Operator catalog
| Operator | Versions | Executor | Trust | Downloads | Actions |
|---|---|---|---|---|---|
piflow_add_uuid_stopPiFlow · Data processing Prepend a distinct 32-character hexadecimal UUID per row. Params: column (new column name). | python9.2 KiB | 6 | |||
piflow_column_checkPiFlow · Data processing Validate full ordered column names against a reference and return left input. Params: none. Inputs: left, right. Left-input buffering limit: 128 MiB. Does not check physical units. | python8.6 KiB | 6 | |||
piflow_complement_by_memcachePiFlow · Data processing Replace selected columns on a lookup key hit. Rows without a matching key keep their original values. | python35.7 KiB | 4 | |||
piflow_consistency_checkPiFlow · Data processing Left-join reference keys; prepend flagField=0 for matches, 1 otherwise. Params: flagField, correlationColumn (comma-separated). Duplicate reference keys retain join multiplicity. 128 MiB input limit each. | python9.7 KiB | 6 | |||
piflow_convert_schemaPiFlow · Data processing Rename columns; does not cast data types. Params: schema, e.g. station_id->station,temperature->air_c. | python9.1 KiB | 6 | |||
piflow_data_sortPiFlow · Data processing Globally sort a bounded input on one or more fields. Ascending order places nulls first, and descending places them last. | python35.7 KiB | 4 | |||
piflow_distinctPiFlow · Data processing Deduplicate records across Arrow batches; retains the first row per key. Params: optional columnNames; empty means all columns. Key-state limit: 128 MiB. | python9.0 KiB | 6 | |||
piflow_drop_fieldPiFlow · Data processing Remove record columns incrementally. Params: columnNames (comma-separated string); absent names ignored. | python9.2 KiB | 6 | |||
piflow_mock_dataPiFlow · Data processing Random Arrow records. source is a trigger stream consumed and discarded (matching upstream unused default input); output independent of input. Params schema e.g. name:String,age:Int,optional:Double:true; count string0..1000000. Types String/Int/Long/Float/Double/Boolean. Batches1024. | python10.3 KiB | 5 | |||
piflow_reichstein_checkPiFlow · Data processing EcoFlow Reichstein u-star screening implemented without Spark: six equal-count temperature groups, twenty wind bins, Pearson<0.4, median valid threshold; flag nighttime low-wind FC. Params soilTemperatureField,frictionWindSpeedField,FC strings,flagValue int32,percentFlux numeric(0,1]. Excludes null/nonfinite/flagged samples from estimation; requires120 valid observations. Fixes upper-bin bound and deterministic threshold order. Bounded128MiB. | python11.4 KiB | 5 | |||
piflow_select_fieldPiFlow · Data processing Select named record columns in order, incrementally by Arrow batch. Params: columnNames (comma-separated string). | python2.6 KiB | 8 | |||
piflow_subtractPiFlow · Data processing EXCEPT DISTINCT over whole rows: return unique left rows absent from right. Nulls compare equal for this set operation. No params. 128 MiB input limit each. | python8.9 KiB | 5 | |||
script_jsonl_projectJSON · Data processing Select top-level fields from JSONL objects without buffering the complete stream. | python9.4 KiB | 1 |