Create ExecutionSettings from Config Block

Load database connection details and execution parameters from config.yml and secrets.yml. Schema info (CDM schema, work schema, cohort table, etc.) comes from config.yml. Credentials (dbms, user, password, server, port, connectionString) come from secrets.yml, keyed by the dbServer field in each config block.

Usage

createExecutionSettingsFromConfig(
  configBlock,
  configFilePath = here::here("config.yml"),
  secretsFilePath = "~/.picard/secrets.yml",
  cdmDatabaseSchema = NULL,
  workDatabaseSchema = NULL,
  tempEmulationSchema = NULL,
  cohortTable = NULL,
  databaseName = NULL,
  pipelineVersion = "prod",
  cohortTableSuffix = NULL,
  executionContext = NULL
)

Arguments

configBlock

Character. The name of the config block to load (e.g., "optum_dod")

configFilePath

Character. Path to the config.yml file. Defaults to config.yml in working directory.

secretsFilePath

Character. Path to the secrets.yml file. Default "~/.picard/secrets.yml".

cdmDatabaseSchema

Character. Override for CDM database schema.

workDatabaseSchema

Character. Override for work database schema.

tempEmulationSchema

Character. Override for temp emulation schema.

cohortTable

Character. Override for cohort table name.

databaseName

Character. Override for human-readable database name.

pipelineVersion

Character. "prod" (the default) or a MAJOR.MINOR.PATCH version means "use the configured production cohort table unchanged". Any other value (e.g. "dev", "develop_ml") is treated as a test namespace and routes to a suffixed cohort table. Ignored when executionContext is supplied.

cohortTableSuffix

Character. Optional suffix for cohort table names in non-semver (test) runs. Normalized to lowercase snake_case via normalizePipelineVersion(). If NULL, the non-semver pipelineVersion is used as the suffix. The derived table name is rejected if it exceeds MAX_TEST_COHORT_TABLE_NAME_LENGTH characters.

executionContext

An optional ExecutionContext for the current run. When supplied, its mode and normalized pipelineVersion control cohort table routing. The legacy pipelineVersion and cohortTableSuffix arguments remain available for direct callers.

Value

An ExecutionSettings object with populated connectionDetails

Details

Credentials are loaded from secrets.yml (default ~/.picard/secrets.yml). The config block's dbServer field is used to look up the server entry in secrets.yml. Schema fields continue to come from config.yml, with parameter overrides taking precedence.

secrets.yml supports two value formats:

• Plain strings: user: "myuser"

• R expressions: password: !expr keyring::key_get("picard", "server_pw")