> But that is not what httrack does. 
> 
> It downloads only ONE copy of each file, no matter
> how many times it appears in the mirror. (Think site
> icon or css files) It then modifies each html, js,
> or css file to point to the where it put the file. 
> 
> Now supposed two different pages have two different
> image.jpg The browser would overwrite the file and
> one page would display differently, where as httrack
> would rename one file and adjust the reference.

It doesn't matter how many copies HTT downloads, that was just an example to
explain it's a simple function to compare what's on one's HDD versus what
still needs to be downloaded. You can also compare timestamps and file size to
check for duplicates as well.

In other words, all the files exist on the HDD in the same directory structure
as on the website. So when HTT went to download a file, it would check the
destination path and see there is already a file there with the same
information as the one it is attempting to retrieve. Now if you wanted to
replace broken files, HTT would compare filesize/timestamp of the one on the
drive with the one on the remote server, see they don't match, and replace the
broken file with the most current one on the site (or whatever option you have
set).