DOM Level 3 Load/Save Implementations report

Date: $Date: 2004/02/05 20:12:07 $

Note: The information contained in this page may be updated without notice to reflect the latest version of the DOM Level 3 Test Suite and implementations. We are still working on the tests to improve them with regards to the W3C Recommendation of DOM Level 3 Load/Save, therefore the results of the tests are not necessarely fully accurate. We appreciate any bug report or comment regarding the tests (use www-dom-ts@w3.org).

As a reminder:

The DOM TS (DOM Test Suite) aims at helping implementors to write applications that support the DOM specifications. In no way are these conformance tests in the sense of providing companies or institutions with certification of DOM support. The only claim that could be made is that a particular implementation is conformant to a particular version of the DOM TS.

The DOM Level 3 Load and Save test suite contains ? tests, as of January 8, 2004. To build and run a DOM test suite, see Test suites.

Disclaimer: The implementations presented on this page were tested during the development of the specification and may contain imperfections.

• Implementation report from Xerces:
  1. parseWithContext and async are not supported but those are optional features of the specification.
  2. Character normalization won't be implemented in the near future, but this is an optional in the specification.
• Implementation report from X-Hive:
  1. Parameters "infoset" and "well-formed" are not supported.
  2. async is not supported but is is an optional feature of the specification.
  3. Character normalization won't be implemented in the near future, but this is an optional in the specification.
• Implementation report for pxdom module (Python)

  Not supported: external resource resolution & loading, validation (optional), asynchronous LSParsers (optional), format-pretty-print. Future versions may support external entities and pretty-printing; the others are unlikely.

  pxdom supports the Character Normalization where the native language libraries contain the functionality - unicodedata.normalize in Python 2.3+.

  Requiring all DOMs support character normalization even on platforms where such libraries are not easily available is not desirable - and I'm not convinced the feature should be 'on if supported' by default in LSSerializer, either. But including it as an optional feature does no harm.