But if you are certain that the content that you are returning is **serializable with JSON**, you can pass it directly to the response class and avoid the extra overhead that FastAPI would have by passing your return content through the `jsonable_encoder` before passing it to the response class.

{* ../../docs_src/custom_response/tutorial001b_py39.py hl[2,7] *}

/// info

    /** Drains the reference queues used by this segment, if any. */
    @GuardedBy("this")
    void maybeDrainReferenceQueues() {}

    /** Clears the reference queues used by this segment, if any. */
    void maybeClearReferenceQueues() {}

    /** Sets the value of the given {@code entry}. */
    void setValue(E entry, V value) {
      this.map.entryHelper.setValue(self(), entry, value);
    }

  return new TF_AttrBuilder(op_name);
}

void TF_DeleteAttrBuilder(TF_AttrBuilder* builder) { delete builder; }

void TF_AttrBuilderSetType(TF_AttrBuilder* builder, const char* attr_name,
                           TF_DataType value) {
  auto iter = builder->attr_names.insert(attr_name).first;
  builder->Set(*iter, static_cast<tensorflow::DataType>(value));
}

			if err != nil {
				return err
			}
			nodeBuckets[idx] = localBuckets
			return nil
		}, idx)
	}

	errs := g.Wait()

	// The list of buckets in a map to avoid duplication
	resultMap := make(map[string]BucketInfo)

	for poolIdx := 0; poolIdx < sys.poolsCount; poolIdx++ {
		perPoolErrs := make([]error, 0, len(sys.peerClients))
		for i, client := range sys.peerClients {

Also, constants are always evaluated as 64-bit unsigned integers.
Thus <code>-2</code> is not the integer value minus two,
but the unsigned 64-bit integer with the same bit pattern.
The distinction rarely matters but
to avoid ambiguity, division or right shift where the right operand's
high bit is set is rejected.
</p>

<h3 id="symbols">Symbols</h3>

<p>
Some symbols, such as <code>R1</code> or <code>LR</code>,

And any model class that doesn't have `table=True` is a **data model**, these ones are actually just Pydantic models (with a couple of small extra features). 🤓

With SQLModel, we can use **inheritance** to **avoid duplicating** all the fields in all the cases.

#### `HeroBase` - the base class { #herobase-the-base-class }

Let's start with a `HeroBase` model that has all the **fields that are shared** by all the models:

	g := errgroup.WithNErrs(len(buckets))
	bucketMetas := make([]BucketMetadata, len(buckets))
	for index := range buckets {
		g.Go(func() error {
			// Sleep and stagger to avoid blocked CPU and thundering
			// herd upon start up sequence.
			time.Sleep(25*time.Millisecond + time.Duration(rand.Int63n(int64(100*time.Millisecond))))

   * {@link Builder} and call {@link Builder#orderRowsBy}, {@link Builder#orderColumnsBy}, and
   * {@link Builder#putAll}
   *
   * <p>Despite the method name, this method attempts to avoid actually copying the data when it is
   * safe to do so. The exact circumstances under which a copy will or will not be performed are
   * undocumented and subject to change.
   */

	erasureLockOwner string

	// List of endpoints provided on the command line.
	endpoints PoolEndpoints

	// String version of all the endpoints, an optimization
	// to avoid url.String() conversion taking CPU on
	// large disk setups.
	endpointStrings []string

	// Total number of sets and the number of disks per set.
	setCount, setDriveCount int
	defaultParityCount      int

    void clearReferenceQueues() {
      if (map.usesKeyReferences()) {
        clearKeyReferenceQueue();
      }
      if (map.usesValueReferences()) {
        clearValueReferenceQueue();
      }
    }

    void clearKeyReferenceQueue() {
      while (keyReferenceQueue.poll() != null) {}
    }

    void clearValueReferenceQueue() {