Subject: Re: Better include them in the XSLT 2.0 spec (Was: Re: Time for an exslt for 2.0?)
From: Colin Paul Adams <colin@xxxxxxxxxxxxxxxxxx>
Date: 13 May 2005 09:53:14 +0100
|
>>>>> "Colin" == Colin Paul Adams <colin@xxxxxxxxxxxxxxxxxx> writes:
Colin> Or are you requesting banning non-pure functions
Colin> altogether? This would mean disallowing calls to extension
Colin> functions, or else insisting extension functions must not
Colin> have side-effects. The latter condition would be
Colin> incompatible with XSLT 1.0, I think
Actually, it's far worse than that - functions such as last() are not
referentially transparent, so you can't have referential transparency.
Given that, would there be any advantage to having f($x) is f($x)
always returning true()?
--
Colin Adams
Preston Lancashire
|