Create output_parser.py

output_parser.py

@@ -0,0 +1,54 @@
+import re
+from collections import defaultdict
+
+def parse_srt_output(srt_content):
+    speakers = defaultdict(lambda: {"utterances": [], "total_duration": 0})
+    current_speaker = None
+    current_utterance = {"start": "", "end": "", "text": ""}
+
+    for line in srt_content.split('\n'):
+        if line.startswith("Speaker"):
+            if ":" in line and "total duration" in line:
+                # This is the summary line at the end
+                speaker, duration = line.split(": total duration ")
+                speakers[speaker]["total_duration"] = duration.strip()
+            else:
+                current_speaker = line.strip()
+        elif line.startswith(" time:"):
+            time_match = re.search(r'\((.+?) --> (.+?)\)', line)
+            if time_match:
+                current_utterance["start"] = time_match.group(1)
+                current_utterance["end"] = time_match.group(2)
+        elif line.startswith(" text:"):
+            current_utterance["text"] = line.replace(" text:", "").strip()
+            if current_speaker:
+                speakers[current_speaker]["utterances"].append(current_utterance.copy())
+            current_utterance = {"start": "", "end": "", "text": ""}
+
+    # Print the parsed output for debugging
+    print("Parsed SRT Output:")
+    for speaker, data in speakers.items():
+        print(f"{speaker}:")
+        print(f"  Total duration: {data['total_duration']}")
+        print(f"  Utterances:")
+        for utterance in data['utterances'][:3]:  # Print first 3 utterances for brevity
+            print(f"    {utterance['start']} - {utterance['end']}: {utterance['text']}")
+        print(f"  Total utterances: {len(data['utterances'])}")
+        print()
+
+    return speakers
+
+def get_speaker_data_for_charts(parsed_output):
+    speaker_data = {}
+    for speaker, data in parsed_output.items():
+        speaker_data[speaker] = {
+            "total_duration": data["total_duration"],
+            "utterance_count": len(data["utterances"]),
+            "average_utterance_length": sum(len(u["text"].split()) for u in data["utterances"]) / len(data["utterances"]) if data["utterances"] else 0
+        }
+    
+    # Print the data for charts
+    print("Data for Charts:")
+    print(speaker_data)
+    
+    return speaker_data