I posted this problem yesterday but didn't really understand how complex it was. Now I'm reposting the problem with the new requirements I've uncovered.<BR><BR>My goal is to extract links from HTML pages to store for validation and cataloging. The situtation is complicated by the fact that for pages that are in the same domain don't have an http:// on them with the proper domain name. <BR><BR>For example:<BR><BR>domain is http://www.churchguides.com<BR><BR>so a couple of example links might be<BR><BR><a href=addlink.asp>Add your link</a><BR><a href="news.asp?type=religious">Religious News</a><BR><a href="http://www.affiliateplan.com">Affiliate Programs</a><BR><BR>for these three sample links I would want to get back:<BR><BR>http://www.churchguides.com/addlink.asp<BR>http://www.churchguides.com/news.asp?type=religious<BR>http://www.affiliateplan.com<BR><BR>As you can see the local links complicate my problem. To make it more challenging I really need to know if a particular link is local to the domain or external. At times I may want to ignore one or the other when I am processing.<BR><BR>One more tricky aspect I'm wrestling with is that the page of the domain I'm on could be http://www.churchguides.com/default.asp or it could be http://www.churchguides.com or even http://www.churchguides.com/framework/webpromo.asp<BR><BR>Obviously this would change how the local pages would need to be added to the root. <BR><BR>I appreciate any help.