"ted sandler" <[EMAIL PROTECTED]> writes:

> Likewise, should one create a new parser for each and every
> document?  One could imagine the parser's internal state getting
> severly confused by "good" documents as well.

I do. But I don't have all that many programs that burn through large
amounts of documents. Usually I'm parsing small numbers of very large
documents. 

The way I set it up to deal with large numbers of documents was to
trap the errors, and only create a new parser on an error, because
apparently the internal state is fine unless there's an exception.

jas.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to