Working TOC but indexing incorrect

app/DocumentGeneration/__init__.py

@@ -12,7 +12,7 @@ if current_dir not in sys.path:
 from pdf_generator import create_pdf_document
 from word_generator import create_word_document
 from chart_generator import parse_embedded_chart_data, create_chart_from_data
-from table_of_content import generate_table_of_contents, add_table_of_contents_page
+from table_of_content import generate_table_of_contents_after_content, add_table_of_contents_page
 from document_helpers import (
     extract_business_name_from_content,
     filter_empty_sections,
@@ -38,7 +38,7 @@ __all__ = [
     'create_word_document',
     'parse_embedded_chart_data',
     'create_chart_from_data',
-    'generate_table_of_contents',
+    'generate_table_of_contents_after_content',
     'add_table_of_contents_page',
     'extract_business_name_from_content',
     'filter_empty_sections',

app/DocumentGeneration/pdf_generator.py

@@ -18,7 +18,7 @@ from document_helpers import (
     extract_business_name_from_content,
     remove_duplicate_headings
 )
-from table_of_content import generate_table_of_contents, add_table_of_contents_page
+from table_of_content import generate_table_of_contents_after_content, add_table_of_contents_page
 from chart_generator import parse_embedded_chart_data, create_chart_from_data
 from markdown_renderer import (
     clean_text_for_pdf,
@@ -34,6 +34,189 @@ from CurrencyMapping.currency_mapping import convert_currency_symbols_to_iso
 # Import sort_sections_by_order from utils module
 from utils import sort_sections_by_order
 
+def calculate_real_page_numbers(request: ExportRequest, filtered_sections):
+    """Calculate real page numbers by creating a temporary PDF with all content INCLUDING charts"""
+    try:
+        from fpdf import FPDF
+    except ImportError:
+        raise Exception("fpdf2 is not installed. Please install it with: pip install fpdf2")
+    
+    # Create temporary PDF to calculate page numbers
+    temp_pdf = FPDF()
+    temp_pdf.set_auto_page_break(auto=True, margin=15)
+    
+    # Add cover page
+    temp_pdf.add_page()
+    
+    # Add TOC page (placeholder)
+    temp_pdf.add_page()
+    
+    # Parse charts from embedded chart data in section content
+    embedded_charts = parse_embedded_chart_data(filtered_sections)
+    
+    # Create charts and map them to sections
+    chart_info_map = {}
+    if embedded_charts:
+        for chart_info in embedded_charts:
+            chart_bytes = create_chart_from_data(chart_info, request.businessIdea or "Business")
+            if chart_bytes:
+                source_section = chart_info["source_section"]
+                if source_section not in chart_info_map:
+                    chart_info_map[source_section] = []
+                chart_info_map[source_section].append({
+                    "chart_bytes": chart_bytes,
+                    "title": chart_info["title"],
+                    "type": chart_info["type"]
+                })
+    
+    # Add all sections to calculate real page numbers INCLUDING charts
+    section_page_map = {}
+    sorted_sections = sort_sections_by_order(filtered_sections)
+    
+    for section in sorted_sections:
+        temp_pdf.add_page()
+        current_page = temp_pdf.page_no()
+        section_page_map[section.title] = current_page
+        
+        # Add section title
+        temp_pdf.set_font("Arial", "B", 20)
+        temp_pdf.set_text_color(0, 0, 0)
+        temp_pdf.set_xy(25, 25)
+        temp_pdf.cell(0, 18, section.title, ln=True)
+        
+        # Add subtle line under section title
+        text_width = temp_pdf.get_string_width(section.title)
+        line_width = min(text_width + 8, 160)
+        temp_pdf.set_draw_color(221, 221, 221)
+        temp_pdf.set_line_width(0.3)
+        temp_pdf.line(25, temp_pdf.get_y(), 25 + line_width, temp_pdf.get_y())
+        temp_pdf.ln(8)
+        
+        # Add section content
+        temp_pdf.set_font("Arial", "", 12)
+        temp_pdf.set_text_color(51, 51, 51)
+        temp_pdf.set_xy(25, temp_pdf.get_y())
+        
+        # Process content
+        try:
+            content_text = clean_text_for_pdf(section.content)
+        except Exception as e:
+            content_text = section.content or ""
+        
+        # Remove chart data markers
+        content_text = re.sub(r'<!--CHARTDATASTART-->.*?<!--CHARTDATAEND-->', '', content_text, flags=re.DOTALL)
+        content_text = re.sub(r'<!--CHARTDATASTART-->.*$', '', content_text, flags=re.DOTALL)
+        
+        # Remove the section title from content to prevent duplication
+        section_title_clean = section.title.strip()
+        content_lines = content_text.split('\n')
+        filtered_lines = []
+        for line in content_lines:
+            line_clean = line.strip()
+            if (line_clean == section_title_clean or 
+                line_clean == f"**{section_title_clean}**" or
+                line_clean == f"# {section_title_clean}" or
+                line_clean == f"## {section_title_clean}" or
+                line_clean == f"### {section_title_clean}" or
+                line_clean == f"#### {section_title_clean}" or
+                line_clean.startswith(f"{section_title_clean}") and len(line_clean) <= len(section_title_clean) + 10):
+                continue
+            filtered_lines.append(line)
+        content_text = '\n'.join(filtered_lines)
+        
+        # Add content with proper markdown rendering
+        if content_text and content_text.strip():
+            render_markdown_to_pdf_grouped(temp_pdf, content_text, (0,0,0), (75,0,130), (51,51,51))
+        
+        # Add charts for this section (same logic as in actual PDF generation)
+        section_charts = chart_info_map.get(section.title, [])
+        
+        # Try different matching strategies for charts
+        if not section_charts:
+            # Case-insensitive match
+            for chart_source, charts in chart_info_map.items():
+                if chart_source.lower() == section.title.lower():
+                    section_charts = charts
+                    break
+            
+            # Partial match
+            if not section_charts:
+                for chart_source, charts in chart_info_map.items():
+                    if (section.title.lower() in chart_source.lower() or 
+                        chart_source.lower() in section.title.lower()):
+                        section_charts = charts
+                        break
+            
+            # Key-based match
+            if not section_charts and hasattr(section, 'key'):
+                for chart_source, charts in chart_info_map.items():
+                    if section.key and section.key.replace('prompt_', '') in chart_source.lower():
+                        section_charts = charts
+                        break
+        
+        # Add charts with same page break logic as actual PDF
+        if section_charts:
+            charts_per_page = 0
+            max_charts_per_page = 5
+            
+            for chart_info in section_charts:
+                try:
+                    chart_height_needed = 95  # Chart height (75) + title (20) + minimal spacing
+                    current_y = temp_pdf.get_y()
+                    charts_per_page += 1
+                    
+                    # Only force page break if absolutely necessary
+                    if (current_y + chart_height_needed > 290) or (charts_per_page > 5):
+                        temp_pdf.add_page()
+                        y_title = 25
+                        charts_per_page = 1
+                    else:
+                        y_title = current_y
+                    
+                    # Add minimal spacing before chart if not at top of page
+                    if y_title > 25:
+                        temp_pdf.ln(3)
+                    
+                    # Add chart title
+                    temp_pdf.set_font("Arial", "B", 16)
+                    temp_pdf.set_text_color(0, 0, 0)
+                    temp_pdf.set_xy(25, y_title)
+                    temp_pdf.cell(0, 14, chart_info["title"], ln=True)
+                    
+                    # Add subtle line under chart title
+                    text_width = temp_pdf.get_string_width(chart_info["title"])
+                    line_width = min(text_width + 8, 160)
+                    temp_pdf.set_draw_color(221, 221, 221)
+                    temp_pdf.set_line_width(0.3)
+                    temp_pdf.line(25, y_title + 14, 25 + line_width, y_title + 14)
+                    
+                    # Add chart image
+                    img_width = 100
+                    x = (210 - img_width) / 2
+                    y_img = y_title + 20
+                    
+                    # Save image to temp file
+                    with tempfile.NamedTemporaryFile(delete=False, suffix=".png") as tmp_img:
+                        tmp_img.write(chart_info["chart_bytes"])
+                        temp_img_path = tmp_img.name
+                    
+                    # Add chart
+                    chart_height = 75
+                    temp_pdf.image(temp_img_path, x=x, y=y_img, w=img_width)
+                    temp_pdf.ln(chart_height + 5)
+                    
+                    # Clean up temp file
+                    try:
+                        os.remove(temp_img_path)
+                    except Exception:
+                        pass
+                        
+                except Exception:
+                    # Skip faulty chart and continue
+                    continue
+    
+    return section_page_map
+
 def create_pdf_document(request: ExportRequest):
     """Create a PROFESSIONAL PDF document from the business plan data using FPDF2
     NOTE: Now uses matplotlib charts exclusively - old base64 chart logic commented out"""
@@ -50,15 +233,13 @@ def create_pdf_document(request: ExportRequest):
     if not filtered_sections:
         raise Exception("No valid sections found after filtering. All sections appear to be empty.")
     
-    # Generate table of contents
-    try:
-        toc_items = generate_table_of_contents(filtered_sections, request.businessIdea)
-    except Exception as e:
-        toc_items = []
-        # Continue with a fallback; TOC page builder already has fallback handling
-    # logging.info(f"Generated TOC with {len(toc_items)} items: {[item['title'] for item in toc_items[:3]]}...")
+    # First pass: Calculate real page numbers by creating a temporary PDF
+    section_page_map = calculate_real_page_numbers(request, filtered_sections)
+    
+    # Generate TOC with real page numbers
+    toc_items = generate_table_of_contents_after_content(filtered_sections, section_page_map, request.businessIdea)
     
-    # Create PDF with professional settings
+    # Second pass: Create final PDF with TOC on page 2
     pdf = FPDF()
     pdf.set_auto_page_break(auto=True, margin=15)  # Reduced margin to allow more charts per page
     
@@ -447,8 +628,8 @@ def create_pdf_document(request: ExportRequest):
         except Exception as e:
             raise Exception(f"Cover page failed: {str(e)}")
             
+        # Add TOC on page 2 with REAL page numbers (already calculated above)
         try:
-            # Ensure TOC items are generated before adding the page
             if toc_items:
                 add_table_of_contents_page(pdf, toc_items, request.businessIdea)
             else:

app/DocumentGeneration/table_of_content.py

@@ -1,5 +1,5 @@
 # Table of Contents Module
-# Contains generate_table_of_contents() and add_table_of_contents_page() functions
+# Contains functions to generate TOC AFTER content is placed with real page numbers
 
 import re
 import os
@@ -11,8 +11,9 @@ sys.path.append(os.path.dirname(os.path.dirname(__file__)))
 from models import ExportSection
 from constants import SECTION_ORDER
 
-def generate_table_of_contents(sections: List[ExportSection], business_idea: str) -> List[Dict[str, Any]]:
-    """Generate a structured table of contents with page numbers following SECTION_ORDER"""
+def generate_table_of_contents_after_content(sections: List[ExportSection], section_page_map: Dict[str, int], business_idea: str) -> List[Dict[str, Any]]:
+    """Generate TOC AFTER all content is placed, using REAL page numbers from section_page_map
+    This ensures accurate page numbers based on actual content placement"""
     toc_items = []
     
     # Create a mapping from section titles to section objects for easy lookup
@@ -50,23 +51,27 @@ def generate_table_of_contents(sections: List[ExportSection], business_idea: str
                         section = section_obj
                         break
         
-        # Always add to TOC, even if section doesn't exist or has no content
-        # Each section starts on a new page, so page number is 3 + i (cover=1, TOC=2, sections start at 3)
-        page_number = 3 + i
-        toc_items.append({
-            "title": expected_title,  # Use the expected title for consistency
-            "level": 1,
-            "page": page_number,
-            "section_type": "section",
-            "has_content": bool(section and section.content and section.content.strip())
-        })
+        # Only add sections that have actual content AND real page numbers - NO DUMMY DATA
+        has_content = bool(section and section.content and section.content.strip())
+        if has_content:
+            # Get the REAL page number from section_page_map
+            real_page = section_page_map.get(section.title, 0)
+            if real_page > 0:  # Only add if we have a real page number
+                toc_items.append({
+                    "title": expected_title,  # Use the expected title for consistency
+                    "level": 1,
+                    "page": real_page,  # REAL page number from actual content placement
+                    "section_type": "section",
+                    "has_content": True,
+                    "section_obj": section  # Store reference to actual section
+                })
     
     return toc_items
-    
+
 def add_table_of_contents_page(pdf, toc_items: List[Dict[str, Any]], business_idea: str):
-    """Add a professional table of contents page following SECTION_ORDER"""
+    """Add a professional table of contents page with REAL page numbers"""
     
-    # TOC items are already in the correct order from generate_table_of_contents
+    # TOC items are already in the correct order and have real page numbers
     sorted_toc_items = toc_items
     
     # Professional color scheme - Refined for bank/investor presentation
@@ -141,7 +146,7 @@ def add_table_of_contents_page(pdf, toc_items: List[Dict[str, Any]], business_id
             pdf.set_xy(x, y_position + 3)
             pdf.cell(1, 1, ".", ln=False)
         
-        # Page number with hyperlink
+        # Page number with hyperlink - NOW USING REAL PAGE NUMBERS
         page_number_width = pdf.get_string_width(str(item["page"]))
         pdf.set_xy(160, y_position)
         try:

app/DocumentGeneration/word_generator.py

@@ -18,7 +18,7 @@ from document_helpers import (
     filter_empty_sections,
     extract_business_name_from_content
 )
-from table_of_content import generate_table_of_contents
+# from table_of_content import generate_table_of_contents  # Not used in word generator
 from chart_generator import parse_embedded_chart_data, create_chart_from_data
 from markdown_renderer import (
     render_markdown_to_docx_grouped,

app/main.py

@@ -118,7 +118,7 @@ from DocumentGeneration.pdf_generator import create_pdf_document
 from DocumentGeneration.word_generator import create_word_document
 from DocumentGeneration.live_pdf_generator import live_pdf_generator
 from DocumentGeneration.chart_generator import parse_embedded_chart_data, create_chart_from_data
-from DocumentGeneration.table_of_content import generate_table_of_contents, add_table_of_contents_page
+from DocumentGeneration.table_of_content import generate_table_of_contents_after_content, add_table_of_contents_page
 from DocumentGeneration.document_helpers import (
     extract_business_name_from_content,
     filter_empty_sections,
@@ -1674,7 +1674,7 @@ def root():
 # @app.get("/test-toc")
 # def test_toc():
 #     """Test endpoint to verify TOC generation"""
-#     from app.main import generate_table_of_contents, ExportSection
+#     from app.main import generate_table_of_contents_after_content, ExportSection
     
 #     # Create test sections
 #     test_sections = [
@@ -1696,7 +1696,7 @@ def root():
 #     ]
     
 #     # Generate TOC
-#     toc_items = generate_table_of_contents(test_sections, "Test Business")
+#     toc_items = generate_table_of_contents_after_content(test_sections, {}, "Test Business")
     
 #     return {
 #         "test_sections": [{"title": s.title, "key": s.key, "content_length": len(s.content)} for s in test_sections],