Update app.py

app.py

@@ -79,7 +79,7 @@ REQUIRED_FIELDS = [
     "exam_type", "content_type", "exam_section", "domain", "subdomain",
     "topic", "difficulty_level", "reading_passage", "question_text",
     "option_a", "option_b", "option_c", "option_d", "correct_answer",
-    "explanation", "is_active"
+    "explanation", "is_active", "source_text"
 ]
 
 class ExamQuestion(BaseModel):
@@ -99,6 +99,7 @@ class ExamQuestion(BaseModel):
     option_d: str
     correct_answer: str
     explanation: str
+    source_text: str  # The original text from which the question was generated
     is_active: bool = True
 
 class ExamQuestionResponse(BaseModel):
@@ -470,6 +471,7 @@ ABSOLUTELY CRITICAL:
 8. The response should be PURE JSON - nothing else
 9. SKIP ANY QUESTIONS that refer to images, diagrams, graphs, figures, or visual elements
 10. If a question mentions "look at the image", "in the picture", "as shown in", etc., DO NOT include it
+11. For each question, include the specific source text that the question is based on
 
 Text to process:
 {text}
@@ -495,6 +497,7 @@ Format EVERY question using this exact JSON structure:
     "option_d": "exact_option_d_from_text",
     "correct_answer": "one_of[A,B,C,D]_determined_from_context",
     "explanation": "explanation_from_text_or_generate_based_on_answer",
+    "source_text": "exact_text_snippet_that_this_question_is_based_on",
     "is_active": true
 }}"""
 
@@ -562,7 +565,8 @@ Format EVERY question using this exact JSON structure:
                 "difficulty_level": "Medium",
                 "reading_passage_title": None,
                 "is_active": True,
-                "source_file": source_file
+                "source_file": source_file,
+                "source_text": text  # Add default source text
             }
             
             for q_idx, q in enumerate(questions, 1):
@@ -581,7 +585,8 @@ Format EVERY question using this exact JSON structure:
                     "option_c",
                     "option_d",
                     "correct_answer",
-                    "explanation"
+                    "explanation",
+                    "source_text"  # Add source_text as a critical field
                 ]
                 
                 # Validate critical fields
@@ -1387,6 +1392,16 @@ def display_question(question, index):
             with col_e:
                 st.markdown(f"**Source:** {question.get('source_file', 'N/A')}")
         
+        # Source text if available
+        if question.get('source_text'):
+            st.markdown("### 📄 Source Text")
+            st.markdown(
+                f"""<div style='background-color: #e8f4f9; padding: 20px; border-radius: 10px; margin: 10px 0; color: #1f1f1f;'>
+                {question['source_text']}
+                </div>""",
+                unsafe_allow_html=True
+            )
+
         # Reading passage if available
         if question.get('reading_passage'):
             st.markdown("### 📖 Reading Passage")