Skip to Content.
Sympa Menu

shibboleth-dev - RE: SHIRE/SHAR/RM proposal

Subject: Shibboleth Developers

List archive

RE: SHIRE/SHAR/RM proposal


Chronological Thread 
  • From: Scott Cantor <>
  • To: 'Derek Atkins' <>
  • Cc:
  • Subject: RE: SHIRE/SHAR/RM proposal
  • Date: Thu, 27 Jun 2002 13:02:21 -0400
  • Importance: Normal
  • Organization: The Ohio State University

> No, I do not believe it supports aliased pointers. OTOH, my
> intention was to ship all the variable information (read:
> attributes) as an XML strings rather than parsed data. Yes,
> it means you need to reparse for every request, but my hope
> is that parsing the SAML isn't a high-overhead operation
> (compared to, say, the network latency).

I understand, I was just curious about ONC for my own reasons.

We haven't done any benchmarks, really. One obvious optimization is not
to revalidate, since the SHAR has already done it once. Might be able to
use SAX to parse the attributes also, which is cheaper than DOM.
Building a DOM is almost a requirement to keep OpenSAML from becoming a
nightmare to write, but just regulping fragments of known content isn't
too bad in SAX.

-- Scott

------------------------------------------------------mace-shib-design-+
For list utilities, archives, subscribe, unsubscribe, etc. please visit the
ListProc web interface at

http://archives.internet2.edu/

------------------------------------------------------mace-shib-design--




Archive powered by MHonArc 2.6.16.

Top of Page