> For the complete documentation index, see [llms.txt](https://quantumphp.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://quantumphp.gitbook.io/docs/packages/httpclient/architecture.md).

# Architecture

The package wraps either `Curl\Curl` or `Curl\MultiCurl` behind the same `HttpClient` API.

## Runtime flow

### 1. Create a client

You begin in one of three ways:

* `createRequest($url)` for one request
* `createMultiRequest()` for a collected batch
* `createAsyncMultiRequest($success, $error)` for callback-driven batches

Each method replaces the current internal client instance.

It does **not** reset the wrapper's other state. The current HTTP method, pending data payload, tracked request headers, collected responses, and collected errors all stay on the same `HttpClient` object until you overwrite them or create a fresh wrapper instance.

## 2. Configure the underlying curl client

`HttpClient` exposes a small native API (`setMethod()`, `setData()`, `start()`) and forwards unknown method calls to the underlying curl-class object through `__call()`.

That means methods such as these come from the wrapped library, not from Quantum itself:

* `setHeader()`
* `setHeaders()`
* `setOpt()`
* `addGet()`
* `addPost()`

If the wrapped client does not have the requested method, `__call()` throws `HttpClientException::methodNotSupported(...)`.

## 3. Start execution

### Single request path

For a single request, `start()` calls an internal `startSingleRequest()` method that:

1. applies `CURLOPT_CUSTOMREQUEST` from `setMethod()`
2. applies `CURLOPT_POSTFIELDS` when `setData()` holds a truthy value
3. executes the request
4. stores headers, cookies, body, and any transport error

### Multi request path

For a normal multi request, `start()` simply calls `MultiCurl::start()`. Each completed request is then passed into `handleResponse()` by the callback registered in `createMultiRequest()`.

### Async multi request path

For async multi requests, Quantum does not install its own completion collector. Your success and error callbacks are the primary output path.

## Response storage model

Quantum stores completed data in two internal arrays:

* `$response[$curlId]`
* `$errors[$curlId]`

Each response entry contains:

* `headers`
* `cookies`
* `body`

Each error entry contains:

* `code`
* `message`

On single requests, public getters collapse the storage down to the current request ID.

On multi requests, public getters return the full ID-keyed maps.

## Header tracking model

The package keeps a separate `$requestHeaders` array for headers you set through proxied:

* `setHeader($name, $value)`
* `setHeaders([...])`

Those keys are lowercased before storage.

This is only a bookkeeping layer for later reads through `getRequestHeaders()`. It does not inspect headers set through unrelated low-level cURL options.

## Guard rails

Several read methods call `ensureSingleRequest()` first.

If the current client is `MultiCurl`, these methods fail instead of guessing which request you meant:

* `getRequestHeaders()`
* `getResponseHeaders()`
* `getResponseCookies()`
* `getResponseBody()`
* `info()`
* `url()`

That contract keeps the single-request API predictable, but it means multi-request consumers must work with the aggregated response arrays instead.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://quantumphp.gitbook.io/docs/packages/httpclient/architecture.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.