Re: Gnopernicus and ISO-Latin2 characters

[Bill Haneman <Bill Haneman Sun COM>]

Hi Remus:

In UTF-8, which is the default encoding on most of our platforms now, 
there is only "one character set", which is the unicode character set. 
(more below)

remus draica wrote:

> Hi,
>
> Bill, for me one thing is still unclear. Is possible to have more than
> one graphical representation for a character (same code as number)
> depending on the character set? In this case, how this is specified? In
> windows I know that a code page is used for this.
>  
Windows systems can and do use unicode as well.

> My impression was (and still is) that all characters are represented
> from 0 to a huge number and some parts of the interval represents
> characters sets.
>  
Yes, but those 'code pages' do not represent different languages.  They 
map approximately to different "scripts".

> 0......x......y.....z......t.....(something huge)
>       [latin1]    [latin2]
> where latin1 is the interval of chars with codes between x and y.
>
> you said "different Latin characters"? What that means? Same character
> in different languages? For example "a" in english and german? Or same
> character with a different sense depending on something else (a "code
> page"?)?
>  
As Samuel pointed out, by 'different Latin characters' I meant the 
various latin code pages in unicode.  They do not overlap, in the sense 
that characters in Latin-2 are not redundant with Latin-1, and most real 
languages that use the Latin character set include characters from more 
than one Latin code page. 

When dealing with UTF-8 content, there is no ambiguity about what 
character is intended by a particular set of bytes; each UTF-8 encoded 
character refers to a specific unicode 'codepoint' or character.  Thus 
the 'a' in German and the 'a' in English and French refer to the same 
unicode point, usually written as U+00061, and both are encoded in UTF-8 
as a single byte, '0x61'.  The 'a' with dieresis (ä) is in the "Latin-1 
Supplement" section of unicode, written U+00E4, and is encoded in UTF-8 
as two bytes, '0xC3' '0xA4'.

regards,

Bill

> Remus