sourcecode 0.13.0__py3-none-any.whl → 0.15.0__py3-none-any.whl

sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Genera mapas de contexto estructurado para agentes IA."""
 
-__version__ = "0.13.0"
+__version__ = "0.15.0"

sourcecode/architecture_analyzer.py

@@ -0,0 +1,347 @@
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Literal, Optional
+
+from sourcecode.schema import (
+    ArchitectureAnalysis,
+    ArchitectureDomain,
+    ArchitectureLayer,
+    BoundedContext,
+    ModuleGraph,
+    SourceMap,
+)
+
+_TOOLING_PREFIXES = (
+    ".claude/",
+    ".vscode/",
+    "bin/",
+    ".git/",
+    "__pycache__/",
+    "node_modules/",
+    ".venv/",
+    "venv/",
+    ".mypy_cache/",
+    ".pytest_cache/",
+    "dist/",
+    "build/",
+)
+_SRC_TRANSPARENT = {"src", "lib", "app", "pkg"}
+_CODE_EXTENSIONS = {
+    ".py", ".js", ".ts", ".tsx", ".jsx", ".mjs", ".cjs",
+    ".go", ".java", ".kt", ".rs", ".rb",
+}
+_GENERIC_NAMES = {"utils", "helpers", "common", "shared", "misc", "core", "root", ""}
+
+DOMAIN_ROLES: dict[str, str] = {
+    "controllers": "HTTP request handlers",
+    "handlers":    "HTTP request handlers",
+    "routes":      "Route definitions",
+    "services":    "Business logic",
+    "usecases":    "Business logic",
+    "application": "Application layer",
+    "repositories": "Data access",
+    "repos":       "Data access",
+    "store":       "Data access",
+    "models":      "Domain models",
+    "entities":    "Domain entities",
+    "domain":      "Domain core",
+    "infra":       "Infrastructure",
+    "infrastructure": "Infrastructure",
+    "adapters":    "Hexagonal adapters",
+    "ports":       "Hexagonal ports",
+    "frontend":    "Frontend/UI layer",
+    "backend":     "Backend/API layer",
+    "components":  "UI components",
+    "pages":       "UI pages/views",
+    "views":       "View layer",
+    "templates":   "View templates",
+    "tests":       "Test suite",
+    "test":        "Test suite",
+}
+
+LAYER_PATTERNS: dict[str, dict[str, list[str]]] = {
+    "mvc": {
+        "controller": ["controller", "controllers", "routes", "views", "handlers"],
+        "model":      ["model", "models", "entity", "entities", "domain"],
+        "view":       ["views", "templates", "pages", "components"],
+    },
+    "layered": {
+        "controller":     ["controller", "controllers", "api", "routes", "handlers", "endpoints"],
+        "service":        ["service", "services", "usecase", "usecases", "application"],
+        "repository":     ["repository", "repositories", "repo", "repos", "store", "storage", "dao"],
+        "infrastructure": ["infra", "infrastructure", "persistence", "db", "database"],
+    },
+    "hexagonal": {
+        "port":    ["port", "ports", "interface", "interfaces"],
+        "adapter": ["adapter", "adapters"],
+        "domain":  ["domain", "core", "model", "models"],
+    },
+    "fullstack": {
+        "frontend": ["frontend", "client", "web", "ui", "pages", "components", "app"],
+        "backend":  ["backend", "server", "api", "services"],
+    },
+}
+
+
+class ArchitectureAnalyzer:
+    """Analiza la arquitectura de un proyecto a partir de su estructura de ficheros y grafo de modulos."""
+
+    def analyze(
+        self,
+        root: Path,
+        sm: SourceMap,
+        graph: Optional[ModuleGraph] = None,
+    ) -> ArchitectureAnalysis:
+        limitations: list[str] = []
+
+        # Step 1: filter paths
+        filtered = self._filter_paths(sm.file_paths)
+        if len(filtered) < 2:
+            return ArchitectureAnalysis(
+                requested=True,
+                pattern="unknown",
+                limitations=["Arquitectura no inferida: proyecto sin archivos de codigo suficientes"],
+            )
+
+        # Step 2: domain clustering
+        domains = self._cluster_domains(filtered)
+
+        # Step 3: layer detection
+        pattern, layers = self._detect_layers(filtered)
+        if pattern in (None, "flat", "unknown"):
+            if pattern == "flat":
+                limitations.append("Patron de capas no detectado: proyecto con estructura plana")
+            elif pattern == "unknown":
+                limitations.append("Patron de capas no reconocido: estructura de directorios sin senales claras")
+
+        # Step 4: bounded context inference
+        bounded_contexts = self._infer_bounded_contexts(domains, graph)
+
+        # Overall confidence
+        confidence: Literal["high", "medium", "low"]
+        if len(domains) >= 3 and pattern not in (None, "unknown"):
+            confidence = "high"
+        elif len(domains) >= 1:
+            confidence = "medium"
+        else:
+            confidence = "low"
+
+        method = "graph+heuristic" if graph is not None else "heuristic"
+
+        return ArchitectureAnalysis(
+            requested=True,
+            pattern=pattern,
+            domains=domains,
+            layers=layers,
+            bounded_contexts=bounded_contexts,
+            confidence=confidence,
+            method=method,
+            limitations=limitations,
+        )
+
+    # ------------------------------------------------------------------
+    # Private helpers
+    # ------------------------------------------------------------------
+
+    def _is_tooling(self, path: str) -> bool:
+        norm = path.replace("\\", "/")
+        return any(norm.startswith(p) for p in _TOOLING_PREFIXES)
+
+    def _filter_paths(self, paths: list[str]) -> list[str]:
+        result = []
+        for p in paths:
+            norm = p.replace("\\", "/")
+            if self._is_tooling(norm):
+                continue
+            ext = Path(norm).suffix.lower()
+            if ext not in _CODE_EXTENSIONS:
+                continue
+            result.append(norm)
+        return result
+
+    def _extract_domain_segment(self, path: str) -> str:
+        parts = path.replace("\\", "/").split("/")
+        if len(parts) == 1:
+            return "root"
+        first = parts[0]
+        if first in _SRC_TRANSPARENT and len(parts) >= 3:
+            return parts[1]
+        return first
+
+    def _cluster_domains(self, paths: list[str]) -> list[ArchitectureDomain]:
+        groups: dict[str, list[str]] = {}
+        for p in paths:
+            seg = self._extract_domain_segment(p)
+            groups.setdefault(seg, []).append(p)
+
+        domains: list[ArchitectureDomain] = []
+        for name, files in groups.items():
+            if len(files) < 2:
+                continue
+            role = DOMAIN_ROLES.get(name, "")
+            domain_confidence: Literal["high", "medium", "low"]
+            if name in DOMAIN_ROLES:
+                domain_confidence = "high"
+            elif len(name) >= 5 and name not in _GENERIC_NAMES:
+                domain_confidence = "medium"
+            else:
+                domain_confidence = "low"
+            domains.append(ArchitectureDomain(name=name, files=files, role=role, confidence=domain_confidence))
+        return domains
+
+    def _detect_layers(self, paths: list[str]) -> tuple[str, list[ArchitectureLayer]]:
+        dir_names: set[str] = set()
+        for p in paths:
+            parts = p.replace("\\", "/").split("/")
+            for part in parts[:-1]:
+                dir_names.add(part.lower())
+
+        best_pattern = ""
+        best_score = 0
+        best_matched: dict[str, list[str]] = {}
+
+        for pattern_name, layer_keys in LAYER_PATTERNS.items():
+            matched: dict[str, list[str]] = {}
+            for layer_key, keywords in layer_keys.items():
+                matched_dirs = [d for d in dir_names if d in keywords]
+                if matched_dirs:
+                    matched[layer_key] = matched_dirs
+            score = len(matched)
+            if score > best_score:
+                best_score = score
+                best_pattern = pattern_name
+                best_matched = matched
+
+        if best_score < 2:
+            # Check depth for flat vs unknown
+            max_depth = max(
+                len(p.replace("\\", "/").split("/")) - 1
+                for p in paths
+            )
+            flat_pattern = "flat" if max_depth <= 2 else "unknown"
+            return flat_pattern, []
+
+        # Build ArchitectureLayer list
+        layer_confidence: Literal["high", "medium", "low"] = "high" if best_score >= 3 else "medium"
+        layers: list[ArchitectureLayer] = []
+        for layer_key, matched_dirs in best_matched.items():
+            matched_files = [
+                p for p in paths
+                if any(
+                    seg.lower() in matched_dirs
+                    for seg in p.replace("\\", "/").split("/")[:-1]
+                )
+            ]
+            layers.append(ArchitectureLayer(
+                name=layer_key,
+                pattern=best_pattern,
+                files=matched_files,
+                confidence=layer_confidence,
+            ))
+
+        return best_pattern or "unknown", layers
+
+    def _infer_bounded_contexts(
+        self,
+        domains: list[ArchitectureDomain],
+        graph: Optional[ModuleGraph],
+    ) -> list[BoundedContext]:
+        # Priority 1: use graph SCCs when available
+        if graph is not None:
+            sccs = self._find_sccs(graph)
+            bc_list: list[BoundedContext] = []
+            for scc in sccs:
+                if len(scc) < 2:
+                    continue
+                # Derive name from most frequent directory segment among nodes
+                seg_counts: dict[str, int] = {}
+                for node_id in scc:
+                    parts = node_id.replace("\\", "/").split("/")
+                    seg = parts[0] if len(parts) == 1 else parts[-2] if len(parts) >= 2 else parts[0]
+                    seg_counts[seg] = seg_counts.get(seg, 0) + 1
+                name = max(seg_counts, key=lambda k: seg_counts[k])
+                if name in _GENERIC_NAMES:
+                    continue
+                bc_conf: Literal["high", "medium", "low"] = "high" if name in DOMAIN_ROLES else "medium"
+                bc_list.append(BoundedContext(name=name, modules=scc, confidence=bc_conf))
+            if bc_list:
+                return bc_list
+
+        # Priority 2: use domains as candidates
+        bc_from_domains: list[BoundedContext] = []
+        for d in domains:
+            if d.name in _GENERIC_NAMES:
+                continue
+            if d.confidence == "low":
+                continue
+            bc_from_domains.append(BoundedContext(
+                name=d.name,
+                modules=d.files,
+                confidence="medium",
+            ))
+        return bc_from_domains
+
+    def _find_sccs(self, graph: ModuleGraph) -> list[list[str]]:
+        """Kosaraju's algorithm for strongly connected components."""
+        import_edges = [
+            (e.source, e.target)
+            for e in graph.edges
+            if e.kind == "imports"
+        ]
+        node_ids = [n.id for n in graph.nodes]
+
+        # Build adjacency lists
+        adj: dict[str, list[str]] = {n: [] for n in node_ids}
+        radj: dict[str, list[str]] = {n: [] for n in node_ids}
+        for src, tgt in import_edges:
+            if src in adj and tgt in adj:
+                adj[src].append(tgt)
+                radj[tgt].append(src)
+
+        visited: set[str] = set()
+        order: list[str] = []
+
+        def dfs1(v: str) -> None:
+            stack = [(v, False)]
+            while stack:
+                node, done = stack.pop()
+                if done:
+                    order.append(node)
+                    continue
+                if node in visited:
+                    continue
+                visited.add(node)
+                stack.append((node, True))
+                for nb in adj[node]:
+                    if nb not in visited:
+                        stack.append((nb, False))
+
+        for n in node_ids:
+            if n not in visited:
+                dfs1(n)
+
+        visited2: set[str] = set()
+        sccs: list[list[str]] = []
+
+        def dfs2(v: str) -> list[str]:
+            comp: list[str] = []
+            stack = [v]
+            while stack:
+                node = stack.pop()
+                if node in visited2:
+                    continue
+                visited2.add(node)
+                comp.append(node)
+                for nb in radj[node]:
+                    if nb not in visited2:
+                        stack.append(nb)
+            return comp
+
+        for n in reversed(order):
+            if n not in visited2:
+                comp = dfs2(n)
+                if len(comp) >= 2:
+                    sccs.append(comp)
+
+        return sccs

sourcecode/architecture_summary.py

@@ -1,4 +1,3 @@
-"""Resumen arquitectonico estatico y compacto para consumo LLM."""
 from __future__ import annotations
 
 import ast

sourcecode/classifier.py

@@ -1,4 +1,3 @@
-"""Clasificacion de tipo de proyecto y scoring de confianza."""
 from __future__ import annotations
 
 from collections.abc import Iterable, Sequence

sourcecode/cli.py

@@ -1,4 +1,3 @@
-"""CLI de sourcecode — interfaz de linea de comandos principal."""
 from __future__ import annotations
 
 import json
@@ -47,7 +46,7 @@ def main(
     compact: bool = typer.Option(
         False,
         "--compact",
-        help="Output reducido (~500 tokens): tipo de proyecto, stacks, entradas y arbol nivel 1",
+        help="Output reducido (~500-700 tokens): tipo, stacks, entradas, arbol nivel 1 y summaries de flags opcionales activos",
     ),
     dependencies: bool = typer.Option(
         False,
@@ -117,6 +116,41 @@ def main(
         "--semantics",
         help="Incluir call graph semantico, linking cross-file de simbolos y resolucion avanzada de imports",
     ),
+    architecture: bool = typer.Option(
+        False,
+        "--architecture",
+        help="Inferencia arquitectonica: dominios funcionales, capas (MVC/layered/hexagonal) y bounded contexts aproximados",
+    ),
+    git_context: bool = typer.Option(
+        False,
+        "--git-context",
+        "-g",
+        help="Incluir contexto git: commits recientes, ficheros mas activos y cambios pendientes",
+    ),
+    git_depth: int = typer.Option(
+        20,
+        "--git-depth",
+        help="Numero de commits recientes a incluir con --git-context (default: 20)",
+        min=1,
+        max=100,
+    ),
+    git_days: int = typer.Option(
+        90,
+        "--git-days",
+        help="Ventana temporal en dias para detectar ficheros mas activos con --git-context (default: 90)",
+        min=1,
+        max=3650,
+    ),
+    env_map: bool = typer.Option(
+        False,
+        "--env-map",
+        help="Incluir mapa de variables de entorno: claves, tipos, categorias y ficheros que las referencian",
+    ),
+    code_notes: bool = typer.Option(
+        False,
+        "--code-notes",
+        help="Extraer anotaciones TODO/FIXME/HACK/NOTE/DEPRECATED/WARNING/BUG/XXX/OPTIMIZE con ubicacion y simbolo envolvente, y detectar ADRs en docs/decisions/, docs/adr/ y similares",
+    ),
 ) -> None:
     """Genera un mapa de contexto estructurado del proyecto en formato JSON o YAML."""
     # Validar formato
@@ -158,6 +192,7 @@ def main(
     from sourcecode.metrics_analyzer import MetricsAnalyzer
     from sourcecode.redactor import SecretRedactor, redact_dict
     from sourcecode.scanner import FileScanner
+    from sourcecode.semantic_analyzer import SemanticAnalyzer
     from sourcecode.schema import (
         AnalysisMetadata,
         DocRecord,
@@ -233,7 +268,6 @@ def main(
     doc_analyzer = DocAnalyzer() if docs else None
     metrics_analyzer = MetricsAnalyzer() if full_metrics else None
 
-    from sourcecode.semantic_analyzer import SemanticAnalyzer
     semantic_analyzer = SemanticAnalyzer() if semantics else None
 
     root_manifests = [
@@ -292,7 +326,7 @@ def main(
         doc_records.extend(root_doc_records)
         doc_summaries.append(root_doc_summary)
 
-    file_metrics_records: list = []
+    file_metrics_records: list[Any] = []
     metrics_summaries = []
     if metrics_analyzer is not None:
         root_metrics_tree = (
@@ -500,6 +534,29 @@ def main(
     sm.project_summary = ProjectSummarizer(target).generate(sm)
     sm.architecture_summary = ArchitectureSummarizer(target).generate(sm)
 
+    # Phase 13 Plan 04: Architectural Inference (--architecture flag)
+    if architecture:
+        from sourcecode.architecture_analyzer import ArchitectureAnalyzer
+        arch_graph = module_graph  # None si --graph-modules no fue pasado
+        sm.architecture = ArchitectureAnalyzer().analyze(target, sm, arch_graph)
+
+    # Git Context (--git-context flag)
+    if git_context:
+        from sourcecode.git_analyzer import GitAnalyzer
+        sm.git_context = GitAnalyzer().analyze(target, depth=git_depth, days=git_days)
+
+    # Env Map (--env-map flag)
+    if env_map:
+        from sourcecode.env_analyzer import EnvAnalyzer
+        env_records, env_summary = EnvAnalyzer().analyze(target, file_tree)
+        sm = replace(sm, env_map=env_records, env_summary=env_summary)
+
+    # Code Notes (--code-notes flag)
+    if code_notes:
+        from sourcecode.code_notes_analyzer import CodeNotesAnalyzer
+        cn_notes, cn_adrs, cn_summary = CodeNotesAnalyzer().analyze(target)
+        sm = replace(sm, code_notes=cn_notes, code_adrs=cn_adrs, code_notes_summary=cn_summary)
+
     # 4. Serializar (con o sin modo compact)
     if compact:
         data = compact_view(sm)
@@ -536,33 +593,34 @@ def main(
     # 5. Escribir output (CLI-04)
     write_output(content, output=output)
 
+
+@app.command("prepare-context")
+def prepare_context_cmd(
+    task: str = typer.Argument(..., help="Descripción de la tarea"),
+    path: Path = typer.Option(".", "--path", "-p", help="Directorio del proyecto"),
+) -> None:
+    """Prepara contexto mínimo optimizado para que un LLM modifique el código."""
+    from dataclasses import asdict
+
     from sourcecode.prepare_context import ContextBuilder
 
-    @app.command("prepare-context")
-    def prepare_context_cmd(
-            task: str = typer.Argument(..., help="Descripción de la tarea"),
-            path: Path = typer.Option(".", help="Directorio del proyecto"),
-    ):
-        """
-        Prepara contexto mínimo optimizado para que un LLM modifique el código.
-        """
-        target = path.resolve()
-
-        if not target.exists() or not target.is_dir():
-            typer.echo(f"Error: '{target}' no es un directorio válido.", err=True)
-            raise typer.Exit(code=1)
+    target = path.resolve()
+
+    if not target.exists() or not target.is_dir():
+        typer.echo(f"Error: '{target}' no es un directorio válido.", err=True)
+        raise typer.Exit(code=1)
 
-        builder = ContextBuilder(target)
-        result = builder.prepare(task)
+    builder = ContextBuilder(target)
+    result = builder.prepare(task)
 
-        output = {
-            "task": result.task,
-            "entry_points": result.entry_points,
-            "relevant_files": [f.__dict__ for f in result.relevant_files],
-            "call_flow": result.call_flow,
-            "snippets": [s.__dict__ for s in result.snippets],
-            "tests": result.tests,
-            "notes": result.notes,
-        }
+    out = {
+        "task": result.task,
+        "entry_points": result.entry_points,
+        "relevant_files": [asdict(f) for f in result.relevant_files],
+        "call_flow": result.call_flow,
+        "snippets": [asdict(s) for s in result.snippets],
+        "tests": result.tests,
+        "notes": result.notes,
+    }
 
-        typer.echo(json.dumps(output, indent=2, ensure_ascii=False))
+    typer.echo(json.dumps(out, indent=2, ensure_ascii=False))