sourcecode 1.1.0__tar.gz → 1.2.0__tar.gz

{sourcecode-1.1.0 → sourcecode-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sourcecode
-Version: 1.1.0
+Version: 1.2.0
 Summary: Deterministic codebase context for AI coding agents
 License:                                  Apache License
                                    Version 2.0, January 2004

{sourcecode-1.1.0 → sourcecode-1.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "sourcecode"
-version = "1.1.0"
+version = "1.2.0"
 description = "Deterministic codebase context for AI coding agents"
 readme = "README.md"
 requires-python = ">=3.9"

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Deterministic codebase context maps for AI coding agents."""
 
-__version__ = "1.1.0"
+__version__ = "1.2.0"

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/architecture_analyzer.py

@@ -183,7 +183,11 @@ class ArchitectureAnalyzer:
         if ddd_result is not None:
             ddd_pattern, ddd_layers, ddd_contexts, ddd_layer_names = ddd_result
             domains_for_ddd = self._cluster_domains(filtered) if len(filtered) >= 2 else []
-            bc_list = [BoundedContext(name=n, confidence="high") for n in ddd_contexts]
+            module_files = self._build_ddd_module_files(sm.file_paths, ddd_contexts)
+            bc_list = [
+                BoundedContext(name=n, modules=module_files.get(n, []), confidence="high")
+                for n in ddd_contexts
+            ]
             return ArchitectureAnalysis(
                 requested=True,
                 pattern=ddd_pattern,
@@ -414,6 +418,22 @@ class ArchitectureAnalyzer:
         ]
         return "ddd", arch_layers, bounded_context_names, ddd_layer_names
 
+    def _build_ddd_module_files(
+        self, paths: list[str], bounded_context_names: list[str]
+    ) -> "dict[str, list[str]]":
+        """Build a mapping of DDD module name → list of file paths."""
+        _DDD_LAYERS = frozenset({"application", "domain", "infrastructure"})
+        module_files: dict[str, list[str]] = {}
+        for p in paths:
+            parts = p.replace("\\", "/").split("/")
+            for i, part in enumerate(parts):
+                if part in _DDD_LAYERS and i >= 2:
+                    mod = parts[i - 1]
+                    if mod in bounded_context_names:
+                        module_files.setdefault(mod, []).append(p)
+                    break
+        return module_files
+
     def _is_tooling(self, path: str) -> bool:
         norm = path.replace("\\", "/")
         return any(norm.startswith(p) for p in _TOOLING_PREFIXES)

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/cli.py

@@ -790,7 +790,7 @@ def main(
     # Require at least 8: src(1)+main(2)+java(3)+com(4)+co(5)+app(6)+module(7)+file.
     _java_manifest_names = {"pom.xml", "build.gradle", "build.gradle.kts"}
     _is_java = any(Path(m).name in _java_manifest_names for m in manifests)
-    _java_min_depth = 8
+    _java_min_depth = 10
     effective_depth = max(depth, _java_min_depth) if _is_java and depth < _java_min_depth else depth
 
     # --agent: enable signal analyzers; output via agent_view (not compact)
@@ -1376,6 +1376,12 @@ def main(
                         ))
         sm = _replace(sm, pipeline_trace=_trace.build_trace())
 
+    # P3-B: Auto-switch to centrality ranking when DDD layout detected
+    if (rank_by == "relevance"
+            and sm.architecture is not None
+            and sm.architecture.pattern == "ddd"):
+        rank_by = "centrality"
+
     # Contract pipeline — runs for mode=contract|standard|deep|hybrid (skip for raw)
     _is_contract_mode = mode in ("contract", "standard")
     if _is_contract_mode:

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/confidence_analyzer.py

@@ -193,6 +193,27 @@ class ConfidenceAnalyzer:
                 impact="low",
             ))
 
+        # ── Java test coverage gap check (P2-A) ──────────────────────────────
+        _java_all = [p for p in sm.file_paths if p.endswith(".java")]
+        _java_tests = [
+            p for p in _java_all
+            if "/test/" in p.replace("\\", "/") or "/tests/" in p.replace("\\", "/")
+            or Path(p).stem.endswith(("Test", "Tests", "IT", "Spec"))
+        ]
+        _java_prod = [p for p in _java_all if p not in set(_java_tests)]
+        if _java_prod and len(_java_prod) >= 10:
+            _ratio = len(_java_tests) / len(_java_prod)
+            if _ratio < 0.05:
+                gaps.append(AnalysisGap(
+                    area="testing",
+                    reason=(
+                        f"Backend test coverage critical: {len(_java_tests)} test files "
+                        f"for {len(_java_prod)} Java files "
+                        f"({_ratio:.1%})"
+                    ),
+                    impact="high",
+                ))
+
         # ── Compute overall confidence ─────────────────────────────────────────
         # Stack: use best manifest-detected stack, fall back to min
         manifest_stacks = [s for s in sm.stacks if s.detection_method != "heuristic"]

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/detectors/java.py

@@ -25,8 +25,18 @@ _CONTROLLER_ADVICE_RE = re.compile(r'@ControllerAdvice\b')
 _WEB_FILTER_RE = re.compile(r'@WebFilter\b')
 _FILTER_BEAN_RE = re.compile(r'FilterRegistrationBean\b')
 # Extracts path from @RequestMapping("/v1/foo"), @GetMapping("/bar"), etc.
+# Handles attribute order: value= may come after method= in legacy @RequestMapping style.
 _HTTP_PATH_RE = re.compile(
-    r'@(?:Request|Get|Post|Put|Delete|Patch)Mapping\s*\(\s*(?:value\s*=\s*)?["\']([^"\']+)["\']'
+    r'@(?:Request|Get|Post|Put|Delete|Patch)Mapping\s*\([^)]*?(?:value\s*=\s*)?["\']([^"\']+)["\']'
+)
+_REQUEST_METHOD_VERB_RE = re.compile(
+    r'method\s*=\s*RequestMethod\.([A-Z]+)'
+)
+# @M3FiltroSeguridad custom security annotation
+_M3_FILTRO_RE = re.compile(r'@M3FiltroSeguridad\b')
+_M3_FILTRO_PARAMS_RE = re.compile(
+    r'@M3FiltroSeguridad\s*\(\s*(?:nombreRecurso\s*=\s*"([^"]*)")?'
+    r'(?:[^)]*nivelRequerido\s*=\s*(\d+))?'
 )
 
 
@@ -149,16 +159,29 @@ class JavaDetector(AbstractDetector):
 
         # Quick pre-filter before running regexes
         if ("Controller" not in content and "Filter" not in content
-                and "ControllerAdvice" not in content):
+                and "ControllerAdvice" not in content
+                and "M3FiltroSeguridad" not in content):
             return []
 
         if _REST_CONTROLLER_RE.search(content):
             http_path_match = _HTTP_PATH_RE.search(content)
             http_path = http_path_match.group(1) if http_path_match else None
+            verb_match = _REQUEST_METHOD_VERB_RE.search(content)
+            if verb_match and http_path:
+                http_path = f"[{verb_match.group(1)}] {http_path}"
+            elif verb_match:
+                http_path = f"[{verb_match.group(1)}]"
+            security_evidence = None
+            m3_match = _M3_FILTRO_PARAMS_RE.search(content)
+            if m3_match:
+                nombre = m3_match.group(1) or ""
+                nivel = m3_match.group(2) or ""
+                security_evidence = f"@M3FiltroSeguridad(nombreRecurso={nombre!r}, nivelRequerido={nivel})"
             return [EntryPoint(
                 path=rel_path, stack="java", kind="rest_controller",
                 source="annotation", confidence="high",
                 http_path=http_path,
+                evidence=security_evidence,
             )]
         if _CONTROLLER_ADVICE_RE.search(content):
             return [EntryPoint(
@@ -168,10 +191,22 @@ class JavaDetector(AbstractDetector):
         if _MVC_CONTROLLER_RE.search(content) and _REQUEST_MAPPING_RE.search(content):
             http_path_match = _HTTP_PATH_RE.search(content)
             http_path = http_path_match.group(1) if http_path_match else None
+            verb_match = _REQUEST_METHOD_VERB_RE.search(content)
+            if verb_match and http_path:
+                http_path = f"[{verb_match.group(1)}] {http_path}"
+            elif verb_match:
+                http_path = f"[{verb_match.group(1)}]"
+            security_evidence = None
+            m3_match = _M3_FILTRO_PARAMS_RE.search(content)
+            if m3_match:
+                nombre = m3_match.group(1) or ""
+                nivel = m3_match.group(2) or ""
+                security_evidence = f"@M3FiltroSeguridad(nombreRecurso={nombre!r}, nivelRequerido={nivel})"
             return [EntryPoint(
                 path=rel_path, stack="java", kind="mvc_controller",
                 source="annotation", confidence="medium",
                 http_path=http_path,
+                evidence=security_evidence,
             )]
         if _WEB_FILTER_RE.search(content):
             return [EntryPoint(

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/env_analyzer.py

@@ -27,6 +27,8 @@ _ENV_EXAMPLE_NAMES = {
 
 # Spring Boot application.properties / application.yml and their profile variants
 _SPRING_CONF_BASE = {"application.properties", "application.yml", "application.yaml"}
+# Matches options/{profile}/ in multi-tenant SAS layout paths
+_OPTIONS_PROFILE_PATH_RE = re.compile(r'options/([a-z0-9_-]+)/', re.IGNORECASE)
 _SPRING_CONF_PROFILE_RE = re.compile(r'^application-([a-z0-9_-]+)\.(properties|ya?ml)$', re.IGNORECASE)
 # Matches ${ENV_VAR} or ${ENV_VAR:default} where ENV_VAR is UPPER_SNAKE_CASE.
 # Group 1 = key, Group 2 = default (may be empty string, absent = no default).
@@ -507,6 +509,10 @@ class EnvAnalyzer:
                 # Spring Boot application.properties / application.yml (incl. profiles)
                 if name_lower in _SPRING_CONF_BASE or _SPRING_CONF_PROFILE_RE.match(name_lower):
                     profile = _extract_spring_profile(name)
+                    # Override profile if path contains options/{profile}/ (multi-tenant SAS layout)
+                    path_profile_match = _OPTIONS_PROFILE_PATH_RE.search(rel)
+                    if path_profile_match:
+                        profile = path_profile_match.group(1)
                     if profile and profile not in profiles_scanned:
                         profiles_scanned.append(profile)
                     count = _parse_spring_config(entry, rel, findings, profile)

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/metrics_analyzer.py

@@ -229,6 +229,29 @@ class MetricsAnalyzer:
                 "null complexity fields are expected, not an error."
             )
 
+        # P2-C: DDD module metrics — group by module, count files/methods per layer
+        _DDD_LAYERS = {"domain", "application", "infrastructure"}
+        ddd_files = [r for r in records if "/ddd/" in r.path.replace("\\", "/")]
+        if ddd_files:
+            module_layer_data: dict[str, dict[str, dict]] = {}
+            for fm in ddd_files:
+                parts = fm.path.replace("\\", "/").split("/")
+                for i, part in enumerate(parts):
+                    if part in _DDD_LAYERS and i >= 2:
+                        module = parts[i - 1]
+                        layer = part
+                        if module not in module_layer_data:
+                            module_layer_data[module] = {lyr: {"files": 0, "methods": 0} for lyr in _DDD_LAYERS}
+                        module_layer_data[module][layer]["files"] += 1
+                        module_layer_data[module][layer]["methods"] += fm.function_count or 0
+                        break
+            ddd_metrics = [
+                {"module": mod, "layers": layers}
+                for mod, layers in sorted(module_layer_data.items())
+            ]
+        else:
+            ddd_metrics = []
+
         summary = MetricsSummary(
             requested=True,
             file_count=len(records),
@@ -238,6 +261,7 @@ class MetricsAnalyzer:
             coverage_records=coverage_records,
             coverage_sources_found=sorted({r.format for r in coverage_records}),
             limitations=limitations,
+            ddd_module_metrics=ddd_metrics,
         )
         return records, summary
 

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/schema.py

@@ -225,6 +225,7 @@ class MetricsSummary:
     coverage_records: list[CoverageRecord] = field(default_factory=list)
     coverage_sources_found: list[str] = field(default_factory=list)
     limitations: list[str] = field(default_factory=list)
+    ddd_module_metrics: list[dict] = field(default_factory=list)
 
 
 @dataclass

{sourcecode-1.1.0 → sourcecode-1.2.0}/src/sourcecode/semantic_analyzer.py

@@ -40,6 +40,34 @@ _MAX_SYMBOLS = 10_000
 # JS/TS keyword and builtin exclusions (Plan 12-03)
 # ---------------------------------------------------------------------------
 
+# ---------------------------------------------------------------------------
+# Java/JVM heuristic regex constants (module-level for performance)
+# ---------------------------------------------------------------------------
+
+_AUTOWIRED_FIELD_RE = re.compile(
+    r'@Autowired\s+(?:private\s+|protected\s+|public\s+)?([A-Z][A-Za-z0-9_<>]*)\s+(\w+)\s*;'
+)
+_MAPPER_IFACE_RE = re.compile(
+    r'@Mapper\b.*?(?:public\s+)?interface\s+([A-Z][A-Za-z0-9_]*)',
+    re.DOTALL,
+)
+_EXTENDS_RE = re.compile(
+    r'(?:class|interface)\s+([A-Z][A-Za-z0-9_]*)\s+extends\s+([A-Z][A-Za-z0-9_]*)'
+)
+_M3_FILTRO_METHOD_RE = re.compile(
+    r'@M3FiltroSeguridad(?:\([^)]*\))?\s+(?:@[^\s]+\s+)*'
+    r'(?:public|private|protected)\s+\w[\w<>\[\]]*\s+([a-z][A-Za-z0-9_]*)\s*\('
+)
+_LOMBOK_CLASS_RE = re.compile(
+    r'(@(?:Data|Slf4j|Builder|AllArgsConstructor|NoArgsConstructor)(?:\([^)]*\))?\s+)*'
+    r'(?:public\s+)?(?:class|interface)\s+([A-Z][A-Za-z0-9_]*)',
+    re.MULTILINE,
+)
+
+# ---------------------------------------------------------------------------
+# JS/TS keyword and builtin exclusions (Plan 12-03)
+# ---------------------------------------------------------------------------
+
 _JS_KEYWORD_EXCLUSIONS: frozenset[str] = frozenset({
     # JS reserved words
     "if", "else", "for", "while", "do", "switch", "case", "break", "continue",
@@ -645,6 +673,29 @@ class SemanticAnalyzer:
                 ),
             }
 
+            # P1-D: Link @Mapper interfaces to their *Mapper.xml files
+            xml_paths = [p for p in all_paths if p.endswith(("Mapper.xml", "MyBatis.xml"))]
+            mapper_xml_index: dict[str, str] = {}
+            for xp in xml_paths:
+                stem = Path(xp).stem  # e.g. "PacienteMapper"
+                mapper_xml_index[stem] = xp
+
+            for sym in all_symbols:
+                if sym.kind == "mapper_interface" and sym.language == "java":
+                    xml_key = sym.symbol
+                    if xml_key in mapper_xml_index:
+                        xml_path = mapper_xml_index[xml_key]
+                        links.append(SymbolLink(
+                            importer_path=Path(sym.path).as_posix(),
+                            symbol=sym.symbol,
+                            source_path=xml_path,
+                            source_line=None,
+                            is_external=False,
+                            confidence="high",
+                            method="heuristic",
+                            workspace=workspace,
+                        ))
+
         # Determine explicit analysis status — never emit silent empty results.
         # An agent must be able to tell "analysis ran and found nothing" from
         # "analysis failed to run" or "significant coverage gap".
@@ -865,6 +916,8 @@ class SemanticAnalyzer:
         """Heuristic Java/Kotlin: detecta class/method declarations y call sites.
 
         method="heuristic", confidence="low" para todos los edges Java.
+        Includes: Lombok synthetic symbols, @Autowired field edges,
+        @Mapper interface detection, inheritance chains, @M3FiltroSeguridad AOP edges.
         """
         _JAVA_KEYWORDS: frozenset[str] = frozenset({
             "if", "for", "while", "switch", "catch", "super", "this", "new",
@@ -923,6 +976,105 @@ class SemanticAnalyzer:
                     method="heuristic",
                 ))
 
+        # P1-D: @Mapper interface detection — emit as mapper_interface symbol
+        for m in _MAPPER_IFACE_RE.finditer(content):
+            iface_name = m.group(1)
+            line = content[: m.start()].count("\n") + 1
+            symbols.append(SymbolRecord(
+                symbol=iface_name,
+                kind="mapper_interface",
+                language="java",
+                path=rel_path,
+                line=line,
+                exported=True,
+            ))
+
+        # P1-E: Inheritance chain — class X extends Y → edge X→Y
+        for m in _EXTENDS_RE.finditer(content):
+            subclass = m.group(1)
+            superclass = m.group(2)
+            line = content[: m.start()].count("\n") + 1
+            calls.append(CallRecord(
+                caller_path=rel_path,
+                caller_symbol=subclass,
+                callee_path=rel_path,
+                callee_symbol=superclass,
+                call_line=line,
+                confidence="low",
+                method="heuristic",
+            ))
+
+        # P1-F: Lombok annotations — emit synthetic symbols
+        for m in _LOMBOK_CLASS_RE.finditer(content):
+            annotations_block = m.group(0)
+            class_name = m.group(2)
+            line = content[: m.start()].count("\n") + 1
+            if "@Slf4j" in annotations_block:
+                symbols.append(SymbolRecord(
+                    symbol="log",
+                    kind="field",
+                    language="java",
+                    path=rel_path,
+                    line=line,
+                    exported=False,
+                ))
+            if "@Builder" in annotations_block:
+                symbols.append(SymbolRecord(
+                    symbol=f"{class_name}.builder",
+                    kind="function",
+                    language="java",
+                    path=rel_path,
+                    line=line,
+                    exported=True,
+                ))
+            if "@AllArgsConstructor" in annotations_block or "@Data" in annotations_block:
+                symbols.append(SymbolRecord(
+                    symbol=f"{class_name}(allArgs)",
+                    kind="function",
+                    language="java",
+                    path=rel_path,
+                    line=line,
+                    exported=True,
+                ))
+            if "@NoArgsConstructor" in annotations_block:
+                symbols.append(SymbolRecord(
+                    symbol=f"{class_name}()",
+                    kind="function",
+                    language="java",
+                    path=rel_path,
+                    line=line,
+                    exported=True,
+                ))
+
+        # P1-G: @Autowired field injection — emit dependency edges
+        for m in _AUTOWIRED_FIELD_RE.finditer(content):
+            field_type = m.group(1)
+            field_name = m.group(2)
+            line = content[: m.start()].count("\n") + 1
+            calls.append(CallRecord(
+                caller_path=rel_path,
+                caller_symbol=field_name,
+                callee_path=rel_path,
+                callee_symbol=field_type,
+                call_line=line,
+                confidence="medium",
+                method="heuristic",
+            ))
+
+        # P3-C: @M3FiltroSeguridad AOP proxy edges
+        for m in _M3_FILTRO_METHOD_RE.finditer(content):
+            method_name = m.group(1)
+            line = content[: m.start()].count("\n") + 1
+            calls.append(CallRecord(
+                caller_path=rel_path,
+                caller_symbol="M3FiltroSeguridadImpl",
+                callee_path=rel_path,
+                callee_symbol=method_name,
+                call_line=line,
+                confidence="medium",
+                method="heuristic",
+            ))
+
         return symbols, calls
 
     # -----------------------------------------------------------------------