url_watch: watch for the appearance of strings in web pages and notify
           via email.

Usage: url_watch <options>

Options:
	-w '<email-list>'	List of email addresses to notify. (space
				separated)

	-u <url>		Add <url> to the urls to watch.

	-p <pattern>		Add <pattern> to the patterns to watch for.

	-f <follow_match>	If embedded URLs match <follow_match>,
				those URLs are searched as well.  

	-l <n>			Loop continously, waiting <n> seconds between
				checks.

	-s <n>			Sleep <n> seconds between URL fetches.

	-t			Trim HTML tags before checking for pattern
				matches.

	-H <file>		Hit file used to avoid repeat matches.
				It is read at startup. A hit is a match 
				*and* a found HREF="..." to the left of the
				match. The HREF url defines the hit, not the
				match itself. <file> has one URL per line.

	-r			Sets Must_Have_HREF.  If a match has no
				corresponding HREF then do not report it.
				Each HTML file is processed and trimmed
				from left to right, so it is possible for a
				match to not have an HREF to its left.
 
	-c <file>		Configuration file <file> (containing code like
				that at the top of this script) is sourced
				before each check iteration.  This allows
				resetting the parameters, URLs, etc without
				having to restart the program.
				File should end with "1;"

	-m <path>		Specify the path to the Unix sendmail program.
				(Default: /usr/lib/sendmail)

	-T <seconds>		Set the timeout for url fetching. See the
				comments in the Configuration section for
				details.

	-C '<command>'		Command to run whenever a match is found.
				See the comments in the Configuration section
				for details.

Notes:
	Use of -c <file> may override parameters supplied on the cmd line.