Re: convert links problem

[Hrvoje Niksic]

[EMAIL PROTECTED] writes:

> This is working well with one major problem.  Links, img tags,
> etc. are not being properly converted.  For example, if there is an
> img tag in the source file such as:
>
> <img src="../images/whatever.gif">
>
> Then it is converted to:
>
> <img
> src="http://www.somedomain.com/home/intadm/public_html/folder/whatever.gif";>

Does Wget download this GIF file?

The link conversion works both ways: the files that are downloaded are
converted to relative references (something like "../foo/bar.gif").
But the files that have not been downloaded are converted to full
links (e.g. "http://server/foo/bar.gif";).

> The domain is the source domain not the domain of the machine
> running wget, and the path, although correct should start as
> ~intadm... to work.

Wget probably found "/home/intadm" in some of the URLs.  Inspect the
input files and you'll almost certainly find it somewhere.

> Also, when a link to another page on a source document is to a
> different domain as such:
>
> <a href="http://www.thisdomain.com/some/path/file.htm";>
>
> it is converted to:
>
> <a href="http://www.somedomain.com/file.htm";>

That sounds like a bug.  Wget 1.7 is several years old.  Could you
please try to compile Wget 1.9 and see if it works better?