XOM API for Processing XML with Java

[Elliotte Harold <elharo AT metalab.unc.edu>]

luca wrote:

> Elliotte Harold wrote:
>
> > There are use cases for not resolving entities, but this is on the 
> > wrong side of XOM's 80/20 split.
>
>
> No, it's not.

I would say more than 80% of the documents I see don't have document 
type declarations. Of those that do, more than 80% use relative URLs 
that point to local files. And of those that don't, more than 80% can be 
rewritten at the source to use local copies instead.


> What about all of the XHTML content out there?. 

TagSoup is your friend. :-)

> All of the config
> files (TLDs,web.xml,server.xml) which we just need to parse
> without hitting the sun.com website?

If these can't be changed to point to local copies, then a SAX 
EntityResolver can do the trick. As I said, I'll look at ways of making 
this simpler in future versions post-1.0.


--
Elliotte Rusty Harold
elharo AT metalab.unc.edu