/**
 * Provides tools for processing Maven dependency injection annotations at compile time.
 * <p>
 * This package contains annotation processors that generate metadata files used by
 * the Maven dependency injection system. The main component is the {@link org.apache.maven.di.tool.DiIndexProcessor},
 * which processes classes annotated with {@link org.apache.maven.api.di.Named} and creates an index file

```

</div>

////

La seule option nouvelle ici est `--workers` qui indique à Uvicorn de démarrer 4 processus workers.

Vous pouvez aussi voir qu'il affiche le **PID** de chaque processus, `27365` pour le processus parent (c'est le **gestionnaire de processus**) et un pour chaque processus worker : `27368`, `27369`, `27370` et `27367`.

## Concepts de déploiement { #deployment-concepts }

func (cs *callbacks) Create() *processor {
	return cs.processors["create"]
}

func (cs *callbacks) Query() *processor {
	return cs.processors["query"]
}

func (cs *callbacks) Update() *processor {
	return cs.processors["update"]
}

func (cs *callbacks) Delete() *processor {
	return cs.processors["delete"]
}

func (cs *callbacks) Row() *processor {
	return cs.processors["row"]
}

- Worker processes. Long-running daemon processes that the Gradle daemon starts to run specific kinds of work, such as compilation or test execution.

The diagram below shows the relationships between the different processes:

```mermaid
    graph LR
        
        gradle["Gradle CLI client"]
    

* Restarts
* **Replication (the number of processes running)**
* Memory
* Previous steps before starting

Up to this point, with all the tutorials in the docs, you have probably been running a **server program**, for example, using the `fastapi` command, that runs Uvicorn, running a **single process**.

## Decision

1. The Gradle distribution will never include a Java runtime.
   Instead, Gradle will leverage Java toolchains for the Daemon and worker processes.
2. The Gradle Daemon, Worker processes, and Tooling API client will continue to support running on different Java versions during a single Gradle invocation.
   The exact versions supported will be determined by the Gradle version.

        List<InterpolationPostProcessor> processors = new ArrayList<>(2);
        if (projectDir != null) {
            processors.add(new PathTranslatingPostProcessor(
                    PROJECT_PREFIXES, TRANSLATED_PATH_EXPRESSIONS,
                    projectDir, pathTranslator));
        }
        processors.add(new UrlNormalizingPostProcessor(urlNormalizer));
        return processors;
    }

     *
     * @return the ingest factory instance
     */
    protected IngestFactory getIngestFactory() {
        return ComponentUtil.getIngestFactory();
    }

    /**
     * Processes a result data object for web/file crawling.
     * Default implementation returns the target unchanged.
     *
     * @param target the result data to process
     * @param responseData the response data from crawling

import org.opensearch.search.aggregations.bucket.terms.Terms;

import com.google.common.io.BaseEncoding;

/**
 * Response object for faceted search results containing query counts and field facets.
 * This class processes OpenSearch aggregations to provide structured facet information
 * for search result filtering and navigation.
 */
public class FacetResponse {
    /**

Como viste en el capítulo anterior sobre [Conceptos de Despliegue](concepts.md), hay múltiples estrategias que puedes usar.

Aquí te mostraré cómo usar **Uvicorn** con **worker processes** usando el comando `fastapi` o el comando `uvicorn` directamente.

/// info | Información