Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-10-11

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

All times shown according to UTC.

Time Nick Message
00:06 dsantiago joined #git
00:06 DolphinDream joined #git
00:08 timvishe` joined #git
00:08 hweaving Or, since no one has suggestions about this situation, how do I avoid a merge conflict to begin with?
00:09 hweaving If master and develop are identical besides one additional commit in develop, why should "git merge develop" identify a "both files changed" conflict?
00:11 hweaving One clue:  Apparently master and develop were NOT identical -- the hashes are different.  However, they contain identical descriptions for each commit.
00:12 hweaving Maybe doing "git merge --no-ff develop" previously caused this?
00:14 Guest29251 joined #git
00:16 hweaving Eventual answer:  My "master" contained the "Signed-off" commits because I used "git am" to apply a patch instead of "git merge develop".
00:16 hweaving Thanks!
00:17 AlexxNica joined #git
00:19 jnewt_ joined #git
00:20 nachoman joined #git
00:21 FirewalkR joined #git
00:21 zivester joined #git
00:22 timvishe` joined #git
00:22 Rodya_ joined #git
00:26 lolmac joined #git
00:26 nowhereman joined #git
00:29 mmmveggies joined #git
00:32 cqi joined #git
00:33 brokensyntax joined #git
00:33 timvishe` joined #git
00:38 ekinmur joined #git
00:38 mmmveggies joined #git
00:39 Rodya_ joined #git
00:39 _ix joined #git
00:45 ruby_ joined #git
00:46 timvishe` joined #git
00:46 rgrinberg joined #git
00:49 Rodya_ joined #git
00:55 timvisher joined #git
00:55 e14 joined #git
00:56 Aces_Charles joined #git
00:58 jaguarmagenta joined #git
00:58 timvisher joined #git
01:00 buzzLightBeer joined #git
01:02 Derperperd joined #git
01:02 anonymuse joined #git
01:06 durham joined #git
01:08 durham joined #git
01:10 watersoul_ joined #git
01:11 Limes_ joined #git
01:12 AlexxNica joined #git
01:14 aarobc joined #git
01:14 Vgr joined #git
01:14 Vgr joined #git
01:15 sunri5e joined #git
01:16 fstd joined #git
01:18 RedneckDonut joined #git
01:20 GandalfTheWizard joined #git
01:20 pks joined #git
01:21 d^sh joined #git
01:24 pikajude joined #git
01:29 _ix joined #git
01:30 arescorpio joined #git
01:31 fscala joined #git
01:31 GandalfTheWizard joined #git
01:41 Doginal joined #git
01:41 EvilPeng1 joined #git
01:42 Ahmed90 joined #git
01:46 netj 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
01:48 rchavik joined #git
01:52 liuzhen joined #git
01:52 timvisher joined #git
01:53 t0n1c joined #git
01:54 nettoweb joined #git
01:56 Vampire0 joined #git
01:57 Gsham joined #git
01:57 nivag joined #git
01:58 laumann joined #git
01:59 evanwang joined #git
01:59 cyphase joined #git
02:01 red82 joined #git
02:01 gambler joined #git
02:03 FirewalkR joined #git
02:06 Jameser joined #git
02:07 sleepee joined #git
02:11 cyphase joined #git
02:11 nettoweb1 joined #git
02:14 mizu_no_oto joined #git
02:16 harish joined #git
02:16 HoierM joined #git
02:20 chachasmooth joined #git
02:22 lagothrix joined #git
02:22 xameada joined #git
02:23 Electrometro joined #git
02:23 cyphase joined #git
02:23 xameada_ joined #git
02:24 flaguy48 left #git
02:25 Rewneyo joined #git
02:26 buzzLightBeer joined #git
02:31 cyphase joined #git
02:34 hexagoxel_ joined #git
02:35 dreiss joined #git
02:39 ochorocho__ joined #git
02:40 dodobrain joined #git
02:40 gfixler joined #git
02:41 cyphase joined #git
02:42 timvisher joined #git
02:43 drmt joined #git
02:50 dsantiago joined #git
02:52 dsdeiz_ left #git
02:54 malcon joined #git
02:59 Wulf joined #git
03:02 red82 joined #git
03:02 jbitdrop joined #git
03:03 flaguy48 joined #git
03:03 danslo joined #git
03:03 liuzhen joined #git
03:08 Kitty- joined #git
03:09 EvilPeng1 joined #git
03:15 Vortex34 joined #git
03:18 lizard_ joined #git
03:18 e14 joined #git
03:19 alex1a joined #git
03:20 raijin joined #git
03:21 Kitty- joined #git
03:21 raijin joined #git
03:24 mehola joined #git
03:27 nivag joined #git
03:28 eizua joined #git
03:30 svm_invictvs joined #git
03:31 timvisher joined #git
03:31 sangy joined #git
03:32 FirewalkR joined #git
03:32 evanwang joined #git
03:35 xall joined #git
03:39 robotroll joined #git
03:40 raijin joined #git
03:42 fedepad joined #git
03:44 nd joined #git
03:44 Electrometro joined #git
03:47 DarkFox_ joined #git
03:49 raijin joined #git
03:50 boombatower joined #git
03:51 bronson joined #git
03:51 a_thakur joined #git
03:55 anonymuse joined #git
03:56 Kitty- joined #git
03:56 jimi_ joined #git
04:01 aikuma joined #git
04:01 offtopik left #git
04:02 danslo joined #git
04:03 _ix joined #git
04:06 Andrew_K joined #git
04:06 mmmveggies joined #git
04:08 e14 joined #git
04:12 rscata joined #git
04:12 govg joined #git
04:13 Wulf joined #git
04:15 pR0Ps joined #git
04:17 firebalrog joined #git
04:18 xall joined #git
04:23 Zumochi joined #git
04:23 Zumochi joined #git
04:23 EgS joined #git
04:23 Neico joined #git
04:23 dilfridge joined #git
04:23 saustin1 joined #git
04:24 Nevik joined #git
04:24 hexagoxel joined #git
04:25 bond joined #git
04:25 tilpner joined #git
04:26 rscata joined #git
04:28 cbreak joined #git
04:28 Bombe joined #git
04:31 firebalrog left #git
04:31 timvisher joined #git
04:32 Muzer joined #git
04:34 eshlox joined #git
04:35 ewet joined #git
04:37 mniip joined #git
04:38 brokensyntax joined #git
04:39 D-Boy joined #git
04:42 Motoko joined #git
04:44 bq joined #git
04:47 Cabanossi joined #git
04:48 oskarkv joined #git
04:55 daey_ joined #git
04:56 dranve joined #git
04:56 anonymuse joined #git
04:59 freimatz joined #git
05:01 nilg joined #git
05:06 bmahe joined #git
05:08 timvisher joined #git
05:09 mmmveggies joined #git
05:09 netj joined #git
05:13 NeverDie joined #git
05:15 qt-x joined #git
05:17 justan0theruser joined #git
05:20 johnny56 joined #git
05:22 overlord_tm joined #git
05:25 cyan__ joined #git
05:27 bocaneri joined #git
05:28 jimi_ joined #git
05:30 zeroed joined #git
05:31 hexagoxel joined #git
05:32 a_thakur joined #git
05:36 Raging_Hog joined #git
05:39 snufft joined #git
05:41 ochorocho__ joined #git
05:42 _ix joined #git
05:43 ThomasLocke_ joined #git
05:44 a_thakur joined #git
05:45 a_thakur joined #git
05:47 xall joined #git
05:49 timvisher joined #git
05:53 nilg` joined #git
05:54 eka joined #git
06:05 Wayne_Shephard joined #git
06:06 ogny joined #git
06:06 liuzhen joined #git
06:09 JeroenT joined #git
06:10 chele joined #git
06:12 a_thakur joined #git
06:16 sea-gull_ joined #git
06:17 Jameser_ joined #git
06:18 LucaTM joined #git
06:18 fedepad joined #git
06:23 eb joined #git
06:23 Khisanth joined #git
06:26 roelmonnens joined #git
06:27 rchavik joined #git
06:29 PCatinean joined #git
06:30 dan2k3k4 joined #git
06:37 hexagoxel joined #git
06:37 rewt joined #git
06:38 cyan__ joined #git
06:38 star_prone joined #git
06:40 star_prone joined #git
06:42 timvisher joined #git
06:43 jnavila joined #git
06:44 mablae__ joined #git
06:44 anonymuse joined #git
06:45 corsibu joined #git
06:46 cqi joined #git
06:46 berndj joined #git
06:47 jknetl joined #git
06:48 maret joined #git
06:48 jave joined #git
06:49 brillenfux joined #git
06:49 nivag joined #git
06:50 Sceorem joined #git
06:53 brillenfux joined #git
06:53 Opertus joined #git
06:55 brokensyntax joined #git
06:56 vuoto joined #git
06:57 fub joined #git
06:59 fub Hi. I have a project where everything was rewritten from scratch in a different language. The old project is in Repository A, the new is in B. I would like to merge project B into A, so there is only one current project. What is the best way to do this?
06:59 fub I dont want to lose any commit which was made
06:59 SporkWitch joined #git
06:59 nivag joined #git
07:00 djb-irc joined #git
07:00 thiago joined #git
07:02 _ikke_ What do you want the end result to be?
07:02 jaequery joined #git
07:02 _ikke_ That project B starts where project A ended?
07:03 dunaeth joined #git
07:03 styler2go joined #git
07:04 gruetzkopf joined #git
07:05 forgot joined #git
07:05 ronskisika joined #git
07:05 abra0 joined #git
07:06 Vampire0 fub, if you want to merge both histories with a merge commit, add B as remote to A, fetch B and then merge B into A
07:07 krabat joined #git
07:07 Vampire0 fub, if you want A to start at the end of B, do the same, but do not merge, instead us a graft to make the last commit of B the parent of the first commit of A and to make it permanent use git rebase. But be aware that this will !rewrite all commits of A
07:07 gitinfo fub: 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
07:07 ToBeFree joined #git
07:08 brillenfux joined #git
07:09 cdown joined #git
07:10 gemie joined #git
07:10 elect joined #git
07:11 gemie hi, If I stand in branch Foo and want to merge branch Bar. Is bar = remote and Foo = local? right?
07:11 Rewneyo joined #git
07:12 SwiftMatt joined #git
07:13 fedepad joined #git
07:13 SporkWitch joined #git
07:15 _ikke_ right, git also calls it theirs and ours
07:15 _ikke_ where Foo is ours, and Bar is theirs
07:16 _ix joined #git
07:18 grimel_ joined #git
07:18 TomyWork joined #git
07:18 \du joined #git
07:18 \du joined #git
07:21 fub _ikke_, Vampire0: I dont want to rewrite history, every commit in Project A is before the commit in Project B (the do not overlap). I just want toe original repository now to only have the current files in it
07:21 ronskisika joined #git
07:21 _ikke_ fub: I don't know of a good way to do it without rewriting history
07:22 _ikke_ fub: things like git replace exist, but they rely on certain refs to exist, that are not pushed by default
07:22 timvisher joined #git
07:22 Alienpruts joined #git
07:22 fub _ikke_: isnt it possible to do something like: move everything in project A to a branch called "old", then move every commit from project B to project  A?
07:23 _ikke_ yes, that's possible
07:24 apotry joined #git
07:24 fub how? Project A: 1) Create branch 2) switch back to master branch 3) delete all? Project B: How to move everything to A?
07:24 eka joined #git
07:24 _ikke_ Easiest is to just fetch everything into Repo B
07:25 _ikke_ git remote add old path/to/A
07:25 danslo joined #git
07:25 _ikke_ git fetch old branch:refs/heads/old
07:27 Filiphe joined #git
07:30 freimatz joined #git
07:31 Balliad joined #git
07:32 nickthetait joined #git
07:32 freimatz Hello, if someone do: a commit, a push to remote, then we dont want to have the commit furthermore. Whats the right way to do this?
07:32 overlord_tm joined #git
07:33 SporkWitch joined #git
07:33 _ikke_ Why do you want them to commit it in the first place then?
07:33 vulture joined #git
07:35 freimatz i don't want it, it happend in past
07:35 freimatz maybe in future past
07:35 dc5ala joined #git
07:36 _ikke_ Ok, and don't you want the commit to exist, or do you want to undo the changes of the commit?
07:37 freimatz hm, i see no difference, lets asome this commit was the last commit
07:37 noidea joined #git
07:37 noidea joined #git
07:37 dreiss joined #git
07:38 danslo freimatz: git revert <commit hash> would be the easiest here
07:38 _ikke_ freimatz: The former requires rewriting history, the latter not (it would create a new commit)
07:38 _ikke_ freimatz: What danslo said is what I mean with the latter
07:38 nikitasius joined #git
07:38 nikitasius hi folks! can someone help me with my regexp..? #regexp is dead, i have no place to ask..
07:38 danslo (sorry, just joined the channel, don't see earlier messages)
07:39 gemie joined #git
07:39 nikitasius here is regexp and i need Blablabla https://gist.github.com/nikitasius/6316a02fab6e57078f2eb9bfcc1cf123 .
07:40 gemie hi, I am looking for a tool I used a long time ago. It's an extension/abstraction for Git. For example if I write git status I get a list and a number on the left so I can do like git add 1 etc. Can't remember the name, anyone know what I talk about?
07:40 _ikke_ gemie: git add --interactive
07:40 miguelfernandes joined #git
07:40 _ikke_ That's part of git
07:40 freimatz git revert sounds good. is there probability to have conflicts?
07:41 _ikke_ yes
07:41 gemie but I know it was an extension, there was more good stuff like git discard 1,2 etc
07:41 _ikke_ gemie: Ok, then I don't know
07:41 _ikke_ gemie: I recall something like a shell extension that does that
07:41 _ikke_ or maybe not a shell extension
07:41 ochorocho__ joined #git
07:42 gemie _ikke_: yes, a shell extension. correct
07:42 danslo freimatz: if it's the last commit, no. otherwise you can treat it like applying any other random changeset to a codebase that has changed before. maybe it conflicts, maybe it doesnt
07:43 freimatz hm, the other way, whats this? locally i can do a git reset
07:44 _ikke_ You can force push when you have reset it locally, if you know what the consequences are
07:44 bn0t joined #git
07:45 overlord_tm joined #git
07:46 kurkale6ka joined #git
07:47 lindenle joined #git
07:47 brokensyntax joined #git
07:48 freimatz i only have an vagoue imaginary
07:48 bocaneri joined #git
07:49 freimatz is it enough that each other do a pull on master?
07:49 fub _ikke_: problem is, this should all be published on github under project A
07:49 tchan joined #git
07:49 fub so moving everything to B is maybe not the best way
07:50 xall joined #git
07:51 jimi_ joined #git
07:51 mikecmpbll joined #git
07:52 _ikke_ It doensn't matter really
07:53 _ikke_ If you want to reuse the same branch name, you have to do a force push any way
07:55 SporkWitch joined #git
07:56 timvisher joined #git
07:56 mehola joined #git
07:56 fub _ikke_: no it can be a different branch name
07:56 ahmedelgabri joined #git
07:56 fub oh no, I want it in master
07:56 fub so yes, same branch name
07:56 watabou joined #git
07:57 finalbeta joined #git
07:57 hahuang61 joined #git
07:57 sweatsuit joined #git
07:57 _ikke_ right, so do the steps I told you, then updat the remote url for origin (or add a new one), and do a force push
07:57 fub _ikke_: but first create a new branch, then switch back to master and remove all files, right?
07:58 aikuma joined #git
08:00 _ikke_ You don't need to remove any files. You basically have two lines of history coexisting
08:01 __main__ joined #git
08:01 troyt joined #git
08:01 sarbs joined #git
08:02 fub okay so I'm in ProjectA. 1) git remote add new path/to/B 2) git fetch new branch:refs/heads/old -- then I have "master" (my old repo) and "new" (my new repo). How do I swap them now?
08:03 netj joined #git
08:03 marcogmonteiro joined #git
08:03 nuck joined #git
08:03 _ikke_ The steps I gave assumed you were in ProjectB
08:04 Kitty- joined #git
08:04 Guest37970 joined #git
08:05 ocbtec joined #git
08:05 redeemed joined #git
08:07 eka joined #git
08:09 cstratak joined #git
08:10 fub _ikke_: right
08:10 fub but it would be nice if I can do this from ProjectA
08:11 fub (because thats also the project on github where I want to push it to)
08:11 navidr joined #git
08:12 bongjovi joined #git
08:12 _ADN_ joined #git
08:12 Xethron joined #git
08:13 lmatteis joined #git
08:13 tvw joined #git
08:14 meLon joined #git
08:18 harish joined #git
08:19 flaviodesousa joined #git
08:20 luke_penn joined #git
08:21 chll_ joined #git
08:23 _ikke_ fub: Then you have to rename the master branch first
08:23 eka_ joined #git
08:24 a_thakur joined #git
08:25 noodle joined #git
08:26 JeroenT joined #git
08:26 fub _ikke_: so this: https://paste.pound-python.org/show/S5GfVrMMX5Xfea5QhzgP/ ?
08:28 baako joined #git
08:29 baako hi guys if i do a git reset --hard c0793e
08:29 baako will that rollback my local branch to that commit version?
08:30 moritz yes
08:30 baako moritz, just my local right?
08:30 elect joined #git
08:30 sarbs joined #git
08:30 mischat joined #git
08:31 baako moritz, thanks
08:31 Thorn__ joined #git
08:31 jnewt__ joined #git
08:32 t-mart_ joined #git
08:32 metabulation joined #git
08:32 Snugglebash joined #git
08:32 fstd_ joined #git
08:32 jstimm joined #git
08:32 moritz baako: yes; basically just "git push" modifies remote branches
08:32 TomyWork joined #git
08:32 anonymuse joined #git
08:32 smithbone_ joined #git
08:32 yaewa joined #git
08:33 jargan joined #git
08:33 moritz (and "git fetch" updates your local copy of the remote branches)
08:33 tristanp_ joined #git
08:34 milki_ joined #git
08:35 brokensyntax_ joined #git
08:35 Infamous- joined #git
08:35 timvisher joined #git
08:36 robogoat joined #git
08:36 govg_ joined #git
08:36 brewmaster joined #git
08:36 solidus-1iver joined #git
08:36 jbondhus_ joined #git
08:36 enleeten_ joined #git
08:36 ftoulemo1 joined #git
08:36 przemoc86 joined #git
08:37 Ratler_ joined #git
08:38 brendyyn joined #git
08:38 Jonuz joined #git
08:38 Ph__ joined #git
08:38 MerinoBailon joined #git
08:38 chimay joined #git
08:38 QwertyDragon_ joined #git
08:38 _ikke_ fub: Looks good
08:38 themayor_ joined #git
08:38 ThomasLocke joined #git
08:38 ThomasLocke joined #git
08:39 blahdodo_ joined #git
08:39 KevinMGranger_ joined #git
08:39 xer0x_ joined #git
08:39 Grum_ joined #git
08:39 parasite__ joined #git
08:40 fakenullie joined #git
08:40 tcurdt_ joined #git
08:40 tmk joined #git
08:40 ploues_cat joined #git
08:40 dvdyzag8 joined #git
08:40 gnarf joined #git
08:40 lep2 joined #git
08:40 bakingbread joined #git
08:40 mwleeds_ joined #git
08:40 ajpiano_ joined #git
08:40 bipolar_ joined #git
08:40 bauruine_ joined #git
08:40 jokajak joined #git
08:40 troyt_ joined #git
08:40 jrahmy- joined #git
08:40 cafuser0m joined #git
08:41 shakalaka_ joined #git
08:41 mycyber joined #git
08:41 pdgwien- joined #git
08:41 kake|| joined #git
08:41 nuck joined #git
08:41 Marbug joined #git
08:42 lohfu1 joined #git
08:42 ayjay_t joined #git
08:43 brokensyntax joined #git
08:43 clr joined #git
08:44 peepsalot joined #git
08:45 moestevens joined #git
08:45 Tsutsukakushi joined #git
08:45 stealthii joined #git
08:45 n-st joined #git
08:45 bdeferme joined #git
08:45 halbomin joined #git
08:45 wizonesolutions joined #git
08:45 Junior joined #git
08:46 hexagoxel joined #git
08:46 barq joined #git
08:46 PCatinean joined #git
08:46 Sleepy63 joined #git
08:46 elijah joined #git
08:47 mehdi___ joined #git
08:47 mgoodwin joined #git
08:47 dinoocch joined #git
08:47 angular_mike_ joined #git
08:47 brokensyntax joined #git
08:48 Shentino joined #git
08:48 mischat joined #git
08:49 harleypig joined #git
08:49 timwis joined #git
08:49 lindenle joined #git
08:49 blackjid joined #git
08:49 Iacobus joined #git
08:49 JackMc joined #git
08:49 ionelmc joined #git
08:49 m1dnight_ joined #git
08:49 Nothing4You joined #git
08:49 cory___ joined #git
08:49 RickyB98 joined #git
08:49 idmistir joined #git
08:49 deltab_ joined #git
08:49 bruce_lee joined #git
08:49 on1x joined #git
08:49 gniourf joined #git
08:49 ozymandias_ joined #git
08:49 xaa joined #git
08:49 xall joined #git
08:49 Muzer joined #git
08:49 Neico joined #git
08:49 Aces_Charles joined #git
08:49 chigang joined #git
08:49 mquin joined #git
08:49 Lor_ joined #git
08:49 aca20031 joined #git
08:49 ij joined #git
08:49 emptynine joined #git
08:49 wwalker_ joined #git
08:49 qws-user-1228 joined #git
08:49 dwmw2_gone joined #git
08:49 doppo joined #git
08:49 Zarthus joined #git
08:49 PaulCapestany joined #git
08:49 belak joined #git
08:49 sfan786_ joined #git
08:49 etcinit joined #git
08:49 CussBot joined #git
08:49 imdsm_ joined #git
08:49 bcardarella joined #git
08:49 dunk joined #git
08:49 bjoe2k4 joined #git
08:49 richardlitt joined #git
08:49 yena joined #git
08:49 kwmiebach__ joined #git
08:49 Tee_Pee joined #git
08:49 flupke joined #git
08:49 gyre007 joined #git
08:49 thecomedian joined #git
08:49 amaged__ joined #git
08:49 yoyoyo____ joined #git
08:49 epic joined #git
08:49 theresajayne joined #git
08:49 kanyeezy joined #git
08:49 addshore joined #git
08:49 karstensrage joined #git
08:49 MatthewAllan93 joined #git
08:49 vrishab joined #git
08:49 BinGOs joined #git
08:49 devhoag joined #git
08:49 r3c0d3x joined #git
08:49 murph joined #git
08:49 POQDavid joined #git
08:49 nsh- joined #git
08:49 devhost joined #git
08:49 gitinfo joined #git
08:49 duarteper joined #git
08:49 Caelum joined #git
08:49 i7c joined #git
08:49 Mikaela joined #git
08:49 n1ax joined #git
08:49 McKraken joined #git
08:49 swift1_ joined #git
08:49 xkr47_ joined #git
08:49 liberpater1 joined #git
08:49 nedbat joined #git
08:49 cschneid- joined #git
08:49 kikijiki[m] joined #git
08:49 Ineentho[m] joined #git
08:49 cm24 joined #git
08:49 [cust] joined #git
08:49 jor joined #git
08:49 quizzi joined #git
08:49 JelmerD joined #git
08:50 RickyB98 joined #git
08:50 cory___ joined #git
08:50 ionelmc joined #git
08:50 JackMc joined #git
08:50 Iacobus joined #git
08:50 timwis joined #git
08:50 harleypig joined #git
08:50 troyt joined #git
08:50 idmistir joined #git
08:50 RickyB98 joined #git
08:51 led_belly joined #git
08:51 DJ_Spies joined #git
08:51 cm24 joined #git
08:51 Lor_ joined #git
08:51 xall joined #git
08:51 tjbp joined #git
08:51 Serg_Penguin joined #git
08:51 xaa joined #git
08:52 dunk joined #git
08:52 cory___ joined #git
08:53 JackMc joined #git
08:53 addshore joined #git
08:53 Serg_Penguin hi ! I have a project with lots of old experimental branches. I want to have archives with project states up to every old branch, with only that branch in git, no others or master.
08:53 ionelmc joined #git
08:53 chigang joined #git
08:53 ojdo joined #git
08:54 Serg_Penguin to get to this, i make new directory, copy .git to it, check out the branch, but how can i safely disconnect remote master and remove master and all other branches ?
08:54 Iacobus joined #git
08:54 moritz git branch -D <branchname>
08:54 richardlitt joined #git
08:54 moritz git remote remove origin
08:54 XDS2010 joined #git
08:55 Serg_Penguin moritz, is it safe ?
08:55 moritz Serg_Penguin: what are you afraid of?
08:55 techntoke_ joined #git
08:56 Mindfart joined #git
08:56 moritz it deletes branches and a remote. If that's what you want, there's no problem with that :-)
08:56 moritz it won't cause your computer to spit out blue flames
08:56 belak joined #git
08:56 moritz at least, it's never done that for me :-)
08:57 osse what about white ones
08:57 timwis joined #git
08:57 harleypig joined #git
08:57 moritz never had that either
08:57 Serg_Penguin i don't want to damage repo on bitbucket, i only want to cook everey old unmerged branch into tar.gz and purge them from repo
08:57 moritz git isn't very sparkly :-)
08:57 gryffus joined #git
08:57 moritz Serg_Penguin: you can't damage the repo on bitbucket by running those commands locally
08:58 noodle joined #git
08:58 moritz Serg_Penguin: you can only damage it with commands such as 'git push'
08:58 Serg_Penguin i will never push from this truncated repo ;)
08:58 gryffus If i do "git reset --hard HEAD" will i lose local unpushed changes?
08:58 moritz gryffus: yes
09:00 _ikke_ You will loose uncomitted changes
09:00 _ikke_ Not unpushed changes that are comitted
09:02 btree joined #git
09:03 fub joined #git
09:03 muthu joined #git
09:04 btree joined #git
09:04 gryffus _ikke_: nice, thank you :)
09:06 fub _ikke_: sorry, had a short disconnect
09:06 _ikke_ fub: That plan you pasted looked good to me
09:06 Serg_Penguin after deleting branches, i have tons of empty lines in git space usage report - just size, no file name. How to get rid ?
09:07 fub _ikke_: great, thanks for your help
09:07 osse what is a git space usage report ?
09:07 Serg_Penguin osse, https://stubbisms.wordpress.com/2009/07/10/git-script-to-show-largest-pack-objects-and-trim-your-waist-line/
09:08 btree joined #git
09:08 btree left #git
09:08 selckin git reflog expire --expire=now --all; git gc --prune=now
09:10 Serg_Penguin selckin, wow, you rock ! nuked many megabytes !
09:10 nilg joined #git
09:10 ghostlight joined #git
09:11 muthu joined #git
09:11 Ratler joined #git
09:12 govg joined #git
09:12 frobnic joined #git
09:17 fub _ikke_: when following my "plan", I get this when trying the git push --delete origin master: ! [remote rejected] master (refusing to delete the current branch: refs/heads/master)
09:17 _ikke_ selckin: There is also --expire-unreachable, which is slightly less nuclear
09:17 _ikke_ fub: Right, I think you just need to force push in this case
09:18 _ikke_ fub: master is the 'default' branch, and you cannot delete that
09:18 selckin _ikke_: going nuclear is the point
09:18 fub _ikke_: so git push --delete --force origin master ?
09:18 _ikke_ fub: No
09:19 _ikke_ fub: no, not force delete
09:19 _ikke_ force push the new master branch
09:19 _ikke_ No need to delete it first
09:20 fub ok
09:21 fub _ikke_: when doing the "git remote add", can this also be a different directory?
09:21 _ikke_ sure, git can fetch from the local file system
09:22 fub I added it as remote, then git fetch, but got: fatal: Couldn't find remote ref branch; fatal: The remote end hung up unexpectedly
09:23 fub oh, I guess 'branch' is not a keyword, I have to replace it with the branch I want to fetch from, right
09:23 fub called git fetch master branch:refs/heads/old
09:24 _ikke_ right
09:26 fub _ikke_: fatal: Refusing to fetch into current branch refs/heads/old of non-bare repository
09:27 _ikke_ Yeah, that's because you already created old locally. I think you want to call that one master now, right?
09:28 a_thakur joined #git
09:31 Wind0r joined #git
09:35 interrobangd joined #git
09:37 Darcidride joined #git
09:39 Electrometro joined #git
09:39 maret joined #git
09:40 BadSnappa joined #git
09:41 nettoweb joined #git
09:44 achadwick joined #git
09:45 corsibu joined #git
09:45 drmt joined #git
09:46 endiruna joined #git
09:47 lamont joined #git
09:51 thiago joined #git
09:52 jimi_ joined #git
09:52 jimi_ joined #git
09:55 elementalest joined #git
09:55 elementalest joined #git
09:58 jimi_ joined #git
09:58 Wulf joined #git
10:00 fub _ikke_: sorry, got a call.
10:00 fub _ikke_: yeah I removed the master branch locally and now I want to fetch from ProjectB and create a new branch called master
10:03 nickthetait joined #git
10:04 fees joined #git
10:04 ieee joined #git
10:05 cdown joined #git
10:06 chimay joined #git
10:06 HelloAtMega left #git
10:09 HelloAtMega joined #git
10:12 brokensyntax joined #git
10:12 halbomin joined #git
10:12 timvisher joined #git
10:14 mmmveggies joined #git
10:15 ams__ joined #git
10:16 liuzhen joined #git
10:16 barteks2x joined #git
10:16 Snugglebash joined #git
10:17 manosnore joined #git
10:18 blahdodo joined #git
10:21 qws-user-1229 joined #git
10:24 _ikke_ so it would be git fetch <remote> master then
10:25 theskillwithin joined #git
10:26 Wayne_Shephard hi, can you help me, plz? i want a file to be in my repo, but i want only the initial version of this file, so that everyone could change it, but it won't change in repo. never. how could i do that?
10:27 _ikke_ Wayne_Shephard: This won't work well in git
10:27 cdown joined #git
10:27 _ikke_ it's better to commit a template under a different name
10:27 a_thakur joined #git
10:28 kiresp joined #git
10:28 Wayne_Shephard ex: .idea/workspace.xml i want it to hold some debug info, but not to be modified in repo every time i open my IDE
10:29 _ikke_ I believe idea makes a difference between local and shared settings
10:29 * osse has seen gitignores with '.idea' in them
10:29 osse not sure it's a good thing to version
10:30 ^7heo it's never a good thing to version.
10:30 ^7heo as a matter of fact
10:30 cbreak so it's a bad idea?
10:30 ^7heo you should redirect any metadata ever to /dev/null
10:30 ^7heo and why not, the content of your files too.
10:30 ^7heo /dev/null has the highest compression ratio ever.
10:30 ^7heo it is EXTREMELY efficient.
10:30 Wayne_Shephard if i ignore it from the start - those who download my repo  have to configure their ide manually..
10:30 ^7heo and moreover, it's webscale.
10:30 _ikke_ It's WEBSCALE
10:30 cbreak Wayne_Shephard: so?
10:30 ^7heo _ikke_: I got it first, sorry.
10:30 _ikke_ :-(
10:30 ^7heo :P
10:30 cbreak Wayne_Shephard: if you ship them, I have to configure xcode too
10:30 Wayne_Shephard assign dr_uart_3 = GPIO2[2];
10:31 Wayne_Shephard so not everyone can do it in my team(
10:31 Wayne_Shephard i have to come and do it for them
10:31 ^7heo Wayne_Shephard: make a script to configure the repo for their IDE
10:31 ^7heo Wayne_Shephard: and version THAT.
10:31 cbreak cmake ftw
10:31 ^7heo cmake sucks
10:31 ^7heo please stop advising cmake
10:31 cbreak it's the least crappy solution of all
10:32 cbreak there's nothing less bad.
10:32 ^7heo it's enough pain with the current amount of projects using t.
10:32 ^7heo s/t\./i&/
10:32 ^7heo cbreak: bare make is better, if used right.
10:32 cbreak no
10:32 cbreak bare make sucks
10:32 ^7heo yes.
10:32 cbreak it's obvious for everyone that ever used it
10:32 ^7heo But you have to know what is "using it right"
10:32 Wayne_Shephard k. i c. i started a kinda holywar)))
10:32 Wayne_Shephard thank you
10:32 corsibu joined #git
10:32 ^7heo cbreak: it's obvious to everyone who has ever used it... wrong.
10:33 ^7heo Wayne_Shephard: it's IRC, those things happen all the time.
10:33 kiresp joined #git
10:34 Wayne_Shephard i know)))
10:34 ppf joined #git
10:34 raijin joined #git
10:34 ppf is git backwards compatible?
10:35 ppf if i've created repos with git 1.8, can i upgrade to 2.7 or above and not break the repors?
10:35 osse Wayne_Shephard: if it's a file that your tool changes without any way of controlling it,then don't version it
10:35 osse ppf: no problem
10:35 ppf osse: cool
10:35 halbomin joined #git
10:35 cbreak ppf: so far, git has not changed the repository format
10:35 cbreak that doesn't mean they never will
10:36 Hink_ joined #git
10:36 cbreak sha1 has been an issue for ages :)
10:36 jbitdrop joined #git
10:36 Wayne_Shephard osse: thanks
10:36 rorro joined #git
10:37 osse Wayne_Shephard: i'd investigate what _ikke_ said.
10:37 tlaxkit joined #git
10:37 timvisher joined #git
10:38 blahdodo joined #git
10:39 HelloShitty joined #git
10:40 raijin joined #git
10:40 _ikke_ cbreak: If they change the hash function, they'll probably do so in a way that's backwards compattible
10:40 cbreak _ikke_: store both and live with less security / integrity with old clients?
10:41 cbreak and potential inconsistent data
10:41 nydel07 joined #git
10:41 _ikke_ It often comes up in the mailing list
10:41 cbreak file formats with accidental redundancy are problematic :(
10:41 cbreak it makes the format more complex and more fragile
10:42 cbreak and unless git is strict, and checks both, it'd be an entry point for chimera file attacks
10:43 nowhereman joined #git
10:43 laumann joined #git
10:43 selckin why wouldn't it check both?
10:43 Murii joined #git
10:45 raijin joined #git
10:45 ppf left #git
10:47 ctmarco joined #git
10:47 aspiers joined #git
10:52 mischat_ joined #git
10:54 spNk joined #git
10:54 raijin joined #git
10:56 mizu_no_oto joined #git
10:57 Seveas cbreak: there have been minor changes, like HEAD no longer being a symlink but a symref
10:58 _ikke_ since 1.8?
10:58 _ikke_ That's older, right?
10:58 Seveas definitely not since 1.8
10:58 Seveas I was more responding to "so far, ..."
10:58 _ikke_ right
10:59 cbreak Seveas: before 1 doen't count :)
10:59 Seveas fair
10:59 Seveas now I'm wondering when packfiles and packed refs were introduced :)
10:59 JeroenT joined #git
11:00 Seveas meh, packfiles definitely very early
11:01 zyxep joined #git
11:02 raijin joined #git
11:02 ki0 joined #git
11:05 roelmonn_ joined #git
11:06 roelmon__ joined #git
11:09 timvisher joined #git
11:10 raijin joined #git
11:11 Samual joined #git
11:12 jgrasser joined #git
11:13 corsibu joined #git
11:14 dan2k3k4 joined #git
11:15 thebloggu joined #git
11:15 metrixx how can i convert "20161011" string date to BigDecimal?
11:15 metrixx i want it in same format but with BigDecimal type
11:15 _ikke_ metrixx: I don't think this is a #git question, right? ;-)
11:16 metrixx ohhhhh
11:16 metrixx sorry wrong channel
11:16 _ikke_ np
11:17 liuzhen joined #git
11:17 ralu joined #git
11:17 re1 joined #git
11:17 eka joined #git
11:18 Filiphe joined #git
11:18 jgrasser joined #git
11:18 joco42 joined #git
11:21 pijiu joined #git
11:21 derphilipp_ joined #git
11:22 pijiu joined #git
11:22 derphilipp_ How is it possible that a merged branch is shown when I try `git branch —no-merged`? It is also shown (correctly) when I look for `git branch —merged master`;
11:23 _ikke_ derphilipp_: Are you currently on branch master?
11:23 pijiu joined #git
11:23 jgrasser joined #git
11:24 derphilipp_ _ikke_: No, if I switch to it, it is displayed correctly
11:24 _ikke_ right, it shows what has(n't) been merged in the current branch
11:24 _ikke_ So the branch is not merged in the branch that you had checked out, but has been merged in master
11:24 gemie joined #git
11:24 derphilipp_ _ikke_: aaah, the _current_ branch, I thought it was ‘has not been merged at all’
11:25 _ikke_ No, it's always relative to some other branch
11:25 pijiu joined #git
11:25 _ikke_ derphilipp_: as a matter of fact, all it does is test for reachability
11:25 derphilipp_ _ikke_: is it possible to get what branches has not been merged at all?
11:25 Guest29223 joined #git
11:25 derphilipp_ _ikke_: ah, I see
11:25 _ikke_ so there might not even be a merge commit in between
11:26 _ikke_ derphilipp_: That's not easy. Basically git branch --no-merge <all_other_branches>
11:26 pijiu joined #git
11:26 derphilipp_ _ikke_: I see, thanks for the hint!
11:26 gemie I am looking for a shell extension for Git. I don't remember the name but for instance if you write gs (git status) you get a list of files but with an associated number so that you can do ga 1,2 to add file 1 and 2. Anyone know an extension like that? I have a blurred memory that it has something with "scm" something in it's name but I can't find it, have been googling for some time today.
11:26 mischat joined #git
11:26 _ikke_ s/no-merge/no/merged
11:27 fahadash joined #git
11:27 derphilipp_ gemie: https://github.com/scmbreeze/scm_breeze ?
11:28 gemie derphilipp_: yes! thank you, I love you!
11:28 derphilipp_ gemie:  you are welcome :) spread the git love
11:28 _ikke_ Nice!
11:28 Kitty- joined #git
11:30 ojdo joined #git
11:33 jaguarmagenta joined #git
11:36 nuck joined #git
11:36 derphilipp_ left #git
11:37 _Vi joined #git
11:38 timvisher joined #git
11:40 geoid joined #git
11:41 alex1a_ joined #git
11:41 johnmilton joined #git
11:42 jaafar joined #git
11:43 Murii joined #git
11:44 nikivi joined #git
11:45 sdothum joined #git
11:46 corsibu joined #git
11:48 a_thakur joined #git
11:50 thiago joined #git
11:50 JeroenT joined #git
11:51 va joined #git
11:52 va Jenkins seems to be using ttys by default for whatever it executes; what setting in git do I need to change so that git always operates as if it had no tty? (like, --color=never and --no-pager for everything)
11:53 osse va: git ... | cat
11:54 va yeah, I wanted to avoid adding that for the many commands that I have
11:54 Murii joined #git
11:54 osse va: { all the commands here; with
11:54 osse new lines and
11:54 osse stuff
11:54 _ikke_ git config core.pager cat; git config color.ui false
11:54 osse } | cat
11:54 va cheers
11:55 _ikke_ Note that other color.* variables override this one, so make sure they're disabled too
11:58 Cavallari joined #git
12:00 theshitisay joined #git
12:02 rpg_ joined #git
12:02 timvisher joined #git
12:03 UKn0Me joined #git
12:03 synthroid joined #git
12:03 mischat_ joined #git
12:05 jgrasser joined #git
12:08 Guest43571 joined #git
12:08 Guest43571 left #git
12:08 kiresp joined #git
12:08 freimatz joined #git
12:09 talin joined #git
12:09 talin hello. how can i diff a file between two branches?
12:10 Wulf talin: git diff b0:file b1:file
12:11 moei joined #git
12:11 inflames joined #git
12:13 Murii joined #git
12:14 talin Wulf: hello seawulf!
12:14 bruce_lee joined #git
12:14 Eryn_1983_FL joined #git
12:14 _ikke_ talin: If it's the same file, then it's just git diff branch1 branch2 file
12:14 geoid_ joined #git
12:15 talin thank you
12:16 mmmveggies joined #git
12:17 jgrasser joined #git
12:19 Kitty- joined #git
12:19 irco joined #git
12:20 saustin1 joined #git
12:21 jeffreylevesque joined #git
12:24 phanimahesh joined #git
12:24 nettoweb1 joined #git
12:25 rgrinberg joined #git
12:25 tlaxkit1 joined #git
12:27 roflincopter joined #git
12:27 _dunaeth joined #git
12:28 Guest29223 joined #git
12:28 Electrometro joined #git
12:28 timvisher joined #git
12:29 jottr joined #git
12:31 eb joined #git
12:33 cdown joined #git
12:34 mischat joined #git
12:37 styler2go joined #git
12:42 saustin1 joined #git
12:42 WayToDoor joined #git
12:43 jokajak joined #git
12:43 DeltaHeavy joined #git
12:44 saustin1 joined #git
12:44 zivester joined #git
12:47 JeroenT joined #git
12:48 timvisher joined #git
12:51 lindenle joined #git
12:51 Jetchisel joined #git
12:51 jottr joined #git
12:54 mycyber joined #git
12:55 manosnore joined #git
12:55 Darcidride joined #git
12:55 canuk joined #git
12:56 N0v4 joined #git
12:56 manosnore joined #git
12:57 fmcgeough joined #git
12:58 manosnore joined #git
12:58 DieguezZ joined #git
12:59 jgrasser joined #git
12:59 manosnore joined #git
13:00 faux___ joined #git
13:01 manosnore joined #git
13:01 porsche944 joined #git
13:02 mischat joined #git
13:02 dances-with-cats joined #git
13:03 john_doe_jr joined #git
13:03 john_doe_jr I'm getting an "out of memory" error when using git…does anyone know why I would be getting this error?
13:03 _ikke_ What operation?
13:03 john_doe_jr The size of my repo is "498M"
13:03 _ikke_ pack operations will require memory
13:04 john_doe_jr _ikke_: I just wanted to put the code under version control..that was the operation
13:05 tobiasvl what git command did you run
13:05 john_doe_jr tobiasvl: git add -A
13:06 _ikke_ How much memory does your machine have?
13:07 _ix joined #git
13:07 rpg joined #git
13:07 timvisher joined #git
13:09 ash_workz joined #git
13:09 Noldorin joined #git
13:11 hariel joined #git
13:11 CheckDavid joined #git
13:12 ilmari joined #git
13:12 john_doe_jr _ikke_: good question..it's a shared server @ 1and1 so I don't really know
13:13 Nd-47-M joined #git
13:13 barteks2x joined #git
13:13 _ikke_ free -m ?
13:13 _ikke_ But I guess not too much
13:13 _ikke_ Do you have large files in the repository?
13:14 john_doe_jr _ikke_: 807
13:14 _ikke_ about 1G then
13:14 john_doe_jr _ikke_: I see…I'm going to put the larger files in the .gitignore and try that
13:16 CEnnis91 joined #git
13:16 ExoUNX joined #git
13:16 Jameser joined #git
13:16 flaguy48 left #git
13:18 DivineHeir joined #git
13:18 john_doe_jr _ikke_: how much memory do u need to put a 1/2 GB repo under version control by the way?
13:19 _ikke_ It depends on how large the files are. If you have lots of commits with small files, not too much, but if you add large files, memory usage increases a lot
13:20 _ikke_ There are settings to alleviate that, but that's more for pack-file created
13:21 chele i wanna clone a svn repo into a local git. i use the command git svn clone -s svn://..../project. The problem is that 'project' is not used as real source but as a root project folder from which all other projects derive. Therefore there is no trunk or branches skeletons inside it. From 'project' hang all other projects which do have trunk branches skeleton. As a consequence I do not get files within 'project' folder in my git which i n
13:21 chele eed. how do i get my files from 'project' or root project folder into local git?
13:22 howdoi joined #git
13:22 BrianBlaze420 joined #git
13:22 e14 joined #git
13:23 Rodya_ joined #git
13:24 sangy joined #git
13:25 thiago joined #git
13:26 timvisher joined #git
13:27 a_thakur joined #git
13:28 nettoweb joined #git
13:28 a_thakur joined #git
13:30 raijin joined #git
13:33 GodGinrai joined #git
13:33 snowkidind joined #git
13:34 jaguarmagenta joined #git
13:34 quickbrownfox joined #git
13:35 gweodoo_ joined #git
13:35 gweodoo joined #git
13:36 saml joined #git
13:36 cbreak chele: git svn clone the project folder
13:37 john_doe_jr _ikke_: what are the common causes of a pre-commit hook not running
13:37 JeroenT_ joined #git
13:37 Murii joined #git
13:37 Gsham joined #git
13:38 ^7heo john_doe_jr: missing execution permission.
13:39 gweodoo_ left #git
13:39 gweodoo_ joined #git
13:40 joki- joined #git
13:40 timvisher joined #git
13:42 meuser joined #git
13:43 mouser_ joined #git
13:44 jaafar joined #git
13:44 guampa joined #git
13:44 john_doe_jr ^7heo: that is exactly what the problem was
13:45 JeroenT joined #git
13:45 buffal0 joined #git
13:46 anonymuse joined #git
13:46 mouser_ Good Morning
13:46 mouser_ I am working on a production issue
13:46 mouser_ and I desperately need to be able to determine what the version was prior to the latest oull
13:47 mouser_ if anyone can help, I would be very grateful
13:47 saustin1 joined #git
13:47 cbreak joined #git
13:48 john_doe_jr mouser_: what does "latest oull" mean?
13:48 mouser_ i am sorry, I mean "pull"
13:48 Murii joined #git
13:48 mouser_ in other words, what version was there before i pulled the latest release?
13:49 anonymuse joined #git
13:49 red82 joined #git
13:50 NeverDie joined #git
13:50 lmat On github, there is a pull request with conflicts. I would like to close it and open a new pull request rebased so that it doesn't have conflicts.
13:51 ^7heo john_doe_jr: you're welcome :)
13:51 lmat Normally what I do to do that is   git clone -t <pull request branch> <other github repo>
13:51 K7 joined #git
13:51 lmat then git checkout <other github repo>/<pull request branch>;  git rebase develop;
13:51 lmat The problem is that I don't have permission to see this <other github repo>!
13:51 lmat I can see the commits on github...how do I download them!?
13:52 joco42 how can i overwrite my current files from the head ?
13:52 joco42 something with force?
13:52 _ikke_ "git checkout ."
13:52 al-damiri joined #git
13:52 joco42 or reset?
13:52 joco42 checkout ?
13:53 joco42 that's it ?
13:53 lindenle joined #git
13:53 joco42 or maybe this ? https://git-scm.com/book/en/v1/Git-Tools-Stashing
13:54 nilg joined #git
13:54 treehug88 joined #git
13:54 lmat joco42: Lots of ways. Depends on exactly what you want to do. git checkout or git reset will replace any of your *modified* files with HEAD
13:55 mischat joined #git
13:55 lmat joco42: If you want to remove *untracked* files (to match HEAD), use git reset. Sometimes I want to get back to a pristine working copy. In that case, I use
13:55 ekinmur joined #git
13:55 lmat joco42: git reset --hard HEAD; git clean -fdx .; (this clean is run from the root of the repository.)
13:55 joco42 lmat, i made some refactoring and then i messed up something but dont know what.... now i want to temporariliy go back to HEAD and see what changed... what did i mess up :)
13:55 faux___ joined #git
13:56 faux___ joined #git
13:56 lmat joco42: If temporarily (you want to get *back* to your current state), you need to create a commit to mark your current spot, then go back.
13:56 roelmonnens joined #git
13:56 joco42 ahh
13:56 joco42 i see
13:56 raijin joined #git
13:56 joco42 nice
13:56 lmat joco42: This can be done very easily with git stash (which quietly creates a "stash" commit), or explicitly   git commit -a .; (or something like that) then git checkout HEAD^;
13:56 joco42 nice
13:56 joco42 this makes a lot of sense
13:56 e14 joined #git
13:56 lmat joco42: Glad you think so :-)
13:57 lmat joco42: Some resist the force.
13:57 joco42 :)
13:57 joco42 ohh
13:57 lmat joco42: However resist the force, you must not.
13:57 joco42 thanks lmat , now i am off to a job interview :)
13:57 joco42 starts in 3 min...
13:58 Xethron joined #git
13:58 relipse joined #git
13:58 GodGinrai joco42: good luck
13:59 spNk joined #git
13:59 Murii joined #git
13:59 bremner in the immortal words of Jimi Hendrix, "don't you be late"
14:00 mischat_ joined #git
14:00 GodGinrai lol
14:00 phaleth joined #git
14:00 timvisher joined #git
14:02 mquin joined #git
14:03 buffal0 joined #git
14:04 raijin joined #git
14:04 AaronMT joined #git
14:06 malcon joined #git
14:07 PCatinean joined #git
14:08 thiago joined #git
14:08 aikuma_ joined #git
14:09 corsibu joined #git
14:09 thiago_ joined #git
14:10 nielsv joined #git
14:10 Murii joined #git
14:10 star_prone joined #git
14:11 nielsv left #git
14:11 Zamyatin joined #git
14:12 buffal0 joined #git
14:12 Gsham joined #git
14:15 raijin joined #git
14:15 irco joined #git
14:15 patarr joined #git
14:16 mmmveggies joined #git
14:17 patarr joined #git
14:18 cdg joined #git
14:19 timvisher joined #git
14:22 cyan__ joined #git
14:23 Murii joined #git
14:25 tlaxkit1 joined #git
14:25 perlpilot joined #git
14:26 oxsyn joined #git
14:30 john_doe_jr How do I clone into a non-empty folder ?
14:30 _ikke_ git refuses to do that
14:30 raijin joined #git
14:30 thiago john_doe_jr: make it empty first
14:30 tlaxkit joined #git
14:31 john_doe_jr thiago: it's a snippet so it's a file that I'm cloning…actually a git hook "pre-commit" file
14:31 danslo joined #git
14:31 osse cd folder; git init; git remote add origin ...; git fetch origin; git checkout -b master origin/master; somethignsomething HEAD
14:31 canuk joined #git
14:32 thiago john_doe_jr: you don't clone files
14:32 thiago john_doe_jr: you clone entire repositories
14:32 IrishPrime joined #git
14:32 GodGinrai john_doe_jr: clone the repo and copy the file.
14:32 john_doe_jr thiago: I get it but it's a snippet on bitbucket that I'm cloning which is a specific file
14:32 john_doe_jr GodGinrai: so clone it to /tmp
14:32 GodGinrai sure
14:32 john_doe_jr and then copy it to the right location
14:33 thiago john_doe_jr: don't care
14:33 thiago john_doe_jr: cloning is always of an entire repository and requires an empty dir
14:33 timvisher joined #git
14:37 dendazen joined #git
14:37 raijin joined #git
14:39 ilmari joined #git
14:39 shpew12 joined #git
14:40 john_doe_jr thiago: thanks for your help…that is exactly what I did
14:40 tlaxkit joined #git
14:40 tcorneli joined #git
14:41 crayon joined #git
14:42 basiclaser joined #git
14:42 Murii joined #git
14:44 Murii joined #git
14:44 Kuukunen joined #git
14:45 leprechau joined #git
14:46 timvisher joined #git
14:47 marijnfs joined #git
14:47 Gsham joined #git
14:48 Hedgework joined #git
14:48 Derperperd joined #git
14:49 patarr joined #git
14:49 Murii joined #git
14:49 raijin joined #git
14:49 goircbot joined #git
14:51 goircbot joined #git
14:51 madewokherd joined #git
14:52 zivester joined #git
14:53 cyan__ joined #git
14:55 raijin joined #git
14:56 Murii joined #git
14:56 digidog joined #git
14:56 goircbot joined #git
14:56 Jameser joined #git
14:58 rbern joined #git
15:00 timvisher joined #git
15:00 Electrometro joined #git
15:01 eivarv joined #git
15:01 sammyo joined #git
15:02 Murii joined #git
15:03 ejb joined #git
15:05 mtpi joined #git
15:05 courrier joined #git
15:06 linelevel joined #git
15:06 crose joined #git
15:07 Murii joined #git
15:07 thebope_ joined #git
15:07 linelevel Hi, is there a command that I can use to move a local ref (like a branch name) to point at an arbitrary commit, but without interrupting my current working tree by checkout out that branch?
15:08 roelmonn_ joined #git
15:08 linelevel I'd like something equivalent to: `git checkout other_branch; git reset <commit_hash>` but without having to actually switch branches.
15:08 john_doe_jr what is the size that a git repo can be before I have trouble?
15:08 osse linelevel: git branch -f
15:10 newuser42 joined #git
15:10 Mixologic joined #git
15:11 thiago linelevel: git push . commitid:branchname
15:11 thiago if you're going backwards, add -f
15:12 linelevel osse: thiago: Thanks, both of those work. I wasn't aware that I could use `.` in place of a remote name in `git push`.
15:12 thiago the plumbing command is git update-ref
15:12 timvisher joined #git
15:12 thiago yes, . is the current directory
15:12 thiago that'st he current repository
15:12 osse the push . is nice because it denies non-fast-forwards and such
15:13 osse useful for updating local branches after a fetch
15:13 linelevel osse: Yeah, but my use case requires `-f` in both cases anyway :)
15:13 linelevel thiago: So `update-ref` seems like exactly what I wanted. Any reason for me to avoid it or prefer the `git push -f` option?
15:14 isysd joined #git
15:14 goircbot joined #git
15:14 osse update-ref requires the full name. you cannot omit refs/heads
15:14 thiago linelevel: I'd use git push when I wanted a fast-forwad
15:14 Rewneyo joined #git
15:14 thiago between push -f and update-ref, they're the same
15:15 linelevel thiago: Got it.
15:15 thiago I don't know if update-ref updates the reflog too by default. It might. I know it does with the -m option.
15:15 thiago push always does
15:15 thiago push also runs the regular hooks
15:15 linelevel Thanks.
15:15 thiago and push will refuse to update the checked out branch
15:15 mischat joined #git
15:16 thiago in general, push is more secure
15:16 thiago as in, foolproof
15:16 SwiftMatt joined #git
15:17 linelevel Great, I'll prefer that.
15:17 marc_v92 joined #git
15:17 InfoTest joined #git
15:17 xaa joined #git
15:17 dunaeth joined #git
15:19 kpease joined #git
15:20 goircbot joined #git
15:20 lindenle joined #git
15:21 goircbot joined #git
15:21 shinnya joined #git
15:24 slayerjain joined #git
15:25 patarr joined #git
15:25 Jellyg00se joined #git
15:25 john_doe_jr left #git
15:25 timvisher joined #git
15:27 Mavericks joined #git
15:27 ejb joined #git
15:28 Mavericks how do I specify range of lines for git checkout ?
15:28 Mixologic joined #git
15:28 goircbot joined #git
15:30 digidog joined #git
15:30 jbitdrop joined #git
15:30 GodGinrai Mavericks: a range of lines?  like... in files?
15:30 ocbtec joined #git
15:31 Mavericks let us say formatting in text editor changed a few lines in a file
15:31 ochorocho__ joined #git
15:31 Mavericks it was accidental and need to get rid of changes in those specific lines
15:32 GodGinrai Mavericks: you already made changes afterwards?
15:32 Mavericks is there something like git checkout <filename> <<start line no><end line no>>
15:32 Mavericks or a another better command
15:32 Mavericks GodGinrai: no changes
15:32 red82 joined #git
15:32 cyan__ joined #git
15:33 Mavericks GodGinrai: i mean no manual changesd
15:33 GodGinrai Mavericks: then can't you use the undo function of your editor?
15:33 King_Hual joined #git
15:33 King_Hual joined #git
15:34 Mavericks GodGinrai: i have tried that no use. blue screen error doesn't let me go back far enough
15:34 faux___ joined #git
15:34 zacts joined #git
15:34 xshuiniu joined #git
15:34 GodGinrai Mavericks: blue screen? o.o
15:34 GodGinrai Mavericks: like, your computer crashed?
15:35 jaguarmagenta joined #git
15:35 davidosomething joined #git
15:35 Murii joined #git
15:35 Mavericks GodGinrai: yes and then i reopened file in text editor
15:35 yqt joined #git
15:36 GodGinrai Mavericks: well, it appears that there is a -p, or --patch, flag for `git checkout`, so you can probably use that.
15:36 fub joined #git
15:36 Mavericks GodGinrai: ok will try that
15:36 GodGinrai Mavericks: I'd turn off the autoformatting in your editor to prevent this happening in the future.
15:37 Mavericks GodGinrai: will do that
15:37 Rodya_ joined #git
15:37 osse Mavericks: try git checkout -p -- file then answer y or n as appropriate
15:38 diogenese joined #git
15:38 timvisher joined #git
15:38 ejb joined #git
15:38 Mixologic joined #git
15:39 b1101 joined #git
15:40 b1101 hello. What's a good way to  reference an older commit in a new one?
15:40 shinnya joined #git
15:40 ilmari 'commit XXXXXXX'
15:40 thiago b1101: do you mean, in the commit message or in the files?
15:40 thiago the SHA-1 of that commit, usually
15:41 GodGinrai I'd use the shorthash
15:41 ilmari things like gitweb and github will make it a link
15:41 marc_v92 joined #git
15:41 thiago I'd use more than the shortest
15:41 thiago use as many as you can
15:42 marcogmonteiro joined #git
15:42 b1101 The commit message should be almost identical as the same bug was affecting both changes
15:42 b1101 so it's almost like a continuation of fixing the same bug
15:42 raijin joined #git
15:43 Ardethian\work joined #git
15:43 Mavericks GodGinrai: osse: thank you both. it worked
15:44 Ardethian\work joined #git
15:45 cdown joined #git
15:45 sshine joined #git
15:45 b1101 ilmari: so do I just paste the sha-1 in the commit message, and it will link it ?
15:45 mischat joined #git
15:45 ilmari b1101: yes
15:45 b1101 nice. Thank You
15:46 ilmari for example http://perl5.git.perl.org/perl.git/commit/11a12be6c2df2fcd9601f9fde8c7df8c7538c956
15:46 mangolisk joined #git
15:46 _ikke_ git uses "hash (subject, date)
15:46 _ikke_ "
15:46 _ikke_ the git project
15:46 a_thakur joined #git
15:46 ilmari github seems to shorten it for display: https://github.com/Perl/perl5/commit/11a12be6c2df2fcd9601f9fde8c7df8c7538c956
15:47 synthroid joined #git
15:47 _ikke_ note that a shortlog is not going to be unambiguous for ever
15:48 sshine with git-svn on windows here, I've got a file that is modified but won't reset because of CRLF/LF issues. I've got core.autocrlf = false. what else do I need to avoid its wicked magic?
15:48 _ikke_ linux already has problems with short hashes around 12 characters
15:48 _ikke_ sshine: is there a .gitattributes file somewhere?
15:49 qpdb joined #git
15:49 GodGinrai ilmari: think that link's got enough perls and gits? XD
15:49 sshine _ikke_, oh wow, yes. one was checked in, and I didn't do it. must've been a remnant from a module import.
15:49 ilmari GodGinrai: not quite
15:50 timvisher joined #git
15:50 _Vi joined #git
15:50 nettoweb joined #git
15:51 sshine hehe. co-worker imports git project into SVN, forgets to remove .gitattributes file, I get affected. :P
15:51 sshine _ikke_, thanks!
15:51 va that'll teach you not to import :p
15:51 GodGinrai sshine: wait
15:51 raijin joined #git
15:52 GodGinrai sshine: you are using git svn to pull a svn repo that was originally a git repo?
15:52 Mavericks osse: GodGinrai: is there a way to undo last git checkout -p -- file ?
15:52 GodGinrai Mavericks: afaik, you can only undo it to the last committed state.
15:52 nilg joined #git
15:52 GodGinrai Mavericks: unless you have a stash laying around
15:53 sshine GodGinrai, no ;) I use git-svn because my co-workers use SVN. my co-worker imported this 3rd party library into the codebase that originates from GitHub and has a bunch of .git* files lying around. his SVN doesn't care. my git does.
15:53 GodGinrai sshine: I see
15:54 Derperperd joined #git
15:54 phanimahesh Are there any legitimate reasons to use svn? Not trying to start a flame war, just.. I can not imagine not having decentralised version control.
15:54 Mavericks GodGinrai: osse: never mind the undo in text editor worked
15:55 GodGinrai Mavericks: yay
15:55 sshine GodGinrai, the .gitignores are pretty good - that way I don't have to remake them. but the .gitattributes assume the wrong things.
15:55 dreiss joined #git
15:55 GodGinrai phanimahesh: Most of them revolve around using tools that have built-in SVN integration
15:55 sshine phanimahesh, I have read some comparison websites that argue that SVN works better for huge repos.
15:56 phanimahesh Oh. okay.
15:56 phanimahesh old tools and integrations are tricky to work around.
15:56 Electrometro joined #git
15:56 raijin joined #git
15:57 sshine phanimahesh, git-svn certainly is a bit quirky at a ~7G repo with 20k commits. I've never worked with a repo this size in pure git, so I don't know.
15:57 muthu joined #git
15:57 Oebele joined #git
15:57 phanimahesh but regarding large repos, mercurial can be taught a few tricks, like facebook did sometime ago I guess.
15:57 _ikke_ And perceived easier to use for people not versed in version control
15:57 phanimahesh 7G of source code? o_o
15:57 jaequery joined #git
15:58 phanimahesh The easier to use argument is not my favorite. partly because I can see it, and I still love the way git does it.
15:59 sshine phanimahesh, for someone who isn't trying to start a flamewar, you're surely arguing a lot. :P
15:59 _ikke_ sshine: It's an honest argument
15:59 madewokherd joined #git
15:59 phanimahesh hahahah. I do that a lot.
15:59 phanimahesh And well, I haven't been in a flamewar in ages, so... :P
16:00 a_thakur joined #git
16:00 BPL joined #git
16:00 _ikke_ People tend to use one SVN repository with lots of projects in there
16:00 BPL Hi everyone
16:01 Rodya_ joined #git
16:01 _ikke_ That's not working well in git (ie, you cannot just get one subdir)
16:01 phanimahesh BPL: Hi!
16:01 phanimahesh _ikke_: Can you get just one dir in svn?
16:01 Mekkis joined #git
16:01 phanimahesh I had to use svn once in college, and cheated by using git-svn.
16:01 _ikke_ yes, you can clone subdirs
16:02 phanimahesh Cool. I wished git could do that, at times.
16:02 BPL I'm trying to understand how to rebase interactively and maybe you guys can help me. For instance, let's say I've already pushed to a cmake's new branch and now I got a new commit I want to squash, what are the typical steps you need to do? First commit and then rebase interactively?
16:02 _ikke_ Which means you get just one dir checked out
16:02 GodGinrai sshine: an "argument" does not necessarily imply heated feelings
16:02 _ikke_ because git commits snapshot the entire worktree, it's not possible to do operations only on subdirs
16:02 phanimahesh And a fuse fs backed by git so that files can be fetched lazily on demand. But git doen't work that way, exactly. It'll be very tricky.
16:03 timvisher joined #git
16:04 kyan joined #git
16:04 phanimahesh BPL: little drawings help. I'm bad with imagination. Can you draw what you did on something like awwapp.com?
16:04 Kitty- joined #git
16:04 Limes_ joined #git
16:04 phanimahesh Never rewrite any commits you made public.
16:05 _ikke_ BPL: Is it like a fix for a certain commit?
16:05 relipse Something is going on, every time i switch branches, and i have my editor open (sublime) it does funky things, i tried doing git rebase master and it said cannot rebase you have unstaged changes and now git status says i deleted a file
16:05 _ikke_ BPL: look at git commit --fixup + rebase --autosquash
16:05 relipse how do i reset the current branch back to what it was
16:05 _ikke_ relipse: deja-vu? :P
16:05 a_thakur joined #git
16:05 phanimahesh relipse: git rebase --abort?
16:05 a_thakur joined #git
16:06 BPL _ikke_: Yeah, I had pushed a new feature and then the experienced main dev suggested me to change some methods, I've change the code and I want to push again, I thought it'd be nice to keep the repo as clean as possible cos it's the same feature
16:06 EvilPeng1 joined #git
16:06 _ikke_ BPL: Makes sense, those features I mention help with this
16:06 relipse phanimahesh i tried that but it said No rebase in progress?
16:07 relipse _ikke_ yes it is the same error , how do i reset my "recurring" branch back to the way it was
16:07 phanimahesh BPL: Do as _ikke_ suggested,and then git push --force-with-lease, because you are overwriting already published stuff.
16:07 _ikke_ exactly
16:07 TomyLobo joined #git
16:07 _ikke_ relipse: git reset --hard HEAD will undo all uncomitted changes to tracked files
16:08 BPL I see, I'll give it a shot, thanks guys. Btw, do you know this tool called Sourcetree from Attlasian? (the one I'm using)
16:08 svm_invictvs joined #git
16:08 dalias who came up with the weird name "with-lease"?
16:08 phanimahesh relipse: can you grab an sha from git reflog? you can git checkout <sha> to look around or git reset [--hard] <sha>
16:08 dalias --force-cas or --force-cmpxchg would make more sense
16:08 phanimahesh dalias: It isn't weird once you look hard at what it does.
16:09 dalias it is. there's no "lease" involved
16:09 dalias it's an atomic compare-and-swap operation
16:09 a_thakur joined #git
16:09 eshlox joined #git
16:09 phanimahesh It is as if you took a lease on the remote ref, and if your lease isn't valid (someone overwrote it), then the push aborts.
16:09 BPL https://answers.atlassian.com/questions/15829483/git-commit---fixup
16:09 phanimahesh umm, yes.
16:10 _ikke_ dalias: http://dl-cdn.alpinelinux.org/alpine/edge/main/x86_64/
16:10 _ikke_ arg
16:10 _ikke_ dalias: https://public-inbox.org/git/1373581580-13651-1-git-send-email-gitster@pobox.com/
16:10 phanimahesh I wouldn't mind --force-only-if either, in case with-lease is weird and cas or cmpxchg are not clear.
16:11 dalias *nod* i still think it's weird counter-intuitive naming
16:11 hdon joined #git
16:11 dalias with-lease would make sense if you took out some stateful lease, but you don't
16:11 dalias a lease would imply immunity to ABA issues
16:11 dalias ala ll/sc style atomics
16:11 _ikke_ dalias: This explains it more: https://public-inbox.org/git/1374561800-938-4-git-send-email-gitster@pobox.com/
16:11 BrianBlaze420 joined #git
16:11 phanimahesh It's helpful to think of it as a lease, though. It isn't an actual lease, yes.
16:11 dalias this is a cas-style atomic, not an ll/sc-style atomic
16:12 dalias the difference is ABA issues
16:12 _ikke_ " You assume you took the lease on the ref when you fetched to decide what the rebased history should be, and you can push back only if the lease has not been broken."
16:12 hdon hi all :) I come from Mercurial, and our branches work differently from git. One thing I miss is that mercurial documents which branch a commit was made to, which helps me understand my revision history at a glance. what can i do now as a git user?
16:12 bremner use commit messages? if you merge from a tag, that will be in the message
16:13 hdon hmm... yeah I guess that will work
16:13 dalias hdon, that's one of the most awful misfeatures of hg :-P
16:13 bremner also from a branch, I guess
16:13 dalias making people's crap temp branch names a permanent part of the history :-P
16:13 BrianBlaze420 joined #git
16:13 hdon bremner, i don't see it in the message when i do, for instance, git show
16:13 hdon or git log
16:13 bremner well, people can edit it out
16:14 hdon i haven't edited anything out
16:14 BPL _ikke_: So, I'm seeing the syntax is `git commit --fixup <commit>`, that means I don't need to specify any commit message to this new fixup little commit, right? <commit> is the parent's one?
16:14 buffal0 joined #git
16:14 timvisher joined #git
16:14 hdon maybe your git config automatically adds this?
16:14 dalias anyway, it's a bikeshed, sorry
16:14 hdon please share if so :)
16:14 phanimahesh hdon: git doesn't care about branches or tags. They aren't recorded in commits, (except in default merge commit messages)
16:14 bremner hdon: it might depend on git version. what version of git are you using? I guess the other point is you probably want --no-ff
16:14 raijin joined #git
16:15 phanimahesh hdon: And git users think it is for good reasons. mercurial people disagree, and there have been many arguments over this.
16:15 hdon phanimahesh, yes i know. but as i tried explaining, i miss seeing this information in the revision history. i think it will work to include this info in the commit message as bremner suggests.
16:15 hdon bremner, 1.9.1
16:15 bremner in a fast-forward merge, there is no commit, hence no commit message
16:15 BrianBlaze420 joined #git
16:15 hdon phanimahesh, i can accept the differences here. i think bremner has offered me a good compromise.
16:15 phanimahesh hdon: you can then write a git hook to add this to the commit mesage.
16:15 hdon bremner, that's is good behavior IMO
16:16 hdon phanimahesh, i'll have to look into this. thanks :)
16:17 thiago joined #git
16:17 phanimahesh dalias: ABA in this context means? Sorry, my mind's a bit rusted, can't recollect any ABAs.
16:17 bremner maybe ABBA?
16:17 jimi_ joined #git
16:17 mmmveggies joined #git
16:17 dalias https://en.wikipedia.org/wiki/ABA_problem
16:19 phanimahesh dalias: Ah. Not an issue here, because sha collisions are relatively unlikely
16:20 dalias not sha collisions
16:20 hdon phanimahesh, is commit-msg the appropriate hook?
16:20 phanimahesh bremner: I've only heard people mention ABBA, haven't heard any of theirs, either.
16:20 dalias my point is that it's not a lease because...
16:21 phanimahesh yep. I get it that it's not a lease. And there can be ABA issues. but they won't matter technically unless we have sha collisions.
16:21 dalias if you observed remote branch's head = your revision, then someone else changes it, then force reverts the change...
16:21 dalias you update despite other intervening changes
16:21 dalias you probably _want_ to update then
16:21 phanimahesh then git doesn't care that those changes happened at all. it makes no difference.
16:21 dalias it's the expected behavior
16:21 Rodya_ joined #git
16:21 dalias but it makes lease the wrong term
16:21 dalias it's a compare-and-swap
16:21 dalias not a lease
16:21 safe joined #git
16:21 phanimahesh yep. agreed.
16:21 phanimahesh hdon: lemme check. one minute.
16:22 anli joined #git
16:22 anli Is a .git file added to a submodule dir?
16:23 phanimahesh hdon: yep. commit-msg is the hook.
16:23 hdon thanks bremner, phanimahesh
16:23 hdon dalias, at least when hg says "branch" we mean it :) in git it's a horrible misnomer
16:23 phanimahesh anli: yep, to help git keep track of it as an independent git repo. that will be a plain file pointing to parent .git, though
16:23 anli ah, true
16:24 phanimahesh hdon: Matter of perspective..
16:24 hdon phanimahesh, i'm just being a jerk :3
16:24 anli Does git keep track of which version the submodule had when it was added?
16:24 _ikke_ yes
16:24 anli And git pull would never update it, I gather
16:24 dalias hdon, hm?
16:24 _ikke_ You need to execute git submodule update
16:24 anli yeah
16:25 anli I like how it is built up
16:25 _ikke_ Funny how you seem to grasp it immediately
16:25 anli haha
16:25 nitric joined #git
16:25 oskarkv joined #git
16:25 _ikke_ The other day we had a discussion with someone saying submodules were broken
16:25 anli ok?
16:26 timvisher joined #git
16:26 e14 joined #git
16:26 _ikke_ He didn't want to have to commit every change to the submodule
16:26 perlpilot submodules are broken ... if you're trying to make them do something they weren't designed for  ;)
16:26 anli hehe
16:27 phanimahesh they are broken from a convenience perspective. but from a safety, no surprises, do everything yourself perspective, they are brilliant.
16:27 dalias submodules/subrepos are a great way to get in big trouble
16:27 phanimahesh people seem to have different expectations from what the tools offer by default. That's the root of all trouble.
16:28 phanimahesh I wouldn't curse my worst enemies to have to work with tools that don't match their needs and mental models.
16:28 diogenese joined #git
16:28 neeca joined #git
16:28 neca joined #git
16:28 Blackpajamas joined #git
16:31 durham joined #git
16:32 anli git cannot store information about which branch or tag that is stable, right?
16:33 anli I have checked out a git repo as submodule, but I do not think the current master is compilable
16:33 anli Sorry for the svn lingo
16:36 _ikke_ no
16:37 fakenullie It is usually stored as branch name
16:37 _ikke_ If master is not compilable, I don't think that project has a stable branch
16:37 eka joined #git
16:37 Gsham joined #git
16:38 anli There are a lot of branches, maybe I should check for latest tag.
16:39 nowhereman joined #git
16:39 mikecmpbll joined #git
16:39 timvisher joined #git
16:40 phanimahesh anli: usually latest tag, or a branch called stable or something that is a named release, or master.
16:40 phanimahesh I look at things in that order.
16:40 Darren_ joined #git
16:41 anli Hm, wonder if only my environment thinks its incompilable, I do not trust qt creator a lot...
16:41 EvilPeng1 joined #git
16:41 anli There was a cmake file in the project
16:41 phanimahesh things would be much better if we all agreed on one thing, but then what's the fun in that
16:41 anli hehehe
16:41 va one could agree on disagreeing
16:41 phanimahesh anli: qmake can be a bit of a pain. check dependencies, qmake and qt versions, etc
16:41 anli ok
16:42 phanimahesh va: I've encountered many people who wouldn't even agree on disagreeing.
16:42 dalias phanimahesh, the awful way submodules always get used is that libraries are submodules and developers are always working with current version of the library, but the top-level project using it is mostly stable, and only rarely updates the submodule reference
16:42 dalias if any complex branching ends up happening then....
16:43 dalias it becomes impossible to trace the actual whole-program revision history that developers were working with
16:43 phanimahesh well, most developers are bad maintainers. (myself included)
16:43 dalias e.g. for bisect, regression tracking, etc.
16:43 Emperor_Earth joined #git
16:43 troulouliou_div2 joined #git
16:44 dalias a good submodule front-end would disallow commits to the checked-out submodule without also making a linked commit in the parent-level repo using it
16:44 dalias and would require matching branches too
16:44 anli Ok, you are obviously aware of qt, where do I find the big user base on that environment?
16:44 anli The #qt channel is like dead fish
16:45 anli A lot of idlers
16:45 phanimahesh My only serious experience with qt was to debug works on my machine, doesn't work there issue for a colleague.
16:45 Guest29223 joined #git
16:46 phanimahesh i was the local linux wizard, widely believed to be able to do anything. I had to pay the price.
16:47 eka joined #git
16:47 phanimahesh (the price is to actually have to do anything. Discworld reference, in case there are any Prachett fans here)
16:47 hahuang61 joined #git
16:48 phanimahesh anli: To answer your question, no idea. :/
16:48 anli ok :)
16:48 thebope_ joined #git
16:48 anli There are qt forums, but they seem a bit how can I say it, braindead
16:49 phanimahesh stackoverflow?
16:49 anli hm
16:50 anli Maybe I should try to ask there
16:51 sbeller <dalias> a good submodule front-end would disallow commits to the checked-out submodule without also making a linked commit in the parent-level repo using it
16:51 sbeller dalias: why would this be a good thing? I guess as a first step a config option would suffice
16:51 timvisher joined #git
16:52 phanimahesh I disagree. if I can't work on a submodule independent of the parent repo, why have it at all?
16:52 osse anli: have you checked the qt wiki?
16:52 freimatz joined #git
16:52 anli I do not expect a lot from that, but thanks for the tip
16:53 nettoweb1 joined #git
16:53 dalias sbeller, to prevent the situation i described above
16:53 dalias where actual history of the project as a whole is untrackable
16:54 phanimahesh that is just due to lack of proper discipline on part of developers. if one works with a different snapshot of a submodule, they commit it.
16:55 phanimahesh trying clean builds in a CI would be simplest way to catch bad commits.
16:55 phanimahesh giving up the flexibility is going too far, IMO.
16:56 anli Lets assume I find a fix for the compilation problems in the submodule. Do I have to clone the submodule to be able to convey the changes to my colleagues then without committing to the submodule repo itself?
16:56 kristofferR joined #git
16:56 dviola joined #git
16:56 littleforest joined #git
16:56 sbeller anli: no you can do the fix in the submodule repo and from there you push it to e.g. your codereview system
16:57 sbeller no need for a separate repo there
16:57 anli well, I have no codereview system
16:58 phanimahesh anli: you can make a new commit in the submodule, push it somewhere accessible to your colleagues, and make a commit in parent repo with changed submodule, and changed url of submodule.
16:58 anli yeah, thats forking then
16:59 phanimahesh yep. you want a change not present upstream, you fork and use the fork till it lands upstream.
16:59 anli yeah
16:59 phanimahesh submodules or not, that's what we live with.
16:59 anli A github fork can be private, right?
16:59 anli Forget the question, it answers itself
16:59 eka joined #git
17:00 ekinmur joined #git
17:00 anli I think some of the automatically included qrc files should not be automatically included :)
17:00 anli sorry for qt specific stuff in here
17:01 timvisher joined #git
17:01 anli I go to #qt afterwards and discuss git if you want ;)
17:01 hdon dalias, well in git a branch is just a symbolic name for a revision. that is not a branch.
17:01 hdon :3
17:02 phanimahesh hdon: It is. A branch is a handy name to a reference, that's the way it was intended. What's in a name, afterall?
17:03 phanimahesh s/.../handy name pointing to a reference/
17:04 geoid_ joined #git
17:06 dalias hdon, no, it's a symbolic name that follows commits to a branch of the overall DAG
17:06 dalias a tag is a symbolic name for a specific revision
17:06 dalias a branch is a tag that moves when you commit to the branch
17:08 patarr joined #git
17:09 MacWinner joined #git
17:09 hahuang65 joined #git
17:10 MacWinner joined #git
17:10 thiago joined #git
17:10 timvisher joined #git
17:11 diogenese joined #git
17:13 star_prone joined #git
17:13 PCatinean_ joined #git
17:13 Darren_ joined #git
17:14 xall joined #git
17:15 Faylite joined #git
17:16 Guest54 joined #git
17:16 Gsham joined #git
17:16 cyan__ joined #git
17:16 astrofog joined #git
17:17 jaequery joined #git
17:18 Limes_ joined #git
17:18 luxPerpetua joined #git
17:18 milkshoes joined #git
17:19 timvisher joined #git
17:19 Derperperd joined #git
17:20 PioneerAxon joined #git
17:21 ejb joined #git
17:22 Mekkis joined #git
17:23 LionsMane joined #git
17:24 ExoUNX joined #git
17:25 AnarchyAo joined #git
17:26 _ikke_ branches and tags are both refs
17:26 _ikke_ Only branches will automatically moved (and can be actually checked out)
17:27 malcon joined #git
17:28 Murii joined #git
17:28 t0by joined #git
17:29 fatalhalt joined #git
17:29 kpease joined #git
17:30 timvisher joined #git
17:31 thiago joined #git
17:31 theskillwithin joined #git
17:33 lus joined #git
17:33 SteffanW joined #git
17:33 xaa joined #git
17:34 EvilPenguin joined #git
17:35 jaguarmagenta joined #git
17:36 Spec-Chum joined #git
17:36 dreiss joined #git
17:36 Spec-Chum joined #git
17:36 mikecmpbll joined #git
17:37 crayon joined #git
17:37 thebope_ joined #git
17:37 rpg joined #git
17:37 tmk joined #git
17:39 danslo_ joined #git
17:40 lus joined #git
17:41 eivarv joined #git
17:42 timvisher joined #git
17:42 Blackpajamas joined #git
17:43 omx joined #git
17:45 flavius joined #git
17:46 dsantiago joined #git
17:47 robotroll joined #git
17:49 Sasazuka joined #git
17:50 timvisher joined #git
17:52 purope joined #git
17:53 ki0 joined #git
17:53 purope Hi, can someone help me with my little Git problem. When I do a git clone --recursive https://repo, currently I'm not getting the updated versions of submodules, and I'm getting the last commits that the submodules have been added to. How can I make it so this command will get the most up to date version of some branch? And how can I specify the branch?
17:53 akanksha_ joined #git
17:55 a_thakur joined #git
17:57 malcon joined #git
17:57 thiago joined #git
17:58 Phylock joined #git
17:59 nedbat some days it seems like this channel should be renamed #submodules :)
17:59 faux___ joined #git
17:59 sbeller purope: git submodule update --remote
17:59 aapaaal joined #git
17:59 timvisher joined #git
18:00 Orion3k joined #git
18:00 jj96 joined #git
18:01 phanimahesh _ikke_: you can checkout a tag, just like you can checkout any git sha/ref
18:02 faux____ joined #git
18:02 _ikke_ phanimahesh: The 'actual' being the keyword
18:02 purope sbeller: What does that do?
18:02 _ikke_ You check out the commit it's refering to, not the tag
18:02 Derperperd joined #git
18:02 adamretter joined #git
18:02 sbeller purope: search for --remote in https://git-scm.com/docs/git-submodule
18:03 sbeller Instead of using the superproject’s recorded SHA-1 to update the submodule, use the status of the submodule’s remote-tracking branch.
18:03 cagedwisdom joined #git
18:03 _Vi joined #git
18:03 adamretter I have a branch, and I want to rewrite the history so that a merge commit, and then subsequent revert merge commit never happened, how can I do that? I thought I could use rebase, but I don't think that is playing nicely
18:03 phanimahesh _ikke_: well, that distinction was sort of obvious and natural.
18:04 phanimahesh adamretter: you can either hard reset and push --force-with-lease, or rebase.
18:04 phanimahesh (rebase and push --force-with-lease)
18:05 adamretter phanimahesh: my question isn't about pushing, I want to know how to correctly edit the history of the branch
18:05 geoid_ joined #git
18:05 phanimahesh if you have no intervening commits and can get away with hard reset, do it. else, rebase is your only option.
18:05 adamretter so I have intervening commits
18:05 phanimahesh adamretter: rebase -i
18:06 phanimahesh And standard caveats apply. !rewrite
18:06 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
18:07 adamretter phanimahesh: sure I get that, but rebase -i looks completely different to `git log`. I think maybe I need the -p option
18:08 _ikke_ adamretter: Be aware of the caveats with -p and -i combined
18:08 timvisher joined #git
18:08 adamretter _ikke_: yeah I need to check those first for sure
18:09 phanimahesh if you want to play it safe, can you reset to just before merge, and cherry pick commits from later?
18:09 _ikke_ My recomendation is not the rewrite history that contains merges
18:10 phanimahesh Aye. I stand with _ikke_.
18:11 cjohnson When I run git branch --merged master, and it lists a bunch of branches, assuming master is up to date from remote and I just ran git fetch, does it mean it's also safe to delete those branches off remote?
18:11 cjohnson for instance if feature/multiple-analytics-keys  is listed, but somebody has pushed changes to that feature since I last updated it
18:12 cjohnson and if not how can I specifically check the merged branches on remote?
18:12 cjohnson git branch --merged origin/master I think will only show those where my *local* branches are merged to origin/master
18:12 cjohnson not the origin/feature/foo
18:13 cjohnson (alternatively does github provide an interface to list prunable branches?)
18:13 thebope_ joined #git
18:15 ViktorRu joined #git
18:16 timvisher joined #git
18:21 _ikke_ git branch --remote --merge
18:21 _ikke_ and git branch --all --merged
18:22 Surfer2010 joined #git
18:24 patarr_ joined #git
18:24 jottr joined #git
18:24 OnBrokenWings joined #git
18:24 timvisher joined #git
18:26 m0viefreak joined #git
18:26 aikuma joined #git
18:28 thiago joined #git
18:28 cjohnson git branch --remote --merged is listing a ton of branches, most of which have already been deleted from remote
18:29 _ikke_ cjohnson: git fetch --prune would remove them
18:29 cjohnson ah, so those are local shadows of those remote branches?
18:29 _ikke_ yes, remote tracking branches
18:29 Ahmed90 joined #git
18:29 cjohnson ah, makes sense. so then when I do git branch --remote --merged  do I need to include master or origin/master at teh end of that?
18:30 cjohnson guessing origin/master
18:30 _ikke_ You supply the branch you want to check is merged
18:30 va left #git
18:30 cjohnson just confirming that it would be origin/master not master
18:30 cjohnson or does it not care
18:30 _ikke_ doesn't matter
18:30 cjohnson k
18:30 _ikke_ if you do master, it will check master, if you do origin/master, it will check origin/master
18:31 cjohnson ok so since I'm trying to delete remote branches I probably want origin/master by default. unless I'm sure local master is up to date
18:31 _ikke_ So if you want to know if there is anything remote that has not been merged, you'd use the remote tracking branches after you fetched
18:31 tgunr left #git
18:31 cjohnson which in this case it is
18:31 _ikke_ right
18:31 cjohnson got it
18:31 timvisher joined #git
18:31 cjohnson last question, is there a combination git branch -D foo and git push origin :foo
18:32 _ikke_ Not that I'm aware of
18:32 _ikke_ I'm fairly certain there isn't
18:32 eivarv joined #git
18:32 cjohnson alright!
18:33 buffal0 joined #git
18:34 [Brain] joined #git
18:35 aikuma git grep -E "&spi[\d]" *   why does this not work?
18:35 _ikke_ aikuma: what do you expect to happen?
18:35 aikuma http://www.regexpal.com/?fam=95967
18:36 aikuma _ikke_: http://www.regexpal.com/?fam=95967 this
18:36 _ikke_ Note that this is not PCRE (Which most of these tools use)
18:37 aikuma _ikke_: i tried with javascript as well
18:37 _ikke_ javascript is also PCRE
18:37 _ikke_ grep is a better tool to test it
18:37 Sasazuka_ joined #git
18:37 aikuma _ikke_: ok, what should i do inorder to get this kind of pattern?
18:37 aikuma in grep
18:38 snfgf joined #git
18:39 timvisher joined #git
18:39 _ikke_ &spi[0-9]
18:39 Guest69334 joined #git
18:40 GodGinrai aikuma: git grep -E '&spi[0-9]'
18:40 aikuma _ikke_: so \d is not supported
18:40 aikuma ok
18:40 _ikke_ right
18:40 _ikke_ http://www.regular-expressions.info/posix.html
18:41 malcon joined #git
18:41 aikuma _ikke_: i was about to ask you this but you answered before that. Clairvoyant.
18:42 _ikke_ :P
18:42 GodGinrai aikuma: you could use [[:digit:]] as well, but that would waste more keystrokes :P
18:42 * _ikke_ wishes everything would support pcre
18:43 * GodGinrai wishes everything would use pcre only when necessary
18:44 fedepad joined #git
18:44 GodGinrai _ikke_: https://swtch.com/~rsc/regexp/regexp1.html
18:45 _ix joined #git
18:45 thebope_ joined #git
18:46 _ADN_ joined #git
18:46 _ikke_ I know pcre is less performant (not regular), but the lack of some features is so frustrating
18:46 timvisher joined #git
18:46 jaequery joined #git
18:46 Xethron joined #git
18:46 Sasazuka joined #git
18:47 Rodya_ joined #git
18:47 GodGinrai _ikke_: I do not disagree that some features from pcre are really nice to have.  But many of the same features can be achieved through smart coding without much effort
18:48 _ikke_ GodGinrai: Sometimes you have to option to code around it
18:48 hariel joined #git
18:48 _ikke_ s/to/no/
18:49 GodGinrai _ikke_: well, given that the Thompson NFA approach is used much less than pcre, I'm not really certain where you wouldn't have an option to code around it.
18:49 GodGinrai except for git grep, I guess
18:49 _ikke_ GodGinrai: Well, situations where you cannot code at all
18:49 GodGinrai but how often do you really need backreferences when using git grep?
18:51 _ikke_ Not backreference
18:52 _ikke_ GodGinrai: Nice reference btw
18:52 GodGinrai yea, it's a great article
18:52 nettoweb joined #git
18:52 timvisher joined #git
18:53 flaguy48 joined #git
18:55 Rodya_ joined #git
18:56 anonymuse joined #git
18:58 nettoweb_ joined #git
18:58 jimi_ joined #git
18:59 EvilPenguin joined #git
18:59 timvisher joined #git
19:00 nwkj86 joined #git
19:00 Balliad joined #git
19:01 EvilPeng1 joined #git
19:02 star_prone joined #git
19:04 SuperSeriousCat joined #git
19:05 geoid_ joined #git
19:06 timvisher joined #git
19:07 Sasazuka joined #git
19:09 magellanicloud joined #git
19:12 LBV_User joined #git
19:12 timvisher joined #git
19:12 svankmajer joined #git
19:12 porsche944_ joined #git
19:13 Kitty- joined #git
19:13 dojo2 joined #git
19:14 limar joined #git
19:14 magellanicloud joined #git
19:14 xall joined #git
19:15 Derperperd joined #git
19:17 Jellyg00se joined #git
19:18 timvisher joined #git
19:20 thiago joined #git
19:20 kpease joined #git
19:21 w00z_ joined #git
19:22 w00z_ left #git
19:23 Sasazuka joined #git
19:23 [sr] joined #git
19:23 [sr] howdy
19:23 [sr] small help
19:23 finalbeta joined #git
19:24 [sr] i've configured git several years ago, and been using since, but now i'm running an x64 sys to have other capabilities, and git gives-me this:  /etc/init.d/git-daemon start warning: git-daemon: unable to open supervise/ok: file does not exist
19:24 [sr] any idea?
19:24 [sr] i've been searching the web but none found that would be my issue
19:24 timvisher joined #git
19:24 [sr] what would i be missing?
19:26 geri joined #git
19:27 Cavallari joined #git
19:27 thiago [sr]: that file is not part of Git.
19:28 preaction sounds like something from bsd daemontools
19:29 isysd joined #git
19:29 [sr] thiago: what file, the init script or the supervise?
19:30 thiago [sr]: the init script
19:30 [sr] thiago: you're right, i copied it from the old system, since i installed git from deb packages and i had no init script
19:30 DeltaHea_ joined #git
19:31 thiago [sr]: probably because you don't need one
19:31 timvisher joined #git
19:31 [sr] thiago: but there's nothing running on git port and didnt found any script to start it
19:31 [sr] well many things may changed... it was about 4 years ago
19:31 thiago please check with #debian
19:32 thiago it might be with systemctl now
19:32 laumann joined #git
19:32 [sr] hum... i've seen that somewhere
19:33 [sr] well i have a starting point, thanks thiago!
19:33 kulelu88 joined #git
19:34 SteffaN joined #git
19:36 jaguarmagenta joined #git
19:36 malwar3hun73r left #git
19:37 Sasazuka_ joined #git
19:37 timvisher joined #git
19:38 ketas joined #git
19:40 danslo_ joined #git
19:40 johnmilton joined #git
19:41 prg3 joined #git
19:41 buffal0 joined #git
19:42 tvw joined #git
19:44 patarr joined #git
19:44 timvisher joined #git
19:45 johnmilton joined #git
19:46 fedepad joined #git
19:46 _ix joined #git
19:46 Socialoutcast joined #git
19:48 LionsMane joined #git
19:48 jottr joined #git
19:49 Jackneill joined #git
19:49 aidalgol joined #git
19:50 timvisher joined #git
19:54 jottr_ joined #git
19:55 timvisher joined #git
19:56 Seveas [sr]: in debian systems, the proper git-daemon initscript sits in the git-daemon-sysv package
19:56 Rodya_ joined #git
19:56 svm_invictvs joined #git
19:56 Seveas there's also git-daemon-run for runit based systems. There doesn't seem to be a proper systemd unit yet
19:57 _ix joined #git
20:00 hashpuppy joined #git
20:00 timvisher joined #git
20:02 Master_Bob joined #git
20:03 Master_Bob Hello. When I checkout a file from a previous commit (i.e. git checkout SHA /path/to/file) is the history for that file retained?
20:04 _ikke_ Master_Bob: The only thing that does is place the contents of that file in the working tree
20:04 _ikke_ No commits are touched
20:04 Lor_ joined #git
20:05 Master_Bob Okay. So, then there's no way for more to restore a file and it's history?
20:05 _ikke_ I'm not following you
20:05 [sr] Seveas: someone on debian helped, thanks for your help!
20:06 _ikke_ Uncomitted changes you will lose, but anything comitted is retained
20:06 patarr joined #git
20:06 timvisher joined #git
20:06 cyan__ joined #git
20:06 e14 joined #git
20:07 nidr0x joined #git
20:07 Master_Bob _ikke_: so, I was working on file X. And instead of editing file X, I made new file Y which is a copy of X. I edited Y. Now I've realized my mistake, and I want to go back and edit X and place the changes made in Y in X, so that way the history of the file is retained.
20:08 Master_Bob Does that make sense?
20:08 jaequery joined #git
20:08 ahmedelgabri joined #git
20:09 _ikke_ Ok, and you have comitted the changes to Y?
20:09 diogenese joined #git
20:09 Master_Bob Yes.
20:11 timvisher joined #git
20:11 DieguezZ joined #git
20:12 _ikke_ You'd have to use interactive rebase to move the changes from Y to X
20:12 fscala joined #git
20:13 Master_Bob oh, fun. Okay. Thanks _ikke_ !
20:14 geoid_ joined #git
20:16 Master_Bob X was also deleted.
20:17 timvisher joined #git
20:18 anonymuse joined #git
20:19 ahmedelgabri joined #git
20:20 kpease joined #git
20:21 lus joined #git
20:22 timvisher joined #git
20:24 cdg joined #git
20:25 Limes_ joined #git
20:25 linelevel joined #git
20:27 HoierM joined #git
20:27 cdg joined #git
20:28 timvisher joined #git
20:29 diogenese joined #git
20:30 svankmajer joined #git
20:33 Sasazuka joined #git
20:34 timvisher joined #git
20:35 ahmedelgabri joined #git
20:38 Sasazuka_ joined #git
20:38 DeltaHea_ joined #git
20:38 Eugene [sr] - git-daemon is the server for the (unencrypted, not secured, not recommended) git:// protocol. Chances are that you want a tool like !gitolite instead
20:38 gitinfo [sr]: Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/
20:39 Socialoutcast joined #git
20:40 timvisher joined #git
20:42 [sr] Eugene: will dig into it
20:44 patarr joined #git
20:45 ki0 joined #git
20:45 buffal0 joined #git
20:46 timvisher joined #git
20:46 jlebar joined #git
20:48 nettoweb joined #git
20:48 msonntag joined #git
20:50 kpease joined #git
20:51 star_prone joined #git
20:51 BPL I've been practicing about squashing commits with sourcetree and it's quite straightforward, now... I was wondering how to squash the 2 first commits like the ones shown here http://screencast.com/t/gmP0m09Oy, when i try to do it on sourcetree I get this dialog, http://screencast.com/t/gmP0m09Oy . As you can see I only see one commit I can't squash with the previous one, any clue?
20:51 Exagone313 joined #git
20:52 Eugene Squashing the initial commit is a special case because it doesn't have any parents to diff against, so I'm unsurprised that some tools fail on it
20:52 shaggycat joined #git
20:52 Eugene If you want to hide the !sausage that thoroughly you should consider making an !orphan branch and just recreate the history that you want
20:52 gitinfo [!sausage_making] Some developers like to "hide the sausage making", transforming their commits before presenting them to the outside world. See http://sethrobertson.github.com/GitBestPractices/#sausage and !perfect
20:52 gitinfo To create an orphaned branch(unrelated to existing history): `git checkout --orphan newbranchname`. This will leave your index/worktree as-is(use `rm .git/index; git clean -dfx` to delete EVERYTHING). If you get 'error: unknown option `orphan`' see !orphan_old. For an empty/null commit see !orphan_null.
20:52 BPL Eugene: I see, so... is it not possible to do? Take a look http://screencast.com/t/f5gvEK6DzSS
20:52 timvisher joined #git
20:52 Eugene Or, just don't care and move on with life. Nobody else cares about your commit messages(really, we don't)
20:53 Eugene I'm not familiar with that tool, sorry
20:53 BPL Eugene: Ok, forget about that tool, is it not possible then?
20:53 Eugene Sure; man git-rebase and see --root
20:53 gitinfo the git-rebase manpage is available at http://jk.gs/git-rebase.html
20:54 BPL alright, ty
20:54 anonymuse joined #git
20:54 Sasazuka joined #git
20:56 humanBird joined #git
20:56 humanBird git-credential-osxkeychain died of signal 15  what the heck does this mean?
20:56 bn0t_ joined #git
20:57 patarr joined #git
20:58 timvisher joined #git
20:59 Doginal joined #git
21:04 timvisher joined #git
21:04 Exagone313 joined #git
21:05 littleforest joined #git
21:07 Sasazuka joined #git
21:07 BPL when i do git rebase -i vim is being used by default, any chance to change it by another text editor?
21:09 timvisher joined #git
21:09 nedbat BPL: define the EDITOR environment variable
21:10 BPL nedbat: Ah, ok, cheers!
21:11 diogenese joined #git
21:12 nettoweb joined #git
21:12 pikajude joined #git
21:13 DarkFox_zzzzZZZz joined #git
21:13 humanBird anyone know what signal 15 is?
21:13 timvisher joined #git
21:14 thebope_ joined #git
21:16 purope joined #git
21:16 alex1a joined #git
21:17 Thorn joined #git
21:19 timvisher joined #git
21:19 shaggycat joined #git
21:20 perlpilot humanBird: yes
21:20 humanBird what is it
21:20 perlpilot SIGTERM
21:20 inflames joined #git
21:21 humanBird well, why is it sig terming
21:21 humanBird why is git doing that
21:21 humanBird git-credential-osxkeychain died of signal 15
21:21 perlpilot I dunno
21:21 btree joined #git
21:24 timvisher joined #git
21:24 NeverDie joined #git
21:25 btree joined #git
21:28 jeffreylevesque joined #git
21:28 mikecmpbll joined #git
21:29 ViktorRu joined #git
21:29 timvisher joined #git
21:31 Axuu joined #git
21:31 lus joined #git
21:32 jeffreylevesque_ joined #git
21:33 Hobbyboy joined #git
21:34 btree left #git
21:34 safe joined #git
21:34 timvisher joined #git
21:35 nocbreak joined #git
21:37 e14 joined #git
21:37 jaguarmagenta joined #git
21:39 Guest37970 joined #git
21:39 timvisher joined #git
21:41 danslo joined #git
21:42 rahtgaz joined #git
21:43 Sasazuka_ joined #git
21:43 cdg_ joined #git
21:44 timvisher joined #git
21:45 lus joined #git
21:45 nowhereman joined #git
21:46 mizu_no_oto joined #git
21:46 pcarrier joined #git
21:47 deepy joined #git
21:47 pcarrier I'm running into something weird. `git diff-index --exit-code HEAD -- foo` exits with status 1, then I run `git status`, then `git diff-index --exit-code HEAD -- foo` exists with status 0
21:48 pcarrier am I missing something?
21:48 Gamecubic joined #git
21:48 e14 joined #git
21:48 pcarrier does `git status` somehow refresh the index?
21:49 timvisher joined #git
21:49 pcarrier (git 2.10.1 on macos sierra)
21:49 nitric_ joined #git
21:52 ketas joined #git
21:53 Socialoutcast joined #git
21:53 _maz joined #git
21:53 DJ_Spies joined #git
21:53 disi joined #git
21:53 timvisher joined #git
21:54 _maz hi, I have a git repo on localhost(mac 10.11.5). I do `git push origin <branch>` and I get the error `remote: env: python2: No such file or directory`
21:54 _maz I have no idea why this just suddenly started happening
21:54 lb1a _maz could you !repro please
21:54 gitinfo [!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
21:55 fmcgeough joined #git
21:55 _maz ok
21:56 _maz git push fails: https://gist.github.com/mazz/52a5838d37314f2365c878fa7e58e575
21:57 jimi_ joined #git
21:58 timvisher joined #git
21:58 lb1a _maz there you go, line 9: pre-receive hook declined
21:58 lb1a look whats in there and causes the trouble
21:58 BPL what's the meaning of `error: cannot spawn plink: No such file or directory` when I'm trying to clone a github repo using ssh protocol?
21:59 _maz lbla, where do I find the pre-receive hook?
21:59 nitric joined #git
22:00 thiago mizu_no_oto: the error comes from your remote repository. Fix your server.
22:01 mizu_no_oto thiago: which error?
22:01 thiago mizu_no_oto: sorry, bad nick completion
22:02 thiago _maz: ^^^ that was for you. The error is in your server.
22:02 dojo2 joined #git
22:02 _maz thiago ok it looks like from docs on the server I should check .git/hooks
22:03 lb1a _maz in the repo you are trying to push to
22:03 timvisher joined #git
22:03 _maz lbla got it
22:04 lb1a _maz the repo you are trying to push to is most likly a bare repo. so no .git/hooks but directly repo.git/hooks
22:06 hahuang65 joined #git
22:06 _maz lbla thanks for the follow-up. the script has  #/usr/bin/env python2 at the beginning. but when I do `/usr/bin/env python2` it resolves OK to python 2.7.10  -- so why does it fail in the script but not from Terminal/zsh ?
22:08 justanotheruser joined #git
22:08 nitric joined #git
22:08 lus joined #git
22:08 lb1a _maz did you try to execute your pre-receive-hook directly with /usr/bin/env python2 pre-receive-hook
22:08 timvisher joined #git
22:09 mangolisk joined #git
22:09 lb1a _maz and btw its l b 1 (number one) a ;) i barely see your mentions :D
22:10 _maz lb1a: got it. yes I did `/usr/bin/env python2 pre-receive-hook`  no issue in zsh
22:10 _maz lb1a: /usr/bin/env python2 pre-receive rather
22:10 lb1a _maz then you'll have to debug it. cannot say without code
22:10 lb1a my bad
22:11 _maz lb1a: thanks, I think your hints set me in right direction
22:11 lb1a _maz different shell env maybe?
22:12 _maz lb1a: possibly, that's what I was thinking. I just noticed that kallithea edited the script. maybe that is the culprit
22:12 canuk joined #git
22:13 lb1a _maz happy debugging :)
22:13 _maz lb1a: I will try making a fresh repo and compare
22:13 timvisher joined #git
22:13 Sasazuka joined #git
22:14 aikuma_ joined #git
22:17 _maz lb1a: fixed it. I just renamed `pre-receive` to `pre-receive_kallithea` .. bogus script and I don't use kallithea anyway
22:17 lb1a _maz ^^ the less the better
22:18 timvisher joined #git
22:19 _maz lb1a: thiago: appreciate the help
22:22 zivester joined #git
22:23 nowhereman joined #git
22:23 timvisher joined #git
22:23 overlord_tm joined #git
22:24 afuentes joined #git
22:24 lb1a yw
22:27 Rzoj joined #git
22:28 timvisher joined #git
22:31 aikuma joined #git
22:31 relipse joined #git
22:32 Sasazuka_ joined #git
22:32 HoierM joined #git
22:32 timvisher joined #git
22:33 EvilPenguin joined #git
22:33 madewokherd joined #git
22:34 samtheknife joined #git
22:35 cdown joined #git
22:35 purope joined #git
22:37 patarr joined #git
22:37 timvisher joined #git
22:39 nettoweb joined #git
22:39 diogenese joined #git
22:39 star_prone joined #git
22:42 diogenese joined #git
22:42 DJ_Spies joined #git
22:42 anuxivm joined #git
22:42 timvisher joined #git
22:44 beagleGuy joined #git
22:44 fedepad joined #git
22:47 timvisher joined #git
22:48 porsche944 joined #git
22:48 eb joined #git
22:50 netj joined #git
22:50 jaequery left #git
22:50 Salander27 joined #git
22:51 kadoban joined #git
22:51 timvisher joined #git
22:55 timvisher joined #git
22:56 green_mars joined #git
22:56 green_mars left #git
22:59 dviola joined #git
22:59 timvisher joined #git
23:04 timvisher joined #git
23:05 e14 joined #git
23:05 quetz joined #git
23:06 buffal0 joined #git
23:07 ColdKeyboard joined #git
23:08 ToBeCloud joined #git
23:10 metrixx joined #git
23:10 e14 joined #git
23:10 snowkidind joined #git
23:15 saustin1 joined #git
23:16 e14 joined #git
23:17 jason237 joined #git
23:18 saustin1 joined #git
23:21 sangy joined #git
23:21 edaught joined #git
23:26 nikivi joined #git
23:29 mrpackethead joined #git
23:29 aax joined #git
23:30 mablae_ joined #git
23:31 KaIdEn13 joined #git
23:33 danlovesppipes joined #git
23:35 wessbrown84 joined #git
23:36 eindoofus joined #git
23:37 saustin1 joined #git
23:38 jaguarmagenta joined #git
23:40 saustin1 joined #git
23:42 thebope joined #git
23:42 saustin1 joined #git
23:45 littleforest joined #git
23:46 EvilPeng1 joined #git
23:47 wololokatamari[m joined #git
23:49 anuxivm joined #git
23:52 Ineentho[m] joined #git
23:52 hallogallo[m] joined #git
23:52 kikijiki[m] joined #git
23:52 M-meznak joined #git
23:52 Remramm joined #git
23:52 M-mistake joined #git
23:52 aviraldg joined #git
23:53 [cust] joined #git
23:53 bcardarella joined #git
23:54 M-shine joined #git
23:54 nikivi joined #git
23:54 flupke joined #git
23:55 kjsage[m] joined #git
23:55 e14 joined #git
23:56 xall joined #git
23:56 theresajayne joined #git
23:57 BPL joined #git

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