Hi,
I have written a simple library that allows me to perform some basic statistical analysis on test results; things like mean execution time, standard deviation and percentiles. It is done relatively efficiently and only does the real processing at the end of the tests, writing a summary of the results to the standard log files. However, I believe there is a risk that these log files will cycle part way through the results being written. There is also the issue of separating out "debug" type information from the "results" type information in the file. (And, of course, there is one log file per Agent machine, although this is only a minor niggle.)
If there was a feature to allow scripts to write to a "summary" or "results" file (one per agent would be fine) and present the file(s) on the results screen for downloading this would meet this need very well.
(There may be some other means of allowing a summary to be presented to the tester which may be equally, or more, valid.)
I realise this is possibly not something that could be implemented in the short term, but might it be added to a wish-list please?
Kind regards,
Colin Main
Colin