using Pkg
Pkg.add(["Optim", "CairoMakie", "PythonCall", "CondaPkg", "RCall"])Multi-Language Computing: Julia, Python, and R Working Together
Showcasing Quarto’s multi-language support and how to seamlessly call Python and R from Julia.
1 Introduction
One of the things I love about Quarto is its first-class support for multiple programming languages. In many ways, Quarto feels like the natural evolution of Jupyter notebooks. The term “Jupyter” itself is a nod to the three core languages it originally supported: JUlia, PYThon, and R. Quarto brings this multi-language philosophy to a content-first publishing platform. With Quarto, you write rich Markdown documents with figures, tables, cross-references and citations while incorporating live code blocks possibly from different programming languages, all in the same document. These code blocks get run and their output gets weaved into the document upon rendering. Quarto can render the document into a variety of formats, including HTML, PDF, and even Microsoft Word. The Microsoft Word output is actually surprisingly good, with proper formatting and embedded figures, which is great for sharing with collaborators who prefer that format for editing and reviewing.
But why would anyone want to use multiple languages in the same document? The answer is simple: each language has its strengths, and sometimes you want to leverage the best tool for the job without switching contexts. When I started coding in Julia back in 2016, Julia was a very young language. Its ecosystem was still growing, especially in visualization and plotting libraries. Having access to Python’s mature plotting libraries meant that I could use Julia for computationally expensive tasks while still creating beautiful visualizations with Python. This multi-language approach allowed me to be productive and get the best of both worlds.
Fast-forward 10 years to 2026 and Julia’s ecosystem has already matured quite a lot. However, R’s ecosystem remains unparalleled for statistical analysis and Python’s ecosystem for deep learning is still largely unmatched with enormous funds poured into it from big tech companies. Of course, both R and Python have their own quirks and limitations, such as the speed of their native code. Native code of both Python and R is very slow which can be a big issue if you are working in a computationally intensive domain. The way R and Python programmers usually work around this issue is by providing bindings to C/C++/Fortran libraries that are well optimized. So only the shell of your code is in R or Python, while the heavy lifting is done by code written in a lower level language. Julia on the other hand, while relatively young, was designed from the ground up to be fast in runtime (albeit precompilation and compilation can be slow), with excellent technical computing support in its standard library, e.g. support for linear algebra, statistics and multi-threading. Being fast and user-friendly at the same time attracted many early adopters from the Python, R, MATLAB and even C++ communities, each building their own Julia packages in their area of expertise.
Given that each language still has its own strengths and weaknesses, you may find yourself needing to use multiple languages in the same project, if for nothing else then at least for benchmarking purposes, when comparing similar libraries across languages. In this post, I will show how you can use Quarto’s multi-language support to call Julia, Python and R in the same notebook. I will then showcase how to call both Python and R from within Julia code using the PythonCall and RCall packages, sharing data structures and memory across language barriers, i.e. no writing to disk or message-passing using network-style protocols. PythonCall and RCall are actually how the native Julia Quarto engine implements multi-language support in Quarto under the hood when you set the engine to julia in a Quarto document. Different Quarto engines of course have different approaches. Beside their use in the Julia Quarto engine, PythonCall and RCall are also very powerful tools on their own, allowing you to call Python and R from any Julia code, not just in Quarto documents.
2 Prerequisites
This document is built using Quarto and the Julia Quarto engine, which allows you to run Julia, Python, and R code blocks in the same document. Before running the code in this document, you need to have Julia, Python, and R installed on your system.
Installing Julia: - Download from julialang.org or use your system’s package manager - Verify installation: julia --version
Installing Python: - Download from python.org or use your system’s package manager - Verify installation: python --version or python3 --version
Installing R: - Download from cran.r-project.org - Verify installation: R --version
Once installed, the Julia Quarto engine will automatically find and use these interpreters.
3 Making R Visible to Quarto and Fixing SSL Certificate Issues on Linux
On my Linux Fedora system, the R installation was not visible to Quarto and the CRAN package downloads were failing due to SSL certificate issues. To fix this, I had to set the R_HOME and LD_LIBRARY_PATH environment variables manually to point to my R installation directory and then rebuild RCall. I also had to set the CURL_CA_BUNDLE environment variable to point to the system’s CA bundle to fix the SSL certificate issues when downloading R packages from CRAN using curl. You can set these environment variables directly in Julia using the following code before building RCall:
# Set R home and library path for RCall to find the R installation on Linux Fedora
ENV["R_HOME"] = "/usr/lib64/R"
ENV["LD_LIBRARY_PATH"] = "/usr/lib64/R/lib:$(get(ENV, "LD_LIBRARY_PATH", ""))"
# Fix SSL certificate issues for R package downloads using curl on Linux Fedora
ENV["CURL_CA_BUNDLE"] = "/etc/pki/tls/certs/ca-bundle.crt"For Ubuntu-based systems, you would set them like this:
# Set R home and library path for RCall to find the R installation on Linux Ubuntu
ENV["R_HOME"] = "/usr/lib/R"
ENV["LD_LIBRARY_PATH"] = "/usr/lib/R/lib:$(get(ENV, "LD_LIBRARY_PATH", ""))"
# Fix SSL certificate issues for R package downloads using curl on Linux Ubuntu
ENV["CURL_CA_BUNDLE"] = "/etc/ssl/certs/ca-certificates.crt"This post is built on a GitHub Actions runner running Ubuntu. Downloading R and setting up the environment variables in the GitHub Actions workflow file was done using the following steps in the workflow:
- name: Install R
uses: r-lib/actions/setup-r@v2
with:
use-public-rspm: true
r-version: 'release'
- name: Set R library path
run: echo "LD_LIBRARY_PATH=$(R RHOME)/lib:$LD_LIBRARY_PATH" >> $GITHUB_ENV
- name: Set CURL_CA_BUNDLE for R on Ubuntu
run: echo "CURL_CA_BUNDLE=/etc/ssl/certs/ca-certificates.crt" >> $GITHUB_ENVThe entire workflow file can be found in the GitHub repository for the personal website, which is linked at the end of this post.
After setting these environment variables, if RCall is already installed but broken (due to the previously missing environment variables), you can re-build RCall to ensure it can find the R installation and is properly configured:
using Pkg
Pkg.build("RCall")To confirm the R installation is correctly found by RCall, you can run a simple command to check the R version:
using RCall
R"version"In this post, we will install RCall below programmatically from within the document itself, which will trigger the build process after installation, so there is no need to manually call Pkg.build("RCall") in this case.
4 Installing Packages
When using the Julia Quarto engine, package management works differently for each language. Julia packages are managed through the Project.toml file, Python packages through CondaPkg.toml, and R packages are installed directly via R commands.
4.1 Julia Packages and Project Environment
Julia packages are managed using the built-in Pkg system. You can add packages from the Julia REPL:
These packages get installed in the document’s environment. Every Quarto document is executed by the Julia engine in an environment, which is defined by a Project.toml file and an optional Manifest.toml file. The Project.toml file specifies the packages and their versions that are required for the document, while the Manifest.toml file records the exact versions of all packages that were installed, including indirect dependencies. This ensures that your document is reproducible, as anyone else can use the same Project.toml and Manifest.toml files to recreate the exact same environment. However, one caveat of the Manifest.toml file is that it does not guarantee reproducibility across different operating systems and hardware architectures. For example, if you create a Manifest.toml file on a Windows machine, it may not work correctly on a Linux machine due to differences in how packages are built and their dependencies. This is something to keep in mind when sharing your document with others who may be using different platforms.
By default, the Julia Quarto engine looks for a Project.toml file in the same directory as the Quarto document. If it finds one, it activates that environment and uses the packages specified there. If it doesn’t, it will keep looking up the directory tree until it finds a Project.toml. If it fails to find any, it will use an isolated environment for that document, created and managed by the Quarto Julia engine itself.
If you create a Project.toml file (and therefore environment) for the document, managing that environment becomes your responsibility. You can launch Julia in that environment before running the document and install packages as you normally would or you can use the Pkg API from within the document itself, like we did above, to install packages programmatically. Adding a new package to your Julia environment will automatically update the Project.toml and Manifest.toml files to reflect the new dependencies. However, keep in mind that the existence of the Project.toml and Manifest.toml files alone does not guarantee that the packages are actually installed on the system in every scenario. For example, one can copy these files from another project, they are just text files, without actually having the packages installed. To ensure that the packages are installed and the environment is ready to use, you need to run Pkg.instantiate() in that environment. This will install all the packages specified in the Project.toml and Manifest.toml files. If the toml files were generated from adding packages and not copied from another project or modified manually, then the packages should already be installed, so running Pkg.instantiate() will just verify that the environment is consistent and ready to use. This instantiation step can be done from the Julia REPL or from within the document itself and it only needs to happen once per environment. If only a Project.toml file is present, calling Pkg.instantiate() will resolve the package versions and create a Manifest.toml file automatically. If both files are present, it will install the exact versions specified in the Manifest.toml file, if they are not already installed.
An example Project.toml file for this document would look like this:
[deps]
CairoMakie = "13f3f980-e62b-5c42-98c6-ff1f3baf88f0"
CondaPkg = "992eb4ea-22a4-4c89-a5bb-47a3300528ab"
Optim = "429524aa-4258-5aef-a3af-852621145aeb"
PythonCall = "6099a3de-0909-46bc-b1f4-468b9a2dfc0d"
RCall = "6f49c342-dc21-5d91-9882-a32aef131414"
[compat]
julia = "1.12"Note that the Julia version is also specified in the compat section. This is important for reproducibility because different Julia versions may have different package compatibility and behavior. For example, the pseudo-random number generator in Julia can change between minor (e.g. 1.12 to 1.13) or even patch (e.g. 1.12.0 to 1.12.1) version updates which can lead to different results when running the same code without it being considered a breaking change, according to the Julia documentation.
To check that RCall is installed and built successfully, you can run the following code:
using RCall
R"version"Precompiling packages... 1811.6 ms ✓ QuartoNotebookWorkerTablesExt (serial) 1 dependency successfully precompiled in 2 seconds Precompiling packages... 869.2 ms ✓ QuartoNotebookWorkerLaTeXStringsExt (serial) 1 dependency successfully precompiled in 1 seconds Precompiling packages... 2209.3 ms ✓ QuartoNotebookWorkerRCallExt (serial) 1 dependency successfully precompiled in 2 seconds
RObject{VecSxp}
_
platform x86_64-pc-linux-gnu
arch x86_64
os linux-gnu
system x86_64, linux-gnu
status
major 4
minor 5.3
year 2026
month 03
day 11
svn rev 89597
language R
version.string R version 4.5.3 (2026-03-11)
nickname Reassured Reassurer Similarly, to check that PythonCall is installed and built successfully, you can run the following code:
using PythonCall
pyimport("sys").versionPrecompiling packages... 1052.5 ms ✓ QuartoNotebookWorkerJSONExt (serial) 1 dependency successfully precompiled in 1 seconds Precompiling packages... 2449.0 ms ✓ QuartoNotebookWorkerPythonCallExt (serial) 1 dependency successfully precompiled in 2 seconds
Python: '3.14.4 | packaged by conda-forge | (main, Apr 8 2026, 02:10:24) [GCC 14.3.0]'4.2 Python Packages
When using the Julia Quarto engine, Python packages are managed through CondaPkg, which integrates with Julia’s package system. The CondaPkg.toml file, much like the Project.toml file for Julia packages, specifies Python dependencies, e.g:
[deps]
scipy = ""
numpy = ""To install Python packages programmatically using PythonCall, you can use the following code:
using PythonCall
using CondaPkg
# Add Python packages via Conda
CondaPkg.add("scipy")
CondaPkg.add("numpy")Calling CondaPkg.add("package_name") will add the specified package to the CondaPkg.toml file and install it in the Python environment that PythonCall uses. You can also specify versions if needed, e.g. CondaPkg.add("scipy", "1.10.0").
If the CondaPkg.toml file is already present with the required dependencies, running using PythonCall in Julia will automatically resolve and install the Python dependencies specified in the CondaPkg.toml file, which is similar to instantiating the environment in Julia. To explicitly instantiate the Python environment and ensure all dependencies are installed before loading PythonCall, you can call CondaPkg.resolve(), which will read the CondaPkg.toml file and install any missing packages.
If instead, you prefer to create your own Conda environment outside of Julia to have full control over your Python environment, you can also do that and then communicate to CondaPkg which existing environment to use by setting some environment variables before running the document. For more details, see the PythonCall and CondaPkg documentation.
4.3 Installing R Packages and Running R Code
You can install R packages using R’s built-in package manager from within Julia using RCall:
using RCall
R"""
library(utils)
options(repos = c(CRAN = "https://cloud.r-project.org"))
install.packages("data.table", dependencies=TRUE)
"""* installing *source* package ‘commonmark’ ...
** this is package ‘commonmark’ version ‘2.0.0’
** package ‘commonmark’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions.c -o extensions.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c wrapper.c -o wrapper.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/cmark.c -o cmark/cmark.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/node.c -o cmark/node.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/iterator.c -o cmark/iterator.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/blocks.c -o cmark/blocks.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/inlines.c -o cmark/inlines.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/scanners.c -o cmark/scanners.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/utf8.c -o cmark/utf8.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/buffer.c -o cmark/buffer.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/references.c -o cmark/references.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/render.c -o cmark/render.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/man.c -o cmark/man.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/xml.c -o cmark/xml.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/html.c -o cmark/html.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/commonmark.c -o cmark/commonmark.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/latex.c -o cmark/latex.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/houdini_href_e.c -o cmark/houdini_href_e.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/houdini_html_e.c -o cmark/houdini_html_e.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/houdini_html_u.c -o cmark/houdini_html_u.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/cmark_ctype.c -o cmark/cmark_ctype.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/arena.c -o cmark/arena.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/linked_list.c -o cmark/linked_list.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/plugin.c -o cmark/plugin.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/registry.c -o cmark/registry.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/syntax_extension.c -o cmark/syntax_extension.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/plaintext.c -o cmark/plaintext.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/footnotes.c -o cmark/footnotes.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c cmark/map.c -o cmark/map.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/autolink.c -o extensions/autolink.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/core-extensions.c -o extensions/core-extensions.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/ext_scanners.c -o extensions/ext_scanners.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/strikethrough.c -o extensions/strikethrough.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/table.c -o extensions/table.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/tagfilter.c -o extensions/tagfilter.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -Icmark -I. -DCMARK_GFM_STATIC_DEFINE -DCMARK_GFM_EXTENSIONS_STATIC_DEFINE -DR_NO_REMAP -DSTRICT_R_HEADERS -I/usr/local/include -fvisibility=hidden -fpic -g -O2 -c extensions/tasklist.c -o extensions/tasklist.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o commonmark.so extensions.o init.o wrapper.o -Lcmark -lstatcmark -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-commonmark/00new/commonmark/libs
** R
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (commonmark)
* installing *source* package ‘R.methodsS3’ ...
** this is package ‘R.methodsS3’ version ‘1.8.2’
** package ‘R.methodsS3’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (R.methodsS3)
* installing *source* package ‘evaluate’ ...
** this is package ‘evaluate’ version ‘1.0.5’
** package ‘evaluate’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (evaluate)
* installing *source* package ‘xfun’ ...
** this is package ‘xfun’ version ‘0.57’
** package ‘xfun’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c base64.c -o base64.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c rand_lcg.c -o rand_lcg.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o xfun.so base64.o init.o rand_lcg.o -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-xfun/00new/xfun/libs
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (xfun)
* installing *source* package ‘bit’ ...
** this is package ‘bit’ version ‘4.6.0’
** package ‘bit’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c attrutil.c -o attrutil.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c bit.c -o bit.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c chunkutil.c -o chunkutil.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c clone.c -o clone.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c integerutil.c -o integerutil.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c merge.c -o merge.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c rle.c -o rle.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c sort.c -o sort.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o bit.so attrutil.o bit.o chunkutil.o clone.o init.o integerutil.o merge.o rle.o sort.o -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-bit/00new/bit/libs
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (bit)
* installing *source* package ‘zoo’ ...
** this is package ‘zoo’ version ‘1.8-15’
** package ‘zoo’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I/usr/local/include -fpic -g -O2 -c coredata.c -o coredata.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I/usr/local/include -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I/usr/local/include -fpic -g -O2 -c lag.c -o lag.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o zoo.so coredata.o init.o lag.o -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-zoo/00new/zoo/libs
** R
** demo
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (zoo)
* installing *source* package ‘yaml’ ...
** this is package ‘yaml’ version ‘2.3.12’
** package ‘yaml’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c api.c -o api.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c dumper.c -o dumper.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c emitter.c -o emitter.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c implicit.c -o implicit.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c loader.c -o loader.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c parser.c -o parser.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c r_emit.c -o r_emit.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c r_ext.c -o r_ext.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c r_parse.c -o r_parse.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c reader.c -o reader.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c scanner.c -o scanner.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I. -DNDEBUG -I/usr/local/include -fpic -g -O2 -c writer.c -o writer.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o yaml.so api.o dumper.o emitter.o implicit.o loader.o parser.o r_emit.o r_ext.o r_parse.o reader.o scanner.o writer.o -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-yaml/00new/yaml/libs
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (yaml)
* installing *source* package ‘data.table’ ...
** this is package ‘data.table’ version ‘1.18.2.1’
** package ‘data.table’ successfully unpacked and MD5 sums checked
** using staged installation
gcc -std=gnu2x 13.3.0
zlib 1.3 is available ok
* checking if R installation supports OpenMP without any extra hints... yes
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c assign.c -o assign.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c between.c -o between.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c bmerge.c -o bmerge.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c chmatch.c -o chmatch.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c cj.c -o cj.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c coalesce.c -o coalesce.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c dogroups.c -o dogroups.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fastmean.c -o fastmean.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fcast.c -o fcast.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fifelse.c -o fifelse.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fmelt.c -o fmelt.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c forder.c -o forder.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frank.c -o frank.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fread.c -o fread.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c freadR.c -o freadR.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c froll.c -o froll.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frollR.c -o frollR.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frolladaptive.c -o frolladaptive.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frollapply.c -o frollapply.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fsort.c -o fsort.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fwrite.c -o fwrite.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fwriteR.c -o fwriteR.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c gsumm.c -o gsumm.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c hash.c -o hash.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c idatetime.c -o idatetime.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c ijoin.c -o ijoin.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c inrange.c -o inrange.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c mergelist.c -o mergelist.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c nafill.c -o nafill.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c negate.c -o negate.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c nqrecreateindices.c -o nqrecreateindices.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c openmp-utils.c -o openmp-utils.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c programming.c -o programming.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c quickselect.c -o quickselect.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c rbindlist.c -o rbindlist.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c reorder.c -o reorder.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c shellsort.c -o shellsort.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c shift.c -o shift.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c snprintf.c -o snprintf.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c subset.c -o subset.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c transpose.c -o transpose.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c types.c -o types.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c uniqlist.c -o uniqlist.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c utils.c -o utils.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c vecseq.c -o vecseq.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c wrappers.c -o wrappers.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o data.table.so assign.o between.o bmerge.o chmatch.o cj.o coalesce.o dogroups.o fastmean.o fcast.o fifelse.o fmelt.o forder.o frank.o fread.o freadR.o froll.o frollR.o frolladaptive.o frollapply.o fsort.o fwrite.o fwriteR.o gsumm.o hash.o idatetime.o ijoin.o init.o inrange.o mergelist.o nafill.o negate.o nqrecreateindices.o openmp-utils.o programming.o quickselect.o rbindlist.o reorder.o shellsort.o shift.o snprintf.o subset.o transpose.o types.o uniqlist.o utils.o vecseq.o wrappers.o -fvisibility=hidden -fopenmp -lz -L/opt/R/4.5.3/lib/R/lib -lR
PKG_CFLAGS = -fvisibility=hidden -fopenmp
PKG_LIBS = -fvisibility=hidden -fopenmp -lz
if [ "data.table.so" != "data_table.so" ]; then mv data.table.so data_table.so; fi
if [ "" != "Windows_NT" ] && [ `uname -s` = 'Darwin' ]; then install_name_tool -id data_table.so data_table.so; fi
installing to /home/runner/work/_temp/Library/00LOCK-data.table/00new/data.table/libs
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (data.table)
* installing *source* package ‘R.oo’ ...
** this is package ‘R.oo’ version ‘1.27.1’
** package ‘R.oo’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
Warning in setGenericS3.default(name, export = exportGeneric, envir = envir, :
Renamed the preexisting function getClasses to getClasses.default, which was defined in environment R.oo.
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (R.oo)
* installing *source* package ‘highr’ ...
** this is package ‘highr’ version ‘0.12’
** package ‘highr’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (highr)
* installing *source* package ‘litedown’ ...
** this is package ‘litedown’ version ‘0.9’
** package ‘litedown’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (litedown)
* installing *source* package ‘bit64’ ...
** this is package ‘bit64’ version ‘4.6.0-1’
** package ‘bit64’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c bsearch.c -o bsearch.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c cache.c -o cache.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c hash64.c -o hash64.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c integer64.c -o integer64.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c sort64.c -o sort64.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fpic -g -O2 -c sortuse64.c -o sortuse64.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o bit64.so bsearch.o cache.o hash64.o init.o integer64.o sort64.o sortuse64.o -lm -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-bit64/00new/bit64/libs
** R
** data
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (bit64)
* installing *source* package ‘xts’ ...
** this is package ‘xts’ version ‘0.14.2’
** package ‘xts’ successfully unpacked and MD5 sums checked
** using staged installation
** libs
using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c add_class.c -o add_class.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c any.c -o any.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c attr.c -o attr.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c binsearch.c -o binsearch.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c coredata.c -o coredata.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c dimnames.c -o dimnames.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c endpoints.c -o endpoints.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c extract_col.c -o extract_col.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c init.c -o init.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c isOrdered.c -o isOrdered.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c isXts.c -o isXts.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c lag.c -o lag.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c merge.c -o merge.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c na.c -o na.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c period_apply.c -o period_apply.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c period_arithmetic.c -o period_arithmetic.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c period_quantile.c -o period_quantile.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c rbind.c -o rbind.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c rollfun.c -o rollfun.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c runSum.c -o runSum.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c startofyear.c -o startofyear.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c subset.c -o subset.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c toperiod.c -o toperiod.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c totalcols.c -o totalcols.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c tryXts.c -o tryXts.o
gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I../inst/include -I'/home/runner/work/_temp/Library/zoo/include' -I/usr/local/include -fpic -g -O2 -c unique.time.c -o unique.time.o
gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o xts.so add_class.o any.o attr.o binsearch.o coredata.o dimnames.o endpoints.o extract_col.o init.o isOrdered.o isXts.o lag.o merge.o na.o period_apply.o period_arithmetic.o period_quantile.o rbind.o rollfun.o runSum.o startofyear.o subset.o toperiod.o totalcols.o tryXts.o unique.time.o -L/opt/R/4.5.3/lib/R/lib -lR
installing to /home/runner/work/_temp/Library/00LOCK-xts/00new/xts/libs
** R
** data
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** checking absolute paths in shared objects and dynamic libraries
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (xts)
* installing *source* package ‘R.utils’ ...
** this is package ‘R.utils’ version ‘2.13.0’
** package ‘R.utils’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
Warning in setGenericS3.default(name, export = exportGeneric, envir = envir, :
Renamed the preexisting function warnings to warnings.default, which was defined in environment R.utils.
Warning in setGenericS3.default("use") :
Renamed the preexisting function use to use.default, which was defined in environment R.utils.
Warning in setMethodS3.default("use", "default", function(pkg = "R.utils", :
Method already existed and was overwritten: use.default
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (R.utils)
* installing *source* package ‘knitr’ ...
** this is package ‘knitr’ version ‘1.51’
** package ‘knitr’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (knitr)
* installing *source* package ‘markdown’ ...
** this is package ‘markdown’ version ‘2.0’
** package ‘markdown’ successfully unpacked and MD5 sums checked
** using staged installation
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (markdown)
┌ Warning: RCall.jl: Installing package into ‘/home/runner/work/_temp/Library’
│ (as ‘lib’ is unspecified)
│ also installing the dependencies ‘commonmark’, ‘R.oo’, ‘R.methodsS3’, ‘evaluate’, ‘highr’, ‘xfun’, ‘litedown’, ‘bit64’, ‘bit’, ‘R.utils’, ‘xts’, ‘zoo’, ‘yaml’, ‘knitr’, ‘markdown’
│
│ trying URL 'https://cloud.r-project.org/src/contrib/commonmark_2.0.0.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/R.oo_1.27.1.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/R.methodsS3_1.8.2.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/evaluate_1.0.5.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/highr_0.12.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/xfun_0.57.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/litedown_0.9.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/bit64_4.6.0-1.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/bit_4.6.0.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/R.utils_2.13.0.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/xts_0.14.2.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/zoo_1.8-15.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/yaml_2.3.12.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/knitr_1.51.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/markdown_2.0.tar.gz'
│ trying URL 'https://cloud.r-project.org/src/contrib/data.table_1.18.2.1.tar.gz'
│
│ The downloaded source packages are in
│ ‘/tmp/Rtmp7a5e6l/downloaded_packages’
└ @ RCall ~/.julia/packages/RCall/fTLHT/src/io.jl:166
RObject{NilSxp}
NULLTo test if the package is installed and can be loaded, you can run the following to check the version of the installed package:
R"""
# Load the package to confirm it was installed correctly
library(data.table)
packageVersion("data.table")
"""RObject{VecSxp}
[1] ‘1.18.2.1’The syntax above R"""...""" is known as a string macro in Julia. For single line commands, R"..." also works. This feature of Julia allows one to define arbitrary syntax and semantics for domain-specific languages (DSLs). In this case, the entire R programming language is treated as a DSL! Note that one has to be careful to use single quotes to define R strings, e.g. 'data.table', when using R"..." to avoid conflicting with the double quotes used for the Julia string macro. However, when using the multi-line string syntax in Julia R"""...""", using double quotes for R strings works fine.
The R"..." string macro allows you to write R code directly in Julia, and it will be executed in the R interpreter. This means that you can install R packages, run R scripts, and even create R plots all from within your Julia code. The main disadvantage of string macros is the lack of syntax highlighting and code completion in the editor, which can make it harder to write and debug R code. However, for simple commands like installing packages, it can be a convenient way to manage R dependencies without leaving the Julia environment. Ensuring the string quote style is correct is also a bit annoying. We will see a better approach below.
The install.packages function is used above to install packages from CRAN, after setting the repos option to specify the CRAN mirror to use. The dependencies=TRUE argument ensures that any dependencies of the package are also installed. You can replace 'data.table' with the name of the package you want to install.
Since the Julia Quarto engine also uses RCall under the hood, one can also directly utilize Quarto’s multi-language code block feature and run an R block instead to install and load the package, which is more straightforward and doesn’t require embedding R code in a Julia string macro. This also allows you to have syntax highlighting and code completion for R code in the editor.
library(utils)
options(repos = c(CRAN = "https://cloud.r-project.org"))
install.packages("data.table", dependencies=TRUE)* installing *source* package ‘data.table’ ... ** this is package ‘data.table’ version ‘1.18.2.1’ ** package ‘data.table’ successfully unpacked and MD5 sums checked ** using staged installation gcc -std=gnu2x 13.3.0 zlib 1.3 is available ok * checking if R installation supports OpenMP without any extra hints... yes ** libs using C compiler: ‘gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0’ gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c assign.c -o assign.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c between.c -o between.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c bmerge.c -o bmerge.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c chmatch.c -o chmatch.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c cj.c -o cj.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c coalesce.c -o coalesce.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c dogroups.c -o dogroups.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fastmean.c -o fastmean.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fcast.c -o fcast.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fifelse.c -o fifelse.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fmelt.c -o fmelt.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c forder.c -o forder.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frank.c -o frank.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fread.c -o fread.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c freadR.c -o freadR.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c froll.c -o froll.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frollR.c -o frollR.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frolladaptive.c -o frolladaptive.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c frollapply.c -o frollapply.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fsort.c -o fsort.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fwrite.c -o fwrite.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c fwriteR.c -o fwriteR.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c gsumm.c -o gsumm.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c hash.c -o hash.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c idatetime.c -o idatetime.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c ijoin.c -o ijoin.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c init.c -o init.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c inrange.c -o inrange.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c mergelist.c -o mergelist.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c nafill.c -o nafill.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c negate.c -o negate.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c nqrecreateindices.c -o nqrecreateindices.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c openmp-utils.c -o openmp-utils.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c programming.c -o programming.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c quickselect.c -o quickselect.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c rbindlist.c -o rbindlist.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c reorder.c -o reorder.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c shellsort.c -o shellsort.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c shift.c -o shift.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c snprintf.c -o snprintf.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c subset.c -o subset.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c transpose.c -o transpose.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c types.c -o types.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c uniqlist.c -o uniqlist.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c utils.c -o utils.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c vecseq.c -o vecseq.o gcc -std=gnu2x -I"/opt/R/4.5.3/lib/R/include" -DNDEBUG -I/usr/local/include -fvisibility=hidden -fopenmp -fpic -g -O2 -c wrappers.c -o wrappers.o gcc -std=gnu2x -shared -L/opt/R/4.5.3/lib/R/lib -L/usr/local/lib -o data.table.so assign.o between.o bmerge.o chmatch.o cj.o coalesce.o dogroups.o fastmean.o fcast.o fifelse.o fmelt.o forder.o frank.o fread.o freadR.o froll.o frollR.o frolladaptive.o frollapply.o fsort.o fwrite.o fwriteR.o gsumm.o hash.o idatetime.o ijoin.o init.o inrange.o mergelist.o nafill.o negate.o nqrecreateindices.o openmp-utils.o programming.o quickselect.o rbindlist.o reorder.o shellsort.o shift.o snprintf.o subset.o transpose.o types.o uniqlist.o utils.o vecseq.o wrappers.o -fvisibility=hidden -fopenmp -lz -L/opt/R/4.5.3/lib/R/lib -lR PKG_CFLAGS = -fvisibility=hidden -fopenmp PKG_LIBS = -fvisibility=hidden -fopenmp -lz if [ "data.table.so" != "data_table.so" ]; then mv data.table.so data_table.so; fi if [ "" != "Windows_NT" ] && [ `uname -s` = 'Darwin' ]; then install_name_tool -id data_table.so data_table.so; fi installing to /home/runner/work/_temp/Library/00LOCK-data.table/00new/data.table/libs ** R ** inst ** byte-compile and prepare package for lazy loading ** help *** installing help indices ** building package indices ** installing vignettes ** testing if installed package can be loaded from temporary location ** checking absolute paths in shared objects and dynamic libraries ** testing if installed package can be loaded from final location ** testing if installed package keeps a record of temporary installation path * DONE (data.table) ┌ Warning: RCall.jl: Installing package into ‘/home/runner/work/_temp/Library’ │ (as ‘lib’ is unspecified) │ trying URL 'https://cloud.r-project.org/src/contrib/data.table_1.18.2.1.tar.gz' │ Content type 'application/x-gzip' length 6023973 bytes (5.7 MB) │ ================================================== │ downloaded 5.7 MB │ │ │ The downloaded source packages are in │ ‘/tmp/Rtmp7a5e6l/downloaded_packages’ └ @ RCall ~/.julia/packages/RCall/fTLHT/src/io.jl:166
library(data.table)
packageVersion("data.table")1-element Vector{Any}:
[1, 18, 2, 1]5 Solving an Optimization Problem
In the rest of this post, we will attempt to solve the same optimization problem in all three languages, Julia, Python and R. The optimization problem we will solve is known as the Rosenbrock function, which is a common test problem for optimization algorithms. The function is defined as:
\[f(x, y) = (1 - x)^2 + 2(y - x^2)^2\]
Let’s visualize the Rosenbrock function with a heatmap:
using CairoMakie
# Define the Rosenbrock function
rosenbrock_two_inputs(x, y) = (1 - x)^2 + 2 * (y - x^2)^2
# Create a grid of points
x = range(-5, 5, length=1000)
y = range(-2, 7, length=1000)
# Create the heatmap with log scale
fig, ax, hm = heatmap(x, y, (x, y) -> log(rosenbrock_two_inputs(x, y) + 1), colormap=:viridis)
scatter!([1.0], [1.0], color=:red, markersize=10, label="Optimum")
ax.xlabel = "x"
ax.ylabel = "y"
ax.title = "Rosenbrock Function (log scale)"
Colorbar(fig[1, 2], hm, label="log(f(x,y) + 1)")
figPrecompiling packages... 6733.7 ms ✓ QuartoNotebookWorkerMakieExt (serial) 1 dependency successfully precompiled in 7 seconds Precompiling packages... 5101.3 ms ✓ QuartoNotebookWorkerCairoMakieExt (serial) 1 dependency successfully precompiled in 5 seconds
The Rosenbrock function is interesting because it has a global minimum at \((x, y) = (1, 1)\) with a value of \(f(1, 1) = 0\). The valley containing the minimum is shallow and narrow, making it a somewhat challenging problem for optimization algorithms.
6 Solving with Julia
Let’s start with Julia. We will use the Optim package, which provides a variety of optimization algorithms. We will be using the Broyden–Fletcher–Goldfarb–Shanno (BFGS) optimization algorithm, which is a popular quasi-Newton method for unconstrained optimization. The code below defines the Rosenbrock function, sets an initial guess, and then calls the optimize function from Optim to find the minimum.
using Optim
# Define the Rosenbrock function
rosenbrock(x) = (1 - x[1])^2 + 2 * (x[2] - x[1]^2)^2
# Initial guess
x0 = [0.0, 0.0]
# Solve using Optim with the BFGS method
result = optimize(rosenbrock, x0, BFGS())
println("Julia (Optim) Results:")
println("Minimum found at: $(Optim.minimizer(result))")
println("Minimum value: $(Optim.minimum(result))")
println("Iterations: $(Optim.iterations(result))")Julia (Optim) Results:
Minimum found at: [1.0000000056884275, 1.0000000119477706]
Minimum value: 3.3010096601180365e-17
Iterations: 87 Solving with Python
Now let’s solve the same problem with Python’s SciPy library:
import numpy as np
from scipy.optimize import minimize
# Define the Rosenbrock function
def rosenbrock(x):
return (1 - x[0])**2 + 2 * (x[1] - x[0]**2)**2
# Initial guess
x0 = np.array([0.0, 0.0])
# Solve using SciPy
result = minimize(rosenbrock, x0, method='BFGS')
print("Python (SciPy) Results:")
print(f"Minimum found at: {result.x}")
print(f"Minimum value: {result.fun}")
print(f"Iterations: {result.nit}")Python (SciPy) Results:
Minimum found at: [0.99999995 0.99999989]
Minimum value: 2.6658832890325714e-15
Iterations: 78 Solving with R
Let’s also solve it with R’s built-in optimization functions:
# Define the Rosenbrock function
rosenbrock <- function(x) {
(1 - x[1])^2 + 2 * (x[2] - x[1]^2)^2
}
# Initial guess
x0 <- c(0, 0)
# Solve using optim
result <- optim(x0, rosenbrock, method = "BFGS")
cat("R (optim) Results:\n")
cat("Minimum found at:", result$par, "\n")
cat("Minimum value:", result$value, "\n")
cat("Iterations:", result$counts[1], "\n")R (optim) Results:
Minimum found at: 0.9999963 0.9999924
Minimum value: 1.39005e-11
Iterations: 41 The difference in the number of iterations is due to different implementation details, such as stopping criteria. However, all 3 optimization functions were able to get close to the global minimum at (1, 1) with a function value close to 0, which is expected for this problem.
9 Transferring and Sharing Data between Julia and Python
We have seen how to run separate code blocks in different languages within the same Quarto document, which under the hood uses PythonCall and RCall. We have also seen how to run R code ourselves from within Julia using RCall. To run Python code from within Julia, we can similarly use PythonCall ourselves directly. In the code below, I define the Rosenbrock function in Julia and then call SciPy’s minimize function directly on the Julia function to minimize it using Python!
using PythonCall
# Define the Rosenbrock function in Julia
julia_rosenbrock(x) = (1 - x[1])^2 + 2 * (x[2] - x[1]^2)^2
# Call SciPy's minimize from Julia
julia_x0 = [0.0, 0.0]
py_result = pyimport("scipy.optimize").minimize(julia_rosenbrock, julia_x0, method="BFGS")
println("Python via PythonCall:")
println("Minimum found at: $(py_result.x)")
println("Minimum value: $(py_result.fun)")Python via PythonCall:
Minimum found at: [0.99999995 0.99999989]
Minimum value: 2.6658832890325714e-15In the above code, there is a lot of magic! Let me explain. First, the Julia function gets converted under the hood to a Python callable that can be used with SciPy’s minimize function. The Julia array julia_x0 also gets converted to a NumPy array under the hood when passed to Python. But NumPy uses 0-based indexing while Julia uses 1-based indexing, so PythonCall automatically creates a wrapper around the NumPy array, called a PyArray, that behaves like a Julia array with 1-based indexing when accessed from Julia. PyArray is a Julia type but it wraps the NumPy array. So the function julia_rosenbrock only sees the PyArray Julia wrapper of the NumPy array, while scipy.optimize.minimize only sees the original NumPy array and the Python callable version of the Julia function 🤯
To verify this, let’s add some print statements in the Julia function to see what it receives when called from Python:
function julia_rosenbrock(x)
println("Inside Julia function, received x: $x")
println("Type of x in Julia: $(typeof(x))")
return (1 - x[1])^2 + 2 * (x[2] - x[1]^2)^2
endjulia_rosenbrock (generic function with 1 method)Now let’s run the optimization for 1 iteration only to see the print statements in action:
pyimport("scipy.optimize").minimize(julia_rosenbrock, julia_x0, method="BFGS", options=Dict("maxiter" => 1))Inside Julia function, received x: [0.0, 0.0]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [1.4901161193847656e-8, 0.0]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [0.0, 1.4901161193847656e-8]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [1.0099999999999998, -1.5050172917918657e-8]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [1.010000014901161, -1.5050172917918657e-8]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [1.0099999999999998, -1.4901172407100112e-10]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [0.41497022746999623, -6.183538296248993e-9]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [0.4149702423711574, -6.183538296248993e-9]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}
Inside Julia function, received x: [0.41497022746999623, 8.717622897598663e-9]
Type of x in Julia: PythonCall.PyArray{Float64, 1, true, true, Float64}Python:
message: Maximum number of iterations has been exceeded.
success: False
status: 1
fun: 0.4015657185419657
x: [ 4.150e-01 -6.184e-09]
nit: 1
jac: [-5.984e-01 -6.888e-01]
hess_inv: [[ 5.376e-01 4.914e-01]
[ 4.914e-01 1.000e+00]]
nfev: 9
njev: 3Note that PyArray is a thin wrapper around the original NumPy array that allows it to be accessed with 1-based indexing in Julia. So they share the same memory and any changes to the PyArray in Julia will reflect in the original NumPy array in Python and vice versa. However, the indexing is different, so when you access x[1] in the Julia function, it corresponds to x[0] in the original NumPy array in Python, and when you access x[2] in Julia, it corresponds to x[1] in the original NumPy array in Python. This is an important detail to keep in mind when working with arrays across language boundaries, as it can lead to off-by-one errors if not handled correctly.
After the optimization is complete, we can read the output back into Julia and compare it with the output from the native Julia optimization:
println("Comparison:")
println("Julia minimum: $(Optim.minimizer(result)), value: $(Optim.minimum(result))")
println("Python minimum: $(py_result.x), value: $(py_result.fun)")Comparison:
Julia minimum: [1.0000000056884275, 1.0000000119477706], value: 3.3010096601180365e-17
Python minimum: [0.99999995 0.99999989], value: 2.6658832890325714e-15Let’s display the Julia types of the results:
println("Data Types:")
println("Julia result type: $(typeof(Optim.minimizer(result)))")
println("Python result type: $(typeof(py_result.x))")Data Types:
Julia result type: Vector{Float64}
Python result type: PythonCall.PyNotice that the Python result is of type PythonCall.Py in Julia, which is a generic Julia wrapper around any Python object. However in this case, since py_result.x is actually a NumPy array on the Python side, we can convert it to a Julia array type in 2 different ways.
First, we can use pyconvert to convert to a native Julia array type:
py_minimum_julia = pyconvert(Array, py_result.x)
println("Python minimum converted to Julia array: $py_minimum_julia")
println("Type after conversion: $(typeof(py_minimum_julia))")Python minimum converted to Julia array: [0.9999999497098492, 0.9999998911497547]
Type after conversion: Vector{Float64}Notice the type after conversion is Vector{Float64}, which is a native Julia array type. This means we can now use this result in any further Julia computations without needing to worry about Python interop. However, note that this conversion does involve copying the data from Python to Julia, which can be inefficient for large arrays.
To avoid copying, one can also convert the result to a PyArray. Recall, PyArray is a Julia type that is meant to wrap NumPy arrays WITH memory sharing, behaving as a Julia mirror of the underlying Python’s NumPy arrays. This allows you to manipulate the same data in both languages without copying. To convert to a PyArray, you can simply call the PyArray constructor on the Python object:
py_minimum_shared = PyArray(py_result.x)2-element PyArray{Float64, 1}:
0.9999999497098492
0.9999998911497547To verify that this is indeed a shared array, we can modify the py_minimum_shared array in Julia and see if it reflects in the original Python result:
py_minimum_shared[1] += 1.0 # Modify the shared array in Julia
println("Modified Python minimum (shared): $py_minimum_shared")
println("Original Python minimum after modification: $(py_result.x)")Modified Python minimum (shared): [1.9999999497098493, 0.9999998911497547]
Original Python minimum after modification: [1.99999995 0.99999989]10 Transferring Data between Julia and R
Similarly, we can also call R code from Julia using RCall and share data between the two languages. For example, we can define the Rosenbrock function in Julia and then call R’s optim function directly on the Julia function to minimize it using R!
using RCall
# Define the Rosenbrock function in Julia
julia_rosenbrock(x) = (1 - x[1])^2 + 2 * (x[2] - x[1]^2)^2
# Call R's optim from Julia
julia_x0 = [0.0, 0.0]
r_result = R"optim"(julia_x0, julia_rosenbrock, method="BFGS")
julia_result_par = rcopy(r_result[:par])
julia_result_value = rcopy(r_result[:value])
println("R via RCall:")
println("Minimum found at: $(julia_result_par)")
println("Minimum value: $(julia_result_value)")R via RCall:
Minimum found at: [0.9999962794690533, 0.9999923884440581]
Minimum value: 1.3900496413906366e-11Similar kind of magic happens under the hood as with PythonCall to manage the conversion of the Julia function and array to their R counterparts. Notice that to access the fields of the R object, we use the [:field_name] syntax, which is how RCall allows you to access R objects from Julia, as opposed to the $field_name syntax used in R itself. This is because $ is reserved for string interpolation in Julia, which can be used to interpolate Julia objects/data into R code when using the R"..." string macro, but it cannot be used to access fields of R objects from Julia.
Now let’s print the types of the results:
println("Data Types:")
println("R result type before conversion: $(typeof(r_result[:par]))")
println("R result type: $(typeof(julia_result_par))")Data Types:
R result type before conversion: Ptr{RCall.RealSxp}
R result type: Vector{Float64}Notice that the R result is just a wrapper around the underlying R vector object. By calling rcopy, we convert it to a native Julia array type Vector{Float64}. This conversion involves copying the data from R to Julia, which can be inefficient for large arrays. However, unlike PythonCall, as far as I know, RCall does not have a built-in way to share memory between R and Julia without copying, so using rcopy is the standard way to get the data into a native Julia format for further computations.
12 Side Note on String Macros in Julia
String macros in Julia have a funny naming convention. Say we want to define our own string macro called foo"...", we can define it using the following syntax:
macro foo_str(str)
# Do something with the string `str`
endBut when we are using the macro, we are using it using the following syntax (without the _str suffix):
foo"some string"And when importing it explicitly from a package like we are doing above, we use the following syntax (with an added @ prefix):
using PackageName: @foo_strSo you refer to string macros in 3 different ways in different contexts! This is arguably a bit of a Julia quirk. But as you saw, string macros are also extremely powerful, so we will let that one slide!
13 Conclusion
With this, hopefully, you see the relative ease of mixing multiple programming languages in the same Quarto document using the Julia engine. While PythonCall and RCall each have their own slightly different syntax and semantics, they both work well enough unlocking the entirety of the Python and R ecosystems from within Julia. So if not having access to that one Python/R package you need was stopping you from exploring Julia, hopefully, I have convinced you that you can have your cake and eat it too with Quarto’s multi-language support and the Julia Quarto engine.
While PythonCall and RCall may seem like magic if you have never used multiple programming languages like this before, under the hood, it’s all careful engineering and management of data structures and memory across language boundaries. Quarto makes things even easier by providing language-specific code blocks with syntax highlighting and native syntax for each language (e.g. using $ in R blocks), while the Julia engine simplifies the sharing and transfer of data across language barriers.
Of course, we barely scratched the surface here. Both PythonCall and RCall have many more features for more complex use cases. The same goes for Quarto which has other engines besides the Julia Quarto engine, that we have used here, with different implementation details and capabilities.
Happy multi-language computing with Quarto and Julia!