Chiusura fase di ristrutturazione e modularizzazione del workflow per favorire sviluppo modulare

harness/tht/evidence/corpus/pipeline.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 
 import hashlib
 import json
+import logging
 import re
 import uuid
 from collections.abc import Mapping, Sequence
@@ -11,17 +12,16 @@ from dataclasses import asdict, dataclass, field
 from datetime import UTC
 from pathlib import Path
 
+import tht.evidence.acquisition as evidence_acquisition
+from tht.evidence.contracts import EvidenceSource, SourceObject, canonical_provenance_uri
 from tht.evidence.corpus.chunk import ChunkPolicy, chunk
 from tht.evidence.corpus.models import CanonicalChunk, CanonicalDocument, CorpusManifest
 from tht.evidence.corpus.normalize import normalize
 from tht.evidence.corpus.store import CorpusStore
-import tht.evidence.acquisition as evidence_acquisition
-from tht.evidence.contracts import EvidenceSource, SourceObject, canonical_provenance_uri
-from tht.ports.vector import VectorStore, VectorWriteRecord
-from tht.vectorstore.records import VectorRecord
 from tht.jobs.models import JobSpec
 from tht.jobs.runner import JobContext, StageArtifacts, run_job, seal_stage_artifacts
-
+from tht.ports.vector import VectorStore, VectorWriteRecord
+from tht.vectorstore.records import VectorRecord
 
 EVIDENCE_STAGE_IDS = (
     "discover",
@@ -33,6 +33,8 @@ EVIDENCE_STAGE_IDS = (
     "retention_cleanup",
 )
 
+logger = logging.getLogger(__name__)
+
 
 class PipelineError(RuntimeError):
     """Credential-free failure at the preprocessing boundary."""
@@ -212,14 +214,14 @@ class CorpusPipeline:
             if purge_vector:
                 try:
                     self.vector_store.delete_generation("evidence", generation, self.workspace_id)
-                except Exception:
+                except Exception:  # noqa: BLE001 - retention reports per-generation failures
                     failures.append({"generation": generation, "error": "vector cleanup failed"})
                     continue
             try:
                 if purge_filesystem:
                     self.store.discard(generation)
                 evicted.append(generation)
-            except Exception:
+            except Exception:  # noqa: BLE001 - retention reports per-generation failures
                 failures.append({"generation": generation, "error": "filesystem cleanup failed"})
         return {"status": "partial" if failures else "succeeded", "dry_run": dry_run,
                 "active_generation": self.store.active_generation(), "evicted": evicted,
@@ -378,7 +380,7 @@ class CorpusPipeline:
 
         try:
             active_assets_valid = active_assets_are_valid(previous)
-        except Exception:
+        except Exception:  # noqa: BLE001 - any corrupt active asset disables reuse
             active_assets_valid = False
         reusable = (
             active_assets_valid
@@ -538,7 +540,7 @@ class CorpusPipeline:
             try:
                 self.vector_store.delete_generation("evidence", generation, self.workspace_id)
             except Exception:
-                pass
+                logger.debug("Failed to clean the compensated vector generation", exc_info=True)
             write(context, "compensated.json", {"generation": generation})
 
         def rotate_compensated_generation(context: JobContext) -> None:
@@ -786,12 +788,12 @@ class CorpusPipeline:
         try:
             self.store.discard(generation)
         except Exception:
-            pass
+            logger.debug("Failed to discard the unpublished evidence generation", exc_info=True)
         if vector_written:
             try:
                 self.vector_store.delete_generation("evidence", generation, self.workspace_id)
             except Exception:
-                pass
+                logger.debug("Failed to delete the unpublished vector generation", exc_info=True)
 
     @staticmethod
     def _vector_record(