https://shenjuexiao.com@shenjuexiao神觉晓http://www.rssboard.org/rss-specificationpython-feedgenhttps://github.githubassets.com/favicons/favicon.svghttps://shenjuexiao.comSat, 18 Jul 2026 08:42:00 +0000神觉晓60神觉晓https://shenjuexiao.com/post/7.html# actions-gh-pages ## 内容来源 - [actions-gh-pages](https://github.com/peaceiris/actions-gh-pages) - [GitHub凭证区别](https://chat.deepseek.com/share/37lldwwwu2m7wobum4)[（Deepseek）](https://chat.deepseek.com/a/chat/s/351bd0f3-9c56-42f0-ac38-91bf90c0e184) - [GitHub令牌区别](https://chat.deepseek.com/share/8m05dxz9pqrkl0wo8s)[（Deepseek）](https://chat.deepseek.com/a/chat/s/26baece4-d43e-48eb-81d4-d6de0ece8f04) actions-gh-pages 这个项目非常不错，利用 `Github Actions` 部署静态页面。https://shenjuexiao.com/post/7.htmlSat, 18 Jul 2026 08:41:32 +0000https://shenjuexiao.com/post/6.html# Astro 模板 ## Astro Paper  高评分、暗黑模式、RSS/SEO 完善 Demo: [astro-paper.pages.dev](https://astro-paper.pages.dev/) Github: [satnaing/astro-paper](https://github.com/satnaing/astro-paper) ## Astrofy  含简历/商店/博客多合一，CV+ 项目画廊 Demo: [astrofy-template.netlify.app](https://astrofy-template.netlify.app/) Github: [manuelernestog/astrofy](https://github.com/manuelernestog/astrofy) ## EV0  极简响应式 Demo: [ev0.gndx.dev](https://ev0.gndx.dev/) Github: [withastro/starlight](https://github.com/gndx/ev0-astro-theme) ## Firefly  专为技术文档设计，支持多语言、搜索、版本切换 Demo: [firefly.cuteleaf.cn](https://firefly.cuteleaf.cn/) Github: [CuteLeaf/Firefly](https://github.com/CuteLeaf/Firefly) Docs: [docs-firefly.cuteleaf.cn](https://docs-firefly.cuteleaf.cn/zh/) ## Pure  极简 + 全文搜索 Demo: [astro-pure.js.org](https://astro-pure.js.org/) Github: [cworld1/astro-theme-pure](https://github.com/cworld1/astro-theme-pure) > 这位作者的 Github 主页，可以追溯到 1990 年，不知道怎么弄出来的。https://shenjuexiao.com/post/6.htmlSat, 18 Jul 2026 05:02:40 +0000https://shenjuexiao.com/post/5.html# The requested URL returned error: 403 # GitHub Actions push 403 权限拒绝解决 ## 内容来源 - [豆包：GitHub Actions 403](https://www.doubao.com/chat/38434639772911618) - [The requested URL returned error: 403](https://github.com/shenjuexiao/giscafer/issues/1) <img src='https://cdn.jsdelivr.net/gh/shenjuexiao/shenjuexiao@main/Github_png/Giscafer_GitHub_Actions_push_403.png' /> ## fatal: unable to access ```yaml Run yarn sync-post yarn sync-post git add . git commit -m 'chore(ci): blog sync' git push shell: /usr/bin/bash -e {0} env: GH_TOKEN: *** GH_USER: *** GH_PROJECT_NAME: *** yarn run v1.22.22 $ node ./scripts/sync-post.js 文章全部同步成功！ 0 Done in 0.33s. ℹ No staged files found. [main 7226b3a] chore(ci): blog sync 3 files changed, 33 deletions(-) delete mode 100644 data/blog/post-1.mdx delete mode 100644 data/blog/post-2.mdx delete mode 100644 data/blog/post-3.mdx remote: Permission to ***/***.git denied to github-actions[bot]. fatal: unable to access 'https://github.com/***/***/': The requested URL returned error: 403 Error: Process completed with exit code 128. ``` ## 文件：`.github/workflows/sync-post.yml` ```yaml name: Sync Post # Controls when the workflow will run on: # schedule: # - cron: '30 1 * * *' # https://docs.github.com/cn/developers/webhooks-and-events/events/issue-event-types issues: types: - opened - closed - renamed - labeled - unlabeled - reopened - committed # 修改？ workflow_dispatch: env: GH_TOKEN: ${{ secrets.GH_TOKEN }} GH_USER: ${{ secrets.GH_USER }} GH_PROJECT_NAME: ${{ secrets.GH_PROJECT_NAME }} jobs: Publish: runs-on: ubuntu-latest steps: - name: Checkout 🛎️ uses: actions/checkout@v2 - name: Git config 🔧 run: | git config --global user.name 'shenjuexiao' git config --global user.email 'shenjuexiao@qq.com' - name: Display runtime info ✨ run: | echo '当前目录：' pwd - name: Install 🔧 run: yarn install # - name: Build ⛏️ # run: yarn build - name: Update blog files ⛏️ run: | yarn sync-post git add . git commit -m 'chore(ci): blog sync' git push ``` **github-actions[bot]** 没有仓库写入推送权限，触发 403 forbidden。https://shenjuexiao.com/post/5.htmlTue, 14 Jul 2026 15:15:02 +0000https://shenjuexiao.com/post/4.html# Gitblog main.py ## 内容来源 - [Deepseek：#Gitblog main.py 代码解析](https://chat.deepseek.com/share/0k49t8521q3gk2acx9) - [The requested URL returned error: 403](https://github.com/shenjuexiao/giscafer/issues/1) ## main.py ```python # main.py # -*- coding: utf-8 -*- import argparse import os import re import markdown from feedgen.feed import FeedGenerator from github import Github from lxml.etree import CDATA from marko.ext.gfm import gfm as marko MD_HEAD = '''## [Gitblog](https://yihong0618.github.io/gitblog/) My personal blog([About Me](https://github.com/yihong0618/gitblog/issues/282)) using issues and GitHub Actions (随意转载，无需署名) [Things I like](https://github.com/yihong0618/gitblog/issues/311)  [RSS Feed](https://raw.githubusercontent.com/{repo_name}/master/feed.xml) ''' BACKUP_DIR = 'BACKUP' ANCHOR_NUMBER = 5 TOP_ISSUES_LABELS = ['Top'] TODO_ISSUES_LABELS = ['TODO'] FRIENDS_LABELS = ['Friends'] ABOUT_LABELS = ['About'] THINGS_LABELS = ['Things'] IGNORE_LABELS = ( FRIENDS_LABELS + TOP_ISSUES_LABELS + TODO_ISSUES_LABELS + ABOUT_LABELS + THINGS_LABELS ) FRIENDS_TABLE_HEAD = '| Name | Link | Desc | \n | ---- | ---- | ---- |\n' FRIENDS_TABLE_TEMPLATE = '| {name} | {link} | {desc} |\n' FRIENDS_INFO_DICT = { '名字': '', '链接': '', '描述': '', } def get_me(user): return user.get_user().login def is_me(issue, me): return issue.user.login == me def is_hearted_by_me(comment, me): reactions = list(comment.get_reactions()) for r in reactions: if r.content == 'heart' and r.user.login == me: return True return False def _make_friend_table_string(s): info_dict = FRIENDS_INFO_DICT.copy() try: string_list = s.splitlines() # drop empty line string_list = [l for l in string_list if l and not l.isspace()] for l in string_list: string_info_list = re.split('：', l) if len(string_info_list) < 2: continue info_dict[string_info_list[0]] = string_info_list[1] return FRIENDS_TABLE_TEMPLATE.format( name=info_dict['名字'], link=info_dict['链接'], desc=info_dict['描述'] ) except Exception as e: print(str(e)) return # help to covert xml vaild string def _valid_xml_char_ordinal(c): codepoint = ord(c) # conditions ordered by presumed frequency return ( 0x20 <= codepoint <= 0xD7FF or codepoint in (0x9, 0xA, 0xD) or 0xE000 <= codepoint <= 0xFFFD or 0x10000 <= codepoint <= 0x10FFFF ) def format_time(time): return str(time)[:10] def login(token): return Github(token) def get_repo(user: Github, repo: str): return user.get_repo(repo) def parse_TODO(issue): body = issue.body.splitlines() todo_undone = [l for l in body if l.startswith('- [ ] ')] todo_done = [l for l in body if l.startswith('- [x] ')] # just add info all done if not todo_undone: return f'[{issue.title}]({issue.html_url}) all done', [] return ( f'[{issue.title}]({issue.html_url})--{len(todo_undone)} jobs to do--{len(todo_done)} jobs done', todo_done + todo_undone, ) def get_top_issues(repo): return repo.get_issues(labels=TOP_ISSUES_LABELS) def get_todo_issues(repo): return repo.get_issues(labels=TODO_ISSUES_LABELS) def get_repo_labels(repo): return [l for l in repo.get_labels()] def get_issues_from_label(repo, label): return repo.get_issues(labels=(label,)) def add_issue_info(issue, md): time = format_time(issue.created_at) md.write(f'- [{issue.title}]({issue.html_url})--{time}\n') def add_md_todo(repo, md, me): todo_issues = list(get_todo_issues(repo)) if not TODO_ISSUES_LABELS or not todo_issues: return with open(md, 'a+', encoding='utf-8') as md: md.write('## TODO\n') for issue in todo_issues: if is_me(issue, me): todo_title, todo_list = parse_TODO(issue) md.write('TODO list from ' + todo_title + '\n') for t in todo_list: md.write(t + '\n') # new line md.write('\n') def add_md_top(repo, md, me): top_issues = list(get_top_issues(repo)) if not TOP_ISSUES_LABELS or not top_issues: return with open(md, 'a+', encoding='utf-8') as md: md.write('## 置顶文章\n') for issue in top_issues: if is_me(issue, me): add_issue_info(issue, md) def add_md_firends(repo, md, me): s = FRIENDS_TABLE_HEAD friends_issues = list(repo.get_issues(labels=FRIENDS_LABELS)) if not FRIENDS_LABELS or not friends_issues: return friends_issue_number = friends_issues[0].number for issue in friends_issues: for comment in issue.get_comments(): if is_hearted_by_me(comment, me): try: s += _make_friend_table_string(comment.body or '') except Exception as e: print(str(e)) pass s = markdown.markdown(s, output_format='html', extensions=['extra']) with open(md, 'a+', encoding='utf-8') as md: md.write( f'## [友情链接](https://github.com/{str(me)}/gitblog/issues/{friends_issue_number})\n' ) md.write('<details><summary>显示</summary>\n') md.write(s) md.write('</details>\n') md.write('\n\n') def add_md_recent(repo, md, me, limit=5): count = 0 with open(md, 'a+', encoding='utf-8') as md: # one the issue that only one issue and delete (pyGitHub raise an exception) try: md.write('## 最近更新\n') for issue in repo.get_issues(sort='created', direction='desc'): if is_me(issue, me): add_issue_info(issue, md) count += 1 if count >= limit: break except Exception as e: print(str(e)) def add_md_header(md, repo_name): with open(md, 'w', encoding='utf-8') as md: md.write(MD_HEAD.format(repo_name=repo_name)) md.write('\n') def add_md_label(repo, md, me): labels = get_repo_labels(repo) # sort lables by description info if it exists, otherwise sort by name, # for example, we can let the description start with a number (1#Java, 2#Docker, 3#K8s, etc.) labels = sorted( labels, key=lambda x: ( x.description is None, x.description == '', x.description, x.name, ), ) with open(md, 'a+', encoding='utf-8') as md: for label in labels: # we don't need add top label again if label.name in IGNORE_LABELS: continue issues = get_issues_from_label(repo, label) issues = list(sorted(issues, key=lambda x: x.created_at, reverse=True)) if len(issues) != 0: md.write('## ' + label.name + '\n\n') i = 0 for issue in issues: if not issue: continue if is_me(issue, me): if i == ANCHOR_NUMBER: md.write('<details><summary>显示更多</summary>\n') md.write('\n') add_issue_info(issue, md) i += 1 if i > ANCHOR_NUMBER: md.write('</details>\n') md.write('\n') def get_to_generate_issues(repo, dir_name, issue_number=None): md_files = os.listdir(dir_name) generated_issues_numbers = [ int(i.split('_')[0]) for i in md_files if i.split('_')[0].isdigit() ] to_generate_issues = [ i for i in list(repo.get_issues()) if int(i.number) not in generated_issues_numbers ] if issue_number: to_generate_issues.append(repo.get_issue(int(issue_number))) return to_generate_issues def generate_rss_feed(repo, filename, me): generator = FeedGenerator() generator.id(repo.html_url) generator.title(f'RSS feed of {repo.owner.login}'s {repo.name}') generator.author( {'name': os.getenv('GITHUB_NAME'), 'email': os.getenv('GITHUB_EMAIL')} ) generator.link(href=repo.html_url) generator.link( href=f'https://raw.githubusercontent.com/{repo.full_name}/master/{filename}', rel='self', ) for issue in repo.get_issues(): if not issue.body or not is_me(issue, me) or issue.pull_request: continue item = generator.add_entry(order='append') item.id(issue.html_url) item.link(href=issue.html_url) item.title(issue.title) item.published(issue.created_at.strftime('%Y-%m-%dT%H:%M:%SZ')) for label in issue.labels: item.category({'term': label.name}) body = ''.join(c for c in issue.body if _valid_xml_char_ordinal(c)) item.content(CDATA(marko.convert(body)), type='html') generator.atom_file(filename) def main(token, repo_name, issue_number=None, dir_name=BACKUP_DIR): user = login(token) me = get_me(user) repo = get_repo(user, repo_name) # add to readme one by one, change order here add_md_header('README.md', repo_name) for func in [add_md_firends, add_md_top, add_md_recent, add_md_label, add_md_todo]: func(repo, 'README.md', me) generate_rss_feed(repo, 'feed.xml', me) to_generate_issues = get_to_generate_issues(repo, dir_name, issue_number) # save md files to backup folder for issue in to_generate_issues: save_issue(issue, me, dir_name) def save_issue(issue, me, dir_name=BACKUP_DIR): md_name = os.path.join( dir_name, f'{issue.number}_{issue.title.replace('/', '-').replace(' ', '.')}.md' ) with open(md_name, 'w') as f: f.write(f'# [{issue.title}]({issue.html_url})\n\n') f.write(issue.body or '') if issue.comments: for c in issue.get_comments(): if is_me(c, me): f.write('\n\n---\n\n') f.write(c.body or '') if __name__ == '__main__': if not os.path.exists(BACKUP_DIR): os.mkdir(BACKUP_DIR) parser = argparse.ArgumentParser() parser.add_argument('github_token', help='github_token') parser.add_argument('repo_name', help='repo_name') parser.add_argument( '--issue_number', help='issue_number', default=None, required=False ) options = parser.parse_args() main(options.github_token, options.repo_name, options.issue_number) ``` 这是一个使用GitHub Issues作为博客系统的Python脚本，名为Gitblog。https://shenjuexiao.com/post/4.htmlTue, 14 Jul 2026 15:13:57 +0000https://shenjuexiao.com/post/3.html# Gmeek 部署 ## 内容来源 - [Gmeek快速上手](https://blog.meekdai.com/post/Gmeek-kuai-su-shang-shou.html) - [Gmeek 部署](https://github.com/gaishanke/gaishanke.github.io/issues/3) ## Gmeek 博客框架 - 超轻量级个人博客模板 - 完全基于 `Github Pages` 、 `Github Issues` 和 `Github Actions`， - 不需要本地部署 ## 一、资源 ## Gmeek - [Gmeek快速上手](https://blog.meekdai.com/post/Gmeek-kuai-su-shang-shou.html) - [Gmeek Github 源码](https://github.com/Meekdai/Gmeek) ### Markdown 语法 - [Basic writing and formatting syntax](https://docs.github.com/en/get-started/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax) - [基本撰写和格式语法](https://docs.github.com/zh/get-started/writing-on-github/getting-started-with-writing-and-formatting-on-github/basic-writing-and-formatting-syntax) ### 在线编辑工具 - [markdown.com.cn](https://markdown.com.cn/editor/) ## 二、安装 ### 创建仓库 - 点击[通过模板创建仓库](https://github.com/new?template_name=Gmeek-template&template_owner=Meekdai)，仓库名称：`<用户名>.github.io` ### 启用 Pages - 仓库页面导航：`Settings` -> `Pages` -> `Build and deployment` - `Source` 下拉选择 `Github Actions` <img src='https://cdn.jsdelivr.net/gh/shenjuexiao/shenjuexiao@main/Github_png/Gmeek_Settings_Github-Pages_Github-Actions.png' /> ### 开始写作 - 仓库页面导航：`Issues` -> `New issue` - `Add a title`（写标题），`Add a description`（写描述）。https://shenjuexiao.com/post/3.htmlTue, 14 Jul 2026 11:44:53 +0000https://shenjuexiao.com/post/2.html# Github Issues 博客项目（不完全统计） ## 内容来源 - [Github Issues 搭建的博客项目列表](https://github.com/gaishanke/gaiashanke.github.io/issues/2) ## @rena.to/github-blog - **技术栈**：TypeScript（核心API库） - **项目链接**：[https://www.npmjs.com/package/@rena.to/github-blog](https://www.npmjs.com/package/@rena.to/github-blog) ## AllanChain/blog - **技术栈**：Astro - **项目链接**：[https://github.com/AllanChain/blog](https://github.com/AllanChain/blog) ## babilema - **技术栈**：Go（静态博客生成器） - **项目链接**：[https://pkg.go.dev/github.com/ByteBakersCo/babilema](https://pkg.go.dev/github.com/ByteBakersCo/babilema) ## ginb - **技术栈**：Next.js, Tailwind CSS, React - **项目链接**：[https://github.com/ChrAlpha/ginb](https://github.com/ChrAlpha/ginb) ## gitblog - **技术栈**：Python - **项目链接**：[https://github.com/yihong0618/gitblog](https://github.com/yihong0618/gitblog) ## gitblog - **技术栈**：JavaScript - **项目链接**：[https://github.com/imuncle/gitblog](https://github.com/imuncle/gitblog) ## Gmeek - **技术栈**：GitHub Pages, GitHub Actions, jinja2, primer.style - **项目链接**：[https://github.com/Ambop/Gmeek](https://github.com/Ambop/Gmeek) ## Issue-Blog-With-Github-Action - **技术栈**：Vuepress - **项目链接**：[https://github.com/Yidadaa/Issue-Blog-With-Github-Action](https://github.com/Yidadaa/Issue-Blog-With-Github-Action) ## qiwihui/blog - **技术栈**：CSS - **项目链接**：[https://github.com/qiwihui/blog](https://github.com/qiwihui/blog) ## Sep0lkit/git-issues-blog - **技术栈**：Python - **项目链接**：[https://github.com/Sep0lkit/git-issues-blog](https://github.com/Sep0lkit/git-issues-blog) ## ttop5/issue-blog - **技术栈**：Vue - **项目链接**：[https://github.com/ttop5/issue-blog](https://github.com/ttop5/issue-blog) ## whyour/issue-blog - **技术栈**：TypeScript（VS Code插件） - **项目链接**：[https://github.com/whyour/issue-blog](https://github.com/whyour/issue-blog) ## xizhibei/blog - **技术栈**：Node.js, Golang - **项目链接**：[https://github.com/xizhibei/blog](https://github.com/xizhibei/blog)。https://shenjuexiao.com/post/2.htmlTue, 14 Jul 2026 11:44:22 +0000https://shenjuexiao.com/post/1.html# Gitblog 用过 Jekyll、Hexo，捣鼓了一阵子，都没能长期使用持续写作，现在转头使用 GIthub Issue 搭建的博客系统。https://shenjuexiao.com/post/1.htmlTue, 14 Jul 2026 11:43:56 +0000