Based on the documentation I read, it appears to me that the Parsing phase is comprised of the parsing pipeline, merging pipeline and the typing pipeline. Is this right? I also wonder at which stage the event is defined. Since the merging pipeline aggregates the lines, I assume that by the end of this merging pipeline, the event is defined. Is this right?
See if this page helps answer you questions:
http://docs.splunk.com/Documentation/Splunk/6.5.3/Admin/Configurationparametersandthedatapipeline
Interesting thing. These pages are a bit different. The second is How Splunk processes data through pipelines and processes
versus -