In message <[EMAIL PROTECTED]>, Hassan 
Ali writes:
>When I am using the ORO package to find matches in a relatively big html page 
>(eg. amazon) it is giving me a stack overflow exception when I run it on Sun S
>parcV. The problem is that the request will run properly in the windows enviro
>nment. 
>Specifically the problem occurs when I try to read in the whole web page using
> 
>
>((?i)<html>(.*\n)*?.*?</html>)
>
>and then try to look for a match in the resulting buffer. 

You have to either increase the stack size used by your JVM (-Xss) or
rewrite your regular expression to be more efficient.  From what I can
tell (<html>.*</html>) compiled with CASE_INSENSITIVE_MASK | SINGLELINE_MASK
will do what you want more efficiently with no recursion.  I would suggest,
however, that it is faster to search for <html>, record the start offset,
search for </html>, record the end offset, and then copy the stuff in between
than it is to have the regular expression do all of that extra matching.

daniel



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

Reply via email to