Skip to Content.
Sympa Menu

xom-interest - Re: [XOM-interest] Performance problems with very large attribute strings

xom-interest AT lists.ibiblio.org

Subject: XOM API for Processing XML with Java

List archive

Chronological Thread  
  • From: Elliotte Rusty Harold <elharo AT ibiblio.org>
  • To: XOM API for Processing XML with Java <xom-interest AT lists.ibiblio.org>
  • Subject: Re: [XOM-interest] Performance problems with very large attribute strings
  • Date: Sun, 6 Oct 2013 11:08:32 -0400

Off the top of my head, I think attributes are provided as simple strings
to XOM by the SAX parser, and XOM stores them as simple strings. I don't
think I've had occasion to figure out how Java handles 25MB string objects.

Which parser are you using? If it's not the default, XOM may be verifying
all the PCDATA in the attribute, which could take a while. This shouldn't
be happening unless you've swapped parsers; but it's worth making sure that
you're using fastAddAttribute instead of addAttribute i the parser does not
look like a bottleneck.




On Sat, Oct 5, 2013 at 6:32 PM, Peter Murray-Rust <pm286 AT cam.ac.uk> wrote:

> I am using XOM to parse SVG files which use
> <image xlink:href="data:image/png;base64,iVB..."/>
> to hold binary images such as PNGs. Some images require 25 Mbytes of data
> and take minutes to read in through Builder().
>
> Is there any workaround? I'd be happy simply to skip the very large images
> - would SAX be faster?
>
>
> --
> Peter Murray-Rust
> Reader in Molecular Informatics
> Unilever Centre, Dep. Of Chemistry
> University of Cambridge
> CB2 1EW, UK
> +44-1223-763069
> _______________________________________________
> XOM-interest mailing list
> XOM-interest AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/xom-interest
>



--
Elliotte Rusty Harold
elharo AT ibiblio.org




Archive powered by MHonArc 2.6.24.

Top of Page