bruno
ff99b8eb2b
📚 论文级功能实现: - ✅ KaTeX数学公式渲染系统 (内联$...1411...1411) - ✅ 高级代码高亮 (行号、标题、语法着色) - ✅ PWA渐进式Web应用支持 - ✅ 多语言代码支持 (20+编程语言) - ✅ 代码行标注系统 (错误/成功/警告/高亮) 🧮 数学公式系统: - LaTeX语法完整支持 - 数学宏定义 (\RR, \NN, \ZZ, \QQ, \CC等) - 复杂公式渲染 (矩阵、积分、求和、分数) - 响应式数学表达式 - 黑暗模式数学公式适配 💻 高级代码展示: - 代码块标题显示 - 行号自动生成 - 语法高亮增强 - 多行高亮标注 - 错误/成功/警告行标记 - 20+编程语言支持 🔧 企业级插件: - @docusaurus/plugin-pwa (离线支持) - rehype-katex (数学公式) - remark-math (数学语法) - remark-gfm (GitHub风格Markdown) - prism-react-renderer (代码高亮) 📊 算法示例升级: - Transformer深度学习模型 - 多头注意力机制实现 - 完整的神经网络架构 - 专业文档注释 - 类型提示和错误处理 🎨 视觉优化: - 数学公式背景框 - 代码行高亮动画 - 专业配色方案 - 渐变边框效果 - 响应式设计适配 📱 PWA功能: - 离线访问支持 - 应用图标配置 - 清单文件完整 - 主题色统一 - 安装提示优化 🚀 性能优化: - 图片质量提升至85% - 代码分割优化 - 字体预加载 - CSS变量系统 - GPU加速渲染 ✨ 用户体验: - 论文级排版标准 - 学术引用格式 - 专业视觉层次 - 无障碍访问支持 - 多设备完美适配
sourcemap-codec
Encode/decode the mappings property of a sourcemap.
Why?
Sourcemaps are difficult to generate and manipulate, because the mappings property – the part that actually links the generated code back to the original source – is encoded using an obscure method called Variable-length quantity. On top of that, each segment in the mapping contains offsets rather than absolute indices, which means that you can't look at a segment in isolation – you have to understand the whole sourcemap.
This package makes the process slightly easier.
Installation
npm install sourcemap-codec
Usage
import { encode, decode } from 'sourcemap-codec';
var decoded = decode( ';EAEEA,EAAE,EAAC,CAAE;ECQY,UACC' );
assert.deepEqual( decoded, [
// the first line (of the generated code) has no mappings,
// as shown by the starting semi-colon (which separates lines)
[],
// the second line contains four (comma-separated) segments
[
// segments are encoded as you'd expect:
// [ generatedCodeColumn, sourceIndex, sourceCodeLine, sourceCodeColumn, nameIndex ]
// i.e. the first segment begins at column 2, and maps back to the second column
// of the second line (both zero-based) of the 0th source, and uses the 0th
// name in the `map.names` array
[ 2, 0, 2, 2, 0 ],
// the remaining segments are 4-length rather than 5-length,
// because they don't map a name
[ 4, 0, 2, 4 ],
[ 6, 0, 2, 5 ],
[ 7, 0, 2, 7 ]
],
// the final line contains two segments
[
[ 2, 1, 10, 19 ],
[ 12, 1, 11, 20 ]
]
]);
var encoded = encode( decoded );
assert.equal( encoded, ';EAEEA,EAAE,EAAC,CAAE;ECQY,UACC' );
License
MIT