Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-06-09

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

All times shown according to UTC.

Time Nick Message
00:00 Lyqyd joined #git
00:01 ash_mz joined #git
00:01 radialneon joined #git
00:01 ash_mz .digraph a->b
00:01 gitinfo .digraph: http://g.jk.gs/dv.png
00:02 milki so good
00:02 ash_mz ironically, I _just_ wanted the url and didn't realize it was in the topic
00:03 milki o
00:03 milki thats why i was executed the other day
00:03 ash_mz I'm sorry to hear that, ghost of milki
00:05 ash_mz stupid; sub \n for actual newlines and you can do `[label=""+\n"Start your label on a new line"]` but not `[label=+\n"Start your label on a new line"]`
00:05 rominronin joined #git
00:07 mikecmpbll joined #git
00:09 fracting joined #git
00:10 lopin joined #git
00:11 drodger joined #git
00:12 boudiccas joined #git
00:13 IWishIKnew joined #git
00:16 mmlb joined #git
00:17 d5sx43 joined #git
00:19 xall joined #git
00:19 drodger joined #git
00:19 griffindy joined #git
00:20 lopin joined #git
00:20 diogenese joined #git
00:24 eletuchy joined #git
00:24 neps joined #git
00:27 fuchstronaut joined #git
00:27 al-damiri joined #git
00:28 blur3d joined #git
00:28 TheVoid joined #git
00:29 ajf- joined #git
00:30 ShekharReddy joined #git
00:32 pwnawannab[NE] joined #git
00:36 boudicca joined #git
00:40 boudiccas joined #git
00:40 diegoaguilar joined #git
00:40 hahuang65 joined #git
00:41 eroussel joined #git
00:42 rkazak joined #git
00:47 fission6 joined #git
00:47 zturner joined #git
00:52 zturner is it possible to have git convert line endings for all staged files?  not interested in core.autocrlf, i just want to run dos2unix or vice versa on every staged file
00:52 watabou joined #git
00:53 droidicus joined #git
00:53 Schwarzbaer_ joined #git
00:54 drodger joined #git
00:55 SimonNa joined #git
00:58 xlegoman joined #git
01:01 xlegoman_ joined #git
01:03 mizu_no_oto joined #git
01:03 xlegoman_ joined #git
01:03 livingstn joined #git
01:04 Wulf4 joined #git
01:04 jaguarmagenta joined #git
01:05 dermoth joined #git
01:07 rominronin joined #git
01:08 jchonig joined #git
01:09 drodger joined #git
01:10 jon-mac joined #git
01:10 eletuchy_ joined #git
01:11 Dougie187 joined #git
01:12 espindola joined #git
01:15 eletuchy joined #git
01:17 Ahmed90 joined #git
01:18 vanquish349 joined #git
01:18 wgrant joined #git
01:19 eletuchy_ joined #git
01:20 fission6 joined #git
01:21 eletuch__ joined #git
01:22 fstd_ joined #git
01:22 watabou joined #git
01:22 pks joined #git
01:23 eletuchy joined #git
01:24 ISmithers joined #git
01:24 fracting joined #git
01:24 ISmithers Morning o/
01:25 rueben joined #git
01:26 italoacasas joined #git
01:27 eletuchy_ joined #git
01:28 milki zturner: why do you need git to do it?
01:28 milki zturner: sounds like something you can just script up
01:29 eletuch__ joined #git
01:29 neps joined #git
01:29 ISmithers I am looking for some advice on general repository structure. We have platform updates for a platform we author and maintain. As an example we roll out an update each quarter, 1.4, 1.42, 1.5 and so forth.
01:30 ISmithers If a client is on 1.3, they need to be applied sequentially (1.5 doesn't include preceding updates, you only ever upgrade from 1.42, to 1.5) What is a good way to set this up in a repository?
01:30 ISmithers Shall I just make each version a new branch?
01:30 aardvark___ joined #git
01:31 ISmithers They would never get merged to master. Or should I make each version a subsequent commit on master? Then during upgrades, you rewind to the desired commit and go from there?
01:32 ensyde joined #git
01:32 eletuchy joined #git
01:32 jeffreylevesque joined #git
01:34 eletuchy_ joined #git
01:34 bilb_ono joined #git
01:35 dermoth joined #git
01:38 mmlb joined #git
01:46 boudicca joined #git
01:47 thiago joined #git
01:51 IWishIKnew joined #git
01:51 romerocesar joined #git
01:53 boudiccas joined #git
01:54 dmto joined #git
01:54 eletuchy joined #git
01:56 drodger joined #git
01:56 xall_ joined #git
01:58 boudicca joined #git
01:58 eletuchy_ joined #git
02:01 eletuchy joined #git
02:01 NeverDie_ joined #git
02:02 edude03 joined #git
02:02 Cata|Work joined #git
02:02 Lynxium joined #git
02:03 drbean_ joined #git
02:03 eletuch__ joined #git
02:07 eletuchy joined #git
02:08 rominronin joined #git
02:08 SwiftMatt joined #git
02:11 Sasazuka joined #git
02:14 hectortrope joined #git
02:15 hectortrope hello all how can I upload my files to git repo ? Help please
02:20 eletuchy joined #git
02:20 espindola joined #git
02:21 lopin joined #git
02:22 eletuchy_ joined #git
02:22 justanotheruser joined #git
02:24 eletuch__ joined #git
02:25 neps joined #git
02:27 eletuchy joined #git
02:29 raijin joined #git
02:30 cdg joined #git
02:31 aardvark___ joined #git
02:31 ahr3n joined #git
02:32 eletuchy_ joined #git
02:34 eletuch__ joined #git
02:35 durham_ joined #git
02:37 drodger joined #git
02:37 eletuchy joined #git
02:39 harish joined #git
02:39 eletuchy_ joined #git
02:41 scoobertron joined #git
02:43 eletuchy joined #git
02:45 interne7y joined #git
02:46 eletuchy_ joined #git
02:46 ensyde__ joined #git
02:48 eletuch__ joined #git
02:48 pmort joined #git
02:50 eletuchy joined #git
02:52 Dougie187 left #git
02:54 eletuchy_ joined #git
02:55 drodger joined #git
02:55 hexagoxel joined #git
02:56 bjpenn joined #git
02:58 chachasmooth joined #git
02:58 sgen how do I remove a files changes from the cache without changing the actual file?
02:58 sgen git rm --cached?
03:00 sgen nvm found i t
03:00 kadoban sgen: 'git status' tells you. It should be 'git reset HEAD -- the_file'
03:00 edude03 joined #git
03:01 durham joined #git
03:02 jstimm joined #git
03:03 neps joined #git
03:03 aavrug joined #git
03:04 drodger joined #git
03:04 aavrug joined #git
03:05 jaguarmagenta joined #git
03:06 sangy joined #git
03:06 inflames joined #git
03:09 rominronin joined #git
03:09 drodger joined #git
03:15 zivester joined #git
03:16 dreiss joined #git
03:20 drodger joined #git
03:20 k3rn31 joined #git
03:22 durham joined #git
03:22 ochorocho__ joined #git
03:22 rnsanchez_ joined #git
03:27 xelxebar joined #git
03:30 Limes_ joined #git
03:30 son joined #git
03:31 eijk_ joined #git
03:32 MDRXB95AP joined #git
03:32 texinwien joined #git
03:33 aardvark___ joined #git
03:33 foo joined #git
03:33 pwnz0r joined #git
03:33 foo If I do git diff <git id> ... where the git id is from 10 commits ago, do I see all changes between current and 10 commits ago?
03:33 bjpenn joined #git
03:34 diogenese joined #git
03:35 shgysk8zer0 joined #git
03:37 drodger joined #git
03:38 eletuchy joined #git
03:38 eletuchy joined #git
03:40 k3rn31 joined #git
03:42 shgysk8zer0 joined #git
03:43 codebam joined #git
03:44 pwnz0r joined #git
03:46 interne7y joined #git
03:46 eletuchy joined #git
03:49 foo Has to be the case, from my tests
03:51 bjpenn joined #git
03:51 son joined #git
03:53 robotroll joined #git
03:54 josuebrunel joined #git
03:55 boudiccas joined #git
03:55 dreiss joined #git
03:57 son joined #git
04:01 boudiccas joined #git
04:03 Mixologic_ joined #git
04:04 son joined #git
04:10 rominronin joined #git
04:11 guardianJ joined #git
04:20 Cabanossi joined #git
04:22 Mechanical_Man joined #git
04:22 Goplat joined #git
04:24 romerocesar joined #git
04:28 Mechanical_Man So I'm having an interesting issue. I moved one file from one location to another, but when I try to stage, the diff is showing a similarity index of 92% and that I removed every line from the file and then readded them. I used a hex editor to compare the previous version from the version I'm trying to commit and they are identical, so I really don't know why the similarity index is showing 92%.
04:29 fuchstronaut joined #git
04:32 ash_mz_ joined #git
04:33 pungi-man joined #git
04:33 aardvark___ joined #git
04:38 shgysk8zer0 joined #git
04:42 govg joined #git
04:43 thiago Mechanical_Man: check that it's showing the correct source path
04:43 thiago is it the file you moved from?
04:44 Mechanical_Man Oh sorry I forgot to attatch the gist of the diff: https://gist.github.com/anonymous/6b131b2e0744569c134c1d59eaf7d71f
04:45 thiago well, git is right and you're wrong
04:45 arescorpio joined #git
04:45 thiago are you on Windows?
04:45 Mechanical_Man yep windows 7
04:45 thiago the difference is probably the line ending
04:46 Mechanical_Man But that would show up in the hex editor right?
04:46 thiago not necessarily
04:46 Mechanical_Man hmm
04:46 thiago if the original file had the wrong line endings, they are now getting corrected
04:46 thiago hence the difference
04:46 Mechanical_Man ah I see.
04:47 interne7y joined #git
04:47 durham joined #git
04:48 Mechanical_Man Appreciate the help thiago. I'll just see if I can find a way to compare the line endings.
04:49 thiago you have to tell Git to stop converting crlf first
04:50 Mechanical_Man I switched the it to use Unix for commiting
04:51 Mechanical_Man thiago: But I'm really surprised that line endings wouldn't show up as differences in a hex editor. I mean I'm assuming that line endings are stored in the actual file right?
04:52 kadoban They would show up, yes.
04:54 Mechanical_Man So if line endings are actually the issue here, then I must be doing something wrong when I viewed the contents of the old version. I'm using Git Extensions and I just saved a copy to my desktop. Just to be sure I'll try checking out the old revision and use the hex editor on the file.
04:58 subhojit777 joined #git
04:58 Mechanical_Man Files are showing up as identical with Hxd
05:00 tinajohnson___ joined #git
05:01 MattMaker joined #git
05:01 dimi1947 joined #git
05:01 muthu joined #git
05:04 Grievar joined #git
05:04 thiago joined #git
05:04 thiago Mechanical_Man: because the file is being corrected on check-in
05:04 thiago you probably have a file with CRLF line-endings
05:05 thiago but when you add it again, Git will convert it to LF on check-in
05:05 Fijit joined #git
05:06 Mechanical_Man Right... Ok so internally inside git it's changing the line endings and not the actual file outside in the directory right?
05:06 thiago correct
05:06 jaguarmagenta joined #git
05:06 Mechanical_Man thiago: Thanks you! Now I understand what's going on :)
05:06 MattMake_ joined #git
05:09 Mechanical_Man left #git
05:10 griffindy joined #git
05:11 rominronin joined #git
05:15 shafox joined #git
05:18 Macaveli joined #git
05:18 przemoc joined #git
05:18 przemoc joined #git
05:18 Macaveli joined #git
05:21 b0x joined #git
05:21 b0x hey all
05:22 ISmithers Ello
05:22 b0x im having trouble with composer installed extension (for behat) .. I'm including the /vendor directory in my repo as I run tests on gitlab/various others, so cant be doing composer install all the time..
05:23 b0x however, one of the extensions has the .git folder inside it, and therefore all its subfolders can't be added
05:23 b0x as its a 'submodule'
05:24 b0x how can i get around this? i've tried force adding, trying to add the .git folder to the gitignore (via./.. 'vendor/**/.git')
05:24 tax joined #git
05:24 b0x its not listed as a submodule (or mentioned at all) in my main repo's .git/config
05:24 b0x so im at a loss :(
05:24 ISmithers .git is the Git repository.
05:25 ISmithers You can't add a repository to a repository.
05:25 b0x i know, but its installed with composer, so i would never use it as a repo
05:25 ISmithers So delete it then.
05:25 ISmithers Or move composer outside your repository.
05:25 b0x but it would just return when i run composer install?
05:25 ISmithers Its a bit strange that the installer adds the repository... where did you get the installer from?
05:26 b0x https://github.com/Vanare/behat-cucumber-formatter
05:26 wxg joined #git
05:26 daxroc joined #git
05:27 metastableB joined #git
05:27 Awesomecase joined #git
05:27 DuncanT joined #git
05:27 GameGamer43 joined #git
05:27 ISmithers That's not an installer, wheres the installer?
05:27 theresajayne joined #git
05:28 stuh84 joined #git
05:29 sauvin joined #git
05:29 rfv joined #git
05:29 addshore joined #git
05:29 kutenai joined #git
05:29 devhost joined #git
05:29 janx joined #git
05:29 boxrick1 joined #git
05:29 angular_mike joined #git
05:29 DarkMukke joined #git
05:30 dunk joined #git
05:30 braderhart joined #git
05:30 zeller joined #git
05:30 angiebadillo joined #git
05:30 bttf joined #git
05:30 yena joined #git
05:31 ajm203_ joined #git
05:31 imanc joined #git
05:31 ams__ joined #git
05:31 contracode joined #git
05:31 shennyg joined #git
05:31 etcinit joined #git
05:31 flupke joined #git
05:31 cloudcell__ joined #git
05:31 barq joined #git
05:31 wizonesolutions joined #git
05:31 epic joined #git
05:32 paperziggurat joined #git
05:32 sauvin joined #git
05:32 vcoinminer joined #git
05:32 ionelmc joined #git
05:32 kwmiebach joined #git
05:32 harleypig joined #git
05:32 twisted` joined #git
05:32 ToeSnacks joined #git
05:32 {AS} joined #git
05:32 worstadmin joined #git
05:32 skelterjohn joined #git
05:32 drodger joined #git
05:32 johtso joined #git
05:32 rossome joined #git
05:32 WillMoogle joined #git
05:33 elijah joined #git
05:33 discopatrick joined #git
05:33 teadrop_ joined #git
05:33 MLM joined #git
05:33 fahadash joined #git
05:33 mehdi_ joined #git
05:33 shafox joined #git
05:33 spr0tsuki joined #git
05:33 cstrahan joined #git
05:33 huhlig joined #git
05:33 parliament joined #git
05:33 noahsussman joined #git
05:34 aendrew joined #git
05:34 Lowell joined #git
05:34 parsnip joined #git
05:34 jonha joined #git
05:34 bcardarella joined #git
05:34 JohnLea___ joined #git
05:34 nezZario joined #git
05:34 da5id joined #git
05:34 FritzLakritz joined #git
05:34 cmosguy joined #git
05:34 aardvark___ joined #git
05:34 elmcrest joined #git
05:34 diogenese joined #git
05:34 bedouin joined #git
05:34 JackMc joined #git
05:35 quackgyver joined #git
05:35 angelsl joined #git
05:35 cspotcode joined #git
05:35 M-meznak joined #git
05:35 jleclanche joined #git
05:35 akitada joined #git
05:35 evanc joined #git
05:35 dimi1947 joined #git
05:36 bbhoss joined #git
05:36 Tee_Pee joined #git
05:36 sfan786 joined #git
05:36 XDS2010 joined #git
05:37 zero7 joined #git
05:37 XDS2010 joined #git
05:37 Meow-J joined #git
05:37 prosodyContexte joined #git
05:38 XDS2010 joined #git
05:38 dimi1947 I 've merged a branch into master...what seems odd to me that staged changes for commit appear...for commits I made in that branch,now they reappear for commit in master branch...is that natural?
05:38 XDS2010 joined #git
05:39 b0x errr
05:39 b0x ISmithers: ill paste bin it
05:39 b0x my composer.json
05:39 XDS2010 joined #git
05:40 b0x http://pastebin.com/GicRD3CH
05:40 b0x then just composer install
05:40 b0x brb
05:40 ISmithers b0x: When you say installer, I imagine something like a win32 installer. Are you using the word 'install' to describe cloning a repository?
05:40 XDS2010 joined #git
05:40 b0x no
05:40 b0x im using 'composer install'
05:41 b0x which reads the composer.json, and downloads everything
05:41 ISmithers So "$ composer require --dev vanare/behat-cucumber-json-formatter" ?
05:41 XDS2010 joined #git
05:41 b0x yes, however i've just added it to the json manually, same effect
05:41 ISmithers I guess I am not following how installing something like that, also brings down the .git repository unless they have misconfigured their package somehow.
05:42 ISmithers Like, I install a heap of stuff using Node and never get the repositories with my packages.
05:42 b0x yes
05:42 b0x thats my confusion :(
05:42 zincrokx joined #git
05:42 ggherdov joined #git
05:43 ISmithers OK so I would say that is an issue with their package. Do the other packages that you have listed in your json, like drupal-extension for example, also come with their .git repositories included?
05:44 ISmithers (I would assume almost certainly not, it sounds very odd to me)
05:46 Raging_Hog joined #git
05:49 PCatinean joined #git
05:49 jaguarmagenta joined #git
05:49 apotry joined #git
05:51 Mobutils joined #git
05:52 ISmithers Still there b0x?
05:53 bjpenn joined #git
05:54 LucaTM joined #git
05:55 ThomasLocke joined #git
05:59 santa-fire joined #git
05:59 mariuscc joined #git
06:00 t0by joined #git
06:00 stuh84 joined #git
06:00 neanderslob joined #git
06:01 aardvark___ joined #git
06:03 fracting joined #git
06:03 dopesong joined #git
06:04 dsdeiz joined #git
06:10 Psychiatrist joined #git
06:10 phanimahesh joined #git
06:10 cyan__ joined #git
06:10 Meow-J joined #git
06:11 sfan786 joined #git
06:11 t0by joined #git
06:11 prosodyContexte joined #git
06:12 brainwave joined #git
06:12 rominronin joined #git
06:12 brainwave Hey guys, I am new to version control
06:14 phanimahesh brainwave: Everyone is new at some point of time. :)
06:14 phanimahesh here to just hang out, or looking for help with a query?
06:16 settermjd joined #git
06:16 rscata joined #git
06:16 snfgf joined #git
06:17 jceb joined #git
06:17 Mobutils joined #git
06:17 brainwave phanimahesh: Had some queries but I found a few things to try. If it doesn't work out I'll ask here, thanks!
06:20 boudiccas joined #git
06:23 zeroed joined #git
06:26 ahmedelgabri joined #git
06:27 boudicca joined #git
06:28 xorox90 joined #git
06:30 fuchstronaut joined #git
06:30 Andrew_K joined #git
06:31 boudiccas joined #git
06:32 GandalfTheWizard joined #git
06:35 ImJune joined #git
06:36 SoRoot joined #git
06:36 fracting joined #git
06:38 boudiccas joined #git
06:39 liuzhen joined #git
06:40 JeroenT joined #git
06:40 boudicca joined #git
06:44 SuperflyR6 joined #git
06:44 leeN joined #git
06:45 boudiccas joined #git
06:45 Alienpruts joined #git
06:47 dimi1947 I 've merged a branch into master...what seems odd to me that staged changes for commit appear...for commits I made in that branch,now they reappear for commit in master branch...is that natural?
06:47 boudicca joined #git
06:47 interne7y joined #git
06:50 rominronin joined #git
06:51 boudiccas joined #git
06:51 jagob joined #git
06:53 LaT0rtue joined #git
06:53 cbreak dimi1947: it's a merge commit
06:53 cbreak the merge commit contains the merged state from both branches
06:53 cbreak if you compare it with the former tip of one of the branches, you see all the changes introduced by the other branch
06:54 sparetire joined #git
06:55 b0x ISmithers: sorry was on lunch
06:55 ISmithers Unforgiveable
06:55 b0x :D
06:55 xorox90 joined #git
06:55 b0x to answer your question
06:55 ISmithers I was just curious if other packages you install also come down with their .git repository, as that seems to me like a broken package.
06:56 b0x yes the drupal extension subfolders have .git folders
06:56 boudiccas joined #git
06:56 ISmithers And those are not ones created by you?
06:56 jokajak joined #git
06:57 dimi1947 cbreak:I do not understand since commits were made in this other bramch and then merge took place what is the point in repeating them in master branch
06:57 b0x correct
06:57 xbob joined #git
06:57 ISmithers OK then I as confused as you on that subject.
06:57 cbreak dimi1947: it doesn't repeat anything
06:57 ISmithers I don't know why they would include the source repository, in the package install.
06:57 ISmithers That makes no sense to me, unless I am missing something.
06:57 ISmithers Maybe others know.
06:58 ISmithers I use Grunt to do a bunch of task processing, and none of the packages I install and use, come with their source repository.
06:58 b0x well
06:58 b0x just checked slack
06:58 b0x one of the other devs got around it somehow
06:58 cbreak dimi1947: it should be obvious that if you compare the merge commit with its ancestors, you see all the changes from the other branch, because this is the whole point of merging
06:58 b0x checking how
06:59 b0x something about git cache
06:59 dimi1947 cbreak:yes but git asks me to make commits in master that I have already done in the non-master branch(the one being merged)
06:59 cbreak no
06:59 cbreak it asks you to make a single commit: the merge commit
06:59 cbreak (if there is no conflict, it will not even ask you, just create it)
06:59 dimi1947 cbrea:oh....
07:00 [Brain] joined #git
07:00 dimi1947 cbreak:so you mean whenever merge happens...the commited changes in the branch are staged again for the merge commit
07:01 cbreak the merge commit will have the state the merge produces. If there are conflicts, you have to make it the way you want it to be yourself
07:01 cbreak dimi1947: no
07:01 cbreak git does not stage changes
07:01 cbreak git does not work via changes
07:01 cbreak it works via state
07:01 dimi1947 cbreak:I am lost again a little
07:01 cbreak the merge commit does not contain changes either, it contains the merged state
07:02 JeroenT joined #git
07:02 cbreak if you have a branch adding file A, and an other file B, and you merge both branches, the result will be a commit containing BOTH file A and file B
07:02 cbreak the merge commit does not contain any changes of course
07:02 cbreak but when comparing it with the first branch, it appears to add file B, and when comparing with the second branch it appears to add file A
07:02 akiva joined #git
07:02 PCatinean joined #git
07:05 dimi1947 cbreak:I understand this example about file A and file B...what I do not understand is that after merge takes place into master git asks me to commit the adding of these 2 files...I thought this would be done automatically upon merging
07:05 cbreak if there is a conflict, you have to resolve it
07:06 SuperflyR6 joined #git
07:07 ahmedelgabri joined #git
07:07 dimi1947 cbreak:so you are saying that this happens WHEN there is a conflict..cause I had one
07:08 liuzhen joined #git
07:09 cbreak dimi1947: if there's no conflict, it would do the same, but commit it without your involvement
07:10 Rish joined #git
07:11 rolha joined #git
07:11 TomyWork joined #git
07:11 dimi1947 cbreak:what do you mean "commit it without your involvement"
07:12 cbreak exactly that
07:12 cbreak you tell git merge, and it will commit the merge
07:12 cbreak and you end up with the merge commit
07:13 shafox left #git
07:14 dimi1947 cbreak:still do not get it...I am presented with unstaged changes,changes that I had made in the merged branch
07:14 cbreak no
07:15 cbreak everything should be staged or conflicting
07:15 moei joined #git
07:15 cdown joined #git
07:16 emPi joined #git
07:16 alcohol joined #git
07:16 encod3 joined #git
07:17 dimi1947 cbreak:I do not understand...
07:17 dimi1947 cbreak:sorry
07:18 cbreak don't see the problem with understanding
07:19 dimi1947 cbreak:be more specific...
07:19 cbreak all is so simple
07:19 cbreak !cs
07:19 gitinfo "Git for Computer Scientists" is a quick introduction to git internals for people who are not scared by phrases like Directed Acyclic Graph. http://eagain.net/articles/git-for-computer-scientists/ See also !concepts !bottomup
07:20 dreiss joined #git
07:20 jceb joined #git
07:21 cbreak !gcs
07:21 gitinfo [!concepts] "Git Concepts Simplified" explains the basic structures used by git, which is very helpful for understanding its concepts. http://gitolite.com/gcs.html
07:23 dimi1947 cbreak:you said "everything should be staged..."can you explain that a little
07:23 cbreak which part?
07:23 cbreak that's the way merging works
07:24 cbreak the new state will be staged
07:24 cbreak (unless it conflicts)
07:25 dimi1947 cbreak:so..staging means that I must commit these changes
07:25 cbreak sure, if you want them in the merge result
07:26 ISmithers I wouldn't say it means you *must* commit so much as it they are ready to be committed.
07:26 cbreak you can unstage them too, or change them
07:26 cbreak but introducing new changes into a merge commit is considered evil
07:26 ISmithers You move files to staging in order to make them part of the next commit.
07:26 yuhlw joined #git
07:27 ISmithers You unstage them if you don't want them to be a part of the next commit and when ready, you make a commit with a message, and all staged files will be a part of it.
07:27 dimi1947 cbreak:so let me see if I got it right...
07:28 imjacobclark joined #git
07:28 imjacobclark joined #git
07:28 dimi1947 cbreak:I have a branch makes some commits to it..then merge into master...and the changes I made to that branch must be commited now into master...
07:30 mablae joined #git
07:33 ahmedelgabri joined #git
07:33 SASDOE joined #git
07:34 SASDOE Anyone familiar with git flow. I get everything except for the release command. When am I supposed to start a new release? Before or after I've coded the features?
07:34 Juzzika joined #git
07:34 cdown_ joined #git
07:37 yuhlw joined #git
07:37 mikecmpbll joined #git
07:39 pmort joined #git
07:40 phanimahesh dimi1947: read man git-merge, specifically the option --commit and --no-commit
07:40 gitinfo dimi1947: the git-merge manpage is available at http://jk.gs/git-merge.html
07:40 circ-user-7QvWA joined #git
07:41 phanimahesh By default, --commit is assumed. So when no confliucts are detected, git makes a commit automaticallyfor you.
07:41 phanimahesh you can prevent it by passing --no-commit to git, where it will just stage all changes and let you make the commit.
07:41 phanimahesh when a conflict happens, git can not make the commit automatically.
07:42 phanimahesh so the behaviour changes to that of --no-commit.
07:42 dimi1947 phanimahesh:now I understand it...
07:43 dimi1947 so you are saying that when conflicts happen on merging...I must make the commit myself(which if no coflict occured on merging this would happen automatically)
07:43 Timvde phanimahesh: what happens if you use --commit with a merge conflict? Or will it just refuse to commit?
07:43 phanimahesh git will refuse to commit because it can not make a clean merge
07:44 Timvde Okay, as I thought :)
07:44 phanimahesh --commit can be read as: if there is a clean merge, please make a commit for me.
07:44 phanimahesh because that's what most people want most of the time, it's the default
07:45 garethdaine joined #git
07:45 josuebrunel joined #git
07:46 phanimahesh and --no-commit should be read as: stage the merge, but don't commit it. If there are conflicts, they are not staged and git prompts you to resolve manually.
07:46 dimi1947 phanimahesh:NOW I GOT IT
07:48 phanimahesh :)
07:50 phanimahesh SASDOE: Sometimes a matter of preference, but git-flow convention is to start a release when you are done with features and have a release candidate.
07:51 phanimahesh every commit on the release branch should be a release candidate or a release, ideally.
07:51 ash_mz_ joined #git
07:52 ToBeCloud joined #git
07:53 troulouliou_div2 joined #git
07:53 troulouliou_div2 joined #git
07:55 GavinMagnus joined #git
07:55 peepsalot joined #git
07:59 mikecmpbll joined #git
08:00 nafg_ joined #git
08:02 eletuchy joined #git
08:02 nafg_ .
08:02 nafg_ Hi can someone help me with line endings, no matter what I do I can't get a clean git diff (meaning to show my actual changes)
08:03 nafg_ #1, is there some way to tell what EOL the file was previously committed with (I'm guessing windows, but just to make sure)?
08:04 SASDOE http://stackoverflow.com/questions/20496084/git-status-ignore-line-endings-identical-files-windows-linux-environment
08:05 garethdaine joined #git
08:06 nafg_ SASDOE: Thanks, but it's not helping
08:06 nafg_ #1, is there some way to tell what EOL the file was previously committed with (I'm guessing windows, but just to make sure)?
08:06 nafg_ #2, unix2dos'ing the file isn't helping either
08:08 nafg_ Like, is the problem that my editor changed the line endings?
08:08 nafg_ Or just because I'm on linux git gets confused?
08:10 intellix joined #git
08:11 rolha joined #git
08:11 phanimahesh nafg_: on linux editors usually write unix line endings.
08:11 phanimahesh some may convert by default
08:12 nafg_ phanimahesh: right but even if i tell it to save otherwise git still complains
08:12 phanimahesh checkout man gitattributes, sections eol and end of line conversion
08:12 gitinfo the gitattributes manpage is available at http://jk.gs/gitattributes.html
08:13 phanimahesh nafg_: Are there any git attributes set? Are you sure the line endings haven't been changed?
08:13 nafg_ phanimahesh: no, I'm trying to get more insight into that
08:13 nafg_ Anyway I just did  git diff -w --ignore-blank-lines > 3easy.diff
08:13 nafg_ So now I want to "start over"
08:14 eletuchy_ joined #git
08:14 phanimahesh use unix diff or a hex viewer to check
08:14 phanimahesh what do you mean by start over?
08:14 nafg_ reset --hard and apply the pax
08:15 nafg_ hex viewer is a good idea, what's installed by default on linux?
08:15 nafg_ ubuntu
08:15 marcogmonteiro joined #git
08:16 a_thakur joined #git
08:16 nafg_ ok after a reset --hard I only see 0a (LF right?) in it
08:16 nafg_ my patch doesn't apply cleanly though
08:16 eletuch__ joined #git
08:17 nafg_ Does .gitattributes have to be committed?
08:17 sajjadg joined #git
08:18 sajjadg hi, is it possible to clone a git directory into another git branch?
08:18 sajjadg I wanna create a new git repo and have a branch for the old repo
08:19 eletuchy joined #git
08:20 chll joined #git
08:20 aswen joined #git
08:21 watabou joined #git
08:21 cbreak dimi1947: no, no changes have to be committed
08:22 cbreak as I told you before, only STATE has to be committed
08:22 troyt joined #git
08:23 b0x joined #git
08:23 jceb joined #git
08:24 eletuchy_ joined #git
08:25 mdw joined #git
08:28 eletuchy joined #git
08:30 Mixologic joined #git
08:30 nafg_ remind me how to reset working dir file to index?
08:31 fuchstronaut joined #git
08:32 Rish joined #git
08:33 eletuchy_ joined #git
08:33 b0x git checkout .
08:33 cbreak nafg_: git checkout filename
08:33 kadoban nafg_: 'git checkout -- the_file'
08:33 b0x or git reset --hard HEAD
08:33 cbreak b0x: no!
08:34 b0x :D
08:34 nafg_ no TO INDEX
08:34 mischat joined #git
08:34 grassass joined #git
08:35 acetakwas joined #git
08:35 boingy joined #git
08:35 diegoaguilar joined #git
08:36 eletuchy joined #git
08:37 Jellyg00se joined #git
08:40 acetakwas joined #git
08:41 soc42 joined #git
08:41 eletuchy_ joined #git
08:43 acetakwas joined #git
08:44 eletuchy joined #git
08:45 rolha joined #git
08:45 BSAlb joined #git
08:46 courrier joined #git
08:47 eletuchy_ joined #git
08:48 Orion3k joined #git
08:48 interne7y joined #git
08:49 eletuch__ joined #git
08:49 Schwarzbaer_ joined #git
08:50 boudiccas joined #git
08:51 eletuchy joined #git
08:54 eletuchy_ joined #git
08:54 mehola joined #git
08:55 ocbtec joined #git
08:55 govg joined #git
08:55 boudicca joined #git
08:56 govg joined #git
08:57 jknetl joined #git
08:58 grift joined #git
08:58 SoRoot joined #git
08:58 boudiccas joined #git
08:58 romerocesar joined #git
08:59 bruce_lee joined #git
09:00 eletuchy joined #git
09:00 notebox joined #git
09:02 sajjadg joined #git
09:02 PatrykDz joined #git
09:03 PatrykDz left #git
09:03 eletuchy_ joined #git
09:03 sajjadg left #git
09:04 Sceorem joined #git
09:04 jceb joined #git
09:06 eletuchy joined #git
09:07 howdoi joined #git
09:08 eletuch__ joined #git
09:08 goTAN joined #git
09:09 SuperflyR6 left #git
09:10 eletuchy_ joined #git
09:11 mdw_ joined #git
09:12 mns joined #git
09:13 eletuchy joined #git
09:15 imjacobclark joined #git
09:16 eletuchy_ joined #git
09:16 c0ded joined #git
09:18 eletuch__ joined #git
09:18 ocbtec joined #git
09:18 c0ded joined #git
09:18 SoRoot joined #git
09:19 osse i wonder if he ever realized
09:19 SoRoot joined #git
09:19 lb sooner or later
09:21 eletuchy joined #git
09:23 paul424 joined #git
09:24 eletuchy_ joined #git
09:25 diegoaguilar joined #git
09:26 kurkale6ka joined #git
09:27 eletuchy joined #git
09:27 nilg joined #git
09:27 M-sbts joined #git
09:27 muthu joined #git
09:28 nnyk joined #git
09:28 Saxon joined #git
09:29 paul424 I get to do " git apply 0001-*.patch  // reapply the patch, without committing, solve the conflict with RELEASE-NOTES.md "
09:29 paul424 how do I solve the confilicts with RELEASE-NOTES.md
09:29 paul424 what's that file for ?
09:29 me joined #git
09:29 netcarver joined #git
09:29 JeroenT joined #git
09:30 lb paul424 like you solve every problem with conflicts?! !conflicts
09:30 lb hm !confilct
09:30 lb !conflict
09:30 gitinfo [!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
09:30 lb there it is
09:31 lb lol the gateway for freenodes webchat was k-lined ? :D
09:32 M-sbts ping
09:32 lb M-sbts pong
09:32 lb !ping
09:32 gitinfo I only respond to !reply_time
09:32 M-sbts lb: thanks
09:32 Psychiatrist joined #git
09:33 Orion3k joined #git
09:37 shiriru joined #git
09:39 scalability-junk joined #git
09:39 joki joined #git
09:40 scalability-junk hey. I am using git https with user:pw within the push/pull repository links. These are added in CI etc. On a successful push git returns the remote url with user:pw and therefore leaks information. any method to prevent that?
09:40 scalability-junk Switching to ssh is not the preferred solution as that would require ssh inside the CI.
09:42 ronskisika joined #git
09:44 notebox joined #git
09:45 paul424 I get : git commit -a 'New Culling method : now culling works for both of the cameras, convex_hull from boost::geometry used  '
09:45 paul424 fatal: Paths with -a does not make sense.
09:45 paul424 what's wrong ?
09:46 cbreak scalability-junk: if someone has access to that "CI" thing, then they can look at how the link is generated anyway
09:46 dsdeiz joined #git
09:46 cbreak scalability-junk: have you thought about not putting the password in the link?
09:47 paul424 paths to files , that's what it means ?
09:47 scalability-junk cbreak: yeah, if they have access that's true. But the output will be logged and having the credentials in there is undesirable, if there is an easy solution to suppress them that would be nice.
09:48 garethdaine joined #git
09:48 cbreak scalability-junk: don't put them in there :)
09:48 scalability-junk cbreak: the call to push looks like this: "git push https://user:pw@gitexample.com/stuff/repo.git master"
09:49 cbreak remove the password
09:49 scalability-junk and provide the password on prompt? how would that be done automatically?
09:50 cbreak sure, you can do that
09:50 cbreak or use the curl native means
09:51 vishnurao joined #git
09:51 cbreak scalability-junk: i.e. via a .netrc file
09:51 * paul424 jumps , please please help me
09:51 muthu_ joined #git
09:51 scalability-junk cbreak: thanks will look into that. So git uses curl under the covers for http?
09:52 cbreak yes
09:52 cbreak libcurl I think
09:52 cbreak or use a credentials helper
09:52 cbreak see man git-config, search for credential
09:52 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
09:53 scalability-junk Thanks.
09:54 King_Hual joined #git
09:54 King_Hual joined #git
09:54 ahr3n joined #git
09:54 cbreak paul424: it means that your command makes no sense
09:54 cbreak why do you try to give git commit both -a (to commit everything), and specify a file?
09:54 paul424 ahh I meant -m
09:54 paul424 git -a -m , now it works
09:58 keltvek joined #git
09:58 eletuchy joined #git
09:59 vishnurao left #git
10:00 pingo joined #git
10:03 Macaveli joined #git
10:03 fuchstronaut joined #git
10:03 son joined #git
10:04 pingo When rebasing a feature branch on master is it possible to specify a merge strategy to always overwrite feature with changes from master *but* only bother me to manally solve conflicts for a few files or directories I specify?
10:04 XDS2010 joined #git
10:05 tvw joined #git
10:05 Alok_ joined #git
10:06 eijk joined #git
10:06 rnsanchez joined #git
10:06 fuchstronaut joined #git
10:08 watabou joined #git
10:11 iLembus joined #git
10:11 rahtgaz joined #git
10:11 ferr1 joined #git
10:12 permalac joined #git
10:12 permalac hello guys, I'm kind of new with git and I have done a stupid thing
10:12 permalac I have deleted a file with rm from command line,
10:13 permalac the git st shows me a D in front of it.
10:13 permalac can I restore the file easily ?
10:13 permalac getting it from the remote would be ok for me.
10:14 philipp64|work joined #git
10:15 Rish joined #git
10:16 alok__ joined #git
10:16 permalac git checkout -- filename.php     #Awsome!!!
10:16 permalac :)
10:17 Rish_ joined #git
10:17 Alok_ Yeah
10:17 Alok_ :)
10:17 ShekharReddy joined #git
10:19 jknetl joined #git
10:21 synthmeat joined #git
10:22 mdw joined #git
10:24 Lynxium joined #git
10:24 i7c Version Control to the rescue! \o/
10:25 davisonio joined #git
10:27 NeverDie joined #git
10:30 ycon_ joined #git
10:30 raijin joined #git
10:31 kedare joined #git
10:31 ycon_ Hi all, I'm trying to do a pip install of a branch. pip install https://github.com/user/repo.git@branch doesn't work
10:32 mostlybadfly joined #git
10:34 XDS2010 joined #git
10:36 jhass sounds like a #python issue, less so a #git one
10:37 khelkun joined #git
10:40 lb ycon_ LMGTFY: http://lmgtfy.com/?q=python+pip+install+from+github
10:43 ojacobson joined #git
10:53 JeroenT joined #git
10:54 espindola joined #git
10:55 iLembus_ joined #git
10:56 djb-irc joined #git
10:58 a_thakur joined #git
10:59 yqt joined #git
10:59 fracting joined #git
10:59 romerocesar joined #git
11:01 mdw joined #git
11:02 ramsub07 joined #git
11:02 barteks2x joined #git
11:03 settermjd joined #git
11:04 dsdeiz_ joined #git
11:04 johnmilton joined #git
11:05 XDS2010 joined #git
11:07 Psychiatrist joined #git
11:08 Tomdarkness joined #git
11:09 watabou joined #git
11:13 dan2k3k4 joined #git
11:13 a_thakur joined #git
11:13 jaguarma_ joined #git
11:13 Saxon joined #git
11:14 Psychiatrist joined #git
11:15 nd joined #git
11:15 boudicca joined #git
11:16 SoRoot joined #git
11:17 goTAN joined #git
11:19 ramsub07 joined #git
11:21 k3rn31 joined #git
11:22 clmsy joined #git
11:25 mizu_no_oto joined #git
11:28 Macaveli joined #git
11:28 ramsub07 joined #git
11:29 bnnb joined #git
11:30 circ-user-7QvWA When I rebase a small patch many  commits down, I get some conflicts in the unrelated code.
11:31 circ-user-7QvWA Can I tell git to ignore other changes, accumulated between two?
11:33 sdothum joined #git
11:33 cdown joined #git
11:35 XDS2010 joined #git
11:35 fission6 joined #git
11:37 j`ey joined #git
11:37 edude03 joined #git
11:37 j`ey I want to do 'git show --stat', but in a way to just get the files names
11:37 j`ey is there a way to do that?
11:37 fission6 joined #git
11:38 davisonio joined #git
11:41 j`ey git diff-tree --no-commit-id --name-only -r HEAD seems to do the trick
11:41 goTAN joined #git
11:44 JoshL joined #git
11:44 reactormonk left #git
11:47 basiclaser joined #git
11:47 ronskisika joined #git
11:47 rbr joined #git
11:48 basiclaser hey y'all, whats the recommended method for installing git on OSX?
11:48 qpdb joined #git
11:49 j`ey left #git
11:49 basiclaser i don't want to install xcode tools particularly
11:50 subhojit777 joined #git
11:52 jaguarmagenta joined #git
11:54 SoRoot joined #git
11:54 baraba joined #git
11:54 cbreak permalac: you checked out the file from the cache, all local state, no remote communication needed :)
11:54 cbreak if you had committed the deletion, you would have had to specify a commit to get it from: git checkout HEAD~1 -- filename
11:54 cbreak again, no remote communication needed
11:55 cbreak git history is all kept local
11:55 jon-mac joined #git
11:56 SoRoot joined #git
11:56 i7c basiclaser, maybe this? https://git-scm.com/download/mac (I have not the slightest idea about mac :)
11:56 basiclaser thanks
11:56 basiclaser it seemed dodgy becuase it went to sourceforge, and was made by an unsigned/ unidentified developer
11:56 cbreak basiclaser: tried to run git on the CLI?
11:56 basiclaser yeh
11:57 cbreak does it not offer to install the command line dev tools?
11:57 SoRoot joined #git
11:57 i7c Indeed, it goes to source forge. Maybe not the best choice as a hosting platform. :D
11:58 SoRoot joined #git
11:58 basiclaser it does but source / unidentified developer
11:58 basiclaser so dodgy
11:58 cbreak I read sourceforge adds malware to the software they offer
11:58 basiclaser indeed
11:59 cbreak basiclaser: I'd get it from apple
11:59 basiclaser my OSX tried to block me from installing it
11:59 cbreak the command line dev tools, if you can't get xcode
11:59 garethdaine joined #git
11:59 basiclaser yeh ill go with apple
11:59 cbreak I personally compile git from source
11:59 basiclaser thx guys
11:59 i7c I just use the very up-to-date packaging of my distro. <3
11:59 i7c But compiling is quite simple as well.
11:59 cbreak git version 2.7.4  is the one bundled with xcode at the moment
11:59 cbreak i7c: what is the version in your distro? :)
12:00 i7c git version 2.8.3
12:00 cbreak i7c: on some of the linux I have to deal with it's still a pre-2 :(
12:00 i7c Arch Linux here :)
12:00 cbreak how unprofessional
12:00 i7c Is it now?
12:00 cbreak RHEL is much more professional! Everything is ancient
12:00 i7c :D
12:01 i7c So very stable, just without all the features.
12:01 LaT0rtue joined #git
12:01 SirCmpwn joined #git
12:02 Sceorem joined #git
12:02 moritz it's so stable that biologists have developed a special word for it
12:03 jbitdrop joined #git
12:04 i7c fossils?
12:05 stasmakarov joined #git
12:09 d10n-work joined #git
12:11 CEnnis91 joined #git
12:14 dbolser_ joined #git
12:14 dbolser_ joined #git
12:14 dbolser_ hello
12:14 rwb joined #git
12:15 astrofog joined #git
12:15 dbolser I have a staged commit I want to apply on top of one commit back
12:16 dbolser let me try again...
12:16 dbolser I have a diff I want to add to the last but one commit
12:17 dbolser I would normally "git commit --amend" to do this to the /last/ commit, but I want to do the same to the /last but one/
12:17 i7c !interactive_rebase
12:17 gitinfo Interactive rebase sounds similar to rebase but has completely different abilities. It can do this to commits: change the order, squash some of them together, remove some, add random existing commits from other branches, split them, and more... it's very powerful. Documentation is in the section "Interactive Rebase" in 'man git-rebase'.
12:17 moritz i7c: I meant "dead", actually
12:17 i7c moritz :D too obvious for me.
12:18 dbolser i7c: so I stage this commit then rebase -i and re-order them?
12:18 i7c dbolser, commit it, rebase -i reorder, and replace "pick" by squash in the appropriate line.
12:18 i7c Yes.
12:18 dbolser or I commit this change .. thanks
12:18 JeroenT joined #git
12:18 i7c dbolser, careful with the order in the interactive rebase. It's just the opposite of what git log does.
12:19 * moritz tends ot use "fixup" much more often than "squash"
12:19 dbolser ic
12:19 i7c I.e. "last" commit is on the bottom.
12:19 cdown joined #git
12:20 i7c Indeed, fixup. Won't prompt you with a weird patched-together commit message :)
12:20 dbolser i7c: beautiful :-D
12:20 lb more love for fixup
12:22 dbolser oic
12:22 dbolser mixxed up commit message actually gave me confidence in what I was doing
12:23 JanC_ joined #git
12:23 jhass bonus points, git commit --fixup=<sha>; git rebase -i --autosquash
12:24 allcentury joined #git
12:25 weox joined #git
12:25 den2204_ joined #git
12:25 pwnawannab[NE] joined #git
12:27 rgrinberg joined #git
12:29 davisonio joined #git
12:30 Alienpruts joined #git
12:30 synthroid joined #git
12:33 gebi joined #git
12:35 gebi hi :), is there a way to get git to resolve DNS names through the configured socks5 proxy?
12:35 XDS2010 joined #git
12:36 jeffreylevesque joined #git
12:39 Mixologic_ joined #git
12:40 scalability-junk cbreak: I was looking into .netrc and am now using credential helpers aka the credential store from git. The credentials are stored in ~/.git_credentials, but they are not read unfortunately. Neither the username nor the password. I could not find more info on why this fails.
12:41 notebox joined #git
12:41 cbreak scalability-junk: did you configure the credentials helper?
12:41 scalability-junk yes, double checked the config file itself.
12:41 cbreak the man page I gave you, man git-config points to man gitcredendials for more info
12:41 gitinfo the git-config manpage is available at http://jk.gs/git-config.html
12:41 cbreak weird
12:42 cbreak well, I've never used any of that
12:42 scalability-junk cbreak: yeah already through that one and through all subpages :P
12:42 cbreak I just do it properly :)
12:42 cbreak (via ssh)
12:43 svankmajer joined #git
12:43 dstarh joined #git
12:44 scalability-junk I don't really see the value in using ssh instead of https. Either I have the keys laying around or the pw/user. The only difference is more stuff inside the built environment :P, but yeah seems like the only solution :(
12:44 allcentury joined #git
12:46 mikecmpbll joined #git
12:46 SoRoot joined #git
12:46 j416 scalability-junk: being able to delete the pubkey from the server if your laptop etc. gets in the wrong hands? :)
12:47 scalability-junk j416: it's for the build environment. Deleting the pubkey is the same as changing the pw/token in my opinion so :P
12:49 monoxane joined #git
12:50 interne7y joined #git
12:50 scalability-junk cbreak: j416 thanks for your help
12:50 scalability-junk seems like git credentials is not used widely. almost no documentation besides the man pages.
12:50 livingstn joined #git
12:52 j416 scalability-junk: I think most use ssh, that's why
12:52 yqt joined #git
12:53 j416 we use ssh pubkeys for build environment too.
12:53 j416 one-time setup, so not an issue
12:53 fission6 joined #git
12:55 scalability-junk j416: but you need ssh in the build environment. for now I only had to have curl, git, ca-certs, static binary doing the build. But yeah, one last try and if that doesn't work, go ssh go :D
12:55 j416 scalability-junk: how do you ssh to the build environment? you have sshd but not ssh? o_O
12:56 permalac cbreak, thanks for the explanation.
12:56 scalability-junk j416: the build environment is a separate minimal container and therefore doesn't need ssh. Neither ssh nor sshd.
12:56 j416 scalability-junk: ah, container. I see.
12:56 Psychiatrist joined #git
12:57 MadLamb joined #git
12:59 MadLamb joined #git
12:59 MadLamb I rebased my branch with another branch, but then I decided I do not want to use that branch anymore. How can I revert this rebase?
13:00 MadLamb I guess reflog is not an option because I had many changes on this branch after the rebase
13:02 espindola joined #git
13:02 tlaxkit joined #git
13:03 italoacasas joined #git
13:03 jhass MadLamb: did you push the version you want to go back to anywhere?
13:03 MadLamb jhass, no
13:03 MadLamb jhass, I did, but the new changes are also there
13:03 jhass then git reflog is your best bet afaik
13:04 MadLamb jhass, I guess I can rebase interactive and not pick the commits of the rebased branch
13:04 CheckDavid joined #git
13:04 italoacasas joined #git
13:05 jhass mh, yeah
13:06 deadnull_ joined #git
13:07 jimi_ joined #git
13:07 k3rn31 joined #git
13:07 bollullera joined #git
13:07 j416 MadLamb: if you did not do anything else (such as rebasing again or .. other operations I don't remember, Git has conveniently put the state before your rebase in ORIG_HEAD)
13:08 j416 er, move that closing parenthesis left a bit. :)
13:08 MadLamb j416, yes, i rebased again :(
13:08 j416 MadLamb: then git-reflog
13:08 MadLamb j416, I was expecting the commits to be removed when the other branch were merged to master, but then I realized was a bad change
13:09 troulouliou_div2 joined #git
13:09 MadLamb j416, git reflog will not help because I can't return to that state, I have other changes that happened later that I need to preserve
13:09 happy-dude joined #git
13:10 j416 MadLamb: ok
13:10 CJKinni joined #git
13:11 JanC_ joined #git
13:12 watabou joined #git
13:13 jknetl joined #git
13:13 khelkun hi. How could I prohibit everyone to commit changes to a file after it was added to the repo please ?
13:14 i7c You could'nt.
13:14 i7c couldn't*
13:14 khelkun fair enough ^^
13:14 khelkun thx
13:15 i7c I mean not if you don't prohibit commits at all. There are tools that make basic access control and prohibit pushing to certain branches for example.
13:15 i7c But none of this ships with git itself, as far as I know.
13:15 j416 khelkun: by doing e.g. code reviews and finding that they did something stupid through that, or if you really want to (I don't think you want to), perhaps something like gitolite can enforce it, but I'm not sure if it can be configured on file path level
13:16 j416 i7c: that said, you can never prevent that people commit locally, but you can take measures to prevent what gets pushed to a shared repo.
13:16 i7c Yup.
13:16 j416 well you could take their keyboard
13:16 j416 maybe
13:16 i7c :)
13:16 i7c They'll just use the onscreen clicky thing then.
13:17 i7c Bad devs always find a way!
13:17 j416 s/bad dev/better dev/
13:17 khelkun I see thanks for the advice. It's about sql schema migration script, I'll use a hook on the local dev workstation to ensure the script syntax is valid, but I don't want people to be able to modify the script as soon as they pushed it to the origin
13:17 khelkun <j416> well you could take their keyboard
13:17 khelkun :)
13:18 j416 khelkun: aha. well, if your devs do that, then it seems you have greater problems. :)
13:18 khelkun I do :)
13:18 j416 I would try to fix those problems instead
13:18 fission6 joined #git
13:19 j416 we put migration files in our repo. We don't have a process to prevent them from changing. But people are sane enough to not modify them :)
13:19 j416 even if they did modify them, well we're using Git, so we can see who modified it and what they did
13:20 Nevik joined #git
13:20 marcelofr joined #git
13:20 rnsanchez_ joined #git
13:20 clobrano joined #git
13:21 clobrano joined #git
13:22 khelkun Right. I'm actually quite new to database versioning, I guess I'll setup this git hook to check the scripts syntax. And try to trust people about not modifying an existing valid script
13:22 fracting joined #git
13:23 ki0 joined #git
13:23 diegoaguilar joined #git
13:23 j416 we don't do git hooks either
13:24 j416 well, if the individual dev prefers to set up a hook, we don't prevent it
13:24 j416 but we don't have any rules about it
13:24 khelkun ok thanks again for sharing your own experience
13:24 j416 personally I don't care for commit hooks, because in my preferred workflow I commit a lot of unfinished things
13:25 ojacobson joined #git
13:25 j416 and then I squash and reorganise when things start to become ready
13:26 khelkun it makes sense
13:28 marcogmonteiro joined #git
13:30 qt-x joined #git
13:31 axiom_1 joined #git
13:34 tlaxkit joined #git
13:36 XDS2010 joined #git
13:37 noidea joined #git
13:37 fjfpvd joined #git
13:39 leeN joined #git
13:40 bollullera left #git
13:41 goTAN joined #git
13:43 Erajo joined #git
13:43 [Brain] joined #git
13:44 [Brain] joined #git
13:44 cyan__ joined #git
13:45 Raging_Hog joined #git
13:45 mbrzuzy joined #git
13:46 irco joined #git
13:46 texinwien joined #git
13:48 kaldoran joined #git
13:50 goTAN joined #git
13:50 _pix joined #git
13:51 ahmedelgabri joined #git
13:52 intellix joined #git
13:53 Macaveli joined #git
13:53 sitaram_ j416: gitolite can enforce at file path level, just FYI
13:53 j416 sitaram_: thanks!
13:53 j416 sitaram_: do you have a hl set on "gitolite"? :D
13:53 T4nk5bb0t joined #git
13:55 thiago joined #git
13:55 phaleth joined #git
13:55 davisonio joined #git
13:55 italoacasas joined #git
13:55 bollullera joined #git
13:56 ronskisika joined #git
13:56 fahadash joined #git
13:57 harish joined #git
13:57 aleskandro joined #git
13:59 AaronMT joined #git
13:59 exarch joined #git
14:00 marcelofr joined #git
14:01 Guest3535 joined #git
14:01 ekinmur joined #git
14:01 aleskandro Hi all I've a git bare in a server. I woul to configure in a shared users way with umask 002. So I set the bashrc in the home of the users I need to have umask 002 and if they login by ssh they have the umask setted to 002. But when they push some objects the umask is still 022. How can I set it?
14:01 SimonNa joined #git
14:01 Ch4rAss joined #git
14:02 mdw joined #git
14:02 cdg joined #git
14:02 drodger joined #git
14:03 cbreak aleskandro: man git-config
14:03 gitinfo aleskandro: the git-config manpage is available at http://jk.gs/git-config.html
14:03 cbreak aleskandro: look for core.sharedRepository
14:05 Ali2 joined #git
14:06 intellix joined #git
14:09 Rish joined #git
14:09 aleskandro cbreak in the server I set git config core.sharedRepository
14:09 aleskandro but I still have rwxr-xr-x
14:09 cbreak what do you set it to?
14:10 MutantMahesh joined #git
14:10 aleskandro I run git config core.sharedRepository
14:10 aleskandro can I git config core.sharedRepository 002?
14:10 cbreak that returns the current value
14:10 cbreak aleskandro: that would only give others access
14:10 cbreak aleskandro: you should set it to 0660 or so
14:10 JeroenT joined #git
14:10 cbreak it's all described in the man page
14:10 Vap joined #git
14:11 Psychiatrist joined #git
14:11 intellix joined #git
14:12 aleskandro cbreak thanks it worked
14:13 watabou joined #git
14:14 maxorator joined #git
14:14 elastix1 joined #git
14:15 SoRoot joined #git
14:15 zincrokx joined #git
14:15 jaguarmagenta joined #git
14:17 Douhan joined #git
14:18 Ogh joined #git
14:18 mohabaks joined #git
14:20 markovh joined #git
14:21 barteks2x joined #git
14:22 jost_ joined #git
14:24 Tomdarkness joined #git
14:24 Gsham joined #git
14:24 Tiwt joined #git
14:25 fluxdude joined #git
14:25 fluxdude left #git
14:27 phanimah1sh joined #git
14:29 eletuchy joined #git
14:30 kpease joined #git
14:31 fracting joined #git
14:31 Zdob joined #git
14:31 mrkat joined #git
14:36 XDS2010 joined #git
14:37 FroMaster joined #git
14:37 ncthom91 joined #git
14:40 zincrokx joined #git
14:41 sitaram_ j416: of course!  And on "sitaram" too
14:42 sitaram_ j416: many many many years ago I used to be more active in the "git" questions but now I lack the time.  However, I'm on the channel and I think I must at least answer or respond to gitolite specific stuff when it comes up :)
14:42 zivester joined #git
14:43 perlpilot sitaram_: When are you going to release a newer, more featureful version of gitolite?   ;)
14:44 hobodave joined #git
14:46 Memto joined #git
14:46 ash_workz joined #git
14:47 bjpenn joined #git
14:49 Psych joined #git
14:50 edude03 joined #git
14:50 garethdaine joined #git
14:50 tim joined #git
14:51 toogley joined #git
14:51 j416 sitaram_: :) nice of you.
14:51 interne7y joined #git
14:53 prooftechnique joined #git
14:54 tim i pulled by accident, and then tried to undo it by running git reset @HEAD{1} but i don't think it did that
14:54 osse tim: it's written HEAD@{1}
14:54 tim oh, my bad
14:54 tim it's that way in my console's history
14:54 prooftechnique left #git
14:55 tim just misread it
14:55 robotroll joined #git
14:55 tim git status says that i have files that are modified but not commited after the reset
14:55 osse you should have used --reset
14:56 osse but it's fine. just follow git status's advice for discarding the changes
14:57 tim sorry, the story still continues: i did some more reset commands, thinking that i didn't choose the correct version
14:57 tim my branch and the origin/master have divergetd and i have no idea what that means
14:57 kballou joined #git
14:58 shgysk8zer0 joined #git
15:00 tim i can still see the line in reflog that i think is correct, so how do i get it that state
15:00 tim which command takes the --reset argument?
15:02 mdw joined #git
15:02 tim git reset --reset or git --reset or what
15:03 mdw_ joined #git
15:03 alex3rd joined #git
15:03 peepsalot i did a git rebase, and had some merge conflicts which i've attempted to resolve.  there are no longer any <<<<<, ======, >>>>>> line in the files.  when i run git diff, sometimes the + sign at the begining of a line has a space before it, is this still indicating "mine" vs "theirs" via that space?
15:04 kilo_bravo joined #git
15:04 bluezone joined #git
15:05 Asho joined #git
15:05 tim peepsalot: maybe it's just the console window showing long lines wierdly?
15:06 peepsalot it looks like the space is all on the changes im merging from remote, and my changes are not spaced
15:07 tim i don't really know the exact format the diff uses
15:07 tim so sorry
15:07 tim but i don't recall seeing a space before the +
15:08 marc_v92 joined #git
15:09 bango2 joined #git
15:09 tim any help on how to do that reset correctly?
15:09 mdw joined #git
15:09 mjfua joined #git
15:09 digidog joined #git
15:10 ojacobson peepsalot: workspace diffs show first parent and second parent as two columns of +-
15:10 ojacobson tim: assuming the work tree was clean when you started, 'git reset --hard HEAD@{1}' would have done it. From here, follow the directions git-status gives you for unstaging changes, or use 'git reset --hard' to obliterate all uncommitted changes to tracked files.
15:11 ojacobson (The changes are irrevocably gone, in either case)
15:11 ochorocho__ joined #git
15:12 mdw__ joined #git
15:12 Vinnie_win joined #git
15:13 timini joined #git
15:13 LionsMane joined #git
15:13 timini Hey I have some modifications and I want to pop my stash
15:14 timini when i try to pop stash i get the error
15:14 timini Please, commit your changes or stash them before you can merge.
15:14 frem_ joined #git
15:14 timini anyway i can pop multiple stashes ontop of one another?
15:15 santa-fire joined #git
15:16 weox joined #git
15:16 tim ojacobson: actually, i'm not sure whether my tree was clean or not but i had already given up on not losing those changes
15:16 tim i think it was nothing valuable
15:16 nnyk joined #git
15:17 lb timini no
15:17 lb timini man git stash ->  The working directory must match the index.
15:17 gitinfo the git-stash manpage is available at http://jk.gs/git-stash.html
15:17 peepsalot ojacobson, ah, thanks for the explanation
15:17 blackwind_123 joined #git
15:18 griffindy joined #git
15:18 timini how can I just view a stash?
15:18 mdw joined #git
15:18 tim so i should do git reset --hard HEAD@{3} now i think, those previous reset's caused more lines in the reflog
15:19 tim which one is the one i should choose, the number with the good commit or the line with the mistaken pull?
15:19 tim does it revert to the stae before or after the operation, that is
15:19 tim *state
15:19 lb timini man git stash
15:19 gitinfo the git-stash manpage is available at http://jk.gs/git-stash.html
15:19 lb git stash show <stash>
15:19 braydenjw joined #git
15:20 Darcidride joined #git
15:20 ki0_ joined #git
15:21 braydenjw left #git
15:21 intellix joined #git
15:22 tyreld joined #git
15:24 eletuchy joined #git
15:24 tim hmm git status still claims that my branch and origin/master have diverged, is this normal
15:25 tinajohnson___ joined #git
15:25 marcogmonteiro joined #git
15:26 phanimah1sh timini: do a git commit, then pop/apply, then reset soft to HEAD~1
15:26 marc_v92 joined #git
15:26 phanimah1sh you'll have the stash applied along with your current changes.
15:28 nnyk_ joined #git
15:30 cmyers joined #git
15:31 cmyers does git gc only consider tags and refs/heads/* for reachability, or would it notice things in, e.g. refs/archive/* as well?
15:31 cmyers (google is not forthcoming with the answer and I'm hoping to avoid reading src)
15:31 romerocesar joined #git
15:32 phanimahesh What does refs/archive contain? Haven't come across it so far. Or I don't remember
15:33 cmyers nothing - it's a non-standard path
15:33 cmyers I am considering "archiving" branches to a non-standard place and wondering if it will cause git prune/gc to clean up stuff
15:34 LucaTM joined #git
15:34 cmyers so we know git prune must consider reachability before cleaning up objects, and it must consider at least the reflog and refs/heads/* - but it might do refs/* or something to be more complete
15:35 phanimahesh if it's nonstandard I guess theonly way to be sure is to find an official doc or read source or test.
15:35 cmyers many git servers use alternative paths to store things, like gitlab/bitbucket store pull requests in refs/pull-requests/id/{from,to}
15:35 cmyers ....or ask this channel full of git devs who might know off the top of their head. =)
15:35 phanimahesh Yep, even github stores in refs/pulls/:id
15:36 phanimahesh Ah, yep that. Let's wait for someone who knows to pitch in, then.
15:36 nilsi joined #git
15:37 XDS2010 joined #git
15:37 phanimahesh (someone please tag me when it's answered, I may be afk)
15:37 cmyers the git gc docs did not say, but in git prune docs it says "This runs git fsck --unreachable using all the refs available in refs/, optionally with additional set of objects specified on the command line, and prunes all unpacked objects unreachable from any of these head objects from the object database."
15:37 cmyers and my impression is that git gc just runs git prune
15:38 cmyers so that is fairly strong evidence it is refs/*, but I'd feel better hearing it from a dev or reading the src (which I am trying to do now)
15:39 sitaram_ perlpilot: features keep getting added, though the pace is much slower now.  What are you needing?  (Don't say "GUI" though!!)
15:39 zincrokx joined #git
15:41 bongjovi joined #git
15:41 phanimahesh cmyers: also man git-gc says git tries very hard to be safe, which implies it probably retains everything under refs/. Would be good to have it clearly spelled out though.
15:41 gitinfo cmyers: the git-gc manpage is available at http://jk.gs/git-gc.html
15:42 cmyers looks like reachable.c has the code for the traversal
15:42 elastix joined #git
15:42 madewokherd joined #git
15:42 cmyers and prune.c calls it on line 142 passing in "revs", which is initialized by "init_revisions"
15:43 cmyers init_revisions takes a prefix
15:43 dsdeiz joined #git
15:44 romerocesar joined #git
15:44 benwbooth joined #git
15:45 cmyers the prefix is passed in to cmd_prune but I don't see where it comes from
15:45 xlegoman joined #git
15:46 cmyers I am feeling pretty confident now that the default prefix is probably refs/
15:46 cmyers though my C fu is too weak to find how that happens
15:47 LaT0rtue joined #git
15:47 LeBlaaanc joined #git
15:47 phanimahesh ag refs? :P
15:47 sangy joined #git
15:47 benwbooth joined #git
15:47 phanimahesh Or grep -R refs . if you prefer grep
15:48 PHPanos_ joined #git
15:48 cmyers "refs/" appears about a million times =|
15:48 phanimahesh When I have no idea where to look, I start with a grep for probable strings.
15:48 phanimahesh Ah. :/
15:48 phanimahesh build ctags and navigate by callers?
15:48 cmyers there are some prune tests in t/ that might show how it works
15:50 dreiss joined #git
15:50 marcogmo_ joined #git
15:50 cmyers bleh, I should just test it by hand
15:54 gfortune left #git
15:56 durham joined #git
15:59 cmyers phanimahesh: confirmed - it uses refs/* for reachability: https://gist.github.com/terabyte/2e504041bae099cabe40fbd48a2d333e
15:59 jetole joined #git
15:59 grubles joined #git
15:59 morenoh149 joined #git
15:59 ochorocho__ joined #git
15:59 harrymoreno1 joined #git
15:59 harrymoreno1 left #git
15:59 phanimahesh cmyers: yayy! That's the sensible thing to do, anyway. PR git docs?
15:59 eletuchy joined #git
15:59 jetole Hey guys. I have a master branch. I created a dev branch of master and both master and dev have been pushed to remote. I have merged dev into master and it has been pushed to remote. I have now created a staging branch from master and have pushed it to remote. Right now all three branches are the same. I would like to have dev be a branch of staging instead of master on my working copy and my remote copy but I don't know how to do that
15:59 netwoodle joined #git
15:59 cmyers seeing as how "git prune" docs already say the right thing, I was just slightly distrustful of them, I'm not sure what to add.  I guess I could replicate the sentiment to git gc docs as well.
15:59 phanimahesh jetole: There is no "branch of another branch". do you want to update the tracking branch?
15:59 phanimahesh the one shown by git status?
16:00 NeverDie_ joined #git
16:00 jetole phanimahesh: I am not sure what you mean. Can I merge dev into staging now even though dev and staging were both branched when master was the working branch
16:00 jetole ?
16:01 [Brain] joined #git
16:01 phanimahesh yep. You can.
16:01 [Brain] joined #git
16:02 kadoban joined #git
16:02 sauvin_ joined #git
16:03 phanimahesh jetole: Did you ever run git log --all --graph ? It'll help you understand. The whole of your repo is just a single graph.
16:03 fracting joined #git
16:03 jetole I will now
16:04 phanimahesh Any two branches can be merged so long as they have some common ancestor. (Not exactly true, you can technically also merge detached ones, but ignore it for now)
16:04 phanimahesh If you are running it I prefer a prettified version.
16:04 phanimahesh git log --all --graph --pretty=format:'%Cred%h%Creset -%C(auto)%d%Creset %s %Cgreen(%cr) %C(bold blue)<%an>%Creset' --abbrev-commit --date=relative
16:05 xlegoman joined #git
16:05 phanimahesh set it as an alias if you like it, that's what I have currently.
16:05 jstimm joined #git
16:06 jetole so it looks like HEAD is pointing to all branches. Does this mean that if I make changes to dev than I can checkout staging and merge dev in and than I can checkout master and merge in staging?
16:06 phanimahesh yep.
16:07 jetole Thank you, phanimahesh
16:07 phanimahesh HEAD points to a commit. each of your branches points to a commit
16:07 phanimahesh currently they happen to all point to the same commit.
16:07 jetole Great
16:07 phanimahesh when you experiment, look at the graph once in a while till you develop an intuition regarding what happens under the hood
16:08 jetole OK. I will
16:08 jetole Thank you
16:08 phanimahesh You're welcome. :)
16:09 f3r70rr35f joined #git
16:10 synthroid joined #git
16:13 Topic for #git is now Welcome to #git, the place for git help and comedy salvation | Public logs at http://goo.gl/BuUi5o | Current stable version: 2.8.3 | First visit? Read: http://jk.gs/git | Getting "cannot send to channel"? /msg gitinfo .voice | git pull a day keeps the conflict's away
16:13 edude03 joined #git
16:14 Lorcatar joined #git
16:15 Wtau joined #git
16:15 rkazak joined #git
16:15 watabou joined #git
16:15 soLucien joined #git
16:16 soLucien hello guys ! Maybe this is off-topic,  but i have a question about github
16:16 perlpilot sitaram_: I don't really need anything, I was just seeing if you had an answer to that gitolite question. :-)  I've been happily using gitolite for a long time now and it's done everything I've needed it to, thanks.
16:16 perlpilot soLucien: try #github ?
16:16 jaguarmagenta joined #git
16:16 perlpilot soLucien: Also, github has quite a bit of help on their website.
16:17 snfgf joined #git
16:18 tomog999 joined #git
16:19 soLucien no response
16:19 Eugene !github
16:19 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)
16:20 Eugene The worst we can do is not answer ;-)
16:20 imachuchu joined #git
16:20 * kadoban preps the knives and salt
16:20 soLucien exactly, here it seems like people are a bit more active
16:20 ftu joined #git
16:21 allcentury joined #git
16:21 soLucien http://pasteboard.co/1yDN18t0.png
16:22 soLucien creating a user with the org's name seems like the thing i would do , but it also means there isn't soemthing better
16:22 soLucien as a feature they offer
16:22 LeBlaaanc joined #git
16:23 kfunk joined #git
16:24 Eugene soLucien - github has an "Organizations" feature. Why not use that?
16:24 soLucien seems like you have to pay for it
16:24 Eugene Only if you want private repos under it
16:25 osse soLucien: https://github.com/organizations/new
16:25 soLucien oh, that's great
16:25 soLucien cool , i was under the impression organization =  you pay for it
16:26 soLucien thanks for the clarification
16:26 Eugene !next
16:26 gitinfo Another satisfied customer. NEXT!
16:26 soLucien ok , another one
16:26 Urpiz joined #git
16:27 soLucien i have a git repo that i host in my private TFS server , where i have set up CI and all that stuff
16:27 soLucien i have things i want to share in the repo, and things i don't want to
16:27 ahmedelgabri joined #git
16:28 soLucien what should be the strategy for me to open source it
16:28 Eugene Private things like keys?
16:28 soLucien no , private things like the test environment
16:28 bjpenn joined #git
16:28 soLucien build devinitions
16:28 osse !sensitive
16:28 gitinfo [!filter_sensitive] You can use filter-branch to remove sensitive data from a repository's history. https://help.github.com/articles/remove-sensitive-data/
16:28 osse !config
16:28 gitinfo [!configfiles] It is recommended to store local configuration data in a file which is not tracked by git, but certain deployment scenarios(such as Heroku) may require otherwise. See https://gist.github.com/1423106 for some ideas
16:29 daslicht joined #git
16:29 Eugene You'll need to remove those from the history so that they're not in the repo at all. You can do this with filtering(see above), or by maintaining a separate "public" repo, starting by copying only the things you want into a new history
16:29 soLucien what should be the strategy for managing the OSS contributions from github, as well
16:29 soLucien yes , the public thing was first thing i came up with
16:30 Eugene I find it easiest to maintain my private things in a configfiles-style manner, IE untracked-in-git
16:30 soLucien but then i have to manage 2 things
16:30 Eugene (never commit them to begin with)
16:30 soLucien they are not private, but rather company-specific .. they are supposed to be in TFS
16:30 soLucien but not on github
16:31 soLucien so set up a new repo
16:31 soLucien for the public . cool
16:32 xorox90 joined #git
16:33 tcsc joined #git
16:35 diegoaguilar joined #git
16:37 XDS2010 joined #git
16:37 rgrinberg joined #git
16:37 crose joined #git
16:40 dmto joined #git
16:40 Mobutils joined #git
16:40 Tim-SEM joined #git
16:41 Tim-SEM joined #git
16:43 pungi-man joined #git
16:45 dhollinger joined #git
16:46 sitaram_ perlpilot: :)
16:47 piling joined #git
16:47 Gitzilla joined #git
16:48 davisonio joined #git
16:50 sebbu joined #git
16:50 sebbu hi
16:50 sebbu i wanted to clone a repository
16:50 sebbu but
16:50 sebbu fatal: write error (File too large)
16:50 sebbu i got that error, and i'm in a 5 character folder at the root of an ext4 partition
16:52 interne7y joined #git
16:53 mikecmpbll joined #git
16:53 johntramp joined #git
16:53 santa-fire sebbu: how did you clone? in command line or gui?
16:54 Saxon joined #git
16:54 fscala joined #git
16:55 qpdb joined #git
16:55 hahuang61 joined #git
16:58 svankmajer joined #git
17:01 freimatz joined #git
17:03 rbr joined #git
17:03 Darcidride_ joined #git
17:03 rominronin joined #git
17:03 bket joined #git
17:04 jimi_ joined #git
17:04 djb-irc joined #git
17:04 pwnawannab[NE] joined #git
17:05 PaulCapestany joined #git
17:06 ki0 joined #git
17:06 SwiftMatt joined #git
17:06 ki0 joined #git
17:07 yqt joined #git
17:08 saml joined #git
17:08 saml i made commits to master instead of creating new branch. what should i do?
17:09 watabou joined #git
17:09 osse saml: https://raw.githubusercontent.com/Osse/git-stuff/master/git-reset-explanation.txt
17:09 saml git checkout origin/master;  git checkout -b feature; git merge master;  and move master back to origin/master ?
17:09 hahuang61 joined #git
17:10 saml ah simpler. thanks osse
17:10 elsevero joined #git
17:11 GavinMagnus joined #git
17:11 xorox90 joined #git
17:13 jetole joined #git
17:13 jimi_ joined #git
17:14 TheVoid joined #git
17:14 mkoskar joined #git
17:16 chris2 joined #git
17:16 italoacasas joined #git
17:18 GavinMagnus joined #git
17:18 GavinMagnus left #git
17:18 diegoviola joined #git
17:19 phanimahesh osse: Do you have a collection of stock responses written up?
17:19 osse phanimahesh: not exactly
17:19 osse phanimahesh: check out that repo i linked to
17:21 eyp joined #git
17:21 thiago joined #git
17:22 phanimahesh osse: cool. Gave it a star. :)
17:23 cdg joined #git
17:23 osse and I have a lot of small variations that I haven't committed
17:23 osse it's starting to become a mess
17:23 allcentury joined #git
17:24 osse phanimahesh: patches welcome, btw
17:24 Grimm_ joined #git
17:26 domgetter joined #git
17:26 weylin How do I make my git e-mail account private in logs for commits and such?
17:26 osse weylin: the only way is to change it
17:26 phanimahesh weylin: Use a fake email.
17:27 BoomZ joined #git
17:27 osse i recommend 'gentleman69@mailinator.com'
17:27 weylin I can't retro on commits, I assume?
17:27 phanimahesh The email is part of the commit metadata. Changing the email will change the commit id.
17:27 osse weylin: you can rewrite history
17:27 BoomZ Hello
17:27 gitinfo BoomZ: 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.
17:28 phanimahesh osse: if you are okay with it, you can do a !filter-branch
17:28 weylin Alright thank you guys.
17:28 osse weylin: man git-filter-branch has an example
17:28 gitinfo weylin: the git-filter-branch manpage is available at http://jk.gs/git-filter-branch.html
17:28 phanimahesh oops meant to tag weylin :P
17:28 osse --env-filter is your friend
17:28 BoomZ I would like to create a git frontend client, is there any resoucres availible to get started?
17:29 weylin I'll read up on it. Appreciate the help. :)
17:29 M-ou-se joined #git
17:30 lukeshu` joined #git
17:30 clemf joined #git
17:31 vude joined #git
17:31 CEnnis91 joined #git
17:33 sebbu santa-fire, command line git clone git@github.com:user/repo.git
17:34 ribasushi joined #git
17:34 ignas joined #git
17:35 TomyLobo joined #git
17:37 Chill_Surf joined #git
17:37 tterrag joined #git
17:37 XDS2010 joined #git
17:38 xorox90 joined #git
17:40 bnnb joined #git
17:42 mikecmpbll joined #git
17:43 dreiss joined #git
17:44 davisonio joined #git
17:45 ramsub07 joined #git
17:45 espindola joined #git
17:46 Dyonisos joined #git
17:47 mdw joined #git
17:48 choki joined #git
17:48 al-damiri joined #git
17:48 rueben_ joined #git
17:52 paul424 joined #git
17:52 cdg joined #git
17:53 nidr0x joined #git
17:54 hahuang61 joined #git
17:54 drodger joined #git
17:57 svankmajer joined #git
17:57 svankmajer joined #git
17:59 cstrahan joined #git
17:59 JohnLea___ joined #git
17:59 djb-irc joined #git
17:59 noahsussman joined #git
18:00 Sasazuka joined #git
18:00 ygeLN joined #git
18:00 CheckDavid joined #git
18:00 realz joined #git
18:00 kline joined #git
18:01 CEnnis91 joined #git
18:01 al-damiri joined #git
18:02 shinnya joined #git
18:03 nezZario joined #git
18:03 cspotcode joined #git
18:03 durham joined #git
18:04 edude03 joined #git
18:04 rominronin joined #git
18:04 epic joined #git
18:05 piling joined #git
18:05 toogley joined #git
18:06 hobodave_ joined #git
18:06 cstrahan joined #git
18:06 fahadash joined #git
18:06 ribasushi joined #git
18:07 Sleepy63 joined #git
18:09 t-mart__ joined #git
18:09 xfbs joined #git
18:10 Soliton joined #git
18:10 MLM joined #git
18:10 ToeSnacks joined #git
18:10 paperziggurat joined #git
18:11 t-mart__ joined #git
18:11 noidea joined #git
18:11 aendrew joined #git
18:12 ramsub07 joined #git
18:13 discopatrick joined #git
18:15 beckyconning_ joined #git
18:16 davisonio joined #git
18:16 jpb joined #git
18:16 Impaloo93 joined #git
18:17 jaguarmagenta joined #git
18:17 Schwarzbaer_ joined #git
18:17 ToeSnacks joined #git
18:18 t-mart joined #git
18:20 blackwind_123 joined #git
18:21 sadmac joined #git
18:21 leeN joined #git
18:22 Impaloo31 joined #git
18:22 Crash1hd_ joined #git
18:22 Crash1hd joined #git
18:23 swilsonau joined #git
18:23 discopatrick joined #git
18:23 Crash1hd hey all so I had updated my gitignore and removed a bunch of stuff from the git ignore yet after running git rm -r --cached it still doesnt add the items when I goto add them
18:24 dcg joined #git
18:24 Crash1hd yet when I run git check-ignore it does not show that the file is being ignored?
18:24 allcentury joined #git
18:24 al-damiri joined #git
18:25 [Brain] joined #git
18:25 mischat joined #git
18:26 benwbooth joined #git
18:26 timini joined #git
18:27 Aerosonic joined #git
18:27 jbitdrop joined #git
18:27 nevodka joined #git
18:28 TinkerTyper joined #git
18:28 xorox90 joined #git
18:30 nevodka left #git
18:31 al-damiri joined #git
18:32 ajf- joined #git
18:32 huhlig joined #git
18:32 dangerousdave joined #git
18:33 SwiftMatt joined #git
18:37 davisonio joined #git
18:37 msonntag joined #git
18:37 dangerousdave joined #git
18:37 iamdevnul joined #git
18:38 XDS2010 joined #git
18:38 T4nk5bb0t joined #git
18:39 dangerousdave joined #git
18:40 dangerousdave joined #git
18:41 Crash1hd hey all so I had updated my gitignore and removed a bunch of stuff from the git ignore yet after running git rm -r --cached it still doesnt add the items when I goto add them
18:41 Crash1hd yet when I run git check-ignore it does not show that the file is being ignored?
18:41 puff joined #git
18:42 badloop joined #git
18:42 badloop left #git
18:42 multi_io joined #git
18:43 t-mart_ joined #git
18:43 Balliad joined #git
18:44 UTAN_dev joined #git
18:45 cdg joined #git
18:45 dangerousdave joined #git
18:45 t-mart joined #git
18:46 rgrinberg joined #git
18:46 iamdevnul joined #git
18:47 ynanm joined #git
18:47 snowinferno joined #git
18:49 alex1a joined #git
18:49 djb-irc joined #git
18:49 spr0tsuki joined #git
18:49 ShekharReddy joined #git
18:50 harleypig joined #git
18:51 Ahmed90 joined #git
18:51 xorox90 joined #git
18:51 ShekharReddy left #git
18:52 ribasushi joined #git
18:52 k-man joined #git
18:52 shauna joined #git
18:53 interne7y joined #git
18:53 cmosguy joined #git
18:53 parliament joined #git
18:54 guampa joined #git
18:55 m0viefreak joined #git
18:55 barq joined #git
18:56 chachasmooth joined #git
18:56 ktosiek joined #git
19:00 dsdeiz joined #git
19:00 paul424 joined #git
19:00 Waskeend joined #git
19:00 edude03 joined #git
19:01 nwkj86 joined #git
19:01 ec joined #git
19:02 Guest81466 joined #git
19:02 espindola joined #git
19:02 Soar joined #git
19:03 jstimm joined #git
19:03 ec joined #git
19:03 cdown joined #git
19:03 rominronin joined #git
19:04 kallisti5 joined #git
19:05 barq joined #git
19:06 cdown_ joined #git
19:06 Dazzyp joined #git
19:09 hahuang61 joined #git
19:09 k3rn31 joined #git
19:10 tinanoo joined #git
19:11 diegoviola joined #git
19:11 yqt joined #git
19:11 ahmedelgabri joined #git
19:11 ocbtec joined #git
19:12 ResidentBiscuit joined #git
19:12 watabou joined #git
19:12 benwbooth joined #git
19:13 LBV_User joined #git
19:13 ResidentBiscuit So how does `rebase --onto` work? If you have branch B which is based on branch A, which is based on master, and you move everything in Branch B not in branch A onto master, wouldn't you be missing work? Since you based your work on commits that do not exist in master
19:17 Orion3k joined #git
19:18 ktosiek joined #git
19:18 dan2k3k4 joined #git
19:18 espindola joined #git
19:18 vals_ joined #git
19:20 espindola joined #git
19:20 shgysk8zer0 joined #git
19:21 tango_ joined #git
19:21 Gsham joined #git
19:24 asimzaidi joined #git
19:25 allcentury joined #git
19:25 bgyl joined #git
19:25 diogenese joined #git
19:25 sangy joined #git
19:26 GavinMagnus joined #git
19:27 barteks2x joined #git
19:27 perlpilot joined #git
19:28 SteffanW joined #git
19:30 fission6 joined #git
19:30 Seveas when rebasing, git only cares about transplanting patches from one place to another. Whether the rebase makes semantic sense is not git's task to judge :)
19:31 nwkj86 joined #git
19:31 philipp64|work this should be simple but I can't figure out a way to do it unambiguously...  I have a repo R on a server, and I've forked it as F.  I then clone it locally as C.
19:31 philipp64|work Periodically I want to pull --rebase from R... but I want a script which figures out what 'R' is.
19:32 ki0_ joined #git
19:32 Seveas what does 'forked as F' mean in this scenario? And what did you clone, R or F?
19:32 philipp64|work I've made a private copy of R on the server.
19:32 philipp64|work and locally I cloned F.
19:33 Seveas then git has no way to automatically determine R, but you could add it as remote to fetch from
19:33 philipp64|work If I got onto the server (which I can't) and looked at which remote F pulls from, it would point at R...
19:33 Seveas git remote add upstream url-of-R
19:33 Seveas then you can do git pull --rebase upstream/somebranch
19:34 philipp64|work right, but I need to know which branch of R I'm working in, especially if I've done a checkout -b in my cloning.
19:34 rgrinberg joined #git
19:34 Seveas git won't know.
19:34 philipp64|work why is that?
19:35 Seveas because it doesn't have that information, you didn't tell it that :)
19:35 philipp64|work I tried using "branch --contains", but I often get multiple hits which I don't understand.
19:35 isy joined #git
19:35 Seveas that means a commit is reachable from multiple heads, which is perfectly normal
19:36 nnyk_ joined #git
19:36 dan2k3k4 joined #git
19:36 mohabaks joined #git
19:37 durham joined #git
19:38 spudowiar gitinfo
19:38 XDS2010 joined #git
19:38 spudowiar !help
19:38 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.
19:41 Gsham joined #git
19:44 kre10s_ joined #git
19:44 kre10s_ Is it possible to inclease the depth of a shallow clone?
19:46 slixpk joined #git
19:46 osse kre10s_: git fetch --depth=newdepth
19:50 irco joined #git
19:50 Crash1hd how come bin ignores folders that have a capital B? like Bin
19:51 Roha joined #git
19:53 bragr joined #git
19:54 LeBlaaanc joined #git
19:55 dan2k3k4 joined #git
19:56 philipp64|work joined #git
19:58 bragr Is there a way to disable SSH password authentication in git and only rely on SSH keys? I am writing scripts that use git and I'd like git to abort if it can't auth with keys because the password prompts are making my scripts hang. I've googled around but I'm getting buried in results about setting up keys, which isn't my issue.
19:59 ajf- joined #git
19:59 madewokherd probably something to configure in ssh, not git
20:00 cdg joined #git
20:01 Crash1hd Hey All is there a way to have git ignore just bin folders and not Bin folders (Notice the casing) :)
20:01 madewokherd PasswordAuthentication setting in 'man ssh_config'
20:01 nitin joined #git
20:01 bragr Right I could do that, but this is supposed to be a utility script for other users to help with their workflow with a variety of git services so I don't control the git servers.
20:02 mdw joined #git
20:02 ojacobson Crash1hd: add bin/ to .gitignore and use a case-sensitive filesystem
20:02 Crash1hd ojacobson, crap stuck in windows
20:02 Crash1hd and it wont work that way above
20:02 ojacobson Then they are the same folder
20:02 madewokherd bragr: oh, I was hoping that option would work on the client, and that there's a way to set ssh options temporarily through your script
20:02 ojacobson either its contents are ignored, or they are not
20:02 Crash1hd yeah dang alright thanks :)
20:03 madewokherd bragr: git config manpage implies there's an SSH_ASKPASS environment variable, maybe you can set that to false?
20:04 A124 joined #git
20:04 rominronin joined #git
20:04 madewokherd nope, that doesn't help
20:06 Crash1hd god I wish I could get away from windows lol
20:06 elastix joined #git
20:08 bragr madewokherd: Thanks, you've sent me down a good path. It looks like new version of git have an environment variable GIT_SSH_COMMAND which lets you set custom ssh commands, and then you can use the SSH BatchMode option to cause it to fail if key auth fails. But it looks like this var is very new so it isn't supported on most distros yet :/
20:09 user083 joined #git
20:09 bragr Basically I've swapped one "Fix your system issue" with another haha
20:09 ramsub07 joined #git
20:09 lohq joined #git
20:12 fission6 is there some generic form of branch clean up, i have a ton locally and some upstream
20:12 madewokherd make a temporary ssh binary and add it to PATH? :/
20:12 fission6 generally speaking i want to delete all locally and upstream that aren't master or have pull requests
20:12 drodger joined #git
20:14 dsdeiz joined #git
20:16 axiom_1 joined #git
20:17 dopesong joined #git
20:18 jaguarmagenta joined #git
20:18 texasmynsted joined #git
20:20 hahuang61 joined #git
20:22 jaime joined #git
20:24 jaime left #git
20:24 ylez joined #git
20:25 allcentury joined #git
20:26 drodger joined #git
20:27 edude03 joined #git
20:30 nivag joined #git
20:31 Limes_ joined #git
20:32 cdhc joined #git
20:35 kisisten joined #git
20:35 freimatz3 joined #git
20:37 sesquipedalian joined #git
20:38 dtqoo joined #git
20:38 radsy joined #git
20:38 nnyk_ joined #git
20:38 nnyk_ joined #git
20:39 XDS2010 joined #git
20:40 ojdo joined #git
20:40 nitin joined #git
20:41 tremblerz joined #git
20:42 shgysk8zer0 joined #git
20:43 tremblerz If I have made some changes in a branch and I haven't commited them, now I want to change my branch but want that uncommited code to stay in that branch itself
20:43 tremblerz how do I do that ?
20:43 ktosiek joined #git
20:44 madewokherd I usually just commit with a message of WIP
20:44 ktosiek joined #git
20:45 madewokherd then when I switch back to that branch and am ready to commit, I amend the WIP commit and give it a real message
20:45 tremblerz In my case I won't be needing that commit for final changes
20:48 tremblerz any other solution madewokherd ?
20:48 kisisten hi, i'm trying to clone a apache repo with git url but I'm getting the master version. I would like to get a older version of the code. How can I do that?
20:49 texasmynsted what is a good way to enforce moving changes from a local feature branch to a public dev branch as a single squashed/consolidated commit?
20:49 kre10s_ kisisten, checkout the relevant branch or tag.
20:49 sesquipedalian joined #git
20:49 diogenese joined #git
20:50 inflames joined #git
20:51 madewokherd no, sorry, I don't think what you're asking is possible and I don't have a good alternative
20:51 nwkj86 joined #git
20:51 tremblerz madewokherd: can git stash help ?
20:51 MadLamb joined #git
20:52 kisisten kre10s_: how do I get the tag 15.3.3? I browse this url > https://git-wip-us.apache.org/repos/asf?p=activemq.git
20:52 dsdeiz joined #git
20:52 dsdeiz joined #git
20:52 madewokherd yes, but I don't know a way around manually running it when you switch branches
20:54 dmitrij_1003 joined #git
20:54 kisisten kre10s_: I got it "git tag -l" thanks
20:54 interne7y joined #git
20:55 kre10s_ so something like git checkout activemq-15.3.3 ? tag -l only lists them
20:56 johntramp joined #git
20:57 NeverDie joined #git
20:57 kisisten kre10s_: if I run that command I get "fatal: '-l' cannot be used with updating paths"
20:58 kisisten never mind, slow day
20:58 kre10s_ which command exactly?
20:58 choki joined #git
20:59 johnmilton joined #git
20:59 diegoviola joined #git
21:00 shgysk8zer0 joined #git
21:00 jeffreylevesque joined #git
21:01 Pyrogerg joined #git
21:04 netj joined #git
21:04 ramsub07 joined #git
21:05 rominronin joined #git
21:08 rueben_ joined #git
21:09 Flush joined #git
21:09 kre10s_ I was sent a patch via email. It won't apply and it looks like it's just due to the whitespace... tabs and stuff. How are you guys applying patches out of email?
21:11 Biov joined #git
21:11 hashpuppy joined #git
21:11 eletuchy joined #git
21:13 eletuchy_ joined #git
21:13 tremblerz left #git
21:13 ahmedelgabri joined #git
21:15 rgrinberg joined #git
21:16 afutr joined #git
21:16 eletuchy joined #git
21:16 johnmilton joined #git
21:17 papna joined #git
21:17 edude03 joined #git
21:18 papna Is there a nice way to add a   fetch =  line to a remote programatically?
21:18 eletuch__ joined #git
21:19 ahr3n joined #git
21:19 osse papna: git config --add remote.rofl.fetch 'omg:lol'
21:20 eletuchy_ joined #git
21:21 neachdainn joined #git
21:21 uucu4 joined #git
21:22 neachdainn I have a branch that has a couple of merges from Master. If I rebase that branch, will the resulting branch still be merged from Master or will I have duplicate commits?
21:22 neachdainn Rebase the branch on itself, that is.
21:23 DrSlony joined #git
21:24 eletuchy joined #git
21:24 DrSlony Hello, how do I "git revert" the last two commits?
21:24 osse git revert HEAD HEAD~1
21:26 AaronMT_ joined #git
21:26 eletuch__ joined #git
21:26 allcentury joined #git
21:28 eletuchy_ joined #git
21:28 edude03 joined #git
21:28 dmitrij_2691 joined #git
21:28 dan2k3k4 joined #git
21:28 nnyk_ joined #git
21:29 bruce_lee joined #git
21:30 github631 joined #git
21:30 nitin joined #git
21:31 eletuchy joined #git
21:33 eletuch__ joined #git
21:36 eletuchy joined #git
21:37 svankmajer joined #git
21:37 svankmajer joined #git
21:39 XDS2010 joined #git
21:39 aidalgol joined #git
21:40 eletuchy_ joined #git
21:40 bittyx joined #git
21:41 rgrinberg joined #git
21:42 aapaaal joined #git
21:43 eletuchy joined #git
21:43 paul424 joined #git
21:44 LeBlaaanc joined #git
21:45 neachdainn Anyone know?
21:46 eletuchy_ joined #git
21:46 osse neachdainn: what do you mean by on itself?
21:46 osse you mean git rebase HEAD~10 or something like that?
21:47 neachdainn Yeah. I have two commits that need to be removed from history, so I'm rebasing the branch onto an earlier commit
21:48 osse neachdainn: the merges will be removed by default. but it sounds like git rebase -p can do the trick
21:48 osse -p tries to preserve merges
21:48 neachdainn Awesome. Thank you
21:49 papna osse: Thanks
21:57 DrSlony Hello, while doing a merge I did "git stash" to check some code on another branch, then I went back to the previous branch, did "git stash pop" and committed the finished merge. Turns out that git no longer saw this as a merge but just as a plain commit, neither did github, and the commit caused problems. I had to revert and re-merge without stashing halfway inbetween.
21:57 DrSlony git 2.7.3-r1
21:57 DrSlony left #git
21:57 King_Hual joined #git
21:57 MacWinner sorry for silly question.. but if I have a bunch of files in my workding master directory that are not committed and not checked in, and I want to move them over to a new branch, is there a simple way to do this?
21:58 osse MacWinner: there is only one way
21:58 aidalgol What does it mean to preserve a merge?
21:58 vishal MacWinner: if they are new files that haven't even been added to the index, you can simply checkout any branch, and git wont touch them
21:59 osse MacWinner: create and checkout a new branch. add files. commit
21:59 vishal aidalgol: context?
21:59 aidalgol vishal: above
21:59 aidalgol osse said git rebase -p tries to preserve merges
22:00 osse aidalgol: it means create a new merge commit that looks like the old one
22:00 vishal ah normally I suppose rebase -i will flatten merges
22:00 aidalgol ah
22:01 hekto joined #git
22:01 bsanford joined #git
22:03 GT joined #git
22:05 Noldorin joined #git
22:06 swati_27 joined #git
22:06 newbsie joined #git
22:06 rominronin joined #git
22:07 xbob joined #git
22:07 newbsie If I do "git reset HEAD^ --hard
22:07 newbsie Will that remove the last commit?
22:07 fstd joined #git
22:08 osse yes
22:08 swati_27 Hello. I made some changes in a file on my system ( on master branch). I want those to be discarded i.e. the file should be same as the file on remote repo. Could this be done?
22:08 i7c And discard all your changes in the working tree.
22:08 newbsie ossebecause when I check "git log --graph --oneline --all"
22:08 rbern joined #git
22:08 newbsie osse: sorry, my fingers aren't complying. So if I do "git log --graph --oneline --all", the commit still shows up?
22:09 osse newbsie: add --decorate
22:09 osse does it say anything next to it?
22:10 newbsie osse: it says "(origin/master, origin/HEAD) "
22:10 osse there you go
22:10 osse you reset master. but origin/master still has it
22:10 osse push --force if you must
22:10 newbsie osse: gotcha, so I need to push
22:10 osse newbsie: git config --global log.decorate true
22:10 osse bam!
22:11 swati_27 Hello. I made some changes in a file on my system ( on master branch). I want those to be discarded i.e. the file should be same as the file on remote repo. How could this be done?
22:11 osse git status tells you how
22:11 newbsie osse: but how did you figure out (origin/master, origin/HEAD) means it is on remote, but not local?
22:11 osse newbsie: because it says origin
22:11 osse !remote_tr > newbsie
22:11 gitinfo [!remote_tracking_branch] 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
22:12 newbsie So what does (HEAD -> master), mean?
22:12 osse !HEAD
22:12 gitinfo HEAD is a 'pointer' to the currently checked out branch (or commit, if HEAD is !detached). In bare repositories it tells clients which branch to checkout initially after cloning. Unlike commonly believed, HEAD is *not* something that exists separately for every branch. It also is *not* necessarily the newest commit in the repo (that's hard to define in a DVCS, anyway...)
22:12 wxg joined #git
22:13 newbsie I see, so (HEAD -> master) is where my branch is at, whereas (origin/master, origin/HEAD) is where the remote is
22:14 edude03 joined #git
22:14 osse HEAD in a remote repo as a different meaning: it indicates what branch you get by default when you clone
22:15 madewokherd joined #git
22:15 watabou joined #git
22:15 newbsie osse: so what shows up in decorate is remote?
22:16 osse newbsie: all sorts of things can show up there
22:16 osse in your case: yes
22:16 nnyk_ joined #git
22:16 osse inside (...) will be whatever branches and tags that point to that commit
22:19 jaguarmagenta joined #git
22:19 swati_27 left #git
22:19 newbsie osse: So HEAD -> master means, local HEAD points to master branch,  origin/master means remote master branch, and origin/HEAD means remote HEAD points to this commit
22:19 ahmedelgabri joined #git
22:19 osse yup
22:20 zh1 joined #git
22:21 newbsie osse: thanks.... I got a lot of learning to do, and this is not intuitive at all. The idea is fine, but the way it is presented seems confusing to me and maybe there is a good reason for that, but I havne't quite seen it yet.
22:21 eletuchy joined #git
22:22 italoacasas joined #git
22:22 ochorocho__ joined #git
22:22 osse my advice is to understand how it works
22:23 osse not focus on remembering long command lines. focus on understanding the principles
22:23 osse after that everything becomes easier
22:23 eletuch__ joined #git
22:23 cmyers I've basically never met someone who was even 1/10th competent at git that didn't understand the data model.  the way it is designed, the implementation "leaks out" and is really important to understand.
22:24 cmyers it is simultaneously it's greatest strenght and it's greatest weakness
22:24 newbsie osse: I think I have a decent grasp of it, but the issue is little things. Like the syntax "HEAD -> master" is different from "origin/HEAD", why isn't it just "HEAD" assuming if it doesn't specify, it is local, or heck just "local/HEAD"
22:25 osse newbsie: because it could just as well been HEAD -> mybranch or HEAD -> bleh
22:25 osse you can have as many branches as you want. one of them is named master
22:25 newbsie but then why isn't it "origin/HEAD -> ..."?
22:26 ahr3n joined #git
22:26 cmyers "origin" is a remote - origin/foo is telling you "the remote origin has a branch named foo"
22:26 eletuchy joined #git
22:26 osse newbsie: I don't have a good explanation for that.  Probably related to the fact that HEAD in a clone and HEAD in a remote mean different things
22:26 cmyers HEAD is a special kind of branch called a "symbolic branch" which can point at commits, like other branches *or* other branches themselves.
22:27 allcentury joined #git
22:27 cmyers origin/foo can only point at a commit, but origin/HEAD could point at a branch or a commit
22:27 cmyers HEAD is special in another way - for any repository, it represents "what is checked out" - so if HEAD points to a branch, that is the branch you have checked out.
22:28 cmyers the next time you run "git commit", the commit created will have the commit pointed to by HEAD as its parent
22:28 newbsie cmyers: so if a branch isn't specified with ->, then it points to the current commit line?
22:28 cmyers but you can be "in a detached head state" - where HEAD points straight at a commit, and there is no branch you have checked out
22:28 eletuchy_ joined #git
22:29 osse newbsie: do git checkout abc435 (pick some valid hash). do the log again
22:29 osse see that HEAD has moved
22:29 osse git checkout master to get back
22:29 cmyers yup, exaclty
22:29 osse !talk
22:29 gitinfo [!talks] Some good video talks about Git: [yt] http://goo.gl/z72s (Linus Torvalds: History&Concepts); [yt] http://goo.gl/R9H2q (Scott Chacon: Git basics, live examples); http://vimeo.com/35778382 (Randal Schwartz: Git basics, descriptional); http://vimeo.com/46010208 (Jesica Kerr: Git basics, descriptional)
22:30 cmyers detached heads are a perfect example of something a bit confusing until you understand branches, symbolic refs, and commits really well
22:30 osse newbsie: the last of the above talks explain this very easily with a whiteboard and some sticky notes
22:30 ki0 joined #git
22:30 osse I highly recommend watching it
22:31 newbsie Do you think we will git be evolved to something easier for beginners over time? Or is this it?
22:31 Dougie187 joined #git
22:31 osse it gets marginally easier all the time
22:31 newbsie osse: I will check out the video, got it noted down.
22:31 osse it was a lot worse 6-7 years ago
22:31 mkelly joined #git
22:32 eletuchy joined #git
22:32 griffindy joined #git
22:32 newbsie osse: I guess I'm privileged... I suppose if you use something worse then your view is, this is soooo easy.
22:33 * vishal remembers when he was new at git, deleting remote branches required you to use/understand the refspec. now one can just use --delete ... somewhere (I dunno I still just use the refspec :))
22:33 osse newbsie: I've used VCSes that are both even less intuitive *and* much much much much worse
22:33 osse and that one even cost a lot of money
22:33 MadLamb joined #git
22:34 eletuch__ joined #git
22:34 osse vishal: have you seen how --delete works under the hood?
22:34 vishal osse: nope
22:34 vishal osse: anything surprising?
22:35 osse vishal: it's so unsurprising it's almost surprising
22:35 osse vishal: https://github.com/git/git/blob/master/builtin/push.c#L91-L94
22:36 vishal osse: hah! of course :)
22:36 cmyers osse: you must be talking about visual source safe?
22:36 cmyers or perhaps clearcase?
22:37 eletuchy joined #git
22:37 cdown joined #git
22:37 SoRoot joined #git
22:37 osse cmyers: clearcase
22:37 SoRoot joined #git
22:37 osse with Unified Cancer Management on top
22:37 svankmajer joined #git
22:37 osse even the ones in charge had no idea how to use it properly
22:39 eletuchy_ joined #git
22:39 newbsie So why did git become more popular than mercurial? They seem pretty similar to me.
22:39 osse Never used VSS. I just now that some guy on SO said "An absolute nightmare of a system. *It made me wish we could go back to VSS!*"
22:39 osse *know
22:39 XDS2010 joined #git
22:39 cmyers newbsie: probably because of Linux's influence plus the appearance of github
22:39 osse newbsie: my guess is that the buzz around it was bigger, since it was written to replace the previous VCS for linx
22:39 cmyers a lot of companies and investment backed git
22:40 cmyers github gave everyone a free SCM server if they used git - which was pretty much unheard of at the time
22:41 cmyers *nix hacker types liked git better because it was more "unixy" in design philosophy, instead of one monolithic binary, it was a toolkit in the truest sense
22:41 cmyers mercural has made quite a resurgence recently though, with facebook's commitment to it
22:42 cbreak facebook commits to the dumbest things
22:42 cbreak like mercurial, or php...
22:42 cmyers yeah, I gotta agree with you on the php thing
22:42 newbsie cbreak: why is mercurial dumb?
22:42 cmyers but their reason for committing to mercurial - from their perspective at least - was the git folks basically told them to sod off and stop being dumb when they said they wanted 50GB repos
22:42 eletuchy joined #git
22:42 newbsie I think the PHP was just, because that is what they started with and so they are heavily invested.
22:44 cmyers there are strong "buildsystem related" reasons to have "everything in one big repo", google does it and facebook does it.  There are alternatives ways to accomplish the same guarantees across multiple repos (amazon does this for example) but it leads to a much more complex build system
22:44 cmyers I favor the amazon way, but it isn't for everyone.
22:44 eletuchy_ joined #git
22:45 durham cmyers: Mercurial also provides some very nice abstractions and extensibility points that make replacing parts of it with more scalable implementations easier
22:45 mase-tech joined #git
22:45 cmyers yeah, which is what FB has done to accomodate their needs
22:45 durham (disclosure: I work on the FB source control team)
22:46 knutstrut joined #git
22:46 cmyers durham: I was almost your teammate then =)
22:46 cmyers interviewed a few weeks back
22:46 eletuch__ joined #git
22:46 ELFrederich joined #git
22:47 durham I hope it was a good experience at least
22:47 osse gaah
22:47 osse suffering from clearcase flashbacks now
22:47 cmyers on the whole it was pretty positive.  I'm actually slightly unsure how I went wrong, but typically when that happens I assume some of my "crazy idealism" leaked out when it shouldn't have =)
22:47 osse I am angry
22:49 cmyers I think I ran into some litmus test questions too.  At least *now* I know what k-d trees are.
22:49 eletuchy joined #git
22:50 cbreak they asked what a kd tree is for source control? what kind of sources did they want to store, 3D meshes? :)
22:50 cmyers it wasn't specifically for source control, I think, it was just a general dev question.  it was like "implement the webservice to return the top 10 closest places to a set of lat,long coordinates"
22:51 cbreak there are geo spatial databases that can answer that quickly
22:51 cmyers I did everything short of inventing k-d trees on the spot, best I can tell from my memory of the interview.
22:51 cmyers I definitely said that
22:51 GavinMagnus left #git
22:51 cmyers even oracle has a geo column type and can build indexes and shit
22:51 cbreak or oct trees, or kd trees, or even uniform grids :)
22:52 ahr3n_ joined #git
22:52 elisa87 joined #git
22:52 cmyers anyways, suffice it to say, and to borrow Steve Yegge terminology, I think I just ran into my "anti-loop"
22:52 cbreak you're probably better off not working for facebook
22:52 elisa87 hey can you please have a look and let me know how to fix it http://pastebin.com/UkAZVfGM here's my repo: https://github.com/lamiastella/monajalal.github.io
22:53 cmyers maybe that too.  I ended up at Cloudera and I am super excited about the sort of problems they have in the build space.
22:53 cmyers like "it's so broken, it's gonna be so cool to fix" type problems.
22:53 BlaXpirit joined #git
22:53 osse do they have a ping pong table?
22:53 cmyers yes.
22:53 * osse applies
22:53 * cbreak got a PS4 with Mortal Kombat at work
22:53 cmyers =D
22:54 osse it's the only thing that matters
22:54 osse I write GUIs in Visual Basic to track IP addresses if I had a ping pong table
22:55 cmyers my team is still hiring for git/buildsystem/tools stuff, so pm me if you are serious =)
22:55 osse I can't say I am
22:55 osse Also it seems it would have to be a big move
22:56 osse The opening closest to me is in Legal :P
22:56 osse I'm not a lawyer, but I've seen one on TV.
22:56 cmyers heh.  they do some remote work, but they won't ship you a ping pong table =)
22:57 cmyers (probably)
22:57 osse it's the thing I miss the most from my old job
22:57 osse that and having my own office
22:57 cmyers I worked a job that spoiled me *forever*
22:58 cmyers 16 squares a week, catered goodness, laundry service, free drink fridge full of whatever you ask for
22:58 osse what's a square in this context?
22:58 BlaXpirit joined #git
22:58 cmyers a full catered meal, variety of stuff, menu changes every day, usually several choices
22:59 cmyers friday breakfast was always bacon + doughnuts.  probably took 5 years off my life.
22:59 osse hmm
22:59 osse sounds like the kind of place you'd be at 12-16 hours a day
23:00 cmyers indeed, that was the trick
23:00 cmyers and they hired young fresh-out-of-stanford kids *heavily*
23:00 elisa87 did you see http://pastebin.com/UkAZVfGM ?
23:00 osse that's not me.  might be related to the fact that I can walk to/from work in 15 minutes :p
23:01 cmyers elisa87: your remote is wrong I think:  I see "git@github.com/lamiastella/monajalal.github.io.git" but you probably mean "git@github.com:lamiastella/monajalal.github.io.git"
23:01 elisa87 I got this error when I added that cmyers fatal: remote origin already exists.
23:02 cmyers you need to modify the existing remote, not re-add it
23:02 elisa87 what's the command for that?
23:02 cmyers you can even just edit .git/config - it's plain text
23:02 osse elisa87: git remote set-url
23:02 cmyers that's it.
23:03 elisa87 $ git remote set-url --add git@github.com:lamiastella/monajalal.github.io.git didn't work
23:04 osse elisa87: you forgot to specify which remote to modify
23:04 elisa87 $ git remote set-url origin  git@github.com:lamiastella/monajalal.github.io.git
23:04 elisa87 worked
23:06 elisa87 error: failed to push some refs to 'git@github.com:lamiastella/monajalal.github.io.git' can you please help me with this?
23:06 elisa87 hint: Updates were rejected because the remote contains work that you do
23:06 elisa87 hint: not have locally. This is usually caused by another repository pushing
23:06 elisa87 hint: to the same ref. You may want to first integrate the remote changes
23:06 osse !force
23:06 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.
23:07 ybden !rewrite
23:07 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
23:07 fission6 joined #git
23:07 elisa87 I did a git pull origin master and then git push and it worked
23:08 rominronin joined #git
23:10 elisa87 any idea why push is stuck at this percentage Writing objects:  64% (29/45), 1.30 MiB | 43.00 KiB/s ?
23:13 cyan__ joined #git
23:13 AfroThundr joined #git
23:14 sangy joined #git
23:15 dokma joined #git
23:16 dokma I have a project that I want to add to git. This folder is my working folder in which I develop. Is this folder where I just 'git init'?
23:17 dokma I need to later publish this repo to production. Is this properly done with 'git co' on the prod machine?
23:18 boombatower joined #git
23:18 rueben_ joined #git
23:19 kre10s_ left #git
23:20 ahmedelgabri joined #git
23:22 rueben_ joined #git
23:25 elisa87 so why I get a 404 error http://monajalal.github.io/
23:26 elisa87 I have the same repo https://github.com/lamiastella/monajalal.github.io
23:26 elisa87 ??
23:27 ahr3n joined #git
23:28 allcentury joined #git
23:28 griffindy joined #git
23:29 WaReZ joined #git
23:29 dokma noone is here atm seems like...
23:31 ki0 joined #git
23:31 vishal dokma: 'yes' to the first question
23:31 vishal and for the second, see !deploy
23:31 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
23:32 Pieter joined #git
23:35 mattfoo joined #git
23:37 ahr3n joined #git
23:37 garethdaine joined #git
23:37 thiago joined #git
23:39 espindola joined #git
23:40 XDS2010 joined #git
23:40 mkelly joined #git
23:42 ash_workz joined #git
23:42 Orion3k joined #git
23:42 eletuchy_ joined #git
23:43 blur3d joined #git
23:44 eletuch__ joined #git
23:45 Grimm_ elisa87 You get 1 user address for github.io which has to be the same as your username. So https://github.com/lamiastella/lamiastella.github.io
23:45 EvilPenguin joined #git
23:45 Grimm_ elisa87 you can have as many project sites as you want though
23:45 elisa87 Grimm_: can I change my github username?
23:46 Grimm_ not sure
23:46 bjpenn left #git
23:46 Grimm_ looks like you can
23:47 Grimm_ elisa87 go to your profile then click on Account on the left side. Box right above Delete account is Change Username
23:47 eletuchy joined #git
23:47 elisa87 I changed my username
23:47 elisa87 thanks for the hint
23:47 Grimm_ cool
23:48 elisa87 but why is this still a 404? http://monajalal.github.io/
23:48 elisa87 Grimm_:
23:48 NeXTSUN joined #git
23:49 Grimm_ elisa87 i've heard it can take up to 20 mins for github DNS to update
23:49 eletuchy_ joined #git
23:49 elisa87 oh I see
23:49 elisa87 is my CNAME correct if it is just www.monajalal.com ?
23:50 ahr3n joined #git
23:51 Grimm_ there is a way to use a custom domain name with your github.io address but i've never done it
23:52 Grimm_ elisa87 https://help.github.com/articles/using-a-custom-domain-with-github-pages/
23:52 eletuchy joined #git
23:55 jimi_ joined #git
23:56 eletuchy_ joined #git
23:56 SwiftMatt joined #git
23:57 k3rn31 joined #git
23:57 ajf- joined #git
23:58 IWishIKnew joined #git
23:58 Tim-SEM joined #git
23:59 blackwind_123 joined #git
23:59 eletuchy joined #git

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