SmartHttpEndpoint
Namespace: Wisej.AI.Endpoints
Assembly: Wisej.AI
Represents an abstract base class for a smart HTTP endpoint, providing common functionality for HTTP-based communication.
- C#
- VB.NET
public class SmartHttpEndpoint : SmartEndpoint
Public Class SmartHttpEndpoint
Inherits SmartEndpoint
This class is designed to facilitate communication with HTTP endpoints by managing authentication, headers, and retry logic. It provides methods for sending requests and handling responses, including error detection and retry mechanisms.
Constructors
SmartHttpEndpoint()
Initializes a new instance of the SmartHttpEndpoint class.
Properties
ApiKey
String: Gets or sets the API key used for authentication. (Default: null)
Authentication
String: Gets or sets the authentication scheme used for requests. (Default: "Bearer")
Headers
NameValueCollection: Gets or sets the collection of additional headers to be included in requests.
MaxRetries
Int32: Gets or sets the maximum number of retry attempts for failed requests. (Default: 10)
RetryDelay
Int32: Gets or sets the delay between retry attempts in milliseconds. (Default: 10000)
URL
String: Gets or sets the base URL of the endpoint. (Default: null)
The URL should not end with a trailing slash. If a trailing slash is present, it will be removed.
Methods
AskAsync(session, messages)
Asynchronously sends a request to the endpoint and returns the response message.
| Parameter | Type | Description |
|---|---|---|
| session | SmartSession | The session managing the AI interaction. |
| messages | IList<Message> | The list of messages to be sent. |
Returns: Task<Message>. A task representing the asynchronous operation, with a Message as the result.
The implementation of this method should call session.Usage.Add(assistantMessage.Usage) to keep track of the input and output tokens utilized at the session level.
Throws:
- ArgumentNullException Thrown when session or messages is null.
AskEmbeddingsAsync(inputs)
Asynchronously requests embeddings for the specified inputs.
| Parameter | Type | Description |
|---|---|---|
| inputs | String[] | The array of input strings for which embeddings are requested. |
Returns: Task<Embedding>. A task representing the asynchronous operation, with an Embedding as the result.
This method sends the inputs to the embeddings endpoint and returns the resulting embeddings. Throws:
- ArgumentNullException Thrown when inputs is null.
- NotSupportedException Thrown when the embedding model is not defined.
BuildPayload(session, messages)
Builds the payload for the request.
| Parameter | Type | Description |
|---|---|---|
| session | SmartSession | The session managing the AI interaction. |
| messages | IList<Message> | The list of messages to be included in the payload. |
Returns: Object. The constructed payload object.
Throws:
- ArgumentNullException Thrown when session or messages is null.
CreateContent(data)
Creates the HTTP content from the given data.
| Parameter | Type | Description |
|---|---|---|
| data | Object | The data to be serialized into the content. |
Returns: HttpContent. The created HttpContent.
DetectContextLimitExceeded(response)
Detects if the context limit has been exceeded in the response.
| Parameter | Type | Description |
|---|---|---|
| response | HttpResponseMessage | The HTTP response message. |
Returns: Boolean. true if the context limit is exceeded; otherwise, false.
DetectRateLimitExceeded(response)
Detects if the rate limit has been exceeded in the response.
| Parameter | Type | Description |
|---|---|---|
| response | HttpResponseMessage | The HTTP response message. |
Returns: Boolean. true if the rate limit is exceeded; otherwise, false.
GetApiKey()
Gets the API key for the endpoint.
Returns: String. The API key as a string.
GetApiURL()
Gets the API URL for the endpoint.
Returns: String. The API URL as a string.
GetEmbeddingsURL()
Gets the URL for embeddings.
Returns: String. The embeddings URL as a string.
Throws:
- NotImplementedException Thrown when the method is not implemented.
PostAsync(url, data)
Asynchronously sends a POST request to the specified URL with the given data.
| Parameter | Type | Description |
|---|---|---|
| url | String | The URL to send the request to. |
| data | Object | The data to be sent in the request body. |
Returns: Task<Response>. A task representing the asynchronous operation, with a Response as the result.
This method handles retries and error detection, including token and rate limit exceeded exceptions. Throws:
- TokenLimitExceededException Thrown when the token limit is exceeded.
- RateLimitExceededException Thrown when the rate limit is exceeded.
- Exception Thrown when the response status code is not OK.
ReadAssistantMessage(response, message)
Reads the assistant's message from the API response.
| Parameter | Type | Description |
|---|---|---|
| response | Response | The response received from the API. |
| message | Message | The message object to populate with the response content. |
Throws:
- ArgumentNullException Thrown when the response or message is null.
ReadUsage(message, reply)
Reads the usage statistics from the API response.
| Parameter | Type | Description |
|---|---|---|
| message | Message | The message object to update with usage data. |
| reply | Object | The dynamic object containing the API response data. |
Inherited By
| Name | Description |
|---|---|
| AmazonBedrockEndpoint | Represents an endpoint for connecting to Amazon Bedrock services. |
| AnthropicEndpoint | Represents an endpoint for connecting to Anthropic services. |
| AzureAIEndpoint | Represents an endpoint for connecting to Azure AI services, specifically designed to interact with OpenAI models. |
| CerebrasEndpoint | Represents a connection to Cerebras endpoints, inheriting from OpenAIEndpoint. |
| DeepSeekEndpoint | Represents a connection to DeepSeek endpoints, inheriting from OpenAIEndpoint. |
| GoogleAIEndpoint | Represents a connection to Google AI endpoints for generating content and embeddings. |
| GroqCloudEndpoint | https://console.groq.com/docs/openai |
| GroqCloudEndpointWhisper | Represents a connection to GroqCloud speech endpoints for audio transcription. |
| HuggingFaceEndpoint | Represents a connection to HuggingFace serverless endpoints for model inference and embeddings. |
| LocalAIEndpoint | Represents a connection to LocalAI endpoints, providing access to various AI models and services. |
| LocalAIEndpointImageGen | Represents an endpoint for connecting to LocalAI Image Generation services. |
| LocalAIEndpointTTS | Represents an endpoint for connecting to LocalAI's speech services. |
| LocalAIEndpointWhisper | Represents an endpoint for connecting to LocalAI's Whisper model for speech-to-text transcriptions. |
| NvidiaAIEndpoint | Represents a connection to NVIDIA AI endpoints, providing access to various AI models and services. |
| OllamaEndpoint | Represents an endpoint that connects to Ollama services, providing functionalities for chat and embeddings. |
| OpenAIEndpoint | Represents an endpoint for connecting to OpenAI services. |
| OpenAIEndpointDallE | Represents an endpoint for connecting to OpenAI DallE services. |
| OpenAIEndpointRealtime | Represents a component that manages communication with OpenAI's real-time API endpoints. |
| OpenAIEndpointTTS | Represents an endpoint for connecting to OpenAI's speech services. |
| OpenAIEndpointWhisper | Represents an endpoint for connecting to OpenAI's Whisper model for speech-to-text transcriptions. |
| SambaNovaEndpoint | Represents a connection to SambaNova endpoints, inheriting from OpenAIEndpoint. |
| TogetherAIEndpoint | Represents a connection to TogetherAI endpoints, providing access to various AI models and services. |
| XAIEndpoint | Represents a connection to X.AI endpoints, extending the functionality of OpenAIEndpoint. |