maziggy 4436349a03 fix(stats): scope per-run filament to the printed plate, not the whole 3MF (#2614) ...

A single plate dispatched from a multi-plate 3MF could log the entire file's
filament against that one plate. When the AMS tracker measured nothing, a
completed run's PrintLogEntry.filament_used_grams fell back to
PrintArchive.filament_used_grams -- the sum over every plate (correct for the
archive card / project rollup, #1593) -- ignoring the archive's plate_id. So
each printed plate of a 22-plate file logged the full ~12 kg; cost inherited
the same whole-file value.

Forward: when the archive has a plate_id and its 3MF is on disk, the completed-
run fallback uses that plate's own slicer estimate (extract_plate_metadata_from_3mf)
and scales cost by the plate's share of the whole. Tracker-measured runs and
single-plate archives are unchanged.

Backfill: a startup migration repairs rows already written -- completed entries
whose stored grams exactly equal the archive's whole-file value, with a plate_id
and an on-disk 3MF, get recomputed to plate-scoped grams + cost. The exact-match
guard never touches tracker-measured or partial rows; idempotent, data-only,
identical on SQLite and Postgres, and logs the correction.

2026-07-20 11:52:18 +02:00

..

app

fix(stats): scope per-run filament to the printed plate, not the whole 3MF (#2614)

2026-07-20 11:52:18 +02:00

tests

fix(stats): scope per-run filament to the printed plate, not the whole 3MF (#2614)

2026-07-20 11:52:18 +02:00

__init__.py

…