Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-31

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

All times shown according to UTC.

Time Nick Message
00:01 Gsham joined #git
00:01 de-facto joined #git
00:01 gunnaro joined #git
00:06 cyberz joined #git
00:13 matoro joined #git
00:13 rk joined #git
00:16 gunnaro joined #git
00:18 nevodka joined #git
00:18 gunnaro joined #git
00:18 rsripada joined #git
00:19 rk joined #git
00:22 mehola joined #git
00:24 thebope joined #git
00:27 rnsanchez joined #git
00:28 netj joined #git
00:28 Sasazuka joined #git
00:33 chipotle_ joined #git
00:33 mattcen joined #git
00:35 bocaneri joined #git
00:40 cyberz joined #git
00:40 ToBeCloud joined #git
00:41 b0nn cjohnson: yes, different branch, but I had checked the different branch out and there's nothing to pull
00:41 b0nn apologies for the delay in responding, got dragged into a meeting :(
00:41 Nicmavr joined #git
00:41 drodger joined #git
00:41 gunnaro joined #git
00:44 ToBeFree joined #git
00:49 SoulOfTheInterne joined #git
00:49 holodoc joined #git
00:51 AstralStorm joined #git
00:52 Rodya_ joined #git
00:54 rchavik joined #git
00:56 Cabanossi joined #git
01:00 chigang joined #git
01:00 netj joined #git
01:00 steeze joined #git
01:01 dan_ joined #git
01:01 roelmonnens joined #git
01:03 b0nn cjohnson: Ignore all that, I've fixed it :)
01:03 durham_ joined #git
01:04 durham_ joined #git
01:04 cheesegits joined #git
01:04 jimi_ joined #git
01:04 ronin_work joined #git
01:05 orbyt_ joined #git
01:06 Sasazuka_ joined #git
01:08 theiostream joined #git
01:10 zerocua joined #git
01:12 nd- joined #git
01:13 webdev007 joined #git
01:15 dsdeiz joined #git
01:17 holodoc joined #git
01:18 thebope joined #git
01:24 jtoy joined #git
01:24 gunnaro joined #git
01:25 Wulf4 joined #git
01:25 Wulf4 joined #git
01:28 fstd_ joined #git
01:29 finalbeta joined #git
01:32 kyan joined #git
01:33 dan__ joined #git
01:36 ozmage joined #git
01:36 Grievre joined #git
01:36 Grievre Is there a shorter way to accomplish the same thing as this: https://hastebin.com/raw/cesazaluyi
01:37 Grievre I usually just do rebase -i but this sequence of commands I've found to be less prone to giving up due to changes in the master branch
01:43 sunri5e joined #git
01:48 ilbot3 joined #git
01:48 Topic for #git is now Welcome to #git, the place for git help. Just git help, apparently. | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.12.2 | Getting "cannot send to channel"? /msg gitinfo .voice | This is serious business. https://madeitwor.se/git-jokes
01:52 dsdeiz joined #git
01:52 dsdeiz joined #git
01:52 faLUCE joined #git
01:54 thiago Grievre: I can't think of a shorter way.
01:54 thiago Grievre: but I wouldn't use squash merges at all
01:54 durham joined #git
01:55 lavalamp joined #git
01:55 Rodya_ joined #git
01:58 maroloccio joined #git
01:59 d^sh joined #git
01:59 danboid joined #git
02:00 dermoth joined #git
02:03 dreiss joined #git
02:06 cyberz joined #git
02:07 juvenal joined #git
02:08 jeffreylevesque joined #git
02:11 jameser joined #git
02:11 Cabanossi joined #git
02:12 thebope joined #git
02:12 holodoc joined #git
02:12 thebope joined #git
02:13 pionen joined #git
02:17 cyberz joined #git
02:19 lagothrix joined #git
02:19 King_Hual joined #git
02:19 King_Hual joined #git
02:19 danboid joined #git
02:20 matoro joined #git
02:20 Atm0spher1c joined #git
02:21 hashtagstrashtag joined #git
02:22 shgysk8zer0_ joined #git
02:22 dtbn joined #git
02:22 mehola joined #git
02:23 peeps[lappy] joined #git
02:25 hexagoxel joined #git
02:25 Chinggis6 joined #git
02:26 Rodya_ joined #git
02:28 holodoc joined #git
02:29 kbs joined #git
02:34 cyberz joined #git
02:35 ronin_work joined #git
02:35 holodoc joined #git
02:37 dsdeiz joined #git
02:38 dtbn joined #git
02:39 CmndrSp0ck joined #git
02:39 Emperor_Earth joined #git
02:42 cyberz joined #git
02:44 lb1c joined #git
02:47 dtbn joined #git
02:54 Vortex34 joined #git
02:58 snowkidind joined #git
03:04 chachasmooth joined #git
03:05 de-facto joined #git
03:06 ojdo joined #git
03:06 alexandre9099 joined #git
03:06 cyberz joined #git
03:08 holodoc joined #git
03:08 sim590 joined #git
03:09 kaldoran joined #git
03:10 ozmage joined #git
03:11 gunnaro joined #git
03:14 cqi joined #git
03:15 rchavik joined #git
03:15 matoro joined #git
03:15 cyberz joined #git
03:17 holodoc joined #git
03:19 jokajak joined #git
03:19 Andrew_K joined #git
03:20 svm_invictvs joined #git
03:20 ayogi joined #git
03:21 aavrug joined #git
03:22 scoobertron joined #git
03:22 sunri5e joined #git
03:22 aavrug joined #git
03:23 aavrug joined #git
03:23 TbobbyZ joined #git
03:24 marianina8 joined #git
03:24 menip joined #git
03:27 ayogi joined #git
03:28 dreiss joined #git
03:33 Eragon joined #git
03:33 Garen joined #git
03:34 Eragon joined #git
03:34 Eragon joined #git
03:36 akkad joined #git
03:39 matoro joined #git
03:39 dnukem joined #git
03:40 Joufflu joined #git
03:40 Joufflu noob here...question how do I replace my local files with those in my git repository ? I tried "git pull" but that didn't work
03:41 Joufflu i also found "git stash" but not 100% what it will actually do and reading the documentation isn't helping me
03:42 hvxgr joined #git
03:42 drodger joined #git
03:43 kadoban Joufflu: Your local files typically are in a git repository, so it's unclear what you mean.
03:44 Joufflu i've made changes in my editor and have not done a git push origin master
03:44 Joufflu so my files in my repo have not changed
03:45 maroloccio joined #git
03:45 Joufflu i want those files in my repo replace my current local files
03:46 Joufflu i'm working with a local ruby rails server (on vagrant) and usually i play around with the code as I make changes to my site but right now I did something and for some reason I can't get rid of this error
03:46 Joufflu so I just want to revert back to what is currently in my repo
03:48 deepwater joined #git
03:49 dsdeiz joined #git
03:49 sbulage joined #git
03:50 sz0 joined #git
03:51 cyberz joined #git
03:54 ISmithers joined #git
03:54 mizu_no_oto joined #git
03:56 Cabanossi joined #git
03:57 snowalpaca joined #git
03:57 Nugget joined #git
03:58 mehola joined #git
04:00 cyberz joined #git
04:01 raspado joined #git
04:02 Rodya_ joined #git
04:06 xall joined #git
04:07 a_thakur joined #git
04:07 ronin_work joined #git
04:07 a_thakur joined #git
04:17 Goplat joined #git
04:22 cyberz joined #git
04:24 robotroll joined #git
04:24 MutantMahesh joined #git
04:27 kbeflo joined #git
04:27 miczac joined #git
04:32 cyberz joined #git
04:34 reznord joined #git
04:35 ayogi joined #git
04:36 sbulage joined #git
04:37 cyberz joined #git
04:47 nckpz joined #git
04:49 cyberz joined #git
04:49 Eryn_1983_FL joined #git
04:53 saul is there a way to limit permissions on part of the repository so i can prevent users from gaining access to the entire source
04:54 saul i'm operating in a high-risk environment in terms of people taking code out of the company
04:55 _ikke_ No
04:55 _ikke_ Not when you want people to be able to clone the repository
04:55 saul yeah didnt think so ;/
04:56 robattila256 joined #git
04:56 saul i wonder how proprietary software shops handle this
04:56 saul i doubt a windows operating system developer can clone and get the complete source for windows :D
04:56 svm_invictvs joined #git
04:57 reznord saul: well, that is the reason they have private repositories :P
04:58 enckse joined #git
04:58 cyberz joined #git
04:58 saul reznord: i'm not sure what you mean by private ... our repo requires a password, it's not a free public github repo for example
04:59 saul but the password lets you clone it all :D
04:59 storrgie joined #git
04:59 saul i'm imagining something similar to file permissions in unix
04:59 saul and having ownership of files or directories per git user
05:00 saul another idea i had was, for example if you're working on something that uses MVC architecture, the Model and Controller code are on a server, and the guy working on views can edit files and upload them and test
05:00 saul but only the view file he's working on, or something
05:00 saul now i'm just rambling, sorry :)
05:00 rscata joined #git
05:02 dtbn joined #git
05:03 Wulf4 joined #git
05:04 _ikke_ saul: I suppose they are not working with one big repository
05:04 _ikke_ But divided it up into multiple projects
05:04 Rodya_ joined #git
05:05 saul _ikke_: that was another thing i was considering
05:05 dec0n joined #git
05:06 saul (just doing multiple repos)
05:13 holodoc joined #git
05:14 karols146 joined #git
05:14 MutantMahesh joined #git
05:14 dsdeiz joined #git
05:15 dec0n joined #git
05:18 dtbn joined #git
05:20 ogny joined #git
05:22 Eryn_1983_FL joined #git
05:22 aidalgol joined #git
05:24 holodoc joined #git
05:26 TbobbyZ joined #git
05:26 Brski joined #git
05:30 spb joined #git
05:33 dtbn joined #git
05:36 overlord_tm joined #git
05:39 Raging_Hog joined #git
05:39 chardan joined #git
05:44 ronin_work joined #git
05:45 dendazen joined #git
05:49 dtbn joined #git
05:50 lordjancso joined #git
05:52 Eryn_1983_FL joined #git
05:53 Paixo joined #git
05:54 mda1 joined #git
05:55 Paixo left #git
06:00 Paixo joined #git
06:00 diogenese joined #git
06:01 AstralStorm joined #git
06:03 Paixo joined #git
06:05 Rodya_ joined #git
06:06 visual morning fellow gits
06:10 dtbn joined #git
06:11 imack joined #git
06:12 Paixo joined #git
06:13 mar77i joined #git
06:14 inflames joined #git
06:15 Paixo joined #git
06:16 luke_nukem joined #git
06:16 Paixo joined #git
06:17 Paixo joined #git
06:17 Paixo joined #git
06:18 Paixo joined #git
06:19 roelmonnens joined #git
06:19 Silmarilion joined #git
06:20 dsdeiz joined #git
06:20 roelmonnens joined #git
06:22 zeroed joined #git
06:23 alcohol joined #git
06:23 jnavila joined #git
06:24 chele joined #git
06:27 aruns joined #git
06:28 MineCoins joined #git
06:28 faLUCE joined #git
06:29 Taikibansei joined #git
06:31 dtbn joined #git
06:33 faLUCE joined #git
06:35 metachr0n joined #git
06:36 Ratler joined #git
06:38 dreiss joined #git
06:40 faLUCE joined #git
06:40 JeroenT joined #git
06:40 drodger joined #git
06:41 Cabanossi joined #git
06:41 mehola joined #git
06:45 jtoy joined #git
06:45 roelmonn_ joined #git
06:47 dtbn joined #git
06:47 mar77i joined #git
06:48 eroux joined #git
06:48 jagob joined #git
06:49 PaulCapestany joined #git
06:50 mehola joined #git
06:50 a_thakur joined #git
06:51 holodoc joined #git
06:51 a_thakur joined #git
06:53 danboid joined #git
06:54 vrach joined #git
06:56 ShapeShifter499 joined #git
07:01 manuelschneid3r joined #git
07:01 cqi joined #git
07:03 dtbn joined #git
07:03 zulutango joined #git
07:06 kegster joined #git
07:07 pinkkis joined #git
07:07 rokups joined #git
07:07 netroxen joined #git
07:07 chipotle joined #git
07:08 JeroenT joined #git
07:10 chardan joined #git
07:11 Cabanossi joined #git
07:14 matoro joined #git
07:15 MutantMa_ joined #git
07:16 elect_ joined #git
07:20 cyberz joined #git
07:20 sea-gull joined #git
07:21 gtristan joined #git
07:24 Calinou joined #git
07:24 amrits joined #git
07:25 sotoz joined #git
07:25 alesan joined #git
07:25 alesan hello
07:26 alesan when in a git log I see one commit with message like "Merge branch 'develop' of https://......... into develop"
07:26 alesan (develop is a branch in this git)
07:26 weualo joined #git
07:27 alesan is that specific commit, with hash 6a5bf35317390c62191b7382bc1d65e3a5f353d8, just a place holder of some sorts? If I do a git show it does not show anything
07:27 alesan git show 6a5bf35317390c62191b7382bc1d65e3a5f353d8 I mean
07:27 jast it's a merge commit
07:27 jast basically it tells Git that two independent lines of development have been joined (git log -g to see the structure)
07:27 jast sorry, that's git log --graph
07:28 alesan so where would I see the changes?
07:28 alesan oh let me try that
07:28 TbobbyZ joined #git
07:28 schleppel joined #git
07:28 jast well, merges have not just one, but two parents, so *by default* the changes aren't shown since Git doesn't know which parent you want to compare to
07:28 jast you can use '-m' to have the commit listed multiple times: once for each parent, with a diff relative to that parent
07:29 alesan I am not sure I'm still following you... let me make it simpler
07:29 alesan the commit *before* this special merge commit is ...5f7295
07:30 jast chronologically, perhaps.. but think of it as a structure
07:30 imack joined #git
07:30 jast each normal commit has a "parent commit" (the commit that preceded it)
07:30 alesan yes
07:30 jast with a merge commit, there are *two* parent. one is probably that commit you mentioned (5f7295), the other is from the other side of the merge
07:31 alesan OK
07:31 jast with --graph you get lines showing this: directly below the merge commit, it splits up into two lines
07:31 alesan problem is, this "merge" commit introduces a behavior that I need to modify
07:31 flaviodesousa joined #git
07:31 alesan and I cannot find out what files have been touched
07:32 jast as I said, try the -m flag on 'git show', and look at the first diff
07:32 alesan basically there is a bug, and I cannot find out what files I need to investigate
07:32 alesan OK
07:32 alesan git show on the merge commit obviously
07:32 jast or you can do it more explicitly: git diff 6a5bf35^ 6a5bf35
07:32 navidr joined #git
07:32 jast "compare first parent of the commit to commit"
07:32 jast yep
07:32 alesan I see OK let me have a look
07:32 alesan ...to understand why -m is not the default ;)
07:33 jast because it shows the commit multiple times... and *usually* the merge doesn't give you any useful information because the changes are also listed in the second line going downwards from that merge
07:34 jast this is a little more verbose, but here's how to see the commits that got merged in:
07:34 jast git log -p 6a5bf35^..6a5bf35^2
07:35 jast that gives you not a single diff with everything mashed together but the individual commits that got merged in
07:35 cyberz joined #git
07:35 jast *usually* they're equivalent to the first diff from the merge, unless someone did the merge poorly and screwed something up :)
07:38 alesan ok but this means that
07:38 alesan ...
07:38 borkr joined #git
07:38 alesan wait :)
07:40 alesan because the changes are also listed in the second line going downwards from that merge
07:40 alesan can you explain that part again?
07:41 alesan should I look "down" on the git log? to have a list of individual commits?
07:41 karols146 joined #git
07:41 Cabanossi joined #git
07:41 theoceaniscool joined #git
07:42 dtbn joined #git
07:44 sz0 joined #git
07:44 cyberz joined #git
07:46 basiclaser joined #git
07:48 crs joined #git
07:49 crs Hey, I am trying to add shell script as an alias - what shell is aliased script executed in? The one that git was evoked from or some default one?
07:49 grawity I think Git just calls system(), which then uses /bin/sh
07:50 Tobbi joined #git
07:51 crs ok, thanks.
07:52 wouter0100 joined #git
07:53 grawity hmm, it uses something custom, but still just sh -c <arg>
07:54 Levex joined #git
07:55 wouter0100 joined #git
07:56 de-facto joined #git
07:56 alexandre9099 joined #git
07:56 crs So, /usr/local/bin/zsh -c 'f() { echo 'test' }; f' works in terminal, but test = !/usr/local/bin/zsh -c 'f() { echo 'test' }; f' and then git test gives /usr/local/bin/zsh -c 'f() { echo 'test' }: -c: line 0: unexpected EOF while looking for matching `''
07:56 exotime joined #git
07:56 JeroenT_ joined #git
07:56 grawity your quoting is a bit weird there
07:56 crs not sure if I can simply replace !sh with that path…
07:57 grawity and overall I'd suggest calling a full shellscript instead
07:57 crs Ah, had to put it all in "".
07:57 grawity (if you name the script git-<something>, you don't even to define an alias for it!)
07:57 crs grawity: As in, outide of git? Have it at the moments, but would really like that to be part of git * family. :)
07:57 kurkale6ka joined #git
07:58 grawity yeah, plenty of third-party tools do that
07:58 crs grawity: Ah, did do that too, but then it wouldn't appear in zsh tab completion…
07:58 grawity that's a problem with zsh's tab completion then
07:58 raynold joined #git
07:59 grawity I did `git help -a` and it shows me a whole section with my custom subcommands
07:59 crs ok, let me try again then.
08:01 D4R5C0D3 joined #git
08:02 karols146 joined #git
08:02 crs OK, so I can see it in 'git commands available from elsewhere on your $PATH' section of git help -a, but still not in tab completion.
08:03 raspado joined #git
08:04 hurricanehrndz joined #git
08:05 roelmonnens joined #git
08:05 sgo11 joined #git
08:05 faLUCE joined #git
08:06 Andrew_K joined #git
08:07 Rodya_ joined #git
08:07 sgo11 Hi, I have an unreliable network and it always interrupts the connection. My package installer in linux will involve git clone and since it's cloning a big repo and it always failed with the error "error: RPC failed; curl 18 transfer closed with outstanding read data remaining". Is that possible to add some options in .gitconfig to overcome this issue? just tell git clone try again and again when connection interrupts occur. thanks a lot.
08:07 theoceaniscool joined #git
08:10 CaptainHeavy joined #git
08:11 faLUCE joined #git
08:11 Cabanossi joined #git
08:12 kevinfagan joined #git
08:13 dvaske joined #git
08:13 Tjark joined #git
08:15 faLUCE joined #git
08:16 mkoskar joined #git
08:17 dysfigured joined #git
08:17 marcogmonteiro joined #git
08:17 sgo11 the error: https://hastebin.com/ujavehitip.pas
08:18 dysfigured joined #git
08:19 xaa joined #git
08:20 faLUCE joined #git
08:21 RxMcDonald joined #git
08:22 mar77i joined #git
08:23 tvw joined #git
08:23 Eragon joined #git
08:24 sbulage joined #git
08:25 ispn joined #git
08:27 dtbn joined #git
08:28 a_thakur joined #git
08:30 indistylo joined #git
08:30 zeroed joined #git
08:30 zeroed joined #git
08:30 finalbeta joined #git
08:32 mikecmpbll joined #git
08:36 jgrasser joined #git
08:36 MineCoins joined #git
08:42 a_thakur joined #git
08:43 mkoskar_ joined #git
08:44 Vortex34 joined #git
08:44 kinlo joined #git
08:45 a_thakur_ joined #git
08:45 jtoy joined #git
08:45 diget joined #git
08:46 dtbn joined #git
08:47 mischat joined #git
08:47 chele joined #git
08:48 dsdeiz joined #git
08:48 dsdeiz joined #git
08:49 Sound joined #git
08:49 jameser_ joined #git
08:49 chele joined #git
08:51 pks joined #git
08:52 bouncy joined #git
08:53 chele joined #git
08:56 redeemed joined #git
08:58 holodoc joined #git
08:59 karols146 joined #git
09:00 theoceaniscool joined #git
09:01 sgo11 joined #git
09:02 dhollin3 joined #git
09:02 onehrxn joined #git
09:04 onehrxn joined #git
09:06 fragMental joined #git
09:06 lmatteis joined #git
09:07 dtbn joined #git
09:08 theiostream joined #git
09:10 dersand joined #git
09:10 thinkpad joined #git
09:11 Cabanossi joined #git
09:12 buzzLightBeer joined #git
09:20 a_thakur joined #git
09:22 gtristan joined #git
09:22 theoceaniscool joined #git
09:22 mischat joined #git
09:25 clmsy joined #git
09:27 JeroenT joined #git
09:28 mar77i joined #git
09:29 sgo11 joined #git
09:31 TbobbyZ joined #git
09:33 faLUCE joined #git
09:34 dnukem joined #git
09:35 mupadhye joined #git
09:38 faLUCE joined #git
09:38 overlord_tm joined #git
09:39 vuoto joined #git
09:44 marcogmonteiro joined #git
09:44 dnukem joined #git
09:45 sgo11 joined #git
09:45 sgo11 joined #git
09:47 holodoc joined #git
09:49 mehola joined #git
09:50 mehola joined #git
09:51 maroloccio joined #git
09:51 mehola joined #git
09:51 mehola joined #git
09:52 mehola joined #git
09:53 mehola joined #git
09:54 mehola joined #git
09:55 mehola joined #git
09:59 jameser joined #git
10:03 rebrec joined #git
10:04 King_Hual joined #git
10:08 Rodya_ joined #git
10:09 xall_ joined #git
10:09 holodoc joined #git
10:10 Raging_Hog joined #git
10:11 Cabanossi joined #git
10:11 imack joined #git
10:14 Eryn_1983_FL uh guys what does this mean? and how do i fix it?
10:15 Eryn_1983_FL https://pastebin.com/xkzC5PUY
10:15 Assault joined #git
10:15 cdleonard joined #git
10:16 marianina8 joined #git
10:16 Renter it means you have a corrupt object
10:16 Renter do you use intellij?
10:16 jast it means your filesystem is corrupted, probably because of a system crash or power failure
10:17 jast at least that's the most likely scenario... not the only one, of course
10:17 Eryn_1983_FL no..
10:17 Eryn_1983_FL yeah power failures
10:17 Eryn_1983_FL how do i recover?
10:17 Renter git fsck might work
10:17 jast fsck will just confirm it
10:17 Eryn_1983_FL i got another orgin repo
10:17 Renter clean clone will fix it obv
10:17 Eryn_1983_FL git clean clone?
10:17 jast do you have any commits you haven't pushed yet?
10:17 Eryn_1983_FL i dont think so i am good about that
10:18 Renter rm -rf repo ; git clone <remote>
10:18 Eryn_1983_FL unhuh
10:18 Eryn_1983_FL there not a nicer way
10:18 jast if you do have stuff you want to keep, it might be possible to fix this without starting over
10:18 grawity clone the remote *elsewhere*
10:18 Renter You might have luck removing the object and git fetch
10:18 jast first off you can remove the file it complains about, it's worthless anyway
10:18 Eryn_1983_FL ok..
10:18 grawity then move its .git/objects/pack/* to the old repo's
10:18 Renter yeah, like grawity said replace the corrupt object from a clean repo
10:18 grawity then remove those two 0-sized files it complains about
10:18 Renter or just all
10:19 jast first steps I'd suggest before you try anything else: remove the corrupt file and 'git reset'
10:19 grawity then run a `git fsck` just to make sure
10:19 jast that might fix it
10:19 jast if not you can always start with the more complicated things :)
10:19 Renter personally I'd just wipe and get a clean one if the repo is not too big
10:20 Renter but I'm used to IDEs, backups or others corrupting my repos
10:20 jast I'm not :)
10:20 sarri joined #git
10:20 sarri joined #git
10:20 Eryn_1983_FL git reset
10:20 Eryn_1983_FL fatal: Could not parse object 'HEAD'
10:20 jast okay, that's another thing you can probably fix. which branch did you have checked out?
10:20 Eryn_1983_FL it found another corrup file
10:20 Darcidride joined #git
10:21 Eryn_1983_FL experimental branch
10:21 jast it's called "experimental"?
10:21 Eryn_1983_FL yes
10:21 jast does this work: git rev-parse experimental
10:21 Eryn_1983_FL erin@terreis:~/puppet$git branch
10:21 Eryn_1983_FL develop
10:21 Eryn_1983_FL high_level_load_sensor
10:21 Eryn_1983_FL master
10:21 Eryn_1983_FL * support/experimental
10:21 Eryn_1983_FL D
10:21 Eryn_1983_FL working
10:21 Eryn_1983_FL git rev-parse  support/experimental
10:21 Eryn_1983_FL 07746dd551ee6802ef2103863cc7cc906cc9fd80
10:22 jast okay, cool, just fix HEAD, then: echo "ref: refs/heads/support/experimental" >.git/HEAD
10:22 jast obviously never do this if you don't have to :)
10:22 Eryn_1983_FL ok sure
10:22 jast then see if 'git reset' works after that
10:22 Eryn_1983_FL nope same bad head message
10:23 jast huh
10:23 Eryn_1983_FL https://pastebin.com/KkP0E530
10:24 jast how about this one: git cat-file -t support/experimental
10:24 Eryn_1983_FL $git cat-file -t support/experimental
10:24 Eryn_1983_FL fatal: git cat-file support/experimental: bad file
10:24 kexmex joined #git
10:25 jast okay, so the topmost commit on that branch is broken, too
10:25 Eryn_1983_FL :(
10:25 jast and in fact that's the object that got corrupted
10:25 jast can you recover the branch contents from elsewhere?
10:25 Eryn_1983_FL yes my servers had batter backup
10:25 Eryn_1983_FL their intact ill just remake the dang repo :(
10:26 jast because then you should be able to simply delete it, checkout another branch and go 'git branch -D support/experimental'
10:26 jast or you can see if that corrupt file is in one of your backups and copy it over
10:27 ferr1 joined #git
10:28 afuentes joined #git
10:29 bejeb what is it with today and corrupt git projects? I just got "Could not get HEAD hash. libgit2 returned: corrupted loose reference file: refs/heads/feature/review" and git status says I'm on my very first commit from 2 years ago. Does anyone have any ideas?
10:29 Eryn_1983_FL fixed i moved my repo to -old and  recloned..
10:30 Eryn_1983_FL Jast is awesome he would know :)
10:32 acetakwas joined #git
10:33 jast bejeb: solar flares? global magnetic interference? anyway, which branch are you supposed to be on, feature/review?
10:34 bejeb jast: haha. I like your theory. I'm meant to be on feature/review
10:34 jast what's that line in 'git status'?
10:34 jast actually never mind, I can imagine how this might have happened
10:35 bejeb I shut down last night on feature/review. This morning I started up, loaded up tortoisegit and got that error. Ran git status and I get the message from my "initial commit" and every file is set "to be committed"
10:36 jast try 'git log -g feature/review'. if it shows things, the first entry is probably the "right" commit.
10:36 bejeb actually strictly `git status` says "initial commit"
10:36 jast ah, yeah, that's what happens when HEAD points to a branch that Git thinks doesn't exist yet
10:37 jast so suppose that command gave you a reasonable commit ID, then you can resurrect your branch (example commit ID abc123): git update-ref refs/heads/feature/review abc123
10:38 bejeb jast: ugh, see https://pastebin.com/raw/15YSy2kP
10:39 bejeb `git reflog -n 10` gives "fatal: your current branch appears to be broken"
10:39 jast okay, how about this: cat .git/logs/refs/heads/feature/review
10:40 jast if that gives you something, the right commit ID is probably the second SHA1 in the last line (verify with 'git show' on that ID)
10:40 zeroed joined #git
10:40 zeroed joined #git
10:40 ankit01ojha joined #git
10:40 ti2 joined #git
10:40 RxMcDonald joined #git
10:41 Cabanossi joined #git
10:41 RxMcDonald can you change the date of a commit in the past and not affect anything if it's not overlapping?
10:41 canton7 RxMcDonald, the only way you can change the date of a commit is be rewriting history, so yes, that's going to affect things
10:41 RxMcDonald i.e. I made 200 commits on master so far and I want to change the data of the commit 123
10:41 jast RxMcDonald: "not affect anything" is a no-go...whenever you change a commit's metadata, all of the descendant commits (later commits in the branch) have to be changed, too
10:41 canton7 git doesn't really care about whether dates re ordered
10:42 jast if you're the only person working on the project that's still kind of feasible (we have great tools to rewrite history), but think hard on whether you *really* need to do it :)
10:43 RxMcDonald I actually made a commit on a Friday 2 months ago and I want to move it to a saturday but i kept working on the branch
10:43 RxMcDonald and i don't want to lose the work I did after that or the commit history
10:43 RxMcDonald there are no commits on saturday so it shouldnt be a problem
10:43 bejeb jast: amazing, git show shows the right commit. So do I just do a git reset --hard onto that commit?
10:44 jast bejeb: probably won't work since the branch ref is broken, but try that update-ref thing I mentioned and then a 'git reset --hard' (no commit ID needed for that one if the first step is done)
10:44 bejeb ok
10:45 jast you can try a 'git reset --hard <commit>' if you like, it's not like anything can break :)
10:45 bejeb jast: `git update-ref refs/heads/feature/review 702059bd` failed with "fatal: update_ref failed for ref 'refs/heads/feature/review': cannot lock ref 'refs/heads/feature/review': unable to resolve reference 'refs/heads/feature/review': reference broken"
10:46 bejeb I am quite happy to trash that ref, if that's possible, and just checkout to another branch?
10:46 jtoy joined #git
10:46 bejeb that branch isn't pushed to a remote yet
10:46 jast okay, you can do it on filesystem level instead, but you're going to have to use the full commit ID in this case: echo 'full commit id' >.git/refs/heads/feature/review
10:46 zerow joined #git
10:47 juvenal joined #git
10:47 jast well no matter which way you go, you probably don't want to have a broken ref forever, so might as well take care of it now :)
10:48 tmg joined #git
10:48 dsdeiz joined #git
10:48 mar77i joined #git
10:49 a_thakur joined #git
10:51 sotoz joined #git
10:51 bejeb jast: that worked, thank you. How do I buy you a libation of your choice? Interestingly I have actually lost one commit.
10:52 sgo11 My package installer in linux will involve git clone and since it's cloning a big repo and it always failed with the error "error: RPC failed; curl 18 transfer closed with outstanding read data remaining". Is that possible to add some options in .gitconfig to overcome this issue?  the error: https://hastebin.com/ujavehitip.pas thank you very much. I tried to fix this error for two days without luck.
10:54 drodger joined #git
10:54 theiostream joined #git
10:54 tvw joined #git
10:54 bejeb jast: that's odd. `git reflog -n 10` shows the same sha for the lost commit as for the one I've restored for the branch (which is the one that showed up in .git/logs/refs/heads/feature/review)
10:55 dtbn joined #git
10:55 bejeb https://pastebin.com/raw/R4Xc9FRb
10:55 sgo11 I even tried http postBuffer = 1600M. it doesn't help at all.
10:55 sgo11 :q
10:56 bremner sgo11: can you clone the repo outside of your package thingy?
10:56 bejeb sgo11: could you git clone mirror first, then git clone with a reference and then remove the reference?
10:56 eihabarabia joined #git
10:57 sgo11 bremner, no. it gave me the same error msg.
10:57 bejeb sorry `git clone --mirror`
10:57 bremner so maybe you need to discuss with the repo owner if their repo is unclonable.
10:57 bremner what's the url?
10:58 bejeb jast: haha, and because I did a git reset --hard without thinking, I lost the file changes from that missing commit - oops!
10:58 sgo11 bremner, https://github.com/Microsoft/vscode   many people can git clone successfully. I don't know why I failed. maybe because of slow and unreliable network. I am not sure. when it's doing git clone, I does see the speed and progress and it just suddently stops.
10:59 bremner yeah, cloning over crap networks is not really a strongpoint of git.
11:00 sgo11 bremner, the package I tried to install is https://aur.archlinux.org/packages/visual-studio-code-oss From the comments, I don't see anyone met the same error as me.
11:00 bremner yes, clone is very fast here (as expected for github)
11:00 sgo11 bremner, so how can I overcome this issue please? this gives me headache. I didn't do anything except trying to fix this issue for two days.
11:00 jast bejeb: you should now be able to view the reflog properly (git log -g feature/review) so if any commit in there is more up-to-date than what you have now, you can reset --hard to it
11:00 jnavila joined #git
11:01 sgo11 bremner, can I add some options in git and curl config to fix it? thanks.
11:01 bremner no idea
11:01 LukasT joined #git
11:01 bremner sgo11: do you have some other host with better network? you can run the clone there, and then transfer somehow
11:01 bremner e.g. rsync a bundle.
11:02 bejeb jast: you're right, I can view the reflog properly - that latest commit just isn't mentioned, despite being in `git reflog` output, albeit with the same sha as the commit before it (which I've restored)
11:02 masuberu joined #git
11:03 bejeb so commit c0b82ecc was what the branch is based on, and shows up in the reflog. 702059b is next (and what I've restored, and what was in the /ref/log/ file for the branch). ??????? was next, but shows up in  git status with the sha 702059b
11:03 bejeb It's not a biggy I guess. It would be awesome not to have to do the work again, but I'll just crack on with that for now.
11:03 bejeb I'm wondering if it means that I have other problems though, and should re-clone the repo?
11:04 sgo11 bremner, I can try that way. I don't know if it will work in my package installer or not. I have to rsync to some tmp directory before the installer quits. not sure if the installer can use that directory or not. I can try that.
11:04 bremner sgo11: yeah, your package installer is more an arch issue than a git issue
11:05 jast bejeb: shows up in git status? where? and why does it matter? ;)
11:05 bejeb sgo11: can you edit the package installer? If so could you first clone it (either locally or remotely and then transfer it) then tell your package installer to use your local repo directory as the remote?
11:05 jast bejeb: generally I'd recommend running 'git fsck' to have Git look at everything, that way if there's a remaining issue it will find it
11:05 sgo11 bremner, right, but it's git issue first. it's git clone fails. git should take care of unreliable network.... my network is working and just unreliable.
11:06 jast in the meantime, lunch break for me!
11:06 bejeb jast: https://pastebin.com/raw/R4Xc9FRb HEAD@{2} is the ??????? commit, but it has the sha 702059bd12
11:06 bremner sgo11: well, that's a problem that's not going to be fixed soon enough to help you with this specific problem
11:06 bejeb thanks jast. I'll just reclone and redo that work. You're awesome
11:06 sgo11 bejeb, thanks for the advice. I don't know much about my package installer. still a newbie to archlinux. I will try that.
11:07 bremner apparently not enough people with bad network want to work on restartable clone...
11:07 jast bejeb: my guess is that commit simply never made it into the reflog proper before your filesystem/crash issue happened
11:07 HoierM joined #git
11:07 bejeb jast: must be
11:07 bejeb jast: thank you so much again
11:07 jast bejeb: but git fsck --unreachable will find commits that are not referenced from anywhere, might help finding it
11:07 xall joined #git
11:08 sgo11 bremner, more than 1 billion people here have this issue. people just ignore or maybe they don't know. :)
11:08 bejeb jast: `git fsck` looks nasty. I'll reclone and do the work again
11:09 Rodya_ joined #git
11:09 sgo11 s/people/git developers/
11:11 bongjovi joined #git
11:12 Andrew_K joined #git
11:12 TomyWork joined #git
11:12 al-damiri joined #git
11:13 tymczenko joined #git
11:14 henriknj joined #git
11:18 EvilDMP joined #git
11:18 mar77i joined #git
11:19 EvilDMP Given a file that is no longer tracked in Git (because something deleted it, at some point) how can I discover exactly which commit was responsible?
11:19 selckin git log -- path/file
11:21 EvilDMP selckin: perfect
11:21 EvilDMP tahnks
11:23 moritz another nice trick, to show all file deletions: git log --stat --diff-filter=D
11:23 moritz (if you aren't sure about the exact name)
11:25 EvilDMP Noted, thanks
11:26 Murii joined #git
11:31 dglambert joined #git
11:31 dglambert yo
11:33 nedbat moritz: wow, the help for --stat is amazingly unhelpful :)
11:34 TbobbyZ joined #git
11:34 jnavila joined #git
11:36 nowhere_man joined #git
11:37 denisMone joined #git
11:37 jimi_ joined #git
11:45 marianina8 joined #git
11:50 jast nedbat: not if you know the diffstat tool ;)
11:52 Tobbi joined #git
11:54 notebox joined #git
11:55 tcorneli joined #git
11:56 mehola joined #git
11:58 a_thakur joined #git
11:59 mabregu joined #git
12:00 Levex joined #git
12:02 sunri5e joined #git
12:03 Raging_Hog joined #git
12:04 raspado joined #git
12:07 Raging_Hog joined #git
12:10 Rodya_ joined #git
12:10 indistylo joined #git
12:15 rchavik joined #git
12:16 aw1 joined #git
12:17 Ardethian\ joined #git
12:17 kpease joined #git
12:29 kbeflo joined #git
12:31 Zialus joined #git
12:33 mar77i joined #git
12:33 buffal0 joined #git
12:34 aw1 joined #git
12:35 theiostream joined #git
12:36 scottjl joined #git
12:37 RxMcDonald what happens if your last commit is today and you add a new commit with --date in the past
12:37 canton7 not a lot
12:38 canton7 there's at least one third-party history that gets confused, but other than that...
12:38 canton7 *history viewing
12:38 canton7 *viewer, grr. haven't finished my afternoon tea...
12:38 selckin you can change dates to anything you want
12:38 RxMcDonald NOICE!!
12:39 selckin you even have 2 of them, commit time and author time
12:39 vuoto joined #git
12:39 canton7 the order in which commits occurred is tracked by the fact that each commit knows what its parent is, rather than date
12:40 RxMcDonald so i can put all the commits in different dates, past future etc and everything will be OK?
12:40 selckin yes
12:40 rchavik joined #git
12:40 jtoy joined #git
12:41 chipotle joined #git
12:41 canton7 RxMcDonald, are you trying to game the github contribution activity graph? :P
12:42 RxMcDonald canton7: actually i messed up so i wanna fix it
12:42 RxMcDonald canton7 u know what happens if ur employer sees ur commit to ur side project in a work day
12:46 yarekt Heh You can tell them that commit time doesn't correlate to when the code was written
12:46 lukasdboer joined #git
12:46 tmsmith joined #git
12:47 lukasdboer joined #git
12:47 Tobbi joined #git
12:47 Darion joined #git
12:49 CussBot joined #git
12:49 jeffreylevesque joined #git
12:50 selckin if they are the invasive in your time then you have bigger problems
12:51 synthroid joined #git
12:53 oleo joined #git
12:55 theoceaniscool joined #git
12:56 Cabanossi joined #git
12:56 faLUCE joined #git
12:58 lukasdboer joined #git
12:58 marcogmonteiro joined #git
12:59 dvaske joined #git
13:00 ExoUNX joined #git
13:05 Gsham joined #git
13:06 tmsmith joined #git
13:10 cheesegits joined #git
13:13 dvaske joined #git
13:14 ThomasLocke joined #git
13:17 courrier joined #git
13:18 codematt joined #git
13:20 thebird joined #git
13:21 dtbn joined #git
13:22 juvenal joined #git
13:24 kbs joined #git
13:26 chigang joined #git
13:27 marianina8 joined #git
13:28 MutantMahesh joined #git
13:28 theoceaniscool joined #git
13:28 s2013 joined #git
13:31 Dougie187 joined #git
13:32 jnavila joined #git
13:33 steeze joined #git
13:34 mischat_ joined #git
13:34 Ice_Strike joined #git
13:36 RxMcDonald left #git
13:39 chachasmooth joined #git
13:39 TbobbyZ joined #git
13:43 jimi_ joined #git
13:46 CroogoBot joined #git
13:46 s2013 joined #git
13:46 marcogmonteiro joined #git
13:47 CroogoBot left #git
13:49 kbeflo_ joined #git
13:50 karols146 joined #git
13:52 neunon joined #git
13:52 weakdan joined #git
13:53 jameser joined #git
13:55 Tobbi joined #git
13:57 YuGiOhJCJ joined #git
13:57 theoceaniscool joined #git
13:57 Es0teric joined #git
13:57 mohabaks joined #git
13:58 mohabaks hello folks can someone explain to me in simpler term what's rebase all about and when would I find it useful than merging. thanks.
14:02 publio joined #git
14:02 mda1 joined #git
14:04 rominronin joined #git
14:07 davidfetter_ge joined #git
14:09 bannakaffalatta joined #git
14:11 davidfetter_ge left #git
14:11 Rodya_ joined #git
14:11 Vampire0 mohabaks, !merge_vs_rebase
14:11 gitinfo [!merge_or_rebase] For a detailed discussion of the dis-/advantages of merge and rebase, and when to best use which, see: https://coderwall.com/p/ny1hia -- and here's how Linus explains which one to use: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg39091.html
14:12 lavalamp joined #git
14:12 sarri joined #git
14:12 sarri joined #git
14:12 planigan joined #git
14:12 dave0x6d joined #git
14:13 dave_uy joined #git
14:13 notebox joined #git
14:14 netj joined #git
14:14 Rodya_ joined #git
14:14 LeBlaaanc joined #git
14:15 dave_uy When I try git diff --submodule=diff I get fatal: bad object  ... Is this a new bug?
14:15 Es0teric joined #git
14:17 thebird joined #git
14:17 kbeflo joined #git
14:18 jagob joined #git
14:19 AaronMT joined #git
14:19 DarkPsydeLord joined #git
14:19 gugah joined #git
14:21 jokajak joined #git
14:25 bes_internal joined #git
14:25 tymczenko joined #git
14:25 `mx joined #git
14:25 notebox joined #git
14:27 senaps joined #git
14:28 eh3 joined #git
14:28 _Vi joined #git
14:31 kbs joined #git
14:32 LionsMane joined #git
14:35 jeffreylevesque_ joined #git
14:36 thebird joined #git
14:37 ResidentBiscuit joined #git
14:39 Rodya_ joined #git
14:41 thebird joined #git
14:42 ash_workz joined #git
14:44 avar Yay my patch for making @{U} synonymous with @{u} is now in git.git master
14:44 avar So much typo'd @{U} cases will now work..
14:45 _ikke_ avar: yay
14:45 sweatsuit joined #git
14:46 tvw joined #git
14:47 adgtl- joined #git
14:49 mischat joined #git
14:50 avar Also hopefully [tag,branch,for-each-ref] --no-contains will land soon: https://github.com/git/git/commit/ac3f5a3468
14:51 roelmonn_ joined #git
14:51 bes_internal left #git
14:52 rchavik joined #git
14:52 Starky joined #git
14:52 tymczenko joined #git
14:52 robotroll joined #git
14:52 ngladitz left #git
14:54 WakiWork joined #git
14:55 LionsMane joined #git
14:55 alburaq joined #git
14:56 tymczenko joined #git
14:56 WakiWork hi, is it possible to do an octopus merge with unrelated histories? even when specifiying --allow-unrelated-histories git says "Unable to find common commit with <first branch specified>"
14:58 Nilesh_ joined #git
14:59 gableroux joined #git
14:59 madewokherd joined #git
15:00 flavius joined #git
15:02 zerow joined #git
15:03 VicMackey joined #git
15:04 oskarkv joined #git
15:05 dave_uy Correction, when I am in a submodule which has it's own submodules and I run git diff -submodule=diff in the submodule I get fatal: bad object.
15:05 alburaq anyone knows how to resolve this conflict? https://github.com/qutebrowser/qutebrowser/pull/2493
15:05 alburaq i'm caveman
15:05 alburaq is that error resolvable by me?
15:06 ExeciN joined #git
15:06 cjohnson github won't let you do a PR if your merge results in conflicts
15:06 cjohnson it won't let you merge a PR that is*
15:06 cjohnson However you can still merge the branch yourself and close the PR
15:06 cjohnson alternately, you can merge the target branch into your local branch and resolve the conflicts there
15:07 EvilDMP left #git
15:07 cjohnson so when you merge back, there are no conflicts, and the PR will let you go through
15:07 alburaq i forked that repo. but when i do git fetch upstream, i don't get those changes locally
15:09 Sound joined #git
15:09 roelmonnens joined #git
15:09 alburaq cjohnson: did github let me yet to do a PR? does that look like i have already made a PR?
15:10 bannakaf_ joined #git
15:10 TbobbyZ joined #git
15:11 xall joined #git
15:12 Atemu joined #git
15:13 mt joined #git
15:15 shgysk8zer0_ joined #git
15:15 steeze joined #git
15:15 s1scha joined #git
15:15 sparetire joined #git
15:16 de-facto joined #git
15:16 cjohnson you can create a PR, you just can't merge it automatically if it conflicts
15:16 cjohnson git fetch upstream only updates your local copies of upstream, if you want to merge those changes into your local branchesy ou have to do that
15:17 cjohnson git checkout my-foo ; git merge upstream/my-foo
15:17 cjohnson etc
15:20 tymczenko joined #git
15:22 s2013 joined #git
15:25 Gustavo6046 joined #git
15:29 mmlb joined #git
15:31 Eragon joined #git
15:32 jstimm joined #git
15:34 theiostream joined #git
15:37 VicMackey joined #git
15:39 thiago joined #git
15:39 dvaske joined #git
15:41 ResidentBiscuit joined #git
15:41 ayogi joined #git
15:42 robotroll joined #git
15:42 Cabanossi joined #git
15:47 Sound_ joined #git
15:48 courrier joined #git
15:49 courrier Hi there! I stashed changes on path/file, I then renamed path/file to new_path/file and commited the renaming, I then stash-poped, but git replied: Dropped refs/stash@{0} (93978a99db898ba355d730d08f5e3bfd735e1a33)
15:49 courrier Is it due to the renaming? Can I still recover my stash?
15:49 rosiu joined #git
15:50 grawity that's what `git stash pop` always says
15:50 immae courrier: stash pop updates the workdir and drops the stash
15:50 grawity since after all it's apply+drop
15:50 immae so your stash is in workdir/staging :p
15:50 grawity you can `git stash store 93978a99db898ba355d730d08f5e3bfd735e1a33` to un-drop it
15:50 Gsham joined #git
15:52 rosiu joined #git
15:52 vuoto joined #git
15:53 synbit joined #git
15:53 rosiu joined #git
15:53 mischat_ joined #git
15:53 vuoto joined #git
15:54 orbyt_ joined #git
15:55 jimklo joined #git
15:56 igeaod joined #git
15:57 courrier grawity: mmmmh, but "git diff" shows nothing after that, I should see the content of my changes
15:58 thiago git diff HEAD
15:58 thebird joined #git
15:59 courrier OK sorry, the first stash was indeed empty, my actual work was one stash below
16:00 jungsubk joined #git
16:02 kbs joined #git
16:03 shgysk8zer0_ joined #git
16:05 King_Hual joined #git
16:06 raspado joined #git
16:09 armyriad joined #git
16:11 rwb joined #git
16:12 blueyed left #git
16:12 blueyed joined #git
16:13 mischat joined #git
16:13 durham joined #git
16:14 masuberu joined #git
16:15 blueyed joined #git
16:15 blueyed Is it possible to reset/discard changes added to the index?  I have used git-stash --keep-index before, and committed them already, and now want to keep only what "git diff" shows, but not what "git diff --cached" shows.
16:17 svm_invictvs joined #git
16:18 regedit joined #git
16:18 Noldorin joined #git
16:21 immae blueyed: you can do something like git diff > patch && git reset --hard && git apply patch, but maybe there is a faster way
16:22 tmsmith joined #git
16:23 immae or you can commit the index, then the diff, then git rebase -i HEAD^^^ and delete the commit from the index
16:25 xall joined #git
16:26 nitric joined #git
16:28 tmsmith joined #git
16:29 chachasmooth joined #git
16:32 tymczenko joined #git
16:33 pbrewczynski joined #git
16:34 kbeflo_ joined #git
16:35 d^sh joined #git
16:39 s2013 joined #git
16:39 mischat joined #git
16:40 perlpilot joined #git
16:40 linyt joined #git
16:42 zerocua_ joined #git
16:43 jstimm joined #git
16:45 blueyed joined #git
16:46 raspado joined #git
16:47 paul424 joined #git
16:47 JJtG joined #git
16:48 shinnya joined #git
16:50 JJtG hey guys I have a question. For some reason I have local commits that I have never made. Maybe the VS git plugin made them I dont know. Question is how do I get rid of those. I read that git reset --hard origin/master should do it. But I guess I need to change the names for origin and master? I think my local branch is called release e.g
16:52 gableroux joined #git
16:52 overlord_tm joined #git
16:54 moritz JJtG: yes. And that will lose all other changes you made, including uncommitted changes
16:54 cjohnson I don't think reset changes working copy does it
16:54 cjohnson I thought it was a metadata change only
16:54 JJtG moritz, thats okay. I do not do code changes, I dont even understand why git keep thinking i do because I dont
16:54 cjohnson well, you are, somewhere
16:55 cjohnson they don't just magicaly appear :P
16:55 JJtG using UE4 I just do content, all I do is pull and compile
16:55 JJtG but somehow changes appear constantly that I have not made
16:55 cjohnson changes? or commits
16:56 cjohnson if you are compiling and they have not set it up to ignore every built and/or changed file then yeah compiling will generate some untracked changes
16:56 cjohnson you need to first clarify if what you have are commits, or untracked changes
16:56 moritz maybe some generated files are under version control, and modified by the build
16:57 tmsmith joined #git
16:57 JJtG both, just now, the git vs plugin showed me that I had changed 120 files. A friend suggested git sync, but that gave a merge conflict. after exiting that, the plugin got stuck on every subsequent pull operation, so I gues I am using the console now
16:58 mischat_ joined #git
16:59 JJtG but yeah git branch -a gives me a list one is green with an asterisk. I assume that is my local branch. And then a bunch of red ones. One of the red ones has a -> origin/release I assume thats the remote. But whats its name origin or release?
16:59 gableroux joined #git
17:00 JJtG so would the command be git reset --hard release/origin ?
17:01 chachasmooth joined #git
17:03 dreiss joined #git
17:03 deepak joined #git
17:05 jstimm joined #git
17:07 JJtG its not git reset --hard release/origin it gives an error saying fatal:ambiguous argument release/origin : unknown revision or path no in the working tree
17:07 gopar joined #git
17:10 darkmist_ joined #git
17:10 Fallen0223 joined #git
17:11 tang^ joined #git
17:11 Cabanossi joined #git
17:12 iveqy joined #git
17:12 TikityTik joined #git
17:15 jrnieder joined #git
17:15 rominronin joined #git
17:16 nowhere_man joined #git
17:17 imack joined #git
17:18 coreyfarrell joined #git
17:19 Relevant joined #git
17:20 darkmist_ joined #git
17:21 nowhere_man joined #git
17:21 gableroux joined #git
17:21 Relevant Hello all. Currently in process of implementing a development process, probably going with some style of scrumban. What I haven't been able to figure out is how to go about merging from QA to Production branches if there are overlapping changes and one is approved and one is not.
17:22 cjohnson Relevant: my approach is, features get branched from master, and merged into QA for testing. once approved, the feature gets merged into master
17:22 cjohnson we never merge QA
17:22 karols146 joined #git
17:22 coreyfarrell hello, does git have a way to do a dry-run cherry-pick to another branch?  I want to find out if current HEAD commit can be cleanly cherry-picked to origin/different-branch without performing a checkout on that branch or actually changing anything.
17:23 Relevant That's what we tried in the past. It seemed to work, not perfectly, but seemed to work
17:23 cjohnson I think merging QA only works if you do "release planning" and then you either have to delay the release until all features are approved, or you have to revert the merges back out of QA, which can be hugely messy
17:23 cjohnson another part of what we do is to rebase the features from master to keep PRs clean and keep the branches up to date
17:23 Relevant cjognson: ugh, exactly! Glad what I said made sense and you get it! :D
17:24 cjohnson The only downside I've found to my approach is that if you have to make a commit into QA because two features conflict with each other or overlap in some specific way
17:25 Relevant Right, that's what I was just typing lol. If it only works in QA, because of another feature that is in QA...
17:25 cjohnson for example in one branch I might refactor an API and go around to all the places that uses the API to update their usage. If another branch introduces a new call to the old style of API, I would have to add a commit to QA which updates just that one usage
17:25 codematt joined #git
17:25 cjohnson Usually when I come into a case like that, I make a secondary branch from master
17:25 cjohnson and merge those two branches into that secondary branch
17:25 cjohnson and put the fixing commit in there
17:25 cjohnson then merge the merged branch into qa for testing, and merge it into master for release
17:25 Relevant So you basically have a QA "version" and a master "version" of the feature?
17:26 cjohnson yeah in a way. I know it means QA is not a perfect test for prod but honestly I don't know of any workflow that guarantees that
17:26 Relevant Yeah, I think you're right. The only way I've thought of is doing full scrum with sprints
17:27 Relevant Or at least some kind of development stop
17:27 cjohnson yeah if you can absolutely guarantee that you can do a proper release schedule it works
17:27 cjohnson I've just never had stakeholders that would cooperate
17:27 cjohnson lol
17:27 Relevant Yep, same situation here
17:27 DarkPsydeLord joined #git
17:27 gableroux joined #git
17:27 nowhere_man joined #git
17:30 nixjdm joined #git
17:32 moL-Ou joined #git
17:33 marianina8 joined #git
17:35 marianina8 joined #git
17:37 s2013 joined #git
17:39 thiago joined #git
17:39 Oatmeal joined #git
17:39 snowalpaca joined #git
17:41 matoro joined #git
17:42 joshszep joined #git
17:43 snowalpaca joined #git
17:43 snowalpa_ joined #git
17:44 thebird joined #git
17:45 synthroid joined #git
17:45 acetakwas joined #git
17:45 shgysk8zer0_ joined #git
17:49 MutantMahesh joined #git
17:50 duderonomy joined #git
17:50 cbreak joined #git
17:51 finalbeta joined #git
17:53 AstralStorm joined #git
17:57 Acerific joined #git
17:57 Es0teric joined #git
17:57 imack joined #git
18:02 diogenese joined #git
18:02 notebox joined #git
18:03 doener joined #git
18:04 thebope joined #git
18:05 jimi_ joined #git
18:05 Doginal joined #git
18:07 duderono_ joined #git
18:07 marianina8 joined #git
18:09 Sasazuka joined #git
18:11 Forlorn joined #git
18:11 frobnic joined #git
18:12 ertesx joined #git
18:15 marianina8 joined #git
18:15 TomyLobo2 joined #git
18:17 thebird joined #git
18:22 terravires joined #git
18:22 zerocua_ joined #git
18:25 Evidlo joined #git
18:26 marianina8 joined #git
18:29 jstimm joined #git
18:33 mar77i_ joined #git
18:34 freimatz joined #git
18:37 giraffe joined #git
18:38 PioneerAxon joined #git
18:40 dtbn joined #git
18:41 Cabanossi joined #git
18:41 kusznir Hi all: I have what I think is a simple git quesiton: on github, where I have forked a project, how do I bring in the commits I am behind on from the original?
18:42 henriknj joined #git
18:43 marianina8 joined #git
18:43 notebox joined #git
18:45 mikecmpbll joined #git
18:46 dsdeiz joined #git
18:46 Evidlo kusznir: https://help.github.com/articles/syncing-a-fork/
18:46 Evidlo actually this one is better http://stackoverflow.com/questions/7244321/how-do-i-update-a-github-forked-repository
18:48 Kardoso joined #git
18:49 BadApe joined #git
18:50 xall joined #git
18:50 BadApe hi, i think i have a valid ssl cert, but i get the message  SSL certificate problem: unable to get local issuer certificate
18:51 coreyfarrell left #git
18:53 dglambert Hey if I git branch foo and then git branch foo again, does it overwrite my original foo
18:53 chachasmooth joined #git
18:53 qqx dglambert: It will report an error saying that the branch already exists.
18:55 Noldorin joined #git
18:55 scoobertron joined #git
18:57 dglambert ok
18:57 dglambert i was worried I overwrote something
18:57 terravires Is it possible to move all files *.meta from a submodule into the parent repo?
18:59 terravires And not move the files?  So they could stay Parent/Submodule/file.meta but it's part of parent's repo and not submodule?
19:00 trewe joined #git
19:01 oskarkv_ joined #git
19:01 Rodya_ joined #git
19:02 _Vi joined #git
19:05 boombatower joined #git
19:05 Evidlo Is it possible to rebase these 9 commits on the pip-installable branch onto master after they've already been merged? http://imgur.com/a/qwRYd
19:05 qqx terravires: Locally you can do `git rm --cached Parent/Submodule; git add Parent/Submodule`. But other repos won't deal well with updating that.
19:06 boombatower joined #git
19:07 terravires qqx, Maybe my structure is wrong.  Someone suggested I use three repos.  I have a Server, Client and share "Core".  Sadly, the Client requires files in Core that I don't need/want in Server.  Is there a way to do that cleanly?
19:07 mathneverends joined #git
19:07 mathneverends hello
19:07 mathneverends I am having an issue
19:07 mathneverends I have a branch off a master in a git repo
19:07 mathneverends that i have been working on from my main computer
19:07 mathneverends but i am temp on a laptop
19:08 mathneverends i cloned the branch to my laptop
19:08 mathneverends but i cannot launch my angular server as files paths are to my main computers local drive
19:08 mathneverends for some reason
19:08 mathneverends (i did not manually set them)
19:08 dtbn joined #git
19:08 mathneverends Anyone know what the heck is going on?
19:08 qqx terravires: Ahh, so it's only some of the files you want to take out of the submodule. No that isn't possible.
19:09 roby10 joined #git
19:09 orbyt_ joined #git
19:09 moritz mathneverends: did you commit some generate files? or use absolute paths anywhere?
19:10 terravires qqx, Yeah, I needed to share all the files except the *.meta files in Core.  I'm kind of lost how to do this... any suggestions?
19:10 mischat joined #git
19:10 mathneverends moritz: yes but unintentionally
19:10 moritz mathneverends: then fix it
19:11 qqx terravires: The closest you could do would be to have a dir in the client which mixes the .meta files with symlinks to the files from the core submodule which is placed elsewhere.
19:14 qqx terravires: Only other option I could think of that might approach that would be to leave those files in the core repository, and use a sparse checkout of that in the server copies.
19:14 terravires qqx: what about combining client+core, and then use some kind of .gitignore for the server using client/core as a submodule?
19:14 qqx But that isn't something that can be setup in the submodule definition that is committed.
19:16 dermoth joined #git
19:18 chris349 joined #git
19:18 chris349 Is there any way to use git stash to only show the changes that would be made, but not actually make them yet?
19:19 julius joined #git
19:19 julius hi
19:19 julius whats the return code for git when i run git pull and there are changes in my local repository?  got no "not updated" git repo currently
19:20 moritz chris349: git stash show -p
19:20 dtbn joined #git
19:20 moritz julius: the exit code is 0, if that's what you want to know
19:21 chris349 moritz, I think that is only showing past stashes. When I try to do a git pull I get this error: error: Your local changes to 'Gemfile.lock' would be overwritten by merge
19:21 chris349 But the git stash show command does not list that file
19:21 moritz chris349: if git pull gives you this error, you have local modifications
19:21 moritz chris349: and 'git diff' and 'git diff --cached' will show them
19:22 moritz stashes aren't relevant for that
19:22 chris349 I want to know what change will be made with git stash. Always when I get this error do do a git stash followed by git pull
19:22 matoro joined #git
19:22 chris349 So how can I be certain git stash will change what files?
19:23 borkr joined #git
19:24 perlpilot chris349: Why are you uncertain?
19:24 Random832 well you could look at the stash afterward with git stash show
19:24 Random832 or generally predict it from the output of git status
19:24 chris349 because git status and git stash show are not saying the same thing
19:25 Random832 well git stash show is for after you make the stash
19:25 jnewt_ joined #git
19:25 xall joined #git
19:25 chris349 So how can I see what git stash will do without actually doing it?
19:25 perlpilot chris349: you don't (I don't think).
19:25 drodger joined #git
19:26 perlpilot chris349: but, git-stash always saves the contents of the working dir and the index.  That's kinda all it does.
19:26 BlaXpirit yeah  just do it and undo it, what could go wrong
19:27 chris349 lol
19:27 perlpilot BlaXpirit: fear is the mind killer :-)
19:27 BlaXpirit chris349, may i interest you in http://www.praqma.com/stories/git-autostash/
19:29 perlpilot BlaXpirit++ I didn't know about --autostash
19:29 BlaXpirit neither did i,  until today, in fact
19:29 chris349 I would rather something like a prompt from git pull: File XYZ conflict, keep source or destination version?
19:29 BlaXpirit so i only tested it once before recommending here.  but what could go wrong
19:29 perlpilot But that comment on "non-trivial conflicts" is quite right.
19:30 BlaXpirit chris349, you will get that conflict when stash tries to reapply
19:30 perlpilot chris349: I often do what that auto-stash article mentions.  git stash; git pull; git stash pop
19:30 chris349 I have never reapplied a stash
19:30 chris349 I just use it to overcome the git pull errors
19:32 perlpilot chris349: in the case of a conflict, it's not always keep change A or B, sometimes it's C--meld the two sides into something else.
19:33 d5sx43 joined #git
19:33 tyreld joined #git
19:33 BlaXpirit yeah i was gonna say that, i'm just incredulous that this never comes up
19:33 cyberz joined #git
19:34 raijin joined #git
19:34 perlpilot chris349: you know, you could always commit the changes in your working dir before you pull
19:35 gtristan joined #git
19:35 BlaXpirit adn get the same problem then
19:35 Random832 use pull --rebase when you do that
19:35 sparetire joined #git
19:35 perlpilot BlaXpirit: it's not exactly the same... at least your changes would be saved in a commit  :)
19:35 Random832 unless you're doing something weird
19:35 BlaXpirit yeah you feel safe, that's good, perlpilot
19:36 Random832 technically stashing does save your changes in a commit
19:37 perlpilot yeah, but it's kind of a second-rate commit.  It doesn't live with others.   The stash is commits on the other side of the tracks.
19:37 BlaXpirit ok   so autostash completely won me over
19:37 al-damiri joined #git
19:37 Es0teric_ joined #git
19:38 PioneerAxon joined #git
19:39 roelmonnens joined #git
19:40 synthmeat joined #git
19:41 TbobbyZ joined #git
19:41 Cabanossi joined #git
19:41 yqt joined #git
19:42 oleo joined #git
19:47 luke_nukem joined #git
19:49 bariscant joined #git
19:50 matthiaskrgr I wonder why during "Writing objects" stage of git gc, CPU is still often the bottleneck
19:50 pks joined #git
19:57 jj995 joined #git
19:58 jj995 is "working copy" the proper term for the files resulting from a git clone possibly after uncommitted/unstaged changes?
19:59 cyberz joined #git
20:00 qqx jj995: man gitglossary refers to that as the "working tree" but I suspect most people would consider that as a fine alternative.
20:00 gitinfo jj995: the gitglossary manpage is available at http://jk.gs/gitglossary.html
20:00 bwn joined #git
20:01 jj995 qqx and gitinfo: thanks!
20:01 gitinfo jj995: you're welcome, but please note that I'm a bot. I'm not programmed to care.
20:02 Murii joined #git
20:05 synthmeat joined #git
20:10 Pseudocrat joined #git
20:10 Pseudocrat joined #git
20:11 Cabanossi joined #git
20:13 sevenever joined #git
20:15 hahuang61 joined #git
20:16 Discovery joined #git
20:16 dtbn joined #git
20:16 ysh joined #git
20:17 marianina8 joined #git
20:17 marianina8 joined #git
20:22 bwn joined #git
20:23 synthmeat joined #git
20:25 govg joined #git
20:26 iota_pi joined #git
20:30 jeffreylevesque joined #git
20:32 Random832 matthiaskrgr, isn't that the part where it finds delta codings for everything?
20:33 matthiaskrgr that is Compression  I though? :O
20:33 Random832 well it's a kind of compression, yeah
20:33 Random832 compression is cpu intensive in general
20:37 Random832 it's got to compare every blob object against every other blob (well, not *every*, but it has to do it a lot) to figure out which is the best one to use as the delta base
20:38 raijin joined #git
20:43 prg3 joined #git
20:44 Sasazuka joined #git
20:45 matoro joined #git
20:46 orbyt_ joined #git
20:47 pifon joined #git
20:47 Fallen0223 joined #git
20:47 tremby joined #git
20:48 benlieb joined #git
20:50 tremby I'm trying to parse git status --porcelain or git status -z. Parsing filenames becomes difficult when they contain a space, because then suddenly the name is quoted. -z gets around this. But then it becomes complicated again when there is a rename staged. The original name is on the line with the status (`RD oldname\0` for example), and then new name is then on the next line alone (`newname\0`). If newname is something like "MM something" I woul
20:50 tremby Am I missing something? Is there a way to get porcelain or -z to *not* tell me the new names for any renames?
20:52 Tobbi joined #git
20:52 Es0teric joined #git
20:53 MrWoohoo joined #git
20:56 Cabanossi joined #git
20:57 durham_ joined #git
20:58 tremby Oh man, and it turns out that an untracked file with spaces in it is *not* quoted. Seems like a bug.
20:58 Fallen0223 joined #git
20:58 seveneve2 joined #git
20:59 bariscant joined #git
21:01 Aleric joined #git
21:01 tremby I had oldname and newname backwards for the -z version above. But there's still the problem if the old name started with samething which looks like a status string
21:03 amdi_ joined #git
21:04 Aleric Hello all. I'm trying to set up a git server on my machine. When I try to access the server for the first time, using 'git push origin master', I get the error "fatal: 'repos/hamukau.git' does not appear to be a git repository". I found that running the push command results in an attempt to run "git-receive-pack 'repos/hamukau.git'" (without the double quotes) on the remote server.
21:05 Aleric And indeed, I get the same error when I do either:
21:05 Aleric sean:~/tmp/foobar>ssh hamukau "git-receive-pack 'repos/hamukau.git'"
21:05 Aleric fatal: 'repos/hamukau.git' does not appear to be a git repository
21:06 Aleric The remote server is a jailroot... The directory looks ok, because I do not get this error when I run:
21:07 Aleric sean:~/tmp/foobar>sudo chroot /opt/gitserver/hamukau/root /bin/git-receive-pack repos/hamukau.git
21:07 Aleric 008c0000000000000000000000000000000000000000 capabilities^{}report-status delete-refs side-band-64k quiet atomic ofs-delta agent=git/2.7.4
21:07 Aleric etc..
21:07 durham joined #git
21:07 Aleric Anyone got an idea what is wrong? :/
21:08 Aleric s/do either:/do:/
21:08 rwb joined #git
21:09 alexandre9099 joined #git
21:11 jimi_ joined #git
21:11 jstimm joined #git
21:12 p4trix joined #git
21:14 tvw joined #git
21:15 jost_ joined #git
21:18 oleo joined #git
21:19 Aleric nm, I figured it out... chroot does a chdir '/', while the ssh ends up in the users "home directory" which wasn't /.
21:20 Fallen0223 joined #git
21:21 askb joined #git
21:21 Atm0spher1c joined #git
21:24 peterbecich joined #git
21:24 miczac joined #git
21:27 roelmonnens joined #git
21:29 drodger joined #git
21:30 durham joined #git
21:32 Fallen0223 joined #git
21:32 tmsmith joined #git
21:35 cyberz joined #git
21:35 alexandre9099 joined #git
21:36 zerocua joined #git
21:39 cdg joined #git
21:40 durham joined #git
21:41 schleppel joined #git
21:45 texasmynsted left #git
21:47 prg3 joined #git
21:54 Achylles joined #git
21:54 YuGiOhJCJ joined #git
22:08 blackwind_123 joined #git
22:10 bannakaffalatta joined #git
22:10 Dougie187 left #git
22:13 m0viefreak joined #git
22:20 miczac joined #git
22:20 Sasazuka_ joined #git
22:22 acetakwas joined #git
22:22 mischat joined #git
22:26 mikecmpbll joined #git
22:27 anuxivm joined #git
22:28 Tobbi joined #git
22:32 martin_g left #git
22:33 Vinnie_win_d joined #git
22:33 gopar joined #git
22:34 drodger joined #git
22:34 gopar joined #git
22:42 raijin joined #git
22:42 Rodya_ joined #git
22:45 zerocua joined #git
22:47 Sasazuka joined #git
22:48 rokups joined #git
22:48 rigid joined #git
22:48 rigid joined #git
22:50 dendazen joined #git
22:50 cahoots joined #git
22:51 Aleric Should a git server be able to create arbitrary files in it's root? Currently I see 'config', 'description' and 'HEAD'.
22:52 cahoots hi, i want to resolve some merge conflicts with the master branch, but i don't actually want to merge the latest master into my code, because that would invalidate all of my build artifacts and force a clean build. is there any way to do this?
22:52 Noldorin joined #git
22:52 cahoots like, even just seeing what merging would look like, so then i can change my code not to cause it would be good
22:54 wizardedit joined #git
22:55 Aleric cahoots: you should be using ccache and not worry about that :p
22:55 bambams joined #git
22:55 bambams joined #git
22:55 Aleric https://linux.die.net/man/1/ccache
22:56 cahoots Aleric, it's not in place yet sadly
22:56 Cabanossi joined #git
22:57 Sasazuka joined #git
23:00 tmg joined #git
23:00 joshszep joined #git
23:00 robattila256 joined #git
23:00 cyphase joined #git
23:01 zulutango joined #git
23:02 svm_invictvs joined #git
23:04 kristofferR joined #git
23:04 gopar joined #git
23:05 cahoots it looks like there are some ways to just merge specific files, but they all look super hacky
23:05 gopar joined #git
23:06 hahuang61 joined #git
23:09 Sasazuka joined #git
23:10 fakenerd joined #git
23:13 tremby left #git
23:16 roelmonnens joined #git
23:19 madewokherd joined #git
23:23 mischat joined #git
23:23 nowhere_man joined #git
23:29 kristofferR joined #git
23:29 Sasazuka_ joined #git
23:32 cahoots_ joined #git
23:33 kristofferR joined #git
23:35 duderonomy joined #git
23:36 anuxivm left #git
23:37 cyphase joined #git
23:37 Sasazuka joined #git
23:39 Oatmeal joined #git
23:41 matoro joined #git
23:41 Cabanossi joined #git
23:41 duderonomy Hello. How to move a commit from an old branch (to master). The old branch has not been worked on for 2 weeks. I will probably hace conflist to resolve. What are my options or approach? thx!
23:42 durham_ joined #git
23:42 afuentes joined #git
23:42 duderonomy I think that I see two options: cherry-pick and rebase a branch. Not sure how to proceed. Trying to research which one is preferable for my case.
23:44 Garen joined #git
23:45 prg3 joined #git
23:49 cahoots joined #git
23:51 jtoy joined #git
23:52 cyphase joined #git
23:53 Sasazuka_ joined #git
23:53 elsevero joined #git
23:54 cyberz joined #git
23:57 joshszep joined #git

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