This is the mail archive of the davenport@berkshire.net mailing list for the Davenport project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: DAVENPORT: FOSI Stylesheets


/ "Beckers, Marc" <mbe@software-ag.de> was heard to say:
| Now, when I try to compile docbookx.dtd v3.1.3, the Adept parser
| trips over the notations and character entities. The only way I can get 
| docbookx.dtd to compile is by commenting the notations and character
| entity declarations out of the DTD.

Here's a little trick you can play.  First note that I've left
public identifiers in the DTD, so you can remap the files that
are pointed to with a catalog. Now, create character entity
files that contain declarations that look like this:

<!ENTITY cularr	"&#38;#x21B6;"> <!-- ANTICLOCKWISE TOP SEMICIRCLE ARROW -->

Tinker with the catalog so that Adept finds these entity files
instead of the ones I supplied, and you should be all set.  Er,
with Adept 8.0, anyway.

The "escaped" ampersand gets passed a limitation in the MarkIt
parser used by dtdgen. And Adept proper does the right thing with
instances.

In fact, this is completely legit according to the 1.0
recommendation.  Numeric character references are really a wierd
sort of "encoding" since they are required to be "completely
expanded" during the parse.

| What has been your experience? How do you deal with character entities /
| notations
| when you use docbookx in Adept?

What's your problem with notations?

                                        Cheers,
                                          norm
P.S. Disclaimer: I work for Arbortext.

-- 
Norman Walsh <ndw@nwalsh.com>      | Don't put all your hypes in one
http://www.oasis-open.org/docbook/ | home page.
Member, DocBook Editorial Board    | 


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]