render-markdown.nvim

render-markdown.nvim

优化Neovim中Markdown文件的渲染效果

render-markdown.nvim插件通过增强Markdown文件的渲染效果,改善了Neovim中的文档浏览体验。该插件完全集成于Neovim环境,无需额外窗口,可在不同模式间切换渲染视图。它支持渲染多种Markdown元素,如标题、代码块、列表和表格,并能处理LaTeX公式。插件还提供自定义处理程序和反隐藏功能,满足个性化需求。render-markdown.nvim致力于为Neovim用户提供更直观、美观的Markdown文件呈现方式。

NeovimMarkdown渲染插件开发Lua配置Github开源项目

render-markdown.nvim

Plugin to improve viewing Markdown files in Neovim

[!NOTE]

Repo has been renamed from markdown.nvim to make things more consistent.

Github will redirect all requests so nothing should break for most users.

For lazy.nvim & tokyonight.nvim users changing the plugin name will remove custom highlights until the following is merged #620.

In the meantime you can fix this by adding the following to your config:

require('tokyonight').setup({ plugins = { markdown = true }, })

For rocks.nvim users migrate to the new LuaRock.

:Rocks prune markdown.nvim :Rocks install render-markdown.nvim
<!-- panvimdoc-ignore-start -->
HeadingTable
QuoteLaTeX
Callout
<!-- panvimdoc-ignore-end -->

Features

  • Functions entirely inside of Neovim with no external windows
  • Changes between rendered view in normal mode and raw view in all other modes
  • Supports anti-conceal behavior, removing any virtual text added by this plugin on the line the cursor is on, this does have a performance penalty and can be disabled
  • Changes window options between rendered and raw view based on configuration
    • Effects conceallevel & concealcursor by default
  • Supports rendering markdown injected into other file types
  • Renders the following markdown components:
    • Headings: highlight depending on level and replaces # with icon
    • Horizontal breaks: replace with full-width lines
    • Code blocks: highlight to better stand out
      • Adds language icon 1, requires icon provider (mini.icons or nvim-web-devicons)
      • Left pad lines within block 1
    • Inline code: highlight to better stand out
    • List bullet points: replace with provided icon based on level
    • Checkboxes: replace with provided icon based on whether they are checked
    • Block quotes: replace leading > with provided icon
    • Tables: replace border characters, handles misaligned tables but does NOT align according to delimiter indicator
    • Callouts
      • Github & Obsidian out of the box, supports user defined as well
    • Custom checkbox states 1, function similar to callouts
    • Adds icon before images / links 1
    • LaTeX blocks: renders formulas if latex parser and pylatexenc are installed
  • Disable rendering when file is larger than provided value
  • Support custom handlers which are ran identically to builtin handlers

Requirements

  • neovim >= 0.9.0 (minimum) >= 0.10.0 (recommended)
  • treesitter parsers:
  • Icon provider plugin (Optional): Used for icon above code blocks
  • System dependencies:
    • pylatexenc (Optional): Used to transform LaTeX strings to appropriate unicode using latex2text

Install

lazy.nvim

{ 'MeanderingProgrammer/render-markdown.nvim', opts = {}, dependencies = { 'nvim-treesitter/nvim-treesitter', 'echasnovski/mini.nvim' }, -- if you use the mini.nvim suite -- dependencies = { 'nvim-treesitter/nvim-treesitter', 'echasnovski/mini.icons' }, -- if you use standalone mini plugins -- dependencies = { 'nvim-treesitter/nvim-treesitter', 'nvim-tree/nvim-web-devicons' }, -- if you prefer nvim-web-devicons }

rocks.nvim

This plugin is available on LuaRocks

:Rocks install render-markdown.nvim

packer.nvim

use({ 'MeanderingProgrammer/render-markdown.nvim', after = { 'nvim-treesitter' }, requires = { 'echasnovski/mini.nvim', opt = true }, -- if you use the mini.nvim suite -- requires = { 'echasnovski/mini.icons', opt = true }, -- if you use standalone mini plugins -- requires = { 'nvim-tree/nvim-web-devicons', opt = true }, -- if you prefer nvim-web-devicons config = function() require('render-markdown').setup({}) end, })

Commands

  • :RenderMarkdown | :RenderMarkdown enable - Enable this plugin
    • Can also be accessed directly through require('render-markdown').enable()
  • :RenderMarkdown disable - Disable this plugin
    • Can also be accessed directly through require('render-markdown').disable()
  • :RenderMarkdown toggle - Switch between enabling & disabling this plugin
    • Can also be accessed directly through require('render-markdown').toggle()
  • :RenderMarkdown expand - Increase anti-conceal margin above and below by 1
    • Can also be accessed directly through require('render-markdown').expand()
  • :RenderMarkdown contract - Decrease anti-conceal margin above and below by 1
    • Can also be accessed directly through require('render-markdown').contract()

Setup

Checkout the Wiki for examples and images associated with different configuration options.

The full default configuration is provided below for reference.

Any part of it can be modified however for many fields this does not make much sense.

Some of the more useful fields are discussed further down.

<details> <summary>Full Default Configuration</summary>
require('render-markdown').setup({ -- Whether Markdown should be rendered by default or not enabled = true, -- Maximum file size (in MB) that this plugin will attempt to render -- Any file larger than this will effectively be ignored max_file_size = 10.0, -- Milliseconds that must pass before updating marks, updates occur -- within the context of the visible window, not the entire buffer debounce = 100, -- Pre configured settings that will attempt to mimic various target -- user experiences. Any user provided settings will take precedence. -- obsidian: mimic Obsidian UI -- lazy: will attempt to stay up to date with LazyVim configuration -- none: does nothing preset = 'none', -- Capture groups that get pulled from markdown markdown_query = [[ (atx_heading [ (atx_h1_marker) (atx_h2_marker) (atx_h3_marker) (atx_h4_marker) (atx_h5_marker) (atx_h6_marker) ] @heading) (thematic_break) @dash (fenced_code_block) @code [ (list_marker_plus) (list_marker_minus) (list_marker_star) ] @list_marker (task_list_marker_unchecked) @checkbox_unchecked (task_list_marker_checked) @checkbox_checked (block_quote) @quote (pipe_table) @table ]], -- Capture groups that get pulled from quote nodes markdown_quote_query = [[ [ (block_quote_marker) (block_continuation) ] @quote_marker ]], -- Capture groups that get pulled from inline markdown inline_query = [[ (code_span) @code (shortcut_link) @shortcut [(inline_link) (full_reference_link) (image)] @link ]], -- The level of logs to write to file: vim.fn.stdpath('state') .. '/render-markdown.log' -- Only intended to be used for plugin development / debugging log_level = 'error', -- Filetypes this plugin will run on file_types = { 'markdown' }, -- Vim modes that will show a rendered view of the markdown file -- All other modes will be uneffected by this plugin render_modes = { 'n', 'c' }, -- Set to avoid seeing warnings for conflicts in health check acknowledge_conflicts = false, anti_conceal = { -- This enables hiding any added text on the line the cursor is on enabled = true, -- Number of lines above cursor to show above = 0, -- Number of lines below cursor to show below = 0, }, latex = { -- Whether LaTeX should be rendered, mainly used for health check enabled = true, -- Executable used to convert latex formula to rendered unicode converter = 'latex2text', -- Highlight for LaTeX blocks highlight = 'RenderMarkdownMath', -- Amount of empty lines above LaTeX blocks top_pad = 0, -- Amount of empty lines below LaTeX blocks bottom_pad = 0, }, heading = { -- Turn on / off heading icon & background rendering enabled = true, -- Turn on / off any sign column related rendering sign = true, -- Determines how the icon fills the available space: -- inline: underlying '#'s are concealed resulting in a left aligned icon -- overlay: result is left padded with spaces to hide any additional '#' position = 'overlay', -- Replaces '#+' of 'atx_h._marker' -- The number of '#' in the heading determines the 'level' -- The 'level' is used to index into the array using a cycle icons = { '󰲡 ', '󰲣 ', '󰲥 ', '󰲧 ', '󰲩 ', '󰲫 ' }, -- Added to the sign column if enabled -- The 'level' is used to index into the array using a cycle signs = { '󰫎 ' }, -- Width of the heading background: -- block: width of the heading text -- full: full width of the window -- Can also be an array of the above values in which case the 'level' is used -- to index into the array using a clamp width = 'full', -- Amount of padding to add to the left of headings left_pad = 0, -- Amount of padding to add to the right of headings when width is 'block' right_pad = 0, -- Minimum width to use for headings when width is 'block' min_width = 0, -- Determins if a border is added above and below headings border = false, -- Highlight the start of the border using the foreground highlight border_prefix = false, -- Used above heading for border above = '▄', -- Used below heading for border below = '▀', -- The 'level' is used to index into the array using a clamp -- Highlight for the heading icon and extends through the entire line backgrounds = { 'RenderMarkdownH1Bg', 'RenderMarkdownH2Bg', 'RenderMarkdownH3Bg', 'RenderMarkdownH4Bg', 'RenderMarkdownH5Bg', 'RenderMarkdownH6Bg', }, -- The 'level' is used to index into the array using a clamp -- Highlight for the heading and sign icons foregrounds = { 'RenderMarkdownH1', 'RenderMarkdownH2', 'RenderMarkdownH3', 'RenderMarkdownH4', 'RenderMarkdownH5', 'RenderMarkdownH6', }, }, code = { -- Turn on / off code block & inline code rendering enabled = true, -- Turn on / off any sign column related rendering sign = true, -- Determines how code blocks & inline code are rendered: -- none: disables all rendering -- normal: adds highlight group to code blocks & inline code, adds padding to code blocks -- language: adds language icon to sign column if enabled and icon + name above code blocks -- full: normal + language style = 'full', -- Determines where language icon is rendered: -- right: right side of code block -- left: left side of code block position = 'left', -- An array of language names for which background highlighting will be disabled -- Likely because that language has background highlights itself disable_background = { 'diff' }, -- Width of the code block background: -- block: width of the code block -- full: full width of the window width = 'full', -- Amount of padding to add to the left of code blocks left_pad = 0, -- Amount of padding to add to the right of code blocks when width is 'block' right_pad = 0, -- Minimum width to use for code blocks when width is 'block' min_width = 0, -- Determins how the top / bottom of code block are rendered: -- thick: use the same highlight as the code body -- thin: when lines are empty overlay the above & below icons border = 'thin', -- Used above code blocks for thin border above = '▄', -- Used below code blocks for thin border below = '▀', -- Highlight for code blocks highlight = 'RenderMarkdownCode', -- Highlight for inline code highlight_inline = 'RenderMarkdownCodeInline', }, dash = { -- Turn on / off thematic break rendering enabled = true, -- Replaces '---'|'***'|'___'|'* * *' of 'thematic_break' -- The icon gets repeated across the window's width icon = '─', -- Width of the generated line: -- <integer>: a hard coded width value -- full: full width of the window width = 'full', -- Highlight for the whole line generated from the icon highlight = 'RenderMarkdownDash', }, bullet = { -- Turn on / off list bullet rendering enabled = true, -- Replaces '-'|'+'|'*' of 'list_item' -- How deeply nested the list is determines the 'level' -- The 'level' is used to index into the array using a cycle -- If the item is a 'checkbox' a conceal is used to hide the bullet instead icons = { '●', '○', '◆', '◇' }, -- Padding to add to the left of bullet point left_pad = 0, -- Padding to add to the right of bullet point right_pad = 0, -- Highlight for the bullet icon highlight = 'RenderMarkdownBullet', }, -- Checkboxes are a special instance of a 'list_item' that start with a 'shortcut_link' -- There are two special states for unchecked & checked defined in the markdown grammar checkbox = { -- Turn on / off checkbox state rendering enabled = true, unchecked = { -- Replaces '[ ]' of 'task_list_marker_unchecked' icon = '󰄱 ', -- Highlight for the unchecked icon highlight = 'RenderMarkdownUnchecked', }, checked = { -- Replaces '[x]' of 'task_list_marker_checked' icon = '󰱒 ', -- Highligh for the checked icon highlight = 'RenderMarkdownChecked', }, -- Define custom checkbox states, more involved as they are not part of the markdown grammar -- As a result this requires neovim >= 0.10.0 since it relies on 'inline' extmarks -- Can specify as many additional states as you like following the 'todo' pattern below -- The key in this case 'todo' is for healthcheck and to allow users to change its values -- 'raw': Matched against the raw text of a 'shortcut_link' -- 'rendered': Replaces the 'raw' value when rendering -- 'highlight': Highlight for the 'rendered' icon custom = { todo = { raw = '[-]', rendered = '󰥔 ', highlight = 'RenderMarkdownTodo' }, }, }, quote = { -- Turn on / off block quote & callout rendering

Footnotes

  1. Requires neovim >= 0.10.0 2 3 4

编辑推荐精选

MarkItDown

MarkItDown

一个支持多种格式转换的工具库

MarkItDown 是一个强大的 Python 工具库,专注于文档格式转换。它能够处理多种类型的文件,如 HTML、Wikipedia 页面以及 Bing 搜索结果页等,将其转换为 Markdown 格式。该项目支持插件扩展,提供了清晰的接口和丰富的功能,为开发者和文档处理人员提供了便捷、高效的文档转换解决方案,能有效提升文档处理效率,是文档转换领域的优秀选择。

Trae

Trae

字节跳动发布的AI编程神器IDE

Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。

AI工具TraeAI IDE协作生产力转型热门
OmniParser

OmniParser

帮助AI理解电脑屏幕 纯视觉GUI元素的自动化解析方案

开源工具通过计算机视觉技术实现图形界面元素的智能识别与结构化处理,支持自动化测试脚本生成和辅助功能开发。项目采用模块化设计,提供API接口与多种输出格式,适用于跨平台应用场景。核心算法优化了元素定位精度,在动态界面和复杂布局场景下保持稳定解析能力。

OmniParser界面解析交互区域检测Github开源项目
Grok3

Grok3

埃隆·马斯克旗下的人工智能公司 xAI 推出的第三代大规模语言模型

Grok3 是由埃隆·马斯克旗下的人工智能公司 xAI 推出的第三代大规模语言模型,常被马斯克称为“地球上最聪明的 AI”。它不仅是在前代产品 Grok 1 和 Grok 2 基础上的一次飞跃,还在多个关键技术上实现了创新突破。

腾讯元宝

腾讯元宝

腾讯自研的混元大模型AI助手

腾讯元宝是腾讯基于自研的混元大模型推出的一款多功能AI应用,旨在通过人工智能技术提升用户在写作、绘画、翻译、编程、搜索、阅读总结等多个领域的工作与生活效率。

AI助手AI对话AI工具腾讯元宝智能体热门 AI 办公助手
Windsurf Wave 3

Windsurf Wave 3

Windsurf Editor推出第三次重大更新Wave 3

新增模型上下文协议支持与智能编辑功能。本次更新包含五项核心改进:支持接入MCP协议扩展工具生态,Tab键智能跳转提升编码效率,Turbo模式实现自动化终端操作,图片拖拽功能优化多模态交互,以及面向付费用户的个性化图标定制。系统同步集成DeepSeek、Gemini等新模型,并通过信用点数机制实现差异化的资源调配。

AI IDE
Cursor

Cursor

增强编程效率的AI代码编辑器

Cursor作为AI驱动的代码编辑工具,助力开发者效率大幅度提升。该工具简化了扩展、主题和键位配置的导入,可靠的隐私保护措施保证代码安全,深受全球开发者信赖。此外,Cursor持续推出更新,不断优化功能和用户体验。

AI开发辅助编程AI工具CursorAI代码编辑器
Manus

Manus

全面超越基准的 AI Agent助手

Manus 是一款通用人工智能代理平台,能够将您的创意和想法迅速转化为实际成果。无论是定制旅行规划、深入的数据分析,还是教育支持与商业决策,Manus 都能高效整合信息,提供精准解决方案。它以直观的交互体验和领先的技术,为用户开启了一个智慧驱动、轻松高效的新时代,让每个灵感都能得到完美落地。

飞书知识问答

飞书知识问答

飞书官方推出的AI知识库 上传word pdf即可部署AI私有知识库

基于DeepSeek R1大模型构建的知识管理系统,支持PDF、Word、PPT等常见文档格式解析,实现云端与本地数据的双向同步。系统具备实时网络检索能力,可自动关联外部信息源,通过语义理解技术处理结构化与非结构化数据。免费版本提供基础知识库搭建功能,适用于企业文档管理和个人学习资料整理场景。

酷表ChatExcel

酷表ChatExcel

大模型驱动的Excel数据处理工具

基于大模型交互的表格处理系统,允许用户通过对话方式完成数据整理和可视化分析。系统采用机器学习算法解析用户指令,自动执行排序、公式计算和数据透视等操作,支持多种文件格式导入导出。数据处理响应速度保持在0.8秒以内,支持超过100万行数据的即时分析。

AI工具酷表ChatExcelAI智能客服AI营销产品使用教程
下拉加载更多