Updating to package template 1.1.1 #177

bsipocz · 2016-06-16T00:29:22Z

No description provided.

bmorris3 · 2016-06-16T17:15:38Z

astroplan/conftest.py

+    version = 'dev'
+
+packagename = os.path.basename(os.path.dirname(__file__))
+TESTED_VERSIONS[packagename] = version


This line seems to be causing a bunch of test failures.

This is due to unrelated conda issues (yet another one, these past couple of days are quite messed up), simply a too old astropy is picked up.

So I would wait with this until those things in astropy/ci-helpers#98 are ironed out.

bsipocz · 2016-06-21T13:52:00Z

@bmorris3 - Can you restart this travis, please?

bmorris3 · 2016-06-21T16:00:04Z

@bsipocz: Things that are still failing

Sphinx build on py 2.7: ImportError: Numpy version 1.6.0 or later must be installed to use Astropy

bsipocz · 2016-06-21T16:02:20Z

Ouch, OK, I'll put in on the todo.

On 21 June 2016 at 17:00, Brett M. Morris notifications@github.com wrote:

@bsipocz https://github.com/bsipocz: Things that are still failing

Sphinx build on py 2.7: ImportError: Numpy version 1.6.0 or later must be
installed to use Astropy

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#177 (comment),
or mute the thread
https://github.com/notifications/unsubscribe/AGeUwrw3iXTF_daPmUjhhEJXLbWgN2bFks5qOAqFgaJpZM4I26_y
.

bsipocz · 2016-06-21T16:09:53Z

OK, this is now rebased, hopefully it will solve the issues.

bmorris3 · 2016-06-21T17:44:17Z

@bsipocz I think the remaining issues are related to an IERS warning for the remote-data tests, which I came across in #167. I'll make a PR to your branch to fix those (remove the tests).

bsipocz · 2016-06-22T00:15:36Z

@bmorris3 - The fail seems to be related to a new conda issue that just popped up now that the previous ones are solved. I'm on it, but have to admit it's not easy to get fixed while trying to focus on dotastro instead ;)

Skipping doctests that aren't useful, causing warnings on py2.7

skipping is_night doctest

bsipocz · 2016-06-30T17:36:54Z

@bmorris3 - I'm not sure what to do with the remaining two failures.
https://travis-ci.org/astropy/astroplan/jobs/141190508#L747

The first one seems totally out of context (is that download triggered somewhere else, but there is a delay with the output?). For the second one I see the warning locally right before the test header is printed out, but still got the failure later. I guess the warnings are set to be raised only once, thus the test can't catch it?

bmorris3 · 2016-06-30T17:42:23Z

I'm not sure what that's all about. I'm hoping to chat with @eteq to track down what the new IERS behavior is in astropy, which may be the culprit.

eteq · 2016-07-07T18:58:12Z

Hmm, the first of the second (:confused:) error is not exactly due to the IERS machinery - it's because when the tests are run in a clean environment they have to download the latest IERS. That's fine except that the first time that happens it prints the message about the download. and that's confusing pytest. So we just have to convince pytest not to worry about that (probably by having a fixture that runs at the start of all the tests that just makes sure the latest IERS are in place.)

The test_iers_download error is more confusing: I'm guessing that's because the new IERS machinery now replaces what we had in astroplan, so the warning isn't getting raised (which is good). But @bmorris3 and I should indeed probably chat to figure out how best to deal with that.

eteq · 2016-07-07T18:59:00Z

Oh, but I think the first error is a travis issue - I've seen that happen sometimes just because travis is messed up temporarily. I restarted it just to see if it resolves itself now.

kvyh · 2016-07-09T01:51:43Z

I was testing what Time.delta_ut1_utc, and once it gets to the end of the IERS_A table it just continues to use the last delta in the table. To fix the second error, we would either need to have it explicitly reference the IERS table during get_IERS_A_or_workaround()and warn when you leave it, or remove the recwarn.pop from the test_iers_download.

kvyh · 2016-07-13T23:18:27Z

The first error occurs every time with python setup.py test --remote-data, but I don't know what's causing it or how it could be fixed. @eteq any ideas on how it could be fixed? I've sent a PR to the branch being PR'ed here bsipocz#4 that fixes the second error. The fix is running into an error with python2.7 not recognizing that OldEarthOrientationDataWarning is a warning.

bsipocz · 2016-07-14T19:50:13Z

@bmorris3 @eteq - What about putting the remote tests to the allowed failures for now?
This is probably triviality, but the remote tests pass nicely with the lts branch of astropy (1.0.x)

kvyh · 2016-07-14T19:56:25Z

I think allowing the remote tests to fail would be the best option unless we can specifically allow the doctest failure that is caused in sun_rise_time, I'm pretty sure my code should work now to fix the test_iers_download (it is acting like it fixed itself overnight).

bsipocz · 2016-07-14T20:07:06Z

Indeed @kvyh! Thanks. I had two open questions regarding the fix, but can merge it asap, too and move the discussion here instead.

fixed _low_precision to throw the warning when off the table

bsipocz · 2016-07-14T21:56:57Z

Something is getting rather awful here. With the fix of @kvyh the second error goes away, but the first error gets even worse when building against the dev version. To start with the behaviour is different locally, with 1.2 it finds IERS_A in the cache, while with dev it goes into the else part

     if IERS_A_in_cache():
         iers.IERS.iers_table = _get_IERS_A_table()
     else:
         Time._get_delta_ut1_utc = _low_precision_utc_to_ut1

The failure with astropy dev:


/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py:678: UnexpectedException
_________________________________________________________________________ [doctest] astroplan.observer.Observer.sun_set_time _________________________________________________________________________
669         Time at next rise/set of ``target``.
670 
671         Parameters
672         ----------
673         time : `~astropy.time.Time` or other (see below)
674             Time of observation. This will be passed in as the first argument to
675             the `~astropy.time.Time` initializer, so it can be anything that
676             `~astropy.time.Time` will accept (including a `~astropy.time.Time`
677             object)
678 
UNEXPECTED EXCEPTION: AttributeError("'module' object has no attribute 'IERS_Auto'",)
Traceback (most recent call last):

  File "/sw/lib/python3.4/doctest.py", line 1318, in __run
    compileflags, 1), test.globs)

  File "<doctest astroplan.observer.Observer.sun_set_time[4]>", line 1, in <module>

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/utils/decorators.py", line 636, in sun_set_time
    func = make_function_with_signature(func, name=name, **wrapped_args)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/units/decorators.py", line 130, in wrapper
    return wrapped_function(*func_args, **func_kwargs)

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py", line 1148, in sun_set_time
    return self.target_set_time(time, get_sun(time), which, horizon)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/utils/decorators.py", line 636, in target_set_time
    func = make_function_with_signature(func, name=name, **wrapped_args)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/units/decorators.py", line 130, in wrapper
    return wrapped_function(*func_args, **func_kwargs)

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py", line 964, in target_set_time
    horizon=horizon))

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py", line 834, in _determine_which_event
    next_event = function(*args('next'))

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py", line 717, in _calc_riseset
    altaz = self.altaz(times, target)

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py", line 480, in altaz
    return coordinate.transform_to(altaz_frame)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/sky_coordinate.py", line 398, in transform_to
    new_coord = trans(self.frame, generic_frame)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/transformations.py", line 918, in __call__
    curr_coord = t(curr_coord, curr_toframe)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/transformations.py", line 706, in __call__
    res = self.func(fromcoord, toframe)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/builtin_frames/cirs_observed_transforms.py", line 66, in cirs_to_altaz
    distance = locitrs.separation_3d(cirs_coo)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/baseframe.py", line 1048, in separation_3d
    other_in_self_system = other.transform_to(self)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/baseframe.py", line 780, in transform_to
    return trans(self, new_frame)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/transformations.py", line 918, in __call__
    curr_coord = t(curr_coord, curr_toframe)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/transformations.py", line 706, in __call__
    res = self.func(fromcoord, toframe)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/builtin_frames/intermediate_rotation_transforms.py", line 77, in cirs_to_itrs
    pmat = cirs_to_itrs_mat(itrs_frame.obstime)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/builtin_frames/intermediate_rotation_transforms.py", line 36, in cirs_to_itrs_mat
    era = erfa.era00(*get_jd12(time, 'ut1'))

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/coordinates/builtin_frames/utils.py", line 143, in get_jd12
    newtime = getattr(time, scale)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/time/core.py", line 1149, in __getattr__
    tm._set_scale(attr)

  File "/Users/bsipocz/.virtualenvs/astropy-dev/lib/python3.4/site-packages/astropy-1.2.dev14833-py3.4-macosx-10.10-x86_64.egg/astropy/time/core.py", line 426, in _set_scale
    args.append(get_dt(jd1, jd2))

  File "/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/utils.py", line 40, in _low_precision_utc_to_ut1
    if self.mjd*u.day not in iers.IERS_Auto.open()['MJD']:

AttributeError: 'module' object has no attribute 'IERS_Auto'

/private/var/folders/dc/hsm7tqpx2d57n7vb3k1l81xw0000gq/T/astroplan-test-id4b0x7a/lib.macosx-10.10-x86_64-3.4/astroplan/observer.py:678: UnexpectedException

… the cache or download the fresh version

bsipocz · 2016-07-14T22:23:32Z

So it's still not clear why there is a behaviour difference between 1.2 and dev astropy, but adding AttributeError to the exceptions and thus raising a warning is I think an acceptable workaround.

Also using iers.IERS_Auto.open() rather than iers.IERS.open() solves the problem of downloading a newer version of the table right at the beginning, so no more errors for the doctests.

However I would appreciate if someone more familiar with the IERS infrastructure would give the last few commits it a thorough review.

bsipocz · 2016-07-15T11:55:14Z

This is weird, the remote tests now pass locally for py3, but not for py2 while both still fail on travis.
So as a temporary solution I've allowed them to fail on travis, so this can be merged while we figure out the fix for the problems.

Do you want me to rebase and remove the rather experimental last few commits that clearly didn't work as intended?

bsipocz · 2016-07-24T19:50:08Z

@bmorris3 @eteq - This is readyto go.

There are a few workarounds included, but those I think we should live with them until they are fixed either upstream (e.g. astropy/astropy#5194), or independently of this PR in astroplan.

bsipocz · 2016-07-25T13:32:31Z

docs/faq/iers.rst

+    # Workaround until astropy/astropy#5194 is solved
+    try:
+        iers_a = IERS_A.open(download_file(IERS_A_URL, cache=True))
+    except HTTPError:


@taldcroft - I though about using IERS_Auto.open() here, but the sake of this example is to show the difference between A and B, etc, and if A is not available this becomes a bit pointless. However with fresh eyes and brain my solution seems even more hacky than yesterday, and inclined to use the astropy machinery.

@bmorris3 @eteq - what do you think?

@bsipocz - I think the intent is that eventually astroplan will switch to the astropy machinery, right @bmorris3 ? So then there's no need for this at all. Basically this IERS stuff was a stand-in until the machinery in astropy got working.

So with that in mind it doesn't really matter that much one way or another I guess. ;)

OK, good to know. I'll try to come back to this in the weekend to clean up.

wtgee · 2016-07-25T20:50:14Z

Sorry to pester, but if this is passing can we get it merged? Myself and others want to do some work with astroplan but it currently won't install from source or from pip because of the ConfigParser issue, which is a quick fix but has left it broken for a month now. Thanks!

bsipocz · 2016-07-25T21:35:11Z

@wtgee - I don't have commit rights, so we need to wait for @bmorris3 or @eteq, but both of them are traveling as far as I know.

wtgee · 2016-07-26T01:47:30Z

@bsipocz Thanks for letting me know. @bmorris3 @eteq Any ideas when this can happen? This is causing all of our travis-ci builds to fail and is very annoying. If it's not going to happen within a day or two then the ConfigParser items should be merged in a separate pull-request without having to wait for resolution on the IERS issues.

bmorris3 · 2016-07-26T21:11:49Z

Sorry for the delay @bsipocz and @wtgee, I'm back in Seattle and ready to push the big green button! Thanks for all of the help @bsipocz

Updating to package template 1.1.1

bmorris3 reviewed Jun 16, 2016
View reviewed changes

bsipocz added 2 commits June 21, 2016 17:09

Updating to package template 1.1.1

4cccb2c

Updating version numbers both for travis and appveyor

35217a2

bsipocz force-pushed the update_template_1.1.1 branch from e6c1b7c to 35217a2 Compare June 21, 2016 16:09

Skipping doctests that aren't useful, causing errors on py2.7

80e6a7e

bmorris3 mentioned this pull request Jun 21, 2016

Skipping doctests that aren't useful, causing warnings on py2.7 bsipocz/astroplan#2

Merged

bsipocz and others added 2 commits June 22, 2016 23:18

Merge pull request #2 from bmorris3/update_template_1.1.1

a9240f8

Skipping doctests that aren't useful, causing warnings on py2.7

skipping is_night doctest

722c72c

bmorris3 mentioned this pull request Jun 23, 2016

skipping is_night doctest bsipocz/astroplan#3

Merged

Merge pull request #3 from bmorris3/update_template_1.1.1

5412e38

skipping is_night doctest

bsipocz mentioned this pull request Jun 27, 2016

Error with distutils config ConfigParser again astropy/package-template#179

Closed

bsipocz added 3 commits June 28, 2016 17:47

Fix configparser in docs/conf

c3bdd81

Changing to use astropy html theme rather than rtd default

7b34f89

Customizing package logo name

2fd57d7

This was referenced Jul 6, 2016

Python3.5.2 causes setup.py to throw an error #186

Closed

changed the way ConfigParser is imported #187

Closed

fixed _low_precision to throw the warning when off the table

eebe16f

fixed issues that came up when the iers table wasn't set up

4bd3458

bsipocz mentioned this pull request Jul 14, 2016

Constraint broadcasting #165

Merged

Merge pull request #4 from kvyh/template_update

9660037

fixed _low_precision to throw the warning when off the table

bsipocz added 2 commits July 14, 2016 23:15

Adding AttributeError to the exceptions, and cleaning up a bit

1f236b1

Using the new iers.IERS_Auto.open() to get the IERS table either from…

a23f2e6

… the cache or download the fresh version

Allow the remote tests fail for now.

d45a636

bsipocz force-pushed the update_template_1.1.1 branch from 4f7f0b9 to d45a636 Compare July 24, 2016 17:22

bsipocz mentioned this pull request Jul 24, 2016

HTTPError: Forbidden for download_IERS_A() #192

Closed

Adding workaround for iers url issue

a223c39

bsipocz mentioned this pull request Jul 24, 2016

Fix ConfigParser in setup.py #191

Closed

bsipocz reviewed Jul 25, 2016
View reviewed changes

bmorris3 merged commit 2736494 into astropy:master Jul 26, 2016

bmorris3 mentioned this pull request Jul 26, 2016

Terminology #175

Merged

bsipocz mentioned this pull request Jul 27, 2016

Swapping PyEphem moon calculations for astropy's #167

Merged

bsipocz pushed a commit to bsipocz/astroplan that referenced this pull request Jul 29, 2016

Merge pull request astropy#177 from bsipocz/update_template_1.1.1

2d50baf

Updating to package template 1.1.1

bmorris3 mentioned this pull request Aug 23, 2016

Update to configparser for py3 compatibility xastropy/xastropy#110

Merged

bsipocz mentioned this pull request Feb 15, 2017

Coveralls reporting is broken #287

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Updating to package template 1.1.1 #177

Updating to package template 1.1.1 #177

bsipocz commented Jun 16, 2016

bmorris3 Jun 16, 2016

bsipocz Jun 16, 2016

bsipocz commented Jun 21, 2016

bmorris3 commented Jun 21, 2016 •

edited

Loading

bsipocz commented Jun 21, 2016

bsipocz commented Jun 21, 2016

bmorris3 commented Jun 21, 2016 •

edited

Loading

bsipocz commented Jun 22, 2016

bsipocz commented Jun 30, 2016

bmorris3 commented Jun 30, 2016

eteq commented Jul 7, 2016

eteq commented Jul 7, 2016

kvyh commented Jul 9, 2016 •

edited

Loading

kvyh commented Jul 13, 2016 •

edited

Loading

bsipocz commented Jul 14, 2016

kvyh commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 15, 2016

bsipocz commented Jul 24, 2016

bsipocz Jul 25, 2016

eteq Jul 29, 2016

bsipocz Jul 29, 2016

wtgee commented Jul 25, 2016

bsipocz commented Jul 25, 2016

wtgee commented Jul 26, 2016

bmorris3 commented Jul 26, 2016

Updating to package template 1.1.1 #177

Updating to package template 1.1.1 #177

Conversation

bsipocz commented Jun 16, 2016

bmorris3 Jun 16, 2016

Choose a reason for hiding this comment

bsipocz Jun 16, 2016

Choose a reason for hiding this comment

bsipocz commented Jun 21, 2016

bmorris3 commented Jun 21, 2016 • edited Loading

bsipocz commented Jun 21, 2016

bsipocz commented Jun 21, 2016

bmorris3 commented Jun 21, 2016 • edited Loading

bsipocz commented Jun 22, 2016

bsipocz commented Jun 30, 2016

bmorris3 commented Jun 30, 2016

eteq commented Jul 7, 2016

eteq commented Jul 7, 2016

kvyh commented Jul 9, 2016 • edited Loading

kvyh commented Jul 13, 2016 • edited Loading

bsipocz commented Jul 14, 2016

kvyh commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 14, 2016

bsipocz commented Jul 15, 2016

bsipocz commented Jul 24, 2016

bsipocz Jul 25, 2016

Choose a reason for hiding this comment

eteq Jul 29, 2016

Choose a reason for hiding this comment

bsipocz Jul 29, 2016

Choose a reason for hiding this comment

wtgee commented Jul 25, 2016

bsipocz commented Jul 25, 2016

wtgee commented Jul 26, 2016

bmorris3 commented Jul 26, 2016

bmorris3 commented Jun 21, 2016 •

edited

Loading

bmorris3 commented Jun 21, 2016 •

edited

Loading

kvyh commented Jul 9, 2016 •

edited

Loading

kvyh commented Jul 13, 2016 •

edited

Loading