Package index • hvtiRdatabuild

Package Overview

Package-level documentation: what the package builds, who it’s for, and how the verify workflow fits together.

hvtiRdatabuild hvtiRdatabuild-package: hvtiRdatabuild: Build and Verify Analytic Datasets for the HVTI CORR Group

Verify: R Build vs. SAS Oracle

Freeze a SAS-built dataset as a checksummed parquet oracle, then compare an R-built dataset against it. Use snapshot_oracle() once per legacy SAS dataset to create the frozen reference; use compare_built() for every R rebuild you want to verify against it.

snapshot_oracle(): Freeze a SAS-built dataset as a checksummed parquet oracle
compare_built(): Compare an R-built dataset against a SAS oracle
print(<built_comparison>): Print a dataset comparison
str(<built_comparison>): Structure of a dataset comparison, without identifiers

Pull: Warehouse to R

Read a study’s configuration, open a credentialed warehouse connection, and pull the modules the study declares. The pull is read-only.

read_study_config(): Read and validate a study configuration
dw_connect(): Open a connection to the data warehouse
dw_modules(): List the available warehouse modules
dw_pull(): Pull the enabled warehouse modules for a study
print(<pull_result>): Print a pull result

Build: Assemble and Derive

Join the pulled modules into a study’s built dataset, run its study-specific derivation code, and carry the cohort’s inclusion and exclusion history through both in order.

build_dataset(): Assemble the patient-level dataset
derive_vars(): Apply study-specific variable derivations
new_cohort_record(): Create an ordered cohort history
append_cohort_stage(): Append an outcome stage to a cohort history
cohort_summary(): Summarize cohort attrition without patient identifiers
print(<build_result>): Print a built dataset result without patient values
str(<build_result>): Safely inspect a built dataset result
print(<derive_result>): Print a derived dataset result without patient values
str(<derive_result>): Safely inspect a derived dataset result
str(<cohort_record>): Safely inspect a cohort record

Analysis sets

Write and read declared, checkpointed selections of a study’s built dataset.

write_analysis_set(): Write an analysis set
read_analysis_set(): Read an analysis set

Dataset publication

Turn a mutable draft into an immutable dated release, or withdraw a release without changing its bytes.

publish_dataset(): Publish an immutable dataset release
withdraw_dataset_release(): Withdraw a published dataset release

Master datasets

Snapshot, lift and correct the master datasets studies read.

read_master_config(): Read a master dataset's configuration
snapshot_master(): Snapshot a master dataset to parquet, with its lineage
lift_master(): Lift a master's parquet snapshot into the warehouse as a view
backfill_corrections(): Backfill a master's corrections from its SAS build, and regenerate its view
propose_correction(): Propose a correction to a single cell of a master
decide_correction(): Decide a correction: accept, reject, supersede or bake it