Implement adjust_offsets on word delimiter graph token filter (#3934)

This commit implements adjust_offsets on word delimiter graph token filter

(cherry picked from commit e410ac9)

Author: codebrain

src/Nest/Analysis/TokenFilters/WordDelimiterGraph/WordDelimiterGraphTokenFilter.cs

@@ -10,6 +10,13 @@ namespace Nest
 	/// </summary>
 	public interface IWordDelimiterGraphTokenFilter : ITokenFilter
 	{
+		/// <summary>
+		/// By default, the filter tries to output subtokens with adjusted offsets to reflect their actual position in the token stream. However, when used in combination with other filters that alter the length or starting position of tokens without changing their offsets (e.g. <see cref="TrimTokenFilter"/>) this can cause tokens with illegal offsets to be emitted. Setting <see cref="AdjustOffsets"/> to false will stop <see cref="WordDelimiterGraphTokenFilter"/> from adjusting these internal offsets.
+		/// </summary>
+		[DataMember(Name ="adjust_offsets")]
+		[JsonFormatter(typeof(NullableStringBooleanFormatter))]
+		bool? AdjustOffsets { get; set; }
+
 		/// <summary>
 		/// If true causes all subword parts to be catenated: "wi-fi-4000" ⇒ "wifi4000". Defaults to false.
 		/// </summary>
@@ -104,6 +111,9 @@ public class WordDelimiterGraphTokenFilter : TokenFilterBase, IWordDelimiterGrap
 	{
 		public WordDelimiterGraphTokenFilter() : base("word_delimiter_graph") { }
 
+		/// <inheritdoc />
+		public bool? AdjustOffsets { get; set; }
+
 		/// <inheritdoc />
 		public bool? CatenateAll { get; set; }
 
@@ -149,6 +159,7 @@ public class WordDelimiterGraphTokenFilterDescriptor
 		: TokenFilterDescriptorBase<WordDelimiterGraphTokenFilterDescriptor, IWordDelimiterGraphTokenFilter>, IWordDelimiterGraphTokenFilter
 	{
 		protected override string Type => "word_delimiter_graph";
+		bool? IWordDelimiterGraphTokenFilter.AdjustOffsets { get; set; }
 		bool? IWordDelimiterGraphTokenFilter.CatenateAll { get; set; }
 		bool? IWordDelimiterGraphTokenFilter.CatenateNumbers { get; set; }
 		bool? IWordDelimiterGraphTokenFilter.CatenateWords { get; set; }
@@ -179,6 +190,9 @@ public WordDelimiterGraphTokenFilterDescriptor GenerateNumberParts(bool? generat
 		public WordDelimiterGraphTokenFilterDescriptor CatenateNumbers(bool? catenateNumbers = true) =>
 			Assign(catenateNumbers, (a, v) => a.CatenateNumbers = v);
 
+		/// <inheritdoc />
+		public WordDelimiterGraphTokenFilterDescriptor AdjustOffsets(bool? adjustOffsets = true) => Assign(adjustOffsets, (a, v) => a.AdjustOffsets = v);
+
 		/// <inheritdoc />
 		public WordDelimiterGraphTokenFilterDescriptor CatenateAll(bool? catenateAll = true) => Assign(catenateAll, (a, v) => a.CatenateAll = v);
 

src/Tests/Tests/Analysis/TokenFilters/TokenFilterTests.cs

@@ -833,6 +833,7 @@ public class WordDelimiterGraphTests : TokenFilterAssertionBase<WordDelimiterGra
 		{
 			public override FuncTokenFilters Fluent => (n, tf) => tf
 				.WordDelimiterGraph(n, t => t
+					.AdjustOffsets()
 					.CatenateAll()
 					.CatenateNumbers()
 					.CatenateWords()
@@ -848,6 +849,7 @@ public class WordDelimiterGraphTests : TokenFilterAssertionBase<WordDelimiterGra
 			public override ITokenFilter Initializer =>
 				new WordDelimiterGraphTokenFilter
 				{
+					AdjustOffsets = true,
 					CatenateAll = true,
 					CatenateNumbers = true,
 					CatenateWords = true,
@@ -863,6 +865,7 @@ public class WordDelimiterGraphTests : TokenFilterAssertionBase<WordDelimiterGra
 			public override object Json => new
 			{
 				type = "word_delimiter_graph",
+				adjust_offsets = true,
 				generate_word_parts = true,
 				generate_number_parts = true,
 				catenate_words = true,