Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-10-03

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

All times shown according to UTC.

Time Nick Message
00:00 pmartnhoj it would help if I used the same units...  so the 40mb repo is coming down at close to 100mbit which is the network speed
00:01 ahr3n joined #git
00:01 thiefg pmartnhoj: oh, then it could be a bottleneck on the disk IO, is 5gb repo mostly hard to compress data?
00:02 pmartnhoj yes it is raw data captured from some accelerometers so not very compressable
00:03 einnjo joined #git
00:04 pmartnhoj I can rsync the data at a high speed though,   is there a way to tell it to not try compress?
00:04 thiefg pmartnhoj: no, git requires the data to be compressed for it to transfer the data
00:05 mischat joined #git
00:06 thiefg pmartnhoj: unless you want to directly transfer the git objects yourself, which will proably break things if not done exactly right
00:06 jimi_ joined #git
00:07 bremner pmartnhoj: if you have ssh access to the host, you could try making a git bundle and rsyncing that. It might also try to compress, I don't know
00:07 GandalfTheWizar1 joined #git
00:08 rjack2 joined #git
00:08 pmartnhoj hmm it is a gitolite server,  I guess I will just have to wait it out
00:09 Kitty- joined #git
00:11 jaguarmagenta joined #git
00:14 ahr3n joined #git
00:17 jottr_ joined #git
00:17 Syn_Don joined #git
00:19 shgysk8zer0 joined #git
00:20 shgysk8zer0 joined #git
00:23 pmartnhoj it got there eventually, took 52 minutes
00:25 kerrick joined #git
00:28 ahr3n joined #git
00:29 Urbany joined #git
00:34 fuchstronaut joined #git
00:35 e14 joined #git
00:37 einnjo joined #git
00:42 mingrammer joined #git
00:43 ahr3n joined #git
00:46 nettoweb joined #git
00:49 leeN joined #git
00:54 hahuang62 joined #git
00:57 ahr3n joined #git
01:00 tzafrir joined #git
01:00 mehola joined #git
01:01 SwiftMatt joined #git
01:02 fuzzmz joined #git
01:04 DolpheenDream joined #git
01:05 ahr3n joined #git
01:05 mischat joined #git
01:07 WizJin joined #git
01:08 johnmilton joined #git
01:08 leeN joined #git
01:16 Vampire0_ joined #git
01:16 d^sh_ joined #git
01:18 leeN joined #git
01:18 jottr_ joined #git
01:20 AnarchyAo joined #git
01:22 ahr3n joined #git
01:23 sunri5e joined #git
01:27 mizu_no_oto joined #git
01:27 fahadash joined #git
01:31 ahr3n joined #git
01:32 fstd_ joined #git
01:33 thiefg left #git
01:33 Lyqyd joined #git
01:35 fuchstronaut joined #git
01:36 dreiss joined #git
01:36 pks joined #git
01:36 mingrammer joined #git
01:39 diogenese joined #git
01:46 shufan joined #git
01:48 ilbot3 joined #git
01:48 Topic for #git is now Welcome to #git, the place for git help and electronically-induced comas | Please take the git survey: https://jk.gs/gsurv | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.10.0 | Getting "cannot send to channel"? /msg gitinfo .voice | git-daemon: the evil rises
01:49 mars_ joined #git
01:50 ahr3n joined #git
01:50 harish joined #git
01:53 theskillwithin export GIT_DIR=$PWD/.git && cd dist && GIT_WORK_TREE=$PWD GIT_INDEX=/tmp/lol && git add . && git push origin -f $(git write-tree):refs/heads/gh-pages && unset GIT_DIR GIT_WORK_TREE GIT_INDEX
01:53 theskillwithin ! [remote rejected] 6e3ee2b7b98db2a97877ff50a19e804c5a7ded1e -> gh-pages (failed) error: failed to push some refs to 'git@github.com:theskillwithin/hrtcup.git'
01:53 theskillwithin so close to figuring this out
01:53 danslo_ joined #git
01:54 romerocesar joined #git
01:57 govg joined #git
01:59 olalonde joined #git
02:00 snthsnth joined #git
02:03 Rodya_ joined #git
02:04 ahr3n joined #git
02:06 mischat joined #git
02:14 mehola joined #git
02:17 Limes_ joined #git
02:19 ahr3n joined #git
02:20 jottr_ joined #git
02:21 t-mart joined #git
02:28 fatalhalt joined #git
02:28 Kitty- joined #git
02:28 hypercube32 joined #git
02:32 lagothri1 joined #git
02:33 ahr3n joined #git
02:34 johnmilton joined #git
02:35 shafun joined #git
02:36 fuchstronaut joined #git
02:37 fatalhalt joined #git
02:41 dodobrain joined #git
02:43 mizu_no_oto joined #git
02:46 hexagoxel joined #git
02:49 hypercube32 joined #git
02:54 tmg joined #git
02:54 bmahe joined #git
02:55 Dougie187 left #git
02:55 hahuang62 joined #git
02:59 romerocesar joined #git
03:01 arescorpio joined #git
03:02 chachasmooth joined #git
03:06 anonymuse joined #git
03:12 aidalgol joined #git
03:14 romerocesar joined #git
03:15 Vortex34 joined #git
03:20 YottaByte joined #git
03:20 jottr_ joined #git
03:21 YottaByte hi, when I do a "git pull" and I have local changes, it tells me to commit them first, so I do that, then I do the pull, and then I have to resolve the merge somehow
03:21 YottaByte once I use a merge tool what do I do then
03:21 YottaByte do I have to git commit again?
03:25 mingrammer joined #git
03:25 archaic joined #git
03:27 ANTI__mosad joined #git
03:28 ANTI__mosad left #git
03:29 ochorocho__ joined #git
03:33 snthsnth joined #git
03:35 star_prone joined #git
03:36 mizu_no_oto joined #git
03:36 zacts joined #git
03:37 fuchstronaut joined #git
03:39 Andrew_K joined #git
03:40 masuberu joined #git
03:40 danslo_ joined #git
03:42 moei joined #git
03:43 theskillwithin export GIT_DIR=$PWD/.git && cd dist && GIT_WORK_TREE=$PWD GIT_INDEX=/tmp/lol && git add . && git push origin -f $(git write-tree):refs/heads/gh-pages && unset GIT_DIR GIT_WORK_TREE GIT_INDEX
03:44 theskillwithin ! [remote rejected] 6e3ee2b7b98db2a97877ff50a19e804c5a7ded1e -> gh-pages (failed) error: failed to push some refs to 'git@github.com:theskillwithin/hrtcup.git'
03:44 xameada joined #git
03:52 Kitty- joined #git
03:58 a_thakur joined #git
04:00 wz520 joined #git
04:02 fedepad joined #git
04:03 xshuiniu joined #git
04:05 xameada joined #git
04:07 daey_ joined #git
04:07 afuentes joined #git
04:08 tax joined #git
04:10 mischat joined #git
04:10 xshuiniu joined #git
04:13 boombatower joined #git
04:16 dzho joined #git
04:19 mingrammer joined #git
04:21 romerocesar joined #git
04:21 jottr_ joined #git
04:22 Andrew_K|2 joined #git
04:25 a_thakur_ joined #git
04:26 a_thakur joined #git
04:27 Masber joined #git
04:31 mehola joined #git
04:37 fuchstronaut joined #git
04:40 ewet joined #git
04:41 danslo_ joined #git
04:46 arand joined #git
04:48 raijin joined #git
04:53 vuoto joined #git
04:53 a3pq51 joined #git
04:56 govg joined #git
04:57 Cabanossi joined #git
04:59 MAXEffort joined #git
05:00 dermoth joined #git
05:06 subhojit777 joined #git
05:10 tzafrir joined #git
05:11 mischat joined #git
05:13 star_prone joined #git
05:15 tristanp joined #git
05:16 Kitty- joined #git
05:16 tmg joined #git
05:17 SwiftMatt joined #git
05:19 timmu_ joined #git
05:19 romerocesar joined #git
05:22 jottr_ joined #git
05:24 einnjo joined #git
05:26 jozwior joined #git
05:27 Mowgli joined #git
05:28 snorge joined #git
05:30 morenoh149 joined #git
05:31 stuh84 joined #git
05:31 themayor joined #git
05:32 nilg joined #git
05:38 fuchstronaut joined #git
05:40 Kitty- joined #git
05:42 mingrammer joined #git
05:42 phanimahesh joined #git
05:44 Milos joined #git
05:45 l4v2 joined #git
05:46 lordjancso joined #git
05:49 jozwior joined #git
05:51 ThomasLocke joined #git
05:51 kerrick joined #git
05:51 ThomasLocke_ joined #git
05:59 alcohol joined #git
06:02 zeroed joined #git
06:02 zeroed joined #git
06:04 Raging_Hog joined #git
06:09 rscata joined #git
06:10 tzafrir joined #git
06:10 cmd-t joined #git
06:11 mischat joined #git
06:12 duderonomy joined #git
06:14 mingrammer joined #git
06:14 shinnya joined #git
06:15 Mowgli joined #git
06:15 tristanp joined #git
06:17 roelmonnens joined #git
06:17 JeroenT joined #git
06:21 theskillwithin export GIT_DIR=$PWD/.git && cd dist && GIT_WORK_TREE=$PWD GIT_INDEX=/tmp/lol && git add . && git push origin -f $(git write-tree):refs/heads/gh-pages && unset GIT_DIR GIT_WORK_TREE GIT_INDEX
06:21 theskillwithin ! [remote rejected] 6e3ee2b7b98db2a97877ff50a19e804c5a7ded1e -> gh-pages (failed) error: failed to push some refs to 'git@github.com:theskillwithin/hrtcup.git'
06:21 t4nk502 joined #git
06:23 jottr_ joined #git
06:24 cbreak theskillwithin: no need to spam
06:24 cbreak it tells you what's wrong: remote rejected
06:24 titankiller joined #git
06:24 theskillwithin yes, why is it rejecting
06:25 cbreak does it not say before?
06:25 cbreak look how the server is configured
06:25 cbreak it probably has a hook that rejects it
06:25 theskillwithin its github
06:26 cbreak theskillwithin: no output?
06:26 cbreak (before this rejection?)
06:26 theskillwithin cbreak https://i.imgur.com/MrAuhgI.png
06:26 cbreak bad ref
06:27 theskillwithin https://i.imgur.com/6arswwK.png
06:27 cbreak pastebin git ls-remote origin
06:28 cbreak what I'd do is delete gh-pages on the remote, then push fresh.
06:28 cbreak that would avoid some issues
06:28 theskillwithin cbreak http://kopy.io/szvxR
06:28 cbreak but maybe not yours
06:28 cbreak your gh-pages is on master? weird :)
06:29 cbreak theskillwithin: try with -f
06:29 theskillwithin this is making a gh-pages brach based on a folder dist, treating dist as root and in gh-pages and adding even tho its in gitignore
06:29 cbreak (before the origin of course)
06:30 cbreak theskillwithin: normally you'd have a separate branch for it
06:30 cbreak you should make one
06:30 cbreak git checkout --orphan gh-pages
06:30 cbreak or even a separate repository
06:31 cbreak you can also create a new view with the new-worktree script
06:31 theskillwithin ok but then I have to have it use just the dist folder  which is my build folder  as root
06:31 xaa joined #git
06:31 theskillwithin https://github.com/theskillwithin/hrtcup  as dist folder is not tracked in repo
06:32 ahr3n joined #git
06:32 cbreak (if you haven't seen what I wrote above, move the -f in front of the remote and ref parameter, into the option flag part of the command)
06:33 roelmonn_ joined #git
06:33 dan2k3k4 joined #git
06:34 thomas_25 joined #git
06:34 dax joined #git
06:35 fedepad joined #git
06:36 jozwior_ joined #git
06:37 Sceorem joined #git
06:38 clandestino joined #git
06:39 fuchstronaut joined #git
06:40 overlord_tm joined #git
06:41 barteks2x joined #git
06:41 QwertyDragon joined #git
06:44 JeroenT joined #git
06:44 danslo_ joined #git
06:47 frobnic joined #git
06:47 Andrew_K joined #git
06:49 star_prone joined #git
06:50 star_prone joined #git
06:52 ahr3n joined #git
06:55 tjone270 joined #git
06:56 Orbitrix joined #git
06:56 Orbitrix joined #git
06:57 Orbitrix joined #git
06:57 hahuang62 joined #git
06:59 cdown joined #git
07:00 Balliad joined #git
07:01 qt-x joined #git
07:05 tehtraveler2321 joined #git
07:05 barteks2x joined #git
07:05 danslo_ joined #git
07:05 theskillwithin cbreak like so:  export GIT_DIR=$PWD/.git && cd dist && GIT_WORK_TREE=$PWD GIT_INDEX=/tmp/lol && git add . && git push origin  $(git write-tree):refs/heads/gh-pages -f && unset GIT_DIR GIT_WORK_TREE GIT_INDEX
07:05 cbreak no
07:06 cbreak theskillwithin: why do you want to push to a branch -f?
07:06 theskillwithin ah
07:06 theskillwithin export GIT_DIR=$PWD/.git && cd dist && GIT_WORK_TREE=$PWD GIT_INDEX=/tmp/lol && git add . -f && git push origin $(git write-tree):refs/heads/gh-pages && unset GIT_DIR GIT_WORK_TREE GIT_INDEX
07:07 cbreak theskillwithin: why do you want to add a file -f?
07:08 cbreak theskillwithin: honestly, remove all that garbage from your commandline
07:08 _joes__ joined #git
07:08 cbreak just do a simple git push -f origin gh-pages
07:09 Seveas theskillwithin: that hack is fragile as hell. Either use a separate worktree for that branch, or use the github api to update the gh-pages branch with new content.
07:10 kugel joined #git
07:11 magellanicloud joined #git
07:11 theskillwithin been trying all sorts of methods haven't got one to work
07:11 theskillwithin https://i.imgur.com/VyZpmU0.png
07:12 SpecChum joined #git
07:12 theskillwithin making dist the root in just that branch even tho that folder is in gitignore
07:12 encod3 joined #git
07:12 Seveas yes, all those hacks are equally fragile.
07:13 theskillwithin http://kopy.io/ZQ5TD
07:14 Seveas still no good
07:14 Seveas use 'git worktree add' to create a new worktree for the gh-pages branch.
07:15 a3pq51_ joined #git
07:18 theskillwithin git worktree add -B gh-pages
07:19 Jameser joined #git
07:19 khmarbaise joined #git
07:22 Jameser joined #git
07:23 jottr_ joined #git
07:24 sl33k joined #git
07:28 JeroenT_ joined #git
07:28 mar77i uh, I added a filter thing to .git/info/attributes. how can I check out the smudged files now?
07:29 mar77i I tried git checkout -f .
07:29 Alienpruts joined #git
07:30 ochorocho__ joined #git
07:31 JeroenT__ joined #git
07:32 mar77i and git checkout HEAD -- .
07:32 MafiaInc joined #git
07:32 dunaeth joined #git
07:33 dreiss joined #git
07:33 PCatinean joined #git
07:35 mar77i and git heh
07:36 mar77i by heh I mean the filter doesn't seem to work on the whole tree... :)
07:36 Choups joined #git
07:37 skyres joined #git
07:37 Choups left #git
07:39 theskillwithin Seveas still lost haha maybe I need to do more googling
07:40 fuchstronaut joined #git
07:40 mischat joined #git
07:41 mar77i why would a git filter on *.py not apply to files in one subdirectory but to files in another?
07:45 Mindfart joined #git
07:46 laumann joined #git
07:49 mar77i after applying all the smudges manually, git status shows them all red
07:51 mar77i heh. adding them all made them disappear. I probably had a set of nothing staged
07:52 Gloomy joined #git
07:54 mar77i I don't get what it was that I got wrong.
07:55 ahmedelgabri joined #git
07:57 tilsa joined #git
07:57 kurkale6ka joined #git
07:58 harish joined #git
07:58 jknetl joined #git
07:59 tilsa left #git
08:00 joki joined #git
08:04 stupidystupid joined #git
08:08 govg joined #git
08:13 OnOut2k16 left #git
08:15 Snugglebash joined #git
08:16 tristanp joined #git
08:16 BadSnappa joined #git
08:21 clmsy joined #git
08:22 mquin joined #git
08:23 MillerBoss joined #git
08:23 tHEMtZ joined #git
08:24 jottr_ joined #git
08:26 noidea joined #git
08:31 encod3 joined #git
08:32 Ezzy joined #git
08:32 dede_ joined #git
08:33 chll_ joined #git
08:34 mingrammer joined #git
08:39 encod3 joined #git
08:41 fuchstronaut joined #git
08:42 JeroenT joined #git
08:42 mischat joined #git
08:44 titankiller joined #git
08:49 jknetl joined #git
08:50 AlHafoudh joined #git
08:51 ochorocho__ joined #git
08:53 eshlox joined #git
08:55 a_thakur joined #git
08:55 a_thakur joined #git
08:57 hahuang62 joined #git
08:57 pl joined #git
08:58 jez9999 joined #git
08:58 jez9999 hi
08:58 jez9999 does the "fetchRecurseSubmodules" option cause git to behave like you typed "git clone --recursive" for that submodule?
09:02 encod3_ joined #git
09:03 encod3__ joined #git
09:04 harish joined #git
09:04 jknetl joined #git
09:06 byobubu joined #git
09:07 danslo_ joined #git
09:07 des_consolado joined #git
09:08 encod3 joined #git
09:08 jaequery joined #git
09:09 M-ou-se joined #git
09:10 achadwick joined #git
09:11 encod3_ joined #git
09:11 byobubu Hi. I have a (for sure stupid) problem with git rebase. I have master branch that I use to follow upstream, and a "personal" branch where I just added some files. I update the master branch (git pull upstream master), then I tried to rebase my personal branch (git rebase master mybranch). The merge fails, and in "git status" I get a "both added" merge error, but this is for a file that does not exist in the master branch!
09:11 byobubu Indeed it does not appear in git ls-files, while it does appear if I checkout mybranch.
09:13 byobubu What could I try?
09:18 mar77i byobubu: looks like upstream added a file with the same name as you?
09:19 mar77i byobubu: can you verify whether that is the case?
09:20 byobubu mar77i: but I did check... there's no such file
09:20 brokensyntax joined #git
09:21 osse check how?
09:21 tuor joined #git
09:22 osse do  git log upstream/master -- file
09:23 xaa joined #git
09:23 encod3 joined #git
09:25 Mowgli joined #git
09:25 jottr_ joined #git
09:27 govg joined #git
09:28 timini joined #git
09:28 pl osse, returns nothing
09:29 timini Hi can you offer me some advice?
09:29 timini I have a branch which has got outof sync with dev
09:30 byobubu http://pastebin.com/26dBpj70
09:30 corpsicle joined #git
09:31 timini when i try to rebase dev onto my branch I get LOAds of merge conflicts
09:31 timini Is there a way to rebase only the stuff without conflicts?
09:32 corpsicle so. if i have 20 git repos, and i want to merge _all_ of them at once, from one branch to another, within each repo. is there a convenient way to create pull requests on all those merges?
09:32 a_thakur joined #git
09:33 MAXEffort left #git
09:36 JeroenT_ joined #git
09:38 encod3 joined #git
09:40 joki joined #git
09:41 fuchstronaut joined #git
09:43 leangjia joined #git
09:43 jbitdrop joined #git
09:46 encod3 joined #git
09:49 Snugglebash joined #git
09:51 Darcidride joined #git
09:56 xshuiniu joined #git
09:59 tuor joined #git
10:01 antlarr joined #git
10:02 mikkeloscar joined #git
10:04 dunaeth joined #git
10:04 Vampire0 corpsicle, if you have a commandline tool that can create PRs like git-spindle you can simply loop through your repos with a bash loop and create the pull requests
10:04 a_thakur joined #git
10:05 corpsicle oh cool, thanks
10:05 mikkeloscar What is a lightweight way (in terms of bandwidth) to get tag info (author,date) of a remote repo? I can use git ls-remote --tags to get the tag names, but I also want to get date and author.
10:06 cbreak joined #git
10:06 mingrammer joined #git
10:07 ocbtec joined #git
10:08 hahuang62 joined #git
10:12 eka joined #git
10:17 Orbitrix joined #git
10:17 osse byobubu: can you do the thing i said?
10:18 Orbitrix joined #git
10:18 Orbitrix joined #git
10:18 Jameser joined #git
10:19 M-ou-se joined #git
10:21 canton7 joined #git
10:21 ams__ joined #git
10:23 encod3 joined #git
10:23 stamina joined #git
10:26 jottr_ joined #git
10:27 cederfjard joined #git
10:27 cbreak joined #git
10:29 CussBot joined #git
10:33 ferr joined #git
10:34 mischat joined #git
10:34 maret joined #git
10:34 thierryp joined #git
10:35 ferr left #git
10:37 JeroenT joined #git
10:38 Akimb joined #git
10:38 danslo_ joined #git
10:39 nettoweb joined #git
10:42 fuchstronaut joined #git
10:43 Mowgli joined #git
10:44 leangjia left #git
10:46 musicmatze joined #git
10:47 musicmatze Hi... Can someone tell me how to get all lines which are left from a version. So basically all lines that were in revision XY and are still in HEAD?
10:47 irco joined #git
10:47 mago_ joined #git
10:47 Murii joined #git
10:48 osse musicmatze: git blame file | grep '^XY'
10:48 alex1a joined #git
10:48 musicmatze osse: this does not work out if I `git mv`ed a lot around... does it?
10:48 ^7heo that's gonna give a pretty hard to read output tho.
10:48 ^7heo and no it does not follow renames.
10:48 osse musicmatze: yes
10:49 ^7heo osse: it follows renames!?
10:49 sentriz joined #git
10:49 osse hmm, i think so
10:49 osse man git-blame
10:49 gitinfo the git-blame manpage is available at http://jk.gs/git-blame.html
10:49 osse "The origin of lines is automatically followed across whole-file renames (currently there is no option to turn the rename-following off). To follow lines moved from one file to another, or to follow lines that were copied and pasted from another file, etc., see the -C and -M options."
10:50 ^7heo osse: I think it only follows renames if the rename is atomic enough that it is not considered a new file.
10:50 ^7heo (i.e. no modifications on the content)
10:50 osse yeah i'm not sure what they mean there
10:51 osse log --follow, diff etc. use 50%
10:51 osse if they literally mean no modification at all then... :/
10:51 budre joined #git
10:51 ^7heo yeah
10:51 govg joined #git
10:53 osse I just tested. It seems to work on partial too
10:54 osse in a repo I renamed a file AND change one line. git blame correctly attributed t the other lines to the previous commit
10:54 Camusensei joined #git
10:54 ^7heo osse: nice :)
10:56 Camusensei Hello guys. I have a strang problem with a branch that is not appearing in my log. I'm using "git log --graph --abbrev-commit --decorate --oneline" and there is only one branch, already tracked that I can't see
10:56 osfe joined #git
10:56 Camusensei I fetched from the remote, I checked with ls-remote (it exists), but the git log command won't show it. what else should I try?
10:56 osse that will only show branches reachable from HEAD
10:57 Camusensei osse: forgot --all
10:57 osse in that case it doesn't exist locally
10:57 osfe Hi there, I'm new to gitlab and I can't get it to work... I set up a remote repo on framagit (?provider?) I followed the instruction to create ssh key and created a new key on framagit. I followed the instructions given for an existing folder. it work except for the push.  I'm asked for my password and then I get a permission denied.  can you please help me here ?
10:58 osse Camusensei: what is its full name according to ls-remote? and what does git config remote.origin.fetch say?
10:58 Camusensei osse: I can even check it out! but it doesn't say detached head... I think I may have created it in a strange way
10:58 Camusensei I did this:            git checkout ovh/Config_correction -b Config_correction
10:59 Camusensei ls-remote says 6f16f676f45e5b698abde00dfea277a52c00ef00        refs/heads/Config_correction
10:59 thiago joined #git
10:59 Camusensei +refs/pull-requests/*/from:refs/remotes/origin/pull-requests/*
10:59 Camusensei +refs/heads/*:refs/remotes/ovh/*       might be more useful here
11:00 Camusensei I guess I'll just drop the local branch and get it from the remote again
11:01 Camusensei yep, now it's back. Thank you for your help osse :)
11:02 manuel joined #git
11:03 bremner osfe: can you push to other remotes? i.e. do you have a problem specifically with gitlab/framagit or with git in general?
11:05 osfe bremner: I guess it's a framagit issue
11:05 greendo joined #git
11:06 osfe I've never been able to find a proper tutorial to setup a git framework...
11:06 phanimahesh joined #git
11:07 osfe framework might not be the right word
11:08 Camusensei left #git
11:08 a_thakur joined #git
11:10 rnsanchez joined #git
11:14 jez9999 left #git
11:14 drbean joined #git
11:15 geoid joined #git
11:16 rjack2_ joined #git
11:17 sdothum joined #git
11:17 xbob joined #git
11:18 Snugglebash joined #git
11:19 WayToDoor joined #git
11:20 osfe ok, dit again the entire procedure, erasing everithing before (git repo + key in sshdir). It looks like it works now
11:21 Limes_ joined #git
11:24 DolpheenDream joined #git
11:27 jottr_ joined #git
11:28 ojdo joined #git
11:34 mizu_no_oto joined #git
11:35 Emma_Gination joined #git
11:35 thomas_25 joined #git
11:36 nuck joined #git
11:36 Emma_Gination Hi guys, is there any way of knowing if i have write permissions to a remote git?
11:36 Emma_Gination (without changing anything)
11:36 selckin no
11:37 Emma_Gination selckin: really? Any idea why this is the case?
11:37 greendo joined #git
11:38 selckin because nobody has a need for it
11:38 jottr_ joined #git
11:38 mingrammer joined #git
11:38 osse Git leaves to to the OS and other facilities to handle that
11:38 zifnab joined #git
11:38 osse but if you're not sure whether you have permission then you most likely don't
11:39 dcpc007 joined #git
11:39 Emma_Gination so to find out if i have write permissions i have to clone, make a change, add, commit and then push - just to see if i have the rights to do so?
11:40 paws_ joined #git
11:40 osse yep
11:40 selckin you don't care if you have the right todo so untill you want to push something
11:40 Emma_Gination selckin: well i do :)
11:41 osse it's probably easier to ask someone
11:42 Emma_Gination The someone whom i should ask isn't sure and he has asked me to check myself - if i don't have the rights then we will have to escalate the issue
11:43 Emma_Gination but for the sake of not bothering a lot of people to find out whether this is resolved i'd like to find out myself
11:43 fuchstronaut joined #git
11:43 osse what are the actual details involved here?
11:43 osse where is the repo hosted? and how? what URL do you use to clone it?
11:44 selckin so work till you have something to push, try and complain if no access
11:45 eduardas_m joined #git
11:46 fees joined #git
11:47 Emma_Gination osse: it is hosted on gitlab and it is a private repo owned by my employers.
11:47 Emma_Gination selckin: guess i'll have to :P
11:47 Emma_Gination thank you both answering <3
11:47 nettoweb joined #git
11:48 osse Emma_Gination: in that case the gitlab web UI can be used to check and configure access
11:49 Emma_Gination osse: really?! thank god!
11:49 Emma_Gination (and thank you)
11:49 Emma_Gination do you know how?
11:49 osse no, its UI confuses the hell out of me
11:49 osse but they have an irc channel and docs online
11:50 osse also, regardless, cloning the repo right now is not useless. you can start working and wait for push access.
11:50 osse *you* don't have to do anything to gain access. the URL is the same, the ssh keys are the same, etc.
11:51 rjack2 joined #git
11:59 jottr_ joined #git
12:02 crameth joined #git
12:04 _dunaeth joined #git
12:04 Camusensei joined #git
12:04 johnmilton joined #git
12:05 Camusensei okay this is actually not good. I keep gtting  * [new branch]      Config_correction -> ovh/Config_correction      when I fetch
12:05 govg joined #git
12:05 Camusensei but I don't see it
12:05 anonymuse joined #git
12:05 johnmilton joined #git
12:06 Camusensei git fetch ovh      * [new branch]      Config_correction -> ovh/Config_correction         git fetch ovh  * [new branch]      config_correction -> ovh/config_correction
12:06 Camusensei notice how the capital letter appears and disapears
12:09 hahuang62 joined #git
12:09 mischat joined #git
12:11 jknetl joined #git
12:12 leeN joined #git
12:13 Camusensei looks like there are two remotes (one with capital and one without) and my windows can't handle case changes :D problem solved
12:13 eihabarabia joined #git
12:14 WayToDoor joined #git
12:15 Emma_Gination Found it! osse and selckin - if you are, yourself interested, you can view your project access levels here https://<gitlab.host.com>/<group>/<project>/project_members
12:15 selckin !github
12:15 gitinfo Note that git != github. Feel free to ask us about Github-specific features (Forks, Pull Requests, Wikis, etc), but there are no guarantees. There is a #github channel, which might help too (again, no guarantees)
12:16 Emma_Gination (or actually, didn't find it - i asked on the gitlab irc :P )
12:16 Emma_Gination yeah, probably not on github - dunno
12:16 selckin also not git
12:16 jaguarmagenta joined #git
12:16 des_consolado Hey if I do `git remote show origin` one of my local branches configured for 'git pull' is aligned with 2 remote branches, how do I remove one of them?
12:17 Emma_Gination selckin: right
12:18 des_consolado !gitlab
12:18 gitinfo Gitlab is Git hosting similar to Github, but for your own server or on gitlab.com, giving you unlimited free private repositories: http://gitlab.com/
12:19 tristanp joined #git
12:25 synthroid joined #git
12:26 yottabyte joined #git
12:27 yottabyte hi, when I do "git pull" and have local changes, it'll tell me to commit them before pull (I guess this happens if some merge conflict occurs?), so I git add and git commit, then I can git pull? do I have to git add/commit after resolving the merge conflict? just trying to figure out how to do it
12:28 nfk joined #git
12:29 altendky yottabyte: if you pull and then get notified of a merge conflict your local state will be in the middle of a merge.  to continue you will have to resolve the conflicts, add the conflicted files, and then commit.
12:30 altendky are git submodules commonly used for separate resource repositories?  i am looking to make sure they are an appropriate recommendation for this pysunspec project and their models repository.  https://github.com/sunspec/pysunspec/issues/17
12:30 Snugglebash joined #git
12:30 des_consolado pull is a merge from remote to local. push is a merge from local to remote.
12:31 Emma_Gination yottabyte: you don't necessarily have a merge conflict - you allways have to commit your local changes before you can pull
12:31 Emma_Gination but as altendky says, you will be notified if you have a merge conflict
12:31 des_consolado is that true that you _always_ have to commit local changes before pulling?
12:31 WhereIsMySpoon joined #git
12:32 WhereIsMySpoon How do I create a diff that I can manually apply later between my current commit and HEAD (I'm currently a couple of commits ahead of HEAD)?
12:32 jimi_ joined #git
12:32 des_consolado yeah that's not true
12:33 des_consolado also git pull seems to bring up to date all other branches on the same repo for me
12:33 Emma_Gination des_consolado: well yes, I you have local changes you will not be allowed to pull before the changes are stashed, committed or reset
12:33 canton7 des_consolado, it updates the remote-tracking branches (origin/XXX), but not your local branches
12:33 buffal0 joined #git
12:33 des_consolado oh right
12:34 des_consolado remote-tracking branches = local references/copies of remote branches?
12:34 canton7 updating a local branch potentially requires a merge, which might potentially conflict, requiring user intervention - so they're never done automatically
12:34 canton7 !remote_tracking_branch
12:34 gitinfo Remote-tracking branches (branches which start with e.g. 'origin/', listed by 'git branch -r') are read-only mirrors of the corresponding branches in another repository. They're updated by 'git fetch'. You can't edit them directly (trying to check them out results in a !detached HEAD), but you can create a new local branch based on a remote-tracking branch using e.g. 'git checkout -b <branch> <remote>/<branch>'
12:34 des_consolado ah, ty
12:35 wrouesnel1 joined #git
12:35 WhereIsMySpoon How do I create a diff that I can manually apply later between my current commit and HEAD (I'm currently a couple of commits ahead of HEAD)?
12:36 WhereIsMySpoon Or patch file or whatever its called
12:36 CussBot joined #git
12:37 des_consolado commit and then cherry-pick I think?
12:37 WhereIsMySpoon do I need a separate patch file for each commit?
12:38 osse WhereIsMySpoon: you are always on HEAD by definition
12:38 WhereIsMySpoon er...I meant the remote head, sorry
12:38 osse WhereIsMySpoon: git diff X Y generates a patch you can apply on X to acheive Y
12:38 osse so git diff origin/master master  for example
12:39 up_the_irons joined #git
12:40 danslo_ joined #git
12:40 jimmyeatworld joined #git
12:43 fmcgeough joined #git
12:43 WayToDoor joined #git
12:43 WhereIsMySpoon ok thanks osse
12:44 fuchstronaut joined #git
12:44 rgrinberg joined #git
12:45 jeffreylevesque joined #git
12:45 cyan__ joined #git
12:46 sangy joined #git
12:48 rgrinberg joined #git
12:50 Raging_Hog joined #git
12:53 roxlu_ joined #git
12:54 freimatz joined #git
12:55 govg joined #git
12:57 josuebrunel joined #git
12:57 patarr joined #git
12:58 WayToDoor joined #git
12:59 languitar joined #git
12:59 WhereIsMySpoon left #git
13:00 eggoez joined #git
13:03 m0viefreak joined #git
13:03 jaguarmagenta joined #git
13:04 e14 joined #git
13:11 PCatinean joined #git
13:12 str joined #git
13:13 PCatinean joined #git
13:14 ExoUNX joined #git
13:17 ahmedelgabri joined #git
13:18 inflames joined #git
13:19 somaReverse joined #git
13:20 einnjo joined #git
13:20 titankiller joined #git
13:26 f3r70rr35f joined #git
13:28 GodGinrai joined #git
13:29 Mowgli joined #git
13:30 geoid joined #git
13:31 GodGinrai joined #git
13:31 dunk If you have a dev branch that has not been merged into master yet, but you want to continue work in another branch (or other branches) based on that work, then what is the canonical way to get those changes into the new dev branch such that it won't cause problems later when merging that branch into master?
13:31 grawity just branch off dev, that's all
13:32 dunk grawity: what if the second dev branch already existed?
13:32 dunk and had some changes in it
13:32 grawity pick better branch names than 'dev' 'dev2' 'dev3'
13:32 dunk grawity: obviously those are just for explanation
13:33 grawity or, if you want to renew that other branch, maybe rebase it on top of current dev?
13:33 mischat joined #git
13:34 dunk so the following is ok then?: make dev-1, make dev-2, n commits into dev-1, n commits into dev-2, create PR for dev-1, rebase dev-2 on dev-1, [dev-1 is merged into master], create PR for dev-2, [dev-2 is merged into master]
13:36 mischat_ joined #git
13:38 bernardio joined #git
13:39 dunk Before creating the PR for dev-2, I would rebase it on master, of course
13:40 jottr_ joined #git
13:40 e14 joined #git
13:42 jaguarmagenta joined #git
13:43 danslo_ joined #git
13:43 faux___ joined #git
13:44 fuchstronaut joined #git
13:45 timini joined #git
13:46 rnsanchez_ joined #git
13:47 Darcidride joined #git
13:48 tuor Hi, I want to see the difference between my staged changes and HEAD. How can I?
13:49 osse git diff --staged
13:49 jaguarmagenta joined #git
13:51 tuor osse, thx!
13:51 Snugglebash joined #git
13:52 blackwind_123 joined #git
13:52 sl33k_ joined #git
13:53 Murii joined #git
13:53 AaronMT joined #git
13:53 tristanp joined #git
13:54 Murii joined #git
13:54 khmarbaise joined #git
13:56 King_Hual joined #git
13:56 King_Hual joined #git
13:56 Murii joined #git
13:56 Masber joined #git
13:57 Murii joined #git
13:58 Murii joined #git
13:59 faux___ joined #git
13:59 Guest39784 joined #git
14:00 Murii joined #git
14:00 maret joined #git
14:00 phaleth joined #git
14:02 JeroenT joined #git
14:02 faux___ joined #git
14:02 lb1a !ping
14:02 gitinfo I only respond to !reply_time
14:03 faux___ joined #git
14:04 osse Obviously it also responds to !ping
14:04 gitinfo I only respond to !reply_time
14:04 lb1a joined #git
14:05 lb1a joined #git
14:05 mischat joined #git
14:05 WayToDoor joined #git
14:07 hgutmann joined #git
14:07 hgutmann left #git
14:10 hahuang62 joined #git
14:13 dave0x6d joined #git
14:16 al-damiri joined #git
14:18 ResidentBiscuit joined #git
14:18 sangy joined #git
14:22 ForeignBiscuit joined #git
14:25 mischat joined #git
14:25 troulouliou_div2 joined #git
14:26 Snugglebash joined #git
14:27 leeN joined #git
14:28 raijin joined #git
14:28 mischat_ joined #git
14:29 ShekharReddy joined #git
14:29 mischat__ joined #git
14:30 Seveas it's a trap
14:30 Seveas !reply_time
14:30 gitinfo I only respond to !ECHO-REQUEST
14:30 Seveas !ECHO-REQUEST
14:30 gitinfo I only respond to !ping
14:30 Seveas -EINFINITELOOP
14:30 ForeignBiscuit !help
14:30 gitinfo What do you need help with? Something with git? Tell us what's going on, and please be specific. For information on how to abuse me (I'm a bot!), see http://jk.gs/git/bot . Please be gentle.
14:30 * gitinfo watches gleefully as Seveas gets stuck pinging forever
14:31 Seveas jast: remove your hand from gitinfo's butt
14:32 Seveas or was jast crazy enough to let osse do that, hmm...
14:32 gitinfo What do you mean? I'm my own free automaton.
14:32 crose joined #git
14:35 thomas_2_ joined #git
14:37 l4v2 joined #git
14:40 jottr_ joined #git
14:43 danslo_ joined #git
14:44 subhojit777 joined #git
14:45 ash_workz joined #git
14:45 fuchstronaut joined #git
14:47 isysd joined #git
14:49 PCatinean joined #git
14:50 troulouliou_div2 joined #git
14:53 jez9999 joined #git
14:53 jez9999 Does anyone know whether there is a way to automatically update subrepos when you git pull on a superrepo?
14:53 est31 joined #git
14:54 EdwardIII joined #git
14:54 est31 so I am doing git submodule update --init trying to sync with a public git project
14:54 est31 but somehow it prompts me for Username for 'https://github.com':
14:54 EdwardIII hey i'm a bit confused. if you have master -> feature, then you take feature2 off of feature, so you branch looks like master -> feature -> feature2 and they all have dev. when you merge feature2 back into master you get the commits from both feature + feature 2, right?
14:54 est31 is it somehow possible that upstream deleted a submodule and I have to do it too?
14:54 canton7 est31, looks like it's trying to fetch a submodule from github?
14:54 est31 whats the command for that
14:55 jez9999 EdwardIII: yes
14:55 est31 and how do I find out which submodule its wondering about
14:55 jez9999 EdwardIII: except the commits made to feature after feature2 was branched
14:55 EdwardIII jez9999: right of course
14:55 Zamyatin joined #git
14:55 Dougie187 joined #git
14:55 EdwardIII so then when i do git master...feature, as long as there are no new commits to feature, i should see no changes, right?
14:56 jez9999 i think it's 2 dots
14:57 PCatinean joined #git
14:57 est31 canton7, any ideas?
14:58 EdwardIII jez9999: hrm but that just shows all differences right?
14:58 EdwardIII the problem is, git is acting like `feature` branch isn't merged in
14:58 EdwardIII and i'm a bit scared to delete it
14:58 jez9999 between the 2 tags specified
14:58 EdwardIII but when i manually flick throuhg it looks kinda like all the relevant stuff is already there
14:59 EdwardIII however when i try and merge or rebase i enter conflict city
14:59 jez9999 how did you merge feature2 in?
14:59 btree joined #git
14:59 EdwardIII i rebased master onto feature2, then merge feature2 into master
14:59 jez9999 why not rebase feature2 onto master
14:59 EdwardIII if i re-write master's history i'll enter into a world of ballache no? other people have been working on master
14:59 EdwardIII (thus the rebase)
15:00 jez9999 yeah that's why you rebase feature2 onto master
15:00 EdwardIII you mean git checkout master && git rebase feature2?
15:00 EdwardIII or the other way around?
15:00 jez9999 other way around
15:01 jez9999 "git rebase master" is saying "rebase this onto the master"
15:01 est31 meh, gave up
15:01 est31 rm -rf ed the whole directory and doing a clean re-clone
15:01 EdwardIII ah yes sorry i often mess those up
15:01 EdwardIII but i think i understand it
15:01 est31 submodules are a mess
15:01 est31 left #git
15:02 EdwardIII but that's what i did. git checkout feature2; git rebase master; # then effectively: # git checkout master; git merge feature # except all on gitlab
15:02 jez9999 if you didnt make changes to feature after the feature2 branch, you dont need to merge feature
15:03 EdwardIII that's what i thought, but i'm a bit scared as git doesn't seem to know about that
15:03 EdwardIII e.g. git branch -d feature says it's not fully merged
15:03 jez9999 look at gitk
15:03 jez9999 see whether feature is on a chain that gets merged back into master.  if it is, it can be safely deleted.  it's just a tag.
15:04 eszett_chatzilla joined #git
15:04 mingrammer joined #git
15:05 EdwardIII hrm so when i do gitk on master i don't see feature mentioned anywhere, only feature3
15:05 EdwardIII *feature2
15:06 jez9999 try gitk feature1
15:07 EdwardIII hrm that just shows feature1 ending at the top of the tree
15:07 Megaf joined #git
15:07 jez9999 you should be able to tell whether there are additional changes to feature1
15:07 Megaf joined #git
15:07 jez9999 but i dont know the details of your repo
15:08 Snugglebash joined #git
15:08 EdwardIII i know for a fact there are no additional changes to feature1
15:08 e14 joined #git
15:08 EdwardIII but, i think i can see duplicate commits
15:08 derpsauce__ joined #git
15:08 EdwardIII lemme take a screenshot
15:09 EdwardIII https://cloud.githubusercontent.com/assets/412107/19042396/c395c398-8983-11e6-8ac3-8b95bf19b316.png
15:09 EdwardIII so "Add infrastructure for next cat in rails" appears twice?
15:10 jez9999 wow that looks ugly compared to the gui gitk
15:10 EdwardIII similar info though right?
15:10 beecushman joined #git
15:11 jez9999 they look like they've both been merged in
15:11 EdwardIII https://cloud.githubusercontent.com/assets/412107/19042448/061eca5c-8984-11e6-9684-0bb2ec2d0c3e.png <- gitk
15:11 EdwardIII oddly exploded & pixelated sorry heh
15:12 tyreld_ joined #git
15:12 EdwardIII next_cat = feature, custom_cat_widgets = feature2
15:13 EdwardIII but if i'm reading this i can see next_cat in git lol, but not in gitk?
15:15 _Vi joined #git
15:17 nsanden joined #git
15:17 PCatinean joined #git
15:18 nsanden after i git pull to my live site, i have to sometimes manually move files from /home/domain1/myapp/web/images/ to /home/domain1/public_html/images/ - what is a good way to do this automatically? git hook? bash script?
15:18 grawity !deploy
15:18 gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
15:18 grawity write a git hook, and make it do whatever you want
15:19 grawity though usually...
15:19 grawity actually, where do you pull from, and where to?
15:20 nsanden i pull from github repo into my live site
15:20 hashpuppy joined #git
15:20 grawity so what's ~/myapp/web and what's ~/public_html
15:22 nsanden /myapp is a clone of the repo and /myapp/web is what files should be web accessible. public_html is where those files need to be copied to actually make them web accessible
15:24 grawity so it's not specific to just images, then?
15:24 zivester joined #git
15:24 btree left #git
15:24 nsanden right, theres css, js, images
15:25 perlpilot nsanden: you might want look at an actual deployment tool like capistrano or ansible or something  (just an idea, not a strong suggestion)
15:25 le_melomane joined #git
15:25 EvilPeng1 joined #git
15:26 grawity nsanden: then a) git hook as a deploy script,
15:27 grawity nsanden: or b) reconfigure your web server to directly serve from ~/myapp/web
15:27 GodGinrai I would advise against b)
15:27 eivarv joined #git
15:27 GodGinrai the git hook is the better option, or as perlpilot said, a deployment tool.
15:27 navidr joined #git
15:27 GodGinrai hell, you could even write your own deployment scripts using rsync
15:29 jez9999 if you're using node.js, use npm
15:29 subhojit_ joined #git
15:30 nsanden anyone have a good tutorial for git hooks. i havent found much
15:30 nsanden if i understand correctly, you create them in .git/hooks? so they are repo dependant, not global?
15:33 LionsMane joined #git
15:33 osse yes
15:33 osse man githooks
15:33 gitinfo the githooks manpage is available at http://jk.gs/githooks.html
15:33 osse it's all there
15:33 osse !hooks
15:33 gitinfo [!hook_pitfalls] Guidelines for writing hooks: 1. Consume all input (cat >/dev/null if you don't want it). 2. If you use any 'cd', also 'unset GIT_DIR'. 3. Don't git-pull in a hook (or any other script).
15:34 osse nsanden: pong
15:35 nsanden post-merge is what id want then i think for git pull
15:35 acetakwas joined #git
15:37 nsanden one more concern - i would only want the hook to execute on the live site, not on my local environment when im developing
15:37 tzafrir joined #git
15:38 synthroid joined #git
15:39 osse that's not gonna happen.
15:39 osse rarely if ever is a merge done on a remote repo
15:39 osse !deploy
15:39 gitinfo Git is not a deployment tool, but you can build one around it (in simple environments) or use it as an object store(for complex ones). Here are some options/ideas to get you started: http://gitolite.com/deploy.html
15:40 SteffanW joined #git
15:41 canton7 remember that merges can and will fail, and need human interaction to fix
15:41 canton7 unless you like your website delivering ">>>>>>>>>>>>>>>>>>" to the user
15:41 kpease joined #git
15:41 jottr_ joined #git
15:42 osse it's not *that* bad. it's more like ">>>>>>>"
15:42 grawity https://twitter.com/badedgecases/status/781928309059969029
15:42 grawity I like the Swift reply
15:42 flavius joined #git
15:43 le_melomane joined #git
15:44 snowkidind joined #git
15:45 GodGinrai nsanden: why use post-merge instead of using server-side hooks?
15:45 jaafar joined #git
15:46 marvi joined #git
15:46 dan2k3k4 joined #git
15:46 nsanden GodGinrai: no idea - would one of those hooks server better?
15:46 danslo_ joined #git
15:47 barteks2x joined #git
15:48 osse see the deploy link
15:48 osse or: what you you want to achieve?
15:49 madewokherd joined #git
15:49 tang^ joined #git
15:50 robotroll joined #git
15:50 GodGinrai nsanden: post-receive
15:50 nettoweb joined #git
15:51 Endarked joined #git
15:51 nsanden GodGinrai: thought thats what happens when something like Github receives a push from my local
15:52 nsanden my setup is local clone -> push to github   and    live server clone -> pull from github
15:52 nsanden osse: looking again at it
15:52 nsanden osse: seems what i want is not recommended - Also note that none of them involve git pull, which a naive user might think is worth trying, but in fact has absolutely no place in an automated setup or script of any kind. (If you don't know why, the short answer is "pull = fetch + merge" and merges cannot be guaranteed to complete without manual fixups).
15:53 osse nsanden: better approach on the server clone:  fetch && reset --hard
15:53 osse you can use github's hook thingy to notify the server somehow (not sure exactly how)
15:54 nsanden oh fetch && reset so no chance of merge conflict?
15:54 osse no chance at all
15:54 tgunr how can search a repo to find when a file got deleted? something taht should be in the repo since 2010 has disappeared from my master and I can't figure out when or why it got deleted
15:54 osse you just say "fuckit, use what was fetched"
15:55 osse tgunr: git log -- path/to/file
15:55 nsanden i like that
15:55 ws2k3__ joined #git
15:55 GodGinrai nsanden: sure.  Set up live as another remote and push to it.
15:55 GodGinrai nsanden: rather than sshing into live and pulling
15:56 faux___ joined #git
15:56 nsanden GodGinrai: i have other devs working on it, id rather manually pull
15:56 tgunr How can that work if the file is missing?
15:56 tgunr [mp]/Sources/Factor (fc_merge): git log basis/db/tuples/tuples.factor
15:56 tgunr fatal: ambiguous argument 'basis/db/tuples/tuples.factor': unknown revision or path not in the working tree.
15:57 osse tgunr: because git looks at the files in the commits, not the files on disk
15:57 GodGinrai nsanden: how would manually pulling make anything safer for those other devs?
15:57 osse use the -- (and make sure to get the exact path right)
15:57 tgunr well, its complaining
15:57 osse or, do -- '**/tuples.factor'
15:57 tgunr the path is correct
15:57 tgunr trying
15:57 nsanden GodGinrai: then they arent pushing at 3am when im asleep and breaking the live site
15:57 oskarkv joined #git
15:57 nsanden only time live site updates is when i do so manually
15:58 prawnzy joined #git
15:58 GodGinrai nsanden: They can't pust to live if you don't let them.
15:58 GodGinrai *push
15:58 tgunr that worked! thanks
15:58 osse nsanden: in any case, the main point is that no actual development takes place on the live server right? for all you care there doesn't actually need to be a repo there as long as the files are there, right?
15:58 osse tgunr: which of them?
15:59 nsanden right osse
15:59 tgunr **/tuples.factor
15:59 dreiss joined #git
15:59 osse tgunr: so the path was wrong then :P
15:59 mingrammer joined #git
15:59 nsanden GodGinrai: i guess i dont understand what your suggesting then. i thought you were saying skip github and push straight to live
15:59 osse nsanden: there are many ways to gently pet that cat
15:59 tgunr no, its always been basis/db/tuples.tuples.factor
16:00 osse in that case  git log -- basis/db/tuples.tuples.factor  would work too, unless you're in a subdir
16:00 nsanden so current thinking is, on live site, ill git fetch and merge and then i guess write a bash script to sync files to web folder
16:00 nsanden or public_html folder rather
16:01 ws2k3 joined #git
16:01 osse no merge!!
16:01 osse did you learn nothing?
16:01 Farelli joined #git
16:01 nsanden err git fetch and reset i mean
16:01 einnjo joined #git
16:01 GodGinrai nsanden: No.  I'm saying have the "origin" remote, which is github, and the "live" remote, which is live
16:01 Snugglebash joined #git
16:01 Electrometro joined #git
16:01 GodGinrai nsanden: and only you can push to the "live" remote
16:02 nsanden GodGinrai: ohh interesting
16:03 ahmedelgabri joined #git
16:03 osse so you can make the live repo a bare one, and then the script you want to write will be essentially one of the ready-made solutions from that link
16:04 nsanden very interesting! that would be nice not to have to ssh in
16:04 tgunr I forgot the --
16:04 l4v2 joined #git
16:05 tgunr interesting, `git log -- basis/db/tuples.tuples.factor' returns nothing
16:05 osse tgunr: makes sense. in any case, the first commit listed will be the one that removed (or renamed...) it
16:05 osse tgunr: one . should be a /
16:06 Electrom_ joined #git
16:06 tgunr doh!, too early in the morning, still on first cup of coffee
16:06 Puffball joined #git
16:06 dan2k3k4 joined #git
16:10 nsanden alright thanks osse, GodGinrai, others. I'll see if i can setup post-receive and push to live
16:14 qguv joined #git
16:15 eivarv joined #git
16:15 a_thakur joined #git
16:16 thierryp joined #git
16:18 mischat joined #git
16:18 Electrometro joined #git
16:18 kyan joined #git
16:19 overlord_tm joined #git
16:19 WayToDoor joined #git
16:20 b1tchcakes joined #git
16:22 Electrometro joined #git
16:23 e14 joined #git
16:25 ansimation joined #git
16:27 jimi_ joined #git
16:27 LeBlaaanc joined #git
16:30 cdg joined #git
16:31 faux___ joined #git
16:32 acetakwas joined #git
16:32 Snugglebash joined #git
16:38 nitric joined #git
16:39 Puffball_ joined #git
16:39 dimi1947 joined #git
16:43 diogenese joined #git
16:43 jottr_ joined #git
16:46 f3r70rr35f joined #git
16:46 fuchstronaut joined #git
16:47 troulouliou_div2 joined #git
16:47 InfoTest joined #git
16:48 danslo_ joined #git
16:48 dwsr joined #git
16:49 sostenuto joined #git
16:50 Puffball joined #git
16:51 cdg joined #git
16:53 mingrammer joined #git
16:53 qguv I'm giving a tutorial to first-year computer science students that may not know what a linked list is. Does anyone have slide recommendations for a very visual and high-level tutorial?
16:54 sea-gull joined #git
16:54 Gamecubic joined #git
16:54 qguv I was reading through the triggers, but the tutorials there are either too broadly scoped or too technical.
16:54 crose joined #git
16:55 faux___ joined #git
16:55 qguv I want them to be able to go home this evening and start tracking changes and making experimental branches. I don't care as much about getting them to collaborate with others.
16:56 NeverDie joined #git
16:56 nanga joined #git
16:59 mkoskar joined #git
16:59 underpantz_ joined #git
17:00 sn1c joined #git
17:02 geoid joined #git
17:05 jaequery joined #git
17:05 troulouliou_div2 joined #git
17:05 thomas_25 joined #git
17:05 thomas_25 left #git
17:07 robotroll joined #git
17:07 inflames joined #git
17:07 Darcidride joined #git
17:08 rgrinberg joined #git
17:08 rangergord joined #git
17:10 pythonist joined #git
17:10 lmatteis joined #git
17:11 AnarchyAo joined #git
17:11 dpower joined #git
17:12 WayToDoor joined #git
17:13 qpdb joined #git
17:16 Puffball joined #git
17:20 f3r70rr35f joined #git
17:21 Puffball_ joined #git
17:25 fedepad joined #git
17:26 qguv if anyone has advice, please mention my name and I'll get a notification, thanks!
17:26 bremner qguv: I've reasonable success walking 3rd year students through man gittutorial
17:26 gitinfo qguv: the gittutorial manpage is available at http://jk.gs/gittutorial.html
17:27 bremner first year is a bit different. The main problem I've noticed is that people's ability to use the command line is *cough* uneven
17:27 rangergord qguv: what's the question?
17:28 subhojit__ joined #git
17:29 LBV_User joined #git
17:29 bremner qguv: I modified the tutorial slightly to have them modify some existing examples and commit and merge them.
17:29 star_prone joined #git
17:30 dsantiago joined #git
17:30 UTAN_dev joined #git
17:30 acetakwas joined #git
17:31 jez9999 Does anyone know whether there is a way to automatically update subrepos when you git pull on a superrepo?
17:32 ahmedelgabri joined #git
17:35 dsantiag_ joined #git
17:35 Lunatrius` joined #git
17:36 laumann joined #git
17:36 fedepad joined #git
17:37 Guest9 joined #git
17:37 mingrammer joined #git
17:38 Noldorin joined #git
17:38 cjohnson jez9999: hooks
17:38 cjohnson or modifying the script that runs the pull
17:38 cjohnson assuming you mean submodules that is
17:38 jez9999 hmm well that sucks.  what about making the supermodule reference a tag rather than a commit hash?
17:38 jez9999 i think until you can do that, it will be a PITA to work with submodules
17:39 cjohnson you can ref a tag but you'll still have the same issue
17:39 cjohnson unless you update that submodule to update its tag
17:39 jez9999 huh?
17:39 cjohnson when you use submodules, the state of the submodule is tied into specific commits in your parent repo
17:39 jez9999 what i mean is, internally git always points to a commit hash, so that when the submodules are updated, git status will report changes in the supermodule
17:40 Achylles joined #git
17:40 jez9999 yes, so the supermodule doesn't reference a tag
17:40 Guest9 joined #git
17:40 cjohnson anyway to answer your question yes submodules are a PITA
17:40 jez9999 well why dont they fix this stuff?
17:40 cjohnson I recommend npm+browserify for JS and composer for PHP
17:40 jez9999 a lot of people have a problem with it
17:41 cjohnson VCS is not dependency management
17:41 cjohnson they don't have the same goals
17:41 jez9999 whats the point in even having submodules?
17:41 jez9999 i usually try to use them to share code
17:41 cjohnson referencing exactly another repository, not to use them as a shippable library link
17:42 Sasazuka__ joined #git
17:42 fedepad joined #git
17:42 cjohnson jez9999: what I do to make submodules more painless is I wrote wrapper scripts for my git commands. no extra stuff except the things that are repetitive
17:42 cjohnson branch foo = git checkout foo ; git submodule sync ; git submodule update --init --recursive
17:42 dreiss joined #git
17:43 jez9999 yeah but i dont see the point of that
17:43 cjohnson the point of VCS is so you can say you know exactly the state of your codebase at X point in time, and track the changes and when they were introduced
17:43 cjohnson you are trying to use VCS as dependency management
17:43 jez9999 what's wrong with storing the info "the current commit pointed to by 'tag'"?
17:43 cjohnson that's why your goals don't align with submodules
17:43 cjohnson it's ok to use git this way but you have to understand it's not the designed purpose
17:43 cjohnson npm or composer are much better suited for what you're describing
17:43 aarobc joined #git
17:43 jez9999 you know the state of your codebase without submodules.
17:44 jez9999 what does submodules add?
17:44 cjohnson when you're using a submodule, the state of your codebase *includes* the state of their codebases
17:44 jottr_ joined #git
17:45 jez9999 yeah
17:45 jez9999 but i don't see why that is useful
17:45 cjohnson that's by design
17:45 jez9999 it's a shit design
17:45 ilmari is there a way to get 'git log' to only show commits that are _not_ merged into a given branch?
17:45 jez9999 it should be expanded
17:45 jez9999 it's good for nothing
17:45 ilmari or a given set of branches
17:45 cjohnson jez9999: ok. git is open source. submit your patches
17:46 jez9999 right now, i don't really know what people use submodules for though.
17:46 cjohnson ilmari: are all those commits in a branch you could compare from?
17:46 jez9999 it's ludicrous that the supermodule has to have a commit log full of "update submodule reference" entries
17:46 cjohnson a la master..feature ?
17:46 jez9999 i mean it's just totally polluting the commit log
17:47 cjohnson jez9999: so use composer
17:47 cdg joined #git
17:47 _ikke_ Why are you using submodules when it's design doesn't fit your needs?
17:47 cjohnson no commit log
17:47 jez9999 i'm not.  i just wish the design did because i use git.
17:47 jez9999 and it seems like a missed opportunity
17:47 Vampire0 jez9999, well, it says exactly what you are doing. You are updating the version of the dependency
17:47 cjohnson "I wish this hammer would drive screws better"
17:47 Vampire0 jez9999, I mean the commit message
17:48 cjohnson "why don't you use a screwdriver?"
17:48 ilmari cjohnson: they are on a bunch of different branches. specifically basically I want to find my unmerged commits and the branches they're on
17:48 ilmari -basically
17:48 jez9999 cjohnson: yeah well i dont exactly buy into the unix philosophy.  for me using a million different tools when mope functionality could be wrapped into one is a pain
17:48 cjohnson ilmari: hrm...I'm not sure. you could do it via script though
17:48 jez9999 it's yet another thing to learn instead of being productive
17:48 cjohnson list all branches that aren't merged (git branch --no-merged), for each, list master..that-branch
17:49 ilmari yeah
17:49 danslo_ joined #git
17:49 cjohnson jez9999: well, then stop using such softwares, like git
17:49 Vampire0 jez9999, well, I'm being a gazillion times more productive on Linux than on Windows where I continuously have to deal with bullshit that could never happen on Linux ;-)
17:49 cjohnson I dunno what else to tell ya
17:50 jez9999 there's no fundamental reason why submodules couldn't allow referencing tags
17:50 jez9999 instead of commits
17:50 Guest23964 joined #git
17:50 jez9999 i want one git repo embedded in another, and i want its latest master
17:50 buffal0 joined #git
17:51 jez9999 it would be so simple for "git pull" to just update them
17:51 Dougie187 The fundamental reason is because that hasn't been tested. Git has no reason to think those things would work together at all.
17:51 Dougie187 But, you could just do `git submodule foreach git pull`
17:52 Dougie187 Or, `git pull --recurse-submodules`
17:52 cjohnson oh nice when was that flag added
17:52 Vampire0 jez9999, that's simply not what submodules are for. I'm sure the others told you so (scroll up and search for hammer and screwdriver ;-) )
17:52 jez9999 Vampire0: i'll have to remember to tell everyone they're worthless then
17:52 jez9999 it's idioptic
17:52 Dougie187 cjohnson: The man page says 1.7.3
17:52 jez9999 they're literally worthless
17:52 jez9999 i wouldnt use them for anything
17:52 cjohnson Dougie187: wow... let me read
17:53 jez9999 but it's misleading
17:53 jez9999 because they look like they could be useful
17:53 Dougie187 jez9999: You're saying everyone in here is useless, or the other tools are useless?
17:53 jez9999 no im saying submodules are
17:53 jez9999 in their current form
17:53 Dougie187 They aren't though.
17:53 cjohnson they're worthless as screwdrivers, yes
17:53 Dougie187 Did you see what I said?
17:53 cjohnson but the label on them says "hammer"
17:53 cjohnson and you are defiant on using them as screwdrivers and proclaiming them bad at the job
17:53 jez9999 well i want them to be screwdrivers god damnit!
17:53 jez9999 it would be useful!
17:53 cjohnson just...get...the...screwdriver
17:53 shinnya joined #git
17:53 Dougie187 jez9999: So... no you didn't read what I wrote?
17:54 _ikke_ submodules create a strong binding between the parent and child repo
17:54 cjohnson damn vampires
17:54 Vampire0 jez9999, submodules are not worthless if you use them for what they are designed.
17:54 jez9999 Dougie187: yeah remembering to write some extra stuff that shouldnt be needed, which doesn't fix the fundamental problem which is that supermodules reference a commit rather than a tag
17:54 cjohnson Dougie187: now do you know if this does submodule sync or if there's an option that does?
17:54 _ikke_ jez9999: a tag and a commit are the same in this context!
17:54 jez9999 Vampire0: lots of people need to embed shared code.  so if submodules arent designed for that, we should have something that is
17:55 jez9999 _ikke_: no they're not.
17:55 Dougie187 _ikke_: But jez9999 is using tag when be means branch.
17:55 Dougie187 Not tag.
17:55 jez9999 Dougie187: yeah sorry, branch
17:55 Vampire0 jez9999, release your shared lib to a repository and use an artifact dependency
17:55 Dougie187 cjohnson: I *think* `git pull --recurse-submodules` is equivalent to `git pull; git submodule update`
17:55 _ikke_ submodules point to a commit so that you know you'll have compattible versions
17:55 Dougie187 but I'm not 100% sure.
17:55 Vampire0 jez9999, or use repo
17:55 jez9999 _ikke_: right.  and i don't want that because it means tons of stupud extra commits for the supermodule.
17:56 _ikke_ then don't use submodules, point
17:56 cjohnson Dougie187: ok so I would still have to git submodule sync in the event that my submodule urls ever change
17:56 jez9999 yeah.  don't worry.  i won't.
17:56 Dougie187 cjohnson: Yeah, I think so.
17:56 _ikke_ doesn't mean they're useless, means your usecase is different
17:56 jez9999 but i want something that does fit my needs.
17:56 jez9999 they should add it to git.
17:56 cjohnson composer
17:56 jez9999 ive never heard of that
17:56 Dougie187 jez9999: You can contribute to git.
17:56 Dougie187 jez9999: It's open source.
17:56 _ikke_ It's a php package manager
17:56 jez9999 " Dependency Manager for PHP "
17:56 jez9999 im using node.js
17:56 cjohnson npm then
17:56 cjohnson does your job already
17:57 jez9999 not really.
17:57 cjohnson package.json, you specify a dependency range
17:57 cjohnson commit that
17:57 jez9999 npm relies on packages uploaded to NPM
17:57 cjohnson nope, you can use all sorts of sources
17:57 jez9999 if you have several local private packages npm is no good
17:57 cjohnson you can host your own repo
17:57 jez9999 ha
17:57 cjohnson it's really not hard
17:57 jez9999 it's a freaking pain compared to git offering that functionality
17:57 Dougie187 jez9999: but also, like _ikke_ said, just because you don't think you can use them doesn't mean that submodules are useless. Personally I use them in several of my projects and they are very useful.
17:57 cjohnson not really, it takes a few minutes to set up
17:57 cjohnson try it
17:58 jez9999 it's just logical for several git repos to be contained in another one
17:58 cjohnson jez9999: https://www.npmjs.com/package/sinopia
17:58 cjohnson here you go
17:58 SpeakerToMeat joined #git
17:58 cjohnson there's even docker images for zero config
17:58 cjohnson anyway, lunch
17:58 eivarv joined #git
17:58 jez9999 Dougie187: so your supermodules are full of commit messages "updating submodule"?
17:58 Dougie187 Not full of them. There are a handful of them
17:58 _ikke_ jez9999: look at !subtree then
17:58 gitinfo jez9999: The subtree merge method is great for incorporating a subsidiary git repo into your current one with "unified" history. Read http://git-scm.com/book/en/v2/Git-Tools-Advanced-Merging#_subtree_merge for more info, see also !git-subtree and !git-stitch-repo.
17:59 jez9999 Dougie187: how is that not annoying?
17:59 jez9999 those commits are 100% redundant
17:59 Dougie187 jez9999: There are typically api changes that are needed in the super proejct as well. So it's good to capture all of them together.
17:59 Limes_ joined #git
17:59 _ikke_ jez9999: Usually you don't update your submodules in isolation
18:00 jez9999 _ikke_: well i just think there should be _the option_ of referencing a branch
18:00 jez9999 i dont see why that cant be introduced
18:00 Dougie187 jez9999: again, you can feel free to contribute it.
18:00 Dougie187 Or request it, and then be happy when it is introduced.
18:00 jez9999 yeah but you lot sound like you would oppose it on principle
18:00 Dougie187 It would be useful, but it's a fragile workflow.
18:01 Dougie187 In general, you wouldn't expect arbitrary versions of a super project and a sub project to work together well.
18:01 Dougie187 As a result, you wouldn't want to have git automatically update your submodule to point to an untested version of the code without you explicitly asking for it.
18:02 Ragnor joined #git
18:02 _ikke_ Right
18:02 _ikke_ imagine upstream introducing breaking changes
18:02 averell joined #git
18:02 jez9999 if im controlling upstream, i know what's going on
18:02 jez9999 so im not going to break stuff
18:02 Dougie187 Then why do you have them separate?
18:03 jez9999 because code sharing.
18:03 Vampire0 jez9999, if you are controlling upstream, that is no the use-case for submodules
18:03 _ikke_ jez9999: what if you want to check an older version?
18:03 __main__ joined #git
18:03 jez9999 code sharing.
18:03 Dougie187 jez9999: OK, so you have other people contributing to your code. In which case you're not 100% in control, and at some point you'll likely run into api changes.
18:03 _ikke_ How do you know what version of the submodule to use so that it works?
18:03 Dougie187 Those api changes would break the super project still.
18:03 kerrick joined #git
18:04 Dougie187 As _ikke_ says, there is likely some range of sub project commits that are valid for a different range of  super project commits.
18:04 jez9999 Dougie187: it's a question of weighing that risk against the annoyance of redundant commits on the supermodule.  i think the latter is more of a problem.  it depends on your API stability i guess.
18:04 Vampire0 jez9999, maybe you want to try repo (the tool google built to compose one code-base out of several git repositories for Android)
18:04 Dougie187 jez9999: Well, it's either more commits in your super project, or additional commands.
18:04 jez9999 why additional commands?
18:04 Dougie187 It's easy enough to update all of your subprojects (you can even have a hook do it for you)
18:04 Dougie187 jez9999: because in general it's not a good workflow.
18:04 jez9999 Dougie187: it pollutes commit history.
18:05 multi_io joined #git
18:05 Dougie187 jez9999: No, you can update your submodule without committing it.
18:05 Dougie187 jez9999: So you could just update locally in your checkout as you want.
18:05 Dougie187 jez9999: Again, `git submodule foreach git pull`
18:05 Dougie187 something like that.
18:05 jez9999 update your submodule without committing it??
18:05 Dougie187 Yes
18:05 Dougie187 You don't want to commit it.
18:05 Dougie187 Because you don't like the commits.
18:06 jez9999 so, git status is going to permanently show changes
18:06 Dougie187 So?
18:06 jez9999 well that's terrible too.  i like my git status clean
18:06 Dougie187 You aren't polluting your repo with "redundant commits"
18:06 sbeller There is also `git submodule update --remote` to follow the upstream branch
18:06 Dougie187 jez9999: Try that then ^^
18:07 jez9999 sbeller: it still won't actually reference a branch will it?
18:07 jez9999 it might *track* one
18:07 sbeller s/follow/check it out for a moment/; you still need to commit the update to the superproejct
18:07 t4nkd joined #git
18:07 Dougie187 jez9999: Though, to be fair, it's not any different from the solution I gave you
18:07 Zamyatin joined #git
18:07 Dougie187 jez9999: git status will still be dirty.
18:08 ShekharReddy joined #git
18:08 Dougie187 It sounds to me though that you don't have many choices because the workflow you have setup is not a generally useful one.
18:08 snthsnth joined #git
18:08 rgrinberg joined #git
18:08 jez9999 here's a workflow: update utilities repo, commit and push, cd to another repo that uses that utilities lib, pull recursively, refresh visual studio, use new library API functions
18:09 jez9999 that is a common workflow for me
18:09 nikivi joined #git
18:09 _ikke_ That's indeed a very poor workflow
18:09 jez9999 (or would be if you could pull recursively and not commit the supermodule)
18:09 Dougie187 If you want to do that as a workflow, that's fine, but again it's not a generally useful workflow.
18:09 jez9999 poor?  pfft, it's a quick and easy way to update a library
18:09 Dougie187 It's very fragile
18:09 apotry joined #git
18:10 jez9999 i guess it can be.
18:10 Dougie187 No, it is
18:10 Dougie187 By definition it's a fragile workflow that is prone to breakage.
18:10 jez9999 you might as well say having a master branch is fragile because someone might check in some broken code.
18:10 Dougie187 Additionally, it defeats one of the main purposes for using version control systems.
18:11 Dougie187 jez9999: No, because you can return to the past and checkout that broken version and fix it.
18:11 jez9999 Dougie187: you can do that with submodules.
18:11 _ikke_ jez9999: You are self master over your own master branch (no pun intended)
18:11 Dougie187 jez9999: In your workflow, a random commit of your super project has no useful commits it points to for the sub projects.
18:11 cspack joined #git
18:11 Dougie187 As a result, you can never bisect your super project to figure out what broke, or how to fix it.
18:12 hahuang62 joined #git
18:12 dviola joined #git
18:12 Dougie187 Since you don't have commits that say "This version of my super project is now compatible with this other version of my sub project"
18:13 Dougie187 So, something as simple as finding / fixing a bug would take significantly longer, or even be impossible with the workflow you described.
18:13 jez9999 *shrug* it has worked for me in the past, is all i can say.  and it would've been a lot less of a pain if i didn't have to keep committing to the supermodule.
18:13 marc_v92 joined #git
18:14 jez9999 if the submodule is changing a lot, committing to the supermodule gets more annoying
18:14 BlaXpirit jez9999, make a library, publish on npm, don't use submodules
18:15 __main__ joined #git
18:15 bongjovi joined #git
18:15 kerrick joined #git
18:15 jez9999 why would i publish on npm?
18:15 jez9999 sometimes i care about my code being private
18:16 Dougie187 jez9999: All I can say is it doesn't sound like you have a good solution to your problem. You could try to get something contributed to git to provide the functionality that you're looking for, but again in general your workflow is fragile and not something you should recommend to other people to do.
18:17 fiddlerwoaroof joined #git
18:17 sbeller Is Jeremy Morton on here?
18:17 jez9999 Dougie187: there are a bunch of other discussions over the last few years of people wanting all submodules to be upgated recursively on a git pull, so it looks like i'm not the only one who wants something like this
18:17 jez9999 sbeller: why?
18:17 sbeller https://public-inbox.org/git/57F29FEF.30700@game-point.net/T/#u
18:18 _ikke_ jez9999: seeing you on the ML
18:18 warthog9 joined #git
18:18 Dougie187 jez9999: Updating submodules recursively is different from what you're asking for, and that functionality exists.
18:18 Dougie187 jez9999: As `git pull --recurse-submodules`
18:18 sbeller Not sure if the discussions cross over here, but it may be easier to stick to one medium
18:18 _ikke_ jez9999: is Jeremy Morton
18:19 jez9999 Dougie187: that's the kind of command i know i'll just forget
18:19 Dougie187 jez9999: So alias it.
18:20 jez9999 i'll just type "git pull"
18:20 Dougie187 jez9999: but even if you do it, it's different from what you're asking.
18:20 jez9999 just like i *always* type "git clone" and then curse because it isnt recursive
18:20 Dougie187 jez9999: I'm just guessing here, but I think it's equivalent to `git pull ; git submodule update`
18:20 sbeller cool :)
18:20 sbeller jez9999: to answer the question: I think that is by design, so you have a specific version that you can get in the future as well. If you want to just track the latest branch, you may be interested in git-repo
18:21 jez9999 sbeller: git help repo?
18:21 Dougie187 No, git-repo
18:21 Dougie187 it's a project from google
18:21 Dougie187 https://code.google.com/p/git-repo/
18:21 stfn joined #git
18:21 kulelu88 joined #git
18:22 jez9999 great, yet another tool
18:22 Dougie187 Doesn't sound like you have many options at this point though
18:22 fahadash joined #git
18:23 jez9999 frankly it's simpler for me just to down the repos separately
18:23 Dougie187 OK
18:23 jez9999 just annoying when git could offer the option
18:23 jaequery joined #git
18:23 Dougie187 Again, you could contribute it.
18:25 einnjo joined #git
18:26 vdv joined #git
18:31 mingrammer joined #git
18:33 __main__ joined #git
18:34 madewokherd joined #git
18:34 jesk joined #git
18:35 romerocesar joined #git
18:37 e14 joined #git
18:38 synthroid joined #git
18:38 King_Hual` joined #git
18:39 dave0x6d joined #git
18:42 t4knd joined #git
18:43 marc_v92 joined #git
18:45 kerrick joined #git
18:45 __main__ joined #git
18:45 jottr_ joined #git
18:48 fuchstronaut joined #git
18:49 kyan joined #git
18:49 danslo_ joined #git
18:50 LosPup joined #git
18:50 _Vi joined #git
18:52 eivarv joined #git
18:52 Endarked joined #git
18:53 _main_ joined #git
18:54 __main__ joined #git
18:55 eihabarabia joined #git
18:55 khmarbaise joined #git
18:56 _main_ joined #git
18:59 Balliad joined #git
19:00 duderonomy joined #git
19:03 Endarked joined #git
19:04 Anja joined #git
19:04 ansimation joined #git
19:05 ansimation joined #git
19:08 sdothum joined #git
19:08 snthsnth joined #git
19:12 nidr0x joined #git
19:15 cjohnson wow seriously this long
19:15 * cjohnson sprinkles garlic all over the place
19:19 Vampire0_ joined #git
19:20 elastix joined #git
19:22 ruby_ joined #git
19:25 mingrammer joined #git
19:25 armyriad joined #git
19:26 robotroll joined #git
19:27 swills joined #git
19:28 Snugglebash joined #git
19:28 swills if anyone is willing to help, i'd really appreciate it. i have a set of branches like this http://pastebin.com/21xxVzVx and want to merge the changes between branches/2016Q3 and swills-2016Q3 onto swills-2016Q4. is that possible and if so, how? i can't seem to get rebase or merge to do what i want, but i'm sure i'm doing something wrong.
19:29 acetakwas joined #git
19:29 thiago to merge: git checkout targetbranch; git checkout branchtobemerged
19:29 mattcen joined #git
19:29 thiago you do not get to choose a range. The merge merges everything that was done since the branch point (or last merge)
19:30 swills thiago: is that for me?
19:30 rangergord yes
19:30 swills that's just two checkout commands
19:32 thiago oops
19:32 thiago the second was git merge
19:32 Dougie187 You could have merges do what you said though.
19:32 Dougie187 It gets more confusing though
19:32 nds_ joined #git
19:33 rangergord thiago: still at Qt?
19:33 nikow joined #git
19:34 thiago rangergord: define "at Qt"
19:34 rangergord thiago: working full-time on Qt (I know you work for Intel)
19:35 thiago no, I am not
19:35 rangergord what do you do these days then, if it's not indiscrete to ask?
19:35 johnmilton joined #git
19:35 lindenle joined #git
19:35 thiago IoT
19:36 swills i guess i should note that the things that are on branches/2016Q3 are also first committed to master and then merged to branches/2016Q3
19:36 swills so the simple merge case fails
19:36 rangergord cool
19:37 Dougie187 swills: How many commits are on swills-2016Q3 but not on branches/2016Q3?
19:38 str joined #git
19:39 hashpuppy joined #git
19:39 cagedwisdom joined #git
19:40 hashpuppy i removed a bunch of tracking branches locally.  how can i remove all the branches that don't exist locally from my remote?
19:40 Vampire0 joined #git
19:40 hashpuppy i tried git remote prune origin --dry-run, but it shows nothing
19:41 johnmilton joined #git
19:41 hashpuppy i can only find info on the reverse scenario
19:41 tzafrir joined #git
19:42 Sadeukko joined #git
19:42 jottr_ joined #git
19:42 nsanden joined #git
19:44 multiproc joined #git
19:44 rangergord hashpuppy: why would you expect that to delete remote branches? Prune IIRC deletes branches unmerged to master. imagine 10 people collaborating. Every user will have their own local branches. user1 might not have branch7, should they be deleting stuff on origin?
19:44 EY joined #git
19:45 nsanden so i have a post-receive hook now and confirm its running, because i have an echo "test" in it and i see that in the output when i git push
19:45 rangergord err, deletes branches alreayd merged to master
19:45 Dougie187 rangergord: prune deletes stale branches that have already been deleted upstream
19:46 nsanden but i also have a git fetch origin master and a git reset --hard origin/master and one or both of these lines isnt really working. they work fine if i ssh in and run them manually though
19:46 Dougie187 hashpuppy: I think you want `git push origin :<branch1> :<branch2> ... :<branchN>
19:46 Dougie187 `
19:46 atomicb0mb joined #git
19:46 hashpuppy yeah, i have like 50 remote branches that don't exist locally and was wondering if there was something already out there.  i'll probably just write a script, then
19:47 Dougie187 hashpuppy: I don't think there's a single command for it, I'd recommend a script if you can write one easily.
19:47 Dougie187 If it's on github, they have a nice interface where you can click to delete the branches
19:49 fuchstronaut joined #git
19:50 timvisher joined #git
19:53 wogi joined #git
19:56 dodobrain joined #git
19:58 synthroid joined #git
19:59 tvw joined #git
20:01 ahr3n joined #git
20:03 publio joined #git
20:07 mattcen joined #git
20:08 swills Dougie187: at the moment just one, but i'm trying to work out the workflow for cases where there would be many many more
20:08 swills so i've just created a test branch and done one commit on it
20:08 acetakwas joined #git
20:12 hahuang62 joined #git
20:12 marc_v92 joined #git
20:14 WayToDoor joined #git
20:15 sn1c left #git
20:17 Rodya_ joined #git
20:18 phillly joined #git
20:19 mingrammer joined #git
20:21 nanga joined #git
20:21 nanga joined #git
20:21 dan2k3k4 joined #git
20:25 elisa87 joined #git
20:25 mattcen joined #git
20:26 m_antis joined #git
20:26 Gitzilla joined #git
20:26 elisa87 hey not sure why it uploaded the gist anonymously. How can I upload it under my github useraccount from terminal? I installed the tools using gem https://gist.github.com/anonymous/2d40d2cd9f20de5a5e98b9ba1ec67f68
20:28 canton7 elisa87, which tools? hub?
20:29 elisa87 canton7: this https://github.com/defunkt/gist
20:29 canton7 did you follow the instructions under 'login'?
20:31 jimi_ joined #git
20:33 jimi_ joined #git
20:35 anonymuse joined #git
20:36 converge joined #git
20:38 converge Im inside a magento/ directory, and created a new git repository, after I did "git add app/design/myDir" , git commit/push.. but when I see the repository, git added all dirs from magento/
20:38 converge did I do something wrong ?
20:38 canton7 git you commit -a perhaps?
20:39 vassagus joined #git
20:39 converge canton7: I did git commit -m 'first commit'
20:39 canton7 !repro
20:39 gitinfo [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
20:39 converge after git push origin master
20:41 Limes_ joined #git
20:42 converge just found the problem -_- git add pub/media/ .
20:43 anonymuse joined #git
20:43 Seveas woops :)
20:43 fiddlerwoaroof joined #git
20:47 rgrinberg joined #git
20:49 boombatower joined #git
20:50 aarobc joined #git
20:50 fuchstronaut joined #git
20:51 danslo_ joined #git
20:53 boombatower joined #git
20:55 DolphinDream joined #git
20:57 garethdaine joined #git
20:58 HanClinto joined #git
21:01 Kitty- joined #git
21:01 Puffball joined #git
21:01 xaa joined #git
21:01 converge just installed magento, the backend is so slow.. it's an imac , 8gb ram. is there some magic to speed it up ?
21:03 dan2k3k4 joined #git
21:03 crayon joined #git
21:03 khmarbaise joined #git
21:03 crayon joined #git
21:04 faux___ joined #git
21:04 Puffball joined #git
21:05 cyphase joined #git
21:07 Puffball joined #git
21:07 Snugglebash joined #git
21:08 Darren_ joined #git
21:11 l4v2 joined #git
21:12 _rnsanchez_ joined #git
21:13 mingrammer joined #git
21:13 Seveas converge: it's magento, it's not meant to be fast :)
21:14 Seveas get some caching module
21:16 xall joined #git
21:17 star_prone joined #git
21:19 snthsnth joined #git
21:20 converge Seveas: nice, ill check it out
21:21 Puffball joined #git
21:22 navidr joined #git
21:23 d3r1ck_ joined #git
21:25 Atm0spher1c joined #git
21:27 converge bin/magento setup:upgrade --keep-generated is resulting in constraint violation (duplicated entry) , is there somewhere I can edit the SQL script ? or how to fix it.. ?
21:29 jeffreylevesque joined #git
21:31 Seveas try a magento channel
21:31 converge ops, sorry
21:31 raz left #git
21:31 * thiago wonders if magento has anything to do with magneto
21:31 treehug88 joined #git
21:32 Seveas they're both evil
21:32 apotry joined #git
21:32 thiago but weren't always?
21:35 ollien joined #git
21:36 Snugglebash joined #git
21:37 mattcen joined #git
21:38 converge magento is love, allow people/business with low resource to have their 1st store
21:38 thiago joined #git
21:39 aarobc joined #git
21:39 blackwind_123 joined #git
21:41 aidalgol joined #git
21:41 Puffball joined #git
21:48 mischat joined #git
21:50 Dougie187 left #git
21:51 fuchstronaut joined #git
21:51 nettoweb1 joined #git
21:52 Anja joined #git
21:52 danslo_ joined #git
21:53 fatalhalt joined #git
21:56 fatalhalt joined #git
21:57 LeBlaaanc joined #git
21:58 Anja joined #git
22:02 nettoweb joined #git
22:02 drawkula joined #git
22:02 mizu_no_oto joined #git
22:04 blackwind_123 joined #git
22:04 ShekharReddy joined #git
22:13 hahuang62 joined #git
22:13 crayon joined #git
22:16 t0n1c joined #git
22:18 jason237 joined #git
22:18 rnsanchez_ joined #git
22:19 star_prone joined #git
22:23 Fr0stBit joined #git
22:23 Fr0stBit Hello o/
22:23 crayon joined #git
22:23 crayon joined #git
22:24 Fr0stBit I am using git lfs and in my .gitattributes file i have a rule with pattern */ext/** to match all the ext folders and assign the git lfs attributes (filter=lfs diff=lfs etc..)
22:24 Fr0stBit I want to exclude a family of filetypes from this list. How can this be done?
22:25 Fr0stBit Documentation says that any rule that comes afterwards will override anything that is set, witch i tested and worked using a rule like */ext/**/*.glsl with attributes (filter= diff= merge=)
22:26 Fr0stBit Is this the correct way to set the attributes to defaults? (By using an empty value)
22:27 Fr0stBit (which* for the grammar nazis that saw it)
22:27 snthsnth_ joined #git
22:27 axl_ joined #git
22:27 mattcen joined #git
22:31 b1tchcakes joined #git
22:33 Limes_ joined #git
22:34 madewokherd joined #git
22:36 rwb joined #git
22:36 sangy joined #git
22:44 rwb joined #git
22:49 anonymuse joined #git
22:49 fatalhalt joined #git
22:49 Kitty- joined #git
22:52 fuchstronaut joined #git
22:52 dviola joined #git
22:54 danslo_ joined #git
22:55 Hink_ joined #git
22:56 raijin joined #git
22:56 fmcgeough joined #git
22:57 blahdodo joined #git
23:01 raspado joined #git
23:02 anuxivm joined #git
23:02 Hink_ joined #git
23:05 macivy joined #git
23:07 SwiftMatt joined #git
23:09 ResidentBiscuit joined #git
23:10 mattcen joined #git
23:13 duderonomy joined #git
23:17 mablae_ joined #git
23:19 star_prone joined #git
23:24 evie_hammond joined #git
23:27 crayon joined #git
23:28 crayon joined #git
23:31 EvilPenguin joined #git
23:31 thott joined #git
23:31 thott joined #git
23:32 einnjo joined #git
23:34 Kitty- joined #git
23:34 atomicb0mb joined #git
23:35 davix joined #git
23:39 mischat joined #git
23:40 _ADN_ joined #git
23:40 Lyqyd joined #git
23:40 boxhead joined #git
23:43 zivester joined #git
23:43 boxhead left #git
23:43 perrier-jouet joined #git
23:44 lanox joined #git
23:44 lanox this is driving me insane. I have dir foo and inside i have 1  and 2  inside 1 i have another directory call bar how can i ignore bar and everything in it
23:44 lanox I have tried /foo/1/bar/ and /foo/1/bar/*
23:44 lanox and nothing works
23:48 unreal joined #git
23:48 apotry joined #git
23:50 lanox joined #git
23:50 rbr joined #git
23:50 lanox anyone
23:50 ojacobson lanox: in a .gitignore in the root of your repo, /foo/1/bar/ will ignore any untracked files in foo/1/bar/ or any child directory.
23:51 ojacobson If you believe this is not happening, can you gist the contents of .gitignore, and the git status command that's surprising you?
23:51 RustyShackleford joined #git
23:51 faux___ joined #git
23:51 ^7heo s/gist/pastebin/
23:52 lanox https://gist.github.com/anonymous/c007d48a14b5b0ccdbb7d04f10d9a175
23:52 lanox that is exactly what I am doing
23:52 * ^7heo mv #git #github
23:52 lanox I have tried with / at the end and with /* and with out /
23:52 fuchstronaut joined #git
23:53 lanox no matter what i
23:53 lanox do it still add content of bin and pkg
23:53 dsdeiz joined #git
23:53 dsdeiz left #git
23:53 dsdeiz joined #git
23:54 NeXTSUN joined #git
23:54 EvilPeng1 joined #git
23:55 RustyShackleford so here's what I did: feature1 branched from master, then I submitted a PR. meanwhile, I branched feature2 from feature1
23:55 Lor_ joined #git
23:55 RustyShackleford when the feature1 PR  was accepted, that gets squashed down to one commit. when trying to sync my repo again, I have a bunch of merge conflicts
23:56 lanox ojacobson:  did you see the gist
23:57 ojacobson No, I was tabbed out, hang on
23:57 RustyShackleford I did a git reset --hard upstream/master on my master and feature1 branches. Now, how to fix feature2?
23:57 ojacobson So where's the surprising git-status output, lanox?
23:59 lanox sorry here is git status
23:59 lanox https://gist.github.com/lanox/7c9814cf6678586163d3ddd082e6e472
23:59 Kaisyu joined #git

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