[ECOS] CVS docs
Bart Veer
bartv@ecoscentric.com
Tue Oct 29 14:03:00 GMT 2002
>>>>> "Iztok" == Iztok Zupet <iztok.zupet@vsr.si> writes:
Iztok> Hello,
Iztok> It's possible to edit the SGML's using the Xmetal 2 or 3.
Iztok> To achieve that, You must import the docbook 3.1 dtd and
Iztok> accompanying files into the Xmetal's Rules directory and
Iztok> modify the catalog inthere like the one in attachment.
Iztok> Since there's no support for dsssl or xsl styles in the
Iztok> Xmetal editor, You must use the docbook.css from the
Iztok> attachment, in order that the preview works. Just copy the
Iztok> css from the attachment into the Xmetal's Display
Iztok> directory.
Iztok> Xmetal has a nice TagOn view and nice support for tag
Iztok> editing, besides WYSIWYG.
I am not familiar with Xmetal, but there is an important issue about
such tools generally. They may not mess about with the SGML contents
of a document. However when outputting the savefile most tools will
completely change the structure. For example tags may be indented
differently, whitespace may get removed or added, paragraphs may get
wrapped using a different number of columns, etc. This will not be
noticeable if you only use a single tool, but other people using e.g.
emacs may be badly affected.
As an example, compare the raw text of e.g.
kernel/current/doc/kernel.sgml and
compat/uitron/current/doc/uitron.sgml. The former was written in
emacs, and I would be rather unhappy if the structure changed
completely because somebody edited the document in a WYSIWYG tool.
Worse, if people start using different tools and the structure is
changed then CVS diffs and histories can become completely useless: as
far as CVS can tell the whole file has been changed. This also makes
it difficult to generate a simple patch. Possibly tools like osgmlnorm
could be used, but I don't know how effective they are.
Bart
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss
More information about the Ecos-discuss
mailing list