Merge pull request #6383 from sushilicious/master

Made default parser a tiny bit mroe robust

Emby.Naming/Common/NamingOptions.cs

@@ -137,8 +137,11 @@ namespace Emby.Naming.Common
 
             CleanStrings = new[]
             {
-                @"[ _\,\.\(\)\[\]\-](3d|sbs|tab|hsbs|htab|mvc|HDR|HDC|UHD|UltraHD|4k|ac3|dts|custom|dc|divx|divx5|dsr|dsrip|dutch|dvd|dvdrip|dvdscr|dvdscreener|screener|dvdivx|cam|fragment|fs|hdtv|hdrip|hdtvrip|internal|limited|multisubs|ntsc|ogg|ogm|pal|pdtv|proper|repack|rerip|retail|cd[1-9]|r3|r5|bd5|bd|se|svcd|swedish|german|read.nfo|nfofix|unrated|ws|telesync|ts|telecine|tc|brrip|bdrip|480p|480i|576p|576i|720p|720i|1080p|1080i|2160p|hrhd|hrhdtv|hddvd|bluray|blu-ray|x264|x265|h264|h265|xvid|xvidvd|xxx|www.www|AAC|DTS|\[.*\])([ _\,\.\(\)\[\]\-]|$)",
+                @"^\s*(?<cleaned>.+?)[ _\,\.\(\)\[\]\-](3d|sbs|tab|hsbs|htab|mvc|HDR|HDC|UHD|UltraHD|4k|ac3|dts|custom|dc|divx|divx5|dsr|dsrip|dutch|dvd|dvdrip|dvdscr|dvdscreener|screener|dvdivx|cam|fragment|fs|hdtv|hdrip|hdtvrip|internal|limited|multisubs|ntsc|ogg|ogm|pal|pdtv|proper|repack|rerip|retail|cd[1-9]|r3|r5|bd5|bd|se|svcd|swedish|german|read.nfo|nfofix|unrated|ws|telesync|ts|telecine|tc|brrip|bdrip|480p|480i|576p|576i|720p|720i|1080p|1080i|2160p|hrhd|hrhdtv|hddvd|bluray|blu-ray|x264|x265|h264|h265|xvid|xvidvd|xxx|www.www|AAC|DTS|\[.*\])([ _\,\.\(\)\[\]\-]|$)",
-                @"(\[.*\])"
+                @"^(?<cleaned>.+?)(\[.*\])",
+                @"^\s*(?<cleaned>.+?)\WE[0-9]+(-|~)E?[0-9]+(\W|$)",
+                @"^\s*\[[^\]]+\](?!\.\w+$)\s*(?<cleaned>.+)",
+                @"^\s*(?<cleaned>.+?)\s+-\s+[0-9]+\s*$"
             };
 
             SubtitleFileExtensions = new[]

Emby.Naming/Video/CleanStringParser.cs

@@ -25,26 +25,35 @@ namespace Emby.Naming.Video
                 return false;
             }
 
-            var len = expressions.Count;
+            // Iteratively apply the regexps to clean the string.
-            for (int i = 0; i < len; i++)
+            bool cleaned = false;
+            for (int i = 0; i < expressions.Count; i++)
             {
                 if (TryClean(name, expressions[i], out newName))
                 {
-                    return true;
+                    cleaned = true;
+                    name = newName.ToString();
                 }
             }
 
-            newName = ReadOnlySpan<char>.Empty;
+            newName = cleaned ? name.AsSpan() : ReadOnlySpan<char>.Empty;
-            return false;
+            return cleaned;
         }
 
         private static bool TryClean(string name, Regex expression, out ReadOnlySpan<char> newName)
         {
             var match = expression.Match(name);
             int index = match.Index;
-            if (match.Success && index != 0)
+            if (match.Success)
+            {
+                var found = match.Groups.TryGetValue("cleaned", out var cleaned);
+                if (!found || cleaned == null)
                 {
-                newName = name.AsSpan().Slice(0, match.Index);
+                    newName = ReadOnlySpan<char>.Empty;
+                    return false;
+                }
+
+                newName = name.AsSpan().Slice(cleaned.Index, cleaned.Length);
                 return true;
             }
 

tests/Jellyfin.Naming.Tests/Video/CleanStringTests.cs

@@ -1,4 +1,4 @@
-using System;
+using System;
 using Emby.Naming.Common;
 using Emby.Naming.Video;
 using Xunit;
@@ -23,6 +23,12 @@ namespace Jellyfin.Naming.Tests.Video
         [InlineData("Crouching.Tiger.Hidden.Dragon.BDrip.mkv", "Crouching.Tiger.Hidden.Dragon")]
         [InlineData("Crouching.Tiger.Hidden.Dragon.BDrip-HDC.mkv", "Crouching.Tiger.Hidden.Dragon")]
         [InlineData("Crouching.Tiger.Hidden.Dragon.4K.UltraHD.HDR.BDrip-HDC.mkv", "Crouching.Tiger.Hidden.Dragon")]
+        [InlineData("[HorribleSubs] Made in Abyss - 13 [720p].mkv", "Made in Abyss")]
+        [InlineData("[Tsundere] Kore wa Zombie Desu ka of the Dead [BDRip h264 1920x1080 FLAC]", "Kore wa Zombie Desu ka of the Dead")]
+        [InlineData("[Erai-raws] Jujutsu Kaisen - 03 [720p][Multiple Subtitle].mkv", "Jujutsu Kaisen")]
+        [InlineData("[OCN] 애타는 로맨스 720p-NEXT", "애타는 로맨스")]
+        [InlineData("[tvN] 혼술남녀.E01-E16.720p-NEXT", "혼술남녀")]
+        [InlineData("[tvN] 연애말고 결혼 E01~E16 END HDTV.H264.720p-WITH", "연애말고 결혼")]
         // FIXME: [InlineData("After The Sunset - [0004].mkv", "After The Sunset")]
         public void CleanStringTest_NeedsCleaning_Success(string input, string expectedName)
         {