[
Lists Home |
Date Index |
Thread Index
]
> My quick reaction is it looks like you're proposing to validate the
> entire file to determine what schema to use for the file. Possibly many
> times. Have I misunderstood?
>
Actually, I have been answering in rush. The current program does not validate
the whole document, it stops validation as soon as it fails; that means that
for the same queries as implemented in nXML, it will parse the root element
if it fails, and the whole document if it succeeds, but the latter is only
once.
Furthermore, 'any' pattern can be easily recognized (will do it in a day)
so that it never parses the whole document if it is only asked to look
at its prologue.
For patterns which actually require parsing till the end, such as checking that
an element with particular namespace is present somewhere in the document, it
will parse and validate the whole document. But again, it is fast. 1 second per
5 Megabytes.
David Tolpin
http://davidashen.net/
|
