Summary

Updates ome_writers to consume the new plate_row/plate_col fields from useq positions and delegate all position naming to useq-schema.

Bridge useq plate_row/plate_col to ome-writers positions

• Propagate useq.Position.plate_row/plate_col → ome_writers.Position.plate_row/plate_column in _build_stage_positions_plan and _pos_with_grid_point.
• Handle both int (0-based index → letter/number) and str (pass-through) via new _plate_row_to_str / _plate_col_to_str helpers.
• Build a Plate object from position plate_row/plate_col annotations in _plate_from_useq when no WellPlatePlan is present.

Use useq grid position names (stop overriding)

• _build_well_plate_positions: uses well_pos.name (from useq's name_pattern) instead of inventing fov{idx}.
• _pos_with_grid_point: uses gp.name instead of overriding to fov{gp_idx} for plate positions.
• All position naming now originates from useq-schema's name_pattern field on grid plans.

Full pipeline

YAML (positions with plate_row/plate_col + grid_plan)
  → useq.MDASequence (names auto-generated from plate coords)
  → useq_to_acquisition_settings()
  → AcquisitionSettings (with Plate + plate-annotated positions)
  → OMEStream → HCS zarr

zarr structure: plate_row/plate_column/grid_name
  e.g., A/1/0000, A/1/0001, B/2/0000, B/2/0001

Example

# useq YAML config
stage_positions:
- x: 1000
  y: 1000
  plate_row: 0
  plate_col: 0
- x: 2000
  y: 2000
  plate_row: 1
  plate_col: 1

grid_plan:
  rows: 2
  columns: 2
  fov_width: 180
  fov_height: 180

Produces zarr paths: A/1/0000, A/1/0001, ..., B/2/0000, B/2/0001, ...

Depends on: pymmcore-plus/useq-schema#268

Test plan

• All 128 existing tests pass (1 pre-existing failure excluded)
• Updated well_plate_with_points test expectations (fov0 → 0000)
• Updated test_well_plate_fov_folder_names zarr path expectations
• End-to-end: verify HCS zarr output with plate-annotated positions + grid_plan

🤖 Generated with Claude Code