Python arrays and sting formatting options

[Marc 'BlackJack' Rintsch]

On Tue, 30 Sep 2008 00:04:18 +0200, Ivan Reborin wrote:

> 1. Multi dimensional arrays - how do you load them in python For
> example, if I had:
> -------
> 1 2 3
> 4 5 6
> 7 8 9
> 
> 10 11 12
> 13 14 15
> 16 17 18
> -------
> with "i" being the row number, "j" the column number, and "k" the ..
> uhmm, well, the "group" number, how would you load this ?
> 
> If fortran90 you would just do:
> 
> do 10 k=1,2
> do 20 i=1,3
> 
> read(*,*)(a(i,j,k),j=1,3)
> 
> 20 continue
> 10 continue
> 
> How would the python equivalent go ?

Well, I don't know if this qualifies as equivalent:

=====
from __future__ import with_statement
from functools import partial
from itertools import islice
from pprint import pprint


def read_group(lines, count):
    return [map(int, s.split()) for s in islice(lines, count)]


def main():
    result = list()
    
    with open('test.txt') as lines:
        # 
        # Filter empty lines.
        # 
        lines = (line for line in lines if line.strip())
        # 
        # Read groups until end of file.
        # 
        result = list(iter(partial(read_group, lines, 3), list()))

    pprint(result, width=30)


if __name__ == '__main__':
    main()
=====

The output is:

[[[1, 2, 3],
  [4, 5, 6],
  [7, 8, 9]],
 [[10, 11, 12],
  [13, 14, 15],
  [16, 17, 18]]]

`k` is the first index here, not the last and the code doesn't use fixed 
values for the ranges of `i`, `j`, and `k`, in fact it doesn't use index 
variables at all but simply reads what's in the file.  Only the group 
length is hard coded in the source code.

Ciao,
	Marc 'BlackJack' Rintsch