Web Test Extraction and Validation rules

Now that most of our major features are complete, we’re focusing on the bug fixing and polish required to get us to RTM (after our upcoming beta, of course).  One area of polish that we’re currently lacking in is the assortment of built-in extraction and validation rules.  Keep in mind that you can always implement your own custom extraction and validation rules in any .NET language, but we want to cover as many of your scenarios as possible right out of the box.

In case you haven’t use them yet, extraction rules extract a value from an HTTP response for use later in the web test.  Some extraction examples include extracting a session ID from a hidden form field, an order number from a checkout page, or even a dynamically generated link that will be navigated to later in the web test.  Specifying what to extract is the hard part.  Some rules use a regular expression, some look for a particular tag/attribute, and others look in the HTTP headers for a value.

Validation rules, on the other hand, are designed to allow you to check whether an HTTP response was correct.  It’s easy to manually check this when playing back a single web test, but validation rules allow you to automate the process and validate responses during a load test.  Some validation examples include verifying a response came back in a reasonable amount of time, checking for certain words or IDs on a page, verifying the contents of a shopping cart, and making sure the server didn’t redirect to an error page.

We need your input!  What built-in extraction and validation rules do you expect us to have?  We’ve already received a request for an extraction rule that extracts a form field’s value and another request for some way to make it possible to extract the text between two tags.  What extraction and validation scenarios are we missing?

Comments (7)

  1. Hi,

    How about finding .Net control values? Eg Label1.text (something that nunitasp.net does by looking at the HTML).

    …and URL querystring parameter values.

    Best regards


  2. BillT says:

    It sure would be nice to have these extraction and validation rules available in the regular .NET API. Are they?

  3. MArcel de Vries says:

    I Totaly agree with Steve. It would be great if you offered all .NET ASP.NET controls as a validation object so we can validate them like in NUNitASP. This would realy boost productivity.

  4. jason buxton says:

    If you could encapsulate the combination of NUnitAsp and HtmlAgility, you’d have me hooked. NUnitAsp for the user controls, etc., and HtmlAgility for everything else. I’ve found that I can do just about anything I want to with these two tools in tandem, including frames and so on…

  5. Krishna says:

    Exposing the ASP.NET server control as objects would sure help, Scott’s earlier post on internal test cases did show similar samples using the DataXX controls..would be nice to have something similar..

  6. gpironi says:

    A new extraction rule that finds the text between two tags and returns multiple occurrences of those values that are found in a list would be helpful.