codemap-go 0.1.0a1__tar.gz

codemap_go-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_go-0.1.0a1/PKG-INFO

@@ -0,0 +1,74 @@
+Metadata-Version: 2.4
+Name: codemap-go
+Version: 0.1.0a1
+Summary: Go indexer plugin for CodeMap
+Project-URL: Homepage, https://github.com/qxbyte/codemap
+Author: CodeMap Contributors
+License: MIT
+Keywords: codemap,go,indexer,tree-sitter
+Classifier: Development Status :: 3 - Alpha
+Classifier: Programming Language :: Go
+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-go>=0.23
+Requires-Dist: tree-sitter>=0.25
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# codemap-go
+
+> A Go indexer for [CodeMap](https://github.com/qxbyte/codemap), shipped
+> as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-go`:
+
+| AST node | Symbol kind |
+|---|---|
+| `function_declaration` | `function` |
+| `method_declaration` | `method` (attached to its receiver type) |
+| `type_declaration` containing `struct_type` | `class` (with `extra.go_kind=struct`) |
+| `type_declaration` containing `interface_type` | `class` (with `extra.go_kind=interface`) |
+| `type_declaration` (other) | `class` (with `extra.go_kind=type`) |
+| `const_declaration` (top-level) | `variable` (with `extra.go_kind=const`) |
+| `var_declaration` (top-level) | `variable` |
+
+`package_clause` is captured as `extra.package` on every symbol-producing
+type so a downstream bridge can use it for cross-file resolution.
+
+Method receivers — both `func (u User) M()` and `func (u *User) M()` —
+attach the method to its `User` type, producing
+`scip-go . . . main.go/User#Login().`.
+
+## SymbolID encoding
+
+```
+scip-go . . . pkg/user/user.go/User#Login().
+└──────┘ └─────────────────────┘ └────┘ └─────┘
+ scheme        file path          type   method
+```
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-go"
+```
+
+After install, `codemap doctor` lists `go` alongside the other indexers
+on identical terms (ADR-004 + ADR-L001).
+
+## Limits
+
+* No `imports` edges. Top-level imports are recorded for the package
+  table but not turned into edges.
+* No generic constraints / type-parameter descriptors.
+* Method bodies are not traversed for call edges yet.
+* Anonymous structs and function-typed fields are skipped.
+
+## License
+
+MIT.

codemap_go-0.1.0a1/README.md

@@ -0,0 +1,54 @@
+# codemap-go
+
+> A Go indexer for [CodeMap](https://github.com/qxbyte/codemap), shipped
+> as an independent PyPI package.
+
+## What it captures
+
+Backed by `tree-sitter-go`:
+
+| AST node | Symbol kind |
+|---|---|
+| `function_declaration` | `function` |
+| `method_declaration` | `method` (attached to its receiver type) |
+| `type_declaration` containing `struct_type` | `class` (with `extra.go_kind=struct`) |
+| `type_declaration` containing `interface_type` | `class` (with `extra.go_kind=interface`) |
+| `type_declaration` (other) | `class` (with `extra.go_kind=type`) |
+| `const_declaration` (top-level) | `variable` (with `extra.go_kind=const`) |
+| `var_declaration` (top-level) | `variable` |
+
+`package_clause` is captured as `extra.package` on every symbol-producing
+type so a downstream bridge can use it for cross-file resolution.
+
+Method receivers — both `func (u User) M()` and `func (u *User) M()` —
+attach the method to its `User` type, producing
+`scip-go . . . main.go/User#Login().`.
+
+## SymbolID encoding
+
+```
+scip-go . . . pkg/user/user.go/User#Login().
+└──────┘ └─────────────────────┘ └────┘ └─────┘
+ scheme        file path          type   method
+```
+
+## Install
+
+```bash
+pip install "git+https://github.com/qxbyte/codemap.git#subdirectory=plugins/codemap-go"
+```
+
+After install, `codemap doctor` lists `go` alongside the other indexers
+on identical terms (ADR-004 + ADR-L001).
+
+## Limits
+
+* No `imports` edges. Top-level imports are recorded for the package
+  table but not turned into edges.
+* No generic constraints / type-parameter descriptors.
+* Method bodies are not traversed for call edges yet.
+* Anonymous structs and function-typed fields are skipped.
+
+## License
+
+MIT.

codemap_go-0.1.0a1/pyproject.toml

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

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

@@ -0,0 +1,8 @@
+"""Go indexer plugin for CodeMap."""
+
+from __future__ import annotations
+
+from codemap_go.indexer import GoIndexer
+
+__all__ = ["GoIndexer"]
+__version__ = "0.1.0"

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

@@ -0,0 +1,315 @@
+"""Go indexer built on tree-sitter-go.
+
+Notable Go-specific behaviour: ``method_declaration`` carries a receiver
+parameter list before the method name. The indexer pulls the receiver's
+type out and attaches the method as ``<TypeName>#methodName()`` so a
+later query like ``callers 'scip-go . . . main.go/User#Login().'`` can
+find every cross-receiver caller via the cross-module bridge.
+"""
+
+from __future__ import annotations
+
+from pathlib import Path, PurePosixPath
+from typing import ClassVar
+
+import tree_sitter
+import tree_sitter_go
+
+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-go"
+LANG = "go"
+
+_GO_LANG = tree_sitter.Language(tree_sitter_go.language())
+
+
+class GoIndexer:
+    name: ClassVar[str] = "go"
+    version: ClassVar[str] = "0.1.0"
+    file_patterns: ClassVar[list[str]] = ["*.go"]
+    languages: ClassVar[list[str]] = [LANG]
+
+    def supports(self, path: Path) -> bool:
+        return path.suffix == ".go"
+
+    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="GO002",
+                        message=f"not valid UTF-8: {exc}",
+                        producer=self.name,
+                    )
+                ]
+            )
+        parser = tree_sitter.Parser(_GO_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="GO001",
+                    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._package: str = ""
+
+    def visit(self, node: tree_sitter.Node) -> None:
+        kind = node.type
+        if kind == "package_clause":
+            self._set_package(node)
+            return
+        if kind == "function_declaration":
+            self._visit_function(node)
+            return
+        if kind == "method_declaration":
+            self._visit_method(node)
+            return
+        if kind == "type_declaration":
+            for spec in node.children:
+                if spec.type == "type_spec":
+                    self._visit_type_spec(spec)
+            return
+        if kind == "const_declaration":
+            self._visit_const_or_var(node, go_kind="const")
+            return
+        if kind == "var_declaration":
+            self._visit_const_or_var(node, go_kind="var")
+            return
+        for child in node.children:
+            self.visit(child)
+
+    # ------------------------------------------------------------ pkg
+
+    def _set_package(self, node: tree_sitter.Node) -> None:
+        for child in node.children:
+            if child.type == "package_identifier":
+                self._package = _node_text(child)
+                return
+
+    # ------------------------------------------------------- functions
+
+    def _visit_function(self, node: tree_sitter.Node) -> None:
+        name_node = node.child_by_field_name("name")
+        if name_node is None:
+            return
+        name = _node_text(name_node)
+        if not name:
+            return
+        sid = self._make_id([], name, kind=DescriptorKind.METHOD)
+        self.symbols.append(
+            Symbol(
+                id=sid,
+                kind="function",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                signature=_function_signature(node, name),
+                extra={"package": self._package} if self._package else {},
+            )
+        )
+
+    def _visit_method(self, node: tree_sitter.Node) -> None:
+        name_node = node.child_by_field_name("name")
+        if name_node is None:
+            return
+        name = _node_text(name_node)
+        if not name:
+            return
+        receiver_type = _extract_receiver_type(node)
+        if receiver_type is None:
+            # Methods without resolvable receivers are dropped — they would
+            # otherwise alias against unrelated top-level functions.
+            return
+        sid = self._make_id([receiver_type], name, kind=DescriptorKind.METHOD)
+        self.symbols.append(
+            Symbol(
+                id=sid,
+                kind="method",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                signature=_function_signature(node, name),
+                extra={
+                    "package": self._package,
+                    "receiver_type": receiver_type,
+                },
+            )
+        )
+
+    # ----------------------------------------------------------- types
+
+    def _visit_type_spec(self, node: tree_sitter.Node) -> None:
+        name_node = node.child_by_field_name("name")
+        if name_node is None:
+            return
+        name = _node_text(name_node)
+        if not name:
+            return
+        body = node.child_by_field_name("type")
+        if body is not None and body.type == "struct_type":
+            go_kind = "struct"
+        elif body is not None and body.type == "interface_type":
+            go_kind = "interface"
+        else:
+            go_kind = "type"
+        sid = self._make_id([], name, kind=DescriptorKind.TYPE)
+        self.symbols.append(
+            Symbol(
+                id=sid,
+                kind="class",
+                language=LANG,
+                file=self.relative_path,
+                range=_node_range(node),
+                extra={"package": self._package, "go_kind": go_kind},
+            )
+        )
+
+    # ---------------------------------------------------- vars / consts
+
+    def _visit_const_or_var(self, node: tree_sitter.Node, *, go_kind: str) -> None:
+        for spec in node.children:
+            if spec.type not in {"const_spec", "var_spec"}:
+                continue
+            for child in spec.children:
+                if child.type != "identifier":
+                    continue
+                name = _node_text(child)
+                if not name:
+                    continue
+                sid = self._make_id([], name, kind=DescriptorKind.TERM)
+                self.symbols.append(
+                    Symbol(
+                        id=sid,
+                        kind="variable",
+                        language=LANG,
+                        file=self.relative_path,
+                        range=_node_range(spec),
+                        extra={"package": self._package, "go_kind": go_kind}
+                        if self._package
+                        else {"go_kind": go_kind},
+                    )
+                )
+
+    # --------------------------------------------------------- helpers
+
+    def _make_id(
+        self,
+        type_chain: list[str],
+        name: str,
+        *,
+        kind: DescriptorKind,
+    ) -> SymbolID:
+        descriptors = list(_path_namespaces(self.relative_path))
+        descriptors.extend(Descriptor(name=t, kind=DescriptorKind.TYPE) for t in type_chain)
+        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 _function_signature(node: tree_sitter.Node, name: str) -> str:
+    params = node.child_by_field_name("parameters")
+    params_text = _node_text(params) if params is not None else "()"
+    result = node.child_by_field_name("result")
+    result_text = (" " + _node_text(result)) if result is not None else ""
+    return f"func {name}{params_text}{result_text}"
+
+
+def _extract_receiver_type(method_node: tree_sitter.Node) -> str | None:
+    """For ``func (u User) M()`` and ``func (u *User) M()`` return ``User``."""
+    receiver = method_node.child_by_field_name("receiver")
+    if receiver is None:
+        # Some versions of the grammar don't expose `receiver` by name —
+        # fall back to the first parameter_list before the method name.
+        for child in method_node.children:
+            if child.type == "parameter_list":
+                receiver = child
+                break
+    if receiver is None:
+        return None
+    for param in receiver.children:
+        if param.type != "parameter_declaration":
+            continue
+        type_node = param.child_by_field_name("type")
+        if type_node is None:
+            # Walk children to find the type
+            for c in param.children:
+                if c.type in {"type_identifier", "pointer_type", "qualified_type"}:
+                    type_node = c
+                    break
+        if type_node is None:
+            continue
+        return _unwrap_type(type_node)
+    return None
+
+
+def _unwrap_type(node: tree_sitter.Node) -> str | None:
+    if node.type == "type_identifier":
+        return _node_text(node)
+    if node.type == "pointer_type":
+        for c in node.children:
+            if c.type == "type_identifier":
+                return _node_text(c)
+            if c.type == "qualified_type":
+                return _unwrap_type(c)
+        return None
+    if node.type == "qualified_type":
+        for c in node.children:
+            if c.type == "type_identifier":
+                return _node_text(c)
+        return None
+    return None

codemap_go-0.1.0a1/tests/test_indexer.py

@@ -0,0 +1,181 @@
+"""Unit tests for the Go indexer plugin."""
+
+from __future__ import annotations
+
+import textwrap
+from pathlib import Path, PurePosixPath
+
+from codemap_go import GoIndexer
+from codemap_go.indexer import SCHEME
+
+from codemap.core.models import IndexResult
+from codemap.indexers.base import IndexContext
+
+
+def _index(source: str, *, path: str = "main.go") -> IndexResult:
+    code = textwrap.dedent(source).lstrip("\n")
+    return GoIndexer().index_file(
+        Path(path),
+        code.encode("utf-8"),
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath(path),
+            language="go",
+        ),
+    )
+
+
+# ---------------------------------------------------------------------------
+# Metadata
+# ---------------------------------------------------------------------------
+
+
+def test_indexer_metadata() -> None:
+    ix = GoIndexer()
+    assert ix.name == "go"
+    assert ix.languages == ["go"]
+    assert "*.go" in ix.file_patterns
+    assert ix.supports(Path("a.go"))
+    assert not ix.supports(Path("a.py"))
+
+
+def test_scheme_is_consistent() -> None:
+    r = _index(
+        """
+        package main
+        func f() {}
+        type T struct {}
+        const C = 1
+        """
+    )
+    for s in r.symbols:
+        assert str(s.id).startswith(f"{SCHEME} ")
+
+
+# ---------------------------------------------------------------------------
+# Functions / methods
+# ---------------------------------------------------------------------------
+
+
+def test_function_declaration() -> None:
+    r = _index(
+        """
+        package main
+        func helper(x int) int { return x + 1 }
+        """
+    )
+    funcs = [s for s in r.symbols if s.kind == "function"]
+    assert len(funcs) == 1
+    assert "helper" in str(funcs[0].id)
+    assert funcs[0].extra.get("package") == "main"
+
+
+def test_method_attached_to_receiver_value() -> None:
+    r = _index(
+        """
+        package main
+        type User struct { Name string }
+        func (u User) Hello() string { return u.Name }
+        """
+    )
+    methods = [s for s in r.symbols if s.kind == "method"]
+    assert len(methods) == 1
+    assert "User#Hello()." in str(methods[0].id)
+    assert methods[0].extra.get("receiver_type") == "User"
+
+
+def test_method_attached_to_pointer_receiver() -> None:
+    r = _index(
+        """
+        package main
+        type User struct {}
+        func (u *User) Login(pw string) bool { return false }
+        """
+    )
+    methods = [s for s in r.symbols if s.kind == "method"]
+    assert "User#Login()." in str(methods[0].id)
+
+
+# ---------------------------------------------------------------------------
+# Types
+# ---------------------------------------------------------------------------
+
+
+def test_struct_type() -> None:
+    r = _index("package main\ntype User struct { ID int }")
+    types = [s for s in r.symbols if s.kind == "class"]
+    assert len(types) == 1
+    assert types[0].extra.get("go_kind") == "struct"
+
+
+def test_interface_type() -> None:
+    r = _index(
+        """
+        package main
+        type Greeter interface { Hello() string }
+        """
+    )
+    types = [s for s in r.symbols if s.kind == "class"]
+    assert types[0].extra.get("go_kind") == "interface"
+
+
+# ---------------------------------------------------------------------------
+# Constants / variables
+# ---------------------------------------------------------------------------
+
+
+def test_const_declaration() -> None:
+    r = _index("package main\nconst MAX = 10")
+    vars_ = [s for s in r.symbols if s.kind == "variable"]
+    assert len(vars_) == 1
+    assert vars_[0].extra.get("go_kind") == "const"
+
+
+def test_var_declaration() -> None:
+    r = _index("package main\nvar counter int")
+    vars_ = [s for s in r.symbols if s.kind == "variable"]
+    assert vars_[0].extra.get("go_kind") == "var"
+
+
+# ---------------------------------------------------------------------------
+# SymbolID + edge cases
+# ---------------------------------------------------------------------------
+
+
+def test_symbol_id_uses_path_namespaces() -> None:
+    r = _index(
+        "package user\nfunc helper() {}",
+        path="pkg/user/u.go",
+    )
+    assert str(r.symbols[0].id) == "scip-go . . . pkg/user/u.go/helper()."
+
+
+def test_empty_file_yields_no_symbols() -> None:
+    r = _index("")
+    assert r.symbols == []
+
+
+def test_invalid_utf8_yields_diagnostic() -> None:
+    ix = GoIndexer()
+    r = ix.index_file(
+        Path("bad.go"),
+        b"\xff\xfe package",
+        IndexContext(
+            project_root=Path("/tmp/proj"),
+            relative_path=PurePosixPath("bad.go"),
+            language="go",
+        ),
+    )
+    assert r.symbols == []
+    assert r.diagnostics[0].code == "GO002"
+
+
+def test_package_captured_on_symbols() -> None:
+    r = _index(
+        """
+        package user
+        type User struct {}
+        """
+    )
+    cls = next(s for s in r.symbols if s.kind == "class")
+    assert cls.extra.get("package") == "user"