That works for pages downloaded with Httrack, but it doen't work for pages
downloaded with other utilities (default for other utilities is relative URL;
default for HTtrack is absolute URL).

I'm thinking it's something with the scanning rules. 

A html file downloaded to disk will link to...

With -file:///C:\my web sites\blahblah

but I want Httrack to read the file as <http://blahblah/blahblah> instead. The
problem is that Httrack can't capture some pages that are in the "deep web",
whereas DownThemAll can.