Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-03-23

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

All times shown according to UTC.

Time Nick Message
00:01 sespiros joined #git
00:02 Cabanossi joined #git
00:07 jeffreylevesque joined #git
00:08 royiv joined #git
00:08 royiv How do I fetch a remote's HEAD?
00:09 SmashingX joined #git
00:09 t0by joined #git
00:11 dtrott Not sure what your asking, why does:    git fetch REMOTE_NAME not work for you ?
00:11 evanextreme joined #git
00:12 a3Dman joined #git
00:12 royiv Nope
00:13 royiv remotes have a "HEAD"; if you `git branch -r`, for each remote $remote you should see $remote/HEAD
00:13 royiv when you do a non-bare clone, the remote's HEAD becomes the default branch
00:13 royiv *becomes the first non-remote branch that's initially checked out
00:14 sespiros joined #git
00:14 royiv One would naively think that `git fetch` ought to have an option to re-fetch this thing, but I can't for the life of me find it
00:15 royiv I could reclone the repo into a temp dir and copy .git/refs/remotes/origin/HEAD from the temp clone to my clone, but feels excessively hacky
00:15 bremner git ls-remote foo | grep HEAD
00:15 royiv GitHub exposes the remotes HEAD in the settings pane under "default branch" (or something. my org hasn't given me permissions to see or update that pane)
00:16 bremner hmm. ls-remote is not symbolic, that's a bit unhelpful.
00:16 royiv bremner: yeah, ls-remote sees it, but it shows me the deref'd hash. Normally it's symbolic.
00:16 bremner I mean, you can guess which branch it is by matching the hash.
00:17 cagomez joined #git
00:17 foreal joined #git
00:18 royiv I can also read .git/refs/yada/yada on a temp clone and not guess at all.
00:18 royiv I'm starting to think this is a bug in fetch.
00:20 bremner because it doesn't provide the same functionality as clone?
00:21 dtrott royiv: I believe the answer here answers your question: http://stackoverflow.com/questions/18726037/what-determines-default-branch-after-git-clone
00:22 royiv bremner: well, I mean, it seems to keep a remote ref for it, so wouldn't one think one would want to keep said ref up-to-date? I mean, I can reference it, but it's presently *wrong*, and not really correctible.
00:23 bremner i see.
00:23 royiv dtrott: oh, that's interesting. `git remote show` gets HEAD correct. so it must connect then to see?
00:24 Nilesh_ joined #git
00:25 ozmage joined #git
00:25 Kaisyu joined #git
00:25 Muzer joined #git
00:25 ozmage_ joined #git
00:26 Fallen022317 joined #git
00:27 hashtagstrashtag joined #git
00:27 bejeb joined #git
00:27 Rotaerk joined #git
00:27 ozmage__ joined #git
00:28 royiv It looks like it should be `git remote set-head --auto origin`; its output even claims "origin/HEAD set to develop", which would be correct, but it doesn't actually set origin/HEAD
00:29 jeffreylevesque joined #git
00:29 royiv why it's on `git-remote` and not `git-fetch` is a bit weird
00:29 Vampire0_ joined #git
00:29 jeffreylevesque_ joined #git
00:29 royiv `git fetch` also isn't fetching all the branches
00:30 royiv and it doesn't have origin/develop, whcih might be why trying to set origin/HEAD to it doesn't work
00:32 NeverDie joined #git
00:32 Cabanossi joined #git
00:33 ntnn joined #git
00:35 strayArch joined #git
00:36 matoro joined #git
00:36 roelmonnens joined #git
00:37 strayArch when solving conflicts for pr to another branch, what's the proper way to handle? Merge the target branch into the pr?
00:38 jeffreylevesque joined #git
00:39 brent__ joined #git
00:41 royiv Finally got it. Lord does git make this painful. The root of my problem was that the repo was shallow, so I needed:
00:42 royiv git fetch --unshallow ; git config remote.origin.fetch '+refs/heads/*:refs/remotes/origin/*' ; git fetch ; git remote set-head --auto origin
00:42 Vampire0 strayArch, depends on how you want the history to look like. I usually prefer doing a rebase in this situation
00:43 Vampire0 royiv, shouldn't the last part be enough?
00:43 brent__ joined #git
00:43 Vampire0 royiv, while I also would have thought fetch will update it
00:43 Vampire0 royiv, you should request this on the mailing list
00:44 netj joined #git
00:44 royiv Me too; the last command is insufficient; running it alone output:
00:44 royiv error: Not a valid ref: refs/remotes/origin/develop
00:44 royiv origin/HEAD set to develop
00:44 royiv but it doesn't actually set origin/HEAD to develop
00:45 royiv (which seems to be a bug; I assume it doesn't want to b/c no such remote branch exists w/ that name)
00:45 a_thakur joined #git
00:47 a_thakur_ joined #git
00:47 Vampire0 royiv, ah, I see, your refspec didn't include the develop branch, then it cannot be set of course, yeah
00:48 royiv "my refspec" <- this is part of what I find really surprising: git-clone changes what branches it tracks/fetches based on what depth you ask it to clone to
00:48 SmashingX left #git
00:48 royiv those two things are not causally related to me: that I ask to make a shallow clone doesn't (to me) imply a different set of refs to fetch.
00:48 royiv (just a different depth to fetch them to)
00:49 MacWinner joined #git
00:49 royiv but clone's behavior seems to be to switch the default from all refs to just the HEAD ref if you specify a clone depth
00:49 Vampire0 royiv, btw. from a very quick look of a non-C-developer to the Git code it seems Git is doing exactly what you found out. Query the refs like ls-remote and then guess from the matching SHA values
00:49 royiv and then git-remote just can't cope with shallow repos.
00:49 Geo joined #git
00:50 Vampire0 royiv, if you e. g. have two branches that point to the same commit and HEAD points to one of them, remote set-head will tell you that it couldn't guess and you should select which one is the right one
00:50 royiv My googling indicated that *old* versions of git tried to deduce the ref from the SHAs, but that newer gits actually fetched them.
00:50 Geo Hi, looking for some advice on a merge strategy for http://pastebin.com/raw/jpctvmUr
00:50 Vampire0 royiv, so even Git itself is only guessing about the HEAD in remote
00:50 Geo is it possible to merge I-J-K as a FF merge into stable/1.8 ?
00:51 Levex joined #git
00:53 Vampire0 royiv, well, I looked at the sources of 2.12.1, but I might have misinterpreted and I tested on 1.7.10.4
00:53 strayArch Vampire0, for now it's saying already up-to date although, bitbucket says that there is a conflict but trying to pull destination into source doesn't do anything
00:54 royiv The googling seemed to say it changed in the 1.8.x line
00:54 royiv Actually, I think it might have been the link mentioned above
00:54 strayArch I'll google your way for now, thanks Vampire0
00:54 btree joined #git
00:54 royiv http://stackoverflow.com/questions/18726037/what-determines-default-branch-after-git-clone
00:54 royiv that one
00:54 Vampire0 Geo, sure, just do a rebase first like `git rebase --onto stable/1.8 develop...release/1.8.1` I think
00:55 Vampire0 strayArch, that will not be any different. Whether you merge or rebase, only the result looks differently. If merge says up-to-date so will rebase.
00:56 Vampire0 strayArch, did you really try to merge from the upstream branch, or did you accidentally try to pull from your forks branch?
00:57 Geo Vampire0 wow, all new commands for me to look at :) thank you, I'll read up
00:59 Vampire0 royiv, well, then I probably misinterpreted the code and my try-out was not helpful as it was with an ancient version :-)
00:59 Geo Vampire0, if someone was already working off of that stable branch, would that rebase break existing checkouts?
01:00 Vampire0 Geo, no, the stable branch will not change history. Your release branch will though. So if someone based work on top of it, he will need to rebase similarly to what you need to do to actually move it in the first place. See !rewrite
01:00 gitinfo Geo: 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
01:01 Vampire0 Geo, you just move commits on top of the stable branch like if you would create new commits
01:01 Vampire0 Geo, so pretty daily business
01:01 Vampire0 Geo, you just move commits on top of the stable branch like if you would create new commits
01:02 Vampire0 ups, ignore the last message
01:03 causasui joined #git
01:03 Geo right, but if F and I in that diagram conflict.. that doesn't allow it
01:03 Geo ?
01:05 Vampire0 royiv, yeah, I repeated my test with a newer Git and it works properly. So the code I have seen must be for the case that the newer Git client is talking to an older Git server that does not transmit the information
01:05 grayjoc joined #git
01:05 Geo (I think your example reverses what I intended, but I get the gist of it)
01:05 Vampire0 Geo, if it reverses what you intended, maybe you stated your question in reverse?
01:06 shinnya joined #git
01:07 Vampire0 Geo you said you want to ff-merge release/1.8.1 into stable/1.8, for that release/1.8.1 must be on top of stable/1.8 and what I told you does this. It takes the commits in release/1.8.1 but not in develop and applies them on top of stable/1.8
01:07 Vampire0 Geo, if there is a conflict you have the same situation for the rebase as you have for a merge. Git will pause and tell you, waiting for you to resolve the conflict and tell it to continue
01:07 cagomez joined #git
01:08 Geo so, stupid question- if I have to resolve conflicts, doesn't that prevent me from having a ff-merge?
01:09 strayArch Vampire0, it looks like atlassian only supports git 1.8.5 and i'm using 2.12 so I'm guessing the conflict is unique to that git version ....
01:09 Vampire0 strayArch, this action has nothing to do with what Atlassian supports
01:10 Vampire0 strayArch, you fetch the upstream you do a local rebase of your branch and you do a force-push. All actions BitBucket should support
01:10 Aaron1011 joined #git
01:10 Vampire0 strayArch, if it tells you up-to-date but it is not, then you probably did not fetch or try to rebase/merge against the wrong remote
01:11 Aaron1011 Is this kind of mailmap line supported? https://github.com/rust-lang/rust/commit/418ef78c5c1eeacdd955bdd118e0e72191ae7f92#diff-c9d540715cff3469b65ddd01f614848bR136
01:11 wizardedit joined #git
01:11 Aaron1011 the documentation doesn't mention specifying multiple name/email pairs on a single line
01:11 Vampire0 Geo, no, you do a rebase first, resolving the conflicts. After you did the rebase, there are no conflicts anymore and you are on-top, so you can do a ff-merge and get linear history
01:12 Some_Person joined #git
01:12 PaulCape_ joined #git
01:12 xaviergmail joined #git
01:12 Geo Vampire0, great info, thank you
01:12 Geo I'll play a little more to get the hang of it
01:16 l4v2 joined #git
01:16 strayArch Vampire0, It supports the actions but merging doesn't result in conflict
01:18 Lyqyd joined #git
01:18 Some_Person left #git
01:19 gugah joined #git
01:19 dsdeiz_ joined #git
01:22 ShapeShifter499 joined #git
01:22 eikhorst joined #git
01:26 Vampire0 Aaron1011, I think it is allowed, but the third pair is simply ignored
01:26 Aaron1011 yeah
01:26 Aaron1011 that seems to be how read_mailmap_line handles it
01:26 Vampire0 strayArch, you didn't say it does not give you a conflict, but you said it says up-to-date, didn't you?
01:26 Aaron1011 I also tested it by creates a new commit
01:27 Aaron1011 anything after the second email is simply ignored
01:30 Vampire0 Aaron1011, yep, I also had a look at the code :-)
01:31 jmesmon joined #git
01:31 Geo Vampire0: invalid upstream develop...release/1.8.1
01:32 Geo from what you gave me earlier
01:32 strayArch Vampire0, correct, bitbucket claims there is a conflict though. I even deleted the repo and merged into source into target w/ no conflict, but bitbucket claims there is..
01:32 Vampire0 Geo, ah, then make it `git rebase --onto stable/1.8 E release/1.8.1`
01:33 Geo E?
01:33 Vampire0 Geo, or maybe `git rebase --onto stable/1.8 $(git merge-base develop release/1.8.1) release/1.8.1`
01:33 Vampire0 Geo, E from your ascii art
01:33 Geo ahh
01:34 Vampire0 Geo or use the last command that should determine E if you didn't do a merge from develop to release/1.8.1 in-between
01:34 rchavik joined #git
01:34 Vampire0 strayArch, !repr
01:34 gitinfo strayArch: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
01:36 raijin joined #git
01:38 ankit01ojha joined #git
01:39 Emperor_Earth joined #git
01:42 juvenal joined #git
01:43 gopar joined #git
01:45 pks joined #git
01:48 diogenese joined #git
01:48 Geo Vampire0, it looks like this is exactly what I needed, thank you very much
01:49 Geo I never would have gotten here
01:49 ecuanaso joined #git
01:54 eikhorst joined #git
01:55 livingstn joined #git
01:55 Gsham joined #git
01:59 jameser joined #git
02:01 arescorpio joined #git
02:03 a_thakur joined #git
02:03 a_thakur joined #git
02:04 Vampire0 Aaah, freshly built 2.12.1 now on this system, nice :-)
02:10 MarioBranco joined #git
02:16 strayArch Vampire0, it ended up being that merging by branch did not cause conflict, but merging by hash did. Which may be how bit bucket checks for conflicts
02:16 NightMonkey joined #git
02:17 d^sh joined #git
02:18 durham_ joined #git
02:20 roelmonnens joined #git
02:22 finalbeta joined #git
02:31 Muzer joined #git
02:32 Cabanossi joined #git
02:33 dsdeiz joined #git
02:37 Goplat joined #git
02:40 sunri5e joined #git
02:41 SmashingX joined #git
02:42 SmashingX I have a repo which in each subdirectory has a submodule when I go into each of the directories the submodules are empty
02:46 SmashingX why when I clone the repository it doesn’t get all the submodules as well?
02:46 SmashingX or what do I have to do to get the submodules?
02:46 rchavik i think you need to do git submodule --update --init
02:46 Eryn_1983_FL joined #git
02:46 fstd joined #git
02:46 durham joined #git
02:46 sin joined #git
02:47 khinashefrei joined #git
02:48 SmashingX rchavik: got it thanks
02:48 SmashingX every single submodule though is in detached head state, I have to go into every submodule and checkout master, do you know why?
02:49 Lyqyd joined #git
02:52 rchavik submodule notes the exact commit, not the branch.. thus it checks outs the registered commit in superproject
02:56 Vampire0 SmashingX, if you want to compose multiple repositories together in which you develop concurrently, submodules are not the right tool in my opinion. submodules are only useful if you integrate 3rd party code and from now to then update it to a newer version
02:56 ayogi joined #git
02:56 SmashingX rchavik: so I’m guessing since I don’t have the same branch in super project as in the modules they go to a detached head state?
02:57 SmashingX I should just then do the submodule init when superproject is in master.
02:58 SmashingX Vampire0: I have to stick with what my company is doing, but thanks for the suggestion
02:58 rchavik not really, you most likely already have the branch locally, it's just that git automatically checks out the specific commit from that branch
02:59 SmashingX ok so to fix that I just checkout all the modules to master then I do a commit from superproject?
02:59 rchavik you'll have to periodically update the submodule, and then tell the superproject to use the new commit
02:59 SmashingX and everytime I clone the main project they all will be in master?
03:00 SmashingX so how can I make all my submodules to point to master when I clone the main project?
03:00 rchavik you can't, you can only tell it to use a specific commit.
03:01 pks joined #git
03:02 Cabanossi joined #git
03:03 SmashingX rchavik: what?
03:03 SmashingX that’s annoying
03:04 a3Dman joined #git
03:04 a3Dman joined #git
03:04 SmashingX every time that I want to do submodule update I have to remember that I have to checkout master
03:05 mizu_no_oto joined #git
03:05 ojdo joined #git
03:05 MarioBranco joined #git
03:05 eclecticjohny joined #git
03:06 rchavik SmashingX, that's basically what Vampire0 said :)
03:07 ayogi joined #git
03:07 SmashingX so what do you guys recommend instead of submodules?
03:08 kadoban I use subtree (contrib command) fairly often
03:08 kadoban It better fits my desires than submodule most of the time
03:08 jeffreylevesque joined #git
03:09 SmashingX Vampire0: what is your suggestion to use then?
03:09 durham joined #git
03:10 SmashingX rchavik: ?
03:10 anuxivm left #git
03:12 Vortex35 joined #git
03:15 lagothri1 joined #git
03:16 Lyqyd joined #git
03:16 chachasmooth joined #git
03:17 thebinary joined #git
03:20 Lyqyd joined #git
03:20 thebinary @moongazer : 'git branch' without any arguments will list branches
03:26 kodo[m] joined #git
03:32 Cabanossi joined #git
03:32 sbulage joined #git
03:40 Vampire0 SmashTheStack, !subre
03:40 gitinfo [!subprojects] So, you want to add git repositories inside of other git repositories? Well, you have four main options. First is to just do it, add the repo to the outer project's .gitignore, and treat them entirely separately. Best if they are entirely separate. Otherwise your best options are "!submodule", "!gitslave", and "!subtree". Try those commands in this channel, or in a PM to avoid flooding.
03:41 Derperperd joined #git
03:42 NightMonkey joined #git
03:43 khinashefrei a
03:44 khinashefrei can anyone help me resolve a gerrit/git situation? I'm trying to understand exactly how patch sets and commits are related but I can't for the life of me figure out how to solve it. for a detailed explanation I put it here: http://pastebin.com/EXp4tERA.  tldr version is that I'm trying to append a patch set to be reviewed in gerrit, but made my cha
03:44 khinashefrei nges in a local repository, so I want to append the commits since that patch from my local to the gerrit repo, but have no idea how to.
03:45 hexagoxel joined #git
03:45 khinashefrei i tried some things like adding the original repo as a remote but it copies all of the commits since the git init
03:45 rchavik joined #git
03:46 davimore joined #git
03:49 SimpleName joined #git
03:50 SimpleName I already add keys to /home/git/.ssh/authorized_keys, but when I push code to it, still need to input password
03:50 acetakwas joined #git
03:52 SimpleName how to look sshd log information
03:53 Derperperd joined #git
03:54 Derperperd joined #git
03:56 mattcen joined #git
03:57 evanextreme joined #git
03:57 finalbeta1 joined #git
03:58 ploop joined #git
03:59 a_thakur joined #git
04:00 a_thakur joined #git
04:02 Cabanossi joined #git
04:02 Derperperd joined #git
04:03 Atm0spher1c joined #git
04:03 learning joined #git
04:03 roelmonnens joined #git
04:06 evanextreme joined #git
04:08 Derperperd joined #git
04:12 raspado joined #git
04:12 fakenerd joined #git
04:15 dsdeiz joined #git
04:15 dsdeiz joined #git
04:15 miczac joined #git
04:17 netj joined #git
04:18 mehola joined #git
04:19 svm_invictvs joined #git
04:21 SimpleName why can’t clone the code in local server
04:22 SimpleName git server and git client in same machine
04:23 b4rbz joined #git
04:24 elinux joined #git
04:24 Derperperd joined #git
04:25 leprechau joined #git
04:28 Gsham joined #git
04:30 luke_nukem joined #git
04:31 robotroll joined #git
04:32 Cabanossi joined #git
04:33 bocaneri joined #git
04:35 menip joined #git
04:35 dsockwell joined #git
04:36 bocaneri joined #git
04:37 rewt joined #git
04:40 roentgen joined #git
04:40 thiago joined #git
04:41 evanextreme joined #git
04:41 mehola joined #git
04:42 learning joined #git
04:45 Cabanossi joined #git
04:46 kyan joined #git
04:46 MarioBranco joined #git
04:48 Vaelatern joined #git
04:52 Darren_ joined #git
04:52 hexagoxel joined #git
04:52 steeze joined #git
04:54 shgysk8zer0 joined #git
05:02 ayogi joined #git
05:04 venmx joined #git
05:07 zezba joined #git
05:11 khinashefrei anyone around?
05:14 stoned I am happy to use git autocorrection.
05:14 stoned I just typed  git bnanch
05:14 stoned hehe
05:17 lucido-cl joined #git
05:21 stoned Also, does anyone use gitlab? I created a public repository and now I want to make it public but I cant' seemto. I've changed the privacy settings to public, and disabled permissions (other option was team members only)
05:21 stoned khinashefrei: hello
05:22 dave0x6d joined #git
05:22 khinashefrei hi
05:23 khinashefrei ive used gitlab before. do you mean you have a private repo you want to make public?
05:23 zezba_ joined #git
05:24 sunrunner20 joined #git
05:25 venmx joined #git
05:26 Soni joined #git
05:26 l4v2 joined #git
05:26 evanextreme joined #git
05:27 SimpleName when I try git clone code in the git server, raise these error  https://dpaste.de/DLeW
05:30 zezba joined #git
05:31 deltab SimpleName: those two lines are in the wrong file: they belong in sshd_config, not ssh_config
05:31 Geo left #git
05:32 stoned khinashefrei: yessir
05:33 stoned I initially started a private repository. I need to share the code base with another, perhaps I should just add another remote to my repository and push there, to a new public repository.
05:33 stoned leave this one as it is. And when I need to show code to people, just push to the public remote
05:33 stoned perhaps I should do that, right?
05:33 stoned What do you ponder, sir?
05:35 SimpleName deltab: so this is the git bug?
05:35 deltab SimpleName: not, it's a problem with ssh configuration
05:35 SimpleName deltab: but I never config it or modify the file
05:36 Muzer joined #git
05:36 deltab what are lines 63 and 64 of /etc/shh/ssh_config?
05:36 deltab I mean /etc/ssh/ssh_config
05:37 khinashefrei @stoned well maybe just double check that the public settings are actually getting properly saved and that you have the privileges to do so first
05:37 stoned I want to understand remote more, so: git add remote_name_here linkto_repo
05:37 stoned Instead of messing w/ my priv repo, I have createad a new public gitlab repo, and now I want to add this as a remote to my other repository
05:37 stoned Trying to understand remote naming.
05:38 stoned Shall I say, git add testing new_remote_link
05:38 stoned instead of git add origin new_remote_link
05:38 stoned https://help.github.com/articles/about-remote-repositories/
05:38 stoned This is github or git or gitlab (this channel)?
05:39 stoned I'm not sure if I'm asking in the correct spot
05:39 stoned I wonder if I am using the correct docs
05:39 deltab there's a separate channel for github
05:39 stoned https://git-scm.com/book/en/v2/Git-Basics-Working-with-Remotes oh this is more detailed
05:40 _ikke_ stoned: a remote is just a shortcut to a url, you can name it however you want. origin is the default name git uses for the remote you cloned from
05:40 deltab but you can ask about git aspects here
05:40 stoned Ahh, thanks.
05:40 stoned i understand, so that's just the remote name.
05:40 stoned there is ony one remote called origin right now. I see. wicked.
05:41 SimpleName deltab: https://dpaste.de/hzV7
05:42 deltab SimpleName: yeah, those shouldn't be in ssh_config
05:43 _ikke_ It should be ServerAliveInterval for the client
05:43 _ikke_ and ServerAliveCountMax
05:43 deltab ls -l /etc/ssh/ssh_config  will show you when the file was last changed
05:44 brent__ joined #git
05:44 stoned Ahh manI like this git. nice
05:44 dsantiago joined #git
05:44 armyriad joined #git
05:48 diogenese joined #git
05:54 dvaske joined #git
05:55 venmx joined #git
05:57 MrcRjs joined #git
05:57 ogny joined #git
05:57 fhackdroid joined #git
05:59 stoned Ok, that was nice. Git is cool.
05:59 stoned I'm going to stone a bit now. See you stoned.
06:00 stoned See you guys.
06:04 learning joined #git
06:04 rscata joined #git
06:05 IndigoTiger joined #git
06:05 raatiniemi joined #git
06:06 dec0n joined #git
06:06 _cyril__ joined #git
06:06 pasky_ joined #git
06:07 snowinferno_ joined #git
06:08 askb_ joined #git
06:08 chamunks joined #git
06:08 cathyjf joined #git
06:08 BeerSerc_ joined #git
06:08 Ruler-_ joined #git
06:08 govg_ joined #git
06:09 heroux_ joined #git
06:09 bravespear|2 joined #git
06:09 Sp4rKy joined #git
06:10 mattcen_ joined #git
06:10 utrack_ joined #git
06:11 pkiller joined #git
06:12 strayArc- joined #git
06:12 Jguy| joined #git
06:12 cydrobolt_ joined #git
06:12 Aaeriele` joined #git
06:12 Sigma[Mtp] joined #git
06:13 th1nkpad joined #git
06:13 selckin1 joined #git
06:13 spriz joined #git
06:13 anbz joined #git
06:13 Darren_ joined #git
06:13 Exagone313 joined #git
06:13 Zialus joined #git
06:13 Vampire0 joined #git
06:14 _ikke_ You live up to your name
06:14 qpdb joined #git
06:14 tsunamie joined #git
06:15 murr4y joined #git
06:15 hue joined #git
06:19 overlord_tm joined #git
06:21 Skyfinn joined #git
06:24 d^sh_ joined #git
06:26 qt-x joined #git
06:26 acetakwas joined #git
06:30 hexagoxel joined #git
06:35 chigang joined #git
06:35 dhollinger joined #git
06:39 Shavik joined #git
06:39 hvn0413 joined #git
06:41 nostrora joined #git
06:42 nostrora joined #git
06:42 imack joined #git
06:45 Cabanossi joined #git
06:47 Silmarilion joined #git
06:48 dersand joined #git
06:49 Shavik Having some difficulties pulling a newly created branch. So I created a new branch on my dev machine and pushed it with a single commit to bitbucket. From there I fetched/pulled to our central repo store for our vpn. From there I'm trying to pull that new branch to a machine on the vpn. I can see the branch via 'git ls-remote origin' but when I fetch it, it's no where to be found via 'git branch -a' or 'git
06:49 Shavik branch -r' I'm not sure what else to try. running git 1.7.1. Appreciate any insight. Thanks!
06:50 TheWild joined #git
06:50 TheWild hello
06:50 Shavik I've also tried doing a 'git fetch origin branch-name:origin/branch-name. and hello :)
06:50 TheWild "fatal: Not a git repository (or any of the parent directories): .git" - is there any way to override repository directory without setting it to current directory?
06:51 Shavik But when I run that, it gives me fatal: Couldn't find remote ref branch-name even though I see it when I 'git ls-remote origin'
06:53 ozmage joined #git
06:55 acetakwas joined #git
06:55 Raging_Hog joined #git
07:02 hue- joined #git
07:03 thiago TheWild: why can't you just cd to the repository?
07:04 MarioBranco joined #git
07:04 Karazhan joined #git
07:04 sns joined #git
07:04 iamthad joined #git
07:04 jimInDevNull joined #git
07:04 Sp4rKy_ joined #git
07:04 abh joined #git
07:04 MineCoins joined #git
07:04 Zialus joined #git
07:05 hecliunyx joined #git
07:05 hanneso joined #git
07:05 ti2 TheWild: Does the '-C' option help? E.g. 'git -C ../path/to/repo ls-remote'?
07:05 c^ joined #git
07:05 thiago ls-remote doesn't need a repository
07:06 TheWild ti2: Yes! The -C option is exactly what I need. Thank you very much.
07:06 TooLmaN joined #git
07:06 thiago Shavik: what fetch command did you run?
07:07 Shavik So I think I might have figured it out
07:07 Shavik :/
07:07 Shavik in the ls-remote response, it listed refs/remotes/origin/branch-name
07:07 Shavik but not refs/heads/branch-name
07:07 Shavik I hadn't actually checked out the branch to my first server. it just knew of the remote branch so I couldn't pull it to the next machine in the chain :/
07:08 spriz joined #git
07:08 justinmrkva joined #git
07:08 inovas joined #git
07:09 thiago you can still fetch it
07:09 learning joined #git
07:09 Chaser joined #git
07:10 Lc0rE joined #git
07:10 dserodio joined #git
07:11 tvw joined #git
07:16 roelmonnens joined #git
07:16 roelmonnens joined #git
07:18 cagomez joined #git
07:19 Brski joined #git
07:22 basiclaser joined #git
07:23 zeroed joined #git
07:24 PCatinean joined #git
07:24 sbulage joined #git
07:25 JeroenT joined #git
07:28 ogny joined #git
07:28 MrWoohoo joined #git
07:37 z8z joined #git
07:37 zezba left #git
07:41 zefferno joined #git
07:42 schleppel joined #git
07:42 tojo joined #git
07:47 freimatz joined #git
07:47 ozmage joined #git
07:50 aidalgol joined #git
07:51 acetakwas joined #git
07:52 tvl joined #git
07:52 a3Dman joined #git
07:56 slayerjain joined #git
07:57 djb-irc joined #git
07:57 Sp4rKy_ left #git
07:58 grayjoc joined #git
08:00 askb joined #git
08:01 alexandre9099_ joined #git
08:04 askb joined #git
08:04 chele joined #git
08:04 khinashefrei joined #git
08:05 Tobbi joined #git
08:05 s2013 joined #git
08:05 jnavila joined #git
08:09 acetakwas joined #git
08:13 a3Dman joined #git
08:13 acetakwas joined #git
08:14 King_Hual joined #git
08:17 dmc joined #git
08:18 alexandre9099 joined #git
08:18 wrouesnel1 joined #git
08:19 dvaske joined #git
08:20 fatalhalt joined #git
08:21 ShapeShifter499 joined #git
08:21 btree_ joined #git
08:22 kiltzman joined #git
08:23 TomyWork joined #git
08:23 lucasem joined #git
08:24 luke_nukem joined #git
08:24 a3Dman joined #git
08:26 prions joined #git
08:26 Gloomy joined #git
08:26 irivqu joined #git
08:30 t0by joined #git
08:30 mknod joined #git
08:30 pothepanda joined #git
08:31 nowhere_man joined #git
08:32 elect__ joined #git
08:33 JeroenT_ joined #git
08:35 selckin joined #git
08:42 varazir joined #git
08:43 varazir Hello, how do I pull a specefic commit from a PR ?
08:44 theoceaniscool joined #git
08:44 Balliad joined #git
08:44 yushyin fetch the branch, cherry-pick the commit?
08:44 varazir I used this git pull origin pull/22232
08:45 varazir but the latest commit on that PS is broken
08:45 varazir PR*
08:45 a_thakur joined #git
08:45 Cabanossi joined #git
08:45 varazir yushyin: something like that
08:45 grawity in this case sounds like "fetch the branch, git merge the commit"
08:46 grawity e.g. `git fetch origin pull/22232 && git merge FETCH_HEAD~1`
08:46 grawity after undoing the pull/merge you already did, of course
08:48 varazir fatal: Couldn't find remote ref pull/22232
08:49 kurkale6ka joined #git
08:52 iasj joined #git
08:52 venmx joined #git
08:53 iasj Hello guys. I have a question. Is there any way to add and commit a symbolic link as its regular destination, instead of adding and commiting it as symbolic link?
08:55 selckin don't think so
08:55 iasj I thought so!
08:56 ShapeShifter499 joined #git
08:56 iasj that means some shell scripting as a work around
08:56 tobiasvl that sounds insecure
08:56 tobiasvl allowing git to write to files outside the repo
08:56 iasj oh
08:57 iasj that makes the hole sense now
08:57 iasj true
08:58 iasj I think I'll make a shell script that delete the link, copy the destination, and then add and commit, and then relink everything back.
08:59 iasj thanks anyway.
08:59 grayjoc joined #git
08:59 selckin maybe you want hardlinks
09:01 mikecmpbll joined #git
09:01 iasj No. I rather not use them. I'm afraid of making a mess.
09:05 afuentes joined #git
09:06 a3Dman joined #git
09:08 iasj left #git
09:09 Xatenev joined #git
09:10 Xatenev Helo
09:10 dfas joined #git
09:11 marcogmonteiro joined #git
09:12 clmsy joined #git
09:12 Tantamounter joined #git
09:13 mikecmpbll is there a way to see why i can't fast forward merge a branch? i can't see why in this instance it's not allowing it, i'm probably just being dense ..
09:13 mikecmpbll like, see merge conflicts or something
09:14 mikecmpbll eugh nvm i am being dense. goti t
09:15 sotoz joined #git
09:15 xaa joined #git
09:15 khinashefrei joined #git
09:18 mindfart joined #git
09:18 infinite joined #git
09:20 miczac joined #git
09:20 zeroed joined #git
09:20 zeroed joined #git
09:20 infinite joined #git
09:20 flaviodesousa joined #git
09:20 bariscant joined #git
09:21 bariscant joined #git
09:21 dvaske joined #git
09:22 infinite joined #git
09:29 Silmarilion Hi, I accidentally created branch B from branch A instead of creating branch B from master branch. I already have a few commits in branch B, is there a way to alter this?
09:29 miczac joined #git
09:30 Xatenev Silmarilion: No safe way I guess. You could try opening a new branch C from master and cherrypicking your single commits?
09:30 Silmarilion Xatenev: ok, Ill try that. Tnx
09:32 pbrewczynski joined #git
09:34 chll_ joined #git
09:35 GodEater left #git
09:37 jameser joined #git
09:38 tmg joined #git
09:38 juvenal joined #git
09:40 a3Dman joined #git
09:40 Vampire0 Silmarilion, you can simply rebase this
09:41 Vampire0 Silmarilion, `git rebase --onto master $(git merge-base A B) B`
09:41 Vampire0 Silmarilion, but be aware that you !rewrite history in case you already pushed B
09:41 gitinfo Silmarilion: 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
09:43 kexmex joined #git
09:44 Silmarilion ok tnx for the info
09:45 JeroenT joined #git
09:45 Cabanossi joined #git
09:46 TheWild I have two projects that are mostly the same but slightly different. To sync changes that I made in first project, I'm creating a patch of last commit (git format-patch -1) and apply it to second project (git am < changes.patch).
09:46 TheWild it mostly works
09:46 JeroenT_ joined #git
09:46 TheWild however, this time didn't and now I'm in the middle of an am session
09:46 TheWild ...fix conflicts and then run "git am --continue"
09:47 TheWild git didn't leave me <<<<<<, ====== and >>>>>>
09:47 howdoi joined #git
09:47 manuelschneid3r joined #git
09:48 Karazhan joined #git
09:50 sagittarian joined #git
09:50 Darcidride joined #git
09:52 Silenced_v2 joined #git
09:53 Vampire0 TheWild, !repro
09:53 gitinfo TheWild: [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
09:55 n1ce joined #git
09:55 Sound joined #git
09:56 Karazhan joined #git
09:56 hary joined #git
09:57 TheWild Vampire0: https://gist.github.com/anonymous/9d02ec5cd1d4d57d7c1a87bae2077bf5
09:57 dsantiago joined #git
09:57 rafalcpp joined #git
09:58 TheWild the file in work2 wasn't touched
09:59 overlord_tm joined #git
10:00 Silenced joined #git
10:00 brent__ joined #git
10:02 Snugglebash joined #git
10:05 n1ce Good morning
10:06 slayerjain left #git
10:07 sunrunner20 joined #git
10:10 Xatenev lol^^
10:10 Xatenev live server: Your branch is ahead of origin/master by 1782 commits
10:12 ertes joined #git
10:15 jameser joined #git
10:16 Murii joined #git
10:17 Skyfinn joined #git
10:19 marcogmonteiro joined #git
10:20 SlashLife^m Morning :)
10:20 aielima joined #git
10:21 SlashLife^m I'm writing a bash script as a git subcommand. I'd like to put some color in the output, but only when running interactive. Unfortunately when running the subcommand through git, it is executed by a program (=git), and thus does not count as interactive by the normal detection methods.
10:22 SlashLife^m Does git somehow communicate to subcommands whether they're being run in an interactive shell or not?
10:23 hexagoxel joined #git
10:25 fahadash joined #git
10:26 jameser joined #git
10:27 mda1 joined #git
10:28 Snugglebash joined #git
10:31 re1 joined #git
10:32 jameser joined #git
10:33 Xatenev SlashLife^m: Im confused. Can't you just simply use shell scripting to determine if its interactive or not with something like [[ $- == *i* ]] ?
10:33 Xatenev Why does git has to commnuicate to subcommands? Git subcommands are standalone executeables
10:34 miczac joined #git
10:35 cdown joined #git
10:37 cdown_ joined #git
10:38 SlashLife^m Xatenev: They are usually run through git, though.
10:39 vrach joined #git
10:39 SlashLife^m Xatenev: "git foobar" -> executes "git-foobar", but it does not do so directly; you run git, git then runs git-foobar.
10:39 SlashLife^m And at that point the information about whether or not the shell that ran git is lost.
10:39 SlashLife^m +is interactive
10:40 demize isatty?
10:40 demize (Is how you check in C.)
10:44 demize actually
10:44 demize SlashLife^m: `if tty -s`
10:44 SlashLife^m demize: Thanks, that helped. Found the shell equivalent for that: if [ -t 1 ]
10:45 demize "shell" is ambiguous ;p
10:45 demize But yeah, that'd be more portable.
10:47 jast [ is a standard tool, though, so it doesn't depend on the shell that much
10:47 TheWild Vampire0: I think I finally got what gone wrong that even "git am -3" was failing.
10:47 mehola joined #git
10:47 TheWild "git am -3" doesn't work very well for sample files that are 5 lines ling.
10:47 TheWild long
10:48 Vampire0 TheWild, ok, good that you solved it
10:49 YuGiOhJCJ joined #git
10:49 buzzLightBeer joined #git
10:50 theoceaniscool joined #git
10:50 Vampire0 jast, actzakky [ is a builtin in bash
10:51 Vampire0 s/actzakky/actually
10:51 jast sure, but if the builtin doesn't implement the standard flags, something is very wrong :)
10:51 Vampire0 indeed
10:52 ferr1 joined #git
10:53 jast anyway, my point was that, at least by default, [ supports this
10:53 fees joined #git
10:54 yqt joined #git
10:56 sbasso joined #git
10:58 lolmac joined #git
11:00 Cabanossi joined #git
11:01 XII-tdv_ joined #git
11:02 theoceaniscool joined #git
11:03 JeroenT joined #git
11:06 JeroenT joined #git
11:06 doener joined #git
11:07 irqq joined #git
11:07 sbasso joined #git
11:09 grift joined #git
11:11 xanadu_ joined #git
11:16 achadwick joined #git
11:18 dsdeiz joined #git
11:18 dsdeiz joined #git
11:18 tcorneli joined #git
11:24 Xatenev left #git
11:24 marianina8 joined #git
11:24 fakenerd_ joined #git
11:24 MineCoins joined #git
11:25 grift joined #git
11:27 jameser joined #git
11:27 dsdeiz joined #git
11:27 Snugglebash joined #git
11:28 z8z joined #git
11:29 xanadu_ joined #git
11:30 ozmage joined #git
11:30 z8z joined #git
11:35 fakenerd joined #git
11:36 oskarkv joined #git
11:38 jfr_ joined #git
11:41 lmatteis joined #git
11:42 juvenal joined #git
11:43 bvtton joined #git
11:44 MineCoins joined #git
11:44 _Vi joined #git
11:46 xanadu_ joined #git
11:46 polyrob_ joined #git
11:47 miczac joined #git
11:49 Cavallari joined #git
11:54 juvenal joined #git
11:56 byte512 joined #git
11:58 synthroid joined #git
12:01 buffal0 joined #git
12:01 masuberu joined #git
12:02 Stutters joined #git
12:03 cho joined #git
12:04 venmx joined #git
12:06 cdown_ joined #git
12:08 Skyfinn joined #git
12:11 navidr joined #git
12:12 miczac joined #git
12:15 livingstn joined #git
12:17 brent__ joined #git
12:20 kpease joined #git
12:21 selckin joined #git
12:21 sespiros joined #git
12:26 MineCoins joined #git
12:27 xanadu_ joined #git
12:28 kpease joined #git
12:29 grayjoc joined #git
12:31 Muzer joined #git
12:36 jameser joined #git
12:38 XII-tdv_ joined #git
12:39 Anthaas_ joined #git
12:42 Anthaas_ I am creating a git repo (git init), then I am adding my files (git add *.lp). I can see my files are added to staging (git status). I then write a commit message in some detail (git commit, edit in nano, ctrl+o, ctrl+x). I then get a message saying "6 files changed, 397 insertions, create mode 100644 x.lp" for all files x. I then write git status and it
12:42 Anthaas_ says "On branch master, nothing to commit, working directory clean" - I haven't pushed, and pushing fails because of this?
12:44 brent__ joined #git
12:44 tobiasvl Anthaas_: pushing fails because of what? how does pushing fail?
12:44 osse Anthaas_: there is nothing wrong wtih what you describe
12:45 osse you must misunderstand
12:46 tobiasvl Anthaas_: you don't mention adding a remote that you push TO. what remote do you add? is it an empty repo, or are you trying to push your one-commit repo to a repo that already has history?
12:46 osse also: who was phone?!
12:47 hvxgr joined #git
12:47 nowhere_man joined #git
12:48 ionl joined #git
12:48 ionl left #git
12:48 oncall-pokemon joined #git
12:49 oncall-pokemon Is `git fetch origin && git rebase -i master` the same as `git checkout master && git pull && git rebase -i master` in terms of outcome of the rebase?
12:49 bwn joined #git
12:49 Anthaas_ Ahhh sorry
12:50 Anthaas_ Connection refused :S
12:50 qqx oncall-pokemon: No. The first one won't change your local master branch, only the remote-tracking branches.
12:50 jast oncall-pokemon: on master, 'git rebase -i master' is a no-op. so, other than updating master, your second example does nothing. your first one doesn't update any branch but rebases the current branch against master (in interactive mode)
12:50 oncall-pokemon qqx: oh I see. Thanks
12:51 oncall-pokemon jast: yeah I missed the part where I move back to the feature branchy and rebase
12:51 oncall-pokemon on the second one
12:52 dglambert joined #git
12:52 jast right
12:52 oncall-pokemon my goal is to make an alias to get the new master code and then rebase feature
12:52 oncall-pokemon so I don't have to type it all in every time
12:52 Anthaas_ Why might I be getting a connection refused?
12:52 jast in normal use, you can't update a local branch if it's not checked out. there are exceptions but if you don't know them, you shouldn't know about them. ;)
12:52 Anthaas_ ssh: connect to host github.com port 22: Connection refused
12:52 Anthaas_ fatal: Could not read from remote repository.
12:52 osse MY guess would be someone or something refused your connection
12:53 jast Anthaas_: because your network is broken, your URL is incorrect, or the remote server is down
12:53 osse is ssh properly set up? does 'ssh sameurlhere' work ?
12:54 Anthaas_ It would seem ssh github.com:MyProfile/my-project.git doesnt work
12:54 oncall-pokemon so jast and qqx  for me to get the new code in master I should move there and do a pull? is there not a way to do it without double checkout
12:54 jast only if you don't have any commits on master that don't exist on the remote
12:54 marianina8 joined #git
12:55 jast and then the syntax is quite special (a good hint you shouldn't normally do this): git fetch; git push . origin/master:master
12:55 PioneerAxon joined #git
12:55 Gsham joined #git
12:56 synchronous joined #git
12:56 oncall-pokemon got ya
12:57 jeffreylevesque joined #git
12:57 qqx You can actually do that in a single command with `git fetch origin master:master`.
12:59 ankit01ojha joined #git
12:59 a3Dman joined #git
12:59 oncall-pokemon thanks looks good just what I needed
12:59 zivester joined #git
13:00 fakenerd joined #git
13:00 osse Anthaas_: you must add git@ in front
13:00 gugah joined #git
13:01 qqx But the presence of that shouldn't affect getting a connection refused error.
13:03 Anthaas_ osse: I did.
13:03 Anthaas_ it came back with it as I posted in the error message
13:03 osse !transcript
13:03 gitinfo Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
13:04 Anthaas_ https://www.irccloud.com/pastebin/jBYX9kSq/
13:06 osse Anthaas_: does this link open in your browser (after you un-cencor it) ? https://github.com/me/project
13:06 jameser joined #git
13:06 qqx Anthaas_: For testing with ssh by itself, remove the : and everything after it.
13:06 osse https://help.github.com/articles/testing-your-ssh-connection/
13:07 Anthaas_ osse: Yes it does
13:07 osse ok, what does git remote -v origin say
13:07 Anthaas_ Ahh ok, so yeah, something is blocking ssh to github.com
13:07 Anthaas_ Thanks
13:07 osse I mean git remote -v
13:07 Anthaas_ I'll go berate the IT guys...
13:08 ThomasLocke joined #git
13:09 netj joined #git
13:10 fbwnd joined #git
13:10 armyriad joined #git
13:10 fbwnnd joined #git
13:11 fbwnd joined #git
13:11 Gloomy joined #git
13:14 jameser joined #git
13:21 juvenal joined #git
13:25 XII-tdv_ joined #git
13:25 ij joined #git
13:26 ij 1. master points to x. 2. `git symbolic-ref master not-x`. 3. master still points to x. Why?
13:26 cdown joined #git
13:28 orbyt_ joined #git
13:28 qqx ij: The symbolic-ref command needs full ref names not dwim shortcuts. Like refs/heads/master
13:31 ij ah!
13:35 ij For both refs that is? "g s-r refs/heads/master refs/heads/not-x"?
13:36 qqx ij: Yes
13:36 qqx You should also check to see if you have a .git/master file created by your previous attempt. That could cause issues for some commands.
13:36 ecuanaso joined #git
13:37 Anthaas_ left #git
13:37 wrouesnel joined #git
13:37 livingstn joined #git
13:38 ij Yeah, that's why I was asking. I immediately did «find .git | grep master» upon «master» is ambiguous.
13:38 ij wrong quoting there ^
13:38 marianina8 joined #git
13:39 osse is it a good idea to turn master into a symbolic ref?
13:39 ij Shouldn't impact much.
13:39 ij Maybe I'll point the arrow otherwise though.
13:40 ij What's the antonym of "symoblic ref"?
13:40 ij Concrete?
13:40 osse not sure it has an antonym
13:41 ij Symbolic-refs are so weird, because upon checking it out, the HEAD resolves to the concrete ref —, seemingly, not what you checked out.
13:41 osse they're just like symlinks
13:41 dave0x6d joined #git
13:41 osse in fact, .git/HEAD used to be a symlink
13:42 GodGinrai joined #git
13:42 ij Right, but it's still not what I'm used to see happening.
13:43 qqx I don't think that saying they're like symlinks explains that behaviour. If I do `vi symlink`, vi will say I'm editing the symlink not what it points to.
13:43 ronator joined #git
13:44 qqx Not a whole lot of tools will use readlink to determine what a symlink actually points at.
13:44 Es0teric joined #git
13:45 Cabanossi joined #git
13:45 nopf joined #git
13:45 ij Hmmmm... git branch shows "* not-x", but HEAD points to "ref: ref/heads/master".
13:46 publio joined #git
13:47 sarri joined #git
13:47 sarri joined #git
13:47 DarkPsydeLord joined #git
13:47 AaronMT joined #git
13:48 Muzer joined #git
13:49 ExoUNX joined #git
13:49 bvtton joined #git
13:51 jameser joined #git
13:52 Gloomy joined #git
13:52 JeroenT_ joined #git
13:52 Gloomy joined #git
13:53 xanadu_ joined #git
13:54 _ikke_ It's ref vs symbolic-ref
13:54 Ryanar joined #git
13:55 ij I'm failing to update-ref it back. http://sprunge.us/KGXZ
13:56 qqx ij: You probably need to use the --no-deref option.
13:57 Bennux joined #git
13:57 ij Knowledge: gained.
13:57 reznord when we rename a file in a project, why doesn't it show up in git status?
13:57 nikitasius joined #git
13:58 _ikke_ reznord: might be that the content changed so much that git does not recognize it as a rename
13:58 osse reznord: what does git status say ?
13:58 fahadash joined #git
13:58 j7k6 joined #git
14:00 reznord osse: _ikke_ here is a screenshot of the whole thing https://cl.ly/223n3E2W2G1f
14:01 osse judging by the font I can only assume you're on OSX
14:01 osse (no offense)
14:01 osse anyway, OSX is not case-sensitive
14:02 reznord yeah, I am no OSX
14:02 reznord * on
14:02 osse try renaming it to somethingelse.js and see what happens then
14:02 Es0teric joined #git
14:03 qqx reznord: If you use `git mv` rather than just `mv` git should pick that up.
14:03 j7k6 joined #git
14:03 _ikke_ Note that git never stores renames
14:04 _ikke_ so it has little significance whether git status mentions it or not
14:04 reznord osse: ah !! that worked
14:04 reznord casesensitive is the problem :P
14:04 qqx _ikke_: It doesn't matter if it shows it as a rename, but it does matter if it sees the change at all.
14:04 j7k6 joined #git
14:04 osse also, that font is growing on me
14:05 osse it looks weird to see a font in that style being monospaced. but weird in a good way.
14:06 j7k6 joined #git
14:07 reznord osse: well, I like this font. so, couldn't help myself in changing it ????
14:08 * osse checks IRC on phone to see what emoji that's supposed to be
14:08 * _ikke_ wonders how valid the strategy of git rebase -i @{u} is, and then changing the todolist to noop, to get rid of all the commits git didn't recognize as duplicate (because of an upstream rebase)
14:09 j7k6 joined #git
14:09 muhannad joined #git
14:09 _ikke_ technically, I could just use a hard reset
14:13 osse _ikke_: to be super duper sure, shouldn't you do the rebase the usual way and use --skip where appropriateP?
14:13 JeroenT joined #git
14:15 learning joined #git
14:16 JeroenT joined #git
14:17 plos joined #git
14:17 _ikke_ perhaps
14:17 j7k6 joined #git
14:17 _ikke_ git already skipps the commits that it recognizes as duplicate
14:19 fhackdroid joined #git
14:19 Random832 if you're doing an interactive rebase it gets annoyed at you if it ends up trying to make a commit that doesn't do anything
14:19 j7k6 joined #git
14:19 JeroenT joined #git
14:20 matoro joined #git
14:20 _ikke_ it just warns you
14:21 _ikke_ If resolving a conflict for example results in an empty commit, git warns you that the commit itself may be superfluous
14:22 jstimm joined #git
14:29 LeBlaaanc joined #git
14:29 Snugglebash joined #git
14:30 JeroenT joined #git
14:33 notebox joined #git
14:34 netj joined #git
14:35 Dolos joined #git
14:37 mazert_ joined #git
14:39 Dolos joined #git
14:40 ggherdov Hello. Is there a way delete, in the local repository, branches such as origin/foo and origin/bar without ever connecting to the remote?
14:40 ggherdov I don't want the remote to be affected, just get rid of those locally as they clutter my otherwise delightful `git log --oneline --graph --all --decorate`.
14:40 ggherdov `git branch -d/-D` doesn't seem to work on those
14:41 canton7 ggherdov, -d -r will. Note that they will be recreated next time you do a 'git fetch' (or 'git pull')
14:41 SlashLife^m How would I check the hash of a branch without switching to it? (I want to check whether a local feature branch has the same state as the upstream before putting upstream away)
14:41 canton7 ggherdov, if they no longer exist on the remote origin, 'git fetch --prune' will get rid of them
14:41 Dolos joined #git
14:41 ggherdov canton7: ok thanks
14:42 canton7 ggherdov, otherwise, I'd suggest fixing your log command to not include those branches (replace the --all with something more specific), rather than deleting them
14:42 e14 joined #git
14:42 _ikke_ you can add --branches instead of --all
14:42 ggherdov _ikke_: thanks!
14:42 ionl joined #git
14:42 _ikke_ and --tags also if you are interested in that
14:42 JanC_ joined #git
14:43 ggherdov ok
14:43 Es0teric joined #git
14:44 _ikke_ SlashLife^m: git rev-parse <branchname>
14:44 _ikke_ you can also use git branch --merged
14:45 Dolos joined #git
14:45 Dolos joined #git
14:46 ionl left #git
14:46 zxd joined #git
14:46 guampa joined #git
14:47 evanextreme joined #git
14:48 SlashLife^m _ikke_: Thanks. branch --merged looks better; it doesn't make much sense to force a user to update a local branch to test for equality with the origin when the next step is abandoning the branch anyway.
14:48 qthwyx joined #git
14:49 xanadu_ joined #git
14:50 Ryanar joined #git
14:50 jozwior joined #git
14:50 Es0teric_ joined #git
14:50 Orphis joined #git
14:51 oleo joined #git
14:51 ThomasLocke joined #git
14:52 ss22ever joined #git
14:53 kjsaihs joined #git
14:53 fbwnd joined #git
14:55 ss22ever Hello I am a new git user can anybody help me to pull changes from the repository ..but git status showing me to be ahed of 1 commit
14:55 osse ss22ever: then there is nothing to upll
14:55 osse pull
14:57 peacememories joined #git
14:58 ss22ever osse: no actually i am ahed of 1 commit in my local repo and I want the changes to be pulled to my local repo only from the actual repo..
14:59 osse ss22ever: what changes? there are no changes to be pull to your local repo. you already have them
14:59 osse do you mean you want to push ?
15:00 matriness joined #git
15:00 al-damiri joined #git
15:00 evanextreme joined #git
15:00 evanextreme left #git
15:00 kjsaihs left #git
15:01 ss22ever osse:no to ensure I am using the latest version of the repo..do I need to clone the whole repo again?
15:01 osse ss22ever: no
15:01 osse ss22ever: you are using the latest version. in fact, you're using something even newer
15:01 armyriad joined #git
15:02 osse that's what "ahead" means
15:02 ss22ever osse:well no I am specifically talking about some other repo qemu
15:02 steeze joined #git
15:03 osse doesn't matter what repo you're talking about. if git status says you're ahead then you're ahead
15:03 ss22ever osse: and there are commited changes which my local repo is unable to have ..just because I dont have them yet?
15:03 osse no, there are no committed changes you don't have
15:03 ss22ever osse:yeah then what does ahed mean
15:03 cdg joined #git
15:03 osse ss22ever: that YOU have committed changes THEY don't have
15:03 qqx ss22ever: Are you using your own fork, e.g. at github and need to get changes from the original repo?
15:05 ss22ever qqx:Yeah !!
15:05 osse !pull_upstream
15:05 gitinfo To incorporate changes from an "upstream" or forked-from repository: git remote add upstream $URL; git fetch upstream; git merge upstream/$BRANCH. https://help.github.com/articles/syncing-a-fork/
15:06 TbobbyZ joined #git
15:07 TbobbyZ joined #git
15:09 adamru joined #git
15:11 bravespear|2 joined #git
15:11 Ranhir joined #git
15:12 raijin joined #git
15:12 JDubs joined #git
15:12 YuGiOhJCJ joined #git
15:14 hotbobby joined #git
15:14 Muzer joined #git
15:14 hotbobby hi if theres a repo that i want to submit pull requests to, do i have to branch it? how would i submit any changes back to the original repo?
15:15 osse hotbobby: you either have to "branch" it (ie. create a fork), or request permission to submit changes back
15:16 orbyt_ joined #git
15:16 cagomez joined #git
15:16 ecuanaso joined #git
15:17 scoobertron joined #git
15:17 bomb-on joined #git
15:17 CACAleedhock joined #git
15:21 Sansay joined #git
15:23 Sansay Hey guys, im working on a git repo thats hosted on a linux machine, and i work on it on a windows machine, im trying to understand how the .gitattributes file works, from my understanding you can have it always pull using LF even when on windows is this correct? I just need to add *.ext and it will always treat the file line endings when pulled from the repo with LF?
15:23 LeBlaaanc joined #git
15:23 peacememories joined #git
15:25 i7c You can add *.ext text eol=lf  to have lf in the working directory.
15:25 i7c And it does not happen during pull but during checkout.
15:25 i7c But overall, yes. :)
15:25 cagomez joined #git
15:25 Sansay let me try to explain my problem and you tell me if this is the correct solution
15:26 i7c https://git-scm.com/docs/gitattributes
15:26 i7c The section "End-of-line conversion" might be enlightning.
15:27 Sansay i have a remote repo on a linux based machine. Using netbeans i have a copy of that on my Windows machine. When i checkout on my windows machine sometimes git diff will say the whole file is changed when its really not due to line endings being different
15:27 i7c *enlightening, hopefully :D
15:27 YuGiOhJCJ joined #git
15:28 i7c Sansay, where you host the remote does not matter
15:28 i7c What matters are the kinds of line endings that get checked in to the repo (committed) and with what settings you check them out again.
15:29 Sansay do you know if i update my attributes file, will those settings take affect when i knewly clone a repo as well?
15:29 Sansay that has this settings file?
15:31 i7c it would take immediate effect, you don’t need to clone again
15:31 i7c Sansayn, tbh you can just set git config core.eol=lf
15:31 i7c If you want to have lf line endings.
15:31 Sansay which way do you think is the best approach?
15:31 i7c Just run a git checkout <something> again and it shoould be updated.
15:32 i7c I would prefer setting the config.
15:32 Sansay would i need to run it against all files?
15:32 i7c No.
15:32 i7c You just checkout the commit again.
15:33 Sansay okay thank you, ill forget the .gitattributes file and set the config globally
15:33 Sansay the default for core.eol is it auto?
15:33 Sansay if i want to reset it
15:33 i7c If you do the command I mentioned, it is also only repo-wide
15:34 i7c if you want globally you have to specify --global
15:34 brent__ joined #git
15:34 i7c The default is native.
15:34 SmashingX joined #git
15:34 i7c man git-config
15:34 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
15:34 Sansay ah okay
15:34 cdg_ joined #git
15:34 Sansay thank you so much for your help
15:37 Dave114 joined #git
15:37 iamthad joined #git
15:38 Gustavo6046 joined #git
15:38 i7c Sansay, actually I think you should use a tool that can handle both crlf and lf line endings seamlessly :P
15:39 destino joined #git
15:41 Sansay do you ever run into this problem?
15:41 Sansay if so how do you solve it?
15:42 KevinMGranger[M] left #git
15:42 jstimm joined #git
15:43 i7c You can have your editor/IDE normalize line endings or you can have git normalize it.
15:43 i7c See the autocrlf option
15:43 i7c Also described here: https://git-scm.com/docs/gitattributes
15:44 i7c Have to leave. bbl
15:44 KevinMGranger joined #git
15:44 borkr joined #git
15:46 Derperperd joined #git
15:46 concerti joined #git
15:47 sbasso joined #git
15:48 SmashingX joined #git
15:50 iamthad joined #git
15:51 SmashingX joined #git
15:52 rominronin joined #git
15:52 matoro joined #git
15:52 TbobbyZ_ joined #git
15:52 rominronin joined #git
15:53 robotroll joined #git
15:54 bannakaffalatta joined #git
15:55 rominronin joined #git
15:56 iamthad joined #git
15:57 shgysk8zer0 joined #git
16:00 Cabanossi joined #git
16:01 roelmonn_ joined #git
16:04 Skyfinn joined #git
16:04 TbobbyZ joined #git
16:06 Tobbi joined #git
16:07 jnavila joined #git
16:08 madewokherd joined #git
16:09 nd joined #git
16:10 jstimm joined #git
16:11 durham joined #git
16:14 GodGinrai joined #git
16:15 raulp joined #git
16:19 raspado joined #git
16:26 Lac3rat3d joined #git
16:27 robattila256 joined #git
16:31 svm_invictvs joined #git
16:31 kfoonamalik joined #git
16:32 iamthad joined #git
16:32 jagob joined #git
16:32 Lac3rat3d left #git
16:34 livingstn joined #git
16:36 astrofog joined #git
16:37 Keytap joined #git
16:38 Keytap Quick, what's the best way to split a branch in two at a specific commit?
16:39 Keytap joined #git
16:41 cagomez joined #git
16:42 cagomez joined #git
16:42 Eugene Keytap - you can create a new branch anywhere you like; git doesn't care. `git branch <newbranch> <starting-point>`
16:42 madewokherd Keytap: git checkout commitid -b newbranch
16:42 Eugene Or the checkout syntax there to create & checkout it at once ^
16:42 Keytap And then rebase that branch to master?
16:42 e14 joined #git
16:42 Keytap (so that it can be merged into master, rather than the origin branch)
16:42 Keytap original*
16:43 madewokherd now I'm confused
16:43 Keytap Okay
16:43 Keytap So I have two branches.  master and foo
16:43 Keytap foo's head split from master ten commits back
16:43 Keytap But starting from commit 5 going forward I worked on something completely unrelated to foo
16:44 Keytap And it will take us a while to get foo merged in
16:44 madewokherd oh, so you want one branch with the first 5 and one with the last 5?
16:44 Keytap So I'd like to take the last five commits of foo, and make it as if I had just branched off of master to begin with
16:44 Keytap Yeah
16:44 SmashingX joined #git
16:45 immae from foo, git checkout -b new_name and git rebase --onto=master new_name^^^^^ new_name
16:45 ToBeCloud joined #git
16:45 iamthad joined #git
16:46 immae the things will stay in foo, you may then do git checkout foo and git reset --hard HEAD^^^^^
16:46 Keytap new_name^^^^^ just means new_name five commits back, right?  Same as new_name~5?
16:47 immae yep
16:47 Keytap Okay.  (it's not actually five commits back)
16:47 Keytap Was just making sure
16:47 immae replace it by a commit hash it's safer :p
16:47 Keytap Sure thing
16:47 diogenese joined #git
16:48 immae actually commit_hash^
16:48 immae otherwise you wont take it
16:48 Keytap What does the karet do heree?
16:48 Keytap Wouldn't that be once commit behind the commit?
16:48 immae yes
16:48 opendev joined #git
16:48 Keytap Huh.  Alright
16:49 immae in the rebase command you don't take the first commit
16:49 madewokherd you want to use last commit that won't be included in the rebase
16:49 opendev left #git
16:56 disi joined #git
16:56 rominronin joined #git
16:56 s1scha joined #git
16:58 Es0teric joined #git
16:59 iffraff joined #git
17:00 Cabanossi joined #git
17:00 iamthad joined #git
17:01 cagomez joined #git
17:01 chardan joined #git
17:02 qqx I'd instead use `git rebase -i master` from the new branch.
17:02 qqx Then just remove the ones that aren't intended for that branch from the instruction sheet.
17:03 qqx Easier than working out the number of commits.
17:03 chardan joined #git
17:04 kexmex joined #git
17:05 t0by joined #git
17:05 t0by joined #git
17:05 WeiJunLi joined #git
17:05 cagomez joined #git
17:06 cagomez joined #git
17:07 iamthad joined #git
17:07 evil-t0by joined #git
17:08 cjohnson qqx: can you clarify that command for me? sounds like something I wish I'd known about sooner
17:08 cjohnson does that find the most recent ancestor from current branch to master and rebase everything since there?
17:09 qqx That would list all of the commits not already on master and let you choose which ones to keep (and even reorder them if desired).
17:10 TikityTik joined #git
17:10 qqx Although looking back a bit closer, the counting was coming up when trying to rewind the existing branch not when trying to remove the inapplicable commits from the new one.
17:11 peacememories joined #git
17:11 qqx But I guess `rebase -i master` could be used for that as well. Would just need to select a different set of commits to keep.
17:12 qqx And if an initial set of commits are all kept (and in the same order) it shouldn't even do an actual rewrite of those.
17:14 cyberz joined #git
17:17 overlord_tm joined #git
17:19 chipotle joined #git
17:21 bariscant joined #git
17:22 cyberz joined #git
17:22 boombatower joined #git
17:22 HardlySeen joined #git
17:23 HardlySeen joined #git
17:25 muhannad joined #git
17:25 fakenerd joined #git
17:26 Discovery joined #git
17:26 yehowyada joined #git
17:27 leeN joined #git
17:28 jstimm joined #git
17:29 SmashingX joined #git
17:31 fakenerd joined #git
17:32 gopar joined #git
17:33 raijin joined #git
17:34 iamthad joined #git
17:36 cyberz joined #git
17:36 skered joined #git
17:36 SmashingX joined #git
17:37 genq joined #git
17:37 marianina8 joined #git
17:38 thiago joined #git
17:39 nowhere_man joined #git
17:39 joshszep joined #git
17:41 algun joined #git
17:41 tang^ joined #git
17:41 matoro joined #git
17:42 cyberz joined #git
17:42 SmashingX joined #git
17:44 Noldorin joined #git
17:44 causasui joined #git
17:45 cjohnson qqx: how does it know if it's on master? if I cherry pick a commit to master will it not show up?
17:45 cjohnson and what about merged commits
17:45 Cabanossi joined #git
17:45 qqx Commits merged into master would definitely be considered as part of master.
17:46 blackwind_123 joined #git
17:46 cjohnson but cherry-picks? where commit ID doesn't match?
17:46 qqx It will attempt to determine if the same commit is present, but with a different ID. But that isn't always successful.
17:47 cjohnson cool good to know
17:47 alexandre9099 joined #git
17:47 qqx But if you identify identical commits in the instruction list you can always remove them before it actually starts to do anything.
17:48 qqx And any that weren't identified will likely result in conflicts. And when the rebase stops for you to resolve them you could use `git rebase --skip` to skip them at that point.
17:51 Es0teric joined #git
17:51 Gsham joined #git
17:52 Es0teric_ joined #git
17:52 robotroll joined #git
17:54 raijin joined #git
17:54 iamthad joined #git
17:54 rominronin joined #git
17:56 durham_ joined #git
17:58 porsche944 joined #git
17:58 cyberz joined #git
17:58 menip joined #git
18:00 marianina8 joined #git
18:01 Keytap Hello again all.  If I wanted to pluck one commit out of a branch and move it to its own branch, how would I go about that?
18:01 Keytap (using its commit hash, preferably)
18:02 n1ce joined #git
18:02 cjohnson cherry-pick
18:02 cjohnson man cherry-pick
18:03 cyberz joined #git
18:03 cjohnson man git-cherry-pick
18:03 gitinfo the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html
18:03 cjohnson sorry
18:03 Keytap ty
18:03 cjohnson if the cherry pick's diff is not compatible with the target branch it will fail and you'll have to manually rectify
18:04 cjohnson then that will ead to conflicts if you end up merging the results of those two branches anywhere
18:04 Keytap So I need to branch off of master, then move cherry pick the commit (which exists in another branch) onto the newly created branch?
18:05 Keytap And does cherry picking remove the commit from the other branch's history?
18:06 la-mogette joined #git
18:07 hotbobby joined #git
18:07 MarioBranco joined #git
18:10 iamthad joined #git
18:12 hvxgr joined #git
18:12 raijin joined #git
18:12 ertesx joined #git
18:13 bwn joined #git
18:14 durham joined #git
18:15 btree_ left #git
18:16 angelo_ts joined #git
18:16 anth0ny__ joined #git
18:17 Sasazuka joined #git
18:18 aielima joined #git
18:18 matoro joined #git
18:19 e14 joined #git
18:20 thedudez0r joined #git
18:20 osse Keytap: No
18:20 m0viefreak joined #git
18:21 osse Ehh that was the answer to your a
18:21 osse Your last question
18:21 TomyLobo2 joined #git
18:22 learning joined #git
18:22 iamthad joined #git
18:23 inflames joined #git
18:23 Keytap What if I'm moving the commit to its own repo?
18:23 Keytap lol
18:23 miczac joined #git
18:24 Sasazuka joined #git
18:26 bwn joined #git
18:26 TbobbyZ_ joined #git
18:26 gopar joined #git
18:27 snits joined #git
18:27 rominron_ joined #git
18:28 manuelschneid3r joined #git
18:28 adamru joined #git
18:29 duderonomy joined #git
18:30 raijin joined #git
18:30 i7c What does that even mean?
18:33 iamthad joined #git
18:33 Sound joined #git
18:34 dhollinger joined #git
18:36 cyberz joined #git
18:38 eroux joined #git
18:39 Silmarilion joined #git
18:39 TheWild joined #git
18:40 m1dnight_ joined #git
18:40 la-mogette joined #git
18:40 synthroid joined #git
18:42 SuperLag joined #git
18:43 SuperLag If you have a large file downloading into repo "foo", and you rename the working branch to something else... is that going to cause problems?
18:45 Cabanossi joined #git
18:45 livingstn joined #git
18:46 theiostream joined #git
18:47 jeffreylevesque joined #git
18:48 peterbecich joined #git
18:48 Keytap Sorry, I should have been more clear.  The commit basically turned into its own composer package for our project.
18:48 Keytap So we moved it into its own commit.
18:50 SmashingX Vampire0: so then what do you recommend to use instead of submodules for dev?
18:52 Keytap *so we moved it into its own repository/package
18:53 m1dnight_ joined #git
18:55 Eryn_1983_FL joined #git
18:56 cyberz joined #git
18:56 oleksiyp joined #git
18:57 Atm0spher1c joined #git
18:58 jstimm joined #git
19:01 gopar joined #git
19:01 Es0teric joined #git
19:02 minn joined #git
19:02 m1dnight_ joined #git
19:02 theiostream joined #git
19:04 Es0teric joined #git
19:04 Sadeukko joined #git
19:05 StuartMI joined #git
19:10 cyberz joined #git
19:11 ariver|ERR left #git
19:11 ariver joined #git
19:12 bgamari joined #git
19:13 fahadash joined #git
19:14 prg318 joined #git
19:14 dreiss joined #git
19:14 Eryn_1983_FL joined #git
19:16 Bennux joined #git
19:17 aidalgol joined #git
19:19 quadz joined #git
19:20 miczac joined #git
19:20 TheWild joined #git
19:22 luke_nukem joined #git
19:24 cyberz joined #git
19:28 willingc joined #git
19:31 Darren_ joined #git
19:34 cyberz joined #git
19:35 la-mogette joined #git
19:37 la-mogette_ joined #git
19:38 d5sx43 joined #git
19:42 cyberz joined #git
19:43 inflames joined #git
19:44 devhost joined #git
19:44 Es0teric joined #git
19:44 LionsMane joined #git
19:44 nixjdm joined #git
19:44 cdown joined #git
19:44 brent__ joined #git
19:46 GT4066 joined #git
19:47 TbobbyZ joined #git
19:48 cagomez joined #git
19:49 satifant joined #git
19:52 SmashingX joined #git
19:52 lucido-cl joined #git
19:53 ecuanaso joined #git
19:54 m1dnight_ joined #git
19:56 renlo joined #git
19:57 Dazzyp joined #git
19:58 Silenced joined #git
19:58 renlo I have git v2.10 (so its > 1.6), when I do 'git checkout some-branch-which-exists-in-remote', absolutely nothing happens. The only thing that works is 'git checkout -b some-branch-which-exists-in-remote origin/some-branch-which-exists-in-remote' I only have *1* remote, named origin
19:58 MrMojit0 joined #git
19:58 cyberz joined #git
19:58 Keytap left #git
19:59 renlo any idea why it's doing this? I have never seen this behavior
19:59 rominron_ joined #git
20:00 Cabanossi joined #git
20:00 cyberz joined #git
20:01 theiostream joined #git
20:02 Aaeriele` left #git
20:02 Aaeriele joined #git
20:06 cdg joined #git
20:08 la-mogette joined #git
20:08 cyberz joined #git
20:09 la-mogette joined #git
20:12 parasite_ joined #git
20:12 Es0teric_ joined #git
20:15 Murii_ joined #git
20:15 luke_nukem joined #git
20:16 la-mogette joined #git
20:16 durham joined #git
20:17 askb joined #git
20:17 dreiss joined #git
20:17 cagomez joined #git
20:17 Rodya_ joined #git
20:17 _Vi joined #git
20:17 ecuanaso joined #git
20:19 ToBeFree joined #git
20:19 ozmage joined #git
20:19 Some_Person joined #git
20:19 la-mogette_ joined #git
20:19 Some_Person Is it just me, or is git rebase somewhat dumb. I have to wonder if there's a better way than manually doing trivial changes in my files
20:21 theiostream joined #git
20:22 ujjain joined #git
20:22 ujjain joined #git
20:22 iffraff1 joined #git
20:22 iffraff joined #git
20:23 learning joined #git
20:24 john-soda joined #git
20:24 irqq joined #git
20:25 luke_nukem joined #git
20:25 rschneberger joined #git
20:25 marianina8 joined #git
20:26 Murr joined #git
20:26 GodGinrai joined #git
20:27 llamapixel What makes it 'dumb' for you Sam Some_Person ?
20:27 willingc joined #git
20:27 Some_Person Seems like I'm having to resolve conflicts that look trivial to resolve
20:28 Some_Person like, where I just make it take everything
20:28 ozmage joined #git
20:28 Some_Person but I dunno, maybe there's a reason it does this
20:30 llamapixel You could check for expectation output in a test repo, provided you work with testing in mind.
20:32 AaronMT joined #git
20:32 Es0teric_ joined #git
20:33 Balliad joined #git
20:33 durham joined #git
20:35 snowalpaca joined #git
20:35 rominronin joined #git
20:36 marianina8 joined #git
20:37 rominronin joined #git
20:38 matoro joined #git
20:38 bamb joined #git
20:38 bamb left #git
20:38 neilthereildeil joined #git
20:38 neilthereildeil hi
20:39 neilthereildeil how can i clone a tree up to a certain commit only? i dont want subsequent commits
20:40 tang^ a clone is a clone
20:40 tang^ but you could clone and reset hard to the commit in question
20:40 luke_nukem joined #git
20:42 la-mogette_ joined #git
20:43 durham joined #git
20:45 Cabanossi joined #git
20:47 la-mogette joined #git
20:48 evanextreme joined #git
20:50 neilthereildeil left #git
20:50 tyreld joined #git
20:51 theiostream joined #git
20:53 hashtagstrashtag joined #git
20:55 HurricaneHarry joined #git
20:58 hwrdprkns joined #git
20:58 neilthereildeil joined #git
20:59 neilthereildeil how can i port changes from one git tree to another, preserving its dates?
21:00 Eugene neilthereildeil - you can't; `git clone` is an all-or-nothing thing for the source repo & its branches. --depth controls how much history you download, but that's the other-way-around from what you're thinking
21:01 ecuanaso joined #git
21:01 Eugene You might be able to do something with a custom fetchspec and only download 1 branch, but that's really not helpful in most scenarios. Why do you want to do this? Git branches are lightweight, so unless there's a lot of divergence between branches or in recent history it shouldn't matter....
21:01 Eugene And re: copying changes: man git-cherry-pick and man git-format-patch
21:01 gitinfo the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html
21:01 gitinfo the git-format-patch manpage is available at http://jk.gs/git-format-patch.html
21:01 neilthereildeil the source branch did some correct stuff and some incorrect stuff. i only want the correct stuff
21:04 livingstn joined #git
21:05 amosbird joined #git
21:05 neilthereildeil ye ai think cherrypick is what im looking for
21:05 ghoti joined #git
21:06 Eugene You probably want to `git clone` the source repo; `git checkout -b mybranch <GOODSHA>` to start your branch from wherever you like; then `git cherry-pick <COMMIT>` for each change you want to hand-pluck in
21:06 neilthereildeil can i cherry pick betwen repos, or only branches in teh same repo?
21:06 Eugene cherry-pick operates locally, like almost all git commands(fetch/push & friends are the only exceptions to this rule)
21:07 qmr ohi Eugene
21:07 * Eugene qmr
21:07 bariscant joined #git
21:07 matoro joined #git
21:07 neilthereildeil Eugene: hmm if i export the changes to a patch file, can i set the date on the commit manually?
21:08 Eugene That's what format-patch & am are for
21:10 osse renlo: is it unambiguous?
21:10 osse renlo: (ie. do you perhaps have more than one remote) ?
21:11 renlo just one remote osse
21:11 osse renlo: and for the given X you try to do 'git checkout X' with, what does 'git for-each-ref | grep X' say?
21:12 dsdeiz joined #git
21:12 jeffreylevesque joined #git
21:12 finalbeta joined #git
21:14 jost__ joined #git
21:14 renlo ohhh jeez i figured out what it was osse .... there's a file named 'some-branch-which-exists-in-remote' ... so I've been checking out the file
21:15 renlo the branch is named dev, apparently there's now a directory called 'dev', so I was checking out the directory
21:15 renlo my bad
21:15 Cabanossi joined #git
21:15 renlo feel kind of silly now
21:15 osse renlo: ahh! try 'git checkout dev --'
21:16 renlo that works! nice
21:17 _flow__ left #git
21:19 muhannad joined #git
21:21 dreiss joined #git
21:21 d5sx43 joined #git
21:22 neilthereildeil can i tell git format-patch to leave a file out of the diff?
21:23 finalbeta joined #git
21:23 osse neilthereildeil: 99% certain you can't. format-patch works on whole commits. you can edit the file after the fact
21:23 hfp Hey, how do I make a fixup with part of the message from the commit I want to squash into? Something like `git commit --fixup:/message/`, except I can't remember not find the correct syntax...
21:23 osse or change the commit first
21:23 hfp s/not/nor
21:24 osse hfp: --fixup=foo or --fixup foo
21:24 Unode joined #git
21:24 osse --fixup=:/message
21:24 n1ce_ joined #git
21:24 Unode is it possible to do a shallow clone (--depth 1) while including a selected few branches?
21:25 Unode I tried --no-single-branch but this gives me a reference to all remote branches.
21:25 hfp osse: yes, thank you
21:25 dirtyroshi joined #git
21:26 aidalgol joined #git
21:30 durham joined #git
21:30 rominronin joined #git
21:31 willingc joined #git
21:33 ozmage joined #git
21:34 zumba_addict joined #git
21:34 zumba_addict good afternoon. Am I right that git checkout branCH is case insensitive?
21:36 Eugene zumba_addict - git is case sensitive for branch names; your file system may or may not be. It is not a good idea to create conflicting names(eg, master and Master).
21:36 Unode left #git
21:38 zumba_addict so looks like my osx is doing some bad things but 99% of the time, I use lowercase
21:38 zumba_addict i was just doing some test
21:39 cyberz joined #git
21:39 nyuszika7h joined #git
21:40 nicolai86 joined #git
21:40 durham joined #git
21:43 tmg joined #git
21:45 Cabanossi joined #git
21:45 _ikke_ zumba_addict: yeah, hfs is by default case insensitive
21:46 zumba_addict yup and I found out that I need to reformat my hdd :( ARGH
21:46 sbasso joined #git
21:47 zumba_addict it's weird that when I git checkout my-FiX-branch, it works but when I execute git branch, it's not selected but git log shows the logs
21:48 le_melomane joined #git
21:50 durham joined #git
21:51 black_13 joined #git
21:52 dsdeiz_ joined #git
21:52 eroux joined #git
21:52 black_13 how do i pulll all the subproject of a git hub project
21:53 sbasso joined #git
21:53 juvenal joined #git
21:55 Rodya_ joined #git
21:56 aw1 joined #git
21:56 mizu_no_oto joined #git
21:57 willingc joined #git
22:00 Rodya__ joined #git
22:01 durham joined #git
22:02 _ikke_ black_13: What do you mean?
22:02 mikecmpbll joined #git
22:05 neilthereildeil is it possible to have 1 main repo with 5 branches, and 5 slave repose, each with only 1 of the 5 branches in main?
22:06 peacememories joined #git
22:06 osse neilthereildeil: git has no concept of master and slave. but nothing is stopping you from having one repo with five branches, and cloning that repo five times with --single-branch
22:07 black_13 _ikke_:  i want pull https://github.com/gsauthof/tree-model and all the submodules
22:08 osse black_13: git clone --recursive
22:09 osse if you already have cloned: git submodule update --init
22:09 willingc joined #git
22:11 alphor joined #git
22:12 puff I was working on a branch for a feature, basically sending an email out when a certain event happens, when I was asked put the email feature on hold, add date tracking for that event as a separate, new branch off master, and then finish the email feature - and include the date info in the email.
22:12 puff So I made a new branch off master, added the date stuff, committed/pushed.
22:13 puff The date branch hasn't been merged into master yet, but I want to finish the email feature.  If I merge the date branch into the email branch, and then later the team lead merges the date branch into master, and I then merge master into email, will that cause problems?
22:14 d5sx43 joined #git
22:15 Cabanossi joined #git
22:16 sbasso joined #git
22:17 osse puff: no
22:17 anderson joined #git
22:21 nidr0x joined #git
22:23 learning joined #git
22:24 Sasazuka joined #git
22:26 Phrohdoh joined #git
22:27 marianina8 joined #git
22:27 n1ce joined #git
22:29 puff osse: Thanks.
22:29 osse puff: why would you merge master into email?
22:30 puff osse: I pull master and rebuild/rerun tests before pushing the feature branch and issuing a merge request.
22:31 osse sounds fine
22:31 n1ce_ joined #git
22:31 osse the final log will be a bit criss-cross but it should work out fine
22:31 t0by joined #git
22:31 t0by joined #git
22:31 puff osse: Basically to make sure there's no bad interaction between my code and any changes somebody else might make/commit/push/merge while I'm working on the feature.
22:32 puff osse: Is that not standard?
22:32 manuelschneid3r joined #git
22:34 atomi joined #git
22:36 thiago joined #git
22:37 Phrohdoh So I have two projects and I want to merge one into the other but _not_ have a "git mv'd project B" commit or lose the commit metadata in project B. So project A is https://github.com/OpenRA/OpenRAModTemplate and project B is https://github.com/OpenRA/d2, I'd like for the root of project B to be <A>/mods/d2 (so that the file <A>/mods/d2/mod.yaml ex
22:37 Phrohdoh ists) but with existing commits containing valid filepaths (so a change to mod.yaml would be recorded/rewritten as a change to ./mods/d2/mod.yaml), how should this be done? http://stackoverflow.com/a/20974621/3816796 looks pretty close but I don't want to have a commit in the middle where project B lived outside of the desired path.
22:37 anuxivm joined #git
22:40 NightMonkey joined #git
22:43 neilthereildeil i have 2 repos based on the same initial commit. how can i clone from repo2 into repo1 in a new branch?
22:44 durham joined #git
22:44 Rodya_ joined #git
22:45 sarbs_ joined #git
22:45 Phrohdoh http://stackoverflow.com/a/21495718/3816796 may be closer to what I want
22:49 Phrohdoh Ok yep I'm pretty close, now I just want to rewrite the commit filepaths to include my new directory prefix
22:50 dave0x6d joined #git
22:50 Es0teric joined #git
22:51 hahuang61 joined #git
22:53 texasmynsted joined #git
22:54 NightMonkey joined #git
22:55 texasmynsted So when somebody does a push -f, and re-writes history for a public branch, is there a way to force a pull to over-write history in my local copy of the branch to match origin?
22:55 texasmynsted pull -f is not the same concept
22:56 ahrs joined #git
22:56 mda1 joined #git
22:56 Sound joined #git
23:00 tmg left #git
23:01 ujjain joined #git
23:01 ujjain joined #git
23:02 mysticjohn joined #git
23:02 rominronin joined #git
23:07 sbasso joined #git
23:08 cbreak texasmynsted: man git reset --hard
23:08 gitinfo texasmynsted: the git-reset manpage is available at http://jk.gs/git-reset.html
23:08 cbreak (after git fetch)
23:09 Nysar joined #git
23:10 joki joined #git
23:11 texasmynsted oh, so git fetch then git reset —hard will force it to match origin rather than an older version of the local
23:12 zq joined #git
23:12 zq how can i git rebase --skip all the conflicts?
23:12 texasmynsted hmm from the man page it does not seem like
23:13 zq all
23:13 zq of the conflicts
23:13 bremner zq: you can't, unless you want to skip all the commits too
23:13 zq all
23:13 zq bremner: nah, i know for certain that the commits i want don't conflict
23:13 texasmynsted ah like this git reset --hard origin/master
23:13 zq because they're in conflict-free areas of the file tree
23:14 miczac joined #git
23:15 bremner I'm not aware of a "skip-on-conflict" option, but you could probably script it given enough motivation.
23:15 zq ah alright
23:16 renlo joined #git
23:16 texasmynsted :( No git reset —hard origin/master did not work
23:16 texasmynsted it did not find the master branch
23:17 texasmynsted well it was not master, but it still did not find it
23:17 texasmynsted this is what worked but it seems wrong
23:17 texasmynsted git branch -D foo
23:17 texasmynsted git checkout foo
23:18 bannakaffalatta joined #git
23:19 zq is there a shortcut to find the most recent ancestor of two refspecs
23:20 zq nvm
23:21 zq left #git
23:22 renlo left #git
23:22 bronson_ joined #git
23:23 neilthereildeil left #git
23:24 royiv Is there a command that computes the difference foo...bar, and then lets me apply it as a patch (as in -p) to the current HEAD/working directory?
23:24 royiv Essentially, I want to cherry-pick -p a whole branch onto a different branch.
23:24 royiv (I have a branch with a bunch of junk I want to pull out of it into a new branch to push out for review.)
23:25 royiv *junk in it
23:25 scoobert1on joined #git
23:25 ramsub07 joined #git
23:27 peacememories joined #git
23:27 Sasazuka joined #git
23:30 cagomez joined #git
23:33 cagomez_ joined #git
23:34 SmashingX joined #git
23:34 bannakaf_ joined #git
23:37 WhiskerBiscuit joined #git
23:38 Sasazuk__ joined #git
23:40 cdown joined #git
23:40 mkoskar_ joined #git
23:41 madewokherd joined #git
23:41 Rotaerk joined #git
23:42 bannakaffalatta joined #git
23:43 cdown_ joined #git
23:45 irqq_ joined #git
23:45 cyberz joined #git
23:45 CicelyQuails joined #git
23:47 Kaisyu joined #git
23:47 Rodya_ joined #git
23:49 alexandre9099 joined #git
23:49 TbobbyZ_ joined #git
23:52 matoro joined #git
23:53 Gsham joined #git
23:54 NenitaMaharg joined #git
23:55 Sasazuka joined #git

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