<groupId>com.ibm.icu</groupId>
			<artifactId>icu4j</artifactId>
			<version>${icu4j.version}</version>
		</dependency>
		<dependency>
			<groupId>org.apache.tika</groupId>
			<artifactId>tika-langdetect-optimaize</artifactId>
			<version>${tika.version}</version>
			<exclusions>
				<exclusion>
					<groupId>javax.annotation</groupId>
					<artifactId>javax.annotation-api</artifactId>
				</exclusion>

				"application/x-texinfo",
				"application/x-tika-msoffice",
				"application/x-tika-msoffice-embedded",
				"application/x-tika-msoffice-embedded;format=ole10_native",
				"application/x-tika-msoffice-embedded;format=comp_obj",
				"application/x-tika-msworks-spreadsheet",
				"application/x-tika-ooxml",
				"application/x-tika-ooxml-protected",
				"application/x-tika-staroffice",
				"application/x-uc2-compressed",

	for _, tt := range RunesTests {
		tin := []byte(tt.in)
		a := Runes(tin)
		if !slices.Equal(a, tt.out) {
			t.Errorf("Runes(%q) = %v; want %v", tin, a, tt.out)
			continue
		}
		if !tt.lossy {
			// can only test reassembly if we didn't lose information
			s := string(a)
			if s != tt.in {
				t.Errorf("string(Runes(%q)) = %x; want %x", tin, s, tin)
			}
		}
	}
}

		{"ends-with-a-dot.", true},
		{"ends-with-a-dash-", true},
		{"-starts-with-a-dash", true},
		{"THIS-BEINGS-WITH-UPPERCASe", true},
		{"tHIS-ENDS-WITH-UPPERCASE", true},
		{"ThisBeginsAndEndsWithUpperCase", true},
		{"una ñina", true},
		{"lalalallalallalalalallalallalala-theString-size-is-greater-than-64", true},
		// cases for which test should fail.
		// passing invalid bucket names.
		{"", false},
		{SlashSeparator, false},
		{"a", false},

# Type of hot thread monitoring (e.g., cpu).
crawler.hotthread.type=cpu
# Metadata fields to exclude from document content.
crawler.metadata.content.excludes=resourceName,X-Parsed-By,Content-Encoding.*,Content-Type.*,X-TIKA.*,X-FESS.*
# Mapping for document metadata names.
crawler.metadata.name.mapping=\
title=title:string\n\
Title=title:string\n\
dc:title=title:string\n\

# html