Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-01-17

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

All times shown according to UTC.

Time Nick Message
00:01 Zialus joined #git
00:02 cqi joined #git
00:07 disi i have 10 branches based on master. when one merges, i want to pull master, delete the branch that merged, then rebase the other 9 branches and force push them. is there a better way to do that other than manually checking out each one, rebasing it then force pushing?
00:07 Laurenceb__ joined #git
00:11 dsdeiz joined #git
00:11 dsdeiz joined #git
00:11 jellycode2 is there a way to generate an ssh key with git for windows without being able to use the git-bash command prompt?
00:11 thethorongil joined #git
00:11 m0viefreak disi: rebase requires a working-tree, so no
00:11 jellycode2 purely from windows command prompt, potentially powershell
00:12 felixjet in github, it is not possible to push directly to a org repo without having to fork and PR ?
00:12 jellycode2 it is possible
00:13 m0viefreak jellycode2: just run ssh-keygen?
00:13 Vampire0 jellycode2, why don't you start bash from Windows command prompt?
00:13 jellycode2 i'm using a cloud vm which gives me a command prompt, but gitbash comnmand doesn't work
00:13 Vampire0 jellycode2, or just run the ssh-keygen from the Windows command prompt, should both work fine, except that it is not in the PATH
00:13 m0viefreak just run ssh-keygen, no need for shellception
00:14 jellycode2 thanks!
00:14 felixjet jellycode2, how?
00:14 disi m0viefreak: ty
00:14 Zarthus jellycode2, windows subsystem or puttygen then converting it to an openssh key
00:14 xaviergmail joined #git
00:15 jellycode2 i saw a bunch of articles that seemed to incdicate "you must launch git bash first..."  very misleading perhaps
00:15 gitinfo felixjet: Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
00:16 m0viefreak jellycode2: it usually not added to PATH outside git-bash by default
00:16 m0viefreak maybe thats why
00:16 jellycode2 good point, m0viefreak is awesome
00:16 jellycode2 do u know answers to any of my other questions freak?
00:17 ome joined #git
00:17 thethorongil joined #git
00:17 m0viefreak uhm the clone one?
00:17 m0viefreak just cd to <path> first and then clone?
00:18 cyan__ joined #git
00:18 jellycode2 might actually be able to do that here :)
00:18 m0viefreak or use git -C
00:19 m0viefreak git -C <path> clone <repourl>
00:19 druonysus joined #git
00:19 m0viefreak which is essentially the same as cd <path> && git clone <repourl>
00:19 m0viefreak see man git
00:19 gitinfo the git manpage is available at http://jk.gs/git.html
00:19 jellycode2 that's the one!
00:20 cdown joined #git
00:26 mehola joined #git
00:26 cdown joined #git
00:26 nick123 joined #git
00:32 YuGiOhJCJ joined #git
00:35 JanC_ joined #git
00:37 kyan joined #git
00:37 mehola joined #git
00:37 arescorpio joined #git
00:39 NeverDie joined #git
00:40 mehola_ joined #git
00:48 a_thakur joined #git
00:50 Noodlewitt joined #git
00:50 lindenle joined #git
00:54 aw1 joined #git
00:56 timvisher joined #git
01:05 rivarun joined #git
01:08 s2013 joined #git
01:10 variable joined #git
01:25 mizu_no_oto joined #git
01:25 JanC joined #git
01:27 a_thakur joined #git
01:28 UnnamedPlayer joined #git
01:29 Darren_ joined #git
01:29 AlineGomes joined #git
01:32 jellycode2 wow, the github "integration" system is confusing :(
01:32 rchavik joined #git
01:35 lvh joined #git
01:35 Eugene joined #git
01:35 amrits joined #git
01:39 pR0Ps joined #git
01:39 chipotle joined #git
01:39 atrigent joined #git
01:40 timvisher joined #git
01:44 portalesman joined #git
01:46 portalesman left #git
01:48 SantaClauze joined #git
01:56 mizu_no_oto joined #git
01:57 Guest38975 joined #git
01:57 Guest38975 left #git
01:58 a_thakur joined #git
02:00 atrigent joined #git
02:00 GT4066 joined #git
02:01 nevodka joined #git
02:01 Vampire0_ joined #git
02:02 chipotle joined #git
02:02 drbean joined #git
02:03 nunchuck joined #git
02:06 Spherical joined #git
02:07 finalbeta joined #git
02:07 Spherical joined #git
02:10 Tobarja joined #git
02:12 TinkerTyper joined #git
02:14 Wulf4 joined #git
02:15 timvisher joined #git
02:18 tchan joined #git
02:18 tchan joined #git
02:19 cqi joined #git
02:21 Cthalupa joined #git
02:21 mischat joined #git
02:22 cqi joined #git
02:23 brent__ joined #git
02:26 chipotle joined #git
02:28 nick123 joined #git
02:29 regedit joined #git
02:32 thethorongil joined #git
02:32 simonides joined #git
02:33 s2013 joined #git
02:37 d^sh joined #git
02:38 Cthalupa joined #git
02:41 askb joined #git
02:42 askb joined #git
02:46 pks joined #git
02:46 shinnya joined #git
02:49 ilbot3 joined #git
02:49 Topic for #git is now Welcome to #git, the place for git help and a fresh start on a new branch | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | This channel may contain users known to the state of California to cause cancer.
02:52 timvisher joined #git
02:53 dermoth_ joined #git
02:54 snowkidind joined #git
02:56 Raineer joined #git
02:56 aavrug joined #git
02:57 fakenerd joined #git
03:01 Vampire0 joined #git
03:02 Vampire0 joined #git
03:04 hahuang61 joined #git
03:04 Vampire0 joined #git
03:07 Sasazuka joined #git
03:08 rchavik joined #git
03:08 fakenerd joined #git
03:09 ericsupreme joined #git
03:09 midgetspy joined #git
03:14 Khisanth joined #git
03:15 Darren_ joined #git
03:17 hexagoxel joined #git
03:22 Vortex34 joined #git
03:23 mischat joined #git
03:23 dreiss joined #git
03:25 hypercube32 joined #git
03:26 [R] joined #git
03:26 [R] i merged a branch into my current branch, and the branch i merged updated the path for a submodule
03:26 [R] i can't figure out how to make it switcch to taht new submodule
03:26 cyphase joined #git
03:27 scarcry joined #git
03:28 jameser joined #git
03:29 nick123 joined #git
03:29 chachasmooth joined #git
03:29 timvisher joined #git
03:30 sunri5e joined #git
03:31 howdoi joined #git
03:33 aw1 joined #git
03:34 jameser joined #git
03:34 lindenle joined #git
03:37 Goplat joined #git
03:37 mizu_no_oto joined #git
03:38 fakenerd_ joined #git
03:40 phanimah1sh joined #git
03:41 zivester joined #git
03:42 fstd_ joined #git
03:47 gugah joined #git
03:47 fahadash joined #git
03:48 jon-mac joined #git
03:49 rnsanchez joined #git
03:51 rnsanchez_ joined #git
03:51 chandu20 joined #git
03:58 tristanp joined #git
03:59 cbreak joined #git
04:00 hqip joined #git
04:00 a_thakur joined #git
04:06 timvisher joined #git
04:06 menip joined #git
04:08 mostlybadfly joined #git
04:08 Raineer joined #git
04:14 phanimahesh joined #git
04:14 atrigent joined #git
04:14 nckpz joined #git
04:15 govg joined #git
04:21 thethorongil joined #git
04:21 chandu20 joined #git
04:21 thethorongil joined #git
04:23 mischat joined #git
04:24 mar77i joined #git
04:28 acetakwas joined #git
04:30 nick123 joined #git
04:33 phanimah1sh joined #git
04:38 phanimahesh joined #git
04:38 raedah joined #git
04:43 bluepixel joined #git
04:44 sbulage joined #git
04:45 timvisher joined #git
04:46 robotroll joined #git
04:47 xall joined #git
04:48 arnsa joined #git
04:51 energizer joined #git
04:51 energizer Is it possible to do git diff ignoring line order?
04:59 svm_invictvs joined #git
05:00 acetakwas joined #git
05:01 zacts joined #git
05:03 ayogi joined #git
05:04 g__k__ joined #git
05:12 jimInDevNull joined #git
05:13 lindenle joined #git
05:13 albel727 joined #git
05:14 fakenerd joined #git
05:15 rewt`` joined #git
05:16 s2013 joined #git
05:21 romerocesar joined #git
05:22 jameser joined #git
05:24 mischat joined #git
05:27 Mowgli joined #git
05:29 timvisher joined #git
05:31 nick123 joined #git
05:34 acetakwas joined #git
05:36 GodGinrai joined #git
05:37 durham joined #git
05:38 Cabanossi joined #git
05:40 _ikke_ energizer: perhaps be creating a custom diff driver
05:40 energizer _ikke_: i'm not familiar with that phrase, what does it mean
05:41 Outsider__ joined #git
05:41 _ikke_ git allows you to specify a diff 'driver', which is basically another program that provides diffs
05:41 _ikke_ But can you tell what you mean exactly with ignoring line order?
05:45 energizer if two files are the same except the lines are rearranged, it should say they're the same
05:46 energizer just like diff <(sort file1) <sort file2)
05:46 energizer except that i want to do --color-words
05:49 fakenerd joined #git
05:49 _ikke_ energizer: https://stackoverflow.com/questions/20377294​/git-diff-not-taking-line-order-into-account
05:50 energizer cool, thanks
05:50 _ikke_ But I suspect --color-words only works with gits built-in diff engine
05:52 atrigent joined #git
05:53 govg joined #git
05:56 cqi joined #git
05:59 mar77i joined #git
05:59 Duder9999 joined #git
06:00 rnsanchez joined #git
06:04 timvisher joined #git
06:05 absinthe joined #git
06:05 rnsanchez joined #git
06:06 pity joined #git
06:10 rscata joined #git
06:11 Raging_Hog joined #git
06:15 jameser_ joined #git
06:15 durham joined #git
06:16 durham joined #git
06:17 Nitesh git apply abc.patch fails . One of the files cannot be applied, is there any command that can git apply a patch in atleast those files where it is successful
06:17 durham joined #git
06:18 _ikke_ --reject
06:18 durham joined #git
06:18 Nitesh _ikke_: Applied. :-) Thnx
06:19 eizua joined #git
06:23 pity joined #git
06:24 tirej joined #git
06:24 tirej joined #git
06:26 rominronin joined #git
06:26 mischat joined #git
06:26 pity Guys, I have a problem. I use git on Mac OS El Captain, when I run 'git clone' or 'git checkout', it hangs for 30 seconds, I've done some debugging but no clue found by now.
06:27 pity git checkout master  0.01s user 0.00s system 0% cpu 30.030 total
06:30 xberg joined #git
06:30 zefferno joined #git
06:31 eper3z joined #git
06:34 tristanp joined #git
06:38 qt-x joined #git
06:40 safe joined #git
06:40 nckpz joined #git
06:41 chandu20 joined #git
06:43 timvisher joined #git
06:47 jnavila joined #git
06:48 dec0n joined #git
06:48 fakenerd joined #git
06:48 adymitruk pity: have you tried via ssh instead of https?
06:50 NeverDie joined #git
06:50 fakenerd joined #git
06:50 hahuang61 joined #git
06:54 ThomasLocke joined #git
06:56 diogenese joined #git
06:56 fakenerd joined #git
06:58 lordjancso joined #git
07:01 zeroed joined #git
07:02 brent__ joined #git
07:04 a_thakur joined #git
07:05 tchan joined #git
07:05 fakenerd joined #git
07:06 karthike1991 joined #git
07:14 overlord_tm joined #git
07:15 pity joined #git
07:17 pks joined #git
07:18 bill99 joined #git
07:19 MineCoins joined #git
07:19 karthickeayan joined #git
07:20 dreiss joined #git
07:20 zeroed joined #git
07:20 ojacobson 30 seconds? suspiciously like DNS or ip6 timeouts
07:21 ojacobson Consider turning off ip6 support on your router, if you're not using it for anything specific. If that helps, then that's a strong indication it's misconfigured.
07:22 timvisher joined #git
07:22 kadoban Why would checkout do network anything though?
07:23 osse pity: ''type git''
07:23 osse pity: and  GIT_TRACE=1 git clone ...
07:24 ojacobson Oh, hm.
07:24 ojacobson Good question.
07:25 kadoban Still might be right though, people have been known to make mistakes in their questions
07:25 JelmerD joined #git
07:26 mischat joined #git
07:27 mohabaks joined #git
07:28 thiago joined #git
07:28 tchan joined #git
07:28 chele joined #git
07:29 _ikke_ You can also set the preference to ipv4 instead of ipv6
07:29 _ikke_ no need to disable it completely
07:29 _ikke_ if that's the issue
07:31 fakenerd_ joined #git
07:31 pity joined #git
07:34 parasite_ joined #git
07:34 M-ou-se joined #git
07:35 nidr0x joined #git
07:36 thiago ipv6 is preferred if you have a native ipv6 connection
07:36 thiago otherwise, ipv4 is preferred
07:37 thiago you don't need to do anything, unless the software in question is buggy
07:37 miczac joined #git
07:39 eduardas_m joined #git
07:41 _ikke_ Right. But for example in certain parts of our netwerk, ipv6 is enabled, but does not yet have internet access
07:41 thiago if you have a global ipv6 address with a global route, then you should have global ipv6 internet
07:41 thiago anything else is grounds for firing the network admin
07:42 JeroenT joined #git
07:42 _ikke_ thiago: It's more due to policy
07:42 _ikke_ And technical limitations
07:42 _ikke_ (hard to go into detail)
07:42 thiago I suppose it's like the office network, where I do have a global ipv4 route but it doesn't lead to the internet
07:42 thiago it requires a proxy
07:42 _ikke_ right
07:42 thiago so use the proxy
07:43 _ikke_ We have transparent proxy
07:43 _ikke_ but only does ipv4
07:43 thiago do you have a global ipv6 address?
07:43 _ikke_ yes
07:43 zeroed joined #git
07:43 zeroed joined #git
07:43 _ikke_ But no direct internet access
07:43 thiago and you have a global route?
07:43 _ikke_ certainly
07:43 thiago and your network admin hasn't been fired?
07:44 _ikke_ Nope
07:44 thiago why couldn't he/she simply configure the routers not to advertise a global route?
07:44 mar77i joined #git
07:44 _ikke_ Because we do use these routes
07:44 thiago you just said they don't work
07:44 _ikke_ Just not for all clients
07:44 thiago why do you have a ::/0 route if you can't reach the internet?
07:45 thiago couldn't you simply have the route for wherever you need to go?
07:46 thiago why do you have a global address if you can't reach the global internet?
07:46 thiago that's why your network admin should be fired for incompetence
07:46 _ikke_ thiago: Because in ipv6, you use global adresses by default
07:46 thiago you should be using a unique-local address instead
07:46 thiago from the fc00::/7 range
07:46 _ikke_ Not really
07:47 thiago then /etc/gai.conf would have sorted IPv4 above IPv6
07:48 _ikke_ Note that it's a temporary issue
07:48 _ikke_ because the transparent proxy does not support ipv6 yet
07:48 romerocesar joined #git
07:49 thiago you can't transparent-proxy ipv6
07:49 _ikke_ Most servers don't have an ipv6 adress yet
07:49 ihradek joined #git
07:49 _ikke_ for windows clients we deployed a registry fix to prefer ipv4
07:49 thiago well, I suppose you could, but you're not suppose to
07:49 thiago for Linux, do this: ip -6 route add unreachable 2000::/3
07:50 thiago this will disable trying to reach the global Internet
07:50 Alienpruts joined #git
07:50 thiago your organisation's local addresses in fc00::/7 will be unaffected, as will any other routes
07:50 _ikke_ This would break our internal network (which consists of globally routable adresses)
07:50 habanera joined #git
07:51 _ikke_ note that we have a global network
07:51 jameser joined #git
07:51 thiago you should really reconsider just moving your network to fc00::/7 and leave the global addresses for when you have an uplink
07:51 jnavila joined #git
07:51 hahuang61 joined #git
07:51 fakenerd joined #git
07:51 _ikke_ The only thing that prevents ipv6 access is a firewall
07:51 _ikke_ It's completely routable for the rest
07:52 thiago if you're supposed to use a proxy, then configure a proxy in your software
07:52 _ikke_ thiago: like I said, we use a transparent proxy
07:52 thiago otherwise, if direct connection is supposed to work, then software is working properly
07:52 thiago it's your network that is misconfigured and your network admin deserves to lose his job if he leaves it like that for any extended period of time
07:52 ^7heo _ikke_: are you coming to fosdem?
07:52 _ikke_ ^7heo: nope
07:53 ProbabilityMoon joined #git
07:55 fakenerd_ joined #git
07:58 xall joined #git
07:59 pity joined #git
07:59 absinthe joined #git
07:59 Erendis joined #git
08:00 Khisanth joined #git
08:00 hhee joined #git
08:01 _ikke_ thiago: apparently fc00::/8 has not been properly defined yet
08:02 elect joined #git
08:03 sbulage joined #git
08:04 timvisher joined #git
08:04 JeroenT joined #git
08:07 a_thakur joined #git
08:09 Sceorem joined #git
08:12 theoceaniscool joined #git
08:12 rez-- joined #git
08:13 solution_ joined #git
08:16 jagob joined #git
08:17 nug700 joined #git
08:17 nug700 using sourcetree on windows, for some reason my push is stuck on "POST git-receive-pack (chunked)"
08:18 nug700 after using "git config http.postBuffer 524288000" on the repo.
08:19 Erendis left #git
08:20 donught joined #git
08:20 karthike1991 joined #git
08:21 marvi joined #git
08:22 RxMcDonald joined #git
08:23 RxMcDonald Hello, anyone knows how to checkout to a commit that was made in a different remote? git checkout fork/master doesn't show the last commit
08:24 moritz RxMcDonald: you might need to 'git fetch --all' first
08:26 nostrora joined #git
08:26 star_prone joined #git
08:26 a_thakur joined #git
08:27 marvi joined #git
08:27 garethdaine joined #git
08:27 aard_ joined #git
08:29 AustinMatherne joined #git
08:30 a_thakur_ joined #git
08:31 donught joined #git
08:35 steve joined #git
08:38 zefferno joined #git
08:38 aax joined #git
08:40 dimi1947 joined #git
08:42 timvisher joined #git
08:45 SwiftMatt joined #git
08:45 Guest10762 joined #git
08:47 anry joined #git
08:47 karthickeayan joined #git
08:47 absinthe joined #git
08:50 Raed|Laptop joined #git
08:51 Seveas joined #git
08:52 WayToDoor joined #git
08:53 a_thakur joined #git
08:54 a_thakur_ joined #git
08:55 Balliad joined #git
08:56 mischat joined #git
08:56 mischat joined #git
08:57 star_prone joined #git
08:57 ispn joined #git
08:58 Endarked joined #git
09:03 Darren_ joined #git
09:05 onmeac joined #git
09:05 Balliad joined #git
09:06 s2013 joined #git
09:06 ronny hi
09:06 bocaneri joined #git
09:07 ronny anyone aware of a git native way to clean up local and remote merged branches? - i did a quick google, but there seems to be a massive crowd of fragile shell scripts instead of a clean/safe solution
09:08 laen joined #git
09:09 kurkale6ka joined #git
09:09 osse ronny: git for-each-ref --merged master --format 'delete %(refname)' | git update-ref --stdin
09:10 garethdaine joined #git
09:10 osse ronny: not 150% native, but I'll shoot myself if that is considered fragile
09:11 serialoverflow joined #git
09:11 osse ronny: might want to limit it to branches
09:11 osse git for-each-ref --merged master --format 'delete %(refname)' refs/heads
09:12 ronny osse: wow, that one is way better than the drunken grep vodoo i see all over the place
09:12 j416 git branch --merged | (delete local and remote)
09:12 ronny j416: wtf? git branch --merged is the broken starting point of most of those fragile shell hells ...
09:13 KevinDF joined #git
09:14 osse ronny: by default for-each-ref iterates over ALL refs. branches, tags, remote branches, your grandma's sock drawer
09:14 osse it can be limited
09:14 osse as for remote branches, a git fetch --prune might get rid of a lot
09:14 ronny osse: its a very lovely tool, much more safe than all of the crazy hacks starting with that stuff
09:14 bongjovi joined #git
09:15 ronny osse: as for remote cleanup, i would like to propagate from the current repo
09:15 ronny btw, is there a sane way to detect rebases/squashes?
09:15 ronny the github repos im interacting with dont always go for plain merge using a commit
09:15 zeroed joined #git
09:16 ronny (and thus the git mechanism is unhelpfull due to lack/use of rebase metadata)
09:17 osse I don't know. What would be the purpose?
09:19 star_prone joined #git
09:20 timvisher joined #git
09:20 ronny osse: i want to automate pruning the branches that i brought trough the review process after they got merged, it works fine with the normal ones, but the ones that get squashed/rebased by github are hidden because git rightfully views them as different changes
09:22 snits joined #git
09:23 Guest36461 joined #git
09:23 mp4 joined #git
09:24 osse ronny: even after they get rebased you still have to merge them, no? Or how else does the content get into the main branch?
09:24 Mowgli joined #git
09:25 ronny osse: the workflow is i submit a commit to gh, gh puts a rebased/squahsed one onto master, my own one is now not matching what comes in from master
09:26 ronny (i.e. its in sane)
09:26 ronny *insane
09:27 osse ronny: in that case you can check if rebasing that branch on master results in all changes being dropped
09:27 mar77i joined #git
09:27 Probabil_ joined #git
09:27 osse since rebase automatically skips changes that are already present in the new base branch
09:28 ronny osse: is there a way to do that without changing the workdir or a ref?
09:28 osse not that I know
09:28 ronny i see
09:28 osse but you can create a secondary worktree and a testing ref
09:28 star_prone joined #git
09:29 m4sk1n joined #git
09:29 osse Or start deleting the branches immediately after receiving a notification they've been accepted :p
09:29 Guest9 joined #git
09:29 star_prone joined #git
09:29 Guest9 joined #git
09:30 Snugglebash joined #git
09:31 jstein joined #git
09:31 phanimahesh joined #git
09:31 thiago joined #git
09:31 clmsy joined #git
09:33 Shashikant86 joined #git
09:33 mahendratech joined #git
09:33 acetakwas joined #git
09:35 ronny osse: im currently working trough a vacation backlog ^^
09:35 osse ahh
09:37 overlord_tm joined #git
09:37 tristanp joined #git
09:39 r4dn3t joined #git
09:39 jellycode joined #git
09:39 overlord_tm joined #git
09:41 marcogmonteiro joined #git
09:41 CaptainQuirk joined #git
09:42 CaptainQuirk Hi guys ! How can I change the locale used by git in my terminal
09:42 CaptainQuirk I switched the local system wide on my ubuntu, but git status continues to display info in french
09:42 osse CaptainQuirk: export LANG=blabla
09:43 wrouesnel2 joined #git
09:43 osse or before you do that, what does `locale` say?
09:43 Mowgli joined #git
09:44 CaptainQuirk osse, it's still system issue then
09:44 jozwior_ joined #git
09:45 CaptainQuirk LANGUAGE seems to do the trick
09:45 KekSi or export LC_ALL="en_US.UTF-8"
09:45 pity joined #git
09:45 CaptainQuirk there must be a place where it was set up at first install
09:45 jozwior__ joined #git
09:46 KekSi locale-gen --no-purge en_US.UTF-8 && export LC_ALL="en_US.UTF-8" && update-locale LANG=en_US.UTF-8
09:46 Shashikant86 joined #git
09:46 KekSi from one of my base Dockerfiles (just to make absolutely sure)
09:46 cyan__ joined #git
09:47 Alienpruts joined #git
09:49 chll_ joined #git
09:51 mogio joined #git
09:51 Mowgli joined #git
09:52 CaptainQuirk doesn't work in new terminal I open
09:52 MrMojito joined #git
09:52 MrMojito joined #git
09:53 hahuang61 joined #git
09:54 pity joined #git
09:54 Kevin_ joined #git
09:55 Mowgli joined #git
09:55 Kevin__ joined #git
09:56 Serpent7776 joined #git
09:56 Mowgli joined #git
09:57 richardlampitt joined #git
09:57 Mowgli joined #git
09:58 oskarkv joined #git
09:59 ronskisika joined #git
10:00 Mowgli joined #git
10:00 cdown joined #git
10:00 chipotle joined #git
10:00 richardlampitt Hello all - I've got an issue on git on windows. I've renamed a few of my files by changing the cases of some of the letters, but every time I checkout a commit that was before the rename occurred, all of the filenames revert to their previous cases. Is there any way that I can convince git to adopt the altered filecases across the entire history of my repository, or at least convince it not...
10:00 richardlampitt ...to alter the cases of files already existing?
10:00 giraffe joined #git
10:01 jast richardlampitt: what does 'git config core.ignorecase' say?
10:01 RxMcDonald left #git
10:02 jast generally, changing the case of a file for all of your history means all clones have to be fixed, too, lest you end up with terrible merges of doom (!rewrite)
10:02 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
10:02 richardlampitt False. I explicitly altered this to false on the suggestion of a stackoverflow thread about the same problem, although I understand that it-s the default value in any case
10:02 richardlampitt (lol case, I made a pun)
10:03 tinti joined #git
10:04 richardlampitt I'm not aware of any clones of this repository. I'm the only person working on it
10:05 quackgyver joined #git
10:06 jast in that case you can go and have fun with git filter-branch (man git-filter-branch)
10:06 gitinfo the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html
10:06 jast renaming a file for all of history isn't exactly as straightforward as removing a file (unless you use --tree-filter which is _slow_), but maybe there are examples somewhere on the internet
10:07 govg joined #git
10:07 WayToDoor joined #git
10:07 jast okay, you had the same issues with ignorecase enabled?
10:08 richardlampitt Dunno, lemme try
10:08 tirej joined #git
10:09 phanimahesh joined #git
10:09 achadwick joined #git
10:11 mehola joined #git
10:12 richardlampitt Yup
10:12 mogio mogioidentify mogio jasjas
10:12 mogio lol
10:12 Ruler- How can I ignore multiple directories from `git diff`? I found how I could ignore one directory like this `git diff -- ':(exclude)Directory/'` but how can I make that ignore other directories also?
10:12 osse Ruler-: have you tried adding more :(exclude) ?
10:13 Ruler- what's the format?
10:13 mogio hello?
10:13 Ruler- just add `:(exclude)` after the first directory?
10:14 mogio joined #git
10:14 rez- joined #git
10:14 osse Ruler-: before each directory to ignore
10:14 richardlampitt jast: Yes, same behaviour, although at least now I can actually check out branches without having to create a temporary branch and then reset it to the commit I want
10:14 jast right
10:15 mogio joined #git
10:15 Ruler- osse, ahhhh. thx
10:15 jast I wasn't completely sure what ignorecase does and doesn't do. seems like you'll have to rewrite your history if case matters more than that.
10:15 mogio hi
10:16 jast mogio: have you changed your password? :}
10:16 richardlampitt Righteo. How would I go about doing that?
10:16 cdown_ joined #git
10:16 zefferno joined #git
10:17 mogio left #git
10:17 jast richardlampitt: here's a slow solution that may work if your repo isn't too big: http://stackoverflow.com/a/9458637/12471
10:17 jast this example is for just one branch
10:17 richardlampitt Small PHP app with perhaps a couple of hundred mainly tiny commits
10:17 atrigent joined #git
10:19 ramsub07 joined #git
10:19 jast and here's the fast version: http://stackoverflow.com/questions/1502858​0/filter-branch-index-filter-always-failin​g-with-fatal-bad-source/15029691#15029691
10:19 jast both should be fine in your case, I guess
10:19 richardlampitt yeah
10:20 mogio joined #git
10:20 mogio hi there
10:21 King_Hual joined #git
10:21 mogio how can I check, if the in .gitconfig defined settings are actually used by git?
10:22 sunri5e joined #git
10:22 borkr joined #git
10:22 canton7 mogio, you can put arbitrary settings in .gitconfig, which third-party tools use, so there's no check to say "this setting isn't known to git"
10:22 mogio to me "git config -l" seems as if it would just read out
10:22 canton7 mogio, what problem are you trying to solve?
10:22 jast mogio: right
10:23 mogio canton7, I have serious issues to run git on one of my client webspaces as the memory limit seems to be to low
10:23 mogio bit I changed it the /.git/config file
10:24 mogio and also in the .gitconfig
10:24 mogio still I receive the memory warning
10:24 mogio "fatal: Out of memory? mmap failed: "
10:24 rnsanchez joined #git
10:24 jast what did you change?
10:24 mogio sorry for my typing ;)
10:25 jast what setting, I mean
10:25 mogio [pack]  windowMemory = 1000m     packSizeLimit = 1000  threads = 1
10:25 mogio [core] packedgitwindowsize=1000m   packedgitlimit=1000m
10:26 jast is this a repo you cloned from somewhere else?
10:26 mogio yes...
10:26 jast possibly the original objects are already too big to fit into memory
10:26 mogio on other servers this is running e
10:26 jast these settings only affect re-compressing
10:26 mogio fine
10:26 Mowgli joined #git
10:27 mogio I receive the warning when pulling
10:27 mogio the memory warning
10:27 jast on the target machine?
10:27 jast (the one you are pulling to)
10:27 mogio how do I affect the memory limit for that
10:27 mogio yes the client git
10:28 mogio bash console of the webspace account
10:28 jast if the server is sending a pack with objects that don't fit into your client's RAM, the settings on the server need to be changed, not on your client
10:28 ToBeFree joined #git
10:28 jast and if any single object/file in the repo is too large for your client's memory, there's basically nothing you can do
10:28 richardlampitt >.> How do I escape from multiline in bash?
10:29 jast richardlampitt: ctrl-c? :)
10:29 richardlampitt Thanks ^^
10:29 mogio ok ... I understand..
10:29 mogio how do you know it is a single file that creates the issue
10:29 mogio ?
10:29 Velizar joined #git
10:29 mogio .. not the whole bunch of data
10:29 jast I don't know. just saying that *if* that's the problem, you can't solve it
10:30 jast if the problem is due to pack window size, you *can* solve it... by settings these things on the server
10:30 mogio ... ok but this was helfful
10:30 mogio p
10:30 jast just so you know, the pack window should be quite a bit smaller than your RAM
10:30 mogio the dev repo contains some psd's
10:30 jast git needs to keep other things in memory in addition to the pack window
10:30 Velizar Can I un-add the deletion of a file? I wish to commit something else, and commit the deletion of that file in a separate commit.
10:30 mogio the pro dows not... so I can try this instead
10:30 mogio does
10:30 jast Velizar: git reset -- <file>
10:31 jast 'git status' tends to have helpful hints about these things
10:31 Velizar huh that worked
10:31 jast of course it did :P
10:31 Velizar idk why it needed '--'
10:31 jast the -- is not always strictly necessary, it tells git "this is not a branch"
10:32 Velizar it didn't work when I tried it earlier without '--'
10:32 jast it's a good habit to get into because it's possible to get ambiguities
10:32 jast e.g. if one of your branch names is equal to a filename
10:32 sarri joined #git
10:32 sarri joined #git
10:33 jast and possibly some of the code paths aren't so smart to detect you're talking about a file. not sure. :)
10:34 finalbeta joined #git
10:35 ome joined #git
10:35 overlord_tm joined #git
10:35 sarri joined #git
10:35 sarri joined #git
10:36 mogio ls
10:36 mogio left #git
10:37 pity joined #git
10:38 diget joined #git
10:41 ferr1 joined #git
10:42 Laurenceb__ joined #git
10:42 Guest9 joined #git
10:44 Guest26 joined #git
10:44 nettoweb joined #git
10:45 manuelschneid3r joined #git
10:46 sarri joined #git
10:46 sarri joined #git
10:47 star_prone joined #git
10:49 Tobbi joined #git
10:54 hahuang61 joined #git
10:54 xall joined #git
10:55 tvw joined #git
10:56 star_prone joined #git
10:57 star_prone joined #git
10:58 drbean joined #git
10:58 star_prone joined #git
10:59 afuentes joined #git
11:00 geoid_ joined #git
11:01 theoceaniscool joined #git
11:02 Shashikant86 joined #git
11:05 timvisher joined #git
11:07 Cavallari joined #git
11:07 JeroenT joined #git
11:09 JeroenT_ joined #git
11:09 leeN joined #git
11:10 mostlybadfly joined #git
11:12 dersand joined #git
11:12 star_prone joined #git
11:13 rivarun joined #git
11:15 nowhereman joined #git
11:16 notebox joined #git
11:16 absinthe joined #git
11:17 chachasmooth joined #git
11:17 govg joined #git
11:17 thiago joined #git
11:19 ankit01ojha joined #git
11:19 Raed|Mobile joined #git
11:21 tvw joined #git
11:21 interrobangd joined #git
11:27 davimore_ joined #git
11:27 acetakwas joined #git
11:30 grift joined #git
11:32 moongazer joined #git
11:32 moongazer Hi
11:32 moongazer git checkout 'branch-name' is not changing the branch for me
11:32 moongazer I am operating git after 3 months and it worked earlier
11:33 canton7 moongazer, !repro
11:33 gitinfo moongazer: [!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
11:34 dsdeiz joined #git
11:34 rorro joined #git
11:34 aw1 joined #git
11:36 quesker_ joined #git
11:36 CaptainQuirk Tried a while before to change my system locale setting
11:37 CaptainQuirk it works for the whole ui but not for my terminal
11:37 grift joined #git
11:37 CaptainQuirk git is affected
11:37 CaptainQuirk I know that it's not git related
11:37 CaptainQuirk but I don't know on which channel to ask
11:38 bremner as your OS support?
11:38 tristanp joined #git
11:39 selckin running 'locale' should give you the relevant variables and their current value
11:41 timvisher joined #git
11:41 fakenerd joined #git
11:41 leah2 joined #git
11:43 ij left #git
11:44 joki joined #git
11:47 NeverDie When is it really worth it to do git rebase?
11:47 tvw joined #git
11:47 ^7heo NeverDie: of course not, you can totally rm -rf
11:47 ^7heo NeverDie: it compresses the data much better.
11:47 NeverDie Lol.
11:47 nuck joined #git
11:48 ^7heo NeverDie: unless you're not about compression, but versionning.
11:48 ^7heo NeverDie: then it is worth it to git rebase or git merge, your preference.
11:48 t0by joined #git
11:48 t0by joined #git
11:49 NeverDie Yeah I usually do git pull origin development && git checkout development && git merge branch-name
11:49 YuGiOhJCJ joined #git
11:49 NeverDie I was told to do git checkout branch-name && git rebase -i && git checkout development && git merge --no-ff branch-name instead
11:49 romerocesar joined #git
11:51 masuberu joined #git
11:52 Snugglebash joined #git
11:52 ertesx joined #git
11:53 JeroenT joined #git
11:54 cdown joined #git
11:57 chandu20 joined #git
11:57 byte512 joined #git
12:00 tvl joined #git
12:13 absinthe joined #git
12:14 a_thakur joined #git
12:15 overlord_tm joined #git
12:16 Gloomy joined #git
12:18 rez- joined #git
12:23 Jellyg00se joined #git
12:23 phanimahesh joined #git
12:23 timvisher joined #git
12:25 acetakwas joined #git
12:25 chipotle joined #git
12:29 aax joined #git
12:29 ramsub07 joined #git
12:30 thinkpad joined #git
12:31 Spherical joined #git
12:33 chipotle joined #git
12:34 NwS joined #git
12:35 str joined #git
12:36 geoid_ joined #git
12:40 fstd joined #git
12:40 fstd joined #git
12:42 mniip joined #git
12:45 a_thakur joined #git
12:45 Esya joined #git
12:46 finalbeta joined #git
12:47 tchan joined #git
12:47 tchan joined #git
12:47 eihabarabia joined #git
12:47 PioneerAxon joined #git
12:48 sdothum joined #git
12:50 tomek joined #git
12:50 tomek joined #git
12:51 rnsanchez joined #git
12:53 str joined #git
12:54 Kevin_ joined #git
12:54 hahuang61 joined #git
12:55 sunri5e joined #git
12:56 pkiller joined #git
12:56 a_thakur_ joined #git
12:56 Masber_080 joined #git
12:57 thiago joined #git
12:58 ojdo joined #git
12:58 Shashikant86 joined #git
13:00 timvisher joined #git
13:01 ReptilianBrain joined #git
13:05 pkiller joined #git
13:07 mischat joined #git
13:08 mischat joined #git
13:13 sunri5e joined #git
13:13 Endarked joined #git
13:14 theoceaniscool joined #git
13:15 chandu20 joined #git
13:16 rorro joined #git
13:20 lindenle joined #git
13:20 onmeac joined #git
13:21 dec0n joined #git
13:22 ertesx joined #git
13:22 s2013 joined #git
13:23 crose joined #git
13:24 justanotheruser joined #git
13:25 dfaught joined #git
13:27 sunri5e joined #git
13:27 mehola joined #git
13:29 ronny joined #git
13:30 sunri5e joined #git
13:31 Mogget joined #git
13:33 synthroid joined #git
13:33 humoroustwo joined #git
13:36 jeffreylevesque joined #git
13:37 ramsub07 joined #git
13:37 noecc joined #git
13:37 JanC_ joined #git
13:38 pbrewczynski joined #git
13:39 Snugglebash joined #git
13:39 tristanp joined #git
13:39 synthroi_ joined #git
13:39 theoceaniscool joined #git
13:39 siddart joined #git
13:41 baris joined #git
13:43 timvisher joined #git
13:43 humoroustwo joined #git
13:45 geoid_ joined #git
13:45 Kevin__ joined #git
13:46 s2013 joined #git
13:50 sea-gull_ joined #git
13:51 romerocesar joined #git
13:51 Raineer joined #git
13:52 aavrug joined #git
13:53 guampa joined #git
13:53 Cryp71c joined #git
13:54 hramrach joined #git
13:56 Gsham joined #git
13:59 tirej joined #git
13:59 mostlybadfly joined #git
14:00 Shashikant86 joined #git
14:01 nettoweb joined #git
14:02 todd_dsm joined #git
14:03 humoroustwo joined #git
14:07 theoceaniscool joined #git
14:07 jimi_ joined #git
14:08 eblip joined #git
14:08 Corosus joined #git
14:09 CaptainQuirk joined #git
14:09 Corosus joined #git
14:13 ablackack joined #git
14:13 davimore joined #git
14:14 Tobbi joined #git
14:15 Shashikant86 joined #git
14:15 dimi1947 joined #git
14:16 Endarked joined #git
14:16 rnsanchez joined #git
14:17 sunri5e joined #git
14:19 timvisher joined #git
14:21 gabi_ joined #git
14:22 jameser joined #git
14:23 Xatenev joined #git
14:24 Xatenev Hi, got following problem: we have translation files live & local, those files can be edited live and local. That means they become modified on the live server, thus we would cause a conflict when we would try a git pull.
14:24 selckin !deploy
14:24 gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
14:25 Xatenev On deployment, we basically want to add the translation files that have been changed on top of what we have. Basically a merge
14:25 Xatenev but merging on the live is probably not the best solution :/
14:25 Xatenev What could we do there? :P
14:25 selckin using git to deploy is already a bad idea
14:26 cdg joined #git
14:26 tobiasvl Xatenev: why are you changing files locally on the live server? also, yes, don't deploy with git
14:26 jast not necessarily
14:26 jast it's just needlessly complicated
14:26 Xatenev tobiasvl: Well, we have translation files that can be changed by editors
14:26 humoroustwo joined #git
14:26 jast but it definitely is an all-around bad idea to have people edit files on the live system
14:26 tobiasvl Xatenev: then either keep them outside of git, or make the changes through git
14:26 Xatenev tobiasvl: I guess the best solution is just to do it twice, live and local and have it gitignored?
14:26 ssitani joined #git
14:27 jast there's simply no way around things be hard to keep up-to-date with a workflow like that
14:27 jameser joined #git
14:27 Xatenev jast: makes sense
14:27 jast plus whenever a file is half-written, the live environment will be broken for a short while
14:27 tobiasvl Xatenev: if you have them gitignored you would only need to do it once – git wouldn't need (or have) a copy
14:27 jast in the case of a pull or something like that, people might end up seeing conflict markers in the live env
14:27 tobiasvl Xatenev: but see !config for solutions like that
14:27 gitinfo Xatenev: [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas
14:27 Xatenev tobiasvl: I mean the developer would have to do it twice
14:27 Xatenev tobiasvl: Change the translation file once locally to test, and then live again
14:28 Xatenev tobiasvl: because the changes are not tracked via git
14:28 jast I can see the appeal of fixing things within a running system, though :) so maybe set up a staging environment for this kind of stuff
14:28 jast developers can then commit the changes made directly in the system, push them, and then you can do a clean update in the live env
14:29 ssitani Hi Paul
14:29 gitinfo ssitani: 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.
14:30 ssitani sorry, wrong channel :)
14:30 gitinfo sorry for auto-responding, then :)
14:30 Xatenev jast: yeap that would be nice
14:30 tobiasvl haha gitinfo achieved sentience
14:30 gitinfo tobiasvl: you must be not very sentient if it took you that long to notice
14:31 * mgedmin is moving away slowly
14:31 ssitani gitinfo: actually your auto-reply gave me some valuable info Thanks:)
14:31 gitinfo ssitani: you're welcome, but please note that I'm a bot. I'm not programmed to care.
14:31 ssitani ^^give this bot a cookie :)
14:32 Xatenev thanks for the help
14:32 dec0n joined #git
14:33 jast !botsnack
14:33 gitinfo Om nom nom
14:34 al-damiri joined #git
14:35 davimore_ joined #git
14:36 tobiasvl oooh, a bot insulted me
14:38 nettoweb joined #git
14:38 osse surely that breaks one of the three laws
14:38 kpease joined #git
14:39 pity joined #git
14:40 perlpilot only if "harm" includes "to the psyche"
14:41 Gsham joined #git
14:41 osse is there presedence in the corpus?
14:42 * osse speed-reads all novels and short stories
14:42 def_jam joined #git
14:44 timvisher joined #git
14:45 thiago joined #git
14:45 bejeb joined #git
14:46 GodGinrai joined #git
14:46 GodGinrai joined #git
14:46 Dougie187 joined #git
14:46 fahadash joined #git
14:47 notebox joined #git
14:47 gabi__ joined #git
14:48 lordjancso joined #git
14:52 mogio joined #git
14:52 mogio funny thing :)
14:53 kyan joined #git
14:53 mogio is it possible that git will pull on an initial pull even files that are mentioned in the .gitignore?
14:53 todd_dsm hey guys, I'm trying to manage a custom Vim build, using git submodules. No matter what though, git status give messages like "untracked content" / "modified content". I have a gist: https://gist.github.com/todd-dsm/​20f11e3c57ee7a53ecbe26782908b9e7. How do you get past this issue to get a solid commit?
14:53 osse mogio: yes
14:54 tobiasvl mogio: !ignore_tracked
14:54 gitinfo mogio: Git only applies ignore patterns to untracked files. You can't use ignore patterns to ignore changes to files that are already tracked by git. To remove files only from git, but keeping them on disk, use git rm --cached <file>. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.
14:54 tvw joined #git
14:54 mogio ok... thats somehow nice... but does it not create problems?
14:54 mentazoom joined #git
14:55 Es0teric joined #git
14:55 osse like what kind?
14:55 acetakwas joined #git
14:55 osse Once a file as been added to the repo in a commit .gitignore does no longer apply
14:55 mogio ok... I am just wondering as the temp folder in pulled but not the config and settings files
14:55 Gloomy joined #git
14:55 hahuang61 joined #git
14:55 tristanp joined #git
14:56 mogio ah ok I see
14:56 mogio so they are downloaded as the online repo just contains them
14:57 mogio but they are ignored straight away by my local git as they are mentioned in the local .gitignore ?... right?
14:57 osse mogio: depends on what you mean by ignored in that question
14:57 pity joined #git
14:57 osse actually, i am pretty sure the answer is no in any case
14:58 Xatenev left #git
14:58 l33ch joined #git
14:58 Snugglebash joined #git
14:59 mogio hmmm... when the gitignore does not change the way already uploaded files are treated ... I then have to remove them from my tracked file list?
14:59 Snugglebash joined #git
14:59 reactormonk joined #git
14:59 mogio to stop them from being updated if necessary?
14:59 bannakaffalatta joined #git
14:59 reactormonk Can I "unmerge" a specific file, aka reset it before I applied the mergetool?
15:00 osse reactormonk: git checkout --merge file
15:00 Snugglebash joined #git
15:00 mogio or does they just exist because they where uploaded at a certain point ..?
15:00 mogio but they will not be updated
15:00 thebope joined #git
15:00 Dougie187 mogio: Git tracks anything that's been explicitly added to the repo.
15:00 tobiasvl yes, you need to remove them from being tracked
15:00 Dougie187 even if it's been added to .gitignore already
15:00 osse mogio: once it is committed .gitignore is *irrelevant*
15:01 Snugglebash joined #git
15:01 thiago _ikke_: right, fc00::/8 hasn't, but fd00::/8 has
15:01 Jackneilll joined #git
15:01 chipotle joined #git
15:01 Snugglebash joined #git
15:02 mogio but having them in the gitignore does stop commits to contain them if there were updated
15:02 Snugglebash joined #git
15:02 osse no
15:02 mogio .. just complete pull by a new systems does download them... right?
15:03 Dougie187 mogio: No. It prevents you from adding them to the repo.
15:03 Dougie187 mogio: .gitignore attempts to prevent things from being added to the repo. Once they are in the repo, it'll track them and changes to them.
15:03 Snugglebash joined #git
15:03 tobiasvl and even then you can still add them with git add -f
15:03 Dougie187 ^^
15:03 Dougie187 That's the *attempts* part.
15:04 tobiasvl yeah
15:04 Snugglebash joined #git
15:04 e14 joined #git
15:05 Snugglebash joined #git
15:05 mogio hmm are u sure... ? with me a settings.php is not uploaded when changed "add" does ignore it, although it was initially tracked
15:05 xkr47 joined #git
15:05 Snugglebash joined #git
15:06 scarcry joined #git
15:06 osse mogio: does 'git update-index --assume-unchanged' ring a bell?
15:06 YourTheBombNow23 joined #git
15:06 l33ch does anyone know of a git service that sends a message to IRC when a push is detected for example?
15:06 timvisher joined #git
15:06 Snugglebash joined #git
15:06 tobiasvl l33ch: there are plenty, just google "git bot irc"
15:07 l33ch ok i'll google, but so far my searches have come up empty.. :(
15:07 Dougie187 mogio: https://gist.github.com/douglasjacob​sen/53511ab16100f7d34ccd86f1d807eb77
15:07 Dougie187 mogio: there is a test, to show you. Once it's added to the .gitignore, if it was previously added to the repo it is tracked still.
15:08 mentazoom joined #git
15:09 griffindy joined #git
15:09 mogio ok..
15:09 mogio Dougie187, thanks... I understand...
15:10 mogio so a post adding is not picked up once a file was already tracked
15:11 mogio "adding to gitignore of course"
15:11 lindenle joined #git
15:11 thiago a tracked file cannot be ignored
15:11 Ryanar joined #git
15:11 giraffe joined #git
15:11 mogio so its needed to remove them also from the tracked files list of git
15:11 thiago a file can be in one of three states: tracked, untracked & unignored, unracked & ignored
15:13 mogio helpful thanks
15:13 marcogmonteiro joined #git
15:14 AaronMT joined #git
15:16 pietv joined #git
15:20 dd joined #git
15:26 Gsham joined #git
15:26 ash_workz joined #git
15:27 saidi joined #git
15:29 kevwil joined #git
15:29 netj joined #git
15:30 timvisher joined #git
15:30 pity osse: Sorry, today is really busy.
15:30 nickabbey joined #git
15:32 pity osse: git is /usr/local/bin/git. Before, it was Apple compiled. After, I installed git by homebrew.
15:32 HcsOmot joined #git
15:33 bannakaf_ joined #git
15:33 pity 23:32:25.253089 git.c:350               trace: built-in: git 'checkout' 'test'
15:33 pity Switched to branch 'test'
15:33 GodGinrai pity: use a pastebin or gist
15:33 pity osse: GIT_TRACE=1 git checkout test it shows just a few lines
15:34 osse and it still takes 30 seconds?
15:34 pity GodGinrai: Just 2 lines
15:34 chandu20 joined #git
15:34 pity osse: yes
15:34 osse pity: ls .git/hooks
15:35 pity osse: it dose not happen in one repo, but in any repo.
15:35 pity osse: 23:35:17.563537 trace.c:435             performance: 30.106185000 s: git command: 'git' 'checkout' 'master'
15:35 HcsOmot hi, i've created a new branch, did work, and realized some of that work needs to be on a separate branch. nothing's yet neither staged nor commited. I've got a clean branch with a bunch of changes waiting to be staged. how do I proceed?
15:36 osse pity: is one repo huuuuuuuuuuuuuuuuge ?
15:36 jast HcsOmot: this is actually easy. 'git checkout -b myseparatebranch', selectively stage and commit, 'git checkout -'
15:36 jast !float
15:36 gitinfo If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there.  If the files you changed differ between branches, the checkout will fail.  In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.
15:36 pity osse: I use dtruss to follow the calls, there's some kevent calls
15:37 jast (no -b if the branch already exists, of course)
15:37 pity osse: no, the repo is very tiny, only on file, with 32 commits.
15:38 BadSnappa joined #git
15:39 madewokherd joined #git
15:39 netj joined #git
15:39 HcsOmot ok, i've done git checkout -b new_branch, and right after I hit enter, I got a list of all my changed files with "M" in front of them. what does that mean?
15:39 MineCoins joined #git
15:39 dreiss joined #git
15:40 Hedgework Modified
15:40 grilled-cheese joined #git
15:42 pity osse: well, I found that!
15:43 HcsOmot ok, so I'm on my new branch now that's supposed to hold some of those changes. can I do more changes to those files while still on this new branch?
15:43 pity osse: because user.email is not set
15:43 ferr2 joined #git
15:45 LJ joined #git
15:45 LJ left #git
15:46 eljay joined #git
15:47 pity osse: I wonder why not one line in dtruss log metioned user.email....
15:48 aviraldg joined #git
15:48 aviraldg joined #git
15:48 Endarked joined #git
15:48 netj joined #git
15:48 grill_ joined #git
15:51 timvisher joined #git
15:51 d^sh joined #git
15:51 ianmethyst joined #git
15:52 kpease joined #git
15:54 manuelschneid3r joined #git
15:55 lvh joined #git
15:56 Masber_080 joined #git
15:56 hahuang61 joined #git
15:57 theoceaniscool joined #git
15:57 jast HcsOmot: yeah. basically uncommitted changes aren't tied to any branch, Git keeps carrying them around when you switch
15:58 marcogmonteiro joined #git
15:59 xall joined #git
15:59 mehola joined #git
15:59 raspado joined #git
16:01 davimore__ joined #git
16:02 zivester joined #git
16:03 lvh joined #git
16:03 humoroustwo joined #git
16:04 rominronin joined #git
16:05 sgfgdf left #git
16:05 LionsMane joined #git
16:09 flavius joined #git
16:10 TomyWork joined #git
16:11 Shashikant86 joined #git
16:11 LeBlaaanc joined #git
16:12 freimatz joined #git
16:14 Soa joined #git
16:14 zacts joined #git
16:17 HcsOmot quick question - how do I unstage changes?
16:17 emg joined #git
16:17 synthroid joined #git
16:17 HcsOmot I just want to not commit something I staged, but want to commit it later
16:17 _ikke_ git status would tell
16:18 HcsOmot i suck at git status :(
16:18 _ikke_ git reset -p you can selectively unstage
16:18 _ikke_ git reset <file> would unstage the entire file
16:18 _ikke_ (all changes to the file)
16:18 alphor joined #git
16:19 HcsOmot i want to keep the changes, just postpone them to another commit.
16:19 tirej joined #git
16:19 brent__ joined #git
16:19 _ikke_ This would do that
16:19 timvisher joined #git
16:19 randuser99 joined #git
16:19 GodGinrai HcsOmot: `git reset` doesn't delete your changes unless you use the --hard flag
16:20 HcsOmot i've changed 3 files and staged them. now I want to unstage 1 but keep the changes and commit that one separately.
16:20 HcsOmot ooooh
16:20 HcsOmot I see.
16:20 HcsOmot thank you
16:21 l33ch HcsMot: you could also use git stash or git shelve
16:22 GodGinrai l33ch: git stash is for unstaged changes
16:22 treia joined #git
16:23 GodGinrai l33ch: and there is no such thing as `git shelve`
16:24 l33ch GodGinrai: oh nvm then
16:24 GavinMagnus joined #git
16:26 Tobbi joined #git
16:27 finalbeta joined #git
16:28 Meteorhead joined #git
16:29 Goplat joined #git
16:30 ianmethyst joined #git
16:30 lvh joined #git
16:32 gugah joined #git
16:33 HcsOmot aaand here we go again. I've don goofed
16:34 HcsOmot i was on branch called 8-feeding_time, which was based of upstream/master
16:35 HcsOmot did work, noticed i need another branch. created a new branch 7.5-fix_bugs off 8-feeding_time.
16:35 HcsOmot now I tried pushing 7.5 to origin/master, but can't
16:36 HcsOmot git push -u origin/7.5-fix_bugs fatal: 'origin/7.5-fix_bugs' does not appear to be a git repository fatal: Could not read from remote repository.  Please make sure you have the correct access rights and the repository exists.
16:36 HcsOmot any way to fix?
16:36 ssitani joined #git
16:37 canton7 yeah, learn the syntax of git push ;)
16:37 canton7 'git push -u origin 7.5-fix_bugs'
16:37 canton7 git push <place to push to> <what to push>
16:37 canton7 "origin/7.5-fix_bugs" is a !remote_tracking_branch - it's a reflection of where we last saw the branch '7.5-fix_bugs' on the remote 'origin'.
16:37 gitinfo Remote-tracking branches (branches which start with e.g. 'origin/', listed by 'git branch -r') are read-only mirrors of the corresponding branches in another repository. They're updated by 'git fetch'. You can't edit them directly (trying to check them out results in a !detached HEAD), but you can create a new local branch based on a remote-tracking branch using e.g. 'git checkout -b <branch> <remote>/<branch>'
16:38 Darren_ joined #git
16:39 umpc joined #git
16:40 robotroll joined #git
16:41 miczac joined #git
16:41 a_thakur joined #git
16:41 rgrinberg joined #git
16:41 GodGinrai HcsOmot: canton7 explained it pretty well.  Basically, you specify the remote you are pushing to, and then what branch you want to push to that remote
16:41 * HcsOmot feels embarrassed...
16:42 kyan joined #git
16:42 f3r70rr35f joined #git
16:43 jccn joined #git
16:43 Zamyatin joined #git
16:44 thiago_ joined #git
16:44 theoceaniscool joined #git
16:44 HcsOmot you guys... any chance you could help me wrap this up?
16:44 HcsOmot I managed to push it to origin
16:44 GodGinrai `git wrap`
16:44 GodGinrai jk
16:44 HcsOmot now I need to create a pr from that new branch to upstream
16:44 timvisher joined #git
16:45 HcsOmot GodGinrai: tried that, didn't work :/
16:45 aax joined #git
16:45 GodGinrai anyway, are you looking for a traditional git PR?  or is this on github or something?
16:45 xall_ joined #git
16:45 ajf- joined #git
16:46 _ikke_ GodGinrai: 95% of the time when people mention PRs here, they mean github PRs :P
16:46 HcsOmot github, sorry
16:46 rgrinberg Is it possible to revert a commit by only applying the diff that would revert it to the working area or index?
16:46 Shashikant86 joined #git
16:46 HcsOmot _ikke_: +1 here obviously
16:46 grawity rgrinberg: git revert --no-commit
16:46 Snugglebash joined #git
16:47 GodGinrai _ikke_: it's best to confirm, tho'  I can't help them if they need to know how to do it on gitlab or bitbucket
16:47 rgrinberg grawity: Awesome! thank you
16:47 Cavallari joined #git
16:47 _ikke_ GodGinrai: Al those systems work mostly the same
16:47 _ikke_ All
16:47 _ikke_ Few minor differences, but the process is the same
16:48 absinthe joined #git
16:48 HcsOmot so I'm on my github profile, in my repo, on my 7.5 branch. I hit "compare & pull request". the "base fork" and "head fork" are what confuses me.
16:48 jnavila joined #git
16:48 HcsOmot "base fork" is the place I want to send my changes to, right?
16:49 HcsOmot and "head fork" is where I'm sending from, right?
16:49 [0xAA] joined #git
16:49 _ikke_ right
16:49 GodGinrai HcsOmot: "base" should be upstream, "head" should be your fork.  Then just choose the master branch on base, and your 7.5 branch on your fork
16:49 * _ikke_ never noticed the terms base and head fork on github
16:49 sharpy joined #git
16:50 nickabbey joined #git
16:52 humoroustwo joined #git
16:53 Guest42 joined #git
16:54 ResidentBiscuit joined #git
16:55 stick-shift joined #git
16:55 anddam joined #git
16:56 anddam hello
16:56 anddam I had a own fork of a public repo, I had my own fork as "origin" remote and the project one as "upstream"
16:57 sharpy joined #git
16:57 anddam I switched the fetch url or origin to the one of the public, since I only need to push on my fork but always pull from upstream
16:58 [0xAA] joined #git
16:59 larske joined #git
16:59 tirej joined #git
16:59 HcsOmot guys, thanks a lot! all done. cheers!
17:00 anddam I then killed a rebase while I noticed it was working heavily, while I expected it to peacefully switch to the other branch
17:00 anddam now I get https://gist.github.com/anddam/3​7b1182137febeb94249a152c0766bdb
17:00 anddam I believe evil forces are at work, can anyone help me out figuring what's happening?
17:00 _ikke_ Nothing evil is going on
17:01 anddam _ikke_: mmm, not sure about that
17:01 _ikke_ git was busy with a rebase, and you interrupted it
17:01 anddam that's correct
17:01 _ikke_ so git is in a middle-of-a-rebase state
17:01 anddam what's the best next move?
17:01 _ikke_ git rebase --abort
17:01 anddam or the least bad one
17:01 _ikke_ That should return you to the situation before the rebase
17:01 anddam ok, my goal is to just checkout upstream/master that now has the same url as origin/master
17:02 anddam I got burned since lately I pulled without rebasing and ended up duplicating 300 commits, that then I gladly pushed upstream
17:02 _ikke_ did upstream rebase?
17:02 anddam so I have an origin where push is my own fork, first, so I have to manually check
17:02 thiago joined #git
17:02 anddam check the command for pushing upstream*
17:03 anddam _ikke_: do you mean locally here on my computer?
17:03 _ikke_ no, I mean the public project
17:03 anddam no they kept the duplicates
17:03 anddam I ran --abort, and now I'm     HEAD detached from refs/heads/master
17:03 anddam now I can checkout upstream/master, right?
17:04 ikelso joined #git
17:04 _ikke_ You need to create 'local' branch for that (or overwrite an existing one)
17:04 anddam I have wip, for any kind of mess-up
17:04 voldz1tic3n joined #git
17:04 rominronin joined #git
17:05 anddam _ikke_: why do I need a local branch btw?
17:05 todd_dsm hey guys, using a bunch of submodules to maintain a custom vim config. Is there a method of verifying that you have the latest code from a given repo/submodule?
17:05 _ikke_ anddam: You work on local branches. (the one prefixed with origin or upstream are called remote tracking branches)
17:05 e14 joined #git
17:06 anddam ok, I have master and "wip" locally
17:07 jagob joined #git
17:07 anddam _ikke_: that satisfy the previous statement of creating a local branch, right?
17:07 anddam i.e. you didn't mean I had to create a new one right now
17:08 _ikke_ anddam: My comment was based on you wanting to checkout upstream/master
17:08 finalbeta joined #git
17:08 _ikke_ which would leave you in a detached HEAD
17:08 anddam ah ok, while 'checkout master' wouldn't since that's the local branch, right?
17:08 _ikke_ correct
17:09 durham joined #git
17:09 anddam _ikke_: point is I learned about checking log upstream/master..master last time (when I got burned) since that would have show the number of changesets pushed
17:09 anddam now that I checked out master I see
17:09 anddam $ git log master..origin/master | wc -l
17:09 _ikke_ Right, that's perfectly valid
17:09 anddam 5052
17:09 anddam and this freaks me out a bit
17:10 _ikke_ So you're a bit behind :P
17:10 anddam I'm not sure if this is because my own fork, from which I'm "upgrading" was stuck a couple weeks before or what else
17:10 anddam yes, that's fine
17:10 anddam the workflow I want is pull locally, change, push on own fork on github, open PR
17:11 _ikke_ Right
17:11 _ikke_ So best is to not work on master at all
17:11 _ikke_ (you can keep it up-to-date from upstream)
17:11 anddam I understand that actually learning how git works would be the proper approach, but right now i'm otherwise worried
17:12 anddam _ikke_: and working where, if not in master?
17:12 hasc joined #git
17:12 _ikke_ What are you worried about?
17:12 _ikke_ The easiest workflow is to always use seperate branches for each PR
17:12 _ikke_ ^^
17:12 _ikke_ These are so called topic branches
17:12 hasc joined #git
17:13 timvisher joined #git
17:15 ramsub07 joined #git
17:15 anddam I'm worried about duplicating hundreds of commits again, once I checkout master I expected 'log master..origin/master' to have no output
17:15 anddam the earlier use of 'worried' was actually 'busy'
17:16 _ikke_ the branch currently checked out has no influence on the output of that command
17:16 _ikke_ because you explicitly mention the branches
17:16 anddam 'Your branch and 'origin/master' have diverged, and have 13 and 747 different commits each, respectively.'
17:17 _ikke_ anddam: Do you have anything in your local master you want to keep?
17:17 anddam no
17:17 anddam I may have stashed some stuff
17:18 _ikke_ ok
17:18 djb-irc joined #git
17:18 anddam but nothing critical, I figure the stash is separated from branches
17:18 _ikke_ then you probably just want to make it equal to origin/master
17:18 _ikke_ anddam: correct
17:18 anddam could only have possible issues reapplying
17:18 anddam but I don't have anything in local master
17:19 anddam I mean no changes I want to keep
17:19 _ikke_ git checkout -B master origin/master
17:19 _ikke_ this would recreate master based on origin/master
17:20 anddam since the previous local master branch was the one left behind, that I have in my own github fork
17:20 anddam I see the point, thanks
17:20 anddam git log origin/master..master | wc -l
17:20 anddam 0
17:20 anddam ahhhhh
17:20 nidr0x joined #git
17:20 synthroid joined #git
17:21 anddam _ikke_: now I feel better and can understand what was going on
17:21 anddam the huge number of commits was for my left-behind branch, right?
17:21 anddam i.e. pushing that would have brought forward my own fork
17:22 _ikke_ You had 13 commits that origin did not have
17:22 _ikke_ and origin had 747 commits you didn't have yet in your local master branch
17:23 anddam btw i thought I had committed those
17:23 _ikke_ commits happen locally
17:23 govg joined #git
17:23 _ikke_ only after you push will they go to the remote
17:23 ramsub08 joined #git
17:24 anddam now should    checkout -B master mine/master     work and restore the previous situation with the 13/747 commits difference?
17:24 anddam mine is a further remote where i have only my github fork URLs
17:24 anddam upstream remote has only the project ones
17:24 anddam and origin is hybrid, fetch from project, push on fork
17:24 anddam does that even make sense?
17:28 crayon joined #git
17:29 cagmz joined #git
17:29 _ikke_ That command I gave you just ignores what you have localy for master, and just points master to whatever origin/master points to
17:33 Electrometro joined #git
17:34 TomyLobo joined #git
17:34 canuk joined #git
17:35 nanga joined #git
17:35 anddam yes, I read the man page
17:35 anddam I was trying to walk back, as experiment
17:35 anddam given that my local branch was pushed to mine/master
17:36 canuk how can I exclude a file pattern from my "git diff"? (e.g .min.css)
17:36 timvisher joined #git
17:36 canuk if I want everything *but* that
17:36 grawity echo "*.min.css -diff" >> .git/info/attributes
17:36 grawity marks those as binary, essentially
17:36 anddam I figure     git checkout -B master mine/master    should get me back to the previous master branch, isn't that right?
17:37 canuk grawity: but it will still version control it right?
17:37 anddam the reason I ask is because I tried and got "fatal: Cannot update paths and switch to branch 'master' at the same time." instead
17:37 grawity yes
17:37 canuk i dont want to .gitignore it
17:37 canuk ok
17:37 anddam oh I see, I hadn't fetched mine/master
17:37 _ikke_ anddam: right
17:38 anddam btw why sometimes I can refer to a branch as   mine/master, like in the checkout
17:38 _ikke_ anddam: this is what git calls 'do what I mean'
17:38 anddam but when i fetch I have to use  "<remote> <branch>"  separated by space?
17:38 GodGinrai canuk: he said attributes, not ignore
17:38 _ikke_ anddam: Two separate things
17:38 grawity anddam: in `git checkout` you're referring to a branch that's *already fetched*
17:38 canuk GodGinrai: yeah just checking, not really familiar with attributes
17:38 _ikke_ anddam: git fetch requires 2 things: a remote and a branch (refspec)
17:39 _ikke_ anddam: git fetch <remote> <refspec>
17:39 xall_ joined #git
17:39 grawity anddam: i.e. the branch 'master' was already fetched to a local copy named 'origin/master' and you're checking out that local copy
17:39 s2013 joined #git
17:39 tirej joined #git
17:39 anddam as newcomers I have to say that's confusing, there should be a unique syntax
17:39 anddam IMHO
17:39 grawity basically `git checkout` doesn't have a clue about remotes
17:39 grawity it lets `git fetch` deal with those
17:39 anddam newcomer*
17:40 grawity anddam: maybe, but Git has slightly different priorities
17:40 anddam Your branch and 'origin/master' have diverged, and have 1 and 747 different commits each, respectively.
17:40 anddam fair point
17:40 grawity though it *does* have a certain amount of DWIM
17:40 anddam _ikke_: so seems that i didn't have something pushed
17:40 grawity (like `git diff a..b` vs `git diff a b`)
17:40 anddam no big deal
17:40 Duder9999 joined #git
17:41 anddam grawity: DWIM is the just mentioend 'do what i mean', but I'm not sure how that would apply in those two cases
17:41 bernardio joined #git
17:41 grawity automatically choosing the most likely meaning of e.g. `git checkout origin master`
17:41 svm_invictvs joined #git
17:41 grawity which currently means "check out file 'master' from local branch 'origin'"
17:41 sdothum joined #git
17:42 grawity ...well, I guess `git fetch origin/master` would be easier
17:42 AlineGomes joined #git
17:42 grawity it's less ambiguous
17:42 fahadash joined #git
17:42 anddam I think I get the point
17:43 anddam now that I somehow have my old local master branch I can    pull --rebase   and push on my own fork, this would bring my fork forward, right?
17:43 diogenese joined #git
17:44 thiago yes
17:46 anddam I have 760 commits listed in    git log mine/master..master
17:47 anode joined #git
17:47 anode left #git
17:48 DolphinDream joined #git
17:48 anddam so I should be able to push
17:48 DolphinDream howdy
17:48 anddam and another "oh jeez" moment
17:48 chandu20 joined #git
17:48 nettoweb joined #git
17:49 DolphinDream any ideas why my __git_ps1 is missing even after installing the bash-completion package (Ubuntu 16) ?
17:49 alip_ left #git
17:49 anddam hint: Updates were rejected because the tip of your current branch is behind its remote counterpart. Integrate the remote changes (e.g. 'git pull ...') before pushing again.
17:50 anddam but if I   pull mine master   I'll rewind all the 760 commits I just pulled
17:51 nickabbey joined #git
17:52 aavrug joined #git
17:52 GodGinrai anddam: no
17:52 GodGinrai anddam: pull doesn't rewind changes
17:54 anddam that may very true, I use the verb "rewind" since I read 'First, rewinding head to replay your work on top of it...'
17:54 anddam and then a bunch of 'Applying:…'
17:54 overlord_tm joined #git
17:54 GodGinrai that's a rebase
17:54 thiago anddam: you have 760 more commits, but you didn't check how many more your repository has.
17:54 thiago i.e., the two have diverged
17:54 anddam thiago: none?
17:55 thiago after a rebase, that's correct
17:55 thiago but push told you that you didn't rebase
17:55 anddam I had rebased origin master, then tried to push to mine master
17:56 anddam git warned me the tip was ahead (not sure why) so I pull-rebased mine master as well
17:56 anddam now this is mostly TIAS, a mad monkey would have tried the same combination
17:58 timvisher joined #git
17:59 wCPO joined #git
17:59 wCPO joined #git
18:00 mehola joined #git
18:00 namname joined #git
18:00 hahuang61 joined #git
18:00 j416 ronny: why is it broken?
18:02 marcogmonteiro joined #git
18:02 aavrug joined #git
18:02 boombatower joined #git
18:04 AustinMatherne joined #git
18:06 e14 joined #git
18:08 anddam ha, I did it again, this time on my own fork
18:08 hashpuppy joined #git
18:09 anddam how do I revert these duplicated commits? https://github.com/anddam/ma​cports-ports/commits/master
18:09 anddam I'm fine wiping the repo off but since the fork is just a sandbox I may as well learn how to revert this
18:09 anddam even better avoiding it in first place
18:10 xall_ joined #git
18:11 mentazoom joined #git
18:11 Noldorin joined #git
18:15 mikeric joined #git
18:16 ronny j416: its human readable output, its a mess to make parsable, using the other command thats for machine consumption is orders of magnitude better
18:17 nettoweb1 joined #git
18:17 ronny j416: by definition is fragile to ask for a mess and fix it up using grep and sed when there is a way to ask for a fixed up starting point
18:18 mikeric joined #git
18:19 dreiss joined #git
18:19 Endarked joined #git
18:19 tirej joined #git
18:20 timvisher joined #git
18:20 thierryp joined #git
18:21 thierryp joined #git
18:22 nostrora joined #git
18:22 duderonomy joined #git
18:23 acetakwas joined #git
18:23 thierryp joined #git
18:25 PioneerAxon joined #git
18:25 Es0teric joined #git
18:27 theoceaniscool joined #git
18:27 Darcidride joined #git
18:27 ski7777 joined #git
18:29 chardan joined #git
18:30 e14 joined #git
18:33 j416 ronny: it's perhaps not optimal. I've just done git branch -r --merged, grepped away HEAD, fed that to xargs to delete branches
18:33 j416 works like a charm
18:34 gugah joined #git
18:35 j416 nice how git-for-each-ref has quoting for different languages
18:36 rominronin joined #git
18:37 vdv joined #git
18:37 kulelu88 joined #git
18:40 Guest42 joined #git
18:42 Rodya_ joined #git
18:42 tyreld joined #git
18:43 done joined #git
18:43 NeverDie joined #git
18:43 timvisher joined #git
18:44 mentazoom joined #git
18:46 ronny j416: but the point is, why should i programm with a output not intended for programming when i can have a output for programming
18:47 j416 because it can be easier for one-off things
18:47 ronny from my pov its like when given the choice between meat and rotten meat to eat, eating the rotten meat
18:47 ronny j416: even for the one off the programmer output is easier
18:47 jost_ joined #git
18:48 j416 dude, I'm just giving you options.
18:49 ronny you keep repeating that a demonstrable mess that works worse than the prior given alternative is "just fine"
18:49 bill99 joined #git
18:51 unbalanced joined #git
18:51 Epx998 joined #git
18:51 cyan__ joined #git
18:52 dave0x6d joined #git
18:53 kpease joined #git
18:53 magellanicloud joined #git
18:55 marcogmonteiro joined #git
18:57 theoceaniscool_ joined #git
18:58 kyan joined #git
18:58 dsantiago joined #git
18:58 graingert joined #git
18:58 guampa joined #git
18:58 j416 because it has worked fine for my use-case.
18:59 tirej joined #git
18:59 ReptilianBrain joined #git
19:01 j416 I don't get why you're so seemingly upset about it. If it doesn't suit you, just don't use it.
19:03 publio joined #git
19:04 miklcct joined #git
19:05 timvisher joined #git
19:08 Zamyatin joined #git
19:09 Sasazuka joined #git
19:11 donught joined #git
19:11 dsantiag_ joined #git
19:11 ertes joined #git
19:11 druonysus joined #git
19:11 donught joined #git
19:12 Remramm joined #git
19:12 Balliad joined #git
19:13 dsantiag_ joined #git
19:13 Remramm joined #git
19:13 WakiMiko joined #git
19:14 bannakaffalatta joined #git
19:15 siddart joined #git
19:18 mablae joined #git
19:19 dreiss joined #git
19:19 nanga joined #git
19:22 aax joined #git
19:24 larske joined #git
19:27 timvisher joined #git
19:27 InfoTest joined #git
19:28 aw1 joined #git
19:30 yottabyte joined #git
19:30 yottabyte what's the difference between git fetch and git fetch origin?
19:31 jarshvor joined #git
19:31 jimi_ joined #git
19:32 Rodya_ joined #git
19:32 yottabyte will git fetch only fetch my local branch? where does it put it?
19:32 yottabyte because I know after I have to do merge or whatever
19:33 grawity "fetching local branches" already sounds like nonsense
19:33 cdown joined #git
19:34 nickabbe_ joined #git
19:34 doadet joined #git
19:34 yottabyte I mean my current branch
19:35 Random832 yottabyte, it fetches the remote which is the upstream of the current branch, or origin if none is defined.
19:36 nickabb__ joined #git
19:36 zivester joined #git
19:36 absinthe joined #git
19:37 Random832 and it stores it in FETCH_HEAD and in refs/remotes/[remote name]/[branch name]
19:37 yottabyte so when I do something like "git fetch origin && git pull origin some-branch," to try to bring my branch up to date, like merge that other branch into mine, if there are merge conflicts, how are those resolved? would I commit it in my branch and then just push? I don't need to pull the other branch again?
19:38 Random832 "git pull" is just fetch followed by merge
19:38 Random832 but yes you can commit it in your branch and then push
19:39 Random832 and for some workflows it may be beneficial to use git pull --rebase
19:39 tirej joined #git
19:44 m0viefreak joined #git
19:46 KV joined #git
19:46 cyan__ joined #git
19:49 ResidentBiscuit joined #git
19:49 KV joined #git
19:50 Guest42 joined #git
19:50 timvisher joined #git
19:50 hhee2 joined #git
19:51 yul joined #git
19:52 absinthe joined #git
19:53 romerocesar joined #git
19:53 ace_33 joined #git
19:55 thiago yottabyte: if there are conflicts, they are solved by you
19:57 Hrf joined #git
19:58 Atm0spher1c joined #git
19:58 druonysus joined #git
20:00 mischat joined #git
20:00 proq joined #git
20:00 Guest42 joined #git
20:00 hamster joined #git
20:03 jorj joined #git
20:04 nickabbey joined #git
20:04 gugah joined #git
20:05 siddart joined #git
20:05 moei joined #git
20:05 synthroi_ joined #git
20:06 Emperor_Earth joined #git
20:07 rominron_ joined #git
20:09 tfrangio2 joined #git
20:10 _Cyclone_ joined #git
20:12 WakiMiko joined #git
20:13 mozzarella joined #git
20:13 Sasazuka_ joined #git
20:13 beneter joined #git
20:13 pR0Ps joined #git
20:15 beneter joined #git
20:16 beneter_ joined #git
20:16 timvisher joined #git
20:16 beneter joined #git
20:17 ramsub07 joined #git
20:18 aidalgol joined #git
20:19 Es0teric joined #git
20:19 tirej joined #git
20:20 _xor joined #git
20:20 Davey joined #git
20:20 Davey joined #git
20:21 jason237 joined #git
20:23 matsaman joined #git
20:24 sq-one joined #git
20:25 aard_ joined #git
20:25 Discovery joined #git
20:26 noecc left #git
20:26 Balliad joined #git
20:26 jason237 joined #git
20:27 m0viefreak joined #git
20:31 synthroid joined #git
20:32 spacelord_ joined #git
20:32 cagedwisdom joined #git
20:39 timvisher joined #git
20:39 bannakaffalatta joined #git
20:41 Snugglebash joined #git
20:41 Electrometro joined #git
20:42 AndChat69225 joined #git
20:46 hasc joined #git
20:46 krillr joined #git
20:51 ssitani joined #git
20:53 Es0teric joined #git
20:56 DownTekno joined #git
20:56 Ijw joined #git
20:59 saml joined #git
20:59 tirej joined #git
20:59 saml is it easy to find base of fast forwarded merge?
21:00 ojacobson it's exactly one of the two commits involved in the merge
21:00 stick-shift joined #git
21:00 Epx998_ joined #git
21:00 Tobbi joined #git
21:00 ojacobson once that information is lost (once the merge has been performed), it's impossible to recover without outside context (reflog entries, for example), but before you perform the merge, it's trivial
21:00 Vampire0 In the reflog
21:01 saml reflog is local, right?
21:01 ojacobson yes
21:01 robattila256 joined #git
21:01 saml hrm maybe i should always do no-ff
21:05 timvisher joined #git
21:05 leprechau joined #git
21:05 donught joined #git
21:06 metachr0n joined #git
21:07 peeps[lappy] joined #git
21:07 e14 joined #git
21:07 PettanShoutaKun joined #git
21:07 PettanShoutaKun I have a bunch of commits that were pushed into master.... what is the best way to uncommit them?
21:07 PettanShoutaKun I was thinking interactive rebase
21:08 PettanShoutaKun but I am not an expert
21:08 PettanShoutaKun what do you recommend
21:08 _ikke_ PettanShoutaKun: You know the consequences of rewriting history?
21:08 PettanShoutaKun mmmmmm.... no? probably not entirely or as well as I should
21:08 _ikke_ !rewrite
21:08 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
21:09 chipotle_ joined #git
21:09 _ikke_ It depends if anyone has fetches those commits and comitted on top of those
21:09 PettanShoutaKun oh yeah
21:09 PettanShoutaKun there's quite a few commits mixed in
21:09 PettanShoutaKun most of the commits are in a single folder
21:09 PettanShoutaKun so in theory we could just delete that whole folder and push
21:10 _ikke_ The commits would still be there
21:10 Epx998_ left #git
21:10 PettanShoutaKun I've never been a git maintainer before though, so I'm wondering what the recommended course of action would be
21:11 moritz what happened? why do you want to undo it?
21:11 PettanShoutaKun It's breaking the build currently
21:11 PettanShoutaKun and the code that has been commited is not finished
21:11 borkr joined #git
21:11 donught joined #git
21:11 PettanShoutaKun they have it in a separate branch but they merged it in prematurely
21:11 ceaa joined #git
21:11 PettanShoutaKun however commits have been made on top of master with these code changes in place
21:12 PettanShoutaKun BUT! most if not all of the changes that have been made should be independent of what these commits changed
21:12 moritz ok, then a rebase that throws out the merge might be feasible
21:12 moritz followed by a git push --force, with the usual caveats
21:12 PettanShoutaKun so we want to pull it out for now to get our build passing, and then recommit it later
21:12 _ikke_ always --force-with-lease
21:13 moritz or maybe you could teach your build system to ignore those faulty files
21:13 PettanShoutaKun so I should interactive rebase and delete the commits and then git push origin/master --force-with-lease ?
21:13 moritz then you could avoid the force-push
21:13 PettanShoutaKun I'd rather get them out of the repo on principal honestly
21:13 cdown joined #git
21:14 PettanShoutaKun most people know to pull in the latest origin/master and merge it into their branch before pushing it back up to master
21:16 donught joined #git
21:19 nettoweb joined #git
21:19 gugah joined #git
21:21 nickabbey joined #git
21:21 cdg joined #git
21:23 rgrinberg joined #git
21:24 ikelso joined #git
21:24 rgrinberg Is there a way to search for commits where a particular file was modified/added/deleted?
21:24 bannakaffalatta joined #git
21:26 donught joined #git
21:27 ikelso_ joined #git
21:27 WaReZ joined #git
21:27 timvisher joined #git
21:28 PettanShoutaKun okay this rebase has a ton of commits on it...
21:28 PettanShoutaKun I have a better idea/question
21:28 PettanShoutaKun so master hase 4 commits on it... the 4th one being the commit that merged all these other commits into it
21:28 Dougie187 joined #git
21:29 PettanShoutaKun what would be the easiest way to go back 4 commits, and then just pull in the top commits on the branch
21:30 PettanShoutaKun or is that not possible/how a merge works
21:30 PettanShoutaKun do I interactively have to drop these out
21:30 troyt joined #git
21:31 PettanShoutaKun I mean in reality one commit basically messed this all up
21:32 PettanShoutaKun is there anyway to undo that one commit or does that commit carry the baggage of every other commit with it?
21:32 minikN joined #git
21:33 tfrangio joined #git
21:34 rivarun joined #git
21:35 dsdeiz joined #git
21:35 dsdeiz joined #git
21:36 donught joined #git
21:38 jason237 joined #git
21:39 donught joined #git
21:39 tirej joined #git
21:39 PettanShoutaKun http://pastebin.com/mG4SGEtf
21:39 PettanShoutaKun ffffffffffffudge
21:39 PettanShoutaKun what do I do?
21:39 rominronin joined #git
21:40 GodGinrai joined #git
21:40 notebox joined #git
21:40 PettanShoutaKun :V
21:41 LionsMane joined #git
21:41 Dougie187 PettanShoutaKun: resolve the conflicts?
21:41 Dougie187 PettanShoutaKun: look at the output of `git status`
21:43 Sasazuka_ joined #git
21:46 druonysus left #git
21:47 druonysus joined #git
21:47 Darren_ joined #git
21:48 timvisher joined #git
21:49 mp3 joined #git
21:49 Balliad joined #git
21:51 PettanShoutaKun what if I missed a commit to drop when I rebase x.x
21:51 PettanShoutaKun do I finish this rebase and rebase again?
21:52 PettanShoutaKun or if I run "git rebase --skip" will that ignore this commit
21:52 PettanShoutaKun it's currently stuck on conflicts for this commit anyway
21:52 sea-gull_ joined #git
21:53 Dougie187 PettanShoutaKun: Then you can `git rebase --abort` and start over.
21:53 Electrometro joined #git
21:53 Dougie187 `git rebase --skip` will ignore the commit you're on
21:54 PettanShoutaKun SWAG
21:54 PettanShoutaKun yeah I just missed one and I don't need this commit so I went for the skip option
21:57 _ng joined #git
21:57 thiago PettanShoutaKun: if you've got conflicts on the commit you were goign to drop anyway, git rbase --skip
21:57 thiago you don't need to abort and restart
21:58 thiago if you're stuck in a commit prior to it, I'd still solve and continue. If it doesn't get dropped, you can drop it later.
21:58 thiago if you're stuck in a commit after it and you're stuck because it wasn't dropped, then abort.
21:58 austin987 joined #git
21:58 gajus joined #git
21:58 hasc joined #git
22:00 _ng joined #git
22:00 msonntag joined #git
22:01 hasc joined #git
22:01 mehola joined #git
22:02 brent__ joined #git
22:03 chardan joined #git
22:05 canuk joined #git
22:05 marchelly joined #git
22:06 blackwind_123 joined #git
22:07 marchelly hi, while running git push post-receive hook if client disconnects by timeout or any other reason what happens? It continues or stopped after last "online" action?
22:07 timvisher joined #git
22:07 marchelly how do this works?
22:08 tinanoo joined #git
22:08 jeffreylevesque joined #git
22:09 ianmethyst joined #git
22:11 PettanShoutaKun so I realize we have branches made AFTER the commits... that will probably end up merged back into master
22:11 PettanShoutaKun and those will probably pull in all the commits I just rebased out
22:11 PettanShoutaKun fffffff
22:13 PettanShoutaKun does this mean everyone on a branch has to do a git rebase master
22:13 PettanShoutaKun and then push?
22:15 askb joined #git
22:17 ianmethyst joined #git
22:18 jamick joined #git
22:19 tirej joined #git
22:23 PettanShoutaKun what do I do? x.x
22:23 jonmorehouse joined #git
22:25 Gsham joined #git
22:29 kyan joined #git
22:29 BSAlb joined #git
22:31 busta joined #git
22:31 duderonomy joined #git
22:33 timvisher joined #git
22:34 jonmorehouse joined #git
22:34 Snugglebash joined #git
22:35 testchat joined #git
22:37 fahadash joined #git
22:37 ianmethyst joined #git
22:41 mirge joined #git
22:41 ResidentBiscuit joined #git
22:42 multi_io joined #git
22:42 mirge howdy - anyone here familiar enough with git merge-base to help me figure out if a behavior i'm seeing is a bug in git or my expectations?
22:43 notebox joined #git
22:43 ToBeCloud joined #git
22:43 Noodlewitt joined #git
22:43 WayToDoor joined #git
22:43 multi_io how can I check out another branch without modifying the working directory?
22:44 grawity that's kinda what checking out does, though
22:44 grawity anyway, `git worktree add ../myproject_dev branchname`
22:45 ResidentBiscuit joined #git
22:45 ioffa joined #git
22:47 multi_io grawity: I essentially just want to move HEAD to some other branch
22:47 mirge you're looking to reset then
22:48 multi_io I guess echo 'ref: refs/heads/branch' >.git/HEAD
22:48 mirge git reset --soft branch
22:48 multi_io mirge: nah, that moves the branch HEAD points at along with it...
22:48 grawity git symbolic-ref HEAD refs/heads/branch && probably git read-tree HEAD
22:48 mirge git reset --soft branch
22:48 grawity mirge: repeating it won't make it any more correct...
22:49 mirge ha - if i'm repeating it's my client not me
22:49 mirge (sorry)
22:49 jimi_ joined #git
22:50 m4sk1n joined #git
22:51 multi_io thanks
22:52 multi_io is "git read-tree HEAD" the same as "git reset HEAD", then?
22:52 cworth joined #git
22:56 gugah joined #git
22:56 cagedwisdom joined #git
22:57 mirge why does `git merge-base --forkpoint <branch>` fail to report a merge base if the forkpoint isn't in the reflog for <branch>?
22:58 ianmethyst joined #git
22:59 tirej joined #git
23:01 fmeerkoetter joined #git
23:02 mostlybadfly joined #git
23:02 jstimm joined #git
23:03 humoroustwo joined #git
23:04 timvisher joined #git
23:04 ResidentBiscuit joined #git
23:06 toydestroyer joined #git
23:07 multi_io grawity: never heard of "git worktree" btw, wtf
23:10 cr34ton joined #git
23:11 rominronin joined #git
23:15 miczac_ joined #git
23:15 sargas joined #git
23:17 Dougie187 left #git
23:18 msonntag joined #git
23:22 phtes joined #git
23:23 Kaisyu joined #git
23:23 phtes Hi having an odd git issue. Had branch-a merged into master then reverted, then branch-a was merged int o branch-b and branch-b was merged into master, then reverted
23:23 phtes now on branch-a if i try to rebase it "git pull --rebase master" its removing files from branch-a
23:24 tobiasvl phtes: that sounds normal, on master there is a commit that reverts the changes in branch-a
23:24 phtes do i need to revert the revert?
23:24 tobiasvl yes
23:24 tobiasvl !faulty_merge
23:24 gitinfo 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/Docu​mentation/howto/revert-a-faulty-merge.txt
23:24 phtes bleck ok
23:24 dsdeiz joined #git
23:24 dsdeiz joined #git
23:24 phtes wasnt us who did it. sigh
23:27 thiago joined #git
23:27 liamonade joined #git
23:29 nscott76 joined #git
23:30 Sammitch joined #git
23:30 nowhereman joined #git
23:32 malide okay, weird question, but what exactly does git hash? It's obviously not something as simple as the file name concatenated with the contents or else it would be easy to make a conflict.
23:32 ojacobson for very large values of "easy"
23:32 malide "easy"?
23:32 ojacobson the hash of a given git object is the hash of a header, a null byte, and the disk representation of the object
23:33 ojacobson the header contains an ascii string naming the type ("blob", "commit", etc), a space (0x20), and an ascii-encoded number indicating the length of the object, in bytes.
23:33 kadoban malide: The hash used for this it actually wouldn't be easy, though it's getting more possible.
23:34 malide ah, that makes sense. Thanks ojacobson and kadoban
23:37 cdown joined #git
23:37 timvisher joined #git
23:37 Sammitch anyone happen to know how to run `git gc` on a bare repo?
23:39 tirej joined #git
23:39 ojacobson `git gc`
23:40 Sammitch yeah... no.
23:40 Sammitch on a bare repo
23:40 GodGinrai joined #git
23:41 pity joined #git
23:41 Sammitch git gc works fine in my working tree, but I can't figure out how to gc the bare repo upstream
23:42 Sammitch everything I've dug up says "don't worry about it, it happens automatically" except that it clearly does not
23:44 Sammitch nvm, `cd /opt/git/repo.git && git --bare gc`
23:45 ojacobson (BARE: master) $ git gc
23:45 ojacobson Nothing new to pack.
23:46 nowhereman joined #git
23:46 Sammitch left #git
23:47 PettanShoutaKun do we have to rebase all the branches that had the same commits we pulled out of master
23:47 PettanShoutaKun or am I over thinking that?
23:50 fatalhalt joined #git
23:55 Anja joined #git

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