cd .github/workflows/multipart/

function cleanup() {
	docker-compose -f docker-compose-site1.yaml rm -s -f || true
	docker-compose -f docker-compose-site2.yaml rm -s -f || true
	for volume in $(docker volume ls -q | grep minio); do
		docker volume rm ${volume} || true
	done

	docker system prune -f || true
	docker volume prune -f || true
	docker volume rm $(docker volume ls -q -f dangling=true) || true
}

cleanup

        shell: bash
        run: |
          running_containers=$(docker ps -q) && \
          if [[ $running_containers == "" ]]; then
            echo "No running containers";
          else
            echo "Running container(s) found" && \
            docker stop $running_containers;
          fi
          docker container prune -f
          docker image prune -af
      - name: Clean repository
        shell: bash

* **安全密码**，默认使用密码哈希
* **JWT 令牌**身份验证
* **SQLAlchemy** 模型（独立于 Flask 扩展，可直接用于 Celery Worker）
* 基础的用户模型（可按需修改或删除）
* **Alembic** 迁移
* **CORS**（跨域资源共享）
* **Celery** Worker 可从后端其它部分有选择地导入并使用模型和代码
* REST 后端测试基于 Pytest，并与 Docker 集成，可独立于数据库实现完整的 API 交互测试。因为是在 Docker 中运行，每次都可从头构建新的数据存储（使用 ElasticSearch、MongoDB、CouchDB 等数据库，仅测试 API 运行）
* Python 与 **Jupyter Kernels** 集成，用于远程或 Docker 容器内部开发，使用 Atom Hydrogen 或 Visual Studio Code 的 Jupyter 插件
* **Vue** 前端：

TFCI_BAZEL_TARGET_SELECTING_CONFIG_PREFIX=linux_cpu
TFCI_BUILD_PIP_PACKAGE_ARGS="--repo_env=WHEEL_NAME=tensorflow_cpu"
TFCI_DOCKER_ENABLE=1
TFCI_DOCKER_IMAGE=tensorflow/build:2.16-python${TFCI_PYTHON_VERSION}
TFCI_DOCKER_PULL_ENABLE=1
TFCI_DOCKER_REBUILD_ARGS="--build-arg PYTHON_VERSION=python$TFCI_PYTHON_VERSION --target=devel tensorflow/tools/tf_sig_build_dockerfiles"
TFCI_INDEX_HTML_ENABLE=1

# despite lacking Nvidia CUDA support.
TFCI_BUILD_PIP_PACKAGE_ARGS="--repo_env=WHEEL_NAME=tensorflow"
TFCI_DOCKER_ENABLE=1
TFCI_DOCKER_IMAGE=gcr.io/tensorflow-sigs/build-arm64:tf-2-16-multi-python
TFCI_DOCKER_PULL_ENABLE=1
TFCI_DOCKER_REBUILD_ARGS="--target=tf ci/official/containers/linux_arm64"
TFCI_INDEX_HTML_ENABLE=1
TFCI_LIB_SUFFIX="-cpu-linux-arm64"
TFCI_OUTPUT_DIR=build_output

MOUNT_DEST="${MOUNT_DEST:-/work}"

read -ra DOCKER_RUN_OPTIONS <<< "${DOCKER_RUN_OPTIONS:-}"

[[ -t 1 ]] && DOCKER_RUN_OPTIONS+=("-it")
[[ ${UID} -ne 0 ]] && DOCKER_RUN_OPTIONS+=(-u "${UID}:${DOCKER_GID}")

# $CONTAINER_OPTIONS becomes an empty arg when quoted, so SC2086 is disabled for the
# following command only
# shellcheck disable=SC2086
"${CONTAINER_CLI}" run \
    --rm \
    "${DOCKER_RUN_OPTIONS[@]}" \
    --init \

!!! tip "Заметка"
    Если вы не знаете, что такое **контейнеры**, Docker или Kubernetes, не переживайте.

    Я поведаю Вам о контейнерах, образах, Docker, Kubernetes и т.п. в главе: [FastAPI внутри контейнеров - Docker](docker.md){.internal-link target=_blank}.

## Шаги, предшествующие запуску

        track: stable
    spec:
      containers:
      - image: fake.docker.io/google-samples/hello-go-gke:1.0
        name: hello
        ports:
        - containerPort: 80
          name: http
        resources: {}
      - image: docker.io/istio/proxy_debug:unittest
        name: istio-proxy
        resources: {}
      initContainers:
      - image: docker.io/istio/proxy_init:unittest-testiop
        name: istio-init

  },
  "features": {
    "ghcr.io/devcontainers/features/docker-outside-of-docker:1": {},
    "ghcr.io/mpriscella/features/kind:1": {}
  },
  "customizations": {
    "vscode": {
      "extensions": [
        "golang.go",
        "rust-lang.rust-analyzer",
        "eamodio.gitlens",
        "zxh404.vscode-proto3",
        "ms-azuretools.vscode-docker",
        "redhat.vscode-yaml",
        "IBM.output-colorizer"

        shell: bash
        run: |
          running_containers=$(docker ps -q) && \
          if [[ $running_containers == "" ]]; then
            echo "No running containers";
          else
            echo "Running container(s) found" && \
            docker stop $running_containers;
          fi
          docker container prune -f
      - name: Clean repository
        shell: bash