Diagnostic check: environmental site names present in data but missing from
the reference sites list are flagged – particularly new sites that also
lack coordinates, which usually signal a data-entry issue rather than a
genuine new site. Names are compared upper-cased and whitespace-squished
(embedded newlines collapsed) on both sides. data is returned unchanged,
with the unmatched sites attached as the "polis_new_sites" attribute, so the
check composes into clean_es() without writing to disk or touching global
state.
Arguments
- data
A cleaned ES data frame (canonical names) carrying
site_name.- sites
Reference site list: a data frame with a site-name column, or a character vector of known site names.
- site_col
Name of the site-name column in
dataandsites(default"site_name").- verbose
Emit a cli warning when unknown sites are found. Default
TRUE.
Value
data, unchanged, with attribute "polis_new_sites": a tibble with
columns site_name (character, the raw site label from data) and
no_coords (logical, TRUE when the site lacks a coordinate value,
NA when no coordinate column is present in data), one row per distinct
unmatched site.
Examples
es <- data.frame(
site_name = c("SITE A", "SITE B"),
site_y_coordinate = c(6.5, NA)
)
out <- validate_es_sites(es, sites = "SITE A")
#> ! 1 ES site not in the reference list (1 without coordinates); flagged, not dropped.
attr(out, "polis_new_sites")
#> # A tibble: 1 × 2
#> site_name no_coords
#> <chr> <lgl>
#> 1 SITE B TRUE