EpwMorpher consumes completed EsgStore extraction outputs and creates
future EPW files through a store-backed morphing workflow.
EpwMorpher$new()Create an EPW morpher.
EpwMorpher$new(
store,
epw,
site_id = NULL,
transform = monthly_transform("original_morphing"),
label = NULL
)storeAn EsgStore object.
epwBaseline EPW path, internal EpwFile, or an external
object inheriting from "Epw".
site_idOptional site identifier.
transformA reusable weather transformation created by a scale-specific transform constructor.
labelOptional source label.
EpwMorpher$preflight()Preflight EPW morphing inputs without writing store state.
EpwMorpher$preflight(
plan_id = NULL,
periods = NULL,
reference_plan_id = NULL,
reference_periods = NULL,
summary_id = NULL,
reference_summary_id = NULL,
observed_plan_id = NULL,
observed_periods = NULL,
observed_summary_id = NULL,
baseline_id = NULL,
by = c("source_id", "experiment_id", "variant_label", "period"),
strict = TRUE
)plan_idOptional extraction plan IDs.
periodsOptional period table from epw_morph_periods().
reference_plan_idOptional reference extraction plan IDs for change-factor backends.
reference_periodsOptional reference period table from
epw_morph_periods().
summary_idOptional climate summary ID.
reference_summary_idOptional reference climate summary ID for change-factor backends.
observed_plan_idOptional observed-weather extraction plan IDs.
observed_periodsOptional observed-weather period table.
observed_summary_idOptional observed-weather summary ID.
baseline_idOptional baseline summary ID.
byClimate grouping columns.
strictWhether required-data issues are errors.
EpwMorpher$summarise_climate()Summarise extracted climate data by period and month.
EpwMorpher$summarise_climate(
plan_id,
periods,
strict = TRUE,
overwrite = FALSE,
input_role = "model_future"
)plan_idExtraction plan IDs.
periodsPeriod table from epw_morph_periods().
strictWhether incomplete extraction coverage is an error.
overwriteWhether to replace existing rows for this summary.
input_roleSource role used to validate raw variables and
frequencies, one of model_future, model_historical, or
observed_reference. Workflow calls select this automatically.
EpwMorpher$plan()Create a morphing plan and monthly factors.
EpwMorpher$plan(
summary_id,
reference_summary_id = NULL,
observed_summary_id = NULL,
baseline_id = NULL,
by = c("source_id", "experiment_id", "variant_label", "period"),
strict = TRUE,
overwrite = FALSE
)summary_idClimate summary ID.
reference_summary_idOptional reference climate summary ID for change-factor backends.
observed_summary_idOptional observed-weather summary ID.
baseline_idBaseline summary ID. If NULL, baseline summary is created.
byClimate grouping columns.
strictWhether missing required variables are blocking errors.
overwriteWhether to replace an existing plan.
EpwMorpher$preview_plan()Preview a morphing plan and monthly factors without writing store state.
EpwMorpher$preview_plan(
summary_id,
reference_summary_id = NULL,
observed_summary_id = NULL,
baseline_id = NULL,
by = c("source_id", "experiment_id", "variant_label", "period"),
strict = TRUE
)summary_idClimate summary ID.
reference_summary_idOptional reference climate summary ID for change-factor backends.
observed_summary_idOptional observed-weather summary ID.
baseline_idBaseline summary ID. If NULL, baseline summary is created.
byClimate grouping columns.
strictWhether missing required variables are blocking errors.
EpwMorpher$run()Execute a morphing plan and write one hourly Parquet file per result member and weather year.
EpwMorpher$write_epw()Write one future EPW file for every persisted result member.
EpwMorpher$write_epw(
morph_id,
dir,
separate = TRUE,
overwrite = FALSE,
resume = TRUE,
reporter = NULL
)morph_idMorphing plan ID.
dirOutput directory inside the store root. Relative paths are
resolved under the store root. If NULL, the workflow stops
after writing morph result Parquet files and does not write EPW
outputs.
separateWhether to create case subdirectories.
overwriteWhether to overwrite existing EPW files.
resumeWhether to reuse complete existing EPW outputs.
reporterOptional workflow reporter used by task-level runs.
EpwMorpher$workflow()Run the store-native EPW morphing workflow.
EpwMorpher$workflow(
plan_id,
periods,
reference_plan_id = NULL,
reference_periods = NULL,
observed_plan_id = NULL,
observed_periods = NULL,
by = c("source_id", "experiment_id", "variant_label", "period"),
strict = TRUE,
dir = "outputs/future-epw",
separate = TRUE,
overwrite = FALSE,
resume = TRUE,
reporter = NULL
)plan_idExtraction plan IDs.
periodsPeriod table from epw_morph_periods().
reference_plan_idOptional reference extraction plan IDs for change-factor backends.
reference_periodsOptional reference period table from
epw_morph_periods().
observed_plan_idOptional observed-weather extraction plan IDs.
observed_periodsOptional observed-weather period table.
byClimate grouping columns.
strictWhether blocking diagnostics should abort the workflow.
dirOutput directory. Relative paths are resolved under the store root.
separateWhether to create case subdirectories.
overwriteWhether to overwrite existing plan, result, and EPW outputs.
resumeWhether to reuse complete existing result and EPW outputs.
reporterOptional workflow reporter used by task-level runs.
EpwMorpher$process_data()Read saved execution products for one morphing plan. The result is a
list named by case ID. Each entry contains the saved plan and case,
runtime options, backend factors, method-specific parts, runtime
diagnostics and provenance, plus the associated result manifest rows
in results. These are execution products, not preview factors.
Factor tables may summarize hourly values. Use method-specific parts
for hourly application parameters where supplied by the backend.
Empty or NULL products mean that the backend did not supply them.
This interface does not reconstruct unrecorded pipeline stages.
Multi-year members share one case payload. Reading checks its SHA256.
Older runs without saved process data require an explicit rerun with
overwrite = TRUE. No climate data are fetched by this method.