Snyk for Python
Snyk provides security scanning on projects for vulnerabilities both through the Snyk CLI and from the Snyk Web UI through different Snyk Integrations.
This page describes how to use Snyk to scan Python projects.
Features of Snyk for Python
How Snyk for Python works
After Snyk builds the tree, Snyk uses the vulnerability database to find vulnerabilities in any of the packages anywhere in the dependency tree.
The way Snyk analyzes and builds the tree varies depending on the language and package manager of the Project, as well as the location of your project. For more information see Snyk CLI for Python Projects.
pip
Snyk requires the full, nested dependency tree in order to run tests. Requirements.txt files contain only the top-level dependencies and not the nested or transitive dependencies. The most efficient way to ensure accuracy is to install the full pip Project.
Snyk runs tests against the specific pip installation used for that specific service or product rather than an unlinked, separate dependency tree.
In order to scan the full dependency tree, Snyk analyzes the installed packages to ensure none are missing. This happens even when not explicitly specified in the manifest file.
Install the missing packages by invoking pip install, for example:
python3 -m pip install -r requirements.txtURLs in requirements.txt files are not supported, as this introduces a security risk.
They are removed before resolving the dependencies in the files.
Poetry
To build the dependency tree for a Poetry application, Snyk uses pyproject.toml and poetry.lock files. Both these files must be present for Snyk to scan Poetry dependencies and identify issues.
Pipenv
To build the dependency tree, run pipenv install as Snyk needs this to create the pipenv graph which is then used to generate the dependency tree.
Snyk uses the built dependency tree to analyze the Pipfile.
setup.py
To build the dependency tree, Snyk analyzes the setup.py file, and detects packages listed in the install_requires key.
There is no auto-discovery for this file. It must be specified manually:
snyk test --file=setup.pyYou can convert setup.py to requirements.txt by installing the packages into a virtual environment and then running pip freeze.
Snyk CLI for Python Projects
The way Snyk analyzes and builds the tree varies depending on the language and package manager of the Project.
Prerequisites for Snyk CLI for Python
Ensure you have installed the relevant package manager before you begin using the Snyk CLI.
Ensure you have included the relevant manifest files supported by Snyk before testing.
Install and authenticate the Snyk CLI to start analyzing projects from your local environment. For more information about Snyk CLI see Getting started with the CLI.
Snyk CLI options for Python
For information about the snyk test options available for use with Python, see Options for Python projects in the test help.
For the available snyk monitor options, see Options for Python projects in the monitor help.
Git services for pip projects
To test Python projects that use pip as a package manager, Snyk analyzes your requirements.txt file. You must have this file in your repository before importing.
If you have renamed your requirements.txt files, for example if you have renamed a file to requirements-dev.txt, Snyk tries to import every file that follows the **/*req*.txt convention as a Python project.
If you have placed your files in a requirements folder, for example, if you have placed your file under requirements/requirements.txt, Snyk tries to import every file that follows the **/requirements/*.txt convention as a Python project.
If you are using a package manager that creates different manifest file formats from requirements.txt, then either convert or import, depending on the package manager/supported files, the manifest file to the requirements.txt format.
Example:
dephell deps convert --from=conda --to=requirements.txtGit services for Poetry projects
To test Python projects that use the Poetry package manager, Snyk analyzes your pyproject.toml and poetry.lock files. You must have these files in your repository before importing.
You can choose whether Snyk should include dev dependencies when scanning your Poetry projects.
Snyk regards non-dev dependencies to be those declared in tool.poetry.dependencies (the implicit main group). All others are classed as dev dependencies.
By default Poetry dev dependencies are not included in scans. To change this adjust your settings as follows:
Log in to your account and navigate to the relevant Group and Organization to manage.
Select Settings > Languages.
Select Edit settings for Python.
Toggle dev dependency scanning on or off under Poetry dev dependencies.
Using different Python versions
Some Python projects may have dependencies that are only valid using Python 3. By default, Snyk scans with Python 3.
You can adjust the version of Python that Snyk uses to scan dependencies in both the CLI and Git integration.
Setting the Python version in the CLI
Add the following option to snyk test or snyk monitor with the value of the python binary:
--command=python3For details see the Test command help and the Monitor command help.
Setting the Python version in Git Projects
By default Snyk tests using Python 3.
To define which Python major version Snyk uses to test your Git imported Projects, use either Organization settings or a .snyk policy file.
To define the Python version for all Projects in an Organization:
Log in to your account and navigate to the relevant Group and Organization to manage.
Select Settings > Languages.
Select Edit settings for Python.
Select Python 2 or Python 3 to use when testing projects for this organization.
Snyk recommends you create different Organizations to work with different Python versions.
If you prefer to use one Organization but require Projects to use different Python versions, you may add a .snyk file to a Project repository and specify the desired version.
The .snyk file must be in the same directory as the project manifest file.
Major and minor versions of Python
On finding a .snyk file, Snyk detects the major version specified and uses this to control whether the Project is tested with Python 2 or Python 3. It does not use the exact version specified.
For example, for Projects imported via Git:
language-settings:
python: '3.7.2'This example tells Snyk to use a recent version of Python 3, but Snyk does not use the exact minor and patch version specified.
Scanning Python dependencies in IDEs
If you are using any of the supported IDEs to write Python, there are some configurations you must add in order to properly scan Python manifest files.
If you are using a virtual environment, it is important that you add the PYTHON_PATH to the Additional Options text input in the Snyk integration settings, for example, --command=.venv/bin/python. Snyk tries to look for a *req*.txt file in the root of the project as seen in the IDE.
However, if you have manifest files in other directories within the root of the project, Snyk is not able to identify them. In order for Snyk to find them, you must to use the --all-projects option. Snyk then recursively searches through each directory within the Project to find all of the manifest files.
If those directories each require a different virtual environment to run, the Snyk scan will not be successful because it will be using one virtual environment to search for installed dependencies. In this case, it is best to use the CLI or the Git integration to get vulnerability information on all of the dependencies listed in each directory of your Project.
Last updated
Was this helpful?