Fess uses OpenSearch/Elasticsearch internally, so you can:

1. Export index mappings and settings from source
2. Apply custom analyzers in Fess's `fess_config.properties`
3. Create synonym dictionaries in Fess's dictionary directory

---

### Apache Solr

For migrating from Apache Solr:

#### Document Export and Import

**Step 1: Export from Solr**

```bash

labels.upgrade_data_migration=Data Migration
labels.upgrade_reindex=Re-indexing
labels.upgrade_start_button=Start
labels.replace_aliases=Replace Aliases
labels.reset_dictionaries=Reset Dictionaries
labels.reindex_start_button=Start
labels.targetVersion=Target Version
labels.target_version=Target Version
labels.sereq_configuration=Request to OpenSearch
labels.sereq_request_file=Request File

labels.upgrade_data_migration=Data Migration
labels.upgrade_reindex=Re-indexing
labels.upgrade_start_button=Start
labels.replace_aliases=Replace Aliases
labels.reset_dictionaries=Reset Dictionaries
labels.reindex_start_button=Start
labels.targetVersion=Target Version
labels.target_version=Target Version
labels.sereq_configuration=Request to OpenSearch
labels.sereq_request_file=Request File

    /** The key of the message: Replace Aliases */
    public static final String LABELS_replace_aliases = "{labels.replace_aliases}";

    /** The key of the message: Reset Dictionaries */
    public static final String LABELS_reset_dictionaries = "{labels.reset_dictionaries}";

    /** The key of the message: Start */

    *   `steps_per_epoch` and `steps` arguments are supported with numpy arrays.
    *   New error message when unexpected keys are used in
        sample_weight/class_weight dictionaries
    *   Losses are scaled in Keras compile/fit and not in the optimizers
        anymore. If you are using custom training loop, we have new utilities to
        help scale losses `tf.nn.compute_average_loss`,

9. Diese *Pfadoperation-Funktion* gibt keine Datenklassen zurück (obwohl dies möglich wäre), sondern eine Liste von Dictionarys mit internen Daten.

    FastAPI verwendet den Parameter `response_model` (der Datenklassen enthält), um die Response zu konvertieren.

9. Esta *path operation function* no está devolviendo dataclasses (aunque podría), sino una lista de diccionarios con datos internos.

    FastAPI usará el parámetro `response_model` (que incluye dataclasses) para convertir el response.

9. Esta *função de operação de rota* não está retornando dataclasses (embora pudesse), mas uma lista de dicionários com dados internos.

    O FastAPI usará o parâmetro `response_model` (que inclui dataclasses) para converter a resposta.

9. Cette *fonction de chemin d'accès* ne renvoie pas des dataclasses (même si elle le pourrait), mais une liste de dictionnaires contenant des données internes.

    FastAPI utilisera le paramètre `response_model` (qui inclut des dataclasses) pour convertir la réponse.

Wenn Sie nun die Dokumentation ansehen, werden dort alle zusätzlichen Metadaten angezeigt:

<img src="/img/tutorial/metadata/image02.png">

### Reihenfolge der Tags { #order-of-tags }

Die Reihenfolge der Tag-Metadaten-Dictionarys definiert auch die Reihenfolge, in der diese in der Benutzeroberfläche der Dokumentation angezeigt werden.