Re: [xsl] ANN: 'Testing XSLT' training course PDF available

[Steve Ball <Steve.Ball@xxxxxxxxxxxxxx>]

Andrew,

My apologies, I missed the footnote in my initial reading of your  
original message.  The two points I was trying to make were:

1.  there are (generally speaking) two kinds of testing that need to  
be performed on stylesheets:
	a) Black-box testing of the whole stylesheet, and
	b) Unit (white-box) testing of the functions/named templates.

2. both of these testing concepts apply to XSLT 1.0 just as much as to  
2.0.

Regards,
Steve Ball

On 17/12/2007, at 8:31 PM, Andrew Welch wrote:

> On 16/12/2007, Steve Ball <Steve.Ball@xxxxxxxxxxxxxx> wrote:
> > On 14/12/2007, at 11:31 PM, Andrew Welch wrote:
> > > Personally I don't think you can unit test XSLT*, it's only really
> > > worthwhile to test the output for a given set of inputs.
> >
> > You can certainly unit test functions, as Mike Kay points out
>
> If you include the starred footnote along with the quote, you'll see
> that I said that:
>
> "* Maybe with 2.0 and user defined functions it might be worthwhile,
> but to ensure your transforms are doing the right thing, I think the
> only approach it to check the output for a given set of inputs."
>
> It's a bit off to quote something that references a relevant footnote
> and not include the footnote itself.
>
>
>
> --
> Andrew Welch
> http://andrewjwelch.com
> Kernow: http://kernowforsaxon.sf.net/