[m-dev.] getting type representation decisions from .int files

[Julien Fischer]

Hi Peter,

On Mon, 5 Jul 2021, Peter Wang wrote:

>> >> + at c This should not be here. It should be replaced by a statement at the top
>> >> + at c that says: these cases list *all* the situations in which S =< T may hold
>> >> + at c without S = T. In every other situation, S =< T if and only if S = T."
>> >>  @samp{P =< Q} holds if @var{P} and @var{Q} are both
>> >>  existentially quantified type variables, and @samp{P = Q}.
>> > 
>> > That should be okay.
>> 
>> Ok. I have rewritten this part of the section using more formal wording,
>> though I have no idea how to make the result pretty. Does anyone
>> know how to nicely typeset this kinds of maths in texinfo?
>
> From a quick look, you can insert TeX commands with @math but that will
> do nothing for HTML and Info output. @sub does subscripts which can be
> rendered in HTML, but results in something like S_{i} in Info.

Does anyone even use Info these days?

> I don't see that as better.

Recent versions of texinfo seem to be able to turn @math into MathJax
when generating HTML.  (That said, recent versions of texinfo almost
don't include the one on the machine that currently builds the rotds.)

We could maybe try using conditionals, e.g. @iftex ... @end iftex, to at
least format maths nicely for those output targets where it is possible.

Julien.