Qodana 2021.2 Help

Qodana for Python Docker image configuration

Docker image paths




Root directory of the project to be analyzed


Directory to store the analysis results, needs to be empty before running Qodana for Python


PyCharm configuration directory


Directory of the PyCharm distributive


Used if a profile wasn't configured by run options or via qodana.yaml. See Order of resolving a profile.

Configuration options

The following arguments are available for the jetbrains/qodana-python command.



-i, --project-dir=

Project root folder (default: current working directory /data/project)

-o, --results-dir=

Save results to folder (default: /data/results)

-r, --report-dir=

Save HTML report to folder (default: /data/results/report)


Set cache folder (default: /data/cache)

-s, --save-report

Generate HTML report

-w, --show-report

Serve HTML report on port 8080

-b, --baseline

Run in baseline mode. Provide the path to an existing SARIF report to be used in the baseline state calculation.


Include in the output report the results from the baseline run that are absent in the current run


Set the number of problems that will serve as a quality gate. If this number is reached, the inspection run is terminated.


Directory to be inspected. If not specified, the whole project is inspected by default

-n, --profile-name

Name of a profile defined the in project. Note that the name of the profile does not necessarily match the name of the containing file stored in .idea/inspectionProfiles. The actual name is stored as the <option name="myName" value="%profileName%" /> value in the profile xml file and is visible in the PyCharm UI. For details on working with inspection profiles in PyCharm, see the PyCharm documentation.

-p, --profile-path

Absolute path to the profile file


Inspect uncommitted changes and report new problems

Examples of execution tuneup

Override the default inspection profile

docker run ... -v <inspection-profile.xml>:/data/profile.xml jetbrains/qodana-python

If no profile is specified, the default qodana.recommended profile is used. For more options of how to specify a profile, see Order of resolving a profile. For more about available profiles, see Set up a profile.

Save a report as HTML

docker run ... jetbrains/qodana-python --save-report

By default, the HTML report is stored in a separate report/ subdirectory under the results directory. This location could be configured with --report-dir.

Display a report in HTML

docker run ... -p 8080:8080 jetbrains/qodana-python --show-report

After the inspection is finished, the container will not exit and will listen on port 8080. You can connect to http://localhost:8080 to see the results. To stop the web server, press Ctrl-C in the Docker console.

Change the Heap size

docker run ... -e _JAVA_OPTIONS=-Xmx6g jetbrains/qodana-python

By default, Heap size is set to 80% of the host RAM.

Log INFO messages to STDOUT

docker run ... jetbrains/qodana-python --property=idea.log.config.file=info.xml

The default log level for STDOUT is WARN.

Use a different 'idea.properties' file

docker run ... -e IDEA_PROPERTIES=/data/project/idea.properties jetbrains/qodana-python

Turn off user statistics

To disable the reporting of usage statistics, adjust the idea.headless.enable.statistics value:

docker run ... jetbrains/qodana-python --property=idea.headless.enable.statistics=false

Manage plugins

You can add any free IntelliJ platform plugins or your custom plugin by using the following command:

docker run ... -v /your/custom/path/%pluginName%:/opt/idea/plugins/%pluginName% jetbrains/qodana-python

To optimize the most common cases, some bundled plugins are disabled by default. You can check the whole list of disabled plugins in /root/.config/idea/disabled_plugins.txt.

The Python and its libraries'/frameworks' plugins are enabled by default. plugins are enabled by default.

To change the plugins list, do any of the following:

  • Override disabled_plugins.txt by mounting your own file:

    docker run ... -v $empty_file$:/root/.config/idea/disabled_plugins.txt jetbrains/qodana-python
  • Use IDE properties idea.required.plugins.id and idea.suppressed.plugins.id:

    docker run ... jetbrains/qodana-python --property=idea.required.plugins.id=JavaScript,org.intellij.grails --property=idea.suppressed.plugins.id=com.intellij.spring.security

Analyze changes

Qodana for Python lets you check only changed files:

docker run ... jetbrains/qodana-python --property=idea.required.plugins.id=Git4Idea,Subversion,hg4idea --changes

You can adjust the idea.required.plugins.id value and keep only the VCS plugin suitable for your project.

Run in baseline mode

In baseline run mode, each new Qodana for Python run is compared to some initial run selected as a "baseline". This can help in situations when you have no possibility to fix old problems and rather want to prevent the appearance of new ones.

docker run ... jetbrains/qodana-python --baseline <baseline-path> [--baseline-include-absent]

where <baseline-path> is the path to a qodana.sarif.json file from an earlier run. If the --baseline-include-absent option is provided, the inspection results will include absent problems, that is the problems detected only in the baseline run but not in the current run.

After the inspection is finished, Qodana for Python displays the summary of the current project state compared to the baseline.

Baseline comparison result - UNCHANGED: 10, NEW: 5, ABSENT: 3

where the detected problems are aggregated as follows:

  • new: The problem was detected only in the current run but not in the baseline run.

  • absent: The problem was detected only in the baseline run but not in the current run.

  • unchanged: The problem was detected both in the current run and in the baseline run.

The SARIF output report will contain the per-problem information on the baseline state.

Set a quality gate

Qodana for Python lets you configure a "quality gate", that is, the number of problems that will act as a threshold. If the threshold number is reached, the inspection run is terminated.

docker run ... jetbrains/qodana-python --fail-threshold <number>

When running in baseline mode, a threshold is calculated as the sum of new and absent problems. Unchanged results are ignored.

Run as non-root

By default, the container is run as the root user so that Qodana for Python can read any volumes bind-mounted with the project and write the results. As a result, files in the results/ folder are owned by the root after the run.

To avoid this, you can run the container as a regular user:

docker run -u $UID ... # or docker run -u $(id -u):$(id -g) ...

Note that in this case the results/ folder on host should already be created and owned by you. Otherwise, Docker will create it as root and Qodana for Python will not be able to write to it.

Cache dependencies

You can decrease the time for a Qodana for Python run by persisting cache from one run to another. For example, package and dependency management tools such as Maven, Gradle, npm, and Yarn keep a local cache of downloaded dependencies.

By default, Qodana for Python would save caches to folder /data/cache inside container. This location could be changed via --cache-dir cli argument. The data inside is per-repository, so you can pass cache from branch-a to build checking branch-b. In this case, only new dependencies would be downloaded, if they were added.

Example for local run:

docker run --rm -it -p 8080:8080 \ -v <source-directory>/:/data/project/ \ -v <output-directory>/:/data/results/ \ -v <cache-directory>/:/data/cache/ \ jetbrains/qodana-python --show-report

In this case mapping the same <cache-directory> would speed up the second run.

In a GitHub workflow you can utilise actions/cache, see full example.

GitLab CI also has cache which can be stored only inside the project directory. In this case, we recommend excluding the cache folder from inspection via qodana.yaml.

Order of resolving a profile

Qodana for Python checks the configuration parameters for resolving the inspection profile in this order:

  1. Profile with the name %name% from the command-line option --profile-name %name%.

  2. Profile by the path %path% from the command-line option --profile-path %path%.

  3. Profile with the name %name% from qodana.yaml.

  4. Profile by the path %path% from qodana.yaml.

  5. Profile mounted to /data/profile.xml.

  6. Fall back to using the default qodana.recommended profile.

Last modified: 14 October 2021