Re: shape.dtd, diagram.dtd

[James K.Lowden <jklowden speakeasy org>]

On Thu, 11 Jul 2002 06:03:31 +0100 (BST), "Andrew Ferrier"
<andrew junk new-destiny co uk> wrote:

> > > Is it undesirable for Dia to use its DTD?
>
> Well, normally, 'using' a DTD would involve validating the
> input XML/SGML (or whatever format is derived from it), to
> ensure that the syntax is valid. That's really the only 'use'
> of a DTD as far as I know.

Same here.  

> In this case, ensuring validation would probably be a waste of
> time unless it was (very) trivial (which, incidentally, it
> might be if the appropriate XML libraries supported it). 

Do you mean "if they did, but they don't"?  

> If it were a _user_ modifying these files, that would be a
> different matter entirely (for example, if a user typically
> modified .dia files it would be well worth validating these
> against a DTD --- or, better, an XML schema, DTD's
> less-well-supported but more featureful cousin).

I have in mind generating Dia files from external sources, "sql2dia" if
you will.  Properly done, that would rely on the DTD to determine the
output structure.  

I think XML transformation tools will continue to improve, making such
things relatively easier over time.  There are many external data stores
that would be easier (for people) to read if they could be exported to
Dia.  

Every time I think about this, I remember that Cyrille was right:
diff/patch for xml would permit updating a .dia file from an external
store (which would not typically have Dia layout information), or merging
two diagrams.  

Thanks for the clarification.

--jkl