问题
- Is iso-8859-1 a proper subset of utf-8?
- What about iso-8859-n?
- What about windows-1252?
If the answer is no to any of the above, what are the disjoint characters? I'm testing some logic that detects charsets and want to write tests to verify the detection is working properly.
回答1:
Is iso-8859-1 a proper subset of utf-8?
The character reportoire of ISO-8859-1 (the first 256 characters of Unicode) is a proper subset of that of UTF-8 (every Unicode character).
However, the characters U+0080 to U+00FF are encoded differently in the two encodings.
- ISO-8859-1 assigns each of these characters a single byte from
80toFF. - UTF-8 encodes the same characters as two-byte sequences
C2 80toC3 BF.
What about iso-8859-n?
These are 15 different encodings that contain a total of 614 distinct characters. Some of these characters occur in multiple "parts" of ISO 8859, and some don't. You'll have to be more specific.
I see that your question is tagged ISO-8859-2. The characters that are in -2 that aren't in -1 are:
Ă㥹ĆćČčĎďĐđĘęĚěĹ弾ŁłŃńŇňŐőŔŕŘřŚśŞşŠšŢţŤťŮůŰűŹźŻżŽžˇ˘˙˛˝
What about windows-1252?
Windows-1252 is just like ISO-8859-1 except that it replaces the rarely used control characters in the 0x80-0x9F range with printable characters. The characters that are in windows-1252 but not in ISO-8859-1 are:
ŒœŠšŸŽžƒˆ˜–—‘’‚“”„†‡•…‰‹›€™
回答2:
Unicode is a superset of all these character sets, and of pretty much all established character sets out there. You can find a list of mappings of all these character sets to Unicode code points here: http://unicode.org/Public/MAPPINGS/.
来源：https://stackoverflow.com/questions/10021594/character-set-special-characters