Update src/quiz_processing.py

src/quiz_processing.py

@@ -7,60 +7,7 @@ from typing import Dict, Any, List, Optional
 from transformers import AutoTokenizer
 from sentence_transformers import SentenceTransformer
 from huggingface_hub import login
-from src.prompts import SYSTEM_PROMPT
-
-# Define the prompt templates directly in this file since they're referenced but missing
-SUMMARY_PROMPT_TEMPLATE = """You are an expert content analyst specialized in creating professional, actionable summaries of educational content.
-
-Please analyze the following text to create a comprehensive yet concise summary that will be valuable to readers. Break down the content into 2-3 meaningful segments, each focused on a key topic or theme.
-
-For each segment of the content, provide:
-1. A descriptive topic name
-2. 3-5 key concepts or terms that are central to understanding this segment
-3. A concise summary paragraph (3-5 sentences) that captures the essential information
-
-The text to analyze is:
-{text}
-
-FORMAT YOUR RESPONSE STRICTLY AS A JSON OBJECT AS FOLLOWS (with no other text, explanation or formatting):
-{
-  "segments": [
-    {
-      "topic_name": "Title for the first segment",
-      "key_concepts": ["Key concept 1", "Key concept 2", "Key concept 3"],
-      "summary": "Concise summary paragraph for this segment that captures the essential information."
-    },
-    {
-      "topic_name": "Title for the second segment",
-      "key_concepts": ["Key concept 1", "Key concept 2", "Key concept 3"],
-      "summary": "Concise summary paragraph for this segment that captures the essential information."
-    }
-  ]
-}"""
-
-QUIZ_PROMPT_TEMPLATE = """You are an expert quiz creator specialized in creating educational assessments.
-
-Please analyze the following text and create 5 multiple-choice quiz questions that test understanding of the key concepts and information presented in the text. For each question:
-1. Write a clear, concise question
-2. Create 4 answer options (A, B, C, D) with exactly one correct answer
-
-The text to analyze is:
-{text}
-
-FORMAT YOUR RESPONSE STRICTLY AS A JSON OBJECT AS FOLLOWS (with no other text, explanation or formatting):
-{
-  "quiz_questions": [
-    {
-      "question": "The full text of the question?",
-      "options": [
-        { "text": "First option text", "correct": false },
-        { "text": "Second option text", "correct": true },
-        { "text": "Third option text", "correct": false },
-        { "text": "Fourth option text", "correct": false }
-      ]
-    }
-  ]
-}"""
+from src.prompts import SUMMARY_PROMPT_TEMPLATE, QUIZ_PROMPT_TEMPLATE
 
 GEMINI_MODEL = "gemini-2.0-flash"
 DEFAULT_TEMPERATURE = 0.7
@@ -132,60 +79,22 @@ def generate_with_gemini(text, api_key, language, content_type="summary"):
         
         try:
             content = response.content
-            # First try to find JSON within code blocks
-            json_match = re.search(r'```(?:json)?\s*([\s\S]*?)\s*```', content)
+            json_match = re.search(r'```json\s*([\s\S]*?)\s*```', content)
             
             if json_match:
                 json_str = json_match.group(1)
             else:
-                # Then try to find JSON with curly braces 
                 json_match = re.search(r'(\{[\s\S]*\})', content)
                 if json_match:
                     json_str = json_match.group(1)
                 else:
-                    # If we still don't have JSON, try to clean and parse the content directly
                     json_str = content
             
-            # Clean up the JSON string
-            json_str = json_str.strip()
-            
-            # Try to parse the JSON
-            try:
-                function_call = json.loads(json_str)
-                return function_call
-            except json.JSONDecodeError:
-                # If direct parsing fails, try to fix common issues
-                # Remove markdown formatting or extra text
-                cleaned_json = re.sub(r'^[^{]*', '', json_str)
-                cleaned_json = re.sub(r'[^}]*$', '', cleaned_json)
-                return json.loads(cleaned_json)
-                
-        except json.JSONDecodeError as e:
-            # Fall back to a default structure
-            if content_type == "summary":
-                return {
-                    "segments": [
-                        {
-                            "topic_name": "Content Analysis",
-                            "key_concepts": ["AI Processing", "Text Analysis"],
-                            "summary": "The model was unable to produce a properly formatted JSON response. Please try again with a different text sample."
-                        }
-                    ]
-                }
-            else:
-                return {
-                    "quiz_questions": [
-                        {
-                            "question": "Unable to generate quiz questions from the provided text.",
-                            "options": [
-                                {"text": "Try again", "correct": true},
-                                {"text": "Use different text", "correct": false},
-                                {"text": "Adjust the prompt", "correct": false},
-                                {"text": "Contact support", "correct": false}
-                            ]
-                        }
-                    ]
-                }
+            # Parse the JSON
+            function_call = json.loads(json_str)
+            return function_call
+        except json.JSONDecodeError:
+            raise Exception("Could not parse JSON from LLM response")
     except Exception as e:
         raise Exception(f"Error calling API: {str(e)}")
 
@@ -193,33 +102,79 @@ def format_summary_for_display(results, language="English"):
     output = []
     
     if language == "Uzbek":
-        segment_header = "QISM"
-        key_concepts_header = "ASOSIY TUSHUNCHALAR"
-        summary_header = "QISQACHA MAZMUN"
+        title_header = "SARLAVHA"
+        overview_header = "UMUMIY KO'RINISH"
+        key_points_header = "ASOSIY NUQTALAR"
+        key_entities_header = "ASOSIY SHAXSLAR VA TUSHUNCHALAR"
+        conclusion_header = "XULOSA"
     elif language == "Russian":
-        segment_header = "СЕГМЕНТ"
-        key_concepts_header = "КЛЮЧЕВЫЕ ПОНЯТИЯ"
-        summary_header = "КРАТКОЕ СОДЕРЖАНИЕ"
+        title_header = "ЗАГОЛОВОК"
+        overview_header = "ОБЗОР"
+        key_points_header = "КЛЮЧЕВЫЕ МОМЕНТЫ"
+        key_entities_header = "КЛЮЧЕВЫЕ ОБЪЕКТЫ"
+        conclusion_header = "ЗАКЛЮЧЕНИЕ"
     else:
-        segment_header = "SEGMENT"
-        key_concepts_header = "KEY CONCEPTS"
-        summary_header = "SUMMARY"
+        title_header = "TITLE"
+        overview_header = "OVERVIEW"
+        key_points_header = "KEY POINTS"
+        key_entities_header = "KEY ENTITIES"
+        conclusion_header = "CONCLUSION"
     
-    segments = results.get("segments", [])
-    if not segments:
-        return "No segments were generated. Please try again with a different text sample."
-        
-    for i, segment in enumerate(segments):
-        topic = segment["topic_name"]
-        segment_num = i + 1 
+    if "summary" not in results:
+        if "segments" in results:
+            segments = results.get("segments", [])
+            for i, segment in enumerate(segments):
+                topic = segment.get("topic_name", f"Section {i+1}")
+                segment_num = i + 1
+                output.append(f"\n\n{'='*40}")
+                output.append(f"SEGMENT {segment_num}: {topic}")
+                output.append(f"{'='*40}\n")
+                
+                if "key_concepts" in segment:
+                    output.append("KEY CONCEPTS:")
+                    for concept in segment["key_concepts"]:
+                        output.append(f"• {concept}")
+                
+                if "summary" in segment:
+                    output.append("\nSUMMARY:")
+                    output.append(segment["summary"])
+            
+            return "\n".join(output)
+        else:
+            return "Error: Could not parse summary results. Invalid format received."
+    
+    summary = results["summary"]
+    if "title" in summary:
         output.append(f"\n\n{'='*40}")
-        output.append(f"{segment_header} {segment_num}: {topic}")
+        output.append(f"{title_header}: {summary['title']}")
         output.append(f"{'='*40}\n")
-        output.append(f"{key_concepts_header}:")
-        for concept in segment["key_concepts"]:
-            output.append(f"• {concept}")
-        output.append(f"\n{summary_header}:")
-        output.append(segment["summary"])
+    
+    # Overview
+    if "overview" in summary:
+        output.append(f"{overview_header}:")
+        output.append(f"{summary['overview']}\n")
+    
+    # Key Points
+    if "key_points" in summary and summary["key_points"]:
+        output.append(f"{key_points_header}:")
+        for theme_group in summary["key_points"]:
+            if "theme" in theme_group:
+                output.append(f"\n{theme_group['theme']}:")
+            if "points" in theme_group:
+                for point in theme_group["points"]:
+                    output.append(f"• {point}")
+    
+    # Key Entities
+    if "key_entities" in summary and summary["key_entities"]:
+        output.append(f"\n{key_entities_header}:")
+        for entity in summary["key_entities"]:
+            if "name" in entity and "description" in entity:
+                output.append(f"• **{entity['name']}**: {entity['description']}")
+    
+    # Conclusion
+    if "conclusion" in summary:
+        output.append(f"\n{conclusion_header}:")
+        output.append(summary["conclusion"])
     
     return "\n".join(output)
 
@@ -238,9 +193,6 @@ def format_quiz_for_display(results, language="English"):
     output.append(f"{'='*40}\n")
     
     quiz_questions = results.get("quiz_questions", [])
-    if not quiz_questions:
-        return "No quiz questions were generated. Please try again with a different text sample."
-        
     for i, q in enumerate(quiz_questions):
         output.append(f"\n{i+1}. {q['question']}")
         for j, option in enumerate(q['options']):
@@ -252,9 +204,6 @@ def format_quiz_for_display(results, language="English"):
 
 def analyze_document(text, gemini_api_key, language, content_type="summary"):
     try:
-        if not text or len(text.strip()) < 100:
-            return "Error: Text is too short to analyze. Please provide a longer text sample.", None, None
-            
         start_time = time.time()
         text_parts = split_text_by_tokens(text)
 
@@ -262,8 +211,7 @@ def analyze_document(text, gemini_api_key, language, content_type="summary"):
         output_tokens = 0
         
         if content_type == "summary":
-            all_results = {"segments": []}
-            segment_counter = 1
+            all_results = {}
             
             for part in text_parts:
                 actual_prompt = SUMMARY_PROMPT_TEMPLATE.format(text=part)
@@ -272,24 +220,17 @@ def analyze_document(text, gemini_api_key, language, content_type="summary"):
                 
                 analysis = generate_with_gemini(part, gemini_api_key, language, "summary")
                 
-                if "segments" in analysis and analysis["segments"]:
-                    for segment in analysis["segments"]:
-                        segment["segment_number"] = segment_counter
-                        all_results["segments"].append(segment)
-                        segment_counter += 1
-                else:
-                    # Add a default segment if none were returned
-                    all_results["segments"].append({
-                        "segment_number": segment_counter,
-                        "topic_name": "Content Analysis",
-                        "key_concepts": ["Text Processing", "AI Analysis", "Document Summarization"],
-                        "summary": "The system was unable to generate detailed segments from this text portion. This may be due to the complexity of the content or formatting issues. Consider breaking the text into smaller, more focused sections."
-                    })
-                    segment_counter += 1
+                if not all_results and "summary" in analysis:
+                    all_results = analysis
+                elif "summary" in analysis:
+                    if "key_points" in analysis["summary"] and "key_points" in all_results["summary"]:
+                        all_results["summary"]["key_points"].extend(analysis["summary"]["key_points"])
+                    if "key_entities" in analysis["summary"] and "key_entities" in all_results["summary"]:
+                        all_results["summary"]["key_entities"].extend(analysis["summary"]["key_entities"])
             
             formatted_output = format_summary_for_display(all_results, language)
-            
-        else:  # Quiz generation
+                
+        else:  
             all_results = {"quiz_questions": []}
             
             for part in text_parts:
@@ -299,23 +240,11 @@ def analyze_document(text, gemini_api_key, language, content_type="summary"):
                 
                 analysis = generate_with_gemini(part, gemini_api_key, language, "quiz")
                 
-                if "quiz_questions" in analysis and analysis["quiz_questions"]:
+                if "quiz_questions" in analysis:
                     remaining_slots = 10 - len(all_results["quiz_questions"])
                     if remaining_slots > 0:
                         questions_to_add = analysis["quiz_questions"][:remaining_slots]
                         all_results["quiz_questions"].extend(questions_to_add)
-                else:
-                    # Add a default question if none were returned
-                    if len(all_results["quiz_questions"]) < 10:
-                        all_results["quiz_questions"].append({
-                            "question": "What is the main purpose of text analysis in educational contexts?",
-                            "options": [
-                                {"text": "To change the original meaning of the text", "correct": False},
-                                {"text": "To extract key concepts and facilitate understanding", "correct": True},
-                                {"text": "To reduce text to exactly half its original length", "correct": False},
-                                {"text": "To eliminate all technical terminology", "correct": False}
-                            ]
-                        })
             
             formatted_output = format_quiz_for_display(all_results, language)