Ok, now this is definitely very weird. Here's a list of my actions and their
results

1) Run the command you gave 3 or 4 times. Get the error document every time
2) Notice that I'm running version 7.9.8 (or some such). Decide I need to
upgrade and then try again
3) Do real paying work for ~5 hours. Eventually get round to updating.
4) curl -V = "curl 7.10.1 (i686-pc-cygwin) libcurl/7.10.1 OpenSSL/0.9.6g
zlib/1.1.4"
5) Run the command. Success! Got the login page.
6) Decide to try from fresh. Delete cahoot.cookie
7) Run the command twice. Get the error document twice.
8) Run the command a further 20 or so times, occasionally pausing to delete
the cookie file. Get the error document every time

Hmm. I had it in my grasp, only to have it torn away. Curse you cahoot ;-)

I guess I'll try again from home tonight (we have a proxy server too) see if
I get different results.

Is your result reproducible?

Cheers,

Rich

-----Original Message-----
From: rmitchell_at_eds.com [mailto:rmitchell_at_eds.com]
Sent: 15 October 2002 06:53
To: curl-users_at_lists.sourceforge.net
Subject: Re: Beaten by cahoot.com

OK, this is probably a bit weird, but I just managed to get the "cahoot log
in"
page by doing this twice:

    curl -v -i -b cahoot.cookie -c cahoot.cookie
https://ibank.cahoot.com/servlet/com.aquarius.security.authentication.servle
t.LoginEntryServlet

First time I get back cookies and the "General Error Page". Second time I
get
the login page. I suppose that the cookies tell it something useful, but I
couldn't begin to tell you how the browser manages to get it right, unless
there's a cookie exchange before it really gets the page.

So, it works, but you have to do it twice, with cookies enabled both times.
I
am in sunny Oklahoma, though, going via corporate proxy, so your milage may
vary, contents may have settled during transit, mind the gap, etc... :)

I got curl-7.10.1 & openssl-0.9.6g here. They got an old copy of Apache
there:

    Server: IBM_HTTP_Server/1.3.12.6 Apache/1.3.12 (Unix)

Ralph Mitchell

Richard Cooper wrote:

> Hi,
>
> I would like to use curl to scrape statement information off my bank's
> website but I'm being beaten at the first hurdle.
>
>
https://ibank.cahoot.com/servlet/com.aquarius.security.authentication.servle
> t.LoginEntryServlet
>
> get me to the login page in my browser (IE 6) but
>
> curl
>
https://ibank.cahoot.com/servlet/com.aquarius.security.authentication.servle
> t.LoginEntryServlet
>
> returns an error page.
>
> -v doesn't produce anything helpful (to me)
> -A to pretend to be IE doesn't help
> the page doesn't seem to rely on having a specific referrer or cookies
>
> Any help would be greatly appreciated.
>
> Cheers,
>
> Rich
>
> --
> Richard Cooper
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
Received on 2002-10-15