Subject: Re: [xsl] // expanding to descendant-or-self::node() From: "Andrew Welch" <andrew.j.welch@xxxxxxxxx> Date: Wed, 17 Sep 2008 20:19:39 +0100 |
2008/9/17 Wendell Piez <wapiez@xxxxxxxxxxxxxxxx>: > At 01:43 PM 9/17/2008, Andrew wrote: >> >> 2008/9/17 David Carlisle <davidc@xxxxxxxxx>: >> > >> > >> >> the distinction between //foo[1] and (//foo)[1]. That has got to be >> >> the biggest downside of the way "//" is defined. >> > >> > and anyway any blemish in the definition of // is minor compared to != >> > which would have been better not being defined, since 9 times out of 10 >> > when it is used it does the wrong thing. >> > >> >> true, perhaps != should be for atomic comparison and much lesser used >> "ne" for set comparison, rather than the other way around... >> >> In Java using != to compare Strings gets underlined as a warning and >> you get a reminder to the use the method equals() instead, so maybe >> the same could be done here > > I'm actually not in favor of changing any of it. I think Mike is right when > he warns us that such "improvements" generally turn out to cost more, and > for longer, than they're worth. sure, me neither, just musing :) that those two particular comparison operators ( != and ne ) would've been better the other way around, as most people would guess right with != and probably not even be aware of "ne" until finding it as a solution to the problem of set comparison. -- Andrew Welch http://andrewjwelch.com Kernow: http://kernowforsaxon.sf.net/
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
Re: [xsl] // expanding to descendan, Wendell Piez | Thread | Re: [xsl] // expanding to descendan, Florent Georges |
Re: [xsl] // expanding to descendan, Wendell Piez | Date | RE: [xsl] Fundimentle Predicate Pro, Michael Kay |
Month |