<!-- note on the offset value here: this can only be as big as
           MimeTypes#getMinLength(). If you set the offset value to larger
           than that size, the magic will only be compared to up to
           MimeTypes#getMinLength() bytes. It should also only start after
           the higher priority "start of file" one above
       -->

    /** The resource name for the MIME types configuration file. */
    protected static final String MIME_TYPES_RESOURCE_NAME = "/org/codelibs/fess/crawler/mime/tika-mimetypes.xml";

    /** The MimeTypes instance for detecting MIME types. */
    protected MimeTypes mimeTypes;

    /** Whether to use the filename for MIME type detection. */
    protected boolean useFilename = false;

    default boolean isHtmlMimetypeForCache(final String mimetype) {
        final String[] mimetypes = getCrawlerDocumentCacheHtmlMimetypes().split(",");
        if (mimetypes.length == 1 && StringUtil.isBlank(mimetypes[0])) {
            return true;
        }
        return stream(mimetypes).get(stream -> stream.anyMatch(s -> s.equalsIgnoreCase(mimetype)));
    }

crawler.document.cache.enabled=true
# Maximum size (bytes) for document cache.
crawler.document.cache.max.size=2621440
# Supported MIME types for document cache.
crawler.document.cache.supported.mimetypes=text/html

    /** The key of the configuration. e.g. text/html */
    String CRAWLER_DOCUMENT_CACHE_SUPPORTED_MIMETYPES = "crawler.document.cache.supported.mimetypes";

    /** The key of the configuration. e.g. text/html */
    String CRAWLER_DOCUMENT_CACHE_HTML_MIMETYPES = "crawler.document.cache.html.mimetypes";

    /** The key of the configuration. e.g. true */
    String INDEXER_THREAD_DUMP_ENABLED = "indexer.thread.dump.enabled";