Re: rethinking @def*

[Patrice Dumas]

On Wed, Aug 10, 2022 at 03:24:12PM +0100, Gavin Smith wrote:
> On Wed, Aug 10, 2022 at 10:37:41AM +0200, Patrice Dumas wrote:
> > Should be do that, which means never have combinations?  If we do that
> > for those commands, it would be logical to do it for other specific
> > indicatric @-commands, such as @option, @file, @env...?  Combinations
> > would only be possible within those commands, and for font commands such
> > as @slanted and similar.  That means that something like @var{@code{}}
> > or @code{@var{}} will always only apply the internal @-command
> > formatting.
> 
> I think it makes sense for @code (and @t) at least, to force an upright

For @code, ok, but for @t, I disagree.  @t is just a font change
command, it should not force anything.  If @t is always upright, also,
then it is impossible to have a slanted typewriter formatting.  It could
be left undefined and format specific, but forcing @t to be always
upright (in @def arguments) seems wrong to me.

> shape and never slant the argument.  It is necessary to change the
> output for @code anyway, because \texttt is not enough: compare
> 
> \texttt{a : b} with \texttt{\frenchspacing a : b}.
> 
> The former has extra space after the :.

This is unrelated to the font issue, it is however, similar to the --
--- ligatures issue.  I will add something to the TODO on that issue.

> We can leave the other commands as they are unless a problem presents itself.

I find that inconsistent, but if you prefer...

I will adapt the code to have embrac work.

-- 
Pat