Add offline regression coverage for audio_ocr_helpers

tests/test_audio_ocr_helpers.py

@@ -1,5 +1,7 @@
+from types import SimpleNamespace
+
 from llm_workflows.audio_text import (
-    convert_time_to_srt_format, create_multiline_srt, split_line_by_char_limit,
+    convert_time_to_srt_format, create_multiline_srt, format_segments, split_line_by_char_limit,
 )
 from utils.full_multi_updated2 import detect_languages_in_text, save_results
 
@@ -22,6 +24,35 @@ def test_multiline_srt_file(tmp_path):
     assert out.read_text(encoding="utf-8") == "1\n00:00:00,000 --> 00:00:01,250\nhello world\nagain\n\n"
 
 
+def test_format_segments_handles_empty_and_optional_word_timestamps():
+    assert format_segments([]) == ([], [], "")
+
+    segments = [
+        SimpleNamespace(text="  first sentence  ", start=0.0, end=1.2),
+        SimpleNamespace(
+            text=" second sentence ",
+            start=1.2,
+            end=2.5,
+            words=[
+                SimpleNamespace(word=" first ", start=1.2, end=1.5),
+                SimpleNamespace(word=" second ", start=1.6, end=2.0),
+            ],
+        ),
+    ]
+
+    sentences, words, transcript = format_segments(segments)
+
+    assert sentences == [
+        {"id": 0, "text": "first sentence", "start": 0.0, "end": 1.2},
+        {"id": 1, "text": "second sentence", "start": 1.2, "end": 2.5},
+    ]
+    assert words == [
+        {"word": "first", "start": 1.2, "end": 1.5},
+        {"word": "second", "start": 1.6, "end": 2.0},
+    ]
+    assert transcript == "first sentence second sentence"
+
+
 def test_detect_languages():
     assert detect_languages_in_text("hello") == ["Latin_Script"]
     assert detect_languages_in_text("ଓଡ଼ିଆ") == ["Odia"]