XML parsing performance

Takuki Kamiya kamiya at rp.open.cs.fujitsu.co.jp
Thu Oct 28 06:41:51 BST 1999


David Brownell wrote:
> 
> Vilya Harvey wrote:
> > 
> > >  2) Using attribute's instead of nodes.
> > 
> > This is probably a bit of a red herring, although it may have some (fairly
> > negligible) impact depending on the parser you use.
> 
> Depends on the DTD you use, actually.  I think attributes are more
> expensive to parse; they certainly need normalization and defaulting,
> which elements don't use.  Parsers fill out an auxiliary data structure
> and then scan it ...
> 
> Don't get me wrong, this shouldn't be a design consideration, but I
> do believe that there's a minor price to pay in _all_ parsers.
> 

Many parsers process "attribute XML without DTD" files in about 10 percent
less time compared to the time for "element XML without DTD" in my experience.
(On the other hand, file size ratio was about 1 : 1.3)

I found that there are a few parsers inclusing Microsoft's whose performance
improves significantly (about as half the time) for attribute XML files. I
do not believe it is coincidence that those parsers appear not to normalize
attribute values at all.

Another caveat is that some parsers' performance degrades a lot for attribute
XML files. I guess those parsers have not been tuned for the heavy use of
attributes.

= Takuki Kamiya  Phone: (045)476-4586 Fax: (045)476-4749   =
= FUJITSU LIMITED (COINS:7128-4217 NIFTY:HHA01731)         =


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)





More information about the Xml-dev mailing list