Removes pings that appear under more than one trip identifier, keeping a single copy.
Usage
drop_overlapping_pings(df, prefer = character(0))Arguments
- df
Prepared fishing points, as returned by
prepare_tracks_for_effort().- prefer
Character vector of trip identifiers whose copy of a shared ping should be kept. Defaults to none.
Details
Complements index_trip_duplicates(), which only removes trips duplicated
in full: when PDS moves part of a track to another trip the two files
overlap without being identical, and the shared pings would otherwise
contribute their time twice.
Pings are matched on the exact (timestamp, latitude, longitude) triple –
two vessels reporting the same position to the metre in the same second is
not a case that occurs in practice, and the triple is compared across trips
precisely because a retired trip carries no device metadata to group on. The
copy kept belongs to a trip listed in prefer where possible, then to the
lowest trip identifier, so the outcome does not depend on download order.
dt_hours is left as computed within each trip: deduplicating before the
intervals were known would turn the removed pings into artificial gaps and
inflate the survivors.
Only pings within the batch are compared, which is why
aggregate_pds_effort() recalls the trips already aggregated that could
share pings with it (recall_overlapping_trips()) rather than relying on
this function alone.