codemap-kotlin 0.1.0a1__tar.gz

codemap_kotlin-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_kotlin-0.1.0a1/PKG-INFO

@@ -0,0 +1,61 @@
+Metadata-Version: 2.4
+Name: codemap-kotlin
+Version: 0.1.0a1
+Summary: Kotlin indexer plugin for CodeMap
+Project-URL: Homepage, https://github.com/qxbyte/codemap
+Author: CodeMap Contributors
+License: MIT
+Keywords: codemap,indexer,kotlin,tree-sitter
+Classifier: Development Status :: 3 - Alpha
+Classifier: Programming Language :: Kotlin
+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-kotlin>=1.1
+Requires-Dist: tree-sitter>=0.25
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# codemap-kotlin
+
+> A Kotlin indexer for [CodeMap](https://github.com/qxbyte/codemap),
+> shipped as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-kotlin`:
+
+| AST node | Symbol kind |
+|---|---|
+| `class_declaration` (keyword `class`) | `class` (with `extra.kotlin_kind=class`) |
+| `class_declaration` (keyword `interface`) | `class` (with `extra.kotlin_kind=interface`) |
+| `object_declaration` | `class` (with `extra.kotlin_kind=object`) |
+| `function_declaration` (free) | `function` |
+| `function_declaration` (inside type) | `method` |
+| `property_declaration` (top-level) | `variable` |
+| `property_declaration` (inside type) | `field` |
+
+`package_header` is captured as `extra.package` on type symbols.
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-kotlin"
+```
+
+## SymbolID encoding
+
+```
+scip-kotlin . . . src/main/kotlin/User.kt/User#hello().
+```
+
+## Limits
+
+* Companion object members are not attached to the enclosing class.
+* Generic-parameter descriptors are dropped.
+
+## License
+
+MIT.

codemap_kotlin-0.1.0a1/README.md

@@ -0,0 +1,41 @@
+# codemap-kotlin
+
+> A Kotlin indexer for [CodeMap](https://github.com/qxbyte/codemap),
+> shipped as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-kotlin`:
+
+| AST node | Symbol kind |
+|---|---|
+| `class_declaration` (keyword `class`) | `class` (with `extra.kotlin_kind=class`) |
+| `class_declaration` (keyword `interface`) | `class` (with `extra.kotlin_kind=interface`) |
+| `object_declaration` | `class` (with `extra.kotlin_kind=object`) |
+| `function_declaration` (free) | `function` |
+| `function_declaration` (inside type) | `method` |
+| `property_declaration` (top-level) | `variable` |
+| `property_declaration` (inside type) | `field` |
+
+`package_header` is captured as `extra.package` on type symbols.
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-kotlin"
+```
+
+## SymbolID encoding
+
+```
+scip-kotlin . . . src/main/kotlin/User.kt/User#hello().
+```
+
+## Limits
+
+* Companion object members are not attached to the enclosing class.
+* Generic-parameter descriptors are dropped.
+
+## License
+
+MIT.

codemap_kotlin-0.1.0a1/pyproject.toml

@@ -0,0 +1,36 @@
+[build-system]
+requires = ["hatchling>=1.21"]
+build-backend = "hatchling.build"
+
+[project]
+name = "codemap-kotlin"
+version = "0.1.0a1"
+description = "Kotlin indexer plugin for CodeMap"
+readme = "README.md"
+requires-python = ">=3.11"
+license = { text = "MIT" }
+authors = [{ name = "CodeMap Contributors" }]
+keywords = ["codemap", "kotlin", "indexer", "tree-sitter"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Kotlin",
+    "Topic :: Software Development",
+]
+dependencies = [
+    "codemap-core>=0.1.0a1,<0.2",
+    "tree-sitter>=0.25",
+    "tree-sitter-kotlin>=1.1",
+]
+
+[project.optional-dependencies]
+dev = ["pytest>=8.0"]
+
+[project.entry-points."codemap.indexers"]
+kotlin = "codemap_kotlin:KotlinIndexer"
+
+[project.urls]
+Homepage = "https://github.com/qxbyte/codemap"
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/codemap_kotlin"]

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

@@ -0,0 +1,8 @@
+"""Kotlin indexer plugin for CodeMap."""
+
+from __future__ import annotations
+
+from codemap_kotlin.indexer import KotlinIndexer
+
+__all__ = ["KotlinIndexer"]
+__version__ = "0.1.0"

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

@@ -0,0 +1,246 @@
+"""Kotlin indexer built on tree-sitter-kotlin.
+
+The Kotlin grammar uses ``class_declaration`` for both ``class`` and
+``interface``. The keyword token disambiguates. ``object_declaration``
+is its own node type for singleton objects.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path, PurePosixPath
+from typing import ClassVar
+
+import tree_sitter
+import tree_sitter_kotlin
+
+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-kotlin"
+LANG = "kotlin"
+
+_KT_LANG = tree_sitter.Language(tree_sitter_kotlin.language())
+
+
+class KotlinIndexer:
+    name: ClassVar[str] = "kotlin"
+    version: ClassVar[str] = "0.1.0"
+    file_patterns: ClassVar[list[str]] = ["*.kt", "*.kts"]
+    languages: ClassVar[list[str]] = [LANG]
+
+    def supports(self, path: Path) -> bool:
+        return path.suffix in {".kt", ".kts"}
+
+    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="KT002",
+                        message=f"not valid UTF-8: {exc}",
+                        producer=self.name,
+                    )
+                ]
+            )
+        parser = tree_sitter.Parser(_KT_LANG)
+        tree = parser.parse(source)
+        visitor = _Visitor(ctx.relative_path)
+        visitor.visit(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="KT001",
+                    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] = []
+        self._type_stack: list[str] = []
+        self._package: str = ""
+
+    def visit(self, node: tree_sitter.Node) -> None:
+        kind = node.type
+        if kind == "package_header":
+            self._set_package(node)
+            return
+        if kind == "class_declaration":
+            self._visit_class_or_interface(node)
+            return
+        if kind == "object_declaration":
+            self._visit_object(node)
+            return
+        if kind == "function_declaration":
+            self._visit_function(node)
+            return
+        if kind == "property_declaration":
+            self._visit_property(node)
+            return
+        for child in node.children:
+            self.visit(child)
+
+    # ----------------------------------------------------- packages
+
+    def _set_package(self, node: tree_sitter.Node) -> None:
+        for child in node.children:
+            if child.type == "qualified_identifier":
+                self._package = _node_text(child)
+                return
+            if child.type == "identifier":
+                self._package = _node_text(child)
+                return
+
+    # --------------------------------------------------- type-level
+
+    def _visit_class_or_interface(self, node: tree_sitter.Node) -> None:
+        kotlin_kind = "class"
+        for child in node.children:
+            if child.type == "interface":
+                kotlin_kind = "interface"
+                break
+        self._visit_type(node, kotlin_kind=kotlin_kind)
+
+    def _visit_object(self, node: tree_sitter.Node) -> None:
+        self._visit_type(node, kotlin_kind="object")
+
+    def _visit_type(self, node: tree_sitter.Node, *, kotlin_kind: str) -> None:
+        name = _first_identifier(node)
+        if name is None:
+            return
+        sid = self._make_id(name, kind=DescriptorKind.TYPE)
+        extra: dict[str, str] = {"kotlin_kind": kotlin_kind}
+        if self._package:
+            extra["package"] = self._package
+        self.symbols.append(
+            Symbol(
+                id=sid,
+                kind="class",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                extra=extra,
+            )
+        )
+        body = _find_class_body(node)
+        if body is None:
+            return
+        self._type_stack.append(name)
+        try:
+            for child in body.children:
+                self.visit(child)
+        finally:
+            self._type_stack.pop()
+
+    # ----------------------------------------------------- functions
+
+    def _visit_function(self, node: tree_sitter.Node) -> None:
+        name = _first_identifier(node)
+        if name is None:
+            return
+        kind: str = "method" if self._type_stack else "function"
+        sid = self._make_id(name, kind=DescriptorKind.METHOD)
+        self.symbols.append(
+            Symbol(
+                id=sid,
+                kind=kind,  # type: ignore[arg-type]
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                signature=f"fun {name}()",
+            )
+        )
+
+    # ---------------------------------------------------- properties
+
+    def _visit_property(self, node: tree_sitter.Node) -> None:
+        # property_declaration > variable_declaration > identifier
+        for child in node.children:
+            if child.type == "variable_declaration":
+                for grand in child.children:
+                    if grand.type == "identifier":
+                        name = _node_text(grand)
+                        if not name:
+                            return
+                        sym_kind: str = "field" if self._type_stack else "variable"
+                        sid = self._make_id(name, kind=DescriptorKind.TERM)
+                        self.symbols.append(
+                            Symbol(
+                                id=sid,
+                                kind=sym_kind,  # type: ignore[arg-type]
+                                language=LANG,
+                                file=self.relative_path,
+                                range=_node_range(node),
+                            )
+                        )
+                        return
+
+    # -------------------------------------------------------- helpers
+
+    def _make_id(self, name: str, *, kind: DescriptorKind) -> SymbolID:
+        descriptors = list(_path_namespaces(self.relative_path))
+        descriptors.extend(Descriptor(name=t, kind=DescriptorKind.TYPE) for t in self._type_stack)
+        descriptors.append(Descriptor(name=name, kind=kind))
+        return SymbolID(scheme=SCHEME, descriptors=tuple(descriptors))
+
+
+# ---------------------------------------------------------------------------
+# 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_identifier(node: tree_sitter.Node) -> str | None:
+    for child in node.children:
+        if child.type == "identifier":
+            return _node_text(child)
+    return None
+
+
+def _find_class_body(node: tree_sitter.Node) -> tree_sitter.Node | None:
+    for child in node.children:
+        if child.type == "class_body":
+            return child
+    return None

codemap_kotlin-0.1.0a1/tests/test_indexer.py

@@ -0,0 +1,154 @@
+"""Unit tests for the Kotlin indexer plugin."""
+
+from __future__ import annotations
+
+import textwrap
+from pathlib import Path, PurePosixPath
+
+from codemap_kotlin import KotlinIndexer
+from codemap_kotlin.indexer import SCHEME
+
+from codemap.core.models import IndexResult
+from codemap.indexers.base import IndexContext
+
+
+def _index(source: str, *, path: str = "src/Main.kt") -> IndexResult:
+    code = textwrap.dedent(source).lstrip("\n")
+    return KotlinIndexer().index_file(
+        Path(path),
+        code.encode("utf-8"),
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath(path),
+            language="kotlin",
+        ),
+    )
+
+
+def test_indexer_metadata() -> None:
+    ix = KotlinIndexer()
+    assert ix.name == "kotlin"
+    assert ix.languages == ["kotlin"]
+    assert ix.supports(Path("a.kt"))
+    assert ix.supports(Path("a.kts"))
+    assert not ix.supports(Path("a.java"))
+
+
+def test_scheme_is_consistent() -> None:
+    r = _index("class A\nfun f() {}\nval X = 1\n")
+    for s in r.symbols:
+        assert str(s.id).startswith(f"{SCHEME} ")
+
+
+def test_class_declaration() -> None:
+    r = _index(
+        """
+        class User(val name: String) {
+            fun hello(): String = name
+        }
+        """
+    )
+    classes = [s for s in r.symbols if s.kind == "class"]
+    assert len(classes) == 1
+    assert classes[0].extra.get("kotlin_kind") == "class"
+    methods = [s for s in r.symbols if s.kind == "method"]
+    assert any("User#hello()." in str(m.id) for m in methods)
+
+
+def test_interface_kept_as_class_with_extra() -> None:
+    r = _index(
+        """
+        interface Greeter {
+            fun hello(): String
+        }
+        """
+    )
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert cls.extra.get("kotlin_kind") == "interface"
+
+
+def test_object_kept_as_class_with_extra() -> None:
+    r = _index(
+        """
+        object Singleton {
+            fun greet() = "hi"
+        }
+        """
+    )
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert cls.extra.get("kotlin_kind") == "object"
+
+
+def test_package_captured_in_extra() -> None:
+    r = _index(
+        """
+        package com.example
+        class A
+        """
+    )
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert cls.extra.get("package") == "com.example"
+
+
+def test_free_function() -> None:
+    r = _index("fun helper(x: Int): Int = x + 1")
+    funcs = [s for s in r.symbols if s.kind == "function"]
+    assert len(funcs) == 1
+    assert "helper" in str(funcs[0].id)
+
+
+def test_top_level_val_is_variable() -> None:
+    r = _index("val MAX = 10")
+    vars_ = [s for s in r.symbols if s.kind == "variable"]
+    assert len(vars_) == 1
+
+
+def test_top_level_var_is_variable() -> None:
+    r = _index("var counter = 0")
+    vars_ = [s for s in r.symbols if s.kind == "variable"]
+    assert len(vars_) == 1
+
+
+def test_method_inside_class_uses_method_kind() -> None:
+    r = _index(
+        """
+        class User {
+            fun hello() = ""
+        }
+        """
+    )
+    methods = [s for s in r.symbols if s.kind == "method"]
+    funcs = [s for s in r.symbols if s.kind == "function"]
+    assert len(methods) == 1
+    assert len(funcs) == 0
+
+
+def test_symbol_id_uses_path_namespaces() -> None:
+    r = _index("class A", path="src/main/kotlin/A.kt")
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert str(cls.id) == "scip-kotlin . . . src/main/kotlin/A.kt/A#"
+
+
+def test_empty_file_yields_no_symbols() -> None:
+    r = _index("")
+    assert r.symbols == []
+
+
+def test_invalid_utf8_yields_diagnostic() -> None:
+    ix = KotlinIndexer()
+    r = ix.index_file(
+        Path("bad.kt"),
+        b"\xff\xfe class",
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath("bad.kt"),
+            language="kotlin",
+        ),
+    )
+    assert r.symbols == []
+    assert r.diagnostics[0].code == "KT002"
+
+
+def test_kts_extension_supported() -> None:
+    r = _index("fun main() {}", path="build.kts")
+    assert any(s.kind == "function" for s in r.symbols)