Install py5

Install py5#

These instructions have been tested on Linux, Windows, and macOS, and they should work for most people. If you have difficulties, please be patient and try to work through it or let us know on GitHub Issues and we’ll do what we can to help. If you hit a snag and figure out a solution, tell us about it on GitHub Discussions and we’ll update the documentation to share what you’ve learned.

Important

There are a few manageable issues related to using py5 on Mac computers. Mac users should read the Special Notes for macOS Users page for more information.

Requirements#

Below are the basic requirements for using py5.

  • Python 3.9+

  • Java 17+

Python 3.9 is the minimum Python version but you can use a newer version if you wish. Java 17 is the minimum Java version but you can install a newer version of that as well. The Java Virtual Machine cannot be a headless JVM.

The best setup for beginners is to use the Thonny Python Editor and the py5 Thonny plugin, created by @tabreturn. For this route, follow the plugin’s installation instructions. The plugin should work correctly on all computers.

Quick Setup#

If you already have Java 17 and Python 3.9+ available on your computer, you can install py5 with the below command.

pip install py5

Extra Dependencies#

If you intend to use py5 with Jupyter Notebooks, you can install py5 and py5jupyter at the same time with this one command:

pip install py5[jupyter]

If you want to use py5’s Python Ecosystem Integrations, you can install py5, py5jupyter, and all of py5’s optional dependencies (except for the SVG Images and Cairo Integration) with this command:

pip install py5[extras]

The [jupyter] or [extras] suffixes tell the Python package installer to install py5’s optional dependencies. If you have py5 installed already you can alternatively use pip install py5jupyter instead of [jupyter], or pip install colour matplotlib py5jupyter shapely trimesh instead of [extras].

Jupyter Notebook Kernels#

If you have py5jupyter installed, you can install one or both of py5’s Jupyter Notebook Kernels. To install the py5 kernel for imported mode Sketches, use this command:

python -m py5jupyter.kernels.py5.install --sys-prefix

To install py5bot for static mode sketches, use this:

python -m py5jupyter.kernels.py5bot.install --sys-prefix

Quick Example#

Here is a quick py5 example to test that everything works. Please test this with a Jupyter Notebook, the IPython REPL, or by saving it to a *.py file and running it from the command line like this: python quick_example.py. Typing py5 python code into the generic python REPL is generally not a good idea, and this specific example will not work for reasons that are too detailed to explain here.

import py5

def setup():
    py5.size(200, 200)
    py5.rect_mode(py5.CENTER)

def draw():
    py5.rect(py5.mouse_x, py5.mouse_y, 10, 10)

py5.run_sketch()

You should see a small window that draws squares as you move your mouse around. If that works, have a look at the tutorials for more interesting examples.

Anaconda or Miniconda Setup#

Anaconda is a widely used platform for working with Python and the open-source ecosystem. It makes it very easy to create and manage Python environments containing various Python libraries such as py5. Anaconda will also make it easy for you to use other popular Python tools such as Jupyter Notebooks.

Miniconda is similar to Anaconda but with a minimal installation.

Download either the Anaconda Installer or the Miniconda Installer for your Operating System.

Anaconda’s installation instructions and Miniconda’s installation instructions are both extensive and should be able to provide the necessary guidance for your computer.

Brief Steps#

You can create a complete Anaconda environment for py5 using one command:

conda env create -n py5coding -f http://py5coding.org/files/install/py5_environment.yml

Feel free to replace py5coding with your prefered name for the Anaconda environment.

If you don’t like using the command line you can also download py5_environment.yml and create the environment using Anaconda Navigator.

That environment file contains the below information, telling Anaconda to create a Python 3.11 environment with Jupyter and many of py5’s required and optional dependencies.

name: py5coding
channels:
  - conda-forge
dependencies:
  - python=3.11
  - colour>=0.1.5
  - jpype1>=1.4
  - jupyterlab
  - line_profiler>=4.0
  - matplotlib>=3.7
  - numpy>=1.24
  - pillow>=9.5
  - pip
  - shapely>=2.0
  - trimesh>=3.23
  - pip:
      - py5[jupyter]

You must activate the environment using conda activate <environment name>. When the environment is active, you will see the environment name (e.g. “py5coding”) in the command prompt.

conda activate py5coding

You will need to install Java 17 if you don’t have it already. Before attempting an installation, first check to see if you already have it. You can do this from a terminal or DOS window using the command java -version.

java -version

The results should be similar to this:

openjdk version "17.0.2" 2022-01-18
OpenJDK Runtime Environment 21.9 (build 17.0.2+8)
OpenJDK 64-Bit Server VM 21.9 (build 17.0.2+8, mixed mode, sharing)

If you get an error or see the version number is something like 1.8 or 11.0.14, you will need to install or upgrade Java. If your Java installation is a headless JVM, the output of java -version may not indicate this but py5 will later raise an Exception when you import the py5 library.

You can install Java any way you like, but note that installing Java through Anaconda has caused problems in the past. One straightforward installation approach is to use the Python library install-jdk. Install it into your Anaconda environment using pip install:

pip install install-jdk

Then use this next command to install Java 17.

python -c "import jdk; print('Java installed to', jdk.install('17'))"

It isn’t necessary to set the JAVA_HOME environment variable, but if it is already set, you will need to make sure it is set to a Java 17 installation because py5 will always use the version that JAVA_HOME points to, even if there is a newer version available elsewhere on your machine. If it is set incorrectly, py5 will provide you with an error message with some debugging information to help you fix it.

Although the JAVA_HOME environment variable is optional, some users have solved their installation problems by setting it.

Now that Java is installed, you can launch jupyter lab and start coding with py5.

jupyter lab

Try testing with the Quick Example to verify everything works.

Before moving on, consider installing one or both of py5’s Jupyter Notebook Kernels: py5 kernel and py5bot.

Detailed Steps#

If the Brief Steps don’t work for you or you want more detailed information, the below steps will provide you with the necessary information to work through any difficulties.

Create Anaconda Environment#

First you must create an Anaconda environment to install the Python packages into. Below, we create a Python environment called py5coding with Python 3.11. This command will also install Jupyter Lab and many of py5’s required and optional dependencies from Anaconda.

conda create -n py5coding python=3.11 colour>=0.1.5 jpype1>=1.4 jupyterlab line_profiler>=4.0 matplotlib>=3.7 numpy>=1.24 pillow>=9.5 pip shapely>=2.0 trimesh>=3.23

After creating the py5coding environment you must “activate” it so that the subsequent commands take place inside of it. You will know you are inside the environment because your terminal prompt will change to include the name of the environment.

conda activate py5coding

Install Java#

You will need to have Java 17 (or later) installed on your computer.

There are many avenues for doing this, starting with the detailed but not particularly readable instructions on the official java.com website. You can use any method you like so long as it works and the java -version command gives the correct results.

Important

It is important that you have Java 17 installed and available because Processing 4 and therefore py5 both depend on it. If now or in the future you have the wrong version, you will see an error message stating that code “has been compiled by a more recent version of the Java Runtime.”

For your convenience, py5 is designed to be compatible with the Python library install-jdk. This library can download and install the correct version of Java and will put it in a location that py5 will check when it is imported. Enter these two commands into the command prompt to install install-jdk and then install Java 17.

pip install install-jdk
python -c "import jdk; print('Java installed to', jdk.install('17'))"

That’s it. It will install Java into the hidden directory .jdk located in your home directory. You may want to set the $JAVA_HOME environment variable to point to this location, but for this case that isn’t necessary because py5 already knows to look here.

Extra Information About How py5 Finds Java#

When import py5 is executed, py5 will start the Java Virtual Machine. Before doing so it will go through a series of steps to locate a valid Java installation on your computer. If you are a Java aficionado you may have multiple versions of Java installed on your machine. The following information outlines the logic py5 employs to select the version it will use. This information will help you fix problems if py5 cannot be imported properly.

  1. If the $JAVA_HOME environment variable is properly set on your computer, you have communicated to both jpype and py5 that this is the installation of Java that you want to use. The JVM startup process will proceed using that version of Java, but if it is not Java 17, you will get an error.

  2. The jpype library has a function called getDefaultJVMPath() that py5 relies on to search the common Java installation directories for your operating system when the $JAVA_HOME environment variable is not properly set. If you have multiple versions of Java installed, jpype will stop at the first installation found. This may not be the most current Java version available on your machine.

  3. If getDefaultJVMPath() finds a Java installation, py5 will evaluate it to get the Java version number. If it is 17 or more, py5 will proceed using this Java installation.

  4. If Java is not found or the Java version was less than 17, py5 will look in your home ($HOME) directory for the .jdk and .jre subdirectories that Python library install-jdk installs Java into and search through both to find a sufficient Java installation to use.

  5. If Java 17 isn’t found, you will get an error. You will also get an error if the architecture (32 bit vs 64 bit) of your Python installation and your Java installation are not the same. If you get an error, you will also see some helpful debug information that you can use to address your situation.

Install py5 library#

Finally, install the py5 and py5jupyter libraries.

pip install py5 py5jupyter

If you are on Windows or on a Mac, you may get an error relating to the dependent line-profiler package. If so, use the following command to resolve the error, then try pip install py5 again.

conda install -c conda-forge line_profiler

After installing py5, try testing with the Quick Example to verify everything works. Also, consider installing one or both of py5’s Jupyter Notebook Kernels: py5 kernel and py5bot.

Jupyter Notebook Kernels#

To use py5’s Jupyter support you must have the py5jupyter library installed. Starting with version 0.8.0, py5’s Jupyter functionality requires the second optional dependency py5jupyter. Use of py5 and Jupyter without py5jupyter is possible but is deprecated.

py5 kernel#

You can optionally install the py5 Jupyter Notebook Kernel. This is a customized Python kernel that will let you write py5 code in Imported Mode. See The Five py5 Modes to learn about the different py5 Modes.

python -m py5jupyter.kernels.py5.install --sys-prefix

The --sys-prefix argument is optional but I recommend you use it. It will install the py5 kernel inside the py5 Anaconda environment and Jupyter will only present it as an option when Jupyter is run in that environment.

py5bot#

You can optionally install py5bot, which is also a Jupyter Notebook Kernel. This is a customized Python kernel that will let you write py5 code in Static Mode.

python -m py5jupyter.kernels.py5bot.install --sys-prefix

Keeping py5 Updated#

Since py5 is a new library, you can expect frequent updates. Later you will want to upgrade your installation, which you can do with this command:

pip install --upgrade py5 py5jupyter
Contents