cadccutout 0.1.0

Cutout library in Python

Cutout library written in Python that uses Astropy APIs.

API

Cutouts are performed unidirectionally, meaning the library assumes an input stream that can only be read once, rather than seeking. If more than one HDU was requested, then each HDU is iterated over and compared from top to bottom. Single HDU requests are shortcircuited using the Astropy astropy.io.fits.getdata() function

Python 3.x has a potential issue when appending to an output stream where Astropy incorrectly sets a mode to prevent appending (See Astropy GitHub 7856).

Example 1

Perform a cutout of a file using the cfitsio cutout string format.

import tempfile
from cadccutout import OpenCADCCutout

test_subject = OpenCADCCutout()
output_file = tempfile.mkstemp(suffix='.fits')
input_file = '/path/to/file.fits'

# Cutouts are in cfitsio format.
cutout_region_string = '[300:800,810:1000]'  # HDU 0 along two axes.

# Needs to have 'append' flag set.  The cutout() method will write out the data.
with open(output_file, 'ab+') as output_writer, open(input_file, 'rb') as input_reader:
    test_subject.cutout(input_reader, output_writer, cutout_region_string, 'FITS')

Example 2 (CADC)

Perform a cutout from an input stream from an HTTP request.

import tempfile
from cadccutout import OpenCADCCutout
from cadcdata import CadcDataClient

test_subject = OpenCADCCutout()
anonSubject = net.Subject()
data_client = CadcDataClient(anonSubject)
output_file = tempfile.mkstemp(suffix='.fits')
archive = 'HST'
file_name = 'n8i311hiq_raw.fits'
input_stream = data_client.get_file(archive, file_name)

# Cutouts are in cfitsio format.
cutout_region_string = '[SCI,10][80:220,100:150]'  # SCI version 10, along two axes.

# Needs to have 'append' flag set.  The cutout() method will write out the data.
with open(output_file, 'ab+') as output_writer:
    test_subject.cutout(input_stream, output_writer, cutout_region_string, 'FITS')

Testing

Docker

The easiest thing to do is to run it with docker. OpenCADC has an AstroQuery docker image available for runtime available in Python 2.7, 3.5, 3.6, and 3.7.

Run tests in Docker

You can mount the local dev directory to the image and run the python test that way. From inside the dev (working) directory:

docker run --rm -v $(pwd):/usr/src/app opencadc/astroquery:3.7-alpine python setup.py test

docker run --rm -v $(pwd):/usr/src/app opencadc/astroquery:2.7-alpine python setup.py test