* value slightly too high than slightly too low. If the value is not exact, the {@link
   * ImmutableDoubleArray} that is built will very likely occupy more memory than strictly
   * necessary; to trim memory usage, build using {@code builder.build().trimmed()}.
   */
  public static Builder builder(int initialCapacity) {
    checkArgument(initialCapacity >= 0, "Invalid initialCapacity: %s", initialCapacity);

With rewrites, redirects, follow-ups and retries, your simple request may yield many requests and responses. OkHttp uses `Call` to model the task of satisfying your request through however many intermediate requests and responses are necessary. Typically this isn’t many! But it’s comforting to know that your code will continue to work if your URLs are redirected or if you failover to an alternate IP address.

Calls are executed in one of two ways:

 * This interface allows users to define and configure various toolchains
 * that can be utilized by Maven during the build process. Toolchains can
 * include compilers, interpreters, and other tools that are necessary
 * for building a project in a specific environment.</p>
 *
 * <p>Toolchains are defined in the Maven toolchains.xml file and can be
 * referenced in the project's POM file. This allows for greater flexibility

import org.junit.jupiter.api.Test;

import static java.util.Objects.nonNull;

/**
 * Pseudo test to generate documentation fragment about supported CLI options. TODO such documentation generation code
 * should not be necessary as unit test but should be run during site generation (Velocity? Doxia macro?)
 */
class CLIManagerDocumentationTest {
    private static final String LS = System.lineSeparator();

      pull-requests: read
    # Set job outputs to values from filter step
    outputs:
      docs: ${{ steps.filter.outputs.docs }}
    steps:
    - uses: actions/checkout@v4
    # For pull requests it's not necessary to checkout the code but for the main branch it is
    - uses: dorny/paths-filter@v3
      id: filter
      with:
        filters: |
          docs:
            - README.md
            - docs/**

After you have tests, then you can upgrade the **FastAPI** version to a more recent one, and make sure that all your code is working correctly by running your tests.

If everything is working, or after you make the necessary changes, and all your tests are passing, then you can pin your `fastapi` to that new recent version.

## About Starlette

You shouldn't pin the version of `starlette`.

  echo "Downloading ztunnel: $1 to $2"
  time ${DOWNLOAD_COMMAND} --header "${AUTH_HEADER:-}" "$1" > "$2"
  chmod +x "$2"

  # Make a copy named just "ztunnel" in the same directory (overwrite if necessary).
  echo "Copying $2 to $(dirname "$2")/${3}"
  cp -f "$2" "$(dirname "$2")/${3}"
  popd

  # Also copy it to out/$os_arch/ztunnel as that's whats used in the build
  echo "Copying '${2}' to ${TARGET_OUT_LINUX}/ztunnel"

///

The only thing the function returned by `GzipRequest.get_route_handler` does differently is convert the `Request` to a `GzipRequest`.

Doing this, our `GzipRequest` will take care of decompressing the data (if necessary) before passing it to our *path operations*.

After that, all of the processing logic is the same.

 * doesn't cache partial responses.
 *
 * ## Force a Network Response
 *
 * In some situations, such as after a user clicks a 'refresh' button, it may be necessary to skip
 * the cache, and fetch data directly from the server. To force a full refresh, add the `no-cache`
 * directive:
 *
 * ```java
 * Request request = new Request.Builder()

    // Optimization - use signed division if dividend < 2^63
    if (dividend >= 0) {
      return dividend / divisor;
    }

    /*
     * Otherwise, approximate the quotient, check, and correct if necessary. Our approximation is
     * guaranteed to be either exact or one less than the correct value. This follows from fact that
     * floor(floor(x)/i) == floor(x/i) for any real x and integer i != 0. The proof is not quite