In the next chapter about [FastAPI in Containers - Docker](docker.md){.internal-link target=_blank} I'll tell some strategies you could use to handle the other **deployment concepts**.

I'll also show you the **official Docker image** that includes **Gunicorn with Uvicorn workers** and some default configurations that can be useful for simple cases.

A wildcard rule was added with a wildcard that is not in leftmost position! We'll need to change the ${PublicSuffixDatabase::class.java.name} to handle this."""
    }
    check(rule.indexOf(WILDCARD_CHAR, 1) == -1) {
      """Wildcard Assertion Failure: '$rule'
A wildcard rule was added with multiple wildcards! We'll need to change ${PublicSuffixDatabase::class.java.name} to handle this."""
    }

  if _os.path.exists(_kernel_dir):
    _ll.load_library(_kernel_dir)

  # Load third party dynamic kernels.
  for _s in _site_packages_dirs:
    _plugin_dir = _os.path.join(_s, "tensorflow-plugins")
    if _os.path.exists(_plugin_dir):
      _ll.load_library(_plugin_dir)
      # Load Pluggable Device Library
      _ll.load_pluggable_device_library(_plugin_dir)

* Memory
* Previous steps before starting

We'll see how they would affect **deployments**.

In the end, the ultimate objective is to be able to **serve your API clients** in a way that is **secure**, to **avoid disruptions**, and to use the **compute resources** (for example remote servers/virtual machines) as efficiently as possible. 🚀

	}

	return nil
}

type logLevelState struct {
	client         *ControlzClient
	outputLogLevel string
}

func (ll *logLevelState) run(_ io.Writer) error {
	scopeInfos, err := newScopeInfosFromScopeLevelPairs(ll.outputLogLevel)
	if err != nil {
		return err
	}
	err = ll.client.PutScopes(scopeInfos)
	if err != nil {
		return err
	}
	return nil
}

type stackTraceLevelState struct {

# CI Pipeline Configuration

## Open & import the project

In your IDEA, `File` - `Open`, select `.teamcity/pom.xml`, `import as project`, and you'll have a Maven project.

## Project structure

Mostly a standard Maven project structure. The entry point `settings.kts` defines the TeamCity project.

There are 3 subprojects in the TeamCity project hierarchy: `Check` for Gradle builds, `Promotion` for releasing Gradle versions, `Util` for miscellaneous utilities.

}

// Convert the generated YAML to pretty JSON.
func yamlToPrettyJSON(yml string) (string, error) {
	// YAML objects are not completely compatible with JSON
	// objects. Let yaml.YAMLToJSON handle the edge cases and
	// we'll re-encode the result to pretty JSON.
	uglyJSON, err := yaml.YAMLToJSON([]byte(yml))
	if err != nil {
		return "", err
	}
	var decoded any
	if uglyJSON[0] == '[' {
		decoded = make([]any, 0)
	} else {

* Also don't worry about style rules, there are already automatized tools checking that.

And if there's any other style or consistency need, I'll ask directly for that, or I'll add commits on top with the needed changes.

### Check the code

* Check and read the code, see if it makes sense, **run it locally** and see if it actually solves the problem.

## Setting up your development environment

In order to make changes to Gradle, you'll need:

* A [Java Development Kit](http://jdk.java.net/) (JDK) **version 11**. Fixed version is required to use [remote cache](#remote-build-cache). 
* A text editor or IDE. We use and recommend [IntelliJ IDEA CE](http://www.jetbrains.com/idea/).  IntelliJ Ultimate will also work. You'll need IntelliJ 2021.2.2 or newer.

That's what we'll solve, let's load the model before the requests are handled, but only right before the application starts receiving requests, not while  the code is being loaded.

## Lifespan

You can define this *startup* and *shutdown* logic using the `lifespan` parameter of the `FastAPI` app, and a "context manager" (I'll show you what that is in a second).