20
20
"""Some functions to enable caching the conversion between unicode to utf8"""
25
_utf8_encode = codecs.getencoder("utf-8")
26
_utf8_decode = codecs.getdecoder("utf-8")
27
def _utf8_decode_with_None(bytestring, _utf8_decode=_utf8_decode):
28
"""wrap _utf8_decode to support None->None for optional strings.
30
Also, only return the Unicode portion, since we don't care about the second
33
if bytestring is None:
36
return _utf8_decode(bytestring)[0]
38
23
# Map revisions from and to utf8 encoding
39
24
# Whenever we do an encode/decode operation, we save the result, so that
45
30
def encode(unicode_str,
46
31
_uni_to_utf8=_unicode_to_utf8_map,
47
_utf8_to_uni=_utf8_to_unicode_map,
48
_utf8_encode=_utf8_encode):
32
_utf8_to_uni=_utf8_to_unicode_map):
49
33
"""Take this unicode revision id, and get a unicode version"""
50
34
# If the key is in the cache try/KeyError is 50% faster than
51
35
# val = dict.get(key), if val is None:
59
43
return _uni_to_utf8[unicode_str]
61
_uni_to_utf8[unicode_str] = utf8_str = _utf8_encode(unicode_str)[0]
45
_uni_to_utf8[unicode_str] = utf8_str = unicode_str.encode('utf-8')
62
46
_utf8_to_uni[utf8_str] = unicode_str
66
50
def decode(utf8_str,
67
51
_uni_to_utf8=_unicode_to_utf8_map,
68
_utf8_to_uni=_utf8_to_unicode_map,
69
_utf8_decode=_utf8_decode):
52
_utf8_to_uni=_utf8_to_unicode_map):
70
53
"""Take a utf8 revision id, and decode it, but cache the result"""
72
55
return _utf8_to_uni[utf8_str]
74
unicode_str = _utf8_decode(utf8_str)[0]
75
_utf8_to_uni[utf8_str] = unicode_str
57
_utf8_to_uni[utf8_str] = unicode_str = utf8_str.decode('utf-8')
76
58
_uni_to_utf8[unicode_str] = utf8_str
89
71
return decode(encode(unicode_str))
92
def get_cached_utf8(utf8_str):
93
"""Return a cached version of the utf-8 string.
95
Get a cached version of this string (similar to intern()).
96
At present, this will be decoded to ensure it is a utf-8 string. In the
97
future this might change to simply caching the string.
99
return encode(decode(utf8_str))
102
def get_cached_ascii(ascii_str,
103
_uni_to_utf8=_unicode_to_utf8_map,
104
_utf8_to_uni=_utf8_to_unicode_map):
105
"""This is a string which is identical in utf-8 and unicode."""
106
# We don't need to do any encoding, but we want _utf8_to_uni to return a
107
# real Unicode string. Unicode and plain strings of this type will have the
108
# same hash, so we can just use it as the key in _uni_to_utf8, but we need
109
# the return value to be different in _utf8_to_uni
110
ascii_str = _uni_to_utf8.setdefault(ascii_str, ascii_str)
111
_utf8_to_uni.setdefault(ascii_str, unicode(ascii_str))
115
74
def clear_encoding_cache():
116
75
"""Clear the encoding and decoding caches"""
76
global _unicode_to_utf8_map, _utf8_to_unicode_map
117
77
_unicode_to_utf8_map.clear()
118
78
_utf8_to_unicode_map.clear()
added
removed
bzrlib/cache_utf8.py
