app.py

#!/usr/bin/env python3
"""
TIDAL Download API - Hugging Face Space
FastAPI endpoint to download TIDAL tracks in various qualities
"""

import os
import sys
import json
import time
import base64
import tempfile
import requests
from fastapi import FastAPI, HTTPException, Query
from fastapi.responses import FileResponse
from xml.etree import ElementTree as ET

# TV session credentials (used by OrpheusDL for lossless access)
TV_CLIENT_ID = "cgiF7TQuB97BUIu3"
TV_CLIENT_SECRET = "1nqpgx8uvBdZigrx4hUPDV2hOwgYAAAG5DYXOr6uNf8="

# Mobile session credentials
MOBILE_CLIENT_ID = "fX2JxdmntZWK0ixT"
MOBILE_CLIENT_SECRET = "1Nm5AfDAjxrgJFJbKNWLeAyKGVGmINuXPPLHVXAvxAg="

# Read tokens from environment variables (Hugging Face Space secrets)
ACCESS_TOKEN = os.environ.get("TIDAL_ACCESS_TOKEN", "")
REFRESH_TOKEN = os.environ.get("TIDAL_REFRESH_TOKEN", "")
CURRENT_CLIENT_ID = TV_CLIENT_ID  # Use TV client by default

app = FastAPI(title="TIDAL Download API", description="Download TIDAL tracks in LOSSLESS/HI_RES quality")

def refresh_access_token():
    """Refresh the access token using the refresh token"""
    global ACCESS_TOKEN, REFRESH_TOKEN, CURRENT_CLIENT_ID
    
    print("🔄 Refreshing access token with TV session credentials...")
    
    response = requests.post(
        "https://auth.tidal.com/v1/oauth2/token",
        data={
            "grant_type": "refresh_token",
            "refresh_token": REFRESH_TOKEN,
            "client_id": TV_CLIENT_ID,
            "client_secret": TV_CLIENT_SECRET
        },
        headers={
            "Content-Type": "application/x-www-form-urlencoded",
            "User-Agent": "TIDAL_ANDROID/1039 okhttp/3.14.9"
        },
        timeout=15
    )
    
    if response.status_code == 200:
        data = response.json()
        ACCESS_TOKEN = data.get('access_token')
        new_refresh_token = data.get('refresh_token')
        if new_refresh_token:
            REFRESH_TOKEN = new_refresh_token
        CURRENT_CLIENT_ID = TV_CLIENT_ID
        print(f"✅ Access token refreshed")
        return True
    else:
        print(f"❌ Failed to refresh token: {response.status_code}")
        return False

def get_headers():
    """Get headers with current access token"""
    return {
        "X-Tidal-Token": CURRENT_CLIENT_ID,
        "Authorization": f"Bearer {ACCESS_TOKEN}",
        "User-Agent": "TIDAL_ANDROID/1039 okhttp/3.14.9"
    }

def decrypt_security_token(security_token):
    """Decrypt TIDAL security token to get AES key and nonce"""
    security_token = base64.b64decode(security_token)
    
    # Key derivation constants
    key = bytes([0x7A, 0x5C, 0x38, 0x5A, 0x7B, 0x25, 0x25, 0x49, 
                 0x43, 0x4B, 0x31, 0x71, 0x5E, 0x24, 0x76, 0x5D])
    
    # Extract nonce (first 12 bytes) and ciphertext
    nonce = security_token[:12]
    ciphertext = security_token[12:-16]  # Remove auth tag
    
    from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
    from cryptography.hazmat.backends import default_backend
    
    decryptor = Cipher(algorithms.AES(key), modes.GCM(nonce), backend=default_backend()).decryptor()
    decrypted = decryptor.update(ciphertext)
    
    # Decrypted data contains key and nonce for actual audio decryption
    audio_key = decrypted[:16]
    audio_nonce = decrypted[16:28]
    
    return audio_key, audio_nonce

def decrypt_file(input_path, output_path, key, nonce):
    """Decrypt a file using AES-GCM"""
    from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
    from cryptography.hazmat.backends import default_backend
    
    with open(input_path, 'rb') as infile:
        encrypted_data = infile.read()
    
    # Remove auth tag (last 16 bytes)
    encrypted_data = encrypted_data[:-16]
    
    decryptor = Cipher(algorithms.AES(key), modes.GCM(nonce), backend=default_backend()).decryptor()
    decrypted_data = decryptor.update(encrypted_data)
    
    with open(output_path, 'wb') as outfile:
        outfile.write(decrypted_data)

def parse_dash_manifest(manifest_data):
    """Parse DASH manifest XML and extract segment URLs and audio format"""
    manifest_str = manifest_data.decode('utf-8')

    base_url = ""
    segments = []
    media_template = ""
    start_number = 1
    audio_format = "flac"
    codecs_detected = ""

    try:
        root = ET.fromstring(manifest_str)

        for elem in root.iter():
            tag = elem.tag
            if '}' in tag:
                tag = tag.split('}')[1]

            if tag == 'BaseURL' or tag.lower() == 'baseurl':
                base_url = elem.text

            if tag == 'Representation' or tag.lower() == 'representation':
                codec = elem.get('codecs', '').upper()
                codecs_detected = codec
                if codec:
                    print(f"   Found Representation with codecs: '{codec}'")
                    if 'MP4A' in codec or 'AAC' in codec:
                        audio_format = "m4a"
                    elif 'FLAC' in codec:
                        audio_format = "flac"
                    elif 'EC3' in codec or 'EAC3' in codec:
                        audio_format = "eac3"
                    print(f"   Detected audio format: {audio_format}")

            if tag == 'SegmentTemplate' or tag.lower() == 'segmenttemplate':
                media_template = elem.get('media', '')
                initialization = elem.get('initialization', '')
                start_number = int(elem.get('startNumber', '1'))
                
                if initialization:
                    segments.append(initialization)
                    print(f"✅ Added initialization segment: {initialization}")

                for timeline in elem.iter():
                    tl_tag = timeline.tag
                    if '}' in tl_tag:
                        tl_tag = tl_tag.split('}')[1]

                    if tl_tag == 'SegmentTimeline' or tl_tag.lower() == 'segmenttimeline':
                        total_segments = 0
                        for s_elem in timeline.iter():
                            s_tag = s_elem.tag
                            if '}' in s_tag:
                                s_tag = s_tag.split('}')[1]

                            if s_tag == 'S' or s_tag.lower() == 's':
                                repeat = int(s_elem.get('r', '0'))
                                total_segments += repeat + 1

                        if media_template and total_segments > 0:
                            for i in range(start_number, start_number + total_segments):
                                segment_url = media_template.replace('$Number$', str(i))
                                segments.append(segment_url)
                            print(f"✅ Generated {len(segments)} segment URLs from template")
    except Exception as e:
        print(f"⚠️ Error parsing DASH manifest: {e}")

    if not segments:
        print("⚠️ Trying regex parsing...")

        init_match = re.search(r'initialization="([^"]+)"', manifest_str)
        if init_match:
            initialization = init_match.group(1)
            segments.append(initialization)
            print(f"✅ Added initialization segment with regex: {initialization}")

        template_match = re.search(r'media="([^"]+\$Number\$[^"]+)"', manifest_str)
        if template_match:
            media_template = template_match.group(1)

            start_num_match = re.search(r'startNumber="(\d+)"', manifest_str)
            if start_num_match:
                start_number = int(start_num_match.group(1))

            timeline_matches = re.findall(r'<S[^>]*d="(\d+)"[^>]*r="(\d+)"', manifest_str)
            total_segments = 0
            for d, r in timeline_matches:
                total_segments += int(r) + 1

            single_segments = re.findall(r'<S[^>]*d="(\d+)"[^>]*[^r][^>]*>', manifest_str)
            total_segments += len(single_segments)

            if total_segments > 0:
                for i in range(start_number, start_number + total_segments):
                    segment_url = media_template.replace('$Number$', str(i))
                    segments.append(segment_url)
                print(f"✅ Generated {len(segments)} segment URLs with regex")

        codecs_match = re.search(r'codecs="([^"]+)"', manifest_str)
        if codecs_match:
            codec = codecs_match.group(1).upper()
            codecs_detected = codec
            if 'MP4A' in codec or 'AAC' in codec:
                audio_format = "m4a"
            elif 'FLAC' in codec:
                audio_format = "flac"
            elif 'EC3' in codec or 'EAC3' in codec:
                audio_format = "eac3"

        mime_match = re.search(r'mimeType="([^"]+)"', manifest_str)
        if mime_match:
            mime = mime_match.group(1).lower()
            if 'mp4' in mime or 'aac' in mime or 'm4a' in mime:
                audio_format = "m4a"
            elif 'flac' in mime:
                audio_format = "flac"

        base_match = re.search(r'<[^>]*BaseURL[^>]*>([^<]+)</', manifest_str)
        if base_match:
            base_url = base_match.group(1)

    print(f"   Final detected codec: {codecs_detected}, format: {audio_format}")
    return base_url, segments, audio_format

def parse_bt_manifest(manifest_data):
    """Parse BT manifest (JSON format)"""
    import json
    
    try:
        manifest = json.loads(manifest_data.decode('utf-8'))
        urls = manifest.get('urls', [])
        if urls:
            return "", urls
        else:
            print("⚠️ BT manifest has no URLs")
            return "", []
    except json.JSONDecodeError as e:
        print(f"❌ Failed to parse BT manifest: {e}")
        return "", []

@app.get("/download/{track_id}")
async def download_track(track_id: int, quality: str = "LOSSLESS"):
    """Download a TIDAL track in specified quality.
    
    Returns JSON with track info and download URL.
    - HIGH/LOW: Returns direct TIDAL CDN URL (time-limited)
    - LOSSLESS/HI_RES_LOSSLESS: Downloads, decrypts, merges, returns mounted file URL
    """
    try:
        print(f"🔍 Download request: track_id={track_id}, quality={quality}")
        
        if not ACCESS_TOKEN:
            raise HTTPException(status_code=401, detail="Access token not set")
        
        headers = get_headers()
        
        response = requests.get(
            f"https://api.tidal.com/v1/tracks/{track_id}/playbackinfopostpaywall/v4",
            params={
                "countryCode": "HK",
                "playbackmode": "STREAM",
                "assetpresentation": "FULL",
                "audioquality": quality,
                "prefetch": "false"
            },
            headers=headers,
            timeout=15
        )
        
        if response.status_code == 401:
            print("🔄 Token expired, refreshing...")
            if refresh_access_token():
                headers = get_headers()
                response = requests.get(
                    f"https://api.tidal.com/v1/tracks/{track_id}/playbackinfopostpaywall/v4",
                    params={
                        "countryCode": "HK",
                        "playbackmode": "STREAM",
                        "assetpresentation": "FULL",
                        "audioquality": quality,
                        "prefetch": "false"
                    },
                    headers=headers,
                    timeout=15
                )
        
        if response.status_code != 200:
            raise HTTPException(status_code=response.status_code, detail="Failed to get playback info")
        
        data = response.json()
        audio_quality = data.get('audioQuality')
        manifest_mime = data.get('manifestMimeType', '')
        manifest_data = base64.b64decode(data.get('manifest', ''))
        
        print(f"✅ Got playback data: audioQuality={audio_quality}, manifestType={manifest_mime}")
        
        if audio_quality != quality:
            print(f"⚠️ Requested {quality} but got {audio_quality}")
        
        track_info = requests.get(
            f"https://api.tidal.com/v1/tracks/{track_id}",
            params={"countryCode": "HK"},
            headers=headers,
            timeout=10
        ).json()
        
        title = track_info.get('title', f"track_{track_id}")
        artist = track_info.get('artist', {}).get('name', 'Unknown')
        duration = track_info.get('duration', 0)
        album = track_info.get('album', {}).get('title', '')
        album_artist = track_info.get('album', {}).get('artist', {}).get('name', '')
        track_number = track_info.get('trackNumber', 1)
        isrc = track_info.get('isrc', '')
        
        # For HIGH/LOW quality, return direct CDN URLs (no server-side download)
        if quality in ['HIGH', 'LOW']:
            print("📦 Parsing manifest for direct URL...")
            base_url, segments, audio_format = parse_dash_manifest(manifest_data)
            
            if segments:
                # Get the first segment URL (initialization segment)
                first_segment = segments[0]
                if first_segment.startswith('http://') or first_segment.startswith('https://'):
                    direct_url = first_segment
                elif base_url:
                    direct_url = base_url + first_segment
                else:
                    direct_url = first_segment
                
                print(f"✅ Direct URL found for {quality}: {direct_url[:100]}...")
                
                return {
                    "success": True,
                    "track": {
                        "id": track_id,
                        "title": title,
                        "artist": artist,
                        "album": album,
                        "albumArtist": album_artist,
                        "trackNumber": track_number,
                        "duration": duration,
                        "isrc": isrc,
                        "quality": audio_quality,
                        "format": audio_format
                    },
                    "downloadUrl": direct_url,
                    "message": f"Direct TIDAL CDN URL for {quality} quality (time-limited)"
                }
            else:
                raise HTTPException(status_code=500, detail="No segments found in manifest")
        
        # For LOSSLESS/HI_RES_LOSSLESS, download and merge segments
        is_lossless = 'dash+xml' in manifest_mime
        
        if is_lossless:
            security_token = data.get('securityToken')
            print(f"   securityToken present: {bool(security_token)}, length: {len(security_token) if security_token else 0}")
            key = None
            nonce = None
            
            if security_token:
                print("🔐 Decrypting security token...")
                key, nonce = decrypt_security_token(security_token)
            
            download_dir = "./downloads"
            os.makedirs(download_dir, exist_ok=True)
            tmpdir = tempfile.mkdtemp(prefix="tidal-segments-")
            temp_files = []
            
            try:
                print("📦 Parsing DASH manifest...")
                base_url, segments, audio_format = parse_dash_manifest(manifest_data)
                print(f"✅ Found {len(segments)} segments, format: {audio_format}")
                
                print("⬇️ Downloading segments...")
                
                for i, seg in enumerate(segments):
                    if seg.startswith('http://') or seg.startswith('https://'):
                        seg_url = seg
                    elif base_url:
                        seg_url = base_url + seg
                    else:
                        seg_url = seg
                    seg_path = os.path.join(tmpdir, f"segment_{i:04d}.mp4")
                    temp_files.append(seg_path)
                    
                    max_retries = 3
                    success = False
                    
                    for attempt in range(max_retries):
                        try:
                            resp = requests.get(seg_url, headers=headers, timeout=60)
                            if resp.status_code == 200:
                                with open(seg_path, 'wb') as f:
                                    f.write(resp.content)
                                success = True
                                break
                            else:
                                print(f"⚠️ Segment {i+1} attempt {attempt+1} failed: {resp.status_code}")
                        except Exception as e:
                            print(f"⚠️ Segment {i+1} attempt {attempt+1} failed: {e}")
                    
                    if not success:
                        raise HTTPException(status_code=500, detail=f"Failed to download segment {i+1}")
                    
                    print(f"   Segment {i+1}/{len(segments)} downloaded")
                
                print(f"🔄 Merging {len(segments)} segments into {audio_format}...")
                print(f"   Temp files count: {len(temp_files)}")
                print(f"   Temp directory: {tmpdir}")
                
                # Check first segment header
                if temp_files:
                    with open(temp_files[0], 'rb') as f:
                        header = f.read(16).hex()
                        print(f"   First segment header (hex): {header}")
                        if header.startswith('0000001866747970'):
                            print("   ✅ First segment is MP4 (ftyp header)")
                        elif header.startswith('0000010c6d6f6f66'):
                            print("   ⚠️ First segment is fragmented MP4 (moof header)")
                        else:
                            print(f"   ❓ Unknown header: {header[:20]}...")
                
                output_filename = f"{artist} - {title}.{audio_format}"
                output_path = os.path.join(download_dir, output_filename)
                print(f"   Output path: {output_path}")
                
                if audio_format == "flac":
                    merged_path = os.path.join(tmpdir, "merged.mp4")
                    print(f"   Merged path: {merged_path}")
                    
                    with open(merged_path, 'wb') as out:
                        print("   Starting merge...")
                        for idx, temp_file in enumerate(temp_files):
                            with open(temp_file, 'rb') as f:
                                out.write(f.read())
                        print("   Merge complete")
                    
                    print("   Starting FFmpeg conversion...")
                    try:
                        import subprocess
                        
                        # First check if the merged file is valid with ffprobe
                        probe_result = subprocess.run([
                            'ffprobe', '-v', 'error', '-show_streams', '-of', 'json', merged_path
                        ], capture_output=True, text=True, timeout=30)
                        print(f"ffprobe return code: {probe_result.returncode}")
                        if probe_result.returncode == 0:
                            print(f"✅ Merged file is valid MP4")
                            print(f"ffprobe output: {probe_result.stdout[:500]}")
                        else:
                            print(f"❌ Merged file analysis failed: {probe_result.stderr}")
                        
                        # Use stream copy mode which is what OrpheusDL does
                        result = subprocess.run([
                            'ffmpeg', '-y', '-i', merged_path, 
                            '-c', 'copy', 
                            '-loglevel', 'error',
                            output_path
                        ], capture_output=True, text=True, timeout=120)
                        
                        print(f"FFmpeg return code: {result.returncode}")
                        print(f"FFmpeg stdout: {result.stdout[:2000] if result.stdout else 'None'}")
                        print(f"FFmpeg stderr: {result.stderr[:2000] if result.stderr else 'None'}")
                        
                        if result.returncode != 0:
                            print(f"⚠️ FFmpeg conversion failed with code {result.returncode}: {result.stderr}")
                            raise HTTPException(status_code=500, detail=f"FFmpeg conversion failed: {result.stderr}")
                        
                        print(f"✅ Converted to FLAC: {output_path}")
                    except FileNotFoundError:
                        print("⚠️ FFmpeg not found, using raw merge")
                        with open(output_path, 'wb') as out:
                            for temp_file in temp_files:
                                with open(temp_file, 'rb') as f:
                                    out.write(f.read())
                    except subprocess.TimeoutExpired:
                        print("⚠️ FFmpeg timeout")
                        raise HTTPException(status_code=500, detail="FFmpeg timeout")
                    except Exception as e:
                        print(f"⚠️ Exception during FFmpeg: {e}")
                        raise
                else:
                    with open(output_path, 'wb') as out:
                        for temp_file in temp_files:
                            with open(temp_file, 'rb') as f:
                                out.write(f.read())
                
                print(f"✅ Download complete: {output_path}")
                
                print("   Cleaning up temp files...")
                import shutil
                shutil.rmtree(tmpdir, ignore_errors=True)
                print("   Cleanup complete")
                
                file_size = os.path.getsize(output_path)
                
                return {
                    "success": True,
                    "track": {
                        "id": track_id,
                        "title": title,
                        "artist": artist,
                        "album": album,
                        "albumArtist": album_artist,
                        "trackNumber": track_number,
                        "duration": duration,
                        "isrc": isrc,
                        "quality": audio_quality,
                        "fileSize": file_size,
                        "format": audio_format
                    },
                    "downloadUrl": f"/download_file/{output_filename}",
                    "message": "File downloaded and ready for download"
                }
            finally:
                import shutil
                shutil.rmtree(tmpdir, ignore_errors=True)
        
        else:
            print("📦 Parsing BT manifest...")
            base_url, segments = parse_bt_manifest(manifest_data)
            print(f"✅ Found {len(segments)} URLs")
            
            if not segments:
                raise HTTPException(status_code=500, detail="No URLs found in BT manifest")
            
            download_url = segments[0] if segments else None
            
            return {
                "success": True,
                "track": {
                    "id": track_id,
                    "title": title,
                    "artist": artist,
                    "album": album,
                    "albumArtist": album_artist,
                    "trackNumber": track_number,
                    "duration": duration,
                    "isrc": isrc,
                    "quality": audio_quality,
                    "format": "m4a"
                },
                "downloadUrl": download_url,
                "message": "Direct TIDAL CDN URL (time-limited)"
            }
    
    except HTTPException:
        raise
    except Exception as e:
        print(f"❌ Unexpected error: {type(e).__name__}: {str(e)}")
        raise HTTPException(status_code=500, detail=str(e))

@app.get("/download_file/{filename}")
async def download_file(filename: str):
    """Serve downloaded file for LOSSLESS/HI_RES tracks"""
    try:
        file_path = os.path.join("./downloads", filename)
        
        if not os.path.exists(file_path):
            raise HTTPException(status_code=404, detail="File not found")
        
        ext = os.path.splitext(filename)[1].lower()
        media_type = "audio/flac" if ext == ".flac" else "audio/mp4"
        
        return FileResponse(
            file_path,
            media_type=media_type,
            filename=filename
        )
    except HTTPException:
        raise
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

@app.get("/search")
async def search_tracks(query: str = Query(..., description="Search query"), limit: int = 5):
    """Search for TIDAL tracks"""
    try:
        headers = get_headers()
        response = requests.get(
            "https://api.tidal.com/v1/search",
            params={
                "query": query,
                "countryCode": "HK",
                "limit": limit,
                "types": "TRACKS"
            },
            headers=headers,
            timeout=15
        )
        
        if response.status_code == 401:
            if refresh_access_token():
                headers = get_headers()
                response = requests.get(
                    "https://api.tidal.com/v1/search",
                    params={
                        "query": query,
                        "countryCode": "HK",
                        "limit": limit,
                        "types": "TRACKS"
                    },
                    headers=headers,
                    timeout=15
                )
        
        if response.status_code != 200:
            raise HTTPException(status_code=response.status_code, detail="Failed to search")
        
        data = response.json()
        tracks = data.get('tracks', {}).get('items', [])
        
        results = []
        for track in tracks:
            results.append({
                "id": track.get('id'),
                "title": track.get('title'),
                "artist": track.get('artist', {}).get('name', 'Unknown'),
                "quality": track.get('audioQuality', 'Unknown'),
                "duration": track.get('duration')
            })
        
        return {"results": results}
    
    except HTTPException:
        raise
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

@app.get("/track/{track_id}")
async def get_track_info(track_id: int):
    """Get detailed track information"""
    try:
        headers = get_headers()
        
        response = requests.get(
            f"https://api.tidal.com/v1/tracks/{track_id}",
            params={"countryCode": "HK"},
            headers=headers,
            timeout=15
        )
        
        if response.status_code == 401:
            if refresh_access_token():
                headers = get_headers()
                response = requests.get(
                    f"https://api.tidal.com/v1/tracks/{track_id}",
                    params={"countryCode": "HK"},
                    headers=headers,
                    timeout=15
                )
        
        if response.status_code != 200:
            raise HTTPException(status_code=response.status_code, detail="Failed to get track info")
        
        return response.json()
    
    except HTTPException:
        raise
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))

@app.get("/")
async def root():
    """API endpoint listing"""
    return {
        "message": "TIDAL Download API",
        "endpoints": {
            "/download/{track_id}": "Download track (query: quality=LOSSLESS/HIGH/HI_RES_LOSSLESS)",
            "/search": "Search tracks (query: query, limit)",
            "/track/{track_id}": "Get track information"
        },
        "usage": "GET /download/5267743?quality=LOSSLESS"
    }

if __name__ == "__main__":
    import uvicorn
    port = int(os.environ.get("PORT", 7860))
    uvicorn.run(app, host="0.0.0.0", port=port)