// If a function call is resolved to an implicit invoke call, the FirImplicitInvokeCall will have the `invoke()` function as the
        // callee and the variable as the explicit receiver. To correctly get all candidates, we need to get the original function
        // call's explicit receiver (if there is any) and callee (i.e., the variable).
        val unwrappedExplicitReceiver = explicitReceiver?.unwrapSmartcastExpression()

But by following the steps above, it will be able to do some performance optimizations.

## Technical Details

Modern versions of Python have support for **"asynchronous code"** using something called **"coroutines"**, with **`async` and `await`** syntax.

Let's see that phrase by parts in the sections below:

* **Asynchronous Code**
* **`async` and `await`**
* **Coroutines**

## Asynchronous Code

        get() = analysisSession.token

    override fun analyseImports(file: KtFile): KtImportOptimizerResult = withValidityAssertion {
        error("FE10 implementation of KtImportOptimizer should not be called from anywhere")
    }

## pytest.mark.anyio

If we want to call asynchronous functions in our tests, our test functions have to be asynchronous. AnyIO provides a neat plugin for this, that allows us to specify that some test functions are to be called asynchronously.

## HTTPX

Even if your **FastAPI** application uses normal `def` functions instead of `async def`, it is still an `async` application underneath.

1. Creating a PR and observing the presubmit test results
2. Running the CI scripts locally, as explained below
3. **Google employees only**: Google employees can use an internal-only tool
called "MLCI" that makes testing more convenient: it can execute any full CI job
against a pending change. Search for "MLCI" internally to find it.

You may invoke a CI script of your choice by following these instructions:

```bash

  // e.g. `meta.k8s.io/v1.DeleteOptions` or `meta.k8s.io/v1.CreateOptions`. This may be
  // different than the options the caller provided. e.g. for a patch request the performed
  // Operation might be a CREATE, in which case the Options will a
  // `meta.k8s.io/v1.CreateOptions` even though the caller provided `meta.k8s.io/v1.PatchOptions`.
  // +optional
  optional k8s.io.apimachinery.pkg.runtime.RawExtension options = 12;
}

import okhttp3.internal.connection.RealCall

/**
 * A concrete interceptor chain that carries the entire interceptor chain: all application
 * interceptors, the OkHttp core, all network interceptors, and finally the network caller.
 *
 * If the chain is for an application interceptor then [exchange] must be null. Otherwise it is for
 * a network interceptor and [exchange] must be non-null.
 */
class RealInterceptorChain(

   *
   * ```java
   * // ensure the response (and underlying response body) is closed
   * try (Response response = client.newCall(request).execute()) {
   *   ...
   * }
   * ```
   *
   * The caller may read the response body with the response's [Response.body] method. To avoid
   * leaking resources callers must [close the response body][ResponseBody] or the response.
   *

Calls are executed in one of two ways:

 * **Synchronous:** your thread blocks until the response is readable.
 * **Asynchronous:** you enqueue the request on any thread, and get [called back](https://square.github.io/okhttp/4.x/okhttp/okhttp3/-callback/) on another thread when the response is readable.

Most of the code is actually the same.

So, we are going to focus only on the differences.

## File structure

Let's say you have a directory named `my_super_project` that contains a sub-directory called `sql_app` with a structure like this:

```
.
└── sql_app
    ├── __init__.py
    ├── crud.py
    ├── database.py
    ├── main.py
    └── schemas.py
```