light_travel_time is broken for Time objects with ndim > 1

Reshaping a Time object to ndim greater than 1 breaks light_travel_time, with the following error:

...
/usr/local/lib/python3.5/site-packages/astropy/coordinates/builtin_frames/icrs_cirs_transforms.py in gcrs_to_icrs(gcrs_coo, icrs_frame)
    171     astrom = erfa.apcs13(jd1, jd2, pv)
    172 
--> 173     i_ra, i_dec = aticq(gcrs_ra, gcrs_dec, astrom)
    174 
    175     if gcrs_coo.data.get_name() == 'unitspherical' or gcrs_coo.data.to_cartesian().x.unit == u.one:

/usr/local/lib/python3.5/site-packages/astropy/coordinates/builtin_frames/utils.py in aticq(ri, di, astrom)
    184 
    185     # Bias-precession-nutation, giving GCRS proper direction.
--> 186     ppr = erfa.trxp(astrom['bpn'], pos)
    187 
    188     # Aberration, giving GCRS natural direction

/usr/local/lib/python3.5/site-packages/astropy/_erfa/core.py in trxp(r, p)
  24006 
  24007     #Create the output array, based on the broadcasted shape, adding the generated dimensions if needed
> 24008     broadcast = numpy.broadcast(numpy.int32(0.0), numpy.int32(0.0), r_in[...,0,0], p_in[...,0])
  24009     trp_out = numpy.empty(broadcast.shape + (3,), dtype=numpy.double)
  24010 

ValueError: shape mismatch: objects cannot be broadcast to a single shape

The issue here is a tricky one, arising because the coordinate transformations are not particularly shape-aware.

So at https://github.com/astropy/astropy/blob/master/astropy/time/core.py#L599 we have

itrs = location.get_itrs(obstime=self)

which for an (N, M) shaped Time gives an ITRS frame with an (N, M) shaped obstime attribute and scalar xyz attributes.

Then at https://github.com/astropy/astropy/blob/master/astropy/time/core.py#L609 we have

gcrs_coo = itrs.transform_to(GCRS(obstime=self))

This goes through the transformation ITRS->CIRS->GCRS. The resulting GCRS frame has (N, M) shaped obstime but the ra and dec properties are shaped (N*M, ) - as does the intermediate CIRS frame. The next transformation from GCRS to ICRS here inevitably fails because of this mismatch.

To solve this issue a large part of the transformation code is going to have to become shape aware, which is a significant code refactor and also should wait until #4123 and #5104 are in, since the refactored code will be much cleaner after code in those PRs has been merged.

Just to say this issue is magically fixed by #5254.

Navigation Menu

Repository navigation

light_travel_time is broken for Time objects with ndim > 1 #5191

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

light_travel_time is broken for Time objects with ndim > 1 #5191

Description

Activity

StuartLittlefair commented on Jul 26, 2016

StuartLittlefair commented on Aug 26, 2016

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions