Skip to content

codeql_database_create.prompt.md

Latest commit

 

History

History
204 lines (137 loc) · 6.54 KB

codeql_database_create.prompt.md

File metadata and controls

204 lines (137 loc) · 6.54 KB
description Reference for the codeql database create command

Command Resource for codeql database create

The codeql database create command is used to create CodeQL databases from source code repositories. This can take a long time, depending upon the size of the codebase from which the data(base) is to be extracted.

Primary use of codeql database create

The following is an example use of the command for creating a database from a Java project:

$ codeql database create --language=java --source-root=/path/to/project mydb

Alternative uses of codeql database create

The codeql database create command can also be used with different language and build configurations:

# Multiple languages
$ codeql database create --language=java,javascript --source-root=. mydb

# With custom build command for compiled languages
$ codeql database create --language=cpp --command="make all" mydb

# Auto-build mode
$ codeql database create --language=java --build-mode=autobuild mydb

Help for codeql database create

Run codeql database create --help for more information. Run codeql database create --help --verbose for much more information.

Commands commonly run BEFORE codeql database create

Commands commonly run AFTER codeql database create

Command Synopsis

codeql database create [OPTIONS] -- <database>

Description

Create a CodeQL database by analyzing source code. The database creation process extracts semantic information from source code, enabling subsequent query analysis. Supports multiple programming languages and build modes.

Key Options

Required Arguments

  • <database> - Mandatory Path where the new database will be created

Language and Build Configuration

  • --language=<lang>[,<lang>...] - Programming languages to analyze (auto-detected from GitHub repos if omitted with GITHUB_TOKEN)
  • --build-mode=<mode> - Build mode for database creation:
    • none: No building required (C#, Java, JavaScript/TypeScript, Python, Ruby)
    • autobuild: Automatic build detection (C/C++, C#, Go, Java/Kotlin, Swift)
    • manual: Manual build command (C/C++, C#, Go, Java/Kotlin, Swift)

Source Code Configuration

  • -s, --source-root=<dir> - Root source code directory (default: current directory)
  • -c, --command=<command> - Build command for compiled languages

Performance Options

  • -j, --threads=<num> - Threads for import operation (default: 1, 0 = cores, -N = leave N cores)
  • -M, --ram=<MB> - Memory for import operation

Advanced Database Options

  • --no-cleanup - Advanced Suppress database cleanup after finalization (debugging)
  • --no-pre-finalize - Advanced Skip pre-finalize script from extractor
  • --[no-]skip-empty - Advanced Warn instead of failing for empty databases
  • --[no-]linkage-aware-import - Advanced Control linkage-aware import (default: enabled)

Baseline and Coverage

  • --[no-]calculate-baseline - Calculate baseline information about analyzed code
  • --[no-]sublanguage-file-coverage - GitHub.com/GHES 3.12+ Use sub-language file coverage

Extractor Configuration

  • --search-path=<dir>[:<dir>...] - Directories containing extractor packs
  • -O, --extractor-option=<name=value> - Set extractor options
  • --extractor-options-file=<file> - JSON/YAML file with extractor options

GitHub Integration

  • -a, --github-auth-stdin - Accept GitHub Apps token for language auto-detection
  • -g, --github-url=<url> - GitHub instance URL (auto-detected from checkout)

Build Customization

  • --working-dir=<dir> - Advanced Directory for build command execution
  • --no-run-unnecessary-builds - Advanced Only run builds when extractors need them
  • --no-tracing - Advanced Don't trace build command
  • --extra-tracing-config=<file> - Advanced Custom tracer configuration

Language-Specific Build Modes

No Build Required

# Python, JavaScript, Ruby
codeql database create --language=python mydb

Auto-build

# Java, C#, Go with automatic build detection
codeql database create --language=java --build-mode=autobuild mydb

Manual Build

# C/C++ with custom build command
codeql database create --language=cpp --command="make all" mydb

Common Usage Patterns

Single language with auto-detection

codeql database create --language=java --source-root=/path/to/project mydb

Multiple languages

codeql database create --language=java,javascript --source-root=. mydb

With custom build command

codeql database create --language=cpp --command="cmake --build build/" mydb

GitHub repository with auto-detection

# Requires GITHUB_TOKEN environment variable
codeql database create --source-root=/path/to/repo mydb

High-performance configuration

codeql database create --language=java --threads=8 --ram=16384 mydb

With extractor options

codeql database create --language=java --extractor-option=java.buildtools.maven.M2_HOME=/usr/local/maven mydb

When to Use

  • Analyzing source code repositories for security vulnerabilities
  • Setting up databases for custom query development
  • CI/CD integration for continuous security analysis
  • Research and investigation of codebases
  • Preparing data for bulk query analysis

Expected Outputs

  • Complete CodeQL database in specified directory
  • Extraction logs and progress information
  • Database finalization and cleanup
  • Error messages for build or extraction failures

Database Structure

mydb/
├── codeql-database.yml    # Database metadata
├── db-<language>/         # Language-specific data
├── log/                   # Extraction logs
├── working/               # Temporary working files (cleaned up)
└── src.zip               # Source code archive

Related Commands