Hey Rex (any everyone else),
The Adoption SC would love any ideas or content you'd like to work on
as part of an instructional cookbook. I'm planning on putting together
some coding examples for the .NET platform of standard operations that
are part of working with the standards. If there are any other coders
out there that would like to contribute examples for other platforms
please let me know. (Java would be a big help). Some ideas I've had for
examples are:
1. Generating proxy classes from a schema.
2. Validating an xml string against a schema.
3. Consuming a web service that uses one of the standards. DM OPEN
would probably be the first choice.
--Josh
Joshua Shows
Software Engineer
Emergency Services integrators (ESi)
699 Broad St, Suite 1011
Augusta, GA 30901
706-823-0911
-----Original Message-----
From: Rex Brooks [mailto:]
Sent: Sunday, March 25, 2007 10:28 AM
To:
Subject: [emergency] schemaLocation for validating local copies of
standards
Hi Everyone,
One of my action items from last week's TC meeting was to post a
message detailing how to provide and specify a locally stored copy of
an XMS Schema from a standard rather than requiring a parser to find
the document on the web in order to validate it.
The problematic part of validating The XML Schema from standards is
that you must cite the namespace. This is specified as a URN (Uniform
Resource NAME), which is pretty straightforward since we are
specifying a NAMEspace. However, a urn just specifies the legal
provenance of the NAME.
This is the point that is usually missed.
You should also specify a URL to the "schemaLocation." If you specify
the location where the standard is supposedly, or is supposed to be
kept you can get into trouble if the site is down for maintenance or
under a denial-of-service attack. Plus you would be wasting bandwidth
that is especially important during an emergency incident.
If your system's validator can't validate the schema from the remote
location, your system could break. So it is better to keep a local
version, which also gives you complete control over the correct
versioning for your applications.
You should keep a correct version of any standard or set of standards
your system uses, or that are included within other standards your
system uses. However, you still need the schema(s) as separate .xsd
files because these are the files you specify in the "schemaLocation"
which allows your validator to validate the schema locally and
prevent inadvertent failures of the kind mentioned above.
For reference, here's an example:
http://www.stylusstudio.com/w3c/schema0/schemaLocation.htm
For CAPv1.1Errata you need the file: cap1.1.xsd
Here's a sample of what I would use:
<c xmlns="urn:oasis:names:tc:emergency:cap1.1">
<xsi:schemaLocation=",/your/path/cap1.1.xsd">
This specifies that in <c:xxx...> "c;" is the abbreviation that
specifies that element xxx refers to xxx from the namespace for
cap1.1.
This is a simplification of something that can be greatly
complexified for any number of reasons
We can't anticipate every need, But, we make sure that we have a
zipped file of all standards referenced, "included" and/or "imported"
in our standard and a zipped file of all .xsd files from those
standards. This way we should be able to prevent the most common
kinds of failures. If we also include a clear explanation of how to
use our standards in simple, base application examples, it will be
much easier to adopt our standards.
I would be happy to work with the Adoption SC to develop this into
part of the cookbook or set of cookbooks for our specifications,
since this particular piece of practical information applies to all
of our specifications.
Cheers,
Rex
--
Rex Brooks
President, CEO
Starbourne Communications Design
GeoAddress: 1361-A Addison
Berkeley, CA 94702
Tel: 510-849-2309
