Perl 6 - the future is here, just unevenly distributed

IRC log for #git, 2017-04-20

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

All times shown according to UTC.

Time Nick Message
00:01 menip joined #git
00:03 thiago joined #git
00:04 Noodlewitt joined #git
00:07 qpdb joined #git
00:09 bruce_lee joined #git
00:09 jimi_ joined #git
00:12 alexandre9099 joined #git
00:15 drodger joined #git
00:16 ToBeCloud joined #git
00:18 HurricaneHarry joined #git
00:20 Starky joined #git
00:23 nexemacs joined #git
00:26 drodger joined #git
00:27 durham_ joined #git
00:30 jstimm joined #git
00:32 unplayednamer joined #git
00:32 ProLoser joined #git
00:37 jimi_ joined #git
00:43 theiostream joined #git
00:46 sammyo joined #git
00:47 Gsham joined #git
00:48 Domhack joined #git
00:49 drodger joined #git
00:52 cdown joined #git
00:56 fstd_ joined #git
01:01 jimi_ joined #git
01:03 duderono_ joined #git
01:03 Cabanossi joined #git
01:07 HoierM joined #git
01:10 justanotheruser joined #git
01:13 tnecniv joined #git
01:13 m1dnight_ joined #git
01:17 batrick joined #git
01:17 NeXTSUN joined #git
01:24 msonntag joined #git
01:24 chipotle joined #git
01:25 nexemacs joined #git
01:26 systemovich joined #git
01:27 finalbeta joined #git
01:31 kadoban joined #git
01:32 LeBlaaanc joined #git
01:46 dsdeiz joined #git
01:46 dsdeiz joined #git
01:48 Cabanossi joined #git
01:51 scarabx joined #git
01:52 dsdeiz_ joined #git
01:52 netj joined #git
01:56 uruk joined #git
01:56 bannakaf_ joined #git
01:57 Kaisyu joined #git
01:59 qpdb joined #git
02:03 ojdo joined #git
02:05 sucks joined #git
02:06 sunri5e joined #git
02:10 Goplat joined #git
02:15 aidalgol joined #git
02:18 tnecniv joined #git
02:18 NoOutlet joined #git
02:20 fission6 joined #git
02:22 fission6 joined #git
02:22 duderonomy joined #git
02:25 dsdeiz joined #git
02:25 dsdeiz joined #git
02:25 jameser joined #git
02:27 cqi joined #git
02:32 chipotle joined #git
02:33 webdev007 joined #git
02:33 danimal2 joined #git
02:36 bruce_lee joined #git
02:36 bruce_lee joined #git
02:37 tinnn joined #git
02:39 dreiss joined #git
02:43 tinnn joined #git
02:43 tinnn how can I add a local branch master and set it's upstream to a remote branch in one go? I can't use clone
02:44 lb1c joined #git
02:47 Saint_Isadore joined #git
02:49 Vortex34 joined #git
02:51 cagedwisdom joined #git
02:52 bronson joined #git
02:53 aavrug joined #git
02:54 lagothri1 joined #git
02:55 oskarkv joined #git
02:55 fbwnd joined #git
02:56 ojdo joined #git
02:58 chachasmooth joined #git
02:59 gregor2 joined #git
02:59 agileadam joined #git
02:59 hexagoxel joined #git
03:01 a_thakur joined #git
03:02 LordRyan git init; git remote add origin <URL>;
03:02 LordRyan I dunno how to set upstream offhand. I'm not at a computer
03:03 jstimm joined #git
03:04 onehrxn joined #git
03:04 Cabanossi joined #git
03:06 sucks_ joined #git
03:07 chachasmooth joined #git
03:11 kbeflo__ joined #git
03:12 ShapeShifter499 joined #git
03:16 ZyX-I joined #git
03:17 tinnn thanks LordRyan
03:25 LeBlaaanc joined #git
03:27 SwiftMatt joined #git
03:30 jimi_ joined #git
03:30 justan0theruser joined #git
03:32 govg joined #git
03:35 durham joined #git
03:39 pifon joined #git
03:40 AfroThundr joined #git
03:47 Cabanossi joined #git
03:48 cdown joined #git
03:50 navidr joined #git
03:52 bilal joined #git
03:53 jost_ joined #git
03:55 alexggordon joined #git
03:56 t-mart joined #git
03:59 nexemacs joined #git
04:06 gregor2 joined #git
04:09 rscata joined #git
04:10 fahadash joined #git
04:13 LeBlaaanc joined #git
04:13 re1_ joined #git
04:26 fakenerd joined #git
04:27 PaulCapestany joined #git
04:28 cdown joined #git
04:38 pbrewczynski_c joined #git
04:41 matoro joined #git
04:44 mda1 joined #git
04:44 ayogi joined #git
04:44 chipotle joined #git
04:45 Darren_ joined #git
04:48 aw1 joined #git
04:51 novice0310 joined #git
04:52 mda1 joined #git
04:52 Noodlewitt joined #git
04:52 diogenese joined #git
04:52 bronson joined #git
04:55 nexemacs joined #git
04:57 Lucky-- joined #git
04:58 re1 joined #git
04:58 Lucky-- So I took over a project from another developer. At the time I had the Nov '16 codebase. I committed that to master, and started refactoring that in a branch called "interface refactor." In between then and now, I had to do some edits to the master branch for the stakeholders. I have now foundout that there was a Dec 16
04:58 freimatz joined #git
04:59 Lucky-- code base and it has substantial edits that I need to add to my branch among other things. Any tips to handle this situation?
04:59 Peetz0r joined #git
04:59 MeltedLux joined #git
05:02 svm_invictvs joined #git
05:02 Cabanossi joined #git
05:06 nexemacs joined #git
05:07 safe joined #git
05:07 peepsalot joined #git
05:07 _ikke_ commit it on top of the original code base, then merge it in, deal with conflicts?
05:10 overlord_tm joined #git
05:13 Paper-Mario joined #git
05:14 tymczenko joined #git
05:14 rwp Lucky--, Like _ikke_ said. That would work. Or my preference would be if my branch was private only that I would rebase my own changes on top of the shared master and deal with the conflicts there.
05:19 ShapeShifter499 joined #git
05:26 teroshan joined #git
05:26 xsddz joined #git
05:27 jstimm joined #git
05:30 marianina8 joined #git
05:31 Saint_Isadore joined #git
05:31 treeman joined #git
05:36 gareppa joined #git
05:44 Silenced joined #git
05:46 jameser joined #git
05:47 mmattice joined #git
05:47 jimmyrcom joined #git
05:47 Cabanossi joined #git
05:48 ShapeShifter499 joined #git
05:49 shgysk8zer0 joined #git
05:49 fakenerd_ joined #git
05:51 ThomasLocke joined #git
05:51 ThomasLocke joined #git
05:53 bronson joined #git
05:53 waveclaw joined #git
05:54 fakenerd joined #git
05:56 Sasazuka joined #git
05:56 alexggordon joined #git
05:56 dreiss joined #git
05:57 inflames joined #git
05:57 fakenerd_ joined #git
05:58 qt-x joined #git
05:59 pietv joined #git
06:02 xall joined #git
06:02 serdar joined #git
06:03 MineCoins joined #git
06:09 peterbec` joined #git
06:09 fakenerd joined #git
06:11 danimal2 joined #git
06:12 nutron joined #git
06:13 Raging_Hog joined #git
06:16 Silmarilion joined #git
06:17 Cabanossi joined #git
06:19 fakenerd joined #git
06:19 malt3 joined #git
06:20 zeroed joined #git
06:20 freimatz joined #git
06:20 serdar joined #git
06:23 JeroenT joined #git
06:24 chele joined #git
06:25 fakenerd_ joined #git
06:26 acetakwas joined #git
06:26 ploop joined #git
06:26 sbulage joined #git
06:27 stoqn4o_pm joined #git
06:30 rokups joined #git
06:31 dermoth joined #git
06:31 aw1 joined #git
06:33 slidercrank joined #git
06:36 zeroed joined #git
06:36 zerow joined #git
06:41 onehrxn joined #git
06:42 jagob joined #git
06:42 sbulage joined #git
06:44 jaziz joined #git
06:46 jaziz hm
06:46 jaziz so I typed "git config credential.helper store"
06:46 jaziz and I changed my mind
06:46 jaziz but the "store" is still there
06:46 jaziz how do I remove?
06:48 osse jaziz: git config --unset credential.helper store
06:50 jaziz mmm
06:50 jaziz osse, thanks
06:52 rominronin joined #git
06:55 Raging_Hog joined #git
06:55 CrummyGummy joined #git
06:56 pigeon joined #git
06:57 alexggordon joined #git
07:00 dersand joined #git
07:00 pbandark joined #git
07:02 sargentmki joined #git
07:03 fakenerd joined #git
07:04 bilal_ joined #git
07:05 pbandark joined #git
07:05 _ng joined #git
07:06 paul424 joined #git
07:09 dvaske joined #git
07:09 lb joined #git
07:10 Tobbi joined #git
07:10 fakenerd joined #git
07:11 axisys_ joined #git
07:12 dvaske_ joined #git
07:14 MissionCritical joined #git
07:15 alexggordon joined #git
07:16 fakenerd_ joined #git
07:17 nevodka joined #git
07:17 imack joined #git
07:20 PtxDK joined #git
07:21 overlord_tm joined #git
07:21 kurkale6ka joined #git
07:22 zefferno joined #git
07:23 petersaints joined #git
07:23 cdown joined #git
07:24 rwp joined #git
07:25 fl0w joined #git
07:25 fl0w left #git
07:29 re1_ joined #git
07:30 MineCoins joined #git
07:31 ertes joined #git
07:32 ertes joined #git
07:36 masuberu joined #git
07:39 Panacea joined #git
07:42 schleppel joined #git
07:45 holodoc joined #git
07:46 Peetz0r joined #git
07:51 yarekt left #git
07:54 cr1tic joined #git
07:57 chardan Strategy question: I have a pretty large set of commits. I need to grab subsets of them into new branches. Is cherry-pick the right tool for this? And/or is there a way to say "hey, take commits a,b,c and move them into a new branch"? Thank you!
07:57 moritz chardan: cherry-pick accepts a range of commits too
07:57 TomyLobo3 joined #git
07:57 avih joined #git
07:57 chardan moritz: Cool, I will keep reading about it! Thanks.
07:58 moritz chardan: or you can start with a copy of another branch, and do an interactive rebase where you drop all the commits you don't want
07:58 chardan moritz: So to do that, I'd do something like "git checkout -b new_br" and then "git rebase -i <oldbr hashtag>"?
07:59 moritz chardan: yes
07:59 chardan moritz: Ok, those sound like they'll probably work for me! Thanks!
07:59 brk_ joined #git
08:03 King_Hual joined #git
08:04 skunkz joined #git
08:04 a_thakur joined #git
08:05 elect joined #git
08:05 matoro joined #git
08:07 stoqn4o_pm joined #git
08:08 mikecmpbll joined #git
08:09 chichou joined #git
08:09 AbleBacon_ joined #git
08:13 permalac joined #git
08:14 marcogmonteiro joined #git
08:16 YuGiOhJCJ joined #git
08:18 Lunatrius joined #git
08:19 Snugglebash joined #git
08:19 a_thakur joined #git
08:21 Introoter joined #git
08:21 clmsy joined #git
08:22 tymczenko joined #git
08:24 HardlySeen joined #git
08:27 marianina8 joined #git
08:27 Murii joined #git
08:28 chardan moritz: Next piece of the puzzle... My situation has a commit that the other commits depend on: D a b c. The eventual fate is that each of a,b,c will be in github PRs. Do I duplicate D so I have Da,Db,Dc, or is there A Better Way(TM)?
08:28 chardan moritz: (Or is the way this usually goes just to say "hey, this depends on this other thing"?)
08:28 Tuor joined #git
08:28 Tuor joined #git
08:29 avih left #git
08:29 chichou joined #git
08:31 jast chardan: so that sounds like you may want to create three branches, each based on D, and cherry-pick one of a, b and c to each. there's no one way that is significantly better than any other, though.
08:32 nod joined #git
08:32 mentazoom joined #git
08:32 svm_invictvs joined #git
08:33 svm_invictvs joined #git
08:33 chardan jast: Ok, that's "just the way it will have to be", then. Thank you!
08:34 svm_invictvs joined #git
08:35 svm_invictvs joined #git
08:35 monq joined #git
08:36 chll_ joined #git
08:37 causasui joined #git
08:38 inovas joined #git
08:38 Lunatrius joined #git
08:42 jstimm_ joined #git
08:46 Lunatrius joined #git
08:47 sarri joined #git
08:47 sarri joined #git
08:52 zylum joined #git
08:52 fakenerd joined #git
08:53 zefferno joined #git
08:55 bilal joined #git
08:56 borkr joined #git
08:57 cqi joined #git
08:57 fakenerd_ joined #git
08:58 onehrxn joined #git
09:00 marcogmonteiro joined #git
09:01 fakenerd joined #git
09:01 afuentes joined #git
09:02 raijin joined #git
09:03 Cabanossi joined #git
09:04 fakenerd_ joined #git
09:05 zxd joined #git
09:05 zxd why isn't 'git status' detecting the file has been renamed? it shows as deleted then as untracked file
09:05 fbwnd joined #git
09:06 grawity because that's really what a rename *is*
09:06 grawity if you add the file under the new name, Git will show it as a rename based on similarity
09:06 grawity or if you use `git mv` the next time instead of regular `mv`
09:06 SantaClauze joined #git
09:08 zxd grawity: I added the file
09:08 zxd still shows as new in green
09:08 Acerific joined #git
09:09 grawity did you change it a lot after renaming?
09:09 fakenerd joined #git
09:12 zxd no\
09:14 graingert is there a nice command to list any duplicate files under a case insensitve FS?
09:15 canton7 graingert, the clue is lots of modified files after cloning
09:15 fakenerd_ joined #git
09:15 graingert canton7: I've only got a case sensitive FS
09:15 graingert CI only has a case sensitive FS
09:15 graingert Prod only has a case sensitive FS
09:15 theoceaniscool joined #git
09:16 graingert all the other devs here are running silly macOs
09:17 canton7 aah, I see. Apparently `find . | sort -f | uniq -di` and `find . | tr '[:upper:]' '[:lower:]' | sort | uniq -cd` will do it
09:18 graingert git ls-files
09:18 graingert not find
09:18 canton7 whatever you want :P You didn't ask anything that was specific to git
09:19 graingert ...
09:20 graingert canton7: nice that works
09:20 acetakwas joined #git
09:22 fakenerd joined #git
09:23 alexggordon joined #git
09:24 Gloomy joined #git
09:24 tvw joined #git
09:27 cr1tic joined #git
09:28 fbwnd joined #git
09:28 cr1tic joined #git
09:29 cr1tic joined #git
09:33 bilal joined #git
09:34 finalbeta joined #git
09:39 skunkz Hi, I'm using git rebase -i for the first time: I have this history http://puu.sh/vqe3y/a70297d1ba.png and I would like to remove the branching so I did `git rebase -i 78a1292` and edited it like this http://puu.sh/vqe5a/71d6c5a056.png. But then I get a lot of conflicts which have already been resolved in my current HEAD, do I need to re-resolve them all ? I'm tempted to git rebase --skip as soon as I get
09:39 skunkz conflicts but I'm pretty sure that's not the right thing to do
09:40 brk_ joined #git
09:48 Cabanossi joined #git
09:49 zefferno joined #git
09:51 cads2 joined #git
09:54 SantaClauze joined #git
09:56 Karazhan joined #git
09:57 digidog joined #git
09:57 govg joined #git
10:01 navidr joined #git
10:03 Silenced joined #git
10:03 ferr joined #git
10:10 theoceaniscool joined #git
10:13 a_thakur joined #git
10:15 sathed_ joined #git
10:18 Cabanossi joined #git
10:20 earnestly joined #git
10:21 Guest85646 joined #git
10:21 cfoch-always joined #git
10:23 kurkale61 joined #git
10:23 hanthings joined #git
10:23 kurkale61 joined #git
10:25 schleppel joined #git
10:26 finalbeta joined #git
10:28 dedicated joined #git
10:29 a_thakur joined #git
10:30 dedicated joined #git
10:31 agileadam joined #git
10:32 Darcidride joined #git
10:33 a_thakur_ joined #git
10:33 telephone joined #git
10:33 telephone joined #git
10:34 pbandark1 joined #git
10:35 i7c Firstly, I want to question your goal to remove a merge. I don’t see the sense. :)
10:37 i7c And then, yes, you probaly will have conflicts because you are effectively changing the order in which changes are applied.
10:38 i7c skunkz, the question is, what do you want the outcome to be like? Do you want all changes from both branches but without a merge or do you want to remove the changes coming from one branch?
10:38 cbreak skunkz: skip would drop that commit
10:39 AfroThundr joined #git
10:39 oskarkv joined #git
10:39 kurkale61 joined #git
10:41 kurkale61 joined #git
10:42 kurkale6ka joined #git
10:48 nrajasekhar joined #git
10:51 dust_ joined #git
10:52 BlueNeXuS joined #git
10:52 nrajasekhar left #git
10:54 alexggordon joined #git
10:56 loskutak joined #git
10:56 theoceaniscool joined #git
10:59 star_prone joined #git
11:00 skunkz joined #git
11:01 kurkale6ka joined #git
11:02 alexandre9099 joined #git
11:03 Cabanossi joined #git
11:03 dust_ joined #git
11:03 skunkz i7c: well I gave up but I wanted to "flatten" the history so that there is no merge commit anymore, but I guess my mistake was not using git rebase -p
11:04 nevodka joined #git
11:04 skunkz so yeah I wanted changes from both branches but without spending time on resolving conflicts I had already resolved when the merge was done
11:07 i7c What’s wrong with a little merge commit ;)
11:08 cads2 joined #git
11:12 courrier joined #git
11:13 fakenerd joined #git
11:15 MarioBranco joined #git
11:16 a_thakur joined #git
11:17 finalbeta1 joined #git
11:18 xall joined #git
11:19 rnsanchez joined #git
11:19 theoceaniscool joined #git
11:19 kurkale6ka joined #git
11:21 schleppel joined #git
11:21 kurkale6ka joined #git
11:21 skunkz Well mostly for my personal understanding of git, and also because people from my team often merge a into a when they could pull --rebase so it would be nice for me to know how to clean this mess lol
11:21 anonymoose joined #git
11:22 anonymoose joined #git
11:22 vuoto joined #git
11:24 Loomaanaatii_ joined #git
11:24 kurkale6ka joined #git
11:26 nowhere_man joined #git
11:27 i7c skunkz, maybe they shouldn’t work all on the same branch and just merge complete feature branches or so
11:27 finalbeta joined #git
11:27 i7c Or yes, rebase, but I think whether you have a merge or rebase workflow mostly depends on how your team works.
11:28 wachpwnski-mobi joined #git
11:28 dsdeiz joined #git
11:28 dsdeiz joined #git
11:28 wachpwnski-mobi Does anyone know how I can untrack a file on git annex?
11:28 manuelschneid3r joined #git
11:28 Mathiasdm joined #git
11:29 i7c We also have a rebase-based workflow here due to how our integration works, however, sometimes I have the feeling some people have a rebase workflow out of sheer fear of merge commits ^^
11:29 Ratler joined #git
11:29 Loomaanaatii_ joined #git
11:29 wachpwnski-mobi I'm on a bare repository and I keep getting this: git annex unannex SHA256E-s329400--62e8f76db705f88d366c653c130b3e690765de4c3367ecc1f67b88f1b2a887f8.par
11:29 wachpwnski-mobi git-annex: Cannot proceed with uncommitted changes staged in the index. Recommend you: git commit
11:29 schleppel joined #git
11:29 gechr joined #git
11:29 cagedwisdom joined #git
11:29 Jonuz joined #git
11:30 timlyo[m] joined #git
11:30 bremner wachpwnski-mobi: can you make a local clone?  working directly on the annexed object you probably need to use plumbing commands
11:30 overlord_tm joined #git
11:30 Loomaanaatii joined #git
11:30 hiq[m] joined #git
11:30 Matsv joined #git
11:30 bremner wachpwnski-mobi: btw, git-annex has it's own channel on oftc
11:30 Xarodon joined #git
11:31 skunkz i7c: I agree, but unfortunately I work with people who don't give a f about git workflow and all work on the same branch, they don't even know how to look at the history, I'm the only one a bit worried about it lol
11:32 i7c skunkz, :( I feel your pain. I have worked with such people before.
11:33 wachpwnski-mobi bremner: I'm not sure I know what you mean?
11:34 bremner wachpwnski-mobi: it looks like you tried to add something from ./git/annex/objects
11:34 skunkz Only 2 more months and I'm free! Do you confirm using git rebase -p -i would have avoided getting the conflicts again, given that the commits are applied in the same order as they were with the merge?
11:36 wachpwnski-mobi bremner:  I am trying to remove missing files, it spams errors on my repo when I try to git annex get
11:36 wachpwnski-mobi It seems like this worked `git annex dropkey SHA256E-s329400--62e8f76db705f88d366c653c130b3e690765de4c3367ecc1f67b88f1b2a887f8.par `
11:37 bremner yes, that's the kind of plumbing command I meant. So you did understand ;)
11:38 wachpwnski-mobi Hah, I have so much to learn.
11:38 xintron left #git
11:38 bremner yes, unfortunately git-annex is a whole layer of stuff on top of git
11:39 ArchNoob joined #git
11:40 ArchNoob Hi, is base the 80c819a characters you see when you say something like `git log`?
11:40 osse "base" ?
11:41 ArchNoob yeah, "BASE"
11:41 fakenerd joined #git
11:41 osse where do you see that prase?
11:42 osse the only BASE I know of is when you resolve merge conflicts, and that refers to something else
11:42 ArchNoob Sorry, I'm making a pull-request using hub command. I see BASE and HEAD I wonder which one should I use to point to the particular recent commit I just made.
11:42 osse so I guess the answer to your question is no
11:43 osse HEAD is the commit you just made
11:43 osse BASE is the branch you want them to merge in to
11:43 Gloomy joined #git
11:45 NeverDie_ joined #git
11:46 jimi_ joined #git
11:47 ArchNoob ooh, Thank you. I understand now.
11:48 _ikke_ Those characters you refer to are the commit hash
11:52 steve joined #git
11:52 dvaske joined #git
11:52 NeverDie joined #git
11:54 star_prone joined #git
11:59 gnixev joined #git
12:00 watersoul joined #git
12:00 MarioBranco joined #git
12:01 dvaske_ joined #git
12:02 darkbit joined #git
12:03 lordjancso joined #git
12:03 MarioBranco joined #git
12:03 darkbit joined #git
12:04 notebox joined #git
12:05 watersoul joined #git
12:05 gregor2 joined #git
12:05 rivarun joined #git
12:07 fakenerd joined #git
12:09 yarekt joined #git
12:10 watersoul joined #git
12:10 yarekt left #git
12:12 ygg joined #git
12:12 jonascj_ Any one familiar with github and pull requests? I've forked a repo, made a feature branch fb on my working copy of the original repo, pushed that branch to my own repo (the github fork), and created a pull request (for the original author to implement my changes).
12:13 jonascj_ Now however, I want to continue to work on other improvements, and I don't know if I should create a new branch for working on other changes. And if I should create a new branch, what about the improvements from my first feature branch? How do I take those with me to feature branch 2?
12:14 i7c Just branch off the other feature branch?
12:14 i7c off of*
12:14 mentazoom joined #git
12:15 watersoul joined #git
12:15 i7c Also this is potentially problematic if the upstream repo refuses your pull request or makes major changes, you will have to deal with the quirks of the rebase, but that being said, you can do that.
12:15 jimi_sanchez joined #git
12:16 jonascj_ i7c: if upstream refuses my requests I'll just maintain my own fork :P
12:16 i7c haha
12:17 jonascj_ i7c: do I have other alternatives, merge my feature branch into master and branch again?
12:17 Cabanossi joined #git
12:17 jonascj_ but right, the easiest for me to understand would be if upstream just merge my changes, and then I can do the same thing over again
12:18 i7c It effectively makes no difference if you merge in your repo or not. Would be a fast-forward merge anyways.
12:18 owlbynight joined #git
12:19 i7c jonascj_, I would just work on top of the first branch, and if upstream merges your branch or changes it and merges or does any other change that affects your work, I would just rebase.
12:19 jonascj_ i7c: fast forward since nothing has happened on master while I've been working on the feature branch?
12:20 i7c exactly
12:20 watersoul joined #git
12:21 i7c jonascj_, also it’s interesting if upstream proceeds work on master and there are conflicts, if they want you to resolve them (i.e. rebase) or if they resolve them when merging.
12:22 i7c But that’s a separate issue :D
12:24 alexggordon joined #git
12:25 watersoul joined #git
12:29 courrier_ joined #git
12:29 jonascj_ i7c: alright, thanks
12:29 ash_workz joined #git
12:29 nowhere_man joined #git
12:30 fahadash joined #git
12:31 zeroed joined #git
12:31 zeroed joined #git
12:31 star_prone joined #git
12:32 d10n-work joined #git
12:33 i7c You’re welcome :)
12:34 jimi_ joined #git
12:35 darkbit joined #git
12:36 seni joined #git
12:36 darkbit joined #git
12:38 darkbit joined #git
12:39 dglambert joined #git
12:39 dglambert yo
12:39 dsdeiz joined #git
12:40 jimi_sanchez joined #git
12:41 Gloomy joined #git
12:43 brk_ joined #git
12:43 livingstn joined #git
12:44 loskutak joined #git
12:45 star_prone joined #git
12:46 jeffreylevesque joined #git
12:47 slidercrank hello. I want to write a gerrit server hook. it´s invoked as ref-updated --oldrev <old rev> --newrev <new rev> --refname <ref name> --project <project name> --submitter <submitter> . how go I get a list of files knowing newrev <sha1>?  In pre-commit I could invoke ¨git diff --cached --name-only¨ for that
12:47 osse slidercrank: git diff --name-only old-rev new-rev
12:47 slidercrank thank you, osse
12:52 mujow joined #git
12:55 ash_workz joined #git
12:55 peacememories joined #git
12:57 nanga joined #git
12:59 schleppel joined #git
13:01 fakenerd joined #git
13:02 ExoUNX joined #git
13:03 vkareh joined #git
13:04 acetakwas joined #git
13:05 marcogmonteiro joined #git
13:05 TheSimonator joined #git
13:10 zxd hi
13:10 zxd what do they mean by 'If you commit at this point then you will lock the submodule into having the new code when other people update.' in  https://git-scm.com/book/en/v2/Git-Tools-Submodules
13:12 jameser joined #git
13:12 monq joined #git
13:12 ratchet joined #git
13:15 synthroid joined #git
13:16 sunri5e joined #git
13:17 Guest37119 The resently added worktree feature help me with my workflow wanted to say thanks for that to whomever is responsible.
13:19 ruxu joined #git
13:19 vrach joined #git
13:20 Gloomy joined #git
13:21 eric____ joined #git
13:22 eric____ hi there.. simple question here... Git keeps asking for my username and password when I push to a branch.. how can I make it so that it will only ask upon reboot of my computer?
13:23 tobiasvl eric____: are you pushing over ssh or https?
13:23 cdg joined #git
13:24 jimmyrcom joined #git
13:24 fission6 joined #git
13:25 fakenerd joined #git
13:25 eric____ tobiasvl : im not sure, i'll try to find out and get back to you.
13:26 tobiasvl OK, should be fairly easy to find out
13:26 eric____ my push URL is a https beanstalkapp.com link
13:26 tobiasvl eric____: then !credential-cache
13:26 gitinfo eric____: Tired of entering your username and password every time you access an HTTP(S) remote and have good reasons not to use SSH? A summary of shortcuts and supported caching methods can be found here: http://stackoverflow.com/a/5343146
13:26 al-damiri joined #git
13:27 marianina8 joined #git
13:28 eric____ tobiasvl : where would i put this in
13:28 eric____ tobiasvl : sorry for my noobness loll
13:28 NeXTSUN joined #git
13:28 tobiasvl eric____: the answer should explain everything, the text with gray background are commands you can run in your shell
13:29 eric____ ok
13:30 tobiasvl man gitcredentials
13:30 gitinfo the gitcredentials manpage is available at https://gitirc.eu/gitcredentials.html
13:30 cqi joined #git
13:32 eric____ https://you:password@github.com/you/example.git
13:32 eric____ this worked perfectly :)
13:32 dave0x6d joined #git
13:33 theoceaniscool joined #git
13:33 ig0r_ joined #git
13:33 tobiasvl yes, that eliminates the prompt
13:34 tobiasvl but saves the password in cleartext in bash history and displays it in the terminal
13:36 NeXTSUN joined #git
13:36 eric____ ok... not best then
13:38 jameser joined #git
13:43 Es0teric joined #git
13:45 Gsham joined #git
13:45 thebinary joined #git
13:46 mentazoom joined #git
13:49 masuberu joined #git
13:49 Dougie187 joined #git
13:50 lewix joined #git
13:51 DelphiWorld joined #git
13:51 DelphiWorld hi giters
13:51 DelphiWorld i have a git data folder backed up from gitlab
13:51 DelphiWorld how can i extract my git repos from it?
13:53 ArchNoob joined #git
13:53 justin3 joined #git
13:54 osse DelphiWorld: you mean a .git dir ?
13:54 DelphiWorld yes, repos.git
13:54 osse you can clone it
13:54 livingstn joined #git
13:54 DelphiWorld from local?
13:54 publio joined #git
13:54 osse yes
13:54 cbreak git clone can clone from any repository
13:55 DelphiWorld haha
13:55 DelphiWorld awesome
13:55 cbreak if it's local, it might only use hardlinks
13:55 cbreak if you want distinct files, you can use file:// urls
13:55 alexggordon joined #git
13:55 cbreak or specify --no-hardlinks
13:55 Sgpf joined #git
13:55 cbreak other than that it's pretty much a normal clone
13:56 DelphiWorld HAHAHA.... awesooooooome
13:56 DelphiWorld thanks guys
13:56 oleo joined #git
13:57 oaaao joined #git
13:58 acetakwas joined #git
13:58 DelphiWorld left #git
14:01 eroux joined #git
14:01 notebox joined #git
14:02 Es0teric joined #git
14:03 xall joined #git
14:03 theoceaniscool joined #git
14:07 schleppel joined #git
14:07 zxd fatal: reference is not a tree: 7a008cfbecc70c71a586ca6f7b34b01ce7e9fadd
14:07 zxd Unable to checkout '7a008cfbecc70c71a586ca6f7b34b01ce7e9fadd' in submodule path 'foo/irssi-smartfilter'
14:09 maryo joined #git
14:10 telephone joined #git
14:10 telephone joined #git
14:12 moei joined #git
14:13 guardian hello, how can I update the index to add/synchronize a submodule to a specific SHA1 without cding into the submodule path and checking out the SHA1 in question
14:16 Emperor_Earth joined #git
14:17 AaronMT joined #git
14:17 Cabanossi joined #git
14:19 Vampire0 guardian, do you just want to save the cd, or also the checkout?
14:20 lewix I have an untracked file (not folder) that being a pain in the butt, git clean -f won't remove it
14:20 Vampire0 guardian, `git -C submodule-path checkout <commit-ish>` would checkout in the submodule without the need of cding into it
14:21 Vampire0 lewix, is it an ignored file?
14:21 lewix nope
14:21 Vampire0 lewix, hm, what happens if you do `git clean -f` and / or what is the output of `git clean -n`?
14:21 lewix it's a test file
14:22 Vampire0 lewix, and do you execute this from the directory (or a parent directory) where the file resides, or from a sibling directory?
14:22 lewix nothing
14:23 lewix from a long lost cousin directory
14:23 Vampire0 lewix, what means lost? where exactly is the file and where exactly do you call the clean?
14:24 tymczenko joined #git
14:24 lewix i see it's working. Now, it means that the directory i was executing it from was a sibling of a sibling of a sibling
14:24 DarkPsydeLord joined #git
14:24 lewix i went to the right directory and it worked
14:24 kpease joined #git
14:25 TheSimonator joined #git
14:25 lewix Vampire0: thank you
14:26 Vampire0 lewix, clean only works on the current directory and its descendents. If you want to catch all untracked files, you need to execute it in the root directory or give the root directory as argument to the clean like `git clean -n $(git rev-parse --show-cdup)`
14:27 menip joined #git
14:28 basiclaser joined #git
14:28 guardian I don't want to commit in the submodule
14:28 lewix Vampire0: thanks
14:29 Nilesh_ joined #git
14:29 guardian I want to update the synchronization SHA1 between parent and submodule
14:30 xall joined #git
14:30 Es0teric joined #git
14:30 star_prone joined #git
14:31 aw1 joined #git
14:37 Vampire0 guardian, just that you checkout does not mean you commit. You just checkout the commit you want to reference in the submodule, then update the reference in the parent and commit in the parent. There will not be any commit in the submodule and you need to checkout in the submodule anyway to have the correct state for the submodule
14:37 fission6 joined #git
14:39 guardian rephrasing: I would like to know whether it's possible to use git update-index since I know the SHA1, instead of doing the dance: cd path/to/submodule ; git checkout SHA1; cd -; git add path/to/submodule; git commit -m 'updated submodule'
14:39 bilal_ joined #git
14:41 a3Dman joined #git
14:41 a3Dman joined #git
14:44 NeXTSUN joined #git
14:44 Vampire0 guardian, to have what as result? You only want to do update-index and the intended result is?
14:47 Es0teric joined #git
14:47 madewokherd joined #git
14:49 Dougie187 left #git
14:49 Vampire0 guardian, I think you should be able to use update-index with --cacheinfo, but I'm not sure that results in what you intend. You will save the checkout, but then the submodules worktree is not in the state you reference, so it will be shown as changed, as it is not in the state expected
14:49 Dougie187 joined #git
14:50 ki0 joined #git
14:50 Es0teric joined #git
14:51 boombatower joined #git
14:52 ResidentBiscuit joined #git
14:52 mozzarella joined #git
14:53 ResidentBiscuit joined #git
14:53 p4trix joined #git
14:54 netj joined #git
14:54 onehrxn_ joined #git
14:56 ZyX-I joined #git
14:56 CheckDavid joined #git
14:57 ozcanesen joined #git
14:57 revoltingPeasant joined #git
14:57 Macaveli joined #git
14:59 revoltingPeasant I'm tracking an svn repo locally using git. here is a copy of it's .git/config https://ptpb.pw/cJDn
14:59 revoltingPeasant I want to push this to a new git repo
15:00 oaaao joined #git
15:00 Es0teric joined #git
15:00 revoltingPeasant I also want my changes to be written to this git repo when I push any further commits
15:00 oaaao joined #git
15:00 thebope joined #git
15:01 revoltingPeasant I'm not sure how to do this and google isn't giving me anything relevant. does anyone know where I can read about this?
15:03 Es0teric joined #git
15:04 grawity revoltingPeasant: literally regular `git remote add ...` and `git push`
15:08 Silenced joined #git
15:09 fission6_ joined #git
15:09 telephone joined #git
15:09 telephone joined #git
15:11 Tobbi joined #git
15:11 weakdan left #git
15:12 ayogi joined #git
15:12 cousteau joined #git
15:13 cousteau sup
15:13 cousteau Does git create one whole copy of all the project for each branch?
15:13 kristofferR joined #git
15:13 Vampire0 cousteau, no
15:13 tsdev joined #git
15:13 Murii joined #git
15:13 Vampire0 cousteau, a branch in Git costs exactly 41 bytes
15:13 cousteau ah
15:14 cousteau I was thinking that creating the whole copy would simplify the process of checking out another branch, and...
15:14 cousteau ok so a branch is just a tag you put on a commit
15:14 MissionCritical joined #git
15:14 Vampire0 cousteau, a branch in Git is just like a post-it note sticked to a commit, the same for tags. The only difference is, that the branch post-it is automatically peeled off and sticked onto the new commit if you create a new commit while the branch is checked out (the branch, not the commit, the branch is pointing to)
15:15 Vampire0 cousteau, if you switch branches, simply the files that are different between the branches are updated to the new branches state
15:15 maryo joined #git
15:15 Vampire0 cousteau, you can also checkout multiple branches in different directories with man git worktree
15:15 gitinfo cousteau: the git-worktree manpage is available at https://gitirc.eu/git-worktree.html
15:15 cousteau I see
15:15 overlord_tm joined #git
15:16 theiostream joined #git
15:16 LeBlaaanc joined #git
15:17 cousteau ok in that case I think I'll ...yeah I'll use a tag, let me look up what's that
15:17 Vampire0 cousteau, it is the exact same thing as a branch technically, with the exception that it never moves automatically
15:17 cousteau I just got my project reviewed so I want to tag this for future reference
15:17 Cabanossi joined #git
15:17 Vampire0 cousteau, yes, then you are probably after a tag
15:18 Silenced joined #git
15:18 borkr joined #git
15:19 quizzi joined #git
15:20 thiago joined #git
15:20 shinnya joined #git
15:21 robattila256 joined #git
15:22 Es0teric joined #git
15:22 netj joined #git
15:24 eroux joined #git
15:25 dedicated joined #git
15:26 alexggordon joined #git
15:26 Achylles joined #git
15:27 tyreld joined #git
15:27 tymczenko joined #git
15:27 Es0teric joined #git
15:31 theoceaniscool joined #git
15:31 ArchNoob joined #git
15:32 theiostream joined #git
15:32 samlamamma joined #git
15:34 moony22 joined #git
15:35 vkareh joined #git
15:35 theiostream joined #git
15:36 a3Dman joined #git
15:40 moony22 Hi, I know this is probably a trivial question but I'm bad at git and don't want to completely mess everything up. I had a remote master which had many commits, and I made changes to a local master which was many commits behind. Then I tried to git pull and it merged instead of rebased, making this in my github network graph (I pushed like an idiot): http://i.imgur.com/UPX1S4F.png - is there a way to revert
15:40 moony22 this
15:40 gregor2 joined #git
15:40 xall_ joined #git
15:41 Es0teric joined #git
15:41 moony22 or is it too late and I can't change it
15:41 _ikke_ moony22: Do you expect anyone to have pulled from master since you pushed?
15:41 moony22 only myself
15:42 moony22 from another pc
15:42 Gustavo6046 joined #git
15:42 moony22 other than that theres no changes or anything, the latest commit is the merge commit
15:42 matsaman joined #git
15:42 Hello71 joined #git
15:43 metachr0n joined #git
15:44 Macaveli joined #git
15:44 _ikke_ then you can still undo it
15:45 vkareh joined #git
15:45 moony22 oh great how can I do it? I don't need the changes I made on the first PC before pulling since I can just copy that file
15:45 osse moony22: use the --force
15:45 tohuw joined #git
15:47 moony22 sorry, I know these tips might be good for someone that properly knows git commands but honestly i've got no idea how to start and where to start from
15:48 Es0teric joined #git
15:48 moony22 i guess I should use git reset
15:48 kadoban joined #git
15:48 rominronin joined #git
15:49 acetakwas joined #git
15:49 rorro joined #git
15:50 moony22 or git revert idk
15:50 Derperperd joined #git
15:55 moony22 crap I reverted the merge but it didnt fix it
15:55 TbobbyZ joined #git
15:55 Es0teric joined #git
15:56 Guest29517 joined #git
15:56 moony22 I guess it's not possible
15:57 nexemacs joined #git
15:57 moony22 is there any way to rewrite history instead?
15:57 GT4066 joined #git
15:57 LeBlaaanc joined #git
15:58 synthroid joined #git
16:01 jagob joined #git
16:01 _ikke_ yes
16:01 hobodave joined #git
16:01 _ikke_ With git reset for example you can undo the revert and the merge
16:01 _ikke_ !fixup
16:01 gitinfo So you lost or broke something or need to otherwise find, fix, or delete commits? Look at http://sethrobertson.github.com/GitFixUm/ for full instructions, or !fixup_hints for the tl;dr. Warning: changing old commits will require you to !rewrite published history!
16:02 peacememories joined #git
16:02 moony22 !rewrite
16:02 gitinfo Rewriting public history is not recommended. Everyone who has pulled the old history will have to do work (and you'll have to tell them to), so it's infinitely better to just move on. If you must, you can use `git push --force-with-lease <remote> <branch>` to force (and the remote may reject that, anyway). See http://goo.gl/waqum
16:02 ki0_ joined #git
16:02 brent__ joined #git
16:03 lewix joined #git
16:03 moony22 can I use git push -f origin #commit:master
16:03 Es0teric joined #git
16:04 _ikke_ Yes, but you also want to fix it locally
16:04 nixjdm joined #git
16:04 moony22 ok so I can use git reset --hard and the commit id of the right one as well im guessing
16:04 TheSimonator joined #git
16:07 saul left #git
16:07 jstimm joined #git
16:08 moony22 it worked great, thanks a lot!
16:10 Es0teric joined #git
16:11 star_prone joined #git
16:13 MineCoins joined #git
16:14 zeroed joined #git
16:14 zeroed joined #git
16:15 HoierM joined #git
16:16 tcsc i'd like to close all the branches i have in my local repo that start with a prefix ("bug/") and have been merged into another branch, which is called inbound because reasons.
16:17 tcsc historically i've done this manually. e.g. checkout each branch, rebase against inbound, then see if branch -d works. this is tedious and i've slacked on it and have about 30 branches that i'd have to do this to, and would rather not do it manually, or at least would be interested in any slightly better way of the manual approach i just described
16:17 dvaske joined #git
16:18 Cabanossi joined #git
16:19 jordila joined #git
16:20 Es0teric joined #git
16:20 _ikke_ tsdev: git branch --merged tells you what branches are reachable from a specific branch
16:20 jordila hi... newbie here, i just want 'develop' becomes 'master' branch ( no merge here ) , which is the easiest way to accomplish it ?
16:21 tcsc _ikke_: hm, and i guess i could -D those with confidence
16:21 durham joined #git
16:21 * jordila ... is wondering... ^^ would a pure branches renaming do the job ?
16:22 orbyt_ joined #git
16:22 _ikke_ jordila: sure, that's a good way to do it
16:22 theoceaniscool joined #git
16:23 revoltingPeasant grawity: Thanks
16:23 star_prone joined #git
16:24 Es0teric joined #git
16:24 jordila _ikke_ i've read here that git doesn't easily allow the deletion of (afterwards needed) 'master' branch  :
16:25 * jordila https://stackoverflow.com/questions/1526794/rename-master-branch-for-both-local-and-remote-git-repositories ... see 'Note:'
16:25 portal_ joined #git
16:25 ambro718 joined #git
16:26 dendazen joined #git
16:27 tcsc _ikke_: ah, so it appears that something (possibly the fact that they're not yet merged in my local tree, due to the patch submission/code review process we use, or possibly the fact that this is really a mercurial tree i'm interacting with via git-cinnabar) is making it so that git branch --merged doesn't contain any of the branches i'm interested in
16:27 TikityTik joined #git
16:27 Es0teric joined #git
16:28 tcsc i know that the approach i outlined above does work, e.g. git co branch-in-question; git rebase inbound; git co inbound; git branch -d branch-in-question
16:28 tcsc if that ... helps
16:28 tcsc i guess i can just write a shell script or something to do that
16:28 InfoTest joined #git
16:29 tcsc or i could just delete the ones i haven't touched in several months outright and hope it works out...
16:32 Es0teric joined #git
16:33 __main__ joined #git
16:36 nexemacs joined #git
16:37 livingstn joined #git
16:37 Introoter joined #git
16:38 nexemacs joined #git
16:38 TheSimonator joined #git
16:39 Darren_ joined #git
16:39 tcsc yep, hope i don't regret doing that
16:40 kfoonamalik joined #git
16:40 ankit01ojha joined #git
16:41 shgysk8zer0 joined #git
16:41 ki0 joined #git
16:42 dreiss joined #git
16:43 Es0teric joined #git
16:45 kfoonamalik joined #git
16:45 peacememories joined #git
16:46 dfas joined #git
16:47 maryo joined #git
16:49 rominronin joined #git
16:50 jstimm joined #git
16:50 theiostream joined #git
16:52 diogenese joined #git
16:52 truckcrash joined #git
16:52 truckcrash anyone know where I could go to get help for gitbash for windows?
16:53 mikecmpbll joined #git
16:54 Murii joined #git
16:56 alexggordon joined #git
16:57 freimatz joined #git
16:57 truckcrash huh. sure is quiet in here for a channel with so many users :)
16:58 truckcrash If anyone knows what is going on with this, or knows where I could get help, I'd like to know
16:58 svm_invictvs joined #git
16:58 truckcrash http://i.imgur.com/6Vb4g8M.png
16:58 truckcrash telnet is installed, and works under the windows command line
17:00 tarkus joined #git
17:02 synthroi_ joined #git
17:03 s1scha joined #git
17:03 tnecniv joined #git
17:03 makinen left #git
17:04 davimore joined #git
17:06 gopar joined #git
17:06 peacememories joined #git
17:08 nexemacs joined #git
17:09 DarkPsydeLord joined #git
17:12 TheSimonator joined #git
17:13 saml joined #git
17:13 saml i want to introduce revert commit(s).   and also want to create a branch out of sha1...sha2
17:13 saml so that I keep working on the new branch and be able to merge back to master
17:15 marianina8 joined #git
17:16 marianina8 joined #git
17:17 vamiry joined #git
17:17 MineCoins joined #git
17:19 Discovery joined #git
17:19 cousteau joined #git
17:19 oskarkv joined #git
17:19 saml A-B-C   want to do A-B-C-revertC-revertB    but also put B-C to a branch in a way that the branch can be merged back, bringing back B-C
17:22 cjohnson have you already pushed the current branch?
17:23 cjohnson pushed B and C that is
17:23 cr1tic joined #git
17:24 cjohnson if you haven't, you can just make a new branch from the current branch at C, and the new branch will be A B C
17:24 cjohnson then rewind the current branch to A and push that
17:24 Ryanar joined #git
17:26 cr1tic joined #git
17:27 ki0 joined #git
17:28 saml cjohnson, yeah things are already pushed. so i want to create a revert commit
17:28 saml looks like more things are pushed...  so I want to revert a merge commit. then create a branch in a way that I can work on it further and merge it back
17:31 cjohnson maybe you could make a separate branch where you just revert the revert commits
17:31 star_prone joined #git
17:32 cjohnson since all the branches will always contain B and C as well as their reverts, any merge won't bring B and C back over when you're ready later to integrate
17:32 ig0r_ joined #git
17:32 cjohnson you'll need to make B and C into entirely new commits
17:33 Cabanossi joined #git
17:33 pks joined #git
17:33 saml how do I turn B-C into new commits?
17:34 saml generate patch somehow and create a commit out of patch?
17:34 jtri joined #git
17:35 cjohnson revert revert
17:35 cjohnson if you revert revertC you should end up with a commit identical to C
17:35 cjohnson in theory
17:35 AaronMT joined #git
17:35 saml thanks
17:35 Vinnie_win joined #git
17:36 jtri hmm, got myself in a tangle, i did git stash, switched branches to master, did some merging, building, etc, then accidentally edited in master, so i just switched back to my branch and tried a git stash pop, my tree also looks a bit harry: http://dpaste.com/1XCS4X5
17:37 jtri wondering how to get back to working in my branch with the changes showing when i type git diff
17:39 phase2 joined #git
17:39 Acerific joined #git
17:39 Acerific joined #git
17:40 jtri that commit has e47f9aa8 is not my stashed changes
17:42 imack joined #git
17:43 dviola joined #git
17:45 GGMethos joined #git
17:46 Ryanar joined #git
17:46 Tobbi joined #git
17:48 gugah joined #git
17:49 livingstn joined #git
17:49 re1 joined #git
17:49 jordila left #git
17:52 doener joined #git
17:52 peterbecich joined #git
17:53 pbandark joined #git
17:55 lewix joined #git
17:57 ki0 joined #git
17:58 ezio joined #git
17:58 ezio is there a key manager for git bash ssh?
17:58 dfas joined #git
17:58 preaction is ssh-agent installed?
17:59 ezio ah thanks
17:59 a3Dman joined #git
17:59 ezio uhh
17:59 ezio yes
17:59 ezio thanks
17:59 safe joined #git
18:00 skered joined #git
18:00 kadoban jtri: Not really clear what trouble you're having. You figure it out yet?
18:00 gregor2 joined #git
18:01 jtri kadoban: yeah, i'm not sure how i got into it, but i had an unapplied stash item which i was able to recover
18:02 Silenced joined #git
18:03 livingstn joined #git
18:04 scoobert1on joined #git
18:04 chovy_ joined #git
18:04 cliluw joined #git
18:04 avar When I run e.g. `git -c <config> grep --recurse-submodules` the <config> is passed to the child processes spawned, but I don't see how that's happening.
18:05 avar via ps auxf the invoked child processes don't get their own -c <config>, wondering how this is passed along..
18:05 teroshan joined #git
18:06 paul424 joined #git
18:06 Fallen0223 joined #git
18:07 ProLoser joined #git
18:08 leeN joined #git
18:08 ProLoser I don't want to have to do `git push origin newbranch -u` every single time i checkout a new branch
18:08 ProLoser How can i get this to work byd efault?
18:08 ProLoser that it pushes to a branch of the same name unless explicitly configured to do otherwise
18:09 ProLoser or like when i create a new branch, it's automatically setup to track a respective branch on the remote of the same name (again, unless explicitly configured otherwise)
18:09 avar look up push-default in git-config(1)
18:10 jfr_ joined #git
18:10 vuoto joined #git
18:10 manoflags joined #git
18:11 TheSimonator joined #git
18:11 Sasazuka joined #git
18:11 cr1tic joined #git
18:14 jstimm joined #git
18:14 manoflags joined #git
18:15 rosiu joined #git
18:15 __main__ joined #git
18:15 lekare joined #git
18:15 nd_ joined #git
18:17 dmj` joined #git
18:17 bilal joined #git
18:18 ok91 joined #git
18:18 bronson joined #git
18:18 schleppel joined #git
18:19 avar Oh man, that config thing is because git itself passes along GIT_CONFIG_PARAMETERS
18:21 rominronin joined #git
18:24 ki0 joined #git
18:25 Silenced joined #git
18:25 Dan0maN_web joined #git
18:26 alexggordon joined #git
18:28 TheSimonator joined #git
18:28 Dan0maN_web hi all.  i'm trying to make a patch for an RPM.  RPMs tend to use the publicly released code from git repos at the time of release.  what i'm trying to get a patch of is a fix from where someone forked down the road from that release commit, fixed something, then PR'd it in
18:29 Dan0maN_web is there a way to do this easily?
18:29 hobodave joined #git
18:32 Es0teric joined #git
18:33 SilverBe joined #git
18:33 govg joined #git
18:36 TbobbyZ_ joined #git
18:39 vkareh joined #git
18:39 cbreak Dan0maN_web: a pull request is already a commit
18:39 a3Dman joined #git
18:39 cbreak you can easily get patches from existing git commits with git format-patch
18:41 Mutter joined #git
18:42 nexemacs joined #git
18:45 manuelschneid3r joined #git
18:46 dvaske joined #git
18:47 hobodave joined #git
18:47 TbobbyZ joined #git
18:49 vkareh left #git
18:50 maryo joined #git
18:51 Macaveli joined #git
18:55 theiostream joined #git
18:58 justin3 joined #git
18:58 brent__ joined #git
18:58 zivester joined #git
18:59 Ryanar joined #git
18:59 kpease joined #git
19:01 dvaske_ joined #git
19:02 Cabanossi joined #git
19:04 Dan0maN_web left #git
19:04 a3Dman joined #git
19:04 notebox joined #git
19:05 chovy_ joined #git
19:06 pbandark1 joined #git
19:06 jimi_ joined #git
19:10 synthroid joined #git
19:12 ozcanesen joined #git
19:14 suck joined #git
19:15 Oatmeal joined #git
19:15 blow_ joined #git
19:16 blow_ joined #git
19:18 peacememories joined #git
19:21 hsiktas joined #git
19:22 __main__ joined #git
19:23 p4trix joined #git
19:24 fission6 joined #git
19:24 p4trix joined #git
19:26 cebor joined #git
19:26 Achylles joined #git
19:26 jstimm joined #git
19:29 Matt``` joined #git
19:31 a3Dman joined #git
19:32 star_prone joined #git
19:32 Matt``` if I make a change in branch A, in branch B if I try to update from master it complains about conflicts so I have to manually duplicate what I did in branch A
19:32 Cabanossi joined #git
19:32 Matt``` even though nothing changed in branch B
19:33 NeverDie_ joined #git
19:34 Peetz0r joined #git
19:36 ali1234 joined #git
19:38 ArchNoob joined #git
19:40 druonysus_ joined #git
19:41 Cavallari joined #git
19:43 davimore joined #git
19:47 DigitallyBorn left #git
19:48 nexemacs joined #git
19:48 overlord_tm joined #git
19:52 rorro joined #git
19:52 rominronin joined #git
19:53 peterbec` joined #git
19:56 Primer joined #git
19:58 Primer Hi. I'm trying to understand how IDEs work with git. For example, it seems that some IDEs associate a directory with a single "project", and since switching branches in git doesn't change the directory, a "project" and concepts associated with it, such as a list of files you've worked on in the branch, don't carry over very well.
19:59 Primer I'm finding that I can't maintain a list of files on a per-branch basis in my IDE, phpstorm
19:59 Primer I'm wondering if I'm just missing something, or if this is in fact some shortcoming of the IDE
20:00 kadoban Hm? Typically each project is a git repository. I don't follow why you'd want per-branch lists of files? For the most part branches share the vast majority of their files.
20:00 _ikke_ Well a branch in git does not represent a work context
20:00 Primer Say what?
20:00 Primer How does a branch in git NOT represent a work context?!
20:00 gregor2 joined #git
20:00 _ikke_ a branch represents comitted content
20:01 Primer Ok...
20:01 _ikke_ Uncomited changes stay behind if you switch branches for example
20:02 Primer So you're saying this is a workflow issue?
20:02 _ikke_ Primer: Well, some IDE's do have features to manage work contexts
20:03 kadoban I personally don't understand the goal ...
20:03 _ikke_ kadoban: branch A, files a, b and c are open. Branch B, files d, e and f are open
20:04 Primer Because it's not uncommon for me to have more than one branch. For example: I start work on a feature. That feature opens 10 files. I then get a bug report. I make a branch of the WIP, commit, switch back to master, create a bug branch, open 10 other files...
20:04 preaction does the ide support git-work-tree?
20:05 Primer Fix the bug, commit, merge to master, switch back to the feature branch...and now I have to reset and open the dirty files
20:05 kadoban _ikke_: Oh is that what they're looking for? Never would have guessed
20:06 _ikke_ "list of files you've worked on in the branch, don't carry over very well."
20:06 * Primer searches for git-work-tree
20:06 gareppa joined #git
20:06 Primer Well, that's doable via git diff master --name-only
20:06 _ikke_ well, git-work-tree also requires IDE specific files to cary over
20:06 Primer presuming the branch is rebased often
20:07 medeirosthiiago joined #git
20:07 medeirosthiiago joined #git
20:07 Primer http://stackoverflow.com/questions/36438333/how-do-i-use-git-worktrees-in-intellij-idea-2016-1
20:07 GT4066 joined #git
20:07 _ikke_ Primer: intellij idea does have contexts
20:07 Primer Honestly, I'm looking for concepts that I feel I've probably just haven't heard of, such as worktree
20:08 Primer phpstorm == intellij
20:08 _ikke_ they are just not tied to branches
20:08 _ikke_ right, same for phpstorm
20:09 Primer So back to workflow then...what is YOUR workflow, and how does it differ from using branches for features/bugs?
20:09 crankslider joined #git
20:09 _ikke_ I do use brances for reatures/bugs, but I do not expect my editor to open all files that have changed in a branch
20:10 Primer I don't expect mine to open them either, but I do want it to track them, preferably in a view that shows all the files I've modified for the branch
20:10 Primer I just want a convenient view of the files touched by the branch. And if I open an untouched file and modify it, that it gets added to this view.
20:10 aielima joined #git
20:11 rorro joined #git
20:11 Primer I have a co-worker that uses...Kate, I believe (despite the fact that has has a license for phpstorm) simply because it supports this concept, and he says he's unwilling to change IDEs because of this
20:11 _ikke_ Right, that's a feature request you'd have to make then
20:12 _ikke_ I would just open a pane with a terminal that shows git diff --name-only output
20:12 MineCoins joined #git
20:12 hahuang65 joined #git
20:12 Primer Yes, that is what I myself have been doing
20:12 _ikke_ (I use tmux with panes)
20:12 Primer But within the context of the IDE
20:12 Primer I can open an IDE view of git diff master --name-only (note master there)
20:12 MineCoins joined #git
20:13 Primer But that view, by default, is a diff viewer, so it's a 2 step process to actually open the file
20:13 Primer I realize this is a shortcoming of the IDE
20:13 MineCoins joined #git
20:13 _ikke_ You can write a plugin for it
20:14 Primer But again, I'm just trying to figure out if 1) I'm doing something incorrectly, 2) I'm just not aware of concepts that make this possible, 3) there's a better workflow that is more conducive to what I need
20:14 MineCoins joined #git
20:14 kadoban I typically just manually open the files I need, though I could see where that'd sometimes be annoying.
20:14 _ikke_ I manuallo do the diff when I need it
20:14 kadoban If I had to, I'd probably just do a quick little command-line script to have it open the files outputted by that 'git diff' I guess
20:15 Primer _ikke_: and by that I'm guessing you don't mean in your IDE, but in a terminal?
20:15 kadoban If I really wanted to get fancy I guess I'd add it to vim as a thing ... though I tend to get creative with my branch names so I couldn't fix "master" as the comparison
20:15 _ikke_ correct, but I do everything in the terminal anyway
20:15 _ikke_ I used to use intellij, but switched to vim completely
20:16 Lunatrius joined #git
20:16 Primer kadoban: I'm just using master as an example here. We also do feature branches and branch off of those.
20:16 kadoban Yeah, it'd be hard to handle that automatically then, because I'm not sure how you'd programatically come up with "the branch I based this on" that easily
20:17 Primer kadoban: having that be a one-time configuration thing should be simple enough
20:17 Primer but that'd have to be tied to the git metadata, I would think
20:17 Primer surely easily done with git config?
20:17 Primer Anyhow
20:18 kadoban Probably could go in the config I suppose, ya.
20:18 Sazhen86 joined #git
20:18 Primer I'm going to read up on worktree
20:18 Primer I had never heard of this until now
20:18 Primer thanks
20:18 _ikke_ I don't think IDEs support work-trees that well
20:19 _ikke_ You'd have to open the project multiple times
20:19 Macaveli joined #git
20:19 kadoban All of the IDE support for git I've seen was very limited and kind of annoying to me. But I never looked that hard either I suppose.
20:19 Primer My co-worker tells me this Kate feature is in fact a plugin
20:20 _ikke_ The only thing I used IDE git integration for is the revert change per hunk option
20:20 _ikke_ and see what lines were changed
20:20 Primer Honestly, I do all my git operations from the terminal too, but it's nice seeing a graphical side-by-side diff with a list of files
20:21 _ikke_ but an actualy diff would be much more useful to me than just the filenames
20:21 Primer git ops such as pull/push/rebase are done in the terminal, but I prefer to do diffs in the IDE
20:21 Primer and by that I mean actual diffs, not just --name-only
20:22 hobodave joined #git
20:23 _ikke_ but if your coworker wants to use kate, let them use kate
20:24 Primer I also have a co-worker that uses nano
20:24 preaction so?
20:24 Primer over ssh
20:24 Primer An IDE has the ability to quickly show, for example, where a method is used, throughout an entire project, using a search index.
20:25 Es0teric joined #git
20:25 preaction so?
20:25 Primer I'm one single key stroke away from finding all usages of a method, for example
20:25 preaction congratulations?
20:25 _ikke_ ctags can do that too
20:25 Primer My nano-using co-worker has to drop to a shell and do a recursive grep of the project
20:25 brigadier7 joined #git
20:25 preaction i mean, i use vim, so i am too
20:25 preaction so?
20:26 Primer Ok, say the method name is "get"
20:26 preaction are you trying to impose your IDE on these people? because how do you intend to approach that conversation without insulting them in some way?
20:26 Primer Good luck filtering out all the false hits from grep?
20:26 preaction condescension. that'll get them on your side, yep.
20:27 Primer So far I've been pretty pragmatic about what I've written here. If you feel insulted, that's on you.
20:27 Primer At no time did I write anything insulting.
20:27 preaction i'm not the one who you're going to insult. it's your cow-orkers, who you have to ork with
20:28 preaction "stop using nano, use my thing instead because it's better" "how is it better" "i can do these things fast" "i don't care about doing those things fast" </scene>
20:29 Primer One time I observed a co-worker spend an hour opening hundreds of files manually and changing text in them manually.
20:29 QwertyDragon joined #git
20:30 Primer I then showed this person how I would have done it with sed, in about 30 seconds.
20:30 osse inb4 "so?"
20:30 cjohnson lol
20:30 cagomez joined #git
20:31 Primer I feel that my nano-using co-worker is wasting a lot of time with his choice to continue doing only what he knows.
20:31 preaction that's an entirely different situation. you're asking someone to take on the arduous task of re-learning their entire development workflow just so they can do one rare thing they may or may not care about "fast"
20:31 chardan joined #git
20:32 darkbit joined #git
20:32 brigadier7 I'm fairly new to git, and I made a branch called freq from master.  Someome made changes to master, and I merged them into my branch on 4/13.  I want to undo that.  What is the best way?
20:32 Primer This same co-worker had to switch from svn to git.
20:32 preaction and again i ask you: How do you intend on telling your co-worker that without insulting them?
20:32 Primer Was my insistence of this equivalent?
20:32 andrew9184 joined #git
20:32 Primer He had a really hard time with it
20:32 preaction in the context of this conversation, yes
20:32 _ikke_ brigadier7: Did you made commitsa after that?
20:33 brigadier7 yes :(
20:33 osse if a person is insulted by another person offering advice then let them be insulted
20:33 cjohnson Primer it sounds like all you need to do is configure your editor to detect workfing file changes and either reload or warn/close the tab for files that disappear
20:33 cr1tic joined #git
20:33 cjohnson Me personally I much prefer to have to manually do it, in case I screw something up
20:33 _ikke_ brigadier7: Did you push your branch?
20:33 cjohnson so I would hate to have your preference of wanting your editor to automate it dictated to me via git
20:33 brigadier7 _ikke: yes :( :(
20:33 Primer cjohnson: agreed. I was trying to figure out if this is an uncommon thing with IDEs. I only have the context of intellij so far.
20:34 cjohnson Most IDEs default to reloading as files change afaik
20:34 davimore joined #git
20:34 cjohnson I don't think vim does though
20:34 _ikke_ brigadier7: Do you expect someone else to have fetched this branch ?
20:34 _ikke_ cjohnson: vim can do it
20:34 cjohnson sorry I just mean, by default
20:34 preaction yes, but not by default, as was said
20:34 kadoban If you have the files open in vim and they change with like git, vim will ask if you want to reload them. If that's what you mean.
20:35 brigadier7 _ikke_: No, it's only me, and there have been no changes on my branch since the merge
20:35 cjohnson yeah that's my preference
20:35 Primer preaction: anyhow, I've had to approach this switch to the IDE diplomatically. I currently have 3 developers that don't want to get on board, but only because they're set in their ways, and are unwilling to change.
20:35 cjohnson detect, warn, and ask me to reload
20:35 cjohnson but give me the option to keep what I have
20:35 brigadier7 Actually, I merged twice: 4/13 and today
20:35 Primer preaction: they have not been required to change, nor will they
20:35 cjohnson Primer: your personal IDE preferences don't need to be spread like gospel to the world
20:35 preaction Primer: and my belief is that it would be better to make the project more open to any kind of development environment, including IDEs or yes, your coworker who uses nano
20:35 cjohnson other people who want to use their own workflow should be free to do so
20:35 tnecniv joined #git
20:35 cjohnson that's the beauty of git
20:36 holdsworth joined #git
20:36 brigadier7 So the state as of 4/12 is what I want
20:36 cjohnson Primer: If you think they are wasting a huge amount of time, take the time to make the case to them for why your approach might help them
20:36 cjohnson but respect that they may disagree
20:36 _ikke_ brigadier7: and any commits that came after it?
20:37 preaction and also respect that it may damage your working relationship with them
20:37 _ikke_ brigadier7: what should happen to those?
20:37 thebope joined #git
20:37 brigadier7 _ikke_: no, there were no code changes except on master.  My branch is pretty much only mine.
20:37 cjohnson Well making a suggestion shouldn't damage any relationship if they are reasonable people
20:37 preaction depends on if the person making the suggestion is making it reasonably
20:37 osse and whether the person getting it is reasonable
20:38 preaction most "you should use my ide instead of vim" suggestions i've received have been decidedly unreasonable conversations
20:38 brigadier7 _iike_: I don't need anything before the merge, but perhaps in 2 months, I'd want to merge master for real.
20:38 _ikke_ nothing before or after the merge?
20:38 preaction that said, i did have to do quite a bit of work to get Adobe Flex working right using standard Unix toolchain... so annoying...
20:39 osse would you consider what damage to your working relationship with someone before suggesting that to the person that he should put on a new pot of coffee if he empties it?
20:39 brigadier7 _ikke_: I meant after.  I'd like the code in the state it was in on 4/12.  No changes were made to the branch except the merges
20:39 cjohnson I would consider damaging that person that took the last of the coffee osse
20:39 preaction "vim is stupid. you should use X instead" <- not reasonable, and hugely common
20:39 cjohnson With a pipe
20:39 cjohnson lol
20:40 _ikke_ brigadier7: git reset --keep <commit> resets it back to that commit
20:40 _ikke_ find the commit hash of the commit you want on that ate
20:42 brigadier7 _ikke_: I just did that, and I'll let you know how it works....
20:44 tymczenko joined #git
20:45 brigadier7 Once I do that, should I somehow push to the server respository?
20:46 nexemacs_ joined #git
20:47 _ikke_ git push --force-with-lease <remote> <branch>
20:48 Primer I only hope that by me demonstrating that I can run circles around them with an IDE that they'll choose to adopt its usage.
20:48 Primer Anyhow, I didn't want to turn this into a debate about IDEs.
20:48 Primer or lack thereof
20:49 cjohnson consider time wasted evangelizing about your IDE. Who knows, maybe they're more efficient than you even if they're slower with their IDE
20:49 cjohnson lol
20:49 jstimm joined #git
20:49 preaction this debate is not about IDEs, it's about people
20:50 cjohnson and coffee
20:50 cjohnson and why I will cut you if you take the last ofi t
20:51 Primer I think you made it about people. I can think of several equivalent scenarios where pointing out that there exist better tools to do your job.
20:51 brigadier7 _ikke_: Thank you so much.  This is so potentially destructive I really feared doing anything, but I was stuck.  I will toast you this evening with an over-caffeinated beverage.
20:52 Primer And said job becoming faster, better, more efficient...heck, even more pleasant
20:53 _ikke_ :)
20:53 zerow joined #git
20:53 Primer This was the case for me. I just want my co-workers to share in my happiness!
20:53 diogenese joined #git
20:54 preaction and those scenarios may involve less change to use. changing how someone does every single aspect of their job is a huge change to ask for
20:54 gnixev joined #git
20:54 cjohnson so did the christians Primer
20:54 preaction but i don't work with you, thank yhwh, so i don't care. good luck
20:54 Primer I feel like I've touched a nerve here
20:54 osse Who said anything about "every single aspect" except you?
20:55 cjohnson batman
20:55 preaction what else is an IDE but "the single place you spend most of your time in as a developer"?
20:56 Primer I also have a co-worker who refuses to use phpstorm specifically because he disagrees with its license. I've had to point out to him that he's not paying for it, but he doesn't care.
20:56 Primer Do you agree that he should be exempted from using it because of that?
20:56 cjohnson licenses are a philosophical issue as well, not just economic
20:56 preaction yes, i do
20:56 osse I combine using an ide and not using that ide every day
20:56 cjohnson Primer: exempted from what?
20:56 cjohnson who died and made you king of the office editor choices?
20:56 Primer cjohnson: exempted from being "encouraged" to use it
20:57 cjohnson Nobody is stopping you from encouraging
20:57 alexggordon joined #git
20:57 cjohnson I'm sure if you took the time to make a positive and constructive case to them they would hear what you have to say
20:57 cjohnson but it sounds like you don't mean encouraging
20:58 Primer This is exactly what I've been doing. In fact, this is exactly why I'm here, to find a way to satisfy a co-worker's desire to manage project/branch based files in the IDE I'm encouraging him to use.
20:58 cjohnson Have them set editor to reload files automatically
20:58 cjohnson done
20:59 Primer Right...if only such an option existed, given the criteria
20:59 Atm0spher1c joined #git
20:59 cjohnson pretty much every editor supports reloading files as they change
20:59 cjohnson if it doesn't then you are encouraging them to use a bad editor :P
21:00 preaction they also want to reload groups of files based on what branch is checked-out (so, some kind of "session")
21:00 osse but are you suggesting that they use that IDE for *every single aspect of their job*? Or just some of it?
21:00 thebope joined #git
21:00 cjohnson lol
21:00 cdown joined #git
21:00 brigadier7 _ikke_: Bingo!  I'm in business again.  You rock
21:00 Dewin joined #git
21:00 cjohnson preaction: oh, lots of editors support that too
21:00 cjohnson Primer ^
21:00 preaction cjohnson; yes, i know
21:00 Silenced joined #git
21:00 Primer The goal was to maintain a "view of project files" which would refresh based on the branch
21:01 cjohnson nah, that's not necessary, just use sessions
21:01 cjohnson one session for 1 feature 1 session for another
21:01 cjohnson lots of editors support that too
21:01 preaction osse: so, Primer will be satisfied if nano-guy uses the IDE just to do search/replace over the code and not actually write code with it?
21:01 Primer osse: I'm only suggesting that they try the IDE and avail of the features that have the potential to make their jobs easier, better, faster
21:01 Primer preaction: that's a start
21:01 livingstn joined #git
21:02 cjohnson If they don't want to, they don't want to. If you continue to push beyond that, you are not doing encouragment
21:02 osse preaction: no idea, but it seems to me that what you're arguing against is not what Primer is saying but quite the hyperbole of it
21:02 preaction trying a new tool takes weeks to get up-to-speed and comfortable. you're requiring them to go through weeks of frustration
21:02 preaction sorry, not requiring, requesting
21:02 osse some tools are worth it
21:03 Primer These guys have had a license for this IDE close to 1 year now. I've not forced anyone to do anything.
21:03 osse how to drive a car comes to mind
21:03 preaction you don't get to make that judgement for me
21:03 osse i don't
21:03 osse and i'm not
21:03 cjohnson Primer: no but you are taking offense that they haven't jumped ship onto what you personally believe is good
21:03 cjohnson that's what is annoying
21:03 Primer I do, however, take the time to point out how I would do tedious and time-consuming takes in the IDE when I see them do something their way.
21:03 cjohnson nothing wrong with making a suggestion, just like there's nothing wrong with them disagreeing and not taking it
21:04 preaction and the more time you spend trying to convince them that you're right, the less likely they'll think you're just trying to be helpful
21:04 osse but I would also argue that refusing to learn these tools shows attitudes that developers should not have
21:04 preaction i refuse to learn phpshark or whatever the ide is
21:04 cjohnson Sure but that's something for the manager to fix, not the coworker
21:04 cjohnson bring it up to the manager if it's such a major issue
21:04 cjohnson if they're wasting such monumental amounts of time
21:04 Primer cjohnson: Taking offense would imply this is something personal. Is asking for more efficiency in your job something that I, as a manager, should not expect?
21:05 preaction are they meeting project deadlines?
21:05 preaction are those deadlines reasonable?
21:05 preaction if yes, then why does it matter?
21:05 cjohnson Primer: I can say definitively that you are definitely taking it personal lol, just based on all your interactions here
21:05 Primer Scotty factor
21:05 Primer I'm guilty of it too
21:06 preaction ... so you're saying if they picked a different tool, you wouldn't have to pad your estimates for unknown variables because... why?
21:07 osse in this case offense is taken, not given.
21:08 osse *even if* deadlines are met, if a dev feels insulted from being given friendly advice (yes, *friendly* is a condition in my opinion) then screw them
21:08 preaction that doesn't make it unoffensive
21:08 Primer I'm trying really hard not to go into hyperbole myself here.
21:08 osse if the advice is not friendly then it's a different story
21:08 preaction repeating advice over and over is not friendly
21:08 cjohnson I agree with that osse
21:08 cjohnson and I agree with that preaction
21:08 cjohnson lol
21:08 cjohnson if you make the case and they don't agree with you, that's that. maybe bringi t up once or twice more
21:09 cjohnson but buying them a license to an editor and continually pushing them to use it is not friendly
21:09 Primer But the only thing I can think of are examples of times when sloppy code has been turned in, code that could have been better if the developer had been using better tools.
21:09 preaction "hey i just found this new thing that really helped me" <- friendly advice
21:09 cjohnson if they are missing deadlines or something and it's entirely because of their editor choices then sure, that's a performance issue
21:09 kyan joined #git
21:09 nexemacs joined #git
21:09 cjohnson I doubt that Primer
21:09 preaction how would it be better if they were using better tools?
21:09 cjohnson sloppy code comes from sloppy coders
21:09 osse that's what CI is for!
21:09 jstimm joined #git
21:10 cjohnson editors, like guns, cannot be blamed for the things the monkeys you give them to do
21:10 kent\n everyone has a CI installation
21:10 Es0teric joined #git
21:10 kent\n just some people also have a production environment as well
21:10 Primer Pardon my ignorance, but...CI?
21:10 preaction if you are having a problem with sloppy code, that's a lot easier to fix: point it out, explain how it should be corrected, and then you even have a good reason to say "the phpstorm thing even does this automatically for you if you want to try it out"
21:10 cjohnson continuous integration Primer
21:10 Primer I'm not being facetious here either
21:10 cjohnson automatically testing their code
21:10 cjohnson If it fails they get warnings traight away to fix it
21:10 Primer yeah, I had a feeling it was something like this
21:10 preaction Travis or Jenkins, like
21:11 preaction the server judges the code and tells them if they live or die!
21:11 cjohnson you're trying to fix this from the wrong end
21:11 cjohnson check it in the release pipeline and don't let shit pass. you can do linting as well
21:11 cjohnson and fail if they fail linting
21:11 osse I like it when people write #include "StuFF.h" and the CI picks up on it because it's running runix
21:12 Primer In a perfect world, every place has all of these things.
21:12 cdown joined #git
21:12 preaction except the mandated IDE
21:12 cjohnson CI is not hard to set up if your project is on github
21:12 cjohnson just plug in the command that runs your tests
21:12 osse "mandated". there's the hyperbole again
21:12 Primer What tests? :)
21:12 cjohnson and if your codebase isn't tested, well, imo, that's a failure of you
21:12 cjohnson for not allocating time and enforcing testing requirements
21:12 cjohnson assuming you're the manager
21:13 preaction osse: if my boss continually harangued me for not using a certain tool, i would consider that a mandate, yes
21:13 osse Primer: on the other hand, if devs come to you and ask for time to set up a build server or the like, you should consider letting them.
21:13 cjohnson if you are letting people push code live that doesn't have associated tests then you are the one generating the sloppy code Primer
21:13 Primer "Hey management, I'd like to setup automated testing."
21:13 osse At our place we've waasting soooo much time trying to do stuff in a way that saves time
21:13 cjohnson Primer: Use *whatever logic* you are currently trying to use to enforce they use an editor
21:13 Primer "Sounds good! We can pencil that into Q4 next year. How does that sound?"
21:13 cjohnson if it's good enough to get them to use a different editor it's good enough to justify spending time testing
21:13 preaction it was different when i thought this was just a dev talking to another dev. this is a boss talking to their devs, that makes "suggestions" into a bit more
21:14 cjohnson if you can't justify testing then you have no valid reason to try to get them to change editors either
21:14 cjohnson code clarity and quality, stability, etc
21:14 cjohnson it's an easy sell. If you are so slammed that it's impossible to fit in then why is it reasonable to spend soe much of your time trying to get them to switch editors? Obviously they are slammed right?
21:15 dsdeiz joined #git
21:15 Primer So far all I've done is obtain licenses for my co-workers and encouraged them to use it. The use is not mandated, nor will it be.
21:15 cjohnson a manager continually hounding you to use something is soft-mandated
21:15 safe joined #git
21:16 cjohnson or at the very least a waste of time
21:16 Primer The nano guy tells me he wants to use the IDE, but that he wants to start using it on a "smaller project". Except, none are on the horizon.
21:16 osse then perhaps let him spend a day of the week on a pet project. something internal maybe
21:16 osse like... a testing suite!
21:16 Primer osse: wonderful idea!
21:17 cjohnson You don't have to have 100% coverage out of hte gate either
21:17 cjohnson set up tests and then mandate that all new code be tested
21:17 cjohnson you can come back next year to test the rest
21:18 cjohnson no major time cost there
21:18 Primer So...mandate testing...what if I meet resistance?
21:18 askb joined #git
21:18 Primer "Sorry, I don't like this idea of testing..."
21:18 cjohnson I thought you were the manager
21:18 cjohnson Well testing has a huge list of pros
21:19 Primer Ok...
21:19 rorro joined #git
21:19 cjohnson You can come up with their entire argument against it ahead of time
21:19 Primer Should I reach for some popcorn?
21:19 cjohnson and write out an argument for why the pros outweigh the cons
21:19 Primer Sorry
21:19 Primer Biting my lip here
21:19 osse Primer: then you quit your job as a manager, start working somewhere else as a dev, because who wants to be a manager in the first place :O
21:19 cjohnson don't hurt yourself
21:19 cjohnson osse on point
21:20 Primer osse: you're firing on all cylinders, man!
21:20 osse :P
21:21 Primer It's funny, but I've had this conversation so many times before. And it all starts from something innocent.
21:21 livingstn joined #git
21:21 cjohnson Seriously though, if your devs are pushing shit/sloppy code to production, it's because you are letting them. To fix that, you can either start code reviewing every commit, which is simply not feasible, or you can set up automated enforcement
21:21 cjohnson tests, linting, etc
21:22 cjohnson and if they refuse to participate, and they refuse to improve the quality of code, and it's not because you are running them ragged and not giving them enough time to do things
21:22 tribly_ joined #git
21:22 cjohnson fire them
21:22 jstimm joined #git
21:23 Primer We've had this project called "major upgrade" on the books now for over 1 year. It's where I get to do all that: company-wide consensus-based code review, automated testing, aggressive pre-commit hooks for linting (I have this, but it's very informal), etc.
21:23 kristofferR joined #git
21:23 medeirosthiiago joined #git
21:23 cjohnson it doesn't need to be anything major, like I said
21:23 Primer Thing is, it's kind of become a running joke.
21:23 cjohnson you don't need 100%  coverage out of hte gate
21:24 cjohnson in the time you have spent in this channel you could have signed up for circle-ci and gotten your app deploying there :P
21:24 Primer Well, it's called major upgrade for a different reason, which has to do with upgrading from stale APIs and libraries.
21:24 cjohnson then spend a little time adding a testing suite
21:24 cjohnson yeah that's a different project
21:24 rominronin joined #git
21:24 cjohnson don't try to mix them into one
21:25 Primer Historically I've had to "ear mark" such things here.
21:25 cjohnson I'm saying that you yourself could get the bare tools needed to do these things up and running in the next day or two
21:25 cjohnson and obviously you've got time to sit on IRC (no offense)
21:26 cjohnson then write a couple tests for a few small things, make sure they work, teach others how to write tests, and then start enforcing it. it doesn't have to be this major thing that people make it out to be
21:26 Primer It's like a bill (a feature request). That gets sent to committee (my boss), and I then "ear mark" what I can onto it. As long as I can make the thing I want land within the projected time frame, it's all good.
21:26 cjohnson come back during your refactor and make sure you add time while refactoring to write tests for the refactored code
21:26 cjohnson yeah that's pretty standard
21:26 Primer But I often have to deal with unrealistic dealines.
21:26 cjohnson Ok so why are you blaming your devs and their editors then? lol
21:27 Primer Part of my job is research. Spending time on IRC is time well spent.
21:27 Snugglebash joined #git
21:27 rahtgaz joined #git
21:27 cjohnson Yeah I know, I agree. I was just saying, you aren't SO slammed that you don't have time to spend the next couple days setting up testing suites and CI
21:27 cjohnson so when people suggest it to you and you snort and act like it's an impossible goal, it's telling
21:28 Primer It's obviously not an impossible goal. It's just that something like this, currently, has to be done out of band.
21:28 cdown joined #git
21:29 cjohnson right, and my point was, you could be using your IRC time as that out of band time
21:29 cjohnson to get it bootstrapped
21:29 Primer I'm lucky in that my boss actually acknowledges when I do this.
21:30 Primer I'm feeling more motivated after this discussion.
21:30 courrier joined #git
21:30 cads2 joined #git
21:31 Primer Thank you cjohnson, osse, preaction, kent\n
21:31 cjohnson !!! good
21:31 cjohnson tests!
21:31 cjohnson do them
21:31 cjohnson or die
21:31 Primer That will be my new motto!
21:31 nexemacs joined #git
21:31 kent\n a test that passes is a useless test.
21:32 kent\n you want tests that fail.... then you fix them so they pass.
21:32 Primer test aut mori motto nostrum est!
21:33 anuxivm joined #git
21:33 Primer kent\n: I excel at failing
21:34 kent\n Cogito ergo sum ego temptare
21:34 kent\n hm, not parity-safe :/
21:35 gopar joined #git
21:36 a3Dman joined #git
21:37 druonysus_ joined #git
21:39 gigq joined #git
21:39 maroloccio joined #git
21:44 tnecniv joined #git
21:47 Cabanossi joined #git
21:48 tymczenko joined #git
21:48 portal_ joined #git
21:48 gopar joined #git
21:50 Vaelatern joined #git
21:53 overlord_tm joined #git
21:55 Dougie187 left #git
21:58 nugb joined #git
22:02 gunnaro joined #git
22:03 zulutango joined #git
22:11 andrew9184 joined #git
22:12 bannakaf_ joined #git
22:20 fahadash joined #git
22:21 NeXTSUN joined #git
22:21 systemovich joined #git
22:24 Darren_ joined #git
22:26 albel727 joined #git
22:27 zivester joined #git
22:28 gopar joined #git
22:30 pandeiro joined #git
22:31 ChrisF79 joined #git
22:32 ChrisF79 Hi everyone.  I have some repos at bitbucket and just reformatted my machine.  I copied my ~/Sites folder back to where it should be but I'm confused how I relink the two.
22:32 gopar joined #git
22:33 kadoban ChrisF79: Hmm? What is in that folder, and why should it be linked to bitbucket? Is it a git repo? Git repos are just files, so if you successfully restored from backups, congratulations, you're all done really.
22:34 ChrisF79 It isn't working.  Let me get the exact message.
22:34 dendazen joined #git
22:34 ChrisF79 actually... it's on the server side
22:35 kadoban Oh. You probably don't have your SSH keys/aliases set up correctly maybe.
22:35 ChrisF79 when I do git pull "git@bitbucket.org:cfarrugia/mybucket.git master" it lists every wordpress file (too many so I can't even see the top) and says at the bottom "Please move or remove them before you can merge.  Aborting"
22:35 ChrisF79 I did have to add an SSH key
22:35 kadoban That doesn't sound server-side
22:37 ChrisF79 well, on my local machine I changed one file.  I added everything and did a git push.  It says my branch is up-to-date with 'origin/master'
22:37 ChrisF79 so then I go to the server to do git pull and that's what I get
22:37 kadoban Oh, you're talking about some server you have?
22:37 ChrisF79 yes
22:39 kadoban Wouldn't know what states your separate repos are in ... not really sure what to help with that. That pull command looks suspicuous, rarely is it a good idea to be specifying a URL like that really, and I ... don't see how that command could even work with the quotes around both the URL and the ref name like that
22:39 ChrisF79 I just quoted it
22:39 ChrisF79 but yeah, it is a weird one
22:39 ChrisF79 Just doing git pull shows this: There is no tracking information for the current branch.
22:40 kadoban Well, how did you set up the "server" repo? Are you using this for deployment or something it sounds like? How does it usually work, what command do you usually run?
22:41 kadoban Are you saying the quotes don't exist in the command you actually ran? Because that would make more sense.
22:41 ChrisF79 yes on the quoets
22:41 bilal_ joined #git
22:41 ChrisF79 Normally what I do is develop locally and as I go, I just commit and do git push
22:41 ChrisF79 so bitbucket has my files... then, I head onto the server and USUALLY just do git pull
22:41 ChrisF79 and suddenly that isn't working
22:42 kadoban Which machine are you talking about that you reformated and restored?
22:42 ChrisF79 my local machine
22:42 kadoban Then that wouldn't have changed how the "server" machine works, so that doesn't make a lot of sense :-/
22:42 ChrisF79 although, I had someone trying to help me and I'm thinking he may have removed my .git folder on the server
22:42 ChrisF79 that's all I can think of
22:42 kadoban Did you run any other commands on the server machine?
22:42 kadoban Oh
22:42 kadoban Well, that sounds like a mess.
22:43 ChrisF79 heh, you're right about that!
22:44 kadoban Can you ask them what they did I guess? But without doing that, I'm not sure what to suggest. I guess I'd start with looking at the server git repo, seeing what state the files are in, if it seems like it matches what's expected and the remotes are set up correctly and etc.
22:44 ChrisF79 The remote on bitbucket is great
22:44 Vampire0 ChrisF79, what does `git status` on the server say?
22:44 bilal joined #git
22:45 kadoban That's good at least. And is the history on your current branch on the server look the same as is on bitbucket? And yeah, that too ^
22:45 ChrisF79 https://www.dropbox.com/s/metm26wu73in8lp/Screenshot%202017-04-20%2018.45.15.png?dl=0
22:45 kadoban By the way, 'git pull' isn't exactly what I'd recommend for deployment
22:45 ChrisF79 kadoban, I'm here to learn.  What do you recommend?
22:46 kadoban Not using git for deployment. I use 'rsync' usually. Typically I track a bash script with the correct rsync command and run that when I want to.
22:46 Vampire0 ChrisF79, there you have it, it tells you all you need to know
22:46 Vampire0 ChrisF79, "On branch master ... Initial commit"
22:46 kadoban But even if you want to use git, pull doesn't make a ton of sense because merging on the server is just a weird thing to want.
22:47 kadoban Yeah, initial commit sounds weird
22:47 Vampire0 ChrisF79, so indeed someone nuked your history, probably by deleting .git and then doing `git init`
22:47 gopar joined #git
22:47 Vampire0 ChrisF79, just re-clone the repo on the server and you are probably fine
22:48 ChrisF79 that's exactly what I think happened
22:48 Vampire0 ChrisF79, and regarding deployment with git, you might want to read !deploy
22:48 gitinfo ChrisF79: 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
22:48 ChrisF79 ok, just googled it and I think I fixed it now that I understand the problem
22:48 kadoban Make a backup of the files first, this kind of thing would usually have untracked files that a re-clone isn't going to have for you
22:49 alexandre9099 joined #git
22:49 Gustavo6046 joined #git
22:49 ChrisF79 yep, I'm working again!
22:50 ChrisF79 although wrongly according to what you guys have taught me
22:50 ChrisF79 but, at least I'm back to where I should be for now
22:50 cagomez joined #git
22:50 ChrisF79 oh, one other question.  I have /blog/wp-config.php on both servers.  I added blog/wp-config.php to the .gitignore file but it still tries to sync
22:51 frobnic joined #git
22:51 kadoban ChrisF79: .gitignore and other similar mechanisms doesn't have any effect at all on already-tracked files.
22:51 kadoban All it really does is make git not list them in 'git status' as untracked and prevents 'git add' from accidentally adding them, that's about it.
22:51 ChrisF79 got it
22:51 ChrisF79 thanks so much
22:51 kadoban Anytime, glad you figured out your thing
22:53 AfroThundr joined #git
22:55 rominronin joined #git
22:56 holdsworth joined #git
23:01 hahuang65 joined #git
23:01 gitinfo ChrisF79: 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.
23:10 monq joined #git
23:17 fakenerd_ joined #git
23:19 ki0 joined #git
23:19 robattila256 joined #git
23:23 gnixev joined #git
23:24 snowkidind joined #git
23:24 dirtyroshi joined #git
23:27 alexggordon joined #git
23:28 oleo joined #git
23:28 fakenerd joined #git
23:28 bket joined #git
23:30 msonntag joined #git
23:30 bl0w3d_0ut joined #git
23:31 Achylles joined #git
23:31 gunnaro joined #git
23:31 Noodlewitt joined #git
23:32 cagomez I have already made changes in my branch. will applying a stash remove those changes?
23:32 cagomez or will I end up with original changes + stash changes?
23:32 menip joined #git
23:33 Cabanossi joined #git
23:33 kadoban The latter. If there are conflicts, you'll have to resolve them
23:34 FrostCandy joined #git
23:35 Gustavo6046_ joined #git
23:37 jaziz joined #git
23:37 Noldorin joined #git
23:43 ki0 joined #git
23:46 maroloccio joined #git
23:47 TbobbyZ_ joined #git
23:48 ki0 joined #git
23:54 Kaisyu joined #git
23:54 Gustavo6046 joined #git
23:55 rmesta joined #git
23:57 anuxivm left #git
23:59 shgysk8zer0 joined #git

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