@peaceroad/markown-figure-num-setting

markown-figure-num-setting

Set sequential numbers for Markdown caption labels (and optionally image alt text).

Caption label detection is based on p7d-markdown-it-p-captions language regex rules.

Install

npm i @peaceroad/markown-figure-num-setting

Usage

import setMarkdownFigureNum from '@peaceroad/markown-figure-num-setting'

// The first argument must be Markdown content text (not a file path).
const outputMarkdown = setMarkdownFigureNum(markdownText, option)

If you have a file path, read the file content first:

import fs from 'node:fs'
import setMarkdownFigureNum from '@peaceroad/markown-figure-num-setting'

const markdownText = fs.readFileSync(markdownPath, 'utf8')
const outputMarkdown = setMarkdownFigureNum(markdownText)

Basic behavior

Input:

Paragraph.

Figure. A caption

![A alt text.](image.jpg)

Paragraph.

Figure. A caption

![A alt text.](image.jpg)

Output (default option):

Paragraph.

Figure 1. A caption

![A alt text.](image.jpg)

Paragraph.

Figure 2. A caption

![A alt text.](image.jpg)

Options

Default options:

{
  img: true,
  video: false,
  table: false,
  'pre-code': false,
  'pre-samp': false,
  blockquote: false,
  slide: false,
  audio: false,
  labelMarkMap: null,
  noSetAlt: true,
  setNumberAlt: false,
  setImgAlt: false,
}

• img, video, table, pre-code, pre-samp, blockquote, slide, audio: Enable numbering for each caption type.
• setNumberAlt: true: Also set numbered label text to related image alt text.
• setImgAlt: true: Alias behavior for alt-text numbering.
• noSetAlt: false: Legacy-compatible switch to enable alt-text numbering.
• labelMarkMap: Optional map to resolve overlapping labels by mark name. Example: { '図': 'pre-samp', 'リスト': 'pre-samp' }.

All type options above are active in the current implementation.

When multiple type options are enabled, counters are maintained per type.

Label overlap note (from p7d-markdown-it-p-captions dictionaries):

• Japanese 図 can match both img and pre-samp.
• Japanese リスト can match both pre-code and pre-samp.

If an input label can match multiple enabled types, the first matched type wins in this order: img -> video -> table -> pre-code -> pre-samp -> blockquote -> slide -> audio.

You can override this default with labelMarkMap.

Alt text numbering example

Input:

図 キャプション

![ALT-A](image.jpg)

図 キャプション

![ALT-B](image.jpg)

Output ({ setNumberAlt: true }):

図1 キャプション

![図1](image.jpg)

図2 キャプション

![図2](image.jpg)

Notes

• Original line endings are preserved, including mixed \r\n and \n.
• Fenced code blocks are skipped:
  • Backtick/tilde fences with 3+ markers (including variable-length fences like `````).
  • Full-line math fences ($, $$, ...).
• Indented code blocks (4 spaces or tab) are currently out of scope. Use fenced code blocks if you need guaranteed skip behavior.
• Non-string input is returned as-is.
• Alt-text rewriting targets inline image syntax such as ![alt](url "title") (including escaped chars and trailing attrs).
• Reference-style image links are out of scope.
• If caption-to-image pairing is ambiguous in dense sections, separate blocks explicitly (for example with <!---->).