Skip to contents

Import and convert recodeflow variables and variable details metadata files to MockData configuration format

Source: R/import_from_recodeflow.R
import_from_recodeflow.Rd

Converts recodeflow variables.csv and variable_details.csv files into MockData configuration format (mock_data_config.csv and mock_data_config_details.csv). Filters variables by role and optionally by database.

Usage

import_from_recodeflow(
  variables_path,
  variable_details_path,
  role_filter = "mockdata",
  database = NULL,
  output_dir = "inst/extdata/"
)

Arguments

variables_path

Character. Path to recodeflow variables.csv file.

variable_details_path

Character. Path to recodeflow variable_details.csv file.

role_filter

Character. Role value to filter variables by. Only variables with this role will be imported. Default is "mockdata". Use regex word boundary matching to avoid partial matches (e.g., "mockdata" won't match "mockdata_test").

database

Character vector or NULL. Database identifier(s) to filter by. If NULL (default), extracts all unique databases from variables.csv databaseStart column. If specified, only imports variables that exist in the specified database(s).

output_dir

Character. Directory where output CSV files will be written. Default is "inst/extdata/". Files will be named mock_data_config.csv and mock_data_config_details.csv.

Value

Invisible list with two data frames: config and details

Details

Column Mapping

variables.csv -> mock_data_config.csv

Direct copy: variable, role, label, labelLong, section, subject, variableType, units, version, description (to notes)

Generated:

  • uid: v_001, v_002, v_003, ...

  • position: 10, 20, 30, ...

  • source_database: extracted from databaseStart based on database filter

  • source_spec: basename of variables_path

  • last_updated: current date

  • mockDataLastUpdated: current date

  • seed: NA

  • rType: NA (user must fill in: integer/double/factor/date/logical/character)

  • corrupt_low_prop, corrupt_low_range, corrupt_high_prop, corrupt_high_range: NA

  • mockDataVersion, mockDataVersionNotes: NA

variable_details.csv -> mock_data_config_details.csv

Direct copy: variable, dummyVariable, catStartLabel (to catLabel), catLabelLong, units, notes

Mapped:

  • recStart: copied from input recStart

  • recEnd: initialized to input recStart (user updates to: valid, distribution, mean, etc.)

Generated:

  • uid: looked up from config by variable name

  • uid_detail: d_001, d_002, d_003, ...

  • proportion: NA (user fills in - must sum to 1.0 per variable)

  • value: NA (user fills in distribution parameters)

  • sourceFormat: NA (user fills in for date variables)

Database Filtering

When database parameter is specified, the function:

  1. Filters variables.csv rows where databaseStart contains the specified database(s)

  2. Filters variable_details.csv rows where databaseStart contains the specified database(s)

  3. Sets source_database to the filtered database(s) in mock_data_config.csv

Examples

if (FALSE) { # \dontrun{
# Import all variables with role "mockdata" from all databases
import_from_recodeflow(
  variables_path = "inst/extdata/cchs/variables_cchsflow_sample.csv",
  variable_details_path = "inst/extdata/cchs/variable_details_cchsflow_sample.csv",
  role_filter = "mockdata"
)

# Import only from specific database
import_from_recodeflow(
  variables_path = "inst/extdata/cchs/variables_cchsflow_sample.csv",
  variable_details_path = "inst/extdata/cchs/variable_details_cchsflow_sample.csv",
  role_filter = "mockdata",
  database = "cchs2015_2016_p"
)

# Import from multiple databases
import_from_recodeflow(
  variables_path = "inst/extdata/cchs/variables_cchsflow_sample.csv",
  variable_details_path = "inst/extdata/cchs/variable_details_cchsflow_sample.csv",
  role_filter = "mockdata",
  database = c("cchs2015_2016_p", "cchs2017_2018_p")
)
} # }