Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-10-10

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

All times shown according to UTC.

Time Nick Message
00:00 clemf joined #git
00:02 dsdeiz joined #git
00:02 dsdeiz joined #git
00:03 nevodka joined #git
00:04 dvdyzag joined #git
00:05 timvisher joined #git
00:09 thierryp joined #git
00:10 mizu_no_oto joined #git
00:13 pigeon joined #git
00:14 jason237 joined #git
00:16 anuxivm left #git
00:16 dodobrain joined #git
00:17 Tonkers Vampire0: sorry, stepped out for a bit. I've made a git repository out of my public_html folder, and cloned it onto my local machine. Are you saying I have to push my local changes to a bare repository, and then merge the files over to my codebase somehow?
00:18 SwiftMatt joined #git
00:18 Vampire0 Tonkers, I'd say you are hitting !deploy
00:18 gitinfo Tonkers: Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
00:21 star_prone joined #git
00:21 xall i'm doing a git add -p
00:21 xall and i want to split further
00:22 nydel07 joined #git
00:22 xall but it just repeats. seen https://i.imgur.com/P8UNqCn.png
00:22 xall how to split it?
00:22 Tonkers feelings about https://github.com/git-deploy/git-deploy ?
00:23 Vampire0 xall, you can only split if there is an unmodified line to split at. If you have a pure-modification block, you cannot split, but you can use edit. You edit and make it look like you want
00:24 Vampire0 Tonkers, no
00:24 Tonkers I'm so used to using Pantheon for all of this stuff... :(
00:26 mmmveggies joined #git
00:26 xall Vampire0: i've never been able to make sense of edit. if i delete the + on the line i want to seperate hunks, i get an error. if i change the + to #, it doesn't seem to split them
00:27 Vampire0 xall edit is not for splitting, edit is for editing
00:27 Vampire0 xall if you decide whether a hunk should be added or not
00:27 xall right, so how do i "make it look like i want"
00:28 Vampire0 xall,you see the hunk that will be applied
00:28 Vampire0 xall, if you want to change what gets applied, use edit and change it to what you want to be applied (i. e. added)
00:28 Vampire0 xall, I don't understand your question
00:29 Vampire0 xall, make it a valid diff hunk that can be applied to the current state and makes it to the state you want added
00:29 xall Vampire0: this hunk includes the end of a class and the entirety of another added class. i don't want to stage the new class. how can i split it up?
00:29 Vampire0 xall if in your screenshot you want to add the first three lines, delete the rest and you are done
00:29 Tonkers Vampire0: so would I point remote origin to a bare repo on my server, and then run export GIT_WORK_TREE=/deploy/dir , git checkout -f master from the bare repo?
00:30 raijin joined #git
00:30 xall Vampire0: ah, so this is just what gets staged? i imagined it would it would affect my working copy
00:30 xall Vampire0: if that's the case, i'm dumb and that makes sense
00:30 Vampire0 xall, no, you just decide what is added to the index. Your working copy will not be touched
00:31 Vampire0 :-)
00:31 Vampire0 xall, glad I was of help
00:31 xall thanks!
00:31 gitinfo xall: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
00:31 xall Vampire0: thanks!
00:32 Vampire0 yw
00:32 Vampire0 Tonkers, that is maybe one way. Yet I never used Git for !deploy. All I can say is read the link
00:32 gitinfo Tonkers: Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
00:32 Tonkers Gotcha. Thanks.
00:35 mmmveggies joined #git
00:35 gitinfo Tonkers: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
00:36 Vampire0 you're welcome tonyj
00:36 Vampire0 you're welcome Tonkers
00:36 Tonkers Vampire0: thanks!
00:36 Vampire0 godd n8 guys
00:36 Tonkers Peace
00:36 Vampire0 I'm going to fetch some sleep
00:36 Tonkers .karma Vampire0
00:36 gitinfo the Genuine Real Life Karma™ REST API results are back! v‍ampire0: 24 (1 in past 30 days) (3 given out)
00:36 Tonkers thanks Vampire0
00:36 Tonkers there did that work
00:36 Tonkers .karma Vampire0
00:36 Vampire0 sure
00:36 gitinfo the Genuine Real Life Karma™ REST API results are back! v‍ampire0: 24 (1 in past 30 days) (3 given out)
00:36 Tonkers oh well
00:37 Tonkers Checkout sleep :)
00:37 Vampire0 But you need several to get one point
00:37 Tonkers gotcha
00:45 diogenese joined #git
00:50 thiefg joined #git
00:50 dodobrain joined #git
00:51 dreiss joined #git
00:51 dsdeiz joined #git
00:51 dsdeiz joined #git
00:53 Motoko joined #git
00:53 theskillwithin lol http://captiongenerator.com/151522/Hitler-uses-Bootstrap
00:54 buffal0 joined #git
00:54 fuchstronaut joined #git
00:55 ekinmur joined #git
00:57 NimeshNeema joined #git
00:59 mmmveggies joined #git
01:02 buddyq joined #git
01:02 danslo joined #git
01:03 timvisher joined #git
01:06 raijin joined #git
01:08 greendo joined #git
01:08 Mixologic_ joined #git
01:09 mrwat joined #git
01:09 Lipp joined #git
01:11 thierryp joined #git
01:11 tristanp joined #git
01:12 inflames joined #git
01:13 FirewalkR joined #git
01:15 red82 joined #git
01:16 sunri5e joined #git
01:18 fstd_ joined #git
01:21 nevodka joined #git
01:22 d^sh joined #git
01:22 nevodka left #git
01:23 pks joined #git
01:24 lindenle joined #git
01:25 ryouma joined #git
01:27 Dazzyo joined #git
01:27 tristanp joined #git
01:28 shock- joined #git
01:28 xall joined #git
01:28 JanC_ joined #git
01:28 buffal0 joined #git
01:28 adante joined #git
01:28 ryouma i have a directory with backup files that look like a.~1~, a.~2~, etc.  each one is a particular version of a file with a particular mtime.  i'd like to convert a subset of these, with mtimes, to git.  my goal is just to be able to do git log and git diff.  are there any scripts out there that can just do this?
01:29 Derperperd_ joined #git
01:29 HoierM joined #git
01:29 holdsworth_ joined #git
01:30 nattyP joined #git
01:30 mycelium joined #git
01:30 harish joined #git
01:31 imMute joined #git
01:31 Rhonda joined #git
01:31 skyres joined #git
01:31 timmu joined #git
01:31 pigeon joined #git
01:31 pasky joined #git
01:31 topi` joined #git
01:32 comps joined #git
01:32 rgrinberg joined #git
01:33 f3r70rr35f joined #git
01:33 underpantz__ joined #git
01:34 Arokh joined #git
01:35 justinmrkva joined #git
01:35 ashnur joined #git
01:35 ashnur joined #git
01:35 ghostlight joined #git
01:35 JrCs joined #git
01:35 dsdeiz joined #git
01:35 dsdeiz joined #git
01:36 lucz joined #git
01:36 chachasmooth joined #git
01:37 nydel07 joined #git
01:38 jrahmy joined #git
01:38 varx1 joined #git
01:39 wloncom joined #git
01:40 timvisher joined #git
01:41 Shentino joined #git
01:46 milki shouldnt be hard to script yourself
01:46 milki but it does sound like a problem that has been generalized to a script before
01:46 milki im not aware of any myself
01:46 ochorocho__ joined #git
01:48 ilbot3 joined #git
01:48 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
01:52 dviola joined #git
01:54 mizu_no_oto joined #git
01:54 fuchstronaut joined #git
01:57 Vampire0_ joined #git
01:58 Wulf4 joined #git
01:58 Shock_ joined #git
01:59 buffal0 joined #git
02:00 ken joined #git
02:01 meLon joined #git
02:02 nanga joined #git
02:02 nanga joined #git
02:03 _ix joined #git
02:05 liuzhen joined #git
02:07 safe joined #git
02:12 evanwang joined #git
02:14 QwertyDragon joined #git
02:15 f3r70rr35f joined #git
02:19 K7 joined #git
02:19 Lipp joined #git
02:20 johnlage joined #git
02:25 chachasmooth joined #git
02:25 blahdodo joined #git
02:27 Jameser joined #git
02:30 timvisher joined #git
02:35 nattyP joined #git
02:37 hexagoxel joined #git
02:40 buffal0 joined #git
02:40 netj joined #git
02:42 f3r70rr35f joined #git
02:47 romerocesar joined #git
02:51 DarkFox_ joined #git
02:54 JanC_ joined #git
02:55 fuchstronaut joined #git
02:57 anonymuse joined #git
02:59 fedepad joined #git
02:59 physikoi joined #git
03:03 faux____ joined #git
03:03 anonymuse joined #git
03:03 mizu_no_oto joined #git
03:04 danslo joined #git
03:07 xameada_ joined #git
03:08 tristanp joined #git
03:10 cqi joined #git
03:10 boombatower joined #git
03:12 thierryp joined #git
03:13 nd_ joined #git
03:13 lindenle joined #git
03:14 red82 joined #git
03:14 Vortex35 joined #git
03:14 eb joined #git
03:16 tristanp joined #git
03:17 tristanp joined #git
03:18 diogenese joined #git
03:20 Geo joined #git
03:20 chid joined #git
03:22 Geo joined #git
03:22 Geo Hi, looking for comments on if there's a better way to do this- a user submits a PR to our repo. Because we're horrible people, it takes us a while to get to it, so by the time we do, it is behind HEAD. We check it out and do git rebase master to get it in synch, then add a few changes. When we try and push that change to the PR, we get the 'tip of your branch is behind...' error.
03:22 e14 joined #git
03:22 Geo So all thats left to do at that point is use push -f . Or so we think.
03:22 Geo Is there some way we could avoid this situation, from that initial scenario?
03:23 Andrew_K joined #git
03:24 EvilPeng1 joined #git
03:24 rchavik joined #git
03:24 anonymuse joined #git
03:25 mmlb joined #git
03:26 arescorpio joined #git
03:30 fahadash joined #git
03:31 raijin joined #git
03:31 timvisher joined #git
03:32 mizu_no_oto joined #git
03:35 FirewalkR joined #git
03:35 FirewalkR joined #git
03:36 dave0x6d joined #git
03:36 FirewalkR joined #git
03:36 joki joined #git
03:37 FirewalkR joined #git
03:38 FirewalkR joined #git
03:38 physikoi left #git
03:38 FirewalkR joined #git
03:39 physikoi joined #git
03:39 FirewalkR joined #git
03:40 xall joined #git
03:40 FirewalkR joined #git
03:41 physikoi Hello #git. Is it possible to merge only "positive" changes?
03:49 snowkidind joined #git
03:49 physikoi joined #git
03:50 physikoi Hello #git. Is it possible to merge positive changes only?
03:51 phroa you can use --no-commit (man git-merge) then edit the result as much as you want before committing.
03:51 gitinfo the git-merge manpage is available at http://jk.gs/git-merge.html
03:52 phroa unless you mean something other and/or more specific than "the good changes, don't merge the bad" by "positive"
03:52 sangy joined #git
03:53 svm_invictvs joined #git
03:55 tristanp joined #git
03:56 jimi_ joined #git
03:56 brokensyntax joined #git
03:57 physikoi phroa: to clarify, i meant positive in the mathematical sense
03:57 star_prone joined #git
03:58 robotroll joined #git
03:58 watabou joined #git
03:59 buffal0 joined #git
04:02 phroa hmm, I'm not the guy then
04:04 deltab joined #git
04:05 InfoTest joined #git
04:05 physikoi phroa: thanks anyway
04:08 rscata joined #git
04:09 Derperperd joined #git
04:09 nevodka joined #git
04:13 kbdkode_ joined #git
04:13 nowhere_man joined #git
04:16 shafun joined #git
04:17 mehola joined #git
04:18 f3r70rr35f joined #git
04:18 Andrew_K|2 joined #git
04:20 mehola joined #git
04:22 raijin joined #git
04:27 Andrew_K joined #git
04:29 raijin joined #git
04:36 ss22ever22 joined #git
04:38 timvisher joined #git
04:39 raijin joined #git
04:41 Derperperd joined #git
04:43 f3r70rr35f joined #git
04:48 Cabanossi joined #git
04:49 justanotheruser joined #git
04:50 holodoc joined #git
04:52 Derperperd joined #git
04:54 f3r70rr35f joined #git
04:56 satya4ever joined #git
04:57 fuchstronaut joined #git
04:57 daey_ joined #git
04:59 mrmadcat joined #git
05:01 hariel joined #git
05:02 Motoko joined #git
05:11 Derperperd joined #git
05:13 jozwior joined #git
05:13 thierryp joined #git
05:14 overlord_tm joined #git
05:15 Derperperd_ joined #git
05:17 werdfriek joined #git
05:17 greendo joined #git
05:17 sea-gull_ joined #git
05:17 yaewa joined #git
05:19 WayToDoor joined #git
05:19 moei joined #git
05:20 ewet joined #git
05:23 cyphase joined #git
05:24 xall joined #git
05:24 mrmadcat joined #git
05:26 a_thakur joined #git
05:30 barteks2x joined #git
05:33 K7 joined #git
05:33 ss22ever22 joined #git
05:35 nilg joined #git
05:36 f3r70rr35f joined #git
05:37 mrmadcat joined #git
05:39 awestwell joined #git
05:40 awestwell hey all I am trying to push a repo to a gitolite server... I get the errors below when running "git push -u gitolite --all"
05:41 awestwell remote: W refs/heads/dev/v1.8.1 test/bx-seam-fm gitolite DENIED by fallthru
05:41 awestwell remote: error: hook declined to update refs/heads/dev/v1.8.1
05:41 awestwell Any suggestions
05:41 j416 awestwell: you're not allowed to; gitolite blocks it.
05:42 awestwell is there a way to allow it?
05:42 j416 why are you pushing everything?
05:42 awestwell failed server
05:42 j416 ?
05:42 awestwell created a new one
05:42 timvisher joined #git
05:43 awestwell is there a way I can allow it for the push
05:43 j416 read this v
05:43 j416 https://help.github.com/articles/importing-a-git-repository-using-the-command-line/
05:43 werdfriek joined #git
05:44 j416 yes, I know you're not using github, but the steps are very similar.
05:44 awestwell ok
05:45 Junior joined #git
05:45 werdfriek How do I do .gitignore? I mean, do I have to create the file? And where is it found / created?
05:45 j416 (assuming that you have a new server to which you are pushing this repo anew)
05:45 mehola joined #git
05:45 j416 werdfriek: create it
05:45 j416 werdfriek: man gitignore explains where it can be
05:45 gitinfo werdfriek: the gitignore manpage is available at http://jk.gs/gitignore.html
05:46 werdfriek j416: cool. Where - on an ubuntu linux machine? In my home dir?
05:46 j416 werdfriek: did you click the link I sent?
05:46 freimatz joined #git
05:46 werdfriek You know I didn't think there would be one for that. Interesting. thx
05:47 freimatz joined #git
05:47 zeroed joined #git
05:47 zeroed joined #git
05:48 j416 werdfriek: tl;dr .gitignore goes in your project directory or a subdirectory of it; not in your home dir. Use a global gitignore file (set in config) to ignore things not related to your project, such as that .DS_Store on a mac or Thumbs.db on Windows.
05:48 ThomasLocke_ joined #git
05:48 ThomasLocke_ joined #git
05:49 werdfriek j416: That's odd ... there is indeed a man page existant, but there isn't a man page for that on my machine.
05:50 j416 werdfriek: what are you typing then?
05:50 werdfriek man gitignore
05:50 gitinfo the gitignore manpage is available at http://jk.gs/gitignore.html
05:50 werdfriek and man 5 gitignore
05:51 j416 then you didn't install those man pages, simply :)
05:51 bq__ joined #git
05:51 werdfriek yeah I know - just now I wonder if there is additional man pages that can be installed for git (and possibly other stuff that I have installed on my machine)
05:51 awestwell so i tried that I did git clone --bare <to the local copy of the repo> and then did a git push --mirror to the remote server
05:52 werdfriek well you learn something everyday. thx
05:52 awestwell the remote still complains with the same remote errors
05:53 alienacorn joined #git
05:55 qt-x joined #git
05:59 LucaTM joined #git
05:59 j416 awestwell: yes.
05:59 Raging_Hog joined #git
05:59 j416 awestwell: you need to configure gitolite to allow you to do that
06:00 bq joined #git
06:00 awestwell I changed the config to be RW+
06:00 j416 I'm no gitolite expert, sorry
06:00 awestwell ok
06:00 j416 don't forget to remove the repo first
06:00 j416 create it anew
06:01 bq joined #git
06:01 j416 or you might end up with refs from both that clone and other clones you might have attempted this with
06:02 bq joined #git
06:02 f3r70rr35f joined #git
06:02 waveclaw joined #git
06:03 bq joined #git
06:08 omps joined #git
06:14 MattE joined #git
06:14 brokensyntax joined #git
06:14 SporkWitch joined #git
06:14 liuzhen joined #git
06:15 omps left #git
06:15 raijin joined #git
06:17 xall joined #git
06:19 dreiss joined #git
06:19 lb1c joined #git
06:21 dc5ala joined #git
06:24 snappy joined #git
06:24 roelmonnens joined #git
06:25 roelmonnens joined #git
06:26 PCatinean joined #git
06:28 MattE joined #git
06:28 chele joined #git
06:30 JeroenT joined #git
06:34 raijin joined #git
06:36 timvisher joined #git
06:38 evanwang joined #git
06:39 Sceorem joined #git
06:39 star_prone joined #git
06:39 jnavila joined #git
06:40 star_prone joined #git
06:42 star_prone joined #git
06:44 f3r70rr35f joined #git
06:46 dan2k3k4 joined #git
06:46 cyan__ joined #git
06:49 thierryp joined #git
06:52 jknetl joined #git
06:52 evanwang joined #git
06:56 tjone270 joined #git
06:57 xall joined #git
06:58 fuchstronaut joined #git
06:59 [Brain] joined #git
07:02 fuchstronaut joined #git
07:02 elect joined #git
07:03 lindenle joined #git
07:04 dunaeth joined #git
07:08 dwmw2_gone joined #git
07:09 titankiller joined #git
07:12 f3r70rr35f joined #git
07:13 grimel joined #git
07:13 grimel joined #git
07:14 vuoto joined #git
07:15 roelmonn_ joined #git
07:17 greendo joined #git
07:17 fedepad joined #git
07:18 titankiller joined #git
07:19 laumann joined #git
07:20 cyphase joined #git
07:21 SwiftMatt joined #git
07:22 TomyWork joined #git
07:24 quenode_ joined #git
07:25 Alienpruts joined #git
07:26 submitnine joined #git
07:27 a_thakur joined #git
07:28 Gloomy joined #git
07:28 PioneerAxon joined #git
07:30 Balliad joined #git
07:30 djb-irc joined #git
07:30 liuzhen joined #git
07:31 timvisher joined #git
07:31 tjone270 joined #git
07:31 _ix joined #git
07:32 watabou joined #git
07:33 overlord_tm joined #git
07:35 redeemed joined #git
07:35 danslo joined #git
07:36 nowhere_man joined #git
07:36 jas4711 joined #git
07:37 fuchstronaut joined #git
07:37 cyphase joined #git
07:39 luxPerpetua joined #git
07:43 Motoko joined #git
07:45 eshlox joined #git
07:45 rbr joined #git
07:47 vuoto joined #git
07:47 Motoko joined #git
07:48 fees joined #git
07:48 DevAntoine joined #git
07:49 djb-irc joined #git
07:49 blaster joined #git
07:50 ocbtec joined #git
07:51 acetakwas joined #git
07:51 roelmonnens joined #git
07:52 cyphase joined #git
07:52 \du joined #git
07:52 \du joined #git
07:54 colona joined #git
07:55 flaviodesousa joined #git
07:56 ogny joined #git
07:57 Mindfart joined #git
07:58 mikecmpbll joined #git
07:59 tvw joined #git
07:59 PioneerAxon joined #git
08:00 ramsub07 joined #git
08:00 _ADN_ joined #git
08:02 nilg joined #git
08:03 babilen joined #git
08:03 cdown joined #git
08:08 ahmedelgabri joined #git
08:11 vuoto joined #git
08:13 vuoto joined #git
08:15 eduardas_m joined #git
08:15 JeroenT joined #git
08:15 cyphase joined #git
08:16 jaequery joined #git
08:17 marcogmonteiro joined #git
08:18 NimeshNeema joined #git
08:19 cdown joined #git
08:21 titankiller joined #git
08:22 vuoto joined #git
08:23 kurkale6ka joined #git
08:25 lss8 joined #git
08:25 noidea joined #git
08:28 Orion3k joined #git
08:29 cyphase joined #git
08:31 Unbeerable left #git
08:31 ogny joined #git
08:32 nilg joined #git
08:32 navidr joined #git
08:32 TheOrangeOne joined #git
08:33 _ix joined #git
08:35 timvisher joined #git
08:36 clmsy joined #git
08:36 bq joined #git
08:42 hexagoxel joined #git
08:42 mizu_no_oto joined #git
08:43 King_Hual joined #git
08:43 mischat joined #git
08:44 qws-user-1228 joined #git
08:45 david__ joined #git
08:45 Mixologic joined #git
08:45 vancrash joined #git
08:45 mischat_ joined #git
08:45 jnewt__ joined #git
08:45 timvishe` joined #git
08:46 faux___ joined #git
08:46 dharriga1 joined #git
08:46 idmistir joined #git
08:46 idmistir joined #git
08:46 ksk_ joined #git
08:47 Kaisyu2 joined #git
08:47 lxnch_ joined #git
08:48 yanope joined #git
08:48 sonOfRa_ joined #git
08:48 ksk_ hola! If I have two remotes in my local git copy, (2x push, 2x pull) and Id like to checkout a branch of one (the one that comes second) how do I tell git?
08:48 bronson_ joined #git
08:49 quetz joined #git
08:49 canton7 ksk_, 'git checkout -b branchname nameofremote/branchname'
08:49 kfunk_ joined #git
08:49 danslo ksk_: git fetch <remote>, git checkout -b <remote>/<branch name>
08:49 canton7 the remote-tracking branches from each remote are namespaces to that remote
08:49 canton7 *are namespaced
08:49 kenansulayman joined #git
08:49 MarkusDB1 joined #git
08:49 ksk_ thanks canton7 danslo
08:49 wwalker_ joined #git
08:49 robogoat_ joined #git
08:49 cite_ joined #git
08:49 brewmast1r joined #git
08:49 bongjovi joined #git
08:49 canton7 so when you're specifying the branch to base your new branch on, it's namespaced to the remote it's on
08:49 stnly_ joined #git
08:49 anoe_ joined #git
08:49 FabTG_ joined #git
08:49 Camusens1i joined #git
08:49 sunny256_ joined #git
08:50 Mathiasd1 joined #git
08:50 cYmen_ joined #git
08:50 KnownSyntax_ joined #git
08:50 KnownSyntax_ joined #git
08:50 bejeb_ joined #git
08:50 Kuukunen- joined #git
08:50 maxb_ joined #git
08:50 tessier_ joined #git
08:50 dsdeiz_ joined #git
08:50 Soliton_ joined #git
08:50 kugel_ joined #git
08:51 Zoxc_ joined #git
08:51 ploop_ joined #git
08:52 jozwior_ joined #git
08:52 Kronuz_ joined #git
08:53 Nile joined #git
08:53 TSS3000RO joined #git
08:53 pks_ joined #git
08:53 zero1985 joined #git
08:53 timvishe` joined #git
08:53 fus_ joined #git
08:53 nebg_ joined #git
08:53 constantinexvi_ joined #git
08:53 ldleworker joined #git
08:53 yossarian joined #git
08:53 cydrobolt_ joined #git
08:53 neunon_ joined #git
08:53 gnarf_ joined #git
08:53 leonarth_ joined #git
08:53 jokajak` joined #git
08:53 SirFunk_ joined #git
08:53 Hink_ joined #git
08:53 gsingh93- joined #git
08:53 rzek- joined #git
08:53 abh_ joined #git
08:53 hfp_ joined #git
08:53 Lorcatar_ joined #git
08:54 Defcronyke_ joined #git
08:54 yossarian joined #git
08:54 Michaela joined #git
08:54 MillerBo55 joined #git
08:54 anonnumberanon_ joined #git
08:54 NightMonkey_ joined #git
08:54 segfaulthunter joined #git
08:54 PaulePan1er joined #git
08:54 emptynine joined #git
08:55 yossarian joined #git
08:55 Afforess joined #git
08:55 corsibu joined #git
08:55 themayor joined #git
08:56 johnnyfive_ joined #git
08:56 Kobaz joined #git
08:56 thierryp joined #git
08:56 enckse joined #git
08:56 qlkzy_ joined #git
08:57 omegak joined #git
08:57 ieee joined #git
08:57 TheMoonMaster_ joined #git
08:57 nilg joined #git
08:57 shakalaka joined #git
08:58 CEnnis91 joined #git
08:58 Iacobus joined #git
08:58 Xethron joined #git
08:58 boxrick1 joined #git
08:58 worstadmin joined #git
08:58 zidane_g joined #git
08:59 huhlig joined #git
08:59 XDS2010 joined #git
08:59 eshlox joined #git
08:59 ionelmc joined #git
08:59 discopatrick joined #git
08:59 rfv joined #git
08:59 ulkesh joined #git
08:59 techntoke_ joined #git
08:59 ec joined #git
08:59 fahadash joined #git
08:59 cory___ joined #git
09:00 inflames joined #git
09:01 Vortex35 joined #git
09:01 varx1 joined #git
09:01 evanwang joined #git
09:01 APic joined #git
09:01 nivag joined #git
09:01 ilogger2 joined #git
09:01 nullie joined #git
09:01 ayjay_t joined #git
09:01 ewet joined #git
09:01 brett_ joined #git
09:01 QwertyDragon joined #git
09:01 buffal0 joined #git
09:01 Sauvin joined #git
09:02 Sonderblade joined #git
09:02 mmattice joined #git
09:02 bdeferme joined #git
09:02 mago_ joined #git
09:02 gnoirzox joined #git
09:02 Sleepy63 joined #git
09:02 Ring0` joined #git
09:02 taargus41 joined #git
09:02 XDS2010 joined #git
09:03 ec joined #git
09:04 mizu_no_oto joined #git
09:04 monokrome joined #git
09:04 delboy1978uk joined #git
09:05 delboy1978uk i need to amend a commit message from 2 commits ago, but how?
09:05 delboy1978uk some sort of rebase, right?
09:05 selckin interactive rebase
09:05 lindenle joined #git
09:06 dinoocch joined #git
09:06 delboy1978uk selckin: thanks
09:07 tomzx joined #git
09:09 buffffal00 joined #git
09:11 danslo delboy1978uk: you might also want to look into autosquashing
09:11 delboy1978uk how do i tell the rebase -i to start from 2 commits ago?
09:12 selckin give it sha of 3 commits ago
09:12 mchl_nix_ joined #git
09:12 delboy1978uk awesome, just found that myself, thanks selckin
09:13 jknetl joined #git
09:14 mchl_nix_ I have a question regarding branches. Given, that I have a develop branch d and a bunch feature branch f1-f5. I find a bug that affects all feature branches in f1. I usually make a hotfix branch h1, where I fix it. How should I then merge the hotfix branch?
09:14 delboy1978uk hooray, my commit message is  compliant and i can push again
09:15 aapaaal joined #git
09:15 mchl_nix_ I usually merge it into develop and all the feature branches, before deleting the hotfix branch. Is that the best practice, or what would be an alternative?
09:15 danslo mchl_nix_: I would merge hotfix to your dev branch, then rebase your feature branches on top of it
09:15 delboy1978uk mchl_nix_: if all feature branches come from the dev branch, i think your hotfix should branch off the dev branch too
09:15 delboy1978uk what danslo said ^^
09:15 selckin too many branches
09:16 mchl_nix_ The hotfix branch does branch from dev, delboy1978uk
09:16 mchl_nix_ What would be the advantage to that, danslo?I don't rebase often.
09:16 mchl_nix_ Less branches in the log?
09:17 danslo mchl_nix_: you could also just merge, but rebasing gives you a bit cleaner history
09:17 delboy1978uk mchl_nix_: in that case merge the fix into dev, and then rebase your feature branches to catch up with the dev branches work
09:17 flaviodesousa joined #git
09:17 mchl_nix_ Looks I should read up on what rebasing is.
09:17 mchl_nix_ *like
09:18 netj joined #git
09:18 delboy1978uk mchl_nix_: merge is just a fetch AND rebase, right?
09:18 mchl_nix_ Thanks, danslo and delboy1978uk :)
09:18 delboy1978uk np
09:18 mchl_nix_ No idea.
09:18 mchl_nix_ Luckily I work on the repo alone, haha.
09:18 delboy1978uk im sure someone here told me that last week
09:18 danslo delboy1978uk: pull = fetch & merge
09:18 danslo that's probably what they said :p
09:18 delboy1978uk danslo: ah yeah loll
09:22 ghostlight joined #git
09:23 manuel_ joined #git
09:24 delboy1978uk left #git
09:24 manuel_ joined #git
09:25 ogny joined #git
09:26 buffal0 joined #git
09:26 manuel_ joined #git
09:27 manuel_ joined #git
09:28 notebox joined #git
09:28 Serpent7776 joined #git
09:29 manuel_ joined #git
09:30 manuel_ joined #git
09:31 Serpent7776 joined #git
09:31 manuel_ joined #git
09:33 manuel_ joined #git
09:34 _ix joined #git
09:34 manuel_ joined #git
09:35 mkoskar joined #git
09:36 jozwior__ joined #git
09:39 Borkr joined #git
09:40 drmt joined #git
09:43 timvishe` joined #git
09:43 peepsalot joined #git
09:44 jozwior_ joined #git
09:44 harish joined #git
09:44 xaa joined #git
09:49 jozwior joined #git
09:49 thesha joined #git
09:51 liuzhen joined #git
09:52 xehbit joined #git
09:53 jbitdrop joined #git
09:55 stupidystupid joined #git
09:58 zero1985 left #git
09:58 Guest26 joined #git
09:59 seaders joined #git
09:59 nettoweb joined #git
09:59 zero1985 joined #git
10:00 seaders hey guys, I ran a `git filter-branch -f --prune-empty --subdirectory-filter android` on a repo I’m on
10:00 zero85 joined #git
10:00 seaders but on pushing to github, github complained about a big file
10:00 seaders but that file had, in history lived in /unity in the big repo
10:00 seaders which, I thought, would have been completely wiped from the history of the repo?
10:01 Murii joined #git
10:01 anpefi joined #git
10:01 mar77i uh, what now. I cloned a remote repository and noticed the subtree is missing, just an empty directory. and I can't pull the subtree into that directory
10:02 mar77i git subtree add --prefix dir url branch // gives: prefix 'dir' already exists.
10:03 mar77i and worse, git subtree pull --prefix dir url branch // gives: fatal: refusing to merge unrelated histories
10:03 thierryp joined #git
10:03 mar77i I just want to pull in the subtree that goes into this stub directoy.
10:05 Snugglebash joined #git
10:06 rafalcpp joined #git
10:08 _ng joined #git
10:10 soc42 joined #git
10:10 bruce_lee joined #git
10:10 bruce_lee joined #git
10:11 wrouesnel1 joined #git
10:11 jodwws joined #git
10:12 cdown_ joined #git
10:16 tcorneli joined #git
10:17 Mixologic joined #git
10:21 mar77i oh, it's a submodule and there's a script that does it for me, which is nice
10:21 Jameser joined #git
10:21 stupidystupid joined #git
10:23 star_prone joined #git
10:26 bauruine joined #git
10:26 nilg joined #git
10:26 star_prone joined #git
10:29 phillly joined #git
10:29 overlord_tm joined #git
10:31 TomyWork joined #git
10:32 dan2k3k4 joined #git
10:33 rominron_ joined #git
10:35 _ix joined #git
10:37 mehola joined #git
10:47 timvishe` joined #git
10:49 fuchstronaut joined #git
10:50 liuzhen joined #git
10:52 rominron_ joined #git
10:53 thesha joined #git
10:55 ieee joined #git
10:56 barteks2x joined #git
10:59 fuchstronaut joined #git
11:00 rafalcpp joined #git
11:00 tlaxkit joined #git
11:01 aapaaal joined #git
11:03 chele joined #git
11:10 Snugglebash joined #git
11:12 somaReverse joined #git
11:12 somaReverse hi
11:13 somaReverse What does this mean https://ptpb.pw/RwYr outputed by "git submodule".
11:13 somaReverse What's the meaning of the prefix "+ / -"
11:14 DJ_Spies joined #git
11:14 tobiasvl somaReverse: it's a diff. what gave you that output? just running "git submodule"?
11:17 danslo joined #git
11:18 thebloggu joined #git
11:18 corsibu joined #git
11:19 somaReverse yes
11:19 rnsanchez joined #git
11:20 Borkr joined #git
11:24 freimatz3 joined #git
11:25 ELFrederich I'm teaching Git some to some colleagues.  They had a question about doing a diff between two versions of a file through the gui.  I showed them that you can launch gitk against a file like "gitk src/foo/bar.c" and will show them visually all history.
11:26 ELFrederich ... from that GUI they can select one commit and right click another and see a "diff" but it isn't a graphical diff.  Any idea how to launch a graphical diff viewer from the GUI?
11:26 ELFrederich I know you can find the SHA vaules yourself and run "git difftool -y <sha-a> <sha-b>", but can you do that from the GUI?
11:31 _ikke_ What is "the GUI"?
11:31 ^7heo _ikke_: "the Graphical User Interface".
11:32 ^7heo _ikke_: you're welcome.
11:32 eizua joined #git
11:32 osse gitk
11:32 raijin joined #git
11:32 FirewalkR joined #git
11:33 _ikke_ ^7heo: -_^
11:33 ^7heo _ikke_: :P
11:34 enckse joined #git
11:35 nuck joined #git
11:35 jozwior_ joined #git
11:35 Juzzika joined #git
11:35 kyan joined #git
11:35 _ix joined #git
11:35 aax joined #git
11:36 Ezzy joined #git
11:36 Jameser joined #git
11:37 freimatz joined #git
11:39 rafalcpp joined #git
11:41 grimel joined #git
11:43 timvishe` joined #git
11:43 achadwick joined #git
11:44 Alienpruts joined #git
11:45 miguelfernandes joined #git
11:46 nowhere_man joined #git
11:47 cdown joined #git
11:47 BadSnappa joined #git
11:48 Infamous joined #git
11:48 johnmilton joined #git
11:49 Snugglebash joined #git
11:50 jimi_ joined #git
11:50 kenansulayman joined #git
11:50 johnmilton joined #git
11:55 synthroid joined #git
11:56 shinnya joined #git
11:58 sgfgdf joined #git
11:58 synthroi_ joined #git
11:59 mischat_ joined #git
12:00 corsibu joined #git
12:02 mischat__ joined #git
12:02 rpg_ joined #git
12:03 dsdeiz joined #git
12:05 hariel joined #git
12:06 mischat joined #git
12:08 ProbabilityMoon joined #git
12:09 mischat joined #git
12:09 paws_ joined #git
12:09 tvw joined #git
12:13 jknetl joined #git
12:14 ELFrederich _ikke_: git gui
12:14 ELFrederich _ikke_: but really... gitk (which can be launched from git gui)
12:15 irco joined #git
12:19 bruce_lee joined #git
12:19 bruce_lee joined #git
12:20 ahmed_elgabri joined #git
12:21 kiresp joined #git
12:21 mischat_ joined #git
12:25 Jameser joined #git
12:26 mischat joined #git
12:27 sdothum joined #git
12:27 mischat joined #git
12:28 Darcidride joined #git
12:31 jnavila joined #git
12:32 re1 joined #git
12:32 timvishe` joined #git
12:34 Jameser joined #git
12:34 thesha left #git
12:34 star_prone joined #git
12:36 _ix joined #git
12:36 thiago_ joined #git
12:37 Jameser joined #git
12:39 mehola joined #git
12:39 perlpilot joined #git
12:40 kiresp joined #git
12:42 dave0x6d joined #git
12:49 perrier-jouet joined #git
12:51 cyan__ joined #git
12:53 rpg joined #git
12:56 Noldorin joined #git
12:57 Endarked joined #git
12:57 FirewalkR joined #git
12:58 ochorocho__ joined #git
12:59 chll_ joined #git
13:03 tjbp joined #git
13:03 seaders hey guys, I’ve run a git filter-branch --subirectory-filter on a repo, but when I look through the history, it’s got history from the other directories, that I specifically wanted it to filter out
13:04 mischat joined #git
13:05 f3r70rr35f joined #git
13:05 seaders I ran `git filter-branch -f --prune-empty --subdirectory-filter android`, but afterwards, if I check `git log --all -- unity/`, it shows the history of commits that were under the “unity” subdirectory
13:06 JeroenT_ joined #git
13:07 ramsub07 joined #git
13:10 ij joined #git
13:11 ij Can I "--bare"-clone without setting config.bare? It feels like this should do, but doesn't. "GIT_DIR=reponame git clone http://host/reponame ."
13:11 kyan joined #git
13:11 malcon joined #git
13:13 _ix joined #git
13:17 timvishe` joined #git
13:18 ExoUNX joined #git
13:18 ExoUNX joined #git
13:20 rgrinberg joined #git
13:20 leeN joined #git
13:20 Snugglebash joined #git
13:21 GodGinrai joined #git
13:23 ogny joined #git
13:23 fmcgeough joined #git
13:24 snowkidind joined #git
13:26 tycoon177 joined #git
13:26 netj joined #git
13:29 tycoon177 I have been rebasing a feature branch as master gets updated. Now that it is ready to be merged into master, what is my best bet for merging them into one so that I can deploy? Can I checkout the master branch and run "git rebase feature_branch" or will that mess things up? Would a git merge be better? I'm just wanting to keep the commit log as clean as can
13:29 tycoon177 be and I worry that merge commit messages might clutter it up
13:29 Sonderblade joined #git
13:30 mmmveggies joined #git
13:30 chele joined #git
13:31 cqi joined #git
13:31 Alienpruts joined #git
13:32 Lipp joined #git
13:32 sonOfRa joined #git
13:33 dcpc007 left #git
13:34 vlt tycoon177: If you want o keep all the single commits from the feature branch then merge (the rebased) feature into master. If you want an even cleaner master log you can --squash the commits into a single one on master.
13:34 quickbrownfox joined #git
13:34 anonymuse joined #git
13:34 Kronuz_ joined #git
13:35 tycoon177 vlt, since it's up to date with master, will there be a merge commit?
13:35 intellix joined #git
13:36 Zamyatin joined #git
13:36 vlt tycoon177: I’m not sure. I propably would branch a test off master and merge there to see how that looks :-D
13:36 tycoon177 didn't think of that! Thanks vlt
13:37 GodGinrai tycoon177: if your feature branch is rebased off of master, then merging it into master should just be a fastforward merge
13:38 tycoon177 Thanks GodGinrai
13:38 GodGinrai np
13:38 corsibu joined #git
13:38 Geo Hi, looking for comments on if there's a better way to do this- a user submits a PR to our repo. Because we're horrible people, it takes us a while to get to it, so by the time we do, it is behind HEAD. We check it out and do git rebase master to get it in synch, then add a few changes. When we try and push that change to the PR, we get the 'tip of your branch is behind...' error.
13:38 jas4711 joined #git
13:38 Geo So all thats left to do at that point is use push -f . Or so we think. Is there some way we could avoid this situation, from that initial scenario?
13:39 raijin joined #git
13:39 GodGinrai Geo: A PR will almost always be behind head, since if you merge one PR, all others become behind HEAD
13:40 ekinmur joined #git
13:40 GodGinrai Geo: are you pulling master before rebasing?
13:41 sangy joined #git
13:42 mischat joined #git
13:45 Geo Yes
13:45 Geo er
13:46 Geo I checkout master, do a git pull origin master
13:46 jaguarmagenta joined #git
13:46 minibar joined #git
13:46 Geo then checkout the PR, and do git rebase master
13:47 qpdb joined #git
13:47 rnsanchez_ joined #git
13:49 patarr joined #git
13:50 nattyP joined #git
13:50 bq joined #git
13:50 rpg joined #git
13:52 red82 joined #git
13:53 plundra joined #git
13:53 sonOfRa joined #git
13:54 plundra What triggers the "Checking out ..."-output when doing a git merge --no-ff? Even with --no-progress. Running 2.10.1; is it really slow disk-io or something?
13:54 plundra We've failed to trigger it our self when testing, but in a cronjob we've hit it.
13:55 GodGinrai Geo: and master is always ahead?
13:55 GodGinrai Geo: after the merge?
13:55 Lartza left #git
13:56 osse plundra: it's based on a timer. if it takes too long then git will print that
13:56 hashpuppy joined #git
13:56 sonOfRa joined #git
13:56 Geo GodGinrai, it seems that way, yes
13:56 GodGinrai Geo: sounds like you need a warden
13:57 GodGinrai Geo: Basically, someone who everything must pass through to go to master.  The only person with push rights to master
13:57 phaleth joined #git
13:57 GodGinrai Geo: He does the merges and pushes to master
13:57 GodGinrai Then no one can push "more recent" branch heads
13:57 plundra osse: Mkay, so even with --no-progress this is triggered?
13:57 GodGinrai for master
13:58 osse plundra: i don't know
13:58 zivester joined #git
13:58 _Vi joined #git
13:59 plundra osse: I know what to look for now at least though, thanks!
13:59 GodGinrai plundra: this seems like an xyproblem
13:59 GodGinrai plundra: !xy
13:59 gitinfo plundra: Woah, slow down for a bit. Are you sure that you need to jump through that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...
13:59 rominron_ joined #git
13:59 GodGinrai Why do you feel you need --no-progress?
13:59 geoid joined #git
13:59 timvishe` joined #git
14:00 osse cronjob?
14:00 osse !deploy
14:00 gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
14:00 seaders hi osse, have you any idea why a `git filter-branch -f --prune-empty --subdirectory-filter …` would still have history from other subdirectories?
14:00 osse ain't nobody got time for doing merges
14:00 GodGinrai osse: yea, but then shouldn't he just use --quiet?
14:00 plundra GodGinrai: Because we got the Checking out (.../...) n%^M over and over again, we added no-progress, it went away. But that was probably accidental if it's timer-based.
14:00 osse seaders: becauser you didn't give the --all switch to work on all branches
14:00 Geo GodGinrai, even in that case, if we get 4 PRs at the same time, 3 of them will be out of date by the time the first is pushed
14:01 seaders osse, I did
14:01 osse GodGinrai: he should fetch and reset
14:01 osse or checkout
14:01 plundra We're doing a fetch, verify all new changes, then do a merge --no-ff up to that point.
14:01 osse seaders: what's the full command, then? and how do you determine the history is not gone?
14:01 seaders full command is `git filter-branch -f --prune-empty --subdirectory-filter android -- --all`
14:02 GodGinrai Geo: sure.  But if one person does all the pushing, the remote master will never be ahead of his own master
14:02 plundra GodGinrai: And we don't want quiet; we want to get the output of what changed.
14:02 seaders if I then do a `git log --all -- unity/tom`
14:02 seaders there’s history available from that file
14:02 seaders even though it only existed within another subdirectory from the original root
14:03 GodGinrai plundra: you hadn't expressed that before.  That's why I asked you to explain why you felt you needed --no-progess
14:03 jimi_ joined #git
14:03 Lipp joined #git
14:03 osse seaders: the --all will include the refs/original hierarchy of refs
14:03 osse aka filter-branche's backups
14:03 plundra GodGinrai: Yeah, no I get that.
14:03 seaders aha
14:04 Lipp joined #git
14:04 seaders what would be a cleanup to blow all of those away?
14:04 plundra GodGinrai: I've been scratching my head with this for a while; you have no context what so ever ;-)
14:04 seaders I’ve tried a million cleanup commands already :D
14:04 osse seaders: git update-ref -d refs/blabla
14:04 hanneso joined #git
14:05 seaders that `blabla` is a commit or a branch?
14:05 seaders I’ve about 2000 branches
14:05 GodGinrai plundra: if you want the output of what changed, could you not throw in a `git diff HEAD~1` after the successful merge?
14:05 osse seaders: see the bottom of man git-filter-branch. it has a xargs thingy to delete them
14:05 gitinfo seaders: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html
14:06 osse seaders: got for-each-ref --format='delete %(refname)' refs/original | git update-ref --stdin
14:06 plundra GodGinrai: Something like that might work, although it feels wrong to do it in different steps, whatever is on disk might have changed in theory. Although we probably have a few other cases of that.
14:07 plundra GodGinrai: Kind of annoying to work around though, I think --no-progress *should* suppress it and it's just in some cases with large changes applied.
14:07 osse plundra: how about running diff/log directly instead of relying on output from merge ?
14:08 FirewalkR joined #git
14:08 Filiphe joined #git
14:09 _ix joined #git
14:09 cdown joined #git
14:09 plundra osse: That might have to be a workaround, yeah. But with quiet, what else is supprsed? We want all warnings, errors, etc.
14:09 osse i have no clue
14:10 osse I'd use >/dev/null :P
14:10 osse and trust that git prints important stuff to stderr
14:10 cmn joined #git
14:10 seaders osse, I’ve just run that git for-each-ref command, and it’s still the same
14:10 osse seaders: !repro
14:10 gitinfo seaders: [!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
14:13 seaders osse: https://gist.github.com/seaders/ee2e69ab17191a8420995115bc5dade1
14:14 seaders unfortunately the repo I’m splitting is private, so I can’t share that exactly to you
14:14 Darcidride joined #git
14:14 seaders but I definitely think I shouldn’t be getting history from other directories when I complete this process
14:15 maret joined #git
14:15 osse what does a plain git for-each-ref print afterwards?
14:16 mischat joined #git
14:16 Geo GodGinrai: sorry for the slow comeback, trying to make sure I'm not asking dumb questions
14:16 _ix joined #git
14:16 Geo we're not pushing to the remote master, we push the change back to the PR
14:16 minibar joined #git
14:16 Geo so that the PR can be merged with the additional change
14:16 seaders you want the whole output?  there’s tons there
14:16 troulouliou_div2 joined #git
14:17 GodGinrai Geo: you're pushing the change back to the PR?
14:17 Geo but it appears that rebasing the PR with the new master changes mucks with its history to where it cant be merged
14:17 Geo yes
14:17 osse seaders: do any of them have "original" in the name ?
14:17 seaders all sorts of `7dd91d3dc4b9b3b84c5433635007874ebfe2194f commitrefs/heads/…`
14:17 seaders no
14:17 seaders only “refs/heads” and “refs/tags"
14:17 seaders hmmm
14:18 GodGinrai Geo: well no wonder you are getting a problem
14:18 seaders I think the tags are what are keeping those bad commits
14:18 osse seaders: do git log -- dirthatshouldnotexist; grab a SHA1 and do  git branch --contains thatsha1here
14:18 osse ahhhh
14:18 GodGinrai Geo: rebasing creates new commits.
14:18 osse yes, tags wil
14:18 osse seaders:  --tag-filter cat
14:18 GodGinrai Geo: you aren't *supposed* to pull from anywhere, rebase the commits, and push back to the same medium
14:18 Geo right
14:19 GodGinrai Geo: once it is public, you don't rebase.
14:19 Geo so think of the problem case of updating a changefile
14:19 Geo *changelog
14:19 red82 joined #git
14:19 Geo everythime something new goes into the master, it changes
14:19 GodGinrai if you were merging with master instead of the PR, the rebasing would be fine
14:19 GodGinrai because master doesn't already have those commits
14:19 Geo and we have to update it in the PR before we can add the PRs new change
14:19 GodGinrai so you would correctly create a fastforward merge situations
14:19 GodGinrai *situation
14:20 Emperor_Earth joined #git
14:20 GodGinrai Geo: if you have to update it in the PR, then you should be performing a merge, not a rebase
14:20 Geo Hmm
14:20 seaders osse, we’ve all sorts of old tags (this is an old BIG repo) so that won’t work, *but* “git for-each-ref --format="%(refname)" refs/tags/ | xargs -n 1 git update-ref -d” does
14:21 Geo so pull the PR, merge the changefile, push that back to PR, then merge that to paster?
14:21 seaders if I run the for-each-ref for getting rid of original, and tags we’re good
14:21 tcorneli joined #git
14:21 Emperor_Earth joined #git
14:21 osse seaders: why won't that work ?
14:21 GodGinrai Geo: pull latest master -> checkout PR -> merge master into PR -> push PR
14:22 osse seaders: oops. I meant --tag-name-filter
14:22 seaders ok, and what would be the command there?
14:22 GodGinrai Geo: and then the merge PR into master after that
14:22 drbean_ joined #git
14:22 seaders not 100% sure what that’s running
14:22 osse seaders: add it to filter-branch
14:22 seaders yup
14:23 osse seaders: "The original tags are not deleted, but can be overwritten; use "--tag-name-filter cat" to simply update the tags."
14:23 seaders ah do
14:23 seaders * d'oh
14:23 seaders sorry, I thought cat was to be replaced with a “category”, or something
14:23 seaders that looks perfect now
14:23 Geo and for clarity of history... can I avoid a merge commit when i merge master into the PR in that scenario?
14:23 seaders going to repro from the start, with that in instead to verify
14:24 osse seaders: you have duplicate fetches and pulls. and btw pull --all just passes --all to the underlying fetch.
14:24 osse also, why create local branches for everything? :O
14:24 drbean joined #git
14:24 seaders because this is going to a new remote
14:24 GodGinrai Geo: no.
14:25 JeroenT joined #git
14:25 seaders we wanted to preserve as much of old remote in new place (in as sensible way as possible)
14:25 GodGinrai Geo: honestly, rebasing doesn't make the history more clear, anyway.  it merely makes it more linear
14:26 Geo for log purposes, I want to avoid every other commit being 'merged master'
14:26 osse seaders: git push newremote oldremote/master:master  :)
14:27 seaders so, the *full* story of this work is…  big fat old repo with sdk/android, sdk/ios …
14:27 GodGinrai Geo: then you shouldn't be updating the PR branch, and should be "accepting" PRs by merging them into master.  If you did that, you could rebase onto master and then push to master
14:27 seaders and in each one of those platforms, they had a “public” submodule folder
14:28 seaders the work I’m doing is splitting that big fat repo into 1 repo per platform, while also bringing in that public repo, as a subdirectory
14:28 seaders and then pushing that godawful mess into a new repo
14:28 GodGinrai Geo: besides, if you need to remove "merged into master" from your logs, there are these great tools like awk and grep that can find these lines and remove them
14:28 GodGinrai Geo: with ease
14:29 seaders there’s a few thousand branches in the private area, and per each one of the public submodules
14:30 Geo GodGinrai, if i merge from local repo, then I lose GH tracking of closing issues and PRs
14:30 Geo blarg
14:30 Geo I can't imagine I'm the only one doing this, heh
14:31 GodGinrai Geo: git log --oneline | grep -v 'Merge remote-tracking branch'
14:31 GodGinrai boom, now you don't see the messages
14:31 mischat_ joined #git
14:32 Geo The point it to avoid them all together
14:32 a_thakur joined #git
14:32 GodGinrai Geo: but there's no reason to.  You are overcomplicating things.
14:33 pR0Ps joined #git
14:33 Derperperd joined #git
14:33 rominron_ joined #git
14:33 GodGinrai Geo: it's always better to have more information at your disposal and to pull from it the information you need than to have less information than you need.
14:34 raijin joined #git
14:34 hashpuppy our junior dev keeps committing to his the tracking branch of our PR branch (develop) and does PRs from there and it bothers this shit out of me and I can't articulate why
14:35 patarr joined #git
14:36 GodGinrai hashpuppy: I'm having trouble understanding what you are saying.  Are you saying that your junior dev commits to your development branch directly and then PRs off of the development branch?
14:36 hashpuppy yeah
14:37 hashpuppy he forked the repos.  and keeps his develop branch in sync w/ upstream/develop.  then commits to develop.  and does PRs from there
14:37 GodGinrai instead of basing things off of master?
14:37 hashpuppy it's obviously an issue when he has to work on multiple things
14:38 jast joined #git
14:38 _ix joined #git
14:38 GodGinrai ahhh, he doesn't make separate branches for each feature?
14:38 hashpuppy yeah
14:38 GodGinrai urk
14:38 GodGinrai that would bother the hell out of me, too
14:38 hashpuppy just trying to stay on topic
14:38 hashpuppy haha
14:40 raijin joined #git
14:40 timvishe` joined #git
14:44 grimel_ joined #git
14:44 jnavila joined #git
14:45 misterpink joined #git
14:46 ash_workz joined #git
14:48 thebloggu joined #git
14:48 fahadash joined #git
14:49 acetakwas joined #git
14:51 aeae64ff8df64d2a joined #git
14:53 sshine joined #git
14:53 Darcidride joined #git
14:54 luke_penn joined #git
14:54 sshine left #git
14:56 e14 joined #git
14:58 Darcidride joined #git
14:59 nikivi joined #git
14:59 basiclaser joined #git
14:59 emg joined #git
15:00 t-mart joined #git
15:01 clemf joined #git
15:01 roelmonn_ joined #git
15:03 kpease joined #git
15:04 concerti joined #git
15:05 K7 joined #git
15:05 Borkr joined #git
15:06 ResidentBiscuit joined #git
15:06 red82 joined #git
15:07 eivarv joined #git
15:09 InfoTest joined #git
15:10 leeN joined #git
15:10 nikivi joined #git
15:11 lindenle joined #git
15:13 Exagone313 Hi, if I have a .git directory, can I extract the files from the latest commit (for example from master)? With git reset maybe?
15:13 Muzer joined #git
15:14 Exagone313 I mean, if all other files are missing, just have a .git
15:15 thiago_ joined #git
15:15 martin- joined #git
15:15 nettoweb joined #git
15:16 selckin git checkout -- .
15:17 Exagone313 what -- means?
15:17 NeverDie joined #git
15:17 selckin seperator between sha and file
15:18 vlt I think "git reset --hard" would do the same.
15:21 zivester joined #git
15:22 sentriz joined #git
15:23 nilg joined #git
15:23 lss8 joined #git
15:23 lss8 joined #git
15:24 timvishe` joined #git
15:24 diogenese joined #git
15:25 FirewalkR joined #git
15:28 Vinnie_win_v joined #git
15:29 PCatinean joined #git
15:30 Exagone313 Thanks you both
15:31 mkoskar joined #git
15:32 NerudaLY joined #git
15:33 buffal0 joined #git
15:34 star_prone joined #git
15:34 Oebele joined #git
15:34 go1111111 joined #git
15:35 faux___ joined #git
15:37 NerudaLY joined #git
15:39 nikivi joined #git
15:39 svm_invictvs joined #git
15:40 zeroed joined #git
15:40 Phylock joined #git
15:42 Lipp joined #git
15:46 acetakwas joined #git
15:47 lindenle joined #git
15:48 freimatz3 joined #git
15:48 littleforest joined #git
15:49 aax joined #git
15:50 dviola joined #git
15:50 kbdkode joined #git
15:51 e14 joined #git
15:52 anaiden joined #git
15:53 jaguarmagenta joined #git
15:54 patarr joined #git
15:54 ansimation joined #git
15:54 kpease joined #git
15:56 Lipp joined #git
15:58 mikel joined #git
15:58 mmmveggies joined #git
15:59 anaiden left #git
16:03 oskarkv joined #git
16:03 sgfgdf hello, guys! i have some feature branch which is behind the one that i've branched off of and i have only one commit which is added. now when i do git pull --rebase origin the-branch-i-branched-off-of it dosn't put my new commit on top of the new changes. am i doing something wrong?
16:05 maucar joined #git
16:05 SwiftMatt joined #git
16:06 thiago sgfgdf: it should, so please show a !repro
16:06 gitinfo sgfgdf: [!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
16:06 maucar joined #git
16:07 Snugglebash joined #git
16:07 Rewneyo joined #git
16:09 mischat joined #git
16:11 timvishe` joined #git
16:12 nitric joined #git
16:13 kulelu88 joined #git
16:13 magellanicloud joined #git
16:14 corsibu joined #git
16:14 seaders osse, I’m much closer now, but there are still other refs / branches with edits not within subdirectory I ran filter-branch —subdirectory-filter on
16:15 seaders they seem to be branches that all but *never* interacted with the android subdirectory
16:16 ochorocho__ joined #git
16:17 Darcidride joined #git
16:19 Electrometro joined #git
16:20 e14_ joined #git
16:22 tyreld joined #git
16:22 Electrometro joined #git
16:26 Ph_ joined #git
16:26 osse seaders: i didn't know that's how it works, but it makes sense
16:26 svm_invictvs joined #git
16:27 seaders slightly, but from the book, https://git-scm.com/book/en/v2/Git-Tools-Rewriting-History
16:27 seaders “Now your new project root is what was in the trunk subdirectory each time. Git will also automatically remove commits that did not affect the subdirectory."
16:27 madewokherd joined #git
16:28 irco joined #git
16:28 seaders I’m currently looking to see if there’s an auto branch prune command, to kill off what shouldn’t have been left behind
16:28 osse but what about commits that never contained that directory at all? :O
16:28 seaders yeah, it’s a brain exercise that’s currently…. melting mine...
16:29 osse man page says "Only look at the history which touches the given subdirectory." To me that reads like the other branches will just be excluded from the processing
16:29 seaders I’ve emailed the mailing list as well
16:29 magellanicloud joined #git
16:29 seaders that does make sense, but when you’re trying to break up a huge repo like this, it certainly doesn’t make it easy!
16:29 seaders there’s gotta be a way to filter out branches, altogether
16:30 le_melomane joined #git
16:32 ronny joined #git
16:32 vuoto joined #git
16:32 ronny hi
16:33 ronny anyone aware of a reasonably simple way to take all commits of a certain author on branch X and cherry pick them onto a new branch?
16:34 osse ronny: git cherry-pick $(git rev-list --reverse --author=certain master..X)
16:35 osse e.g.. I wrote master.. because I assume you want to end the search somewhere
16:36 ronny osse: does that work even if x is not a ancester of master (bit instead has a common ancestor)
16:37 thiago if X were an ancestor, the list would be empty
16:37 osse ronny: it works *only* if it has a common ancestor with master
16:37 thiago it's supposed to be either a descendant, or diverging from
16:37 Noldorin joined #git
16:37 nevodka joined #git
16:37 thiago master..X means all commits in X that aren't in master (all the way from the branch point or last merge point)
16:38 Lipp_ joined #git
16:39 eindoofus joined #git
16:39 LionsMane joined #git
16:42 ronny thanks for the explanations
16:42 ronny things work now
16:43 patarr joined #git
16:43 eindoofus hi, i'm slightly confused about switching branches. i have a branch i'm working on (let's call it xyz) that has some commited/uncommited changes but i need to go to master to update a stub since that's what we're all sharing. then need to push up that one change on master. will my current changes to xyz show up when i switch to master? if not, will i loose them? i believe i can do a stash, but is this the only way to safely switch branches?
16:44 moritz eindoofus: the committed changes won't show up on master
16:44 ksk hey again. How do I please use "git-mv" I tried both "git-mv" and "git git-mv" but "does not work" - am I doing it wrong? (basicly, I have some file in git but Id like to rename it)
16:44 moritz eindoofus: and git tries to "float" the uncommitted changes to master, and the switching of branches aborts if that fails
16:44 ksk eh, its "git mv" internet says, probably gonna fix my problem. thanks!
16:45 astrofog joined #git
16:45 moritz ksk: the internet is right! (says the guy on the internet :-)
16:45 eindoofus moritz, ah, how can i do that that switch without the changes floating? sounds like floating would cause them to easily get added accidentally to master
16:45 EvilPeng1 joined #git
16:46 raijin joined #git
16:46 e14 joined #git
16:46 moritz eindoofus: commit or stash them
16:47 Faylite joined #git
16:48 ronny left #git
16:49 eindoofus moritz, never understood how the devs i work with switch/commit so quickly. do most people have separate projects for each branch open?
16:49 moritz eindoofus: I certainly don't
16:49 moritz eindoofus: but sometimes I just commit stuff I have, even though it's not perfect, and later combine it with more changes, and change the message
16:49 dreiss joined #git
16:50 eindoofus moritz, that never occurred me. so you can add changes to a previous commit.. interesting
16:50 moritz eindoofus: sometimes I just float the changes. I always do a 'git add -p' and never 'git commit -a', so it's relativ safe
16:50 moritz eindoofus: as long as you don't push it yet, sure
16:51 timvishe` joined #git
16:51 mikecmpbll joined #git
16:51 moritz I sometimes have separate checkouts for long-lived branches (several weeks), but that happens rarely
16:53 troulouliou_div2 joined #git
16:55 raijin joined #git
16:56 ochorocho__1 joined #git
16:57 mutex joined #git
16:58 Fisd joined #git
17:02 fedepad joined #git
17:02 SteffanW joined #git
17:04 jaequery joined #git
17:07 dan2k3k4 joined #git
17:12 malcon joined #git
17:14 annoymouse joined #git
17:15 geoid joined #git
17:17 mischat_ joined #git
17:18 eivarv joined #git
17:20 gambler joined #git
17:20 timvishe` joined #git
17:22 relipse joined #git
17:22 relipse for some reason I tried to git merge other-branch and it failed saying a file was in use or something I closed my editor and tried again and now i have like 10 files that are modified 1 deleted, but why? a merge isn't suppoesd to do that
17:23 K7 joined #git
17:23 sentriz joined #git
17:23 Faylite joined #git
17:23 annoymouse joined #git
17:23 _ikke_ relipse: What does git status say?
17:24 _ikke_ You might be in a mid-merge stage
17:25 rgrinberg joined #git
17:26 relipse error: Your local changes to the following files would be overwritten by merge: ... aborting
17:27 relipse so what do I do, how do i go back to the way it was and then try merging again
17:27 GodGinrai relipse: when git merge aborts, I'm pretty sure it remains in a partially-merged state
17:27 raijin joined #git
17:27 GodGinrai unless you abort the merge manually
17:28 a_thakur joined #git
17:28 relipse how do I do that
17:28 _ikke_ git merge --abort
17:28 _ikke_ Should fix it
17:28 aikuma joined #git
17:28 relipse fatal: There is no merge to abort (MERGE_HEAD missing).
17:29 _ikke_ Did you have uncomitted changes before the merge?
17:29 relipse no i just checked out master branch
17:29 aikuma wondering how can i see the contents of the stash?
17:29 relipse its possible i might of left something
17:29 relipse but i'm pretty sure I committed all
17:29 _ikke_ aikuma: git stash show -p
17:29 _ikke_ relipse: git reset --hard HEAD is easiest then
17:30 sea-gull_ joined #git
17:30 _ikke_ relipse: Is this something you can reproduce?
17:31 skalpin I'm trying to rebase a topic branch onto develop, but "git rebase origin develop" isn't doing anything. Why?
17:31 skalpin git log ..origin/develop shows 3 commits that are not in my topic branch
17:32 _ikke_ skalpin: git rebase <base> <branch_to_checkout>
17:32 _ikke_ skalpin: so git rebase develop topic
17:33 _ikke_ ("topic" is optional when topic is already checked out)
17:33 skalpin _ikke_: that didn't do anything either... could it be that I only have 1 commit?
17:33 _ikke_ ah sorry
17:34 _ikke_ git rebase origin/develop
17:35 a_thakur_ joined #git
17:36 luxPerpetua joined #git
17:38 aard_ joined #git
17:39 jaequery joined #git
17:40 emg joined #git
17:40 ahmed_elgabri joined #git
17:40 skalpin _ikke_: I just tested it using local folders. rebasing a branch with only one commit doesn't seem to be doing anything.
17:41 _ikke_ Number of commits doesn't matter
17:41 Nd-47-M joined #git
17:43 SporkWitch joined #git
17:43 oojo joined #git
17:44 skalpin you're right... there must be something else going on
17:46 dsantiago joined #git
17:46 synthroid joined #git
17:47 nettoweb joined #git
17:48 Lipp joined #git
17:49 jaequery joined #git
17:52 laumann joined #git
17:53 AlexxNica joined #git
17:53 timvishe` joined #git
17:54 jaguarmagenta joined #git
17:55 TomyLobo joined #git
17:56 xall joined #git
17:56 Snugglebash joined #git
17:56 mangolisk joined #git
17:57 Sasazuka joined #git
17:57 SporkWitch joined #git
17:58 Goplat joined #git
17:58 bernardio joined #git
17:58 _ikke_ git log --graph --oneline --decorate --all -20
18:00 tessier joined #git
18:01 moestevens joined #git
18:01 deltab_ joined #git
18:03 Orion3k joined #git
18:03 Camusens1i left #git
18:04 FirewalkR joined #git
18:05 spudowiar joined #git
18:05 Camusensei joined #git
18:05 Camusensei left #git
18:06 multi_io_ joined #git
18:06 zivester joined #git
18:06 tvw joined #git
18:06 mikecmpbll joined #git
18:09 WayToDoor joined #git
18:09 dreiss joined #git
18:11 Lipp joined #git
18:12 jaequery joined #git
18:12 ejb joined #git
18:13 xall joined #git
18:18 Spacew00t joined #git
18:19 GandalfTheWizard joined #git
18:19 GandalfTheWizard joined #git
18:19 nettoweb joined #git
18:20 laumann joined #git
18:26 ahmed_elgabri joined #git
18:26 sanketdg joined #git
18:27 Jackneill joined #git
18:28 kyan joined #git
18:30 timvishe` joined #git
18:30 Geo Is it possible to copy a file from a different branch to the current, via git commandline?
18:30 star_prone joined #git
18:32 _ikke_ Geo: git checkout <branc>
18:32 shafun joined #git
18:32 _ikke_ this would put the file as it was in that branch in your working tree
18:32 Kartagis joined #git
18:32 Kartagis ehlo
18:33 _ikke_ if you want to store it under a different name, you could do: git checkout <branch>:<file> >othername
18:33 patarr joined #git
18:33 Kartagis how do I get a specific hash? git checkout?
18:33 luxPerpetua joined #git
18:33 osse Kartagis: define "get"
18:33 grawity hash -> files, or files -> hash?
18:33 grawity commit hash -> files? git checkout <commit>
18:34 Kartagis osse: clone up to a specific hash
18:34 e14 joined #git
18:35 grawity that doesn't make any more sense in comparison
18:35 Geo _ikke_: so git checkout branch:file gives me that file ?
18:35 grawity but I'm still guessing you want 'git checkout <commit>' after the clone
18:36 Kartagis osse: http://stackoverflow.com/a/3489576/1384283 maybe?
18:37 jost joined #git
18:37 altendky joined #git
18:38 rahtgaz joined #git
18:39 rominron_ joined #git
18:39 theskillwithin joined #git
18:39 ronny joined #git
18:39 ronny hi
18:40 ronny anyone aware of a listing of git versions one should support in poreclain commands?
18:40 Anja joined #git
18:40 dodobrain joined #git
18:40 mniip joined #git
18:41 Cassiopaya joined #git
18:41 preaction i try to support as far back as i can without undue hardship, which currently is 1.7.4 i think
18:43 mikecmpbll joined #git
18:43 Fomina joined #git
18:47 Dazzyp joined #git
18:47 malisper joined #git
18:48 _sjs joined #git
18:49 frp joined #git
18:51 timvisher joined #git
18:54 Gsham joined #git
18:54 nikivi joined #git
18:54 timvishe` joined #git
18:55 alex1a joined #git
18:57 apotry joined #git
18:57 monoxane joined #git
18:57 monoxane joined #git
19:00 apotry joined #git
19:02 raspado joined #git
19:02 ramsub07 joined #git
19:03 ahmed_elgabri joined #git
19:03 aca20031 joined #git
19:03 buddyq joined #git
19:06 e14 joined #git
19:07 rjb joined #git
19:07 sarath_babu joined #git
19:08 led_belly joined #git
19:09 e14 joined #git
19:11 hacman joined #git
19:11 bathtub_shark joined #git
19:13 ronny is there soemthing more reliable than git desribe --always --dirty, on travis git this one is wrong :/
19:14 patarr joined #git
19:15 Murii joined #git
19:17 eindoofus_ joined #git
19:19 Derperperd joined #git
19:19 synthroi_ joined #git
19:19 mizu_no_oto joined #git
19:19 serialoverflow joined #git
19:22 Limes_ joined #git
19:22 alex1a joined #git
19:23 tvw joined #git
19:23 ronny is there any git command that tells me about unstaged and staged changes for use from scripts?
19:24 preaction git status --porcelain
19:25 d^sh joined #git
19:25 grayjoc joined #git
19:27 timvishe` joined #git
19:28 Gsham joined #git
19:28 synthroid joined #git
19:28 _ikke_ Geo: Yes, it will output the contents of the file
19:31 stuh84 joined #git
19:31 apotry joined #git
19:34 eindoofus joined #git
19:35 manuels joined #git
19:36 manuels joined #git
19:37 Gsham joined #git
19:37 manuels joined #git
19:38 ekinmur joined #git
19:38 aikuma joined #git
19:39 manuels joined #git
19:39 Deliri0m joined #git
19:40 Deliri0m who is the file FIFO in repo?
19:40 d^sh joined #git
19:40 rominron_ joined #git
19:41 mizu_no_oto joined #git
19:42 NimeshNeema joined #git
19:44 eb joined #git
19:48 cdown joined #git
19:48 fahadash joined #git
19:48 ViktorRu joined #git
19:50 apotry joined #git
19:50 star_prone joined #git
19:50 timvishe` joined #git
19:51 xall joined #git
19:52 hashpuppy joined #git
19:53 Lor_ joined #git
19:54 eindoofus joined #git
19:54 rominron_ joined #git
19:54 jaguarmagenta joined #git
19:57 _ikke_ git has no FIFO file
19:57 eindoofus joined #git
19:57 mizu_no_oto joined #git
19:58 malisper left #git
20:01 mwleeds joined #git
20:02 Deliri0m _ikke_, see https://github.com/ehamberg/simple-gitbot
20:03 elastix joined #git
20:03 _ikke_ Deliri0m: It's a linux thing
20:04 _ikke_ http://man7.org/linux/man-pages/man7/fifo.7.html
20:04 red82 joined #git
20:04 m0viefreak joined #git
20:07 rpg_ joined #git
20:10 metrixx_ joined #git
20:10 metrixx_ hi.
20:10 gitinfo metrixx_: 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.
20:10 metrixx_ svn vs git. which one do you prefer?
20:10 _ikke_ metrixx_: I think most people here prefer git
20:14 concerti !backup
20:14 gitinfo Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.com/GitBestPractices/#backups
20:14 GodGinrai _ikke_: can't tell if troll or not... *squints*
20:15 diogenese joined #git
20:16 _ikke_ GodGinrai: You think?
20:16 patarr joined #git
20:17 timvishe` joined #git
20:24 saml joined #git
20:28 NeverDie joined #git
20:31 mischat joined #git
20:31 ViktorRu joined #git
20:32 nilg joined #git
20:33 brokensyntax joined #git
20:36 geoid joined #git
20:38 ocbtec joined #git
20:40 fscala joined #git
20:41 snfgf joined #git
20:41 timvishe` joined #git
20:41 snfgf good evenigns
20:41 _ikke_ ohai
20:43 alexclark joined #git
20:43 axisys anyone know of a wrapper that will edit a file with vim and then invoke git commit and prompt you for a git commit message? I have been using a wrapper I call ``change'' which call ``co'' and then ``vim'' to edit a file and at the end would call ``ci'' to put it in revision control
20:44 axisys I am planning to replace rcs with git
20:44 alexclark im working on a project with submodules and im trying to verify that my shell script to update the submodule if someone else pinned a newer version works
20:44 alexclark does anyone know a way to bring in a change to the submodule rev hash without actually updating the submodule?
20:46 grawity a) git update-index, commit
20:46 grawity b) cd mysubmodule, check out an old commit >_>
20:46 _ikke_ axisys: I know this is how RCS works, but do you want to only commit one file at a time?
20:47 _ikke_ axisys: git commits entire trees in one go, creating stable snapshots
20:47 nowox joined #git
20:47 axisys _ikke_: I am new to git, so let me explain my goal
20:47 _ikke_ axisys: sure
20:47 aidalgol joined #git
20:47 axisys _ikke_: we manage many tacacs config file .. all under same dir /etc
20:48 axisys _ikke_: we are few in our team and at any time one of us could be modifying a file
20:48 mizu_no_oto joined #git
20:48 axisys _ikke_: however most of the time we run a script that modifies all the files .. like adding a user on multiple tacacs config file
20:49 _ikke_ Ok, it would make sense to have all these changes in one commit then
20:49 axisys _ikke_: only once in a while we have to actually add a new group definition to only one of the tacacs config file and that is not scripted today
20:49 _ikke_ axisys: btw, have you looked at !etckeeper btw?
20:49 gitinfo axisys: etckeeper is a collection of tools to let /etc be stored in a git, mercurial, darcs, or bzr repository. It hooks into various package managers. http://etckeeper.branchable.com
20:49 _ikke_ Never used it myself though
20:49 grawity so I don't really see how this is any specific
20:50 grawity vs regular `git commit`
20:50 _ikke_ axisys: change, add, commit
20:51 grawity `git commit <file>` (avoiding the index) might help with concurrent edits
20:51 _ikke_ axisys: I'd use git add -p, to see what I'm comitting, and then git commit -m "message" (and add -e to edit the message more)
20:51 grawity one important thing is that git *does not* do 'locking' like RCS `co` does
20:51 metrixx joined #git
20:51 axisys I know of etckeeper or something similar from past.. but I am not putting everything under /etc into revision.. only those about 20 tacacs config files
20:51 _ikke_ axisys: alright
20:52 _ikke_ axisys: Comitting a change to a single file is not something that I think needs to be scripted
20:53 axisys _ikke_: right.. only when adding a user to multiple tacacs config file do we run script today.. so yes that make sense
20:53 axisys so I did not about the -p and -e switches.. thanks
20:53 axisys _ikke_: ^
20:54 axisys grawity: right.. no locking here..
20:54 _ikke_ git add -p is imo very powerful
20:54 _ikke_ It lets you review what you are going to commit while staging it
20:54 _ikke_ hunk by hunk
20:54 Snugglebash joined #git
20:54 _ikke_ especially if you fat finger something, it ought to turn up in the diff
20:55 axisys so when one of my team member edit a file today they run ``change'', but when we need to add a user we run a script
20:55 Lipp joined #git
20:56 axisys so script probably can do the git commit automatically, but change probably can be updated to make the git commit.. I think I am getting a better idea on how to handle my situation more clearly now...
20:57 fatalhalt joined #git
20:57 _Vi joined #git
20:57 axisys change is just a bash wrapper that does the checkout/edit/checkin today
20:59 alexclark joined #git
20:59 axisys I guess I could put a lock on the script change, so not two people can run it at the same time.. as a workaround of having no lock
20:59 mycelium joined #git
21:00 nwkj86 joined #git
21:00 timvishe` joined #git
21:01 mrmadcat joined #git
21:02 metrixx joined #git
21:04 Mobutils joined #git
21:10 amdi_ joined #git
21:11 alexclark joined #git
21:12 yqt joined #git
21:13 frobnic joined #git
21:15 porsche944 joined #git
21:15 porsche944 how can view which branches a commit is on
21:16 j416 porsche944: git branch --contains <commit>
21:16 porsche944 thanks!
21:16 j416 welcome o/
21:18 fscala joined #git
21:20 werdfriek joined #git
21:20 red82 joined #git
21:21 Snugglebash joined #git
21:21 timvishe` joined #git
21:24 rns joined #git
21:24 werdfriek I don't get what the problem is ... I created a .gitignore in my home directory and placed *.pro.user* ( have QT project files ) -- but I still see the same project file showing up as an untracked file. What gives?
21:26 grawity Git doesn't read ~/.gitignore by default
21:26 grawity it looks for .gitignore in the repository itself
21:27 werdfriek grawity: is there a way to make some things global?
21:27 grawity ~/.config/git/ignore
21:27 eindoofus joined #git
21:28 werdfriek sigh - I read something like that but ... well idk how I missed it. Thank youl
21:33 bilb_ono joined #git
21:34 bilb_ono how do you specify the ssh key to use when cloning a treatment?
21:34 bilb_ono sorry, cloning a repo
21:34 dsdeiz_ joined #git
21:35 moritz bilb_ono: git uses the normal ssh client, so if you can configure it through .ssh/config, that's generally the easiest
21:36 moritz bilb_ono: and git honors the GIT_SSH_COMMAND environment variable, so you can set it to 'git -i yourkey' or something. man git for more.
21:36 gitinfo bilb_ono: the git manpage is available at http://jk.gs/git.html
21:37 FirewalkR joined #git
21:37 werdfriek grawity: So nothing from config/* on existed on my system. I need clarification, is the */ignore  a file or a dir? If a dir then what is the file name to include inside it?
21:37 werdfriek Thanks
21:37 grawity it's a file.
21:38 snfgf can i git add/git commit from outside of the work tree?
21:38 grawity no
21:38 vlt ~/.config/git/ignore is a thing? \o/
21:38 vlt grawity: Thank you!
21:38 werdfriek grawity: Ok, so, when I created */git/ignore (ignore as a file) there was no change in the output of git status (file still shows up as untracked).
21:38 grawity vlt: core.excludesFile has been a thing for long before it, too
21:39 grawity werdfriek: um, what
21:39 grawity what's "*/git/ignore"?
21:39 grawity is that your shorthand for ~/.config/git/ignore, or something else?
21:39 DeltaHeavy joined #git
21:39 werdfriek grawity: ~/.config/git/ignore   <-- yes my shorthand (sorry_
21:40 grawity werdfriek: how old is your git?
21:40 werdfriek and I try to typ too fast .  It was installed (in a strange way I might add) about 4 days ago.
21:41 nettoweb joined #git
21:41 nilg joined #git
21:41 werdfriek I watched a youtube video where some e. indian dude was having me install from source or something.
21:43 eindoofus joined #git
21:44 timvishe` joined #git
21:45 bilb_ono moritz: once I do that, is there a way when I clone to confirm that its actually using the command I want?
21:45 bilb_ono sorry that its actually using the new config that I created?
21:46 Kitty- joined #git
21:47 snfgf --set-upstream, add upstream tracking reference used by argument-less git-pull
21:47 snfgf ^wut
21:47 snfgf what does it mean
21:47 tgunr joined #git
21:48 werdfriek wow
21:48 EmeraldExplorer joined #git
21:49 cdown_ joined #git
21:50 seitensei Evening, folks.
21:50 HoierM joined #git
21:50 werdfriek I put "ignore" ( a file ) in the working directory and I get "ignore" show up as an untracked effin file. I do cp ignore .gitignore (change the filename) and I end up with ".gitignore" as an untracked file. Seriously??
21:51 werdfriek excuse me - I ran mv ignore .gitignore    not cp
21:51 BlaXpirit ok so?
21:52 werdfriek uhh, is there not a way to have git ignore stuff any longer?
21:52 BlaXpirit put its filename in the .gitignore file
21:52 seitensei or directory
21:52 BlaXpirit but the file being named 'ignore' doesnt make it anything special
21:53 seitensei Make sure .gitignore doesn't have any extensions
21:53 werdfriek And regex is permitted inside the file right? So, for example, if I want to ignore all files ending in .txt I can specify *.txt right?
21:53 seitensei Yep.
21:53 werdfriek seitensei: What's an extension?
21:54 seitensei .txt, .rtf...etc.
21:55 werdfriek seitensei: I see what you mean. No, it doesn't (I'm working from the bash command line in linux so no it won't add anything that isn't explicitly specified in the filename)
21:55 seitensei werdfriek, what software is your box running
21:55 seitensei ok, linux
21:55 werdfriek seitensei: Ubuntu 16.04.
21:55 jaguarmagenta joined #git
21:55 seitensei make sure the .gitignore file in your repository root
21:55 seitensei on the same level as the .git folder
21:56 werdfriek git version 2.10.0.GIT
21:56 a3Dman joined #git
21:56 seitensei It's not going to work retroactively
21:56 alvarezp joined #git
21:56 seitensei so anything you've already added to your commits/remote will still be there
21:56 werdfriek seitensei: I think that is the problem. Do I have so "clear" git status (so to say)?
21:57 seitensei you need to remove it from cache
21:57 anuxivm joined #git
21:57 werdfriek so what? git rm < filename > ?
21:57 snfgf guys, i have created a new branch, pushed it, then gone into a previous clone, and run 'git branch -a', but i can't see the remote branch i pushed another clone
21:58 seitensei git rm --cached file
21:59 alvarezp Hi, I have merge commit in which a file was lost. One branch had the file created and the other branch didn't know anything about the file. In the resulting merge the file is not there. The commit doesn't have a "Merge ..." message. I expected to have the removal revealed using git log --all --follow --stat -- file, but I only see the creation. Any clues as to how to diagnose this?
22:00 drizztbsd joined #git
22:01 alvarezp (the only thing I can think of is that the other branch did a git rm of a parent directory)
22:01 werdfriek seitensei: cool.  Ok, so I guess you have to tell git to ignore .gitignore (include it in the file). I didn't have to rm anything and all is well --
22:02 seitensei You want your .gitignore tracked
22:02 seitensei Not sure why you would want it untracked
22:03 werdfriek So how could I make it global then?  I have a .config in my home directory but nothing after that. I tried a file named .gitignore as well as a file named ignore earlier (after creating the aforementioned git directory inside .config). Neither of them did a single thing.
22:04 _ADN_ joined #git
22:04 werdfriek So Im  a little confused what he right way to ignore stuff globally is
22:05 patarr joined #git
22:08 Bob8989|2 joined #git
22:08 WayToDoor joined #git
22:09 timvishe` joined #git
22:09 cdown joined #git
22:09 seitensei werdfriek, git config --global core.excludesfile /path/to/global/.gitignore
22:10 eb joined #git
22:10 seitensei You don't want to ignore gitignores, since they should contain repo/project specific ignore rules
22:11 seitensei very important if you're sharing your git repo with someone else
22:11 seitensei since they might not have the same global ignore
22:11 cdown joined #git
22:13 brokensyntax joined #git
22:16 HoierM joined #git
22:18 endiruna joined #git
22:19 werdfriek seitensei: So what does git prioritze then? The reop level ignore file, or the global level?
22:19 werdfriek seitensei: Sorry, phone call.
22:24 werdfriek joined #git
22:25 mquin joined #git
22:25 werdfriek left #git
22:25 werdfriek joined #git
22:28 red82 joined #git
22:28 werdfriek joined #git
22:28 werdfriek seitensei: Hey, thanks, that worked. I just made / set git to look in my home dir ( like old days ).
22:28 FirewalkR joined #git
22:29 timvishe` joined #git
22:30 mikecmpbll joined #git
22:31 Infamous joined #git
22:32 tilpner joined #git
22:33 kenansulayman joined #git
22:34 ericsupreme joined #git
22:39 fatalhalt joined #git
22:39 sangy joined #git
22:44 Orbitrix joined #git
22:44 Orbitrix joined #git
22:44 snfgf hello
22:45 theskillwithin joined #git
22:46 alvarezp Is it at all possible to have a file removed without showing the removal in git log?
22:46 timvishe` joined #git
22:47 bremner you sound like you need a lawyer, not tech support ;)
22:48 bremner you can rewrite history, but mostly, it's not worth the trouble if the history is already shared (pushed)
22:48 bremner alvarezp: see man git-filter-branch for all of the gory details
22:48 gitinfo alvarezp: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html
22:49 red82 joined #git
22:51 CBX joined #git
22:52 nevodka joined #git
22:52 CBX left #git
22:53 alvarezp ;-) no, it's not something I want to do, but we are seeing some kind of weird merge where a file "just" disappeared after the merge but there is no indication of git rm in git log --follow --all for the file. I tried replicating it in an isolated repo but no win. Tried some combinations but it always works ok.
22:55 xameada_ joined #git
22:56 str joined #git
22:57 cdown_ joined #git
22:57 lizard_ joined #git
22:59 anonymuse joined #git
23:00 mmlb joined #git
23:01 DarkFox_ joined #git
23:02 Rodya_ joined #git
23:04 ojacobson alvarezp: git's purely snapshot-based, so no
23:05 ojacobson if a file was present and is now absent, there will be at least one earliest commit in which the file is absent
23:05 ojacobson you could use git-bisect to find it
23:05 ojacobson (I say "at least one" because the file could have been removed on two or more parallel branches)
23:05 zacts joined #git
23:07 cstratak joined #git
23:08 timvishe` joined #git
23:09 mattp_ joined #git
23:09 alvarezp A-ha! If I do "git merge --no-commit otherbranch; git rm somefile; git commit -am Merge;" the resulting commit will not have somefile but no deletion record will show up on git log --follow --stat -- somefile
23:11 kadoban joined #git
23:13 justanotheruser joined #git
23:13 e14 joined #git
23:15 alvarezp Yup, if during a merge conflict resolution a file is deleted, git log --stat will not show any sign of this.
23:16 madewokherd joined #git
23:17 chigang joined #git
23:18 Rodya_ joined #git
23:19 SwiftMatt joined #git
23:20 lizard_ joined #git
23:21 jbitdrop joined #git
23:28 rgrinberg joined #git
23:29 timvishe` joined #git
23:29 dsdeiz joined #git
23:30 KevinMGranger joined #git
23:30 fedepad joined #git
23:31 mablae__ joined #git
23:33 drmt joined #git
23:34 FirewalkR joined #git
23:34 nckpz joined #git
23:34 dsdeiz_ joined #git
23:37 DolphinDream joined #git
23:38 red82 joined #git
23:42 nettoweb joined #git
23:48 hweaving joined #git
23:48 timvishe` joined #git
23:49 hweaving I'm getting mind-boggled about a merge issue that is hard to describe
23:50 hweaving I have a fully merged "develop" branch with N commits.  I have a fully merged "master" branch with N+1 commits.
23:50 hweaving If I do "git checkout master; git merge develop" I see a single merge conflict.  I am unsure why this is a conflict, since I simply want the newer version to be completely accepted.
23:50 hweaving Fine, I manually edit the file to use the new lines, and do "git commit".
23:51 hweaving At this point, my "master" branch suddenly has about 12 new commits duplicating old commits from "develop".  One possible difference is that many of the duplicate entries have my "Signed-off" line, when I didn't even intend to add that line.
23:51 hweaving Any ideas?
23:53 Lyqyd joined #git
23:53 hweaving To be thorough, I also did "git add CONFLICT_FILE" before the "git commit" command.
23:53 hweaving What could cause the "Signed-off" lines to be added?  I did not use "git commit -s".
23:56 jaguarmagenta joined #git
23:57 hweaving I cannot find any way to avoid these "Signed-off" commits :(

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