Stephen Reindl <[EMAIL PROTECTED]> writes:
| > | Wouldn't this have an enormous space impact on those of us who don't
| > | need it ?
|
| > enormous? no I don't think so.
| > Try to set LyXParagraph::value_type to int and see how large impact is
| > has.
I just did a test... Userguide with char: 6684, with int 82xx, and in
my book this is not enormous.
One thing... an initial runtime size of 4M with no documents loaded seems very
large too me. I guess the only way to reduce this is to use more lazy
loading.
| After spending several days in reading Unicode and ICU
| (oss.software.ibm.com) documentation I think it will will be enough to
| use a 16bit word. The unicode standard
| defines 21 bit to be used for unicode characters. The UTF-16 encoding
| uses special encodings
| to use characters above 0xffff.
Perhaps... but remember that wchar_t is 32bit on most architectures.
| > | I thought the point of UTF-8 was that ASCII could be stored in 7-bit
| > | values ?
|
| > yes, but we do not really want to use utf-8 internally.
|
| Yes, but it will be good enough to be used for the LyX-file format I
| think.
yes.
--
Lgb
