Line lengths with polyglossia in Cantonese and Japanese ?

Max Chernoff mseven at telus.net
Fri May 31 07:52:39 CEST 2024


Hi Ken,

(unrelated to your question, but hopefully still helpful for the other
parts of your project)

On Fri, 2024-05-31 at 00:54 +0100, zarniwhoop at ntlworld.com wrote:
> I've intermittently been documenting various
> TTF and OTF fonts for use on Linux: how they look, which codepoints
> they contain, and from that which modern languages are covered -
> showing at a minimum the alphabet [...]
>
> To list all the possible codepoints I care about for the various
> languages I use XeLaTeX.  [...] Adding one codepoint at a
> time from yet another font which had patchy overage convinced me
> that XeLaTeX, not LuaLaTeX, was the way to go (much quicker, and
> did not fail if nothing found in the font).

I have two TeX Stack Exchange answers that might be helpful to you, with
code at:

    (1) https://tex.stackexchange.com/a/707031

    (2) https://tex.stackexchange.com/a/715598

and sample output at:

    (1) https://tug.org/~mseven/files/all-characters.pdf

    (2) https://tug.org/~mseven/files/georgian-fonts.pdf

These don't help with the linebreaking at all (they both just typeset a
grid of characters), but they might be helpful for the alphabet and
codepoint coverage portions of the project.

Both answers use LuaTeX (or LuaMetaTeX), but they're pretty quick to run
after the caches are built: (1) loads all 231 Noto fonts and outputs
83 020 distinct characters in under 5 seconds, and (2) loads literally
every single font on the system and outputs ~8 000 characters in under
30 seconds. The first run is quite painful though -- up to 30 minutes
for (2) -- but after that, the caches will be available for all future
runs/documents.

Thanks,
-- Max



More information about the tex-live mailing list.