35 lines
1.5 KiB
Markdown
35 lines
1.5 KiB
Markdown
# Margdarshak
|
|
|
|
## Steps to Run
|
|
|
|
To run the project, use the following commands:
|
|
|
|
```sh
|
|
# Run the project
|
|
API_KEY=<openai api key> cargo run
|
|
```
|
|
|
|
## The project uses the following dependencies:
|
|
|
|
- `anyhow`: For error handling.
|
|
- `async-trait`: For async functions in traits.
|
|
- `bytes`: For working with byte buffers.
|
|
- `http-body-util`: For working with HTTP bodies.
|
|
- `hyper`: For HTTP server and client implementation.
|
|
- `tokio`: For asynchronous runtime.
|
|
- `tracing`: For application-level tracing and logging.
|
|
|
|
## Project Logic
|
|
|
|
The project is structured to handle HTTP requests and process them using a wizard model. Here is a brief overview of the logic:
|
|
|
|
1. **Initialization**: The `run` function initializes the `TargetRuntime` with HTTP, file system, and environment IO components. It then scrapes data from a specified URL and processes it into a query.
|
|
|
|
2. **Wizard Interaction**: A `Wizard` instance is created using the processed query and an API key from the environment. The wizard is used to ask questions and get responses.
|
|
|
|
3. **HTTP Server**: The `AppCtx` struct, which holds the `Wizard` instance and the scraped data, is passed to the HTTP server. The server listens for incoming requests.
|
|
|
|
4. **Request Handling**: When a request is received, the `handle_request` function processes the request body, creates a `Question` instance, and uses the wizard to get a response. The response is then sent back to the client.
|
|
|
|
This setup allows the project to dynamically process and respond to HTTP requests using the wizard model.
|