Saving and Reading a DuckDB Database in R

DuckDB is an open-source, columnar relational database that provides fast performance for both small-scale ad-hoc queries and large-scale analytics workloads. As its popularity grows, users are exploring ways to save and load data into the DuckDB database. In this article, we will delve into the process of saving a DuckDB database in R and reading from it.

Introduction

DuckDB offers several benefits over traditional relational databases, including:

• Fast query performance
• Low memory requirements for large datasets
• Support for columnar storage
• Integration with popular data science libraries like tidyverse

However, saving a DuckDB database in R can be challenging due to its unique configuration and connection mechanisms. In this article, we will explore the correct approach to saving and reading from a DuckDB database using R.

Saving a DuckDB Database

To save a DuckDB database, you need to create a DuckDB connection object (con) using the dbConnect function and pass it to the dbWriteTable function. The dbWriteTable function takes several parameters, including:

• con: the connection object to the DuckDB database
• name: the name of the table to save (in this case, "diamonds.dbi")
• data: the data to be saved into the table
• append: a logical value indicating whether to append new data to the existing table or overwrite it

Here is an example code snippet that saves a DuckDB database:

library(tidyverse)
library(duckdb)

# Create a DuckDB connection object
drv <- duckdb(dbdir = "database.duckdb")
con <- dbConnect(drv, dbdir = drv$dbDir)

# Load the data
diamonds <- dplyr::tbl(con, sql("select * from diamonds"))

# Save the table to the database
duckdb::dbWriteTable(
  con,
  "diamonds.dbi",
  diamonds,
  append = TRUE
)

Reading a DuckDB Database

To read from a saved DuckDB database, you need to create another connection object using the dbConnect function and pass it the path to the database file. You can then use the tbl function to access specific tables in the database.

Here is an example code snippet that reads from a saved DuckDB database:

library(tidyverse)
library(duckdb)

# Create a connection object to the database
drv <- duckdb(dbdir = "database.duckdb")
con <- dbConnect(drv, dbdir = drv$dbDir)

# Load the table from the database
tbl(con, sql("select * from 'diamonds.dbi'"))

Important Considerations

When working with DuckDB databases in R, there are a few important considerations to keep in mind:

• Data Type: Make sure that your data types match those supported by DuckDB. For example, datetime values should be stored as strings.
• Schema: The schema of the table must match the schema used when saving the data. This can be achieved using the sql function in combination with the tbl function.
• References: When reading from a saved DuckDB database, make sure to wrap references to tables or columns in single quotes (`’’).

Conclusion

Saving and reading a DuckDB database in R requires careful consideration of the connection configuration and data types. By following these guidelines and using the correct functions (e.g., dbWriteTable, dbConnect, tbl), you can efficiently save and read from your DuckDB databases.

Last modified on 2025-02-20