* removed in one segment while checking another, in which case the table was never actually
     * empty at any point. (The sum ensures accuracy up through at least 1<<31 per-segment
     * modifications before recheck.)  Method containsValue() uses similar constructions for
     * stability checks.
     */
    long sum = 0L;
    Segment<K, V, E, S>[] segments = this.segments;

     * removed in one segment while checking another, in which case the table was never actually
     * empty at any point. (The sum ensures accuracy up through at least 1<<31 per-segment
     * modifications before recheck.)  Method containsValue() uses similar constructions for
     * stability checks.
     */
    long sum = 0L;
    Segment<K, V, E, S>[] segments = this.segments;

   * {@code iterable} contains fewer than {@code numberToSkip} elements, the returned iterable skips
   * all of its elements.
   *
   * <p>Modifications to the underlying {@link Iterable} before a call to {@code iterator()} are
   * reflected in the returned iterator. That is, the iterator skips the first {@code numberToSkip}

   * {@code iterable} contains fewer than {@code numberToSkip} elements, the returned iterable skips
   * all of its elements.
   *
   * <p>Modifications to the underlying {@link Iterable} before a call to {@code iterator()} are
   * reflected in the returned iterator. That is, the iterator skips the first {@code numberToSkip}

[graalvm]: https://www.graalvm.org/
[graalvm_21]: https://www.graalvm.org/release-notes/21_0/
[graalvm_22]: https://www.graalvm.org/release-notes/22_2/
[idna_15_1_0]: https://www.unicode.org/reports/tr46/#Modifications
[kotlin_1_4_20]: https://github.com/JetBrains/kotlin/releases/tag/v1.4.20
[kotlin_1_5_31]: https://github.com/JetBrains/kotlin/releases/tag/v1.5.31
[kotlin_1_6_10]: https://github.com/JetBrains/kotlin/releases/tag/v1.6.10

   * {@code k} to {@code function.apply(k)}. The {@code keySet}, {@code values}, and {@code
   * entrySet} views of the returned map iterate in the same order as the backing set.
   *
   * <p>Modifications to the backing set are read through to the returned map. The returned map
   * supports removal operations if the backing set does. Removal operations write through to the

   * {@code k} to {@code function.apply(k)}. The {@code keySet}, {@code values}, and {@code
   * entrySet} views of the returned map iterate in the same order as the backing set.
   *
   * <p>Modifications to the backing set are read through to the returned map. The returned map
   * supports removal operations if the backing set does. Removal operations write through to the

Ce faisant, notre `GzipRequest` se chargera de décompresser les données (si nécessaire) avant de les transmettre à nos *chemins d'accès*.

Après cela, toute la logique de traitement est identique.

Mais grâce à nos modifications dans `GzipRequest.body`, le corps de la requête sera automatiquement décompressé lorsque **FastAPI** le chargera, si nécessaire.

     * removed in one segment while checking another, in which case the table was never actually
     * empty at any point. (The sum ensures accuracy up through at least 1<<31 per-segment
     * modifications before recheck.) Method containsValue() uses similar constructions for
     * stability checks.
     */
    long sum = 0L;
    Segment<K, V>[] segments = this.segments;

//
// We expose this here to allow fine-grained synchronization in multi-threaded
// workloads, which is required since the Python implementation depends on the
// above mutation methods. This allows us to prevent modifications to nodes in
// the graph after the session has been made aware of them.
TF_CAPI_EXPORT extern void TF_ExtendSession(TF_Session* session,
                                            TF_Status* status);