Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-12-22

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

All times shown according to UTC.

Time Nick Message
00:00 nicksloan joined #git
00:02 xaviergmail joined #git
00:02 ochorocho__ joined #git
00:06 mephistorium joined #git
00:07 dsantiag_ joined #git
00:07 cagomez joined #git
00:10 cagomez joined #git
00:11 howitdo joined #git
00:11 anuxivm left #git
00:11 nd_ joined #git
00:15 cagomez joined #git
00:19 mgoodwin How important do you think it is if it's been a while since you've committed on a project to break up your work with add -p
00:19 mgoodwin Or do you typically just commit everything that changed and write a detailed message
00:21 jw4242 joined #git
00:23 thiago how likely is that someone is going to read that commit, ever?
00:23 thiago you in the future, or someone else
00:25 jasondockers__ joined #git
00:29 cagedwisdom joined #git
00:29 ejb joined #git
00:30 WhiskerBiscuit joined #git
00:30 ToBeCloud joined #git
00:35 dermoth_ joined #git
00:37 johnmilton joined #git
00:38 SpecChum joined #git
00:38 newbie1 joined #git
00:40 clemf_ joined #git
00:40 Grum_ joined #git
00:40 dsantiago joined #git
00:41 frobnic_ joined #git
00:41 arand_ joined #git
00:41 anonymuse joined #git
00:43 sunny256 joined #git
00:43 sword` joined #git
00:44 rominronin joined #git
00:45 Werring` joined #git
00:46 blueyed joined #git
00:46 qpdb joined #git
00:46 diogenese joined #git
00:46 trobotham joined #git
00:46 Tsutsukakushi joined #git
00:46 marlinc joined #git
00:47 vdl joined #git
00:47 nikow joined #git
00:48 crayon joined #git
00:48 meZee joined #git
00:49 Sonderblade joined #git
00:49 dodobrain joined #git
00:50 fling joined #git
00:51 hahuang61 joined #git
00:53 ghostlight joined #git
00:53 mizu_no_oto joined #git
00:54 bouncy joined #git
00:55 uu joined #git
00:55 Sleepy63 joined #git
00:56 mikecmpbll joined #git
00:56 cagomez I have project/file.py, and project/morefiles/. how do I ignore project/__pycache__/?
00:56 nickabbey joined #git
00:57 thiago add project/__pycache__ to .gitignore
00:57 rewt joined #git
00:57 cagomez thiago: do I need a trailing slash?
00:58 thiago no
00:58 cagomez ok thanks
01:00 anonymuse joined #git
01:00 b1tchcakes joined #git
01:03 ochorocho__ joined #git
01:03 Jerem0 joined #git
01:06 MrWoohoo joined #git
01:07 _flow__ joined #git
01:07 midgetspy joined #git
01:07 dreiss joined #git
01:07 salamanderrake joined #git
01:07 stfn joined #git
01:07 hexagoxel joined #git
01:07 PaulCapestany joined #git
01:07 SpEcHiDeR joined #git
01:07 M-ou-se joined #git
01:07 Mogget joined #git
01:07 ixian joined #git
01:07 JelmerD joined #git
01:07 fuzzmz joined #git
01:07 Anthaas joined #git
01:07 phroa joined #git
01:07 eltopo[m] joined #git
01:07 Ineentho[m] joined #git
01:07 Sheogorath[m] joined #git
01:07 jgrasser joined #git
01:07 mquin joined #git
01:07 qpdb joined #git
01:07 Karazhan joined #git
01:09 Jerem0 left #git
01:10 albel727 joined #git
01:10 shgysk8zer0 joined #git
01:11 ghostlight joined #git
01:14 uu joined #git
01:15 pikajude joined #git
01:15 WhiskerBiscuit joined #git
01:16 SwiftMatt joined #git
01:16 bronson joined #git
01:24 anonymuse joined #git
01:29 webdev007 joined #git
01:32 Darren_ joined #git
01:33 Sasazuka joined #git
01:38 Poppa joined #git
01:38 moei joined #git
01:38 eizua joined #git
01:39 tmsmith joined #git
01:41 sanjoyd joined #git
01:42 sanjoyd I have a merge related question.  I often see these kinds of conflicts produced by git: https://ghostbin.com/paste/5fzdx
01:42 sanjoyd during a merge operation
01:43 sanjoyd and it seems to me that git should be able to resolve these trivially -- both upstream/master and HEAD added lines to a single location, and it seems to be obvious to take the union of these lines
01:44 sanjoyd Is there anything I can do to avoid manual work, short of writing a custom merge driver?
01:44 kyan joined #git
01:44 cdg joined #git
01:44 thiago no
01:44 thiago it isn't obvious to git that union is the correct solution
01:46 pypi_ joined #git
01:46 jaguarmagenta joined #git
01:48 donught joined #git
01:51 treehug88 joined #git
01:52 morenoh149 joined #git
02:01 amdi_ joined #git
02:03 ochorocho__ joined #git
02:04 pks joined #git
02:11 dviola joined #git
02:16 rominronin joined #git
02:17 Lyqyd joined #git
02:18 Limes joined #git
02:19 oksuruk joined #git
02:22 donught joined #git
02:27 finalbeta joined #git
02:34 d^sh joined #git
02:35 pks joined #git
02:40 anonymuse joined #git
02:42 xaviergmail joined #git
02:46 Vampire0_ joined #git
02:47 diogenese joined #git
02:48 MattE joined #git
02:49 dirtyroshi joined #git
02:49 Gsham joined #git
02:51 frostythesnowman joined #git
02:54 QuestionTime joined #git
02:54 kent\n joined #git
02:59 derk0pf joined #git
02:59 Gsham joined #git
03:00 bruce_lee joined #git
03:01 aavrug joined #git
03:02 aavrug joined #git
03:03 clemf joined #git
03:04 Neelixx joined #git
03:05 ochorocho__ joined #git
03:15 unbalancedparen joined #git
03:15 chachasmooth joined #git
03:17 bronson joined #git
03:18 xaviergmail joined #git
03:25 Goplat joined #git
03:25 Goplat joined #git
03:29 webdev007 joined #git
03:29 prg3 joined #git
03:29 plugwash joined #git
03:30 plugwash how can I find commits with no author?
03:32 clemf joined #git
03:35 Vortex35 joined #git
03:38 amdi_ joined #git
03:39 duderonomy joined #git
03:40 _Vi joined #git
03:41 morenoh149 joined #git
03:42 clemf joined #git
03:46 peeps[lappy] joined #git
03:47 rominronin joined #git
03:47 hexagoxel joined #git
03:47 llamapixel git log --author=“” would be my first guess ;)
03:48 drbean joined #git
03:48 jaguarmagenta joined #git
03:49 llamapixel On test that appears to find all
03:50 griffindy joined #git
03:50 GT4066 joined #git
03:51 svm_invictvs joined #git
03:52 llamapixel If you had to work the other way by omitting specific authors plugwash  http://dymitruk.com/blog/2012/0​7/18/filtering-by-author-name/
03:52 plugwash llamapixel, well I found a soloution myself
03:52 plugwash git log shows everything
03:52 plugwash git log --author="" only shows commits that have an author field
03:52 llamapixel ok
03:53 plugwash so if I feed the output of those two commands to diff I get the commits that don't have an author field
03:53 llamapixel I need to make another fake user for my git folder test to catch that, thanks
03:54 kpease joined #git
03:54 jon-mac joined #git
03:55 a_thakur joined #git
03:55 unked joined #git
03:56 shgysk8zer0 joined #git
03:58 sunri5e joined #git
04:01 a_thakur joined #git
04:04 SantaClauze joined #git
04:04 Neelixx_ joined #git
04:05 SantaClauze joined #git
04:08 meLon joined #git
04:09 jaguarmagenta joined #git
04:22 venmx joined #git
04:25 Jon28 joined #git
04:26 Shentino joined #git
04:27 whomp_ joined #git
04:27 whomp_ hi, when i do a git diff --name-only, it seems that i don't see which dirs were deleted. how can i show that?
04:29 clemf joined #git
04:30 fstd_ joined #git
04:31 cagomez joined #git
04:36 duderonomy joined #git
04:37 subhojit777 joined #git
04:39 clemf joined #git
04:42 romerocesar joined #git
04:44 Andrew_K joined #git
04:46 robattila256 joined #git
04:46 romerocesar joined #git
04:46 Gsham joined #git
04:50 Siegfried joined #git
04:51 aruns joined #git
04:52 clemf joined #git
04:52 deltab whomp_: git doesn't track that, just file contents and pathnames
04:53 nickabbey joined #git
04:56 thesha joined #git
04:57 romerocesar joined #git
04:57 knot joined #git
04:58 clemf joined #git
05:00 Gsham joined #git
05:00 bronson joined #git
05:01 ayogi joined #git
05:02 pypi_ joined #git
05:02 Hink joined #git
05:04 Neelixx joined #git
05:05 ochorocho__ joined #git
05:06 Hink joined #git
05:09 _Vi joined #git
05:10 romerocesar joined #git
05:13 llamapixel https://github.com/git/git
05:17 Rodya_ joined #git
05:17 Mowgli joined #git
05:18 griffindy joined #git
05:18 nckpz joined #git
05:19 Shentino joined #git
05:19 dreiss joined #git
05:20 rominronin joined #git
05:20 Cabanossi joined #git
05:22 jstimm joined #git
05:30 morenoh149 joined #git
05:33 _Vi joined #git
05:35 bket joined #git
05:36 howitdo joined #git
05:41 netj joined #git
05:47 nickabbey joined #git
05:52 _Vi joined #git
05:54 Gsham joined #git
05:55 Gsham joined #git
05:55 gkatsev joined #git
05:57 knot joined #git
06:00 Castor_Troy joined #git
06:00 bocaneri joined #git
06:01 Castor_Troy Hello all. I have a question. I have 2 identical branches. I made a change on branch2, didn't add and didn't commit. When I checkout branch1, I can see the change.
06:01 Castor_Troy How is that possible. I thought branch 1 should reflect till last commits and should not show changes made on other branches until I merge
06:02 spinningarrow Castor_Troy: When you make a change, it isn't actually "on a branch" until you commit it.
06:02 brent__ joined #git
06:02 menip joined #git
06:02 Castor_Troy If I am working on a feature in branch2 and made some modifications to files, they are shown in Branch 1 ?
06:03 kadoban Castor_Troy: !float (though it's just a longer-winded way of saying what spinningarrow already did
06:03 gitinfo Castor_Troy: If you have made a change in your working directory and have NOT YET COMMITTED, you may "float" that change over to another (`git checkout oldbranch`) or new (`git checkout -b newbranch`) branch and commit it there.  If the files you changed differ between branches, the checkout will fail.  In that case, `git stash` then checkout, and `git stash apply` and go through normal conflict resolution.
06:03 Castor_Troy wow, sounds like advanced stuff.
06:03 spinningarrow kadoban: I really need to spend some time knowing all the bot commands :P
06:04 Castor_Troy I'm assuming that branches are complete seperate copies and are independent of changes made in each, until I merge
06:04 khagler joined #git
06:04 kadoban Haha, it's hard. 1/2 the time I forget what they even are, but that one comes up a lot xD
06:04 freimatz joined #git
06:04 kadoban Castor_Troy: It sounds more complicated than it is. The commits on the branches are, but what's in the working directory, the files you actually see, isn't really *on* a branch at all, certainly changes you've made but haven't commited aren't.
06:05 Neelixx joined #git
06:05 spinningarrow Castor_Troy: No branches are not separate copies or completely independent. Let me see if there's a bot command that explains that well; 1 sec haha
06:05 kadoban Castor_Troy: Think about it this way: what should happen to those changes when you switch branches? What would you expect to happen?
06:05 spinningarrow !branch
06:05 gitinfo A branch and a tag are just convenient ways of spelling the name of a particular commit.  A commit represents a specific set of files and the history of all commits which came before it, and the SHA-1 hash tag official name provides cryptographic assurance of the lineage of a particular commit (and thus branch or tag).  A branch's reference may change.  A tag usually doesn't.
06:05 Castor_Troy so, if i want to get back on branch 1 and work on it, i should do a pull to be safe ?
06:06 kadoban Castor_Troy: What? Almost certainly not, pull doesn't seem to make sense here.
06:06 ochorocho__ joined #git
06:06 Castor_Troy kadoban, suppose I added a file readme.txt on branch2, when i come to branch1 it shouldnt be there.
06:07 spinningarrow Castor_Troy: that's basically correct with one difference: it won't be there as long as you *commit* it.
06:07 kadoban Castor_Troy: Okay, but what should happen to it if you haven't commited? When you change branch should git just delete it? You'd never be able to get it back. You'd have effectively lost all that work immediately.
06:07 Castor_Troy but what if i accidentalyl commit on branch1... readme.txt gets committed though its being worked on branch 2 and is not completed yet
06:08 llamapixel branches and nests of birds with eggs is how I got it.
06:08 deepy that's why I commit early, commit often, and rewrite history all the time
06:08 Castor_Troy kadoban, that makes sense. i thought git could save the session orsomething.
06:08 deepy and if I don't commit I stash
06:08 llamapixel what bird and nest are you sitting on ;)
06:08 mojava- left #git
06:08 kadoban Yeah, not really. You can use 'git stash', but it's not done automatically, it'd be a little confusing if it was probably.
06:09 kadoban Castor_Troy: As to accidentially commiting stuff you didn't mean to, that's part of why you should stage changes with 'git add' and check what you're commiting. I habitually run 'git diff --cached' to review what I'm actually going to commit before I do.
06:09 Castor_Troy But that doesn't make sense. Lets say I am working on two features at once. Like adding admin pages and working on documentation. I will create branches for both.
06:09 llamapixel If you don’t have a repo that you use to test ideas and angles, you are committing to the irc time waste branch
06:09 llamapixel Does your company have the docs as a sub branch?
06:10 kadoban Castor_Troy: Okay, with you so far.
06:10 Castor_Troy I am just using it as an example.
06:10 llamapixel Branches are usually as good as office locations in most situations
06:11 Castor_Troy so, if i start adding readme.txt and filemenuhelp.txt on branch2... and get back on branch1 to work to add login.php, finish it and commit it... readme and filemenuhelp also goes
06:11 Castor_Troy ok, so you are saying i should really be careful on what is getting staged and committed..
06:11 rscata joined #git
06:11 kadoban Castor_Troy: Yes. You should likely commit or stash those changes before you switch branches in that case.
06:12 Castor_Troy OK. Havent read about stash but may be it will help.
06:12 Castor_Troy also, it seems, two persons cant work on same computer on different branches, since there is no good way of knowing what is getting modified
06:12 kadoban It's kind of what you're expecting to happen automatically, except it's manual. It could be what you want
06:12 llamapixel a reset of your local svae
06:12 Darren_ joined #git
06:12 kadoban Castor_Troy: They can, but they should likely each have their own working repository.
06:12 llamapixel save * cool inventing new lexicon
06:13 CheckDavid joined #git
06:13 Castor_Troy yes, but kadoban like you said, they should commit or stash changes so as to not mess up others work in
06:13 kadoban It'd be relatively chaotic trying to have two people use the same git repo directly.
06:13 llamapixel Do you guys use a pre emptive master or a DMZ?
06:13 kadoban Castor_Troy: Hm? That's not really related, two people shouldn't be using the same repo at once.
06:13 llamapixel https://gist.github.com/djs​piewak/9f2f91085607a4859a66
06:14 Castor_Troy kadoban, i mean branches.. seperate branches on same computer.. anyway, i guess i understood
06:14 llamapixel same computer? both ssh’d?
06:14 Castor_Troy llamapixel, i am just guessing, i dont have any real world experience
06:14 plugwash Whether it's on the same computer or not you probablly want three repos
06:14 Castor_Troy OK, in real world, do you use command line or some GUI tool while working with git?
06:15 llamapixel how about looking at the pre staging document above ;)
06:15 kadoban It rather depends what you mean on the same computer. Usually if two people use the same computer, they have completely separate home directories and such, which is where you'd have git repos you're working on.
06:15 llamapixel always cmdline
06:15 plugwash one for each person and a main repo that people push/pull from/to
06:15 kadoban #git is pretty heavily biased towards the CLI
06:15 llamapixel all guis have a problem
06:15 llamapixel they are an extension from the real information
06:16 Derderderd joined #git
06:16 llamapixel Would you add a random movement to your nuclear power plant robot arms?
06:17 llamapixel Graphical user interfaces ( WIMP / DEV / DEV Brain / Code execution / Source / Bug Report )
06:17 llamapixel 2 many layers for my sanity lads.
06:17 Castor_Troy I liked github gui tool interface, since the timeline is pretty.. i am using smartgit now, but it is not nice when it comes to branches..
06:18 llamapixel I don’t mind using a third party repository to sore things but I prefer the command line to make sure it is safe.
06:18 kadoban I just use the git command line tool, and gitk for visualization occasionally.
06:19 llamapixel https://github.com/stevenjack/cig is the only thing I like to interupt that. https://github.com/stevenjack/cig
06:19 llamapixel https://github.com/llamapixel
06:19 gkatsev joined #git
06:20 Castor_Troy ok guys, thanks a lot, i will go back to studying
06:20 Raging_Hog joined #git
06:20 llamapixel If it helps make a test project, I still have one today for obscure questions
06:21 llamapixel Fred Joe Harry in branch RED Green Blue on sub projects ZXY
06:23 venmx joined #git
06:23 rtm1046 joined #git
06:29 svm_invictvs joined #git
06:35 ogny joined #git
06:37 ayogi joined #git
06:37 Mobutils joined #git
06:43 orfeo joined #git
06:44 vdv joined #git
06:44 misgit joined #git
06:44 ThomasLocke joined #git
06:46 menip joined #git
06:49 Shentino joined #git
06:51 rominronin joined #git
06:51 Domhack joined #git
06:53 _Vi joined #git
06:58 nisseni joined #git
07:00 morenoh149 joined #git
07:00 dermoth joined #git
07:05 dreiss joined #git
07:05 JeroenT joined #git
07:05 jaafar joined #git
07:06 qt-x joined #git
07:06 darkbit joined #git
07:08 inflames joined #git
07:09 JeroenT joined #git
07:09 _Vi joined #git
07:10 freimatz joined #git
07:10 algun joined #git
07:11 nellicus_ joined #git
07:12 tuor joined #git
07:12 lb1c joined #git
07:14 jknetl joined #git
07:15 darkbit joined #git
07:17 Shentino joined #git
07:17 rchavik joined #git
07:18 Hetu joined #git
07:19 lightstalker joined #git
07:19 spriz joined #git
07:20 zefferno joined #git
07:23 star_prone joined #git
07:23 SwiftMatt joined #git
07:24 lordjancso joined #git
07:25 star_prone joined #git
07:25 cdown joined #git
07:27 lmat joined #git
07:28 zeroed joined #git
07:28 venmx joined #git
07:30 pwnz0r joined #git
07:33 diogenese joined #git
07:35 ayogi joined #git
07:37 rominronin joined #git
07:37 sawk joined #git
07:38 nckpz joined #git
07:39 lb1c joined #git
07:45 phanimahesh joined #git
07:46 Sceorem joined #git
07:46 gareppa joined #git
07:47 pwnz0r joined #git
07:47 swalladge joined #git
07:49 venmx joined #git
07:53 oskarkv joined #git
07:54 spacelord_ joined #git
07:56 xaviergmail joined #git
07:57 mischat joined #git
07:59 eduardas_m joined #git
08:01 theoceaniscool joined #git
08:01 mda1 joined #git
08:02 nellicus_ joined #git
08:07 ochorocho__ joined #git
08:08 SwiftMatt joined #git
08:10 phanimah1sh joined #git
08:10 finalbeta joined #git
08:11 aswen joined #git
08:12 Box_of_Foxes joined #git
08:13 NimeshNeema joined #git
08:14 fffghhhhh joined #git
08:16 fffghhhhh left #git
08:17 www-bukolay-com joined #git
08:17 phanimahesh joined #git
08:18 Mir4g3 joined #git
08:19 Singmyr joined #git
08:20 rly joined #git
08:20 Guest7354 joined #git
08:22 JeroenT joined #git
08:27 JeroenT joined #git
08:28 Hink joined #git
08:30 Jackneill joined #git
08:31 nickabbey joined #git
08:31 _Vi joined #git
08:33 bl0w3d_0ut joined #git
08:34 Jackneill joined #git
08:34 InfoTest joined #git
08:36 ams__ joined #git
08:40 cdown_ joined #git
08:40 inflames joined #git
08:42 Balliad joined #git
08:42 vlt joined #git
08:44 SteffanW joined #git
08:44 bronson joined #git
08:44 Balliad joined #git
08:44 bernardio joined #git
08:45 netj joined #git
08:46 phanimahesh joined #git
08:46 Mement joined #git
08:48 Mement Is there a report what an average of total commits per day and how many lines of code a developer does? Or some best practice? Or is this not trackable
08:48 selckin completely useless metric
08:48 Panacea joined #git
08:48 RichiH joined #git
08:49 selckin but you can find many program to do it, google git stats for example
08:49 morenoh149 joined #git
08:49 jast there's no built-in way to generate a report like that, you can write your own script that parses the output from git
08:49 Mement selckin, Ok thanks for clearing that up. What metric would be beneficial if I like to prove that current Git usage is not right for a continuous integration workflow?
08:50 jast that's a question with many possible answers. what's wrong with the current git usage in your project, in your opinion?
08:51 tvw joined #git
08:51 Mement My opinion right now is that the employees are not committing often enough. Some only do it at the end of the day, to make sure it is 'stored' and can continue work tomorrow safely.
08:52 selckin stats are not gonna help that, you need to teach them that every commit should be a logical change, that compiles/works/etc
08:52 Mement This is done cause of the lack of knowledge what Git actually is and should not be seen as some storage folder. But I like to have some stats/report to show numbers and have some prove when I tell my founding
08:53 Mement Yes true, I understand
08:53 phroa maybe do a training session?  numbers can't fix people
08:53 selckin its not realted to git, they would have this issue with every source control
08:53 jast it depends very much on the complexity of the tasks being done. on more complex things I sometimes go a week or two without committing
08:53 jast it's actually preferrable to not commit (or at least not push) incomplete changes, if the incomplete change would break the build
08:54 aib joined #git
08:54 selckin s/commit/push/
08:54 kurkale6ka joined #git
08:54 Mement jast, But during that period you are able to pull other changes most of the time, cause of good communication that no one else will interfere with your work?
08:55 Mement Or no pulls at all?
08:55 Neelixx joined #git
08:55 jast the important bit, in my opinion, is to train people to understand which units of change make sense
08:55 jast well, typically I make these kinds of changes on a feature branch
08:55 jast you need a lot of coordination if two people are working on the same feature at the same time, no matter which process and tools you use
08:56 a_thakur joined #git
08:56 jast and I regularly rebase before pushing, sometimes even after pushing if people are aware I'm going to do it, so yes, I typically do use some type of update operation
08:56 willlondon joined #git
08:56 Mement It is not rarely done here as well, just curious :)
08:56 Mement it is rarely done here as well*
08:57 jast anyway, one of the least useful ways to use commits is to think of them as "snapshot of what I've been doing this past $unit-of-time"
08:58 jast it works for doing backups, basically, but commits that are self-contained, logical units of change are much, much more useful if you ever need to go back and understand what happened, or juggle a whole bunch of features being worked on
08:58 Mement Ok, maybe I was on the wrong track. Over here there is just no continuous integration and barely testing present. My assignment is implementing these.
08:58 jast that's kind of what you said about people making sure it's "stored" at the end of the day :)
08:58 vlt left #git
08:59 Mement Now I am looking for some way to show that implementing this will improve the software development and working environment.
08:59 nellicus_ joined #git
08:59 jast who are you needing to convince, just developers or others, too?
08:59 Mement But that hasn't to do much with #git anymore I guess. :)
08:59 Mement Developers (and college)
08:59 venmx joined #git
09:00 bl0w3d_0ut joined #git
09:01 jast okay. the thing is, numbers alone often don't convince people if, for example, the numbers contradict their personal experience or understanding of the situation
09:01 ki0 joined #git
09:01 Shentino joined #git
09:01 jast ideally you can make people see that there are concrete practical advantages to them if they do adopt CI and testing
09:02 jast it might be relevant to find out from them what problems they're currently having, what things aren't going smoothly, particularly in the direction of software stability and robustness, maintenance, that sort of thing
09:03 Mement I found some advantages and increase in productivity statistics using a annual report from Puppet. So I got something yes, but that's why I also like to try something that refers to my actual place
09:03 jast and arm yourself with case stories (from real projects talked about on the internet, for instance) that show how others have managed to fix these types of problems
09:03 Mement Got it
09:04 jast yeah, that's a little tricky because the actual place only has data for the "bad" practices and not the "good" ones :)
09:04 Random832 joined #git
09:04 Mement That's exactly what I wanted to type hehe
09:04 jast so the next best thing you can do is find stories from similar places
09:05 Mement It's somewhat hard using their current environment, cause they don't have any 'good' practice besides VCS. No testing, no automatic deployment, no integration
09:06 Mement I'll focus on that and stop my search (for now) trying to use some software metric that analyses their current environment
09:06 Mement Thank you :)
09:06 jast so, carefully (not too heavy-handedly) make them more aware of all the problems they're having right now because these things are missing
09:06 jast and when they're beginning to feel the pain, have things to tell them ;)
09:07 juboba joined #git
09:07 Mement Not too heavy... ok hehe. ;)
09:08 bl0w3d_0ut joined #git
09:08 bl0w3d_0ut1 joined #git
09:08 marcogmonteiro joined #git
09:09 jast it may take a while, I've found that patience pays off with these things
09:09 juboba hey guys. I'm trying the following: I have a bare repo with a script that will: cd to a working-directory, pull from this repo (both are in the same host, so pull /path/to/dir) and then compile and deploy some code
09:09 juboba but when trying to pull I get "fatal: no repo"
09:09 bl0w3d_0ut1 joined #git
09:10 juboba so I tried: git --work-tree=$WORKING_DIR --git-dir=$BARE_REPO pull
09:10 juboba but it doesn't work
09:10 Mement jast, Can I ask one last question? (totally unrelated to #git and VCS, but about software metrics as I get the idea you have some knowledge about this)
09:10 jast juboba: !deploy might help you get started with this type of thing. there are a bunch of examples there.
09:10 gitinfo juboba: 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
09:11 jast Mement: sure. I have to warn you that I'm very sceptical of performance metrics in general... :)
09:12 juboba jast: I know this. I'm just deploying in development ENV
09:12 juboba we use jenkins for real deploying
09:12 juboba but, how could I achieve pulling from the bare repo in the script?
09:12 jast the link has example scripts
09:12 jast have you looked at them?
09:12 BadSnappa joined #git
09:12 scarcry joined #git
09:13 juboba I am
09:13 juboba thanks
09:14 PioneerAxon joined #git
09:14 Mement jast, I've been advised software metrics to generate a score, rating their current environment. This score would look at amount of manual tasks need to be performed, how much time is needed and what failure rate is. (Subject is deploying a website from dev to prod)
09:14 jast there are some trade-offs with the various approaches to getting files from a bare repo, which is why there are a bunch of different examples and quite a bit of text that explains the various advantages and disadvantages
09:15 Mement Is this a good thing to track? If so, you have recommendations for me where I can learn to do this?
09:15 jast Mement: well, that actually seems like a very reasonable thing to keep track of :)
09:15 juboba thanks a lot jast
09:17 metachr0n joined #git
09:18 Neelixx_ joined #git
09:18 jaguarmagenta joined #git
09:19 jast Mement: it's difficult to automatically generate numbers, but you could use a very short questionnaire (the shorter, the better) and hand it to people doing the actual deployment, to fill out after they're done. example: how long did this deployment take (roughly)? how much of the time did you have to actively do something? were there any surprising/tricky bits (none, few, many)? on a scale of 0-10, how easy did you find doing this deployment? possibly a free-form fi
09:19 jast if you have plenty of deployment situations this should allow you to collect data fairly quickly
09:20 jast and if you can figure out a way to collect the data anonymously, it won't get distorted by people feeling as though their personal performance is being monitored
09:21 Mement Ah good one about the anonymity
09:21 jast e.g. have people hold onto their questionnaires after they've filled them out, and collect all of them at once later (and be sure to let them know about this, so they know they won't be judged individually)
09:22 Mement I think I got enough. I was on the wrong track by looking at total commits and all that stuff. :)
09:22 jast the problem with those kinds of numbers is that they're very indirect information
09:22 inflames joined #git
09:22 jast first step in collecting data: figure out which data will help you the most ;)
09:22 Mement :)
09:23 elastix joined #git
09:23 _Vi joined #git
09:26 bl0w3d_0ut joined #git
09:26 jast oh, one more thing. I don't know how involved you've been in the project in the past. getting a project into shape so CI doesn't fail due to old baggage can be very frustrating and time-consuming, so that needs attention
09:26 bl0w3d_0ut1 joined #git
09:26 nickabbey joined #git
09:28 juboba jast: that worked like a charm!
09:28 jast great
09:28 jakesyl_mobile joined #git
09:29 achadwick joined #git
09:31 _ADN_ joined #git
09:31 _jan joined #git
09:31 JeroenT joined #git
09:32 b100s joined #git
09:34 overlord_tm joined #git
09:34 troulouliou_div2 joined #git
09:36 mischat joined #git
09:36 mischat joined #git
09:41 jantje joined #git
09:42 YuGiOhJCJ joined #git
09:43 diogenese joined #git
09:44 bronson joined #git
09:46 bl0w3d_0ut joined #git
09:46 bl0w3d_0ut1 joined #git
09:47 navidr joined #git
09:54 ki0_ joined #git
09:55 nopf joined #git
09:59 HcsOmot joined #git
10:00 bronson joined #git
10:04 HcsOmot hi. I don't feel like experimenting too much, so I thought I'd ask here for help. What would happen to tracked files if I deleted .git directory? I foolishly issued git add . command inside my .config directory, and some of the files were updated, and I committed and pushed those changes.
10:04 Gloomy joined #git
10:04 HcsOmot I realized I don't need to keep track of everything in .config. How do I discard all but a handful of files I want to keep tracking?
10:05 m0viefreak joined #git
10:05 jast if you delete the .git repository, you'll essentially turn your local repository into just a bunch of normal files, without any git history/index at all
10:06 jast what are you trying to do, undo the push?
10:07 systemovich joined #git
10:07 bernardio joined #git
10:07 NeverDie joined #git
10:08 HcsOmot ok, but what about the contents of the files? there was initial content. files got tracked, commited and pushed. some files changed, changes got committed and then pushed. now some of those files changed again, but without commit. if i delete .git dir, will I be left with whatever is the current version on the branch, or will they revert to whatever the content was before the initial commit/push?
10:08 ochorocho__ joined #git
10:08 canton7 deleting the .git dir doesn't magically change the other files on your disk - that's not how filesystems work
10:09 HcsOmot no, I don't mind the push, I can go straight to the server and delete the whole repo. I'll probably start over with just tracking a handful of files I care about.
10:09 canton7 if you delete the .git dir, the other files on your filesystem stay exactly as they were at the point that you deleted the .git dir
10:09 HcsOmot ok, but i'm talking about the files that were tracked, changed, and committed.
10:10 PHPanos joined #git
10:10 HcsOmot the current version of the file was committed. so the staging area is clean. i delete .git dir, and I'll be left with current state, right?
10:11 canton7 I'm not sure how else to explain this. Any files currently on your filesystem are left untouched. Anything that's stored in the .git  - branches, history, commits, etc - is lost
10:11 howitdo joined #git
10:11 canton7 you'll be left with whatever files are currently on your filesystem
10:11 canton7 take a look inside a file. That content will stay there when you delete the .git dir
10:11 HcsOmot great! that's what I'm looking for! thanks!
10:11 bernardio joined #git
10:12 canton7 it's like... I have a box of cookies. I take a cookie out and put it on the table. I then throw the box away. The cookie I put on the table is still on the table, because I didn't throw it away
10:12 jast great, now I'm hungry, thanks canton7
10:12 canton7 ><
10:13 HcsOmot but just out of curiosity - if I wanted to be left with a previous version of a file, I could do a rollback to whatever commit I'm looking for, and then delete .git dir. that will keep the content that's been committed at that point in time, right?
10:13 jast sure
10:13 canton7 yes. Deleting the .git dir does not change anything *outside* the .git dir
10:13 jast you don't even need to delete .git
10:13 canton7 whatever content is in files outside of the .git dir is left untouched when you delete the .git dir
10:13 jast because there are a bunch of tools to rewrite your repository's history if you don't like it the way it is
10:14 jast almost like time travel, but more affordable
10:16 bernardio_ joined #git
10:17 pcemnky joined #git
10:19 HcsOmot actually, i'm not sure that's the problem. I pushed the complete .config dir, and there's a lot of stuff in there I don't actually need to keep the track of. I'm mostly interested in config files for just a couple of apps that I customized for my workflow, and a couple of system config files. everything else is dispensable
10:20 HcsOmot so I was planning on just removing the whole repo from the server and starting over with just the stuff I need
10:21 Serpent7776 joined #git
10:23 venmx joined #git
10:24 pcemnky left #git
10:25 canton7 joined #git
10:25 kedare joined #git
10:26 canton7 joined #git
10:28 Hetu joined #git
10:32 troulouliou_div2 joined #git
10:34 theoceaniscool joined #git
10:36 chele joined #git
10:36 rtm1046 joined #git
10:38 morenoh149 joined #git
10:40 d0nn1e joined #git
10:43 Derderderd joined #git
10:43 Panacea joined #git
10:44 JeroenT_ joined #git
10:44 JeroenT_ joined #git
10:44 manuelschneid3r joined #git
10:46 Panacea joined #git
10:51 nettoweb joined #git
10:53 Doginal joined #git
10:55 Funeral joined #git
10:56 phanimahesh joined #git
11:01 xaviergmail joined #git
11:05 Singmyr joined #git
11:11 bernardio joined #git
11:12 onmeac joined #git
11:17 Anja joined #git
11:18 ki0 joined #git
11:19 jaguarmagenta joined #git
11:19 WayToDoor joined #git
11:24 venmx joined #git
11:24 mikecmpbll joined #git
11:24 JuliusN how can i choose which account (rsa key) shh and/or git uses when i try to connect to remote (bitbucket in this case)?
11:25 belak JuliusN: BB cloud or server?
11:25 somaReverse joined #git
11:25 grawity JuliusN: ssh -i
11:25 JuliusN git@bitbucket.org
11:25 grawity JuliusN: .ssh/config
11:26 belak And are the keys attached to different user accounts or are some of them deploy keys
11:27 JuliusN what's a deploy key?
11:27 belak They're read only keys usually used to clone repos during deployments
11:27 belak I assume you're asking about this because you have multiple user accounts?
11:28 JuliusN yes, i have my personal one and then the one i need to use at work
11:28 belak Then you can just use your username rather than git@
11:28 JuliusN oh
11:28 belak We just added that a few months ago
11:29 JuliusN that seems to work
11:30 belak Good to hear
11:30 belak I'd be worried if it didn't
11:30 JuliusN it might be related to ssh or git configs that the connection was authenticated with my personal account by default
11:31 belak git@ will guess the user based on the keys used. Whichever key is sent first (that has a user associated with it) will be used
11:31 belak If you specify a username we filter the keys we check against and only use keys for that username
11:35 Doginal joined #git
11:38 bernardio_ joined #git
11:41 pcre joined #git
11:43 spacelord_ joined #git
11:44 clmsy joined #git
11:44 Derderderd joined #git
11:45 chll_ joined #git
11:50 afuentes joined #git
11:51 jeffreylevesque joined #git
11:52 CtrlC joined #git
11:52 CtrlC joined #git
11:53 DrRobotic joined #git
11:54 deadnull joined #git
11:59 elastix joined #git
11:59 cyphase joined #git
12:00 vuoto joined #git
12:01 PHPanos joined #git
12:01 quackgyver joined #git
12:03 Cavallari joined #git
12:05 Neelixx joined #git
12:09 ochorocho__ joined #git
12:09 hariel joined #git
12:11 ferr1 joined #git
12:11 diogenese joined #git
12:12 hariel joined #git
12:19 NeverDie_ joined #git
12:19 vuoto joined #git
12:25 venmx joined #git
12:26 bl0w3d_0ut1 joined #git
12:27 morenoh149 joined #git
12:29 a_thakur_ joined #git
12:29 bl0w3d_0ut joined #git
12:31 bl0w3d_0ut joined #git
12:31 bl0w3d_0ut1 joined #git
12:32 rnsanchez joined #git
12:33 bl0w3d_0ut1 joined #git
12:35 Macaveli joined #git
12:39 roelmonnens joined #git
12:39 NeverDie joined #git
12:44 dermoth_ joined #git
12:45 Derderderd joined #git
12:47 Panacea joined #git
12:48 ktoverso joined #git
12:48 a_thakur joined #git
12:49 Gsham joined #git
12:49 Neelixx_ joined #git
12:53 byte512 joined #git
12:55 cdown joined #git
12:56 ojdo joined #git
12:58 ktoverso joined #git
12:59 cqi joined #git
13:01 cqi joined #git
13:01 johnmilton joined #git
13:05 Derderderd joined #git
13:06 noecc joined #git
13:07 ayogi joined #git
13:10 mikecmpbll joined #git
13:12 nettoweb joined #git
13:13 netj joined #git
13:17 donught joined #git
13:19 jaguarmagenta joined #git
13:20 google-bukolay-y joined #git
13:21 donught joined #git
13:24 cqi joined #git
13:24 howdoi joined #git
13:24 bl0w3d_0ut joined #git
13:24 bl0w3d_0ut1 joined #git
13:26 venmx joined #git
13:27 ispn joined #git
13:28 _ikke_ New git command: git commend (commit + ammend) :D
13:29 Anticom joined #git
13:29 selckin real?
13:29 _ikke_ no
13:29 _ikke_ I just typed it
13:29 _ikke_ my brain merged the two
13:30 Anticom Hi all. My HEAD is currently at e2dd7c0fcc41a008aaa164d4273b29802a041869 but when i do `git describe --always --long --tags` i get v1.4-0-ge2dd7c0 Why is the abbreviated hash prefixed with 'g' ?
13:30 selckin think i have every permutation of rebase and commit aliased to them
13:30 donught joined #git
13:31 _ikke_ Anticom: That's part of the format
13:31 Anticom _ikke_: okay so this is the way it's supposed to be (?)
13:31 selckin The "g" prefix stands for "git"
13:31 selckin from the man page
13:32 selckin <tag>-<commits>-g<sha>
13:32 khagler joined #git
13:32 _ikke_ Anticom: Note that g is not a hex character, so it cannot be part of the hash
13:33 Neelixx joined #git
13:33 Anticom _ikke_: good point. I was just wondering, why that might not lead to confusion. But that explains it
13:33 theoceaniscool joined #git
13:37 lss8 joined #git
13:37 lss8 joined #git
13:39 GodGinrai joined #git
13:41 a_thakur joined #git
13:43 synthroid joined #git
13:47 sdothum joined #git
13:47 xaviergmail joined #git
13:47 jeffreylevesque joined #git
13:47 jaguarmagenta joined #git
13:49 Neelixx joined #git
13:52 Macaveli joined #git
13:52 flavius joined #git
13:53 jcelerier joined #git
13:53 jcelerier hello :)
13:53 jcelerier how can one remove all commits, and data, of a given branch ?
13:54 QuestionTime joined #git
13:54 canton7 delete the branch?
13:54 jcelerier (explanation: I have a specific branch on my repository where my continuous integration tool pushes the output of doxygen at each commit I make on master)
13:54 jcelerier canton7: will it also delete the data of each commit in this branch ?
13:55 canton7 if the commits are no longer referenced by any branches (or tags, etc), they will eventually be garbage collected
13:55 canton7 ... but you can force a branch to point to a specific commit, which sounds more like what you want?
13:55 jcelerier okay
13:55 jcelerier yes maybe :p
13:55 canton7 (remember that a branch is just a pointer to a commit)
13:56 jcelerier but I don't want to keep the previous commit
13:56 jcelerier (because it uses up disk space for nothing and my repo becomes more and more slow to clone every day)
13:56 ktoverso joined #git
13:57 canton7 so you probably want to configure your build server to reset the doxygen branch to point to some specific commit (maybe the commit being built?), then commit the doxygen to it, then force-push
13:57 canton7 although tracking build *output* in a git repo smells a bit...
13:57 jcelerier yes, that would be it
13:57 jcelerier well that's the only way I've found to host my doxygen cheaply
13:57 jcelerier (e.g. on github's gh-pages branch)
13:57 Juzzika joined #git
13:59 canton7 anyway, 'git checkout doxygen-branch; git reset --hard someCommit'
13:59 canton7 or 'git branch -f doxygen-branch someCommit; git checkout doxygen-branch'
13:59 ochorocho__ joined #git
13:59 jcelerier hmm, so my reset --hard could be on the first commit on that branch for instance
14:00 jcelerier (its history is separate from the "code" branches's history)
14:00 canton7 sure. First figure out what you want the history of the branch to look like, then the git commands will follow
14:00 aavrug joined #git
14:00 jleclanche does anyone know if there's an option to backup uncommitted changes as diffs in .git whenever they're being thrown away?
14:01 selckin stash
14:01 jleclanche no, i mean to actively prevent it
14:01 jleclanche like to prevent the situation of "oops i just git reset --hard a bunch of changes i didn't mean to"
14:01 canton7 jleclanche, "whenever they're being thrown away", e.g. by 'git reset' or 'git checkout'?
14:01 jleclanche either
14:01 minikN joined #git
14:02 canton7 sure, I was giving examples in the name of clarity :P
14:02 jleclanche yeah i misread you. yes exactly
14:02 canton7 I'm not aware of any way... don't think any hooks will run for that (man githooks)
14:02 gitinfo the githooks manpage is available at http://jk.gs/githooks.html
14:03 jleclanche i was talking to someone about their git gui and wondering why the gui didn't do exactly this. then i wondered why git itself doesn't do exactly this, or doesn't have an option for it somewhere
14:03 DarsVaeda joined #git
14:03 theoceaniscool joined #git
14:03 rnsanchez_ joined #git
14:03 BrownBear joined #git
14:03 selckin because throwing thsoe changes away is usually the point
14:04 DarsVaeda hi I have an existing git repo and want to start a new development in it for a new version that is not depended on the prior version at all, basically it is a fresh start, what would be the best option to start?
14:04 canton7 DarsVaeda, first off, are you sure you don't want a new git repo?
14:04 jleclanche selckin: usually being the keyword. just like when i delete a commit the point is usually to delete it. but it's still here in the reflog and that's super useful
14:04 DarsVaeda yes
14:05 canton7 DarsVaeda, you can create a new branch which isn't based on anything using 'git checkout --orphan newbranch && git rm -rf .'
14:05 jcelerier mh, I have a single commit in my git log (the first of my doxygen branch). However there is already some data in it, that I want to get rid of
14:05 DarsVaeda my current thinking is, that I start a new ...
14:05 DarsVaeda ah that's what I had in mind too
14:05 jcelerier could I just create a "new" empty commit ?
14:05 DarsVaeda thanks :)
14:06 jcelerier when I search for "git empty commit" I only find stuff about pushing commits with no changes
14:06 ktoverso joined #git
14:06 Derderderd joined #git
14:06 canton7 jcelerier, you might want to do something similar: 'git branch -D doxygen-branch && git checkout --orphan doxygen-branch && git rm -rf . && <add and commit doxygen stuff> && git push -f remotename HEAD'
14:06 jcelerier ahh
14:07 jcelerier git checkout --orphan is what I wanted
14:07 jcelerier thanks !
14:07 canton7 strange how questions of one type tend to come in waves
14:07 jcelerier quantum causality at its finest
14:09 jeffreylevesque anyone here use BFG cleaner before?
14:09 ShekharReddy joined #git
14:09 canton7 jeffreylevesque, I'm sure someone has
14:09 jeffreylevesque just wondering if there is any magic to it
14:10 jeffreylevesque so i install it, and run a command to remove all history of a file
14:10 canton7 magic in what sense?
14:10 jeffreylevesque I'm looking at https://rtyley.github.io/bfg-repo-cleaner/
14:11 jeffreylevesque so i can just run `bfg --delete-files id_{dsa,rsa}  my-repo.git`, and it will remove "'id_rsa' or 'id_dsa'" and all it's history?
14:12 anonymuse joined #git
14:12 canton7 if that's what the docs say, yep
14:14 Raging_Hog joined #git
14:15 ExoUNX joined #git
14:16 morenoh149 joined #git
14:18 jimi_ joined #git
14:19 Nitesh joined #git
14:20 marcogmonteiro joined #git
14:21 AlineGomes joined #git
14:22 Macaveli joined #git
14:22 mda1 joined #git
14:25 Macaveli joined #git
14:25 adino joined #git
14:26 cdg joined #git
14:26 venmx joined #git
14:29 Siegfried joined #git
14:29 gribouille joined #git
14:29 gribouille hi
14:29 gitinfo gribouille: 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.
14:30 plos joined #git
14:31 nfm joined #git
14:31 gribouille left #git
14:31 gribouille joined #git
14:32 gribouille how should I write .git/hooks/pre-commit to build my program before each commit?
14:32 _Cyclone_ joined #git
14:32 systemovich joined #git
14:36 f3r70rr35f joined #git
14:40 plos joined #git
14:41 nickabbey joined #git
14:42 _ikke_ Nothing really special
14:43 _ikke_ Only thing might be that you need to cd to the working tree
14:43 gribouille _ikke_, what do you mean?
14:43 mdw joined #git
14:44 newbie1 joined #git
14:44 gribouille _ikke_, I want to build the program as it is recorded in the index, not the version in the working treee
14:46 _ikke_ gribouille: Then you have to check out the index to a temporary directory
14:46 danielmahon joined #git
14:47 gribouille _ikke_, that's what I'm doing right now, but each time I change a file and commit, I have to rebuild the whole program
14:47 Panacea joined #git
14:48 eduardas_ joined #git
14:49 jcelerier thanks canton7, it worked admirably
14:50 canton7 sweet!
14:52 adino joined #git
14:52 google-bukolay-y joined #git
14:54 nikivi joined #git
14:56 plos joined #git
14:57 Neelixx joined #git
14:57 plos joined #git
14:58 sagerdearia joined #git
14:58 fahadash When you commit something to a branch, do a fetch, find out that someone else also committed something to the same branch around that time you did, and you end up with two heads on your local on that branch. Is there a good name/proper term for this issue? I call it Double-Head Issue.
14:59 Dougie187 joined #git
15:00 star_prone joined #git
15:01 Juzzika joined #git
15:01 plos joined #git
15:01 ochorocho__ joined #git
15:01 UTAN_dev_ joined #git
15:02 _ikke_ git calls it diverged branches
15:02 plos1 joined #git
15:03 Es0teric joined #git
15:03 re1 joined #git
15:05 ejb joined #git
15:05 gribouille is there a way to apply the staged changed in a directory?
15:05 ResidentBiscuit joined #git
15:05 orfeo joined #git
15:06 qqx gribouille: git checkout-index
15:06 adino joined #git
15:06 Rapture joined #git
15:07 plos joined #git
15:08 plos joined #git
15:08 Gsham joined #git
15:08 shinnya joined #git
15:08 Derderderd joined #git
15:11 plos joined #git
15:11 rominronin joined #git
15:11 axisys_ joined #git
15:13 marc_v92 joined #git
15:13 kryojenik joined #git
15:14 kryojenik Quick question.  Can I use a clone of a repro as the origin for another machine to clone from?
15:14 nikivi joined #git
15:14 marlinc joined #git
15:14 ejb joined #git
15:15 gribouille git checkout-index doesn't delete files that have been removed by git rm
15:16 ki0 joined #git
15:18 qqx gribouille: Good point. You might be able to use `git diff --cached | git apply`
15:19 e14 joined #git
15:19 Vortex34 joined #git
15:23 refried_ joined #git
15:25 plos joined #git
15:26 timvisher joined #git
15:27 King_Hual joined #git
15:27 venmx joined #git
15:30 jast kryojenik: in principle, yes. one thing that might trip you up is that you'll be cloning only the local branches, i.e. the ones you've already worked on in the first clone (only the default branch, e.g. master, by default)
15:32 jast side note on the other topic of discussion: checkout-index is a low-level ("plumbing") tool. the same can be done with 'git checkout'.
15:32 plos joined #git
15:32 jstein_ joined #git
15:33 jast gribouille: checkout-index (and git checkout in its second form with a commit argument) only look at the index, which is precisely where 'git rm' removes things from :) to restore a file from the most recent commit, use 'git checkout HEAD -- path/to/file'
15:33 jast gah
15:33 jast *without a commit argument
15:33 King_Hual joined #git
15:33 star_prone joined #git
15:33 jast side note, 'git status' describes in each section of its output how to undo what happened to the files in that section
15:34 star_prone joined #git
15:34 plos joined #git
15:35 marc_v92 joined #git
15:36 Noldorin joined #git
15:36 fahadash Thanks _ikke_
15:36 WayToDoor joined #git
15:37 al-damiri joined #git
15:37 _ikke_ Double-Head issue sounds to me like it's a problem, while this happens regulary in normal workflows
15:38 nikivi joined #git
15:39 ki0 joined #git
15:40 google-bukolay-y joined #git
15:40 morenoh149 joined #git
15:41 kryojenik Thx jast
15:44 mdw joined #git
15:44 Derderderd joined #git
15:44 nettoweb joined #git
15:45 mdw_ joined #git
15:46 mdw__ joined #git
15:48 shgysk8zer0 joined #git
15:49 Darren_ joined #git
15:50 GodGinrai joined #git
15:50 clemf joined #git
15:51 Es0teric joined #git
15:51 likewhoa joined #git
15:52 Derderderd joined #git
15:53 plos joined #git
15:56 plos joined #git
15:56 ki0 joined #git
15:56 cyphase joined #git
15:57 finalbeta joined #git
15:58 plos joined #git
15:59 yehowyada joined #git
15:59 plos joined #git
15:59 Panacea joined #git
16:00 jasondockers joined #git
16:00 jasondockers joined #git
16:01 crayon joined #git
16:01 e14 joined #git
16:02 gribouille jast, does it explain why git checkout-index doesn't remove files that have been removed with git rm?
16:03 a_thakur joined #git
16:03 chll_ joined #git
16:03 Darren_ joined #git
16:03 roelmonn_ joined #git
16:04 mdw joined #git
16:04 Rodya_ joined #git
16:05 plos joined #git
16:08 qqx gribouille: That doesn't have any information about history, only the current state of the index.
16:09 qqx The only way it could remove files that were removed from git would be to remove anything that's unknown. Which would make it useless for your use case, you'd still need to rebuild everything each time.
16:10 GodGinrai joined #git
16:11 subhojit777 joined #git
16:11 Rodya_ joined #git
16:12 albel727 joined #git
16:12 jast 163306 < jast> gribouille: checkout-index (and git checkout in its second form with a commit argument) only look at the index, which is precisely where 'git rm' removes things from :) to restore a file from the most recent commit, use 'git checkout HEAD -- path/to/file'
16:12 jast wait, what do you mean "remove"? aren't they already gone after you used 'git rm'?
16:13 gribouille jast, yes they did
16:13 jast I'm guessing you're trying to update files in some other location based on changes being made in your repo
16:13 saml joined #git
16:14 gribouille jast, git checkout-index with the --prefix option
16:14 plos joined #git
16:15 jast anyway, the simple answer is that the index doesn't know what used to be in the index, only what's in there right now
16:16 gribouille jast git checkout-index --prefix=tmp/ doesn't remove the file tmp/foo, although I did git rm foo
16:16 jast so, checkout-index can't distinguish "this file used to be in the index and so now I should remove it" and "this file never was in the index, so it's untracked and should stay where it is"
16:16 thunderbirdtr joined #git
16:17 gribouille jast: there should be an option to remove files that are not in the index
16:17 SpeakerToMeat joined #git
16:17 jast there's a separate command for that, it operates on the working tree, though
16:18 jast if tmp/ truly is temporary, you could remove it and then run checkout-index :)
16:18 thebope joined #git
16:18 Rodya_ joined #git
16:19 marc_v92 joined #git
16:20 Rodya__ joined #git
16:22 gribouille jast, no, because git checkout-index doesn't preserve timestgamps
16:23 jast neither does any other operation in git that updates files
16:23 gribouille jast, git diff --cached | git apply --directory=tmp/ preserves the timestamps on unmodified files
16:24 jast actually doesn't that do what you're looking for?
16:24 gribouille jast, yes it does
16:25 ki0 joined #git
16:25 jast well, that's good then, I guess
16:25 gribouille yes it is
16:26 rominronin joined #git
16:26 narendraj9 joined #git
16:26 Funeral joined #git
16:27 rominronin joined #git
16:28 rominron_ joined #git
16:28 venmx joined #git
16:29 darkbit joined #git
16:30 nydel07 joined #git
16:30 Dandelock joined #git
16:30 clemf joined #git
16:32 ejb joined #git
16:34 knot joined #git
16:36 kryojenik left #git
16:36 traxo-xx joined #git
16:38 Rodya_ joined #git
16:39 synthroid joined #git
16:40 svm_invictvs joined #git
16:40 refried_ joined #git
16:47 neunon joined #git
16:47 ngrin joined #git
16:48 refried_ joined #git
16:51 e14 joined #git
16:52 treehug88 joined #git
16:53 kab joined #git
16:53 Derderderd joined #git
16:57 Salander27 joined #git
16:58 jnavila joined #git
16:58 sathed joined #git
17:02 ochorocho__ joined #git
17:05 gzim joined #git
17:06 relipse joined #git
17:07 relipse how do I finish resolving a merge? I modified the file in git status, i git add and git commit and git merge --commit it says "Already up-to-date." but when i look at the branch i'm on it is off kilter and says feature/foobar ↑·85|…2⚑ 7
17:08 cagmz joined #git
17:11 howitdo joined #git
17:11 Jackneill joined #git
17:13 jedahan joined #git
17:16 xaviergmail joined #git
17:17 bronson joined #git
17:17 ayogi joined #git
17:17 moritz relipse: what does 'git status' say?
17:18 gargola joined #git
17:18 Macaveli joined #git
17:21 thunderbirdtr joined #git
17:23 anonymuse joined #git
17:26 jccn joined #git
17:26 timvisher is there an argument to `make install-man` which changes the prefix?
17:27 algun joined #git
17:27 snowkidind joined #git
17:29 venmx joined #git
17:30 Derderderd joined #git
17:30 whomp_ joined #git
17:30 nikivi joined #git
17:31 rominronin joined #git
17:35 menip joined #git
17:35 Tsutsukakushi joined #git
17:35 alip joined #git
17:37 nidr0x joined #git
17:37 sea-gull_ joined #git
17:39 WayToDoor joined #git
17:40 robogoat joined #git
17:40 e14 joined #git
17:40 edaught joined #git
17:41 boombatower joined #git
17:43 mikecmpbll joined #git
17:45 Disgeae joined #git
17:46 buckowski joined #git
17:46 Tsutsukakushi joined #git
17:46 Ardethian joined #git
17:46 Balliad joined #git
17:47 Rodya_ joined #git
17:47 dreiss joined #git
17:49 freimatz3 joined #git
17:50 kpraveen joined #git
17:51 cdg joined #git
17:52 duderonomy joined #git
17:55 hue joined #git
17:57 kier joined #git
17:58 jedahan joined #git
18:00 buckowski joined #git
18:01 e14 joined #git
18:01 jnavila joined #git
18:01 emg joined #git
18:02 star_prone joined #git
18:03 knot joined #git
18:04 spacelord_ joined #git
18:04 jstimm joined #git
18:06 cdg joined #git
18:07 sathed left #git
18:09 ki0 joined #git
18:10 synthroi_ joined #git
18:11 Fissure joined #git
18:12 pypi_ joined #git
18:12 dviola joined #git
18:15 jedahan joined #git
18:16 Es0teric joined #git
18:18 nisseni joined #git
18:18 Spec-Chum joined #git
18:19 f3r70rr35f joined #git
18:23 tyreld joined #git
18:25 nisseni_ joined #git
18:25 vdv joined #git
18:25 sea-gull_ joined #git
18:27 [0xAA] joined #git
18:28 cdown joined #git
18:28 [0xAA] joined #git
18:28 [0xAA] joined #git
18:29 [0xAA] joined #git
18:30 venmx joined #git
18:30 Balliad joined #git
18:35 nisseni joined #git
18:36 bannakaffalatta joined #git
18:39 vuoto joined #git
18:39 idlemind joined #git
18:39 ktoverso joined #git
18:40 idlemind using the official windows client how do i tell what ssh key it will use during remote operations?
18:40 cdown joined #git
18:40 bannakaffalatta joined #git
18:41 hahuang61 joined #git
18:42 Derderderd joined #git
18:42 SpeakerToMeat joined #git
18:43 cyan__ joined #git
18:44 Ardethian joined #git
18:47 cagedwisdom joined #git
18:48 finalbeta joined #git
18:50 Rodya_ joined #git
18:54 crayon joined #git
18:57 manuelschneid3r joined #git
18:58 fk joined #git
18:58 rominronin joined #git
19:01 Gsham joined #git
19:01 rominron_ joined #git
19:02 ochorocho__ joined #git
19:03 sammyo joined #git
19:04 Sasazuka joined #git
19:04 nwkj86 joined #git
19:05 j416 I'm guessing it'll try all of them, like regular ssh
19:06 ktoverso_ joined #git
19:06 regedit joined #git
19:08 cagedwisdom joined #git
19:08 navidr joined #git
19:12 beastd joined #git
19:12 crayon joined #git
19:17 morenoh149 joined #git
19:18 Singmyr joined #git
19:18 synthroid joined #git
19:22 beastd left #git
19:22 venmx joined #git
19:25 JanC_ joined #git
19:26 JanC__ joined #git
19:27 shgysk8zer0 joined #git
19:28 willlondon joined #git
19:30 bronson joined #git
19:32 Doginal joined #git
19:36 jstein_ joined #git
19:37 WhiskerBiscuit joined #git
19:37 wizardedit joined #git
19:38 MrEnergy joined #git
19:38 emg joined #git
19:41 dave0x6d joined #git
19:43 Random832 joined #git
19:45 jens__ joined #git
19:45 ktoverso joined #git
19:48 multi_io_ joined #git
19:50 kyan joined #git
19:52 Rodya_ joined #git
19:52 jacob123 joined #git
19:53 whomp_ joined #git
19:54 umpc joined #git
19:55 knot joined #git
19:56 Guest32_ joined #git
19:56 BSaboia joined #git
20:01 Es0teric joined #git
20:02 anonymus_ joined #git
20:06 rewt joined #git
20:07 e14 joined #git
20:10 e14 joined #git
20:10 Es0teric joined #git
20:12 Derderderd joined #git
20:13 BrianJ joined #git
20:14 theoceaniscool joined #git
20:14 e14 joined #git
20:16 Random832 joined #git
20:17 e14 joined #git
20:18 tang^ joined #git
20:19 star_prone joined #git
20:23 sathed joined #git
20:26 xaviergmail joined #git
20:27 sathed Is there an easy way to test out some groovy hooks for gitblit?
20:29 Es0teric joined #git
20:29 comps joined #git
20:30 Rapture joined #git
20:30 bronson joined #git
20:30 daynaskully joined #git
20:32 jasondockers__ joined #git
20:33 rominron_ joined #git
20:33 aidalgol joined #git
20:33 jasondockers__ joined #git
20:33 polyrob_ joined #git
20:35 hahuang65 joined #git
20:36 madewokherd joined #git
20:37 synthroi_ joined #git
20:38 cdg joined #git
20:38 nisseni joined #git
20:40 dodobrain joined #git
20:41 Learath2 joined #git
20:41 ajf- joined #git
20:41 derk0pf joined #git
20:41 UTAN_dev joined #git
20:41 ejb joined #git
20:41 derk0pf joined #git
20:41 cdg joined #git
20:44 e14 joined #git
20:45 derk0pf joined #git
20:47 Rodya_ joined #git
20:47 e14 joined #git
20:49 austin987 joined #git
20:51 Darren_ joined #git
20:53 alhariel joined #git
20:53 Atm0spher1c joined #git
20:55 venmx joined #git
20:57 Vortex34 joined #git
21:00 jccn joined #git
21:01 msafi joined #git
21:01 m0viefreak joined #git
21:01 invisbl_ joined #git
21:04 ToBeCloud joined #git
21:04 msafi Hey folks, I think I'm having issues with this git repo submodules. Basically, other members in the team when they want to switch branches, they delete the repo and reclone it, switch to the branch they want, and then do `git submodule update --init --recursive` to ensure they get the submodules that match the branch
21:04 msafi I'm wondering if there's a way to get the right submodules after switching branches without having to re-clone the repo?
21:04 _ikke_ hmm, that's an odd workflow
21:04 _ikke_ git submodule update after git checkout should be enough
21:05 _ikke_ (and deleting repositories is for me by far the last solution I would take
21:05 _ikke_ )
21:05 isysd joined #git
21:05 moritz I did remember switch branches being painful, when for example subrepo remote URLs changed, or stuff like that
21:07 msafi Okay, thank you. I just wanted to confirm
21:07 theoceaniscool joined #git
21:07 cdown joined #git
21:07 _ikke_ It used to be hard when one branch had the submodule, and the other branch didn't
21:08 dingsundbums joined #git
21:08 _ikke_ but that got fixed a couple of years ago
21:08 invisbl joined #git
21:09 dingsundbums joined #git
21:09 e14 joined #git
21:10 cdg_ joined #git
21:10 ktoverso joined #git
21:10 howitdo joined #git
21:12 e14 joined #git
21:15 gareppa joined #git
21:15 dreiss joined #git
21:16 PettanShoutaKun joined #git
21:18 PettanShoutaKun we have a company private repo thing that we are adding a pre commit hook too. I thought I could just 'git add .git/hooks/pre-commit' to make the hooks run when people push, but such is not the case. What is the best way to as automagically as possible put git hooks in place after you checkout a repo? Also this repo will only be cloned and worked with on windows machines.
21:19 _ikke_ PettanShoutaKun: git does not have any mechanisms to automatically deploy hooks
21:19 _ikke_ out of security
21:19 gareppa joined #git
21:19 _ikke_ You could provide a install_hooks script or something
21:19 _ikke_ But people will have to manually execute it
21:21 tmg joined #git
21:21 PettanShoutaKun so all that script does is copy the hook to their ./.git/hooks/ folder?
21:21 PettanShoutaKun or put a symlink there to the script located elsewhere?
21:22 _ikke_ whatever floats your boat
21:23 MuffinMedic joined #git
21:24 e14 joined #git
21:24 kulelu88 joined #git
21:26 shgysk8zer0 joined #git
21:28 Guest32_ joined #git
21:28 e14 joined #git
21:30 venmx joined #git
21:31 bronson joined #git
21:32 jacob123 joined #git
21:33 Derderderd joined #git
21:33 e14 joined #git
21:34 whomp__ joined #git
21:35 e14 joined #git
21:37 sagerdearia joined #git
21:39 rnsanchez_ joined #git
21:42 b100s joined #git
21:45 e14 joined #git
21:45 leftriver49 joined #git
21:46 binBlob joined #git
21:47 binBlob Hi, how can I skip the editing of a commit message while executing git rebase --continue?
21:48 fatalhalt joined #git
21:50 WaReZ joined #git
21:52 snowkidind joined #git
21:52 e14 joined #git
21:52 Gsham joined #git
21:52 ochorocho__ joined #git
21:52 jeffreylevesque joined #git
21:53 enckse joined #git
21:54 storrgie joined #git
21:54 Vampire0 binBlob, --no-edit
21:55 Rodya_ joined #git
21:56 e14 joined #git
21:58 binBlob Vampire0: I also thought that that should be the command, but no it is not
21:59 binBlob error: unknown option `no-edit'
22:00 e14 joined #git
22:00 binBlob if I use -no-edit instead of --no-edit git even asks: error: did you mean `--no-edit` (with two dashes ?)
22:01 venmx joined #git
22:02 aw1 joined #git
22:02 crawler joined #git
22:02 Seveas dodgy hack that I think should work: GIT_EDITOR=: git rebase --continue
22:04 binBlob that worked
22:04 rominronin joined #git
22:05 binBlob so is the missing option a bug or is there some deeper meaning to this?
22:06 sysadmin joined #git
22:07 nowhere_man joined #git
22:08 commande` joined #git
22:10 ktoverso joined #git
22:12 Rodya_ joined #git
22:13 [Saint] joined #git
22:14 dreiss joined #git
22:14 Rodya_ joined #git
22:15 manuelschneid3r joined #git
22:19 PettanShoutaKun I have a branch on my repo called 66962 I want to setup a hook to watch this branch... I notice by default the hook looks at */master so do I want to watch */66962 ? I noticed that a stack overflow question the guy looks at **feature/* to build his feauture branch
22:19 PettanShoutaKun what am I looking for here?
22:20 PettanShoutaKun what exactly am I trying to wildcard?
22:20 Keytap joined #git
22:20 whomp_ joined #git
22:20 diogenese joined #git
22:20 johnmilton joined #git
22:22 Keytap Evening gents.  Quick question for you.  When you do something like `git checkout HEAD~1` or `git checkout "$BRANCH"@'{'"$DATE"'}'`, what is that called?
22:22 Keytap I'm trying to find a way to checkout the previous commit on a specific branch relative to the commit I have currently checked out.
22:27 ktoverso_ joined #git
22:29 catsup joined #git
22:30 invisbl joined #git
22:30 sea-gull_ joined #git
22:30 invisbl joined #git
22:32 jknetl joined #git
22:32 Hello71 man gitrevisions
22:32 gitinfo the gitrevisions manpage is available at http://jk.gs/gitrevisions.html
22:32 whomp_ joined #git
22:33 holgersson joined #git
22:34 roygbiv joined #git
22:36 catsup joined #git
22:37 Gsham joined #git
22:39 ktoverso joined #git
22:39 Keytap Though I would still like to know what that's called for future reference, I just would up resorting to a quick and dirty bash script to loop a counter to iterate through the commits.
22:42 catsup joined #git
22:42 jeffreylevesque_ joined #git
22:42 holgersson_ joined #git
22:42 invisbl joined #git
22:43 diogenese joined #git
22:44 whomp_ joined #git
22:44 Noodlewitt joined #git
22:46 moei joined #git
22:48 nettoweb joined #git
22:48 crawler joined #git
22:48 catsup joined #git
22:51 SwiftMatt joined #git
22:52 venmx joined #git
22:54 overlord_tm joined #git
22:55 catsup joined #git
22:59 thiago_ joined #git
22:59 xaviergmail joined #git
23:00 whomp_ joined #git
23:01 fmeerkoetter joined #git
23:01 catsup joined #git
23:04 askb joined #git
23:04 MeltedLux joined #git
23:08 catsup joined #git
23:08 kulelu88 joined #git
23:11 Dougie187 left #git
23:12 diluted joined #git
23:14 catsup joined #git
23:15 [Saint] joined #git
23:16 finalbeta1 joined #git
23:16 nowhere_man joined #git
23:19 whomp_ joined #git
23:20 catsup joined #git
23:20 Guest32_ joined #git
23:21 mikecmpbll joined #git
23:22 nowhereman joined #git
23:26 catsup joined #git
23:31 VoidFox joined #git
23:32 cluelessperson left #git
23:32 bronson joined #git
23:32 catsup joined #git
23:32 GodGinrai joined #git
23:33 pwnz0r joined #git
23:34 mellernoia joined #git
23:36 Anja joined #git
23:36 cdown joined #git
23:38 fayimora_ joined #git
23:38 catsup joined #git
23:39 inflames joined #git
23:39 snowkidind joined #git
23:44 jimi_ joined #git
23:44 paolodedios joined #git
23:44 catsup joined #git
23:51 catsup joined #git
23:51 dling joined #git
23:53 eljefeloco_ joined #git
23:53 paolodedios joined #git
23:54 LiMp^ joined #git
23:56 catsup joined #git
23:58 arescorpio joined #git

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