prem2resolve/premiere_to_resolve.py

#!/usr/bin/env python3
"""
Premiere to Resolve XML Converter

Processes Adobe Premiere Pro XML exports to make them compatible with DaVinci Resolve on Linux.
Updates file paths and optionally converts MP4 files to MOV format.
"""

import os
import sys
import xml.etree.ElementTree as ET
import urllib.parse
import subprocess
from pathlib import Path
from typing import Dict, Tuple, Set


def extract_video_references(xml_file: str) -> Set[Tuple[str, str]]:
    """
    Parse XML and extract all unique video file references.
    
    Returns a set of tuples: (filename, pathurl)
    """
    try:
        tree = ET.parse(xml_file)
        root = tree.getroot()
    except ET.ParseError as e:
        print(f"Error parsing XML file {xml_file}: {e}", file=sys.stderr)
        return set()
    
    video_refs = set()
    
    # Find all <file> elements that contain <pathurl> tags
    for file_elem in root.iter('file'):
        name_elem = file_elem.find('name')
        pathurl_elem = file_elem.find('pathurl')
        
        if name_elem is not None and name_elem.text:
            filename = name_elem.text.strip()
            pathurl = ""
            
            if pathurl_elem is not None and pathurl_elem.text:
                pathurl = pathurl_elem.text.strip()
            
            # Only include video files (check extension)
            if filename.lower().endswith(('.mp4', '.mov', '.avi', '.mxf', '.mts', '.mkv')):
                video_refs.add((filename, pathurl))
    
    return video_refs


def decode_pathurl(pathurl: str) -> str:
    """
    Decode a file:// URL path to a regular path and extract filename.
    
    Example: file://localhost/C%3a/Users/.../file.mp4 -> file.mp4
    """
    if not pathurl:
        return ""
    
    try:
        # Remove file://localhost/ or file:/// prefix
        if pathurl.startswith('file://localhost/'):
            path_part = pathurl[17:]  # Remove 'file://localhost/'
        elif pathurl.startswith('file:///'):
            path_part = pathurl[7:]  # Remove 'file:///'
        elif pathurl.startswith('file://'):
            path_part = pathurl[7:]  # Remove 'file://'
        else:
            path_part = pathurl
        
        # URL decode
        decoded = urllib.parse.unquote(path_part)
        
        # Extract filename
        filename = os.path.basename(decoded)
        return filename
    except Exception as e:
        print(f"Error decoding pathurl {pathurl}: {e}", file=sys.stderr)
        return ""


def find_local_files(video_refs: Set[Tuple[str, str]], directory: str) -> Dict[str, Dict[str, str]]:
    """
    Match XML references with files in directory.
    
    Returns a dict mapping original filename to:
    {
        'local_path': actual path found,
        'original_pathurl': original pathurl from XML,
        'needs_conversion': True if MP4 needs conversion
    }
    """
    directory_path = Path(directory)
    file_mapping = {}
    
    # Create a case-insensitive mapping of files in directory
    local_files = {}
    for file_path in directory_path.iterdir():
        if file_path.is_file():
            local_files[file_path.name.lower()] = str(file_path.resolve())
    
    for filename, pathurl in video_refs:
        # Try exact match first
        local_path = None
        if filename in local_files:
            local_path = local_files[filename]
        elif filename.lower() in local_files:
            local_path = local_files[filename.lower()]
        
        if local_path:
            needs_conversion = filename.lower().endswith('.mp4')
            file_mapping[filename] = {
                'local_path': local_path,
                'original_pathurl': pathurl,
                'needs_conversion': needs_conversion
            }
    
    return file_mapping


def convert_mp4_to_mov(input_file: str, output_file: str) -> bool:
    """
    Convert MP4 to MOV using ffmpeg with minimal quality loss.
    
    Strategy:
    1. Try copy codecs first (lossless if compatible)
    2. Fallback to high-quality encoding
    """
    # Check if output file already exists
    if os.path.exists(output_file):
        response = input(f"Output file {output_file} already exists. Overwrite? (y/n): ")
        if response.lower() != 'y':
            return False
    
    # Try copy codecs first (lossless if compatible)
    cmd_copy = [
        'ffmpeg', '-i', input_file,
        '-c:v', 'copy',
        '-c:a', 'copy',
        '-y',  # Overwrite output file
        output_file
    ]
    
    try:
        result = subprocess.run(
            cmd_copy,
            capture_output=True,
            text=True,
            timeout=3600  # 1 hour timeout
        )
        
        if result.returncode == 0:
            print(f"  Converted (copy codecs): {os.path.basename(input_file)} -> {os.path.basename(output_file)}")
            return True
    except subprocess.TimeoutExpired:
        print(f"  Timeout converting {input_file}", file=sys.stderr)
        return False
    except FileNotFoundError:
        print("Error: ffmpeg not found. Please install ffmpeg.", file=sys.stderr)
        return False
    
    # Fallback to high-quality encoding
    print(f"  Copy codecs not compatible, using high-quality encoding for {os.path.basename(input_file)}")
    cmd_encode = [
        'ffmpeg', '-i', input_file,
        '-c:v', 'libx264',
        '-crf', '18',  # High quality (lower = better quality)
        '-preset', 'slow',  # Better compression
        '-c:a', 'copy',  # Copy audio
        '-y',
        output_file
    ]
    
    try:
        result = subprocess.run(
            cmd_encode,
            capture_output=True,
            text=True,
            timeout=3600
        )
        
        if result.returncode == 0:
            print(f"  Converted (encoded): {os.path.basename(input_file)} -> {os.path.basename(output_file)}")
            return True
        else:
            print(f"  Error converting {input_file}: {result.stderr}", file=sys.stderr)
            return False
    except subprocess.TimeoutExpired:
        print(f"  Timeout converting {input_file}", file=sys.stderr)
        return False
    except Exception as e:
        print(f"  Error converting {input_file}: {e}", file=sys.stderr)
        return False


def update_xml_paths(xml_file: str, file_mapping: Dict[str, Dict[str, str]]) -> bool:
    """
    Update XML pathurl tags with new Linux file:// URLs.
    
    Returns True if successful, False otherwise.
    """
    try:
        tree = ET.parse(xml_file)
        root = tree.getroot()
    except ET.ParseError as e:
        print(f"Error parsing XML file {xml_file}: {e}", file=sys.stderr)
        return False
    
    updated_count = 0
    
    # Find all <file> elements and update their pathurl and name if needed
    for file_elem in root.iter('file'):
        name_elem = file_elem.find('name')
        pathurl_elem = file_elem.find('pathurl')
        
        if name_elem is not None and name_elem.text:
            filename = name_elem.text.strip()
            
            if filename in file_mapping:
                mapping = file_mapping[filename]
                new_path = mapping['local_path']
                new_filename = os.path.basename(new_path)
                
                # Update name tag if filename changed (e.g., MP4 -> MOV)
                if new_filename != filename:
                    name_elem.text = new_filename
                
                # Convert to file:// URL format
                # On Linux, file:// URLs should be file:///absolute/path (three slashes)
                # Encode the path properly, but preserve forward slashes
                # urllib.parse.quote with safe='/' will preserve slashes, so /path becomes /path
                # Then file:///path gives us the correct three-slash format
                encoded_path = urllib.parse.quote(new_path, safe='/')
                new_pathurl = f"file://{encoded_path}"
                
                if pathurl_elem is not None:
                    pathurl_elem.text = new_pathurl
                    updated_count += 1
                else:
                    # Create pathurl element if it doesn't exist
                    pathurl_elem = ET.SubElement(file_elem, 'pathurl')
                    pathurl_elem.text = new_pathurl
                    updated_count += 1
    
    if updated_count > 0:
        try:
            # Write back to file
            tree.write(xml_file, encoding='UTF-8', xml_declaration=True)
            print(f"  Updated {updated_count} path references in {xml_file}")
            return True
        except Exception as e:
            print(f"Error writing XML file {xml_file}: {e}", file=sys.stderr)
            return False
    
    return True


def check_ffmpeg() -> bool:
    """Check if ffmpeg is available."""
    try:
        subprocess.run(['ffmpeg', '-version'], capture_output=True, timeout=5)
        return True
    except (FileNotFoundError, subprocess.TimeoutExpired):
        return False


def main():
    """Main workflow."""
    # Get current directory
    current_dir = os.getcwd()
    print(f"Working directory: {current_dir}\n")
    
    # Find all XML files
    xml_files = [f for f in os.listdir(current_dir) if f.lower().endswith('.xml')]
    
    if not xml_files:
        print("No XML files found in current directory.")
        return
    
    print(f"Found {len(xml_files)} XML file(s):")
    for xml_file in xml_files:
        print(f"  - {xml_file}")
    print()
    
    # Extract all video references from XML files
    all_video_refs = set()
    xml_to_refs = {}
    
    for xml_file in xml_files:
        refs = extract_video_references(xml_file)
        all_video_refs.update(refs)
        xml_to_refs[xml_file] = refs
        print(f"Found {len(refs)} video reference(s) in {xml_file}")
    
    print(f"\nTotal unique video files referenced: {len(all_video_refs)}\n")
    
    # Find local files
    file_mapping = find_local_files(all_video_refs, current_dir)
    
    # Separate found and missing files
    found_files = set(file_mapping.keys())
    missing_files = {filename for filename, _ in all_video_refs if filename not in found_files}
    
    # Show summary
    print("File status:")
    print(f"  Found: {len(found_files)}")
    print(f"  Missing: {len(missing_files)}")
    
    if found_files:
        print("\nFound files:")
        for filename in sorted(found_files):
            mapping = file_mapping[filename]
            status = " (needs MP4→MOV conversion)" if mapping['needs_conversion'] else ""
            print(f"  - {filename}{status}")
    
    if missing_files:
        print("\nMissing files:")
        for filename in sorted(missing_files):
            print(f"  - {filename}")
    
    print()
    
    # Check for MP4 files that need conversion
    mp4_files = {f: m for f, m in file_mapping.items() if m['needs_conversion']}
    
    # Prompt for conversion
    should_convert = False
    if mp4_files:
        response = input(f"Convert {len(mp4_files)} MP4 file(s) to MOV? (y/n): ")
        should_convert = response.lower() == 'y'
        
        if should_convert:
            if not check_ffmpeg():
                print("Error: ffmpeg not found. Please install ffmpeg to convert files.", file=sys.stderr)
                should_convert = False
            else:
                print("\nConverting files...")
                conversion_map = {}
                
                for filename, mapping in mp4_files.items():
                    input_path = mapping['local_path']
                    # Create output filename with .mov extension
                    base_name = os.path.splitext(filename)[0]
                    output_filename = f"{base_name}.mov"
                    output_path = os.path.join(current_dir, output_filename)
                    
                    if convert_mp4_to_mov(input_path, output_path):
                        # Update mapping to point to converted file
                        conversion_map[filename] = output_filename
                        file_mapping[filename]['local_path'] = output_path
                        file_mapping[filename]['needs_conversion'] = False
                
                if conversion_map:
                    print(f"\nSuccessfully converted {len(conversion_map)} file(s)")
                    # Update file_mapping for converted files
                    for original_filename, new_filename in conversion_map.items():
                        if original_filename in file_mapping:
                            new_path = os.path.join(current_dir, new_filename)
                            file_mapping[original_filename]['local_path'] = new_path
    else:
        print("No MP4 files found that need conversion.")
    
    print()
    
    # Prompt for XML updates
    response = input("Update XML file paths? (y/n): ")
    should_update = response.lower() == 'y'
    
    if should_update:
        print("\nUpdating XML files...")
        for xml_file in xml_files:
            # Only update paths for files referenced in this XML
            xml_specific_mapping = {
                filename: mapping
                for filename, mapping in file_mapping.items()
                if filename in {f for f, _ in xml_to_refs[xml_file]}
            }
            
            if xml_specific_mapping:
                update_xml_paths(xml_file, xml_specific_mapping)
            else:
                print(f"  No matching files found for {xml_file}")
    
    # Final report
    print("\n" + "="*60)
    print("Final Report")
    print("="*60)
    
    if missing_files:
        print(f"\nMissing files ({len(missing_files)}):")
        for filename in sorted(missing_files):
            print(f"  - {filename}")
    else:
        print("\nAll referenced files were found!")
    
    if should_convert and mp4_files:
        converted_count = sum(1 for f, m in file_mapping.items() 
                             if f in mp4_files and not m['needs_conversion'])
        print(f"\nConverted {converted_count} MP4 file(s) to MOV")
    
    if should_update:
        print(f"\nUpdated {len(xml_files)} XML file(s)")


if __name__ == '__main__':
    main()