Read the operation log

get_profile_dir() / set_profile_dir() locate the interpolation profile log: four CSV files (interp_runs.csv, interp_stages.csv, interp_maps.csv, interp_params.csv) that every interpolate_model() call appends its per-stage / per-map / per-parameter timing and memory readings to (R heap, per-stage heap peak, OS process RSS and its process-lifetime peak). When unset, the tables land next to the operation log if one is set, else under <project>/logs/ when a project registry exists; with no sink resolvable nothing is written.

The operation log is an optional, project-wide record of what ran: set_log_file() names a CSV file, and from then on interpolate_model(), solve_scenario() (and through it solve_model()), and solve_recursive() append one line each — timestamp, op, object, status, duration_sec, details. read_log() reads it back as a tibble with parsed timestamps.

Levcost's internal mini-model solves are not logged (their scratch scenarios are dot-prefixed). Per-run solver detail lives in each run's run.yml (see scenario_runs()); the log records the sequence.

Usage

set_log_file(path = NULL)

get_log_file()

set_profile_dir(path = NULL)

get_profile_dir()

read_log(file = get_log_file())

Arguments

path

character, the file or directory to use; NULL clears the setting and restores the default.

file

character, the log file (default: the log_file option).

Value

a tibble, one row per logged operation (empty when the file does not exist).

See also

Other options: en_config_path(), en_config_show(), get_default_solver(), get_exchange_format(), get_storage_format(), isVerbose(), set_option(), set_path_builder(), set_registry_file(), set_reports_path(), set_scenarios_path(), set_solver_path()

Examples

if (FALSE) { # \dontrun{
set_log_file("energyRt_log.csv")
scen <- interpolate_model(mod) |> solve_scenario()
read_log()
} # }