Re: use ENABLE_ENCODING for Info only

[Patrice Dumas]

On Sun, Dec 25, 2022 at 10:03:51PM +0000, Gavin Smith wrote:
> On Sun, Dec 25, 2022 at 10:45:04PM +0100, Patrice Dumas wrote:
> > Hello,
> > 
> > If I recall well, there was some previous discussions, I believe Gavin
> > proposed something similar.
> > 
> > As it is now --enable-encoding (ENABLE_ENCODING customization option)
> > means something very different for Info/Plaintext and HTML/XML based
> > formats.  For Info/Plaintext is means using actual characters and not
> > ASCII transliterations as far as poosible.  For HTML/XML
> > --enable-encoding means using characters instead of entities.  It means
> > something very different, in HTML/XML encodings support is enabled in
> > any case (though not necessarily fully when encoding is US-ASCII), but
> > the option is really about entities or not.
> 
> I think this is a good idea.  You might be referring to this mail I
> sent:
> 
> https://lists.gnu.org/archive/html/bug-texinfo/2022-10/msg00065.html
> From: Gavin Smith
> Subject: Re: with HTML output, @minus{} is converted to a hyphen instead of a 
> real minus character
> Date: Thu, 13 Oct 2022 19:30:07 +0100

Exactly.

> > I would propose I sthat ENABLE_ENCODING only says something on supporting
> > encodings or not, and that it has not much effect except for Info and
> > Plaintext.  I propose to use a customization variable for HTML/XML,
> > for example named OUTPUT_CHARACTERS.
> 
> Would the use of ENABLE_ENCODING simply be renamed to something else
> in the code for HTML/XML?

I think so, but I need to check, as ENABLE_ENCODING is used more
generally for conversion to Text, which is common to more than one
format.  As a side note, to workaround that ENABLE_ENCODING is
in general not set for HTML/XML, but we want to convert to Text as if it
was set, there is a trick to do as if ENABLE_ENCODING was set even if
it is not, by setting the second argument of
Texinfo::Convert::Text::copy_options_for_convert_text
to 1. Having a separate variable would allow to avoid needing that
trick.  So it may not be only renaming to something else, but the
code could be simplified too.

> If that's the case, then we could easily
> change the name of the variable later if we think of a better name.

Ok.

-- 
Pat