Re: Acquiring UTF-8 string length
- From: Ulrich Eckhardt <eckhardt@xxxxxxxxxxxxxx>
- Date: Mon, 02 Apr 2007 09:39:25 +0200
Coder Guy wrote:
Well as in the example at
http://en.wikipedia.org/wiki/Multi-byte_character_set, this UTF-8 string
actually has three different lengths:
// I{heart}NY
char str[] = { 0x49, 0xE2, 0x99, 0xA5, 0x4E, 0x59, 0x00 };
count of bytes = 7, obtained by sizeof()
Right.
count of code points = 6, obtained by strlen()
Wrong. strlen() only returns the number of chars up to the first NUL char.
The number of codepoints is four, plus the terminating NUL.
count of characters = 4, no API af far as I can tell
Four characters plus a terminating NUL. Note that there are codepoinst that
don't resolve to a character and characters that resolve to more than one
codepoint.
How would I get the number of characters in this string? Or how would
I go about reversing the characters in this string? Do I have to really
implement my own UTF-8 decoder/encoder?
There are libraries out there that help you handle all the various facets of
Unicode. You might want to take a look at ICU, for example.
Another example might be that I am reading a file and it specifies a
code page at the top... arn't there any APIs which will help me manage
by per-character rather than per-code point?
Not that I was aware of.
[snipped fullquote with signature]
Stop this misbehaviour please, it is considered impolite on the Usenet.
Uli
.
- References:
- Re: Acquiring UTF-8 string length
- From: Igor Tandetnik
- Re: Acquiring UTF-8 string length
- From: Coder Guy
- Re: Acquiring UTF-8 string length
- Prev by Date: Re: Acquiring UTF-8 string length
- Next by Date: MSVC hangs at linking
- Previous by thread: Re: Acquiring UTF-8 string length
- Next by thread: Re: Acquiring UTF-8 string length
- Index(es):
Relevant Pages
|
