Load generic-wide data¶

Goal: read data where each scan is a column, sharing one radius column.

Prerequisites: Installation.

The data file¶

wide-example/scans.csv

radius_cm,scan_001,scan_002,scan_003
5.90,0.0120,0.0080,0.0055
5.92,0.0185,0.0110,0.0075
5.94,0.0410,0.0240,0.0160
5.96,0.0925,0.0670,0.0450
5.98,0.1840,0.1510,0.1190

The manifest¶

Wide format requires an explicit column mapping. Metadata is never inferred from column-name conventions.

wide-example/manifest.json

{
  "schema_version": "1.0",
  "format": "generic-wide",
  "data_file": "scans.csv",
  "experiment": {
    "experiment_id": "wide-example-001",
    "name": "Wide-format example",
    "experiment_type": "sedimentation_velocity"
  },
  "defaults": {
    "optical_system": "absorbance",
    "signal_unit": "absorbance_unit"
  },
  "columns": {
    "radius": "radius_cm",
    "scans": [
      {"column": "scan_001", "scan_id": "scan_001", "elapsed_seconds": 0},
      {"column": "scan_002", "scan_id": "scan_002", "elapsed_seconds": 600},
      {"column": "scan_003", "scan_id": "scan_003", "elapsed_seconds": 1200}
    ]
  }
}

Each entry in columns.scans maps one data column to one scan and may carry elapsed_seconds, wavelength_nm, optical_system, rotor_speed_rpm, temperature_c, cell, channel and source_scan_id.

Load it¶

import openauc

experiment = openauc.load("wide-example")
print(experiment.observations.mode)     # RadiusAxisMode.SHARED
print(experiment.validate_structure())  # structural validation: OK (no issues)

Why wide is shared-axis only¶

A wide table has one radius column, so by construction every scan is read against the same radial positions. That is the format's defining property, not a limitation of the reader.

It follows that an experiment whose scans have different radius vectors cannot be written wide without placing them on a common grid — which means interpolating or resampling measured data.

openauc refuses rather than resamples.

from openauc.synthetic import (
    SyntheticExperimentConfig, generate_experiment, write_generic_wide,
)

ragged = generate_experiment(
    SyntheticExperimentConfig(scenario="per-scan-radius", n_scans=4)
)
write_generic_wide(ragged, "out/wide")

SyntheticWriteError: wide layout needs one shared radius axis, but this
experiment uses 'per_scan' axes. Writing it wide would require resampling onto
a common grid, which openauc never does; use write_generic_long or write_aucx
instead.

This follows ADR-0002: any operation placing per-scan data onto a common grid must be an explicit, opt-in, recorded transformation. No such operation exists in the library today.

Use generic-long or AUCX for ragged data — both represent per-scan axes natively.

Common failure modes¶

Error	Cause	Fix
`ManifestError` / missing `columns`	Wide format needs a column mapping	Add the `columns` block
`ParseError: missing required column`	A mapped column is absent	Check the header matches `column` values exactly
`DataConflictError: duplicate radius`	Repeated values in the radius column	Deduplicate at source
`SyntheticWriteError: wide layout needs one shared radius axis`	Ragged data	Use long or AUCX