436 lines
17 KiB
C#
436 lines
17 KiB
C#
using AngleSharp.Parser.Html;
|
|
using DuoVia.FuzzyStrings;
|
|
using FireSharp.Config;
|
|
using FireSharp.Interfaces;
|
|
using Herse.Models;
|
|
using Newtonsoft.Json;
|
|
using Newtonsoft.Json.Linq;
|
|
using System;
|
|
using System.Collections.Generic;
|
|
using System.Configuration;
|
|
using System.IO;
|
|
using System.Linq;
|
|
using System.Net;
|
|
using System.Text;
|
|
using System.Text.RegularExpressions;
|
|
using System.Threading.Tasks;
|
|
|
|
namespace BillboardPlaylistUpdater
|
|
{
|
|
class Program
|
|
{
|
|
static List<Song> songs = null;
|
|
static List<SongList> songList = null;
|
|
|
|
static void Main(string[] args)
|
|
{
|
|
//args = new string[] { "mbrucedogs" };
|
|
if (args.Length != 1)
|
|
{
|
|
Console.WriteLine("usage: songcrawler partyid songspath");
|
|
return;
|
|
}
|
|
string controller = args[0];
|
|
|
|
IFirebaseConfig config = new FirebaseConfig
|
|
{
|
|
AuthSecret = ConfigurationManager.AppSettings["Firebase.Secret"],
|
|
BasePath = ConfigurationManager.AppSettings["Firebase.Path"]
|
|
};
|
|
FireSharp.FirebaseClient client = new FireSharp.FirebaseClient(config);
|
|
songs = client.Get(string.Format("controllers/{0}/songs", controller)).ResultAs<List<Song>>();
|
|
|
|
string firepath = "songList";
|
|
Console.WriteLine("Loading current library");
|
|
songList = client.Get(firepath).ResultAs<List<SongList>>();
|
|
|
|
if (songList != null)
|
|
Console.WriteLine(string.Format("{0} songList loaded", songList.Count));
|
|
else
|
|
songList = new List<SongList>();
|
|
|
|
////RunTest();
|
|
|
|
//// TEST MODE: Only process first list and don't save back to Firebase
|
|
//Console.WriteLine("TEST MODE: Processing only first list...");
|
|
//if (songList != null && songList.Count > 0)
|
|
//{
|
|
// var firstList = songList[0];
|
|
// Console.WriteLine("********************************************************");
|
|
// Console.WriteLine(string.Format("Matching Controllers Songs for {0}", firstList.Title));
|
|
// Console.WriteLine("********************************************************");
|
|
// Search(firstList);
|
|
|
|
// // Show results summary
|
|
// Console.WriteLine("\n=== TEST RESULTS SUMMARY ===");
|
|
// int totalMatches = 0;
|
|
// foreach (var song in firstList.Songs)
|
|
// {
|
|
// Console.WriteLine($"{song.Title} - {song.Artist}: {song.FoundSongs.Count} matches");
|
|
// totalMatches += song.FoundSongs.Count;
|
|
// }
|
|
// Console.WriteLine($"Total matches across all songs: {totalMatches}");
|
|
//}
|
|
//else
|
|
//{
|
|
// Console.WriteLine("No song lists found to test.");
|
|
//}
|
|
|
|
// Commented out Firebase saves for testing
|
|
client.Set(firepath, songList);
|
|
client.Set(string.Format("controllers/{0}/songList", controller), songList);
|
|
UpdateSearchLists();
|
|
client.Set(string.Format("controllers/{0}/songList", controller), songList);
|
|
}
|
|
|
|
static void UpdateSearchLists()
|
|
{
|
|
//update the controller SongLists using parallel processing
|
|
Console.WriteLine($"Processing {songList.Count} lists in parallel...");
|
|
|
|
var options = new ParallelOptions { MaxDegreeOfParallelism = Environment.ProcessorCount };
|
|
|
|
Parallel.ForEach(songList, options, list =>
|
|
{
|
|
Console.WriteLine($"********************************************************");
|
|
Console.WriteLine($"Matching Controllers Songs for {list.Title}");
|
|
Console.WriteLine($"********************************************************");
|
|
Search(list);
|
|
});
|
|
}
|
|
|
|
static void Search(SongList list)
|
|
{
|
|
if (list.Songs == null || songs == null)
|
|
return;
|
|
|
|
// Pre-filter disabled songs once
|
|
var availableSongs = songs.Where(s => !s.Disabled).ToList();
|
|
Console.WriteLine($"Searching through {availableSongs.Count} available songs for {list.Songs.Count} songs in list...");
|
|
|
|
// Build fast lookup indexes
|
|
var titleIndex = BuildTitleIndex(availableSongs);
|
|
var artistIndex = BuildArtistIndex(availableSongs);
|
|
|
|
// Process each song in the list sequentially
|
|
foreach (var songListItem in list.Songs)
|
|
{
|
|
Console.WriteLine($"Searching for: {songListItem.Title} - {songListItem.Artist}");
|
|
|
|
var songMatches = new List<(Song song, double score)>();
|
|
|
|
// Get candidate matches using indexes
|
|
var titleCandidates = GetTitleCandidates(songListItem.Title, titleIndex);
|
|
var artistCandidates = GetArtistCandidates(songListItem.Artist, artistIndex);
|
|
|
|
// Combine and deduplicate candidates
|
|
var allCandidates = titleCandidates.Union(artistCandidates).Distinct().ToList();
|
|
|
|
Console.WriteLine($" Found {allCandidates.Count} candidates to evaluate...");
|
|
|
|
// Evaluate only the candidates (much faster than checking all songs)
|
|
foreach (var song in allCandidates)
|
|
{
|
|
// Calculate similarity scores for title and artist
|
|
double titleSimilarity = CalculateSimilarity(songListItem.Title, song.Title);
|
|
double artistSimilarity = CalculateSimilarity(songListItem.Artist, song.Artist);
|
|
|
|
// Combined score (weighted average - title is more important)
|
|
double combinedScore = (titleSimilarity * 0.7) + (artistSimilarity * 0.3);
|
|
|
|
// If combined score is above threshold, consider it a match
|
|
if (combinedScore >= 0.85) // Adjustable threshold
|
|
{
|
|
songMatches.Add((song, combinedScore));
|
|
}
|
|
}
|
|
|
|
// Sort matches by relevance and file type priority
|
|
songListItem.FoundSongs = songMatches
|
|
.OrderByDescending(x => x.score)
|
|
.ThenBy(x => GetFileTypePriority(x.song))
|
|
.ThenBy(x => GetChannelPriority(x.song))
|
|
.Take(10) // Limit to top 10 matches
|
|
.Select(x => x.song)
|
|
.ToList();
|
|
|
|
Console.WriteLine($" Total matches found: {songListItem.FoundSongs.Count}");
|
|
}
|
|
}
|
|
|
|
static double CalculateSimilarity(string str1, string str2)
|
|
{
|
|
if (string.IsNullOrEmpty(str1) || string.IsNullOrEmpty(str2))
|
|
return 0.0;
|
|
|
|
// Normalize strings for comparison
|
|
str1 = NormalizeString(str1);
|
|
str2 = NormalizeString(str2);
|
|
|
|
// Use multiple fuzzy string algorithms and combine results
|
|
double diceCoeff = str1.DiceCoefficient(str2);
|
|
double levenshtein = 1.0 - (str1.LevenshteinDistance(str2) / (double)Math.Max(str1.Length, str2.Length));
|
|
|
|
// Use Longest Common Subsequence as a third algorithm
|
|
double lcsScore = CalculateLCSSimilarity(str1, str2);
|
|
|
|
// Weighted average of different algorithms
|
|
return (diceCoeff * 0.4) + (levenshtein * 0.3) + (lcsScore * 0.3);
|
|
}
|
|
|
|
static double CalculateCombinedSimilarity(SongListSong songListItem, Song song)
|
|
{
|
|
double titleSimilarity = CalculateSimilarity(songListItem.Title, song.Title);
|
|
double artistSimilarity = CalculateSimilarity(songListItem.Artist, song.Artist);
|
|
|
|
// Weight title more heavily than artist
|
|
return (titleSimilarity * 0.7) + (artistSimilarity * 0.3);
|
|
}
|
|
|
|
static string NormalizeString(string input)
|
|
{
|
|
if (string.IsNullOrEmpty(input))
|
|
return string.Empty;
|
|
|
|
// Remove common karaoke prefixes/suffixes
|
|
var normalized = input.ToLowerInvariant();
|
|
|
|
// Remove common karaoke indicators
|
|
normalized = Regex.Replace(normalized, @"\b(karaoke|karaoke version|instrumental|backing track)\b", "", RegexOptions.IgnoreCase);
|
|
|
|
// Remove extra whitespace and punctuation
|
|
normalized = Regex.Replace(normalized, @"\s+", " ");
|
|
normalized = Regex.Replace(normalized, @"[^\w\s]", "");
|
|
|
|
return normalized.Trim();
|
|
}
|
|
|
|
static double CalculateLCSSimilarity(string str1, string str2)
|
|
{
|
|
if (string.IsNullOrEmpty(str1) || string.IsNullOrEmpty(str2))
|
|
return 0.0;
|
|
|
|
int lcsLength = CalculateLCSLength(str1, str2);
|
|
int maxLength = Math.Max(str1.Length, str2.Length);
|
|
|
|
if (maxLength == 0)
|
|
return 1.0; // Both strings are empty
|
|
|
|
return (double)lcsLength / maxLength;
|
|
}
|
|
|
|
static int CalculateLCSLength(string str1, string str2)
|
|
{
|
|
int[,] dp = new int[str1.Length + 1, str2.Length + 1];
|
|
|
|
for (int i = 1; i <= str1.Length; i++)
|
|
{
|
|
for (int j = 1; j <= str2.Length; j++)
|
|
{
|
|
if (str1[i - 1] == str2[j - 1])
|
|
{
|
|
dp[i, j] = dp[i - 1, j - 1] + 1;
|
|
}
|
|
else
|
|
{
|
|
dp[i, j] = Math.Max(dp[i - 1, j], dp[i, j - 1]);
|
|
}
|
|
}
|
|
}
|
|
|
|
return dp[str1.Length, str2.Length];
|
|
}
|
|
|
|
static int GetFileTypePriority(Song song)
|
|
{
|
|
// MP4 files get priority (lower number = higher priority)
|
|
if (song.FileType == FileType.MP4)
|
|
return 0;
|
|
else
|
|
return 1;
|
|
}
|
|
|
|
static int GetChannelPriority(Song song)
|
|
{
|
|
// Channel priorities in order (lower number = higher priority)
|
|
string[] channelPriorities = {
|
|
"Sing King Karaoke",
|
|
"KaraFun Karaoke",
|
|
"Stingray Karaoke"
|
|
};
|
|
|
|
// Extract folder name from path
|
|
string folderName = ExtractFolderName(song.Path);
|
|
|
|
// Find the priority index
|
|
for (int i = 0; i < channelPriorities.Length; i++)
|
|
{
|
|
if (folderName.IndexOf(channelPriorities[i], StringComparison.OrdinalIgnoreCase) >= 0)
|
|
{
|
|
return i; // Return the priority index (0 = highest priority)
|
|
}
|
|
}
|
|
|
|
// If not found in priority list, give it lowest priority
|
|
return channelPriorities.Length;
|
|
}
|
|
|
|
static string ExtractFolderName(string path)
|
|
{
|
|
if (string.IsNullOrEmpty(path))
|
|
return string.Empty;
|
|
|
|
try
|
|
{
|
|
// Get the directory name from the path
|
|
string directory = Path.GetDirectoryName(path);
|
|
if (string.IsNullOrEmpty(directory))
|
|
return string.Empty;
|
|
|
|
// Get the last folder name
|
|
return Path.GetFileName(directory);
|
|
}
|
|
catch
|
|
{
|
|
return string.Empty;
|
|
}
|
|
}
|
|
|
|
// Index building methods for fast lookup
|
|
static Dictionary<string, List<Song>> BuildTitleIndex(List<Song> songs)
|
|
{
|
|
var index = new Dictionary<string, List<Song>>();
|
|
|
|
foreach (var song in songs)
|
|
{
|
|
var normalizedTitle = NormalizeString(song.Title);
|
|
var words = normalizedTitle.Split(' ', (char)StringSplitOptions.RemoveEmptyEntries);
|
|
|
|
// Index by each word in the title
|
|
foreach (var word in words)
|
|
{
|
|
if (word.Length >= 3) // Only index words with 3+ characters
|
|
{
|
|
if (!index.ContainsKey(word))
|
|
index[word] = new List<Song>();
|
|
index[word].Add(song);
|
|
}
|
|
}
|
|
|
|
// Also index by first few characters for prefix matching
|
|
for (int i = 3; i <= Math.Min(8, normalizedTitle.Length); i++)
|
|
{
|
|
var prefix = normalizedTitle.Substring(0, i);
|
|
if (!index.ContainsKey(prefix))
|
|
index[prefix] = new List<Song>();
|
|
index[prefix].Add(song);
|
|
}
|
|
}
|
|
|
|
return index;
|
|
}
|
|
|
|
static Dictionary<string, List<Song>> BuildArtistIndex(List<Song> songs)
|
|
{
|
|
var index = new Dictionary<string, List<Song>>();
|
|
|
|
foreach (var song in songs)
|
|
{
|
|
var normalizedArtist = NormalizeString(song.Artist);
|
|
var words = normalizedArtist.Split(' ', (char)StringSplitOptions.RemoveEmptyEntries);
|
|
|
|
// Index by each word in the artist name
|
|
foreach (var word in words)
|
|
{
|
|
if (word.Length >= 3) // Only index words with 3+ characters
|
|
{
|
|
if (!index.ContainsKey(word))
|
|
index[word] = new List<Song>();
|
|
index[word].Add(song);
|
|
}
|
|
}
|
|
|
|
// Also index by first few characters for prefix matching
|
|
for (int i = 3; i <= Math.Min(8, normalizedArtist.Length); i++)
|
|
{
|
|
var prefix = normalizedArtist.Substring(0, i);
|
|
if (!index.ContainsKey(prefix))
|
|
index[prefix] = new List<Song>();
|
|
index[prefix].Add(song);
|
|
}
|
|
}
|
|
|
|
return index;
|
|
}
|
|
|
|
static List<Song> GetTitleCandidates(string searchTitle, Dictionary<string, List<Song>> titleIndex)
|
|
{
|
|
var candidates = new HashSet<Song>();
|
|
var normalizedSearch = NormalizeString(searchTitle);
|
|
var searchWords = normalizedSearch.Split(' ', (char)StringSplitOptions.RemoveEmptyEntries);
|
|
|
|
// Find songs that contain any of the search words
|
|
foreach (var word in searchWords)
|
|
{
|
|
if (word.Length >= 3 && titleIndex.ContainsKey(word))
|
|
{
|
|
foreach (var song in titleIndex[word])
|
|
{
|
|
candidates.Add(song);
|
|
}
|
|
}
|
|
|
|
// Also check prefixes
|
|
for (int i = 3; i <= Math.Min(8, word.Length); i++)
|
|
{
|
|
var prefix = word.Substring(0, i);
|
|
if (titleIndex.ContainsKey(prefix))
|
|
{
|
|
foreach (var song in titleIndex[prefix])
|
|
{
|
|
candidates.Add(song);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
return candidates.ToList();
|
|
}
|
|
|
|
static List<Song> GetArtistCandidates(string searchArtist, Dictionary<string, List<Song>> artistIndex)
|
|
{
|
|
var candidates = new HashSet<Song>();
|
|
var normalizedSearch = NormalizeString(searchArtist);
|
|
var searchWords = normalizedSearch.Split(' ', (char)StringSplitOptions.RemoveEmptyEntries);
|
|
|
|
// Find songs that contain any of the search words
|
|
foreach (var word in searchWords)
|
|
{
|
|
if (word.Length >= 3 && artistIndex.ContainsKey(word))
|
|
{
|
|
foreach (var song in artistIndex[word])
|
|
{
|
|
candidates.Add(song);
|
|
}
|
|
}
|
|
|
|
// Also check prefixes
|
|
for (int i = 3; i <= Math.Min(8, word.Length); i++)
|
|
{
|
|
var prefix = word.Substring(0, i);
|
|
if (artistIndex.ContainsKey(prefix))
|
|
{
|
|
foreach (var song in artistIndex[prefix])
|
|
{
|
|
candidates.Add(song);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
return candidates.ToList();
|
|
}
|
|
}
|
|
}
|
|
|