codemap-cpp 0.1.0a1__tar.gz

codemap_cpp-0.1.0a1/.gitignore

@@ -0,0 +1,43 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Build artifacts
+build/
+dist/
+*.egg-info/
+*.egg
+.eggs/
+
+# Test / coverage
+.pytest_cache/
+.coverage
+.coverage.*
+htmlcov/
+coverage.xml
+.tox/
+.mypy_cache/
+.ruff_cache/
+.benchmarks/
+
+# Virtualenv
+.venv/
+venv/
+env/
+
+# uv / pdm lockfiles (commit uv.lock once we settle)
+# uv.lock
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+
+# OS
+.DS_Store
+Thumbs.db
+
+# CodeMap own index when dogfooding
+.codemap/

codemap_cpp-0.1.0a1/PKG-INFO

@@ -0,0 +1,61 @@
+Metadata-Version: 2.4
+Name: codemap-cpp
+Version: 0.1.0a1
+Summary: C++ language indexer plugin for CodeMap
+Project-URL: Homepage, https://github.com/qxbyte/codemap
+Author: CodeMap Contributors
+License: MIT
+Keywords: c++,codemap,cpp,indexer,tree-sitter
+Classifier: Development Status :: 3 - Alpha
+Classifier: Programming Language :: C++
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Software Development
+Requires-Python: >=3.11
+Requires-Dist: codemap-core<0.2,>=0.1.0a1
+Requires-Dist: tree-sitter-cpp>=0.23
+Requires-Dist: tree-sitter>=0.25
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# codemap-cpp
+
+> A C++ language indexer for [CodeMap](https://github.com/qxbyte/codemap),
+> shipped as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-cpp`:
+
+| AST node | Symbol kind |
+|---|---|
+| `namespace_definition` | namespace prefix (recursed into) |
+| `class_specifier` (named, with body) | `class` (`extra.cpp_kind=class`) |
+| `struct_specifier` (named, with body) | `class` (`extra.cpp_kind=struct`) |
+| `union_specifier` (named, with body) | `class` (`extra.cpp_kind=union`) |
+| `enum_specifier` | `class` (`extra.cpp_kind=enum`) |
+| `function_definition` (top-level or in namespace) | `function` |
+| `function_definition` (inside class body) | `method` |
+| `field_declaration` (data member) | `field` |
+| `template_declaration` wrapping any of the above | unwraps to the inner declaration |
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-cpp"
+```
+
+## File patterns
+
+* `*.cpp`, `*.cc`, `*.cxx`, `*.hpp`, `*.hh`, `*.hxx`
+
+## Limits
+
+* Out-of-class method definitions (``void Foo::bar() { ... }``) appear as
+  free functions, not as members of ``Foo``.
+* `using` declarations and aliases are not emitted as symbols.
+* Macro-only ``#define`` constants are not captured (see ``codemap-c``).
+
+## License
+
+MIT.

codemap_cpp-0.1.0a1/README.md

@@ -0,0 +1,41 @@
+# codemap-cpp
+
+> A C++ language indexer for [CodeMap](https://github.com/qxbyte/codemap),
+> shipped as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-cpp`:
+
+| AST node | Symbol kind |
+|---|---|
+| `namespace_definition` | namespace prefix (recursed into) |
+| `class_specifier` (named, with body) | `class` (`extra.cpp_kind=class`) |
+| `struct_specifier` (named, with body) | `class` (`extra.cpp_kind=struct`) |
+| `union_specifier` (named, with body) | `class` (`extra.cpp_kind=union`) |
+| `enum_specifier` | `class` (`extra.cpp_kind=enum`) |
+| `function_definition` (top-level or in namespace) | `function` |
+| `function_definition` (inside class body) | `method` |
+| `field_declaration` (data member) | `field` |
+| `template_declaration` wrapping any of the above | unwraps to the inner declaration |
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-cpp"
+```
+
+## File patterns
+
+* `*.cpp`, `*.cc`, `*.cxx`, `*.hpp`, `*.hh`, `*.hxx`
+
+## Limits
+
+* Out-of-class method definitions (``void Foo::bar() { ... }``) appear as
+  free functions, not as members of ``Foo``.
+* `using` declarations and aliases are not emitted as symbols.
+* Macro-only ``#define`` constants are not captured (see ``codemap-c``).
+
+## License
+
+MIT.

codemap_cpp-0.1.0a1/pyproject.toml

@@ -0,0 +1,36 @@
+[build-system]
+requires = ["hatchling>=1.21"]
+build-backend = "hatchling.build"
+
+[project]
+name = "codemap-cpp"
+version = "0.1.0a1"
+description = "C++ language indexer plugin for CodeMap"
+readme = "README.md"
+requires-python = ">=3.11"
+license = { text = "MIT" }
+authors = [{ name = "CodeMap Contributors" }]
+keywords = ["codemap", "cpp", "c++", "indexer", "tree-sitter"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Programming Language :: C++",
+    "Programming Language :: Python :: 3",
+    "Topic :: Software Development",
+]
+dependencies = [
+    "codemap-core>=0.1.0a1,<0.2",
+    "tree-sitter>=0.25",
+    "tree-sitter-cpp>=0.23",
+]
+
+[project.optional-dependencies]
+dev = ["pytest>=8.0"]
+
+[project.entry-points."codemap.indexers"]
+cpp = "codemap_cpp:CppIndexer"
+
+[project.urls]
+Homepage = "https://github.com/qxbyte/codemap"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/codemap_cpp"]

codemap_cpp-0.1.0a1/src/codemap_cpp/__init__.py

@@ -0,0 +1,8 @@
+"""C++ indexer plugin for CodeMap."""
+
+from __future__ import annotations
+
+from codemap_cpp.indexer import CppIndexer
+
+__all__ = ["CppIndexer"]
+__version__ = "0.1.0"

codemap_cpp-0.1.0a1/src/codemap_cpp/indexer.py

@@ -0,0 +1,376 @@
+"""C++ indexer built on tree-sitter-cpp.
+
+The visitor recurses into ``namespace_definition`` so that descriptors of
+nested types are prefixed with the containing namespace chain.
+``template_declaration`` wrappers are unwrapped so the underlying
+function / class / struct surfaces as a normal symbol.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path, PurePosixPath
+from typing import ClassVar
+
+import tree_sitter
+import tree_sitter_cpp
+
+from codemap.core.models import Diagnostic, Edge, IndexResult, Range, Symbol
+from codemap.core.symbol import Descriptor, DescriptorKind, SymbolID
+from codemap.indexers.base import IndexContext
+
+SCHEME = "scip-cpp"
+LANG = "cpp"
+
+_CPP_LANG = tree_sitter.Language(tree_sitter_cpp.language())
+
+
+class CppIndexer:
+    name: ClassVar[str] = "cpp"
+    version: ClassVar[str] = "0.1.0"
+    file_patterns: ClassVar[list[str]] = [
+        "*.cpp",
+        "*.cc",
+        "*.cxx",
+        "*.hpp",
+        "*.hh",
+        "*.hxx",
+    ]
+    languages: ClassVar[list[str]] = [LANG]
+
+    def supports(self, path: Path) -> bool:
+        return path.suffix in {".cpp", ".cc", ".cxx", ".hpp", ".hh", ".hxx"}
+
+    def index_file(
+        self,
+        path: Path,
+        source: bytes,
+        ctx: IndexContext,
+    ) -> IndexResult:
+        try:
+            source.decode("utf-8")
+        except UnicodeDecodeError as exc:
+            return IndexResult(
+                diagnostics=[
+                    Diagnostic(
+                        severity="error",
+                        file=ctx.relative_path,
+                        code="CPP002",
+                        message=f"not valid UTF-8: {exc}",
+                        producer=self.name,
+                    )
+                ]
+            )
+        parser = tree_sitter.Parser(_CPP_LANG)
+        tree = parser.parse(source)
+        visitor = _Visitor(ctx.relative_path)
+        visitor.visit_root(tree.root_node)
+        diagnostics = list(visitor.diagnostics)
+        if tree.root_node.has_error:
+            diagnostics.append(
+                Diagnostic(
+                    severity="warning",
+                    file=ctx.relative_path,
+                    range=Range(start_line=1, end_line=1),
+                    code="CPP001",
+                    message="tree-sitter reported parse errors; symbols may be incomplete",
+                    producer=self.name,
+                )
+            )
+        return IndexResult(
+            symbols=visitor.symbols,
+            edges=visitor.edges,
+            diagnostics=diagnostics,
+        )
+
+
+class _Visitor:
+    def __init__(self, relative_path: PurePosixPath) -> None:
+        self.relative_path = relative_path
+        self.symbols: list[Symbol] = []
+        self.edges: list[Edge] = []
+        self.diagnostics: list[Diagnostic] = []
+
+    def visit_root(self, root: tree_sitter.Node) -> None:
+        for child in root.children:
+            self._visit(child, scope=[])
+
+    def _visit(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+    ) -> None:
+        kind = node.type
+        if kind == "namespace_definition":
+            self._visit_namespace(node, scope=scope)
+        elif kind == "template_declaration":
+            # Unwrap: a template is metadata around a single declaration.
+            inner = _template_inner(node)
+            if inner is not None:
+                self._visit(inner, scope=scope)
+        elif kind == "function_definition":
+            self._emit_function(node, scope=scope, container_kind=None)
+        elif kind == "class_specifier":
+            self._emit_record(node, scope=scope, cpp_kind="class")
+        elif kind == "struct_specifier":
+            self._emit_record(node, scope=scope, cpp_kind="struct")
+        elif kind == "union_specifier":
+            self._emit_record(node, scope=scope, cpp_kind="union")
+        elif kind == "enum_specifier":
+            self._emit_enum(node, scope=scope)
+
+    def _visit_namespace(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+    ) -> None:
+        name = _first_child_text(node, "namespace_identifier")
+        new_scope = list(scope)
+        if name is not None:
+            new_scope.append(Descriptor(name=name, kind=DescriptorKind.NAMESPACE))
+        body = _first_child(node, "declaration_list")
+        if body is None:
+            return
+        for child in body.children:
+            self._visit(child, scope=new_scope)
+
+    def _emit_function(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+        container_kind: str | None,
+    ) -> None:
+        name = _function_declarator_name(node)
+        if name is None:
+            return
+        descriptors = [*list(scope), Descriptor(name=name, kind=DescriptorKind.METHOD)]
+        kind = "method" if container_kind == "class" else "function"
+        self.symbols.append(
+            Symbol(
+                id=self._make_id(descriptors),
+                kind=kind,
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                signature=f"{name}()",
+            )
+        )
+
+    def _emit_record(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+        cpp_kind: str,
+    ) -> None:
+        name = _first_child_text(node, "type_identifier")
+        body = _first_child(node, "field_declaration_list")
+        if name is None or body is None:
+            return
+        record_desc = Descriptor(name=name, kind=DescriptorKind.TYPE)
+        descriptors = [*list(scope), record_desc]
+        self.symbols.append(
+            Symbol(
+                id=self._make_id(descriptors),
+                kind="class",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                extra={"cpp_kind": cpp_kind},
+            )
+        )
+        inner_scope = list(descriptors)
+        for child in body.children:
+            if child.type == "function_definition":
+                self._emit_function(child, scope=inner_scope, container_kind="class")
+            elif child.type == "field_declaration":
+                self._emit_field(child, scope=inner_scope)
+            elif child.type == "template_declaration":
+                inner = _template_inner(child)
+                if inner is not None:
+                    if inner.type == "function_definition":
+                        self._emit_function(inner, scope=inner_scope, container_kind="class")
+                    elif inner.type == "field_declaration":
+                        self._emit_field(inner, scope=inner_scope)
+
+    def _emit_field(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+    ) -> None:
+        # field_declaration may carry either a field_identifier (data member)
+        # or a function_declarator (method declaration without body).
+        decl = _first_child(node, "function_declarator")
+        if decl is not None:
+            name = _first_child_text(decl, "field_identifier")
+            if name is None:
+                name = _first_child_text(decl, "identifier")
+            if name is None:
+                return
+            self.symbols.append(
+                Symbol(
+                    id=self._make_id(
+                        [*list(scope), Descriptor(name=name, kind=DescriptorKind.METHOD)]
+                    ),
+                    kind="method",
+                    language=LANG,
+                    file=self.relative_path,
+                    range=_node_range(node),
+                    signature=f"{name}()",
+                )
+            )
+            return
+        name = _first_child_text(node, "field_identifier")
+        if name is None:
+            return
+        self.symbols.append(
+            Symbol(
+                id=self._make_id([*list(scope), Descriptor(name=name, kind=DescriptorKind.TERM)]),
+                kind="field",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+            )
+        )
+
+    def _emit_enum(
+        self,
+        node: tree_sitter.Node,
+        *,
+        scope: list[Descriptor],
+    ) -> None:
+        name = _first_child_text(node, "type_identifier")
+        if name is None:
+            return
+        enum_desc = Descriptor(name=name, kind=DescriptorKind.TYPE)
+        self.symbols.append(
+            Symbol(
+                id=self._make_id([*list(scope), enum_desc]),
+                kind="class",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                extra={"cpp_kind": "enum"},
+            )
+        )
+        body = _first_child(node, "enumerator_list")
+        if body is None:
+            return
+        for child in body.children:
+            if child.type == "enumerator":
+                enum_name = _first_child_text(child, "identifier")
+                if enum_name is None:
+                    continue
+                self.symbols.append(
+                    Symbol(
+                        id=self._make_id(
+                            [
+                                *list(scope),
+                                enum_desc,
+                                Descriptor(name=enum_name, kind=DescriptorKind.TERM),
+                            ]
+                        ),
+                        kind="field",
+                        language=LANG,
+                        file=self.relative_path,
+                        range=_node_range(child),
+                        extra={"cpp_kind": "enumerator"},
+                    )
+                )
+
+    def _make_id(self, descriptors: list[Descriptor]) -> SymbolID:
+        full = list(_path_namespaces(self.relative_path))
+        full.extend(descriptors)
+        return SymbolID(scheme=SCHEME, descriptors=tuple(full))
+
+
+# ---------------------------------------------------------------------------
+# Pure helpers
+# ---------------------------------------------------------------------------
+
+
+def _path_namespaces(path: PurePosixPath) -> list[Descriptor]:
+    return [Descriptor(name=part, kind=DescriptorKind.NAMESPACE) for part in path.parts]
+
+
+def _node_range(node: tree_sitter.Node) -> Range:
+    sr, sc = node.start_point
+    er, ec = node.end_point
+    return Range(
+        start_line=sr + 1,
+        start_col=sc,
+        end_line=max(er + 1, sr + 1),
+        end_col=ec,
+    )
+
+
+def _node_text(node: tree_sitter.Node) -> str:
+    return node.text.decode("utf-8") if node.text is not None else ""
+
+
+def _first_child(node: tree_sitter.Node, kind: str) -> tree_sitter.Node | None:
+    for child in node.children:
+        if child.type == kind:
+            return child
+    return None
+
+
+def _first_child_text(node: tree_sitter.Node, kind: str) -> str | None:
+    found = _first_child(node, kind)
+    return _node_text(found) if found is not None else None
+
+
+_INNER_DECL_TYPES = frozenset(
+    {
+        "function_definition",
+        "class_specifier",
+        "struct_specifier",
+        "union_specifier",
+        "enum_specifier",
+        "field_declaration",
+        "template_declaration",
+    }
+)
+
+
+def _template_inner(node: tree_sitter.Node) -> tree_sitter.Node | None:
+    for child in node.children:
+        if child.type in _INNER_DECL_TYPES:
+            return child
+    return None
+
+
+def _function_declarator_name(node: tree_sitter.Node) -> str | None:
+    declarator = _first_child(node, "function_declarator")
+    if declarator is None:
+        for child in node.children:
+            if child.type in {
+                "pointer_declarator",
+                "reference_declarator",
+                "parenthesized_declarator",
+            }:
+                name = _function_declarator_name(child)
+                if name is not None:
+                    return name
+        return None
+    # function_declarator's first identifier is the function name.
+    for child in declarator.children:
+        if child.type in {"identifier", "field_identifier"}:
+            return _node_text(child)
+        if child.type == "qualified_identifier":
+            # ``Foo::bar`` — take the rightmost component.
+            return _qualified_name(child)
+    return None
+
+
+def _qualified_name(node: tree_sitter.Node) -> str | None:
+    rightmost: str | None = None
+    for child in node.children:
+        if child.type in {"identifier", "field_identifier"}:
+            rightmost = _node_text(child)
+        elif child.type == "qualified_identifier":
+            rightmost = _qualified_name(child) or rightmost
+    return rightmost

codemap_cpp-0.1.0a1/tests/test_indexer.py

@@ -0,0 +1,173 @@
+"""Unit tests for the C++ indexer plugin."""
+
+from __future__ import annotations
+
+import textwrap
+from pathlib import Path, PurePosixPath
+
+from codemap_cpp import CppIndexer
+from codemap_cpp.indexer import SCHEME
+
+from codemap.core.models import IndexResult
+from codemap.indexers.base import IndexContext
+
+
+def _index(source: str, *, path: str = "src/lib.cpp") -> IndexResult:
+    code = textwrap.dedent(source).lstrip("\n")
+    return CppIndexer().index_file(
+        Path(path),
+        code.encode("utf-8"),
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath(path),
+            language="cpp",
+        ),
+    )
+
+
+def test_indexer_metadata() -> None:
+    ix = CppIndexer()
+    assert ix.name == "cpp"
+    assert ix.languages == ["cpp"]
+    for ext in (".cpp", ".cc", ".cxx", ".hpp", ".hh", ".hxx"):
+        assert ix.supports(Path(f"a{ext}"))
+    assert not ix.supports(Path("a.py"))
+
+
+def test_scheme_is_consistent() -> None:
+    r = _index("int x = 1; int f() { return 0; }")
+    for s in r.symbols:
+        assert str(s.id).startswith(f"{SCHEME} ")
+
+
+def test_free_function() -> None:
+    r = _index("int add(int a, int b) { return a + b; }")
+    funcs = [s for s in r.symbols if s.kind == "function"]
+    assert len(funcs) == 1
+    assert funcs[0].id.descriptors[-1].name == "add"
+
+
+def test_class_with_method_and_field() -> None:
+    r = _index(
+        """
+        class Foo {
+            int counter;
+            void bar() {}
+        };
+        """
+    )
+    classes = [s for s in r.symbols if s.kind == "class"]
+    methods = [s for s in r.symbols if s.kind == "method"]
+    fields = [s for s in r.symbols if s.kind == "field"]
+    assert len(classes) == 1
+    assert classes[0].extra["cpp_kind"] == "class"
+    assert {m.id.descriptors[-1].name for m in methods} == {"bar"}
+    assert {f.id.descriptors[-1].name for f in fields} == {"counter"}
+
+
+def test_struct_treated_as_class() -> None:
+    r = _index(
+        """
+        struct Point {
+            int x;
+            int y;
+        };
+        """
+    )
+    classes = [s for s in r.symbols if s.kind == "class"]
+    assert classes[0].extra["cpp_kind"] == "struct"
+
+
+def test_enum_class_with_enumerators() -> None:
+    r = _index("enum class Color { RED, GREEN };")
+    classes = [s for s in r.symbols if s.kind == "class"]
+    fields = [s for s in r.symbols if s.kind == "field"]
+    assert classes[0].extra["cpp_kind"] == "enum"
+    assert {f.id.descriptors[-1].name for f in fields} == {"RED", "GREEN"}
+
+
+def test_namespace_prefixes_descriptors() -> None:
+    r = _index(
+        """
+        namespace ns {
+            int gx = 1;
+            class Foo {
+                int bar;
+            };
+        }
+        """
+    )
+    # The outer namespace should appear as a NAMESPACE descriptor on
+    # Foo's symbol id.
+    foo = next(s for s in r.symbols if s.kind == "class")
+    name_chain = [d.name for d in foo.id.descriptors]
+    assert "ns" in name_chain
+    assert "Foo" in name_chain
+
+
+def test_nested_namespace_chain() -> None:
+    r = _index(
+        """
+        namespace outer {
+            namespace inner {
+                int compute() { return 0; }
+            }
+        }
+        """
+    )
+    compute = next(s for s in r.symbols if s.kind == "function")
+    name_chain = [d.name for d in compute.id.descriptors]
+    # outer and inner appear before compute().
+    assert name_chain.index("outer") < name_chain.index("inner") < name_chain.index("compute")
+
+
+def test_template_function_unwrapped() -> None:
+    r = _index(
+        """
+        template<typename T>
+        T identity(T value) { return value; }
+        """
+    )
+    funcs = [s for s in r.symbols if s.kind == "function"]
+    assert {f.id.descriptors[-1].name for f in funcs} == {"identity"}
+
+
+def test_template_class_unwrapped() -> None:
+    r = _index(
+        """
+        template<typename T>
+        class Box {
+            T value;
+        };
+        """
+    )
+    classes = [s for s in r.symbols if s.kind == "class"]
+    assert {c.id.descriptors[-1].name for c in classes} == {"Box"}
+
+
+def test_symbol_id_uses_path_namespaces() -> None:
+    r = _index("int x = 1;", path="src/util/math.cpp")
+    # Top-level declarations without init are not emitted; need a class for path test.
+    r = _index("struct S { int x; };", path="src/util/math.cpp")
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert str(cls.id) == "scip-cpp . . . src/util/math.cpp/S#"
+
+
+def test_empty_file_yields_no_symbols() -> None:
+    r = _index("")
+    assert r.symbols == []
+
+
+def test_invalid_utf8_yields_diagnostic() -> None:
+    ix = CppIndexer()
+    r = ix.index_file(
+        Path("bad.cpp"),
+        b"\xff\xfe class",
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath("bad.cpp"),
+            language="cpp",
+        ),
+    )
+    assert r.symbols == []
+    assert r.diagnostics[0].code == "CPP002"