#
# This program reads rekordbox CUE files and output a label track to be imported in Audacity.
# Usage: cue2audacity.py directory (where directory is the path where the CUE files are stored)
# Author: Manuel Weiser
# Date: 2025-07-03
#

import argparse
import glob
import re
import os

# Get argument on command line
parser = argparse.ArgumentParser()
parser.add_argument("thePathOfCueFiles", help="The path of the directory with the CUE files to process")
thePath = parser.parse_args().thePathOfCueFiles

# Read *.cue files in specified directory
for cueFileName in glob.glob(os.path.join(thePath, "*.cue")):
    #print(cueFileName)
    with open(cueFileName, "r", encoding="utf-8") as cueFile:
        cueFileContent = cueFile.readlines()
    #print(cueFileContent)

    # Create label file
    labelFileName = os.path.splitext(cueFileName)[0] + "_audacity.txt"
    tracklistFileName = os.path.splitext(cueFileName)[0] + "_tracklist.txt"
    
    with open(labelFileName, "w", encoding="utf-8") as labelFile, \
         open(tracklistFileName, "w", encoding="utf-8") as tracklistFile:
        #print ("###" + labelFileName + "###")

        # Process file
        theTitle = "" # initialize title now in order to keep it available when scanning subsequent INDEX line
        thePerformer = "" # initialize performer
        trackCount = 0
        lastTitles = []  # Track the last few titles to avoid duplicates within a few tracks
        maxTrackDistance = 2  # Filter duplicates within this many tracks
        
        for lineNum, line in enumerate(cueFileContent, 1):
            originalLine = line
            line = line.strip()  # Remove leading/trailing whitespace

            # Match TITLE - try multiple patterns for different indentation (tabs or spaces)
            theMatchTitle = re.search(r'TITLE\s+"([^"]+)"', line) 
            if theMatchTitle is not None:
                theTitle = theMatchTitle.group(1)

            # Match PERFORMER - try multiple patterns for different indentation (tabs or spaces)
            theMatchPerformer = re.search(r'PERFORMER\s+"([^"]+)"', line) 
            if theMatchPerformer is not None:
                thePerformer = theMatchPerformer.group(1)

            # Match INDEX 01 - try multiple patterns for different formats (tabs or spaces)
            theMatchIndex = re.search(r'INDEX\s+01\s+(\d{1,2}):(\d{2}):(\d{2})', line) 
            if theMatchIndex is not None:
                theHours = int(theMatchIndex.group(1))      # Actually hours in HH:MM:SS format
                theMinutes = int(theMatchIndex.group(2))    # Minutes
                theSeconds = int(theMatchIndex.group(3))    # Seconds
                
                # Convert HH:MM:SS to total seconds
                theStartTime = (theHours * 3600) + (theMinutes * 60) + theSeconds
                
                # Check for duplicates: skip if same title appears within the last few tracks
                isDuplicate = theTitle in lastTitles[-maxTrackDistance:]
                
                if not isDuplicate:
                    trackCount += 1
                    
                    # format theStartTime
                    theStartTimeFormatted = "{:.6f}".format(theStartTime)
                    
                    # Format time for tracklist (MM:SS)
                    timeMinutes = theStartTime // 60
                    timeSeconds = theStartTime % 60
                    timeFormatted = f"{int(timeMinutes):02d}:{int(timeSeconds):02d}"
                    
                    # Write time and title to Audacity label file
                    labelFile.write(theStartTimeFormatted + "\t" + theStartTimeFormatted + "\t" + theTitle + "\n")
                    
                    # Write to tracklist file: TrackNr. Artist - Track (Time)
                    if thePerformer:
                        tracklistFile.write(f"{trackCount:02d}. {thePerformer} - {theTitle} ({timeFormatted})\n")
                    else:
                        tracklistFile.write(f"{trackCount:02d}. {theTitle} ({timeFormatted})\n")
                    
                    # Add to title history
                    lastTitles.append(theTitle)
                else:
                    print(f"Skipping duplicate: '{theTitle}' at {theHours:02d}:{theMinutes:02d}:{theSeconds:02d} (appeared within last {maxTrackDistance} tracks)")
        
        print(f"Created {trackCount} labels in {labelFileName}")
        print(f"Created tracklist in {tracklistFileName}")

        # end for line in cueFileContent
    
# end for cueFileName in glob.glob(thePath + "\\*.cue")