Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-04-19

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

All times shown according to UTC.

Time Nick Message
00:03 Some_Person joined #git
00:06 juvenal joined #git
00:06 Some_Person left #git
00:07 drodger joined #git
00:08 LeBlaaanc joined #git
00:10 loskutak joined #git
00:11 cahoots joined #git
00:12 bruce_lee joined #git
00:12 fish3485 joined #git
00:13 BenderRodriguez joined #git
00:14 drodger joined #git
00:15 Anja joined #git
00:16 watabou joined #git
00:20 Cabanossi joined #git
00:20 alexandre9099 joined #git
00:23 alexggordon joined #git
00:24 Noodlewitt joined #git
00:27 TbobbyZ joined #git
00:32 ispn joined #git
00:34 Noodlewitt joined #git
00:36 ToBeCloud joined #git
00:41 TikityTik joined #git
00:42 Noodlewitt joined #git
00:44 DarkPsydeLord joined #git
00:45 Noodlewitt joined #git
00:49 warthog9 joined #git
00:50 Cabanossi joined #git
00:51 peterbec` joined #git
00:51 alexandre9099 joined #git
00:52 watersoul joined #git
00:52 TikityTik left #git
00:53 demize joined #git
00:53 bongo joined #git
00:54 devhost joined #git
00:54 wenzowski joined #git
00:54 thecomedian joined #git
00:54 gajus joined #git
00:54 TikityTik joined #git
00:54 boshhead joined #git
00:56 Esya joined #git
00:56 brewmaster joined #git
00:56 tctara joined #git
00:57 lmat joined #git
00:57 Ragnor joined #git
00:58 fstd_ joined #git
00:58 mar77i joined #git
00:59 bongo joined #git
00:59 __name__ joined #git
00:59 ChrisEadle joined #git
01:00 cahoots_ joined #git
01:04 AstralStorm joined #git
01:07 Murr joined #git
01:17 thiago joined #git
01:19 LeBlaaanc joined #git
01:20 Cabanossi joined #git
01:20 nicolai86 joined #git
01:28 TbobbyZ joined #git
01:32 om_henners joined #git
01:33 cqi joined #git
01:37 durham_ joined #git
01:39 eikhorst joined #git
01:43 Shavik joined #git
01:46 TikityTik joined #git
01:46 ertes joined #git
01:46 nick123 joined #git
01:52 peterbec` joined #git
01:53 jameser joined #git
01:53 jeffreylevesque joined #git
01:53 rchavik joined #git
01:59 watabou joined #git
02:00 sunri5e joined #git
02:02 Godot joined #git
02:02 Godot left #git
02:02 fahadash joined #git
02:05 Cabanossi joined #git
02:07 HoierM joined #git
02:08 jargonjunkie joined #git
02:11 a3Dman joined #git
02:16 hashtagstrashtag joined #git
02:18 plasticle joined #git
02:18 Cabanossi joined #git
02:18 cqi joined #git
02:19 brent__ joined #git
02:20 mentor joined #git
02:20 mentor joined #git
02:22 TbobbyZ joined #git
02:22 Goplat joined #git
02:23 plasticle joined #git
02:25 mentazoom joined #git
02:26 ojdo joined #git
02:26 dsdeiz joined #git
02:26 dsdeiz joined #git
02:27 dreiss joined #git
02:33 dreaddymck joined #git
02:34 inflames joined #git
02:40 jason237 joined #git
02:40 Sasazuka joined #git
02:41 DarkPsydeLord joined #git
02:42 fbwnd joined #git
02:44 samba_ joined #git
02:44 lb1c joined #git
02:45 samba_ hello I've a rebase problem with git
02:45 samba_ when I'm trying to do a rebase -i HEAD~2
02:45 samba_ I can pick the commit I want and drop the unwanted
02:46 samba_ but when I complete the rebase and I back to the master branch I can't push
02:46 samba_ instead I need to git pull and this brings me to the begin, where I need to rebase again
02:47 samba_ what I'm not understanding?
02:47 kadoban Sounds like you're rewriting history that you already pushed, then pulling which apparently brings back in the commit you're dropping or something of that sort.
02:48 kadoban So, you're not understanding pull or why it's giving you that message or why not to blindly follow its advice :) Is anyone else using the repo you're pushing to? Is this a public or shared repo/branch?
02:48 metachr0n joined #git
02:49 Vortex35 joined #git
02:49 samba_ yes, I did a fork from the afew repo https://github.com/afewmail/afew
02:50 samba_ my fork https://github.com/opensamba/afew/
02:50 CEnnis91 joined #git
02:51 samba_ "blindly follow its advice" :)
02:52 kadoban Do you understand the consequences of changing published history? Is this actually what you want to do? Is it likely that anyone else is using your fork?
02:53 peterbec` joined #git
02:53 cahoots joined #git
02:53 porfiriopaiz joined #git
02:54 Cabanossi joined #git
02:54 samba_ I understand in the moment you publish you cannot back, yes. I wanted send a fix and did a push, but then I realized 1day later the fix was not fixing and I did another commit+push. now I'd like to have just one single commit
02:54 samba_ And, no. no-one else is using my fork
02:56 mizu_no_oto joined #git
02:56 kadoban Then you probably just want to do 'push -f' ... but what you're describing you want is precisely what you seem to think you cannot do "I understand in the moment you publish you cannot back" <-- that's what you're trying to do though.
02:57 kadoban If nobody is using your published stuff though, you usually don't care.
02:58 phase2 joined #git
02:59 chachasmooth joined #git
03:00 hexagoxel joined #git
03:01 samba_ yes, you are right. but I'm still trying to understand how to manage multiple commits and when make sense to push
03:02 samba_ anyway the push -f did the trick
03:02 duderonomy joined #git
03:04 aavrug joined #git
03:04 _FifniR_ joined #git
03:08 netj joined #git
03:08 ayogi joined #git
03:13 theiostream joined #git
03:17 bronson joined #git
03:18 durham joined #git
03:19 pifon joined #git
03:20 Cabanossi joined #git
03:20 telephone joined #git
03:20 telephone joined #git
03:23 LeBlaaanc joined #git
03:26 davimore joined #git
03:26 bronson joined #git
03:27 andrew710 joined #git
03:31 TbobbyZ joined #git
03:33 SwiftMatt joined #git
03:34 peterbec` joined #git
03:38 Sasazuka joined #git
03:41 a_thakur joined #git
03:41 hexagoxel joined #git
03:46 xsddz joined #git
03:48 Sasazuka joined #git
03:48 Cabanossi joined #git
03:51 bronson joined #git
03:51 Saint_Isadore joined #git
03:52 jost joined #git
03:57 robotroll joined #git
03:57 drodger joined #git
04:00 choki joined #git
04:01 Kaisyu joined #git
04:04 pietv joined #git
04:05 dsdeiz joined #git
04:05 childnode joined #git
04:08 fakenerd joined #git
04:09 shgysk8zer0 joined #git
04:09 cagedwisdom joined #git
04:09 Motoko joined #git
04:10 ISmithers joined #git
04:12 rscata joined #git
04:17 MineCoins joined #git
04:17 DarkPsydeLord joined #git
04:18 timvisher joined #git
04:18 safe joined #git
04:23 nivag joined #git
04:24 hashtagstrashtag joined #git
04:25 childnode left #git
04:30 childnoDe joined #git
04:34 juvenal joined #git
04:40 gregor2 joined #git
04:41 dsdeiz joined #git
04:42 ayogi joined #git
04:45 Noodlewitt joined #git
04:48 Cabanossi joined #git
04:49 fakenerd joined #git
04:51 freimatz joined #git
04:52 snits joined #git
04:53 bodkan joined #git
04:53 watabou joined #git
05:02 dersand joined #git
05:03 MrC joined #git
05:04 Guest26 joined #git
05:06 diogenese joined #git
05:07 svm_invictvs joined #git
05:07 Mutter joined #git
05:09 mentazoom joined #git
05:11 bket joined #git
05:12 xsddz joined #git
05:12 howdoi joined #git
05:13 Shakeel_ joined #git
05:13 Mutter_ joined #git
05:18 todd_dsm_ joined #git
05:18 xsddz left #git
05:20 xsddz joined #git
05:21 xsddz left #git
05:24 xsddz joined #git
05:24 jozwior joined #git
05:25 Shakeel_ joined #git
05:26 PioneerAxon joined #git
05:31 luke_penn joined #git
05:33 relgar left #git
05:33 juvenal joined #git
05:35 ogny joined #git
05:35 overlord_tm joined #git
05:35 PioneerAxon joined #git
05:36 loskutak joined #git
05:38 hashtagstrashtag joined #git
05:41 fakenerd joined #git
05:44 brent__ joined #git
05:45 sbulage joined #git
05:46 Raging_Hog joined #git
05:46 ThomasLocke joined #git
05:47 TbobbyZ joined #git
05:48 fakenerd joined #git
05:49 Cabanossi joined #git
05:52 GT4066 joined #git
05:52 MineCoins joined #git
05:56 imack joined #git
06:00 dermoth joined #git
06:01 lb1d joined #git
06:01 slidercrank joined #git
06:02 qt-x joined #git
06:02 slidercrank hello. I am trying to write a pre-commit hook in bash. How do I get the name of a file (or files) being committed?
06:04 pietv joined #git
06:04 peterbec` joined #git
06:05 slidercrank I probably should use commit-msg. Because $1 is not empty there
06:07 ChrisEadle joined #git
06:07 Gabemo joined #git
06:08 slidercrank ah, no. I get ¨.git/COMMIT_EDITMSG¨ instead of a filename to be committed
06:08 pks joined #git
06:08 _ikke_ git diff HEAD --name-only should return the names
06:09 slidercrank ah, I see. Thank you
06:09 _ikke_ or better, git diff --cached --name-only
06:12 SeanC joined #git
06:12 _ng joined #git
06:13 inflames joined #git
06:16 JeroenT joined #git
06:17 aruns joined #git
06:17 narcberry joined #git
06:22 fakenerd joined #git
06:22 tarkus joined #git
06:24 absinthe joined #git
06:27 leeN joined #git
06:28 freimatz joined #git
06:28 chele joined #git
06:29 fakenerd_ joined #git
06:30 fakenerd joined #git
06:38 Mindiell left #git
06:38 JeroenT joined #git
06:38 bodkan joined #git
06:39 earnestly joined #git
06:39 bronson joined #git
06:41 Squarism Im forgetting to push changes a bit too often at work. Anyone got some nice trick to avoid it?
06:42 immae Squarism: is your problem the fact that you cannot then work from somewhere else or that you risk loosing your work?
06:42 Xatenev__ joined #git
06:43 Xatenev__ Is it possible to create a git alias with a parameter?
06:43 Xatenev__ (If not - I will try to create a shell script that just calls git)
06:43 Squarism l
06:43 Squarism immae, my problem is that I risk colleques accepting pull request that arent complete
06:45 immae ok so you would like to "push" each time you commit something?
06:45 Squarism immae, possibly yes.
06:45 zeroed joined #git
06:45 zeroed joined #git
06:46 immae how about a less invading one: showing the status after each commit?
06:46 Squarism so this isnt a problem other people battle? Working on 1-5 branches every day for smaller features
06:46 Squarism immae, also good idea
06:46 immae you can add that to a hook then ;)
06:46 immae "git status"
06:48 dreiss joined #git
06:48 kadoban Squarism: Uhm, don't request people pull stuff before you actually check it then? Also shouldn't their reviews catch that even if you do?
06:49 Squarism kadoban, all communications is primarly done using JIRA.
06:49 Squarism +bitbucket
06:50 JeroenT joined #git
06:51 kadoban Squarism: I don't understand how that relates. There is no way to check what you're requesting someone pull before you do it, or?
06:51 chichou joined #git
06:52 Squarism kadoban, sure. Their review should catch it. But every so often you find some aspect of project A something. Make a commit. Person comes in, context switch. Wednesday you work on project B. All details about project A gets forgotten.
06:52 jagob joined #git
06:53 immae Squarism: do you know how to add hooks?
06:53 Squarism immae, i havent done it but i get the idea.
06:53 immae ok
06:53 immae good then
06:53 waveclaw joined #git
06:54 Squarism immae, what type of hook did you have in mind?
06:54 immae .git/hooks/post-commit
06:54 immae that script is ran after each commit
06:55 immae here you can either "git push origin <branch>" or "git status"
06:55 _ikke_ Xatenev__: Yes, you can
06:56 immae http://stackoverflow.com/a/28042939
06:56 immae a nice way to auto-push
06:56 immae (I prefer the status one less invasive, but it's up to you)
06:58 Squarism immae, maybe auto push would be a bit drastic for me - not to seldom i undo commits. But hooks could definently help. I use in an IDE - and do git work there too. I prolly write a unity-applet that shows warnings on non pushed branches
06:59 Xatenev__ _ikke_, Got it already :)
07:01 immae ok, you have the idea now, I'm not used to IDE's, so it's your choices now ;)
07:01 zefferno joined #git
07:02 tarkus How to tweak a remote to make it "fetch" only?
07:03 aavrug joined #git
07:03 Squarism immae, thanks alot. Colleque interrupted, so sorry for delayed thanks
07:03 mkalfon_ joined #git
07:04 zeroed joined #git
07:04 zeroed joined #git
07:05 Nilesh_ joined #git
07:05 serdar joined #git
07:05 Tobbi joined #git
07:12 watabou joined #git
07:17 alexandre9099 joined #git
07:19 Cabanossi joined #git
07:20 CheckDavid joined #git
07:27 imack joined #git
07:28 mattcen joined #git
07:29 SilverBe joined #git
07:31 chardan joined #git
07:31 ZyX-I joined #git
07:33 a_thakur joined #git
07:34 brk_ joined #git
07:38 Chuguniy joined #git
07:39 Jugurtha joined #git
07:39 waveclaw joined #git
07:40 bilal joined #git
07:40 Jugurtha Hi.. I was working on something and wanted to go back to a previous state without committing, so I did a "git stash". When I wanted to go back where I was, I did a "git stash apply" and the code was full with ">>>>>>> Stashed changes". How do I look at this?
07:43 cbreak Jugurtha: those are just conflicts
07:43 cbreak the stashed changes conflict with some local changes
07:43 cbreak you have to resolve them like any other conflicts
07:44 mikecmpbll joined #git
07:45 brent__ joined #git
07:45 schleppel joined #git
07:47 Jugurtha cbreak: Thank you.
07:48 Jugurtha cbreak: I found this https://help.github.com/articles/resolving-a-merge-conflict-using-the-command-line/
07:50 Jugurtha cbreak: So I have to go wherever I find these things and decide what to keep (i.e: manually modify).
07:50 fakenerd_ joined #git
07:51 vamiry_ joined #git
07:52 j`ey left #git
07:52 dvaske joined #git
07:53 acetakwas joined #git
07:54 cbreak Jugurtha: yes
07:54 cbreak Jugurtha: typically done with git diff and a text editor, or a merge tool
07:55 inflames joined #git
07:55 jimmyrcom joined #git
07:55 JeroenT joined #git
07:56 mikecmpbll joined #git
07:57 Jugurtha cbreak: What if I do a git checkout HEAD -- file.txt and want to recover.
07:58 babilen joined #git
07:59 kurkale6ka joined #git
07:59 ambro718 joined #git
08:00 Jugurtha cbreak: Did I lose what I stashed?
08:00 cbreak no
08:00 cbreak you lost what you had in the working dir
08:00 cbreak stashes are not affected by checkout
08:01 cbreak git stash list should show them still, and git apply should apply them again
08:01 cbreak git stash pop or drop remove stashes, but only on success
08:01 aavrug joined #git
08:02 Tee_Pee joined #git
08:02 pbandark joined #git
08:02 kristofferR joined #git
08:03 wxg joined #git
08:04 Cabanossi joined #git
08:05 Jugurtha Suppose that I did a git stash, then for one file I did a "git checkout HEAD -- file", how do I cancel this last operation?
08:05 waveclaw joined #git
08:05 mindfart joined #git
08:06 peterbec` joined #git
08:06 manuelschneid3r joined #git
08:06 grawity with no edits in between, it did nothing
08:07 Jugurtha grawity: It did. I'm seeing the conflicts disappear and the work I did before stashing disappear to the previous version (the one that's already been committed before improvement).
08:08 King_Hual joined #git
08:09 theoceaniscool joined #git
08:09 Jugurtha grawity, cbreak: Okay, I just reopened the file in Sublime and Ctrl-Z, now I have it the way it was.
08:10 Jugurtha cbreak, grawity: Man, powerful tools in the hands of a primate like myself sure are something.
08:10 marcogmonteiro joined #git
08:11 acetakwas joined #git
08:11 skunkz joined #git
08:14 portal_ joined #git
08:19 theoceaniscool joined #git
08:20 Tuor joined #git
08:23 govg joined #git
08:24 zulutango joined #git
08:25 armyriad joined #git
08:27 cbreak git stash apply would apply the stash again
08:28 Xatenev__ left #git
08:30 rasto joined #git
08:36 rominronin joined #git
08:38 gareppa joined #git
08:40 dsdeiz joined #git
08:40 clmsy joined #git
08:41 notebox joined #git
08:42 Snugglebash joined #git
08:42 ispn joined #git
08:44 djb-irc joined #git
08:44 kurkale6ka joined #git
08:46 Calinou joined #git
08:47 sami joined #git
08:48 bilal joined #git
08:48 choki joined #git
08:48 kent\n joined #git
08:49 sami left #git
08:49 andrew710 joined #git
08:49 marcogmonteiro joined #git
08:51 maroloccio joined #git
08:52 jeffreylevesque joined #git
08:54 fakenerd joined #git
08:54 Introoter joined #git
08:56 sshine joined #git
08:57 TbobbyZ joined #git
08:58 sshine hi. trying to add a remote ssh repo for easy deployment: [remote "web"] url = user@foo.bar:dir/subdir   fetch = +refs/heads/*:refs/remotes/web/*, but doing 'git push web master' gives "bash: git-receive-pack: command not found. fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists."
08:59 sshine I suspect that I'm specifying the remote directory wrong.
08:59 Raging_Hog joined #git
08:59 Seveas no, you need to install git on the remote box. Also !deploy
08:59 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
09:00 JeroenT_ joined #git
09:01 mentazoom joined #git
09:01 sshine the thing is, I'm doing freelance work for a company with no devs, so a simple deployment process while I'm here is better than setting up an actual deployment tool for now.
09:01 sshine also, git isn't installed on the remote box. that was silly of me. thanks a lot! :)
09:02 pbandark1 joined #git
09:03 fakenerd joined #git
09:03 serdar joined #git
09:03 Cabanossi joined #git
09:04 gunnaro joined #git
09:05 bruce_lee joined #git
09:05 fide joined #git
09:06 Raging_Hog joined #git
09:06 Murii joined #git
09:07 peterbec` joined #git
09:11 fakenerd joined #git
09:14 PaulCape_ joined #git
09:16 a_thakur joined #git
09:18 theoceaniscool joined #git
09:18 MACscr joined #git
09:20 luke_penn joined #git
09:21 portal_ joined #git
09:22 cyberz joined #git
09:24 neunon joined #git
09:26 lb1c joined #git
09:27 aidalgol joined #git
09:28 a_thakur joined #git
09:28 peacememories joined #git
09:29 kurkale6ka joined #git
09:30 mentazoom joined #git
09:32 oskarkv joined #git
09:32 waveclaw joined #git
09:34 mschorm joined #git
09:34 aidalgol joined #git
09:38 Tobbi joined #git
09:39 troyt joined #git
09:39 monq joined #git
09:42 chll_ joined #git
09:45 theoceaniscool joined #git
09:45 byte512 joined #git
09:45 fahadash joined #git
09:47 sbulage joined #git
09:48 snits joined #git
09:49 lordjancso joined #git
09:49 ferr joined #git
09:50 xsddz left #git
09:50 peacememories joined #git
09:51 zefferno joined #git
09:51 a_thakur joined #git
09:52 rorro joined #git
09:53 Rusky_ joined #git
09:55 qis joined #git
09:55 qis How can I enable git-lfs support on FreeBSD?
09:55 qis I installed the git-lfs package with pkg(1) and ran `git lfs install` as both, root and user.
09:55 qis But there is no `git-lfs-authenticate` script and git push fails with:
09:55 qis Remote "origin" does not support the LFS locking API.
09:55 qis batch request: git-lfs-authenticate: Command not found.: exit status 1
09:57 a_thakur joined #git
09:58 CheckDavid joined #git
09:59 qis There is no git-lfs-authenticate command on linux either
09:59 fragMental joined #git
10:05 fakenerd joined #git
10:06 indistylo joined #git
10:06 qis Nevermind. I understood how it's working.
10:08 Gloomy joined #git
10:09 h12o joined #git
10:11 a_thakur joined #git
10:15 kurkale6ka joined #git
10:15 fakenerd joined #git
10:16 kurkale6ka joined #git
10:18 timewalker joined #git
10:18 Cabanossi joined #git
10:21 navidr joined #git
10:22 tvw joined #git
10:23 PioneerAxon joined #git
10:26 fakenerd joined #git
10:27 alexandre9099 joined #git
10:31 peacememories joined #git
10:33 h12o joined #git
10:35 peacememories joined #git
10:35 sshine left #git
10:37 theoceaniscool joined #git
10:39 Mutter joined #git
10:39 Tobbi joined #git
10:42 pbrewczynski__ joined #git
10:43 nowhere_man joined #git
10:49 earnestly joined #git
10:50 peacememories joined #git
10:50 Tuor joined #git
10:50 Tuor joined #git
10:53 sarri joined #git
10:53 sarri joined #git
10:53 serses joined #git
10:54 Puffball joined #git
10:54 serses Hi, I've tried to push a modified file but I get "error src refspec name_of_my_branch does not match any"
10:55 serses but now if I do get status the change isn't there anymore.
10:56 osse serses: you do not push files, and you definitely cannot push modified content (if by modified you mean what git status means=
10:56 jast that just means you have committed it, git status shows only uncommitted changes
10:56 Jugurtha joined #git
10:58 serses So the file is commited ? I tried with "git log" and I can see commit Date and Title
10:58 Taikibansei joined #git
10:58 osse serses: did you commit it?
10:59 serses yes
10:59 osse what does git status say ?
10:59 serses it's ok the changes aren't there anymore
11:00 serses but my question is why the error "error src refspec name_of_my_branch does not match any" right after i've done the push
11:00 kurkale6ka joined #git
11:01 osse !repro
11:01 gitinfo [!transcript] Please paste (using https://gist.github.com/ or similar) a transcript ( https://git.io/viMGr ) of your terminal session so we can see exactly what you see
11:03 serses https://gist.github.com/anonymous/22cbeabd734b6aeba2b2aa9a2c680df3
11:04 osse serses: "resopnsive"
11:04 serses yeah I know :) but it's the right branch
11:05 osse but your prompt says "responsive"
11:06 serses oh crap they are going into "responsive" . How I can change them to go into resopnsive?
11:06 osse commit it in the right branch
11:07 serses ok thanks :)
11:07 osse you might find cherry-pick a suitable tool for "copying" a commit from one place to another
11:07 osse oh
11:08 TbobbyZ joined #git
11:08 osse serses: your local branch has the riht name and the remote branch does not ?
11:08 osse *right
11:08 osse serses: git push origin responsive-newsletters:resopnsive-newsletters
11:08 osse serses: add -u to be able to just type 'git push' next time
11:09 qis left #git
11:09 peterbec` joined #git
11:11 serses yeah the local has the correct name but the remote didn't, that was the problem thanks again
11:12 Raging_Hog joined #git
11:12 MarioBranco joined #git
11:13 skunkz joined #git
11:16 sunri5e joined #git
11:16 jokester joined #git
11:17 jokester joined #git
11:18 dsdeiz joined #git
11:18 dsdeiz joined #git
11:18 Tobbi joined #git
11:18 Cabanossi joined #git
11:19 fakenerd joined #git
11:20 maroloccio joined #git
11:20 borkr joined #git
11:23 chipotle joined #git
11:24 serses left #git
11:24 Takumo joined #git
11:24 Takumo joined #git
11:24 todd_dsm joined #git
11:26 Chuguniy joined #git
11:26 Hanonim joined #git
11:28 kurkale6ka joined #git
11:28 MarioBranco joined #git
11:29 kurkale6ka joined #git
11:29 gregor2 joined #git
11:30 Hanonim left #git
11:31 Tobbi joined #git
11:31 theoceaniscool joined #git
11:34 dsdeiz joined #git
11:34 dsdeiz joined #git
11:34 YuGiOhJCJ joined #git
11:35 re1 joined #git
11:39 dsdeiz joined #git
11:39 dsdeiz joined #git
11:42 jmin joined #git
11:43 netj joined #git
11:45 dglambert joined #git
11:45 dglambert yo
11:46 HoierM joined #git
11:47 masuberu joined #git
11:48 PioneerAxon_ joined #git
11:48 Cabanossi joined #git
11:50 Raging_Hog joined #git
11:54 visual heyyy
11:54 Vampire0 houuu
11:54 visual HELLO FROM MUMBAI
11:55 visual no im kidding
11:55 visual trying to formulate the question
11:55 Vampire0 !justask
11:55 gitinfo [!just_ask] You can just ask your question. If anybody knows the answer, they will answer soon (most of the time)
11:55 visual right, so when checking out another branch, you dont lose the changes, right?
11:56 visual do they merge?
11:56 visual or how can you handle the merge?
11:56 moritz !float
11:56 gitinfo 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.
11:56 Gloomy joined #git
11:56 visual awesome
11:56 visual thx
11:58 visual alright, did git stash
11:58 gitinfo visual: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx", "ty" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
11:58 visual then git checkout, but it aborts with "error: The following untracked working tree files would be overwritten by checkout:"
11:58 kegster joined #git
11:58 visual moritz: thx
11:58 Vampire0 visual, well, as the message says
11:59 visual can you ++ your own carma? :^)
11:59 Vampire0 you have untracked file in your workspace that exist in the branch you want to switch to
11:59 visual Vampire0: yeah it says git stash, git checkout, git stash apply
11:59 Vampire0 vishal, I don't think you can
11:59 Vampire0 visual, yeah, but stash only saves changes tracked files by default
11:59 visual :O!
12:00 Vampire0 visual, you can tell it to also save untracked files though
12:00 visual so git add . ?
12:00 visual to track them, then stash?
12:00 Vampire0 visual, that would probably also be an option
12:00 Vampire0 visual, then you can steer which of the files get stashed opposed to stashing all untracked files
12:01 IntruderSRB joined #git
12:01 synthroid joined #git
12:02 boombatower joined #git
12:02 visual wew
12:02 visual it did something
12:03 apeiros joined #git
12:03 visual "CONFLICT (rename/add): Rename {file} in Updated upstream. {file} added in Stashed changes. Adding as {file~Stashed changes} instead
12:03 visual what did it mean by this?
12:03 kpease joined #git
12:04 visual i see it created a new file, but why didnt it run the editor to invoke a merge?
12:04 apeiros hi all. I try to disable the message one gets when doing `git checkout somebranch` ("Switched to branch 'somebranch'\nYour branch is up to date …yadda…"). I tried with `git config --global advice.statusHints false`, but that didn't suppress those messages
12:04 apeiros any idea on how to disable/suppress them?
12:04 moritz visual: the conflict was that one of the changes was a rename, the other an addition of the renamed file
12:05 visual what?
12:05 visual one of the changes was a rename - i got that part
12:05 moritz so, it renamed a to b
12:05 visual the other change was an addition of the renamed file? addition .. of code?
12:06 moritz and then in the other, a file a was added or changed
12:07 visual what?
12:07 pmtamal joined #git
12:07 visual it renamed the file, yeah
12:07 jast apeiros: I'm not sure it can be turned off individually but you can set status.short to true
12:07 visual but why did it create a new file?
12:07 Vampire0 visual, what do you not understand? In your other branch file b was moved to file a, in your stashed changes file a was newly added. There is no chance to merge those as far as Git is concerned
12:08 visual shit this is confusing
12:08 visual here's the git error:
12:08 visual CONFLICT (rename/add): Rename resources/views/employee/calendar/newEvent.blade.php->resources/views/employee/calendar/newAppointment.blade.php in Updated upstream. resources/views/employee/calendar/newAppointment.blade.php added in Stashed changes
12:08 visual Adding as resources/views/employee/calendar/newAppointment.blade.php~Stashed changes instead
12:09 visual oops, thought i had the link in the clipboard - https://kopy.io/8CgRh
12:09 apeiros jast: re "individually" - if I'd turn them off - how would I do that and what do you think would be also affected?
12:09 visual so it renamde newEvent -> newAppointment
12:09 visual AND created a new newAppointment~Stashed changes file
12:09 visual why?
12:09 visual what's up with the new file? :I
12:09 Puffball joined #git
12:09 quizzi joined #git
12:09 jast apeiros: short mode (enabled with that option) condenses status output way down, showing only a few flags for each file. you can look at it without setting the option: git status --short
12:10 TbobbyZ joined #git
12:10 moritz visual: maybe your stash also touched a newAppointment file?
12:10 peterbec` joined #git
12:11 choki joined #git
12:11 visual my stash did indeed have a newAppointment file
12:12 visual anyway i dont want to waste your time, im only 1/3 of the way through the git book, so maybe it'll become obvious along the way
12:12 visual Vampire0: thx
12:12 visual moritz: thx
12:12 visual )
12:12 apeiros jast: setting that option doesn't reduce that output either :-/
12:12 apeiros jast: thanks for the hint, though :)
12:12 visual oh right, so what if i checkout back to the prev. branch that i stashed the changes from
12:12 visual does that branch still have the changes?
12:12 Vampire0 apeiros, according to the source - if I didn't misinterpret from a quick skim - you can only suppress by using -q or --quiet
12:13 apeiros Vampire0: ok, that works
12:13 apeiros so I guess I'll have an alias for checkout
12:13 Vampire0 visual, the new file is the one that you stashed, just compare and merge it manually with what you have checked out from the branch you switched to
12:14 apeiros probably means I'll also have to figure how to get autocompletion to work with the alias :-(
12:14 visual yeah did that, the new file had all the needed changes, so just replaced them
12:14 apeiros thanks Vampire0
12:14 Vampire0 visual, if you switch back to the old branch, the change is of course not there, as the branch does not have the changes, but if you apply the stashed changes there, they will of course be back
12:14 visual so the stash is still in the "clipboard"? how can i check it?
12:14 visual even after apply
12:14 Vampire0 apeiros, with autocompletion you mean bash completion? That works fine with aliases
12:15 Vampire0 visual, git stash list
12:15 apeiros Vampire0: great, less work :)
12:15 apeiros (and yes, that's what I meant)
12:15 visual wew, i have something stashed from way back when, how can i remove them? )
12:15 visual git stash remove 1 ?
12:15 osse drop
12:15 Vampire0 visual, and yes, if you only do git stash apply, the changes stay in the stash. If you do git stash pop and they apply without conflict they are removed from the stash. If you do git stash drop they are removed from the stash without being applied. See man git stash
12:15 gitinfo visual: the git-stash manpage is available at https://gitirc.eu/git-stash.html
12:16 visual cool beans
12:16 visual thanks Vampire0
12:16 Vampire0 visual, yw. but not `1` but `stash@{5}`
12:16 Vampire0 visual, like shown in `git stash list`
12:16 visual got it, thx
12:17 osse i think a patch was made to be able to write simply 5
12:17 osse or maybe it was dropped
12:17 Vampire0 Ah, nice, didn't know of that
12:17 visual apparently not
12:17 visual bash-4.3$ git stash drop 0
12:17 visual fatal: ambiguous argument '0': unknown revision or path not in the working tree.
12:17 visual git version 2.6.3.windows.1
12:18 Vampire0 Also in 2.9.3.windows.1 this does not work
12:19 Cabanossi joined #git
12:19 osse since 2.11 only
12:19 Vampire0 ah, ok, should update then :-)
12:19 Vampire0 thx osse
12:20 osse the commit message says "This patch includes the superior implementation provided by {my name}"
12:20 osse except due to some misunderstanding it's not actually my name
12:20 osse ::D
12:20 Vampire0 visual, btw. be aware that the indexes change as soon as you drop or pop a stashed change if it is not the oldest one. Important in case you want to drop multiple ones consecutively
12:21 dsdeiz joined #git
12:22 jast apeiros: try --no-branch (there's an option for that, too), this is the default though (and I didn't mention it earlier because it affect only --short mode)
12:22 jameser joined #git
12:22 visual yeah noticed )
12:22 jast apeiros: hey, are you trying to script this stuff? you almost certainly want to use 'git ls-files', then
12:22 visual btw
12:22 visual http://prntscr.com/ey98ky
12:23 visual ^ what did it mean by 0 additions 0 deletions? there's a lot of additions
12:23 nickel_dime_look joined #git
12:24 ankit01ojha joined #git
12:25 jast that's one for the github blooper reel
12:25 visual *gitlab
12:25 jast or that
12:25 fl0w joined #git
12:26 gugah joined #git
12:26 theiostream joined #git
12:26 gugah joined #git
12:29 King_Hual joined #git
12:31 dsdeiz joined #git
12:31 dsdeiz joined #git
12:32 King_Hual` joined #git
12:33 theiostream joined #git
12:34 Chuguniy joined #git
12:36 apeiros jast: no, I have a bash prompt which already shows me all that information, so I want to suppress redundant info
12:37 Clickys joined #git
12:37 Clickys Hello iam trying to push my local project to my github account but it wont let me, i got this error
12:37 Clickys ! [rejected]        master -> master (non-fast-forward)
12:37 Clickys error: failed to push some refs to 'https://github.com/Clickys/to-do-app.git'
12:37 Clickys hint: Updates were rejected because the tip of your current branch is behind
12:37 Clickys hint: its remote counterpart. Integrate the remote changes (e.g.
12:37 Clickys hint: 'git pull ...') before pushing again.
12:37 Clickys hint: See the 'Note about fast-forwards' in 'git push --help' for details.
12:38 Clickys should i use git push -f to force it?
12:40 osse Clickys: is the repo on github newly created?
12:40 Clickys yes
12:40 osse dod you accept to let github generate a default readme and/or .gitignore?
12:40 osse *did
12:41 Clickys yes , i use -f and now its ok
12:41 Clickys it push the project
12:41 osse yes, then -f is the correct medicine
12:41 Clickys ty man
12:42 theoceaniscool joined #git
12:42 Tobbi joined #git
12:42 MineCoins joined #git
12:45 ShapeShifter499 joined #git
12:45 a3Dman joined #git
12:45 aruns__ joined #git
12:46 cdown joined #git
12:47 loskutak joined #git
12:48 acetakwas joined #git
12:49 Cabanossi joined #git
12:52 rorro joined #git
12:55 Tobbi joined #git
12:56 cdown_ joined #git
13:01 ExoUNX joined #git
13:06 marianina8 joined #git
13:08 jameser joined #git
13:08 TheSimonator joined #git
13:09 peacememories joined #git
13:13 jeffreylevesque joined #git
13:14 sanguisdex joined #git
13:14 zeroed joined #git
13:14 zeroed joined #git
13:15 DarkPsydeLord joined #git
13:16 githubmgw joined #git
13:16 githubmgw left #git
13:18 Tuor joined #git
13:18 Tuor joined #git
13:21 sanguisdex I have a merge comming up that I know is going to be filled with conflicts. is there a merge strategy that will tell (OURS|THEIRS)  to just merge in above or below  the other changes?
13:21 peacememories joined #git
13:23 Vampire0 sanguisdex, I don't think so. Does not sound too useful. But actually you will have exactly this in the conflict state, so just remove the conflict markers and you should have the result you desire
13:23 robattila256 joined #git
13:23 Tobbi joined #git
13:24 sanguisdex Vampire0:  that was the plan if there was no out of the box stratagy
13:24 sanguisdex thanks
13:24 relipse joined #git
13:25 gitinfo sanguisdex: This channel tracks karma based on who has gotten lots of thanks for being helpful. If you want to help someone reach karmic nirvana, please mention their name when thanking them with "thank you", "thankyou", "thanks", "thx", "ty" or "cheers". Try ".karma <nick>" or ".topkarma" to show karma status of a person.
13:26 sanguisdex Vampire0 thanks*
13:26 cdown joined #git
13:26 sanguisdex its to early in the day to get schooled by a bot
13:27 a_thakur joined #git
13:27 mentazoom joined #git
13:31 fl0w joined #git
13:36 durham joined #git
13:37 Acerific joined #git
13:37 Darcidride joined #git
13:39 ayogi joined #git
13:39 s17 .karma s17
13:39 gitinfo the karma of the given users is shrouded in the mists of uncertainty.
13:39 s17 thanks s17
13:40 s17 .karma gitinfo
13:40 gitinfo the Genuine Real Life Karma™ REST API results are back! g‍itinfo: 31 (0 in past 30 days)
13:41 cjohnson .karma cjohnson
13:41 gitinfo the Genuine Real Life Karma™ REST API results are back! c‍johnson: 2 (0 in past 30 days) (1 given out)
13:42 cjohnson thanks for nothing gitinfo
13:42 gitinfo cjohnson: you're welcome, but please note that I'm a bot. I'm not programmed to care.
13:42 cjohnson Me either
13:42 cjohnson :(
13:45 ephimetheus joined #git
13:46 notebox joined #git
13:46 lknd joined #git
13:47 lknd hey guys, sort of a really stupid question
13:47 lknd if I am working with various branches, but all pull requests must go to the dev branch before being merged into master
13:47 lknd do I need to create a branch based off of dev?
13:47 lknd and work on top of that?
13:47 s17 checkout the dev branch
13:48 lknd well the thing is I need to create each new development (including bug fixes)
13:48 lknd into a new branch
13:48 lknd that's why I'm a little bit confused too
13:48 lknd so for example, git checkout -b bugfix-30
13:48 cjohnson lknd: depends on your setup. In my case, I make the branch from master, then merge into dev for testing, and once it's approved, I merged the branch back into master
13:48 scarabx joined #git
13:48 cjohnson this way my branch doesn't accidentally include changes that aren't ready to go out
13:49 lknd that's basically our case!
13:49 AaronMT joined #git
13:49 Endarked joined #git
13:49 lknd I just don't understand how git merge works entirely
13:49 a_thakur joined #git
13:49 lknd I'm a bit afraid that if I merge my bug-fix branch into dev, but my branch is based off of master that something will go south real quick
13:49 cjohnson nope, should be fine
13:50 cjohnson in effect, a merge just finds commits that aren't yet in the target branch that are in the source branch and copies them over
13:50 osse lknd: you will merge in your stuff + whatever is in master that isn't in dev
13:50 cjohnson it's a little more nuanced than that
13:50 lknd osse: when you mean "whatever is in master that isn't in dev" you mean commit-wise right?
13:50 lknd thanks cjohnson
13:50 cjohnson yes, it's all commits
13:50 lknd in my head I thought it was just some kind of copy pasterino
13:51 cjohnson it never just copies arbitrary working states from one place to the other
13:51 acetakwas joined #git
13:51 lknd got it
13:51 lknd yeah we pass everything through dev and it will be reviewed before passing it to the master branch
13:51 lknd so everything in master should be found in dev
13:52 gregor2 joined #git
13:52 TomyLobo3 joined #git
13:52 cjohnson Yeah in that case you probably want to branch from master, if you branch from dev then your branch will include everything that was on dev at the time
13:52 cjohnson whether it's approved or not
13:53 lknd makes sense, thanks a lot
13:53 lknd that cleared up a whole lot
13:53 lknd I thought it would overwrite whatever was already changed in dev
13:53 lknd but if it is keeping track through the commits then should be good
13:55 Tobbi joined #git
13:56 bannakaffalatta joined #git
13:57 theoceaniscool joined #git
13:58 oleo joined #git
13:59 ping0ra joined #git
14:00 Es0teric joined #git
14:00 bannakaf_ joined #git
14:02 peacememories joined #git
14:04 Cabanossi joined #git
14:08 bannakaffalatta joined #git
14:09 TikityTik joined #git
14:11 hugdru joined #git
14:11 Tobbi joined #git
14:12 overlord_tm joined #git
14:12 LeBlaaanc joined #git
14:13 cdg joined #git
14:17 sunri5e joined #git
14:18 MineCoins joined #git
14:20 Paraxial joined #git
14:20 dunpeal joined #git
14:20 dunpeal Hi. How do I view and change the default upstream for some branches?
14:23 publio joined #git
14:27 gregor2 joined #git
14:28 crose joined #git
14:30 Aeterne joined #git
14:30 Kaisyu joined #git
14:32 a_thakur joined #git
14:32 IRQ` joined #git
14:34 tomatopeel left #git
14:36 a_thakur joined #git
14:37 thiago joined #git
14:39 storrgie joined #git
14:40 peterbec` joined #git
14:40 fl0w joined #git
14:40 enckse joined #git
14:43 cdown_ joined #git
14:44 lewix joined #git
14:44 Tobbi joined #git
14:44 mekhami joined #git
14:45 mekhami why doesn't git reset --hard fail/error if you have no commits? i just lost an admittedly trivial project cause i was stupid but...
14:48 Es0teric joined #git
14:48 ssarah joined #git
14:48 awc737 joined #git
14:49 juvenal joined #git
14:49 h12o joined #git
14:51 jakesyl_mobile joined #git
14:53 TbobbyZ joined #git
14:53 Dougie187 joined #git
14:55 ping0ra joined #git
14:57 theoceaniscool joined #git
15:00 borkr joined #git
15:01 a_thakur joined #git
15:02 gregor2 joined #git
15:04 Cabanossi joined #git
15:04 mikecmpbll joined #git
15:04 fl0w mekhami: Not sure what you did there, but for me git reset --hard does nothing without commits.
15:04 fl0w git 2.10.1
15:04 Dougie187 `git reset --hard` should be equivalent to `git reset --hard HEAD`
15:05 loskutak joined #git
15:06 fl0w I understand it as he had no commits - and that should "do nothing" AFAIK?
15:06 IntruderSRB joined #git
15:06 sborza joined #git
15:07 fl0w I'm guessing there was an "init commit" and the ran reset --hard, which understandibly works as expected
15:07 fl0w or rather, an empty init commit.
15:07 mekhami that definitely didn't happen
15:07 mekhami i had a repository with no commits
15:07 mekhami i had just init'd it
15:07 Gloomy joined #git
15:07 mekhami i had staged everything though
15:07 mekhami and when i did a reset hard
15:08 mekhami everything that was tracked by git was gone
15:08 ZyX-I joined #git
15:09 durham joined #git
15:11 pbandark i created orphan branch and kept only required files. but now I am unable to change the branch
15:11 pbandark https://paste.fedoraproject.org/paste/QCA-nDLyXKOZfOung2xpVl5M1UNdIGYhyRLivL9gydE=
15:12 ping0ra joined #git
15:12 pbandark is that expected ?
15:13 theoceaniscool joined #git
15:13 pbandark how i can fix the issue ?
15:13 grawity that's normal
15:13 grawity those files you kept aren't on any branch yet – you forgot to git add/commit
15:13 grawity or rather
15:14 grawity if it complains about files you *didn't* want to keep, then you forgot to `rm` them
15:16 masuberu joined #git
15:17 TbobbyZ joined #git
15:17 masuberu joined #git
15:17 synthroid joined #git
15:18 synthroi_ joined #git
15:19 masuberu joined #git
15:19 pbandark grawity: i want to create new branch for specific file. the new branch should only contain files related to current work. for ex. i have created new branch for elastic-search. so i want to only store files which are related to elastic-search. the branch should not contain any other files
15:20 doener joined #git
15:20 masuberu joined #git
15:20 pbandark hence, I created orphan branch, remove other files using `git rm --cached` and commited
15:21 pbandark is it correct aproch ?
15:23 qqx pbandark: I'd expect use of `git rm` without --cached for that. Although a separate repo may be even better.
15:24 pbandark qqx: so `git rm` without --cached will remove file from the file system right ? will I be able to get it once I switch to master/other branch
15:24 al-damiri joined #git
15:24 qqx pbandark: Yes by default `git rm` will remove files from both the file system and from future commits.
15:25 tlaxkit joined #git
15:25 qqx When you switch to a commit that contained those files they would be restored to the committed version.
15:25 pbandark ok
15:26 qqx And having the file in the file system, but not tracked (due to using rm --cached) would prevent switching to a commit which had that file as tracked.
15:26 pbandark i am getting "fatal: pathspec 'pillar/mysql.sls' did not match any files" when i do `git rm` on file
15:26 pbandark but the file is present
15:27 qqx pbandark: Yeah, since you'd already removed it from tracking `git rm` won't find it. You can use just `rm` (without `git`) to remove it from the filesystem.
15:27 pbandark ok
15:28 pbandark thanks qqx grawity
15:32 TheSimonator joined #git
15:33 Karthick joined #git
15:34 cdown joined #git
15:35 jagob joined #git
15:38 juvenal joined #git
15:40 jstimm joined #git
15:40 Mutter joined #git
15:41 mikecmpbll joined #git
15:41 peterbec` joined #git
15:42 theoceaniscool joined #git
15:43 PettanShoutaKun joined #git
15:44 PettanShoutaKun how do tags work? .... I'm just finding out now that forever ago someone tagged our repo.... and it says 1.16.7-489-g61dc96e
15:44 cdg joined #git
15:44 PettanShoutaKun but what does that number actually mean?
15:44 whitby Tag names are user defined
15:44 PettanShoutaKun did someone tag that as 1.16.7 it's 489 commits ahead of g61dc96e?
15:45 whitby Someone tagged it '1.16.7-489-g61dc96e' if that's what Git shows as the name name
15:45 whitby It could've been autogenerated, but afaik it wouldn
15:45 whitby 't have been Git that generated it
15:47 shinnya joined #git
15:48 qqx PettanShoutaKun: That output from `git describe` says that you're at commit 61dc96e which is 489 commits ahead of tag 1.16.7.
15:50 dreiss joined #git
15:52 PettanShoutaKun wow!
15:52 PettanShoutaKun that's awesome
15:53 tarkus joined #git
15:53 drodger joined #git
15:54 theoceaniscool joined #git
15:54 PettanShoutaKun is there a way to get just the tag?
15:54 PettanShoutaKun like can I auto increment the patch numbeg of each commit tag
15:55 PettanShoutaKun because that would be sweet!
15:55 peterbecich joined #git
15:55 _ikke_ You can complete define how your tag is called
15:56 _ikke_ git tag foo creates a (lightweight) tag called foo on the current commit
15:59 bannakaffalatta joined #git
15:59 imack joined #git
16:00 TomyLobo3 joined #git
16:01 orbyt_ joined #git
16:02 dsheets joined #git
16:03 TomyLobo3 joined #git
16:04 PettanShoutaKun joined #git
16:06 mikedep333 joined #git
16:06 PettanShoutaKun I currently have a tag set 1.16.7 apparently
16:06 PettanShoutaKun is there a way to remove this tag?
16:06 theiostream joined #git
16:06 h12o joined #git
16:06 mikedep333 So my coworker tells me that it is bad for performance reasons to have a git repo where 2 branches are extremely different. (Even though they both have only plaintext.) Is that true?
16:06 _ikke_ git tag -d 1.16.7
16:06 PettanShoutaKun Also sorry, I disconnected earlier
16:06 PettanShoutaKun can I push the removal of that tag?
16:06 Introoter joined #git
16:06 PettanShoutaKun also are tags pushed by default?
16:07 _ikke_ mikedep333: nah, should not matter at all
16:07 mekhami left #git
16:07 mikedep333 _ikke_: Thank you.
16:07 HedgeMage mikedep333: It's not a big deal. The differences are so small as to be ignore-able even on legacy hardware.
16:07 _ikke_ PettanShoutaKun: You can, but note that others who have fetched the tag would still have that tag
16:07 theoceaniscool joined #git
16:07 PettanShoutaKun crap...
16:07 mikedep333 HedgeMage: Thank you.
16:07 PettanShoutaKun if I set a new tag does it overwrite the old tag?
16:07 _ikke_ no
16:07 PettanShoutaKun or do I get some form of frankentag?
16:07 PettanShoutaKun oh man
16:07 _ikke_ yes
16:07 PettanShoutaKun this is gonna get weird
16:08 _ikke_ tags are meant to be stable
16:08 _ikke_ ie, always point to the same commit
16:08 PettanShoutaKun hnnng
16:08 PettanShoutaKun wait!
16:08 PettanShoutaKun so if I have that 467 commits ago commit tagged
16:08 PettanShoutaKun and I tag THIS commit as 1.0
16:08 rokups joined #git
16:08 PettanShoutaKun when I run git describe --tag
16:09 PettanShoutaKun will I see the closest tag? i.e. 1.0-Bla-bae9482whatever
16:09 PettanShoutaKun or will both tags show up?
16:09 _ikke_ yeah, git will pick the first tag that it finds
16:09 mr-fantastic joined #git
16:09 qqx It will try to pick the closest tag, but I find it doesn't always to a good job of that.
16:10 dreiss joined #git
16:11 PettanShoutaKun well crap
16:11 PettanShoutaKun that would be unfortunate if it didn't
16:11 PettanShoutaKun I'm trying to use this to make release artifactory versions
16:11 PettanShoutaKun so if it picks the wrong tag, that could be bad....
16:12 thiago joined #git
16:12 madewokherd joined #git
16:13 qqx It should always use a tag that is an exact match for the current version, which is what you'd probably want to be doing for releases.
16:13 _ikke_ As long as you tag in the same branch line, it should not matter
16:13 _ikke_ right
16:13 Gustavo6046 joined #git
16:13 drodger joined #git
16:13 mr-fantastic Hey all, I have a problem with git that's been plaguing me for a couple weeks now and I'm in over my head at this point. I was wondering if someone could help me root cause what would make git-upload-pack hang for eternity?
16:13 PettanShoutaKun shafuuuuuuuu.... Are tags preserved on merges?
16:13 PettanShoutaKun and squashes?
16:14 _ikke_ PettanShoutaKun: merges create new commits, the tags just stash behind
16:14 _ikke_ stays*
16:14 synthroid joined #git
16:14 _ikke_ if you squash commits that have been tagged, the tag still points to the commit history that was not squashed
16:15 _ikke_ (you should not tag commits that you intend to squash anyway)
16:15 Vpa joined #git
16:15 PettanShoutaKun that's not good... especially considering I'd like people to be able to push their tags to master
16:15 mr-fantastic I don't mean to be a bother, so if anyone has the time I'll be in this chat or you can /msg me. Thanks all.
16:15 _ikke_ PettanShoutaKun: That does not make sense, you don't push tags to branches
16:15 todd_dsm joined #git
16:15 _ikke_ mr-fantastic: It often helps if you already start asking a question
16:15 PettanShoutaKun you can though... right?
16:16 _ikke_ PettanShoutaKun: branches and tags are both refs
16:16 theoceaniscool joined #git
16:16 _ikke_ A tag points to a commit, it does not care what branch that commit is on
16:16 mr-fantastic _ikke_: git-upload-pack on the server seems to hang forever, preventing people from doing a git pull or git push roughly 50% of the time until all git processes are killed.
16:16 qqx PettanShoutaKun: Tags point at a specific commit. Tags are *never* contained in branches.
16:16 mr-fantastic the problem seems to resurface after about 45 minutes to an hour.
16:16 svm_invictvs joined #git
16:17 _ikke_ mr-fantastic: Some kind of job running on the server?
16:17 PettanShoutaKun fffffuuuuuuuuuuuudge....
16:17 _ikke_ ...
16:17 qqx PettanShoutaKun: But any number of branches can contain a commit that is tagged.
16:17 XenophonF joined #git
16:18 MineCoins joined #git
16:18 _ikke_ PettanShoutaKun: What is the problem with that?
16:18 mr-fantastic _ikke_: nothing is in /etc/crontab. The server is also running bitbucket, but according to their support there doesn't seem to be anything wrong on that end.
16:18 PettanShoutaKun I'm really just trying to implement the ability to automatically version
16:18 PettanShoutaKun and tags seemed like a cool way to pin specific commits to specific versions in artifactory
16:18 Cabanossi joined #git
16:19 PettanShoutaKun as well as provide an auto release versioning system of sorts
16:19 _ikke_ And what is preventing you from doing that?
16:19 drodger joined #git
16:20 _ikke_ When you tag a specific commit, you really mean *that* commit, not something that has changed in the mean time
16:20 PettanShoutaKun but it seems like they get left behind which would be undesired... the job that pulls down the repo would need to be able to see the tags in master
16:20 PettanShoutaKun and people can't push straight to master.... so they can't directly push tags
16:20 PettanShoutaKun all commits into master are merge commits from pull requests
16:21 PettanShoutaKun or squash commits from merge reqests
16:21 _ikke_ PettanShoutaKun: You can push tags regardless on what branch they are
16:21 blackpajamas joined #git
16:21 _ikke_ tags live on their own
16:21 PettanShoutaKun and they would NEED to be there otherwise the artifactory publish would error when it runs git describe --tag and doesn't get a tag back
16:21 PettanShoutaKun and only gets a commit
16:23 lambada355 joined #git
16:23 _ikke_ But it's not common that arbitrary people create tags
16:23 _ikke_ especially if they don't have access to the stable branch
16:23 re1 joined #git
16:23 jstimm joined #git
16:23 PettanShoutaKun yeah but this means someone with push access to master has to push the tag up to master?
16:24 PettanShoutaKun I guess?
16:24 theoceaniscool joined #git
16:24 nexemacs joined #git
16:25 qqx PettanShoutaKun: I'd expect in that case you'd merge a change into master, then tag the merge commit.
16:25 PettanShoutaKun my problem is.... I want to trust people who want to version to be able to version....
16:25 qqx Exactly how that needs to be done would depend on how the repo is managed.
16:25 PettanShoutaKun but I don't want to trust them all to be able to merge into master
16:25 PettanShoutaKun *to push straight into master
16:26 qqx It generally doesn't make sense to tag something before it is merged.
16:26 PettanShoutaKun I wonder if I can add some auto incrementing tag system to my jenkins job on master
16:26 _ikke_ right
16:26 qqx That is likely to lead to multiple people trying to use the same version or versions being out of order.
16:26 _ikke_ PettanShoutaKun: who is going to decide what to call version 1.0?
16:26 drodger joined #git
16:27 PettanShoutaKun I don't know <_>
16:27 PettanShoutaKun that's the problem
16:27 _ikke_ Usually someone who is making the releases takes care of that
16:27 PettanShoutaKun everyone is freaking out because everyone is like WE NEED CONTINUOUS RELEASE RIGHT NOW
16:27 ChrisEadle joined #git
16:27 PettanShoutaKun but none of the systems for this are currently in place
16:27 PettanShoutaKun so people are freaking out when "snapshots" go into "release"
16:28 osse you can release a commit. it's version is simply the output of "git describe". And for ASRs (Actually Serious Release) then the commit will have nice tag, so git describe will print a nice string.
16:28 osse The End.
16:28 _ikke_ right
16:29 _ikke_ you cannot call something v1.0 before you know you going to release it as v1.0, so before you ran your tests and completed your build, it does not make sense to tag anything
16:30 _ikke_ PettanShoutaKun: You can always create a separate tag namespace ofcourse, but it's kind of useless to require tags for this
16:30 PettanShoutaKun that makes sense
16:30 _ikke_ usually the CI system after completing the build and performing all necessary steps tags that version
16:31 _ikke_ not before
16:31 _ikke_ That's where the build number comes from some projects use in their versioning
16:32 osse I once had to "re-tag" something because I found a big in how we created installers
16:32 PettanShoutaKun so what I should do.... I think.... is in the Jenkins job that runs all the tests and stuff....when it gets to deploy, it should get the current tag, increment it, commit that tag, push that tag?, then run release publish?
16:32 _ikke_ you don't commit a tag
16:32 osse up to when we had tags like v1.2.3-beta and v1.2.3-rc.  Once v1.2.3 came along creating installers broke because there was no dash in the tag name :D
16:32 _ikke_ but I understand what you mean
16:33 _ikke_ osse: d'oh
16:33 _ikke_ that's anoying
16:33 _ikke_ Why would that be a requirement :-/
16:33 _ikke_ PettanShoutaKun: but that's the basic gist
16:34 Mewho joined #git
16:34 _ikke_ PettanShoutaKun: You should think about whether each build should do a coplete version bump, or just increase a build number
16:34 TheSimonator joined #git
16:35 _ikke_ PettanShoutaKun: ie, if I create a pull request, you merge it, would that mean a new version of the software is released?
16:35 Noldorin joined #git
16:35 Mewho Hi, I'm trying to use rpmbuild to make an rpm from git 2.12.0 source, but it doesn't seem like a spec file is included in the source any longer.
16:36 PettanShoutaKun hmmmm _ikke_ I'm not sure how I would make that decision honestly
16:36 PettanShoutaKun we're sort of pre 1.0 at this point....
16:36 _ikke_ Mewho: https://public-inbox.org/git/4c42a1f4-4f03-0fdf-8bd2-8a7f1f978073@accelya.com/
16:36 Mewho Thanks
16:37 hobodave joined #git
16:37 PettanShoutaKun so I'm almost wondering if I could just tag something as 0.1 and then just use the auto generated tag/hash
16:37 HardlySeen joined #git
16:37 PettanShoutaKun and never change the tag until we actually hit a 1.0 release
16:37 _ikke_ PettanShoutaKun: what git describe returns is not a tag though
16:38 zeroed joined #git
16:38 PettanShoutaKun it just returns a string?
16:39 _ikke_ yes
16:40 _ikke_ the string often includes a tag of course
16:40 Mewho So based on that email thread, is there currently no way to use rpmbuild to make a git 2.12 0 rpm from source?
16:40 _ikke_ Mewho: Not an rpmbuild provided by git
16:40 _ikke_ Mewho: specific distros have their own rpmbuilds
16:41 Mewho But where does the spec file come from?
16:42 _ikke_ the maintainers of the packes for the respective distros create and maintain them
16:42 _ikke_ Mewho: just like git does not provide build files for debian, archlinux, alpine, etc. Each distro creates their own
16:42 Mewho So where would I get the CentOS spec file?
16:43 _ikke_ Probably, yes
16:43 peterbec` joined #git
16:43 PettanShoutaKun is there a way to commit a tag inline? like without vim popping up
16:43 Mewho I'm just confused now
16:43 PettanShoutaKun create a tag?
16:44 _ikke_ PettanShoutaKun: -m to provide the message
16:44 PettanShoutaKun what am I trying to say? Then how do I do that <_>
16:44 PettanShoutaKun thank you!
16:44 _ikke_ Mewho: git provides the Makefile to actually build git
16:44 _ikke_ after that, it's up to the distro to create a package out of that
16:44 cdown joined #git
16:45 Mewho So what do I need to do to create a CentOS rpm then?
16:46 _ikke_ Mewho: I think you need to get the source repo for git on centos
16:46 _ikke_ source package*
16:46 _ikke_ https://wiki.centos.org/HowTos/RebuildSRPM
16:47 _ikke_ This is more of a centOS question than a git question thogh
16:47 _ikke_ though
16:47 h12o joined #git
16:48 Mewho OK, thanks
16:48 Cabanossi joined #git
16:49 indistylo joined #git
16:49 PettanShoutaKun What should I provide as a message for a tag?
16:49 Darren_ joined #git
16:49 PettanShoutaKun "Auto-tagged by Jenkins CI" ?
16:51 _ikke_ Any relevant information from the build itself?
16:51 s1scha joined #git
16:52 tyreld joined #git
16:53 qqx It's usually good to include the tag name in the message as well. Since the ref name that will be used isn't part of the calculation for the sha1 that IDs the tag.
16:54 _ikke_ right, that's what I do too
16:54 _ikke_ qqx: Was looking for a reason why, but that's a very good one
16:54 ambro718 joined #git
16:57 re1_ joined #git
16:57 sz0 joined #git
16:57 prg3 joined #git
16:59 flavius joined #git
17:03 matsaman joined #git
17:04 gareppa joined #git
17:04 NeXTSUN joined #git
17:05 paws_ joined #git
17:07 mr-fantastic _ikke_: I have gathered some facts about the server git is running on, as well as my attempt to debug it with dtrace. https://pastebin.com/gp02NGn7
17:08 ShapeShifter499 joined #git
17:08 hashtagstrashtag joined #git
17:10 mr-fantastic I also checked some other servers we have for crontab. The redmine one does a pull every 30 minutes.
17:12 tnecniv joined #git
17:13 Es0teric joined #git
17:15 IntruderSRB joined #git
17:17 peepsalot joined #git
17:18 synthroi_ joined #git
17:18 serdar joined #git
17:21 ChrisEadle joined #git
17:22 gnixev joined #git
17:22 mikedep333 left #git
17:22 h12o joined #git
17:23 freimatz joined #git
17:24 tnecniv joined #git
17:26 TheSimonator joined #git
17:27 cd-rum joined #git
17:29 duderonomy joined #git
17:30 cdown joined #git
17:33 pks joined #git
17:34 drodger joined #git
17:39 pijiu joined #git
17:40 ResidentBiscuit joined #git
17:43 maxxe joined #git
17:44 peterbec` joined #git
17:45 maxxe is it ok to commit image files (.png etc) to a repo?
17:46 nedbat maxxe: if they are part of your work product, then why not?
17:46 nedbat maxxe: that is, yes.
17:46 acetakwas joined #git
17:46 maxxe okey
17:47 marianina8 joined #git
17:47 marianina8 joined #git
17:48 maxxe have been careful to not commit any type of binaries before
17:48 Cabanossi joined #git
17:49 _ikke_ git does not mind binary files
17:49 _ikke_ it's just that large often changing binary files will bloat the repo
17:50 maxxe yeah, ok
17:50 MarioBranco joined #git
17:51 nedbat maxxe: and there's no reason to commit built binaries, only edited binaries.
17:52 hobodave joined #git
17:52 maroloccio joined #git
17:53 h12o joined #git
17:54 wizardedit joined #git
17:54 maxxe ok. not sure what you mean with edited binaries..
17:56 drodger joined #git
17:56 publio joined #git
17:56 orbyt_ joined #git
17:57 serdar joined #git
17:58 Es0teric joined #git
17:59 Emperor_Earth joined #git
18:00 _ikke_ maxxe: files you change
18:00 JanC_ joined #git
18:00 Sasazuka joined #git
18:01 maxxe no reason to commit built binaries but only files that change?
18:03 thiago joined #git
18:04 Murii joined #git
18:05 bronson joined #git
18:05 nedbat maxxe: there are files that are your original work: source files, .png files you draw, etc.
18:05 teroshan joined #git
18:05 nedbat maxxe: then there are files that are produced as part of a build process: .o, .dll, .exe, .zip, etc.  Those are built files.
18:05 nedbat maxxe: there's no need to commit built files, only edited files, the original work.
18:06 maxxe alright. I see
18:06 maxxe now I understand
18:07 maxxe want to provide precompiled binaries for windows though. but I'll host them on some server.
18:07 preaction if you're using github, you can use github releases or github pages or git lfs
18:07 synthroid joined #git
18:08 maxxe yes. thanks.
18:09 YuGiOhJCJ joined #git
18:11 awc737 joined #git
18:12 rastos3 joined #git
18:15 nexemacs_ joined #git
18:16 nixjdm joined #git
18:16 dminuoso joined #git
18:16 dminuoso Hi. Is there a way to set a change aside in such a way to keep in the in the wd but outside the index, and not have it considered as dirty?
18:17 dminuoso Basically I got a modification that I need for a while in development, and I really dont want to see it in git add -p all the time.
18:17 dminuoso (I mean I suppose I could just commit it to a branch and rebase it out at the end, but I'd like to know whether there's an alternate way)
18:18 h12o joined #git
18:20 qqx dminuoso: Committing it and later rebasing it out is the best way I know of.
18:20 qqx And avoiding accidentally pushing those types of commits somewhere undesired was my main motivation for implementing pre-push hooks.
18:21 xPucTu4 joined #git
18:21 dminuoso qqx: What kind of hook would that be?
18:21 dminuoso qqx: Or do you tag those commits in such a way, that the hook would prevent a push?
18:22 qqx I start the message for those types of commits with 'WIP', and have a pre-push hook which will refuse to send those to important branches.
18:23 qqx The example pre-push hook that's included when you initialize a new repo should already be setup for that.
18:23 xPucTu4 after upgrading to Visual Studio 2017 my git does not work
18:24 xPucTu4 i get this message "error: cannot spawn .git/hooks/pre-push: No such file or directory"
18:24 xPucTu4 and cannot find answers on google
18:24 dminuoso qqx: Imagine that. The timing.. :-)
18:24 qqx Yeah.
18:25 qqx No idea about that though, I don't do Windows.
18:25 Fallen0223 joined #git
18:25 ShapeShifter499 joined #git
18:25 xPucTu4 also the channel's web page there is a link to a public log, but it does not work too (timeout)
18:27 Eugene !logs
18:27 gitinfo [!irclog] Public logs of #git are kept at: http://colabti.org/irclogger/irclogger_log/git or look at stats for the last 30 days: http://itvends.com/irc/git.html or since Jan 1, 2012: http://itvends.com/irc/git-all.html
18:27 Eugene I guess I could post my logs directory
18:28 Eugene I'm not setting up a fancy browser like colabti had
18:28 bannakaf_ joined #git
18:32 xPucTu4 if you can give us access to the cannel logs it will be great. or at least check if some of the logs contains "cannot spawn .git/hooks/pre-push: No such file"
18:36 ski7777 joined #git
18:39 theoceaniscool joined #git
18:40 tristanp_ joined #git
18:41 lekare joined #git
18:44 paul424 joined #git
18:46 Eugene .trigger_edit irclog Public logs of #git are kept at: https://itvends.com/irc/git/ or see stats for the last 30 days: http://itvends.com/irc/git.html or all logs: http://itvends.com/irc/git-all.html
18:46 gitinfo Eugene: Okay.
18:46 h12o joined #git
18:46 TbobbyZ_ joined #git
18:49 tnecniv joined #git
18:50 nexemacs joined #git
18:52 raynold joined #git
18:59 nexemacs joined #git
19:00 gareppa joined #git
19:01 cr34ton joined #git
19:01 TinkerTyper_ joined #git
19:03 Cabanossi joined #git
19:05 Silenced_v2 joined #git
19:05 Silenced_v2 joined #git
19:06 TbobbyZ joined #git
19:09 Acerific joined #git
19:10 finalbeta joined #git
19:10 cdown joined #git
19:10 bodkan joined #git
19:14 gopar joined #git
19:20 MrWoohoo joined #git
19:21 TheSimonator joined #git
19:26 tymczenko joined #git
19:26 MindSpark joined #git
19:27 MindSpark hey, do I have to have a fork in order to create a PR using git request-pull ?
19:27 _moogs joined #git
19:27 Dougie187 git request-pull has nothing to do with github pull requests
19:27 chipotle joined #git
19:29 MindSpark oh, then I must've misunderstood
19:30 TheSimonator joined #git
19:30 MindSpark ok, I'll try to rephrase then. I can edit files on a github on a specific repository and when I do so it automatically opens a pull request. Somehow without me forking the project.
19:31 MindSpark Is there a way to clone it locally, do the changes then create the PR similarly?
19:31 Eugene MindSpark - "Pull Requests" are a feature on Github that is emulated elsewhere, and works from changes in a different branch or repository. "git request-pull" is a `git` command(to be run on your local machine) that summarizes changes, so that you can write an  email requesting somebody to pull your changes. THey are similar in concept, but not in function
19:32 Eugene Sure, `git clone` the repo locally; make your changes on a different branch, push them into github on that branch, and then go create the PR on the website
19:32 Dougie187 Where clone and push might be to a fork.
19:32 Dougie187 Depending on if you have permissions to the repo or not.
19:33 MindSpark ah, so request-pull doesn't really *do* anything besides outputting summary info?
19:33 Dougie187 Right
19:34 MindSpark ok... but if I don't have access to the repo. I still can do requests on github, by editing the file on the website... is that restricted to only the website?
19:34 TSS_ joined #git
19:34 Dougie187 No, you can even fork, clone, branch, commit, push, PR
19:36 MindSpark alright, thanks. Getting kicked out of the office
19:36 Dougie187 NP
19:36 MindSpark will have to continue this tomorrow
19:36 MindSpark thanks a lot for the enlightenment :D
19:39 afuentes joined #git
19:41 sathed joined #git
19:44 nickel_dime_look joined #git
19:45 Atm0spher1c joined #git
19:46 drodger joined #git
19:46 peterbec` joined #git
19:49 Cabanossi joined #git
19:49 Anja joined #git
19:51 Anja joined #git
19:51 vuoto joined #git
19:54 livingstn joined #git
19:55 moei joined #git
19:56 gopar joined #git
19:58 hahuang65 joined #git
19:58 jstimm joined #git
19:59 gregor2 joined #git
19:59 mikecmpbll joined #git
20:03 Tobbi joined #git
20:04 gopar joined #git
20:04 idlemind joined #git
20:04 drodger joined #git
20:05 mda1 joined #git
20:07 nick123 joined #git
20:09 mr-fantastic Hey all, I'm back. I have an update to my original question about git pull hanging. Here's a client side trace I just learned how to make that shows it hanging on 'git-upload-pack'. Does anyone have any advice on how to continue to root cause my problem? https://pastebin.com/V1P1iJkD
20:11 mr-fantastic for context, on the client side 'git pull' will sometimes hang. This is what it looks like with the trace variables set.
20:12 holodoc joined #git
20:13 raijin joined #git
20:13 Eugene mr-fantastic - you can replace your git-update-pack binary with a wrapper script that adds debug flags
20:13 Eugene (on the server)
20:14 drodger joined #git
20:14 lewix joined #git
20:14 mr-fantastic Eugene: excellent idea, I'll write that up and get back to you.
20:14 Eugene Or, invoke it manually by calling `ssh <host> git-upload-pack` with the flags you need
20:14 cdown joined #git
20:15 Eugene mr-fantastic - a common problem that might save you some time: how big is the repository and how much RAM does your server have?
20:16 shgysk8zer0 joined #git
20:16 marianina8 joined #git
20:17 matsaman joined #git
20:17 satifant joined #git
20:17 mr-fantastic Eugene: The largest git repo we have is 632M, the server has about 140 GB of ram, of which at any moment about 80GB is free.
20:17 Shayanjm joined #git
20:18 Shayanjm Hi there. I have two branches I'd like to merge while preserving merge history. See: https://gist.github.com/shayanjm/7ed72a044d433cea398d60c89dc209ad
20:18 Eugene So probably not OOMkiller, heh. We see this question from people on a 256MB VPS about once a month
20:19 Es0teric joined #git
20:19 Shayanjm basically, branch1 and branch2's "initial commits" are functionally identical. I'd like to swap branch1's existing "initial commit" for the one in branch2
20:19 Shayanjm how would I go about doing that without squashing all of the commits stacked on top?
20:24 Dougie187 left #git
20:24 serdar joined #git
20:27 re1 joined #git
20:28 mattcen joined #git
20:30 ajeremias ei.. i have a server with a custom ssh port.. and git fails to clone git clone git@framagit.org:myusername/nameofrepo.git:22
20:30 ajeremias says: ssh: connect to host framagit.org port 1700: Network is unreachable
20:30 ajeremias how can I set the port?
20:31 narcberry joined #git
20:31 MineCoins joined #git
20:31 AaronMT joined #git
20:31 ambro718 joined #git
20:31 tymczenko joined #git
20:33 qqx ajeremias: Use a ssh:// URL: ssh://git@framagit.org:port/myusername/nameofrepo.git
20:34 canton7 ajeremias, (see "git urls" in man git clone)
20:34 gitinfo ajeremias: the git-clone manpage is available at https://gitirc.eu/git-clone.html
20:35 ajeremias awesome :D
20:35 Shayanjm any ideas on this?: https://gist.github.com/shayanjm/7ed72a044d433cea398d60c89dc209ad
20:35 canton7 .. and you'll probably want ssh://git@framagit.org:port/~username/repo.git, with the tilde
20:39 Es0teric joined #git
20:41 mr-fantastic Eugene: https://pastebin.com/tqpfXW6T
20:42 mr-fantastic It seems to hang at the last 0000
20:42 mr-fantastic sometimes the last line will be 0000fatal: The remote end hung up unexpectedly, but most of the time it'll just stay there.
20:44 bilal_ joined #git
20:44 mr-fantastic The trace points here: https://github.com/git/git/blob/v2.11.0/pkt-line.c#L80 if I'm to understand the trace correctly.
20:44 zerow joined #git
20:45 Anja_ joined #git
20:45 bilal_ left #git
20:47 peterbec` joined #git
20:48 bilal joined #git
20:50 snowkidind joined #git
20:50 DigitallyBorn joined #git
20:50 h12o joined #git
20:51 DigitallyBorn Crazy question: can I somehow use environment variables in a git commit template?
20:51 DigitallyBorn I'm thinking not
20:53 gregor2 joined #git
20:56 canton7 DigitallyBorn, there's a hook to generate commit messages
20:56 canton7 and I'm sure you can use env vars there :)
20:58 ertesx joined #git
20:59 DigitallyBorn ooo .. ok
20:59 ChrisEadle joined #git
21:01 Lunatrius` joined #git
21:01 mr-fantastic DigitallyBorn: for an example, see https://gist.github.com/bartoszmajsak/1396344
21:01 re1_ joined #git
21:01 mr-fantastic note: `sed -i` is non-portable tho
21:02 Darren_ joined #git
21:03 Darren_ joined #git
21:03 cdg_ joined #git
21:04 aidalgol joined #git
21:05 overlord_tm joined #git
21:05 jeffreylevesque joined #git
21:06 Eugene mr-fantastic - You are two steps deeper into the git C code than I have previously ventured. Good luck.
21:07 mr-fantastic Eugene: Oh geez
21:07 mr-fantastic this is gonna end with a pull request, I just know it. :)
21:07 anuxivm joined #git
21:07 dsdeiz joined #git
21:07 dsdeiz joined #git
21:08 fission6 joined #git
21:09 fission6 a colleague has a small feature branch, i want to check it out and basically bring the branch and its file changes all back to masters versions and then adjust one or two files how can i do that. usually i would just do a new branch off master and delete the old one but feel not sure if this is as good as what im proposing
21:10 jml1017 joined #git
21:10 fission6 whats the best workflow for a situation where a colleague has a topic branch and PR but nearly all their changes are not right and i want to revert it all back to master and then make a few small changes
21:10 fission6 (again i can just blow the branch away if thats the right thing to do and reissue a new PR)
21:10 TheSimonator joined #git
21:11 kfoonamalik joined #git
21:13 Random832 joined #git
21:16 mr-fantastic DigitallyBorn: Your question piqued my interest so I wanted to see if I could do a quick one liner for a commit message.
21:16 mr-fantastic You can see what I came up with here: https://pastebin.com/h8NNbLfj
21:17 DigitallyBorn mr-fantastic: https://pastebin.com/m73r7aEF
21:17 SegFaultAX joined #git
21:18 DigitallyBorn mr-fantastic: I didn't actually need an environment variable once I noticed it was a shell script
21:19 Cabanossi joined #git
21:19 DigitallyBorn mr-fantastic: ... but that's what I'm doing ... adding what I'm listening to at the bottom of my commit message :D
21:19 DigitallyBorn mr-fantastic: Just a moment of distraction for a little fun
21:20 mr-fantastic Neat. Note tho, your script will actually put the data at the end of the file, after the part of the commit message that ignores the following lines.
21:20 mr-fantastic replace line 4 of your script with: printf "1i\nListening to: $artist - $track\n.\nwq\n" | ed -s "$1"
21:21 phase2 joined #git
21:22 askb joined #git
21:23 Darcidride joined #git
21:23 fission6 any ideas on my workflow above?
21:25 snowalpaca joined #git
21:26 qqx fission6: Unless there's some reason that you need to preserve the info that your colleague worked on that, I'd likely just start anew and manually pull in the usable bits.
21:32 Celelibi joined #git
21:37 fission6 ok
21:39 TheSimonator joined #git
21:42 mattcen joined #git
21:45 Endarked joined #git
21:46 Dougie187 joined #git
21:49 alexandre9099 joined #git
21:49 rossome joined #git
21:52 dmarr left #git
21:55 jimi_ joined #git
21:56 deltab mr-fantastic: I thought it was only #-prefixed lines that are removed, hence "Show unified diff [...] at the bottom of the commit message template. Note that this diff output doesn’t have its lines prefixed with #."
21:57 re1 joined #git
21:58 mr-fantastic deltab: Yeah, sometimes I forget I have an alias for commit that goes 'commit --verbose', which adds all the source that changed and a line that says "do not remove the above line all lines below will be removed" or something to that effect.
22:03 Dougie187 left #git
22:04 finalbeta joined #git
22:04 deltab mr-fantastic: heh, if I use --cleanup=verbatim, the comments stay, but the diff is still removed!
22:07 sunri5e joined #git
22:08 mr-fantastic gotta go. Thanks for the help all. It got me a little bit closer to an answer/fix.
22:08 masuberu joined #git
22:09 Garen joined #git
22:12 kristofferR joined #git
22:14 manuelschneid3r joined #git
22:16 cdg joined #git
22:18 sonOfRa joined #git
22:19 Xenophon1 joined #git
22:19 gregor2 joined #git
22:19 kristofferR joined #git
22:21 peterbec` joined #git
22:26 kelthan joined #git
22:26 bannakaffalatta joined #git
22:28 phase2 joined #git
22:28 diogenese joined #git
22:35 gopar joined #git
22:39 glebihan_ joined #git
22:41 MrWoohoo joined #git
22:43 atomi joined #git
22:43 awc737 joined #git
22:44 kbeflo joined #git
22:44 Fallen0223 joined #git
22:47 gopar joined #git
22:49 pifon joined #git
22:54 msonntag joined #git
22:55 porfiriopaiz joined #git
22:57 jimi_ joined #git
22:58 LordRyan joined #git
23:00 SpeakerToMeat joined #git
23:02 tnecniv joined #git
23:02 cads2 joined #git
23:03 gopar joined #git
23:04 cads2 hey guys, quick question - suppose that I have beautified my code or otherwise made some global change which affects a lot of lines of code, and then I made the mistake of making /further/ changes which affects the function of the code. The file is now ahead of the previous commit by a large number of changes and it's difficult to tell which changes are simply cosmetic and which changes have altered the behavior of the code. How should I resolve this?
23:04 cads2 My current approach is as follows:
23:05 jstvz joined #git
23:05 cads2 first, stash the current work. Then make the cosmetic changes and commit those. Finally, merge the stashed changes back in, resolving conflicts as necessary.
23:06 rwp cads2, Sounds like a good plan to me. Let me cheer you on from the sidelines.
23:06 dirtyroshi joined #git
23:07 rwp Stashing is fine but personally I tend to check those into a real branch for later use.
23:07 cads2 I've done this a couple times now, it seems to work - The rationale is that the stash and the beautified commit code will for the most part agree on all of the lines except the functionally different ones
23:07 Derperperd joined #git
23:07 pietv joined #git
23:07 rwp Sounds good.l
23:07 rwp Sounds good.
23:08 rwp I also make heavy use of 'git add -p' in order to split changes and commit them separately.
23:10 jameser joined #git
23:11 samgaw_ joined #git
23:12 shadow-x joined #git
23:12 chipotle joined #git
23:12 gopar joined #git
23:14 cads2 rwp, yeah, I certainly love me some patch mode
23:14 jstvz joined #git
23:15 cads2 I sometimes add multiple features and only later realize "hey, I'd really like it if the commit narrative would make me look a little less ADHD here"
23:16 Derperperd joined #git
23:16 LBV_User joined #git
23:16 shadow-x joined #git
23:23 unplayednamer joined #git
23:23 shadow-x joined #git
23:23 kbeflo_ joined #git
23:26 shadow-x_ joined #git
23:38 drodger joined #git
23:41 sunri5e joined #git
23:44 Fallen0223 joined #git
23:48 drodger joined #git
23:48 heroux joined #git
23:48 Cabanossi joined #git
23:49 phase2 joined #git
23:50 Fallen0223 joined #git
23:51 netj joined #git
23:54 jimi_ joined #git
23:57 Domhack joined #git
23:57 sturner joined #git
23:59 uruk joined #git

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