62
62
return _utf8_to_uni[utf8_str]
64
_utf8_to_uni[utf8_str] = unicode_str = _utf8_decode(utf8_str)[0]
64
unicode_str = _utf8_decode(utf8_str)[0]
65
_utf8_to_uni[utf8_str] = unicode_str
65
66
_uni_to_utf8[unicode_str] = utf8_str
78
79
return decode(encode(unicode_str))
82
def get_cached_utf8(utf8_str):
83
"""Return a cached version of the utf-8 string.
85
Get a cached version of this string (similar to intern()).
86
At present, this will be decoded to ensure it is a utf-8 string. In the
87
future this might change to simply caching the string.
89
return encode(decode(utf8_str))
92
def get_cached_ascii(ascii_str,
93
_uni_to_utf8=_unicode_to_utf8_map,
94
_utf8_to_uni=_utf8_to_unicode_map):
95
"""This is a string which is identical in utf-8 and unicode."""
96
# We don't need to do any encoding, but we want _utf8_to_uni to return a
97
# real Unicode string. Unicode and plain strings of this type will have the
98
# same hash, so we can just use it as the key in _uni_to_utf8, but we need
99
# the return value to be different in _utf8_to_uni
100
ascii_str = _uni_to_utf8.setdefault(ascii_str, ascii_str)
101
_utf8_to_uni.setdefault(ascii_str, unicode(ascii_str))
81
105
def clear_encoding_cache():
82
106
"""Clear the encoding and decoding caches"""
83
107
_unicode_to_utf8_map.clear()
added
removed
bzrlib/cache_utf8.py
