Downloads 257this week

GitHub stars 3

Extension repository on GitHub

Extension descriptor (YAML)

Parse and analyze source code ASTs from 20+ programming languages with tree-sitter grammars

Installing and Loading

INSTALL sitting_duck FROM community;
LOAD sitting_duck;

Example

-- Parse Python code and extract function definitions
SELECT name, type, depth
FROM parse_ast('
def hello():
    return "hello world"
', 'python')
WHERE type = 'function_definition';

-- Parse JavaScript from a file
SELECT type, name, start_row, end_row
FROM read_ast('src/**/*.js')
WHERE type IN ('function_declaration', 'class_declaration');

-- Analyze code complexity by counting AST nodes
SELECT file_path, COUNT(*) as node_count
FROM read_ast('src/**/*.py')
GROUP BY file_path
ORDER BY node_count DESC;

-- Find all function calls in TypeScript
SELECT name, file_path, start_row
FROM read_ast('app.ts', language := 'typescript')
WHERE type = 'call_expression';

-- Cross-language analysis
SELECT language, COUNT(*) as definitions
FROM read_ast(['src/**/*.py', 'src/**/*.js', 'src/**/*.ts', 'src/**/*.go'])
WHERE type LIKE '%definition%' OR type LIKE '%declaration%'
GROUP BY language;

About sitting_duck

Sitting Duck is a DuckDB extension for parsing source code into Abstract Syntax Trees (ASTs) using tree-sitter grammars. It provides a SQL interface to analyze code structure across 20+ programming languages.

Core Table Functions:

• read_ast(file_pattern, language := NULL) - Parse source files into AST rows
• parse_ast(content, language) - Parse source code strings into AST rows

All pasring is highly flexible via parameters (see: https://github.com/teaguesterling/sitting_duck/blob/main/docs/api/parameters.md)

Supported Languages (27): | Category | Languages | |———-|———–| | Web | JavaScript, TypeScript, HTML, CSS | | Systems | C, C++, Go, Rust, Zig | | Scripting | Python, Ruby, PHP, Lua, R, Bash | | Enterprise | Java, C#, Kotlin, Swift | | Mobile | Dart | | Infrastructure | HCL (Terraform), JSON, TOML, GraphQL | | Documentation | SQL, Markdown |

AST Schema Fields: See: https://github.com/teaguesterling/sitting_duck/blob/main/docs/api/output-schema.md

• type - Node type (e.g., function_definition, class_declaration)
• name - Extracted identifier name when applicable
• file_path - Source file path
• language - Detected or specified language
• start_line, start_column - Node start position
• end_end, end_column - Node end position
• depth - Nesting depth in the AST
• parent_type - Type of parent node
• descendant_count - Number of descendant nodes
• peek - Customizable preview of node source
• semantic_type - Normalized node type across all langauges
• qualified_name - Extracted name and available scope qualifiers
• signature_type - Extracted type string for variables, expressions, and functions as available
• parameters - Extraacted function parameters, arguments, parent clases, etc. as available
• modifiers - Class, variable, parameter modifiers as available
• annotations - Decorators and annotations as avaialble

See https://github.com/teaguesterling/sitting_duck/blob/main/docs/native_extraction_semantics.md for additional details on "native" extraction for each langauge. Note: this is an active area of development.

Key Features:

• Automatic language detection from file extensions
• Glob pattern support for multi-file analysis
• Semantic type normalization across languages (https://github.com/teaguesterling/sitting_duck/blob/main/docs/api/semantic-types.md)
• Native, locally aware annotation of key node types for easy handling (https://github.com/teaguesterling/sitting_duck/blob/main/docs/native_extraction_semantics.md)
• Efficient tree-sitter parsing with pre-generated grammars
• DuckDB native SQL integration for complex queries

Perfect for code analysis, refactoring tools, documentation generation, codebase exploration, and AI-assisted development workflows.

Added Functions