Bug in units for NASA Exoplanet Archive results

[bmorris3]

Using astroquery 0.4.7.dev9069, I ran into an incorrect unit in the NASA Exoplanet Archive module today:

>>> from astroquery.ipac.nexsci.nasa_exoplanet_archive import NasaExoplanetArchive as NEA

>>> result = NEA.query_object('K2-141 b')

>>> result['pl_trandur']
[  nan   nan   nan 0.868 0.94 ] d

The pl_trandur column is the transit duration, noted as "T_14" on the rendered webpage, with units of hours (fifth row from the bottom):

I dug into the source and I'm not sure why it doesn't work with the existing logic. I also checked that the Exoplanet Archive table claims units of hours, and it does. Should we put a one-off fix for this column in _fix_units, here?

https://github.com/astropy/astroquery/blob/ab7fdaa48b7197df781671ce34a03527269bef4b/astroquery/ipac/nexsci/nasa_exoplanet_archive/core.py#L534-L535

Related to: #2130, #2218

[bsipocz]

cc @rickynilsson

[rickynilsson]

I'll look into this.

[bmorris3]

Any updates? Thanks!

[rickynilsson]

Any updates? Thanks!

Thanks for the ping. I'll try to get to this when I'm back from eclipse travel later this week.

[bmorris3]

Friendly reminder @rickynilsson.

[rickynilsson]

@bmorris3 - There seems to be an inconsistency in the units for Transit Duration in the Exoplanet Archive, which are in units of hours in the overviews and PS ICE table, but in units of days in the underlying db table (and hence in TAP queries used by astroquery's Exoplanet Archive module). As you say, it should be hours everywhere. I've brought it up with the team and we'll fix the unit in the underlying db as soon as possible. Thanks for bringing this to our attention.

[rmhuch]

Hi! I ran into this exact problem today.. wondering if there is an estimated timeline when this will be fixed? Thanks!

[rickynilsson]

Yeah, it's been fixed in dev and waiting for merge into production next week or so.