• Achilleas Margaritis <Achilleas_member pathlink.com> Apr 22 2004
• Ben Hinkle <bhinkle4 juno.com> Apr 22 2004
• "Scott Egan" <scotte tpg.com.aux> Apr 22 2004

Achilleas Margaritis <Achilleas_member pathlink.com> writes:

The unicode standard says that UTF8 and UTF16 characters vary in size. How D
handles this ? is it assumed that UTF8 chars are always 8-bits and UTF16 chars
are always 16 bits ?

Ben Hinkle <bhinkle4 juno.com> writes:

On Thu, 22 Apr 2004 11:57:31 +0000 (UTC), Achilleas Margaritis
<Achilleas_member pathlink.com> wrote:

The unicode standard says that UTF8 and UTF16 characters vary in size. How D
handles this ? is it assumed that UTF8 chars are always 8-bits and UTF16 chars
are always 16 bits ?


In std.utf 
http://www.digitalmars.com/d/phobos.html#utf
there are functions like
 dchar decode(char[] s, inout uint idx)
that take a UTF8 char[] and an index and return the UTF32 codepoint
and advances the index by one or more bytes. The regular array
indexing [] doesn't know about multi-slot characters.

-Ben

"Scott Egan" <scotte tpg.com.aux> writes:

It doesn't although they are called UTF-8 and UTF-16 they are just arrays of
appropriate lengh chars.

The O/S is what really has to deal with them as Unicode.

This means of course that using indexes against the char[] and mucking
aroung with the data you may end up with invalid unicode.

telle est la vie


"Achilleas Margaritis" <Achilleas_member pathlink.com> wrote in message
news:c68bvb$1vgk$1 digitaldaemon.com...
 The unicode standard says that UTF8 and UTF16 characters vary in size. How


 handles this ? is it assumed that UTF8 chars are always 8-bits and UTF16


 are always 16 bits ?