code-graph-rag 0.0.187__tar.gz → 0.0.209__tar.gz

{code_graph_rag-0.0.187/code_graph_rag.egg-info → code_graph_rag-0.0.209}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-graph-rag
-Version: 0.0.187
+Version: 0.0.209
 Summary: The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
 License-Expression: MIT
 Keywords: rag,retrieval-augmented-generation,knowledge-graph,code-analysis,tree-sitter,mcp,mcp-server,llm,graph-database,semantic-search,codebase,memgraph,developer-tools,monorepo
@@ -38,6 +38,7 @@ Requires-Dist: protobuf>=6.33.5
 Requires-Dist: defusedxml>=0.7.1
 Requires-Dist: huggingface-hub[hf-xet]>=1.7.2
 Requires-Dist: griffe<2,>=1.0
+Requires-Dist: pathspec>=1.0.4
 Provides-Extra: test
 Requires-Dist: pytest>=8.4.1; extra == "test"
 Requires-Dist: pytest-asyncio>=1.0.0; extra == "test"
@@ -99,7 +100,7 @@ The package installs a `cgr` command.
 **Start Memgraph, parse a repo, and query it:**
 
 ```bash
-docker compose up -d                       # start Memgraph
+cgr daemon up                              # start Memgraph + Qdrant
 cgr start --repo-path ./my-project \
           --update-graph --clean           # parse & launch interactive chat
 ```

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209}/PYPI_README.md

@@ -34,7 +34,7 @@ The package installs a `cgr` command.
 **Start Memgraph, parse a repo, and query it:**
 
 ```bash
-docker compose up -d                       # start Memgraph
+cgr daemon up                              # start Memgraph + Qdrant
 cgr start --repo-path ./my-project \
           --update-graph --clean           # parse & launch interactive chat
 ```

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209}/README.md

@@ -20,7 +20,7 @@
     <img src="https://img.shields.io/github/forks/vitali87/code-graph-rag?style=social" alt="GitHub forks" />
   </a>
   -->
-  <!-- Codecov, SonarCloud, and OpenSSF Scorecard badges were fed by GitHub-side CI. They no longer update after the move to Codeberg. Restore once equivalent CI is in place on Codeberg (Woodpecker).
+  <!-- Codecov, SonarCloud, and OpenSSF Scorecard badges are fed by GitHub-side CI. Uncomment once that CI is confirmed to publish to these services again.
   <a href="https://codecov.io/gh/vitali87/code-graph-rag">
     <img src="https://codecov.io/gh/vitali87/code-graph-rag/graph/badge.svg" alt="Codecov" />
   </a>
@@ -39,7 +39,7 @@
   <a href="https://pepy.tech/projects/code-graph-rag">
     <img src="https://static.pepy.tech/personalized-badge/code-graph-rag?period=total&units=INTERNATIONAL_SYSTEM&left_color=BLACK&right_color=GREEN&left_text=downloads" alt="PyPI Downloads" />
   </a>
-  <!-- OpenSSF Scorecard only tracks GitHub-hosted repos. Restore if the project is mirrored back to GitHub or scorecard.dev adds Codeberg support.
+  <!-- OpenSSF Scorecard only tracks GitHub-hosted repos. Uncomment once the Scorecard workflow is confirmed running on the authoritative GitHub repo.
   <a href="https://scorecard.dev/viewer/?uri=github.com/vitali87/code-graph-rag">
     <img src="https://api.scorecard.dev/projects/github.com/vitali87/code-graph-rag/badge" alt="OpenSSF Scorecard" />
   </a>
@@ -187,7 +187,7 @@ volumes (`memgraph_data`, `memgraph_log`, `qdrant_storage`).
 ### Local development install
 
 ```bash
-git clone https://codeberg.org/vitali87/code-graph-rag.git
+git clone https://github.com/vitali87/code-graph-rag.git
 cd code-graph-rag
 ```
 
@@ -292,7 +292,7 @@ ollama pull llama3.2
 
 4. **Start Memgraph database**:
 ```bash
-docker compose up -d
+cgr daemon up
 ```
 
 5. **Verify installation**:
@@ -331,6 +331,7 @@ Use the Makefile for common development tasks:
 | `make format` | Run ruff format |
 | `make typecheck` | Run type checking with ty |
 | `make check` | Run all checks: lint, typecheck, test |
+| `make release` | Build, verify, and publish the current pyproject version to PyPI, then tag and create a GitHub Release |
 | `make pre-commit` | Run all pre-commit checks locally (comprehensive test before commit) |
 <!-- /SECTION:makefile_commands -->
 
@@ -674,6 +675,7 @@ The knowledge graph uses the following node types and relationships:
 | ModuleInterface | `{qualified_name: string, name: string, path: string, absolute_path: string}` |
 | ModuleImplementation | `{qualified_name: string, name: string, path: string, absolute_path: string, implements_module: string}` |
 | ExternalPackage | `{name: string, version_spec: string}` |
+| ExternalModule | `{qualified_name: string, name: string, path: string}` |
 <!-- /SECTION:node_schemas -->
 
 ### Language-Specific Mappings
@@ -703,7 +705,7 @@ The knowledge graph uses the following node types and relationships:
 | Project, Package, Folder | CONTAINS_MODULE | Module |
 | Module | DEFINES | Class, Function |
 | Class | DEFINES_METHOD | Method |
-| Module | IMPORTS | Module |
+| Module | IMPORTS | Module, ExternalModule |
 | Module | EXPORTS | Class, Function |
 | Module | EXPORTS_MODULE | ModuleInterface |
 | Module | IMPLEMENTS_MODULE | ModuleImplementation |
@@ -713,6 +715,8 @@ The knowledge graph uses the following node types and relationships:
 | ModuleImplementation | IMPLEMENTS | ModuleInterface |
 | Project | DEPENDS_ON_EXTERNAL | ExternalPackage |
 | Function, Method | CALLS | Function, Method |
+| Module, Function, Method | REFERENCES | Function, Method |
+| Module, Function, Method | INSTANTIATES | Class |
 <!-- /SECTION:relationship_schemas -->
 
 ## 🔧 Configuration
@@ -754,27 +758,28 @@ Configuration is managed through environment variables in `.env` file:
 
 ### Custom Ignore Patterns
 
-You can specify additional directories to exclude by creating a `.cgrignore` file in your repository root:
+You can specify additional files and directories to exclude by creating a `.cgrignore` file in your repository root. Patterns follow `.gitignore` conventions:
 
 ```
 # Comments start with #
 vendor
-.custom_cache
-my_build_output
+*.gen.ts
+docs/*.md
+/generated
+!bin/keep.py
 ```
 
-- One directory name per line
-- Lines starting with `#` are comments
-- Blank lines are ignored
-- Patterns are exact directory name matches (not globs)
-- Patterns from `.cgrignore` are merged with `--exclude` flags and auto-detected directories
+- Gitignore syntax: `*` matches within a segment, `**` crosses segments, bare names match at any depth, slash-containing patterns are anchored to the root, trailing slash matches directories only
+- Lines starting with `!` un-ignore paths that a default exclusion would skip (explicit excludes always win)
+- Lines starting with `#` are comments; blank lines are ignored
+- Patterns from `.cgrignore` are merged with `--exclude` flags (same syntax) and auto-detected directories
 
 ### Key Dependencies
 
 <!-- SECTION:dependencies -->
 - **loguru**: Python logging made (stupidly) simple
 - **mcp**: Model Context Protocol SDK
-- **pydantic-ai**: Agent Framework / shim to use Pydantic with LLMs
+- **pydantic-ai**: AI Agent Framework, the Pydantic way
 - **pydantic-settings**: Settings management using Pydantic
 - **pymgclient**: Memgraph database adapter for Python language
 - **python-dotenv**: Read key-value pairs from a .env file and set them as environment variables
@@ -792,6 +797,7 @@ my_build_output
 - **defusedxml**: XML bomb protection for Python stdlib modules
 - **huggingface-hub**: Client library to download and publish models, datasets and other repos on the huggingface.co hub
 - **griffe**: Signatures for entire Python programs. Extract the structure, the frame, the skeleton of your project, to generate API documentation or find breaking changes in your API.
+- **pathspec**: Utility library for gitignore style pattern matching of file paths.
 <!-- /SECTION:dependencies -->
 
 ## 🤖 Agentic Workflow & Tools
@@ -835,7 +841,7 @@ Code-Graph-RAG makes it easy to add support for any language that has a Tree-sit
 
 > **⚠️ Recommendation**: While you can add languages yourself, we recommend waiting for official full support to ensure optimal parsing quality, comprehensive feature coverage, and robust integration. The languages marked as "In Development" above will receive dedicated optimization and testing.
 
-> **💡 Request Support**: If you want a specific language to be officially supported, please [submit an issue](https://codeberg.org/vitali87/code-graph-rag/issues) with your language request.
+> **💡 Request Support**: If you want a specific language to be officially supported, please [submit an issue](https://github.com/vitali87/code-graph-rag/issues) with your language request.
 
 #### Quick Start: Add a Language
 

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209/code_graph_rag.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-graph-rag
-Version: 0.0.187
+Version: 0.0.209
 Summary: The ultimate RAG for your monorepo. Query, understand, and edit multi-language codebases with the power of AI and knowledge graphs
 License-Expression: MIT
 Keywords: rag,retrieval-augmented-generation,knowledge-graph,code-analysis,tree-sitter,mcp,mcp-server,llm,graph-database,semantic-search,codebase,memgraph,developer-tools,monorepo
@@ -38,6 +38,7 @@ Requires-Dist: protobuf>=6.33.5
 Requires-Dist: defusedxml>=0.7.1
 Requires-Dist: huggingface-hub[hf-xet]>=1.7.2
 Requires-Dist: griffe<2,>=1.0
+Requires-Dist: pathspec>=1.0.4
 Provides-Extra: test
 Requires-Dist: pytest>=8.4.1; extra == "test"
 Requires-Dist: pytest-asyncio>=1.0.0; extra == "test"
@@ -99,7 +100,7 @@ The package installs a `cgr` command.
 **Start Memgraph, parse a repo, and query it:**
 
 ```bash
-docker compose up -d                       # start Memgraph
+cgr daemon up                              # start Memgraph + Qdrant
 cgr start --repo-path ./my-project \
           --update-graph --clean           # parse & launch interactive chat
 ```

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209}/code_graph_rag.egg-info/SOURCES.txt

@@ -44,9 +44,11 @@ codebase_rag/parsers/call_processor.py
 codebase_rag/parsers/call_resolver.py
 codebase_rag/parsers/definition_processor.py
 codebase_rag/parsers/dependency_parser.py
+codebase_rag/parsers/export_detection.py
 codebase_rag/parsers/factory.py
 codebase_rag/parsers/function_ingest.py
 codebase_rag/parsers/import_processor.py
+codebase_rag/parsers/python_source_roots.py
 codebase_rag/parsers/stdlib_extractor.py
 codebase_rag/parsers/structure_processor.py
 codebase_rag/parsers/type_inference.py
@@ -61,12 +63,14 @@ codebase_rag/parsers/class_ingest/parent_extraction.py
 codebase_rag/parsers/class_ingest/relationships.py
 codebase_rag/parsers/class_ingest/utils.py
 codebase_rag/parsers/cpp/__init__.py
+codebase_rag/parsers/cpp/type_inference.py
 codebase_rag/parsers/cpp/utils.py
 codebase_rag/parsers/cpp_frontend/__init__.py
 codebase_rag/parsers/cpp_frontend/constants.py
 codebase_rag/parsers/cpp_frontend/frontend.py
 codebase_rag/parsers/cpp_frontend/qn.py
 codebase_rag/parsers/go/__init__.py
+codebase_rag/parsers/go/type_inference.py
 codebase_rag/parsers/go/utils.py
 codebase_rag/parsers/handlers/__init__.py
 codebase_rag/parsers/handlers/base.py

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209}/code_graph_rag.egg-info/requires.txt

@@ -18,6 +18,7 @@ protobuf>=6.33.5
 defusedxml>=0.7.1
 huggingface-hub[hf-xet]>=1.7.2
 griffe<2,>=1.0
+pathspec>=1.0.4
 
 [semantic]
 qdrant-client>=1.9.0

{code_graph_rag-0.0.187 → code_graph_rag-0.0.209}/codebase_rag/constants.py

@@ -185,6 +185,7 @@ KEY_EXPORTED_AT = "exported_at"
 KEY_PARSER = "parser"
 KEY_NAME = "name"
 KEY_QUALIFIED_NAME = "qualified_name"
+KEY_IS_PROPERTY = "is_property"
 KEY_QUERY = "query"
 KEY_RESPONSE = "response"
 KEY_START_LINE = "start_line"
@@ -201,7 +202,6 @@ KEY_TO_VAL = "to_val"
 KEY_VERSION_SPEC = "version_spec"
 KEY_PREFIX = "prefix"
 KEY_PROJECT_NAME = "project_name"
-KEY_IS_EXTERNAL = "is_external"
 
 ERR_SUBSTR_ALREADY_EXISTS = "already exists"
 ERR_SUBSTR_CONSTRAINT = "constraint"
@@ -227,6 +227,7 @@ ONEOF_FUNCTION = "function"
 ONEOF_METHOD = "method"
 ONEOF_FILE = "file"
 ONEOF_EXTERNAL_PACKAGE = "external_package"
+ONEOF_EXTERNAL_MODULE = "external_module"
 ONEOF_MODULE_IMPLEMENTATION = "module_implementation"
 ONEOF_MODULE_INTERFACE = "module_interface"
 ONEOF_INTERFACE = "interface_node"
@@ -389,6 +390,34 @@ DUP_QN_MARKER = "@"
 PATH_CURRENT_DIR = "."
 PATH_PARENT_DIR = ".."
 GLOB_ALL = "*"
+# (H) JS/TS import specifier schemes that name genuinely external code (node
+# (H) builtins, package registries, URLs). A specifier with any OTHER scheme
+# (H) (`ext:` deno-runtime aliases, bundler virtual modules) points at first-party
+# (H) code under a non-file-path name, so its unresolved calls defer to the trie.
+JS_EXTERNAL_IMPORT_SCHEMES: frozenset[str] = frozenset(
+    {"node", "npm", "jsr", "bun", "http", "https", "data", "file", "blob"}
+)
+# (H) Module file extensions stripped when turning a tsconfig `paths` target into a
+# (H) module qn (`src/util.ts` -> `src/util`), longest first so `.d.ts`-like
+# (H) compound suffixes are handled before the bare `.ts`.
+JS_TS_MODULE_EXTENSIONS: tuple[str, ...] = (
+    ".d.ts",
+    ".tsx",
+    ".ts",
+    ".jsx",
+    ".mjs",
+    ".cjs",
+    ".js",
+)
+TSCONFIG_FILENAMES: tuple[str, ...] = (
+    "tsconfig.json",
+    "tsconfig.base.json",
+    "jsconfig.json",
+)
+JS_INDEX_STEM = "index"
+TS_COMPILER_OPTIONS_KEY = "compilerOptions"
+TS_PATHS_KEY = "paths"
+TS_BASE_URL_KEY = "baseUrl"
 PATH_RELATIVE_PREFIX = "./"
 PATH_PARENT_PREFIX = "../"
 CPP_IMPORT_PARTITION_PREFIX = "import :"
@@ -417,12 +446,14 @@ class QueryFormat(StrEnum):
 
 
 # (H) Decorators whose presence marks a function/method as an implicit entry point
-# (H) (web routes, task/flow handlers, fixtures, CLI commands, event listeners).
+# (H) (web routes, task/flow handlers, fixtures, CLI commands, event listeners, and
+# (H) Pydantic validators/serializers the framework invokes by registration).
 DEFAULT_ROOT_DECORATORS: frozenset[str] = frozenset(
     {
         "route",
         "get",
         "post",
+        "callback",
         "put",
         "delete",
         "patch",
@@ -435,11 +466,37 @@ DEFAULT_ROOT_DECORATORS: frozenset[str] = frozenset(
         "app",
         "on_event",
         "listener",
+        "validator",
+        "field_validator",
+        "model_validator",
+        "root_validator",
+        "field_serializer",
+        "model_serializer",
+        "abstractmethod",
     }
 )
 
-# (H) Substrings in a node's file path that mark it as test code.
-TEST_PATH_PATTERNS: tuple[str, ...] = ("test_", "_test", "conftest", "/tests/")
+# (H) Base classes that mark a class as a structural interface: its method stubs
+# (H) are never call targets themselves (callers resolve to the implementations),
+# (H) so dead-code analysis roots every method the class defines.
+# (H) ponytail: direct bases only; transitive Protocol subclassing is not chased.
+PROTOCOL_BASE_QNS: tuple[str, ...] = ("typing.Protocol", "typing_extensions.Protocol")
+
+# (H) Substrings in a node's file path that mark it as test code. Covers Python
+# (H) (test_, _test, conftest, /tests/), the JS/TS filename convention
+# (H) (foo.test.ts, foo.spec.tsx), and the Jest __tests__/ directory so those
+# (H) files are not reported as dead. Singular /test/ and /spec/ directories are
+# (H) intentionally excluded: they collide with product code (a domain "spec"
+# (H) module), which would misclassify live code as test.
+TEST_PATH_PATTERNS: tuple[str, ...] = (
+    "test_",
+    "_test",
+    "conftest",
+    "/tests/",
+    ".test.",
+    ".spec.",
+    "__tests__",
+)
 
 
 class NodeLabel(StrEnum):
@@ -458,6 +515,7 @@ class NodeLabel(StrEnum):
     MODULE_INTERFACE = "ModuleInterface"
     MODULE_IMPLEMENTATION = "ModuleImplementation"
     EXTERNAL_PACKAGE = "ExternalPackage"
+    EXTERNAL_MODULE = "ExternalModule"
 
 
 _NODE_LABEL_UNIQUE_KEYS: dict[NodeLabel, UniqueKeyType] = {
@@ -476,6 +534,7 @@ _NODE_LABEL_UNIQUE_KEYS: dict[NodeLabel, UniqueKeyType] = {
     NodeLabel.MODULE_INTERFACE: UniqueKeyType.QUALIFIED_NAME,
     NodeLabel.MODULE_IMPLEMENTATION: UniqueKeyType.QUALIFIED_NAME,
     NodeLabel.EXTERNAL_PACKAGE: UniqueKeyType.NAME,
+    NodeLabel.EXTERNAL_MODULE: UniqueKeyType.QUALIFIED_NAME,
 }
 
 _missing_keys = set(NodeLabel) - set(_NODE_LABEL_UNIQUE_KEYS.keys())
@@ -501,6 +560,7 @@ class RelationshipType(StrEnum):
     IMPLEMENTS = "IMPLEMENTS"
     OVERRIDES = "OVERRIDES"
     CALLS = "CALLS"
+    REFERENCES = "REFERENCES"
     INSTANTIATES = "INSTANTIATES"
     DEPENDS_ON_EXTERNAL = "DEPENDS_ON_EXTERNAL"
 
@@ -721,6 +781,7 @@ FIELD_DECLARATOR = "declarator"
 FIELD_PARAMETERS = "parameters"
 FIELD_RECEIVER = "receiver"
 FIELD_TYPE = "type"
+FIELD_RESULT = "result"
 # (H) Rust impl `trait`/`type` fields and a trait's supertrait `bounds`.
 FIELD_TRAIT = "trait"
 FIELD_BOUNDS = "bounds"
@@ -1059,6 +1120,9 @@ IGNORE_SUFFIXES = frozenset(
     {".tmp", "~", ".pyc", ".pyo", ".o", ".a", ".so", ".dll", ".class"}
 )
 
+# (H) pathspec style for .cgrignore / --exclude patterns (#495).
+GITWILDMATCH_STYLE = "gitignore"
+
 PAYLOAD_NODE_ID = "node_id"
 PAYLOAD_QUALIFIED_NAME = "qualified_name"
 
@@ -1080,7 +1144,7 @@ CYPHER_DELETE_CALLS = "MATCH ()-[r:CALLS]->() DELETE r"
 # (H) Removes external import-target Module nodes that no module imports anymore
 # (H) (e.g. an imported name that was renamed/removed on an incremental rebuild).
 CYPHER_DELETE_ORPHAN_EXTERNAL_MODULES = (
-    "MATCH (m:Module) WHERE m.is_external = true AND NOT (m)<--() DETACH DELETE m"
+    "MATCH (m:ExternalModule) WHERE NOT (m)<--() DETACH DELETE m"
 )
 
 # (H) Queries for orphan pruning — returns all paths stored in the graph
@@ -1088,13 +1152,62 @@ CYPHER_ALL_FILE_PATHS = (
     "MATCH (f:File) RETURN f.path AS path, f.absolute_path AS absolute_path"
 )
 CYPHER_ALL_MODULE_PATHS_INTERNAL = (
-    "MATCH (m:Module) WHERE m.is_external IS NULL OR m.is_external = false "
-    "RETURN m.path AS path, m.qualified_name AS qualified_name"
+    "MATCH (m:Module) RETURN m.path AS path, m.qualified_name AS qualified_name"
 )
 CYPHER_ALL_FOLDER_PATHS = (
     "MATCH (f:Folder) RETURN f.path AS path, f.absolute_path AS absolute_path"
 )
 
+# (H) Rehydrate the in-memory function registry on an incremental run: returns
+# (H) every definition node's qualified name and label so call/instantiation
+# (H) resolution can see symbols defined in files that were not re-parsed.
+CYPHER_ALL_DEFINITION_QNS = (
+    "MATCH (n) WHERE n:Function OR n:Method OR n:Class OR n:Interface "
+    "OR n:Enum OR n:Type OR n:Union "
+    "RETURN n.qualified_name AS qualified_name, head(labels(n)) AS label, "
+    "n.is_property AS is_property"
+)
+
+# (H) Inbound reference edges (from unchanged files) into symbols defined in one
+# (H) of $paths. Captured BEFORE a changed file's subtree is deleted so the exact
+# (H) edges can be restored verbatim afterwards (issue #532, inbound half).
+# (H) Re-resolving the callers instead would diverge from a clean index, because
+# (H) cgr's call resolution is context-sensitive (protocol vs concrete receiver,
+# (H) import granularity); the original edges already match a clean re-index.
+CYPHER_INBOUND_EDGES = (
+    "MATCH (caller)-[r:CALLS|REFERENCES|INSTANTIATES|IMPORTS|INHERITS|OVERRIDES]->(target) "
+    "WHERE target.path IN $paths AND caller.qualified_name IS NOT NULL "
+    "AND (caller.path IS NULL OR NOT caller.path IN $paths) "
+    "RETURN head(labels(caller)) AS caller_label, "
+    "caller.qualified_name AS caller_qn, type(r) AS rel, "
+    "head(labels(target)) AS target_label, target.qualified_name AS target_qn"
+)
+# (H) Rehydrate class_inheritance on an incremental run: every INHERITS edge
+# (H) (child -> base) with resolved qns, so protocol dispatch and inherited-method
+# (H) resolution still see the hierarchy of classes defined in files that were not
+# (H) re-parsed. Without it, editing a caller drops the protocol/inheritance
+# (H) redirect (issue #532 residual): a call resolves to the Protocol stub instead
+# (H) of the concrete implementer because _protocol_classes() is empty. Ordered by
+# (H) base_index so multiple-inheritance base order matches the original source,
+# (H) which method resolution and override attribution depend on.
+CYPHER_ALL_INHERITS = (
+    "MATCH (child)-[r:INHERITS]->(base) "
+    "WHERE child.qualified_name IS NOT NULL AND base.qualified_name IS NOT NULL "
+    "RETURN child.qualified_name AS child_qn, base.qualified_name AS base_qn, "
+    "r.base_index AS base_index "
+    "ORDER BY child_qn, base_index"
+)
+KEY_CHILD_QN = "child_qn"
+KEY_BASE_QN = "base_qn"
+KEY_BASE_INDEX = "base_index"
+
+CYPHER_PARAM_PATHS = "paths"
+KEY_CALLER_LABEL = "caller_label"
+KEY_CALLER_QN = "caller_qn"
+KEY_REL = "rel"
+KEY_TARGET_LABEL = "target_label"
+KEY_TARGET_QN = "target_qn"
+
 REALTIME_LOGGER_FORMAT = (
     "<green>{time:YYYY-MM-DD HH:mm:ss.SSS}</green> | "
     "<level>{level: <8}</level> | "
@@ -1124,6 +1237,9 @@ DIST_DIR = "dist"
 BYTES_PER_MB_FLOAT = 1024 * 1024
 
 PYPROJECT_PATH = "pyproject.toml"
+PYPROJECT_KEY_TOOL = "tool"
+PYPROJECT_KEY_SETUPTOOLS = "setuptools"
+PYPROJECT_KEY_PACKAGE_DIR = "package-dir"
 TREESITTER_EXTRA_KEY = "treesitter-full"
 TREESITTER_PKG_PREFIX = "tree-sitter-"
 
@@ -1764,6 +1880,18 @@ class CppNodeType(StrEnum):
     CONSTRUCTOR_OR_DESTRUCTOR_DECLARATION = "constructor_or_destructor_declaration"
     INLINE_METHOD_DEFINITION = "inline_method_definition"
     OPERATOR_CAST_DEFINITION = "operator_cast_definition"
+    TYPE_IDENTIFIER = "type_identifier"
+    PARAMETER_LIST = "parameter_list"
+    PARAMETER_DECLARATION = "parameter_declaration"
+    OPTIONAL_PARAMETER_DECLARATION = "optional_parameter_declaration"
+    INIT_DECLARATOR = "init_declarator"
+    TEMPLATE_TYPE = "template_type"
+    FIELD_EXPRESSION = "field_expression"
+    COMPOUND_STATEMENT = "compound_statement"
+    THIS = "this"
+    TYPE_DEFINITION = "type_definition"
+    ALIAS_DECLARATION = "alias_declaration"
+    TYPE_DESCRIPTOR = "type_descriptor"
 
 
 CPP_MODULE_EXTENSIONS = (".ixx", ".cppm", ".ccm", ".mxx")
@@ -1796,6 +1924,13 @@ CPP_KEYWORD_CLASS = "class"
 CPP_KEYWORD_STRUCT = "struct"
 CPP_EXPORTED_CLASS_KEYWORDS = frozenset({CPP_KEYWORD_CLASS, CPP_KEYWORD_STRUCT})
 
+# (H) A C/C++ class/struct/union tag with no body is a forward declaration
+# (H) (`class Widget;`); it must not become its own node, or it collides with the
+# (H) real definition's qn and fragments one class into several same-named nodes.
+CPP_TYPE_SPECIFIER_NODE_TYPES = frozenset(
+    {"class_specifier", "struct_specifier", "union_specifier"}
+)
+
 CPP_FALLBACK_OPERATOR = "operator_unknown"
 CPP_FALLBACK_DESTRUCTOR = "~destructor"
 CPP_OPERATOR_TEXT_PREFIX = "operator"
@@ -1916,6 +2051,10 @@ TS_VARIABLE_DECLARATOR = "variable_declarator"
 TS_CALL_EXPRESSION = "call_expression"
 TS_EXPORT_CLAUSE = "export_clause"
 TS_EXPORT_SPECIFIER = "export_specifier"
+TS_EXPORT_DEFAULT = "default"
+TS_ACCESSIBILITY_MODIFIER = "accessibility_modifier"
+TS_PRIVATE = "private"
+TS_PRIVATE_PROPERTY_IDENTIFIER = "private_property_identifier"
 
 # (H) Tree-sitter Java import node types
 TS_IMPORT_DECLARATION = "import_declaration"
@@ -1989,11 +2128,27 @@ TS_GO_TYPE_ALIAS = "type_alias"
 TS_GO_STRUCT_TYPE = "struct_type"
 TS_GO_INTERFACE_TYPE = "interface_type"
 TS_GO_PARAMETER_DECLARATION = "parameter_declaration"
+TS_GO_FUNC_LITERAL = "func_literal"
 TS_GO_SOURCE_FILE = "source_file"
 TS_GO_FUNCTION_DECLARATION = "function_declaration"
 TS_GO_METHOD_DECLARATION = "method_declaration"
 TS_GO_CALL_EXPRESSION = "call_expression"
 TS_GO_IMPORT_DECLARATION = "import_declaration"
+TS_GO_PARAMETER_LIST = "parameter_list"
+TS_GO_VAR_DECLARATION = "var_declaration"
+TS_GO_VAR_SPEC = "var_spec"
+TS_GO_SHORT_VAR_DECLARATION = "short_var_declaration"
+TS_GO_EXPRESSION_LIST = "expression_list"
+TS_GO_COMPOSITE_LITERAL = "composite_literal"
+TS_GO_UNARY_EXPRESSION = "unary_expression"
+TS_GO_POINTER_TYPE = "pointer_type"
+# (H) Go composite types a method may return; a chained call lands on the CONTAINER,
+# (H) not its element, so return-type inference must not unwrap these to an element
+# (H) name (a `[]Command` return must not resolve `.Run()` to `Command.Run`).
+TS_GO_CONTAINER_TYPES: frozenset[str] = frozenset(
+    {"slice_type", "array_type", "map_type", "channel_type", "function_type"}
+)
+FIELD_OPERAND = "operand"
 
 # (H) Tree-sitter Scala node types
 TS_SCALA_CLASS_DEFINITION = "class_definition"
@@ -2036,6 +2191,7 @@ TS_PHP_OBJECT_CREATION_EXPRESSION = "object_creation_expression"
 TS_PHP_NAMESPACE_DEFINITION = "namespace_definition"
 TS_PHP_NAMESPACE_USE_DECLARATION = "namespace_use_declaration"
 TS_PHP_NAMESPACE_USE_CLAUSE = "namespace_use_clause"
+TS_PHP_FUNCTION = "function"
 TS_PHP_INCLUDE_EXPRESSION = "include_expression"
 TS_PHP_INCLUDE_ONCE_EXPRESSION = "include_once_expression"
 TS_PHP_REQUIRE_EXPRESSION = "require_expression"
@@ -2070,6 +2226,10 @@ TS_CPP_BINARY_EXPRESSION = "binary_expression"
 TS_CPP_UNARY_EXPRESSION = "unary_expression"
 TS_CPP_UPDATE_EXPRESSION = "update_expression"
 TS_CPP_FUNCTION_DECLARATOR = "function_declarator"
+# (H) Substring shared by C++ declarator node types (pointer_declarator,
+# (H) reference_declarator, parenthesized_declarator, ...), used to unwrap a
+# (H) parameter declarator down to its bound identifier.
+CPP_DECLARATOR_SUFFIX = "declarator"
 
 # (H) Tree-sitter Java node types for language_spec
 TS_JAVA_METHOD_INVOCATION = "method_invocation"
@@ -2085,6 +2245,7 @@ TS_CLASS_HERITAGE = "class_heritage"
 TS_IMPLEMENTS_CLAUSE = "implements_clause"
 TS_EXTENDS_CLAUSE = "extends_clause"
 TS_MEMBER_EXPRESSION = "member_expression"
+TS_SELECTOR_EXPRESSION = "selector_expression"
 TS_EXTENDS = "extends"
 TS_ARGUMENTS = "arguments"
 TS_EXTENDS_TYPE_CLAUSE = "extends_type_clause"
@@ -2103,6 +2264,16 @@ FIELD_OPERATOR = "operator"
 # (H) Derived node type tuples for class ingestion
 CPP_CLASS_TYPES = (CppNodeType.CLASS_SPECIFIER, TS_STRUCT_SPECIFIER)
 CPP_COMPOUND_TYPES = (*CPP_CLASS_TYPES, TS_UNION_SPECIFIER, TS_ENUM_SPECIFIER)
+# (H) Node types that open their own variable scope; C++ local-variable inference must
+# (H) not descend into them, or a name declared inside a lambda / nested function /
+# (H) local class body would be attributed to the enclosing function's scope.
+CPP_NESTED_SCOPE_NODE_TYPES = frozenset(
+    (
+        TS_CPP_FUNCTION_DEFINITION,
+        TS_CPP_LAMBDA_EXPRESSION,
+        *CPP_COMPOUND_TYPES,
+    )
+)
 JS_TS_PARENT_REF_TYPES = (TS_IDENTIFIER, TS_MEMBER_EXPRESSION)
 
 # (H) Import processor function names
@@ -2452,6 +2623,10 @@ TS_PY_TYPED_DEFAULT_PARAMETER = "typed_default_parameter"
 TS_PY_ATTRIBUTE = "attribute"
 TS_PY_CALL = "call"
 TS_PY_LIST = "list"
+TS_PY_DICTIONARY = "dictionary"
+TS_PY_PAIR = "pair"
+TS_PY_SET = "set"
+TS_PY_TUPLE = "tuple"
 TS_PY_LIST_COMPREHENSION = "list_comprehension"
 TS_PY_FOR_STATEMENT = "for_statement"
 TS_PY_FOR_IN_CLAUSE = "for_in_clause"
@@ -2461,6 +2636,7 @@ PY_RETURN_QUERY = "(return_statement) @return_stmt"
 TS_PY_CLASS_DEFINITION = "class_definition"
 TS_PY_BLOCK = "block"
 TS_PY_FUNCTION_DEFINITION = "function_definition"
+TS_PY_LAMBDA = "lambda"
 TS_PY_RETURN_STATEMENT = "return_statement"
 TS_PY_RETURN = "return"
 TS_PY_KEYWORD = "keyword"
@@ -2509,6 +2685,10 @@ PY_NONE = "None"
 # (H) Python keyword identifiers
 PY_KEYWORD_SELF = "self"
 PY_KEYWORD_CLS = "cls"
+# (H) Visibility by naming convention: a leading underscore marks a private
+# (H) symbol, while a dunder (__x__) is public API invoked by the runtime.
+PY_NAME_UNDERSCORE = "_"
+PY_NAME_DUNDER = "__"
 # (H) typing.Protocol base name and the conventional XxxProtocol class suffix
 # (H) used to map a Protocol to its concrete implementer.
 PY_PROTOCOL = "Protocol"
@@ -2526,6 +2706,7 @@ HIGHER_ORDER_BUILTINS: frozenset[str] = frozenset({"sorted", "min", "max", "redu
 
 # (H) Python attribute prefixes
 PY_SELF_PREFIX = "self."
+PY_CLS_PREFIX = "cls."
 
 # (H) Python type inference patterns
 PY_VAR_PATTERN_ALL = "all_"
@@ -2550,8 +2731,13 @@ GUARD_INHERITED_METHOD = "_inherited_method_guard"
 # (H) JS/TS ingest node types
 TS_PAIR = "pair"
 TS_OBJECT = "object"
+TS_ARRAY = "array"
 TS_FUNCTION_EXPRESSION = "function_expression"
 TS_ARROW_FUNCTION = "arrow_function"
+TS_REQUIRED_PARAMETER = "required_parameter"
+TS_OPTIONAL_PARAMETER = "optional_parameter"
+TS_FIELD_PATTERN = "pattern"
+TS_FIELD_PARAMETER = "parameter"
 TS_MODULE = "module"
 TS_CLASS_BODY = "class_body"
 TS_STATIC = "static"