[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

Re: [xsl] Speeding up collection()


Subject: Re: [xsl] Speeding up collection()
From: Andrew Welch <andrew.j.welch@xxxxxxxxx>
Date: Wed, 10 Jun 2009 09:08:34 +0100

2009/6/10 Jesper Tverskov <jesper.tverskov@xxxxxxxxx>:
> Andrew Welch wrote:
>> You could call out to Java to get the directory list:
>
> Yes, I can use one of the general programming languages my XSLT
> processor supports.
>
> But is that really the method for the task at hand except when we only
> have very few xml files to load?

It's a feature of the language that document(), doc() and
doc-available() all parse the document completely and then (as I
understand it) hold the document in memory, hence the need for the
saxon:discard-document() extension when you process any reasonable
number in one go.  (It's a shame there isn't an inbuilt equivalent of
saxon:discard-document() as it's essential for batch processing)

I don't know whether with collection() the processor has the freedom
not to load the document into memory until its needed, but I suspect
it would have to be a specific optimisation to not load it for
document-uri()

-- 
Andrew Welch
http://andrewjwelch.com
Kernow: http://kernowforsaxon.sf.net/


Current Thread
Keywords