Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-05-20

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

All times shown according to UTC.

Time Nick Message
00:00 rkazak joined #git
00:01 eletuchy joined #git
00:02 CJKinni joined #git
00:10 Lipp joined #git
00:12 joako joined #git
00:17 fuchstronaut joined #git
00:18 rgrinberg joined #git
00:20 CJKinni joined #git
00:25 boombatower joined #git
00:30 tomog999_ joined #git
00:33 Gamecubic joined #git
00:35 madewokherd joined #git
00:38 miesco joined #git
00:39 miesco Hi.  Is it good to compress a tarball if you are gpging it?
00:39 miesco I know gpg compresses...
00:39 polyzen gpg doesn't compress
00:39 polyzen afaik o.O
00:40 polyzen miesco, are you signing the tarball or encrypting it?
00:40 miesco polyzen: encrypting
00:40 polyzen probably good to compress regardless
00:40 polyzen http://superuser.com/questions/427264/d​oes-gnu-gpg-compress-by-default#427301
00:40 polyzen it does compress
00:43 ash_workz joined #git
00:43 polyzen If this option is not used, the default behavior is to examine the recipient key preferences to see which algorithms the recipient supports. If all else fails, ZIP is used for maximum compatibility.
00:44 liuzhen joined #git
00:45 arahael4 joined #git
00:45 raininja joined #git
00:45 Lipp joined #git
00:48 jetpack_joe joined #git
00:49 Internet13 joined #git
00:52 mwhooker joined #git
00:54 blackwind_123 joined #git
00:54 dmto joined #git
00:58 justanotheruser joined #git
00:59 arahael joined #git
01:01 ril joined #git
01:01 cdg joined #git
01:02 hanshenrik joined #git
01:03 sssilver joined #git
01:03 rkazak joined #git
01:03 boombatower joined #git
01:03 pks joined #git
01:04 thiago joined #git
01:04 hanshenrik git is stupid
01:04 hanshenrik err, i mean, i have a problem with git*
01:06 hanshenrik i had a project, then i made another project in the parent folder of the first project, and added the subporj/.git/* to .gitignore of the parent folder project, and... the sub folder's project isn't being tracked in the parent project's folder's project... something
01:06 hanshenrik hmm, maybe i should find a less confusing way to say that..
01:06 Eugene !subproject
01:06 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.
01:07 hanshenrik further, i tried to delete the subproject's .git folder/files.. the parent project still refuse to track the subproject's files... how do i get the parent project to track those files?
01:07 hanshenrik i don't care about the subproject's history
01:08 Eugene If .git/ doesn't exist then git won't consider it to be a repository
01:08 Eugene Did you perhaps accidnetally make it into a submodule?
01:08 eletuchy joined #git
01:08 hanshenrik it did exist last time i did "git add *" "git commit" "git push"
01:09 hanshenrik then i deleted the subproject's .git
01:09 eletuchy joined #git
01:09 hanshenrik and... then i did "git add *" "git commit" again, but the subproject's files aren't tracked still
01:10 johntramp joined #git
01:10 hanshenrik i guess git is just ignoring everything happening in that subfolder since
01:11 hanshenrik C:\tibia\Blackd-Proxy-CLASSIC [master ≡]> git add *
01:11 hanshenrik fatal: Pathspec '*' is in submodule 'Blackd-Proxy-CLASSIC'
01:11 johntramp joined #git
01:11 hanshenrik yeah its probably a submodule then
01:11 hanshenrik how do i... un-submodule it? :p
01:11 Gsham joined #git
01:11 johntramp hi, is there an easy way to get a version number similar to git describe --tags --long, except without the git hash on the end?  so just the tag and number of commits
01:11 flaguy48 joined #git
01:12 sssilver joined #git
01:13 haway joined #git
01:15 sssilver joined #git
01:16 Eugene You could just regex it off....
01:16 hanshenrik seems the answer to my question (how to un-submodule) is here ^^ http://stackoverflow.com/a/1789374/1067003
01:16 hanshenrik Eugene, thanks for the help ^^
01:18 fuchstronaut joined #git
01:19 sssilver joined #git
01:19 italoacasas joined #git
01:20 Tanger_ joined #git
01:22 Lipp joined #git
01:26 MrWoohoo joined #git
01:28 chris_wot joined #git
01:30 bjpenn joined #git
01:32 clemf joined #git
01:32 Shentino joined #git
01:35 nandi_ge___ joined #git
01:36 ochorocho__ joined #git
01:38 prsn joined #git
01:38 diegoviola joined #git
01:39 csd_ joined #git
01:39 bjpenn joined #git
01:41 mwhooker joined #git
01:42 rominronin joined #git
01:46 nocbreak joined #git
01:46 jaguarmagenta joined #git
01:46 dmto joined #git
01:46 SimonNa joined #git
01:46 arescorpio joined #git
01:49 ilbot3 joined #git
01:49 Topic for #git is now Welcome to #git, the place for git help and zerg rushes | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.8.2 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | YOU MUST SUBMIT ADDITIONAL JOKES https://madeitwor.se/git-jokes
01:49 sangy joined #git
01:51 eletuchy joined #git
01:53 ozzloy joined #git
01:53 Asiajey joined #git
01:54 SimonNaa joined #git
01:58 Lipp joined #git
01:59 dmto joined #git
02:01 sd512 joined #git
02:02 harish joined #git
02:04 mda1 joined #git
02:10 joki joined #git
02:11 xlegoman joined #git
02:12 dmto joined #git
02:13 Gsham joined #git
02:13 Ali1 joined #git
02:14 lindenle joined #git
02:16 watabou_ joined #git
02:18 acetakwas joined #git
02:19 fuchstronaut joined #git
02:19 bronson joined #git
02:20 Saylen joined #git
02:21 Goplat joined #git
02:23 Gsham joined #git
02:24 lukeholder joined #git
02:29 nilg joined #git
02:29 eletuchy joined #git
02:31 Sleepy63 joined #git
02:31 jimi_ joined #git
02:31 Yuken joined #git
02:32 Yuken did someone say zerg rushes?
02:34 cdg joined #git
02:35 netj joined #git
02:35 BombStrike joined #git
02:36 Lipp joined #git
02:37 ochorocho__ joined #git
02:39 govg joined #git
02:39 otho joined #git
02:41 Dougie187 left #git
02:43 MacWinner joined #git
02:45 chachasmooth joined #git
02:51 fscala joined #git
02:57 dermoth_ joined #git
03:00 csd_ joined #git
03:01 mbreedlove joined #git
03:04 lukeholder joined #git
03:05 hexagoxel joined #git
03:05 Khisanth joined #git
03:12 Lipp joined #git
03:15 Yuken left #git
03:15 ceaucari joined #git
03:19 sesquipedalian joined #git
03:20 fuchstronaut joined #git
03:21 mwhooker joined #git
03:29 mbreedlove joined #git
03:31 mehola joined #git
03:31 mbreedlove joined #git
03:32 Gsham joined #git
03:38 ochorocho__ joined #git
03:39 Mobutils joined #git
03:42 acetakwas joined #git
03:42 |ifei5g00d joined #git
03:43 rominronin joined #git
03:43 jaguarmagenta joined #git
03:43 blackwell joined #git
03:45 fracting joined #git
03:47 sesquipedalian joined #git
03:48 harish joined #git
03:49 Lipp joined #git
03:51 Limes_ joined #git
03:51 Raging_Hog joined #git
03:54 Gsham joined #git
03:56 joako joined #git
03:56 mehola joined #git
03:58 unforgiven512 joined #git
03:58 cyan__ joined #git
03:59 jonmorehouse joined #git
04:04 lukeholder joined #git
04:09 Mobutils joined #git
04:12 jagob joined #git
04:13 dEPy joined #git
04:14 DirtyCajun joined #git
04:16 mwhooker joined #git
04:19 baba_ joined #git
04:19 jonmorehouse joined #git
04:20 fuchstronaut joined #git
04:21 baba_ joined #git
04:24 shauna_ joined #git
04:24 rxo joined #git
04:24 ojacobson joined #git
04:24 Lipp joined #git
04:25 autofly joined #git
04:26 irqq joined #git
04:27 darlinger joined #git
04:29 faux___ joined #git
04:30 LeMike joined #git
04:38 dmto joined #git
04:40 aavrug joined #git
04:41 _ikke_ deanj: git tag + git describe tends to solve this pretty well
04:43 tinajohnson___ joined #git
04:43 baba_ joined #git
04:43 arahael joined #git
04:44 rominronin joined #git
04:45 Cabanossi joined #git
04:45 Rish joined #git
04:45 robogoat joined #git
04:49 dmto joined #git
04:50 nnyk_ joined #git
04:50 arahael1 joined #git
04:54 bjpenn joined #git
04:56 grassass joined #git
04:56 eletuchy joined #git
04:57 dmto joined #git
05:00 SpeakerToMeat joined #git
05:01 Lipp joined #git
05:04 arahael1 joined #git
05:10 lukeholder joined #git
05:11 baba_ joined #git
05:14 mwhooker joined #git
05:14 baba_ joined #git
05:15 kanzure joined #git
05:15 thiago joined #git
05:16 qt-x joined #git
05:18 mwhooker joined #git
05:19 den2204 joined #git
05:20 den2204_ joined #git
05:20 dmto joined #git
05:21 fuchstronaut joined #git
05:24 jonmorehouse joined #git
05:25 Rish joined #git
05:31 amdi_ joined #git
05:31 woodruffw joined #git
05:31 woodruffw joined #git
05:32 dmto joined #git
05:38 dmto joined #git
05:38 subhojit777 joined #git
05:39 ochorocho__ joined #git
05:39 Lipp joined #git
05:39 dsdeiz joined #git
05:39 dsdeiz joined #git
05:39 mda1 joined #git
05:40 mohabaks joined #git
05:40 eletuchy joined #git
05:41 muthu joined #git
05:41 apotry joined #git
05:41 Lvjasha joined #git
05:42 SimonNa joined #git
05:42 ThomasLocke joined #git
05:44 jaguarmagenta joined #git
05:44 gajop joined #git
05:44 Lvjasha joined #git
05:45 rominronin joined #git
05:45 gajop i'm trying to pull an upstream repo and my git is stuck https://gist.github.com/gajop/8​d803219817bc642fb1ba1e2fc8c17a2
05:46 gajop been like this for a while, what to do/how to debug?
05:48 gajop i expect it to create a merge conflict which i should then resolve
05:48 baba_ joined #git
05:49 mwhooker joined #git
05:51 LucaTM joined #git
05:52 smithbone joined #git
05:53 romankl joined #git
05:53 jonmorehouse joined #git
05:53 Mobutils joined #git
05:54 jaguarmagenta joined #git
05:54 _joes__ joined #git
05:54 Lvjasha joined #git
05:55 stangeland joined #git
05:55 robogoat joined #git
05:56 stantonk joined #git
05:59 settermjd joined #git
06:00 gajop i hacked around it ^_^
06:00 docnvk joined #git
06:02 wcpan joined #git
06:05 dmto joined #git
06:11 rscata joined #git
06:13 lamppid joined #git
06:13 baba_ joined #git
06:15 Lipp joined #git
06:15 weyer joined #git
06:17 arahael1 joined #git
06:19 GavinMagnus joined #git
06:21 roelmonnens joined #git
06:22 fuchstronaut joined #git
06:22 FrontPsych joined #git
06:23 bronson joined #git
06:24 leeN joined #git
06:24 nilsi joined #git
06:24 mangolisk__ joined #git
06:24 dopesong joined #git
06:25 PaulCape_ joined #git
06:26 MajorPayne joined #git
06:28 me_and joined #git
06:28 relipse joined #git
06:28 Smirnov joined #git
06:28 Smirnov joined #git
06:28 kisonecat joined #git
06:29 _W_ joined #git
06:29 Ariadeno joined #git
06:29 LaT0rtue joined #git
06:29 angelsl joined #git
06:29 justanotheruser joined #git
06:29 MaddHatter joined #git
06:30 drodger joined #git
06:31 neanderslob joined #git
06:31 cdown joined #git
06:32 cwillu_at_work joined #git
06:32 Nugget joined #git
06:32 Corosus__ joined #git
06:32 LeMike joined #git
06:32 PaulCapestany joined #git
06:32 Cyclohexane joined #git
06:33 led_belly joined #git
06:33 jceb joined #git
06:33 DolphinDream joined #git
06:33 jonmorehouse joined #git
06:33 canine joined #git
06:34 kriskropd joined #git
06:35 GavinMagnus joined #git
06:35 BrianBlaze420 joined #git
06:38 eijk joined #git
06:38 murr4y joined #git
06:39 chachasmooth joined #git
06:41 gfixler joined #git
06:41 lukeholder joined #git
06:42 Sceorem joined #git
06:42 jast joined #git
06:44 IndigoTi- joined #git
06:45 circ-user-bUyXf joined #git
06:46 chachasmooth joined #git
06:46 stuivi joined #git
06:46 rominronin joined #git
06:46 xelxebar joined #git
06:48 bjpenn joined #git
06:49 rominron_ joined #git
06:49 fees joined #git
06:51 nnyk_ joined #git
06:52 arahael2 joined #git
06:52 mwhooker joined #git
06:52 Tanger_ joined #git
06:54 PCatinean joined #git
06:56 Nitesh joined #git
06:57 hanthings_ joined #git
06:58 GavinMagnus joined #git
06:58 cdown joined #git
06:59 baba_ joined #git
06:59 lindenle joined #git
07:00 Xethron joined #git
07:00 arahael3 joined #git
07:00 acolson joined #git
07:01 ItSANg___ joined #git
07:01 robogoat joined #git
07:04 Lipp joined #git
07:05 arahael4 joined #git
07:05 Alienpruts joined #git
07:07 jsolano joined #git
07:07 dan2k3k4 joined #git
07:08 zadock joined #git
07:08 cdown joined #git
07:10 chris_wot joined #git
07:10 Lipp joined #git
07:10 harish joined #git
07:11 TomyWork joined #git
07:11 arahael joined #git
07:11 Alienpruts joined #git
07:14 JeroenT joined #git
07:15 fracting joined #git
07:15 arahael1 joined #git
07:15 bjpenn joined #git
07:18 xbob joined #git
07:20 nilsi joined #git
07:21 dsantiago joined #git
07:21 jceb joined #git
07:23 fuchstronaut joined #git
07:23 arahael2 joined #git
07:27 arahael3 joined #git
07:31 hahuang65 joined #git
07:31 harish joined #git
07:32 josuebrunel joined #git
07:32 arahael4 joined #git
07:33 Lipp joined #git
07:33 roflin joined #git
07:34 ahmedelgabri joined #git
07:37 arahael4 joined #git
07:39 ochorocho__ joined #git
07:42 arahael joined #git
07:43 SpaghettiCat joined #git
07:43 slappymcfry joined #git
07:44 Balliad joined #git
07:45 DevAntoine joined #git
07:45 jas4711 joined #git
07:45 leonarth joined #git
07:46 DevAntoine joined #git
07:46 arahael1 joined #git
07:47 SpaghettiCat Good Friday morning friends, if I'm currently working on branch "br_feature1" and want to know what updates have been made to branch "pu" (without merging them to my local "pu") . How can I do this using `git fetch`?
07:49 emPi joined #git
07:51 arahael2 joined #git
07:51 kurkale6ka joined #git
07:52 osse first step: run git fetch
07:52 elisa87 joined #git
07:52 elisa87 hey why do I get this error: error: failed to push some refs to 'git@github.com:lamiastella/bashscripts.git'
07:53 radsy joined #git
07:53 troulouliou_div2 joined #git
07:53 cbreak elisa87: read the rest of the messages
07:54 elisa87 remote: fatal error in commit_refs cbreak
07:54 cbreak pastebin the whole output
07:54 cbreak and post the link here
07:55 colona joined #git
07:56 davisonio joined #git
07:57 arahael2 joined #git
07:57 bongjovi joined #git
07:58 elisa87 can you guys see the videos https://www.lynda.com/Bash-tutorials​/Working-flags/142989/156661-4.html
07:58 baba_ joined #git
07:58 jceb joined #git
07:58 elisa87 http://pastebin.com/50BL9YTF cbreak
07:59 jsolano hi there! I am trying to work locally mixing code from 2 remotes but pushing changes separately so that they are kept separate, most of the code is non-overlapping. Could anybody point me in the right direction?
07:59 cbreak elisa87: the remote doesn't want you to allow to push
07:59 cbreak elisa87: it says "remote rejected"
07:59 elisa87 but I just pushed to it
07:59 elisa87 what is the reason?
07:59 elisa87 remote is myself haha cbreak
07:59 cbreak the remote has problems with commit_refs, and rejects it because of that
08:00 cbreak elisa87: is the target repo's volume full?
08:00 cbreak or do you have permission issues?
08:00 elisa87 I don't think so
08:00 elisa87 no
08:00 elisa87 as I said I just pushed to it
08:01 cbreak I'd recommend checking the permissions in the filesystem of the target repository
08:01 elisa87 I just pushed to it 1 hour ago cbreak https://github.com/lamiastella/bash​scripts/commits?author=lamiastella
08:01 elisa87 I don't know how :(
08:01 cbreak oh, so it's not your repo
08:01 cbreak it's github's
08:01 cbreak don't think they give enough access for that
08:01 arahael3 joined #git
08:02 elisa87 lamiastella is my account!
08:02 elisa87 and that is my repo
08:02 clmsy joined #git
08:02 cbreak elisa87: you can try git push with -f to force, but that might not help in this situation
08:02 cbreak elisa87: it's not yours because you can't look at it :)
08:02 cbreak github doesn't permit enough access to debug this properly (
08:02 elisa87 yeah
08:02 cbreak it's quite annoying, really
08:03 elisa87 :( is it because it is bash scripts?
08:03 cbreak no
08:03 cbreak I would suspect it to be some github related issue
08:03 osse elisa87: try ssh -T git@github.com
08:04 mbreedlo_ joined #git
08:04 selckin joined #git
08:05 weyer joined #git
08:08 ocbtec joined #git
08:09 Fijit joined #git
08:09 baxx joined #git
08:10 WayToDoor joined #git
08:10 fracting joined #git
08:10 jonmorehouse joined #git
08:11 baba__ joined #git
08:12 Olipro joined #git
08:13 arahael3 joined #git
08:14 Alienpruts joined #git
08:15 Shashikant86 joined #git
08:15 Lipp joined #git
08:16 _pix joined #git
08:17 mbreedlove joined #git
08:19 cdown joined #git
08:19 bjpenn joined #git
08:19 david007 joined #git
08:19 mbreedlove joined #git
08:20 mischat joined #git
08:20 aavrug joined #git
08:22 harish joined #git
08:25 marcogmonteiro joined #git
08:26 dcpc007 left #git
08:26 bronson joined #git
08:26 hanthings_ joined #git
08:31 br0kenman joined #git
08:31 aavrug joined #git
08:33 djb-irc joined #git
08:35 baba_ joined #git
08:36 gaen joined #git
08:36 m1dnight_ joined #git
08:37 parv joined #git
08:38 baba_ joined #git
08:40 courrier joined #git
08:40 netcarver joined #git
08:41 cdown joined #git
08:41 wrouesnel2 joined #git
08:42 dsdeiz joined #git
08:43 zeroed joined #git
08:45 glauxosdever joined #git
08:46 djb-irc joined #git
08:48 nnyk_ joined #git
08:48 notebox joined #git
08:49 rivarun joined #git
08:50 cdown joined #git
08:50 JeroenT joined #git
08:51 notebox joined #git
08:55 acetakwas joined #git
08:58 jaguarmagenta joined #git
08:58 platzhirsch joined #git
08:58 platzhirsch Can someone explain "I enforce rebasing on all pull requests and reject all merges in all repos. This keeps a very clean git history; but, more to the point, rebasing moves your local commits until after the vendor updates." ? This is regarding getting added vendor libraries out of pull requests
08:59 platzhirsch I have no clue what it means in terms of rebasing, probably need to read up on rebase again, just conceptionally its confusing
08:59 selckin keep a linear history
08:59 selckin think svn
09:02 lindenle joined #git
09:04 tobiasvl basically I guess it means that they don't want merge commits
09:04 fracting joined #git
09:05 platzhirsch tobiasvl: which would mean no pull requests?
09:06 cdown joined #git
09:06 tobiasvl platzhirsch: are you talking about github now?
09:06 kadoban platzhirsch: It would mean your PRs should always be rebased on top of the tip of 'master' or whatever you expect them to be merged into, such that there is no need for a merge commit, they can be fast-forwarded.
09:07 tobiasvl platzhirsch: github supports squashing of pull requests, https://help.github.com/articles/a​bout-pull-request-merge-squashing/
09:07 tobiasvl presumably the person you quoted uses that
09:08 tobiasvl or well that actually squashes all into one commit, I thought it just merged using --rebase
09:08 kadoban tobiasvl: That sounds like a different (somehow even worse) way to get a linear history than what the quote implied.
09:08 tobiasvl yeah I saw that now too
09:08 kadoban Ah okay
09:09 jceb joined #git
09:10 razieliyo joined #git
09:10 razieliyo joined #git
09:13 EY joined #git
09:13 Vinnie_win joined #git
09:15 JeroenT joined #git
09:17 mbreedlove joined #git
09:18 Xethron joined #git
09:19 bjoe2k4 joined #git
09:19 mbreedlove joined #git
09:20 platzhirsch tobiasvl: yeah basically
09:20 platzhirsch the origin of the discussion was that I got enjoyed about commiting vendored libraries pollutes pull requests
09:20 cgundersson joined #git
09:23 acetakwas joined #git
09:24 chrisml joined #git
09:24 baba_ joined #git
09:24 fuchstronaut joined #git
09:25 sans_s3r1f joined #git
09:26 acetakwas joined #git
09:28 jceb joined #git
09:30 dmto joined #git
09:32 bjoe2k4 joined #git
09:32 baba_ joined #git
09:33 Xethron joined #git
09:34 JeroenT joined #git
09:34 jbitdrop joined #git
09:36 WayToDoor joined #git
09:37 mase-tech joined #git
09:39 subhojit777 joined #git
09:40 cyan__ joined #git
09:40 ochorocho__ joined #git
09:40 caog joined #git
09:41 aavrug joined #git
09:44 davisonio joined #git
09:45 selckin that sentance makes no sence whatsoerver
09:52 mdw joined #git
09:56 fees joined #git
09:57 irqq joined #git
09:59 RominRonin joined #git
10:01 Shashikant86 joined #git
10:02 EY joined #git
10:02 csd_ joined #git
10:02 digidog joined #git
10:04 realz joined #git
10:07 fuchstronaut joined #git
10:08 saedelaere joined #git
10:09 MattMaker joined #git
10:09 fuchstronaut joined #git
10:11 fuchstronaut joined #git
10:12 fracting joined #git
10:14 notebox joined #git
10:15 mniip joined #git
10:15 talin joined #git
10:16 talin hello. i am struggling with rebase and merge... i have a branch foo and master... so i am in branch foo, i merge in master. then i want to fetch the latest foo before i push... the question is, should i rebase or merge, then? if i try to rebase, i get some 3-way merge messages and a few conflicts...
10:16 digidog joined #git
10:17 talin feel like i don't know rebase/merge well enough to think clearly about this
10:24 klaas__ joined #git
10:24 tobiasvl talin: !eek
10:24 gitinfo talin: [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, *don't panic*. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-​2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice! A nice video explaining merge conflicts: https://www.youtube.com/watch?v=zz7NuSCH6II
10:25 AbsurdTech joined #git
10:25 tobiasvl you probably want to rebase in your scenario, if you want to make it look like the merge from master was done before you merge in the changes from upstream foo, but it depends
10:25 glauxosdever Hello, I'm wondering where is the list of commits in a branch stored.
10:26 tobiasvl talin: perhaps you also actually wanted to rebase when merging from master, if foo is a feature branch
10:26 Shashikant86 joined #git
10:26 penna left #git
10:26 AbsurdTech joined #git
10:27 glauxosdever Is it in some refs, objects...
10:27 tobiasvl glauxosdever: there is no list of commits, but the commit that the branch points to has a pointer to its parent(s), and then recursively back
10:27 glauxosdever If a commit has two parents?
10:27 AbsurdTech joined #git
10:27 tobiasvl then it points to both its parents
10:27 grawity glauxosdever: do `git cat-file -p <some_commit_id>`
10:28 grawity you'll see the "parent ..." lines in the raw object
10:28 bronson joined #git
10:29 rahtgaz joined #git
10:29 glauxosdever Since I don't have a commit with two parents so I can't see, which is the first of them?
10:29 glauxosdever I mean, the first parent.
10:29 mase-tech joined #git
10:30 baba__ joined #git
10:31 tobiasvl the base commit, ie. the branch which was merged into
10:31 glauxosdever Ok, so I'm interested in the first only.
10:32 grawity `git log` and `git rev-list` have --first-parent, in case that's useful
10:35 talin tobiasvl: hmm, right. so i do: checkout master, branch foo, <work on foo>, checkout master, fetch, rebase, checkout foo, merge master, checkout foo... and then before pushing foo, i wanna fetch and either merge or rebase it before i push
10:36 andjjj23 joined #git
10:38 andjjj23 How can I specify a literal escape sequence in the regex provided to git-log's -G option?
10:38 davisonio joined #git
10:39 andjjj23 i need to do something like: git log -G 'a ton of text that will be recognized as regular expression characters like ().*[], but i just want to literally search for this exact string'
10:39 grawity escape everything™
10:40 andjjj23 only option huh?
10:40 andjjj23 some regex support literal escape sequences e.g. \QThis is a literal escape sequence\E
10:40 grawity there's also -S, which has slightly different semantics, but might be close enough
10:40 andjjj23 -S isn't an option, already tried it and doesn't do what i want
10:40 osse andjjj23: one easy way is to make the shell help you: git log -G $'blah\emoreblah'
10:41 osse or  git log -G blah$'\e'blah  if you wish
10:41 selckin could try with -F not sure if thats applies to -G etc
10:42 noidea joined #git
10:42 andjjj23 osse: huh, i have no idea what that is doing. i'll have to look it up.
10:42 osse andjjj23: it's a specific form of shell quoting. try  echo $'hello\t\t\tworld'
10:42 osse the argument to git will have a literal escape (0x1b) in it
10:43 osse andjjj23: /msg greybot !$'
10:44 lss8 joined #git
10:44 david007 joined #git
10:44 andjjj23 hm, that makes sense to me for things like \t and \n but does that really help with regard to regular expression things? Are you saying if I git log -G  $'(].*'  it will search for the literal string (].*  ?
10:45 * andjjj23 goes to try it
10:45 grawity no
10:45 grawity it's more like ... the exact opposite.
10:45 grawity it doesn't perform escaping; it performs *un*-escaping.
10:45 ferr1 joined #git
10:45 andjjj23 yeah, that's what i thought.
10:45 osse andjjj23: it's not a regex feature and it's not a git feature. it's a shell feature.
10:46 osse the argument after -G will have an ESC in it. That's all.
10:46 andjjj23 I think what it boils down to is I need to manually escape ( [ . * etc
10:46 grawity osse: so a literal ESC will be understood by the regex parser then?
10:46 osse andjjj23: sorry, I've misunderstood your question completely
10:46 JeroenT joined #git
10:47 osse grawity: a literal ESC is not a special regex character so I assume it matches itself, just like a b or c
10:47 osse andjjj23: disregard my suggestion; it's not relevant to your question
10:47 grawity so why were you suggesting it, then
10:47 andjjj23 basically i need to be able to git log -G '\Q.*][{)\E' and have it search for literally .*][{)
10:47 osse grawity: misunderstanding.
10:48 selckin andjjj23: try adding -F
10:48 andjjj23 selckin: no dice :(
10:48 selckin gonna need to find a cli tool to escape it then or do it manually
10:48 andjjj23 to be clear, git uses posix regex?
10:49 osse selckin: I suppose the -F only applies to --grep
10:50 osse andjjj23: it uses extended POSIX regular expressions
10:52 MadHatter42 joined #git
10:52 LaT0rtue joined #git
10:53 aguimaraes1986 joined #git
10:54 dan2k3k4 joined #git
10:54 apotry joined #git
10:54 andjjj23 Ah ok. Well that's a bummer. I'm sure I can hack together something that escapes the regex characters, but it's going to feel very brittle
10:55 glauxosdever https://www.samba.org/~jelmer/dulwich/apid​ocs/dulwich.repo.BaseRepo.html#get_parents <- How can I make good use of this function to get a list of commits in a branch?
10:55 andjjj23 thanks for the help, though
10:56 johnmilton joined #git
10:58 jaguarmagenta joined #git
10:59 andjjj23 I believe the following sed will do the trick, but we'll see: sed 's/[][{}()\.^$?*+]/\\&/g'
11:00 anbz joined #git
11:03 lindenle joined #git
11:04 cdown joined #git
11:04 jast glauxosdever: I'm not familiar with that API, seems to be third party stuff, but seems like a straightforward depth-first traversal should do it
11:08 sdothum joined #git
11:09 klaas__ can I stop git from removing remote branches that are not local on push?
11:11 davisonio joined #git
11:12 rnsanchez joined #git
11:13 reverse_light joined #git
11:14 vls-xy joined #git
11:15 linuxmint joined #git
11:15 cgundersson joined #git
11:17 SimonNa joined #git
11:18 maucar joined #git
11:18 ForgottenPlayer joined #git
11:20 andyhoang1 joined #git
11:20 Psychiatrist joined #git
11:21 AbsurdTech joined #git
11:22 aavrug joined #git
11:23 mdw joined #git
11:23 alcohol klaas__: I don't think git does that..
11:24 fracting joined #git
11:25 leonarth joined #git
11:25 nemesit|znc joined #git
11:25 Pulp joined #git
11:25 andjjj23 left #git
11:26 fwdit joined #git
11:27 Marny joined #git
11:27 Shashikant86 joined #git
11:27 klaas__ alcohol: it does here, but I think I figured it out on my own, I had set mirror = true
11:27 CROOKED^HILLARY^ joined #git
11:29 davisonio joined #git
11:30 h264 joined #git
11:30 bibble joined #git
11:35 cgundersson joined #git
11:36 leonarth joined #git
11:39 robogoat joined #git
11:41 ochorocho__ joined #git
11:41 xenthree3 joined #git
11:41 xenthree3 left #git
11:41 bibble left #git
11:42 megatronous joined #git
11:44 bibble joined #git
11:45 maxis joined #git
11:47 Xethron_ joined #git
11:48 Asiajey joined #git
11:49 hoxolotl joined #git
11:49 nnyk_ joined #git
11:50 leonarth joined #git
11:50 hoxolotl hi there, is there a git documentation somewhere that was written from a user perspective?
11:51 selckin all of them?
11:51 osse !book
11:51 gitinfo There are several good books available about git; 'Pro Git' is probably the best: http://git-scm.com/book but also look at !bottomup !cs !gcs !designers !gitt !vcbe and !parable
11:51 osse !tutorial
11:51 gitinfo A list of interactive git tutorials: http://try.github.com
11:51 osse man gittutorial
11:51 gitinfo the gittutorial manpage is available at http://jk.gs/gittutorial.html
11:51 bjoe2k4_ joined #git
11:52 mdw joined #git
11:54 hoxolotl that last one was the user-perspective one I was looking for :)
11:54 hoxolotl thanks
11:55 narendraj9 joined #git
11:55 ojdo joined #git
11:57 MattMaker joined #git
11:57 * hoxolotl is hitting brain barriers.
11:58 cdown joined #git
11:58 hoxolotl I got too comfy using svn ;)
11:59 fstd joined #git
11:59 liquidjamm joined #git
12:00 hoxolotl https://jk.gs/gittutorial-2.html is a 404
12:01 davisonio joined #git
12:01 narendra` joined #git
12:02 Meteorhead joined #git
12:02 hoxolotl is there a way to pull such as it overwrites the local changes?
12:03 deadnull_ joined #git
12:03 tobiasvl hoxolotl: just the local, unstaged changes? or the local branch's commits that the remote doesn't have?
12:04 hoxolotl git fetch --all... maybe *reads some more*
12:04 deadnull_ joined #git
12:04 lindenle joined #git
12:05 tobiasvl fetch is not a pull. but if you want the second thing I said, you can fetch first and then git reset --hard to the remote-tracking branch, for example
12:05 * hoxolotl will run a lot into the "please move or remove files before you can merge" as he is using Unity, and Unity modifies files.
12:05 Darcidride joined #git
12:05 glauxosdever If anyone is familiar with the dulwich python API, can he tell me how can I get the list of files that are present in a commit tree? (Not only the files that were changed, but also unmodified files from previous commits.)
12:05 livingstn joined #git
12:05 deadnull_ joined #git
12:06 hoxolotl basically I want to overwrite my local copy with everything from the dev branch, no questions asked.
12:07 Sceorem joined #git
12:07 tobiasvl hoxolotl: what you said above, with "please move or remove files before you can merge", makes me think you just want to throw away the unstaged/uncommited local changes. what you said next though, "I want to overwrite my local copy with everything from the dev branch, no questions asked", makes me think you want to throw away all local history too and replace it with the remote dev branch.
12:07 tobiasvl which is it?
12:08 hoxolotl also, warning: I am an artist, if there is something I should not do because *reason given in cryptic coder language* there is a very high probability I will do just that because *it's logical for the common mortal*.
12:08 synthroid joined #git
12:09 * hoxolotl just had a small epyphany
12:09 hoxolotl about git being a local branch/svn server too -ish
12:09 thana joined #git
12:09 thana hello
12:09 gitinfo thana: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
12:09 * hoxolotl also head desked.
12:10 |2701 joined #git
12:10 tobiasvl hoxolotl: git is a distributed version control system, yes
12:10 baba_ joined #git
12:10 hoxolotl tobiasvl, this means I will be able to mess up twice instead of just once.
12:11 tobiasvl hehe.
12:11 hoxolotl once on my local copy, and once in the development branch.
12:11 hoxolotl woohoo.
12:11 thana im trying to cope with wnidows linenedings on linux. i added a .gitattributes with the single line "*.java eol=crlf". after that git consideres every single line in ebvery single java file changed -- why is that? (of course the line breaks ARE \r\n)
12:14 arahael4 joined #git
12:14 eletuchy joined #git
12:15 fracting joined #git
12:15 hoxolotl I want to "download" everything from the development branch and overwrite the local files. However I do NOT want it to delete/change the uncommitted files, or even care about those.
12:15 notebox joined #git
12:15 wxg joined #git
12:15 _rgn git reset --hard origin/development-branch
12:16 Guest76930 joined #git
12:17 rwb joined #git
12:17 * hoxolotl just realised what he explained he wanted was wrong.
12:18 hoxolotl I just need the new files, and ignore the conflicted files.
12:18 cdesai joined #git
12:19 leonarth joined #git
12:20 _rgn that is more complicated
12:20 nettoweb joined #git
12:21 jost__ joined #git
12:21 hoxolotl also, everytime I hear a command I'm looking up the documentation, not going to blindly type in things before I knoew exactly what they do ;)
12:23 dopesong joined #git
12:24 Guest92035 joined #git
12:24 parv joined #git
12:26 * hoxolotl reads some more.
12:26 jimmybuffet joined #git
12:26 hoxolotl great I'm going to have to make a feature request.
12:26 fuchstronaut joined #git
12:26 DevAntoine joined #git
12:27 moritz can't you merge with -X ours ?
12:27 moritz that resolves all conflicts in favor of the current branch
12:27 fwdit joined #git
12:30 bronson joined #git
12:30 hoxolotl moritz, I'm looking into what you're suggesting, this might take a while. Thanks.
12:31 Eoim4 joined #git
12:32 CheckDavid joined #git
12:33 thana is it possible to get hexdumps of diffs (of textfiles)
12:33 thana git diffs shows differences i doubt... i'd love to compare them on binary level :/
12:34 tobiasvl git diff --binary ?
12:34 Ali1 joined #git
12:34 thana this doesnt change anything
12:35 thana i'd expect and output like that of xxd for the things git thinks have changed
12:35 davisonio joined #git
12:35 bollullera joined #git
12:36 ahmedelgabri joined #git
12:36 bollullera left #git
12:37 ewet joined #git
12:37 rchavik joined #git
12:38 xanadu joined #git
12:40 thana ok got it
12:40 parv_ joined #git
12:40 thana niot that  what i was looking for bt at leas i now know the changes
12:40 tobiasvl what did the trick?
12:41 thana i didnt got the output i expected. but i still found out what changed. its is REALLY annoying and distracting that git diff does not properly ahndly windows line breaks
12:41 thana i configured to use crlf so i would expect the ^M chars do not appear in the output of git difff
12:41 bitkiller joined #git
12:41 tlaxkit joined #git
12:42 tinajohnson___ joined #git
12:42 ahmedelgabri joined #git
12:43 cagedwisdom joined #git
12:44 jagob joined #git
12:49 espindola joined #git
12:49 italoacasas joined #git
12:49 Gsham joined #git
12:50 Gsham joined #git
12:51 baba_ joined #git
12:51 jimmybuffet joined #git
12:52 Psychiatrist joined #git
12:54 dmto joined #git
12:54 docnvk joined #git
12:55 digidog joined #git
12:55 kus joined #git
12:55 davisonio joined #git
12:56 thana left #git
12:56 ocbtec joined #git
12:56 deed02392 i have run a git apply on a single stash with git in cygwin and git.exe hasn't returned for several minutes now...
12:56 deed02392 the stash only took a few seconds to make - what do?!
12:56 jbitdrop joined #git
12:57 deed02392 it's using no CPU
12:57 rchavik joined #git
12:58 MadHatter42 joined #git
12:58 nanga joined #git
12:59 jaguarmagenta joined #git
13:00 dcpc007 joined #git
13:01 ph88_ joined #git
13:02 lolmac joined #git
13:02 ph88_ hey guys, i want to change my default name/email by running this (suggested by git) command: git config --global --edit    however it says:   error: cannot spawn nedit: No such file or directory   error: unable to start editor 'nedit'    i don't even like nedit .. how can i see which editors are available and pick one? I'm on windows 10
13:04 nnyk_ joined #git
13:04 ToxicFrog ph88_: it's probably basing that on the core.editor setting, or failing that on the GIT_EDITOR, VISUAL, or EDITOR environment variables (probably in that order)
13:04 ToxicFrog ph88_: so try: git config --global --get core.editor
13:05 ToxicFrog And if that does nothing, try 'env | grep nedit' to see which environment variable it's coming from.
13:05 ToxicFrog Or just 'git config --global core.editor nano' or whatever and see if that fixes it.
13:06 anu0 joined #git
13:07 dangerousdave joined #git
13:07 marc_v92 joined #git
13:08 unbalancedparen joined #git
13:10 jimi_ joined #git
13:11 ph88_ ToxicFrog, git config --global --get core.editor   gives back nothing     env | grep nedit   gives back:   ED=nedit   EDITOR=nedit   VISUAL=nedit
13:11 dcpc007 hi, i got a bug with aliases like git st "fatal git-st permission denied. i found on google that git seems to check execution permission on each folder in the PATH before trying to use an alias ????
13:12 ph88_ ToxicFrog, git config --global core.editor nano    also does nothing
13:12 dcpc007 i have an old git version (working centos 6.7 ...), is it confirmed ? seems near like a bug
13:12 ph88_ ToxicFrog, ah actually it sets nano ... but now it says: unable to find nano
13:12 _ikke_ ph88_: You need to set it to something that is installed
13:13 _ikke_ An editor on your pc
13:13 ph88_ vim was installed ^^
13:13 ph88_ can i use sublime text too ?
13:13 _ikke_ sure
13:13 _ikke_ But you need to set the full path I think
13:13 _ikke_ (and perhaps use subl.exe, not sublime.exe)
13:15 ph88_ like this   git config --global core.editor /c/Program\ Files/Sublime\ Text\ 3/subl.exe   or like this?   git config --global core.editor "C:\Program Files\Sublime Text 3\subl.exe"  ?
13:15 rchavik joined #git
13:15 ph88_ or even    git config --global core.editor "C/c/Program Files/Sublime Text 3/subl.exe"  ?
13:17 txdv why is there no option to just check out a file of a specific version and put it right there
13:17 txdv why does git have to be so shuitty
13:17 espindol_ joined #git
13:17 perlpilot txdv: eh?
13:17 _ikke_ txdv: like git checkout <version> -- <file>?
13:17 hoxolotl txdv, because it's built to keep you frome messing up other people's work ;)
13:18 txdv why is it staging it then?
13:18 perlpilot txdv: I think one of those "it"s is probably you.
13:18 _ikke_ Because it's built to do that
13:18 cgundersson joined #git
13:18 Gloomy joined #git
13:19 _ikke_ If you want to skip the index: git show <version>:file >file
13:19 txdv no, i just did git checkout <hash> -- <file>
13:19 txdv and it just staged it immediately
13:19 _ikke_ yes
13:19 tvw joined #git
13:19 _ikke_ git checkout stages, correct
13:20 perlpilot txdv: git's whole purpose is to *save stuff*.   you don't want it to act contrary to that purpose usually.
13:20 _ikke_ "Updates files in the working tree to match the version in the index or the specified tree."
13:20 kus joined #git
13:20 txdv the file is already commited
13:20 txdv perlpilot: how does staging accomplish your point?
13:21 _ikke_ txdv: the simple explenation is because it's not a common workflow
13:21 ToxicFrog ph88_: it's been years since I used git on windows, but msys and cygwin both understand windows paths, so I think something like: git config --global core.editor "C:/Program Files/Sublime Text 3/subl.exe"
13:21 _ikke_ and so noone bothered to implement skipping the index
13:21 txdv so how do i unstage it?
13:21 ToxicFrog ph88_: except you probably want to add a -w and quote the path since it has spaces in it
13:22 ToxicFrog ph88_: so actually: git config --global core.editor "'C:/Program Files/Sublime Text 3/subl.exe' --wait"
13:22 _ikke_ txdv: git reset file
13:22 ph88_ thanks ToxicFrog that worked
13:23 txdv thanks _ikke_
13:23 txdv it totally sucks but whatever
13:24 Gsham joined #git
13:24 osse txdv: make an alias that does both in one swoop.
13:25 txdv i will call it git justdoit
13:25 _ikke_ dwim
13:25 txdv or rather JUST_DO_IT
13:25 ToxicFrog ph88_: huzzah
13:25 cdg joined #git
13:25 ph88_ :D
13:25 osse git config alias.dwim '!git checkout "$1" -- "$2" && git reset HEAD -- "$2" && :'
13:25 osse git dwim abc1234 path/to/file.txt
13:27 _ikke_ why the && :?
13:28 _ikke_ looks like a noop to me
13:28 osse in case you eff up and define $3 and $4
13:28 osse because git not only defines the positional arguments; it *also* appends them to the end of the command line
13:29 _ikke_ ah ok
13:29 osse so you either have to do &&: or define a function in the alias
13:29 dunk If you're using a forking scheme, and github requires manual conflict resolution, it suggests to do a --no-ff  merge locally, and then push to origin master. But in a forking scheme origin master isn't the repo where the conflict is a problem - that's in upstream. Can I just replace that last line with "git push upstream master"
13:29 dunk I'm a bit lost
13:30 osse that said:  git config alias.dwim '!hash=$1; shift; git checkout "$hash" -- "$@" && git reset HEAD -- "$@" && :'
13:30 osse git dwim abc1234 path/to/file.txt other/path.txt
13:32 txdv where does it store that alias?
13:32 Woody_ joined #git
13:32 txdv cant find it in ~/.gitconfig
13:32 osse in .git/config
13:32 troulouliou_div2 joined #git
13:32 _ikke_ if you don't use --global, it stores it in .git/config
13:32 mmlb joined #git
13:32 osse I forgot to put --global
13:33 baba_ joined #git
13:33 txdv interesting default setting
13:34 osse _ikke_: alternative: git config alias.dwim '!f() { hash=$1; shift; git checkout "$hash" -- "$@" && git reset HEAD -- "$@"; }; f'
13:34 strk joined #git
13:34 MarkyC joined #git
13:34 osse not sure which is most common. I think the && : way looks less bad
13:34 strk what does it take to make a repository cloneable over http ?
13:34 strk or, why can't I clone https://strk.kbt.io/tmp/hello/.git/ ?
13:34 acetakwas joined #git
13:35 _ikke_ you probably need to run git update-server-info
13:35 _ikke_ (because of 'dumb' http)
13:35 osse txdv: there are a bunch of config things that only make sense to have in .git/config. I suppose you could argue that Git could infer --global based on which specific setting you modify, but I think that would just lead to even more confusion
13:36 Gamecubic joined #git
13:36 b00^wk joined #git
13:36 b00^wk hello
13:38 strk how do I find out what "git update-server-info" did ? (I usually create a git repo and git diff to tell, but sounds like a recursive problem here :)
13:38 osse strk: it updates stuff in .git/info
13:38 osse man git-update-server-info
13:38 gitinfo the git-update-server-info manpage is available at http://jk.gs/git-update-server-info.html
13:39 strk osse: thanks :)
13:42 ochorocho__ joined #git
13:42 tvw joined #git
13:44 leeN joined #git
13:45 djb-irc joined #git
13:45 MattMaker joined #git
13:47 BlessJah joined #git
13:47 AaronMT joined #git
13:48 baba_ joined #git
13:50 docnvk joined #git
13:51 ferr1 joined #git
13:53 ph88_ left #git
13:54 GavinMagnus joined #git
13:57 digidog joined #git
13:57 GavinMagnus left #git
13:59 Tom___ joined #git
14:00 timmu joined #git
14:00 b00^wk hey, say i have a local repo, and i have one remote origin; then i go and  do git remote add --mirror=push nusvr xxxx;   how will that new remote be treated, how will it work ?
14:01 b00^wk I've added another remote like that, to my local repo, with intention to just always push to both places, but now seems something got screwed up ..
14:01 frem_ joined #git
14:01 JJM joined #git
14:01 Shashikant86 joined #git
14:02 phaleth joined #git
14:03 ash_workz joined #git
14:05 parv_ joined #git
14:06 dopesong joined #git
14:06 leehambley joined #git
14:06 bruce_lee joined #git
14:06 bruce_lee joined #git
14:07 SpaghettiCat left #git
14:09 guardian hello, I would like to write down a commit with instructions on how to use git replace to graft histories. What's the idiomatic command to get the SHA1 of the commit that comes just after the initial commit of a ref?
14:09 Shashikant86 joined #git
14:10 bjoe2k4 joined #git
14:10 guardian git rev-list HEAD | tail -2 | head -1 is too scary already
14:11 dmto joined #git
14:13 sd512 joined #git
14:14 j416 guardian: you can't.
14:14 skulltip joined #git
14:15 hoxolotl yay...
14:15 GavinMagnus joined #git
14:15 osse git rev-list --reverse HEAD | sed -n 2p   \o/
14:15 hoxolotl I deleted files locally, how do I get git to go get those from the main branch? pull doesn't work.
14:16 j416 guardian: it can be zero or many
14:16 j416 guardian: or one or two
14:16 j416 hoxolotl: git checkout -- thosefiles
14:16 jason237 joined #git
14:16 j416 hoxolotl: git is distributed, you already have the files.
14:16 hoxolotl revert all the files?
14:17 j416 hoxolotl: did I say revert? I think I said git checkout -- thosefilers
14:17 j416 files* even
14:17 j416 cant type today
14:18 skulltip can i use git gui on windows to do work on a git repo on a linux server? it seems windows wants to create its own repository, what is a good tool to use
14:18 hoxolotl ok, I *think* that worked.
14:19 fahadash joined #git
14:20 jaafar joined #git
14:20 guardian hmm alrught I'll just commit instructions using git replace XXX YYY with explicit SHA1s then
14:20 guardian thanks
14:20 ahmedelgabri joined #git
14:21 * hoxolotl grocks the checkout is checking out from the local branch on his PC, and is not checking out from the main remote development branch. and hopes he grocked correctly.
14:21 jstimm joined #git
14:22 fuchstronaut joined #git
14:23 bjpenn joined #git
14:23 dopesong joined #git
14:23 maestrojed joined #git
14:25 thiago joined #git
14:27 chrisshattuck joined #git
14:27 Kulrak joined #git
14:27 Kulrak joined #git
14:28 mekhami i'm not understanding something. I have a commit which removed a bunch of folders and their contents. I used git rm -rf folder/ to do this. I committed that, pulled it down to another computer, and those folders are all there again, even though the commit is there and the diff shows those folders being removed
14:29 sssilver joined #git
14:31 marc_v92 joined #git
14:33 baba_ joined #git
14:35 bgamari joined #git
14:35 shinnya joined #git
14:36 Panacea joined #git
14:36 IWishIKnew joined #git
14:37 bremner skulltip: git is mainly about working on local repos.
14:37 ToxicFrog mekhami: what does "git status" say?
14:37 thorsten` joined #git
14:37 thorsten` how can I make `git clone` fail if it clones an empty repository?
14:37 bgamari joined #git
14:37 mekhami ToxicFrog: Nothing noticeable?
14:38 mekhami ToxicFrog: https://gist.github.com/mekhami/​495b91fa0ae626ab03bf72a3ca204542
14:38 ToxicFrog mekhami: so, the directory still appears in your working copy, but "git log" and "git show" show it being removed, and "git status" doesn't list it as "untracked" or the like?
14:38 mekhami correct
14:38 aard joined #git
14:38 osse thorsten`: !xy
14:38 gitinfo thorsten`: 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...
14:38 ToxicFrog What does "git log -- path/to/directory" say?
14:39 xelxebar joined #git
14:39 pelegreno joined #git
14:40 mekhami it shows the commit where i removed the folder
14:41 osse thorsten`: if this is for some scripting purpuse I suppose you could grep the output of git clone.
14:41 dsdeiz_ joined #git
14:41 ToxicFrog mekhami: is it in gitignore?
14:42 bgamari joined #git
14:42 thorsten` osse: https://aur.archlinux.org/ offers git repositories. e.g. https://aur.archlinux.org/somename.git . However, if the repository does not exist one receives an empty repo. And indeed in a script, I want to detect that
14:42 thorsten` I want to detect the case 'there is no package/repo with that name'
14:42 Seveas thorsten`: that's a broken server setup.
14:42 mekhami ToxicFrog: no
14:42 b00^wk if i have mirror=true for a remote in the .git/config, how will it affect the pushes i make ?
14:42 Seveas before cloning, check with ls-remote whether there is an actual repo there.
14:43 Seveas and yell at the aur people to fix their brokennes...
14:44 ToxicFrog mekhami: what shows up in git status if you 'git add' or 'git rm --cached -r' that directory?
14:44 mekhami git add nothing
14:44 yoe joined #git
14:44 thorsten` Seveas: thanks for ls-remote!
14:44 mekhami git rm --cached -r folder/ => fatal: pathspec 'folder' did not match any files
14:44 osse wow that is weird
14:45 osse will ls-remote also trigger some weird creation of a repo?
14:45 hashpuppy joined #git
14:45 * osse runs ls-remote in a loop against AUR
14:45 thorsten` osse: no, ls-remote does not create any repo
14:45 osse how do you know? :p
14:45 b00^wk if I replace the line fetch = +refs/heads/*:refs/remotes/origin/* with  mirror=true for a second remote in the .git/config,, when i do git push --dryrun, when its mirror it wants to seems create 2 branches, * [new branch]      origin/HEAD -> origin/HEAD* [new branch]      origin/master -> origin/master
14:45 qt-x joined #git
14:45 b00^wk but without mirror, not
14:46 b00^wk can someone help, me lost
14:46 crameth joined #git
14:46 thorsten` osse: and I doubt that it's created on demand. just for the case of clone, an empty repo is delivered
14:47 narendraj9 joined #git
14:47 csd_ joined #git
14:47 Seveas thorsten`: ls-remote and clone do similar things.
14:47 Seveas so it may just create things on the remote end
14:47 mekhami ToxicFrog: i've never seen anything like this @.@
14:47 thorsten` Seveas: git ls-remote -h https://aur.archlinux.org/somename.git # has empty output
14:48 stantonk joined #git
14:48 ToxicFrog mekhami: ok, so the message from 'git rm --cached' indicates that it's not present in HEAD even if it is present on disk
14:48 mekhami yeah
14:48 ToxicFrog mekhami: and in fact you can confirm this with 'git ls-tree HEAD'
14:48 ToxicFrog ...what OS are you on?
14:48 mekhami i just removed the entire directory above it, then reset --hard and got only the files i want
14:48 mekhami that's a weird one
14:48 mekhami os x
14:48 hashpuppy i forked the main repos.  my coworker john also forked the main repos.  i want to try out some of his stuff.  if i add his repos as a remote to mine, named "john", how can i checkout his develop branch and rename it to not confuse it with my develop branch?
14:48 osse thorsten`: yes, but why doesn't it rather say "unable to access"
14:48 osse like it does for http://dfbgsgfdfasg.git
14:48 ToxicFrog mekhami: Do you by any chance have other files/directories with the same name/path but different capitalization?
14:49 mekhami ToxicFrog: no
14:49 ToxicFrog git behaves really weirdly on case-insensitive filesystems.
14:49 ToxicFrog Huh.
14:50 aard__ joined #git
14:50 mekhami so i guess what happened is
14:50 durham joined #git
14:50 mekhami just replaying that commit won't remove the files from your local filesystem
14:50 mekhami git rm -rf just gets rid of the file from the git snapshot
14:50 robotroll joined #git
14:50 thorsten` osse: is this a serious question?
14:50 mekhami so you still have to go in and manually remove the file, it's not tracked anymore
14:51 osse thorsten`: yes. By all accounts ls-remote against this host has the same weirdness as clone has (ie. thinking there is an empty repo there)
14:51 clemf joined #git
14:52 mmlb joined #git
14:52 digidog joined #git
14:52 thorsten` osse: ok. then the answer is: for http://dfbgsgfdfasg.git there is no host with that name so name resolving fails. and even if name resolving didn't fail, there is no git-daemon to talk to
14:53 osse hmm
14:53 ToxicFrog mekhami: the thing is, it should, and in my experience it does
14:53 rkazak joined #git
14:53 osse indeed, git ls-remote http://google.com/somename.git gives a different error
14:53 ToxicFrog If I delete a file on one machine, push that commit, and then fetch and merge that commit on another machine, the files vanish
14:54 mekhami that's how i would think it works too
14:54 ToxicFrog If I have uncommited changes to them, the merge fails; if I have committed changes, it enters conflict resolution mode.
14:54 mekhami indeed it didn't in this case..
14:54 osse the same error you git from http://github.com/validuser/invalidrepo.git
14:54 osse heh, git
14:54 andyhoang1 joined #git
14:54 ToxicFrog Making things even weirder is the thing where it's not in HEAD and is on disk but doesn't appear in git status and doesn't appear in any .gitignore
14:54 ToxicFrog That doesn't seem like it should even be possible.
14:55 mekhami definitely
14:55 ToxicFrog Every file in the working copy but not in HEAD that isn't ignored should appear in the status output.
14:55 mekhami i'm at a loss
14:55 digidog joined #git
14:55 thorsten` osse: anyway. I'm quite happy with pre-checking ls-remote :)
14:56 ToxicFrog mekhami: so am I. I don't suppose you have a tarchive of the repo in this state you can share?
14:57 mekhami i really wouldn't recommend it heh. i'm removing several thousand files that shouldn't have ever been in the repo to begin with
14:57 mekhami which now exist on 10+ branches
14:57 mekhami the file would be enormous
14:57 osse thorsten`: yes I think that's the cleanest thing to do
14:57 strk left #git
14:57 osse (if that by any chance leaves a mess on the host, well, then they asked for it)
14:58 b00^wk this looks nightmarish ..git push --prune git@example.com:/new-location.git +refs/remotes/origin/*:refs/heads/* +refs/tags/*:refs/tags/*
14:58 digidog joined #git
14:58 ahmedelgabri joined #git
14:59 bjoe2k4 joined #git
15:00 kpease joined #git
15:00 bjoe2k4 joined #git
15:00 jaguarmagenta joined #git
15:00 fuchstronaut joined #git
15:00 ForgottenPlayer joined #git
15:03 foodSurprise joined #git
15:03 bjoe2k4_ joined #git
15:04 foodSurprise can anyone help me? I did a git rm... and I want to take away those changes staged for commit
15:04 foodSurprise I did git reset but they still show up in red, just not staged
15:04 selckin git status output tells you
15:04 diegoviola joined #git
15:05 rangergord joined #git
15:05 foodSurprise it tells me to do git checkout
15:05 selckin indeed
15:05 foodSurprise but that dosen't change them
15:05 rangergord are local git repos fully portable? Can I zip one, give it to someone running another distro, and they can use it without any problems?
15:05 foodSurprise theys till show up as deleted:...
15:05 _ikke_ rangergord: srue
15:05 selckin then you aren't running checkout proeprly
15:05 perlpilot rangergord: yes
15:05 _ikke_ sure
15:05 rangergord or are they dependent on the version of git that did the clone
15:05 lmat joined #git
15:06 perlpilot rangergord: nope
15:06 sssilver joined #git
15:06 NeverDie joined #git
15:06 digidog joined #git
15:06 perlpilot rangergord: at least not since the earliest days of git.
15:06 nicechap joined #git
15:06 rangergord PerlJam: so, me on my dinky ubuntu 08.04, clone and edit a repo with some archaic git, I give the .git to someone running ubuntu 16.04 with latest git, no problems?
15:06 nicechap Why does this not work? vim < git show 94bcf1642e2ba89:WCN/lib/WCN/Cont​roller/Channel/URIActionGroup.pm
15:06 foodSurprise git checkout master dosen't do it
15:06 preaction nicechap: git show ... | vim
15:06 kilo_bravo joined #git
15:07 nicechap preaction: hmmmm okay trying it out
15:07 perlpilot rangergord: correct
15:07 SporkWitch joined #git
15:07 circ-user-nYr1x joined #git
15:07 preaction nicechap: < is for file redirection. that's a command, so you need pipe
15:07 rangergord impressive...thanks
15:07 lindenle joined #git
15:07 psyphen joined #git
15:07 Meteorhead I have a conflict on my pull request, strangely enough the atlassian frontend instructed me to git fetch myrepo.git master
15:07 rangergord perlpilot: out of curiosity, what about other OSes? Git for Windows, OSX, etc?
15:07 rangergord surely there's differences on Windows!
15:08 Meteorhead and then git checkout master  && git merge FETCH_HEAD
15:08 foodSurprise Looks like I actually deleted the files on my local working copy
15:08 perlpilot rangergord: Not sure about Windows, but I've copied repos between OSX and ubuntu before without problem
15:08 foodSurprise i should have done rm --index
15:08 Meteorhead thing is I don't get anything, only already up-to-date
15:08 foodSurprise now I need to pull it back agian
15:08 rangergord thanks
15:08 selckin foodSurprise: git checkout -- file
15:09 inflames joined #git
15:09 subhojit777 joined #git
15:10 nicechap preaction: I get a warning something along the lines of 'input is not from a terminal' and then something flashes, as if the file was opened and then vim crashes with 'Vim: Error reading input, exiting...'
15:10 mmlb joined #git
15:10 lindenle joined #git
15:10 preaction nicechap: then ask the fine folk at #vim how to open stdin for reading, as i don't know. or redirect the output of git show to a file and then open that
15:10 Shashikant86 joined #git
15:10 foodSurprise thanks
15:10 nicechap thanks preaction will try
15:11 perlpilot nicechap: vim <( ... )
15:12 Meteorhead can I force push like so: git push -f myremote-repo.git HEAD
15:12 Meteorhead ?
15:13 selckin usually specify a branch
15:13 sangy joined #git
15:13 Meteorhead selckin: it's on master
15:13 nicechap perlpilot: it exits with 'bash: syntax error near unexpected token `('
15:13 perlpilot nicechap: show me the exact command you used
15:13 selckin Meteorhead: so why not do master instead of head
15:14 Kicer86 joined #git
15:14 _noblegas joined #git
15:14 fornax joined #git
15:14 prsn joined #git
15:14 nicechap perlpilot:  vim < (git show 94bcf1642e2ba89:Path/To/My/File.txt)
15:14 anu1 joined #git
15:14 nicechap perlpilot: all fine if I only run this, git show 94bcf1642e2ba89:Path/To/My/File.txt
15:15 Meteorhead selckin: did and that and it fucking worked, amazing man, you saved my shitty job!
15:15 perlpilot nicechap: you added a space.  Don't do that :)
15:15 nicechap then it shows the file, but not in vim
15:15 Meteorhead selckin: thanks a lot man
15:15 perlpilot nicechap: vim <(git show 94bcf1642e2ba89:Path/To/My/File.txt)
15:15 selckin Meteorhead: np
15:15 nicechap perlpilot: darn
15:15 nicechap yes, without space it works
15:15 nicechap WTF
15:15 nicechap :-/ I are confused
15:15 perlpilot syntax.  it matters :)
15:16 Gsham joined #git
15:16 nicechap anyhow, thanks perlpilot, I have lived in a world where spaces didnt matter much
15:16 perlpilot nicechap: man bash   #  ;-)
15:16 nicechap been a lesson
15:16 nicechap thanks a lot
15:17 parv joined #git
15:17 kus joined #git
15:17 nicechap perl -e 'print "thanks perlpilot\n" x 100;'
15:17 nicechap :)
15:18 Cata|Work joined #git
15:19 nilsi joined #git
15:19 mmlb joined #git
15:19 cato joined #git
15:21 cato Hello, can I create a new branch from an existing branch on a remote repository without cloning it my local machine? How?
15:21 dmto joined #git
15:21 selckin no
15:21 _ikke_ If you talk the git protocol, you can
15:21 _ikke_ but the git client doesn't
15:21 ForgottenPlayer joined #git
15:21 Gsham joined #git
15:22 NeverDie joined #git
15:22 perlpilot cato: why would you want to make things more complicated?
15:22 nettoweb joined #git
15:23 thorsten` left #git
15:23 akushner joined #git
15:23 cato perlpilot: it is a rather large repo (.git has 3.4G)
15:24 wviana_ joined #git
15:24 bjoe2k4_ joined #git
15:24 perlpilot cato: If it's on github (or you're using the same UI as github), you can create the branch from the UI
15:24 perlpilot I'd guess that other UIs may do something similiar too
15:25 zeroed joined #git
15:25 platzhirsch joined #git
15:25 cato gitlab here, yes, that is probably the easiest way
15:27 armenzg joined #git
15:28 armenzg Hello everyone
15:28 gitinfo armenzg: hi! I'd like to automatically welcome you to #git, a place full of helpful gits. Got a question? Just ask it — chances are someone will answer fairly soon. The topic has links with more information about git and this channel. NB. it can't hurt to do a backup (type !backup for help) before trying things out, especially if they involve dangerous keywords such as --hard, clean, --force/-f, rm and so on.
15:28 tcsc any reason use git fetch --all over git remote update? i've always used the latter but apparently that's weird
15:29 _ikke_ I don't think it matters
15:30 _ikke_ except from that fetch gives you additional parameters
15:30 foodSurprise joined #git
15:31 robotroll joined #git
15:31 liamonade joined #git
15:32 bronson joined #git
15:33 flavius joined #git
15:33 LionsMane joined #git
15:33 EY___ joined #git
15:33 MattMaker joined #git
15:34 sssilver joined #git
15:34 LionsMane with git-svn, is the only way to add a svn-remote is via the .git/config file?
15:35 selckin multiple svn repos in 1 git?
15:36 bjoe2k4_ joined #git
15:37 netj joined #git
15:37 LionsMane Adding a svn repo after a git repo has been initialized, selckin
15:40 faux___ joined #git
15:41 bjoe2k4 joined #git
15:42 bjoe2k4 joined #git
15:42 Gsham joined #git
15:42 ochorocho__ joined #git
15:43 caog left #git
15:46 happy-dude joined #git
15:47 TomyLobo joined #git
15:50 Gsham joined #git
15:50 InfoTest joined #git
15:50 _3by8 joined #git
15:51 _3by8 I can't clone any repos for some reason. Whenever I run clone, I get two errors: 'fatal: 'origin' does not appear to be a git repository' and 'fatal: Could not read from remote repository.' Can someone help me figure out what's happening?
15:51 bjoe2k4 joined #git
15:52 grawity _3by8: can you `git ls-remote` those repos?
15:52 perlpilot _3by8: show your clone command.
15:52 bjoe2k4 joined #git
15:52 _3by8 perlpilot: I tried git clone git://git.kernel.org/pub/scm/git/git.git
15:52 _3by8 grawity: one moment
15:53 bjoe2k4 joined #git
15:53 CROOKED^HILLARY^ joined #git
15:53 perlpilot _3by8: maybe a firewall is blocking the git protocol?
15:53 EY___ joined #git
15:53 DoubleMan joined #git
15:54 Ali1 joined #git
15:54 grawity also can you ls-remote (or clone) either https://git.kernel.org/pub/scm/git/git.git or https://kernel.googlesourc​e.com/pub/scm/git/git.git
15:55 synthroid joined #git
15:55 troulouliou_div2 joined #git
15:56 _3by8 grawity: I can use git ls-remote on all of those repos and it returns version numbers and hashes.
15:56 _3by8 Still can't clone anything.
15:57 ril joined #git
15:59 Mobutils joined #git
16:00 ochorocho__ joined #git
16:00 _3by8 perlpilot: is there any way I could test that?
16:00 bjoe2k4 joined #git
16:01 cato left #git
16:02 fuchstronaut joined #git
16:02 Meteorhead okay, so I have figured out commits that messed up the repo, I Have a merges and commits, will git revert them work ?
16:02 Meteorhead I dont know about merges seems an option -m has to be specified
16:02 Eugene Meteorhead - !fixup
16:02 gitinfo Meteorhead: 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!
16:03 thiago _3by8: please !repro
16:03 gitinfo _3by8: [!transcript] Please paste (using https://gist.github.com/ or https://vomitb.in/ ) a transcript ( https://gist.github.com/2415442 ) of your terminal session so we can see exactly what you see
16:03 Eugene Meteorhead - !revert probably does not mean what you're used to
16:03 gitinfo Meteorhead: That's a rather ambiguous question... options: a) make a commit that "undoes" the effects of an earlier commit [man git-revert]; b) discard uncommitted changes in the working tree [git reset --hard]; c) undo committing [git reset --soft HEAD^]; d) restore staged versions of files [git checkout -p]; e) move the current branch to a different point(possibly losing commits)[git reset --hard $COMMIT]?
16:03 LeBlaaanc joined #git
16:03 madewokherd joined #git
16:03 skapunker joined #git
16:04 _3by8 !repro
16:04 gitinfo [!transcript] Please paste (using https://gist.github.com/ or https://vomitb.in/ ) a transcript ( https://gist.github.com/2415442 ) of your terminal session so we can see exactly what you see
16:04 _3by8 Oh...
16:04 Meteorhead thank you very much for your help, well I'd rather go back to a working commit (I know its hash) but Im afraid that if I want to push these changes, remote would refuse them because they are behind the tip of the branch
16:04 Eugene !force
16:04 gitinfo [!force_push] If you need to overwrite the history of a remote git repository (very bad idea, see !rewrite), you can do so with `git push --force-with-lease`. Note the remote server may reject this. See man git-config and search for receive.denyNonFastForwards. Best practice is for upstream servers to denyNonFastForwards.
16:04 Eugene (the bot is intentionally a bit scary)
16:05 _3by8 thiago: https://gist.github.com/threebyeigh​t/25cb6f19dfcfd42446a9e277993c7ba2
16:05 Meteorhead Eugene: thanks for this, I'll try force-with-lease :)
16:06 thiago _3by8: could it be a firewall or anti-virus checker?
16:06 acetakwas joined #git
16:06 thiago _3by8: are you on a corporate network?
16:06 _3by8 thiago: home network
16:07 _3by8 thiago: no antivirus on this VM and the firewall hasn't been touched.
16:07 Meteorhead Eugene: pardon my ignorance but I did a reset hard, would this be the command: git push -u --force-with-lease origin master ?
16:07 Eugene Yup
16:08 _3by8 I'm going to try disabling my hardware firewall and see if that works, I may disappear for a minute or two...
16:08 _3by8 Test...
16:08 _3by8 thiago: still nothing
16:08 bjoe2k4 joined #git
16:09 Meteorhead Eugene: says : error: unknown option `force-with-lease' :/
16:10 Eugene Use regular --force then; your git is too old for the newer fancy one
16:11 Meteorhead Eugene: lol ok thanks!
16:11 ajf- joined #git
16:13 f3r70rr35f joined #git
16:14 CROOKED^HILLARY^ joined #git
16:14 diegoaguilar joined #git
16:15 garethdaine joined #git
16:15 liquidjamm joined #git
16:16 PaulCapestany joined #git
16:16 thiago _3by8: I've never seen this error
16:16 _3by8 thiago: lucky me
16:17 bjoe2k4 joined #git
16:17 marcaddeo joined #git
16:17 blackwind_123 joined #git
16:18 mdw joined #git
16:18 knutstrut joined #git
16:18 marc_v92 joined #git
16:19 bjoe2k4 joined #git
16:20 pwnz0r joined #git
16:21 fracting joined #git
16:22 wrouesnel joined #git
16:23 bjpenn joined #git
16:24 maestrojed joined #git
16:25 xelxebar joined #git
16:25 annoymouse joined #git
16:25 StuartMI joined #git
16:26 _3by8 Is there some sort of global git configuration file that might contain some reference to 'origin' that's screwing up my attempts at cloning?
16:26 irqq joined #git
16:26 osse _3by8: /etc/gitconfig, ~/.gitconfig
16:27 osse the word 'origin' shouldn't appear the the git clone command
16:27 _ikke_ git config --show-origin -l
16:27 _3by8 _ikke_: unknown option...
16:27 _3by8 Have either of you seen my gist? https://gist.github.com/threebyeigh​t/25cb6f19dfcfd42446a9e277993c7ba2
16:28 griffindy joined #git
16:28 _3by8 My .gitconfig looks fine...
16:28 _ikke_ does just git config -l show something?
16:29 _3by8 _ikke_: shows the contents of .gitconfig
16:29 _ikke_ But nothing with remotes?
16:30 _ikke_ _3by8: Does it only happen with that repository, or with all repositories?
16:30 _3by8 _ikke_: all repositories
16:30 osse post checkout hook?
16:30 osse is git a wrapper?
16:30 osse type git
16:31 _3by8 returns help
16:31 _3by8 usage, etc
16:31 _3by8 _ikke_: nothing with remotes
16:31 osse _3by8: type 'type git'
16:31 _ikke_ which git
16:32 bjoe2k4 joined #git
16:32 _3by8 both of those commands here: https://gist.github.com/threebyeigh​t/fde5037ecae965a1a2f099aaaa74b41f
16:32 glauxosdever Using the dulwich python API, I'm wondering how can I get the list of files that are present in a commit tree? (Not only the files that were changed, but also unmodified files from previous commits.)
16:32 Meteorhead Running out of ideas here, I have removed some faulty commits (git reset --hard) on local and remote branch but when I attempt to do a pull request on upstream, it rejects my changes, saying that my ( n - 3) is already there
16:32 osse _3by8: try  GIT_TRACE=1 git clone ...
16:32 imachuchu joined #git
16:33 Meteorhead I'm using git version 1.8.4.3
16:33 bjoe2k4 joined #git
16:33 xbob joined #git
16:33 _3by8 osse: https://gist.github.com/threebyeigh​t/041408199b7d9643ce19c184d66b63d0
16:35 _ikke_ _3by8: what version of git do you have?
16:35 _3by8 _ikke_: 2.7.4
16:35 parv_ joined #git
16:35 Gsham joined #git
16:36 _ikke_ I added the output I get
16:38 _ikke_ What OS?
16:38 _ikke_ oh, ubuntu
16:38 guardian it's interesting git rev-list -n 1 gets applied before --reverse
16:38 _3by8 _ikke_: so the difference is that the first command your version of git is running is git-remote-https vs mine which is git-upload-pack
16:38 skulltip using windows git gui,  do i clone a repository, make changes, then commit it back to the linux server?
16:38 _ikke_ _3by8: correct
16:38 fahadash joined #git
16:38 skulltip or is it called 'checking out'?
16:38 _3by8 _ikke_: do you think it's a problem with the version? If I built from source do you think that'd fix it?
16:39 _ikke_ _3by8: Let me check
16:39 bjoe2k4 joined #git
16:39 _ikke_ skulltip: You clone a repository once, then you can work on it, locally. then you commit (locally), and then you can push those changes back to the server
16:39 mmlb joined #git
16:39 _ikke_ You checkout branches
16:40 CROOKED^HILLARY^ joined #git
16:40 _ikke_ _3by8: And I see the protocol also doesn't matter, right
16:40 skulltip ty _ikke_
16:40 _ikke_ the first time you used git:// the other time https
16:40 _3by8 _ikke_: correct
16:41 cgundersson joined #git
16:41 compeman joined #git
16:41 EduardoMartins joined #git
16:41 _ikke_ building 2.7.4 now
16:42 skulltip if two or more push, how does it know not to overlay each other's code
16:42 _ikke_ The first one wins, the second one first needs to merge the history from the first one
16:43 _ikke_ (this is per branch)
16:44 _3by8 _ikke_: I appreciate the help.
16:44 guardian it's interesting git rev-list -n 1 gets applied before --reverse <-- I expected git rev-list --reverse -n 1 to give me the root commit
16:44 _ikke_ _3by8: still uses git-remote-https for me
16:44 skulltip ok more dumb questions, a branch is like an experimental version?
16:45 _ikke_ skulltip: branches can be used for many things
16:45 skulltip a new tree can grow from a branch i take it
16:45 _3by8 _ikke_: what could it be? This is driving me absolutely nuts. What's causing mine to use some alternate command?
16:45 _3by8 I'm going to remove Git and reinstall from source.
16:45 garethdaine joined #git
16:45 parv_ joined #git
16:46 sebboh left #git
16:46 _ikke_ first try a local installation :-)
16:46 bjoe2k4 joined #git
16:46 _ikke_ make && make install installs it in your home dir
16:46 peepsalot joined #git
16:48 _3by8 _ikke_: will do
16:49 davisonio joined #git
16:50 davisonio joined #git
16:52 Jellyg00se joined #git
16:52 eijk joined #git
16:52 Terens joined #git
16:52 baba_ joined #git
16:53 kus joined #git
16:55 bjoe2k4 joined #git
16:55 glauxosdever Using the dulwich python API, I'm wondering how can I get the list of files that are present in a commit tree? (Not only the files that were changed, but also unmodified files from previous commits.)
16:57 glauxosdever I asked in their channel, no response (only 13 people there (13 including me)).
16:57 glauxosdever So I thought this is a more appropriate place.
16:58 glauxosdever Yes, I looked at the documentation, but it's so big.
16:58 moritz I think most folks here are more familia with the command line client
16:59 sssilver joined #git
16:59 glauxosdever moritz: I'm pretty sure about this.
16:59 _ikke_ glauxosdever: What do you have already?
16:59 glauxosdever In regards to what aspect?
16:59 glauxosdever I can read the list of branches, and the list of commits.
16:59 thiago joined #git
17:00 _ikke_ If you have a commit, you can get a tree from that
17:00 glauxosdever Yes. I don't know how can I do it.
17:00 ajf- joined #git
17:00 _ikke_ commit.tree
17:01 jaguarmagenta joined #git
17:01 Terens joined #git
17:01 bjoe2k4 joined #git
17:02 moritz ... sounds complicated :-)
17:02 liamonade joined #git
17:02 bjoe2k4 joined #git
17:02 fuchstronaut joined #git
17:03 akushner joined #git
17:03 tlaxkit joined #git
17:03 glauxosdever If you finally find the correct submodule it is easy.
17:04 jonmorehouse joined #git
17:06 mdw_ joined #git
17:06 glauxosdever https://www.dulwich.io/apidocs/dulwich.html <- Except I don't find the commit submodule.
17:06 Noldorin joined #git
17:06 glauxosdever moritz: Now you see how complicated it is...
17:07 _3by8 _ikke_: I'm awful with the command line, it's installed in bin/ within my home directory. I can execute it with ./git in that directory but how do I execute it outside of that directory?
17:07 glauxosdever _3by8: You provide a relative path to it.
17:07 bjoe2k4 joined #git
17:07 _ikke_ _3by8: full path
17:07 glauxosdever If you are in ~/somedir, then you run it with ../git
17:07 _3by8 glauxosdever: I tried ../bin/git but that didn't work, what's the syntax to execute it from a relative path?
17:08 _ikke_ _3by8: I usually do: GIT=$PWD/git
17:08 _ikke_ then you can use $GIT anywhere
17:08 glauxosdever Ah, it's inside bin.
17:08 glauxosdever So from the ~ directory, ./bin/git
17:09 _ikke_ Gamecubic: t = r.get_object(o.tree)
17:09 CJKinni joined #git
17:09 _ikke_ That gives you a tree object
17:09 _ikke_ o is a commit object
17:09 glauxosdever _ikke_: Was that by chance to me?
17:09 _ikke_ glauxosdever: yes
17:10 _ikke_ r is a repo object
17:10 cdg joined #git
17:11 dmto joined #git
17:11 solf1re joined #git
17:12 solf1re joined #git
17:12 glauxosdever Is it a sha?
17:12 cdown joined #git
17:12 glauxosdever (The commit object.)
17:13 bjoe2k4 joined #git
17:13 _ikke_ The object is a Commit object
17:14 _3by8 _ikke_: I think I've figured out the problem. I didn't try cloning from other directories. I forgot that my projects folder is actually an HGFS share and for some reason the user is set to root on all folders within it. Not sure how to fix that yet but I have a feeling once I get the user sorted on that, it should be fine.
17:14 glauxosdever And what is its representation?
17:14 _3by8 I mean I know how to change a user but it keeps changing it back because of HGFS T_T
17:15 ojacobson _3by8: is the host filesystem a Windows filesystem?
17:15 _3by8 ojacobson: yessir!
17:15 solf1re joined #git
17:15 ojacobson VMWare's hgfs doesn't map linux perms to windows perms, it just squashes to root and rwxrwxrwx by default
17:15 ojacobson you can twiddle the mount options (eg. in fstab) to change the uid and perms, if I remember right
17:16 _3by8 ojacobson: sounds like fun, thanks for the tip :)
17:16 sosedoff_ joined #git
17:16 ojacobson Sharing a git repo between two hosts is fraught, though  :)
17:17 _ikke_ glauxosdever: It represents a commit..
17:17 bjoe2k4 joined #git
17:17 glauxosdever Yes, that's obvious. :p
17:18 glauxosdever The question was in what form.
17:18 _ikke_ glauxosdever: https://www.dulwich.io/apidoc​s/dulwich.objects.Commit.html
17:18 glauxosdever But anyway, it seems it is irrelevant.
17:18 glauxosdever Yes, I know...
17:18 Eugene git+hgfs does not end well. Been there, done that.
17:18 tobiasvl what form dulwich represents git objects in is a dulwich api question and not a git question
17:18 tobiasvl maybe #dulwich is better for that
17:19 glauxosdever Except no one answers there (any question since 25 hours...)
17:19 glauxosdever Anyway, I thought someone knows. :-)
17:19 gmurop joined #git
17:19 dopesong joined #git
17:20 discopatrick joined #git
17:20 _ikke_ glauxosdever: I'm giving you an answer..
17:20 baba_ joined #git
17:21 _ikke_ r = dulwich.repo.Repo('repository')
17:21 glauxosdever Ok, I know to do it.
17:21 _ikke_ head = r.refs['HEAD']
17:22 _ikke_ commmit = r.get_object(head)
17:22 _ikke_ tree = r.get_object(commit.tree)
17:22 _ikke_ files = tree.items()
17:22 bjoe2k4 joined #git
17:23 bjoe2k4 joined #git
17:24 glauxosdever Now another question arises. Now the list of commits I get are in the form of sha's. Is there a simple way to return a list of Commit objects in a branch?
17:24 glauxosdever Since it's better to work with objects and not sha's.
17:25 glauxosdever commmit = r.get_object(head) # is to get the commit from working branch head.
17:25 _ikke_ Just run repo.get_object() on each hash
17:25 ajf- joined #git
17:25 ojacobson something like [r.get_object(hash) for hash in hashes] because python has real list comprehensions
17:25 _ikke_ :-)
17:26 ojacobson it does seem like a lot of this is "how do I python" - have you talked to #python? they're full of great advice on that front :)
17:26 alkisg joined #git
17:26 glauxosdever Seems I don't know python that well finally...
17:27 glauxosdever I didn't use it to its full potential.
17:27 mjuszczak joined #git
17:27 liamonade joined #git
17:27 alkisg Hi, I'm running: git clone git://anongit.freedesktop.org/accountsservice; cd accountsservice; git log
17:27 alkisg And I'm seeing commits from 2015, while https://cgit.freedesktop.org/accountsservice/ has commits from 2016. What am I doing wrong?
17:28 ojacobson glauxosdever: so go we all. The only way to get better is by doing, so you're on the right track.
17:28 bjoe2k4 joined #git
17:29 baba_ joined #git
17:29 grawity alkisg: both dates are right actually
17:29 grawity alkisg: git log --format=fuller
17:29 alkisg thanks, checking...
17:30 fscala joined #git
17:30 alkisg Yup that did it, thank you grawity, I'll read about the format=fuller part
17:31 alkisg AuthorDate vs CommitDate, got it
17:31 elastix joined #git
17:31 eletuchy joined #git
17:32 djb-irc joined #git
17:34 glauxosdever Still, the way to get the hashes is pretty ugly. What is the recommented way?
17:34 _3by8 Hi, I have an HGFS mount on an Ubuntu guest which runs on a Windows host. I understand that when it comes to permissions, everything gets squashed to root since the Windows and Linux filesystems aren't compatible but I've been told that with some tweaks to the fstab in regards to UID, I can maybe get the user and permissions just right. Any clues?
17:34 _3by8 Whoops! lmao
17:34 ahmedelgabri joined #git
17:34 _3by8 Meant for #vmware
17:35 ojacobson _3by8: man mount
17:35 ojacobson have a look at the generic options that are available for any filesystem
17:35 bronson joined #git
17:35 _3by8 ojacobson: risque
17:35 _3by8 Will do though.
17:35 bjoe2k4 joined #git
17:35 ojacobson I'm not going to dignify that. :)
17:35 EvilPenguin joined #git
17:36 _3by8 You already did ;)
17:37 glauxosdever Still, the way to get the hashes is pretty ugly. What is the recommented way?
17:38 _ikke_ What way are you getting the hashes?
17:38 parv_ joined #git
17:38 elijah Why would a single line file with `<php define('MVPD_VERSION', '576.33');` and `<php define('MVPD_VERSION', '99.9.9')` get a merge conflict? Even if I add a full docblock it still gets a merge conflict.
17:39 glauxosdever https://gist.github.com/anonymous​/d1a740318d1b70febc8640ca80fe2618
17:39 _ikke_ elijah: Because you edit the same line?
17:39 ojacobson elijah: the file has changed since the last common parent, on both arms of the merge
17:39 perlpilot elijah: two commits modify it and git can't tell which is canon.
17:39 Sasazuka joined #git
17:40 ojacobson glauxosdever: have a look at dulwich's graph walker
17:40 foodSurprise whats the best way to make a git backup? Just doing a gitclone?
17:40 ojacobson foodSurprise: it's just files, so whatever you're doing for files already will handle git just fine
17:41 PryMar56 joined #git
17:42 _ikke_ glauxosdever: https://www.dulwich.io/apido​cs/dulwich.walk.Walker.html
17:42 bjoe2k4 joined #git
17:43 bjoe2k4 joined #git
17:43 Gege joined #git
17:43 LionsMane joined #git
17:44 ahmedelgabri joined #git
17:44 elijah hrm, thanks for the feedback, I don't follow what is happening still but I will process your comments some more and try out some more experimentation
17:44 ramsub07 joined #git
17:45 ramsub07 hello, i'm trying to pull from upstream and i'm getting this error, http://pastebin.com/eXkuk25f
17:45 ReenignE joined #git
17:45 ReenignE joined #git
17:45 _ikke_ elijah: perhaps !eek helps
17:45 gitinfo elijah: [!eekaconflict] Merge conflicts are a natural part of collaboration. When facing one, *don't panic*. Read "How to resolve conflicts" in man git-merge and http://git-scm.com/book/ch3-​2.html#Basic-Merge-Conflicts then carefully go through the conflicts. Picking one side verbatim is not always the right choice! A nice video explaining merge conflicts: https://www.youtube.com/watch?v=zz7NuSCH6II
17:45 ojacobson ramsub07: finish or abandon your existing merge before starting a new one with git-pull
17:46 ojacobson git-status will tell you how
17:46 ramsub07 ojacobson: i tried git stash, didn't work
17:46 timmu joined #git
17:47 elijah I am actually trying to automate a Jenkin's script to merge back into master. This file will change a lot. I need to find a way where it won't have a conflict.
17:47 ramsub07 ojacobson: git status is asking me to commit, but i don't want to make any commits
17:47 ojacobson yes, you can't stash a merge (because of design limitations in stash, basically)
17:47 ojacobson either complete the merge or abandon it
17:47 ramsub07 ojacobson:  i want to ignore the commits
17:47 glauxosdever Ok, but if I want to get commits from one specific branch?
17:47 ojacobson glauxosdever: start the walk from the branch tip
17:47 kadoban joined #git
17:47 ramsub07 ojacobson: how do i abandon ?
17:48 ojacobson 'git merge --abort' will throw away the current merge
17:48 ojacobson including any conflict resolution you may have done
17:48 foodSurprise joined #git
17:48 akushner joined #git
17:48 ojacobson elijah: conflicts are a normal outcome of merges - they're the cases that are non-automatable
17:49 _ikke_ elijah: automatic and merging don't go well together, precisely because of this problem
17:49 ojacobson "how will I automate merges so that there are never conflicts" is an unanswerable question, because it starts with the wrong assumptions
17:49 mohabaks joined #git
17:49 ojacobson (i.e. that a conflict is in some way a mistake or a problem)
17:49 elijah Hrm, I tried again without test values, so jumping from 15.0.4 to 16.0.1 and the merge had no conflicts
17:50 elijah Yeah, it is just that I don't expect that to be a conflict
17:50 ojacobson elijah: for some demo cases, have a look at https://github.com/ojacobson/conflicts
17:50 alkisg left #git
17:50 bjoe2k4 joined #git
17:50 nnyk_ joined #git
17:51 parv___ joined #git
17:51 _ikke_ elijah: It's a conflict when the line has changed in both branches
17:51 |2701 joined #git
17:52 elijah *_ikke_ okay, thanks, maybe it did change before. I did reset both branches this last time I tried.
17:52 btyne79 joined #git
17:52 ramsub08 joined #git
17:54 akushner joined #git
17:55 bjoe2k4 joined #git
17:55 flavius joined #git
17:57 Meteorhead still nothing, I have reverted my changes on local and my remote but to update the upstream or central repo I get (already up to date)
17:57 ahmed_elgabri joined #git
17:57 bjoe2k4 joined #git
17:57 Meteorhead but then I found this on setting: Reject all force pushes (git push --force) to this repository Enabled
17:58 Meteorhead if I disable this would I be able to create a pull request with n - 3 commits ?
18:01 fuchstronaut joined #git
18:03 dmto joined #git
18:04 bjoe2k4 joined #git
18:04 djb-irc joined #git
18:04 ajf- joined #git
18:04 diegoaguilar joined #git
18:04 livingstn joined #git
18:05 djb-irc left #git
18:05 italoacasas joined #git
18:06 babilen joined #git
18:06 timmu joined #git
18:06 ceaucari joined #git
18:07 mohabaks joined #git
18:07 eletuchy joined #git
18:07 mwhooker joined #git
18:07 drawks joined #git
18:08 dopesong joined #git
18:08 drawks given a commit id what is the easiest way to determine which branches have been changed?
18:08 davisonio joined #git
18:08 ochorocho__ joined #git
18:09 glauxosdever head = repo["refs/heads/{}".format (branch)].id
18:09 glauxosdever walker = Walker (Commit, include=head)
18:09 glauxosdever What am I doing wrong?
18:10 bjoe2k4 joined #git
18:10 ceaucari joined #git
18:10 darkbit joined #git
18:12 glauxosdever What does ObjectStore do?
18:14 Gsham joined #git
18:15 dopesong joined #git
18:16 bjoe2k4 joined #git
18:16 realz joined #git
18:17 bjoe2k4 joined #git
18:18 ramsub07 joined #git
18:18 nwkj86 joined #git
18:21 dmto joined #git
18:21 timmu joined #git
18:22 bjoe2k4 joined #git
18:23 bjpenn joined #git
18:24 cdown joined #git
18:25 razieliyo joined #git
18:25 razieliyo joined #git
18:26 eletuchy joined #git
18:27 tobiasvl Meteorhead: what do you mean by a pull request with n - 3 commits?
18:28 tobiasvl Meteorhead: disabling rejection of force pushes will let you push a divergent history to the remote overwriting what is already there
18:29 bjpenn joined #git
18:29 bjoe2k4 joined #git
18:29 dreiss joined #git
18:31 Limes_ joined #git
18:31 relipse2 joined #git
18:31 Meteorhead tobiasvl: thank god, someone answered, first thank you! The thing is I have a local & remote repositories and a central repo with a web frontend. Ive noticed some faulty commits and Ive fixed those with hard resets on my local and remote
18:31 SteffanW joined #git
18:32 neunon joined #git
18:32 Meteorhead but when I attempt to do a pull request on the central repo it tells me that changes are already there because well, central repo is  at n commits and I wanna go back with it 3 commits back
18:32 tobiasvl Meteorhead: but didn't you do a hard reset on the remote too?
18:33 baba__ joined #git
18:34 MattMaker joined #git
18:34 eletuchy_ joined #git
18:34 Meteorhead tobiasvl: I tried pushing to remote now but I got failed to push to refs... what I did is git rebase -i (trying many things actually) if you have any ideas Im taking
18:34 tobiasvl Meteorhead: to be able to (force) push, you will need to disable the hook you talked about earlier.
18:35 tobiasvl as you said yourself, it's a setting that disables force pushes
18:35 tobiasvl so change that if that's what you want
18:35 bjoe2k4 joined #git
18:36 Meteorhead tobiasvl: thanks a lot man! Im gonna try this now :)
18:36 foodSurprise joined #git
18:36 Mateon1 joined #git
18:37 bjoe2k4 joined #git
18:38 ahmed_elgabri joined #git
18:39 liamonade joined #git
18:44 CJKinni` joined #git
18:45 bjoe2k4 joined #git
18:45 kfunk joined #git
18:46 Marny joined #git
18:46 bjoe2k4 joined #git
18:48 cdown joined #git
18:48 timmu joined #git
18:48 baba_ joined #git
18:49 mwhooker joined #git
18:52 dreiss joined #git
18:52 baba__ joined #git
18:52 bjoe2k4 joined #git
18:53 MrWoohoo joined #git
18:53 FabTG joined #git
18:53 charlesr joined #git
18:54 ams__ joined #git
18:54 mischat_ joined #git
18:57 baba_ joined #git
18:58 dopesong joined #git
18:59 elastix joined #git
18:59 NeverDie joined #git
18:59 bjoe2k4 joined #git
19:00 sgen joined #git
19:00 davisonio joined #git
19:01 dopesong_ joined #git
19:01 NwS joined #git
19:01 jaguarmagenta joined #git
19:05 oncall-pokemon joined #git
19:07 bjoe2k4 joined #git
19:07 auka joined #git
19:11 nfk joined #git
19:12 mwhooker joined #git
19:12 bjoe2k4 joined #git
19:13 bjoe2k4_ joined #git
19:14 bjoe2k4 joined #git
19:15 auka I'm merging master into a local branch and commits seem to be lost when the merge happens. If I do git log _file_, the commit I want is no longer there. If I use git merge -s ours, the commit shows up. What is happening here?
19:15 auka If I do git log --all _file_ without using -s ours, it shows up on another branch that was previously merged in this branch
19:15 auka What's the best way to follow what happened to a file between merges across multiple branches?
19:15 cdown joined #git
19:15 auka I can't figure out where in the history that commit got removed. It doesn't appear to have been reverted since the git log shows no reverts on that file.
19:16 multi_io joined #git
19:17 sssilver joined #git
19:17 djb-irc joined #git
19:17 tobiasvl auka: commits that are reverted still show up in history (they're not removed, just negated with the opposite diff)
19:17 tobiasvl auka: !lol
19:17 gitinfo auka: A nifty view of branches, tags, and other refs: git log --oneline --graph --decorate --all
19:17 fuchstronaut joined #git
19:18 tobiasvl and !repro – how are you merging? no squashing or rebasing?
19:18 gitinfo [!transcript] Please paste (using https://gist.github.com/ or https://vomitb.in/ ) a transcript ( https://gist.github.com/2415442 ) of your terminal session so we can see exactly what you see
19:18 bjoe2k4_ joined #git
19:19 auka I'm doing git merge master from the command line
19:20 Gsham joined #git
19:20 auka when using git log --oneline --graph --decorate --all, what does a branch beside it mean? i.e, 3b49130 (origin/some_branch_here) my commit message
19:21 tobiasvl it means that's where the branch is pointing
19:21 tobiasvl the branch is that commit
19:21 auka if I do a normal git log, that does not appear, is there a reason for that?
19:22 Eugene auka - because you didn't use --decorate? It's kinda right there in the command....
19:23 csd_ joined #git
19:23 auka that commit is not there entirely
19:23 ojacobson 'git log' logs commits reachable from HEAD
19:24 ojacobson 'git log --all' logs commits reachable from any name
19:24 Eugene `git log` defaults to showing you the log of HEAD. `git log --all` shows everything reachable from any branch/tag/ref
19:24 ojacobson (branches, tags, whatever)
19:24 ceaucari joined #git
19:24 Eugene Again, it's right there in the command you typed. !tryit ;-)
19:24 gitinfo [!tias] Try it and see™. You learn much more by experimentation than by asking without having even tried. If in doubt, make backups before you experiment (see !backup). http://gitolite.com/tias.html may help with git-specific TIAS.
19:24 BSaboia joined #git
19:25 auka okay
19:25 bjoe2k4_ joined #git
19:25 auka after I commit the merge, that commit is no longer reachable from HEAD
19:25 auka why would that happen?
19:26 auka so git log _file_ shows the commit I want before commiting the merge (although, the changes aren't actually in the file which is weird), after I commit the merge, the commit is not there at all when doing git log _file_
19:26 bjoe2k4_ joined #git
19:27 auka If I do git merge -s ours, that commit stays reachable from HEAD
19:27 auka but on a normal merge, there isn't a conflict in that file
19:27 marksei joined #git
19:27 ceaucari joined #git
19:27 auka so why does -s ours change the behaviour of the merge for that specific file/commit?
19:27 jimi_ joined #git
19:28 a3Dman joined #git
19:28 glauxosdever _ikke_: Sadly, your way of getting the list of files doesn't work out for my case. I'd like to look into subdirectories too. (With the least code possible.)
19:29 bjoe2k4_ joined #git
19:29 _ikke_ glauxosdever: Trees are trees, you have to go through them recursively
19:29 mmlb joined #git
19:30 glauxosdever Isn't there a function call that does this?
19:30 mwhooker joined #git
19:30 _ikke_ I'd have to look
19:31 troulouliou_dev joined #git
19:31 Puffball_ joined #git
19:32 bjoe2k4_ joined #git
19:32 HerrThees joined #git
19:33 FilipNortic joined #git
19:33 glauxosdever I have a potential candidate for this.
19:34 auka If I change the merge strategy to --resolve, I get an actual unmerged path in that file that says "both added"
19:34 auka the default is recursive and doesn't have that
19:34 bjpenn joined #git
19:35 dmto joined #git
19:35 m0viefreak joined #git
19:35 bjoe2k4_ joined #git
19:37 auka What does both added mean with the resolve strategy?
19:37 ochorocho__ joined #git
19:38 bjoe2k4_ joined #git
19:38 bronson joined #git
19:39 ramsub07 joined #git
19:40 glauxosdever https://www.dulwich.io/apidocs/du​lwich.objects.Tree.html#iteritems <- What can the generator be used?
19:42 mikeric joined #git
19:42 tobiasvl glauxosdever: can you rephrase your nonsensical question?
19:43 tobiasvl auka: I asked you for some terminal output earlier. I think that would help get some answers.
19:43 glauxosdever Erm, what can the generator be used for in this case?
19:43 _ikke_ The same as the list..
19:43 EvilPenguin joined #git
19:44 tobiasvl the same as any other generator (and really this turned into #python long ago)
19:44 unforgiven512 joined #git
19:45 unforgiven512 joined #git
19:45 glauxosdever Sorry, this was not my intention at all.
19:46 bjoe2k4_ joined #git
19:46 glauxosdever It's just the module git-oriented, so I ask here.
19:46 glauxosdever Hmm, on the other side it's for python.
19:46 bjoe2k4_ joined #git
19:47 eletuchy joined #git
19:47 tobiasvl no worries. but we are really just qualified to answer the git-specific questions you might have about dulwich. not things that are part of the language of python.
19:48 LeBlaaanc joined #git
19:48 bjoe2k4_ joined #git
19:50 Dyonisos joined #git
19:50 mwhooker joined #git
19:50 kelar joined #git
19:50 bjoe2k4_ joined #git
19:51 Dyonisos joined #git
19:52 faux___ joined #git
19:52 jonmorehouse joined #git
19:52 bjoe2k4_ joined #git
19:53 bjoe2k4_ joined #git
19:55 dopesong joined #git
19:55 sssilver joined #git
19:56 bjoe2k4_ joined #git
19:57 settermjd joined #git
19:57 LionsMane I added [svn-remote "svn"] (with the rest of the config, did a git svn fetch, and after an hour-ish the process of transferring the svn repository completed.  However, I don't know where the files actually went?  Or, what git checkout do I need to see the files?
19:58 dmto joined #git
19:59 HerrThees joined #git
19:59 EvilPenguin joined #git
19:59 LionsMane AH when I fetched, I needed to add the branch I wanted to check it out into?
20:00 drawks left #git
20:01 mar77i_ joined #git
20:01 ojacobson LionsMane: git branch --all will show all the fetched branches
20:02 scarabx joined #git
20:02 auka tobiasvl: I can't show git history since commits have sensitive company information
20:02 LionsMane Whoa, I got a lot of output.  I am brand new to the git-svn functionality.
20:02 bjoe2k4_ joined #git
20:03 tobiasvl auka: just show the output of log? censor the commit messages if you have to?
20:03 Terens joined #git
20:03 auka you want the output of git log only?
20:03 auka I'll show you the log before and after the merge
20:04 bjoe2k4_ joined #git
20:04 tobiasvl show whatever you think is relevant for us to help you.
20:04 zivester joined #git
20:04 auka alright
20:05 psyphen left #git
20:05 LionsMane ojacobson: So I see a list of remotes/origin/*  How is the origin in the remote git repo different than the remotes/origin from the svn fetch?
20:06 Sasazuka__ joined #git
20:06 EvilPeng1 joined #git
20:07 mr_chris left #git
20:09 jonmorehouse joined #git
20:10 mwhooker joined #git
20:10 bjoe2k4 joined #git
20:10 dmto joined #git
20:11 krnex joined #git
20:12 khmarbaise joined #git
20:12 mda1 joined #git
20:13 EvilPeng1 joined #git
20:14 dmto joined #git
20:15 mda1 joined #git
20:16 auka tobiasvl: https://gist.github.com/arianitu/​a05a10f73dffc0f37d5c2d4b8cebfc80
20:17 bjoe2k4 joined #git
20:18 stantonk joined #git
20:18 sssilver joined #git
20:18 auka so commits 3a1fb35 and 3b49120 are no longer reachable from HEAD after merging master
20:18 auka Does that mean those commits were reverted, does it mean the files they touched were deleted? How do I find out?
20:19 djb-irc joined #git
20:19 ojacobson auka: reasoning about branch history while restricting the history to a single file is not a great idea
20:20 Aboba joined #git
20:20 ojacobson take the filename out of the picture (and don't worry about the state of history after a conflicting merge but before you commit the merge; it's the same as the state before the merge)
20:21 mischat joined #git
20:22 nettoweb joined #git
20:22 auka How do I reason about the branch history then?
20:22 auka I just want to figure out why commit 3b49120 is no longer reachable from HEAD
20:23 bjoe2k4 joined #git
20:23 stantonk joined #git
20:24 ojacobson who says it isn't? you're limiting what git-log can show you by limiting it to commits affecting a single file.
20:24 ojacobson intuitively, the same commits _should_ appear in history, but I've seen weirder things.
20:25 bjoe2k4 joined #git
20:26 jonmorehouse joined #git
20:27 mostlybadfly joined #git
20:27 auka This doesn't seem solvable without a good understanding of git
20:27 auka so I guess I'll do that
20:28 bjoe2k4 joined #git
20:29 auka Is there a similar problem I could perhaps look at that might help me?
20:30 auka Specifically, merges leading to lost changes
20:30 AbsurdTech joined #git
20:30 AbsurdTech joined #git
20:31 mwhooker joined #git
20:32 Puffball joined #git
20:33 josuebrunel joined #git
20:35 bjoe2k4 joined #git
20:36 sssilver joined #git
20:36 cagmz joined #git
20:37 jonmorehouse joined #git
20:37 cdown joined #git
20:38 ramsub07 joined #git
20:38 bruce_lee joined #git
20:38 ochorocho__ joined #git
20:41 bjoe2k4 joined #git
20:42 bjoe2k4 joined #git
20:42 ojacobson > This doesn't seem solvable without a good understanding of git
20:42 ojacobson 4:27 PM
20:42 ojacobson git is effectively unusable without a good understanding of git, IMO
20:42 ojacobson the UI externalizes most of the complexity of source control (i.e. it's "simple" in the sense of containing very little complexity)
20:43 ojacobson (rather than "simple" in the sense of managing complexity)
20:45 diego1 joined #git
20:46 LionsMane When using git svn fetch in combination with a remote git repo, looking at git branches -a output, how do I tell which branches are in the svn-remote and which ones are in the remote git repo?
20:46 ajf- joined #git
20:46 EvilPeng1 joined #git
20:46 auka You can have a decent understanding of git and accomplish a lot.
20:47 auka I've read the git book, but I don't think that's enough to solve this problem
20:47 auka and most people on the team can handle https://www.kernel.org/pub/software/scm/​git/docs/howto/revert-a-faulty-merge.txt
20:47 auka whatever I'm dealing with is a lot more complicated
20:47 MaddHatter LionsMane: newer git-svns will put all the svn stuff into an svn remote, iirc
20:48 diego1 joined #git
20:49 boombatower joined #git
20:50 nnyk_ joined #git
20:51 bjoe2k4 joined #git
20:52 MaddHatter (I think git-svn was originally envisioned in the context of SVN being the repo's only remote)
20:52 dopesong joined #git
20:52 cdown joined #git
20:54 nanga joined #git
20:57 LionsMane MaddHatter: Yes I get they are supposed to live together.
20:58 diego1 joined #git
20:58 bjoe2k4 joined #git
20:58 boombatower joined #git
20:58 nettoweb joined #git
20:58 satifant joined #git
20:59 LionsMane MaddHatter: I created the element in the .git/config file and I was able to do the fetch.  But still, when I look at the branches, I am pretty confused.
21:00 LionsMane Do all the svn remote branches work exactly the same as local branches?  If I want to post a fetched svn-remote branch, I need to check it out and the push it to the remote repo?
21:01 arooni joined #git
21:02 MaddHatter LionsMane: created what element in .git/config? The remote branches are mirrors of SVN's tags/branches. You probably won't ever push to those branches, since making commits to SVN can't be done via push.
21:02 BlackMaria joined #git
21:02 jaguarmagenta joined #git
21:03 bjoe2k4 joined #git
21:04 LionsMane I added the [svn-remote …] section to my local git repo and then did a fetch against trunk.
21:04 plasticle joined #git
21:04 LionsMane Well, I did a fetch without parameters, MaddHatter
21:04 LionsMane err git svn fetch
21:05 mjuszczak joined #git
21:05 MaddHatter I've not set up git-svn by editing my config; I've always used the commands to do it.
21:06 ahmed_elgabri joined #git
21:06 MaddHatter if you look in .git/svn/refs you should be able to see what refs belong to git-svn, vs. which (notable by their absence from that location) belong to other remotes, assuming you've done the initial fetch with an old enough version of git-svn that it didn't set up a distinct remote for the svn stuff
21:07 LionsMane Ahhh what are the commands, MaddHatter.  The only one I found was git svn init or git svn clone.  Both create a completely new repo.
21:08 MaddHatter init is what I had in mind. What are you trying to do?
21:09 bjoe2k4 joined #git
21:10 LionsMane The big big picture is that I have a GitHub git repo.  On the corporate netowrk, we have an SVN repo that my group needs to get stuff from and merge into our work.  I want to have the fetches eventually make it into a git branch so we can handle merging in a sane way.
21:10 bjoe2k4 joined #git
21:10 jonmorehouse joined #git
21:10 bjoe2k4_ joined #git
21:10 LionsMane Plus, the hackhandles over on the SVN side are checking in BINARIES so we can't use diff's.
21:11 MadHatter42 joined #git
21:11 bjoe2k4 joined #git
21:12 MaddHatter So git svn init/clone the SVN thing, git remote add the github thing, fetch both, checkout a local branch from one of them, merge the branch from the other?
21:12 MaddHatter if there's a utility that can address conflict resolution for the binaries, you can probably use filters, but that's not a thing I've very familiar with.
21:13 bjoe2k4 joined #git
21:13 mwhooker joined #git
21:14 bjpenn joined #git
21:14 AbsurdTech joined #git
21:15 mjuszczak joined #git
21:16 bjoe2k4 joined #git
21:17 bjoe2k4 joined #git
21:18 dsantiago joined #git
21:18 invisbl joined #git
21:18 AbsurdTech joined #git
21:19 MattMaker joined #git
21:19 bjoe2k4 joined #git
21:20 bjoe2k4 joined #git
21:22 marcelofr joined #git
21:23 tvw joined #git
21:23 bjoe2k4 joined #git
21:25 bjoe2k4 joined #git
21:26 bjoe2k4 joined #git
21:27 bjoe2k4 joined #git
21:28 bjoe2k4 joined #git
21:28 ToBeCloud joined #git
21:28 chachasmooth joined #git
21:28 bjoe2k4 joined #git
21:29 MattMaker joined #git
21:31 bjoe2k4 joined #git
21:33 durham joined #git
21:34 Gsham joined #git
21:34 stantonk joined #git
21:34 durham joined #git
21:34 mwhooker joined #git
21:35 bjoe2k4 joined #git
21:36 bjoe2k4 joined #git
21:36 bjoe2k4 joined #git
21:38 mjuszczak joined #git
21:39 ochorocho__ joined #git
21:39 bronson joined #git
21:39 hanshenrik is there some way to tell git that a certain type of files (*.map) should not be tracked/updated after initial commit? just keep the originals?
21:39 bjoe2k4 joined #git
21:40 bjoe2k4 joined #git
21:41 saedelaere joined #git
21:42 maestrojed joined #git
21:43 fscala joined #git
21:44 d4rklit3 joined #git
21:44 eletuchy joined #git
21:44 dzho you mean, even if you change them?
21:45 dzho hanshenrik: I seem to recall that adding something to .gitignore after it has gone into the repo, that'll be how it works
21:45 bjoe2k4 joined #git
21:45 LionsMane MaddHatter: Our group, the consumer, will not be messing with the binaries.  But they need to be present for the code we are importing.
21:45 LionsMane MaddHatter: I guess I am not sure how to turn a git svn fetch'ed branch into a local branch that I can then push to the remote.
21:47 tobiasvl dzho: no, !ignore_tracked
21:47 gitinfo dzho: 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.
21:47 bjoe2k4 joined #git
21:47 tobiasvl and hanshenrik ^
21:48 hanshenrik oki, thanks ^^
21:48 bjoe2k4 joined #git
21:48 johnmilton joined #git
21:50 tgunr joined #git
21:50 bjoe2k4 joined #git
21:51 bjoe2k4 joined #git
21:51 nb_bez___ joined #git
21:52 bjoe2k4 joined #git
21:54 bjoe2k4 joined #git
21:54 bjoe2k4 joined #git
21:55 bjoe2k4 joined #git
21:55 peeps joined #git
21:57 foodSurprise joined #git
21:57 knutstrut joined #git
21:58 bjoe2k4 joined #git
21:59 cdown joined #git
21:59 bjoe2k4 joined #git
21:59 bjoe2k4 joined #git
21:59 cyan__ joined #git
22:00 bjoe2k4 joined #git
22:02 bjoe2k4 joined #git
22:03 bjoe2k4 joined #git
22:04 elisa87 joined #git
22:05 bjoe2k4 joined #git
22:05 mjuszczak joined #git
22:05 johnmilton joined #git
22:06 mjuszczak joined #git
22:06 ocbtec joined #git
22:07 ocbtec joined #git
22:07 bjoe2k4_ joined #git
22:08 bjoe2k4 joined #git
22:08 ahmed_elgabri joined #git
22:09 tgunr joined #git
22:10 diegoviola joined #git
22:10 apotry joined #git
22:11 LionsMane OK, I added an svn-remote to my git repository and then I did a git svn fetch.
22:11 LionsMane git pulled down all sorts of revisions.  I have no idea where the files are.
22:13 LionsMane The files in the .git folder are much larger, but in terms of what do I type to see the files? Do I need to to a git checkout?  I looked at the branches (all the ones I got from SVN are remotes/origin/trunk or remotes/origin/promo_test and are in red)
22:14 mwhooker joined #git
22:17 durham joined #git
22:19 chachasmooth joined #git
22:19 mikeric joined #git
22:19 faux___ joined #git
22:21 mjuszczak joined #git
22:21 paule32 joined #git
22:22 paule32 hello
22:22 paule32 how can i push/init/reinit project to github.com?
22:23 rainbyte joined #git
22:24 mjuszczak joined #git
22:28 diegoaguilar joined #git
22:31 diegoviola joined #git
22:31 diegoviola joined #git
22:40 ochorocho__ joined #git
22:48 cdown joined #git
22:49 jonmorehouse joined #git
22:52 madewokherd joined #git
22:53 italoacasas joined #git
22:53 sargas joined #git
22:54 gaen joined #git
22:59 MattMaker joined #git
23:00 nettoweb joined #git
23:01 HerrThees_ joined #git
23:02 MattMaker joined #git
23:02 cbreak paule32: just git push -f what ever you want
23:02 mablae joined #git
23:03 jaguarmagenta joined #git
23:03 Sasazuka joined #git
23:05 sargas joined #git
23:06 shinnya joined #git
23:06 mwhooker joined #git
23:07 mellernoia joined #git
23:09 NeverDie joined #git
23:09 ahmed_elgabri joined #git
23:10 holodoc joined #git
23:11 jonmorehouse joined #git
23:14 anu0 joined #git
23:16 ceaucari joined #git
23:16 otho joined #git
23:17 peepsalot i have a patch file in git repo.  if i edit a line in it and try to git diff, git seems confused that I changed thousands of lines.  what is the proper way to go about this
23:23 SimonNa joined #git
23:24 noyb joined #git
23:24 kadoban joined #git
23:25 rgrinberg joined #git
23:25 eletuchy joined #git
23:27 JanC_ joined #git
23:29 MattMaker joined #git
23:35 gaen joined #git
23:36 hahuang65 joined #git
23:36 devster31 I'm having trouble with a diff, I added a new remote with git remote add <name> <url> , then I did git fetch <name> and I tried git diff origin/master <name>/<branch> but it tells me "unknown revision or path not in the working tree" , I thought I did those steps correctly
23:37 MattMaker joined #git
23:37 fahadash joined #git
23:39 pwnz0r joined #git
23:39 frekoya_ joined #git
23:39 mablae_ joined #git
23:39 EvilPeng1 joined #git
23:40 arahael4 left #git
23:40 CheckDavid joined #git
23:40 cdg joined #git
23:40 ochorocho__ joined #git
23:41 bronson joined #git
23:42 hahuang61 joined #git
23:43 Arahael joined #git
23:46 devster31 i forgot the fetch -a flag
23:47 rkazak joined #git
23:49 mwhooker joined #git
23:56 Ali1 joined #git
23:56 dopesong joined #git
23:57 Shentino joined #git
23:58 mattfoo joined #git
23:58 nwkj86 joined #git
23:59 kus joined #git
23:59 fstd joined #git

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