import org.apache.logging.log4j.LogManager;
import org.apache.logging.log4j.Logger;
import org.codelibs.core.lang.StringUtil;
import org.codelibs.fess.suggest.analysis.SuggestAnalyzer;
import org.codelibs.fess.suggest.converter.ReadingConverter;
import org.codelibs.fess.suggest.entity.SuggestItem;
import org.codelibs.fess.suggest.exception.SuggesterException;
import org.codelibs.fess.suggest.index.contents.querylog.QueryLog;

                return directive;
            }
        }
        return null;
    }

    /**
     * Adds a directive to the robots.txt rules.
     * The user-agent pattern in the directive is converted to a regular expression pattern,
     * where '*' is replaced with '.*' for pattern matching, and stored case-insensitively.
     *
     * @param directive The directive to add to the robots.txt rules
     */

Declaras la "forma" de los datos como clases con atributos.

Y cada atributo tiene un tipo.

Entonces creas un instance de esa clase con algunos valores y validará los valores, los convertirá al tipo adecuado (si es el caso) y te dará un objeto con todos los datos.

Y obtienes todo el soporte del editor con ese objeto resultante.

Un ejemplo de la documentación oficial de Pydantic:

    * Path parameters.
    * Query parameters.
    * Cookies.
    * Headers.
    * Forms.
    * Files.
* <abbr title="also known as: serialization, parsing, marshalling">Conversion</abbr> of output data: converting from Python data and types to network data (as JSON):
    * Convert Python types (`str`, `int`, `float`, `bool`, `list`, etc).
    * `datetime` objects.
    * `UUID` objects.
    * Database models.

 * 2^64 + x}). The methods for which signedness is not an issue are in {@link Longs}, as well as
 * signed versions of methods for which signedness is an issue.
 *
 * <p>In addition, this class provides several static methods for converting a {@code long} to a
 * {@code String} and a {@code String} to a {@code long} that treat the {@code long} as an unsigned
 * number.
 *

                    that `password`.
                * Validation: whatever you return will be serialized with the
                    `response_model`, converting any data as necessary to generate the
                    corresponding JSON. But if the data in the object returned is not
                    valid, that would mean a violation of the contract with the client,

          }
        }
      }
    },
    /*
     * This one comes last to work around b/367716565. (One *possible* alternative would be to not
     * declare any methods in this enum, converting assertFailsToDecode into a static method that is
     * implemented with a `switch`. I haven't tested that.)
     */
    @GwtIncompatible // decodingStream(Reader)
    DECODING_STREAM {
      @Override

Luego convertirá y validará los datos. Así que, cuando uses ese objeto `settings`, tendrás datos de los tipos que declaraste (por ejemplo, `items_per_user` será un `int`).

### Usar el `settings` { #use-the-settings }

You can return a `dict`, `list`, singular values as `str`, `int`, etc.

You can also return Pydantic models (you'll see more about that later).

There are many other objects and models that will be automatically converted to JSON (including ORMs, etc). Try using your favorite ones, it's highly probable that they are already supported.

### Step 6: Deploy it { #step-6-deploy-it }

El LLM probablemente traducirá esto mal. Lo interesante es si mantiene la traducción corregida al volver a traducir.

///

////

//// tab | Información

La persona que diseña el prompt puede elegir si quiere convertir comillas neutras a comillas tipográficas. También está bien dejarlas como están.

Consulta por ejemplo la sección `### Quotes` en `docs/de/llm-prompt.md`.

////