diff options
| author | Walter Dörwald <walter@livinglogic.de> | 2002-06-04 15:16:29 (GMT) |
|---|---|---|
| committer | Walter Dörwald <walter@livinglogic.de> | 2002-06-04 15:16:29 (GMT) |
| commit | 474458da4894b542ff63fb8eaafeaefad7744005 (patch) | |
| tree | 6fdffec2184e0680da735a75281529ec429a7267 /Lib/codecs.py | |
| parent | bc48826dc2fdcf53a837dbf6c35eff4ab69d08bf (diff) | |
| download | cpython-474458da4894b542ff63fb8eaafeaefad7744005.zip cpython-474458da4894b542ff63fb8eaafeaefad7744005.tar.gz cpython-474458da4894b542ff63fb8eaafeaefad7744005.tar.bz2 | |
Add constants BOM_UTF8, BOM_UTF16, BOM_UTF16_LE, BOM_UTF16_BE,
BOM_UTF32, BOM_UTF32_LE and BOM_UTF32_BE that represent the Byte
Order Mark in UTF-8, UTF-16 and UTF-32 encodings for little and
big endian systems.
The old names BOM32_* and BOM64_* were off by a factor of 2.
This closes SF bug http://www.python.org/sf/555360
Diffstat (limited to 'Lib/codecs.py')
| -rw-r--r-- | Lib/codecs.py | 49 |
1 files changed, 32 insertions, 17 deletions
diff --git a/Lib/codecs.py b/Lib/codecs.py index 9178db9..b089e90 100644 --- a/Lib/codecs.py +++ b/Lib/codecs.py @@ -18,29 +18,44 @@ except ImportError, why: 'Failed to load the builtin codecs: %s' % why __all__ = ["register", "lookup", "open", "EncodedFile", "BOM", "BOM_BE", - "BOM_LE", "BOM32_BE", "BOM32_LE", "BOM64_BE", "BOM64_LE"] + "BOM_LE", "BOM32_BE", "BOM32_LE", "BOM64_BE", "BOM64_LE", + "BOM_UTF8", "BOM_UTF16", "BOM_UTF16_LE", "BOM_UTF16_BE", + "BOM_UTF32", "BOM_UTF32_LE", "BOM_UTF32_BE"] ### Constants # -# Byte Order Mark (BOM) and its possible values (BOM_BE, BOM_LE) +# Byte Order Mark (BOM = ZERO WIDTH NO-BREAK SPACE = U+FEFF) +# and its possible byte string values +# for UTF8/UTF16/UTF32 output and little/big endian machines # -BOM = struct.pack('=H', 0xFEFF) -# -BOM_BE = BOM32_BE = '\376\377' -# corresponds to Unicode U+FEFF in UTF-16 on big endian -# platforms == ZERO WIDTH NO-BREAK SPACE -BOM_LE = BOM32_LE = '\377\376' -# corresponds to Unicode U+FFFE in UTF-16 on little endian -# platforms == defined as being an illegal Unicode character -# -# 64-bit Byte Order Marks -# -BOM64_BE = '\000\000\376\377' -# corresponds to Unicode U+0000FEFF in UCS-4 -BOM64_LE = '\377\376\000\000' -# corresponds to Unicode U+0000FFFE in UCS-4 +# UTF-8 +BOM_UTF8 = '\xef\xbb\xbf' + +# UTF-16, little endian +BOM_LE = BOM_UTF16_LE = '\xff\xfe' + +# UTF-16, big endian +BOM_BE = BOM_UTF16_BE = '\xfe\xff' + +# UTF-32, little endian +BOM_UTF32_LE = '\xff\xfe\x00\x00' + +# UTF-32, big endian +BOM_UTF32_BE = '\x00\x00\xfe\xff' + +# UTF-16, native endianness +BOM = BOM_UTF16 = struct.pack('=H', 0xFEFF) + +# UTF-32, native endianness +BOM_UTF32 = struct.pack('=L', 0x0000FEFF) + +# Old broken names (don't use in new code) +BOM32_LE = BOM_UTF16_LE +BOM32_BE = BOM_UTF16_BE +BOM64_LE = BOM_UTF32_LE +BOM64_BE = BOM_UTF32_BE ### Codec base classes (defining the API) |