These are chat archives for astropy/astropy

23rd
Mar 2017
matrixbot
@matrixbot
Mar 23 2017 00:02
Cadair bsipocz: you have been on a PR / issue closing rampage I see :D
bsipocz I blame @hamogu to start this, but he is right some stuff needs to be closed.
Cadair :D
Cadair you can come do SunPy as well if you want!!!
bsipocz We still have 117 open PRs ;)
Cadair I am sad that SunPy has cleared 60
matrixbot
@matrixbot
Mar 23 2017 00:07
bsipocz don't worry, you will catch up, at least the monthly stat predicts it (https://github.com/sunpy/sunpy/pulse/monthly)
Cadair +5
Cadair I must close some!!
bsipocz -13 :)
bsipocz (without the ones closed without merge)
Cadair impressive!
matrixbot
@matrixbot
Mar 23 2017 06:18
mohanagr on Freenode Cadair I was looking into your PR no. #5770 for drafting my GSoC application. 4 plugins have been separated out. Idea is basically to include all of them in a separate package right?
mohanagr on Freenode I'll try and fix that PR since some tests are failing for Python2 as pllim mentioned. Also Cadair/bsipocz I'll be glad if one of you could tell me if the mailing list is active. I'll be thankful if I could receive some guidance for my application for the Test helper package.
matrixbot
@matrixbot
Mar 23 2017 11:04
Cadair My original idea was not to have them in a separate package, but to allow packages that depend in astropy to pick and choose which ones they want to use
Cadair But it enables that as well
Aarya Patil
@AustereCuriosity
Mar 23 2017 14:27
@hamogu @eteq @taldcroft, I had a few doubts regarding the Seamless Combination of SkyCoord, Table, WCS and FITS project.
I am working on its proposal and I wanted to know the exact goals as the issue discussion encompasses a lot of things
Aarya Patil
@AustereCuriosity
Mar 23 2017 16:00
Is this a good place to contact them?
Zé Vinícius
@mirca
Mar 23 2017 17:14
Hi @AustereCuriosity definitely a great place
However, I’m affraid @hamogu does not hear this channel
But you should definitely contact your potential mentors by email as well =)
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:26
Personal email would do?
I've tried the mailing list. Thanks though :smile:
matrixbot
@matrixbot
Mar 23 2017 17:47
bsipocz @austerecuriosity - are you here?
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:47
yes :smile:
matrixbot
@matrixbot
Mar 23 2017 17:48
bsipocz what do you have for the following command (to check which forks you have):
git remote -v
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:49
aaryas-MacBook-Pro:astropy Aarya$ git remote -v
AustereCuriosity https://github.com/AustereCuriosity/astropy.git (fetch)
AustereCuriosity https://github.com/AustereCuriosity/astropy.git (push)
astropy git://github.com/astropy/astropy.git (fetch)
astropy git://github.com/astropy/astropy.git (push)
matrixbot
@matrixbot
Mar 23 2017 17:49
bsipocz excellent
bsipocz The first fetch everything to be up to date:
bsipocz git fetch --all
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:50
Done
Fetching AustereCuriosity
Fetching astropy
remote: Counting objects: 5, done.
remote: Compressing objects: 100% (5/5), done.
remote: Total 5 (delta 0), reused 0 (delta 0), pack-reused 0
Unpacking objects: 100% (5/5), done.
From git://github.com/astropy/astropy
d97b09b..4541e01 master -> astropy/master
I'd done git fetch astropy before
matrixbot
@matrixbot
Mar 23 2017 17:51
bsipocz great
bsipocz git rebase -i astropy/master
bsipocz this will bring up an editor windows
bsipocz window
bsipocz oh, wait
bsipocz making a copy of the branch may be useful before you do a rebase
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:53
I've done that beforehand
matrixbot
@matrixbot
Mar 23 2017 17:53
bsipocz ok, so then from the commits you have, I see 12
bsipocz however many of those are not relevant as are due to a previous rebase attempt
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:55
Yes. Many of them were because there were merge conflicts
matrixbot
@matrixbot
Mar 23 2017 17:55
bsipocz so in the editor remove those commits, I think that will help a bit
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:56
So all of those which involved failed rebase attempts?
matrixbot
@matrixbot
Mar 23 2017 17:57
bsipocz yes. those are not your original code changes, are most of them removed in later commits anyway.
Aarya Patil
@AustereCuriosity
Mar 23 2017 17:57
Okay just a minute
matrixbot
@matrixbot
Mar 23 2017 17:57
bsipocz also while we're at it, remove all traces of the big test file from the history
bsipocz for that you move the last commit (with pick f136e96c2 Removed test file)
bsipocz underneath the commit you've added the tests (pick 9fc481ff2 added tests)
bsipocz change the "pick" of the latter commit to "fixup"
bsipocz so it will merge the two commits together, basically getting rid of the file
bsipocz (we have to repeat this for removing the png file, too, but don't yet have a commit for that).
bsipocz let me know if you are ready
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:05
Just a minute
I added the test file after the commit 9fc481ff2
That is the commit after added test file I think
matrixbot
@matrixbot
Mar 23 2017 18:09
bsipocz yes, you added the file in 9fc481ff2 and removed it in f136e96c2. These two should be merged ideally.
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:10
Oh so they basically remove the traces
matrixbot
@matrixbot
Mar 23 2017 18:10
bsipocz if ready you shall see 8 non commented lines in the editor
bsipocz save and exit
bsipocz first commit will result a conflict
bsipocz both modified: docs/io/ascii/fast_ascii_io.rst
bsipocz (sorry if I'm rushing, but got to go in 15 mins)
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:11
pick 9c21649 Fixes issue 5845
pick 09ccb30 minor changes
pick 66cba12 Default Keyword Argument for Backward compatibility
pick 9fc481f added tests
fixup f136e96 Removed test file
pick 6e0d025 Corrected failed tests
pick 0ce5838 Removed tabs, trailing whitespace,blank line whitespace
pick 3df986d Docstring formatting
pick 308b787 Remove test.png
pick 20f8350 Merge conflict
Should I remove test.png now itself?
matrixbot
@matrixbot
Mar 23 2017 18:12
bsipocz nope, you haven't pushed all of these up (thus I didn't see them)
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:13
Yes, so these are 10 non commented lines
matrixbot
@matrixbot
Mar 23 2017 18:13
bsipocz move 308b787 under 0ce5838 and change it to "fixup" too
bsipocz also remove the last merge conflict commit (as I assume it was just the side effect of another rebase gone wrong)
bsipocz so you have 9 commits
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:14
Done
pick 9c21649 Fixes issue 5845
pick 09ccb30 minor changes
pick 66cba12 Default Keyword Argument for Backward compatibility
pick 9fc481f added tests
fixup f136e96 Removed test file
pick 6e0d025 Corrected failed tests
pick 0ce5838 Removed tabs, trailing whitespace,blank line whitespace
pick 308b787 Remove test.png
pick 3df986d Docstring formatting
matrixbot
@matrixbot
Mar 23 2017 18:15
bsipocz save and exit
bsipocz nope
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:15
oh wait
matrixbot
@matrixbot
Mar 23 2017 18:15
bsipocz change "pick" to "fixup" for "308b787"
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:15
test.png should be
fixup
yeah figured
Done
matrixbot
@matrixbot
Mar 23 2017 18:16
bsipocz and also remove the last docstring formatting, as I see that's another remnant of a rebase
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:16
8 non-commented lines
two fixups
done
matrixbot
@matrixbot
Mar 23 2017 18:17
bsipocz ok, great
bsipocz save and exit the editor
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:18

error: could not apply 9c21649... Fixes issue 5845

When you have resolved this problem, run "git rebase --continue".
If you prefer to skip this patch, run "git rebase --skip" instead.
To check out the original branch and stop rebasing, run "git rebase --abort".
Could not apply 9c216497277a0dfef2f8530b0372db4040c3397d... Fixes issue 5845

matrixbot
@matrixbot
Mar 23 2017 18:19
bsipocz check git status
bsipocz it will show that there is conflict in one file
bsipocz both modified: docs/io/ascii/fast_ascii_io.rst
bsipocz open that file in an editor to resolve the conflict
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:20
Okay
matrixbot
@matrixbot
Mar 23 2017 18:20
bsipocz look for the str <<<<<<< HEAD
bsipocz that's where the conflict starts
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:20
Yes got that
So what exactly do I change there

<<<<<<< HEAD

.. _fast_conversion_opts:

=======

In order to optimize memory usage, specify the dtype of columns when reading in data using the fast readers.
This can be done using the converters paramater which is consistent with the converters paramater used in basic read.

import numpy as np
converters = {'col1': [ascii.convert_numpy(np.uint)],
... 'col2': [ascii.convert_numpy(np.float32)]}
ascii.read('file.dat', format = 'fast_basic', converters=converters)

9c21649... Fixes issue 5845

matrixbot
@matrixbot
Mar 23 2017 18:21
bsipocz So that section is the docs you wrote, right?
bsipocz figure out what's the best place for it in the new modified version
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:21
Yes, I'd written that part. I had tried removing it and rebasing but failed
matrixbot
@matrixbot
Mar 23 2017 18:22
bsipocz don't remove it, just try to find the right place for it.
bsipocz What you need to remove is the strings "<<<<<<< HEAD" and "======" and at the bottom ">>>>>>> 9c2164972... Fixes issue 5845"
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:23
What exactly would mean a right place>
matrixbot
@matrixbot
Mar 23 2017 18:23
bsipocz so this docs rendered here:
bsipocz currently you've insterted the example right under the title here:
bsipocz if you think that's a good place for it then you can leave it, otherwise move around in the file whereever it fits best
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:26
Oh okay
matrixbot
@matrixbot
Mar 23 2017 18:26
bsipocz I got to go in 5 mins, so will rush through the next steps
bsipocz alternatively can come back to gitten in about 2 hours time
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:26
Now the parallel and fast options section seems better
I have moved the contents there
matrixbot
@matrixbot
Mar 23 2017 18:28
bsipocz 1) when you're happy with the file, save it and exit the editor
2) in the shell add the modified file to be committed git add docs/io/ascii/fast_ascii_io.rst
3) use command git rebase --continue
4) this brings up an editor where you can change the commit message. you can leave it as is if you like, exit the editor
bsipocz repeat this requence until no commit is left
bsipocz (there is a conflict coming in the next one as well)
bsipocz I'll be back in ~2 hours
Aarya Patil
@AustereCuriosity
Mar 23 2017 18:46
Yes no problem
Aarya Patil
@AustereCuriosity
Mar 23 2017 19:21
I got an error in the first rebase --continue itself

5 files changed, 70 insertions(+), 14 deletions(-)
error: could not apply 09ccb30... minor changes

When you have resolved this problem, run "git rebase --continue".
If you prefer to skip this patch, run "git rebase --skip" instead.
To check out the original branch and stop rebasing, run "git rebase --abort".
Could not apply 09ccb30496fe8ec8e46c59901cf4d272da02c542... minor changes

It is way past midnight here so if possible could you help me out tomorrow whenever you have time?
Mohan Agrawal
@mohanagr
Mar 23 2017 20:53
Where can I contact Cadair and pllim? Can't ping them directly from here.
Is there a personal mail? Thanks.
matrixbot
@matrixbot
Mar 23 2017 20:53
Cadair I am here
Mohan Agrawal
@mohanagr
Mar 23 2017 20:53
@Cadair How will the testing for new package be structured like?
matrixbot
@matrixbot
Mar 23 2017 20:54
Cadair What do you mean?
Mohan Agrawal
@mohanagr
Mar 23 2017 20:54
I mean, won't there be tests for the pytest_plugins in the new package?
4 separate plugins.
matrixbot
@matrixbot
Mar 23 2017 20:56
Cadair Yeah
Cadair I assume you can just test the plugins in a normal fashion with pytest
Mohan Agrawal
@mohanagr
Mar 23 2017 21:01
What timezone are you in btw?
matrixbot
@matrixbot
Mar 23 2017 21:02
Cadair Utc
Cadair Till Sunday then UTC + 1 because daylight savings
Mohan Agrawal
@mohanagr
Mar 23 2017 21:02
Wanting to know when I can reach you. I'm in UTC +5:30! Gonna be 3 AM
matrixbot
@matrixbot
Mar 23 2017 21:03
Cadair :D
Mohan Agrawal
@mohanagr
Mar 23 2017 21:04
I have working on a couple of FITS IO issues. Every time I work on one, another one pops up.
matrixbot
@matrixbot
Mar 23 2017 21:04
Cadair You might be better off emailing the astropy gsoc list, pllim had the original idea to move the code out of astropy.
Cadair Ahhh fits :D
Mohan Agrawal
@mohanagr
Mar 23 2017 21:04
I did email.
No replies yet.
matrixbot
@matrixbot
Mar 23 2017 21:04
Cadair Not sure she is active on here.
Cadair OK maybe she is not active on the list either XD
Mohan Agrawal
@mohanagr
Mar 23 2017 21:04
pllim is very active on GH comments!
matrixbot
@matrixbot
Mar 23 2017 21:05
Cadair Truth!
Mohan Agrawal
@mohanagr
Mar 23 2017 21:05
I will try and submit a patch for your #5770
matrixbot
@matrixbot
Mar 23 2017 21:06
Cadair Excellent! That would be a great help, I have completely run out of time on that particular side quest
Mohan Agrawal
@mohanagr
Mar 23 2017 21:06
pllim said it fails because of some python2 incompatibility
matrixbot
@matrixbot
Mar 23 2017 21:08
Cadair makes sense, don't think I tested it locally with 2.x and I have been coding 3 exclusively for a few months at this point. I forget what I can't do XD
Mohan Agrawal
@mohanagr
Mar 23 2017 21:10
As of now, what I've understood about the plugin splitting thing is :-
1) implement astropy utils and a a few other astropy helper classes separately which plugins depend on
2) make those 4 plugins in 4 importable modules
3) include extensive documentation (doctests also? )
4) add tests for each module
5) package it
I was thinking if we are to make this available for the benefit of other projects, it'd be nice if any extra feature/plugins that can be included?
Can you suggest something?
Mohan Agrawal
@mohanagr
Mar 23 2017 21:15
@Cadair
matrixbot
@matrixbot
Mar 23 2017 21:17
Cadair That sounds pretty comprehensive
Mohan Agrawal
@mohanagr
Mar 23 2017 21:18
About the extra additions, anything you have in mind?
I should just tag pllim in some GH comment to get help on this? :D
matrixbot
@matrixbot
Mar 23 2017 21:23
Cadair There is a lot of features as a part of setup.py test
Mohan Agrawal
@mohanagr
Mar 23 2017 21:24
For example?
Also I forgot to include logging.
matrixbot
@matrixbot
Mar 23 2017 21:24
Cadair I mean available through that interface.
Mohan Agrawal
@mohanagr
Mar 23 2017 21:24
to make it easy to debug. FITS io has been some pain.
matrixbot
@matrixbot
Mar 23 2017 21:24
Cadair Open file handles, etc etc
Cadair I can't remember off the top of my head.
Mohan Agrawal
@mohanagr
Mar 23 2017 21:25
open files handles is a plugin already. (part of those 4)
matrixbot
@matrixbot
Mar 23 2017 21:26
Cadair Sure
Cadair Ok
Mohan Agrawal
@mohanagr
Mar 23 2017 21:26
Also, how about automated testing?
matrixbot
@matrixbot
Mar 23 2017 21:26
Cadair For the plugins?
Mohan Agrawal
@mohanagr
Mar 23 2017 21:26
Since multiple projects will be using these modules.
Yes.
matrixbot
@matrixbot
Mar 23 2017 21:26
Cadair I am not sure. Just need to write tests using pytest
Mohan Agrawal
@mohanagr
Mar 23 2017 21:31
Yes. There really won't be any time consuming tests.