Implement a more complete "driver" for the ResultParser(s)
centis opened this issue · 0 comments
centis commented
With the current implementation, any new results from DiskSpd must be added to both XmlResultsParse and ResultsParser. This is error-prone, and it seems there are instances where the output of the two is not consistent.
IResultParser is a good start, but it only implements a single top-level API. It would be better to implement a single ResultParser base class that drives the overall structure of the output and each derived class just adds/outputs values.
It might also be nice to change the suffix to "Generator" or something - it's not really a parser.