evita la extracció quan no es necesari

core/pipeline.py

@@ -6,7 +6,7 @@ import shutil
 
 import rarfile
 
-from core.archive import detect_real_format, extract_archive, repack_as_cbz, ArchiveError
+from core.archive import detect_real_format, extract_archive, repack_as_cbz, ArchiveError, list_archive_names
 from core.collision import CollisionPolicy, resolve_collision
 from core.result import ComicResult, StepResult
 from processors.validator import validate_archive
@@ -69,6 +69,18 @@ class Pipeline:
 
         return {}
 
+    def _needs_extraction(self, step_results: list, real_format: str) -> bool:
+        for step in self.steps:
+            if step in ("normalize_pages", "normalize_images", "convert_images"):
+                return True
+            if step == "convert" and needs_conversion(real_format, self.desired_format):
+                return True
+            if step == "clean":
+                trash = next((r for r in step_results if r.step == "check_trash"), None)
+                if trash and trash.warnings:
+                    return True
+        return False
+
     def run(self, path: str, confirm_fn=None) -> ComicResult:
         step_results = []
 
@@ -80,20 +92,31 @@ class Pipeline:
 
         real_format = detect_real_format(path)
 
-        # 2. Extraer una sola vez
+        # 2. Obtener lista de miembros sin extraer
+        try:
+            names = list_archive_names(path)
+        except Exception as exc:
+            step_results.append(StepResult(step="list", changed=False, errors=[str(exc)]))
+            return ComicResult(original_path=path, final_path=None, steps=step_results)
+
+        # 3. Ejecutar siempre los 4 content checks sobre los nombres (sin extraer)
+        step_results += [
+            check_trash(names),
+            check_page_numbering(names),
+            check_image_extensions(names),
+            check_comicinfo(names),
+        ]
+
+        # 4. Pre-flight: si ningún step necesita extracción, salir sin tocar el archivo
+        if not self._needs_extraction(step_results, real_format):
+            return ComicResult(original_path=path, final_path=path, steps=step_results)
+
+        # 5. Extraer una sola vez
         temp_dir = tempfile.mkdtemp()
         try:
             extract_archive(path, temp_dir)
 
-            # 3. Ejecutar siempre los 4 content checks
-            step_results += [
-                check_trash(temp_dir),
-                check_page_numbering(temp_dir),
-                check_image_extensions(temp_dir),
-                check_comicinfo(temp_dir),
-            ]
-
-            # 4. Aplicar cada fix step sobre el directorio temporal
+            # 6. Aplicar cada fix step sobre el directorio temporal
             any_changed = False
 
             if "clean" in self.steps:
@@ -152,7 +175,7 @@ class Pipeline:
                     if conv_result.changed:
                         any_changed = True
 
-            # 5. Reempaquetar si hubo cambios o conversión de formato
+            # 7. Reempaquetar si hubo cambios o conversión de formato
             needs_repack = any_changed or (
                 "convert" in self.steps
                 and needs_conversion(real_format, self.desired_format)