Merge PR #732: feat: add Groovy and Spock support

graphify/detect.py

@@ -18,7 +18,7 @@ class FileType(str, Enum):
 
 _MANIFEST_PATH = "graphify-out/manifest.json"
 
-CODE_EXTENSIONS = {'.py', '.ts', '.js', '.jsx', '.tsx', '.mjs', '.ejs', '.go', '.rs', '.java', '.cpp', '.cc', '.cxx', '.c', '.h', '.hpp', '.rb', '.swift', '.kt', '.kts', '.cs', '.scala', '.php', '.lua', '.luau', '.toc', '.zig', '.ps1', '.ex', '.exs', '.m', '.mm', '.jl', '.vue', '.svelte', '.dart', '.v', '.sv', '.sql', '.r', '.f', '.F', '.f90', '.F90', '.f95', '.F95', '.f03', '.F03', '.f08', '.F08'}
+CODE_EXTENSIONS = {'.py', '.ts', '.js', '.jsx', '.tsx', '.mjs', '.ejs', '.go', '.rs', '.java', '.groovy', '.gradle', '.cpp', '.cc', '.cxx', '.c', '.h', '.hpp', '.rb', '.swift', '.kt', '.kts', '.cs', '.scala', '.php', '.lua', '.luau', '.toc', '.zig', '.ps1', '.ex', '.exs', '.m', '.mm', '.jl', '.vue', '.svelte', '.dart', '.v', '.sv', '.sql', '.r', '.f', '.F', '.f90', '.F90', '.f95', '.F95', '.f03', '.F03', '.f08', '.F08'}
 DOC_EXTENSIONS = {'.md', '.mdx', '.txt', '.rst', '.html', '.yaml', '.yml'}
 PAPER_EXTENSIONS = {'.pdf'}
 IMAGE_EXTENSIONS = {'.png', '.jpg', '.jpeg', '.gif', '.webp', '.svg'}

graphify/extract.py

@@ -726,6 +726,18 @@ _JAVA_CONFIG = LanguageConfig(
     import_handler=_import_java,
 )
 
+_GROOVY_CONFIG = LanguageConfig(
+    ts_module="tree_sitter_groovy",
+    class_types=frozenset({"class_declaration", "interface_declaration"}),
+    function_types=frozenset({"method_declaration", "constructor_declaration"}),
+    import_types=frozenset({"import_declaration"}),
+    call_types=frozenset({"method_invocation"}),
+    call_function_field="name",
+    call_accessor_node_types=frozenset(),
+    function_boundary_types=frozenset({"method_declaration", "constructor_declaration"}),
+    import_handler=_import_java,
+)
+
 _C_CONFIG = LanguageConfig(
     ts_module="tree_sitter_c",
     class_types=frozenset(),
@@ -1877,6 +1889,122 @@ def extract_java(path: Path) -> dict:
     return _extract_generic(path, _JAVA_CONFIG)
 
 
+def _is_spock_file(path: Path, ts_result: dict) -> bool:
+    """Return True when the file contains Spock-style ``def "feature"()`` methods
+    that tree-sitter-groovy cannot parse, detected by checking the raw source."""
+    import re as _re
+    _SPOCK_FEATURE_RE = _re.compile(r"""^\s*def\s+[\"']""", _re.MULTILINE)
+    try:
+        return bool(_SPOCK_FEATURE_RE.search(path.read_text(errors="replace")))
+    except OSError:
+        return False
+
+
+def _extract_spock_fallback(path: Path, ts_result: dict) -> dict:
+    """Regex-based fallback for Spock spec files where tree-sitter-groovy cannot parse
+    ``def "feature name"()`` methods. Merges import edges from the tree-sitter pass
+    (which survive reliably) with class and feature-method nodes extracted via regex.
+    """
+    import re as _re
+    source = path.read_text(errors="replace")
+    str_path = str(path)
+    stem = _file_stem(path)
+
+    # Only keep the file node from the tree-sitter pass (guaranteed present and
+    # correctly IDed) plus all import edges.  All other ts nodes are discarded to
+    # avoid orphaned method/constructor nodes whose parent edges were dropped.
+    file_node = next((n for n in ts_result.get("nodes", []) if n.get("label") == path.name), None)
+    nodes: list[dict] = [file_node] if file_node else []
+    edges: list[dict] = [e for e in ts_result.get("edges", []) if e.get("context") == "import"]
+    seen_ids: set[str] = {n["id"] for n in nodes}
+
+    def _add_node(nid: str, label: str, line: int) -> None:
+        if nid not in seen_ids:
+            seen_ids.add(nid)
+            nodes.append({
+                "id": nid,
+                "label": label,
+                "file_type": "code",
+                "source_file": str_path,
+                "source_location": f"L{line}",
+            })
+
+    def _add_edge(src: str, tgt: str, relation: str, line: int,
+                  confidence: str = "EXTRACTED") -> None:
+        edges.append({
+            "source": src,
+            "target": tgt,
+            "relation": relation,
+            "confidence": confidence,
+            "source_file": str_path,
+            "source_location": f"L{line}",
+            "weight": 1.0,
+        })
+
+    lines_text = source.splitlines()
+
+    # Extract class declarations
+    class_re = _re.compile(r"^\s*(?:[\w@]+\s+)*class\s+(\w+)")
+    # Extract Spock feature methods: def "..." () or def '...' ()
+    # Two separate capture groups per quote style so apostrophes inside
+    # double-quoted names (e.g. "shouldn't") are captured correctly.
+    feature_re = _re.compile(r"""^\s*def\s+(?:\"([^\"]+)\"|'([^']+)')\s*\(""")
+    # Extract plain def methods (non-string names) as well
+    plain_method_re = _re.compile(r"""^\s*def\s+(\w+)\s*\(""")
+
+    current_class_nid: str | None = None
+    file_nid = _make_id(str_path)
+
+    # Ensure the file node exists (tree-sitter pass may have emitted it)
+    if file_nid not in seen_ids:
+        _add_node(file_nid, path.name, 1)
+
+    for lineno, line_text in enumerate(lines_text, start=1):
+        cm = class_re.match(line_text)
+        if cm:
+            class_name = cm.group(1)
+            class_nid = _make_id(stem, class_name)
+            _add_node(class_nid, class_name, lineno)
+            _add_edge(file_nid, class_nid, "contains", lineno)
+            current_class_nid = class_nid
+            continue
+
+        if current_class_nid is None:
+            continue
+
+        fm = feature_re.match(line_text)
+        if fm:
+            method_name = fm.group(1) or fm.group(2)
+            method_label = f'"{method_name}"'
+            method_nid = _make_id(current_class_nid, method_name)
+            _add_node(method_nid, method_label, lineno)
+            _add_edge(current_class_nid, method_nid, "method", lineno)
+            continue
+
+        pm = plain_method_re.match(line_text)
+        if pm:
+            method_name = pm.group(1)
+            if method_name not in ("if", "while", "for", "switch", "catch"):
+                method_label = f".{method_name}()"
+                method_nid = _make_id(current_class_nid, method_name)
+                _add_node(method_nid, method_label, lineno)
+                _add_edge(current_class_nid, method_nid, "method", lineno)
+
+    return {"nodes": nodes, "edges": edges}
+
+
+def extract_groovy(path: Path) -> dict:
+    """Extract classes, methods, constructors, and imports from a .groovy/.gradle file.
+
+    Falls back to a regex-based Spock extractor when tree-sitter-groovy cannot parse
+    ``def "feature name"()`` methods (common in Spock specification classes).
+    """
+    result = _extract_generic(path, _GROOVY_CONFIG)
+    if _is_spock_file(path, result):
+        result = _extract_spock_fallback(path, result)
+    return result
+
+
 def extract_c(path: Path) -> dict:
     """Extract functions and includes from a .c/.h file."""
     return _extract_generic(path, _C_CONFIG)
@@ -4183,6 +4311,8 @@ _DISPATCH: dict[str, Any] = {
     ".go": extract_go,
     ".rs": extract_rust,
     ".java": extract_java,
+    ".groovy": extract_groovy,
+    ".gradle": extract_groovy,
     ".c": extract_c,
     ".h": extract_c,
     ".cpp": extract_cpp,

pyproject.toml

@@ -21,6 +21,7 @@ dependencies = [
     "tree-sitter-go",
     "tree-sitter-rust",
     "tree-sitter-java",
+    "tree-sitter-groovy",
     "tree-sitter-c",
     "tree-sitter-cpp",
     "tree-sitter-ruby",

tests/fixtures/sample.groovy

@@ -0,0 +1,21 @@
+package com.nicklastrange.example
+
+import com.nicklastrange.Processor
+import com.nicklastrange.util.Helper
+
+class SampleService {
+    Processor processor
+
+    SampleService(Processor processor) {
+        this.processor = processor
+    }
+
+    String process(String input) {
+        def result = processor.transform(input)
+        return Helper.clean(result)
+    }
+
+    private void reset() {
+        processor.reset()
+    }
+}

tests/fixtures/sample_spock.groovy

@@ -0,0 +1,42 @@
+package com.nicklastrange.example
+
+import spock.lang.Specification
+
+class SampleSpec extends Specification {
+
+    def setup() {
+        // common setup
+    }
+
+    def "should process valid input"() {
+        given:
+        def input = "hello"
+
+        when:
+        def result = input.toUpperCase()
+
+        then:
+        result == "HELLO"
+    }
+
+    def "should not change value when it's already correct"() {
+        given:
+        def value = "HELLO"
+
+        when:
+        def result = value.toUpperCase()
+
+        then:
+        result == value
+    }
+
+    def "should handle #input and return #expected"() {
+        expect:
+        input.toUpperCase() == expected
+
+        where:
+        input   | expected
+        "hello" | "HELLO"
+        "world" | "WORLD"
+    }
+}

tests/test_languages.py

@@ -6,6 +6,7 @@ from graphify.extract import (
     extract_java, extract_c, extract_cpp, extract_ruby,
     extract_csharp, extract_kotlin, extract_scala, extract_php,
     extract_swift, extract_go, extract_julia, extract_js, extract_fortran,
+    extract_groovy,
 )
 
 FIXTURES = Path(__file__).parent / "fixtures"
@@ -896,3 +897,70 @@ def test_markdown_no_dangling_edges():
     node_ids = {n["id"] for n in r["nodes"]}
     for e in r["edges"]:
         assert e["source"] in node_ids, f"Dangling source: {e}"
+
+
+# ── Groovy ───────────────────────────────────────────────────────────────────
+
+
+def test_groovy_no_error():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    assert "error" not in r
+
+
+def test_groovy_finds_class():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    assert any("SampleService" in l for l in _labels(r))
+
+
+def test_groovy_finds_methods():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    labels = _labels(r)
+    assert any("process" in l for l in labels)
+    assert any("reset" in l for l in labels)
+
+
+def test_groovy_finds_imports():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    assert "imports" in _relations(r)
+
+
+def test_groovy_import_edges_have_import_context():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    import_edges = _edges_with_relation(r, "imports", "imports_from")
+    assert import_edges
+    assert all(e.get("context") == "import" for e in import_edges)
+
+
+def test_groovy_no_dangling_edges():
+    r = extract_groovy(FIXTURES / "sample.groovy")
+    node_ids = {n["id"] for n in r["nodes"]}
+    for e in r["edges"]:
+        assert e["source"] in node_ids
+
+
+def test_groovy_spock_finds_class():
+    r = extract_groovy(FIXTURES / "sample_spock.groovy")
+    assert any("SampleSpec" in l for l in _labels(r))
+
+
+def test_groovy_spock_finds_feature_methods():
+    r = extract_groovy(FIXTURES / "sample_spock.groovy")
+    feature_labels = [l for l in _labels(r) if l.startswith('"')]
+    assert len(feature_labels) >= 2
+
+
+def test_groovy_spock_finds_method_with_apostrophe():
+    r = extract_groovy(FIXTURES / "sample_spock.groovy")
+    assert any("it's" in l for l in _labels(r))
+
+
+def test_groovy_spock_preserves_import_edges():
+    r = extract_groovy(FIXTURES / "sample_spock.groovy")
+    assert "imports" in _relations(r)
+
+
+def test_groovy_spock_no_dangling_edges():
+    r = extract_groovy(FIXTURES / "sample_spock.groovy")
+    node_ids = {n["id"] for n in r["nodes"]}
+    for e in r["edges"]:
+        assert e["source"] in node_ids