1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/lib/__builtins__/unicode.py Mon Dec 12 22:33:15 2016 +0100
1.3 @@ -0,0 +1,60 @@
1.4 +#!/usr/bin/env python
1.5 +
1.6 +"""
1.7 +Unicode objects.
1.8 +
1.9 +Copyright (C) 2015, 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 +from __builtins__.str import basestring
1.26 +from posix.iconv import Converter
1.27 +
1.28 +class utf8string(basestring):
1.29 +
1.30 + "A character string representation based on UTF-8."
1.31 +
1.32 + def encode(self, encoding):
1.33 +
1.34 + "Encode the string to the given 'encoding'."
1.35 +
1.36 + from_utf8 = Converter("UTF-8", encoding)
1.37 + try:
1.38 + from_utf8.feed(self)
1.39 + return str(from_utf8)
1.40 + finally:
1.41 + from_utf8.close()
1.42 +
1.43 +def unicode(s, encoding):
1.44 +
1.45 + "Convert 's' to a Unicode object, interpreting 's' as using 'encoding'."
1.46 +
1.47 + if isinstance(s, utf8string):
1.48 + return s
1.49 +
1.50 + # Obtain a string representation.
1.51 +
1.52 + s = s.__str__()
1.53 +
1.54 + # Convert the string to UTF-8.
1.55 +
1.56 + to_utf8 = Converter(encoding, "UTF-8")
1.57 + try:
1.58 + to_utf8.feed(s)
1.59 + return utf8string(str(to_utf8))
1.60 + finally:
1.61 + to_utf8.close()
1.62 +
1.63 +# vim: tabstop=4 expandtab shiftwidth=4