Re: 255UInt16 spec

Also, in this clause:

  An encoder SHOULD choose shorter encodings, and MUST be consistent in
choice of encoding for the same value, as this will tend to compress better.

I think the second part (MUST) is excessive.

On Sat, Mar 28, 2015 at 2:55 PM, Behdad Esfahbod <behdad@google.com> wrote:

> Hi all,
>
> I suppose this is the latest draft:
>
>   http://dev.w3.org/webfonts/WOFF2/spec/#DataTypes
>
> Reading the section on 255UInt16 encoding, I found a few issues:
>
> The table is really confusing:
>
> Data TypeSyntaxDescription and CommentsUInt8   Codeif (Code < 253) Value
> = Code; /* [0..252] */if ((Code == 254) || (Code == 255))UInt8   Value1if
> (Code == 255) Value = 253 + Value1; /* [253..508] */
> if (Code == 254) Value = 506 + Value1; /* [506..761] */else if (Code ==
> 253)UInt16   ValueValue; /* [0..65535] */
>
> IMO this is a very clumsy description of the encoding.  What do "Data
> Type" and "Syntax" columns mean in this context?!  I understand the
> encoding from putting everything together, but think the table should be
> rewritten.
>
> In the example following it is all bogus:
>
>   For example, the value 506 can be encoded as [255, 203], [254, 0], and
> [253, 1, 250]
>
> I know it can't be changed now, but why was it designed in a way that 253
> and 506-508 can be represented redundantly using one / two byte sequences?
>
> Thanks,
> behdad
>