Re: NamesList.txt as data source

On 3/11/2016 9:37 AM, Oren Watson wrote:
> Ok, so let me see if I understand this correctly. Suppose I'm writing
> a editor for math equations, and I want the user to be able to press a
> "Doublestruck" button and then type an C or D to get a ℂ or 𝔻
> respectively. There is apparently no official source containing a
> machine-readable table of the doublestruck equivalents of each
> character that has such an equivalent. Such a table might also include
> { -> ⦃ and such.
>
> This seems like something that would be very convenient to have
> centralized and standardized.
>

O.k., it is taking more time to talk about this than to just make the lists.
See attached list, which took about 5 minutes to cull.

That lists the 24 "unifications" mentioned on page 7 of UTR #25, Unicode
Support for Mathematics:

http://www.unicode.org/reports/tr25/

It matches the 24 explicit cross-references listed in the Unicode names
list.

If the ability to pull out such a list and make it "machine-readable" in
a few minutes
doesn't suffice, and you need something which counts as a more "official
source", then the best way forward would be to engage with the UTC during
the next update cycle for UTR #25, when its associated data table needs
to be checked for the 9.0 repertoire additions, and advocate that some
further documentation
be made explicitly for those 24 mappings.

BTW, all 24 *are* already present in MathClassEx-14.txt:

http://www.unicode.org/Public/math/revision-14/MathClassEx-14.txt

as commented-out entry lines. So an even faster way to get a centralized
(if not "official") list, is to take MathClassEx-14.txt and

% grep #1D MathClassEx-14.txt | grep reserved > maplistout.txt

See also attached.

As for starting down the road of suggesting additional equivalences,
e.g. for
double-struck parentheses, that is certainly something somebody could do,
and might be interesting content to add to UTR #25 -- but it goes beyond the
formal unification issue for the 24 mathematical alphabet letters already
encoded in the 2100 block.

--Ken