RE: Non-zero loca transformLength from Levantovsky, Vladimir on 2015-03-30 (public-webfonts-wg@w3.org from March 2015)

From: Levantovsky, Vladimir <Vladimir.Levantovsky@monotype.com>
Date: Mon, 30 Mar 2015 15:46:07 +0000
To: Jonathan Kew <jfkthame@gmail.com>, Khaled Hosny <khaledhosny@eglug.org>, WebFonts WG <public-webfonts-wg@w3.org>
Message-ID: <f75d231b01af401494f44bcdda412f64@wob-maildb-03.agfamonotype.org>

Agree. 
I added a new "conform-mustRejectLoca" conformance case for UA when either the transformLength has a non-zero value or if the origLength size doesn’t match the calculated  size (based on numGlyphs and indexFormat values).

Thank you,
Vlad

-----Original Message-----
From: Jonathan Kew [mailto:jfkthame@gmail.com] 
Sent: Sunday, March 29, 2015 5:19 PM
To: Khaled Hosny; WebFonts WG
Subject: Re: Non-zero loca transformLength

On 29/3/15 21:56, Khaled Hosny wrote:
> Hi,
>
>>From [1], what User Agent should do if the value is not zero, reject the
> font?

In my opinion, yes: it's invalid, and must not be used.

If the WG agrees, we should make this explicit in the spec.

JK

>
> Regards,
> Khaled
>
> 1. http://dev.w3.org/webfonts/WOFF2/spec/#conform-transformedLocaMustBeZero:

>
>

Received on Monday, 30 March 2015 15:47:19 UTC