Re: [hlx] Suggestion for a few changes

[tom s. <to...@as...>]

Ryszard Kubiak <ry...@bi...> wrote:

> 1. I think it would be nice to have a variable like:
> 
>    (defvar hyperlatex-xml-charset "UTF-8"
>      "The name of character encoding for an entire XML document")
> 
> so that a user may override the UTF-8 encoding which is now
> hard-wired in the code.

At the moment, my understanding is that the code only supports UTF-8
(and that only because UTF-8 is backward compatible with ASCII).  I am
far from knowledgable about encodings, especially where emacs is
involved, so if I'm wrong, please explain it to me (or make me a patch
that supports other encodings), and I'll be happy to change it.

> 2. There is a bug around line 1850 of hyperlatex.el. The following
> cannot work properly in Emacs since - is a meta-character in regular
> expressions:
> 
>    (string-match "^[a-zA-Z][a-zA-Z0-9_-.:]+$" label)
> 
> A consequence of the error is that HyperLatex converts too many
> labels to a form idN which in many cases is not what we want.
> A patch might look like:
> 
>    (string-match "^[a-zA-Z][a-zA-Z0-9\-\_\.\:]*$" label)
> 
> Note that _ has moved to a second posiiton, after \-, because of
> another bug, in Emacs this time. Should _ (no matter whether
> escaped or not) stay at the first position after 9 Emacs
> would fail to string-match labels like "k_node".

Thank you, that seems to work just as you say (including the emacs
bug).  I'll change it.

> 3. I've just noticed that a \label command generates some output.
> I very much believe that it should should not! For example,
> \label{id4} generates <a name="id4">&nbsp;</a> while
> <a name="id4"></a> should suffice. At least it does with
> FireFox while the spurious &nbsp; badly pollutes layout.

Can anyone on the list suggest why this should not be done?  Does anyone
know why the space was in there?

Thanks,

 -tom

-- 
 ------------------------
 tomfool at as220 dot org
 http://sgouros.com  
 http://whatcheer.net