Re: WOFF 1.0 clarifications

Thank you for posting the 27 July 2010 updated draft which includes a 
number of excellent improvements.

The second concern I had (exact definition of WOFFHeader.totalSfntSize) 
has been clarified.

The first concern, however, has not. Moreover I have realized (after 
sending my original email) that the concern is broader than I first 
indicated. Specifically, the broader concern is that the spec does not 
explicitly say that all length and offset values are in units of bytes. 
While many implementers will likely assume such, the required padding of 
WOFF font tables would permit length and offset values to be in units of 
16- or even 32-bit words, so it would be good to clarify the matter.

NB: Some of the Conformance Requirements in Section 8 indirectly imply 
that certain length or offset values are in units of bytes, but this 
does not cover all the length and offset values defined by the spec.

Regards,

Bob

On 2010-07-29  at 2:41  Levantovsky, Vladimir wrote:
>
> Hello Bob, all,
>
> Thank you very much for reviewing the WOFF specification and submitting your comments - they are invaluable part of the WOFF spec development as a W3C Recommendation.
>
> I'd like to announce that the first public working draft of the WOFF spec (containing all the changes and revisions to date) has just been published by W3C at http://www.w3.org/TR/WOFF/. The WebFonts WG would very much appreciate the comments from www-font list participants.
>
> Thank you,
> Vladimir
>
>
>    
>> -----Original Message-----
>> From: www-font-request@w3.org [mailto:www-font-request@w3.org] On
>> Behalf Of Bob Hallissy
>> Sent: Wednesday, July 28, 2010 7:19 PM
>> To: www-font@w3.org
>> Subject: WOFF 1.0 clarifications
>>
>> Reading WOFF File Format 1.0
>> (http://www.w3.org/Submission/2010/SUBM-WOFF-20100408/), a couple of
>> questions:
>>
>> 1) There are several places where a "length" item is specified (e.g.,
>> WOFFHeader.length, TableDirectoryEntry.compLength,
>> TableDirectoryEntry.origLength) but no where (that I can find) is it
>> said what the units are. I will guess, and most implementors would
>> probably assume, bytes -- but it would be good if the spec were
>> explicit.
>>
>> 2) WOFFHeader.totalSfntSize is documented as "Total size needed for the
>> uncompressed font data, including the sfnt header, directory, and
>> tables."  Is this number to include the table-ending padding or is it
>> the sum of the table sizes (plus header and directory sizes) ?  Again,
>> I
>> would probably get it right by guessing, but it would be good to change
>> "...header, directory, and tables." to something like  "...header,
>> directory, and tables (as padded)." if that is what is intended.
>>
>> Regards,
>>
>> Bob Hallissy
>> SIL International
>>
>>      
>

Received on Friday, 30 July 2010 21:27:16 UTC