I know we are only talking milliseconds here and what I'm currently working on
is just over 100 docs. But when it comes down to the other projects I work on
and it is now 10's of thousands if not 100's of thousands of docs then
efficiency might come into play. (it is sometimes hard to explain to a manager
why the process is taking so long, concept of time out the window the higher on
the corporate ladder).

So the processor I have is a dual core 2.00Ghz Intel chip and 2GB Ram. Not bad
for a laptop.

Marijan (Mario) Madunic

Quoting Michael Kay <mike@xxxxxxxxxxxx>:

> Efficiency and speed depend entirely on the processor you are using: as
> always, the rule is to try some experiments that are representative of your
> particular situation.
> 
> I would have thought collection() was more convenient here, but doc() more
> portable.
> 
> Michael Kay
> http://www.saxonica.com/ 
> 
> > -----Original Message-----
> > From: Mario Madunic [mailto:hajduk@xxxxxxxx] 
> > Sent: 26 June 2008 14:52
> > To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> > Subject:  which is more effiecient and fast doc() or collection()
> > 
> > Working on converting an XSD to SVG (will hopefully look like 
> > TURBOXML's lovely view). I'm mainly using doc() with a regex 
> > to parse the various directories of XSD files. Since the XSD 
> > is built modularly and each element in its own file (I 
> > personally do not like reading XSDs or XSLTs (for that 
> > matter) in one large file and have to scroll up and down to 
> > make sense of it all, but that is another
> > discussion) meaning there are a lot of files. So before I 
> > test it out and spend time on switching from doc() to 
> > collection() I was hoping for some feedback. 
> > 
> > Which is the better method? Which is faster or more efficient?
> > 
> > Marijan (Mario) Madunic