The eph package provides some useful functions, classes and tools to retrieve, parse and manipulate ephemerides in an astropy-compatible way.

See eph-howto (jupyter notebook) for more info.

import eph

req = eph.JplReq() # create the request
req.read('eph.ini', section='jplparams') # read parameters from 'jplparams' section in 'eph.ini'
req.set({
    'COMMAND': 'venus',
    'START_TIME': '2007-11-17',
    'STOP_TIME': '2017-4-22'
    'STEP_SIZE': '10d'
    }) # set parameters from dictionary
req['OBJ_DATA'] = False # set parameter dict-like
req.csv = True # set parameter as attributes
req.set(
    TABLE_TYPE='V',
    VEC_LABELS=False,
    VEC_TABLE=1
) # set position vectors output

res = req.query() # perform the request obtaining a response from Jpl Horizons service
e = res.parse() # parse the ephemeris in an astropy QTable

from astropy.io import ascii

ascii.write(e, format='csv') # write output data

The content of eph.ini can be something like this (see ftp://ssd.jpl_process.nasa.gov/pub/ssd/horizons_batch_example.long for a complete description of JPL parameters)

[jplparams]
MAKE_EPHEM=YES
REFERENCE_PLANE=ECLIPTIC
REF_SYSTEM=J2000
OUT_UNITS=AU-D

eph package defines also some useful shortcut functions to easily access Jpl Horizons data. Instead of building a JplReq and get back a JplRes to parse, you can get an astropy QTable with

from eph import *

e = get('2000-1-1', '2018-1-1', 299, step=100).

This returns by default an observer table. If you want vectors, type

e = vec('2000-1-1', '2018-1-1', 299, step=100).

eph package also provides a command line tool:

$ eph 2007-11-17 2017-4-22 venus

This command gives you an ephemeris table of Venus starting from 2007-11-17 to 2017-4-22. You can also change the reference frame, the time-step size, the output etc.. through the options provided. Check available options typing

$ eph --help