• Docker Desktop 4.30.0
• GitHub CLI gh version 2.50.0
• wget (brew install wget)
• jq (brew install jq)

Note: This current setup only supports arm64 architecture.

• Create a .deepsource.toml file in the root of the repository. Ensure all values and metadata are configured properly in .deepsource.toml file for accurate analysis results. Refer to the respective analyzer's docs for more information. Refer to the .deepsource.toml configuration file examples section below for example configurations.

• Authenticate gh with GitHub - gh auth login

• Create a Classis PAT with scope read:packages (Only Classic PATs are allowed; GitHub Container Registry does not support fine-grained access tokens yet)

• Create an env variable called GH_PAT_DEEPSOURCE with the GitHub PAT token

export GH_PAT_DEEPSOURCE=<PAT>

• Change directory to the repository root on which analysis is to be run
• Setup DeepSource analysis environment

wget https://raw.githubusercontent.com/QuackatronHQ/analyze/master/run.sh -O run.sh
chmod +x run.sh
./run.sh setup

• From the current working directory as repository root, run the following command to start the analysis:

./run.sh analyze . <LANGUAGE>

Note: Currently supported LANGUAGE values are: go, python, java, ruby, javascript, docker, csharp

• Once the analysis is complete, the results will be available in analysis_results_<LANGUAGE>.json and analysis_results_<LANGUAGE>_pretty.json files in the current working directory.

• The results file is JSON formatted and contains the issues in issues key. Each issue has the following keys:
  • issue_code: The unique identifier for the issue.
  • issue_text: The one-line description of the issue. Detailed description of each issue can be found in the Analyzer Directory. For example: description of issue code GO-2307 can be found at app.deepsource.com/directory/analyzers/go/issues/GO-S2307. DeepSource login not required.
  • location: The location of the issue in the code.
    • path: The path of the file where the issue is found.
    • position: The position of the issue in the file.
    • begin: The starting position of the issue.
      • line: The line number where the issue starts.
      • column: The column number where the issue starts.
    • end: The ending position of the issue.
      • line: The line number where the issue ends.
      • column: The column number where the issue ends.

• This setup supports analyzing only one language per repository at any given time. If you need to analyze multiple languages within the same repository, please conduct each analysis sequentially. Each analyzer's results are stored as analysis_results_<LANGUAGE>.json and analysis_results_<LANGUAGE>_pretty.json in the current working directory.
• Analysis is run on all files of the repository (excluding the patterns configured in exclude_patterns). Individual file analysis is not supported, unless it is the only file in the directory/repository.

To cleanup all DeepSource generated files, run the following command:

./run cleanup

Note: This will not cleanup the locally downloaded docker images.

version = 1

test_patterns = [
  "tests/**",
  "test_*.py"
]

exclude_patterns = [
  "migrations/**",
  "**/examples/**"
]

[[analyzers]]
name = "python"

  [analyzers.meta]
  runtime_version = "3.x.x"

version = 1

test_patterns = [
  "tests/*_test.go",
  "**/*_test.go"
]

[[analyzers]]
name = "go"
enabled = true

  [analyzers.meta]
  import_root = "github.com/DeepSourceCorp/marvin-go"

version = 1

test_patterns = [
  "test/**",
  "*Test.java"
]

[[analyzers]]
name = "java"

  [analyzers.meta]
  runtime_version = "21"

version = 1

test_patterns = [
  "test/**",
  "*_test.rb"
]

exclude_patterns = [
  "vendor/**",
  "**/examples/**"
]

[[analyzers]]
name = "ruby"

version = 1

test_patterns = ["*/test/**"]

exclude_patterns = [
    "public/**,",
    "dist/**"
]

[[analyzers]]
name = "javascript"

version = 1

test_patterns = [
  "tests/**",
  "*Test.cs"
]

exclude_patterns = [
  "**/examples/**"
]

[[analyzers]]
name = "csharp"

version = 1

[[analyzers]]
name = "docker"