Skip to contents

Resolve one flagged review row (multi-analyte and/or multi-CAS)

Source: R/cleaning_pipeline.R
resolve_review_row.Rd

Single coordinated engine behind resolve_multi_analyte_row(). Splits the selected Name value and, when pairing = "position", assigns CAS values from spec$cas_parts across the resulting rows so a row flagged for both reasons cannot be split twice into garbage.

Usage

resolve_review_row(df, name_cols, row_index, spec, cas_cols = character(0))

Arguments

df

Cleaned data frame.

name_cols

Character vector of Name-tagged column names.

row_index

One-based row position in df.

spec

List describing the disposition:

  • name_action: "split", "keep", or "rename".

  • name_parts: split parts or rename value (character or newline/; text). For split, NULL/empty falls back to suggest_multi_analyte_parts().

  • cas_parts: CAS values to assign (character or newline/; text). Ignored unless pairing = "position".

  • pairing: "position" pairs name/CAS parts one-to-one (counts must match, or one side has length 1); "broadcast" (default) leaves CAS columns intact.

cas_cols

Character vector of CASRN-tagged column names.

Value

List with cleaned_data and audit_trail.