Setting MJDParameter from Time

[paulray]

As I was writing docs, I found this:

PINT/src/pint/models/parameter.py

Line 852 in 30d916e

result = val

When you set an MJDParameter to an astropy.Time that is in a scale different than par.time_scale, it is not converted. I think probably it should be.

par.quantity = Time(58000.0,format='mjd',scale='tdb')
print("Quantity       ", par.quantity, type(par.quantity))
print("Value          ", par.value)
print(par)
par.quantity 

returns

Quantity        58000.0 <class 'astropy.time.core.Time'>
Value           58000.0
MJDParameter(   TZRMJD          58000.0000000000000000 (d) frozen=True)
<Time object: scale='tdb' format='mjd' value=58000.0>

So I think probably the line should be

    result = Time(val,scale=self.time_scale)

But this probably needs some discussion.

One question is how the time_scale is handled for TZRMJD. For most TZRSITE values, TZRMJD should be a pulsar_mjd in UTC, but fro TZRSITE at the barycenter, TZRMJD should be in the TDB timescale. Is this done correctly currently?

[luojing1211]

I agree with you @paulray, it should convert to the right scale. However, the location in the timing scale is a tricky part. Should we allow MJDparameter has a location?

[paulray]

Yeah, this is a bit tricky!

Nearly all MJD parameters are in TDB and should not have a location (e.g. T0, PEPOCH, TASC, etc...)

I think only TZRMJD is something that might want a location. However, its location is really specified by TZRSITE, so I think the TZRMJD parameter should not have a location associated with its Time object. Instead, when a TOA is constructed from TZRMJD, TZRSITE, TZRFRQ, it should get the location from TZRSITE.

[luojing1211]

Maybe in the future, astropy can allow a location at pulsar barycenter. and do the conversion for us. You know what, that is called astropy.pint module.