v0.2.1

Install codeanalyzer-python v0.2.1

Shell script (installs the canpy CLI via uv / pipx / pip):

curl --proto '=https' --tlsv1.2 -LsSf https://github.com/codellm-devkit/codeanalyzer-python/releases/latest/download/canpy-installer.sh | sh

PyPI:

pip install codeanalyzer-python==0.2.1

For the optional live Neo4j push (--emit neo4j --neo4j-uri ...):

pip install 'codeanalyzer-python[neo4j]==0.2.1'

Download

File	Description
codeanalyzer_python-0.2.1-py3-none-any.whl	Python wheel
codeanalyzer_python-0.2.1.tar.gz	Source distribution
canpy-installer.sh	Shell installer (uv / pipx / pip)
schema.json	Neo4j schema contract

What's Changed

🚀 Features

• Analysis venv (uv + Jedi wiring), external_symbols, app-scoped prune, --no-venv (#44 #45 #46 #47) by @rahlk in #48

Full Changelog: v0.2.0...v0.2.1

v0.2.0

Install codeanalyzer-python v0.2.0

Shell script (installs the canpy CLI via uv / pipx / pip):

curl --proto '=https' --tlsv1.2 -LsSf https://github.com/codellm-devkit/codeanalyzer-python/releases/latest/download/canpy-installer.sh | sh

PyPI:

pip install codeanalyzer-python==0.2.0

For the optional live Neo4j push (--emit neo4j --neo4j-uri ...):

pip install 'codeanalyzer-python[neo4j]==0.2.0'

Download

File	Description
codeanalyzer_python-0.2.0-py3-none-any.whl	Python wheel
codeanalyzer_python-0.2.0.tar.gz	Source distribution
canpy-installer.sh	Shell installer (uv / pipx / pip)
schema.json	Neo4j schema contract

What's Changed

🚀 Features

• feature/neo4j: property-graph output with Py/PY_-namespaced schema; rename CLI to canpy by @rahlk in #33

🛠 Other Changes

• Fix CodeQL call-graph edges dropped on (file, start_line) join miss (#25) by @rahlk in #26

New contributors and full diff: https://github.com/codellm-devkit/codeanalyzer-python/releases/tag/v0.2.0

v0.1.14

Release Notes (from CHANGELOG.md)

Added

• Call graph in analysis output: PyApplication.call_graph: List[PyCallEdge]. Every run now produces a call graph in addition to the symbol table. Edges carry source, target (both PyCallable.signature), weight, and provenance (jedi / codeql / joern).
• call_graph module (codeanalyzer.semantic_analysis.call_graph) with to_digraph / from_digraph networkx adapters, jedi_call_graph_edges, and merge_edges. Endpoints absent from the symbol table become ghost nodes so RPC / third-party / framework edges are preserved.
• CodeQL Python query rewritten against the CodeQL Python library (was Java idioms before). Resolves direct calls and constructor calls via ClassValue.lookup("__init__"), using the modern Value.getACall() predicate (CodeQL Python 7.x).
• augment_call_sites: when --codeql is enabled, CodeQL backfills PyCallsite.callee_signature entries Jedi left unresolved.
• resolve_unresolved_constructors: heuristic fallback that walks the symbol table by class short-name and scope to fill in constructor sites neither Jedi nor CodeQL resolved (common for classes nested inside functions/methods). Synthesizes <class>.__init__ signatures.
• iter_classes_in_symbol_table: full recursive walker over classes — including inner classes, classes nested in functions, and classes nested in class methods.

Changed

• BREAKING: Removed --analysis-level / analysis_level. The call graph is built unconditionally; use --codeql/--no-codeql to control CodeQL participation. Jedi-derived edges are always available.
• Jedi constructor calls now resolve to <class>.__init__ (was: bare <class>). When script.infer() returns a class, the qualified name is rewritten to point at the constructor — matching where method PyCallables actually live in the symbol table. PyCallsite.is_constructor_call now reflects Jedi's type inference (was: method_name == "__init__", only true for explicit obj.__init__() calls).
• _call_sites scope correctness: replaced naive ast.walk with _iter_calls_in_scope, which stops at nested FunctionDef / AsyncFunctionDef / ClassDef bodies (those have their own PyCallable.call_sites). Decorators, default arguments, return annotations, base classes and class keyword args are still walked since they execute in the enclosing scope. Previously, outer functions over-attributed every call from every nested definition.
• CodeQL CLI binary is now downloaded into <cache_dir>/codeql/bin/ (per-project, respecting --cache-dir) and discovered before any CodeQL operation — including when the database cache is reused. The downloaded archive is removed after extraction.
• CodeQLQueryRunner now accepts the resolved binary path instead of relying on PATH. The temporary .ql file is written inside a per-project qlpack (<cache_dir>/codeql/qlpack/) whose codeql/python-all dependency is resolved once via codeql pack install, eliminating the lockfile / search-path gymnastics.

Fixed

• zipfile extraction dropped Unix permissions on the CodeQL CLI launcher, causing PermissionError on first query run. Entries are now extracted with their stored external_attr mode applied, plus a defensive chmod +x on the resolved binary.
• rglob("codeql") matched the bundled codeql/codeql/ directory before the launcher file, returning a directory instead of an executable. Both CodeQLLoader and _ensure_codeql_bin now filter to is_file().
• CodeQLQueryRunner crashed on subprocess errors with 'NoneType' object has no attribute 'stderr' because stderr=None returns None from communicate(). Now captures stderr=PIPE and decodes bytes safely.

---

Detailed Changes (auto-generated)

🚀 Features

• Release 0.1.14: call graph + CodeQL integration
  • PR: #23

v0.1.13

Release Notes (from CHANGELOG.md)

Improved

• CLI Help Documentation: Comprehensive help text added for all command-line options
  • Added descriptive help messages for all CLI parameters including --output, --format, --analysis-level, etc.
  • Enhanced user experience with clear option descriptions in --help output
  • Improved CLI parameter organization using Annotated type hints for better maintainability
  • Added case-insensitive support for --format option
  • Updated verbosity option help to clearly indicate multiple usage (-v, -vv, -vvv)

Technical Details

• Refactored CLI function signature to use consistent Annotated type hint pattern
• Added comprehensive help text for all 12 command-line options
• Improved code organization and type safety in CLI parameter definitions

---

Detailed Changes (auto-generated)

v0.1.12

Release Notes (from CHANGELOG.md)

Changed

• BREAKING CHANGE: Refactored Codeanalyzer constructor to use AnalysisOptions dataclass in response to #12
  • Replaced multiple individual parameters with single AnalysisOptions object for cleaner API
  • Improved type safety and configuration management through centralized options structure
  • Enhanced maintainability and extensibility for future configuration additions
  • Updated CLI integration to create and pass AnalysisOptions instance
  • Maintained backward compatibility in terms of functionality while improving code architecture

Added

• New AnalysisOptions dataclass in codeanalyzer.options module in response to #12
  • Centralized configuration structure with all analysis parameters
  • Type-safe configuration with proper defaults and validation
  • Support for OutputFormat enum integration
  • Clean separation between CLI and library configuration handling

Technical Details

• Added new codeanalyzer.options package with AnalysisOptions dataclass
• Updated Codeanalyzer.__init__() to accept single options parameter instead of 9 individual parameters
• Modified CLI handler in __main__.py to create AnalysisOptions instance from command line arguments
• Improved code organization and maintainability for configuration management
• Enhanced API design following best practices for parameter object patterns

---

Detailed Changes (auto-generated)

v0.1.11

Release Notes (from CHANGELOG.md)

Fixed

• CRITICAL: Fixed NumPy build failure on Python 3.12+ (addresses #19)
  • Updated NumPy dependency constraints to handle Python 3.12+ compatibility
  • Split NumPy version constraints into three tiers:
    • numpy>=1.21.0,<1.24.0 for Python < 3.11
    • numpy>=1.24.0,<2.0.0 for Python 3.11.x
    • numpy>=1.26.0,<2.0.0 for Python 3.12+ (requires NumPy 1.26+ which supports Python 3.12)
  • Resolves ModuleNotFoundError: No module named 'distutils' errors on Python 3.12+
  • Ensures compatibility with Python 3.12 which removed distutils from the standard library
• Fixed Pydantic v1/v2 compatibility issues in JSON serialization throughout codebase
  • Added comprehensive Pydantic version detection and compatibility layer
  • Introduced model_dump_json() and model_validate_json() helper functions for cross-version compatibility
  • Fixed PyApplication.parse_raw() deprecated method usage (replaced with model_validate_json())
  • Updated CLI output methods to use compatible serialization functions
  • Resolved forward reference updates only for Pydantic v1 (v2 handles these automatically)

Changed

• Enhanced Pydantic compatibility infrastructure in schema module
  • Added runtime Pydantic version detection using importlib.metadata
  • Created compatibility abstraction layer for JSON serialization/deserialization
  • Improved forward reference resolution logic to work with both Pydantic v1 and v2
  • Updated all JSON serialization calls to use new compatibility functions
  • Better error handling for missing Pydantic dependency

Technical Details

• Added packaging dependency for robust version comparison
• Enhanced schema module with runtime version detection and compatibility helpers
• Updated core analysis caching system to use compatible Pydantic JSON methods
• Improved CLI output formatting with cross-version Pydantic support

---

Detailed Changes (auto-generated)

🐛 Fixes

• Fix issue #19
  • PR: #20

v0.1.10

Release Notes (from CHANGELOG.md)

Added

• Ray distributed processing support for parallel symbol table generation (addresses #16)
• --ray/--no-ray CLI flag to enable/disable Ray-based distributed analysis
• --skip-tests/--include-tests CLI flag to control whether test files are analyzed (improves analysis performance)
• --file-name CLI flag for single file analysis (addresses part of #16)
• Incremental caching system with SHA256-based file change detection
  • Automatic caching of analysis results to analysis_cache.json
  • File-level caching with content hash validation to avoid re-analyzing unchanged files
  • Significant performance improvements for subsequent analysis runs
  • Cache reuse statistics logging
• Custom exception classes for better error handling in symbol table building:
  • SymbolTableBuilderException (base exception)
  • SymbolTableBuilderFileNotFoundError (file not found errors)
  • SymbolTableBuilderParsingError (parsing errors)
  • SymbolTableBuilderRayError (Ray processing errors)
• Enhanced PyModule schema with metadata fields for caching:
  • last_modified timestamp tracking
  • content_hash for precise change detection
• Progress bar support for both serial and parallel processing modes
• Enhanced test fixtures including xarray project for comprehensive testing
• Comprehensive __init__.py exports for syntactic analysis module
• Smart dependency installation with conditional logic:
  • Only installs requirements files when they exist (requirements.txt, requirements-dev.txt, dev-requirements.txt, test-requirements.txt)
  • Only performs editable installation when package definition files are present (pyproject.toml, setup.py, setup.cfg)
  • Improved virtual environment setup with better dependency detection and installation logic

Changed

• BREAKING CHANGE: Updated Python version requirement from >=3.10 to >=3.9 for broader compatibility (closes #17)
• BREAKING CHANGE: Updated dependency versions with more conservative constraints for better stability:
  • pydantic downgraded from >=2.11.7 to >=1.8.0,<2.0.0 for stability
  • pandas constrained to >=1.3.0,<2.0.0
  • numpy constrained to >=1.21.0,<1.24.0
  • rich constrained to >=12.6.0,<14.0.0
  • typer constrained to >=0.9.0,<1.0.0
  • Other dependencies updated with conservative version ranges for better compatibility
• Major Architecture Enhancement: Complete rewrite of analysis caching system
  • analyze() method now implements intelligent caching with PyApplication serialization
  • Symbol table building redesigned to support incremental updates and cache reuse
  • File change detection using SHA256 content hashing for maximum accuracy
• Enhanced Codeanalyzer constructor signature to accept file_name parameter for single file analysis
• Refactored symbol table building from monolithic build() method to cache-aware file-level processing
• Enhanced Codeanalyzer constructor signature to accept skip_tests and using_ray parameters
• Improved error handling with proper context managers in core analyzer
• Updated CLI to use Pydantic v1 compatible JSON serialization methods
• Reorganized syntactic analysis module structure with proper exception handling and exports
• Enhanced virtual environment detection with better fallback mechanisms
• Symbol table builder now sets metadata fields (last_modified, content_hash) for all PyModule objects

Fixed

• Fixed critical symbol table bug for nested functions (closes #15)
  • Corrected _callables() method recursion logic to properly capture both outer and inner functions
  • Previously, only inner/nested functions were being captured in the symbol table
  • Now correctly processes module-level functions, class methods, and all nested function definitions
• Fixed nested method/function signature generation in symbol table builder
  • Corrected _callables() method to properly build fully qualified signatures for nested structures
  • Fixed issue where nested functions and methods were getting incorrect signatures (e.g., main.__init__ instead of main.outer_function.NestedClass.__init__)
  • Added prefix parameter to _callables() and _add_class() methods to maintain proper nesting context
  • Signatures now correctly reflect the full nested hierarchy (e.g., main.outer_function.NestedClass.nested_class_method.method_nested_function)
  • Updated class method processing to pass class signature as prefix to nested callable processing
  • Improved path relativization to project directory for cleaner signature generation
• Fixed Pydantic v2 compatibility issues by reverting to v1 API (json() instead of model_dump_json())
• Fixed missing import statements and type annotations throughout the codebase
• Fixed symbol table builder to support individual file processing for distributed execution
• Improved error handling in virtual environment detection and Python interpreter resolution
• Fixed schema type annotations to use proper string keys for better serialization
• Enhanced import ordering and removed unnecessary blank lines in CLI module
• Improved virtual environment setup reliability:
  • Fixed unnecessary pip installs by adding conditional logic to only install when dependencies are available
  • Only attempts to install requirements files if they actually exist in the project
  • Only performs editable installation when package definition files are present
  • Prevents errors and warnings from attempting to install non-existent dependencies

Technical Details

• Added Ray as a core dependency for distributed computing capabilities (addresses #16)
• Implemented @ray.remote decorator for parallel file processing
• Comprehensive caching system implementation:
  • _load_pyapplication_from_cache() and _save_analysis_cache() methods for PyApplication serialization
  • _file_unchanged() method with SHA256 content hash validation
  • Cache-aware symbol table building with selective file processing
  • Automatic cache statistics and performance reporting
• Enhanced progress tracking for both serial and parallel execution modes with Rich progress bars
• Updated schema to use Dict[str, PyModule] instead of dict[Path, PyModule] for better serialization
• Extended PyModule schema with optional last_modified and content_hash fields for caching metadata
• Added comprehensive exception hierarchy for better error classification and handling
• Refactored symbol table building into modular, file-level processing suitable for distribution
• Enhanced Python interpreter detection with support for multiple version managers (pyenv, conda, asdf)
• Added hashlib integration for file content hashing throughout the codebase
• Enhanced virtual environment setup logic:
  • Modified _add_class() method to accept prefix parameter and pass class signature to method processing
  • Updated _callables() method signature to include prefix parameter for nested context tracking
  • Enhanced signature building logic to use prefix when available, falling back to Jedi resolution for top-level definitions
  • Fixed recursive calls to pass current signature as prefix for proper nesting hierarchy
  • Implemented conditional dependency installation with existence checks for requirements files and package definition files

Notes

• This release significantly addresses the performance improvements requested in #16:
  • ✅ Ray parallelization implemented
  • ✅ Incremental caching with SHA256-based change detection implemented
  • ✅ --file-name option for single-file analysis implemented
  • ❌ --nproc options not yet included (still uses all available cores with Ray)
• ✅ Critical bug fix for nested function detection (#15) is now included in this version
• Expected performance improvements: 2-10x faster on subsequent runs depending on code change frequency
• Enhanced symbol table accuracy ensures all function definitions are properly captured
• Virtual environment setup is now more robust and only installs dependencies when they are actually available

---

Detailed Changes (auto-generated)

• no changes

v0.1.9

Release Notes (from CHANGELOG.md)

Fixed

• Fixed AttributeError: 'OutputFormat' object has no attribute 'casefold' when using --format flag with case-insensitive options
• Changed OutputFormat enum to inherit from str to support typer's case-insensitive string processing

---

Detailed Changes (auto-generated)

• no changes

v0.1.8

Release Notes (from CHANGELOG.md)

Added

• Added missing config module with OutputFormat enum for better code organization
• Added proper __init__.py and config.py files to the config directory

Fixed

• Fixed missing config directory files that were not previously tracked in git

---

Detailed Changes (auto-generated)

• no changes

v0.1.7

Release Notes (from CHANGELOG.md)

Changed

• Relaxed Python version requirement from ==3.10.* to >=3.10 for improved flexibility
• Enhanced compatibility to support Python 3.10+ versions while maintaining backward compatibility

---

Detailed Changes (auto-generated)

• no changes