Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-09-08

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

All times shown according to UTC.

Time Nick Message
00:00 quetz joined #git
00:00 JanC_ joined #git
00:01 UserOO7 joined #git
00:01 copumpkin joined #git
00:06 stonercoder joined #git
00:06 deadprez joined #git
00:07 JamesBaxter joined #git
00:09 rishi` joined #git
00:11 rominronin joined #git
00:12 joki joined #git
00:13 SwiftMatt joined #git
00:14 deadprez joined #git
00:15 romerocesar joined #git
00:15 red82 joined #git
00:15 deadprez joined #git
00:15 _August_ joined #git
00:15 deadprez_ joined #git
00:16 _August_ joined #git
00:16 _August_ how can i pull from a branch and ignore local changes?
00:16 _August_ do i just reset?
00:17 kadoban _August_: Sounds like you possibly want a reset. Do you just want your local branch to exactly equal the remote one, and dump any commits or changes you may have made?
00:18 _August_ yeah i want to dump all the commits, somehow they got stuck on the remote server
00:19 mehola joined #git
00:19 _August_ i was thinking something like git reset --hard and then pull again
00:20 kadoban _August_: If you want the local branch blah to be exactly like the remote branch origin/blah, you can just do 'git checkout blah && git reset --hard origin/blah'  be careful that that's actually what you want though, any stuff on the local blah will be gone.
00:21 kadoban You maybe should fetch first, though it probably doesn't matter. 'git fetch'
00:22 ksefchik joined #git
00:23 UserOO7 joined #git
00:24 red82 joined #git
00:33 MattMaker joined #git
00:33 anuxivm joined #git
00:39 mingrammer joined #git
00:40 zacts joined #git
00:44 diogenese joined #git
00:44 a_thakur joined #git
00:45 star_prone joined #git
00:48 JamesBaxter joined #git
00:52 shmuli joined #git
00:56 _August_ i keep getting  The following untracked working tree files would be overwritten by merge: when i try to pull
00:57 _August_ any suggestions?
00:58 oxfouzer joined #git
00:59 thiago _August_: the untracked file would be overwritten by the merge
00:59 thiago either don't merge or remove the offending file
00:59 _August_ and when i do git fetch nothing comes back
00:59 thiago because git fetch does not modify the working tree
00:59 _August_ how do i remove them?
00:59 thiago rm
00:59 _August_ you actually delete the files?
00:59 thiago yes
01:00 sleepee joined #git
01:00 m1m3-50 joined #git
01:01 _August_ sigh...
01:06 GavinMagnus left #git
01:07 pouet joined #git
01:08 cyan__ joined #git
01:09 cqi joined #git
01:09 actinide joined #git
01:10 relipse joined #git
01:10 actinide joined #git
01:12 rominronin joined #git
01:12 MattMaker joined #git
01:12 oxfouzar joined #git
01:15 mingrammer joined #git
01:15 BlaXpirit joined #git
01:17 zacts joined #git
01:18 KnownSyntax joined #git
01:18 pabs3 joined #git
01:18 pabs3 in a script, how can I tell if the master branch is in sync with its upstream?
01:18 durham joined #git
01:19 dtrott joined #git
01:20 dodobrain joined #git
01:22 ochorocho__ joined #git
01:22 ruby_ joined #git
01:22 sweatsuit_ joined #git
01:25 pouet joined #git
01:25 ruby__ joined #git
01:27 pks joined #git
01:29 duderonomy joined #git
01:31 zacts joined #git
01:31 mizu_no_oto joined #git
01:36 Sleepy63 joined #git
01:36 blackwind_123 joined #git
01:36 citizen-stig joined #git
01:37 Dispatch joined #git
01:40 hahuang61 joined #git
01:41 dreiss joined #git
01:46 MattMaker joined #git
01:46 PioneerAxon joined #git
01:46 Vampire0_ joined #git
01:46 d^sh joined #git
01:49 ilbot3 joined #git
01:49 Topic for #git is now Welcome to #git, the place for git help and 90s cartoons | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.10.0 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | It's big, it's heavy, it's wood! git-log, from BLAMMO!
01:51 romerocesar joined #git
01:51 zacts joined #git
01:52 phoenixz joined #git
01:56 cebor joined #git
01:56 shmuli joined #git
02:03 actinide joined #git
02:04 shmuli joined #git
02:06 lachm joined #git
02:07 justanotheruser joined #git
02:08 lachm joined #git
02:09 Gianormagantrous I thought there were a way to commit with the last commit message as the message. I cannot find docs on it.
02:10 kadoban You want to just repeat the last commit message?
02:10 Gianormagantrous git commit -m"Fixes 123" && cd ../otherProjet && git commit -sameMessage
02:10 Gianormagantrous I'm sure I saw that somewhere
02:10 norotartagen joined #git
02:11 norotartagen thanks, git is working great
02:11 Gianormagantrous sort of a global
02:11 norotartagen just merged some stuff
02:11 kadoban I doubt there's a flag for that. git repos aren't related in general
02:11 norotartagen no problem
02:11 norotartagen left #git
02:12 kadoban Gianormagantrous: If it's not in your global git config, I no of no place it exists, and commit messages are definitely not recorded there.
02:12 Gianormagantrous I know they are not "related" but I have this remembrance of a sort of ~/.git_last_commit_message
02:13 annoymouse4210 joined #git
02:13 kadoban I'd be shocked if that existed, or anything like it.
02:13 rominronin joined #git
02:13 Gianormagantrous Oh well, I must've been dreaming...
02:14 kadoban It'd be much more beliveable inside of one repo, but then I'm not sure why you'd want to repeat the same commit message twice in a row.
02:14 polyzen joined #git
02:14 Gianormagantrous It indeed does not make sense in a single repo
02:15 dreiss joined #git
02:15 A5101 joined #git
02:16 hexagoxel joined #git
02:17 actinide joined #git
02:19 mehola joined #git
02:19 d0nn1e joined #git
02:20 fstd_ joined #git
02:20 snowkidind joined #git
02:21 _August_ joined #git
02:22 Vampire0 Gianormagantrous,    git commit -C HEAD
02:23 mycelium joined #git
02:23 Vampire0 Gianormagantrous, not exactly the last commit message, but the commit message of the currently checked out commit and thus maybe the last commit message
02:23 Vampire0 Gianormagantrous, at least if you didn't checkout anything else since the last commit
02:24 ruby_ joined #git
02:24 fakenerd joined #git
02:25 actinide joined #git
02:25 Vampire0 Gianormagantrous, you can use a commit-msg hook though to write the commit message to a file on commit and a prepare-commit-msg hook to read that file and insert it into the commit message template on the fly, either as-is or as comment
02:25 Vampire0 Gianormagantrous, wouldn't be too hard
02:26 Gianormagantrous Thanks Vampire0, but I thought there were a file accessible globally (in my vision a file named something like .git_last_commit_message") and kept thinking about what could it be useful for. And now, working with multiple repos and submodule all inter-related, I think I would find that useful
02:26 Gianormagantrous Yes, good idea Vampire0
02:28 Gianormagantrous Again, I must have seen something totally unrelated…
02:33 snowkidind joined #git
02:34 pabs3 left #git
02:34 Goplat joined #git
02:36 lagothrix joined #git
02:40 stonercoder joined #git
02:40 arescorpio joined #git
02:42 fakenerd_ joined #git
02:43 deadprez joined #git
02:43 jeffreylevesque joined #git
02:45 a_thakur joined #git
02:47 red82 joined #git
02:50 jceb joined #git
02:50 Bl4ckpajamas joined #git
02:54 ruby_ joined #git
02:57 chachasmooth joined #git
03:02 m1m3-50 joined #git
03:02 aidalgol joined #git
03:04 aavrug joined #git
03:04 rgrinberg joined #git
03:05 jeromelanteri hi, after modify two files, i see that i want this file to stay modified only in my local repo and not on bitbucket... so i add them on .gitignore. then for this should be view from bitbucket repo and then updated on VPS host, i add .gitignore and commit and push. OK. but actually, these 2 files i changed stay in red color in the list of files who has been modified. How to said git to actually take care of the list of
03:05 jeromelanteri ignored file and stop track these 2 modified files ?
03:05 aavrug joined #git
03:06 stonercoder joined #git
03:06 watabou joined #git
03:07 rewt .gitignore only applies to untracked files, so you have to remove them from the repo
03:08 RiPer joined #git
03:08 pouet joined #git
03:09 cqi joined #git
03:10 deadprez joined #git
03:11 ShekharReddy joined #git
03:12 zacts joined #git
03:12 deadprez joined #git
03:14 chrisshattuck joined #git
03:14 Vortex35 joined #git
03:14 rominronin joined #git
03:17 Lunatrius joined #git
03:18 ansimation joined #git
03:18 Psychiatrist joined #git
03:21 nettoweb joined #git
03:21 stux|RC-only joined #git
03:23 ochorocho__ joined #git
03:23 cyan__ joined #git
03:23 chrisshattuck joined #git
03:24 cschneid joined #git
03:25 evanwang joined #git
03:25 citizen-stig joined #git
03:30 chartractegg joined #git
03:31 chartractegg joined #git
03:34 scottxu joined #git
03:36 justan0theruser joined #git
03:36 mycelium joined #git
03:40 deadprez joined #git
03:40 pouet joined #git
03:41 hahuang61 joined #git
03:44 deadprez joined #git
03:45 mehola joined #git
03:46 Bl4ckpajamas joined #git
03:47 mehola_ joined #git
03:50 Andrew_K joined #git
03:50 mehola joined #git
03:51 a_thakur joined #git
03:52 romerocesar joined #git
03:52 Kamilion joined #git
03:53 ruby_ joined #git
03:53 ksefchik joined #git
03:55 _sjs joined #git
03:58 harish joined #git
04:01 romerocesar joined #git
04:01 a_thakur joined #git
04:03 oxfouzar joined #git
04:03 _28_ria joined #git
04:05 jeromelanteri rewt and how to remove them from the bitbucket repo and only from the bitbucket repo (not from my own local repo) ? because the interest of non track files is not to delete them... if i want to delete them i use git rm... i just want it not looking at them anymore.
04:05 _28_ria joined #git
04:06 jeromelanteri if i do a git rm --cached Gemfile.lock (fr example), it remove it from bitbucket repo... yes, ut also from my own local repo (that i doesn't want).
04:08 _28_ria joined #git
04:09 thiago joined #git
04:09 mrpackethead joined #git
04:09 Cabanossi joined #git
04:10 zivester joined #git
04:15 roxtrongo joined #git
04:16 rominronin joined #git
04:16 BlaXpirit joined #git
04:24 SwiftMatt joined #git
04:24 rewt jeromelanteri, you could move them out of the local dir temporarily, and them move them back after they're gone from the repo and .gitignore is set up
04:26 azerus joined #git
04:27 realz joined #git
04:29 fakenerd joined #git
04:30 mangolisk joined #git
04:36 romerocesar joined #git
04:36 Bl4ckpajamas joined #git
04:38 _August_ joined #git
04:39 martinkl_ joined #git
04:41 jeromelanteri rewt, yes, i do a copy of this and make it back. but i'm really surprise git doesn't offer a easy way to do administration things like that. If you don't follow a strict and tight protocol, you are locked quickly...
04:42 jeromelanteri locked or loose time with lot of commands instead of coding your application.
04:42 _xor joined #git
04:42 jeromelanteri i dream to have something like git, but for humans.
04:44 pouet joined #git
04:45 azerus joined #git
04:49 _28_ria joined #git
04:50 ruby_ joined #git
04:51 t2mkn joined #git
04:51 pouet joined #git
04:53 _28_ria joined #git
04:54 Duder9999 joined #git
04:54 a3Dman joined #git
04:55 constant joined #git
04:56 _28_ria joined #git
04:57 govg joined #git
04:57 daey_ joined #git
04:59 sdothum joined #git
05:02 e14 joined #git
05:02 dec0n joined #git
05:04 m1m3-50 joined #git
05:05 contrapumpkin joined #git
05:06 _28_ria joined #git
05:09 _28_ria joined #git
05:09 realz__ joined #git
05:11 subhojit777 joined #git
05:12 deadprez joined #git
05:14 realz__ joined #git
05:14 _28_ria joined #git
05:14 citizen-stig joined #git
05:16 stompleee joined #git
05:16 _28_ria joined #git
05:16 romerocesar joined #git
05:17 rominronin joined #git
05:19 realz__ joined #git
05:19 _28_ria joined #git
05:20 crawler joined #git
05:20 nilg joined #git
05:20 scottxu test
05:21 deadprez joined #git
05:23 govg joined #git
05:23 deadprez_ joined #git
05:23 ochorocho__ joined #git
05:24 vrishab joined #git
05:25 m1m3-50 joined #git
05:26 nilg joined #git
05:26 red_ joined #git
05:27 hahuang61 joined #git
05:27 sanketdg joined #git
05:30 dec0n joined #git
05:33 Duder9999 joined #git
05:35 bmahe joined #git
05:36 stux|RC-only joined #git
05:36 nilg joined #git
05:37 dsantiago joined #git
05:40 mingrammer joined #git
05:41 scottxu #wrokman
05:41 ThomasLocke joined #git
05:42 zeroed joined #git
05:42 stux|RC-only joined #git
05:44 jason237 joined #git
05:46 jason237 joined #git
05:46 aeo1ack joined #git
05:48 Raging_Hog joined #git
05:49 _August_ joined #git
05:50 sQVe joined #git
05:50 qt-x joined #git
05:50 chele joined #git
05:50 bocaneri joined #git
05:50 sQVe joined #git
05:51 MattMaker joined #git
05:55 phanimahesh joined #git
05:57 cyan__ joined #git
05:57 mohabaks joined #git
05:59 pouet joined #git
06:00 _August_ joined #git
06:00 danmackay joined #git
06:00 sQVe joined #git
06:00 sQVe joined #git
06:01 satya4ever joined #git
06:02 phanimahesh joined #git
06:03 crawler joined #git
06:04 rscata joined #git
06:04 Alienpruts joined #git
06:05 sQVe joined #git
06:05 nopevms joined #git
06:06 circ-user-OBsh3 joined #git
06:07 MattMaker joined #git
06:10 e14 joined #git
06:13 sQVe joined #git
06:13 lb1c joined #git
06:13 phanimahesh joined #git
06:14 davidosomething joined #git
06:16 e14 joined #git
06:18 rominronin joined #git
06:18 roelmonnens joined #git
06:19 star_prone joined #git
06:21 JeroenT joined #git
06:21 glebihan joined #git
06:30 dsdeiz joined #git
06:30 dsdeiz joined #git
06:32 rominronin joined #git
06:33 fullstack joined #git
06:34 MattMaker joined #git
06:41 Sceorem joined #git
06:41 cbreak jeffreylevesque: git will track history without fail
06:41 cbreak sorry, Jeromelanteri
06:42 jeromelanteri cbreak, sorry for what ?
06:42 cbreak jeromelanteri: what I said was meant for you
06:42 cbreak git will track history without fail
06:42 cbreak it will sync history via fetch / push
06:43 constant joined #git
06:43 cbreak having different history on a remote repo from the local history makes no sense to git
06:43 freimatz joined #git
06:43 jeromelanteri cbreak, yes but git should be better to just do what the user want...
06:43 jeromelanteri and that's all.
06:43 cbreak if you remove a file from git, and sync that history somewhere else, it obviously remembers that the file is removed
06:43 cbreak because that's obviously what the user wante
06:43 cbreak otherwise the user would not sync history of the deletion somewhere else
06:44 jeromelanteri cbreak, no, all the users doesn't want the same thing all the time. and git or what ever shouldn't have to said what user want. you are not all the user,
06:44 cbreak see?
06:44 dreamyspell joined #git
06:45 cbreak jeromelanteri: if a user of git doesn't want to use git, then that user is obviously using the wrong tool
06:45 cbreak the whole point of git is to have consistent history
06:45 jeromelanteri the problem is that, if you doesn't follow a strick protocol thgen you will do errors, then you will be screwed by git.
06:45 gusnan joined #git
06:45 cbreak if it couldn't guarantee that, then git would be worthless
06:45 cbreak git will do its best to avoid such issues
06:45 _ng joined #git
06:45 cbreak it will NOT lose changes that were committed by the user
06:46 jeromelanteri much more... if you want for example to just update a file inside all the branch or a list of branch you choose... you can not easy, you need to go inside each branch for do edit and commit and... wtf !
06:46 cbreak jeromelanteri: there's no strict protocol required
06:46 cbreak jeromelanteri: nope
06:46 ToBeCloud joined #git
06:46 cbreak jeromelanteri: you just make the change in the merge-base, then merge that commit into all branches
06:46 jeromelanteri ok, cbreak, please, ignore me now, i don't like to loose my time with people who denie facts.
06:46 jeromelanteri bye
06:46 cbreak jeromelanteri: then stop denying facts
06:47 jeromelanteri you are ignored now. bye
06:48 jeromelanteri was kicked by cbreak: stop being both dumb and rude please
06:49 cbreak wonder why people are always arguing with facts and then get rude and refuse to accept reality :(
06:49 GT4066 joined #git
06:50 rishi joined #git
06:51 danslo joined #git
06:54 jeffreylevesque_ joined #git
06:54 dreiss joined #git
06:59 submitnine joined #git
07:00 mikecmpbll joined #git
07:00 stuivi joined #git
07:01 moritz because reality sucks, and people are emotionally driven
07:01 Darcidride joined #git
07:01 star_prone joined #git
07:03 citizen-stig joined #git
07:04 star_prone joined #git
07:04 star_prone joined #git
07:05 crawler joined #git
07:05 mrpackethead joined #git
07:06 laumann joined #git
07:08 JeroenT joined #git
07:08 lordjancso joined #git
07:10 elect joined #git
07:10 aax joined #git
07:13 DieguezZ joined #git
07:14 m1m3-50 joined #git
07:14 e14 joined #git
07:14 PCatinean joined #git
07:18 jknetl joined #git
07:18 harish joined #git
07:19 e14 joined #git
07:20 star_prone joined #git
07:22 ShapeShifter499 joined #git
07:24 apotry joined #git
07:24 ochorocho__ joined #git
07:25 quetz joined #git
07:26 _kreo joined #git
07:27 e14 joined #git
07:27 tristanp joined #git
07:28 hahuang61 joined #git
07:28 phanimahesh joined #git
07:29 davidosomething joined #git
07:30 Defcronyke joined #git
07:30 sweatsuit joined #git
07:30 citizen-stig joined #git
07:33 bauruine joined #git
07:35 JeroenT joined #git
07:37 zacts joined #git
07:37 noodle joined #git
07:39 chovy joined #git
07:40 chovy is it possible to accept the key when i do git clone?
07:40 chovy i'm trying to automate something
07:42 iveqy chovy: you mean the ssh key?
07:44 sQVe joined #git
07:44 inflames joined #git
07:44 diogenese joined #git
07:45 stuh84 joined #git
07:45 ahmedelgabri joined #git
07:46 davidosomething joined #git
07:48 Balliad joined #git
07:50 chovy yeah
07:50 chovy its asking me if i want to trust the host
07:50 chovy before it clones
07:50 chovy iveqy: ^
07:51 iveqy chovy: you know that this is a one time question only?
07:51 chovy yeah. but i'm writing a script to deploy a server. and i'd like to just force it to YES
07:51 kurkale6ka joined #git
07:52 iveqy chovy: it's certainly possible. I'm not sure how much git has support for but you can always use ssh to ignore that warning
07:52 iveqy chovy: the easy way is of course to use https instead of ssh when cloning
07:52 chovy i want to commit back though
07:52 osse export GIT_SSH_COMMAND='ssh -o "StrictHostKeyChecking no"'
07:53 iveqy osse: nice!
07:53 chovy how do i use that?
07:54 iveqy chovy: run that line in your shell just before you run your git commands
07:54 chovy oh. i get it
07:54 chovy works! thank you!
07:54 clmsy joined #git
07:54 chovy now if i can just figure out how to add a key to bitbucket from bash
07:54 iveqy chovy: I'm sure bitbucket has a nice api
07:54 GavinMagnus joined #git
07:55 ki0 joined #git
07:55 JeroenT joined #git
07:57 mikecmpbll joined #git
08:01 eduardas_m joined #git
08:02 rjack2 joined #git
08:02 ahmed_elgabri joined #git
08:03 encod3 joined #git
08:04 chovy yeah. i have to figure out how to auth though.
08:04 chovy save that one for another day
08:04 clmsy joined #git
08:04 ksefchik joined #git
08:04 davidosomething joined #git
08:08 chll_ joined #git
08:08 guardian hello, where can I learn more about why git spends lot of time packing objects on cloning a large repo?
08:09 ferr joined #git
08:10 marcogmo_ joined #git
08:14 u1106 joined #git
08:16 bongjovi joined #git
08:17 MattMaker joined #git
08:17 buzzLightBeer joined #git
08:18 phanimahesh joined #git
08:19 _main_ joined #git
08:19 mehola joined #git
08:19 iveqy guardian: if noone answers you here, use !list
08:19 gitinfo guardian: [!mailing_list] The mailing list can be reached via git@vger.kernel.org. You don't need to subscribe to the list, you will always be put in cc on reply. Read archives at http://j.mp/gitlist
08:21 shirty joined #git
08:22 ThunderChicken joined #git
08:23 encod3 joined #git
08:29 ahmed_elgabri joined #git
08:33 _main_ joined #git
08:37 Ezzy joined #git
08:37 mozzarella joined #git
08:37 decav joined #git
08:37 diogenese joined #git
08:38 mattsl1 left #git
08:41 guardian thanks
08:41 al-damiri joined #git
08:41 uranellus joined #git
08:41 uranellus joined #git
08:41 sQVe joined #git
08:43 citizen-stig joined #git
08:46 erguven joined #git
08:47 MattMaker joined #git
08:48 berndl joined #git
08:48 roelmonn_ joined #git
08:52 ahmedelgabri joined #git
08:54 sQVe joined #git
08:54 leeN joined #git
08:57 danslo joined #git
08:59 davisonio joined #git
09:00 marcogmo_ joined #git
09:00 _ng joined #git
09:01 star_prone joined #git
09:02 marcogmo_ joined #git
09:03 Sceorem joined #git
09:04 settermjd joined #git
09:05 marcogmo_ joined #git
09:07 MattMaker joined #git
09:10 ahmedelgabri joined #git
09:11 Wizek joined #git
09:13 ProbabilityMoon joined #git
09:13 harish joined #git
09:17 MattMaker joined #git
09:23 Wizek_ joined #git
09:24 sQVe joined #git
09:24 grindhold_ left #git
09:25 ochorocho__ joined #git
09:27 stonercoder joined #git
09:28 pijiu3 joined #git
09:28 hahuang61 joined #git
09:28 necku joined #git
09:30 johnmccabe joined #git
09:31 e14 joined #git
09:32 MattMaker joined #git
09:33 mohabaks joined #git
09:35 ichux_ joined #git
09:36 e14 joined #git
09:37 irqq joined #git
09:39 notebox joined #git
09:39 kugel_ joined #git
09:41 t-ask joined #git
09:42 t-ask Hi, I pulled an upstream branch some days ago, played around with the source, but I don't want to keep my changes anymore, just resyncing my repo with the upstream one. How can I do it?
09:43 e14 joined #git
09:44 _sjs joined #git
09:44 t-ask I tried pull, but this led to merge conflicts. git reset --hard HEAD didn't seem to work either, while I thought reset + pull should do it
09:45 Hink joined #git
09:45 tobiasvl t-ask: did you commit? if so, hard-resetting to HEAD obviously won't work since HEAD has moved
09:45 tobiasvl you can hard-reset to the upstream remote-tracking branch
09:45 heftig t-ask: git reset --hard @{u}
09:46 e14 joined #git
09:48 nivag joined #git
09:50 t-ask nice, I must have done something wrong. I did reset hard with upstream, it worked. Strange I thought I did this several times before
09:52 ayjay joined #git
09:52 t-ask anyways thank you
09:52 a3pq51 joined #git
09:54 sQVe joined #git
09:57 MattMaker joined #git
09:58 kugel_ joined #git
09:58 pouet joined #git
10:00 stamina joined #git
10:04 lmatteis_ left #git
10:05 roxlu_ joined #git
10:07 MattMaker joined #git
10:09 govg joined #git
10:13 m0nkmaster joined #git
10:18 DieguezZ joined #git
10:18 star_prone joined #git
10:19 mrBen2k2k2k joined #git
10:20 pouet joined #git
10:23 Pieplay joined #git
10:23 pspletinckx joined #git
10:23 Hink joined #git
10:24 ahmed_elgabri joined #git
10:25 ash_workz joined #git
10:27 rahtgaz joined #git
10:32 mingrammer joined #git
10:34 david007 joined #git
10:37 MattMaker joined #git
10:38 rafalcpp joined #git
10:38 dimi1947 joined #git
10:39 Hink joined #git
10:44 kyan joined #git
10:45 _sjs joined #git
10:46 martinkl_ joined #git
10:46 cdown joined #git
10:47 JeroenT joined #git
10:47 martinkl_ joined #git
10:48 martinkl_ joined #git
10:49 martinkl_ joined #git
10:49 govg joined #git
10:51 tlaxkit joined #git
10:51 notebox joined #git
10:52 nwkj86 joined #git
10:53 star_prone joined #git
10:55 _nwkj86 joined #git
10:56 crised joined #git
10:56 crised M       .idea/workspace.xml
10:56 crised this is driving me nuts, I already added .idea to .gitignore
10:57 crised How can I totally untrack & delete .idea directory?
10:57 canton7 git rm -r --cached .idea
10:57 canton7 that'll untrack it, but not delete it from disk
10:57 crised canton7: from master branch?
10:57 canton7 from whatever branch is currently checked out
10:58 crised canton7: rm '.idea/workspace.xml'
10:58 mingrammer joined #git
10:58 crised canton7: I'm good now?
10:59 mingrammer joined #git
10:59 osse no
10:59 osse now git will report it as missing
11:00 crised D       .idea/workspace.xml Switched to branch 'mgUpdate'
11:00 crised I still see this when creating a new branch
11:00 osse try doing what was suggested
11:00 canton7 why didn't you do what I suggested?
11:00 canton7 also, you'll need to commit the deletion, of course
11:00 crised canton7: I did exactly that
11:00 crised canton7: I wrote the output to you
11:01 canton7 right, because it looked like you ran `rm '.idea/workspace.xml'`
11:01 osse ooooh
11:01 canton7 anyway, you still need to commit the deletion, as with any other file change
11:01 crised canton7:  delete mode 100644 .idea/workspace.xml
11:01 osse maybe git rm shouldn't print something that looks like a command :P
11:02 crised canton7: now it seems to be working fine
11:03 canton7 cool!
11:03 nwkj86 joined #git
11:04 rnsanchez joined #git
11:06 tesuji joined #git
11:08 chele is there a way to detect a merge conflict before the actual merge happens?
11:08 hahuang61 joined #git
11:08 Anticom joined #git
11:09 chele say we have the following scenario: dev1 makes a pull request to repo A. dev 2 makes a pull request to repo A. manager has to merge the requests from repos of dev1 and dev2. Is there a way for the manager to resolve the conflict at that point of time?
11:12 rnsanchez_ joined #git
11:17 _ikke_ What is wrong with actually merging?
11:18 _ikke_ git has no built-in ways to detect it accept for just executing the merge
11:18 osse s/accept/except/
11:18 osse #rekt
11:18 hahuang61 joined #git
11:19 _ikke_ lol, fingers on autopilot :P
11:19 chele _ikke_ yes that's clear. maybe there some 'on top of' server which have this. say github has a ui with fork a repo for example which i think is not built-in functionality.
11:19 _ikke_ chele: correct. They have some custom code to detect it
11:20 akiTendo joined #git
11:20 _ikke_ They need to compare the diffs of both branches and check if they share lines
11:20 osse git merge-tre
11:20 osse e
11:20 akiTendo Anyone have a recommendation for a git gui?  I was using Sourcetree on Max OSX and I'm now running Debian 8.5 (jessie)
11:20 chele yep
11:21 nwkj86 joined #git
11:21 chele _ikke_ have you ever heard of servers which implement such a thing?
11:22 MattMaker joined #git
11:23 Andrew___ joined #git
11:25 Andrew___ hi, there
11:25 gitinfo Andrew___: 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.
11:26 sevenfourk joined #git
11:27 nilg joined #git
11:28 ahmed_elgabri joined #git
11:28 mikecmpbll joined #git
11:28 afuentes joined #git
11:29 MadHatter42 joined #git
11:30 greendo joined #git
11:32 nwkj86 joined #git
11:32 govg joined #git
11:34 m0viefreak joined #git
11:35 JeroenT joined #git
11:36 star_prone joined #git
11:38 a3Dman joined #git
11:38 settermjd joined #git
11:40 ahmed_elgabri joined #git
11:40 ksefchik joined #git
11:44 gniourf joined #git
11:48 star_prone joined #git
11:51 crem joined #git
11:51 HalfEatenPie joined #git
11:51 david007 joined #git
11:52 Sceorem joined #git
11:52 MattMaker joined #git
11:52 fefa2k joined #git
11:52 fefa2k hello
11:52 gitinfo fefa2k: 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.
11:53 tmg joined #git
11:53 tmg Hi, I started a bisect but now I need to do some other work; is there a way to "pause" the bisect?
11:54 monty joined #git
11:55 tmg RTFM is the answer here, I think
11:55 tmg thanks!
11:55 tmg left #git
11:58 justrohu joined #git
11:59 justrohu Hello all, I am getting detached from #Hash  in my git ... I don't understand why
11:59 bremner justrohu: probably from checking out things that are not branches
11:59 _ikke_ justrohu: means you are checking out a commit instead of a branch
12:00 justrohu _ikke_ yes I did do git checkout #commit
12:00 justrohu _ikke_, bremner What should I do
12:00 _ikke_ !detached
12:00 gitinfo A detached HEAD(aka "no branch") occurs when your HEAD does not point at a branch. New commits will NOT be added to any branch, and can easily be !lost. This can happen if you a) check out a tag, remote tracking branch, or SHA; or b) if you are in a submodule; or you are in the middle of a c) am or d) rebase that is stuck/conflicted. See !reattach
12:01 _ikke_ justrohu: what's the reason you are checking out a commit?
12:01 dimi1947 joined #git
12:01 justrohu _ikke_, I am just practicing
12:02 _ikke_ justrohu: Alright
12:02 _ikke_ It's not bad to checkout a commit if you just want to check how it looks like
12:02 _ikke_ but if you want to make commits, you should make sure you are on a branch
12:02 justrohu _ikke_, ok .. but How Do I reattach it
12:03 osse !reattach
12:03 gitinfo Letters refer to !detached. (a) and (b): 'git checkout branchname' to continue working on another branch, or 'git checkout -b branchname' to start a new one here; (c) git am --continue; (d) git rebase --continue
12:04 justrohu osse, what does git am does
12:04 quetz joined #git
12:04 justrohu sorry my questions must be pretty basic since I am noob to git
12:04 osse man git-am
12:04 gitinfo the git-am manpage is available at http://jk.gs/git-am.html
12:04 learath2 joined #git
12:04 Alienpruts joined #git
12:06 BlaXpirit joined #git
12:07 fefa2k I want to download a repository, make some changes and ask for a pull request. The initial clone, should be a fork?
12:07 osse fefa2k: make a fork on github. clone your fork. work. push.
12:08 jast fefa2k: assuming you're talking about github, yes. a fork will give you your own repo on github that you can do whatever you want in, and then send pull requestsd
12:08 elsevero joined #git
12:08 fefa2k perfect
12:08 fefa2k thanks guys
12:08 fefa2k ^^
12:09 bencevan_ joined #git
12:09 rbr joined #git
12:10 Anticom joined #git
12:10 JeroenT_ joined #git
12:11 johnmilton joined #git
12:12 johnmilton joined #git
12:14 infra-red joined #git
12:15 bruce_lee joined #git
12:15 bruce_lee joined #git
12:15 nilg joined #git
12:19 ksefchik joined #git
12:19 dermoth_ joined #git
12:19 jimi_ joined #git
12:19 m0nkmaster joined #git
12:20 drbean_ joined #git
12:21 BlaXpirit joined #git
12:22 star_prone joined #git
12:22 BlaXpirit joined #git
12:23 git_student joined #git
12:24 drbean joined #git
12:25 pouet joined #git
12:26 drbean- joined #git
12:26 BlaXpirit joined #git
12:27 GodGinrai joined #git
12:29 drbean_ joined #git
12:30 sanketdg joined #git
12:31 nikivi joined #git
12:33 _sjs joined #git
12:33 star_prone joined #git
12:34 nanga joined #git
12:34 nanga joined #git
12:35 kyan joined #git
12:35 phanimahesh joined #git
12:36 anonymuse joined #git
12:37 MattMaker joined #git
12:38 deadprez joined #git
12:38 danslo_ joined #git
12:38 dvl I am looking at hooks used on our git server in preparation to moving to a Github Enterprise appliance. Most of the post-update hooks invoke git-update-server-info.  My reading of https://git-scm.com/docs/git-update-server-info says that's only for 'dumb servers' and we don't need to do that when using GHE. Correct?
12:39 fmcgeough joined #git
12:39 star_prone joined #git
12:40 bremner !github
12:40 gitinfo 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)
12:40 bremner in case you didn't know about that channel.
12:40 deadprez joined #git
12:41 fakenerd joined #git
12:41 danslo__ joined #git
12:41 nanga joined #git
12:41 j416 dvl: figure out your needs and go from there. If you just need basic git stuff, you probably don't need the hooks -- migrate using git, and the github backend, and you will be fine
12:42 dvl bremner: Did not know.  :)
12:42 jeffreylevesque joined #git
12:43 dvl j416: cheers.  Trouble I'm facing, historical repos, with original users long gone, nobody seems to know what is required.  ;)  Hence, my investigations.  Thank you.
12:44 j416 dvl: clone them and push back to the new server
12:44 e14 joined #git
12:44 dvl j416: Done that part.
12:44 j416 then done
12:44 j416 (don't forget tags etc)
12:44 dvl woot!  beers everyone.
12:44 phanimahesh joined #git
12:44 dvl j416: tags? what about tags?
12:44 j416 dvl: as long as it works, what's the issue? :)
12:45 j416 dvl: if there are tags, you may want to push those as well
12:45 j416 and any branches that are not the default branch
12:45 j416 perhaps there is some shorthand fancy flag to push all that, not sure
12:45 j416 that guy bremner probably knoww
12:45 j416 knows*
12:46 dvl j416: I did a git clone with --bare...
12:46 mizu_no_oto joined #git
12:46 dvl Compared old and new, seemed identical.
12:46 j416 compared how?
12:46 j416 by listing all the refs?
12:47 MattMaker joined #git
12:47 dvl j416: looked the list of branches and tags..
12:47 nettoweb joined #git
12:48 j416 sounds like it'd be good enough, aye.
12:48 dvl Good start then.
12:48 j416 there's --mirror to clone and push, not sure if that's what you need
12:48 j416 haven't used it much
12:48 dsdeiz joined #git
12:49 dvl j416: Yes, I saw --mirror... the process I saw was git clone --bare, then git push --mirror
12:49 cdown joined #git
12:49 j416 sounds sane, even though I've never tried it
12:49 dvl with a git remote add in the middle.
12:49 j416 nod
12:49 j416 nod
12:50 dvl And a creation of the repo on the server manually.
12:50 j416 nod
12:52 dvl j416: Thank you for the sanity checks.
12:53 dvl Ahh, also, there is nothing special about hooks in code, no chroot magic? I ask because many of our hooks are referencing scripts I cannot locate, e.g. exec /usr/local/bin/git-notify.pl
12:54 crem left #git
12:55 j416 dvl: well, there might be something to your setup that relies on those
12:55 j416 dvl: as far as git and working with the repos is concerned though, none of that is needed
12:56 j416 then again, relying on there being server-side hooks seems like a fragile way to do it
12:57 zivester joined #git
12:57 Gurkenglas joined #git
12:57 MattMaker joined #git
12:57 kadoban joined #git
12:57 dvl j416: It looks like some of the stuff is sending to mailing lists etc.
12:58 ahmedelgabri joined #git
13:02 danslo__ left #git
13:02 danslo joined #git
13:07 MattMaker joined #git
13:07 ExoUNX joined #git
13:11 LukeLR joined #git
13:11 JeroenT joined #git
13:11 pouet joined #git
13:15 Megaf joined #git
13:16 JamesBaxter joined #git
13:16 cdg joined #git
13:17 BrianJ joined #git
13:17 _ng joined #git
13:19 ralismark^ joined #git
13:20 sdothum joined #git
13:21 hahuang61 joined #git
13:22 MattMaker joined #git
13:22 yac joined #git
13:22 yac It's not possible to make a commit with date before christ, is it?
13:22 rgrinberg joined #git
13:24 Opertus joined #git
13:26 Jellyg00se joined #git
13:26 _ADN_ joined #git
13:26 ochorocho__ joined #git
13:27 danslo_ joined #git
13:27 j416 yac: try it
13:28 yac well, I've been unlucky so far and docs state the internal format is unix timestamp
13:28 kpease joined #git
13:28 j416 why would you want this?
13:29 yac git abuse as data visualization tool
13:29 yac for a history timeline, would be cool
13:31 j416 doesn't seem like you can go before epoch
13:32 ralismark^ yac: not really - http://stackoverflow.com/a/24977895
13:32 yac yeah, looks like it kind of underflows or something
13:33 inflames joined #git
13:35 crelix joined #git
13:35 yac maybe it wouldn't be too difficult to patch around
13:36 shinnya joined #git
13:38 red82 joined #git
13:38 jaafar joined #git
13:38 mda1_ joined #git
13:40 rnsanchez_ joined #git
13:41 Grav joined #git
13:42 griffindy joined #git
13:42 Grav Hi, anyone knows if its possible to work with git on mounted disks?
13:45 mohabaks joined #git
13:45 moneylotion joined #git
13:46 texasmynsted Grav, not sure what you mean
13:46 AaronMT joined #git
13:47 Grav I have a project with git repository on mac, mounted the folder under linux. When I go check git status on linux it treats every file as modified.
13:48 govg joined #git
13:48 texasmynsted you might be able to change your mount options
13:48 texasmynsted like no-atime and that sort of thing
13:48 Kicer86 joined #git
13:49 texasmynsted That sounds like it is not git, but rather a problem with the way the disk is mounted
13:50 Grav i will look more into it then
13:50 ruby_ joined #git
13:50 Dougie187 joined #git
13:52 MattMaker joined #git
13:54 deadprez joined #git
13:55 j416 Grav: executable bit, perhaps
13:56 Grav executable bit?
13:56 j416 git diff would say something like "file mode changed"
13:58 Grav git diff  shows old mode /new mode for every file
13:59 JeroenT joined #git
14:00 ams__ joined #git
14:00 chrisshattuck joined #git
14:00 j416 Grav: then it's the executable bit
14:00 j416 Grav: your way of mounting made all files executable
14:01 JeroenT joined #git
14:01 j416 not a git issue
14:01 j416 Grav: that said, running git over a mount like that can be slow
14:01 j416 I'd recommend running git where the files actually are
14:01 j416 it'll give better performance
14:02 t4nkd joined #git
14:02 Grav i know, i just dont like working on mac and had everything setup on my linux anyway
14:03 VagabondAzulien joined #git
14:03 SourTrout joined #git
14:03 j416 move the repo then
14:03 j416 :)
14:04 Vgr joined #git
14:04 Grav i need to compile it on ios
14:04 j416 mount the other way round maybe.
14:05 Grav maybe i will try
14:05 j416 or, you know, just use the mac
14:05 j416 it's got a terminal too..
14:05 Grav the stuff outside terminal annoys me
14:05 j416 so don't use it
14:05 chrisshattuck joined #git
14:06 Grav and i got a lot of habits
14:06 the_hawk joined #git
14:06 dhk joined #git
14:06 Grav do you know how to maybe mount the disk in some other way that wont cause those git problems?
14:08 j416 I'm not familiar with mounting disks
14:08 j416 I'm a mac user, see. :)
14:08 JeroenT joined #git
14:08 POQDavid joined #git
14:08 j416 it's not a git issue, either way, so perhaps you would have more luck in a linux channel
14:08 davidosomething joined #git
14:08 Grav well i guess
14:08 Grav thanks
14:09 j416 o/
14:09 Grav left #git
14:09 POQDavid Hi guys i am on Windows 10 64bit and installed the new git 64bit and it keeps saying error: gpg failed to sign the data
14:09 POQDavid fatal: failed to write commit object
14:09 nettoweb joined #git
14:09 j416 POQDavid: how are you committing?
14:09 POQDavid yes
14:10 j416 that's not an answer to the question
14:10 POQDavid git commit -a -S -m 'Updated submodules'
14:10 j416 why are you using -S ?
14:10 POQDavid gpg
14:10 j416 apparently gpg is not working
14:10 pijiu3 joined #git
14:10 j416 you'd need to configure that first
14:10 hashpuppy joined #git
14:11 POQDavid you saying i got to downgrade?
14:11 j416 what do you mean?
14:11 DJ_Spies joined #git
14:11 j416 either don't use gpg, or set gpg up properly
14:11 j416 simple as that
14:11 pijiu3 joined #git
14:13 POQDavid j416: my gpg used to work
14:13 POQDavid before update
14:13 le_melomane joined #git
14:13 POQDavid after i installed Git-2.10.0-64-bit.exe it stoped
14:14 Wizek joined #git
14:14 Wizek joined #git
14:15 Ultra_DuckZ joined #git
14:16 Ultra_DuckZ hi, is there a way to do git submodule add without cloning the new repo immediately?
14:16 Ultra_DuckZ oh... nvm... that wouldn't make much sense, I guess
14:18 kba I have two remotes, one public (master) and one private (dev). I often make several commits/pushes to the private repo, and then I later want to take all those commits and create one new commit that I can push to the public repo. What's the best way of doing that?
14:18 Wizek_ joined #git
14:19 tabrez joined #git
14:19 POQDavid the thing is everything was fine before update and i updated because it said gpg signing was improved
14:21 phaleth joined #git
14:21 surdaft joined #git
14:21 _sjs joined #git
14:22 nwkj86 joined #git
14:22 MattMaker joined #git
14:23 anonymuse joined #git
14:23 ResidentBiscuit joined #git
14:23 e14 joined #git
14:24 surdaft hello, I am having an issue if someone would be able to help me please? I tried to rename the file 'PayPal.php' to 'Paypal.php' and no matter how I do it, whether using `git mv --force` or what it keeps saying the file is untracked and will be removed when I change branch?
14:25 DomeMasterYolo joined #git
14:26 le_melomane joined #git
14:26 rivqaa joined #git
14:26 moritz surdaft: try using a case-senstive file system
14:27 Wizek__ joined #git
14:28 e14 joined #git
14:28 surdaft Is there anything else we can try first? I currently don't have access to that
14:29 VagabondAzulien joined #git
14:29 mjacob joined #git
14:29 DomeMasterYolo joined #git
14:29 subhojit777 joined #git
14:30 kba surdaft: Maybe you can rename it to something else first
14:30 kba if you rename it PayPal.php -> Paypal1.php -> Paypal.php, it won't just be the case that changes
14:30 dtrott joined #git
14:31 mjacob is it possible to clone a public repository, but get the objects from someone being in the same local network as me?
14:32 surdaft I tried renaming it to test then comitting and renaming it back to the version I want then comit again and the error still ocurrs
14:32 MattMaker joined #git
14:33 sssilver joined #git
14:33 osse mjacob: clone --reference
14:33 shirty joined #git
14:34 mjacob osse: i still want to have an independent, full copy on my machine at the end
14:34 e14 joined #git
14:35 antranigv I still don't get all that git merge humor;
14:35 antranigv like this one: https://media.giphy.com/media/s4AxMEKhn5AfC/giphy.gif
14:35 antranigv I never had a problem while merging
14:35 antranigv when does the problem happens?
14:35 mjacob osse: ah, maybe if i do a --dissociate afterwards it does what i want
14:36 encod3 joined #git
14:37 DomeMasterYolo joined #git
14:37 r4z joined #git
14:38 liuzhen joined #git
14:40 ayjay joined #git
14:40 Gurkenglas joined #git
14:41 phoenixz joined #git
14:41 osse antranigv: someone posted a more positive way of visualizing a git merge. in one branch a cat gets glasses, in another a hat. then at the end the cat has both glasses and a hat :3
14:41 osse but now I can't find it
14:42 eivarv joined #git
14:43 cyan__ joined #git
14:43 DomeMasterYolo joined #git
14:45 cdown joined #git
14:46 pspletinckx joined #git
14:46 Pieplay joined #git
14:49 constant joined #git
14:50 e14 joined #git
14:53 jrdavid joined #git
14:53 antranigv osse: awww so cute =^_^=
14:54 dsdeiz joined #git
14:54 dsdeiz joined #git
14:54 applepi joined #git
14:55 ShekharReddy joined #git
14:55 _ng joined #git
14:56 clemf joined #git
14:56 applepi hi all..  I'm trying to pull a remote upstream into my github repo, but use their changes for EVERYTHING.  (It's a beaglebone linux branch, and my repo is 140k commits behind, and I really only need to re-apply about 4 changes once it's done.)
14:57 tyreld joined #git
14:57 selckin make patch files, fork again?
14:58 applepi I've tried a git pull -X theirs [url], while in my checkout of my github fork, but I still don't get a copy of what I get if I checkout the beaglebone repo.
14:58 applepi selckin, would it just be easier to re-fork into a new repo?
14:59 selckin you hadn't really described a problem yet, so i assumed you wanted to avoid a fetch + upload
14:59 selckin sounds like you just want to pull --rebase or similar
15:00 kadoban applepi: Just save a reference to your commits that you want to cherry-pick or whatever your plan is, then reset the branch to be equal to the remote one, and cherry-pick (or whatever) your commits on top.
15:00 LionsMane joined #git
15:01 applepi oh hm..  so a git pull --rebase [beaglebone] rather than trying to have it apply all the updates and merge would be the way to go?
15:01 applepi that makes sense
15:01 roelmonnens joined #git
15:02 zivester joined #git
15:02 thiago applepi: are you just behind or have you diverged?
15:03 applepi thiago, very very mildly diverged, I have freshly checked out their source and it's trivial to apply my changes from that
15:04 thiago applepi: yeah, so you want to rebase
15:05 applepi i'll give this a shot, thanks all.  that definitely sounds like it makes more sense than what i have been trying to do.
15:08 jceb joined #git
15:15 ahmedelgabri joined #git
15:17 rnsanchez_ joined #git
15:17 clemf joined #git
15:19 encod3 joined #git
15:19 dmto joined #git
15:19 israelzuniga joined #git
15:20 mangolisk joined #git
15:21 PCatinean joined #git
15:21 berndl` joined #git
15:22 hahuang61 joined #git
15:22 ahmed_elgabri joined #git
15:22 drbean joined #git
15:22 MattMaker joined #git
15:24 nettoweb joined #git
15:24 bencevan_ joined #git
15:25 e14 joined #git
15:25 GingerBeardGeek joined #git
15:26 drbean_ joined #git
15:27 ochorocho__ joined #git
15:27 drbean joined #git
15:28 romerocesar joined #git
15:30 PhilippN joined #git
15:30 rubie joined #git
15:30 drbean- joined #git
15:31 phaleth joined #git
15:32 t4nkd joined #git
15:34 ochorocho__ joined #git
15:35 InfoTest joined #git
15:35 zeroed joined #git
15:35 zeroed joined #git
15:35 sssilver joined #git
15:36 bauruine joined #git
15:36 drbean joined #git
15:37 MattMaker joined #git
15:37 danslo joined #git
15:37 e14 joined #git
15:38 rominronin joined #git
15:39 e14 joined #git
15:39 drbean_ joined #git
15:41 e14 joined #git
15:43 drbean joined #git
15:43 PresidentTrump joined #git
15:43 PresidentTrump hi
15:43 gitinfo PresidentTrump: 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.
15:44 rubie joined #git
15:44 PresidentTrump are there any git commit policy documents that I can adopt for my projects
15:44 ahmed_elgabri joined #git
15:44 PhilippN joined #git
15:44 spudowiar joined #git
15:44 PresidentTrump I am basically after something I can give to my developers as guidelines without having to write it myself
15:44 bremner sure, what are you paying?
15:44 ^7heo PresidentTrump: no, they are open source, they cannot be stolen.
15:45 bremner (just kidding. not such a crazy request)
15:45 QwertyDragon joined #git
15:45 govg joined #git
15:46 PresidentTrump what ^7heo ?
15:46 drbean_ joined #git
15:46 PhilippN Hi
15:46 gitinfo PhilippN: 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.
15:47 PresidentTrump I have seen many big projects have ones customized for their company with logos etc
15:47 PresidentTrump I'm after one not customized for any company
15:47 madewokherd joined #git
15:48 PhilippN I am having trouble with Git (using git bash) on Windows. I would like to use SSH connection to Gitlab. The connection seems to work, I can connect with ssh git@gitlab.com just fine, but when i try to push or pull, I get prompted to add the server's key to local cache, answer y and then it just hangs there.
15:49 ^7heo PhilippN: what are the remotes like, in your git repo?
15:49 drbean- joined #git
15:49 actinide joined #git
15:49 ichux_ joined #git
15:49 PhilippN ^7heo: Sorry, not sure what you mean by that.
15:50 m0viefreak joined #git
15:50 ^7heo PhilippN: cat .git/config
15:50 ^7heo PhilippN: what's in there for the remotes?
15:50 PhilippN ^7heo: [remote "gitlab"]
15:50 PhilippN url = git@gitlab.com:city-of-waseca/wasecatools.git
15:50 PhilippN fetch = +refs/heads/*:refs/remotes/gitlab/*
15:51 drbean joined #git
15:51 durham joined #git
15:52 ^7heo PhilippN: so you are indeed using ssh for it
15:52 e14 joined #git
15:52 ^7heo PhilippN: and you are using the same user/domain
15:52 ^7heo PhilippN: and not a different port
15:52 ^7heo also
15:53 ^7heo getting prompted to add an ssh server key to .ssh/known_hosts AFTER the initial connection is VERY fishy.
15:53 jeremyhall joined #git
15:53 ^7heo did you connect to git@gitlab.com with the option to turn server key verification to off?
15:53 actinide joined #git
15:54 PhilippN <^7heo> that's what I thought. I think the problem may be that it's not using the internal ssh to Git Bash, but rather plink from Putty? but not sure how to change that.
15:54 PhilippN when I do push or pull, that it
15:54 grawity set/change/unset $GIT_SSH
15:54 ^7heo PhilippN: well, it depends what your path looks like, I'd say
15:55 ^7heo because I'm pretty sure that the path of git bash and the path of windows are different.
15:55 andlabs joined #git
15:56 drbean_ joined #git
15:57 ichux_ joined #git
15:58 drbean joined #git
15:59 PrashantJ joined #git
15:59 PhilippN Yep, initial state is $ echo $GIT_SSH
15:59 PhilippN C:\Program Files (x86)\PuTTY\plink.exe
15:59 PhilippN When I set it to /usr/bin/ssh it works fine. Is there a reason it defaults to Putty? Any problems if I just use /usr/bin/ssh?
15:59 PrashantJ left #git
16:00 akkad joined #git
16:01 dreiss joined #git
16:01 drbean_ joined #git
16:02 durham joined #git
16:02 ^7heo PhilippN: I'd say: since I do not install putty and use git bash instead, I have no such problem. Maybe your solution is the same?
16:03 drbean- joined #git
16:03 ^7heo I mean, putty was great when we didn't have ssh compiled for windows
16:03 ^7heo (with mingw)
16:03 ^7heo but now we do.
16:03 ^7heo So...
16:04 PhilippN Good point. never thought about it that way but I think I may go that way too.
16:05 PhilippN Thanks for the help ^7heo & grawity!
16:06 ^7heo Glad I could help.
16:07 MattMaker joined #git
16:07 drbean joined #git
16:09 s34n I'm struggling to organize separate but related repositories. My conundrum is summarized here: https://gist.github.com/smcmurray/78db2c444bd05376ea6f2971c7293394
16:10 Agialsa joined #git
16:10 Agialsa left #git
16:10 s34n I don't know of any way to clone only specific paths within a repo
16:11 s34n And I don't know of a way to get multiple repos to share a directory
16:11 nitric joined #git
16:11 drbean_ joined #git
16:11 PresidentTrump left #git
16:12 ruby_ joined #git
16:12 mycelium joined #git
16:12 MattMaker joined #git
16:12 drbean- joined #git
16:14 satbyy joined #git
16:14 grawity use symlinks
16:14 Electrometro joined #git
16:14 grawity put repo b under plugins/b, then ln -s plugins/b/b2 b2
16:15 drbean joined #git
16:17 drbean_ joined #git
16:19 d0nn1e joined #git
16:19 drbean- joined #git
16:20 GavinMagnus joined #git
16:21 fakenerd joined #git
16:22 MattMaker joined #git
16:23 ruby_ joined #git
16:23 shirty joined #git
16:24 drbean joined #git
16:25 theskillwithin joined #git
16:25 drbean_ joined #git
16:26 Marco_Ramius left #git
16:27 JanC_ joined #git
16:27 s34n grawity: that may be my best option. Has any work been done to explore path-based partial clones?
16:27 Spec-Chum joined #git
16:27 MattMaker joined #git
16:27 drbean- joined #git
16:29 e14 joined #git
16:29 drbean joined #git
16:31 grawity afaik, none
16:31 sssilver_ joined #git
16:31 grawity it's a bit difficult in Git's storage model
16:31 drbean_ joined #git
16:33 drbean- joined #git
16:33 Eugene s34n - you can't fetch arbitary objects, so that wouldn't work well
16:34 Topic for #git is now Welcome to #git, the place for git help and NP-hard problems | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.10.0 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | Now with more indirect cycling giraffes!
16:34 spudowiar joined #git
16:34 yoh is there some GIT_TRACE variable/setting which would trace values passed to e.g. cat-file --batch ?
16:35 drbean joined #git
16:35 robotroll joined #git
16:36 shmuli joined #git
16:37 romerocesar joined #git
16:37 Eugene You mean besides GIT_TRACE and friends in man git? No, not really...
16:37 gitinfo the git manpage is available at http://jk.gs/git.html
16:37 drbean_ joined #git
16:37 Ringer joined #git
16:38 charlesrose joined #git
16:38 fahadash joined #git
16:38 durham joined #git
16:39 ruby_ joined #git
16:40 PaulCapestany joined #git
16:40 jdurango joined #git
16:41 mohabaks joined #git
16:41 drbean joined #git
16:42 MattMaker joined #git
16:42 drbean- joined #git
16:43 bauruine joined #git
16:44 fees joined #git
16:46 e14 joined #git
16:46 bernardio joined #git
16:47 drbean joined #git
16:47 Ringer joined #git
16:49 JCMA_mejia joined #git
16:49 m0rf3o joined #git
16:50 m0rf3o Hi, i have a proyect with git and always we merge changes the conflicts mark all file when changes are only in one line
16:50 m0rf3o Its problem of configuration?
16:50 drbean_ joined #git
16:51 kadoban m0rf3o: Most likely your line endings are messed up, there's settings for that.
16:52 Silenced joined #git
16:52 SteffanW joined #git
16:53 drbean- joined #git
16:54 m0rf3o kadoban: yeah!, it should be the problem
16:55 m0rf3o I will test and comment
16:55 drbean joined #git
16:55 mehola joined #git
16:55 durham_ joined #git
16:55 GavinMagnus left #git
16:55 POQDavid guys since the update i cant use my gpg why is that?
16:56 Silenced joined #git
16:57 shinnya joined #git
16:57 POQDavid since the update i cant see git icon and looks like the update messed up everything for my git
16:59 rgrinberg joined #git
16:59 drbean_ joined #git
17:01 antispaceman joined #git
17:01 BSaboia I have a branch that I want to merge onto master, and I know that it will have a lot of conflicts, but I want to just fast-forward my branch to it (meaning that effectively I want my branch to be the tip of master). how can i do it?
17:01 xaa joined #git
17:01 jungsubk joined #git
17:02 kadoban BSaboia: Quite a bit of that sounded contradictory.
17:02 drbean joined #git
17:02 BSaboia kadoban, well, simply put, all the merge conflicts would take my branch version
17:03 BSaboia actually I don't even want to merge, I want my branch to be the tip of master
17:03 kadoban BSaboia: Sounds like you want reset --hard, most likely. Be careful you understand what that'll do though.
17:04 a_thakur joined #git
17:04 BSaboia kadoban, I do, but it's not what I want
17:04 BSaboia kadoban, the full history is the following: what I have right now on master will be moved to an experimental branch. I have an older, more stable branch which should actually be the master right now
17:04 lss8 joined #git
17:05 kadoban Sounds even more like reset --hard
17:05 durham joined #git
17:06 drbean_ joined #git
17:06 thm joined #git
17:06 aax joined #git
17:07 thm hi. when I have two different changes in two adjacent lines, and I want to add only one of them using git add -p, I have to manually edit the chunk and reorder the lines. is there a simpler way?
17:08 kadoban thm: Not really, think that's the only way.
17:09 thm with magit (nice emacs package) I can select single lines, still that yields the wrong order...
17:09 thiago joined #git
17:09 kadoban The wrong order?
17:10 tabrez joined #git
17:10 ShekharReddy joined #git
17:11 thm it's a bit hard to explain ... consider this chunk: https://paste.fedoraproject.org/424014/33546671/
17:11 thm now, when I select in magit the lines 6 and 8, the result is wrong.
17:12 POQDavid guys what ever was the problem i had to downgrade to Git-2.6.2
17:12 EvilPenguin joined #git
17:13 kadoban thm: Sounds weird
17:16 thm no, if I select 6 and 8, that means that 7 stays (remove the '-') and line 9 is discarded. as a result, the two lines get swapped.
17:16 thm this is perfectly reasonable, but not what I want ;)
17:17 thm same goes for git add -p, without manually reordering.
17:18 dreiss joined #git
17:20 freimatz joined #git
17:21 kus joined #git
17:22 hahuang61 joined #git
17:23 spudowiar I rewrote the history in a branch, how can I rebase another branch (which includes the old history) on this new branch?
17:23 spudowiar Well, what's the cleanest way, I mean
17:24 le_melomane joined #git
17:24 jdurango spudowiar: you can create the new branch from the actual brach and this new branch would contain your changes
17:24 spudowiar jdurango: but I already have another branch?
17:24 spudowiar jdurango: wait, let me show you what I mean
17:25 matsaman joined #git
17:25 jdurango so you need merge that changes
17:25 berndl` joined #git
17:25 spudowiar jdurango: I have a branch called "base" A -> B -> C, and a branch called "topic" A -> B -> C -> X, I rewrote "base" to be A -> D -> E, and I want "topic" to be A -> D -> E -> Z
17:26 spudowiar Ah, wait, I think I got it
17:26 spudowiar I'm not sure
17:26 spudowiar No harm in trying I guess! :)
17:27 ajpiano joined #git
17:28 jdurango spudowiar: try in your new branch: git merge old_branch
17:28 spudowiar jdurango: I don't want a merge
17:28 spudowiar jdurango: I want a rebase
17:30 thm kadoban: maybe what I really want, is to see this diff instead: https://paste.fedoraproject.org/424029/47335582/
17:31 mikecmpbll joined #git
17:31 kadoban thm: Maybe. I'm not sure that's possible though. Maybe there's some place to influence the diffing algo?
17:31 kadoban spudowiar: Just do it? If you have to, you can specify the "cutting point" or whatever it's called.
17:32 spudowiar kadoban: do what?
17:32 kadoban spudowiar: rebase
17:32 spudowiar kadoban: I was trying to work out how! :)
17:32 spudowiar I think I've got the command right
17:32 spudowiar Let me try
17:32 MattMaker joined #git
17:33 jdurango spudowiar: and what happened when you did git rebase branch
17:33 BSaboia kadoban, reset --hard passing the branch as an argument?
17:33 spudowiar jdurango: that won't work
17:33 spudowiar I don't think so, anyway
17:34 spudowiar As I suspected, it just doubled the commits
17:34 kadoban BSaboia: "git reset --hard something" says "change the branch I'm on to point exactly to <something>". So whatever you want it to point to, in your case yeah the other branch sounds likely.
17:35 BSaboia kadoban, but the other branch was never merged
17:35 BSaboia how can I reset master to it? I think I am missing something here
17:35 BSaboia I know what git reset --hard do, but I am not seeing how could I use it on my case
17:36 kadoban BSaboia: I think despite that assurance, you're missing what 'git reset --hard' does ...
17:36 BSaboia kadoban, ok, so I should reset to the other branch?
17:36 BSaboia is that what you're saying?
17:37 kadoban If you want your local master to be exactly what the other branch currently is, yes.
17:37 kadoban I'm not sure if that's actually what you want, but it's the most consistent thing I can come up with out of your desires specified earlier.
17:37 Agialsa joined #git
17:37 Agialsa left #git
17:37 nwkj86 joined #git
17:38 rominronin joined #git
17:39 BSaboia kadoban, well, here's my more detailed scenario: some changes were done to master that now proved to be too much unstable to go live. so, I created a feature branch from an older, more stable branch (let's call it B). now that I am finished, I want to have master tip to be B (the state, I don't care about having the commits)
17:40 kadoban BSaboia: Is anyone sharing the same master branch? Have you pushed it elsewhere perhaps and would someone be basing work on it, or using it for something?
17:40 Sasazuka joined #git
17:40 mingrammer joined #git
17:41 spudowiar kadoban: the command I need was `git rebase --onto base old-base topic` :)
17:41 BSaboia kadoban, this is a "me and a teammate" project, so he's using it but he's aware that some drastic changes may be needed
17:41 spudowiar *needed
17:41 BSaboia I don't care about losing refs or things like that. for instance, I am considering changing master to this branch
17:41 BSaboia B
17:42 kadoban BSaboia: Then probably yeah, do the reset --hard, push it and your teammate will also have to reset --hard to the same thing.
17:42 kadoban spudowiar: Yep, sounds right.
17:42 danslo_ joined #git
17:42 spudowiar kadoban: I always mess up the second argument, that's where my problems came from
17:42 BSaboia kadoban, so I can actually reset to a non-pushed branch
17:42 spudowiar kadoban: rather than old-base, I do the commit after that on the topic branch ("the first commit")
17:43 kadoban I don't understand the question. You can reset --hard to ... anything you want. It's a very powerful and dangerous command.
17:43 spudowiar kadoban: and obviously get a ton of merge conflicts :)
17:43 spudowiar But I'm learning
17:43 spudowiar BSaboia: you can reset --hard to any tree
17:43 spudowiar BSaboia: it just points the branch elsewhere
17:43 BSaboia spudowiar, ok
17:43 BSaboia thanks
17:43 BSaboia so kadoban was right, and I had a wrong impression of what the command did heh
17:43 blackjid joined #git
17:43 spudowiar Yes
17:43 spudowiar :)
17:44 r4z joined #git
17:44 devhoag joined #git
17:44 e14 joined #git
17:44 boltR_ left #git
17:45 skarn_ joined #git
17:45 AimHere joined #git
17:45 AimHere left #git
17:47 le_melomane joined #git
17:48 AaronMT joined #git
17:49 SpeakerToMeat joined #git
17:52 le_melomane joined #git
17:53 AaronMT joined #git
17:53 SwiftMatt joined #git
17:54 irqq joined #git
17:56 multi_io joined #git
17:56 DolphinDream joined #git
17:56 dtrott left #git
17:57 MattMaker joined #git
17:59 sin3t joined #git
18:00 saedelaere joined #git
18:00 Xizor joined #git
18:01 actinide joined #git
18:01 zeuuo joined #git
18:02 e14 joined #git
18:06 Mannn joined #git
18:06 jeffreylevesque joined #git
18:07 MattMaker joined #git
18:07 govg joined #git
18:08 LionsMane joined #git
18:10 jimi_ joined #git
18:10 cdown joined #git
18:10 devhoag joined #git
18:11 t2mkn joined #git
18:15 graft joined #git
18:15 graft hi all, i accidentally did git checkout -b new-branch from another branch instead of master
18:15 le_melomane joined #git
18:16 KevinMGranger Oh no! Just kidding, that's easy to fix
18:16 KevinMGranger Have you made any commits yet?
18:16 jason237 joined #git
18:18 durham joined #git
18:18 durham joined #git
18:21 dhollinger joined #git
18:21 MattMaker joined #git
18:23 graft joined #git
18:23 graft joined #git
18:23 stamina joined #git
18:24 graft blargh, cosmic ray fried my computer. as i was saying, i accidentally made a new branch on top of an existing branch
18:24 graft how do i move the last commit from my current branch to a new branch (off master)
18:24 Ekh joined #git
18:24 j416 graft: git rebase --onto newbranch master
18:25 j416 where newbranch is a new branch that exists
18:25 j416 or just any old commit
18:26 e14 joined #git
18:26 HarryHallman joined #git
18:27 graft j416: i'm afraid
18:27 j416 graft: !backup
18:27 gitinfo graft: Worried about your data while trying stuff out in your repo? The repository in its entirety lives inside the .git directory in the root of your work tree so to backup everything `cp -a path/to/workdir path/to/backup` or equivalent will suffice as long as the repo is not modified during backup. See also http://sethrobertson.github.com/GitBestPractices/#backups
18:29 berndl` joined #git
18:29 meejah joined #git
18:30 raz joined #git
18:30 graft j416: what i'm worried about is this:
18:30 kfunk joined #git
18:30 graft 51ed841 - working version (22 hours ago)
18:30 graft and 490982f - working version (64 minutes ago)
18:30 mizu_no_oto joined #git
18:30 graft these are the same commit that are now in my two branches with different commit hashes
18:31 j416 then they are not the same commit
18:31 j416 what is your question?
18:31 graft the first one was created when i did git branch -b
18:31 raijin joined #git
18:31 graft off the branch containing the second
18:31 j416 not it wasn't. Git branch -b does not create commits.
18:31 j416 s/Git/git/
18:31 graft well where did it come from?
18:31 j416 you did "git commit" at some point and created it.
18:32 osse it was only 64 minutes ago
18:32 graft i have only typed git commit -m "working version" once if that's what you mean
18:32 j416 osse: no that's the second one
18:32 osse maybe you can look through your shell history
18:32 graft and it was 22 hours ago
18:32 meejah what's the best way to answer the question "when did branch X diverge from master"? ideally I'd like to be able to tag this somehow in my "git log"-using alias
18:32 j416 graft: once yes, creating the first commit.
18:32 j416 graft: go on
18:33 grawity meejah: git merge-base (--fork-point)
18:33 j416 meejah: git merge-base X master
18:33 meejah great, thanks :)
18:33 graft j416: ok, then i did git branch -b new-branch (about an hour ago), and now i have those two commits with different hashes
18:33 j416 grawity: didn't know about --fork-point. thanks!
18:33 j416 graft: you did something else inbetween
18:33 grawity it's new
18:33 j416 graft: such as rebase
18:34 * grawity wonders if graft's reflog would have hints
18:34 j416 graft: or after, for that matter
18:34 j416 graft: !repro
18:34 gitinfo graft: [!transcript] Please paste (using https://gist.github.com/ or https://vomitb.in/ ) a transcript ( https://gist.github.com/2415442 ) of your terminal session so we can see exactly what you see
18:34 j416 might help
18:35 mingrammer joined #git
18:35 boombatower joined #git
18:35 EvilDMP joined #git
18:37 graft hmm, i tried to rebase but then aborted
18:37 graft with git rebase --abort
18:37 oncall-pokemon joined #git
18:38 osse !lol might give some clues
18:38 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
18:38 EvilDMP I have a project on a server, with an updated version of the same (a few changed files) locally. Stupidly, I forgot to turn the server version into a Git repository before I started making the changes, so although subsequently I made each version into a branch, I can't actually do pull requests or diffs with Git, because they have completely different version histories: https://github.com/evildmp/aikidocardiff/compare/master...responsive
18:38 graft i should just stash this commit and start over
18:38 EvilDMP Is there some way to get round this?
18:38 e14 joined #git
18:39 osse EvilDMP: make patches with git format-patch and apply them on the target repo with git am
18:39 EvilDMP osse: can you expand on what you mean by "make patches"?
18:39 EvilDMP or will this be obvious from the git format-patch docs?
18:40 osse I did. "make patches with git format-patch" :D
18:40 osse it creates a file with a diff and the commit message etc.
18:40 rominronin joined #git
18:40 EvilDMP (to be honest, I rarely find anythng obvious from the docs...)
18:40 EvilDMP OK, I will play, thanks osse
18:40 osse that file can recreate an equivalent commit in another repo with git ma
18:40 osse am
18:41 mjuszczak joined #git
18:41 osse EvilDMP: it produces files looking roughly similar to what git show would show
18:41 osse try it and see
18:42 berndl` joined #git
18:44 dviola joined #git
18:44 le_melomane joined #git
18:45 aard_ joined #git
18:46 anonymuse joined #git
18:47 thm what is the reason for submodules to have .git a 'symbolic link' to ../.git/modules/foo ?
18:47 thm and, can that be switched off?
18:47 Sasazuka joined #git
18:47 PioneerAxon joined #git
18:47 star_prone joined #git
18:48 JamesBaxter joined #git
18:48 sbeller thm: to be able to deinit a submodule, but keeping its git_dir
18:48 osse thm: just guesses here:  To have all the data in one place.  And, I suppose you can do rm .git; mv ../.git/modules/foo .git; and git itself wouldn't care
18:48 sbeller the old way to put the git dir directly into the submodule is still supported
18:49 thm sbeller: is there a cmdline option or similar (besides moving stuff manually, as osse suggested)?
18:49 raspado joined #git
18:49 sbeller osse: additionally you need to unset the workingdir config value in .git/config
18:49 pouet joined #git
18:49 sbeller thm: no there is not
18:49 raspado I created a branch, how can I determine what branch I created the branch from?
18:50 applepi Hi all..  I have a fork of the beaglebone linux on github that I'm trying to rebase to get up to date (several years of updates), and it appears to rebase (in 'git log' the latest commit is the beaglebone commit), but then I start getting patches from years ago popping up for some reason that fail?
18:50 SpeakerToMeat joined #git
18:50 thm sbeller: this behavior breaks pip: https://github.com/pypa/pip/issues/3908
18:51 King_Hual joined #git
18:51 King_Hual joined #git
18:51 PioneerAxon joined #git
18:51 kyan joined #git
18:51 sbeller thm: well in that case you have to copy everything, not just the submodule, as submodules are not completely independent
18:52 sbeller when copying everything (read including the superproject) the gitlink relative link still works
18:52 duderonomy joined #git
18:52 romerocesar joined #git
18:52 applepi Is there a way to get it to JUST rebase, get my fork up to date with the upstream, and not try to reapply any patches?
18:53 satifant joined #git
18:53 thm sbeller: but should some (random) software know it is inside a subproject? it is not clear to me how a general solution would look like.
18:54 SourTrout joined #git
18:54 EvilPenguin joined #git
18:56 sbeller thm: good point; there should be a "git is-submodule" thing, maybe "git rev-parse --get-superproject-path" that prints out the superproject if any
18:57 m0viefreak joined #git
18:58 spriz joined #git
18:58 osse isn't it by design that submodules are unaware that they are submodules?
18:58 Pyrogerg joined #git
19:00 sgen joined #git
19:00 sbeller yes, and I think it is a bit of bad design as it is not completely independent. So if you want to move this repository you need to know how to adapt the git file pointer. To do that you need to either know it is a submodule or there needs to be a command to move a repository for you
19:01 sbeller but that command doesn't exist, as "mv/cp" should be fine, right?
19:01 forgot joined #git
19:01 nettoweb joined #git
19:04 berndl` joined #git
19:06 nidr0x joined #git
19:09 elastix joined #git
19:10 diogenese joined #git
19:12 osse git submodule init --do-not-meddle \o/
19:13 sbeller "/grumpy_cat_NO"
19:13 osse why? :O
19:13 SwiftMatt joined #git
19:14 b1tchcakes joined #git
19:14 govg joined #git
19:16 sbeller because I'd rather be bike shedding about the name; meddle doesn't sound git-ish enough ;)
19:16 osse git submodule init --in-place \o/
19:16 osse --git-dir-in-place
19:17 osse --stay
19:17 osse --bad-dog
19:19 cagedwisdom joined #git
19:19 israelzuniga joined #git
19:21 e14_ joined #git
19:21 nilg` joined #git
19:22 shinnya joined #git
19:23 ansimation joined #git
19:23 hahuang61 joined #git
19:25 berndl` joined #git
19:26 cdown joined #git
19:26 qws-user-1229 joined #git
19:27 johnmilton joined #git
19:27 adante joined #git
19:32 Rodya_ joined #git
19:33 sbeller heh, I don't know. If you're serious about that discussion, why not post to the mailing list?
19:33 sbeller I mean the original problem of thm may be caused by using submodules, but submodules in itself are not the only occurrence for this. see worktrees for example, they make heavy use of the git dir file pointer
19:33 groundnuty hey, I've just discovered smudge/clean options. I got really entusiastic about it but when I googled it was hard to find any nice examples/real applciations of it
19:33 AnarchyAo joined #git
19:34 groundnuty I was thinking of replacing values/templating
19:34 chrisshattuck joined #git
19:34 j416 groundnuty: what's your use case?
19:35 groundnuty j416: we have documentation using gitbook. gitbook does not allow templated in markdown ~~~ ~~~ code sections
19:35 groundnuty we replace things like version number etc by hand
19:35 groundnuty or maybe we wil use sed -i
19:36 j416 do you not have a deployment procedure?
19:36 groundnuty my first impression was that clean/smudge were perfect for it
19:36 actinide_ joined #git
19:36 j416 clean/smudge is an ok last-minute resort; I would avoid it if I could.
19:36 groundnuty j416: we do. I was thinking to keep $verson variables in lets say develop branch
19:36 berndl` joined #git
19:37 groundnuty and in a master or release/1.1 have them repalced by something
19:37 vuoto joined #git
19:37 EvilDMP osse: well, I got *somewhere* http://dpaste.com/0ZQT9ZP - it seems it doesn't like it when files exist in both branches, or something like that (am I doing this in roughly the right way?)
19:37 j416 groundnuty: why? because, it means what is in your working dir does not match what is in the repo. Causes confusion. And if say one user doesn't have the right repo configuration, well, there'll be no replacement
19:37 qws-user-1228 joined #git
19:38 j416 groundnuty: I would simply create a script around sed or the like to do the replacing upon deploy or whatnot.
19:38 mizu_no_oto joined #git
19:39 groundnuty j416: sure, just got excited about those options :)
19:39 j416 :)
19:41 groundnuty one more matter, before I got all excited and distracted form what I wanted to do in the first place.... :)
19:41 rominronin joined #git
19:42 groundnuty I was reading about gitattributes example. 'Diffing Binary Files' in git pro
19:42 groundnuty and wondring can it be nicely applied to .md diles
19:42 groundnuty files
19:42 Fabricio20 joined #git
19:42 laumann joined #git
19:43 groundnuty I wanted to wrice a piece of code that will take all the .md files in our documentation and list how long ago each of them was eddited
19:43 groundnuty how frequently, who was last, what is the last/average diff
19:44 SpeakerToMeat joined #git
19:46 shmuli joined #git
19:46 shmuli joined #git
19:47 phanimahesh joined #git
19:47 cdown joined #git
19:48 berndl` joined #git
19:49 dtrott joined #git
19:49 j416 groundnuty: for that you don't need gitattributes or a custom diff driver. Just script it.
19:50 j416 groundnuty: a custom diff driver can be used to produce custom diffs. Say you want to be able to diff two images, for example. Or if your file format is special somehow, so that you want to do special diffing.
19:52 mehola joined #git
19:52 dtrott I am doing:  git commit-tree -p $TAG …. <OTHER_ARGS>
19:52 dtrott This works fine if $TAG is a lightweight tag but fails if its an annotated tag, is there any syntax I can use with commit-tree to pass the commit hash under the annotated tag?
19:53 stamina joined #git
19:53 e14_ joined #git
19:54 jdurango joined #git
19:54 dviola joined #git
19:55 dtrott Hmm, I think its:   $TAG^{commit}     is that correct ?
19:55 mikecmpbll joined #git
19:58 j416 dtrott: yes, see man gitrevisions for more info on that
19:58 gitinfo dtrott: the gitrevisions manpage is available at http://jk.gs/gitrevisions.html
19:59 e14_ joined #git
19:59 dtrott thanks
20:00 gargola joined #git
20:00 mehola joined #git
20:00 e14_ joined #git
20:01 spudowiar joined #git
20:03 DolphinDream joined #git
20:03 mingrammer joined #git
20:03 moneylotion joined #git
20:07 cdown joined #git
20:12 osse EvilDMP: I think so
20:12 osse EvilDMP: but I think you've created a different patch than intended. try git format-patch -1 {hash here}
20:13 osse sbeller: i was just being silly. I have no strong opinions on the matter.
20:13 EvilPenguin joined #git
20:14 PrashantJ joined #git
20:14 e14_ joined #git
20:14 dodobrain joined #git
20:15 e14_ joined #git
20:16 bongjovi joined #git
20:16 cdg joined #git
20:17 hahuang61 joined #git
20:19 King_Hual joined #git
20:19 King_Hual joined #git
20:21 hahuang62 joined #git
20:21 Raed|Mobile joined #git
20:24 jdurango_ joined #git
20:24 t0by joined #git
20:26 e14_ joined #git
20:30 Dougie187 Anyone know why worktrees mess with aliases?
20:31 qws-user-1229 joined #git
20:31 skylite_ joined #git
20:32 berndl` joined #git
20:32 clemf_ joined #git
20:33 Dougie187 e.g. Make a work tree, do `git config alias.st status`, cd into a subdirectory of the worktree, and `git st`. Then you'll get some weird message about how the worktree is already set.
20:33 Dougie187 Ah, looks like it should be fixed in newer versions.
20:34 e14_ joined #git
20:34 Rodya_ joined #git
20:40 shmuli joined #git
20:42 rominronin joined #git
20:42 qws-user-1228 joined #git
20:42 berndl` joined #git
20:44 dodobrain joined #git
20:48 Groscheri joined #git
20:48 POQDavid guys keep testing and when i install 2.10 it wont sign my commits and i have to downgrade to 2.9
20:50 DieguezZ joined #git
20:53 berndl` joined #git
20:55 greendo joined #git
20:57 imacinto joined #git
20:59 theskillwithin joined #git
21:00 imacinto is there a maximum number of commits one can squash?
21:01 imacinto someone just asked me this question, and I said "probably not one you would ever encounter," imagining it to be in the tens of thousands -- if at all
21:01 osse i would be surprised if there even is a limit
21:02 osse if anything the limit would be available disk space or memory, but since it's just a repeated commit --amend I find that hard to believe too
21:04 SourTrou_ joined #git
21:04 imacinto osse: ok. so in other words, probably not
21:04 clemf joined #git
21:04 EvilPenguin joined #git
21:05 osse imacinto: you mean using 'squash' in rebase -i , right?
21:05 imacinto osse: yep
21:05 m0viefreak joined #git
21:06 osse imacinto: in that case the limit is probably the size of the COMMIT_MSG you're asked to edit :P
21:06 osse in any case, there's no set limit like 10, or 20 or 512 or whatever
21:07 imacinto hah! fair point :)
21:10 hahuang61 joined #git
21:11 xcyclist joined #git
21:12 xcyclist Is it practical to create a repository on my linux directory, and then create the github repository using shell git commands from that source area?
21:13 osse xcyclist: it's so impractical it's literally impossible
21:13 lmat I would like to rebase a few commits onto my branch.
21:13 sharkman joined #git
21:13 lmat I have branches a,b somewhere in the history, and I want them on top of my current branch, br.
21:13 xcyclist Well, that's a solid answer.  Thank you.
21:13 lmat git rebase --onto br a^ b;  does sort of the right thing, but it leaves br behind (and leaves me in detached HEAD).
21:13 Vampire0 xcyclist, install git-spindle. It has git subcommands for amongst others GitHub
21:14 xcyclist Ok.  Thanks.
21:14 Vampire0 xcyclist, with it you can do almost everything that is possible through the API and it works great
21:14 lmat It may be that I'm abusing git rebase here...
21:14 lmat because I'm not rebasing some branch around 'b' to br...maybe I need a cherry-pick?
21:15 xcyclist I don't see git-spindle as a native ubuntu package.
21:15 Vampire0 lmat, what do you mean by "leaving br behind"? Why should br change, you just rebase onto it
21:15 Vampire0 xcyclist, it is in some ppa that Seveas provides afair
21:16 osse readme lays it all out
21:16 xcyclist Definitely interesting.  It's on my short list.
21:16 xcyclist Once again thank you.
21:16 jeffreylevesque joined #git
21:16 lmat Vampire0: Because I want those commits "on" br. I want br to include those changes.
21:17 Vampire0 xcyclist, sudo add-apt-repository ppa:dennis/python && sudo add-apt-repository ppa:dennis/devtools && sudo apt-get update && sudo apt-get install git-spindle
21:17 osse xcyclist: i forgot about git spindle. but fundamentally it's true that git has no native way of creating a repository somewhere else than "here". I think hg does, though.
21:17 clemf joined #git
21:17 osse i think if you push to the middle of nowhere then hg will create a repo there
21:18 Vampire0 osse,  heavily depends on the management software. There are ones that create remote repos on the fly simply when trying to push to them
21:18 blaster joined #git
21:19 osse oh nice
21:19 Vampire0 lmat, then you either want to rebase br onto those changes or merge those changes into br, or if it is only some specific commands, you are after man git cherry-pick. Most likely the last one is what you are after
21:19 gitinfo lmat: the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html
21:20 netcarver joined #git
21:21 Vampire0 osse, gitolite should be capable of this if you configure it like that afair
21:24 jwest joined #git
21:30 BlaXpirit can I tell  git diff  not to print any text after @@ lines?
21:30 Vampire0 osse, yep, called wild repos: http://gitolite.com/gitolite/wild.html :-)
21:31 Vampire0 BlaXpirit, you mean suppressing the function name?
21:31 BlaXpirit Vampire0, well yeah except it prints just the previous line in my case
21:33 xcyclist joined #git
21:33 xcyclist So, I just mucked up my local repository by forgetting to take out some huge test data files before the initial git add.
21:33 efco joined #git
21:34 osse BlaXpirit: echo '* diff=mystuff' > .git/info/attributes; git config diff.mystuff.xfuncname '^$'
21:34 lmat Vampire0: Thanks.
21:34 xcyclist I think there is some way to get rid of the cruft I just made, but I don't recall it.
21:34 osse (untested)
21:34 imacinto joined #git
21:34 imacinto joined #git
21:34 osse xcyclist: git rm --cached
21:34 xcyclist I don't want to git push several gigabytes of fake data.
21:34 xcyclist Ok...
21:34 osse have you committed yet?
21:34 BlaXpirit osse, uhhhhh no, I'm looking for an argument, not a permanent solution
21:34 xcyclist I had to cancel the commit, because it was bringing down mysystem.
21:35 osse xcyclist: I'd just rm -rf .git; git init and start over
21:35 xcyclist Ok.  Thanks.
21:35 osse BlaXpirit: there is none
21:35 BlaXpirit ok thanks
21:35 Rodya_ joined #git
21:36 osse BlaXpirit: git diff | sed 's/^\(@@.*@@ \).*/\1/
21:36 osse \o/
21:37 BlaXpirit I get the point, thanks
21:37 deadprez joined #git
21:37 Vampire0 xcyclist, if you didn't commit yet, just use man git reset
21:37 gitinfo xcyclist: the git-reset manpage is available at http://jk.gs/git-reset.html
21:38 osse If it was possible to define temporary attributes on the CLI one could  create an alias
21:38 osse BlaXpirit: but what I showed earlier is actually meant for configuring what you *want* to be showed. And that might be worth a permanent config
21:39 mellernoia joined #git
21:39 xcyclist I cannot use the git init, as I'm pulling down from github to check in there...which is again the eariler topic.
21:39 xcyclist So I just cloned it again.
21:39 BlaXpirit osse, no, I don't mind it showing, i want to store the diff, not read it,  in this particular scripted occasion
21:39 osse ah
21:40 osse in that case I've no better suggestion that stripping it out with sed or equivalent
21:40 laumann joined #git
21:40 xcyclist Perhaps I asked that wrong though.  Perhaps I can create the github repository there, and do the rest from the shell, if I just specify the URL of the repository somewhere locally.
21:40 jimi_ joined #git
21:41 _ADN_ joined #git
21:42 Vampire0 xcyclist, as I said, with git-spindle you can do everything from shell. without git-spindle or a similar tool (like hub, similar but worse) you can just do git operations, meaning you have to create the repo on GitHub, then you can use it locally, either by adding it as remote to an existing repo, or by cloning it
21:42 xcyclist Yes, it looks like I just need to use:  git remote add origin remote repository URL
21:42 r4z joined #git
21:42 xcyclist Then the push will work.
21:43 osse xcyclist: from an empty repo that will not work unless the github repo is also empty to begin with
21:43 xcyclist Yes, thank you Vampire0, that reinforces it.  However, I was mistaken about how I was asking the question.
21:43 buzzLightBeer joined #git
21:43 xcyclist What if the github repo has a Readme file it automatically added, and nothing else?
21:44 xcyclist Bydefinition, not empty.
21:44 jason237 joined #git
21:44 rominronin joined #git
21:44 osse git is never fuzzy about these things :p
21:44 thomas joined #git
21:44 rgrinberg joined #git
21:44 xcyclist That is reassuring.  So, I could delete and recreate it empty though.
21:44 osse This is why I recommend not allowing github to create a template readme and/or license.
21:45 osse it just confuses people when they have an existing repo already
21:45 xcyclist I shall strongly consider that ethic in my future behaviors.
21:45 qws-user-1228 joined #git
21:45 osse Well, now you know, so do whatever feels right in each scenario :)
21:48 davidfetter joined #git
21:48 davidfetter hi
21:49 davidfetter i want to undo some of the changes to an unstaged file
21:49 davidfetter is there some relatively straightforward way to do this in git?
21:49 osse git checkout -p file
21:49 EvilDMP left #git
21:49 osse works like git add -p, but asks for what you want to discard
21:49 davidfetter perfect :)
21:49 davidfetter thanks :) :)
21:50 diogenese joined #git
21:50 osse davidfetter: if it had been staged you could do  git checkout -p HEAD file
21:50 osse davidfetter: and if it had been staged, and you wanted to unstage parts but not modify the file: git reset -p file
21:51 davidfetter osse, thanks. this is super handy stuff :)
21:51 Dougie187 left #git
21:52 xcyclist So here's another bother:  git config --global user.name myusername doesn't make the prompting go away from git push origin master
21:52 sangy joined #git
21:52 xcyclist How do I tell it that stuff so it doesn't need to ask me each time?
21:52 aax joined #git
21:52 osse xcyclist: that config has nothing to do with access.
21:52 osse xcyclist: man gitcredentials
21:52 gitinfo xcyclist: the gitcredentials manpage is available at http://jk.gs/gitcredentials.html
21:53 osse or use SSH with keys
21:53 xcyclist Ok...thanks.
21:53 osse xcyclist: that config only controls what metadata is put in commits (author etc)'
21:53 ansimation joined #git
21:55 xcyclist So, as long as I'm asking too many questions, is there a way to make the credentials differ for different repositories I have on disk?  Like if I have a node1 from github, and a node99 from codeplex?
21:55 osse i've no clue :P I've never used gitcredentials. I prefer SSH
21:56 osse but I assume yes.
21:56 osse anything else sounds weirx
21:56 osse d
21:56 xcyclist Well, I'm confined to a world where sometimes I need to do things on W$.
21:57 xcyclist So, I need to be able to have different areas like that.
21:57 qws-user-1229 joined #git
21:57 xcyclist It would be possible to have different logins, but less helpful.
21:58 osse from what I can tell the config is per-repo
22:05 andlabs joined #git
22:08 groundnuty how can I make git to display similarity index 95% ?
22:08 groundnuty I mean the similarity index
22:09 ShapeShifter499 joined #git
22:10 EY__ joined #git
22:12 ThiefMaster joined #git
22:12 nilg` joined #git
22:13 iveqy joined #git
22:13 garyserj joined #git
22:14 nullie joined #git
22:14 obiwahn joined #git
22:14 garyserj there are github videos on youtube that purport to teach git but they are full of so much misinformation
22:14 garyserj one video said that git add -A deletes files
22:15 cyan__ joined #git
22:21 aidalgol joined #git
22:23 ojdo joined #git
22:25 star_prone joined #git
22:25 oeau joined #git
22:26 SwiftMatt joined #git
22:26 johnmilton joined #git
22:29 imacinto joined #git
22:30 imacinto joined #git
22:30 anuxivm joined #git
22:32 aidalgol If I understand the .gitignore globbing rules correctly "lib/" should only match lib in the root folder of the repository, but that is not the case on my system.  (git version 2.10.0)
22:35 Dispatch joined #git
22:35 sanjoyd joined #git
22:35 charlesrose joined #git
22:36 Rodya_ joined #git
22:37 linuxmint joined #git
22:37 sanjoyd I'm considering reverting a merge commit.  docs/howto/revert-a-faulty-merge gives me a good sense of what to do; but I have a follow-on question:
22:37 sanjoyd That document mentions that bisection may become harder if we have a revert of a merge, and its revert in the history of the repo.
22:38 sanjoyd Are there are other things like that that can become workflow issues in the future?
22:40 _ikke_ sanjoyd: Might be clear, but merging that branch again will not work
22:40 EvilPenguin joined #git
22:40 nwkj86 joined #git
22:41 sanjoyd _ikke_: I understand that I'll have to revert the revert before I try merging again.
22:42 sanjoyd _ikke_: anything other than that?  I'm trying to do some due diligence to ensure that I won't realize six weeks later that things are broken badly. :)
22:42 _ikke_ I cannot think of anything else
22:42 _ikke_ It just creates a new commit
22:44 stonercoder joined #git
22:45 cdown joined #git
22:45 rominronin joined #git
22:47 sanjoyd _ikke_: thank you for your time!
22:47 ResidentBiscuit joined #git
22:51 groundnuty how can I make git to display similarity index ?
22:51 inflames joined #git
22:51 paanii joined #git
22:53 holodoc joined #git
22:57 ycyclist joined #git
22:57 deadprez joined #git
22:58 deadprez joined #git
22:59 mkoskar_ joined #git
22:59 dalias joined #git
22:59 Rodya_ joined #git
23:02 astrofog joined #git
23:04 stonercoder joined #git
23:07 _ikke_ regarding what?
23:08 r4z joined #git
23:10 mizu_no_oto joined #git
23:11 shmuli joined #git
23:16 imacinto joined #git
23:19 realz joined #git
23:20 BackEndCoder joined #git
23:20 mattcen joined #git
23:23 ducman89 joined #git
23:24 clemf joined #git
23:24 dtrott joined #git
23:26 cagmz joined #git
23:30 akkad joined #git
23:30 codezomb joined #git
23:32 Guest9 joined #git
23:35 dsantiago joined #git
23:35 greendo joined #git
23:37 sharkman joined #git
23:45 zivester joined #git
23:46 rominronin joined #git
23:46 neanderslob joined #git
23:46 rafalcpp joined #git
23:48 cagedwisdom joined #git
23:48 citizen-stig joined #git
23:51 inflames joined #git
23:56 pandeiro joined #git

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