traverse and download 1

[Mag0007]

is it possible to give a URL to a script, and it will pick out the link for a ".doc" or ".xls" and download it?

For example.

% find_file ".doc .xls"

http://local_site/foo_docs_1.htm

and it will goto foo_docs_1.htm and look for .doc and .xls files and download it for me?


thanks in advance

 

[p5wizard]

I'd write a script based on wget. Depending on how easy it is to grab the URLs needed to download the documents it would not be that hard.

first use wget to put the origina html document in a unix file and see what you need to awk.grep out the URLs for the documents, then use wget to doanload the documents one by one.



HTH,

p5wizard

 

[feherke]

Hi

[tt]wget[/tt] with the -A or --accept option :

wget -r -p -A.doc,.xls [URL unfurl="true"]http://local_site/foo_docs_1.htm[/URL]

[tt]lynx[/tt] with the -dump option, then [tt]wget[/tt] :

lynx -dump [URL unfurl="true"]http://local_site/foo_docs_1.htm[/URL] | \
sed '1,/References/d; s/.* //' | \
grep -E "\.(doc|xls)$" | \
wget -i -

The second may look strange, but sometimes that was what I need.

Feherke.

http://rootshell.be/~feherke/

 

[Mag0007]

thanks for the replies...I did mine something like this


lynx -dump

http://www.somepage.com/foo.html

| grep .xls | awk {'print $2'}|xargs wget

this seems to work, but i like your way too!


thanks!