TFCI_BAZEL_BAZELRC_ARGS=
TFCI_BAZEL_COMMON_ARGS=
TFCI_BAZEL_TARGET_SELECTING_CONFIG_PREFIX=
TFCI_BUILD_PIP_PACKAGE_BASE_ARGS=
TFCI_BUILD_PIP_PACKAGE_WHEEL_NAME_ARG=
# Mostly meant to be used for Windows.
# For Windows, the same wheel is uploaded to both tensorflow_cpu, and tensorflow
TFCI_BUILD_PIP_PACKAGE_ADDITIONAL_WHEEL_NAMES=
TFCI_DOCKER_ARGS=
TFCI_DOCKER_ENABLE=
TFCI_DOCKER_IMAGE=
TFCI_DOCKER_PULL_ENABLE=
TFCI_DOCKER_REBUILD_ARGS=

   * (unless the executor has been shutdown).
   *
   * <p>The returned executor is backed by the executor returned by {@link
   * MoreExecutors#newDirectExecutorService} and subject to the same constraints.
   *
   * <p>Although all tasks are immediately executed in the thread that submitted the task, this
   * {@code ExecutorService} imposes a small locking overhead on each task submission in order to

- **Protocol Layers**: Separate SMB1 (`jcifs.internal.smb1/`) and SMB2/3 (`jcifs.internal.smb2/`) implementations
- **Authentication**: NTLM, Kerberos, SPNEGO unified subsystem
- **Network Layer**: Transport abstraction, NetBIOS name resolution, session management

### Contributing

We welcome contributions! Please:
1. Fork the repository
2. Create a feature branch
3. Make your changes with appropriate tests

///

## Advanced User Guide { #advanced-user-guide }

There is also an **Advanced User Guide** that you can read later after this **Tutorial - User guide**.

The **Advanced User Guide** builds on this one, uses the same concepts, and teaches you some extra features.

But you should first read the **Tutorial - User Guide** (what you are reading right now).

    assertEquals(if (connectionType == H2) 1 else 0, client.connectionPool.connectionCount())

    cancelLatch.await()

    val events = listener.eventSequence.filter { isConnectionEvent(it) }.map { it.name }
    listener.clearAllEvents()

    assertThat(events).startsWith("CallStart", "ConnectStart", "ConnectEnd", "ConnectionAcquired")
    if (cancelMode == CANCEL) {
      assertThat(events).contains("Canceled")

   * {@code put()} or {@code putAll()}, or {@code setValue()} on its entries.
   *
   * <p>In contrast, the maps returned by {@code rowMap().get()} have the same behavior as those
   * returned by {@link #row}. Those maps may support {@code setValue()}, {@code put()}, and {@code
   * putAll()}.
   *
   * @return a map view from each row key to a secondary map from column keys to values

////

## 파이썬에서 환경 변수 읽기

파이썬 **바깥**인 터미널에서(다른 도구로도 가능) 환경 변수를 생성도 할 수도 있고, 이를 **파이썬에서 읽을 수 있습니다.**

예를 들어 다음과 같은 `main.py` 파일이 있다고 합시다:

```Python hl_lines="3"
import os

name = os.getenv("MY_NAME", "World")
print(f"Hello {name} from Python")
```

/// tip | 팁

<a href="https://docs.python.org/3.8/library/os.html#os.getenv" class="external-link" target="_blank">`os.getenv()`</a> 의 두 번째 인자는 반환할 기본값입니다.

 *
 * <p>This uses the same efficient implementation as {@link Ordering#leastOf(Iterable, int)},
 * offering expected O(n + k log k) performance (worst case O(n log k)) for n calls to {@link
 * #offer} and a call to {@link #topK}, with O(k) memory. In comparison, quickselect has the same
 * asymptotics but requires O(n) memory, and a {@code PriorityQueue} implementation takes O(n log

        String ip = "192.168.1.3";

        // Make attempts from same IP with different users
        for (int i = 1; i <= 5; i++) {
            assertTrue(rateLimiter.checkAttempt("user" + i, ip), "Attempt " + i + " should be allowed");
            rateLimiter.recordFailure("user" + i, ip);
        }

        // 6th attempt from same IP should be blocked

nt),l=-1!==e.placement.indexOf('-'),f=t?a||l||s%2==d%2?r:Z:p,m=t?r:p;return{left:f(1==s%2&&1==d%2&&!l&&t?n.left-1:n.left),top:m(n.top),bottom:m(n.bottom),right:f(n.right)}}function K(e,t,o){var n=D(e,function(e){var o=e.name;return o===t}),i=!!n&&e.some(function(e){return e.name===o&&e.enabled&&e.order<n.order});if(!i){var r='`'+t+'`';console.warn('`'+o+'`'+' modifier is required by '+r+' modifier in order to work, be sure to include it before '+r+'!')}return i}function z(e){return'end'===e?'sta...