Re: rethinking @def*

[Gavin Smith]

On Sun, Jul 31, 2022 at 07:46:02PM +0200, pertusus@free.fr wrote:
> On Sun, Jul 31, 2022 at 02:38:17PM +0100, Gavin Smith wrote:
> > 
> > I've got a different proposal now for the difference between
> > @deftypefn and @deffn (described above).
> 
> I tried to implement that proposal (at least what I had understood...)
> both in LaTeX and HTML.

You made changes like this

-        $tree = $self->gdt("\@strong{{name}} \@emph{{arguments}}", {
+        $tree = $self->gdt("\@code{{name}} \@r{\@slanted{{arguments}}}", {

I am not sure if that leads to an improvement in the output.  In the
reference test results there are changes like


-<dt class="deffn defun-alias-deffn" id="index-function"><span 
class="category-def">Function: </span><span><strong 
class="strong">function</strong> <em class="emph">(arg1, arg2)</em><a 
class="copiable-link" href=\'#index-function\'> &para;</a></span></dt>
+<dt class="deffn defun-alias-deffn" id="index-function"><span 
class="category-def">Function: </span><span><code class="code">function</code> 
<span class="r"><i class="slanted">(arg1, arg2)</i></span><a 
class="copiable-link" href=\'#index-function\'> &para;</a></span></dt>

As you noted, the name is in a typewriter font due to <code> rather than
<strong> being used.  Maybe we discussed this already, but I think we
should reconsider this change.  It should be easier to understand this
issue than that of how to format the argument list.

What was the benefit of changing <em class="emph"> to
<span class="r"><i class="slanted">?  Isn't the former much simpler?