Skip to contents

Strip terminal enclosures (parentheticals and brackets) from name fields

Source: R/cleaning_pipeline.R
strip_terminal_enclosures.Rd

Removes terminal (...) and [...] from Name-tagged columns, with protection for chemical names containing "yl" (except exception words), percentages, Roman oxidation states, recognized identity tokens, and enclosures attached directly to the name with no preceding whitespace (e.g. Cyclo(L-Phe-L-Pro)). Preserves stripped content in formula_extract_{source} columns.

Usage

strip_terminal_enclosures(df, name_cols)

Arguments

df

Dataframe with name columns

name_cols

Character vector of Name-tagged column names

Value

List with cleaned_data, audit_trail, and new_tags (empty list)

Details

Recognized identity tokens protect the entire enclosure, including mixed annotation text. Supported forms include 13C/14C, 15N, 17O/18O, 2H/3H, 34S and 37Cl with optional one- or two-digit counts, and D/d with a count. Isotope symbols are case-sensitive; ring-/U- prefixes are case-insensitive, with optional spaces between isotope components. Congener codes include Parlar/P, TMX, Hp-Sed/Hx-Sed, Andrews-Vetter B codes, T/Tox, and PCB/BZ/CB/PBDE/BDE/PBB/IUPAC numbers, matched case-insensitively. Congener tokens cannot immediately follow a hyphen in a longer name; name-attached isotope tokens remain supported. Unsupported synonym enclosures such as Tryptophan-P-1 follow ordinary annotation stripping. Single-digit charges with a preceding or following sign are also protected. Numeric comma lists ending in a hyphen (e.g. 1,3- or 2,4,6-) are protected as positional locants, not charges. Comma spacing affects recognition only; single locants, primed locants, and general structural-name parsing are not included in this extension. Reporting-basis phrases such as "as PCB6" are preserved separately from individual-congener evidence, including mixed annotations. The observed PCBtotal qualifier has no established composition in the source workbook; preserving it does not certify congener identity or aggregate composition. Recognition does not rewrite the original enclosure text.

Examples

df <- tibble::tibble(chemical_name = c("Acetone (ACS reagent)", "ethanol (ethyl alcohol)"))
strip_terminal_enclosures(df, "chemical_name")
#> $cleaned_data
#> # A tibble: 2 × 2
#>   chemical_name           formula_extract_chemical_name
#>   <chr>                   <chr>                        
#> 1 Acetone                 ACS reagent                  
#> 2 ethanol (ethyl alcohol) NA                           
#> 
#> $audit_trail
#> # A tibble: 1 × 6
#>   row_id field         step                      original_value new_value reason
#>    <int> <chr>         <chr>                     <chr>          <chr>     <chr> 
#> 1      1 chemical_name strip_terminal_enclosures Acetone (ACS … Acetone   Strip…
#> 
#> $new_tags
#> list()
#>