Skip to Content.
Sympa Menu

xom-interest - Re: [XOM-interest] Canonicalizer Performance

xom-interest AT lists.ibiblio.org

Subject: XOM API for Processing XML with Java

List archive

Chronological Thread  
  • From: "Michael Kay" <mike AT saxonica.com>
  • To: "'Wolfgang Hoschek'" <wolfgang.hoschek AT mac.com>
  • Cc: 'xom-interest' <xom-interest AT lists.ibiblio.org>, 'Elliotte Harold' <elharo AT metalab.unc.edu>
  • Subject: Re: [XOM-interest] Canonicalizer Performance
  • Date: Mon, 28 Aug 2006 20:04:44 +0100

> I don't know the document in question, but the statement
> below doesn't seem to necessarily imply each element having
> 1000 namespace prefixes in scope.

I only said "if"...

> Admittedly not the kind of input one might hope for, but at
> the same time, not something that should cause dramatic
> degradation or failure.

I don't think there's any "should" about it. There's no moral obligation for
software designers to handle pathological cases well (or at all); on the
contrary, one has to optimize for the most common case. I don't think it's
unreasonable to take the view that this example is pathological.

Saxon's NamePool at one time had a limit of 256 prefixes bound to a single
URI. Someone blew that limit and I raised it to 1024, because I had some
spare bits and other users weren't adversely affected. But the performance
with 1024 is still likely to be pretty atrocious; and if someone asks for it
to be raised to 65536 (say) then the answer is going to be sorry but no.

Of course if such usage became mainstream then I would have to reconsider.
Meanwhile, you can write a simple SAX filter to reduce the number of
prefixes.

Michael Kay
http://www.saxonica.com/





Archive powered by MHonArc 2.6.24.

Top of Page