mirror of
https://github.com/rembo10/headphones.git
synced 2026-03-21 20:29:27 +00:00
55 lines
1.5 KiB
Python
55 lines
1.5 KiB
Python
# -*- coding: utf-8 -*-
|
|
"""Transliterate Unicode text into plain 7-bit ASCII.
|
|
|
|
Example usage:
|
|
>>> from unidecode import unidecode:
|
|
>>> unidecode(u"\u5317\u4EB0")
|
|
"Bei Jing "
|
|
|
|
The transliteration uses a straightforward map, and doesn't have alternatives
|
|
for the same character based on language, position, or anything else.
|
|
|
|
In Python 3, a standard string object will be returned. If you need bytes, use:
|
|
>>> unidecode("Κνωσός").encode("ascii")
|
|
b'Knosos'
|
|
"""
|
|
Cache = {}
|
|
|
|
def unidecode(string):
|
|
"""Transliterate an Unicode object into an ASCII string
|
|
|
|
>>> unidecode(u"\u5317\u4EB0")
|
|
"Bei Jing "
|
|
"""
|
|
|
|
retval = []
|
|
|
|
for char in string:
|
|
codepoint = ord(char)
|
|
|
|
if codepoint < 0x80: # Basic ASCII
|
|
retval.append(char)
|
|
continue
|
|
|
|
if codepoint > 0xeffff:
|
|
continue # Characters in Private Use Area and above are ignored
|
|
|
|
section = codepoint >> 8 # Chop off the last two hex digits
|
|
position = codepoint % 256 # Last two hex digits
|
|
|
|
try:
|
|
table = Cache[section]
|
|
except KeyError:
|
|
try:
|
|
mod = __import__('lib.unidecode.x%03x'%(section), [], [], ['data'])
|
|
except ImportError:
|
|
Cache[section] = None
|
|
continue # No match: ignore this character and carry on.
|
|
|
|
Cache[section] = table = mod.data
|
|
|
|
if table and len(table) > position:
|
|
retval.append( table[position] )
|
|
|
|
return ''.join(retval)
|