### [Fast Fallback](https://square.github.io/okhttp/4.x/okhttp/okhttp3/-ok-http-client/-builder/fast-fallback/)

Since version 5.0, `OkHttpClient` supports fast fallback, which is our implementation of Happy Eyeballs [RFC 6555](https://datatracker.ietf.org/doc/html/rfc6555).

With fast fallback, OkHttp attempts to connect to multiple web servers concurrently. It keeps whichever route connects first and cancels all of the others. Its rules are:

    val mediaType = parse("text/plain; charset=utf-8; charset=utf-16")
    assertEquals("UTF-8", mediaType.charsetName())
  }

  /**
   * This is invalid according to RFC 822. But it's what Chrome does and it avoids a potentially
   * unpleasant IllegalCharsetNameException.
   */
  @Test fun testCharsetNameIsSingleQuoted() {
    val mediaType = parse("text/plain;charset='utf-8'")

        assertArrayEquals(result1, result2);
    }

    @Test
    void testHMACT64WithKnownTestVector() throws NoSuchAlgorithmException {
        // Test with known test vector
        // Using test vector from RFC 2104 adapted for HMACT64
        byte[] key = new byte[16];
        Arrays.fill(key, (byte) 0x0b);
        byte[] data = "Hi There".getBytes();

import okhttp3.internal.ws.WebSocketProtocol.PAYLOAD_SHORT
import okhttp3.internal.ws.WebSocketProtocol.toggleMask
import okio.Buffer
import okio.BufferedSource
import okio.ByteString

/**
 * An [RFC 6455][rfc_6455]-compatible WebSocket frame reader.
 *
 * This class is not thread safe.
 *
 * [rfc_6455]: http://tools.ietf.org/html/rfc6455
 */
class WebSocketReader(
  private val isClient: Boolean,

   * a graceful shutdown was already underway or if the web socket is already closed or canceled.
   *
   * @param code Status code as defined by
   *     [Section 7.4 of RFC 6455](http://tools.ietf.org/html/rfc6455#section-7.4).
   * @param reason Reason for shutting down, no longer than 123 bytes of UTF-8 encoded data (**not** characters) or null.

**Registration**:
```java
extractorFactory.addExtractor("text/html", htmlExtractor, 2);  // Weight 2
extractorFactory.addExtractor("text/html", tikaExtractor, 1);  // Fallback
```

### Helpers

**RobotsTxtHelper**: RFC 9309 parsing, user-agent matching, crawl-delay, sitemaps
**SitemapsHelper**: Sitemap XML parsing, index handling
**MimeTypeHelper**: MIME detection via Tika
**EncodingHelper**: Charset detection with BOM

    // alternative subject starting with a japanese kanji character. This is
    // not supported by Android because the underlying implementation from
    // harmony follows the definition from rfc 1034 page 10 for alternative
    // subject names. This causes the code to drop all alternative subjects.
    assertThat(verifier.verify("bar.com", session)).isTrue()
    assertThat(verifier.verify("a.bar.com", session)).isFalse()

    message = "moved to val",
    replaceWith = ReplaceWith(expression = "priorResponse"),
    level = DeprecationLevel.ERROR,
  )
  fun priorResponse(): Response? = priorResponse

  /**
   * Returns the RFC 7235 authorization challenges appropriate for this response's code. If the
   * response code is 401 unauthorized, this returns the "WWW-Authenticate" challenges. If the

/**
 * A Cache-Control header with cache directives from a server or client. These directives set policy
 * on what responses can be stored, and which requests can be satisfied by those stored responses.
 *
 * See [RFC 7234, 5.2](https://tools.ietf.org/html/rfc7234#section-5.2).
 */
class CacheControl internal constructor(
  /**
   * In a response, this field's name "no-cache" is misleading. It doesn't prevent us from caching

 *  Fix: Follow HTTP 307 and 308 redirects on methods other than GET and POST. We're reluctant to
    change OkHttp's behavior in handling common HTTP status codes, but this fix is overdue! The new
    behavior is now consistent with [RFC 7231][rfc_7231_647], which is newer than OkHttp itself.
    If you want this update with the old behavior use [this interceptor][legacy_interceptor].