XOM API for Processing XML with Java

[Wolfgang Hoschek <whoschek AT lbl.gov>]

On Nov 14, 2004, at 4:32 AM, Elliotte Harold wrote:

> Wolfgang Hoschek wrote:
>
> > #1 Right, the Verifier URI sanity checks are quite involved, 
> > unfortunately. The JDOM checks are less involved (but also less 
> > strict). If you look at how java.net.URI is implemented (someone 
> > seems to have worked real hard on that code) you might, with luck, 
> > find some tricks that can help.
>
> That makes a lot more sense. It occurs to me now that I could probably 
> speed this up for the common case by caching three or four namespace 
> URIs (it's a rare program that uses more than that) and not 
> reverifying them for each and every element. Then however expensive 
> the URI verification code is, you'd only need to pay the price once, 
> at least most of the time. That might remove the need for one of your 
> patches.

A small cache array might help quite a bit, indeed. Would need to 
synchronize or ThreadLocal reads and writes, since all in Verifier is 
static. Lock contention might become an issue.

> I'll look at your benchmark code and see what I can do. Does the code 
> include the sample documents as well?

No, but quite a few are linked from the javadoc. You might want to 
start with the XBIS datasets, e.g. soap.xml which is one of the 
"toughest".

> One thing I've noticed while looking at this is that a lot of my tests 
> cases I've been using to optimize are namespace free. I do need to do 
> some more testing with documents that use namespaces.

There may be some surprises then. The Hashmap & Treesets manipulations 
typically also show up prominently on iteration for serialization, and 
also on deserialization. Tool-generated xml often contains lots of 
redundant namespace declarations.

Wolfgang.