Hi Germo,
If we are collecting all the links of a web page and navigating to
each link via click, don't you think you are sending http request
twice ? (via click and http.request_get)
Instead cannot we figure out page is broken or not without sending the
extra request? (dont want to verify page
Thanks for all the replys.
At first I was quite excited then realized they would make additional
calls - the cookies, session, conversation elements would all be a
nightmare. Not to mention that the load balancer may give valid
results sometimes and invalid the next (say if one server is bad).
Oh, you're right! I forgot to think anything about sessions and
cookies :/ In that way it should be indeed easier just to scan page
html...
But anyway, I don't see why such things are needed. Why won't you just
create normal tests, which assert multiple things and they would fail
for sure if
Maybe we need to make some distinctions here.
I view Watir as a 'browser driver' in the sense that it talks to the
DocumentObjectModel of a browser and doesn't care how the browser
deals with http protocol (well, we can maybe peek a bit). Watir API is
talking to the final composition of html, js,
I also wondered why this regular expression script was in Watir's
source, since it is actually really easy to get status with Net::HTTP.
Anyway, here are 2 examples I just did which could be used. First one
returns response status code and response class. It will throw
exception if web server
Ok, I played a little more so it would also work with https.
Just needed to add few extra lines to existing script:
def page_status url
url = URI.parse(url)
http = Net::HTTP.new(url.host, url.port)
if url.scheme.downcase == https
http.verify_mode = OpenSSL::SSL::VERIFY_NONE
Sorry, you also need to replace require 'net/http' with require 'net/
https'
Jarmo
--~--~-~--~~~---~--~~
You received this message because you are subscribed to the Google Groups
Watir General group.
To post to this group, send email to
can you get openssl to work on windows. I dont remember if this is easy or
hard.
Watir is different to net/http. If you did a watir request, and then a
net/http request, thats 2 requests. And you need to get all the cookies etc
from the browser into net/http. And you wont get any javascript or
Hi Paul,
Would suggest you use fiddler which is a free tool.(http://
www.fiddler2.com/fiddler2/)
You could run fiddler while your scripts are executing and capture the
status code, urls hit etc from a different process.
You could also try httpwatch (not free) http://www.httpwatch.com/
This
i could never find a way of getting the status code easily. If you look
through the watir source it does something like use a regular expression
against the page title to try and figure it out. At one point watir would
raise an exception when the page finished loading if it was an error, I
think
10 matches
Mail list logo
