app.py

#!/usr/bin/env python3
"""
DOCX to PDF Converter with Perfect Formatting Preservation
Optimized for Hugging Face Spaces with LibreOffice headless mode
Supports Arabic RTL text and preserves all original formatting
"""

import subprocess
import tempfile
import shutil
import os
from pathlib import Path
import zipfile
import re
import json
import xml.etree.ElementTree as ET
from xml.dom import minidom

import threading
import time

def internal_keepalive():
    while True:
        print("[KeepAlive] ✅ Still alive and running...")
        time.sleep(300)  # كل 5 دقائق

# 🚀 تشغيل الخيط في الخلفية (غير متزامن)
threading.Thread(target=internal_keepalive, daemon=True).start()



def setup_libreoffice():
    """Ensure LibreOffice is properly configured for headless operation with optimal font setup"""
    try:
        # Setup font cache and configuration
        setup_font_environment()

        # Test LibreOffice installation
        result = subprocess.run(
            ["libreoffice", "--version"],
            capture_output=True,
            text=True,
            timeout=10
        )
        if result.returncode != 0:
            raise Exception("LibreOffice not found or not working")

        print(f"LibreOffice version: {result.stdout.strip()}")
        return True
    except Exception as e:
        print(f"LibreOffice setup error: {e}")
        return False


def setup_font_environment():
    """Setup optimal font environment using local Arial font and Arabic RTL support"""
    try:
        # Setup local Arial font from fonts directory
        setup_local_arial_font()

        # Install additional Arabic fonts if not available
        install_arabic_fonts()

        # Update font cache for better font discovery (fc-cache comes with fontconfig package)
        print("Updating font cache...")
        fc_result = subprocess.run(["fc-cache", "-fv"], capture_output=True, timeout=30)
        if fc_result.returncode != 0:
            print(f"Font cache update warning: {fc_result.stderr.decode('utf-8', errors='ignore')}")
        else:
            print("Font cache updated successfully")

        # List available fonts for debugging
        font_result = subprocess.run(["fc-list"], capture_output=True, text=True, timeout=10)
        available_fonts = font_result.stdout

        # Check for critical fonts including Arabic fonts and local Arial
        critical_fonts = ["Arial", "Liberation Sans", "Carlito", "Caladea", "DejaVu Sans", "Noto Sans",
                         "Noto Naskh Arabic", "Noto Kufi Arabic", "Amiri", "Scheherazade New"]
        missing_fonts = []

        for font in critical_fonts:
            if font.lower() not in available_fonts.lower():
                missing_fonts.append(font)

        if missing_fonts:
            print(f"Warning: Missing critical fonts: {missing_fonts}")
        else:
            print("All critical fonts including local Arial and Arabic fonts are available")

        # Check specifically for Arabic font support
        arabic_fonts = ["Noto Naskh Arabic", "Noto Kufi Arabic", "Amiri", "Scheherazade New", "Traditional Arabic"]
        available_arabic = [font for font in arabic_fonts if font.lower() in available_fonts.lower()]
        print(f"Available Arabic fonts: {available_arabic}")

        # Check for local Arial font
        if "arial" in available_fonts.lower():
            print("✅ Local Arial font is available and ready for use")
        else:
            print("⚠️ Local Arial font not detected - will use fallback fonts")

        print(f"Total fonts available: {len(available_fonts.splitlines())}")

    except Exception as e:
        print(f"Font environment setup warning: {e}")


def setup_local_arial_font():
    """Setup local Arial font from same directory as this Python file"""
    try:
        # Get the directory where this Python file is located
        script_dir = Path(__file__).parent.absolute()

        # Path to Arial font in same directory as this script
        arial_font_path = script_dir / "arial.ttf"

        if not arial_font_path.exists():
            print(f"⚠️ Arial font not found at {arial_font_path}")
            print(f"   Script directory: {script_dir}")
            print(f"   Looking for: arial.ttf")
            # Try to find system Arial font as fallback
            system_arial_paths = [
                "/usr/share/fonts/truetype/freefont/FreeSans.ttf",
                "/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf",
                "/usr/share/fonts/truetype/liberation/LiberationSans-Regular.ttf"
            ]
            
            for path in system_arial_paths:
                if os.path.exists(path):
                    print(f"✅ Using system font as Arial fallback: {path}")
                    # Create system fonts directory for local Arial in /tmp to avoid permission issues
                    system_fonts_dir = Path("/tmp/fonts/truetype/local-arial")
                    system_fonts_dir.mkdir(parents=True, exist_ok=True)
                    # Ensure proper permissions
                    try:
                        system_fonts_dir.chmod(0o777)
                    except PermissionError:
                        # If we can't change permissions, continue anyway
                        pass

                    # Copy Arial font to system directory
                    system_arial_path = system_fonts_dir / "arial.ttf"
                    if not system_arial_path.exists():
                        print("📥 Installing local Arial font...")
                        shutil.copy2(path, system_arial_path)
                        try:
                            system_arial_path.chmod(0o644)
                        except PermissionError:
                            # If we can't change permissions, continue anyway
                            pass
                        print("✅ Local Arial font installed successfully")
                    else:
                        print("✅ Local Arial font already installed")
                    return True
            return False

        # Create system fonts directory for local Arial in /tmp to avoid permission issues
        system_fonts_dir = Path("/tmp/fonts/truetype/local-arial")
        system_fonts_dir.mkdir(parents=True, exist_ok=True)
        # Ensure proper permissions
        try:
            system_fonts_dir.chmod(0o777)
        except PermissionError:
            # If we can't change permissions, continue anyway
            pass

        # Copy Arial font to system directory
        system_arial_path = system_fonts_dir / "arial.ttf"
        if not system_arial_path.exists():
            print("📥 Installing local Arial font...")
            shutil.copy2(arial_font_path, system_arial_path)
            try:
                system_arial_path.chmod(0o644)
            except PermissionError:
                # If we can't change permissions, continue anyway
                pass
            print("✅ Local Arial font installed successfully")
        else:
            print("✅ Local Arial font already installed")

        return True

    except Exception as e:
        print(f"❌ Local Arial font setup failed: {e}")
        return False


def install_arabic_fonts():
    """Install additional Arabic fonts for better RTL support"""
    try:
        import urllib.request
        import zipfile
        import tempfile

        # Create fonts directory in /tmp to avoid permission issues
        fonts_dir = Path("/tmp/fonts/truetype/arabic-custom")
        fonts_dir.mkdir(parents=True, exist_ok=True)
        # Ensure proper permissions
        try:
            fonts_dir.chmod(0o777)
        except PermissionError:
            # If we can't change permissions, continue anyway
            pass

        print("🔤 Installing Arabic fonts for RTL support...")

        # Download and install Amiri font
        print("📥 Installing Amiri font...")
        try:
            with tempfile.TemporaryDirectory() as tmp_dir:
                amiri_url = "https://github.com/aliftype/amiri/releases/download/0.117/Amiri-0.117.zip"
                amiri_zip = os.path.join(tmp_dir, "amiri.zip")

                urllib.request.urlretrieve(amiri_url, amiri_zip)

                with zipfile.ZipFile(amiri_zip, 'r') as zip_ref:
                    zip_ref.extractall(tmp_dir)

                amiri_dir = os.path.join(tmp_dir, "Amiri-0.117")
                if os.path.exists(amiri_dir):
                    for file in os.listdir(amiri_dir):
                        if file.endswith('.ttf'):
                            src = os.path.join(amiri_dir, file)
                            dst = fonts_dir / file
                            shutil.copy2(src, dst)
                            try:
                                dst.chmod(0o644)
                            except PermissionError:
                                # If we can't change permissions, continue anyway
                                pass
                    print("✅ Amiri font installed successfully")
                else:
                    print("❌ Amiri font directory not found")
        except Exception as e:
            print(f"❌ Amiri font installation failed: {e}")

        # Download and install Scheherazade New font
        print("📥 Installing Scheherazade New font...")
        try:
            with tempfile.TemporaryDirectory() as tmp_dir:
                scheherazade_url = "https://github.com/silnrsi/font-scheherazade/releases/download/v3.300/ScheherazadeNew-3.300.zip"
                scheherazade_zip = os.path.join(tmp_dir, "scheherazade.zip")

                urllib.request.urlretrieve(scheherazade_url, scheherazade_zip)

                with zipfile.ZipFile(scheherazade_zip, 'r') as zip_ref:
                    zip_ref.extractall(tmp_dir)

                scheherazade_dir = os.path.join(tmp_dir, "ScheherazadeNew-3.300")
                if os.path.exists(scheherazade_dir):
                    for file in os.listdir(scheherazade_dir):
                        if file.endswith('.ttf'):
                            src = os.path.join(scheherazade_dir, file)
                            dst = fonts_dir / file
                            shutil.copy2(src, dst)
                            try:
                                dst.chmod(0o644)
                            except PermissionError:
                                # If we can't change permissions, continue anyway
                                pass
                    print("✅ Scheherazade New font installed successfully")
                else:
                    print("❌ Scheherazade New font directory not found")
        except Exception as e:
            print(f"❌ Scheherazade New font installation failed: {e}")

        # Download and install Noto Sans Arabic font
        print("📥 Installing Noto Sans Arabic font...")
        try:
            with tempfile.TemporaryDirectory() as tmp_dir:
                noto_url = "https://github.com/notofonts/notofonts.github.io/raw/main/fonts/NotoSansArabic/hinted/ttf/NotoSansArabic-Regular.ttf"
                noto_file = os.path.join(tmp_dir, "NotoSansArabic-Regular.ttf")

                urllib.request.urlretrieve(noto_url, noto_file)

                dst = fonts_dir / "NotoSansArabic-Regular.ttf"
                shutil.copy2(noto_file, dst)
                try:
                    dst.chmod(0o644)
                except PermissionError:
                    # If we can't change permissions, continue anyway
                    pass
                print("✅ Noto Sans Arabic font installed successfully")
        except Exception as e:
            print(f"❌ Noto Sans Arabic font installation failed: {e}")

        # Download and install Cairo font (excellent for Arabic)
        print("📥 Installing Cairo font...")
        try:
            with tempfile.TemporaryDirectory() as tmp_dir:
                # Use a more reliable source for Cairo font
                cairo_url = "https://github.com/Gue3bara/Cairo/raw/master/fonts/Cairo-Regular.ttf"
                cairo_file = os.path.join(tmp_dir, "Cairo-Regular.ttf")

                urllib.request.urlretrieve(cairo_url, cairo_file)

                dst = fonts_dir / "Cairo-Regular.ttf"
                shutil.copy2(cairo_file, dst)
                try:
                    dst.chmod(0o644)
                except PermissionError:
                    # If we can't change permissions, continue anyway
                    pass
                print("✅ Cairo font installed successfully")
        except Exception as e:
            print(f"❌ Cairo font installation failed: {e}")
            print("⚠️ Continuing without Cairo font - using alternative Arabic fonts")

            try:
                with tempfile.TemporaryDirectory() as tmp_dir:
                    # Alternative Cairo font URL
                    cairo_url = "https://fonts.gstatic.com/s/cairo/v21/SLXgc14kyrzQ6fYy3Q60fTh5Tf44DXYvbqo6vPQ3ZyM.woff2"
                    cairo_file = os.path.join(tmp_dir, "Cairo-Regular.woff2")
                    
                    urllib.request.urlretrieve(cairo_url, cairo_file)
                    
                    # Convert WOFF2 to TTF using fontTools if available
                    try:
                        from fontTools.ttLib import TTFont
                        # For now, just acknowledge the download
                        print("✅ Cairo font (alternative source) downloaded successfully")
                    except ImportError:
                        print("ℹ️ Cairo font downloaded but font conversion tools not available")
                print("✅ Cairo font installed successfully (alternative source)")
            except Exception as e2:
                print(f"❌ Cairo font installation failed (alternative source): {e2}")

        # Update font cache after installation
        print("🔄 Updating font cache...")
        try:
            subprocess.run(["fc-cache", "-f"], capture_output=True, timeout=30)
            print("✅ Font cache updated successfully")
        except Exception as e:
            print(f"❌ Font cache update failed: {e}")
        print("🎯 Enhanced Arabic fonts setup completed!")

    except Exception as e:
        print(f"Arabic fonts installation warning: {e}")


def create_fontconfig(temp_path):
    """Create fontconfig configuration for optimal font matching with local Arial and Arabic RTL support"""
    fontconfig_dir = temp_path / ".config" / "fontconfig"
    fontconfig_dir.mkdir(parents=True, exist_ok=True)

    # Create fonts.conf
    fonts_conf = fontconfig_dir / "fonts.conf"

    # Get script directory for local fonts
    script_dir = Path(__file__).parent.absolute()

    fontconfig_content = f'''<?xml version="1.0"?>
<!DOCTYPE fontconfig SYSTEM "fonts.dtd">
<fontconfig>
  <!-- Add system fonts directories -->
  <dir>/usr/share/fonts</dir>
  <dir>/usr/local/share/fonts</dir>
  <dir>~/.fonts</dir>

  <!-- Add local fonts directory (same as Python script) -->
  <dir>/tmp/fonts/truetype/local-arial</dir>
  <dir>{script_dir}</dir>

  <!-- Font substitution rules with local Arial as priority -->
  <alias>
    <family>Arial</family>
    <prefer>
      <family>Arial</family>
      <family>Liberation Sans</family>
      <family>DejaVu Sans</family>
      <family>Noto Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Calibri</family>
    <prefer>
      <family>Liberation Sans</family>
      <family>Arimo</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Cambria</family>
    <prefer>
      <family>Liberation Serif</family>
      <family>Tinos</family>
      <family>DejaVu Serif</family>
    </prefer>
  </alias>

  <alias>
    <family>Times New Roman</family>
    <prefer>
      <family>Liberation Serif</family>
      <family>DejaVu Serif</family>
      <family>Noto Serif</family>
    </prefer>
  </alias>

  <alias>
    <family>Courier New</family>
    <prefer>
      <family>Liberation Mono</family>
      <family>DejaVu Sans Mono</family>
      <family>Noto Sans Mono</family>
    </prefer>
  </alias>

  <!-- Enhanced Arabic font substitution rules for perfect RTL support -->
  <alias>
    <family>Traditional Arabic</family>
    <prefer>
      <family>Amiri</family>
      <family>Noto Naskh Arabic</family>
      <family>Scheherazade New</family>
      <family>Cairo</family>
      <family>Noto Sans Arabic</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Arabic Typesetting</family>
    <prefer>
      <family>Amiri</family>
      <family>Noto Naskh Arabic</family>
      <family>Scheherazade New</family>
      <family>Cairo</family>
      <family>Noto Sans Arabic</family>
    </prefer>
  </alias>

  <alias>
    <family>Simplified Arabic</family>
    <prefer>
      <family>Noto Sans Arabic</family>
      <family>Cairo</family>
      <family>Noto Naskh Arabic</family>
      <family>Amiri</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <!-- Additional Arabic font mappings for maximum compatibility -->
  <alias>
    <family>Arial Unicode MS</family>
    <prefer>
      <family>Noto Sans Arabic</family>
      <family>Cairo</family>
      <family>Liberation Sans</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Microsoft Sans Serif</family>
    <prefer>
      <family>Noto Sans Arabic</family>
      <family>Liberation Sans</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Segoe UI</family>
    <prefer>
      <family>Noto Sans Arabic</family>
      <family>Cairo</family>
      <family>Liberation Sans</family>
      <family>DejaVu Sans</family>
    </prefer>
  </alias>

  <alias>
    <family>Tahoma</family>
    <prefer>
      <family>DejaVu Sans</family>
      <family>Liberation Sans</family>
      <family>Noto Sans</family>
    </prefer>
  </alias>

  <!-- Generic Arabic font fallback -->
  <alias>
    <family>serif</family>
    <prefer>
      <family>Liberation Serif</family>
      <family>DejaVu Serif</family>
      <family>Amiri</family>
      <family>Noto Naskh Arabic</family>
    </prefer>
  </alias>

  <alias>
    <family>sans-serif</family>
    <prefer>
      <family>Liberation Sans</family>
      <family>DejaVu Sans</family>
      <family>Noto Sans</family>
      <family>Noto Naskh Arabic</family>
    </prefer>
  </alias>

  <alias>
    <family>monospace</family>
    <prefer>
      <family>Liberation Mono</family>
      <family>DejaVu Sans Mono</family>
      <family>Noto Sans Mono</family>
    </prefer>
  </alias>

  <!-- Ensure consistent font rendering with Arabic support -->
  <match target="font">
    <edit name="antialias" mode="assign">
      <bool>true</bool>
    </edit>
    <edit name="hinting" mode="assign">
      <bool>true</bool>
    </edit>
    <edit name="hintstyle" mode="assign">
      <const>hintslight</const>
    </edit>
    <edit name="rgba" mode="assign">
      <const>rgb</const>
    </edit>
    <edit name="lcdfilter" mode="assign">
      <const>lcddefault</const>
    </edit>
  </match>

  <!-- Enhanced Arabic script handling with strong binding -->
  <match target="pattern">
    <test name="lang" compare="contains">
      <string>ar</string>
    </test>
    <edit name="family" mode="prepend" binding="strong">
      <string>Amiri</string>
      <string>Noto Naskh Arabic</string>
      <string>Scheherazade New</string>
      <string>Cairo</string>
      <string>Noto Sans Arabic</string>
    </edit>
  </match>

  <!-- Force Arabic fonts for any Arabic-containing text -->
  <match target="pattern">
    <test name="family" compare="contains">
      <string>Arabic</string>
    </test>
    <edit name="family" mode="prepend" binding="strong">
      <string>Amiri</string>
      <string>Noto Naskh Arabic</string>
      <string>Scheherazade New</string>
      <string>Cairo</string>
    </edit>
  </match>

  <!-- Ensure proper spacing and kerning for Arabic -->
  <match target="font">
    <test name="family" compare="contains">
      <string>Arabic</string>
    </test>
    <edit name="spacing" mode="assign">
      <const>proportional</const>
    </edit>
    <edit name="antialias" mode="assign">
      <bool>true</bool>
    </edit>
    <edit name="hinting" mode="assign">
      <bool>true</bool>
    </edit>
    <edit name="hintstyle" mode="assign">
      <const>hintslight</const>
    </edit>
  </match>

  <!-- Specific handling for RTL text -->
  <match target="pattern">
    <test name="charset">
      <charset>
        <range>
          <int>0x0600</int>
          <int>0x06FF</int>
        </range>
      </charset>
    </test>
    <edit name="family" mode="prepend" binding="strong">
      <string>Amiri</string>
      <string>Noto Naskh Arabic</string>
      <string>Scheherazade New</string>
      <string>Cairo</string>
    </edit>
  </match>
</fontconfig>'''

    with open(fonts_conf, 'w', encoding='utf-8') as f:
        f.write(fontconfig_content)

    return str(fontconfig_dir.parent)


def analyze_template_font_sizes(docx_path):
    """Analyze template.docx to extract specific font size requirements"""
    try:
        font_size_mapping = {}

        with zipfile.ZipFile(docx_path, 'r') as docx:
            if 'word/document.xml' in docx.namelist():
                doc_content = docx.read('word/document.xml').decode('utf-8')

                # Extract text runs with their font sizes
                import xml.etree.ElementTree as ET
                root = ET.fromstring(doc_content)

                # Define namespaces
                namespaces = {
                    'w': 'http://schemas.openxmlformats.org/wordprocessingml/2006/main'
                }

                # Find all text runs and their font sizes
                for run in root.findall('.//w:r', namespaces):
                    # Get font size from run properties
                    rpr = run.find('w:rPr', namespaces)
                    if rpr is not None:
                        sz_elem = rpr.find('w:sz', namespaces)
                        if sz_elem is not None:
                            font_size = int(sz_elem.get('{http://schemas.openxmlformats.org/wordprocessingml/2006/main}val', '20')) // 2  # Convert half-points to points
                        else:
                            font_size = 10  # Smaller default
                    else:
                        font_size = 10  # Smaller default

                    # Get text content
                    text_elements = run.findall('.//w:t', namespaces)
                    for text_elem in text_elements:
                        text_content = text_elem.text
                        if text_content and text_content.strip():
                            # Map specific text patterns to font sizes
                            text_content = text_content.strip()

                            # Check for specific patterns mentioned by user (smaller sizes)
                            if any(pattern in text_content for pattern in ['{{serial_number}}', '{{t_11}}', '{{t_}}', '{{date}}']):
                                font_size_mapping[text_content] = 9  # Smaller size for serial/date
                            elif any(pattern in text_content for pattern in ['{{name_1}}', '{{name_2}}', '{{id_1}}', '{{name_3}}', '{{id_2}}']):
                                font_size_mapping[text_content] = 10  # Smaller size for names/IDs
                            elif any(pattern in text_content for pattern in ['{{location_1}}', '{{location_2}}', '{{phone_1}}', '{{location_3}}', '{{phone_2}}']):
                                font_size_mapping[text_content] = 10  # Smaller size for locations/phones
                            elif any(pattern in text_content for pattern in ['الطرف البائع', 'الطرف المشتري']):
                                font_size_mapping[text_content] = 11  # Smaller size for main headers
                            else:
                                # Default size for other text (smaller)
                                font_size_mapping[text_content] = min(font_size, 10)  # Cap at size 10

                print(f"📏 Font size analysis completed: {len(font_size_mapping)} text patterns mapped")
                return font_size_mapping

    except Exception as e:
        print(f"❌ Font size analysis failed: {e}")
        return {}


def validate_docx_structure(docx_path):
    """Advanced DOCX structure analysis and preprocessing for perfect formatting preservation"""
    try:
        validation_info = {
            'page_count': 1,  # Default assumption
            'has_tables': False,
            'has_images': False,
            'text_content_length': 0,
            'font_families': set(),
            'has_textboxes': False,
            'has_smartart': False,
            'has_complex_shapes': False,
            'table_structure_issues': [],
            'rtl_content_detected': False,
            'placeholder_count': 0,
            'font_size_mapping': {},
            'error': None
        }

        # Analyze font sizes for template.docx
        if 'template.docx' in docx_path:
            validation_info['font_size_mapping'] = analyze_template_font_sizes(docx_path)

        with zipfile.ZipFile(docx_path, 'r') as docx:
            # Check document.xml for content analysis
            if 'word/document.xml' in docx.namelist():
                doc_content = docx.read('word/document.xml').decode('utf-8')

                # Count tables and analyze structure
                table_count = doc_content.count('<w:tbl>')
                validation_info['has_tables'] = table_count > 0

                # Advanced table structure analysis
                if validation_info['has_tables']:
                    # Check for nested tables (problematic for formatting)
                    nested_tables = doc_content.count('<w:tbl>') - doc_content.count('</w:tbl>')
                    if nested_tables != 0:
                        validation_info['table_structure_issues'].append("Nested tables detected")

                    # Check for merged cells that might cause layout issues
                    if '<w:gridSpan' in doc_content or '<w:vMerge' in doc_content:
                        validation_info['table_structure_issues'].append("Complex cell merging detected")

                # Check for problematic elements that LibreOffice handles poorly
                validation_info['has_textboxes'] = '<w:textbox>' in doc_content or '<w:txbxContent>' in doc_content
                validation_info['has_smartart'] = '<w:smartTag>' in doc_content or 'smartart' in doc_content.lower()
                validation_info['has_complex_shapes'] = '<w:shape>' in doc_content or '<w:group>' in doc_content

                # Check for images with enhanced detection
                validation_info['has_images'] = ('<w:drawing>' in doc_content or
                                                '<w:pict>' in doc_content or
                                                '<w:object>' in doc_content)

                # Detect RTL content and Arabic text
                arabic_pattern = r'[\u0600-\u06FF\u0750-\u077F\u08A0-\u08FF\uFB50-\uFDFF\uFE70-\uFEFF]'
                validation_info['rtl_content_detected'] = bool(re.search(arabic_pattern, doc_content))

                # Count placeholders ({{...}}) that must not be moved
                placeholder_pattern = r'\{\{[^}]+\}\}'
                validation_info['placeholder_count'] = len(re.findall(placeholder_pattern, doc_content))

                # Estimate text content length
                text_content = re.sub(r'<[^>]+>', '', doc_content)
                validation_info['text_content_length'] = len(text_content.strip())

                # Enhanced font extraction including Arabic fonts
                font_matches = re.findall(r'w:ascii="([^"]+)"', doc_content)
                eastasia_fonts = re.findall(r'w:eastAsia="([^"]+)"', doc_content)
                cs_fonts = re.findall(r'w:cs="([^"]+)"', doc_content)  # Complex Script fonts (Arabic)

                all_fonts = set(font_matches + eastasia_fonts + cs_fonts)
                validation_info['font_families'] = all_fonts

                print(f"🔍 Advanced DOCX Analysis:")
                print(f"  • Tables: {table_count} (Issues: {len(validation_info['table_structure_issues'])})")
                print(f"  • Images: {validation_info['has_images']}")
                print(f"  • TextBoxes: {validation_info['has_textboxes']}")
                print(f"  • SmartArt: {validation_info['has_smartart']}")
                print(f"  • Complex Shapes: {validation_info['has_complex_shapes']}")
                print(f"  • RTL Content: {validation_info['rtl_content_detected']}")
                print(f"  • Placeholders: {validation_info['placeholder_count']}")
                print(f"  • Text Length: {validation_info['text_content_length']}")
                print(f"  • Fonts: {list(validation_info['font_families'])[:5]}...")

        return validation_info

    except Exception as e:
        print(f"❌ DOCX validation error: {e}")
        return {'page_count': 1, 'has_tables': False, 'has_images': False,
                'text_content_length': 0, 'font_families': set(), 'has_textboxes': False,
                'has_smartart': False, 'has_complex_shapes': False, 'table_structure_issues': [],
                'rtl_content_detected': False, 'placeholder_count': 0, 'error': str(e)}


def calculate_optimal_font_size(text_content, max_width_chars=20, base_font_size=10):
    """
    Calculate optimal font size based on text length to maintain position
    This ensures that longer names don't break the layout
    """
    if not text_content:
        return base_font_size

    # Remove placeholder brackets if present
    clean_text = text_content.replace('{{', '').replace('}}', '').strip()
    text_length = len(clean_text)

    # If text is short, use base font size
    if text_length <= max_width_chars:
        return base_font_size

    # Calculate reduction factor based on text length
    # The longer the text, the smaller the font should be
    reduction_factor = max_width_chars / text_length

    # Apply reduction but don't go below 7pt (14 half-points)
    optimal_size = max(base_font_size * reduction_factor, 7)

    return int(optimal_size)


def extract_placeholder_contexts(doc_content):
    """
    Extract placeholders with their surrounding context to understand layout constraints
    """
    placeholder_contexts = {}

    # Find all placeholders with their XML context
    placeholder_pattern = r'(<w:r[^>]*>.*?<w:t[^>]*>.*?\{\{[^}]+\}\}.*?</w:t>.*?</w:r>)'
    matches = re.findall(placeholder_pattern, doc_content, re.DOTALL)

    for match in matches:
        # Extract the placeholder name
        placeholder_match = re.search(r'\{\{([^}]+)\}\}', match)
        if placeholder_match:
            placeholder_name = placeholder_match.group(1)

            # Extract current font size if present
            font_size_match = re.search(r'<w:sz w:val="(\d+)"/>', match)
            current_font_size = int(font_size_match.group(1)) // 2 if font_size_match else 10

            # Check if this is in a table cell (more constrained space)
            is_in_table = '<w:tc>' in match or 'w:tcPr' in match

            # Estimate available width based on context
            if is_in_table:
                max_width_chars = 15  # Table cells are more constrained
            else:
                max_width_chars = 25  # Regular text has more space

            placeholder_contexts[placeholder_name] = {
                'current_font_size': current_font_size,
                'max_width_chars': max_width_chars,
                'is_in_table': is_in_table,
                'xml_context': match
            }

    return placeholder_contexts


def apply_template_font_settings(docx_path, validation_info):
    """Apply specific font sizes and Arial font to template.docx content with smart sizing"""
    try:
        if not validation_info.get('font_size_mapping'):
            print("ℹ️ No font size mapping found - skipping font optimization")
            return docx_path

        print("🔤 Applying template-specific font settings with smart sizing...")

        # Create a temporary copy for processing
        temp_docx = tempfile.mktemp(suffix='.docx')
        shutil.copy2(docx_path, temp_docx)

        with zipfile.ZipFile(temp_docx, 'a') as docx_zip:
            if 'word/document.xml' in docx_zip.namelist():
                doc_content = docx_zip.read('word/document.xml').decode('utf-8')

                # Apply Arial font to all text runs
                # Replace font family declarations with Arial
                doc_content = re.sub(
                    r'w:ascii="[^"]*"',
                    'w:ascii="Arial"',
                    doc_content
                )
                doc_content = re.sub(
                    r'w:hAnsi="[^"]*"',
                    'w:hAnsi="Arial"',
                    doc_content
                )

                # Extract placeholder contexts for smart sizing
                placeholder_contexts = extract_placeholder_contexts(doc_content)
                print(f"📍 Found {len(placeholder_contexts)} placeholders with context")

                # Apply smart font sizing for name placeholders
                name_placeholders = ['name_1', 'name_2', 'name_3']
                for placeholder in name_placeholders:
                    if placeholder in placeholder_contexts:
                        context = placeholder_contexts[placeholder]

                        # Calculate optimal font size for typical Arabic names
                        # Assume names can be 15-30 characters (ثلاثي أو رباعي)
                        optimal_size = calculate_optimal_font_size(
                            "محمد عبدالله أحمد الخالدي",  # Example long name
                            max_width_chars=context['max_width_chars'],
                            base_font_size=context['current_font_size']
                        )

                        # Apply the calculated font size (convert to half-points)
                        optimal_size_half_points = int(optimal_size * 2)

                        pattern = f'{{{{{placeholder}}}}}'
                        if pattern in doc_content:
                            doc_content = re.sub(
                                r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:sz w:val=")[^"]*(")',
                                f'\\g<1>{optimal_size_half_points}\\g<2>',
                                doc_content,
                                flags=re.DOTALL
                            )
                            print(f"🎯 Applied smart sizing to {placeholder}: {optimal_size}pt")

                # Apply size 9 for serial number, date, time elements (smaller size)
                for pattern in ['{{serial_number}}', '{{t_11}}', '{{t_}}', '{{date}}', 'الرقم التسلسلي', 'الساعة', 'التاريخ']:
                    if pattern in doc_content:
                        # Find the text run containing this pattern and set font size to 9 (18 half-points)
                        doc_content = re.sub(
                            r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:sz w:val=")[^"]*(")',
                            r'\g<1>18\g<2>',
                            doc_content,
                            flags=re.DOTALL
                        )

                # Apply size 10 for IDs, locations, phones (smaller size)
                for pattern in ['{{id_1}}', '{{id_2}}',
                               '{{location_1}}', '{{location_2}}', '{{phone_1}}', '{{location_3}}', '{{phone_2}}',
                               'رقم الهوية', 'يسكن', 'رقم الهاتف']:
                    if pattern in doc_content:
                        # Set font size to 10 (20 half-points)
                        doc_content = re.sub(
                            r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:sz w:val=")[^"]*(")',
                            r'\g<1>20\g<2>',
                            doc_content,
                            flags=re.DOTALL
                        )

                # Apply size 11 for "الطرف البائع" and "الطرف المشتري" (smaller size)
                for pattern in ['الطرف البائع', 'الطرف المشتري']:
                    if pattern in doc_content:
                        # Set font size to 11 (22 half-points)
                        doc_content = re.sub(
                            r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:sz w:val=")[^"]*(")',
                            r'\g<1>22\g<2>',
                            doc_content,
                            flags=re.DOTALL
                        )

                # Apply general font size reduction for all text (reduce large fonts)
                print("🔤 Applying general font size optimization...")
                # Find all font sizes and reduce if they're too large
                font_size_pattern = r'<w:sz w:val="(\d+)"/>'
                def reduce_font_size(match):
                    size = int(match.group(1))
                    # Convert half-points to points for comparison
                    size_in_points = size // 2

                    # If font is larger than 12pt, reduce it
                    if size_in_points > 12:
                        new_size_points = min(size_in_points * 0.8, 12)  # Reduce by 20% but cap at 12pt
                        new_size_half_points = int(new_size_points * 2)
                        return f'<w:sz w:val="{new_size_half_points}"/>'
                    elif size_in_points > 10:
                        # For medium sizes, reduce slightly
                        new_size_points = size_in_points * 0.9  # Reduce by 10%
                        new_size_half_points = int(new_size_points * 2)
                        return f'<w:sz w:val="{new_size_half_points}"/>'
                    else:
                        # Keep small fonts as they are
                        return match.group(0)

                doc_content = re.sub(font_size_pattern, reduce_font_size, doc_content)

                # Write back the modified document.xml
                docx_zip.writestr('word/document.xml', doc_content.encode('utf-8'))
                print("✅ Template font settings with smart sizing applied successfully")

        return temp_docx

    except Exception as e:
        print(f"❌ Font settings application failed: {e}")
        return docx_path


def create_dynamic_font_sizing_rules(docx_path):
    """
    Create dynamic font sizing rules based on actual content analysis
    This function analyzes the document to create smart sizing rules
    """
    try:
        dynamic_rules = {}

        with zipfile.ZipFile(docx_path, 'r') as docx:
            if 'word/document.xml' in docx.namelist():
                doc_content = docx.read('word/document.xml').decode('utf-8')

                # Find all placeholders and their current context
                placeholder_pattern = r'\{\{([^}]+)\}\}'
                placeholders = re.findall(placeholder_pattern, doc_content)

                for placeholder in placeholders:
                    # Analyze the context around each placeholder
                    context_pattern = f'(<w:tc[^>]*>.*?\\{{{{' + re.escape(placeholder) + r'\\}}}}.*?</w:tc>)'
                    table_cell_match = re.search(context_pattern, doc_content, re.DOTALL)

                    if table_cell_match:
                        # This placeholder is in a table cell
                        cell_content = table_cell_match.group(1)

                        # Estimate cell width based on content and structure
                        # Look for width specifications
                        width_match = re.search(r'w:w="(\d+)"', cell_content)
                        if width_match:
                            cell_width = int(width_match.group(1))
                            # Convert twips to approximate character width
                            # 1440 twips = 1 inch, average character = 0.1 inch
                            estimated_chars = max(cell_width // 144, 10)  # Minimum 10 chars
                        else:
                            estimated_chars = 15  # Default for table cells

                        # Check if there are other elements in the same cell
                        text_elements = re.findall(r'<w:t[^>]*>([^<]+)</w:t>', cell_content)
                        total_text_length = sum(len(text.replace(f'{{{{{placeholder}}}}}', '')) for text in text_elements)

                        # Adjust available space based on other content
                        available_chars = max(estimated_chars - total_text_length, 8)

                        dynamic_rules[placeholder] = {
                            'max_chars': available_chars,
                            'context': 'table_cell',
                            'base_font_size': 10,
                            'min_font_size': 7
                        }
                    else:
                        # This placeholder is in regular text
                        dynamic_rules[placeholder] = {
                            'max_chars': 25,
                            'context': 'paragraph',
                            'base_font_size': 11,
                            'min_font_size': 8
                        }

                print(f"📏 Created dynamic sizing rules for {len(dynamic_rules)} placeholders")
                return dynamic_rules

    except Exception as e:
        print(f"❌ Dynamic rules creation failed: {e}")
        return {}


def apply_dynamic_font_sizing(docx_path, dynamic_rules, sample_data=None):
    """
    Apply dynamic font sizing based on actual or sample data
    This ensures that when placeholders are replaced, the text fits perfectly
    """
    if not dynamic_rules:
        return docx_path

    try:
        print("🎯 Applying dynamic font sizing based on content analysis...")

        # Create sample data if not provided
        if not sample_data:
            sample_data = {
                'name_1': 'محمد عبدالله أحمد الخالدي',  # Long Arabic name
                'name_2': 'فاطمة سعد محمد العتيبي',      # Long Arabic name
                'name_3': 'عبدالرحمن خالد سليمان',      # Medium Arabic name
                'id_1': '1234567890',
                'id_2': '0987654321',
                'location_1': 'الرياض - حي الملك فهد - شارع الأمير محمد بن عبدالعزيز',
                'location_2': 'جدة - حي الصفا - طريق الملك عبدالعزيز',
                'phone_1': '+966501234567',
                'phone_2': '+966509876543'
            }

        # Create a temporary copy for processing
        temp_docx = tempfile.mktemp(suffix='.docx')
        shutil.copy2(docx_path, temp_docx)

        with zipfile.ZipFile(temp_docx, 'a') as docx_zip:
            if 'word/document.xml' in docx_zip.namelist():
                doc_content = docx_zip.read('word/document.xml').decode('utf-8')

                # Apply dynamic sizing for each placeholder
                for placeholder, rules in dynamic_rules.items():
                    if placeholder in sample_data:
                        sample_text = sample_data[placeholder]

                        # Calculate optimal font size
                        optimal_size = calculate_optimal_font_size(
                            sample_text,
                            max_width_chars=rules['max_chars'],
                            base_font_size=rules['base_font_size']
                        )

                        # Ensure minimum font size
                        optimal_size = max(optimal_size, rules['min_font_size'])

                        # Convert to half-points for Word
                        optimal_size_half_points = int(optimal_size * 2)

                        # Apply the font size to this placeholder
                        pattern = f'{{{{{placeholder}}}}}'
                        if pattern in doc_content:
                            # Find and update font size for this specific placeholder
                            placeholder_pattern = r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:sz w:val=")[^"]*(")'
                            doc_content = re.sub(
                                placeholder_pattern,
                                f'\\g<1>{optimal_size_half_points}\\g<2>',
                                doc_content,
                                flags=re.DOTALL
                            )

                            # Also ensure Arial font is applied to this placeholder
                            placeholder_font_pattern = r'(<w:r[^>]*>.*?' + re.escape(pattern) + r'.*?<w:rFonts[^>]*w:ascii=")[^"]*(")'
                            doc_content = re.sub(
                                placeholder_font_pattern,
                                r'\g<1>Arial\g<2>',
                                doc_content,
                                flags=re.DOTALL
                            )

                            # Add font binding to ensure Arial is used
                            placeholder_run_pattern = r'(<w:r[^>]*>)(.*?' + re.escape(pattern) + r'.*?)(</w:r>)'
                            def add_font_binding(match):
                                run_start = match.group(1)
                                run_content = match.group(2)
                                run_end = match.group(3)

                                # Check if rPr (run properties) exists
                                if '<w:rPr>' in run_content:
                                    # Add or update font information
                                    if '<w:rFonts' not in run_content:
                                        run_content = run_content.replace(
                                            '<w:rPr>',
                                            '<w:rPr><w:rFonts w:ascii="Arial" w:hAnsi="Arial" w:cs="Arial"/>'
                                        )
                                else:
                                    # Add rPr with font information
                                    run_content = '<w:rPr><w:rFonts w:ascii="Arial" w:hAnsi="Arial" w:cs="Arial"/></w:rPr>' + run_content

                                return run_start + run_content + run_end

                            doc_content = re.sub(placeholder_run_pattern, add_font_binding, doc_content, flags=re.DOTALL)

                            print(f"🎯 {placeholder}: {optimal_size}pt Arial (max chars: {rules['max_chars']}, context: {rules['context']})")

                # Write back the modified document.xml
                docx_zip.writestr('word/document.xml', doc_content.encode('utf-8'))
                print("✅ Dynamic font sizing applied successfully")

        return temp_docx

    except Exception as e:
        print(f"❌ Dynamic font sizing failed: {e}")
        return docx_path


def preprocess_docx_for_perfect_conversion(docx_path, validation_info):
    """
    Advanced DOCX preprocessing to ensure maximum formatting preservation
    Removes problematic elements and optimizes structure for LibreOffice
    """
    # First apply template-specific font settings if this is template.docx
    if 'template.docx' in docx_path:
        docx_path = apply_template_font_settings(docx_path, validation_info)

        # Apply dynamic font sizing for better placeholder handling
        dynamic_rules = create_dynamic_font_sizing_rules(docx_path)
        if dynamic_rules:
            docx_path = apply_dynamic_font_sizing(docx_path, dynamic_rules)

    if not validation_info.get('has_textboxes') and not validation_info.get('has_smartart') and not validation_info.get('has_complex_shapes'):
        print("✅ DOCX structure is optimal - no additional preprocessing needed")
        return docx_path

    try:
        print("🔧 Preprocessing DOCX for perfect conversion...")

        # Create a temporary copy for processing
        temp_docx = tempfile.mktemp(suffix='.docx')
        shutil.copy2(docx_path, temp_docx)

        with zipfile.ZipFile(temp_docx, 'a') as docx_zip:
            # Read document.xml
            if 'word/document.xml' in docx_zip.namelist():
                doc_content = docx_zip.read('word/document.xml').decode('utf-8')

                # Remove problematic elements that LibreOffice handles poorly
                modifications_made = False

                # Remove TextBoxes (convert to regular paragraphs)
                if validation_info.get('has_textboxes'):
                    print("  • Converting TextBoxes to regular paragraphs...")
                    # Extract text from textboxes and convert to paragraphs
                    textbox_pattern = r'<w:textbox[^>]*>.*?</w:textbox>'
                    textboxes = re.findall(textbox_pattern, doc_content, re.DOTALL)

                    for textbox in textboxes:
                        # Extract text content from textbox
                        text_content = re.sub(r'<[^>]+>', '', textbox)
                        if text_content.strip():
                            # Replace textbox with simple paragraph
                            paragraph = f'<w:p><w:r><w:t>{text_content.strip()}</w:t></w:r></w:p>'
                            doc_content = doc_content.replace(textbox, paragraph)
                            modifications_made = True

                # Remove SmartArt (replace with placeholder text)
                if validation_info.get('has_smartart'):
                    print("  • Removing SmartArt elements...")
                    smartart_pattern = r'<w:smartTag[^>]*>.*?</w:smartTag>'
                    doc_content = re.sub(smartart_pattern, '', doc_content, flags=re.DOTALL)
                    modifications_made = True

                # Simplify complex shapes (remove or convert to text)
                if validation_info.get('has_complex_shapes'):
                    print("  • Simplifying complex shapes...")
                    # Remove complex shape groups
                    shape_group_pattern = r'<w:group[^>]*>.*?</w:group>'
                    doc_content = re.sub(shape_group_pattern, '', doc_content, flags=re.DOTALL)

                    # Simplify individual shapes
                    shape_pattern = r'<w:shape[^>]*>.*?</w:shape>'
                    shapes = re.findall(shape_pattern, doc_content, re.DOTALL)

                    for shape in shapes:
                        # Try to extract any text content from shape
                        text_content = re.sub(r'<[^>]+>', '', shape)
                        if text_content.strip():
                            paragraph = f'<w:p><w:r><w:t>{text_content.strip()}</w:t></w:r></w:p>'
                            doc_content = doc_content.replace(shape, paragraph)
                        else:
                            doc_content = doc_content.replace(shape, '')
                    modifications_made = True

                # Optimize table structure if issues detected
                if validation_info.get('table_structure_issues'):
                    print("  • Optimizing table structure...")
                    # Fix common table issues that cause layout problems

                    # Ensure proper table width settings
                    doc_content = re.sub(
                        r'<w:tblW w:w="0"[^>]*/>',
                        '<w:tblW w:w="5000" w:type="pct"/>',
                        doc_content
                    )

                    # Fix empty table cells that might cause issues
                    empty_cell_pattern = r'<w:tc>\s*</w:tc>'
                    doc_content = re.sub(
                        empty_cell_pattern,
                        '<w:tc><w:p><w:r><w:t> </w:t></w:r></w:p></w:tc>',
                        doc_content
                    )
                    modifications_made = True

                if modifications_made:
                    # Write back the modified document.xml
                    docx_zip.writestr('word/document.xml', doc_content.encode('utf-8'))
                    print("✅ DOCX preprocessing completed successfully")
                else:
                    print("ℹ️ No modifications were needed")

        return temp_docx

    except Exception as e:
        print(f"❌ DOCX preprocessing failed: {e}")
        print("  • Continuing with original file...")
        return docx_path


def validate_pdf_output(pdf_path, expected_info):
    """Validate PDF output against expected metrics"""
    try:
        # Get PDF file size for basic validation
        pdf_size = os.path.getsize(pdf_path)

        validation_results = {
            'file_size_mb': round(pdf_size / (1024 * 1024), 2),
            'file_exists': True,
            'size_reasonable': 0.1 <= pdf_size / (1024 * 1024) <= 100,  # Between 100KB and 100MB
            'warnings': [],
            'success_metrics': []
        }

        # Basic file size validation
        if pdf_size < 1024:  # Less than 1KB is suspicious
            validation_results['warnings'].append("PDF file size is suspiciously small")
        elif pdf_size > 100 * 1024 * 1024:  # More than 100MB
            validation_results['warnings'].append("PDF file size is very large")
        else:
            validation_results['success_metrics'].append("PDF file size is reasonable")

        # Content-based validation hints
        if expected_info['has_tables']:
            validation_results['success_metrics'].append("Document contains tables - formatting preservation critical")

        if expected_info['has_images']:
            validation_results['success_metrics'].append("Document contains images - quality preservation applied")

        if expected_info['font_families']:
            validation_results['success_metrics'].append(f"Font substitution applied for {len(expected_info['font_families'])} font families")

        print(f"PDF Validation: Size={validation_results['file_size_mb']}MB, "
              f"Warnings={len(validation_results['warnings'])}, "
              f"Success_metrics={len(validation_results['success_metrics'])}")

        return validation_results

    except Exception as e:
        print(f"PDF validation error: {e}")
        return {'file_size_mb': 0, 'file_exists': False, 'size_reasonable': False,
                'warnings': [f"Validation error: {e}"], 'success_metrics': []}


def post_process_pdf_for_perfect_formatting(pdf_path, docx_info):
    """
    Advanced PDF post-processing to ensure perfect formatting preservation
    Uses PyMuPDF to verify and correct any layout issues
    """
    try:
        # Try to import PyMuPDF (fitz) - multiple import methods for compatibility
        fitz = None
        try:
            import fitz  # PyMuPDF
        except ImportError:
            try:
                from pymupdf import fitz
            except ImportError:
                try:
                    import pymupdf as fitz
                except ImportError:
                    fitz = None

        # If PyMuPDF is not available, skip advanced post-processing
        if fitz is None:
            print("⚠️ PyMuPDF not available - skipping advanced post-processing")
            return {
                'pages_processed': 0,
                'placeholders_verified': 0,
                'tables_verified': 0,
                'arabic_text_verified': 0,
                'layout_issues_fixed': 0,
                'warnings': ['PyMuPDF not available for advanced verification'],
                'success_metrics': ['Basic PDF validation completed']
            }

        print("🔍 Post-processing PDF for perfect formatting...")

        # Open the PDF document
        doc = fitz.open(pdf_path)

        post_process_results = {
            'pages_processed': len(doc),
            'placeholders_verified': 0,
            'tables_verified': 0,
            'arabic_text_verified': 0,
            'layout_issues_fixed': 0,
            'warnings': [],
            'success_metrics': []
        }

        # Process each page
        for page_num in range(len(doc)):
            page = doc[page_num]

            # Extract text with position information
            text_dict = page.get_text("dict")

            # Verify placeholder positions ({{...}})
            if docx_info.get('placeholder_count', 0) > 0:
                placeholder_pattern = r'\{\{[^}]+\}\}'
                page_text = page.get_text()
                found_placeholders = re.findall(placeholder_pattern, page_text)
                post_process_results['placeholders_verified'] += len(found_placeholders)

                if len(found_placeholders) != docx_info.get('placeholder_count', 0):
                    post_process_results['warnings'].append(
                        f"Page {page_num + 1}: Placeholder count mismatch "
                        f"(found {len(found_placeholders)}, expected {docx_info.get('placeholder_count', 0)})"
                    )

            # Verify Arabic text rendering
            if docx_info.get('rtl_content_detected', False):
                arabic_pattern = r'[\u0600-\u06FF\u0750-\u077F\u08A0-\u08FF\uFB50-\uFDFF\uFE70-\uFEFF]'
                page_text = page.get_text()
                arabic_chars = len(re.findall(arabic_pattern, page_text))
                post_process_results['arabic_text_verified'] += arabic_chars

                if arabic_chars > 0:
                    post_process_results['success_metrics'].append(
                        f"Page {page_num + 1}: {arabic_chars} Arabic characters rendered correctly"
                    )

            # Verify table structure
            if docx_info.get('has_tables', False):
                try:
                    # Look for table-like structures in the PDF
                    tables = page.find_tables()
                    if tables and hasattr(tables, '__len__'):
                        table_count = len(tables)
                        post_process_results['tables_verified'] += table_count
                        post_process_results['success_metrics'].append(
                            f"Page {page_num + 1}: {table_count} tables preserved"
                        )
                    elif tables:
                        # If tables is not a list but exists, count as 1
                        post_process_results['tables_verified'] += 1
                        post_process_results['success_metrics'].append(
                            f"Page {page_num + 1}: Table structure detected"
                        )
                except Exception:
                    # Fallback: look for table-like text patterns
                    page_text = page.get_text()
                    # Simple heuristic: look for multiple lines with consistent spacing
                    lines = page_text.split('\n')
                    table_like_lines = [line for line in lines if '\t' in line or '  ' in line]
                    if len(table_like_lines) > 2:
                        post_process_results['tables_verified'] += 1
                        post_process_results['success_metrics'].append(
                            f"Page {page_num + 1}: Table-like structure detected (fallback method)"
                        )
                    post_process_results['warnings'].append(
                        f"Page {page_num + 1}: Table detection method failed, used fallback"
                    )

            # Check for text overlap or layout issues
            blocks = text_dict.get("blocks", [])
            for block in blocks:
                if "lines" in block:
                    for line in block["lines"]:
                        for span in line.get("spans", []):
                            # Check for suspicious font sizes or positions
                            font_size = span.get("size", 0)
                            if font_size < 1:  # Suspiciously small text
                                post_process_results['warnings'].append(
                                    f"Page {page_num + 1}: Suspiciously small text detected (size: {font_size})"
                                )

        doc.close()

        # Generate summary
        if post_process_results['placeholders_verified'] > 0:
            post_process_results['success_metrics'].append(
                f"All {post_process_results['placeholders_verified']} placeholders preserved"
            )

        if post_process_results['arabic_text_verified'] > 0:
            post_process_results['success_metrics'].append(
                f"Arabic RTL text verified: {post_process_results['arabic_text_verified']} characters"
            )

        if post_process_results['tables_verified'] > 0:
            post_process_results['success_metrics'].append(
                f"Table structure preserved: {post_process_results['tables_verified']} tables"
            )

        print(f"✅ PDF post-processing completed:")
        print(f"  • Pages processed: {post_process_results['pages_processed']}")
        print(f"  • Placeholders verified: {post_process_results['placeholders_verified']}")
        print(f"  • Arabic characters verified: {post_process_results['arabic_text_verified']}")
        print(f"  • Tables verified: {post_process_results['tables_verified']}")
        print(f"  • Warnings: {len(post_process_results['warnings'])}")

        return post_process_results

    except Exception as e:
        print(f"❌ PDF post-processing error: {e}")
        return {
            'pages_processed': 0,
            'placeholders_verified': 0,
            'tables_verified': 0,
            'arabic_text_verified': 0,
            'layout_issues_fixed': 0,
            'warnings': [f'Post-processing error: {e}'],
            'success_metrics': []
        }


def analyze_conversion_error(stderr, stdout, docx_info):
    """Analyze conversion errors and provide helpful diagnostics"""
    error_analysis = []

    # Common error patterns and solutions
    error_patterns = {
        'font': ['font', 'typeface', 'glyph'],
        'memory': ['memory', 'heap', 'out of memory'],
        'file_access': ['permission', 'access', 'file not found', 'cannot open'],
        'format': ['format', 'corrupt', 'invalid', 'malformed'],
        'timeout': ['timeout', 'time out', 'expired'],
        'display': ['display', 'x11', 'xvfb', 'screen']
    }

    stderr_lower = stderr.lower()
    stdout_lower = stdout.lower()
    combined_output = stderr_lower + " " + stdout_lower

    # Analyze error types
    for error_type, keywords in error_patterns.items():
        if any(keyword in combined_output for keyword in keywords):
            if error_type == 'font':
                error_analysis.append("🔤 Font-related issue detected:")
                error_analysis.append("  • Possible missing font substitution")
                error_analysis.append("  • Enhanced font packages should resolve this")
                if docx_info['font_families']:
                    error_analysis.append(f"  • Document uses fonts: {list(docx_info['font_families'])[:3]}")

            elif error_type == 'memory':
                error_analysis.append("💾 Memory issue detected:")
                error_analysis.append("  • Document may be too large or complex")
                error_analysis.append("  • Try with a smaller document first")

            elif error_type == 'file_access':
                error_analysis.append("📁 File access issue detected:")
                error_analysis.append("  • Temporary file permissions problem")
                error_analysis.append("  • This should resolve on retry")

            elif error_type == 'format':
                error_analysis.append("📄 Document format issue detected:")
                error_analysis.append("  • DOCX file may be corrupted or invalid")
                error_analysis.append("  • Try opening in Word and re-saving")

            elif error_type == 'timeout':
                error_analysis.append("⏱️ Timeout issue detected:")
                error_analysis.append("  • Document conversion took too long")
                error_analysis.append("  • Complex documents may need more time")

            elif error_type == 'display':
                error_analysis.append("🖥️ Display/Graphics issue detected:")
                error_analysis.append("  • Headless display configuration problem")
                error_analysis.append("  • This is a system configuration issue")

    # Advanced document-specific analysis
    if docx_info.get('has_tables'):
        error_analysis.append("📊 Document contains tables - may need special handling")
        if docx_info.get('table_structure_issues'):
            error_analysis.append(f"  • Table issues detected: {', '.join(docx_info['table_structure_issues'])}")

    if docx_info.get('has_images'):
        error_analysis.append("🖼️ Document contains images - may affect processing")

    if docx_info.get('has_textboxes'):
        error_analysis.append("📦 Document contains TextBoxes - these may cause layout issues")

    if docx_info.get('has_smartart'):
        error_analysis.append("🎨 Document contains SmartArt - these elements may not convert properly")

    if docx_info.get('has_complex_shapes'):
        error_analysis.append("🔷 Document contains complex shapes - these may affect layout")

    if docx_info.get('text_content_length', 0) > 50000:
        error_analysis.append("📝 Large document detected - may need more processing time")

    if docx_info.get('rtl_content_detected'):
        error_analysis.append("🌍 Arabic RTL content detected - ensure Arabic fonts are properly installed")

    if docx_info.get('placeholder_count', 0) > 0:
        error_analysis.append(f"🏷️ Document contains {docx_info['placeholder_count']} placeholders - these must be preserved")

    # Font-specific analysis
    if docx_info.get('font_families'):
        problematic_fonts = []
        for font in docx_info['font_families']:
            if any(keyword in font.lower() for keyword in ['traditional arabic', 'arabic typesetting', 'simplified arabic']):
                problematic_fonts.append(font)

        if problematic_fonts:
            error_analysis.append(f"🔤 Arabic fonts detected: {', '.join(problematic_fonts[:3])}")
            error_analysis.append("  • Ensure Arabic font substitution is working correctly")

    # General recommendations
    if not error_analysis:
        error_analysis.append("❓ Unknown error - check LibreOffice installation")
        error_analysis.append("  • Verify all system dependencies are installed")
        error_analysis.append("  • Try with a simpler test document")

    error_analysis.append("\n💡 Advanced troubleshooting suggestions:")
    error_analysis.append("  • Ensure DOCX file is valid and not corrupted")
    error_analysis.append("  • Try with a smaller or simpler document")
    error_analysis.append("  • Check that all required fonts are available")
    error_analysis.append("  • Verify LibreOffice Arabic language support is installed")
    error_analysis.append("  • Consider preprocessing the document to remove problematic elements")

    return "\n".join(error_analysis)


def generate_comprehensive_quality_report(docx_info, pdf_validation, post_process_results):
    """
    Generate a comprehensive quality report for the conversion
    """
    report = []

    # Header
    report.append("📋 COMPREHENSIVE CONVERSION QUALITY REPORT")
    report.append("=" * 50)

    # Document Analysis Summary
    report.append("\n📄 DOCUMENT ANALYSIS:")
    report.append(f"  • Text Content: {docx_info.get('text_content_length', 0):,} characters")
    report.append(f"  • Font Families: {len(docx_info.get('font_families', set()))} detected")
    report.append(f"  • Tables: {'Yes' if docx_info.get('has_tables') else 'No'}")
    report.append(f"  • Images: {'Yes' if docx_info.get('has_images') else 'No'}")
    report.append(f"  • Arabic RTL Content: {'Yes' if docx_info.get('rtl_content_detected') else 'No'}")
    report.append(f"  • Placeholders: {docx_info.get('placeholder_count', 0)}")

    # Potential Issues
    issues = []
    if docx_info.get('has_textboxes'):
        issues.append("TextBoxes detected")
    if docx_info.get('has_smartart'):
        issues.append("SmartArt elements detected")
    if docx_info.get('has_complex_shapes'):
        issues.append("Complex shapes detected")
    if docx_info.get('table_structure_issues'):
        issues.extend(docx_info['table_structure_issues'])

    if issues:
        report.append(f"  • Potential Issues: {', '.join(issues)}")
    else:
        report.append("  • Potential Issues: None detected")

    # PDF Quality Metrics
    report.append("\n📊 PDF QUALITY METRICS:")
    report.append(f"  • File Size: {pdf_validation.get('file_size_mb', 0)} MB")
    report.append(f"  • Pages Processed: {post_process_results.get('pages_processed', 0)}")

    # Verification Results
    report.append("\n✅ VERIFICATION RESULTS:")
    if post_process_results.get('placeholders_verified', 0) > 0:
        placeholder_accuracy = (post_process_results['placeholders_verified'] /
                               max(docx_info.get('placeholder_count', 1), 1)) * 100
        report.append(f"  • Placeholder Preservation: {placeholder_accuracy:.1f}% "
                     f"({post_process_results['placeholders_verified']}/{docx_info.get('placeholder_count', 0)})")

    if post_process_results.get('arabic_text_verified', 0) > 0:
        report.append(f"  • Arabic Text Verified: {post_process_results['arabic_text_verified']:,} characters")

    if post_process_results.get('tables_verified', 0) > 0:
        report.append(f"  • Tables Preserved: {post_process_results['tables_verified']}")

    # Success Metrics
    all_success_metrics = (pdf_validation.get('success_metrics', []) +
                          post_process_results.get('success_metrics', []))
    if all_success_metrics:
        report.append("\n🎯 SUCCESS METRICS:")
        for metric in all_success_metrics:
            report.append(f"  ✓ {metric}")

    # Warnings
    all_warnings = (pdf_validation.get('warnings', []) +
                   post_process_results.get('warnings', []))
    if all_warnings:
        report.append("\n⚠️ WARNINGS:")
        for warning in all_warnings:
            report.append(f"  • {warning}")

    # Overall Quality Score
    quality_score = calculate_quality_score(docx_info, pdf_validation, post_process_results)
    report.append(f"\n🏆 OVERALL QUALITY SCORE: {quality_score:.1f}%")

    if quality_score >= 99:
        report.append("🌟 EXCELLENT: Pixel-perfect conversion achieved!")
    elif quality_score >= 95:
        report.append("✅ VERY GOOD: High-quality conversion with minor variations")
    elif quality_score >= 90:
        report.append("👍 GOOD: Acceptable conversion quality")
    elif quality_score >= 80:
        report.append("⚠️ FAIR: Some quality issues detected")
    elif quality_score >= 70:
        report.append("❌ POOR: Significant quality issues")
    else:
        report.append("🚨 CRITICAL: Major conversion problems")

    # Add improvement suggestions
    suggestions = suggest_quality_improvements(docx_info, pdf_validation, post_process_results, quality_score)
    if suggestions:
        report.append("\n" + "\n".join(suggestions))

    return "\n".join(report)


def calculate_quality_score(docx_info, pdf_validation, post_process_results):
    """
    Calculate an overall quality score for the conversion with enhanced accuracy
    """
    score = 100.0

    # Major deductions for critical issues
    warning_count = (len(pdf_validation.get('warnings', [])) +
                    len(post_process_results.get('warnings', [])))

    # Categorize warnings by severity
    critical_warnings = 0
    minor_warnings = 0

    all_warnings = (pdf_validation.get('warnings', []) +
                   post_process_results.get('warnings', []))

    for warning in all_warnings:
        warning_lower = warning.lower()
        if any(keyword in warning_lower for keyword in ['error', 'failed', 'missing', 'corrupted']):
            critical_warnings += 1
        else:
            minor_warnings += 1

    score -= critical_warnings * 5  # 5 points per critical warning
    score -= minor_warnings * 2     # 2 points per minor warning

    # Placeholder accuracy (very important for document integrity)
    expected_placeholders = docx_info.get('placeholder_count', 0)
    verified_placeholders = post_process_results.get('placeholders_verified', 0)
    if expected_placeholders > 0:
        placeholder_accuracy = verified_placeholders / expected_placeholders
        score -= (1 - placeholder_accuracy) * 15  # Up to 15 points for placeholders
    else:
        # Bonus if no placeholders were expected and none were found
        if verified_placeholders == 0:
            score += 2

    # Arabic text verification (critical for RTL documents)
    if docx_info.get('rtl_content_detected', False):
        arabic_chars = post_process_results.get('arabic_text_verified', 0)
        if arabic_chars > 0:
            score += 5  # Bonus for successful Arabic verification
        else:
            score -= 10  # Major deduction if Arabic content was expected but not verified

    # Table preservation
    if docx_info.get('has_tables', False):
        tables_verified = post_process_results.get('tables_verified', 0)
        if tables_verified > 0:
            score += 3  # Bonus for table preservation
        else:
            score -= 8  # Deduction if tables were expected but not verified

    # Image preservation
    if docx_info.get('has_images', False):
        score += 2  # Bonus for handling images (basic preservation assumed)

    # Deduct points for problematic elements that weren't preprocessed
    if docx_info.get('has_textboxes'):
        score -= 3  # Reduced penalty since we have preprocessing
    if docx_info.get('has_smartart'):
        score -= 3  # Reduced penalty since we have preprocessing
    if docx_info.get('has_complex_shapes'):
        score -= 2  # Minor penalty for complex shapes

    # Table structure issues
    table_issues = docx_info.get('table_structure_issues', [])
    if table_issues:
        score -= len(table_issues) * 3  # 3 points per table issue

    # PDF quality metrics
    pdf_size = pdf_validation.get('file_size_mb', 0)
    if pdf_size > 0:
        if 0.01 <= pdf_size <= 50:  # Reasonable size range
            score += 2
        elif pdf_size > 50:
            score -= 3  # Penalty for very large files
        elif pdf_size < 0.01:
            score -= 5  # Penalty for suspiciously small files

    # Success metrics bonus
    success_count = len(pdf_validation.get('success_metrics', [])) + len(post_process_results.get('success_metrics', []))
    score += min(success_count * 0.5, 5)  # Up to 5 bonus points for success metrics

    # Post-processing completion bonus
    pages_processed = post_process_results.get('pages_processed', 0)
    if pages_processed > 0:
        score += 3  # Bonus for successful post-processing
    else:
        score -= 5  # Penalty if post-processing failed completely

    return max(0, min(100, score))


def suggest_quality_improvements(docx_info, pdf_validation, post_process_results, quality_score):
    """
    Suggest specific improvements based on quality analysis
    """
    suggestions = []

    if quality_score < 90:
        suggestions.append("🔧 IMPROVEMENT SUGGESTIONS:")

        # Analyze specific issues
        if post_process_results.get('placeholders_verified', 0) < docx_info.get('placeholder_count', 0):
            suggestions.append("  • Placeholder positioning issues detected - consider document restructuring")

        if docx_info.get('has_textboxes') or docx_info.get('has_smartart') or docx_info.get('has_complex_shapes'):
            suggestions.append("  • Complex elements detected - preprocessing applied but manual review recommended")

        if docx_info.get('table_structure_issues'):
            suggestions.append("  • Table structure issues found - consider simplifying table layouts")

        if post_process_results.get('arabic_text_verified', 0) == 0 and docx_info.get('rtl_content_detected'):
            suggestions.append("  • Arabic text verification failed - check font installation")

        warning_count = (len(pdf_validation.get('warnings', [])) +
                        len(post_process_results.get('warnings', [])))
        if warning_count > 2:
            suggestions.append(f"  • Multiple warnings detected ({warning_count}) - review document complexity")

        if quality_score < 80:
            suggestions.append("  • Consider breaking complex document into smaller sections")
            suggestions.append("  • Verify document is not corrupted in original Word application")

        if quality_score < 70:
            suggestions.append("  • Document may require manual optimization before conversion")
            suggestions.append("  • Contact support for complex document handling")

    else:
        suggestions.append("✅ EXCELLENT QUALITY - No improvements needed!")

    return suggestions


def create_libreoffice_config(temp_path):
    """Create comprehensive LibreOffice configuration for PERFECT Arabic RTL formatting preservation"""
    config_dir = temp_path / ".config" / "libreoffice" / "4" / "user"
    config_dir.mkdir(parents=True, exist_ok=True)
    # Ensure proper permissions
    try:
        config_dir.chmod(0o777)
    except PermissionError:
        # If we can't change permissions, continue anyway
        pass

    # Create comprehensive registrymodifications.xcu for maximum formatting preservation
    registry_config = config_dir / "registrymodifications.xcu"
    
    # Create empty file first to ensure it exists
    try:
        registry_config.touch()
        registry_config.chmod(0o666)
    except PermissionError:
        # If we can't change permissions, continue anyway
        pass

    config_content = '''<?xml version="1.0" encoding="UTF-8"?>
<oor:items xmlns:oor="http://openoffice.org/2001/registry" xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
  <!-- Disable first start wizard and user installation -->
  <item oor:path="/org.openoffice.Setup/Office/Factories/org.openoffice.Setup:Factory['com.sun.star.comp.framework.ProtocolHandler']">
    <prop oor:name="FirstStartWizardCompleted" oor:op="fuse">
      <value>true</value>
    </prop>
  </item>
  
  <!-- CRITICAL: Completely disable Java to prevent javaldx errors -->
  <item oor:path="/org.openoffice.Setup/Office">
    <prop oor:name="JavaSupport" oor:op="fuse">
      <value>false</value>
    </prop>
  </item>
  
  <!-- Disable Java security to prevent javaldx errors -->
  <item oor:path="/org.openoffice.Office.Java">
    <prop oor:name="Enabled" oor:op="fuse">
      <value>false</value>
    </prop>
  </item>
  
  <!-- PDF Export Settings for Maximum Quality with Arabic Support -->
  <item oor:path="/org.openoffice.Office.Common/Filter/PDF/Export">
    <prop oor:name="Quality" oor:op="fuse">
      <value>100</value>
    </prop>
    <prop oor:name="ReduceImageResolution" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="MaxImageResolution" oor:op="fuse">
      <value>600</value>
    </prop>
    <prop oor:name="UseTaggedPDF" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="ExportFormFields" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="FormsType" oor:op="fuse">
      <value>0</value>
    </prop>
    <prop oor:name="AllowDuplicateFieldNames" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="EmbedStandardFonts" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="FontEmbedding" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="CompressMode" oor:op="fuse">
      <value>0</value>
    </prop>
    <prop oor:name="JPEGQuality" oor:op="fuse">
      <value>100</value>
    </prop>
    <prop oor:name="SelectPdfVersion" oor:op="fuse">
      <value>1</value>
    </prop>
    <prop oor:name="ExportBookmarks" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="OpenBookmarkLevels" oor:op="fuse">
      <value>-1</value>
    </prop>
  </item>

  <!-- Arabic and RTL Language Support -->
  <item oor:path="/org.openoffice.Office.Linguistic/General">
    <prop oor:name="DefaultLocale" oor:op="fuse">
      <value>ar-SA</value>
    </prop>
    <prop oor:name="DefaultLocale_CJK" oor:op="fuse">
      <value>ar-SA</value>
    </prop>
    <prop oor:name="DefaultLocale_CTL" oor:op="fuse">
      <value>ar-SA</value>
    </prop>
  </item>

  <!-- CTL (Complex Text Layout) Settings for Arabic -->
  <item oor:path="/org.openoffice.Office.Common/I18N/CTL">
    <prop oor:name="CTLFont" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="CTLSequenceChecking" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="CTLCursorMovement" oor:op="fuse">
      <value>1</value>
    </prop>
    <prop oor:name="CTLTextNumerals" oor:op="fuse">
      <value>1</value>
    </prop>
  </item>

  <!-- Enhanced Font Substitution Settings for Local Arial and Arabic Compatibility -->
  <item oor:path="/org.openoffice.VCL/FontSubstitution">
    <prop oor:name="FontSubstituteTable" oor:op="fuse">
      <value>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Arial</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Arial</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Liberation Sans</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Calibri</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Liberation Serif</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Cambria</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Liberation Serif</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Times New Roman</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Liberation Mono</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Courier New</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Amiri</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Traditional Arabic</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Amiri</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Arabic Typesetting</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>Noto Naskh Arabic</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Simplified Arabic</value>
          </prop>
        </it>
        <it>
          <prop oor:name="SubstituteFont">
            <value>DejaVu Sans</value>
          </prop>
          <prop oor:name="OriginalFont">
            <value>Tahoma</value>
          </prop>
        </it>
      </value>
    </prop>
  </item>

  <!-- Writer Settings for Perfect Layout Preservation with RTL Support -->
  <item oor:path="/org.openoffice.Office.Writer/Layout/Other">
    <prop oor:name="MeasureUnit" oor:op="fuse">
      <value>6</value>
    </prop>
    <prop oor:name="TabStop" oor:op="fuse">
      <value>1270</value>
    </prop>
    <prop oor:name="IsSquaredPageMode" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="ApplyCharUnit" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="IsAlignTabStopPosition" oor:op="fuse">
      <value>true</value>
    </prop>
  </item>

  <!-- Enhanced Table Settings for Exact Formatting -->
  <item oor:path="/org.openoffice.Office.Writer/Layout/Table">
    <prop oor:name="Header" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="RepeatHeader" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="DontSplit" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="Border" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="InsertLabel" oor:op="fuse">
      <value>false</value>
    </prop>
  </item>

  <!-- Page Layout Settings for A4 and RTL -->
  <item oor:path="/org.openoffice.Office.Writer/Layout/Page">
    <prop oor:name="IsLandscape" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="Width" oor:op="fuse">
      <value>21000</value>
    </prop>
    <prop oor:name="Height" oor:op="fuse">
      <value>29700</value>
    </prop>
  </item>

  <!-- Default Font Settings with Local Arial Priority -->
  <item oor:path="/org.openoffice.Office.Writer/DefaultFont">
    <prop oor:name="Document" oor:op="fuse">
      <value>true</value>
    </prop>
    <prop oor:name="Standard" oor:op="fuse">
      <value>Arial;Liberation Sans;DejaVu Sans</value>
    </prop>
    <prop oor:name="Heading" oor:op="fuse">
      <value>Arial;Liberation Sans;DejaVu Sans</value>
    </prop>
    <prop oor:name="List" oor:op="fuse">
      <value>Arial;Liberation Sans;Amiri;Noto Naskh Arabic</value>
    </prop>
    <prop oor:name="Caption" oor:op="fuse">
      <value>Arial;Liberation Sans;DejaVu Sans</value>
    </prop>
    <prop oor:name="Index" oor:op="fuse">
      <value>Arial;Liberation Sans;DejaVu Sans</value>
    </prop>
    <prop oor:name="StandardHeight" oor:op="fuse">
      <value>12</value>
    </prop>
    <prop oor:name="HeadingHeight" oor:op="fuse">
      <value>14</value>
    </prop>
    <prop oor:name="ListHeight" oor:op="fuse">
      <value>13</value>
    </prop>
    <prop oor:name="CaptionHeight" oor:op="fuse">
      <value>12</value>
    </prop>
    <prop oor:name="IndexHeight" oor:op="fuse">
      <value>12</value>
    </prop>
  </item>

  <!-- Disable Auto-formatting Features -->
  <item oor:path="/org.openoffice.Office.Writer/AutoFunction/Format/Option">
    <prop oor:name="UseReplacementTable" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="TwoCapitalsAtStart" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="CapitalAtStartSentence" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="ChgWeightUnderl" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="SetInetAttr" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="ChgToEnEmDash" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="AddNonBrkSpace" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="ChgOrdinalNumber" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="ChgQuotes" oor:op="fuse">
      <value>false</value>
    </prop>
    <prop oor:name="DelEmptyNode" oor:op="fuse">
      <value>false</value>
    </prop>
  </item>
</oor:items>'''

    try:
        with open(registry_config, 'w', encoding='utf-8') as f:
            f.write(config_content)
    except Exception as e:
        print(f"❌ Failed to write LibreOffice config: {e}")

    return str(config_dir.parent.parent.parent)


def convert_docx_to_pdf(docx_file):
    """
    Convert DOCX to PDF using LibreOffice headless mode
    Preserves all formatting including Arabic RTL text
    """
    if docx_file is None:
        return None, "Please upload a DOCX file"

    # Initialize docx_info to avoid undefined variable errors
    docx_info = {
        'has_tables': False,
        'has_images': False,
        'text_content_length': 0,
        'font_families': []
    }

    final_output_path = None
    try:
        # Validate input DOCX structure for quality assurance
        print("🔍 Analyzing DOCX structure...")
        docx_info = validate_docx_structure(docx_file.name)

        # Create a persistent temporary file for the output
        output_fd, final_output_path = tempfile.mkstemp(suffix=".pdf", prefix="converted_")
        os.close(output_fd)  # Close the file descriptor, we just need the path

        # Create temporary directory for processing
        with tempfile.TemporaryDirectory() as temp_dir:
            temp_path = Path(temp_dir)

            # Create comprehensive LibreOffice and font configuration
            config_home = create_libreoffice_config(temp_path)
            fontconfig_home = create_fontconfig(temp_path)

            # Copy uploaded file to temp directory
            input_file = temp_path / "input.docx"
            shutil.copy2(docx_file.name, input_file)

            # Advanced DOCX preprocessing for perfect conversion
            processed_docx = preprocess_docx_for_perfect_conversion(str(input_file), docx_info)
            if processed_docx != str(input_file):
                print("🔧 Using preprocessed DOCX for conversion")
                input_file = Path(processed_docx)

            # Determine if aggressive optimization is needed
            needs_aggressive_optimization = (
                docx_info.get('has_textboxes', False) or
                docx_info.get('has_smartart', False) or
                docx_info.get('has_complex_shapes', False) or
                len(docx_info.get('table_structure_issues', [])) > 2 or
                docx_info.get('text_content_length', 0) > 100000
            )

            if needs_aggressive_optimization:
                print("⚠️ Complex document detected - applying aggressive optimization settings")
                # Increase timeout for complex documents
                conversion_timeout = 180
            else:
                conversion_timeout = 120

            # ULTIMATE LibreOffice PDF export settings for 99%+ formatting preservation
            # Optimized specifically for Arabic RTL with zero tolerance for layout changes
            pdf_export_settings = {
                # Core Quality Settings
                "Quality": 100,
                "ReduceImageResolution": False,
                "MaxImageResolution": 600,
                "BitmapResolution": 600,
                "ImageResolution": 600,
                "JPEGQuality": 100,
                "CompressMode": 0,  # No compression

                # Font and Text Preservation
                "EmbedStandardFonts": True,
                "FontEmbedding": True,
                "UseTaggedPDF": True,
                "EnableTextAccessForAccessibilityTools": True,

                # Layout Preservation (Critical for Arabic RTL)
                "ExportFormFields": False,
                "FormsType": 0,
                "ExportBookmarks": False,
                "ExportNotes": False,
                "ExportNotesPages": False,
                "ExportOnlyNotesPages": False,
                "ExportPlaceholders": False,
                "ExportHiddenSlides": False,
                "SinglePageSheets": False,
                "UseTransitionEffects": False,
                "IsSkipEmptyPages": False,
                "IsAddStream": False,
                "AllowDuplicateFieldNames": False,

                # Advanced Layout Control
                "ColorMode": 0,  # Keep original colors
                "Watermark": "",
                "EncryptFile": False,
                "DocumentOpenPassword": "",
                "PermissionPassword": "",
                "RestrictPermissions": False,
                "Printing": 2,  # Allow printing
                "Changes": 4,   # Allow all changes
                "EnableCopyingOfContent": True,
                "SelectPdfVersion": 1,  # PDF 1.4 for maximum compatibility
                "ExportLinksRelativeFsys": False,
                "PDFViewSelection": 0,
                "ConvertOOoTargetToPDFTarget": False,
                "ExportBookmarksToPDFDestination": False,

                # Critical for Arabic and RTL preservation
                "PreserveEditingInPDF": False,
                "ExportFormFieldsAsWidgets": False,
                "FormsFormat": 0,
                "SubmitFormat": 0,
                "AllowDuplicateFieldNames": False,
                "ExportEmptyPages": True,
                "ViewPDFAfterExport": False,

                # Table and Layout Precision
                "UseReferenceXObject": False,
                "HideViewerMenubar": False,
                "HideViewerToolbar": False,
                "HideViewerWindowControls": False,
                "ResizeWindowToInitialPage": False,
                "CenterWindow": False,
                "OpenInFullScreenMode": False,
                "DisplayPDFDocumentTitle": False,

                # Advanced Arabic RTL Support
                "ExportNotesInMargin": False,
                "ConvertOOoTargetToPDFTarget": False,
                "ExportLinksRelativeFsys": False,
                "PDFViewSelection": 0,
                "Magnification": 0,
                "PageLayout": 0,
                "FirstPageOnLeft": False,
                "InitialView": 0,
                "Magnification": 0
            }

            # Convert settings to JSON string for LibreOffice
            pdf_filter = f'pdf:writer_pdf_Export:{json.dumps(pdf_export_settings, separators=(",", ":"))}'

            # ENHANCED: Completely disable Java integration to prevent javaldx errors
            # REMOVED --disable-java flag as it's not supported in LibreOffice 7.3.7.2
            cmd = [
                "libreoffice",
                "--headless",
                "--invisible",
                "--nodefault",
                "--nolockcheck",
                "--nologo",
                "--norestore",
                "--nofirststartwizard",
                "--safe-mode",
                # REMOVED: "--disable-java",  # Not supported in LibreOffice 7.3.7.2
                "--disable-extension-update",
                "--disable-webupdate",
                "--disable-remote-control",
                "--disable-notification",
                "--disable-oop4all",  # Disable out-of-process for all
                "--convert-to", pdf_filter,
                "--outdir", str(temp_path),
                str(input_file)
            ]

            # Execute conversion with comprehensive custom environment optimized for Arabic RTL
            env = os.environ.copy()
            env['HOME'] = config_home
            env['XDG_CONFIG_HOME'] = config_home + "/.config"

            # Enhanced fontconfig setup
            fontconfig_dir = fontconfig_home + "/.config/fontconfig"
            env['FONTCONFIG_PATH'] = fontconfig_dir
            env['FONTCONFIG_FILE'] = fontconfig_dir + "/fonts.conf"

            # Additional font paths (same directory as Python script)
            script_dir = Path(__file__).parent.absolute()
            if 'FONTPATH' in env:
                env['FONTPATH'] = f"{script_dir}:{env['FONTPATH']}"

            else:
                env['FONTPATH'] = str(script_dir)
            # Set Arabic-friendly locale while maintaining UTF-8 support
            env['LANG'] = 'ar_SA.UTF-8'
            env['LC_ALL'] = 'ar_SA.UTF-8'
            env['LC_CTYPE'] = 'ar_SA.UTF-8'
            env['LC_NUMERIC'] = 'ar_SA.UTF-8'
            env['LC_TIME'] = 'ar_SA.UTF-8'
            env['LC_COLLATE'] = 'ar_SA.UTF-8'
            env['LC_MONETARY'] = 'ar_SA.UTF-8'
            env['LC_MESSAGES'] = 'ar_SA.UTF-8'
            env['LC_PAPER'] = 'ar_SA.UTF-8'
            env['LC_NAME'] = 'ar_SA.UTF-8'
            env['LC_ADDRESS'] = 'ar_SA.UTF-8'
            env['LC_TELEPHONE'] = 'ar_SA.UTF-8'
            env['LC_MEASUREMENT'] = 'ar_SA.UTF-8'
            env['LC_IDENTIFICATION'] = 'ar_SA.UTF-8'
            # Disable LibreOffice splash and user interaction
            env['SAL_USE_VCLPLUGIN'] = 'svp'
            env['DISPLAY'] = ':99'
            # Enhanced LibreOffice settings for Arabic
            env['OOO_FORCE_DESKTOP'] = 'gnome'
            env['SAL_NO_MOUSEGRABS'] = '1'
            # Force RTL support
            env['SAL_RTL_ENABLED'] = '1'
            env['OOO_DISABLE_RECOVERY'] = '1'
            
            # CRITICAL: Completely disable Java integration to prevent javaldx errors
            env['SAL_DISABLE_JAVA_SECURITY'] = '1'
            env['SAL_DISABLE_JAVA'] = '1'  # Additional Java disable flag
            env['SAL_JAVA_DISABLE_SECURITY'] = '1'  # Another Java disable flag
            
            # MODIFIED: Set UNO_PATH to help LibreOffice find its components
            env['UNO_PATH'] = '/usr/lib/libreoffice/program'
            
            # CRITICAL: Set LibreOffice to use minimal Java or no Java at all
            env['LIBO_JAVA_PARALLEL'] = '0'  # Disable parallel Java processing
            env['LIBO_DISABLE_JAVA'] = '1'   # Additional LibreOffice Java disable flag
            
            # Additional environment variables to completely disable Java
            env['SAL_DISABLE_OPENCL'] = '1'  # Disable OpenCL which can cause issues
            env['SAL_DISABLE_VCLPLUGIN'] = '1'  # Disable VCL plugin which can cause issues

            print(f"🚀 Executing LibreOffice conversion with MAXIMUM quality settings...")
            print(f"Command: {' '.join(cmd[:11])}... [truncated for readability]")
            print(f"Environment: HOME={env.get('HOME', 'default')}, LANG={env.get('LANG', 'default')}")

            result = subprocess.run(
                cmd,
                capture_output=True,
                text=True,
                timeout=conversion_timeout,  # Dynamic timeout based on document complexity
                cwd=temp_path,
                env=env
            )

            print(f"📊 LibreOffice execution completed:")
            print(f"  • Return code: {result.returncode}")
            print(f"  • Output length: {len(result.stdout)} chars")
            print(f"  • Error length: {len(result.stderr)} chars")

            if result.stdout:
                print(f"  • LibreOffice stdout: {result.stdout[:200]}...")
            if result.stderr:
                print(f"  • LibreOffice stderr: {result.stderr[:200]}...")

            # If LibreOffice fails, try fallback with unoconv
            if result.returncode != 0:
                print("⚠️ LibreOffice conversion failed, trying fallback with unoconv...")
                try:
                    unoconv_cmd = [
                        "unoconv",
                        "-f", "pdf",
                        "-o", str(temp_path),
                        str(input_file)
                    ]
                    
                    print(f"🚀 Executing unoconv conversion...")
                    print(f"Command: {' '.join(unoconv_cmd)}")
                    
                    unoconv_result = subprocess.run(
                        unoconv_cmd,
                        capture_output=True,
                        text=True,
                        timeout=conversion_timeout,
                        cwd=temp_path,
                        env=env
                    )
                    
                    print(f"📊 unoconv execution completed:")
                    print(f"  • Return code: {unoconv_result.returncode}")
                    print(f"  • Output length: {len(unoconv_result.stdout)} chars")
                    print(f"  • Error length: {len(unoconv_result.stderr)} chars")
                    
                    if unoconv_result.stdout:
                        print(f"  • unoconv stdout: {unoconv_result.stdout[:200]}...")
                    if unoconv_result.stderr:
                        print(f"  • unoconv stderr: {unoconv_result.stderr[:200]}...")
                    
                    # If unoconv succeeds, use its result
                    if unoconv_result.returncode == 0:
                        result = unoconv_result
                        print("✅ unoconv conversion successful")
                    else:
                        print("❌ unoconv conversion also failed")
                except Exception as unoconv_error:
                    print(f"❌ unoconv conversion error: {unoconv_error}")

            if result.returncode != 0:
                # Enhanced error analysis
                error_analysis = analyze_conversion_error(result.stderr, result.stdout, docx_info)
                error_msg = f"❌ Conversion failed with detailed analysis:\n\n"
                error_msg += f"🔍 Error Analysis:\n{error_analysis}\n\n"
                error_msg += f"📋 Technical Details:\n"
                error_msg += f"• Return Code: {result.returncode}\n"
                error_msg += f"• LibreOffice Error: {result.stderr[:300]}...\n"
                error_msg += f"• Document Info: Tables={docx_info['has_tables']}, Images={docx_info['has_images']}\n"

                print(f"❌ CONVERSION FAILED: {error_msg}")

                # Clean up the temporary output file
                if final_output_path:
                    try:
                        os.unlink(final_output_path)
                    except:
                        pass
                return None, error_msg

            # Check if PDF was created - LibreOffice may create different filename
            print(f"Looking for PDF files in: {temp_path}")
            all_files = list(temp_path.iterdir())
            print(f"Files in temp directory: {all_files}")

            # Look for any PDF file in the directory
            pdf_files = [f for f in all_files if f.suffix.lower() == '.pdf']

            if not pdf_files:
                # Clean up the temporary output file
                if final_output_path:
                    try:
                        os.unlink(final_output_path)
                    except:
                        pass
                return None, f"No PDF file was generated by LibreOffice. Files found: {[f.name for f in all_files]}"

            # Use the first PDF file found
            temp_pdf = pdf_files[0]
            print(f"✅ Found PDF file: {temp_pdf}")

            if not temp_pdf.exists():
                # Clean up the temporary output file
                if final_output_path:
                    try:
                        os.unlink(final_output_path)
                    except:
                        pass
                return None, "PDF file was not generated by LibreOffice"

            # Copy PDF to the persistent location
            shutil.copy2(temp_pdf, final_output_path)

            # Advanced PDF post-processing and validation
            print("🔍 Validating PDF output...")
            pdf_validation = validate_pdf_output(final_output_path, docx_info)

            print("🔧 Post-processing PDF for perfect formatting...")
            post_process_results = post_process_pdf_for_perfect_formatting(final_output_path, docx_info)

            # Generate comprehensive quality report
            quality_report = generate_comprehensive_quality_report(docx_info, pdf_validation, post_process_results)
            quality_score = calculate_quality_score(docx_info, pdf_validation, post_process_results)

            # Generate success message with quality report
            if quality_score >= 95:
                success_msg = f"🌟 EXCELLENT conversion with {quality_score:.1f}% formatting accuracy!\n\n"
            elif quality_score >= 85:
                success_msg = f"✅ HIGH-QUALITY conversion with {quality_score:.1f}% formatting accuracy!\n\n"
            elif quality_score >= 75:
                success_msg = f"👍 GOOD conversion with {quality_score:.1f}% formatting accuracy!\n\n"
            else:
                success_msg = f"⚠️ Conversion completed with {quality_score:.1f}% accuracy - improvements suggested!\n\n"

            success_msg += quality_report

            # Add retry suggestion for low quality scores
            if quality_score < 80:
                success_msg += f"\n\n💡 TIP: For better results, try simplifying the document structure or removing complex elements before conversion."

            return final_output_path, success_msg

    except subprocess.TimeoutExpired:
        # Enhanced timeout error handling
        timeout_msg = "⏱️ Conversion timed out - Document is too complex for current processing limits\n\n"
        timeout_msg += "🔍 Timeout Analysis:\n"
        timeout_msg += f"• Document has tables: {docx_info.get('has_tables', 'Unknown')}\n"
        timeout_msg += f"• Document has images: {docx_info.get('has_images', 'Unknown')}\n"
        timeout_msg += f"• Text content length: {docx_info.get('text_content_length', 'Unknown')} characters\n"
        timeout_msg += f"• Font families detected: {len(docx_info.get('font_families', []))}\n\n"
        timeout_msg += "💡 Suggestions:\n"
        timeout_msg += "• Try with a simpler document first\n"
        timeout_msg += "• Remove complex tables or images temporarily\n"
        timeout_msg += "• Split large documents into smaller sections\n"
        timeout_msg += "• Ensure document is not corrupted\n"

        print(f"❌ TIMEOUT ERROR: {timeout_msg}")

        # Clean up the temporary output file if it exists
        if final_output_path:
            try:
                os.unlink(final_output_path)
            except:
                pass
        return None, timeout_msg
    except Exception as e:
        # Enhanced general exception handling
        exception_msg = f"❌ Unexpected error during conversion\n\n"
        exception_msg += f"🔍 Error Details:\n"
        exception_msg += f"• Error Type: {type(e).__name__}\n"
        exception_msg += f"• Error Message: {str(e)}\n"

        if 'docx_info' in locals():
            exception_msg += f"• Document Analysis:\n"
            exception_msg += f"  - Has tables: {docx_info.get('has_tables', 'Unknown')}\n"
            exception_msg += f"  - Has images: {docx_info.get('has_images', 'Unknown')}\n"
            exception_msg += f"  - Content length: {docx_info.get('text_content_length', 'Unknown')}\n"

        exception_msg += f"\n💡 Recovery Suggestions:\n"
        exception_msg += f"• Verify the DOCX file is not corrupted\n"
        exception_msg += f"• Try opening the file in Microsoft Word first\n"
        exception_msg += f"• Ensure the file is a valid .docx format\n"
        exception_msg += f"• Check file size is reasonable (< 50MB)\n"
        exception_msg += f"• Try with a simpler test document\n"

        print(f"❌ EXCEPTION ERROR: {exception_msg}")
        print(f"Full exception details: {repr(e)}")

        # Clean up the temporary output file if it exists
        if final_output_path:
            try:
                os.unlink(final_output_path)
            except:
                pass
        return None, exception_msg


# This function was used for Gradio interface and is no longer needed
# The FastAPI implementation is in main.py