Mechanize users - Jun 2009 - ruby (mechanize , net/http)

Khomich Vlad wrote:
> 
> Hi guys.
>  I`m using mechanize to get some images from the web application. I have
> 2 problems:
>  1-st : I can`t catch the links. The whole link looks like <a href >
"https://.../reports/switch_accounts.html?..."> Full
report</a> , i can see it in the page source i get with mech
> 
> And there`re couple of these that i try to catch with a singe regexp
> 
> i tried this :
> <code>
> final.links_with(:href => %r{/switch_accounts.html/} ).each do |link|
> puts link.href, link.text
> end
> </code>
The "%r{/switch_accounts.html/}" part seems wrong. Maybe this works?
> final.links_with(:href => /switch_accounts\.html/ ).each do |link|
(not tested)

Vlad,

This regular expression:
>
> %r{/switch_accounts.html/}
>
doesn''t match this href:
>
> https://.../reports/switch_accounts.html?...
>
Because the regular expression has a trailing ''/'' after the
accounts.html. I think you could just change to this:

%r{/switch_accounts.html}

and get a match.

Also, something that I do which is useful for me, is to load a copy of
the page in irb, using something like this:

irb> page = Mechanize::Page.new(nil, {''content-type'' =>
''text/html''},
File.read(''my.html''), 200)

(From memory, may have a detail wrong.)

Then, you can test your mechanize code interactively against the page,
and make sure it''s working the way you expect, before you put it in
real code.

Aaron

Vlad,
> The second problem - how can i actually save the images to a selected
> folder(assuming i have the links for them)?
You don''t need mechanize for this. Just use open-uri.

Aaron