fix: resolve type hint mismatch in deduplicator and broaden embedder exceptions (#259)

backend/app/main.py

@@ -1401,6 +1401,14 @@ async def _run_repo_scan_task(
             except ValueError:
                 epsilon = 0.15
 
+
+            # CORRECT WAY:
+            if not disable_dedup and SENTENCE_TRANSFORMERS_AVAILABLE:
+    # The assertion must be safely inside the block body
+              assert all(isinstance(f, Finding) for f in findings), \
+              f"Expected Finding objects, got {set(type(f).__name__ for f in findings)}"
+
+            findings = deduplicate(findings, epsilon)
             if not disable_dedup and SENTENCE_TRANSFORMERS_AVAILABLE:
                 findings = deduplicate(findings, epsilon)
 

backend/app/ml/deduplicator.py

@@ -1,5 +1,10 @@
 from collections import defaultdict
 
+from typing import Union, List, Dict, Any
+from sklearn.cluster import DBSCAN
+from app.models import Finding
+
+
 from sklearn.cluster import DBSCAN
 
 
@@ -28,6 +33,11 @@ def embed_findings(findings):
 
 
 def deduplicate(
+
+    findings: List[Union[Finding, Dict[str, Any]]],
+    epsilon: float = 0.15,
+) -> List[Union[Finding, Dict[str, Any]]]:
+
     findings: list[dict],
     epsilon: float = 0.15,
 ) -> list[dict]:
@@ -73,3 +83,6 @@ def deduplicate(
         results.append(representative)
 
     return results
+
+    return results
+

backend/app/ml/embedder.py

@@ -1,18 +1,54 @@
+
+import logging
+
 import numpy as np
 
 try:
     from sentence_transformers import SentenceTransformer
 
     MODEL = SentenceTransformer("all-MiniLM-L6-v2")
+
+except Exception:
+    MODEL = None
+    logging.getLogger(__name__).warning(
+        "Failed to load sentence-transformers model: all-MiniLM-L6-v2", 
+        exc_info=True
+    )
+
+
+def _extract_text(finding) -> str:
+    """Safely extracts title and description from either a Pydantic Finding object or a raw dict."""
+    if isinstance(finding, dict):
+        return f"{finding.get('title', '')} {finding.get('description', '')}".strip()
+    return f"{getattr(finding, 'title', '')} {getattr(finding, 'description', '')}".strip()
+
+
+def embed_findings(findings: list) -> np.ndarray:
+
 except ImportError:
     MODEL = None
 
 
 def embed_findings(findings: list[dict]) -> np.ndarray:
+
     """
     Convert findings into embeddings.
 
     Each finding is converted to:
+
+    "{title} {description}"
+
+    Returns:
+    		np.ndarray of shape (n, 384)
+    """
+    if MODEL is None:
+        raise RuntimeError(
+            "sentence-transformers is not installed or failed to initialize. "
+            "Install it using: pip install sentence-transformers"
+        )
+
+    texts = [_extract_text(finding) for finding in findings]
+
     "{rule_id} {message} {file_path}"
 
     Returns: