On Thu 6 May 2004 13:03, William Ballard wrote:
> On Thu, May 06, 2004 at 12:50:20PM -0400, Rick Pasotto wrote:
> > On Thu, May 06, 2004 at 09:23:35AM -0700, William Ballard wrote:
> > > Has anybody tried 'lynx google.com' lately?
> > >
> > > You get HTTP/1.0 400 Bad Request. It only started happening
> > > about 2 weeks ago.
> >
> > 'links google.com' works. The problem may be with lynx.
>
> Yeah, but lynx not supporting tables is sometimes useful to get a
> site to serve up a different version of the site, if I'm going to
> dump it and parse it. lynx worked up until last week. Do you think
> it's using HTTP/1.0 and google stopped supporting it? Haven't
> bothered tracing it.
>
> Incidentally, I use this script to get around google's text ads:
Maybe they caught on to your little trick and put an end to it? :) Or at
least tried to, since you seem to be spoofing the UA.
>
> wget -O- -U 'Mozilla/4.0 (compatible; Lynx 2.7.2)' \
> "http://www.google.com/search?q=$*"; 2>/dev/null | \
> sed -e 's/Froogle//g' -e 's/<img src=[^>]*>//g' \
> -e 's/.*\(<div>.*\)/\1/' \
> -e 's/href=\//href=http:\/\/google.com\//g' \
^
Does this line actually work? To me it looks like you're missing an
escape before the second '/' before the second 'href'.
> -e 's/\([^>]\)<br>/\1/g' > ~/search.html
> /x/b/moz/cur/firefox -remote "openURL(/home/tom/search.html)"
>
> I used to use 'lynx -dump' instead of wget and then I didn't need the
> <div> sed clause, because google wouldn't serve up the text ads if
> you didn't have tables.
>
> Try my little script, keep a small terminal open beside Moz and use
> this script for your google searches. It makes the results take up
> the whole screen again and strips out the text ads!
--
David P James
Ottawa, Ontario
http://david.jamesnet.ca
ICQ: #42891899, Jabber: davidpjames@jabber.org
If you've lost something, you had to lose it, not loose it.
Attachment:
pgp0tFBs9swRb.pgp
Description: signature