Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2016-12-28

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

All times shown according to UTC.

Time Nick Message
00:00 martinjlowm rizonz: say, merge x commits into one. Could be a number of commits on a feature branch that you want to merge or rebase onto the production branch
00:01 kadoban dshap: Might be .git/objects/pack/ though IIRC they're much less easily edited/seen in there. I don't recall the format.
00:01 martinjlowm instead of having many small commits that describe a subset of a larger change, you can squash them into one commit describing all the changes in detail
00:01 dshap kadoban: roger that. thx!
00:01 bronson joined #git
00:03 rizonz martinjlowm: yap, but I need to look more into branches, now I mostly let my app update it;s code from trunk when using SVN but I think I can use versions/branches now
00:03 rizonz I deploy my app from my repos
00:04 psyb0t joined #git
00:04 cdown joined #git
00:05 Sasazuka joined #git
00:05 martinjlowm sure :)
00:05 rizonz martinjlowm: some people think it's stupid but I think it's ideal!
00:06 rizonz why have some deploy management when git or SVN could do as well
00:06 Noldorin joined #git
00:06 toothe joined #git
00:07 jimi_ joined #git
00:07 toothe are there any extensions I can add to my shell (ksh) to show what branch I am in?
00:07 marcogmonteiro joined #git
00:07 meadmoon joined #git
00:07 ertes rizonz: good enough for simple scenarios, but it's not necessarily ideal if you have dependencies, target-specific peculiarities, parametric deployment, etc.
00:10 GavinMagnus left #git
00:10 joki joined #git
00:10 martinjlowm toothe: are you forced to use ksh?
00:11 rizonz ertes: how so ? I use pull it using puppet and go from there
00:11 rizonz done
00:11 howitdo joined #git
00:11 ertes rizonz: ok then; i thought you use only git =)
00:11 rizonz ertes: no!
00:11 rizonz haha
00:11 rizonz sorrry
00:12 rizonz I can't manage 300 servers without foreman/puppet
00:13 martinjlowm toothe: otherwise, there are plenty of options for zsh
00:13 rtm1046 joined #git
00:14 martinjlowm maybe liquidprompt works for ksh, I'm not sure though. You can check that out
00:16 askb joined #git
00:16 toothe no, I am not.
00:16 toothe I just happened to be using it on FreeBSD. I could always switch to bash.
00:16 kyan joined #git
00:17 metachr0n joined #git
00:18 martinjlowm okay, at least, liquidprompt works for bash. You can see how it looks on its github page
00:18 toothe what are the options?
00:19 toothe oh, liquidprompt.
00:19 toothe okay!
00:19 toothe thank you
00:20 rivarun joined #git
00:21 musicmatze1 joined #git
00:22 BSaboia joined #git
00:23 Padawan joined #git
00:25 dsdeiz joined #git
00:25 dsdeiz joined #git
00:25 nuck joined #git
00:25 Rodya_ joined #git
00:26 dshap Hmm, does anyone know -- I used the BFG repo cleaner to remove all traces of a file and I pushed to Github per the instructions, but now I'm trying to see how a copy of that file could still be accessed via the read-only pull request refs that Github maintains. I tried checking out an old pull request, back when the (now deleted) file existed, but it's not showing up. Does that file actually have to be modified in the pull request for this issu
00:26 dshap apply?
00:27 austin987 joined #git
00:28 bakers dshap, What are you trying to do? Remove all traces a file existed in the repo?
00:28 bakers I had to do that once (had passwords in a file)... fixing the repo locally by doing a rebase, and then a git push --force worked for me
00:28 dshap bakers: Yep, which I believe I did, but then I read about a caveat about these pull request refs and I'm trying to verify that it's an issue. I'm trying to verify that I have to contact Github support to delete my pull requests
00:29 dshap bakers: since those refs are read-only
00:29 tmsmith joined #git
00:29 bakers Is this your repo, or a remote repo?
00:29 bakers Where do the pull requests live
00:31 dshap pull requests live on Github
00:31 dshap it's a Github thing
00:31 dshap I'm chatting with someone in #github, which might be more appropriate for this
00:32 dshap basically they are read-only refs
00:32 bakers Ah ok... probably
00:32 dirtyroshi joined #git
00:32 dshap so once I use BFG locally, i can't push to them on my remote
00:32 dshap like i can with all other branches/refs
00:32 watabou joined #git
00:34 cqi joined #git
00:34 tmsmith joined #git
00:39 Lyqyd joined #git
00:39 Sasazuka_ joined #git
00:39 Rodya_ joined #git
00:40 fstd joined #git
00:42 dling` joined #git
00:43 Raed joined #git
00:43 esc joined #git
00:44 HelloShitty joined #git
00:44 hramrach joined #git
00:44 Spinkzeit joined #git
00:44 iamtakingiteasy joined #git
00:44 rawtaz joined #git
00:44 quite joined #git
00:45 denimsoft joined #git
00:46 nettoweb joined #git
00:47 dreiss joined #git
00:50 Isla_de_Muerte joined #git
00:53 Padawan I have this local repo which has some submodules of github projects. Now, i need to make changes on one of these submodules, i can do that, but when i try to push my local to remote and clone it to another machine... the submodule doesnt get in, and i get errors. What am i missing
00:55 jimi_ joined #git
00:56 arescorpio joined #git
00:56 satbyy left #git
00:56 AciD` joined #git
01:02 bronson joined #git
01:03 Mathieu_Du joined #git
01:06 sbeller Padawan: the superproject needs to have an updated pointer to the submodule, i.e. in the superproject: git add <submodule> (with the correct new sha1 checked out in the submodule)
01:07 rwb joined #git
01:09 moei joined #git
01:18 dodobrain joined #git
01:20 dsdeiz joined #git
01:20 dsdeiz joined #git
01:21 Rodya_ joined #git
01:21 endiruna joined #git
01:26 dar123 joined #git
01:28 theskillwithin joined #git
01:29 tmcmahon joined #git
01:30 [log] joined #git
01:32 nickabbey joined #git
01:33 arescorpio joined #git
01:35 endiruna joined #git
01:37 a_thakur joined #git
01:38 Sasazuka joined #git
01:43 a_thakur joined #git
01:46 JamesBaxter joined #git
01:46 fahadash joined #git
02:00 Sasazuka_ joined #git
02:03 bronson joined #git
02:06 finalbeta joined #git
02:08 unreal_ joined #git
02:09 aw1 joined #git
02:12 endiruna joined #git
02:12 rtm1046 joined #git
02:12 andrew710 joined #git
02:14 Rodya__ joined #git
02:15 grumble joined #git
02:17 NeverDie joined #git
02:18 cagedwisdom joined #git
02:18 netj joined #git
02:23 pks joined #git
02:23 muuse joined #git
02:24 sdothum joined #git
02:29 manuelmrtz joined #git
02:29 d^sh joined #git
02:32 metachr0n joined #git
02:32 cyphase joined #git
02:33 looper joined #git
02:35 jw4242 joined #git
02:37 marlinc joined #git
02:40 Vampire0 joined #git
02:41 franks-reich joined #git
02:43 cqi joined #git
02:46 manuelmrtz joined #git
02:48 ilbot3 joined #git
02:48 Topic for #git is now Welcome to #git, the place for git help and serious business | Public logs at http://jk.gs/glog | First visit? Read: http://jk.gs/git | Current stable version: 2.11.0 | Getting "cannot send to channel"? /msg gitinfo .voice | The git-jokes project has been suspended for lack of contribution. Help revive it! https://madeitwor.se/git-jokes
02:52 Goplat joined #git
02:52 aavrug joined #git
02:54 ayjay_t omg ntfs folders showing up as 0 bytes and giving me weird permissions, what do
02:54 ayjay_t git no work
03:02 beecushm_ joined #git
03:03 moestevens joined #git
03:04 bronson joined #git
03:06 NeverDie joined #git
03:07 chachasmooth joined #git
03:26 unreal joined #git
03:28 Vortex35 joined #git
03:29 Nilesh_ joined #git
03:32 fk joined #git
03:34 justanotheruser joined #git
03:36 jstimm joined #git
03:40 ayjay_t joined #git
03:41 hexagoxel joined #git
03:41 raedah joined #git
03:42 raedah I have serveral branches that each contain a patch I am submitting upstream as an individual pull request. How do I make a new branch that has the changes from all of my branches so that I can run it and test it?
03:43 Lyqyd joined #git
03:45 phroa man git-cherry-pick to copy each commit, then optionally squish them together with man git-rebase /INTERACTIVE
03:45 gitinfo the git-cherry-pick manpage is available at http://jk.gs/git-cherry-pick.html
03:45 gitinfo the git-rebase manpage is available at http://jk.gs/git-rebase.html
03:47 tmsmith joined #git
03:51 sunri5e joined #git
03:51 Emperor_Earth joined #git
03:54 a_thakur joined #git
03:54 meloc joined #git
03:54 ToBeCloud joined #git
03:55 mizu_no_oto joined #git
03:57 raedah cool, thats working
04:00 denimsoft joined #git
04:02 looper joined #git
04:05 bronson joined #git
04:05 justanotheruser joined #git
04:07 rtm1046 joined #git
04:08 chachasmooth joined #git
04:08 nd joined #git
04:10 raedah how do I remove a cherry-pick I dont want anymore? delete a commit
04:12 [log] joined #git
04:13 raedah delete the branch and start over maybe
04:18 fstd_ joined #git
04:18 xall joined #git
04:23 phroa raedah: git rebase -i
04:23 phroa man git-rebase
04:23 gitinfo the git-rebase manpage is available at http://jk.gs/git-rebase.html
04:25 JanC_ joined #git
04:25 nowhere_man joined #git
04:25 dsdeiz joined #git
04:26 cyphase joined #git
04:27 dave0x6d joined #git
04:33 fatalhalt joined #git
04:38 netj joined #git
04:46 subhojit777 joined #git
04:48 slash_ joined #git
04:49 slash_ Hey guys, what would happen if I pull a feature branch1 to local, rebase feature branch2 on top of feature branch 2 and try to push it?
04:49 slash_ *on top of feature branch 1
04:49 preaction try to push it where?
04:50 ayogi joined #git
04:50 kadoban slash_: It's unclear what you're asking exactly, but ... not much interesting?
04:50 preaction you'd have rewritten feature branch 2, causing it to have all new commits in it, which means if you pushed the old fb2 anywhere, you'd have to force-push the new fb2 there if you wanted to push it
04:51 slash_ preaction: I've never pushed fb2 previously. This is only on local.
04:51 preaction then you'd push fb2
04:51 slash_ kadoban: I'm sorry if it's not interesting enough for you.
04:51 slash_ preaction: Is that the recommended workflow if I had to work on top of an existing feature branch?
04:52 preaction that isn't what kadoban meant, i'm sure.
04:52 kadoban (it's not)
04:52 preaction also, as far as i know, it's the only way to work "on top" of an existing feature branch, unless you meant merging the branches instead, at which point it's up to your team
04:53 preaction i prefer to keep a linear history, so i rebase often (and force-push often). other teams prefer to merge instead
04:57 slash_ kadoban: Sorry for the abrasiveness then :)
04:58 slash_ preaction: I prefer linear history as well. So since I'd be working on fb2 alone, would what I've done currently work fine
04:58 kadoban No worries, sorry for giving the wrong impression
04:58 preaction i mean, if you want to know if you like how the history looks, use git log and look at the history
04:58 preaction consider using the --one-line and --graph options if you want to see if it's linear or not
05:00 preaction rebasing in this way will create a linear history, if that's what you're asking
05:01 slash_ I see a really long lie when I use --graph. That's linear? How would branched history look?
05:02 slash_ *line
05:06 bronson joined #git
05:09 dreiss joined #git
05:11 duderonomy joined #git
05:12 holodoc joined #git
05:14 Cabanossi joined #git
05:14 holodoc I remember being able to clone a git repo into a file and then unpack it on another location. Could anyone point me in the right direction what I might have used back then? I don't think it was git-archive.
05:15 kadoban holodoc: Sounds like git bundle
05:16 holodoc kadoban, Yes! git bundle!
05:16 kadoban :)
05:17 holodoc Can't believe I wasn't able to remember that effing name
05:17 kadoban It doesn't get used too much, heh.
05:17 holodoc Thanks kadoban!
05:17 kadoban Though when you need it, it works great. Anytime.
05:17 holodoc I need it to move a large repo
05:17 holodoc And I don't have many other options to do it
05:17 holodoc So I remembered using git bundle for that a long time ago
05:25 svm_invictvs joined #git
05:26 duderonomy joined #git
05:27 nowhere_man joined #git
05:29 Mowgli joined #git
05:32 sbulage joined #git
05:32 nowhere_man joined #git
05:33 Limes_ joined #git
05:38 bocaneri joined #git
05:39 jon-mac joined #git
05:40 nowhere_man joined #git
05:44 shubhnik joined #git
05:53 fatalhalt joined #git
05:57 [log] joined #git
05:59 bannakaffalatta joined #git
05:59 weckstrom joined #git
06:01 zefferno joined #git
06:07 bronson joined #git
06:08 NeverDi__ joined #git
06:10 ogny joined #git
06:12 justanotheruser joined #git
06:12 duderonomy joined #git
06:15 BrianJ joined #git
06:15 cagedwisdom joined #git
06:19 cyan__ joined #git
06:21 clemf joined #git
06:32 pks joined #git
06:34 Tobarja joined #git
06:36 freimatz joined #git
06:39 overlord_tm joined #git
06:43 DonqKonq joined #git
06:49 wcpan joined #git
06:49 netj joined #git
06:50 wcpan joined #git
06:53 ThomasLocke joined #git
06:53 Ozymandy joined #git
07:01 dminuoso joined #git
07:02 dminuoso joined #git
07:02 dminuoso Hi, I have some github repository that I want to fork for a small modification - however I don't want to be bothered to constantly pay attention to any commits from upstream.
07:03 kadoban Why would you be bothered? I have several forks that I don't use for anything anymore, github doesn't seem to care.
07:03 dminuoso kadoban: Because I still want updates as they come in.
07:04 Sircle_ joined #git
07:04 Sircle_ Hi
07:04 Sircle_ I had a working git with bitbucket. I just added my new gitlab remote by 'git remote add localOrigin Http://path/to/gitlab/localrepo.git'. I see commits in 'activity' page but none in tree/master page or commits/master or /project page. Why is that?
07:04 Sircle_ I still see git instructions on the project page
07:04 dminuoso kadoban: My dependency tool accepts github repositories, so if I specified by own fork I would not get upstream updates unless I merged them myself all the time. If I specified the upstream I would not get my own modification.
07:05 kadoban dminuoso: Then get upstream to take your change and maintain it, otherwise what you want doesn't sound possible, or sensible in the git model.
07:06 dminuoso kadoban: That would be interesting - the change is a little extreme (I essentially castrate a library to mock out half of its features to dramatically speed things up) so Ill see how it goes.
07:06 dminuoso Haven't even considered sending the change upstream.
07:08 bronson joined #git
07:08 Anja joined #git
07:09 * Sircle_ snores
07:10 basiclaser joined #git
07:23 prawnzy I'm oversimplifying but if the singularity happens and whole-body gene therapies become a thing you can reset to whatever you are now
07:23 prawnzy or when you iced your cells
07:23 prawnzy oops wrong channel
07:24 diogenese joined #git
07:25 ayogi what's the difference between "git pull" and "git pull origin master"
07:25 GT4066 joined #git
07:25 ayogi what should be used to deploy in production master branch
07:25 JeroenT joined #git
07:25 dminuoso ayogi: git should not be used for deployment directly at all.
07:25 dminuoso It is not a deployment tool.
07:25 dminuoso !deploy
07:25 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
07:26 ayogi dminuoso: i mean i am updating the code in the production
07:26 ayogi how do i update code, i need to do pull?
07:26 dminuoso ayogi: Again, it is not a deployment tool.
07:27 kadoban ayogi: Certainly anything that involves a pull into production code sounds insane.
07:27 dminuoso ayogi: We are running a setup that for each deployment run checks out the entire application into a separate directory with a unique name, and then use symlinks to switch between releases.
07:27 kadoban If this is a website for instance, that's just going to break things all the time ...
07:27 dminuoso We then keep the last 5 deployed versions around, allowing us to instantly roll back between versions.
07:28 davideb joined #git
07:28 ayogi dminuoso: okay. but for now please tell me what's the difference
07:28 navidr joined #git
07:28 dminuoso ayogi: You mean the difference between using a knife and scissors to cut paper? They are different things.
07:31 ayogi dminuoso: yeah, what's the difference and efficient
07:31 prawnzy In the analogy one is better suited for the task, safer, much more efficient, more can go wrong with the other one
07:32 prawnzy not just efficiency
07:32 prawnzy if you don't care about the trade-offs and just want to cut paper, keep using the knife, but know the scissor exists and you should probably use it
07:32 dminuoso ayogi: git is in no way designed to facilitate deployment. You will just get conflicts and issues. Just imagine at some point you decide to rewrite history for whatever reason.
07:32 dminuoso It is really easy to build a simplistic deployment tool around git.
07:32 dminuoso even if its a small bash script akin to what we do
07:34 howitdo joined #git
07:34 howitdo joined #git
07:36 ayogi dminuoso: i can not find any difference between "git pull" and "git pull origin master" after reading: http://stackoverflow.com/questions/23953125/diffe​rence-between-git-pull-and-git-pull-origin-master
07:36 ayogi if default is set
07:36 skered joined #git
07:36 aavrug joined #git
07:37 skered Can git-format-patch generate a patch that isn't/won't be used for email submission or would that just be git-diff output?
07:38 skered ex: can git-format-patch generate a patch minus the email headers.  Bugzilla recognize git-format-match attachments as mbox files.
07:38 Junior joined #git
07:38 skered recognizes*
07:40 duderonomy joined #git
07:44 xall joined #git
07:46 Sircle_ I had a working git with bitbucket. I just added my new gitlab remote by 'git remote add localOrigin Http://path/to/gitlab/localrepo.git'. I see commits in 'activity' page but none in tree/master page or commits/master or /project page. Why is that? I still see git instructions on the project page
07:46 davideb left #git
07:53 lu_zero joined #git
07:54 Mannn joined #git
07:56 dave0x6d joined #git
07:57 lu_zero joined #git
07:57 lu_zero joined #git
08:02 a_thakur joined #git
08:09 bronson joined #git
08:10 freimatz joined #git
08:11 lordjancso joined #git
08:13 somaReverse joined #git
08:15 PCatinean joined #git
08:17 _rht joined #git
08:17 nidr0x joined #git
08:17 grawity skered: does that actually cause problems with bugzilla?
08:17 grawity the format happens to be the same as consumed by `git am`
08:22 sarbs joined #git
08:22 Anja joined #git
08:22 aidalgol joined #git
08:29 Anja joined #git
08:31 shubhnik joined #git
08:33 mattp_ joined #git
08:35 Sadeukko joined #git
08:40 cagedwisdom joined #git
08:45 Balliad joined #git
08:45 Panacea joined #git
08:45 Balliad joined #git
08:46 SteffanW joined #git
08:49 overlord_tm joined #git
08:50 skered grawity: No, because you just change it to be a patch
08:50 skered But I would suspect if it didn't have the email headers it would auto detech it as a patch.
08:50 skered Just added extra step if you attach a patch when you create a bug
08:51 Silox| joined #git
08:51 tvw joined #git
08:51 bremner Sircle_: I guess you need to find a gitlab specific support channel/forum/mailing-list
08:52 skered But if I could get that same output minus the headers via a simple flag to format-patch or via a .gitconfig alias I would use that instead of git-format-patch
09:05 NimeshNeema joined #git
09:06 cqi joined #git
09:07 LarArT joined #git
09:07 jnavila_ joined #git
09:07 clmsy joined #git
09:09 bronson joined #git
09:14 marcogmonteiro joined #git
09:17 robattila256 joined #git
09:18 fbwnd joined #git
09:18 t0by joined #git
09:19 danslo joined #git
09:26 osse skered: won't that be the same as git show?
09:26 osse hmm, never mind. but i think a custom --format to git show/log is the way to go
09:30 cartman joined #git
09:32 Macaveli joined #git
09:34 grawity skered: but then the committer would need to do extra steps to copy the message/authorship/etc.
09:38 jstein_ joined #git
09:41 Junior joined #git
09:42 MrLawrence joined #git
09:43 skered grawity: Well for this case I don't think it's git on the otherside.
09:43 Panacea joined #git
09:43 skered So losing the meta data isn't that big of a deal.
09:45 skered I think I might bark up the bugzilla tree to see what can be done to understand a git-format-patch patch or better options during bug creation.
09:45 ferr1 joined #git
09:47 nisseni joined #git
09:51 robotroll joined #git
09:52 anddam I have a project remote called 'upstream' and a copy of mine at origin, 'git log upstream/master..master' shows only one log that means that if I push on upstream I'm only pushing that one commit, is that right?
09:53 anddam that's my understanding but I'm asking as double check since last time I messed up big time
09:54 selckin git push --dry-run
09:54 selckin gitk <copy paste hashes it shows>
09:54 selckin thats what it'll do
09:55 skered osse: git show -u --format=full ...
09:55 kurkale6ka joined #git
09:55 skered Appears nice.
09:57 jnavila_ joined #git
09:57 YuGiOhJCJ joined #git
09:57 skered I think with some shell scripting it could also generate directly to a file.
09:58 skered It good I might have a alterative but I'd like to see if Bugzilla could better support format-patch output.
09:58 skered We'll see what they have to say.
09:58 cdown joined #git
09:59 skered I'd guess the only argument might be that they can't tell the difference between an mbox file and a format-patch file.
10:00 skered Only because they're the samethign.
10:01 nettoweb joined #git
10:01 cdown_ joined #git
10:05 dubing joined #git
10:07 anddam selckin: I see, what's 'gitk'?
10:07 Jebs joined #git
10:09 Jebs Hi ALl
10:09 gitinfo Jebs: 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.
10:10 chll_ joined #git
10:10 bronson joined #git
10:11 aw1 joined #git
10:14 shiriru joined #git
10:16 howitdo joined #git
10:18 tojo joined #git
10:22 absence joined #git
10:24 PtxDK joined #git
10:25 absence https://github.com/julien-t​ruffaut/Monocle/tree/v1.3.2 <- this repo seems to have a tag called v1.3.2, but when i do git tag -l (after git fetch to get the latest changes), v1.3.2 isn't in the list, and i can't do git checkout v1.3.2 either. what could be wrong?
10:25 selckin git fetch --tags
10:26 absence selckin: ah, thanks. why was that necessary? it seems to have fetched all the other tags before i did that
10:27 selckin afaik its not default to fetch them, can be configured
10:29 anddam selckin: was gitk a typo?
10:29 absence weird, i've never done anything explicitly to get tags, and yet i had most of them
10:29 absence oh well, file under "yet another git mystery" :)
10:29 selckin i don't know
10:31 mgedmin the original git clone gets the tags that existed when you cloned, iirc
10:32 mgedmin wait, I'm overthinking this
10:32 jnavila_ joined #git
10:32 mgedmin "By default, any tag that points into the histories being fetched is also fetched"
10:32 mgedmin so, if v1.3.2 points to a commit that's not part of the master branch (assuming you fetched the master branch), it wouldn't get fetched
10:33 Macaveli joined #git
10:35 dimi1947 joined #git
10:36 acetakwas joined #git
10:40 DieguezZ joined #git
10:43 a_thakur joined #git
10:45 watmm Anyone know why a git merge might introduce special characters to filenames/paths that aren't decoded properly (e.g. German letters with umlauts being translated as '\220'), when the characters aren't visible on the source repository?
10:46 selckin because the pc you're running the merge on can't handle them?
10:48 watmm I can't find where they come from though, they don't appear to be intentional
10:48 Jackneill joined #git
10:50 Mowgli joined #git
10:58 bket joined #git
11:02 d0nn1e joined #git
11:04 JeroenT joined #git
11:06 rtm1046 joined #git
11:09 DrRobotic joined #git
11:11 oaao joined #git
11:11 bronson joined #git
11:14 msonntag joined #git
11:15 cmn_ joined #git
11:19 cmn_ joined #git
11:21 netj joined #git
11:24 davimore joined #git
11:26 manuelschneid3r joined #git
11:32 Greg364 joined #git
11:35 ddv left #git
11:36 Seveas which os are you on, and which filesystem is teh repo on?
11:37 JeroenT joined #git
11:38 Seveas also, are these characters part of the path, or part of the output of  git command?
11:39 m0viefreak joined #git
11:39 rahtgaz joined #git
11:43 ok91 joined #git
11:43 leeN joined #git
11:44 rvalue left #git
11:50 cmn_ joined #git
11:51 ertesx joined #git
11:52 xaa joined #git
11:53 absence mgedmin: i guess that's it. useful to know, thanks!
11:53 sostenuto joined #git
11:54 fbwnd joined #git
11:59 Nitesh I sent a patch using git send-mail and then another patch with the same Subject but a v2. i.e. the only difference between the subject lines of two patches was "v2". Still when I sent the second patch, it came in the mailing list as one below the other as if it were the same messages sent twice
12:00 Nitesh Not sure why this happened, also the subject lines came as same for both the patches in the mailing list. Am I missing something/doing something wrong
12:01 kpraveen joined #git
12:01 SwingShock joined #git
12:01 thiago joined #git
12:03 pixaal joined #git
12:03 SwingShock Hi. I was trying to get rid of a commit i did earlier and go back to the commit before it and set it as the master. I tried to use revert <commit> to that commit and i am now stuck at (master|REVERTING). What can i do next ?
12:04 rnsanchez joined #git
12:05 pixaal hey all, i'm having a bit of trouble authenticating to github. trying to clone one of my repos gives a 'Permission denied (publickey)' error, even though i've added the contents of id_rsa.pub to github and doing 'ssh -T git@github.com' shows "Hi <username>! You've successfully authenticated..."
12:06 JeroenT joined #git
12:06 osse pixaal: what is the URL you're trying to use=
12:07 pixaal git@github.com:gregzaal/Gaffer.git
12:07 MenoPiu joined #git
12:08 MenoPiu left #git
12:08 osse pixaal: what does 'git config core.sshCommand' say?
12:08 osse and env | grep GIT
12:08 maialino4 joined #git
12:08 donught joined #git
12:08 maialino4 hey, how can I rebase commits with empty message?
12:08 somaReverse joined #git
12:09 donught joined #git
12:09 pixaal osse, doesn't show anything, but neither does my other pc where it is working fine
12:10 osse pixaal: does git clone https://github.com/gregzaal/Gaffer.git  work ?
12:10 osse and show the output of ''GIT_TRACE=1 git clone git@github...''
12:10 pixaal probably, but i'd rather not have to type in my username & password every time
12:10 SwingShock I learned to reset instead. It is risky but it worked for now. Bye.
12:10 * pixaal will try
12:11 maialino4 osse: yep it works
12:12 pixaal osse, https://hastebin.com/imelujatef.pas
12:12 bronson joined #git
12:13 pixaal cloning https does work
12:14 Jackneill_ joined #git
12:15 kpraveen left #git
12:19 osse pixaal: well that looks suspicious
12:19 osse that thingy before git@
12:19 osse nuke it from orbit
12:21 pixaal indeed
12:21 pixaal but how
12:22 JamesBaxter joined #git
12:22 osse maybe you accidentially typed AltGr-Shift-Escape-ASpace-Ctrl as you were typing "git" then just used "up and enter" ever since
12:22 mgedmin copy/paste error?  one half of a UTF-8 sequence for NO-BREAK-SPACE?
12:22 pixaal ahhhhh
12:22 pixaal yes
12:23 pixaal the old copypasta
12:23 mgedmin which is \xC2\xA0 in UTF-8, but \xA0 in Latin-1
12:23 osse but there is a regular space there too... otherwise he'd get "clone git@..." not a git command
12:24 pixaal yes the must've been some encoding screwup with copy paste, thanks guys o/
12:24 pixaal working now
12:24 pixaal strange that it didn't screw up the https too tho
12:27 osse it can happen to the worst of us. my bet is a pebcak and that it has nothing to do with actual copy-pasting.
12:28 mgedmin as a linux user, I forget that there are ways to input no-break spaces that do not involve copy-pasting
12:29 osse Alt-Space is one I think, at least in gnome. I usually change the keyboard layout to let Space produce a Space no matter what keys I may be pressing at the same time... The amount of time spent on fixing nbsp bugs in latex when typing $ ... GRRRR
12:29 grawity alt-space is the system menu
12:29 grawity shift-space might be NBSP?
12:29 grawity I keep forgetting too
12:30 mgedmin alt-space opens the window menu in gnome
12:30 osse hmm, then it's AltGr-Space
12:30 osse (I belong to the AltGr cult)
12:30 mgedmin altgr-space is just plain space in my Lithuanian layout, but this is getting off topic and I'm sorry
12:31 grawity osse: heathen
12:31 * grawity : Compose user
12:31 grawity ahhh here it is
12:31 grawity there's a whole *ton* of Xkb options for it: https://i.imgur.com/j6dsfeG.png
12:32 osse Anything an op wishes to discuss is on topic by definition!
12:34 pixaal i'm using git bash on windows. never really trusted copy-pasting with terminals on windows in general, but it's gotten better in win10
12:34 uskerine joined #git
12:36 uskerine hi, I am new to git. I have created my repo in bitbucket. I have installed SourceTree and succesfully see my new repo. I added my first source files to that repo and I can see them in SourceTree. I have "Staged" all them. Now what do I have to do?
12:36 selckin commit
12:36 uskerine selckin, I press commit and nothing happens
12:36 selckin read some documentation
12:37 selckin !gui
12:37 gitinfo Graphical user interfaces are not supported here. If you want to get support, it needs to be through the git CLI. Reasons: 1) Because very few people here use the graphical interface. 2) Because giving instructions for GUI's is difficult. 3) The command line gives you a history of what commands you have executed.
12:38 somaReverse joined #git
12:38 uskerine I am asking about conceptual issues of git
12:38 uskerine not about the gui
12:39 thiago uskerine: once you stage, you commit
12:39 thiago or unstage
12:39 thiago after you commit, you can push. Or not.
12:40 selckin which is the first thing any doc/tutorial will go over, you are trying to guess how it works from the gui
12:40 oskarkv joined #git
12:41 andrew710 joined #git
12:43 uskerine thanks thiago
12:43 uskerine what does stage/unstage mean?
12:43 ToBeCloud joined #git
12:44 selckin its not very efficient for us to type out the entire documetnation, read one and then come with specific questions
12:44 thiago uskerine: to put them into the staging area for the next commit or to remove from it
12:45 uskerine understood
12:45 osse pixaal: I definitely suggest you use the mintty and not the regular cmd window thingy
12:46 Padawan_ joined #git
12:51 pixaal osse, i use the git bash thing, not the normal cmd thing
12:52 Panacea joined #git
12:54 OnceMe joined #git
12:54 OnceMe hello, can someone help me with git?
12:54 Muzer joined #git
12:54 OnceMe Im not sure should I checkout new branch
12:54 mgedmin I love mintty except it badly breaks python repl :'-(
12:54 OnceMe because I just have cloned it from github
12:55 pixaal git bash comes with some bash commands to, maybe from cygwin idk.
12:55 thiago from cygwin, correct
12:55 thiago OnceMe: do you want to check out a new branch?
12:55 mgedmin msys2 rather than cygwin these days, afaiu
12:55 OnceMe I dont know
12:56 ojdo joined #git
12:56 OnceMe thiago: I just want to push it
12:56 JeroenT joined #git
12:57 OnceMe but I cloned main git repo, and I have my branch on github which is under blah/tree/myusername
12:57 OnceMe so when I did git add . -A and git commit -m "m"
12:57 OnceMe I did git push origin myusername but it fails
12:57 shams joined #git
12:58 thiago OnceMe: push what?
12:58 thiago ah
12:58 thiago you cloned from the main repo and that's where you've tried to push back to
12:58 thiago I assume you don't have the rights to do that
12:59 OnceMe well I tried with git push origin myusername
12:59 OnceMe I guess it can push only to master, so I need to swtch branches if I'm right?
12:59 shams how to bring changes in a file back?
12:59 thiago you need to first git remote add -f personal git@github.com/myusername/myreponame
12:59 thiago then you can do: git push perosnal master
12:59 thiago shams: explain more, please
12:59 OnceMe thiago: I dont want to push on remote
13:00 OnceMe and I already have my tree, Im added into project
13:00 OnceMe basically I thikn I need to checkuot the branch
13:00 OnceMe I get non-fast forward thing
13:00 thiago OnceMe: you don't need to change your local branch, but you can
13:00 shams thiago, i changed a .c file then did git add,commit,now i want my file should be same as previous
13:01 synthroid joined #git
13:01 thiago you don't need to change branches to push to a different branch
13:01 thiago anyway, your problem is that your local copy is old and doesn't have the latest from upstream. Taht's the "non-fast-forward" thing (which you should have mentioned before).
13:01 thiago git pull again
13:01 thiago shams: do you want to remove the commit you've made?
13:02 thiago shams: or do you want to just rollback the changes to the file? Temporarily or permanently?
13:02 shams thiago, i created a patch but i don't want to send it
13:02 finalbeta joined #git
13:02 OnceMe thiago: didnt fix
13:02 shams i want my .c file as previous
13:02 shams permanently
13:03 shams thiago, are there two methods for doing this temp and permanent?
13:04 thiago shams: yes
13:04 OnceMe thiago: what to do?
13:04 thiago shams: do you want to remove the commit from existence?
13:04 shams yes
13:04 OnceMe do I commit to my branch as git push origin/mybranchusername or?
13:04 thiago OnceMe: obviously something you didn't tell me is not working properly.
13:05 Jackneill joined #git
13:05 thiago OnceMe: what is your local branch called?
13:05 OnceMe how do I find that?
13:05 thiago shams: have you pushed this commit anywhere other people can see?
13:05 thiago OnceMe: git branch
13:05 shams thiago, no its in my local
13:06 OnceMe onceme
13:06 thiago shams: git reset --hard HEAD~
13:06 thiago shams: this will remove the current commit from existence
13:06 thiago shams: and remove any changes in the work tree
13:06 shams thiago , and what about my .c file?
13:06 thiago OnceMe: git push -u origin onceme:remotebranchname
13:06 thiago shams: what about it?
13:07 thiago shams: the reset --hard command will make your work tree look exactly like the commit second to last (which will be the last & current after the command)
13:08 shams thiago,this --reset command will also make my .c file as earlier?
13:08 thiago shams: it will make all your files look like as they were in HEAD~
13:08 OnceMe thiago: can we do it step by step
13:08 OnceMe so I clone: git clone https://git.com/.../blah.git
13:09 OnceMe cd blah
13:09 OnceMe I make change to one file, so how can I see all available remote branches?
13:09 thiago OnceMe: git branch -r
13:09 shams thiago, but when i type git status ,it says HEAD detached
13:09 thiago shams: that's not the reset command. Something else did it.
13:09 rtm1046 joined #git
13:09 OnceMe thanks
13:09 OnceMe I know whats the issue
13:10 OnceMe I tried git push origin myusername when I needed git push origin/myusername
13:10 thiago OnceMe: that's not your issue
13:10 shams thiago, yeah i did previously git checkout <commit_id>
13:10 OnceMe thiago: maybe I need to switch to my branch first?
13:10 OnceMe how can I do that?
13:10 thiago OnceMe: I know that because the second command is incorrect, therefore cannot be the solution
13:10 thiago OnceMe: git checkout otherbranchname
13:10 thiago shams: that does detaching
13:11 OnceMe git checkout origin/myusername ?
13:11 thiago shams: check out your regular branch first, then reset
13:11 stamina joined #git
13:11 thiago OnceMe: no, git checkout myusername
13:11 OnceMe ok Im on it
13:12 OnceMe now I make changes and then what? git add . -A ?
13:12 thiago yes
13:12 shams thiago, how do i bring back the head to commit_id?
13:13 OnceMe and then git commit -m "All good"
13:13 thiago shams: git reset --hard commit_id
13:13 OnceMe and then git push origin myusername ?
13:13 thiago shams: --hard will check out after bringing HEAD to commit_id
13:13 OnceMe is that all good commands?
13:13 thiago OnceMe: yes
13:13 bronson joined #git
13:15 shams thiago, i didn't get you
13:15 Makko_fR left #git
13:19 CheckDavid joined #git
13:19 mindfart joined #git
13:22 SteffanW joined #git
13:25 SteffanW joined #git
13:25 plos1 joined #git
13:25 pixaal joined #git
13:26 shams thiago, done
13:27 shams thiago, one more question
13:29 sdothum joined #git
13:30 shams what is better to follow- to remove your commits permanently and do it again correctly or make you r changes in a file and do commit again without removing the previous commit in you local
13:32 narendraj9 joined #git
13:33 linuxdaemon joined #git
13:35 JeroenT joined #git
13:35 danslo joined #git
13:35 jeffreylevesque joined #git
13:36 shams thiago, which approach you used while contributing upstream?
13:38 desertSniper87 joined #git
13:38 OnceMe how can I avoid commiting/pushing .metadata dir changes in Eclipse from terminal?
13:38 grawity remove the directory from the repo (git rm -rf --cached .metadata), add it to gitignore (echo .metadata >> .gitignore), then add&commit the .gitignore file
13:39 troulouliou_div2 joined #git
13:40 OnceMe so basically: echo ".metadata" > .gitignore && git add . -A && git commit -m "ok" && git push origin username?
13:42 grawity assuming .metadata isn't *yet* tracked by the repo
13:42 OnceMe it is
13:43 grawity then untrack it using the command given above
13:43 Jackneill_ joined #git
13:44 masuberu joined #git
13:44 sobersabre joined #git
13:45 sobersabre hi. is there some kind of version related change that happened and I missed it or is there a default config that changed in how git handles staged vs non-staged changes to a file ?
13:45 sobersabre I remember I could interactively add only selected changes to a file
13:45 JeroenT joined #git
13:45 sobersabre then commit, and the non-staged changes would remain unstaged and uncommitted.
13:45 grawity you used `git add -p` or `git add -i` for that
13:45 Seveas sobersabre: git add -p (or --interactive)
13:45 sobersabre Seveas: I'm talking about how commit behaves...
13:45 sobersabre Something is weird
13:46 j416 sobersabre: what is weird, specifically?
13:46 grawity commit always commits everything that is staged
13:46 Seveas git commit -p :) (which is really add -p + commit)
13:46 grawity well, you can tell it to commit specific files bypassing the staging area
13:46 grawity but I don't think you can do the same at -p level
13:46 Seveas correct, -p manipulates the index
13:47 grawity the whole idea of committing individual chunks seems to kinda require a staging area
13:48 sobersabre j416: when I ran "git commit -s ." I did not see unstaged changes in a file.
13:48 sobersabre when I ran now "git commit -s", I saw only staged changes.
13:48 sobersabre so if I say git commit ., it takes all files not even staged?!
13:48 grawity the former is a case of `git commit <file>`
13:48 grawity which bypasses the staging area
13:48 j416 ^
13:48 sobersabre grawity: I remember it used not to bypass.
13:48 systemovich joined #git
13:48 grawity I don't remember it having ever done so
13:49 Seveas sobersabre: your memory is faulty.
13:49 grawity in fact I think that always has been the whole point of this mode?
13:49 sobersabre Seveas: I now understand what's the mystery, this is already a plus
13:49 sobersabre I am thinking I mentally remember this to work in the past.
13:49 j416 how can you remember something if not mentally?
13:49 sobersabre maybe the change crawled up slowly, and I didn't notice.
13:49 * j416 ponders
13:50 sobersabre I could record a session, or video, and remember having it, that is with a hard copy.
13:50 Seveas j416: well, some people think with other parts than their brain :)
13:50 j416 Seveas: lol
13:50 sobersabre Seveas: that too, aren't we all sometimes?
13:50 j416 sobersabre: fair point
13:50 j416 sobersabre: not Seveas.
13:50 astrofog joined #git
13:50 sobersabre j416: you can't think with what you don't have.
13:51 j416 sobersabre: Seveas is practically made out of brain tissue.
13:51 * j416 runs
13:52 sobersabre j416: you made me this mental(again) image of a certains scene of "Starship Troopers". clearly made by a biased person...
13:52 Seveas /nick Krang
13:52 sobersabre :))))
13:52 j416 heh. I don't think I have seen that.
13:52 Seveas I've only read starship troopers, never seen the movie
13:52 j416 or read it.
13:53 Seveas I've heard the movie described as "so unlike the book, that's it's best described as 'being inspired by the cover of the book'"
13:53 osse Seveas: is the book satirical and a bit black comedy-ish? The movie is like Robocop, basically. dark satire all the way
13:53 wogi joined #git
13:53 Seveas no
13:54 rwb joined #git
13:54 osse i remember thinking it was a bad movie when i was younger. but now i think i've missed much of the point. i need to watch it again
13:54 osse so that I can *grok* it.
13:54 sobersabre Seveas: starship troopers is one of the books I haven't read b/c the movie was too weird. Generally I do like Heinlein.
13:54 Seveas Read the book
13:55 sobersabre Seveas: +1 thanks.
13:57 osse "While the original novel has been accused of promoting militarism, fascism and military rule, the film satirizes these concepts by featuring grandiose displays of nationalism as well as news reports that are intensely fascist, xenophobic, and propagandistic"   this surely wooshed by my 15 year old head
13:58 synthroi_ joined #git
13:59 grawity osse: I wonder if nowadays it'd be written "...attempts to satirize these concepts by featuring grandiose displays of nationalism as well as news reports that are intensely fascist, xenophobic, and propagandistic, but falls short due to Poe's law"
13:59 afuentes joined #git
13:59 osse haha, could be
14:00 SwingShock joined #git
14:00 osse But Robocop still holds its ground, and it's the same director.
14:00 leeN joined #git
14:01 shinnya joined #git
14:01 SwingShock When i opened the git folder with git gui after a long time since i been using bash the whole time, it shows there are lots of loose objects and it needs to be compressed to maintain optimal performance. Is this adivasable to do ?
14:02 osse SwingShock: you can do it, but git usually does it by itself
14:02 SwingShock can it be checked and done from time to time with git bash instead ?
14:04 rnsanchez_ joined #git
14:05 Seveas SwingShock: just run 'git gc'
14:06 Asenar joined #git
14:06 SwingShock That worked nice. It didn't ask any confirmations too.
14:07 osse git runs git gc --auto by itself sometimes. that's usually more than enough
14:07 osse when you see that message you mentioned that usually means git does it then and there
14:07 SwingShock I had 320 loose objects.
14:08 osse i think the automatic gc only kicks in after passing 5000
14:08 acetakwas joined #git
14:09 SwingShock So it's really unnecessary to do. Then it was shown in git gui maybe due to it's lower check limit there.
14:11 Jackneill_ joined #git
14:12 NeverDie joined #git
14:12 khfeng_ joined #git
14:13 cdg joined #git
14:14 bronson joined #git
14:14 synthroid joined #git
14:15 HoierM joined #git
14:16 cdown joined #git
14:18 plos1 joined #git
14:25 dynamic joined #git
14:26 pbrewczynski joined #git
14:26 ShekharReddy joined #git
14:27 glebihan joined #git
14:27 treehug88 joined #git
14:28 Spascho joined #git
14:28 al-damiri joined #git
14:28 jimi_ joined #git
14:28 nidr0x joined #git
14:29 ExoUNX joined #git
14:30 Jackneill_ joined #git
14:31 AlineGomes joined #git
14:31 nidr0x joined #git
14:33 relipse joined #git
14:33 Es0teric joined #git
14:33 relipse I deleted a whole directory, how do I tell git to restore all the deleted files "Not staged for commit"
14:34 xall joined #git
14:36 cyphase joined #git
14:36 f3r70rr35f joined #git
14:37 nowhere_man joined #git
14:42 systemovich `git checkout -- directory/`
14:43 denisMone joined #git
14:45 ac3takwas joined #git
14:45 sostenuto joined #git
14:46 glebihan joined #git
14:47 JamesBaxter joined #git
14:48 cdown joined #git
14:48 sostenuto joined #git
14:50 Panacea joined #git
14:50 delboy1978uk joined #git
14:50 delboy1978uk hi everyone, some work got pushed to the wrong branch, and otrhers have since committed. whats the easiest way to remove a branch that was merged?
14:51 selckin !revert_merge
14:51 gitinfo [!faulty_merge] If you have published a merge commit that turned out to be faulty and you'd like to get rid of it, you're probably not going to have a good time. Here's the full story: https://raw.github.com/git/git/master/Docu​mentation/howto/revert-a-faulty-merge.txt
14:53 sharpy joined #git
14:54 ispn joined #git
14:55 rtm1046 joined #git
14:58 ThomasTang joined #git
14:58 Lyqyd joined #git
14:59 cfoch joined #git
15:01 cfoch Hi. If I restore a file using "git checkout 45343^ -- path/to/file" and then I commit it, will blame tell that the author of all the lines was the previous original authors or I?
15:01 selckin i
15:02 cfoch selckin: is there a way to recover the file but keeping the original authors
15:02 cfoch ?
15:02 delboy1978uk selckin: thanks very much for that, highly informative
15:02 zefferno joined #git
15:03 selckin cfoch: its generally not usefull to look at blame without context
15:03 delboy1978uk cfoch: if you commit it, you ae now responsible.
15:03 Rapture joined #git
15:03 selckin delboy1978uk: no.
15:04 delboy1978uk wont it show him in blame?
15:04 ThomasTang joined #git
15:04 selckin that does not make you responsible
15:05 delboy1978uk hmm, could disagree with that. guess it depends on office protocol
15:05 selckin if you rename a method, are you responsible for every usage of it?
15:05 delboy1978uk theoretically though, if the code has your name on it..
15:05 selckin the team owns the project
15:06 delboy1978uk selckin: then you'd be responsible for line x
15:06 selckin having code that is owned by a person is the biggest sign of a disfunctional org
15:06 cdown_ joined #git
15:06 delboy1978uk i totally get you
15:06 newbie1 joined #git
15:06 Noldorin joined #git
15:06 delboy1978uk but git blame helps show who committed the code, then you can speak to team member x and ask why
15:07 NeverDie_ joined #git
15:07 delboy1978uk thats all my point is
15:07 selckin and you look at the commit mesage and see "revert x", and blame the parent commit
15:07 selckin or "rename x"
15:07 selckin you know generally be a functional person
15:07 delboy1978uk lol
15:08 cfoch How can I revert an old file from a previous commit but keeping the authors for each line?
15:08 selckin you can't
15:09 delboy1978uk unless... could he rewind to a point where it was still in existance? then branch off, and cherry pick in the new stuff?
15:09 selckin you can rewrite history to anything you want
15:09 delboy1978uk then merge that in the branch that doesnt have the file?
15:09 ThomasTang joined #git
15:09 delboy1978uk wouldnt that keep the blame history?
15:10 selckin if you rewrite history as if the commit never happened yes offcourse
15:10 aavrug joined #git
15:14 ogny joined #git
15:15 bronson joined #git
15:18 glebihan joined #git
15:20 dsdeiz joined #git
15:20 aavrug joined #git
15:21 ertes can't you revert deletes?
15:21 Seveas anything you commit can be reverted
15:23 synthroi_ joined #git
15:23 shams joined #git
15:24 Seveas cfoch, this may work, but is untested. Do it in a scratch copy of the repo and make sure you understand what the commands do before trying them: git checkout -b temp-branch last-commit-that-had-the-file; git checkout master -- .; git checkout HEAD -- path/to/file-to-restore; git commit -m "Restore path/to/file"; git checkout master; git merge temp-branch
15:25 Seveas i.e., create temporary branch at the point the file existed, then copy the contents of master into it, but keeping the deleted file, then merge that temporary branch into master.
15:25 hashpuppy joined #git
15:25 delboy1978uk left #git
15:26 selckin thats gonna be your name
15:27 snowkidind joined #git
15:28 Seveas git reset --hard selckin
15:28 selckin its gonna show what committer/author is when you commit in that flow
15:28 selckin your 5th command
15:29 danslo joined #git
15:31 aavrug_ joined #git
15:31 Seveas the commit we create changes all files (potentially). Except the one we're restoring, as it's been reset to the current contents
15:31 Seveas so it shouldn't affect blame
15:31 Seveas but as I said, it's untested. And I can't test it now as I need to run to a train :)
15:33 ertes cfoch: perhaps reverting the commit that deleted the file could keep the authors
15:34 aavrug_ left #git
15:38 nettoweb joined #git
15:39 anonymuse joined #git
15:42 YuGiOhJCJ joined #git
15:46 nettoweb joined #git
15:46 Jackneill_ joined #git
15:48 dviola joined #git
15:48 cyan__ joined #git
15:49 narendraj9 joined #git
15:51 nettoweb1 joined #git
15:53 shams joined #git
15:53 rts-sander what do I do when a tag does not exist on a certain repository?
15:54 rts-sander I pushed the tag, I pulled with fetch --tags but the tag is not there
15:54 zefferno joined #git
15:55 freimatz3 joined #git
15:56 grawity check with `git ls-remote` whether it actually exists or not
15:56 narendraj9 joined #git
15:56 rts-sander it doesn't
15:56 rts-sander I removed a random tag and did git fetch --tags and it only retrieved that tag again
15:57 subhojit777 joined #git
15:58 Mowgli joined #git
15:59 rts-sander it's a problem with the remote, I got it
15:59 grawity well, if `git ls-remote` doesn't show it, then fetching isn't going to retrieve it either
15:59 grawity either you pushed it wrong, or the remote "forgot" it somehow
16:00 grawity maybe try pushing again, specifically to refs/tags/<thename>
16:00 nettoweb joined #git
16:00 rts-sander the remote points to a directory on the filesystem instead a server URI
16:00 grawity shouldn't matter
16:02 rts-sander why have a repository point to another repository on the filesystem that points to the actual server URI?
16:02 jnavila joined #git
16:03 selckin because thats what you did?
16:03 Cavallari joined #git
16:04 rts-sander coworker, he's on vacation and I'm trying to reverse engineer his automatic deployment system because I broke it by force pushing
16:04 rts-sander but it's all good now
16:04 Shock- joined #git
16:06 grawity sounds like !deploy recipe #4.2
16:06 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
16:07 a_thakur joined #git
16:09 JamesBaxter joined #git
16:10 adbjesus joined #git
16:11 clemf joined #git
16:11 netcarver joined #git
16:12 theskill_ joined #git
16:12 nettoweb1 joined #git
16:15 shgysk8zer0 joined #git
16:16 bronson joined #git
16:18 synthroid joined #git
16:20 pcre joined #git
16:22 thebope joined #git
16:23 Nerbrun joined #git
16:25 doener joined #git
16:35 jnavila joined #git
16:37 Oatmeal joined #git
16:39 manuelschneid3r joined #git
16:39 gtristan joined #git
16:39 Rr4sT joined #git
16:39 dar123 joined #git
16:40 snowkidind joined #git
16:41 porsche944 joined #git
16:43 shubhnik joined #git
16:43 a_thakur joined #git
16:44 shgysk8zer0 joined #git
16:53 shgysk8zer0_ joined #git
16:53 WayToDoor joined #git
16:54 RiPer I have checked out a repo, updated a file, the push a pull request, but it is including 2 other commits that are not part of my edit. My branch and the branch I am pushing to are identical except for the single file I changed, thoughts on why it is including two other commits?
16:54 RiPer checked out = forked oops
16:54 nicksloan joined #git
16:55 cdown joined #git
16:55 nowhere_man joined #git
16:55 jccn joined #git
16:57 jccn joined #git
16:57 zifnab joined #git
16:57 tmsmith joined #git
16:58 pks joined #git
16:58 svm_invictvs joined #git
16:58 Darren_ joined #git
16:58 RiPer ahh figured it out, I created a new branch with my commit, I have to merge to the branch I want to pull request with.
17:01 mda1 joined #git
17:03 freimatz3 joined #git
17:03 xall joined #git
17:04 sq-one joined #git
17:05 synthroi_ joined #git
17:05 PCatinean joined #git
17:06 HarryPothead joined #git
17:08 Goplat joined #git
17:08 duderonomy joined #git
17:09 shubhnik joined #git
17:11 howitdo joined #git
17:12 boombatower joined #git
17:14 slayerjain joined #git
17:15 meloc joined #git
17:17 imachuchu joined #git
17:17 bronson joined #git
17:19 cdown joined #git
17:19 Rapture joined #git
17:20 joki joined #git
17:20 ResidentBiscuit joined #git
17:20 Oebele joined #git
17:23 duderonomy joined #git
17:23 Guest14887 joined #git
17:25 sea-gull_ joined #git
17:26 algun joined #git
17:27 ac3takwas joined #git
17:27 JanC_ joined #git
17:28 Macaveli joined #git
17:29 prestorium joined #git
17:30 prestorium I thought it would be worth to share
17:30 prestorium A developer just did a merge of repo A into repo B.
17:30 prestorium I was speechless
17:30 xall joined #git
17:30 prestorium I have absolutely no idea how he achieved that. :)
17:31 prestorium and I'm afraid to known
17:31 a_thakur joined #git
17:34 maxb It's not that unusual
17:34 grawity probably just did `git pull <wrong_url>`
17:34 grawity Git only recently added a warning about that
17:34 maxb Entirely reasonable if two projects merge into a single repo
17:35 Macaveli_ joined #git
17:35 Torsho_ joined #git
17:36 synthroid joined #git
17:48 kyan joined #git
17:51 jnavila joined #git
17:51 kyan Hi! If I have a commit tagged, then delete that tag, and recreate the tag pointing to a different commit, how can I see the previous revision of the tag? It doesn't show up in the commit history... Thanks!
17:52 grawity tags don't have "revisions"
17:52 grawity more like the opposite, they're something holds revisions (like read-only branches)
17:53 nettoweb joined #git
17:53 kyan Ok... how can I see those changes, then?
17:54 kyan (Like, if I screw something up in the files in a repository, I can see what it used to be, and go back to it. I'd like to be able to fix similar screw-ups to the tags...)
17:55 grawity well, tags are meant to never change in the first place
17:56 kyan Right, but if I (or someone with access to my comp) deletes one and create a new one with that name, it'd change it, I'd think, at least as far as downstream users are concerned
17:57 kyan I could include a listing of tags in the repository
17:57 grawity "someone with access to my computer" is a bigger problem...
17:58 grawity for downstream users which already have the tag, Git won't update it (unless they delete the old one manually)
17:58 tvl joined #git
17:58 kyan Oh, interesting
17:58 kyan OHH, that's what it meant in the manual about how push --tags skips ones they have
17:59 kyan cool, thanks :D
17:59 grawity well downstream users would be fetching them instead, but same general idea
17:59 grawity (and of course if you had a policy of always pgp-signing tags, then "someone"s fake tag would get noticed quicker)
18:00 kyan mm
18:01 grawity but anyway, going back to the original question
18:01 grawity tags are just names attached *to* a revision (commit)
18:02 grawity they could in theory have a local reflog like branches do
18:02 grawity although Git deletes reflogs when you delete a ref, so that wouldn't help much
18:02 grawity logging ref updates *on the server* would help more
18:03 grawity or, for that matter, forbidding tag changes on the server
18:04 kyan thanks!
18:04 BlackMaria joined #git
18:09 Rapture joined #git
18:09 shgysk8zer0_ joined #git
18:10 aw1 joined #git
18:10 thebope joined #git
18:18 bronson joined #git
18:19 ac3takwas joined #git
18:20 duderonomy joined #git
18:21 mrnuke joined #git
18:24 shams joined #git
18:25 NeXTSUN joined #git
18:26 Es0teric joined #git
18:27 wogi joined #git
18:27 nettoweb joined #git
18:29 aarobc joined #git
18:29 morenoh149 joined #git
18:29 morenoh149 how do I checkout a file from a previous revision?
18:31 Spec-Chum joined #git
18:32 rudi_s morenoh149: git checkout id -- path
18:32 rudi_s man git-checkout
18:32 gitinfo the git-checkout manpage is available at http://jk.gs/git-checkout.html
18:38 toothe left #git
18:39 coasterz joined #git
18:39 publio joined #git
18:43 crayon joined #git
18:43 madewokherd joined #git
18:43 jimi_ joined #git
18:44 zeroed joined #git
18:49 sammyo joined #git
18:49 Balliad joined #git
18:52 Padawan joined #git
18:53 ResidentBiscuit joined #git
18:56 Salander27_ joined #git
18:58 weckstrom joined #git
18:59 todd_dsm joined #git
18:59 prestorium joined #git
19:00 Peuc joined #git
19:01 Surfer2010 joined #git
19:03 v-y joined #git
19:03 joshszep joined #git
19:06 Surfer2010_ joined #git
19:06 dsantiago joined #git
19:07 ochorocho__ joined #git
19:09 weckstro_ joined #git
19:10 tyreld joined #git
19:11 weckstro_ joined #git
19:12 Pulp joined #git
19:13 rominronin joined #git
19:14 Learath2 joined #git
19:15 languitar left #git
19:17 snowkidind http://pastebin.com/DTsk0dkr
19:17 snowkidind i deleted a file on my server but the origin has the one i want to be there.
19:17 desertSniper87 joined #git
19:17 Panacea joined #git
19:18 snowkidind and i edited my .gitignore file on the server but i want to sync that with the origin
19:18 jknetl joined #git
19:18 bronson joined #git
19:21 Silenced joined #git
19:21 Silenced joined #git
19:22 Sasazuka joined #git
19:23 guest joined #git
19:25 hahuang65 joined #git
19:26 fission6 joined #git
19:26 fission6 how do i stash an untracked file
19:26 dsdeiz joined #git
19:26 dsdeiz joined #git
19:29 tmsmith joined #git
19:29 hashpuppy kind of confused.  i'm trying to clean up my branches that have already been merged.  i'm doing git push origin :feature1 and then git fetch origin -p and it still shows the upstream branch i deleted
19:29 hashpuppy same thing when i try to git remote | xargs -n 1 git remote prune
19:30 joki joined #git
19:30 duderonomy joined #git
19:30 hashpuppy or what's the best way to delete both at once?
19:32 freimatz joined #git
19:34 cagedwisdom joined #git
19:35 morenoh149 joined #git
19:37 weckstrom joined #git
19:37 GavinMagnus joined #git
19:39 ac3takwas joined #git
19:40 neunon joined #git
19:40 cyan__ joined #git
19:41 WaReZ joined #git
19:42 watabou joined #git
19:44 _czyzewski joined #git
19:44 mda1 joined #git
19:44 _czyzewski left #git
19:44 tvw joined #git
19:46 neunon_ joined #git
19:50 Achylles joined #git
19:54 kyan joined #git
19:54 Panacea joined #git
19:55 ResidentBiscuit joined #git
20:00 BlueProtoman joined #git
20:03 idr0p joined #git
20:05 grumble joined #git
20:07 Darcidride joined #git
20:12 Goplat joined #git
20:16 kyan joined #git
20:17 Oatmeal joined #git
20:17 dirtyroshi joined #git
20:19 bronson joined #git
20:20 regedit joined #git
20:20 WayToDoor joined #git
20:21 Peuc joined #git
20:22 windsurf_ joined #git
20:23 PHPanos joined #git
20:23 windsurf_ project has two developers, each committed with different indentation (tabs/spaces). We want to fix indentation to tabs. If I reserve a commit for only whitespace fixes (change all indentation to tabs), and commit with ignore-space-change, then from then-on commit without ignore-space-change, will the history keep the tab fixes but not consider the whole file to have changed? History preserved elegantly?
20:24 DrRobotic joined #git
20:25 Silenced joined #git
20:26 dsantiago joined #git
20:35 rwp windsurf_, Without answering your question (because I don't have a good answer) let me suggest trying what you want on a branch. Then poke around on it and see if the result is acceptable. If not then no problem just delete the experimental branch.
20:35 windsurf_ rwp good idea!
20:36 rwp That is one of the features I like about git over many other systems. With git you can try it easily with no long term risk. Then decide later on keeping it or discarding it.
20:37 kulelu88 joined #git
20:37 anonymuse joined #git
20:40 Darcidride joined #git
20:40 ToBeCloud joined #git
20:40 watmm1 joined #git
20:46 rewt commit with ignore-space-change would leave the existing whitespace in the file, resulting in not changing the committed state, and then the first commit without ignore-space-change will commit the whitespace changes
20:47 [cust] joined #git
20:49 Rapture joined #git
20:50 overlord_tm joined #git
20:51 windsurf_ rewt thank you. What's the best way to make my whitespace the same and preserve history elegantly? Is that even possible?
20:52 rewt what do you mean by preserve history elegantly?
20:52 rewt just change the whitespace, commit, and done
20:53 mdw joined #git
20:53 anonymuse joined #git
20:54 BrianJ joined #git
20:55 watabou joined #git
20:55 shgysk8zer0_ joined #git
21:00 jottr joined #git
21:01 Tatou_ joined #git
21:01 Tatou_ Howdy
21:01 Tatou_ I've got something I've commited to github - but there's a mistake on it
21:02 Tatou_ Only the one commit. But what's the way I can silently re-write history here?
21:02 tobiasvl Tatou_: silently, not really. once history is made public, you should consider it public
21:02 tobiasvl Tatou_: is there just a mistake or did you commit !sensitive data?
21:02 gitinfo Tatou_: [!filter_sensitive] You can use filter-branch to remove sensitive data from a repository's history. https://help.github.com/arti​cles/remove-sensitive-data/
21:03 Tatou_ Just a mistake
21:03 tobiasvl well, for one commit filter-branch is a bit overkill anyway
21:03 Tatou_ But I am choosing to learn how to rewrite history here
21:03 tobiasvl Tatou_: !reewrite
21:03 tobiasvl Tatou_: !rewrite
21:03 gitinfo Tatou_: 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
21:03 Tatou_ I'm still learning git, tobiasvl .. So I'm looking at my options
21:04 Tatou_ There's nothing sensitive or anything. I just want to see what git is capable of :)
21:04 Zialus joined #git
21:04 clemf_ joined #git
21:05 tobiasvl Tatou_: well, if it's the latest commit, it's really simple. fix the mistake, add the file, run "git commit --amend", and then "git push --force-with-lease" to force github to update with the rewritten history
21:06 Tatou_ cool, let me try that!
21:06 Tatou_ Damn I'm on the wrong laptop -_-
21:06 tobiasvl wrong laptop? this is git, you can just clone :P
21:07 Tatou_ I know, I know..
21:07 Tatou_ Sigh I'll not be lazy
21:07 Atm0spher1c joined #git
21:08 GavinMagnus left #git
21:10 mp3 joined #git
21:10 howitdo joined #git
21:10 nowhere_man joined #git
21:11 unborn joined #git
21:11 mp3 left #git
21:11 tmsmith joined #git
21:13 BlueProtoman left #git
21:14 fatalhalt joined #git
21:14 ok91 joined #git
21:15 dsdeiz joined #git
21:15 dsdeiz joined #git
21:18 cyphase joined #git
21:20 jknetl joined #git
21:20 bronson joined #git
21:21 askb joined #git
21:21 Oatmeal joined #git
21:23 jccn joined #git
21:25 Rodya_ joined #git
21:25 jottr joined #git
21:28 windsurf_ rewt I mean, if I have commits A, B and C (chronologically), and B only fixes whitespace, then when diffing changes between B and A or C and A, it will look like almost the whole file has changed, so the history before B is impacted pretty significantly from a practical standpoint
21:28 ResidentBiscuit joined #git
21:30 rewt well, if you change anything, including whitespace, it will be a change, no way around that... but you can diff ignoring whitespace and you won't see the whitespace changes
21:30 rewt i think it's -B, but check man git diff
21:30 gitinfo the git-diff manpage is available at http://jk.gs/git-diff.html
21:31 rewt -b
21:33 Vortex35 joined #git
21:33 GavinMagnus joined #git
21:34 Es0teric joined #git
21:38 jagob joined #git
21:41 cyphase joined #git
21:43 mdw joined #git
21:43 endiruna joined #git
21:47 GavinMagnus left #git
21:48 satifant joined #git
21:48 windsurf_ rewt ok that's something
21:48 windsurf_ thx
21:48 windsurf_ left #git
21:50 kulelu88 left #git
21:51 jstein__ joined #git
21:52 jeffreylevesque joined #git
21:52 evil-t0by joined #git
21:55 sLite joined #git
21:57 c^ joined #git
21:57 drgeb joined #git
21:57 drgeb Hi
21:57 gitinfo drgeb: 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.
21:59 drgeb What is best way to deal with one git repo lets say  the project is called DiscoverTheWorld but right now while I am starting I have many little pieces of the big puzzle that I neeed to Discover before I get there. My manager wants just one repo where as I would create as many as needed for each part of the puzzle that we are trying to solve and buildup the whole Architecture as we go. Any suggestions on this ???
22:00 BrianJ joined #git
22:01 WayToDoor joined #git
22:06 ResidentBiscuit joined #git
22:06 YuGiOhJCJ joined #git
22:10 aidalgol joined #git
22:14 Cassiopaya joined #git
22:16 iAmerikan joined #git
22:16 iAmerikan is there any method faster than git pull to check if the remote has changes? fetch / pull --dry-run both appear to be slower
22:17 uva joined #git
22:17 uva How would I codify git diff "--no-prefix" into ~/.gitconfig?
22:17 IlNominePatre joined #git
22:17 kadoban ls-remote, it's not really user-friendly though in the slightest.
22:18 iAmerikan then compare hashes?
22:19 ac3takwas joined #git
22:19 kadoban Yes
22:21 bronson joined #git
22:22 anli joined #git
22:22 iAmerikan sweet, thanks
22:23 anli With rabbitvcs, I miss one thing: "restore after commit", is it a hidden function by any chance?
22:23 anli Meaning that a file does not change, even if you edit the diff before committing it
22:23 jagob joined #git
22:24 kadoban anli: What?
22:25 ResidentBiscuit ^
22:25 kadoban I can't tell if you're using some other tool and looking for some feature from git, or using git and looking for some feature from some other tool, or ... something else, and the details of what you're looking for.
22:26 iAmerikan anyone have any lower level or more advanced git resource go-to's?
22:28 llamapixel packtpub books at the moment iAmerikan are $5
22:29 llamapixel If you want branching management concepts - http://nvie.com/posts/a-succ​essful-git-branching-model/  ||  https://gist.github.com/djs​piewak/9f2f91085607a4859a66
22:30 llamapixel What specifically are you looking for?
22:30 llamapixel https://www.atlassian.com/git​/tutorials/advanced-overview/
22:31 iAmerikan nothing specific. that atlassian tutorial has been my go to, was looking to keep going down the rabbit hole.
22:31 kini joined #git
22:31 helpD joined #git
22:31 overlord_tm joined #git
22:31 rewt anli, this was covered yesterday and the day before;  when you do a partial commit, the file you had before the commit is the file you end up with after, with both the committed and uncommitted changes
22:32 helpD left #git
22:33 llamapixel iAmerikan: and gitfu ? https://gitfu.wordpress.com/
22:33 jimi_ joined #git
22:34 iAmerikan these are all awesome links, thanks!
22:34 rewt anli, git's staging is same as the copy you'd edit with rabbitcvs that you'd commit
22:35 llamapixel https://www.packtpub.com/all?search=git
22:35 llamapixel np iAmerikan
22:35 jottr joined #git
22:35 Gianormagantrous joined #git
22:35 Gianormagantrous joined #git
22:36 rewt anli, this is the way git actually works; it's not a hidden feature at all
22:38 jagob joined #git
22:42 tang^ joined #git
22:43 roygbiv joined #git
22:43 jw4242 joined #git
22:44 Samual joined #git
22:48 WaReZ joined #git
22:50 masuberu joined #git
22:51 gtristan joined #git
22:52 bekks joined #git
22:52 bekks hi all
22:54 pks joined #git
22:55 bekks I've started learning git today, using gitlab and eclipse. Until now, I've used SVN. For a very small project (<50 files, <30 "releases" until now), I'm trying to "rebuild" that entire project and its releases using a git repo and tags (for the release versions). In SVN, I've used "x.y.z" in the file headers and replaced it with the according version string. How would I go about this using git? Basically I want to identify the release by some value in ...
22:55 bekks ... the file header.
22:57 bekks I'd be glad for any hints please :)
23:00 theskillwithin joined #git
23:01 nidr0x joined #git
23:01 fmeerkoetter joined #git
23:03 dsdeiz joined #git
23:04 anjen joined #git
23:05 Electrometro joined #git
23:07 Electrometro joined #git
23:08 theskillwithin joined #git
23:08 llamapixel so you are looking to merge or migrate SVN to git bekks ?   https://www.atlassian.com/git/​tutorials/migrating-overview/   || https://www.atlassian.com/git/t​utorials/migrating-synchronize/
23:08 PHPanos joined #git
23:09 bekks llamapixel: No no no, I'm not looking for a migration tutorial, I will migrate the current SVN upstream to git. I am looking for on how to handle releases after that.
23:10 nicksloan joined #git
23:14 jagob joined #git
23:15 PHPanos joined #git
23:16 jagob joined #git
23:16 sbeller joined #git
23:21 masuberu joined #git
23:21 cdown joined #git
23:21 Samual joined #git
23:22 bronson joined #git
23:27 watabou joined #git
23:29 denimsoft joined #git
23:30 Sasazuka_ joined #git
23:32 pks joined #git
23:35 Sasazuka joined #git
23:36 imachuchu bekks: With git you just tag the commit that the release is based off of with whatever version number you want and then deploy it however you do that. There isn't any standard format for versions, so you can make it whatever you want so like "Release v1.2.3"
23:36 bekks Yeah.
23:36 unborn hello guys, I have really idiotic question.. when I create git and push it to remote, everything works, now when I change anything in my code.. how to push updates to remote? is there any single command for this?
23:37 bekks The only point I'm strugling with, is: having whatever crap in a git repo, can I just tag a few files, and not the whole repo?
23:38 jagob joined #git
23:38 imachuchu unborn: yes/no. First you "add" the changes (to tell git you want it to store them), then "commit" the changes (to make the object it stores those changes in), then "push" to the remote (to give it the commit you made)
23:39 unborn imachuchu: like git add * -m "changes" and then just git push? that is all? (I mean remote is in config already)
23:40 unborn I mean git add * then git commit -m "commit message" and then just git push?
23:40 imachuchu bekks: not really. Add the files you want in a commit and tag it, you can have other stuff uncommited, or on a different branch. Git thinks of a commit as a snapshot of workfolder at some point in time (at least the stuff you told it to snapshot). This means you can't tag just some files, but you can create a commit with them and tag it
23:41 unborn or is it git push origin master? that is where I am confused
23:42 anonymuse joined #git
23:42 imachuchu unborn: well.. they're the same thing. "git push" says "with the branch I've currently checkedout, push it to the remote I've said it's tracking", "git push origin master" says "push the branch master to the remote origin"
23:43 Gianormagantrous joined #git
23:43 Gianormagantrous joined #git
23:43 imachuchu unborn: assuming that you've cloned from origin, then you're currently on the branch master and it's been setup to track origin, so the two commands do the same thing, just one is you specifying instead of going with it's defaults
23:44 bekks imachuchu: So having my workfolder, I could change, commit and tag files A and C - how would I "add" file B to this tag then?
23:44 unborn oh okay... but I do not have to check out every time do I? I mean I had issue to update one of my repos that is why I am asking what is the step by step with command line
23:45 unborn imachuchu: so it follows like this: git add * then git commit "commit msg" and lastly git push.. is that all?
23:46 imachuchu unborn: yes, to push your changes to remote. If there are other people working on the same remote repo then it gets a bit more complex, but that's a different bridge to cross
23:47 FooMunki joined #git
23:47 imachuchu (cause you have to integrate their changes with your code, and they might have pushed before you did, etc. Standard stuff, but a different topic)
23:48 bekks imachuchu: So maybe I'm better off using a branch for every "release" then, and just transfer every file from my working state to that branch, then tag that branch, and release it?
23:48 FooMunki Anyone know any reason why `git submodule [status | init | update]` are executing silently?
23:49 imachuchu bekks: probably not. So why, say, do you want file x in release 1.2.3 but not in 1.2.4?
23:49 unborn imachuchu: nah its just my private repos.. as I am webdev whole development gets really bad mess for me so I started using git couple months ago, however I was using those fancy tools aka smartgit but lately doing 60gb of repos is damn bad with any gui.. I wrote script which does init for me and create all repo however I was not sure how to simply push any changes to my remote
23:50 imachuchu bekks: !xy
23:50 gitinfo bekks: Woah, slow down for a bit. Are you sure that you need to jump through that particular hoop to achieve your goal? We suspect you don't, so why don't you back up a bit and tell us about the overall objective...
23:50 imachuchu sorry, not bekks, that was for unborn
23:50 rwb joined #git
23:51 imachuchu unborn: 60gb? It sounds like you're storing the wrong thing it git. Git's really great for code, really bad for binaries/images/blobs of any kind
23:51 BlessJah unborn: how did you ended with 60GiB of stuff? you might be interested in !annex
23:51 gitinfo unborn: git-annex, git-media, and Git LFS are some solutions to the !binary problem. They work by keeping the blobs outside of the repo, storing a reference to the blob in the repo instead. http://git-annex.branchable.com https://github.com/alebedev/git-media http://git.io/git-lfs
23:52 unborn imachuchu: ah no :) most repos are 15 megs max..
23:52 BlessJah s/ended/end up/
23:52 unborn but mostly very small.. like bash scripts and bunch of js and html and css scripts
23:52 jagob joined #git
23:54 bekks imachuchu: So my overall objective is: I have project consisting of a bunch of files. I am constantly editing these files, and now and then I want to release a subset of these files as a tarball for "a new release". In addition, I want to have the tag name in the headers of these released files.
23:54 BlessJah bekks: why would you release just subset of files?
23:55 unborn I do not use it for binary stuff and if I do, its only some gif or small zipped file occasionally BlessJah years of coding, backups and other projects I contributed to.. together its load
23:55 bekks BlessJah: Because some of the unreleased files are like notes (long text) to a specific implementation, and such, which are irrelevant for the users of the release.
23:56 BlessJah unborn: borg serves me well for backups
23:56 unborn BlessJah: yeah I have normal backups - those are mods for configs for clients and friends or stuff like that..
23:57 BlessJah bekks: will users have access to said git repo? or just tarball releases?
23:57 unborn is it possible to do git commands with && aka git add * && git commit "msg" && git push? possibly another idiotic question from me.. im sorry.
23:57 bekks BlessJah: Users will have access to the tarballs only.
23:58 BlessJah unborn: git is just an another command in your shell
23:58 unborn that is brilliant!
23:59 BlessJah unborn: also, there is git for windows bundled with bash (just in case you're on windoze)

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