Add Attachment type

Cherry pick commits from 2.x
- 8bcfb0a
- 9105620

detect language not required when explicitly sending language

Author: russcam

src/Nest/Mapping/Types/Specialized/Attachment/Attachment.cs

@@ -0,0 +1,192 @@
+using System;
+using System.Collections.Generic;
+using Newtonsoft.Json;
+
+namespace Nest
+{
+	[JsonConverter(typeof(AttachmentConverter))]
+	public class Attachment
+	{
+		/// <summary>
+		/// The author.
+		/// </summary>
+		[JsonProperty("author")]
+		public string Author { get; set; }
+
+		/// <summary>
+		/// The base64 encoded content. Can be explicitly set
+		/// </summary>
+		[JsonProperty("content")]
+		public string Content { get; set; }
+
+		/// <summary>
+		/// The length of the content before text extraction.
+		/// </summary>
+		[JsonProperty("content_length")]
+		public long? ContentLength { get; set; }
+
+		/// <summary>
+		/// The content type of the attachment. Can be explicitly set
+		/// </summary>
+		[JsonProperty("content_type")]
+		public string ContentType { get; set; }
+
+		/// <summary>
+		/// The date of the attachment.
+		/// </summary>
+		[JsonProperty("date")]
+		public DateTime? Date { get; set; }
+
+		/// <summary>
+		/// The keywords in the attachment.
+		/// </summary>
+		[JsonProperty("keywords")]
+		public string Keywords { get; set; }
+
+		/// <summary>
+		/// The language of the attachment. Can be explicitly set.
+		/// </summary>
+		[JsonProperty("language")]
+		public string Language { get; set; }
+
+		/// <summary>
+		/// Detect the language of the attachment. Language detection is
+		/// disabled by default.
+		/// </summary>
+		[JsonProperty("detect_language")]
+		public bool? DetectLanguage { get; set; }
+
+		/// <summary>
+		/// The name of the attachment. Can be explicitly set
+		/// </summary>
+		[JsonProperty("name")]
+		public string Name { get; set; }
+
+		/// <summary>
+		/// The title of the attachment.
+		/// </summary>
+		[JsonProperty("title")]
+		public string Title { get; set; }
+
+		/// <summary>
+		/// Determines how many characters are extracted when indexing the content.
+		/// By default, 100000 characters are extracted when indexing the content.
+		/// -1 can be set to extract all text, but note that all the text needs to be
+		/// allowed to be represented in memory
+		/// </summary>
+		[JsonProperty("indexed_chars")]
+		public long? IndexedCharacters { get; set; }
+
+		/// <summary>
+		/// Whether the attachment contains explicit metadata in addition to the
+		/// content. Used at indexing time to determine the serialized form of the
+		/// attachment.
+		/// </summary>
+		[JsonIgnore]
+		public bool ContainsMetadata => !Name.IsNullOrEmpty() ||
+		                                !ContentType.IsNullOrEmpty() ||
+		                                !Language.IsNullOrEmpty() ||
+										DetectLanguage.HasValue ||
+										IndexedCharacters.HasValue;
+	}
+
+	internal class AttachmentConverter : JsonConverter
+	{
+		public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
+		{
+			var attachment = (Attachment)value;
+			if (!attachment.ContainsMetadata)
+			{
+				writer.WriteValue(attachment.Content);
+			}
+			else
+			{
+				writer.WriteStartObject();
+				writer.WritePropertyName("_content");
+				writer.WriteValue(attachment.Content);
+
+				if (!string.IsNullOrEmpty(attachment.Name))
+				{
+					writer.WritePropertyName("_name");
+					writer.WriteValue(attachment.Name);
+				}
+
+				if (!string.IsNullOrEmpty(attachment.ContentType))
+				{
+					writer.WritePropertyName("_content_type");
+					writer.WriteValue(attachment.ContentType);
+				}
+
+				if (!string.IsNullOrEmpty(attachment.Language))
+				{
+					writer.WritePropertyName("_language");
+					writer.WriteValue(attachment.Language);
+				}
+
+				if (attachment.DetectLanguage.HasValue)
+				{
+					writer.WritePropertyName("_detect_language");
+					writer.WriteValue(attachment.DetectLanguage.Value);
+				}
+
+				if (attachment.IndexedCharacters.HasValue)
+				{
+					writer.WritePropertyName("_indexed_chars");
+					writer.WriteValue(attachment.IndexedCharacters.Value);
+				}
+
+				writer.WriteEndObject();
+			}
+		}
+
+		public override object ReadJson(JsonReader reader, Type objectType, object existingValue, JsonSerializer serializer)
+		{
+			if (reader.TokenType == JsonToken.String)
+			{
+				return new Attachment { Content = (string)reader.Value };
+			}
+
+			if (reader.TokenType == JsonToken.StartObject)
+			{
+				var attachment = new Attachment();
+				while (reader.Read())
+				{
+					if (reader.TokenType == JsonToken.PropertyName)
+					{
+						var propertyName = (string)reader.Value;
+						switch (propertyName)
+						{
+							case "_content":
+								attachment.Content = reader.ReadAsString();
+								break;
+							case "_name":
+								attachment.Name = reader.ReadAsString();
+								break;
+							case "_content_type":
+								attachment.ContentType = reader.ReadAsString();
+								break;
+							case "_language":
+								attachment.Language = reader.ReadAsString();
+								break;
+							case "_detect_language":
+								attachment.DetectLanguage = reader.ReadAsBoolean();
+								break;
+							case "_indexed_chars":
+								reader.Read();
+								attachment.IndexedCharacters = (long?)reader.Value;
+								break;
+						}
+					}
+					if (reader.TokenType == JsonToken.EndObject)
+					{
+						break;
+					}
+				}
+				return attachment;
+			}
+			return null;
+		}
+
+		public override bool CanConvert(Type objectType) => objectType == typeof(Attachment);
+	}
+}

src/Nest/Mapping/Visitor/PropertyWalker.cs

@@ -137,6 +137,9 @@ private IProperty InferProperty(Type type)
 			if (type == typeof(CompletionField))
 				return new CompletionProperty();
 
+			if (type == typeof(Attachment))
+				return new AttachmentProperty();
+
 			return new ObjectProperty();
 		}
 

src/Nest/Nest.csproj

@@ -850,6 +850,7 @@
     <Compile Include="Mapping\Types\CorePropertyDescriptorBase.cs" />
     <Compile Include="Mapping\Types\PropertyDescriptorBase.cs" />
     <Compile Include="Mapping\Types\PropertyJsonConverter.cs" />
+    <Compile Include="Mapping\Types\Specialized\Attachment\Attachment.cs" />
     <Compile Include="Mapping\Types\Specialized\Attachment\AttachmentAttribute.cs" />
     <Compile Include="Mapping\Types\Specialized\Attachment\AttachmentProperty.cs" />
     <Compile Include="Mapping\Types\Specialized\Completion\CompletionAttribute.cs" />

src/Tests/Document/Single/Attachment/AttachmentApiTests.cs

@@ -8,13 +8,13 @@
 
 namespace Tests.Mapping.Types.Specialized.Attachment
 {
-	public class AttachmentMappingTests : TypeMappingTestBase<Tests.Framework.MockData.Attachment>
+	public class AttachmentMappingTests : TypeMappingTestBase<Nest.Attachment>
 	{
 		protected override object ExpectJson => new
 		{
 			properties = new
 			{
-				file = new
+				content = new
 				{
 					type = "attachment",
 					fields = new
@@ -66,14 +66,14 @@ protected override void AttributeBasedSerializes()
 			// TODO: Implement
 		}
 
-		protected override Func<PropertiesDescriptor<Framework.MockData.Attachment>, IPromise<IProperties>> FluentProperties => p => p
+		protected override Func<PropertiesDescriptor<Nest.Attachment>, IPromise<IProperties>> FluentProperties => p => p
 			.Attachment(a => a
-				.Name(n => n.File)
+				.Name(n => n.Content)
 				.AuthorField(d => d
 					.Name(n => n.Author)
 				)
 				.FileField(d => d
-					.Name(n => n.File)
+					.Name(n => n.Content)
 				)
 				.ContentLengthField(d => d
 					.Name(n => n.ContentLength)

src/Tests/Document/Single/Attachment/Attachment_Test_Document.pdf

@@ -242,7 +242,6 @@
     <Compile Include="Framework\EndpointTests\Clusters\TypeOfCluster.cs" />
     <Compile Include="Framework\ManagedElasticsearch\Process\ElasticsearchNode.cs" />
     <Compile Include="Framework\ManagedElasticsearch\Process\ObservableProcess.cs" />
-    <Compile Include=".\Framework\MockData\Attachment.cs" />
     <Compile Include=".\Framework\MockData\CommitActivity.cs" />
     <Compile Include=".\Framework\MockData\Developer.cs" />
     <Compile Include=".\Framework\MockData\Gender.cs" />

src/Tests/Framework/MockData/Attachment.cs

@@ -7,4 +7,4 @@ elasticsearch_version: 5.0.0-alpha2
 force_reseed: false
 # do not spawn nodes as part of the test setup if we find a node is already running
 # this is opt in during development in CI we never want to see our tests running against an already running node
-test_against_already_running_elasticsearch: true
+test_against_already_running_elasticsearch: true