June 24, 2004
@ 04:15 PM
Comments [15]
What Would You Like To See in System.Xml in Orcas/Longhorn?

I'm in the end stages of doing the spec work for the various components in the System.Xml namespace  I am responsible for in the Whidbey betas. After the 4th of July holidays we plan to start doing initial brain storming for what feature work we should do in Orcas/Longhorn. I thought it would be valuable to have various users of XML in the .NET Framework suggest what they'd like us to do in the Orcas version of System.Xml. What changes would people like to see? For example, I'm putting Schematron and XPathReader on the 'nice to have' list. No idea is too unconventional since this is the early brainstorming and prototyping phase.

Caveat: The fact that a technology is mentioned as being on our 'nice to have' list or is suggested in a comment to this post is not an indication that it will be implemented in future versions of the .NET Framework.


 

Categories: Life in the B0rg Cube | XML

« A Good Day | Home | Guess Who's Back »
Thursday, 24 June 2004 17:13:21 (GMT Daylight Time, UTC+01:00)
Schematron would be awesome. I think it hasn't received the attention it deserves, largely because of the lack of implementations. It would be a great addition to System.Xml.
Patrick Cauldwell
Thursday, 24 June 2004 18:59:20 (GMT Daylight Time, UTC+01:00)
- Automatic design-time and/or run-time IXPathNavigable generation for any object based on XmlSerialization attributes. Not a generic one like the ObjectXPathNavigator, but a truly "typed" one. Just like the XmlSerializer itself does by generating a specialized (serialization) reader.
- Some kind of API for building specialized derived XML parsers that avoid relying completely on generic XSD for validation, and that allows exposing higher-level constructs from the parsing layer. This would include an easy to use statemachine, i.e. like the Quantum hierarchical state machine.
- RDF API & query (http://www.w3.org/Submission/2004/SUBM-RDQL-20040109/)

My 2 cents ;)
Daniel Cazzulino
Thursday, 24 June 2004 19:00:42 (GMT Daylight Time, UTC+01:00)
- Automatic design-time and/or run-time IXPathNavigable generation for any object based on XmlSerialization attributes. Not a generic one like the ObjectXPathNavigator, but a truly "typed" one. Just like the XmlSerializer itself does by generating a specialized (serialization) reader.
- Some kind of API for building specialized derived XML parsers that avoid relying completely on generic XSD for validation, and that allows exposing higher-level constructs from the parsing layer. This would include an easy to use statemachine, i.e. like the Quantum hierarchical state machine.
- RDF API & query (http://www.w3.org/Submission/2004/SUBM-RDQL-20040109/)

My 2 cents ;)
Daniel Cazzulino
Thursday, 24 June 2004 19:01:46 (GMT Daylight Time, UTC+01:00)
- Automatic design-time and/or run-time IXPathNavigable generation for any object based on XmlSerialization attributes. Not a generic one like the ObjectXPathNavigator, but a truly "typed" one. Just like the XmlSerializer itself does by generating a specialized (serialization) reader.
- Some kind of API for building specialized derived XML parsers that avoid relying completely on generic XSD for validation, and that allows exposing higher-level constructs from the parsing layer. This would include an easy to use statemachine, i.e. like the Quantum hierarchical state machine.
- RDF API & query (http://www.w3.org/Submission/2004/SUBM-RDQL-20040109/)

My 2 cents ;)
Daniel Cazzulino
Friday, 25 June 2004 06:26:49 (GMT Daylight Time, UTC+01:00)
A lightweight C parser.
MSXML Lover
Friday, 25 June 2004 20:07:02 (GMT Daylight Time, UTC+01:00)
+1 to Schematron, especially if you can wire Schematron rules into xsd:appinfo for stronger validation rules than plain schema.
+1 to Daniel's 2nd point.

* RELAX-NG support, if it's not already planned. With Schematron and RELAX-NG, you'd have support for every major validation scheme out there. Personally, when I have to write an XSD, I write it in RELAX and convert it to xsd with James Clark's Trang.
Gordon Weakliem
Sunday, 27 June 2004 02:06:07 (GMT Daylight Time, UTC+01:00)
I would like the xsd.exe tool work in conjunction with the framework to create objects which are created and populated lazily; when I index a value in an array or whatever xsd produces, thats when the xml is parsed.

Further, an option to have generators of forward-only iterators rather than a data structure, with the same sort of behavior.
doug ransom
Wednesday, 30 June 2004 13:08:45 (GMT Daylight Time, UTC+01:00)
RelaxNG would be way cool, indeed. SUN also provides the MSV (multi-schema validator) that supports all validation languages: DTD, XSD, RelaxNG and Schematron.
RelaxNG supports more features than XSD when it comes to unordered content models as well as conditional content models (i.e. if "abstract" attribute is present, dom't allow certain child elements).
Daniel Cazzulino
Wednesday, 30 June 2004 21:52:23 (GMT Daylight Time, UTC+01:00)
+1 for RelaxNG support
G Mladenov
Thursday, 01 July 2004 00:20:05 (GMT Daylight Time, UTC+01:00)
Schematron support sounds great! And if RELAX NG and/or Schematron will be supported, it should be with support of choosing schema on the fly, particularly NRL. http://www.thaiopensource.com/relaxng/nrl.html
Fumiaki Yoshimatsu
Sunday, 11 July 2004 07:34:41 (GMT Daylight Time, UTC+01:00)
OK, Dare, my list was too long to type into your little edit box. See http://www.gotdotnet.com/team/dbox/default.aspx?key=2004-07-11T06:37:50Z for my list.

DB
Don Box
Monday, 12 July 2004 04:14:22 (GMT Daylight Time, UTC+01:00)
Relax NG, Schematron would be nice. But a big one would be creating XMLDOM objects without requiring a XmlDocument. For instance, w/ XmlSerialization, you often have generic XmlNode objects which can't be instantiated w/ new XmlText() or the like. Rather you have to create and deal w/ the XmlDocument. Kinda kludgy.
Randy Charles Morin
Monday, 12 July 2004 07:31:24 (GMT Daylight Time, UTC+01:00)
Wow Randy, are you really advocating that we add public constructors to XmlElement and friends? I imagine the next thing you'll be asking for is to allow an XmlNode to be inserted into a random document, not just the one it was created for.

Oh the humanity...

DB

PS: That is in fact my tongue planted firmly in cheek.

Don Box
Tuesday, 13 July 2004 23:16:40 (GMT Daylight Time, UTC+01:00)
Either XSLT 2.0, an extension system for the XSLT debugger or a rich set of premade functions like EXSLT that work with the XSLT debugger.

However XSLT 2.0 is what I'd like most.
Orion Adrian
Monday, 19 July 2004 16:26:44 (GMT Daylight Time, UTC+01:00)
I would like the xml serializer and the binary serializer to share a method for versioning my objects and running my code when version numbers don't match.

Dan Golick
Comments are closed.