it can't be done directly from wget. If the domain doesn't appear in the pages then sed or perl can be enough.
Another solution/hack can be to place the files you have downloaded on a local web server, that handles everything as a static file, in a way that "http://foo.com/bar/baz"; can be accessed as "http://localhost/bar/baz"; and fool your DNS so "foo.com" points to "127.0.0.1". At this point it will be enough to repeat the original wget command (and hope it will work). Better solutions are welcome. Cheers, Giuseppe Peng Yu <pengyu...@gmail.com> writes: > I have download a web directory (with option --mirror --timestamping > --no-parent). But I forget to specify the option --convert-links. > Is there a way to post-processing the already downloaded files to > convert the links without having to redownload all the files?
