Merge branch 'issue-1'

Frends.OpenAI.CallChatGPT/CHANGELOG.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## [1.0.1] - 2024-01-05
+### Changed
+- Moved some input parameters to options
+- Moved API key from options to input
+
 ## [1.0.0] - 2023-12-28
 ### Changed
 - Initial implementation

Frends.OpenAI.CallChatGPT/Frends.OpenAI.CallChatGPT.Tests/BasicPropertyTests.cs

@@ -43,18 +43,29 @@ public void InputProperties_SetAndGetCorrectly()
         // Act
         input.Model = "gpt-3.5-turbo";
         input.Messages = messages;
-        input.MaxTokens = 100;
-        input.N = 3;
-        input.Seed = 12345;
-        input.User = "test-user";
 
         // Assert
         Assert.AreEqual("gpt-3.5-turbo", input.Model);
         CollectionAssert.AreEqual(messages, input.Messages);
-        Assert.AreEqual(100, input.MaxTokens);
-        Assert.AreEqual(3, input.N);
-        Assert.AreEqual(12345, input.Seed);
-        Assert.AreEqual("test-user", input.User);
+    }
+
+    [Test]
+    public void OptionsProperties_SetAndGetCorrectly()
+    {
+        // Arrange
+        var options = new Options();
+
+        // Act
+        options.MaxTokens = 100;
+        options.N = 3;
+        options.Seed = 12345;
+        options.User = "test-user";
+
+        // Assert
+        Assert.AreEqual(100, options.MaxTokens);
+        Assert.AreEqual(3, options.N);
+        Assert.AreEqual(12345, options.Seed);
+        Assert.AreEqual("test-user", options.User);
     }
 
     [Test]

Frends.OpenAI.CallChatGPT/Frends.OpenAI.CallChatGPT/Definitions/Input.cs

@@ -17,6 +17,14 @@ public class Input
     [DefaultValue("gpt-3.5-turbo")]
     public string Model { get; set; }
 
+    /// <summary>
+    /// Chat GPT API key.
+    /// </summary>
+    /// <example>1234567qwerty</example>
+    [DisplayFormat(DataFormatString = "Text")]
+    [PasswordPropertyText]
+    public string ApiKey { get; set; }
+
     /// <summary>
     /// Array of messages to send to Chat GPT.
     /// </summary>
@@ -27,49 +35,6 @@ public class Input
     /// ]
     /// </example>
     public InputMessage[] Messages { get; set; }
-
-    /// <summary>
-    /// The maximum number of tokens that can be generated in the chat
-    /// completion. Set to 0 for unlimited tokens.
-    /// The total length of input tokens and generated tokens is still limited
-    /// by the model's context length.
-    /// </summary>
-    /// <example>100</example>
-    [DefaultValue(null)]
-    public int? MaxTokens { get; set; }
-
-    /// <summary>
-    /// How many chat completion choices to generate for each input message.
-    /// Note that you will be charged based on the number of generated tokens
-    /// across all of the choices. Keep n as 1 to minimize costs.
-    /// </summary>
-    /// <internalnote>
-    /// This property name comes directly from ChatGPT API, thus is called
-    /// just N and not somethings else. We want to keep the task as close to
-    /// the API as possible.
-    /// </internalnote>
-    /// <example>3</example>
-    [DefaultValue(1)]
-    public int N { get; set; } = 1;
-
-    /// <summary>
-    /// If specified, ChatGPT system will make a best effort to sample
-    /// deterministically, such that repeated requests with the same seed and
-    /// parameters should return the same result. Determinism is not guaranteed,
-    /// and you should refer to the system_fingerprint response parameter to
-    /// monitor changes in the backend.
-    /// Set to null for non-deterministic results.
-    /// </summary>
-    /// <example>12345</example>
-    [DefaultValue(null)]
-    public int? Seed { get; set; }
-
-    /// <summary>
-    /// A unique identifier representing your end-user, which can help OpenAI to
-    /// monitor and detect abuse.
-    /// </summary>
-    /// <example>User12345</example>
-    public string User { get; set; }
 }
 
 /// <summary>

Frends.OpenAI.CallChatGPT/Frends.OpenAI.CallChatGPT/Definitions/Options.cs

@@ -9,12 +9,47 @@
 public class Options
 {
     /// <summary>
-    /// Chat GPT API key.
+    /// The maximum number of tokens that can be generated in the chat
+    /// completion. Set to 0 for unlimited tokens.
+    /// The total length of input tokens and generated tokens is still limited
+    /// by the model's context length.
     /// </summary>
-    /// <example>1234567qwerty</example>
-    [DisplayFormat(DataFormatString = "Text")]
-    [PasswordPropertyText]
-    public string ApiKey { get; set; }
+    /// <example>100</example>
+    [DefaultValue(null)]
+    public int? MaxTokens { get; set; }
+
+    /// <summary>
+    /// How many chat completion choices to generate for each input message.
+    /// Note that you will be charged based on the number of generated tokens
+    /// across all of the choices. Keep n as 1 to minimize costs.
+    /// </summary>
+    /// <internalnote>
+    /// This property name comes directly from ChatGPT API, thus is called
+    /// just N and not somethings else. We want to keep the task as close to
+    /// the API as possible.
+    /// </internalnote>
+    /// <example>3</example>
+    [DefaultValue(1)]
+    public int N { get; set; } = 1;
+
+    /// <summary>
+    /// If specified, ChatGPT system will make a best effort to sample
+    /// deterministically, such that repeated requests with the same seed and
+    /// parameters should return the same result. Determinism is not guaranteed,
+    /// and you should refer to the system_fingerprint response parameter to
+    /// monitor changes in the backend.
+    /// Set to null for non-deterministic results.
+    /// </summary>
+    /// <example>12345</example>
+    [DefaultValue(null)]
+    public int? Seed { get; set; }
+
+    /// <summary>
+    /// A unique identifier representing your end-user, which can help OpenAI to
+    /// monitor and detect abuse.
+    /// </summary>
+    /// <example>User12345</example>
+    public string User { get; set; }
 
     /// <summary>
     /// Whether to throw an exception if the API returns an error. Otherwise

Frends.OpenAI.CallChatGPT/Frends.OpenAI.CallChatGPT/Frends.OpenAI.CallChatGPT.cs

@@ -45,9 +45,17 @@ public static async Task<Result> CallChatGPT(
         [PropertyTab] Options options,
         CancellationToken cancellationToken)
     {
-        using var client = CreateClient(options);
+        using var client = CreateClient(input);
         var request = new RestRequest("v1/chat/completions");
-        request.AddJsonBody(input);
+        request.AddJsonBody(new
+        {
+            input.Messages,
+            input.Model,
+            options.MaxTokens,
+            options.N,
+            options.Seed,
+            options.User,
+        });
 
         var response = await client.ExecutePostAsync<ChatCompletion>(request, cancellationToken);
         return response.IsSuccessful
@@ -69,9 +77,9 @@ private static Result HandleErrorResponse(Options options, RestResponse<ChatComp
         return new Result(false, errorText, null);
     }
 
-    private static RestClient CreateClient(Options options)
+    private static RestClient CreateClient(Input input)
     {
-        var auth = new JwtAuthenticator(options.ApiKey);
+        var auth = new JwtAuthenticator(input.ApiKey);
         var restClientOptions = new RestClientOptions
         {
             // S1075 is complaining about hardcoded URL, but this is the host