Multi line checking
I want to compare the content of this policy, I want to use the whole policy as the token instead having it line by line. route-policy Policy_test if community matches-any (100) then drop else pass endif end-policy
Let's say that I have a 10 policies in a router. I want to create a test that validate the contents of this policy. The test case is call a policy and the policy name is an argument. The problem is, I can't do compare a multi line compare. I listed a policy but not in a single line but 5-15 lines per policy. I don't know what will be the best approach in here.
Do you need to use multi line regexp ![alt text][1] This could be done by using iTest custom rule, by clicking on add analysis rule on the response tab ![alt text][2] select custom and click on Extract Using: regular expression, In regular expression we have two options(portion of match(es) to extract) 1. Numbered group (
http://wiki.tcl.tk/986) Nothing but indicies to select in tcl, tcl support 0 to 9 numbered group, based on the our requirement may changed. 2. Full match This match compare full match in the regular expression, you may consider the according Based on your requirement you can use your own regular expression, here i use .+ to select what every displayed ![alt text][3] The output is look this ![alt text][4] [1]:
http://itestforums.spirent.com/storage/attachments/2290-MultilineSeletion.png [2]:
http://itestforums.spirent.com/storage/attachments/2291-Custom.png [3]:
http://itestforums.spirent.com/storage/attachments/2292-numberd_group.png [4]:
http://itestforums.spirent.com/storage/attachments/2293-MultilineSeletion.png Thanks
http://tclscripting.blogspot.in/
There are methods to do this in iTest; however, I would argue that doing a multi-line comparison may not yield reliable results nor give you the ability to specifically tell what's not right with the policy. Multi-line comparisons often fail due to issues with spacing and tabs. This can often lead to false negatives in testing. Instead, checking each "part" of the policy not only eliminates this issue, but also provides you with a way to determine exactly why the failure occurred. To do this, use the response maps to parse the response and provide you with a list of policies. You can use the policy name as the key. Then, you can use a series of analysis rules to test each part of the policy to ensure it's accuracy. I'd be glad to create a mock up of this if you'd provide your 10 policy response.
12 People are following this question.
