Skip to contents

Process metadata file to extract variable information

Source: R/process_data.R
process_metadata_file.Rd

process_metadata_file() reads a metadata file and extracts information about the column names, column types, decimal positions, and variable definitions.

Usage

process_metadata_file(file)

Arguments

file

(String) The path to the metadata file.

Value

A list containing the scenario (e.g., "single", "single_multiple", "single_multiple_single") and a tibble with variable information.

Details

process_metadata_file() processes metadata files following specific rules to handle encoding, remove unnecessary information, and extract variable details. It detects the scenario based on the occurrence of double asterisks in variable names.

Examples

# \donttest{

  path <- tempfile("fcadata")
  dir.create(path)

  download_data(
    year = 2025,
    month = "September",
    dest = path
  )
#> Files successfully downloaded into /tmp/RtmpNBmo7f/fcadata406d2729792b

  process_metadata_file(file.path(path, "D_RC1.TXT"))
#> $scenario
#> [1] "single"
#> 
#> $vars_info
#> # A tibble: 58 × 7
#>    ColumnName ColumnType DecimalPosition Definition       MultipleOccurrenceCo…¹
#>    <chr>      <chr>      <chr>           <chr>            <lgl>                 
#>  1 SYSTEM     Numeric    0               System Code      FALSE                 
#>  2 DIST       Numeric    0               District Code    FALSE                 
#>  3 ASSOC      Numeric    0               Association Code FALSE                 
#>  4 MONTH      Numeric    0               Month of Report  FALSE                 
#>  5 YEAR       Numeric    0               Year of Report   FALSE                 
#>  6 UNINUM     Numeric    0               System, Distric… FALSE                 
#>  7 TYPRE      Numeric    0               Loans - Product… FALSE                 
#>  8 TYPPROINT  Numeric    0               Loans - Product… FALSE                 
#>  9 TYPCOOP    Numeric    0               Loans - Agibusi… FALSE                 
#> 10 TYPPROCMKT Numeric    0               Loans - Agibusi… FALSE                 
#> # ℹ 48 more rows
#> # ℹ abbreviated name: ¹​MultipleOccurrenceColumn
#> # ℹ 2 more variables: CodeColumn <lgl>, ColumnTypeSQL <chr>
#> 

# }