TST: Mark coordinates tests as requiring remote data#8033
TST: Mark coordinates tests as requiring remote data#8033pllim merged 2 commits intoastropy:masterfrom
Conversation
|
Check results are now reported in the status checks at the bottom of this page. |
Codecov Report
@@ Coverage Diff @@
## master #8033 +/- ##
=========================================
- Coverage 86.92% 86.73% -0.2%
=========================================
Files 384 384
Lines 57995 57995
Branches 1060 1060
=========================================
- Hits 50413 50302 -111
- Misses 6967 7078 +111
Partials 615 615
Continue to review full report at Codecov.
|
|
This is not release critical, re-milestoning. |
|
@pllim - can you rebase? I need to do a new release soonish and can't see a reason why this should be kept open rather than merged in. |
11b3bb2 to
7669ca2
Compare
|
UPDATE: I reran the tests with and without this patch and confirmed that this is still necessary to silence those warnings. In fact, I caught one more (probably it was added since I originally ran this). I think this is good to merge if you don't see anything wrong with it. |
| altazframe = AltAz(location=EarthLocation(lat=0*u.deg, lon=0*u.deg, height=0*u.m), | ||
| obstime=Time('J2000')) | ||
| return fullstack_icrs.transform_to(altazframe) | ||
| with warnings.catch_warnings(): # Ignore remote_data warning |
There was a problem hiding this comment.
My fixture-fu is not great, so if there is a better way to do this, please let me know.
|
|
||
| @pytest.mark.xfail | ||
| @pytest.mark.remote_data | ||
| @pytest.mark.xfail(reason="Current output is completely incorrect") |
|
p.s. I am not convinced that this needs to be backported all the way to 2.x. Please remilestone to something less painful if straight backport is not possible. |
|
Looks like this is now needed because of IERS being down and is holding up other PRs because the test is running when it is not supposed to without this patch. Merging. |
TST: Mark coordinates tests as requiring remote data
TST: Mark coordinates tests as requiring remote data
As part of #7928 but more controversial than #7989. This one is to address the following warnings I get when I run
python setup.py test -P coordinates(without--remote-data).I looked at one of the tests and traced that one to such a call that invokes IERS, which in turns invokes Internet connection.
I am not sure if this is a proper way to address such warnings but at least it is how I could get rid of them with minimal change. With this patch, here is a log of
python setup.py test -P coordinatesshowing that the tests are now skipped without--remote-data:ztstlog.txt