[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: XSLT paradox - XSLT as one way encryption scheme?

• To: "'Jens Jakob Andersen, PDI'" <jens.jakob.andersen@p...>,<xml-dev@l...>
• Subject: RE: XSLT paradox - XSLT as one way encryption scheme?
• From: "Michael Kay" <michael.h.kay@n...>
• Date: Thu, 22 Nov 2001 10:41:04 -0000
• Importance: Normal
• In-reply-to: <E6CDEF4DFA31524D83641C6B9550AF77BFDF2A@e...>
• Reply-to: <michael.h.kay@n...>

I have a question/observation about XSLT.
 
If we look at the equation, where :
XML(1)*XSLT(1)=Output(1)
as well as
Output(1)/XSLT(1)<>XML(1)
 
we find that XSLT in fact is an (effective?) one way encryption scheme. 
 
I doubt it's a very strong encryption scheme, though that no doubt depends
on the stylesheet. I tend to think of a stylesheet as a function output =
xslt(input). And it's certainly a one-way function.
 
Is this intended behavious, should i t be so, and is this addressed in XSLT
2.0?
 
I think a reversible transformation would be a very different animal, and
would require a much more restrictive language. For example, how could you
possibly reverse the effect of the stylesheet:
 
<total xsl:version="1.0" xmlns:xsl="..." value="{sum(//item/@value)}"/>
 
Mike Kay

<<attachment: winmail.dat>>

• References:
  • XSLT paradox - XSLT as one way encryption scheme?
    • From: "Jens Jakob Andersen, PDI" <jens.jakob.andersen@p...>

• Prev by Date: looking for converters
• Next by Date: Re: Default namespace declaration in DTD and MSXML
• Previous by thread: XSLT paradox - XSLT as one way encryption scheme?
• Next by thread: looking for converters
• Index(es):
  • Date
  • Thread