Subject: Re: [xsl] functions and XSLT From: Colin Muller <colin@xxxxxxxxxxxxxx> Date: Wed, 14 Mar 2001 09:34:44 +0800 |
Jeni Tennison wrote: > Yesterday, I suggested a complicated third way that would involve > having a core of peer-reviewed, tried and tested functions within each > module. Other functions could be added as per David's suggestion, so > that they're available quickly, but these would only be moved into the > core following review (on version changes). If you based the repository around conformance tests, you'd partly remove the need for human review of version changes and new implementations. If no function (and associated implementation/s) is added to the repository without at least an alpha version of a conformance test for it (which could be as simple as xml, xsl and expected output), then people wanting to make use of an implementation can easily check for themselves whether any implementation passes version x.y of the test, and the test can progress towards a version 1.0 as implementations reveal its weak spots and these are patched. The test version number in this scenario is more important than any implementation version number. This could reduce (somewhat at least) the workload on the reviewers needed to achieve what you suggest above, and so speed up the process of adding implementations to the repository - and increase users' confidence in what the repository provides. Sort of like regression testing, and very much like XP's unit tests. Colin XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: [xsl] functions and XSLT, Jeni Tennison | Thread | RE: [xsl] functions and XSLT, Michael Kay |
[xsl] Resend: Is there a way ..., Allen Chang | Date | RE: [xsl] Removing duplicates for u, Christopher Go |
Month |