Flag rows matching reference list entries

Two-pass matching: exact match first (tolower comparison), then substring match. Only active=TRUE reference entries are matched against. Match type and source are recorded in audit trail.

Usage

flag_reference_matches(df, name_cols, reference_list, flag_type, flag_label)

Arguments

df

Dataframe with name columns

name_cols

Character vector of Name-tagged column names

reference_list

Tibble with columns: term, source, active

flag_type

Either "warning" or "blocking"

flag_label

Human-readable label for the flag (e.g., "functional category")

Value

List with cleaned_data and audit_trail

Examples

df <- tibble::tibble(chemical_name = c("plasticizer", "dibutyl phthalate plasticizer"))
ref <- tibble::tibble(term = "plasticizer", source = "app_default", active = TRUE)
flag_reference_matches(df, c("chemical_name"), ref, "warning", "functional category")
#> $cleaned_data
#> # A tibble: 2 × 2
#>   chemical_name                 cleaning_flag                        
#>   <chr>                         <chr>                                
#> 1 plasticizer                   WARN: functional category [exact]    
#> 2 dibutyl phthalate plasticizer WARN: functional category [substring]
#> 
#> $audit_trail
#> # A tibble: 2 × 6
#>   row_id field         step         original_value              new_value reason
#>    <int> <chr>         <chr>        <chr>                       <chr>     <chr> 
#> 1      1 chemical_name flag_warning plasticizer                 WARN: fu… Match…
#> 2      2 chemical_name flag_warning dibutyl phthalate plastici… WARN: fu… Match…
#>