reading an array with missing data and spaces in the first column

Question

I have a .txt file I want to read using pyhon. The file is an array. It contains data on comets. I copied 3 rows out of the 3000 rows.

P/2011 U1 PANSTARRS               1.54 0.5   14.21 145.294 352.628 6098.07
P/2011 VJ5 Lemmon                 4.12 0.5    2.45 139.978 315.127 5904.20 *
149P/Mueller 4                    3.67 0.1    5.32  85.280  27.963 6064.72

I am reading the array using the the following code:

import numpy as np
list_comet = np.genfromtxt('jfc_master.txt', dtype=None)

I am facing 2 different problems:

First, in row 1 the name of the comet is: P/2011 U1 PANSTARRS. If I type: list_comet[0][1] the result will be P/2011. How should I tell python how to read the name of each comet? Note that the longest name is 31 characters. So what is the command to tell python that column 1 is 31 characters long?

Second, in row 2 that value of the last column is *. When I read the file I am receiving an error which says that:

Line #2941 (got 41 columns instead of 40)

(note that the above data is not the complete data, the total number of columns I have in my original data is 38). I guess I am receiving this error due to the * found in certain rows. How can I fix this problem?

Answer 1

You didn't mention what data structure you're looking for, i.e. what operations you intend to perform on the parsed data. In the simplest case, you could massage the file into a list of 8-tuples - the last element being either '*' or an empty string. That is as simple as

import string

def tokenize(s):
    if s[-1] == '*':
        return string.rsplit(s, None, 7)
    else:
        return string.rsplit(s, None, 6) + ['']

tokens = (tokenize(line.rstrip()) for line in open('so21712204.txt'))

To be fair, this doesn't make tokens a list of 8-tuples but rather a generator (which is more space efficient) of lists, each of which having 8 elements.