XOM API for Processing XML with Java

[Wolfgang Hoschek <whoschek AT lbl.gov>]

On Jan 6, 2005, at 3:28 PM, Elliotte Harold wrote:

> Wolfgang Hoschek wrote:
>
> > Take JDBC PreparedStatements as an analogy. A query is compiled and 
> > optimized once (an expensive process in particular for optimizing 
> > engines such as Nux/Saxon), then executed many times, each time with 
> > different variables aka parameters. Say variables customerID, SSN, 
> > sectionTitle, timeRange, etc. For example, the Nux XQueryPool 
> > addresses such usage scenarios in a easy-to-use thread-safe manner.
>
> But you notice I'm not preparing statements. I'm just passing in 
> strings. I think I'll  leave such advanced usage to NUX. :-)

Assembling the query via string concatenations of fixed parts and 
variable parts isn't particularly easy to use, IMHO.

The other aspect is that repetitive query parsing and optimization 
often takes considerable time which would make XPath navigation less 
appealing over hand crafted tree navigation code. For example, consider 
complex queries over many tiny documents. Here, parsing and 
optimization can be a large fraction of overall execution time. 
Minimizing XPath/XQuery overhead leads to fewer arguments in favour of 
manual navigation.

Wolfgang.