In mkoohafkan/arcpy: Interface to 'ArcGIS' 'Python' Modules

knitr::opts_chunk$set(echo = TRUE)

Getting started with {arcpy} in R is easy. If you're using ArcGIS Pro, you'll first need to create a Conda environment that will link to your ArcGIS Pro install. If you're not familiar with Conda or the reticulate package, check the vignettes Calling Python from R and Python Version Configuration.

For example, here's how you would create a Conda environment to link to an ArcGIS Pro 3.1 install.

library(arcpy)

install_arcpy(version = "3.1")

If you don't specify the python version, the package will use the latest compatible Python version based on what is reported in the arcpy module build. By default, the package creates a new environment named "r-arcpy", but this can be overridden.

{arcpy} automatically provides a reticulate module object called arcpy. This object provides the interface to ArcGIS.

arcpy
arcpy$GetInstallInfo()$ProductName

Once you are connected to your ArcGIS installation, using arcpy functions and classes is as seamless as using any other Python module via reticulate.

# get and set the arcpy environment
arcpy$env$workspace = tempdir()
arcpy$env$workspace

{arcpy} also re-exports reticulate's py_help function so that you can access the arcpy documentation.

# get help on arcpy functions
py_help(arcpy$Exists)

Happy scripting!

mkoohafkan/arcpy documentation built on July 16, 2025, 4:56 p.m.