Options
Testing scripts with dynamic data
Brian Donahue
Posts: 6,590 Bronze 1
Hi,
I noticed this with script tester -- if you parse any content from a webpage and use it in your test script, for instance if your web page returns something like a session ID in a hidden field and you want to post it back to the server, script tester will use stale data when you try to view the compiled script in IE.
For example, I test this forum using ANTS Load. PHPBB returns a session in a hidden field. I examine WebClient.HttpResponse.Content and pull the session ID from the GET request, so it can be posted to the server.
When the script 'compiles', the session ID is valid. But when you then step through the script in IE, PHPBB creates a new session ID, but the script tester is using the original one.
The result, in my case, is that my test script doesn't post new topics, for example, because the session ID is invalid, and PHPBB continually redirects me to the login page.
In ANTS Load, however, the script runs great and posts new topics.
Maybe I can find a workaround for this in the future! :-)
I noticed this with script tester -- if you parse any content from a webpage and use it in your test script, for instance if your web page returns something like a session ID in a hidden field and you want to post it back to the server, script tester will use stale data when you try to view the compiled script in IE.
For example, I test this forum using ANTS Load. PHPBB returns a session in a hidden field. I examine WebClient.HttpResponse.Content and pull the session ID from the GET request, so it can be posted to the server.
When the script 'compiles', the session ID is valid. But when you then step through the script in IE, PHPBB creates a new session ID, but the script tester is using the original one.
The result, in my case, is that my test script doesn't post new topics, for example, because the session ID is invalid, and PHPBB continually redirects me to the login page.
In ANTS Load, however, the script runs great and posts new topics.
Maybe I can find a workaround for this in the future! :-)
