#!/bin/bash

# Enhanced Website to PDF Converter with Browser Support
# Supports both web crawling and local HTML processing with Cloudflare bypass

# set -euo pipefail

# Configuration
SCRIPT_NAME="$(basename "$0")"
readonly VERSION="2.1"
MAX_DEPTH=2
MAX_PAGES=100
readonly WEASYPRINT_TIMEOUT=120
readonly BROWSER_TIMEOUT=60

# Color codes for output
readonly RED="\033[0;31m"
readonly GREEN="\033[0;32m"
readonly YELLOW="\033[1;33m"
readonly BLUE="\033[0;34m"
readonly NC="\033[0m" # No Color

# Global variables
TEMP_DIR=""
KEEP_TEMP=0
VERBOSE=0
MODE=""
SOURCE=""
OUTPUT_PDF=""
USE_BROWSER=0

# Logging functions
log_info() {
    echo -e "${BLUE}[INFO]${NC} $1"
}

log_success() {
    echo -e "${GREEN}[SUCCESS]${NC} $1"
}

log_warning() {
    echo -e "${YELLOW}[WARNING]${NC} $1"
}

log_error() {
    echo -e "${RED}[ERROR]${NC} $1" >&2
}

log_verbose() {
    if [ "$VERBOSE" -eq 1 ]; then
        echo -e "${BLUE}[VERBOSE]${NC} $1"
    fi
}

# Help function
show_help() {
    cat << EOF
$SCRIPT_NAME v$VERSION - Enhanced Website to PDF Converter with Browser Support

USAGE:
    $SCRIPT_NAME web <start_url> <output_pdf> [options]
    $SCRIPT_NAME local <html_directory> <output_pdf> [options]

MODES:
    web             Crawl website starting from URL
    local           Process locally saved HTML files

ARGUMENTS:
    start_url       Starting URL for web crawling
    html_directory  Directory containing HTML files for local processing
    output_pdf      Output PDF file path

OPTIONS:
    --keep-temp     Keep temporary files for debugging
    --verbose       Enable verbose output
    --max-depth N   Maximum crawling depth (default: $MAX_DEPTH)
    --max-pages N   Maximum pages to crawl (default: $MAX_PAGES)
    --browser       Use browser automation to bypass Cloudflare protection
    --help          Show this help message

EXAMPLES:
    $SCRIPT_NAME web https://example.com/docs docs.pdf --verbose
    $SCRIPT_NAME local ./saved_website/ website.pdf --keep-temp
    $SCRIPT_NAME web https://blog.exploit.org/caster-legless/ article.pdf --browser --verbose

DEPENDENCIES:
    curl, weasyprint, pdfunite (poppler-utils)
    For --browser mode: python3, selenium, chromium-browser

EOF
}

# Cleanup function
cleanup_on_signal() {
    local exit_code=$?
    log_warning "Script interrupted, cleaning up..."
    if [ -n "$TEMP_DIR" ] && [ -d "$TEMP_DIR" ]; then
        rm -rf "$TEMP_DIR"
        log_verbose "Cleaned up temporary directory: $TEMP_DIR"
    fi
    exit $exit_code
}

# Manual cleanup function
cleanup_temp() {
    if [ -n "$TEMP_DIR" ] && [ -d "$TEMP_DIR" ]; then
        if [ "$KEEP_TEMP" -eq 0 ]; then
            rm -rf "$TEMP_DIR"
            log_verbose "Cleaned up temporary directory: $TEMP_DIR"
        else
            log_info "Temporary directory preserved: $TEMP_DIR"
        fi
    fi
}

# Set up signal handlers for interruption only
trap cleanup_on_signal INT TERM

# Dependency check
check_dependencies() {
    local missing_deps=()

    if [ "$MODE" = "web" ] && ! command -v curl &> /dev/null; then
        missing_deps+=("curl")
    fi

    if ! command -v weasyprint &> /dev/null; then
        missing_deps+=("weasyprint")
    fi

    if ! command -v pdfunite &> /dev/null; then
        missing_deps+=("pdfunite (poppler-utils)")
    fi

    if [ "$USE_BROWSER" -eq 1 ]; then
        if ! command -v python3 &> /dev/null; then
            missing_deps+=("python3")
        fi
        
        # Check if selenium is available
        if ! python3 -c "import selenium" 2>/dev/null; then
            missing_deps+=("python3-selenium")
        fi
    fi

    if [ ${#missing_deps[@]} -gt 0 ]; then
        log_error "Missing dependencies: ${missing_deps[*]}"
        log_error "Please install the required packages and try again."
        if [ "$USE_BROWSER" -eq 1 ]; then
            log_error "For browser mode, install: pip3 install selenium"
        fi
        exit 1
    fi
}

# Parse command line arguments
parse_arguments() {
    if [ $# -lt 3 ]; then
        show_help
        exit 1
    fi

    MODE="$1"
    SOURCE="$2"
    OUTPUT_PDF="$3"
    shift 3

    # Validate mode
    if [ "$MODE" != "web" ] && [ "$MODE" != "local" ]; then
        log_error "Invalid mode: $MODE. Use \'web\' or \'local\'"
        exit 1
    fi

    # Parse options
    while [ $# -gt 0 ]; do
        case "$1" in
            --keep-temp)
                KEEP_TEMP=1
                ;;
            --verbose)
                VERBOSE=1
                ;;
            --browser)
                USE_BROWSER=1
                ;;
            --max-depth)
                if [ $# -lt 2 ] || ! [[ "$2" =~ ^[0-9]+$ ]]; then
                    log_error "--max-depth requires a numeric argument"
                    exit 1
                fi
                MAX_DEPTH="$2"
                shift
                ;;
            --max-pages)
                if [ $# -lt 2 ] || ! [[ "$2" =~ ^[0-9]+$ ]]; then
                    log_error "--max-pages requires a numeric argument"
                    exit 1
                fi
                MAX_PAGES="$2"
                shift
                ;;
            --help)
                show_help
                exit 0
                ;;
            *)
                log_error "Unknown option: $1"
                exit 1
                ;;
        esac
        shift
    done
}

# Validate inputs
validate_inputs() {
    if [ "$MODE" = "web" ]; then
        if ! [[ "$SOURCE" =~ ^https?:// ]]; then
            log_error "Invalid URL format: $SOURCE"
            exit 1
        fi
    elif [ "$MODE" = "local" ]; then
        if [ ! -d "$SOURCE" ]; then
            log_error "Directory not found: $SOURCE"
            exit 1
        fi
        SOURCE="$(realpath "$SOURCE")"
    fi

    # Check if output directory exists
    local output_dir="$(dirname "$OUTPUT_PDF")"
    if [ ! -d "$output_dir" ]; then
        log_error "Output directory does not exist: $output_dir"
        exit 1
    fi
}

# Initialize working environment
init_environment() {
    TEMP_DIR=$(mktemp -d)
    log_verbose "Created temporary directory: $TEMP_DIR"

    # Create log files
    mkdir -p "$TEMP_DIR/logs"
    touch "$TEMP_DIR/logs/crawl_errors.log"
    touch "$TEMP_DIR/logs/weasyprint_errors.log"
    touch "$TEMP_DIR/logs/processing.log"
    touch "$TEMP_DIR/logs/browser_errors.log"
}

# Browser-based page fetcher
fetch_page_with_browser() {
    local url="$1"
    local output_file="$2"
    
    log_verbose "Fetching $url with browser automation"
    
    # Create Python script for browser automation
    cat > "$TEMP_DIR/browser_fetch.py" << EOF
import sys
import time
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
from selenium.common.exceptions import TimeoutException, WebDriverException

def fetch_page(url, output_file, timeout=60):
    chrome_options = Options()
    chrome_options.add_argument(\'--headless\')
    chrome_options.add_argument(\'--no-sandbox\')
    chrome_options.add_argument(\'--disable-dev-shm-usage\')
    chrome_options.add_argument(\'--disable-gpu\')
    chrome_options.add_argument(\'--window-size=1920,1080\')
    chrome_options.add_argument(\'--user-agent=Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36\')
    
    driver = None
    try:
        driver = webdriver.Chrome(options=chrome_options)
        driver.set_page_load_timeout(timeout)
        
        print(f"Loading page: {url}", file=sys.stderr)
        driver.get(url)
        
        # Wait for page to load and bypass Cloudflare if needed
        time.sleep(5)
        
        # Check if we\'re still on a Cloudflare challenge page
        if "Just a moment" in driver.title or "cf-browser-verification" in driver.page_source:
            print("Waiting for Cloudflare challenge to complete...", file=sys.stderr)
            WebDriverWait(driver, timeout).until(
                lambda d: "Just a moment" not in d.title and "cf-browser-verification" not in d.page_source
            )
        
        # Get the final page source
        html_content = driver.page_source
        
        with open(output_file, \'w\', encoding=\'utf-8\') as f:
            f.write(html_content)
        
        print(f"Successfully saved page to {output_file}", file=sys.stderr)
        return True
        
    except TimeoutException:
        print(f"Timeout while loading {url}", file=sys.stderr)
        return False
    except WebDriverException as e:
        print(f"WebDriver error: {e}", file=sys.stderr)
        return False
    except Exception as e:
        print(f"Unexpected error: {e}", file=sys.stderr)
        return False
    finally:
        if driver:
            driver.quit()

if __name__ == "__main__":
    if len(sys.argv) != 3:
        print("Usage: python3 browser_fetch.py <url> <output_file>", file=sys.stderr)
        sys.exit(1)
    
    url = sys.argv[1]
    output_file = sys.argv[2]
    
    success = fetch_page(url, output_file)
    sys.exit(0 if success else 1)
EOF

    # Run the browser script
    if python3 "$TEMP_DIR/browser_fetch.py" "$url" "$output_file" 2>> "$TEMP_DIR/logs/browser_errors.log"; then
        return 0
    else
        return 1
    fi
}

# URL normalization functions (for web mode)
normalize_url() {
    local url="$1"
    local base_url="$2"
    local current_url="$3"

    if [[ "$url" =~ ^/ ]]; then
        url="$base_url$url"
    elif [[ ! "$url" =~ ^https?:// ]]; then
        # Corrected sed command: use a different delimiter for the s command
        url=$(echo "$url" | sed 's|^\./||')
        local current_dir="$(dirname "$current_url" | sed "s|^$base_url||")"
        url="$base_url$current_dir/$url"
    fi

    # Clean up URL
    url=$(echo "$url" | sed 's/#.*$//; s|/$||; s/\?.*$//')
    echo "$url"
}

url_to_filepath() {
    local url="$1"
    local base_url="$2"
    local domain="$3"

    local path=$(echo "$url" | sed "s|^$base_url/||; s|/$||; s|/|_|g")
    if [ -z "$path" ]; then
        path="index"
    fi
    echo "$TEMP_DIR/$domain/$path.html"
}

calculate_depth() {
    local url="$1"
    local path=$(echo "$url" | sed "s|^https\?://[^/]*||; s|^/||; s|/$||")
    if [ -z "$path" ]; then
        echo 0
    else
        echo "$path" | awk -F'/' '{print NF}'
    fi
}

# Function to download and process external resources
download_external_resources() {
    local html_file="$1"
    local base_url="$2"
    local html_dir="$(dirname "$html_file")"
    
    log_verbose "Processing external resources for $(basename "$html_file")"
    
    # Create assets directory
    mkdir -p "$html_dir/assets"
    
    # Download CSS files
    grep -o 'href=["\047][^"\047]*\.css[^"\047]*["\047]' "$html_file" 2>/dev/null | sed 's/.*href=["\047]\([^"\047]*\)["\047].*/\1/' | while read -r css_url; do
        if [[ "$css_url" =~ ^https?:// ]]; then
            local css_filename=$(basename "$css_url" | sed 's/\?.*$//')
            local css_path="$html_dir/assets/$css_filename"
            if curl -s -L --fail --max-time 10 "$css_url" > "$css_path" 2>/dev/null; then
                log_verbose "Downloaded CSS: $css_url"
                # Update HTML to use local CSS
                sed -i "s|href=[\"']${css_url}[\"']|href=\"assets/$css_filename\"|g" "$html_file"
            fi
        elif [[ "$css_url" =~ ^/ ]]; then
            local full_css_url="$base_url$css_url"
            local css_filename=$(basename "$css_url" | sed 's/\?.*$//')
            local css_path="$html_dir/assets/$css_filename"
            if curl -s -L --fail --max-time 10 "$full_css_url" > "$css_path" 2>/dev/null; then
                log_verbose "Downloaded CSS: $full_css_url"
                sed -i "s|href=[\"']${css_url}[\"']|href=\"assets/$css_filename\"|g" "$html_file"
            fi
        fi
    done
    
    # Download images
    grep -o 'src=["\047][^"\047]*\.\(png\|jpg\|jpeg\|gif\|svg\|webp\)[^"\047]*["\047]' "$html_file" 2>/dev/null | sed 's/.*src=["\047]\([^"\047]*\)["\047].*/\1/' | while read -r img_url; do
        if [[ "$img_url" =~ ^https?:// ]]; then
            local img_filename=$(basename "$img_url" | sed 's/\?.*$//')
            local img_path="$html_dir/assets/$img_filename"
            if curl -s -L --fail --max-time 10 "$img_url" > "$img_path" 2>/dev/null; then
                log_verbose "Downloaded image: $img_url"
                sed -i "s|src=[\"']${img_url}[\"']|src=\"assets/$img_filename\"|g" "$html_file"
            fi
        elif [[ "$img_url" =~ ^/ ]]; then
            local full_img_url="$base_url$img_url"
            local img_filename=$(basename "$img_url" | sed 's/\?.*$//')
            local img_path="$html_dir/assets/$img_filename"
            if curl -s -L --fail --max-time 10 "$full_img_url" > "$img_path" 2>/dev/null; then
                log_verbose "Downloaded image: $full_img_url"
                sed -i "s|src=[\"']${img_url}[\"']|src=\"assets/$img_filename\"|g" "$html_file"
            fi
        fi
    done
}

# Function to inject CSS for better PDF rendering
inject_pdf_css() {
    local html_file="$1"
    
    # Create CSS for better PDF rendering
    # Using a temporary file for the CSS content to avoid complex escaping issues with newlines in sed
    local pdf_css_file="$TEMP_DIR/pdf_styles.css"
    cat > "$pdf_css_file" << 'PDF_CSS_EOF'
<style>
/* PDF-specific styles */
@page {
    size: A4;
    margin: 2cm;
}

body {
    font-family: Arial, sans-serif;
    line-height: 1.6;
    color: #333;
    max-width: none;
}

/* Image scaling */
img {
    max-width: 100% !important;
    height: auto !important;
    display: block;
    margin: 10px 0;
    page-break-inside: avoid;
}

/* Table improvements */
table {
    width: 100%;
    border-collapse: collapse;
    margin: 10px 0;
    page-break-inside: avoid;
}

table, th, td {
    border: 1px solid #ddd;
}

th, td {
    padding: 8px;
    text-align: left;
}

/* Code blocks */
pre, code {
    background-color: #f4f4f4;
    padding: 10px;
    border-radius: 4px;
    font-family: monospace;
    font-size: 0.9em;
    overflow-wrap: break-word;
    word-wrap: break-word;
}

/* Headings */
h1, h2, h3, h4, h5, h6 {
    page-break-after: avoid;
    margin-top: 20px;
    margin-bottom: 10px;
}

/* Lists */
ul, ol {
    margin: 10px 0;
    padding-left: 30px;
}

/* Links */
a {
    color: #0066cc;
    text-decoration: none;
}

/* Prevent widows and orphans */
p {
    orphans: 3;
    widows: 3;
}

/* Hide navigation and other non-content elements */
nav, .nav, .navigation, .menu, .sidebar, .footer, .header, .ads, .advertisement {
    display: none !important;
}

/* Ensure content flows properly */
.content, .main, .article, .post {
    width: 100% !important;
    max-width: none !important;
    margin: 0 !important;
    padding: 0 !important;
}
</style>
PDF_CSS_EOF
    
    # Insert CSS before closing head tag or at the beginning of body
    if grep -q "</head>" "$html_file"; then
        sed -i "/\/head>/i\n$(cat "$pdf_css_file")\n" "$html_file"
    else
        sed -i "/body>/a\n$(cat "$pdf_css_file")\n" "$html_file"
    fi
}

# Web crawling function
crawl_website() {
    local start_url="$SOURCE"
    local domain=$(echo "$start_url" | grep -oE "(https?://[^/]+)" | sed 's|https\?://||')
    local base_url=$(echo "$start_url" | grep -oE 'https?://[^/]+')

    log_info "Starting web crawl from: $start_url"
    log_info "Domain: $domain, Max depth: $MAX_DEPTH, Max pages: $MAX_PAGES"
    if [ "$USE_BROWSER" -eq 1 ]; then
        log_info "Using browser automation to bypass Cloudflare protection"
    fi

    local queue_file="$TEMP_DIR/queue.txt"
    local visited_file="$TEMP_DIR/visited.txt"
    local crawl_log="$TEMP_DIR/logs/crawl_errors.log"

    echo "$start_url" > "$queue_file"
    touch "$visited_file"

    mkdir -p "$TEMP_DIR/$domain"

    local page_count=0

    while [ -s "$queue_file" ] && [ "$page_count" -lt "$MAX_PAGES" ]; do
        local current_url=$(head -n 1 "$queue_file")
        sed -i '1d' "$queue_file"

        if grep -Fx "$current_url" "$visited_file" >/dev/null; then
            log_verbose "Skipping already visited: $current_url"
            continue
        fi

        echo "$current_url" >> "$visited_file"

        local depth=$(calculate_depth "$current_url")
        log_verbose "Processing $current_url (depth $depth)"

        if [ "$depth" -gt "$MAX_DEPTH" ]; then
            log_verbose "Skipping (depth $depth > $MAX_DEPTH): $current_url"
            continue
        fi

        local file_path=$(url_to_filepath "$current_url" "$base_url" "$domain")
        log_verbose "Downloading $current_url to $file_path"

        local download_success=0
        
        # Try browser method first if enabled
        if [ "$USE_BROWSER" -eq 1 ]; then
            if fetch_page_with_browser "$current_url" "$file_path"; then
                download_success=1
                log_verbose "Successfully downloaded with browser: $current_url"
            else
                log_warning "Browser download failed for $current_url, trying curl"
            fi
        fi
        
        # Fallback to curl if browser failed or not enabled
        if [ "$download_success" -eq 0 ]; then
            if curl -s -L --fail --retry 3 --retry-delay 2 --max-time 30 "$current_url" > "$file_path" 2>> "$crawl_log"; then
                download_success=1
                log_verbose "Successfully downloaded with curl: $current_url"
            fi
        fi

        if [ "$download_success" -eq 1 ]; then
            ((page_count++))
            log_info "Crawled page $page_count: $current_url"

            # Download external resources and inject PDF CSS
            download_external_resources "$file_path" "$base_url"
            inject_pdf_css "$file_path"

            # Extract and process links (only if not at max depth)
            if [ "$depth" -lt "$MAX_DEPTH" ]; then
                local links=$(grep -o '<a[^>]*href=["\047][^"\047]*["\047]' "$file_path" 2>/dev/null | sed 's/.*href=["\047]\([^"\047]*\)["\047].*/\1/' | sort -u)

                for link in $links; do
                    if [ -z "$link" ] || [[ "$link" =~ ^(javascript:|mailto:|#|data:) ]]; then
                        continue
                    fi

                    local normalized_link=$(normalize_url "$link" "$base_url" "$current_url")

                    if [[ "$normalized_link" =~ ^https?://$domain ]]; then
                        local local_path=$(url_to_filepath "$normalized_link" "$base_url" "$domain")
                        local relative_path=$(realpath --relative-to="$(dirname "$file_path")" "$local_path" 2>/dev/null || echo "$local_path")

                        # Update link in HTML file
                        # Using a different delimiter for sed to avoid issues with '/' in URLs
                        local escaped_link=$(echo "$link" | sed 's/[&/\.]/\\&/g') # Escape &, /, and .
                        sed -i "s@href=[\"']${escaped_link}[\"']@href=\"$relative_path\"@g" "$file_path" 2>> "$crawl_log"

                        if ! grep -Fx "$normalized_link" "$visited_file" >/dev/null; then
                            echo "$normalized_link" >> "$queue_file"
                            log_verbose "Queued: $normalized_link"
                        fi
                    fi
                done
            fi
        else
            log_warning "Failed to download $current_url"
        fi
    done

    log_success "Crawled $page_count pages"
}

# Local HTML processing function
process_local_html() {
    local html_dir="$SOURCE"
    log_info "Processing local HTML files from: $html_dir"

    # Find all HTML files recursively
    local html_files=()
    while IFS= read -r -d '' file; do
        html_files+=("$file")
    done < <(find "$html_dir" -type f \( -name "*.html" -o -name "*.htm" \) -print0)

    if [ ${#html_files[@]} -eq 0 ]; then
        log_error "No HTML files found in $html_dir"
        exit 1
    fi

    log_info "Found ${#html_files[@]} HTML files"

    # Copy HTML files to temp directory, maintaining structure
    local temp_html_dir="$TEMP_DIR/html"
    mkdir -p "$temp_html_dir"

    for html_file in "${html_files[@]}"; do
        local rel_path=$(realpath --relative-to="$html_dir" "$html_file")
        local dest_path="$temp_html_dir/$rel_path"
        local dest_dir=$(dirname "$dest_path")

        mkdir -p "$dest_dir"
        cp "$html_file" "$dest_path"
        
        # Inject PDF CSS for better rendering
        inject_pdf_css "$dest_path"
        
        log_verbose "Copied: $rel_path"
    done

    # Also copy any associated assets (CSS, JS, images)
    local asset_extensions=("css" "js" "png" "jpg" "jpeg" "gif" "svg" "ico" "woff" "woff2" "ttf" "eot")
    for ext in "${asset_extensions[@]}"; do
        while IFS= read -r -d '' file; do
            local rel_path=$(realpath --relative-to="$html_dir" "$file")
            local dest_path="$temp_html_dir/$rel_path"
            local dest_dir=$(dirname "$dest_path")

            mkdir -p "$dest_dir"
            cp "$file" "$dest_path"
            log_verbose "Copied asset: $rel_path"
        done < <(find "$html_dir" -type f -name "*.${ext}" -print0 2>/dev/null)
    done

    log_success "Processed local HTML files"
}

# Try alternative conversion for problematic files
try_alternative_conversion() {
    local html_file="$1"
    local pdf_file="$2"
    local html_dir="$3"
    local weasyprint_log="$4"

    log_verbose "Trying alternative conversion for $(basename "$html_file")"

    # Try with simpler options (no base URL, different media type)
    if timeout 30 weasyprint \
        --media-type screen \
        --pdf-version 1.4 \
        "$html_file" "$pdf_file" 2>> "$weasyprint_log"; then
        return 0
    fi

    # Try with minimal options
    if timeout 30 weasyprint \
        "$html_file" "$pdf_file" 2>> "$weasyprint_log"; then
        return 0
    fi

    return 1
}

# PDF generation function
generate_pdfs() {
    local html_files=()
    local pdf_files=()

    # Find HTML files in temp directory
    while IFS= read -r -d '' file; do
        html_files+=("$file")
    done < <(find "$TEMP_DIR" -type f \( -name "*.html" -o -name "*.htm" \) -print0)

    if [ ${#html_files[@]} -eq 0 ]; then
        log_error "No HTML files found for PDF generation"
        exit 1
    fi

    # Sort files for consistent processing order
    IFS=$'\n' html_files=($(sort <<<"${html_files[*]}"))
    unset IFS

    log_info "Converting ${#html_files[@]} HTML files to PDF"

    local success_count=0
    local failed_count=0
    local current_file=0
    local weasyprint_log="$TEMP_DIR/logs/weasyprint_errors.log"

    for html_file in "${html_files[@]}"; do
        ((current_file++))
        local html_dir=$(dirname "$html_file")
        local pdf_file="${html_file%.html}.pdf"
        pdf_file="${pdf_file%.htm}.pdf"
        local filename=$(basename "$html_file")

        log_info "[$current_file/${#html_files[@]}] Converting $filename to PDF..."

        local start_time=$(date +%s)

        log_verbose "Running WeasyPrint on: $html_file"
        
        # Use optimized WeasyPrint settings for better CSS and image handling
        if timeout "$WEASYPRINT_TIMEOUT" weasyprint \
            --base-url "file://$html_dir/" \
            --media-type screen \
            --presentational-hints \
            --pdf-version 1.7 \
            --verbose \
            "$html_file" "$pdf_file" 2>> "$weasyprint_log"; then

            local end_time=$(date +%s)
            local duration=$((end_time - start_time))

            if [ -f "$pdf_file" ] && [ -s "$pdf_file" ]; then
                pdf_files+=("$pdf_file")
                ((success_count++))
                log_success "✓ $filename converted successfully (${duration}s)"
            else
                log_warning "✗ $filename: PDF file is empty or missing"
                ((failed_count++))
            fi
        else
            local end_time=$(date +%s)
            local duration=$((end_time - start_time))

            if [ $duration -ge $WEASYPRINT_TIMEOUT ]; then
                log_warning "✗ $filename: Conversion timed out after ${WEASYPRINT_TIMEOUT}s"
            else
                log_warning "✗ $filename: Conversion failed (${duration}s)"
            fi

            # Try alternative conversion
            if try_alternative_conversion "$html_file" "$pdf_file" "$html_dir" "$weasyprint_log"; then
                if [ -f "$pdf_file" ] && [ -s "$pdf_file" ]; then
                    pdf_files+=("$pdf_file")
                    ((success_count++))
                    log_success "✓ $filename converted with alternative method"
                else
                    ((failed_count++))
                fi
            else
                ((failed_count++))
            fi
        fi

        # Show progress
        if [ $((current_file % 10)) -eq 0 ] || [ $current_file -eq ${#html_files[@]} ]; then
            log_info "Progress: $current_file/${#html_files[@]} (Success: $success_count, Failed: $failed_count)"
        fi
    done

    if [ ${#pdf_files[@]} -eq 0 ]; then
        log_error "No PDFs were generated successfully"
        exit 1
    fi

    log_success "Generated $success_count PDFs successfully"
    if [ $failed_count -gt 0 ]; then
        log_warning "$failed_count files failed to convert"
    fi

    # Merge PDFs
    log_info "Merging ${#pdf_files[@]} PDFs into $OUTPUT_PDF..."
    if pdfunite "${pdf_files[@]}" "$OUTPUT_PDF" 2>> "$TEMP_DIR/logs/processing.log"; then
        log_success "Successfully created $OUTPUT_PDF"
        if command -v du &> /dev/null; then
            local file_size=$(du -h "$OUTPUT_PDF" | cut -f1)
            log_info "Final PDF size: $file_size"
        fi
    else
        log_error "Failed to merge PDFs"
        exit 1
    fi
}

# Copy debug files
copy_debug_files() {
    local debug_files=(
        "crawl_errors.log"
        "weasyprint_errors.log"
        "processing.log"
        "browser_errors.log"
    )

    for file in "${debug_files[@]}"; do
        if [ -f "$TEMP_DIR/logs/$file" ]; then
            cp "$TEMP_DIR/logs/$file" "./$file"
        fi
    done

    # Copy other useful files
    if [ -f "$TEMP_DIR/queue.txt" ]; then
        cp "$TEMP_DIR/queue.txt" "./queue.log"
    fi

    if [ -f "$TEMP_DIR/visited.txt" ]; then
        cp "$TEMP_DIR/visited.txt" "./visited.log"
    fi

    log_info "Debug files saved to current directory"
}

# Main function
main() {
    log_info "$SCRIPT_NAME v$VERSION starting..."

    parse_arguments "$@"
    validate_inputs
    check_dependencies
    init_environment

    if [ "$MODE" = "web" ]; then
        crawl_website
    elif [ "$MODE" = "local" ]; then
        process_local_html
    fi

    generate_pdfs
    copy_debug_files
    cleanup_temp

    log_success "Process completed successfully!"
    log_info "Output: $OUTPUT_PDF"
}

# Run main function with all arguments
main "$@"