Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-12-12

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

All times shown according to UTC.

Time Nick Message
00:01 quetz joined #git
00:05 jimi_ joined #git
00:10 bdonnahue joined #git
00:11 arescorpio joined #git
00:12 thebope joined #git
00:13 maroloccio joined #git
00:13 maroloccio hi. if projectA has CI verification, should each commit in branchA of projectA pass said CI checks or only merge commits? (git usage question)
00:14 tmg joined #git
00:15 tmg is there a natural way for `git format-patch --root` to not treat the initial commit as a PATCH?
00:15 mizu_no_oto joined #git
00:16 dikiaap joined #git
00:17 albel727 joined #git
00:18 particleGolem joined #git
00:20 mackiesurfacepro joined #git
00:23 CEnnis91 joined #git
00:25 CrazyMonkey201 joined #git
00:26 thott joined #git
00:29 milki maroloccio: most commonly, both the merge commit and the tip of the branch to be merged should pass CI checks
00:30 milki maroloccio: you probably want CI checks to pass before you consider a branch to be merged
00:30 milki maroloccio: and you want your master to always pass CI
00:30 breezy_ joined #git
00:39 ChiLL_SurF joined #git
00:40 rishi joined #git
00:42 CussBot joined #git
00:43 mizu_no_oto joined #git
00:47 zugzug joined #git
00:48 mischat joined #git
00:52 buffal0 joined #git
00:53 ncthom91 joined #git
01:00 davimore joined #git
01:02 _28_ria joined #git
01:04 _28_ria joined #git
01:04 DolphinDream joined #git
01:06 anuxivm left #git
01:07 eagle_mike joined #git
01:09 maroloccio milki: but you'd be ok with commits on a merged branch being "broken commits"?
01:10 nettoweb joined #git
01:10 lanking joined #git
01:11 raijin joined #git
01:17 dodobrain joined #git
01:23 maroloccio milki: thanks for you initial answer, by the way
01:24 a_thakur joined #git
01:28 cqi joined #git
01:28 jason237 joined #git
01:28 realtime joined #git
01:34 SwiftMatt joined #git
01:41 b3kNik joined #git
01:41 PaulCapestany joined #git
01:42 buffal0 joined #git
01:43 armyriad joined #git
01:45 Nilesh_ joined #git
01:46 m1dnight_ joined #git
01:47 ismaris joined #git
01:47 tmsmith joined #git
01:48 unbalanced joined #git
01:49 mischat joined #git
01:51 stfn joined #git
01:53 romerocesar joined #git
01:53 linuxmodder joined #git
01:55 Siegfried joined #git
01:56 dsdeiz joined #git
01:58 cagedwisdom joined #git
01:58 milki maroloccio: yeah, branches encompass work done which can lead to partially incomplete changes to a code base
01:59 milki maroloccio: only the whole is considered to be merged, not each individual change by themselves
02:00 ncthom91 joined #git
02:00 converge joined #git
02:01 converge I want to make a pull request, but in my last commit/push, I added a wrong file to the repository, is there a easy way to fix it?
02:01 milki converge: !fixup
02:01 gitinfo converge: 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!
02:01 milki converge: following that should lead to git commit --amend i think
02:02 nettoweb joined #git
02:03 Jameser joined #git
02:08 converge milki, tks
02:10 venmx joined #git
02:10 gusnan joined #git
02:12 drbean joined #git
02:13 thebope joined #git
02:16 tmg left #git
02:18 d^sh joined #git
02:21 snowkidind joined #git
02:32 Limes_ joined #git
02:39 anonymuse joined #git
02:46 andrew710 joined #git
02:46 tnecniv joined #git
02:50 mischat joined #git
02:51 Lyqyd_ joined #git
02:53 romerocesar joined #git
02:54 buffal0 joined #git
02:54 Gsham joined #git
02:54 pks joined #git
02:59 Emperor_Earth joined #git
03:04 romerocesar joined #git
03:04 pks joined #git
03:07 snowkidind joined #git
03:08 armyriad joined #git
03:09 VoidFox joined #git
03:13 buffal0 joined #git
03:14 thebope joined #git
03:15 Goplat joined #git
03:18 lindenle joined #git
03:20 basiclaser joined #git
03:21 Siegfried joined #git
03:22 chachasmooth joined #git
03:23 steelsw0rd joined #git
03:26 Gsham joined #git
03:27 texasmynsted joined #git
03:32 raijin joined #git
03:32 aavrug joined #git
03:32 aavrug joined #git
03:35 breezy_ joined #git
03:37 sunri5e joined #git
03:40 pks joined #git
03:41 oskarkv joined #git
03:46 Sample joined #git
03:50 mischat joined #git
03:51 fahadash joined #git
03:57 a_thakur joined #git
03:58 pks joined #git
03:59 hexagoxel joined #git
04:02 bdonnahue left #git
04:06 maroloccio joined #git
04:07 realz joined #git
04:07 Andrew_K joined #git
04:11 venmx joined #git
04:15 Vortex34 joined #git
04:18 fstd_ joined #git
04:21 anonymuse joined #git
04:21 hahuang62 joined #git
04:23 pks joined #git
04:29 ayogi joined #git
04:31 robotroll joined #git
04:33 larrybolt joined #git
04:39 aruns joined #git
04:40 subhojit777 joined #git
04:44 Noodlewitt joined #git
04:47 tarkus joined #git
04:48 JanC_ joined #git
04:51 mischat joined #git
04:52 rnat joined #git
04:55 Gsham joined #git
05:00 a_thakur joined #git
05:04 arnsa joined #git
05:06 pks joined #git
05:11 dreiss joined #git
05:12 daey joined #git
05:12 mischat joined #git
05:12 navidr joined #git
05:14 sbulage joined #git
05:16 thebope joined #git
05:16 jimi_ joined #git
05:17 a_thakur joined #git
05:25 wgrant joined #git
05:33 dodobrain joined #git
05:33 dodobrain joined #git
05:34 dodobrain joined #git
05:45 axisys joined #git
05:45 mahendratech joined #git
05:47 hahuang62 joined #git
05:48 jfr joined #git
05:48 holodoc joined #git
05:50 Cabanossi joined #git
05:52 kyan joined #git
05:53 murgi joined #git
05:54 clevodearia joined #git
05:55 morfin joined #git
05:55 morfin hello
05:56 morfin if you want to return reverted changes what you would do? Revert reverting commit?
05:56 _ikke_ correct
05:56 pks joined #git
05:58 jimi_sanchez joined #git
06:03 aavrug joined #git
06:04 aavrug joined #git
06:04 tarkus joined #git
06:07 a_thakur_ joined #git
06:09 romerocesar joined #git
06:11 venmx joined #git
06:13 mischat joined #git
06:15 brillenfux joined #git
06:16 QuestionTime joined #git
06:16 thebope joined #git
06:17 romerocesar joined #git
06:19 qt-x joined #git
06:20 brillenfux joined #git
06:20 mda joined #git
06:21 johnny56_ joined #git
06:21 jaguarmagenta joined #git
06:22 blackwind_123 joined #git
06:22 lordjancso joined #git
06:22 mehola joined #git
06:36 pks joined #git
06:38 duderino_ joined #git
06:39 baryon joined #git
06:41 johnny56_ joined #git
06:41 iliv joined #git
06:46 OOPMan joined #git
06:49 darkbit joined #git
06:53 PCatinean joined #git
06:53 overlord_tm joined #git
07:01 Raging_Hog joined #git
07:03 rscata joined #git
07:04 alcohol joined #git
07:05 darkbit joined #git
07:06 SteveCharleston joined #git
07:07 Andrew_K|2 joined #git
07:10 tarkus_ joined #git
07:13 mischat joined #git
07:16 jmpp joined #git
07:17 thebope joined #git
07:18 hahuang62 joined #git
07:19 lindenle joined #git
07:20 Flowinho joined #git
07:22 Stummi joined #git
07:24 romerocesar joined #git
07:24 JeroenT joined #git
07:25 aruns joined #git
07:25 themill joined #git
07:26 venmx joined #git
07:27 eduardas_m joined #git
07:30 quetz joined #git
07:33 ThomasLocke joined #git
07:33 ThomasLocke joined #git
07:37 swalladge joined #git
07:39 andrew710 joined #git
07:39 jnavila joined #git
07:40 zeroed joined #git
07:40 zeroed joined #git
07:42 GavinMagnus joined #git
07:44 GT4066 joined #git
07:47 inflames joined #git
07:49 KV joined #git
07:49 elect_ joined #git
07:52 ingenious joined #git
07:55 jmpp joined #git
07:55 jmpp joined #git
07:57 Macaveli joined #git
08:01 wgrant joined #git
08:02 acetakwas joined #git
08:03 chele joined #git
08:04 JeroenT_ joined #git
08:05 JeroenT_ joined #git
08:05 PCatinean joined #git
08:06 jmpp joined #git
08:06 jmpp joined #git
08:08 bocaneri joined #git
08:08 d0nn1e joined #git
08:09 murgi joined #git
08:11 Sceorem joined #git
08:12 lanking joined #git
08:14 mischat joined #git
08:18 anddam I'm reading about the definition of staging into git, from what I read I'd say that the staging area is what "diff --cached" refers to but I've read on a SO answer (now lost in the browsing history) that staged files are "almost" those referred by --cached
08:18 anddam which one is it?
08:20 Alienpruts joined #git
08:20 Andrew_K joined #git
08:21 _ikke_ anddam: What do you mean with which one is it?
08:22 _ikke_ Note that there is not really a thing called staging area in git (though we refer to it because that's how it looks like to users)
08:22 grawity `git diff --cached` *is* a diff between the staging area and the working tree (current files)
08:22 grawity _ikke_: close enough, just another name for the index
08:22 djb-irc joined #git
08:22 _ikke_ I would argue that the diff between the index and the working tree is what to users is the staging area
08:22 tobiasvl anddam: --cached shows changes staged for the next commit relative to the named commit (default is HEAD). if there is no named commit, and there also is no HEAD, then it shows all staged changes :P
08:23 grawity oh, it's between index and HEAD?
08:23 tobiasvl yeah
08:23 _ikke_ indeed, between the index and HEAD
08:23 anddam _ikke_: I mean which one of the scenario, that --cached was referring to the staging area or that they were similar but not exactly overlapped
08:24 tobiasvl anddam: the SO answer probably referred to the fact that it's a diff between HEAD (or another commit/treeish you give it) and the index
08:24 anddam ok, I now lack the definition of index
08:24 tobiasvl unless there is no HEAD or given commit, in which case it actually is the index
08:24 tobiasvl well, "staging area"
08:27 brillenfux joined #git
08:28 arnsa joined #git
08:28 tarkus joined #git
08:29 dpower joined #git
08:36 _ikke_ anddam: the index is a structure in git which it uses to build the next commit
08:37 _ikke_ anddam: any time you add a file (or changes), they are added to the index (as a snapshot, not a diff)
08:37 _ikke_ anddam: the staging area is just a concept, which is derrived from the index
08:38 Mutter joined #git
08:38 djb-irc joined #git
08:38 mikecmpbll joined #git
08:39 _28_ria joined #git
08:42 JeroenT joined #git
08:42 tnecniv joined #git
08:42 dikiaap joined #git
08:44 Balliad joined #git
08:45 brk_ joined #git
08:46 ahmedelgabri joined #git
08:46 tarkus_ joined #git
08:46 systemov` joined #git
08:51 theoceaniscool joined #git
08:54 marlinc joined #git
08:57 djb-irc joined #git
08:57 ahmedelgabri joined #git
08:59 willlondon joined #git
09:00 SwiftMatt joined #git
09:05 encod3 joined #git
09:08 Snugglebash joined #git
09:08 kfoonamalik joined #git
09:08 knot joined #git
09:10 tarkus_ How to ignore a folder only when it's found in the root directory?
09:11 roelmonnens joined #git
09:11 _ng joined #git
09:11 Zarthus tarkus_: .gitignore?
09:11 tarkus_ something like .gitignore ^data ... will this work?
09:11 Zarthus why not just `data` or `data/` or `data/.gitignore` with '*' in it
09:12 jcadduono start it with a / ? ex. /data
09:12 tarkus_ Zarthus, what if i have another folder with the same name, e.g. src/data/foo.js
09:12 Zarthus heh, I never really considered that type of collision could happen.
09:14 kurkale6ka joined #git
09:14 jcadduono so "/data" would ignore data/all/the/stuff but not /other/data/the/good/stuff
09:14 cbreak yes.
09:15 nettoweb joined #git
09:15 jcadduono i use the .gitignore in the linux kernel for some good examples
09:15 tnecniv joined #git
09:17 cbreak I am sure the linux people have some stupid ignore rules in there... they still don't want to accept zfs :(
09:19 ismithers joined #git
09:19 jmpp joined #git
09:19 ams__ joined #git
09:19 hahuang62 joined #git
09:19 Orion3k joined #git
09:20 knot left #git
09:20 tnecniv joined #git
09:21 a_thakur joined #git
09:21 alcohol thats because btrfs is obviously better
09:21 alcohol even the name implies that
09:21 tmk joined #git
09:21 alcohol :p
09:22 tmk i am getting error while using git send-email command
09:22 tmk Unable to initialize SMTP properly. Check config and use --smtp-debug. VALUES: server=smtp.gmail.com encryption=tls hello=fedora.localdomain port=587 at /usr/libexec/git-core/git-send-email line 1383.
09:23 tmk please answer if you know
09:24 mikecmpbll joined #git
09:24 spaceone joined #git
09:24 thief joined #git
09:26 TooL69 joined #git
09:27 spaceone how can i see a git-log of multiple commits in a squashed view?
09:27 spaceone e.g. (pseudo) git log --grep bug12345 --patch --squash
09:28 flaviodesousa joined #git
09:31 maskeraid joined #git
09:31 TooL69 left #git
09:32 mischat joined #git
09:32 maskeraid joined #git
09:32 Sasazuka joined #git
09:35 Sasazuka joined #git
09:36 ferr1 joined #git
09:37 NeverTired joined #git
09:38 maskeraid joined #git
09:39 a_thakur joined #git
09:40 _Vi joined #git
09:40 anddam _ikke_: I see, thanks
09:40 CheckDavid joined #git
09:43 marcogmonteiro joined #git
09:44 acetakwas joined #git
09:46 tnecniv joined #git
09:50 mazert joined #git
09:50 thesquib joined #git
09:52 clmsy joined #git
09:52 inflames joined #git
09:57 chll_ joined #git
10:00 a_thakur joined #git
10:01 iliv joined #git
10:06 tnecniv joined #git
10:06 Serpent7776 joined #git
10:07 Serpent7776 joined #git
10:08 _ikke_ spaceone: I don't think has a way to generate a single diff over multiple non-sequential commits
10:09 Serpent7776 joined #git
10:09 TomyWork joined #git
10:09 sdothum joined #git
10:11 a_thakur joined #git
10:11 JeroenT joined #git
10:15 spinningarrow joined #git
10:15 JeroenT joined #git
10:16 [0xAA] joined #git
10:16 spinningarrow My git is using the pager for commands like 'show' even though core.pager is unset. Is there some other place where this is set
10:16 spinningarrow ?
10:16 _ikke_ spinningarrow: You need to explicitly set it to false
10:16 _ikke_ not unset it
10:16 Olipro joined #git
10:16 _ikke_ git uses a pager by default
10:17 spinningarrow Is this a recent change? I don't remember ever setting it before
10:17 _ikke_ no, that's the behaviour for as long as I know
10:17 spinningarrow I tried git on another machine and git show isn't paginated and core.pager isn't set either
10:17 spinningarrow This is on macOS if it makes any difference
10:17 _ikke_ Might be
10:18 Virox spinningarrow: what version of git are you running?
10:18 moei joined #git
10:18 tmk joined #git
10:18 spinningarrow okay just to clarify - git show is paginated on both machines. However `git show --stat` is paginated on my machine but not the other one I tried
10:18 _ikke_ PAGER and GIT_PAGER environment variables also influence this
10:19 spinningarrow I'm using 2.10.2 on my machine. The other one is using 2.10.0
10:19 acetakwas joined #git
10:19 holdsworth joined #git
10:19 JeroenT_ joined #git
10:19 _ikke_ spinningarrow: sounds like a different LESS environment variable
10:19 spinningarrow _ikke_: both variables are unset on both machines :/
10:20 _ikke_ By default git sets LESS to not page when the output is less than one term
10:20 spinningarrow ahhhhh
10:20 spinningarrow You're so right
10:20 spinningarrow LESS on my laptop was set to `-R`
10:20 spinningarrow unsetting fixed it
10:20 _ikke_ FRX is the default that git uses
10:20 JeroenT_ joined #git
10:21 spinningarrow cool
10:21 spinningarrow I think I'll set mine to that
10:21 maskeraid joined #git
10:21 spinningarrow thanks people!
10:22 [0xAA] joined #git
10:22 neanderslob joined #git
10:25 tnecniv joined #git
10:25 compeman joined #git
10:27 netj joined #git
10:28 tusskan joined #git
10:29 a_thakur_ joined #git
10:29 subhojit777 joined #git
10:31 tsunamie joined #git
10:32 goTaN joined #git
10:32 prg318 joined #git
10:32 a_thakur_ joined #git
10:32 sangy joined #git
10:33 mischat_ joined #git
10:34 goTaN can I apply a patch even I only have the bare files and not a git repo?
10:34 bumpy joined #git
10:36 Guest86595 left #git
10:36 General joined #git
10:36 tobiasvl goTaN: you don't have a git repo? is this a git related question or are you just looking for patch(1)
10:36 treaki joined #git
10:37 _ikke_ goTaN: git apply works without a repo
10:37 General how to test whether my git config is able to send mail
10:37 goTaN yea
10:37 goTaN I have a git repo
10:37 goTaN but I "export" the complete project and give it to the customer
10:38 goTaN can I then create a patch from my project and give it to the customer so that they can apply it somehow?
10:38 goTaN maybe they create a bare repo and add the files to it, and then apply the patch?
10:39 a_thakur joined #git
10:39 tobiasvl you can't add files to a bare repo
10:39 goTaN something like that... so my actual quesiton is, can they apply a patch even if there repo is not cloned (or somehow related) from the repo where the patch is created
10:40 goTaN tobiasvl: yea sorry, the terminology was wrong
10:40 tobiasvl just use man git format-patch, give them the patch and tell them to apply it regularly (without git, but with patch)?
10:40 gitinfo the git-format-patch manpage is available at http://jk.gs/git-format-patch.html
10:40 ^7heo tobiasvl: it's funny how our brain comes up with working interpretations of impossible stuff just because IRC.
10:40 ^7heo tobiasvl: I actually read: create a bare repo, clone it, add the files to the clone and push it.
10:41 tobiasvl goTaN: or just pipe "git diff" to a file, and give them that file as a patch file! that's simpler.
10:41 ^7heo tobiasvl: but doesn't keep the author.
10:41 tobiasvl hehe ;)
10:41 ^7heo git format-patch --stdout would.
10:41 ^7heo but for that, ofc, one has to commit.
10:42 JeroenT joined #git
10:42 tobiasvl yeah. seems goTaN doesn't need author information though.
10:42 tobiasvl if the customer only has the flatfiles anyway
10:42 ^7heo one doesn't simply ignore the author of a git diff.
10:42 * ^7heo hides
10:42 JeroenT joined #git
10:42 tnecniv joined #git
10:42 diogenese joined #git
10:42 ^7heo but yeah, whatever works.
10:42 goTaN ok, so it really does not matter if the repo of the customer is a clone from the repo where the patch is created?
10:43 ^7heo git works with diffs anyway
10:43 goTaN I am wondering how that works...
10:43 goTaN yea that is why I dont understand it
10:43 tobiasvl goTaN: I thought your customer didn't have a clone
10:43 ^7heo if there's no conflict in the diff you're submitting... that will work.
10:43 goTaN tobiasvl: yea
10:43 tobiasvl or are we talking about ^7heo's crazy solution now :)
10:43 ^7heo goTaN: you can also use git diff and pipe that to patch -p1
10:44 ^7heo that will not require git at all on the receiving end, and I think that's what tobiasvl was thinking of.
10:44 tobiasvl yeah
10:44 goTaN tobiasvl: basically they dont have git, they just have a bunch of files. But I can tell them to create a git repo and to apply my patch. But I am confused how that will work
10:44 ^7heo so go for the solution from tobiasvl
10:44 ^7heo it will work like a charm.
10:44 ^7heo git diff, send that via whateveryouwant, tell them to patch -p1 and you're set.
10:44 goTaN hm even when there alot of changes?
10:45 goTaN on alot of files
10:45 ^7heo the problem isn't the lot of changes or files.
10:45 ^7heo the problem is the conflicts, if any.
10:45 goTaN yea
10:45 ^7heo if you're SURE that nothing/noone else than you changed those files at those offsets, you're good.
10:45 ^7heo (or approximate offsets since it includes context)
10:45 goTaN ok, that will help
10:47 converge joined #git
10:47 goTaN and what if I would use the format-patch? So in theory I can tell them to create a git repo with the files and then they just apply the patch I created to their new created repo?
10:47 goTaN would that work with conflicts?
10:47 ^7heo if you use format patch, it will really help in case there is *already* a git repo, but it's not worth it IMHO to create a git repo just to have the author of one diff.
10:48 jmpp joined #git
10:48 jmpp joined #git
10:48 ^7heo essentially, git has better conflict resolution than patch (obviously, or git wouldn't have been needed that much to begin with)
10:48 ^7heo but again, it really depends what they changed on their side.
10:48 goTaN well but how do they apply the patch when I do a git diff -p1?
10:49 ^7heo in both or either cases, you can't just drop changes on them and pray.
10:49 ^7heo p1 isn't a valid git diff option.
10:49 rnsanchez joined #git
10:50 goTaN well I am talking about ur solution, "you can also use git diff and pipe that to patch -p1"
10:50 ^7heo ah yes.
10:51 goTaN because I can make sure that there are no conflicts
10:51 ^7heo that is tobiasvl's solution, mine was to use "git format-patch --stdout"
10:51 Macaveli joined #git
10:52 goTaN ah, well so which one is the one to prefere>
10:52 murgi joined #git
10:52 ^7heo and that has to be applied with git apply
10:52 goTaN yea
10:52 ^7heo it depends on the situation.
10:52 goTaN but then they would need git and the files in a git repo right?
10:52 Silox| joined #git
10:52 goTaN well they need git anyway
10:53 tobiasvl no they don't need git anyway
10:53 goTaN ah ok
10:53 ^7heo patch isn't git patch.
10:53 ^7heo patch is patch
10:53 ^7heo git patch doesn't exist, there's git apply instead.
10:53 goTaN yea
10:53 tobiasvl (which is why I called it "patch(1)" earlier, just to make sure we were on the same page)
10:54 ^7heo patch is a POSIX utility (http://pubs.opengroup.org/onlinepub​s/9699919799/utilities/patch.html)
10:55 goTaN ok, so I do a git diff HEAD..someCommit | "dont know the syntax yet"
10:55 goTaN and they just apply the patch
10:55 goTaN ok
10:56 tobiasvl git diff HEAD..someCommit > patchfile.patch
10:56 tobiasvl is the syntax
10:56 tobiasvl and you give them the patch file and they apply it
10:56 goTaN yea thanks I would have looked that up
10:57 goTaN ok thank you guys you really helped me
10:57 tobiasvl np
10:57 davimore_ joined #git
10:58 mischat joined #git
11:00 tnecniv joined #git
11:03 los joined #git
11:04 los joined #git
11:04 zeks joined #git
11:05 zeks how do I say to git to stop tracking the files but in such a way that it wont delete them when I checkout that branch? after rm --cached/ .gitignore  the moment I switch to that bracnh ignored files get deleted
11:05 tobiasvl zeks: !ignore_tracked
11:05 gitinfo zeks: Git only applies ignore patterns to untracked files. You can't use ignore patterns to ignore changes to files that are already tracked by git. To remove files only from git, but keeping them on disk, use git rm --cached <file>. Still, see https://gist.github.com/1423106 for ways people have worked around the problem.
11:06 tobiasvl perhaps that link has some ideas
11:06 zeks damnit, our ISP is blocking github atm :(
11:06 Ekho- joined #git
11:06 absinthe joined #git
11:07 patte joined #git
11:08 tier2villain_ joined #git
11:08 achadwick joined #git
11:09 kfoonamalik joined #git
11:09 FilipNortic joined #git
11:10 yonatankoren joined #git
11:10 Mixologic joined #git
11:11 daishan joined #git
11:11 salparadise joined #git
11:11 Ekho joined #git
11:11 mischat_ joined #git
11:12 tsunamie joined #git
11:13 DieguezZ joined #git
11:14 jfr joined #git
11:14 compeman joined #git
11:15 zeks is there any reason I should not go with --assume-unchamged route?
11:16 canton7 yeah, lots of different operations (e.g. reset) will flip it back
11:16 zeks damn
11:16 selckin not clear what the problem is
11:16 canton7 it's meant as an optimization, not as a way of ignoring local changes to files, and git's usage of it reflects that
11:16 canton7 !config
11:16 gitinfo [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas
11:16 nkouevda joined #git
11:16 King_Hual joined #git
11:17 zeks problem is - I have a bunch of config files already in repo
11:17 zeks and I really need git to just forget about them
11:17 zeks but not delete them each time the brach changes
11:17 mischat joined #git
11:17 selckin delete them from all branches
11:18 zeks and?
11:18 selckin done?
11:18 zeks if I delete them then any checkout will delete files on disk too
11:18 selckin it will conflict
11:19 selckin if they don't have local changes, then why care
11:19 manuels joined #git
11:19 zeks after I deleted them on one branch with rm --cached, the momemnt a switched to anotehr branch and back it just deletes all those files
11:19 zeks even though they are in gitignore
11:20 selckin commit the delete
11:20 zeks ofc I did that
11:20 selckin delete them on the other branch too
11:20 thebope joined #git
11:20 particleGolem joined #git
11:20 selckin your story doesn't make sense because it should be refusing the checkout the other branch since that would overwrite files not tracked
11:21 zeks I think it remembers that they were deleted
11:21 hahuang62 joined #git
11:21 zeks and just does that
11:22 kryl99 joined #git
11:23 selckin https://paste.ee/p/GEnsf
11:24 zeks your test case doesnt deal with gitignore
11:24 selckin proof it
11:26 romerocesar joined #git
11:26 c1arknova joined #git
11:28 chll_ joined #git
11:30 eduardas_m joined #git
11:31 General joined #git
11:32 brillenfux_ joined #git
11:32 TomyWork joined #git
11:33 brillenfux__ joined #git
11:33 spb joined #git
11:34 Macaveli joined #git
11:34 theoceaniscool joined #git
11:36 Sceorem_ joined #git
11:36 buffal0 joined #git
11:38 brillenfux_ joined #git
11:38 Sceorem__ joined #git
11:39 General Anybody uses git here?
11:39 selckin no, hg is best
11:39 kryl99 joined #git
11:39 General hg??
11:39 Andrew_K General: good joke dude
11:40 ikonia I suspect most people in #git IRC channel will use git one way or another
11:40 General can you guys help me
11:40 brillenfux__ joined #git
11:41 selckin google 'smart qeustions' and read it
11:41 InfoTest joined #git
11:42 brillenfux joined #git
11:43 General I am not asking you blindly,i searched a lot but still i don't able to figure out
11:44 selckin some next level trolling
11:45 ahmed_elgabri joined #git
11:45 algun_ joined #git
11:46 ajpiano joined #git
11:46 brillenfux_ joined #git
11:49 General Andrew_K: i am a newbie
11:50 Andrew_K General: i'm new to git as well
11:51 buffffal00 joined #git
11:52 General Andrew_K: ok
11:52 JeroenT joined #git
11:52 zeks I guess deleting file on all branches is the only solution, except it's very ugly and git really, REALLY needs some simpler mechanism to "just stop tracking this please"
11:53 coachmanny joined #git
11:53 selckin no.
11:54 zeks selckin, yes. if you google the probelm, it is being asked a LOT and there is no satisfactory answer
11:54 General selckin: you use hg instead of git?
11:54 selckin General: go away
11:54 General what happen?
11:54 cbreak what problem?
11:55 zeks and the very fact that you can't find an answer of how to stop tracking without having problem when you pull later down the line clearly shows that tehre is i nfact a problem
11:55 cbreak zeks: oh, that's easy
11:55 cbreak stop tracking with git rm --cached
11:56 zeks cbreak, sadly, no
11:56 cbreak commit the "deletion"
11:56 zeks my steps:
11:56 cbreak pulling later will result in a merge conflict, as expected
11:56 cbreak so no problem there
11:56 zeks 1. rm --cached
11:56 zeks 2. gitgnore
11:56 cbreak (if the file was changed)
11:56 zeks 3. checkout of the branch tha tstill has the file tracked
11:56 zeks 4. checkout of te branch that ignores the file
11:57 zeks the files gets _deleted_
11:57 zeks silently
11:57 cbreak switching to a branch that tracks the file with an untracked file with different contents shouldn't work in the first place
11:57 cbreak git will give you an error
11:57 zeks see step 2
11:58 zeks the file is in gitignore so it switches silently without awarning
11:58 zeks and when you switch back it just gets deleted silently again
11:59 zeks I happened on this problem when I decided to delete .ini config files from my repo that were modified too frequently
11:59 jaguarmagenta joined #git
12:00 zeks it turns out I can't, not without having a ton of problems later when I switch branches
12:00 garethdaine joined #git
12:01 cbreak the problem you have seems to be caused by ignoring the file
12:01 jmpp joined #git
12:02 zeks the whole idea is that I want to ignore this file.....
12:02 cbreak git seems to ignore the file too good if you tell it to ignore it
12:02 zeks I want to say to git: "stop tracking changes in this file please"
12:02 Darcidride joined #git
12:02 cbreak git rm --cached does that
12:03 bartmon joined #git
12:03 cbreak (it will remove the file from any future commits)
12:03 zeks it will mark the file as untracked though
12:03 zeks each time I want to switch branches
12:03 cbreak yes... because stopping to track the file makes it untracked.
12:03 zeks making it a major nuisance
12:04 zeks imagine having 1X untracked files each time you want to switch
12:04 zeks this si clearly not manageable
12:05 andrew711 joined #git
12:05 bartmon hi. i have a file in my local branch which has 2 patches.  I'd like to revert one but keep the other. Does "git checkout -p file" support this interactive use-case like "git add -p" does?
12:06 Thorn joined #git
12:06 bartmon it seems so, i didn't read the enough of the man page!
12:09 cbreak bartmon: you can also git revert one commit
12:12 arnsa joined #git
12:12 patte joined #git
12:12 Flowinho joined #git
12:13 howitdo joined #git
12:13 elastix joined #git
12:16 Gloomy joined #git
12:16 [0xAA] joined #git
12:18 GavinMagnus joined #git
12:19 Learath2 joined #git
12:20 Akimb joined #git
12:20 dreiss joined #git
12:21 subhojit777 joined #git
12:23 CEnnis91 joined #git
12:23 ogny joined #git
12:23 denisMone joined #git
12:24 tvw joined #git
12:25 CheckDavid joined #git
12:26 mischat_ joined #git
12:32 twisted` joined #git
12:33 armin joined #git
12:33 zeroed joined #git
12:35 astrofog joined #git
12:35 YuGiOhJCJ joined #git
12:36 Naan joined #git
12:37 jozwior joined #git
12:38 AustinMatherne joined #git
12:39 KV joined #git
12:40 venmx joined #git
12:40 GavinMagnus joined #git
12:42 Darcidride joined #git
12:43 a_thakur joined #git
12:46 bouncy joined #git
12:46 skelterjohn joined #git
12:46 bumpy joined #git
12:47 Iacobus__ joined #git
12:48 General joined #git
12:49 manuels joined #git
12:49 anddam joined #git
12:50 ToBeCloud joined #git
12:50 dsdeiz joined #git
12:50 dsdeiz joined #git
12:53 mehdi___ joined #git
12:54 czart joined #git
12:54 tarkus joined #git
12:54 Macaveli joined #git
12:56 ingenious left #git
12:58 johnmilton joined #git
12:58 masuberu joined #git
12:59 roelmonn_ joined #git
13:00 danslo joined #git
13:00 tarkus_ joined #git
13:00 fees joined #git
13:01 Starky joined #git
13:04 isysd joined #git
13:05 wizonesolutions joined #git
13:08 systemov` joined #git
13:09 WaReZ joined #git
13:09 synthroid joined #git
13:10 kfoonamalik joined #git
13:11 QuestionTime joined #git
13:15 henriquev joined #git
13:16 jknetl joined #git
13:17 JeroenT joined #git
13:17 nkouevda joined #git
13:18 clmsy joined #git
13:20 Macaveli joined #git
13:20 GavinMagnus left #git
13:21 willlondon joined #git
13:22 hahuang62 joined #git
13:22 thebope joined #git
13:22 lindenle joined #git
13:25 zeller joined #git
13:27 angular_mike_ joined #git
13:28 elmcrest joined #git
13:29 SteveCharleston joined #git
13:31 ash_workz joined #git
13:32 jmpp joined #git
13:33 teadrop_ joined #git
13:35 beatdown joined #git
13:35 richardlitt joined #git
13:36 CrazyMonkey201 joined #git
13:36 Awesomecase joined #git
13:36 theoceaniscool joined #git
13:36 thunderbirdtr joined #git
13:37 _Vi joined #git
13:38 CrazyMonkey2011 joined #git
13:40 Surfer2010 joined #git
13:40 patte joined #git
13:41 fmcgeough joined #git
13:41 zeroed joined #git
13:41 Anja joined #git
13:43 treehug88 joined #git
13:44 roelmonnens joined #git
13:44 onmeac joined #git
13:45 mischat joined #git
13:45 onmeac joined #git
13:46 sunri5e joined #git
13:48 unbalanced joined #git
13:48 Snugglebash joined #git
13:49 jimi_ joined #git
13:52 tarkus joined #git
13:53 onmeac joined #git
13:53 _Cyclone_ joined #git
13:54 lamont joined #git
13:58 ToeSnacks joined #git
13:58 Gloomy joined #git
13:59 masuberu joined #git
14:01 jaguarmagenta joined #git
14:03 derk0pf joined #git
14:04 rnsanchez joined #git
14:04 intellix joined #git
14:05 howdoi joined #git
14:06 Snugglebash joined #git
14:06 mischat joined #git
14:07 joco42 joined #git
14:07 Zamerick joined #git
14:08 converge joined #git
14:09 leeN joined #git
14:09 sbulage joined #git
14:10 cdg joined #git
14:14 sshine joined #git
14:14 dikiaap joined #git
14:15 sshine hi. I'm trying to git pull from an upstream repo and I get the error that I must specify the branch then. I'd like to have a default branch, but how?
14:15 ExoUNX joined #git
14:17 clearcut joined #git
14:17 sshine ah, I guess 'git pull --rebase upstream master' works quite easily.
14:17 chamunks- joined #git
14:17 pdgwien- joined #git
14:18 a_thakur joined #git
14:18 osse sshine: doesn't the error message tell you how to configure that?
14:19 Derderderd joined #git
14:19 Gsham joined #git
14:20 anonymuse joined #git
14:20 Vauff joined #git
14:21 Orphis joined #git
14:23 Werring joined #git
14:23 pepijndevos joined #git
14:23 dminuoso joined #git
14:23 tabakhase joined #git
14:23 loglaunch joined #git
14:23 ThiefMaster joined #git
14:23 selckin joined #git
14:23 obiwahn joined #git
14:24 vishal joined #git
14:24 noctux joined #git
14:24 DelphicOkami joined #git
14:24 Someguy123 joined #git
14:24 eduardas_m joined #git
14:25 Silex joined #git
14:25 Sonderblade joined #git
14:25 rvolosatovs joined #git
14:25 ShekharReddy joined #git
14:26 dcpc007 joined #git
14:28 tmsmith joined #git
14:29 bauruine joined #git
14:30 clearcut left #git
14:35 breezy_ joined #git
14:35 armin so my company is considering git, finally. unfortunately they consider building a git server "with high-availability". am i the only one thinking this clearly doesn't make any sense?
14:35 tobiasvl armin: how so
14:36 armin tobiasvl: where's the point?
14:36 tobiasvl armin: what do you mean?
14:36 tobiasvl they want a central repo, which is understandable for a company, and they want it to be available at all times?
14:36 armin tobiasvl: it's decentralized and fully usable while "the main server" is down anyways, so why enter the evil depth of HA for this?
14:36 tobiasvl how is HA evil?
14:37 armin it adds additional complexity, possible split brains, manual interventions in case of those, ...
14:37 armin really, where's the point? git is decentralized by design, why consider HA?
14:37 netj joined #git
14:37 tobiasvl and using git completely decentralized does not lead to split brains and manual interventions?
14:37 tobiasvl I don't understand what you mean. why does github exist?
14:37 Gonzo89 joined #git
14:37 _ikke_ armin: Because most people do not know how to work in a non-decentralized way with git
14:38 armin _ikke_: that's my thought.
14:38 sshine osse, it just asks to specify a branch on the command-line, but not exactly how. I just went with that instead of supplying a default branch.
14:38 tobiasvl and even if they do, it's understandable that a company don't want that. the company wants to deploy from a central repo. continuous deployment, most likely.
14:38 _ikke_ If you have all your PRs on a centralized git hosting, than that being offline causes a big inconvenience
14:38 tobiasvl they want pull requests in one place, centralized approvals/gateways for changes to master, etc
14:39 pypi_ joined #git
14:39 Es0teric joined #git
14:39 tobiasvl I love git being decentralized but mostly because it means I can work locally with the repo offline, I don't want to exchange patches and branches with a myriad of colleagues decentralized and then try to merge all this upstream when the main repo decides to come back up
14:39 armin i have a feeling adding another layer of complexity on top of git just so you have HA is a very crazy thing.
14:40 _ikke_ armin: Note that such a server provides more then just repository hosting
14:40 baryon joined #git
14:40 armin _ikke_: depends on the server i would say.
14:40 _ikke_ it provides a web interface, user authentication, CI/CD
14:40 _ikke_ armin: of course
14:40 venmx joined #git
14:40 osse sshine: ah ok, it's:  git branch -u origin/master master  (for example)
14:41 armin in the simplest form it would be something that creates a bare repository, accessible via SSH anyways.
14:41 synthroi_ joined #git
14:41 _ikke_ armin: Right, but chances are the company is going to want more than just that
14:41 jaguarmagenta joined #git
14:41 armin so really, where's the point in putting all the pitfalls on top of this?
14:42 armin _ikke_: unfortunately that's likely true, yeah.
14:42 _ikke_ We use gitlab here, which really provides value imo
14:42 GodGinrai joined #git
14:43 armin _ikke_: i actually wrote my own git server which is nothing more than a shellscript being run by a command= stanza in an authorized_keys file.
14:43 armin _ikke_: and if you really want to know why: because i consider gitlab bloated.
14:43 _ikke_ armin: check !gitolite
14:43 gitinfo armin: 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/
14:43 perlpilot armin: gitolite is nice
14:43 _ikke_ I think you will like that if you are not opposed to perl either
14:44 armin that's what i've been using for years. i considered it counter-intuitive, bloated and conceptionally weird.
14:44 armin so i wrote my own.
14:44 mbeierl1 left #git
14:44 armin and yes i hate perl.
14:44 _ikke_ Then you are beyond help ;-)
14:44 armin :(
14:46 jmpp joined #git
14:46 AaronMT joined #git
14:47 eduardas_ joined #git
14:47 _ikke_ Just accept that most people want more than the bare minumun :)
14:47 _ikke_ minimum*
14:47 _ikke_ makes your life much easier
14:48 Starky joined #git
14:49 synthroid joined #git
14:50 tobiasvl armin: if you consider gitlab bloated then I think the HA part isn't the one you will be least satisfied with in a central setup
14:50 uranellus joined #git
14:50 uranellus joined #git
14:51 griffindy joined #git
14:51 sharms joined #git
14:53 anonymus_ joined #git
14:54 a_thakur joined #git
14:54 thebope joined #git
14:55 Cavallari joined #git
14:55 armin oh wow. so i need to take care of redis and postgres being HA-ready just so users can access that web-interface while one of both gitlab-nodes is down, while git already allows users to continue to work if that happens, by design?
14:55 armin i have major problems accepting that this just "exceeds the bare minimum".
14:56 rishi joined #git
14:56 armin to me that seems like opening a whole can of worms just "because we do HA"...
14:57 armin consider me headstuck, but i really fail to see the point here.
14:57 tobiasvl define "continue to work". as _ikke_ said, gitlab and other solutions offer more than just git.
14:57 armin tobiasvl: that's why i pointed out that gitlab will additionally need postgres and redis...
14:58 tobiasvl yes, but git does not allow users to continue to do everything that gitlab offers, by design
14:58 armin but yeah, my dream of reduced complexity when switching to git was just disturbed by having to accept that corporate needs exceed mine... ;)
14:58 armin true.
14:58 tobiasvl hehe, exactly
14:59 nickabbey joined #git
15:00 romerocesar joined #git
15:01 rishi joined #git
15:01 steelsw0rd joined #git
15:02 tobiasvl even linux and git itself are dependent on infrastructure around git, even if that's a mailing list and a central repo that can more easily be swapped out if it goes down
15:03 osse "The bureaucracy is expanding to meet the needs of the expanding bureaucracy"
15:03 tobiasvl indeed – but stuff like CI/CD and a system around peer review/approvals on PRs are good things IMO
15:04 DMorg joined #git
15:04 osse oh i was just quoting for teh lulz
15:04 tobiasvl hehe
15:04 osse git is nice, but is only a piece of the larger whole imho, albeit an important one
15:06 e14 joined #git
15:06 HoierM joined #git
15:07 GodGinrai joined #git
15:07 DMorg Sorry for the noob question. I have a master branch and a feature branch. I've made commits to the master that I want to incorporate into the feature branch. Is this a merge operation?
15:07 tobiasvl DMorg: yes
15:07 GodGinrai DMorg: yes.
15:08 GodGinrai Anytime you want the contents of one branch in another existing branch, it is a merge
15:08 tobiasvl GodGinrai: weeeell, cherry-pick and rebase exist
15:09 GodGinrai tobiasvl: that's true
15:09 GodGinrai tobiasvl: I forgot about cherry-pick
15:09 GodGinrai rebase is essentially a merge (you can even tell it to generate a merge commit)
15:09 DMorg so i do merge from master into feature branch?
15:10 GodGinrai just not a normal merge
15:10 tobiasvl DMorg: yes
15:10 GodGinrai DMorg: yes
15:10 jmpp joined #git
15:10 jmpp joined #git
15:10 GodGinrai DMorg: and then if there are any conflicts, git will tell you, and then you can resolve them and complete the merge
15:10 DMorg thanks very much
15:10 GodGinrai DMorg: and then if there are any conflicts, git will tell you, and then you can resolve them and complete the mergehttps://git-scm.com/book/en/v2/Git-Branching​-Basic-Branching-and-Merging#_basic_merging
15:11 DMorg been trying to read up on git but sometimes you just need a human to tell you you're not about to do something stupid
15:12 kfoonamalik joined #git
15:12 GodGinrai DMorg: the site I linked is probably one of the best resources for getting up to speed on git you can read.
15:12 ojdo joined #git
15:13 star_prone joined #git
15:14 notze joined #git
15:14 raijin joined #git
15:14 notze hey guys, can i somehow autoupdate subtrees on checkout?
15:14 notze on clone
15:15 osse notze: checkout or clone?
15:15 notze clone
15:15 osse --recursive
15:15 notze everythime on clone get the latest version
15:16 tobiasvl the latest? you want to make a new commit when you clone that updates all the subtrees to a new rev?
15:16 notze tobiasvl, yes more or less
15:17 notze i want an essential repository shared with all my git repos and this will checkout the latest essential version to all
15:19 GodGinrai You use the word essential, but I'm not sure you understand what it means
15:19 GodGinrai How about you run us through your use-case with a hypothetical
15:19 GodGinrai ?
15:20 notze GodGinrai, i dont search a new name for my modules, i want to know how to update subtree automatically to latest version
15:20 notze https://help.github.com/articles/abou​t-git-subtree-merges/#platform-linux
15:20 vrlx joined #git
15:20 notze this tells me to manually do it for eahc repos
15:20 converge joined #git
15:20 converge joined #git
15:20 notze thats not what we're looking for
15:21 GodGinrai notze: I know what you *want*, I don't know why.
15:21 GodGinrai notze: I'm fairly sure this is an !xy problem
15:21 gitinfo notze: Woah, slow down for a bit. Are you sure that you need to jump through that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...
15:22 notze GodGinrai, we have one modules repository that should be in all our project repos included always in the latets version
15:22 GodGinrai notze: so if you have repos A, B, and C, and module repo M...
15:23 notze no
15:23 GodGinrai notze: A, B, and C all have M as a subtree?
15:23 hahuang62 joined #git
15:23 notze i have repos A and B
15:23 notze B is cotainer for all shared modules
15:23 tobiasvl submodules work like this: the main repo (superrepo?) needs to know and record what commit each submodule is at
15:23 notze and a should on clone always get latets version from B
15:23 osse tobiasvl: submodule != subtree.  I got it wrong with my answer
15:23 notze tobiasvl, submodules we used before but the subtree looked more intresting
15:24 tobiasvl oh hehe
15:24 Semtex-H joined #git
15:24 tobiasvl don't know how I managed to read that wrong every time it was mentioned – probably the use of "modules"
15:24 tobiasvl carry on then
15:24 notze problem with submodules is the more complicated handling and same problem with updating the ref
15:24 GodGinrai notze: so "all our project repos" means one repo, A?
15:24 notze no
15:24 notze 20 diffrent all linking to B
15:25 notze any client project
15:25 GodGinrai so then wouldn't A, B, and C be accurate? (since it shows there is more than one repo that uses module repo M)
15:26 notze so
15:26 notze do you know a solution
15:27 Es0teric joined #git
15:27 finalbeta joined #git
15:27 GodGinrai notze: not yet.  I'm trying to make sure I understand what you are trying to do so I can think of one.
15:27 notze we have repos a -Z and repo 007
15:28 notze repo a-Z should always on clone get the latest version from 007
15:30 GodGinrai notze: I'm thinking your best bet might be to get your team to adopt an alias or script that clones and updates that they can use instead of just cloning.  Or are you wanting this to do the same for your clients as well?
15:30 notze so you want to say git is not able to offer a solution to this taskl
15:31 Orion3k joined #git
15:31 mischat_ joined #git
15:31 snowkidind joined #git
15:31 Nugget joined #git
15:32 GodGinrai notze: I won't say that.  I haven't used subtrees much, so if I did, I might be misinforming you.  However, since the others here don't seem to know a solution either, I'm thinking it is likely.
15:33 lordjancso joined #git
15:33 cbreak git clone always gets the latest.
15:34 guardian joined #git
15:34 GodGinrai cbreak: he wants the latest code from the modules repository in his subtree.  You're saying that git clone does that automatically?
15:34 cbreak he's using git submodules?
15:34 GodGinrai no
15:34 GodGinrai he's using subtrees
15:35 cbreak what's that?
15:35 cbreak the subtree merge thing?
15:35 cbreak or some third party extension?
15:35 cbreak subtree merging doesn't involve separate repositories when you clone
15:35 cbreak you will get 100% what was on the remote you clone
15:36 GodGinrai cbreak: Actually, subtrees can have remotes
15:36 cbreak remotes are not cloned
15:36 GodGinrai cbreak: and that's what he's wanting
15:37 cbreak they are not needed to be cloned
15:37 jason237 joined #git
15:37 cbreak it seems he doesn't want subtree merging
15:37 cbreak more like submodules with pull config
15:37 thecomedian joined #git
15:37 cbreak or even a script that clones sub repositories like google's repo
15:38 GodGinrai cbreak: I agree
15:38 garethdaine joined #git
15:40 al-damiri joined #git
15:41 Snugglebash joined #git
15:43 Ranhir joined #git
15:43 a_thakur joined #git
15:44 Ranhir joined #git
15:46 thebope joined #git
15:48 subhojit777 joined #git
15:48 dsdeiz joined #git
15:48 dsdeiz joined #git
15:49 nickabbey joined #git
15:50 denisMone joined #git
15:51 regedit joined #git
15:53 nickabbey joined #git
15:53 pypi_ joined #git
15:57 nettoweb joined #git
15:57 chardan joined #git
15:58 al-damiri joined #git
15:58 ekneuss joined #git
15:58 vrlx joined #git
15:58 nickthetait joined #git
15:59 peeps[lappy] joined #git
16:00 jaafar joined #git
16:00 ekneuss Hello there. I'm in the following scenario: I merged branch A into branch B, which caused some conflicts that I fixed and commited. Several commits later on B, it turns out that the way to fix the conflict was not correct, what is the best approach to resolve this?
16:00 vrlx joined #git
16:00 ekneuss For instance, if I revert the merge commit and merge the branch again, would the conflict reappear as such? I would assume it is going to tell me: everything up to date.
16:02 osse ekneuss: the easiest and one you should seriously consider: own up to your mistake, make a new commit that fixes the problem and explain it in the message
16:02 osse reverting merges can be troublesome
16:02 maskeraid joined #git
16:03 GodGinrai Yea, I would just make a new commit explaining that the conflict resolution was not the best approach
16:03 ekneuss That makes sense, and sounds the simplest. But it also means the fixes are going to be fully manual.
16:04 osse ekneuss: so will the next attempt at merging it
16:04 osse are you a single developer? what i mean is, can you push --force and wreck havoc at will?
16:04 Guest12386 joined #git
16:04 ekneuss no, push -f is not an option here
16:05 GodGinrai plus, that's not a good habit to get into :P
16:06 ekneuss I was wondering if by some magical option I would be able to merge a branch as if it was never merged before and get a new conflict that I could then resolve correctly.
16:06 osse ekneuss: I don't think the advantage of being able to re-merge it outweighs the problems with reverting it in the first place. so i'd do a manual fixup
16:06 Salander27 joined #git
16:06 roelmonn_ joined #git
16:06 utrack joined #git
16:06 osse ekneuss: you could create a new tmp branch at the point immediately before the merge on B, then merge A into that.
16:07 ekneuss right, but B is what I care about, so the state of B is always going to contain that previous, and incorrect, merge
16:07 osse ekneuss: then perhaps go back to B, and do  git checkout -p tmp -- affected files
16:07 kfoonamalik joined #git
16:07 ekneuss mmh I see
16:07 GodGinrai ekneuss: You *can* revert the merge if you want, using the -m flag
16:08 ekneuss osse, that sounds like the simplest approach! I will try that
16:08 ekneuss osse, GodGinrai, thanks!
16:08 GodGinrai ekneuss: but beware, the -m flag has its idiosyncrasies
16:08 osse !revert_merge > ekneuss
16:08 gitinfo [!faulty_merge] If you have published a merge commit that turned out to be faulty and you'd like to get rid of it, you're probably not going to have a good time. Here's the full story: https://raw.github.com/git/git/master/Docu​mentation/howto/revert-a-faulty-merge.txt
16:09 qqx Another approach for that would be to redo the merge as suggested by osse, then use `git reset <original-merge-id>`, then commit the resulting changes, and cherry-pick the resulting commit.
16:10 tnecniv joined #git
16:12 kbs2 joined #git
16:14 kbs2 left #git
16:14 iei joined #git
16:14 kbs joined #git
16:15 kbs By any chance - is there a public "single-use-throwaway" git hosting service? (ie, more like pastebin rather than github; where I could git push instead of pasteing some text.)
16:15 canton7 kbs, gist.github.com ?
16:16 bremner gist?
16:16 kbs canton7: does it let me push a full git repo? (I assumed they were just single files)
16:16 canton7 kbs, yep
16:16 kbs ah, thanks
16:16 canton7 gists are git repos. they can be cloned, and pushed to
16:16 danslo kbs: a gist can consist of multiple files, there's also a bunch of commandline tools that help you around that
16:16 saml joined #git
16:17 Derderderd joined #git
16:17 overlord_tm joined #git
16:17 kbs awesome - thanks all
16:17 danslo (this one's popular: https://github.com/defunkt/gist - see the examples)
16:18 PioneerAxon joined #git
16:18 GavinMagnus joined #git
16:18 theoceaniscool joined #git
16:18 menip joined #git
16:19 patte joined #git
16:20 CrazyMonkey201 joined #git
16:20 dreiss joined #git
16:21 CrazyMonkey201 left #git
16:21 star_prone joined #git
16:23 star_prone joined #git
16:24 star_prone joined #git
16:26 angelo_ joined #git
16:27 snits joined #git
16:28 Ekho- joined #git
16:29 afuentes joined #git
16:30 LionsMane joined #git
16:30 kbs So, it appears that I cannot push to an anonymous gist - is that correct? (test: git remote add origin git@gist.github.com:2df0d99d68e​f2d28e81f860c229ec6ca.git; git push origin master. gist == https://gist.github.com/anonymous/​2df0d99d68ef2d28e81f860c229ec6ca)
16:30 Snugglebash joined #git
16:30 tsunamie joined #git
16:31 daishan joined #git
16:31 boombatower joined #git
16:31 synthroid joined #git
16:32 canton7 kbs, I would have though that's correct - how does it know to let you push to it if it's anonymous?
16:32 dreiss joined #git
16:32 a_thakur_ joined #git
16:32 canton7 kbs, ... but what's the output from git?
16:32 Macaveli joined #git
16:33 kbs well - so at the top-bit, that's the functionality I'd like - an anonymous (paste-bin-unlogged-in style thing) but to answer your question: "Permission denied (publickey).... etc."
16:33 kbs
16:34 kbs (I realize I can log in to github and use a logged-in gist, but then I may as well use a regular repo)
16:34 canton7 well, apart from the way in which gists are displayed, can be "secret" (non-browsable), etc
16:35 GavinMagnus left #git
16:35 canton7 but yeah, I don't know of any provider which gives you anonymous throwaway git repos you can push to
16:35 GodGinrai kbs: http://stackoverflow.com/questions/2648​4337/upload-a-file-to-a-gist-with-bash
16:35 aarobc joined #git
16:36 steelsw0rd joined #git
16:36 kbs *nod* gotcha. thanks canton7
16:36 miczac joined #git
16:36 GodGinrai this does it for a file, but I'm sure you can get an idea from that on how to do it using git
16:36 jstimm joined #git
16:37 GodGinrai kbs: this might also be useful: https://developer.github.com/v3/gists/
16:37 e14 joined #git
16:38 kbs Thanks GodGinrai. At first glance all that seems a bit sideways from what I'm looking for, but will poke around for anything - appreciate the pointers.
16:38 GodGinrai kbs: from what I can tell, you should be able to create the anon gist by POSTing to /gists, then take the url it returns and use that with git to properly push your code to it
16:41 venmx joined #git
16:41 miczac joined #git
16:43 GodGinrai kbs: I just did a post to the api, and it does in fact return a "git push url"
16:43 GodGinrai which should be suitable for your purposes
16:44 Ardethian\ joined #git
16:44 kbs GodGinrai: were you actually able to push to it?
16:44 kbs https://gist.github.com/anonymous​/cde67ba439fa50be1191a78877ca6e70
16:44 GodGinrai kbs: I'll try now
16:45 ash_workz joined #git
16:45 rbern joined #git
16:46 jnavila joined #git
16:47 Dougie187 joined #git
16:47 garethdaine joined #git
16:49 [0xAA] joined #git
16:49 tang^ joined #git
16:50 ahmedelgabri joined #git
16:52 kbs GodGinrai: fwiw - I get https://gist.github.com/anonymous​/d3807058dfe16b9cca3d6be0cbdc2ac4 -- (I bail out at the credentials step, and empty answers here give me no joy either) but if you/someone see a way, would love to learn how. *heigh-ho heigh-ho off to work*
16:52 regedit joined #git
16:53 kadoban joined #git
16:54 robotroll joined #git
16:55 saml how can I clone a repo's one directory at the HEAD only?
16:55 kulelu88 joined #git
16:55 saml instead of entire HEAD,  is there a way to clone specific sub directory?
16:56 saml i googled hehehehhe
16:57 los joined #git
16:58 mario_ joined #git
16:59 canton7 saml, no
16:59 Shentino joined #git
16:59 saml there's sparse checkout
16:59 Es0teric joined #git
16:59 canton7 that doesn't avoid cloning everything though - it just controls what is checked out
16:59 saml ah i see
16:59 saml so clone --depth=1 is what i want?
16:59 cagedwisdom joined #git
16:59 saml to minimize cloning stuff
16:59 Snugglebash joined #git
16:59 canton7 dunno, do you want all of thie history or not?
17:00 saml i just want few image files in another repo during building this app
17:00 canton7 --depth=1 is probably best, then
17:03 JrCs joined #git
17:04 dikiaap joined #git
17:05 Xarodon joined #git
17:05 saml thanks
17:06 GodGinrai joined #git
17:06 netj joined #git
17:06 barq When I do git submodule update --init nothing happens, only a dir is created with a submodule and a .git file is created, but nothing is downloaded
17:07 barq how can I get more logging or fix this?
17:07 barq Normal clone works, so it's not my network
17:07 Derderderd joined #git
17:07 CheckDavid joined #git
17:08 GavinMagnus joined #git
17:09 canton7 joined #git
17:09 howitdo joined #git
17:10 Shentino joined #git
17:10 paperziggurat left #git
17:13 pbrewczynski joined #git
17:14 los joined #git
17:15 nickabbey joined #git
17:17 checkit joined #git
17:17 ikonia win 10
17:17 ikonia oops
17:18 theoceaniscool joined #git
17:19 lacour joined #git
17:20 Zamerick joined #git
17:21 Ordentlig joined #git
17:21 tarkus joined #git
17:23 hahuang62 joined #git
17:24 zza joined #git
17:25 GodGinrai joined #git
17:25 Shentino joined #git
17:27 SwiftMatt joined #git
17:28 jordila joined #git
17:28 a_thakur joined #git
17:29 synthroi_ joined #git
17:29 madewokherd joined #git
17:30 garethdaine joined #git
17:31 albel727 joined #git
17:31 cdown joined #git
17:31 mischat joined #git
17:32 gemie joined #git
17:32 gemie hi, cant I do a rebase interactive if there is no tracking on the branch?
17:33 gemie when doing git rebase --interactive I get an error saying "There is no tracking information for the current branch.
17:33 gemie Please specify which branch you want to rebase against."
17:33 nikivi joined #git
17:33 grawity you can, but you need to specify which branch to rebase against
17:33 gemie its a local branch that I want to clean up.
17:33 grawity it's still branched off *something*, isn't it?
17:34 grawity though, rebase doesn't strictly require a branch
17:34 Neiz joined #git
17:34 gemie ye it should be of develop.
17:34 Snugglebash joined #git
17:34 grawity any old commit will do, and you'll be rebasing commits since that one
17:34 gemie grawity: why does it then say it? I have never had a problem with rebase interactive, done it 100 times.
17:34 gemie but first I thought something wassup with emacs-magit because I got errors, then tried it manual from terminal and got that error.
17:34 gemie how can I check?
17:34 grawity it says that because it has to say that
17:35 Snugglebash joined #git
17:35 grawity there is no 'tracking information' so it doesn't know which branch to rebase against
17:35 gemie what I can see all other branches is like feature/foo, release/bar etc, mine is just foobar
17:35 grawity doesn't matter
17:35 gemie grawity: how can I add that?
17:35 gemie weird that it doesnt have tho, it's based on develop some time ago.
17:35 durham joined #git
17:35 Snugglebash joined #git
17:36 gemie but anyway, rebase interactive is a "local thing"?
17:36 grawity yes
17:36 jordila say i have  master, mybranch. Created child 'mysubbranches1,2...' from parent 'mybranch'. Am i wrong saying that deleting mybranch deletes its child mysubbranches1,2... too ?
17:36 gemie grawity: what should I do? :)
17:36 Snugglebash joined #git
17:36 grawity any sort of rebase, and much of everything in git, is a "local thing"
17:36 grawity you should tell `git rebase` which branch to rebase against
17:37 grawity "develop", in this case
17:37 jccn joined #git
17:37 gemie ok so I don't need to set an upstream branch to develop? git rebase develop --interactive ?
17:37 onmeac joined #git
17:37 Snugglebash joined #git
17:37 grawity that should work
17:37 gemie let me try
17:38 Gsham joined #git
17:38 Snugglebash joined #git
17:39 nkouevda joined #git
17:39 Snugglebash joined #git
17:39 Snugglebash joined #git
17:40 Snugglebash joined #git
17:41 Snugglebash joined #git
17:41 barq Is there any way to show the % complete when doing git submodule update, which does a clone?
17:42 Snugglebash joined #git
17:43 thebope joined #git
17:43 jordila left #git
17:45 gemie grawity: worked, thanks!
17:47 grawity barq: it already shows the same output as with regular clones
17:47 barq No, it doesn't show percent
17:48 barq It shows cloning ... and then at some point cloning the next, but now it is stuck for an hour cloning one and I don't know whether it crashed or is still working
17:48 grawity more likely because you're cloning over dumb-HTTP
17:49 barq can I clone the repo manually to skip this step?
17:49 barq How can I tell what protocol I am using?
17:49 grawity I mean, try with GIT_CURL_VERBOSE=1, if you see a ton of requests collecting the repo piece by piece it's dumb-HTTP and it doesn't know what "100%" *is*
17:49 grawity (that would mean that the server's admins have not configured git-http-backend correctly)
17:49 sdothum joined #git
17:50 GodGinrai this is why ssh is the way to git
17:52 aruns joined #git
17:52 timothy joined #git
17:52 barq Yeah, it looks like it's stuck
17:52 barq upload completely sent off: 9996 out of 9996 bytes < HTTP/1.1 200 OK
17:53 barq Can I clone the repo manually into the externals dir?
17:54 barq I can see the name of the submodule but not the repository
17:55 lindenle joined #git
17:55 grawity eh, I think you can just clone it directly into the (empty) placeholder directory
17:56 barq How do I know the remote?
17:56 barq Right now the dir does not exist
17:57 mablae joined #git
17:57 los joined #git
17:57 brokensyntax joined #git
17:57 albel727 joined #git
17:57 barq got it, .gitmodules
17:58 TSS_ joined #git
17:58 barq Now it works, but does not work with git subdmodule update
17:59 shgysk8zer0 joined #git
18:01 star_prone joined #git
18:01 nkouevda joined #git
18:02 khfeng_ joined #git
18:02 lindenle joined #git
18:03 austin987 joined #git
18:04 thebope joined #git
18:05 nicksloan joined #git
18:05 nickabbey joined #git
18:06 tanuki joined #git
18:06 thebope joined #git
18:07 dikiaap joined #git
18:08 hfp joined #git
18:11 mda1 joined #git
18:11 converge joined #git
18:15 BrownBear joined #git
18:16 BSaboia joined #git
18:16 aartist joined #git
18:17 dreiss joined #git
18:18 Groscheri joined #git
18:18 hahuang62 joined #git
18:20 m0viefreak joined #git
18:26 dunpeal joined #git
18:26 Guest37120 joined #git
18:28 derk0pf joined #git
18:30 dunpeal Hi. I have a feature branch foo that diverged from master N commits ago. How do I view, in my difftool, all the changes that foo introduces?  To clarify, this is not a simple diff, since I'm not interested in any changes made by master since the merge base.
18:30 _ikke_ dunpeal: Not sure how it works with your difftoolk
18:30 kyan joined #git
18:30 vrlx joined #git
18:30 _ikke_ dunpeal: but git diff master...foo
18:30 _ikke_ Note the 3 dots
18:31 nivag joined #git
18:31 Hello71 that's not ...
18:31 * j416 〆(... )
18:31 intellix joined #git
18:31 _ikke_ ...
18:31 Hello71 git diff master..foo
18:31 dunpeal _ikke_: so that means "diff between foo and its merge-base with master"?
18:31 j416 Hello71: three dots is correct.
18:31 _ikke_ Hello71: It should be *3* dots
18:32 Limes_ joined #git
18:32 Pseudocrat joined #git
18:32 Pseudocrat joined #git
18:32 _ikke_ dunpeal: correct
18:32 Guest55 joined #git
18:32 kadoban Note that it means something different from what ... means elsewhere, unfortunately.
18:32 joshszep joined #git
18:32 Hello71 "I'm not interested in any changes made by master since the merge base."
18:32 _ikke_ because dots with diff usually do not make any sense anyway
18:32 * Hello71 grumbles
18:33 kadoban Yeah
18:33 dunpeal OK, this is super useful, thanks.
18:33 j416 Hello71: "git diff foo..bar" is equivalent to "git diff foo bar"
18:33 Hello71 that's totally consistent.
18:33 dunpeal And thanks for inventing Git!
18:33 Hello71 /s.
18:33 j416 Hello71: i.e. just the tips
18:33 nickabbey joined #git
18:34 rominronin joined #git
18:34 svm_invictvs joined #git
18:35 kier joined #git
18:36 sbeller_ joined #git
18:37 garethdaine joined #git
18:39 Derderderd joined #git
18:41 zzz joined #git
18:41 dreampwnzor joined #git
18:41 anddam joined #git
18:41 venmx joined #git
18:44 aard_ joined #git
18:44 einnjo joined #git
18:44 nivag joined #git
18:46 e14 joined #git
18:47 vdv joined #git
18:47 rominronin joined #git
18:47 robotroll joined #git
18:48 LionsMane joined #git
18:49 nickabbey joined #git
18:49 fatalhalt joined #git
18:50 tnecniv joined #git
18:52 Sample joined #git
18:52 [0xAA] joined #git
18:52 ahmedelgabri joined #git
18:53 zzz joined #git
18:54 nd_ joined #git
18:54 jrnieder joined #git
18:55 nivag joined #git
18:55 [0xAA] joined #git
18:56 SteveCharleston joined #git
18:56 rominronin joined #git
18:57 raspado joined #git
18:58 tvw joined #git
18:58 _Vi joined #git
18:58 [0xAA] joined #git
19:00 mda joined #git
19:00 Balliad joined #git
19:01 railssmith joined #git
19:01 nicksloan joined #git
19:02 romerocesar joined #git
19:02 ahmed_elgabri joined #git
19:03 thebope joined #git
19:04 cyan__ joined #git
19:05 rominronin joined #git
19:06 navidr joined #git
19:06 e14 joined #git
19:08 paws_ joined #git
19:08 thebope joined #git
19:08 BlackMaria joined #git
19:10 inflames joined #git
19:10 Sasazuka joined #git
19:12 raspado joined #git
19:12 nidr0x joined #git
19:16 Gsham joined #git
19:18 Tanfrieze joined #git
19:19 Dougie187 joined #git
19:19 thebope joined #git
19:19 converge joined #git
19:19 bill99 joined #git
19:20 einnjo joined #git
19:20 jrnieder joined #git
19:23 raspado joined #git
19:23 raspado joined #git
19:23 Sasazuka_ joined #git
19:23 ski7777 joined #git
19:24 raspado joined #git
19:26 rominronin joined #git
19:31 Es0teric joined #git
19:32 Rodya_ joined #git
19:32 aidalgol joined #git
19:33 jnavila joined #git
19:33 BenderRodriguez joined #git
19:33 fcrick3 joined #git
19:34 jason237 joined #git
19:35 fcrick3 Hi there - I was trying to figure out a way to get a blame view of all the lines touched by a commit, but I guess I'm having trouble explaining to google what I want
19:35 _ikke_ fcrick3: A diff comes closest to that...
19:35 fcrick3 like, I want to effective just build up all the line ranges touched by a specific commit, then run git blame from the commit's parent on that list of files and line ranges
19:36 jccn joined #git
19:37 fcrick3 _ikke_: how do you mean? can you show previous authors in a diff?
19:37 sbeller joined #git
19:37 _ikke_ No, but all lines touched by a commit.,.
19:38 _ikke_ And the author of that commit is per difinition the author of those lines
19:38 manuels joined #git
19:38 fcrick3 right - that author is me - I want to know the previous author
19:38 _ikke_ fcrick3: Perhaps look into the blame view of tig
19:39 tnecniv joined #git
19:39 KV joined #git
19:39 Es0teric joined #git
19:40 sitaram joined #git
19:40 sitaram joined #git
19:40 venmx joined #git
19:41 dsantiago joined #git
19:43 GodGinrai fcrick3: you don't want the blame of a commit.  You want to get the blame state from before you made the commit
19:44 GodGinrai fcrick3: although I don't see how this would be useful
19:44 fcrick3 GodGinrai: correct
19:44 fcrick3 well, I have a large change that makes very targeted changes to dozens of files, and I was going to send code reviews to the authors of the code i'm changing
19:45 mrmadcat joined #git
19:45 fcrick3 so I'm trying to determine, for each file in the change, who was most recently responsible for the lines of code my change affects
19:46 rominronin joined #git
19:46 GodGinrai fcrick3: have you already made the commit?
19:46 _ikke_ fcrick3: use git diff with 0 context to find the line ranges of changed code
19:46 fcrick3 locally, but I can just reset HEAD^ if that helps
19:46 _ikke_ and then feed it to git blame
19:47 _ikke_ No need to reset
19:47 GodGinrai _ikke_: that will just show his name for all of the changes
19:48 GodGinrai fcrick3: you could make a separate branch and reset up so that you don't screw up your branch
19:48 GodGinrai fcrick3: then do something like this:
19:48 GodGinrai soft reset, mind you
19:48 GodGinrai so you have the modified files
19:48 GodGinrai anyway
19:48 dsdeiz joined #git
19:49 luuk joined #git
19:49 GodGinrai `git status | awk '/modified/ { print $2 }' | git blame`
19:49 _ikke_ GodGinrai: Right, you need to use the parent commit of course as base
19:49 GodGinrai ^ that would give you the blame for the files
19:49 GodGinrai not the particular lines, tho'
19:50 fcrick3 yeah the files are all huge and in each I changed a tiny fraction of the file
19:50 monokrome joined #git
19:50 e14 joined #git
19:51 fcrick3 and most have many authors that didn't touch the parts I changed
19:51 rominronin joined #git
19:52 fcrick3 git diff -U0 HEAD^ seems like a good starting point at least
19:52 _ikke_ fcrick3: combination of git diff -U0 and git blame -L <commit>
19:52 _ikke_ yes
19:53 gustafl joined #git
19:53 thebope joined #git
19:53 GodGinrai you'd still need some way for git blame to know you don't want the most recent person
19:53 GodGinrai if you are not using reset
19:53 _ikke_ GodGinrai: By specifying the parent commit
19:53 _ikke_ GodGinrai: git blame can take any commit
19:53 fcrick3 can't i just checkout hEAD^?
19:54 _ikke_ sure, but why should you, no need
19:54 GodGinrai _ikke_: I don't even see a commit hash param in the manpage for git-blame
19:54 gustafl testing
19:54 fcrick3 still fairly new to the git ecosystem...years in perforce
19:54 _ikke_ GodGinrai:  [<rev> | --contents <file> | --reverse <rev>..<rev>]
19:55 dminuoso joined #git
19:55 GodGinrai _ikke_: ah, I missed that
19:56 gustafl Newbie here. Is there a way to remove all files from my reposity that is not currently not in my working directory? To save space.
19:56 GodGinrai gustafl: do you mean remove all untracked files?
19:56 _ikke_ GodGinrai: Yeah, I had to look where it wa
19:56 _ikke_ was
19:56 _ikke_ gustafl: man git clean
19:56 gitinfo gustafl: the git-clean manpage is available at http://jk.gs/git-clean.html
19:56 _ikke_ This command is destructive
19:57 gustafl Um, I'm not sure. I'm thinking of using git as a backup solution. And push updates to a clone on an external harddrive.
19:57 GodGinrai !backup
19:57 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.c​om/GitBestPractices/#backups
19:57 GodGinrai oh
19:57 _ikke_ GodGinrai: not what you meant :P
19:57 GodGinrai that's not the factoid I was looking for
19:57 jrnieder joined #git
19:57 GodGinrai I thought we had one that said "Git is not a backup solution!"
19:58 GodGinrai in more words
19:58 gustafl So when I delete files from my computer, I want to push the same deletes to my external harddisk. And ideally, remove every trace of them from the .git folder too.
19:58 _ikke_ !wisdom
19:58 gitinfo 1. If it's not stored in git, it might as well be gone. 2. Git is not a backup.
19:58 GodGinrai _ikke_: not the one I remember, but I like it ;)
19:58 grawity 2. If it *is* stored in git, it still has the risk of being gone.
19:58 _ikke_ Yup
19:58 grawity GodGinrai: http://jk.gs/git/bot/trigger.php
19:59 GodGinrai grawity: 3. <.<
19:59 GodGinrai also thanks for that link
19:59 GodGinrai I wish phrik had a link like this
19:59 grawity no, I'd say that's more fitting in place of current #2
19:59 grawity phrik has !listkeys and !listvals
19:59 _ikke_ (archlinux bot)
20:00 GodGinrai grawity: listkeys and listvals requires you have some idea of what you are looking for.  It is horrible for discovery/
20:00 GodGinrai *.
20:00 ahmed_elgabri joined #git
20:00 Hello71 !trig
20:00 gitinfo [!triggers] Please don't spam me! I'm just a poor bot! Here's everything I know: http://jk.gs/git/bot/trigger.php -- that's a list of these nifty keywords like "!bot" that you can use in the channel, in case you were wondering.
20:01 _ikke_ That page also has the history of the triggers
20:01 Hello71 wait, it does?
20:01 _ikke_ click on a factoid and you'll see
20:01 rominronin joined #git
20:01 WayToDoor joined #git
20:02 Hello71 shit, I been here, uh... years I think now
20:02 _ikke_ :-)
20:02 GodGinrai lol
20:03 _ikke_ June 2014 is my first record of you being here :P
20:03 Balliad joined #git
20:04 gustafl left #git
20:04 _ikke_ Sorry, my archived logs go back further
20:04 _ikke_ 2013-09-13 02:58:28     -->     Hello71 (Hello71@wikipedia/Hello71) has joined #git
20:05 Hello71 that's (more than) three years, isn't it
20:05 dreiss joined #git
20:05 _ikke_ Yeah, bit more than 3 years
20:06 _ikke_ 2011-12-20 08:04:37     -->     _ikke_ (~kevin@****.nl) has joined #git
20:06 zzz joined #git
20:06 Hello71 _ikke_: UTC is true clock [20:58:29] *** Joins: Hello71 (Hello71@wikipedia/Hello71)
20:06 ejb joined #git
20:06 Es0teric joined #git
20:07 _ikke_ Hello71: I'm 1-2 hours from UTC
20:07 Hello71 that was 20130912.log
20:08 breezy_ joined #git
20:12 rominronin joined #git
20:12 anonymuse joined #git
20:14 cagedwisdom joined #git
20:19 tnecniv joined #git
20:20 pothepanda joined #git
20:24 jrnieder joined #git
20:25 askb joined #git
20:25 TSS_ joined #git
20:27 Doginal joined #git
20:27 tmsmith joined #git
20:28 thebope_ joined #git
20:28 ajf- joined #git
20:28 ajf- What is the recommended way to keep a remote so that I can get updates of the software while allowing my own modifications?
20:29 grawity the usual, just merge from the remote branch every now and then
20:29 kadoban ajf-: I usually have a branch that just tracks exactly what the upstream has, separate from any branch I touch directly.
20:29 ojacobson I let git-remote manage those ;)
20:30 kadoban Then either you merge upstream in occasionally, or you rebase your stuff on upstream occasionally, depending on what the goal is and what the workflow is.
20:31 e14 joined #git
20:31 steve joined #git
20:32 thebope joined #git
20:33 thebope_ joined #git
20:34 HoierM joined #git
20:36 thebope joined #git
20:36 jason237 joined #git
20:36 meLon joined #git
20:36 tnecniv joined #git
20:37 worstadmin_ joined #git
20:39 worstadmin_ joined #git
20:41 sea-gull_ joined #git
20:42 thiago joined #git
20:44 converge joined #git
20:44 b1tchcakes joined #git
20:47 [0xAA] joined #git
20:47 LionsMane1 joined #git
20:48 Ordentlig joined #git
20:49 worstadmin_ left #git
20:50 worstadmin joined #git
20:52 _28_ria joined #git
20:54 e14 joined #git
20:55 Shentino joined #git
20:55 kyan joined #git
20:55 tristanp joined #git
20:56 Darren_ joined #git
20:57 Atm0spher1c joined #git
20:57 tnecniv joined #git
21:05 cdg joined #git
21:07 dodobrain joined #git
21:10 _nwkj86 joined #git
21:11 fahadash joined #git
21:11 HoierM joined #git
21:12 robattila256 joined #git
21:12 FroMaster joined #git
21:12 star_prone joined #git
21:12 howitdo joined #git
21:14 e14 joined #git
21:15 jstimm joined #git
21:16 s0c joined #git
21:16 porsche944 joined #git
21:16 ajf- joined #git
21:16 ismaris joined #git
21:18 GavinMagnus joined #git
21:18 msonntag joined #git
21:20 dsdeiz joined #git
21:22 chachasmooth joined #git
21:25 blackpajamas joined #git
21:27 blackpajamas joined #git
21:30 chardan joined #git
21:31 arand joined #git
21:32 heroux joined #git
21:36 johnmilton joined #git
21:38 d0nn1e joined #git
21:40 davimore__ joined #git
21:41 tnecniv joined #git
21:47 e14 joined #git
21:49 Gloomy joined #git
21:50 roelmonnens joined #git
21:50 shgysk8zer0 joined #git
21:52 zivester joined #git
21:53 bamb joined #git
21:53 neilthereildeil joined #git
21:53 bamb left #git
21:53 neilthereildeil hi
21:53 neilthereildeil i clone a tree, and since then, the source has had some updates
21:54 svm_invictvs joined #git
21:54 dminuoso neilthereildeil, a clone automatically adds a remote that you can fetch and merge from.
21:55 neilthereildeil i did git pull, but when i try to checkout the latest commit in the remote tree, it tells me head is detached
21:56 neilthereildeil i get this error: http://pastebin.com/4PdreGJU
21:56 dminuoso neilthereildeil, all HEAD is is a pointer. You can point it back to your local master branch by "git checkout master"
21:56 neilthereildeil any idea what im doing wrong?
21:57 neilthereildeil but my master should be updated when i pull, right?
21:57 neilthereildeil it should mirror the source tree i thought
21:57 dminuoso neilthereildeil, possibly. Did you make any modifications to your local repository?
21:57 neilthereildeil i added tags
21:57 neilthereildeil but im expecting git checkout -f master will take me to the latest commit in the remote tree
21:58 neilthereildeil is that an incorrect assumption on my part?
21:58 vaskozl joined #git
21:58 dminuoso neilthereildeil, correct in that the assumption is incorrect.
21:58 neilthereildeil hmm
21:58 neilthereildeil what is happening in reality?
21:58 vaskozl Hey what is the correct way of making a repository on a server public (do not want to use github or the like)?
21:59 dminuoso neilthereildeil, "git checkout master" will just have HEAD point to master, which itself points to some commit.
21:59 vaskozl So that it is easily clonable, any links appreciated.
21:59 neilthereildeil but it master updated when i pull?
21:59 HedgeWork vaskozl: There are many acceptable ways.  If you aren't sure what to use, I'd start with gitlab if you have newbs who need a gui, or gitolite if you want good granular security and no-fuss setup.
21:59 roygbiv joined #git
22:00 dminuoso neilthereildeil, yes. Effectively it just does "git fetch && git merge FETCH_HEAD" by default
22:00 HedgeWork vaskozl: Wait, I misread, you aren't worried about team/contributor access, just public clone-ability?
22:00 vaskozl Yes, should I make a git user and allow anyone to log in trough ssh?
22:01 vaskozl I just want to host my own repository basically.
22:02 neilthereildeil heres the error i got when i first tean that command:
22:02 neilthereildeil http://pastebin.com/1ZRdbCPU
22:02 neilthereildeil ran*
22:02 neilthereildeil it said im leaving a commit behind
22:02 dminuoso neilthereildeil, a detached head just means that HEAD is no longer pointing to a branch but some specific commit or tag.
22:02 HedgeWork vaskozl: I'd have to have a think about that.  I've never actually put something up without also setting up access for contributors.
22:02 HedgeWork Maybe someone else here has a good answer :)
22:02 neilthereildeil but the o ijust pasted was more sinister
22:02 dminuoso neilthereildeil, it's quite explicit.
22:02 dminuoso neilthereildeil, what else beyond what git is already telling you in such a verbose way would you want me to tell yoy?
22:03 cagedwisdom joined #git
22:03 cgundersson joined #git
22:03 neilthereildeil why wasnt my target checkout part of an of my branches?
22:03 neilthereildeil i thought it would be the newest tip of master after a git pull
22:05 neilthereildeil ok the difference i see is that master now points to the latest commit, which might have solved the issue
22:05 neilthereildeil how can i make sure master is updated every time i pull?
22:07 ljc joined #git
22:09 texasmynsted if I add  "add -b origin/develop" to my get submodule add, then it should create a remote named origin, and try to align with the commit from develop, right?
22:11 Mutter joined #git
22:11 Dougie187 joined #git
22:15 nidr0x joined #git
22:17 nickabbey joined #git
22:19 Mutter joined #git
22:24 steelsw0rd joined #git
22:24 Mutter joined #git
22:31 Mutter joined #git
22:32 Zamyatin joined #git
22:32 breezy_ joined #git
22:32 JanC_ joined #git
22:39 tang^ joined #git
22:42 rwb joined #git
22:47 Noodlewitt joined #git
22:47 CheckDavid joined #git
22:48 romerocesar joined #git
22:54 [Saint] joined #git
22:56 texasmynsted Anybody know if there is a way to check some kinds of config changes into the repo its self?
22:57 texasmynsted example: I would like to create an alias so that whenever git push is performed, a git push --recurse-submodules=check is actually performed.
22:58 tnecniv joined #git
22:58 texasmynsted Oh I see this git config push.recurseSubmodules check
22:58 sbeller texasmynsted: for that one you can use an alias for now or the config
22:59 sbeller there is a change pending upstream to make --recurse-submodules=check the default option :)
22:59 texasmynsted but this should work, right?  git config push.recurseSubmodules check
22:59 sbeller sure\
23:01 star_prone joined #git
23:01 fmeerkoetter joined #git
23:02 mischat joined #git
23:02 MrLawrence joined #git
23:03 MrLawrence Hello. What would be a good heuristic for pushing changes to a repository?
23:03 MrLawrence Is there a foss project or company with guidelines for this?
23:04 jj995 joined #git
23:04 SilenceDogood joined #git
23:05 jj995 does gitcredentials work with ssh address's (e.g. git@git.example.com:foo/bar.git) or only https addresses?
23:06 kadoban Typically for SSH you just use public key authentication and your OS's ssh agent.
23:06 jj995 kadoban: yeah, I'm just trying to hack something together to get a travis build to work with a non-github submodule
23:06 llamapixel MrLawrence:  some initial reading http://nvie.com/posts/a-succ​essful-git-branching-model/   |  https://gist.github.com/djs​piewak/9f2f91085607a4859a66   I like a hybrid of the first with the dmz added.
23:07 SilenceDogood Hey folks, Git seems to be ignoring this line https://github.com/BenMcLean/Planet-Ge​nerator-Web/blob/master/.gitignore#L72
23:07 kadoban jj995: Then I'm not sure. I'd suspect it can work, but honestly I've never used the thing even.
23:07 venmx joined #git
23:07 SilenceDogood I want git to ignore everything in every folder called build, except I want to include the contents of the folder html/build/dist/ in my repository
23:08 MrLawrence llamapixel, thanks
23:08 llamapixel np m8
23:08 Eugene vaskozl - !gitolite is our recommendation for hosting a repo yourself
23:08 gitinfo vaskozl: 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/
23:08 SilenceDogood Doing it the way I linked in that line is what I find on Google and it doesn't work
23:08 kadoban SilenceDogood: line 66 appears like it'll screw up that plan
23:08 Eugene If you want to set up a public (read-only, unauthenticated) git:// server then you want to look into git-daemon; its setup is beyond the scope of IRC support ;-)
23:10 SilenceDogood kadoban, OK, I removed that line. But why?
23:12 star_prone joined #git
23:12 kadoban SilenceDogood:
23:12 kadoban "An optional prefix "!" which negates the pattern; any matching file excluded by a previous pattern will become included again. It is not possible to re-include a file if a parent directory of that file is excluded. Git doesn’t list excluded directories for performance reasons, so any patterns on contained files have no effect, no matter where they are defined."
23:12 kadoban From man git ignore
23:12 gitinfo the git manpage is available at http://jk.gs/git.html
23:13 kadoban Specifically the second sentence there
23:13 kadoban Hmm, that link wasn't the right one man gitignore
23:13 gitinfo the gitignore manpage is available at http://jk.gs/gitignore.html
23:13 SilenceDogood kadaban, thanks for the info
23:13 kadoban It trips up a *lot* of people. But you can work around it.
23:13 SilenceDogood Somebody told me I can use git check-ignore -v -- to find out which line of a gitignore is providing the rule to ignore some file. That solved it :)
23:14 kadoban Yep, that's a useful tool.
23:14 SilenceDogood But knowing that info from the manual is good to know also
23:17 Siegfried joined #git
23:20 ejb joined #git
23:21 whomp joined #git
23:23 thebope joined #git
23:23 justanotheruser joined #git
23:24 SwiftMatt joined #git
23:25 SilenceDogood thanks everybody
23:25 SilenceDogood left #git
23:28 chachasmooth joined #git
23:29 einnjo joined #git
23:31 dirtyroshi joined #git
23:31 Derderderd joined #git
23:36 thebope joined #git
23:38 Ordentlig joined #git
23:38 pbrewczynski joined #git
23:38 romerocesar joined #git
23:39 chachasmooth joined #git
23:42 romerocesar joined #git
23:43 romerocesar joined #git
23:43 Dougie187 left #git
23:45 dsdeiz joined #git
23:46 neilthereildeil left #git
23:47 fmcgeough joined #git
23:47 snowkidind joined #git
23:52 romerocesar joined #git
23:55 a3Dman joined #git
23:55 jrnieder joined #git
23:56 Zamyatin joined #git

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