These are chat archives for astropy/astropy

24th
Mar 2017
Aarya Patil
@AustereCuriosity
Mar 24 2017 04:15
@bsipocz Do tell me when you get some time. :smile:
Also, I haven't been able to get across @eteq @hamogu @taldcroft. I tried the mailing list and emailing them too.
matrixbot
@matrixbot
Mar 24 2017 11:41
bsipocz I think you've got an answer in the meantime.
bsipocz As a rule of thumb there is always answers, but since we all do this besides our day jobs, sometimes there is a few hours/days before an answer.
bsipocz I have some very limited time today if you want to finish up that rebase (though looking at it, there isn't any more conflict after the second one, which is a trivial one, so we should be able to get it done in 10mins max)
Aarya Patil
@AustereCuriosity
Mar 24 2017 14:10
Yeah the second one is failing
Aarya Patil
@AustereCuriosity
Mar 24 2017 14:24
And yes I did get an answer :smile:
matrixbot
@matrixbot
Mar 24 2017 18:04
bsipocz @austerecuriosity - are you still here?
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:05
Yes
matrixbot
@matrixbot
Mar 24 2017 18:05
bsipocz do you want to finish the rebase now?
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:05
Yes would sure want to
matrixbot
@matrixbot
Mar 24 2017 18:07
bsipocz so now you have a second conflict, right?
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:07
Yes, the commit after the rebase attempts fails
matrixbot
@matrixbot
Mar 24 2017 18:08
bsipocz let me ho back to that point
bsipocz as I remember it's a conflict in one file, yes?
bsipocz git status
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:08
Yes in one file
matrixbot
@matrixbot
Mar 24 2017 18:09
bsipocz ok, open it in an editor and look for the <<<<<< HEAD string
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:09

rebase in progress; onto 4541e01
You are currently rebasing branch 'astropy_issue_5845' on '4541e01'.
(fix conflicts and then run "git rebase --continue")
(use "git rebase --skip" to skip this patch)
(use "git rebase --abort" to check out the original branch)

Changes to be committed:
(use "git reset HEAD <file>..." to unstage)

modified:   CHANGES.rst
modified:   astropy/io/ascii/fastbasic.py

Unmerged paths:
(use "git reset HEAD <file>..." to unstage)
(use "git add <file>..." to mark resolution)

both modified:   astropy/io/ascii/core.py
Found it
matrixbot
@matrixbot
Mar 24 2017 18:13
bsipocz you see that the conflict is a very simple one, and due to on a minor PEP8 change
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:14
Yes the one regarding the ValueError String
matrixbot
@matrixbot
Mar 24 2017 18:14
bsipocz keep what was is HEAD, remove all lines with the conflict decorating strings ("<<<<..." "=====", ">>>>>")
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:16
Done
matrixbot
@matrixbot
Mar 24 2017 18:16
bsipocz save and exit
bsipocz git rebase --continue
bsipocz there shouldn't be any more conflict
bsipocz (ohh, don't forget to git add astropy/io/ascii/core.py
bsipocz before continue
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:18
Successfully rebased and updated refs/heads/astropy_issue_5845.
:smile:
matrixbot
@matrixbot
Mar 24 2017 18:19
bsipocz great
Zé Vinícius
@mirca
Mar 24 2017 18:19
@AustereCuriosity congrats! =)
matrixbot
@matrixbot
Mar 24 2017 18:19
bsipocz now you have to push it back to github and you're done
bsipocz use --force when pushing
bsipocz (you need this was the history was rewritten, so it won't be an incremental change as usual, but involve removing already pushed commits)
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:21
To https://github.com/AustereCuriosity/astropy.git
  • f136e96...afdc2f9 astropy_issue_5845 -> astropy_issue_5845 (forced update)
So basically force will change the entire commits history instead of creating a new one?
matrixbot
@matrixbot
Mar 24 2017 18:22
bsipocz yes.
bsipocz the commit hashes changes even for commits that didn't have any conflict
bsipocz and the content of commits may have also changed if there were conflicts.
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:24
Thanks for the help
It felt so good seeing the rebase message
matrixbot
@matrixbot
Mar 24 2017 18:25
bsipocz in the long term getting a git GUI can be useful to see what's going on different branches and to have a quick peek into commits. I use gitx (available on on osx), and gitk. But there are others out there.
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:26
So all the changes that you told me to make, the commit messages and everything, you monitored that through gitx?
matrixbot
@matrixbot
Mar 24 2017 18:26
bsipocz No, I can't see what you have locally with you.
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:27
No, I meant the commits relating to pushed changes
matrixbot
@matrixbot
Mar 24 2017 18:27
bsipocz But yes, checked out the branch you opened the PR from and rebased it along with you
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:29
Rebasing has a lot more to it than I initially thought
And I learnt a lot :D
matrixbot
@matrixbot
Mar 24 2017 18:30
bsipocz you'll get the hang of it, git is very powerfull
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:31
It sure is
I merge the remaining commits as well
matrixbot
@matrixbot
Mar 24 2017 18:48
bsipocz @AustereCuriosity why did you do a merge commit again?
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:49
There were some commits in astropy/master
matrixbot
@matrixbot
Mar 24 2017 18:49
bsipocz you don't have to worry about those
bsipocz the only case is when there are changes to the part you're patching, too
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:50
I thought I had to keep it updated to the current master
matrixbot
@matrixbot
Mar 24 2017 18:51
bsipocz still, merging it into your feature branch is not the recommended way, basically you should never use git merge
bsipocz however there are several test failures due to your code, you need to fix those
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:51
So if I have to update my branch I should just do it if there are merge conflict
matrixbot
@matrixbot
Mar 24 2017 18:52
bsipocz (you can run the tests locally: python setup.py test)
bsipocz you update your branch with a rebase, but need only do it when asked by a maintainer
Aarya Patil
@AustereCuriosity
Mar 24 2017 18:53
Oh, I'll keep that in mind
I'll fix the tests
I'd fixed all tests earlier, so are these failing because of the rebase?
matrixbot
@matrixbot
Mar 24 2017 18:56
bsipocz yes, possible that functions you've relied on have changed. Also the testing environments changed in the meantime
Aarya Patil
@AustereCuriosity
Mar 24 2017 19:37
Rather, I don't know where exactly to mention this should not cause a ValueError