#!/usr/bin/env python

# License:
# 2006 Lenny Domnitser, all rights waived.

'''
Convert named character references (entities) in HTML to Unicode.
'''

import htmlentitydefs
import re

__all__ = 'convert_entities',

RE_ent = re.compile(r'&(\w+);')


def _ent_match_to_unicode(match):
    name = match.group(1)
    if name not in ('lt', 'gt', 'amp', 'quot', 'apos') and name in htmlentitydefs.name2codepoint:
        return unichr(htmlentitydefs.name2codepoint[name])
    else:
        return '&%s;' % name


def convert_entities(html):
    return RE_ent.sub(_ent_match_to_unicode, html)


def main():
    import optparse
    import sys

    parser = optparse.OptionParser()
    parser.set_description(__doc__.strip())
    parser.add_option('-m', '--modify', dest='modify', action='store_true',
                      help='modify files in place')
    parser.add_option('-e', '--output-encoding', dest='output_encoding', metavar='CHARSET', default='utf-8',
                      help='write out in CHARSET. Default is UTF-8.')
    parser.add_option('--input-encoding', dest='input_encoding', metavar='CHARSET', default='utf-8',
                      help='input is in CHARSET. Default is UTF-8.')
    options, filenames = parser.parse_args()

    def do_it(infile):
        return convert_entities(infile.read().decode(options.input_encoding)).encode(options.output_encoding)

    if not filenames:
        sys.stdout.write(do_it(sys.stdin))
    else:
        for filename in filenames:
            html = do_it(open(filename))
            if options.modify:
                outfile = open(filename, 'w')
            else:
                outfile = sys.stdout
            outfile.write(html)

if __name__ == '__main__':
    main()