/*
   * Ideally, this class would have exposed only constructors that require a non-null cause. See
   * https://github.com/jspecify/jspecify-reference-checker/blob/61aafa4ae52594830cfc2d61c8b113009dbdb045/src/main/java/com/google/jspecify/nullness/NullSpecTransfer.java#L789
   * and https://github.com/jspecify/jspecify/issues/490.
   *
   * (That would also have ensured that its cause was always an Error, rather than possibly another

That's why the current versions are still `0.x.x`, this reflects that each version could potentially have breaking changes. This follows the <a href="https://semver.org/" class="external-link" target="_blank">Semantic Versioning</a> conventions.

You can create production applications with **FastAPI** right now (and you have probably been doing it for some time), you just have to make sure that you use a version that works correctly with the rest of your code.

Let's say you have a file structure as described in [Bigger Applications](bigger-applications.md){.internal-link target=_blank}:

```
.
├── app
│   ├── __init__.py
│   └── main.py
```

In the file `main.py` you have your **FastAPI** app:


{* ../../docs_src/app_testing/app_a_py39/main.py *}

### Testing file { #testing-file }

	}

	for _, count := range vidMap {
		// do we have enough common versions
		// that have enough quorum to satisfy
		// the etag.
		if count >= quorum {
			return etags
		}
	}

	return make([]string, len(partsMetadata))
}

// Extracts list of times from FileInfo slice and returns, skips
// slice elements which have errors.

}

// Read - reads single record.
// Once Read is called the previous record should no longer be referenced.
func (r *PReader) Read(dst sql.Record) (sql.Record, error) {
	// If we have have any records left, return these before any error.
	for len(r.current) <= r.recordsRead {
		if r.err != nil {
			return nil, r.err
		}
		// Move to next block
		item, ok := <-r.queue
		if !ok {
			r.err = io.EOF

Notice that the default value is still `None`, so the parameter is still optional.

But now, having `Query(max_length=50)` inside of `Annotated`, we are telling FastAPI that we want it to have **additional validation** for this value, we want it to have maximum 50 characters. 😎

/// tip

If you have ktlint errors, you can automatically fix them by running `mvn com.github.gantsign.maven:ktlint-maven-plugin:1.1.1:format`.

## How the configuration works

We use Kotlin portable DSL to store TeamCity configuration, which means you can easily create a new pipeline
based on a specific branch. Currently, we have two pipelines: `master` and `release`, but you can easily create

            continue@eachQueue
          }

          // If we already have more than one task, that's enough work for now. Stop searching.
          readyTask != null -> {
            multipleReadyTasks = true
            break@eachQueue
          }

          // We have a task to execute when we complete the loop.
          else -> {
            readyTask = candidate
          }

}

@test "API compatibility test passes, ensuring no unexpected changes to the TF API" {
    bazel test $TFCI_BAZEL_COMMON_ARGS //tensorflow/tools/api/tests:api_compatibility_test
    echo "You have to re-generate the TF API goldens and have the API changes reviewed."
    echo "Look at the instructions for ':api_compatibility_test -- --update_goldens=True'"
}

teardown_file() {
    bazel shutdown

     *       thread
     *   <li>If this field's value == null because of unsafe publication, we know that it isn't the
     *       object associated with our thread, because if it was the publication wouldn't have been
     *       unsafe and we'd have seen our thread as the value. This state is also why a new
     *       ThreadConfinedTaskQueue object must be created for each inline execution, because