energy_analysis.chile_fleet.io#

Strict hourly schema validation and bounded, per-site Arrow scans.

energy_analysis.chile_fleet.io.read_table(path)[source]#

Read a small registry or CSV; production hourly inputs should be Parquet.

Parameters:

path (Path)

Return type:

DataFrame

energy_analysis.chile_fleet.io.normalize_ids(values)[source]#

Keep stable string identifiers; reject missing and whitespace-only IDs.

Parameters:

values (Series)

Return type:

Series

energy_analysis.chile_fleet.io.parse_utc(values)[source]#

Reject naive civil times, then normalize explicit offsets to UTC.

Parameters:

values (Series)

Return type:

DatetimeIndex

energy_analysis.chile_fleet.io.validate_hourly(frame, *, require_actual=True)[source]#

Validate site-hour keys without replacing zero, missing or negative values.

Parameters:
  • frame (DataFrame)

  • require_actual (bool)

Return type:

DataFrame

energy_analysis.chile_fleet.io.boolean_values(values, label)[source]#

Decode bool/0/1/string booleans without Python’s truthy-string trap.

Parameters:
  • values (Series)

  • label (str)

Return type:

Series

energy_analysis.chile_fleet.io.load_sites(path)[source]#

Preserve every registry row, including sites lacking coordinates or data.

Parameters:

path (Path)

Return type:

DataFrame

class energy_analysis.chile_fleet.io.SiteDataSource(path, *, require_actual=True)[source]#

Bases: object

Read one site at a time, using Arrow predicate pushdown for Parquet.

CSV is a convenience for small files (at most one million rows). Partition production datasets by site_id to avoid scanning unrelated row groups.

Parameters:
  • path (Path)

  • require_actual (bool)

site(site_id)[source]#
Parameters:

site_id (str)

Return type:

DataFrame

energy_analysis.chile_fleet.io.input_files(path)[source]#
Parameters:

path (Path)

Return type:

Iterator[Path]

energy_analysis.chile_fleet.io.fingerprint(path, *, contents=True)[source]#

Hash all source files once; no tokens or request URLs enter provenance.

Parameters:
  • path (Path)

  • contents (bool)

Return type:

dict[str, object]

energy_analysis.chile_fleet.io.json_safe(value)[source]#

JSON serialization with explicit nulls for nonfinite scientific values.

Parameters:

value (object)

Return type:

object

energy_analysis.chile_fleet.io.write_json(path, value)[source]#
Parameters:
  • path (Path)

  • value (object)

Return type:

None

energy_analysis.chile_fleet.io.row_mapping(frame)[source]#
Parameters:

frame (DataFrame)

Return type:

Iterator[dict[str, object]]

energy_analysis.chile_fleet.io.write_fleet_csv(frame, path)[source]#

Escape spreadsheet formula prefixes in labels; Parquet preserves exact IDs.

Parameters:
  • frame (DataFrame)

  • path (Path)

Return type:

None

energy_analysis.chile_fleet.io.validate_output_location(output, *inputs)[source]#

Keep generated Parquet out of a directory scanned as an input dataset.

Parameters:
  • output (Path)

  • inputs (Path)

Return type:

None