Lidarr optimizations

UmlautAdaptarr/Controllers/SearchController.cs

@@ -184,7 +184,7 @@ namespace UmlautAdaptarr.Controllers
                     if (categories.Split(',').Any(category => AUDIO_CATEGORY_IDS.Contains(category)))
                     {
                         var mediaType = "audio";
-                        searchItem = await searchItemLookupService.GetOrFetchSearchItemByExternalId(mediaType, title.ToLower());
+                        searchItem = await searchItemLookupService.GetOrFetchSearchItemByExternalId(mediaType, title.GetLidarrTitleForExternalId());
                     }
                 }
             }

UmlautAdaptarr/Models/SearchItem.cs

@@ -12,6 +12,7 @@ namespace UmlautAdaptarr.Models
         public bool HasUmlaut => Title?.HasUmlauts() ?? false;
         public string ExpectedTitle { get; set; }
         public string? ExpectedAuthor { get; set; }
+        // TODO rename GermanTitle into Foreign or LocalTitle?
         public string? GermanTitle { get; set; }
         public string[] TitleSearchVariations { get; set; }
         public string[] TitleMatchVariations { get; set; }
@@ -42,6 +43,11 @@ namespace UmlautAdaptarr.Models
                 if (expectedTitle.Contains(expectedAuthor))
                 {
                     var titleWithoutAuthorName = expectedTitle.Replace(expectedAuthor, string.Empty).RemoveExtraWhitespaces().Trim();
+
+                    if (titleWithoutAuthorName.Length < 2)
+                    {
+                        // TODO log warning that this album can't be searched for automatically
+                    }
                     TitleMatchVariations = GenerateVariations(titleWithoutAuthorName, mediaType).ToArray();
                 }
                 else
@@ -86,13 +92,18 @@ namespace UmlautAdaptarr.Models
             }
         }
 
-        private IEnumerable<string> GenerateVariations(string? germanTitle, string mediaType)
+        private IEnumerable<string> GenerateVariations(string? title, string mediaType)
         {
-            if (germanTitle == null)
+            if (title == null)
+            {
+                return [];
+            }
+            var cleanTitle = title.GetCleanTitle();
+
+            if (cleanTitle?.Length == 0)
             {
                 return [];
             }
-            var cleanTitle = germanTitle.RemoveAccentButKeepGermanUmlauts().GetCleanTitle();
 
             // Start with base variations including handling umlauts
             var baseVariations = new List<string>
@@ -121,11 +132,16 @@ namespace UmlautAdaptarr.Models
                 });
             }
 
-            // If a german title starts with der/die/das also accept variations without it
-            if (mediaType != "audio" && cleanTitle.StartsWith("Der") || cleanTitle.StartsWith("Die") || cleanTitle.StartsWith("Das"))
+            // If a title starts with der/die/das also accept variations without it
+            // Same for english the, an, a
+            if (cleanTitle.StartsWith("Der ") || cleanTitle.StartsWith("Die ") || cleanTitle.StartsWith("Das ")
+                || cleanTitle.StartsWith("The ") || cleanTitle.StartsWith("An "))
             {
-                var cleanTitleWithoutArticle = germanTitle[3..].Trim();
+                var cleanTitleWithoutArticle = title[3..].Trim();
                 baseVariations.AddRange(GenerateVariations(cleanTitleWithoutArticle, mediaType));
+            } else if (cleanTitle.StartsWith("A "))
+            {
+                var cleanTitleWithoutArticle = title[2..].Trim();
             }
 
             // Remove multiple spaces

UmlautAdaptarr/Providers/LidarrClient.cs

@@ -42,16 +42,17 @@ namespace UmlautAdaptarr.Providers
 
                     var lidarrAlbumUrl = $"{_lidarrHost}/api/v1/album?artistId={artistId}&apikey={_lidarrApiKey}";
 
-                    if (cache.TryGetValue(lidarrAlbumUrl, out List<dynamic>? albums))
-                    {
-                        logger.LogInformation($"Using cached albums for {UrlUtilities.RedactApiKey(lidarrAlbumUrl)}");
-                    }
-                    else
-                    {
+                    // Disable cache for now as it can result in problems when adding new albums that aren't displayed on the artists page initially
+                    //if (cache.TryGetValue(lidarrAlbumUrl, out List<dynamic>? albums))
+                    //{
+                    //    logger.LogInformation($"Using cached albums for {UrlUtilities.RedactApiKey(lidarrAlbumUrl)}");
+                    //}
+                    //else
+                    //{
                     logger.LogInformation($"Fetching all albums from artistId {artistId} from Lidarr: {UrlUtilities.RedactApiKey(lidarrArtistsUrl)}");
                     var albumApiResponse = await httpClient.GetStringAsync(lidarrAlbumUrl);
-                        albums = JsonConvert.DeserializeObject<List<dynamic>>(albumApiResponse);
-                    }
+                    var albums = JsonConvert.DeserializeObject<List<dynamic>>(albumApiResponse);
+                    //}
 
                     if (albums == null)
                     {
@@ -74,7 +75,7 @@ namespace UmlautAdaptarr.Providers
                         string[]? aliases = null;
 
                         // Abuse externalId to set the search string Lidarr uses
-                        var externalId = expectedTitle.RemoveGermanUmlautDots().RemoveAccent().RemoveSpecialCharacters().RemoveExtraWhitespaces().ToLower();
+                        var externalId = expectedTitle.GetLidarrTitleForExternalId();
 
                         var searchItem = new SearchItem
                         (

UmlautAdaptarr/Utilities/Extensions.cs

@@ -47,20 +47,46 @@ namespace UmlautAdaptarr.Utilities
             return stringBuilder.ToString().Normalize(NormalizationForm.FormC);
         }
 
-        // TODO possibly replace GetCleanTitle with RemoveSpecialCharacters
+        public static string GetLidarrTitleForExternalId(this string text)
+        {
+            text = text.RemoveGermanUmlautDots()
+                       .GetCleanTitle()
+                       .ToLower();
+
+            // Lidarr removes the, an and a
+            return TitlePrefixRegex()
+                .Replace(text, "")
+                .RemoveExtraWhitespaces()
+                .Trim();
+        }
+
         public static string GetCleanTitle(this string text)
         {
-            return text.Replace("(", "").Replace(")", "").Replace("?","").Replace(":", "").Replace("'", "");
+            return text
+                .Replace(".", " ")
+                .Replace(":", " ")
+                .RemoveAccentButKeepGermanUmlauts()
+                .RemoveSpecialCharacters(removeUmlauts: false)
+                .RemoveExtraWhitespaces()
+                .Trim();
         }
 
         public static string NormalizeForComparison(this string text)
         {
+            // TODO see if we can replace RemoveGermanUmlautDots() with RemoveSpecialCharacters(removeUmlauts: false);
             return text.RemoveGermanUmlautDots().RemoveAccent().RemoveSpecialCharacters().Replace(" ", "").Trim().ToLower();
         }
 
-        public static string RemoveSpecialCharacters(this string text)
+        public static string RemoveSpecialCharacters(this string text, bool removeUmlauts = true)
         {
-            return SpecialCharactersRegex().Replace(text, "");
+            if (removeUmlauts)
+            {
+                return NoSpecialCharactersRegex().Replace(text, "");
+            }
+            else
+            {
+                return NoSpecialCharactersExceptUmlautsRegex().Replace(text, "");
+            }
         }
 
 
@@ -101,9 +127,15 @@ namespace UmlautAdaptarr.Utilities
         }
 
         [GeneratedRegex("[^a-zA-Z0-9 ]+", RegexOptions.Compiled)]
-        private static partial Regex SpecialCharactersRegex();
+        private static partial Regex NoSpecialCharactersRegex();
+
+        [GeneratedRegex("[^a-zA-Z0-9 öäüßÖÄÜ]+", RegexOptions.Compiled)]
+        private static partial Regex NoSpecialCharactersExceptUmlautsRegex();
 
         [GeneratedRegex(@"\s+")]
         private static partial Regex MultipleWhitespaceRegex();
+
+        [GeneratedRegex(@"\b(the|an|a)\b", RegexOptions.IgnoreCase, "de-DE")]
+        private static partial Regex TitlePrefixRegex();
     }
 }