W3C home > Mailing lists > Public > xmlschema-dev@w3.org > June 2008

Re: schema arrangement or architecture

From: Andrew Welch <andrew.j.welch@gmail.com>
Date: Tue, 3 Jun 2008 10:48:33 +0100
Message-ID: <74a894af0806030248n443e5618t182343c03f691401@mail.gmail.com>
To: "Boris Kolpackov" <boris@codesynthesis.com>
Cc: xmlschema-dev@w3.org

2008/6/3 Boris Kolpackov <boris@codesynthesis.com>:
> My ideal schema has all its types defined globally (e.g., no
> anonymous types), and all its elements and attributes --
> locally, except for the elements that are valid document roots.

Currently mine is pretty much in this order:

- all elements
- some attributes (where the types are user defined - I'm wondering
whether it should be all of them)
- simple types
- complex types

I like this arrangement because the element and attribute list form a
reference / index at the top of the schema.

On the current project the schema is just over a 1000 lines, all
elements are from the same namespace, but could be grouped into a few
logical areas (which they are in the single file - as in elements are
not in alphabetical order, but in their logical groups... perhaps they
should just be in order...)

The actual file size is fine for me, but for scalability and
maintenance reasons I wondered how others approached it, so far it
seems there is no common approach - in the mean time the one big file
approach is fine for me and I'm happy that I'm not missing something.

One thing that would really help is control-clicking the type or ref
attribute values to jump to their respective targets - in oXygen I can
click around the diagram to jump from say element declaration to type
definition which is fine, but often I just minimize the diagram view
and use the text editor so control-clicking the attribute values would
be good.



-- 
Andrew Welch
http://andrewjwelch.com
Kernow: http://kernowforsaxon.sf.net/
Received on Tuesday, 3 June 2008 09:49:09 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:56:14 UTC