utils.io_utils

utils.io_utils

IO Utils - ParquetWriter and Convenience Wrappers

Expose ParquetWriter and convenience functions for reading/writing tables.

Functions

Name	Description
get_writer	Get ParquetWriter instance.
read_table	Read table from parquet storage.
save_viz_data_json	Save visualization data as tidy JSON next to a PNG.
write_table	Write DataFrame to parquet storage.

get_writer

utils.io_utils.get_writer(base_dir: str | Path | None = None)

Get ParquetWriter instance.

Args: base_dir: Base data directory (uses default from config if None)

read_table

utils.io_utils.read_table(table: str, base_dir: str | Path | None = None)

Read table from parquet storage.

Args: table: Table name (e.g., “gold.facilities_production_annual”) base_dir: Base data directory (uses default from config if None)

save_viz_data_json

utils.io_utils.save_viz_data_json(
    df: pd.DataFrame,
    output_path: Path,
    orient: str = 'records',
)

Save visualization data as tidy JSON next to a PNG.

Args: df: DataFrame containing the data actually plotted output_path: Path to the PNG file (JSON will reuse the same stem) orient: pandas to_json orientation (“records” is best for Observable)

Returns: Path to the JSON file

write_table

utils.io_utils.write_table(
    df: pd.DataFrame,
    table: str,
    base_dir: str | Path | None = None,
)

Write DataFrame to parquet storage.

Args: df: DataFrame to write table: Table name (e.g., “gold.facilities_production_annual”) base_dir: Base data directory (uses default from config if None)