Sample Script
	SS_URLs

Purpose
	Collects URLs referenced by a URL.

Source Code
#####################################################################
# SCRIPT: SS_URLs
#
# This script extracts a list of URL addresses from one URL.
# The URL is assigned using FVA (Forward Variable Assignment)
# for str variable $URL. The value of $URL is of form
#	"http://www.xxx.yyy"  or  "http://xxx.yyy.zzz/www.../qqq.html"
#
# The list of domains to be ignored is passed using FVA for str variabls
# $ignore_domains. The format is <domain>|<domain>|<domain> ...
# Each <domain> is in the form "http://www.abc.def" .
#
# All URLs and domains passed to this script need to be in the standard format
#         http://www.abc.def/ or http://www.abc.def/...../page.html
# The http:// part IS REQUIRED. The page name extension can be anything such as
# .html, .aspx, .js, etc.
#
# The list of URL addresses found on this web page is written
# to output stream, one URL address per line.
#
# This script can be stored, and edited as needed, in a file called
# SS_URLs.txt. The script can then be called as
#
#	script SS_URLs.txt URL("<URL>") ignore_domains("<domain>|<domain>|...")
#
#####################################################################

var str URL		# Name of the URL
var str ignore_domains	# List of domains to be ignored. Domains are separated by |.

var str foundURL	# URLs found, one at a time

var str content		# Collects the contents of the URL here.
			# $content is progressively cut down as each new URL is found.

repro $URL >$content

# We will change the value of $wsep to suit our purpose.
# But, we will save the original value, so we can restore it
# after we are done.
var str saved_wsep
set $saved_wsep = $wsep
set $wsep="<> \t,;?&\\\"\r\n"

while ( { sen -c "^href=\"^" $content } > 0 )
do
  # Strip off the portion upto and including "href=\""
  # We will use case-insensitive so it can match any of href, HREF, etc.
  stex -c "^href=\"^]" $content > null

  # The remaining $content now starts with a URL address followed by
  # a double quote (") or a question mark (?).

  wex "1" $content > $foundURL	# Get the URL address

  # Remove anything beginning with #.
  stex "[^#^" $foundURL >null

  # Process this URL further only if it is not empty, and it is not the same URL as $URL ?
  if ( ($foundURL <> "") AND ($foundURL <> $URL ) )
  do

    # Output the found URL. But first check if this is
    # an absolute or relative URL.
    var str fullURL
    var str s
    set $s = { stex -c -p "[^http:/^" $foundURL }
    if ( $s == $foundURL )
    do
      # This is absolute URL.
      set $fullURL = $foundURL
      done
    else
    do
      # Prepend URL with our own address.
      set $fullURL = $URL+"/"+$foundURL
      done
    endif

    # Cure the format of this URL.
    script SS_CureURL.txt URL($fullURL) > $fullURL

    # Check if this URL is to be ignored. The SS_IgnoreURL will assign an
    # empty value to $fullURL if this URL is to be ignored.
    script SS_IgnoreURL.txt URL($fullURL) ignore_domains($ignore_domains) > $fullURL

    # Is this URL (name) empty ?
    if ( $fullURL <> "" )
      echo $fullURL
    endif

    done
  endif

  done

# Restore original value of $wsep.
set $wsep = $saved_wsep