Re: Decryption Transform processing question

On Wednesday 29 May 2002 04:06 pm, merlin wrote:
> Right, but if you are constructing an XPointer then the
> dummy node still exists from an XPath-processing perspective
> so if you wish to write full XPointer exceptions, the form
> #xpointer(/dummy/Blah) will be needed. I think this is a
> rare and documentable case.

Ok, apologies for my brain block on the iteration, step 4 is very clear. 

> It is, as I say, contrived. The inner Data element containing
> arbitrary data is first encrypted (EncryptedData#2). Then
> EncryptedData#1 is generated and its ciphertext is put in a
> new Data element that happens to have the same Id as the one
> that is now encrypted. So there never was an invalid document.
> Now, run this through the decryption transform as currently
> specified and the behaviour is indeterminate.

Would doing something as simple as saying "select the first e in document 
order" for each iteration solve this problem?

Received on Wednesday, 29 May 2002 17:03:50 UTC