About CodeQL databases
To analyze a project, you need to select a CodeQL database for that project. You can select a database locally (from a ZIP archive or an unarchived folder), from a public URL, or from a project's URL on GitHub.com. To read about how to create a database with the CodeQL CLI, see "Preparing your code for CodeQL analysis."
Downloading a database from GitHub
GitHub stores CodeQL databases for over 200,000 repositories on GitHub.com.
You can check if a repository has any CodeQL databases available for download, and if so download it, using the REST API. For more information, see "List CodeQL databases for a repository" and "Get a CodeQL database for a repository" in the GitHub REST API documentation.
Choosing a database to analyze
-
Hover over the title bar of the "Databases" view and choose the appropriate icon to select your database. You can select a local database (from a ZIP archive or an unarchived folder), from a public URL, or from a project's URL on GitHub.com.
-
Once you've chosen a database, it will be displayed in the "Databases" view. To see the menu options for interacting with a database, right-click an entry in the list. You can select multiple databases at once.
Note: You can also analyze test databases. Test databases (folders with a .testproj
extension) are generated when you run regression tests on custom queries using the CodeQL CLI. If a query fails a regression test, you may want to import the test database into Visual Studio Code to debug the failure. For more information about running query tests, see "Testing custom queries."
Filtering databases and queries by language
Optionally, to see databases containing a specific language and queries written for that language, you can apply a language filter using the language selector.
-
To see available language filters, in the sidebar, open the "Language" view.
-
Hover over the language filter you would like to apply, then click Select.
Next steps
To learn how to use the extension to analyze your projects by running queries on CodeQL databases, see "Running CodeQL queries."