Doc Release

# 只有 MindSpore 有 tag
# [repo_name, has_git, source_branch, target_html_branch, tag, url_name]
# ['tutorials', False, 'master', 'r2.10.0', '', 'tutorials'],
# ['mindscience', True, 'master', 'r0.7', '', 'mindscience'],
# ['mindflow', False, 'master', 'r0.3', '', 'mindflow'],
# ['mindearth', False, 'master', 'r0.3', '', 'mindearth'],
# ['mindformers', True, 'master', 'r2.0.0', '', 'mindformers'],
# ['mindspore', True, 'master', 'r2.10.0', 'v2.10.0', 'docs'],
# ['mindspore-lite', True, 'master', 'r2.10.0', 'r2.10', 'lite'],
# ['vllm-mindspore', True, 'master', 'r0.5.0', '', 'vllm_mindspore'],
# ['golden-stick', True, 'master', 'r1.3.0', '', 'golden_stick'],
# ['mindquantum', True, 'master', 'r0.11', '', 'mindquantum'],
# ['mindstudio', False, 'master', '81RC1', '', 'mindstudio'],

https://www.mindspore.cn/tutorials/zh-CN/master/index.html
                          ^^^^^^^^^         ^^^^^^
                          url_name          source_branch

https://www.mindspore.cn/tutorials/zh-CN/r2.10.0/index.html

"source": [
  "[![View Source On AtomGit](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source_en.svg)](https://atomgit.com/mindspore/docs/blob/master/tutorials/source_en/dataset/record.ipynb)\n"
]

import os
import shutil

DOCS_COMPONENTS_DIR = "./docs/docs"

# 构建保留列表：sample_code + 用户指定的组件白名单
keep_list = ["sample_code"] + {{ parameters.components_to_keep }}

print(f"保留的组件: {keep_list}")

if os.path.isdir(DOCS_COMPONENTS_DIR):
    for entry in os.listdir(DOCS_COMPONENTS_DIR):
        full_path = os.path.join(DOCS_COMPONENTS_DIR, entry)
        if os.path.isdir(full_path) and entry not in keep_list:
            print(f"删除组件: {entry}")
            shutil.rmtree(full_path)
else:
    print(f"目录不存在: {DOCS_COMPONENTS_DIR}")

import os
import re
import logging
import requests
from concurrent.futures import ThreadPoolExecutor, as_completed

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

# ============ 链接检测函数 ============
def check_url(url, timeout=5):
    """检测 URL 是否可访问（HEAD 请求）"""
    try:
        resp = requests.head(url, timeout=timeout, allow_redirects=True)
        return resp.status_code < 400
    except Exception:
        return False

def validate_links(link_list, max_workers=5):
    """并发检测链接列表"""
    if not link_list:
        return

    logger.info(f"开始检测 {len(link_list)} 个链接...")
    failed_links = []

    with ThreadPoolExecutor(max_workers=max_workers) as executor:
        future_to_url = {executor.submit(check_url, url): url for url in link_list}
        for future in as_completed(future_to_url):
            url = future_to_url[future]
            try:
                is_valid = future.result()
                if not is_valid:
                    failed_links.append(url)
                    logger.warning(f"⚠️ 链接可能无效: {url}")
            except Exception as e:
                failed_links.append(url)
                logger.error(f"检测失败 {url}: {e}")

    if failed_links:
        logger.error(f"发现 {len(failed_links)} 个可能无效的链接，请检查:")
        for link in failed_links:
            logger.error(f"  - {link}")
    else:
        logger.info("✅ 所有检测的链接均有效")

# ============ 主替换逻辑 ============
def replace_urls(target_paths, sub_list, enable_check=True):
    """遍历目标目录，执行链接替换"""
    all_replaced_links = []

    for tp in target_paths:
        if not os.path.exists(tp):
            logger.warning(f"路径不存在: {tp}")
            continue

        for root, dirs, files in os.walk(tp):
            for file in files:
                file_path = os.path.join(root, file)
                if not file.endswith(('.md', '.txt', '.rst', '.html', '.yml', '.yaml', '.ipynb')):
                    continue

                try:
                    with open(file_path, 'r', encoding='utf-8') as f:
                        content = f.read()

                    new_content = content
                    for pattern, replacement in sub_list:
                        try:
                            new_content = re.sub(pattern, replacement, new_content, flags=re.DOTALL)
                        except re.error as e:
                            logger.error(f"正则表达式错误: {pattern}, {e}")
                            continue

                    if new_content != content:
                        with open(file_path, 'w', encoding='utf-8') as f:
                            f.write(new_content)
                        logger.info(f"已更新: {file_path}")

                        if enable_check:
                            atomgit_links = re.findall(
                                r'https?://atomgit\.com/[^\s<>"\'\)]+',
                                new_content
                            )
                            all_replaced_links.extend(atomgit_links)

                except UnicodeDecodeError:
                    continue
                except Exception as e:
                    logger.error(f"处理失败 {file_path}: {e}")
                    continue

    if enable_check and all_replaced_links:
        validate_links(list(set(all_replaced_links)))

# ============ 配置区域 ============
target_paths = [
    './docs/docs',
    './docs/tutorials',
    './docs/install',
]

SOURCE_BRANCH = "{{ parameters.source_branch }}"
ENABLE_LINK_CHECK = {{ parameters.enable_link_check | default(True) | lower }}

# 组件配置：[repo_name, has_git, target_html_branch, tag, url_name]
# 只有 MindSpore 有 tag
component_configs = [
    {% for component, config in parameters.component_branch_map.items() %}
    [
        "{{ config.repo_name }}",
        {{ config.has_git | lower }},
        "{{ config.target_html_branch }}",
        "{{ config.tag if config.tag else '' }}",
        "{{ config.url_name }}"
    ],
    {% endfor %}
]

all_sub_list = []

# ============================================
# 第一部分：OBS 链接替换（通用）
# 使用 target_html_branch
# ============================================
for repo_name, has_git, target_html_branch, tag, url_name in component_configs:
    # OBS website-images 链接
    # 示例: website-images/master/resource/ -> website-images/r2.10.0/resource/
    all_sub_list.append([
        rf'(mindspore-website\.obs\.cn-north-4\.myhuaweicloud\.com/website-images)/{SOURCE_BRANCH}',
        rf'\1/{target_html_branch}'
    ])

    # OBS notebook 链接
    # 示例: notebook/master/tutorials/ -> notebook/r2.10.0/tutorials/
    all_sub_list.append([
        rf'(mindspore-website\.obs\.cn-north-4\.myhuaweicloud\.com/notebook)/{SOURCE_BRANCH}(/[\w\d/_.-]*?)',
        rf'\1/{target_html_branch}\2'
    ])

# ============================================
# 第二部分：各组件专属链接替换
# ============================================
for repo_name, has_git, target_html_branch, tag, url_name in component_configs:

    # ----- 处理 mindspore（docs 仓链接）-----
    if repo_name == 'mindspore':
        # atomgit docs 仓链接 - 使用 target_html_branch
        # 示例: docs/blob/master/tutorials/ -> docs/blob/r2.10.0/tutorials/
        all_sub_list.append([
            rf'(atomgit\.com/mindspore/docs/(?:blob|tree))/{SOURCE_BRANCH}(/[\w\d/_.-]*?)',
            rf'\1/{target_html_branch}\2'
        ])

    # ----- 处理 tutorials -----
    if repo_name == 'tutorials':
        # 官网教程链接 - 使用 target_html_branch
        # 示例: mindspore.cn/tutorials/zh-CN/master/ -> mindspore.cn/tutorials/zh-CN/r2.10.0/
        all_sub_list.append([
            rf'(mindspore\.cn/tutorials/[\w\d/_.-]*?)/{SOURCE_BRANCH}',
            rf'\1/{target_html_branch}'
        ])
        # 带 www 前缀
        all_sub_list.append([
            rf'(www\.mindspore\.cn/tutorials/[\w\d/_.-]*?)/{SOURCE_BRANCH}',
            rf'\1/{target_html_branch}'
        ])
        # atomgit docs 仓教程链接
        all_sub_list.append([
            rf'(atomgit\.com/mindspore/docs/(?:blob|tree))/{SOURCE_BRANCH}(/[\w\d/_.-]*?tutorials)',
            rf'\1/{target_html_branch}\2'
        ])

    # ----- 处理普通组件（非教程）-----
    if repo_name != 'tutorials':
        # 官网链接 - 使用 target_html_branch
        # 示例: mindspore.cn/docs/zh-CN/master/ -> mindspore.cn/docs/zh-CN/r2.10.0/
        all_sub_list.append([
            rf'(mindspore\.cn/{url_name}/[\w\d/_.-]*?)/{SOURCE_BRANCH}',
            rf'\1/{target_html_branch}'
        ])
        # 带 www 前缀
        all_sub_list.append([
            rf'(www\.mindspore\.cn/{url_name}/[\w\d/_.-]*?)/{SOURCE_BRANCH}',
            rf'\1/{target_html_branch}'
        ])

        # 仓库链接处理（仅当有独立仓库且有 tag 时）
        if has_git and tag:
            # atomgit 仓库链接 - 使用 tag
            # 示例: atomgit.com/mindspore/mindspore/blob/master/ -> blob/v2.10.0/
            all_sub_list.append([
                rf'(atomgit\.com/mindspore/{repo_name}/(?:blob|tree))/{SOURCE_BRANCH}(/[\w\d/_.-]*?)',
                rf'\1/{tag}\2'
            ])

            # git clone 命令（仅替换带 -b 参数的）
            # 示例: git clone -b master https://atomgit.com/mindspore/mindspore.git
            #    -> git clone -b v2.10.0 https://atomgit.com/mindspore/mindspore.git
            all_sub_list.append([
                f'git clone -b {SOURCE_BRANCH} https://atomgit.com/mindspore/{repo_name}.git',
                f'git clone -b {tag} https://atomgit.com/mindspore/{repo_name}.git'
            ])
        elif has_git and not tag:
            # 有独立仓库但没有 tag，只处理官网链接
            logger.info(f"⚠️ 组件 {repo_name} 有独立仓库但未配置 tag，仓库链接暂不替换")

        # 无独立仓库的组件（has_git=False），只处理官网链接

# ============================================
# 打印所有替换规则用于调试
# ============================================
logger.info("========== 替换规则列表 ==========")
logger.info(f"源分支: {SOURCE_BRANCH}")
logger.info(f"链接检测: {'启用' if ENABLE_LINK_CHECK else '禁用'}")
for pattern, replacement in all_sub_list:
    logger.info(f"替换: {pattern} -> {replacement}")
logger.info("==================================")

# ============================================
# 执行替换
# ============================================
replace_urls(target_paths, all_sub_list, enable_check=ENABLE_LINK_CHECK)

https://www.mindspore.cn/tutorials/zh-CN/master/index.html
→ https://www.mindspore.cn/tutorials/zh-CN/r2.10.0/index.html