Export Complete Study Data to Disk — export_study

Writes a longitudinal_study object to a structured folder hierarchy. Each snapshot date becomes a subdirectory containing up to four subfolders:

Usage

export_study_data(
  study,
  output_dir = ".",
  study_folder = NULL,
  format = "csv",
  overwrite = FALSE,
  save_rds = FALSE,
  verbose = FALSE
)

Arguments

study: A longitudinal_study object (output of create_longitudinal_study or quick_longitudinal_study), or a multiple_longitudinal_studies object (output of create_multiple_longitudinal_studies). When a collection of studies is passed, each study is exported to its own subfolder under output_dir.
output_dir: Root directory under which the study folder is created. Defaults to the current working directory.
study_folder: Optional name for the top-level study folder. Defaults to study$study_id with characters that are invalid in folder names replaced by underscores.
format: Export format. Options are "csv" (default), "parquet", or "both". When "parquet" is used, the arrow package must be installed.
overwrite: If TRUE, existing files are silently overwritten. If FALSE (default), an error is raised when the target study folder already exists.
save_rds: If TRUE, an analytics_full.rds file is also written per snapshot, preserving any non-data.frame objects (workflow lists, summaries, etc.) that cannot be expressed as flat files. Defaults to FALSE.
verbose: If TRUE, prints progress messages. Defaults to FALSE.

Value

Invisibly returns the path to the top-level study folder (for a single study) or a named list of paths (for multiple studies).

Details

<output_dir>/<study_id>/
  <snapshot_date>/
    raw/          # Raw_*.csv  (from study$raw_data)
    mapped/       # Mapped_*.csv  (from study$analytics[[date]]$mapped)
    analytics/    # <metric>_<table>.csv  (from study$analytics[[date]]$results)
    reporting/    # Reporting_*.csv  (from study$reporting[[date]])

Folders are only created when the corresponding data actually exists. Non-data.frame leaves inside analytics are silently skipped; they can be preserved alongside the CSVs by setting save_rds = TRUE, which writes a companion analytics_full.rds per snapshot.