Re: Query for Validity of Thai Sequence

From: John Hudson (
Date: Fri Feb 16 2007 - 15:01:38 CST

  • Next message: Markus Scherer: "Re: BOCU-1 spec"

    Lokesh Joshi wrote:

    > Here's the Microsoft Thai OT spec. section about invalid combining marks:

    > Therefore, I'm inclined to think that more relaxed checking is in order.
    > This, unfortunately, will require a redesign of the Thai engine. Perhaps
    > this should be done in conjunction w/ the work for Thai and Lao OpenType
    > processing, and perhaps the generic mark checking. (Tickets 2382, 2386,
    > 4740)

    I dob't think that Thai OT spec is sufficiently recent to be used as a reliable reference
    (note document date: February 2002). Most of the OT script-specific specs are acknowledged
    by MS to be obsolete and in need of revision, since they do not necessarily reflect
    current shaping engine behaviour. I know that the Thai engine in Uniscribe was revised for
    the Office 2007 and Vista releases because I was testing it against some older fonts and
    found that some of my older methods for kerning tall vowels off stacks were no longer
    working, requiring updates to the fonts. [Note: I don't blame MS for this; the kerning
    method I had used wasn't the most sensible, it just happened to work with the older
    version of the shaping engine. The new method I came up with to work with the new engine
    is a lot more robust.] I don't know if any of these changes relate specifically to
    definition of invalid sequences.

    John Hudson

    Tiro Typeworks
    Vancouver, BC
    Do not begin to paddle unless you intend always to paddle.
              - St Jean de Brébeuf, instructions for missionaries, 1637

    This archive was generated by hypermail 2.1.5 : Fri Feb 16 2007 - 15:03:19 CST