com.openai.unity 8.4.0 (#303)

- Add realtime support - Added `o1`, `o1-mini`, `gpt-4o-mini`, and `gpt-4o-realtime`, `gpt-4o-audio` model convenience properties - Fixed some bugs with function invocations - Fixed `strict` for built in `FunctionAttribute` defined tools - Fixed `FunctionAttribute` tool generated names so they aren't too long - Refactored `Tools` and `ToolCalls`. There is more of a distinction now in `ChatResponses` - Refactored `SpeechRequest`, and deprecated `SpeechVoice` enum in favor of new `Voice` class - Refactored `OpenAI.Chat` to support new audio modalities and output audio
RageAgainstThePixel · Nov 15, 2024 · d8afd72 · d8afd72
1 parent 27a2d3c
commit d8afd72
Show file tree

Hide file tree

Showing 208 changed files with 11,699 additions and 1,065 deletions.
diff --git a/Documentation~/README.md b/Documentation~/README.md
diff --git a/Runtime/Assistants/AssistantExtensions.cs b/Runtime/Assistants/AssistantExtensions.cs
diff --git a/Runtime/Assistants/AssistantFileResponse.cs b/Runtime/Assistants/AssistantFileResponse.cs
diff --git a/Runtime/Assistants/AssistantResponse.cs b/Runtime/Assistants/AssistantResponse.cs
@@ -123,15 +123,6 @@ internal AssistantResponse(
         [JsonProperty("tool_resources")]
         public ToolResources ToolResources { get; }
 
-        /// <summary>
-        /// A list of file IDs attached to this assistant.
-        /// There can be a maximum of 20 files attached to the assistant.
-        /// Files are ordered by their creation date in ascending order.
-        /// </summary>
-        [JsonIgnore]
-        [Obsolete("Files removed from Assistants. Files now belong to ToolResources.")]
-        public IReadOnlyList<string> FileIds => null;
-
         /// <summary>
         /// Set of 16 key-value pairs that can be attached to an object.
         /// This can be useful for storing additional information about the object in a structured format.
@@ -161,7 +152,7 @@ internal AssistantResponse(
 
         /// <summary>
         /// Specifies the format that the model must output.
-        /// Setting to <see cref="ChatResponseFormat.Json"/> enables JSON mode,
+        /// Setting to <see cref="ChatResponseFormat.Json"/> or <see cref="ChatResponseFormat.JsonSchema"/> enables JSON mode,
         /// which guarantees the message the model generates is valid JSON.
         /// </summary>
         /// <remarks>
@@ -172,10 +163,11 @@ internal AssistantResponse(
         /// which indicates the generation exceeded max_tokens or the conversation exceeded the max context length.
         /// </remarks>
         [Preserve]
-        [JsonProperty("response_format")]
         [JsonConverter(typeof(ResponseFormatConverter))]
+        [JsonProperty("response_format", DefaultValueHandling = DefaultValueHandling.Ignore)]
         public ResponseFormatObject ResponseFormatObject { get; }
 
+        [Preserve]
         [JsonIgnore]
         public ChatResponseFormat ResponseFormat => ResponseFormatObject ?? ChatResponseFormat.Auto;
 

diff --git a/Runtime/Assistants/AssistantsEndpoint.cs b/Runtime/Assistants/AssistantsEndpoint.cs
@@ -2,8 +2,6 @@
 
 using Newtonsoft.Json;
 using OpenAI.Extensions;
-using OpenAI.Files;
-using System;
 using System.Threading;
 using System.Threading.Tasks;
 using Utilities.WebRequestRest;
@@ -105,83 +103,5 @@ public async Task<bool> DeleteAssistantAsync(string assistantId, CancellationTok
             response.Validate(EnableDebug);
             return response.Deserialize<DeletedResponse>(client)?.Deleted ?? false;
         }
-
-        #region Files (Obsolete)
-
-        /// <summary>
-        /// Returns a list of assistant files.
-        /// </summary>
-        /// <param name="assistantId">The ID of the assistant the file belongs to.</param>
-        /// <param name="query"><see cref="ListQuery"/>.</param>
-        /// <param name="cancellationToken">Optional, <see cref="CancellationToken"/>.</param>
-        /// <returns><see cref="ListResponse{AssistantFile}"/>.</returns>
-        [Obsolete("Files removed from Assistants. Files now belong to ToolResources.")]
-        public async Task<ListResponse<AssistantFileResponse>> ListFilesAsync(string assistantId, ListQuery query = null, CancellationToken cancellationToken = default)
-        {
-            var response = await Rest.GetAsync(GetUrl($"/{assistantId}/files", query), new RestParameters(client.DefaultRequestHeaders), cancellationToken);
-            response.Validate(EnableDebug);
-            return response.Deserialize<ListResponse<AssistantFileResponse>>(client);
-        }
-
-        /// <summary>
-        /// Attach a file to an assistant.
-        /// </summary>
-        /// <param name="assistantId"> The ID of the assistant for which to attach a file. </param>
-        /// <param name="file">
-        /// A <see cref="FileResponse"/> (with purpose="assistants") that the assistant should use.
-        /// Useful for tools like retrieval and code_interpreter that can access files.
-        /// </param>
-        /// <param name="cancellationToken">Optional, <see cref="CancellationToken"/>.</param>
-        /// <returns><see cref="AssistantFileResponse"/>.</returns>
-        [Obsolete("Files removed from Assistants. Files now belong to ToolResources.")]
-        public async Task<AssistantFileResponse> AttachFileAsync(string assistantId, FileResponse file, CancellationToken cancellationToken = default)
-        {
-            if (file?.Purpose?.Equals(FilePurpose.Assistants) != true)
-            {
-                throw new InvalidOperationException($"{nameof(file)}.{nameof(file.Purpose)} must be 'assistants'!");
-            }
-
-            var payload = JsonConvert.SerializeObject(new { file_id = file.Id }, OpenAIClient.JsonSerializationOptions);
-            var response = await Rest.PostAsync(GetUrl($"/{assistantId}/files"), payload, new RestParameters(client.DefaultRequestHeaders), cancellationToken);
-            response.Validate(EnableDebug);
-            return response.Deserialize<AssistantFileResponse>(client);
-        }
-
-        /// <summary>
-        /// Retrieves an AssistantFile.
-        /// </summary>
-        /// <param name="assistantId">The ID of the assistant who the file belongs to.</param>
-        /// <param name="fileId">The ID of the file we're getting.</param>
-        /// <param name="cancellationToken">Optional, <see cref="CancellationToken"/>.</param>
-        /// <returns><see cref="AssistantFileResponse"/>.</returns>
-        [Obsolete("Files removed from Assistants. Files now belong to ToolResources.")]
-        public async Task<AssistantFileResponse> RetrieveFileAsync(string assistantId, string fileId, CancellationToken cancellationToken = default)
-        {
-            var response = await Rest.GetAsync(GetUrl($"/{assistantId}/files/{fileId}"), new RestParameters(client.DefaultRequestHeaders), cancellationToken);
-            response.Validate(EnableDebug);
-            return response.Deserialize<AssistantFileResponse>(client);
-        }
-
-        /// <summary>
-        /// Remove an assistant file.
-        /// </summary>
-        /// <remarks>
-        /// Note that removing an AssistantFile does not delete the original File object,
-        /// it simply removes the association between that File and the Assistant.
-        /// To delete a File, use the File delete endpoint instead.
-        /// </remarks>
-        /// <param name="assistantId">The ID of the assistant that the file belongs to.</param>
-        /// <param name="fileId">The ID of the file to delete.</param>
-        /// <param name="cancellationToken">Optional, <see cref="CancellationToken"/>.</param>
-        /// <returns>True, if file was removed.</returns>
-        [Obsolete("Files removed from Assistants. Files now belong to ToolResources.")]
-        public async Task<bool> RemoveFileAsync(string assistantId, string fileId, CancellationToken cancellationToken = default)
-        {
-            var response = await Rest.DeleteAsync(GetUrl($"/{assistantId}/files/{fileId}"), new RestParameters(client.DefaultRequestHeaders), cancellationToken);
-            response.Validate(EnableDebug);
-            return response.Deserialize<DeletedResponse>(client)?.Deleted ?? false;
-        }
-
-        #endregion Files (Obsolete)
     }
 }
diff --git a/Runtime/Assistants/CreateAssistantRequest.cs b/Runtime/Assistants/CreateAssistantRequest.cs
@@ -1,9 +1,7 @@
 // Licensed under the MIT License. See LICENSE in the project root for license information.
 
 using Newtonsoft.Json;
-using Newtonsoft.Json.Schema;
 using OpenAI.Extensions;
-using System;
 using System.Collections.Generic;
 using System.Linq;
 using UnityEngine.Scripting;
@@ -102,19 +100,6 @@ public CreateAssistantRequest(
         {
         }
 
-        [Obsolete("use new .ctr")]
-        public CreateAssistantRequest(
-            AssistantResponse assistant,
-            string model,
-            string name,
-            string description,
-            string instructions,
-            IEnumerable<Tool> tools,
-            IEnumerable<string> files,
-            IReadOnlyDictionary<string, string> metadata)
-        {
-        }
-
         /// <summary>
         /// Constructor.
         /// </summary>
@@ -279,7 +264,7 @@ public CreateAssistantRequest(
 
         /// <summary>
         /// Specifies the format that the model must output.
-        /// Setting to <see cref="ChatResponseFormat.Json"/> enables JSON mode,
+        /// Setting to <see cref="ChatResponseFormat.Json"/> or <see cref="ChatResponseFormat.JsonSchema"/> enables JSON mode,
         /// which guarantees the message the model generates is valid JSON.
         /// </summary>
         /// <remarks>
@@ -294,6 +279,7 @@ public CreateAssistantRequest(
         [JsonProperty("response_format", DefaultValueHandling = DefaultValueHandling.Ignore)]
         public ResponseFormatObject ResponseFormatObject { get; internal set; }
 
+        [Preserve]
         [JsonIgnore]
         public ChatResponseFormat ResponseFormat => ResponseFormatObject ?? ChatResponseFormat.Auto;
 
@@ -302,6 +288,7 @@ public CreateAssistantRequest(
         /// This can be useful for storing additional information about the object in a structured format.
         /// Keys can be a maximum of 64 characters long and values can be a maximum of 512 characters long.
         /// </summary>
+        [Preserve]
         [JsonProperty("metadata")]
         public IReadOnlyDictionary<string, string> Metadata { get; }
     }

diff --git a/Runtime/Audio/AudioEndpoint.cs b/Runtime/Audio/AudioEndpoint.cs
@@ -118,10 +118,6 @@ void StreamCallback(Response partialResponse)
             return new Tuple<string, AudioClip>(cachedPath, clip);
         }
 
-        [Obsolete("Use CreateTranscriptionTextAsync or CreateTranscriptionJsonAsync instead.")]
-        public async Task<string> CreateTranscriptionAsync(AudioTranscriptionRequest request, CancellationToken cancellationToken = default)
-            => await CreateTranscriptionTextAsync(request, cancellationToken);
-
         /// <summary>
         /// Transcribes audio into the input language.
         /// </summary>
@@ -201,10 +197,6 @@ private async Task<string> Internal_CreateTranscriptionAsync(AudioTranscriptionR
             return response.Body;
         }
 
-        [Obsolete("Use CreateTranslationTextAsync or CreateTranslationJsonAsync instead.")]
-        public async Task<string> CreateTranslationAsync(AudioTranslationRequest request, CancellationToken cancellationToken = default)
-            => await CreateTranslationTextAsync(request, cancellationToken);
-
         /// <summary>
         /// Translates audio into English.
         /// </summary>

diff --git a/Runtime/Audio/SpeechRequest.cs b/Runtime/Audio/SpeechRequest.cs
@@ -19,11 +19,11 @@ public sealed class SpeechRequest
         /// <param name="responseFormat">The format to audio in. Supported formats are mp3, opus, aac, flac, wav and pcm.</param>
         /// <param name="speed">The speed of the generated audio. Select a value from 0.25 to 4.0. 1.0 is the default.</param>
         [Preserve]
-        public SpeechRequest(string input, Model model = null, SpeechVoice voice = SpeechVoice.Alloy, SpeechResponseFormat responseFormat = SpeechResponseFormat.MP3, float? speed = null)
+        public SpeechRequest(string input, Model model = null, Voice voice = null, SpeechResponseFormat responseFormat = SpeechResponseFormat.MP3, float? speed = null)
         {
             Input = !string.IsNullOrWhiteSpace(input) ? input : throw new ArgumentException("Input cannot be null or empty.", nameof(input));
             Model = string.IsNullOrWhiteSpace(model?.Id) ? Models.Model.TTS_1 : model;
-            Voice = voice;
+            Voice = string.IsNullOrWhiteSpace(voice?.Id) ? OpenAI.Voice.Alloy : voice;
             ResponseFormat = responseFormat;
             Speed = speed;
         }
@@ -49,8 +49,8 @@ public SpeechRequest(string input, Model model = null, SpeechVoice voice = Speec
         /// </summary>
         [Preserve]
         [JsonProperty("voice", DefaultValueHandling = DefaultValueHandling.Include)]
-        [FunctionProperty("The voice to use when generating the audio.", true)]
-        public SpeechVoice Voice { get; }
+        [FunctionProperty("The voice to use when generating the audio.", true, "alloy", "echo", "fable", "onyx", "nova", "shimmer")]
+        public string Voice { get; }
 
         /// <summary>
         /// The format to audio in. Supported formats are mp3, opus, aac, flac, wav and pcm.

diff --git a/Runtime/Audio/SpeechVoice.cs b/Runtime/Audio/SpeechVoice.cs
@@ -1,7 +1,10 @@
 // Licensed under the MIT License. See LICENSE in the project root for license information.
 
+using System;
+
 namespace OpenAI.Audio
 {
+    [Obsolete("Use OpenAI.Voice instead.")]
     public enum SpeechVoice
     {
         Alloy = 0,

diff --git a/Runtime/Audio/TranscriptionSegment.cs b/Runtime/Audio/TranscriptionSegment.cs
@@ -3,6 +3,9 @@
 
 namespace OpenAI.Audio
 {
+    /// <summary>
+    /// Segment of the transcribed text and their corresponding details.
+    /// </summary>
     [Preserve]
     public sealed class TranscriptionSegment
     {

diff --git a/Runtime/Audio/TranscriptionWord.cs b/Runtime/Audio/TranscriptionWord.cs
@@ -3,6 +3,9 @@
 
 namespace OpenAI.Audio
 {
+    /// <summary>
+    /// Extracted word and their corresponding timestamps.
+    /// </summary>
     [Preserve]
     public sealed class TranscriptionWord
     {

diff --git a/Runtime/Authentication/OpenAIAuthentication.cs b/Runtime/Authentication/OpenAIAuthentication.cs
@@ -178,8 +178,8 @@ public override OpenAIAuthentication LoadFromDirectory(string directory = null,
 
                     var lines = File.ReadAllLines(filePath);
                     string apiKey = null;
-                    string organizationId = null;
                     string projectId = null;
+                    string organizationId = null;
 
                     foreach (var line in lines)
                     {