libxml from Daniel Veillard on 2000-10-30 (www-xml-linking-comments@w3.org from October to December 2000) (original) (raw)

        XML Linking Implementation Questionnaire
        ========================================

General

What is the name of your organization?

The Gnome Project (and W3C)

How should we contact you on linking matters?

Mail at Daniel.Veillard@w3.org

What is the name of the linking technology?

libxml, the Gnome XML Library

Please describe its functionality and development status briefly.

This library allows to manipulate XML files. It includes support to read, modify and write XML and HTML files. There is DTDs support this includes parsing and validation even with complex DtDs, either at parse time or later once the document has been modified. The output can be a simple SAX stream or and in-memory DOM like representations. In this case one can use the built-in XPath and XPointer implementation to select subnodes or ranges. A flexible Input/Output mechanism is available, with existing HTTP and FTP modules combined to an URI library.

Is it publicly available? If so, what is its current location?

It is publicly available with source at http://xmlsoft.org/

XLink

How does the technology use XLink?

There is no XLink support so far.

XPointer

How does the technology use XPointer?

It provides an API to evaluate XPointer expressions and return the results.

Does (or will) the technology claim XPointer conformance?

yes, it does now.

Please provide details about the parts of XLink implemented.

Bare names yes
Child sequences yes
id() function yes
Full XPath yes
range-to yes
string-range yes
Namespace support yes but not well tested and probably needs improvements
In programming API yes
In user interface no
- Selecting/showing multiple non-contiguous ranges
Any other comments/details?

What difficulties did you discover in using the XPointer specification?

Couple of errors in the examples
Range-to wasn't specified clearly enough but discussion with Michael Kay led to a good description
lack of any examples for the functions in section 5.4.3 Additional Range-Related Functions http://www.w3.org/TR/xptr#N2698

What difficulties did you discover in implementing XPointer?

this was implemented on top of the libxml XPath implementation and forced me to revamp some of the early interfaces for better expandability

XML Base

How does the technology use XML Base?

there is a simple function to get the XML Base lookup, this is an unified function in the sense that it does it independantly of the type of document (HTML or XML).

Does (or will) the technology claim XML Base conformance?

I would prefer to have it being reused (for example in an XInclude module) bofeore claiming conformance.

Please provide details about the parts of XML Base implemented.

This is actually quite small, correct support for entity boundaries checking need to be refined.

What difficulties did you discover in using the XML Base specification?

none, this is fairly small and easy.

What difficulties did you discover in implementing XML Base?

I don't have specific XHTML support, and so far I'm afraid that if parsed as XML it will use xml:base to do the base lookup and if parsed with the HTML parser it will use the base element.

Integration

How does the technology uses XLink, XPointer, and XML Base in conjunction?

Without XLink nor XInclude support there isn't much integration. For the moment it's mostly separate APIs in the toolbox. I hope to implement XInclude soon.

Please provide details about the implementation.

Handling single nodes as participating resources
Handling ranges of nodes as participating resources
Handling participating resources made up of non-contiguous pieces
Any other comments/details? What difficulties did you discover in using their specifications together? What difficulties did you discover in implementing them together?

Conclusion

Do you have any other comments you would like to share with us?

I personally think that both specs are ready for general public consumption but adding a few more examples would not harm.

Daniel

Received on Monday, 30 October 2000 05:48:09 UTC