httpd-docs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From André Malo>
Subject Re: cvs commit: httpd-2.0/docs/manual/style common.dtd
Date Mon, 16 Dec 2002 00:18:09 GMT
* Erik Abele wrote:

> Okay, you're partially right :-) There is another, not so obvious
> advantage with this: if we consider direct usage of the XML docs in the
> future and forget about the ridiculous Mozilla bug, we don't need to
> distribute the entity and DTD files (.ent/.dtd) later on, since they are
> referenced with their PUBLIC identifier. We just 'fake' the online
> references while building the XHTML files so there is no need to fetch
> these files on every build. This is just for our convenience ;-) The
> online XML docs are still referenced to the (always up-to-date)
> resources, as they should IMO.
> Well, since we're not using the XML docs directly up to now this is not
> very important, but later on it can save resources...

*hrm*. At first: currently the transformation process doesn't care about 
the catalog and loads the entities from the w3. (ISDN, not very funny).

The mozilla bug is not really a bug. AFAICS it's intended. mozilla loads 
external documents only from urls that use the chrome: scheme. 
Unfortunately I don't find the url again, where I read this ;-(

However. I'm not happy with that solution, since it adds a dependency to 
the process that is (currently) not needed. If we consider sometime not to 
distribute the entity files (that are only required for the transformation 
process), we may simply re-add the stuff, that's now in.
The .dtds are already referenced absolute and with publicId, so there's no 

I think further, that the build process is our own. It references to some 
of our own dtds, that use "our own" entities (currently not really, I know, 
but it's the same argument, what happens in future?)

hmm. Other opinions?

print "Just Another Perl Hacker";

# André Malo, <> #

View raw message