functional tests

Jesus Rodriguez jesusr at redhat.com
Thu Feb 25 18:51:13 UTC 2010


On Thu, Feb 25, 2010 at 10:32:51AM -0400, Devan Goodwin wrote:
> May as well ask the question, do we want to keep two levels of
> functional testing? I've got mixed feelings on those Java tests. On
> the one hand I like that we have control over the database and code
> internals for test setup and such. On the other I don't like that it's
> deploying in a separate manner that ends up not showing problems we
> later see when we go deploy into Tomcat and JBoss.
> 
> Personally I'll likely be doing my functional/integration testing via
> the Python suites, I like knowing it's hitting an actual deployed
> instance of CP and it gives me better access to what the JSON/XML
> looks like as it's getting passed around.

Makes sense to me, the purpose of the functional tests is to show
a deployed service and run against its api external like a client
would. The 'functional java tests' don't really do that, it's a made
up scenario.

And I bet we could easily add a task to run the python functional tests
from buildr like we did with checkstyle and schema generation.

The one thing the java functional tests have is you don't have to 
deploy candlepin first, it just runs.

-- 
jesus m. rodriguez          | jesusr at redhat.com
principal software engineer | irc: zeus
red hat systems management  | 919.754.4413 (w)
rhce # 805008586930012      | 919.623.0080 (c)
+---------------------------------------------+
|   "Those who cannot remember the past       |
|    are condemned to repeat it."             |
|                        -- George Santayana  |
+---------------------------------------------+



More information about the candlepin mailing list