RE : Re: saxon:try()

Colin Paul Adams wrote:

> Now, that is assuming that ex:error-state is an
> instruction. And indeed, you created it as such.

> So if ex:error-state is an instruction, then it should
> expect to see errors caused by evaluating variables within
> that instruction.

> But you seem to be arguing that the error trapping should
> be lexically associated. In which case it should be part
> of a declaration, not an instruction.

  Yes, I misused "lexicaly".  I had in mind the simplest
case:

    <xsl:variable name="v" .../>
    <ex:error-safe>
       <ex:try>
          <xsl:sequence select="$v"/>

  You are right, this is instead related to the "sequence
construction".

  But my point was that because the variable declaration is
not evaluated in the sequence constructor of the ex:try, the
error it could raise should not be caught (even if a
processor can evaluating the value of the variable while
evaluating the seq ctor).

  Regards,

--drkm























      _____________________________________________________________________________ 
Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail http://mail.yahoo.fr