perl -W hgrepurl.pl http://www.senopt.com
Subroutine Cwd::fastcwd redefined at c:/ActivePerl/site/lib/Cwd.pm lin
+e 812.
Subroutine Cwd::getcwd redefined at c:/ActivePerl/site/lib/Cwd.pm line
+ 812.
Subroutine Cwd::abs_path redefined at c:/ActivePerl/site/lib/Cwd.pm li
+ne 812
main::get_html() called too early to check prototype at hgrepurl.pl li
+ne 27.
[download]

perl -c hgrepurl.pl http://www.senopt.com
hgrepurl.pl syntax OK
[download]

Viewing the source of http://www.senopt.com shows "links" like:

<a href="senopt/VPS/vps.html" target="_blank">vps</a>
<br>
<a href="senopt/DogTraining/dogtraining.html" target="_blank">dog trai
+ning</a>
[download]

You're going to have to do a bit more work, I'm afraid. You may want to start by reading More robust link finding than HTML::LinkExtor/HTML::Parser?.

HTH,

planetscape

Do you want to say that senopt.com does not have real links?

I tried

perl hgrepurl.pl http://www.txtlinks.com/
[download]

which is quite encouraging.
txtlinks is a web directory and it returns first a portion of url encoded stuff, like:

...http://www.txtlinks.com/%3Chtml%3E%0A%3Chead%3E%0A%20%20%3Ctitle%3E
+TXT%20Links%20Pure%20Links%20Di.......
[download]

without line brakes. Then it returns a bunch of normal links from this site.
What is the first portion?
Is it possible to get all real links residing in a site on all depth levels with this program?