Re: entities in Nautilus (was Re: reduce redundancy?)



* John Fleck (jfleck@inkstain.net) wrote at 17:25 on 22/08/00:
> On Tue, Aug 22, 2000 at 12:22:02AM +0300, Ali Abdin wrote:
> > * John Fleck (jfleck@inkstain.net) wrote at 00:19 on 21/08/00:
> > > On Sun, Aug 20, 2000 at 05:52:39PM +0200, Karl Eichwalder wrote:
> > > 
> > > > 
> > > > OTOH, the browser doesn't have to work with the raw SGML/XML files
> > > > coming from the author.  We should install normalized SGML/XML files
> > > > with entities resolved (`sgmlnorm' and/or `sx').
> > > > 
> > > 
> 
> > 
> > Ummmmmmmmmmmmmmm - Doesn't this defeat the whole purpose of having
> > 'gnome-db2html2' in the first place. Now instead of using 'jade' we use
> > 'sgmlnorm' or 'sx'. Yes, its probably faster - but the whole goal was to
> > eliminate the 'compile' step of the docs (wasn't it?)
> > 
> > The proper approach is to handle entities in gnome-db2html2. Unfortunately,
> > this aspect of the code I do not understand. DV told me he'd look into it, but
> > apparently he hasn't :(
> > 
> 
> Ali is right. The more I think about this, the less I like it. :(
> 
> Using sx or sgmlnorm might work, but it would involve essentially
> keeping two sets of docs - the writer's original, using entities, and
> the final distribution version, which has been normalized to remove
> the entities. This seems like a huge headache.

Actually - if we can't get entities done you /could/ maintain two sets of docs
- the 'original' one and a 'without-entities' one that is "sx/sgmlnorm'd" -
  You make all your modifications to the original docs, but sx/sgmlnorm the
'installed' copy (not as part of the 'compile step' but the authors would have
to do it manually).

I don't like this approach at all - but it might be the best if there is no
entity support.

Regards,
Ali





[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]