Note from Paul Hunkar - sent April 19 2019

Not sure if you were aware of this work, but Archie is working for the certification group and generating the updated needed for the CTT to be able to calculate aggregates (and thus we would be able to implement all of the testing for aggregates). He is taking the .net code that Randy has to generate aggregates and converting it to C++ (calculation/status calc parts), since that is what the CTT needs (multi-platform and all). During his conversion he has noticed an issue – the Github project has all data for Historian2 using UseSlopedExtrapolation = true, but the specification has UseSlopedExtrapolation = false for all historians (at least that is what is in the released 1.04 spec). Is this something we changed as part of mantis issue? It would make sense to have one of the example include a sloppedExtrapolation, even though most of the aggregates would not change. If the table in the doc match what randy generate and UseSlopedExtrapolation = true I would just change the table definition to match. When I took a quick look most aggregates return Bad_NoData – so they do not use the flag. Only interpolated, TimeAverage, Total use the flag, so only the tables for these three aggregates for historian 2 for the single example that goes passed the collected time would not match what randy’s program generate. From my look at the spec it, it appears that the spec is following what is defined in it vs what the github project has.

Any thoughts would be good? We could easily change the spec since it is just an example.

Agreed to make this change -- needs 1.04 Errata.

Fixed in part 13 1.05.2

Historian 2 example has been reverted to have UseSlopedExtrapolation set to true. Two locations needed to be changed.
TimeAverage 12:01:30.000 is 96.250 and Interpolative 12:01:35.000 is 102.5.

Errata changes are in 1.04.8

Agreed to 1.05 changes and 1.04.11 Errata in Virtual F2F.