Skip to content

Latest commit

 

History

History
62 lines (39 loc) · 2.44 KB

File metadata and controls

62 lines (39 loc) · 2.44 KB

SolarNetwork Reading Aggregate Validation Tool changes

This document describes changes in each release of the SolarNetwork Reading Aggregate Validation Tool.

New features

  • Add new --dry-run option to run the validation without making any changes in SolarNetwork.
  • Add new --generate-reset-datum-min-gap=N option to support automatically creating Reset datum auxiliary records to cope with gaps in data.

Other changes

  • Better query rate limit throughput using X-SN-Rate-Limit-Retry-After response header.

New features

  • More detailed reporting on task status: completed, partially completed, and never started states are reported instead of the general No validation problems found message.

Fixes

  • 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.

Other changes

  • Change the default for --min-days-offset from 1 to 5 to not follow current events so closely.

New features

  • The --node-id argument is now optional. If not provided all node IDs that match the given --source-id values 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-stale option, 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).

Other changes

  • 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.