1.1 --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
     1.2 +++ b/imiptools/text.py	Sun Jan 31 00:45:26 2016 +0100
     1.3 @@ -0,0 +1,56 @@
     1.4 +#!/usr/bin/env python
     1.5 +
     1.6 +"""
     1.7 +Parsing of textual content.
     1.8 +
     1.9 +Copyright (C) 2016 Paul Boddie <paul@boddie.org.uk>
    1.10 +
    1.11 +This program is free software; you can redistribute it and/or modify it under
    1.12 +the terms of the GNU General Public License as published by the Free Software
    1.13 +Foundation; either version 3 of the License, or (at your option) any later
    1.14 +version.
    1.15 +
    1.16 +This program is distributed in the hope that it will be useful, but WITHOUT
    1.17 +ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
    1.18 +FOR A PARTICULAR PURPOSE.  See the GNU General Public License for more
    1.19 +details.
    1.20 +
    1.21 +You should have received a copy of the GNU General Public License along with
    1.22 +this program.  If not, see <http://www.gnu.org/licenses/>.
    1.23 +"""
    1.24 +
    1.25 +import re
    1.26 +
    1.27 +# Parsing of lines to obtain functions and arguments.
    1.28 +
    1.29 +line_pattern_str = r"(?:" \
    1.30 +                   r"(?:'(.*?)')" \
    1.31 +                   r"|" \
    1.32 +                   r'(?:"(.*?)")' \
    1.33 +                   r"|" \
    1.34 +                   r"([^\s]+)" \
    1.35 +                   r")+" \
    1.36 +                   r"(?:\s+|$)"
    1.37 +line_pattern = re.compile(line_pattern_str)
    1.38 +
    1.39 +def parse_line(text):
    1.40 +
    1.41 +    """
    1.42 +    Parse the given 'text', returning a list of words separated by whitespace in
    1.43 +    the input, where whitespace may occur inside words if quoted using single or
    1.44 +    double quotes.
    1.45 +    """
    1.46 +
    1.47 +    parts = []
    1.48 +
    1.49 +    # Match the components of each part.
    1.50 +
    1.51 +    for match in line_pattern.finditer(text):
    1.52 +
    1.53 +        # Combine the components by traversing the matching groups.
    1.54 +
    1.55 +        parts.append(reduce(lambda a, b: (a or "") + (b or ""), match.groups()))
    1.56 +
    1.57 +    return parts
    1.58 +
    1.59 +# vim: tabstop=4 expandtab shiftwidth=4