Skip to contents

Persist per-WSG output into the province-wide habitat tables

Source: R/lnk_pipeline_persist.R
lnk_pipeline_persist.Rd

Copies the per-WSG staging tables (<schema>.streams, <schema>.streams_habitat) from the working schema into the persistent province-wide tables (<persist_schema>.streams, <persist_schema>.streams_habitat_<sp>, one per species). Wide-per- species pivot — fresh's long-format streams_habitat (one row per segment-species) becomes one row per segment in each per-species table.

Usage

lnk_pipeline_persist(
  conn,
  aoi,
  cfg,
  species,
  schema = paste0("working_", tolower(aoi))
)

Arguments

conn

DBI connection.

aoi

Watershed group code (e.g. "LRDO").

cfg

An lnk_config object with cfg$pipeline$schema set.

species

Character vector of species codes to persist. Should match what lnk_persist_init() was called with — typically lnk_pipeline_species() output for the AOI.

schema

Working schema (per-WSG staging). Default paste0("working_", tolower(aoi)).

Value

conn invisibly.

Details

Idempotent: each call DELETEs all rows for the given AOI before INSERTing the fresh ones, so re-running a WSG cleanly replaces its data without affecting other WSGs.

Column projection is driven by cols_streams + cols_habitat (named vectors at the top of R/lnk_persist_init.R) — single source of truth shared with lnk_persist_init().

Call after lnk_pipeline_connect() in the per-WSG orchestrator, before computing any rollup queries that should reflect the final per-species classification.