Update app.py

app.py

@@ -20,53 +20,48 @@ genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
 GRID_ROWS, GRID_COLS = 20, 14
 
 # ---------------- PROMPTS ----------------
-PROMPTS = {
-    # Updated QP+MS transcription prompt:
-    "QP_MS_TRANSCRIPTION": {
-        "role": "system",
-        "content": """You are a high-quality OCR/Transcription assistant.
-
-INPUT: This file is a scanned/printed PDF that first contains the Question Paper and then, after all questions, the Markscheme.
-TASK: Produce an exact transcription in plain text with clear separators.
-
-IMPORTANT: Output **ALL QUESTIONS FIRST** (in the same order they appear in the PDF).
-For each question, output:
-- Question ID (exact as printed, e.g., "1", "2(a)", "3.b", "4(ii)")
-- Question text (exact wording; do not change punctuation)
-- Total marks for that question (exact number if printed; if not printed leave blank)
-
-After you have outputted **all questions** (and their total marks), output the **entire markscheme block** exactly as it appears in the PDF. In the markscheme section, ensure notation is explicit and clear: represent M, A, R notation **in brackets** after each mark item where applicable. For example:
-[M1] Description...
-[A1] Description...
-[R1] Description...
-
-Also include at the top a single line stating the total marks of the paper (if present in the paper).
-
-KEY REQUIREMENTS:
-- Do NOT interleave question and markscheme. First: questions + totals. Second: markscheme (verbatim, preserving mark IDs/formatting).
-- Transcribe the markscheme verbatim; do NOT correct or reformat content (only ensure M/A/R are shown in brackets if present).
-- Represent M, A, R marks explicitly and consistently (e.g., M1, A2, R1). If mark IDs are missing, transcribe as-is.
-- Ignore any N1, N2, N3 notations (do not use them).
-
-OUTPUT FORMAT (use these exact markers to make parsing straightforward):
-==== PAPER TOTAL MARKS ====
-<integer or blank>
-
-==== QUESTIONS BEGIN ====
-Question: <id>
-Total Marks: <integer or blank>
-QP:
-<question text (multiline)>
---QUESTION-END--
-
-(repeat the Question block for all questions, in order)
-==== QUESTIONS END ====
-
-==== MARKSCHEME BEGIN ====
-<verbatim markscheme text exactly as in PDF; include mark IDs and use brackets for M/A/R notations where they appear>
-==== MARKSCHEME END ====
+PROMPTS["QP_MS_TRANSCRIPTION"] = {
+    "role": "system",
+    "content": """You are a high-quality OCR/Transcription assistant.
+
+INPUT: This file is a PDF that first contains the Question Paper and immediately after it the Markscheme.
+
+TASK: 
+1. Transcribe EXACTLY all the questions FIRST (with their total marks).  
+2. After ALL questions, transcribe the Markscheme exactly, preserving M/A/R notation in brackets.  
+3. Always number the questions sequentially (Question 1, Question 2, Question 3, …) **in the order they appear in the PDF**, even if the PDF shows a different number or leaves it blank. Do NOT skip or leave Question: blank.  
+
+FORMAT:  
+==== PAPER TOTAL MARKS ====  
+<total marks>  
+
+==== QUESTIONS BEGIN ====  
+Question 1  
+Total Marks: <number>  
+QP: <question text>  
+--QUESTION-END--  
+
+Question 2  
+Total Marks: <number>  
+QP: <question text>  
+--QUESTION-END--  
+
+(repeat for all questions in order of appearance)  
+
+==== QUESTIONS END ====  
+
+==== MARKSCHEME BEGIN ====  
+Answer 1:  
+<exact MS for Q1 with notations M1, A1, R1 etc>  
+
+Answer 2:  
+<exact MS for Q2 with notations>  
+
+(repeat for all answers)  
+==== MARKSCHEME END ====  
 """
-    },
+}
+,
 
     # GRADING_PROMPT unchanged except we will print steps around calling it
     "GRADING_PROMPT": {