Can't access hdu from tesscut without saving file

[Nschanche]

Problem description

When creating a Tesscut cutout, you can not read it in directly as a lightkurve object. You must first save out the file and then read it back in. Ideally, you could skip this middle step.

Example

import lightkurve as lk
from astroquery.mast import Tesscut, Observations
from astropy.time import Time
from astropy.coordinates import SkyCoord

grb_time = Time(['2018-10-22T17:31:49.9'], format='isot', scale='utc').jd - 2400000
coordinates = SkyCoord("03h47m10.09s -29d22m46.8s", frame = "icrs") #defaults to ICRS frame
obs = Observations.query_criteria(coordinates=coordinates, obs_collection="TESS", target_name = "TESS FFI", t_min = [0,grb_time[0]])
hdulist = Tesscut.get_cutouts(coordinates=coordinates, size=31, sector=4)

tpf = lk.read(hdulist[0])

---------------------------------------------------------------------------
LightkurveError                           Traceback (most recent call last)
Cell In[8], line 1
----> 1 lk.read(hdulist[0])

File /opt/conda/envs/tess/lib/python3.8/site-packages/lightkurve/io/read.py:131, in read(path_or_url, **kwargs)
    126         raise LightkurveError(
    127             f"Data product f{path_or_url} of type {filetype} is not supported " "in this version of Lightkurve."
    128         ) from exc
    129 else:
    130     # if these keywords don't exist, raise `ValueError`
--> 131     raise LightkurveError(
    132         "Not recognized as a supported data product:\n"
    133         f"{path_or_url}\n"
    134         "This file may be corrupt due to an interrupted download. "
    135         "Please remove it from your disk and try again."
    136     )

LightkurveError: Not recognized as a supported data product:
[<astropy.io.fits.hdu.image.PrimaryHDU object at 0x7f98eff0bdc0>, <astropy.io.fits.hdu.table.BinTableHDU object at 0x7f98f17219a0>, <astropy.io.fits.hdu.image.ImageHDU object at 0x7f98f1740ee0>]
This file may be corrupt due to an interrupted download. Please remove it from your disk and try again.

Expected behavior

To read it in, you have to first save it as follows:

hdulist[0].writeto('hdutest.fits')
tpf = lk.read('hdutest.fits')

Ideally, I would like to skip the middle step. The reason appears to be because lightkurve assumes that the hdu object has a 'filetype' keyword that is set upon saving. Perhaps there is a way to either adjust or remove this requirement.

[christinahedges]

Hi @Nschanche I think this is just a case of not loading the object in the right way, because you're trying to pass an HDUList which is loaded into memory, you should pass it directly to the TessTargetPixelFile class, not try to pass it to the reader.

There's a little bug in the detect.py file, I'll fix it!

import lightkurve as lk
from astroquery.mast import Tesscut, Observations
from astropy.time import Time
from astropy.coordinates import SkyCoord

grb_time = Time(['2018-10-22T17:31:49.9'], format='isot', scale='utc').jd - 2400000
coordinates = SkyCoord("03h47m10.09s -29d22m46.8s", frame = "icrs") #defaults to ICRS frame
obs = Observations.query_criteria(coordinates=coordinates,
                                  obs_collection="TESS",
                                  target_name="TESS FFI",
                                  t_min=[0, grb_time[0]])
hdulists = Tesscut.get_cutouts(coordinates=coordinates, size=3, sector=4)
tpf = lk.TessTargetPixelFile(hdulists[0])