On Friday 04 July 2003 14:49, TAKAHASHI Naoto wrote:
> Alexej Kryukov writes:
> > First of all, Greek ano teleia (#x0387) is not
> > converted from Unicode.
>
> Tell me ano teleia's Ibycus notation and I will add that conversion
> to the rules.
It looks like greek.el already contains the correct rule:
(";" ?О‡) ; ano teleia
which, however, for some reasons doesn't work in the conversion
routine.
>
> > Second, many redundant pairs of \eng{greek} ... \begin{greek}
> > are produced. The algorithm used removes these pairs
> > on line breaks, but not on common punctuation characters,
> > like period or comma.
>
> It is designed to do so and nobody has complained about it so far.
I think, nobody used it for large texts. Please, take a look
at the attached Unicode document and how it was converted to
the Ibycus notation. Is it correct or normal?
> If you suggest a better behaviour and other people second, I will
> change it.
Well, as I've already said, to my mind cgreek shouldn't insert
these markup commands at all. However, if it does, you have to include
additional operations for removing redundant \end{greek} ...
\begin{greek}. For example:
(goto-char (point-min))
(while (re-search-forward "\\\\end{greek}\\([,
]*\\)\\\\begin{greek}"
nil t)
(replace-match "\\1"))
(goto-char (point-min))
(while (re-search-forward "\\\\end{greek}\\([\.
]*\\)\\\\begin{greek}"
nil t)
(replace-match "\\1"))
Attachment:
cgreek-test.tar.gz
Description: application/tgz