[XTsat]

Resolved #10577

选择镜像站时，现在能直接看到每个 GitHub 加速地址和 PyPI 源通不通、有多快，不用再一个个试

Modifications / 改动点

后端

• astrbot/dashboard/services/stat_service.py：新增 test_pypi_index_connection（只取响应头、使用 pip 的 UA、拒绝非 http(s) 输入）和 measure_pypi_index_speed（从索引页解析出最新的 -py3-none-any.whl 并真实下载）；新增 measure_ghproxy_speed，通过代理流式下载一个真实的仓库文件；test_ghproxy_connection 现在把空 proxy_url 视作"直连 GitHub"，使"不使用加速"与各镜像在同等条件下被测量；统一采样预算（6 秒 / 4 MiB），避免一次点击拉取几十 MB 流量
• astrbot/dashboard/api/stats.py：新增 POST /api/v1/stats/pypi-index/test、/pypi-index/speed、/ghproxy/speed，均为 system scope
• astrbot/dashboard/schemas.py：新增 PypiIndexTestRequest
• openspec/openapi-v1.yaml、dashboard/src/api/generated/openapi-v1/*、dashboard/src/api/v1.ts：契约、重新生成的客户端、statsApi 封装

前端

• dashboard/src/components/shared/PypiMirrorSelector.vue（新增）：官方、阿里云、腾讯云、中科大、清华五个预设加自定义 URL，每项带延迟与 MB/s chip，并提供两个批量测试按钮
• dashboard/src/components/shared/ProxySelector.vue：把两级开关拍平为单个单选列表（none / 预设下标 / custom）；自定义输入内容独立存放并使用独立 localStorage 键，因此切换预设不会丢失，未被选中时也仍可被测试；每条线路（含"不使用代理"）都显示延迟与 MB/s chip
• dashboard/src/views/Settings.vue：新增 PyPI 源独立卡片；pypi_index_url 从通用 network 组移除，保证该字段只渲染一次
• zh-CN、en-US、ja-JP、ru-RU 四套 i18n

入口对照（旧 → 新）

• 设置 → 网络 →「GitHub 加速」开关 → 「选择 GitHub 镜像源」扁平列表，其中"不使用 GitHub 镜像"成为一等选项。

• 设置 → 网络 → 通用 pypi_index_url 配置行 → 「PyPI 软件仓库地址」卡片。

• This is NOT a breaking change. / 这不是一个破坏性变更。

Screenshots or Test Results / 运行截图或测试结果

---

Checklist / 检查清单

• 😊 If there are new features added in the PR, I have discussed it with the authors through issues/emails, etc.
  / 如果 PR 中有新加入的功能，已经通过 Issue / 邮件等方式和作者讨论过。

• 👀 My changes have been well-tested, and "Verification Steps" and "Screenshots" have been provided above.
  / 我的更改经过了良好的测试，并已在上方提供了“验证步骤”和“运行截图”。

• 📚 I checked the affected WebUI instructions and screenshots in docs/zh and docs/en against the changed navigation, page structure, and labels, and updated them in this PR (or explained why no documentation update is needed). For renamed, moved, or merged entry points, I included an old entry → new entry mapping in the documentation and changelog.
  / 我已对照变化后的 WebUI 入口、页面结构和术语，核对并在本 PR 中更新 docs/zh 和 docs/en 的相关操作说明与截图（或说明无需更新文档的原因）。入口改名、移动或合并时，已在文档和 changelog 中补充 旧入口 → 新入口 对照。

• 🤓 I have ensured that no new dependencies are introduced, OR if new dependencies are introduced, they have been added to the appropriate locations in requirements.txt and pyproject.toml.
  / 我确保没有引入新依赖库，或者引入了新依赖库的同时将其添加到 requirements.txt 和 pyproject.toml 文件相应位置。

• 😮 My changes do not introduce malicious code.
  / 我的更改没有引入恶意代码。

Summary by Sourcery

Add network source pickers that let users compare PyPI mirrors and GitHub routes by availability, latency, and download speed.

New Features:

• Add PyPI mirror selection with official, preset, and custom indexes.
• Add connection latency and download-speed testing for PyPI indexes and GitHub routes.
• Expose direct GitHub access as a selectable and measurable network route.

Bug Fixes:

• Prevent PyPI index configuration from being rendered twice in Settings.
• Preserve custom proxy and mirror URLs when switching between presets.

Enhancements:

• Flatten GitHub proxy selection into a single radio list with status and throughput indicators.
• Apply bounded sampling to network speed tests to limit download time and data usage.

Chores:

• Extend the statistics API contract, generated client, schemas, and localized settings labels for network source testing.

[sourcery-ai]

Hey - I've found 2 issues

Prompt for AI Agents

Please address the comments from this code review:

## Individual Comments

### Comment 1
<location path="astrbot/dashboard/services/stat_service.py" line_range="655" />
<code_context>
+                        raise StatServiceError(
+                            f"Failed. Status code: {response.status}"
+                        )
+                    page = await response.text()
+
+                wheels: list[tuple[Version, str]] = []
</code_context>
<issue_to_address>
**Large index pages exhaust memory**

When a configured or caller-supplied PyPI index returns a very large `/pip/` page, `measure_pypi_index_speed` reads the complete response into `page` with `response.text()` before applying any size limit. Large or overlapping responses consume dashboard memory and can exhaust the process.

Read the response with a byte limit and stop or reject it once that limit is reached.
</issue_to_address>

### Comment 2
<location path="astrbot/dashboard/services/stat_service.py" line_range="658" />
<code_context>
+                    page = await response.text()
+
+                wheels: list[tuple[Version, str]] = []
+                for href in re.findall(r'href="([^"#]+)', page):
+                    file_name = href.rsplit("/", 1)[-1]
+                    suffix = "-py3-none-any.whl"
</code_context>
<issue_to_address>
**Valid index markup fails speed test**

When a custom PyPI index serves its links with single-quoted `href` attributes or uppercase `HREF`, `measure_pypi_index_speed()` only extracts lowercase, double-quoted `href` attributes, so a valid simple-index page using single quotes or uppercase `HREF` yields no wheel links and the speed test reports “No pip wheel found.”

Parse the page with an HTML parser that handles valid attribute quoting and casing.
</issue_to_address>

Sourcery assessment

Needs a human reviewer. 2 findings to address first, and the new system-scoped speed probe fetches the index URL and then follows wheel links from its page, including absolute links, so a supplied index can make the server send GET requests to internal hosts. Reverting removes the probe but cannot undo any side effects those requests may already have triggered.

Blocking findings: astrbot/dashboard/services/stat_service.py:655, astrbot/dashboard/services/stat_service.py:658

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook