This document describes changes in each release of the SolarNetwork Reading Aggregate Validation Tool.
- Add new
--dry-runoption to run the validation without making any changes in SolarNetwork. - Add new
--generate-reset-datum-min-gap=Noption to support automatically creating Reset datum auxiliary records to cope with gaps in data.
- Better query rate limit throughput using
X-SN-Rate-Limit-Retry-Afterresponse header.
- More detailed reporting on task status: completed, partially completed, and never started states are reported instead of the general No validation problems found message.
- Use absolute dates in all API calls, to avoid issues with daylight saving time.
- Fix reading comparison to use reading statistics, not calendar properties. This resulted in false-positives when the two measurement styles differed.
- Better cope with time ranges that split on large gaps in data.
- When marking one hour/day stale to compensate for a higher-level aggregate mismatch, find the first hour of each day with actual data to mark as stale, otherwise marking an hour without any data will have no effect and the higher-level aggregate will not be reprocessed.
- Change the default for
--min-days-offsetfrom 1 to 5 to not follow current events so closely.
- The
--node-idargument is now optional. If not provided all node IDs that match the given--source-idvalues will be included. - SolarNetwork HTTP rate-limit errors are now handled by pausing a short amount of time before retrying the operation.
- Add new
--incremental-mark-staleoption, to mark each stream's invalid time ranges immediately after its validation completes (from finding all ranges or reaching the change limit or running out of time).
- Better native image support.
- Tidier and more consistent output formatting.
- Improve performance of day level hour-by-hour difference comparisons.
This is the first public release.