Native mode

By default, Qodana runs its linters using Docker based on Linux images. In specific cases, you have to deal with private packages or run Qodana on the operating systems that provide incomplete support for Docker.

To overcome this, Qodana supports the native mode for the Qodana for .NET linter. In this case, Qodana reuses its execution environment, which lets you execute Qodana in exactly the same environment as you use for building the projects, use the correct operating system, have access to all repository credentials, and resolve dependencies.

Before you start

Make sure that you have a proper version of the .NET SDK and all required dependencies installed on your machine.

Build the project before inspecting it using Qodana. You can do it by using the bootstrap key of the qodana.yaml file.

The project building and artifact packaging stages should occur before Qodana or simultaneously with it. Because running Qodana may affect the project state and its files, we recommend that you avoid reusing the same directory in your build pipelines any further.

You can also provide Qodana with a pre-built project, or specify the build steps in your CI/CD pipeline. In this case, in your repository create the empty qodana.yaml file to eliminate warnings related to project building.

In your operating system, save the QODANA_TOKEN environment variable containing the Qodana Cloud project token.

Install Qodana CLI on the machine where you plan to run Qodana locally.

Starting from the version 2023.3 of Qodana, the sanity inspection will report in case the qodana.yaml file containing the bootstrap key is missing in your project directory. The bootstrap key should contain instructions for building the project. If you do not wish to build the project, disable this inspection using the --disable-sanity option, add this inspection to a baseline, or create the qodana.yaml file that will contain the ide: QDNET configuration.

We recommend running the native mode on the same machine where you build a project because this can guarantee that Qodana has access to private NuGet feeds.

How it works

You can enable the native mode ba saving this configuration in the qodana.yaml file:

ide: QDNET

This configuration tells Qodana to download and employ the required JetBrains IDE binary file while running.

Below are the examples showing how you can run Qodana in the native mode:

Make sure that the QODANA_TOKEN variable is defined in the environment and refers to a proper project token. If necessary, you can define it:
QODANA_TOKEN=<cloud-project-token>
If you have already enabled the native mode using the qodana.yaml file, use this command:
qodana scan
You can also run Qodana without configuring the qodana.yaml file:
qodana scan \ --ide QDNET

If you have already enabled the native mode using the qodana.yaml file, you can use a basic configuration sample from the GitHub Actions section.

To run Qodana without configuring the qodana.yaml file, in your GitHub repository navigate to a workflow configuration file and specify the --ide,QDNET option:

        name: Qodana
        on:
          workflow_dispatch:
          pull_request:
          push:
            branches:
              - master
              - 'releases/*'
        jobs:
          qodana:
            runs-on: ubuntu-latest
            permissions:
              contents: write
              pull-requests: write
              checks: write
            steps:
              - uses: actions/checkout@v3
                with:
                  ref: ${{ github.event.pull_request.head.sha }}  # to check out the actual pull request commit, not the merge commit
                  fetch-depth: 0  # a full history is required for pull request analysis
              - name: 'Qodana Scan'
                uses: JetBrains/qodana-action@v2024.1
                with:
                  args: --ide,QDNET
                env:
                  QODANA_TOKEN: ${{ secrets.QODANA_TOKEN }}
        

Last modified: 20 June 2024