离线 · 无上传 · 浏览器本地完成Offline · no uploads · runs in the browser
Excel 转 Markdown 表格——离线、多工作表、不上传Excel to Markdown table offline, multi-sheet, no upload
把 Excel 贴进 README、Issue 或文档里，需要的是一张 Markdown 管道表。这条链路先把工作簿写成 HTML 表格（SheetJS 的 `sheet_to_html`，日期格在此之前已被改成 ISO 文本），再交给 Markdown 的表格规则输出。多工作表在这里是**支持**的：每个表前面会写一行 `## 工作表名`，各成一表——终点是 Markdown，不经过「只接受数组」那一闸，所以不会像 HTML 转 CSV 那样在多张表上失败。代价也说清楚：管道表没有跨列，合并区被覆盖的格子根本不产生单元格，右边整排会左移一格。Putting a spreadsheet into a README, an issue or a document means a Markdown pipe table. This route first writes the workbook as HTML tables (SheetJS’s `sheet_to_html`, with date cells already turned into ISO text) and then lets the Markdown table rule emit them. Multiple worksheets are **supported** here: each one gets a `## Sheet name` line above its table, because the endpoint is Markdown and never passes the “arrays only” gate that makes HTML to CSV fail on several tables. The cost is stated plainly: a pipe table has no column spanning, and the cells a merge covers emit nothing, so everything right of a merge shifts one column left.
这条链路怎么走How the route runs
Excel (.xlsx)Excel (.xlsx) HTMLHTML MarkdownMarkdown
共 2 步，由扩展的路径搜索自动选出，中间产物不落盘。2 steps, chosen by the extension's own path search; intermediate results never touch the disk.
会保留什么What carries over
- 每个工作表一张管道表，多 sheet 时前面各写一行 `## 工作表名`——多张表在这条路上是正常输出，不是错误One pipe table per worksheet, each preceded by a `## Sheet name` line when the workbook holds several — multiple tables are a normal result on this route, not an error
- 日期按 ISO 文本落地（`2024-01-05`，带时间的是 `2024-01-05 14:30:00`），不会被改回序列号；作者显式设定的显示格式照原样保留Dates land as ISO text (`2024-01-05`, or `2024-01-05 14:30:00` with a time) instead of being rewritten into serials, and a format the workbook author set explicitly is respected rather than overridden
- 单元格取的是工作簿里的值与显示文本，所以公式那格给的是算好的数，不是 `=SUM(...)`Cells carry the sheet’s value and display text, so a formula cell gives the computed number rather than `=SUM(...)`
- 首行后紧跟 `---` 分隔线，列数按最宽的一行补齐；格子里的竖线写成 `\|`、反斜杠成对写，内容撑不破表The first row is followed by the `---` rule and rows are padded to the widest one; a pipe in a cell becomes `\|` and a backslash is doubled, so content cannot break the table
如实说明的限制What it does not do
- 合并单元格左移：HTML 那一步用 colspan / rowspan 表达合并，但纯文本没有跨列概念，被覆盖的格子一个都不产生，于是合并区右边的列整体左移一格——要列位置稳定，先取消合并再转Merged ranges shift columns: the HTML step expresses them as colspan / rowspan, but plain text has no spanning and the covered cells emit nothing, so the columns right of a merge shift one place left — unmerge first for stable positions
- 图片、图表、批注、条件格式与样式都不在 SheetJS 读出的单元格里：它们不是「在这里被丢掉」，而是从来没进过这一步Images, charts, comments, conditional formatting and styling are not in the cell grid SheetJS reads: they are not lost at this step, they never entered it
- 格子里的换行与连续空白折成单空格，行内格式也不保留——管道表一行就是一行Newlines and runs of whitespace inside a cell collapse to one space and inline formatting is not carried over — one pipe-table row is one line
- 表头行里重复或为空的表头会互相覆盖（同名的后一格赢），首行没写到的列不会存在A repeated or empty entry in the header row overwrites its twin (the later cell wins), and a column the first row does not name will not exist
- 一张表最多读 50 万格、整个工作簿最多 100 万格，超过直接报错，不会给一张截断的表At most 500,000 cells per sheet and 1,000,000 per workbook are read; past that it fails rather than emitting a truncated table
实操提示Working with it
- 一批最多 200 个文件，单个文件上限 100 MB；多个工作簿逐个产出各自的 .md，想一次拿走就打成一个 ZIPUp to 200 files per batch and 100 MB per file; several workbooks yield one .md each, and the set can be taken as a single ZIP
- 要的是能排版的成品文档而不是表格，走 Excel 转 Word；只要第一个工作表的纯文本，走 Excel 转文本For a document that keeps its layout, take Excel to Word; for the first sheet as plain text, take Excel to Text
- 想让数据进 CSV 或 JSON：Excel 转 CSV 会把每个工作表写成一份 CSV 打包 ZIP，Excel 转 JSON 则一个文件按工作表名分组——那两条都不经过 HTML 提取，所以也不会撞上「多张表」那道闸For the data as CSV or JSON: Excel to CSV writes one CSV per worksheet inside a ZIP, and Excel to JSON returns one file grouped by sheet name — neither goes through HTML extraction, so neither meets the “several tables” gate either
常见问题Questions people actually ask
工作簿里有五个工作表，会不会像 HTML 转 CSV 那样失败？Five worksheets in the workbook — will it fail like HTML to CSV does?
不会。那条失败是因为多张表会被并成一个分组对象，而 CSV 那一步只接受数组；这条的终点是 Markdown，每张表各成一段并冠一行小标题，没有那道闸。No. That route fails because several tables merge into one grouped object and the CSV stage accepts only arrays. This endpoint is Markdown: each table becomes its own section under a heading line, and there is no such gate.
合并过的单元格会变成什么？What does a merged cell become?
合并区只有左上那一格有内容，其余格子根本不产出——所以合并区右边的列会整体左移一格，Markdown 管道表也没有跨列的写法。想让列位置不变，先在源文件里取消合并。Only the top-left cell of a merged range carries content and the covered cells emit nothing, so the columns to the right shift one place left — and a pipe table has no syntax for spanning anyway. Unmerge in the source file to keep column positions.
公式那一格会给公式还是结果？Does a formula cell give the formula or the result?
给结果：工作簿存下的显示文本 / 计算值。公式本身不会进文本类产物，需要公式就保留那份 .xlsx。The result — the display text or value the workbook stores. Formulas do not travel into text artefacts, so keep the .xlsx if you need them.
含个人信息的表会在本机以外处理吗？Is a sheet containing personal data processed anywhere off this machine?
不会。读工作簿、生成表格、写 Markdown 都在扩展自己的标签页里完成，只申请 storage 权限、没有 host 权限，全程零网络请求。No. Reading the workbook, building the tables and writing the Markdown all happen inside the extension’s own tab: the storage permission only, no host permissions, and no requests at any point.
不需要账号，也不需要联网：装好扩展后点图标，在新标签页里打开工作台，文件从磁盘直接选。No account and no network: install the extension, click its icon, and the workbench opens in a new tab reading straight off your disk.
从 Chrome 应用商店安装Install from the Chrome Web Store 安装步骤Install steps GitHub 源码Source on GitHub