W3C home > Mailing lists > Public > www-dom-ts@w3.org > November 2001

Re: More thoughts on multi media-type tests

From: Mary Brady <mbrady@nist.gov>
Date: Wed, 28 Nov 2001 11:17:52 -0500
Message-ID: <00c001c17828$3ec0eb20$293b0681@HAPPY>
To: "Curt Arnold" <carnold@houston.rr.com>, <www-dom-ts@w3.org>
Comments inlined.

--Mary
  ----- Original Message ----- 
  From: Curt Arnold 
  To: www-dom-ts@w3.org 
  Sent: Wednesday, November 28, 2001 10:36 AM
  Subject: More thoughts on multi media-type tests


  I've had some more thoughts about tests that can work on different media-types.  For example, most of the core tests can work on both XML and SVG implementations, most of the HTML tests could work on both XHTML and HTML implementations.

  However, since some processors could support distinct DOM implementations for multiple document classes (for example, a browser could have a distinct implementation for XHTML and HTML), you can't just stiff the browser and pick one document class.

  [mb] In original versions of the test suite, we loaded documents differently depending on the name of the test -- we will of course need some method to 
  indicacte how a document should be loaded.

  Another issue is that HTML implementations are case-insensitive and XHTML implementations are case-sensitive (at least on the Core methods).  The base tests should use lower case tagnames for parameters.  However, there should be some methods that test the case-insensitivity of HTML implementations.

  [mb] It's easy enough to change all tags to lowercase so that XHTML can be tested as well.  What is case-insensitive in XHTML -- just the tags, or the contents 
  as well?  Do you think it would be possible to have a transformation to handle this conversion, rather than separate tests for each -- XHTML, HTML -- that is, 
  have everything lowercase in the xml-ized version of the test, and let the transformation convert to uppercase to test HTML, and leave as is for XHTML.

  To declare that a test depends on a case-insensitive implementation (i.e. HTML not XHTML), I would suggest adding an:

  <implementationAttribute name="caseInsensitive" value="true"/>

  to the prologue of a test.  Again most HTML tests should not have this which would allow it to test both XHTML and HTML implementations.

  [mb] If we used the above scenario, this probably wouldn't be necessary.  We would have equivalent tests for both XHTML and HTML.  Do we need to have 
  an implemenationAttribute to indicate that it is for the HTML module?

  For tests that could apply to multiple media-types, probably would add a <mediaType> to test suites that would allow you to have an XHTML suite and an HTML (or XML and SVG) suite without having distinct XHTML and HTML versions of the tests.  Will think about that a little more.

  [mb] Not sure I understand the mediaType angle -- could you give an example?

  --Mary
Received on Wednesday, 28 November 2001 11:17:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 6 April 2009 12:58:45 GMT