From d0a201476e11167b03a46e835785058f238e3c2d Mon Sep 17 00:00:00 2001
From: Santhosh Janardhanan <santhoshj@gmail.com>
Date: Mon, 13 Apr 2026 14:33:26 -0400
Subject: [PATCH] fix: add IndexError guard and error handling test

---
 src/companion/rag/search.py |  7 ++++++-
 tests/test_search.py        | 23 +++++++++++++++++++++++
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/src/companion/rag/search.py b/src/companion/rag/search.py
index dd21d6e..598ceb1 100644
--- a/src/companion/rag/search.py
+++ b/src/companion/rag/search.py
@@ -67,7 +67,12 @@ class SearchEngine:
         k = top_k or self.default_top_k
 
         try:
-            query_embedding = self.embedder.embed([query])[0]
+            embeddings = self.embedder.embed([query])
+            if not embeddings:
+                raise RuntimeError(
+                    "Failed to generate embedding for query: embedder returned empty result"
+                )
+            query_embedding = embeddings[0]
         except RuntimeError as e:
             raise RuntimeError(f"Failed to generate embedding for query: {e}") from e
 
diff --git a/tests/test_search.py b/tests/test_search.py
index f86bf03..6812c4a 100644
--- a/tests/test_search.py
+++ b/tests/test_search.py
@@ -2,6 +2,8 @@ import tempfile
 from pathlib import Path
 from unittest.mock import MagicMock, patch
 
+import pytest
+
 from companion.rag.search import SearchEngine
 from companion.rag.vector_store import VectorStore
 
@@ -32,3 +34,24 @@ def test_search_returns_results(mock_embedder_cls):
         results = engine.search("hello")
         assert len(results) == 1
         assert results[0]["source_file"] == "a.md"
+
+
+@patch("companion.rag.search.OllamaEmbedder")
+def test_search_raises_on_embedder_failure(mock_embedder_cls):
+    mock_embedder = MagicMock()
+    mock_embedder.embed.side_effect = RuntimeError("Connection failed")
+    mock_embedder_cls.return_value = mock_embedder
+
+    with tempfile.TemporaryDirectory() as tmp:
+        store = VectorStore(uri=tmp, dimensions=4)
+        engine = SearchEngine(
+            vector_store=store,
+            embedder_base_url="http://localhost:11434",
+            embedder_model="dummy",
+            embedder_batch_size=32,
+            default_top_k=5,
+            similarity_threshold=0.0,
+            hybrid_search_enabled=False,
+        )
+        with pytest.raises(RuntimeError, match="Failed to generate embedding"):
+            engine.search("hello")