Subject: RE: [xsl] Re: Incrementing a Global variable From: "Michael Kay" <mhk@xxxxxxxxx> Date: Thu, 28 Aug 2003 15:40:31 +0100 |
> If you have a function f(x) and and expression > > f(x) + f(x) > > it may be optimised to 2 * f(x) which means you only have to > calculate the function once. > > However you can't do that if there is a possibility of the > "function" doing something strange like incrementing teh > value of some second variable y as in the first case y goes > up by 2 and in the second it goes up by 1. Unfortunately when it's f(x) is f(x) then you do have to evaluate the function twice; expressions in XPath 2.0 are allowed to have the limited side-effect of creating new, distinct nodes. The formal semanticists seem happy that this is under control, but I can't say I've worked out all the implications for my optimizer yet! Michael Kay XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: [xsl] Re: Incrementing a Global, Wendell Piez | Thread | Re: [xsl] Re: Incrementing a Global, David Carlisle |
RE: [xsl] Re: What would be the bes, SANWAL, ABHISHEK (HP | Date | RE: [xsl] tail-recursion, Michael Kay |
Month |