RE: [cgmo-webcgm] Errors in XCF DTD?

["Cruikshank, David W" <david.w.cruikshank@boeing.com>]

 Oh...and you'll also note that in the attached dtd the DOCTYPE is
commented out....need to uncomment that and the close bracket at the end
of the file.  For some reason the products didn't like the DOCTYPE
included.

Thx...Dave

-----Original Message-----
From: Lofton Henderson [mailto:lofton@rockynet.com] 
Sent: Tuesday, October 25, 2005 7:50 AM
To: Cruikshank, David W
Cc: cgmo-webcgm@lists.oasis-open.org
Subject: RE: [cgmo-webcgm] Errors in XCF DTD?

At 04:37 AM 10/25/2005 -0700, Cruikshank, David W wrote:
>[...]
>I would be interesting to know what parser Mary uses to validate.

You'll have to ask her.  I don't know.  Note that she hasn't tested our
stuff yet.  In fact, I'm unsure whether she would test it embedded in
the text.  But Ch.4 says that we're going to extract the DTD and put it
stand-alone in OASIS space.

>We've
>been running the XCF dtd through multiple (4) parsers and 2 of them 
>don't like the null defintions for all the "EXT" entities, but XMLspy 
>and TurboXML pass them.  Xmllint and Omnimark object to them.
>
>Also, on all the parsers we have to change the UTF-16 to UTF-8 to get 
>them to accept the dtd??

I don't know about that.

Are you giving the parser a DTD of one-byte or two-byte codes?  If you
are only extracting the DTD from Ch.4 text in a typical editor, then I
suspect you are getting ISO Latin1 -- amaya "Document Information" on
Ch.4 says the CharSet is ISO-8859-1 (whose codes are also 1-byte codes
in UTF-8 encoding).

If you had an editor that would let you force the encoding, then you
could SaveAs UTF-16.

Should we change the XML declaration on the DTD to UTF-8, so that people
won't have SaveAs UTF-16 to parse?

-Lofton.


>-----Original Message-----
>From: Lofton Henderson [mailto:lofton@rockynet.com]
>Sent: Monday, October 24, 2005 4:33 PM
>To: Cruikshank, David W
>Cc: cgmo-webcgm@lists.oasis-open.org
>Subject: RE: [cgmo-webcgm] Errors in XCF DTD?
>
>At 03:24 PM 10/24/2005 -0700, Cruikshank, David W wrote:
> >  There are a couple of typos:
> >
> >Right after the styleproperties entity there is a line that has "<"
> >bunch of spaces ">" - DELETE IT
>
>fixed.
>
> >The ATTLIST statement for webcgm has an "!" missing just before the 
> >"ATTLIST"
>
>fixed.
>
>
> >One other problem I haven't figure out yet....
> >
> ><!ELEMENT layer (%layerEXT;)*
> >
> >%layerEXT is defined as "" so the ELEMENT content is illegal to a 
> >parser.  We may need to define the generic value for layerEXT as 
> >EMPTY somehow.  I haven't worked it out yet.
>
>Please do this asap, or assign it if you can't.  The text freeze for CS

>will be in a couple of days, followed by a short TC-only review 
>(requested by Dieter).
>
>This WILL prevent CS, according to Mary -- any DTDs, schemas, etc must 
>parse or the spec will send it back to us.
>
>-Lofton.