[macem-release]

🤖 I have created a release beep boop

0.12.0 (2026-11-01)

⚠ BREAKING CHANGES

• CIFImportError gains the case inconsistentSymmetryOperations (an exhaustive switch needs it). CIFImport.Policy gains operationMetricTolerance (defaulted, source compatible). Files whose operation list does not fit the cell, is not closed or has another crystal class than the IT number are refused where they were read before; the metric check can be relaxed through the policy.
• CIFImportError gains the case inconsistentSymmetryOperations (an exhaustive switch needs it). CIFImport.Policy gains operationMetricTolerance (defaulted, source compatible). Files whose operation list does not fit the cell, is not closed or has another crystal class than the IT number are refused where they were read before; the metric check can be relaxed through the policy.

Features

• a CIF read warns when the operation loop and the Hall symbol disagree (#59) (ea54e1c)
• FriedelCentre, the centre of a pattern with a blocked beam (#53) (cec5d4b)
• FriedelCentre, the centre of a pattern with a blocked beam (py4DSTEM get_origin_friedel) (cec5d4b)
• Orientation from Euler angles, dot, dotOuter and distanceMatrix (#44) (c6c65a9)
• read Hall symbols, expand a Hall-only CIF with its group (#48) (2c72ae8)

Bug Fixes

• ACOM Metal argmax keeps the lowest bin on an exact tie (#63) (bc22dc5)
• PatternCentre.find refuses a blocked beam instead of returning a Bragg spot or the stop rim (#55) (b210444)
• refuse a CIF operation list that is not its space group (#49) (1319b34)
• ring profile fit stops runs that cannot improve (R 600 px from 24 s to 5 s) (#51) (2171e36)

Performance Improvements

• ACOM Metal path 1.9x faster at 600 beam directions, 3.2x at 2400 (#61) (e3a8508)

Fixtures and tests

• name why orix Misorientation.reduce misses the smallest angle (#43) (7354509)
• no corner of the ACOM zone is sparser; a pattern between templates lands within the hole (#60) (c426184)
• retire the apps-parity dumps, name tools, fixtures and tests by subject (#35) (f0dd2ec)
• ring-fit bias follows the gap to the neighbouring ring, shared by py4DSTEM (#52) (8ddbb98)
• ring-list shape tests moved in from an app (938731a)
• ring-list shape tests moved in from macTEM (#38) (938731a)
• the large ACOM off-grid failures are coincidence twins of the truth (#46) (b87b5d5)
• the nine review regression CIF headers say which structure each holds and what it tests (comment lines only; no pin or value changes) (8bd454a)

---

This PR was generated with Release Please. See documentation.