refactor(FuzzySharp): rename DomainTypes to Sources and DomainTermMapping to SynonymMapping

Yanan Wang · Yanan Wang · commit 3162be4e00bc · 2025-11-13T14:17:20.000-06:00
- Rename DomainTypes → Sources
- Rename DomainTermMapping → SynonymMapping
- Update namespace: BotSharp.Abstraction.FuzzSharp → BotSharp.Plugin.FuzzySharp.FuzzSharp
- Rename DomainTermMatcher.cs → SynonymMatcher.cs
diff --git a/src/Infrastructure/BotSharp.Abstraction/Knowledges/IPhraseCollection.cs b/src/Infrastructure/BotSharp.Abstraction/Knowledges/IPhraseCollection.cs
@@ -3,5 +3,5 @@ namespace BotSharp.Abstraction.Knowledges;
 public interface IPhraseCollection
 {
     Task<Dictionary<string, HashSet<string>>> LoadVocabularyAsync();
-    Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadDomainTermMappingAsync();
+    Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadSynonymMappingAsync();
 }
diff --git a/src/Infrastructure/BotSharp.Abstraction/Knowledges/Models/SearchPhrasesResult.cs b/src/Infrastructure/BotSharp.Abstraction/Knowledges/Models/SearchPhrasesResult.cs
@@ -1,11 +1,11 @@
 
-namespace BotSharp.Abstraction.Knowledges.Models
+namespace BotSharp.Abstraction.Knowledges.Models;
+
+public class SearchPhrasesResult
 {
-    public class SearchPhrasesResult
-    {
-        public string Token { get; set; } = string.Empty;
-        public List<string> DomainTypes { get; set; } = new();
-        public string CanonicalForm { get; set; } = string.Empty;
-        public double Confidence { get; set; }
-    }
+    public string Token { get; set; } = string.Empty;
+    public List<string> Sources { get; set; } = new();
+    public string CanonicalForm { get; set; } = string.Empty;
+    public string MatchType { get; set; } = string.Empty;
+    public double Confidence { get; set; }
 }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Constants/MatchReason.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Constants/MatchReason.cs
@@ -4,9 +4,9 @@ namespace BotSharp.Plugin.FuzzySharp.Constants;
 public static class MatchReason
 {
     /// <summary>
-    /// Token matched a domain term mapping (e.g., HVAC -> Air Conditioning/Heating)
+    /// Token matched a synonym term (e.g., HVAC -> Air Conditioning/Heating)
     /// </summary>
-    public const string DomainTermMapping = "domain_term_mapping";
+    public const string SynonymMatch = "synonym_match";
 
     /// <summary>
     /// Token exactly matched a vocabulary entry
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Controllers/FuzzySharpController.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Controllers/FuzzySharpController.cs
@@ -21,13 +21,13 @@ public FuzzySharpController(
     }
 
     /// <summary>
-    /// Analyze text for typos and entities using domain-specific vocabulary.
-    /// 
+    /// Analyze text for typos and entities using vocabulary.
+    ///
     /// Returns:
     /// - `original`: Original input text
     /// - `tokens`: Tokenized text (only included if `include_tokens=true`)
     /// - `flagged`: List of flagged items (each with `match_type`):
-    ///     - `domain_term_mapping` - Business abbreviations (confidence=1.0)
+    ///     - `synonym_match` - Business abbreviations (confidence=1.0)
     ///     - `exact_match` - Exact vocabulary matches (confidence=1.0)
     ///     - `typo_correction` - Spelling corrections (confidence less than 1.0)
     /// - `processing_time_ms`: Processing time in milliseconds
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Arguments/TextAnalysisRequest.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Arguments/TextAnalysisRequest.cs
@@ -1,53 +1,13 @@
-using System.ComponentModel.DataAnnotations;
-using System.Text.Json.Serialization;
 
-namespace BotSharp.Abstraction.FuzzSharp.Arguments;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp.Arguments;
 
 public class TextAnalysisRequest
 {
-    /// <summary>
-    /// Text to analyze
-    /// </summary>
-    [Required]
-    [JsonPropertyName("text")]
     public string Text { get; set; } = string.Empty;
-
-    /// <summary>
-    /// Folder path containing CSV files (will read all .csv files from the folder or its 'output' subfolder)
-    /// </summary>
-    [JsonPropertyName("vocabulary_folder_name")]
     public string? VocabularyFolderName { get; set; }
-
-    /// <summary>
-    /// Domain term mapping CSV file
-    /// </summary>
-    [JsonPropertyName("domain_term_mapping_file")]
-    public string? DomainTermMappingFile { get; set; }
-
-    /// <summary>
-    /// Min score for suggestions (0.0-1.0)
-    /// </summary>
-    [JsonPropertyName("cutoff")]
-    [Range(0.0, 1.0)]
+    public string? SynonymMappingFile { get; set; }
     public double Cutoff { get; set; } = 0.82;
-
-    /// <summary>
-    /// Max candidates per domain (1-20)
-    /// </summary>
-    [JsonPropertyName("topk")]
-    [Range(1, 20)]
     public int TopK { get; set; } = 5;
-
-    /// <summary>
-    /// Max n-gram size (1-10)
-    /// </summary>
-    [JsonPropertyName("max_ngram")]
-    [Range(1, 10)]
     public int MaxNgram { get; set; } = 5;
-
-    /// <summary>
-    /// Include tokens field in response (default: False)
-    /// </summary>
-    [JsonPropertyName("include_tokens")]
     public bool IncludeTokens { get; set; } = false;
 }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/INgramProcessor.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/INgramProcessor.cs
@@ -1,25 +1,25 @@
-using BotSharp.Abstraction.FuzzSharp.Models;
+using BotSharp.Plugin.FuzzySharp.FuzzSharp.Models;
 
-namespace BotSharp.Abstraction.FuzzSharp;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp;
 
 public interface INgramProcessor
 {
     /// <summary>
     /// Process tokens and generate all possible n-gram match results
     /// </summary>
     /// <param name="tokens">List of tokens to process</param>
-    /// <param name="vocabulary">Vocabulary (domain type -> vocabulary set)</param>
-    /// <param name="domainTermMapping">Domain term mapping</param>
-    /// <param name="lookup">Lookup table (lowercase vocabulary -> (canonical form, domain type list))</param>
+    /// <param name="vocabulary">Vocabulary (source -> vocabulary set)</param>
+    /// <param name="synonymMapping">Synonym term Mapping</param>
+    /// <param name="lookup">Lookup table (lowercase vocabulary -> (canonical form, source list))</param>
     /// <param name="maxNgram">Maximum n-gram length</param>
     /// <param name="cutoff">Minimum confidence threshold for fuzzy matching</param>
     /// <param name="topK">Maximum number of matches to return</param>
     /// <returns>List of flagged items</returns>
     List<FlaggedItem> ProcessNgrams(
         List<string> tokens,
         Dictionary<string, HashSet<string>> vocabulary,
-        Dictionary<string, (string DbPath, string CanonicalForm)> domainTermMapping,
-        Dictionary<string, (string CanonicalForm, List<string> DomainTypes)> lookup,
+        Dictionary<string, (string DbPath, string CanonicalForm)> synonymMapping,
+        Dictionary<string, (string CanonicalForm, List<string> Sources)> lookup,
         int maxNgram,
         double cutoff,
         int topK);
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/IResultProcessor.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/IResultProcessor.cs
@@ -1,6 +1,6 @@
-using BotSharp.Abstraction.FuzzSharp.Models;
+using BotSharp.Plugin.FuzzySharp.FuzzSharp.Models;
 
-namespace BotSharp.Abstraction.FuzzSharp;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp;
 
 /// <summary>
 /// Result processor interface
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/ITokenMatcher.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/ITokenMatcher.cs
@@ -1,4 +1,4 @@
-namespace BotSharp.Abstraction.FuzzSharp;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp;
 
 public interface ITokenMatcher
 {
@@ -24,8 +24,8 @@ public record MatchContext(
     int StartIndex,
     int NgramLength,
     Dictionary<string, HashSet<string>> Vocabulary,
-    Dictionary<string, (string DbPath, string CanonicalForm)> DomainTermMapping,
-    Dictionary<string, (string CanonicalForm, List<string> DomainTypes)> Lookup,
+    Dictionary<string, (string DbPath, string CanonicalForm)> SynonymMapping,
+    Dictionary<string, (string CanonicalForm, List<string> Sources)> Lookup,
     double Cutoff,
     int TopK);
 
@@ -34,6 +34,6 @@ public record MatchContext(
 /// </summary>
 public record MatchResult(
     string CanonicalForm,
-    List<string> DomainTypes,
+    List<string> Sources,
     string MatchType,
     double Confidence);
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Models/FlaggedItem.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Models/FlaggedItem.cs
@@ -1,50 +1,13 @@
-using System.Text.Json.Serialization;
 
-namespace BotSharp.Abstraction.FuzzSharp.Models;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp.Models;
 
 public class FlaggedItem
 {
-    /// <summary>
-    /// Token index in the original text
-    /// </summary>
-    [JsonPropertyName("index")]
     public int Index { get; set; }
-
-    /// <summary>
-    /// Original token text
-    /// </summary>
-    [JsonPropertyName("token")]
     public string Token { get; set; } = string.Empty;
-
-    /// <summary>
-    /// Domain types where this token was found (e.g., ['client_Profile.Name', 'data_ServiceType.Name'])
-    /// </summary>
-    [JsonPropertyName("domain_types")]
-    public List<string> DomainTypes { get; set; } = new();
-
-    /// <summary>
-    /// Type of match: 'domain_term_mapping' (business abbreviations, confidence=1.0) | 
-    /// 'exact_match' (vocabulary matches, confidence=1.0) | 
-    /// 'typo_correction' (spelling corrections, confidence less than 1.0)
-    /// </summary>
-    [JsonPropertyName("match_type")]
+    public List<string> Sources { get; set; } = new();
     public string MatchType { get; set; } = string.Empty;
-
-    /// <summary>
-    /// Canonical form or suggested correction
-    /// </summary>
-    [JsonPropertyName("canonical_form")]
     public string CanonicalForm { get; set; } = string.Empty;
-
-    /// <summary>
-    /// Confidence score (0.0-1.0, where 1.0 is exact match)
-    /// </summary>
-    [JsonPropertyName("confidence")]
     public double Confidence { get; set; }
-
-    /// <summary>
-    /// N-gram length (number of tokens in this match). Internal field, not included in JSON output.
-    /// </summary>
-    [JsonIgnore]
     public int NgramLength { get; set; }
 }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Models/TextAnalysisResponse.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzSharp/Models/TextAnalysisResponse.cs
@@ -1,31 +1,10 @@
-using System.Text.Json.Serialization;
 
-namespace BotSharp.Abstraction.FuzzSharp.Models;
+namespace BotSharp.Plugin.FuzzySharp.FuzzSharp.Models;
 
 public class TextAnalysisResponse
 {
-    /// <summary>
-    /// Original text
-    /// </summary>
-    [JsonPropertyName("original")]
     public string Original { get; set; } = string.Empty;
-
-    /// <summary>
-    /// Tokenized text (only included if include_tokens=true)
-    /// </summary>
-    [JsonPropertyName("tokens")]
-    [JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingNull)]
     public List<string>? Tokens { get; set; }
-
-    /// <summary>
-    /// Flagged items (filter by 'match_type' field: 'domain_term_mapping', 'exact_match', or 'typo_correction')
-    /// </summary>
-    [JsonPropertyName("flagged")]
     public List<FlaggedItem> Flagged { get; set; } = new();
-
-    /// <summary>
-    /// Processing time in milliseconds
-    /// </summary>
-    [JsonPropertyName("processing_time_ms")]
     public double ProcessingTimeMs { get; set; }
 }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzySharpPlugin.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/FuzzySharpPlugin.cs
@@ -1,4 +1,4 @@
-using BotSharp.Abstraction.FuzzSharp;
+using BotSharp.Plugin.FuzzySharp.FuzzSharp;
 using BotSharp.Abstraction.Knowledges;
 using BotSharp.Abstraction.Plugins;
 using BotSharp.Plugin.FuzzySharp.Services;
@@ -23,7 +23,7 @@ public void RegisterDI(IServiceCollection services, IConfiguration config)
         services.AddScoped<IPhraseService, PhraseService>();
         services.AddScoped<IPhraseCollection, CsvPhraseCollectionLoader>();
         services.AddScoped<ITokenMatcher, ExactMatcher>();
-        services.AddScoped<ITokenMatcher, DomainTermMatcher>();
+        services.AddScoped<ITokenMatcher, SynonymMatcher>();
         services.AddScoped<ITokenMatcher, FuzzyMatcher>();
     }
 }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/CsvPhraseCollectionLoader.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/CsvPhraseCollectionLoader.cs
@@ -36,25 +36,25 @@ public async Task<Dictionary<string, HashSet<string>>> LoadVocabularyAsync()
         }
 
         // Load each CSV file
-        foreach (var (domainType, filePath) in csvFileDict)
+        foreach (var (source, filePath) in csvFileDict)
         {
             try
             {
                 var terms = await LoadCsvFileAsync(filePath);
-                vocabulary[domainType] = terms;
-                _logger.LogInformation($"Loaded {terms.Count} terms for domain type '{domainType}' from {filePath}");
+                vocabulary[source] = terms;
+                _logger.LogInformation($"Loaded {terms.Count} terms for source '{source}' from {filePath}");
             }
             catch (Exception ex)
             {
-                _logger.LogError(ex, $"Error loading CSV file for domain type '{domainType}': {filePath}");
+                _logger.LogError(ex, $"Error loading CSV file for source '{source}': {filePath}");
             }
         }
 
         return vocabulary;
     }
 
     [SharpCache(60)]
-    public async Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadDomainTermMappingAsync()
+    public async Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadSynonymMappingAsync()
     {
         string filename = "";
         var result = new Dictionary<string, (string DbPath, string CanonicalForm)>();
@@ -81,7 +81,7 @@ public async Task<Dictionary<string, HashSet<string>>> LoadVocabularyAsync()
 
             if (!HasRequiredColumns(csv))
             {
-                _logger.LogWarning("Domain term mapping file missing required columns: {FilePath}", filePath);
+                _logger.LogWarning("Synonym mapping file missing required columns: {FilePath}", filePath);
                 return result;
             }
 
@@ -106,11 +106,11 @@ public async Task<Dictionary<string, HashSet<string>>> LoadVocabularyAsync()
                 result[key] = (dbPath, canonicalForm);
             }
 
-            _logger.LogInformation("Loaded domain term mapping from {FilePath}: {Count} terms", filePath, result.Count);
+            _logger.LogInformation("Loaded synonym mapping from {FilePath}: {Count} terms", filePath, result.Count);
         }
         catch (Exception ex)
         {
-            _logger.LogError(ex, "Error loading domain term mapping file: {FilePath}", filePath);
+            _logger.LogError(ex, "Error loading synonym mapping file: {FilePath}", filePath);
         }
 
         return result;
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/ExactMatcher.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/ExactMatcher.cs
@@ -1,4 +1,4 @@
-using BotSharp.Abstraction.FuzzSharp;
+using BotSharp.Plugin.FuzzySharp.FuzzSharp;
 using BotSharp.Plugin.FuzzySharp.Constants;
 
 namespace BotSharp.Plugin.FuzzySharp.Services.Matching;
@@ -13,7 +13,7 @@ public class ExactMatcher : ITokenMatcher
         {
             return new MatchResult(
                 CanonicalForm: match.CanonicalForm,
-                DomainTypes: match.DomainTypes,
+                Sources: match.Sources,
                 MatchType: MatchReason.ExactMatch,
                 Confidence: 1.0);
         }
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/FuzzyMatcher.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/FuzzyMatcher.cs
@@ -1,4 +1,4 @@
-using BotSharp.Abstraction.FuzzSharp;
+using BotSharp.Plugin.FuzzySharp.FuzzSharp;
 using System.Text.RegularExpressions;
 using FuzzySharp;
 using FuzzySharp.SimilarityRatio;
@@ -19,20 +19,20 @@ public class FuzzyMatcher : ITokenMatcher
             return null;
         }
 
-        var (canonicalForm, domainTypes, confidence) = match.Value;
+        var (canonicalForm, sources, confidence) = match.Value;
         return new MatchResult(
             CanonicalForm: canonicalForm,
-            DomainTypes: domainTypes,
+            Sources: sources,
             MatchType: MatchReason.TypoCorrection,
             Confidence: confidence);
     }
 
     /// <summary>
     /// Check typo correction using fuzzy matching
     /// </summary>
-    private (string CanonicalForm, List<string> DomainTypes, double Confidence)? CheckTypoCorrection(
+    private (string CanonicalForm, List<string> Sources, double Confidence)? CheckTypoCorrection(
        string contentSpan,
-       Dictionary<string, (string CanonicalForm, List<string> DomainTypes)> lookup,
+       Dictionary<string, (string CanonicalForm, List<string> Sources)> lookup,
        double cutoff)
     {
         // Convert cutoff to 0-100 scale for FuzzySharp
@@ -56,9 +56,9 @@ public class FuzzyMatcher : ITokenMatcher
             return null;
         }
 
-        // Get the canonical form and domain types from lookup
+        // Get the canonical form and sources from lookup
         var match = lookup[result.Value];
-        return (match.CanonicalForm, match.DomainTypes, Math.Round(result.Score / 100.0, 3));
+        return (match.CanonicalForm, match.Sources, Math.Round(result.Score / 100.0, 3));
     }
 
     /// <summary>
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/SynonymMatcher.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Matching/SynonymMatcher.cs
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/PhraseService.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/PhraseService.cs
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Processors/NgramProcessor.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Processors/NgramProcessor.cs
diff --git a/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Processors/ResultProcessor.cs b/src/Plugins/BotSharp.Plugin.FuzzySharp/Services/Processors/ResultProcessor.cs

Original file line number	Diff line number	Diff line change
`@@ -3,5 +3,5 @@ namespace BotSharp.Abstraction.Knowledges;`
`3`	`3`	`public interface IPhraseCollection`
`4`	`4`	`{`
`5`	`5`	`Task<Dictionary<string, HashSet<string>>> LoadVocabularyAsync();`
`6`		`- Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadDomainTermMappingAsync();`
	`6`	`+ Task<Dictionary<string, (string DbPath, string CanonicalForm)>> LoadSynonymMappingAsync();`
`7`	`7`	`}`