sourcecode 1.16.0__tar.gz → 1.18.0__tar.gz

{sourcecode-1.16.0 → sourcecode-1.18.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sourcecode
-Version: 1.16.0
+Version: 1.18.0
 Summary: Deterministic codebase context for AI coding agents
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -221,7 +221,7 @@ Description-Content-Type: text/markdown
 
 **Compressed AI-ready context for Java/Spring enterprise codebases.**
 
-![Version](https://img.shields.io/badge/version-1.16.0-blue)
+![Version](https://img.shields.io/badge/version-1.18.0-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -255,7 +255,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.16.0
+# sourcecode 1.18.0
 ```
 
 ---

{sourcecode-1.16.0 → sourcecode-1.18.0}/README.md

@@ -2,7 +2,7 @@
 
 **Compressed AI-ready context for Java/Spring enterprise codebases.**
 
-![Version](https://img.shields.io/badge/version-1.16.0-blue)
+![Version](https://img.shields.io/badge/version-1.18.0-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -36,7 +36,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.16.0
+# sourcecode 1.18.0
 ```
 
 ---

{sourcecode-1.16.0 → sourcecode-1.18.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "sourcecode"
-version = "1.16.0"
+version = "1.18.0"
 description = "Deterministic codebase context for AI coding agents"
 readme = "README.md"
 requires-python = ">=3.9"

{sourcecode-1.16.0 → sourcecode-1.18.0}/src/sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Deterministic codebase context maps for AI coding agents."""
 
-__version__ = "1.16.0"
+__version__ = "1.18.0"

{sourcecode-1.16.0 → sourcecode-1.18.0}/src/sourcecode/cli.py

@@ -511,6 +511,11 @@ def main(
         "--agent",
         help="Structured noise-free JSON for AI agents: identity, entry points, dependencies, confidence, gaps.",
     ),
+    full: bool = typer.Option(
+        False,
+        "--full",
+        help="Remove truncation limits on transactional_boundaries, mybatis.dto_mappers, and other capped lists.",
+    ),
     trace_pipeline: bool = typer.Option(
         False,
         "--trace-pipeline",
@@ -1504,12 +1509,12 @@ def main(
         else:
             content = json.dumps(data, indent=2, ensure_ascii=False)
     elif agent:
-        data = agent_view(sm)
+        data = agent_view(sm, full=full)
         if not no_redact:
             data = redact_dict(data)
         content = json.dumps(data, indent=2, ensure_ascii=False)
     elif compact:
-        data = compact_view(sm, no_tree=no_tree)
+        data = compact_view(sm, no_tree=no_tree, full=full)
         if not no_redact:
             data = redact_dict(data)
         content = json.dumps(data, indent=2, ensure_ascii=False)
@@ -1607,6 +1612,17 @@ def prepare_context_cmd(
         "-c",
         help="Copy output to system clipboard after a successful run. No-op when clipboard is unavailable.",
     ),
+    output_path: Optional[Path] = typer.Option(
+        None,
+        "--output",
+        "-o",
+        help="Write output to a file instead of stdout (UTF-8, avoids PowerShell BOM on Windows).",
+    ),
+    symptom: Optional[str] = typer.Option(
+        None,
+        "--symptom",
+        help="(fix-bug) Keyword hint for the bug: boosts matching files and surfaces related code notes.",
+    ),
 ) -> None:
     """Task-specific context for AI coding agents.
 
@@ -1673,7 +1689,7 @@ def prepare_context_cmd(
     from dataclasses import asdict
 
     builder = TaskContextBuilder(target)
-    output = builder.build(task, since=since)
+    output = builder.build(task, since=since, symptom=symptom)
 
     # Task-specific content-filter: each task emphasizes different output fields.
     # Fields marked False are suppressed from this task's output to reduce noise.
@@ -1757,16 +1773,23 @@ def prepare_context_cmd(
         out["affected_entry_points"] = output.affected_entry_points
     if output.limitations:
         out["limitations"] = output.limitations
+    if output.symptom:
+        out["symptom"] = output.symptom
+    if output.related_notes:
+        out["related_notes"] = output.related_notes
     if llm_prompt:
         out["llm_prompt"] = builder.render_prompt(output)
 
-    import sys as _sys
     _pc_content = json.dumps(out, indent=2, ensure_ascii=False)
-    _pc_bytes = _pc_content.encode("utf-8")
-    _sys.stdout.buffer.write(_pc_bytes)
-    if not _pc_content.endswith("\n"):
-        _sys.stdout.buffer.write(b"\n")
-    _sys.stdout.buffer.flush()
+    if output_path is not None:
+        output_path.write_text(_pc_content, encoding="utf-8")
+    else:
+        import sys as _sys
+        _pc_bytes = _pc_content.encode("utf-8")
+        _sys.stdout.buffer.write(_pc_bytes)
+        if not _pc_content.endswith("\n"):
+            _sys.stdout.buffer.write(b"\n")
+        _sys.stdout.buffer.flush()
 
     if copy:
         _trimmed = _pc_content.strip()

{sourcecode-1.16.0 → sourcecode-1.18.0}/src/sourcecode/detectors/java.py

@@ -272,6 +272,16 @@ class JavaDetector(AbstractDetector):
             frameworks.append(FrameworkDetection(name="Jakarta EE", source=source))
         if "mybatis" in text:
             frameworks.append(FrameworkDetection(name="MyBatis", source=source))
+        if "spring-boot-starter-security" in text or "spring-security-core" in text:
+            frameworks.append(FrameworkDetection(name="Spring Security", source=source))
+        if "spring-boot-starter-data-jpa" in text or "spring-data-jpa" in text:
+            frameworks.append(FrameworkDetection(name="Spring Data JPA", source=source))
+        if "spring-ldap-core" in text or "spring-security-ldap" in text:
+            frameworks.append(FrameworkDetection(name="Spring LDAP", source=source))
+        if "spring-aspects" in text or "spring-aop" in text:
+            frameworks.append(FrameworkDetection(name="Spring AOP", source=source))
+        if "spring-boot-starter-activemq" in text or "activemq-broker" in text or "activemq-client" in text:
+            frameworks.append(FrameworkDetection(name="ActiveMQ", source=source))
         return frameworks
 
     def _collect_entry_points(self, context: DetectionContext) -> list[EntryPoint]:

{sourcecode-1.16.0 → sourcecode-1.18.0}/src/sourcecode/prepare_context.py

@@ -321,6 +321,8 @@ class TaskOutput:
     why_these_files: dict[str, str] = field(default_factory=dict)  # path → why relevant
     changed_files: list[str] = field(default_factory=list)         # delta task only
     affected_entry_points: list[str] = field(default_factory=list) # delta task only
+    symptom: Optional[str] = None                                  # fix-bug only
+    related_notes: list[dict] = field(default_factory=list)        # fix-bug + symptom only
 
 
 # ─────────────────────────────────────────────────────────────────────────────
@@ -385,7 +387,7 @@ class TaskContextBuilder:
     def __init__(self, root: Path) -> None:
         self.root = root
 
-    def build(self, task_name: str, *, since: Optional[str] = None) -> TaskOutput:
+    def build(self, task_name: str, *, since: Optional[str] = None, symptom: Optional[str] = None) -> TaskOutput:
         if task_name not in TASKS:
             raise ValueError(
                 f"Unknown task '{task_name}'. Available: {', '.join(TASKS)}"
@@ -630,6 +632,56 @@ class TaskContextBuilder:
             delta_files=_delta_files,
         )
 
+        # ── 6b. Symptom keyword boost + related notes (fix-bug + --symptom) ──
+        symptom_keywords: list[str] = []
+        related_notes: list[dict] = []
+        if task_name == "fix-bug" and symptom:
+            import re as _re
+            symptom_keywords = [
+                w.lower() for w in _re.split(r"[\s\W]+", symptom)
+                if len(w) > 2
+            ]
+            if symptom_keywords:
+                # Surface code notes whose text contains any keyword
+                for _n in cn_notes_for_ranking:
+                    _text = (getattr(_n, "text", "") or "").lower()
+                    if any(kw in _text for kw in symptom_keywords):
+                        related_notes.append({
+                            "kind": getattr(_n, "kind", ""),
+                            "path": getattr(_n, "path", ""),
+                            "line": getattr(_n, "line", None),
+                            "text": getattr(_n, "text", ""),
+                        })
+                # Secondary pass: inject files whose path matches symptom keywords
+                # but weren't in the candidate pool (no structural/git signals).
+                _existing_paths = {rf.path for rf in relevant_files}
+                for _p in all_paths:
+                    if _p in _existing_paths:
+                        continue
+                    if Path(_p).suffix.lower() not in _ALL_EXTENSIONS:
+                        continue
+                    _p_lower = _p.lower()
+                    _matching_kws = [kw for kw in symptom_keywords if kw in _p_lower]
+                    if not _matching_kws:
+                        continue
+                    _boost = 0.2 * len(_matching_kws)
+                    _injected_score = round(min(0.5 + _boost, 1.0), 2)
+                    _first_kw = _matching_kws[0]
+                    relevant_files.append(RelevantFile(
+                        path=_p,
+                        role="symptom_match",
+                        score=_injected_score,
+                        reason=f"path matches symptom keyword: {_first_kw}",
+                        why=f"symptom injection: {', '.join(_matching_kws)}",
+                    ))
+                    _existing_paths.add(_p)
+
+                # Re-rank all relevant_files: boost files whose path matches keywords
+                def _symptom_score(rf: "RelevantFile") -> float:
+                    path_lower = rf.path.lower()
+                    return rf.score + 0.2 * sum(1.0 for kw in symptom_keywords if kw in path_lower)
+                relevant_files = sorted(relevant_files, key=lambda rf: -_symptom_score(rf))
+
         # ── 7. Test gaps (generate-tests only) ────────────────────────────
         test_gaps: list[str] = []
         if task_name == "generate-tests":
@@ -708,6 +760,8 @@ class TaskContextBuilder:
             why_these_files=why_these_files,
             changed_files=changed_files,
             affected_entry_points=affected_entry_points,
+            symptom=symptom if task_name == "fix-bug" and symptom else None,
+            related_notes=related_notes,
         )
 
     def render_prompt(self, output: TaskOutput) -> str:
@@ -988,12 +1042,18 @@ class TaskContextBuilder:
         return Path(path).suffix.lower() in _SOURCE_EXTENSIONS
 
     def _get_git_changed_files(self, since: Optional[str] = None) -> list[str]:
-        """Get files changed since a git ref (default: HEAD~1) relative to root."""
+        """Get files changed since a git ref (default: HEAD~1) relative to self.root.
+
+        Uses --relative so paths are relative to cwd (self.root), not the git repo
+        root. This is critical for monorepos where self.root is a subpath of the
+        git root and git diff would otherwise return prefixed paths that don't match
+        the scanned file tree.
+        """
         import subprocess
         ref = since or "HEAD~1"
         try:
             result = subprocess.run(
-                ["git", "diff", "--name-only", ref, "HEAD"],
+                ["git", "diff", "--name-only", "--relative", ref, "HEAD"],
                 cwd=str(self.root),
                 capture_output=True,
                 text=True,
@@ -1011,7 +1071,7 @@ class TaskContextBuilder:
         # Fallback: uncommitted changes
         try:
             result = subprocess.run(
-                ["git", "diff", "--name-only"],
+                ["git", "diff", "--name-only", "--relative"],
                 cwd=str(self.root),
                 capture_output=True,
                 text=True,

{sourcecode-1.16.0 → sourcecode-1.18.0}/src/sourcecode/serializer.py

@@ -42,6 +42,14 @@ _MAX_DEFAULT_CONTRACTS = 20  # max contracts in default/standard contract output
 _MAX_HARD_SIGNALS_DEFAULT = 20  # max hard_signals entries in default output
 
 
+def _truncate_note(text: str, limit: int) -> str:
+    """Truncate note text at word boundary, appending … when cut."""
+    if len(text) <= limit:
+        return text
+    cut = text[:limit].rsplit(" ", 1)[0]
+    return cut.rstrip(".,;") + "…"
+
+
 def _cap_contracts_for_output(
     contracts: list[Any],
     max_count: int = _MAX_DEFAULT_CONTRACTS,
@@ -301,7 +309,7 @@ def _is_dto_mapper(path: str, root: Optional[Path] = None) -> bool:
     return False
 
 
-def _mybatis_pairing(sm: "SourceMap") -> "Optional[dict[str, Any]]":
+def _mybatis_pairing(sm: "SourceMap", *, full: bool = False) -> "Optional[dict[str, Any]]":
     """Lightweight MyBatis mapper interface <-> XML file pairing from file_paths.
 
     Separates genuine @Mapper interfaces (need XML) from DtoMapper bean-mapping
@@ -343,7 +351,11 @@ def _mybatis_pairing(sm: "SourceMap") -> "Optional[dict[str, Any]]":
     if missing_xml:
         result["missing_xml"] = missing_xml[:5]
     if dto_mappers:
-        result["dto_mappers"] = dto_mappers[:10]
+        _total_dto = len(dto_mappers)
+        result["dto_mappers"] = dto_mappers if full else dto_mappers[:10]
+        result["dto_mappers_total"] = _total_dto
+        if _total_dto > 10 and not full:
+            result["dto_mappers_truncated"] = True
     return result
 
 
@@ -371,16 +383,27 @@ def _spring_profiles_context(sm: "SourceMap") -> "Optional[dict[str, Any]]":
     if not profiles:
         return None
 
-    # Per-profile variants: Java files whose name contains a profile name (case-insensitive)
+    # Per-profile variants: Java/XML files whose stem or path contains the profile name.
+    # Portal profiles (e.g. "ingesa-portal") use dash in directory/resource names but
+    # never in Java class names — match on full path instead of stem only.
     per_profile: dict[str, list[str]] = {}
     for profile in profiles:
         pfx = profile.lower()
-        matches = [
-            p for p in sm.file_paths
-            if pfx in Path(p).stem.lower() and p.endswith(".java")
-        ]
+        is_portal = "-" in profile
+        if is_portal:
+            matches = [
+                p for p in sm.file_paths
+                if pfx in p.lower() and (p.endswith(".java") or p.endswith(".xml"))
+            ]
+        else:
+            matches = [
+                p for p in sm.file_paths
+                if pfx in Path(p).stem.lower() and p.endswith(".java")
+            ]
         if matches:
             per_profile[profile] = [Path(p).name for p in matches[:5]]
+        elif is_portal:
+            per_profile[profile] = []
 
     result: dict[str, Any] = {"detected": profiles}
     if per_profile:
@@ -398,26 +421,63 @@ def _spring_profiles_context(sm: "SourceMap") -> "Optional[dict[str, Any]]":
     return result
 
 
-def _transactional_summary(sm: "SourceMap") -> "Optional[dict[str, Any]]":
+def _transactional_summary(sm: "SourceMap", *, full: bool = False) -> "Optional[dict[str, Any]]":
     """Surface @Transactional class boundaries from the Java stack detection."""
     for s in sm.stacks:
         classes = getattr(s, "transactional_classes", [])
         if classes:
             total = len(classes)
             result: dict[str, Any] = {"count": total, "classes": classes}
-            if total > 10:
+            if total > 10 and not full:
                 result["classes"] = classes[:10]
                 result["truncated"] = True
-                result["note"] = f"showing 10 of {total}; use --full for complete list"
+                result["note"] = f"showing 10 of {total}; use --full to see all {total}"
             return result
     return None
 
 
-def _security_surface_from_eps(eps: list) -> "Optional[dict[str, Any]]":
+def _resolve_java_constant(symbol: str, root: "Optional[Path]", file_paths: "Optional[list[str]]") -> str:
+    """Resolve a Java constant reference like ClassName.FIELD_NAME to its string value."""
+    import re as _re
+    if not root or not file_paths or "." not in symbol:
+        return symbol
+    parts = symbol.rsplit(".", 1)
+    if len(parts) != 2:
+        return symbol
+    class_name, field_name = parts
+    if not class_name or not field_name or not field_name.isupper():
+        return symbol
+    target_file = f"{class_name}.java"
+    candidates = [p for p in file_paths if Path(p).name == target_file]
+    if not candidates:
+        return symbol
+    _CONST_RE = _re.compile(
+        r'\b' + _re.escape(field_name) + r'\s*=\s*"([^"]+)"'
+    )
+    for rel_path in candidates:
+        try:
+            abs_path = root / rel_path
+            content = abs_path.read_text(encoding="utf-8", errors="replace")
+            m = _CONST_RE.search(content)
+            if m:
+                return m.group(1)
+        except OSError:
+            continue
+    return symbol
+
+
+def _security_surface_from_eps(
+    eps: list,
+    *,
+    root: "Optional[Path]" = None,
+    file_paths: "Optional[list[str]]" = None,
+) -> "Optional[dict[str, Any]]":
     """Extract @M3FiltroSeguridad resource names from entry point evidence strings."""
     import re as _re
     _NOMBRE_RE = _re.compile(r"nombreRecurso=[\"']([^\"']+)[\"']")
+    _CONST_SYMBOL_RE = _re.compile(r'^[\w]+\.[\w]+$')
     resource_names: list[str] = []
+    unresolved: list[str] = []
     seen: set[str] = set()
     for ep in eps:
         evidence = getattr(ep, "evidence", None)
@@ -425,20 +485,30 @@ def _security_surface_from_eps(eps: list) -> "Optional[dict[str, Any]]":
             continue
         for m in _NOMBRE_RE.finditer(evidence):
             nm = m.group(1)
-            if nm and nm not in seen:
-                seen.add(nm)
+            if not nm or nm in seen:
+                continue
+            seen.add(nm)
+            if _CONST_SYMBOL_RE.match(nm):
+                resolved = _resolve_java_constant(nm, root, file_paths)
+                if resolved != nm:
+                    resource_names.append(resolved)
+                else:
+                    resource_names.append(nm)
+                    unresolved.append(nm)
+            else:
                 resource_names.append(nm)
     if not resource_names:
         return None
-    return {
+    result: dict[str, Any] = {
         "schema": (
             "Values used in @M3FiltroSeguridad(nombreRecurso=VALUE) on REST controller "
             "methods. Each value names a permission resource checked at runtime."
         ),
-        # resource_names kept for backward compatibility; resources is the canonical key
         "resource_names": resource_names,
-        "resources": resource_names,
     }
+    if unresolved:
+        result["resource_names_unresolved"] = unresolved
+    return result
 
 
 def _bootstrap_structured(eps: list) -> "Optional[dict[str, Any]]":
@@ -702,6 +772,7 @@ def _file_relevance(sm: SourceMap, *, limit: int = _FILE_RELEVANCE_LIMIT) -> lis
             "category": file_class.category if file_class else "source",
             "confidence": file_class.confidence if file_class else "low",
             "relevance": relevance_val,
+            "score": relevance_val,
             "reason": file_class.reason if file_class else (fs.reasons[0] if fs.reasons else "source file"),
             "evidence": file_class.evidence if file_class else [],
         }
@@ -712,6 +783,18 @@ def _file_relevance(sm: SourceMap, *, limit: int = _FILE_RELEVANCE_LIMIT) -> lis
         if ranking_reasons:
             item["ranking_reasons"] = ranking_reasons
 
+        # Override metadata for known M3 base controller classes
+        if any(k in stem for k in ("GenericRestController", "GenericCRUDRestController")):
+            item["category"] = "runtime_core"
+            item["relevance"] = 0.95
+            item["score"] = 0.95
+            item["reason"] = (
+                "base class for all REST controllers — extends this to get "
+                "centralized exception handling via handlerException()"
+            )
+            item["evidence"] = ["base_rest_controller"]
+            item["ranking_reasons"] = ["universal base class", "exception handling contract"]
+
         scored.append((combined, item))
 
     # Deterministic sort: score desc, then path asc
@@ -831,7 +914,7 @@ def _section_confidence(sm: SourceMap) -> dict[str, str]:
     }
 
 
-def compact_view(sm: SourceMap, *, no_tree: bool = False) -> dict[str, Any]:
+def compact_view(sm: SourceMap, *, no_tree: bool = False, full: bool = False) -> dict[str, Any]:
     """Context package ready for prompt or handoff (~300-500 tokens).
 
     Answers: what it is, where it enters, what depends on what,
@@ -914,7 +997,7 @@ def compact_view(sm: SourceMap, *, no_tree: bool = False) -> dict[str, Any]:
                     "kind": getattr(n, "kind", ""),
                     "path": getattr(n, "path", ""),
                     "line": getattr(n, "line", None),
-                    **({"text": getattr(n, "text", "")[:60]} if getattr(n, "text", "") else {}),
+                    **({"text": _truncate_note(getattr(n, "text", ""), 120)} if getattr(n, "text", "") else {}),
                 }
                 for n in _sorted_notes[:_CODE_NOTES_CAP]
             ]
@@ -960,7 +1043,12 @@ def compact_view(sm: SourceMap, *, no_tree: bool = False) -> dict[str, Any]:
     conf_dict: Any = None
     if sm.confidence_summary is not None:
         cs = sm.confidence_summary
-        conf_dict = {"overall": cs.overall, "stack": cs.stack_confidence, "entry_points": cs.entry_point_confidence}
+        conf_dict = {
+            "overall": cs.overall,
+            "stack": cs.stack_confidence,
+            "entry_points": cs.entry_point_confidence,
+            "sections": _section_confidence(sm),
+        }
         if cs.anomalies:
             conf_dict["anomalies"] = cs.anomalies
 
@@ -987,9 +1075,10 @@ def compact_view(sm: SourceMap, *, no_tree: bool = False) -> dict[str, Any]:
         if _app_server:
             _deployment["app_server_hint"] = _app_server
     _deploy_risks = _project_deployment_risks(sm)
-    _security_surface = _security_surface_from_eps(sm.entry_points)
-    _mybatis = _mybatis_pairing(sm)
-    _transactional = _transactional_summary(sm)
+    _sec_root = Path(sm.metadata.analyzed_path) if sm.metadata.analyzed_path else None
+    _security_surface = _security_surface_from_eps(sm.entry_points, root=_sec_root, file_paths=sm.file_paths)
+    _mybatis = _mybatis_pairing(sm, full=full)
+    _transactional = _transactional_summary(sm, full=full)
     _git_ctx = _compact_git_context(sm)
     _spring_profiles = _spring_profiles_context(sm)
 
@@ -1321,7 +1410,7 @@ def validate_cross_analyzer_consistency(
     return findings
 
 
-def agent_view(sm: SourceMap) -> dict[str, Any]:
+def agent_view(sm: SourceMap, *, full: bool = False) -> dict[str, Any]:
     """Opinionated output for AI agents — structured, noise-free, gap-aware.
 
     Output order:
@@ -1505,13 +1594,14 @@ def agent_view(sm: SourceMap) -> dict[str, Any]:
         signals["semantic_graph"] = sem_info
 
     # Java/Spring: security surface, ORM structure, transactional boundaries
-    _sec_surf = _security_surface_from_eps(sm.entry_points)
+    _av_root = Path(sm.metadata.analyzed_path) if sm.metadata.analyzed_path else None
+    _sec_surf = _security_surface_from_eps(sm.entry_points, root=_av_root, file_paths=sm.file_paths)
     if _sec_surf:
         signals["security_surface"] = _sec_surf
-    _mb = _mybatis_pairing(sm)
+    _mb = _mybatis_pairing(sm, full=full)
     if _mb:
         signals["mybatis"] = _mb
-    _txn = _transactional_summary(sm)
+    _txn = _transactional_summary(sm, full=full)
     if _txn:
         signals["transactional_boundaries"] = _txn