lantz.stringparser

A stand alone module used by lantz. (website)

Motivation

The stringparser module provides a simple way to match patterns and extract information within strings. As patterns are given using the familiar format string specification PEP 3101, writing them is much easier than writing regular expressions (albeit less powerful).

Examples

You can build a reusable parser object:

>>> parser = Parser('The answer is {:d}')
>>> parser('The answer is 42')
42
>>> parser('The answer is 54')
54

Or directly:

>>> Parser('The answer is {:d}')('The answer is 42')
42

You can retrieve many fields:

>>> Parser('The {:s} is {:d}')('The answer is 42')
('answer', 42)

And you can use numbered fields to order the returned tuple:

>>> Parser('The {1:s} is {0:d}')('The answer is 42')
(42, 'answer')

Or named fields to return an OrderedDict:

>>> Parser('The {a:s} is {b:d}')('The answer is 42')
OrderedDict([('a', 'answer'), ('b', 42)])

You can ignore some fields using _ as a name:

>>> Parser('The {_:s} is {:d}')('The answer is 42')
42

Limitations

  • From the format string: [[fill]align][sign][#][0][minimumwidth][.precision][type] only type, sign and # are currently implemented. This might cause trouble to match certain notation like:
    • decimal: ‘-4’ written as ‘- 4’
    • etc
  • Lines are matched from beginning to end. {:d} will NOT return all the numbers in the string. Use regex for that.