Re: GSFromUnicode/GSToUnicode buffers [Was: [bug #4208]Applications won't start]

[Richard Frith-Macdonald]

On Tuesday, July 8, 2003, at 03:31 AM, Alexander Malmberg wrote:

> When looking at this bug, I noticed in the GSFromUnicode and 
> GSToUnicode
> documentation:
>
> (input to GSToUnicode)
> ""
>  * <p>The slen argument is the length (bytes) of the 8-bit character
> string
>  * which is to be converted to 16-bit unicode.
>  * This is measured in characters, not bytes.
> ""
>
> and (output from GSFromUnicode)
>
> ""
>  * <p>The size argument is a pointer to the initial size of the
> destination
>  * buffer.  If the function changes the buffer size, this value will be
>  * altered to the new size.  This is measured in characters, not bytes.
> ""
>
> which makes no sense to me. It seems that both of these should be in
> bytes.

I think that what was meant was that these counts are supposed to be in 
bytes for non-unicode date, and unichars for unicode data as 
appropriate ...

ie. lengths of ucs2 strings should be the number of 16-bit unichars in 
the strings, but lengths of
strings in other encodings should be the number of 8-bit bytes in the 
strings (which is the same
thing as the number of characters for most encodings).

I haven't checked what the code actually does though.