Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-10-20

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

All times shown according to UTC.

Time Nick Message
00:00 alex1a joined #git
00:04 dsdeiz_ joined #git
00:04 Dohplay joined #git
00:05 buddyq osse, it's working... only kink is that git fetch to github from my hosts ask for a password. I thought the ssh key was to bypass that.
00:08 jimi_ joined #git
00:09 Coka joined #git
00:09 mischat joined #git
00:11 nettoweb joined #git
00:12 hahuang65 joined #git
00:13 star_prone joined #git
00:15 peey joined #git
00:17 peey I created a subdirectory within my git repo and did a git init in that directory and pushed the subdirectory to remote. I then pushed the changes to original repo to the remote, but when I try to clone it it gives me No submodule mapping found in .gitmodules for path 'new'. What should I do?
00:18 unbalancedparen joined #git
00:20 Rodya_ joined #git
00:22 alex1a joined #git
00:24 denimsoft joined #git
00:27 CussBot joined #git
00:28 KnownSyntax joined #git
00:31 jstein_ joined #git
00:33 zivester joined #git
00:33 Derperperd joined #git
00:34 nettoweb joined #git
00:34 e14 joined #git
00:40 kulelu88 joined #git
00:40 BPL Any cmake contributor over here?
00:41 hahuang65 joined #git
00:43 lagothrix joined #git
00:45 anonymuse joined #git
00:48 thebope joined #git
00:51 Goplat joined #git
00:53 einnjo joined #git
00:55 evanwang joined #git
00:58 fstd_ joined #git
00:59 CussBot_ joined #git
01:03 pks joined #git
01:04 Jameser joined #git
01:05 dreiss joined #git
01:06 sunri5e joined #git
01:07 ochorocho__ joined #git
01:10 mischat joined #git
01:11 d^sh joined #git
01:12 Motoko joined #git
01:12 al-damiri joined #git
01:13 krs93 joined #git
01:17 krs93 joined #git
01:21 Dougie187 joined #git
01:21 CussBot_ joined #git
01:21 plos joined #git
01:26 parsnip joined #git
01:26 axisys joined #git
01:26 CussBot joined #git
01:29 finalbeta joined #git
01:34 plos joined #git
01:39 ResidentBiscuit joined #git
01:47 brokensyntax joined #git
01:49 thebope joined #git
01:50 drbean joined #git
01:53 Vampire0_ joined #git
01:54 liuzhen joined #git
01:55 Limes_ joined #git
01:56 BPL joined #git
02:02 pks joined #git
02:04 sunri5e joined #git
02:06 Jameser joined #git
02:06 aspiers joined #git
02:07 Wulf joined #git
02:07 msonntag joined #git
02:10 plos joined #git
02:13 star_prone joined #git
02:13 jstimm joined #git
02:15 kyan joined #git
02:20 Rodya_ joined #git
02:22 svm_invictvs joined #git
02:24 Rodya_ joined #git
02:25 safe joined #git
02:26 hexagoxel joined #git
02:28 shakalaka joined #git
02:33 Dougie187 joined #git
02:38 a3Dman_ joined #git
02:39 Gsham joined #git
02:40 Gsham joined #git
02:42 evanwang joined #git
02:43 Anja joined #git
02:43 dsdeiz_ left #git
02:43 dsdeiz joined #git
02:43 fedepad joined #git
02:44 xall joined #git
02:46 dsantiago joined #git
02:48 howdoi joined #git
02:49 chachasmooth joined #git
02:50 lagothri1 joined #git
02:50 thebope joined #git
02:52 chartractegg joined #git
02:53 Gsham joined #git
02:56 Wulf4 joined #git
02:56 brokensyntax joined #git
02:57 littleforest joined #git
02:59 mangolisk joined #git
03:01 xshuiniu joined #git
03:02 ResidentBiscuit joined #git
03:02 harish_ joined #git
03:03 aavrug joined #git
03:03 blackwind_123 joined #git
03:06 e14 joined #git
03:07 brokensyntax joined #git
03:09 e14 joined #git
03:09 chartrac_ joined #git
03:12 mischat joined #git
03:14 gasull joined #git
03:14 Vortex35 joined #git
03:14 nettoweb joined #git
03:15 buddyq joined #git
03:18 Andrew_K joined #git
03:20 justrohu joined #git
03:21 chartractegg joined #git
03:23 govg joined #git
03:24 fahadash joined #git
03:26 alex1a joined #git
03:28 rchavik joined #git
03:28 AndreSomers joined #git
03:31 boombatower joined #git
03:36 neunon joined #git
03:40 krs93 joined #git
03:41 brokensyntax joined #git
03:42 PioneerAxon joined #git
03:42 demo joined #git
03:43 xall joined #git
03:47 parsnip demo: hi
03:51 thebope joined #git
03:52 dreiss joined #git
03:54 mehola joined #git
03:54 a_thakur joined #git
03:57 mizu_no_oto joined #git
04:03 Motoko joined #git
04:07 shaggycat joined #git
04:08 peterR`` joined #git
04:10 Rodya_ joined #git
04:11 LeBlaaanc joined #git
04:12 thiago joined #git
04:13 mischat joined #git
04:20 Rodya_ joined #git
04:24 svm_invictvs joined #git
04:25 cagedwisdom joined #git
04:25 eshlox joined #git
04:25 littleforest joined #git
04:28 chartractegg joined #git
04:31 Anja joined #git
04:37 Cabanossi joined #git
04:41 ToBeCloud joined #git
04:43 mycelium joined #git
04:45 daey_ joined #git
04:51 brillenfux joined #git
04:52 thebope joined #git
04:52 phanimahesh joined #git
04:52 xameada joined #git
04:54 dsantiago joined #git
05:00 dermoth joined #git
05:00 ayogi joined #git
05:01 sea-gull joined #git
05:01 GT4066 joined #git
05:02 govg joined #git
05:02 msonntag joined #git
05:04 frobnic joined #git
05:05 romerocesar joined #git
05:09 bmahe joined #git
05:10 svm_invictvs joined #git
05:11 lindenle joined #git
05:13 anotheryou_ joined #git
05:13 shaggycat joined #git
05:14 mischat joined #git
05:17 hahuang65 joined #git
05:22 dan2k3k4 joined #git
05:31 _geoff joined #git
05:32 Karunamon joined #git
05:34 parsnip joined #git
05:36 BanaShip3hs joined #git
05:37 BanaShip3hs left #git
05:41 Karunamon joined #git
05:45 zeroed joined #git
05:46 spb joined #git
05:46 diogenese joined #git
05:46 Karunamon joined #git
05:46 robogoat joined #git
05:49 brokensyntax joined #git
05:49 qt-x joined #git
05:49 mure joined #git
05:50 dreiss joined #git
05:50 brokensyntax joined #git
05:51 Murii joined #git
05:52 ThomasLocke joined #git
05:52 ThomasLocke joined #git
05:52 me_and joined #git
05:53 thebope joined #git
05:54 diogenese joined #git
05:54 kaldoran joined #git
05:55 liuzhen joined #git
05:55 Anja joined #git
05:57 Raging_Hog joined #git
06:00 jaguarmagenta joined #git
06:01 LeBlaaanc joined #git
06:03 LucaTM joined #git
06:04 Repox joined #git
06:04 Karunamon joined #git
06:04 cdown joined #git
06:04 jaguarmagenta joined #git
06:04 subhojit777 joined #git
06:05 jaguarmagenta joined #git
06:08 ahmedelgabri joined #git
06:08 lb1c joined #git
06:09 jaguarma_ joined #git
06:09 cdown_ joined #git
06:09 fedepad joined #git
06:09 ahmedelgabri joined #git
06:10 rendar joined #git
06:10 Rhonda joined #git
06:11 rscata joined #git
06:12 chele joined #git
06:13 qt-x joined #git
06:14 gniourf joined #git
06:14 liuzhen joined #git
06:14 raspado joined #git
06:15 mischat joined #git
06:18 roelmonnens joined #git
06:20 aidalgol joined #git
06:21 jnavila joined #git
06:22 JeroenT joined #git
06:26 cdown joined #git
06:28 cdown joined #git
06:32 zulutango joined #git
06:32 freimatz joined #git
06:33 pitastrudl hello
06:33 pitastrudl should i have my gitignore file synced in the repo?
06:38 bocaneri joined #git
06:38 kadoban pitastrudl: Usually, yeah.
06:38 Wulf joined #git
06:39 bruce_lee joined #git
06:39 bruce_lee joined #git
06:39 pitastrudl i was wondering how LF and CLRF work, i get the warning "LF will be replaced by CRLF in git "
06:39 pitastrudl like will windows/unix care or will it become unreadable by either
06:39 cyan__ joined #git
06:41 bocaneri joined #git
06:42 bmahe joined #git
06:42 star_prone joined #git
06:43 gtristan joined #git
06:45 star_prone joined #git
06:46 ISmithers1 joined #git
06:46 cdown_ joined #git
06:47 elect joined #git
06:47 brokensyntax joined #git
06:50 qpdb joined #git
06:50 corsibu joined #git
06:51 cagedwisdom joined #git
06:51 cdown joined #git
06:52 fees joined #git
06:53 cdown_ joined #git
06:53 thebope joined #git
06:57 overlord_tm joined #git
06:57 OCram85 joined #git
07:01 JeroenT_ joined #git
07:03 xall joined #git
07:06 shirty joined #git
07:07 TomyWork joined #git
07:07 alphor joined #git
07:07 WayToDoor joined #git
07:08 Jameser joined #git
07:09 lordjancso joined #git
07:10 liuzhen joined #git
07:11 drbean joined #git
07:12 eduardas_m joined #git
07:12 harish_ joined #git
07:15 cdown_ joined #git
07:16 vuoto joined #git
07:20 cdown joined #git
07:22 _ikke_ pitastrudl: Depends on the program
07:22 _ikke_ pitastrudl: On linux, crlf might show up as two new lines
07:23 _ikke_ lf might not show as new line in windows at all (in notepad for example)
07:23 pitastrudl i see
07:24 pitastrudl will have to find a way on how to go about it i guess
07:24 pitastrudl im getting that warning when im trying to sync the gitignore file
07:24 pitastrudl i edited it in vim, on git bash in windows
07:24 osse pitastrudl: what does 'file .gitignore' say ?
07:25 pitastrudl ASCII text
07:25 osse that's it?
07:25 pitastrudl thats it
07:25 Haliucinas joined #git
07:26 osse hmm, open it in vim. what's does ':set ff?' say?
07:26 pitastrudl osse do i know u from somewhere, i feel like i've seen u on another irc channel
07:26 pitastrudl fileformat=unix
07:27 osse pitastrudl: what does 'git show HEAD:.gitignore | file -' say?
07:27 ahmedelgabri joined #git
07:27 pitastrudl fatal: Path '.gitignore' exists on disk, but not in 'HEAD'.
07:27 pitastrudl /dev/stdin: no read permission
07:27 pitastrudl huh
07:28 Balliad joined #git
07:28 elastix joined #git
07:28 osse oh it's a new file?
07:29 osse i'd commit it and then re-run that command
07:29 pitastrudl ok
07:29 osse I have a feeling some of git's warning messages say "foo will be replaced by bar" when it should've been the other way around.
07:30 pitastrudl git show HEAD:.gitignore | file -
07:30 pitastrudl /dev/stdin: ASCII text
07:30 osse or maybe repo settings mean it should be crlf and git is confused
07:30 osse ok, good.
07:30 pitastrudl why does it matter if its in ascii?
07:30 osse now remove the file. then git checkout HEAD -- .gitignore
07:30 cdown joined #git
07:31 osse It doesn't. in this case file will print either "ASCII text" or "ASCII text with CRLF terminators".
07:31 systemovich joined #git
07:31 jnewt joined #git
07:31 pitastrudl oh wait
07:31 pitastrudl remove it from where?
07:31 osse disk
07:31 osse rm .gitignore
07:31 osse so that git recreates it
07:31 pitastrudl oh, forgot to do that before i used "git checkout HEAD -- .gitignore"
07:31 pitastrudl so i should git reset?
07:31 osse it's fine
07:32 osse rm and do it again
07:32 eka joined #git
07:32 buddyq joined #git
07:32 cdown joined #git
07:33 pitastrudl ok i deleted the gitignore, and now?
07:33 pitastrudl how do i pull it back?
07:33 osse git checkout HEAD -- .gitignore
07:34 pitastrudl ohh
07:34 pitastrudl what did that line do?
07:34 johnny56 joined #git
07:34 osse as you put it, it "pulled it back"
07:34 pitastrudl well now it's ".gitignore: ASCII text, with CRLF line terminators"
07:34 pitastrudl so that's good i guess
07:34 osse aka. get it from HEAD and put in disk
07:34 pitastrudl cool
07:35 osse everything is good i think
07:35 osse \o/
07:35 pitastrudl wooo
07:35 pitastrudl :D
07:35 pitastrudl thank you
07:36 osse now gimme a karambit
07:36 pitastrudl nuuuuuuh
07:36 pitastrudl muh pixels
07:37 encod3 joined #git
07:37 maret joined #git
07:38 grift joined #git
07:39 cizra joined #git
07:42 GavinMagnus joined #git
07:43 cizra Hello! I used git filter-branch --subdirectory-filter to split out a subdirectory into a new repo. However, after cloning the filtered repo to a new one, it's still big. I used the one-liner solution (answer #2) from http://stackoverflow.com/questions/10622179/how-to-find-identify-large-files-commits-in-git-history to find large objects -- indeed, I do have binary junk in the history. How do I 1) figure out
07:44 cizra how my repo references it (these objects are not visible in "git log") and 2) get rid of them?
07:46 encod3 joined #git
07:47 King_Hual joined #git
07:47 King_Hual joined #git
07:48 moop joined #git
07:49 Alienpruts joined #git
07:50 qpdb joined #git
07:50 Endarked joined #git
07:50 grayjoc joined #git
07:51 brk_ joined #git
07:53 flaviodesousa joined #git
07:54 thebope joined #git
07:54 ahmedelgabri joined #git
07:55 djb-irc joined #git
07:56 anuxivm joined #git
07:57 hfr joined #git
07:57 brokensyntax joined #git
07:58 davimore_ joined #git
07:58 rominronin joined #git
07:58 anuxivm left #git
07:58 aswen joined #git
07:59 Pieplay joined #git
07:59 kurkale6ka joined #git
07:59 acetakwas joined #git
07:59 davimore__ joined #git
08:02 jknetl joined #git
08:02 luke_penn joined #git
08:03 rominronin joined #git
08:06 ahmedelgabri joined #git
08:07 tlaxkit joined #git
08:08 Limes_ joined #git
08:10 jfr joined #git
08:10 mikecmpbll joined #git
08:10 mischat joined #git
08:11 corsibu joined #git
08:12 nd joined #git
08:12 dan2k3k4 joined #git
08:13 _ADN_ joined #git
08:13 JeroenT joined #git
08:14 JeroenT joined #git
08:14 ki0 joined #git
08:15 govg joined #git
08:15 qws-user-1228 joined #git
08:15 djb-irc joined #git
08:20 cdown joined #git
08:20 halbomin joined #git
08:22 star_prone joined #git
08:24 JeroenT joined #git
08:24 sveinse joined #git
08:25 lss8 joined #git
08:25 sveinse Mercurial has a function to show incoming commits upstream without fetching/pulling (hg incoming). Is there an git equivalent of that?
08:26 kadoban sveinse: Not without fetching, no
08:26 corsibu joined #git
08:26 kadoban You could check that what you have is not the same as the remote, but that's all you can really see without fetching.
08:27 barteks2x joined #git
08:27 ayjay_t joined #git
08:28 marcogmonteiro joined #git
08:29 sveinse all right. this is on a bare mirror, can I then clone the mirror, fetch from (the first) origin and diff them somehow to show what has changed?
08:29 tvw joined #git
08:30 kadoban sveinse: I don't understand I don't think. You should be able to just fetch and then check what's different if you like? Must be missing something.
08:32 sveinse kadoban: We have a local bare mirror in house of an upstream repo, because we make medical validated SW and it must be controlled. So before I pull in an update from upstream, I need to know in advance what has changed (in any branch) before I can get the authorization to fetch the update.
08:33 buddyq joined #git
08:34 Serpent7776 joined #git
08:34 \du joined #git
08:34 \du joined #git
08:34 kadoban You can't even fetch without authorization of some kind? That sounds rather terrible. So you can do anything except mess with your "local bare mirror", I assume, right?
08:35 brokensyntax joined #git
08:35 sveinse I can do anything on my own computer to fetch and manipulate whatever I'd like
08:36 sveinse I just need to come up with some way of evaluating what changes there are between our mirror and the upstream repo
08:36 kadoban sveinse: Then just clone one of them, and add the other as a remote. Then fetch. Then you can compare branches fine
08:37 cdown_ joined #git
08:37 kadoban !lol should work fine, or gitk --all, both good first glances.
08:37 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
08:38 sveinse kadoban: right. Thanks
08:39 JeroenT joined #git
08:39 kadoban Any logs or diffs really are possible, you just need to use the right refs  origin/whatever otherremote/whatever, etc.
08:39 Wulf joined #git
08:41 a_thakur joined #git
08:42 chll_ joined #git
08:45 b100s joined #git
08:47 liuzhen joined #git
08:47 phanimahesh joined #git
08:47 serycjon joined #git
08:49 Muzer joined #git
08:49 acetakwas joined #git
08:50 troulouliou_div2 joined #git
08:50 osse fetching is quite harmless, and you have to do that at some point anyway
08:51 xall joined #git
08:51 skunkz joined #git
08:53 osse ohh
08:54 osse you need auth to just download stuff without using them ? :O
08:55 thebope joined #git
08:57 albel727 joined #git
08:57 JeroenT joined #git
08:59 mischat joined #git
09:01 JeroenT_ joined #git
09:05 sagerdearia joined #git
09:06 liuzhen joined #git
09:06 netj joined #git
09:07 nettoweb joined #git
09:08 _ikke_ Note that fetches do not update the local branches (hg fetch == git pull)
09:08 jbitdrop joined #git
09:08 _ikke_ So the changes you fetch are isolated from the local branches
09:08 jgrasser joined #git
09:09 kajnc joined #git
09:09 omegak joined #git
09:10 ochorocho__ joined #git
09:11 brk_ joined #git
09:15 JeroenT joined #git
09:15 lindenle joined #git
09:16 liuzhen joined #git
09:16 kajnc joined #git
09:18 achadwick joined #git
09:19 alcohol joined #git
09:20 govg joined #git
09:23 qpdb joined #git
09:27 drizztbsd joined #git
09:28 skyfaller joined #git
09:30 courrier joined #git
09:32 afuentes joined #git
09:32 ahr3n joined #git
09:33 drizztbsd joined #git
09:33 nowhereman joined #git
09:35 buddyq joined #git
09:35 drizztbsd joined #git
09:40 jstimm_ joined #git
09:44 TheHackOps joined #git
09:44 TheHackOps gpg: skipped "XXXXXXXXXXXX": secret key not available
09:44 TheHackOps Anyone ever fixed this pain in the butt
09:45 shaggycat joined #git
09:47 nd joined #git
09:48 drizztbsd joined #git
09:48 systemovich joined #git
09:50 drizztbsd joined #git
09:51 nikivi joined #git
09:51 PCatinean joined #git
09:51 tobiasvl TheHackOps: is your secret key available?
09:52 tobiasvl what are you trying to do, signing a commit?
09:52 drizztbsd joined #git
09:52 TheHackOps tobiasvl, yep
09:52 TheHackOps ok
09:52 TheHackOps whoops
09:53 TheHackOps wrong window
09:53 TheHackOps tobiasvl, Yes I have generated it with gpg2 --key-gen and verified it exists
09:54 TheHackOps kewl
09:54 TheHackOps Anyway back to the point, been googleing lots and everyone just says the same things, I have tried setting it all up using git configure
09:54 TheHackOps Seems to give the same error, am i missing something?
09:55 tobiasvl TheHackOps: what have you configured?
09:55 tobiasvl should be enough to set user.signingkey
09:55 tobiasvl also make sure $GNUPGHOME is set
09:55 liuzhen joined #git
09:55 TheHackOps AH
09:55 TheHackOps Maybe i missed that last one
09:56 thebope joined #git
09:57 star_prone joined #git
10:00 makinen joined #git
10:01 makinen git commit is going to add some files I don't want to be added. How do I prevent it?
10:01 makinen I probably added them by accident in first the place
10:02 qpdb joined #git
10:03 corsibu joined #git
10:03 JuliusN makinen: do you want to unstage staged files before committing?
10:04 JuliusN git reset filename
10:04 grawity and later, add them to .gitignore and commit that
10:04 JuliusN git reset without filename unstages all changes
10:05 makinen yea probably they're some temporary files by emacs
10:05 selckin 'git status' outputs tells you how
10:07 makinen I want to leave the files in the file system as they were
10:07 makinen but I don't want them to be committed or pushed to the origin
10:08 makinen I'm not sure if git reset is what I want to do
10:08 selckin it is
10:09 makinen ah yes it was, thx :)
10:10 JuliusN i have an another issue. yesterday i merged my branch with origin/master and resolved all conflicts, but today when i try to merge again i get (seemingly) the same conflicts. i was wondering if i may have made some common newbie mistake?
10:11 JuliusN yesterday i created a new branch before committing the merge
10:11 JuliusN this is the branch in which i'm currently trying to pull from master
10:12 Macaveli joined #git
10:13 rnsanchez joined #git
10:13 brk_ joined #git
10:14 BadSnappa joined #git
10:15 Jameser joined #git
10:19 jeanres joined #git
10:19 jeanres left #git
10:20 jeanre joined #git
10:20 jeanre hi all, we merged a branch and reverted it, now we cannot merge it again, it says there are no changes
10:20 jeanre but when I go into the branch I want to merge into master
10:20 jeanre the changes are there
10:20 jeanre I just can't do git merge [Branch]
10:21 netj joined #git
10:21 jnewt joined #git
10:22 tlaxkit joined #git
10:28 _ikke_ jeanre: a revert does not undo the merge, only the changes in that merge
10:29 _ikke_ jeanre: from git points of view, everyting from that branch has already been merged (only later reverted again)
10:29 alhariel joined #git
10:29 _ikke_ jeanre: Easiest is to revert the revert again
10:29 jeanre yeah
10:29 ShekharReddy joined #git
10:29 jeanre did that
10:29 jeanre but its messy
10:32 dan2k3k4 joined #git
10:33 corsibu joined #git
10:36 freimatz joined #git
10:39 anotheryou_ joined #git
10:40 ferr joined #git
10:41 JeroenT joined #git
10:41 nilg joined #git
10:42 Guest26 joined #git
10:44 Ryanar joined #git
10:45 tobiasvl jeanre: !faulty_merge
10:45 gitinfo jeanre: If you have published a merge commit that turned out to be faulty and you'd like to get rid of it, you're probably not going to have a good time. Here's the full story: https://raw.github.com/git/git/master/Documentation/howto/revert-a-faulty-merge.txt
10:45 tobiasvl maybe we should have a rerere trigger
10:45 jimi_ joined #git
10:46 tobiasvl JuliusN: but you might want man git rerere
10:46 gitinfo JuliusN: the git-rerere manpage is available at http://jk.gs/git-rerere.html
10:46 gniourf joined #git
10:47 JuliusN my coworker suggested rebase
10:47 NitroLux joined #git
10:47 dmc second time hearing about rerere in two days.. better see wth it is
10:47 dmc oic
10:48 rominronin joined #git
10:48 Alienpruts joined #git
10:49 tobiasvl JuliusN: sure, you can rebase, that's an alternative to merging
10:49 rominron_ joined #git
10:49 jeanres joined #git
10:49 tobiasvl re-reading your initial question I agree with your coworker, looks like you want to rebase
10:50 ieee joined #git
10:51 JuliusN thanks
10:53 JuliusN also he suggested that we should keep something he called 'research master', a branch that we should merge with origin/master regularly, and derive all our feature branches from the research branch
10:53 makinen left #git
10:54 tobiasvl sure, that's not an uncommon workflow – although usually this is called a "dev" branch, in case you run into it in online literature
10:54 tobiasvl !workflow
10:54 gitinfo Finding the right workflow for you is critical for the success of any SCM project.  Git is very flexible with respect to workflow.  See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.
10:57 thebope joined #git
10:59 soerenb_ joined #git
10:59 grift left #git
10:59 grift joined #git
11:04 xkr47_ left #git
11:05 xkr47 joined #git
11:05 CussBot joined #git
11:06 Jameser joined #git
11:08 roelmonn_ joined #git
11:08 irco joined #git
11:09 soerenb_ joined #git
11:09 roelmonn_ joined #git
11:11 mischat joined #git
11:11 rafalcpp joined #git
11:11 ochorocho__ joined #git
11:13 mischat_ joined #git
11:15 Jameser joined #git
11:17 Vinnie_win_o joined #git
11:19 soerenb_ joined #git
11:19 ocbtec joined #git
11:21 JeroenT joined #git
11:25 rominronin joined #git
11:25 SpeakerToMeat joined #git
11:25 fuchstronaut joined #git
11:27 mizu_no_oto joined #git
11:29 skyfaller joined #git
11:29 skyfaller joined #git
11:29 stfn joined #git
11:30 skunkz joined #git
11:30 ojdo joined #git
11:30 fuchstronaut joined #git
11:31 Alienpruts joined #git
11:31 umpc joined #git
11:32 Macaveli joined #git
11:32 qpdb joined #git
11:36 nuck joined #git
11:37 buddyq joined #git
11:38 rominronin joined #git
11:43 liuzhen joined #git
11:47 corsibu joined #git
11:48 Gonzo89 joined #git
11:48 brokensyntax joined #git
11:52 aupo joined #git
11:52 brk_ joined #git
11:53 aupo Hey, I have a question on the inner workings of git URLs if anyone could enlighten me.
11:53 aupo What does git do with this URL exactly: "git+ssh://git@git.rd.company.com:repo/package.git#branch"
11:53 aupo Most specifically the ":repo/package..." part, how is that interpreted and or how should it be interpreted?
11:54 canton7 "git+ssh://" isn't a valid git url, afaik
11:54 canton7 ':repo/package.git' is a file path, relative to 'git's home directory, to the repository to interact with
11:54 aupo Oh yeah, shit, I copied that from a very old file
11:54 canton7 but the '#branch' stuff is non-standard as well
11:55 aupo "ssh://git@git.rd.company.com:repo/package.git" this is what I should be asking, of course
11:55 canton7 that's also invalid
11:56 canton7 ssh:// urls interpret the stuff after ':' as a port
11:56 canton7 see 'git urls' in man git clone
11:56 gitinfo the git-clone manpage is available at http://jk.gs/git-clone.html
11:56 corsibu joined #git
11:56 aupo This one works, though... Althought with npm, so I guess I'm actually barking at the wrong tree here.
11:56 canton7 'git@git.rd.company.com:repo/package.git' would be valid, as would 'ssh://git@git.rd.company.com/~/repo/package.git' and 'ssh://git@git.rd.company.com:22/~/repo/package.git'
11:57 aupo Sorry to bother you folks, I guess I'll go bark at #node.js next :)
11:57 canton7 it would be really weird if npm had different url syntax to git :
11:57 aupo It most definitely seems to have.
11:57 thebope joined #git
11:58 canton7 https://docs.npmjs.com/files/package.json#git-urls-as-dependencies yeah, utterly different
11:58 aupo Myup..
11:59 aupo Guess I'll go read the source code to see, what they actually do with that.
12:02 aupo left #git
12:04 synthroid joined #git
12:04 rendar joined #git
12:06 johnmilton joined #git
12:07 BPL joined #git
12:07 BPL Is it possible to know what are the branches of a certain commit hash?
12:07 Endarked joined #git
12:08 _ikke_ git branch --contains <hash>
12:08 Seban joined #git
12:09 anonymuse joined #git
12:13 Shentino joined #git
12:14 clmsy joined #git
12:14 ExoUNX joined #git
12:15 fxer joined #git
12:16 Seban joined #git
12:16 t4nk712 joined #git
12:17 Eryn_1983_FL joined #git
12:20 skyfaller joined #git
12:25 re1_ joined #git
12:26 WayToDoor joined #git
12:26 Macaveli joined #git
12:29 brokensyntax joined #git
12:31 grayjoc joined #git
12:31 modlin joined #git
12:34 mizu_no_oto joined #git
12:36 dsdeiz joined #git
12:37 frobnic joined #git
12:38 davimore_ joined #git
12:38 Gamecubic joined #git
12:38 tjone270 joined #git
12:39 f3r70rr35f joined #git
12:39 buddyq joined #git
12:39 mischat joined #git
12:40 cjbrambo joined #git
12:41 chele I am reading the manual of gitolite and I see that you can specify a permissions rule for a specific ref relatively to refs/heads giving just the name of it ,e.g master, hotfix-bug111 which would mean refs/master or any branch. you can also specify a rule with an absolute path relative to refs/ e.g. by saying refs/admin/secret-path. what does this mean that the admin/secret-path branch is created under refs/ folder and not under refs/he
12:41 chele ads ?
12:41 AndrewAlexMac joined #git
12:41 ahmedelgabri joined #git
12:43 osse chele: it means that git does not consider it a branch. you won't see it's log when you do git log --branches. it doesn't appear in 'git branch' etc.
12:43 ij joined #git
12:44 PioneerAxon joined #git
12:44 traxo-xx joined #git
12:45 chele so there's is something else to handle such creation of 'it'. And what's this 'it'?
12:45 ij Can I do the show/log --format=medium + something else, without repeating the whole thing by hand?
12:45 chele osse maybe you can point out to an example ...
12:47 osse chele: refs that are in refs/heads are called branches. refs that are in refs/tags are called tags. other refs are just called refs.
12:47 JamesBaxter joined #git
12:47 osse not sure what the question is
12:48 chele how do you create a ref?
12:49 osse chele: if you're badass: echo somesha1value > .git/refs/admin/secret-path
12:49 osse chele: normal people use git update-ref
12:51 shinnya joined #git
12:54 xameada joined #git
12:55 gho5t joined #git
12:55 osse chele: e.g. git filter-branch uses git update-ref when it creates the backups in refs/original
12:57 concerti joined #git
12:58 eindoofus joined #git
12:59 jimi_ joined #git
12:59 thebope joined #git
12:59 ij Is git log just a commit range generator that "maps" the range over to git show?
13:00 joki joined #git
13:01 roelmonnens joined #git
13:01 nowhereman joined #git
13:01 tlaxkit joined #git
13:03 _ikke_ I don't think that's an accurate description
13:03 _ikke_ git show can show all sorts of objects
13:05 chele osse so you could have a ref to a refs/private-area/awesome-hacks/ (for instance) where you do dirty stuff and when finished your offer to some integration guy to take it from there and merge to master branch and thus avoiding the creation of a specific repo, right?
13:05 anonymuse joined #git
13:05 osse chele: yes
13:05 osse chele: that said, I'd rather create my own repo to avoid that :p
13:06 chele is not best practice or?
13:06 osse just seems messy to me
13:07 chele osse well haven't found a way to do it w/o installing say standalone gitlab.
13:07 chele cause forking is not in git
13:07 harish_ joined #git
13:08 cdg joined #git
13:09 sorenb_ joined #git
13:10 LBV_User joined #git
13:11 chele osse that or you could still create branches prefixing your username (uniq. dev id) and so this integration guy can merge from it..
13:11 osse chele: if you don't have a fancy github/gitlab-like thingy then maybe this is easier.
13:12 chele osse no I don't
13:12 osse I think gitolite can do it as well (ie. easy creation of forks) but I don't know holw
13:12 chele hm
13:13 chele yes i was exploring the configuration of gitolite and noticed the permission rules allow that.
13:13 canton7 gitolite can do that, yep
13:13 Eryn_1983_FL hi guys so i wrote a script to do my pushing and pull for releases in puppet repo
13:14 Eryn_1983_FL can you check it out real quick? if i am doing it right
13:14 Eryn_1983_FL fyi this is in my repo /home/THRACE/erin/puppet/manifests/release.txt
13:14 Eryn_1983_FL http://sprunge.us/JMFC
13:16 govg joined #git
13:16 Gsham joined #git
13:17 acetakwas joined #git
13:17 Noldorin joined #git
13:18 lindenle joined #git
13:18 Haliucinas joined #git
13:19 anotheryou__ joined #git
13:19 CEnnis91 joined #git
13:20 troulouliou_div2 joined #git
13:20 joki joined #git
13:23 mischat joined #git
13:25 sagerdearia joined #git
13:26 armin joined #git
13:26 selckin thats gonna fuck so much up if any of them fail
13:27 ayogi joined #git
13:29 Seban joined #git
13:31 serycjon left #git
13:34 chele osse would you state some reasons behind forking own repo vs having personal branching architecture?
13:34 star_prone joined #git
13:34 frobnic joined #git
13:36 buddyq joined #git
13:37 osse chele: my main reasons are that you can force-push as much as you like and that you don't pullute other people's repos
13:37 osse when they run 'git branch -a' or -r they'll see your crap
13:37 nettoweb joined #git
13:38 Dougie187 joined #git
13:38 corsibu joined #git
13:40 arooni joined #git
13:41 griffindy joined #git
13:42 BPL _ikke_: Tyvm, it works, i guess git branch --contains hash will return the local ones only, isn't?
13:42 Darcidride joined #git
13:44 navidr joined #git
13:44 _ikke_ yes, use -a to also include remote tracking branches
13:44 malcon joined #git
13:45 frobnic_ joined #git
13:46 EdwardIII left #git
13:46 anotheryou joined #git
13:47 cso joined #git
13:47 DeltaHeavy joined #git
13:48 frobnic joined #git
13:48 saml joined #git
13:49 yqt joined #git
13:50 grayjoc joined #git
13:50 joki joined #git
13:50 troulouliou_div2 joined #git
13:50 kajnc joined #git
13:51 cso left #git
13:52 mischat_ joined #git
13:52 sdothum joined #git
13:53 ash_workz joined #git
13:54 kajnc joined #git
13:55 mischat joined #git
13:55 nettoweb joined #git
13:55 boombatower joined #git
13:55 liuzhen joined #git
13:57 mischat_ joined #git
13:57 frobnic_ joined #git
13:57 mischat_ joined #git
13:57 jeffreylevesque joined #git
13:58 anonymuse joined #git
13:58 Gsham joined #git
14:00 zivester joined #git
14:00 raphink joined #git
14:01 VladGh joined #git
14:02 star_prone joined #git
14:02 Xethron joined #git
14:03 fahadash joined #git
14:04 sorenb_ joined #git
14:04 Gsham_ joined #git
14:04 b_jonas joined #git
14:05 sangy joined #git
14:05 b_jonas Hi. git help rev-parse documents all the magical syntax I can use in git when it expects an object, right?
14:06 b_jonas But I hear there's also some magical syntax for when git wants filenames. Where is that documented?
14:07 Eryn_1983_FL selckin:  yeah..  trying to figure out what to do about that
14:08 davimore__ joined #git
14:08 ResidentBiscuit joined #git
14:08 AaronMT joined #git
14:10 davimore joined #git
14:11 Eryn_1983_FL selckin:  so what do you suggest i do if i cant do a script?
14:11 PioneerAxon joined #git
14:11 selckin at least push only at the end
14:13 modlin joined #git
14:14 Eryn_1983_FL ok
14:14 matsaman joined #git
14:15 LocusAsaf joined #git
14:16 buddyq so, I just did a git push origin origin/my_branch and it said: * [new branch]      origin/Buddys_branch -> origin/Buddys_branch. But the branch already existed
14:16 LocusAsaf Let's say I have Two branches that I consider trunks, qa & dev which most people will branch from for normal development.  Is it possible to detect with 100% certainty which trunk the branch is from?
14:17 buddyq Then it says: Your branch is ahead of 'origin/Buddys_branch' by 2 commits.
14:18 patarr joined #git
14:18 thiago joined #git
14:18 mangolisk joined #git
14:18 thebloggu joined #git
14:19 nikivi joined #git
14:20 joki joined #git
14:22 osse buddyq: u don fukt up.
14:22 osse just kidding
14:23 osse buddyq: you probably have made branch called 'origin/Buddys_branch'. that's quite different from a branch called 'Buddys_branch' located at 'origin'
14:23 buddyq nah, I resolved it actually. Looks I got the ssh key thing worked out.
14:23 buddyq ah!
14:23 buddyq really??
14:23 buddyq maybe thats what happened...
14:23 fb| joined #git
14:23 phanimahesh yep. * [new branch]      origin/Buddys_branch -> origin/Buddys_branch was a hint
14:23 osse buddyq: do git for-each-ref | grep Buddys
14:23 buddyq How should I push my local brnach to github?
14:23 ^7heo Is my submodule directory (.git/module/<module>/) supposed to contain the files that are checked in my submodule?
14:24 _ikke_ buddyq: git push origin <branch>
14:24 LocusAsaf anyone know the answer to my question, or is it not possible?
14:24 buddyq i get 3 results
14:24 ^7heo or is it only supposed to contain what would normally be the .git folder of that submodule?
14:24 _ikke_ ^7heo: It's supposed to contain the contents of the .git folder for thos submodules
14:24 _ikke_ right
14:24 buddyq i did git push origin origin/Buddys_branch. Is that wrong?
14:24 phaleth joined #git
14:24 ^7heo How the fuck did my index content end up there then?!
14:24 ^7heo =/
14:24 ^7heo for some reason
14:24 ^7heo the index points there...
14:25 osse buddyq: yes.  git push origin Buddys_branch
14:25 buddyq They all look correct. I have refs/remotes/origin/Buddys_branch, refs/remotes/host/Buddys_branch, refs/remotes/Buddys_branch
14:25 buddyq ok
14:25 phanimahesh LocusAsaf: git only works with commit graphs. You can look at the history and make an educated guess, but that's all.
14:26 phanimahesh commits aren't coupled to branches, so you can never be certain on what branch a commit was made.
14:26 kajnc joined #git
14:26 buddyq ah okay. cool.... working now
14:26 LocusAsaf phanimahesh: Well I'm not focusing on the commits, I just want to branch to branch
14:26 ferr joined #git
14:26 LocusAsaf want to know*
14:27 phanimahesh LocusAsaf: A branch is just a text file in .git/refs with the id of commit.
14:27 phanimahesh So if commits don't track something, branches don't, either.
14:28 rominronin joined #git
14:28 thiago joined #git
14:28 LocusAsaf speaking in terms of the graph, is it possible to follow the path along until I find a branch that == $SOME_TRUNK ?
14:28 rominron_ joined #git
14:29 LocusAsaf programmatically of course
14:29 phanimahesh the graph is of commits, each pointing to their previous commit (or commits, in case of merges)
14:29 ^7heo _ikke_: ok, for the reference, my worktree in the .git/module/<module>/config was pointing at .git/module/<module>/ rather than at ./<module>
14:29 ^7heo _ikke_: that's how things fucked up.
14:30 phanimahesh you can follow it, but why would you want to do it?
14:30 osse LocusAsaf: to do what?
14:30 anonymuse joined #git
14:31 LocusAsaf phanimahesh: so if I keep following the previous commit and check the branch for each one I would eventually hit another branch, a trunk in my example git process?
14:31 nettoweb joined #git
14:31 Endarked joined #git
14:31 _ikke_ in git, we don't really use the word trunk
14:31 LocusAsaf osse: I want to determine which trunk a feature was branched from
14:31 _ikke_ It's all branches
14:32 ^7heo _ikke_: depends, if it's in German.
14:32 LocusAsaf _ikke_: can't I use the word if that's how I defined my git process for development
14:32 plasticboy joined #git
14:32 phanimahesh LocusAsaf: you first have to start at your "trunk" branch, follow back and mark commits as belonging to "trunk". Then trace feature branches backwards till you cross a commit in "trunk".
14:32 _ikke_ LocusAsaf: Not if it's confusing to other what you mean
14:32 perlpilot LocusAsaf: sounds like you want `git merge-base`
14:32 ^7heo _ikke_: i.e. trunken
14:32 Bysmyyr joined #git
14:33 osse LocusAsaf: find the earliest commit on that branch not on any of the trunks: (git rev-list --reverse branch --not trunk1 trunk2 ), then find which trunk its parent belongs to (git branch --contains foo~1)
14:34 eivarv joined #git
14:34 phanimahesh Or, you can do as osse suggests.
14:34 osse foo = first line of output of the rev-list commadn
14:34 LocusAsaf _ikke_: I'll run my scenario again in case you missed it, sorry for any confusion.  We use branches qa & dev and treat them as trunks.  Both branches current commit corresponds to a deployed version of the code.  All features are branched off one of the "trunks".  WHat I was asking is a programatic method to determine which trunk a feature was branched from
14:35 Bysmyyr hi, I have strange question. If I have file from somewhere git history(only that file). Then I have full git repo. Can I find which commit made that file look like that what I have now. in other words, can I find a last commit of loose file
14:35 LocusAsaf osse: I'll give that a try
14:35 osse Bysmyyr: that's so easy it's not even funny.
14:35 zacts joined #git
14:36 osse Bysmyyr: git hash-object -- file
14:36 Bloomberg joined #git
14:36 osse Bysmyyr: now you know the SHA1 of its blob. now recurse through history to find a commit that has that blob
14:36 JeroenT joined #git
14:37 Bloomberg joined #git
14:37 Bloomberg Hello Room!!
14:37 gitinfo Bloomberg: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
14:37 mischat joined #git
14:38 Bloomberg http://imgur.com/a/PYMhB
14:38 Bloomberg someopne know what means that symbol in my git?
14:38 Bloomberg doesn't working the autodeploy setting
14:38 osse Bysmyyr: hmm, something like.  blob=$(git hash-object file); git rev-list --all | while read hash; do if git ls-tree -r $hash | grep -q "blob $blob"; then echo "OMG $hash contains file!! \o/"; fi; done
14:39 _ikke_ Bloomberg: It's part of you shell prompt, no clue what it means (your font does not support the characters)
14:39 Darcidride_ joined #git
14:39 osse Bysmyyr: this assumes there is a commit in which the file is IDENTICAL. otherwise you'd have to diff and do some heuristics
14:39 Bysmyyr osse: ok, thats good, thanks
14:40 Bloomberg I have several sites in my server, there are setting with autodeploy, but just this site doesn't works the autodeploy and show this symbol
14:41 synthroid joined #git
14:41 converge joined #git
14:41 converge joined #git
14:41 brokensyntax joined #git
14:42 vuoto joined #git
14:43 thebope joined #git
14:43 Endarked joined #git
14:44 shgysk8zer0 joined #git
14:44 dreiss joined #git
14:45 mrwat joined #git
14:45 mischat_ joined #git
14:46 davimore_ joined #git
14:50 CheckDavid joined #git
14:50 IrishPrime joined #git
14:50 rominronin joined #git
14:51 IrishPrime joined #git
14:52 iuk joined #git
14:52 joki joined #git
14:52 raspado joined #git
14:52 isysd joined #git
14:53 e14 joined #git
14:53 overlord_tm joined #git
14:53 LocusAsaf osse: for the contains command, is my answer always going to be the last line of output?
14:54 iuk Hi! Assuming that the commit history is linear, how to find the nth successor of a commit? Is there a notation analogous to ^ but working in the opposite direction?
14:54 ahmedelgabri joined #git
14:54 canton7 iuk, no
14:55 canton7 git doesn't track the child of a commit
14:55 osse LocusAsaf: I don't know.
14:55 ShekharReddy joined #git
14:55 _ikke_ A commit can have an infinite number of children
14:56 iuk _ikke_: I’m OK iterating over all of them.
14:56 _ikke_ which will be created after that commit
14:56 LocusAsaf osse: following your instructions the contains command seems to print out several branches, not just one which is why I asked
14:56 matsaman ain't that the truth
14:56 osse LocusAsaf: a commit can "belong" to several branches
14:56 _ikke_ iuk: It's basically iterating over all commits and check the parents
14:57 iuk _ikke_: I just thought there was a shortcut for that.
14:57 re1 joined #git
14:57 _ikke_ nop
14:57 _ikke_ iuk: One assumption can be all commits that have a date later then the commit, but that assumption is easily broken
14:58 tlaxkit joined #git
14:58 iamthad joined #git
14:58 ochorocho__ joined #git
14:58 iuk _ikke_: I’m not looking for a general solution. I have a repo with linear history.
14:59 perlpilot iuk: if the history is linear, why do you want iterate in reverse?
14:59 jstimm_ joined #git
14:59 iuk perlpilot: It’s a pretty lengthy history. I know the hash of the commit I’m interested in, and I want to know what was right after it.
15:00 iuk Perhaps it’s possible to make `git log` scroll down to a particular commit?
15:01 Guest6802 joined #git
15:01 iuk I guess I should just dump the whole log into a text file and search.
15:01 perlpilot /SHA1-of-commit-I-want  :-)
15:01 iuk perlpilot: Yeah)
15:02 iuk perlpilot: Right! Right!
15:02 iuk perlpilot: Now I get it. Thanks!
15:02 perlpilot sure
15:02 JeroenT joined #git
15:03 leeN joined #git
15:03 LocusAsaf osse: okay, I think perhaps I unintentionally left out some important information from my scenario. So as I said the two features are qa & dev.  The piece I forgot was that dev is branched from qa so dev knows about commits in qa as well. Therefore when I followed your instructions and followed a feature of qa up and asked it what branches know about the qa, basically it hit qa, dev & features of both.  Anecdotally it seems like the right ans
15:03 LocusAsaf wer is the last branch output but i"m not sure
15:03 re1 joined #git
15:04 gtristan joined #git
15:04 Renard_Chase joined #git
15:06 dviola joined #git
15:06 Renard_Chase Is there a command that reverts all changes I that I commited today?
15:06 LocusAsaf according to git-branch documentation the default sort order is "Sort order defaults to sorting based on the full refname (including refs/... prefix)" but I'm not certain what that means
15:06 matsaman Renard_Chase: git reset --hard shasumtocommityouwant
15:06 subhojit777 joined #git
15:06 jwest joined #git
15:07 aarobc joined #git
15:07 jimi_ joined #git
15:08 Renard_Chase matsaman : Thank you
15:09 SrPx_ joined #git
15:10 jeanres joined #git
15:10 stomplee joined #git
15:14 patarr joined #git
15:14 barteks2x joined #git
15:15 danslo joined #git
15:20 al-damiri joined #git
15:20 Gsham joined #git
15:23 Jameser joined #git
15:24 timothy joined #git
15:28 crose joined #git
15:29 shirty left #git
15:29 ahmedelgabri joined #git
15:31 e14 joined #git
15:31 matsaman joined #git
15:32 diogenese joined #git
15:32 JamesBaxter joined #git
15:32 fuchstronaut joined #git
15:32 fuchstronaut joined #git
15:33 bernardio joined #git
15:33 rbern joined #git
15:35 maret joined #git
15:36 lindenle joined #git
15:37 Derperperd joined #git
15:38 eivarv joined #git
15:39 Ryanar joined #git
15:41 sdlnv_ joined #git
15:41 kajnc joined #git
15:43 star_prone joined #git
15:44 kpease joined #git
15:46 rnsanchez joined #git
15:48 danlovesppipes joined #git
15:50 danimal2 joined #git
15:51 danimal2 hi guys, question..if i do a git pull from a repo on github, then run composer update, will that update all code so any changes i might have made own my own, would be overwritten? i have not forked the repo or do PR's etc..
15:51 oskarkv joined #git
15:51 rominronin joined #git
15:52 patarr joined #git
15:52 danimal2 essentially there is some key code i want added in, that the author wont merge...and i want to keep my repo up to date with my changes and the author changes...except for my tweaks.
15:52 qpdb joined #git
15:53 alex1a_ joined #git
15:54 Learath2 joined #git
15:55 littleforest joined #git
15:55 synthroid joined #git
15:55 afuentes joined #git
15:56 synthroi_ joined #git
15:57 re1_ joined #git
15:57 shiriru joined #git
15:58 Gsham joined #git
15:58 einnjo joined #git
15:59 brokensyntax joined #git
16:00 pumafied joined #git
16:00 osse pull performs a merge
16:01 osse between whatever you pull and HEAD
16:02 ygeLN joined #git
16:02 danimal2 right, thanks osse. so best practice would be fork the repo..commit my changes, likely 1 time changes..then how does my fork stay up to date with the author's commits on github?
16:02 osse danimal2: you pull from the original and push to your fork
16:02 chamunks joined #git
16:02 osse if it's just one commit you need as a patch to make things work I suggest pull --rebase
16:03 osse that way your patch is always on top
16:03 xavca joined #git
16:03 danimal2 oh i get it..makes sense. thanks for that rebase tip.ill try that.
16:03 danimal2 osse++
16:03 osse !karma
16:03 gitinfo This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
16:03 osse .karma osse
16:03 gitinfo the Genuine Real Life Karma™ REST API results are back! o‍sse: 250 (3 in past 30 days) (6 given out)
16:03 osse awwww yisss
16:04 Remram joined #git
16:04 osse oh, it doesn't count nick++ ?
16:04 danimal2 not sure..im not an irc guru either
16:04 danimal2 i was trying to give you good karma
16:04 EvilPenguin joined #git
16:04 Erati joined #git
16:05 osse jast: plz fix
16:05 osse i need my karma
16:06 danimal2 thank you osse
16:06 danimal2 lol!
16:06 osse that should do it
16:06 ochorocho__ joined #git
16:06 Zarthus Does anyone know of a good resource for a 5-10 minute introduction to git. Something that doesn't explain it entirely, but does give them an introduction. Something like "homework you should look at before I tell you how to use it". I don't expect to advance beyond pull/push.
16:06 albel727 joined #git
16:07 Zarthus A video might be ideal, but I don't really care in what format it is in.,
16:07 osse Zarthus: any prior knowledge with VCSes?
16:07 Zarthus None, they're complete novices.
16:07 subhojit777 joined #git
16:07 Zarthus I plan to explain it to them in person, but won't see them until next week.
16:07 danimal2 Zarthus, im sure there's bazillions on YT..just my .02
16:08 mischat joined #git
16:08 madewokherd joined #git
16:08 Zarthus I looked at the GitHub youtube account and it's boring as hell.
16:08 osse !videos
16:08 gitinfo [!talks] Some good video talks about Git: [yt] http://goo.gl/z72s (Linus Torvalds: History&Concepts); [yt] http://goo.gl/R9H2q (Scott Chacon: Git basics, live examples); http://vimeo.com/35778382 (Randal Schwartz: Git basics, descriptional); http://vimeo.com/46010208 (Jesica Kerr: Git basics, descriptional)
16:08 jknetl joined #git
16:08 uofxj joined #git
16:08 Zarthus I bet the Linus Torvalds one is going to be fun
16:08 osse the first few minutes of the Jessica Kerr one are good imho
16:08 osse the whole is longer than 10 minutes though
16:09 Zarthus I don't really mind.
16:09 danimal2 Zatrthus,what helped me alot as a newb was Sourcetree if they are working on windows.
16:09 Zarthus I'm recommending they install Github For Desktop/SmartGit/Sourcetree, but that's kinda offtopic here.
16:11 Guest26 joined #git
16:11 osse I don't have much experience with any of those, but I do recommend that you stay as far away from tortoisegit as possible
16:11 _cyril_ joined #git
16:11 Zarthus yep, I had horrible experiences with that.
16:11 Zarthus SmartGit was how I got introduced to git, it's pretty neat -- until I broke things and someone else had to fix things.
16:12 Zarthus I love how Jessica Kerr is shitting on subversion
16:12 danimal2 yeah Tortoisegit i remember hosed me more than once...years ago when i fooled with it.
16:13 shakalaka joined #git
16:13 grayjoc joined #git
16:13 Zarthus I fear this video might still be a bit too technical.
16:14 miklcct joined #git
16:14 osse Zarthus: btw, she's given this talk in a more refiner manner later. but i like the whiteboard etc. it makes stuff more intuitive. I've personally had much success in writing "HEAD" and "master" on a couple of stickies to explain stuff
16:15 mischat joined #git
16:15 Zarthus They're telling us to start a 6 man project without any explanation of version control. And I'm not looking forward to using dropbox on a 6 man project.
16:15 osse Zarthus: the one by scott chacon is really good too.
16:16 _ikke_ Zarthus: Same when I went to school. Discovered version control ourselves
16:16 jeanres joined #git
16:16 Zarthus _ikke_: in week 6 of the first year? :P
16:16 _ikke_ Zarthus: probably
16:16 Zarthus Did you use any resources to teach them/yourself?
16:17 _ikke_ Started with SVN at the time, but for git, !books has good resources
16:17 gitinfo [!book] There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
16:17 _ikke_ and questions on stackoverflow
16:17 danimal2 Zarthus, i dont work as part of a team,just solo dev with my own repos..what made it clear for me is just make ur changes, test, when all good, Commit, then push to your remote repo if necessary.
16:18 _ikke_ !parable is good if you want to explain or learn the principles of version control
16:18 gitinfo 'The git parable' provides some good reasoning behind git.  http://tom.preston-werner.com/2009/05/19/the-git-parable.html
16:19 albel727 joined #git
16:19 davimore__ joined #git
16:20 danimal2 nice link, thank you _ikke_
16:21 Guest26 joined #git
16:22 Zarthus I've dumped some resources in our distributed dropbox. Hopefully they take the time to read/watch some of it
16:22 Zarthus thanks!
16:22 Shentino joined #git
16:22 EvilPenguin joined #git
16:23 _ikke_ yw
16:24 nitric joined #git
16:25 _geoff joined #git
16:25 Exagone313 Hi, I'd like to fork and maintain a project that uses git, but I'm not very familiar with. I wrote this scheme of local repo I could use http://dpaste.com/0VMN2QJ Is it bad? I don't know yet how would I succeed to do that but I'll search. Thanks for your help.
16:28 alfism joined #git
16:28 eshlox joined #git
16:29 cdg joined #git
16:31 jstimm joined #git
16:32 Gsham joined #git
16:32 JamesBaxter joined #git
16:32 LionsMane joined #git
16:33 [Brain] joined #git
16:34 hahuang65 joined #git
16:34 mischat_ joined #git
16:34 blackwind_123 joined #git
16:35 pks joined #git
16:35 Derperperd joined #git
16:37 robotroll joined #git
16:37 a_thakur joined #git
16:38 synthroid joined #git
16:38 manuel joined #git
16:38 svm_invictvs joined #git
16:40 LeBlaaanc joined #git
16:40 mischat joined #git
16:42 mischat joined #git
16:42 _August_ joined #git
16:42 nilg joined #git
16:42 brokensyntax joined #git
16:43 hahuang65 joined #git
16:45 _August_ I have a branch that im trying to merge the changes to another branch but im wondering how can I merge the change into another branch under a single commit instead of all the commits from the branch im merging from. So how can I take 5 commits and combine them into a single commit?
16:46 osse merge --squash
16:47 brokensyntax joined #git
16:47 _August_ so git merge --squash "merge title" ?
16:47 anonymuse joined #git
16:47 _August_ osse ^^
16:48 kulelu88 joined #git
16:48 osse where did you get that from?
16:48 osse write the merge liek you would otherwise, but add --squash to it
16:48 _August_ ohh ok
16:48 durham joined #git
16:49 overlord_tm joined #git
16:51 mischat_ joined #git
16:52 _cyril_ joined #git
16:52 LeBlaaanc joined #git
16:52 skunkz joined #git
16:53 Skryll joined #git
16:54 flavius joined #git
16:54 shakalaka joined #git
16:56 SteffanW joined #git
16:58 cyan__ joined #git
16:58 yohnnyjoe joined #git
16:59 foist joined #git
16:59 _August_ osse: Why when i try to push the merge it keeps saying everything up to date and doesnt push?
16:59 notebox joined #git
16:59 _August_ i pull everything to make sure it was current
17:00 SamuelAlderson joined #git
17:00 _August_ wait... when i squash everything do i have to recommit them?
17:01 X-Kent joined #git
17:01 roelmonnens joined #git
17:02 X-Kent hi, I have checked out, master, then did "git checkout remotes/origin/name_of_branch". Now I am in "detached" state and it didn't seem to update anything, how do I switch a remote branch ?
17:02 osse _August_: yes, you have to run git commit after
17:02 osse X-Kent: !detached
17:02 gitinfo X-Kent: A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach
17:02 _August_ Got it, thanks
17:03 X-Kent !reattach
17:03 gitinfo Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue
17:03 SamuelAlderson left #git
17:03 osse X-Kent: if you only wanted to checkout the remote branch to look at it, test it, compile it, whatever, then being detached is completely fine
17:03 osse if you want to work on it, do the checkout -b thing
17:04 X-Kent osse, nope, I did my code on the master and now I need to deliver it to another branch, I can't figure out how to get anything pointing to that remote branch so I can merge and commit
17:05 X-Kent osse, I tried checking out master again and then branching but it became detached
17:05 Gsham joined #git
17:05 osse X-Kent: to update a remote branch you push
17:06 X-Kent osse, yea I mean commit and then push, but I can't get a working copy that has that branch as a remote
17:06 albel727 joined #git
17:06 osse then you did something wrong
17:06 osse !repro
17:06 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
17:07 X-Kent is there a way to just checkout that new branch ?
17:07 tyreld joined #git
17:07 osse which new branch?
17:07 osse the remote one? no.
17:08 dendazen joined #git
17:08 X-Kent osse, when I do "branch -a" in my master I see it as "remote/originin/branch_name"
17:08 albel727 joined #git
17:08 albel727 joined #git
17:08 X-Kent and git shows that my current is "remotes/origin/HEAD"
17:09 X-Kent how do I switch it to "remote/origin/branch_name" ?
17:09 osse what happened when you tried to do what you were recommended ?
17:09 X-Kent I tried "git branch remote/origin/branch_name", now it says it's detached and I can't pull/push
17:10 osse you need to create a LOCAL branch named 'branch_name'
17:10 thiago joined #git
17:10 osse which git checkout -b branch_name will do if you have checked out remotes/blablba previously
17:11 X-Kent I didn't checked out remotes/blala, always worked on master
17:11 mikecmpbll joined #git
17:11 osse the first thing you said when you join was that you did
17:11 X-Kent so, in my master I do "get checkout -b branch_name" and then pull ?
17:12 CalimeroTeknik joined #git
17:13 osse X-Kent: "I have checked out, master, then did "git checkout remotes/origin/name_of_branch". Now I am in "detached" state"
17:13 osse is this still true?
17:13 X-Kent osse, I already deleted that and checkout again clean master
17:13 osse git checkout -b branch_name origin/branch_name
17:14 anonymuse joined #git
17:14 X-Kent just a sec
17:14 lus joined #git
17:15 durham_ joined #git
17:17 brokensyntax joined #git
17:17 X-Kent osse, it says "Branch branch_name set up to track remote branch branch_name from origin." and "Switched to a new branch 'branch_name'"
17:17 osse fantastic
17:17 X-Kent osse, I can now pull ?
17:17 osse now you have a local branch you can work in, and if you run 'git push' stuff gets push to the remote equivalent
17:17 durham_ joined #git
17:17 osse X-Kent: yes
17:17 X-Kent osse, that's perfect, thank you !
17:17 Alienpruts joined #git
17:18 X-Kent osse, git is really hard when you are used to svn
17:19 osse X-Kent: since you've already done work in master you can copy that over to this branch by using cherry-pick
17:19 synthroid joined #git
17:19 X-Kent osse, I will just run "meld" and merge it there manually :-)
17:20 dreiss joined #git
17:20 joki joined #git
17:21 b_jonas If I run (git diff --cached -- foo) then the foo is relative to the working directory, but in (git diff --cached -- :/foo) the foo is relative to the repository root.
17:21 b_jonas Where is that syntax documented in the git manpages?
17:21 b_jonas I know that git help rev-parse documents all the syntax for arguments that denote objects, but this is a filename argument.
17:22 b_jonas I'd like to know what all the syntax for filename arguments is.
17:22 pks joined #git
17:22 rominronin joined #git
17:23 osse b_jonas: man gitglossary, under 'pathspec'
17:23 gitinfo b_jonas: the gitglossary manpage is available at http://jk.gs/gitglossary.html
17:23 osse (don't blame me. I didn't put it there)
17:24 buddyq joined #git
17:24 raphink joined #git
17:24 b_jonas osse: thank you, that's where it is!
17:24 b_jonas I'll have to read that carefully
17:24 b_jonas I must admit, I don't really like this magic syntax, because :/ means something completely different as a revision argument
17:24 osse b_jonas: very. it's nearly incomprehensible :p
17:25 wget joined #git
17:25 wget joined #git
17:25 b_jonas so without the -- it gets utterly ambiguous
17:25 cdown joined #git
17:26 EvilPenguin joined #git
17:27 buddyq joined #git
17:28 chrisgeorge joined #git
17:28 shakalaka joined #git
17:28 albel727 joined #git
17:30 stuh84 joined #git
17:30 e14 joined #git
17:31 durham_ joined #git
17:32 durham_ joined #git
17:32 LocusAsaf joined #git
17:32 Guest6802 joined #git
17:33 LocusAsaf osse: just wanted to let you know that I worked out a reasonable way to solve my problem. I'm walking down the output of revlist until I hit the first commit hash that equals the hash of my trunk heads.
17:34 encod3 joined #git
17:35 LocusAsaf also, I appreciate the help before
17:35 davimore joined #git
17:35 kireevco_ joined #git
17:36 tang^ joined #git
17:36 iosnerd joined #git
17:36 maxxe joined #git
17:37 JamesBaxter joined #git
17:37 kireevco joined #git
17:37 cagedwisdom joined #git
17:38 Remram joined #git
17:40 a_thakur joined #git
17:40 kajnc joined #git
17:40 duderonomy joined #git
17:40 diogenese joined #git
17:41 ronny hi
17:41 ronny is there a poreclain command that tells me of a git repo is shallow?
17:41 patarr joined #git
17:42 guampa joined #git
17:43 cbreak ronny: does git config count? :)
17:44 nettoweb joined #git
17:44 PCatinean joined #git
17:44 iosnerd joined #git
17:45 cbreak hmm... I guess you can just look at .git/shallow
17:45 raspado_ joined #git
17:45 ronny cbreak: i want to provide an option to user to deide what to do in shallow clones (i.e. fail auto-unshallow or warn that data is incorrect)
17:45 WakiMiko joined #git
17:47 anonymuse joined #git
17:50 EvilPenguin joined #git
17:51 Sasazuka joined #git
17:51 shakalaka joined #git
17:52 pothepanda joined #git
17:53 a_thakur joined #git
17:54 davimore_ joined #git
17:54 jimi_ joined #git
17:54 dsantiago joined #git
17:56 PCatinean joined #git
17:56 dsantiag_ joined #git
17:57 thiago joined #git
17:57 albel727 joined #git
18:00 duderono_ joined #git
18:01 nettoweb joined #git
18:03 mountpoint joined #git
18:05 converge joined #git
18:07 hahuang61 joined #git
18:07 albel727 joined #git
18:08 saml joined #git
18:09 khfeng joined #git
18:10 hypercube32 joined #git
18:11 BSaboia joined #git
18:12 artisanIndia joined #git
18:13 artisanIndia I have committed some files and then made changes in all of them and then I added some of them into gitignore, now if I do a git reset --soft will that discard the changes as well ?
18:13 osse LocusAsaf: Sounds reasonable
18:14 fuchstronaut joined #git
18:14 WayToDoor joined #git
18:17 mibr0 joined #git
18:18 artisanIndia how do I ignore permission changes from the push ?
18:18 eyou joined #git
18:18 Chloe left #git
18:18 fedepad joined #git
18:21 einnjo joined #git
18:22 vannorden joined #git
18:22 btyne79 joined #git
18:24 Darren_ joined #git
18:24 albel727 joined #git
18:24 jstimm joined #git
18:26 canton7 artisanIndia, git doesn't track permissions (except the execute bit)
18:26 fahadash joined #git
18:26 vuoto joined #git
18:26 artisanIndia ok
18:27 Exagone313 Hi, I'd like to fork and maintain a "mod" of a game, that has a repo on Github.  They have multiple branches and I want to clone/pull their branch master (this is not a problem), let's say into the branch "original" (not master).  Then, if what I get is compilable, I want to merge with my changes in the local branch named "dev".  When I get something compilable in dev, I want to merge into my local "master".  At any time, I want to push my branches dev and
18:27 Exagone313 master (if possible, not original branch)  into *another* distant repo.  Also, the original repo uses two submodules, and I'd like to keep them in my other branches dev and master, and if it's possible, to not clone them too many times (I think it can track it is the same commit for submodules, so it would keep the submodule directories when changing the current working branch).  Is there something wrong in my idea?  Can you help me with that?  Thanks for your
18:27 Exagone313 help.
18:28 hariel joined #git
18:28 moritz Exagone313: which part do you need help with?
18:29 monomon[m] joined #git
18:29 sea-gull joined #git
18:29 ygeLN joined #git
18:30 ToBeFree joined #git
18:31 Exagone313 basically what I currently know and use is clone/pull/add/commit/push.  I can read more docs about it (also about merging branches) on my side.  What I don't know is if I can pull into original the code I fork, work on dev, and push dev/master into another repo of mine (not on github)
18:31 ronny cbreak: checking jsut for the file was perfect, thanks
18:32 ronny left #git
18:34 emptynine joined #git
18:34 texasmynsted joined #git
18:35 moritz Exagone313: you can add the original repo as a remote, let's say with name upstream
18:35 moritz Exagone313: then you can git fetch upstream; git merge upstream/master
18:36 cagedwisdom joined #git
18:36 cyan__ joined #git
18:38 overo joined #git
18:39 _August_ why would i get a not something we can merge
18:39 _August_ error?
18:40 Groscheri joined #git
18:40 overo hi! Brief question: I'm in branch MyFeatureA. If I 'git add' new updated files for a work-in-progress feature, then I 'git commit' them and then I 'git checkout -b MyFeatureA-experiment', have I committed the new updated files to my new (local) branch MyFeatureA-experiment?
18:41 lungaro minor confusion w/ git config. Is there a wayto use --list w/ a pattern prefix or section?
18:41 albel727 joined #git
18:41 Exagone313 so, something like `git remote add -t master upstream url` then when I'm in branch original, `git fetch upstream; git merge upstream/master`.  What do I do to set origin only for my dev and master to be able to push? Does original have to be pushed too?
18:41 overo or should I had 'git checkout -b newbranch' just before 'git commit' ?
18:41 Exagone313 moritz:
18:41 nikivi joined #git
18:41 alex1a joined #git
18:42 _August_ nvm i got it
18:46 Exagone313 I think it's dumb, just have to push everything to my own distant repo
18:46 QwertyDragon joined #git
18:47 mountpoint joined #git
18:49 mountpoint joined #git
18:52 abec0 joined #git
18:52 mountpoint joined #git
18:53 Repox joined #git
18:53 e14 joined #git
18:54 rominronin joined #git
18:55 mountpoint joined #git
18:57 crayon joined #git
18:58 ronny joined #git
18:58 ronny hi
18:58 ronny another question, whats the most reliable poreclain command to figure the current branch?
19:00 mablae joined #git
19:00 Rodya_ joined #git
19:01 mischat joined #git
19:03 moritz git branch, git status
19:06 moritz if you meant plumbing, git rev-parse --symbolic-full-name --abbrev-ref HEAD
19:06 khfeng joined #git
19:07 eivarv joined #git
19:09 Seveas ronny: and if scripting: do make sure you handle detached heads (== no current branch) properly :)
19:10 maret joined #git
19:11 diogenese joined #git
19:11 Overand joined #git
19:12 nilg joined #git
19:12 Overand Is there a reasonable way for me to get "git" to default to a particular remote, i.e. "git clone SomeNonsense" doing "git clone https://github.com/Overand/SomeNonsense" ?  (Or allowing 'git clone Overand/SomeNonsense' ?)
19:13 Overand or am I just better off writing a little alias "ghub" that does e.g. "git clone https://github.com/$@
19:13 Joel joined #git
19:13 Joel I have two copies of a repository, are there any gui (I know, I know) tools, which will walk them, and graph out the commit history, showing where they intersect and don't?
19:14 Overand Ouch! That sounds like a fun problem.
19:14 Overand I'm curious what the answers will be
19:14 aarobc joined #git
19:15 johnnyfive joined #git
19:15 kadoban Joel: Fetch one into the other and use 'gitk --all' ?
19:15 kajnc joined #git
19:15 fuchstronaut joined #git
19:15 star_prone joined #git
19:15 kadoban Overand: I have an SSH alias for github, so I just do 'git clone github:kadoban/whatever' for example.
19:16 kadoban You can also have rewrite rules on URLs in git, you'd have to look in the config for the syntax though, I don't remember it.
19:17 dreiss joined #git
19:19 m0viefreak joined #git
19:20 isysd joined #git
19:20 _ADN_ joined #git
19:22 nidr0x joined #git
19:22 [ADN] joined #git
19:25 Endarked joined #git
19:25 menip joined #git
19:30 le_melomane joined #git
19:32 crayon joined #git
19:32 crayon joined #git
19:35 duderonomy joined #git
19:36 zeroed joined #git
19:36 zeroed joined #git
19:38 Haliucinas joined #git
19:38 _ikke_ Joel: Easiest is to let git do the work
19:38 _ikke_ either add one as a remote to the other, and fetch from it, or create a new repository and fetch both in there
19:39 johnmilton joined #git
19:44 elastix joined #git
19:45 davimore joined #git
19:46 nikivi joined #git
19:46 Joel _ikke_, I'm trying to figure out why I'm getting unexpected behavior
19:46 dreiss joined #git
19:46 eivarv joined #git
19:48 overo hey all, I committed a feature to the wrong branch. Is there a way to move the commit onto the right branch?
19:48 ^7heo not "move" per se no
19:49 overo oh I see
19:49 ^7heo I mean, depends if you pushed
19:49 overo I could 1. checkout on the commit 2. push it to the right branch too 3. revert the commit from the wrong branch, maybe?
19:49 ^7heo nah wait
19:49 overo k
19:49 ^7heo go on the branch where you want the commit
19:49 overo k
19:50 ^7heo cherry-pick it
19:50 modlin joined #git
19:50 overo git cherry-pick ? Do I need any other parameter?
19:50 Guest26_ joined #git
19:51 ^7heo the sha of the commit
19:51 ^7heo like
19:51 ^7heo git log -n1 --online branch-where-the-commit-is
19:51 ^7heo will show you the sha
19:51 ^7heo or the shorthand
19:51 overo oh by the way, it's still in local repo
19:51 overo do I need --online anyway?
19:52 ^7heo no
19:52 overo I now have the sha
19:52 ^7heo if it's just in local it's gonna be very easy
19:52 ^7heo so
19:53 ^7heo go to the branch where you want it
19:53 overo ok
19:53 overo I am on it
19:53 ^7heo git cherry-pick $sha
19:53 ^7heo then you have the branch as you want, right?
19:54 overo uhm, it doesn't autocomplete the sha, is that a bad sign?
19:54 ^7heo (and we'll clean the other branch right after)
19:54 ^7heo depends on your shell
19:54 overo autocompletion shows branches by name
19:54 overo I'm on bash
19:54 ^7heo I never use bash
19:54 ^7heo I dunno
19:54 ^7heo try
19:54 ^7heo at worse, the cherry pick will complain and abort
19:55 overo error: could not apply 306fe52...
19:55 overo indeed :D
19:55 ^7heo damn
19:55 ^7heo okay
19:55 overo oh my
19:55 ^7heo I'm sorry I do other things
19:55 overo now I have the commit in the right branch too, though :D
19:55 ^7heo ok?
19:55 ^7heo do a `git status` to check all is in order
19:56 khfeng joined #git
19:56 overo there are two changes to be committed (green) and one unmerged path (red)
19:56 overo but the new file is being recognized (the one I committed to the wrong branch)
19:56 overo recognized as waiting to be committed, and I'm on the right branch now, I mean
19:57 overo can I 'git commit -m "stuff"'?
19:57 converge joined #git
19:57 ^7heo overo: sorry I'm really in a rush
19:58 overo okok
19:58 finalbeta joined #git
19:58 overo it seems it works though
19:58 overo tyvm
19:58 ^7heo overo: I gave you the complex instructions, you might have been ok with a merge or rebase
19:58 ^7heo cherry-pick isn't the easiest
19:58 ^7heo but
19:58 ^7heo you want to resolve the delta
19:58 overo git-scm.com good for learning?
19:58 ^7heo and call `git cherry-pick --continue`
19:58 ^7heo yeah I thnink
19:58 overo uhmkay, I'll note that down
19:58 overo "resolve the delta, ..."
19:59 cdg joined #git
19:59 osse !eek
19:59 gitinfo [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, *don't panic*. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice! A nice video explaining merge conflicts: https://www.youtube.com/watch?v=zz7NuSCH6II
19:59 osse overo: run 'git diff
19:59 osse some parts of the file have <<<< ===== >>>> in them
19:59 osse fix it so it's how it's supposed to be
19:59 chrisgeorge_ joined #git
20:00 overo osse: I've committed featureB to branchA, instead of branchB and I wanted to "move" the commit from branchA to branchB
20:00 patarr joined #git
20:02 overo and it worked as suggested by ^7heo !!!
20:03 ^7heo overo: btw
20:03 overo I mean, I don't know if the commit have been moved around, but now the branchA has only featureA, and the branchB has only featureB !
20:03 ^7heo when you want to remove the commit from the other branch
20:03 ^7heo if it's on the tip of the branch
20:03 ^7heo git reset --hard HEAD~
20:03 ^7heo if it has commits on top
20:03 cdown joined #git
20:03 ^7heo git rebase -i $SHA~
20:03 _August_ joined #git
20:03 overo HEAD~ is the parent of current HEAD?
20:03 ^7heo then edit the file and 'drop' the commit you don't want
20:03 ^7heo overo: yes
20:04 EvilPeng1 joined #git
20:04 jknetl joined #git
20:04 TomyLobo joined #git
20:04 ronny Seveas: bascially i want to know the branch im on, and its ok to assume master in case of detached
20:06 Seveas ronny: git symbolic-ref HEAD 2>/dev/null || echo 'master'
20:06 fb|
20:10 jstimm joined #git
20:14 Ryanar joined #git
20:16 nikivi joined #git
20:16 fuchstronaut joined #git
20:19 nilez joined #git
20:23 Groscheri joined #git
20:24 dsteplight__ joined #git
20:25 rominronin joined #git
20:27 nilez joined #git
20:27 Balliad joined #git
20:28 dbkaplun joined #git
20:28 dbkaplun i have a long list of files that were once checked in but are now removed/renamed. how do i find what commit removed/renamed them?
20:28 star_prone joined #git
20:29 davimore_ joined #git
20:30 _ikke_ git log --diff-filter=D --all <file>
20:31 iosnerd_ joined #git
20:33 hashpuppy joined #git
20:36 khfeng joined #git
20:36 thiago joined #git
20:38 garzer joined #git
20:39 aarobc joined #git
20:40 gesker joined #git
20:44 heroux joined #git
20:44 ahr3n joined #git
20:46 nilez joined #git
20:46 Rodya_ joined #git
20:47 patarr joined #git
20:47 Karazhan joined #git
20:48 davimore__ joined #git
20:49 aafu joined #git
20:49 dbkaplun thanks
20:49 EricPoe joined #git
20:50 khfeng joined #git
20:51 aidalgol joined #git
20:52 jast osse: ++ is counted
20:53 andlabs joined #git
20:54 andlabs Hi. A few weeks ago my coworker changed some lines in a file, and those changed lines don't show up in my local branch. But git doesn't think they changed, so pushing my own changes - even to that file - don't overwrite those older changes, and a fresh clone reveals the correct lines. What's going on?
20:54 dodobrain joined #git
20:55 moritz andlabs: maybe your coworker didn't push his changes? or to a different branch?
20:58 andlabs ...and suddenly they show up NOW
20:58 andlabs I have no idea; sorry for the noise
21:00 LeBlaaanc joined #git
21:10 ilbot3 joined #git
21:10 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
21:15 khfeng joined #git
21:16 overlord_tm joined #git
21:16 anonymuse joined #git
21:17 overlord_tm joined #git
21:18 garzer joined #git
21:18 ayjay_t joined #git
21:19 overlord_tm joined #git
21:19 yohnnyjoe joined #git
21:23 netj joined #git
21:24 peterR`` joined #git
21:24 ffffffff11111111 joined #git
21:25 nwkj86 joined #git
21:26 Karazhan joined #git
21:27 HoierM joined #git
21:29 mischat joined #git
21:29 GavinMagnus left #git
21:30 mischat joined #git
21:30 ffffffff11111111 hi, could you make "make test" honor NO_PERL ? TIA
21:32 EvilPenguin joined #git
21:34 l4v2 joined #git
21:35 anuxivm joined #git
21:37 e14 joined #git
21:41 _ikke_ ffffffff11111111: The !mailing_list is a better revenue for such requests
21:41 gitinfo ffffffff11111111: The mailing list can be reached via git@vger.kernel.org. You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist
21:41 _ikke_ s/revenue/venue/
21:42 manuel joined #git
21:43 Seveas ffffffff11111111: no, it cannot.
21:43 Seveas core test functionality relies on perl
21:46 ffffffff11111111 _ikke_, thank you for info ; I'm patching the test suite ... Seveas, thats not true ; I'm at t5150-request-pull.sh already - it took me a few hours so far and 16 simple patches - "sed" and "tr"; "tac", "xxd" and "dd" on a single occasion each
21:46 nezZario joined #git
21:47 ffffffff11111111 I can easily use a few lines of c code utils or "awk" in all cases
21:47 raspado joined #git
21:47 NeverDie joined #git
21:47 nezZario Hello gentlemen.. I am wondering, I set my push.default option to 'current'... Why is it still telling me 'There is no tracking information for the current branch...' ?  I thought the point of push.default current was that it always pushed/pulled from a branch with the same name on the remote.
21:48 xall joined #git
21:48 nezZario That is, if i'm on a branch called 'develop' then I should 'git pull' from the origin/develop and 'git push' to the origin/develop
21:49 dmc shouldn't this include git:// https://git-scm.com/book/en/v2/Git-Internals-Transfer-Protocols
21:50 Seveas ffffffff11111111: do note that perl is used for constructs where writing portable sed equivalents is not possible (e.g. the ones in test-lib-functions.sh)
21:50 misterpink joined #git
21:50 iosnerd joined #git
21:51 Seveas and tac and xxd are not available everywhere and/or portable
21:51 raijin joined #git
21:52 nezZario Are we talking about the best portable scripting language?  Because there really isn't one.
21:52 _ikke_ nezZario: We aren't
21:53 nezZario i think perl is probably the most widely distributed language that actually has a tit of power
21:53 Seveas nezZario: no, just why some parts of the testsuite rely on perl for portability reasons
21:53 nezZario anyone know what's up with the push.default question?  i'm just tired of having to --set-upstream-to=....
21:53 ffffffff11111111 Seveas, the test suite itself uses "tr" and "sed" - adding "perl" is kind of pointless to me. Using "perl" for the sake of it is kind of sad :). See, "nul_to_q ()" @ "test-lib-functions.sh" uses "perl", a few lines later "q_to_cr ()" uses "tr" - inconsistent in my POV. Anyway, I'll listen to the mailing list advice. 10x for your time
21:53 nezZario my git workflow is simple.  i want to push/pull one branch at a time with the exact same name on origin
21:54 Seveas ffffffff11111111: that's not using it for the sake of it. That one specifically is impossible to do with sed, as not all sed's handle NULL bytes properly
21:54 Seveas same with handling cr and lf
21:54 ffffffff11111111 Seveas, I see your point. I'll try and make a consisten patch. Thank you.
21:55 ffffffff11111111 *consistent
21:55 Seveas ffffffff11111111: apparently you don't as you still think it's possible :)
21:55 Seveas If it were, we would have gotten rid of perl in the testsuite a long time ago
21:55 _ikke_ There are smart people working on git
21:55 _ikke_ right
21:56 Seveas nezZario: current means 'current branch, if and only if it's actually configured with an upstream'
21:56 _ikke_ Seveas: That's not clear from the docs at least
21:56 rominronin joined #git
21:57 Seveas it's also wrong. I keep mixing up current and upstream. Sigh :)
21:58 elsevero joined #git
21:59 _ikke_ nezZario: It works for me
21:59 Electrometro joined #git
22:00 _ikke_ nezZario: https://gist.github.com/Ikke/2efc37dd77af47939e0af5a445fa47c9
22:01 EvilPenguin joined #git
22:01 thebope joined #git
22:02 nowhereman joined #git
22:02 _ikke_ g2g
22:03 f3r70rr35f joined #git
22:04 Seveas ffffffff11111111: https://github.com/git/git/commit/f8fc0ee314d0e0c993d2edac8848a4cfe72c771f
22:04 nikivi joined #git
22:06 ISmithers joined #git
22:09 ffffffff11111111 _ikke_, I have no doubt there are smart people working on git. Then again, the people who don't make mistakes don't exist yet IMHO as well as the ones who wrote the perfect code that can not be improved. Seveas, I need to complete my patch in order to observe my ignorance. Thank you both for your time. Back to the coding table.
22:10 mikecmpbll what'd be the best way to get the tag before last?
22:10 ffffffff11111111 left #git
22:11 e14 joined #git
22:12 Seveas depends on your definition of 'before' and 'last'
22:12 Seveas for before: temporally, topologically, or sorted as version numbers.
22:12 Seveas For last: most recent in time, closest reachable from HEAD or maybe something else
22:13 mikecmpbll good point. temporally or topologically
22:14 mikecmpbll the same way that git describe --tags get's the 'latest'
22:15 mikecmpbll "the most recent tag that is reachable from a commit."
22:15 LandonJH joined #git
22:15 Seveas topologically, with actually some changes in more recent gits
22:15 Seveas giving side branches less of a priority
22:15 jstimm joined #git
22:17 Seveas dodgy hack actually using git-describe: git describe $(git describe | sed -e 's/-[0-9]\+-g[0-9a-f]\+$//')^ | sed -e 's/-[0-9]\+-g[0-9a-f]\+$//'
22:18 mikecmpbll cor blimey
22:18 Seveas much less dodgy (yay manpages): git describe --abbrev=0 $(git describe --abbrev=0)^
22:18 wgrant joined #git
22:19 safe joined #git
22:20 mikecmpbll Seveas : <33
22:20 mikecmpbll duh, why didn't i think of that.
22:21 chris_ joined #git
22:22 Wulf joined #git
22:22 Darren_ joined #git
22:23 patarr joined #git
22:26 kulelu88 joined #git
22:27 courrier joined #git
22:30 anotheryou_ joined #git
22:30 cdown_ joined #git
22:30 mischat joined #git
22:33 codfection joined #git
22:35 anuxivm left #git
22:36 codfection joined #git
22:36 EricPoe joined #git
22:37 Lor_ joined #git
22:38 mizu_no_oto joined #git
22:39 cagedwisdom joined #git
22:39 khfeng joined #git
22:39 garzer joined #git
22:40 codfection joined #git
22:41 madewokherd joined #git
22:41 codfection 1
22:41 cdown__ joined #git
22:42 t-mart joined #git
22:44 f3r70rr35f joined #git
22:45 nettoweb joined #git
22:45 Seveas 1
22:45 theskillwithin joined #git
22:46 kulelu88 joined #git
22:47 Rodya_ joined #git
22:48 yqt joined #git
22:49 andlabs joined #git
22:50 qsx 2
22:50 cdown_ joined #git
22:51 catsup joined #git
22:52 xall joined #git
22:52 cdown____ joined #git
22:52 codfection joined #git
22:54 Seveas 3
22:54 ojacobson π
22:54 Seveas fail
22:57 star_prone joined #git
22:57 pumafied joined #git
22:58 peeps[lappy] joined #git
23:00 heroux joined #git
23:01 metrixx_ joined #git
23:02 thebope joined #git
23:07 theskillwithin joined #git
23:08 _cyril_ joined #git
23:11 ki0 joined #git
23:12 f3r70rr35f joined #git
23:14 Seveas was kicked by Eugene: 5
23:14 reim joined #git
23:14 Topic for #git is now Welcome to #git, the place for git help and public-access television programming | 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-bisect: No, not like that. Grow up.
23:18 heroux_ joined #git
23:19 qsx 5
23:20 themill joined #git
23:22 arescorpio joined #git
23:22 Derperperd joined #git
23:24 bremner suckup
23:25 f3r70rr35f joined #git
23:25 dsdeiz joined #git
23:28 Mixologic_ joined #git
23:28 Jameser joined #git
23:29 littleforest joined #git
23:29 rominronin joined #git
23:29 duderonomy joined #git
23:30 HoierM joined #git
23:31 nezZario Then what in the world did I do ?
23:31 nezZario Why does everytime I go to push it makes me --set-upstream-to=
23:31 LeBlaaanc joined #git
23:32 mischat joined #git
23:35 Dohplay joined #git
23:36 hahuang61 joined #git
23:39 mancha joined #git
23:39 milki you probaly dont have a goodd default
23:39 milki git config push.default
23:39 Remram joined #git
23:40 mancha hi folks. does "git reset --hard HEAD^" nuke any trace of the last commit? in case it contains sensitive info.
23:40 cdown_ joined #git
23:40 mancha and if not, how to purge all traces (in objects etc) of the last commit.
23:40 milki mancha: !sensitive can help you
23:40 gitinfo mancha: [!filter_sensitive] You can use filter-branch to remove sensitive data from a repository's history. https://help.github.com/articles/remove-sensitive-data/
23:41 duderono_ joined #git
23:41 milki mancha: if you havent pushed to a public repo yet, what you did is probably sufficient to ensure you dont push your sensitive commit
23:41 milki mancha: you can folow up with the cleanup steps in the article
23:42 mancha i've not pushed. i did re-commit which created a file with the same name (without the sensitive info)
23:42 dodobrain joined #git
23:42 mancha those instructions you link seem to rely on the path of the file to do its cleaning. but filename is re-used. :/
23:44 mancha is there an wasy way to troll the caches and objects and show them in human form so i can see if the sensitive data is anywhere locally?
23:45 snowkidind joined #git
23:46 duderono_ joined #git
23:47 mancha oh yeh, i see it there. i found it with reflog. so i guess i need to do some cleaning but not sure how.
23:48 Rodya_ joined #git
23:50 milki mancha: basically you need to remove references to anything that points to the original file before you can force a garbage collection
23:50 milki mancha: you already removed the old commit from your branch, but you also need to cleanup your reflog
23:50 mancha right.
23:50 mancha how?  :)
23:50 milki mancha: step 9 shows that
23:51 milki mancha: you can skip the for-each-ref since you didnt do the filter-branch steps
23:53 mancha ok, this doesn't work well
23:53 mancha because sensitive commit was #5
23:53 mancha if i use the time for that won't it expire reflogs for all <5?
23:55 milki you are going to expire all reflogs
23:56 milki you might be able to expire a single reflog
23:56 milki um
23:56 milki guess not
23:56 * milki shrugs
23:56 mancha ok
23:57 mancha i think the easiest is to save the commits (maybe as patches) and then basically replay the commit process on a new repo.
23:57 mancha is there an easy way to replay commits or should i just script this in bash or similar?
23:57 milki the alternative to 9 is listed at the end of the step
23:57 milki just clone the repo
23:57 mancha that'll be clean of cruft?
23:57 CussBot joined #git
23:57 railssmith joined #git
23:57 milki your new repo wont have cruft
23:57 mancha ah nice.
23:58 milki fresh clones never have cruft
23:58 milki probably
23:58 mancha hah
23:58 milki you cant clone unreachable objects at least. that would be a security violation
23:59 dbkaplun left #git
23:59 milki and clone doesnt use reflogs
23:59 mancha let's give it a whirl
23:59 milki if you know what your object id was, you can verify its gone in the new clone with git cat-file

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