Hi
An approach that you might like to consider in relation to fonts is that it is
possible to have in a font a Description field that consists of plain text.
It is stored twice in the font, in two different ways, one of which is just
plain text, possibly just ASCII.
So if you had text such as
$$$PUAB
and so on in that Description field than a software application could search
for all occurrences of $$$ and gather information for each set of data in that
way, without needing separate OpenType tables.
As an example of how information can be stored in the Description field here is
a link to a font that I made years ago.
If you download the font and open it is WordPad, the text can be read.
The direct link is as follows.
www.users.globalnet.co.uk/~ngo/SPANGBLU.TTF
The font is also linked from the following web page, about a quarter of the way
down the page.
http://www.users.globalnet.co.uk/~ngo/fonts.htm
The web pages encoded in the font are for three of the songs linked from the
following page.
http://www.users.globalnet.co.uk/~ngo/song0001.htm
Best regards,
William Overington
Friday 24 August 2018
----Original message----
>From : [email protected]
Date : 2018/08/21 - 19:23 (GMTDT)
To : [email protected]
Subject : Re: Private Use areas
On Tue, Aug 21, 2018 at 10:21 AM, Janusz S. Bień via Unicode
<[email protected]> wrote:
I think PUA users should provide the
properties of the characters used in a form analogical to the Unicode
itself, and the software should be able to use this additional
information.
I already provide this myself for my uses of the PUA as well as the CSUR and
any vendor-specific agreements I can find:
http://www.kreativekorp.com/charset/PUADATA/
Of course there is no way to get software to use this information. I have
entertained the idea of being able to embed this information into the font
itself as OpenType tables, e.g.:
PUAB -> Blocks.txt
PUAC -> CaseFolding.txt
PUAW -> EastAsianWidth.txt
PUAL -> LineBreak.txt
PUAD -> UnicodeData.txt
I've actually invented table names for the majority of UCD files, but those are
probably the most relevant. The table names for the more obscure files get
rather... creative, e.g.:
PUA[ -> BidiBrackets.txt
PUA] -> BidiMirroring.txt
That alone may get some people to think twice about this idea. :P
