Implemented deterministic transcript normalization before the LLM stages

tests/test_validation.py

@@ -1,7 +1,7 @@
 import pytest
 
 from audita.errors import AuditaValidationError
-from audita.schemas import parse_glossary_yaml, parse_transcript_json
+from audita.schemas import parse_glossary_yaml, parse_source_transcript_json, parse_transcript_json
 
 
 def test_valid_transcript_parses():
@@ -102,6 +102,70 @@ def test_transcript_rejects_empty_input():
         parse_transcript_json("[]")
 
 
+def test_source_transcript_accepts_missing_ids():
+    segments = parse_source_transcript_json(
+        """
+        [
+          {"speaker": "Eric", "start": 0.0, "end": 1.0, "text": "Hi"}
+        ]
+        """
+    )
+
+    assert segments[0].id is None
+    assert segments[0].speaker == "Eric"
+
+
+def test_source_transcript_accepts_present_nonsequential_ids():
+    segments = parse_source_transcript_json(
+        """
+        [
+          {"id": 10, "speaker": "Eric", "start": 0.0, "end": 1.0, "text": "Hi"},
+          {"id": 4, "speaker": "Mike", "start": 1.0, "end": 2.0, "text": "There"}
+        ]
+        """
+    )
+
+    assert [segment.id for segment in segments] == [10, 4]
+
+
+def test_source_transcript_rejects_extra_fields():
+    with pytest.raises(AuditaValidationError):
+        parse_source_transcript_json(
+            """
+            [
+              {"speaker": "Eric", "start": 0.0, "end": 1.0, "text": "Hi", "extra": true}
+            ]
+            """
+        )
+
+
+def test_source_transcript_rejects_bad_timestamps():
+    with pytest.raises(AuditaValidationError):
+        parse_source_transcript_json(
+            """
+            [
+              {"speaker": "Eric", "start": 2.0, "end": 1.0, "text": "Hi"}
+            ]
+            """
+        )
+
+
+def test_source_transcript_rejects_empty_values():
+    with pytest.raises(AuditaValidationError):
+        parse_source_transcript_json(
+            """
+            [
+              {"speaker": "", "start": 0.0, "end": 1.0, "text": "Hi"}
+            ]
+            """
+        )
+
+
+def test_source_transcript_rejects_invalid_json():
+    with pytest.raises(AuditaValidationError):
+        parse_source_transcript_json("{")
+
+
 def test_valid_glossary_parses():
     glossary = parse_glossary_yaml(
         """