:38 preaction yeah, that all sounds terrible
19:39 jesk image 5 users all want to share files with each other
19:39 jesk *imagine
19:39 cjohnson git is not a file sharing solution
19:39 jesk and all are out of their homedirs
19:39 cjohnson that said, I think symlinks are your best bet here
19:39 preaction right, which is why i do my dot files with symlinks
19:40 cjohnson jesk: Unless everybody is truly treating this like a software respository
19:40 jesk so either they all have a complete copy of the same homedir or it wont work
19:40 cjohnson they could branch to their hearts content, make feature branches off master
19:40 jesk if not everyone wants all the things of all others
19:40 cjohnson have their own local integration branch
19:40 preaction if you absolutely want to do it that their homedir is a git repo, yes. but that's not a good idea at all
19:40 cjohnson only merge to master when it's tested
19:41 niels__ joined #git
19:41 cjohnson regularly rebase master from integration to pull in changes from other people
19:41 cjohnson never ever merge integration back into master
19:41 jesk its just hard to push to a centralized repo with restrictions that is not all shit in the homedir
19:41 cjohnson that setup would work if they're all decent at git
19:42 cjohnson jesk: that's another reason for symlinks
19:42 cjohnson my dotfiles repo goes to ~/.dotfiles, and then ~/.vimrc -> ~/.dotfiles/.vimrc
19:42 kulelu88 you don't want the homdir as a git repo. someone will accidentally git commit their pr0n folder
19:42 sargas joined #git
19:43 m0r0n joined #git
19:43 jesk ok thx for thoughts sharing
19:43 cjohnson you can .gitignore and force add every file by hand
19:43 cjohnson but then you lose track of stuff in your git status
19:43 mikecmpbll joined #git
19:44 jesk true
19:44 kulelu88 jesk: sounds like a terrible problem you're dealing with.
19:44 LionsMane joined #git
19:45 kulelu88 problem: "adapting to everyones specific/fine-tuned needs
19:45 planet_en joined #git
19:46 snthsnth joined #git
19:47 mizu_no_oto joined #git
19:48 annoymouse joined #git
19:49 johnmilton joined #git
19:49 eivarv joined #git
19:50 cdown joined #git
19:55 ShekharReddy joined #git
19:57 le_melomane joined #git
19:57 cyphase joined #git
19:58 mjhlobdell joined #git
19:59 SwiftMatt joined #git
19:59 mjhlobdell left #git
20:04 le_melomane joined #git
20:05 hobodave_ joined #git
20:06 yottabyte joined #git
20:06 yottabyte so when I do like git push origin master, does that create a pull request to master?
20:12 cdown joined #git
20:12 fahadash joined #git
20:15 eivarv joined #git
20:15 Prisoner-2460_1 no that just pushes it directly in
20:17 _ikke_ A pull request is basically just a message to someone to fetch and merge a branch
20:18 mingrammer joined #git
20:18 _ikke_ There is nothing in git itself that's called a merge request (though, there is a command called request-pull, the sends an e-mail)
20:18 mablae joined #git
20:20 danimal2 joined #git
20:21 [Brain] joined #git
20:21 fuchstronaut joined #git
20:23 cdg joined #git
20:24 mikecmpbll joined #git
20:25 hashpuppy joined #git
20:29 LionsMane joined #git
20:33 planet_e1 joined #git
20:34 aard_ joined #git
20:35 planet_e1 left #git
20:35 robotroll joined #git
20:38 Balliad joined #git
20:39 m0viefreak joined #git
20:39 MattMaker joined #git
20:40 dsantiago joined #git
20:41 boombatower joined #git
20:45 Sc0tty- joined #git
20:48 Pasi joined #git
20:48 cdown joined #git
20:53 Guest60806 joined #git
20:54 shinnya joined #git
20:55 isysd joined #git
20:56 blackjid joined #git
20:57 cdg joined #git
20:58 niels__ joined #git
20:59 Pasi joined #git
21:00 cdg joined #git
21:01 spat joined #git
21:02 niels__ joined #git
21:02 spat I try to create a post-receive hook but i notice that the git acctions I put in there are not realy performed. This can simply be reproduced by doing: touch ../test.txt;git clean -fd
21:03 spat test.txt apears in the working tree O_o
21:03 m0r0n joined #git
21:03 spat when manualy running git clean it gets cleaned up.
21:03 snthsnth joined #git
21:05 WayToDoor joined #git
21:07 cdg joined #git
21:08 raijin joined #git
21:11 Atm0spher1c joined #git
21:11 thebope joined #git
21:12 mingrammer joined #git
21:12 thebope joined #git
21:13 nikivi joined #git
21:15 Pasi joined #git
21:21 ismaris joined #git
21:22 fuchstronaut joined #git
21:22 fuchstronaut joined #git
21:27 Algebr` joined #git
21:29 cdown joined #git
21:29 tgunr joined #git
21:30 Faylite joined #git
21:31 einnjo joined #git
21:33 thebope joined #git
21:33 ruskie joined #git
21:33 thebope joined #git
21:35 ochorocho__ joined #git
21:38 mizu_no_oto joined #git
21:39 Balliad joined #git
21:39 przemoc joined #git
21:40 krillr joined #git
21:40 tgunr joined #git
21:41 lolmac joined #git
21:41 Gsham joined #git
21:44 m0r0n joined #git
21:46 eivarv joined #git
21:48 rgrinberg joined #git
21:55 e14 joined #git
21:57 Algebr` joined #git
22:01 littlebit joined #git
22:02 niels__ joined #git
22:02 Gsham joined #git
22:04 przemoc joined #git
22:04 przemoc joined #git
22:04 ismaris joined #git
22:06 andlabs joined #git
22:06 littlebit hi people, I'm a begginer at git and started to make my first steps. And I have the following git log tree http://dpaste.com/3RZWW9F when checking out 1d3ba2.. (the very first commit), git tells me that I'm in detached head state. What is to do?
22:07 anuxivm joined #git
22:08 therealGent joined #git
22:08 RingZer0 joined #git
22:09 RingZer0 Old company failed to complete job.  Need "clone" the repo (all commits and what was commited when) from a GitLab install.
22:09 RingZer0 Any thoughts on what that recursive-commit-download command would look like?
22:12 vishal littlebit: !detached
22:12 gitinfo littlebit: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach
22:12 sartan littlebit: you can probably do a local git merge
22:13 madewokherd joined #git
22:13 vishal RingZer0: what do you mean? something more than a 'git clone'?
22:13 littlebit !reattach
22:13 gitinfo Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue
22:15 vishal littlebit: what was your intent when checking out that first commit, and what have you done since then, and what do you want to get to next?
22:17 diogomr joined #git
22:18 littlebit vishal: I wanted to get back there and make a new branch.
22:18 vishal back at the newest commit, or make a new branch from the old 1d3ba2 commit?
22:19 malwar3hun73r joined #git
22:20 malwar3hun73r i have a repository i want to clone that is not a software project, but rather a collection of documents. Is there git way to have updates automatigically pulled?
22:21 ffabi joined #git
22:21 vishal what do you mean by 'udates automatigically pulled'
22:21 qws-user-1229 joined #git
22:21 windsurf_ joined #git
22:21 malwar3hun73r (*updates...) every time there is a change on the remote, i want to have the changes pulled locally
22:22 windsurf_ After changing my password via the git host's website, I get "Permission denied (publickey)" when I try to push. Do I need to generate a new key or can I update it somehow?
22:22 malwar3hun73r i know i could easily write a script and use cron, but i was curious if git supported something like this natively (it looks like it may for submodules...)
22:22 vishal you have to do a git pull to get the updates. git doesn't have a way to automate polling of remotes or anything such, so you have to do it externally, script it or something
22:23 malwar3hun73r vishal, perfect - exactly what i was looking for, thank you!
22:23 fuchstronaut joined #git
22:23 windsurf_ malwar3hun73r some hosts let you add hooks where the host will do an http post to a url on your own web server. You could put a script at that location to pull if your server is the one that needs to git pull
22:24 vishal windsurf_: sounds like a question for your git host, not native to git :)
22:25 vishal though if you specify what git host, there may be someone who's used that..
22:26 malwar3hun73r windsurf_, thanks for the input
22:26 sangy joined #git
22:28 alex1a joined #git
22:28 Apteryx joined #git
22:29 ffabi joined #git
22:34 tgunr joined #git
22:35 dendazen joined #git
22:38 xbob joined #git
22:39 d5sx43 joined #git
22:42 dsantiago joined #git
22:43 chachasmooth joined #git
22:45 DJ_Spies joined #git
22:46 tgunr joined #git
22:47 Milijus joined #git
22:49 jlebar joined #git
22:49 durham_ joined #git
22:51 Darren__ joined #git
22:53 mizu_no_oto joined #git
22:55 texasmynsted joined #git
22:55 aspiers joined #git
22:57 Milijus joined #git
22:58 ochorocho__ joined #git
22:59 Algebr` joined #git
23:00 mingrammer joined #git
23:01 Noldorin joined #git
23:01 sonikspin joined #git
23:02 elastix joined #git
23:02 SwiftMatt joined #git
23:03 TomyWork joined #git
23:04 btree joined #git
23:04 niels__ joined #git
23:04 btree left #git
23:06 prawnzy joined #git
23:08 olalonde joined #git
23:10 prawnzy joined #git
23:15 ResidentBiscuit joined #git
23:15 joki joined #git
23:15 NeverDie joined #git
23:17 prawnzy joined #git
23:18 aidalgol joined #git
23:20 ap0c joined #git
23:22 wxg joined #git
23:22 red82 joined #git
23:23 prawnzy joined #git
23:23 fuchstronaut joined #git
23:24 ffabi joined #git
23:29 isysd joined #git
23:29 raijin joined #git
23:30 joki joined #git
23:32 sagerdearia joined #git
23:32 arescorpio joined #git
23:33 user123abc joined #git
23:36 fahadash joined #git
23:37 Marbug joined #git
23:38 prawnzy joined #git
23:40 joki joined #git
23:43 wxg joined #git
23:44 fedepad joined #git
23:46 Pyrogerg joined #git
23:47 Rodya_ joined #git
23:50 tristanp joined #git
23:50 chachasmooth joined #git
23:51 shgysk8zer0 joined #git
23:52 a3Dman_ joined #git
23:53 zhalla joined #git
23:54 mingrammer joined #git
23:58 zhalla left #git
23:58 red82 joined #git

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