Apply the configured pre-join steps to the fetched products

Steps run in order, each against the product object holding the covariate it names, so a later step sees what an earlier one produced.

Usage

apply_prejoin_steps(per_dataset, config)

Arguments

per_dataset

a list of sf POINT objects, one per fetched product

config

a config list, as returned by load_config()

Value

a list of sf POINT objects, ready to join

Details

A step naming one covariate of a product that carries several splits it out first. datamatch::upscale_grid() returns only the columns it was asked to resample, so regridding chlorophyll in place would silently drop the other plankton variables sharing its dataset. Splitting means the rest stay on their own grid and reach the join intact.

See also

prejoin_steps() for the step types