aiqclib.nrtqc.step2_run_qc package

Submodules

aiqclib.nrtqc.step2_run_qc.dataset_all module

This module defines the QCDataSetAll class, the concrete implementation of the NRT QC module’s step 2 (running the configured QC items) for Copernicus CTD data.

class aiqclib.nrtqc.step2_run_qc.dataset_all.QCDataSetAll(config, input_data=None)

Bases: QCDataSetBase

A specialized implementation of aiqclib.nrtqc.step2_run_qc.qc_base.QCDataSetBase that applies the configured QC items to all observations.

Variables:

expected_class_name (str) – The class identifier used for configuration matching.

Parameters:

• config (ConfigBase)

• input_data (DataFrame | None)

expected_class_name: str = 'QCDataSetAll'

aiqclib.nrtqc.step2_run_qc.qc_base module

This module provides the QCDataSetBase class, the core of the NRT QC module’s step 2: applying the configured QC items to the input data.

Each enabled QC item is resolved through the feature registry (the items are ordinary feature classes under prepare/features, registered with qc_-prefixed names), instantiated with its configured parameters, and run over the full input frame. The produced flag columns are joined back onto the data, yielding one column per item/variable combination.

aiqclib.nrtqc.step2_run_qc.qc_base.DEFERRED_QC_ITEMS: tuple = ('temp_to_psal',)

QC items that need the aggregated per-variable flags and therefore run during the flag aggregation step (step 3) instead of step 2.

class aiqclib.nrtqc.step2_run_qc.qc_base.QCDataSetBase(config, input_data=None)

Bases: DataSetBase

Base class for running the configured NRT QC items (step "qc").

Takes the validated input data from step 1 and applies every item enabled in the configuration’s qc_item_set — except the deferred propagation items, which need the aggregated flags of step 3. The result, stored in qc_data, is the input frame plus one flag column per item/variable combination.

Parameters:

• config (ConfigBase)

• input_data (DataFrame | None)

default_file_name: str

The default name for the intermediate flag file.

input_data: DataFrame | None

output_file_name: str

The resolved output path for the intermediate flag file.

qc_data: DataFrame | None

The input data with all QC item flag columns appended.

qc_item_columns()

Return the flag columns added by run_qc_items().

Returns:

The columns of qc_data absent from the input.

Return type:

List[str]

Raises:

ValueError – If run_qc_items() has not been run yet.

run_qc_items()

Apply every enabled QC item and append its flag columns.

Items are applied in configuration order. Each item receives the raw input frame (items are independent of each other’s flags), and its flag columns are joined onto the accumulating result by the observation keys. Deferred items (DEFERRED_QC_ITEMS) are skipped here and handled by step 3.

Raises:

ValueError – If an enabled item has no registered feature class (qc_{name} missing from the registry).

Return type:

None

write_qc_data()

Write the intermediate flag frame to a Parquet file.

Raises:

ValueError – If qc_data is empty, indicating the QC items have not been run.

Return type:

None