On Sat, 05 Mar 2005 Hrvoje Niksic wrote: > -D filters the URLs encountered with -r. Specifying an input file is > the same as specifying those URLs on the command-line. If you need to > exclude domains from the input file, I guess you can use something > like `grep -v'.
Hi Hrvoje, thanks - but grep is no suitable option. I'd have to combine it with any other perl/sed/awk first in order both to merge <a href tags longer than one line and split lines that have more than one href within, just to make sure that only the desired domains are listed. It's sad that filter options do work on -r only, but not on -i (as a special type of -r -l1). Application example: I got a mailbox file which includes those URLs. I'd like to download all from a certain site. One workaround might be to convert this mailbox to a basic html file and read if via http in order to force the -r -H -D branch, instead of using -D -F -i locally. Is there a easier solution in order to tell wget that -i is 'within' the recursive path? Thanks, Martin
