Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-09

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

All times shown according to UTC.

Time Nick Message
00:00 Derperperd joined #git
00:02 GodGinrai joined #git
00:04 peterbecich joined #git
00:04 Anja joined #git
00:05 MeltedLux joined #git
00:06 menip joined #git
00:11 themiddleman_ joined #git
00:11 themiddleman_ left #git
00:11 duderono_ joined #git
00:12 themiddleman joined #git
00:15 madewokherd joined #git
00:15 snowalpaca joined #git
00:16 themiddleman Hey everyone!
00:16 gitinfo themiddleman: 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.
00:17 marianina8 joined #git
00:18 themiddleman We're in a really weird state. A few days ago someone had git crash and now we have a commit with a null parent commit. When we do gitk on a file it shows that commit as the original commit that created the file, but if we do git log all the commits show up from all time. Any idea how we can modify the parent of a commit?
00:20 ocelotsloth joined #git
00:20 rudi_s themiddleman: !backup and then git fsck and check if the repository is intact.
00:20 gitinfo themiddleman: 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
00:22 themiddleman backing up is probably a good idea, I'll report back in a few mins
00:23 Glowball joined #git
00:25 raijin joined #git
00:26 pur3eval joined #git
00:30 hahuang65 joined #git
00:33 BSaboia joined #git
00:35 Derperperd joined #git
00:36 foul_owl joined #git
00:36 foul_owl does "git tag" list all tags across all branches?
00:36 foreal joined #git
00:37 foul_owl Including all remote branches?
00:37 Sasazuka joined #git
00:37 buck1 joined #git
00:38 rudi_s foul_owl: Yes.
00:39 Sazhen86 joined #git
00:39 livingstn joined #git
00:39 Anja joined #git
00:39 foul_owl Even if I have never checked out any of those remote branches? Thanks!
00:41 rominronin joined #git
00:45 jungsubk joined #git
00:46 themiddleman rudi_s: Okay, I ran git fsck and it spit out thousands of lines like: dangling blob (commit hash)
00:46 rudi_s foul_owl: Yes, you only need to fetch them, e.g. with git remote update or similar.
00:47 rudi_s themiddleman: Dangling blocks are not an issue, it just means the blob is unused. To clean that up you can run git gc (but it shouldn't be related to the original issue).
00:47 rudi_s s/blocks/blobs/
00:48 brent__ joined #git
00:48 rudi_s Now check the problematic commits with git show --format=raw $commit and check what's exactly broken.
00:49 oaao joined #git
00:49 nomadsoul joined #git
00:50 TbobbyZ_ joined #git
00:52 livingstn joined #git
00:55 Nicmavr joined #git
00:55 arescorpio joined #git
00:55 Guest83939 joined #git
00:55 themiddleman rudi_s: I ran git gc and I still see a bunch of of those messages when running git fsck. The raw show for the messed up commit shows that every file in our repo was added with this: index 0000000..6a50463
00:57 mischat joined #git
00:58 whaley joined #git
00:59 mehola joined #git
00:59 gugah joined #git
01:00 rudi_s themiddleman: What about previous commits?
01:00 Sasazuka_ joined #git
01:02 anuxivm left #git
01:04 xaviergmail_ joined #git
01:04 themiddleman rudi_s: Okay, older commits are still intact and fine, they also have a parent (hash) as the second line of git show --format=raw. The bad commit does not have the parent (hash) line.
01:04 themiddleman I believe this bad hash thinks it's the root commit of the repo since it has no parent.
01:06 rudi_s themiddleman: Yes.
01:06 rudi_s However I don't think that was the result of corruption, but an accident by the committer.
01:07 rudi_s But I'm curious how you get access to the "older" commits? Is it on a separate branch?
01:07 themiddleman Running git log shows all the commits even before the bad commit.
01:08 rudi_s themiddleman: That sounds weird as the "bad commit" has no parent, how would git know where to look. Maybe git log --decorate helps.
01:10 wxg joined #git
01:11 ojacobson !lol
01:11 gitinfo A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
01:11 ojacobson probably worth inspecting git's idea of the total commit graph
01:11 ojacobson orphan commits aren't _that_ abnormal, it's likely feasible to reassemble the graph from that once you understand what shape it's presently in
01:11 ojacobson (there's always at least one orphan commit in every non-empty repo)
01:12 nickabbey joined #git
01:12 psilo2 joined #git
01:13 Sasazuka joined #git
01:14 DolphinDream joined #git
01:14 psilo2 While rebasing, how can I find my progress as I chew through the commits? As a percentage or number of remaining commits, something like that. Having trouble coming up with good google queries about this
01:15 Darren_ joined #git
01:17 ncthom91 joined #git
01:18 cdown_ joined #git
01:18 faux___ joined #git
01:22 aidalgol joined #git
01:28 chipotle joined #git
01:32 whaley joined #git
01:37 Sasazuka_ joined #git
01:38 BSAlb joined #git
01:40 Gunni joined #git
01:40 ahr3n joined #git
01:44 Sasazuka joined #git
01:46 dandaman joined #git
01:46 PaulCapestany joined #git
01:50 Sasazuka_ joined #git
01:52 planigan joined #git
01:54 Cabanossi joined #git
01:54 jameser joined #git
01:55 lagothrix joined #git
01:58 sargas joined #git
02:00 StuartMI joined #git
02:01 dendazen joined #git
02:02 GodGinrai joined #git
02:02 ShapeShifter499 joined #git
02:09 brent__ joined #git
02:12 gugah joined #git
02:12 rominronin joined #git
02:15 Groovin joined #git
02:15 Groovin joined #git
02:15 pur3eval joined #git
02:15 finalbeta joined #git
02:28 burzum2 joined #git
02:30 d^sh joined #git
02:30 pks joined #git
02:35 mmattice joined #git
02:39 jimklo_ joined #git
02:42 brent__ joined #git
02:47 SwiftMatt joined #git
02:49 svm_invictvs joined #git
02:50 fstd_ joined #git
02:52 nevyn ac
02:54 Cabanossi joined #git
02:57 sunri5e joined #git
02:57 hhee joined #git
02:59 aavrug joined #git
02:59 mehola joined #git
03:02 nowhere_man joined #git
03:04 cqi joined #git
03:04 ojdo joined #git
03:06 geek2nerd joined #git
03:07 fstd_ joined #git
03:08 dave0x6d joined #git
03:12 xissburg joined #git
03:21 nowhere_man joined #git
03:21 duderonomy joined #git
03:23 dreiss joined #git
03:26 swalladge joined #git
03:26 Noodlewitt joined #git
03:27 lagothri1 joined #git
03:27 Vortex34 joined #git
03:32 ahr3n joined #git
03:33 pks joined #git
03:33 sargas joined #git
03:34 chachasmooth joined #git
03:34 kyan joined #git
03:36 gugah_ joined #git
03:38 Gsham joined #git
03:38 SwiftMatt joined #git
03:40 mizu_no_oto joined #git
03:40 rchavik joined #git
03:41 howdoi joined #git
03:42 mda1 joined #git
03:44 durham joined #git
03:44 durham joined #git
03:46 cdg joined #git
03:47 peterbecich joined #git
03:47 muhannad_______ joined #git
03:47 rominronin joined #git
03:47 ResidentBiscuit joined #git
03:49 b3gott3n joined #git
03:49 b3gott3n i have a crazy git question
03:49 b3gott3n so i was using heroku to deploy an app and init a git and then changed my app name so that the git init name changed as well
03:50 b3gott3n how do i init with this new app name on the heroku master?  i asked this in heroku but was thinking that maybe this is more a git problem that i can fix?
03:50 gelei007_ joined #git
03:51 phroa what's the 'git init name', the name of the folder in the repo?  the URL of your remote (git@xxxx.heroku.com:xxx/xxxx)?
03:51 phroa *name of the folder holding the repo
03:52 kadoban man git init
03:52 gitinfo the git-init manpage is available at http://jk.gs/git-init.html
03:52 Derperperd joined #git
03:53 Lyqyd joined #git
03:56 Emperor_Earth joined #git
04:00 hexagoxel joined #git
04:00 gelei007_ joined #git
04:02 cdg_ joined #git
04:03 Noodlewitt joined #git
04:03 GodGinrai joined #git
04:04 pur3eval joined #git
04:06 raspado joined #git
04:07 blackwind_123 joined #git
04:09 Cabanossi joined #git
04:11 Darren_ joined #git
04:11 Gsham joined #git
04:15 chipotle joined #git
04:15 MattMaker joined #git
04:17 rossome joined #git
04:17 pur3eval joined #git
04:19 dreiss joined #git
04:19 lucido-cl joined #git
04:20 robotroll joined #git
04:23 clemf joined #git
04:29 ResidentBiscuit joined #git
04:29 jameser_ joined #git
04:29 Gsham joined #git
04:33 hexagoxel joined #git
04:36 VladGh joined #git
04:38 oskarkv joined #git
04:44 jstimm joined #git
04:49 ojacobson b3gott3n: `heroku git:remote -a <your new app name>` will fix it
04:52 govg joined #git
04:57 mehola joined #git
04:57 mischat joined #git
04:58 metachr0n joined #git
04:58 kasual joined #git
05:09 BeRoKr joined #git
05:11 tvw joined #git
05:12 sargas joined #git
05:20 rominronin joined #git
05:20 ayogi joined #git
05:20 Andrew_K joined #git
05:23 inflames joined #git
05:24 cagedwisdom joined #git
05:28 Cabanossi joined #git
05:31 Goplat joined #git
05:35 Guest65382 joined #git
05:36 sharkz joined #git
05:37 imack joined #git
05:41 rkazak_ joined #git
05:42 auzty joined #git
05:44 peterbecich joined #git
05:46 MattMaker joined #git
05:47 Sample joined #git
05:51 robattila256 joined #git
05:52 BeRoKr left #git
05:55 netj joined #git
05:59 Lunatrius joined #git
06:01 GreySunshine joined #git
06:02 clemf joined #git
06:03 GodGinrai joined #git
06:05 peeps[lappy] joined #git
06:06 aavrug joined #git
06:07 aavrug joined #git
06:10 qt-x joined #git
06:11 thebird joined #git
06:11 rapidshot64 joined #git
06:12 rscata joined #git
06:13 barnett joined #git
06:13 Cabanossi joined #git
06:17 sbulage joined #git
06:18 jungsubk joined #git
06:18 wagle joined #git
06:20 Jellyg00se joined #git
06:22 Drzacek joined #git
06:22 Drzacek Hello
06:23 Drzacek Can I create new branch if I have uncommited changes in my current branch? I would like to put those changes in new branch and NOT commit to current one. I'm afraid that if I create new branch now, my changes will be lost. Any help?
06:23 preaction yes, you can create a new branch with uncommitted changes in your working tree
06:24 dec0n joined #git
06:26 watabou joined #git
06:27 ghoti joined #git
06:27 Drzacek preaction, and the uncommited changes will stay and I will be able to commit them in new branch?
06:28 marianina8 joined #git
06:28 preaction that's what you explained that you wanted, yes
06:28 Xenophon1 joined #git
06:28 preaction if you're unsure, make a !backup
06:28 gitinfo 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
06:29 dandaman joined #git
06:30 jost joined #git
06:33 Murii joined #git
06:33 Drzacek thanks
06:33 bocaneri joined #git
06:34 systmkor joined #git
06:34 Glowball joined #git
06:34 conspervin joined #git
06:34 hahuang65 joined #git
06:35 Guest65382 left #git
06:36 xall joined #git
06:37 conspervin left #git
06:37 bocaneri joined #git
06:38 govg joined #git
06:40 alexandre9099 joined #git
06:44 thiago joined #git
06:44 severance joined #git
06:45 osse !float
06:45 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.
06:47 absinthe joined #git
06:48 MattMaker joined #git
06:48 flobin joined #git
06:48 muhannad_______ joined #git
06:50 freimatz joined #git
06:50 Endarked joined #git
06:50 MineCoins joined #git
06:51 rominronin joined #git
06:55 a_thakur joined #git
06:56 muhannad_______ joined #git
06:58 rapidshot64 joined #git
07:01 Silmarilion joined #git
07:02 Guest39686 joined #git
07:05 snowalpaca joined #git
07:05 chele joined #git
07:06 snowalpaca joined #git
07:06 irqq joined #git
07:06 elect joined #git
07:07 byte512 joined #git
07:10 souther joined #git
07:12 skirk joined #git
07:13 Cabanossi joined #git
07:14 hexagoxel joined #git
07:19 joeyDB joined #git
07:19 variable joined #git
07:19 variable does anyone see a way for me to speed up this script? http://sprunge.us/bDRP
07:19 variable something I'm considering is using something like new-workdir and running it in parallel
07:19 variable but I'm curious about single-threaded optimizations first
07:20 cagedwisdom joined #git
07:20 StuartMI joined #git
07:24 skirk joined #git
07:25 JeroenT joined #git
07:25 dvaske joined #git
07:27 thiago joined #git
07:28 Khisanth joined #git
07:28 nevodka joined #git
07:29 xall joined #git
07:29 Guest2979 joined #git
07:32 zefferno joined #git
07:33 xall joined #git
07:35 markzen joined #git
07:35 markzen hi
07:36 markzen forked some small project on github, made significant modifications and now I would like to rename it -- possibly preserving the info that it was a fork, but there won't be PR or anything
07:36 markzen what's the good way to do that?
07:36 tmg joined #git
07:37 P4nther joined #git
07:38 osse markzen: #github is a better place for that. But I think you can do it from the repo's settings page
07:39 wouter0100 joined #git
07:40 markzen yeah I know how to rename; the question was more, what happens when you rename a fork
07:40 acetakwas joined #git
07:43 Dumblez0r joined #git
07:43 markzen off to #github
07:43 markzen left #git
07:43 osse create a fork and rename it :p
07:44 shampine joined #git
07:48 MattMaker joined #git
07:50 masuberu joined #git
07:55 dvaske_ joined #git
07:55 Darcidride joined #git
07:56 jnavila joined #git
07:57 chigang joined #git
07:58 osse gitinfo: thanks
07:58 gitinfo osse: you're welcome, but please note that I'm a bot. I'm not programmed to care.
07:58 Nilesh_ joined #git
07:59 morfin joined #git
08:02 srandon1213 joined #git
08:03 srandon1213 how can i open a file from a very old commit without rewriting the current   ? i want to have it for example with another name or view it temporarily
08:04 osse srandon1213: git show commit:file
08:04 GodGinrai joined #git
08:04 osse srandon1213: git show commit:file > /tmp/kek
08:04 King_Hual joined #git
08:05 osse GIT_WORK_TREE=/tmp GIT_DIR=.git git checkout commit -- file
08:05 srandon1213 what about viewing the  list of files of a commit  ?
08:05 srandon1213 for example a specific commit '
08:05 srandon1213 ?
08:05 osse git ls-tree commit
08:06 osse add -r or --name-only as you wish
08:06 morfin left #git
08:08 louiz joined #git
08:08 xall joined #git
08:09 Dumblez0r joined #git
08:10 aielima joined #git
08:10 absinthe joined #git
08:10 Glowball joined #git
08:13 Balliad joined #git
08:14 wxg joined #git
08:16 TomyWork joined #git
08:17 luke_nukem joined #git
08:17 dersand joined #git
08:19 PHPanos joined #git
08:20 overlord_tm joined #git
08:21 srandon1213 how can i check commits of another branch...like another branch history  ?
08:21 j416 srandon1213: git log
08:22 xall joined #git
08:23 rominronin joined #git
08:24 xcesariox joined #git
08:26 bruce_lee joined #git
08:26 flobin joined #git
08:28 murii joined #git
08:29 flobinator joined #git
08:30 eroux joined #git
08:31 rominronin joined #git
08:31 Tobbi joined #git
08:32 gtristan joined #git
08:32 JeroenT joined #git
08:33 flaviodesousa joined #git
08:34 srandon1213 when i do ls-tree close to each  file i see an hash ... why ?  can i restore that file in my directory with another name ?  in order to prevent overwrites  ?
08:36 j416 srandon1213: read the part about the git object model in the Pro Git book, it should explain that
08:36 j416 available at git-scm.com
08:36 j416 also, this seems like an xy problem
08:37 j416 what are you trying to accomplish?
08:38 kurkale6ka joined #git
08:38 nettoweb joined #git
08:39 fees joined #git
08:39 ahr3n joined #git
08:39 thiago srandon1213: why do you care that there's more information than you need?
08:40 gomdyoi joined #git
08:41 evie_hammond joined #git
08:42 scoobertron joined #git
08:42 Dumblez0r joined #git
08:43 nowhere_man joined #git
08:44 a_thakur joined #git
08:45 a_thakur_ joined #git
08:47 xcesariox joined #git
08:47 absinthe joined #git
08:48 srandon1213 thiago: i wanted to know how acn i use that information... i mean can i restore the file from the hash ?
08:48 xall joined #git
08:48 louiz left #git
08:49 variable joined #git
08:49 acetakwas joined #git
08:49 MattMaker joined #git
08:49 variable joined #git
08:50 achadwick joined #git
08:51 variable joined #git
08:51 jin_ joined #git
08:52 grayjoc joined #git
08:52 GodGinrai joined #git
08:53 variable joined #git
08:54 muhannad_______ joined #git
08:54 manuelschneid3r joined #git
08:55 dimi1947 joined #git
08:58 Tobbi joined #git
08:59 cousteau joined #git
09:00 JeroenT joined #git
09:00 _ADN_ joined #git
09:01 cousteau How do I make git not mess up the timestamps of files?  For some strange reason when I do a commit, git `touch`es files
09:01 steven noobish question, for some fucked reason git pull aborts and git merge origin/branch aborts as well
09:01 steven since I really just want the origin's state of the repo I did a fetch and get reset --HARD origin/branch
09:02 steven would that be any different then git pull the branch? so pull vs fetch && hard reset?
09:02 MrMojit0 joined #git
09:03 cousteau I understand the reasons why keeping track of file timestamps is a bad idea from a `make` point of view, and for that same reason it feels weird as hell that git is modifying timestamps of files
09:03 JeroenT joined #git
09:04 cousteau (I think this happens when I `git add` a file, either directly or indirectly via `git commit -a`)
09:04 Pieplay joined #git
09:07 clmsy joined #git
09:07 joeyDB joined #git
09:07 tobiasvl steven: yes it's different. pull does a fetch + merge
09:08 tobiasvl steven: if you never want to do/keep local changes, then hard reset is OK. but why does merge (and pull, which does a merge) abort? what's the error?
09:09 mikecmpbll joined #git
09:10 marcogmonteiro joined #git
09:10 CptHaddock joined #git
09:10 CptHaddock_ joined #git
09:11 acetakwas joined #git
09:11 thiago joined #git
09:11 CptHaddock joined #git
09:18 Tobbi joined #git
09:18 _ADN_ joined #git
09:18 srandon1213 what can i do with the hash of a file ?  is there a command to retrieve a file from its hash  ?
09:18 a_thakur joined #git
09:19 tobiasvl srandon1213: what are you making?
09:20 steven tobiasvl: back
09:20 srandon1213 tobiasvl: what  ? i 'm learning git
09:20 steven no idea, really. and I know it does fetch and merge, thats why I did fetch manually and tried to merge it but it simply said "abort"
09:20 srandon1213 i was wondering what can i use for the hash of a file
09:20 tobiasvl srandon1213: you mostly use it for stuff under the hood, porcelain commands etc
09:20 tobiasvl steven: huh, nothing more than "abort"? that's weird
09:21 lmat joined #git
09:21 steven yeah, it printed the file list and in the middle of the list it ended and the new line said abort
09:21 steven cant scroll back in the terminal but it was like
09:21 sbulage joined #git
09:22 steven path/to/file1
09:22 steven path/to/file2
09:22 steven path/t
09:22 steven abort
09:22 jast srandon1213: git cat-file blob <hash of file> to retrieve its contents
09:23 jast same works for the contents of commit, tree and tag objects (just change the 'blob' keyword accordingly)
09:23 ronny joined #git
09:23 upik joined #git
09:24 jast internally git puts the hash in the index and in commits because it's the only way to look up the file contents
09:24 jast well, not in commits, technically, but in the tree objects referenced by the commit object and tree objects in the hierarchy
09:24 srandon1213 thanks jast
09:25 GT4066 joined #git
09:25 masuberu joined #git
09:26 jgrasser joined #git
09:27 srandon1213 guys how can i move a specific commit to the head branch of my history ?  let's say i want a new commit as head which was in the past... how can i do that  ?
09:27 moritz srandon1213: you can use interactive rebase to reorder your commit
09:28 moritz !rewrite applies, though
09:28 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
09:28 Hello71 joined #git
09:28 SupaHam joined #git
09:29 ResidentBiscuit joined #git
09:29 srandon1213 moritz: i do not need to necessarily rewrite history.. i could for example copy that commit and put it as new...
09:29 srandon1213 how can i do this ?
09:29 minibar joined #git
09:30 tobiasvl srandon1213: sooo the past commit was reverted somehow, and you want to re-introduce it?
09:32 srandon1213 exacttly tobiasvl i did some modifications... but now i want to put back that status and commit it...
09:32 finalbeta joined #git
09:32 jast so you basically want to make a new commit that effectively goes back to the contents of an old commit?
09:34 srandon1213 jast: exactly
09:34 cores joined #git
09:34 nettoweb joined #git
09:34 jast okay, there are several ways to do that, let's have a look at them
09:34 seni13 joined #git
09:35 jast the most "high-level" way is to do a bunch of reverts ('git revert', which creates a commit with the reverse of the changes in another commit) and squeeze them into one commit
09:36 jast let's say you want to restore the file contents from commit abc123, then you could step backwards through all of the commits done in the meantime and apply the reverse to your working tree without creating a commit each time: git revert -n <commit>
09:36 jast once you've done that for all the commits you made in the meantime, you can commit the result
09:37 chll_ joined #git
09:37 acetakwas joined #git
09:38 jast the second option uses high-level commands, too, but under some circumstances it can leave your working tree in a somewhat unclean state. basically you checkout everything in the old commit (from the root of your repo: git checkout abc123 -- .) and commit. this does not actually make git remove files that got added in the meantime, though, and that's something you have to pay attention to.
09:39 jast the third option uses low-level commands and is guaranteed to get the right result in the commit but you might still have files left in your working tree that you no longer want. the incantation is 'git read-tree --reset -u abc123' followed by committing
09:39 jast this basically overwrites the index with what's in the old commit, and updates the working tree to match
09:41 mischat joined #git
09:42 srandon1213 thanks jast
09:44 mischat joined #git
09:48 jast I quite like the read-tree trick but I've only needed to use it once or twice, really
09:49 GK___1wm____SU joined #git
09:50 MattMaker joined #git
09:50 aard_ joined #git
09:51 finalbeta joined #git
09:52 cqi joined #git
09:52 m0rdonez joined #git
09:53 snowalpaca joined #git
09:53 m0rdonez joined #git
09:56 GK___1wm____SU joined #git
09:59 testman joined #git
09:59 GK___1wm____SU joined #git
10:01 joeyDB joined #git
10:02 GK___1wm____SU joined #git
10:04 severance left #git
10:05 tomg2 left #git
10:06 Snugglebash joined #git
10:06 GK___1wm____SU joined #git
10:07 a_thakur joined #git
10:11 cev joined #git
10:12 ahr3n joined #git
10:14 m0rdonez joined #git
10:16 dwmw2_gone joined #git
10:21 sbulage joined #git
10:23 m0rdonez_ joined #git
10:25 mischat_ joined #git
10:26 _ADN_ joined #git
10:31 grayjoc joined #git
10:32 stemid left #git
10:34 Tobbi joined #git
10:41 mischat joined #git
10:43 finalbeta joined #git
10:43 _ADN_ joined #git
10:43 Snugglebash joined #git
10:46 forgon joined #git
10:46 forgon joined #git
10:46 forgon joined #git
10:47 forgon How could I undo all changes I made to a local repository?
10:47 forgon I want to go back to the stage where I downloaded it.
10:47 irqq joined #git
10:47 _ikke_ reset all branches to their remote equivelent
10:47 _ikke_ equivalent*
10:48 forgon As given by reflog?
10:49 osse remove all branches, create a new branch by the same name as what origin/HEAD points to
10:49 osse yaay
10:50 forgon mmh
10:50 forgon I don't get that.
10:51 MattMaker joined #git
10:51 crawler joined #git
10:51 osse that's basically what happens when you clone
10:52 rnsanchez joined #git
10:53 tobiasvl forgon: what have you done since you cloned it?
10:53 forgon Too much shit
10:53 Learath2 git reset --hard origin/master ?
10:53 Learath2 you'll lose all changes obv
10:54 tobiasvl could be forgon has done stuff in multiple branches
10:54 forgon That looks like what I want.
10:54 neuro_sys git fetch -p; git reset --hard origin/master; clean -fdv
10:54 neuro_sys will make it all shiny and new
10:54 forgon Less likely.
10:54 osse you're all assume origin/master!
10:54 forgon I just have no clue how to use git and am merely trying to report some bugs I spotted.
10:54 neuro_sys git* clean -fdv
10:54 osse but then again I couldn't find a clean way of dereferencing origin/HEAD
10:54 neuro_sys if it's a seperate branch just replace origin/master with that
10:55 neuro_sys git show-ref master is a good way to compare HEADS, after a fetch.
10:55 tobiasvl this is why I asked forgon what they'd done since they cloned it ;)
10:55 trudko joined #git
10:56 m0rdonez joined #git
10:56 trudko Hi anyone played Git Game here ? need a tip with one level (I would PM to not spoil the for other people)
10:57 _ikke_ Might played on or two levels, can't recall anymoer
10:57 _ikke_ anymore
10:58 mischat_ joined #git
10:58 m0rdonez joined #git
10:59 trudko _ikke_: nvm then got to lvl 8 even found a clue not sure what to do with it
11:01 joeyDB joined #git
11:06 Paraxial joined #git
11:07 pks joined #git
11:10 tcorneli joined #git
11:11 raijin joined #git
11:13 sbasso joined #git
11:13 acetakwas joined #git
11:13 Cabanossi joined #git
11:14 theoceaniscool joined #git
11:14 acetakwas joined #git
11:15 murii_ joined #git
11:16 tmg joined #git
11:16 acetakwas joined #git
11:17 murii__ joined #git
11:17 ferr1 joined #git
11:22 Derperperd joined #git
11:23 hyperair joined #git
11:23 tvw joined #git
11:26 murii_ joined #git
11:26 dsdeiz joined #git
11:28 ahr3n joined #git
11:29 leeN joined #git
11:30 ResidentBiscuit joined #git
11:31 albel727 joined #git
11:31 covey left #git
11:32 freimatz joined #git
11:33 acetakwas joined #git
11:34 platzhirsch joined #git
11:34 covey joined #git
11:34 platzhirsch What's a the best git hook to trigger a deploy when a tag has been pushed?
11:34 platzhirsch Would post-receive work on the client-side?
11:35 tobiasvl platzhirsch: post-receive works on the end that receives the push
11:35 srandon1213 hello all.. can somebody explain to me in practice when it is better a rebase and when a merge  ?
11:35 srandon1213 i tried reading online... but still have doubts about it
11:35 platzhirsch But that only works for self-hosted remote repositories @tobiasvl, doesn't it
11:36 murii joined #git
11:36 platzhirsch oh well, guess I just write a script
11:36 MarioBranco joined #git
11:36 tobiasvl platzhirsch: it only works for remote repos where you have access to add hooks, but that doesn't necessarily mean just self-hosted
11:36 tobiasvl many online services offer webhooks too
11:37 covey joined #git
11:38 marenz_ joined #git
11:39 dwmw2_gone joined #git
11:41 cev joined #git
11:42 platzhirsch tobiasvl: makes sense :) thanks
11:42 srandon1213 let's say i'm working  on a software in which i have a test environment, a production environment and a pre-release environment... would it be a good idea to have those on 3 different branches ?
11:43 Cabanossi joined #git
11:45 Tobbi joined #git
11:46 borkr joined #git
11:47 mischat joined #git
11:48 bongjovi joined #git
11:48 Salz left #git
11:49 tobiasvl srandon1213: sure, depends on your !workflow
11:49 gitinfo srandon1213: Finding the right workflow for you is critical for the success of any SCM project.  Git is very flexible with respect to workflow.  See http://sethrobertson.github.com/GitBestPractices/#workflow for a list of references about choosing branching and distributed workflows.
11:50 tvw joined #git
11:51 srandon1213 tobiasvl: my question was about... if a developer in the develop branch has made some changes... how can i integrate those changes without destructing its branch  ? should i pull ? i mean how can i tell git to only pull certain things and not pull others  ? for example i could have configuration files i want to keep
11:51 rsv joined #git
11:51 tobiasvl srandon1213: lots of differemt questions there. see !configfiles for the last one
11:51 gitinfo srandon1213: 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
11:52 MattMaker joined #git
11:52 tobiasvl srandon1213: and in general: !deploy
11:52 gitinfo srandon1213: 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
11:52 rsv my current branch is default, my remote is origin. how can i push the changes to the git server
11:52 CussBot joined #git
11:52 thebird joined #git
11:53 tobiasvl rsv: what happens if you just "git push"?
11:53 cdown joined #git
11:55 srandon1213 ok thanks tobiasvl
11:55 srandon1213 tobiasvl: if git is not a deployment tooll.. which are common deployment tools ?
11:55 mda1 joined #git
11:56 rsv tobiasvl: i get the message: error: src refspec am437x-pdi does not match any
11:57 jameser joined #git
11:57 rsv tobiasvl: when i do git branch, i see remotes/m/am437x-pdi -> origin/am437x-pdi
11:57 rsv so i presume that i need to push to origin/am437x-pdi
11:59 sarri joined #git
11:59 sarri joined #git
11:59 Snugglebash joined #git
11:59 PHPanos joined #git
12:00 Snugglebash joined #git
12:01 Snugglebash joined #git
12:02 Snugglebash joined #git
12:02 Snugglebash joined #git
12:03 pks joined #git
12:03 mischat_ joined #git
12:07 srandon1213 how can i take updates from other branches without merging/rebasing?  a merge/rebase would delete the branch... instead i want to keep them... because i have branches like "stable", "testing" and so on...
12:07 dwmw2_gone joined #git
12:08 osse merge or rebase doesn't delete
12:09 osse what do you mean by that?
12:09 osse there are only two options left: cherry-pick and manually getting files and commiting then
12:09 srandon1213 osse: i have to keep thosebranches... they are long branches in my workflow... one is testing branch one is production one is the stable and so on
12:09 srandon1213 if i do git merge testing...
12:10 srandon1213 it will delete the testing branch am i wrong '
12:10 srandon1213 ?
12:10 osse yes you are wrong
12:10 osse testing will remain exactly like it is
12:12 thiago joined #git
12:13 eduardas_m joined #git
12:13 masuberu joined #git
12:15 d0nn1e joined #git
12:17 YuGiOhJCJ joined #git
12:18 MarioBranco joined #git
12:19 leitao joined #git
12:19 hexagoxel joined #git
12:20 _ADN_ joined #git
12:21 a_thakur joined #git
12:22 dviola joined #git
12:22 lmatteis joined #git
12:26 TikityTik joined #git
12:29 gugah joined #git
12:29 gugah joined #git
12:29 tmsmith joined #git
12:30 gugah joined #git
12:31 wrouesnel1 joined #git
12:31 dsdeiz joined #git
12:31 dsdeiz joined #git
12:31 gugah joined #git
12:31 gugah joined #git
12:32 netj joined #git
12:32 mischat joined #git
12:34 djb-irc joined #git
12:35 rossome joined #git
12:36 rossome joined #git
12:39 IndigoTiger_ joined #git
12:39 tvw joined #git
12:39 netj_ joined #git
12:39 Jackneilll joined #git
12:40 Tobbi joined #git
12:41 jameser_ joined #git
12:42 notebox joined #git
12:43 chipotle joined #git
12:43 Cabanossi joined #git
12:48 kpease_ joined #git
12:49 topcat joined #git
12:49 topcat I'm on master, how do I undo "git merge --no-commit template-2"
12:50 pbrewczynski joined #git
12:51 ThiefMaster git merge --abort
12:52 ThiefMaster that's the first thing i'd try
12:52 _ADN_ joined #git
12:53 topcat There's nothing to abort the merge worked successful, I just don't actally want master and template-2 together, i wanted the contents of template-2 without a commit i.e. unstaged changes
12:53 canton7 just undo all uncommitted changes, then
12:53 jast the classic approach is removing all uncommmitted changes: git reset --hard
12:54 jast actually, do we have a trigger for that yet?
12:54 jast !reset
12:54 gitinfo reset does two things! (1) without file/path argument: 'git reset [flags] [<commit>]' = make the current branch point to <commit> (default: HEAD). --soft = don't do anything else. --mixed (default) = overwrite the index to match. --hard = overwrite the working files to match. (2) 'git reset [<commit>] -- <path>' = overwrite the index entries for <path> with the content from <commit> (default: HEAD)
12:54 jast !resethard
12:54 BSab joined #git
12:54 Cavallari joined #git
12:54 topcat they are committed i didnt know it would do that. i did cherry-pick --no-commit earlier and it undstages the changes
12:55 canton7 merge --no-commit will not commit anything
12:55 jast 'git merge --no-commit' does not commit
12:55 topcat How do I see where master was before then to move it back
12:56 canton7 again, merge --no-commit does not commit. You do not want to move it back
12:56 jungsubk joined #git
12:56 jast no matter why you do have a commit now, the easiest way to go back is to use the reflog, it makes a record each time you change a branch in some way: git log -g master
12:56 jast just take the commit from before the merge and go back to that: git reset --hard <commit>
12:57 marcogmonteiro joined #git
12:57 dviola joined #git
12:58 cevelans joined #git
12:58 gugah joined #git
12:58 str joined #git
12:59 jast .trigger_edit reset--hard You can go back to the last commit, or any other commit (effectively getting rid of commits which can cause issues, see !rewrite), using 'git reset --hard' with an optional commit/ref argument. This will irrevocably destroy all uncommitted changes, so please be careful!
12:59 gitinfo jast: Okay.
13:01 topcat aw I've f'd it now
13:02 Silmarilion joined #git
13:02 DolphinDream joined #git
13:02 dwmw2_gone joined #git
13:03 scottjl joined #git
13:03 jameser joined #git
13:03 sotoz joined #git
13:04 ankit01ojha joined #git
13:05 jast topcat: what happened? thanks to the reflog we can get almost anything fixed :)
13:06 xall joined #git
13:06 fbwnd joined #git
13:08 twoninefive joined #git
13:08 sotoz joined #git
13:10 felixjet joined #git
13:12 Tobbi joined #git
13:12 shinnya joined #git
13:12 kent\n joined #git
13:12 dglambert joined #git
13:12 dglambert yo
13:12 sunri5e joined #git
13:13 jast ...gurt
13:13 thebird joined #git
13:15 ExoUNX joined #git
13:18 osse yo feel the flow. que pasa paradox
13:18 * bremner dials back osse's drip
13:18 seni13 joined #git
13:18 treehug88 joined #git
13:19 ngladitz joined #git
13:19 ngladitz joined #git
13:20 osse bremner: https://www.youtube.com/watch?v=Bolq56sSGt4
13:22 geek2nerd joined #git
13:22 leitao joined #git
13:24 str joined #git
13:26 jimi_ joined #git
13:27 justanotheruser joined #git
13:27 srandon1213 is there a way to check if there are commits before merging   ?
13:27 Chopper69 joined #git
13:28 Tobbi joined #git
13:28 topcat joined #git
13:28 osse commits where? merging what?
13:28 dviola joined #git
13:29 topcat Sorry PC fell asleep, merge --no-commit guy again here: look I made a sweet diagram of what needs fixing... http://imgur.com/a/wNk4Z
13:30 zxd joined #git
13:30 geek2ne57 joined #git
13:31 dglambert So it looks like there are a few ways to clean my current working dir back to the last commit, but I want to confirm which way is best
13:31 geek2ne72 joined #git
13:32 dglambert I have tracked and untracked file in my WD
13:32 dwmw2_gone joined #git
13:32 zxd I noticed that if I add a file to index with 'git add'  then switch branch the file isn't removed from index isn't it supposed to clear index and have it look exactly like it was from the last commit of the branch I moved to?
13:32 dglambert Lets even say I had staged files too
13:32 nettoweb joined #git
13:33 osse topcat: you want to move template-2 back to 6231 and master back to 2fb5 ?
13:33 topcat zxd: add goes to the stage. just a special area marked for what git is about to commit. nothing happens until commit. zxd what you describe is more like stash
13:33 _ikke_ zxd: git checkout does not touch uncomitted changes if it doesn't have to (if it needs to, it aborts)
13:34 dglambert I could do a git checkout --
13:34 geek2ne80 joined #git
13:34 dglambert I could also do a git clean
13:34 _rgn not the same thing
13:34 dglambert git reset HEAD would just move the staged to unstaged
13:34 grayjoc joined #git
13:35 osse dglambert: git clean handles untracked files only. reset/checkout handles tracked files only.
13:35 _ikke_ git clean -fd + git reset --hard HEAD
13:36 dglambert ok i think I understood that
13:36 dglambert so clean for untracked, reset for tracked
13:36 n1ce joined #git
13:37 troulouliou_div2 joined #git
13:37 osse you could also rm -rf {everything except .git}; git checkout HEAD -- .
13:37 dglambert But is there a 3rd command that does reverts the dir to the last commit regardless of tracked/untracked?
13:37 Vampire0 srandon1213, you mean the commits that will be merged? `git log --oneline master..feature` will log all commits that are in feature but not in master, meaning the commits that will be merged if you merge feature in master
13:37 Derperperd joined #git
13:38 osse dglambert: no
13:38 sunnny joined #git
13:40 a_thakur joined #git
13:40 dglambert ikke what is -fd
13:40 _rgn man git-clean
13:40 gitinfo the git-clean manpage is available at http://jk.gs/git-clean.html
13:40 dglambert ty
13:41 marcogmonteiro joined #git
13:41 InnovativeCoder joined #git
13:42 dglambert man git-checkout
13:42 gitinfo the git-checkout manpage is available at http://jk.gs/git-checkout.html
13:42 dglambert man git-reset
13:42 gitinfo the git-reset manpage is available at http://jk.gs/git-reset.html
13:42 dglambert is checkout just a clever way to reset
13:43 theoceaniscool joined #git
13:43 sotoz joined #git
13:43 _rgn it's for more granular "resetting"
13:43 dglambert so theres not like a git checkout -a
13:44 _ikke_ checkout has two 'modes'
13:44 _ikke_ whole branches, or individual files
13:44 dglambert got it
13:44 topcat so, I have a fork, with 2 commits in a branch, that are pull requested. I want to replace entire those 2 commit in the PR. Do I move my branch, force push it and that update the PR or should I close/reopen
13:44 _rgn or even individual hunks within files
13:44 rapidshot64 joined #git
13:44 _rgn git checkout -p -- path/to/file
13:45 dglambert so if I want to revert my whole WD to the last commit, reset is probably better
13:46 geek2nerd joined #git
13:49 EV3RYDAYPR0GRESS joined #git
13:50 jameser joined #git
13:50 geek2ne78 joined #git
13:50 ghoti joined #git
13:52 ocelotsloth joined #git
13:53 MattMaker joined #git
13:53 InnovativeCoder left #git
13:54 Oatmeal joined #git
13:55 pippijn joined #git
13:55 pippijn hi
13:55 pippijn is there a way to get the branch graph is a machine readable format?
13:56 geek2nerd joined #git
13:56 pippijn git log --oneline --all --graph --decorate # this gives me almost exactly what I want, but in human readable format that's a bit hard to parse
13:56 topcat pippijn: you'd want to look at parsing the --porcelain outputs
13:56 osse pippijn: machnine readable in the sense that you want to draw it yourself?
13:56 _ikke_ what exactly is it that you want to parse?
13:56 pippijn topcat: there is no --porcelain on git log
13:56 pippijn osse: sort of, I want to operate on it
13:56 topcat rats. well i tried
13:56 osse pippijn: you can print the parents of a commit and then reconstruct the graph yourself
13:57 pippijn osse: ok, yeah
13:57 pippijn I'll end up doing that unless there is a better way
13:58 jeffreylevesque joined #git
13:58 jagob joined #git
13:58 jameser joined #git
13:58 osse pippijn: what kind of operations do you have in mind ?
13:58 King_Hual joined #git
13:58 ocelotsloth joined #git
13:59 pippijn osse: I have a root node (branch) that all other branches (a large tree without merges) are based on
13:59 pippijn and I want to rebase the root node together with the whole tree
13:59 gechr joined #git
13:59 JeroenT joined #git
13:59 pippijn so the whole tree is then based on another commit
13:59 pippijn but otherwise looks exactly the same, with all the branching the same way
13:59 osse pippijn: rebase the root first, then for all other branches do  git rebase --onto root root@{1}
14:00 pippijn osse: right, so I have root -> branch1 -> branch2 and root -> branch1 -> branch3 and root -> branch4
14:00 pippijn so I need to: rebase root, then rebase branch1 on root, then branch2 on branch1, etc.
14:00 osse then you have to git rebase --onto root root@{1} branch1; then git rebase --onto branch1 branch1@{1} branch2
14:01 osse etc.
14:01 pippijn (then branch3 on branch1, and in parallel/before/after, branch4 on root)
14:01 pippijn osse: exactly
14:01 dwmw2_gone joined #git
14:01 pippijn and I want to do that automatically
14:01 Rhonda joined #git
14:01 pippijn because I want to do that often, and the tree has around 10 nodes
14:02 Rhonda joined #git
14:03 yqt joined #git
14:04 jeffreylevesque_ joined #git
14:05 dviola joined #git
14:06 aiqmi joined #git
14:06 JeroenT joined #git
14:07 Gsham joined #git
14:09 dglambert Hey is there a bash command to pull up all my command history
14:09 dglambert for a current session
14:10 jast with default settings it doesn't distinguish by session but the command is 'history' :)
14:11 dglambert ugh, I closed and reopened a diffrent window
14:11 gtristan joined #git
14:11 jast in my experience bash doesn't save its history when the shell exits unexpectedly
14:11 jast personally I use zsh which is pretty easy to configure to update its history file after every single command
14:12 jast I don't know that much about configuring bash...
14:13 livingstn joined #git
14:13 topcat How do I move forward the commit a branch was from?
14:14 tobiasvl topcat: what?
14:14 jeffreylevesque joined #git
14:14 Tobbi joined #git
14:15 DarkPsydeLord joined #git
14:16 grawity sounds like you're looking for `git rebase`
14:16 topcat 1 branch from master. master now has 4 more commits. I want the branch back ontop of master
14:16 Dumblez0r joined #git
14:17 porsche944 joined #git
14:17 tobiasvl oh yes "git rebase" then.
14:19 topcat right ok. I had to set the tracking branch
14:19 MineCoins joined #git
14:20 mischat_ joined #git
14:20 mbrumbelow joined #git
14:20 Gloomy joined #git
14:21 GodGinrai joined #git
14:24 nellicus_ joined #git
14:24 jeffreylevesque_ joined #git
14:24 dwmw2_gone joined #git
14:25 osse pippijn: I think I would do git log --decorate root..branch for each branch separately and parse the decorations you encounter. Those branches who have no decorations between root and themselves can be rebased first.
14:26 osse pippijn: Not exactly sure exact how to go from there. But I suppose you then rebase the ones with one decoration
14:27 inflames joined #git
14:29 JeroenT joined #git
14:29 Gsham joined #git
14:30 HardlySeen joined #git
14:30 Dumblez0r joined #git
14:30 mda1 joined #git
14:31 topcat oh man, ok that's all moves and based etc thanks for your input everyone
14:32 topcat so, can i replace the commits to a PR with an entirely different branch and base commit?
14:32 Derperperd joined #git
14:33 topcat because that would be force pushing and I don't think I should do that now the PR has been sent to the other repo
14:33 LionsMane joined #git
14:34 jeffreylevesque joined #git
14:35 tobiasvl topcat: github? more of a github question I guess. usually we here in #git will say you shouldn't force push stuff that people have pulled, but I guess maybe github PRs are a bit special? they're refs in the target repo, but you kinda "own" your fork, right? I dunno what the github rules of thumb is
14:36 permalac joined #git
14:37 topcat tobiasvl: yeh github.
14:38 ahrs joined #git
14:39 flobin joined #git
14:41 notebox joined #git
14:43 Cabanossi joined #git
14:44 geek2ne53 joined #git
14:44 jeffreylevesque_ joined #git
14:44 pks joined #git
14:45 topcat tobiasvl: short answer: went to #github. Yes, and I've just done it so yes you can do that as described.
14:45 tobiasvl alright
14:48 sargas joined #git
14:49 dglambert what is that command call for looking at your servers?
14:50 dglambert like your remote repos
14:50 mischat joined #git
14:50 moritz git ls-remote ?
14:50 moritz git remote -v
14:50 kaldoran joined #git
14:50 moritz depends on what you want
14:50 dglambert remote -v
14:50 dglambert thanks
14:50 sim590 joined #git
14:51 Gsham joined #git
14:53 Dougie187 joined #git
14:53 jameser joined #git
14:54 pippijn osse: yeah, I'm doing that now
14:54 MattMaker joined #git
14:54 jeffreylevesque joined #git
14:55 synthroid joined #git
14:55 govg joined #git
14:55 ocelotsloth joined #git
14:56 geek2nerd joined #git
14:57 m0rdonez joined #git
14:58 Takumo joined #git
14:58 Takumo joined #git
14:58 snowalpaca joined #git
15:00 ujjain joined #git
15:00 ujjain joined #git
15:03 AaronMT joined #git
15:04 geek2ne23 joined #git
15:04 jeffreylevesque_ joined #git
15:04 brent__ joined #git
15:04 phos1 joined #git
15:05 crose joined #git
15:06 phos1 I want to put 2 projects in 1 GIT repository. Is it possible to have GIT push / Pull only a specific folder (1 website is partly .net and partly Php, so it’s a related task but 2 areas)?
15:06 _ikke_ no
15:07 slackR joined #git
15:07 phos1 That’s what i thought it will want to push everything everywhere.
15:10 geek2nerd joined #git
15:10 ejb joined #git
15:11 LeBlaaanc joined #git
15:11 geek2nerd joined #git
15:12 Stutters joined #git
15:12 Stutters does anybody know how to set up netbeans and git to work together?
15:13 gigq joined #git
15:13 Stutters I have files on my server already and I want to add a versioning system to back it up
15:13 Stutters I'm running a GitLab server for that
15:13 Silmarilion Stutters: netbeans works out of the box with Git
15:14 Stutters yes, but I can't get it to work :|
15:14 Stutters I thought, maybe if I describe my steps somebody can kick me in the butt for doing bad things
15:14 dglambert Hey I got my command history in Mingw with the history command, does anyone know if there is a log ?
15:14 al-damiri joined #git
15:14 Silmarilion Stutters: you can do git init . in the projects directory or use TEAM => Git
15:14 dglambert of the actual output of all my commands and everything
15:14 jeffreylevesque joined #git
15:15 Silmarilion Stutters: after you do git init Netbeans will recognize the repository
15:15 Stutters so existing files on a remote machine -> init?
15:15 Stutters Silmarilion: okay, I went so far
15:16 zxd how to empty index file ?
15:16 thiago zxd: why do you want to do that?
15:16 Silmarilion Stutters: so you want to fetch the code from the remote Gitlab repository?
15:17 zxd thiago: I want to start all clean
15:17 Stutters Silmarilion: it's my own server GitLab is running on, the project is empty (nothing to clone from except for the emtpy project). My files do exist on the very same server though so apache can access them
15:17 thiago zxd: start what?
15:18 thiago zxd: do you already have commits in this branch?
15:18 Stutters I do have my project opened in netbeans, I'm not sure I can set git up on an existing project or if I should also create a new project
15:19 nettoweb joined #git
15:19 geek2nerd joined #git
15:19 Silmarilion Stutters: just run git init in the projects directory, and commit the files. Next follow gitlab guides how to add remote and push to it
15:19 sotoz joined #git
15:20 dglambert Ugh, well I can't scroll back far enough anymore, but I merged and had maybe 5 files with conflicts (via git status) so I did git mergetool and corrected two of them, then the other 3 disappeared
15:21 dglambert my merge tool is kdiff
15:21 zxd thiago: never mind, just experimenting stuff
15:21 dglambert do I have to git clean and git reset then remerge
15:22 mbrumbelow joined #git
15:22 thiago dglambert: what do you menan by "disappeared" ?
15:22 darkbit joined #git
15:23 gusnan joined #git
15:23 zxd I noticed if I create a new branch and checkout to it  then 'git add <file>'   and checkout back to previous branch  modify some more tracked files here then the file I had added while working in the branch before also gets commited! have to be careful with that
15:23 zxd maybe I didn't itent to include the file I had added but forgot to commit to the branch i switched after
15:23 thiago zxd: !float
15:23 gitinfo zxd: 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.
15:24 paissad joined #git
15:24 jagobdue joined #git
15:24 Silmarilion Bare repo vs normal repo for code deployment on remote servers?
15:24 _ikke_ !deploy
15:24 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
15:24 Silmarilion !deploy
15:24 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
15:25 jeffreylevesque_ joined #git
15:25 Silmarilion _ikke_: tnx
15:25 paissad hello guys, i just did a "git svn clone http://svnrepo" .... and now a have a directory with my svn content. Is this directory a git repo i could reach via git protocol ? ... or is it simply a git working dir
15:25 paissad actually, i want to import the contents into a new GITLAB project
15:25 paissad that's the goal
15:26 thiago paissad: you could access it via one of the Git protocols
15:26 King_Hual joined #git
15:26 jcelerier joined #git
15:26 jcelerier hello :)
15:26 thiago paissad: for example, by pushing it to a server (like GitLab)
15:26 jcelerier in my repository's .git folder, I have a lot of old submodules that were since removed from the repository that are still referenced
15:26 jcelerier how can I clean them ujp ?
15:26 thiago jcelerier: rm
15:26 thiago (not git rm)
15:27 paissad thiago: so basically, i should only create a new gitlab project .. and then push the recently "git svn cloned" one ?
15:27 dglambert thiago: I had 5 conflicts, 2 of them I got to merge using kdiff the rest I don't see now when I do a git status
15:27 thiago paissad: yes
15:28 paissad thiago: thanks mate
15:28 thiago dglambert: kdiff3 sometimes fixes a conflict without opening the window
15:28 jcelerier thiago: they are also in various .git/config files
15:28 gopar joined #git
15:28 m0rdonez joined #git
15:28 jcelerier but well, I guess this means there is no built-in command for this
15:28 dglambert oh
15:28 thiago jcelerier: don't think so
15:29 ayogi joined #git
15:29 thiago dglambert: depending on the resolution, it might disappear from git status too: if the resultion is to make it exactly like the current commit
15:29 Some_Person I rebased a branch in order to pull in commits that were made to master. A coworker attempted to pull it, and is now in merge hell. He has no local commits in this branch that he needs to keep, but he has made commits in other branches that haven't been pushed. What should he do?
15:29 kexmex joined #git
15:29 dglambert thiago: got it
15:29 dglambert not sure how kdiff can make that decision though
15:29 _ikke_ Some_Person: Is the merge still ongoing?
15:30 Some_Person yes
15:30 _ikke_ git merge --abort
15:30 Some_Person then what?
15:30 davidosomething joined #git
15:30 Balliad joined #git
15:30 ResidentBiscuit joined #git
15:32 Some_Person basically, it seems like he just needs to throw away this branch and get the latest from source control
15:32 _ikke_ If you rebase the branch, git pull --rebase should fix it
15:32 Gustavo6046 joined #git
15:33 mischat joined #git
15:34 TomyLobo joined #git
15:34 NeverDie joined #git
15:35 jeffreylevesque joined #git
15:36 dreiss joined #git
15:37 brent__ joined #git
15:42 jeffreylevesque_ joined #git
15:46 mischat_ joined #git
15:46 mischat_ joined #git
15:47 Chrismeister joined #git
15:47 fbwnd joined #git
15:47 zxd what do the dots .. mean in $ git merge hotfix Updating f42c576..3a0874c
15:48 gugah_ joined #git
15:48 permalac joined #git
15:48 Rhonda zxd: from-to commitish seperator.
15:49 mischat__ joined #git
15:49 Dougie187 left #git
15:49 canton7 zxd, !dots
15:49 gitinfo zxd: A..B = stuff that happened between A and B (if A and B are related; otherwise refer to "man gitrevisions"), A...B = (a) in history viewers: stuff that isn't in both A and B yet; (b) in "git diff": stuff that happened in B since the two diverged; (c) in "git checkout": the merge base of A and B. "master.." is the same as "master..HEAD" and "..master" is the same as "HEAD..master", and so forth.
15:51 a_thakur joined #git
15:51 permalac joined #git
15:52 jeffreylevesque joined #git
15:52 Derperperd joined #git
15:53 nd joined #git
15:54 sunnny left #git
15:54 permalac joined #git
15:55 MattMaker joined #git
15:55 Dougie187 joined #git
15:57 jeffreylevesque_ joined #git
15:58 Bizkit joined #git
15:58 Derperperd joined #git
15:59 rafalcpp_ joined #git
16:00 max3 joined #git
16:00 mischat joined #git
16:01 errst joined #git
16:01 errst joined #git
16:02 tirej joined #git
16:02 mmlb joined #git
16:02 tirej joined #git
16:02 tirej hi guys
16:02 gitinfo tirej: 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.
16:02 tirej i've conflict in pr how can i solve it ?
16:02 phos1 left #git
16:02 bsanford joined #git
16:03 tirej i've branched from master and opened pr to dev
16:03 impliednude joined #git
16:06 Seveas joined #git
16:07 jeffreylevesque joined #git
16:07 jimi_ joined #git
16:08 muhannad_______ joined #git
16:09 jedahan joined #git
16:09 Tobbi joined #git
16:09 metiscus joined #git
16:09 _ikke_ You have to solve the conflict locally
16:10 jamiejackson joined #git
16:10 _ikke_ tirej: usually by rebasing the branch
16:11 jamiejackson hi folks. a colleague's team is mostly able to access a client's git repo, but some team members can't. is it possible to set up a secondary or intermediate git server that the whole offsite team can access?
16:11 Glooomy joined #git
16:13 jamiejackson i'm assuming so, two levels of upstream, but i'm still not super knowledgeable about git
16:14 skalpin joined #git
16:14 Some_Person I have 2 branches, let's call them A and B. B was originally branched from A, and has had commits added. Meanwhile, A has been rebased and had commits added as well. How do I get B on top of everything currently in A?
16:15 osse Some_Person: git rebase A. If that fails or tries to use too many commits:  git rebase --onto=A {hash}  where hash is the first commit that you consider not part of B
16:15 _ikke_ jamiejackson: Sure, git is distributed, so that's completely possible
16:16 _ikke_ jamiejackson: Someone or something needs to make sure that the two repos keep in sync
16:16 MacGyver Some_Person: git checkout B && git rebase A
16:16 MacGyver Some_Person: From there, everything osse said.
16:17 dgonzo joined #git
16:17 clemf joined #git
16:17 GodGinrai joined #git
16:17 eh3 joined #git
16:17 jeffreylevesque_ joined #git
16:17 kriops1 joined #git
16:18 ak77 joined #git
16:18 TbobbyZ joined #git
16:19 Some_Person It runs into a ton of conflicts when I try to do 'git rebase A'
16:19 s1scha joined #git
16:19 kriops1 Hi! I have a school project and we're attempting to cooperate on github. Is there a guide that explains step-by-step from cloning to submitting a pull request? I am not able to make sense of the documentation.
16:19 bsanford joined #git
16:19 tirej _ikke_: which branch ?
16:20 _rgn kriops1: i googled this myself http://hisham.hm/2016/01/01/how-to-make-a-pull-request-on-github-a-quick-tutorial/
16:20 Some_Person conflicts that don't even make much sense. I think it might be trying to rebase all the commits common between A and B, and not just the 2 unique ones in B
16:20 _rgn kriops1: contents seem reasonable
16:21 kriops1 _rgn thank you! My google-fu failed me on this one.
16:21 ak77 I want to translate hg subrepos to git submodules, to correct revisions in entire history of the repo. using filter-branch's --tree-filter, I check changes in .hgsub and add missing submodule with git submodule... but it doesn't stick and so it wants to be added in each step of the history.
16:22 ak77 I check for a presenseof the submodule with git submodule status --cached $path
16:22 _ikke_ tirej: The branch you want to create the PR for
16:23 jamiejackson thanks, _ikke_
16:23 Some_Person situation is like this: http://pastebin.com/X1n11Sak
16:24 robotroll joined #git
16:25 snits joined #git
16:25 Some_Person I know how to get this to work in mercurial, but not git
16:25 thiago joined #git
16:26 osse then follow my advice with the --nto
16:26 osse --onto
16:28 Some_Person osse: I don't understand how that helps, if I'm specifying a commit in A
16:28 jeffreylevesque joined #git
16:28 brent__ joined #git
16:28 Bizkit joined #git
16:28 osse Some_Person: because then git will only try to rebase the commits in {hash}..B
16:28 osse and put those on top of A
16:29 Some_Person osse: But then which hash would I specify?
16:29 osse Some_Person: like i wrote, first commit that you consider not part of B.
16:30 dgonzo joined #git
16:30 osse Some_Person: you said B only has two commits. So I guess that means the third one
16:30 osse (aka. B~2)
16:30 brent__ joined #git
16:31 osse which means Git will try to put the second most recent commit (B~1) and the most recent (B itself) on top of A
16:32 Some_Person osse: This is the situation: http://pastebin.com/nz5Ue7yx
16:32 kexmex joined #git
16:33 raynold joined #git
16:33 Some_Person Except I made a slight mistake. Should be Branch B at the bottom
16:34 Some_Person I basically want to take the "commits in B only" and put them on top of A as it currently stands.
16:34 Some_Person and make that the 'new' B
16:34 sargas joined #git
16:34 bernardio joined #git
16:35 aboccia joined #git
16:36 Some_Person does that make sense? can you see why I'm kinda confused?
16:37 AaronMT joined #git
16:37 dgonzo joined #git
16:37 Some_Person hg rebase lets you specify source and destination commits. Is there a way to specify source with git, since that seems to be what I need?
16:38 gugah joined #git
16:38 jeffreylevesque_ joined #git
16:38 rafalcpp joined #git
16:40 rominronin joined #git
16:40 Murii joined #git
16:41 permalac joined #git
16:42 Guest2979 left #git
16:42 Some_Person I seem to have gotten what I want by running 'git rebase --skip' 6 times
16:43 j08nY joined #git
16:43 Derperperd joined #git
16:44 Some_Person is that really the only way?
16:44 shgysk8zer0 joined #git
16:46 dviola joined #git
16:47 karstensrage joined #git
16:47 ejb joined #git
16:48 jeffreylevesque joined #git
16:48 jorn joined #git
16:49 jorn hey, my `git commit --int` "patch" doesn't allow me to select files anymore...
16:49 jorn is that intended?
16:49 jorn (2.12.0)
16:49 ruxu joined #git
16:52 cdown joined #git
16:53 publio joined #git
16:54 dreiss joined #git
16:54 PHPanos joined #git
16:55 dgonzo joined #git
16:55 cdg joined #git
16:55 ferr1 joined #git
16:56 osse Some_Person: I still think my solution would exactly do what you want
16:56 MattMaker joined #git
16:56 madewokherd joined #git
16:56 hmota57 joined #git
16:57 de-vri-es joined #git
16:57 osse What you say you basically want is exactly my suggestion. The hash I was talking about is where branch B "stops" so to speak
16:57 dgonzo joined #git
16:58 theoceaniscool joined #git
16:59 xissburg joined #git
16:59 ocelotsloth joined #git
17:00 guardian hello, I have a file in the index that adds trailing spaces, and in the work tree I removed the trailing spaces
17:00 jimi_ joined #git
17:00 guardian when I do git add path/to/file the version in the index keeps having trailing spaces
17:00 guardian what could be the reason please?
17:02 gopar joined #git
17:02 gopar joined #git
17:02 durham joined #git
17:04 s2013 joined #git
17:04 jorn my `git add --interactive` / patch doesn't allow me to select files anymore... is there any way to get that back?
17:04 hyuvkr joined #git
17:05 jorn i can't find any relevant config options :-/
17:07 tcorneli joined #git
17:07 _ikke_ jorn: Nothing has changed around that
17:07 _ikke_ jorn: What does git status say?
17:07 jorn _ikke_: 4 files changed...
17:08 jorn i noticed this a while ago, but now i just want to commit a line quickly without a huge other diff
17:08 jorn i was pretty used to selecting the files...
17:08 jorn any env vars taht could cause that?
17:08 thebird joined #git
17:09 dgonzo joined #git
17:09 muhannad_______ joined #git
17:09 jorn (i just tried moving my ~/.gitconfig, same result...)
17:09 dviola joined #git
17:11 ChadTaljaardt joined #git
17:13 shgysk8zer0 joined #git
17:13 durham joined #git
17:13 multi_io joined #git
17:15 j416 jorn: what does all of git status say?
17:16 WayToDoor joined #git
17:16 pgou joined #git
17:17 hhkkaa joined #git
17:17 jorn j416: http://pastebin.com/Y7qcWJQz
17:18 jorn if i then enter `git commit --int`, [p], it doesn't ask me which file to patch, but jumps straight into hunk mode
17:19 j416 that's how it always worked
17:19 jorn same for `git add --interactive` or `git add --patch`
17:19 BackEndCoder joined #git
17:19 jorn nope
17:19 j416 `git add -p` at least
17:19 jorn it used to ask me to enter numbers to select the files i want to patch
17:19 mischat joined #git
17:19 j416 never used -i
17:20 j416 I ser
17:20 j416 I see
17:20 nixjdm joined #git
17:20 jorn ok, possible, i never used --patch ;)
17:20 j416 well, is it a problem?
17:20 jorn it's certainly not as usable as it was for me
17:21 j416 how so?
17:21 _ikke_ jorn: What do you get when you run git add -i
17:21 jorn _ikke_: same
17:22 dgonzo joined #git
17:22 jorn http://pastebin.com/CK1SF7KE
17:22 Darren__ joined #git
17:23 jorn j416: it's not that bad with my "simplistic reproduction", but in another repo i have 12 changed files, out of which i want to commit parts from 3
17:23 oyemohit joined #git
17:23 Derperperd joined #git
17:23 _ikke_ jorn: And where does it differ from what it normally did?
17:23 impliednude joined #git
17:23 jorn with the old mode i could do `git commit --int`, [p], [1,2,7] and then quickly get the hunks of those
17:24 _ikke_ ok, so after you enter 'p', you could select files?
17:24 j416 jorn: you can skip a file with 'd'
17:25 Noldorin joined #git
17:26 dgonzo joined #git
17:26 Glooomy joined #git
17:26 jorn i just switched back to git 2.10.1, then the patch mode looks like this: http://pastebin.com/Pnvrm2jf
17:26 nowhere_man joined #git
17:26 rodan joined #git
17:27 j416 interesting
17:27 Qlai joined #git
17:27 j416 perhaps the changelog provides a clue?
17:27 jorn j416: yupp, but it was pretty convenient to just select the files (for me) like [1,2,5-8]
17:28 rajkosto joined #git
17:28 rajkosto any channel for git-lfs ?
17:28 jorn will have a look
17:28 MarioBranco joined #git
17:28 VladGh joined #git
17:30 dviola joined #git
17:30 _ikke_ jorn: otherwise, you can try git bisect to find what commit made that disappear
17:30 j416 rajkosto: not sure. Maybe here, maybe #github?
17:30 jorn 2.12.0
17:30 rajkosto isnt on github
17:30 j416 what do you mean?
17:30 rajkosto my own git server with lfs, but its not a server problem
17:30 j416 afaik github built git-lfs
17:31 rajkosto i basically added commits while a push was going
17:31 SuperLag joined #git
17:31 tirej joined #git
17:31 rajkosto so now it thinks there are lfs objects on the server when they arent
17:31 j416 I may be mistaken
17:32 j416 interesting. I'll add that to my list of reasons to avoid it.
17:32 dgonzo joined #git
17:32 VladGh joined #git
17:33 SuperLag If I say "git config --global core.autocrlf false" shouldn't that setting stick? when I do "git config --list" after that first command, I see both core.autocrlf=true and core.autocrlf=false in the output.
17:33 ejb joined #git
17:33 SuperLag I'm working from Windows 10, but 100% of the code I'm working on is for (and runs on) Linux machines... so I *always* want stuff to be LF and not CRLF.
17:33 flobin joined #git
17:34 svm_invictvs joined #git
17:35 plugwash joined #git
17:35 marianina8 joined #git
17:35 plugwash what is the best way to create a commit with an existing commit as a parent but none of the files from that commit?
17:36 chardan joined #git
17:36 dgonzo joined #git
17:37 irqq joined #git
17:37 mehola joined #git
17:38 flobin joined #git
17:38 dgonzo joined #git
17:40 tmsmith joined #git
17:40 rominronin joined #git
17:43 bgerber joined #git
17:43 raspado joined #git
17:43 marianina8 joined #git
17:43 Vampire0 plugwash, delete all files and commit?
17:44 Rodya_ joined #git
17:45 mischat joined #git
17:46 mischat joined #git
17:47 geek2ne32 joined #git
17:53 salamanderrake joined #git
17:54 ertesx joined #git
17:54 gopar joined #git
17:56 bgerber joined #git
17:56 MattMaker joined #git
17:57 ahrs joined #git
17:58 pfrench joined #git
18:01 hahuang61 joined #git
18:02 gopar joined #git
18:02 borkr joined #git
18:03 gopar joined #git
18:05 jnavila joined #git
18:05 j416 ^
18:06 j416 SuperLag: unless your repo config overrides it (cat .git/config | grep crlf)
18:07 osse UUOC
18:07 math-alpha joined #git
18:09 Derperperd joined #git
18:09 orbyt_ joined #git
18:09 SuperLag j416: been reading a bit. Just want to make sure I understand this right. If I set it to "auto", Windows will open $FILE as CRLF, but git knows it's originally LF, and will save it that back to LF on commit?
18:11 peeps[lappy] joined #git
18:12 jimklo joined #git
18:12 j416 I don't remember; I don't use windows. Sorry.
18:13 jtoy joined #git
18:13 SuperLag No worries.
18:13 jimklo joined #git
18:14 clemf joined #git
18:14 Cassiopaya joined #git
18:14 aielima joined #git
18:14 mikecmpbll joined #git
18:14 jedahan joined #git
18:14 jimi_ joined #git
18:14 elect joined #git
18:15 ejb joined #git
18:15 a3Dman joined #git
18:16 oyemohit joined #git
18:16 jnavila joined #git
18:19 Mobutils joined #git
18:19 math-alpha left #git
18:20 Seveas joined #git
18:25 dviola joined #git
18:25 dviola joined #git
18:28 rodan joined #git
18:28 satifant joined #git
18:29 robotroll joined #git
18:30 duderonomy joined #git
18:30 TikityTik joined #git
18:30 jeffreylevesque joined #git
18:31 Mobutils left #git
18:31 vuoto joined #git
18:33 Mindi joined #git
18:35 iffraff joined #git
18:36 iffraff hi I'd like to do something like this -> git branch -d git branch --merged.  is this possible?
18:36 dviola joined #git
18:37 Sample joined #git
18:38 jost joined #git
18:39 jwest joined #git
18:39 mda1 joined #git
18:39 osse iffraff: git for-each-ref --merged HEAD --format='%(refname:short)' refs/heads | grep -v master | xargs git branch -d
18:39 osse that's one way
18:40 osse replace master with the one currently checked out
18:40 osse sometimes I wish a branch was not concidered as merged with itself. oh well.
18:41 osse actually, since that's a pain in the ass anyway, you might as well do:  git branch --merged | grep -v master | xargs git branch -d
18:41 iffraff wow, that's kind of complicated.
18:41 osse conceptually it's very
18:41 iffraff but I guess it makes sesne.
18:41 osse conceptually it's very simple. it just looks comlicated because it's one big shell pipeline
18:42 osse you could split it up in managable pieces. or write it in a scripting language you're more comfy with
18:42 joeyDB joined #git
18:43 dviola joined #git
18:43 mischat joined #git
18:43 osse iffraff: simpler still:  git branch -d $(git branch --merged | grep -v '^\*')
18:44 iffraff cool, yea, thank you.  I was AHA that's sort of what I was thinking!
18:44 iffraff thanks!
18:44 absinthe joined #git
18:44 osse technically branches name can contain chars that will be a problem, but fuck it
18:45 cdown joined #git
18:46 BlackMaria joined #git
18:46 iffraff yea deal with those alacarte
18:47 thebird joined #git
18:48 osse also i wish that for-each-ref's default format was %(refname:short)
18:51 tirej left #git
18:52 Murii joined #git
18:52 cdown joined #git
18:52 ispn joined #git
18:53 Gloomy joined #git
18:53 gugah_ joined #git
18:54 jeffreylevesque_ joined #git
18:55 gigq joined #git
18:55 m0viefreak joined #git
18:55 causasui joined #git
18:56 max3 joined #git
18:56 aard_ joined #git
18:57 MattMaker joined #git
18:59 overlord_tm joined #git
18:59 aesthetik joined #git
19:00 avar mm, forgotten, what's the most idiomatic git command to get a blob hash from a <path> at a given <commit>?
19:01 aesthetik left #git
19:03 theoceaniscool joined #git
19:03 dviola joined #git
19:04 oyemohit joined #git
19:05 aboccia joined #git
19:05 avar Ah, git rev-parse commit:path
19:06 zezba9000 joined #git
19:08 durham_ joined #git
19:09 avar now how about the tree a commit points to...
19:09 inflames joined #git
19:09 UTAN_dev joined #git
19:10 sargas joined #git
19:11 qqx git rev-parse 'commit^{tree}'
19:12 rominronin joined #git
19:12 avar whee, thanxs
19:12 metalraiden34 joined #git
19:13 Sasazuka joined #git
19:14 flobin joined #git
19:14 gopar_ joined #git
19:14 aw1 joined #git
19:15 boombatower joined #git
19:16 WeiJunLi joined #git
19:16 jnavila joined #git
19:16 osse here got rev-parse commit: also works
19:16 Gustavo6046 joined #git
19:16 osse :/ doesn't and not :. either
19:17 dreiss joined #git
19:18 avar that gives you the commit, but not the tree it points to
19:18 BlackMaria joined #git
19:18 osse not here
19:18 durham joined #git
19:19 avar huh? You get different results (another commit) for `git --no-pager log --oneline -1 HEAD` & `git rev-parse HEAD` ?
19:19 avar what git version is this?
19:19 osse the colon is important
19:19 osse http://sprunge.us/QjFX
19:19 qqx `git rev-parse commit:` gives me the tree id as well, even with a pretty ancient git.
19:20 osse but I guess ^{tree} is the proper way
19:20 avar Ah, I misunderstood..
19:20 grayjoc joined #git
19:21 NeverDie joined #git
19:21 avar "<rev>:<path>, e.g. HEAD:README, :README, master:./README"
19:21 qqx avar: Basically just another case of the same method you found for getting the SHA for a blob.
19:21 avar via git help rev-parse
19:21 avar Yeah makes sense.
19:21 avar I'm hacking on my patch to add a `git {branch,tag,for-each-ref} --no-contains <commit>`
19:22 max3 joined #git
19:23 blackwind_123 joined #git
19:23 avar So you can do e.g. `git branch --contains v1.0 --no-contains v2.0' to find all branches that forked off from dev between 1.0..2.0
19:23 avar or `git tag --no-contains <commit>` to find releases without buggy <commit>
19:23 theoceaniscool joined #git
19:25 Tobbi joined #git
19:26 pwndave joined #git
19:27 kpease joined #git
19:27 UTAN_dev joined #git
19:28 durham joined #git
19:28 Cabanossi joined #git
19:29 Sasazuka joined #git
19:29 snowalpaca joined #git
19:34 ejb joined #git
19:34 bernardio joined #git
19:36 pfrench joined #git
19:39 oyemohit joined #git
19:42 thebird joined #git
19:42 irqq_ joined #git
19:43 lowercaseman joined #git
19:44 bvcosta joined #git
19:46 tang^ joined #git
19:48 UTAN_dev joined #git
19:50 SourTrout joined #git
19:50 dglambert can I git mergetool one file at a time
19:50 knod joined #git
19:51 dglambert I have 5 files listed, but last time it only let me manually merge two
19:51 dglambert supposedly automatically did the other 3
19:51 knod If I push to a repo with a file 'myfile.js' or whatever and, after that, and 'myfile.js' to the .gitignore, will the file be removed from the remote repo, or will it just never be updated?
19:52 moritz knod: neither
19:52 moritz knod: .gitignore only applies to untracked files
19:52 Sample joined #git
19:53 moritz knod: so if the file is already under version control, the .gitignore entry won't have any effect
19:53 knod Aw, dang. So I have to get rid of the file, then .gitignore it and /then/ it won't update? @moritz
19:53 jimi_ joined #git
19:53 knod Well, won't be added
19:53 moritz right
19:53 knod Thanks moritz, exactly what I needed to know
19:54 UTAN_dev joined #git
19:54 relipse joined #git
19:55 relipse I have a file composer.lock that differs from master, how do I grab the one from master and replace the modified one on another branch
19:55 manuelschneid3r joined #git
19:57 Murii joined #git
19:58 clemf joined #git
19:58 MattMaker joined #git
19:58 Murii joined #git
19:59 Dougie187 left #git
19:59 Murii joined #git
19:59 hashtagstrashtag joined #git
20:02 absinthe joined #git
20:03 qqx relipse: git checkout master composer.lock
20:03 pluszak joined #git
20:04 WeiJunLi joined #git
20:04 dviola joined #git
20:04 relipse qqx: i tried that but git status shows no differences
20:07 hashtagstrashtag left #git
20:07 UTAN_dev joined #git
20:07 hashtagstrashtag joined #git
20:09 hashtagstrashtag Hey all - Is it possible to pass arguments to a filter from .gitattributes? I'm attempting to handle specific subdirectories in a custom way, but don't want to duplicate the scripts/definitions
20:09 hashtagstrashtag Or, is it possible to get the .gitattributes path from a filter so I can use it's path for relative lookups
20:10 knod left #git
20:10 Murii joined #git
20:11 Murii joined #git
20:14 PHPanos joined #git
20:15 Vampire0_ joined #git
20:18 mischat joined #git
20:19 ccolorado joined #git
20:20 JanC_ joined #git
20:23 oyemohit joined #git
20:24 Seveas joined #git
20:27 marianina8 joined #git
20:27 Seveas_ joined #git
20:28 lucido-cl joined #git
20:28 govg joined #git
20:30 MattMaker joined #git
20:31 Thorn joined #git
20:31 alexandre9099 joined #git
20:32 jedahan joined #git
20:33 Seveas joined #git
20:34 UTAN_dev joined #git
20:34 kexmex joined #git
20:35 relipse what is the problem when master is different than origin/master ?
20:35 Seveas that is not a problem
20:35 aw1 joined #git
20:35 relipse it is a problem because I thought master git pull would pull everything from origin/master
20:36 osse it does
20:36 osse pull means merge
20:36 Seveas git pull does two things: fetch all commits from origin, and merge origin/master
20:36 Seveas so if you have local commits, master will still be ahead of origin/master
20:36 relipse then why did a whole lot change when I did git merge origin/master bu tnothing changed when i did git checkout master and git pull and git checkout otherbranch and git merge master
20:37 Seveas because those are two entirely different commands?
20:37 relipse no first id id git merge master second I did git merge origin/master
20:37 relipse and origin/master pulled in a lot more stuff
20:37 WayToDoor joined #git
20:40 whaley joined #git
20:41 theoceaniscool joined #git
20:43 cdown joined #git
20:43 rominronin joined #git
20:47 nowhere_man joined #git
20:47 ocelotsloth joined #git
20:47 apotry joined #git
20:49 Thorn left #git
20:49 rwp relipse, If you have local changes committed to master then you would need to push those to origin in order for them to appear in origin/master.
20:50 dglambert Hey
20:50 dglambert So I just did a merge, and went through using kdiff3 to deal with conflicts
20:50 rwp relipse, But it scares me that you don't know what changes are on the different brances. Pushing in those conditions would be like using a chainsaw to carve a turkey while blindfolded.
20:50 dglambert I don't like the way I handled one of the files and I want to go back and re-compare
20:50 dglambert can I git checkout just that file
20:51 dglambert then re-merge
20:51 dglambert get the conflict again
20:51 dglambert then open it in kdiff3?
20:51 aidalgol joined #git
20:51 dglambert sorry, I could have probably put that in one or two sentences
20:52 dglambert I pretty much want to revert to the unstaged version of the file waiting to be resolved
20:53 luke_nukem joined #git
20:54 Noldorin joined #git
20:54 UTAN_dev joined #git
20:54 cbreak dglambert: sure
20:54 Rodya_ joined #git
20:54 cbreak dglambert: man git-checkout
20:54 gitinfo dglambert: the git-checkout manpage is available at http://jk.gs/git-checkout.html
20:54 cbreak dglambert: look for -m
20:55 cbreak there's also --conflight=XXX
20:55 dglambert I'm more comfortable with the checkout portion, now I am thinking this through
20:55 dglambert my merge is still staged
20:55 ocelotsloth joined #git
20:55 dglambert if I revert the one file i want, then remerge, will everything in my Working Dir get overwritten
20:55 pfrench joined #git
20:55 cbreak reverting here makes zero sense
20:56 cbreak just do a git checkout -m on the file
20:56 geek2nerd joined #git
20:56 frankkahle joined #git
20:56 WeiJunLi joined #git
20:56 Sasazuka_ joined #git
20:57 dglambert well wont it be trying to merge the REMOTE and the LOCAL which is now some frankenstein file I tried merging already
20:58 dglambert VS if I 'revert' the local file back to the last commit then do the checkout -m on the file
20:58 dglambert it will be the REMOTE and LOCAL it merged against when I originally merged
20:58 dglambert make sense?
20:58 cbreak no
20:58 dglambert so file A
20:58 cbreak just do a git checkout -m on the file
20:59 frankkahle Good afternoon, so i setup an ubuntu server and have git installed, and gitlist, and i can see the repositories.  do i need to have a daemon running on the server for people to check in and out of?
20:59 cbreak that will give you the file as it was at the start of the merge attempt
20:59 anjen joined #git
20:59 dglambert got it
20:59 cbreak before you changed it in any way
20:59 dglambert so it ignores what ever i have done to it so far
20:59 dglambert that was my question
20:59 cbreak frankkahle: git does not need a running server
20:59 cbreak frankkahle: just ssh access
21:00 frankkahle ok cool
21:00 cbreak dglambert: no, it doesn't ignore it, it replaces it
21:00 frankkahle I'm the IT guy trying to set thisup not a developer
21:00 Eugene frankkahle - !gitolite or one of the !host_gui packages is probably what you want
21:00 gitinfo frankkahle: Gitolite is a tool to host git repos on a server. It features fine-grained access control, custom hooks, and can be installed without root. Download: https://github.com/sitaramc/gitolite Docs: http://gitolite.com/gitolite/ Quick example: http://gitolite.com/gitolite/overview/#basic-use-case
21:00 gitinfo frankkahle: There are several options for self-hosting git repositories with a web GUI: !gitblit !gitlab !rhode_code !gerrit and !gogs. Keep in mind that these all take far more work than a (non-GUI) !gitolite install. An overview of nearly all GUIs is available at: https://git.wiki.kernel.org/index.php/Interfaces,_frontends,_and_tools
21:00 UTAN_dev joined #git
21:00 cbreak frankkahle: well, if you want permissions, maybe an additional layer would be better
21:00 Eugene Using bare git by itself on a server is not particularly user-friendly. We use(and I recommend) gitblit for my dayjob
21:00 cbreak git on its own doesn't deal with that
21:01 cbreak but gitolite for example does
21:01 cbreak we used to use gitolite before switching to bitbucket
21:01 cbreak gitlab also works I think
21:04 snowalpaca joined #git
21:04 snowalpaca joined #git
21:05 dsdeiz joined #git
21:05 dsdeiz joined #git
21:05 navidr joined #git
21:06 ahr3n joined #git
21:06 fstd_ joined #git
21:08 _ADN_ joined #git
21:14 dglambert cbreak git checkout -m MainBranch --A.File
21:14 dglambert So i want to merge A.File from MainBranch back into my FeatureBranch
21:15 jimklo joined #git
21:16 cdown_ joined #git
21:19 mischat joined #git
21:19 askb joined #git
21:20 UTAN_dev joined #git
21:21 guampa joined #git
21:21 mda1 joined #git
21:24 lmat I'm trying to make a shared directory for several people to use on a server.
21:24 lmat They're not planning to use it remotely.
21:25 lmat It's configuration, sort of like version controlling /etc, but for a web application configuration files for various test environment configurations.
21:25 lmat Anyway, I'm setting one up somewhere else as a trial.
21:25 lmat git init --shared=group .;
21:25 chent joined #git
21:25 alexandre9099_ joined #git
21:26 ocelotsloth joined #git
21:26 lmat (as user1) add a file, commit.  as user 2, change file, can't commit. error: fatal: could not open '.git/COMMIT_EDITMSG': Permission denied
21:26 dglambert that seemed to work cbreak, I just had to swap the -m and --, git checkout -m MainBranch -- A.file
21:26 lmat The COMMIT_EDITMSG is owned by user1. I did   chgrp -R . <common group>;  and the error persists.
21:26 dvaske joined #git
21:26 lmat What am I doing wrong? This is how it's supposed to work, right?
21:27 ujjain joined #git
21:27 ujjain joined #git
21:27 einnjo joined #git
21:27 osse --shared is for bare repos so that more than one person can push to it
21:28 lmat osse: What should I do in my situation?
21:28 osse i think you should give up. it will be very painful to use the same repo as more than one person
21:28 osse alternatively, make them use the same user
21:28 lmat osse: ouch
21:29 osse that's what she said
21:29 lmat osse: That's what we're doing on the "test" server where that's possible. We don't have access to a common user in production...perhaps we should petition for that.
21:30 osse what is the repo used for?
21:30 chent left #git
21:32 lmat osse: Configuration properties for a web application.
21:32 osse lmat: and the files that live there will be the very same physical files that are read by this app ?
21:33 osse i mean if you had it your way?
21:33 lmat osse: exactly.
21:33 PCatinean joined #git
21:33 lmat osse: They're read when the application is moved into position and placed onto the servers whence they are served.
21:34 lmat osse: (they're not read while the application is running.)
21:34 osse lmat: I suggest a bare repo somewhere, with a hook that installs the files into place on each push
21:34 osse lmat: sort of like what people do for web sites
21:34 osse lmat: !deploy
21:34 gitinfo lmat: 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
21:34 hahuang61 joined #git
21:34 lmat osse: yeah, good idea.
21:36 Atm0spher1c joined #git
21:39 TSS_ joined #git
21:39 qqx Huh, just did some testing. And I was able to get a shared non-bare repo working by having the directory setgid and owned by the correct group before running `git init`. Still seems like a bad idea though.
21:41 UTAN_dev joined #git
21:41 alzagros joined #git
21:41 bgerber joined #git
21:41 Jupelius joined #git
21:42 lmat qqx: Why a bad idea?
21:42 qqx Too likely for changes by multiple people to get grouped together into a single commit.
21:43 lmat qqx: What is directory setgid?
21:43 lmat is that different from chgrp and chmod g+w?
21:43 qqx chmod g+x
21:44 qqx Correction chmod g+s
21:44 lmat qqx: That just allows the group to list files in that dir?
21:44 lmat Ohh, different ^_^
21:44 ruxu joined #git
21:44 theoceaniscool joined #git
21:44 lmat qqx: What does it do?
21:44 qqx Causes entries created within it to inherit the group ownership.
21:44 synthroid joined #git
21:45 lmat okay...sounds sticky ^_^
21:45 DammitJim joined #git
21:45 DammitJim what is the best way for me to clone a repo from one server to another and then change the name of the repo?
21:45 qqx And directories created within it also inherit the setgid.
21:45 DammitJim gosh, people are too lazy these days
21:46 lmat sheesh... I'm having trouble with my groups. I added myself to a group, then   ssh localhost;  id; shows that I'm in the group.
21:46 nowhere_man joined #git
21:46 lmat DammitJim: repos don't have names.
21:46 DammitJim lmat, so that's just a folder name?
21:46 lmat When I run tmux; then id;  I'm not in the group anymore :-(
21:46 lmat DammitJim: then use mv like with any folder.
21:46 DammitJim lmat, thanks
21:46 qqx lmat: You need to restart the tmux server.
21:46 lmat git clone https://whatever  wherever;  mv wherever somewhere;
21:47 lmat qqx: Oh yeah, forgot about that daemon stuff.
21:47 DammitJim so, if my repo on my test server is in: /srv/git/test_learn
21:47 lmat qqx: ahh, like butter.
21:47 lmat DammitJim: gti clone test_server:/srv/git/test_learn  newname;
21:47 lmat DammitJim: git clone test_server:/srv/git/test_learn  newname;
21:48 UTAN_dev joined #git
21:48 DammitJim thanks lmat... can I do that with ssh vs https?
21:48 lmat DammitJim: yup. "Whatever" ;-)
21:48 tmg joined #git
21:48 DammitJim oh, that's what you meant LOL
21:51 lmat qqx: When user1 creates a new file in the wc, it is   user1:user1   Any magic for that?
21:51 lmat ...not the water closet...
21:51 ruxu joined #git
21:51 lmat qqx: nm, that's not a big deal.
21:51 DammitJim cool. you are the best!
21:52 lmat qqx: But a big deal *is* COMMIT_EDITMSG is rw-r--r-- user1:user1.
21:52 TSS_ joined #git
21:52 lmat qqx: Maybe it's a one-time fix?
21:53 lmat qqx: Yeah. Now that the permissions are right, user1 and user2 can commit as they like. When they create new files, they'll need to make sure to g+w
21:54 UTAN_dev joined #git
21:54 ToBeCloud joined #git
21:54 qqx lmat: What's your umask?
21:55 qqx For shared editing to work, everyone's umask should be no more restrictive than 007.
21:55 qqx Which could be a security issue for other files if everyone doesn't have a private group as their primary group.
21:56 lmat qqx: Thanks for the tip.
21:57 lmat 0022 of course.
21:57 DammitJim left #git
21:57 s2013 joined #git
21:58 lmat so it's missing the g+w perm as expected.
22:00 anuxivm joined #git
22:01 dandaman joined #git
22:01 dodgems joined #git
22:04 DammitJim joined #git
22:04 DammitJim I did the git clone
22:04 Doow joined #git
22:04 jeffreylevesque joined #git
22:04 keepguessing joined #git
22:04 DammitJim can I start using my new git clone as a normal git repository?
22:05 moei joined #git
22:05 qqx DammitJim: it is a normal git repository.
22:06 Hello71 what other kinds are there
22:06 keepguessing I have done a git merge. When I do git log I find a lot of git commits now that came from the merged branch. How do I filter that out?
22:06 keepguessing DammitJim: it is a normal git repo
22:07 qqx keepguessing: git log --first-parent
22:07 Derperperd joined #git
22:07 DammitJim I didn't know if one has to give access or what not
22:07 t0by joined #git
22:07 t0by joined #git
22:07 DammitJim or when one does git -init or something like that
22:07 WayToDoor joined #git
22:07 UTAN_dev joined #git
22:08 qqx `git init` is for creating a new repository. clone actually does that as part of the job.
22:08 keepguessing DammitJim: when you git clone you have the access.
22:08 ertes joined #git
22:08 DammitJim thanks
22:09 ahr3n joined #git
22:09 keepguessing qqx: How do I see which ref the branch was created?
22:10 qqx keepguessing: I don't understand the question.
22:10 thiago joined #git
22:10 thiago joined #git
22:12 keepguessing qqx: I want to see the first commit of the branch.
22:12 bgerber joined #git
22:13 Cabanossi joined #git
22:14 raijin joined #git
22:14 UTAN_dev joined #git
22:14 DammitJim left #git
22:15 rominronin joined #git
22:16 avar git log --reverse <branch>
22:20 ISmithers joined #git
22:20 mischat joined #git
22:21 UTAN_dev joined #git
22:21 tvw joined #git
22:23 geek2ne79 joined #git
22:24 ldleworker joined #git
22:24 dsdeiz joined #git
22:27 hvn0413 joined #git
22:27 TSS_ joined #git
22:30 mrslaughter1775 joined #git
22:32 t-mart joined #git
22:32 snowalpaca joined #git
22:35 foul_owl I'm trying to see what branch jenkins is using (for debugging, I don't think it's checking out the right branch) I'm having jenkins run: "git branch" but it's printing "detached from <hash>" how do I actually see what brnach it is using?
22:35 phillco joined #git
22:35 osse foul_owl: git branch --contains hash
22:35 geek2nerd joined #git
22:35 VladGh joined #git
22:35 osse you can probably do that locally
22:35 osse a hash is a hash
22:36 foul_owl Ah ok! I didn't know how to tell what branch a hash belongs to
22:36 foul_owl THank you!
22:36 osse \o/
22:39 NeverDie joined #git
22:43 gopar joined #git
22:43 askb joined #git
22:47 cliluw joined #git
22:51 jimklo joined #git
22:52 ocelotsloth joined #git
22:53 murii_ joined #git
22:54 markelite joined #git
22:55 jimklo joined #git
22:56 pfrench joined #git
22:56 a_thakur joined #git
22:57 geek2ne16 joined #git
22:57 dvaske joined #git
23:01 tmsmith joined #git
23:07 Tacoder joined #git
23:09 jeffreylevesque joined #git
23:09 geek2nerd joined #git
23:14 UTAN_dev joined #git
23:16 jimklo_ joined #git
23:20 UTAN_dev joined #git
23:21 geek2nerd joined #git
23:21 hahuang65 joined #git
23:21 Kaisyu joined #git
23:22 mischat joined #git
23:22 om_henners joined #git
23:27 Gsham joined #git
23:27 Sasazuka_ joined #git
23:29 halftroll joined #git
23:30 dsdeiz joined #git
23:30 dsdeiz joined #git
23:30 ejb joined #git
23:31 halftroll Hello.. I did commit my changes, pull, merge.. and out of mistake I did a bad merge so I wanted to go back before the pull and did git reset --hard.. well I went farther than what I needed and lost a day of work
23:31 halftroll I wonder if doing reset --hard i lost the commit that I did before doing pull & merge
23:31 duderono_ joined #git
23:33 Hello71 !fixup
23:33 gitinfo So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!
23:33 rewt joined #git
23:33 geek2nerd joined #git
23:35 halftroll thanks
23:39 dviola joined #git
23:40 fatalhalt joined #git
23:44 PCatinean joined #git
23:46 WayToDoor joined #git
23:46 geek2nerd joined #git
23:47 rominronin joined #git
23:47 raijin joined #git
23:48 ResidentBiscuit joined #git
23:50 PCatinean joined #git
23:51 dtrott joined #git
23:52 halftroll in case someone is interested, what I did was git reflog then git checkout <where I wanted to go>
23:57 hvn0413 left #git
23:58 geek2nerd joined #git
23:59 justanotheruser joined #git

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