feat: first take of adding caching support

2024-08-15 23:04:45 -05:00
parent c0dec5f7de
commit ed97ea95dc
10 changed files with 293 additions and 46 deletions
@@ -19,6 +19,11 @@ public abstract class BaseMessageRequest
  /// </summary>
  public string? System { get; init; } = null;

+  /// <summary>
+  /// Gets the messages to send to the model.
+  /// </summary>
+  public List<TextContent>? SystemMessages { get; init; } = null;
+
  /// <summary>
  /// Gets the messages to send to the model.
  /// </summary>
@@ -74,41 +79,21 @@ public abstract class BaseMessageRequest

  [JsonConstructor]
  internal BaseMessageRequest() { }
-
-  /// <summary>
-  /// Initializes a new instance of the <see cref="BaseMessageRequest"/> class.
-  /// </summary>
-  /// <param name="model">The model ID to use for the request.</param>
-  /// <param name="messages">The messages to send to the model.</param>
-  /// <param name="maxTokens">The maximum number of tokens to generate.</param>
-  /// <param name="system">The system ID to use for the request.</param>
-  /// <param name="metadata">The metadata to include with the request.</param>
-  /// <param name="temperature">The temperature to use for the request.</param>
-  /// <param name="topK">The top-K value to use for the request.</param>
-  /// <param name="topP">The top-P value to use for the request.</param>
-  /// <param name="toolChoice">The tool choice mode to use for the request.</param>
-  /// <param name="tools">The tools to use for the request.</param>
-  /// <param name="stream">A value indicating whether the message should be streamed.</param>
-  /// <param name="stopSequences">The prompt stop sequences.</param>
-  /// <exception cref="ArgumentException">Thrown when the model ID is invalid.</exception>
-  /// <exception cref="ArgumentNullException">Thrown when the model or messages is null.</exception>
-  /// <exception cref="ArgumentException">Thrown when the messages contain no messages.</exception>
-  /// <exception cref="ArgumentException">Thrown when the max tokens is less than one.</exception>
-  /// <exception cref="ArgumentException">Thrown when the temperature is less than zero or greater than one.</exception>
-  /// <returns>A new instance of the <see cref="BaseMessageRequest"/> class.</returns>
-  protected BaseMessageRequest(
+  
+  private BaseMessageRequest(
    string model,
    List<Message> messages,
-    int maxTokens = 1024,
-    string? system = null,
-    Dictionary<string, object>? metadata = null,
-    decimal temperature = 0.0m,
-    int? topK = null,
-    decimal? topP = null,
-    ToolChoice? toolChoice = null,
-    List<Tool>? tools = null,
-    bool stream = false,
-    List<string>? stopSequences = null
+    int maxTokens,
+    string? system,
+    List<TextContent>? systemMessages,
+    Dictionary<string, object>? metadata,
+    decimal temperature,
+    int? topK,
+    decimal? topP,
+    ToolChoice? toolChoice,
+    List<Tool>? tools,
+    bool stream,
+    List<string>? stopSequences
  )
  {
    ArgumentValidator.ThrowIfNull(model, nameof(model));
@@ -138,6 +123,7 @@ public abstract class BaseMessageRequest
    Messages = messages;
    MaxTokens = maxTokens;
    System = system;
+    SystemMessages = systemMessages;
    Metadata = metadata;
    Temperature = temperature;
    TopK = topK;
@@ -147,4 +133,108 @@ public abstract class BaseMessageRequest
    Stream = stream;
    StopSequences = stopSequences ?? [];
  }
+
+  /// <summary>
+  /// Initializes a new instance of the <see cref="BaseMessageRequest"/> class.
+  /// </summary>
+  /// <param name="model">The model ID to use for the request.</param>
+  /// <param name="messages">The messages to send to the model.</param>
+  /// <param name="maxTokens">The maximum number of tokens to generate.</param>
+  /// <param name="system">The system prompt to use for the request.</param>
+  /// <param name="metadata">The metadata to include with the request.</param>
+  /// <param name="temperature">The temperature to use for the request.</param>
+  /// <param name="topK">The top-K value to use for the request.</param>
+  /// <param name="topP">The top-P value to use for the request.</param>
+  /// <param name="toolChoice">The tool choice mode to use for the request.</param>
+  /// <param name="tools">The tools to use for the request.</param>
+  /// <param name="stream">A value indicating whether the message should be streamed.</param>
+  /// <param name="stopSequences">The prompt stop sequences.</param>
+  /// <exception cref="ArgumentException">Thrown when the model ID is invalid.</exception>
+  /// <exception cref="ArgumentNullException">Thrown when the model or messages is null.</exception>
+  /// <exception cref="ArgumentException">Thrown when the messages contain no messages.</exception>
+  /// <exception cref="ArgumentException">Thrown when the max tokens is less than one.</exception>
+  /// <exception cref="ArgumentException">Thrown when the temperature is less than zero or greater than one.</exception>
+  /// <returns>A new instance of the <see cref="BaseMessageRequest"/> class.</returns>
+  protected BaseMessageRequest(
+    string model,
+    List<Message> messages,
+    int maxTokens = 1024,
+    string? system = null,
+    Dictionary<string, object>? metadata = null,
+    decimal temperature = 0.0m,
+    int? topK = null,
+    decimal? topP = null,
+    ToolChoice? toolChoice = null,
+    List<Tool>? tools = null,
+    bool stream = false,
+    List<string>? stopSequences = null
+  ) : this(
+    model,
+    messages,
+    maxTokens,
+    system,
+    null,
+    metadata,
+    temperature,
+    topK,
+    topP,
+    toolChoice,
+    tools,
+    stream,
+    stopSequences
+  )
+  {
+  }
+
+  /// <summary>
+  /// Initializes a new instance of the <see cref="BaseMessageRequest"/> class.
+  /// </summary>
+  /// <param name="model">The model ID to use for the request.</param>
+  /// <param name="messages">The messages to send to the model.</param>
+  /// <param name="maxTokens">The maximum number of tokens to generate.</param>
+  /// <param name="systemMessages">The system messages to send to the model to be used as the system prompt.</param>
+  /// <param name="metadata">The metadata to include with the request.</param>
+  /// <param name="temperature">The temperature to use for the request.</param>
+  /// <param name="topK">The top-K value to use for the request.</param>
+  /// <param name="topP">The top-P value to use for the request.</param>
+  /// <param name="toolChoice">The tool choice mode to use for the request.</param>
+  /// <param name="tools">The tools to use for the request.</param>
+  /// <param name="stream">A value indicating whether the message should be streamed.</param>
+  /// <param name="stopSequences">The prompt stop sequences.</param>
+  /// <exception cref="ArgumentException">Thrown when the model ID is invalid.</exception>
+  /// <exception cref="ArgumentNullException">Thrown when the model or messages is null.</exception>
+  /// <exception cref="ArgumentException">Thrown when the messages contain no messages.</exception>
+  /// <exception cref="ArgumentException">Thrown when the max tokens is less than one.</exception>
+  /// <exception cref="ArgumentException">Thrown when the temperature is less than zero or greater than one.</exception>
+  /// <returns>A new instance of the <see cref="BaseMessageRequest"/> class.</returns>
+  protected BaseMessageRequest(
+    string model,
+    List<Message> messages,
+    int maxTokens = 1024,
+    List<TextContent>? systemMessages = null,
+    Dictionary<string, object>? metadata = null,
+    decimal temperature = 0.0m,
+    int? topK = null,
+    decimal? topP = null,
+    ToolChoice? toolChoice = null,
+    List<Tool>? tools = null,
+    bool stream = false,
+    List<string>? stopSequences = null
+  ) : this(
+    model,
+    messages,
+    maxTokens,
+    null,
+    systemMessages,
+    metadata,
+    temperature,
+    topK,
+    topP,
+    toolChoice,
+    tools,
+    stream,
+    stopSequences
+  )
+  {
+  }
 }