Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-10-12

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

All times shown according to UTC.

Time Nick Message
00:02 nedbat joined #git
00:03 alex1a joined #git
00:04 nitric joined #git
00:08 quizzi joined #git
00:10 MatthewAllan93 joined #git
00:11 _ix joined #git
00:11 Neico joined #git
00:11 SporkWitch joined #git
00:12 poz3idon joined #git
00:14 wololokatamari[m joined #git
00:14 M-shine joined #git
00:15 M-mistake joined #git
00:15 diogenese joined #git
00:17 M-mistake joined #git
00:18 kikijiki[m] joined #git
00:19 lus joined #git
00:19 Ahmed90 joined #git
00:20 Ineentho[m] joined #git
00:20 hallogallo[m] joined #git
00:21 kjsage[m] joined #git
00:21 durham joined #git
00:22 anonymuse joined #git
00:22 M-mistake joined #git
00:23 manosnore joined #git
00:24 ToBeCloud joined #git
00:24 M-meznak joined #git
00:25 imdsm_ joined #git
00:25 eka joined #git
00:25 fscala joined #git
00:27 Mixologic_ joined #git
00:28 star_prone joined #git
00:28 cyphase joined #git
00:29 duarteper joined #git
00:29 Remramm joined #git
00:29 wololokatamari[m joined #git
00:30 MatthewAllan93 joined #git
00:31 McKraken_zzzzz joined #git
00:32 Urbany joined #git
00:34 flopsec joined #git
00:36 diego1 joined #git
00:36 diego1 joined #git
00:36 svankmajer joined #git
00:37 Aces_Charles joined #git
00:38 jor joined #git
00:38 jason237 joined #git
00:39 saustin1 joined #git
00:40 Samual joined #git
00:42 thecomedian joined #git
00:43 yena joined #git
00:43 thebope joined #git
00:43 saustin1 joined #git
00:44 ionelmc joined #git
00:44 Remramm joined #git
00:44 raspado joined #git
00:45 hallogallo[m] joined #git
00:45 kjsage[m] joined #git
00:47 JelmerD joined #git
00:47 n1ax joined #git
00:48 wololokatamari[m joined #git
00:49 xkr47 joined #git
00:49 M-meznak joined #git
00:52 duarteper joined #git
00:53 aviraldg joined #git
00:54 MatthewAllan93 joined #git
00:56 e14 joined #git
00:57 Ineentho[m] joined #git
00:58 eindoofus joined #git
00:59 Remramm joined #git
01:00 gbw459 joined #git
01:01 gbw459 hello all.  "git status" only shows me my untracked folder. I'd like to see that folder and all its contents. Is there a way?
01:03 wololokatamari[m joined #git
01:03 hallogallo[m] joined #git
01:03 Mikaela joined #git
01:03 gbw459 got it! "git status -uall".
01:04 kikijiki[m] joined #git
01:04 M-mistake joined #git
01:06 malcon joined #git
01:07 holodoc joined #git
01:09 oreo joined #git
01:09 saustin1 left #git
01:10 roguewolf_ joined #git
01:10 WizJin joined #git
01:11 roguewolf_ left #git
01:11 DarkFox_ joined #git
01:11 duarteper joined #git
01:12 Cryptek joined #git
01:12 Cryptek Hello?
01:12 gitinfo Cryptek: 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.
01:12 kjsage[m] joined #git
01:12 Cryptek How can I clone and push to a git repository of my organization that I am admin of using the linux git
01:13 mizu_no_oto joined #git
01:13 boombatower joined #git
01:13 Bubbles https://git-scm.com/docs
01:14 Cryptek i tried the remote with that and it gave me some error about fast forwarding
01:14 fstd_ joined #git
01:15 Bubbles !paste
01:15 M-meznak joined #git
01:15 gitinfo Please use a pastebin for snippets longer than one line. https://gist.github.com and https://vomitb.in both work well.
01:15 Bubbles could you paste the error?
01:15 sunri5e joined #git
01:16 oreo joined #git
01:16 kadoban Cryptek: Also be specific what exactly you were trying to do. Did you clone successfully once?
01:16 Cryptek kadoban No I am trying to update a file on the organization and it said could not merge automatically but I dont have the repo on my computer
01:16 Cryptek from the organization
01:17 Cryptek https://vomitb.in/pkI5in9vVW
01:17 kadoban So you never cloned, have no local repo ... and you tried to do what? push? From ... what?
01:18 Bubbles https://i.imgur.com/iZcUNxH.mp4
01:18 Cryptek kadoban the only local repo of the organizations repo is the one I forked, so when I send a pull request from my fork to the official repo in the organization it says we have to manual merge,the organization owner didnt know how to do it, so he made me admin so I could do it and I cant even figure out how to get the organizations repo on my PC with git for linux
01:19 pks joined #git
01:20 yoyoyo____ joined #git
01:20 d^sh joined #git
01:22 kadoban Cryptek: Well, you use 'git clone WhateverUrl' to clone it. I suspect it's going to be a longer process than just that though, you should likely read up on some of the basic operations of git first, https://git-scm.com/book/en/v2 is a decent intro.
01:22 Tee_Pee joined #git
01:22 Cryptek I read the whole thing...
01:22 M-mistake joined #git
01:23 anonymuse joined #git
01:23 kadoban Alright. Well, you need to clone the repo, and do a merge then. Not clear what merge you have to do, that's up to your exact situation and what you're trying to accomplish.
01:23 etcinit joined #git
01:23 kwmiebach__ joined #git
01:24 kadoban As a side note, there was no reason that the other person needed to give you admin rights to do the thing on the upstream repo. You could do a merge on your end and submit a different PR and it'd be fine.
01:24 UKn0Meh joined #git
01:24 UKn0Meh joined #git
01:24 brokensyntax joined #git
01:25 Nugget joined #git
01:25 xameada_ joined #git
01:25 M-mistake joined #git
01:26 Samual joined #git
01:26 ochorocho__ joined #git
01:27 Wulf4 joined #git
01:27 lindenle joined #git
01:27 hallogallo[m] joined #git
01:28 DolpheenDream joined #git
01:28 dendazen joined #git
01:28 acr4 joined #git
01:29 EvilPeng1 joined #git
01:29 M-mistake joined #git
01:30 JelmerD joined #git
01:30 zivester joined #git
01:32 finalbeta joined #git
01:34 sunri5e joined #git
01:35 snowkidind joined #git
01:38 _Vi joined #git
01:38 dodobrain joined #git
01:38 dodobrain joined #git
01:38 jaguarmagenta joined #git
01:46 danslo joined #git
01:46 dodobrain joined #git
01:46 thebope_ joined #git
01:49 ilbot3 joined #git
01:49 Topic for #git is now Welcome to #git, the place for git help and electronically-induced comas | 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: 2.10.1 | Getting "cannot send to channel"? /msg gitinfo .voice | git-daemon: the evil rises
01:50 Jameser joined #git
01:52 watabou joined #git
01:53 dwmw2_gone joined #git
01:53 wololokatamari[m joined #git
01:53 diogenese joined #git
01:53 inflames joined #git
01:54 stack_push joined #git
01:55 Vampire0_ joined #git
01:56 cqi joined #git
01:58 kikijiki[m] joined #git
01:58 stack_push left #git
01:59 Samual joined #git
01:59 Ineentho[m] joined #git
01:59 durham_ joined #git
02:04 Ineentho[m] joined #git
02:04 harish joined #git
02:05 evanwang joined #git
02:06 cdg joined #git
02:07 Nothing4You joined #git
02:07 M-meznak joined #git
02:10 liuzhen joined #git
02:11 sagerdearia joined #git
02:11 star_prone joined #git
02:11 Samual joined #git
02:12 drmt joined #git
02:14 kjsage[m] joined #git
02:14 flaguy48 left #git
02:15 DopeMeston joined #git
02:16 DopeMeston when i do 'git add' how can i add a folder with files? i assume 'git add folder/file.stuff' would add the actual file and ignore the folder itself.
02:18 hho_ joined #git
02:19 cdg joined #git
02:19 [gmi] joined #git
02:19 kadoban DopeMeston: It's recursive, so if you just do 'git add folder/' it'll do everything in there. Is that what you mean?
02:20 StuartMI joined #git
02:21 chachasmooth joined #git
02:21 e14 joined #git
02:21 lagothri1 joined #git
02:21 mangolisk joined #git
02:22 nettoweb joined #git
02:24 DopeMeston kadoban: yes, that is what i meant. thank you
02:24 deltab DopeMeston: it'll record the whole path from the root of the repo
02:26 Goplat joined #git
02:27 cyphase joined #git
02:30 drupald joined #git
02:31 Ardethian\work joined #git
02:32 arescorpio joined #git
02:32 flaguy48 joined #git
02:34 McKraken_zzzzz joined #git
02:35 hexagoxel joined #git
02:35 tristanp joined #git
02:36 ahr3n joined #git
02:36 cyphase_eviltwin joined #git
02:38 Rodya_ joined #git
02:40 svm_invictvs joined #git
02:42 sunri5e joined #git
02:42 tristanp joined #git
02:44 lb1c joined #git
02:45 fedepad joined #git
02:46 thebope_ joined #git
02:47 MrWoohoo joined #git
02:47 fuzzmz joined #git
02:48 etcinit joined #git
02:52 M-shine joined #git
02:55 Zarthus joined #git
02:55 DarkFox_ joined #git
02:57 e14 joined #git
03:01 wololokatamari[m joined #git
03:03 duarteper joined #git
03:03 aidalgol joined #git
03:04 Samual joined #git
03:04 nettoweb1 joined #git
03:05 e14 joined #git
03:06 _ix joined #git
03:07 DarkFox_ joined #git
03:08 hexagoxel joined #git
03:08 Remramm joined #git
03:09 meloc joined #git
03:11 nettoweb joined #git
03:11 KnownSyntax joined #git
03:11 KnownSyntax joined #git
03:12 aviraldg joined #git
03:14 Vortex35 joined #git
03:14 Iacobus joined #git
03:15 wololokatamari[m joined #git
03:17 eka joined #git
03:19 sunri5e joined #git
03:19 Samual joined #git
03:21 M-mistake joined #git
03:26 MatthewAllan93 joined #git
03:27 mehola joined #git
03:31 Rewneyo joined #git
03:32 perrier-jouet joined #git
03:33 lindenle joined #git
03:36 Samual joined #git
03:36 hallogallo[m] joined #git
03:39 samtheknife joined #git
03:39 jaguarmagenta joined #git
03:43 xameada joined #git
03:44 danslo joined #git
03:50 Samual joined #git
03:50 a_thakur joined #git
03:51 nowhereman joined #git
03:51 rchavik joined #git
03:59 xall joined #git
04:02 a_thakur_ joined #git
04:06 Samual joined #git
04:07 Samual joined #git
04:08 cqi joined #git
04:08 laumann joined #git
04:08 duarteper joined #git
04:08 aviraldg joined #git
04:09 nd_ joined #git
04:09 mangolisk joined #git
04:11 kjsage[m] joined #git
04:12 rscata joined #git
04:13 alex1a joined #git
04:14 nilg joined #git
04:17 manosnore joined #git
04:18 red82 joined #git
04:19 gitinfo joined #git
04:19 Kitty- joined #git
04:23 brillenfux joined #git
04:24 mmlb joined #git
04:26 gyre007 joined #git
04:26 xall joined #git
04:28 phanimahesh joined #git
04:33 red82 joined #git
04:34 DarkFox_ joined #git
04:36 romerocesar joined #git
04:37 Samual joined #git
04:38 fscala joined #git
04:38 NoOutlet joined #git
04:40 jaguarmagenta joined #git
04:44 duarteper joined #git
04:45 dsantiago joined #git
04:46 Cabanossi joined #git
04:47 kjsage[m] joined #git
04:47 wololokatamari[m joined #git
04:47 thebope joined #git
04:48 M-meznak joined #git
04:48 hallogallo[m] joined #git
04:48 eka joined #git
04:48 bmahe joined #git
04:49 _Vi joined #git
04:50 Samual joined #git
04:53 Corosus joined #git
04:54 daey_ joined #git
04:55 duarteper joined #git
04:56 freimatz joined #git
05:00 svankmajer joined #git
05:00 dermoth joined #git
05:01 arand joined #git
05:03 _Vi joined #git
05:03 firebalrog joined #git
05:06 eizua joined #git
05:06 alex1a joined #git
05:08 phanimahesh joined #git
05:10 navidr joined #git
05:11 LucaTM joined #git
05:15 jnewt joined #git
05:16 satya4ever joined #git
05:18 mycelium joined #git
05:20 archaic joined #git
05:21 virengos joined #git
05:22 shaggycat joined #git
05:22 kikijiki[m] joined #git
05:23 overlord_tm joined #git
05:25 phanimahesh joined #git
05:26 lindenle joined #git
05:26 SwiftMatt joined #git
05:27 mehola joined #git
05:28 DarkFox_ joined #git
05:28 Corosus joined #git
05:30 Raging_Hog joined #git
05:32 evanwang_ joined #git
05:35 Ardethian\\work joined #git
05:35 kbdkode joined #git
05:36 jast joined #git
05:37 watersoul joined #git
05:38 artgoeshere joined #git
05:38 habbasi joined #git
05:38 yanope joined #git
05:41 ThomasLocke joined #git
05:45 _Vi joined #git
05:46 danslo joined #git
05:47 watabou joined #git
05:48 thebope joined #git
05:49 liuzhen joined #git
05:51 mangolisk joined #git
05:56 jknetl joined #git
05:56 bocaneri joined #git
05:57 smithbone joined #git
05:57 jknetl joined #git
05:58 Doginal joined #git
05:59 fedepad joined #git
06:00 M-meznak joined #git
06:00 hallogallo[m] joined #git
06:05 xall joined #git
06:09 khfeng joined #git
06:10 _Vi joined #git
06:11 zeroed joined #git
06:12 brokensyntax joined #git
06:13 kbdkode joined #git
06:16 aax joined #git
06:16 qt-x joined #git
06:17 freimatz joined #git
06:17 roelmonnens joined #git
06:19 fub joined #git
06:19 ahr3n joined #git
06:20 phanimahesh joined #git
06:22 [Brain] joined #git
06:27 aax joined #git
06:28 finalbeta joined #git
06:28 JeroenT joined #git
06:29 JeroenT joined #git
06:29 jnavila joined #git
06:36 cyan__ joined #git
06:39 Doginal_ joined #git
06:40 ogny joined #git
06:40 Sceorem joined #git
06:43 _Vi joined #git
06:44 maret joined #git
06:47 Fijit joined #git
06:48 chele joined #git
06:49 eshlox joined #git
06:50 PCatinean joined #git
06:51 tvw joined #git
06:52 evanwang joined #git
06:53 Raed joined #git
06:55 ome joined #git
06:56 StuartMI joined #git
06:58 elect joined #git
06:58 jaguarmagenta joined #git
06:59 lrvick joined #git
06:59 lrvick Is there not a way to use "git submodule" commands without changing $PWD to the repo root?
07:00 dan2k3k4 joined #git
07:00 _ikke_ I might misremember, but I thought newer versions of git don't require this anymore
07:00 lrvick git --git-dir=/home/lrvick/rootrepo/.git --work-tree=/home/lrvick/rootrepo submodule add git@github.com:lrvick/subrepo.git subfolder
07:00 lrvick this fails with
07:01 lrvick fatal: $program_name cannot be used without a working tree.
07:01 lrvick Terrible error as well
07:01 thiago joined #git
07:01 _ikke_ What version of git do you have?
07:01 lrvick 2.10.0
07:01 _ikke_ ok
07:01 _ikke_ lrvick: can you try -C instead of --git-dir and --work-tree ?
07:02 Alienpruts joined #git
07:05 lrvick That appears to do what I want.
07:05 diogenese joined #git
07:05 lrvick now to try to script this in a backwards-compatible way
07:05 lrvick thanks!
07:05 twb joined #git
07:05 King_Hual joined #git
07:05 King_Hual joined #git
07:06 _ikke_ (cd dir && git ...) :P
07:06 diogenese joined #git
07:06 twb OK so just doing "git clone git://..." works for me.  But some repos, I don't need a working tree, just to git grep the source
07:07 phanimahesh joined #git
07:07 twb So I do "git clone --bare git://...", which works initially.
07:07 twb But "git fetch" seems to not DWIM
07:07 _ikke_ What do you expect it to do?
07:07 lrvick _ikke_: when was -C introduced? Any git versions not supporting it still in widespread use?
07:07 twb _ikke_: download any more patches that have landed upstream since I last cloned/fetched
07:07 osse lrvick: 1.9.5
07:07 goTAN joined #git
07:08 twb It might just be one repo that's behaving oddly -- git://github.com/systemd/systemd  is where I notice it
07:08 dunaeth joined #git
07:08 star_prone joined #git
07:08 twb Looking in the resulting systemd.git, it seems like there's no remotes set up from "git clone --bare" ?
07:09 _ikke_ osse: I have 1.8.5 in my doc
07:09 twb The actual symptom I see is after a month someone will be talking about a new systemd release and I won't see it at all, and "git fetch" won't make it appear, but if I do a fresh clone I can see the commits and tags.
07:09 _ikke_ Yeah, relnotes also say 1.8.5
07:10 twb Oh and this is on the storage pool's ridiculously old git -- 1.7.0.4
07:10 _ikke_ On a bare repository, the refspecs should be set up to update the local refs
07:10 ij left #git
07:11 _ikke_ twb: ^
07:11 osse _ikke_: now I got flashbacks to the last time I got this wrong
07:11 twb _ikke_: OK, but should "git fetch" pull new stuff from where I originally cloned from?
07:11 dreiss joined #git
07:11 _ikke_ http://tinyurl.com/git-changes
07:11 _ikke_ twb: It should, yes
07:12 _ikke_ That URL is not up-to-date though
07:12 twb OK, thanks for confirming :-)
07:12 star_prone joined #git
07:12 overlord_tm joined #git
07:12 goTAN joined #git
07:13 thebope joined #git
07:13 TomyWork joined #git
07:13 danslo joined #git
07:13 twb Here's what I see: http://sprunge.us/IRSQ
07:14 _ikke_ Right, you either have to specify a remote, or --all
07:15 _ikke_ but after you just cloned, fetching does not do much indeed
07:15 twb _ikke_: shouldn't git find a copy of the original clone URL *somewhere* tho?
07:15 _ikke_ twb: You can use git remote -v to show the remotes
07:15 twb git remove -v has no output
07:15 _ikke_ hmm, that's odd
07:15 twb Maybe this is the fault of ~/.gitconfig
07:15 twb I'll try again with that moved away
07:16 qpdb joined #git
07:16 _ikke_ No, that should not matter
07:16 _ikke_ the remote is setup in the local config
07:16 _ikke_ Let me test
07:16 twb I'll also do a test on my netbook with a newer git
07:19 _ikke_ Yeah, for me, the remotes are added
07:19 _ikke_ So might be a problem with the older version
07:19 kbdkode joined #git
07:19 twb OK, at least it's not me going mad
07:20 twb Reproduced with a smaller repo: http://sprunge.us/NjVQ
07:22 _ikke_ On the same box / same git version?
07:22 twb different hosts & versions, version is shown in the test
07:22 _Vi joined #git
07:22 twb That evidence is good enough for me to jsut dismiss this as "1.7 is weird"
07:23 mithenks joined #git
07:23 _ikke_ Right
07:24 dc5ala joined #git
07:24 WayToDoor joined #git
07:24 fedepad joined #git
07:25 apotry joined #git
07:27 mikecmpbll joined #git
07:27 _ikke_ osse: What kind of flashbacks?
07:29 Balliad joined #git
07:30 osse _ikke_: not the fancy vietnam kind. just the "oh, like that time when" style of flashbacks
07:33 Murii joined #git
07:33 _ikke_ right, wast just curious about the story
07:34 MatthewAllan93 joined #git
07:34 BadSnappa joined #git
07:34 JeroenT joined #git
07:34 gtristan joined #git
07:34 Celelibi joined #git
07:37 moei joined #git
07:38 osse _ikke_: http://sprunge.us/bRKf
07:38 cyan__ joined #git
07:38 Doginal joined #git
07:38 rcn joined #git
07:39 elect joined #git
07:39 gtristan Hmmm, so I have this odd issue I wonder if I can solve with git in some way
07:39 Dazzyp joined #git
07:39 lrvick What is the most succient way to write a condition that says "Am I in the top level of a git repo"? [[ -d .git ]] worked until I started playing with submodules where .git is a file
07:39 gtristan The problem revolves around a huge git repository in which various work is done, some of it is shared but some of it clearly belongs to separate projects
07:39 lrvick then I figure git must have a more sane built-in way to do this
07:40 lrvick [[ $(git rev-parse --git-dir) == "$PWD" ]]
07:40 lrvick is the best I have come up with so far
07:40 osse lrvick: git rev-parse --show-toplevel
07:40 gtristan The essence of it is a build system, imagine something like yocto for instance, with many different things which can be built and some common components, it doesnt all belong together; so I'm thinking submodules or subtrees
07:42 gtristan So I would have a dependency chain where recursively... the low level module provides the bootstrapping glibc/gcc and kernel builds, then there is some middleware modules which depend on it (or include the bootstrap as a submodule), and in turn some full systems depend on the middleware, like a recursive submodule setup
07:42 corsibu joined #git
07:43 gtristan The problem is this: I would really like for A.) The depending module to be in charge of what commit of it's dependency it's going to use (what submodules provide already) but... B.) I would really like the checkouts to be *adjacent* to eachother
07:43 gtristan not nested
07:44 gtristan So I could have a directory structure where in the project root I would have: bootstrap, middleware, os... all checked out beside eachother
07:44 _ikke_ osse: oh lol, same exact feature :P
07:44 gtristan Could I have a submodule that initializes at a path such as "../bootstrap" ?
07:45 gtristan (sorry for being so wordy in my inquiry, I'm just trying to ponder what the solution might be)
07:45 _ikke_ I default to using the RelNotes, and fall back to grepping history if that fails
07:45 osse _ikke_: what does "use" mean there? grepping?
07:46 _ikke_ correct
07:46 osse I got this habit working on completion scripts. finding out where some undocumented stuff comes from etc
07:46 _ikke_ right
07:47 osse For a small while  --find-renames was called --detect-renames for example. It was never in a release, yet made it into zsh's completion
07:47 _ikke_ Oh, that's annoying
07:47 dan2k3k4 joined #git
07:48 elect joined #git
07:48 star_prone joined #git
07:50 ahr3n joined #git
07:50 phanimahesh gtristan: submodules can't be adjacent to parent. But you can create a parent repo with the dependency and the dependant as adjacent submodules.
07:52 omx joined #git
07:52 encod3 joined #git
07:53 goTAN joined #git
07:53 gtristan phanimahesh, then I would have 2 repositories for any module which depends on another module correct ?
07:53 gtristan I could do that
07:53 mithenks joined #git
07:54 gtristan I see the message fatal: ../bootstrap-repo: '../bootstrap-repo' is outside repository
07:54 osse Also it has completion for some log options that were never, and never will be, documented. I wonder how they guy who wrote that worked
07:54 yaiyan joined #git
07:54 gtristan so I cannot add a submodule programmed to be checked out in an adjacent directory indeed
07:54 _ikke_ osse: Probably following the mailing list or something?
07:55 _ikke_ osse: Or parsing the git option parser logic
07:55 kurkale6ka joined #git
07:56 gtristan One separate inquiry... I've been told that subtrees are more user friendly than submodules; but after looking into it, I have this 'eek' feeling that I am mangling the payload/history of what should be in another project into the parent project owning a subtree
07:56 _ikke_ gtristan: Yeah, they have completely different usecases
07:56 gtristan Is there a way to use subtrees which does not consume the child repo ?
07:56 _ikke_ no
07:57 _ikke_ It's a merge algorithm
07:57 gtristan Ok, that was my initial gut feeling
07:57 _ikke_ It merges history
07:57 _ikke_ So you cannot compare the two
07:57 gtristan I see, I dont like it for this purpose, my goal is to cleanly separate what is already an entangled mess
07:57 mikecmpbll joined #git
08:00 ahmedelgabri joined #git
08:00 ocbtec joined #git
08:01 nilg joined #git
08:04 bongjovi joined #git
08:04 Andrew_K joined #git
08:06 star_prone joined #git
08:06 dcpc007 joined #git
08:07 finalbeta joined #git
08:07 eka joined #git
08:08 a_thakur joined #git
08:09 Yuki_ left #git
08:09 pijiu joined #git
08:11 laumann joined #git
08:11 _ADN_ joined #git
08:11 _Vi joined #git
08:11 ki0 joined #git
08:11 pijiu joined #git
08:12 pijiu joined #git
08:13 * gtristan considers a thoroughly horrible hack where a git repo holds a symlink at the root "${repo}/project-root -> .." and then git submodule add <repo> project-root/dependency
08:13 gtristan heh
08:13 gtristan mostly for comic relief I guess
08:13 pijiu joined #git
08:15 pijiu joined #git
08:15 z8z joined #git
08:16 pijiu joined #git
08:17 Mir4g3 joined #git
08:17 qws-user-1228 joined #git
08:17 re1 joined #git
08:18 duarteper joined #git
08:19 marcogmonteiro joined #git
08:20 eka_ joined #git
08:21 JeroenT_ joined #git
08:23 finalbeta joined #git
08:23 redeemed joined #git
08:24 noidea joined #git
08:24 noidea joined #git
08:25 re1 joined #git
08:26 jbitdrop joined #git
08:27 leeN joined #git
08:28 mischat joined #git
08:29 vuoto joined #git
08:31 mquin joined #git
08:34 JeroenT joined #git
08:36 qpdb joined #git
08:37 corsibu joined #git
08:39 hello_world_hell joined #git
08:44 wololokatamari[m left #git
08:47 \du joined #git
08:49 halbomin joined #git
08:50 soerenb joined #git
08:50 Orphis joined #git
08:51 fscala joined #git
08:51 M-meznak joined #git
08:53 lqi joined #git
08:53 lqi where can I find "git gui" or "git-gui" on mac? thanks
08:55 Remramm joined #git
08:56 qpdb joined #git
08:59 GavinMagnus joined #git
08:59 jaguarmagenta joined #git
09:01 z8z left #git
09:01 JeroenT joined #git
09:02 evanwang_ joined #git
09:04 dunaeth joined #git
09:04 KnownSyntax joined #git
09:04 KnownSyntax joined #git
09:06 eb joined #git
09:09 thiago joined #git
09:10 evanwang joined #git
09:10 mischat joined #git
09:10 fees joined #git
09:13 xbob joined #git
09:13 miguelfernandes joined #git
09:13 thebope joined #git
09:14 duarteper joined #git
09:14 wagle what does it mean when a DIRECTORY is untracked (but the file inside it is there fine)?
09:15 selckin directories are never tracked
09:15 wagle $ git status
09:15 wagle On branch master
09:15 wagle Untracked files:
09:15 wagle (use "git add <file>..." to include in what will be committed)
09:15 wagle mozilla/addons/webextensions/
09:15 wagle nothing added to commit but untracked files present (use "git add" to track)
09:16 selckin whats the problem
09:16 _ikke_ It means that non of the files in webextensions is tracked
09:17 _ikke_ use git status -uall to see all files, instead of them grouped together
09:17 wagle its got one file in it
09:18 wagle oic, I just modified that file though
09:18 Ineentho[m] joined #git
09:19 wagle how odd
09:20 wagle this wiki uses git to track pages, it must be confused
09:20 wagle .. about editing files that arent tracked
09:20 gniourf joined #git
09:20 t4nk488 joined #git
09:21 wagle thanks for the -uall
09:21 t4nk488 left #git
09:25 xall joined #git
09:27 sylario joined #git
09:28 rahtgaz joined #git
09:29 ieee joined #git
09:29 lindenle joined #git
09:30 elect joined #git
09:31 WayToDoor joined #git
09:36 star_prone joined #git
09:37 JeroenT joined #git
09:38 mischat_ joined #git
09:38 star_prone joined #git
09:38 goTAN joined #git
09:39 dodobrain joined #git
09:40 JeroenT joined #git
09:41 Wayne_Shephard joined #git
09:42 duarteper joined #git
09:43 Filiphe joined #git
09:44 ionelmanolache joined #git
09:44 mischat joined #git
09:48 nidr0x joined #git
09:49 rorro joined #git
09:50 DJ_Spies joined #git
09:50 phanimahesh joined #git
09:50 maret joined #git
09:51 Darcidride joined #git
09:52 thiago joined #git
09:53 liuzhen joined #git
09:56 qt-x joined #git
09:58 qt-x joined #git
10:01 Silenced_v2 joined #git
10:01 freimatz joined #git
10:02 Silenced_v2 joined #git
10:03 cdown joined #git
10:09 amaged__ joined #git
10:10 DeaTheAngel joined #git
10:10 finalbeta joined #git
10:12 bittin joined #git
10:12 bittin Github talks in Stockholm today :)
10:14 phanimahesh joined #git
10:15 kjsage[m] joined #git
10:16 thebope joined #git
10:18 duarteper joined #git
10:18 courrier joined #git
10:19 manuel joined #git
10:24 cagedwisdom joined #git
10:27 Ineentho[m] joined #git
10:29 Gonzo89 joined #git
10:30 Snugglebash joined #git
10:32 finalbeta joined #git
10:33 dispix joined #git
10:34 star_prone joined #git
10:34 dispix Hey guys, do any of you had the opportunity to work with submodules, and even nested submodules ?
10:34 tobiasvl I've worked with submodules, but not nested
10:35 lqi some experiences with submodules
10:35 dispix What did you think of it ? Was it helpful or a pain in the ass ?
10:35 demize bittin: When/where?
10:36 barteks2x joined #git
10:37 tobiasvl both
10:38 fub joined #git
10:38 manuel joined #git
10:38 theskillwithin joined #git
10:40 manuel joined #git
10:40 a_thakur joined #git
10:43 lqi dispix: too general question...
10:43 M-meznak joined #git
10:44 manuel joined #git
10:46 JeroenT joined #git
10:50 Andrew_K joined #git
10:51 Xethron joined #git
10:51 harish joined #git
10:54 kam270 joined #git
10:55 theskillwithin joined #git
10:55 shaggycat joined #git
10:59 drmt joined #git
11:00 mehola joined #git
11:00 jaguarmagenta joined #git
11:01 DarkFox_ joined #git
11:02 acetakwas joined #git
11:04 acetakwas joined #git
11:04 anuxivm joined #git
11:05 BPL joined #git
11:06 acetakwas joined #git
11:06 BPL Hi, anyone could help me here? It seems I've rebased one feature of mine in the wrong branch ("next") instead of "master", here's how it looks now http://screencast.com/t/u3kusBWlsstd , how could i fix it up?
11:08 osse BPL: what is the name of your feature?
11:09 acetakwas joined #git
11:09 osse BPL: doesn't really matter. check it out, then do  git rebase --onto=master next
11:09 BPL osse: The top one, 'cmExtraSublimeTextGenerator: Feature request (#16351)', as you can see the parent is 'Merge topic 'extract-cmRulePlaceholderExpander' into next'
11:09 BPL osse: And i'd want the parent to be 'CMake Nightly Date Stamp' i guess
11:10 osse yup, this'll do the trick
11:10 dan2k3k4 joined #git
11:10 BPL alright, let me try! :)
11:10 acetakwas joined #git
11:11 dsfreak93 joined #git
11:12 BPL osse: Just to clarify, I've checked out "st2-exclude-patterns-variable" branch and now I'll do `git rebase --onto=master`, is that correct? Or in you previous statement you meant `git rebase --onto=master next`
11:12 corsibu joined #git
11:13 liuzhen joined #git
11:13 osse BPL: I meant what I wrote :P I see how it's ambiguous. Yes, next is part of the command line
11:14 BPL hehe, ok, I asked cos there is this branch called "next", alright
11:14 wrksx joined #git
11:14 osse BPL: https://raw.githubusercontent.com/Osse/git-stuff/master/rebase-onto.txt
11:14 BPL osse: Mmm, what's the meaning of this http://screencast.com/t/ej6mWWcUa4 ?
11:15 osse f
11:15 osse oos
11:16 osse Oh, you need to write stage/next or origin/next
11:16 BPL i see, sorry about these questions, as you can see I'm a total noob with git, guess this is pretty basic stuff :P
11:17 eduardas_m joined #git
11:18 BPL osse: Mmm, i think this is good now? http://screencast.com/t/0HIC4x5ZOvJ , should i do git push --force-with-lease now?
11:20 osse Looks good to me. Yes.
11:20 gtristan So everyone really hates submodules, and I dont know what to tell them; the result is people doing nasty things around git subtree, wrapping it up with "git-subrepo", and working around submodules because of painful experiences
11:21 gtristan I want submodules, it makes sense for the project workflow; but people will get bitten, and most automatically reject the whole idea based on the fact that the proposal text contains "git submodule" at all
11:21 tvw joined #git
11:22 WayToDoor joined #git
11:22 P4nther joined #git
11:22 gtristan perhaps there is some project like git subrepo, which instead of trying to virtualize submodules with subtrees, improves the work flow on top of actual submodules ?
11:22 phanimahesh joined #git
11:23 _ikke_ gtristan: Do you have an idea how submodules could be improved?
11:24 aviraldg joined #git
11:24 _ikke_ A lot of things people run against are consience decissions
11:24 gtristan _ikke_, if you search https://github.com/ingydotnet/git-subrepo/blob/master/doc/intro-to-subrepo.swim for the text "Bad:", there are some starting points, some of which I'm skeptical to believe, even
11:24 BPL osse: Sorry, one last question, what's the meaning of being 3 ahead and 235 behind?
11:24 Seveas gtristan: take a step back: why are you using submodules, and can the problem you're solving with submodules be solved in a different way?
11:24 jaguarmagenta joined #git
11:25 gtristan One thing is, probably --recursive should be available in more options than just clone (i.e. git checkout --recursive ?)
11:25 a_thakur_ joined #git
11:25 khisanth_ joined #git
11:26 gtristan Seveas, Are you familiar with yocto for instance ? This is not yocto for the record, but is similar
11:26 Seveas gtristan: I have no idea what yocto is
11:26 gtristan Seveas, I have a huge tree of build instructions which is basically capable of bootstrapping every possible linux based OS under the sun
11:27 Seveas sounds interesting
11:27 gtristan So, I have directories which contain the gcc cross/native build and glibc and busybox etc
11:27 gtristan I have directories with coreutils and then the build tools like make/automake etc
11:27 lb1a BPL it means, your branch has 3 commits, that are not in it's tracking branch, but the tracking branch has 235 commit that your branch doesn't have
11:28 Seveas gtristan: so you're using submodules for dependency management?
11:28 gtristan Seveas, these "layer" onto eachother
11:28 gtristan Seveas, No, I am currently living in a sort of hell where we are dragging everything along in the same git repo
11:28 gtristan That's what I want to change
11:28 Seveas so you copied the make/busybox/glibc etc sources into your git tree?
11:28 osse BPL: it means the current branch has 3 commits the other one doesn't, and the other one has 235 commits this branch doesn't
11:28 Seveas Hell would be an apt description :)
11:28 BPL lb1a: What's the tracking branch? How do i check it out using git in the command line?
11:29 gtristan Seveas, no, the build instructions, not the sources
11:29 buzzLightBeer joined #git
11:29 ash_workz joined #git
11:29 _ikke_ A tracking branch is branch that keeps track of what you fetched from the remote from that branch
11:29 ojdo joined #git
11:30 lb1a BPL !tracking
11:30 gitinfo [!remote_tracking_branch] Remote-tracking branches (branches which start with e.g. 'origin/', listed by 'git branch -r') are read-only mirrors of the corresponding branches in another repository. They're updated by 'git fetch'. You can't edit them directly (trying to check them out results in a !detached HEAD), but you can create a new local branch based on a remote-tracking branch using e.g. 'git checkout -b <branch> <remote
11:30 gtristan Seveas, each "chunk" refers to a git repo where the sources for every module is stored, and yeah for that we also use exclusively git (we mirror everything, tarballs, svn, cvs into a normalized git source)
11:30 rorro joined #git
11:30 _ikke_ BPL: git branch -vv shows what branch and the status
11:30 goTAN joined #git
11:31 qpdb joined #git
11:32 lindenle joined #git
11:32 BPL _ikke_: Alright, I see this http://screencast.com/t/IqNESAKT7YYk , I guess * indicates the branch I'm working on but still I don't understand what's comparing, ahead&behind with respect to which branch?
11:33 anonymuse joined #git
11:33 osse compared to the one inside the [ ]
11:33 osse which looks to be the "same" branch but the one you have in your fork.
11:34 gtristan Seveas, So basically its a huge tree which describes how the entire world can be built. I want for example, to have a recursive submodule situation, where the gnu/linux base module has a dedicated maintainer, coreutils another, and that a big GNOME system for example, would have freedesktop & X11 as a submodule, which would in turn have coreutils and built utilities as a submodule, and recursively submodule down to the bootstrap
11:35 Seveas gtristan: I'd keep everythin in one big repo in this case
11:35 gtristan The idea is that when I maintain an OS build, I know exactly what version I'm depending on, and changes from down below cannot be *pushed* upon me, I need to track a newer version if I want to build with newer coreutils or such
11:35 gtristan That is tried and horrible
11:35 BPL osse: Ok, right, +/- clear... so the left column is for local branches and the right column remote ones, this is --> st2-exclude-patterns-variable (local)  and brupelo_fork/st2-exclude-patterns-variable (remote)
11:35 aviraldg joined #git
11:35 _ikke_ right
11:36 Darcidride joined #git
11:36 _ikke_ brupelo_fork is the name of the remote that branch came from
11:36 gtristan Seveas, what happens is some guy decides that in master, systemd should not be built --with-compat-libs... and since everything is building in the same repo, most of those end operating system builds break
11:36 osse BPL: the reason the remote one has so many commits is that it's still based on next.
11:36 osse you're right that you should push --force(-with-lease)
11:37 gtristan until the person concerned with their OS finally comes around and sees that someone else broke it by changing how some shared component is built, or by upgrading some library to an API incompatible version
11:37 BPL _ikke_: Alright, so git will always know which remote to compare from, right? I mean, you always create a local branch from 1 remote
11:37 nuck joined #git
11:38 _ikke_ Every remote has it's own set of remote tracking branches, but each local branch only tracks one remote branch
11:39 BPL _ikke_: Cool, I get it now  osse: Any clue about this http://screencast.com/t/G4MAfhISNs?
11:39 dede_ joined #git
11:39 gtristan Seveas, another problem with keeping everything in the same repo is that it creates this sort of cult where one group of people is in control of everything
11:39 mizu_no_oto joined #git
11:40 gtristan Seveas, if we want people to adopt the tech, we would prefer to allow people to consume "this part" to build "their OS or appliance" on top of
11:40 gtristan forking the huge thing is such a mess
11:40 gtristan and results (like with openembedded/yocto) in a huge repository of deadcode
11:40 osse BPL: that's an issue with your git install
11:41 osse you've told git to use plink, but it cannot be found
11:41 gtristan one cannot look at a file and say "hmm how does this contribute to the build", no; 90% chance, it does not effect your build at all
11:41 BPL osse: I see, ok, I'll add it PATH
11:41 osse BPL: are you using the Git Bash thingy?
11:42 BPL osse: Indeed, I got an env var called GIT_SSH=plink, I also got one called SSH_AUTH_SOCK=c:\cygwin64\tmp\.ssh-pageant one (i think this was to use cygwin's ssh in the past)
11:42 WayToDoor joined #git
11:43 jimi_ joined #git
11:43 BPL osse: Right now I'm using the git bash thingy, yeah, ie: "C:\Users\KneDa\AppData\Local\Atlassian\SourceTree\git_local\mingw32\bin\git.exe" , the one provided with sourcetree
11:43 aapaaal joined #git
11:44 BPL osse: To be honest, nowadays I don't know very well the real meaning of GIT_SSH and SSH_AUTH_SOCK , i'll google it
11:45 nuc joined #git
11:45 osse BPL: for command line stuff I suggest installing the regular Git For Windows thingy, and pick openssh as its ssh implementation
11:45 osse you get nice start menu shortcuts, it just works, etc.
11:46 BPL osse: But the one provided by Attlasian sourcetree is the latest git available, i mean --> git version 2.8.1.windows.1
11:48 gtristan So I am getting the recommendation that I should instead of using submodules, I should use git subrepo: With some custom git hooks installed to ensure that someone can never make any commits which touch files in a subrepo: I.e. ensuring that users/collaborators are forced to push changes to the upstream module instead
11:48 gtristan Because people are generally scared shitless of submodules
11:48 gtristan probably with good reason, I'm not the one with the horror stories
11:49 osse BPL: 1) No it isn't.  2) It may have other unknown changes to its behaviour to make it easier to work with from sourcetree 3) It uses plink for some reason.
11:49 BPL osse: _ikke_ : Ok, now it looks alright (i've added plink to env.path) http://screencast.com/t/cMtp6Ikg , thanks guy, you've been really helpful :)
11:50 tcorneli joined #git
11:50 BPL now buildbot doesn't complain anymore \o/
11:52 heroux joined #git
11:52 DolpheenDream joined #git
11:53 johnmilton joined #git
11:54 M-meznak joined #git
11:54 johnmilton joined #git
11:56 prillian5 joined #git
11:56 prillian5 what is an good git-gui on linux ?
11:56 jknetl joined #git
11:57 dunaeth joined #git
11:57 liuzhen joined #git
11:59 lb1a prillian5 https://git-scm.com/downloads/guis
11:59 prillian5 lb1a: I've seen this. But which would you recommend
12:00 rgrinberg joined #git
12:01 moldy joined #git
12:01 moldy hi
12:01 nothingmuch joined #git
12:02 moldy how do you clean up old merged branches?  ``for b in $(git br --merged|grep -v '^*'); do git br -D $b; done`` is there a better way to do it?
12:02 nothingmuch mid rebase, after a commit has failed to apply cleanly, is the sensible way to defer it till later to --edit-todo, add a pick line, and then --skip?
12:02 red82 joined #git
12:02 nothingmuch moldy: i use git branch -d on all of them
12:03 moldy nothingmuch: manually?
12:03 moldy nothingmuch: that's rather annoying if you have dozends of them :)
12:03 nothingmuch moldy: git branch -d $( git branch -l ) and I've been ignoring the fact that branch '*' doesn't exist for years now =)
12:04 moldy nothingmuch: ah, i see
12:05 moldy nothingmuch: what is the purpose of -l here?
12:05 courrier joined #git
12:05 nothingmuch releasing endorphins for muscle memory i guess =P
12:06 nothingmuch the -d is the rational part =)
12:06 sybariten git bash on Win is slightly frustrating in the sense that if you search online , there seems to be like eight different methods for setting HOME ,...
12:06 moldy hmm, i don't understand what that option does
12:06 fahadash joined #git
12:06 nothingmuch apparently nothing, i don't understand why i type it now
12:07 eihabarabia joined #git
12:08 Ahmed90 joined #git
12:10 swift1_ joined #git
12:11 spNk joined #git
12:14 Jameser joined #git
12:15 mizu_no_oto joined #git
12:16 canuk joined #git
12:17 thebope joined #git
12:18 Rodya_ joined #git
12:20 hallogallo[m] joined #git
12:21 prillian5 how can I link from Readme.md to Changelog.md on gitlab?
12:21 a_thakur joined #git
12:22 ayogi joined #git
12:23 ayogi how can i do a git pull from another directory
12:23 firebalrog joined #git
12:23 ayogi i am writing a bash script which sits in home folder, i want to add a command to do git pull in this script
12:24 ayogi the git repo is at some other location
12:24 osse ayogi: git -C other/location pull
12:25 ayogi git version 1.7.1, throws Unknown option: -C
12:25 M-meznak joined #git
12:25 ayogi osse, ^
12:25 _ikke_ 1.8.5
12:26 osse ayogi: git --git-dir=... --work-tree=... pull
12:26 moldy nothingmuch: ahh, ok. maybe you thought it meant --list :)
12:26 osse ayogi: GIT_DIR=... GIT_WORK_TREE=... git pull
12:26 osse ayogi: (cd ... && git upll)
12:26 ayogi osse, what is the difference between git-dir and work-tree
12:26 DolpheenDream joined #git
12:27 clmsy joined #git
12:27 HardlySe1n joined #git
12:27 osse ayogi: the difference between  other/location/.git  and  other/location
12:27 DarkFox_ joined #git
12:27 jeffreylevesque joined #git
12:27 ayogi osse, so i need to pass both argument or any one of them
12:27 osse both
12:29 ayogi osse, fatal: /usr/libexec/git-core/git-pull cannot be used without a working tree.
12:29 mehola joined #git
12:29 osse you got the paths wrong
12:29 soerenb joined #git
12:30 ayogi git --git-dir=/home/admin/website/.git --work-tree=/home/admin/website/ pull
12:30 t0by joined #git
12:30 ayogi osse, is this correct?
12:30 osse AFAIK yes
12:31 osse you can try without the trailing slash
12:31 osse other than that I don't know why it won't work
12:31 osse since it's a bash script you can do the cd
12:31 osse looks prettier too
12:31 ayogi its throwing fatal error man
12:31 buffal0 joined #git
12:32 ayogi osse, what is the issue
12:32 osse i don't know
12:33 sdothum joined #git
12:34 manosnore joined #git
12:35 ayogi osse, git --git-dir=/home/admin/website/.git --work-tree=/home/admin/website/ pull - works when i execute from the repo folder itself
12:35 manosnore joined #git
12:36 ayogi git --git-dir=/home/admin/website/.git --work-tree=/home/admin/website/ fetch - works instead of pull ?
12:36 manosnore joined #git
12:37 osse pull is a script. fetch is builtin. maybe it's a bug in how those flags are treated then
12:37 osse your git is ancient so it has probably been fixed :p
12:37 osse Just do  (cd /home/admin/website && git pull)
12:38 manosnore joined #git
12:39 manosnore joined #git
12:40 joki- joined #git
12:41 yottabyte joined #git
12:41 yottabyte if I have a dozen .gitignore files in the same repo, how can I consolidate them? do I need to run any command after I move all of the contents into a top level one (fixing the hierarchy in the file paths etc)
12:42 _ikke_ No, git reads .gitignore each time it's run
12:42 tobiasvl nope
12:42 ayogi osse, okay now i am using cd instead and its working
12:42 yottabyte will the gitignore show up as something unstaged?
12:42 tobiasvl yes
12:42 _ikke_ If you track .gitignore (which you should), yes
12:42 yottabyte well I remember before I had to type something because a file was added to gitignore after it was already being tracked
12:43 tobiasvl you probably had to untrack it? gitignore doesn't ignore tracked files
12:44 M-meznak joined #git
12:44 tobiasvl that has nothing to do with this though
12:44 treehug88 joined #git
12:44 Murii joined #git
12:46 notebox joined #git
12:47 osse if  subdir/.gitignore contains '*.exe' you cannot just add 'subdir/*.exe' to the toplevel .gitignore. you need 'subdir/**/*.exe' AFAIK
12:47 osse other than that, go for it
12:48 jknetl joined #git
12:48 chimay joined #git
12:49 kjsage[m] joined #git
12:49 hallogallo[m] joined #git
12:49 matrixa1 joined #git
12:51 mizu_no_oto joined #git
12:53 SwiftMatt joined #git
12:57 qt-x joined #git
12:57 thiago joined #git
12:58 PCatinean joined #git
12:58 MapMan joined #git
12:58 ayogi osse, there is a mistry: git --git-dir=/home/test/page/.git pull works in another machine
12:59 M-meznak joined #git
12:59 Rodya_ joined #git
12:59 tobiasvl what git version is that then
12:59 ayogi same 1.7.1
13:01 ayogi and  git --git-dir=/home/test/page/.git pull for one repo says "error: Your local changes to 'test1.php' would be overwritten by merge."
13:01 zivester joined #git
13:01 fmcgeough joined #git
13:01 ayogi and when i have not change any thing in that file
13:03 geoid joined #git
13:03 tobiasvl surely the file is changed
13:04 fscala joined #git
13:05 DeltaHeavy joined #git
13:06 firebalrog joined #git
13:06 nilg joined #git
13:06 ayogi but i did not change file manually, some how this command is messing with it
13:06 tobiasvl git status
13:07 ayogi i tested 10 times, but same thing again, when i do git pull it works, but when i do "git --git-dir=/home/test/page/.git pull" it says "error: Your local changes to 'test1.php' would be overwritten by merge."
13:07 ayogi git status is clean before pulling
13:07 thiago joined #git
13:07 thiago joined #git
13:08 tobiasvl git --git-dir=/home/test/page/.git status
13:10 nilg` joined #git
13:11 nettoweb joined #git
13:11 ExoUNX joined #git
13:11 absence joined #git
13:11 leehambley joined #git
13:12 absence how can i amend the last commit to update the date to current time?
13:12 dstolfa joined #git
13:12 grawity the author date, or the committer date?
13:12 grawity the former is --date=, the latter is updated automatically on amend
13:13 absence the one you see in git log
13:13 moldy left #git
13:13 absence i've seen the date parameter mentioned, but do i have to specify a date, or can it use current time?
13:14 absence maybe if i just reset and recommit
13:15 firebalrog joined #git
13:15 gtristan joined #git
13:15 bsanford joined #git
13:17 irco joined #git
13:17 Darcidride joined #git
13:18 jknetl joined #git
13:19 canton7 absence, that one's the commit date, I think
13:19 canton7 so just 'git commit --amend'
13:19 canton7 or 'git commit --amend --no-edit'
13:19 cdg joined #git
13:19 Snugglebash joined #git
13:19 goTAN joined #git
13:20 on1x joined #git
13:22 osse --reset-author works
13:22 ayogi tobiasvl, i observed when there is something to pull then only "git --git-dir=/home/test/page/.git pull" throws error
13:23 ayogi and when there is nothing to update it says Already up-to-date.
13:23 osse aha mhm yes yes but what does 'git --git-dir=/home/test/page/.git status' say ?
13:24 ayogi its says # On branch master
13:24 ayogi nothing to commit (working directory clean)
13:24 ayogi should i test again do a dummy commit and try to pull using "git --git-dir=/home/test/page/.git pull"
13:25 ayogi and execute git --git-dir=/home/test/page/.git status before pulling
13:26 GodGinrai joined #git
13:27 serialoverflow joined #git
13:29 DeltaHea_ joined #git
13:29 absence canton7: ok, i already recommitted, but i'll keep that in mind for next time. thanks!
13:32 xbob joined #git
13:32 ayogi osse, man, git --git-dir=/home/test/page/.git status prints lot of things, but when i do git status from inside the directory it says clean
13:32 ayogi osse, are we getting nearer to the problem?
13:34 osse ayogi: afaik when you specify -git-dir you also have to specify --work-tree
13:34 osse so I think you're comparing one repo with the working tree of another
13:35 ams__ joined #git
13:36 mischat joined #git
13:37 ayogi osse, okay i specified --work-tree, then it says clean
13:39 ayogi osse, now when i do git --git-dir=/home/test/page/.git --work-tree==/home/test/page/ pull
13:39 ayogi fatal: /usr/libexec/git-core/git-pull cannot be used without a working tree.
13:40 tobiasvl you have a double =
13:41 sshine left #git
13:41 linuxdaemon joined #git
13:41 ayogi sorry that's a typo here only, so git --git-dir=/home/test/page/.git --work-tree=/home/test/page/ status says clean and git --git-dir=/home/test/page/.git --work-tree=/home/test/page/ pull throws error
13:42 Ryanar joined #git
13:43 shaggycat joined #git
13:44 kjsage[m] joined #git
13:44 JoshL joined #git
13:44 Ryanar so I checked out a branch, made a commit, then checked out a new branch from that branch and made a commit.  The second branch now carries the commit from the first branch, how can I remove that? I meant to checkout the second branch from master
13:44 overlord_tm joined #git
13:45 Noldorin joined #git
13:45 osse ayogi: I though tyou started using cd
13:45 tobiasvl Ryanar: !rebase_interactive can fix that
13:45 tobiasvl Ryanar: !interactive_rebase
13:45 gitinfo Ryanar: Interactive rebase sounds similar to rebase but has completely different abilities. It can do this to commits: change the order, squash some of them together, remove some, add random existing commits from other branches, split them, and more... it's very powerful. Documentation is in the section "Interactive Rebase" in 'man git-rebase'.
13:46 ayogi osse, i stared using, but i want to nail the problem behind this
13:46 tobiasvl Ryanar: or you can just make a new branch (or hard-reset) and cherry-pick the commit you wanted
13:46 buffal0 joined #git
13:46 ayogi osse, fetch works because it will not require working tree
13:46 Zamyatin joined #git
13:47 ayogi what does that actually means, i cloned the repo, so will it not have working tree
13:47 osse ayogi: is it a bare repo?
13:47 osse that explains everything
13:48 ayogi git rev-parse --is-bare-repository - false
13:49 nfk joined #git
13:50 ayogi what is working tree that git is complaining about
13:51 ayogi and only when i use git --git-dir=/home/test/page/.git --work-tree=/home/test/page/ pull command, git pull inside the repo works perfectly fine
13:52 sangy joined #git
13:52 moongazer joined #git
13:52 moongazer How do I undo git rm "filename"
13:52 jozwior_ joined #git
13:53 malcon joined #git
13:53 _ikke_ git checkout filename
13:53 red82 joined #git
13:54 AaronMT joined #git
13:55 moongazer _ikke_, pathspec 'file' did not match any file(s) known to git.
13:55 canton7 moongazer, 'git status' tells you what to type, btw
13:56 osse git checkout HEAD filename
13:56 _ikke_ git checkout HEAD <file>
13:56 _ikke_ slow :P
13:56 ayogi osse, you are working on my problem? or should i leave it unsolved
13:57 osse ayogi: Didn't you give git-dir to rev-parse?
13:57 moongazer how do I FIX the last commit message\
13:57 osse moongazer: commit --amend
13:57 cdown joined #git
13:58 Ryanar thanks the interactive rebase worked, and now I see the difference between rebase and rebase interactive
13:58 ayogi osse, yeah i gave its false
14:00 notebox joined #git
14:00 osse ayogi: I don't know, then
14:01 IrishPrime joined #git
14:01 IrishPrime joined #git
14:02 robotroll joined #git
14:02 ayogi osse, and when i do "git --git-dir=/home/cfadministrator/cf-website/.git/ pull" w/o work-tree it will throw me that commit error file has been updated
14:03 osse ayogi: Is there a git repo in the directory you run that command in?
14:03 crose joined #git
14:04 ocbtec joined #git
14:04 ayogi osse, i am running from home folder(~) there may sub-directory here, i am sure there are many subfolders how can i check
14:04 jbitdrop joined #git
14:04 isysd joined #git
14:05 tjone270_ joined #git
14:05 SoRoot joined #git
14:05 jknetl joined #git
14:06 shinnya joined #git
14:06 osse ayogi: but is there a ~/.git ?
14:07 ^7heo yeah, let's use ~/.git and submodules for each git repo.
14:07 ^7heo that way, everything is always versionned.
14:07 t0by joined #git
14:07 ^7heo and don't forget to put *.mp4, *.mpg, *.wmv, *.ogg, etc. to your ~/.gitignore.
14:07 ayogi osse, there is no .git folder
14:07 ^7heo or you might end up with a *big* repo.
14:08 t0by joined #git
14:08 osse ayogi: !repro
14:08 gitinfo ayogi: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
14:08 osse i am confused
14:09 anonymuse joined #git
14:09 InfoTest joined #git
14:09 osse show the differetn commands,including the one with cd
14:09 Gloomy joined #git
14:10 ResidentBiscuit joined #git
14:10 aviraldg joined #git
14:11 a_thakur joined #git
14:12 bsanford joined #git
14:14 hallogallo[m] joined #git
14:14 TomyLobo2 joined #git
14:15 e14 joined #git
14:16 ayogi osse, https://p.ngx.cc/83
14:16 osse ayogi: git --git-dir=/home/test/page/.git/ --work-tree=/home/test/page/ rev-parse --show-toplevel
14:17 osse what does it print?
14:17 shinnya joined #git
14:17 ayogi /home/test/page
14:18 duarteper joined #git
14:19 Lunatrius joined #git
14:19 marc_v92 joined #git
14:19 osse ayogi: and the same command works on a differetn computer with the same git version ?
14:19 osse rerun the pull but with GIT_TRACE=1 in front of the whole command
14:20 thebope joined #git
14:20 TomyLobo2 joined #git
14:22 _ix joined #git
14:22 ayogi i ran: git --git-dir=/home/test/page/.git/ --work-tree=/home/test/page/ pull GIT_TRACE=1
14:22 ayogi fatal: /usr/libexec/git-core/git-pull cannot be used without a working tree.
14:22 yottabyte what does "**.tmp/" mean in gitignore? .tmp is a directory
14:23 osse ayogi: in front
14:23 osse not at the back
14:23 NeverDie joined #git
14:23 ayogi trace: exec: 'git-pull'
14:23 ayogi trace: run_command: 'git-pull'
14:23 ayogi trace: built-in: git 'rev-parse' '--git-dir'
14:23 ayogi fatal: /usr/libexec/git-core/git-pull cannot be used without a working tree.
14:25 osse it's a bug
14:25 ayogi osse, we an remove another computer case from the problem, there also the same error now, but lets focus on this computer only
14:26 osse it's a bug in Git 1.7.1. The flags aren't properly communicated down to git pull
14:26 shinnya joined #git
14:26 thiago joined #git
14:26 osse try stash or rebase instead of pull. i expect similar errors
14:27 ayogi but that is what comes by default from centos repo
14:28 patarr joined #git
14:28 ayogi osse, yeah same error with stash and rebase
14:28 ayogi osse, will fetch work instead of pull
14:28 osse yes
14:29 njbair joined #git
14:29 HoierM joined #git
14:29 ayogi osse, are you sure its a bug and not a problem at my end
14:29 patarr joined #git
14:30 ayogi osse, ubuntu comes with 1.9.1
14:30 ehynds joined #git
14:30 ayogi and centos-6 with git 1.7.1
14:30 ayjay_t joined #git
14:31 ayogi osse, shall i upgrade git and try same command
14:31 osse ayogi: no i am not sure
14:31 osse maybe 85%
14:32 ayogi osse, so i can try updating git and see if it works?
14:32 osse sure
14:32 shinnya joined #git
14:32 ayogi osse, its a prod machine, i am anxious
14:32 quenode_ left #git
14:33 osse ayogi: then use cd and leave it be
14:33 osse on the other hand, you should definitely upgrade since security holes have been fixed
14:34 littleforest joined #git
14:34 ayogi osse, upgrading git will not break any other things right? do you have documentation to do an upgrade in centos
14:34 msonntag joined #git
14:35 dviola joined #git
14:35 osse ayogi: you can install the new git in a special location and only use that in your scripts
14:35 osse ayogi: no I don't. the INSTALL in the repo is good
14:35 kjsage[m] joined #git
14:36 patarr joined #git
14:36 thiago joined #git
14:36 hhee joined #git
14:36 ayogi osse, INSTALL in the repo ?? i did not get that
14:37 [sr] left #git
14:37 jknetl_ joined #git
14:38 dwts joined #git
14:38 osse ayogi: https://github.com/git/git/blob/master/INSTALL
14:39 yottabyte what does "**.tmp/" mean in gitignore? .tmp is a directory
14:39 _ikke_ .tmp in any directory
14:39 osse yottabyte: looks like a typo to me. should probably be **/.tmp
14:39 Filiphe joined #git
14:39 hhee guys, can i revert one file from commit? my commit contains several files, but i need only one to be reverted.
14:40 yottabyte what's **/.tmp?
14:40 ayogi osse, i executed the same command in ubuntu-14.04 with git version 1.9.1 and it is working
14:40 hhee (just begun worked with git)
14:40 emg joined #git
14:40 ayogi osse, that proves 1.7.1 has bug, can we claim that now?
14:41 hariel joined #git
14:41 osse ayogi: yes
14:41 osse yottabyte: **/ means * or */* or */*/* or */*/*/* or ...  so it matches .tmp in any subdir
14:42 ayogi osse, okay man thanks a lot, i learnt something today
14:42 bremner ayogi: in case no one mentioned it, 1.7.1 is from 2010
14:43 osse ayogi: if you really want to have fun you can compile git yourself and use git bisect to track down the commit that fixes the rror
14:43 _dunaeth joined #git
14:43 mischat joined #git
14:45 Vgr joined #git
14:45 Vgr joined #git
14:45 faux___ joined #git
14:47 qpdb joined #git
14:47 ayogi should i file a bug regarding this in github
14:48 osse No, for two reasons: 1) It's already been fixed. 2) Git doesn't use github for issue tracking (or anything other than mirroring)
14:48 bremner ayogi: you could complain to your distro
14:48 Ahmed90 joined #git
14:49 git345 joined #git
14:49 TomyLobo2 joined #git
14:49 ayogi mine is centos-6.8 so where should i log the complain
14:49 bremner with centos
14:50 git345 left #git
14:50 git339 joined #git
14:50 git339 left #git
14:50 ayogi bremner, i have never done that, can you tell me the endpoint where i can post this,
14:50 kjsage[m] joined #git
14:51 bremner no, I cannot help with centos support.
14:51 eivarv joined #git
14:52 GodGinrai ayogi: http://lmgtfy.com/?q=centos+report+bug&amp;l=1
14:54 osse do you want to file a distro bug for an old distro version because one of the pieces of software in its repos has a bug?
14:55 thebope_ joined #git
14:55 supsup joined #git
14:55 eycsigfy joined #git
14:55 bremner well. I mainly want to stop talking about it here tbh ;)
14:56 GodGinrai hehe
14:56 GodGinrai he pinged out
14:56 GodGinrai guess that's one way for the discussion to end
14:56 supsup hi any idea how after i do a pull --rebase on a repo i get modified files? (repo was clean before command)
14:56 _ikke_ supsup: Because you get a conflict?
14:56 buffal0 joined #git
14:56 supsup it did not say there was a conflict
14:56 _ikke_ supsup: I cannot assume that
14:57 supsup i'll paste results in pastebin
14:57 _ikke_ Does the pulled changes include changes in .gitattribute files?
14:58 supsup no
14:58 supsup they are .html and .md files
14:58 supsup http://pastebin.com/2qLC3m99
14:59 thebope joined #git
15:00 snowkidind joined #git
15:00 supsup any idea?
15:00 supsup it's so weird
15:00 _ikke_ line ending normalization I suspect
15:00 Foryah92 joined #git
15:00 supsup how do i solve that?
15:01 Foryah92 Question : What happens if I merge two branches which have the same changes in different commits ?
15:01 _ikke_ What does git diff show?
15:01 Foryah92 ikke, you're talking with me ? :)
15:02 canton7 Foryah92, I answered that question in #github - I'm assuming that was you as well?
15:02 _ikke_ Foryah92: No, was talking to supsup
15:02 supsup ikke what should I look for in the log file?
15:02 mischat joined #git
15:02 GodGinrai Foryah92: if it is the exact same change, then nothing should happen.
15:02 Foryah92 Ok, so, all the changes should be there, and which commit will be in the tree ? Both of them ?
15:03 corpsicle joined #git
15:03 GodGinrai Foryah92: both
15:03 canton7 Foryah92, 'merge' does not remove any commits
15:03 canton7 so both commits are still there
15:03 thebope_ joined #git
15:03 GodGinrai Foryah92: merge just creates a new commit that is an endpoint for multiple branches.  The commits in those branches are not changed.
15:03 star_prone joined #git
15:04 corpsicle should i force push ?
15:05 fedepad joined #git
15:06 GodGinrai corpsicle: that's almost never the right answer
15:06 GodGinrai corpsicle: what lead to this?
15:06 Foryah92 GodGinari: Ok, so, if I have some changes in production and some in staging, but they have been pushed in different commits ( copy - pasted and comited ), when I merge staging in production, it should be fine, right ?
15:06 corpsicle yea so i gather
15:06 Foryah92 GodGinrai: Ok, so, if I have some changes in production and some in staging, but they have been pushed in different commits ( copy - pasted and comited ), when I merge staging in production, it should be fine, right ?
15:06 _ikke_ You can if you know what you are doing, but the question is why
15:06 supsup i tried to do: git config --global core.autocrlf true  but and go back 3 commits but i get the same error
15:06 fedepad_ joined #git
15:07 corpsicle GodGinrai: well, we have a "release" branch, that sort of turned into a mess. so basicly i want to overwrite it with a developement branch which ive verified contains all it should. to prepare for the next release
15:07 corpsicle i mean, a release-branch should be temporary right?
15:07 GodGinrai Foryah92: probably.  But I would merge in a separate branch, test that it works, and then merge into production once you confirm it works.  That way, production just sees a fast-forward merge.
15:08 LionsMane joined #git
15:08 corpsicle or should i just create a new branch from the developement branch
15:08 corpsicle and throw the old release branch away
15:08 Foryah92 GodGinrai: Ok, thanks ! :)
15:08 Darcidride joined #git
15:08 Foryah92 left #git
15:08 goTAN joined #git
15:09 thebope joined #git
15:09 tyreld joined #git
15:09 TalkToMe joined #git
15:10 GodGinrai corpsicle: I would throw the old branch away in that case.  Better to just "start anew" than to force push, imo
15:10 fedepad joined #git
15:11 TalkToMe joined #git
15:11 _ikke_ GodGinrai: There is no different between throwing it away and pushing a new branch, than to force push
15:11 jimi_ joined #git
15:11 _ikke_ Both will result in the same effect when other people pull that branch
15:12 zivester joined #git
15:14 eb_ joined #git
15:14 SporkWitch joined #git
15:15 Zamyatin joined #git
15:16 raspado joined #git
15:18 mustmodify joined #git
15:19 KevinMGranger_ left #git
15:21 GodGinrai _ikke_: yes, but deleting and creating makes more sense from a process standpoint
15:21 corpsicle GodGinrai: alright, thanks =)
15:21 GodGinrai np
15:22 osse do you then end up with a branch name that is weird because you cannot call it the same thing because that makes less sense from a from a process standpoint ?
15:22 lindenle joined #git
15:22 GodGinrai osse: how can you not call it the same thing?
15:22 _ikke_ GodGinrai: But like you say it, it looks like if force push is bad and should be avoided, while deleting the branch and pushing a new branch is ok
15:23 _ikke_ while the same caveats apply
15:23 kulelu88 joined #git
15:24 GodGinrai although honestly, release branches should never be so dirty that force pushing or deleting is necessary
15:24 osse GodGinrai: if the first branch name was clean-up-temp-files the new one could be remove-temp-files
15:24 osse oh release branches
15:24 GodGinrai yea
15:24 _ikke_ GodGinrai: That I agree with
15:25 freimatz joined #git
15:27 eycsigfy joined #git
15:30 Borkr joined #git
15:30 jnavila joined #git
15:31 thiago joined #git
15:31 mehola joined #git
15:33 nowhereman joined #git
15:35 Salander27 joined #git
15:37 waveclaw joined #git
15:37 Remramm joined #git
15:38 phaleth joined #git
15:39 afuentes joined #git
15:40 nonconvergent joined #git
15:40 thiago joined #git
15:41 nonconvergent Hey, if I reset hard then push, will that clean up upstream?
15:42 ResidentBiscuit joined #git
15:43 _ikke_ nonconvergent: You'd have to force-push
15:43 hallogallo[m] joined #git
15:43 _ikke_ !rewrite applies
15:43 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
15:43 kikijiki[m] joined #git
15:43 M-meznak joined #git
15:43 M-shine joined #git
15:43 Ineentho[m] joined #git
15:43 M-mistake joined #git
15:43 aviraldg joined #git
15:43 kjsage[m] joined #git
15:43 ResidentBiscuit joined #git
15:44 a_thakur joined #git
15:44 jknetl_ joined #git
15:44 svankmajer joined #git
15:45 durham joined #git
15:47 durham joined #git
15:47 monoxane joined #git
15:50 al-damiri joined #git
15:54 lungaro joined #git
15:54 lungaro is it possible to checkout just specific directories (and children) when cloning?
15:54 lungaro trying to speed up checkouts
15:55 netj joined #git
15:55 adamretter joined #git
15:56 Ryanar joined #git
15:58 patarr joined #git
15:58 Electrometro joined #git
15:58 nonconvergent _ikke_: So if I reset hard, then pull
15:59 nonconvergent it'll go back to the HEAD from upstream?
16:00 eycsigfy joined #git
16:03 xall joined #git
16:04 finalbeta joined #git
16:04 _ikke_ !HEAD
16:04 gitinfo HEAD is a 'pointer' to the currently checked out branch (or commit, if HEAD is !detached). In bare repositories it tells clients which branch to checkout initially after cloning. Unlike commonly believed, HEAD is *not* something that exists separately for every branch. It also is *not* necessarily the newest commit in the repo (that's hard to define in a DVCS, anyway...)
16:04 synthroid joined #git
16:04 fedepad joined #git
16:04 nonconvergent Okay, so what I really want to do is just revert the last 9 commits?
16:05 nonconvergent In LIFO?
16:05 _ikke_ Did you push those 9 commits?
16:05 nonconvergent Someone else did, I pulled them, they broke things. Some of them are reverts and reverts of reverts themselves.
16:06 nonconvergent It's a weird problem and I can't debug and fix it with good code. I need to take it back to a working commit
16:06 _ikke_ Ok
16:06 _ikke_ But to clean this up, you need to coordinate this with the others working on this repository\
16:06 theshitisay joined #git
16:07 nonconvergent _ikke_: What do you mean?
16:07 nonconvergent Just reset head, force push and then tell them to...?
16:07 _ikke_ nonconvergent: I think the easiest is to tell them to run git pull --rebase
16:08 _ikke_ This should in most cases rebase their local commits on top of the new history
16:08 EvilPeng1 joined #git
16:08 _ikke_ But you should verify if the result is correct
16:08 nonconvergent Okay.
16:09 nonconvergent That's the hard part. It's an EMR project. So i'll make a new branch, reset that, push it to our build server and test it in AWS, then merge into master.
16:10 fub left #git
16:10 batrick joined #git
16:11 e14 joined #git
16:12 madewokherd joined #git
16:12 PCatinean joined #git
16:12 armin joined #git
16:12 LeRieur joined #git
16:13 thebope_ joined #git
16:14 ochorocho__ joined #git
16:14 mustmodify left #git
16:14 armin ohai. i have created a custom .git/hooks/post-receive script where i simply echo "$@" and throw all output (stdout and stderr) into some logging file. this script gets run, but it seems like it doesn't get called with any additional arguments, at all. does anyone have a clue why so? i looked at the post-receive-email example script in the examples directory of the git distribution and that one reads $1 - $3 so i
16:14 armin assume my attempt wasn't totally wrong. any hint?
16:14 _ikke_ If you just reset the branch, then there is nothing to merge into master
16:15 _ikke_ What are you pushing?
16:15 thebope__ joined #git
16:17 nettoweb joined #git
16:17 User458764 joined #git
16:17 WayToDoor joined #git
16:17 armin _ikke_: from your questions and the fact i just joined seconds before i assume you don't mean me?
16:18 Oebele joined #git
16:18 thebope joined #git
16:18 anonymuse joined #git
16:19 LeRieur joined #git
16:19 _ikke_ armin: Sorry, the last one was
16:19 canton7 armin, "It takes no arguments, but gets the same information as the pre-receive hook does on its standard input"
16:19 canton7 (from man githooks)
16:19 gitinfo the githooks manpage is available at http://jk.gs/githooks.html
16:20 canton7 armin, there is no sample post-receive hook
16:21 armin canton7: i have /usr/share/git-core/contrib/hooks/post-receive-email on my system
16:21 canton7 ah, that one, not the sample in .git/hooks
16:21 armin canton7: that's why i wrote "post-receive-email example script in the examples directory of the git distribution"
16:22 canton7 it's almost home time and I've been reading convoluted specs for the past few hours ><
16:22 armin canton7: no worries. :-)
16:22 ResidentBiscuit joined #git
16:23 Silenced_v2 joined #git
16:23 Silenced_v2 joined #git
16:24 canton7 armin, https://github.com/git/git/blob/master/contrib/hooks/post-receive-email#L746: it allows reading from arguments for testing, but reads from stdin in real usage
16:25 LeRieur joined #git
16:25 Electrometro joined #git
16:26 metrixx_ joined #git
16:27 overlord_tm joined #git
16:27 xall joined #git
16:27 crayon joined #git
16:28 anuxivm joined #git
16:29 nitric joined #git
16:29 Darren_ joined #git
16:29 canton7 armin, make sense?
16:30 kpease_ joined #git
16:31 oskarkv joined #git
16:31 nettoweb1 joined #git
16:33 armin canton7: aaaaaah! fair enough! thanks!
16:33 armin canton7: you deserve your home time now.
16:35 canton7 armin, heh, planning on it! :D
16:35 mustmodify joined #git
16:35 mustmodify left #git
16:36 svankmajer joined #git
16:37 drmt joined #git
16:38 freimatz joined #git
16:38 svm_invictvs joined #git
16:39 armin canton7: well, me too, it's 18:38 here already and i'm still in the office. i should go to the pub. now.
16:39 canton7 go!
16:41 smw_ joined #git
16:41 Samual joined #git
16:42 ^7heo same time here
16:42 ^7heo and I'm trying to setup debian
16:42 manosnore joined #git
16:42 canton7 I was in that timezone this morning :P
16:42 GingerGeek joined #git
16:42 GingerGeek joined #git
16:42 ^7heo but the packages are too old
16:42 ^7heo it fails =/
16:42 ^7heo debian, how to fail before even starting.
16:42 duarteper joined #git
16:43 ilmari ^7heo: the git packages? the ones from sid install fine on jessie
16:44 smw_ Hi all, how can I create a shallow clone with submodules at a specific commit? So far I have: git clone --recursive --depth=1 --branch=smw MYREPO. The issue is that 1. I want to give a commit, not branch and 2. the submodule isn't shallow.
16:44 ^7heo ilmari: bind9
16:45 smw_ This is for a deployment system. I am trying to reduce the amount downloaded for each deploy.
16:45 bket joined #git
16:46 patarr joined #git
16:47 nikivi joined #git
16:48 yohnnyjoe joined #git
16:48 svankmajer joined #git
16:49 kjsage[m] joined #git
16:49 milki_ left #git
16:50 yqt joined #git
16:50 milki joined #git
16:50 leeN joined #git
16:50 eshlox joined #git
16:50 lulu joined #git
16:52 cjohnson smw_: I would advise not using git then if that's your goal
16:52 nitric joined #git
16:52 cjohnson not using git as your deployment tool that is
16:52 cjohnson I would tar up a build/release and scp it onto the machine. or use rsync with compression
16:52 smw_ cjohnson: it isn't as a deployment tool. I am setting up a build pipeline where the builder fetches directly from source.
16:53 cjohnson Right and if you're trying to minimize how much goes from source -> build, you would go to the source and tar up there
16:53 cjohnson instead of just trying to use git to get the stuff onto the machine
16:53 cjohnson if you care about reducing that size, that is
16:54 dwmw2_gone joined #git
16:54 Dazzyp joined #git
16:54 smw_ cjohnson: the source is a git repo... there is no other way to get it so I can tar it up. I suppose I could keep a server around with a local copy... but that seems silly. For now I guess size isn't too big of a deal :-\
16:55 cjohnson yeah that would be my second suggestion, an intermediate build server which speeds things up
16:55 cjohnson then to update all you do is run git pull on that build server and make a new tarball
16:56 smw_ this is making me sad. I feel like this should be much easier. Anyways, the optimization isn't worth it right now. In the future it might be.
16:56 * smw_ just wants reproducible builds/deployments
16:57 goTAN joined #git
16:58 smw_ cjohnson: build/deploy currently happens from people's local machines. You can guess how well that works :-).
16:59 P4nther joined #git
16:59 ViktorRu joined #git
16:59 eivarv joined #git
16:59 ahmedelgabri joined #git
16:59 robotroll joined #git
17:00 joki- joined #git
17:01 LBV_User joined #git
17:02 EvilPenguin joined #git
17:03 anuxivm joined #git
17:04 iamtony joined #git
17:04 cjohnson smw_: ah yep that's generally not a great approach imo. I've done it but utlimately it's finnicky and rather insecure
17:04 cjohnson and flaky
17:05 cjohnson your local wifi craps out mid-upgrade and now you might have a bunk system depending on whether you can perform an entire build across all components in a "transaction"
17:06 cjohnson spend some time setting up a build server and centralize your build process and it will also give you free lunch regarding build sizes being too big/slow
17:06 smw_ cjohnson: the biggest issue is that this is sort of a pet project handled by a bunch of different people. The end result is that no one knows what is deployed and when it goes down you just need to hope HEAD is ok so you can make a change.
17:06 Samual joined #git
17:06 cjohnson YUP exactly. then there's also the issue of what happens if two people deploy at once
17:06 CheckDavid joined #git
17:06 geoid joined #git
17:06 cjohnson with no centralization ther'es no way to guarantee it's locked out
17:06 smw_ cjohnson: we use App Engine Flexible so we are protected from a lot of the stuff you are talking about.
17:07 smw_ cjohnson: the biggest problem is reproducibility and understanding what is in production. GAE handles all the stuff you are talking about.
17:08 Limes_ joined #git
17:08 cjohnson never used, but maybe you could implement this build process within that? I'm assuming this is some sort of centralized server/service
17:09 anuxivm joined #git
17:10 nettoweb joined #git
17:10 thebope_ joined #git
17:11 smw_ cjohnson: yeah, I have an abnormal setup so have to wire things together myself. I am using Google Container Builder for most of this. It is the Container Builder service that is doing the checkout.
17:12 SteffanW joined #git
17:13 cyan__ joined #git
17:13 mehola joined #git
17:16 eshlox joined #git
17:17 fscala joined #git
17:17 InfoTest joined #git
17:22 anuxivm joined #git
17:23 Raging_Hog joined #git
17:23 thebope joined #git
17:23 huhlig joined #git
17:23 roentgen joined #git
17:23 dwigton joined #git
17:25 sharp555 joined #git
17:25 thebope_ joined #git
17:26 sharp555 I'm trying to do a "3-way merge" but the automatic part of the merge isn't working properly (it's not asking me to resolve all of the problems). What is the best way to deal with this?
17:26 xbob joined #git
17:27 Awesomecase joined #git
17:28 dwigton I just realized that I had been coding a bunch of stuff in an un committed branch. So I added the things I needed to commit, committed, switched to master and merged then switched back to work on my uncommitted stuff. I then realized to my horror I had absent mindedly commit -a. So all my unstable stuff got merged. Not a big deal but is there a way to walk that back and do it correctly?
17:28 e14 joined #git
17:29 lus joined #git
17:30 mikecmpbll joined #git
17:31 tycoon177 joined #git
17:32 hashpuppy joined #git
17:33 thebope joined #git
17:33 kadoban joined #git
17:33 Mixologic_ joined #git
17:34 ahmedelgabri joined #git
17:34 canton7 dwigton, !fixup
17:34 gitinfo dwigton: So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!
17:34 xall joined #git
17:35 laumann joined #git
17:35 Orion3k joined #git
17:36 jalt joined #git
17:36 patarr joined #git
17:37 barajasfab joined #git
17:37 duarteper joined #git
17:37 cdg joined #git
17:37 zidane_g joined #git
17:40 mustmodify joined #git
17:40 mustmodify crud.
17:40 rahtgaz joined #git
17:40 mustmodify A few days ago I was saying I wanted a command to absolutely reset files to a tag so I could do a live coding demo and have an easy rescue.
17:41 goTAN joined #git
17:41 alvarezp left #git
17:41 mustmodify Now I realized I needed to tag step0 before my step1.
17:41 mustmodify Is there a way to do that? I'm thinking no...
17:42 zidane_g joined #git
17:42 mustmodify Meh, forget it. Not worth the trouble. I'll just wave a magic wand and be done with it.
17:42 mustmodify left #git
17:43 anuxivm joined #git
17:45 thiago git checkout <commit> -- <filename>
17:45 anuxivm joined #git
17:49 anuxivm joined #git
17:50 vrishab joined #git
17:51 P4nther joined #git
17:51 duarteper joined #git
17:51 PCatinean joined #git
17:51 Kitty- joined #git
17:53 hanthings_ joined #git
17:54 Raging_Hog joined #git
17:55 laumann joined #git
17:55 nonconvergent joined #git
17:57 anuxivm joined #git
17:58 Pursuit joined #git
17:59 anuxivm joined #git
17:59 thebope joined #git
17:59 joe543456 joined #git
18:00 Rodya_ joined #git
18:00 seventy joined #git
18:00 thebloggu joined #git
18:00 piling joined #git
18:00 EvilPeng1 joined #git
18:01 ahmedelgabri joined #git
18:03 dsantiago joined #git
18:04 Raging_Hog joined #git
18:04 Ryanar joined #git
18:07 moongazer left #git
18:10 atomi joined #git
18:10 dreiss joined #git
18:11 Ryanar joined #git
18:11 jost joined #git
18:12 roentgen joined #git
18:13 joe543456 Hi, I was wondering if I could ask a question about git-svn?
18:13 bremner yes
18:13 mehola joined #git
18:14 patarr joined #git
18:14 GodGinrai !ask
18:14 gitinfo Yes, it's okay to ask questions here.... in fact, you just asked one! ;-) Pretty much any question is fine. We're not terribly picky, but we might be asleep. Please be patient and you should get an answer soon.
18:15 nandi_ge___ joined #git
18:15 Faylite joined #git
18:16 thebope_ joined #git
18:17 dan2k3k4 joined #git
18:19 Remramm joined #git
18:19 Remramm joined #git
18:19 jalt Hi, if I clone an existing project (and thus I track it by default), and then manually add another remote, what do I need to do so that my default workflow becomes: pull from both origin (original source of the clone) and alt (new remote), and push only to alt? Basically I want to track upstream changes in origin and regularly merge them to my local repo, and regularly push the unchanged "upstream" branch to my alt, along with any other branches I might be working a
18:20 jalt I know how to do it manually, I just need to prevent accidental mistakes, like pushing to origin (where I don't even have commit rights).
18:20 joki- joined #git
18:20 jaafar joined #git
18:21 joe543456 I have an svn repo that I am moving to git. I was able to run "git svn clone my-svn-url" and "git remote add new-git-url". I make changes to the new git master and I can svn dcommit just fine. However, as soon as I run "git pull origin master" I haven't been able to run and git svn commands (such as git svn rebase, git svn fetch) with the error message "Unable to determine upstream SVN information from HEAD history." From the
18:21 joe543456 blogs I read, this has something to do with disjoint histories. The solutions I've come across at http://brandon.dimcheff.com/2009/01/04/commit-a-linear-git-history-to-subversion.html did not work for me, nor have any other answers I've come across. Can someone help me debug this?
18:24 Ryanar joined #git
18:25 sea-gull_ joined #git
18:25 cagedwisdom joined #git
18:26 Derperperd joined #git
18:26 Jackneill joined #git
18:28 Balliad joined #git
18:28 ViktorRu joined #git
18:29 EmeraldExplorer joined #git
18:30 anuxivm joined #git
18:30 red82 joined #git
18:31 cdown joined #git
18:31 eivarv joined #git
18:34 cdown_ joined #git
18:37 m0viefreak joined #git
18:37 rgrinberg joined #git
18:37 Ryanar joined #git
18:40 joki- joined #git
18:42 Sasazuka joined #git
18:44 PocketKiller-rip left #git
18:46 Remramm Yeah by pulling you created a merge, and non-linear histories can't be put into SVN
18:48 Remramm stop using pull or configure it to merge (pull.rebase)
18:48 Remramm To fix your current predicament, you'll have to run git rebase
18:49 vuoto joined #git
18:49 KnownSyntax joined #git
18:49 KnownSyntax joined #git
18:49 EvilPeng1 joined #git
18:49 _dunaeth joined #git
18:50 Doginal joined #git
18:50 joe543456 Thanks Remramm, I'll try that out
18:51 shaggycat joined #git
18:54 dodobrain joined #git
18:54 lungaro is it possible when cloning to checkout just specific directory hierarchies?
18:54 bremner no
18:54 bremner !narrow
18:54 lungaro :(
18:54 lungaro booo
18:54 bremner after you clone you can checkout part of the work tree
18:54 bremner look for "narrow checkout"
18:55 lungaro gotcha. But the clone is what takes so long
18:55 bremner ack
18:55 lungaro i'll check it out though. something may work perhaps w/ a shallow clone
18:55 lungaro and narrow checkout
18:56 ocbtec joined #git
18:58 littleforest joined #git
19:00 iota_pi joined #git
19:00 svm_invictvs joined #git
19:02 Lor_ joined #git
19:02 jeffreylevesque joined #git
19:04 _ikke_ sparse checkout
19:04 dreiss joined #git
19:07 LionsMane joined #git
19:07 star_prone joined #git
19:07 ResidentBiscuit joined #git
19:08 abyx joined #git
19:08 abyx why would `git push origin HEAD` push changes on ALL branches and not just the currently checked-out branch?
19:08 leeN joined #git
19:10 hariel joined #git
19:10 joki- joined #git
19:10 avar do "man git-config" and search for push.default
19:10 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
19:11 hariel joined #git
19:12 nwkj86 joined #git
19:14 hariel joined #git
19:14 _ikke_ avar: push.default should only take effect when not specifying branches
19:15 drmt joined #git
19:16 dodobrain joined #git
19:21 dreiss joined #git
19:22 mikecmpbll joined #git
19:24 vuoto joined #git
19:29 GavinMagnus left #git
19:29 SwiftMatt joined #git
19:33 ikarius joined #git
19:33 rgrinberg joined #git
19:34 ikarius hello,  I’m having trouble pulling a specific directory from a different branch in my repository
19:34 yqt joined #git
19:34 WayToDoor joined #git
19:34 ikarius the google results I’ve found indicate this should be doable with “git checkout other-branch somedirectory/“
19:35 _ikke_ yes
19:35 AnarchyAo joined #git
19:35 InfoTest joined #git
19:35 ikarius but, when I do that, I get messages “error: pathspec ‘other-branch’ did not match any file(s) known to git.
19:35 ikarius and the same for somedirectory/
19:36 ikarius I’m on git version 1.7.10.4, if that affects it
19:36 _ikke_ That's quite old
19:36 _ikke_ It should not affect how the command works though
19:36 ikarius yeah, though I’m seeing google results indicating this solution should work dating 2009ish?
19:37 buffal0 joined #git
19:37 _ikke_ It should work even with older versions of git
19:37 ikarius essentially, looks like git assumes both other-branch and somedirectory/ are file paths
19:37 finalbeta joined #git
19:37 _ikke_ ikarius: try: git checkout other-branch -- somedirectory
19:38 _ikke_ note that you have to pass the full path relative to the root of the repository
19:38 ikarius fatal: invalid reference: other-branch
19:38 _ikke_ It does not know other-branch then
19:38 mikecmpbll joined #git
19:38 _ikke_ are you sure that branch exists?
19:39 ikarius yep, git branch -a | grep other-branch shows me remotes/origin/other-branch
19:39 _ikke_ That's not other-branch
19:39 _ikke_ thats origin/other-branch
19:41 ilbot3 joined #git
19:41 Topic for #git is now Welcome to #git, the place for git help and electronically-induced comas | 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: 2.10.1 | Getting "cannot send to channel"? /msg gitinfo .voice | git-daemon: the evil rises
19:41 ikarius I was loosing my mind trying to figure out why git wasn’t doing what all the solutions indicated it should be doing
19:42 ikarius it never occurred to me that it was breaking because I pulled a clean git clone
19:42 anonymuse joined #git
19:42 patarr joined #git
19:42 ikarius just to confirm....
19:43 ikarius this mechanism will preserve the history of the files pulled over to the new branch, correct?
19:43 Discovery joined #git
19:43 ikarius as opposed to simply copying the directories from the other branch in and using “git add”
19:43 TomyLobo joined #git
19:44 _ikke_ It's just a local file operation
19:44 anli Can I move a submodule directory?
19:44 ikarius hmm
19:44 _ikke_ !submodule_mv
19:44 gitinfo [!submodules_mv] Renaming git submodules is harder than it perhaps should be.  All commands are in the superproject.  Edit .gitmodules to reflect the new path.  Then `git mv oldsubmodulepath newsubmodulepath; git commit -m "Moved submodule from old to new"`  Inspect .git/config to see if anything looks likely for changing.
19:45 ikarius what I’m trying to do is preserve the change history of the files as I migrate them from one branch to another
19:45 IlNominePatre joined #git
19:45 ikarius unfortunately, I cant do a git merge as the branch where this work was done was pulled from… a branch I don’t want to merge.
19:45 anli Maybe I should remove the submodule and clone it again
19:45 cagedwisdom joined #git
19:46 sbeller git mv *ought* to work just fine for moving submodules nowadays (I remember fixing a bug in that)
19:46 ikarius but in “somebranch”, git log mydirectory/pom.xml shows me a list of changes, in new branch, if I do “git log mydirectory/pom.xml”… nothing
19:46 _ikke_ sbeller: That trigger is indeed a bit old
19:46 ikarius this is after the git checkout operation.
19:47 ikarius … is there a better way to accomplish what I’m trying to do?
19:47 KnownSyntax joined #git
19:47 KnownSyntax joined #git
19:47 rancyd joined #git
19:47 mischat joined #git
19:47 _ikke_ ikarius: Perhaps cherry-picking is better
19:48 Faylite_ joined #git
19:48 ikarius _ikke_: ahh, right, that might be the ticket.
19:48 marlun1 left #git
19:48 patarr joined #git
19:58 Ryanar joined #git
19:58 ikarius thanks again for the help!
19:59 anonymuse joined #git
20:00 i7c joined #git
20:02 dan2k3k4 joined #git
20:02 bejeb joined #git
20:04 crayon joined #git
20:04 cdg joined #git
20:04 7F1AAEIMN joined #git
20:08 duarteper joined #git
20:08 Rodya_ joined #git
20:09 ahr3n joined #git
20:10 joki- joined #git
20:11 hariel joined #git
20:12 buffal0 joined #git
20:14 crayon joined #git
20:14 crayon joined #git
20:14 mehola joined #git
20:15 hariel joined #git
20:18 xall joined #git
20:24 drmt joined #git
20:26 patarr joined #git
20:26 ieee joined #git
20:27 reizuki joined #git
20:27 ResidentBiscuit joined #git
20:28 crose joined #git
20:31 jimi_ joined #git
20:31 fatalhalt joined #git
20:33 rgrinberg joined #git
20:34 boombatower joined #git
20:34 wriggle left #git
20:36 Snugglebash joined #git
20:39 hariel joined #git
20:40 Xethron joined #git
20:40 joki- joined #git
20:41 Electrometro joined #git
20:42 Balliad joined #git
20:42 re1 joined #git
20:44 DarkFox_ joined #git
20:44 hariel joined #git
20:45 Zamyatin joined #git
20:45 ViktorRu joined #git
20:46 Guest55 joined #git
20:47 GavinMagnus joined #git
20:48 hariel joined #git
20:49 jalt left #git
20:51 esc joined #git
20:51 ntnn joined #git
20:51 spaceone joined #git
20:53 lamont joined #git
20:55 rgrinberg joined #git
20:56 star_prone joined #git
20:59 LBV_User joined #git
21:00 spNk joined #git
21:02 shaggycat joined #git
21:03 EvilPenguin joined #git
21:05 Noldorin joined #git
21:08 nilg joined #git
21:08 rgrinberg joined #git
21:08 HardlySe2n joined #git
21:09 Scott_ joined #git
21:09 _ADN_ joined #git
21:10 Electrom_ joined #git
21:10 joki joined #git
21:10 ieee1 joined #git
21:10 metrixx_ joined #git
21:11 bjoe2k4_ joined #git
21:11 Lunatrius` joined #git
21:11 sorenb joined #git
21:11 perlpilot_ joined #git
21:12 RickyB98|BNC joined #git
21:13 trelane joined #git
21:13 trelane joined #git
21:13 deltab__ joined #git
21:13 salamanderrake joined #git
21:13 dunk_ joined #git
21:13 BinGOs_ joined #git
21:14 wwalker joined #git
21:14 relgar joined #git
21:14 pikajude_ joined #git
21:14 nuck joined #git
21:14 ahf_ joined #git
21:14 zidane_g_ joined #git
21:14 hashpuppy_ joined #git
21:14 matsaman joined #git
21:14 laumann joined #git
21:15 tjbp_ joined #git
21:15 Sasazuka_ joined #git
21:15 BinGOs_ left #git
21:15 grift joined #git
21:15 MrC joined #git
21:15 sjm joined #git
21:16 polyzen joined #git
21:16 Snugglebash joined #git
21:16 aca_ joined #git
21:16 jargan joined #git
21:16 kpease joined #git
21:16 cbreak joined #git
21:16 nikow joined #git
21:17 Snugglebash joined #git
21:17 ghostlight_ joined #git
21:17 cschneid joined #git
21:17 Jonuz joined #git
21:17 Snugglebash joined #git
21:18 CussBot_ joined #git
21:18 yanome joined #git
21:18 karstensrage_ joined #git
21:18 BinGOs joined #git
21:18 MatthewAllan93_ joined #git
21:18 emptyn1ne joined #git
21:18 idmistir_ joined #git
21:18 JexSys joined #git
21:18 Ratler joined #git
21:18 zyxep^ joined #git
21:18 ozymandi- joined #git
21:19 meLon joined #git
21:19 PalTale joined #git
21:19 pks_ joined #git
21:19 buffal0 joined #git
21:19 nedbat_ joined #git
21:19 doppo_ joined #git
21:19 GingerGeek_ joined #git
21:19 GingerGeek_ joined #git
21:19 holodoc- joined #git
21:19 idmistir joined #git
21:19 r3c0d3x_ joined #git
21:19 nsh joined #git
21:20 Remramm joined #git
21:20 ocbtec_ joined #git
21:20 hallogallo[m] joined #git
21:20 Mixologic_ joined #git
21:20 deltab__ joined #git
21:21 on1x joined #git
21:21 M-meznak joined #git
21:21 malcon joined #git
21:21 Pursuit_ joined #git
21:21 dermoth joined #git
21:21 devhost_ joined #git
21:21 re1 joined #git
21:21 bsanford joined #git
21:21 liberpater1 joined #git
21:22 Caelum joined #git
21:22 epic joined #git
21:23 DJ_Spies joined #git
21:23 mischat joined #git
21:23 sysanthrope joined #git
21:23 MeltedLux joined #git
21:23 DANtheBEASTman joined #git
21:23 JoshL joined #git
21:23 trobotham joined #git
21:23 ujjain joined #git
21:23 ujjain joined #git
21:23 Cerise joined #git
21:23 Cerise joined #git
21:23 kikijiki[m] joined #git
21:24 kjsage[m] joined #git
21:24 demize joined #git
21:24 nyuszika7h joined #git
21:24 overlord_tm joined #git
21:24 Anja joined #git
21:24 SuperSeriousCat joined #git
21:24 lagothrix joined #git
21:24 M-shine joined #git
21:25 Ineentho[m] joined #git
21:25 zidane_g_ joined #git
21:25 m1dnight_ joined #git
21:25 Cork joined #git
21:25 aviraldg joined #git
21:25 murph joined #git
21:25 saml joined #git
21:25 zero7 joined #git
21:25 led_belly joined #git
21:26 wizonesolutions joined #git
21:26 Awesomecase joined #git
21:26 saintaquinas[m] joined #git
21:26 hariel joined #git
21:26 sfan786_ joined #git
21:26 M-mistake joined #git
21:27 XDS2010 joined #git
21:27 KnownSyntax joined #git
21:27 KnownSyntax joined #git
21:27 ketas joined #git
21:27 belak joined #git
21:27 devhoag joined #git
21:28 vrishab joined #git
21:28 Muzer joined #git
21:28 dreiss joined #git
21:28 POQDavid_ZNC joined #git
21:28 Pursuit_ joined #git
21:28 llamapixel joined #git
21:28 _flow_ joined #git
21:28 hegge joined #git
21:28 Jupelius joined #git
21:28 causasui joined #git
21:28 forrestv joined #git
21:28 bremner joined #git
21:28 Derega_ joined #git
21:28 de-vri-es joined #git
21:28 meZee joined #git
21:28 zapb_ joined #git
21:28 geko joined #git
21:28 ojacobson joined #git
21:28 Fissure joined #git
21:28 alphor joined #git
21:28 cdesai joined #git
21:28 asdofindia joined #git
21:28 wcpan joined #git
21:28 jleclanche joined #git
21:28 evanc joined #git
21:28 sostenuto joined #git
21:28 moop joined #git
21:28 iveqy joined #git
21:28 gusnan joined #git
21:28 PasiFist joined #git
21:28 techntoke_ joined #git
21:28 techntoke_ joined #git
21:28 JelmerD joined #git
21:28 PasiFist joined #git
21:28 arand joined #git
21:28 karstensrage_ left #git
21:29 blackjid joined #git
21:30 fusionx86 joined #git
21:30 nettoweb joined #git
21:30 karstensrage joined #git
21:30 D-Boy joined #git
21:31 slester joined #git
21:31 canuk joined #git
21:31 dcpc007 joined #git
21:31 slester hi there. is it possible to know what files have changed from non-merge commits? if I edit file a, but merge in changes to b and c, is there a way to just get 'a' in the file list somehow?
21:31 Brando753-o_O_o joined #git
21:32 harleypig joined #git
21:32 forrestv joined #git
21:33 kanyeezy joined #git
21:33 cbreak slester: git log -p shows changes in commits
21:33 cbreak slester: --no-merges makes it not show merges
21:33 cbreak also look at --first-parent
21:33 slester cbreak, great, thanks! I swear I've looked at the man page and didn't find --first-parent...
21:34 sostenuto joined #git
21:34 matsaman some systems fail at 'man git foo', FYI, requiring instead 'man git-foo'
21:34 gitinfo the git manpage is available at http://jk.gs/git.html
21:34 JackMc joined #git
21:34 matsaman very informative =P
21:34 littleforest joined #git
21:34 duarteper joined #git
21:36 inflames joined #git
21:38 _dunaeth joined #git
21:38 snowkidind joined #git
21:39 hariel joined #git
21:40 joki joined #git
21:42 Kitty- left #git
21:43 empedokles78 joined #git
21:44 empedokles78 should I first clone a git repo from gitlab and then install a virtualenv or vice versa?
21:44 _ikke_ I don't think that matters
21:47 Urbany joined #git
21:47 unclechu[m] joined #git
21:47 alex1a joined #git
21:47 rich joined #git
21:48 matsaman if it matters he'll know
21:49 aeae64ff8df64d2a joined #git
21:50 buffal0 joined #git
21:51 GavinMagnus joined #git
21:52 msonntag joined #git
21:52 metrixx joined #git
21:54 nevodka joined #git
21:54 phos1 joined #git
21:55 phos1 I have a local branch i usually push up to a DEB branch, and then merge it in. Someone else has updated dev branch, can i just pull DEV down and merge it into my local branch without causing any issues (I don’t want to affect dev branch)
21:55 atomi joined #git
21:57 matsaman git merge should tell you if there will be any issues
21:57 _ikke_ phos1: pull --rebase might more sense
21:58 matsaman you can also always backup your local working dir/.git/repo to do terrible testing
21:58 [Brain] joined #git
21:58 phos1 I have local changes i haven’t pushed up yet, So I want to merge in to see how they work together
21:58 SpeakerToMeat joined #git
21:58 quadgrande joined #git
21:58 _ikke_ phos1: pull --rebase reapplies your changes on top of the change comming from remote
21:59 spNk joined #git
21:59 empedokles78 _ikke_, my git ignore is under development/projectx, strangely if I go to my laptop and create the development/projectx and clone in that folder it creates an aditional projectx folder inside.
21:59 lus joined #git
22:01 _ikke_ yes, git does that by default
22:02 empedokles78 _ikke_, so I have to clone in my development directory and don't get any trouble if I have another project?
22:02 sangy joined #git
22:04 phos1 left #git
22:06 xall joined #git
22:06 mrBen2k2k2k joined #git
22:07 tvw joined #git
22:07 rgrinberg joined #git
22:07 joe543546 joined #git
22:07 NeverDie joined #git
22:08 CamonZ joined #git
22:08 johnmilton joined #git
22:09 GavinMagnus left #git
22:12 e14 joined #git
22:13 azerus joined #git
22:14 buffal0 joined #git
22:15 mehola joined #git
22:16 cagedwisdom joined #git
22:16 pks joined #git
22:18 DarkFox_ joined #git
22:20 Balliad joined #git
22:23 azerus joined #git
22:25 relipse joined #git
22:26 Derperperd joined #git
22:27 publio joined #git
22:27 ghostlight joined #git
22:28 Samual joined #git
22:30 azerus joined #git
22:30 Ardethian\\work joined #git
22:31 grumblebee joined #git
22:32 azerus joined #git
22:32 Sasazuka_ joined #git
22:33 atomi joined #git
22:34 oxsyn joined #git
22:38 DJ_Spies joined #git
22:38 _xor joined #git
22:38 netj joined #git
22:38 _xor joined #git
22:39 DeltaHeavy joined #git
22:39 joe543546 Hi, When trying to mirror my git repo to svn using git svn dcommit, the first time I run the command I get Error from SVN Wrong or unexpected property value: Cannot accept non-LF line endings in 'svn:log' property. But the second time I run the command, the commit goes through without errors. Any thoughts?
22:39 madewokherd joined #git
22:41 ISmithers joined #git
22:42 nettoweb joined #git
22:43 FilipNortic joined #git
22:44 buffal0 joined #git
22:47 cdown_ joined #git
22:48 thiefg joined #git
22:49 DeltaHeavy joined #git
22:49 deltab__ joined #git
22:49 Mixologic_ joined #git
22:49 cdg joined #git
22:49 Vgr joined #git
22:49 isysd joined #git
22:49 Guest40834 joined #git
22:49 Mikaela joined #git
22:49 slester left #git
22:51 DeltaHeavy Hey, I'm using git-for-windows SDK and whenever I run 'git status' or 'git diff' it takes a forever to do anything. It's a very very large project but it wasn't always this bad. I recently disabled the indexing service on Windows 7 because it was constantly running my VM's CPU at 100%. Anybody have any suggestions? I have nothing in my prompt to get the branch number. I think it has to do with looking up files.
22:52 matsaman so you want a solution other than the prompt keeping track
22:53 shinnya joined #git
22:56 Pulp joined #git
22:56 spNk joined #git
22:59 nettoweb1 joined #git
22:59 nowhereman joined #git
23:00 buffal0 joined #git
23:00 shinnya joined #git
23:04 DeltaHeavy matsaman I already removed it. I just want to speed things up
23:04 DeltaHeavy Had to afk for a moment.
23:05 matsaman course, IRC =)
23:06 Jakie joined #git
23:06 matsaman well if you did the prompt "right", it'd actually save you time
23:06 matsaman even if it'd be artificially
23:06 matsaman but I'm guessing you just want it faster as it was anyway
23:06 DeltaHeavy Thanks for understanding :p Some people don't see it as a passive-esque protocol
23:07 DeltaHeavy matsaman I'd rather deal with adding my prompt after I get it sped up.
23:07 DeltaHeavy Was thinking I'd use this but I haven't tried yet but I have it aliased: git symbolic-ref --short HEAD
23:07 DeltaHeavy I'm also using zsh w/ oh-my-zsh for the record.
23:07 DeltaHeavy bash seems to be just as slow though
23:08 matsaman some people do indeed seem to miss the fact that IRC has a straightforward linear progression =P
23:08 DeltaHeavy Yup lol
23:08 matsaman could you um
23:08 matsaman it _could_ be a waste of time, _but_
23:08 matsaman the first thing I'd probably do is compare performance on a GNU/Linux box
23:08 matsaman sad but true
23:08 nettoweb joined #git
23:09 * matsaman has Windows reliability trust issues
23:09 DeltaHeavy matsaman OSX would be good enough right?
23:09 DeltaHeavy Same
23:09 matsaman I would think so, yeah
23:09 Vgr joined #git
23:09 Vgr joined #git
23:09 matsaman for this particular slack "test"
23:09 Vgr joined #git
23:09 Vgr joined #git
23:09 DeltaHeavy matsaman I'll copy it off my VM and move it onto my host OS. Over the network mount it seems to act all weird, the shell that is.
23:09 matsaman oh it's in a VM, too?
23:10 magellanicloud joined #git
23:10 matsaman could just be a network or VM overhead thing, in that case
23:10 matsaman but yeah, worth testing locally vs not, not virtualized vs virtualized
23:10 matsaman not windows vs windows
23:10 DeltaHeavy matsaman I'm using git-for-windows SDK in my VM so I don't have to deal with that layer of stuff.
23:10 DeltaHeavy Could still be VM overhead but my co-workers seem to have this issue too who are running Windows 7 as their host OS, no VM.
23:10 matsaman which layer? Sorry I don't know git-for-windows
23:11 DeltaHeavy You know git bash/
23:11 matsaman git bash?
23:11 DeltaHeavy Like....when you use git on Windows it comes with some configured msys (Linux emulatorish thing) instance.
23:11 DeltaHeavy Copying the files now, might take a while.
23:11 matsaman oh okay
23:11 matsaman I think I suffered that experience once
23:11 matsaman so not actual emulated hardware
23:11 fusionx8_ joined #git
23:12 matsaman you're on Windows... 7?
23:13 troulouliou_dev joined #git
23:14 DeltaHeavy Not me, my co-workers, some of them anyway. I'm on OSX running Windows 7 on parallels.
23:14 matsaman hrmmm
23:14 matsaman but why?
23:14 matsaman sorry maybe I missed something reading
23:15 matsaman why wouldn't you just use git in macOS normally?
23:15 Mixologic_ joined #git
23:15 cdg joined #git
23:15 isysd joined #git
23:15 Guest40834 joined #git
23:15 Mikaela joined #git
23:16 DeltaHeavy matsaman The shell acts funky over the virtual network mount
23:16 DeltaHeavy I should give it another go, maybe they updated it.
23:16 eb_ joined #git
23:16 matsaman so you have macOS, and you have a Windows VM, and then you also have a 3rd installation?
23:16 DeltaHeavy 'cd /Volumes/C/www/genome'  ...just hanging
23:17 matsaman is that the Windows?
23:17 DeltaHeavy OSX
23:17 matsaman I mean the path
23:17 matsaman is that to the data on the Windows guest?
23:17 DeltaHeavy Yes
23:17 matsaman okay
23:17 matsaman why can't the data be just... on your plain macOS hard disk somewhere?
23:18 matsaman s/hard disk/root fs/
23:20 DeltaHeavy matsaman Cause the back end is SQL Server and C#.NET on IIS
23:20 DeltaHeavy And it causes issues when it's grabbing stuff from my OSX volumes
23:20 matsaman oh dear
23:21 leeN joined #git
23:22 DeltaHeavy It's only the first git status/diff unless I let it sit for an hour btw
23:22 DeltaHeavy Running git status on OSX
23:22 DeltaHeavy running...
23:22 matsaman and you don't want to, for example
23:23 DeltaHeavy Still...
23:23 matsaman pull from your "remote", & push back to it every time you want to check something?
23:23 DeltaHeavy It's hanging on OSX
23:23 DeltaHeavy What do you mean?
23:24 matsaman mmm, hrmmm
23:24 matsaman so there's one remote backend that you all share?
23:24 matsaman oh you all have copies of it?
23:24 matsaman and your copy is under parallels
23:25 matsaman if it were me, man...
23:25 _ADN_ joined #git
23:25 mizu_no_oto joined #git
23:25 grumblebee left #git
23:25 DeltaHeavy matsaman No, we all have our own VM if we're on OSX or our local environment is our dev environment.
23:25 matsaman I'd probably get all the tsql & c# running natively on macOS
23:25 DeltaHeavy It won't
23:25 DeltaHeavy It's a LOT of code
23:26 matsaman horrors
23:26 matsaman well, you could, for example, try...
23:26 DeltaHeavy Yeah :/ It's pretty good for how long it's been going and what it is.
23:27 Rodya_ joined #git
23:27 matsaman you could try using vmware instead, or virtualbox instead
23:27 matsaman or... I don't follow docker but it's _possible_ they have some interface with vbox if you like docker
23:27 matsaman other things you could:
23:27 matsaman you could put an ssh server on the windows install
23:27 matsaman and mount via sshfs
23:28 matsaman or an nfs server, and mount via nfs, possibly
23:28 DeltaHeavy The shell I use most of the time is on the guest OS
23:28 DeltaHeavy We're using Parallels for virtualization
23:28 DeltaHeavy And it's just as slow on OSX
23:28 mablae__ joined #git
23:28 matsaman so the shell is slow regardless of whether you're at /Volumes/C or someplace local to the guest?
23:29 matsaman how big is the repo in gigs?
23:29 DeltaHeavy matsaman The shell I primarily use is ON my Windows OS
23:30 DeltaHeavy Checking now
23:30 matsaman so the problem is the shell/guest is slow, and not that access to the guest FS from the host is slow?
23:30 Zamyatin joined #git
23:31 DeltaHeavy 15G
23:31 matsaman I would be surprised if that isn't down to the virtualization overhead being too great / your hardware being too weak
23:31 DeltaHeavy Well I just ran it on my host OS, host fs, host shell, and it was still slow
23:31 mizu_no_oto joined #git
23:31 matsaman if it's become much slower without exponentially growing in size, it could be your hard disk or something related is wearing out, though
23:31 emg joined #git
23:32 matsaman you could try putting it on an SSD :/
23:32 Opertus joined #git
23:32 matsaman with Windows, you'd probably want to go with 64GB at least, but it might all fit on 32
23:32 inflames joined #git
23:34 matsaman I mean if your coworkers are using Windows alone, and you're using macOS with virtualized Windows, I would already expect your setup to be considerably slower
23:34 matsaman and that's just with hardware parity
23:34 matsaman realistically, non-apple boxes will have better hardware, because it was cheaper :/
23:34 littleforest joined #git
23:34 matsaman frequently that is offset by Windows sucking so hard, but still
23:34 DeltaHeavy matsaman It is on an SSD
23:35 matsaman I've an iMac and a non-apple box at work, for example, and the non-apple has twice the ram =P
23:35 matsaman DeltaHeavy: oof
23:35 matsaman do you have another SSD you can try it on?
23:35 DolpheenDream left #git
23:35 EvilPenguin joined #git
23:37 matsaman so basically I think, if the most significant change has been the passage of time and not the growth of the codebase, then it's probably the slow demise of your hardware, or the slow demise of your software (update/system maintenance cruft) that's to blame
23:37 astrofog joined #git
23:37 matsaman the latter is quite easy to test against, if you have the time: make a fresh install or two & compare
23:37 matsaman the former is more of a PITA unless you have spare hardware
23:38 DeltaHeavy matsaman My hardware isn't old
23:38 matsaman I don't mean it's old, I mean it, like all hardware, has aged
23:38 DeltaHeavy I should check things out on the new dev's machine.
23:38 DeltaHeavy It's less than a year old mine is
23:39 matsaman that can be a long time for an OS install
23:39 matsaman QA isn't always what it should be
23:39 matsaman new dev also using macOS?
23:40 DeltaHeavy matsaman Nope, Windows 7
23:41 Peuc2 joined #git
23:42 matsaman you might look into whether it's using a sparse file or not (more performant)
23:42 matsaman whether something odd has happened, like fragmentation
23:42 Opertus joined #git
23:43 matsaman & you can maybe get an idea from others in #macosx what sort of io/cpu usage is the norm for parallels
23:43 * matsaman shrugs
23:43 matsaman 's'all I got, someone else's turn =)
23:45 DeltaHeavy matsaman Well it's slow on Windows 7 nativly too so I don't think it's a virtualization thing
23:46 matsaman the git status stuff is?
23:46 DeltaHeavy Yeah, git diff and stuff too
23:46 matsaman oh?
23:46 DeltaHeavy Only the FIRST one
23:46 matsaman I'm used to git branch being a little slow
23:46 matsaman git diff is usually very fast
23:47 DeltaHeavy If I 'git status', it'll hang. It'll then run then I can run it again or git diff really fast
23:47 DeltaHeavy So I think it's some indexing shit or something maybe
23:47 DeltaHeavy idk how git really works under the hood as far as mapping files goes.
23:47 matsaman http://stackoverflow.com/questions/4485059/git-bash-is-extremely-slow-in-windows-7-x64 ?
23:47 matsaman via =P http://www.google.com/search?q=windows%20git%20slow
23:48 matsaman seems a common problem
23:48 DeltaHeavy Did that
23:48 DeltaHeavy Maybe I need to do it with every terminal instance though...
23:50 matsaman hrmmm
23:50 matsaman if I had to guess, git config -l will reflect it's already aware of it
23:50 * matsaman shrugs
23:51 matsaman there are lots of explanations, though, eg: http://stackoverflow.com/questions/31848137/why-is-git-slow-on-my-windows-machine
23:51 matsaman if you don't have the latest git client, definitely try upgrading
23:51 eindoofus joined #git
23:51 matsaman nice rundown on github: https://github.com/msysgit/msysgit/wiki/Diagnosing-why-Git-is-so-slow
23:52 anonymuse joined #git
23:52 matsaman ooh or this: https://github.com/gitextensions/gitextensions/issues/2438
23:52 matsaman overzealous AV scanning every read/access
23:52 matsaman that would definitely slow things down, and would definitely also be virtually Windows specific
23:53 matsaman and AV changes are the kind of silly things sysadmins like to change without telling anyone
23:55 metrixx_ joined #git
23:59 DeltaHeavy That looks proper
23:59 DeltaHeavy Thank you very much. This looks like it
23:59 DeltaHeavy I'll talk to the sysadmins and see if the AV could be toned down but idk...lots of regulations we need to meet

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