Added a subclass of astropy.time to support utime #2409
Conversation
|
Will need a test too.
…On Wed, Jan 17, 2018 at 11:34 AM, James Mason ***@***.***> wrote:
Implemented handling of utime by creating a custom subclass of
astropy.time.Time, which uses the third code example from their
documentation
<http://docs.astropy.org/en/stable/time/#writing-a-custom-format> as a
template.
------------------------------
You can view, comment on, or merge this pull request online at:
#2409
Commit Summary
- Added a subclass of astropy.time to support utime
File Changes
- *A* sunpy/time/utime.py
<https://github.com/sunpy/sunpy/pull/2409/files#diff-0> (21)
Patch Links:
- https://github.com/sunpy/sunpy/pull/2409.patch
- https://github.com/sunpy/sunpy/pull/2409.diff
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
<#2409>, or mute the thread
<https://github.com/notifications/unsubscribe-auth/AA8CF8DfcOADbB6UmYlE6RNz1aAuH7zOks5tLiEcgaJpZM4Rhn_i>
.
|
|
Also, maybe a short explanation of why this is needed.
On Wed, Jan 17, 2018 at 11:40 AM, Jack Ireland <jackireland@gmail.com>
wrote:
… Will need a test too.
On Wed, Jan 17, 2018 at 11:34 AM, James Mason ***@***.***>
wrote:
> Implemented handling of utime by creating a custom subclass of
> astropy.time.Time, which uses the third code example from their
> documentation
> <http://docs.astropy.org/en/stable/time/#writing-a-custom-format> as a
> template.
> ------------------------------
> You can view, comment on, or merge this pull request online at:
>
> #2409
> Commit Summary
>
> - Added a subclass of astropy.time to support utime
>
> File Changes
>
> - *A* sunpy/time/utime.py
> <https://github.com/sunpy/sunpy/pull/2409/files#diff-0> (21)
>
> Patch Links:
>
> - https://github.com/sunpy/sunpy/pull/2409.patch
> - https://github.com/sunpy/sunpy/pull/2409.diff
>
> —
> You are receiving this because you are subscribed to this thread.
> Reply to this email directly, view it on GitHub
> <#2409>, or mute the thread
> <https://github.com/notifications/unsubscribe-auth/AA8CF8DfcOADbB6UmYlE6RNz1aAuH7zOks5tLiEcgaJpZM4Rhn_i>
> .
>
|
|
Will do. It's needed because @Cadair told me so? 😛 |
|
Knowing very little about this, can you construct it with the float? A test for that use would be great as well. (Seeing how that's the most likely use of this, not to save out utime as we are not monsters :p ) |
|
I was afraid of you asking for that.. that's what I spent a couple hours on yesterday (but for a new yyyydoy_sod subclass) and that is much harder to figure out since I haven't found any comprehensive astropy examples. I'll keep at it though. |
|
This PR is now dependent on the resolution of issue 7092 in astropy that will enable, in this case, utime as an input. |
sunpy/time/utime.py
Outdated
|
|
||
|
|
||
| class TimeUTime(TimeFromEpoch): | ||
| """Seconds from 1979-01-01 00:00:00 UTC. Same as Unix time |
There was a problem hiding this comment.
I forget, should there not be a new line after the docstring quotes?
There was a problem hiding this comment.
PEP257 doesn't specify but I can change it if that's a standard recommended in sunpy or in the science stack.
There was a problem hiding this comment.
It's a Cadair standard, which means most of SunPy is like that ;)
sunpy/time/utime.py
Outdated
| from astropy.time import Time | ||
| t = Time('2000-01-01T13:53:23') | ||
| print(t.utime) | ||
| >>> 662738003.0 |
There was a problem hiding this comment.
This looks wrong for example syntax.
I think it should be somrthing like:
Examples
--------
>>> code line 1
>>> code line 2
I forget how you add the output.
There was a problem hiding this comment.
I think you're right. I couldn't find a PEP standard but I found a random example of how it's done in scipy. For example:
>>> from scipy.interpolate import BSpline
>>> k = 2
>>> t = [0, 1, 2, 3, 4, 5, 6]
>>> c = [-1, 2, 0, -1]
>>> spl = BSpline(t, c, k)
>>> spl(2.5)
array(1.375)
>>> bspline(2.5, t, c, k)
1.375
sunpy/time/utime.py
Outdated
| This time format is included for historical reasons. Some | ||
| people in solar physics prefer using this epoch. | ||
|
|
||
| Example: |
There was a problem hiding this comment.
I think example should be underlined
Example
-------
|
|
||
|
|
||
| def test_utime_random_date(): | ||
| assert Time('2018-01-13T13:32:56').utime == 1231853576.0 |
There was a problem hiding this comment.
Could you please add a test which is marked with @pytest.mark.xfail of the reverse conversion? i.e. the open astropy issue?
sunpy/time/tests/test_utime.py
Outdated
| from __future__ import absolute_import, division, print_function | ||
|
|
||
| from astropy.time import Time | ||
|
|
There was a problem hiding this comment.
should this not explicitly import the utime format to register it with astropy?
|
@jmason86 can you merge master into this to fix the CI please? |
|
Can you add a changelog entry for this please? Other than that this looks great. |
[ci skip]
Implemented handling of utime by creating a custom subclass of
astropy.time.Time, which uses the third code example from their documentation as a template.