We need to create a more comprehensive test for the ConvertToEvent class so that we can accurately determine if the new method developed for OpenNMS 18 is faster. Here's what I would like to see in this test:
- Create the test in a branch from 'master' so that it can be run against the OpenNMS 17 code.
- Add a test syslog message file containing 200 - 1000 sample syslog messages.
- Write the test so that it feeds the sample syslog messages into the ConvertToEvent method in a loop. Test the method thousands of times and calculate the mean time to process each message (preferably with nanosecond precision). It may also be useful to use a dropwizard-metrics histogram.
- Merge the test forward into 'develop' and compare the performance of the latest code with the OpenNMS 17 code in 'master'.