how to cat only the http:// link into my webpage?

frenchn00b · 04-12-2009, 12:34 PM

Hello,

I am trying the cat the *.gif of the webpage.

So here how I do :

Code:

wget -k "http://passions.mettavant.fr/pdabureautique.htm"

Code:

sh script.sh pdabureautique.htm

with script.sh

Code:

#!/bin/sh
cat "$1"   | grep -o 'http:[^"]*'

but the problem is that the cat of hte htm page has src wihtout the full link/url ... without http...

Code:

...

  <tr>
      <td bgcolor="#F3F2FD" valign="top">
        <div align="center"><font face="Tahoma" size="1"><a href="#note"><img src="Images/pda/icones/carnetouvert.gif" width="34" height="34" border="0" style="filter:alpha(opacity=40)" onMouseover="makevisible(this,0)" onMouseout="makevisible(this,1)"></a><br>
          bloc notes<br>
          note taker - pad</font></div>

...

How to make it ??

I am blocked

unspawn, the king, any ideas?

thank you!

acid_kewpie · 04-12-2009, 02:43 PM

what does "cat the *.gif" mean???

what you want to do is exactly what the -k option does, and after testing it myself with the page you linked, it works just fine. does your version of wget actually support -k?

frenchn00b · 04-13-2009, 03:23 AM

Quote:

Originally Posted by acid_kewpie

what does "cat the *.gif" mean???

what you want to do is exactly what the -k option does, and after testing it myself with the page you linked, it works just fine. does your version of wget actually support -k?

I would like to wget teh page first.

then "cat" (list) allllll, completely all, the http links/url that are in the webpage

as u can see in this example, the gif files are not seen or listed by my script.sh

that's bit difficult to make it

any ideas?

best regards and happy eastern

Su-Shee · 04-13-2009, 03:34 AM

You want a combination of wget and lynx.

Lynx has the very handy feature of "lynx -dump" (just dumping _all_ links of a webpage) and lynx -image-links (dumping all image links).

So, of you want to fetch all images from webpage XY, you usally have to do a combination like this:

Code:

lynx -dump -image-links http://www.somewebsite.foo | \
for img in `egrep -o ".*\.jpg"`; \
do `wget -x  $img`; \
done

Check the website with lynx first, adjust the egrep-stuff and stuff it into wget and wrap a loop around it.

To just list all links of a page, just do lynx -dump.

I really do _love_ the power of the command line.

acid_kewpie · 04-13-2009, 03:38 AM

as I said, wget's -k option is exactly what you want, so if that's not what you're seeing you're doing it wrong or your version of wget doesn't support -k properly.

frenchn00b · 04-13-2009, 11:21 PM

Quote:

Originally Posted by acid_kewpie

as I said, wget's -k option is exactly what you want, so if that's not what you're seeing you're doing it wrong or your version of wget doesn't support -k properly.

thanks!

hmmm interesting... let's check it:

Code:

wget -k "http://passions.mettavant.fr/pdabureautique.htm"

Ok

Code:

$ wget --version
GNU Wget 1.11.4

Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GN

my try was without wget -k, apparently. :?
I get it working now with wget -k

So it seems we are getting few solutions.
I guess I also did try curl when I had access to my box... but it did do that -k ... dont recall