Diffy relies too much in the content-type header

[camiloribeiro]

I was running some tests here to know how it works, and in one of my tests I have the following setup:

service primary:

• content-type header => application/json
• response-body => { "hello_world" : "primary" }

service secondary:

• content-type header => application/json
• response-body => { "hello_world" : "secondary" }

service candidate:

• content-type header => application/xml
• response-body => { "hello_world" : "candidate" }

In this scenario the diffy never checks the response and hangs forever.

Another test that I did could give you some more context on that issue. The test data in that case is the following:

service primary:

• response code => 401
• content-type header => application/json
• response-body => { "hello_world" : "primary" }

service secondary:

• response code => 402
• content-type header => application/xml
• response-body => { "hello_foo" : "secondary" }

service candidate:

• response code => 403
• content-type header => application/text
• response-body => { "hello_bar" : "candidate" }

In this case diffy holds for a long time, something around 40 seconds, and returns saying that everything is fine and there is no difference in the responses.

In a third scenario for this issue, if I use the same response code for primary and secondary, it returns fast as usual, but still without differences.

Please let me know if you have any questions regarding the scenarios. I am using my docker-compose setup to test it, so if you need feel free to clone it.

[puneetkhanduri]

Issue resolved in https://github.com/opendiffy/diffy