Subject: Re: [xsl] Output a default value for an empty node From: David Carlisle <davidc@xxxxxxxxx> Date: Thu, 7 Oct 2004 10:14:41 +0100 |
I thought the usual answer to this kind of query was to use normalize-space(): If you really are testing for empty then normalize-space is likely to be relatively expensive (ie time proportional to the length of the string) as it will have to go through checking every character to normalise it. whereas checking for text() is probably (never looked at the code of the processor) just checking that some structure has a reference to a child object or not, ie constant time check. Of course, if by "empty" you include elements containing spaces and tabs etc, then you do have to pay that cost. David ________________________________________________________________________ This e-mail has been scanned for all viruses by Star. The service is powered by MessageLabs. For more information on a proactive anti-virus service working around the clock, around the globe, visit: http://www.star.net.uk ________________________________________________________________________
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
RE: [xsl] Output a default value fo, Andrew Welch | Thread | RE: [xsl] Output a default value fo, Andrew Welch |
Re: [xsl] Ampersand in attributes, David Carlisle | Date | [xsl] XPATH wildcard problem with n, Ranjan K. Baisak |
Month |