R

The ggsql R package provides bindings to ggsql, allowing you to run ggsql queries directly from R and render visualisations from data frames. It also contains a knitr engine for supporting ggsql blocks, with facilities for bidirectional data flow between R, Python, and ggsql blocks.

Installation

ggsql can be installed from CRAN with:

pak::pak("ggsql")

To get the development version, change “ggsql” to “posit-dev/ggsql”.

Basic usage

When you load ggsql into R in your Rmarkdown/Quarto document the knitr engine is registered. This allows you to use ggsql directly in Rmarkdown and Quarto documents.

While one of the core appeals of the ggsql R package is the knitr engine it provides, you can also use it directly in R to execute visual queries:

library(ggsql)

# Create an in-memory DuckDB reader
reader <- duckdb_reader()

# Register a dataset in it
ggsql_register(reader, mtcars, "amazing_data")

# Visualize it with a query
ggsql_execute(reader, "
VISUALIZE mpg AS x, disp AS y FROM amazing_data
DRAW point
")

Further reading

Full documentation for the R package is available at r.ggsql.org.