1
0
mirror of https://github.com/moparisthebest/SickRage synced 2024-11-11 03:45:01 -05:00
SickRage/lib/guessit/country.py

113 lines
3.6 KiB
Python
Raw Normal View History

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
#
# GuessIt - A library for guessing information from filenames
# Copyright (c) 2012 Nicolas Wack <wackou@gmail.com>
#
# GuessIt is free software; you can redistribute it and/or modify it under
# the terms of the Lesser GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# GuessIt is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# Lesser GNU General Public License for more details.
#
# You should have received a copy of the Lesser GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
from __future__ import unicode_literals
from guessit import UnicodeMixin, base_text_type, u
from guessit.fileutils import load_file_in_same_dir
import logging
__all__ = [ 'Country' ]
log = logging.getLogger(__name__)
# parsed from http://en.wikipedia.org/wiki/ISO_3166-1
#
# Description of the fields:
# "An English name, an alpha-2 code (when given),
# an alpha-3 code (when given), a numeric code, and an ISO 31666-2 code
# are all separated by pipe (|) characters."
_iso3166_contents = load_file_in_same_dir(__file__, 'ISO-3166-1_utf8.txt')
country_matrix = [ l.strip().split('|')
for l in _iso3166_contents.strip().split('\n') ]
country_matrix += [ [ 'Unknown', 'un', 'unk', '', '' ],
[ 'Latin America', '', 'lat', '', '' ]
]
country_to_alpha3 = dict((c[0].lower(), c[2].lower()) for c in country_matrix)
country_to_alpha3.update(dict((c[1].lower(), c[2].lower()) for c in country_matrix))
country_to_alpha3.update(dict((c[2].lower(), c[2].lower()) for c in country_matrix))
# add here exceptions / non ISO representations
# Note: remember to put those exceptions in lower-case, they won't work otherwise
country_to_alpha3.update({ 'latinoamérica': 'lat',
'brazilian': 'bra',
'españa': 'esp',
'uk': 'gbr'
})
country_alpha3_to_en_name = dict((c[2].lower(), c[0]) for c in country_matrix)
country_alpha3_to_alpha2 = dict((c[2].lower(), c[1].lower()) for c in country_matrix)
class Country(UnicodeMixin):
"""This class represents a country.
You can initialize it with pretty much anything, as it knows conversion
from ISO-3166 2-letter and 3-letter codes, and an English name.
"""
def __init__(self, country, strict=False):
country = u(country.strip().lower())
self.alpha3 = country_to_alpha3.get(country)
if self.alpha3 is None and strict:
msg = 'The given string "%s" could not be identified as a country'
raise ValueError(msg % country)
if self.alpha3 is None:
self.alpha3 = 'unk'
@property
def alpha2(self):
return country_alpha3_to_alpha2[self.alpha3]
@property
def english_name(self):
return country_alpha3_to_en_name[self.alpha3]
def __hash__(self):
return hash(self.alpha3)
def __eq__(self, other):
if isinstance(other, Country):
return self.alpha3 == other.alpha3
if isinstance(other, base_text_type):
try:
return self == Country(other)
except ValueError:
return False
return False
def __ne__(self, other):
return not self == other
def __unicode__(self):
return self.english_name
def __repr__(self):
return 'Country(%s)' % self.english_name