Update process_interview.py

process_interview.py

@@ -505,153 +505,161 @@ def generate_report(analysis_data: Dict) -> str:
 
 
 # --- NEW, ENHANCED PDF GENERATION FUNCTION ---
+# --- Make sure these imports are at the top of your process_interview.py file ---
+from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle, PageBreak, Image
+from reportlab.lib.pagesizes import letter
+from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
+from reportlab.lib.units import inch
+from reportlab.lib import colors
+import time
+import re
+import io
+
+# --- New, Enhanced PDF Generation Function ---
 def create_pdf_report(analysis_data: Dict, output_path: str, gemini_report_text: str):
     try:
         doc = SimpleDocTemplate(output_path, pagesize=letter,
-                                rightMargin=inch/2, leftMargin=inch/2,
-                                topMargin=inch, bottomMargin=inch/2)
+                                rightMargin=0.75*inch, leftMargin=0.75*inch,
+                                topMargin=1*inch, bottomMargin=1*inch)
 
         styles = getSampleStyleSheet()
-        styles.add(ParagraphStyle(name='Justify', alignment=4)) # TA_JUSTIFY = 4
-        h1 = ParagraphStyle(name='Heading1', fontSize=18, leading=22, spaceAfter=20, alignment=1, textColor=colors.HexColor('#003366'))
-        h2 = ParagraphStyle(name='Heading2', fontSize=14, leading=18, spaceBefore=12, spaceAfter=10, textColor=colors.HexColor('#336699'))
-        body_text = ParagraphStyle(name='BodyText', parent=styles['Normal'], spaceAfter=6)
+        h1 = ParagraphStyle(name='Heading1', fontSize=20, leading=24, spaceAfter=18, alignment=1, textColor=colors.HexColor('#00205B'))
+        h2 = ParagraphStyle(name='Heading2', fontSize=14, leading=18, spaceBefore=12, spaceAfter=6, textColor=colors.HexColor('#003366'))
+        body_text = ParagraphStyle(name='BodyText', parent=styles['Normal'], fontSize=10, leading=14, spaceAfter=6)
         bullet_style = ParagraphStyle(name='Bullet', parent=body_text, leftIndent=18, bulletIndent=9)
-
+        
         story = []
 
-        # --- Header and Footer ---
+        # --- Header and Footer Logic ---
         def header_footer(canvas, doc):
             canvas.saveState()
             # Footer
             canvas.setFont('Helvetica', 9)
-            canvas.drawString(inch, 0.5 * inch, f"Page {doc.page} | EvalBot Analysis")
+            canvas.setFillColor(colors.grey)
+            canvas.drawString(doc.leftMargin, 0.5 * inch, f"Page {doc.page} | EvalBot Confidential Report")
             # Header line
-            canvas.setStrokeColorRGB(0, 0.2, 0.4)
+            canvas.setStrokeColor(colors.HexColor('#003366'))
             canvas.setLineWidth(1)
-            canvas.line(doc.leftMargin, doc.height + doc.topMargin - inch/2, doc.width + doc.leftMargin, doc.height + doc.topMargin - inch/2)
+            canvas.line(doc.leftMargin, doc.height + 0.75*inch, doc.width + doc.leftMargin, doc.height + 0.75*inch)
             canvas.restoreState()
 
-        # --- First Page: Summary ---
-        story.append(Paragraph("EvalBot Interview Analysis Report", h1))
-        story.append(Spacer(1, 0.1 * inch))
-        story.append(Paragraph(f"Analysis Date: {time.strftime('%Y-%m-%d')}", styles['Normal']))
-        story.append(Spacer(1, 0.3 * inch))
+        # --- First Page: Title and Summary ---
+        story.append(Paragraph("Interview Performance Analysis", h1))
+        story.append(Paragraph(f"Analysis Date: {time.strftime('%Y-%m-%d')}", ParagraphStyle(name='Date', alignment=1, fontSize=9, textColor=colors.grey)))
+        story.append(Spacer(1, 0.4 * inch))
 
         acceptance_prob = analysis_data.get('acceptance_probability')
         if acceptance_prob is not None:
             story.append(Paragraph("Candidate Evaluation Summary", h2))
-            prob_color = colors.green if acceptance_prob >= 70 else (colors.orange if acceptance_prob >= 40 else colors.red)
-            story.append(Paragraph(f"<font size=14>Estimated Acceptance Probability: <b><font color='{prob_color.hexval()}'>{acceptance_prob:.2f}%</font></b></font>",
+            prob_color = colors.green if acceptance_prob >= 70 else (colors.darkorange if acceptance_prob >= 40 else colors.red)
+            story.append(Paragraph(f"Estimated Acceptance Probability: <font size=14 color='{prob_color.hexval()}'><b>{acceptance_prob:.2f}%</b></font>",
                                  ParagraphStyle(name='Prob', fontSize=12, spaceAfter=10)))
             if acceptance_prob >= 80:
-                story.append(Paragraph("This indicates a very strong candidate with high potential.", body_text))
+                story.append(Paragraph("<b>Overall Assessment:</b> This indicates a very strong candidate with high potential. Recommended for the next round.", body_text))
             elif acceptance_prob >= 50:
-                story.append(Paragraph("This candidate shows solid potential with areas for improvement.", body_text))
+                story.append(Paragraph("<b>Overall Assessment:</b> This candidate shows solid potential but has key areas for improvement.", body_text))
             else:
-                story.append(Paragraph("This candidate may require significant development or may not be an ideal fit.", body_text))
+                story.append(Paragraph("<b>Overall Assessment:</b> This candidate may require significant development or may not be the ideal fit at this time.", body_text))
         
         story.append(PageBreak())
 
-        # --- Second Page and beyond: Detailed Analysis ---
-        story.append(Paragraph("1. Detailed Voice Analysis", h2))
+        # --- Second Page: Detailed Analysis ---
+        story.append(Paragraph("Detailed Analysis", h1))
+        
+        story.append(Paragraph("1. Voice & Speech Metrics", h2))
         voice_analysis = analysis_data.get('voice_analysis', {})
         if voice_analysis and 'error' not in voice_analysis:
-            # Table for voice metrics
+            # --- This is the corrected table ---
             table_data = [
                 ['Metric', 'Value', 'Interpretation'],
-                ['Speaking Rate', f"{voice_analysis.get('speaking_rate', 0):.2f} words/sec", 'Average rate'],
-                ['Filler Words', f"{voice_analysis.get('filler_ratio', 0) * 100:.1f}%", '% of total words'],
+                ['Speaking Rate', f"{voice_analysis.get('speaking_rate', 0):.2f} words/sec", 'Indicator of pace and confidence.'],
+                ['Filler Words Ratio', f"{voice_analysis.get('filler_ratio', 0) * 100:.1f}%", 'Measures use of "um", "uh", etc.'],
                 ['Anxiety Level', voice_analysis.get('interpretation', {}).get('anxiety_level', 'N/A').upper(), f"Score: {voice_analysis.get('composite_scores', {}).get('anxiety', 0):.3f}"],
                 ['Confidence Level', voice_analysis.get('interpretation', {}).get('confidence_level', 'N/A').upper(), f"Score: {voice_analysis.get('composite_scores', {}).get('confidence', 0):.3f}"],
-                ['Fluency', voice_analysis.get('interpretation', {}).get('fluency_level', 'N/A').upper(), 'Overall speech flow']
+                ['Fluency Level', voice_analysis.get('interpretation', {}).get('fluency_level', 'N/A').upper(), 'Overall speech flow and coherence.']
             ]
-            table = Table(table_data, colWidths=[1.5*inch, 1.5*inch, 3*inch])
+            table = Table(table_data, colWidths=[1.6*inch, 1.2*inch, 3.7*inch])
             table.setStyle(TableStyle([
-                ('BACKGROUND', (0,0), (-1,0), colors.HexColor('#4682B4')),
+                ('BACKGROUND', (0,0), (-1,0), colors.HexColor('#003366')),
                 ('TEXTCOLOR',(0,0),(-1,0),colors.whitesmoke),
-                ('ALIGN', (0,0), (-1,-1), 'CENTER'),
+                ('ALIGN', (0,0), (-1,-1), 'LEFT'),
                 ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
                 ('FONTNAME', (0, 0), (-1, 0), 'Helvetica-Bold'),
-                ('BOTTOMPADDING', (0, 0), (-1, 0), 12),
+                ('FONTSIZE', (0, 0), (-1, -1), 9),
+                ('BOTTOMPADDING', (0, 0), (-1, 0), 10),
+                ('TOPPADDING', (0, 0), (-1, 0), 10),
                 ('BACKGROUND', (0, 1), (-1, -1), colors.HexColor('#F0F8FF')),
-                ('GRID', (0,0), (-1,-1), 1, colors.black)
+                ('GRID', (0,0), (-1,-1), 1, colors.lightgrey)
             ]))
             story.append(table)
             story.append(Spacer(1, 0.2 * inch))
 
-            # Chart generation
             chart_buffer = io.BytesIO()
             generate_anxiety_confidence_chart(voice_analysis.get('composite_scores', {}), chart_buffer)
             chart_buffer.seek(0)
             img = Image(chart_buffer, width=4*inch, height=2.5*inch)
+            img.hAlign = 'CENTER'
             story.append(img)
         else:
             story.append(Paragraph("Voice analysis data not available.", body_text))
 
         story.append(PageBreak())
-
-        # --- Parse and display Gemini report sections ---
+        
+        # --- Gemini Report Parsing and Display ---
         sections = {}
+        # Pre-populate to maintain order
+        section_titles = ["Executive Summary", "Voice Analysis Insights", "Content Analysis & Strengths/Areas for Development", "Actionable Recommendations"]
+        for title in section_titles:
+            sections[title] = []
+
+        # Use a more robust way to capture content under each heading
+        # This regex captures the heading line itself to exclude it from the content
+        report_parts = re.split(r'(\s*\*\*\s*\d\.\s*.*?\s*\*\*)', gemini_report_text)
+        
         current_section = None
-        # Simplified patterns to find sections
-        section_patterns = {
-            'Executive Summary': r'executive summary',
-            'Voice Analysis Insights': r'voice analysis insights',
-            'Content Analysis & Strengths/Areas for Development': r'content analysis|strengths/areas for development',
-            'Actionable Recommendations': r'actionable recommendations'
-        }
-        # Pre-populate sections to maintain order
-        for name in section_patterns.keys():
-            sections[name] = []
-
-        # Parse text into sections based on keywords
-        for line in gemini_report_text.split('\n'):
-            line_lower = line.lower()
-            matched = False
-            for name, pattern in section_patterns.items():
-                if re.search(pattern, line_lower):
-                    current_section = name
-                    matched = True
+        for part in report_parts:
+            if not part.strip(): continue
+            
+            is_heading = False
+            for title in section_titles:
+                # Check if the part is a heading
+                if title.lower() in part.lower():
+                    current_section = title
+                    is_heading = True
                     break
-            if not matched and current_section:
-                sections[current_section].append(line)
+            
+            if not is_heading and current_section:
+                sections[current_section].append(part.strip())
 
-        # Display Content Analysis and Recommendations
-        story.append(Paragraph("2. Content Analysis", h2))
+        # Display Content and Recommendations
+        story.append(Paragraph("2. Content Analysis (from Gemini)", h2))
         if sections['Content Analysis & Strengths/Areas for Development']:
             for line in sections['Content Analysis & Strengths/Areas for Development']:
-                line = line.strip()
-                if not line: continue
                 if line.startswith(('-', '•', '*')):
                     story.append(Paragraph(line.lstrip('-•* ').strip(), bullet_style))
                 else:
                     story.append(Paragraph(line, body_text))
         else:
-            story.append(Paragraph("Content analysis not provided in the report.", body_text))
+            story.append(Paragraph("Content analysis not provided.", body_text))
 
-        story.append(Spacer(1, 0.2*inch))
-
-        story.append(Paragraph("3. Recommendations", h2))
+        story.append(Spacer(1, 0.3*inch))
+        
+        story.append(Paragraph("3. Actionable Recommendations (from Gemini)", h2))
         if sections['Actionable Recommendations']:
             for line in sections['Actionable Recommendations']:
-                line = line.strip()
-                if not line: continue
                 if line.startswith(('-', '•', '*')):
                     story.append(Paragraph(line.lstrip('-•* ').strip(), bullet_style))
                 else:
                     story.append(Paragraph(line, body_text))
         else:
-            story.append(Paragraph("Recommendations not provided in the report.", body_text))
-
+            story.append(Paragraph("Recommendations not provided.", body_text))
+        
         doc.build(story, onFirstPage=header_footer, onLaterPages=header_footer)
         return True
     except Exception as e:
         logger.error(f"Enhanced PDF creation failed: {str(e)}", exc_info=True)
         return False
 
-
-
 def convert_to_serializable(obj):
     if isinstance(obj, np.generic): return obj.item()
     if isinstance(obj, dict): return {k: convert_to_serializable(v) for k, v in obj.items()}