Read Paradox Database Files into R.

Rparadox: A Modern Interface for Reading Paradox Databases in R

Rparadox provides a simple and efficient way to read data from Paradox database files (.db) directly into R as modern tibble data frames. It uses the underlying pxlib C library to handle the low-level file format details and provides a clean, user-friendly R interface.

This package is designed to “just work” for the most common use case: extracting the full dataset from a Paradox table, including its associated BLOB/memo file (.mb).

Features

• Direct Reading: Reads Paradox .db files without needing database drivers or external software.
• Tibble Output: Returns data in the tibble format, which is fully compatible with the Tidyverse ecosystem.
• Automatic BLOB Handling: Automatically detects, attaches, and reads data from associated memo/BLOB (.mb) files.
• Character Encoding Control: Automatically handles character encoding conversion to UTF-8 and allows the user to manually override the source encoding for files with incorrect headers.
• Type Conversion: Correctly maps Paradox data types to their corresponding R types, including Date, Time (hms), Timestamp (POSIXct), Logical, Integer, Numeric, and binary blob objects.

Installation

# stable version from CRAN
install.packages("Rparadox")

You can install the development version of Rparadox from GitHub using the devtools package.

# install.packages("devtools")
devtools::install_github("celebithil/Rparadox")

Usage

Basic Usage: The read_paradox() function

The easiest way to read a Paradox file is with the high-level read_paradox() function. It handles opening the file, reading the data, and closing the connection in a single step.

# 1. Load the package
library(Rparadox)

# 2. Get the path to an example database
db_path <- system.file("extdata", "biolife.db", package = "Rparadox")

# 3. Read the data directly into a tibble
# This automatically finds 'biolife.mb' and handles data types.
biolife_data <- read_paradox(db_path)

# 4. View the data
print(biolife_data)
#> # A tibble: 28 × 8
#>    `Species No` Category      Common_Name `Species Name` `Length (cm)` Length_In
#>           <dbl> <chr>         <chr>       <chr>                  <dbl>     <dbl>
#>  1        90020 Triggerfish   Clown Trig… Ballistoides …            50     19.7 
#>  2        90030 Snapper       Red Emperor Lutjanus sebae            60     23.6 
#>  3        90050 Wrasse        Giant Maor… Cheilinus und…           229     90.2 
#>  4        90070 Angelfish     Blue Angel… Pomacanthus n…            30     11.8 
#>  5        90080 Cod           Lunartail … Variola louti             80     31.5 
#>  6        90090 Scorpionfish  Firefish    Pterois volit…            38     15.0 
#>  7        90100 Butterflyfish Ornate But… Chaetodon Orn…            19      7.48
#>  8        90110 Shark         Swell Shark Cephaloscylli…           102     40.2 
#>  9        90120 Ray           Bat Ray     Myliobatis ca…            56     22.0 
#> 10        90130 Eel           California… Gymnothorax m…           150     59.1 
#> # ℹ 18 more rows
#> # ℹ 2 more variables: Notes <chr>, Graphic <blob>

Handling Incorrect Character Encoding

If you have a legacy file where the encoding is specified incorrectly in the header, you can manually override it using the encoding parameter with read_paradox().

library(Rparadox)
# This tells the package to interpret the source data as CP866
data <- read_paradox("path/to/your/file.db", encoding = "cp866")

This ensures that all text fields are correctly converted to UTF-8 in the final tibble.

Advanced Usage

For more control, you can use the lower-level functions. This is useful if you want to inspect metadata before reading the full dataset.

library(Rparadox)
db_path <- system.file("extdata", "biolife.db", package = "Rparadox")

# 1. Open the file and get a handle
pxdoc <- pxlib_open_file(db_path)

if (!is.null(pxdoc)) {
  # 2. Get metadata without reading all the data
  metadata <- pxlib_metadata(pxdoc)
  cat("Number of records:", metadata$num_records, "\n")
  
  # 3. Read the actual data
  data_table <- pxlib_get_data(pxdoc)
  
  # 4. Always close the file when you're done
  pxlib_close_file(pxdoc)
}
#> Number of records: 28

Links

• pxlib C library: https://github.com/steinm/pxlib
• CRAN page: https://cran.r-project.org/package=Rparadox
• Bug reports: https://github.com/celebithil/Rparadox/issues.