Re: scripting an interactive web page

Steffen Schulz <[email protected]> Thu, 4 Mar 2010 14:59:06 +1100
Newsgroups gmane.org.user-groups.slug.chat
Message-ID <[email protected]>
On 100304 at 09:45, Ken Foskey wrote:
> On Tue, 2010-03-02 at 22:48 +1100, Voytek Eymont wrote:
> > I have access to a user/pswd authenticated web site, where, information
> > may or may not be present
> > 
> > so, I need to log on, and see if any information is present.
> > 
> > what can I use to auto check periodically, so, it can notify me if any
> > information is present?
> 
> As suggested, curl will get the file for you and you have to interpret
> the file once retrieved.
> 
> I have found the easiest solution to parsing HTML files is HTML::Parser
> in perl.  You hook up a method to the reader and check where you are in
> the file then extract the data you want.


Or if you like to work with shell script, 'xml2' will translate xml into
a line-oriented format for easy parsing with grep/sed etc.

Alternative to curl is to use good old wget, which might look like:

wget --post-data 'username=me&password=foo' --save-cookies session.tmp www.mysite.com/login.php
wget -O newpage --load-cookies session.tmp www.mysite/request.php

cmp newpage oldpage || echo mysite changed|mail -s "page changed" [email protected]
mv newpage oldpage


Easy peasy

/steffen
-- 
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
Subscription info and FAQs: http://slug.org.au/faq/mailinglists.html