Use tokenizer for emote parsing :HideThePain:

2024-05-20 20:14:41 +02:00
parent aab4977ecf
commit 47a4c5ab8d
2 changed files with 147 additions and 14 deletions
@@ -202,26 +202,42 @@ internal partial class Message
                continue;
            }

-            var builder = new StringBuilder();
-            foreach (var word in text.Content.Split(" "))
+            var wordBuilder = new StringBuilder();
+            var sentenceBuilder = new StringBuilder();
+            foreach (var token in Tokenizer.PrecedenceBasedRegexTokenizer.Tokenize(text.Content))
            {
-                if (checkForEmotes && EmoteCache.Exists(word) && !Plugin.Config.BlockedEmotes.Contains(word))
+                if (token.TokenType == Tokenizer.TokenType.StringValue)
                {
-                    // We add all the previous collected text parts
-                    AddContentAfterURLCheck(builder.ToString(), text, chunk);
-                    builder.Clear();
-
-                    AddChunkWithMessage(new TextChunk(chunk.Source, EmotePayload.ResolveEmote(word), word) { FallbackColour = text.FallbackColour });
-                    builder.Append(' ');
+                    wordBuilder.Append(token.Value);
                    continue;
                }

-                builder.Append($"{word} ");
-            }
+                var word = wordBuilder.ToString();
+                wordBuilder.Clear();

-            // We add the leftovers
-            // Removing the last whitespace as it is set by us
-            AddContentAfterURLCheck(builder.ToString()[..^1], text, chunk);
+                if (checkForEmotes && EmoteCache.Exists(word) && !Plugin.Config.BlockedEmotes.Contains(word))
+                {
+                    // Add the previous punctuation, including whitespaces
+                    AddContentAfterURLCheck(sentenceBuilder.ToString(), text, chunk);
+                    AddChunkWithMessage(new TextChunk(chunk.Source, EmotePayload.ResolveEmote(word), word) { FallbackColour = text.FallbackColour });
+
+                    // Append our current match as it is a special split symbol
+                    sentenceBuilder.Clear();
+                    sentenceBuilder.Append(token.Value);
+                    continue;
+                }
+
+                // Append match if we haven't reached end of string yet
+                if (token.TokenType != Tokenizer.TokenType.SequenceTerminator)
+                {
+                    sentenceBuilder.Append(word);
+                    sentenceBuilder.Append(token.Value);
+                    continue;
+                }
+
+                // End of string reached, we add our leftover
+                AddContentAfterURLCheck(sentenceBuilder.Append(word).ToString(), text, chunk);
+            }
        }

        return newChunks;
@@ -0,0 +1,117 @@
+using System.Text.RegularExpressions;
+
+namespace ChatTwo.Util;
+
+// Modified from: https://jack-vanlightly.com/blog/2016/2/24/a-more-efficient-regex-tokenizer
+public static class Tokenizer
+{
+    public enum TokenType
+    {
+        CloseParenthesis,
+        Comma,
+        Dot,
+        QuestionMark,
+        ExclamationMark,
+        Semicolon,
+        Whitespace,
+        Equals,
+        OpenParenthesis,
+        StringValue,
+        Leftover,
+        SequenceTerminator
+    }
+
+    public class Token(TokenType tokenType, string value)
+    {
+        public Token(TokenType tokenType) : this(tokenType, string.Empty) { }
+
+        public TokenType TokenType { get; } = tokenType;
+        public string Value { get; } = value;
+    }
+
+    public static class PrecedenceBasedRegexTokenizer
+    {
+        private static readonly List<TokenDefinition> TokenDefinitions;
+
+        static PrecedenceBasedRegexTokenizer()
+        {
+            TokenDefinitions = new List<TokenDefinition>
+            {
+                new(TokenType.CloseParenthesis, "\\)", 1),
+                new(TokenType.Comma, ",", 1),
+                new(TokenType.Dot, "\\.", 1),
+                new(TokenType.QuestionMark, "\\?", 1),
+                new(TokenType.ExclamationMark, "!", 1),
+                new(TokenType.Semicolon, ";", 1),
+                new(TokenType.Whitespace, "\\s", 1),
+                new(TokenType.Equals, "=", 1),
+                new(TokenType.OpenParenthesis, "\\(", 1),
+                new(TokenType.StringValue, "\\p{IsBasicLatin}", 2),
+                new(TokenType.Leftover, ".", 3)
+            };
+        }
+
+        public static IEnumerable<Token> Tokenize(string lqlText)
+        {
+            var tokenMatches = FindTokenMatches(lqlText);
+
+            var groupedByIndex = tokenMatches.GroupBy(x => x.StartIndex)
+                .OrderBy(x => x.Key)
+                .ToList();
+
+            TokenMatch? lastMatch = null;
+            foreach (var t in groupedByIndex)
+            {
+                var bestMatch = t.OrderBy(x => x.Precedence).First();
+                if (lastMatch != null && bestMatch.StartIndex < lastMatch.EndIndex)
+                    continue;
+
+                yield return new Token(bestMatch.TokenType, bestMatch.Value);
+
+                lastMatch = bestMatch;
+            }
+
+            yield return new Token(TokenType.SequenceTerminator);
+        }
+
+        private static List<TokenMatch> FindTokenMatches(string lqlText)
+        {
+            var tokenMatches = new List<TokenMatch>();
+
+            foreach (var tokenDefinition in TokenDefinitions)
+                tokenMatches.AddRange(tokenDefinition.FindMatches(lqlText).ToList());
+
+            return tokenMatches;
+        }
+    }
+
+    private class TokenDefinition(TokenType returnsToken, string regexPattern, int precedence)
+    {
+        private readonly Regex Regex = new(regexPattern, RegexOptions.IgnoreCase|RegexOptions.Compiled);
+
+        public IEnumerable<TokenMatch> FindMatches(string inputString)
+        {
+            var matches = Regex.Matches(inputString);
+            for(var i = 0; i < matches.Count; i++)
+            {
+                yield return new TokenMatch
+                {
+                    StartIndex = matches[i].Index,
+                    EndIndex = matches[i].Index + matches[i].Length,
+                    TokenType = returnsToken,
+                    Value = matches[i].Value,
+                    Precedence = precedence
+                };
+            }
+        }
+    }
+
+    private class TokenMatch
+    {
+        public TokenType TokenType { get; set; }
+        public string Value { get; set; }
+        public int StartIndex { get; set; }
+        public int EndIndex { get; set; }
+        public int Precedence { get; set; }
+    }
+}